U.S. patent application number 10/126448 was filed with the patent office on 2003-08-07 for method for diversifying the chemical composition of proteins produced in vivo by genetically disabling the editing function of their aminoacyl trna synthetases.
Invention is credited to Crecy-Lagard, Valerie de, Doring, Volker, Hendrickson, Tamara L., Marliere, Philippe, Nangle, Leslie A., Schimmel, Paul.
Application Number | 20030148422 10/126448 |
Document ID | / |
Family ID | 37401359 |
Filed Date | 2003-08-07 |
United States Patent
Application |
20030148422 |
Kind Code |
A1 |
Doring, Volker ; et
al. |
August 7, 2003 |
Method for diversifying the chemical composition of proteins
produced in vivo by genetically disabling the editing function of
their aminoacyl tRNA synthetases
Abstract
The present invention is directed to a method to diversify the
chemical composition of proteins produced in vivo comprising the
step of disabling, particularly by mutagenesis, the editing
function of one of its aminoacyl tRNA synthetases. The present
invention is also directed to nucleic acid sequences encoding such
mutated aminoacyl tRNA synthetases having their editing site
mutated and capable of mischarging its cognate tRNA with a
noncanonical amino acid. Also described herein is an improved
method for obtaining transformed cells capable of synthetizing in
vivo proteins comprising at least a noncanonical amino acid and
their use for the production of such proteins.
Inventors: |
Doring, Volker; (Paris,
FR) ; Nangle, Leslie A.; (San Diego, CA) ;
Hendrickson, Tamara L.; (Baltimore, MD) ;
Crecy-Lagard, Valerie de; (La Jolla, CA) ; Schimmel,
Paul; (La Jolla, CA) ; Marliere, Philippe;
(Etiolles, FR) |
Correspondence
Address: |
BOZICEVIC, FIELD & FRANCIS LLP
200 MIDDLEFIELD RD
SUITE 200
MENLO PARK
CA
94025
US
|
Family ID: |
37401359 |
Appl. No.: |
10/126448 |
Filed: |
April 19, 2002 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
60285495 |
Apr 19, 2001 |
|
|
|
Current U.S.
Class: |
435/69.1 ;
435/252.3; 435/254.2; 435/320.1; 435/325; 435/455; 435/471;
536/23.1 |
Current CPC
Class: |
C12N 9/93 20130101; C12N
15/10 20130101; C12R 2001/19 20210501; C12N 1/205 20210501; C12P
21/02 20130101; C12N 15/67 20130101 |
Class at
Publication: |
435/69.1 ;
435/455; 435/471; 435/320.1; 435/252.3; 435/254.2; 435/325;
536/23.1 |
International
Class: |
C12P 021/02; C12N
001/21; C12N 005/06; C07H 021/04; C12N 001/18; C12N 015/74; C12N
015/85 |
Claims
We claim:
1. A method to diversify the chemical composition of proteins
produced in vivo by a cell comprising the step of disabling the
editing function of at least one aminoacyl tRNA synthetase of the
cell.
2. A method for producing in vivo proteins comprising at least one
noncanonical amino acid comprising the step of: a) selecting a cell
strain wherein the editing function of at least one aminoacyl tRNA
synthetase of the cell has been disabled by mutagenesis, said
disabled editing function allowing the aminoacyl tRNA synthetase to
mischarge the cognate tRNA with said at least one noncanonical
amino acid; b) culturing the selected strain in a culture medium
comprising said noncanonical amino acid, or one of its precursors,
under conditions favorable for the growth of said strain; and c)
recovering from the culture medium or from the cells obtained in
step b) the proteins containing said noncanonical amino acid.
3. The method according to claim 1 or 2, wherein the cell strain
comprises a mutation in the DNA sequence encoding the editon domain
of said disabled aminoacyl tRNA synthetase compared to the wild
type aminoacyl tRNA synthetase coding sequence.
4. The method according to claim 3, wherein said DNA mutation leads
to at least an amino acid mutation in the editing domain of said
aminoacyl tRNA synthetase.
5. The method according to claim 3, wherein said DNA mutation leads
to at least an amino acid substitution in the editing domain of
said aminoacyl tRNA synthetase.
6. The method according to claim 5, wherein said mutation leads to
a single amino acid substitution in the editing domain of said
aminoacyl tRNA synthetase.
7. The method according to one of claims 1 or 2, wherein the
disabled aminoacyl tRNA synthetase is capable of mischarging its
cognate tRNA with a canonical amino acid sterically similar to the
amino acid charged by the wild type aminoacyl tRNA synthetase.
8. The method according to one of claims 1 or 2, wherein the
disabled aminoacyl tRNA synthetase is capable of mischarging its
cognate tRNA with a noncanonical amino acid sterically similar to
the amino acid charged by the wild type aminoacyl tRNA synthetase
on its cognate tRNA.
9. A method for obtaining cells capable of producing in vivo
proteins comprising at least one noncanonical amino acid comprising
the step of mutagenizing the DNA sequence encoding the editing
domain of an aminoacyl tRNA synthetase in a cell, said mutagenizing
leading to an aminoacyl tRNA synthetase variant having an amino
acid mutation in its editing domain and said mutation allowing the
aminoacyl tRNA synthetase variant to mischarge its cognate tRNA
with said at least one noncanonical amino acid.
10. A method for obtaining cells capable of producing in vivo
proteins comprising at least one noncanonical amino acid according
to claim 9, comprising the steps of: a) assaying the ability of an
aminoacyl tRNA synthetase variant having an amino acid mutation in
its editing domain for its ability to mischarge its cognate tRNA
with a noncanonical amino acid; b) mutagenizing the DNA sequence
encoding the editing domain of said aminoacyl tRNA synthetase in a
cell, said mutagenizing leading to the aminoacyl tRNA synthetase
variant assayed in step a) and capable of producing detectable
noncanonical amino acid mischarging; c) optionally, identifying,
selecting and/or cloning the cells containing such aminoacyl tRNA
synthetase variant having the ability to mischarge one noncanonical
amino acid.
11. The method of one of claims 1, 2, 9 or 10, wherein the cell is
a microbial or animal cell.
12. The method of claim 11, wherein said microbial is a bacterium,
a yeast or a fungus.
13. The method of claim 12, wherein said bacterium is Escherichia
coli or Acinetobacter.
14. The method of one of claims 1, 2, 9 or 10, wherein the editing
domain of the aminoacyl tRNA synthetase has been disabled or
mutagenized by homologous recombination.
15. The method of one of claims 1, 2, 9 or 10, wherein the editing
domain of the aminoacyl tRNA synthetase has been mutagenized by
recombination into the genome of the target cell using an allelic
replacement vector.
16. A method for selecting an aminoacyl tRNA synthetase variant
capable of mischarging its cognate tRNA with a noncanonical amino
acid, comprising the steps of: a) elaborating a DNA construct
comprising a DNA sequence encoding an aminoacyl tRNA synthetase
variant wherein said aminoacyl tRNA synthetase variant has at least
an amino acid mutation in its editing domain compared with the wild
type aminoacyl tRNA synthetase; b) transforming a host cell with
the DNA construct of step a) and, after a step of culturing said
transformed host cell, recovering and, optionally, purifying the
recombinant aminoacyl tRNA synthetase variant expressed by the host
cell; c) assaying the ability of the recombinant aminoacyl tRNA
synthetase variant recovered in step b) for its ability to
mischarge its cognate TRNA with a noncanonical amino acid; and d)
selecting said aminoacyl tRNA synthetase variant if detectable
mischarging has been produced in step c).
17. Isolated aminoacyl tRNA synthetase variant capable of
mischarging its cognate tRNA with a noncanonical amino acid,
obtainable by the method of claim 16.
18. Isolated nucleic acid sequence encoded the aminoacyl tRNA
synthetase variant of claim 17.
19. Vector comprising a nucleic acid of claim 18.
20. Transformed cell comprising a nucleic acid encoding an
aminoacyl tRNA synthetase variant of claim 17.
21. Method for the production of proteins comprising a noncanonical
amino acid characterized in that said method comprises the
following steps: a) culturing a cell of claim 20 in a culture
medium comprising the noncanonical amino acid, or one of its
precursors, capable of being mischarged by the cognate tRNA of the
aminoacyl tRNA synthetase variant contained in said cell; and b)
recovering the proteins comprising said noncanonical amino acid
from the culture medium or from the cells of step a).
22. An isolated protein comprising a noncanonical amino acid
obtained by the method of claims 1, 2 or 21.
23. The method of one of claims 1, 2, 9, 10 or 21, wherein the
aminoacyl tRNA synthetase is an aminoacyl tRNA synthetase selected
from the group consisting of the aminoacyl tRNA synthetase
valyl-tRNA synthetase, isoleucyl-tRNA synthetase, leucyl-tRNA
synthetase, alanyl-tRNA synthetase, prolyl-tRNA synthetase,
threonyl-tRNA synthetase, phenyl-tRNA synthetase and lysyl-tRNA
synthetase.
Description
CROSS-REFERENCE TO RELATED APPLICATIONS
[0001] This application claims the benefit of earlier-filed U.S.
provisional application serial No. 60/285,495, filed Apr. 19, 2001,
which application is incorporated herein by reference in its
entirety.
FIELD OF THE INVENTION
[0002] The present invention is directed to a method to diversify
the chemical composition of proteins produced in vivo, especially
to methods comprising the step of disabling, particularly by
mutagenesis, the editing function of one of its aminoacyl tRNA
synthetases. The present invention is also directed to nucleic acid
sequences encoding such mutated aminoacyl tRNA synthetases having
their editing site mutated and capable of mischarging its cognate
tRNA with a noncanonical amino acid. Also described herein is an
improved method for obtaining transformed cells capable of
synthesizing in vivo proteins comprising at least a noncanonical
amino acid and their use for the production of such proteins.
BACKGROUND OF THE INVENTION
[0003] Aminoacyl tRNA synthetases establish the rules of the
genetic code by catalyzing the aminoacylation of transfer RNAs. The
chemical invariance of the twenty amino acid building blocks of
proteins is well established. The only known extensions to this
invariant set are formyl-methionine (1) and selenocysteine (2),
both incorporated in response to punctuation signals during
translation in certain organisms. Thus, although species have
colonized dissimilar terrestrial habitats throughout geological
times, this diversification has not been mirrored in the evolution
of organisms to include specialized sets of amino acids. For
instance, thermophilic, mesophilic, and psychrophilic organisms all
assemble proteins by combining the same types of twenty canonical
amino acids into different protein sequences. Standing as the
"missing link" between alanine and valine (3), aminobutyrate (Abu,
also known as butyrine) can be generated by transamination from the
physiological metabolite 2-oxo-butyrate and should thus be
considered as a latent metabolite (4). Its absence is therefore
particularly conspicuous in the proteins of extant organisms.
