U.S. patent application number 14/536282 was filed with the patent office on 2015-08-13 for rnase h-based assays utilizing modified rna monomers.
The applicant listed for this patent is Integrated DNA Technologies, Inc.. Invention is credited to Mark Aaron BEHLKE, Josesph DOBOSY, Scott ROSE, Joseph Alan WALDER.
Application Number | 20150225782 14/536282 |
Document ID | / |
Family ID | 42285407 |
Filed Date | 2015-08-13 |
United States Patent
Application |
20150225782 |
Kind Code |
A1 |
WALDER; Joseph Alan ; et
al. |
August 13, 2015 |
RNASE H-BASED ASSAYS UTILIZING MODIFIED RNA MONOMERS
Abstract
The present invention pertains to novel oligonucleotide
compounds for use in various biological assays, such as nucleic
acid amplification, ligation and sequencing reactions. The novel
oligonucleotides comprise a ribonucleic acid domain and a blocking
group at or near the 3' end of the oligonucleotide. These compounds
offer an added level of specificity previously unseen. Methods for
performing nucleic acid amplification, ligation and sequencing are
also provided. Additionally, kits containing the oligonucleotides
are also disclosed herein.
Inventors: |
WALDER; Joseph Alan;
(Chicago, IL) ; BEHLKE; Mark Aaron; (Coralville,
IA) ; ROSE; Scott; (Coralville, IA) ; DOBOSY;
Josesph; (Coralville, IA) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
Integrated DNA Technologies, Inc. |
Coralville |
IA |
US |
|
|
Family ID: |
42285407 |
Appl. No.: |
14/536282 |
Filed: |
November 7, 2014 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
12507142 |
Jul 22, 2009 |
8911948 |
|
|
14536282 |
|
|
|
|
12433896 |
Apr 30, 2009 |
|
|
|
12507142 |
|
|
|
|
61049204 |
Apr 30, 2008 |
|
|
|
Current U.S.
Class: |
506/2 ; 435/6.11;
435/6.12 |
Current CPC
Class: |
C12Q 1/6848 20130101;
C12Q 1/6844 20130101; C12Q 1/686 20130101; C12Q 1/6844 20130101;
C12Q 1/6853 20130101; C12Q 1/6858 20130101; C12Q 1/6853 20130101;
C12Q 2549/101 20130101; C12Q 2549/101 20130101; C12Q 1/6844
20130101; C12Q 2521/301 20130101; C12Q 2525/121 20130101; C12Q
2525/186 20130101; C12Q 2525/186 20130101; C12Q 2525/186 20130101;
C12Q 2549/101 20130101; C12Q 2525/186 20130101; C12Q 2525/121
20130101; C12Q 2549/101 20130101; C12Q 1/6853 20130101; C12Q
2521/301 20130101 |
International
Class: |
C12Q 1/68 20060101
C12Q001/68 |
Claims
1. A method of amplifying a target DNA sequence, said method
comprising the steps of: (a) providing a reaction mixture
comprising (i) an oligonucleotide primer having a cleavage domain
positioned 5' of a blocking group, said blocking group linked at or
near the end of the 3'-end of the oligonucleotide primer wherein
said blocking group prevents primer extension and/or inhibits the
primer from serving as a template for DNA synthesis, (ii) a sample
nucleic acid that may or may not have the target sequence, (iii) a
cleaving enzyme and (iv) a polymerase wherein said cleaving enzyme
is a hot start cleaving enzyme which is thermostable and has
reduced activity at lower temperatures; (b) hybridizing the primer
to the target DNA sequence to form a double-stranded substrate; (c)
cleaving the hybridized primer with said cleaving enzyme at a point
within or adjacent to the cleavage domain to remove the blocking
group from the primer; and (d) extending the primer with the
polymerase.
2. The method of claim 1, wherein the hot start cleaving enzyme is
an RNase H enzyme.
3. The method of claim 2, wherein said RNase H enzyme is an RNase
H2 enzyme.
4. The method of claim 3, wherein said RNase H2 enzyme inherently
has lower activity at reduced temperature, is reversibly
inactivated by chemical modification or by a blocking antibody.
5. The method of claim 1, wherein said cleaving enzyme is a
sequence specific double stranded endonuclease.
6. The method of claim 5, wherein said sequence specific double
stranded endonuclease is a restriction enzyme.
7. The method of claim 1, wherein the blocking group is attached to
the 3'-terminal nucleotide of the primer.
8. The method of claim 1, wherein the blocking group is attached 5'
of the 3'-terminal residue.
9. The method of claim 8, wherein the blocking group includes one
or more abasic residues or modified nucleosides.
10. The method of claim 9, wherein the abasic residue is a C3
spacer.
11. The method of claim 9, wherein the modified nucleoside is a
2'-O-methyl ribose residue.
12. The method of claim 1, wherein the blocking group includes a
label permitting detection of the amplification reaction.
13. The method of claim 12, wherein the label is a fluorophore, a
quencher, biotin, or a hapten.
14. The method of claim 12, wherein the label is a mass tag for
detection of the amplification reaction by mass spectrometry.
15. The method of claim 2, wherein the cleavage domain is a
continuous sequence of 3 or more RNA residues.
16. The method of claim 15, wherein said cleavage domain further
comprises one or more of the following moieties: a DNA residue, an
abasic residue, a modified nucleoside, or a modified phosphate
internucleotide linkage.
17. The method of claim 3, wherein the cleavage domain is a single
RNA residue or two adjacent RNA residues.
18. The method of claim 3, wherein the cleavage domain lacks an RNA
residue.
19. The method of claim 18, wherein the cleavage domain comprises
one or more 2'-modified nucleosides.
20. The method of claim 19, wherein said 2'-modified nucleoside is
a single 2'-fluoronucleoside.
21. The method of embodiment 19, wherein the cleavage domain is two
adjacent 2'-fluoronucleoside residues.
22. The method of claim 18, wherein the cleavage reaction is
carried out in the presence of one or more of the following
divalent cations: manganese, cobalt, nickel or zinc.
23. The method of claim 22, wherein magnesium is also present in
the reaction mixture.
24. The method of claim 17, wherein a sequence within or flanking
the cleavage domain contains one or more internucleoside linkages
resistant to nuclease cleavage.
25. The method of claim 24, wherein said nuclease resistant linkage
is phosphorothioate, phosphorodithioate, methylphosphonate or an
abasic residue.
26. The method of claim 24, wherein said nuclease resistant linkage
is on the 3' side of the cleavage domain.
27. The method of claim 1, further comprising a second primer in
reverse orientation from the first primer to support PCR.
28. The method of claim 27, wherein the second primer is an
unmodified DNA primer.
29. The method of claim 27, wherein the second primer comprises a
cleavage domain (new) positioned 5' of a blocking group, said
blocking group linked at or near the end of the 3'-end of the
oligonucleotide primer wherein said blocking group prevents primer
extension.
30. The method of claim 27, wherein the PCR assay is used to
discriminate between variant alleles.
31. The method of claim 30, wherein a secondary mutation site is
incorporated within or flanking the cleavage domain to enhance
detection of the variant allele.
32. The method of claim 30, wherein a modified nucleoside is
incorporated within or flanking the cleavage domain to enhance
detection of the variant allele.
33. The method of claim 32, wherein said modified nucleoside is a
2'-O-methyl ribose residue.
34. The method of claim 30, wherein a nuclease resistant linkage is
incorporated on the 3'-side of the cleavage domain.
35. The method of claim 27, wherein the PCR assay is used to
quantitate the abundance of the target nucleic sequence in the
sample.
36. The method of embodiment 27, wherein the PCR assay is a
primer-probe PCR assay.
37. The method of claim 36, wherein the primer having a 5' label
domain includes a cleavage domain.
38. The method of claim 37, wherein the cleavage domain is an RNase
H cleavage domain.
39. The method of claim 38, wherein the RNase H cleavage domain is
an RNase H2 cleavage domain.
40. A method of amplifying a target DNA sequence, said method
comprising the steps of: (a) providing a reaction mixture
comprising (i) an oligonucleotide primer having a cleavage domain
with a cleavage site which is cleavable by an RNase H2 enzyme,
wherein said cleavage site is positioned 5' of a blocking group at
or near the 3'-end of the oligonucleotide primer which prevents
primer extension and/or PCR, and wherein said cleavage domain
includes an RNA residue and an abasic residue, (ii) a sample
nucleic acid that may or may not have the target sequence, (iii) a
DNA polymerase, (iv) an RNase H2 enzyme; and (v) optionally a
second primer in reverse orientation to support PCR; (b)
hybridizing the blocked primer to the target DNA sequence in said
reaction mixture to form a double-stranded substrate; (c) cleaving
the hybridized blocked primer with said RNase H2 to remove the
blocking group from the primer; and (d) extending the cleaved
primer with the DNA polymerase.
41. The method of claim 40 wherein said cleavage domain includes a
sequence 5' to 3' Rx or RDx, where R is an RNA residue, D is a DNA
residue, and x is an abasic residue.
42. The method of claim 41 wherein said cleavage domain has the
sequence RDxxD.
43. The method of claim 41 wherein x is a C3 spacer.
44. The method of claim 40 wherein the assay is used to
discriminate between variant alleles.
45. The method of claim 44 wherein said allelic variants are single
nucleotide polymorphisms.
Description
[0001] This application is a continuation application of U.S.
application Ser. No. 12/507,142, filed Jul. 2, 2009 (now U.S. Pat.
No. 8,911,948), which in turn is a continuation-in-part of U.S.
application Ser. No. 12/433,896, filed Apr. 30, 2009, which in turn
claims benefit under 35 U.S.C. 119(e) to U.S. provisional
application No. 61/049,204 filed on Apr. 30, 2008. All of the
above-identified applications are hereby incorporated by reference
in their entirety.
FIELD OF THE INVENTION
[0002] This invention pertains to methods of cleaving a nucleic
acid strand to initiate, assist, monitor or perform biological
assays.
BACKGROUND OF THE INVENTION
[0003] The specificity of primer-based amplification reactions,
such as the polymerase chain reaction (PCR), largely depends on the
specificity of primer hybridization with a DNA template. Under the
elevated temperatures used in a typical amplification reaction, the
primers ideally hybridize only to the intended target sequence and
form primer extension products to produce the complement of the
target sequence. However, amplification reaction mixtures are
typically assembled at room temperature, well below the temperature
needed to insure primer hybridization specificity. Under lower
temperature conditions, the primers may bind non-specifically to
other partially complementary nucleic acid sequences or to other
primers and initiate the synthesis of undesired extension products,
which can be amplified along with the target sequence.
Amplification of non-specific primer extension products can compete
with amplification of the desired target sequences and can
significantly decrease the efficiency of the amplification of the
desired sequence. Non-specific amplification can also give rise in
certain assays to a false positive result.
[0004] One frequently observed type of non-specific amplification
product in PCR is a template independent artifact of the
amplification reaction known as "primer dimers". Primer dimers are
double-stranded fragments whose length typically is close to the
sum of the two primer lengths and are amplified when one primer is
extended over another primer. The resulting duplex forms an
undesired template which, because of its short length, is amplified
efficiently.
[0005] Non-specific amplification can be reduced by reducing the
formation of primer extension products (e.g., primer dimers) prior
to the start of the reaction. In one method, referred to as a
"hot-start" protocol, one or more critical reagents are withheld
from the reaction mixture until the temperature is raised
sufficiently to provide the necessary hybridization specificity. In
this manner, the reaction mixture cannot support primer extension
at lower temperatures. Manual hot-start methods, in which the
reaction tubes are opened after the initial high temperature
incubation step and the missing reagents are added, are labor
intensive and increase the risk of contamination of the reaction
mixture.
[0006] Alternatively, a heat sensitive material, such as wax, can
be used to separate or sequester reaction components, as described
in U.S. Pat. No. 5,411,876, and Chou et al., 1992, Nucl. Acids Res.
20(7):1717-1723. In these methods, a high temperature pre-reaction
incubation melts the heat sensitive material, thereby allowing the
reagents to mix.
[0007] Another method of reducing the formation of primer extension
products prior to the start of PCR relies on the heat-reversible
inactivation of the DNA polymerase. U.S. Pat. Nos. 5,773,258 and
5,677,152, both incorporated herein by reference, describe DNA
polymerases reversibly inactivated by the covalent attachment of a
modifier group. Incubation of the inactivated DNA polymerase at
high temperature results in cleavage of the modifier-enzyme bond,
thereby releasing an active form of the enzyme. Non-covalent
reversible inhibition of a DNA polymerase by DNA
polymerase-specific antibodies is described in U.S. Pat. No.
5,338,671, incorporated herein by reference.
[0008] One objective of the present invention can be used, for
example, to address the problem of carry-over cross contamination
which is a significant concern in amplification reactions,
especially PCR wherein a large number of copies of the amplified
product are produced. In the prior art, attempts have been made to
solve this problem in a number of ways. For example, direct UV
irradiation can effectively remove contaminating DNA (Rys &
Persing, 1993, J Clin Microbiol. 31(9):2356-60 and Sarkar &
Sommer, 1990 Nature. 343(6253):27) but the irradiation of the PCR
reagents must take place before addition of polymerase, primers,
and template DNA. Furthermore, this approach may be inefficient
because the large numbers of mononucleotides present in the
reaction will absorb much of the UV light. An alternative, the "UNG
method", incorporates dUTP into the amplified fragments to alter
the composition of the product so that it is different from native,
naturally occurring DNA (Longo et al. 1990, Gene, 93(1): 125-128).
The enzyme Uracil-N-Glycosylase (UNG) is added together with the
other components of the PCR mixture. The UNG enzyme will cleave the
uracil base from DNA strands of contaminating amplicons before
amplification, and render all such products unable to act as a
template for new DNA synthesis without affecting the sample DNA.
The UNG enzyme is then heat-inactivated and PCR is then carried
out. The requirement for dUTP and the UNG enzyme adds significantly
to the cost of performing PCR.
[0009] Another objective of the present invention is to provide PCR
assays in which a hot-start reaction is achieved through a coupled
reaction sequence with a thermostable RNase H.
Ribonuclease Enzymes
[0010] Ribonucleases (RNases) are enzymes that catalyze the
hydrolysis of RNA into smaller components. The enzymes are present
internally; in bodily fluids; on the surface of skin; and on the
surface of many objects, including untreated laboratory glasswear.
Double-stranded RNases are present in nearly all intracellular
environments and cleave RNA-containing, double-stranded constructs.
Single-stranded RNases are ubiquitous in extracellular
environments, and are therefore extremely stable in order to
function under a wide range of conditions.
[0011] The RNases H are a conserved family of ribonucleases which
are present in all organisms examined to date. There are two
primary classes of RNase H: RNase H1 and RNase H2. Retroviral RNase
H enzymes are similar to the prokaryotic RNase H1. All of these
enzymes share the characteristic that they are able to cleave the
RNA component of an RNA:DNA heteroduplex. The human and mouse RNase
H1 genes are 78% identical at the amino acid level (Cerritelli, et
al., (1998) Genomics, 53, 300-307). In prokaryotes, the genes are
named rnha (RNase H1) and rnhb (RNase H2). A third family of
prokaryotic RNases has been proposed, rnhc (RNase H3) (Ohtani, et
al. (1999) Biochemistry, 38, 605-618).
[0012] Evolutionarily, "ancient" organisms (archaeal species) in
some cases appear to have only a single RNase H enzyme which is
most closely related to the modern RNase H2 enzymes (prokaryotic)
(Ohtani, et al., J Biosci Bioeng, 88, 12-19). Exceptions do exist,
and the archaeal Halobacterium has an RNase H1 ortholog (Ohtani, et
al., (2004) Biochem J, 381, 795-802). An RNase H1 gene has also
been identified in Thermus thermophilus (Itaya, et al., (1991)
Nucleic Acids Res, 19, 4443-4449). RNase H2 enzymes appear to be
present in all living organisms. Although all classes of RNase H
enzymes hydrolyze the RNA component of an RNA:DNA heteroduplex, the
substrate and co-factor requirements are different. For example,
the Type II enzymes utilize Mg.sup.++, Mn.sup.++, Co.sup.++ (and
sometimes Ni.sup.++) as cofactor, while the Type I enzymes require
Mg.sup.++ and can be inhibited by Mn.sup.++ ions. The reaction
products are the same for both classes of enzymes: the cleaved
products have a 3'-OH and 5'-phosphate (See FIG. 1). RNase III
class enzymes which cleave RNA:RNA duplexes (e.g., Dicer, Ago2,
Drosha) result in similar products and contain a nuclease domain
with similarity to RNase H. Most other ribonucleases, and in
particular single stranded ribonucleases, result in a cyclic
2',3'-phosphate and 5'-OH products (see FIG. 2).
Type I RNase H
[0013] E. coli RNase H1 has been extensively characterized. A large
amount of work on this enzyme has been carried out, focusing on
characterization of substrate requirements as it impacts antisense
oligonucleotide design; this has included studies on both the E.
coli RNase H1 (see Crooke, et al., (1995) Biochem J, 312 (Pt 2),
599-608; Lima, et al., (1997) J Biol Chem, 272, 27513-27516; Lima,
et al., (1997) Biochemistry, 36, 390-398; Lima, et al., (1997) J
Biol Chem, 272, 18191-18199; Lima, et al., (2007) Mol Pharmacol,
71, 83-91; Lima, et al., (2007) Mol Pharmacol, 71, 73-82; Lima, et
al., (2003) J Biol Chem, 278, 14906-14912; Lima, et al., (2003) J
Biol Chem, 278, 49860-49867) and the Human RNase H1 (see Wu, et
al., (1998) Antisense Nucleic Acid Drug Dev, 8, 53-61; Wu, et al.,
(1999) J Biol Chem, 274, 28270-28278; Wu, et al., (2001) J Biol
Chem, 276, 23547-23553). In tissue culture, overexpression of human
RNase H1 increases potency of antisense oligos (ASOs) while
knockdown of RNase H1 using either siRNAs or ASOs decreases potency
of antisense oligonucleotides.
[0014] Type I RNase H requires multiple RNA bases in the substrate
for full activity. A DNA/RNA/DNA oligonucleotide (hybridized to a
DNA oligonucleotide) with only 1 or 2 RNA bases is inactive. With
E. coli RNase H1 substrates with three consecutive RNA bases show
weak activity. Full activity was observed with a stretch of four
RNA bases (Hogrefe, et al., (1990) J Biol Chem, 265, 5561-5566). An
RNase H1 was cloned from Thermus thermophilus in 1991 which has
only 56% amino acid identity with the E. coli enzyme but which has
similar catalytic properties (Itaya, et al., (1991) Nucleic Acids
Res, 19, 4443-4449). This enzyme was stable at 65.degree. C. but
rapidly lost activity when heated to 80.degree. C.
[0015] The human RNase H1 gene (Type I RNase H) was cloned in 1998
(Genomics, 53, 300-307 and Antisense Nucleic Acid Drug Dev, 8,
53-61). This enzyme requires a 5 base RNA stretch in DNA/RNA/DNA
chimeras for cleavage to occur. Maximal activity was observed in 1
mM Mg.sup.++ buffer at neutral pH and Mn.sup.++ ions were
inhibitory (J Biol Chem, 274, 28270-28278). Cleavage was not
observed when 2'-modified nucleosides (such as 2'-OMe, 2'-F, etc.)
were substituted for RNA.
[0016] Three amino acids (Asp-10, Glu-48, and Asp-70) make up the
catalytic site of E. coli RNase H1 which resides in the highly
conserved carboxy-terminal domain of the protein (Katayanagi, et
al., (1990) Nature, 347, 306-309); this domain has been evaluated
by both site directed mutagenesis and crystal structure
determination. The same amino acids are involved in coordination of
the divalent ion cofactor.
[0017] Interestingly, 2'-modification of the substrate duplex
alters the geometry of the helix and can adversely affect activity
of RNase H1. 2'-O-(2-methoxyl)ethyl (MOE) modifications flanking
the RNA segment reduce cleavage rates, presumably due to
alterations in the sugar conformation and helical geometry. Locked
nucleic acid (LNA) bases perturb helical geometry to a greater
degree and impacted enzyme activity to a greater extent (Mol
Pharmacol, 71, 83-91 and Mol Pharmacol, 71, 73-82). Damha (McGill
University) has studied the effects of 2'-F modified nucleosides
(2'-deoxy-2'-fluoro-b-D-ribose) when present in the substrate
duplex and finds that this group cannot be cleaved by RNase H1
(Yazbeck, et al., (2002) Nucleic Acids Res, 30, 3015-3025).
Formulas A and B illustrate the two different mechanisms that have
been proposed for RNase H1 cleavage, both of which require
participation of the 2'OH group.
##STR00001##
Formulas A and B
[0018] Damha's studies are consistent with the known active site of
the enzyme, wherein the reaction mechanism involves the 2'-OH
group. The enzyme active site resides within a cluster of lysine
residues which presumably contribute to electrostatic binding of
the duplex. Interaction between the binding surface and negatively
charged phosphate backbone is believed to occur along the minor
grove of the RNA:DNA heteroduplex (Nakamura, et al., (1991) Proc
Natl Acad Sci USA, 88, 11535-11539); changes in structure that
affect the minor groove should therefore affect interactions
between the substrate and the active site. For example, the minor
groove width is 7.5 .ANG. in a B-form DNA:DNA duplex, is 11 .ANG.
in a pure A-form RNA:RNA duplex, and is 8.5 .ANG. in the hybrid
A-form duplex of an RNA:DNA duplex (Fedoroff et al., (1993) J Mol
Biol, 233, 509-523). 2'-modifications protrude into the minor
groove, which may account for some of the behavior of these groups
in reducing or eliminating activity of modified substrates for
cleavage by RNase H1. Even a 2'-F nucleoside, which is the most
"conservative" RNA analog with respect to changing chemical
structure, adversely affects activity.
Type II RNase H
[0019] The human Type II RNase H was first purified and
characterized by Eder and Walder in 1991 (Eder, et al., (1991) J
Biol Chem, 266, 6472-6479). This enzyme was initially designated
human RNase H1 because it had the characteristic divalent metal ion
dependence of what was then known as Class I RNases H. In the
current nomenclature, it is a Type II RNase H enzyme. Unlike the
Type I enzymes which are active in Mg.sup.++ but inhibited by
Mn.sup.++ ions, the Type II enzymes are active with a wide variety
of divalent cations. Optimal activity of human Type II RNase H is
observed with 10 mM Mg.sup.++, 5 mM Co.sup.++, or 0.5 mM
Mn.sup.++.
[0020] Importantly, the substrate specificity of the Type II RNase
H (hereafter referred to as RNase H2) is different from RNase H1.
In particular, this enzyme can cleave a single ribonucleotide
embedded within a DNA sequence (in duplex form) (Eder, et al.,
(1993) Biochimie, 75, 123-126). Interestingly, cleavage occurs on
the 5' side of the RNA residue (See FIG. 3). See a recent review by
Kanaya for a summary of prokaryotic RNase H2 enzymes (Kanaya (2001)
Methods Enzymol, 341, 377-394).
[0021] The E. coli RNase H2 gene has been cloned (Itaya, M. (1990)
Proc Natl Acad Sci U SA, 87, 8587-8591) and characterized (Ohtani,
et al., (2000) J Biochem (Tokyo), 127, 895-899). Like the human
enzyme, the E. coli enzyme functions with Mn.sup.++ ions and is
actually more active with manganese than magnesium.
[0022] RNase H2 genes have been cloned and the enzymes
characterized from a variety of eukaryotic and prokaryotic sources.
The RNase H2 from Pyrococcus kodakaraensis (KOD 1) has been cloned
and studied in detail (Haruki, et al., (1998) J Bacteriol, 180,
6207-6214; Mukaiyama, et al., (2004) Biochemistry, 43,
13859-13866). The RNase H2 from the related organism Pyrococcus
furious has also been cloned but has not been as thoroughly
characterized (Sato, et al., (2003) Biochem Biophys Res Commun,
309, 247-252).
[0023] The RNase H2 from Methanococcus jannaschii was cloned and
characterized by Lai (Lai, et al., (2000) Structure, 8, 897-904;
Lai et al., (2003) Biochemistry, 42, 785-791). Isothermal titration
calorimetry was used to quantitatively measure metal ion binding to
the enzyme. They tested binding of Mn.sup.++, Mg.sup.++, Ca.sup.++,
and Ba.sup.++ and in all cases observed a 1:1 molar binding ratio,
suggesting the presence of only a single divalent metal ion
cofactor in the enzyme's active site. The association constant for
Mn.sup.++ was 10-fold higher than for Mg.sup.++. Peak enzyme
activity was seen at 0.8 mM MnCl.sub.2.
[0024] Nucleic acid hybridization assays based on cleavage of an
RNA-containing probe by RNase H such as the cycling probe reaction
(Walder et al., U.S. Pat. No. 5,403,711) have been limited in the
past by background cleavage of the oligonucleotide by contaminating
single-stranded ribonucleases and by water catalyzed hydrolysis
facilitated by Mg.sup.2+ and other divalent cations. The effect of
single-stranded ribonucleases can be mitigated to a certain degree
by inhibitors such as RNasin that block single-stranded
ribonucleases but do not interfere with the activity of RNase
H.
[0025] Single-stranded ribonucleases cleave 3' of an RNA residue,
leaving a cyclic phosphate group at the 2' and 3' positions of the
ribose (See FIG. 2). The same products are produced by spontaneous
water catalyzed hydrolysis. In both cases, the cyclic phosphate can
hydrolyze further forming a 3'-monophosphate ester in the enzyme
catalyzed reaction, or a mixture of the 3'- and 2'-monophosphate
esters through spontaneous hydrolysis. The difference between the
cleavage products formed by RNase H (FIG. 1) and those formed by
nonspecific cleavage of the probe (FIG. 2) provides a basis for
distinguishing between the two pathways. This difference is even
more pronounced when comparing cleavage by RNase H2 and
single-stranded ribonucleases with substrates having only a single
RNA residue. In that case, RNase H2 and single-stranded
ribonucleases attack at different positions along the phosphate
backbone (See FIG. 3).
[0026] RNase H has been used as a cleaving enzyme in cycling probe
assays, in PCR assays (Han et al., U.S. Pat. No. 5,763,181; Sagawa
et al., U.S. Pat. No. 7,135,291; and Behlke and Walder, U.S. Pat.
App. No. 20080068643) and in polynomial amplification reactions
(Behlke et al., U.S. Pat. No. 7,112,406). Despite improvements
offered by these assays, there remain considerable limitations. The
PCR assays utilize a hot-start DNA polymerase which adds
substantially to the cost. Moreover, each time an alternative DNA
polymerase is required a new hot-start version of the enzyme must
be developed. In addition, the utility of these various assays has
been limited by undesirable cleavage of the oligonucleotide probe
or primer used in the reaction, including water and divalent metal
ion catalyzed hydrolysis 3' to RNA residues, hydrolysis by
single-stranded ribonucleases and atypical cleavage reactions
catalyzed by Type II RNase H enzymes at positions other than the
5'-phosphate of an RNA residue. The present invention overcomes
these limitations and offers further advantages and new assay
formats for use of RNase H in biological assays.
[0027] The current invention provides novel biological assays that
employ RNase H cleavage in relation to nucleic acid amplification,
detection, ligation, sequencing, and synthesis. Additionally, the
invention provides new assay formats to utilize cleavage by RNase H
and novel oligonucleotide substrates for such assays. The
compounds, kits, and methods of the present invention provide a
convenient and economic means of achieving highly specific
primer-based amplification reactions that are substantially free of
nonspecific amplification impurities such as primer dimers. The
methods and kits of the present invention avoid the need for
reversibly inactivated DNA polymerase and DNA ligase enzymes.
BRIEF SUMMARY OF THE INVENTION
[0028] One objective of the present invention is to enable hot
start protocols in nucleic acid amplification and detection assays
including but not limited to PCR, OLA (oligonucleotide ligation
assays), LCR (ligation chain reaction), polynomial amplification
and DNA sequencing, wherein the hot start component is a
thermostable RNase H or other nicking enzyme that gains activity at
the elevated temperatures employed in the reaction. Such assays
employ a modified oligonucleotide of the invention that is unable
to participate in the reaction until it hybridizes to a
complementary nucleic acid sequence and is cleaved to generate a
functional 5'- or 3'-end. Compared to the corresponding assays in
which standard unmodified DNA oligonucleotides are used the
specificity is greatly enhanced. Moreover the requirement for
reversibly inactivated DNA polymerases or DNA ligases is
eliminated.
[0029] In the case of assays involving primer extension (e.g., PCR,
polynomial amplification and DNA sequencing) the modification of
the oligonucleotide inhibiting activity is preferably located at or
near the 3'-end. In some embodiments where the oligonucleotides are
being used as primers, the oligonucleotide inhibiting activity may
be positioned near the 3' end of the oligonucleotide, e.g., up to
about 10 bases from the 3' end of the oligonucleotide of the
invention. In other embodiments, the oligonucleotide inhibiting
activity may be positioned near the 3' end, e.g., about 1-6 bases
from the 3' end of the oligonucleotide of the invention. In other
embodiments, the oligonucleotide inhibiting activity may be
positioned near the 3' end, e.g., about 1-5 bases from the 3' end
of the oligonucleotide of the invention. In other embodiments, the
oligonucleotide inhibiting activity may be positioned near the 3'
end, e.g., about 1-3 bases from the 3' end of the oligonucleotide
of the invention. In other embodiments, the precise position (i.e.,
number of bases) from the 3' end where the oligonucleotide
inhibiting activity may be positioned will depend upon factors
influencing the ability of the oligonucleotide primer of the
invention to hybridize to a shortened complement of itself on the
target sequence (i.e., the sequence for which hybridization is
desired). Such factors include but are not limited to Tm, buffer
composition, and annealing temperature employed in the
reaction(s).
[0030] For ligation assays (e.g., OLA and LCR) the modification
inhibiting activity may be located at or near either the 3'- or
5'-end of the oligonucleotide. In other embodiments, for ligation
assays, modification inhibitory activity, if used, is preferably
placed within the domain that is 3' to the cleavable RNA base in
the region that is removed by probe cleavage. In other embodiments,
for ligation assays, C3 spacers may be positioned close to the RNA
base in the oligonucleotide probes of the invention to improve
specificity that is helpful for improving mismatch discrimination.
In other embodiments, in an OLA assay, where readout depends upon a
PCR assay to amplify the product of a ligation event, any blocking
group may be placed in the domain of the oligonucleotide of the
invention that is removed by RNase H cleavage. In such embodiments,
in an OLA assay where readout depends upon a PCR assay to amplify
the product of a ligation event, the precise position of the
blocking group in the RNase H cleavable domain may be adjusted to
alter specificity for cleavage and precise placement of the
blocking group relative to the cleavable RNA bases may alter the
amount of enzyme needed to achieve optimal cleavage rates.
[0031] Yet a further objective of the present invention is to
provide novel modifications of oligonucleotides to interfere with
primer extension and ligation.
[0032] Yet a further objective of the present invention is to
provide modifications of oligonucleotides that prevent the
oligonucleotide from serving as a template for DNA synthesis and
thereby interfere with PCR.
[0033] Yet a further objective of the invention is to provide
modified oligonucleotide sequences lacking RNA that are cleaved by
RNase H. In one such embodiment, the oligonucleotide contains a
single 2'-fluoro residue and cleavage is mediated by a Type II
RNase H enzyme. In a more preferred embodiment the oligonucleotide
contains two adjacent 2'-fluoro residues.
[0034] Yet a further objective of the present invention is to
provide oligonucleotides for use in the above mentioned assays that
are modified so as to inhibit undesired cleavage reactions
including but not limited to water and divalent metal ion catalyzed
hydrolysis 3' to RNA residues, hydrolysis by single-stranded
ribonucleases and atypical cleavage reactions catalyzed by Type II
RNase H enzymes at positions other than the 5'-phosphate of an RNA
residue (see FIG. 3). In one such embodiment the 2'-hydroxy group
of an RNA residue is replaced with an alternative functional group
such as fluorine or an alkoxy substituent (e.g., O-methyl). In
another such embodiment the phosphate group 3' to an RNA residue is
replaced with a phosphorothioate or a dithioate linkage. In yet
another embodiment the oligonucleotide is modified with nuclease
resistant linkages further downstream from the 3'-phosphate group
of an RNA residue or on the 5'-side of an RNA residue to prevent
aberrant cleavage by RNase H2. Nuclease resistant linkages useful
in such embodiments include phosphorothioates, dithioates,
methylphosphonates, and abasic residues such as a C3 spacer.
Incorporation of such nuclease resistant linkages into
oligonucleotide primers used in PCR assays of the present invention
has been found to be particularly beneficial (see Examples 25, 27
and 28).
[0035] Yet a further objective of the invention is to provide
oligonucleotides for use in the above-mentioned assays that are
modified at positions flanking the cleavage site to provide
enhanced discrimination of variant alleles. Such modifications
include but are not limited to 2'-O-methyl RNA residues and
secondary mismatch substitutions (see Example 23).
[0036] Yet a further objective is to provide oligonucleotides and
assay formats for use in the present invention wherein cleavage of
the oligonucleotide can be measured by a change in fluorescence. In
one such embodiment a primer cleavable by RNase H is labeled with a
fluorophore and a quencher and the assay is monitored by an
increase in fluorescence (see Examples 19-21).
[0037] Yet a further objective of the invention is to provide RNase
H compositions and protocols for their use in which the enzyme is
thermostable and has reduced activity at lower temperatures.
[0038] In yet a further embodiment a Type II RNase H is employed in
a cycling probe reaction in which the RNA residue in the probe is
replaced with a 2'-fluoro residue. In a more preferred embodiment a
probe with two adjacent 2'-fluoro residues is used.
[0039] Many of the aspects of the present invention relating to
primer extension assays, ligation assays and cycling probe
reactions are summarized in Tables 1, 2, and 3, respectively.
[0040] In yet a further embodiment of the invention Type II RNase H
enzymes are used in novel methods for DNA sequencing.
[0041] In yet a further embodiment of the invention Type II RNase H
enzymes are used in novel methods for DNA synthesis.
BRIEF DESCRIPTION OF THE FIGURES
[0042] FIG. 1 depicts the cleavage pattern that occurs with an
RNase H enzyme on a substrate containing multiple RNA bases.
[0043] FIG. 2 depicts the cleavage pattern that occurs with a
single-stranded ribonuclease enzyme or through water catalyzed
hydrolysis, wherein the end-product results in a cyclic phosphate
group at the 2' and 3' positions of the ribose.
[0044] FIG. 3 depicts the cleavage sites for RNase H2 and
single-stranded ribonucleases on a substrate containing a single
RNA base.
[0045] FIGS. 4A and 4B are photographs of SDS 10% polyacrylamide
gels that illustrate the induced protein produced from five
Archaeal RNase H2 synthetic genes. FIG. 4A shows induced protein
for Pyrococcus furiosus and Pyrococus abyssi. FIG. 4B shows induced
protein for Methanocaldococcus jannaschii, Sulfolobus solfataricus,
Pyrococcus kodadarensis.
[0046] FIG. 5 shows a Coomassie Blue stained protein gel showing
pure, single bands after purification using nickel affinity
chromatography of recombinant His tag RNase H2 proteins.
[0047] FIG. 6 shows a Western blot done using anti-His tag
antibodies using the protein gel from FIG. 5.
[0048] FIG. 7 is a photograph of a gel that shows the digestion of
a duplex, containing a chimeric 11 DNA-8 RNA-11 DNA strand and a
complementary DNA strand, by recombinant RNase H2 enzymes from
Pyrococcus kodakaraensis, Pyrococcus furiosus, and Pyrococcus
abyssi.
[0049] FIGS. 8A and 8B are photographs of gels that show the
digestion of a duplex, containing a chimeric 14 DNA-1 RNA-15 DNA
strand and a complementary DNA strand, by recombinant RNase H2
enzymes from Pyrococcus abyssi, Pyrococcus furiosus, and
Methanocaldococcus jannaschii (FIG. 8A) and Pyrococcus
kodakaraensis (FIG. 8B).
[0050] FIG. 9 shows the effects of incubation at 95.degree. C. for
various times on the activity of the Pyrococcus abyssi RNase H2
enzyme.
[0051] FIG. 10 is a photograph of a gel that shows the relative
amounts of cleavage of a single ribonucleotide-containing substrate
by Pyrococcus abyssi RNase H2 at various incubation
temperatures.
[0052] FIG. 11 is a graph showing the actual quantity of substrate
cleaved in the gel from FIG. 10.
[0053] FIG. 12 is a photograph of a gel that shows cleavage by
Pyrococcus abysii RNase H2 of various single 2' modified substrates
in the presence of different divalent cations.
[0054] FIG. 13 is a photograph of a gel that shows cleavage by
Pyrococcus abyssi RNase H2 of single 2'-fluoro or double 2'-fluoro
(di-fluoro) modified substrates. The divalent cation present was
Mn.
[0055] FIG. 14 is a graph quantifying the relative cleavage by
Pyrococcus abyssi RNase H2 of all 16 possible di-fluoro modified
substrates.
[0056] FIG. 15 is a graph quantifying the relative cleavage by
Pyrococcus abyssi RNase H2 of rN substrates with a variable number
of 3' DNA bases (i.e., number of DNA bases on the 3' side of the
RNA residue).
[0057] FIG. 16 is a graph quantifying the relative cleavage by
Pyrococcus abyssi RNase H2 of rN substrates with a variable number
of 5' DNA bases (i.e., number of DNA bases on the 5' side of the
RNA residue).
[0058] FIG. 17 is a graph quantifying the relative cleavage by
Pyrococcus abyssi RNase H2 of di-fluoro substrates with a variable
number of 3' DNA bases (i.e., number of DNA bases on the 3' side of
the fUfC residues).
[0059] FIG. 18 is a reaction schematic of RNase H2 activation of
blocked PCR primers.
[0060] FIG. 19 is a photograph of a gel that shows the products of
an end point PCR reaction performed with a single rU-containing
blocked primer. The suffix 2D, 3D, etc. represents the number of
DNA bases between the rU residues and the 3'-end of the primer. The
primer is blocked with a dideoxy C residue.
[0061] FIGS. 20A-B are PCR amplification plots for a 340 bp
amplicon within the human HRAS gene, using both unmodified and
blocked rN primers, without RNase H2 (FIG. 20A) and with RNase H2
(FIG. 20B). Cycle number is shown on the X-axis and relative
fluorescence intensity is shown on the Y-axis.
[0062] FIGS. 21A-B are PCR amplification plots for a 184 bp
amplicon within the human ETS2 gene, using both unmodified and
blocked rN primers, without RNase H2 (FIG. 21A) and with RNase H2
(FIG. 21B). Cycle number is shown on the X-axis and relative
fluorescence intensity is shown on the Y-axis.
[0063] FIGS. 22A-B are PCR amplification plots for a synthetic 103
bp amplicon, using both unmodified and 3'-fN modified primers,
without RNase H2 (FIG. 22A) and with RNase H2 (FIG. 22B).
[0064] FIG. 23 shows HPLC traces of a rN primer containing a single
phosphorothioate internucleoside modification. The top panel shows
the original synthesis product demonstrating resolution of the two
isomers. The middle panel is the purified Rp isomer and the bottom
panel is the purified Sp isomer.
[0065] FIG. 24 shows the relationship between RNase H2 versus RNase
A enzymatic cleavage with substrates having (SEQ ID NOs 321 and
121, respectively, in order of appearance) having a single RNA base
and different phosphorothioate stereoisomers.
[0066] FIG. 25 shows a photograph of a polyacrylamide gel used to
separate products from PCR reactions done using standard and
blocked/cleavable primers on a HCV amplicon showing that use of
standard primers results in formation of undesired small
primer-dimer species while use of blocked primers results in
specific amplification of the desired product. The nucleic acids
were imaged using fluorescent staining and the image was inverted
for clarity.
[0067] FIG. 26 is a graph quantifying the relative cleavage by
Pyrococcus abyssi RNase H2 of a radiolabeled rC containing
substrate in buffer containing different detergents at different
concentrations (expressed as % vol:vol).
[0068] FIG. 27 is a reaction schematic of RNase H2 activation of
fluorescence-quenched (F/Q) blocked PCR primers.
[0069] FIG. 28 is an amplification plot showing the fluorescence
signal resulting from use of unblocked primers with a
fluorescence-quenched dual-labeled probe (DLP) compared with a
blocked fluorescence-quenched cleavable primer for a 103 base
synthetic amplicon. Cycle number is shown on the X-axis and
relative fluorescence intensity is shown on the Y-axis.
[0070] FIG. 29 is an amplification plot showing the fluorescence
signal resulting from use of a F/Q configuration blocked
fluorescence-quenched cleavable primer compared with a Q/F
configuration blocked fluorescence-quenched cleavable primer for a
103 base synthetic amplicon. Cycle number is shown on the X-axis
and relative fluorescence intensity is shown on the Y-axis.
[0071] FIG. 30 is an amplification plot showing the fluorescence
signal resulting from use of F/Q configuration blocked
fluorescence-quenched cleavable primers to distinguish DNA
templates that differ at a single base within the SMAD 7 gene.
Panel (A) shows results from the FAM channel where the FAM-labeled
"C" allele probe was employed. Panel "B" shows results from the HEX
channel wherein the HEX-labeled "T" allele probe was employed.
Cycle number is shown on the X-axis and relative fluorescence
intensity is shown on the Y-axis.
[0072] FIG. 31 is a reaction schematic of RNase H2 cleavage of
fluorescence-quenched (FQT) primer used in a primer probe assay.
The Primer Domain is complementary to the target nucleic acid and
serves to primer DNA synthesis. The Reporter Domain is
non-complementary to target and contains a RNA base positioned
between a reporter dye and a quencher group. The Reporter Domain
remains single-stranded until conversion to double-stranded form
during PCR where this domain now serves as template. Conversion to
double-stranded form converts the Reporter Domain into a substrate
for RNase H2; cleavage by RNase H2 separates reporter from quencher
and is a detectable event.
[0073] FIG. 32 shows amplification plots of qPCR reactions done
with primers specific for the human Drosha gene using HeLa cell
cDNA. A) Reactions performed using unmodified primers and a
fluorescence-quenched dual-labeled probe (DLP), 5'-nuclease assay
format. The reaction was performed with or without template (HeLa
cDNA) as indicated. B) Reactions performed using a
fluorescence-quenched FQT For primer and an unmodified Rev primer
in a primer-probe assay format. Reactions were performed with or
without the addition of RNase H2 as indicated. Cycle number is
shown on the X-axis and relative fluorescence intensity is shown on
the Y-axis.
[0074] FIG. 33 shows the sequences of cleavable-blocked primers
that are either perfect match or contain a mismatch at position+2
relative to the single RNA base (2 bases 3'- to the
ribonucleotide). SMAD7 target sequences at SNP site rs4939827 are
aligned below the primers to indicate how this strategy results in
the presence of a single mismatch when primers hybridize with one
allele vs. a double mismatch when hybridize with the second allele.
DNA bases are uppercase, RNA bases are lowercase, and SpC3 is a
Spacer C3 modification. FIG. 33 discloses SEQ ID NOS 250-251, 253,
252, 322, 254-255, 257, 256 and 322, respectively, in order of
appearance.
[0075] FIG. 34 is a graph that shows the relative functional
activity of different oligonucleotide compositions to prime DNA
synthesis in a linear primer extension reaction.
[0076] FIG. 35 shows the scheme for performing cycles of DNA
sequencing by ligation using RNase H2 cleavable ligation probes.
FIG. 35 discloses the "3'-AGTCCAGGTCA" sequence as SEQ ID NO:
323.
[0077] FIG. 36 shows the scheme for hybridization, ligation, and
subsequent cleavage by RNase H2 of RNA-containing cleavable
ligation probes of a set of specific exemplary synthetic sequences.
(SEQ ID NOS 272, 274, 273, 275, 276-278, 324, 277, 325 and 277,
respectively, in order of appearance).
[0078] FIG. 37 shows a photograph of a polyacrylamide gel used to
separate products from ligation reactions done using cleavable
ligation probes on a synthetic template showing that the 9 mer
probes are efficiently ligated to the acceptor nucleic acid (ANA)
and that the ligation product is efficiently cleaved by RNase H2,
leaving an ANA species that is lengthened by one base. The nucleic
acids were imaged using fluorescent staining and the image was
inverted for clarity.
[0079] FIG. 38 shows the scheme for hybridization and ligation of
RNA-containing cleavable ligation probes containing either three or
four 5-nitroindole residues. FIG. 38 discloses SEQ ID NOS. 276,
279, 326, 277, 327 and 277, respectively, in order of
appearance.
[0080] FIG. 39 shows a photograph of a polyacrylamide gel used to
separate products from ligation reactions done using cleavable
ligation probes on a synthetic template showing that an 8 mer probe
containing three 5-nitroindole (3.times.5NI) bases is efficiently
ligated to an acceptor nucleic acid (ANA) whereas an 8 mer probe
containing four 5-nitroindole (4.times.5NI) bases is not. The
nucleic acids were imaged using fluorescent staining and the image
was inverted for clarity.
[0081] FIG. 40 shows a photograph of a polyacrylamide gel used to
separate ligation products from reactions done using cleavable
ligation probes on a synthetic template showing that an 8 mer probe
containing a single fixed DNA base at the 5'-end, four random
bases, and 3 universal base 5-nitroindoles can specifically ligate
to the target as directed by the single fixed DNA base.
[0082] FIGS. 41A and 41B show the scheme for a traditional
oligonucleotide ligation assay (OLA). FIG. 41A shows the three
oligonucleotides needed to interrogate a two allele target system.
FIG. 41B shows the steps involved in making a ligation product.
[0083] FIGS. 42A, 42B and 42C show the scheme for the RNase H2
cleavable oligonucleotide ligation assay (OLA) of the present
invention. FIG. 42A shows the four oligonucleotides needed to
interrogate a two allele target system. FIG. 42B shows the steps
involved in making a ligation product using the RNase H2 method.
FIG. 42C illustrates how this method tests the identity of the base
polymorphism twice.
[0084] FIG. 43 shows alignment of sequences (SEQ ID NOS 287,
328-329, 328, 319, 329, 328, 330, 328 and 330-331, respectively, in
order of appearance) used in the present Example during each step
of the RNase H2 cleavable probe OLA using fluorescence microbeads
and a Luminex L100 system to detect the ligation products.
[0085] FIG. 44 is a chart that shows the resulting fluorescent
signal detected by a Luminex L100 system to assess identity of the
reaction products generated from the RNase H2 allelic
discrimination OLA shown in FIG. 43.
DETAILED DESCRIPTION OF THE INVENTION
[0086] The current invention provides novel nucleic acid compounds
having a cleavage domain and a 3' or 5' blocking group. These
compounds offer improvements to existing methods for nucleic acid
amplification, detection, ligation, sequencing and synthesis. New
assay formats comprising the use of these novel nucleic acid
compounds are also provided.
DEFINITIONS
[0087] To aid in understanding the invention, several terms are
defined below.
[0088] The terms "nucleic acid" and "oligonucleotide," as used
herein, refer to polydeoxyribonucleotides (containing
2-deoxy-D-ribose), polyribonucleotides (containing D-ribose), and
to any other type of polynucleotide which is an N glycoside of a
purine or pyrimidine base. There is no intended distinction in
length between the terms "nucleic acid", "oligonucleotide" and
"polynucleotide", and these terms will be used interchangeably.
These terms refer only to the primary structure of the molecule.
Thus, these terms include double- and single-stranded DNA, as well
as double- and single-stranded RNA. For use in the present
invention, an oligonucleotide also can comprise nucleotide analogs
in which the base, sugar or phosphate backbone is modified as well
as non-purine or non-pyrimidine nucleotide analogs.
[0089] Oligonucleotides can be prepared by any suitable method,
including direct chemical synthesis by a method such as the
phosphotriester method of Narang et al., 1979, Meth. Enzymol.
68:90-99; the phosphodiester method of Brown et al., 1979, Meth.
Enzymol. 68:109-151; the diethylphosphoramidite method of Beaucage
et al., 1981, Tetrahedron Lett. 22:1859-1862; and the solid support
method of U.S. Pat. No. 4,458,066, each incorporated herein by
reference. A review of synthesis methods of conjugates of
oligonucleotides and modified nucleotides is provided in Goodchild,
1990, Bioconjugate Chemistry 1(3): 165-187, incorporated herein by
reference.
[0090] The term "primer," as used herein, refers to an
oligonucleotide capable of acting as a point of initiation of DNA
synthesis under suitable conditions. Such conditions include those
in which synthesis of a primer extension product complementary to a
nucleic acid strand is induced in the presence of four different
nucleoside triphosphates and an agent for extension (e.g., a DNA
polymerase or reverse transcriptase) in an appropriate buffer and
at a suitable temperature. Primer extension can also be carried out
in the absence of one or more of the nucleotide triphosphates in
which case an extension product of limited length is produced. As
used herein, the term "primer" is intended to encompass the
oligonucleotides used in ligation-mediated reactions, in which one
oligonucleotide is "extended" by ligation to a second
oligonucleotide which hybridizes at an adjacent position. Thus, the
term "primer extension", as used herein, refers to both the
polymerization of individual nucleoside triphosphates using the
primer as a point of initiation of DNA synthesis and to the
ligation of two oligonucleotides to form an extended product.
[0091] A primer is preferably a single-stranded DNA. The
appropriate length of a primer depends on the intended use of the
primer but typically ranges from 6 to 50 nucleotides, preferably
from 15-35 nucleotides. Short primer molecules generally require
cooler temperatures to form sufficiently stable hybrid complexes
with the template. A primer need not reflect the exact sequence of
the template nucleic acid, but must be sufficiently complementary
to hybridize with the template. The design of suitable primers for
the amplification of a given target sequence is well known in the
art and described in the literature cited herein.
[0092] Primers can incorporate additional features which allow for
the detection or immobilization of the primer but do not alter the
basic property of the primer, that of acting as a point of
initiation of DNA synthesis. For example, primers may contain an
additional nucleic acid sequence at the 5' end which does not
hybridize to the target nucleic acid, but which facilitates cloning
or detection of the amplified product. The region of the primer
which is sufficiently complementary to the template to hybridize is
referred to herein as the hybridizing region.
[0093] The terms "target, "target sequence", "target region", and
"target nucleic acid," as used herein, are synonymous and refer to
a region or sequence of a nucleic acid which is to be amplified,
sequenced or detected.
[0094] The term "hybridization," as used herein, refers to the
formation of a duplex structure by two single-stranded nucleic
acids due to complementary base pairing. Hybridization can occur
between fully complementary nucleic acid strands or between
"substantially complementary" nucleic acid strands that contain
minor regions of mismatch. Conditions under which hybridization of
fully complementary nucleic acid strands is strongly preferred are
referred to as "stringent hybridization conditions" or
"sequence-specific hybridization conditions". Stable duplexes of
substantially complementary sequences can be achieved under less
stringent hybridization conditions; the degree of mismatch
tolerated can be controlled by suitable adjustment of the
hybridization conditions. Those skilled in the art of nucleic acid
technology can determine duplex stability empirically considering a
number of variables including, for example, the length and base
pair composition of the oligonucleotides, ionic strength, and
incidence of mismatched base pairs, following the guidance provided
by the art (see, e.g., Sambrook et al., 1989, Molecular Cloning-A
Laboratory Manual, Cold Spring Harbor Laboratory, Cold Spring
Harbor, N.Y.; Wetmur, 1991, Critical Review in Biochem. and Mol.
Biol. 26(3/4):227-259; and Owczarzy et al., 2008, Biochemistry, 47:
5336-5353, which are incorporated herein by reference).
[0095] The term "amplification reaction" refers to any chemical
reaction, including an enzymatic reaction, which results in
increased copies of a template nucleic acid sequence or results in
transcription of a template nucleic acid. Amplification reactions
include reverse transcription, the polymerase chain reaction (PCR),
including Real Time PCR (see U.S. Pat. Nos. 4,683,195 and
4,683,202; PCR Protocols: A Guide to Methods and Applications
(Innis et al., eds, 1990)), and the ligase chain reaction (LCR)
(see Barany et al., U.S. Pat. No. 5,494,810). Exemplary
"amplification reactions conditions" or "amplification conditions"
typically comprise either two or three step cycles. Two step cycles
have a high temperature denaturation step followed by a
hybridization/elongation (or ligation) step. Three step cycles
comprise a denaturation step followed by a hybridization step
followed by a separate elongation or ligation step.
[0096] As used herein, a "polymerase" refers to an enzyme that
catalyzes the polymerization of nucleotides. Generally, the enzyme
will initiate synthesis at the 3'-end of the primer annealed to a
nucleic acid template sequence. "DNA polymerase" catalyzes the
polymerization of deoxyribonucleotides. Known DNA polymerases
include, for example, Pyrococcus furiosus (Pfu) DNA polymerase
(Lundberg et al., 1991, Gene, 108:1), E. coli DNA polymerase I
(Lecomte and Doubleday, 1983, Nucleic Acids Res. 11:7505), T7 DNA
polymerase (Nordstrom et al., 1981, J. Biol. Chem. 256:3112),
Thermus thermophilus (Tth) DNA polymerase (Myers and Gelfand 1991,
Biochemistry 30:7661), Bacillus stearothermophilus DNA polymerase
(Stenesh and McGowan, 1977, Biochim Biophys Acta 475:32),
Thermococcus litoralis (Tli) DNA polymerase (also referred to as
Vent DNA polymerase, Cariello et al., 1991, Nucleic Acids Res, 19:
4193), Thermotoga maritima(Tma) DNA polymerase (Diaz and Sabino,
1998 Braz J. Med. Res, 31:1239), Thermus aquaticus (Taq) DNA
polymerase (Chien et al., 1976, J. Bacteoriol, 127: 1550),
Pyrococcus kodakaraensis KOD DNA polymerase (Takagi et al., 1997,
Appl. Environ. Microbiol. 63:4504), JDF-3 DNA polymerase (Patent
application WO 0132887), and Pyrococcus GB-D (PGB-D) DNA polymerase
(Juncosa-Ginesta et al., 1994, Biotechniques, 16:820). The
polymerase activity of any of the above enzymes can be determined
by means well known in the art.
[0097] As used herein, a primer is "specific," for a target
sequence if, when used in an amplification reaction under
sufficiently stringent conditions, the primer hybridizes primarily
to the target nucleic acid. Typically, a primer is specific for a
target sequence if the primer-target duplex stability is greater
than the stability of a duplex formed between the primer and any
other sequence found in the sample. One of skill in the art will
recognize that various factors, such as salt conditions as well as
base composition of the primer and the location of the mismatches,
will affect the specificity of the primer, and that routine
experimental confirmation of the primer specificity will be needed
in many cases. Hybridization conditions can be chosen under which
the primer can form stable duplexes only with a target sequence.
Thus, the use of target-specific primers under suitably stringent
amplification conditions enables the selective amplification of
those target sequences which contain the target primer binding
sites.
[0098] The term "non-specific amplification," as used herein,
refers to the amplification of nucleic acid sequences other than
the target sequence which results from primers hybridizing to
sequences other than the target sequence and then serving as a
substrate for primer extension. The hybridization of a primer to a
non-target sequence is referred to as "non-specific hybridization"
and is apt to occur especially during the lower temperature,
reduced stringency, pre-amplification conditions, or in situations
where there is a variant allele in the sample having a very closely
related sequence to the true target as in the case of a single
nucleotide polymorphism (SNP).
[0099] The term "primer dimer," as used herein, refers to a
template-independent non-specific amplification product, which is
believed to result from primer extensions wherein another primer
serves as a template. Although primer dimers frequently appear to
be a concatamer of two primers, i.e., a dimer, concatamers of more
than two primers also occur. The term "primer dimer" is used herein
generically to encompass a template-independent non-specific
amplification product.
[0100] The term "reaction mixture," as used herein, refers to a
solution containing reagents necessary to carry out a given
reaction. An "amplification reaction mixture", which refers to a
solution containing reagents necessary to carry out an
amplification reaction, typically contains oligonucleotide primers
and a DNA polymerase or ligase in a suitable buffer. A "PCR
reaction mixture" typically contains oligonucleotide primers, a DNA
polymerase (most typically a thermostable DNA polymerase), dNTP's,
and a divalent metal cation in a suitable buffer. A reaction
mixture is referred to as complete if it contains all reagents
necessary to enable the reaction, and incomplete if it contains
only a subset of the necessary reagents. It will be understood by
one of skill in the art that reaction components are routinely
stored as separate solutions, each containing a subset of the total
components, for reasons of convenience, storage stability, or to
allow for application-dependent adjustment of the component
concentrations, and that reaction components are combined prior to
the reaction to create a complete reaction mixture. Furthermore, it
will be understood by one of skill in the art that reaction
components are packaged separately for commercialization and that
useful commercial kits may contain any subset of the reaction
components which includes the blocked primers of the invention.
[0101] For the purposes of this invention, the terms
"non-activated" or "inactivated," as used herein, refer to a primer
or other oligonucleotide that is incapable of participating in a
primer extension reaction or a ligation reaction because either DNA
polymerase or DNA ligase cannot interact with the oligonucleotide
for their intended purposes. In some embodiments when the
oligonucleotide is a primer, the non-activated state occurs because
the primer is blocked at or near the 3'-end so as to prevent primer
extension. When specific groups are bound at or near the 3'-end of
the primer, DNA polymerase cannot bind to the primer and extension
cannot occur. A non-activated primer is, however, capable of
hybridizing to a substantially complementary nucleotide
sequence.
[0102] For the purposes of this invention, the term "activated," as
used herein, refers to a primer or other oligonucleotide that is
capable of participating in a reaction with DNA polymerase or DNA
ligase. A primer or other oligonucleotide becomes activated after
it hybridizes to a substantially complementary nucleic acid
sequence and is cleaved to generate a functional 3'- or 5'-end so
that it can interact with a DNA polymerase or a DNA ligase. For
example, when the oligonucleotide is a primer, and the primer is
hybridized to a template, a 3'-blocking group can be removed from
the primer by, for example, a cleaving enzyme such that DNA
polymerase can bind to the 3' end of the primer and promote primer
extension.
