U.S. patent application number 09/945681 was filed with the patent office on 2002-05-30 for compositions and methods for recombinant adeno-associated virus production.
Invention is credited to Chadeuf, Gilliane, Moullier, Philippe, Nony, Pascale, Salvetti, Anna.
Application Number | 20020064878 09/945681 |
Document ID | / |
Family ID | 23000350 |
Filed Date | 2002-05-30 |
United States Patent
Application |
20020064878 |
Kind Code |
A1 |
Salvetti, Anna ; et
al. |
May 30, 2002 |
Compositions and methods for recombinant adeno-associated virus
production
Abstract
The present invention relates to methods and compositions for
the production of recombinant Adeno-Associated Viruses (rAAV). In
particular, the invention discloses nucleic acid constructs and
packaging cells having improved properties for rAAV production, as
well as novel methods of titration and characterization of rAAV
preparations. The invention also describes novel sequences which
promote or increase the packaging of nucleic acids in rAAV, and
their use for producing rAAV with high efficiency. The invention
can be used for producing or testing high quality rAAV
preparations, for biological, preclinical, clinical or
pharmaceutical uses.
Inventors: |
Salvetti, Anna; (Nantes,
FR) ; Nony, Pascale; (Nantes, FR) ; Chadeuf,
Gilliane; (Nantes, FR) ; Moullier, Philippe;
(Basse Goulaine, FR) |
Correspondence
Address: |
YOUNG & THOMPSON
745 SOUTH 23RD STREET 2ND FLOOR
ARLINGTON
VA
22202
|
Family ID: |
23000350 |
Appl. No.: |
09/945681 |
Filed: |
September 5, 2001 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
09945681 |
Sep 5, 2001 |
|
|
|
PCT/EP00/01854 |
Mar 3, 2000 |
|
|
|
PCT/EP00/01854 |
Mar 3, 2000 |
|
|
|
09263093 |
Mar 5, 1999 |
|
|
|
Current U.S.
Class: |
435/456 ;
435/235.1; 536/23.72 |
Current CPC
Class: |
C12N 15/86 20130101;
C12N 2750/14143 20130101 |
Class at
Publication: |
435/456 ;
435/235.1; 536/23.72 |
International
Class: |
C12N 015/861; C07H
021/04; C12N 007/01 |
Claims
1. An isolated Replication Encapsidation Sequence ("RES"), which
provides or promotes the packaging of a nucleic acid operably
linked thereto into an Adeno-Associated Virus (AAV) particle, and
which consists of the fragment from nucleotide position 190 to
nucleotide position 540 of an AAV genome of SEQ ID No: 10.
2. An isolated Replication Encapsidation Sequence ("RES"), which is
a fragment of the RES of claim I and which provides or promotes the
packaging of a nucleic acid operably linked thereto into an
Adeno-Associated Virus (AAV) particle.
3. The isolated Replication Encapsidation Sequence ("RES") of claim
2, which consists of the fragment from nucleotide position 190 to
nucleotide position 350 of an Adeno-Associated Virus (AAV) genome
of SEQ ID No: 10.
4. The isolated Replication Encapsidation Sequence ("RES") of claim
2, which consists of the fragment from nucleotide position 230 to
nucleotide position 350 of an Adeno-Associated Virus (AAV) genome
of SEQ ID No: 10.
5. The isolated Replication Encapsidation Sequence ("RES") of claim
2, which consists of the fragment from nucleotide position 230 to
nucleotide position 300 of an Adeno-Associated Virus (AAV) genome
of SEQ ID No: 10.
6. The isolated Replication Encapsidation Sequence ("RES") of claim
2, which consists of the fragment from nucleotide position 250 to
nucleotide position 300 of an Adeno-Associated Virus (AAV) genome
of SEQ ID No: 10.
7. An isolated Replication Encapsidation Sequence ("RES"), which is
a functional variant of the RES of claim 1, and which differs from
the fragment of the Adeno-Associated Virus (AAV) genome by 1, 2 or
3 mutation(s) or deletion(s).
8. An isolated Replication Encapsidation Sequence ("RES"), which is
a functional variant of the RES of claim 2, and which differs from
the fragment of the Adeno-Associated Virus (AAV) genome by 1, 2 or
3 mutation(s) or deletion(s).
9. The Replication Encapsidation Sequence ("RES") according to any
of claims 1 to 8, wherein said RES comprises a Rep Binding Element
and a terminal resolution site.
10. The Replication Encapsidation Sequence ("RES") of claim 9,
wherein said RES further comprises a palindromic region.
11. The Replication Encapsidation Sequence ("RES") according to any
of claims 1 to 8, which comprises SEQ ID NO: 1 or a functional
variant thereof, wherein said functional variant binds to Rep78
and/or Rep68.
12. The Replication Encapsidation Sequence ("RES") according to any
of claims 1 to 8, which comprises SEQ ID NO: 3 or a functional
variant thereof, wherein said functional variant binds to Rep78
and/or Rep68.
13. The Replication Encapsidation Sequence ("RES") according to any
of claims 1 to 8, which comprises SEQ ID NO: 2 or a functional
variant thereof, wherein said functional variant forms a hairpin
structure.
14. The isolated Replication Encapsidation Sequence ("RES") of any
one of claims 1 to 8, which consists of less than 400 bp.
15. A nucleic acid consisting of one or several Replication
Encapsidation Sequence(s) ("RES") according to any one of claims 1
to 8, operably linked to a heterologous polynucleotide lacking a
functional ITR sequence.
16. A recombinant Adeno-Associated Virus (rAAV) vector plasmid,
comprising a recombinant AAV genome and one or several Replication
Encapsidation Sequence(s) ("RES") according to any one of claims 1
to 8.
17. The recombinant Adeno-Associated Virus (rAAV) vector plasmid of
claim 16, comprising a Replication Encapsidation Sequence ("RES")
in antisense orientation.
18. A composition comprising a recombinant Adeno-Associated Virus
(rAAV) particle comprising a recombinant nucleic acid, wherein said
recombinant nucleic acid comprises one or several Replication
Encapsidation Sequence(s) ("RES") according to any one of claims 1
to 8.
19. The composition of claim 18, wherein said recombinant nucleic
acid is a recombinant Adeno-Associated Virus genome comprising a
Replication Encapsidation Sequence ("RES") in antisense
orientation.
20. A method of producing recombinant Adeno-Associated Viruses
(rAAVs) comprising: (i) co-transfecting a cell culture with: a rAAV
vector plasmid, a rep-cap plasmid devoid of ITR, containing a
rep-cap unit consisting of residues 190-4484 of the AAV genome or
fragments thereof encoding functional Rep and Cap proteins, and an
adenovirus plasmid containing the entire adenoviral genome or a
genome lacking the left and right ITRs and the packaging region,
and (ii) recovering the rAAV produced, wherein the rAAV vector
plasmid comprises a Replication Encapsidation Sequence ("RES")
according to any one of claims 1 to 8, in antisense
orientation.
21. The method of claim 20, wherein said adenovirus plasmid lacks
the E1 region.
22. A method of producing recombinant Adeno-Associated Viruses
(rAAVs) comprising: (i) co-transfecting a culture of cells which
contain, in their genome, nucleic acid construct(s) encoding the
Rep and/or Cap functions, with: a rAAV vector plasmid, and a helper
adenovirus or an adenovirus plasmid containing the entire
adenoviral genome or a genome lacking the left and right ITRs and
the packaging region, and (ii) recovering the rAAV produced,
wherein the rAAV vector plasmid comprises a Replication
Encapsidation Sequence ("RES") according to any one of claims 1 to
8, in antisense orientation.
23. The method of claim 22, wherein said helper adenovirus or
adenovirus plasmid lacks the E1 region.
Description
[0001] This a continuation of international application No.
PCT/EP00/01854, filed Mar. 3, 2000, which designated the United
States of America, and a continuation-in-part of copending
application Ser. No. 09/263,093, filed Mar. 5, 1999.
[0002] The present invention relates to methods and compositions
for the production of recombinant Adeno-Associated Viruses (rAAV).
In particular, the invention discloses nucleic acid constructs and
packaging cells having improved properties for rAAV production, as
well as novel methods of titration and characterization of rAAV
preparations. The invention also describes novel sequences which
promote or increase the packaging of nucleic acids in rAAV, and
their use for producing rAAV with high efficiency. The invention
can be used for producing or testing high quality rAAV
preparations, for biological, preclinical, clinical or
pharmaceutical uses.
BACKGROUND AND INTRODUCTION
[0003] Wild type Adeno-Associated Virus (wtAAV) is a naturally
defective parvovirus which requires co-infection with a helper
virus, such as adenovirus or herpes virus, in order to establish a
productive infection. The virus is not associated with any human
disease and has been shown to have a broad host range of infection
in vitro. AAV has a relatively simple genome organization composed
of two major genes coding for the regulatory (Rep) and structural
(Cap) proteins. Three viral promoters located at map unit 5 (p5),
19 (p19) and 40 (p40) control the synthesis of mRNA coding for the
four Rep and the three Cap proteins. The viral genome is flanked by
145 bases of inverted terminal repeats (ITRs) which contain
palindromic sequences necessary in cis for replication of the viral
genome (Leonard and Berns, 1994).
[0004] Recombinant AAV viruses (rAAV) are derived by deleting the
rep and cap genes which are replaced by the transgene and the
transcriptional control elements needed for its expression. The
only viral sequences retained in cis are the viral ITRs (Muzyczka,
1992). The ability of rAAV to efficiently transduce tissues in mice
such as the muscle, the retina or the liver (Fisher et al., 1997;
Flannery et al., 1997; Kessler et al., 1996; Koeberl et al., 1997;
Snyder et al., Herzog et al., 1997; 1997; Xiao et al., 1996;
Zolotukhin et al., 1996) and to lead to a prolonged gene expression
with little to no pathology makes this virus unique among the
family of viral vectors. In this respect, rAAV can be used in vitro
for recombinant protein production, gene regulation studies, AAV
protein production (to be used in non-viral gene delivery systems),
etc. rAAV can also be used ex vivo or in vivo, to deliver genes of
interest for biological, toxicological, prophylactic or therapeutic
indications, production (to be used in non-viral gene delivery
systems), etc. rAAV can also be used ex vivo or in vivo, to deliver
genes of interest for biological, toxicological, prophylactic or
therapeutic indications, for instance. In this regard, it should be
noted that several clinical trials are currently ongoing using a
rAAV gene delivery vector.
[0005] However, widespread use of rAAV is hampered by the
relatively cumbersome and inefficient procedure needed to produce
it at high titers and in sufficient amount and quality. The
standard procedure relies upon the transfection of 293 cells with
two plasmids: a plasmid providing in trans the Rep and Cap
functions, and the rAAV vector plasmid itself. After subsequent
infection with an adenovirus, rAAV particles are assembled in the
nuclei of the cells concomitantly with adenoviral particles. rAAV
stocks are obtained after purification from total cell lysates
through CsCl gradients (Snyder et al., 1996).
[0006] However, these methods represent relatively long and complex
procedures, which cannot be easily scaled-up. Furthermore, because
of the number of constructs required, recombination events have
been observed leading to rAAV preparations which are contaminated
with replicating AAV particles and with adenoviruses. There is
therefore a need for improved methods of producing rAAV, for
biological, preclinical, clinical or pharmaceutical uses. In
particular, there is a need for methods of producing rAAV
preparations with high titers of infectious particles and which are
essentially free of adenoviruses. There is also a need for methods
to produce rAAV preparations with significantly reduced
contamination by replication competent or recombined AAV. There is
also a need for improved methods of titration of rAAV preparations,
and for methods of characterization of such preparations, i.e., for
use in Quality Control steps, as well as for detecting rAAV or
contaminants in biological fluids, for instance.
[0007] The present invention now provides novel methods and
compositions for producing and characterizing rAAV preparations. In
particular, the invention provides methods of producing rAAV with
very high yields of infectious particles and essentially free of
detectable adenovirus contamination.
DESCRIPTION OF THE INVENTION
[0008] The present invention relates to compositions and methods
for producing and/or characterizing rAAV preparations of improved
quality. The invention relates more particularly to improved
nucleic acid constructs that provide for an efficient production of
rAAVs, as well as to compositions, cell lines and methods for
characterizing rAAV preparations. The invention can be used to
produce rAAV for biological, preclinical or clinical uses, e.g.,
pharmaceutically acceptable rAAV preparations.
[0009] Within the context of the present invention, a recombinant
Adeno-Associated Virus (rAAV) designates an AAV virus which
comprises at least a recombinant nucleic acid genome. More
specifically, rAAVs generally comprise a recombinant genome lacking
a functional rep and/or cap region, and comprising a heterologous
nucleic acid. Most conventional rAAVs comprise a recombinant genome
lacking the entire Rep and Cap regions, which are replaced by the
heterologous nucleic acid. Such recombinant genomes thus usually
comprise the heterologous nucleic acid flanked by the left and
right Inverted Terminal Repeats of AAV. rAAVs may comprise
additional modifications, such as artificial or heterologous capsid
proteins, for instance. Furthermore, the recombinant genome may
comprise, in replacement or in addition to the ITR sequence, RES
elements as disclosed in the present invention. rAAVs may be
derived from different serotypes of AAV, such as for instance
AAV-2, AAV-3, AAV-4 or AAV-6.
[0010] In the present invention, except otherwise indicated, all
references to nucleotide positions of the AAV genome are made with
respect to the sequence of wild-type AAV-2 available at Genebank
under number AF043303.
[0011] A recombinant Adeno-Associated Virus vector plasmid (rAAV
vector plasmid) designates a nucleic acid construct comprising a
copy of the genetic information to be packaged into AAV capsids, to
form the rAAV. The rAAV vector plasmid is therefore any nucleic
acid construct comprising the recombinant genome of the rAAV as
defined above, preferably a heterologous nucleic acid flanked by
one or two AAV Inverted Terminal Repeats (ITR) and/or, optionally,
one or several RES elements. The rAAV vector plasmid can be
autonomously replicating, conditionally replicating, or stably
integrated into the genome of the packaging cell.
[0012] A rep-cap plasmid designates any nucleic acid construct
encoding the Rep and/or Cap proteins, which provide in trans the
AAV complementing functions lacking in the rAAV vector plasmid.
Generally, the rep-cap plasmid encodes Rep78, Rep68, Rep52, Rep40,
VP1, VP2 and VP3. The rep-cap plasmid can be a single construct
encoding all the required Rep and Cap proteins, under the control
of the same or separate promoters. The rep-cap plasmid can also be
a mixture of distinct nucleic acid constructs encoding one or
several Rep and Cap proteins. The rep-cap plasmid can be
autonomously replicating, conditionally replicating, or stably
integrated into the genome of the packaging cell.
[0013] As indicated above, a first aspect of the invention resides
in a method of characterizing rAAV preparations or stocks. Indeed,
because of the complex methods and constructs needed to produce
rAAV, it is important to have efficient and accurate methods of
characterizing the rAAV preparations obtained, especially for
biological uses. Previous methods known in the art essentially
focus on the determination of the contamination by adenoviruses,
and/or the number of infectious AAV particles. Also, in determining
the number of infectious rAAV particles, most prior art methods
rely on the detection of the expression product of the nucleic acid
inserted in the rAAV genome and are therefore transgene-dependent.
The invention now provides a new method of characterizing rAAV
preparations, which is transgene-independent, sensitive, accurate,
and allows the measure of adenovirus and recombined AAV
contaminants. This method can be used in any rAAV production
method, or as a Quality Control in biological processes, to check
the quality of a preparation and, optionally, allow the improvement
of the production parameters.