[0004] The selection of amino acids for protein synthesis is done
by aminoacyl tRNA synthetases. Typically, each of twenty
synthetases catalyzes the attachment of its cognate amino acid to
the 3'-end of its cognate tRNA and amino acids are, in this way,
associated with specific triplets of the genetic code (5). The
active site of several of these enzymes inherently lack the
capacity to discriminate between closely similar amino acids at a
level sufficient to explain the high accuracy of the code. For that
reason, a given enzyme may misactivate closely similar (in size and
shape) amino acids at a low frequency (0.1 to 1%) (6). To correct
these errors, in many cases, a hydrolytic editing function, at a
separate active site, has developed (7-10). One example of a
synthetase that has editing activity is valyl-tRNA synthetase
(ValRS), which misactivates the isosteric natural amino acid Thr
(9), as well as the non-natural Abu Val (11). Misactivation of
these amino acids leads to transient mischarging of tRNA followed
by hydrolytic deacylation (editing) of the mischarged amino acid
from the tRNA.
[0005] The present work aimed to establish conditions of artificial
selection that promoted usage of non-canonical amino acids, such as
Abu, that were not retained by natural selection. Others have
attempted to incorporate a non-canonical amino acid into a protein
by introducing a foreign, "orthogonal" tRNA/synthetase pair that
can insert the amino acid at a specialized stop codon (12).
[0006] However, such approaches are laborious, as they require
selection, identification, cloning, and study of individual mutant
strains.
[0007] In order to facilitate the in vivo production of proteins
comprising noncanonical amino acids, it would be desirable to have
a rapid and generalized method allowing to genetically modify and
select cells capable of achieving the in vivo production of such
proteins.
[0008] Such a desirable method will allow to enlarge the chemistry
of translation by having a non-canonical amino acid "infiltrate"
all of the codons normally associated with one of the natural amino
acids. Indeed, by assigning two amino acids (a cognate and a
non-cognate) to a specific set of codons so as to provide a
selective advantage to the reprogrammed cells, global changes in
the amino acid compositions of all cellular proteins could be made.
The present invention addresses this need.
SUMMARY OF THE INVENTION
[0009] The present invention provides a general method to diversify
the chemical composition of proteins produced in vivo by a cell
comprising the step wherein the editing function of an aminoacyl
tRNA synthetase of said cell has been disabled.
[0010] This rapid and general method may be used to obtain cells
comprising a mutation in the DNA sequence encoding the editon
domain of said disabled aminoacyl tRNA synthetase compared to the
wild type aminoacyl tRNA synthetase coding sequence.
[0011] In general, the method of the present invention includes: a)
selecting a cell strain wherein the editing function of at least
one of the cell's aminoacyl tRNA synthetases has been disabled,
said disabled editing function allowing the aminoacyl tRNA
synthetase to mischarge the cognate tRNA with said at least one
noncanonical amino acid; b) culturing the selected strain in a
culture medium comprising said noncanonical amino acid, or one of
its precursor, under conditions favourable for the growth of said
strain; and c) recovering from the culture medium or from the cells
obtained in step b) the proteins containing said noncanonical amino
acid.
[0012] In various embodiments, the editing function of the
aminoacyl tRNA synthetases has been disabled by mutagenizing the
DNA sequence encoding the editing domain of an aminoacyl tRNA
synthetase in the target cell, said mutagenesis being carried out
in the cell preferably by homologous recombination or allele
replacement vector leading to an aminoacyl tRNA synthetase variant
having an amino acid mutation in its editing domain, said mutation
allowing the aminoacyl tRNA synthetase variant to mischarge its
cognate tRNA with one noncanonical amino acid.
[0013] In a particular related aspect, the present invention is
directed to a method for selecting an aminoacyl tRNA synthetase
variant capable of mischarging its cognate tRNA with a noncognate
amino acid, preferably a noncanonical amino acid, including the
steps of: a) elaborating a DNA construct encoding an aminoacyl tRNA
synthetase variant having an amino acid mutation in its editing
domain; b) transforming a host cell with said DNA construct; c)
assaying the ability of the recombinant aminoacyl tRNA synthetase
variant produced by said transformed host cell for its ability to
mischarge its cognate tRNA with a noncognate amino acid, preferably
a noncanonical amino acid; and d) if appropriate, selecting the
assayed aminoacyl tRNA synthetase variant if said assayed aminoacyl
tRNA synthetase variant is capable of mischarging its cognate tRNA
(tRNA(s) associated with the assayed aminoacyl tRNA synthetase)
with a noncognate amino acid, preferably with a noncanonical amino
acid.
[0014] In a further aspect, the invention relates to isolated
aminoacyl tRNA synthetase variants capable of mischarging its
cognate tRNA with a noncognate amino acid, preferably with a
noncanonical amino acid, wherein the nucleic fragment encoding the
editing site comprises at least one mutation leading to an amino
acid mutation, preferably an amino acid substitution, in the
editing site of said aminoacyl tRNA synthetase. Isolated nucleic
acid encoding such aminoacyl tRNA synthetase variants, vectors,
such as plasmid, and cell comprising such nucleic acid also form
part of the present invention.
[0015] In another preferred embodiment, the present invention is
directed to a method for the production of proteins comprising a
noncanonical amino acid including the general steps of: a)
culturing, in a culture medium containing said noncanonical amino
acid, or one of its precursors, a transformed host cell comprising
an aminoacyl tRNA synthetase allele variant capable of mischarging
its cognate tRNA with said noncanonical amino acid; and b)
recovering and, if appropriate, purifying the proteins comprising
said noncanonical amino acid from the culture medium (supernatant)
and/or from the cells (from cells pellet) of step a).
[0016] In a final aspect, the invention provides proteins
comprising a noncanonical amino acid obtained by the above
method.
[0017] Other features and advantages of the invention will be
apparent from the following description of the preferred
embodiments thereof, and from the claims.
BRIEF DESCRIPTION OF THE DRAWINGS
[0018] FIGS. 1A to 1C. Suppression and toxicity phenotypes. Amino
acid gradient plates were prepared using minimal medium (27).
[0019] FIG. 1A: A schematic showing the structures of cysteine,
S-carbamoyl-cysteine, valine, threonine and a-aminobutyric
acid.
[0020] FIG. 1B: Photographs showing the results of experiments
demonstrating the cysteine-suppression phenotype. Cysteine (100
.mu.l (0.4 M)) was loaded in a central well after spreading and
drying 0.5 ml of a 5/1000 dilution of an overnight culture (in MS
glucose medium containing thymidine (0.3 mM)) of 15456
(thyA::erm+.DELTA.nrdD::kan+valS:- T222P pTS13 (bla+thyA:146GUA)).
Ile-Val (0.3 mM) was added in one plate as a control. The dipeptide
Ile-Val was purchased from Bachem AG (Bubendorf, Switzerland).
Plates were incubated for 2 days at 30.degree. C.
[0021] FIG. 1C: Photographs showing the results of growth of a
strain carrying the valS:T222P allele in the presence of
L-threonine or Abu. Minimal medium plates supplemented with
thymidine (0.3 mM) were pretreated with amino acid solutions by
streaking either with Thr (50 .mu.l (0.2 M)) or Abu (50 .mu.l (0.1
M)) vertically along the diameter of the plate to create an amino
acid gradient. Mutant (.beta.5456) and wild-type (.beta.5419
(thyA::erm+AnrdD::kan+pTS13 (bla+thyA:146GUA)) strains were then
streaked horizontally across the plates and incubated for 2 days at
37.degree. C.
[0022] FIGS. 2A to 2C. Point mutations in the editing site and
their consequences.
[0023] FIG. 2A: A schematic showing the positions of the five point
mutations isolated in the editing site of ValRS are shown. The
IleRS editing site (CP1) (28, 29) that intersects the alternating
.beta.-strands (pentagons) and .alpha.-helices (rectangles) of the
catalytic domain is shown. Alignment of residues in the editing
sites of IleRS and ValRS is also shown, with the strictly conserved
residues among all published sequences labeled with a colon.
Abbreviations are Ec, Escherichia coli; Sc, Saccharomyces
cerivisiae; Hs, Homo sapiens.
[0024] FIG. 2B: A graph showing misaminoacylation of tRNA.sup.Val
with Thr by the T222P mutant enzyme at pH 7.5 and 37.degree. C. The
wild-type (WT) and mutant alleles were cloned under the control of
a PBAD promoter (30). The enzymes were partially purified from a
laboratory strain lacking the chromosomal copy of the valS gene
(AvalS::kan+). The purification and aminoacylation procedures were
adapted from Hendrickson et al. (31). (Main panel)
Misaminoacylation of tRNA.sup.Val with Thr by the two enzymes.
(Inset) Aminoacylation of tRNA.sup.Val with Val by the two
enzymes.
[0025] FIG. 2C: A schematic showing MALDI analysis demonstrating In
vivo incorporation of aminobutyrate. The His-tagged protein AlaXp
was expressed in two .DELTA.ilv strains containing the wild-type
valS or the mutant valS:T222P allele, in MS medium containing
Ile-Leu (0.3 mM), Ile-Val (0.02 mM) and Abu (0.2 mM). AlaXp was
purified with Ni-NTA agarose (Qiagen GmbH, Hilden, Germany), cut
out of a SDS-PAGE preparative gel and prepared for MALDI and
.mu.-LC-MS/MS mass analysis (32). The MALDI-MS analyses were
performed in a Voyager-Elite time-of-flight mass spectrometer with
delayed extraction (PerSeptive Biosystems, Inc., Framingham,
Mass.). The spectrum for peptide Lysl56-Arg172 with mass 2097.04 is
shown on the bottom panel (wild-type cells). The top panel shows
the peptide resolved into two components when isolated from cells
bearing the T222P allele of the gene for ValRS. The second
component has a mass of 2083.04, exactly 14 mass units less then
the "wild-type peptide". Multiple peaks correspond to .sup.13C
isotopic forms that separate peptides differing by 1, 2, 3, etc.
mass units.
[0026] FIG. 3. A graph showing the results of growth of the valine
auxotroph CU505 in the presence of a limiting supply of valine and
increasing concentrations of Abu. Overnight cultures of .DELTA.ilv
auxotrophs containing the WT va/s () or the mutant va/S:T222P
allele (J) and grown in MS glucose medium with limiting valine
(0.04 mM Ile-Val. 0.3 mM Ile-Leu) were diluted 1/1, the Val
concentration was adjusted to 0.02 mM and the biomass was
determined by measuring the optical density at 600 nm after 24 h of
growth at 30.degree. C.