[0103] The term "cleavage domain" or "cleaving domain," as used
herein, are synonymous and refer to a region located between the 5'
and 3' end of a primer or other oligonucleotide that is recognized
by a cleavage compound, for example a cleavage enzyme, that will
cleave the primer or other oligonucleotide. For the purposes of
this invention, the cleavage domain is designed such that the
primer or other oligonucleotide is cleaved only when it is
hybridized to a complementary nucleic acid sequence, but will not
be cleaved when it is single-stranded. The cleavage domain or
sequences flanking it may include a moiety that a) prevents or
inhibits the extension or ligation of a primer or other
oligonucleotide by a polymerase or a ligase, b) enhances
discrimination to detect variant alleles, or c) suppresses
undesired cleavage reactions. One or more such moieties may be
included in the cleavage domain or the sequences flanking it.
[0104] The term "RNase H cleavage domain," as used herein, is a
type of cleavage domain that contains one or more ribonucleic acid
residue or an alternative analog which provides a substrate for an
RNase H. An RNase H cleavage domain can be located anywhere within
a primer or oligonucleotide, and is preferably located at or near
the 3'-end or the 5'-end of the molecule.
[0105] An "RNase H1 cleavage domain" generally contains at least
three residues. An "RNase H2 cleavage domain" may contain one RNA
residue, a sequence of contiguously linked RNA residues or RNA
residues separated by DNA residues or other chemical groups. In one
embodiment, the RNase H2 cleavage domain is a 2'-fluoronucleoside
residue. In a more preferred embodiment the RNase H2 cleavable
domain is two adjacent 2'-fluoro residues.
[0106] The terms "cleavage compound," or "cleaving agent" as used
herein, refers to any compound that can recognize a cleavage domain
within a primer or other oligonucleotide, and selectively cleave
the oligonucleotide based on the presence of the cleavage domain.
The cleavage compounds utilized in the invention selectively cleave
the primer or other oligonucleotide comprising the cleavage domain
only when it is hybridized to a substantially complementary nucleic
acid sequence, but will not cleave the primer or other
oligonucleotide when it is single stranded. The cleavage compound
cleaves the primer or other oligonucleotide within or adjacent to
the cleavage domain. The term "adjacent," as used herein, means
that the cleavage compound cleaves the primer or other
oligonucleotide at either the 5'-end or the 3' end of the cleavage
domain. Cleavage reactions preferred in the invention yield a
5'-phosphate group and a 3'-OH group.
[0107] In a preferred embodiment, the cleavage compound is a
"cleaving enzyme." A cleaving enzyme is a protein or a ribozyme
that is capable of recognizing the cleaving domain when a primer or
other nucleotide is hybridized to a substantially complementary
nucleic acid sequence, but that will not cleave the complementary
nucleic acid sequence (i.e., it provides a single strand break in
the duplex). The cleaving enzyme will also not cleave the primer or
other oligonucleotide comprising the cleavage domain when it is
single stranded. Examples of cleaving enzymes are RNase H enzymes
and other nicking enzymes.
[0108] The term "nicking," as used herein, refers to the cleavage
of only one strand of the double-stranded portion of a fully or
partially double-stranded nucleic acid. The position where the
nucleic acid is nicked is referred to as the "nicking site" (NS). A
"nicking agent" (NA) is an agent that nicks a partially or fully
double-stranded nucleic acid. It may be an enzyme or any other
chemical compound or composition. In certain embodiments, a nicking
agent may recognize a particular nucleotide sequence of a fully or
partially double-stranded nucleic acid and cleave only one strand
of the fully or partially double-stranded nucleic acid at a
specific position (i.e., the NS) relative to the location of the
recognition sequence. Such nicking agents (referred to as "sequence
specific nicking agents") include, but are not limited to, nicking
endonucleases (e.g., N.BstNB).
[0109] A "nicking endonuclease" (NE), as used herein, thus refers
to an endonuclease that recognizes a nucleotide sequence of a
completely or partially double-stranded nucleic acid molecule and
cleaves only one strand of the nucleic acid molecule at a specific
location relative to the recognition sequence. In such a case the
entire sequence from the recognition site to the point of cleavage
constitutes the "cleavage domain".
[0110] The term "blocking group," as used herein, refers to a
chemical moiety that is bound to the primer or other
oligonucleotide such that an amplification reaction does not occur.
For example, primer extension and/or DNA ligation does not occur.
Once the blocking group is removed from the primer or other
oligonucleotide, the oligonucleotide is capable of participating in
the assay for which it was designed (PCR, ligation, sequencing,
etc). Thus, the "blocking group" can be any chemical moiety that
inhibits recognition by a polymerase or DNA ligase. The blocking
group may be incorporated into the cleavage domain but is generally
located on either the 5'- or 3'-side of the cleavage domain. The
blocking group can be comprised of more than one chemical moiety.
In the present invention the "blocking group" is typically removed
after hybridization of the oligonucleotide to its target
sequence.
[0111] The term "fluorescent generation probe" refers either to a)
an oligonucleotide having an attached fluorophore and quencher, and
optionally a minor groove binder or to b) a DNA binding reagent
such as SYBR.RTM. Green dye.
[0112] The terms "fluorescent label" or "fluorophore" refers to
compounds with a fluorescent emission maximum between about 350 and
900 nm. A wide variety of fluorophores can be used, including but
not limited to: 5-FAM (also called 5-carboxyfluorescein; also
called Spiro(isobenzofuran-1(3H), 9'-(9H)xanthene)-5-carboxylic
acid, 3',6'-dihydroxy-3-oxo-6-carboxyfluorescein);
5-Hexachloro-Fluorescein;
([4,7,2',4',5',7'-hexachloro-(3',6'-dipivaloyl-fluoresceinyl)-6-carboxyli-
c acid]); 6-Hexachloro-Fluorescein;
([4,7,2',4',5',7'-hexachloro-(3',6'-dipivaloylfluoresceinyl)-5-carboxylic
acid]); 5-Tetrachloro-Fluorescein;
([4,7,2',7'-tetra-chloro-(3',6'-dipivaloylfluoresceinyl)-5-carboxylic
acid]); 6-Tetrachloro-Fluorescein;
([4,7,2',7'-tetrachloro-(3',6'-dipivaloylfluoresceinyl)-6-carboxylic
acid]); 5-TAMRA (5-carboxytetramethylrhodamine); Xanthylium,
9-(2,4-dicarboxyphenyl)-3,6-bis(dimethyl-amino); 6-TAMRA
(6-carboxytetramethylrhodamine);
9-(2,5-dicarboxyphenyl)-3,6-bis(dimethylamino); EDANS
(5-((2-aminoethyl)amino)naphthalene-1-sulfonic acid); 1,5-IAEDANS
(5-((((2-iodoacetyl)amino)ethyl)amino)naphthalene-1-sulfonic acid);
Cy5 (Indodicarbocyanine-5); Cy3 (Indo-dicarbocyanine-3); and BODIPY
FL
(2,6-dibromo-4,4-difluoro-5,7-dimethyl-4-bora-3a,4a-diaza-s-indacene-3-pr-
oprionic acid); Quasar.RTM.-670 dye (Biosearch Technologies); Cal
Fluor.RTM. Orange dye (Biosearch Technologies); Rox dyes; Max dyes
(Integrated DNA Technologies), as well as suitable derivatives
thereof.
[0113] As used herein, the term "quencher" refers to a molecule or
part of a compound, which is capable of reducing the emission from
a fluorescent donor when attached to or in proximity to the donor.
Quenching may occur by any of several mechanisms including
fluorescence resonance energy transfer, photo-induced electron
transfer, paramagnetic enhancement of intersystem crossing, Dexter
exchange coupling, and exciton coupling such as the formation of
dark complexes. Fluorescence is "quenched" when the fluorescence
emitted by the fluorophore is reduced as compared with the
fluorescence in the absence of the quencher by at least 10%, for
example, 15%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, 95%, 98%,
99%, 99.9% or more. A number of commercially available quenchers
are known in the art, and include but are not limited to DABCYL,
Black Hole.TM. Quenchers (BHQ-1, BHQ-2, and BHQ-3), Iowa Black.RTM.
FQ and Iowa Black.RTM. RQ. These are so-called dark quenchers. They
have no native fluorescence, virtually eliminating background
problems seen with other quenchers such as TAMRA which is
intrinsically fluorescent.
[0114] The term "ligation" as used herein refers to the covalent
joining of two polynucleotide ends. In various embodiments,
ligation involves the covalent joining of a 3' end of a first
polynucleotide (the acceptor) to a 5' end of a second
polynucleotide (the donor). Ligation results in a phosphodiester
bond being formed between the polynucleotide ends. In various
embodiments, ligation may be mediated by any enzyme, chemical, or
process that results in a covalent joining of the polynucleotide
ends. In certain embodiments, ligation is mediated by a ligase
enzyme.
[0115] As used herein, "ligase" refers to an enzyme that is capable
of covalently linking the 3' hydroxyl group of one polynucleotide
to the 5' phosphate group of a second polynucleotide. Examples of
ligases include E. coli DNA ligase, T4 DNA ligase, etc.
[0116] The ligation reaction can be employed in DNA amplification
methods such as the "ligase chain reaction" (LCR), also referred to
as the "ligase amplification reaction" (LAR), see Barany, Proc.
Natl. Acad. Sci., 88:189 (1991); and Wu and Wallace, Genomics 4:560
(1989) incorporated herein by reference. In LCR, four
oligonucleotides, two adjacent oligonucleotides which uniquely
hybridize to one strand of the target DNA, and a complementary set
of adjacent oligonucleotides, that hybridize to the opposite strand
are mixed and DNA ligase is added to the mixture. In the presence
of the target sequence, DNA ligase will covalently link each set of
hybridized molecules. Importantly, in LCR, two oligonucleotides are
ligated together only when they base-pair with sequences without
gaps. Repeated cycles of denaturation, hybridization and ligation
amplify a short segment of DNA. A mismatch at the junction between
adjacent oligonucleotides inhibits ligation. As in other
oligonucleotide ligation assays this property allows LCR to be used
to distinguish between variant alleles such as SNPs. LCR has also
been used in combination with PCR to achieve enhanced detection of
single-base changes, see Segev, PCT Public. No. WO9001069
(1990).
Novel Oligonucleotides and Compounds of the Present Invention.
[0117] In one embodiment, the novel oligonucleotides of the present
invention are primers for DNA replication, as for example in PCR,
DNA sequencing and polynomial amplification, to name a few such
applications. In this embodiment, the primers have an inactive
configuration wherein DNA replication (i.e., primer extension) is
blocked, and an activated configuration wherein DNA replication
proceeds. The inactive configuration of the primer is present when
the primer is either single-stranded, or the primer is hybridized
to the DNA sequence of interest and primer extension remains
blocked by a chemical moiety that is linked at or near to the 3'
end of the primer. The activated configuration of the primer is
present when the primer is hybridized to a nucleic acid sequence of
interest and subsequently acted upon by RNase H or other cleaving
agent to remove the blocking group and allow for an enzyme (e.g., a
DNA polymerase) to catalyze primer extension.
[0118] A number of blocking groups are known in the art that can be
placed at or near the 3' end of the oligonucleotide (e.g., a
primer) to prevent extension. A primer or other oligonucleotide may
be modified at the 3'-terminal nucleotide to prevent or inhibit
initiation of DNA synthesis by, for example, the addition of a 3'
deoxyribonucleotide residue (e.g., cordycepin), a
2',3'-dideoxyribonucleotide residue, non-nucleotide linkages or
alkane-diol modifications (U.S. Pat. No. 5,554,516). Alkane diol
modifications which can be used to inhibit or block primer
extension have also been described by Wilk et al., (1990, Nucleic
Acids Res., 18 (8):2065), and by Arnold et al., (U.S. Pat. No.
6,031,091). Additional examples of suitable blocking groups include
3' hydroxyl substitutions (e.g., 3'-phosphate, 3'-triphosphate or
3'-phosphate diesters with alcohols such as 3-hydroxypropyl), a
2'3'-cyclic phosphate, 2' hydroxyl substitutions of a terminal RNA
base (e.g., phosphate or sterically bulky groups such as
triisopropyl silyl (TIPS) or tert-butyl dimethyl silyl (TBDMS)).
2'-alkyl silyl groups such as TIPS and TBDMS substituted at the
3'-end of an oligonucleotide are described by Laikhter et al., U.S.
patent application Ser. No. 11/686,894 which is incorporated herein
by reference. Bulky substituents can also be incorporated on the
base of the 3'-terminal residue of the oligonucleotide to block
primer extension.
[0119] Blocking groups to inhibit primer extension can also be
located upstream, that is 5', from the 3'-terminal residue.
Sterically bulky substituents which interfere with binding by the
polymerase can be incorporated onto the base, sugar or phosphate
group of residues upstream from the 3'-terminus. Such substituents
include bulky alkyl groups like t-butyl, triisopropyl and
polyaromatic compounds including fluorophores and quenchers, and
can be placed from one to about 10 residues from the 3'-terminus.
Alternatively abasic residues such as a C3 spacer may be
incorporated in these locations to block primer extension. In one
such embodiment two adjacent C3 spacers have been employed (see
Examples 27 and 28).
[0120] In the case of PCR, blocking moieties upstream of the
3'-terminal residue can serve two functions: 1) to inhibit primer
extension, and 2) to block the primer from serving as a template
for DNA synthesis when the extension product is copied by synthesis
from the reverse primer. The latter is sufficient to block PCR even
if primer extension can occur. C3 spacers placed upstream of the
3'-terminal residue can function in this manner (see Examples 26
and 27).
[0121] A modification used as a blocking group may also be located
within a region 3' to the priming sequence that is
non-complementary to the target nucleic acid sequence.
[0122] The oligonucleotide further comprises a cleavage domain
located upstream of the blocking group used to inhibit primer
extension. An RNase H cleavage domain is preferred. An RNase H2
cleavage domain comprising a single RNA residue or replacement of
the RNA base with one or more alternative nucleosides is most
preferred.
[0123] In one embodiment, RNase H2 can be used to cleave duplexes
containing a single 2'-fluoro residue. Cleavage occurs on the 5'
side of the 2'-fluoro residue. In a preferred embodiment, an RNase
H2 cleavage domain comprising two adjacent 2'-fluoro residues is
employed (see Example 6). The activity is enhanced when two
consecutive 2'-fluoro modifications are present. In this embodiment
cleavage occurs preferentially between the 2'-fluoro residues.
Unlike oligonucleotides containing unmodified RNA residues,
oligonucleotides with 2'-fluoro groups are not cleaved by
single-stranded ribonucleases and are resistant to water catalyzed
cleavage and completely stable at high temperatures. Enhanced
cleavage has also been found when a 2'-fluoro modified RNA residue
is used with a 2' LNA modified RNA residue. 2'-fluoro-containing
oligonucleotides have been found to be further advantageous in
certain applications compared to RNA-containing oligonucleotides in
offering greater discrimination with respect to mismatches between
the oligonucleotide and the target sequence.
[0124] Alternatives to an RNA residue that can be used in the
present invention wherein cleavage is mediated by an RNase H enzyme
include but are not limited to 2'-O-alkyl RNA nucleosides,
preferably 2'-O-methyl RNA nucleosides, 2'-fluoronucleosides,
locked nucleic acids (LNA), 2'-ENA residues (ethylene nucleic
acids), 2'-alkyl nucleosides, 2'-aminonucleosides and
2'-thionucleosides. The RNase H cleavage domain may include one or
more of these modified residues alone or in combination with RNA
bases. DNA bases and abasic residues such as a C3 spacer may also
be included to provide greater performance.
[0125] If the cleaving agent is an RNase H1 enzyme a continuouse
sequence of at least three RNA residues is preferred. A continuous
sequence of four RNA residues generally leads to maximal activity.
If the cleaving agent is an RNase H2 enzyme a single RNA residue or
2 adjacent 2'-fluoro residues are preferred.
[0126] One objective of incorporating modified residues within an
RNase H cleavage domain is to suppress background cleavage of a
primer or probe due to water catalyzed hydrolysis or cleavage by
single stranded ribonucleases. Replacement of the 2'-hydroxyl group
with a substituent that cannot attack the adjacent phosphate group
of an RNA residue can accomplish this goal. Examples of this
approach include the use of the 2'-substituted nucleosides listed
above, such as 2'-fluoro and 2'-O-methyl nucleosides. This is
particularly advantageous when cleavage is mediated by RNase H2 and
there is a single RNA residue within the cleavage domain. As shown
in FIG. 3, in this case cleavage by single stranded ribonucleases
or water catalyzed hydrolysis occurs at a different position than
cleavage by RNase H2.
[0127] Other examples of modifications that can be used to suppress
cleavage by single stranded ribonucleases and water catalyzed
hydrolysis at RNA residues include substitution of the 5' oxygen
atom of the adjacent residue (3'- to the RNA base) with an amino
group, thiol group, or a methylene group (a phosphonate linkage).
Alternatively one or both of the hydrogen atoms on the 5' carbon of
the adjacent residue can be replaced with bulkier substituents such
as methyl groups to inhibit background cleavage of a ribonucleotide
residue. In another such embodiment, the phosphate group at the
3'-side of an RNA residue can be replaced with a phosphorothioate,
phosphorodithioates or boronate linkage. In the case of a
phosphorothioate the S stereoisomer is preferred. Combinations of
these various modifications may also be employed.
[0128] It should be noted that background cleavage at RNA residues
by single stranded ribonucleases or water catalyzed hydrolysis
leads to a blocked 3'-end (see FIG. 3) that cannot serve as a
primer for DNA synthesis. This mitigates the occurrence of false
positive results even if such cleavage does occur.
[0129] The cleavage domain may include the blocking group provided
that cleavage occurs on the 5'-side of the blocking group and
generates a free 3'-OH. Generally however the cleavage domain and
the blocking group are separated by one to about 15 bases. After
cleavage takes place the portion of the primer 3' from the cleavage
site containing the blocking group dissociates from the template
and a functional 3'-hydroxyl group is exposed, capable of being
acted on by a polymerase enzyme. The optimal distance between the
cleavage site and the blocking group will depend on the cleaving
agent and the nature of the blocking group. When cleavage of the
oligonucleotide is mediated by RNase H2 at a single RNA residue a
distance of 3 to about 8 bases between the cleavage site and the
blocking group is preferred. If the blocking group is sterically
small, for example a phosphodiester at the 3' terminal nucleotide
as in the following structure
##STR00002##
a cleavage site 5 bases from the 3'-end is generally optimal. If
the blocking group is larger it is advantageous to position the
cleavage site further from it.
[0130] In a preferred embodiment, a thermophilic RNase H2 enzyme is
utilized to cleave the oligonucleotide. In yet a more preferred
embodiment, a thermophilic RNase H2 enzyme is used which is less
active at room temperature than at elevated temperatures. This
allows a hot-start type of reaction to be achieved in PCR and other
primer extension assays using the blocked primers of the present
invention without actually requiring a hot start, i.e., reversibly
inactivated, DNA polymerase. Standard less expensive DNA polymerase
polymerases such as Taq polymerase can be used instead of the much
more expensive hot start versions of the enzyme. Moreover, for
different applications alternative DNA polymerases may be
preferred. Utilizing RNase H as the hot start component of the
assay obviates the need to develop a new reversibly inactivated
analog of each different DNA polymerase.
[0131] Hot start properties of the enzyme may be intrinsic to the
protein as in the case of Pyrococcus abysii RNase H2 (see Example
4). Alternatively the enzyme may be reversibly inactivated by
chemical modification using, for example, maleic acid anhydride
analogs such as citroconic anhydride. These compounds react with
amino groups of the protein and at high temperature are released
restoring activity. In yet another embodiment antibodies against an
RNase H which block the enzyme may be employed which are denatured
at elevated temperatures.
[0132] In yet another embodiment, the oligonucleotide of the
present invention has a cleavage domain that is recognized and
cleaved by a sequence specific nicking agent, e.g., a nicking
enzyme. The nicking agent also can be designed to cleave an
oligonucleotide (e.g., a primer) at a modified nucleic acid or
grouping of modified nucleic acids. In this embodiment, the
oligonucleotide is designed to be recognized by a nicking agent
upon hybridization with the target nucleic acid, and the nicking of
the oligonucleotide/target duplex can be used to remove a blocking
group and allow for oligonucleotide extension. The nicking site
(NS) is preferably located at or near the 3'-end of the
oligonucleotide, specifically, one to about 15 bases from the
3'-end of the oligonucleotide.
[0133] Exemplary nicking agents include, without limitation, single
strand nicking restriction endonucleases that recognize a specific
sequence such as N.BstNBI; or repair enzymes such as Mut H, MutY
(in combination with an AP endonuclease), or uracil-N-glycosylase
(in combination with an AP Lyase and AP endonucleases); and the
geneII protein of bacteriophage fl.
[0134] The blocked primers of the present invention minimize
non-specific reactions by requiring hybridization to the target
followed by cleavage before primer extension. If a primer
hybridizes incorrectly to a related sequence, cleavage of the
primer is inhibited especially when there is a mismatch that lies
at or near the cleavage site. This reduces the frequency of false
priming at such locations and thereby increases the specificity of
the reaction. It should be noted that with Pyrococcus abysii Type
II RNase H and other RNase H enzymes used in the present invention
some cleavage does occur even when there is a mismatch at the
cleavage site. Reaction conditions, particularly the concentration
of RNase H and the time allowed for hybridization and extension in
each cycle, can be optimized to maximize the difference in cleavage
efficiencies between the primer hybridized to its true target and
when there is a mismatch. This allows the methods of the present
invention to be used very effectively to distinguish between
variant alleles, including SNPs (see Examples 12-14, 22-25).
[0135] As noted above, background cleavage of the primer does not
lead to false-positive priming when RNA residues are incorporated
into the oligonucleotide, because the 2',3'-cyclic phosphate (or 2'
or 3'-phosphate) formed at the 3' end of the cleaved primer blocks
primer extension. A freely accessible 3' OH group is needed to form
a substrate for DNA polymerase. The formation of primer-dimers, a
common side reaction occurring in PCR, can also be inhibited using
the 3' blocked primers of the present invention. This allows for a
greater degree of multiplexing in PCR (e.g., detecting multiple
target sequences in the case of a DNA detection/amplification
assay).
[0136] Without being bound by any theory, it has been observed that
atypical cleavage can occur at a low frequency 3' to an RNA residue
when there is a mismatch, presumably catalyzed by RNase H2, to
generate a free 3'-OH and lead to primer extension. This can result
in a decrease in the specificity of the reaction. To mitigate this
effect nuclease resistant residues can be incorporated into the
primer 3' to the RNA residue (see Example 22, 25 and 28). Such
groups include but are not limited to one or more
phosphorothioates, phosphorodithioates, methyl phosphonates and
abasic residues such as a C3 spacer.
[0137] Other substitutions both 5' and 3' to the RNA residue can
also be utilized to enhance the discrimination and detection of
variant alleles in the methods of the present invention. Such
substitutions include but are not limited to 2'-O-methyl RNA and
secondary mismatches (see Example 23).
[0138] The nature of the blocking group which prevents primer
extension is not critical. It can be placed at the 3'-terminal
residue or upstream from it. Labeling groups can be incorporated
within the blocking group or attached at other positions on the
3'-segment of the oligonucleotide primer which dissociates from the
template after cleavage occurs. Such labeling groups include, but
are not limited to, fluorophores, quenchers, biotin, haptens such
as digoxigenin, proteins including enzymes and antibodies, mass
tags which alter the mass of the cleavage fragment for detection by
mass spectrometry, and radiolabels such as .sup.14C, .sup.3H,
.sup.35S, .sup.32P and .sup.33P. These labeling groups can also be
attached to the primer 5' to the cleavage site, in which case they
will be incorporated within the extension product.
[0139] In one embodiment, the blocking group at or near the 3'-end
of the oligonucleotide can be a fluorescent moiety. In this case,
release of the fluorescent molecule can be used to monitor the
progress of the primer extension reaction. This is facilitated if
the oligonucleotide also contains a quencher moiety on the 5'-side
of the cleavage site. Cleavage of the oligonucleotide during the
reaction separates the fluorophore from the quencher and leads to
an increase in fluorescence. If the quencher is itself a
fluorophore, such as Tamra, a decrease in its fluorescence may also
be observed.
[0140] In yet a further embodiment, the oligonucleotide is labeled
with a fluorescent molecule on the 5'-side of the cleavage domain,
and the blocking group located at or near the 3'-end of the
molecule is a quencher such as Iowa Black.RTM., Black Hole.TM., or
Tamra to name a few. Again, cleavage of the quencher from the
oligonucleotide (e.g., a primer) leads to an increase in
fluorescence which can be used to monitor the progress of the
oligonucleotide extension reaction. Moreover, in this case, the
primer extension product is fluorescently labeled.
[0141] In yet a further embodiment, the blocked primers of the
present invention are used for nucleic acid sequencing. As in the
case of DNA amplification reactions, the specificity of primer
extension for DNA sequencing is also increased when using the
oligonucleotides of the present invention. In one sequencing
embodiment, 2',3' dideoxynucleotide triphosphates that are
fluorescently labeled and used as chain terminators and the nested
fragments produced in the reaction are separated by
electrophoresis, preferably capillary electrophoresis.
[0142] In yet another embodiment, an oligonucleotide primer of the
present invention is labeled with a fluorescent group and the 3'
dideoxynucleotide triphosphate chain terminators are unlabeled. In
this embodiment, the blocking group can be a quencher, in which
case background fluorescence is reduced because the primer itself
is not fluorescent. Only the extension products are
fluorescent.
[0143] Another aspect of the invention includes the incorporation
of alternative divalent cations such as Mn.sup.2+, Ni.sup.2+ or
Co.sup.2+, with or without Mg.sup.2+, into the assay buffer. In
certain embodiments of the invention, when such alternative
divalent cations are present, the effectiveness of the particular
assay is increased due to enhanced cleavage by RNase H2. In one
embodiment, when two adjacent 2'-fluoronucleoside residues
constitute the RNase H2 cleavable domain, 0.3-1 mM MnCl.sub.2 with
2-4 mM MgCl.sub.2 gave optimal performance in the assay (see
Example 3).
The Methods of the Present Invention
[0144] The primers, probes and other novel oligonucleotides
described herein can be utilized in a number of biological assays.
Although the following list is not comprehensive, the majority of
the methods of the present invention fall into six general
categories: (1) primer extension assays (including PCR, DNA
sequencing and polynomial amplification), (2) oligonucleotide
ligation assays (OLA), (3) cycling probe reactions, (4) sequencing
by ligation, (5) sequencing by generation of end-labeled fragments
using RNase H enzymes, and (6) synthesis by ligation.
[0145] The primers, probes and other novel oligonucleotides
described herein can be utilized in a number of primer extension
assays.
Primer Extension Assays
[0146] In one embodiment of the present invention, a method of
amplifying a target DNA sequence of interest is provided. The
method comprises the steps of:
(a) providing a reaction mixture comprising a primer having a
cleavage domain and a blocking group linked at or near to the 3'
end of the primer which prevents primer extension, a sample nucleic
acid having the target DNA sequence of interest, a cleaving enzyme
and a polymerase; (b) hybridizing the primer to the target DNA
sequence to form a double-stranded substrate; (c) cleaving the
hybridized primer with the cleaving enzyme at a point within or
adjacent to the cleavage domain to remove the blocking group from
the primer; and (d) extending the primer with the polymerase.
PCR in General
[0147] When used in PCR, a 3'-blocked primer containing a cleavage
domain first hybridizes to the target sequence. In this embodiment,
the primer cannot extend until cleavage of the 3' blocking group
occurs after hybridization to the complementary DNA sequence. For
example, when an RNase H cleavage domain is present in the primer,
an RNase H enzyme will recognize the double-stranded substrate
formed by the primer and target and cleave the primer within or
adjacent to the cleavage domain. The primer can then extend and
amplification of the target can then occur. Because the primer
needs to be recognized and cleaved by RNase H before extension,
non-specific amplification is reduced.
[0148] In conventional PCR, a "hot start" polymerase is often used
to reduce primer dimers and decrease non-specific amplification.
Blocked primers of the present invention requiring cleavage by
RNase H can confer the same advantage. A thermophilic RNase H
enzyme with little or no activity at lower temperatures is
preferred. Activation of the primers occurs only after
hybridization to the target sequence and cleavage at elevated
temperatures. Advantages of this approach compared to the use of a
hot start reversibly inactivated DNA polymerase have been described
above. Of course a hot start RNase H enzyme and a hot start DNA
polymerase can be used in conjunction, if desired.
[0149] Three types of hot start RNase H enzymes are described here
(see Tables 1, 2, and 3): 1) a thermostable RNase H enzyme that has
intrinsically little or no activity at reduced temperatures as in
the case of Pyrococcus abysii RNase H2; 2) a thermostable RNase H
reversibly inactivated by chemical modification; and 3) a
thermostable RNase H reversibly inactivated by a blocking antibody.
In addition, through means well-known in the art, such as random
mutagenesis, mutant versions of RNase H can be synthesized that can
further improve the traits of RNase H that are desirable in the
assays of the present invention. Alternatively, mutant strains of
other enzymes that share the characteristics desirable for the
present invention could be used.
[0150] In one embodiment, the cleavage domain within the primer is
cleavable by RNase H. In yet a further embodiment, the RNase H
cleavage domain consists of a single RNA residue and cleavage of
the primer is mediated by a Type II RNase H enzyme, preferably by a
thermophilic Type II RNase H enzyme, and even more preferably a
thermophilic Type II RNase H enzyme which is less active at room
temperature than at elevated temperatures. In yet a further
embodiment, the RNase H2 cleavage domain consists of two adjacent
2'-fluoro nucleoside residues. In yet a more preferred embodiment
of the present invention in which the cleavage domain consists of
two adjacent 2'-fluoro nucleoside residues, the PCR is carried out
in buffers containing alternative divalent cations, including but
not limited to, Mn.sup.2+, Ni.sup.2+ or Co.sup.2+ in addition to
Mg.sup.2+. In an additional embodiment, the novel 3'-blocked
primers of the present invention comprising a cleavage domain can
be utilized in a variation of hot start PCR in which a thermophilic
nicking enzyme is used and the cleavage domain is a nicking
site.
[0151] Alternatively, a cleavage enzyme that lacks hot start
characteristics can be used in the present invention with
traditional hot-start methods such as adding the enzyme at an
elevated temperature, encasing a necessary reagent or enzyme in
wax, or with a hot start reversibly inactivated DNA polymerase.
[0152] The increased specificity of the present invention, when
used in amplification reactions, enables real-time PCR applications
to achieve more specific results, as compared to conventional
real-time PCR with standard DNA primers. For example,
double-stranded DNA-binding dye assays, such as SYBR.RTM. Green
assays, have a disadvantage in that a signal is produced once the
dye binds to any double-stranded product produced by PCR (e.g., a
primer dimer) and can thereby give rise to a false positive result.
But when a primer of the current invention is used, non-specific
amplification and primer-dimer formation is reduced, and the
intensity of the signal of the double-stranded DNA-binding dye will
reflect amplification only of the desired target (see Example
17).
[0153] The reagent concentrations and reaction conditions of the
assay can be varied to maximize its utility. The relative
efficiency of PCR using the blocked primers of the present
invention relates to the concentration of the unblocking enzyme and
the dwell time at the anneal/extend reaction temperature (where
unblocking proceeds). With low amounts of enzyme and short dwell
times, cleavage can be incomplete and the reactions with blocked
primers have lower efficiency than those with unblocked primers. As
either enzyme concentration or dwell time increases, the reaction
efficiency with blocked primers increases and becomes identical to
unblocked primers. The use of even more enzyme or longer dwell
times can decrease the specificity of the assay and lessen the
ability of the system to discriminate mismatches at the cleavage
site or within the surrounding sequence (see Example 4). This
results because there is an increase in the efficiency of cleavage
of the primer when it is hybridized to a mismatch sequence.
Cleavage at the true target site cannot be further increased
because it is already at 100% each cycle. Thus the assay can be
tuned for SNP assays requiring higher specificity, or for
quantitation of expression levels of mRNA requiring less
specificity.
[0154] In another embodiment, a primer pair having one blocked
primer and one unblocked primer, can be used. In another
embodiment, an enzyme can be selected that has less sequence
specificity and can cleave various sequences. In yet another
embodiment, an additional mismatch flanking the cleavage site can
be added to increase the ability to discriminate variant alleles.
Modified bases such as 2'-O-methyl nucleosides can also be
introduced into the primer on either side of the cleavage site to
increase specificity (see Example 23).
[0155] The reactions of the various assays described herein can be
monitored using fluorescent detection, detection by mass tags,
enzymatic detection, and via labeling the probe or primer with a
variety of other groups including biotin, haptens, radionucleotides
and antibodies to name a few. In one embodiment, the progress of
PCR using the modified primers of the present invention is
monitored in real time using a dye intercelating assay with, for
example, SYBR.RTM. Green. In yet a further embodiment, the progress
of PCR using the modified primers of the present invention is
monitored using a probe labeled with a fluorophore and a quencher
such as a molecular beacon or, as in the 5' nuclease assay where
cleavage of the probe occurs. Alternatively, a dual labeled probe
which is cleavable by RNase H2 may be employed. In the latter case,
cleavage of both the hybridized primers and the probe can be
mediated by the same enzyme. The RNase H cleavage domain within the
probe may comprise only RNA residues. In general, all of the
combinations of residues useful in the cleavage domain of the
blocked primers of the present invention can be used as the
cleavage domain within the probe. In particular, when RNase H2 is
used as the cleavage enzyme, a single RNA residue or two adjacent
2'-F residues are preferred as the cleavage domain within the
probe. Such a modified oligonucleotide probe is particularly useful
in real-time PCR and can be employed with standard DNA primers or
with the blocked primers of the present invention. In such
real-time PCR assays, thermophilic versions of RNase H2 are
preferred, especially thermophilic RNase H2 enzymes having lower
activity at reduced temperatures. In the examples provided herein,
a number of thermophilic RNase H2 enzymes have been isolated and
have shown to be stable under thermocycling conditions and useful
in PCR. When used with the blocked primers of the present
invention, the need for a specific hot-start DNA polymerase can be
eliminated. This results in a significant decrease in assay
cost.
[0156] In another embodiment, the blocked primers of the present
invention can be used in the primer-probe assay format for PCR
described in U.S. Patent App. 2009/0068643. In this case, the
primer also contains a label domain on the 5' end of the
oligonucleotide which may or may not be complementary to the target
nucleic acid. The product generated by extension of the primer
serves as a template for synthesis by the reverse primer in the
next cycle of PCR. This converts the label domain into a double
stranded structure. In one such embodiment a fluorophore and a
quencher are attached to the label domain and the reaction is
monitored by an increase in fluorescence resulting from an increase
in the distance between the fluorophore and quencher in the double
stranded form compared to the single stranded state. In yet another
such embodiment the label domain contains a cleavage domain located
between the fluorophore and quencher. Cleavage occurs only when the
cleavage domain is double stranded. Again the reaction is monitored
by an increase in fluorescence. In this instance the cleaving agent
may be one that cleaves both strands, the primer and its
complement, such as a restriction enzyme. Alternativley the
cleaving agent may be a nicking agent that cleaves only the primer,
preferably an RNase H enzyme, and even more preferably a
thermostable RNase H2 enzyme. There are two cleavage domains within
the primer in this assay format: one 5' of the blocking group at
which cleavage occurs to activate the primer and allow extension
and the second within the label domain. Cleavage at both sites can
be mediated by the same cleaving agent. The label domain may also
contain other labeling groups including but not limited to biotin,
haptens and enzymes to name a few. Alternatively the 5' fragment
released by cleavage within the label domain may serve as a mass
tag for detection by mass spectrometry.
[0157] In yet another embodiment, the blocked primers of the
present invention can be used in the template-probe assay format
for PCR described in U.S. Patent App. 2009/0068643.
[0158] In another embodiment of the invention, RNase H2 cleavable
blocked oligonucleotides are used to detect 5-methylcytosine
residues by PCR analysis of sodium bisulfite treated nucleic acids,
including but not limited to DNA and RNA. Previous work has
established that treatment of nucleic acid template with bisulfite
will rapidly deaminate cytosines that are not methylated on the 5'
carbon of the base. This deamination reaction converts the
unmethylated cytosines into uracil, resulting in a functional
C->T transition mutation in the nucleic acid sequence. It is
also known that 5-methylcytosine is highly resistant to this
deamination, resulting in preservation of the 5-methylcytosine
nucleotide as a cytosine, rather than conversion to a thymine
Numerous methods have been employed to detect 5' cytosine
methylation modifications following the bisulfite conversion
technique. Examples include, but are not limited to, standard
mismatch-specific quantitative and non-quantitative PCR methods, as
well as subcloning and sequencing of the generated sodium bisulfite
reaction products.
[0159] In the present invention, the template is bisulfite treated
by methods that are well known to those in the art. If the starting
template was RNA, a complementary cDNA strand is generated by any
well known reverse transcription method. Blocked cleavable
oligonucleotides that will either match or discriminate against the
target template cytosines (now converted to uracils) or
5-methylcytosines are added to a PCR reaction containing the RNase
H2 enzyme and the bisulfite treated template. Amplification of the
mismatched (converted cytosine>uracil or unconverted
5-methylcytosine>5-methylcytosine) base containing template is
highly reduced relative to the matched base template due to the
mismatch discrimination of RNase H2 cleavage reaction. Incomplete
bisulfite conversion of cytosines to uracils, a consistent concern
with the sodium bisulfite conversion technique, can be detected by
the designing blocked cleavable oligonucleotides that target known
non-5'-methylated cytosines in the bisulfite converted template.
PCR amplification of unconverted cytosines with these primers
should display greater discrimination relative to standard
unblocked primers. The present invention is expected to
significantly increase the discrimination of the methylated and
unmethylated cytosines.
Allele Specific PCR
[0160] The blocked primers of the present invention can also be
used in allele-specific PCR (AS-PCR). In general, AS-PCR is used to
detect variant alleles of a gene, especially single base mutations
such as SNPs (see for example U.S. Pat. No. 5,496,699). SNP
locations in the genome, as well as sequences of mutated oncogenes,
are known in the art and PCR primers can be designed to overlap
with these regions.
[0161] Detection of single base mismatches is a critical tool in
diagnosing and correlating certain diseases to a particular gene
sequence or mutation. Although AS-PCR has been known in the
biological arts for more than a decade (Bottema et al., 1993,
Methods Enzymol., 218, pp. 388-402), tools are still needed to more
accurately discriminate between particular mismatches and fully
complementary sequences. The present invention addresses this
need.
[0162] In AS-PCR a primer is utilized which overlaps the variant
locus. Generally the primer is designed such that the 3'-terminal
nucleotide is positioned over the mutation site. Alternatively, the
mutation site is sometimes located over one or two bases from the
3'-end. If there is a mismatch at or near the 3'-end, primer
extension and hence PCR are inhibited. The difference between the
efficiency of amplification when there is an exact match with the
primers versus an allelic variant where there is one or more
mismatches can in some cases be measured by end point PCR in which
case the final amplification products are analyzed by, for example,
gel electrophoresis. More commonly real time PCR is used to
determine the efficiency of amplification. A fluorescence based
method of detection of the amplicon in real time such as a DNA dye
binding assay or a dual labeled probe assay is most often used. The
PCR cycle where fluorescence is first detectable above background
levels (the Cp, or crossing point) provides a measure of
amplification efficiency. If there is a mismatch between the primer
and the target DNA, amplification efficiency is reduced and the Cp
is delayed. Generally an increase in Cp of 4 to 5 cycles is
sufficient for discrimination of SNPs.
[0163] In one AS-PCR embodiment of the present invention, the
primer contains a single RNA residue, and the mismatch can be
aligned directly over the RNA residue of the primer. The difference
in crossing point (Cp) values between a perfect match and a
mismatch, correlating to a cleavage differential, is readily
apparent (see Example 13). In some instances, aligning the mismatch
one base to either the 5' side or the 3' side of the RNA residue
increases the difference in Cp values. When the mismatch is located
on the 5' side of the RNA residue, the subsequent RNase H2 cleavage
would leave the mismatch as the last base of the 3' end of the
cleaved primer. Surprisingly, having the mismatch directly on top
of the RNA residue is more effective in most cases than locating
the mismatch to the 5' side of the RNA residue.
[0164] In another embodiment, the primer contains multiple RNA
residues or two adjacent 2'-fluoro residues and detection of the
mismatch follows the same principles as with a primer containing
one RNA residue; the mismatch preferably is located near or on top
of the expected point of cleavage.
[0165] In another embodiment, a second mismatch is used to increase
the sensitivity of the assay. In yet a further embodiment, the
second mismatch is placed to the 3' side of the mismatch directly
over the SNP site. In yet a further embodiment, the second mismatch
is placed one or two bases from the mismatch directly over the SNP
site (see Example 23).
[0166] In yet another embodiment, modified residues are
incorporated into the primer on the 5'- or 3'-side of the base
located over the mutation site. In one such embodiment of the
present invention a 2'-O-methyl ribonucleoside is placed
immediately 5' to the RNA base within the primer (see Example
22).
[0167] The sensitivity of the assay can also be increased through
incorporation of nuclease resistant analogs into the primer on the
3'-side of the base over the mutation site. Such nuclease resistant
analogs include, but are not limited to, phosphorothioates,
phosphorodithioates, methylphosphonates and abasic residues such as
a C3 spacer. In one such embodiment of the present invention,
phosphorothioate internucleotide linkages are incorporated at each
position from the RNA base over the mutation site to the 3'-end of
the primer. In yet another such embodiment phosphorothioate
linkages or phosphoroditioate are incorporated at all positions
from the base on the 3'-side of the RNA residue to the 3'-end of
the primer. In yet another such embodiment a single
phosphorothioate or phosphorodithioates is introduced on the
3'-side of the residue immediately downstream from the RNA base
within the primer. In one embodiment, the phosphorothioate bonds
are placed between each monomer 3' to the RNA monomer directly over
the SNP site, as well as between the RNA monomer and the base 3' to
the RNA base (see Example 25).
[0168] The assay sensitivity can also be improved by optimizing the
placement of the 3' blocking group or groups. In one embodiment, a
blocking group is placed internal to the 3' end of the
oligonucleotide. In a further embodiment, more than on blocking
group is placed internal to the 3' terminus. In yet a further
embodiment, an RNA monomer sits directly over the SNP site, with a
DNA monomer 3' to the RNA monomer, followed by two C3 spacers, and
finally followed by a 3' terminal base (see Example 28).
[0169] In one embodiment of the allele-specific PCR, the primers
can be designed to detect more than one mismatch. For example, the
forward primer can detect a first mismatch, and the reverse primer
could detect a second mismatch. In this embodiment, the assay can
be used to indicate whether two mismatches occur on the same gene
or chromosome being analyzed. This assay would be useful in
applications such as determining whether a bacterium of interest is
both pathogenic and antibiotic resistant.
Reverse Transcriptase PCR (RT-PCR)
[0170] In yet another embodiment the methods of the present
invention can be used in coupled reverse transcription-PCR
(RT-PCR). In one such embodiment reverse transcription and PCR are
carried out in two disctinct steps. First a cDNA copy of the sample
mRNA is synthesized using either an oligo dT primer or a sequence
specific primer. Random hexamers and the like can also be used to
prime cDNA synthesis. The resulting cDNA is then used as the
substrate for PCR employing the blocked primers and methods of the
present invention.
[0171] Alternatively reverse transcription and PCR can be carried
out in a single closed tube reaction. In one such embodiment three
primers are employed, one for reverse transcription and two for
PCR. The primer for reverse transcription binds to the mRNA 3' to
the position of the PCR amplicon. Although not essential, the
reverse transcription primer can include RNA residues or modified
analogs such as 2'-O-methyl RNA bases which will not form a
substrate for RNase H when hybridized to the mRNA. Preferably an
RNase H2 enzyme which has decreased activity at lower temperatures
is used as the cleaving agent.
[0172] In the three primer RT-PCR assay it is desirable to inhibit
the RT-primer from participating in the PCR reaction. This can be
accomplished by utilizing an RT-primer having a lower Tm than the
PCR primers so it will not hybridize under the PCR conditions.
Alternatively, a non-replicable primer incorporating, for example,
two adjacent C3 spacers can be used as the RT-primer (as in
polynomial amplification, see U.S. Pat. No. 7,112,406). In this
case when the cDNA is copied by extension of the forward PCR primer
it will not include the binding site for the RT-primer.
[0173] In one embodiment, only the reverse PCR primer is blocked
utilizing the compositions and methods of the present invention. In
yet another embodiment both the forward and reverse PCR primers are
blocked. The reverse PCR primer is blocked in the 3 primer RT-PCR
assay to prevent it from being utilized for reverse transcription.
If desired, modified bases such as 2'-O-methyl RNA residues can be
incorporated in the reverse PCR primer although any such
modification must allow the primer sequence to serve as a template
for DNA synthesis and be copied.
[0174] In the two primer RT-PCR assays of the present invention,
only the forward PCR is blocked. The reverse PCR primer also serves
as the RT-primer and therefore can not be blocked.
[0175] While not comprehensive, Table 1 illustrates how variations
in the blocking groups, labeling groups, cleavage site embodiments,
modifications to the cleavage site or other regions of the
oligonucleotide, buffer conditions and enzyme can further optimize
assay formats depending on their particular application. Examples
of assay formats and applications include PCR; real-time PCR
utilizing double-stranded DNA-binding dyes such as SYBR.RTM. Green,
5' nuclease assays (Taqman.TM. assays) or molecular beacons;
primer-probe and template-probe assays (see U.S. Patent Application
2009/0068643); polynomial or linked linear amplification assays;
gene construction or fragment assembly via PCR; allele-specific PCR
and other methods used to detect single nucleotide polymorphisms
and other variant alleles; nucleic acid sequencing assays; and
strand displacement amplification. In these various assays,
cleavage of the primers of the present invention can be used to
enhance the specificity of the particular reaction.
TABLE-US-00001 TABLE 1 PCR/Primer Extension/Polyamp Primer RNase H
Flanking DNA Blocking Labeling Cleavage sequence Divalent Poly-
Group Group Site modifications cation merase RNase H Sample Use
Assay Format None None RNA None Mg.sup.2+ Hot Start RNase H1
Genomic Sample Prep No additional Fluorophore 1. Single Nuclease-
1. Ab RNase H2 DNA Coupled probe Fluorophore/ RNA resistant
linkages 2. Chemi- 1. Non- amplification 1.Detention of Quencher
residue 1. cally thermostable to reverse primer cleavage 2.
Multiple Phosphorothioate modified 2. Thermostable transcription A.
Fluorescence RNA 2. Dithioate A. Hot Start B. Mass Spec residues 3.
Methyl- i. Intrinsic C. Electrophoresis phosphonate ii. Ab 2.
Dye-binding 4. Non-nucleotide iii. Chemically assay spacers
modified A. Sybr Green B. Non-Hot Start Modification Enzyme
Modified Alter- Non-Hot RNase H3 and Mitochon- Quanti- With an
internal of 3'-terminal 1. Horse- residues: native Start other
catalysts drial/ fication probe residue radish 1. 2 divalent that
cleave chloroplast of target 1.Taqman .RTM. 1. C3 spacer peroxidase
adjacent cation RNA/DNA DNA nucleic acid 2. Fluorescence- 2.
Alkaline 2' F +/- Mg.sup.2+ heteroduplexes sequence quenched
phosphatase residues 1. Chromoso- linear probe mal copy 3.
Molecular number beacon 2. mRNA 4. RNase Biotin 2' OMe cDNA
Detection of H-cleavable Hapten variant allele 1. Deoxigenin
Upstream Antibody Secondary RNase H RNA Gene/ Tempro Assay
modification Mass Tag mismatches mutants having 1. mRNA Fragment 1.
RNase H2- 1. Adjacent to Radiolabel alterered cleav- construction
cleavable probe the 3'-terminal .sup.32P, .sup.14C, .sup.3H, age
specificity residue .sup.35S, etc. 1. Enhanced 2. Further cleavage
of upstream 2'-F substrates
Cycling Probe Reactions
[0176] Cycling probe reactions are another technique for detecting
specific nucleic acid sequences (see U.S. Pat. No. 5,403,711). The
reaction operates under isothermal conditions or with temperature
cycling. Unlike PCR products accumulate in a linear fashion.
[0177] Table 2 illustrates a non-comprehensive set of possible
elements of the current invention to improve assays based on the
cycling probe reaction. New features of the invention include 1)
use of a hot start RNase H enzyme; 2) cleavage of novel sequences
by RNase H enzymes (e.g., cleavage of substrates containing
2'-fluoronucleosides by Type II RNases H); and 3) introduction of
modifications and secondary mismatches flanking an RNase H cleavage
domain to enhance specificity and/or suppress nonspecific cleavage
reactions. Such modifications and secondary mismatches are
particularly useful when cleavage is mediated by a Type II RNase H
and the cleavage domain is a single RNA residue or two adjacent
2'-fluoro residues.
TABLE-US-00002 TABLE 2 Cycling Probe Reaction Primer Extension
RNase H Flanking Blocking Labeling Cleavage sequence Divalent-
Assay Group Group Site mods cation RNase H Sample Use Format None
None RNA None Mg.sup.2+ RNase H1 Genomic Quantification Stand-alone
Modification Fluorophore 1. Single Nuclease- RNase H2 DNA of target
nucleic 1. Isothermal of 3'-terminal Fluorophore/ RNA resistant 1.
Non- Mito- acid sequence 2. Temper- residue Quencher residue
linkages thermostable chondrial/ 1. Chromosomal ature 1. C3 spacer
2. Multiple 1.Phos- 2. Thermostable chloroplast copy number cycling
RNA phorothioate A. Hot Start DNA 2. mRNA residues 2. Dithioate i.
Intrinsic 3. Methyl- ii. Ab phosphonate iii. Chemically 4. Non-
modified nucleotide B. Non-Hot spacers Start Upstream Enzyme
Modified Alter- RNase H3 and Detection of Coupled to modification
1. Horseradish residues: native other catalysts variant allele
Amplifi- 1. Adjacent peroxidase 1. 2 diva- that cleave cation to
the 2. Alkaline adjacent lent RNA/DNA 1. PCR 3'-terminal
phosphatase 2' F cation heteroduplexes 2. LCR residue Biotin
residues 2' OMe +/- RNase H cDNA 3. Polyamp 2. Further Hapten
Mg.sup.2+ mutants upstream 1. Deoxigenin having altered Antibody
Secondary cleavage Mass Tag mismatches specificity Radiolabel 1.
Enhanced .sup.32P, .sup.14C, .sup.3H, cleavage of .sup.35S, etc.
2'-F substrates
DNA Ligation Assays
[0178] The present invention can also serve to increase the
specificity of DNA ligation assays. Donor and/or acceptor
oligonucleotides of the present invention can be designed which
bind adjacent to one another on a target DNA sequence and are
modified to prevent ligation. Blocking groups on the acceptor
oligonucleotide useful to inhibit ligation are the same as those
used to prevent primer extension. Blocking the donor
oligonucleotide can be readily accomplished by capping the 5'-OH
group, for example as a phosphodiester, e.g.:
##STR00003##
Other 5' blocking groups include 5'-O-alkyl substituents such as
5'-O-methyl or 5'-O-trityl groups, 5'-O-heteroalkyl groups such as
5'-OCH.sub.2CH.sub.2OCH.sub.3, 5'-O-aryl groups, and 5'-O-silyl
groups such as TIPS or TBDMS. A 5' deoxy residue can also be used
to block ligation.
[0179] Sterically bulky groups can also be placed at or near the
5'-end of the oligonucleotide to block the ligation reaction. A
5'-phosphate group cannot be used to block the 5'-OH as this is the
natural substrate for DNA ligase. Only after hybridization to the
target DNA sequence are the blocking groups removed by, for example
cleavage at an RNase H cleavable domain, to allow ligation to
occur. Preferably cleavage is mediated by an RNase H Type II
enzyme, and even more preferably a thermophilic Type II RNase H
enzyme. More preferably, a thermophilic Type II RNase H enzyme
which is less active at room temperature than at elevated
temperature is utilized to mediate cleavage and thereby activation
of the acceptor and/or donor oligonucleotide. Alternatively, a
sequence specific nicking enzyme, such as a restriction enzyme, may
be utilized to mediate cleavage of the donor and/or acceptor
oligonucleotide.
[0180] In a further embodiment, the cleaving reaction is first
carried out at a higher temperature at which only one of the two
oligonucleotides hybridizes to the target sequence. The temperature
is then lowered, and the second oligonucleotide hybridizes to the
target, and the ligation reaction then takes place.
[0181] In yet a further embodiment in which there is a cleavage
domain located in the donor oligonucleotide, this oligonucleotide
is not blocked at or near the 5'-end, but simply has a free 5'-OH.
This oligonucleotide cannot serve as a donor in the ligation
reaction; to do so requires a 5'-phosphate group. Thus, the 5'-end
is functionally blocked. Cleavage by RNase H generates a
5'-phosphate group allowing the donor oligonucleotide to
participate in the ligation reaction.
[0182] An important advantage of the present invention is that it
allows double interrogation of the mutation site, and hence greater
specificity, than standard ligation assays. There is an opportunity
for discrimination of a variant allele both at the cleavage step
and the ligation step.
[0183] Table 3 illustrates a non-comprehensive set of possible
elements of the current invention to improve oligonucleotide
ligation assays.
TABLE-US-00003 TABLE 3 Oligonucleotide Ligation Assay Donor
Acceptor Oligonucleotide Oligonucleotide Labeling RNase H Flanking
sequence Blocking Group Blocking Group Group Cleavage Site mods
Divalentcation DNA Ligase RNase H None None None RNA None Mg.sup.2+
Hot Start RNase H1 (5'-phosphate) 1. Single 1.Ab 5'-OH Modification
of Fluorophore RNA residues Nuclease-resistant 2.Chem- RNase H2
(Functional block) 3'-terminal 2. Multiple linkages ically 1.
Non-thermostable Modification of residue Fluorophore/ RNA residues
1. Phos- modifed 2. Thermostable 5'-residue 1. C3 spacer Quencher
phorothioate A. Hot Start 1. C3 spacer 2. Dithioate i. Intrinsic 3.
Methyl- ii. Ab phosphonate iii. Chemically 4. Non-nucleotide
modified spacers B. Non-Hot Start Downstream Upstream Enzyme
Modified Alternative Non-Hot RNase H3 and other modification
modification 1. Horse-radish residues: divalent Start catalysts
that cleave 1. Adjacent to 1. Adjacent to peroxidase 1. 2 adjacent
2' cation +/- RNA/DNA the 5'-terminal the 3'-terminal 2. Alkaline F
residues Mg.sup.2+ heteroduplexes residue residue phosphatase 2.
Further 2. Further Biotin 2' OMe RNase H mutants downstream
upstream Hapten having altered 1. Deoxigenin cleavage specificity
Antibody Secondary 1. Enhanced cleavage Mass Tag mismatches of 2'-F
substrates Radiolabel .sup.32P, .sup.14C, .sup.3H, .sup.35S, etc.
Reaction Sample Use Conditions Assay Format Genomic Quanitifcation
RNase H cleavage Stand-alone DNA of target and DNA ligation 1.
Single cycle nucleic acid at single 2. Linear sequence temperature
Amplification Mito-chondrial/ 1. Chromosomal 3. LCR chloropast copy
number DNA 2. mRNA Detection of RNase H cleavage Coupled to variant
at elevated primer allele temperature extension (reduced 1. PCR
temperature for 2. Reverse cDNA DNA ligation) transcription 3.
Polyamp
Sequencing Reactions
[0184] In one embodiment, a method of sequencing a target DNA of
interest is provided. The method entails
(a) providing a reaction mixture comprising a primer having a
cleavage domain and a blocking group linked at or near to the 3'
end of the primer which prevents primer extension, a sample nucleic
acid comprising the target DNA sequence of interest, a cleaving
enzyme, nucleotide triphosphate chain terminators (e.g., 3'
dideoxynucleotide triphosphates) and a polymerase, (b) hybridizing
the primer to the target nucleic acid to form a double-stranded
substrate; (c) cleaving the hybridized primer with the cleaving
enzyme at a point within or adjacent to the cleavage domain to
remove the blocking group from the primer; and (d) extending the
primer with the polymerase.
[0185] In one embodiment, the invention is used in a "next
generation" sequencing platform. One type of next generation
sequencing is "sequencing by synthesis", wherein genomic DNA is
sheared and ligated with adapter oligonucleotides or amplified by
gene-specific primers, which then are hybridized to complementary
oligonucleotides that are either coated onto a glass slide or are
placed in emulsion for PCR. The subsequent sequencing reaction
either incorporates dye-labeled nucleotide triphosphates or is
detected by chemiluminescence resulting from the reaction of
pyrophosphate released in the extension reaction with ATP
sulfurylase to generate ATP and then the ATP-catalyzed reaction of
luciferase and its substrate luciferin to generate oxyluciferin and
light.
[0186] A second type of next generation sequencing is "sequencing
by ligation", wherein four sets of oligonucleotides are used,
representing each of the four bases. In each set, a
fluorophore-labeled oligonucleotide of around 7 to 11 bases is
employed in which one base is specified and the remaining are
either universal or degenerate bases. If, for example, an 8-base
oligonucleotide is used containing 3 universal bases such as
inosine and 4 degenerate positions, there would be 4.sup.4 or 256
different oligonucleotides in each set each with a specified base
(A, T, C or G) at one position and a fluorescent label attached to
either the 5'- or 3'-end of the molecule or at an internal position
that does not interfere with ligation. Four different labels are
employed, each specific to one of the four bases. A mixture of
these four sets of oligonucleotides is allowed to hybridize to the
amplified sample DNA. In the presence of DNA ligase the
oligonucleotide hybridized to the target becomes ligated to an
acceptor DNA molecule. Detection of the attached label allows the
determination of the corresponding base in the sample DNA at the
position complementary to the base specified within the
oligonucleotide.
[0187] In one embodiment of the present invention, a donor
oligonucleotide of about 7-11 bases contains a specified base at
the 5' end of the oligonucleotide. The remaining bases are
degenerate or universal bases, and a label specific to the
specified base is incorporated on the 3' side of the specified
base. The 3' end of the probe is irreversibly blocked to prevent
the donor oligonucleotide from also acting as an acceptor. In some
cases this may be accomplished by the labeling group. The second
base from the 5' end of the oligonucleotide, i.e., the residue next
to the specified base is a degenerate mixture of the 4 RNA bases.