[0014] More specifically, an object of the present invention is a
method of characterizing a rAAV preparation, said method
comprising:
[0015] a) contacting a sample of said preparation with a culture of
cells expressing the Rep proteins;
[0016] b) contacting a sample of said preparation with a culture of
cells expressing the Rep proteins, co-infected with an
adenovirus;
[0017] c) contacting a sample of said preparation with a culture of
cells which do not express Rep proteins, co-infected with an
adenovirus;
[0018] and measuring the presence of viruses in cultures a), b) and
c).
[0019] As will be discussed in more details below, the rAAV
preparation can be any preparation of rAAVs produced by any method.
It is, preferably, a purified rAAV stock obtained from a rAAV
producing cell culture extract. The rAAV preparation can be, for
instance, a stock of rAAVs, to be assayed before administration to
a mammalian, including a human being, for clinical or
pharmaceutical purposes. In this regard, "characterizing" means
within the context of the above method, determining both (i) the
number of infectious rAAV particles and (ii) the presence of
contaminating viruses, in particular contaminating adenoviruses and
rep-positive rAAVs in the preparation.
[0020] In the characterization method of this invention, the sample
being contacted with the above indicated cell cultures can be a
pure or a diluted sample of the rAAV preparation. In a preferred
embodiment, serial dilutions of the rAAV preparation are being
used, comprising for instance from about 5.10.sup.4 to 50
infectious particles/ml.
[0021] For carrying out the claimed method, different cell cultures
can be used. Preferably, of course, the cells are permissive to
AAV, i.e., can replicate the AAV genome in appropriate conditions
(i.e., in the presence of adenoviral helper functions). Suitable
cell cultures include culture of human primary or established cell
lines, preferably established cell lines; other mammalian cell
lines or cultures, including canine or murine cells. Example of
cells which can be used include for instance human cells such as
nerve cells, fibroblasts, hepatocytes, myoblasts or the like
preferably established as cell lines. More preferred cell lines
include the HEK cells, Hela cells, Huh7, HT1080, J82, T98G, and the
like, or any human cells harboring part of human papilloma virus
such as SIHA and CASKI cells, for instance.
[0022] In conducting the method of this invention, it is preferred
to use in all three contacting tests the same cell type (i.e.,
HT1080, Huh7, HeLa cells, etc.). In a preferred embodiment, the
cell culture used in a), b) and c) is a culture of HeLa cells.
[0023] In an even more preferred embodiment, the cell cultures used
in a) and b) are cultures of the same cell populations. More
preferably, the cell culture used in c) is also essentially
identical to those used in a) and b), except for the rep
status.
[0024] As explained above, the cells used in a) and b) express the
Rep proteins, i.e., the proteins of AAV which are involved in the
replication of the genome. The rep region of AAVs produces
essentially 4 major proteins, Rep78, Rep 68, Rep 52 and Rep 40. All
these proteins are involved in the replication of AAVs and should
be present in the cell culture, to ensure maximum efficacy.
Accordingly, in a particular embodiment, the cells expressing AAV
Rep proteins used in the invention are cells which express Rep78,
Rep 68, Rep 52 and Rep 40. Preferably, the Rep proteins are
expressed under the control of the AAV p5 and p19 promoters. In a
more particular embodiment, the cells express the AAV Rep proteins
encoded by nucleotides 190-2278 of the AAV genome. Alternative
embodiments use cells which express only some of the AAV Rep
proteins, such as Rep78/Rep68, for instance, which are known to be
essential for replication, or any other combination thereof which
allows replication of AAV genome.
[0025] In addition, in the method of this invention, it is
preferred to use cells which also express AAV Cap proteins, i.e.,
the proteins of AAV involved in the formation of the capsid. The
Cap proteins (VP1, VP2 and VP3) are encoded by nucleotides
1850-4484 of AAV. These proteins can be expressed under the control
of the natural AAV p40 promoter, or any heterologous promoter.
Preferably, cells used in a) and b) also express the AAV Cap
protein expressed by nucleotides 1700-4484 of the AAV genome.
[0026] In a variant, the invention relates to a method of
characterizing a rAAV preparation, said method comprising:
[0027] a) contacting a first sample of said preparation with a
culture of mammalian cells expressing the AAV Rep proteins encoded
by nucleotides 190-2278 of the AAV genome;
[0028] b) contacting a second sample of said preparation with
another culture of the cells of a), co-infected with an
adenovirus;
[0029] c) contacting a third sample of said preparation with a
culture of mammalian cells which do not express Rep proteins,
co-infected with an adenovirus;
[0030] and measuring the presence of viruses in cultures a), b) and
c).
[0031] More preferably, the mammalian cells in a) and b) above also
express the AAV Cap proteins encoded by nucleotides 1850-4484 of
the AAV genome. In a particularly preferred embodiment, the
mammalian cells used in a) and b) comprise a nucleic acid sequence,
integrated in their genome, which codes for the Rep and Cap
proteins of AAV. The nucleic acid has for instance the sequence of
nucleotides 190-4484 of the AAV genome. More preferably, the
mammalian cells in a), b) and c) are HeLa cells, such as cells
HeLaRC32, disclosed in the Examples.
[0032] In assays b) and c) of the instant method, the cell cultures
are co-infected with an adenovirus. Usually, the adenovirus is of
the group C, even more preferably of the serotype Ad2, Ad5, Ad7 or
Ad12. Other types of adenoviruses can be used, such as for instance
canine adenoviruses (CAV-2) which are known to complement AAV
replication. Furthermore, the adenovirus can be wild-type or
modified, in particular temperature sensitive. The doses of
adenoviruses used can be adapted by the skilled artisan, depending
on the cell types, the rAAV preparation, and the nature of the
adenovirus. Generally, the adenovirus is used at an multiplicity of
infection (MOI) of between 5 and 1000, preferably below 800, more
preferably between 50 and 600. All these adenoviruses were titered
on 293 cells using the standard procedures (Graham and Prevec,
1991). Finally, the adenovirus can be replaced with an adenoviral
plasmid, i.e., a plasmid or combination of plasmids encoding the
adenoviral functions necessary for AAV replication, although the
use of a virus is preferred.
[0033] After the contacting of the rAAV preparation with the cell
cultures, the presence of viruses is determined. In this regard,
the measuring comprises measuring the presence of rAAV viruses
within each test a), b) and c). More preferably, the measuring of
the presence of viruses in cultures a), b) and c) comprises
measuring (i.e., detecting) the presence of rAAV replicating DNA in
the cells. This measure is usually accomplished by using
rAAV-specific probes, optionally after amplification of the
cellular nucleic acids with rAAV-specific primers. In a particular
embodiment, a probe is used which is complementary to all or part
of the rAAV genome, in particular to all or part of the
heterologous nucleic acid present in the rAAV genome. The probe
comprises preferably at least 100 bp, to ensure higher selectivity.
The probe can be labeled with any conventional technique
(enzymatic, fluorescent, radioactive, etc.). In a preferred
embodiment, the detection of the presence of rAAV replicating DNA
in the cells of a), b) and c) is performed after cell lysis without
amplification of the cellular nucleic acids. Detection without
amplification allows a better quantitative evaluation of the number
of rAAV genomes present within the cells. Of course, amplification
can be performed, prior to the detection, by using primers specific
for regions of the heterologous nucleic acid present in the rAAV
genome.
[0034] The detection of rAAV DNA within the cells is preferably
accomplished by hybridization of the cellular nucleic acids with a
probe as defined above. Preferably, the hybridization is performed
on a solid support, such as a membrane, filter, etc., onto which
the cellular nucleic acids have been transferred or onto which the
cells have been directly lysed. The hybridization conditions can be
either of medium or, preferably, of high stringency. Typical
hybridization conditions under high stringency are as follows:
Dextran sulfate 5%, SSC 5%, SDS 0,1% liquid block 10%, (Amersham).
The support is prehybridized 30' at 65.degree. C., and the
denatured probes are then hybridized overnight at 65.degree. C. It
should be understood that the hybridization conditions can be
adjusted, for instance, by reducing the temperature or the washing
conditions, by the skilled artisan.
[0035] Generally, the measure is performed 12-96 hours after the
contacting, preferably less than 48 hours. Typically, when
measuring comprises measuring the AAV replicating nucleic acids
within the cell culture, the measuring is performed at about 24
hours post-contacting. It allows replication, but no release of AAV
which could infect other cells.
[0036] The above method is particularly advantageous since it
provides not only the titer of a preparation in infectious
particles, regardless of the nature of the heterologous nucleic
acid contained in the vector, but also the level of contamination
by adenoviruses and rep-positive AAVs.
[0037] Indeed, the contacting with cell culture a), i.e., with a
cell culture expressing Rep proteins but in the absence of helper
adenovirus functions, allows the identification of the presence of
complementing adenoviruses within the rAAV preparation, i.e., the
presence of contaminating adenoviruses. Indeed, it is known that
AAV cannot replicate their genome in the absence of helper
adenovirus functions. In this cell culture a), the replication of
rAAV implies the presence of helper adenoviral function in the
medium, i.e., of contaminating adenoviruses within the preparation.
The number of cells in which rAAV is replicating can be directly
correlated to the level of adenovirus contamination.
[0038] The contacting with cell culture b), i.e., with a cell
culture expressing Rep proteins, in the presence of helper
adenovirus functions, allows the titration of infectious rAAV
particles present in the preparation.
[0039] The contacting with cell culture c), i.e., with a cell
culture which do not express Rep proteins, in the presence of
helper adenovirus functions, allows the identification of the
presence of rep-positive AAV viruses within the rAAV preparation,
i.e., the presence of contaminating rAAV which contain a
Rep-encoding nucleic acid or a Rep protein.
[0040] The method is therefore very efficient and provides
immediate information regarding the quality of a rAAV preparation,
as illustrated in the Examples. Another object of the present
invention therefore lies also in a method as described above, for
detecting the presence of rAAV and/or rep-positive AAVs and/or
adenoviral particles in biological fluids, in particular after in
vivo administration of rAAV preparations in animals and/or human
subjects. Such biological fluids are, more particularly, serum,
urine, stool, saliva, broncho alveolar fluids, etc. The method is
performed as described above, by contacting, in tests a), b) and
c), samples of the biological fluids or dilutions or concentrates
or derivatives thereof. It can thus be used to monitor safety
issues during preclinical, clinical or pharmaceutical settings.
[0041] Furthermore, the above method allowed the inventors to
establish improved conditions and improved nucleic acid constructs
for producing rAAV stocks. Moreover, in performing the above
characterization method, the inventors discovered the presence of
rep-positive AAVs within the preparation, which led them to try to
understand the molecular mechanisms responsible therefor. Indeed,
because the nucleic acid constructs used in the production method
are essentially free of overlapping regions between each other,
homologous recombination events are essentially impossible.
Non-homologous recombination events between the rep-cap plasmid and
the rAAV vector plasmid might account for the presence of such
rep-positive rAAVs in the final preparation. Surprisingly, the
applicants found that the presence of rep-positive particles could
be the result of an ITR-independent packaging of rep nucleic acid.
Surprisingly, the inventors have now discovered the existence of
non-ITR packaging regions within the AAV genome. Both the improved
method and non-ITR packaging regions are also encompassed by the
present application, as will be discussed below.
[0042] The invention therefore relates also to a method of
producing rAAV preparations, comprising:
[0043] a) producing rAAVs in a cell culture expressing the Rep and
Cap functions and the adenovirus helper functions; and
[0044] c) characterizing the rAAVs produced according to the method
disclosed above.
[0045] More preferably, the method comprises:
[0046] a) producing rAAVs in a cell culture expressing the Rep and
Cap functions and the adenovirus helper functions;
[0047] b) purifying the rAAVs produced; and
[0048] c) characterizing the rAAVs produced according to the method
disclosed above.
[0049] The production of the rAAVs can be performed according to
various methods, including conventional methods known in the art.
For instance, step a) can be accomplished by co-transfection of a
cell culture with a rep-cap plasmid, the rAAV vector plasmid and a
helper adenovirus. Step a) can also be performed using a culture of
cells which contain, stably integrated into their genome, nucleic
acid construct(s) encoding the Rep and/or Cap proteins. Also, the
helper adenovirus can be either wild-type adenovirus, or a
replication deficient adenovirus, such as a E1-deficient
adenovirus, for instance. In this embodiment, the cells used
preferably produce the function(s) lacking in the adenovirus. In
particular, the 293 cells are frequently used, which express the E1
functions of adenoviruses.
[0050] In a particular embodiment, in the producing step a), a
rep-cap plasmid is used, which lack any functional ITR region, in
particular any adenoviral ITR region. In this respect and, contrary
to previous observations, the inventors have now shown that the use
of a rep-cap plasmid containing the adenoviral ITR regions does not
increase the yield of infectious particles. It is therefore a
preferred embodiment of this invention to use a rep-cap plasmid
lacking adenoviral ITR regions. Such plasmids are disclosed in the
examples. In particular, a preferred rep-cap plasmid within the
context of the instant invention is a plasmid containing a rep-cap
unit consisting of residues 190-4484 of the AAV genome or fragments
thereof encoding functional Rep and Cap proteins. Furthermore, the
plasmids used may contain the homologous transcriptional promoter
regions of the AAV rep and cap genes (i.e., promoters p5, p19 and
p40), or any heterologous promoter region. Particular examples of
such plasmids are, for instance:
[0051] plasmid pspRC, which contains the ITR-deleted AAV genome
position 190-4484 (FIG. 1), and
[0052] plasmid pspRCC, which contains the rep gene (190-2278 bp of
wtAAV) followed by the bGH polyadenylation signal and the CMV
promoter leading the expression of the cap ORF (1882-4484 bp of
wtAAV).
[0053] In this regard, a particular object of this invention also
resides in plasmids pspRC and pspRCC.
[0054] In another particular embodiment of this invention, step a)
comprises the co-transfection of a cell culture with a rep-cap
plasmid, a rAAV vector plasmid and a helper adenovirus.
[0055] In a further particular embodiment of the present invention,
step a) comprises the use of adenovirus helper plasmids, instead of
a helper adenovirus. Indeed, as indicated before, while the
production of rAAV requires the presence of adenovirus helper
functions, the use of a helper adenovirus generally leads to a
contamination of rAAV preparations by adenoviruses (defective,
replicating and/or wild-type). In order to avoid such a
contamination, the inventors have now found that adenovirus
plasmids can be used, without significantly affecting the yields of
rAAV produced. In this respect, particular plasmids can be used,
such as plasmids carrying the entire adenoviral genome or portions
thereof, which are sufficient to supply the functions required for
rAAV production. In a particular embodiment, a plasmid is used
carrying the entire adenovirus genome. Such a plasmid is for
instance pAdc disclosed on FIG. 1. In another particular
embodiment, a plasmid is used, which contains the adenoviral genome
lacking the left and right ITRs and the packaging region. In a
further preferred embodiment, a plasmid is used, which contains a
defective adenoviral genome lacking the left and right ITRs, the
packaging region and the E1 region. Such a plasmid is, for
instance, plasmid pAd.DELTA. (FIG. 1).
[0056] Surprisingly, the inventors have now shown, for the first
time, that the use of such plasmids in replacement of a helper
adenovirus does not reduce the yields of infectious rAAV particles
produced. Furthermore, the use of such plasmids avoids the
production of contaminating adenoviruses in the preparations, as
demonstrated in the examples. This embodiment thus represents a
preferred way of carrying out the instant invention.
[0057] In this regard, a particular object of this invention also
resides in plasmids pAdc and pAd.DELTA..
