Note: Descriptions are shown in the official language in which they were submitted.
CA 02294619 1999-12-21
WO 98/59060 PCT/US98/12966
METHODS OF PREFORMING HOMOLOGOUS RECOMBINATION BASED
MODIFICATION OF NUCLEIC ACIDS IN RECOMBINATION DEFICIENT CELLS
AND USE OF THE MODIFIED NUCLEIC ACID PRODUCTS THEREOF
GOVERNMENTAL SUPPORT
The research leading to the present invention was supported, at least in part,
by a grant from
the National Science Foundation Grant No. MCB-9316625. Accordingly, the
Government
may have certain rights in the invention.
FIELD OF THE INVENTION
This invention relates generally to methods of modifying genes with
specificity in
recombination deficient cells by transiently enabling homologous recombination
in the cells.
Included in the invention are conditional replication shuttle vectors which
bestow transient
recombination capabilities to an otherwise recombination deficient cell. The
independent
origin based cloning vectors containing the modified genes and methods of
using the
independent origin based cloning vectors containing the modified genes are
also included in
the presentinvention.
BACKGROUND OF THE INVENTION
Functional analyses of genes in vivo frequently involve the introduction of
modified genomic
DNA into the germline to generate transgenic animals [Jaenisch et al., Science
240:1468
(1985); Brinster, Ce1141:343 (1985)]. The genomic DNA sequences containing
introns and
essential regulatory sequences have been shown to be expressed in vivo in
cases where
simple cDNA constructs cannot be expressed [Brinster et al.,
Proc.Natl.Acad.Sci. 85:836-840
( 1988)]. Furthermore, the size of the genomic DNA that can be readily
manipulated in vitro
and introduced into the germline can be a critical determinant of the outcome
of the
functional analysis of a gene since elements that are important for high
level, tissue specific
and position-independent expression of the transgene may be located at a long
distance from
the gene itself [Dillon et al., Trends Genet. 9:134 (1993); Kennison, Trends
Genet. 9:75
(1993); Wilson et al., Annu.Rev.Cell.Biol. 6:679 (1990)].
On the other hand, the use of such large genomic transgenes has several
practical problems.
For example, the size of the transgene is presently limited due to constraints
on the sequence
CA 02294619 1999-12-21
WO 98/59060 PCT/US98/12966
2
length that can be cloned and stably maintained in a conventional plasmid or a
cosmid. Thus
DNA sequences suspected of being nonessential are often omitted when designing
the
constructs to be transferred because of the size limitation. In addition, in
vitro manipulations
of large DNAs oftentimes lead to mechanical shear [Peterson et al., TIG 13:61-
66].
Yeast artificial chromosomes (YACs) allow large genomic DNA to be modified and
used for
generating transgenic animals [Burke et al., Science 236:806; Peterson et al.,
Trends Genet.
13:61 (1997); Choi, et al., Nat. Genet., 4:117-223 (1993), Davies, et al.,
Biotechnology
11:911-914 (1993), Matsuura, et al., Hum. Mol. Genet., 5:451-459 (1996),
Peterson et al.,
Proc. Natl. Acad. Sci., 93:6605-6609 (1996); and Schedl, et al., Cell, 86:71-
82 (1996)].
Other vectors also have been developed for the cloning of large segments of
mammalian
DNA, including cosmids, and bacteriophage P1 [Sternberg et al., Proc. Natl.
Acad. Sci.
U.S.A., 87:103-107 (1990)]. YACs have certain advantages over these
alternative large
capacity cloning vectors [Burke et al., Science, 236:806-812 (1987)]. The
maximum insert
size is 35-30 kb for cosmids, and 100 kb for bacteriophage P1, both of which
are much
smaller than the maximal insert for a YAC. However, there are several critical
limitations in
the YAC system including difficulties in manipulating YAC DNA, chimerism and
clonal
instability [Green et al., Genomics, 11:658 (1991); Kouprina et al., Genomics
21:7 {1994);
Larionov et al., Nature Genet. 6:84 ( 1994)]. As a result, generating
transgenic mice with an
intact YAC remains a challenging task [Burke et al., Science 236:806; Peterson
et al., Trends
Genet. 13:61 (1997)].
An alternative to YACs are E. coli based cloning systems based on the E. coli
fertility factor
that have been developed to construct large genomic DNA insert libraries. They
are bacterial
artificial chromosomes (BACs) and P-1 derived artificial chromosomes (PACs)
[Mejia et al.,
Genome Res. 7:179-186 (1997); Shizuya et al., Proc. Natl. Acad. Sci. 89:8794-
8797
(1992);Ioannou et al., Nat. Genet., 6:84-89 (1994); Hosoda et al., Nucleic
Acids Res. 18:3863
(1990)]. BACs are based on the E. coli fertility plasmid (F factor); and PACs
are based on
the bacteriophage P 1. The size of DNA fragments from eukaryotic genomes than
can be
stably cloned in Escherichia coli as plasmid molecules has been expanded by
the advent of
PACs and BACs. These vectors propagate at a very low copy number (1-2 per
cell) enabling
ger.:~:nic inserts up to 300 kb in size to be stably maintained in
recombination deficient hosts
(most clones in human genomic libraries fall within the 100-200kb size range).
The host cell
is required to be recombination deficient to ensure that non-specific and
potentially
CA 02294619 1999-12-21
WO 98/59060 PCT/US98/12966
3
deleterious recombination events are kept to a very minimum. As a result,
libraries of PACs
and BACs are relatively free of the high proportion of chimeric or rearranged
clones typical
in YAC libraries, [Monaco et al., Trends Biotechnol 12:280-286 (1994); Boyseu
et al.,
Genome Research, 7:330-338 (1997)]. In addition, isolating and sequencing DNA
from
PACs or BACs involves simpler procedures than for YACs, and PACs and BACs have
a
higher cloning efficiency than YACs [Shizuya et al., Proc. Natl. Acad. Sci.
89:8794-8797
(1992);Ioannou et al., Nat. Genet., 6:84-89 (1994); Hosoda et al., Nucleic
Acids Res. 18:3863
(1990)]. Such advantages have made BACs and PACs important tools for physical
mapping
in many genomes [Woo et al., Nucleic Acids Res., 22:4922 (1994); Kim et al.,
Proc.Natl.Acad.Sci. 93:6297-6301 (1996); Wang et al., Genomics 24:527 (1994);
Wooster et
al., Nature 378:789 (1995)]. Furthermore, the PACs and BACs are circular DNA
molecules
that are readily isolated from the host genomic background by classical
alkaline lysis
[Birnboim et al., Nucleic Acids Res. 7:1513-1523 (1979].
Functional characterization of a gene of interest contained by a PAC or BAC
clone generally
entails transferring the DNA into a eukaryotic cell for transient or long-term
expression. A
transfection reporter gene, e.g., a gene encoding lacZ, together with a
selectable marker, e.g.,
neo, can be inserted into a BAC [Mejia et al., Genome Res. 7:179-186 (1997).
Transfected
cells can be then detected by staining for X-Gal to verify DNA uptake. Stably
transformed
cells are selected for by the antibiotic 6418.