[0027] FIG. 4. A graph showing hydrolysis of valyl-tRNAile by wild
type and mutated IleRS. This figure shows the large decrease of the
editing activity of the mutant IleRS having 11 alanine residues in
position aa 239-250 compared with the editing activity of the wild
type IleRS. (square; .box-solid.):IleS ala 11 (mutant IleRS having
11 alanine residues in position aa 239-250); (diamond,
.diamond-solid.): IleS wt (wild type IleRS)
[0028] FIGS. 5A and 5B: Correlation of Abu toxicity phenotypes and
effectiveness of deacylation by mutant enzymes.
[0029] FIG. 5A: Noncognate amino acid toxicity phenotypes. ValS
null strains harboring each of the ValRS mutant enzymes in trans
were plated on minimal medium. Abu (alpha-aminobutyrate) was loaded
into a central well and plates were incubated for 24 hours at 42
deg. C. Degree of toxicity in response to Abu was evaluated by the
diameter of the region of inhibited cell growth. Strains bearing
the K277Q and T222P mutant valS alleles showed the most severe
response to Abu. The growth of D230N vals was inhibited by high
concentrations of Abu but the effect was not observed at lower
concentrations. The V276A mutation introduces a slight sensitivity
to Abu. Similar phenotypes were observed in response to exogenous
threonine (data not shown).
[0030] FIG. 5B: Histogram representation of the inhibition zone
diameters of DvalS::kanR strains harboring each of the ValRS
editing mutants on plasmid in response to exogenous Abu. Effects
range from mild to severe. Histogram representation of the
percentage of mischarged Thr-tRNAVal remaining after incubation
with 2 nM enzyme for 15 minutes at room temperature. Both show a
distinct range in the severity of editing defects caused by these
point mutations. Taken together this illustrates a correlation
between the ability of the mutant enzyme to deacylate mischarged
amino acids from tRNAVal in vitro and the degree of in vivo
toxicity.
DETAILED DESCRIPTION OF THE INVENTION
[0031] The present invention provides a method to diversify the
chemical composition of proteins produced in vivo by a cell
comprising the step of disabling the editing function of one of its
aminoacyl tRNA synthetases.
[0032] The phrase "a step of disabling the editing function of an
aminoacyl tRNA synthetase", is intended to designate a step through
which a mutant strain is obtained wherein an allele coding for an
aminoacyl tRNA synthetasecomprises a mutation in its editing site
or domain resulting in:
[0033] the loss of the role of the editing function in restricting
the genetic code to 20 amino acids;
[0034] the loss of the role of the editing function in preventing
the invasion of noncognate amino acid; and/or
[0035] the misactivation of noncognate amino acids or/and the non
hydrolyzation of generated "noncognate amino acid-tRNA" normally
hydrolyzed by the editing function of the wild type cognate
aminoacyl tRNA synthetase,
[0036] these alterations of the editing function inducing the
potential misincorporation of a noncognate amino acid, preferably a
noncanonical amino acid, in place of the cognate amino acid charged
by the aminoacyl tRNA synthetase encoded by the wild type
allele.
[0037] For example, such a step of disabling the editing function
of one of an aminoacyl tRNA synthetase may comprise a step of:
[0038] selecting a mutant strain comprising an aminoacyl tRNA
synthetase natural variant having a mutation in its editing domain
resulting to these above cited editing function alterations; or
preferably
[0039] selecting a mutant strain comprising an aminoacyl tRNA
synthetase variant having a mutation in its editing domain
resulting to these above cited editing function alterations, this
variant being obtained by mutagenesis, such as by homologous
recombination or replacement allele vector or by any mutagenesis
methods known by the skilled person capable of introducing such a
mutation, preferably integrated into the genome of the cell.
[0040] The term "cognate amino acid" as used in the present
specification is intended to designate the amino acid normally
charged by the tRNA corresponding to (or associated with) the
aminoacyl tRNA synthetase wild type (for example: the cognate amino
acid for valyl-tRNA synthetase is valine).
[0041] By the term "cognate tRNA" as used in the present
specification is intended to designate the tRNA corresponding to
(or associated with) the aminoacyl tRNA synthetase wild type (for
example: the cognate tRNA for valyl-tRNA synthetase is
tRNAVal).
[0042] In an additional embodiment of the present invention, is
provided a method for producing in vivo proteins comprising at
least one noncanonical amino acid comprising the step of:
[0043] a) selecting a cell strain wherein the editing function of
one of its aminoacyl tRNA synthetases has been disabled by
mutagenesis, said disabled editing function allowing the aminoacyl
tRNA synthetase to mischarge the cognate tRNA with said at least
one noncanonical amino acid;
[0044] b) culturing the selected strain in a culture medium
comprising said noncanonical amino acid, or one of its precursor,
under favourable conditions for the growth of said strain; and
[0045] c) recovering from the culture medium or from the cells
obtained in step b) the proteins containing said noncanonical amino
acid.
[0046] By "favourable growth conditions" is meant an environment
that is relatively favorable for cell growth and/or viability. Such
conditions take into account the relative availability of nutrients
and optimal temperature, atmospheric pressure, presence or absence
of gases (such as oxygen and carbon dioxide), and exposure to
light, as required by the organism being studied.
[0047] By "precursor" is meant a compound which can be efficiently
converted in vivo by the cell in said noncanonical amino acid.
[0048] In a preferred embodiment, the cell strain wherein the
editing function of one of its aminoacyl tRNA synthetases has been
disabled, comprises a mutation in the DNA sequence encoding the
editing domain of said disabled aminoacyl tRNA synthetase compared
to the wild type aminoacyl tRNA synthetase coding sequence.
[0049] In a more preferred embodiment, said DNA mutation leads to
an amino acid mutation, preferably an amino acid substitution, more
preferably an amino acid single substitution, in the editing domain
of said aminoacyl tRNA synthetase.
[0050] In another preferred embodiment is provided a method
according to the present invention, wherein the disabled aminoacyl
tRNA synthetase is capable of mischarging its cognate tRNA with a
canonical amino acid sterically similar to the amino acid charged
by the wild type aminoacyl tRNA synthetase on its cognate tRNA.
[0051] In another preferred embodiment is provided a method
according to the present invention, wherein the disabled aminoacyl
tRNA synthetase is capable of mischarging its cognate tRNA with a
noncanonical amino acid sterically similar to the amino acid
charged by the wild type aminoacyl tRNA synthetase on its cognate
tRNA.
[0052] In another aspect of the present invention, there is
provides a method for obtaining cells capable of producing in vivo
proteins comprising at least one noncanonical amino acid comprising
the step of mutagenizing the DNA sequence encoding the editing
domain of an aminoacyl tRNA synthetase in a cell, said mutagenizing
leading to an aminoacyl tRNA synthetase variant having an amino
acid mutation in its editing domain and said mutation allowing the
aminoacyl tRNA synthetase variant to mischarge its cognate tRNA
with said at least one noncanonical amino acid.
[0053] In preferred embodiment, there is provides a method for
obtaining cells capable of producing in vivo proteins comprising at
least one noncanonical amino acid according to the present
invention, comprising the steps of:
[0054] a) assaying the ability of an aminoacyl tRNA synthetase
variant mutated in its editing domain for its ability to mischarge
its cognate tRNA with a noncanonical amino acid;
[0055] b) mutagenizing the DNA sequence encoding the editing domain
of said aminoacyl tRNA synthetase in a cell, said mutagenizing
leading to replace the allele encoding the wild type aminoacyl tRNA
synthetase by an allele encoding the aminoacyl tRNA synthetase
variant assayed in step a) and capable of producing detectable
noncanonical amino acid mischarging;
[0056] c) optionally, identifying, selecting and/or cloning the
cells containing such aminoacyl tRNA synthetase variant having the
ability to mischarge one noncanonical amino acid.
[0057] In a preferred embodiment, the cell wherein the editing
domain of the aminoacyl tRNA synthetase has been disabled,
preferably by mutagenesis, is a microbial or animal cell,
preferably a bacterium, a yeast or a fungus, more preferably a
bacterium such as Escherichia coli or Acinetobacter.
[0058] In another preferred embodiment of the present invention, is
provided a method, wherein the editing domain of the aminoacyl tRNA
synthetase of the target cell has been disabled by by homologous
recombination or by recombination into the genome of the target
cell using an allelic replacement vector.
[0059] The oligonucleotides comprising the nucleic fragment
encoding the mutated editing site, or portion thereof, and which
contain the mutation to be introduced into the wild type allele of
the target cell, can be chemically synthetized or synthetized by
Polymerase Chain Reaction (PCR).
[0060] By microbial or animal cells are preferred cells that
undergo homologous recombination. Such cells may be of bacterial,
mycobacterial, yeast, fingal, algal, plant, or animal origin.
[0061] By "homologous recombination" is meant a process by which an
exogenously introduced DNA molecule integrates into a target DNA
molecule in a region where there is identical or near-identical
nucleotide sequence between the two molecules. Homologous
recombination is mediated by complementary base-pairing, and may
result in either insertion of the exogenous DNA into the target DNA
(a single cross-over event), or replacement of the target DNA by
the exogenous DNA (a double cross-over event).
[0062] By "allelic replacement vector" is meant any DNA element
that can be used to introduce mutations into the genome of a target
cell by specific replacement of a native gene with a mutated copy.
For example, gene replacement in bacteria is commonly performed
using plasmids that contain a target gene containing a mutation and
a negative selectable marker outside of the region of homology.
Such a plasmid integrates into the target chromosome by homologous
recombination (single cross-over). Appropriate selection yields
cells that have lost the negative selection marker by a second
homologous recombination event (double cross-over) and contain only
a mutant copy of the target gene.
[0063] In another embodiment, the cell wherein the editing domain
of the aminoacyl tRNA synthetase has been disabled or mutagenized
may contain a selectable marker gene for identifying and selecting
the cells containing the mutagenized DNA. The identification and
selection of that cells wherein the editing domain of the aminoacyl
tRNA synthetase has been disabled or mutagenized may be based upon
the ability of the cells to grow on selective medium, wherein a
cell containing the selectable marker can grow on selective medium,
and a cell lacking this selectable marker cannot grow, or grows
more slowly, on selective medium.