[0058] Another object of this invention also resides in a method of
producing rAAVs comprising:
[0059] (i) co-transfecting a cell culture with:
[0060] a rAAV vector plasmid,
[0061] a rep-cap plasmid devoid of ITR, containing preferably a
rep-cap unit consisting of residues 190-4484 of the AAV genome or
fragments thereof encoding functional Rep and Cap proteins, and
[0062] an adenovirus plasmid containing the entire adenoviral
genome or a genome lacking the left and right ITRs, the packaging
region and, optionally, the E1 region, and
[0063] (ii) recovering the rAAV produced.
[0064] In another particular embodiment of the present invention,
step a) above comprises the use of a culture of cells which
contain, in their genome, nucleic acid construct(s) encoding the
Rep and/or Cap functions, preferably the Rep and Cap functions. In
this embodiment, the production step a) comprises the
co-transfection of this cell culture with the rAAV vector plasmid
and with a helper adenovirus or an adenovirus plasmid as described
above. This embodiment is advantageous in that it avoids the need
for a rep-cap plasmid. Suitable cells for carrying this embodiment
include any cells encoding the AAV REP and CAP proteins encoded by
nucleotides 190-4484 of the AAV genome. These cells can be derived,
as disclosed above, from human cells, such as embryonic cells, or
even from other mammalian cells. Preferred cells are obtained from
293 cells, HeLa cells, A549 cell, Huh7 cells, HT1080, J82, T98G,
HER, or any human cells harboring part of human papilloma virus
such as SIHA and CASKI cells. A specific example is the HeLaRC32
cells disclosed above, for instance.
[0065] Another object of this invention therefore resides also in a
method of producing rAAVs comprising
[0066] (i) co-transfecting a culture of cells which contain, in
their genome, nucleic acid construct(s) encoding the Rep and/or Cap
functions, preferably the Rep and Cap functions, with:
[0067] a rAAV vector plasmid, and
[0068] a helper adenovirus or an adenovirus plasmid containing the
entire adenoviral genome or a genome lacking the left and right
ITRs, the packaging region and, optionally, the E1 region, and
[0069] (ii) recovering the rAAV produced.
[0070] Step b) of the above production method comprises the
purification of the rAAV produced. Said purification can be
performed according to various techniques, including methods known
in the art such as centrifugation, clarification, and cesium
chloride gradient purification. In this regard, the CsCl
purification procedure disclosed in the art essentially comprises
the centrifugation of the rAAV cell extract at 41 000 rpm for 48
hours (rotor sw41). The inventors have now shown that it is
possible to significantly reduce the length of the centrifugation
step, when the other parameters are adjusted. In particular, the
inventors have now shown that when the CsCl centrifugation is
maintained for 6 hours at between 60 000 and 70 000 rpm, preferably
between 65 000 and 70 000 rpm, the same level of purity is obtained
than with previous methods, requiring 48 hours centrifugation.
[0071] A preferred embodiment of the invention therefore comprises
centrifuging the rAAV preparation in a cesium chloride gradient for
less than 12 hours, at between 60 000 and 70 000 rpm, preferably
between 65 000 and 70 000 rpm.
[0072] Other purification methods can be used such as, in
particular, chromatographic techniques. In this respect, a
particular purification method uses anion exchange chromatography
optionally combined with heparin substrate, and further optionally
combined to exclusion chromatography. A specific purification
protocol comprises for instance the loading of the rAAV preparation
on an anionic exchange chromatography column, such as for instance
Resource Q (Amersham, Pharmacia Biotech) and heparin substrate,
preceded by an exclusion chromatography, for instance onto a
Sephacryl (S500) column, to exclude the viral particles and
separate them out of residual protein and lipid contaminants.
[0073] A particular object of the present invention therefore also
lies in a method of purifying rAAVs from a biological sample,
comprising treating said sample in a cesium chloride gradient
centrifugation at between 60 000 and 70 000 rpm, preferably for
less than 12 hours, and recovering the fraction(s) containing the
purified rAAVs.
[0074] Another particular object of this invention lies in a method
of purifying rAAVs from a biological sample, comprising treating
said sample at least by anion exchange chromatography and a heparin
column and exclusion chromatography.
[0075] Furthermore, as indicated above, the invention also relates
to novel nucleic acids with packaging activity. Indeed, the
inventors have now discovered that nucleic acid regions, distinct
from the ITRs, can mediate and/or increase the packaging of nucleic
acids within AAV capsids. These regions, termed "Replication
Encapsidation Sequence" or "RES" as well as their use for producing
rAAVs, represent another particular object of the instant
invention.
[0076] More particularly, within the context of the present
invention, a "Replication Encapsidation Sequence" represents a
sequence different from an AAV ITR sequence, which promotes and/or
facilitates the packaging of a nucleic acid into an AAV capsid.
Preferably, a RES element comprises a Rep Binding Element (RBE),
where Rep78/Rep68 proteins can bind, and a terminal resolution site
(trs), for the binding of endonucleases. Even more preferably a RES
comprises a RBE site, a trs site and a palindromic sequence.
Surprisingly, the Applicants have now found that regions favoring
packaging of a nucleic acid into an AAV particle are present in the
genome of viruses, such as AAV. These regions are distinct from the
ITRs, and can promote the packaging of an ITR-free nucleic acid
into an AAV particle. The invention therefore discloses new nucleic
acids with packaging activity, that can be used to package nucleic
acids into AAV capsids, or to improve the packaging efficiency of
conventional rAAV vectors.
[0077] A particular object of this invention lies in an isolated
Replication Encapsidation Sequence, wherein said RES is a nucleic
acid sequence distinct from an AAV ITR sequence, which provides or
promotes the packaging of a nucleic acid operably linked thereto
into an AAV particle.
[0078] Preferably, a RES of this invention comprises a Rep Binding
Element and a terminal resolution site, and, optionally, a
palindromic sequence (nucleotides 231-250 of SEQ ID NO:8). A
preferred RES according to this invention is an isolated nucleic
acid, wherein, said nucleic acid provides or promotes the packaging
of a polynucleotide operably linked thereto into an AAV
particle.
[0079] A particular example is a RES comprising a region having the
sequence of SEQ ID NO: 1 (GCC CGA GTG AGC ACG CAG) or a functional
variant thereof.
[0080] More preferably, a RES of this invention comprises a region
having the sequence of SEQ ID NO: 1 or a functional variant thereof
and further comprises a region having the sequence of SEQ ID NO: 2
(GCG ACA CCA TGT GGT CAC GC) or a functional variant thereof.
[0081] A particular RES of this invention is a nucleic acid
comprising SEQ ID NO: 3 (GCC CGA GTG AGC ACG CAG GGT CTC CAT TTT
GM) or a functional variant thereof, which nucleic acid having a
packaging activity.
[0082] Even more preferably, a RES of this invention comprises SEQ
ID NO: 3 or a functional variant thereof and SEQ ID NO: 2 or a
functional variant thereof.
[0083] The term "functional variant" means any sequence comprising
one or several structural modifications, that still retains the
activity of the sequence, i.e., the binding to Rep78 or Rep68 for
SEQ ID NOS: 1 and 3, and the ability to form a hairpin structure
for SEQ ID NO: 2. More preferred functional variants retains at
least 50% identity with the sequence disclosed in the examples.
Particular examples of functional variants are sequences with one
or several mutations, additions or deletions in the above
sequences, in particular 1, 2 or 3 mutations.
[0084] Other typical variants are sequences which hybridize with
the above sequences and retain the RES activity. To search for
sequences in the genome of other viruses having some homologies
with the RES fragment and be considered as variants, hybridization
in low stringency conditions using the RES fragment as a probe are
performed. As an example, hybridization is done overnight at
56.degree. C. instead of 65.degree. C. in the hybridization
solution recommended by the manufacturer's instructions (Amersham),
which is 5.times.SSC, 0.1% (W/V) SDS, 5% (W/V) Dextran sulphate and
{fraction (1/20)} of liquid block. Washes are performed also in low
stringency conditions, for instance: 1.times.SSC, 0,1 % SDS at
56.degree. C. The RES can be artificial, semi-synthetic, viral or
of any other origin.
[0085] Other functional variants are RES sequences derived from
other parvoviruses which exhibit the functional properties of the
above RES sequences.
[0086] Preferably, RES elements contain less than 400 bp. A
specific example is provided on FIG. 8A, which consists of
nucleotides 190-540 of AAV genome (i.e., 350 bp). Fragments of this
RES are presented in FIG. 10, which represent particular
embodiments of this invention, such as nucleotides 190-350,
230-350, 230-300, 250-300 or functional variants thereof, for
instance.
[0087] As indicated above, the RES elements of this invention can
be used either to promote the packaging of a nucleic acid into an
AAV capsid, or to improve the packaging efficiency of rAAV vector
plasmids. Accordingly, a particular object of this invention is a
nucleic acid consisting of one or several RES elements fused to a
heterologous polynucleotide lacking a functional ITR sequence.
[0088] The heterologous polynucleotide lacking a functional ITR
sequence can be any nucleic acid sequence of interest, such as a
nucleic acid coding for a protein or RNA of interest for instance.
Fused means that the two elements are operably linked together, in
the same or reverse orientation. The RES elements can be fused at
the 5' or 3' ends of the heterologous nucleic acid, or inserted
within said polynucleotide. Examples of such nucleic acids include
for instance plasmid RES+CMV-LacZ or plasmid RES-CMV-LacZ as shown
on FIG. 9. Any similar construct in which the CMV promoter is
replaced with another promoter and/or the LacZ gene is replaced
with another gene can be produced by the skilled artisan.
[0089] Another object of this invention is a rAAV vector plasmid,
comprising a recombinant AAV genome and one or several RES element.
A particular construct is, for instance, a plasmid comprising the
following operably linked elements: 5'-ITR-RES-heterologous
polynucleotide-ITR-3'.
[0090] The invention also relates to a rAAV particle comprising a
recombinant nucleic acid genome, wherein said recombinant nucleic
acid genome comprises one or several RES elements.
[0091] The invention can be used to produce rAAV and to
characterize rAAV preparations, for use in various technical areas,
such as experimental biology, preclinical studies, clinical studies
or pharmaceutical indications.
[0092] Other advantages and uses of the present invention be
disclosed in the following experimental section, which should be
regarded as illustrative and not limitative.
LEGEND TO THE TABLES
[0093] TABLE 1. Characterization of the 18 large scale rAAV stocks.
Each vector was produced from 25 15-cm plates of 293 cells except
for stocks marked with an asterisk which were produced from 50
15-cm plates of cells. The vector name indicates the promoter and
the transgene inserted between AAV ITRs. The second column
indicates the rAAV size (ITR to ITR); the third, the rep-cap
construct used for the production and the fourth whether adenovirus
(wtAd5 or Ad.dts) or an adenoviral plasmid (pAdc) was used. The
rAAV titer was measured by:.sup.1 dot blot, .sup.2 RCA.
Contaminations.sup.3 with adenovirus and rep-positive AAV particles
were also measured by RCA. The last column indicates the final
volume of virus after CsCl gradient purification and dialysis. rAAV
stocks listed below the darker line in the middle of the table were
purified following the centrifugation conditions described in
Materials and Methods. Those listed above were purified following
the protocol described by Snyder et al. (1996). p./ml:
particles/ml; i.p./ml: infectious particles/mi.
[0094] TABLE 2. Evaluation of different rep-cap expression plasmids
for rAAV production. rAAV was produced from two 15-cm plates of 293
cells transfected with the AAVCMVnlsLacZ vector, the adenoviral
plasmid pAdc and the indicated rep-cap construct. Cells were
collected three days after transfection and cell extracts purified
as described in Materials and Methods. Each stock was titered by:
.sup.1 dot blot, .sup.2 RCA and .sup.3 by an LFU assay performed on
HeLa cells. The final volume of virus was of 7 ml for each
stock.
[0095] TABLE 3. Comparison of the two adenoviral plasmids pAdc and
pAd.DELTA. for rAAV production. rAAV was produced from two 15-cm
plates of 293 cells transfected with the pspRC construct, the
AAVCMVnlsLacZ vector and the indicated adenoviral plasmid. Cells
were collected 72, 96 and 120 hours after transfection and cell
extracts purified as described in Materials and Methods. Each stock
was titered by: .sup.1 dot blot and .sup.2 RCA.
[0096] TABLE 4. Measure of infectious and transducing
rAAVCMVnlsLacZ particles produced using either Ad.dl324 (vAd) or an
adenoviral plasmid (pAdc). Rep-cap functions were provided by the
pspRC construct. In the case of the stock obtained with vAd, the
virus was produced from 20 15-cm plates of 293 cells and the final
volume was of 6.8 ml. In the case of the stock obtained with pAdc,
the virus was produced from 50 15-cm plates of 293 cells and the
final volume was of 13.4 ml. Recombinant AAV was titered by: .sup.1
dot blot, .sup.2 RCA, .sup.3 and .sup.4 LFU assay on HeLa cells in
the presence or in the absence of wtAd 5. ND, not done.
LEGEND TO THE FIGURES
[0097] FIG. 1: Constructs used for rAAV production. A. Rep-cap
constructs. Constructs pAAV/Ad and plM45 are described in
references (Samulski et al., 1989) and (Pereira et al., 1997)
respectively. The pspRC and the RepCMVCap constructs are described
in Materials and Methods. Numbers on the top correspond to position
on the wild type AAV genome. CMVp: CMV promoter, pA: poly A signal
from the bovine growth hormone gene. B. rAAV plasmids derived from
the psub201 plasmid (Samulski et al., 1989). Numbers refer to wild
type AAV sequences maintained in this plasmid and flanking the
viral ITRs. C. Adenoviral plasmids: pAdc has the entire wild type
Ad5 genome. Plasmid pAd.DELTA. has a deletion of the 5' and 3'
ITRs, the .PSI. and E1 regions until position 917 of wtAd5.
[0098] FIG. 2: rAAV production protocol. rAAV was produced using
two different protocols: A. 293 cells were transfected with the
rAAV vector and the rep-cap plasmid (1:1 ratio) and infected 6
hours later with adenovirus. When a cytopathic effect (CPE) was
evident, cells were harvested and processed as described in
Materials and Methods. B. 293 cells were transfected with the rAAV
vector, the rep-cap plasmid and the adenoviral plasmid (1:1:2
ratio), washed after 6 hours, collected three days later and
processed. After centrifugation rAAV containing fraction were
pooled, dialyzed and titered by dot blot and RCA as described in
Materials and Methods. Asterisks indicate the steps which were
modified from the original protocol (Snyder et al, 1996).
[0099] FIG. 3: Characterization of rAAV by a modified Replication
Center Assay (RCA). HeLaRC32 cells, harboring two integrated copies
of an ITR-deleted AAV genome or control HeLa cells were infected
with different dilutions of the rAAV stocks and either infected or
not with wtAd5 (see Materials and Methods for details). Twenty-four
hours later cells are trypsinized, filtered through a membrane,
lysed and the filters were hybridized overnight with a transgene
probe. The number of infectious rAAV particles is determined by
counting the number of spots on HeLa32RC cells in the presence of
adenovirus. The same assay performed in the absence of adenovirus
gives the level of contamination with infectious adenoviral
particles. No signal is detected when HeLaRC32 cells are infected
with adenovirus alone. Finally, the result on control HeLa cells in
the presence of adenovirus gives the level of contamination with
rep-positive AAV.
[0100] FIG. 4: Comparison of rAAV yields produced with different
rep-cap plasmids and either adenovirus or an adenoviral plasmid.
rAAV stocks described in Table 1 were titered by dot blot and RCA
to measure, respectively, the number of particles and infectious
particles. The x axis indicates the size of each rAAV and the
production protocol followed: pAAV/Ad+vAd. transfection of the rAAV
vector with the pAAV/Ad plasmid followed by infection with
adenovirus; pspRC+vAd. transfection of the rAAV vector with the
pspRC plasmid followed by infection with adenovirus; pspRC+pAdc.
transfection of the rAAV vector, the pspRC construct and the
adenoviral plasmid pAdc harboring the entire adenoviral genome. All
the rAAV stocks were prepared from 25 15-cm plates of 293 except
for stocks marked with an asterisk which were prepared from 50
15-cm plates of cells.