However, while PACs and BACs have cloning capacities up to 350kb, performing
homologous recombination to introduce mutations into a gene of interest has
not been
demonstrated [Peterson et al., TIG 13:61-66]. Indeed, although BACs or PACs
have become
an important source of large genomic DNA in genome research, there are still
no methods
available to modify the BACs or PACs. Furthermore, no gerrnline transmission
of intact
BACs or PACs in transgenic mice have been reported. These, as well as other
disadvantages
of BACs and PACs greatly limit their potential use for functional studies.
Therefore, there is
a need for an improved cloning vector for germline transmission of selected
genes in
transgenic animals. More particularly there is a need for a cloning vector
that has the
capacity to contain greater than 100 kilobases of DNA, which can be readily
manipulated and
isolated, but still can be stably stored in libraries relatively free of
rearranged clones. In
addition, there is a need to provide methodology for generating such cloning
vectors. There
CA 02294619 1999-12-21
WO 98/59060 PCT/US98/12966
4
is also a need to apply such vectors to improve current technologies such as
gene targeting.
Gene targeting has been used in various systems, from yeast to mice, to make
site specific
mutations in the genome. Gene targeting is not only useful for studying
function of proteins
in vivo, but it is also useful for creating animal models for human diseases,
and in gene
therapy. The technique involves the homologous recombination between DNA
introduced
into a cell and the endogenous chromosomal DNA of the cell. However, in the
vertebrate
system, the rate of homologous recombination is very low, as compared to
random
integration. The only cell line that allows a relatively high homologous
recombination rate
and maintains the ability to populate the germline is the murine 129 embryonic
stem cells
(ES cells). Using this specialized cell, mice can be generated with a targeted
mutation
(Gene Targeting, a practical approach Ed. by A. Joyner, IRL Press: Oxford, New
York,
Tokyo). However, the rate of homologous recombination for some gene loci in ES
cells is
still extremely low ( < 1 % ), the procedure is labor intensive, and the cost
of generating
targeted mutant mice is very expensive. Moreover, since there are no ES cells
available for
vertebrates other than mice, gene targeting in a germline is still not
possible for other
vertebrates.
The major limitation for gene targeting in vertebrate cells remain to be the
low targeting
frequency. One critical factor affecting the targeting frequency is the total
length of
homology. Deng and Capecchi (MCB, 12:3365-3371) have shown that gene targeting
frequency is linearly-dependent on the logarithm of the total homology length
over
homology lengths of 2.8kb to 14.6kb. Since the curve did not plateau at the
14.6kb
homology, it is likely that incorporating greater homology lengths into the
targeting vector
will further increase the homologous recombination rate. Using a mathematical
model
developed by Fujitani et al, [Genetics, 140:797-809, (1995)], an estimate can
be made that
with a total homology of 100kb isogenous DNA (i. e. , DNA from the same strain
of mice),
the gene targeting rate in ES cells would be 10% . This is a dramatic
improvement over the
conventional 14.6kb targeting vector, which only yields a corresponding rate
of
only 0.03 % . Further support for the present strategy i. e. , using a large
DNA construct for
gene targeting rate comes from an experiment with Mycobacterium tuberculosis,
the causal
agent of tuberculosis. Like vertebrate cells, gene targeting in TB has a very
low rate,
mainly due to the predominance of random integration over homologous
recombination. It
has been demonstrated that using a 40-SO kb linear targeting construct, a 6%
targeting
CA 02294619 1999-12-21
WO 98/59060 PCT/US98/12966
frequency could be obtained, whereas no targeting event was obtained at all
with a smaller
( < lOkb) targeting construct [Balasubramanian et al., J. of Bacteriology
178:273-279
(1996)]. Therefore, there is a need to construct large gene targeting
constructs to allow
efficient gene targeting in many biological systems.
The citation of any reference herein should not be construed as an admission
that such
reference is available as "Prior Art" to the instant application.
SUMMARY OF THE INVENTION
The present invention provides a novel and efficient method of modifying
independent origin
based cloning vectors for in vitro and in vivo gene expression. In its
broadest embodiment,
the present invention provides a method of selectively performing homologous
recombination on a particular nucleotide sequence contained in a recombination
deficient
host cell, i.e., a cell that cannot independently support homologous
recombination. The
method employs a recombination cassette which contains a nucleic acid that
selectively
integrates into the particular nucleotide sequence when the recombination
deficient host cell
I 5 is induced to support homologous recombination. The method comprises
introducing the
recombination cassette into the recombination deficient host cell, and
inducing the
recombinantly deficient host cell to transiently support homologous
recombination, thereby
allowing the nucleic acid to integrate into the particular nucleotide
sequence. In a preferred
embodiment, unselected nucleotide sequence rearrangements and deletions, which
are
characteristic of host cells that support homologous recombination, are not
evident with
restriction endonuclease digestion map analysis with a restriction enzyme such
as HindIII,
EcoRI, XhoI, or AvrII. In a more preferred embodiment, unselected nucleotide
sequence
rearrangements and deletions are not evident with restriction endonuclease
digestion map
analysis with two or more restriction enzymes.
In a particular aspect of the present invention, the recombination deficient
host cell cannot
independently support homologous recombination because the host cell is RecA-.
In this
aspect of the invention, inducing the host cell to transiently support
homologous
recombination comprises inducing the transient expression of a RecA-like
protein in the host
cell. In a preferred embodiment, inducing the transient expression of the RecA-
like protein
can be performed with a conditional replication shuttle vector. In a more
preferred
CA 02294619 1999-12-21
WO 98/59060 PCT/US98/12966
6
embodiment the conditional replication shuttle vector is a temperature
sensitive shuttle
vector (TSSV) that replicates at a permissive temperature, but does not
replicate at a
non-permissive temperature.
In one particular embodiment of this type, inducing the transient expression
of the RecA-like
protein comprises transforming the host cell with the TSSV at a permissive
temperature, and
growing the host cell at a non-permissive temperature. The TSSV encodes a RecA-
like
protein that is expressed in the host cell and supports the homologous
recombination between
a nucleic acid contained in a recombination cassette and the particular
nucleotide sequence
contained in the host cell. The TSSV encoding the RecA-like protein is diluted
out when the
host cell is grown at the non-permissive temperature. In one particular
embodiment of this
type the permissive temperature is 30°C and the non-permissive
temperature is 43 °C.
In a more intricate version of the present invention, the particular
nucleotide sequence which
has been selected to undergo homologous recombination is contained in an
independent
origin based cloning vector (IOBCV) that is comprised by the host cell, and
neither the
independent origin based cloning vector alone, nor the independent origin
based cloning
vector in combination with the host cell, can independently support homologous
recombination. In a particular embodiment of this type both the independent
origin based
cloning vector and the host cell are RecA-, and inducing the host cell to
transiently support
homologous recombination comprises inducing the transient expression of the
RecA-like
protein to support homologous recombination in the host cell. In one
particular embodiment
the independent origin based cloning vector is a Bacterial or Bacteriophage-
Derived
Artificial Chromosome (BBPAC) and the host cell is a host bacterium.
In a preferred embodiment, inducing the transient expression of the RecA-like
protein is
performed with a conditional replication shuttle vector that encodes the RecA-
like protein.