[0064] In still another embodiment, the cell wherein the editing
domain of the aminoacyl tRNA synthetase has been disabled or
mutagenized may contain a reporter gene, for identifying and
selecting the cells containing the mutagenized DNA. The
identification and selection of that cells wherein the editing
domain of the aminoacyl tRNA synthetase has been disabled or
mutagenized may be based on a reporter gene assay, wherein the
expression by the cell of the reporter gene confirms the disabling
or the mutagenesis and a cell lacking the disabling or the
mutagenesis does not express the reporter gene.
[0065] By "selectable marker" is meant a gene that alters the
ability of a cell harboring this gene to grow or survive in a given
growth environment relative to a similar cell lacking the
selectable marker. Such a marker may be a positive or negative
selectable marker. For example, a positive selectable marker (e.g.,
an antibiotic resistance or auxotrophic growth gene) encodes a
product that confers growth or survival abilities in selective
medium (e.g., containing an antibiotic or lacking an essential
nutrient). A negative selectable marker, in contrast, prevents
cells harbouring this gene from growing in negative selection
medium, when compared to cells not harbouring this gene. A
selectable marker may confer both positive and negative
selectability, depending upon the medium used to grow the cell. The
use of selectable markers in prokaryotic and eukaryotic cells is
well known by those of skill in the art.
[0066] By "reporter gene" is meant any gene which encodes a product
whose expression is detectable and/or quantitatable by
immunological, chemical, biochemical, biological, or mechanical
assays. A reporter gene product may, for example, have one of the
following attributes, without restriction: fluorescence (e.g.,
green fluorescent protein), enzymatic activity (e.g.,
lacZ/.beta.-galactosidase, luciferase, chloramphenicol
acetyltransferase, alkaline phosphatase), toxicity (e.g., ricin),
or an ability to be specifically bound by a second molecule (e.g.,
biotin or a detectably labelled antibody). It is understood that
any engineered variants of reporter genes, which are readily
available to one skilled in the art, are also included, without
restriction, in the foregoing definition.
[0067] By "identifying cells containing mutagenized DNA" is meant
exposing the population of cells transformed with the mutagenized
DNA to selective pressure (such as growth in the presence of an
antibiotic or the absence of a nutrient) consistent with a
selectable marker carried by the cell containing the recombined
mutagenized DNA (e.g., an antibiotic resistance gene or auxotrophic
growth gene known to those skilled in the art). Identifying cells
containing the recombined mutagenized DNA may also be done by
subjecting transformed cells to a reporter gene assay. Selections
and screens may be employed to identify cells containing the
recombined mutagenized DNA, although selections are preferred.
[0068] In another aspect, the present invention is directed to a
method for selecting an aminoacyl tRNA synthetase variant capable
of mischarging its cognate tRNA with a noncognate and/or a
noncanonical amino acid, comprising the steps of:
[0069] a) elaborating a DNA construct comprising a DNA sequence
encoding an aminoacyl tRNA synthetase variant wherein said
aminoacyl tRNA synthetase variant has at least an amino acid
mutation, preferably a substitution, more preferably a single
substitution, in its editing domain compared with the wild type
aminoacyl tRNA synthetase;
[0070] b) transforming a host cell with the DNA construct of step
a) and, after a step of culturing said transformed host cell,
recovering and, optionally, purifying the recombinant aminoacyl
tRNA synthetase variant expressed by the host cell;
[0071] c) assaying the ability of the recombinant aminoacyl tRNA
synthetase variant recovered in step b) for its ability to
mischarge its cognate tRNA with a noncognate and/or a noncanonical
amino acid; and
[0072] d) selecting said aminoacyl tRNA synthetase variant if
detectable mischarging has been produced in step c).
[0073] Such methods for assaying the ability of the recombinant
aminoacyl tRNA synthetase variant recovered in step b) for its
ability to mischarge its cognate tRNA with a noncognate and/or a
noncanonical amino acid, are disclosed for example in the following
examples (see FIG. 2B). The methods known to those skilled in the
art for assaying such an ability may be used.
[0074] By "transformation" is meant any method for introducing
foreign molecules, such as DNA, into a cell. Lipofection,
DEAE-dextran-mediated transfection, microinjection, protoplast
fusion, calcium phosphate precipitation, retroviral delivery,
electroporation, natural transformation, and biolistic
transformation are just a few of the methods known to those skilled
in the art which may be used.
[0075] In another aspect, the invention provides an isolated
aminoacyl tRNA synthetase variant capable of mischarging its
cognate tRNA with a noncanonical amino acid, obtainable by the
method for selecting an aminoacyl tRNA synthetase variant according
to the present invention.
[0076] The isolated nucleic acid sequence encoding the aminoacyl
tRNA synthetase variant according to the invention or vectors
comprising a nucleic acid encoding said aminoacyl tRNA synthetase
variant form also part of the present invention.
[0077] In another aspect, the invention provides a transformed cell
comprising a nucleic acid encoding an aminoacyl tRNA synthetase
variant according to the present invention.
[0078] In a preferred embodiment, said transformed cells are
characterized in that said nucleic acid encoding an aminoacyl tRNA
synthetase variant according to the present invention is integrated
in the genome of said cell.
[0079] In a more preferred embodiment, said transformed cells are
characterized in that the nucleic fragment comprising the mutation,
preferably a substitution, more preferably a single substitution,
leading to the alteration of the editing function of the aminoacyl
tRNA synthetase variant, has been integrated into the genome of
said transformed cell by using mutagenesis, such as homologous
recombination, replacement allele vector or any mutagenesis methods
for introducing nucleic acid molecules such as DNA, into a cell,
known to those skilled in the art, such as lipofection,
DEAE-dextran-mediated transfection, microinjection, protoplast
fusion, calcium phosphate precipitation, retroviral delivery,
electroporation, natural transformation and biolistic
transformation.
[0080] The invention also relates to isolated prokaryotic or
eukaryotic cells capable of producing a protein the amino acid
sequence of which comprises at least one noncanonical amino acid,
characterized in that they comprise an aminoacyl-tRNA synthetase
variant which is capable of charging onto one of its cognate tRNAs
a noncanonical amino acid or an amino acid other than the cognate
amino acid, and in that the nucleic acid sequence of the allele
encoding said aminoacyl-tRNA synthetase variant includes at least
one mutation, preferably a substitution, more preferably a single
substitution, compared with the sequence of the corresponding
wild-type allele, said at least one mutation integrated into the
genome being located on the editing site of said aminoacyl-tRNA
synthetase and having been introduced by a technique of
mutagenesis, such as genetic recombination.
[0081] In another aspect, the invention also comprises the use of a
transformed cell according to the invention for producing protein,
in particular recombinant protein, the amino acid sequence of which
comprises at least one unconventional amino acid.
[0082] In a preferred embodiment, the present invention is directed
to a method for the production of proteins comprising a
noncanonical amino acid characterized in that said method comprises
the following steps:
[0083] a) culturing a transformed cell comprising an allele
encoding an aminoacyl tRNA synthetase variant according to the
invention in a culture medium containing a noncanonical amino acid
capable of being mischarged by the cognate tRNA of the aminoacyl
tRNA synthetase variant contained in said cell, in a culture medium
and under culture conditions which allow the growth of said cell;
and
[0084] b) recovering and, optionally, purifying the proteins
comprising said noncanonical amino acid from the culture medium
(supernatant) or from the cells (cells pellet) of step a).
[0085] Proteins comprising a noncanonical amino acid obtained by
the above method according to the invention form also part of the
present invention.
[0086] The processes for purifying protein, which may be natural or
recombinant, conventionally used by those skilled in the art
generally employ methods used individually or in combination, such
as fractionation, chromatography methods, immunoaffinity techniques
using specific mono- or polyclonal antibodies, etc.
[0087] The presence of a noncanonical amino acid on the protein to
be purified, which noncanonical amino acid could have a specific
functional group, may facilitate its purification by reacting
selectively with the purification support without modifying the
activity of the protein.
[0088] Among the proteins which can be produced by a process
according to the invention, mention may be made, but without being
limited thereto, of proteins which, through the incorporation of at
least one noncanonical amino acid, make it possible to obtain a
desired activity which a protein the sequence of which includes
only canonical amino acids does not make it possible to obtain. The
term "activity" is intended to refer, in general, to any activity
such as a physiological or biological activity, even partial,
relating to unicellular or multicellular organisms, such as for
example a structural or biochemical activity, for example an
enzymatic or antigenic activity, an activity of antibody type, or
an activity which modulates, regulates or inhibits biological
activity, or such that it allows the implementation thereof in a
process for biosynthesizing or for biodegrading chemical or
biochemical compounds.
[0089] Among the proteins which can be produced by a process
according to the invention, mention may also be made of proteins
for which the incorporation of at least one noncanonical amino acid
is carried out such that there results therefrom no substantial
modification of the biological activity of the corresponding
unmodified protein. Besides the conserved biological activity of
the corresponding unmodified protein, these proteins according to
the invention will have a noncanonical amino acid with specific
properties which may be advantageously exploited.
[0090] Among the specific properties conferred by the presence of a
noncanonical amino acid, mention may be made in particular of the
properties linked to the presence of a functional group on said
noncanonical amino acid, capable of reacting easily and
specifically with a chemical or biochemical compound under
conditions which make it possible not to modify the activity of the
protein or which avoid modifying the conventional amino acids.
[0091] The presence of this specific functional group may
advantageously be used, for example, for:
[0092] (i) purifying any protein, in particular any recombinant
protein, which incorporates said unconventional amino acid;
[0093] (ii) coupling such a protein to a solid support;
[0094] (iii) coupling to such a protein molecules capable of being
detected, such as spectroscopic probes of varied nature;
[0095] (iv) coupling to such a protein lipophilic or hydrophilic
polymers which allow the solubilization thereof in solvents or
which allow masking against recognition by antibodies;
[0096] (v) coupling such a protein to a polynucleotide;
[0097] (vi) coupling such a protein to a chemical or biochemical
compound the presence of which makes it possible to increase, to
decrease, to modify, to regulate or to target the biological
activity of said protein, or to modify the bioavailability thereof
as a compound for therapeutic use; or
[0098] (vii) permanently attaching to such a protein a coenzyme
which otherwise would diffuse in solution.
[0099] Also included in the present invention are the processes
according to the invention, characterized in that said transformed
cell according to the present invention comprises a homologous or
heterologous gene of interest the coding sequence of which includes
at least one codon encoding an amino acid the cognate tRNA of which
can be mischarged by the aminoacyl tRNA synthetase variant
contained in said transformed cell.