[0101] FIG. 5: Analysis of the replication of rAAVCMVnlsLacz in 293
cells. 293 cells were infected with rAAVCMVnlsLacZ (m.o.i. of 100
as defined by RCA), produced either with adenovirus (vAd) or the
adenoviral plasmid (pAdc), in the presence or in the absence of
Ad.dl324 (m.o.i. of 10). Three days later, cells were harvested,
lysed, and low molecular weight DNA was extracted, run on a gel,
transferred to a membrane and hybridized to a LacZ probe as
described in Materials and Methods. ss DNA: single-stranded DNA;
mRF: monomer double-stranded DNA; dRF: dimer double-stranded
DNA.
[0102] FIG. 6: Nuclear targeted .beta.-galactosidase expression in
the rat muscle after injection of rMVCMVnlsLacZ. Tibialis anterior
muscles of three 9 weeks-old Wistar rats were injected with 2.5
10.sup.8 rAAVCMVnlsLacZ infectious particles each (three sites of
injection per muscle). Animals were sacrificed 4 weeks after
injection. Muscles were fixed with paraformaldehyde, stained
overnight at 37.degree. C. with X-Gal, paraffin embedded and
sectioned into 4-.mu.m sections which were counterstained with
Kernechtrot solution. (magnifications: panel .times.100; inset
panel .times.600).
[0103] FIG. 7: Description of plasmids BS-RC, dITR-RC and dITR-RCA.
Plasmid BS-RC comprises a rep-cap unit (nucleotides 190-4492) of
AAV, inserted in the Bluescript plasmid. Plasmid dITR-RC comprises
the same rep-cap unit as plasmid BS-RC, but flanked by the
adenovirus type-5 ITRs, also inserted in the Bluescript plasmid.
Plasmid dITR-RC.DELTA. is identical to plasmid dITR-RC, with a
deletion of 350 bp in the rep unit (nucleotides 190-540).
[0104] FIG. 8: Schematic representation of the RES element. The RES
element represented corresponds to nucleotides 190-540 of wild-type
AAV genome (AF043303). RBE (8A). FIG. 8B gives the nucleic acid
sequence of this RES element.
[0105] FIG. 9: Description of plasmids pCMV-LacZ, pRES+/CMV-LacZ
and pRES-/CMV-LacZ. Grey box: LacZ gene linked to a nuclear
localization signal; black box: Promoter and enhancer of the
cytomegalovirus (CMV) (Ndel-BamHI fragment of the pRC/CMV plasmid);
empty box: double strand oligonucleotide (20 bp) harboring a
translation stop codon in each of the three opening reading frames;
thick hatched box: polyadenylation signal of the bovine growth
hormone (BamHI-Pvull fragment of the pRC/CMV plasmid); thin hatched
box: RES element (nucleotides 190-540 of AAV).
[0106] FIG. 10: Representation of deleted and mutated RES elements.
Only the mutation introduced in the 190-350 fragment are
represented. Simultaneous mutations at various positions can also
be done.
MATERIALS AND METHODS
[0107] 1. Cell lines and viruses
[0108] 293 and HeLa cell lines were maintained in DMEM medium
(SIGMA) supplemented with 10% heat-inactivated foetal calf serum
(FCS, SIGMA) and 1% (vol/vol) penicillin/ streptomycin (GIBCO BRL,
5000 U/ml).
[0109] Adenoviruses used are: wild type Adenovirus type 5 (wtAd5)
(ATCC VR-5), Ad.dl324 (a gift from Transgne, France) and the double
thermosensitive Ad.dts (Moullier P., unpublished data) which
cumulates the ts125 mutation in the E2a region and the ts149
mutation located in the E2b DNA polymerase (Ginsberg et al., 1977).
All these adenoviruses were produced and titered on 293 cells using
the standard procedures (Graham and Prevec, 1991).
[0110] Cells and adenoviruses were tested for the absence of wild
type AAV (wtAAV) by PCR as indicated below.
[0111] HeLaRC32 cells are HeLa cells containing a rep-cap construct
stably integrated into their genome. These cells were constructed
as follows:
[0112] A stable cell line harboring the rep and cap genes was
obtained by co-transfecting HeLa cells by the calcium-phosphate
precipitation method with two plasmids:
[0113] plasmid PGK-Neo which harbors the neomycin resistance gene
under the control of the mouse phospho-glycerate kinase-1
promoter;
[0114] plasmid pspRC which harbors the ITR-deleted AAV genome (from
position 190 to 4492 bp of the wild-type AAV genome) inserted in
the Psp72 plasmid (FIG. 1).
[0115] Clones isolated following G418 selection (1 mg/ml for three
weeks) were screened for the presence of stably integrated rep-cap
copies by Southern blot using a rep probe. Only four among 45 HeLa
cell clones were positive by Southern blot analysis. Clone HeLaRC32
had the highest copy number per cell genome.
[0116] To document Rep and Cap protein expression, the HeLaRC32
cells were infected with wild-type adenovirus and analyzed by
immunofluorescence using an anti-rep 78/52 monoclonal antibody and
an anti-cap serum. About 50% of the cells were positive for cap
proteins. Most of the cap positive cells (70%) also stained
positive for Rep 78/52. These results indicated the presence of Rep
and Cap proteins upon adenovirus infection of HeLaRC32 cells.
[0117] The production of functional Rep proteins by the HeLaRC32
cells was also tested by looking at their ability to replicate an
AAV vector (ITR-transgene-ITR). For this, the cells were
transfected with the rAAV-CMV-LacZ vector plasmid and subsequently
infected with wild type adenovirus. Two days later, low molecular
weight DNA was extracted following the Hirt procedure, digested
with Dpnl, and analyzed by Southern blot using a LacZ probe. The
result indicated the presence of the typical AAV replicative forms
(the monomer and the dimer double-stranded DNA) upon adenovirus
infection of HeLaRC32 cells, whereas no signal was seen when cells
were uninfected or when they were not transfected with the
AAVCMV-LacZ vector plasmid.
[0118] These results showed that cells prepared following the above
protocols, especially starting with HeLa cells, stably produce
functional Re and Cap proteins. HeLaRC32 cells have been used in
the experimental section.
[0119] 2. DNA constructs
[0120] 2.1. rep-cap plasmids:
[0121] the pspRC plasmid (FIG. 1A) contains the ITR-deleted AAV
genome (positions 190 to 4484 bp). It was excised as an Xba I
fragment from the psub201 plasmid (Samulski et al., 1989) and
inserted into the Xbal site of pSP72 plasmid (Promega).
[0122] The pspRCC plasmid (FIG. 1A) contains the rep gene (190 to
2278 bp of wt AAV) followed by the bovine growth hormone gene polyA
signal and by the CMV promoter leading the expression of the cap
ORF (1882 to 4484 bp of wt AAV). This plasmid was derived from
pspRC by partially deleting the Cap ORF with Xhol which cuts at
position 2232 of wtAAV (upstream of the stop signal for Rep 68 and
Rep40) and further downstream in the plasmid backbone. The 324 bp
polyA signal from the bovine growth hormone gene was then inserted
downstream of the rep ORF to give the pspRep/pA plasmid. This
construct codes for Rep 78 and Rep 52 proteins and contains the p40
promoter of AAV. In order to complete the Rep ORF, a 90 bp PCR
fragment including region 2193 to 2278 bp of wtAAV was obtained
using the following primers: 5'atgatttaaatcaggttgggctgccg3' (SEQ ID
NO: 4; positions 2187 to 2212 of wtAAV) and
5'gctctagatgagcttccaccactgtc3' (SEQ ID NO: 5; positions 2278 to
2251 of wtAAV). This PCR fragment which includes a mutated VP1
start site (underlined in the primer sequence) was inserted between
the Swal and Xbal sites of the pspRep/pA plasmid to give plasmid
pspRep/pA .DELTA.VP1. To obtain plasmid pspRCC, a cassette composed
of the cap ORF (1882 to 44874 bp of wtAAV) placed under the control
of the CMV promoter (873 bp) was inserted downstream the poly
signal of pspRep/pA (.DELTA.VP1) at the unique Pvull site.
[0123] 2.2. rAAV vector plasmids:
[0124] rAAV vector plasmids were derived from psub201 (Samulski et
al., 1989) by deleting the rep-cap Xbal or SnaBI region and
replacing it with different expression cassettes (FIG. 1B).
[0125] 2.3. Adenoviral plasmids:
[0126] Two adenoviral plasmids were generated (FIG. 1C): i) plasmid
pAdc contains the complete adenoviral genome cloned into the
SuperCos plasmid (Stratagene); ii) the pAdA plasmid contains an
adenoviral genome with both ITRs, the packaging signal (.PSI.) and
the E1 region deleted, also cloned into the SuperCos plasmid.
[0127] 3. rAAV production
[0128] rAAV is produced using the procedure detailed in FIG. 2: on
day 1, twenty five 15-cm plates of 293 cells (at approximately 80%
of confluence) are co-transfected by the calcium phosphate method
with the rep-cap and the vector plasmids (12.5 .mu.g each). Six
hours later, cells are washed with DMEM and infected with
adenovirus (moi of 10) in DMEM 5% FCS (FIG. 2A). Under these
conditions, a cytopathic effect (CPE) is visible approximately 3
days later. When using the Ad.dts adenovirus, the cells are
incubated at 32.degree. C. which is the permissive temperature for
adenoviral growth. Alternatively, when rAAV is produced using an
adenoviral plasmid (FIG. 2B), each dish is transfected on Day 1
with three plasmids: the rep-cap, the vector (12.5 .mu.g each) and
the pAdc or pAd.DELTA. plasmids (25 .mu.g). Six hours later, cells
are washed and incubated in DMEM 5% FCS. No cytopathic effect is
evident under these conditions and cells are usually harvested 3
days later unless otherwise indicated.
[0129] 4. rAAV purification
[0130] To purify rAAV particles, cellular pellets (each
corresponding to six 15-cm plates) are resuspended in 20 ml of 10
mM Hepes pH 7.6/150 mM NaCl buffer and lysed by three cycles of
freeze/thawing (dry ice with ethanol/37.degree. C. water bath). The
cell lysate is then centrifuged at 3,000 rpm for 15 mn to remove
cellular debris and further clarified by centrifuging at 10,000 g
(Beckman rotor JA17) for 10 mn at 4.degree. C. The supernatant is
then precipitated by addition of the same volume of cold saturated
(NH.sub.4).sub.2SO.sub.4 (pH 7.0) and incubation for 20 mn on ice.
After centrifugation at 12,000 g for 20 mn at 4.degree. C. (Beckman
rotor JA17), the supernatant is removed and the pellet is
resuspended in 3 ml of Phosphate Buffered Saline pH 7.0 (PBS) which
are then loaded on top of a CsCl step gradient composed of 3 ml of
1.5 g/ml and 3 ml of 1.35 CsCl in PBS (Beckman Optiseal.TM.
centrifuge tubes). The gradients are centrifuged for 6 hours
(minimum time required to reach equilibrium) to overnight at 67,000
rpm at 15.degree. C. (Beckman rotor 90 Ti). After centrifugation,
an adenoviral band is visible in the middle of the tube, whichever
adenoviral helper system was used (virus or plasmid). Ten fractions
(20 drops each) are recovered from the bottom of the tube using the
Beckman Fraction Recovery System and analyzed by dot blot to
identify those containing rAAV genomes (see below). Usually, rAAV
particles are concentrated within 6 fractions located below the
adenoviral band. The rAAV containing fractions are then pooled and
dialyzed for 24 hours against three changes of PBS supplemented
with 0.9 mM CaCl.sub.2 and 0.5 mM MgCl.sub.2. The viral suspension
is then aliquoted and stored at -80.degree. C. without adding any
stabilizer. The final titer of the rAAV preparation is determined
using a frozen aliquot of virus following the methods described
below.
[0131] An alternative purification method comprises chromatographic
treatment of the rAAV preparation. In a typical experiment, the
rAAV particles are collected on an iodixanol gradient or a
"streamline" system (Pharmacia). When a iodixanol gradient is used,
the pool of rAAV is first loaded onto an exclusion column (G25) to
eliminate the iodixanol. The running buffer consists of Tris 10 mM,
NaCl 80 mM, Ca.sup.++ 1 mM and Mg.sup.++ 1 mM.
[0132] The rAAV fractions, detected by dot blot, are pooled and
loaded onto an anionic exchange column (Resource Q, Amersham
Pharmacia Biotech). Flow rates vary between 1 ml/mn to 10 ml/mn.
The setting allows low and medium pressure liquid chromatography.
The loading buffer is made of Tris 10 mM, pH 8.0-8.5, NaCl 80 mM,
Ca.sup.++ 1 mM and Mg.sup.++1mM. Once loaded, elution of the rAAV
particles is obtained using the elution buffer: Tris 10 mM, pH
8.0-8.5, NaCl 160 mM, Ca.sup.++ 1 mM and Mg.sup.++ 1 mM.
[0133] For this anionic exchange chromatography, a 1 ml capacity
column is used to load an equivalent of 1.times.10.sup.8 to
1.times.10.sup.9 infectious particles total of rAAV. A 6 ml
capacity column is used to load approximately 1.times.10.sup.10
infectious particles at once. Higher viral loads can be treated for
instance by fractionating the initial viral load.
[0134] The rAAV fractions eluting between NaCl 100 to 160 mM are
pooled. The pooled rAAV are then loaded onto a Sephacryl (S500)
column to exclude the viral particles and separate them out of
residual protein and lipid contaminants.
[0135] Another column comprises an Heparin column to which the AAV
crude extract is loaded, washed in PBS-MgCl.sub.2-CaCl.sub.2 1 mM
each, and finally eluted in PBS-NaCl 500 mM.
[0136] Finally, rAAV can be concentrated on two sucrose cushions
30% and 50% after ultracentrifugation at 39,000 rpm overnight at
4.degree. C. Recombinant AAV pellets are resuspended in
lactate-Ringer under gentle stirring at 4.degree. C. for 4
hours.
[0137] 5. Titration of rAAV stocks
[0138] Two different methods are used to measure the rAAV titer: 1)
the dot blot analysis to measure the number of particles/ml based
on the quantification of viral DNA; 2) a modified Replication
Center Assay (RCA) to measure the number of infectious
particles/ml, as well as the level of contaminating adenovirus and
rep-positive AAV.
[0139] 5.1. Dot blot analysis:
[0140] 1 and 10 .mu.l of the viral stock are incubated with 20 U of
DNasel (Boehringer Manheim) in 200 .mu.l of DMEM for 1 hr at
37.degree. C. Two hundred .mu.l of 2.times. Proteinase K solution
(20 mM Tris-HCl pH 8.0, 20 mM EDTA pH 8.0, 1% SDS) are then added
and the samples incubated further for 1 hr at 37.degree. C. Viral
nucleic acid are then purified by a phenol/chloroform extraction,
precipitated after addition of NaOAc/ethanol and incubation for 20
mn at -80.degree. C. After 30 mn of centrifugation at 15,000 g, the
nucleic acid pellet is washed in 75% ethanol and resuspended in 400
.mu.l of 0.4M NaOH, 10 mM EDTA. After heating at 100.degree. C. for
5 mn, the DNA is loaded on a Zetaprobe membrane (Biorad) using a
dot blot apparatus. As a standard for the determination of the
amount of viral DNA, several dilutions of rAAV vector plasmid used
to produce the virus are loaded on the same membrane. After
blotting, the membrane is prehybridized for 30 mn at 60.degree. C.