In a more preferred embodiment the conditional replication shuttle vector is a
temperature
sensitive shuttle vector {TSSV) that replicates at a permissive temperature,
but does not
replicate at a non-permissive temperature. In one particular embodiment of
this type the
permissive temperature is 30°C and the non-permissive temperature is
43°C.
In one embodiment the RecA-like protein is controlled by an inducible promoter
and the
transient expression of the RecA-like protein is achieved by the transient
induction of the
CA 02294619 1999-12-21
WO 98/59060 PCT/US98/12966
7
inducible promoter in the host cell. In another embodiment, the RecA-like
protein is
controlled by a constitutive promoter with the transient expression induced by
the TSSV.
In a preferred embodiment the TSSV also comprises a recombination cassette and
a first
gene which bestows resistance to a host cell that contains the TSSV against a
first toxic
agent. In addition, the first gene can be counter-selected against. The
recombination
cassette, the RecA-like protein gene, and the first gene are linked together
on the TSSV such
that when the nucleic acid integrates (i.e. resolved) into the particular
nucleotide sequence,
the RecA-like protein gene and the first gene remain linked together, and
neither the
RecA-like protein gene nor the first gene remain linked to the integrated
nucleic acid.
In a particular embodiment of this type, the independent origin based cloning
vector is a
BBPAC and the host cell is a bacterium. The BBPAC further contains a second
gene that
bestows resistance to the host cells against a second toxic agent. Introducing
the
recombination cassette into the host cells is performed by transforming the
host cell with the
TSSV. Inducing the transient expression of the RecA-like protein to support
homologous
recombination comprises: (i) incubating the host cells at a permissive
temperature in the
presence of the first toxic agent and the second toxic agent, wherein
transformed host cells
containing the TSSV and the BBPAC are selected for and wherein the RecA-like
protein is
expressed. A first homologous recombination event occurs between the
recombination
cassette and the particular nucleotide sequence forming a co-integrate between
the TSSV and
the BBPAC, wherein the TSSV is either free or part of a co-integrate; (ii)
incubating the
transformed host cells at a non-permissive temperature in the presence of the
first toxic agent
and the second toxic agent, wherein host cells containing a TSSV co-integrate
are selected
for, and wherein free TSSV cannot replicate; (iii) selecting a host cell
containing a co-
integrate between the TSSV and the BBPAC by Southern analysis; (iv) incubating
the host
cells containing a co-integrate between the TSSV and the BBPAC at a non-
permissive
temperature in the presence of the second toxic agent, wherein a second
homologous
recombination event occurs between the recombination cassette and the
particular nucleotide
sequence, therein integrating the nucleic acid into the particular nucleotide
sequence and
forming a resolved host cell, i.e., a host cell containing a resolved BBPAC;
and (v)
incubating the host cells containing the resolved BBPAC in the presence of the
second toxic
agent, and a counter-selecting agent, and wherein the counter-selecting agent
is toxic to host
cells containing the first gene, and wherein host cells containing the RecA-
like protein gene
CA 02294619 1999-12-21
WO 98/59060 PCT/US98/12966
8
are removed. Another embodiment further comprises selecting a host cell
containing the
resolved BBPAC by colony hybridization with a labeled probe that binds to a
DNA
homologue of the nucleic acid, an mRNA homologue of the nucleic acid, and/or a
protein
encoded by the nucleic acid. In a particular embodiment, the permissive
temperature is
30°C, the non-permissive temperature is 43°C. In a preferred
embodiment the incubating of
host cells containing the resolved BBPAC in the presence of the second toxic
agent and
counter-selecting agent is performed at 37°C.
Preferred embodiments further comprise the generating of the recombination
cassette by
placing a first genomic fragment 5' of the specific nucleic acid that is to
selectively integrate
into the particular nucleotide sequence, and placing a second genomic fragment
3' of the
specific nucleic acid. The first genomic fragment corresponds to a region of
the particular
nucleotide sequence that is 5' to the region of the particular nucleotide
sequence that
corresponds to the second genomic fragment. Thus, both the first genomic
fragment and the
second genomic fragment contain portions of the particular nucleotide
sequence. In one such
embodiment, both the first genomic fragment and the second genomic fragment
contain 250
or more basepairs of the particular nucleotide sequence. In a preferred
embodiment, the first
and second genomic fragments are about the same size. In another embodiment,
both the
first genomic fragment and the second genomic fragment contain 500 or more
basepairs of
the particular nucleotide sequence. In still another embodiment, both the
first genomic
fragment and the second genomic fragment contain 1000 or more basepairs of the
particular
nucleotide sequence. In one particular embodiment the recombination cassette
is generated
in a building vector and the recombination cassette is subsequently
transferred to the TSSV.
In a particular embodiment the first gene confers tetracycline resistance and
the counter-
selecting agent is fusaric acid. In a preferred embodiment the RecA-like
protein is recA. In
the more preferred embodiment the TSSV is pSVl.RecA having the ATCC no. 97968.
In a related aspect of the present invention the RecA-like protein is
controlled by an
inducible promoter, and the transient expression of the RecA-like protein is
achieved by the
transient induction of the inducible promoter in the host cell. In one
embodiment of this
type, the independent origin based cloning vector is a BBPAC and the
recombination
deficient host cell is an E. coli bacterium. In a preferred embodiment the
RecA-like protein
is recA.
CA 02294619 1999-12-21
WO 98/59060 PCT/US98/12966
9
The present invention also provides a conditional replication shuttle vector
that encodes a
RecA-like protein. In one such embodiment the RecA-like protein is controlled
by an
inducible promoter. In a preferred embodiment the conditional replication
shuttle vector is a
temperature sensitive shuttle vector (TSSV). The RecA-like protein of the TSSV
can be
controlled by either a constitutive promoter or by an inducible promoter. In
one embodiment
the TSSV contains a gene that can be counter-selected against. In a specific
embodiment of
this type the TSSV contains a gene that confers tetracycline resistance. In
another
embodiment the TSSV contains a RecA-like protein that is recA. In still
another
embodiment the TSSV contains both a gene that confers tetracycline resistance
and a RecA-
like protein that is recA. In a preferred embodiment the TSSV is pSVl.RecA
having the
ATCC no. 97968.
The present invention also provides an independent origin based cloning vector
that contains
a particular nucleotide sequence that has undergone homologous recombination
with a
conditional replication shuttle vector in a RecA- host cell, wherein the
conditional replication
shuttle vector encodes a RecA-like protein. In one such embodiment the
particular
nucleotide sequence is part of the gene that encodes the murine zinc finger
gene, RU49
which is contained by the independent origin cloning vector. In one prefer ed
embodiment
the independent origin based cloning vector has undergone homologous
recombination with
a temperature sensitive shuttle vector in a RecA- host cell, wherein the
temperature sensitive
shuttle vector encodes a RecA-like protein. In another embodiment the
independent origin
based cloning vector is a BBPAC, and more preferably a BAC. In a specific
embodiment of
this type the independent origin based cloning vector has undergone homologous
recombination with a temperature sensitive shuttle vector that is pSVl.RecA
having the
ATCC no. 97968.
The present invention also provides methods of using the modified independent
origin based
cloning vectors of the present invention to make transgenic animals, perform
gene targeting,
or perform gene therapy. The independent origin based cloning vectors or
linearized nucleic
acid inserts derived from the IOBCVs, for example, can be introduced into a
eukaryotic cell
or animal.