[0100] In general, the homologous or heterologous gene of interest,
which can be isolated by any conventional technique, such as
cloning or PCR (Polymerase Chain Reaction), or chemically
synthesized, may be chosen from genes encoding any protein which
can be used as a therapeutic or cosmetic compound, or as a
diagnostic reagent or as a compound which can be used in a
biosynthesis or biodegradation process. The protein of interest can
consist of a mature protein, a precursor, and in particular a
precursor intended to be secreted and comprising a signal peptide,
a truncated protein, a chimeric protein originating from the fusion
of sequences of diverse origins, or a mutated protein having
improved and/or modified biological properties.
[0101] The invention also comprises a process for producing a
protein according to the invention, characterized in that the
culture medium of step a) also comprises the compounds required for
inducing the synthesis of the protein encoded by said homologous or
heterologous gene of interest. These compounds are known to those
skilled in the art and depend, in particular, on the cell and on
the homologous or heterologous gene selected.
[0102] In any of the method according to the present invention, it
is preferred that said aminoacyl tRNA synthetase is an aminoacyl
tRNA synthetase selected from the group consisting of the aminoacyl
tRNA synthetase comprising an editing function corresponding to an
editing site or domain encoded by a portion of the DNA encoded said
aminoacyl tRNA synthetase, preferably encoded by a DNA portion
having at least conserved residues compared after alignment with
the editing site of the valyl-tRNA synthetase and isoleucyl-tRNA
synthetase as shown in FIG. 2A, preferably selected from the group
consisting of the aminoacyl tRNA synthetase valyl-tRNA synthetase,
isoleucyl-tRNA synthetase, leucyl-tRNA synthetase, alanyl-tRNA
synthetase, prolyl-tRNA synthetase, threonyl-tRNA synthetase,
phenyl-tRNA synthetase and lysyl-tRNA synthetase which are known to
have an editing site or domain (see for Ile RS Baldwin, A. N. and
Berg, P. (1966) J. Biol. Chem. 241, 839-845 and Eldred, E. W. and
Schimmel, P. R. (1972) J. Biol. Chem. 247, 2961-2964; for Val RS,
Fersht, A. R. and Kaethner, M. M. (1976) Biochemistry. 15 (15),
3342-3346; for Leu RS, English, S. et al., (1986) Nucleic Acids
Research. 14 (19), 7529-7539; for Ala RS, Tsui, W.-C. and Fersht,
A. R. (1981) Nucleic Acids Research. 9, 7529-7539; for Pro RS,
Beuning, P. J. and Musier-Forsyth, K. (2000) PNAS. 97 (16),
8916-8920; for Thr RS, Sankaranarayanan, R. et al., (2000) Nat.
Struct. Biol. 7, 461-465 and Musier-Foryth, K. and Beuning, P. J.
(2000) Nat. Struct. Biol. 7, 435-436; for PheRS, Yarus, M. (1972)
PNAS. 69, 1915-1919 and for LysRS, Jakubowski, H. (1997)
Biochemistry. 36, 11077-11085.
EXAMPLES
Example 1
[0103] A direct selection for restoring an enzymatic activity
through incorporation of Abu could not be easily set up because its
aliphatic side chain lacks chemical reactivity and therefore cannot
act as a catalytic residue. We thus resorted to an indirect scheme
based on the structural resemblance of Abu with Cys (FIG. 1A), an
essential catalytic residue in numerous enzymes. Selecting a
synthetase mutant that mischarged its cognate tRNA with Cys might
result in Abu being mischarged by the mutant synthetase.
[0104] We took advantage of the thyA conditional selection screen
in E. coli, based on the absolute requirement for an active
thymidylate synthase when thymidine is not provided as a growth
factor (13). This same screen was used previously to assess the
potency of suppressor Cys-tRNAs in codon misreading (14) and to
enforce phenotypic suppression by the non-canonical azaleucine
(15). An entire set of plasmid-borne thya alleles with all 64
different codons at position 146 was constructed for altering the
catalytic site occupied by an essential Cys (16, 17). Each allele
was tested for its ability to restore growth to an E. coli strain
(lacking the chromosomal copy of thyA) on mineral glucose medium in
the absence of thymidine (14, 15). Three of the 64 plasmid-borne
thyA alleles restored growth. These had one of three codons-UGU,
UGC, or UGA. The growth responses of the UGU and UGC alleles were
expected, as these code for cysteine. The positive response of UGA
(a termination codon in E. coli) likely results from read-through
by Cys-tRNA, and thus demonstrates the sensitivity of the selection
assay.
[0105] Strains bearing inactive alleles of thyA were then tested to
see if they could be suppressed by supplying them with excess
L-cysteine in mineral medium devoid of thymidine. Shallow growth
was reproducibly observed on L-cysteine gradient plates (18) for
the missense alleles having any of the eight codons AUN and GUN
alone. Growth was stronger with alleles bearing any of the four Val
codons (GUU, GUC, GUA, and GUG) than for those with the three Ile
codons (AUU, AUC, and AUA) or for the Met codon AUG.
Cysteine-suppression of the three Ile codon-bearing alleles and of
the Met codon-bearing allele was abolished by addition of exogenous
L-isoleucine and L-methionine, respectively. Suppression of the
four Val codon-bearing alleles was abolished by addition of
exogenous L-valine plus L-isoleucine but not by L-isoleucine alone
(18). The four Val146 alleles gave a similar growth response in
cysteine gradient plates, despite being decoded by three different
tRNAVal isoacceptors, thus suggesting that Cys is being mischarged
onto all three Val isoacceptor tRNAs by ValRS. Altogether, these
results suggested that ValRS catalyzed the formation of Cys-tRNAVal
in vivo at a rate sufficient for active thymidylate synthase
production and that this mischarging reaction was prevented by
increasing the intracellular concentration of L-valine. This
interpretation is in line with earlier reports of Cys misactivation
by ValRS in vitro (11).
[0106] Suppression of thyA:Val146 alleles was weak on plates and
required high L-cysteine concentrations (developed from a gradient
starting at a concentration of 100 mM). It could thus be
anticipated that a scarce L-cysteine supply should select for an
enhanced efficiency of phenotypic suppression. Two experimental
procedures were followed to this end. In the first procedure strain
.beta.5519 (thyA::erm+AnrdD::kan+pTS13 (bla+thyA:146GUA) was
propagated over 100 generations in serial liquid culture with
limiting cysteine (1.5 mM) under anaerobic conditions resulting in
isolation of strain p5420 (19). The second procedure relied on a
one-step selection under aerobic conditions on plates containing a
non-oxidizable precursor that is inefficiently converted into
cysteine by E. coli (S-carbamoyl-cysteine (Sec, FIG. 1A) (19)). A
mutator marker (dnaQ or muts) was introduced into the test strain
P5519 to increase the frequency above 10.sup.-10 of
Scc-suppressible thymidine auxotrophs (19). This approach resulted
in the isolation of four strains: .beta.5456, .beta.5479,
.beta.5485 and .beta.5486. All five isolated strains were found not
to grow at 42, in agreement with the heat sensitivity generally
caused by translational errors (20). The cysteine-suppression
phenotype is shown for one of these strains (.beta.5456) in FIG.
1B, together with its abolition by L-valine (supplied as an Ile-Val
peptide).
[0107] Judging from these phenotypes, mutations in the vals gene
for valyl-tRNA synthetase were suspected for each isolated strain.
To test for this possibility, we took advantage of the
nrdD::kan+marker (19) located 0.4 min from vals on the E. coli
chromosome. For the five mutants, the trait of cysteine- or
Scc-suppressible thymidine auxotrophy was indeed found to
co-transduce with the nrdD::kan+marker in a proportion of about
45%. Further characterization of the valS mutations was performed
by PCR-amplification followed by sequencing of five
nrdD::kan+transductants exhibiting Scc-suppressible thymidine
auxotrophy derived from the five isolated strains (21). As shown in
Table 1, each of the five mutants contained a single amino acid
substitution at positions within the conserved editing domain
(known as CP1) of ValRS (22). The following mutations were
identified: T222P, R223H, D230N, V276A, and K277Q. Remarkably, two
of these positions (T222 and D230) align with conserved positions
in IleRS that had been demonstrated previously to be involved in
the hydrolytic editing of misacylated Val-tRNA.sup.Ile (23) (FIG.
2A).
1TABLE 1 Mutations of the valS gene selected by suppression of the
thymidine auxotrophy of strain .beta.5419 (.DELTA.thyA::erm.sup.+
.DELTA.nrdD::kan.sup.+pTS13 (bla.sup.+thyA:146GUA)). Method of
Mutator Codon of valS Strain isolation genotype 222 223 230 276 277
.beta.5419 -- wt ACC CGT GAT GTG AAA Thr Arg Asp Val Lys .beta.5456
Selection .DELTA.dnaQ CCC CGT GAT GTG AAA on Scc Pro .beta.5479
Selection .DELTA.mutS ACC CAT GAT GTG AAA on Scc His .beta.5486
Selection .DELTA.mutS ACC CGT AAT GTG AAA on Scc Asn .beta.5485
Selection .DELTA.mutS ACC CGT GAT GCG AAA on Scc Ala .beta.5520
Selection wt ACC CGT GAT GTG CAA on Cys under Gln anaerobiosis
[0108] ValRS is known to misactivate Thr and generate
Thr-tRNA.sup.Val, which normally is hydrolyzed by the ValRS editing
activity (9). If the ValRS mutants in Table 1 are impaired for
editing, then strains bearing each of these mutations should
misincorporate Thr into protein, and L-threonine would then be
toxic in these strains. The growth of the five different vals
strains was therefore tested in the presence of L-threonine. All
displayed high sensitivity towards exogenous L-threonine (at 2 mM),
while the parent strain .beta.5419 was insensitive to L-threonine
at all concentrations. The results for the strain carrying the
valS:T222P allele are shown in FIG. 1C.
[0109] The phenotype of the valS:T222P allele suggested that the
T222P enzyme mischarged tRNA.sup.Val with Thr and Cys in vivo. The
T222P mutant enzyme was, therefore, expressed and partially
purified so that it could be directly assayed for the ability to
misacylate tRNA.sup.Val. The purified enzyme had the same activity
as the wild-type enzyme for charging with valine (FIG. 2B, inset).
In contrast, the T222P mutant enzyme misacylated tRNA.sup.Val with
Thr to give Thr-tRNA.sup.Val while the wild-type enzyme produced no
detectable mischarged tRNA.sup.Val (FIG. 2B). Misacylation of
tRNA.sup.Val with cysteine was also catalyzed only by the mutant
enzyme.