A denatured fluorescein-labelled probe (Amersham, Gene Images
random prime labeling module) corresponding to the cDNA included in
the rAAV vector is then added and incubated overnight at 60.degree.
C. The following day the membrane is processed according to the
manufacturer's protocol (Amersham, Gene Images CDP-star detection
module) and exposed to an autoradiography film.
[0141] 5.2. Modified RCA:
[0142] HeLaRC32 (a stable HeLa cell clone expressing rep-cap) and
control HeLa cells are seeded the previous day in a 48-well plate
(8.times.10.sup.4 cells/well), in DMEM containing 10% fetal calf
serum (FCS) (for each rAAV preparation: 3 wells with HelaRC32 for
titration; 1 well with HelaRC32 to measure adenovirus
contamination; 2 wells with HeLa to measure rep-positive AAV
contamination).
[0143] The following day, cells are infected, as appropriate, with
the rAAV samples and with or without helper adenovirus.
[0144] In this regard, the stocks of adenoviruses are prepared as
follows, generally under P3 confinement conditions. The helper
virus used in these experiments is a wild-type adenovirus type 5
(Adwt5) at a Multiplicity of Infection (MOI) of about 500. A mix
containing the number of pfu necessary to infect all the wells into
the appropriate volume of medium (DMEM, 5% FCS) is prepared. For
adenovirus infection, the medium is removed from the cells and 200
.mu.l of the solution (DMEM, 5% FCS) containing the adenovirus at
the selected MOI are added.
[0145] For rAAV titration, three wells of HelaRC32 cells, infected
with adenoviruses as described above, are infected with 3 dilutions
of the rAAV preparation (10.sup.-4, 10.sup.-5 and 10.sup.-6)
prepared by serial dilution in DMEM, 5% FCS (10 .mu.l of the rAAV
preparation into 90 .mu.l of medium). 2 .mu.l of each rAAV sample
are added to the 200 .mu.l of medium with adenovirus.
[0146] For measuring adenovirus contamination, one well of HelaRC32
cells, not infected with adenoviruses, is infected with the rAAV
preparation (pure sample). 2 .mu.l of each rAAV sample are
therefore added to the 200 .mu.l of medium without adenovirus.
[0147] For measuring rep-positive AAV contamination, two wells of
HeLa cells, infected with adenoviruses as described above, are
infected with samples of the rAAV preparation (pure sample and
10.sup.-1 dilution in DMEM, 5% FCS). 2 .mu.l of each rAAV sample
are added to the 200 .mu.l of medium with adenovirus.
[0148] Twenty four hrs later, the medium is removed from each well,
and the wells are washed twice in PBS. The cells are trypsinized by
addition of 200 .mu.l trypsin for 10 minutes at 37.degree. C. 300
.mu.l DMEM, 5% FCS are added and the cells are detached, by
pipetting up and down several times. The cell suspension is then
added to an eppendorf tube. The wells are washed again with 500
.mu.l of PBS and this volume is added to the cell suspension. The
eppendorf tubes are centrifuged (3000 rpm, 5 minutes), and the
supernatant is removed with a pipet. The cell pellets are suspended
in 1 ml PBS by pipetting up and down several times, and stored on
ice.
[0149] Each suspension is filtered under vacuum through a
Zeta-probe membrane (Biorad). The filters are put on a Whatman
paper soaked in 0.5 M NaOH/1.5 M NaCl for 5 mn, and then
neutralized in a 1 M Tris HCl (pH 7.5)/2.times.SSC (0.3 M NaCl,
0.03 sodium citrate) solution. Filters are dried on dry Whatman
paper and hybridized overnight to a fluorescein-labelled probe
corresponding to the cDNA included in the rAAV vector (Amersham,
Gene Images random prime labeling module). The filters are then
processed as described by the manufacturer (Amersham, Gene Images
CDP-Star detection module).
[0150] 6. LacZ Forming Unit (LFU) assay
[0151] To measure the transducing activity of rAAV harboring the
.beta.-galactosidase gene (LacZ), HeLa cells, seeded the day before
at 2.times.10.sup.5 cells per well (24-well plate), were infected
with pure or diluted rAAV in DMEM 5% FCS in the presence or in the
absence of wtAd5 (m.o.i. of 50). Twenty-four hours later cells are
washed in PBS, fixed with 0.5% glutaraldehyde for 5 mn at room
temperature and then stained with
5-bromo-4-chloro-3-indolyl-.beta.-D-galactopyranoside (X-Gal) for 6
hrs at 37.degree. C.
[0152] 7. Southern blots on low molecular weight DNA
[0153] For extracting low molecular weight DNA, cells are
trypsinized and lysed in a solution of 10 mM TrisHCl (pH 8.0)/10 mM
EDTA/1% SDS for 30 mn at 37.degree. C. After addition of Proteinase
K at 500 .mu.g/ml final (Boehringer Manheim), the lysate is
incubated for 2 hrs at 37.degree. C. To precipitate high molecular
weight DNA, 5 M NaCl is added to the cell lysate (final
concentration of 1.1 M) and incubated overnight at 4.degree. C.
High molecular weight DNA is pelleted by centrifugation at 10000
RPM for 20 min at 4.degree. C. and the supernatant is extracted
twice with chloroform at room temperature. The nucleic acids are
then precipitated with ethanol and resuspended in 10 mM Tris pH
8.0/1 mM EDTA containing 200 .mu.g/ml of RNase (Boehringer
Manheim). After incubation for 15 mn at room temperature, the DNA
is extracted twice with chloroform and ethanol/NaOAc precipitated.
The final pellet is resuspended in 10 mM Tris pH 8.0/1 mM EDTA and
stored at -20.degree. C. For analysis, the DNA is digested with
Dpnl (which cleaves only input methylated plasmid DNA), run on a 1%
agarose gel and transferred under alkaline conditions on a Hybond
N.sup.+ membrane (Amersham). The membrane is hybridized to a
fluorescein-labelled probe and processed as described above.
[0154] 8. Detection of rep sequences by PCR
[0155] Cells and adenoviral stocks were routinely assayed for the
presence of rep-positive AAV by PCR. The PCR primers were: Rep1
(5'-TATTTAAGCCCGAGTGAGCA-3'; SEQ ID NO: 6) which corresponds to
positions 255 to 275 of wild type AAV in the p5 promoter; and Rep3
(5'-AAAGTTCTCATTGGTCCAGT-3'; SEQ ID NO: 7) which corresponds to
positions 1417 to 1397 of wild type AAV in the Rep52/40 ORF. PCR
was carried using Taq polymerase (Gibco BRL) for 25 cycles (30 sec.
at 94.degree. C., 30 sec. at 55.degree. C., 33 sec. at 72.degree.
C.) in a Perkin Elmer thermocycler (Gene Amp PCR System 9600).
RESULTS
[0156] rAAV was initially produced following the protocol described
by Snyder et al. (1996). Briefly, twenty five 15-cm plates of 293
cells are transfected with two plasmids: one harboring the rep and
cap genes and the other the rAAV vector (FIG. 2A). Six hours after
transfection, cells are infected with adenovirus (wild type, or
Ad.dts). When a cytopathic effect appeared, cells were harvested,
lysed and extracts were purified through two cesium chloride
gradients. A first technical change was introduced in the
purification step on the CsCl gradient, reducing the centrifugation
time down to six hours to reach equilibrium. This modification did
not affect the rAAV yields but instead increased the final volume
of virus. Furthermore, beside this technical improvement, several
additional major modifications were introduced in the rAAV
production procedure (FIG. 2): the first one concerns the titration
method employed to measure the number of infectious particles
produced; the second one is related to the use of different rep-cap
expression constructs; the third modification concerns the use of
an adenoviral plasmid instead of an adenoviral particle to provide
helper functions needed for rAAV replication and assembly; the
fourth modification relates to the discovery of RES sequence and
their introduction into rAAV vectors to increase the efficiency of
the method. Another preferred characteristic of the present
invention relates to the use of stable rep-cap cell lines to
provide the AAV transcomplementing functions and thus, eliminate
the need for a rep-cap plasmid. Preferably, the stable rep-cap cell
line may also contain part of human papilloma viruses including
E.sub.1, E.sub.6 and E.sub.7 ORFs. More preferably, the use of
stable rep-cap cell lines will require an infectious adenovirus,
wild type or mutant, instead of the adenoviral plasmid, to provide
the helper functions.
[0157] A summary of the 18 large scale rAAV stocks produced is
presented in Table 1. A detailed analysis of these results is
presented below.
[0158] 1. Use of a stable rep-cap HeLa cell clone for titration and
characterization of rAAV stocks
[0159] Quantification of viral DNA by dot blot is generally used to
measure the amount of rAAV particles. This assay however, does not
provide any information about the number of infectious rAAV
particles which relies either on the detection of the rAAV
transducing activity or in the measure of infectious particles in a
Replication Center Assay (RCA). This latter method is particularly
interesting since it does not depend upon the expression of the
transgene but only upon the ability of rAAV to infect a target cell
(generally 293 cells) and to replicate its genome in the presence
of adenovirus and Rep proteins. In the originally described RCA
(McLaughlin et al., 1988; Yakobson et al., 1987), Rep proteins are
provided by adding wtAAV which requires a restricted area in order
to prevent any contamination. Our rationale was to circumvent the
use of wtAAV in this assay by developing a stable cell clone
expressing Rep proteins. A stable HeLa cell clone with two
integrated copies of an ITR-deleted AAV genome (HeLaRC32 cells) was
thus generated. In the modified RCA, HeLaRC32 or control HeLa cells
are infected with different dilutions of rAAV in the presence or in
the absence of wild type adenovirus, and individually analyzed for
the presence of replicating rAAV using a transgene probe. A typical
result obtained using this assay is presented in FIG. 3 where the
number of spots obtained with the HeLaRC32 cells infected with rAAV
and adenovirus can be translated as the number of infectious rAAV
particles; and the number of spots obtained with the HeLaRC32 cells
infected with rAAV, but in the absence of adenovirus, represents
the number of contaminating infectious adenoviral particles. As
expected, no signal was detected when HeLaRC32 cells were infected
with adenovirus alone. Finally, when control HeLa cells were
infected with rAAV and adenovirus, some signal was always detected
suggesting that rAAV was amplified in these cells (FIG. 3). A
plausible interpretation of this result is that rAAV stocks are
contaminated with rep-positive AAV particles. This last result was
thus used to measure the number of rep-positive AAV particles in
all the large scale rAAV stocks. The development of this modified
RCA allowed us to fully characterize the effect of each
modification introduced in the rAAV production protocol as
described below.
[0160] 2. Evaluation of different rep-cap expressing plasmids for
rAAV production
[0161] The pAAV/Ad plasmid, described in the standard rAAV
production method (Samulski et al., 1989), harbors the AAV rep-cap
sequences flanked by the adenovirus ITRs (FIG. 1A). After producing
three rAAV stocks with the pAAV/Ad plasmid, the following rAAV
stocks were produced using the pspRC construct which harbors only
the rep-cap sequences in a psp72 backbone (FIG. 1A). The use of the
pspRC plasmid for large scale rAAV production did not affect the
total number of physical particles recovered, which ranged between
10.sup.11 and 5.times.10.sup.12 total particles, but instead
increased the infectious particles yields, as measured by RCA
(Table 1 and FIG. 4). Indeed, particles to infectious particles
ratios ranged between 10.sup.3 and 10.sup.4 when the pAAV/Ad was
the trans-complementing plasmid, whereas, rAAV stocks produced with
the pspRC plasmid consistently showed a ratio below 50 and this
independently of the adenoviral helper functions provided
(adenoviral virions or adenoviral plasmid).
[0162] Other rep-cap constructs were further tested on small scale
rAAV productions performed on two 15-cm plates of 293 cells. The
pspRC plasmid was compared to two other constructs: i) plasmid
plM45 (Pereira et al., 1997) which harbors approximately the same
rep-cap region extended on the 5' end of 45 bp from wtAAV (FIG.
1A); ii) plasmid pspRCC which harbors the rep ORF under the control
of the native AAV promoters followed by the cap ORF under the
transcriptional control of the CMV promoter (FIG. 1A). This last
construct was used to test whether expression of the cap gene under
the control of a strong heterologous promoter can increase the
viral titer as recently reported (Vincent et al., 1997). rAAV
produced on small scale experiments was purified on a CsCl gradient
and characterized by dot blot, RCA and by a LacZ Forming Unit assay
(LFU) on HeLa cells. As indicated in Table 2, approximately the
same number of particles was measured by dot blot. Similarly, the
number of infectious or transducing particles was not significantly
affected by the rep-cap plasmid used.
[0163] 3. Use of an adenoviral plasmid for rAAV production
[0164] Recombinant AAV stocks were initially produced using either
wtAd5 or Ad.dts adenovirus. None of these virions did actually make
any difference in terms of rAAV production (Table 1). Subsequently,
we tested whether adenoviral virions could be replaced by plasmid
pAdc harboring the complete adenoviral genome from wtAd5 (FIG. 1C).
The production protocol, thus, consisted in an initial transfection
of three plasmids (FIG. 2B): the rAAV vector plasmid, the rep-cap
plasmid and the adenoviral plasmid (1:1:2 ratio). No cytopathic
effect was observed even five days after transfection.
Consequently, to purify rAAV, cells were usually scraped from the
plates three days after transfection and processed as previously
described.
[0165] Titration of rAAV stocks, produced using this new procedure,
indicated that replacement of adenoviral virions by the pAdc
plasmid did not affect the particles yields which ranged between
10.sup.11 and 5.times.10.sup.12 total particles (Table 1 and FIG.
4). Similarly, the infectious particles yields remained unchanged
(FIG. 4), although some variability in the production yields using
the pAdc plasmid was observed, essentially because of some
variability in transfection efficiencies.
[0166] We next compared the pAdc plasmid with plasmid pAd.DELTA.
which harbors an adenoviral genome lacking the two ITRs, the
packaging signal and the E1 region (FIG. 1C). Two 15-cm plates of
293 were transfected with the rAAV vector plasmid, the pspRC
plasmid and either the pAdc or the pAd.DELTA. constructs. To
determine the optimal harvesting time for rAAV production, the
cells were collected at 72, 96 and 120 hours post-transfection.
Cellular extracts were purified on a CsCl gradient and the results
of the dot blot and of the RCA are presented in Table 3. The same
rAAV titer ranging between 10.sup.10 and 5.times.10.sup.10
infectious particles/ml was obtained with both plasmids. Incubation
of the cells for more than three days did not significantly affect
the particles/ml titer, however, a slight reduction in the
infectious particles yields was observed.
[0167] Both the pAdc and the pAd.DELTA. adenoviral plasmids,
generated a slight band observed at equilibrium in the CsCl
gradients at a position similar to mature adenoviral particles. We
were not able, however, to detect plaques after incubation of 293
cells with an aliquot of the band obtained with the pAdc plasmid.
Furthermore, and more importantly, no adenoviral contamination was
observed in most of the large scale rAAV stocks (7/9) produced
using the pAdc plasmid as detected by RCA (Table 1). Also, the low
level of adenoviral contamination (2.5 10.sup.3 i.p./ml) observed
in two rAAV stocks has not yet been confirmed by other methods.