Alternatively, any anaolog recognized by RNase H2, such as a
2'-fluoronucleoside may be substituted at this position. A
universal base such as riboinosine or ribo-5-nitroindole, may also
be incorporated at this location. The probe first hybridizes to the
target sequence and becomes ligated to the acceptor DNA fragment as
in the standard sequencing by ligation reaction. After detection of
the specified base, RNase H2 is added which cleaves the probe on
the 5'-side of the RNA residue leaving the specified base attached
to the 3' end of the acceptor fragment. The end result is that the
acceptor fragment is elongated by one base and now is in position
to permit the determination of the next base within the sequence.
The cycle is repeated over and over, in each case moving the
position of hybridization of the donor oligonucleotide one base 3'
down the target sequence. The specificity is increased compared to
traditional sequencing by ligation because the specified base is
always positioned at the junction of the ligation reaction.
[0188] The donor oligonucleotide probe can optionally contain
universal bases including, but not limited to, 5-nitroindole,
ribo-5'-nitroindole, 2'-O-methyl-5-nitroindole, inosine,
riboinosine, 2'-O-methylriboinosine and 3-nitropyrrole. This
reduces the number of different oligonucleotides in each set
required for the assay by a factor of four for every degenerate
position on the probe substituted with a universal base. The method
can also include a capping step between the ligation reaction and
the RNase H2 cleaving step. The capping reaction can be performed
by introducing a DNA polymerase and a chain terminator, thereby
capping any of the acceptor fragment molecules that did not ligate
with a donor oligonucleotide probe in the previous step.
[0189] In the above example the ligation reactions and hence the
sequencing readout proceeds in the 5'- to 3'-direction one base at
a time. Alternatively the donor oligonucleotide can be designed so
that two bases are determined in each cycle. In this case the first
two bases on the 5'-end of the donor oligonucleotide are specified
(for example, pA-C-R-N-N-N-I-I-X, where R=a degenerate mixture of
all 4 RNA bases, N=a degenerate DNA base, I=inosine, and X is a
fluorophore). As in all cases there is a 5'-phosphate (p) to permit
ligation of the donor oligonucleotide to the acceptor. Sixteen such
oligonucleotide sets are required, one for each of the sixteen
possible dinucleotides. Each of the sixteen can be labeled with a
different fluorophore. Alternatively ligation reactions can be
carried out with 4 separate pools each having four such sets of
oligonucleotides. In that case, only four different fluorophores
are required.
[0190] In another embodiment for sequencing in the 5'- to
3'-direction a donor oligonucleotide of the following type can be
used: pA-N-R-N-N-N-I-I-X wherein p, N, R, I and X are as defined in
the previous example. One base is determined at each cycle but at
alternate positions: 1, 3, 5, etc. This may be adequate for
identification of the sequence if compared to a reference database.
If desired, the remaining bases (positions 2, 4, 6, etc.) can be
determined by repeating the sequencing reaction on the same
template with the original acceptor oligonucleotide shifted one
base upstream or downstream. In a related example a donor
oligonucleotide of the following type can be used:
p-A-F-FN-N-N-I-I-X wherein p, N, I and X are as defined above and F
is a degenerate mixture of all four 2'-fluoronucleosides. Following
ligation, cleavage by RNase H2 results in the addition of two bases
to the 3'-end of the acceptor (i.e., AF). After the next ligation
reaction, the sequence at the 3'-end of the acceptor would be . . .
A-F-S-F-F-N-N-I-I-X where S is the specified base at position 3,
and X would be a different fluorophore from the previous cycle if
the specified base were not A. Cleavage with RNase H2 next occurs
between the two 2'-fluororesidues. Cleavage by RNase H2 at the
isolated 2'-fluororesidue occurs much more slowly and can be
avoided by adjusting the RNase H2 concentration and reaction
time.
[0191] A variant of the above method can be performed in which
sequencing proceeds in the 3'- to 5'-direction. In this case an
acceptor oligonucleotide is added at each cycle as in the following
structure: X-I-I-N-N-N-F-F-S-OH wherein the specified base (S) is
at the 3'-end of the oligonucleotide. The 5'-end is blocked to
prevent the oligonucleotide from acting as a donor. Cleavage by
RNase H2 leaves the sequence pF-S at the 5'-end of the donor
fragment which is prepared for the next sequencing cycle. A capping
step can be included in the cycle before the cleavage reaction
using a phosphatase to remove the 5'-phosphate of the donor
oligonucleotide if ligation to the acceptor failed to occur.
[0192] In a further embodiment, the invention provides an
improvement for DNA sequencing using ribotriphosphates (or
alternative analogs which provide a substrate for RNase H2, such as
2'-fluoronucleoside triphosphates) in conjunction with a
fluorescently labeled primer. Similar to traditional sequencing
methods known in the art, the triphosphate residue would be
incorporated by a DNA polymerase. The concentration of the ribo
triphosphate, or the alternative analog providing a substrate for
RNase H2, is adjusted to a concentration such that on average one
such base is incorporated randomly within each extension product
produced by the polymerase. The nested family of fragments
originating from the primer is generated by cleavage with RNase H2
and then separated by electrophoresis as in standard DNA sequencing
methods. Alternatively, multiple RNA residues or modified
nucleosides such as 2'-fluoronucleosides may be incorporated into
the extension product and the subsequent digestion with RNase H2 is
limited so that on average each strand is cut only once. Four
separate reactions are run, each substituting one of the bases with
a different ribotriphosphate (A, C, T or G) or other RNase H2
cleavable analog. In this assay, use of expensive fluorescently
labeled dideoxy triphosphate chain terminators is obviated.
[0193] In another embodiment of the present invention, an improved
method for oligonucleotide synthesis is provided. Using similar
techniques as described above, a composition acting as a donor
oligonucleotide can be ligated to an acceptor fragment in order to
add additional bases to the 3'-end of the acceptor fragment. It is
the acceptor fragment that is the growing polynucleotide undergoing
synthesis. In this case, the composition of the donor fragment is
preferably a single-stranded oligonucleotide that forms a hairpin
to provide a double-stranded region with an overhang of about 1-8
bases on the 3'-end. The base at the 5' end would be the desired
base to add to the growing acceptor fragment. For synthesis of a
polynucleotide containing all four bases (A, C, T and G), four
different donor fragments are employed which can have the identical
sequence except varying in the 5' base. Preferably the donor is
blocked at the 3'-end so it cannot react as an acceptor. The
blocking group placed at or near the 3'-end of the donor can be a
label to allow monitoring of the reaction. Four different labels
can be used corresponding to the four different bases at the 5'-end
of the donor. The base adjacent to the desired base at the 5'-end
is a RNA base or an alternative analog such as a
2'-fluoronucleoside which provides a substrate for RNase H2. The
overhang at the 3'-end can be random (degenerate) bases or
universal bases or a combination of both. The donor fragment binds
to the acceptor fragment, through hybridization of the 3'-end of
the acceptor to the 3'-overhang of the donor oligonucleotide. A DNA
ligase enzyme is then used to join the two fragments. Next a Type
II RNase H is used to cleave the product on the 5'-side of the
RNase H2 cleavage site, transferring the 5' base of the donor to
the 3'-end of the acceptor. Optionally, a third step can be
included in the cycle between the ligase and RNase H2 cleavage
reactions in which molecules of the growing polynucleotide chain
which may have failed to ligate are capped by reaction with a
dideoxynucleotide triphosphate (or other chain terminator)
catalyzed by a DNA polymerase. In one embodiment the DNA polymerase
is a deoxynucleotide terminal transferase. The cycle is repeated,
and the acceptor fragment can continue to be extended in a 5' to 3'
direction. To facilitate isolation of the growing polynucleotide at
each step the acceptor can be attached to a solid support such as
controlled pore glass or polystyrene
[0194] Similar to the sequence method described above, a donor
oligonucleotide can be used to add two bases to the 3'-end of the
acceptor oligonucleotide at each cycle.
[0195] In this case the RNase H2 cleavable residue would be
positioned 3' from the 5' end of the donor. This enzymatic
synthesis method is particularly advantageous for synthesis of
longer DNA molecules. The hairpin reagents corresponding to each
base can be collected for reuse in further cycles or additional
syntheses. Because the system does not use organic solvents, waste
disposal is simplified.
Kits of the Present Invention
[0196] The present invention also provides kits for nucleic acid
amplification, detection, sequencing, ligation or synthesis that
allow for use of the primers and other novel oligonucleotides of
the present invention in the aforementioned methods. In some
embodiments, the kits include a container containing a cleavage
compound, for example a nicking enzyme or an RNase H enzyme;
another container containing a DNA polymerase and/or a DNA ligase
and preferably there is an instruction booklet for using the kits.
In certain embodiments, the kits include a container containing
both a nicking enzyme or an RNase H enzyme combined with a DNA
polymerase or DNA ligase. Optionally, the modified oligonucleotides
used in the assay can be included with the enzymes. The cleavage
enzyme agent, DNA polymerase and/or DNA ligase and oligonucleotides
used in the assay are preferably stored in a state where they
exhibit long-term stability, e.g., in suitable storage buffers or
in a lyophilized or freeze dried state. In addition, the kits may
further comprise a buffer for the nicking agent or RNase H, a
buffer for the DNA polymerase or DNA ligase, or both buffers.
Alternatively, the kits may further comprise a buffer suitable for
both the nicking agent or RNase H, and the DNA polymerase or DNA
ligase. Buffers may include RNasin and other inhibitors of single
stranded ribonucleases. Descriptions of various components of the
present kits may be found in preceding sections related to various
methods of the present invention.
[0197] Optionally, the kit may contain an instruction booklet
providing information on how to use the kit of the present
invention for amplifying or ligating nucleic acids in the presence
of the novel primers and/or other novel oligonucleotides of the
invention. In certain embodiments, the information includes one or
more descriptions on how to use and/or store the RNase H, nicking
agent, DNA polymerase, DNA ligase and oligonucleotides used in the
assay as well as descriptions of buffer(s) for the nicking agent or
RNase H and the DNA polymerase or DNA ligase, appropriate reaction
temperature(s) and reaction time period(s), etc.
[0198] Accordingly, in one embodiment, a kit for the selective
amplification of a nucleic acid from a sample is provided. The kit
comprises
(a) a first and a second oligonucleotide primer, each having a 3'
end and 5' end, wherein each oligonucleotide is complementary to a
portion of a nucleic acid to be amplified or its complement, and
wherein at least one oligonucleotide comprises a RNase H cleavable
domain, and a blocking group linked at or near to the 3' end of the
oligonucleotide to prevent primer extension and/or to prevent the
primer from being copied by DNA synthesis directed from the
opposite primer; (b) an RNase H enzyme; and (c) an instruction
manual for amplifying the nucleic acid. The kit may optionally
include a DNA polymerase.
[0199] In a further embodiment, the kit for selective amplification
of a nucleic acid includes an oligonucleotide probe having a 3' end
and a 5' end comprising an RNase H cleavable domain, a fluorophore
and a quencher, wherein the cleavable domain is positioned between
the fluorophore and the quencher, and wherein the probe is
complementary to a portion of the nucleic acid to be amplified or
its complement.
[0200] In yet another embodiment, the present invention is directed
to a kit for the ligation of an acceptor oligonucleotide and a
donor oligonucleotide in the presence of a target nucleic acid
sequence. The kit comprises
(a) a donor oligonucleotide and an acceptor oligonucleotide in
which one or both of the oligonucleotides comprise an RNase H
cleavable domain and a blocking group preventing ligation; (b) an
RNase H enzyme; and (c) an instruction manual for ligating the
acceptor and donor oligonucleotides in the presence of a target
nucleic acid sequence.
[0201] In a further embodiment, the kit may optionally include a
DNA ligase enzyme.
[0202] In a further ligation kit embodiment, the donor
oligonucleotide contains an RNase H cleavage domain, but lacks a
blocking group at or near the 5'-end and instead has a free
5'-OH.
EXAMPLES
[0203] The present invention is further illustrated by reference to
the following Examples. However, it should be noted that these
Examples, like the embodiments described above, are illustrative
and are not to be construed as restricting the enabled scope of the
invention in any way.
Example 1
Cloning of Codon Optimized RNase H2 Enzymes from Thermophilic
Organisms
[0204] This example describes the cloning of codon optimized RNase
H2 enzymes from thermophilic organisms.
[0205] To search for functional novel RNase H2 enzymes with
potentially new and useful activities, candidate genes were
identified from public nucleotide sequence repositories from
Archaeal hyperthermophilic organisms whose genome sequences had
previously been determined. While RNase H2 enzymes do share some
amino acid homology and have several highly conserved residues
present, the actual homology between the identified candidate genes
was low and it was uncertain if these represented functional RNase
H2 enzymes or were genes of unknown function or were non-functional
RNase H2 genes. As shown in Table 4, five genes were selected for
study, including two organisms for which the RNase H2 genes have
not been characterized and three organisms to use as positive
controls where the RNase H2 genes (rnhb) and functional proteins
have been identified and are known to be functional enzymes.
Although two uncharacterized predicted rnhb genes were selected for
this initial study, many more Archaeal species have had their
genome sequences determined whose rnhb genes are uncharacterized
which could similarly be studied.
TABLE-US-00004 TABLE 4 Five candidate RNase H2 (rnhb) genes from
thermophilic bacteria Organism Accession # Length Comments
Pyrococcus AB012613 687 bp, 228 See References kodakaraensis AA
(1-3) below Pyrococcus AE010276 675 bp, 224 See Reference (4)
furiosus AA below and UA20040038366A1 Methanocaldococcus U67470 693
bp, 230 See References jannaschii AA (5, 6) below Pyrococcus
AJ248284 675 bp, 224 uncharacterized abyssi AA Sulfolobus AE006839
639 bp, 212 uncharacterized solfataricus AA Bp = base pairs; AA =
amino acids
References 1-6: 1) Haruki, M., Hayashi, K., Kochi, T., Muroya, A.,
Koga, Y., Morikawa, M., Imanaka, T. and Kanaya, S. (1998) Gene
cloning and characterization of recombinant RNase HII from a
hyperthermophilic archaeon. J Bacteriol, 180, 6207-6214; 2) Haruki,
M., Tsunaka, Y., Morikawa, M. and Kanaya, S. (2002) Cleavage of a
DNA-RNA-DNA/DNA chimeric substrate containing a single
ribonucleotide at the DNA-RNA junction with prokaryotic RNases HII.
FEBS Lett, 531, 204-208; 3) Mukaiyama, A., Takano, K., Haruki, M.,
Morikawa, M. and Kanaya, S. (2004) Kinetically robust monomeric
protein from a hyperthermophile. Biochemistry, 43, 13859-13866 4)
Sato, A., Kanai, A., Itaya, M. and Tomita, M. (2003) Cooperative
regulation for Okazaki fragment processing by RNase HII and FEN-1
purified from a hyperthermophilic archaeon, Pyrococcus furiosus.
Biochem Biophys Res Commun, 309, 247-252; 5) Lai, B., Li, Y., Cao,
A. and Lai, L. (2003) Metal ion binding and enzymatic mechanism of
Methanococcus jannaschii RNase HII. Biochemistry, 42, 785-791; and
6) Lai, L., Yokota, H., Hung, L. W., Kim, R. and Kim, S. H. (2000)
Crystal structure of archaeal RNase HII: a homologue of human major
RNase H. Structure, 8, 897-904.
[0206] The predicted physical properties of the proteins encoded by
the rnhb genes listed above are shown in Table 5 (Pace, C. N. et
al., (1995) Protein Sci., 4, p. 2411).
TABLE-US-00005 TABLE 5 Characteristics of five RNase H2 enzymes #
residues Mol. Molecules/.mu.g Trp, Tyr, .epsilon. 280 nm Organism
weight protein Cys M.sup.-1 cm.sup.-1 Pyrococcus 25800.5 2.3E13 1,
7, 0 15930 kodakarensis Pyrococcus 25315.2 2.4E13 2, 8, 0 22920
furiosus Methanocaldococcus 26505.8 2.3E13 1, 9, 3 19285 jannaschii
Pyrococcus 25394.2 2.4E13 3, 7, 0 26930 abysii Sulfolobus 23924.8
2.5E13 3, 10, 0 31400 solfataricus
[0207] The amino acid similarity between RNase H2 enzymes (or
candidate enzymes) from different Archaeal species within this set
of 5 sequences ranges from 34% to 65%. An amino-acid identity
matrix is shown in Table 6 below.
TABLE-US-00006 TABLE 6 Amino acid identity between five Archaeal
RNase H2 proteins P. kod. P. fur. M. jann. P. ab. S. solf. P.
kodakarensis -- 0.570 0.595 0.358 0.333 P. furiosus 0.570 -- 0.654
0.410 0.362 M. jannaschii 0.595 0.654 -- 0.380 0.363 P. abysii
0.358 0.410 0.380 -- 0.336 S. solfataricus 0.333 0.362 0.363 0.336
--
[0208] Codons of the native gene sequence were optimized for
expression in E. coli using standard codon usage tables. The
following sequences were assembled and cloned into plasmids as
artificial genes made from synthetic oligonucleotides using
standard methods. DNA sequence identity was verified on both
strands. Sequences of the artificial DNA constructs are shown
below. Lower case letters represents linker sequences, including a
Bam HI site on the 5'-end and a Hind III site on the 3'-end. Upper
case letters represents coding sequences and the ATG start codons
are underlined.
[0209] SEQ ID NO: 1--codon optimized rnhb gene from Pyrococcus
kodakaraensis
TABLE-US-00007 ggatccgATGAAGATTGCTGGCATCGATGAAGCCGGCCGTGGCCCGGTA
ATTGGTCCAATGGTTATCGCTGCGGTAGTCGTGGACGAAAACAGCCTGC
CAAAACTGGAAGAGCTGAAAGTGCGTGACTCCAAGAAACTGACCCCGAA
GCGCCGTGAAAAGCTGTTTAACGAAATTCTGGGTGTCCTGGACGATTAT
GTGATCCTGGAGCTGCCGCCTGATGTTATCGGCAGCCGCGAAGGTACTC
TGAACGAGTTCGAGGTAGAAAACTTCGCTAAAGCGCTGAATTCCCTGAA
AGTTAAACCGGACGTAATCTATGCTGATGCGGCTGACGTTGACGAGGAA
CGTTTTGCCCGCGAGCTGGGTGAACGTCTGAACTTTGAAGCAGAGGTTG
TTGCCAAACACAAGGCGGACGATATCTTCCCAGTCGTGTCCGCGGCGAG
CATTCTGGCTAAAGTCACTCGTGACCGTGCGGTTGAAAAACTGAAGGAA
GAATACGGTGAAATCGGCAGCGGTTATCCTAGCGATCCTCGTACCCGTG
CGTTTCTGGAGAACTACTACCGTGAACACGGTGAATTCCCGCCGATCGT
ACGTAAAGGTTGGAAAACCCTGAAGAAAATCGCGGAAAAAGTTGAATCT
GAAAAAAAAGCTGAAGAACGTCAAGCAACTCTGGACCGTTATTTCCGTA AAGTGaagctt
[0210] SEQ ID NO: 2--codon optimized rnhb gene from Pyrococcus
furiosus
TABLE-US-00008 ggatccgATGAAGATTGGTGGCATCGACGAAGCCGGCCGTGGTCCGGCG
ATCGGTCCGCTGGTAGTAGCTACTGTTGTAGTGGATGAAAAAAACATCG
AAAAACTGCGTAACATCGGCGTAAAAGACTCCAAACAGCTGACGCCGCA
CGAACGTAAAAACCTGTTTTCCCAGATCACCTCCATTGCGGATGATTAC
AAGATCGTAATCGTGTCTCCGGAAGAAATTGACAACCGTAGCGGTACCA
TGAACGAGCTGGAAGTTGAAAAATTCGCGCTGGCGCTGAACTCTCTGCA
GATCAAGCCGGCTCTGATCTACGCAGACGCAGCAGATGTTGATGCAAAC
CGCTTCGCATCCCTGATCGAACGTCGCCTGAACTATAAAGCCAAAATCA
TCGCGGAACACAAAGCAGACGCAAAGTACCCGGTCGTTTCTGCGGCGAG
CATTCTGGCGAAGGTTGTGCGTGACGAAGAAATCGAAAAGCTGAAAAAG
CAATATGGCGACTTTGGCAGCGGTTACCCGAGCGACCCGAAAACGAAGA
AATGGCTGGAGGAGTATTACAAGAAACATAACAGCTTCCCACCGATCGT
TCGTCGTACGTGGGAAACTGTCCGCAAAATTGAAGAGTCCATCAAAGCC
AAAAAGTCCCAGCTGACCCTGGATAAATTCTTCAAGAAACCGaagctt
[0211] SEQ ID NO: 3--codon optimized rnhb gene from
Methanocaldococcus jannaschii
TABLE-US-00009 ggatccgATGATTATCATTGGTATCGATGAAGCTGGCCGTGGTCCTGTA
CTGGGCCCGATGGTTGTATGTGCGTTCGCTATCGAGAAGGAACGTGAAG
AAGAACTGAAAAAGCTGGGCGTTAAAGATTCTAAAGAACTGACGAAGAA
TAAACGCGCGTACCTGAAAAAGCTGCTGGAGAACCTGGGCTACGTGGAA
AAGCGCATCCTGGAGGCTGAGGAAATTAACCAGCTGATGAACAGCATTA
ACCTGAACGACATTGAAATCAACGCATTCAGCAAGGTAGCTAAAAACCT
GATCGAAAAGCTGAACATTCGCGACGACGAAATCGAAATCTATATCGAC
GCTTGTTCTACTAACACCAAAAAGTTCGAAGACTCTTTCAAAGATAAAA
TCGAAGATATCATTAAAGAACGCAATCTGAATATCAAAATCATTGCCGA
ACACAAAGCAGACGCCAAGTACCCAGTAGTGTCTGCGGCGAGCATTATC
GCGAAAGCAGAACGCGACGAGATCATCGATTATTACAAGAAAATCTACG
GTGACATCGGCTCTGGCTACCCATCTGACCCGAAAACCATCAAATTCCT
GGAAGATTACTTTAAAAAGCACAAGAAACTGCCGGATATCGCTCGCACT
CACTGGAAAACCTGCAAACGCATCCTGGACAAATCTAAACAGACTAAAC
TGATTATCGAAaagctt
[0212] SEQ ID NO: 4--codon optimized rnhb gene from Pyrococcus
abysii
TABLE-US-00010 ggatccgATGAAAGTTGCAGGTGCAGATGAAGCTGGTCGTGGTCCAGTT
ATTGGTCCGCTGGTTATTGTTGCTGCTGTTGTGGAGGAAGACAAAATCC
GCTCTCTGACTAAGCTGGGTGTTAAAGACTCCAAACAGCTGACCCCGGC
GCAACGTGAAAAACTGTTCGATGAAATCGTAAAAGTACTGGATGATTAC
TCTGTGGTCATTGTGTCCCCGCAGGACATTGACGGTCGTAAGGGCAGCA
TGAACGAACTGGAGGTAGAAAACTTCGTTAAAGCCCTGAATAGCCTGAA
AGTTAAGCCGGAAGTTATTTACATT ATTCCGCTGATGTTAAAGCTGAA
CGTTTCGCTGAAAACATTCGCAGCCGTCTGGCGTACGAAGCGAAAGTTG
TAGCCGAACATAAAGCGGATGCGAAGTATGAGATCGTATCCGCAGCCTC
TATCCTGGCAAAAGTTATCCGTGACCGCGAGATCGAAAAGCTGAAAGCC
GAATACGGTGATTTTGGTTCCGGTTACCCGTCTGATCCGCGTACTAAGA
AATGGCTGGAAGAATGGTATAGCAAACACGGCAATTTCCCGCCGATCGT
GCGTCGTACTTGGGATACTGCAAAGAAAATCGAAGAAAAATTCAAACGT
GCGCAGCTGACCCTGGACAACTTCCTGAAGCGTTTTCGCAACaagctt
[0213] SEQ ID NO: 5--codon optimized rnhb gene from Sulfolobus
solfataricus
TABLE-US-00011 ggatccgATGCGCGTTGGCATCGATGAAGCGGGTCGCGGTGCCCTGATC
GGCCCGATGATTGTTGCTGGTGTTGTAATCTCTGACACTAAACTGAAGT
TTCTGAAAGGCATCGGCGTAAAAGACTCTAAACAGCTGACTCGCGAGCG
TCGTGAAAAGCTGTTTGATATTGTTGCTAACACTGTGGAAGCATTCACT
GTCGTTAAAGTTTTCCCTTATGAAATCGACAACTATAACCTGAATGACC
TGACCTACGACGCAGTTTCTAAAATCATCCTGAGCCTGTCTAGCTTTAA
CCCAGAAATTGTAACGGTTGATAAAGTGGGCGATGAGAAACCGGTTATC
GAACTGATTAATAAGCTGGGCTACAAAAGCAACGTCGTACACAAGGCAG
ATGTACTGTTTGTAGAAGCCTCCGCTGCTAGCATCATTGCGAAAGTTAT
TCGTGATAACTACATTGACGAACTGAAACAAGTATACGGTGACTTTGGT
AGCGGTTACCCAGCTGATCCTCGCACTATCAAATGGCTGAAATCTTTCT
ACGAAAAGAATCCGAATCCGCCGCCAATCATTCGTCGTTCCTGGAAGAT
TCTGCGTTCTACCGCCCCGCTGTATTACATTTCCAAAGAAGGTCGCCGT CTGTGGaagctt
Example 2
Expression of Recombinant RNase H2 Peptides
[0214] The following example demonstrates the expression of
recombinant RNase H2 peptides.
[0215] The five synthetic gene sequences from Example 1 were
subcloned using unique Bam HI and Hind III restriction sites into
the bacterial expression vector pET-27b(+) (Novagen, EMD
Biosciences, La Jolla, Calif.). This vector places six histidine
residues (which together comprise a "His-tag") (SEQ ID NO: 292) at
the carboxy terminus of the expressed peptide (followed by a stop
codon). A "His-tag" permits use of rapid, simple purification of
recombinant proteins using Ni affinity chromatography, methods
which are well known to those with skill in the art. Alternatively,
the synthetic genes could be expressed in native form without the
His-tag and purified using size exclusion chromatography,
anion-exchange chromatography, or other such methods, which are
also well known to a person of ordinary skill in the art.
[0216] BL21(DE3) competent cells (Novagen) were transformed with
each plasmid and induced with 0.5 mM
isopropyl-.beta.-D-thio-galactoside (IPTG) for 4.5 hours at
25.degree. C. For all clones, 5 mL of IPTG induced culture was
treated with Bugbuster.RTM. Protein Extraction Reagent and
Benzonase.RTM. Nuclease (Novagen) to release soluble proteins and
degrade nucleic acids according to the manufacturer's instructions.
The recovered protein was passed over a Ni affinity column
(Novagen) and eluted with buffer containing 1M imidazole according
to protocols provided by the manufacturer.
[0217] Both "total" and "soluble" fractions of the bacterial lysate
were examined using SDS 10% polyacrylamide gel electrophoresis.
Proteins were visualized with Coomassie Blue staining. Following
IPTG induction, large amounts of recombinant proteins were produced
from all 5 Archaeal RNase H2 synthetic genes. Using this method of
purification, protein was recovered in the soluble fraction for 4
enzymes, Pyrococcus kodakaraensis, Pyrococcus furiosus,
Methanocaldococcus jannaschii, and Pyrococcus abyssi. No soluble
protein was recovered for Sulfolobus solfataricus RNase H2 using
this lysis procedure. Examples of induced RNase H2 proteins are
shown in FIGS. 4A and 4B.
[0218] Improved methods to produce and purify the recombinant
proteins were developed to produce small scale amounts of the
proteins for characterization as follows. To maximize the amount of
soluble protein obtained for each clone, an induction temperature
of 37.degree. C. is used for 6 hours. For Pyrococcus kodakaraensis,
Methanocaldococcus jannaschii, and Sulfolobus solfataricus,
CelLytic.TM. B 10.times. lysis reagent (Sigma-Aldrich, St. Louis,
Mo.) is used for lysis. A 10 fold dilution in 500 mM NaCl, 20 mM
TrisHCl, 5 mM imidazole, pH 7.9 is made and 10 mL is used per 0.5 g
of pelleted bacterial paste from induced cultures. For Pyrococcus
furiosus and Pyrococcus abyssi, 5 mL of Bugbuster.RTM. Protein
Extraction Reagent (Novagen) per 100 mL of induced culture is used
for cell lysis. In addition, per 100 mL induced culture for all
clones, 5KU rLysozyme.TM. (Novagen) and 250U DNase I (Roche
Diagnostics, Indianapolis, Ind.) is used to enhance bacterial cell
lysis and decrease the viscosity of the solution. Following
centrifugation to remove cell debris, the lysates are heated for 15
minutes at 75.degree. C. to inactive the DNase I and any other
cellular nucleases present. The lysates are then spun at
16,000.times.g for 20 minutes to sediment denatured protein
following heat treatment. The centrifugation step alone provides a
large degree of functional purification of the recombinant
thermostable enzymes.
[0219] The resulting soluble supernatant is passed over a Ni
affinity column containing His.Bind.RTM. Resin (Novagen) and eluted
with an elution buffer containing 200 mM imidazole. The purified
protein is then precipitated in the presence of 70% ammonium
sulfate and resuspended in storage buffer (10 mM Tris pH 8.0, 1 mM
EDTA, 100 mM NaCl, 0.1% Triton X-100, 50% Glycerol) to concentrate
and stabilize the protein for long term storage. The concentrated
protein is dialyzed 2.times.2 hours (.times.250 volumes each)
against the same storage buffer to remove residual salts. The final
purified protein is stored at -20.degree. C. Using these protocols,
for Pyrococcus abysii, 200 mL of IPTG induced culture yields
.about.2 mg of soluble protein. After passing over a Ni column,
.about.0.7 mg of pure protein is recovered. For functional use, the
concentrated enzyme stocks were diluted in storage buffer and added
1:10 in all enzymatic reactions studied. Therefore all reaction
buffers contain 0.01% Triton X-100 and 5% Glycerol.
[0220] Recombinant protein was made and purified for each of the
cloned RNase H2 enzymes as outlined above. Samples from Pyrococcus
kodakaraensis, Pyrococcus furiosus, Pyrococcus abyssi, and
Sulfolobus solfataricus were examined using SDS 10% polyacrylamide
gel electrophoresis. Proteins were visualized with Coomassie Blue
staining. Results are shown in FIG. 5. If the expression and
purification method functioned as predicted, these proteins should
all contain a 6.times. Histidine tag (SEQ ID NO: 292), which can be
detected using an anti-His antibody by Western blot. The gel shown
in FIG. 5 was electroblot transferred to a nylon membrane and a
Western blot was performed using an anti-His antibody. Results are
shown in FIG. 6. All of the recombinant proteins were recognized by
the anti-His antibody, indicating that the desired recombinant
protein species were produced and purified.
[0221] Large scale preparations of the recombinant proteins can be
better expressed using bacterial fermentation procedures well known
to those with skill in the art. Heat treatment followed by
centrifugation to sediment denatured proteins will provide
substantial purification and final purification can be accomplished
using size exclusion or anion exchange chromatography without the
need for a His-tag or use of Ni-affinity chromatography.
Example 3
RNase H2 Activity for the Recombinant Peptides
[0222] The following example demonstrates RNase H2 activity for the
recombinant peptides.
[0223] RNase H enzymes cleave RNA residues in an RNA/DNA
heteroduplex. All RNase H enzymes can cleave substrates of this
kind when at least 4 sequential RNA residues are present. RNase H1
enzymes rapidly lose activity as the RNA "window" of a chimeric
RNA/DNA species is shortened to less than 4 residues. RNase H2
enzymes, on the other hand, are capable of cleaving an RNA/DNA
heteroduplex containing only a single RNA residue. In all cases,
the cleavage products contain a 3'-hydroxyl and a 5'-phosphate (see
FIG. 1). When multiple RNA residues are present, cleavage occurs
between RNA bases, cleaving an RNA-phosphate linkage. When only a
single RNA residue is present, cleavage occurs only with Type II
RNase H enzymes. In this case cleavage occurs on the 5'-side of the
RNA base at a DNA-phosphate linkage (see FIG. 3). RNase H enzymes
require the presence of a divalent metal ion cofactor. Typically,
RNase H1 enzymes require the presence of Mg.sup.++ ions while RNase
H2 enzymes can function with any of a number of divalent cations,
including but not limited to Mg.sup.++, Mn.sup.++, Co.sup.++ and
Ni.sup.++.
[0224] The recombinant RNase H2 proteins described in Example 2
were tested for both types of RNase H activity and were examined
for the characteristics listed above.
[0225] Cleavage of a Substrate with Multiple RNA Bases.
[0226] The following synthetic 30 bp substrate was used to test the
activity of these enzymes for cleavage of a long RNA domain. The
substrate is an "11-8-11" design, having 11 DNA bases, 8 RNA bases,
and 11 DNA bases on one strand and a perfect match DNA complement
as the other strand. The oligonucleotides employed are indicated
below, where upper case letters represent DNA bases and lower case
letters represent RNA bases.
[0227] SEQ ID NO: 6
TABLE-US-00012 5'-CTCGTGAGGTGaugcaggaGATGGGAGGCG-3'
[0228] SEQ ID NO: 7
TABLE-US-00013 5'-CGCCTCCCATCTCCTGCATCACCTCACGAG-3'
[0229] When annealed, these single-stranded (ss) oligonucleotides
form the following "11-8-11" double-stranded (ds) substrate:
[0230] SEQ ID NOS 6 and 7, respectively, in order of appearance
TABLE-US-00014 5'-CTCGTGAGGTGaugcaggaGATGGGAGGCG-3'
3'-GAGCACTCCACTACGTCCTCTACCCTCCGC-5'
[0231] Aliquots of each of the recombinant protein products were
incubated with single-stranded or double-stranded oligonucleotide
substrates in an 80 .mu.l reaction volume in buffer 50 mM NaCl, 10
mM MgCl.sub.2, and 10 mM Tris pH 8.0 for 20 minutes at 45.degree.
C. or 70.degree. C. Reactions were stopped with the addition of gel
loading buffer (formamide/EDTA) and separated on a denaturing 7M
urea, 15% polyacrylamide gel. Gels were stained using GelStar.TM.
(Lonza, Rockland, Me.) and visualized with UV excitation. All 5
recombinant peptides showed the ability to cleave an 8 base RNA
sequence in an RNA/DNA heteroduplex (11-8-11) substrate.
Importantly, the recombinant proteins did not degrade the single
stranded RNA-containing oligonucleotide (SEQ ID No. 6), indicating
that a double-stranded substrate was required. Further, a dsDNA
substrate was not cleaved.
[0232] Cleavage was not observed in the absence of a divalent
cation (e.g., no activity was observed if Mg.sup.++ was absent from
the reaction buffer). A Mg.sup.++ titration was performed and high
enzyme activity was observed between 2-8 mM MgCl.sub.2.
[0233] Optimal activity was observed between 3-6 mM MgCl.sub.2.
Cleavage activity was also detected using other divalent cations
including Mn.sup.++ and Co.sup.++. In MnCl.sub.2, good activity was
seen from 0.3 mM to 10 mM divalent cation concentration. Enzyme
activity was optimal in the range of 300 nM to 1 mM. For
CoCl.sub.2, activity was seen in the range of 0.3 mM to 2 mM, with
optimal activity in the range of 0.5-1 mM. The isolated enzymes
therefore show RNase H activity, and divalent cation requirements
that are characteristic of the RNase H2 class.
[0234] Digestion of the 11-8-11 substrate by recombinant RNase H2
enzymes from Pyrococcus kodakaraensis, Pyrococcus furiosus, and
Pyrococcus abyssi is shown in FIG. 7.
[0235] Substrate cleavage by RNase H enzymes is expected to result
in products with a 3'-OH and 5'-phosphate. The identity of the
reaction products from the new recombinant RNase H2 proteins was
examined by mass spectrometry. Electrospray ionization mass
spectrometry (ESI-MS) has near single Dalton resolution for nucleic
acid fragments of this size (accuracy of +/-0.02%). The
oligonucleotide 11-8-11 substrate (SEQ ID NOS 6 and 7) was examined
by ESI-MS both before and after digestion with the three Pyrococcus
sp. RNase H enzymes. The primary masses observed are reported in
Table 7 along with identification of nucleic acid species
consistent with the observed masses.
TABLE-US-00015 TABLE 7 Mass of species observed after RNase H2
digestion of SEQ ID No. 6 and 7 RNase H2 Predicted Observed
Treatment Sequence Mol Wt Mol Wt None 5' CTCGTGAGGTGa 9547 9548
(control) ugcaggaGATGGGAG GCG (SEQ ID NO: 6) 3' GAGCACTCCACT 8984
8984 ACGTCCTCTACCCTC CGC (SEQ ID NO: 7) Pyrococcus 5' CTCGTGAGGTGa
3717 3719 kodakaraensis (SEQ ID NO: 9) 5' P-aGATGGGAGG 3871 3871 CG
(SEQ ID NO: 10) 3' GAGCACTCCACT 8984 8984 ACGTCCTCTACCCTC CGC (SEQ
ID NO: 7) Pyrococcus 5' CTCGTGAGGTGa 3717 3719 furiosus (SEQ ID NO:
9) 5' P-aGATGGGAGG 3871 3872 CG (SEQ ID NO: 10) 3' GAGCACTCCACT
8984 8984 ACGTCCTCTACCCTC CGC (SEQ ID NO: 7) Pyrococcus 5'
CTCGTGAGGTGa 3717 3719 abyssi (SEQ ID NO: 9) 5' P-aGATGGGAGG 3871
3872 CG (SEQ ID NO: 10) 3' GAGCACTCCACT 8984 8984 ACGTCCTCTACCCTC
CGC (SEQ ID NO: 7)
[0236] Major species identified are shown. DNA bases are indicated
with upper case letters, RNA bases are indicated with lower case
letters, and phosphate="P". Molecular weights are rounded to the
nearest Dalton. In the absence of other notation, the nucleic acids
strands end in a 5'-hydroxyl or 3'-hydroxyl.
[0237] In all cases, the DNA complement strand was observed intact
(non-degraded). The RNA-containing strands were efficiently cleaved
and the observed masses of the reaction products are consistent
with the following species being the primary fragments produced: 1)
a species which contained undigested DNA residues and a single
3'-RNA residue with a 3'-hydroxyl groups (SEQ ID No. 9), and 2) a
species with a 5'-phosphate, a single 5'-RNA residue, and
undigested DNA residues (SEQ ID No. 10). The observed reaction
products are consistent with the known cleavage properties of both
RNase H1 and RNase H2 enzymes.
[0238] SEQ ID NO: 9
TABLE-US-00016 5' CTCGTGAGGTGa 3'
[0239] SEQ ID NO: 10
TABLE-US-00017 5' P-aGATGGGAGGCG 3'
[0240] Cleavage of a Substrate with a Single RNA Base.
[0241] RNase H2 enzymes characteristically cleave a substrate that
contains a single RNA residue while RNase H1 enzymes cannot. The
following synthetic 30 bp substrates were used to test the activity
of these enzymes for cleavage at a single RNA residue. The
substrates are a "14-1-15" design, having 14 DNA bases, 1 RNA base,
and 15 DNA bases on one strand and a perfect match DNA complement
as the other strand. Four different substrates were made from 8
component single-stranded oligonucleotides comprising each of the 4
RNA bases: C, G, A, and U. The oligonucleotides employed are
indicated below, where upper case letters represent DNA bases and
lower case letters represent RNA bases.
For rC:
[0242] SEQ ID NO: 11
TABLE-US-00018 5'-CTCGTGAGGTGATGcAGGAGATGGGAGGCG-3'
[0243] SEQ ID NO: 12
TABLE-US-00019 5'-CGCCTCCCATCTCCTGCATCACCTCACGAG-3'
[0244] When annealed, these single-stranded (ss) oligonucleotides
form the following "14-1-15 rC" double-stranded (ds) substrate:
[0245] SEQ ID NOS 11 and 12, respectively, in order of
appearance
TABLE-US-00020 5'-CTCGTGAGGTGATGcAGGAGATGGGAGGCG-3'
3'-GAGCACTCCACTACGTCCTCTACCCTCCGC-5'
For rG:
[0246] SEQ ID NO: 14
TABLE-US-00021 5'-CTCGTGAGGTGATGgAGGAGATGGGAGGCG-3'
[0247] SEQ ID NO: 15
TABLE-US-00022 5'-CGCCTCCCATCTCCTCCATCACCTCACGAG-3'
[0248] When annealed, these single-stranded (ss) oligonucleotides
form the following "14-1-15 rG" double-stranded (ds) substrate:
[0249] SEQ ID NOS 14 and 15, respectively, in order of
appearance
TABLE-US-00023 5'-CTCGTGAGGTGATGgAGGAGATGGGAGGCG-3'
3'-GAGCACTCCACTACCTCCTCTACCCTCCGC-5'
For rA:
[0250] SEQ ID NO: 17
TABLE-US-00024 5'-CTCGTGAGGTGATGaAGGAGATGGGAGGCG-3'
[0251] SEQ ID NO: 18
TABLE-US-00025 5'-CGCCTCCCATCTCCTTCATCACCTCACGAG-3'
[0252] When annealed, these single-stranded (ss) oligonucleotides
form the following "14-1-15 rA" double-stranded (ds) substrate:
[0253] SEQ ID NOS 17 and 18, respectively, in order of
appearance
TABLE-US-00026 5'-CTCGTGAGGTGATGaAGGAGATGGGAGGCG-3'
3'-GAGCACTCCACTACTTCCTCTACCCTCCGC-5'
For rU:
[0254] SEQ ID NO: 20
TABLE-US-00027 5'-CTCGTGAGGTGATGuAGGAGATGGGAGGCG-3'
[0255] SEQ ID NO: 21
TABLE-US-00028 5'-CGCCTCCCATCTCCTACATCACCTCACGAG-3'
[0256] When annealed, these single-stranded (ss) oligonucleotides
form the following "14-1-15 rU" double-stranded (ds) substrate:
[0257] SEQ ID NOS 20 and 21, respectively, in order of
appearance
TABLE-US-00029 5'-CTCGTGAGGTGATGuAGGAGATGGGAGGCG-3'
3'-GAGCACTCCACTACATCCTCTACCCTCCGC-5'
[0258] Aliquots of each of the recombinant protein products were
incubated with the single-stranded and double-stranded
oligonucleotide substrates indicated above in an 80 .mu.l reaction
volume in buffer 50 mM NaCl, 10 mM MgCl.sub.2, and 10 mM Tris pH
8.0 for 20 minutes at 70.degree. C. Reactions were stopped with the
addition of gel loading buffer (formamide/EDTA) and separated on a
denaturing 7M urea, 15% polyacrylamide gel. Gels were stained using
GelStar.TM. (Lonza, Rockland, Me.) and visualized with UV
excitation. All 5 recombinant peptides showed the ability to cleave
a single RNA base in an RNA/DNA heteroduplex (14-1-15). Each of the
4 RNA bases functioned as a cleavable substrate with these enzymes.
Importantly, the recombinant proteins did not degrade the single
stranded RNA-containing oligonucleotides (SEQ ID Nos. 11, 14, 17,
20), indicating that a double-stranded substrate was required. The
isolated enzymes therefore show RNase H2 activity. Titration of
divalent cations was tested and results were identical to those
obtained previously using the 8-11-8 substrate.
[0259] Digestion of the four 14-1-15 substrates (SEQ ID NOS 11-12,
14-15, 17-18 and 20-21) and the 11-8-11 substrate (SEQ ID NOS 6 and
7) by recombinant RNase H2 enzymes from Pyrococcus abyssi,
Pyrococcus furiosus, and Methanocaldococcus jannaschii is shown in
FIG. 8A and from Pyrococcus kodakaraensis in FIG. 8B.
[0260] Substrate cleavage by RNase H enzymes is expected to result
in products with a 3'-OH and 5'-phosphate. Further, cleavage of a
substrate containing a single ribonucleotide by RNase H2 enzymes
characteristically occurs at the DNA linkage 5'- to the RNA
residue. The identity of the reaction products using a single
ribonucleotide substrate from the new recombinant RNase H2 proteins
was examined by mass spectrometry. The oligonucleotide 14-1-15 rC
substrate (SEQ ID NOS 11 and 12) was examined by ESI-MS both before
and after digestion with the three Pyrococcus sp. RNase H2 enzymes
and the Methanocaldococcus jannaschii enzyme. The primary masses
observed are reported in Table 8 along with identification of
nucleic acid species consistent with the observed masses.
TABLE-US-00030 TABLE 8 Mass of species observed after RNase H2
digestion of SEQ ID NOS 11 and 12 RNase H2 Predicted Observed
Treatment Sequence Mol Wt Mol Wt None 5' CTCGTGAGGTGATG 9449 9450
(control) cAGGAGATGGGAGGCG (SEQ ID NO: 11) 3' GAGCACTCCACTAC 8984
8984 GTCCTCTACCCTCCGC (SEQ ID NO: 12) Pyrococcus 5' CTCGTGAGGTGATG
4334 4335 kodakaraensis (SEQ ID NO: 23) 5' P-cAGGAGATGGGA 5132 5133
GGCG (SEQ ID NO: 24) 3' GAGCACTCCACTAC 8984 8984 GTCCTCTACCCTCCGC
(SEQ ID NO: 12) Pyrococcus 5' CTCGTGAGGTGATG 4334 4335 furiosus
(SEQ ID NO: 23) 5' P-cAGGAGATGGGA 5132 5132 GGCG (SEQ ID NO: 24) 3'
GAGCACTCCACTAC 8984 8984 GTCCTCTACCCTCCGC (SEQ ID NO: 12)
Pyrococcus 5' CTCGTGAGGTGATG 4334 4335 abyssi (SEQ ID NO: 23) 5'
P-cAGGAGATGGGA 5132 5133 GGCG (SEQ ID NO: 24) 3' GAGCACTCCACTAC
8984 8984 GTCCTCTACCCTCCGC (SEQ ID NO: 12) Methano- 5'
CTCGTGAGGTGATG 4334 4335 caldococcus (SEQ ID NO: 23) jannaschii 5'
P-cAGGAGATGGGA 5132 5133 GGCG (SEQ ID NO: 24) 3' GAGCACTCCACTAC
8984 8984 GTCCTCTACCCTCCGC (SEQ ID NO: 12)
[0261] Major species identified are shown. DNA bases are indicated
with upper case letters, RNA bases are indicated with lower case
letters, and phosphate="P". Molecular weights are rounded to the
nearest Dalton. In the absence of other notation, the nucleic acids
strands end in a 5'-hydroxyl or 3'-hydroxyl.
[0262] In all cases, the DNA complement strand was observed intact
(non-degraded). The RNA-containing strands were efficiently cleaved
and the observed masses of the reaction products are consistent
with the following species being the primary fragments produced: 1)
a species which contained undigested DNA residues with a
3'-hydroxyl (SEQ ID No. 23), and 2) a species with a 5'-phosphate,
a single 5'-RNA residue, and undigested DNA residues (SEQ ID No.
24). The observed reaction products are consistent with the known
cleavage properties of RNase H2 class enzymes.
[0263] SEQ ID NO: 23
TABLE-US-00031 5' CTCGTGAGGTGATG 3'
[0264] SEQ ID NO: 24
TABLE-US-00032 5' P-cAGGAGATGGGAGGCG 3'
[0265] In summary, the cloned, codon-optimized rnhb genes predicted
to encode RNase H2 enzymes from 5 Archaeal species all produced
recombinant protein products which displayed enzyme activities
consistent with that expected for members of the RNase H2 family.
1) The enzymes required divalent cation to function (the
experiments presented here were done using Mg.sup.++). Activity is
also present using Mn.sup.++ or Co.sup.++ ions; 2) Single-stranded
nucleic acids are not degraded; 3) Double-stranded heteroduplex
nucleic acids are substrates where one strand contains one or more
RNA bases; 4) For substrates containing 2 or more consecutive RNA
bases, cleavage occurs in a DNA-RNA-DNA chimera between RNA
linkages; for substrates containing a single RNA base, cleavage
occurs immediately 5'- to the RNA base in a DNA-RNA-DNA at a DNA
linkage; and 6) Reaction products have a 3-hydroxyl and
5'-phosphate.
Example 4
Reaction Temperature Optimization and Thermal Stability of
Pyrococcus abyssi RNase H2
[0266] For this example and all subsequent work, the amount of the
enzyme employed was standardized based upon the following unit
definition, where: [0267] 1 unit is defined as the amount of enzyme
that results in the cleavage of 1 nmole of a heteroduplex substrate
containing a single rC residue per minute at 70.degree. C. in a
buffer containing 4 mM Mg.sup.2+ at pH 8.0.
[0268] Substrate SEQ ID NOS 11 and 12 were employed for
characterizing RNase H2 enzyme preparation for the purpose of
normalizing unit concentration. The following standardized buffer
was employed unless otherwise noted. "Mg Cleavage Buffer": 4 mM
MgCl.sub.2, 10 mM Tris pH 8.0, 50 mM NaCl, 10 .mu.g/ml BSA (bovine
serum albumin), and 300 nM oligo-dT (20 mer poly-dT
oligonucleotide). The BSA and oligo-dT serve to saturate
non-specific binding sites on plastic tubes and improve the
quantitative nature of assays performed.
[0269] Purified recombinant Pyrococcus abyssi RNase H2 enzyme was
studied for thermal stability. Aliquots of enzyme were incubated at
95.degree. C. for various periods of time and then used to cleave
the single rC containing substrate SEQ ID NOS 11 and 12. The RNA
strand of the substrate was radiolabeled with .sup.32P using 6000
Ci/mmol .gamma.-.sup.32P-ATP and the enzyme T4 Polynucleotide
Kinase (Optikinase, US Biochemical). Trace label was added to
reaction mixtures (1:50). Reactions were performed using 100 nM
substrate with 100 microunits (.mu.U) of enzyme in Mg Cleavage
Buffer. Reactions were incubated at 70.degree. C. for 20 minutes.
Reaction products were separated using denaturing 7M urea, 15%
polyacrylamide gel electrophoresis (PAGE) and visualized using a
Packard Cyclone.TM. Storage Phosphor System (phosphorimager). The
relative intensity of each band was quantified using the
manufacturer's image analysis software and results plotted as a
fraction of total substrate cleaved. Results are shown in FIG. 9.
The enzyme retained full activity for over 30 minutes at 95.degree.
C. Activity was reduced to 50% after 45 minutes incubation and to
10% after 85 minutes incubation.
[0270] These results demonstrate that the Pyrococcus abyssi RNase
H2 enzyme is sufficiently thermostable to survive prolonged
incubation at 95.degree. C. and would therefore survive conditions
typically employed in PCR reactions.
[0271] The temperature dependence of the activity of the Pyrococcus
abyssi RNase H2 enzyme was next characterized. The activity was
studied over a 40.degree. C. temperature range from 30.degree. C.
to 70.degree. C. The RNA strand of the rC substrate SEQ ID NOS 11
and 12 was radiolabeled as described above. Reactions were
performed using 100 nM substrate with 200 microunits (.mu.U) of
enzyme in Mg Cleavage Buffer. Reactions were incubated at
30.degree. C., 40.degree. C., 50.degree. C., 60.degree. C., or
70.degree. C. for 10 minutes. Reactions were stopped with the
addition of cold EDTA containing formamide gel loading buffer.
Reaction products were then separated using denaturing 7M urea, 15%
polyacrylamide gel electrophoresis (PAGE) and visualized using a
Packard Cyclone.TM. Storage Phosphor System (phosphorimager). The
resulting gel image is shown in FIG. 10. The relative intensity of
each band was quantified using the manufacturer's image analysis
software and results plotted as a fraction of total substrate
cleaved (see FIG. 11). The enzyme shows only .about.0.1% activity
at 30.degree. C. and does not attain appreciable activity until
about 50 to 60.degree. C.
[0272] Therefore, for practical purposes the enzyme is functionally
inactive at room temperature. Reactions employing this enzyme can
therefore be set up on ice or even at room temperature and the
reactions will not proceed until temperature is elevated. If
Pyrococcus abyssi RNase H2 cleavage were linked to a PCR reaction,
the temperature dependent activity demonstrated herein would
effectively function to provide for a "hot start" reaction format
in the absence of a modified DNA polymerase.
Example 5
Cleavage at Non-Standard Bases by RNase H2
[0273] The natural biological substrates for RNase H1 and RNase H2
are duplex DNA sequences containing one or more RNA residues.
Modified bases containing substitutions at the 2'-position other
than hydroxyl (RNA) have not been observed to be substrates for
these enzymes. The following example demonstrates that the
Pyrococcus abyssi RNase H2 enzyme has activity against modified
RNA-containing substrates.
[0274] The following 14-1-15 substrates containing modified bases
were tested to determine if RNase H2 could recognize single non-RNA
2'-modified bases as sites for cleavage. The modifications are
located on the 2' position of the base and include locked nucleic
acid (LNA), 2'-O-methyl (2'OMe), and 2'-fluoro (2'F); the single
ribo-C containing substrate was employed as positive control.
Hereafter, LNA bases will be designated with a "+" prefix (+N),
2'OMe bases will be designated with a "m" prefix (mN), 2'F bases
will be designated with a "f" prefix (fN), and 2'-amino bases with
an "a" prefix (aN).
Ribo-C Substrate
[0275] SEQ ID NOs 11 and 12, respectively, in order of
appearance
TABLE-US-00033 5'-CTCGTGAGGTGATGcAGGAGATGGGAGGCG-3'
3'-GAGCACTCCACTACGTCCTCTACCCTCCGC-5'
LNA-C Substrate
[0276] SEQ ID NOs 25 and 293, respectively, in order of
appearance
TABLE-US-00034 5'-CTCGTGAGGTGATG(+C)AGGAGATGGGAGGCG-3'
3'-GAGCACTCCACTAC G TCCTCTACCCTCCGC-5'
2'OMe-C Substrate
[0277] SEQ ID NOs 26 and 293, respectively, in order of
appearance
TABLE-US-00035 5'-CTCGTGAGGTGATG(mC)AGGAGATGGGAGGCG-3'
3'-GAGCACTCCACTAC G TCCTCTACCCTCCGC-5'
2'F-C Substrate
[0278] SEQ ID NOs 27 and 293, respectively, in order of
appearance
TABLE-US-00036 5'-CTCGTGAGGTGATG(fC)AGGAGATGGGAGGCG-3'
3'-GAGCACTCCACTAC G TCCTCTACCCTCCGC-5'
[0279] The above 4 substrates were incubated in an 80 .mu.l
reaction volume in various buffers for 20 minutes at 70.degree. C.
with the recombinant Pyrococcus abyssi RNase H2 enzyme. Buffers
tested included 50 mM NaCl, 10 mM Tris pH 8.0 with either 10 mM
MgCl.sub.2, 10 mM CoCl.sub.2, or 10 mM MnCl.sub.2. Reactions were
stopped with the addition of gel loading buffer (formamide/EDTA)
and separated on a denaturing 7M urea, 15% polyacrylamide gel. Gels
were stained using GelStar.TM. (Lonza, Rockland, Me.) and
visualized with UV excitation. Results are shown in FIG. 12. The
control substrate with a single ribo-C residue was 100% cleaved.
The substrates containing a single LNA-C or a single 2'OMe-C
residue were not cleaved. However, the substrate containing a
single 2'-F-C residue was cleaved to a small extent. This cleavage
occurred only in the manganese containing buffer and was not seen
in either cobalt or magnesium buffers.
[0280] Cleavage at a 2'-F-C base was unexpected. Cleavage of
2'-fluoro bases was investigated further using the following
substrates.
[0281] Ribo-C Substrate
[0282] SEQ ID NOs 11 and 12, respectively, in order of
appearance
TABLE-US-00037 5'-CTCGTGAGGTGATGcAGGAGATGGGAGGCG-3'
3'-GAGCACTCCACTACGTCCTCTACCCTCCGC-5'
2'F-C Substrate
[0283] SEQ ID NOs 27 and 293, respectively, in order of
appearance
TABLE-US-00038 5'-CTCGTGAGGTGATG(fC)AGGAGATGGGAGGCG-3'
3'-GAGCACTCCACTAC G TCCTCTACCCTCCGC-5'
2'F-U Substrate
[0284] SEQ ID NOs 28 and 294, respectively, in order of
appearance
TABLE-US-00039 5'-CTCGTGAGGTGATG(fU)AGGAGATGGGAGGCG-3'
3'-GAGCACTCCACTAC A TCCTCTACCCTCCGC-5'
2'F-C+2'FU (fCfU) Substrate
[0285] SEQ ID NOS 29 and 295, respectively, in order of
appearance
TABLE-US-00040 5'-CTCGTGAGGTGATG(fCfU)GGAGATGGGAGGCG-3'
3'-GAGCACTCCACTAC G A CCTCTACCCTCCGC-5'
[0286] The above 4 substrates were incubated in an 80 .mu.l
reaction volume in a buffer containing 50 mM NaCl, 10 mM Tris pH
8.0 and 10 mM MnCl.sub.2 for 20 minutes at 70.degree. C. with
either the recombinant Pyrococcus abyssi RNase H2 enzyme or the
recombinant Pyrococcus furiosus RNase H2 enzyme. Reactions were
stopped with the addition of gel loading buffer (formamide/EDTA)
and separated on a denaturing 7M urea, 15% polyacrylamide gel. Gels
were stained using GelStar.TM. (Lonza, Rockland, Me.) and
visualized with UV excitation. Results are shown in FIG. 13. The
control substrate with a single ribo-C residue was 100% cleaved.
The substrates containing a single 2'-F-C or single 2'-F-U residue
were cleaved to a small extent. The di-fluoro substrate containing
adjacent 2'-F-C and 2'-F-U residues (fCfU) was cleaved nearly 100%.
Further, both the Pyrococcus abyssi and Pyrococcus furiosus RNase
H2 enzymes cleaved the modified substrate in an identical fashion.
This example demonstrates that the unexpected cleavage of the fC
group was not restricted to fC but also occurred with fU. More
importantly, a combination of 2 sequential 2'-fluoro modified bases
was a far better substrate for RNase H2. This novel cleavage
property was seen for both the P. abyssi and P. furiosus enzymes.
Cleavage of such atypical substrates may be a property common to
all Archaeal RNase H2 enzymes.