[0110] We anticipated that ValRS mutants that misincorporated Cys
would also misincorporate Abu. Indeed, the strain carrying the
valS:T222P allele on the chromosome (p5456) was sensitive to Abu
(FIG. 1C) (whereas its wild-type counterpart was not), suggesting
incorporation of Abu in response to Val codons. With this in mind,
we showed that Abu could contribute to the relief of L-valine
auxotrophy of the .DELTA.ilv strain CU505, but only in the presence
of the valS:T222P allele (19). Whole cell protein was isolated from
strain CU505 grown in the presence of Abu (0.2 mM), in the presence
or absence of the valS:T222P allele in the host cell. Analysis of
amino acid composition showed that 24% of the valine was replaced
by Abu only in the strain harboring the mutant allele (Table 2).
Finally, the valine-rich yeast protein AlaXp (swissprot:P53960) was
overexpressed and purified from strains containing the valS:T222P
allele grown in the presence of Abu. The protein samples were
digested by trypsin and analyzed by mass spectrometry. MALDI
analysis showed that, when AlaXp was produced in the strain
carrying the T222P mutation, it contained a mixture of Val and
misincorporated Abu (FIG. 2C). For a given peptide the degree of
misincorporation ranged between 9.5% and 18% per Val codon.
Sequencing of several Abu-containing peptides confirmed that Abu
was specifically misincorporated into positions designated by Val
codons.
[0111] Table 2. Incorporation of Abu into cells bearing wild-type
and T222P mutant alleles of valS. Cultures of the .DELTA.ilv strain
CU505 (19) and the isogenic strain P5498 carrying the valS:T222P
allele were grown overnight in minimal medium (27) with Ile-Leu
(0.3 mM) and limiting valine (0.04 mM Ile-Val), diluted (1/2), and
adjusted with Ile-Val (0.02 mM) with or without Abu (0.2 mM). After
24 hours of growth, total protein was extracted as follows. Cells
were first harvested by centrifugation and washed in cold 10%
trichloroacetic acid (TCA, 1/2 of the culture volume). Cells were
then re-centrifuged at 4000.times.g for 10 min, resuspended in cold
10% TCA ({fraction (1/10)} of the culture volume) and centrifuged
again. The washed cells were resuspended in 5% TCA, heated at
95.degree. C. for 30 minutes and centrifuged (4000.times.g, 10
min). The precipitate was washed three times with cold acetone and
dissolved in 50 mM NH.sub.4HCO.sub.3. Proteins were hydrolyzed in 6
N HCl-0.2% phenol at 110.degree. C. for 20 h in sealed tubes.
Norleucine was added as an internal standard. Aliquots of the
hydrolysates were analyzed on a Beckman 6300 Amino Acid Analyzer.
Amino acids were quantified by appropriate standards and values are
presented relative the wild-type (WT) control that lacked Abu.
2 Amino acid incorporated WT T222P WT + Abu T222P + Abu Abu 0.0 0.0
0.0 0.24 Val 1.0 0.95 1.0 0.73 Val .+-. Abu 1.0 0.95 1.0 0.97 Ile
1.0 1.0 1.0 1.0 Ala 1.0 0.97 1.0 0.92
Example 2
[0112] Directed evolution scheme. To avoid a decrease of cysteine
concentration in the medium due to oxidation, the selections were
carried out under strictly anaerobic conditions. The biomass of
cells in cultures relative to the cysteine concentration (measured
as optical density at 600 nm after 24 h of growth at 30.degree. C.)
gradually increased four-fold when the thyA strain .beta.5366
(AthyA::erm.sup.+ pTS13 (bla.sup.+ thyA:146GUA)) was propagated by
serial transfer in mineral glucose medium supplemented with 1.5 mM
cysteine. A high concentration of cysteine was required because the
input population could hardly be propagated otherwise. After 16
inoculations, each at a dilution of {fraction (1/100)} (a total of
about 100 generations), single colonies were isolated on mineral
glucose plates supplemented with thymidine (a representative of
which was designated strain .beta.5520).
[0113] One-step selection scheme. For the one-step selections,
cysteine oxidation under aerobiosis and subsequent cystine
precipitation from the growth medium was avoided by use of the
non-oxidizable precursor S-carbamoyl-cysteine (Scc). Scc is a poor
precursor of cysteine and sustains growth of a Cys-auxotrophic E.
coli strain less efficiently than cysteine. However, at
concentrations above 1 mM, Sce gave rise to suppression of the
thymidine auxotrophy of strain .beta.5520, while in the absence of
thymidine no growth was detectable below 1 mM Scc. When 35419 cells
bearing the thyA:Val146GUA allele on a high copy plasmid were
plated on mineral glucose plates supplemented with Scc (3 mM), no
colonies grew after prolonged incubation at 30.degree. C. (The
experiment was designed to detect a mutant at a frequency of
10.sup.-10.) To increase the mutation frequency, a mutS::spc+ (33)
mutator allele was introduced in the genetic background of strain
135419 by P1 transduction, yielding strain .beta.5432 (the
mutS::spc+ disruption disables the mismatch repair system and leads
to random transitions and frameshifts (34)). Colonies then appeared
on the same medium with a frequency of about 10.sup.-8. No colonies
were found on plates lacking Scc. A comparable frequency of
Scc-suppressible clones was obtained after introduction by P1
transduction (33) of a dnaQ mutator allele (33) (to give strain
135435).
[0114] Abu misincorporation. The valine auxotroph CU505 (33) was
grown in the presence of a limiting supply of valine and increasing
concentrations of Abu (FIG. 3). The biomass of cells in cultures
relative to the valine concentration in the medium did not change
up to a 1 mM concentration of Abu. In contrast, when the valST222P
allele was introduced into the chromosome (strain .beta.5498), the
yield of cells was diminished in the absence of Abu but increased
up to 30% when Abu (0.2 mM) was added.
[0115] Thus, E. coli strains that proliferate only because of
infiltration of the Val coding pathway were selected and all
contained mutations leading to single amino acid substitutions in
the editing site of ValRS. This observation is consistent with a
central role for editing in restricting the genetic code to twenty
amino acids, by preventing the invasion of other amino acids such
as Abu. Indeed, the editing sites in IleRS and ValRS are rigorously
conserved in even the most deeply branched organisms in the tree of
life. However the translation accuracy maintained by editing may
prevent further chemical diversification of proteins. Thus,
disabling the editing function of a synthetase, as demonstrated in
the present work, offers a a powerful approach to diversify the
chemical composition of proteins produced in vivo.
[0116] For some synthetases, accuracy depends critically on an
editing function at a site distinct from the aminoacylation site.
Mutants of Escherichia coli that mischarge tRNA.sup.Val with
cysteine were sought by random mutagenesis of the whole chromosome.
All mutations obtained were located in the editing site of
valyl-tRNA synthetase. Over 20% of the valine in cellular proteins
from such an editing mutant organism could be replaced with the
noncanonical aminobutyrate, sterically similar to cysteine. Thus,
the editing function may have played a central role in restricting
the genetic code to twenty amino acids. Disabling this editing
function offers a powerful new approach for diversifying the
chemical composition of proteins and for emulating ambiguous
evolutionary stages of translation.
Example 3
Correlation Between Toxicity of .alpha.-Aminobutyrate for E. coli
Strains Harboring Different valS Alleles Mutated in the Editing
Domain and Results of in Vitro Editing Assays Carried Out with the
Corresponding ValRS Enzyme Variants
[0117] To investigate more deeply the editing phenotypes of the
five aforementioned mutant ValRSs, and the relationship between
cell viability and editing, plasmids harboring genes for mutant and
wild-type ValRSs were constructed and placed under control of an
arabinose-inducible promoter. These constructs were then used to
investigate the in vivo phenotypes of the mutant enzymes.
Separately, His-tagged versions of the mutant enzymes were
constructed in parallel for use in purifying the enzymes for
studies of aminoacylation and editing activities in vitro.
[0118] Construction of vals knockout strains: A 1.7 kb portion of
the valS gene from pET16valS that contained the region encoding the
editing domain of ValRS was excised using SalI and XhoI. This
region was then replaced with a kanamycin marker (1.2 kb in size)
liberated from pUC4K by flanking SalI restriction sites.
Overhanging regions of the valS gene of 406 and 708 base pairs
respectively were left to allow for homologous recombination
(resulting plasmid named pLAN362). The kanamycin marker contained
its own promoter and was inserted in the forward orientation with
respect to the valS gene. To increase the amount of linearized
product used for homologous recombination, the valS::kan from
pLAN362 was PCR-amplified using primers annealing to the T7
promoter and terminator. The products of this reaction were
visualized and purified in a 1% agarose TAE gel.
[0119] To get recombination into the E. coli chromosome, the
linearized valS::kan was transformed by electroporation into a
hyper recombinant strain of E. coli (JC8679, Stewart et al.)
containing plasmid-bome valS from Haemophilus influenzae
(pSU18valS). Recombinants were selected on Luria Broth agar plates
supplemented with chloramphenicol (strain JC8679 is CmR) and low
kanamycin (25 ug/ml) because efficiency of kanR markers can be
decreased when it recombines into the chromosome, one recombinant
chosen and named strain PS2838. PS2838 was infected with a PI phage
stock, phage were allowed to propagate for 3 hours then were stored
in CHCl3 at 4C. Transfer of genes and resistance markers was done
by P1 phage transduction using standard protocols (Miller 1972)
into MG1655 wild-type E. coli containing each of the valSpBAD
constructs expressing either wild-type ValRS, T222P, R223H, D230N,
V276A, and K277Q. Transductants were selected by plating on Luria
Broth agar plates supplemented with kanamycin (25 ug/ml) and 0.2%
arabinose, incubating at 25C for 48 hours.
[0120] As expected transduction of the MG1655 strain alone did not
yield colony growth. All transductants were restreaked onto media
supplemented with 50 ug/ml kanamycin to ensure kanR and insertion
of kanamycin marker into the E. coli chromosomal valS gene was
verified by amplifying with a forward primer that annealed to a
region on the chromosome just upstream of the valS gene (valS. 103)
and a reverse primer annealing to the marker itself (puc4k.o14).
Transduction yielded the following strains of valS::kanR: wt
ValRS-PS2847, T222P-PS2849, R223H-PS2862, D230N-PS2865,
V276A-PS2851, K277Q-PS2853.
[0121] Plasmid Construction: A cautious approach was taken in this
construction due to a lack of success introducing valS editing
mutants previously. With this in mind initial constructs were done
in a variation of pBAD 18, a plasmid allowing for tight regulation
of expression under control of an arabinose inducible promoter.