CA 02294619 1999-12-21
WO 98/59060 PCT/US98/12966
In one such embodiment the eukaryotic cell is a fertilized zygote. In another
embodiment the
eukaryotic cell is a mouse ES cell. The gene targeting can be performed to
modify a
particular gene, or to totally disrupt the gene to form a knockout animal.
In this aspect of the present invention, the independent origin based cloning
vector contains a
5 nucleic acid that has undergone homologous recombination with a conditional
replication
shuttle vector in a RecA- whole cell, in which the conditional replication
shuttle vector
includes a RecA like protein. In a preferred embodiment the independent origin
based
cloning vector is a BBPAC. In a more preferred embodiment, the BBPAC has
undergone
homologous recombination with a TSSV. In the most preferred embodiment, the
BBPAC
10 has undergone homologous recombination with the TSSV that is pSVl.RecA
having the
ATCC no. 97968.
One particular embodiment is a method of using the BBPAC to introduce the
nucleic acid
into an animal to make a transgenic animal comprising pronuclear injecting of
the BBPAC
(or a linearized nucleic acid insert derived from the BBPAC) into a fertilized
zygote. In one
embodiment the animal is a mammal. In a more preferred embodiment the mammal
is a
mouse. In a specific embodiment of this type the independent origin based
cloning vector is
a BBPAC and the fertilized zygote is a C57BL/6 mouse zygote. In a preferred
embodiment
of this type two picoliters (pl) of less than one ug/ml BBPAC DNA is injected.
In a more
preferred embodiment 2p1 of 0.6 lcg/ml of DNA is injected.
The present invention also includes a method of using the BBPAC of the
invention to
perform gene targeting in a vertebrate cells comprising introducing the BBPAC
into the
vertebrate cell wherein the nucleic acid that has undergone homologous
recombination with
the conditional shuttle vector, undergoes homologous recombination with the
endogenous
chromosomal DNA of the vertebrate cell. In preferred embodiments of this type
the
vertebrate cell is a mammalian cell. In a more preferred embodiment of this
type the
mammalian cell is a human cell. In a related embodiment the vertebrate cell is
a fertilized
zygote and the nucleic acid contains a disrupted gene. In a preferred
embodiment the
conditional shuttle vector is a TSSV. In a more preferred embodiment the TSSV
is
pSVl.RecA having the ATCC no. 97968.
CA 02294619 1999-12-21
WO 98/59060 PCT/US98/12966
11
The present invention also contains kits for performing homologous
recombination on
selected nucleotide sequences contained on an independent origin based cloning
vector, such
as a BBPAC. In one particular embodiment, the kit comprises a conditional
replication
shuttle vector and a building vector. In a preferred embodiment of this type,
the kit further
contains a restriction map for the shuttle vector and/or a restriction map for
one or more of
the building vectors. In a more preferred embodiment, the kit further includes
a protocol for
using the contents of the kit to perform homologous recombination.
A particular embodiment of the kit contains a TSSV, such as pSV 1.RecA and a
building
vector. in one such embodiment the building vector is pBV.IRES.LacZ.PA. In
another such
embodiment the building vector is pBV.EGFPl. In yet another such embodiment
the
building vector is pBV.IRES.EGFPl. In still another such embodiment the
building vector is
pBV.pGK.Neo.PA.
In a preferred embodiment two or more building vectors are included in the
kit. In a more
preferred embodiment all four of the above-listed building vectors are
included in the kit.
Restriction maps for one or more of the building vectors or the TSSV may also
be included
in the kits. In addition, the kits may also include a protocol for using the
contents of the kit
to perform homologous recombination. In one specific embodiment, a kit
contains
pSVl.RecA and one or more of the above-listed vectors also contains fusaric
acid and/or
chloro-tetracycline.
Accordingly, it is a principal object of the present invention to provide a
method for readily
and specifically modifying an independent origin based cloning vector in a
recombination
deficient host cell.
It is a further object of the present invention to provide a method of
transiently expressing a
RecA-like protein in a RecA~ host cell to allow the specific modification of a
gene of interest
contained by an independent origin based cloning vector.
It is a further object of the present invention to provide a method of
generating deletions,
substitutions, and/or point mutations in a specific gene contained by the
independent origin
based cloning vector in a RecA- cell
CA 02294619 1999-12-21
WO 98/59060 PCTNS98/12966
12
It is a further object of the present invention to provide a conditional
replication shuttle
vector which encodes a RecA-like protein, and which further contains a
specific nucleic acid
in a recombination cassette that selectively undergoes homologous
recombination with an
independent origin based cloning vector when both vectors are present in a
recombination
S deficient host cell.
It is a further object of the present invention to provide a temperature
dependent shuttle
vector which encodes a RecA-like protein.
It is a further object of the present invention to provide a temperature
dependent shuttle
vector which encodes a RecA-like protein, which further contains a specific
nucleic acid in a
recombination cassette that can selectively undergo homologous recombination
with a gene
of interest contained by an independent origin based cloning vector, when both
vectors are
placed in a recombination deficient host cell.
It is a further object of the present invention to provide a temperature
sensitive shuttle vector
that is pSVl.RecA having the ATCC no. 97968.
It is a further object of the present invention to provide a modified
independent origin based
cloning vector that can be used for the pronuclear injection of a nucleic acid
contained by
IOBCV into an animal zygote.
It is a further object of the present invention to provide a modified
independent origin based
cloning vector that can be transfected into an embryonic stem cell.
It is a further object of the present invention to provide a method of
introducing a linearized
nucleic acid insert from a modified independent origin based cloning vector
into a fertilized
zygote of an animal.
It is a further object of the present invention to provide a method of
introducing a modified
independent origin based cloning vector into an embryonic stem cell.
It is a further object of the present invention to provide a method of
purifying a large
linearized BBPAC.
CA 02294619 1999-12-21
WO 98/59060 PCT/US98/12966
13
These and other aspects of the present invention will be better appreciated by
reference to the
following drawings and Detailed Description.
BRIEF DESCRIPTION OF THE DRAWINGS
Fig~.:.~e 1 shows the strategy for targeted BAC modification. (I) Two cloning
steps are
involved in constructing the shuttle vector. The recombination cassette
(genomic fragments
A and B; and IRES-LacZP-Poly A marker gene) is first constructed in the
building vector and
then subcioned into the temperature sensitive pSVl.RecA shuttle vector. (II)
Co-integrate
formation: Co-integrates can be formed through homologous recombination at
either the
homology A or the homology B site, with only the former case illustrated.
(III) Resolution:
Resolved BACs are selected by growth on plates containing fusaric acid and
chloramphenicol. Correctly resolved clones are identified by colony
hybridizations with an
insert specific probe (e.g., a PGK polyA probe).
Figure 2 shows a schematic representation of targeted modifications of the BAC
169, which
contains the murine zinc forger gene, RU49. BAC169 containing RU49 was
obtained from
screening of the mouse 129 strain BAC genomic DNA library (Research Genetics).