[0287] The identity of the cleavage products of the di-fluoro fCfU
substrate was studied using mass spectrometry using the methods
described in Example 3. Using traditional ribonucleotide
substrates, cleavage by RNase H enzymes results in products with a
3'-OH and 5'-phosphate. The fCfU substrate (SEQ ID NOS 29 and 295)
were examined by ESI-MS both before and after digestion by the
recombinant Pyrococcus abyssi RNase H2 enzyme. The primary masses
observed are reported in Table 9 along with identification of
nucleic acid species consistent with the observed masses.
TABLE-US-00041 TABLE 9 Mass of species observed after RNase H2
digestion of SEQ ID NOs 29 and 295 RNase H2 Predicted Observed
Treatment Sequence Mol Wt Mol Wt None 5'-CTCGTGAGGTGATG(fCfU) 9446
9446 (control) GGAGATGGGAGGCG-3' (SEQ ID NO: 29) 3'-GAGCACTCCACTAC
G A 8993 8994 CCTCTACCCTCCGC-5' (SEQ ID NO: 295) Pyrococcus 5'
CTCGTGAGGTGATG(fC) 4642 4643 abyssi (SEQ ID NO: 296) 5'
P-(fU)GGAGATGGGAGGCG 4822 4823 (SEQ ID NO: 297) 3' GAGCACTCCACTAC G
A 8993 8994 CCTCTACCCTCCGC (SEQ ID NO: 295) Major species
identified are shown. DNA bases are indicated with upper case
letters, 2'-F bases are indicated as fC or fU, and phosphate = "P".
Molecular weights are rounded to the nearest Dalton. In the absence
of other notation, the nucleic acids strands end in a 5'-hydroxyl
or 3'-hydroxyl.
[0288] The mass spectrometry data indicates that digestion of a
di-fluoro substrate such as the fCfU duplex studied above by RNase
H2 results in cleavage between the two fluoro bases. Further, the
reaction products contain a 3'-hydroxyl and 5'-phosphate, similar
to the products resulting from digestion of RNA containing
substrates.
[0289] Cleavage of the modified bases was not observed in the
absence of a divalent cation. A titration was performed and enzyme
activity was observed between 0.25-10 mM MnCl.sub.2 and 0.25-1.5 mM
CoCl.sub.2. Enzyme activity was optimal in the range of 0.5 mM to 1
mM for both MnCl.sub.2 and CoCl.sub.2. Hereafter 0.6 mM MnCl.sub.2
was employed in reactions or 0.5 mM CoCl.sub.2. Reduced activity
for cleavage of the modified substrate was observed using Mg
buffers. Overall, optimum activity was observed using Mn buffers
for cleavage of the di-fluoro (fNfN) substrates whereas Mg buffers
were superior for cleavage of ribonucleotide (rN) substrates.
[0290] The ability of the RNase H2 enzymes to cleave at single or
double 2'-F bases was unexpected. The Pyrococcus abyssi RNase H2
enzyme was next tested for the ability to cleave a greater variety
of modified substrates using the same methods described above in
this example. The modified strand of the substrate was radiolabeled
as described above. Reactions were performed using 100 nM substrate
and 480-1000 mU of recombinant enzyme in Mn Cleavage Buffer (10 mM
Tris pH 8.0, 50 mM NaCl, 0.6 mM MnCl.sub.2, 10 .mu.g/ml BSA).
Reactions were incubated at 70.degree. C. for 20 minutes. Reaction
products were separated using denaturing 7M urea, 15%
polyacrylamide gel electrophoresis (PAGE) and visualized using a
Packard Cyclone.TM. Storage Phosphor System (phosphorimager). The
relative intensity of each band was quantified using the
manufacturer's image analysis software and results plotted as a
fraction of total substrate cleaved are shown in Table 10.
TABLE-US-00042 TABLE 10 Cleavage of substrates containing
2'-modification by Pyrococcus abyssi RNase H2 using increased
amounts of enzyme 2'-Mod Oligo Sequence SEQ ID No. Cleavage fN-fN
5'-CTCGTGAGGTGAT(fNfN)AGGAGATGGGAGGCG-3' SEQ ID No. 30 +++++
3'-GAGCACTCCACTA N N TCCTCTACCCTCCGC-5' SEQ ID No. 298 fU-LNA-C
5'-CTCGTGAGGTGAT(fU+C)AGGAGATGGGAGGCG-3' SEQ ID No. 31 ++++
3'-GAGCACTCCACTA A G TCCTCTACCCTCCGC-5' SEQ ID No. 299 mU-fC
5'-CTCGTGAGGTGAT(mUfC)AGGAGATGGGAGGCG-3' SEQ ID No. 32 +++
3'-GAGCACTCCACTA A G TCCTCTACCCTCCGC-5' SEQ ID No. 299 mU-LNA-C
5'-CTCGTGAGGTGAT(mU+C)AGGAGATGGGAGGCG-3' SEQ ID No. 33 ++
3'-GAGCACTCCACTA A G TCCTCTACCCTCCGC-5' SEQ ID No. 299 mU-mN
5'-CTCGTGAGGTGAT(mUmN)AGGAGATGGGAGGCG-3' SEQ ID No. 34 ++
3'-GAGCACTCCACTA A N TCCTCTACCCTCCGC-5' SEQ ID No. 300 Amino-U-
5'-CTCGTGAGGTGAT(aU+C)AGGAGATGGGAGGCG-3' SEQ ID No. 35 + LNA-C
3'-GAGCACTCCACTA A G TCCTCTACCCTCCGC-5' SEQ ID No. 299 fN 5'-
CTCGTGAGGTGATG(fN)AGGAGATGGGAGGCG-3' SEQ ID No. 36 + 3'-
GAGCACTCCACTAC N TCCTCTACCCTCCGC-5' SEQ ID No. 301 mU-Amino-C
5'-CTCGTGAGGTGAT(mUaC)AGGAGATGGGAGGCG-3' SEQ ID No. 37 +/-
3'-GAGCACTCCACTA A G TCCTCTACCCTCCGC-5' SEQ ID No. 299 LNA-T-fC
5'-CTCGTGAGGTGAT(+TfC)AGGAGATGGGAGGCG-3' SEQ ID No. 38 -
3'-GAGCACTCCACTA A G TCCTCTACCCTCCGC-5' SEQ ID No. 299 Amino-U 5'-
CTCGTGAGGTGATG(aU)AGGAGATGGGAGGCG-3' SEQ ID No. 39 - 3'-
GAGCACTCCACTAC A TCCTCTACCCTCCGC-5' SEQ ID No. 294 LNA-T-
5'-CTCGTGAGGTGAT(+T+C)AGGAGATGGGAGGCG-3' SEQ ID No. 40 - LNA-C
3'-GAGCACTCCACTA A G TCCTCTACCCTCCGC-5' SEQ ID No. 299 fU-mC
5'-CTCGTGAGGTGAT(fUmC)AGGAGATGGGAGGCG-3' SEQ ID No. 41 -
3'-GAGCACTCCACTA A G TCCTCTACCCTCCGC-5' SEQ ID No. 299 LNA-T-mC
5'-CTCGTGAGGTGAT(+TmC)AGGAGATGGGAGGCG-3' SEQ ID No. 42 -
3'-GAGCACTCCACTA A G TCCTCTACCCTCCGC-5' SEQ ID No. 299 Uppercase
letters = DNA; fN = 2'-F bases, +N = LNA bases, mN = 2'OMe bases,
aN = 2'-amino bases Use of "N" base indicates that every possible
base (A, G, C, U/T) was tested with the appropriate perfect match
complement. Efficiency of cleavage was rated from "+++++" (100%
cleavage) to "-" (no cleavage). The mUmN substrates did not cleave
equally well and the "++" rating applies to the best cleaving
dinucleotide pair, mUmU. The rank order of cleavage for this
substrate design was mUmU > mUmA > mUmC > mUmG.
[0291] It is clear from the above results that many different
2'-modifications can be cleaved by RNase H2 enzymes that were not
heretofore appreciated. Of the 2'-modified substrates, the
di-fluoro compounds (those with 2 sequential 2'-fluoro bases) were
most active. Additional substrates were tested, including some with
3 or 4 sequential 2'-fluoro bases. No icrease in activity was seen
when increasing the 2'-fluoro content above 2 residues.
[0292] A similar series of experiments was performed using lower
amounts of enzyme. The experiment below was conducted using an
identical protocol except that 148 .mu.U of recombinant Pyrococcus
abyssi RNase H2 was employed instead of the 480 mU previously
employed (3000-fold less enzyme) and the buffer contained a mixture
of divalent cations (3 mM MgCl.sub.2+0.6 mM MnCl.sub.2). Under
these conditions, a substrate containing a single ribonucleotide
residue is completely cleaved whereas modified substrates are not.
Results are shown in Table 11. RNase H2 is more active in cleaving
substrates containing an RNA base than in cleaving the 2'-modified
bases.
TABLE-US-00043 TABLE 11 Cleavage of substrates containing
2'-modification by Pyrococcus abyssi RNase H2 using small amounts
of enzyme 2'-Mod Oligo Sequence SEQ ID NOS Cleavage rN
5'-CTCGTGAGGTGATGnAGGAGATGGGAGGCG-3' SEQ ID No. 43 +++++
3'-GAGCACTCCACTACNTCCTCTACCCTCCGC-5' SEQ ID No. 301 fU-rC
5'-CTCGTGAGGTGAT(fUrC)AGGAGATGGGAGGCG-3' SEQ ID No. 44 +++++
3'-GAGCACTCCACTA A G TCCTCTACCCTCCGC-5' SEQ ID No. 294 rU-fC
5'-CTCGTGAGGTGAT(rUfC)AGGAGATGGGAGGCG-3' SEQ ID No. 45 ++++
3'-GAGCACTCCACTA A G TCCTCTACCCTCCGC-5' SEQ ID No. 294 fN-fN
5'-CTCGTGAGGTGAT(fNfN)AGGAGATGGGAGGCG-3' SEQ ID No. 30 +
3'-GAGCACTCCACTA N N TCCTCTACCCTCCGC-5' SEQ ID No. 298 fN 5'-
CTCGTGAGGTGATG(fN)AGGAGATGGGAGGCG-3' SEQ ID No. 36 - 3'-
GAGCACTCCACTAC N TCCTCTACCCTCCGC-5' SEQ ID No. 301 Uppercase
letters = DNA; fN = 2'-F bases. Use of "N" base indicates that
every possible base (A, G, C, U/T) was tested with the appropriate
perfect match complement. Efficiency of cleavage was rated from
"+++++" (100% cleavage) to "-" (no cleavage).
[0293] Thus, Pyrococcus abyssi RNase H2 can be used to cleave
substrates which do not contain any RNA bases but instead contain
2'-modified bases. Of the compounds studied, di-fluoro (fNfN)
containing substrates performed best. Use of the modified
substrates generally requires increased amounts of enzyme, however
the enzyme is catalytically very potent and it presents no
difficulty to employ sufficient enzyme to achieve 100% cleavage of
a di-fluoro substrate.
[0294] The 2'-modified substrate described in this example are not
susceptible to cleavage by typical RNase enzymes. As such they can
be employed in novel assay formats where cleavage events are
mediated by RNase H2 using substrates that are completely resistant
to cleavage by other RNase enzymes, particularly single stranded
ribonucleases.
Example 6
Base Preferences for Cleavage of the Di-Fluoro fNfN Substrate
[0295] The following example demonstrates that all 16 possible
2'-fluoro dinucleotides can be cleaved by RNase H2. Distinct base
preferences are observed.
[0296] As shown in Table 12, the following 16 substrates were
synthesized and tested for efficiency of cleavage using the
recombinant Pyrococcus abyssi RNase H2 enzyme.
TABLE-US-00044 TABLE 12 fNfN Sequence SEQ ID No. AA
5'-CTCGTGAGGTGATG(fAfA)GGAGATGGGAGGCG-3' SEQ ID No. 46
3'-GAGCACTCCACTAC T T CCTCTACCCTCCGC-5' SED ID No. 332 AC
5'-CTCGTGAGGTGATG(fAfC)GGAGATGGGAGGCG-3' SEQ ID No. 47
3'-GAGCACTCCACTAC T G CCTCTACCCTCCGC-5' SED ID No. 333 AG
5'-CTCGTGAGGTGATG(fAfG)GGAGATGGGAGGCG-3' SEQ ID No. 48
3'-GAGCACTCCACTAC T C CCTCTACCCTCCGC-5' SED ID No. 334 AU
5'-CTCGTGAGGTGATG(fAfU)GGAGATGGGAGGCG-3' SEQ ID No. 49
3'-GAGCACTCCACTAC T A CCTCTACCCTCCGC-5' SED ID No. 335 CA
5'-CTCGTGAGGTGATG(fCfA)GGAGATGGGAGGCG-3' SEQ ID No. 50
3'-GAGCACTCCACTAC G T CCTCTACCCTCCGC-5' SED ID No. 336 CC
5'-CTCGTGAGGTGATG(fCfC)GGAGATGGGAGGCG-3' SEQ ID No. 51
3'-GAGCACTCCACTAC G G CCTCTACCCTCCGC-5' SED ID No. 337 CG
5'-CTCGTGAGGTGATG(fCfG)GGAGATGGGAGGCG-3' SEQ ID No. 52
3'-GAGCACTCCACTAC G C CCTCTACCCTCCGC-5' SED ID No. 338 CU
5'-CTCGTGAGGTGATG(fCfU)GGAGATGGGAGGCG-3' SEQ ID No. 29
3'-GAGCACTCCACTAC G A CCTCTACCCTCCGC-5' SED ID No. 295 GA
5'-CTCGTGAGGTGATG(fGfA)GGAGATGGGAGGCG-3' SEQ ID No. 53
3'-GAGCACTCCACTAC C T CCTCTACCCTCCGC-5' SED ID No. 339 GC
5'-CTCGTGAGGTGATG(fGfC)GGAGATGGGAGGCG-3' SEQ ID No. 54
3'-GAGCACTCCACTAC C G CCTCTACCCTCCGC-5' SED ID No. 340 GG
5'-CTCGTGAGGTGATG(fGfG)GGAGATGGGAGGCG-3' SEQ ID No. 55
3'-GAGCACTCCACTAC C C CCTCTACCCTCCGC-5' SED ID No. 341 GU
5'-CTCGTGAGGTGATG(fGfU)GGAGATGGGAGGCG-3' SEQ ID No. 56
3'-GAGCACTCCACTAC C A CCTCTACCCTCCGC-5' SED ID No. 342 UA
5'-CTCGTGAGGTGATG(fUfA)GGAGATGGGAGGCG-3' SEQ ID No. 57
3'-GAGCACTCCACTAC A T CCTCTACCCTCCGC-5' SED ID No. 343 UC
5'-CTCGTGAGGTGATG(fUfC)GGAGATGGGAGGCG-3' SEQ ID No. 58
3'-GAGCACTCCACTAC A G CCTCTACCCTCCGC-5' SED ID No. 344 UG
5'-CTCGTGAGGTGATG(fUfG)GGAGATGGGAGGCG-3' SEQ ID No. 59
3'-GAGCACTCCACTAC A C CCTCTACCCTCCGC-5' SED ID No. 345 UU
5'-CTCGTGAGGTGATG(fUfU)GGAGATGGGAGGCG-3' SEQ ID No. 60
3'-GAGCACTCCACTAC A A CCTCTACCCTCCGC-5' SED ID No. 346
[0297] The modified strand of each substrate was radiolabeled as
described above. Reactions were performed using 100 nM substrate
with 25 mU of recombinant enzyme in Mn Cleavage Buffer (10 mM Tris
pH 8.0, 50 mM NaCl, 0.6 mM MnCl.sub.2, 10 .mu.g/ml BSA). Reactions
were incubated at 70.degree. C. for 20 minutes. Reaction products
were separated using denaturing 7M urea, 15% polyacrylamide gel
electrophoresis (PAGE) and visualized using a Packard Cyclone.TM.
Storage Phosphor System (phosphorimager). The relative intensity of
each band was quantified, and results plotted as a fraction of
total substrate cleaved are shown in FIG. 14. The enzyme amount was
titrated so that the most active substrate cleaved at 90-95%
without having excess enzyme present so that accurate assessment
could be made of relative cleavage efficiency for less active
substrates.
[0298] All 16 dinucleotide fNfN pairs were cleaved by RNase H2,
however clear substrate preferences were observed. In general,
substrates having the sequence fNfU performed worse, indicating
that placement of a fU base at the 3'-position of the dinucleotide
pair was unfavorable. The least active substrate was fUfU, which
showed 10% cleavage under conditions that resulted in >90%
cleavage of fAfC or fAfG substrates.
[0299] Using greater amounts of enzyme, the relative differences of
cleavage efficiency between substrates is minor and 100% cleavage
can readily be achieved for all substrates studied here.
Example 7
Optimization of 3'- and 5'-Base Lengths for Cleavage of rN and fNfN
Substrates
[0300] The following example shows the optimization of the
placement of the cleavable domain relative to the 3' and 5' ends of
a primer or probe sequence. In the prior examples, the substrates
all had 14 or 15 DNA bases on both the 5'- and 3'-sides flanking
the cleavable domain. For use in designing cleavable probes and
primers, it may at times be beneficial to make these flanking
sequences as short as possible, in order to control Tm
(hybridization temperature) or to improve specificity of priming
reactions. It is therefore important to define the minimum length
of duplex needed to obtain efficient enzymatic cleavage.
[0301] In this experiment, the synthetic substrate duplexes shown
in Table 13 were made having a single rC cleavable base, a fixed
domain of 25 DNA bases 5'-flanking the ribonucleotide and a
variable number of bases on the 3'-side.
TABLE-US-00045 TABLE 13 3'-End Sequence (rC) SEQ ID NOS 3'-D1
5'-CTGAGCTTCATGCCTT SED ID No. 61 TACTGTCCTcT-3'
3'-GACTCGAAGTACGGAA SED ID No. 302 ATGACAGGACA-5' 3'-D2
5'-CTGAGCTTCATGCCTT SED ID No. 62 TACTGTCCTcTC-3'
3'-GACTCGAAGTACGGAA SED ID No. 303 ATGACAGGACAG-5' 3'-D3
5'-CTGAGCTTCATGCCTT SED ID No. 63 TACTGTCCTcTCC-3'
3'-GACTCGAAGTACGGAA SED ID No. 304 ATGACAGGACAGG-5' 3'-D5
5'-CTGAGCTTCATGCCTT SED ID No. 64 TACTGTCCTcTCCTT-3'
3'-GACTCGAAGTACGGAA SED ID No. 305 ATGACAGGACAGGAA-5' 3'-D6
5'-CTGAGCTTCATGCCTT SED ID No. 65 TACTGTCCTcTCCTTC-3'
3'-GACTCGAAGTACGGAA SED ID No. 306 ATGACAGGACAGGAAG-5'
[0302] The modified strand of each substrate was radiolabeled as
described above Reactions were performed using 100 nM substrate
with 100 .mu.U of recombinant enzyme in Mg Cleavage Buffer (10 mM
Tris pH 8.0, 50 mM NaCl, 4 mM MgCl.sub.2, 10 .mu.g/ml BSA).
Reactions were incubated at 70.degree. C. for 20 minutes. Reaction
products were separated using denaturing 7M urea, 15%
polyacrylamide gel electrophoresis (PAGE) and visualized using a
Packard Cyclone.TM. Storage Phosphor System (phosphorimager). The
relative intensity of each band was quantified, and results plotted
as a fraction of total substrate cleaved are shown in FIG. 15.
Maximal cleavage occurred with 4-5 DNA bases flanking the
ribonucleotide on the 3'-side.
[0303] In the next experiment, the synthetic substrate duplexes
shown in Table 14 were made having a single rU cleavable base with
a fixed domain of 25 base-pairs flanking the ribonucleotide on the
3' side and 2-14 base-pairs on the 5'-side. A minimum of 5 unpaired
bases (dangling ends) were left on the unmodified complement to
simulate hybridization to a long nucleic acid sample.
TABLE-US-00046 TABLE 14 5'-End Sequence (rU) SEQ ID NOS 5'-D1
5'-CuCCTGAGCTTCATGCCTTTA SED ID No. 66 CTGTCC-3'
3'-ACGTAGAATGGACAGAAGGAG SED ID No. 307 GACTCGAAGTACGGAAATGACAGG
ACGTA-5' 5'-D2 5'-CCuCCTGAGCTTCATGCCTTT SED ID No. 67 ACTGTCC-3'
3'-ACGTAGAATGGACAGAAGGAG SED ID No. 307 GACTCGAAGTACGGAAATGACAGG
ACGTA-5' 5'-D3 5'-TCCuCCTGAGCTTCATGCCTT SED ID No. 68 TACTGTCC-3'
3'-ACGTAGAATGGACAGAAGGAG SED ID No. 307 GACTCGAAGTACGGAAATGACAGG
ACGTA-5' 5'-D4 5'-TTCCuCCTGAGCTTCATGCCT SED ID No. 69 TTACTGTCC-3'
3'-ACGTAGAATGGACAGAAGGAG SED ID No. 307 GACTCGAAGTACGGAAATGACAGG
ACGTA-5' 5'-D5 5'-CTTCCuCCTGAGCTTCATGCC SED ID No. 70 TTTACTGTCC-3'
3'-ACGTAGAATGGACAGAAGGAG SED ID No. 307 GACTCGAAGTACGGAAATGACAGG
ACGTA-5' 5'-D6 5'-TCTTCCuCCTGAGCTTCATGC SED ID No. 71
CTTTACTGTCC-3' 3'-ACGTAGAATGGACAGAAGGAG SED ID No. 307
GACTCGAAGTACGGAAATGACAGG ACGTA-5' 5'-D8 5'-TGTCTTCCuCCTGAGCTTCAT
SED ID No. 72 GCCTTTACTGTCC-3' 3'-ACGTAGAATGGACAGAAGGAG SED ID No.
307 GACTCGAAGTACGGAAATGACAGG ACGTA-5' 5'-D10
5'-CCTGTCTTCCuCCTGAGCTTC SED ID No. 73 ATGCCTTTACTGTCC-3'
3'-ACGTAGAATGGACAGAAGGAG SED ID No. 307 GACTCGAAGTACGGAAATGACAGG
ACGTA-5' 5'-D12 5'-TACCTGTCTTCCuCCTGAGCT SED ID No. 74
TCATGCCTTTACTGTCC-3' 3'-ACGTAGAATGGACAGAAGGAG SED ID No. 307
GACTCGAAGTACGGAAATGACAGG ACGTA-5' 5'-D14 5'-CTTACCTGTCTTCCuCCTGAG
SED ID No. 75 CTTCATGCCTTTACTGTCC-3' 3'-ACGTAGAATGGACAGAAGGAG SED
ID No. 307 GACTCGAAGTACGGAAATGACAGG ACGTA-5'
[0304] The modified strand of each substrate was radiolabeled as
previously described. Reactions were performed using 100 nM
substrate with 123 .mu.U of recombinant enzyme in a mixed buffer
containing both Mg and Mn cations (10 mM Tris pH 8.0, 50 mM NaCl,
0.6 mM MnCl.sub.2, 3 mM MgCl.sub.2, 10 .mu.g/ml BSA). Reactions
were incubated at 70.degree. C. for 20 minutes. Reaction products
were separated using denaturing 7M urea, 15% polyacrylamide gel
electrophoresis (PAGE) and visualized using a Packard Cyclone.TM.
Storage Phosphor System (phosphorimager). The relative intensity of
each band was quantified and the results plotted as a fraction of
total substrate cleaved in FIG. 16. Little cleavage was seen with
the short substrates. Activity increased with length of the 5'-DNA
domain until maximum cleavage was obtained at around 10-12 bases of
duplex flanking the rU base on the 5'-side.
[0305] Similar experiments were done to determine the optimal
length of the 3'-DNA domain needed for cleavage of di-fluoro (fNfN)
substrates. The duplexes shown in Table 15 were synthesized and
tested to functionally define the length of DNA bases needed at the
3'-end of a fUfC di-fluoro substrate. A fixed domain of 22 base
pairs was positioned at the 5'-end and the 3'-domain was varied
from 2-14 bases.
TABLE-US-00047 TABLE 15 3'-End Sequence (fUfC) SEQ ID NOS 3'-D2
5'-CTGAGCTTCATGCCTTTACTGT(fUfC)CC-SpC3-3' SED ID No. 76
3'-GACTCGAAGTACGGAAATGACA A G GGGCTGTGTGTCGAG-5' SED ID No. 308
3'-D4 5'-CTGAGCTTCATGCCTTTACTGT(fUfC)CCCG-SpC3-3' SED ID No. 77
3'-GACTCGAAGTACGGAAATGACA A G GGGCTGTGTGTCGAG-5' SED ID No. 308
3'-D5 5'-CTGAGCTTCATGCCTTTACTGT(fUfC)CCCGA-SpC3-3' SED ID No. 78
3'-GACTCGAAGTACGGAAATGACA A G GGGCTGTGTGTCGAG-5' SED ID No. 308
3'-D6 5'-CTGAGCTTCATGCCTTTACTGT(fUfC)CCCGAC-SpC3-3' SED ID No. 79
3'-GACTCGAAGTACGGAAATGACA A G GGGCTGTGTGTCGAG-5' SED ID No. 308
3'-D8 5'-CTGAGCTTCATGCCTTTACTGT(fUfC)CCCGACAC-SpC3-3' SED ID No. 80
3'-GACTCGAAGTACGGAAATGACA A G GGGCTGTGTGTCGAG-5' SED ID No. 308
3'-D10 5'-CTGAGCTTCATGCCTTTACTGT(fUfC)CCCGACACAC-SpC3-3' SED ID No.
81 3'-GACTCGAAGTACGGAAATGACA A G GGGCTGTGTGTCGAG-5' SED ID No. 308
3'-D12 5'-CTGAGCTTCATGCCTTTACTGT(fUfC)CCCGACACACAG-SpC3-3' SED ID
No. 82 3'-GACTCGAAGTACGGAAATGACA A G GGGCTGTGTGTCGAG-5' SED ID No.
308 3'-D14 5'-CTGAGCTTCATGCCTTTACTGT(fUfC)CCCGACACACAGCT-SpC3-3'
SED ID No. 83 3'-GACTCGAAGTACGGAAATGACA A G GGGCTGTGTGTCGAG-5' SED
ID No. 308
[0306] The modified strand of each substrate was radiolabeled as
above. Reactions were performed using 100 nM substrate with 37 mU
of recombinant enzyme in a mixed buffer containing both Mg and Mn
cations (10 mM Tris pH 8.0, 50 mM NaCl, 0.6 mM MnCl.sub.2, 3 mM
MgCl.sub.2, 10 .mu.g/ml BSA). Reactions were incubated at
70.degree. C. for 20 minutes. Reaction products were separated
using denaturing 7M urea, 15% polyacrylamide gel electrophoresis
(PAGE) and visualized using a Packard Cyclone.TM. Storage Phosphor
System (phosphorimager). The relative intensity of each band was
quantified and the results plotted as a fraction of total substrate
cleaved in FIG. 17. No cleavage was seen with the substrate having
2 DNA bases on the 3'-side of the cleavable domain. Cleavage was
seen with 4 DNA bases and steadily increased until maximal cleavage
was obtained when 8-10 DNA bases were present on the 3'-side of the
fUfC cleavage domain. Interestingly, the optimal length of DNA
bases on the 3'-side of the cleavage domain is longer for the
di-fluoro substrates (8-10 bases) compared with the single
ribonucleotide substrates (4-5 bases).
[0307] In summary, for ribonucleotide containing substrates,
maximal cleavage activity is seen when at least 4-5 DNA residues
are positioned on the 3'-side and 10-12 DNA residues are positioned
on the 5'-side of the cleavable domain. For di-fluoro substrates,
maximal cleavage activity is seen when at least 8-10 DNA residues
are positioned on the 3'-side of the cleavable domain; from prior
examples it is clear that activity is high when 14-15 DNA residues
are positioned on the 5'-side of the cleavable domain.
Example 8
Application to DNA Primers: Primer Extension Assay Format and
Potential Utility in DNA Sequencing
[0308] The examples above characterized the ability of a
thermostable RNase H2 enzyme to cleave a duplex nucleic acid at a
single internal ribonucleotide or at a 2'-fluoro dinucleotide.
Example 7 establishes parameters for designing short
oligonucleotides which will be effective substrates in this
cleavage reaction. These features can be combined to make cleavable
primers that function in primer extension assays, such as DNA
sequencing, or PCR. A single stranded oligonucleotide is not a
substrate for the cleavage reaction, so a modified oligonucleotide
primer will be functionally "inert" until it hybridizes to a target
sequence. If a cleavable domain is incorporated into an otherwise
unmodified oligonucleotide, this oligonucleotide could function to
prime PCR and will result in an end product wherein a sizable
portion of the primer domain could be cleaved from the final PCR
product, resulting in sterilization of the reaction (lacking the
priming site, the product will no longer be a template for PCR
using the original primer set). If the cleavable domain is
incorporated into an oligonucleotide which is blocked at the
3'-end, then this primer will not be active in PCR until cleavage
has occurred. Cleavage will "activate" the blocked primer. As such,
this format can confer a "hot start" to a PCR reaction, as no DNA
synthesis can occur prior to the cleavage event. Example 4 showed
that this cleavage event is very inefficient with Pyrococcus abysii
RNase H2 until elevated temperatures are attained. Additionally,
the linkage between the cleavage reaction and primer extension
confer added specificity to the assay, since both steps
requireenzymatic recognition of the duplex formed when the primer
hybridizes to the template. A schematic of this reaction is shown
in FIG. 18. Note that this schema applies to both simple primer
extension reactions as well as PCR. It can also be exploited in
other kinds of enzymatic assays such as ligation reactions.
[0309] The following example demonstrates the use of an RNase H2
cleavable primer for DNA sequencing. The most common method of DNA
sequencing in use today involves sequential DNA synthesis reactions
(primer extension reactions) done in the presence of dideoxy
terminator nucleotides. The reaction is done in a thermal cycling
format where multiple cycles of primer extension are performed and
product accumulates in a linear fashion.
[0310] DNA sequencing was done using the Big Dye.TM. Terminator
V3.1 Cycle Sequencing Kit (Applied Biosystems, Foster City,
Calif.). The following primers were used:
M13(-27)
[0311] SEQ ID No. 84
TABLE-US-00048 5'-CAGGAAACAGCTATGAC-3'
M13(-27)-rC
[0312] SEQ ID No. 85
TABLE-US-00049 5'-CAGGAAACAGCTATGACcATGA-SpC3-3'
[0313] As before, DNA bases are indicated in upper case, RNA bases
are indicated in lower case, and SpC3 is a spacer C3 blocking group
placed at the 3'-end of the oligonucleotide. The blocked cleavable
primer contains 17 DNA bases on the 5'-side of the ribonucleotide
and 4 DNA bases on the 3'-side of the ribonucleotide (17-1-4
design) and so conforms to the optimized design rules established
in Example 7.
[0314] Sequencing reactions were set up in 20 .mu.l volume
comprising 0.75.times. ABI Reaction buffer, 160 nM primer,
0.5.times. Big Dye Terminators and 230 ng plasmid DNA template.
Optionally, 4 mM additional MgCl.sub.2 was supplemented into the
reaction, with or without 14, 1.4, or 0.14 mU of recombinant
Pyrococcus abyssi RNase H2. The following cycle sequencing program
was employed: 96.degree. C. for 30 seconds followed by 25 cycles of
[96.degree. C. for 5 seconds, 50.degree. C. for 10 seconds,
55.degree. C. for 4 minutes]. The DNA sequencing reactions were run
on an Applied Biosystems model 3130.times.1 Genetic Analyzer. The
resulting sequencing traces were examined for quality and read
length. Results are summarized in Table 16 below.
TABLE-US-00050 TABLE 16 Results of cycle sequencing using a rC
blocked cleavable primer Read length in ABI Read length in ABI
Primer RNase H2 Buffer Buffer + 4 mM MgCl.sub.2 M13(-27) 0 >800
~500 SEQ ID No. 84 0.14 mU >800 >800 1.4 mU >800 >800
14 mU >800 >800 M13(-27)-rC 0 0 0 SEQ ID No. 85 0.14 mU 0 0
1.4 mU 0 ~300 14 mU ~300 >800
[0315] Control reactions using an unmodified primer resulted in
high quality DNA sequence traces with usable read lengths slightly
exceeding 800 bases. The addition of RNase H2 enzyme to these
reactions did not compromise reaction quality. The manufacturer
(Applied Biosystems) does not disclose the cation content of the
buffer provided in the sequencing kits, so actual reaction
conditions are not certain. Supplementation of the reactions with
an additional 4 mM MgCl.sub.2 had no effect. The rC blocked
cleavable primer did not support DNA sequencing without the
addition of RNase H2. With the addition of RNase H2, high quality
sequencing reactions were obtained using 14 mU of enzyme in the 20
.mu.l reaction. Use of lower amounts enzyme resulted in lower
quality reactions or no functional reaction at all. Supplementing
magnesium content of the reaction buffer was necessary to obtain
cleavage and primer extension reactions using the blocked primers.
The amount of enzyme employed here is 100-fold higher than is
needed to achieve 100% cleavage of a rN substrate under optimal
conditions (70.degree. C., 20 minute incubation). In the cycle
sequencing reactions performed herein, primer annealing was run at
50.degree. C. and extension reactions were run at 55.degree. C. for
10 seconds and 4 minutes, respectively. These lower temperatures
are suboptimal for Pyrococcus abyssi RNase H2 (see Example 4
above). Performing the cycle sequencing reaction at higher
temperatures will require less enzyme but is not necessary.
[0316] This example demonstrates that blocked primers containing an
internal cleavage site for RNase H2 can be used with
primer-extension based sequencing methods, such as dideoxy (Sanger)
sequencing, and are compatible with use of existing high throughput
fluorescent sequencing protocols. Use of blocked primers and the
method of the present invention can confer added specificity to the
sequencing reaction, thus permitting sequencing to be performed for
more cycles and on highly complex nucleic acid samples that work
poorly with unmodified primers.
Example 9
Application to DNA Primers: rN Primers in PCR and Quantitative
Real-Time PCR
[0317] Example 8 demonstrated that RNase H2 could be used to cleave
a blocked primer and that this system could be linked to DNA
synthesis and primer extension reactions, including DNA sequencing.
The following example demonstrates the utility of this method in
PCR. The first system demonstrates use in an end point PCR format
and the second system demonstrates use in a quantitative real-time
PCR format.
[0318] The primers shown in Table 17, were made for use in a
synthetic end-point PCR assay. The Syn-For and Syn-Rev primers are
unmodified control primers specific for an artificial amplicon (a
synthetic oligonucleotide template). The Syn-For primer is paired
with the unmodified control Syn-Rev primer or the different
modified Syn-Rev primers. A set of modified Syn-Rev primers were
made which contain a single rU (cleavable) base followed by 2-6 DNA
bases, all ending with a dideoxy-C residue (ddC). The ddC residue
functions as a blocking group that prevents primer function. The
ddC blocking group is removed with cleavage of the primer at the rU
base by the action of RNase H2 (the unblocking step, shown in FIG.
18). The synthetic template is a 103-base long oligonucleotide,
shown below (SEQ ID No. 93). Primer binding sites are
underlined.
TABLE-US-00051 TABLE 17 Name Sequence SEQ ID No. Syn-For
5'-AGCTCTGCCCAAAGATT SEQ ID No. 86 ACCCTG-3' Syn-Rev
5'-CTGAGCTTCATGCCTTT SEQ ID No. 87 ACTGT-3' Syn-Rev-rU-2D
5'-CTGAGCTTCATGCCTTT SEQ ID No. 88 ACTGTuCC-ddC-3' Syn-Rev-rU-3D
5'-CTGAGCTTCATGCCTTT SEQ ID No. 89 ACTGTuCCC-ddC-3' Syn-Rev-rU-4D
5'-CTGAGCTTCATGCCTTT SEQ ID No. 90 ACTGTuCCCC-ddC-3' Syn-Rev-rU-5D
5'-CTGAGCTTCATGCCTTT SEQ ID No. 91 ACTGTuCCCCG-ddC-3' Syn-Rev-rU-6D
5'-CTGAGCTTCATGCCTTT SEQ ID No. 92 ACTGTuCCCCGA-ddC-3' DNA bases
are shown in uppercase. RNA bases are shown in lowercase. ddC
indicates a dideoxy-C residue which functions as a blocking
group.
Synthetic Template
[0319] SEQ ID No. 93
TABLE-US-00052 AGCTCTGCCCAAAGATTACCCTGACAGCTAAGTGGCAGTGG
AAGTTGGCCTCAGAAGTAGTGGCCAGCTGTGTGTCGGGGAA CAGTAAAGGCATGAAGCTCAG
[0320] PCR reactions were performed in 20 .mu.l volume using 200 nM
primers, 2 ng template, 200 .mu.M of each dNTP (800 .mu.M total), 1
unit of Immolase (a thermostable DNA polymerase, Bioline), 50 mM
Tris pH 8.3, 50 mM KCl, and 3 mM MgCl.sub.2. Reactions were run
either with or without 100 .mu.U of Pyrococcus abyssi RNase H2.
Reactions were started with a soak at 95.degree. C. for 5 minutes
followed by 35 cycles of [95.degree. C. for 10 seconds, 60.degree.
C. for 30 seconds, and 72.degree. C. for 1 second]. Reaction
products were separated on a 10% non-denaturing polyacrylamide gel
and visualized using GelStar staining. Results are shown in FIG.
19. Unmodified control primers produced a strong band of the
correct size. 3'-end blocked rU primers did not produce any
products in the absence of RNase H2. In the presence of RNase H2,
blocked primers produced a strong band of the correct size using
the D4, D5, and D6 primers. No signal was seen using the D2 or D3
primers. This example demonstrates that blocked primers can be used
in PCR reactions using the method of the present invention.
Further, this example is consistent with results obtained using
cleavage of preformed duplex substrates in Example 7, where the
presence of 4-5 3'-DNA bases were found to be optimal for cleavage
of rN containing primers.
[0321] The same synthetic PCR amplicon assay system described above
was next tested in a quantitative real-time PCR assay using
SYBR.RTM. Green detection. Reactions were done in 384 well format
using a Roche Lightcycler.RTM. 480 platform. Reactions comprised
1.times. BIO-RAD iQ.TM. SYBR.RTM. Green Supermix (BIO-RAD,
Hercules, Calif.), 200 nM of each primer (for+rev),
2.times.10.sup.6 copies of synthetic template oligonucleotide (SEQ
ID No. 93), and 5 mU of Pyrococcus abyssi RNase H2 in 10 .mu.l
volume. Thermal cycling parameters included an initial 5 minutes
soak at 95.degree. C. and then 45 cycles were performed of
[95.degree. C. for 10 seconds+60.degree. C. for 20
seconds+72.degree. C. for 30 seconds]. All reactions were run in
triplicate and reactions employed the same unmodified For primer
(SEQ ID No. 86). The Rev primer was varied between the unmodified
and 2-6D modified primers (SEQ ID Nos. 87-92). Cp values, the PCR
cycle number where a positive reaction is first detected, in these
experiments are shown in Table 18 below. The Cps were essentially
identical for control reactions done using unmodified For+Rev
primers and the coupled cleavage PCR reactions performed using the
D4, D5, or D6 blocked primers in the presence of RNase H2. In the
absence of RNase H2, no positive signal was detected using the
blocked primers. As was seen in the end point assay, performance
was reduced for the primers having shorter 3'-DNA domains (D2 or
D3).
TABLE-US-00053 TABLE 18 Cp values for SYBR .RTM. Green qPCR
reactions using cleavable blocked primers in a synthetic amplicon
system with RNase H2 present Reverse Primer SEQ ID No. Cp Value
Syn-Rev (Control) SEQ ID No. 87 17.7 Syn-Rev-rU-2D SEQ ID No. 88
23.4 Syn-Rev-rU-3D SEQ ID No. 89 23.0 Syn-Rev-rU-4D SEQ ID No. 90
16.8 Syn-Rev-rU-5D SEQ ID No. 91 16.6 Syn-Rev-rU-6D SEQ ID No. 92
16.9 All reactions used the same unmodified For primer, SEQ ID No.
86
[0322] The following example demonstrates use of RNase H2 cleavage
using rN blocked primers (both For and Rev) in a quantitative
real-time PCR assay format using an endogenous human gene target
and HeLa cell cDNA as template. The primers shown in Table 19
specific for the human HRAS gene (NM.sub.--176795) were designed
and synthesized. In this case a C3 spacer was use das the blocking
group.
TABLE-US-00054 TABLE 19 Name Sequence SEQ ID No. HRAS-618-For
5'-ACCTCGGCCAAGACCC-3' SEQ ID No. 94 HRAS-916-Rev
5'-CCTTCCTTCCTTCCTT SEQ ID No. 95 GCTTCC-3' HRAS-618-For-
5'-ACCTCGGCCAAGACCC SEQ ID No. 96 rG-D4 gGCAG-SpC3-3' HRAS-916-Rev-
5'-CCTTCCTTCCTTCCTT SEQ ID No. 97 rG-D4 GCTTCCgTCCT-SpC3-3'
Uppercase represents DNA bases, lowercase represents RNA bases.
SpC3 is a spacer C3 placed as a blocking group on the 3'-end.
[0323] These primers define a 340 bp amplicon within the HRAS gene
as shown below. Primer binding sites are underlined.
HRAS Assay Amplicon
[0324] SEQ ID No. 98
TABLE-US-00055 ACCTCGGCCAAGACCCGGCAGGGCAGCCGCTCTGGCTCTAG
CTCCAGCTCCGGGACCCTCTGGGACCCCCCGGGACCCATGT
GACCCAGCGGCCCCTCGCGCTGGAGTGGAGGATGCCTTCTA
CACGTTGGTGCGTGAGATCCGGCAGCACAAGCTGCGGAAGC
TGAACCCTCCTGATGAGAGTGGCCCCGGCTGCATGAGCTGC
AAGTGTGTGCTCTCCTGACGCAGCACAAGCTCAGGACATGG
AGGTGCCGGATGCAGGAAGGAGGTGCAGACGGAAGGAGGAG
GAAGGAAGGACGGAAGCAAGGAAGGAAGGAAGG
[0325] Reactions were performed in 10 l volume in 384 well format
using a Roche Lightcycler.RTM. 480 platform. Reactions comprised
1.times. BIO-RAD iQ.TM. SYBR.RTM. Green Supermix (BIO-RAD,
Hercules, Calif.) using the iTAQ DNA polymerase at 25 U/ml, 3 mM
MgCl.sub.2, 200 nM of each primer (for+rev), 2 ng cDNA (made from
HeLa cell total RNA), with or without 5 mU of Pyrococcus abyssi
RNase H2. Thermal cycling parameters included an initial 5 minutes
soak at 95.degree. C. and then 50 cycles were performed of
[95.degree. C. for 10 seconds+60.degree. C. for 20
seconds+72.degree. C. for 30 seconds]. All reactions were run in
triplicate. Using unmodified primers, the crossing point (Cp)
occurred at cycle 27. In the absence of RNase H2, reactions done
with blocked primers did not support PCR and no fluorescence signal
was detected during the 50 cycle reaction. In the presence of RNase
H2, reactions done with blocked primers produced detectable signal
at cycle 27.4, essentially identical to the control unblocked
primers. Real time PCR fluorescence plots are shown in FIG. 20.
[0326] The following example demonstrates use of RNase H2 cleavage
using rN blocked primers in a quantitative real-time PCR assay
format using another endogenous human gene target and HeLa cell
cDNA as the template. The primers specific for the human ETS2 gene
(NM.sub.--005239) shown in Table 20 were designed and
synthesized.
TABLE-US-00056 TABLE 20 Name Sequence SEQ ID No. ETS2-300-For
5'-CCCTGTTTGCTGTTTTT SEQ ID CCTTCTC-3' No. 99 ETS2-463-Rev
5'-CGCCGCTGTTCCTTTTT SEQ ID GAAG-3' No. 100 ETS2-300-For-
5'-CCCTGTTTGCTGTTTTT SEQ ID rU-D4 CCTTCTCuAAAT-SpC3-3' No. 101
ETS2-463-Rev- 5'-CGCCGCTGTTCCTTTTT SEQ ID rC-D4 GAAGcCACT-SpC3-3'
No. 102 Uppercase represents DNA bases, lowercase represents RNA
bases. SpC3 is a spacer C3 placed as a blocking group on the
3'-end
[0327] These primers define a 184 bp amplicon within the ETS2 gene
as shown below. Primer binding sites are underlined.
ETS2 Assay Amplicon
[0328] SEQ ID No. 103
TABLE-US-00057 CCCTGTTTGCTGTTTTTCCTTCTCTAAATGAAGAGCAAACA
CTGCAAGAAGTGCCAACAGGCTTGGATTCCATTTCTCATGA
CTCCGCCAACTGTGAATTGCCTTTGTTAACCCCGTGCAGCA
AGGCTGTGATGAGTCAAGCCTTAAAAGCTACCTTCAGTGGC TTCAAAAAGGAACAGCGGCG
[0329] Reactions were performed in 10 .mu.l volume in 384 well
format using a Roche Lightcycler.RTM. 480 platform. Reactions
comprised 1.times. BIO-RAD iQ.TM. SYBR.RTM. Green Supermix
(BIO-RAD, Hercules, Calif.) using the iTAQ DNA polymerase at 25
U/ml, 3 mM MgCl.sub.2, 200 nM of each primer (for+rev), 2 ng cDNA
(made from HeLa cell total RNA), with or without 5 mU of Pyrococcus
abyssi RNase H2. Thermal cycling parameters included an initial 5
minutes soak at 95.degree. C. and then 50 cycles were performed of
[95.degree. C. for 10 seconds+60.degree. C. for 20
seconds+72.degree. C. for 30 seconds]. All reactions were run in
triplicate. Using unblocked primers, the Cp occurred at cycle 25.7.
In the absence of RNase H2, reactions done with blocked primers did
not support PCR and no fluorescence signal was detected out to 50
cycles. In the presence of RNase H2, reactions done with blocked
primers produced detectable signal at cycle 31.7, a delay of 6
cycles from the unmodified control primers. Reactions done using
one blocked primer (unmodified For+blocked Rev or blocked
For+unmodified Rev) showed intermediate Cp values. Real time PCR
fluorescence plots are shown in FIG. 21.
[0330] Using the present reaction conditions, the HRAS assay
performed identically using unmodified vs. blocked primers.
However, the ETS2 assay showed a delay between unmodified vs.
blocked primers. In the setting of a PCR reaction where rapid
thermal cycling occurs, primer hybridization and cleavage kinetics
play a significant role in the efficiency of the overall reaction
for reactions which employ the blocked primers. DNA synthesis is
linked to the unblocking event, and unblocking requires
hybridization, binding of RNase H2, and substrate cleavage before
primers become activated and are capable of priming DNA synthesis.
It should be possible to increase the amount of cleaved primer
produced each cycle by either increasing the amount of RNase H2
enzyme present or by increasing the anneal time of the reaction.
DNA synthesis occurs at the anneal temperature (60.degree. C.)
nearly as well as at the extension temperature (72.degree. C.) used
in the above examples. However, unblocking can only take place
during the duration of the anneal step (60.degree. C.) and not
during the extend step (72.degree. C.) due to the Tm of the primers
employed which only permit formation of a double-stranded substrate
for RNase H2 during the anneal step but not at 72.degree. C. (where
the primers only exist in single-stranded form).
[0331] PCR cycle parameters were changed to a 2 step reaction with
anneal/extend as a single event done at 60.degree. C. and the
duration of the anneal/extend step was varied to see if changing
these reaction parameters could allow the blocked ETS2 primers to
perform with similar efficiency as the unmodified control primers.
Reactions were done in 10 .mu.l volume in 384 well format using a
Roche Lightcycler.RTM. 480 platform. Reactions comprised 1.times.
BIO-RAD iQ.TM. SYBR.RTM. Green Supermix (BIO-RAD, Hercules, Calif.)
using the iTAQ DNA polymerase at 25 U/ml, 3 mM MgCl.sub.2, 200 nM
of each primer (for+rev), 2 ng cDNA (made from HeLa cell total
RNA), with or without 5 mU of Pyrococcus abyssi RNase H2. Thermal
cycling parameters included an initial 5 minutes soak at 95.degree.
C. and then 45 cycles were performed of [95.degree. C. for 10
seconds+60.degree. C. for 20-120 seconds]. All reactions were run
in triplicate. The differences between the Cp values obtained for
the blocked primers and the unmodified control primers (.DELTA.Cp)
are summarized in Table 21 below.
TABLE-US-00058 TABLE 21 .DELTA.Cp values for SYBR .RTM. Green qPCR
ETS2 reactions comparing unmodified and cleavable blocked primers
Combined time at 60.degree. C. (anneal/extend) .DELTA.Cp Value 20
seconds 6.1 60 seconds 1.2 90 seconds 0.6 120 seconds 0.4
[0332] Minor adjustment of the cycling parameters and increasing
the duration of the 60.degree. C. anneal step from 20 seconds to
1-2 minutes led to uniform performance between the
blocked-cleavable primers and the control unmodified primers.
Similar experiments were performed keeping the cycling parameters
fixed and increasing enzyme. As predicted, it was possible to
improve performance of the blocked primers using higher amounts of
enzyme. Doubling the amount of enzyme employed to 10 mU RNase H
resulted in minimal difference between control unblocked and
blocked cleaveable primers when using a 30 second anneal step at
60.degree. C.
[0333] The above example demonstrates that blocked primers
containing a single ribonucleotide residue of the optimized design
taught in Example 7 can be used with RNase H2 in quantitative
real-time PCR assays.
Example 10
Application to DNA Primers: fNfN Primers in PCR and Quantitative
Real-Time PCR
[0334] Example 9 above demonstrated utility of RNase H2 mediated
cleavage for use of rN blocked primers in end point and
quantitative real time PCR assays. The present example demonstrates
utility using fNfN blocked primers in quantitative real time PCR
assays.
[0335] Since cleavage of the di-fluoro substrate by RNase H2
results in a species having a 3'-OH end, this product should also
be able to support PCR reactions using the same reaction format as
described in Example 9, assuming that primers bearing a single 2'-F
base (fN) are capable of priming DNA synthesis. Cleavage of a
di-fluoro substrate proceeds best in the presence of manganese
cations, whereas PCR reactions generally are performed in the
presence of magnesium cations. PCR reactions using unmodified
primers were tested using standard qPCR buffer containing 3 mM
MgCl.sub.2 and a modified buffer containing 3 mM MgCl.sub.2+0.6 mM
MnCl.sub.2. Reaction performance was identical and the presence of
this low amount of manganese did not adversely affect the
quantitative nature of the reaction.
[0336] The ability of a terminal 3'-fN primer to function in PCR
was investigated using the synthetic PCR amplicon system described
in example 9. The following primers shown in Table 22 were
tested:
TABLE-US-00059 TABLE 22 Name Sequence SEQ ID No. Syn-For
5'-AGCTCTGCCCAAA SEQ ID No. 86 GATTACCCTG-3' Syn-Rev
5'-CTGAGCTTCATGC SEQ ID No. 87 CTTTACTGT-3' Syn-Rev-fU
5'-CTGAGCTTCATGC SEQ ID No. 104 CTTTACTGT(fU)-3' DNA bases are
shown in uppercase. 2'-fluoro bases are indicated as fN.
Synthetic Template
[0337] SEQ ID No. 93
TABLE-US-00060 AGCTCTGCCCAAAGATTACCCTGACAGCTAAGTGGCAGTGG
AAGTTGGCCTCAGAAGTAGTGGCCAGCTGTGTGTCGGGGAA CAGTAAAGGCATGAAGCTCAG
[0338] Reactions were done in 10 .mu.l volume in 384 well format
using a Roche Lightcycler.RTM. 480 platform. Reactions comprised
1.times. BIO-RAD iQ.TM. SYBR.RTM. Green Supermix (BIO-RAD,
Hercules, Calif.) using the iTAQ DNA polymerase at 25 U/ml, 3 mM
MgCl.sub.2, 0.6 mM MnCl.sub.2, 200 nM of each primer (for+rev),
2.times.10.sup.6 copies of synthetic oligonucleotide target, with
or without 1.75 U of Pyrococcus abyssi RNase H2. Thermal cycling
parameters included an initial 5 minutes soak at 95.degree. C. and
then 30 cycles were performed of [95.degree. C. for 10
seconds+60.degree. C. for 120 seconds+72.degree. C. for 120
seconds]. All reactions were run in triplicate. Results are shown
in FIG. 22. In the absence of RNase H2, the primer having a 2'-F
base at the 3'-end supported PCR with identical efficiency compared
with the unmodified primer. However, in the presence of RNase H2,
the 2'-F modified primer showed a 3.5 Cp delay compared with the
unmodified primer. This results not from the inhibition of DNA
synthesis by RNase H2, but from a low level of cleavage of the
primer from the amplification product by RNase H2. Following DNA
synthesis, incorporation of a fN-containing primer into the newly
formed DNA product creates a potential substrate for RNase H2 (see
example 5 above). Cleavage at the 2'-F base will remove the priming
site from this strand of the amplicon, effectively sterilizing this
product so that any products made from it will be incapable of
further priming events. It is this reaction sequence which occurs
in polynomial amplification. Cleavage of substrates containing a
single 2'-F residue is relatively inefficient, however, so only a
modest decrease in PCR reaction efficiency is seen. Extending
incubation at 72.degree. C. following PCR should result in total
cleavage of the primer from the amplification product, completely
blocking the ability of further amplification to occur and thereby
sterilizing the product. This should be useful in
cross-contamination control of PCR reactions.
[0339] Given that the cleavage of a single 2'-F residue is
inefficient, use of lower amounts of enzyme, or eliminating the
72.degree. C. elongation step permits cleavage of a difluoro
blocked primer by RNase H2 without significantly cleaving the
primer extension reaction product containing a single 2' fluoro
residue. Alternatively, it should be possible to block this
cleavage event by selective placement of a phosphorothioate
modification between the terminal 2'-F residue and the adjacent DNA
base.
[0340] The ability of a di-fluoro blocked primer to support qPCR
was demonstrated using the primers shown in Table 23, in the
synthetic oligonucleotide amplicon system, described in Example 9
above.
TABLE-US-00061 TABLE 23 Name Sequence SEQ ID No. Syn-For
5'-AGCTCTGCCCAAA SEQ ID No. 86 GATTACCCTG-3' Syn-Rev-fU
5'-CTGAGCTTCATGC SEQ ID No. 104 CTTTACTGT(fU)-3' Syn-Rev-
5'-CTGAGCTTCATGC SEQ ID No. 105 fUfC-D10 CTTTACTGT(fUfC)C
CCGACACAC-SpC3-3' DNA bases are shown in uppercase. 2'-F bases are
indicated as fN. SpC3 indicates a spacer C3 group employed to block
the 3'-end
Synthetic Template
[0341] SEQ ID No. 93
TABLE-US-00062 AGCTCTGCCCAAAGATTACCCTGACAGCTAAGTGGCAGTGG
AAGTTGGCCTCAGAAGTAGTGGCCAGCTGTGTGTCGGGGAA CAGTAAAGGCATGAAGCTCAG
[0342] Reactions were done in 10 .mu.l volume in 384 well format
using a Roche Lightcycler.RTM. 480 platform. Reactions comprised
1.times. BIO-RAD iQ.TM. SYBR.RTM. Green Supermix (BIO-RAD,
Hercules, Calif.) using the iTAQ DNA polymerase at 25 U/ml, 3 mM
MgCl.sub.2, 0.6 mM MnCl.sub.2, 200 nM of each primer (for+rev),
2.times.10.sup.6 copies of synthetic oligonucleotide target, with
or without 1.75 U of Pyrococcus abyssi RNase H2. Thermal cycling
parameters included an initial 5 minutes soak at 95.degree. C. and
then 45 cycles were performed of [95.degree. C. for 10
seconds+60.degree. C. for 120 seconds+72.degree. C. for 120
seconds]. All reactions were run in triplicate. The reactions run
with the control primer having a single 2'-fluoro base at the
3'-end (which mimics the cleavage product of the fNfN blocked
primer) had a Cp of 20. Reactions run with the blocked fUfC primer
also had a Cp of 20.
[0343] The amount of RNase H2 enzyme needed in the di-fluoro primer
cleavage assay was next studied in more detail. Reactions were done
in 10 .mu.l volume in 384 well format using a Roche
Lightcycler.RTM. 480 platform. Reactions comprised 1.times. BIO-RAD
iQ.TM. SYBR.RTM. Green Supermix (BIO-RAD, Hercules, Calif.) using
the iTAQ DNA polymerase at 25 U/ml, 3 mM MgCl.sub.2, 0.6 mM
MnCl.sub.2, 200 nM of each primer (for+rev), 2.times.10.sup.6
copies of synthetic oligonucleotide target. The same unmodified
Syn-For primer was used in all reactions. Recombinant Pyrococcus
abyssi RNase H2 was added from 0 to 600 mU per reaction. Thermal
cycling parameters included an initial 5 minutes soak at 95.degree.
C. and then 45 cycles were performed of [95.degree. C. for 10
seconds+60.degree. C. for 120 seconds+72.degree. C. for 120
seconds]. All reactions were run in triplicate. Cp values
corresponding to the varying amounts of RNase H2 for each primer
are shown in Table 24.
TABLE-US-00063 TABLE 24 Optimization of the amount of RNase H2 for
qPCR reactions using a fUfC blocked primer Amount of RNase H2 used
per reaction Primer 600 mU 400 mU 200 mU 100 mU 0 mU Syn-Rev 17.9
17.7 17.2 17.1 17.0 Syn-Rev-fU 25.6 23.2 19.9 18.5 17.0 Syn-Rev-fU
24.6 22.9 21.3 21.9 ND fC-D10 ND = not detected.
[0344] The optimal amount of RNase H2 is 200 mU (Cp=21.3 shown in
bold and underlined). At higher concentrations of RNase H2 PCR
reaction is less efficient, and to a similar degree, with both the
3' fluoroU primer and the blocked difluoro primer. Presumably this
is due to a low level of cleavage at the fU set within the PCR
product as discussed above.