Parent plasmid was a Histidine tagged vals construction received
from Jack Horowitz which we verified by sequence analysis to done
be pET16b (Novagen). Utilizing an internal restriction site and a
restriction site located downstream of the valS gene in the
multiple cloning site of pET16b, a large BamHI fragment was
subcloned into pUC19 (pSC02). In parallel, the parent plasmid was
digested with SalI and BglII, restriction sites that flank the
region of valS encoding the CP1, and this small segment was ligated
into pUC19 (pSCO5).
[0122] The QuikChange mutagenesis kit (Stratagene, La Jolla,
Calif.) was used on pSCO5 to introduce an A to C base change at
nucleotide 663 to achieve the Thr to Pro substitution in residue
222 of ValRS. Both the wild-type and T222P SalI-BglII fragments
were subcloned into pSCO2. The resulting vectors were digested with
NcoI and SmaI to release fragments of the entire downstream region
of valS which were then cloned into pVDC441, a variation of pBAD 18
into which the upstream portion of valS from the pET16bvalS plasmid
had been cloned using EcoRI and KpnI. The His tagged T222P
construct was constructed by subcloning the mutated region of valS
(DraI/XhoI) from pVDC447 and ligating it into the pET16bvalS. The
wild-type valS constructs, pBAD 18valS and pET16bvalS, were
eventually used as templates in the QuickChange mutagenesis
protocol to generate mutations in valS corresponding to the mutants
R223H, D230N, V276A, and K277Q. The entire valS gene for these
vectors was then sequenced for each of these vectors to verify
their integrity.
[0123] Mutant Toxicity Response to High Levels of
.alpha.-Aminobutyrate: .DELTA.valS::kan+ strains expressing
wild-type ValRS, T222P, R223H, D230N, V276A, and K277Q were each
isolated on mineral standard medium (MS) (Richaud et al, 1993)
supplemented with 0.2% glycerol, 0.02% arabinose and ampicillin
(100 ug/mL). Single colonies from each strain were inoculated into
liquid cultures of the same media composition and grown overnight
to saturation. Cells were diluted in media 1:100 and a lawn of
cells was spread and allowed to dry onto MS medium plates.
.alpha.-aminobutyrate (50 uL of 0.1M) was added to a central well
created in each plate and plates were then incubated for 24 hours
at 42C. The relative response of each strain to high levels of
noncognate .alpha.-aminobutyrate could then be evaluated based on
diameter of toxicity halo observed.
[0124] Expression and Purification of ValRS Proteins: pET16b
plasmids corresponding to the wild-type ValRS and each of the ValRS
mutants were transformed into BL21 competent cells (Novagen). These
cells were cultured in 250 mL of Luria Broth supplemented with
ampicillin (100 ug/mL), when cells reached an optical density at
600 nm of 1.0 expression of ValRS was induced with 1 mM IPTG for 5
hours. Cells were then stored at -80C until purification. Cells
were resuspended (50 mM Na.sub.2PO.sub.4, 300 mM NaCl, 50 mM B-ME,
30 mM Imidazole pH 7.4) and lysed 2.times. in French Press. Lysates
were bound a Ni-NTA affinity column and eluted in lysis buffer with
a gradient ranging from 30 mM to 250 mM Imidazole. Collected
fractions were visualized on an 8% SDS-polyacyrlamide gel stained
with coomassie brilliant blue to ensure purity, purest fractions
were pooled and dialyzed into 25 mM Tris-HCl, 1 mM B-ME pH 7.5.
Enzyme concentrations were determined by Bradford assay.
[0125] Aminoaclyation Assay, Misacylation, and Deacylation:
Aminoacylation assays were performed at 37C in a 100 uL volume
containing buffer (20 mM HEPES, 0.1 mM EDTA disodium salt, 0.15M
NH.sub.4Cl, 10 ug/mL BSA pH 7.5), 2 M MgCl.sub.2, 0.7 .mu.M
[.sup.3H]Val, 20 .mu.M cold Val, 2 .mu.M tRNA.sub.val (Sigma) and
20 nM ValRS enzyme (adapted from Hendrickson et al. 2000). Aliquots
(10L) of the reaction mixture were precipitated with
tricholoracetic acid, and the level of aminoacylation of the tRNA
was determined by scintillation counting.
[0126] Misacylation of Thr onto tRNA.sub.val was performed in the
same conditions except 5.86 .mu.M [.sup.3H]Thr was substituted
forVal in the reaction. To test deacylation rates for these enzymes
it was necessary to generate [.sup.3H]Thr-tRNA.sub.val. This was
done using ValRS with the T222P substitution purified previously
(Doring et al, 2001), this enzyme was shown to form the
[.sup.3H]Thr-tRNAval complex. Enzyme was incubated with 2.5 .mu.M
tRNA.sub.val and 45 uM [.sup.3H]Thr, incubated at 37C for 45
minutes, extracted twice in phenol:chloroform, and ethanol
precipitated. Pellet was dissolved in 100 .mu.L sterile H.sub.2O
and scintillation counting was used to determine the success of the
reaction. Deacylation reactions, done in triplicate for each
enzyme, were performed in 150 mM Tris-HCl (pH 7.5), 100 .mu.g/ml
BSA, 10 mM MgCl2. At room temperature 2 nM enzyme was combined with
[.sup.3H]Thr-tRNA.sub.val, at 3, 6, 9, 16, and 30 minutes aliquots
(9 .mu.L) of the reaction mixture were precipitated with
tricholoracetic acid, and the amount of misaminoacylated
[3H]Thr-tRNA.sub.val remaining in the sample was determined by
scintillation counting. A no enzyme control reaction was included
to provide reference point.
[0127] When the percentage of Thr-tRNA.sup.Val hydrolyzed by each
enzyme in deacylation assays in vitro was compared to the observed
Abu-induced inhibition zone diameters in vivo, a pattern was clear
(FIG. 5A). The T222P and K277Q ValRS mutant proteins have the most
severe defects, for example, in deacylation of Thr tRNA.sup.Val in
vitro (FIG. 5B) Strains bearing these mutations show a toxic
response to only slightly raised levels of Abu in vivo. These two
enzymes also accumulated high levels of mischarged Thr-tRNA.sup.Val
in vitro. The D230N mutation appears to be less detrimental to the
deacylase activity, exhibiting a slowed rate of deacylation.
Similarly, toxicity in vivo was limited to intermediate levels of
Abu. Having the V276A mutation appears to impart the least toxic
response to exogenous Abu. This phenotype is reflected in a
deacylation rate that is closer to wild-type ValRS than to that of
the other mutant enzymes. Both D230N and V276A ValRS showed low
levels of mischarging with respect to the other mutant enzymes as
well. Thus, the ability of the mutant enzymes to hydrolyze in vitro
misaminoacylated amino acids from tRNA.sup.Val correlated with the
toxicity in vivo of exogenous noncognate amino acids that were
added to cells bearing the same mutations in their ValRS
Example 4
Construction of Strains of E. coli Expressing an Isoleucyl-tRNA
Synthetase Mutated in the Editing Site
[0128] Several studies have identified the editing domain of IleRS
(Schmidt at Schimmel, 1004, Science), and more particularly the
amino acids of the 239 to 250 region which are extremely conserved
in the IleRS sequences of different organisms. Point mutations in
this region generate enzymes partially or totally deficient in the
editing function (Hendrickson, 2000). An artificial allele of the
gene iles coding for isoleucyl-tRNA E. coli synthetase containing a
succession fine residues replacing residues 239 to 250 has been
constructed in the following manner: the the pVDC433 (Hendrickson
et al. 2001), derived from plasmid pBAD (Guzman et al., 1995) by
insertion of the wild type ileS gene is digested by restriction
enzymes ClaI and SpeI according to the instructions of the
supplier, thus eliminating nucleotides 717 to 750 of the gene.
[0129] The 5' phosphorylated oligonucleotides (Invitrogen Life
Technologies):
3 5'pCTAGTAATCGCGGCGGCGGCGGCGGCGGCGGCGGCGGCGGCGCGCGC AATAT and
5'pCGATATTGCGCGCGCCGCCGCCGCCGCCGCCGCCGCCGCCGC- CGCGA TTA
[0130] are hybridized and ligated with the plasmid pVDC433,
previously cut by SlaI and SpeI under the following conditions: 0.3
pmoles of vector and 3 pmoles of each oligonucleotide are incubated
for 10 minutes at 70 degrees C. after precipitation and
resuspension in 10 ul of H.sub.2O, after return to room
temperature; the ligation is carried out using standard protocols.
The ligation mixture is utilized to transform competent cells of
the SureTM strain (Stratagene) by electroporation following the
protocol of the supplier. The transformants are obtained and
plasmid DNA prepared (Qiaprep Kit, Qiagen). The plasmid obtained in
this manner, pTLH33, is sequenced and the insertion of 11 alanine
codons is verified.
[0131] The pTLH33 plasmid is inserted in the wild type E. coli
strain MG1655 to obtain strain PS2419. The allele obtained by
deletion of the ileS gene, deltaileS203::kan, from strain IQ839
(Shiba and Schimmel, 1992) is introduced in strain PS2419 by P1
transduction according to a standard protocol (Miller, 1972), and a
transducant, strain PS 2449, is obtained in rich medium
supplemented with kanamycin (25 mg/L) and arabinose (0.02%). In the
same way, strain PS2306 is constructed by insertion of the pVDC433
plasmid in strain MG1655 (resulting strain: PS27752) and P1
transduction of allele delta ileS203::kan in strain PS2752. Growth
of the two strains deleted in the ileS locus, PS2306 and PS2449,
requires the presence of arabinose, thus demonstrating the
controlled expression of the wild type ileS gene (plasmid PVDC433,
strain PS2306) and the mutated gene ileS alal 1 (plasmid pTLH33,
strain PS2449) respectively.
Example 5
Sensitivity of the Isoleucyl-tRNA Synthetase to Non Canonic Amino
Acids
[0132] Strains PS2306 and PS2449 are tested for their sensitivity
to artificial amino acids whic present a steric resemblance to
isoleucine. The cells are cultivated in MS mineral medium,
succinate (0.2%), arabinose (0.2%) for 24 h at 37 degrees
centigrade and diluted at {fraction (1/250)} in MS mineral medium.