Figure
2A depicts a restriction map of the BAC I 69. The position of several exons
are shown. The
region of homology A1 (Ikb PCR fragment) and homology B1 (l.6kb Xba-Hind
fragment)
are indicated. Abbreviations: XhoI (Xh), EcoRI (R), HindIII (H), XbaI (X),
NotI (Not) and
PmeI (Pore). Figure 2B depicts a map of the modified BAC 169 with IRES LacZ
PolyA
insertion (BAC169. ILPA). An extra PmeI site is inserted with the marker gene
(asterisk).
The size of the two Pme-Not fragments and the PmeI fragment are indicated.
Since the
marker gene (4kb) is less than the deleted genomic region (7kb), the total
size of the
modified BAC ( 128kb) is smaller than the original BAC ( I 31 kb).
Figure 3 shows Southern blot analyses of BAC co-integrates and resolved BACs.
Figure 3A
shows a schematic representation of expected Southern blot fragments in
BAC169, in co-
integrates through homology B1, and in correctly resolved BACs. When analyzing
recombination through homology B I, an EcoRI digest is used and homology B 1
is used as
the probe; when analyzing the recombination through homology A1, a HindIII
digest is used
and the homology A1 is used as probe. Figure 3B shows homology B1 co-
integrates. The
EcoRI digest of BAC clones and controls are probed with homology B 1. 1-4
represent four
CA 02294619 1999-12-21
WO 98/59060 PCT/US98/12966
14
clones. BAC 169 and pSV 1 with the recombination cassette were used as
controls. Figure
3C shows the analyses of the 5' ends of the resolved BACs. Resolved BAC clones
( 1-8)
were digested with HindIII and probed with homology A1. The controls are
homology B 1
co-integrates (CI), BAC 169 and the shuttle vector with recombination
cassettes. Figure 3D
shows the analyses of the 3' ends of the resolved BACs. The same procedure is
used as
described above except the resolved BAC clones were digested with EcoRI and
probed with
homology B 1.
Figure 4 shows pulsed field gel electrophoresis analyses of modified 169 with
the ILPA
insc: Lion. DNA for two independent clones of BAC 169. ILPA (L 1 and L2) and
BAC 169
were prepared by alkaline lysis, and then digested with NotI, PmeI and XhoI
(in a standard
buffer supplemented with 2.5 mM spermidine). The digested DNA were separated
by pulsed
field gel electrophoresis (Bio-Rad's CHEF-DRII, 5 to lSs, 15 hours at
14°C) and blotted on
to nitrocellulose filter (Stratagene). The same filter was probed separately
with three probes.
L1 and L2 are lacZl and LacZ2 which are independent clones which correspond to
clones 1
and 2 respectively in Figures 3C and 3D. Figure 4A shows the use of the BAC169
probe
which revealed all the restriction fragments. Figure 4B shows the use of the
pgkpoly A probe
which only hybridized to the ILPA insert fragment. Figure 4C shows the use of
the A2 probe
which hybridized to a fragment outside the region of modification. The
position of the
markers are indicated.
Figure 5 shows the production of BAC transgenic mice. Figure SA depicts
purified
linearized BAC L1 128 kb Not I insert for pronuclear injection. The pulsed
field gel is
probed with pgkpolyA probe. The numbers represent different fractions. The
smear below
the intact fragment represent degradation and undigested DNA. Figure SB shows
Southern
blot analyses of the founder transgenic mice with the lacZ probe. The tail DNA
were
digested with Bam HI and Southern blot analysis was performed. The negative
control
consisted of littermates of Y3, Y7 and Y9 mice. The positive control was a
conventional
transgenic mouse with the IacZ transgene. Figures SC and SD show the results
of using PCR
to determine the presence of BAC ends in the transgenic mice. The DNA at each
end
corresponding to the vector sequence is amplified and probed with a third
oligonucleotide in
the middle of the fragment. The appropriate size fragment is indicated. The
negative
controls are littermates. The positive control was BAC169 DNA. Figure SE shows
the
germline transmission of the IacZ transgene in the Y7 mouse line. Tail DNA
from two litters
CA 02294619 1999-12-21
WO 98/59060 PCT/US98/12966
having eight mice each were prepared and digested with BamHI. Southern blot
analysis was
performed with the lacZ probe.
Figure 6 shows the expression of the lacZ transgene in the brain of the Y7 BAC
transgenic
line. P6 mice brain from Y7 transgenic mice (Figure 6A) and a wild type
control litter mate
5 (Figure 6B) were whole mount stained to reveal lacZ expression in the Y7
cerebellum. Thick
saggital sections (Smm) from Y7 transgenic mice were also stained for lacZ
expression.
Figure 6C shows the low magnification and Figure 6D shows the high
magnification of the
rectangle area indicated in Figure 6C. Expression in the cerebellum, the
detate gyros and the
lineage of the olfactory bulb are indicated (i.e. SVZ, RMS and the OB).
Abbreviation Ce,
10 cerebellum; SC, superior collicoli; IC, inferior colliculi; DG, dentate
gyros; VZ, ventricular
zone; SVZ, subventricular zone; LV, lateral ventricle; RMS, rostral migratory
tract; OB,
olfactory bulb; Co, cortex.
Figure 7 is a schematic diagram containing Figure 7A which depicts a
hypothetical map of a
gene of interest within a selected BAC; Figure 7B which depicts the first
targeted
15 modification to introduce the positive selection marker gene; and Figure 7C
which depicts
the second modification to delete the promoter of the gene and to generate the
short arm.
Figure 8 is the restriction map of pSV l.RecA. This temperature sensitive
shuttle vector is
based on the pMB096 vector originally constructed by M. O'Connor et al.
[Science,
244:1307-1312 (1989)].
Figure 9 is the restriction map of pBV.IRES.LacZ.PA. This vector was modified
from the
pWHlO vector originally constructed by Kim et al. [MCB, 12:3636-3643 (1992)].
Figure 10 is the restriction map of pBV.EGFPl. The plasmid is based on
pBluescript.KS(+).
EGFP 1 was from Clonetech.
Figure 11 is the restriction map of pBV.IRES.EGFP 1. The plasmid is based on
the
pBluescript.KS back bone. EGFP 1 was from Clonetech.
Figure 12 is the restriction map of pBV.PGK.Neo.PA. The vector is based on a
pBS.KS
backbone. The pGK.Neo.PA sequences was excised from a pKS.NT vector by
digestion with
CA 02294619 1999-12-21
WO 98/59060 PCT/US98/12966
16
HindIII and BamHI and subcloned into the HindIII/Bam fragment of the
pBV.IRES.LacA.PA.
DETAILED DESCRIPTION OF THE INVENTION
The present invention provides a simple method for directly modifying an
independent origin
based cloning vector (IOBCV) in recombination deficient host cells including
generating
deletions, substitutions, and/or point mutations in a specific gene contained
in the
independent origin based cloning vector. Such modifications may be performed
with great
specificity. The modified independent origin based cloning vectors of the
present invention
can be used to introduce a modified heterologous gene into a host cell. One
specific use of
such a modified vector is for the production of a germline transmitted
independent origin
based cloning vector transgenic animal.