[0345] Generally about 200 mU of Pyrococcus abysii RNase H2 per 10
.mu.l is the optimal enzyme concentration for a coupled RNase
H2-PCR with blocked primers wherein the RNase H2 cleavage domain is
two consecutive 2'-fluoronucleosides. An increase in Cp compared to
standard unmodified DNA primers of between 2 and 6 cycles is
typically observed. This small difference has no effect on assay
performance because results are always compared to a standard curve
of Cp vs. target copy number generated with the same primers as
used to test unknown samples.
[0346] In conclusion, this example has demonstrated that blocked
fNfN primers can support qPCR reactions using RNase H2 cleavage
with the methods of the present invention and defines optimal
amounts of RNase H2 and cycling conditions to employ.
Example 11
Improved Specificity Using rN Blocked Primers in PCR Reactions
[0347] In theory, PCR has an almost unlimited potential for
amplification and a PCR reaction should only be limited by
consumption of reagents in the reaction mix. In actual practice,
PCR reactions are typically limited to 40-45 cycles to help
preserve specificity. The amplification power of PCR is enormous
and, as cycle number exceeds 40-45, it becomes increasingly common
for mispriming events to give rise to amplification of undesired
products and false positive signals. This example demonstrates how
use of cleavable blocked primers with the methods of the present
invention improves reaction specificity and permits use of a
greater number of PCR cycles, thereby increasing the potential
sensitivity of PCR.
[0348] In this example, we studied PCR reactions specific for 3
human genes and compared the specificity of each set of primer
pairs in amplification using human and rat cDNA as the template.
Traditional unmodified oligonucleotides were compared with the new
cleavable blocked primers of the present invention. The following
primers, as shown in Table 25, were employed. DNA bases are shown
in upper case, RNA bases in lower case, and the 3'-blocking group
employed was a C3 spacer (SpC3).
[0349] The gene targets studied were human ETS2, NM.sub.--005239
(rat homolog NM.sub.--001107107), human HRAS, NM.sub.--176795 (rat
homolog NM.sub.--001061671), and human ACACA, NM.sub.--198834 (rat
homolog NM.sub.--022193).
TABLE-US-00064 TABLE 25 Gene Primer SEQ ID No. Sequence ETS2
hETS2-For SEQ ID No. 106 CCCTGTTTGCTGTTTTTCCTTCTC hETS2-For-rU SEQ
ID No. 107 CCCTGTTTGCTGTTTTTCCTTCTCuAAAT-SpC3 hETS2-Rev SEQ ID No.
108 CGCCGCTGTTCCTTTTTGAAG hETS2-Rev-rC SEQ ID No. 109
CGCCGCTGTTCCTTTTTGAAGcCACT-SpC3 HRAS hHRAS-For SEQ ID No. 110
ACCTCGGCCAAGACCC hHRAS-For-rG SEQ ID No. 111
ACCTCGGCCAAGACCCgGCAG-SpC3 hHRAS-Rev SEQ ID No. 112
CCTTCCTTCCTTCCTTGCTTCC hHRAS-Rev-rG SEQ ID No. 113
CCTTCCTTCCTTCCTTGCTTCCgTCCT-SpC3 ACACA hACACA-For SEQ ID No. 114
GCATTTCTTCCATCTCCCCCTC hACACA-For-rU SEQ ID No. 115
GCATTTCTTCCATCTCCCCCTCuGCCT-SpC3 hACACA-Rev SEQ ID No. 116
TCCGATTCTTGCTCCACTGTTG hACACA-Rev-rG SEQ ID No. 117
TCCGATTCTTGCTCCACTGTTGgCTGA-SpC3
[0350] PCR reactions were done in 384 well format using a Roche
Lightcycler.RTM. 480 platform. Reactions comprised 1.times. BIO-RAD
iQ.TM. SYBR.RTM. Green Supermix (BIO-RAD, Hercules, Calif.), 200 nM
of each primer (For+Rev), and 1.3 mU of Pyrococcus abyssi RNase H2
in 10 .mu.l volume. Template DNA was either 2 ng of human HeLa cell
cDNA or 2 ng of rat spinal cord cDNA. Thermal cycling parameters
included an initial 5 minutes soak at 95.degree. C. and then 60
cycles were performed of [95.degree. C. for 10 seconds+60.degree.
C. for 90 seconds]. Under these conditions, the Cp value observed
for human cDNA represents a true positive event. If any signal was
detected using rat cDNA, it was recorded as a false positive event.
For these 3 genes, the human and rat sequences are divergent at the
primer binding sites. Therefore detection of a PCR product in rat
cDNA using human gene specific primers is an undesired, false
positive result that originates from mispriming Results are shown
in Table 26 below.
TABLE-US-00065 TABLE 26 False detection of products in rat cDNA
using human gene specific primers in a 60 cycle qRT-PCR reaction
Observed Cp Observed Cp Primers (For/Rev) Human cDNA Rat cDNA
.DELTA.Cp ETS2 23.6 56.4 32.8 ETS2-blocked 24.9 ND >assay HRAS
25.2 35.5 10.3 HRAS-blocked 26.1 ND >assay ACACA 26.2 52.3 26.1
ACACA-blocked 26.3 ND >assay ND = not detected
[0351] Using unmodified primers, detection of the human targets in
human cDNA was successful and Cp's of 23-26 were observed. For all
3 PCR assays, the human gene-specific primers also detected
products in rat cDNA when cycling was continued, and Cp's of 35-56
were observed. These represent undesired false positive signals
which limit the ability of the PCR assays to detect low levels of
true positive signal.
[0352] Using modified primers, detection of the desired product in
human cDNA was successful and Cp's were all within 1 of the values
obtained for unmodified primers. However, no false positive signals
were seen using rat cDNA with the modified primers, even at 60
cycles. Use of the RNase H2 blocked-cleavable primers resulted in
improved specificity, permitting use of longer, more sensitive PCR
reactions (in this case up to 60 cycles) without detection of false
priming events. This allows for a much greater ability to detect
variant alleles in the presence of a larger excess of the wild type
sequence.
Example 12
Mismatch Discrimination for a rC Substrate Under Steady State
Conditions
[0353] Example 11 demonstrated the ability of the methods of the
invention to improve specificity of a qPCR reaction in the face of
background mispriming events. The present example demonstrates the
specificity of the RNase H2 cleavage reaction with respect to
single-base differences (SNPs). The ability of the Pyrococcus
abyssi RNase H2 enzyme to distinguish base mismatches in a duplex
substrate containing a single rC base was tested under steady state
conditions. The following substrates were .sup.32P-end labeled and
incubated in "Mg Cleavage Buffer" as described in Example 4 above.
Reactions comprised 100 nM substrate with 100 .mu.U of enzyme in 20
.mu.L volume and were incubated at 70.degree. C. for 20 minutes.
Reaction products were separated using denaturing 7M urea, 15%
polyacrylamide gel electrophoresis (PAGE) and visualized using a
Packard Cyclone.TM. Storage Phosphor System (phosphorimager). The
relative intensity of each band was quantified and results plotted
as a fraction of total substrate cleaved.
[0354] Ten duplexes were studied, including the perfect match
(rC:G, SEQ ID NOS 11 and 12) as well as each possible base mismatch
at the rC base (3 duplexes, SEQ ID Nos. 11 and 118-120), at
position+1 relative to the rC (3 duplexes, SEQ ID Nos. 11 and
121-123), and at position-1 relative to the rC (3 duplexes, SEQ ID
Nos. 11 and 124-126). Results were normalized for perfect
match=100% and are shown in Table 27 below.
TABLE-US-00066 TABLE 27 Cleavage of rC substrates with and without
mismatches under steady state conditions Duplex Identity SEQ ID NOS
Substrate Sequence Cleavage SEQ ID No. 11 5' CTCGTGAGGTGATGc 100%
AGGAGATGGGAGGCG 3' SEQ ID No. 12 3' GAGCACTCCACTACG TCCTCTACCCTCCGC
5' SEQ ID No. 11 5' CTCGTGAGGTGATGc 46% AGGAGATGGGAGGCG 3' SEQ ID
No. 118 3' GAGCACTCCACTACA TCCTCTACCCTCCGC 5' SEQ ID No. 11 5'
CTCGTGAGGTGATGc 35% AGGAGATGGGAGGCG 3' SEQ ID No. 119 3'
GAGCACTCCACTACT TCCTCTACCCTCCGC 5' SEQ ID No. 11 5' CTCGTGAGGTGATGc
23% AGGAGATGGGAGGCG 3' SEQ ID No. 120 3' GAGCACTCCACTACC
TCCTCTACCCTCCGC 5' SEQ ID No. 11 5' CTCGTGAGGTGATGc 19%
AGGAGATGGGAGGCG 3' SEQ ID No. 121 3' GAGCACTCCACTAAG
TCCTCTACCCTCCGC 5' SEQ ID No. 11 5' CTCGTGAGGTGATGc 65%
AGGAGATGGGAGGCG 3' SEQ ID No. 122 3' GAGCACTCCACTATG
TCCTCTACCCTCCGC 5' SEQ ID No. 11 5' CTCGTGAGGTGATGc 22%
AGGAGATGGGAGGCG 3' SEQ ID No. 123 3' GAGCACTCCACTAGG
TCCTCTACCCTCCGC 5' SEQ ID No. 11 5' CTCGTGAGGTGATGc 61%
AGGAGATGGGAGGCG 3' SEQ ID No. 124 3' GAGCACTCCACTACG
ACCTCTACCCTCCGC 5' SEQ ID No. 11 5' CTCGTGAGGTGATGc 91%
AGGAGATGGGAGGCG 3' SEQ ID No. 125 3' GAGCACTCCACTACG
CCCTCTACCCTCCGC 5' SEQ ID No. 11 5' CTCGTGAGGTGATGc 46%
AGGAGATGGGAGGCG 3' SEQ ID No. 126 3' GAGCACTCCACTACG
GCCTCTACCCTCCGC 5' DNA bases are shown as uppercase. RNA bases are
shown as lowercase. Mismatches are shown in bold font and are
underlined.
[0355] Pyrococcus RNase H2 was able to discriminate between single
base mismatches under these conditions. The precise degree of
discrimination varied with which bases were paired in the mismatch.
Interestingly, mismatches at position-1 (one base 5' to the rC
base) showed relatively good mismatch discrimination while
mismatches at position +1 (one base 3' to the rC base) were in
general less effective. Although the selectivity appears relatively
modest, it becomes greatly amplified with repeated cycles of
PCR.
Example 13
Mismatch Discrimination for rN Substrates During Thermal
Cycling
[0356] The ability of the Pyrococcus abyssi RNase H2 enzyme to
distinguish base mismatches for a rC substrate under steady state
conditions was described in Example 12. The ability of this enzyme
to distinguish base mismatches for all rN containing substrates
under conditions of thermal cycling was examined in the present
example. In these conditions, the cleavable substrate is only
available for processing by the enzyme for a short period of time
before temperature elevation disrupts the duplex. Mismatch
discrimination was assessed in the setting of a fluorescent
quantitative real-time PCR assay. We found that base mismatch
discrimination was greatly improved under these kinetically limited
conditions than were observed under steady-state conditions.
[0357] The following nucleic acids were employed in this example.
Oligonucleotides were synthesized to provide coverage for all
nearest neighbor pairs and mismatches.
Unmodified for Primer:
[0358] SEQ ID No. 86
TABLE-US-00067 5' AGCTCTGCCCAAAGATTACCCTG 3'
[0359] Blocked rN substrate rev primers (C3 spacer blocking group
at the 3'-end) are shown below. DNA bases are uppercase and RNA
bases are lower case. Regions of variation are indicated by bold
and underlined. At total of 28 blocked primers containing a single
RNA residue were synthesized.
rA Series:
[0360] SEQ ID No. 127
TABLE-US-00068 5' CTGAGCTTCATGCCTTTACTGTaCCCC-SpC3 3'
[0361] SEQ ID No. 128
TABLE-US-00069 5' CTGAGCTTCATGCCTTTACTGAaCCCC-SpC3 3'
[0362] SEQ ID No. 129
TABLE-US-00070 5' CTGAGCTTCATGCCTTTACTGCaCCCC-SpC3 3'
[0363] SEQ ID No. 130
TABLE-US-00071 5' CTGAGCTTCATGCCTTTACTGGaCCCC-SpC3 3'
[0364] SEQ ID No. 131
TABLE-US-00072 5' CTGAGCTTCATGCCTTTACTGTaTCCC-SpC3 3'
[0365] SEQ ID No. 132
TABLE-US-00073 5' CTGAGCTTCATGCCTTTACTGTaGCCC-SpC3 3'
[0366] SEQ ID No. 133
TABLE-US-00074 5' CTGAGCTTCATGCCTTTACTGTaACCC-SpC3 3'
rU Series:
[0367] SEQ ID No. 134
TABLE-US-00075 5' CTGAGCTTCATGCCTTTACTGTuCCCC-SpC3 3'
[0368] SEQ ID No. 135
TABLE-US-00076 5' CTGAGCTTCATGCCTTTACTGAuCCCC-SpC3 3'
[0369] SEQ ID No. 136
TABLE-US-00077 5' CTGAGCTTCATGCCTTTACTGCuCCCC-SpC3 3'
[0370] SEQ ID No. 137
TABLE-US-00078 5' CTGAGCTTCATGCCTTTACTGGuCCCC-SpC3 3'
[0371] SEQ ID No. 138
TABLE-US-00079 5' CTGAGCTTCATGCCTTTACTGTuTCCC-SpC3 3'
[0372] SEQ ID No. 139
TABLE-US-00080 5' CTGAGCTTCATGCCTTTACTGTuGCCC-SpC3 3'
[0373] SEQ ID No. 140
TABLE-US-00081 5' CTGAGCTTCATGCCTTTACTGTuACCC-SpC3 3'
rC Series:
[0374] SEQ ID No. 141
TABLE-US-00082 5' CTGAGCTTCATGCCTTTACTGTcCCCC-SpC3 3'
[0375] SEQ ID No. 142
TABLE-US-00083 5' CTGAGCTTCATGCCTTTACTGAcCCCC-SpC3 3'
[0376] SEQ ID No. 143
TABLE-US-00084 5' CTGAGCTTCATGCCTTTACTGCcCCCC-SpC3 3'
[0377] SEQ ID No. 144
TABLE-US-00085 5' CTGAGCTTCATGCCTTTACTGGcCCCC-SpC3 3'
[0378] SEQ ID No. 145
TABLE-US-00086 5' CTGAGCTTCATGCCTTTACTGTcTCCC-SpC3 3'
[0379] SEQ ID No. 146
TABLE-US-00087 5' CTGAGCTTCATGCCTTTACTGTcGCCC-SpC3 3'
[0380] SEQ ID No. 147
TABLE-US-00088 5' CTGAGCTTCATGCCTTTACTGTcACCC-SpC3 3'
rG Series:
[0381] SEQ ID No. 148
TABLE-US-00089 5' CTGAGCTTCATGCCTTTACTGTgCCCC-SpC3 3'
[0382] SEQ ID No. 149
TABLE-US-00090 5' CTGAGCTTCATGCCTTTACTGAgCCCC-SpC3 3'
[0383] SEQ ID No. 150
TABLE-US-00091 5' CTGAGCTTCATGCCTTTACTGCgCCCC-SpC3 3'
[0384] SEQ ID No. 151
TABLE-US-00092 5' CTGAGCTTCATGCCTTTACTGGgCCCC-SpC3 3'
[0385] SEQ ID No. 152
TABLE-US-00093 5' CTGAGCTTCATGCCTTTACTGTgTCCC-SpC3 3'
[0386] SEQ ID No. 153
TABLE-US-00094 5' CTGAGCTTCATGCCTTTACTGTgGCCC-SpC3 3'
[0387] SEQ ID No. 154
TABLE-US-00095 5' CTGAGCTTCATGCCTTTACTGTgACCC-SpC3 3'
[0388] The unblocked control Rev primer (mimicing reaction product
of blocked primers after cleavage by RNase H2) employed was:
[0389] SEQ ID NO: 309
TABLE-US-00096 5' CTGAGCTTCATGCCTTTACTG 3'
[0390] The following perfect-matched and mismatched synthetic
templates were employed. The locations of varying bases are
indicated in bold font with underline. Unique templates were made
for each possible base variation at the ribonucleotide or one base
5' or one base 3' of the ribonucleotide. In total, 28 templates
were synthesized and tested.
rA Templates:
[0391] SEQ ID No. 155
TABLE-US-00097 5' AGCTCTGCCCAAAGATTACCCTGACAGCTAAGTGGCAGTGG
AAGTTGGCCTCAGAAGTAGTGGCCAGCTGTGTGTCGGGGTACAG
TAAAGGCATGAAGCTCAG-3'
[0392] SEQ ID No. 156
TABLE-US-00098 5'AGCTCTGCCCAAAGATTACCCTGACAGCTAAGTGGCA
GTGGAAGTTGGCCTCAGAAGTAGTGGCCAGCTGTGTGTC
GGGGTCCAGTAAAGGCATGAAGCTCAG-3'
[0393] SEQ ID No. 157
TABLE-US-00099 5'AGCTCTGCCCAAAGATTACCCTGACAGCTAAGTGGCA
GTGGAAGTTGGCCTCAGAAGTAGTGGCCAGCTGTGTGTC
GGGGTTCAGTAAAGGCATGAAGCTCAG-3'
[0394] SEQ ID No. 158
TABLE-US-00100 5'AGCTCTGCCCAAAGATTACCCTGACAGCTAAGTGGCA
GTGGAAGTTGGCCTCAGAAGTAGTGGCCAGCTGTGTGTC
GGGGTGCAGTAAAGGCATGAAGCTCAG-3'
[0395] SEQ ID No. 159
TABLE-US-00101 5'AGCTCTGCCCAAAGATTACCCTGACAGCTAAGTGGCA
GTGGAAGTTGGCCTCAGAAGTAGTGGCCAGCTGTGTGTC
GGGCTACAGTAAAGGCATGAAGCTCAG-3'
[0396] SEQ ID No. 160
TABLE-US-00102 5'AGCTCTGCCCAAAGATTACCCTGACAGCTAAGTGGCA
GTGGAAGTTGGCCTCAGAAGTAGTGGCCAGCTGTGTGTC
GGGATACAGTAAAGGCATGAAGCTCAG-3'
[0397] SEQ ID No. 161
TABLE-US-00103 5'AGCTCTGCCCAAAGATTACCCTGACAGCTAAGTGGCA
GTGGAAGTTGGCCTCAGAAGTAGTGGCCAGCTGTGTGTC
GGGTTACAGTAAAGGCATGAAGCTCAG-3'
[0398] rU templates:
[0399] SEQ ID No. 162
TABLE-US-00104 5'AGCTCTGCCCAAAGATTACCCTGACAGCTAAGTGGCA
GTGGAAGTTGGCCTCAGAAGTAGTGGCCAGCTGTGTGTC
GGGGAACAGTAAAGGCATGAAGCTCAG-3'
[0400] SEQ ID No. 163
TABLE-US-00105 5'AGCTCTGCCCAAAGATTACCCTGACAGCTAAGTGGCA
GTGGAAGTTGGCCTCAGAAGTAGTGGCCAGCTGTGTGTC
GGGGATCAGTAAAGGCATGAAGCTCAG-3'
[0401] SEQ ID No. 164
TABLE-US-00106 5'AGCTCTGCCCAAAGATTACCCTGACAGCTAAGTGGCA
GTGGAAGTTGGCCTCAGAAGTAGTGGCCAGCTGTGTGTC
GGGGACCAGTAAAGGCATGAAGCTCAG-3'
[0402] SEQ ID No. 165
TABLE-US-00107 5'AGCTCTGCCCAAAGATTACCCTGACAGCTAAGTGGCA
GTGGAAGTTGGCCTCAGAAGTAGTGGCCAGCTGTGTGTC
GGGGAGCAGTAAAGGCATGAAGCTCAG-3'
[0403] SEQ ID No. 166
TABLE-US-00108 5'AGCTCTGCCCAAAGATTACCCTGACAGCTAAGTGGCA
GTGGAAGTTGGCCTCAGAAGTAGTGGCCAGCTGTGTGTC
GGGCAACAGTAAAGGCATGAAGCTCAG-3'
[0404] SEQ ID No. 167
TABLE-US-00109 5'AGCTCTGCCCAAAGATTACCCTGACAGCTAAGTGGCA
GTGGAAGTTGGCCTCAGAAGTAGTGGCCAGCTGTGTGTC
GGGAAACAGTAAAGGCATGAAGCTCAG-3'
[0405] SEQ ID No. 168
TABLE-US-00110 5'AGCTCTGCCCAAAGATTACCCTGACAGCTAAGTGGCA
GTGGAAGTTGGCCTCAGAAGTAGTGGCCAGCTGTGTGTC
GGGTAACAGTAAAGGCATGAAGCTCAG-3'
rG Templates:
[0406] SEQ ID No. 169
TABLE-US-00111 5'AGCTCTGCCCAAAGATTACCCTGACAGCTAAGTGGCA
GTGGAAGTTGGCCTCAGAAGTAGTGGCCAGCTGTGTGTC
GGGGCACAGTAAAGGCATGAAGCTCAG-3'
[0407] SEQ ID No. 170
TABLE-US-00112 5'AGCTCTGCCCAAAGATTACCCTGACAGCTAAGTGGCA
GTGGAAGTTGGCCTCAGAAGTAGTGGCCAGCTGTGTGTC
GGGGCTCAGTAAAGGCATGAAGCTCAG-3'
[0408] SEQ ID No. 171
TABLE-US-00113 5'AGCTCTGCCCAAAGATTACCCTGACAGCTAAGTGGCAGTGGAA
GTTGGCCTCAGAAGTAGTGGCCAGCTGTGTGTCGGGGCCCAGTAA
AGGCATGAAGCTCAG-3'
[0409] SEQ ID No. 172
TABLE-US-00114 5'AGCTCTGCCCAAAGATTACCCTGACAGCTAAGTGGCAGTGGAA
GTTGGCCTCAGAAGTAGTGGCCAGCTGTGTGTCGGGGCGCAGTAA
AGGCATGAAGCTCAG-3'
[0410] SEQ ID No. 173
TABLE-US-00115 5'AGCTCTGCCCAAAGATTACCCTGACAGCTAAGTGGCAGTGGAA
GTTGGCCTCAGAAGTAGTGGCCAGCTGTGTGTCGGGACACAGTAA
AGGCATGAAGCTCAG-3'
[0411] SEQ ID No. 174
TABLE-US-00116 5'AGCTCTGCCCAAAGATTACCCTGACAGCTAAGTGGCAGTGGAA
GTTGGCCTCAGAAGTAGTGGCCAGCTGTGTGTCGGGTCACAGTAA
AGGCATGAAGCTCAG-3'
[0412] SEQ ID No. 175
TABLE-US-00117 5'AGCTCTGCCCAAAGATTACCCTGACAGCTAAGTGGCAGTGGAA
GTTGGCCTCAGAAGTAGTGGCCAGCTGTGTGTCGGGCCACAGTAA
AGGCATGAAGCTCAG-3'
rC Templates
[0413] SEQ ID No. 176
TABLE-US-00118 5'AGCTCTGCCCAAAGATTACCCTGACAGCTAAGTGGCAGTGGAA
GTTGGCCTCAGAAGTAGTGGCCAGCTGTGTGTCGGGGGACAGTAA
AGGCATGAAGCTCAG-3'
[0414] SEQ ID No. 177
TABLE-US-00119 5'AGCTCTGCCCAAAGATTACCCTGACAGCTAAGTGGCAGTGGAA
GTTGGCCTCAGAAGTAGTGGCCAGCTGTGTGTCGGGGGTCAGTAA
AGGCATGAAGCTCAG-3'
[0415] SEQ ID No. 178
TABLE-US-00120 5'AGCTCTGCCCAAAGATTACCCTGACAGCTAAGTGGCAGTGGAA
GTTGGCCTCAGAAGTAGTGGCCAGCTGTGTGTCGGGGGCCAGTAA
AGGCATGAAGCTCAG-3'
[0416] SEQ ID No. 179
TABLE-US-00121 5'AGCTCTGCCCAAAGATTACCCTGACAGCTAAGTGGCAGTGGAA
GTTGGCCTCAGAAGTAGTGGCCAGCTGTGTGTCGGGGGGCAGTAA
AGGCATGAAGCTCAG-3'
[0417] SEQ ID No. 180
TABLE-US-00122 5'AGCTCTGCCCAAAGATTACCCTGACAGCTAAGTGGCAGTGGAA
GTTGGCCTCAGAAGTAGTGGCCAGCTGTGTGTCGGGAGACAGTAA
AGGCATGAAGCTCAG-3'
[0418] SEQ ID No. 181
TABLE-US-00123 5'AGCTCTGCCCAAAGATTACCCTGACAGCTAAGTGGCAGTGGAA
GTTGGCCTCAGAAGTAGTGGCCAGCTGTGTGTCGGGTGACAGTAA
AGGCATGAAGCTCAG-3'
[0419] SEQ ID No. 182
TABLE-US-00124 5'AGCTCTGCCCAAAGATTACCCTGACAGCTAAGTGGCAGTGGAA
GTTGGCCTCAGAAGTAGTGGCCAGCTGTGTGTCGGGCGACAGTAA
AGGCATGAAGCTCAG-3'
[0420] Together, these nucleic acids (SEQ ID NOS 86, 310, 134 and
87, respectively, in order of appearance) comprise PCR assays set
up as indicated:
##STR00004##
[0421] The terminal C3 spacer group (indicated by "x") blocks the
rU containing oligonucleotide to serve as a primer. When hybridized
to the template, the duplex becomes a substrate for RNase H2 and
cleavage occurs immediately 5'- to the rU residue, resulting in a
functional primer as shown (.rarw.).
[0422] Quantitative real time PCR reactions were performed using
unmodified primer SEQ ID No. 86 and pairwise combinations of rN
containing primers SEQ ID Nos. 127-154 and templates SEQ ID Nos.
155-182. Reactions were done in 384 well format using a Roche
Lightcycler.RTM. 480 platform. Reactions comprised 1.times. BIO-RAD
iQ.TM. SYBR.RTM. Green Supermix (BIO-RAD, Hercules, Calif.), 200 nM
of each primer (for +rev), and 1.3 mU of Pyrococcus abyssi RNase H2
in 10 .mu.l volume. Thermal cycling parameters included an initial
5 minutes soak at 95.degree. C. and then 45 cycles were performed
of [95.degree. C. for 10 seconds+60.degree. C. for 20
seconds+72.degree. C. for 30 seconds]. Under these conditions, the
Cp value was identical for control reactions done using For +Rev
(unmodified) primers and control coupled RNase H2 cleavage-PCR
reactions done using the perfect match For (unmodified)+rN Rev
(blocked) primers. Thus the reaction conditions employed had
sufficient incubation time and RNase H2 concentration to cleave the
perfect match species within the kinetic constraints of the real
time thermal cycling and any deviations from this point will
represent a change in reaction efficiency imparted by base
mismatches present between the blocked primer and the various
templates.
[0423] Pairwise combinations of primers and templates were run as
described above and results are summarized below showing .DELTA.Cp,
which is the difference of cycle threshold observed between control
and mismatch reactions. Since each Cp represents a cycle in PCR
(which is an exponential reaction under these conditions), a
.DELTA.Cp of 10 represents a real differential of 2.sup.10, or a
1024 fold change in sensitivity. A .DELTA.Cp of 4 to 5 cycles is
generally sufficient to discriminate between SNPs in allele
specific PCR assays.
[0424] Results for tests done varying bases at the central position
over the rN base are shown below in Table 28 (SEQ ID NOS 311 and
312, respectively, in order of appearance):
##STR00005##
TABLE-US-00125 TABLE 28 .DELTA.Cp for all possible base mismatches
at the rN position Template A C G T rA 14.9 9.4 13.6 0 rC 7.4 9.2 0
6.6 rG 13.9 0 12.7 14.5 rU 0 12.2 10.9 5.3
[0425] Very large differences in reactive efficiency are seen in
RNase H2 cleavage of a rN substrate under thermal cycling
conditions, ranging from a difference of around 40-fold (.DELTA.Cp
5.3) to over a 30,000 fold difference (.DELTA.Cp 14.9). None of the
assays showed a .DELTA.Cp less than 5 cycles. Thus the RNase H2 rN
cleavage reaction shows far greater specificity in the setting of a
kinetic assay (qPCR) than under steady state conditions and much
greater selectivity than allele specific PCR with standard DNA
primers. Added specificity may be conferred by the design of the
primers as described in the detailed description of the invention
and demonstrated in the examples below.
[0426] Results for tests done varying bases at the -1 position
relative to the rN base are shown below in Table 29 (SEQ ID NOS 313
and 314, respectively, in order of appearance):
##STR00006##
TABLE-US-00126 TABLE 29 .DELTA.Cp for all possible base mismatches
at position -1 relative to a rU base Template A C G T A(rU) 16.1
8.7 12.6 0 C(rU) 7.6 3.9 0 12.0 G(rU) 13.8 0 12.4 5.9 T(rU) 0 5.2
2.4 6.2
[0427] Results for tests done varying bases at the +1 position
relative to the rN base are shown below in Table 30 (SEQ ID NOS 315
and 316, respectively, in order of appearance):
##STR00007##
TABLE-US-00127 TABLE 30 .DELTA.Cp for all possible base mismatches
at position +1 relative to a rU base Template A C G T (rU)A 11.4
2.5 12.2 0 (rU)C 6.4 10.4 0 9.0 (rU)G 13.8 0 4.5 3.0 (rU)T 0 11.1
11.9 2.9
[0428] Pairwise combinations were similarly tested for all sequence
variants listed above for the -1 and +1 positions relative to the
rN base, including the rA, rC, and rG probes. Results are shown in
Tables 31-36 below.
TABLE-US-00128 TABLE 31 .DELTA.Cp for all possible base mismatches
at position -1 relative to a rA base Template A C G T A(rA) 14.2
8.6 11.8 0 C(rA) 6.9 12.6 0 6.8 G(rA) 12.8 0 12.6 8.9 T(rA) 0 5.1
1.4 8.6
TABLE-US-00129 TABLE 32 .DELTA.Cp for all possible base mismatches
at position +1 relative to a rA base Template A C G T (rA)A 3.1 1.0
6.12 0 (rA)C 9.3 10.2 0 8.3 (rA)G 13.2 0 2.5 5.9 (rA)T 0 5.0 7.1
4.0
TABLE-US-00130 TABLE 33 .DELTA.Cp for all possible base mismatches
at position -1 relative to a rC base Template A C G T A(rC) 13.0
8.2 10.5 0 C(rC) 5.0 3.3 0 3.5 G(rC) 8.3 0 7.0 0.8 T(rC) 0 5.4 2.1
4.6
TABLE-US-00131 TABLE 34 .DELTA.Cp for all possible base mismatches
at position +1 relative to a rC base Template A C G T (rC)A 5.6 1.8
10.2 0 (rC)C 8.8 9.6 0 8.6 (rC)G 9.8 0 3.2 0.3 (rC)T 0 2.1 0.2
0.0
TABLE-US-00132 TABLE 35 .DELTA.Cp for all possible base mismatches
at position -1 relative to a rG base Template A C G T A(rG) 12.4
4.8 10.4 0 C(rG) 4.5 11.1 0 2.5 G(rG) 10.3 0 10.1 3.8 T(rG) 0 3.5
2.2 5.3
TABLE-US-00133 TABLE 36 .DELTA.Cp for all possible base mismatches
at position +1 relative to a rG base Template A C G T (rG)A 6.2 3.0
11.4 0 (rG)C 9.5 7.3 0 4.7 (rG)G 13.1 0 6.0 3.2 (rG)T 0 4.5 11.5
0.3
[0429] The relative change in reaction efficiency of cleavage of a
rN substrate by Pyrococcus abyssi RNase H2 in the setting of a
single base mismatch varies with the identity of the paired bases,
the relative position of the mismatch to the cleavage site, and the
neighboring bases. The mismatch charts defined in this example can
be used to design optimal mismatch detection assays which maximize
the expected differential (.DELTA.Cp) between mismatch and matched
loci, and can be built into an algorithm to automate optimization
of new assay designs.
Example 14
Mismatch Discrimination for fUfU Substrate Under Steady State
Conditions
[0430] The ability of the Pyrococcus abyssi RNase H2 enzyme to
distinguish base mismatches in a duplex substrate containing a fUfU
dinucleotide pair was tested under steady state conditions. The
following substrates were .sup.32P-end labeled and incubated in "Mn
Cleavage Buffer" as described in Examples 5 and 6 above. Reactions
comprised 100 nM substrate with 1 U of enzyme in 20 .mu.L volume
and were incubated at 70.degree. C. for 20 minutes. Reaction
products were separated using denaturing 7M urea, 15%
polyacrylamide gel electrophoresis (PAGE) and visualized using a
Packard Cyclone.TM. Storage Phosphor System (phosphorimager). The
relative intensity of each band was quantified and results plotted
as a fraction of total substrate cleaved.
[0431] Fourteen duplexes shown in Table 37, were studied, including
the perfect match (SEQ ID NOS 60 and 201), mismatches within the
2'-fluoro dinucleotide pair (SEQ ID Nos. 60 and 183-189), and
mismatches adjacent to the 2'-fluoro dinucleotide pair (SEQ ID Nos.
60 and 190-195). Results were normalized for a perfect
match=100%.
TABLE-US-00134 TABLE 37 Cleavage of fUfU substrates with and
without mismatches under steady state conditions Duplex Identity
Substrate Sequence Cleavage SEQ ID No. 60 5'
CTCGTGAGGTGAT(fUfU)AGGAGATGGGAGGCG 3' 100% SEQ ID No. 201 3'
GAGCACTCCACTA A A TCCTCTACCCTCCGC 5' SEQ ID No. 60 5'
CTCGTGAGGTGAT(fUfU)AGGAGATGGGAGGCG 3' 5% SEQ ID No. 183 3'
GAGCACTCCACTA A G TCCTCTACCCTCCGC 5' SEQ ID No. 60 5'
CTCGTGAGGTGAT(fUfU)AGGAGATGGGAGGCG 3' 14% SEQ ID No. 184 3'
GAGCACTCCACTA A C TCCTCTACCCTCCGC 5' SEQ ID No. 60 5'
CTCGTGAGGTGAT(fUfU)AGGAGATGGGAGGCG 3' 1% SEQ ID No. 185 3'
GAGCACTCCACTA A T TCCTCTACCCTCCGC 5' SEQ ID No. 60 5'
CTCGTGAGGTGAT(fUfU)AGGAGATGGGAGGCG 3' 2% SEQ ID No. 186 3'
GAGCACTCCACTA C T TCCTCTACCCTCCGC 5' SEQ ID No. 60 5'
CTCGTGAGGTGAT(fUfU)AGGAGATGGGAGGCG 3' 0% SEQ ID No. 187 3'
GAGCACTCCACTA G G TCCTCTACCCTCCGC 5' SEQ ID No. 60 5'
CTCGTGAGGTGAT(fUfU)AGGAGATGGGAGGCG 3' 0% SEQ ID No. 188 3'
GAGCACTCCACTA C C TCCTCTACCCTCCGC 5' SEQ ID No. 60 5'
CTCGTGAGGTGAT(fUfU)AGGAGATGGGAGGCG 3' 0% SEQ ID No. 189 3'
GAGCACTCCACTA T T TCCTCTACCCTCCGC 5' SEQ ID No. 60 5'
CTCGTGAGGTGAT(fUfU)AGGAGATGGGAGGCG 3' 8% SEQ ID No. 190 3'
GAGCACTCCACTC A A TCCTCTACCCTCCGC 5' SEQ ID No. 60 5'
CTCGTGAGGTGAT(fUfU)AGGAGATGGGAGGCG 3' 4% SEQ ID No. 191 3'
GAGCACTCCACTG A A TCCTCTACCCTCCGC 5' SEQ ID No. 60 5'
CTCGTGAGGTGAT(fUfU)AGGAGATGGGAGGCG 3' 4% SEQ ID No. 192 3'
GAGCACTCCACTT A A TCCTCTACCCTCCGC 5' SEQ ID No. 60 5'
CTCGTGAGGTGAT(fUfU)AGGAGATGGGAGGCG 3' 2% SEQ ID No. 193 3'
GAGCACTCCACTA A A GCCTCTACCCTCCGC 5' SEQ ID No. 60 5'
CTCGTGAGGTGAT(fUfU)AGGAGATGGGAGGCG 3' 8% SEQ ID No. 194 3'
GAGCACTCCACTA A A CCCTCTACCCTCCGC 5' SEQ ID No. 60 5'
CTCGTGAGGTGAT(fUfU)AGGAGATGGGAGGCG 3' 2% SEQ ID No. 195 3'
GAGCACTCCACTA A A ACCTCTACCCTCCGC 5' DNA bases are shown as
uppercase. 2'-F bases are shown as fU. Mismatches are shown in bold
font and are underlined.
[0432] Pyrococcus RNase H2 was able to discriminate very
efficiently between single base mismatches under these conditions.
The precise degree of discrimination varied with which bases were
paired in the mismatch. Interestingly, mismatches at both positions
-1 and +1 (relative to the fUfU domain) were effective. Specificity
for cleavage using the fUfU substrate was significantly higher
under steady state assay conditions than was the rC substrate
(Example 12 above).
[0433] The study above employed the fUfU dinucleotide pair, which
was previously shown in Example 6 to be the least efficient
di-fluoro substrate for cleavage of the 16 possible dinucleotide
pairs. This may impact the mismatch results. Similar experiments
were conducted using the same complement strands, substituting a
fUfC di-fluoro substrate strand. RNase H2 was reduced to 20 mU due
to the increased activity of cleavage seen for fUfC compared to
fUfU substrates. Results are shown in Table 38 below.
TABLE-US-00135 TABLE 38 Cleavage of fUfC substrates with and
without mismatches under steady state conditions Duplex Identity
Substrate Sequence Cleavage SEQ ID No. 58 5'
CTCGTGAGGTGAT(fUfC)AGGAGATGGGAGGCG 3' 100% SEQ ID No. 317 3'
GAGCACTCCACTA A G TCCTCTACCCTCCGC 5' SEQ ID No. 58 5'
CTCGTGAGGTGAT(fUfC)AGGAGATGGGAGGCG 3' 0% SEQ ID No. 196 3'
GAGCACTCCACTA T G TCCTCTACCCTCCGC 5' SEQ ID No. 58 5'
CTCGTGAGGTGAT(fUfC)AGGAGATGGGAGGCG 3' 7% SEQ ID No. 197 3'
GAGCACTCCACTA C G TCCTCTACCCTCCGC 5' SEQ ID No. 58 5'
CTCGTGAGGTGAT(fUfC)AGGAGATGGGAGGCG 3' 0% SEQ ID No. 198 3'
GAGCACTCCACTA G G TCCTCTACCCTCCGC 5' SEQ ID No. 58 5'
CTCGTGAGGTGAT(fUfC)AGGAGATGGGAGGCG 3' 1% SEQ ID No. 199 3'
GAGCACTCCACTA A T TCCTCTACCCTCCGC 5' SEQ ID No. 58 5'
CTCGTGAGGTGAT(fUfC)AGGAGATGGGAGGCG 3' 0% SEQ ID No. 200 3'
GAGCACTCCACTA A C TCCTCTACCCTCCGC 5' SEQ ID No. 58 5'
CTCGTGAGGTGAT(fUfC)AGGAGATGGGAGGCG 3' 2% SEQ ID No. 201 3'
GAGCACTCCACTA A A TCCTCTACCCTCCGC 5' SEQ ID No. 58 5'
CTCGTGAGGTGAT(fUfC)AGGAGATGGGAGGCG 3' 0% SEQ ID No. 202 3'
GAGCACTCCACTA T C TCCTCTACCCTCCGC 5' SEQ ID No. 58 5'
CTCGTGAGGTGAT(fUfC)AGGAGATGGGAGGCG 3' 0% SEQ ID No. 203 3'
GAGCACTCCACTT A G TCCTCTACCCTCCGC 5' SEQ ID No. 58 5'
CTCGTGAGGTGAT(fUfC)AGGAGATGGGAGGCG 3' 4% SEQ ID No. 204 3'
GAGCACTCCACTC A G TCCTCTACCCTCCGC 5' SEQ ID No. 58 5'
CTCGTGAGGTGAT(fUfC)AGGAGATGGGAGGCG 3' 0% SEQ ID No. 205 3'
GAGCACTCCACTG A G TCCTCTACCCTCCGC 5' SEQ ID No. 58 5'
CTCGTGAGGTGAT(fUfC)AGGAGATGGGAGGCG 3' 2% SEQ ID No. 206 3'
GAGCACTCCACTA A G ACCTCTACCCTCCGC 5' SEQ ID No. 58 5'
CTCGTGAGGTGAT(fUfC)AGGAGATGGGAGGCG 3' 4% SEQ ID No. 207 3'
GAGCACTCCACTA A G CCCTCTACCCTCCGC 5' SEQ ID No. 58 5'
CTCGTGAGGTGAT(fUfC)AGGAGATGGGAGGCG 3' 2% SEQ ID No. 208 3'
GAGCACTCCACTA A G GCCTCTACCCTCCGC 5' DNA bases are shown as
uppercase. 2'-F bases are shown as fU. Mismatches are shown in bold
font and are underlined.
[0434] Again, Pyrococcus abyssi RNase H2 was able to discriminate
very efficiently between single base mismatches. The precise degree
of discrimination varied with which bases were paired in the
mismatch. As before, mismatches at both positions -1 and +1
(relative to the fUfC domain) were effective. Specificity for
cleavage using the fUfC substrate was significantly higher under
steady state assay conditions than was the rC substrate (Example 12
above) and also showed slightly greater specificity than the fUfU
substrate. Under kinetic assay conditions during thermal cycling,
mismatch assays using di-fluoro substrates may show even greater
selectivity.
Example 15
Selective Placement of Phosphorothioate Internucleotide
Modifications in the Substrate
[0435] The effect of incorporation of a phosphorothioate
internucleoside linkage was tested for several different
substrates. Phosphorothioate (PS) bonds are typically considered
relatively nuclease resistant and are commonly used to increase the
stability of oligonucleotides in nuclease containing solutions,
such as serum. PS bonds form two stereoisomers, Rp and Sp, which
usually show different levels of stabilization for different
nucleases.
[0436] The di-fluoro substrate was examined with a PS bond between
the two modified bases. A mixture of both diastereomers was
employed for the present study.
Unmodified fUfC Substrate:
[0437] SEQ ID NOs 58 and 317, respectively, in order of
appearance
TABLE-US-00136 5'-CTCGTGAGGTGAT(fUfC)AGGAGATGGGAGGCG-3'
3'-GAGCACTCCACTA A G TCCTCTACCCTCCGC-5'
[0438] PS modified fU*fC substrate ("*"=PS bond):
[0439] SEQ ID NOs 209 and 318, respectively, in order of
appearance
TABLE-US-00137 5'-CTCGTGAGGTGAT(fU*fC)AGGAGATGGGAGGCG-3'
3'-GAGCACTCCACTA A G TCCTCTACCCTCCGC-5'
[0440] (note--gaps in sequence are for alignment purposes)
[0441] The above substrates were incubated for 1 hour at 70.degree.
C. in "Mn Cleavage Buffer" using 160 pmoles of substrate in 120
.mu.l volume (1.3 .mu.M) and 4 units of the recombinant Pyrococcus
RNase H2 enzyme. Reactions were stopped with the addition of gel
loading buffer (formamide/EDTA) and separated on a denaturing 7M
urea, 15% polyacrylamide gel. Gels were stained using GelStar.TM.
(Lonza, Rockland, Me.) and visualized with UV excitation. The
unmodified substrate was 100% cleaved under these conditions;
however the PS-modified substrate was essentially uncleaved. The
phosphorothioate modification can effectively block cleavage of a
di-fluoro substrate.
[0442] A substrate containing a single rC residue was studied next,
testing placement of the PS modification on either side of the RNA
base (5'- or 3'-side as indicated). A mixture of both diastereomers
were employed for the present study.
Unmodified rC Substrate:
[0443] SEQ ID NOs 11 and 12, respectively, in order of
appearance
TABLE-US-00138 5'-CTCGTGAGGTGATTcAGGAGATGGGAGGCG-3'
3'-GAGCACTCCACTAAGTCCTCTACCCTCCGC-5'
PS Modified 5' *rC Substrate:
[0444] SEQ ID NOs 210 and 347, respectively, in order of
appearance
TABLE-US-00139 5'-CTCGTGAGGTGATT*cAGGAGATGGGAGGCG-3'
3'-GAGCACTCCACTAA GTCCTCTACCCTCCGC-5'
PS Modified 3' rC* Substrate:
[0445] SEQ ID NOs 211 and 348, respectively, in order of
appearance
TABLE-US-00140 5'-CTCGTGAGGTGATTc*AGGAGATGGGAGGCG-3'
3'-GAGCACTCCACTAAG TCCTCTACCCTCCGC-5'
[0446] The above substrates were incubated for 1 hour at 70.degree.
C. in "Mg Cleavage Buffer" using 160 pmoles of substrate in 120
.mu.l volume (1.3 .mu.M) and 4 units of the recombinant Pyrococcus
RNase H2 enzyme. Reactions were stopped with the addition of gel
loading buffer (formamide/EDTA) and separated on a denaturing 7M
urea, 15% polyacrylamide gel. Gels were stained using GelStar.TM.
(Lonza, Rockland, Me.) and visualized with UV excitation. The
unmodified substrate was 100% cleaved under these conditions. Both
the 5'-*rC and 3'-rC* PS-modified substrates were approximately 50%
cleaved under these conditions. These results are most consistent
with one stereoisomer, Rp or Sp, being more resistant to cleavage
than the other isomer.
[0447] The 3'-rC* substrate was studied in greater detail. Since
RNase H2 cleaves this substrate on the 5'-side of the
ribonucleotide while other RNases (such as RNase A, RNase 1, etc.)
cleave this substrate on the 3'-side of the ribonucleotide, it may
be possible to use the PS modification as a way of protecting the
substrate from unwanted degradation by other nucleases while
leaving it available as an RNase H2 substrate. It is well known
that cleavage of RNA substrates by RNase A and other
single-stranded ribonucleases is inhibited to a greater extent by
the Sp phosphorothioate isomer than the Rp isomer. The relative
effects of the Sp vs. Rp isomer on RNase H2 cleavage have not been
known. Therefore the two stereoisomers were purified and the
relative contributions of the Sp and Rp isomers on 3'-rC* substrate
stability were studied.
[0448] It is well known that phosphorothioate isomers can be
separated by HPLC techniques and that this separation is readily
done if only a single PS bond exists in an oligonucleotide. HPLC
was therefore employed to purify the two PS isomers of the 3'-rC*
substrate, SEQ ID No. 211. A mass of 7 nmoles of the
single-stranded 3'-rC* containing oligonucleotide was employed.
Characterization showed that the test material had a molecular
weight of 9464 Daltons (calculated 9465) by ESI-MS with a molar
purity of 95% by capillary electrophoresis. This material was
injected into a 4.6 mm.times.50 mm Xbridge.TM. C18 column (Waters)
with 2.5 micron particle size. Starting mobile phase (Buffer A) was
100 mM TEAA pH 7.0 with 5% acetonitrile and which was mixed with
pure acetonitrile (Buffer B) at 35.degree. C. The HPLC method
employed clearly resolved two peaks in the sample which were
collected and re-run to demonstrate purity. HPLC traces of the
mixed isomer sample and purified specimens are shown in FIG. 23.
Both the "A" and "B" peaks had an identical mass of 9464 Daltons by
ESI-MS. From the original sample, 1.3 nmoles of peak "A" and 3.6
nmoles of peak "B" were recovered.
[0449] It was not possible based upon mass or HPLC data to identify
which peak was the Rp and which peak was the Sp isomer. Relative
resistance to degradation by RNase A was employed to assign isomer
identity to the purification fractions. The Sp isomer is known to
confer relatively greater resistance to RNase A degradation than
the Rp isomer. Purified products were studied in the
single-stranded form. The substrate was radiolabeled with .sup.32P
using 6000 Ci/mmol .gamma.-.sup.32P-ATP and the enzyme T4
Polynucleotide Kinase (Optikinase, US Biochemical). Trace label was
added to reaction mixtures (1:50). Reactions were performed using
100 nM substrate in 20 .mu.l volume with 1 pg (72 attomoles) of
RNase A in Mg Cleavage Buffer. Reactions were incubated at
70.degree. C. for 20 minutes. Reaction products were separated
using denaturing 7M urea, 15% polyacrylamide gel electrophoresis
(PAGE) and visualized using a Packard Cyclone.TM. Storage Phosphor
System (phosphorimager). The relative intensity of each band was
quantified and results plotted as a fraction of total substrate
cleaved. Peak "A" was more completely degraded by RNase A than peak
"B"; peak "A" was therefore assigned identity as the Rp isomer and
peak "B" was assigned as the Sp isomer.
[0450] The relative susceptibility of each stereoisomer to RNase H2
cleavage was studied. The RNA-containing strand of the substrate
was radiolabeled with .sup.32P using 6000 Ci/mmol
.gamma.-.sup.32P-ATP and the enzyme T4 Polynucleotide Kinase
(Optikinase, US Biochemical). Trace label was added to reaction
mixtures (1:50). Reactions were performed using 100 nM substrate in
20 .mu.l volume with 100 .mu.U of recombinant Pyrococcus abyssi
RNase H2 in Mg Cleavage Buffer. Substrates were employed in both
single-stranded and duplex form. Reactions were incubated at
70.degree. C. for 20 minutes. Reaction products were separated
using denaturing 7M urea, 15% polyacrylamide gel electrophoresis
(PAGE) and visualized using a Packard Cyclone.TM. Storage Phosphor
System (phosphorimager). The relative intensity of each band was
quantified and results plotted as a fraction of total substrate
cleaved. As expected, single-stranded substrates were not cleaved
by the RNase H2 enzyme. The control unmodified rC duplex (SEQ ID
NOS 11 and 12) were 100% cleaved under the conditions employed. The
Sp isomer 3'-rC* duplex substrate (peak "B") was cleaved .about.30%
whereas the Rp isomer (peak "A") was cleaved <10% under these
conditions. Therefore the relative susceptibility to cleavage of
racemically pure phosphorothioate modified substrates at this
position (3'- to the ribonucleotide) is exactly opposite for RNase
H2 vs. RNase A. The Sp isomer is more readily cleaved by RNase H2
while the Rp isomer is more readily cleaved by RNase A. Therefore
single ribonucleotide containing substrates having a racemically
pure Sp isomer phosphorothioate modification on the 3'-side of the
ribonucleotide could be employed to protect this bond from unwanted
degradation by single-stranded nucleases (such as RNase A) while
still being a functional substrate for cleavage by RNase H2. The
relationship between enzyme cleavage and phosphorothioate
stereoisomer is summarized in FIG. 24.
Example 16
Utility of rN Containing Dual-Labeled Probes in qPCR Assays
[0451] The following example illustrates a real time PCR assay
utilizing a rU-containing dual labeled probe. Previously, we
demonstrated in Example 9 the feasibility for use of rN blocked
primers in qPCR using a SYBR.RTM. Green detection format. Cleavage
of blocked oligonucleotides using the method of the present
invention can also be applied to the dual-labeled probe assay
format. Use of RNase H1 to cleave a dual-labeled probe containing a
4 RNA base cleavage domain in an isothermal cycling probe assay
format has been described by Harvey, J. J., et al. (Analytical
Biochemistry, 333:246-255, 2004). Another dual-labeled probe assay
using RNase H has been described, wherein a molecular beacon
containing a single ribonucleotide residue was employed to detect
polymorphisms in an end-point PCR format using RNase H2 (Hou, J.,
et al., Oligonucleotides, 17:433-443, 2007). In the present example
we will demonstrate use of single ribonucleotide containing
dual-labeled probes in a qPCR assay format that relies upon RNase
H2 cleavage of the probe.
[0452] The following oligonucleotides shown in Table 39, were used
as probes and primers in a qPCR assay with a dual-labeled
fluorescence-quenched probe. The target was a synthetic
oligonucleotide template.
TABLE-US-00141 TABLE 39 Name Sequence SEQ ID No. Syn-For
5'-AGCTCTGCCCAAAGATTA SEQ ID No. 86 CCCTG-3' Syn-Rev
5'-CTGAGCTTCATGCCTTTA SEQ ID No. 87 CTGT-3' Syn-Probe
5'-FAM-TTCTGAGGCCAACT SEQ ID No. 212 CCACTGCCACTTA-IBFQ-3'
Syn-Probe-rU 5'-FAM-TTCTGAGGCCAACu SEQ ID No. 213
CCACTGCCACTTA-IBFQ-3' DNA bases are shown in uppercase. RNA bases
are shown in lowercase. FAM is 6-carboxyfluorescein and IBFQ is a
dark quencher (Integrated DNA Technologies).
[0453] Synthetic template (primer and probe binding sites are
underlined).
[0454] SEQ ID No. 93
TABLE-US-00142 AGCTCTGCCCAAAGATTACCCTGACAGCTAAGTGGCAGTGGAAGTT
GGCCTCAGAAGTAGTGGCCAGCTGTGTGTCGGGGAACAGTAAAGGC ATGAAGCTCAG
[0455] Quantitative real time PCR reactions were performed using
unmodified primers SEQ ID Nos. 86 and 87 and probes Seq ID Nos. 212
and 213. Reactions were done in 384 well format using a Roche
Lightcycler.RTM. 480 platform. Reactions comprised 200 nM of each
primer (for+rev) and 200 nM probe, 2.times.10.sup.6 copies of the
synthetic template, and 5 mU of Pyrococcus abyssi RNase H2 in 10
.mu.l volume. Thermal cycling parameters included an initial 10
minutes incubation at 95.degree. C. and then 45 cycles were
performed of [95.degree. C. for 10 seconds+60.degree. C. for 30
seconds+72.degree. C. for 1 seconds]. The buffer employed varied
with the polymerase used.
[0456] If PCR is performed using a thermostable DNA polymerase
having 5'-exonuclease activity the polymerase will degrade the
probe. Under these conditions, a DNA probe should perform the same
as a rN modified probe. This reaction constitutes a positive
control. If a DNA polymerase is employed which is lacking
5'-exonuclease activity, then neither probe should be degraded.
This reaction constitutes a negative control. A PCR reaction using
the exo-negative polymerase with RNase H2, however, should degrade
the rN containing probe but not the DNA probe, demonstrating
function of the invention. For the present study, the following two
thermostable polymerases were used: Immolase (intact 5' nuclease
activity, Bioline) and Vent Exo.sup.- (5'-exonuclease negative
mutant, New England Biolabs). Buffers employed were the
manufacturer's recommended buffers for the DNA polymerases and were
not optimized for RNase H2 activity. For Immolase, the buffer
comprised 16 mM (NH.sub.4).sub.2SO.sub.4, 67 mM Tris pH 8.3, and 3
mM MgCl.sub.2. For Vent Exo.sup.-, the buffer comprised 10 mM
(NH.sub.4).sub.2SO.sub.4, 20 mM Tris pH 8.8, 10 mM KCl, and 3 mM
MgSO.sub.4.
[0457] qPCR reactions were run as described and results are shown
below in Table 40.
TABLE-US-00143 TABLE 40 Cp values of qPCR reactions comparing DNA
or rU Dual-Labeled Probes Probe Polymerase Minus RNase H2 Plus
RNase H2 Syn-Probe Immolase Exo.sup.+ 21.1 21.0 Vent Exo.sup.- ND
ND Syn-Probe-rU Immolase Exo.sup.+ 21.0 20.7 Vent Exo.sup.- ND 21.1
ND = not detectable
[0458] Using the exonuclease positive polymerase, both probes
showed similar functional performance and gave similar Cp values,
both with or without RNase H2. Using the exonuclease deficient
mutant polymerase, however, the DNA probe did not produce any
detectable fluorescent signal; the rU probe failed to produce
fluorescent signal in the absence of RNase H2, but in the presence
of RNase H2 was cleaved and resulted in signal at the expected Cp
value. Similar results can be obtained using di-fluoro containing
probes. If the RNase H2 cleavage domain is placed over a mutation
site such probes can be used to distinguish variant alleles.
[0459] RNase H-cleavable probes can also be linked with the use of
blocked primers of the present invention to additively increase the
specificity of amplification based assay systems.
Example 17
Utility of rN Containing Blocked Primer to Prevent Primer-Dimer
Formation
[0460] Formation of primer-dimers or other small target independent
amplicons can be a significant problem in both endpoint and
real-time PCR. These products can arise even when the primers
appear to be well designed. Further, it is sometimes necessary to
employ primers which have sub-optimal design because of sequence
constraints for selection of primers which hybridize to specific
regions. For example, PCR assays for certain viruses can be subtype
or serotype specific if primers are chosen in areas that are
variable between strains. Conversely, PCR reactions can be designed
to broadly amplify all viral strains if primers are placed in
highly conserved regions of the viral genome. Thus the sequence
space available to choose primers may be very limited and "poor"
primers may have to be employed that have the potential to form
primer dimers. Use of "hot start" PCR methods may eliminate some
but not all of these problems.
[0461] The following example derives from one such case cited in
U.S. patent Ser. No. 06/001,611 where primer-dimers were found to
be a significant problem during development of a PCR-based nucleic
acid detection assay for the Hepatitis C virus (HCV) using sites in
conserved domains that permit detection of a wide range of viral
serotypes. We demonstrate herein that use of cleavable blocked
primers can prevent unwanted primer dimer formation, specifically
in the absence of a "hot-start" DNA polymerase.
[0462] The following oligonucleotides, as shown in Table 41, were
used as primers in a PCR assay. The target was a cloned synthetic
amplicon isolated from a plasmid.
TABLE-US-00144 TABLE 41 Name Sequence SEQ ID No. ST280A-for
5'-GCAGAAAGCGTCTAGCCA SEQ ID No. 214 TGGCGTTA ST778AA-rev
5'-GCAAGCACCCTATCAGGC SEQ ID No. 215 AGTACCACAA ST280A-for-B
5'-GCAGAAAGCGTCTAGCCA SEQ ID No. 216 TGGCGTTAgTATG-SpC3
ST778AA-rev-B 5'-GCAAGCACCCTATCAGGC SEQ ID No. 217
AGTACCACAAgGCCT-SpC3 DNA bases are shown in uppercase. RNA bases
are shown in lowercase. SpC3 is a C3 spacer. The "B" designation
indicates a blocked, cleavable primer.
Cloned Synthetic Target (Primer Binding Sites are Underlined).