[0168] 4. Detection of rep-positive AAV in the rAAV stocks
[0169] The rAAV stocks were tested in a modified RCA for the
presence of contaminating rep-positive AAV. Infection of control
HeLa cells with different dilutions of rAAV and adenovirus resulted
in the detection of some level of replicating rAAV (detected with a
transgene probe) in all the rAAV stocks (FIG. 3). This suggested
that some Rep activity had been transferred to these cells. This
level of contamination ranged between 10.sup.4 and 5.times.10.sup.7
infectious rep-positive AAV particles/ml and this independently of
the use of adenovirus or of an adenoviral plasmid (Table 1). To
substantiate this result, 293 cells were infected with
rAAVCMVnlsLacZ, produced either with adenovirus or the pAdc
plasmid, in the absence or in the presence of Ad.dl324 and low
molecular weight DNA was analyzed on a Southern blot using a LacZ
probe. As shown in FIG. 5, the typical AAV replicative forms were
detected when cells were co-infected with adenovirus, whereas only
input single stranded DNA was seen in cells infected with
rAAVCMVnlsLacZ alone. This result suggested that particles
containing rep sequences and/or Rep proteins are present in the
rAAV stocks. To check for the presence of rep sequences, DNA was
extracted from 10 .mu.l of a rAAV stock (approximately 10.sup.9
particles) and analyzed by PCR using primers located in the p5
promoter and the rep gene (see Materials and Methods). As expected,
an amplification product was detected in all the rAAV stocks (data
not shown). Altogether, these data suggest that some rep DNA is
packaged during rAAV assembly in 293 cells. Furthermore, the
results presented show that the use of a rep-cap plasmid devoid of
adenoviral ITRs according to the present invention significantly
reduces the contamination by rep.sup.+ AAVs in the stocks
obtained.
[0170] 5. Identification. cloning and characterization of RES
sequences
[0171] This example discloses the discovery, cloning and
characterization of non-ITR sequences which are implicated in cis
in the encapsidation of the AAV genome into AAV capsids. This
example also illustrates how these sequences can be used to package
heterologous polynucleotides (free of functional AAV ITRs) into AAV
capsids and to increase the yields of rAAV production methods.
[0172] 5.1. Encapsidation of AAV rep-cap genome in the absence of
the viral ITRs.
[0173] To demonstrate that a rep-cap genome can be packaged in the
absence of the AAV ITRs, 293 cells were transfected with a plasmid
harboring an ITR-deleted rep-cap genome associated with or without
the adenovirus ITRs (FIG. 7, plasmids dITR-RC and BS-RC
respectively). Cells were then infected with adenovirus, and AAV
particles were purified on a CsCl gradient. After treatment of the
particles with DNase I, the viral DNA packaged inside the particles
was extracted and analyzed on a Southern blot using either a rep or
an adenovirus ITR probe.
[0174] Using a rep probe, a hybridization signal was detected with
viral DNA extracted from viral preparations produced using both
plasmids (BS-RC or dITR-RC). However, a more intense signal was
observed using the dITR.RC plasmid. The position and the pattern of
this hybridization signal on the Southern blot was similar to that
observed in the case of single-stranded viral DNA. This result
indicated that a rep-cap genome can be found associated with AAV
capsids in the absence of the AAV ITRs.
[0175] When the adenovirus ITR probe was used, a specific band
migrating at the same position as that observed using a rep probe
was observed only for the preparation made with the dITR.RC
construct. This result indicates that in this case, the rep-cap
genome found associated with AAV capsids also harbored the
adenovirus ITRs.
[0176] Both these results indicate that a rep-cap genome can be
packaged into AAV capsids despite the absence of the AAV ITRs, thus
suggesting the presence of specific packaging signals in the
rep-cap sequence.
[0177] 5.2. Identification and cloning of viral sequences involved,
in cis, in the encapsidation of an ITR-deleted rep-cap genome.
[0178] To be packaged, the rep-cap genome has first to be
replicated in order to generate single-stranded DNA which is the
form encapsidated into AAV capsids. In the case of wild-type AAV,
replication is initiated from the viral ITRs which harbor the
binding element for Rep78168 (RBE) and its nicking site (terminal
resolution site or trs). Interestingly, a Rep binding element
associated with a cryptic trs is present in the p5 promoter,
approximately 50 bp downstream from the 5' ITR. Moreover, we
discovered in this same region the presence of a putative
palindromic structure upstream to the RBE and trs (FIG. 8B)
(nucleotides 231-250 in SEQ ID NO:8). The respective position of
these three elements (palindrome, RBE and TRs) is similar to that
observed in the AAV ITR. These observations led us to concentrate
on the role of these three elements of the p5 promoter in the
encapsidation of an ITR-deleted rep-cap genome.
[0179] These three elements (palindrome, RBE and trs) were first
deleted from the plasmid dITR-RC by removing a 350 bp region
extending from the 5' end of the p5 promoter to the beginning of
the rep coding region (nucleotides 190 to 540 of wtAAV). This new
construct, named dITR.RCA (FIG. 7) was transfected into 293 cells
which were subsequently infected with adenovirus. Because Rep
proteins could not be produced by the dITR.RCA construct, they were
supplied by co-transfection of the BS-RC plasmid which harbors the
entire rep-cap coding region. The analysis of viral DNA packaged
into AAV capsids was performed by Southern blot using an adenovirus
ITR probe. Opposite to what was observed with the dlTR.RC plasmid,
no signal was detected using the dITR.RCA construct. These results
indicated that the 350 bp deleted from the rep-cap genome contains
essential elements for its replication and/or encapsidation in the
absence of the AAV ITRs. This 350 bp region was named RES
(Replication Encapsidation Sequence).
[0180] 5.3. Use of RES to package a polynucleotide into AAV
capsids
[0181] The previous experiment demonstrated that the deletion of
the RES element prevents the encapsidation of the rep-cap genome.
We next investigated whether the RES element could confer to any
heterologous sequence the ability to be packaged into AAV
capsids.
[0182] Three constructs were made (FIG. 9):
[0183] pCMV-LacZ, harboring only the LacZ expression cassette;
[0184] pRES.sup.+/CMV-LacZ, harboring the LacZ cassette and a RES
element, in the sense orientation, inserted upstream of the CMV
promoter;
[0185] pRES.sup.-/CMV-LacZ, harboring the LacZ cassette and a RES
element, in the antisense orientation, inserted upstream of the CMV
promoter.
[0186] To determine if any of these constructs could be packaged
into AAV capsids, they were individually transfected into HeLaRC32
cells which have integrated into their genome a copy of the
ITR-deleted AAV genome (see Materials and Methods). These cells,
which provide Rep and Cap functions, were used to limit the
recombination events between the plasmids harboring the RES
elements and the rep-cap genome. After transfection, the cells were
infected with adenovirus, and AAV particles were purified on a CsCl
gradient. The DNA packaged into AAV capsids was analyzed on a
Southern blot using a LacZ probe. The results obtained indicated
the absence of a specific signal when cells were transfected with
plasmid pCMV-LacZ. In contrast, a band hybridizing to the LacZ
probe was specifically detected when cells were transfected with
plasmid pRES.sup.+/CMV-LacZ and pRES.sup.-/CMV-LacZ. This result
indicated that the RES element can confer to the CMV-LacZ sequence
the ability to be packaged into AAV capsids.
[0187] To further confirm this observation, AAV particles obtained
using each of the three plasmids (pCMV-LacZ, pRES.sup.+/CMV-LacZ
and pRES.sup.-/CMV-LacZ) were used to infect HeLa cells in the
presence of adenovirus. Twenty-four hours after infection, the
cells were fixed and stained with X-Gal to detect cells expressing
the LacZ gene. No blue cells were observed when using the
pCMV-LacZ-derived viral preparations. In contrast blue cells were
seen when using viral stocks produced using the pRES.sup.+/CMV-LacZ
and pRES.sup.-/CMV-LacZ plasmids. This result indicated that
infectious particles that could transfer the LacZ expression
cassette were present in the preparations (30 LacZ-forming units
total were obtained with each plasmid for 6.times.10.sup.7
transfected cells). This observation further implies that the
single-stranded DNA packaged into these particles had been
converted into a transcriptionally active double-stranded form. In
the case of wild-type AAV or of recombinant AAV, this event is
primarily mediated by the presence of the AAV ITRs which, in their
palindromic conformation, provide a 3' OH free end necessary to
start the polymerization process. A similar mechanism might be
mediated by the small palindrome in the RES.sup.+/CMV-LacZ and
RES.sup.-/CMV-LacZ genomes. An alternative possibility would be
that the double-stranded DNA form was generated by the annealing of
two complementary single-stranded molecules as observed in cells
infected with wtAAV.
[0188] In conclusion, our observations demonstrate that the RES
element can confer to an heterologous sequence the ability to be
packaged into AAV particles and that at least a portion of these
particles are infectious.
[0189] 5.4. Use of RES sequences to improve rAAV yields.
[0190] Since the RES sequence can mediate the replication and/or
encapsidation of any nucleic acid into AAV particles, it can also
be used to increase the packaging efficiency of rAAV vector
plasmids (i.e., harboring AAV ITRs) and thus, to improve the yield
of infectious recombinant AAV particles. For that purpose, one or
several RES elements are inserted into an AAV vector
(ITR-transgene-ITR) to analyze its effect in the context of the AAV
ITRs. Indeed, it is expected that the RES element cooperates with
the AAV ITRs to optimize the encapsidation of single-stranded DNA
into AAV capsids. AAV particles comprising either a standard AAV
vector (ITR-transgene-ITR) or a RES-containing vector
(ITR-RES-transgene-ITR) are produced using adenovirus-infected
HeLaRC32 cells. The yield of infectious particles produced using
the two kinds of vector is measured and compared following standard
procedures.
[0191] The introduction of the RES element, in a sense or antisense
orientation, improves the rAAV physical titer yield 5 to 10
fold.
[0192] 5.5. Constructions of RES variants
[0193] This section illustrates the construction of RES elements
having, for instance, reduced size, while retaining their packaging
activity.
[0194] Deletions and point mutations of the RES element are
introduced into the dITR.RC construct and their effect on its
replicative activity and on its encapsidation are analyzed. The
mutations introduced into the RES element are also analyzed for
their effect on the encapsidation of a CMV-LacZ cassette. More
specifically, the following genetically modified RES sequences are
prepared (FIG. 10):
[0195] RES 190-350
[0196] RES 281-540
[0197] RES 350-540
[0198] RES 255-540
[0199] RES 190-255/350-540
[0200] RES 190-281/350-540
[0201] These variants are prepared by conventional techniques,
using restriction enzymes and ligases. The nucleotide positions are
given with reference to the sequence of FIG. 8 (SEQ ID NO: 8 and
9).
[0202] Additional modified RES sequences are produced, comprising
one or more mutations, as compared to the sequence of FIG. 2. More
preferably, these RES variants comprise mutations outside of the
functional palindromic, RBE and trs domains. Particular examples of
mutated elements are represented in FIG. 10.
[0203] 5. 6. Design of new trans-complementing rep-cap plasmids
able to provide the viral functions for particle formation but
unable to recombine with the new RES+AAV vectors
(ITR-RES-transgene-ITR).
[0204] In order to further avoid any potential homologous
recombination events between the RES element of the rAAV plasmid
vector and the rep gene present in the rep-cap plasmids, the
following constructions are made.
[0205] 5.6.0 Introduction of the RES sequence in an antisense
orientation.
[0206] The improved packaging is maintained and no rep-positive
particles are generated.
[0207] 5.6.1 Deletion of the RES minimal elements from the rep-cap
genome.
[0208] If no deleterious effects are observed on the expression of
the rep and cap genes, a first construction is a rep-cap plasmid
devoid of any overlapping region with RES. For instance, a rep-cap
plasmid having a rep-cap unit with a 5' end starting at nucleotide
position 300 of the AAV genome can be prepared, and used to produce
a rAAV comprising a 5' ITR-RES unit consisting of nucleotides 1-299
of the AAV genome. Such rep-cap plasmids usually comprise a
heterologous promoter instead of p5.
[0209] 5.6.2 Use of non-homologous RES sequences in the rAAV
vector.
[0210] A second approach is based on the use of altered RES
elements lacking significant homology with the rep-cap plasmid. For
that purpose, mutation variants of RES can be prepared as described
in Example 5.5, having reduced sequence homology with any
potentially overlapping region present in the rep-cap plasmid.
Also, artificial RES elements can be prepared by removing most
non-essential sequences.
[0211] As a further alternative, non-homologous RES sequences can
also be used, i.e., RES elements derived from other parvoviruses,
with no significant sequence homology to the AAV genome, but which
are functional.
[0212] 6. Assay of rAAV transducing activity in vitro and vivo
[0213] To relate the RCA and the dot blot titers to the in vitro
transducing activity of rAAV, we used a vector encoding the nlsLacZ
gene under the control of the CMV promoter. Two large scale stocks
of rAAVCMVnlsLacz, produced either with adenovirus (vAd) or with
the pAdc adenoviral plasmid, were assayed in vitro in an infectious
LacZ Forming Unit assay (LFU) on HeLa cells (Table 4). Typically, a
ratio ranging from 10 to 50 was observed between the infectious
particles and the LacZ forming units (LFU) measured on HeLa cells
in the presence of adenovirus. This ratio was further increased ten
fold if the LFU assay was performed in the absence of adenovirus.
The particles to LFUs ratio ranges for both rAAV stocks from
4.times.10.sup.2 to 6.times.10.sup.2. These data indicate that, at
least in vitro, rAAV produced with either adenovirus or an
adenoviral plasmid is functional. Further in vitro assays also
indicate that the rAAVCMVnlsLacZ virus produced with the pAdc
plasmid is, as generally described for rAAV, resistant to heat
treatment (30 mn at 56.degree. C.), to repeated freeze and thaw (at
least twice), and is also stable at least three days at 4.degree.
C. (data not shown).
[0214] A different sensitivity between these two assays, the RCA
and the LFU, can explain the discrepancy between the number of
infectious and transducing particles (Couffinhal et al., 1997).
However two other hypothesis can also be evoked: i) not all the
rAAV genomes able to replicate and thus detected in the RCA, can
lead to the production of the LacZ protein 24 hours after
infection; ii) the Rep proteins produced in the HeLaRC32 cells,
used for the RCA but absent in the control HeLa cells used for the
LFU assay, account for this difference. To test this last
hypothesis, rAAVCMVnlsLacZ was used to infect either HeLa or
HeLaRC32 cells in the presence or absence of adenovirus, followed
by an X-Gal staining 24 hours later. LFU titers remained
essentially unchanged using the HeLaRC32 cells as compared with
control cells. This result indicates that even in the presence of
Rep proteins, the rAAV titer as measured by the LFU assay is not
equivalent to the number of infectious particles measured by the
RCA.
[0215] To test the in vivo transducing activity of rAAVCMVnlsLacZ
(produced with the pAdc plasmid), 2.5.times.10.sup.8 infectious
particles (measured by RCA) were injected in the rat tibialis
anterior muscle (three animals were injected). Animals were
sacrificed one month post injection and X-gal staining revealed the
presence of transduced fibers in most of the tissue sections (FIG.
6). Transduction efficiency was evaluated to range between 10-20%
of the fibers.
[0216] Other in vivo data were obtained after injection in the
mouse muscle of rAAV encoding the murine erythropoietin (mEpo)
under the control or not of a doxycycline inducible promoter (Bohl
D. et al., Blood, 92 (1998) 1512). Two different vectors were used
in this study: i) the rAAVCMVEpo/rtTa (produced with adenovirus)
which harbors the mEpo cDNA under the control of the tetO-CMV
promoter, as well as the reverse transactivator (rtTA) under the
control of the Moloney LTR. This vector was injected into the
tibialis anterior muscle of mice (2.1.times.10.sup.9 or
4.2.times.10.sup.9 infectious particles/muscle) and transgene
expression was monitored by measuring Epo secretion and increase of
the hematocrit. The data obtained up to 7 months after injection,
indicate that Epo secretion can be switched on and off depending on
the presence or absence of doxycycline in the drinking water; ii)
Epo secretion was also achieved after injection in the mouse muscle
of rAAVCMVEpo (produced with the pAdc adenoviral plasmid) in which
the mEpo cDNA is placed under the control of the Moloney LTR. In
summary, these in vivo data indicate that rAAV produced with
adenovirus or the adenoviral plasmid, pAdc, is functional.