Targeted independent origin based cloning vector modification can be used for
functional
studies in diverse biological systems. The ability to efficiently modify a
independent origin
based cloning vector and generate an IOBCV-transgenic animal has important
applications
for functional analyses of genes in vivo. First, modified independent origin
based cloning
vectors can be used to study regulation of genes or gene complexes in
transgenic animals
such as mice. Since modified independent origin based cloning vectors can be
used to study
gene function in vivo, a deletion, substitution and point mutation within a
given gene can be
made in a independent origin based cloning vector, and the independent origin
based cloning
vector containing the modified gene can be reintroduced in vivo in its
endogenous expression
pattern. Furthermore, targeted independent origin based cloning vector
modification can be
used to create targeted expression of a selected gene, in the expression
pattern of another
gene, without prior knowledge of all of the regulatory elements of the
selected gene. An
important application of this type is targeted expression of the cre
recominase for tissue/cell
type specific gene targeting [Kuhn et al., Science 269:1427 (1995); Tsien et
al., Cell 87:1317
(1996)]. Finally, modified independent origin based cloning vectors can be
used to generate
large DNA constructs particularly for gene targeting in ES cells and in vivo.
In one specific embodiment of the present invention the independent origin
based cloning
vector is a Bacterial Artificial Chromosome (BAC) modified in a host E.coli
cell. A targeted
BAC modification system has several advantages over a conventional yeast based
CA 02294619 1999-12-21
WO 98/59060 PCT/US98/12966
17
modification system. First, a modified BAC automatically returns to the
recombination
deficient state after modification, ensuring stable maintenance of the
modified BAC in the
host strain. Second, BAC DNA can be very easily purified in relatively large
quantities and
high quality, allowing for use in biological experimentation including
pronuclear injection.
Third, since it is much easier to construct a BAC library than a YAC library,
there are many
more BAC libraries available from different species of animal, plants and
microbes [Woo et
al., Nucleic Acids Res., 22:4922 (1994); Wang et al., Genomics 24:527 (1994);
Wooster et
al., Nature 378:789 (1995)]. Most BACs also include all the necessary
regulatory elements
(i.e. LCRs and enhancers) to obtain dose dependent and integration site
independent
transgene expression [Dillon et al. Trends Genet. 9:134 (1993); Wilson et al.,
Annu. Rev.
Cell. Biol. 6:679 (1990); Bradley et al., Nature Genet. 14:121 (1997)].
Targeted BAC
modification can be applied successively to dissect these elements. In
addition, such a
modified BAC may be used to generate a transgenic animal. The BAC (or PAC)
stably
integrates into the animal cell genome. The transgenic animal can be used for
functional
studies, or for generating a desired gene product, such as producing a human
protein in the
milk of a transgenic mammal [Drohan et al. U.S. Patent No. 5,589,604, Issued
December 31,
1996]. Alternatively such modified BACs or PACs may be used for delivering a
specific
gene in gene therapy. In the Example below, a modified BAC has been
successfully inserted
into a murine subject animal, and in vivo heterologous gene expression has
been
demonstrated.
The methodology of the present invention is very general. Whereas the targeted
independent
origin based cloning vector modification is demonstrated on BACs, the system
is readily
applicable to BBPACs in general including PACs, P 1 and other vectors
propagated in the
recombination deficient E.coli. In addition, the BAC modification exemplified
herein, is also
apropo to Mammalian Artificial Chromosomes. For example, Harrington et al.
[Nature
Genetics, 15:345-355 (1997)] have used BAC derived DNA as a component of their
Human
Artificial Chromosome. Therefore, the use of such human artificial chromosomes
can
include the BAC modification taught by the present invention.
In accordance with the present invention there may be employed conventional
molecular
biology, microbiology, and recombinant DNA techniques within the skill of the
art. Such
techniques are explained fully in the literature. See, e.g., Sambrook, Fritsch
& Maniatis,
Molecular Cloning: A Laboratory Manual, Second Edition ( 1989) Cold Spring
Harbor
CA 02294619 1999-12-21
WO 98/59060 PCT/US98/12966
18
Laboratory Press, Cold Spring Harbor, New York (herein "Sambrook et al.,
1989"); DNA
Cloning: A Practical Approach, Volumes I and II (D.N. Glover ed. 1985);
Oligonucleotide
Synthesis (M.J. Gait ed. 1984); Nucleic Acid Hybridization [B.D. Hames & S.J.
Higgins eds.
(1985)]; Transcription And Translation [B.D. Hames & S.J. Higgins, eds.
(1984)]; Animal
Cell Culture [R.I. Freshney, ed. (1986)]; hnmobilized Cells And Enzymes [IRL
Press,
(1986)]; B. Perbal, A Practical Guide To Molecular Cloning (1984); F.M.
Ausubel et al.
(eds.), Current Protocols in Molecular Biology, John Wiley & Sons, Inc.
(1994).
As used herein an "IOBCV" is an independent origin based cloning vector. One
example of
such a cloning vector is a BBPAC defined below. An IOBCV generally comprises a
nucleic
acid insert which either is or contains a gene of interest.
A "vector" is a replicon, such as plasmid, phage or cosmid, to which another
DNA segment
may be attached so as to bring about the replication of the attached segment.
A "replicon" is
any genetic element (e.g., plasmid, chromosome, virus) that functions as an
autonomous unit
of DNA replication in vivo, i.e., capable of replication under its own
control.
As used herein, a "Bacterial or Bacteriophage-Derived Artificial Chromosome"
or "BBPAC"
denotes a vector that is derived from a bacterium or bacteriophage such as a
Bacterial
Artificial Chromosome (BAC) which is an E. coli F element based cloning
system, a P1-
Derived Artificial Chromosome (PAC) or a lambda-based cosmid. In one
embodiment, the
BBPAC encodes from 500 to 700 kilobases of genomic sequences. In another
embodiment,
the BBPAC encodes up to 500 kilobases of genomic sequences. In a preferred
embodiment,
the BBPAC encodes between 120 to 180 kilobases of genomic sequences. In one
particular
embodiment the BBPAC encodes 130 kilobases of genomic sequences. A BBPAC used
for
gene targeting can be referred to as a "BBPAC targeting construct" and
contains a nucleic
acid insert comprising the gene targeting construct.
A "gene targeting construct" as used herein is used interchangeably with
"targeting
construct" and is a nucleic acid that when introduced into a cell undergoes
homologous
recombination with the endogenous chromosomal DNA of the cell. The nucleic
acid is
introduced into the cell to induce a modification of a particular gene
contained on the
endogenous chromosomal DNA, including in particular cases, to disrupt that
gene to create a
knockout animal.
CA 02294619 1999-12-21
WO 98/59060 PCTNS98/12966
19
As used herein a recombinant deficient host cell is "RecA-" when the host cell
is unable to
express a RecA-like protein, including recA itself, which can support
homologous
recombination. In the simplest case, the gene encoding the RecA-like protein
has been
deleted in a RecA- host cell. Alternatively the RecA-host cell contains a
mutation in the recA
gene that impairs its function.
A "RecA-like protein" is defined herein to have the meaning generally accepted
in the art
except as used herein the recA protein itself is included as being a specific
RecA-like protein.