[0463] SEQ ID No. 218
Hepatitis C virus subtype 1b amplicon (242 bp):
TABLE-US-00145 gcagaaagcgtctagccatggcgttagtatgagtgtcgtgcagcct
ccaggaccccccctcccgggagagccatagtggtctgcggaaccgg
tgagtacaccggaattgccaggacgaccgggtcctttcttggacta
aacccgctcaatgcctggagatttgggcgtgcccccgcgagactgc
tagccgagtagtgttgggtcgcgaaaggccttgtggtactgcctga tagggtgcttgc
[0464] PCR reactions were done in 384 well format using a Roche
Lightcycler.RTM. 480 platform. Reactions comprised 1.times. New
England Biolabs (Beverly, Mass.) DyNAmo reaction mix with DyNAmo
DNA polymerase, 200 nM of each primer (For+Rev), with or without
1.3 mU of Pyrococcus abyssi RNase H2 in 10 .mu.l volume. Template
DNA was either 2000 copies of the linearized HCV plasmid amplicon
or no target control. Thermal cycling parameters included an
initial 2 minutes soak at 95.degree. C. and then 50 cycles were
performed of [95.degree. C. for 15 seconds+60.degree. C. for 30
seconds]. Samples were separated on an 8% polyacrylamide
non-denaturing gel and visualized using GelStar stain. Results are
shown in FIG. 25. The unblocked standard primers produced multiple
products having sizes ranging from 55 bp to 90 bp in size and no
desired full length product was seen. In the absence of RNase H2,
use of the blocked primers did not result in any amplified product.
With RNase H2, the blocked primers produced a single strong
amplicon of the expected size and no undesired small species were
seen.
[0465] The DyNAmo is a non hot-start DNA polymerase. Use of RNase
H2 blocked primer of the present invention with a hot-start RNase
H2 having reduced activity at lower temperatures eliminated
undesired primer-dimers from the reaction and resulted in formation
of the desired amplicon whereas standard unblocked primers failed
and produced only small, undesired species.
Example 18
Use of Detergent in RNase H2 Assay Buffers
[0466] The presence of detergent was found to be beneficial to
cleavage by the Pyrococcus abyssi RNase H2 enzyme. Different
detergents were tested at different concentrations to optimize the
reaction conditions.
[0467] Aliquots of each of the recombinant RNase H2 enzymes were
incubated with the single-stranded and double-stranded
oligonucleotide substrates indicated above in an 80 .mu.l reaction
volume in buffer 50 mM NaCl, 10 mM MgCl.sub.2, and 10 mM Tris pH
8.0 for 20 minutes at 70.degree. C. Reactions were stopped with the
addition of gel loading buffer (formamide/EDTA) and separated on a
denaturing 7M urea, 15% polyacrylamide gel. The RNA strand of the
substrate SEQ ID NOS 11 and 12 was radiolabeled with .sup.32P.
Reactions were performed using 100 nM substrate with 100 microunits
(.mu.U) of enzyme in Mg Cleavage Buffer with different detergents
at varying concentrations. Detergents tested included Triton-X100,
Tween-20, Tween-80, CTAB, and N-lauryol sarcosyl. Results with
Pyrococcus absii RNase H2 are shown in FIG. 26. Additional
experiments were done to more finely titrate CTAB detergent
concentration. Optimum levels of detergent to obtain highest enzyme
activity were (vol:vol): Triton-X100 0.01%, Tween-20 0.01%, and
CTAB 0.0013%. The detergents Tween-80 and N-lauryol sarcosyl did
not perform as well as the other detergents tested. Thus both
non-ionic (Triton, Tween) and ionic (CTAB) detergents can be
employed to stabilize thermophilic RNase H2 enzymes of the present
invention.
Example 19
Use of Fluorescence-Quenched (F/Q) Cleavable Primers in qPCR
[0468] In Example 9 above, it was demonstrated that cleavable
blocked primers function in PCR and further can be employed in
real-time quantitative PCR (qPCR) using SYBR green detection. In
this example we demonstrate use of fluorescence-quenched cleavage
primers where the primer itself generates detectable signal during
the course of the PCR reaction.
[0469] FIG. 18 illustrates the scheme for performing PCR using
blocked cleavable primers. FIG. 27 illustrates the scheme for
performing PCR using fluorescence-quenched cleavable primers. In
this case one primer in the pair is detectably labeled with a
fluorescent dye. A fluorescence quencher is positioned at or near
the 3'-end of the primer and effectively prevents priming and DNA
synthesis when the probe is intact. A single ribonucleotide base is
positioned between the dye and the quencher. Cleavage at the
ribonucleotide by RNase H2 separates the reporter and quencher,
removing quenching, resulting in a detectable signal.
Concomitantly, cleavage activates the primer and PCR proceeds.
[0470] The following synthetic oligonucleotides shown in Table 42,
were employed to demonstrate this reaction using a synthetic
template. As a control the 5'-nuclease Taqman.RTM. assay was
performed with unmodified primers and a standard
fluorescence-quenched probe. Three variants of the synthetic
fluorescence-quenched cleavable primers were compared, having 4, 5,
or 6 DNA bases 3' to the RNA base. It was previously established
that 4 DNA bases 3' to the RNA base was optimal using
oligonucleotide substrates having a C3 spacer or ddC end group. It
was possible that the presence of a bulky hydrophobic quencher
group at or near the 3'-end might change the optimal number of DNA
residues needed in this domain.
TABLE-US-00146 TABLE 42 Name Sequence SEQ ID No. Syn-For
5'-AGCTCTGCCCAAAGATTA SEQ ID No. 86 CCCTG Syn-Rev
5'-CTGAGCTTCATGCCTTTA SEQ ID No. 87 CTGT Syn-Probe
5'-FAM-TTCTGAGGCCAACT SEQ ID No. 219 CCACTGCCACTTA-IBFQ Syn-For
5'-FAM-CTGAGCTTCATGCC SEQ ID No. 220 F/Q-4D TTTACTGTuCCCC-IBFQ
Syn-For 5'-FAM-CTGAGCTTCATGCC SEQ ID No. 221 F/Q-5D
TTTACTGTuCCCCG-IBFQ Syn-For 5'-FAM-CTGAGCTTCATGCC SEQ ID No. 222
F/Q-6D TTTACTGTuCCCCGA-IBFQ
DNA bases are shown in uppercase. RNA bases are shown in lowercase.
FAM is 6-carboxyfluorescein. IBFQ is Iowa Black FQ, a dark
quencher.
Synthetic Template
[0471] SEQ ID No. 93
TABLE-US-00147 AGCTCTGCCCAAAGATTACCCTGACAGCTAAGTGGCAGTGGAAGTT
GGCCTCAGAAGTAGTGGCCAGCTGTGTGTCGGGGAACAGTAAAGGC ATGAAGCTCAG
[0472] PCR reactions were performed in 10 .mu.l volume using 200 nM
primers, 200 .mu.M of each dNTP (800 .mu.M total), 1 unit of iTaq
(BIO-RAD), 50 mM Tris pH 8.3, 50 mM KCl, and 3 mM MgCl.sub.2.
Reactions were run either with or without varying amounts of
Pyrococcus abyssi RNase H2 on a Roche Lightcycler.RTM. 480 platform
with 2.times.10.sup.6 copies of synthetic template/target
oligonucleotide (SEQ ID No. 93). Reactions were started with a soak
at 95.degree. C. for 5 minutes followed by 45 cycles of [95.degree.
C. for 10 seconds, 60.degree. C. for 30 seconds, and 72.degree. C.
for 1 second]. The For and Rev primers (SEQ ID Nos. 86 and 87) were
used with the internally placed DLP (SEQ ED No. 219).
Alternatively, the For primer (SEQ ID No. 86) was used with the FQ
primers (individually) (SEQ ID Nos. 220-222).
[0473] Use of the F/Q cleavable primers resulted in detectable
fluorescence signal in real time during PCR similar to that
obtained using the traditional dual-labeled probe (DLP) (SEQ ID No.
219) in the 5'-nuclease assay format. Primer SEQ ID No. 220, with 4
DNA residues 3' to the RNA base, showed delayed amplification
relative to the unmodified primers. Primers SEQ ID No. 221 and 222,
with 5 and 6 DNA bases 3' to the RNA base, were more efficient and
performed equally well. It is therefore preferable to use an
oligonucleotide design with 5 DNA bases 3' to the RNA base in this
assay format as opposed to the 3-4 DNA base design optimal when the
3'-blocking group is smaller. In previous Examples using a SYBR
Green assay format, 1.3 mU of RNase H2 resulted in priming
efficiency identical to unmodified primers. In the present F/Q
assay format, use of 1.3 mU of RNase H2 resulted in delayed
amplification whereas use of 2.6 mU of RNase H2 resulted in
identical results compared to unmodified primers. Increasing the
amount of RNase H2 for the F/Q assay format is therefore preferred.
Both amplification and detection of signal was RNase H2
dependent.
[0474] Examples of amplification plots for qPCR reactions run using
the 5'-nuclease assay DLP (SEQ ID No. 219) and the F/Q cleavable 5D
primer (SEQ ID No. 221) are shown in FIG. 28. It is evident that
amplification efficiency is similar between both methods as the Cp
values where fluorescence is first detected is identical (20.0).
Interestingly, the .DELTA.Rf (the magnitude of fluorescence signal
detected) peaked at slightly higher levels using the DLP than the
FQ primer. One possible explanation for the difference in maximal
fluorescence signal release is that the fluorescent dye on the FQ
primer remained partially quenched at the end of the reaction. In
the 5'-nuclease assay, the probe is degraded and the reporter dye
is released into the reaction mixture attached to a single-stranded
short nucleic acid fragment. In the FQ primer assay format the
fluorescent reporter dye remains attached to the PCR amplicon and
is in double-stranded format. DNA can quench fluorescein emission,
so this configuration might lower the final signal.
[0475] We therefore tested if changing dye/quencher configuration
on the primer would alter the fluorescence signal, comparing F/Q
vs. Q/F versions of the same primer. In the synthetic amplicon
assay used above, the preferred 5-DNA probe has a "G" residue
present at the 3'-end. G residues tend to quenche FAM, whereas
other bases have little effect on FAM fluorescence. The amplicon
was therefore modified to change this base. The sequences in Table
43, were synthesized and tested in a fluorescent real-time PCR
assay format.
TABLE-US-00148 TABLE 43 Name Sequence SEQ ID No. Syn-For
5'-AGCTCTGCCCAAAGATT SEQ ID No. 86 ACCCTG Syn-For(C)
5'-FAM-CTGAGCTTCATGC SEQ ID No. 223 F/Q-5D CTTTACTGTuCCCCC-IBFQ
Syn-For(C) 5'-IBFQ-CTGAGCTTCATG SEQ ID No. 224 Q/F-5D
CCTTTACTGTuCCCCC-FAM DNA bases are shown in uppercase. RNA bases
are shown in lowercase. FAM is 6-carboxyfluorescein. IBFQ is Iowa
Black FQ, a dark quencher.
Synthetic Template
[0476] SEQ ID No. 225
TABLE-US-00149 AGCTCTGCCCAAAGATTACCCTGACAGCTAAGTGGCAGTGGAAGTT
GGCCTCAGAAGTAGTGGCCAGCTGTGTGTGGGGGAACAGTAAAGGC ATGAAGCTCAG
[0477] PCR reactions were performed in 10 .mu.l volume using 200 nM
primers, 200 .mu.M of each dNTP (800 .mu.M total), 1 unit of iTaq
(BIO-RAD), 50 mM Tris pH 8.3, 50 mM KCl, and 3 mM MgCl.sub.2.
Reactions were run with 2.6 mU of Pyrococcus abyssi RNase H2 on a
Roche Lightcycler.RTM. 480 platform with 2.times.10.sup.6 copies of
synthetic template/target oligonucleotide (SEQ ID No. 225).
Reactions were started with a soak at 95.degree. C. for 5 minutes
followed by 45 cycles of [95.degree. C. for 10 seconds, 60.degree.
C. for 30 seconds, and 72.degree. C. for 1 second]. The For primer
(SEQ ID No. 86) was used with either the FQ primer (SEQ ID No. 223)
or the QF primer (SEQ ID No. 224).
[0478] Use of the F/Q and Q/F cleavable primers resulted in an
identical Cp, indicating that both primers performed with equal
efficiency in the reaction. As predicted, the Q/F primer showed
increased ARf relative to the F/Q primer. Both versions of the
primer work equally well in the assay.
Example 20
Use of Fluorescence-Quenched (F/Q) Cleavable Primers in Multiplex
qPCR
[0479] Multiplex assays are commonly employed today to streamline
experiments and increase throughput. It is particularly common to
combine a qPCR assay specific for an experimental gene of interest
with a second qPCR assay specific for an internal reference control
gene for normalization purposes. One weakness of SYBR Green
detection for qPCR is that multiplex reactions are not possible.
The use of dye-labeled fluorescence-quenched probes or primers does
permit such multiplex reactions to be run. Real time PCR cycling
and detection equipment is available today that permits combination
of 2, 3, or 4 different fluorophores into the same reaction tube.
This example demonstrates the utility of fluorescence-quenched
(F/Q) cleavable primers in multiplex qPCR.
[0480] The following oligonucleotide reagents shown in Table 44,
were synthesized to perform multiplex qPCR using either a
dual-labeled probe with the 5'-nuclease assay or an F/Q cleavable
primer. One assay was specific for the human MYC gene
(NM.sub.--002476) and the second assay was specific for the human
SFRS9 gene (NM.sub.--003769), a splicing factor which is a commonly
used internal normalization control gene.
TABLE-US-00150 TABLE 44 Name Sequence SEQ ID No. MYC-For
5'-TCGGATTCTCTGCTCT SEQ ID No. 226 CCT MYC-Rev 5'-CCTCATCTTCTTGTTC
SEQ ID No. 227 CTCC MYC-Probe 5'-FAM-CCACCACCAGCA SEQ ID No. 228
GCGACTCTGA-IBFQ MYC-For-FQ 5'-FAM-TCGGATTCTCTG SEQ ID No. 229
CTCTCCTcGACGG-IBFQ MYC-Rev-B 5'-CCTCATCTTCTTGTTC SEQ ID No. 230
CTCCuCAGA-SpC3 SFRS9-For 5'-TGTGCAGAAGGATGGA SEQ ID No. 231 GT
SFRS9-Rev 5'-CTGGTGCTTCTCTCAG SEQ ID No. 232 GATA SFRS9-Probe
5'-MAX-TGGAATATGCCC SEQ ID No. 233 TGCGTAAACTGGA-IBFQ SFRS9-For-FQ
5'-MAX-TGTGCAGAAGGA SEQ ID No. 234 TGGAGTgGGGAT-IBFQ SFRS9-Rev-B
5'-CTGGTGCTTCTCTCAG SEQ ID No. 235 GATAaACTC-SpC3 DNA bases are
shown in uppercase. RNA bases are shown in lowercase. FAM is
6-carboxyfluorescein. IBFQ is Iowa Black FQ, a dark quencher. MAX
is a red reporter dye. SpC3 is a C3 spacer.
[0481] PCR reactions were performed in 10 .mu.l volume using 200 nM
primers (and probe where appropriate), 200 .mu.M of each dNTP (800
.mu.M total), 1 unit of iTaq (BIO-RAD), 50 mM Tris pH 8.3, 50 mM
KCl, and 3 mM MgCl.sub.2. Reactions were run with 10 mU of
Pyrococcus abyssi RNase H2 on a Roche Lightcycler.RTM. 480 platform
with 2 ng of cDNA made from total HeLa cell RNA. Reactions were
started with a soak at 95.degree. C. for 5 minutes followed by 45
cycles of [95.degree. C. for 10 seconds, 60.degree. C. for 30
seconds, and 72.degree. C. for 1 second].
[0482] The multiplex reactions for the 5'-nuclease assays included
the MYC For and Rev primers+MYC probe (SEQ ID Nos. 226-228) and the
SFRS9 For and Rev primers+SFRS9 probe (SEQ ID Nos. 231-233). The
multiplex reactions for the FQ-cleavable primer assays included the
MYC-For-FQ and MYC-Rev-B blocked primers (SEQ ID Nos. 229 and 230)
and the SFRS9-For-FQ and SFRS9-Rev-B blocked primers (SEQ ID Nos.
234 and 235). All assays were also run in singleplex format for
comparison. The FAM primers and probes were detected in the
fluorescein dye channel while the MAX primers and probes were
detected in the HEX dye channel. Both the multiplexed DLP
5'-nuclease assays and the multiplexed FQ-cleavable primer assays
worked well and resulted in very similar data, which is summarized
in Table 45 below.
TABLE-US-00151 TABLE 45 Multiplex qPCR reactions for MYC and SFRS9
Cp Value Cp Value Reaction FAM Channel HEX Channel MYC FAM DLP 25.7
-- SFRS9 MAX DLP -- 24.8 MYC FAM DLP + 24.6 23.9 SFRS9 MAX DLP MYC
FAM FQ-Primer 27.2 -- SFRS9 MAX FQ Primer -- 28.0 MYC FAM FQ-Primer
+ 27.9 26.1 SFRS9 MAX FQ Primer
[0483] RNase H concentration was titrated and higher levels of
enzyme were needed to maintain reaction efficiency in multiplex
format. For example, blocked primers in singleplex SYBR Green
detection format required 1.3 mU of enzyme. Blocked FQ primers in
singleplex format required 2.6 mU of enzyme. Blocked FQ primers in
multiplex format required 10 mU of enzyme. It is therefore
important to titrate the amount of RNase H2 enzyme employed when
cleavable primers are used in different assay formats.
[0484] Another application where use of multiplex probes is common
practice is allelic discrimination SNPs. The following assay was
designed to distinguish a SNP pair for the SMAD7 gene at a site
that is known to be relevant for development of colorectal
carcinoma, rs4939827. FQ blocked primers were designed and
synthesized at this site using the standard design features taught
in the above examples without any further optimization to
discriminate between the "C" and "T" alleles in this gene.
Sequences are shown below in Table 46.
TABLE-US-00152 TABLE 46 Name Sequence SEQ ID No. rs4939827 Rev
5'-CTCACTCTAAACCCCAG SEQ ID No. 236 CATT rs4939827
5'-FAM-CAGCCTCATCCAA SEQ ID No. 237 C-FAM-FQ-For
AAGAGGAAAcAGGA-IBFQ rs4939827 5'-HEX-CAGCCTCATCCAA SEQ ID No. 238
T-HEX-FQ-For AAGAGGAAAuAGGA-IBFQ DNA bases are shown in uppercase.
RNA bases are shown in lowercase. FAM is 6-carboxyfluorescein. IBFQ
is Iowa Black FQ, a dark quencher. MAX is a red reporter dye.
[0485] The above primers target the following 85 bp region of the
SMAD7 gene (NM.sub.--005904). Primer binding sites are underlined
and the SNP location is highlighted as bold italic.
rs4939827 (SMAD7) C allele (SEQ ID No. 239)
TABLE-US-00153 CAGCCTCATCCAAAAGAGGAAA AGGACCCCAGAGCTCCCTCAGA
CTCCTCAGGAAACACAGACAATGCTGGGGTTTAGAGTGAG
rs4939827 (SMAD7) T allele (SEQ ID No. 240)
TABLE-US-00154 CAGCCTCATCCAAAAGAGGAAA AGGACCCCAGAGCTC
CCTCAGACTCCTCAGGAAACACAGACAATGCTGGGGTT TAGAGTGAG
[0486] PCR reactions were performed in 10 .mu.l volume using 200 nM
FQ-For and unmodified Rev primers, 200 .mu.M of each dNTP (800
.mu.M total), 1 unit of iTaq (BIO-RAD), 50 mM Tris pH 8.3, 50 mM
KCl, and 3 mM MgCl.sub.2. Reactions were run with 2.6 mU of
Pyrococcus abyssi RNase H2 on a Roche Lightcycler.RTM. 480 platform
with 2 ng of target DNA. Target DNA was genomic DNA made from cells
homozygous for the two SMAD7 alleles (Coreill 18562 and 18537). The
"C" and "T" alleles (SEQ ID Nos. 239 and 240) were tested
individually (homozygote) and together (heterozygote). Reactions
were started with a soak at 95.degree. C. for 5 minutes followed by
45 cycles of [95.degree. C. for 10 seconds, 60.degree. C. for 30
seconds, and 72.degree. C. for 1 second]. Data acquisition was set
for multiplex mode detecting the FAM and HEX channels.
[0487] Results are shown in FIG. 30. It is clear that the
FAM-labeled "C" probe detected the presence of the "C" target DNA
but not the "T" target DNA and that the HEX "T" probe detected the
presence of the "T" target DNA but not the "C" target DNA. Thus FQ
cleavable primers can be used in multiplex formats to distinguish
SNPs.
Example 21
Use of Fluorescence-Quenched Cleavable Primers in the Primer-Probe
Assay
[0488] We previously described a method of detecting nucleic acid
samples using fluorescence quenched primers comprising two distinct
but linked elements, a Reporter Domain positioned towards the
5'-end and a Primer Domain, positioned at the 3'-end of the nucleic
acid molecule (US patent application US 2009/0068643). The Primer
Domain is complementary to and will bind to a target nucleic acid
under conditions employed in PCR. It is capable of priming DNA
synthesis using the complementary target as template, such as in
PCR. The Reporter Domain comprises a sequence which can be
complementary to the target or can be unrelated to the target
nucleic acid and does not hybridize to the target. Furthermore the
Reporter Domain includes a detectable element, such as a
fluorescent reporter dye, and a quencher. The reporter dye and
quencher are separated by a suitable number of nucleotides such
that fluorescent signal from the reporter dye is effective
suppressed by the quencher when the Reporter Domain is in
single-stranded random coil conformation. During PCR, the Primer
Domain will prime DNA synthesis and the FQT synthetic
oligonucleotide is thereby incorporated into a product nucleic
acid, which itself can is used as template in the next cycle of
PCR. Upon primer extension during the next cycle of PCR, the entire
FQT probe is converted to double-stranded form, including the
Reporter Domain. Formation of a rigid double-stranded duplex
physically increases the distance between the fluoropohore and the
quencher, decreasing the suppression of fluorescence emission
(hence increasing fluorescent intensity). Thus conversion of the
FQT primer to double-stranded form during PCR constitutes a
detectable event. Further increases in fluorescent signal can be
achieved by cleavage of the Reporter Domain at a site between the
reporter dye and the quencher, such that the reporter dye and the
quencher become physically separated and are no longer covalently
linked on the same nucleic acid molecule. This cleavage event is
dependent upon formation of double-stranded nucleic acid sequence
so that cleavage cannot occur if the FQT primer is in its original
single-stranded state. Suitable methods to separate reporter and
quencher include, for example, use of a restriction endonuclease to
cleave at a specific sequence in dsDNA. Alternatively, an RNase H2
cleavage domain can be placed between the fluorophore and quencher.
Placement of a single ribonucleotide residue between the
fluorophore and the quencher would make the FQT primer a suitable
substrate for RNase H2 during PCR. The scheme for this reaction is
shown in FIG. 31. The present example demonstrates use of a
thermostable RNase H2 to mediate cleavage of a
fluorescence-quenched primer in a primer-probe real time PCR
assay.
[0489] A qPCR assay was designed for the human Drosha gene
including unmodified For and Rev primers with an internally
positioned dual-labeled probe suitable for use in the 5'-nuclease
assay. The For primer was also synthesized as an FQT forward primer
using the same Primer Domain sequence as the unmodified For primer
and adding a Reporter Domain on the 5'-end comprising a reporter
dye (Fluorescein-dT) and a dark quencher (IBFQ) separated by 11
bases including a centrally positioned rU base (cleavage site).
Sequences are shown below in Table 47.
TABLE-US-00155 TABLE 47 Name Sequence SEQ ID No. Drosha-For
5'-ACCAACGACAAGAC SEQ ID No. 241 CAAGAG Drosha-Rev
5'-TCGTGGAAAGAAGC SEQ ID No. 242 AGACA Drosha-probe
5'-FAM-ACCAAGACCT SEQ ID No. 243 TGGCGGACCTTT-IBFQ Drosha-For-FQT
5'-IBFQ-TTTCCuGGT SEQ ID No. 244 TT(Fl-dT)ACCAACGA CAAGACCAAGAG DNA
bases are shown in uppercase. RNA bases are shown in lowercase.
Fl-dT in an internal Fluorescein-dT modified base. IBFQ is Iowa
Black FQ, a dark quencher. The portion of the FQT probe that is
complementary to the Drosha target is underlined (i.e., the Primer
Domain).
[0490] The above primers target the following 141 bp region of the
human Drosha gene (RNASEN, NM.sub.--013235). Primer binding sites
are underlined and the internal probe binding site for the
5'-nuclease assay is in bold font.
Drosha amplicon (SEQ ID No. 245)
TABLE-US-00156 ACCAACGACAAGACCAAGAGGCCTGTGGCGCTTCGCAC
CAAGACCTTGGCGGACCTTTTGGAATCATTTATTGCAG
CGCTGTACATTGATAAGGATTTGGAATATGTTCATACT
TTCATGAATGTCTGCTTCTTTCCACGA
[0491] 5'-Nuclease qPCR reactions were performed in 10 .mu.l volume
using 200 nM unmodified For and Rev primers with 200 nM probe, 200
.mu.M of each dNTP (800 .mu.M total), 1 unit of iTaq (BIO-RAD), 50
mM Tris pH 8.3, 50 mM KCl, and 3 mM MgCl.sub.2. FQT qPCR reactions
were performed in 10 .mu.l volume using 200 nM FQT-For primer and
200 nM unmodified Rev primer, 200 .mu.M of each dNTP (800 .mu.M
total), 1 unit of iTaq (hot start thermostable DNA polymerase,
BIO-RAD), 50 mM Tris pH 8.3, 50 mM KCl, and 3 mM MgCl.sub.2.
Reactions were run with or without 2.6 mU of Pyrococcus abyssi
RNase H2 on a Roche Lightcycler.RTM. 480 platform. Reactions were
run with or without 10 ng of cDNA made from HeLa total cellular
RNA. Reactions were started with a soak at 95.degree. C. for 5
minutes followed by 45 cycles of [95.degree. C. for 10 seconds,
60.degree. C. for 30 seconds, and 72.degree. C. for 1 second].
[0492] Results for the 5'-nuclease qPCR reaction are shown in FIG.
32A. A positive signal was seen at cycle 26. Results for the FQT
primer qPCR reactions are shown in FIG. 32B. A positive signal was
seen at cycle 27, nearly identical to the 5'-nuclease assay
results. In this case, signal was dependent upon RNase H2 cleavage.
Thus cleavage at an internal RNA residue by RNase H2 can be used to
generate signal from FQT primers that have a distinct
fluorescence-quenched reporter domain.
Example 22
Use of Modified Bases in Cleavable Blocked Primers to Improve
Mismatch Discrimination
[0493] We demonstrated that blocked cleavable primers can be used
in qPCR to distinguish single base mismatches in the SYBR Green
assay format in Example 13 and in the fluorescence-quenched (FQ)
assay format in Example 20. Depending upon the precise base
mismatch and the sequence context, detectable signal for the
mismatch target occurred from 5 to 15 cycles after detection of the
perfect match target. There may be circumstances where greater
levels of mismatch discrimination are desired, such as detection of
a rare mutant allele in the background of predominantly wild type
cells. We demonstrate in this example that selective placement of
2'OMe RNA modified residue within the cleavable primer can improve
mismatch discrimination.
[0494] Example 5 above demonstrated that modified bases could be
compatible with cleavage of a heteroduplex substrate by RNase H2
depending upon the type of modification employed and placement
relative to the cleavage site. Here we demonstrate in greater
detail use of the 2'OMe modification in blocked primers having a
single unmodified ribonucleotide base. The following primers, shown
below in Table 48, were synthesized and used in qPCR reactions in
the SYBR Green format with a synthetic oligonucleotide template.
Blocked cleavable primers having a single rU residue were
synthesized either without additional modification (SEQ ID No. 134)
or with a 2'OMe base 5'- to the rU (SEQ ID No. 247) or with a 2'OMe
base 3'- to the rU (SEQ ID No. 248). If the 2'OMe residue is
positioned 5'- to the ribonucleotide, then it will remain in the
final primer which results from cleavage by RNase H2. Therefore a
Syn-Rev-mU primer was made specific for the synthetic template
bearing a 3'-2'OMe U residue at the 3'-end to mimic this reaction
product (SEQ ID No. 246).
TABLE-US-00157 TABLE 48 Name Sequence SEQ ID No. Syn-For
5'-AGCTCTGCCCA SEQ ID No. 86 AAGATTACCCTG Syn-Rev 5'-CTGAGCTTCAT
SEQ ID No. 87 GCCTTTACTGT Syn-Rev-mU 5'-CTGAGCTTCAT SEQ ID No. 246
GCCTTTACTG(mU) Syn-Rev-rU-C3 5'-CTGAGCTTCAT SEQ ID No. 134
GCCTTTACTGTuCC CC-SpC3 Syn-Rev-mUrU-C3 5'-CTGAGCTTCAT SEQ ID No.
247 GCCTTTACTG(mU) uCCCC-SpC3 Syn-Rev-rUmC-C3 5'-CTGAGCTTCAT SEQ ID
No. 248 GCCTTTACTGTu (mC)CCC-SpC3 DNA bases are shown in uppercase.
RNA bases are shown in lowercase. 2'OMe RNA bases are indicated as
(mN).
[0495] The following synthetic oligonucleotide was used as
template. Primer binding sites are underlined.
Synthetic template, SEQ ID No. 162:
TABLE-US-00158 AGCTCTGCCCAAAGATTACCCTGACAGCTAAGTGGCAG
TGGAAGTTGGCCTCAGAAGTAGTGGCCAGCTGTGTGTC
GGGGAACAGTAAAGGCATGAAGCTCAG
[0496] PCR reactions were performed in 10 .mu.l volume using 200 nM
unmodified For primer pairwise with 200 nM of each of the different
Rev primers shown above in Bio-Rad SYBR Green master mix. Reactions
were run with or without 1.3-200 mU of Pyrococcus abyssi RNase H2
on a Roche Lightcycler.RTM. 480 platform with no target or
2.times.10.sup.6 copies of the synthetic oligonucleotide template.
Reactions were started with a soak at 95.degree. C. for 5 minutes
followed by 45 cycles of [95.degree. C. for 10 seconds, 60.degree.
C. for 20 seconds, and 72.degree. C. for 30 seconds]. Results are
summarized in Table 49.
TABLE-US-00159 TABLE 49 Cp values of qPCR reactions comparing
blocked primers with or without a 2'OMe base flanking a cleavable
ribonucleotide. Syn-Rev- Syn-Rev- Syn-Rev- Syn-Rev Syn-Rev-mU rU-C3
mUrU-C3 rUmC-C3 RNase SEQ ID No. SEQ ID No. SEQ ID SEQ ID SEQ ID H2
No. 87 246 No. 134 No. 247 No. 248 None 17.8 19.8 >40 >40
>40 50 mU 17.8 19.8 17.2 21.6 >40 100 mU 17.8 19.6 17.2 19.7
>40 150 mU 17.8 19.9 17.2 19.5 >40 200 mU 17.8 19.8 17.2 19.1
>40
[0497] The unblocked primer with a 3'-terminal 2'OMe base (SEQ ID
No. 246) showed a 2 cycle delay relative to the unmodified primer
(SEQ ID No. 87), indicating that the terminal 2'OMe base slightly
decreased priming efficiency but nevertheless was functional as a
PCR primer. The blocked primer containing a single rU base (SEQ ID
No. 134) performed as expected (see Example 13) and worked well
with low concentrations of RNase H2 (data not shown). For the 2'
OMe RNA containing primers a higher concentration of RNase H2 was
needed. The primer having a 2'OMe residue 5'- to the ribonucleotide
(SEQ ID No. 247) showed good activity at 50 mU RNase H2 and
performed identically to the unblocked 2'OMe control primer (SEQ ID
No. 246) when 100 mU or higher RNase H2 was employed. The primer
having a 2' OMe residue 3'- to the ribonucleotide (SEQ ID No. 248)
did not function at any level of RNase H2 tested. The primer having
a 2' OMe residue 5'- to the ribonucleotide (SEQ ID No. 247) was
next tested in a mismatch discrimination qPCR assay.
[0498] The standard configuration blocked RNase H2 cleavable primer
(SEQ ID No. 134) was compared with the 5'-2'OMe version of this
sequence (SEQ ID No. 247). These two "Rev" primers were used with
the unmodified "For" primer (SEQ ID No. 86) together with 3
different synthetic oligonucleotide templates (originally used in
defining mismatch discrimination potential in Example 13). These
templates provide a perfect match control (Template SEQ ID No.
162), a T/U mismatch (Template SEQ ID No. 155), or a G/U mismatch
(Template SEQ ID No. 176). The 3 templates oligonucleotides are
shown below with the cleavable blocked primer (SEQ ID No. 134)
aligned beneath to illustrate the regions of match vs.
mismatch.
Synthetic Template, SEQ ID No. 162 (A:U Match):
TABLE-US-00160 [0499] AGCTCTGCCCAAAGATTACCCTGACAGCTAAGTGGCA
GTGGAAGTTGGCCTCAGAAGTAGTGGCCAGCT
GTGTGTCGGGGAACAGTAAAGGCATGAAGCTCAG-3' |||||||||||||||||||||||||||
3'-C3-CCCCuTGTCATTTCCGTACTTCGAGTC-5'
Synthetic Template, SEQ ID No. 155 (T:U Mismatch):
TABLE-US-00161 [0500] AGCTCTGCCCAAAGATTACCCTGACAGCTAAGTGGCA
GTGGAAGTTGGCCTCAGAAGTAGTGGCCAGCT
GTGTGTCGGGGTACAGTAAAGGCATGAAGCTCAG-3' |||| ||||||||||||||||||||||
3'-C3-CCCCuTGTCATTTCCGTACTTCGAGTC-5'
Synthetic template, SEQ ID No. 176 (G:U mismatch):
TABLE-US-00162 AGCTCTGCCCAAAGATTACCCTGACAGCTAAGTGGCA
GTGGAAGTTGGCCTCAGAAGTAGTGGCCAGCT
GTGTGTCGGGGGACAGTAAAGGCATGAAGCTCAG-3' |||| ||||||||||||||||||||||
3'-C3-CCCCuTGTCATTTCCGTACTTCGAGTC-5'
[0501] PCR reactions were performed in 10 .mu.l volume using 200 nM
unmodified For primer with 200 nM of cleavable blocked Rev primer
(SEQ ID No. 134) or 5'mU containing cleavable blocked Rev primer
(SEQ ID No. 247) in Bio-Rad SYBR Green master mix. Reactions were
run with 1.3 mU (primer SEQ ID No. 134) or 100 mU (primer SEQ ID
No. 247) of Pyrococcus abyssi RNase H2. Reactions were run on a
Roche Lightcycler.RTM. 480 platform with 2.times.10.sup.6 copies of
the different synthetic oligonucleotide templates (SEQ ID Nos. 155,
162, or 176). Reactions were started with a soak at 95.degree. C.
for 5 minutes followed by 45 cycles of [95.degree. C. for 10
seconds, 60.degree. C. for 20 seconds, and 72.degree. C. for 30
seconds]. Results are summarized in Table 50 and are shown as
.DELTA.Cp (.DELTA.Cp=Cp mismatch-Cp match).
TABLE-US-00163 TABLE 50 Cp values of qPCR reactions comparing
mismatch discrimination of blocked primers with or without a 2'OMe
base on the 5'-side of an RNA residue. 1.3 mU RNase H2 100 mU RNase
H2 SEQ ID No. 134 SEQ ID No. 247 rU primer mUrU primer Match (A:U)
0 0 Mismatch (T:U) 5.3 12.7 Mismatch (G:U) 10.9 14.4 (.DELTA.Cp =
Cp mismatch - Cp match)
[0502] In both cases tested, addition of a 2'OMe residue directly
5' to the cleavable ribonucleotide significantly improved mismatch
discrimination. The T/U mismatch improved from a .DELTA.Cp of 5.3
to 12.7 and the G/U mismatch improved from a .DELTA.Cp of 10.9 to
14.4. This new primer design required use of 100 mU of RNase H2
compared with 1.3 mU (in a 10 ul assay), however the enzyme is
inexpensive and the boost in reaction specificity was considerable.
We conclude that the use of chemically modified residues in select
positions within the cleavable primer can significantly improve the
mismatch discrimination capability of the assay.
Example 23
Use of Double-Mismatch Design in Cleavable Blocked Primers to
Improve Mismatch Discrimination
[0503] Some nucleic acid probes that are complementary to a wild
type (WT) sequence will bind to both the perfect match WT target
and a mutant target bearing a single base mismatch with
sufficiently similar affinity that the two sequences (WT and
mutant) are not easily distinguished. While a single mismatch
introduced between the probe and target sequence may not
significantly disrupt binding to the wild type target (which has 1
mismatch with the probe) disrupts binding to the mutant target
(which now has 2 mismatches with the probe). This strategy has been
used to improve selectivity of hybridization based assays as well
as assays dependent upon interaction with nucleic acid binding
proteins. The present example demonstrates use of a double-mismatch
strategy to improve base discrimination with use of
cleavable-blocked primers of the present invention.
[0504] For the present study, the SMAD7 qPCR SNP discrimination
assay presented in Example 20 was employed as a model system,
except that the SYBR Green detection format was used instead of the
FQ format. Blocked-cleavable primers were synthesized with the base
mismatch in the positioned at the cleavable ribonucleotide. Using
the present probe design, any mismatch placed 5'- to the cleavage
site (RNA base) will be retained in the primer extension product
and thus will be replicated during PCR. In order to maintain the
presence of the double-mismatch during PCR, the new mismatch must
be positioned 3'- to the cleavable RNA residue in the domain that
is cleaved off and is not retained in daughter products. It is
desirable that the intentionally added second mismatch not disrupt
function of the primer with a perfect match target. It was
demonstrated in Example 13 that mismatches present in the "+1
position" (i.e., immediately 3'- to the RNA base) can have a
significant impact upon cleavage and functional primer efficiency.
The double mismatch was therefore placed at the "+2 position" 3'-
to the RNA base with the expectation that this configuration would
not be disruptive as a single mismatch but would be disruptive as a
double mismatch.
[0505] Blocked-cleavable primers were designed and synthesized at
this site using standard design features to discriminate between
the "C" and "T" alleles in the SMAD7 gene (SNP locus rs4939827).
The same unmodified Rev primer was used in all assays (SEQ ID No.
236). The perfect match "C" allele primer is SEQ ID No. 250 and the
perfect match "T" allele primer is SEQ ID No. 254. Next, a series
of primers were made bearing a mutation at position+2 relative to
the ribonucleotide (2 bases 3'- to the RNA residue). It was
anticipated that the identity of the base mismatch would alter the
relative perturbation that having a mismatch at this position would
introduce into the assay. Therefore, perfect match (wild type) and
all 3 possible base mismatches were synthesized and studied (SEQ ID
Nos. 251-253 and 255-257). Sequences are shown below in Table
51.
TABLE-US-00164 TABLE 51 Name Sequence SEQ ID No. rs4939827
5'-CTCACTCTAAACCCC SEQ ID No. 236 Rev AGCATT rs4939827
5'-CAGCCTCATCCAAAA SEQ ID No. 249 For GAGGAAA rs4939827
5'-CAGCCTCATCCAAAA SEQ ID No. 250 C-For WT GAGGAAAcAGGA-SpC3
rs4939827 5'-CAGCCTCATCCAAAA SEQ ID No. 251 C-For CAA
GAGGAAAcAAGA-SpC3 rs4939827 5'-CAGCCTCATCCAAAA SEQ ID No. 252 C-For
CAC GAGGAAAcACGA-SpC3 rs4939827 5'-CAGCCTCATCCAAAA SEQ ID No. 253
C-For CAT GAGGAAAcATGA-SpC3 rs4939827 5'-CAGCCTCATCCAAAA SEQ ID No.
254 T-For WT GAGGAAAuAGGA-SpC3 rs4939827 5'-CAGCCTCATCCAAAA SEQ ID
No. 255 T-For UAA GAGGAAAuAAGA-SpC3 rs4939827 5'-CAGCCTCATCCAAAA
SEQ ID No. 256 T-For UAC GAGGAAAuACGA-SpC3 rs4939827
5'-CAGCCTCATCCAAAA SEQ ID No. 257 T-For UAT GAGGAAAuATGA-SpC3 DNA
bases are shown in uppercase. RNA bases are shown in lowercase.
SpC3 is a spacer C3 used as a 3'-blocking group. Mutations
introduced to create double-mismatches are indicated with bold
underline.
[0506] The above primers target the following 85 bp region of the
SMAD7 gene (NM.sub.--005904). Primer binding sites are underlined
and the SNP location is highlighted as bold italic. Primers are
aligned with target in FIG. 33 to help illustrate the scheme of the
"double mutant" approach to improve SNP discrimination.
Rs4939827 (SMAD7) C allele (SEQ ID No. 239)
TABLE-US-00165 CAGCCTCATCCAAAAGAGGAAA AGGACCCCAGAGC
TCCCTCAGACTCCTCAGGAAACACAGACAATGCTGG GGTTTAGAGTGAG
Rs4939827 (SMAD7) T allele (SEQ ID No. 240)
TABLE-US-00166 CAGCCTCATCCAAAAGAGGAAA AGGACCCCAGAGC
TCCCTCAGACTCCTCAGGAAACACAGACAATGCTGGG GTTTAGAGTGAG
[0507] PCR reactions were performed in 10 .mu.l volume using 200 nM
of the unmodified Rev primer (SEQ ID No. 236) and the series of
cleavable blocked For primers (SEQ ID Nos. 250-257) in Bio-Rad SYBR
Green master mix. Reactions were run with 2.6 mU of Pyrococcus
abyssi RNase H2 on a Roche Lightcycler.RTM. 480 platform with 2 ng
of target DNA. Target DNA was genomic DNA made from cells
homozygous for the two SMAD7 alleles (Coreill 18562 and 18537). The
"C" and "T" alleles (SEQ ID Nos. 239 and 240) were tested
individually. Reactions were started with a soak at 95.degree. C.
for 5 minutes followed by 80 cycles of [95.degree. C. for 10
seconds, 60.degree. C. for 30 seconds, and 72.degree. C. for 1
second]. Results are shown in Table 52 below.
TABLE-US-00167 TABLE 52 Cp and .DELTA.Cp values of qPCR reactions
comparing mismatch discrimination of blocked primers with or
without a second mutation at position +2 relative to the
ribonucleotide. Cp "C" Allele Cp "T" Allele Cp SEQ ID No. Unblocked
27.5 26.5 -- 249 control SEQ ID No. rCAG (WT) 29.2 39.9 10.7 250
SEQ ID No. rCAA 29.0 47.8 18.8 251 SEQ ID No. rCAC 31.6 45.4 13.8
252 SEQ ID No. rCAT 30.2 42.8 12.6 253 SEQ ID No. rUAG (WT) 42.6
29.2 13.4 254 SEQ ID No. rUAA 49.3 40.1 9.2 255 SEQ ID No. rUAC
74.1 49.9 24.2 256 SEQ ID No. rUAT 62.5 45.3 17.2 257 (.DELTA.Cp =
Cp mismatch - Cp match)
[0508] For the "C" allele, the standard design perfectly matched
probe (SEQ ID No. 250) showed amplification efficiency similar to
unmodified control primers and the mismatch discrimination was 10.7
cycles (.DELTA.Cp=10.7) against the "T" target. The mismatch
primers showed a minor decrease in detection efficiency with the
"C" allele target (a shift of up to 2.4 cycles was observed) but
mismatch discrimination at the SNP site increased significantly
with a .DELTA.Cp of 18.8 cycles seen for the rCA primer (SEQ ID No.
251).
[0509] For the "T" allele, the standard design perfect match probe
(SEQ ID No. 254) also showed amplification efficiency similar to
unmodified control primers and the mismatch discrimination was 13.4
cycles (.DELTA.Cp=13.4) against the "C" target. However, unlike the
"C" allele, the mismatch primers for the "T" allele showed a large
decrease in detection efficiency with the "T" allele target. Shifts
as large as 20 cycles were observed. Nevertheless the relative SNP
discrimination was improved with a .DELTA.Cp of 24.2 cycles seen
for the rUA primer (SEQ ID No. 256). For this region of the SMAD7
gene, the "T" allele creates an "AT-rich" stretch at the site of
the cleavable RNA base and this sequence has low thermal stability.
The presence of a mismatch at the +2 position must destabilize the
structure in this region much more for the "T" allele than the
higher stability "C" allele, which would account for the observed
increase Cp for the "T" allele probes against the "T" target.
However, this shift in Cp values does not limit utility of the
assay. Given the inherent increased specificity of the
blocked-cleavable primers (see Example 11), there should be no
problem with routinely extending reactions to 60-80 or more cycles.
In certain settings, the increased discrimination power of the
double-mismatch format will be of sufficient value to accept the
lower overall reaction efficiency. In "AT-rich" regions like the
SMAD7 "T" allele, it might also be useful to position the double
mismatch at the +3 position, removing its disruptive effects
further from the cleavable ribonucleotide.
Example 24
Identity of Reaction Products Made by PCR Amplification at SNP
Sites Using Cleavable-Blocked Primers
[0510] For use in PCR or any primer extension application, if a
base mismatch (SNP site) is positioned directly at the
ribonucleotide residue in blocked-cleavable primers, then a
cleavage event that occurs 5'- to the RNA base will result in a
primer extension product that reproduces the base variant present
in the template nucleic acid. A cleavage event that occurs 3'- to
the RNA base will result in a primer extension product that changes
the product to the RNA base present in the primer, creating an
error that will be replicated in subsequent PCR cycles. Cleavage on
the 3'-side of the ribonucleotide is therefore an undesired event.
Given the enormous amplification power of PCR, even a small amount
of 3'-cleavage could lead to the accumulation of a sizeable amount
of products containing a sequence error. For example, cleavage at a
rate of 0.1% would lead to 1 out of 1000 molecules having the
"wrong" base at the site of the RNA residue which would then be
detectable as "perfect match" in subsequent PCR cycles. This would
equate to a 10 cycle shift (.DELTA.Cp=10) in a qPCR reaction. Using
the design parameters taught in Example 13, cycle shifts for SNP
discrimination varied from 5-15. Thus a small amount of undesired
and unsuspected 3'-cleavage could easily account for the delayed
false-positive signals seen in Example 13 during SNP
interrogation.
[0511] A false positive signal in an allele-specific SNP
discrimination reaction could arise from two sources. First,
ongoing inefficient cleavage at the "normal" RNase H2 cleavage site
at the 5'-side of the RNA base (see FIG. 3) in spite of the
mismatch. This reaction will result in primer extension products
identical to the starting target. Second, a false positive signal
in an allele-specific SNP discrimination reaction could also arise
from inefficient cleavage at an "abnormal" position anywhere on the
3'-side of the RNA base. This reaction would produce primer
extension products identical to the primer and which would then
amplify with high efficiency using this same primer. If the first
scenario were correct, then the products from a reaction performed
using allele "A" primer with allele "B" target should produce
mostly allele "B" products, which would continue to amplify
inefficiently with allele "A" primers. If the second scenario were
correct, then the products from a reaction performed using allele
"A" primer with allele "B" target should produce mostly allele "A"
products, which would amplify efficiently with allele "A"
primers.
[0512] To distinguish between these possibilities, a
re-amplification experiment was performed wherein a first round of
PCR amplification was performed using a SMAD7 "T" allele primer
with SMAD7 "T" allele target DNA or with SMAD7 "C" allele target
DNA. The reaction products were diluted 10.sup.8 fold and
re-amplification was performed using the "T" vs. "C" allele primers
to determine if the identity of the SNP base present in the
reaction products changed during the first round of amplification.
The SMAD7 rs4939827 allele system was employed using the following
primers and target DNAs, which are shown below in Table 53.
TABLE-US-00168 TABLE 53 Name Sequence SEQ ID No. rs4939827
5'-CTCACTCTAAACCCC SEQ ID No. 236 Rev AGCATT rs4939827
5'-CAGCCTCATCCAAAA SEQ ID No. 249 For GAGGAAA rs4939827
5'-CAGCCTCATCCAAAA SEQ ID No. 250 C-For WT GAGGAAAcAGGA-SpC3
rs4939827 5'-CAGCCTCATCCAAAA SEQ ID No. 254 T-For WT
GAGGAAAuAGGA-SpC3 DNA bases are shown in uppercase. RNA bases are
shown in lowercase. SpC3 is a spacer C3 used as a 3'-blocking
group.
[0513] The above primers target the following 85 bp region of the
SMAD7 gene (NM 005904). Synthetic oligonucleotides were synthesized
for use as pure targets in the SMAD7 system and are shown below.
Primer binding sites are underlined and the SNP location is
highlighted as bold italic.
rs4939827 (SMAD7) C allele (SEQ ID No. 239)
TABLE-US-00169 CAGCCTCATCCAAAAGAGGAAA AGGACCCCAGAGC
TCCCTCAGACTCCTCAGGAAACACAGACAATGCTGG GGTTTAGAGTGAG
rs4939827 (SMAD7) T allele (SEQ ID No. 240)
TABLE-US-00170 CAGCCTCATCCAAAAGAGGAAA AGGACCCCAGAGC
TCCCTCAGACTCCTCAGGAAACACAGACAATGCTGG GGTTTAGAGTGAG
[0514] PCR reactions were performed in 10 .mu.l volume using 200 nM
of the unmodified Rev primer (SEQ ID No. 236) and the "T" allele
cleavable blocked For primer (SEQ ID No. 254) in Bio-Rad SYBR Green
master mix. Reactions were run with 2.6 mU of Pyrococcus abyssi
RNase H2 on a Roche Lightcycler.RTM. 480 platform with
6.6.times.10.sup.5 copies of synthetic oligonucleotide target SMAD7
"C" allele (SEQ ID No. 239) or SMAD7 "T" allele (SEQ ID No. 249).
Reactions were started with a soak at 95.degree. C. for 5 minutes
followed by 80 cycles of [95.degree. C. for 10 seconds, 60.degree.
C. for 30 seconds, and 72.degree. C. for 1 second]. Results of qPCR
amplifications done at this SNP site are shown in Table 54
below.
TABLE-US-00171 TABLE 54 Cp and .DELTA.Cp values of qPCR reactions
showing mismatch discrimination of cleavable-blocked primers at a
SMAD7 C/T allele. Cp for: Cp for: "T" Target "C" Target SEQ ID No.
240 SEQ ID No. 239 .DELTA.Cp Primer 32.5 18.9 13.6 rs4939827 T-For
WT SEQ ID No. 254
[0515] The "T" allele primer performed similar to pervious results
showing a .DELTA.Cp of 13.6 between reactions run using the match
"T" allele target DNA and the mismatch "C" allele target DNA.
[0516] This experiment was repeated using a 10.sup.8 dilution of
the reaction products from the above PCR amplifications as target
DNA. If cleavage at the mismatch site occurred at the expected
position 5'- to the ribonucleotide, then the reaction products
should remain "true" and "T" allele product would be made from
input "T" allele template and "C" allele product would be made for
input "C" allele template. However, if any appreciable cleavage
occurred 3'- to the ribonucleotide, then the reaction products
should be converted to the sequence of the primer at the SNP site.
In this case, a "T" allele product would be made from a "C" allele
target.
[0517] PCR reactions were performed in 10 .mu.l volume using 200 nM
of the unmodified Rev primer (SEQ ID No. 236) and the "T" allele
cleavable blocked For primer (SEQ ID No. 254) or the "C" allele
cleavable blocked For primer (SEQ ID No. 250) in Bio-Rad SYBR Green
master mix. Reactions were run with 2.6 mU of Pyrococcus abyssi
RNase H2 on a Roche Lightcycler.RTM. 480 platform. Input target DNA
was a 10.sup.8 dilution of the reaction products shown in Table 54
above. Reactions were started with a soak at 95.degree. C. for 5
minutes followed by 45 cycles of [95.degree. C. for 10 seconds,
60.degree. C. for 30 seconds, and 72.degree. C. for 1 second].
Results of qPCR amplifications done at this SNP site are shown in
Table 55 below.
TABLE-US-00172 TABLE 55 Cp and .DELTA.Cp values of qPCR reactions
showing mismatch discrimination of cleavable-blocked primers at a
SMAD7 C/T allele. Cp for: T Cp for: Target amplified "C" Target
amplified by "T" primer by "T" primer .DELTA.Cp Primer 27.7 29.2
1.5 rs4939827 T-For WT SEQ ID No. 254 Primer 38.6 38.5 0.1
rs4939827 C-For WT SEQ ID No. 250
[0518] The reactions products previously made (Table 54) using the
"T" allele primer with both the "T" allele target and the "C"
allele target now show nearly identical amplification efficiency
using the "T" allele primer whereas previously a .DELTA.Cp of 13.6
was observed between the two different starting target DNAs. This
is most consistent with the product nucleic acids having similar
sequence, i.e., both are now predominantly "T" allele. Consistent
with this hypothesis, both of these samples now show similar
delayed Cp using the "C" allele primer. Thus it appears that the
product from the "T" allele primer amplification using the "C"
allele target was largely converted to "T" allele, consistent with
that product originating with a primer cleavage event occurring 3'-
to the ribonucleotide base. The reaction products from the original
amplification using the "T" allele primer (Table 54) were subcloned
and DNA sequence determined. All clones identified had the "T"
allele present, whether the starting template was the "T" allele or
the "C" allele, adding further support to this conclusion.
Example 25
Use of Phosphorothioate Modified Internucleotide Linkages in
Cleavable Blocked Primers to Improve Mismatch Discrimination
[0519] The results form Example 24 indicates that PCR performed
with a mismatched primer/target combination can produce a product
with sequence matching the primer instead of the target. The most
likely scenario that would result in this kind of product starts
with cleavage of the mismatched primer at a position 3'- to the
ribonucleotide residue. Use of chemical modifications that prevent
unwanted cleavage in this domain of the primer may improve
performance of the cleavable-blocked primers especially in SNP
discrimination. The following primers, as shown in below in Table
56, were synthesized with nuclease-resistant phosphorothioate (PS)
modified internucleotide linkages placed at positions 3'- to the
ribonucleotide as indicated. It was established in Example 15 that
placement of a PS bond at the 3'-linkage directly at the RNA base
can decrease cleavage efficiency. This modification survey
therefore focused on the DNA linkages further 3'- to this site. The
synthetic amplicon system previously used in Example 13 was
employed.
TABLE-US-00173 TABLE 56 Name Sequence SEQ ID No. Syn-For
5'-AGCTCTGCCCAAAGAT SEQ ID No. 86 TACCCTG Syn-Rev-rU-C3
5'-CTGAGCTTCATGCCTT SEQ ID No. 134 TACTGTuCCCC-SpC3 Syn-Rev-rU-
5'-CTGAGCTTCATGCCTT SEQ ID No. 320 C*CCC-C3 TACTGTuC*CCC-SpC3
Syn-Rev-rU- 5'-CTGAGCTTCATGCCTT SEQ ID No. 258 CC*CC-C3
TACTGTuCC*CC-SpC3 Syn-Rev-rU- 5'-CTGAGCTTCATGCCTT SEQ ID No. 259
CCC*C-C3 TACTGTuCCC*C-SpC3 Syn-Rev-rU- 5'-CTGAGCTTCATGCCTT SEQ ID
No. 260 C*C*C*C-C3 TACTGTuC*C*C*C-SpC3 DNA bases are shown in
uppercase. RNA bases are shown in lowercase. "*" indicates a
phosphorothioate (PS) modified internucleotide linkage.
[0520] The standard configuration blocked RNase H2 cleavable primer
(SEQ ID No. 134) was compared with PS-modified versions of this
sequence (SEQ ID Nos. 320 and 258-260). This set of "Rev" primers
were used with the unmodified "For" primer (SEQ ID No. 86) together
with two different synthetic oligonucleotide templates (originally
used in defining mismatch discrimination potential in Example 13).
These templates provide a perfect match control (Template SEQ ID
No. 162) and a T/U mismatch (Template SEQ ID No. 155). The two
templates and oligonucleotides are shown below with the cleavable
blocked primer (SEQ ID No. 134) aligned beneath to illustrate the
regions of match vs. mismatch.
Synthetic template, SEQ ID No. 162 (A:U match):
TABLE-US-00174 AGCTCTGCCCAAAGATTACCCTGACAGCTAAGTGG
CAGTGGAAGTTGGCCTCAGAAGTAGTGGCCAGCT
GTGTGTCGGGGAACAGTAAAGGCATGAAGCTCAG-3' |||||||||||||||||||||||||||
3'-C3-CCCCuTGTCATTTCCGTACTTCGAGTC-5'
Synthetic template, SEQ ID No. 155 (T:U mismatch):
TABLE-US-00175 AGCTCTGCCCAAAGATTACCCTGACAGCTAAGTGG
CAGTGGAAGTTGGCCTCAGAAGTAGTGGCCAGCT
GTGTGTCGGGGTACAGTAAAGGCATGAAGCTCAG-3' |||| ||||||||||||||||||||||
3'-C3-CCCCuTGTCATTTCCGTACTTCGAGTC-5'
[0521] PCR reactions were performed in 10 .mu.l volume using 200 nM
of the unmodified For primer (SEQ ID No. 86) and the different
cleavable blocked Rev primers shown above (SEQ ID Nos. 134, 320 and
258-260) in Bio-Rad SYBR Green master mix. Reactions were run with
1.3 mU of Pyrococcus abyssi RNase H2 on a Roche Lightcycler.RTM.
480 platform. Input target DNA was 2.times.10.sup.6 copies of the
synthetic target sequences shown above (SEQ ID Nos. 155 and 162).
Reactions were started with an incubation at 95.degree. C. for 5
minutes followed by 45 cycles of [95.degree. C. for 10 seconds,
60.degree. C. for 30 seconds, and 72.degree. C. for 1 second].
Results of qPCR amplifications done at this SNP site are shown in
Table 57 below.
TABLE-US-00176 TABLE 57 Cp values of qPCR reactions comparing
mismatch discrimination of blocked primers with or without PS
linkages 3'- to the cleavable ribonucleotide. Match Target Mismatch
Target (A:U) (T:U) SEQ ID No. 162 SEQ ID No. 155 .DELTA.Cp SEQ ID
No. 134 18.5 26.2 7.7 CCCC Primer SEQ ID No. 320 19.5 31.9 12.4
C*CCC Primer SEQ ID No. 258 18.2 26.7 8.5 CC*CC Primer SEQ ID No.