0.5 ml of this cell suspension are spread on a Petri dish
containing 25 ml of MS medium supplemented with succinate (0.2%)
and arabinose (0.2%). A well is then prepared in the middle of the
dish and filled with 0.1 ml of an amino acid solution:
[0133] 1) 25 mM S-methyl cysteine
[0134] 2) 25 mM homocysteine
[0135] 3) 25 mM o-methyl-L-serine
[0136] 4) 25 mM Norleucine
[0137] 5) 25 mM Norvaline
[0138] The Petri dishes are then incubated at 37 degrees centigrade
for 24 hours and the appearance of a zone of inhibition around the
well is recorded. The diameters of the zones of attenuated growth
are measured as follows:
4 TABLE 3 Amino acid PS2306 PS2449 S-methyl-L-cysteine 1, 3 cm 3, 0
cm Homocysteine 2, 1 cm 2, 9 cm O-methyl-L-serine 2, 2 cm 5, 6 cm
Norleucin 2, 9 cm 3, 4 cm Norvaline 3, 7 cm 8, 0 cm
[0139] All 5 non-canonic amino acids tested inhibit the growth of
the two strains, but it is noted that in all cases there is a
stronger inhibition on the strain expressing the mutated allele of
the ileS gene in the editing site. Thus the mutated isoleucyl-tRNA
synthetase seems to have an increased specificity for the substrate
amino acid capable of loading tRNAile with non-natural amino
acids.
Example 6
Biochemical Characterization of the IleRS239-250Ala Mutant
[0140] The wild type and mutant IleRS enzymes were purified from
strains PS2306 and PS2449 by chromatography using the procedure
described by Hendrickson et al. (2002). Ile-tRNA loaced with the
amino acid Val (Val-t-RNAIle) can be produced in large quantity by
utilizing the editing mutant IleRS T242P (Hendrickson 2000). The
utilzation of this Val-tRNA/Ile has allowed testing of deacylation
activity of several enzymes using the protocol described by
Hendrickson, 2000. The results obtained, shown in FIG. 4, clearly
show that the IleRS239-250 mutant is deficient in the deacylation
function of Val-tRNAIle.
BIBLIOGRAPHY
[0141] 1. U. L. RajBhandary, J. Bacteriol. 176, 547-52 (1994).
[0142] 2. A. Bock, et al., Mol. Microbiol. 5, 515-20 (1991).
[0143] 3. A. L. Weber, S. L. Miller, J. Mol. Evol. 17, 273-84
(1981).
[0144] 4. I. G. Fotheringham, N. Grinter, D. P. Pantaleone, R. F.
Senkpeil, P. P. Taylor, Biorg. Med. Chem. 7, 2209-13 (1999).
[0145] 5. M. Ibba, D. Soll, Annu. Rev. Biochem. 69, 617-50
(2000).
[0146] 6. R. B. Loftfield, D. Vanderjagt, Biochem. J. 128,
1353-1356 (1972).
[0147] 7. A. N. Baldwin, P. Berg, J. Biol. Chem. 241, 839-845
(1966).
[0148] 8. E. W. Eldred, P. R. Schimmel, J. Biol. Chem. 247, 2961-4
(1972).
[0149] 9. A. Fersht, in Structure and mechanism in Protein Science.
(Freeman, N.Y., 1999) pp. 389-399.
[0150] 10. K. Musier-Forsyth, P. J. Beuning, Nat. Struct. Biol. 7,
435-436 (2000).
[0151] 11. H. Jakubowski, A. R. Fersht, Nucleic Acids Res. 9,
3105-17 (1981).
[0152] 12. D. R. Liu, P. G. Schultz, Proc. Natl. Acad. Sci. U.S.A.
96, 4780-5 (1999).
[0153] 13. M. Belfort, G. Maley, J. Pedersen-Lane, F. Maley, Proc.
Natl. Acad. Sci. U.S.A. 80, 4914-8 (1983).
[0154] 14. V. Doring, P. Marliere, Genetics 150, 543-51 (1998).
[0155] 15. B. Lemeignan, P. Sonigo, P. Marliere, J. Mol. Biol. 231,
161-6 (1993).
[0156] 16. I. K. Dev, B. B. Yates, J. Leong, W. S. Dallas, Proc.
Natl. Acad. Sci. U.S.A 85, 1472-6 (1988).
[0157] 17. The 64 alleles of thyA with different codons at position
146 of the coding sequence were constructed as follows. First, a
unique NheI site was introduced through a G429.fwdarw.A
substitution in the thyA coding region by site-directed mutagenesis
(24) of plasmid pTSO (14) to yield plasmid pTSO1. Oligonucleotides
THY1 (5'-CTGGATAAAATGGCGCTAGCACCGTG- CCATGCATTC-3') and THY2
(5'-TCTGCCACATAGAACTGGAAGAATGCATGGCACGGT-3') were used for this
mutation which preserved the sense of the codon thus mutated.
Plasmid pTSO1 was then digested with NheI and NsiI to remove from
the thyA coding region an 18 bp fragment containing codon 146
(UGC). All 64 oligonucleotides of the thyA coding sequence from
nucleotides 427 to 444 and the 64 oligonucleotides of the partial
reverse sequence were constructed (GENAXIS Biotechnology, Montigny
le Bretonneux, France). The 64 pairs of complementary
oligonucleotides were annealed and ligated with the digested
plasmid pTSO1.
[0158] 18. Cysteine gradient plates were done as described in FIG.
1B legend. The effects of L-valine alone could not be directly
examined because exogenous L-valine is known to inhibit growth of
E. coli K12 in minimal medium (25). This inhibition is relieved if
L-isoleucine is also supplied. Thus, the Ile-Val dipeptide was used
as a valine source, because this dipeptide is transported across
the cell membrane and then broken down to isoleucine and valine
(26).
[0159] 19. V. Doring, et al., Supplementary data can be found at
the Science Web site.
[0160] 20. M. J. Pine, Antimicrobio. Agents Chemother. 13, 676-85
(1978).
[0161] 21. E. coli chromosomal DNA was extracted using a DNeasy
Tissue Kit (Qiagen GmbH, Hilden, Germany) following the
instructions of the manufacturer. PCR to amplify the valS gene was
performed as follows: denaturation at 94.degree. C. for 3 min,
followed by 30 cycles of 30 sec at 94.degree. C., annealing at
57.degree. C. for 30 s, and primer extension at 72.degree. C. for
200 sec. The final step was a primer extension at 72.degree. C. for
600 sec. The reaction was carried out using 2 units of Vent DNA
polymerase (New England Biolabs, Beverly, Mass.) and 100 ng of
chromosomal DNA in a 100 .mu.l reaction mixture. The following
primers were used: VAL1 (5'-GGGGAATTCGGTGTGTGAAATTGCCGCAGAACG-3-
'), and VAL2 (5'-GGCAAGCTTTCAGGTATTTGCTGCCCAGATCGA-3'). Two
independent PCR amplification products of each mutant were
sequenced (GENAXIS Biotechnology, Montigny le Bretonneux,
France).
[0162] 22. L. Lin, S. P. Hale, P. Schimmel, Nature 384, 33-4
(1996).
[0163] 23. O. Nureki, et al., Science 280, 578-82 (1998).
[0164] 24. M. Ansaldi, M. Lepelletier, V. Mejean, Anal. Biochem.
234, 110-1 (1996).
[0165] 25. M. De Felice, et al., J. Mol. Biol. 156, 1-7 (1977).
[0166] 26. A. J. Sussman, C. Gilvarg, Annu. Rev. Biochem. 40,
397-408 (1971).
[0167] 27. C. Richaud, et al., J. Biol. Chem. 268, 26827-35
(1993).
[0168] 28. E. Schmidt, P. Schimmel, Biochemistry 34, 11204-10
(1995).
[0169] 29. L. Lin, P. Schimmel, Biochemistry 35, 5596-601
(1996).
[0170] 30. L. M. Guzman, D. Belin, M. J. Carson, J. Beckwith, J.
Bacteriol. 177, 4121-30 (1995).
[0171] 31. T. L. Hendrickson, T. K. Nomanbhoy, P. Schimmel,
Biochemistry 39, 8180-8186 (2000).
[0172] 32. K. Gevaert, J. Vandekerckhove, Electrophoresis 21,
1145-54 (2000).
[0173] 33. mutS::spc+ allele (gift of F. Taddei, Hopital
Necker-Enfants, Paris); .DELTA.nrdD::kan+ allele, laboratory
collection; AdnaQ::tet allele (gift of J. Shapiro, University of
Chicago, Ill.); CU505 (.DELTA.(ilvE-ilvC)2049 leu455 galT1
IN(rrnD-rrnE)1) gift of Dr. M. Berlyn from the E. coli Genetic
Stock Center (New Haven, Conn.). Transfer of genes and of
resistance markers by P1 transduction were carried out using
standard protocols (35).
[0174] 34. P. Modrich, Annu. Rev. Genet. 25, 229-53 (1991).
[0175] 35. J. H. Miller, Experiments in Molecular Genetics (Cold
Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.,
1972).
[0176] 36. Guzman, L. M., Belin, D., Carson, M. J., and Beckwith,
J. (1995). Tight regulation, modulation, and high-level expression
by vectors containing the arabinose PBAD promoter. J. Bacteriol.
177, 4121-4130
[0177] 37. Hendrickson, T. L., Nomanbhoy, T. K., and Schimmel, P.
(2000). Errors from selective disruption of the editing center in a
tRNA synthetase. Biochemistry 39, 8180
[0178] 38. Hendrickson, T. L., Nomanbhoy, T. K., de Crecy-Lagard,
V., Fukai, S., Nureki, O., Yokoyama, S., and Schimmel, P. (2002).
Mutational separation of two pathways for editing by a class I tRNA
synthetase. Mol. Cell 9, 353-362
[0179] 39. Schmidt, E., and Schimmel, P. (1994). Mutational
isolation of a sieve for editing in a transfer RNA synthetase.
Science 264, 265-267
[0180] 40. Shiba, K., and Schimmel, P. (1992). Functional assembly
of a randomly cleaved protein. Proc. Natl. Acad. Sci. USA 89,
9964-9968
Sequence CWU 1
1
6 1 52 DNA Artificial Sequence synthetic oligonucleotide 1
ntagtaatcg cggcggcggc ggcggcggcg gcggcggcgg cgcgcgcaat at 52 2 50
DNA Artificial Sequence synthetic oligonucleotide 2 ngatattgcg
cgcgccgccg ccgccgccgc cgccgccgcc gccgcgatta 50 3 36 DNA Artificial
Sequence synthetic oligonucleotide 3 ctggataaaa tggcgctagc
accgtgccat gcattc 36 4 36 DNA Artificial Sequence synthetic
oligonucleotide 4 tctgccacat agaactggaa gaatgcatgg cacggt 36 5 33
DNA Artificial Sequence primer 5 ggggaattcg gtgtgtgaaa ttgccgcaga
acg 33 6 33 DNA Artificial Sequence primer 6 ggcaagcttt caggtatttg
ctgcccagat cga 33
* * * * *