RecA-like proteins are proteins involved in homologous recombination and are
homologs to
recA [Clark et al., Critical Reviews in Microbiology 20:125-142 (1994)]. The
recA protein is
the central enzyme in prokaryotic homologous recombination. It catalyzes
pairing and strand
exchange between homologous DNA molecules, and functions in both DNA repair
and
genetic recombination [McKee et al., Chromosoma 7:479-488 (1996)]. A number of
RecA-
like proteins have been found in eukaryotic organisms and yeast [Reiss et
al.,Proc.Natl.Acad.Sci. 93:3094-3098 (1996)] . Two RecA-like proteins in yeast
are Rad51
and l~mc I [McKee et al. ( 1996) supra]. RadS 1 is a highly conserved RecA-
like protein in
eukaryotes [Peakman et al., Proc.Natl.Acad.Sci. 93:10222-10227 (1996)].
As used herein a "gene of interest" is a gene contained by a host cell genome
or more
preferably an independent origin based cloning vector that has been selected
to undergo
homologous recombination with a specific nucleic acid contained in a
recombination
cassette. A gene of interest can be either specifically placed into the host
cell or independent
origin based cloning vector for this purpose, or already contained by the host
cell or
independent origin based cloning vector .
As used herein a "marker" is an indicator, whose presence or absence can be
used to
distinguish the presence or absence of a particular nucleic acid and
preferably the
corresponding presence or absence of a larger DNA which contains and/or is
linked to the
specific nucleic acid. In a preferred embodiment the marker is a protein or a
gene encoding
the protein, and thus can be more specifically termed a "marker protein" or a
"marker gene".
The term "marker" (and thus marker protein or marker gene) is meant to be used
extremely
broadly and includes fluorescent proteins such as green fluorescent protein,
enzymes such as
luciferase, and further includes drug resistant proteins, whose presence or
absence may not
solely be regarded as a means to detect cells that contain the drug resistance
protein; and/or
CA 02294619 1999-12-21
WO 98/59060 PCT/US98/12966
he genes that encode such proteins. However, drug resistance proteins and/or
their
corresponding genes can allow the preferential growth of cells that contain
the drug resistant
gene (or alternatively allow the counter-selection of cells that do not
contain the drug
resistant gene) and therefore bestow a type of selectable distinction which is
meant to fall
5 within the present definition of a marker.
The term "a gene which encodes a marker protein" is used herein
interchangeably with the
term "marker protein gene" and denotes a nucleic acid which encodes a marker
protein.
A "cassette" refers to a segment of DNA that can be inserted into a vector at
specific
restriction sites. The segment of DNA encodes a polypeptide of interest, and
the cassette and
10 restriction sites are designed to ensure insertion of the cassette in the
proper reading frame
for transcription and translation. The present invention provides a
recombination cassette
that includes two homology fragments interrupted by an insertion, deletion or
mutation
sequence.
"Heterologous" DNA refers to DNA not naturally located in the cell, or in a
chromosomal
15 site of the cell. Preferably, the heterologous DNA includes a gene foreign
to the cell.
A ''nucleic acid molecule" refers to the phosphate ester polymeric form of
ribonucleosides
(adenosine, guanosine, uridine or cytidine; "RNA molecules") or
deoxyribonucleosides
(deoxyadenosine, deoxyguanosine, deoxythymidine, or deoxycytidine; "DNA
molecules"), or
any phosphoester analogues thereof, such as phosphorothioates and thioesters,
in either
20 single stranded form, or a double-stranded helix. Double stranded DNA-DNA,
DNA-RNA
and RNA-RNA helices are possible. The term nucleic acid molecule, and in
particular DNA
or RNA molecule, refers only to the primary and secondary structure of the
molecule, and
does not limit it to any particular tertiary forms. Thus, this term includes
double-stranded
DNA found, inter alia, in linear or circular DNA molecules (e.g., restriction
fragments),
plasmids, and chromosomes. In discussing the structure of particular double-
stranded DNA
molecules, sequences may be described herein according to the normal
convention of giving
only the sequence in the 5' to 3' direction along the nontranscribed strand of
DNA (i.e., the
strand having a sequence homologous to the mRNA). A "recombinant DNA molecule"
is a
DNA molecule that has undergone a molecular biological manipulation.
CA 02294619 1999-12-21
WO 98/59060 PCT/US98/12966
21
"Homologous recombination" refers to the insertion of a modified or foreign
DNA sequence
contained by a first vector into another DNA sequence contained in second
vector, or a
chromosome of a cell. The first vector targets a specific chromosomal site for
homologous
recombination. For specific homologous recombination, the first vector will
contain
sufficiently long regions of homology to sequences of the second vector or
chromosome to
allow complementary binding and incorporation of DNA from the first vector
into the DNA
of the second vector, or the chromosome. Longer regions of homology, and
greater degrees
of sequence similarity, may increase the efficiency of homologous
recombination.
A DNA "coding sequence" is a double-stranded DNA sequence which is transcribed
and
translated into a polypeptide in a cell in vitro or in vivo when placed under
the control of
appropriate regulatory sequences. The boundaries of the coding sequence are
determined by
a start codon at the 5' (amino) terminus and a translation stop codon at the
3' (carboxyl)
terminus. A coding sequence can include, but is not limited to, prokaryotic
sequences,
cDNA from eukaryotic mRNA, genomic DNA sequences from eukaryotic (e.g.,
mammalian)
DNA, and even synthetic DNA sequences. If the coding sequence is intended for
expression
in a eukaryotic cell, a polyadenylation signal and transcription termination
sequence will
usually be located 3' to the coding sequence.
Transcriptional and translational control sequences are DNA regulatory
sequences, such as
promoters, enhancers, terminators, and the like, that provide for the
expression of a coding
sequence in a host cell. In eukaryotic cells, polyadenylation signals are
control sequences.
A "promoter sequence" is a DNA regulatory region capable of binding RNA
polymerase in a
cell and initiating transcription of a downstream (3' direction) coding
sequence. For
purposes of defining the present invention, the promoter sequence is bounded
at its 3'
terminus by the transcription initiation site and extends upstream (5'
direction) to include the
minimum number of bases or elements necessary to initiate transcription at
levels detectable
above background. Within the promoter sequence will be found a transcription
initiation site
(conveniently defined for example, by mapping with nuclease S 1 ), as well as
protein binding
domains (consensus sequences) responsible for the binding of RNA poiymerase.
A coding sequence is "under the control" of transcriptional and translational
control
sequences in a cell when RNA polymerase transcribes the coding sequence into
mRNA,
CA 02294619 1999-12-21
WO 98/59060 PCT/US98/12966
22
which is then trans-RNA spliced and translated into the protein encoded by the
coding
sequence.
A "signal sequence" is included at the beginning of the coding sequence of a
protein to be
expressed on the surface of a cell. This sequence encodes a signal peptide, N-
terminal to the
mature polypeptide, that directs the host cell to translocate the polypeptide.
The term
"translocation signal sequence" is used herein to refer to this sort of signal
sequence.
Translocation signal sequences can be found associated with a variety of
proteins native to
eukaryotes and prokaryotes, and are often functional in both types of
organisms.
A particular nucleotide sequence comprising a gene of interest, whether
genomic DNA or
cDNA, can be isolated from any source, particularly from a human cDNA or
genomic
library. In view and in conjunction with the present teachings, methods well
known in the
art, as described above can be used for obtaining such genes from any source
(see, e.g.,
Sambrook et al., 1989, supra).