259 18.4 26.3 7.9 CCC*C Primer SEQ ID No. 260 18.5 29.1 10.6
C*C*C*C Primer (.DELTA.Cp = Cp mismatch - Cp match
[0522] Placement of a PS modified linkage at the 3' "+1" position
(rUC*CCC) led to almost a 5 cycle improvement in SNP discrimination
in this assay (SEQ ID No. 260 vs. 134), demonstrating that
increasing nuclease stability in the domain 3'- to the
ribonucleotide can significantly improve assay performance.
Modification o the linkages further 3' from the ribonucleotide had
minimal impact. Modification of all of the linkages in this area
(rUC*C*C*C, nucleotides 23-27 of SEQ ID No. 260) also showed
benefit, improving relative SNP discrimination by 3 cycles, but
unexpectedly showed less benefit than using just a single
modification at the 3'+1 linkage. This may relate to the lowered
binding affinity Tm that also results from the PS modification.
[0523] Thus, adding nuclease resistant modifications at the
linkages 3'- to the cleavable ribonucleotide can increase SNP
discrimination for the RNase H2 mediated cleavable-blocked primer
PCR assay. Typically, only one of the two stereoisomers at a PS
linkage (the Rp or Sp isomer) confers benefit. Improved activity
might therefore be realized by isolation a chirally pure PS
compound here, as was demonstrated in Example 15. Other nuclease
resistant modifications may be suitable in this area, such as the
non-chiral phosphorodithioate linkage, the methyl phosphonate
linkage, the phosphoramidate linkage, a boranophosphate linkage,
and abasic residues such as a C3 spacer to name a few.
Example 26
Use of Cleavable Primers Having an Unblocked 3'-Hydroxyl in a qPCR
Assay
[0524] In the above Examples, a blocking group was placed at the
3'-end of the primer to prevent primer extension from occurring
prior to RNase H2 cleavage. For certain primer designs and
applications, it may not be necessary or even desirable to employ a
3'-blocking group. We have previously described a method of nucleic
acid amplification termed polynomial amplification that employs
primers that are chemically modified in ways that block template
function while retaining primer function. A variety of groups can
be used for this purpose, including internal C3 spacers and
internal 2'OMe RNA bases. Using nested primers, high specificity is
achieved and amplification power is dependent upon the number of
nested primers employed, with amplification occurring according to
a polynomial expansion instead of the exponential amplification
seen in PCR (see U.S. Pat. No. 7,112,406 and pending US Patent
applications 2005/0255486 and 2008/0038724). Combining elements of
the polynomial amplification primers with an RNase H2 cleavable
domain of the present invention results in a novel primer design
that has an unblocked 3'-hydroxyl and is capable of supporting
primer extension yet cannot support PCR. Upon cleavage, the
template blocking groups are removed and primer function for use in
PCR is restored. The present example demonstrates use of cleavable
template-blocked primers having a 3'-hydroxyl in qPCR.
[0525] The following primers, as shown below in Table 58, were
synthesized for use with the artificial synthetic amplicon used in
previous Examples.
TABLE-US-00177 TABLE 58 SEQ ID Name Sequence No. Syn-For
5'-AGCTCTGCCCAAAGATTACCCTG SEQ ID No. 86 Syn-Rev
5'-CTGAGCTTCATGCCTTTACTGT SEQ ID No. 87 Syn-For-rA-
5'-AGCTCTGCCCAAAGATTACCCTG SEQ ID C3 aCAGC-SpC3 No. 261 Syn-For-rA-
5'-AGCTCTGCCCAAAGATTACCCTG SEQ ID iC3-D1 aCAGC(SpC3-SpC3)A No. 262
Syn-For-rA- 5'-AGCTCTGCCCAAAGATTACCCTG SEQ ID iC3-D2
aCAGC(SpC3-SpC3)AG No. 263 Syn-For-rA- 5'-AGCTCTGCCCAAAGATTACCCTG
SEQ ID iC3-D4 aCAGC(SpC3-SpC3)AGTG No. 264 Syn-For-rA-
5'-AGCTCTGCCCAAAGATTACCCTG SEQ ID iC3-D5 aCAGC(SpC3-SpC3)AGTGG No.
265 DNA bases are shown in uppercase. RNA bases are shown in
lowercase. SpC3 is a Spacer C3 group, positioned either internal
within the primer or at the 3'-end.
[0526] The synthetic amplicon oligonucleotide template (SEQ ID No.
162) is shown below with the unmodified and various modified
cleavable For primers shown aligned above and the unmodified Rev
primer aligned below. DNA bases are uppercase, RNA bases are
lowercase, and "x" indicates a Spacer-C3 group.
TABLE-US-00178 5'AGCTCTGCCCAAAGATTACCCTG SEQ ID No. 86 unmodified
5'AGCTCTGCCCAAAGATTACCCTGaCAGC-x SEQ ID No. 261 3'-block
5'AGCTCTGCCCAAAGATTACCCTGaCAGCxxA SEQ ID No. 262 Int D1
5'AGCTCTGCCCAAAGATTACCCTGaCAGCxxAG SEQ ID No. 263 Int D2
5'AGCTCTGCCCAAAGATTACCCTGaCAGCxxAGTG SEQ ID No. 264 Int D4
5'AGCTCTGCCCAAAGATTACCCTGaCAGCxxAGTGG SEQ ID No. 265 Int D5
|||||||||||||||||||||||||||| |||||
5'AGCTCTGCCCAAAGATTACCCTGACAGCTAAGTGGCAGTGGAAGTTGGCCTCAGAAGTAGTGGCCAG
CTGTGTGTCGGGGAACAGTAAAGGCATGAAGCTCAG-3' ||||||||||||||||||||||
TGTCATTTCCGTACTTCGAGTC-5' SEQ ID No. 87
[0527] PCR reactions were performed in 10 .mu.l volume using 200 nM
of the individual For primers (SEQ ID Nos. 86, 261-65) and the
unmodified Rev primer (SEQ ID No. 87) in Bio-Rad SYBR Green master
mix. Reactions were run with or without 1.3 mU of Pyrococcus abyssi
RNase H2 on a Roche Lightcycler.RTM. 480 platform. Input target DNA
was 2.times.10.sup.6 copies of the synthetic target shown above
(SEQ ID No. 162). Reactions were started with an incubation at
95.degree. C. for 5 minutes followed by 60 cycles of [95.degree. C.
for 10 seconds, 60.degree. C. for 30 seconds, and 72.degree. C. for
1 second]. Results of qPCR amplifications are shown in Table 59
below.
TABLE-US-00179 TABLE 59 Cp values of qPCR reactions comparing
performance of cleavable primers having a 3'-blocking group vs.
cleavable primers having internal template blocking groups. Without
+1.3 mU RNase H2 RNase H2 Unblocked For SEQ ID No. 86 17.0 17.2
Unblocked Rev SEQ ID No. 87 3'-blocked For SEQ ID No. 261 >60
17.1 Unblocked Rev SEQ ID No. 87 Int-blocked For SEQ ID No. 262
(D1) >60 17.1 Unblocked Rev SEQ ID No. 87 Int-blocked For SEQ ID
No. 263 (D2) >60 17.1 Unblocked Rev SEQ ID No. 87 Int-blocked
For SEQ ID No. 264 (D4) >60 17.1 Unblocked Rev SEQ ID No. 87
Int-blocked For SEQ ID No. 265 (D5) >60 17.9 Unblocked Rev SEQ
ID No. 87
[0528] The unblocked primers gave detectable signal at around cycle
17 in this assay system. Using the unblocked Rev primer with the
3'-blocked For primer, no signal was detected within the 60 cycle
PCR run without RNase H2, however with RNase H2 a similar cycle
detection time of around 17 was seen. The internally blocked For
primers that had a free 3'-hydroxyl group behaved identically to
the 3'-modified primer. In spite of the unblocked 3'-hydroxyl,
primer cleavage with RNase H2 was required for function in PCR,
presumably due to the loss of template function imposed by the
internal C3 spacers. C3 spacers placed near the 3'-end may also
inhibit primer extension to a certain degree. No signal was
detected in the absence of RNase H2; with RNase H2, cleavage and
amplification proceeded normally.
[0529] This example demonstrates that cleavable primers do not need
to be modified at the 3'-terminal residue to function in a
cleavable-primer PCR assay and that primers having internal
modifications that disrupt template function can perform equally
well. Given the significance of this finding to primer design, a
similar experiment was performed using an endogenous human gene
target using human genomic DNA to ensure that these results could
be generalized.
[0530] The following primers, as shown below in Table 60, were
synthesized based upon the human SMAD7 gene used in previous
Examples, using only the "C" allele.
TABLE-US-00180 TABLE 60 Name Sequence SEQ ID No. rs4939827
5'-CTCACTCTAAACCCC SEQ ID No. 236 Rev AGCATT rs4939827
5'-CAGCCTCATCCAAAA SEQ ID No. 249 For GAGGAAA rs4939827
5'-CAGCCTCATCCAAAA SEQ ID No. 250 C-For-C3 GAGGAAAcAGGA-SpC3
rs4939827 5'-CAGCCTCATCCAAAA SEQ ID No. 266 C-For-iC3-D1
GAGGAAAcAGGA(SpC3- SpC3)C rs4939827 5'-CAGCCTCATCCAAAA SEQ ID No.
267 C-For-iC3-D2 GAGGAAAcAGGA(SpC3- SpC3)CC rs4939827
5'-CAGCCTCATCCAAAA SEQ ID No. 268 C-For-iC3-D4 GAGGAAAcAGGA(SpC3-
SpC3)CCAG rs4939827 5'-CAGCCTCATCCAAAA SEQ ID No. 269 C-For-iC3-D5
GAGGAAAcAGGA(SpC3- SpC3)CCAGA DNA bases are shown in uppercase. RNA
bases are shown in lowercase. SpC3 is a Spacer C3 group, positioned
either internal within the primer (template block) or at the 3'-end
(primer block).
[0531] The SMAD7 amplicon sequence (SEQ ID No. 239) is shown below
with the unmodified and various modified cleavable For primers
shown aligned above. DNA bases are uppercase, RNA bases are
lowercase, and "x" indicates a Spacer-C3 group.
TABLE-US-00181 5'CAGCCTCATCCAAAAGAGGAAA SEQ ID No. 249 unmodified
5'CAGCCTCATCCAAAAGAGGAAAcAGGA-x SEQ ID No. 250 3'-block
5'CAGCCTCATCCAAAAGAGGAAAcAGGAxxC SEQ ID No. 266 Int-block
5'CAGCCTCATCCAAAAGAGGAAAcAGGAxxCC SEQ ID No. 267 Int-block
5'CAGCCTCATCCAAAAGAGGAAAcAGGAxxCCAG SEQ ID No. 268 Int-block
5'CAGCCTCATCCAAAAGAGGAAAcAGGAxxCCAGA SEQ ID No. 269 Int-block
||||||||||||||||||||||||||| |||||
5'CAGCCTCATCCAAAAGAGGAAACAGGACCCCAGAGCTCCCTCAGACTCCTCAGGAAACACAGACAATGCT
GGGGTTTAGAGTGAG-3'
[0532] PCR reactions were performed in 10 .mu.l volume using 200 nM
of the individual For primers (SEQ ID Nos. 249-50, 266-69) and the
unmodified Rev primer (SEQ ID No. 236) in Bio-Rad SYBR Green master
mix. Reactions were run with or without 2.6 mU of Pyrococcus abyssi
RNase H2 on a Roche Lightcycler.RTM. 480 platform. Input target DNA
was 2 ng of genomic DNA from a human cell line (Coreill 18562,
SMAD7 "C" allele). Reactions were started with an incubation at
95.degree. C. for 5 minutes followed by 60 cycles of [95.degree. C.
for 10 seconds, 60.degree. C. for 30 seconds, and 72.degree. C. for
1 second]. Results of qPCR amplifications are shown in Table 61
below.
TABLE-US-00182 TABLE 61 Cp values of qPCR reactions comparing
performance of cleavable primers having a 3'-blocking group vs.
cleavable primers having internal template blocking groups. Without
+2.6 mU RNase H2 RNase H2 Unblocked For SEQ ID No. 249 25.8 25.5
Unblocked Rev SEQ ID No. 236 3'-blocked For SEQ ID No. 250 >60
26.3 Unblocked Rev SEQ ID No. 236 Int-blocked For SEQ ID No. 266
(D1) >60 26.3 Unblocked Rev SEQ ID No. 236 Int-blocked For SEQ
ID No. 267 (D2) >60 26.2 Unblocked Rev SEQ ID No. 236
Int-blocked For SEQ ID No. 268 (D4) >60 26.2 Unblocked Rev SEQ
ID No. 236 Int-blocked For SEQ ID No. 269 (D5) >60 26.7
Unblocked Rev SEQ ID No. 236
[0533] The unblocked primers gave detectable signal around cycle 26
in this assay system using human genomic DNA. Using the unblocked
Rev primer with the 3'-blocked For primer, no signal was detected
within the 60 cycle PCR run without RNase H2, however with RNase H2
a similar cycle detection time of around 26 was seen. All of the
internally blocked For primers that had a free 3'-hydroxyl group
behaved identically to the 3'-modified primer. No signal was
detected in the absence of RNase H2; with RNase H2, cleavage and
amplification proceeded normally with detection occurring around 26
cycles.
[0534] This example further demonstrates that cleavable primers do
not need to be modified at the 3'-end to function in the cleavable
primer qPCR assay. Primers having internal modifications that
disrupt template function still require a primer cleavage event to
function as primers in the assay. When cleavage is done using RNase
H2 at an internal cleavable residue, like a single RNA base,
amplification efficiency is identical to that seen using unmodified
primers. This novel version of template-blocked cleavable primers
can be employed to perform PCR in complex nucleic acid samples like
human genomic DNA.
Example 27
Cleavable Primers with Internal Template Blocking Groups and a
3'-Hydroxyl can Prime DNA Synthesis
[0535] The cleavable template-blocked primers disclosed in Example
26 have an unblocked 3'-hydroxyl group that should permit the
oligonucleotides to function as primers in linear primer extension
reactions but the internal template-blocking groups prevent
function in PCR as most of the primer cannot be replicated.
Consequently, no primer binding site exists in the daughter
products. Cleavage of the primer by RNase H2 removes the domain
containing the template-blocking groups and restores normal primer
function. The present example demonstrates that these compositions
can function to prime DNA synthesis.
[0536] The following primers shown below in Table 62 were employed
to perform linear primer extension reactions using the artificial
synthetic amplicon system used in previous Examples.
TABLE-US-00183 TABLE 62 SEQ ID Name Sequence No. Syn-For
5'-AGCTCTGCCCAAAGA SEQ ID TTACCCTG No. 86 Syn-For-
5'-AGCTCTGCCCAAAGA SEQ ID rA-C3 TTACCCTGaCAGC-SpC3 No. 261 Syn-For-
5'-AGCTCTGCCCAAAGA SEQ ID rA-iC3-D1 TTACCCTGaCAGC No. 262
(SpC3-SpC3)A Syn-For- 5'-AGCTCTGCCCAAAGA SEQ ID rA-iC3-D2
TTACCCTGaCAGC No. 263 (SpC3-SpC3)AG Syn-For- 5'-AGCTCTGCCCAAAGA SEQ
ID rA-iC3-D4 TTACCCTGaCAGC No. 264 (SpC3-SpC3)AGTG Syn-For-
5'-AGCTCTGCCCAAAGA SEQ ID rA-iC3-D5 TTACCCTGaCAGC No. 265
(SpC3-SpC3)AGTGG DNA bases are shown in uppercase. RNA bases are
shown in lowercase. SpC3 is a Spacer C3 group, positioned either
internal within the primer or at the 3'-end.
[0537] A newly synthesized 103 mer oligonucleotide template was
made which was complementary to the Syn-For primers above (SEQ ID
No. 270), which is shown below with the unmodified and various
modified cleavable For primers aligned above. DNA bases are
uppercase, RNA bases are lowercase, and "x" indicates a Spacer-C3
group.
TABLE-US-00184 5'AGCTCTGCCCAAAGATTACCCTG SEQ ID No. 86 unmodified
5'AGCTCTGCCCAAAGATTACCCTGaCAGC-x SEQ ID No. 261 3'-block
5'AGCTCTGCCCAAAGATTACCCTGaCAGCxxA SEQ ID No. 262 Int D1
5'AGCTCTGCCCAAAGATTACCCTGaCAGCxxAG SEQ ID No. 263 Int D2
5'AGCTCTGCCCAAAGATTACCCTGaCAGCxxAGTG SEQ ID No. 264 Int D4
5'AGCTCTGCCCAAAGATTACCCTGaCAGCxxAGTGG SEQ ID No. 265 Int D5
|||||||||||||||||||||||||||| |||||
3'TCGAGACGGGTTTCTAATGGGACTGTCGATTCACCGTCACCTTCAACCGGAGTCTTCATCACCGGTCGACAC
ACAGCCCCTTGTCATTTCCGTACTTCGAGTC-5'
[0538] The six For primers shown above were radiolabeled with
.sup.32P as described above. Primer extension reactions were
performed in a 20 .mu.L volume using 0.8 U iTaq polymerase
(Bio-Rad), 800 .mu.M dNTPs, 3 mM MgCl.sub.2, in 1.times. iTaq
buffer (20 mM Tris pH 8.4, 50 mM KCl) and 2 nM primer and template
(40 fmole of each oligonucleotide in the 20 .mu.L reaction).
Reactions were started with an incubation at 95.degree. C. for 5
minutes followed by 35 cycles of [95.degree. C. for 10 seconds,
60.degree. C. for 30 seconds, and 72.degree. C. for 1 second] on an
MJ Research PTC-100 thermal cycler. Reactions were stopped with the
addition of cold EDTA containing formamide gel loading buffer.
Reaction products were separated using denaturing 7M urea, 15%
polyacrylamide gel electrophoresis (PAGE) and visualized using a
Packard Cyclone.TM. Storage Phosphor System (phosphorimager). The
relative intensity of each band was quantified as above and results
plotted as a fraction of total radioactive material present in the
band representing the primer extension product. Results are shown
in FIG. 34.
[0539] Under these reaction conditions, 61% of the control
unblocked primer (SEQ ID No. 86) was converted into a higher
molecular weight primer extension product. As expected, the 3'-end
blocked cleavable primer (SEQ ID No. 261) did not show any primer
extension product. Similarly, the D1 and D2 cleavable primers with
internal C3 groups and a 3'-hydroxyl (SEQ ID Nos. 262-3) also did
not support primer extension. The cleavable primers having a
slightly longer terminal DNA domains (the D4 and D5 sequences, SEQ
ID Nos. 264-5) did support primer extension with the D4 showing 47%
conversion and the D5 showing 60% conversion of the primer into an
extension product, a reaction efficiency identical to the
unmodified control primer. Thus when internal C3 spacers are placed
very near the 3'-end both priming and template function are
disrupted. When placed more than 4 residues from the 3'-end only
template function is blocked.
Example 28
Use of Cleavable Primers with Internal Template Blocking Groups and
a 3'-Hydroxyl to Improve Mismatch Discrimination
[0540] Example 24 demonstrated that cleavage of an RNA-containing
primer on the 3'-side of the RNA base by RNase H2 is an undesired
event that can contribute to late arising false positive signals in
a qPCR SNP discrimination assay. Example 25 demonstrated that
modifications which confer nuclease resistance to this domain can
improve SNP discrimination. The novel compositions described in
examples 26 and 27 place internal C3 groups on the 3'-side of the
cleavable ribonucleotide which disrupts template function of the
primer in a domain that is removed by RNase H2 cleavage. This
example demonstrates that positioning the C3 spacer groups close to
the RNA base improves performance of the cleavable primer in SNP
discrimination using a format that leaves the probe "unblocked",
having an unmodified 3'-hydroxyl.
[0541] The following primers, as shown below in Table 63, were
synthesized for the human SMAD7 gene similar to previous Examples.
Primers were made specific for the "C" allele and were tested on
both "C" allele and "T" allele genomic DNA targets.
TABLE-US-00185 TABLE 63 Name Sequence SEQ ID No. rs4939827
5'-CTCACTCTAAACCCCAG SEQ ID No. Rev CATT 236 rs4939827
5'-CAGCCTCATCCAAAAGA SEQ ID No. For GGAAA 249 rs4939827
5'-CAGCCTCATCCAAAAGA SEQ ID No. C-For-C3 GGAAAcAGGA-SpC3 250
rs4939827 5'-CAGCCTCATCCAAAAGA SEQ ID No. C-For-A(C3C3)A
GGAAAcA(SpC3-SpC3)A 271 DNA bases are shown in uppercase. RNA bases
are shown in lowercase. SpC3 is a Spacer C3 group, positioned
either internal within the primer or at the 3'-end.
[0542] The SMAD7 amplicon sequence (SEQ ID No. 239, "C" target) is
shown below with the unmodified and two modified cleavable For
primers aligned above it. DNA bases are uppercase, RNA bases are
lowercase, and "x" indicates a Spacer-C3 group. The site of the
rs4939827 C/T SNP is indicated with bold underline.
TABLE-US-00186 5'CAGCCTCATCCAAAAGAGGAAA SEQ ID No. 249 unmodified
5'CAGCCTCATCCAAAAGAGGAAAcAGGA-x SEQ ID No. 250 3'-block
5'CAGCCTCATCCAAAAGAGGAAAcAxxA SEQ ID No. 271 Int-block
|||||||||||||||||||||||| |
5'CAGCCTCATCCAAAAGAGGAAACAGGACCCCAGAGCTCCCTCAGACTCCTCAGGAAACAC
AGACAATGCTGGGGTTTAGAGTGAG-3'
[0543] The same primers are aligned with the mismatch SMAD7
amplicon sequence (SEQ ID No. 240, "T" target).
TABLE-US-00187 5'CAGCCTCATCCAAAAGAGGAAA SEQ ID No. 249 unmodified
5'CAGCCTCATCCAAAAGAGGAAAcAGGA-x SEQ ID No. 250 3'-block
5'CAGCCTCATCCAAAAGAGGAAAcAxxA SEQ ID No. 271 Int-block
|||||||||||||||||||||| | |
5'CAGCCTCATCCAAAAGAGGAAATAGGACCCCAGAGCTCCCTCAGACTCCTCAGGAAACAC
AGACAATGCTGGGGTTTAGAGTGAG-3'
[0544] PCR reactions were performed in 10 .mu.l volume using 200 nM
of the individual For primers (SEQ ID Nos. 249-50, 266-69) and the
unmodified Rev primer (SEQ ID No. 236) in Bio-Rad SYBR Green master
mix. Reactions were run with or without 2.6 mU ofPyrococcus abyssi
RNase H2 on a Roche Lightcycler.RTM. 480 platform. Input target DNA
was 2 ng of genomic DNA from human cell lines homozygous for the
SMAD7 "C" and "T" alleles (Coreill 18562 and 18537). Reactions were
started with an incubation at 95.degree. C. for 5 minutes followed
by 75 cycles of [95.degree. C. for 10 seconds, 60.degree. C. for 30
seconds, and 72.degree. C. for 1 second]. Results of qPCR
amplifications are shown in Table 64 below.
TABLE-US-00188 TABLE 64 Cp values of qPCR reactions comparing
performance of cleavable primers having a 3'-blocking group vs.
cleavable primers having internal template blocking groups in a SNP
discrimination assay. Unmodified Control 3'-C3 Blocked Int-C3,
3'-unblocked SEQ ID No. 249 SEQ ID No. 250 SEQ ID No. 271 "C" "T"
"C" "T" "C" "T" Allele Allele .DELTA.Cp Allele Allele .DELTA.Cp
Allele Allele .DELTA.Cp No 27.5 25.9 -- >75 >75 -- >75
>75 -- RNaseH 2.6 mU 27.3 26.1 -- 27.8 37.8 10.0 41.8 68.4 26.6
RNaseH 10 mU 27.1 25.8 -- 27.0 40.0 13.0 29.9 53.7 23.8 RNaseH 50
mU 27.1 26.0 -- 27.0 28.5 1.5 27.6 53.3 25.7 RNaseH 200 mU 27.1
25.8 -- 27.0 26.1 -- 27.5 41.4 13.9 RNaseH (.DELTA.Cp = Cp mismatch
"T"- Cp match "C")
[0545] The unmodified primers are designed to be non-discriminatory
and amplified both alleles with similar efficiency, producing a
detectable signal at around 26-27 cycles. Both cleavable primers
were dependent upon RNase H2 for function and did not produce any
detectable signal for either allele in the absence of cleaving
enzyme. Using low amounts of RNase H2 (2.6-10 mU), the 3'-blocked
cleavable primer (SEQ ID No. 250) produced detectable signal around
cycle 27 for the match "C" allele and showed a delayed Cp of 38-40
cycles for the mismatch "T" allele (.DELTA.Cp of 10-13). Using
higher amounts of RNase H2, specificity was lost and both alleles
amplified with similar efficiency. The cleavable primer having two
C3 spacers 3'- to the ribonucleotide (SEQ ID No. 271) required
higher levels of RNase H2 for efficient cleavage/priming and showed
delayed Cp values even for the perfect match "C" allele using 2.6
and 10 mU of enzyme. It is not surprising that modifications of
this kind near the RNA cleavable site require higher amounts of
enzyme. Example 22 demonstrated that placing a 2'OMe modification
adjacent to the ribonucleotide required 100 mU of RNase H2 to
achieve full activity. Using higher amounts of enzyme resulted in
efficient cleavage and positive signal was detected at .about.27
cycles using 50 or 200 mU of RNase H2. Importantly, SNP
discrimination was markedly improved using this primer design, with
the .DELTA.Cp for the "T" allele being around 25 cycles using RNase
H2 in the concentration range of 2.6-50 mU. Mismatch discrimination
decreased when using 200 mU of the enzyme; however, SNP
discrimination was still almost at a 14 cycle .DELTA.Cp. Optimal
enzyme concentration was 50 mU, at which point priming efficiency
was similar to unmodified primers and SNP discrimination showed a
25.7 cycle .DELTA.Cp.
[0546] Therefore the present cleavable primer design with two
internal C3 spacer groups near the ribonucleotide and an unblocked
3'-hydroxyl, "RDxxD", showed significantly improved mismatch
discrimination over the original primer design, "RDDDD-x" (where
R=RNA base, D=DNA base, and x=C3 spacer). Related designs, such as
"RDDxxD" or "RDxxDD", may show similarly improved function and
small optimizations in design may be beneficial depending upon the
precise sequence context of the SNP of interest. Utilizing chemical
modifying groups like the C3 spacer that disrupt template function
but leave the 3'-hydroxyl unmodified can enhance the specificity of
cleavage at the ribonucleotide by RNase H2 and improve SNP
discrimination.
Example 29
Use of RNase H2-Cleavable Ligation Probes in DNA Sequencing
Methods
[0547] The previous Examples described the use of RNase H2
cleavable oligonucleotide compositions for applications as primers
where the cleavable oligonucleotide primes a DNA synthesis
reaction. Applications disclosed in the above Examples include both
end-point and real time PCR in several different detection formats.
Example 8 showed use of cleavable primers in a DNA sequencing
application using the Sanger sequencing method with DNA polymerase
and dideoxynucleotide terminators; in this case the RNase
H2-cleavable oligonucleotide also functioned as a primer. RNase
H2-cleavable oligonucleotides can also be used in ligation format
assays as well. One such application is DNA sequencing using
cleavable ligation probes. The current Example demonstrates use of
RNase H2-cleavable ligation probes in a format suitable for use in
DNA sequencing.
[0548] The use of ligation probes to sequentially interrogate the
identity of bases in an unknown nucleic acid sequence (i.e., DNA
sequencing) has been described (see U.S. Pat. No. 5,750,341 and
U.S. Pat. No. 6,306,597 and US application 2008/0003571). The basic
scheme for sequencing in the 5' to 3' direction by ligation begins
with a nucleic acid acceptor molecule hybridized to an unknown
nucleic acid sequence. A series of base interrogation probes are
hybridized to this sequence which have a known fixed DNA base at
the 5'-end followed by random bases or universal bases to permit
stable nucleic acid hybridization of the probe to the target
nucleic acid of unknown sequence. Hybridization and subsequent
ligation reactions are dependent upon perfect or near perfect match
between the ligation probe and the target; perfect match is
required at the site of ligation. Ligation leads to a detectable
event which permits identification of the specific base present at
the ligation site. An RNase H2 cleavable site is contained within
the ligation probe. Following ligation the probe is cleaved by
RNase H2, releasing the bulk of the probe but leaving the newly
identified base ligated to the acceptor nucleic acid sequence,
which has now been elongated by one residue as a result of the
cycle of ligation and cleavage. This series of enzymatic and
chemical events is repeated through multiple cycles of ligation,
base identification, and cleavage and the unknown nucleic acid
sequence is thereby determined.
[0549] While the patent references cited above teach methods for
sequencing by ligation, the methods suggested therein to achieve
cleavage and release of the ligation probe permitting multiple
cycles of ligation/detection are inefficient and difficult to
perform. RNase H2-cleavable oligonucleotides using the methods of
the present invention offer an improvement over pre-existing method
and permit construction of less costly, easier to use cleavable
ligation probes for DNA sequencing. One scheme for DNA sequencing
using RNase H2 cleavable ligation probes is shown in FIG. 35.
[0550] The RNase H2 cleavable ligation probes in this method
contain a fixed known DNA base (or bases) at the 5'-end. The fixed
known base(s) can be the single 5'-most base or can include 2 or 3
or more bases towards the 5'-end. The present Example employs a
system wherein only the single DNA base at the 5'-end of the probe
is fixed. The synthetic oligonucleotide has a 5'-phosphate to
permit enzymatic ligation using a DNA ligase. An activated
adenylated form of the probe can also be used. As mentioned, the
first base at the 5'-end is fixed (known). Thus four independent
probes are needed to perform DNA sequencing, an "A" probe, a "C"
probe, a "T" probe, and a "G" probe. Obviously more probes will be
needed if the number of fixed bases are greater than one (for
example, 16 ligation probes will be needed if the first 2 bases are
used as fixed known sequence, one for each possible dinucleotide
pair). The first base following the fixed known DNA residue (in
this case, the second base from the 5'-end) is a residue which is
cleavable by RNase H2. In the present Example, an RNA base is
employed, however a 2'-F residue or other cleavable modified base
(such as are described in previous Examples) can also be used. The
remaining bases in the probe will be random bases (heterogeneous
mixes of the 4 DNA bases) and/or universal bases (such as inosine,
5-nitroindole, or other such groups as are well known to those with
skill in the art). Total length of the probe will usually be around
8-9 bases, however longer or shorter probe length is possible
depending on the particular ligase enzyme employed. When using T4
DNA ligase, a length of 8 is sufficient to achieve efficient
hybridization and enzymatic ligation. Longer probes can also be
used.
[0551] Complexity of the probe population increases according to
4.sup.N, where N=the number of random bases employed. For example,
the probe "pTn" has a fixed "T" base at the 5'-end, a single "n"
RNA base, and 6 "N" DNA bases, totaling 7 random residues
(p=phosphate, n=RNA, N=DNA). This presents a complexity of 4.sup.7
molecules (16,384) in the population. The complexity of the probe
can be decreased by substituting universal base groups for random N
bases. This is particularly effective towards the 3'-end. For
example, using 3 inosine residues would convert the above probe to
"pTnNNNIII" (as before, with I=inosine). This probe has a
complexity of 4.sup.4 molecules (256). It will require a
significantly lower mass input of ligation probe to achieve 100%
ligation with a probe having a complexity of 256 than one having a
complexity of 16,384. Use of one or more universal bases is
generally preferred. Finally, the ligation probe has a dye molecule
at or near the 3'-end to provide a detectable signal that can be
resolved following ligation. The 3'-modifying group also serves to
block ligation at the 3'-end so that the ligation probe itself
cannot serve as an acceptor nucleic acid.
[0552] Use of RNase H2 cleavable ligation probes of this design in
DNA sequencing is shown schematically in FIG. 35. A universal
primer or acceptor nucleic acid is hybridized to the unknown
nucleic acid. Attachment of a universal adaptor sequence on the end
of the unknown sequence may be required to permit hybridization of
the acceptor molecule, and this strategy permits use of the same
acceptor nucleic acid for all reactions. The acceptor nucleic acid
must have a 3'-hydroxyl group available for ligation. The mix of
ligation probes is introduced into the reaction in molar excess
(>256 fold excess for the 8 mer inosine containing probe design
described above) and T4 DNA ligase is used to perform enzymatic
ligation. Free probe is removed by washing and retained fluorescent
signal is measured. The color of the dye retained identifies which
probe (A vs. G vs. C vs. T) was attached during the ligation
reaction. RNase H2 is then used to cleave the probe, removing the
"N" bases and universal bases but leaving the known base attached
to the acceptor nucleic acid. In this manner the identity of the
corresponding base within the template is determined, the acceptor
nucleic acid has been extended by one base, and an accessible
3'-hydroxyl is once again available for ligation, permitting
cycling of the process.
[0553] The following oligonucleotides shown below in Table 65, were
made as a representative synthetic system to demonstrate ligation
and subsequent cleavage of RNA-containing fluorescent ligation
probes using the methods of the present invention. The ligation
probes here have a fixed 9 base sequence (without any "N" bases or
universal base modifications). The designation "CLP" indicates
"cleavable ligation probe". The designation "ANA" indicates an
"acceptor nucleic acid" which provides the 3'-hydroxyl acceptor
site for a ligation reaction. "Targ-A" is a target nucleic acid,
which directs a ligation reaction involving the complementary "T"
ligation probe ("CLP-T-Cy3"). "Targ-T" is a target nucleic acid,
which directs a ligation reaction involving the complementary "A"
ligation probe ("CLP-A-FAM").
TABLE-US-00189 TABLE 65 CLP-C-TR 5'-pCaGCTGAAG-TR SEQ ID No. 272
CLP-G-Cy5 5'-pGaGCTGAAG-Cy5 SEQ ID No. 273 CLP-A-FAM
5'-pAaGCTGAAG-FAM SEQ ID No. 274 CLP-T-Cy3 5'-pTaGCTGAAG-Cy3 SEQ ID
No. 275 ANA 5'-CCCTGTTTGCTGTT SEQ ID TTTCCTTCTC No. 276 Targ-A
5'-AGTGTTTGCTCTTC SEQ ID AGCTAGAGAAGGAAAAA No. 277 CAGCAAACAGGG
Targ-T 5'-AGTGTTTGCTCTTC SEQ ID AGCTTGAGAAGGAAAAA No. 278
CAGCAAACAGGG DNA bases are shown in uppercase. RNA bases are shown
in lowercase. "p" is 5'-phosphate. TR is the fluorescent dye Texas
Red. Cy5 is the fluorescent dye Cyanine-5. Cy3 is the fluorescent
dye Cyanine-3. FAM is the fluorescent dye 6-carboxyfluorescein. The
position of base variation between Targ-A and Targ-T is underlined,
which is complementary to the 5'-base of the corresponding ligation
probe.
[0554] FIG. 36 shows the predicted results for a ligation-cleavage
reaction cycle using the synthetic oligonucleotide sequences shown
above. "Targ-A" (SEQ ID No. 277) will direct hybridization and
ligation of the "CLP-T-Cy3" probe (SEQ ID No. 275) while "Targ-T"
(SEQ ID No. 278) will direct hybridization and ligation of the
"CLP-A-FAM" probe (SEQ ID No. 274). Assuming that the reactions
have high specificity, the remaining two ligation probes do not
have a matching target in this experiment and so should not
participate in the ligation reaction. Following ligation, the newly
formed fusion of the "ANA"+"CLP" product will become a substrate
for RNase H2. Cleavage by RNase H2 will result in release of the
3'-end of the ligation probe (including the RNA base and the
fluorescent reporter dye), leaving the "ANA" molecule longer by one
base.
[0555] The "T" target nucleic acid (SEQ ID No. 278) or the "A"
target nucleic acid (SEQ ID No. 277) and the "ANA" acceptor nucleic
acid (SEQ ID No. 276) were mixed at 1.75 .mu.M and all 4 ligation
probes (SEQ ID Nos. 272-75) were added to a concentration of 3.5
.mu.M (each) in T4 DNA Ligase buffer (50 mM Tris-HCl pH 7.5, 10 mM
MgCl.sub.2, 10 mM dithiothreitol, 1 mM ATP) in a volume of 80
.mu.L, heated to 70.degree. C. for 3 minutes and cooled slowly to
25.degree. C. Ligation reactions were incubated at 37.degree. C.
for 5 minutes with or without 140 units of T4 DNA Ligase. The
reactions were stopped by heating at 65.degree. C. for 10 minutes.
Reaction volumes were then adjusted to 200 .mu.L with the addition
of RNase H2 buffer [Tris-HCl pH 8.0 (final concentration 10 mM),
NaCl (final concentration 50 mM), MgCl.sub.2 (final concentration 4
mM)] and 20 units of RNase H2 was added to each tube. Reaction
mixtures were incubated at 60.degree. C. for 30 minutes, followed
by desalting over a Sephadex G25 column, and the samples were
lyophilized. Samples were rehydrated in 70 .mu.L of water and 10
.mu.L aliquots were analyzed on a 20% acrylamide, 7M urea,
denaturing gel, followed by visualization using GelStar stain
(#50535 GelStar Nucleic Acid Gel Stain, Lonza). The remainder of
the reactions was saved at -20.degree. C. for future testing,
including mass spectrometry or other methods as needed.
[0556] The gel image is shown in FIG. 37. Lanes 1 and 5 show the
component oligonucleotides in the absence of enzymes to visualize
migration relative to size markers (lane M). Lanes 2 and 3 are
duplicate reactions where Targ-A (SEQ ID No. 277) was incubated
with the 4 cleavage ligation probes (SEQ ID Nos. 272-75) in the
presence of T4 DNA Ligase. An upward size shift of the acceptor
nucleic acid (ANA, SEQ ID No. 276) is clearly seen which represents
ligation with CLP-T-Cy3 (SEQ ID No. 275) and is identified as the
ligation product. Specific ligation with the correct CLP-T-Cy3
probe and not the other 3 probes (mismatched bases) occurred, which
was verified by visual inspection of the color of the dye (this
cannot be appreciated in the black and white image shown in FIG.
37) and was further verified by mass spectrometry. Similarly, lanes
7 and 8 are duplicate reactions where Targ-T (SEQ ID No. 278) was
incubated with the 4 cleavage ligation probes (SEQ ID Nos. 272-75)
in the presence of T4 DNA Ligase. An upward size shift of the
acceptor nucleic acid (ANA, SEQ ID No. 276) is clearly seen which
represents ligation with CLP-A-FAM (SEQ ID No. 274) and is
identified as the ligation product. Specific ligation with the
correct CLP-A-FAM probe and not the other 3 probes (mismatched
bases) occurred, which was verified again by visual inspection of
the color of the dye and confirmed by mass spectrometry analysis.
Finally, lanes 4 and 8 demonstrate that these ligation products are
reduced in size when treated with RNase H2, indicating that
cleavage occurred. Note that the resulting bands show slightly
reduced mobility compared with the original ANA band, indicating
that this new species is longer than the starting material. Mass
spectrometry confirmed that actual mass of the reaction products in
lanes 4 and 8 were consistent with the predicted 1-base elongation
of the starting ANA nucleic acid, that the correct base was
inserted, and that the new "ANA+1" species had a 3'-hydroxyl. The
new ANA+1 species is now prepared for a second cycle of
ligation/cleavage.
[0557] This example has therefore demonstrated that short
RNA-containing short probes can be specifically ligated to an
acceptor nucleic acid in the presence of a complementary target
nucleic acid. Ligation is sensitive to the identify of the template
base matching the 5'-terminal base of the ligation probe and
specific ligation of the correct complementary probe can be
detected from within a heterogeneous mix of different probe
sequences. Finally, RNase H2 can cleave the ligation probe at the
5'-side of the RNA base, releasing the bulk of the probe, resulting
in an acceptor nucleic acid molecule which has been extended by one
base in length. The extended acceptor nucleic acid contains a
3'-hydroxyl and can be used in repeated cycles of
ligation/cleavage.
Example 30
Use of Universal Bases in RNase H2-Cleavable Ligation Probes
[0558] In Example 29 above it was proposed that universal bases,
such as 5'-nitroindole or inosine, could be used in cleavable
ligation probes. The present example demonstrates use of the
universal base 5-nitroindole in a model system where the probe
sequence is fixed (does not contain random N-bases). The
oligonucleotides, shown below in Table 66, were synthesized based
upon the synthetic probe/template system in Example 29. Cleavage
ligation probes were designed as 8 mers with a 5'-phosphate, an "A"
base at the 5'-end (to direct ligation to the "T" target), a single
ribonucleotide, and 2 or 3 additional fixed DNA bases. Three or
four 5-nitroindole bases were positioned towards the 3'-end. A FAM
fluorescent dye was attached at the 3'-end. The same acceptor
nucleic acid (ANA) and T-target nucleic acid were employed as in
Example 29. A reaction scheme showing alignment of oligonucleotide
components for this example is shown in FIG. 38.
TABLE-US-00190 TABLE 66 CLP-A-FAM- 5'-pAaGCTXXX-FAM SEQ ID 3x5NI
No. 279 CLP-A-FAM- 5'-pAaGCXXXX-FAM SEQ ID 4x5NI No. 280 ANA
5'-CCCTGTTTGCTGT SEQ ID TTTTCCTTCTC No. 276 Targ-T 5'-AGTGTTTGCTCTT
SEQ ID CAGCTTGAGAAGGAAA No. 278 AACAGCAAACAGGG DNA bases are shown
in uppercase. RNA bases are shown in lowercase. "p" is
5'-phosphate. "X" is the universal base 5-nitroindole. FAM is the
fluorescent dye 6-carboxyfluorescein. The position of base
hybridization with the 5'-end of the ligation probe is underlined
on the target.
[0559] The "T" target nucleic acid (SEQ ID No. 278) and the "ANA"
acceptor nucleic acid (SEQ ID No. 276) were mixed at 2 .mu.M with
the 3.times. or 4.times.5'-nitroindole containing CLPs (cleavable
ligation probes, SEQ ID Nos. 279-80) in T4 DNA Ligase buffer (50 mM
Tris-HCl pH 7.5, 10 mM MgCl.sub.2, 10 mM dithiothreitol, 1 mM ATP).
The reactions were heated at 70.degree. C. for 5 minutes and cooled
slowly to 25.degree. C. T4 DNA Ligase (New England Biolabs) was
added at a range of 7.5-120 units and the ligation reactions were
incubated at 25 or 37.degree. C. for 5 minutes. The reactions were
terminated by the addition of EDTA to a final concentration of 50
mM. Final reaction volumes were 50 .mu.L. An equal volume of 90%
formamide, 1.times. TBE loading buffer was added to each sample,
which were then heat denatured at 70.degree. C. for 3 minutes and
cooled on ice. Samples were separated on a denaturing 7M urea, 20%
polyacrylamide gel. Gels were stained using GelStar.TM. stain and
visualized with UV excitation. The gel image is shown in FIG.
39.
[0560] The 8 mer cleavable ligation probe with three 5-nitroindole
universal bases (SEQ ID No. 279) worked well and showed near 100%
ligation efficiency using the higher enzyme amounts (60-120 units
T4 DNA Ligase). In contrast, the 8 mer cleavable ligation probe
with four 5-nitroindole universal bases (SEQ ID No. 280) did not
ligate to the acceptor nucleic acid using any amount of enzyme. The
same results were seen at 25.degree. C. and at 37.degree. C.
suggesting that this difference in reactivity does not relate to
difference in Tm of the two probes. It is more likely that the
differential reactivity relates to substrate preferences for the T4
DNA Ligase enzyme. This Example demonstrates that three
5-nitroindole bases can be positioned at the 3'-end of an 8 mer
ligation probe and retain good function. This same experiment was
repeated using 9 mer ligation probes. In this case, a probe having
"six DNA+three 5-nitroindole bases" and a probe having "five
DNA+four 5-nitroindole bases" were both substrates for T4 DNA
Ligase but a probe with "four DNA bases+five 5-nitroindole bases"
did not (data not shown), consistent with the idea that T4 DNA
Ligase requires 5 fixed DNA bases towards the 5'-end of the
ligation probe to function well and that 5'-nitroindole bases can
be introduced after this requirement is met. The precise optimal
probe design can vary with different ligase enzymes.
[0561] These findings are significant as it permits synthesis of
lower complexity pools of ligation probes.
Example 31
Use of Random Bases and Universal Bases in RNase H2-Cleavable
Ligation Probes
[0562] Examples 29 and 30 demonstrated use of RNase H2-cleavable
ligation probes where some or all of the probe sequence was a
perfect match to the target. In sequencing a nucleic acid of
unknown sequence, it is necessary to use probes that contain
primarily random bases so that probe hybridization can occur for
any sequence encountered. The present Example demonstrates use of 8
mer cleavable ligation probes having a random base (Nmer) domain, a
universal base (5-nitroindole) domain and only a single fixed DNA
base at the 5'-end. The following oligonucleotides shown in Table
67 were employed:
TABLE-US-00191 TABLE 67 CLP-A-FAM 5'-pAnNNNXXX-FAM SEQ ID 4N +
3x5NI No. 281 CLP-T-Cy3 5'-pTnNNNXXX-Cy3 SEQ ID 4N + 3x5NI No. 282
CLP-G-Cy5 5'-pGnNNNXXX-Cy5 SEQ ID 4N + 3x5NI No. 283 ANA
5'-CCCTGTTTGCTGTTTTTCCTTCTC SEQ ID No. 276 Targ-T
5'-AGTGTTTGCTCTTCAGCTTGAGAA SEQ ID GGAAAAACAGCAAACAGGG No. 278 DNA
bases are shown in uppercase. RNA bases are shown in lowercase. "p"
is 5'-phosphate. "N" represents a random mix of the DNA bases A, G,
C, and T. "n" represents a random mix of the RNA bases A, C, G, and
U. "X" is the universal base 5-nitroindole. FAM is the fluorescent
dye 6-carboxyfluorescein. Cy5 is the fluorescent dye Cyanine-5. Cy3
is the fluorescent dye Cyanine-3. The position of base
hybridization with the 5'-end of the ligation probe is underlined
on the target.
[0563] The "T" target nucleic acid (SEQ ID No. 278) and the "ANA"
acceptor nucleic acid (SEQ ID No. 276) were mixed together at a
final concentration of 0.4 .mu.M each and the three cleavable
ligation probes (SEQ ID Nos. 281-83) were individually added at a
final concentration of 50 .mu.M (125-fold excess over the target
and acceptor) in T4 DNA Ligase buffer in a final reaction volume of
50 .mu.L. Reactions were heated to 70.degree. C. for 5 minutes and
cooled slowly to 25.degree. C. T4 DNA Ligase was added (400 U) and
the ligation reactions were incubated at 37.degree. C. for 30
minutes. The reactions were stopped by heating at 65.degree. C. for
10 minutes followed by desalting over a Sephadex G25 column, after
which the samples were lyophilized and rehydrated in 10 .mu.L of
water mixed with 10 .mu.L of 90% formamide, 1.times. TBE loading
buffer. Samples were heat denatured at 70.degree. C. for 3 minutes
and cooled on ice. Reaction products were separated on a 20%
acrylamide 7M urea denaturing gel, followed by visualization using
GelStar stain with UV transillumination (50535 GelStar Nucleic Acid
Gel Stain, Lonza). Results are shown in FIG. 40.
[0564] The target nucleic acid contained a "T" base at the site
complementary to the point of ligation. This template correctly
directed ligation of the "A-FAM" ligation probe (SEQ ID No. 281)
but not the mismatch "T-Cy3" (SEQ ID No. 282) or "G-Cy5" (SEQ ID
No. 283) ligation probes. Ligation specificity was directed by a
single fixed DNA base at the 5'-end of the ligation probes which
otherwise comprised random "N" bases or universal 5-nitroindole
bases. The ligation probes were added to the ligation reactions at
125-fold molar excess over the target and the acceptor nucleic
acids. The ligation probes contain a 4-base "N" domain, so the
complexity of the nucleic acid mixture was 4.sup.4 (256). Thus the
reaction theoretically contained sufficient perfect matched probe
to ligate with only about 50% of the input acceptor nucleic acid.
It is evident from the relative fluorescent images in FIG. 40 that
approximate half of the acceptor was present in the longer ligation
product species and half was unreacted, indicating that the
reaction proceeded as expected. If mismatched sequences ligated to
the acceptor with any appreciable efficiency, then the 125-fold
excess of ligation probe would most likely have reacted with
>50% of the acceptor nucleic acid molecules, which was not
observed. Thus ligation reactions using cleavable ligation probes
of this design were both efficient and specific.
Example 32
Use of RNase H2-Cleavable Probes in an Oligonucleotide Ligation
Assay (OLA)
[0565] Use of cleavable ligation probes in DNA sequencing
represents just one potential format/application for this general
class of assay. The sequencing application is unique in that the
target nucleic acid is of unknown sequence. More typically,
oligonucleotide ligation assays are employed to determine the
presence or absence of a known nucleic acid sequence within a
sample nucleic acid of interest. For example, an OLA can be
employed to detect the presence of a nucleic acid sequence specific
for pathogenic organisms in the background of human DNA. Another
example would be to determine the presence or absence of a known,
defined polymorphism at a specific target nucleic acid locus (e.g.,
an allelic discrimination assay or SNP assay). In all of these
applications, one ligation probe is positioned so that the 3'-most
or 5'-most base aligns with the SNP site and a second perfect-match
nucleic acid is positioned adjacent so that if the probe sequence
is a match for the SNP base, then a ligation event can occur. If
the probe sequence is a mismatch for the SNP base, then ligation is
inhibited. The ligation event results in formation of a detectable
species.
[0566] An allelic discrimination (SNP) assay is shown in this
Example to demonstrate utility of the novel RNase H2 cleavable
ligation oligonucleotide probes of the present invention. Sequence
designs shown herein place the SNP site towards the 3'-end of the
acceptor ligation probe.
[0567] A traditional OLA employs three synthetic oligonucleotides
to discriminate between two alleles (FIG. 41A). If the SNP site
comprises a "C" allele and an "A" allele, then two acceptor
oligonucleotides are required, one bearing a "G" base (match for
the "C" allele) and one bearing a "T" base (match for the "A"
allele). The acceptor oligonucleotides have a free 3'-hydroxyl
group. A third oligonucleotide (a donor nucleic acid) is employed
that hybridizes to the target so as to place its 5'-end adjacent to
the 3'-end of the ligation probe. The acceptor nucleic acid will
have a 5'-phosphate; generally the 3'-end of the donor
oligonucleotide is blocked so that it cannot participate in a
ligation reaction. In this way, perfect match hybridization of both
a acceptor and the donor probes on the target will position the two
oligonucleotides in a head-to-tail fashion that enables ligation
between the 3'-hydroxyl of the acceptor with the 5'-phosphate of
the donor (FIG. 41B). In contrast, a mismatch at the SNP site
disrupts this structure and inhibits ligation. In the traditional
OLA, the identity of the SNP base is interrogated once at the time
of hybridization/ligation and specificity is entirely dependent
upon the ability of the DNA Ligase to perform ligation on the
perfect matched but not the mismatched species. Typically the three
oligonucleotides (two ligation probes and the acceptor) have a
similar Tm so that they can function together with the target
nucleic acid under identical conditions.
[0568] The new RNase H2 OLA of the present invention employs four
synthetic nucleic acids to discriminate between two alleles (FIG.
42A). If the SNP site comprises a "C" allele and an "A" allele,
then two cleavable acceptor ligation probes are required in this
embodiment, one bearing a "G" base (match for the "C" allele) and
one bearing a "T" base (match for the "A" allele). The cleavable
acceptor ligation probes have a single RNA base positioned towards
the 3'-end of the molecule that is aligned to be complementary or
not (match vs. mismatch) with the base at the target SNP site.
Additional DNA bases are positioned 3'- to the RNA base (preferably
four DNA bases, all being complementary to the target) and a
blocking group is placed at the 3'-end to prevent ligation. The
general design and function of the cleavable ligation probe is
similar to the cleavable primers demonstrated in Example 13 in a
qPCR format SNP discrimination assay. The cleavable ligation probes
can also be designed using various chemically modified bases and
abasic residues as outlined in the above Examples to improve SNP
discrimination at the RNase H2 cleavage site (see Examples 22, 23,
25, and 28). Preferably the cleavable ligation probes will be
designed to have a Tm in the range of 60-70.degree. C. (in RNase H2
cleavage buffer) to permit hybridization of the cleavable probe
with target in the optimal temperature range for the enzyme.
[0569] Unlike the traditional OLA format, the donor
oligonucleotides in the RNase H2 OLA format are also SNP
interrogation probes. Thus two donor probes are required, one
bearing a "G" base (match for the "C" allele) and one bearing a "T"
base (match for the "A" allele). Both donor or probes have a
phosphate at the 5'-end to enable ligation and optionally are
blocked at the 3'-end (FIG. 42A). The two donor ligation probes can
have a lower Tm than the RNase H2 cleavable ligation probes so that
hybridization of the cleavable ligation probes and the donor
ligation probes with the target nucleic acid can be differentially
regulated by control of reaction temperature. These donor probes in
the assay format do not interact with RNase H2.
[0570] To perform an RNase H2 OLA, all four OLA probes are mixed in
the presence of the target nucleic acid in a buffer compatible with
RNase H2 activity (see above examples). Preferably this will be
done around 60-70.degree. C. The RNase H2 cleavable acceptor
oligonulceotide is complementary to and will hybridize to the
target nucleic acid under these conditions. If the RNA base of the
acceptor probe and the base at the target SNP site match, then
RNase H2 cleavage can occur (FIG. 42B). It is preferred that the
donor ligation probe (the non-cleavable probe) has a lower Tm than
the cleavable probe. The first stage of the reaction (hybridization
of the acceptor oligonucleotide and cleavage by RNase H2) can then
be carried out at a temperature that is sufficiently above the Tm
of the non-cleavable donor probes that they do not hybridize to
target. Cleavage of the acceptor probe by RNase H2 removes the RNA
base and uncovers the SNP site, making it available to hybridize
with the non-cleavable ligation probe (the donor
oligonucleotide).
[0571] Once the RNase H2 cleavage phase of the OLA is complete,
reaction temperature is lowered to permit hybridization of the
non-cleavable ligation probe to the target. In the presence of DNA
Ligase, the 5'-end of the non-cleavable probe will ligate to the
3'-end of the adjacent cleaved probe (FIG. 42B), if the base at the
5'-end of the donor probe pairs with the base at the SNP site. Thus
the RNase H2 OLA assay interrogates the identity of the base at the
SNP site twice, once during RNase H2 mediated cleavage of the
acceptor oligonucleotide probe and again at the ligation reaction
(FIG. 42C). Double interrogation of the identity of the SNP base by
two different enzymatic events provides for greater specificity
than can be achieved using a traditional OLA.
Example 33
SNP Discrimination Using RNase H2-Cleavable Probes in an OLA
[0572] A variety of methods exist that enable detection of OLA
products. In the present Example, fluorescence detection is
performed in a bead capture assay format to perform an RNase H2 OLA
allelic discrimination assay as outlined in Example 32. Sequences
were designed that were compatible for use with the Luminex xMAP
fluorescent microbead system with detection on a Luminex L100
detection system (Luminex, Austin, Tex.).
[0573] The "OLA-C-antitag" and "OLA-T-antitag" sequences (SEQ ID
Nos. 284-5) were made with a 5'-amino modifier to permit
conjugation to carboxylate xMAP fluorescent beads using
carbodiimide coupling chemistry. The "OLA-T-Tag" and "OLA-C-Tag"
sequences (SEQ ID Nos. 288-9) which serve as donor oligonucleotides
in the ligation reaction have a 12-base sequence towards the 5'-end
which is complementary to the target sequence and positions the SNP
site (C/T base) at the 5'-end. Tm for these 12-base domains is
estimated to be 50-53.degree. C. (in 10 mM Mg.sup.++ containing
buffer). Both sequences have a 5'-phosphate to permit ligation. The
3'-end of these sequences is a "tag" sequence which is
complementary to the "antitag" sequence and permits capture to
antitag bearing beads by hybridization. The "OLA-C" and "OLA-T"
probes (SEQ ID Nos. 286-7) serve as the acceptor fragment and are
complementary to the target and position the single ribonucleotide
base (rC or rU) at the SNP site. Tm for the cleavable ligation
probes is estimated to be .about.75.degree. C. (in 10 mM Mg.sup.++
containing buffer). Both of the oligonucleotide probes have a
biotin at the 5'-end which will enable binding of a reporter dye,
Streptavidin-phycoerythrin, for detection by the Luminex L100
system. Synthetic 98 mer oligonucleotide targets corresponding to
the "C" allele (G base in the target, SEQ ID No. 290) and "T"
allele (A base in the target, SEQ ID No. 291) were employed in this
Example. The sequences corresponding to SEQ ID Nos. 284-291 are
shown below in Table 68. Alignment and interaction of the different
probe, target, tag, and antitag sequences during the various step
in this assay are shown in FIG. 43.
TABLE-US-00192 TABLE 68 OLA-C 5' aminoC12-GATTTG SEQ ID antitag
TATTGATTGAGATTAAAG No. 284 OLA-T 5' aminoC12-GATTGT SEQ ID antitag
AAGATTTGATAAAGTGTA No. 285 rs4939827 5' Biotin-CACCATGC SEQ ID OLA
C TCACAGCCTCATCCAAAA No. 286 GAGGAAAcAGGA-x rs4939827 5'
Biotin-CACCATGC SEQ ID OLA T TCACAGCCTCATCCAAAA No. 287
GAGGAAAuAGGA-x rs4939827 5' pCAGGACCCCAGACT SEQ ID OLA C
TTAATCTCAATCAATACA No. 288 Tag AATC-x rs4939827 5' pTAGGACCCCAGATA
SEQ ID OLA T CACTTTATCAAATCTTAC No. 289 Tag AATC-x Targ-C
5'CCCAGCATTGTCTGTG SEQ ID TTTCCTGAGGAGTCTGAG No. 290
GGAGCTCTGGGGTCCTGT TTCCTCTTTTGGATGAGG CTGTGAGCATGGTGGATT AGAGACAGCC
Targ-T 5'CCCAGCATTGTCTGTG SEQ ID TTTCCTGAGGAGTCTGAG No. 291
GGAGCTCTGGGGTCCTAT TTCCTCTTTTGGATGAGG CTGTGAGCATGGTGGATT AGAGACAGCC
DNA bases are uppercase. RNA bases are lowercase. Biotin is a
Biotin-TEG group. X represents a C3 spacer. For the OLA C/T Tag
oligonucleotides, the portion of the sequence which is the "tag"
and binds the "antitag" sequence is underline. The site of the SNP
with the target sequence under interrogation is underlined and in
bold.