DISCUSSION
[0217] Because of some unique properties among viral vectors, the
use of rAAV for gene transfer is becoming widespread. However,
efficient production of these vectors still relies upon methods
which are cumbersome and have to be performed at a large scale in
order to obtain a sufficient amount of virus. Furthermore, the need
for helper adenovirus for rAAV assembly, leads to the concomitant
production of adenoviral particles which are difficult to separate
from rAAV particles. The present invention now provides novel
methods and compositions for improved production and
characterization of rAAV stocks.
[0218] A first major technical improvement of this invention is the
centrifugation step needed to purify rAAV. By changing the CsCl
gradient conditions, we considerably shortened the protocol since
equilibrium in the gradient can be reached by centrifuging 6 hours
instead of 48 hours as previously described. Alternatively, or in
combination therewith, the use of chromatography or affinity
columns should further improve the rAAV purification procedure
(Tamayose et al., 1996).
[0219] The second improvement described in this study is the
characterization of rAAV stocks. In many studies, titers, and
consequently m.o.i., are given as genome particles per ml (measured
by dot blot). This parameter does not provide information on the
rAAV infectivity and on the level of contamination with adenovirus
and rep-positive AAV. The lack of this information when performing
in vitro and in vivo experiments, using total cellular extracts or
purified virus, makes any comparative evaluation of rAAV-mediated
gene transfer quite difficult. We developed a general titration
method based on the use of an HeLa rep-cap stable cell line. This
method can be applied to any viral stock produced whichever
transgene is present. It can also be applied to monitor (measure)
viral safety issues in biological fluids, after in vivo
administration of a rAAV preparation in animals and/or human
subjects. It allows the measurement of infectious rAAV particles as
well as of the level of contamination with infectious adenovirus
and rep-positive AAV (FIG. 3). In the titration method described by
Clark et al. (Clark et al., 1996; Clark et al., 1995), replicative
forms were analyzed on a Southern blot, 60 hours after infection of
the HeLa rep-cap cells (that is late in the AAV growth cycle) with
different dilutions of rAAV and adenovirus. In our assay, cells are
individually analyzed (FIG. 3) for the presence of replicating rAAV
DNA 24 hours after infection, which is the minimal time to allow
replication of viral DNA but is short enough to prevent the virus
from being released in the culture medium, spreading to other cells
in the well (Carter, 1990). In addition, in the titration assay
developed by Clark et al. (1996), measurements of the level of
contamination with adenovirus and rep-positive AAV were lacking.
The RCA of the present invention provides for the first time
information regarding (i) infectious rAAV particles, (ii)
adenovirus contamination and (iii) rep-positive rAAV
contamination.
[0220] The number of infectious particles measured by our RCA is
approximately 50 fold higher than the number of transducing
particles as measured for example by the LacZ forming units with
the rAAVCMVnlsLacZ (Table 4). Whether this difference is linked to
the assay used to detect transgene expression or to a general
property of rAAV remains still unclear. It is possible that only
part of the pool of replicating DNA is available for transgene
expression, at least in vitro. Another possibility is the presence
in the rAAV stock of defective interfering genomes which, as
described for wild type AAV, would have internal deletions and
still retain the viral ITRs (Carter et al., 1979; Laughlin et al.,
1979).
[0221] This RCA was used to carefully monitor the effect of two
major modifications introduced in the rAAV production protocol: 1)
the use of different rep-cap expression plasmids; 2) the
replacement of adenovirus by an adenoviral plasmid.
[0222] The study by Samulski et al. (1989) describes the pAAV/Ad
plasmid as more efficient than a plasmid without the adenoviral
ITR. Yields were assessed by looking at rAAV transducing activity
(conferring neomycin resistance) using a non purified cell lysate.
Our results obtained with different rAAV vectors produced on large
scale and purified through a cesium gradient clearly do not confirm
this initial observation. Indeed, surprisingly, not only the total
particles yield was not affected but also the number of infectious
particles was increased when the two adenoviral ITRs were removed
from the rep-cap construct. Indeed, in all the rAAV stocks produced
with the pspRC plasmid, the particles to infectious particles
ratios were always inferior to 50, while they ranged between
10.sup.3 and 10.sup.4 in the rAAV stocks produced with the pAAV/Ad
plasmid (Table 1).
[0223] Rep and Cap protein levels are an important parameter to
consider for rAAV production. In our production protocol, Rep and
Cap proteins can be expressed under the control of either the
native AAV promoters or heterologous promoters. In particular,
expression of rep can be controlled by:
[0224] (i) the native AAV p5 and p19 promoters (pspRC construct).
This configuration was chosen to preserve, as much as possible, the
natural cascade of trans-activations and/or repressions occurring
during AAV life cycle in the presence of adenovirus (Pereira et
al., 1997);
[0225] (ii) a heterologous promoter, to produce low levels of rep
and/or inactivate potential RES sequences present in this
region.
[0226] Expression of Cap proteins can also be regulated either
by:
[0227] (i) the native AAV p40 promoter (pspRC construct), or
[0228] (ii) a heterologous promoter, such as CMV (pspRCC
construct), to provide for a strong expression of Cap proteins.
Indeed, at the opposite of Rep, overexpression of CAP proteins by
using an heterologous promoter has also been shown to increase the
rAAV yield (Vincent et al., 1997).
[0229] Preferred rep-cap constructs, however, comprise native AAV
promoters, optionally modified in order to inactivate potential RES
sequences.
[0230] Another preferred characteristic of the rAAV production
method of this invention resides in the replacement of adenovirus
with an adenoviral plasmid. Two constructs were tested: one
harboring the entire adenoviral genome and the second harboring
deletions of the ITRs, the T and E1 regions (FIG. 1C). Despite the
large size of these plasmids (over 40 kb), rAAV production was not
decreased after transfection into 293 cells. rAAV obtained under
these conditions displayed the same physical properties than rAAV
produced with adenovirus (infectious particles to LFU ratio and
heat stability). Recently, other investigators have also described
the rAAV production using adenoviral plasmids which harbor either
an E1 deleted adenoviral genome or only the minimal adenoviral
functions needed for rAAV production (Grimm D. et al., 1998)
(Ferrari et al.,1997). In our hands, all rAAV stocks but two
obtained using the pAdc plasmid were free of detectable adenoviral
contamination as determined by RCA.
[0231] In vivo experiments using rAAV produced with the pAdc
plasmid indicate that these viruses are competent for transducing
muscle cells in mice and rats. It is possible, however, that this
relatively pure rAAV displays a different kinetic and transduction
efficiency in vivo as compared to rAAV produced with
adenovirus.
[0232] Contamination of the rAAV stocks with rep-positive AAV is a
challenging issue. Indeed, some studies (Allen et al., 1997;
Halbert et al., 1997) have reported that rAAV preparation are
contaminated with such particles revealed either by RCA or by dot
blot using in both cases a rep-cap probe. The extent of
contamination published ranges from 0.0001 to 10% of the rAAV
stocks. In the study of Allen et al. (1997), the contaminating
rep-positive AAV has been characterized following sequential
amplification in adenovirus-infected cells. All rep-positive AAV
genomes sequenced have at least a portion of the AAV ITRs and a non
homologous recombination leading to the insertion of rep-cap
sequences close to AAV ITRs was proposed to explain the emergence
of such contaminant (Allen et al., 1997).
[0233] Our RCA assay indicates that amplification of the vector can
occur in HeLa cells, in the presence of adenovirus. Furthermore,
replicative forms were also detected in low molecular weight DNA
extracted from rAAV and adenovirus-infected 293 cells (FIG. 5).
Both these observations suggest the presence of rep-positive
particles in rAAV stocks. The extent of this contamination,
however, represents on average 0.001% of infectious rAAV particles
(also measured by RCA), which seems much lower than in rAAV stocks
prepared with previous methods.
[0234] Another preferred characteristic of the rAAV production
method of this invention is the ability to improve rAAV titers by
introducing the RES sequence, preferably in an antisense
orientation, which avoids generation of rep-positive particles.
REFERENCES
[0235] ALLEN, J. M., DEBELAK, D. J., REYNOLDS, T. C. AND MILLER, A.
D. (1997). Identification and elimination of replication-competent
Adeno-Associated Virus (AAV) that can arise by non homologous
recombination during AAV vector production. J. Virol. 71,
6816-6822.
[0236] CARTER, B. J. (1990). The growth cycle of adeno-associated
virus. In Handbook of Parvoviruses. P. Tjissen, eds (CRC Press,
Boca Raton, Fla.) pp.155-168.
[0237] CARTER, B. J., LAUGHLIN, C. A., DE LA MAZA, L. M. AND MYERS,
M. (1979). Adeno-associated virus autointerference. Virology 92,
449.
[0238] CLARK, K. R., VOULGAROPOULOU, F. AND JOHNSON, P. R. (1996).
A stable cell line carrying adenovirus-inducible rep and cap genes
allows for infectivity titration of adeno-associated virus vectors.
Gene Ther. 3, 1124-1132.
[0239] CLARK, K. R., VOULGAROPOULOU, F., FRALEY, D. M. AND JOHNSON,
P. R. (1995). Cell lines for the production of recombinant
adeno-associated virus. Hum. Gene Ther. 6, 1329-1341.
[0240] COUFFINHAL, T., KEARNEY, M., SULLIVAN, A., SILVER, M.,
TSURUMI, Y. AND ISNER, J. M. (1997). Histochemical staining
following LacZ gene transfer underestimates transfection
efficiency. Hum. Gene Ther. 8, 929-934.
[0241] FERRARI, F. K., XIAO, X., McCARTHY, D. AND SAMULSKI, R. J.
(1997). New developments in the generation of Ad-free, high-titer
rAAV gene therapy vectors. Nature Med. 3, 1295-1297.
[0242] FISHER, K. J., JOOSS, K., ALSTON, J., YANG, Y., HAECKER, S.
E., HIGH, K., PATHAK, R., RAPER, S. E. AND WILSON, J. M. (1997).
Recombinant adeno-associated virus for muscle directed gene
therapy. Nature Med. 3, 306-312.
[0243] FLANNERY, J. G., ZOLOTUKHIN, S., VAGUERO, M. I., LA VAIL, M.
M., MUZYCZKA, N. AND HAUSWIRTH, W. W. (1997). Efficient
photoreceptor-targeted gene expression in vivo by recombinant
adeno-associated virus. Proc. Natl. Acad. Sci. USA 94,
6916-6921.
[0244] GINSBERG, H. S., LUNDHOLM, U. AND LINNE, T. (1977).
Adenovirus DNA-binding protein in cells infected with wild-type 5
adenovirus and two DNA-minus, temperature-sensitive mutants, H5ts
125 and H5ts 149. J.Virol. 23, 142-151.
[0245] GRAHAM, F. L. AND PREVEC, L. (1991). Manipulation of
adenovirus vectors. In Gene Transfer and Protocol. E. J. Murray,
eds (The Humana Press, Inc., Clifton, N.J.) pp.109-128.
[0246] GRIMM D., KERN A., RITTNER K., KLEINSCHMIDT (1998). Novel
tool for production and purification of recombinant adenoassociated
virus vectors. Hum. Gene. Ther. 9, 2745-2760.
[0247] HALBERT, C. L., STANDAERT, T. A., AITKEN, M. L., ALEXANDER,
I. E., RUSSELL, D. W. AND MILLER, A. D. (1997). Transduction by
Adeno-Associated Virus vectors in the rabbit airway: efficiency,
persistence and readministration. J.Virol. 71, 5932-5941.
[0248] HERZOG, R. W., HAGSTROM, J. N., KUNG, S.-H., TAI, S. J.,
WILSON, J. M., FISHER, K. J. AND HIGH, K. A. (1997). Stable gene
transfer and expression of human blood coagulation factor IX after
intramuscular injection of recombinant adeno-associated virus.
Proc. Natl. Acad. Sci. USA 94, 5804-5809.
[0249] KESSLER, P. D., PODSAKOFF, G. M., CHEN, X., MCQUISTON, S.
A., COLOSI, P. C., MATELIS, L. A., KURTZMAN, G. J. AND BYRNE, B. J.
(1996). Gene delivery to skeletal muscle results in sustained
expression and systemic delivery of a therapeutic protein. Proc.
Natl. Acad. Sci. USA 93, 14082-14087.
[0250] KOEBERL, D. D., ALEXANDER, I. E., HALBERT, C. L., RUSSELL,
D. W. AND MILLER, A. D. (1997). Persistent expression of human
clotting factor IX from mouse liver after intravenous injection of
adeno-associated virus vectors. Proc. Natl. Acad. Sci. USA 94,
1426-1431.
[0251] LAUGHLIN, C. A., MYERS, M. W., RISIN, D. L. AND CARTER, B.
J. (1979). Defective-interfering particles of the human parvovirus
adeno-associated virus. Virology 94, 162.
[0252] LEONARD, C. J. AND BERNS, K. I. (1994). Adeno-associated
virus type 2: a latent life cycle. Prog. Nucleic Acid Res. Mol.
Biol. 48, 29-53.
[0253] LI, J., SAMULSKI, R. J. AND XIAO, X. (1997). Role for highly
regulated rep gene expression in adeno-associated virus vector
production. J.Virol. 71, 5236-5243.
[0254] MCLAUGHLIN, S. K., COLLIS, P., HERMONAT, P. L. AND MUZYCZKA,
N. (1988). Adeno-associated virus general transduction vectors:
analysis of proviral structures. J. Virol. 62, 1963-1973.
[0255] MUZYCZKA, N. (1992). Use of adeno-associated virus as a
general transduction vector for mammalian cells. Curr. Top.
Microbiol. Immunol. 158, 97-129.
[0256] PEREIRA, D. J., McCARTY, D. M. AND MUZYCZKA, N. (1997). The
adeno-associated virus (AAV) rep protein acts as both a repressor
and an activator to regulate AAV transcription during a productive
infection. J.Virol. 71, 1079-1088.
[0257] SAMULSKI, R. J., CHANG, L. S. AND SHENK, T. (1989).
Helper-free stocks of recombinant adeno-associated viruses: normal
integration does not require viral gene expression. J. Virol. 63,
3822-3828.
[0258] SNYDER, R., XIAO, X. AND SAMULSKI, R. J. (1996). Production
of recombinant adeno-associated viral vectors. In Current Protocols
in Human Genetics. N. Dracopoli, J. Haines, B. Krof, D. Moir, C.
Morton, C. Seidman, J. Seidman and D. Smith, eds (John Wiley and
Sons Publisher, New York, N.J.) pp.12.1.1-12.1.23.
[0259] SNYDER, R. O., MIAO, C. H., PATIJN, G. A., SPRATT, S. K.,
DANOS, O., NAGY, D., GOWN, A. M., WINTHER, B., MEUSE, L., COHEN, L.
K., THOMPSON, A. R. AND KAY, M. A. (1997). Persistent and
therapeutic concentrations of human factor IX in mice after hepatic
gene transfer of recombinant AAV vectors. Nature Genet. 16,
270-276.
[0260] TAMAYOSE, K., HIRAI, Y. AND SHIMADA, T. (1996). A new
strategy for large-scale preparation of high-titer recombinant
adeno-associated virus vectors by using packaging cell lines and
sulfonated cellulose column chromatography. Hum. Gene Ther. 7,
507-513.
[0261] VINCENT, K. A., PIRAINO, S. T. AND WADSWORTH, S. C. (1997).