[0574] Coupling of Antitag Oligos to xMAP Microspheres.
[0575] Anti-tag oligonucleotides containing a 5' amino group were
coupled to 1.25.times.10.sup.7 xMAP Multi-Analyte COOH Microspheres
(L100-C127-01 and L100-C138-01, Luminex, Austin, Tex.) using 3
mg/mL N-(3-Dimethylaminopropyl)-N'-ethylcarbodiimide hydrochloride
(03449-1G, Sigma Aldritch), in 0.1 M MES, pH 4.5 buffer (M-8250
Sigma-Aldritch) at room temperature for 90 minutes in the dark
(modified manufacturer's protocol). After coupling, the
microspheres were washed once with 0.02% Tween20, and then once
with 0.1% SDS. Microspheres were re-suspended in 200 .mu.L of TE pH
7.5. The concentration of microspheres was determined by counting
with a hemocytometer under a light microscope (Nikon.TM.S, Freyer
Company, Carpentersville, Ill.). Successful coupling was determined
by hybridizing 25-250 fmoles of complementary oligonucleotides
containing a 5' biotin modification and detecting the hybrids with
2 .mu.g/mL streptavidin R-phycoerythrin conjugate (S866 1 mg/mL,
Invitrogen, Carlsbad, Calif.). Mean fluorescence intensity had to
increase in a concentration dependent manner. No cross
hybridization was observed between the two anti-tag sequences.
[0576] OLA Assay.
[0577] RNase H2 digestion mixtures (10 .mu.L) were prepared
containing rs4939827 OLA C and rs4939827 OLA T oligos (SEQ ID Nos.
286-7) at a final concentration of 250 nM, and either C, T or C/T
mix template oligonucleotides (SEQ ID Nos. 290-91) at 125 nM in a
20 mM Tris-HCl (pH 7.6 at 25.degree. C.), 25 mM KAc, 10 mM MgAc, 10
mM DTT, 1 mM NAD, and 0.1% Triton X-100 buffer (Taq DNA Ligase
buffer, New England Biolabs, Ipswitch, Mass.). Samples were
incubated for 30 minutes at 65.degree. C. with or without 5 mU of
Pyrococcus abyssi RNase H2. For each RNase H2 digestion reaction,
the volume was increased to 25 .mu.L by adding 2.5 pmoles of
rs4939827 OLA 12C Tag and 2.5 pmoles rs4939827 OLA 12T Tag
oligonucleotides (SEQ ID Nos. 288-9) (100 nM final concentration
for each oligo), with or without 40 U of Taq DNA Ligase (New
England Biolabs, Ipswitch, Mass.), maintaining a final buffer
composition of 20 mM Tris-HCl (pH 7.6 at 25.degree. C.), 25 mM KAc,
10 mM MgAc, 10 mM DTT, 1 mM NAD, and 0.1% Triton X-100. The
ligation reactions were incubated at 45.degree. C. for 30
minutes.
[0578] Capture of Ligation Product on Fluorescent Beads and
Detection of Signal.
[0579] 10 .mu.L of each ligation mixture was combined with 15 .mu.L
of H.sub.2O, and 25 .mu.L of the xMAP bead mixture (Bead sets 127
and 138) at a density of 100 beads of each type/.mu.L. The samples
were heated to 70.degree. C. for 90 seconds followed by 50.degree.
C. for 30 minutes. The samples were transferred to a Millipore
Multiscreen filtration plate (MABVN1250, Millipore, Bedford,
Mass.), and washed two times with 100 .mu.L of 50.degree. C. 0.2 M
NaCl, 0.1 M Tris pH 8.0, 0.08% Triton X-100 buffer. Microspheres
were incubated at 50.degree. C. for 15 minutes with 75 .mu.L of a 2
.mu.g/mL solution of streptavidin-R phycoerythrin (S866 1 mg/mL,
Invitrogen, Carlsbad, Calif.). Mean fluorescence was measured on a
Luminex L100 detection system (Luminex, Austin, Tex.).
[0580] Results are shown in FIG. 44. Fluorescent beads bearing the
"C" allele antitag sequences showed positive fluorescent signal
only when the reaction was run in the present of the "C" allele
target or the "C/T" mix. Fluorescent beads bearing the "T" allele
antitag sequences showed positive fluorescent signal only when the
reaction was run in the present of the "T" allele target or the
"C/T" mix. Signal was dependent on use of RNase H2 and was not
observed in the absence of target DNA. Thus the RNase H2 cleavable
oligonucleotide ligation assay of the invention was demonstrated to
be effective at distinguishing the presence of a C/T SNP present in
a target DNA in a highly specific fashion.
Additional Acknowledgements
[0581] All references, including publications, patent applications,
and patents, cited herein are hereby incorporated by reference to
the same extent as if each reference were individually and
specifically indicated to be incorporated by reference and were set
forth in its entirety herein.
[0582] The use of the terms "a" and "an" and "the" and similar
referents in the context of describing the invention (especially in
the context of the following claims) are to be construed to cover
both the singular and the plural, unless otherwise indicated herein
or clearly contradicted by context. The terms "comprising,"
"having," "including," and "containing" are to be construed as
open-ended terms (i.e., meaning "including, but not limited to,")
unless otherwise noted. Recitation of ranges of values herein are
merely intended to serve as a shorthand method of referring
individually to each separate value falling within the range,
unless otherwise indicated herein, and each separate value is
incorporated into the specification as if it were individually
recited herein. All methods described herein can be performed in
any suitable order unless otherwise indicated herein or otherwise
clearly contradicted by context. The use of any and all examples,
or exemplary language (e.g., "such as") provided herein, is
intended merely to better illuminate the invention and does not
pose a limitation on the scope of the invention unless otherwise
claimed. No language in the specification should be construed as
indicating any non-claimed element as essential to the practice of
the invention.
[0583] Preferred embodiments of this invention are described
herein, including the best mode known to the inventors for carrying
out the invention. Variations of those preferred embodiments may
become apparent to those of ordinary skill in the art upon reading
the foregoing description. The inventors expect skilled artisans to
employ such variations as appropriate, and the inventors intend for
the invention to be practiced otherwise than as specifically
described herein. Accordingly, this invention includes all
modifications and equivalents of the subject matter recited in the
claims appended hereto as permitted by applicable law. Moreover,
any combination of the above-described elements in all possible
variations thereof is encompassed by the invention unless otherwise
indicated herein or otherwise clearly contradicted by context.
Sequence CWU 1
1
3481697DNAArtificial SequenceSynthetic 1ggatccgatg aagattgctg
gcatcgatga agccggccgt ggcccggtaa ttggtccaat 60ggttatcgct gcggtagtcg
tggacgaaaa cagcctgcca aaactggaag agctgaaagt 120gcgtgactcc
aagaaactga ccccgaagcg ccgtgaaaag ctgtttaacg aaattctggg
180tgtcctggac gattatgtga tcctggagct gccgcctgat gttatcggca
gccgcgaagg 240tactctgaac gagttcgagg tagaaaactt cgctaaagcg
ctgaattccc tgaaagttaa 300accggacgta atctatgctg atgcggctga
cgttgacgag gaacgttttg cccgcgagct 360gggtgaacgt ctgaactttg
aagcagaggt tgttgccaaa cacaaggcgg acgatatctt 420cccagtcgtg
tccgcggcga gcattctggc taaagtcact cgtgaccgtg cggttgaaaa
480actgaaggaa gaatacggtg aaatcggcag cggttatcct agcgatcctc
gtacccgtgc 540gtttctggag aactactacc gtgaacacgg tgaattcccg
ccgatcgtac gtaaaggttg 600gaaaaccctg aagaaaatcg cggaaaaagt
tgaatctgaa aaaaaagctg aagaacgtca 660agcaactctg gaccgttatt
tccgtaaagt gaagctt 6972685DNAArtificial SequenceSynthetic
2ggatccgatg aagattggtg gcatcgacga agccggccgt ggtccggcga tcggtccgct
60ggtagtagct actgttgtag tggatgaaaa aaacatcgaa aaactgcgta acatcggcgt
120aaaagactcc aaacagctga cgccgcacga acgtaaaaac ctgttttccc
agatcacctc 180cattgcggat gattacaaga tcgtaatcgt gtctccggaa
gaaattgaca accgtagcgg 240taccatgaac gagctggaag ttgaaaaatt
cgcgctggcg ctgaactctc tgcagatcaa 300gccggctctg atctacgcag
acgcagcaga tgttgatgca aaccgcttcg catccctgat 360cgaacgtcgc
ctgaactata aagccaaaat catcgcggaa cacaaagcag acgcaaagta
420cccggtcgtt tctgcggcga gcattctggc gaaggttgtg cgtgacgaag
aaatcgaaaa 480gctgaaaaag caatatggcg actttggcag cggttacccg
agcgacccga aaacgaagaa 540atggctggag gagtattaca agaaacataa
cagcttccca ccgatcgttc gtcgtacgtg 600ggaaactgtc cgcaaaattg
aagagtccat caaagccaaa aagtcccagc tgaccctgga 660taaattcttc
aagaaaccga agctt 6853702DNAArtificial SequenceSynthetic 3ggatccgatg
attatcattg tatcgatgaa gctggccgtg gtcctgtact gggcccgatg 60gttgtatgtg
cgttcgctat cgagaaggaa cgtgaagaag aactgaaaaa gctgggcgtt
120aaagattcta aagaactgac gaagaataaa cgcgcgtacc tgaaaaagct
gctggagaac 180ctgggctacg tggaaaagcg catcctggag gctgaggaaa
ttaaccagct gatgaacagc 240attaacctga acgacattga aatcaacgca
ttcagcaagg tagctaaaaa cctgatcgaa 300aagctgaaca ttcgcgacga
cgaaatcgaa atctatatcg acgcttgttc tactaacacc 360aaaaagttcg
aagactcttt caaagataaa atcgaagata tcattaaaga acgcaatctg
420aatatcaaaa tcattgccga acacaaagca gacgccaagt acccagtagt
gtctgcggcg 480agcattatcg cgaaagcaga acgcgacgag atcatcgatt
attacaagaa aatctacggt 540gacatcggct ctggctaccc atctgacccg
aaaaccatca aattcctgga agattacttt 600aaaaagcaca agaaactgcc
ggatatcgct cgcactcact ggaaaacctg caaacgcatc 660ctggacaaat
ctaaacagac taaactgatt atcgaaaagc tt 7024685DNAArtificial
SequenceSynthetic 4ggatccgatg aaagttgcag gtgcagatga agctggtcgt
ggtccagtta ttggtccgct 60ggttattgtt gctgctgttg tggaggaaga caaaatccgc
tctctgacta agctgggtgt 120taaagactcc aaacagctga ccccggcgca
acgtgaaaaa ctgttcgatg aaatcgtaaa 180agtactggat gattactctg
tggtcattgt gtccccgcag gacattgacg gtcgtaaggg 240cagcatgaac
gaactggagg tagaaaactt cgttaaagcc ctgaatagcc tgaaagttaa
300gccggaagtt atttacattg attccgctga tgttaaagct gaacgtttcg
ctgaaaacat 360tcgcagccgt ctggcgtacg aagcgaaagt tgtagccgaa
cataaagcgg atgcgaagta 420tgagatcgta tccgcagcct ctatcctggc
aaaagttatc cgtgaccgcg agatcgaaaa 480gctgaaagcc gaatacggtg
attttggttc cggttacccg tctgatccgc gtactaagaa 540atggctggaa
gaatggtata gcaaacacgg caatttcccg ccgatcgtgc gtcgtacttg
600ggatactgca aagaaaatcg aagaaaaatt caaacgtgcg cagctgaccc
tggacaactt 660cctgaagcgt tttcgcaaca agctt 6855649DNAArtificial
Sequencesynthetic 5ggatccgatg cgcgttggca tcgatgaagc gggtcgcggt
gccctgatcg gcccgatgat 60tgttgctggt gttgtaatct ctgacactaa actgaagttt
ctgaaaggca tcggcgtaaa 120agactctaaa cagctgactc gcgagcgtcg
tgaaaagctg tttgatattg ttgctaacac 180tgtggaagca ttcactgtcg
ttaaagtttt cccttatgaa atcgacaact ataacctgaa 240tgacctgacc
tacgacgcag tttctaaaat catcctgagc ctgtctagct ttaacccaga
300aattgtaacg gttgataaag tgggcgatga gaaaccggtt atcgaactga
ttaataagct 360gggctacaaa agcaacgtcg tacacaaggc agatgtactg
tttgtagaag cctccgctgc 420tagcatcatt gcgaaagtta ttcgtgataa
ctacattgac gaactgaaac aagtatacgg 480tgactttggt agcggttacc
cagctgatcc tcgcactatc aaatggctga aatctttcta 540cgaaaagaat
ccgaatccgc cgccaatcat tcgtcgttcc tggaagattc tgcgttctac
600cgccccgctg tattacattt ccaaagaagg tcgccgtctg tggaagctt
649630DNAArtificial SequenceSynthetic 6ctcgtgaggt gaugcaggag
atgggaggcg 30730DNAArtificial SequenceSynthetic 7cgcctcccat
ctcctgcatc acctcacgag 30860DNAArtificial SequenceSynthetic
8ctcgtgaggt gaugcaggag atgggaggcg gagcactcca ctacgtcctc taccctccgc
60912DNAArtificial SequenceSynthetic 9ctcgtgaggt ga
121012DNAArtificial SequenceSynthetic oligonucleotide 10agatgggagg
cg 121130DNAArtificial SequenceSynthetic oligonucleotide
11ctcgtgaggt gatgcaggag atgggaggcg 301230DNAArtificial
SequenceSynthetic oligonucleotide 12cgcctcccat ctcctgcatc
acctcacgag 301360DNAArtificial SequenceSynthetic oligonucleotide
13ctcgtgaggt gatgcaggag atgggaggcg gagcactcca ctacgtcctc taccctccgc
601430DNAArtificial SequenceSynthetic oligonucleotide 14ctcgtgaggt
gatggaggag atgggaggcg 301530DNAArtificial SequenceSynthetic
oligonucleotide 15cgcctcccat ctcctccatc acctcacgag
301660DNAArtificial SequenceSynthetic oligonucleotide 16ctcgtgaggt
gatggaggag atgggaggcg gagcactcca ctacttcctc taccctccgc
601730DNAArtificial SequenceSynthetic oligonucleotide 17ctcgtgaggt
gatgaaggag atgggaggcg 301830DNAArtificial SequenceSynthetic
oligonucleotide 18cgcctcccat ctccttcatc acctcacgag
301960DNAArtificial SequenceSynthetic oligonucleotide 19ctcgtgaggt
gatgaaggag atgggaggcg gagcactcca ctacttcctc taccctccgc
602030DNAArtificial SequenceSynthetic oligonucleotide 20ctcgtgaggt
gatguaggag atgggaggcg 302130DNAArtificial SequenceSynthetic
oligonucleotide 21cgcctcccat ctcctacatc acctcacgag
302260DNAArtificial SequenceSynthetic oligonucleotide 22ctcgtgaggt
gatguaggag atgggaggcg gagcactcca ctacatcctc taccctccgc
602314DNAArtificial SequenceSynthetic oligonucleotide 23ctcgtgaggt
gatg 142416DNAArtificial SequenceSynthetic oligonucleotide
24caggagatgg gaggcg 162530DNAArtificial SequenceSynthetic
oligonucleotide 25ctcgtgaggt gatgcaggag atgggaggcg
302630DNAArtificial SequenceSynthetic oligonucleotide 26ctcgtgaggt
gatgcaggag atgggaggcg 302730DNAArtificial SequenceSynthetic
oligonucleotide 27ctcgtgaggt gatgcaggag atgggaggcg
302830DNAArtificial SequenceSynthetic oligonucleotide 28ctcgtgaggt
gatguaggag atgggaggcg 302930DNAArtificial SequenceSynthetic
oligonucleotide 29ctcgtgaggt gatgcuggag atgggaggcg
303030DNAArtificial SequenceSynthetic oligonucleotide 30ctcgtgaggt
gatnnaggag atgggaggcg 303130DNAArtificial SequenceSynthetic
oligonucleotide 31ctcgtgaggt gatucaggag atgggaggcg
303230DNAArtificial SequenceSynthetic oligonucleotide 32ctcgtgaggt
gatucaggag atgggaggcg 303330DNAArtificial SequenceSynthetic
oligonucleotide 33ctcgtgaggt gatucaggag atgggaggcg
303430DNAArtificial SequenceSynthetic oligonucleotide 34ctcgtgaggt
gatunaggag atgggaggcg 303530DNAArtificial SequenceSynthetic
oligonucleotide 35ctcgtgaggt gatucaggag atgggaggcg
303630DNAArtificial SequenceSynthetic oligonucleotide 36ctcgtgaggt
gatgnaggag atgggaggcg 303730DNAArtificial SequenceSynthetic
oligonucleotide 37ctcgtgaggt gatucaggag atgggaggcg
303830DNAArtificial SequenceSynthetic oligonucleotide 38ctcgtgaggt
gattcaggag atgggaggcg 303930DNAArtificial SequenceSynthetic
oligonucleotide 39ctcgtgaggt gatguaggag atgggaggcg
304030DNAArtificial SequenceSynthetic oligonucleotide 40ctcgtgaggt
gattcaggag atgggaggcg 304130DNAArtificial SequenceSynthetic
oligonucleotide 41ctcgtgaggt gatucaggag atgggaggcg
304230DNAArtificial SequenceSynthetic oligonucleotide 42ctcgtgaggt
gattcaggag atgggaggcg 304330DNAArtificial SequenceSynthetic
oligonucleotide 43ctcgtgaggt gatgnaggag atgggaggcg
304430DNAArtificial SequenceSynthetic oligonucleotide 44ctcgtgaggt
gatucaggag atgggaggcg 304530DNAArtificial SequenceSynthetic
oligonucleotide 45ctcgtgaggt gatucaggag atgggaggcg
304630DNAArtificial SequenceSynthetic oligonucleotide 46ctcgtgaggt
gatgaaggag atgggaggcg 304730DNAArtificial SequenceSynthetic
oligonucleotide 47ctcgtgaggt gatgacggag atgggaggcg
304830DNAArtificial SequenceSynthetic oligonucleotide 48ctcgtgaggt
gatgagggag atgggaggcg 304930DNAArtificial SequenceSynthetic
oligonucleotide 49ctcgtgaggt gatgauggag atgggaggcg
305030DNAArtificial SequenceSynthetic oligonucleotide 50ctcgtgaggt
gatgcaggag atgggaggcg 305130DNAArtificial SequenceSynthetic
oligonucleotide 51ctcgtgaggt gatgccggag atgggaggcg
305230DNAArtificial SequenceSynthetic oligonucleotide 52ctcgtgaggt
gatgcgggag atgggaggcg 305330DNAArtificial SequenceSynthetic
oligonucleotide 53ctcgtgaggt gatggaggag atgggaggcg
305430DNAArtificial SequenceSynthetic oligonucleotide 54ctcgtgaggt
gatggcggag atgggaggcg 305530DNAArtificial SequenceSynthetic
oligonucleotide 55ctcgtgaggt gatgggggag atgggaggcg
305630DNAArtificial SequenceSynthetic oligonucleotide 56ctcgtgaggt
gatgguggag atgggaggcg 305730DNAArtificial SequenceSynthetic
oligonucleotide 57ctcgtgaggt gatguaggag atgggaggcg
305830DNAArtificial SequenceSynthetic oligonucleotide 58ctcgtgaggt
gatgucggag atgggaggcg 305930DNAArtificial SequenceSynthetic
oligonucleotide 59ctcgtgaggt gatgugggag atgggaggcg
306030DNAArtificial SequenceSynthetic oligonucleotide 60ctcgtgaggt
gatguuggag atgggaggcg 306127DNAArtificial SequenceSynthetic
oligonucleotide 61ctgagcttca tgcctttact gtcctct 276228DNAArtificial
SequenceSynthetic oligonucleotide 62ctgagcttca tgcctttact gtcctctc
286329DNAArtificial SequenceSynthetic oligonucleotide 63ctgagcttca
tgcctttact gtcctctcc 296431DNAArtificial SequenceSynthetic
oligonucleotide 64ctgagcttca tgcctttact gtcctctcct t
316532DNAArtificial SequenceSynthetic oligonucleotide 65ctgagcttca
tgcctttact gtcctctcct tc 326627DNAArtificial SequenceSynthetic
oligonucleotide 66cucctgagct tcatgccttt actgtcc 276728DNAArtificial
SequenceSynthetic oligonucleotide 67ccucctgagc ttcatgcctt tactgtcc
286829DNAArtificial SequenceSynthetic oligonucleotide 68tccucctgag
cttcatgcct ttactgtcc 296930DNAArtificial SequenceSynthetic
oligonucleotide 69ttccucctga gcttcatgcc tttactgtcc
307031DNAArtificial SequenceSynthetic oligonucleotide 70cttccucctg
agcttcatgc ctttactgtc c 317132DNAArtificial SequenceSynthetic
oligonucleotide 71tcttccucct gagcttcatg cctttactgt cc
327234DNAArtificial SequenceSynthetic oligonucleotide 72tgtcttccuc
ctgagcttca tgcctttact gtcc 347336DNAArtificial SequenceSynthetic
oligonucleotide 73cctgtcttcc ucctgagctt catgccttta ctgtcc
367438DNAArtificial SequenceSynthetic oligonucleotide 74tacctgtctt
ccucctgagc ttcatgcctt tactgtcc 387540DNAArtificial
SequenceSynthetic oligonucleotide 75cttacctgtc ttccucctga
gcttcatgcc tttactgtcc 407626DNAArtificial SequenceSynthetic
oligonucleotide 76ctgagcttca tgcctttact gtuccc 267728DNAArtificial
SequenceSynthetic oligonucleotide 77ctgagcttca tgcctttact gtuccccg
287829DNAArtificial SequenceSynthetic oligonucleotide 78ctgagcttca
tgcctttact gtuccccga 297930DNAArtificial SequenceSynthetic
oligonucleotide 79ctgagcttca tgcctttact gtuccccgac
308032DNAArtificial SequenceSynthetic oligonucleotide 80ctgagcttca
tgcctttact gtuccccgac ac 328134DNAArtificial SequenceSynthetic
oligonucleotide 81ctgagcttca tgcctttact gtuccccgac acac
348236DNAArtificial SequenceSynthetic oligonucleotide 82ctgagcttca
tgcctttact gtuccccgac acacag 368338DNAArtificial SequenceSynthetic
oligonucleotide 83ctgagcttca tgcctttact gtuccccgac acacagct
388417DNAArtificial SequenceSynthetic primer 84caggaaacag ctatgac
178522DNAArtificial SequenceSynthetic primer 85caggaaacag
ctatgaccat ga 228623DNAArtificial SequenceSynthetic primer
86agctctgccc aaagattacc ctg 238722DNAArtificial SequenceSynthetic
primer 87ctgagcttca tgcctttact gt 228826DNAArtificial
SequenceSynthetic primer 88ctgagcttca tgcctttact gtuccc
268927DNAArtificial SequenceSynthetic primer 89ctgagcttca
tgcctttact gtucccc 279028DNAArtificial SequenceSynthetic primer
90ctgagcttca tgcctttact gtuccccc 289129DNAArtificial
SequenceSynthetic primer 91ctgagcttca tgcctttact gtuccccgc
299230DNAArtificial SequenceSynthetic primer 92ctgagcttca
tgcctttact gtuccccgac 3093103DNAArtificial SequenceSynthetic
polynucleotide 93agctctgccc aaagattacc ctgacagcta agtggcagtg
gaagttggcc tcagaagtag 60tggccagctg tgtgtcgggg aacagtaaag gcatgaagct
cag 1039416DNAArtificial SequenceSynthetic primer 94acctcggcca
agaccc 169522DNAArtificial SequenceSynthetic primer 95ccttccttcc
ttccttgctt cc 229621DNAArtificial SequenceSynthetic primer
96acctcggcca agacccggca g 219727DNAArtificial SequenceSynthetic
primer 97ccttccttcc ttccttgctt ccgtcct 2798320DNAArtificial
SequenceSynthetic polynucleotide 98acctcggcca agacccggca gggcagccgc
tctggctcta gctccagctc cgggaccctc 60tgggaccccc cgggacccat gtgacccagc
ggcccctcgc gctggagtgg aggatgcctt 120ctacacgttg gtgcgtgaga
tccggcagca caagctgcgg aagctgaacc ctcctgatga 180gagtggcccc
ggctgcatga gctgcaagtg tgtgctctcc tgacgcagca caagctcagg
240acatggaggt gccggatgca ggaaggaggt gcagacggaa ggaggaggaa
ggaaggacgg 300aagcaaggaa ggaaggaagg 3209924DNAArtificial
SequenceSynthetic primer 99ccctgtttgc tgtttttcct tctc
2410021DNAArtificial SequenceSynthetic primer 100cgccgctgtt
cctttttgaa g 2110129DNAArtificial SequenceSynthetic primer
101ccctgtttgc tgtttttcct tctcuaaat 2910226DNAArtificial
SequenceSynthetic primer 102cgccgctgtt cctttttgaa gccact
26103180DNAArtificial SequenceSynthetic polynucleotide
103ccctgtttgc tgtttttcct tctctaaatg aagagcaaac actgcaagaa
gtgccaacag 60gcttggattc catttctcat gactccgcca actgtgaatt gcctttgtta
accccgtgca 120gcaaggctgt gatgagtcaa gccttaaaag ctaccttcag
tggcttcaaa aaggaacagc 18010423DNAArtificial SequenceSynthetic
primer 104ctgagcttca tgcctttact gtu 2310534DNAArtificial
SequenceSynthetic primer 105ctgagcttca tgcctttact gtuccccgac acac
3410624DNAArtificial SequenceSynthetic primer 106ccctgtttgc
tgtttttcct tctc 2410729DNAArtificial SequenceSynthetic primer
107ccctgtttgc tgtttttcct tctcuaaat 2910821DNAArtificial
SequenceSynthetic primer 108cgccgctgtt cctttttgaa g
2110926DNAArtificial SequenceSynthetic primer 109cgccgctgtt
cctttttgaa gccact 2611016DNAArtificial SequenceSynthetic primer
110acctcggcca agaccc 1611121DNAArtificial SequenceSynthetic primer
111acctcggcca agacccggca g 2111222DNAArtificial SequenceSynthetic
primer 112ccttccttcc ttccttgctt cc 2211327DNAArtificial
SequenceSynthetic primer 113ccttccttcc ttccttgctt ccgtcct
2711422DNAArtificial SequenceSynthetic primer 114gcatttcttc
catctccccc tc 2211527DNAArtificial SequenceSynthetic primer
115gcatttcttc catctccccc tcugcct 2711622DNAArtificial
SequenceSynthetic primer 116tccgattctt gctccactgt tg
2211727DNAArtificial SequenceSynthetic primer 117tccgattctt
gctccactgt tggctga 2711830DNAArtificial SequenceSynthetic
oligonucleotide 118cgcctcccat attccacatc acctcacgag
3011930DNAArtificial SequenceSynthetic oligonucleotide
119cgcctcccat ctcctacatc acctcacgag 3012027DNAArtificial
SequenceSynthetic oligonucleotide 120cgcctcccat ctccatcacc tcacgag
2712130DNAArtificial SequenceSynthetic oligonucleotide
121cgcctcccat ctcctgaatc acctcacgag 3012230DNAArtificial
SequenceSynthetic oligonucleotide 122cgcctcccat ctcctgtatc
acctcacgag 3012330DNAArtificial SequenceSynthetic oligonucleotide
123cgcctcccat ctcctggatc acctcacgag 3012430DNAArtificial
SequenceSynthetic oligonucleotide 124cgcctcccat ctcccgcatc
acctcacgag 3012530DNAArtificial SequenceSynthetic oligonucleotide
125cgcctcccat ctcccgcatc acctcacgag 3012630DNAArtificial
SequenceSynthetic oligonucleotide 126cgcctcccat ctccggcatc
acctcacgag 3012727DNAArtificial SequenceSynthetic primer
127ctgagcttca tgcctttact gtacccc 2712827DNAArtificial
SequenceSynthetic primer 128ctgagcttca tgcctttact gaacccc
2712927DNAArtificial SequenceSynthetic primer 129ctgagcttca
tgcctttact gcacccc 2713027DNAArtificial SequenceSynthetic primer
130ctgagcttca tgcctttact ggacccc 2713127DNAArtificial
SequenceSynthetic primer 131ctgagcttca tgcctttact gtatccc
2713227DNAArtificial SequenceSynthetic primer 132ctgagcttca
tgcctttact gtagccc 2713327DNAArtificial SequenceSynthetic primer
133ctgagcttca tgcctttact gtaaccc 2713427DNAArtificial
SequenceSynthetic primer 134ctgagcttca tgcctttact gtucccc
2713527DNAArtificial SequenceSynthetic primer 135ctgagcttca
tgcctttact gaucccc 2713627DNAArtificial SequenceSynthetic primer
136ctgagcttca tgcctttact gcucccc 2713727DNAArtificial
SequenceSynthetic primer 137ctgagcttca tgcctttact ggucccc
2713827DNAArtificial SequenceSynthetic primer 138ctgagcttca
tgcctttact gtutccc 2713927DNAArtificial SequenceSynthetic primer
139ctgagcttca tgcctttact gtugccc 2714027DNAArtificial
SequenceSynthetic primer 140ctgagcttca tgcctttact gtuaccc
2714127DNAArtificial SequenceSynthetic primer 141ctgagcttca
tgcctttact gtccccc 2714227DNAArtificial SequenceSynthetic primer
142ctgagcttca tgcctttact gaccccc 2714327DNAArtificial
SequenceSynthetic primer 143ctgagcttca tgcctttact gcccccc
2714427DNAArtificial SequenceSynthetic primer 144ctgagcttca
tgcctttact ggccccc 2714527DNAArtificial SequenceSynthetic primer
145ctgagcttca tgcctttact gtctccc 2714627DNAArtificial
SequenceSynthetic primer 146ctgagcttca tgcctttact gtcgccc
2714727DNAArtificial SequenceSynthetic primer 147ctgagcttca
tgcctttact gtcaccc 2714827DNAArtificial SequenceSynthetic primer
148ctgagcttca tgcctttact gtgcccc 2714927DNAArtificial
SequenceSynthetic primer 149ctgagcttca tgcctttact gagcccc
2715027DNAArtificial SequenceSynthetic primer 150ctgagcttca
tgcctttact gcgcccc 2715127DNAArtificial SequenceSynthetic primer
151ctgagcttca tgcctttact gggcccc 2715227DNAArtificial
SequenceSynthetic primer 152ctgagcttca tgcctttact gtgtccc
2715327DNAArtificial SequenceSynthetic oligonucleotide
153ctgagcttca tgcctttact gtggccc 2715427DNAArtificial
SequenceSynthetic primer 154ctgagcttca tgcctttact gtgaccc
27155103DNAArtificial SequenceSynthetic polynucleotide
155agctctgccc aaagattacc ctgacagcta agtggcagtg gaagttggcc
tcagaagtag 60tggccagctg tgtgtcgggg tacagtaaag gcatgaagct cag
103156103DNAArtificial SequenceSynthetic polynucleotide
156agctctgccc aaagattacc ctgacagcta agtggcagtg gaagttggcc
tcagaagtag 60tggccagctg tgtgtcgggg tccagtaaag gcatgaagct cag
103157103DNAArtificial SequenceSynthetic polynucleotide
157agctctgccc aaagattacc ctgacagcta agtggcagtg gaagttggcc
tcagaagtag 60tggccagctg tgtgtcgggg ttcagtaaag gcatgaagct cag
103158103DNAArtificial SequenceSynthetic polynucleotide
158agctctgccc aaagattacc ctgacagcta agtggcagtg gaagttggcc
tcagaagtag 60tggccagctg tgtgtcgggg tgcagtaaag gcatgaagct cag
103159103DNAArtificial SequenceSynthetic polynucleotide
159agctctgccc aaagattacc ctgacagcta agtggcagtg gaagttggcc
tcagaagtag 60tggccagctg tgtgtcgggc tacagtaaag gcatgaagct cag
103160103DNAArtificial SequenceSynthetic polynucleotide
160agctctgccc aaagattacc ctgacagcta agtggcagtg gaagttggcc
tcagaagtag 60tggccagctg tgtgtcggga tacagtaaag gcatgaagct cag
103161103DNAArtificial SequenceSynthetic polynucleotide
161agctctgccc aaagattacc ctgacagcta agtggcagtg gaagttggcc
tcagaagtag 60tggccagctg tgtgtcgggt tacagtaaag gcatgaagct cag
103162103DNAArtificial SequenceSynthetic polynucleotide
162agctctgccc aaagattacc ctgacagcta agtggcagtg gaagttggcc
tcagaagtag 60tggccagctg tgtgtcgggg aacagtaaag gcatgaagct cag
103163103DNAArtificial SequenceSynthetic polynucleotide
163agctctgccc aaagattacc ctgacagcta agtggcagtg gaagttggcc
tcagaagtag 60tggccagctg tgtgtcgggg atcagtaaag gcatgaagct cag
103164103DNAArtificial SequenceSynthetic polynucleotide
164agctctgccc aaagattacc ctgacagcta agtggcagtg gaagttggcc
tcagaagtag 60tggccagctg tgtgtcgggg accagtaaag gcatgaagct cag
103165103DNAArtificial SequenceSynthetic polynucleotide
165agctctgccc aaagattacc ctgacagcta agtggcagtg gaagttggcc
tcagaagtag 60tggccagctg tgtgtcgggg agcagtaaag gcatgaagct cag
103166103DNAArtificial SequenceSynthetic polynucleotide
166agctctgccc aaagattacc ctgacagcta agtggcagtg gaagttggcc
tcagaagtag 60tggccagctg tgtgtcgggc aacagtaaag gcatgaagct cag
103167103DNAArtificial SequenceSynthetic polynucleotide
167agctctgccc aaagattacc ctgacagcta agtggcagtg gaagttggcc
tcagaagtag 60tggccagctg tgtgtcggga aacagtaaag gcatgaagct cag
103168103DNAArtificial SequenceSynthetic polynucleotide
168agctctgccc aaagattacc ctgacagcta agtggcagtg gaagttggcc
tcagaagtag 60tggccagctg tgtgtcgggt aacagtaaag gcatgaagct cag
103169103DNAArtificial SequenceSynthetic polynucleotide
169agctctgccc aaagattacc ctgacagcta agtggcagtg gaagttggcc
tcagaagtag 60tggccagctg tgtgtcgggg cacagtaaag gcatgaagct cag
103170103DNAArtificial SequenceSynthetic polynucleotide
170agctctgccc aaagattacc ctgacagcta agtggcagtg gaagttggcc
tcagaagtag 60tggccagctg tgtgtcgggg ctcagtaaag gcatgaagct cag
103171103DNAArtificial SequenceSynthetic polynucleotide
171agctctgccc aaagattacc ctgacagcta agtggcagtg gaagttggcc
tcagaagtag 60tggccagctg tgtgtcgggg cccagtaaag gcatgaagct cag
103172103DNAArtificial SequenceSynthetic polynucleotide
172agctctgccc aaagattacc ctgacagcta agtggcagtg gaagttggcc
tcagaagtag 60tggccagctg tgtgtcgggg cgcagtaaag gcatgaagct cag
103173103DNAArtificial SequenceSynthetic polynucleotide
173agctctgccc aaagattacc ctgacagcta agtggcagtg gaagttggcc
tcagaagtag 60tggccagctg tgtgtcggga cacagtaaag gcatgaagct cag
103174103DNAArtificial SequenceSynthetic polynucleotide
174agctctgccc aaagattacc ctgacagcta agtggcagtg gaagttggcc
tcagaagtag 60tggccagctg tgtgtcgggt cacagtaaag gcatgaagct cag
103175103DNAArtificial SequenceSynthetic polynucleotide
175agctctgccc aaagattacc ctgacagcta agtggcagtg gaagttggcc
tcagaagtag 60tggccagctg tgtgtcgggc cacagtaaag gcatgaagct cag
103176103DNAArtificial SequenceSynthetic polynucleotide
176agctctgccc aaagattacc ctgacagcta agtggcagtg gaagttggcc
tcagaagtag 60tggccagctg tgtgtcgggg gacagtaaag gcatgaagct cag
103177103DNAArtificial SequenceSynthetic polynucleotide
177agctctgccc aaagattacc ctgacagcta agtggcagtg gaagttggcc
tcagaagtag 60tggccagctg tgtgtcgggg gtcagtaaag gcatgaagct cag
103178103DNAArtificial SequenceSynthetic polynucleotide
178agctctgccc aaagattacc ctgacagcta agtggcagtg gaagttggcc
tcagaagtag 60tggccagctg tgtgtcgggg gccagtaaag gcatgaagct cag
103179103DNAArtificial SequenceSynthetic polynucleotide
179agctctgccc aaagattacc ctgacagcta agtggcagtg gaagttggcc
tcagaagtag 60tggccagctg tgtgtcgggg ggcagtaaag gcatgaagct cag
103180103DNAArtificial SequenceSynthetic polynucleotide
180agctctgccc aaagattacc ctgacagcta agtggcagtg gaagttggcc
tcagaagtag 60tggccagctg tgtgtcggga gacagtaaag gcatgaagct cag
103181103DNAArtificial SequenceSynthetic polynucleotide
181agctctgccc aaagattacc ctgacagcta agtggcagtg gaagttggcc
tcagaagtag 60tggccagctg tgtgtcgggt gacagtaaag gcatgaagct cag
103182103DNAArtificial SequenceSynthetic polynucleotide
182agctctgccc aaagattacc ctgacagcta agtggcagtg gaagttggcc
tcagaagtag 60tggccagctg tgtgtcgggc gacagtaaag gcatgaagct cag
10318330DNAArtificial SequenceSynthetic oligonucleotide
183cgcctcccat ctcctgaatc acctcacgag 3018430DNAArtificial
SequenceSynthetic oligonucleotide 184cgcctcccat ctcctcaatc
acctcacgag 3018530DNAArtificial SequenceSynthetic oligonucleotide
185cgcctcccat ctcctatatc acctcacgag 3018630DNAArtificial
SequenceSynthetic oligonucleotide 186cgcctcccat ctccttcatc
acctcacgag 3018730DNAArtificial SequenceSynthetic oligonucleotide
187cgcctcccat ctcctggatc ccatcacgag 3018830DNAArtificial
SequenceSynthetic oligonucleotide 188cgcctcccat ctcctccatc
ccatcacgag 3018930DNAArtificial SequenceSynthetic oligonucleotide
189cgcctcccat ctcctttatc acctcacgag 3019030DNAArtificial
SequenceSynthetic oligonucleotide 190cgcctcccat ctcctaactc
acctcacgag 3019130DNAArtificial SequenceSynthetic oligonucleotide
191cgcctcccat ctcctaagtc acctcacgag 3019227DNAArtificial
SequenceSynthetic oligonucleotide 192cgcctcccat ctaattcacc tcacgag
2719330DNAArtificial SequenceSynthetic oligonucleotide
193cgcctcccat ctccgaaatc acctcacgag 3019430DNAArtificial
SequenceSynthetic oligonucleotide 194cgcctcccat ctcccaaatc
acctcacgag 3019530DNAArtificial SequenceSynthetic oligonucleotide
195cgcctcccat ctccaaaatc ccatcacgag 3019630DNAArtificial
SequenceSynthetic oligonucleotide 196cgcctcccat ctcctgtatc
accactcgag 3019730DNAArtificial SequenceSynthetic oligonucleotide
197cgcctcccat ctcctgcatc accactcgag 3019830DNAArtificial
SequenceSynthetic oligonucleotide 198cgcctcccat ctcctggatc
ccatcacgag 3019930DNAArtificial SequenceSynthetic oligonucleotide
199cgcctcccat ctcctcacta acctcacgag 3020031DNAArtificial
SequenceSynthetic oligonucleotide 200cgcctcccat ctcctcacat
cacctcacga g 3120130DNAArtificial SequenceSynthetic oligonucleotide
201cgcctcccat ctcctaaatc ccatcacgag 3020230DNAArtificial
SequenceSynthetic oligonucleotide 202cgcctcccat ctcctctatc
acctcacgag 3020330DNAArtificial SequenceSynthetic oligonucleotide
203cgcctcccat ctcctgattc acctcacgag 3020430DNAArtificial
SequenceSynthetic oligonucleotide 204cgcctcccat ctcctgactc
acctcacgag 3020530DNAArtificial SequenceSynthetic oligonucleotide
205cgcctcccat ctcctgagtc acctcacgag 3020630DNAArtificial
SequenceSynthetic oligonucleotide 206cgcctcccat ctccagaatc
acctcacgag 3020730DNAArtificial SequenceSynthetic oligonucleotide
207cgcctcccat ctcccgaatc acctcacgag 3020830DNAArtificial
SequenceSynthetic oligonucleotide 208cgcctcccat ctccggaatc
acctcacgag 3020930DNAArtificial SequenceSynthetic oligonucleotide
209ctcgtgaggt gatucaggag atgggaggcg 3021030DNAArtificial
SequenceSynthetic oligonucleotide 210ctcgtgaggt gattcaggag
atgggaggcg 3021130DNAArtificial SequenceSynthetic oligonucleotide
211ctcgtgaggt gattcaggag atgggaggcg 3021227DNAArtificial
SequenceSynthetic probe 212ttctgaggcc aactccactg ccactta
2721327DNAArtificial SequenceSynthetic probe 213ttctgaggcc
aacuccactg ccactta 2721426DNAArtificial SequenceSynthetic primer
214gcagaaagcg tctagccatg gcgtta 2621528DNAArtificial
SequenceSynthetic primer 215gcaagcaccc tatcaggcag taccacaa
2821631DNAArtificial SequenceSynthetic primer 216gcagaaagcg
tctagccatg gcgttagtat g 3121733DNAArtificial SequenceSynthetic
primer 217gcaagcaccc tatcaggcag taccacaagg cct
33218242DNAArtificial SequenceSynthetic polynucleotide
218gcagaaagcg tctagccatg gcgttagtat gagtgtcgtg cagcctccag
gaccccccct 60cccgggagag ccatagtggt ctgcggaacc ggtgagtaca ccggaattgc
caggacgacc 120gggtcctttc ttggactaaa cccgctcaat gcctggagat
ttgggcgtgc ccccgcgaga 180ctgctagccg agtagtgttg ggtcgcgaaa
ggccttgtgg tactgcctga tagggtgctt 240gc
24221927DNAArtificial SequenceSynthetic oligonucleotide
219ttctgaggcc aactccactg ccactta 2722027DNAArtificial
SequenceSynthetic oligonucleotide 220ctgagcttca tgcctttact gtucccc
2722128DNAArtificial SequenceSynthetic oligonucleotide
221ctgagcttca tgcctttact gtuccccg 2822229DNAArtificial
SequenceSynthetic oligonucleotide 222ctgagcttca tgcctttact
gtuccccga 2922328DNAArtificial SequenceSynthetic oligonucleotide
223ctgagcttca tgcctttact gtuccccc 2822428DNAArtificial
SequenceSynthetic oligonucleotide 224ctgagcttca tgcctttact gtuccccc
28225103DNAArtificial SequenceSynthetic polynucleotide
225agctctgccc aaagattacc ctgacagcta agtggcagtg gaagttggcc
tcagaagtag 60tggccagctg tgtgtggggg aacagtaaag gcatgaagct cag
10322619DNAArtificial SequenceSynthetic primer 226tcggattctc
tgctctcct 1922720DNAArtificial SequenceSynthetic primer
227cctcatcttc ttgttcctcc 2022822DNAArtificial SequenceSynthetic
probe 228ccaccaccag cagcgactct ga 2222925DNAArtificial
SequenceSynthetic primer 229tcggattctc tgctctcctc gacgg
2523025DNAArtificial SequenceSynthetic primer 230cctcatcttc
ttgttcctcc ucaga 2523118DNAArtificial SequenceSynthetic primer
231tgtgcagaag gatggagt 1823220DNAArtificial SequenceSynthetic
primer 232ctggtgcttc tctcaggata 2023325DNAArtificial
SequenceSynthetic probe 233tggaatatgc cctgcgtaaa ctgga
2523424DNAArtificial SequenceSynthetic primer 234tgtgcagaag
gatggagtgg ggat 2423525DNAArtificial SequenceSynthetic primer
235ctggtgcttc tctcaggata aactc 2523621DNAArtificial
SequenceSynthetic primer 236ctcactctaa accccagcat t
2123727DNAArtificial SequenceSynthetic primer 237cagcctcatc
caaaagagga aacagga 2723827DNAArtificial SequenceSynthetic primer
238cagcctcatc caaaagagga aauagga 2723985DNAHomo sapiens
239cagcctcatc caaaagagga aacaggaccc cagagctccc tcagactcct
caggaaacac 60agacaatgct ggggtttaga gtgag 8524085DNAHomo sapiens
240cagcctcatc caaaagagga aataggaccc cagagctccc tcagactcct
caggaaacac 60agacaatgct ggggtttaga gtgag 8524120DNAArtificial
SequenceSynthetic primer 241accaacgaca agaccaagag
2024219DNAArtificial SequenceSynthetic primer 242tcgtggaaag
aagcagaca 1924322DNAArtificial SequenceSynthetic probe
243accaagacct tggcggacct tt 2224432DNAArtificial SequenceSynthetic
primer 244tttccuggtt ttaccaacga caagaccaag ag 32245141DNAHomo
sapiens 245accaacgaca agaccaagag gcctgtggcg cttcgcacca agaccttggc
ggaccttttg 60gaatcattta ttgcagcgct gtacattgat aaggatttgg aatatgttca
tactttcatg 120aatgtctgct tctttccacg a 14124622DNAArtificial
SequenceSynthetic primer 246ctgagcttca tgcctttact gu
2224727DNAArtificial SequenceSynthetic primer 247ctgagcttca
tgcctttact guucccc 2724827DNAArtificial SequenceSynthetic primer
248ctgagcttca tgcctttact gtucccc 2724922DNAArtificial
SequenceSynthetic primer 249cagcctcatc caaaagagga aa
2225027DNAArtificial SequenceSynthetic primer 250cagcctcatc
caaaagagga aacagga 2725127DNAArtificial SequenceSynthetic primer
251cagcctcatc caaaagagga aacaaga 2725227DNAArtificial
SequenceSynthetic primer 252cagcctcatc caaaagagga aacacga
2725327DNAArtificial SequenceSynthetic primer 253cagcctcatc
caaaagagga aacatga 2725427DNAArtificial SequenceSynthetic primer
254cagcctcatc caaaagagga aauagga 2725527DNAArtificial
SequenceSynthetic primer 255cagcctcatc caaaagagga aauaaga
2725627DNAArtificial SequenceSynthetic primer 256cagcctcatc
caaaagagga aauacga 2725727DNAArtificial SequenceSynthetic primer
257cagcctcatc caaaagagga aauatga 2725827DNAArtificial
SequenceSynthetic primer 258ctgagcttca tgcctttact gtucccc
2725927DNAArtificial SequenceSynthetic primer 259ctgagcttca
tgcctttact gtucccc 2726027DNAArtificial SequenceSynthetic primer
260ctgagcttca tgcctttact gtucccc 2726128DNAArtificial
SequenceSynthetic primer 261agctctgccc aaagattacc ctgacagc
2826228DNAArtificial SequenceSynthetic primer 262agctctgccc
aaagattacc ctgacagc 2826328DNAArtificial SequenceSynthetic primer
263agctctgccc aaagattacc ctgacagc 2826428DNAArtificial
SequenceSynthetic primer 264agctctgccc aaagattacc ctgacagc
2826528DNAArtificial SequenceSynthetic primer 265agctctgccc
aaagattacc ctgacagc 2826627DNAArtificial SequenceSynthetic primer
266cagcctcatc caaaagagga aacagga 2726727DNAArtificial
SequenceSynthetic oligonucleotide 267cagcctcatc caaaagagga aacagga
2726827DNAArtificial SequenceSynthetic primer 268cagcctcatc
caaaagagga aacagga 2726927DNAArtificial SequenceSynthetic primer
269cagcctcatc caaaagagga aacagga 27270103DNAArtificial
SequenceSynthetic polynucleotide 270ctgagcttca tgcctttact
gttccccgac acacagctgg ccactacttc tgaggccaac 60ttccactgcc acttagctgt
cagggtaatc tttgggcaga gct 10327124DNAArtificial SequenceSynthetic
primer 271cagcctcatc caaaagagga aaca 242729DNAArtificial
SequenceSynthetic oligonucleotide 272cagctgaag 92739DNAArtificial
SequenceSynthetic oligonucleotide 273gagctgaag 92749DNAArtificial
SequenceSynthetic oligonucleotide 274aagctgaag 92759DNAArtificial
SequenceSynthetic oligonucleotide 275tagctgaag 927624DNAArtificial
SequenceSynthetic oligonucleotide 276ccctgtttgc tgtttttcct tctc
2427743DNAArtificial SequenceSynthetic oligonucleotide
277agtgtttgct cttcagctag agaaggaaaa acagcaaaca ggg
4327843DNAArtificial SequenceSynthetic oligonucleotide
278agtgtttgct cttcagcttg agaaggaaaa acagcaaaca ggg
432798DNAArtificial SequenceSynthetic oligonucleotide 279aagctnnn
82808DNAArtificial SequenceSynthetic oligonucleotide 280aagcnnnn
82818DNAArtificial SequenceSynthetic oligonucleotide 281annnnnnn
82828DNAArtificial SequenceSynthetic oligonucleotide 282tnnnnnnn
82838DNAArtificial SequenceSynthetic oligonucleotide 283gnnnnnnn
828424DNAArtificial SequenceSynthetic oligonucleotide 284gatttgtatt
gattgagatt aaag 2428524DNAArtificial SequenceSynthetic
oligonucleotide 285gattgtaaga tttgataaag tgta 2428638DNAArtificial
SequenceSynthetic oligonucleotide 286caccatgctc acagcctcat
ccaaaagagg aaacagga 3828738DNAArtificial SequenceSynthetic
oligonucleotide 287caccatgctc acagcctcat ccaaaagagg aaauagga
3828836DNAArtificial SequenceSynthetic oligonucleotide
288caggacccca gactttaatc tcaatcaata caaatc 3628936DNAArtificial
SequenceSynthetic oligonucleotide 289taggacccca gatacacttt
atcaaatctt acaatc 3629098DNAArtificial SequenceSynthetic
oligonucleotide 290cccagcattg tctgtgtttc ctgaggagtc tgagggagct
ctggggtcct gtttcctctt 60ttggatgagg ctgtgagcat ggtggattag agacagcc
9829198DNAArtificial SequenceSynthetic oligonucleotide
291cccagcattg tctgtgtttc ctgaggagtc tgagggagct ctggggtcct
atttcctctt 60ttggatgagg ctgtgagcat ggtggattag agacagcc
982926PRTArtificial SequenceSynthetic 6xHis tag 292His His His His
His His 1 5 29330DNAArtificial SequenceSynthetic oligonucleotide
293cgcctcccat ctcctgcatc acctcacgag 3029430DNAArtificial
SequenceSynthetic oligonucleotide 294cgcctcccat ctcctacatc
acctcacgag 3029530DNAArtificial SequenceSynthetic oligonucleotide
295cgcctcccat ctccagcatc acctcacgag 3029615DNAArtificial
SequenceSynthetic oligonucleotide 296ctcgtgaggt gatgc
1529715DNAArtificial SequenceSynthetic oligonucleotide
297uggagatggg aggcg 1529830DNAArtificial SequenceSynthetic
oligonucleotide 298cgcctcccat ctcctnnatc acctcacgag
3029930DNAArtificial SequenceSynthetic oligonucleotide
299cgcctcccat ctcctgaatc acctcacgag 3030030DNAArtificial
SequenceSynthetic oligonucleotide 300cgcctcccat ctcctnaatc
acctcacgag 3030130DNAArtificial SequenceSynthetic oligonucleotide
301cgcctcccat ctcctncatc acctcacgag 3030227DNAArtificial
SequenceSynthetic oligonucleotide 302acaggacagt aaaggcatga agctcag
2730328DNAArtificial SequenceSynthetic oligonucleotide
303gacaggacag taaaggcatg aagctcag 2830429DNAArtificial
SequenceSynthetic oligonucleotide 304ggacaggaca gtaaaggcat
gaagctcag 2930531DNAArtificial SequenceSynthetic oligonucleotide
305aaggacagga cagtaaaggc atgaagctca g 3130629DNAArtificial
SequenceSynthetic oligonucleotide 306gaaggacaga gtaaaggcat
gaagctcag 2930750DNAArtificial SequenceSynthetic oligonucleotide
307atgcaggaca gtaaaggcat gaagctcagg aggaagacag gtaagatgca
5030839DNAArtificial SequenceSynthetic oligonucleotide
308gagctgtgtg tcggggaaca gtaaaggcat gaagctcag 3930921DNAArtificial
SequenceSynthetic primer 309ctgagcttca tgcctttact g
21310105DNAArtificial SequenceSynthetic polynucleotide
310agctctgccc aaagattacc ctgacagcta agtggcagtg gaagttggcc
tcagaagtag 60tggccagctg tgtgtgtcgg ggaacagtaa aggcatgaag ctcag
10531127DNAArtificial SequenceSynthetic primer 311ctgagcttca
tgcctttact gtncccc 2731227DNAArtificial SequenceSynthetic
oligonucleotide 312ggggnacagt aaaggcatga agctcag
2731327DNAArtificial SequenceSynthetic primer 313ctgagcttca
tgcctttact gtncccc 2731427DNAArtificial SequenceSynthetic
oligonucleotide 314ggggancagt aaaggcatga agctcag
2731527DNAArtificial SequenceSynthetic primer 315ctgagcttca
tgcctttact gtccccc 2731627DNAArtificial SequenceSynthetic
oligonucleotide 316gggnaacagt aaaggcatga agctcag
2731731DNAArtificial SequenceSynthetic oligonucleotide
317cgcctcccat ctcctgacat cacctcacga g 3131830DNAArtificial
SequenceSynthetic oligonucleotide 318cgcctcccat ctcctgaatc
acctcacgag 3031936DNAArtificial SequenceSynthetic oligonucleotide
319taggacccca gatacacttt atcaaatctt acaatc 3632027DNAArtificial
SequenceSynthetic primer 320ctgagcttca tgcctttact gtucccc
2732130DNAArtificial SequenceSynthetic oligonucleotide
321ctcgtgaggt gattcaggag atgggaggcg 3032233DNAArtificial
SequenceSynthetic oligonucleotide 322ctggggtcct rtttcctctt
ttggatgagg ctg 3332311DNAArtificial SequenceSynthetic
oliogonucleotide 323actggacctg a 1132433DNAArtificial
SequenceSynthetic oligonucleotide 324ccctgtttgc tgtttttcct
tctctagctg aag 3332525DNAArtificial SequenceSynthetic
oligonucleotide 325ccctgtttgc tgtttttcct tctct 2532632DNAArtificial
SequenceSynthetic oligonucleotide 326ccctgtttgc tgtttttcct
tctcaagctn nn 3232732DNAArtificial SequenceSynthetic
oligonucleotide 327ccctgtttgc tgtttttcct tctcaagcnn nn
3232866DNAArtificial SequenceSynthetic oligonucleotide
328gaggagtctg agggagctct ggggtcctat ttcctctttt ggatgaggct
gtgagcatgg 60tggatt 6632933DNAArtificial SequenceSynthetic
oligonucleotide 329caccatgctc acagcctcat ccaaaagagg aaa
3333069DNAArtificial SequenceSynthetic oligonucleotide
330caccatgctc acagcctcat ccaaaagagg aaataggacc ccagatacac
tttatcaaat 60cttacaatc 6933124DNAArtificial SequenceSynthetic
oligonucleotide 331gattgtaaga tttgataaag tgta 2433230DNAArtificial
SequenceSynthetic 332cgcctcccat ctccttcatc acctcacgag
3033330DNAArtificial SequenceSynthetic 333cgcctcccat ctccgtcatc
acctcacgag 3033430DNAArtificial SequenceSynthetic 334cgcctcccat
ctccctcatc acctcacgag 3033530DNAArtificial SequenceSynthetic
335cgcctcccat ctcctgcatc acctcacgag 3033630DNAArtificial
SequenceSynthetic 336cgcctcccat ctccggcatc acctcacgag
3033730DNAArtificial SequenceSynthetic 337cgcctcccat ctccggcatc
acctcacgag 3033830DNAArtificial SequenceSynthetic 338cgcctcccat
ctcccgcatc acctcacgag 3033930DNAArtificial SequenceSynthetic
339cgcctcccat ctcctccatc acctcacgag 3034030DNAArtificial
SequenceSynthetic 340cgcctcccat ctccgccatc acctcacgag
3034130DNAArtificial SequenceSynthetic 341cgcctcccat ctcccccatc
acctcacgag 3034230DNAArtificial SequenceSynthetic 342cgcctcccat
ctccaccatc acctcacgag 3034330DNAArtificial SequenceSynthetic
343cgcctcccat ctcctacatc acctcacgag 3034430DNAArtificial
SequenceSynthetic 344cgcctcccat ctccgacatc acctcacgag
3034530DNAArtificial SequenceSynthetic 345cgcctcccat ctcccacatc
acctcacgag 3034630DNAArtificial SequenceSynthetic 346cgcctcccat
ctccaacatc acctcacgag 3034730DNAArtificial SequenceSynthetic
347cgcctcccat ctcctgaatc acctcacgag 3034830DNAArtificial
SequenceSynthetic 348cgcctcccat ctcctgaatc acctcacgag 30
* * * * *