Analysis of recombinant Adeno-Associated Virus packaging and
requirements for rep and cap gene product. J. Virol. 71,
1897-1905.
[0262] XIAO, X., LI, J. AND SAMULSKI, R. J. (1996). Efficient
long-term gene transfer into muscle tissue of immunocompetent mice
by adeno-associated virus vector. J.Virol. 70, 8098-8108.
[0263] YAKOBSON, B., KOCH, T. AND WINOCOUR, E. (1987). Replication
of adeno-associated virus in synchronized cells without the
addition of helper virus. J. Virol. 61, 972-981.
[0264] ZOLOTUKHIN, S., POTTER, M., HAUSWIRTH, W. W., GUY, J. AND
MUZYCZKA, N. (1996). A "humanized" green fluorescent protein cDNA
adapted for high-level expression in mammalian cells. J.Virol. 70,
4646-4654.
1 TABLE 1 rAAV Contaminations.sup.3 size rep-cap Virus or Titer Ad.
rep+ AAV Vol. vector name (b) plasmid plasmid p./ml.sup.1
i.p./ml.sup.2 ratio i.p./ml i.p./ml (ml) AAVCMVLacZ 4873 pAAV/Ad
wtAd5 1.2 10.sup.12 1.0 10.sup.9 1.2 10.sup.3 5 10.sup.4 4.5
10.sup.7 1.6 APACMVGDNF 3000 pAAV/Ad wtAd5 2.5 10.sup.11 1.5
10.sup.7 1.7 10.sup.4 7.5 10.sup.4 5.5 10.sup.5 1.7 AAVCMVGDNF 3000
pAAV/Ad wtAd5 8.0 10.sup.11 2.0 10.sup.6 4.0 10.sup.3 1.0 10.sup.5
4.0 10.sup.7 2.8 AAVCMVLacZ 4873 pspRC wtAd5 1.6 10.sup.12 9.2
10.sup.10 17.3 7 10.sup.3 1.0 10.sup.5 1.1 AAVPGK.beta.GLU 3854
pspRC Ad.dts 3.0 10.sup.11 9.5 10.sup.9 31.5 3.1 10.sup.4 1.0
10.sup.5 3.0 AAVPGKhALD 3565 pspRC Ad.dts 1.4 10.sup.11 1.8
10.sup.10 7.8 1.3 10.sup.4 1.0 10.sup.4 3.1 AAVCMVApoE4 2609 pspRC
wtAd5 4.7 10.sup.10 1.1 10.sup.10 4.2 1.1 10.sup.4 1.5 10.sup.4 3.6
AAVCMVEpo/rtTa 5017 pspRC wtAd5 2.5 10.sup.11 2.1 10.sup.10 11.9
2.3 10.sup.4 1.3 10.sup.4 4.1 AAVPGKnlsLacZ 4712 pspRC wtAd5 2.5
10.sup.11 9.0 10.sup.10 2.8 2.2 10.sup.4 5.0 10.sup.5 4.2
AAVCMVnlsLacZ 4641 pupRC pAdc 5.1 10.sup.11 1.6 10.sup.10 31.8
<5 10.sup.2 4.5 10.sup.5 7.6 AAVLTBApOE 3700 pspRC pAdc 3.0
10.sup.11 3.7 10.sup.10 8.1 2.5 10.sup.3 5.5 10.sup.5 7.6 AAVLTREpo
3310 pspRC pAdc 3.7 10.sup.11 4.2 10.sup.10 8.8 <5 10.sup.2 4.0
10.sup.5 7.2 AAVCMVnlsLacZ* 4641 pspRC pAdc 3.0 10.sup.11 2.5
10.sup.10 12.0 <5 10.sup.3 7.0 10.sup.4 13.4 AAVCMV.beta.GLU
4717 pspRC pAdc 3.8 10.sup.10 1.5 10.sup.10 2.5 <5 10.sup.2 4.5
10.sup.4 6.8 AAVCMVEpo 2238 pspRC pAdc 1.2 10.sup.11 5.4 10.sup.10
2.2 <5 10.sup.2 5.0 10.sup.4 6.9 AAVCMVvEGF 2080 pspRC pAdc 1.8
10.sup.11 7.4 10.sup.10 2.4 <5 10.sup.2 2.5 10.sup.5 7.5
AAVCMVnlsLacZ* 4641 pspRC pAdc 7.8 10.sup.10 5.5 10.sup.10 1.4
<5 10.sup.2 1.5 10.sup.5 14.0 AAVCMV.beta.GLU* 4717 pspRC pAdc
3.4 10.sup.11 6.6 10.sup.10 5.1 2.5 10.sup.3 1.3 10.sup.5 11.4
[0265]
2 TABLE 2 part/ml.sup.1 inf.part./ml.sup.2 LPU/ml.sup.3 pspRC 1.7
10.sup.10 1.2 10.sup.9 2.0 10.sup.6 pIM45 8.5 10.sup.9.sup. 3.2
10.sup.8 6.6 10.sup.5 RepCMVCap 3.4 10.sup.10 5.8 10.sup.8 2.0
10.sup.5
[0266]
3 TABLE 3 72 hrs 96 hrs 120 hrs inf.part/ inf.part/ inf.part/
part/ml.sup.1 ml.sup.2 part/ml.sup.1 ml.sup.2 part/ml.sup.1
ml.sup.2 pAdc 3.5 10.sup.10 1.4 10.sup.9 2.5 10.sup.10 6.0 10.sup.8
2.1 10.sup.10 3.5 10.sup.7 pAd.DELTA. 3.5 10.sup.10 2.2 10.sup.9
2.0 10.sup.10 1.2 10.sup.9 2.5 10.sup.10 1.0 10.sup.8
[0267]
4 TABLE 4 inf. LFU/ml.sup.3 LFU/ml.sup.3 part/ inf. part/
part/ml.sup.1 part/ml.sup.2 (+wtAd5) (-wtAd5) inf. part LFU vAd 1.7
10.sup.11 6.6 10.sup.9 4.1 10.sup.8 ND 25 16 pAdc 3.0 10.sup.1 2.5
10.sup.10 5.0 10.sup.8 1.4 10.sup.7 12 50
[0268]
Sequence CWU 1
1
10 1 18 DNA adeno-associated virus 1 gcccgagtga gcacgcag 18 2 20
DNA adeno-associated virus 2 gcgacaccat gtggtcacgc 20 3 33 DNA
adeno-associated virus 3 gcccgagtga gcacgcaggg tctccatttt gaa 33 4
26 DNA Oligonucleotide 4 atgatttaaa tcaggttggg ctgccg 26 5 26 DNA
Oligonucleotide 5 gctctagatg agcttccacc actgtc 26 6 20 DNA
Oligonucleotide 6 tatttaagcc cgagtgagca 20 7 20 DNA Oligonucleotide
7 aaagttctca ttggtccagt 20 8 20 DNA adeno-associated virus 8
gcgacaccat gtggtcacgc 20 9 18 DNA adeno-associated virus 9
gcccgagtga gcacgcag 18 10 4679 DNA adeno-associated virus 2 10
ttggccactc cctctctgcg cgctcgctcg ctcactgagg ccgggcgacc aaaggtcgcc
60 cgacgcccgg gctttgcccg ggcggcctca gtgagcgagc gagcgcgcag
agagggagtg 120 gccaactcca tcactagggg ttcctggagg ggtggagtcg
tgacgtgaat tacgtcatag 180 ggttagggag gtcctgtatt agaggtcacg
tgagtgtttt gcgacatttt gcgacaccat 240 gtggtcacgc tgggtattta
agcccgagtg agcacgcagg gtctccattt tgaagcggga 300 ggtttgaacg
cgcagccgcc atgccggggt tttacgagat tgtgattaag gtccccagcg 360
accttgacga gcatctgccc ggcatttctg acagctttgt gaactgggtg gccgagaagg
420 aatgggagtt gccgccagat tctgacatgg atctgaatct gattgagcag
gcacccctga 480 ccgtggccga gaagctgcag cgcgactttc tgacggaatg
gcgccgtgtg agtaaggccc 540 cggaggccct tttctttgtg caatttgaga
agggagagag ctacttccac atgcacgtgc 600 tcgtggaaac caccggggtg
aaatccatgg ttttgggacg tttcctgagt cagattcgcg 660 aaaaactgat
tcagagaatt taccgcggga tcgagccgac tttgccaaac tggttcgcgg 720
tcacaaagac cagaaatggc gccggaggcg ggaacaaggt ggtggatgag tgctacatcc
780 ccaattactt gctccccaaa acccagcctg agctccagtg ggcgtggact
aatatggaac 840 agtatttaag cgcctgtttg aatctcacgg agcgtaaacg
gttggtggcg cagcatctga 900 cgcacgtgtc gcagacgcag gagcagaaca
aagagaatca gaatcccaat tctgatgcgc 960 cggtgatcag atcaaaaact
tcagccaggt acatggagct ggtcgggtgg ctcgtggaca 1020 aggggattac
ctcggagaag cagtggatcc aggaggacca ggcctcatac atctccttca 1080
atgcggcctc caactcgcgg tcccaaatca aggctgcctt ggacaatgcg ggaaagatta
1140 tgagcctgac taaaaccgcc cccgactacc tggtgggcca gcagcccgtg
gaggacattt 1200 ccagcaatcg gatttataaa attttggaac taaacgggta
cgatccccaa tatgcggctt 1260 ccgtctttct gggatgggcc acgaaaaagt
tcggcaagag gaacaccatc tggctgtttg 1320 ggcctgcaac taccgggaag
accaacatcg cggaggccat agcccacact gtgcccttct 1380 acgggtgcgt
aaactggacc aatgagaact ttcccttcaa cgactgtgtc gacaagatgg 1440
tgatctggtg ggaggagggg aagatgaccg ccaaggtcgt ggagtcggcc aaagccattc
1500 tcggaggaag caaggtgcgc gtggaccaga aatgcaagtc ctcggcccag
atagacccga 1560 ctcccgtgat cgtcacctcc aacaccaaca tgtgcgccgt
gattgacggg aactcaacga 1620 ccttcgaaca ccagcagccg ttgcaagacc
ggatgttcaa atttgaactc acccgccgtc 1680 tggatcatga ctttgggaag
gtcaccaagc aggaagtcaa agactttttc cggtgggcaa 1740 aggatcacgt
ggttgaggtg gagcatgaat tctacgtcaa aaagggtgga gccaagaaaa 1800
gacccgcccc cagtgacgca gatataagtg agcccaaacg ggtgcgcgag tcagttgcgc
1860 agccatcgac gtcagacgcg gaagcttcga tcaactacgc agacaggtac
caaaacaaat 1920 gttctcgtca cgtgggcatg aatctgatgc tgtttccctg
cagacaatgc gagagaatga 1980 atcagaattc aaatatctgc ttcactcacg
gacagaaaga ctgtttagag tgctttcccg 2040 tgtcagaatc tcaacccgtt
tctgtcgtca aaaaggcgta tcagaaactg tgctacattc 2100 atcatatcat
gggaaaggtg ccagacgctt gcactgcctg cgatctggtc aatgtggatt 2160
tggatgactg catctttgaa caataaatga tttaaatcag gtatggctgc cgatggttat
2220 cttccagatt ggctcgagga cactctctct gaaggaataa gacagtggtg
gaagctcaaa 2280 cctggcccac caccaccaaa gcccgcagag cggcataagg
acgacagcag gggtcttgtg 2340 cttcctgggt acaagtacct cggacccttc
aacggactcg acaagggaga gccggtcaac 2400 gaggcagacg ccgcggccct
cgagcacgac aaagcctacg accggcagct cgacagcgga 2460 gacaacccgt
acctcaagta caaccacgcc gacgcggagt ttcaggagcg ccttaaagaa 2520
gatacgtctt ttgggggcaa cctcggacga gcagtcttcc aggcgaaaaa gagggttctt
2580 gaacctctgg gcctggttga ggaacctgtt aagacggctc cgggaaaaaa
gaggccggta 2640 gagcactctc ctgtggagcc agactcctcc tcgggaaccg
gaaaggcggg ccagcagcct 2700 gcaagaaaaa gattgaattt tggtcagact
ggagacgcag actcagtacc tgacccccag 2760 cctctcggac agccaccagc
agccccctct ggtctgggaa ctaatacgat ggctacaggc 2820 agtggcgcac
caatggcaga caataacgag ggcgccgacg gagtgggtaa ttcctcggga 2880
aattggcatt gcgattccac atggatgggc gacagagtca tcaccaccag cacccgaacc
2940 tgggccctgc ccacctacaa caaccacctc tacaaacaaa tttccagcca
atcaggagcc 3000 tcgaacgaca atcactactt tggctacagc accccttggg
ggtattttga cttcaacaga 3060 ttccactgcc acttttcacc acgtgactgg
caaagactca tcaacaacaa ctggggattc 3120 cgacccaaga gactcaactt
caagctcttt aacattcaag tcaaagaggt cacgcagaat 3180 gacggtacga
cgacgattgc caataacctt accagcacgg ttcaggtgtt tactgactcg 3240
gagtaccagc tcccgtacgt cctcggctcg gcgcatcaag gatgcctccc gccgttccca
3300 gcagacgtct tcatggtgcc acagtatgga tacctcaccc tgaacaacgg
gagtcaggca 3360 gtaggacgct cttcatttta ctgcctggag tactttcctt
ctcagatgct gcgtaccgga 3420 aacaacttta ccttcagcta cacttttgag
gacgttcctt tccacagcag ctacgctcac 3480 agccagagtc tggaccgtct
catgaatcct ctcatcgacc agtacctgta ttacttgagc 3540 agaacaaaca
ctccaagtgg aaccaccacg cagtcaaggc ttcagttttc tcaggccgga 3600
gcgagtgaca ttcgggacca gtctaggaac tggcttcctg gaccctgtta ccgccagcag
3660 cgagtatcaa agacatctgc ggataacaac aacagtgaat actcgtggac
tggagctacc 3720 aagtaccacc tcaatggcag agactctctg gtgaatccgg
gcccggccat ggcaagccac 3780 aaggacgatg aagaaaagtt ttttcctcag
agcggggttc tcatctttgg gaagcaaggc 3840 tcagagaaaa caaatgtgga
cattgaaaag gtcatgatta cagacgaaga ggaaatcagg 3900 acaaccaatc
ccgtggctac ggagcagtat ggttctgtat ctaccaacct ccagagaggc 3960
aacagacaag cagctaccgc agatgtcaac acacaaggcg ttcttccagg catggtctgg
4020 caggacagag atgtgtacct tcaggggccc atctgggcaa agattccaca
cacggacgga 4080 cattttcacc cctctcccct catgggtgga ttcggactta
aacaccctcc tccacagatt 4140 ctcatcaaga acaccccggt acctgcgaat
ccttcgacca ccttcagtgc ggcaaagttt 4200 gcttccttca tcacacagta
ctccacggga caggtcagcg tggagatcga gtgggagctg 4260 cagaaggaaa
acagcaaacg ctggaatccc gaaattcagt acacttccaa ctacaacaag 4320
tctgttaatg tggactttac tgtggacact aatggcgtgt attcagagcc tcgccccatt
4380 ggcaccagat acctgactcg taatctgtaa ttgcttgtta atcaataaac
cgtttaattc 4440 gtttcagttg aactttggtc tctgcgtatt tctttcttat
ctagtttcca tggctacgta 4500 gataagtagc atggcgggtt aatcattaac
tacaaggaac ccctagtgat ggagttggcc 4560 actccctctc tgcgcgctcg
ctcgctcact gaggccgggc gaccaaaggt cgcccgacgc 4620 ccgggctttg
cccgggcggc ctcagtgagc gagcgagcgc gcagagaggg agtggccaa 4679
* * * * *