Accordingly, any animal cell potentially can serve as the nucleic acid source
for the
molecular cloning of any selected gene. The DNA may be obtained by standard
procedures
known in the art from cloned DNA (e.g., a DNA "library"), and preferably is
obtained from a
cDNA library prepared from tissues with high level expression of the protein
by chemical
synthesis, by cDNA cloning, or by the cloning of genomic DNA, or fragments
thereof,
purified from the desired cell (See, for example, Sambrook et al., 1989,
supra; Glover, D.M.
(ed.), 1985, DNA Cloning: A Practical Approach, MRL Press, Ltd., Oxford, U.K.
Vol. I, II).
Clones derived from genomic DNA may contain regulatory and intron DNA regions
in
addition to coding regions; clones derived from cDNA will not contain intron
sequences.
The present invention provides methods for selectively performing homologous
recombination in a cell that normally cannot independently support homologous
recombination. A specific nucleic acid is inserted into a recombination
cassette that
selectively integrates into a particular nucleotide sequence when the
recombination deficient
cell is transiently induced to support homologous recombination. More
particularly, the
present invention allows the integration of a specific nucleic acid into a
particular nucleotide
sequence of a gene of interest. The methods provided by the present invention
minimize the
CA 02294619 1999-12-21
WO 98/59060 PCTlTJS98/12966
23
nonspecific nucleotide sequence rearrangements and deletions, which are
characteristic of
other systems which involve host cells that normally support homologous
recombination.
In one case the specific nucleic acid can encode an entirely different protein
than the gene of
interest, and the gene of interest may be selected for the tissue specificity
of its promoter, for
example for use in generating a transgenic animal, or in a gene therapy
protocol. In one such
embodiment the rat preproenkephalin gene may be used as the gene of interest
since the
preproenkephalin promoter has been shown to confer brain expression and
synaptic
regulation in transgenic mice. [Donovan et al., Proc.Natl.Acad.Sci. 89:2345-
2349 (1992)).
In the Example below, the murine zinc finger gene, RU49 was used as the gene
of interest.
Alternatively, the specific nucleic acid can be constructed so as to cause a
deliberate and
specific modification in the sequence of the gene of interest, for example for
inducing a
change in the amino acid sequence of the gene product, such as is typically
done in
site-directed mutagenesis protocols.
In one aspect of the present invention, the recombination deficient host cell
cannot
independently support homologous recombination because the host cell is RecA-.
However,
as any person skilled in the art would readily understand, alternative causes
for
recombination deficiency may be rectified by methods that are analogous to
those taught by
the present invention and/or readily apparent in view of such teachings. For
example
recombination deficiency may be due to a deficiency of an alternative
recombination protein
such as another Rec protein including recB, recC, recD, and recE [Clark et
al., Critical
Reviews in Microbiol. 20:125-142 (1994)] which may be manipulated in a manner
that is
analogous to that taught herein for RecA-like proteins.
In the case of a RecA- host cell, inducing the host cell to transiently
support homologous
recombination comprises inducing the transient expression of a RecA-like
protein in the host
cell. Such induction may be performed by expressing a RecA-like protein
contained by the
recombination deficient host that is under the control of an inducible
promoter.
In a preferred aspect of the invention inducing the transient expression of
the RecA-like
protein is performed with a conditional replication shuttle vector that
encodes the RecA-like
protein. Conditional replication shuttle vectors can also include pBR322 in a
polyA
temperature-sensitive bacterial strain. Preferably the conditional replication
shuttle vector is
CA 02294619 1999-12-21
WO 98/59060 PCT/US98/12966
24
a temperature sensitive shuttle vector (TSSV) that replicates at a permissive
temperature, but
does not replicate at a non-permissive temperature.
Inducing the transient expression of the RecA-like protein consists of
transforming the host
cell with the TSSV at a permissive temperature, and growing the host cell at a
non-
permissive temperature. The TSSV encodes a RecA-like protein that is expressed
in the host
cell and supports the homologous recombination between a specific nucleic acid
contained in
a recombination cassette and the particular nucleotide sequence contained in
the host cell.
The TSSV encoding the RecA-like protein is diluted out when the host cell is
grown at the
non-permissive temperature.
In a more intricate version of the present invention, the particular
nucleotide sequence which
has been selected to undergo homologous recombination is contained by an
independent
origin based cloning vector (IOBCV) that is comprised by the host cell, and
neither the
independent origin based cloning vector alone, nor the independent origin
based cloning
vector in combination with the host cell, can independently support homologous
recombination. In a particular embodiment of this type both the independent
origin based
cloning vector and the host cell are RecA-, and inducing the host cell to
transiently support
homologous recombination comprises inducing the transient expression of the
RecA-like
protein to support homologous recombination in the host cell. The independent
origin based
cloning vector can be a BBPAC, such as the BAC exemplified below and the host
cell can be
a host bacterium, such as E. coli.
The independent origin based cloning vectors for use in the methods of the
present invention
can be obtained from a number of sources. For example, E. coli-based
artificial
chromosomes for human libraries have been described [Shizuya et al., Proc.
Natl. Acad. Sci.
89:8794-8797 (1992); Ioannou et al., In Current Protocols in Human Genetics
(ed. Dracopoli
et al.) 5.15.1-5.15.24 John Wiley & Sons, New York (1996); Kim et al.,
Genomics 34:213-
218 (1996)]. Libraries of PACs and BACs have been constructed [reviewed in
Monaco et al.,
Trends Biotechol., 12:280-286 (1994)], that are readily isolated from the host
genomic
background for example by classical alkaline lysis plasmid preparation
protocols [Bimboim
et al., Nucleic Acids Res. 7:1513-1523 (1979)], or alternatively, with the use
of a
nucleobond kit, a boiling Prep, or by cessium gradient (Maniatis, supra). BAC,
PAC, and P 1
libraries are also available for a variety of species (e.g. Research Genetics,
Inc., Genome
CA 02294619 1999-12-21
WO 98/59060 PCT/US98/12966
Research, Inc., Texas A&M has a BAC center to make a BAC library for livestock
and
important crops). Also BACs can be used as a component of mammalian artificial
chromosomes.
An independent origin based cloning vector that is a BAC can be isolated using
a cDNA or
5 genomic DNA probe to screen a BAC genomic DNA library, for example. The use
of a
mouse genomic BAC library from Research Genetics is exemplified below. A
positive BAC
can generally be obtained in a few days. To insert a gene of interest into a
selected locus in
the BAC, the region of insertion can be mapped for restriction enzyme sites.
Whereas
subcloning is necessary for detailed mapping, it is generally unnecessary
since rough
10 mapping is usually sufficient. As is readily apparent, other independent
origin based cloning
vector genomic libraries can be screened and the isolated independent origin
based cloning
vectors manipulated in an analogous fashion.
The conditional replication shuttle vectors of the present invention are
constructed so as to
contain a recombination cassette that can selectively integrate into the
nucleotide sequence of
15 the gene of interest encoded by the independent origin based cloning
vector. Such
conditional replication shuttle vectors can be constructed by inserting a PCR
amplified RecA-
like gene into an appropriate conditional replication shuttle vector which
either contains a
specific drug resistant gene or can be subsequently modified to contain one.
In a preferred
embodiment the drug resistant gene can also be counter-selected against, such
as with,
20 tetracycline and fusaric acid. Alternatively, in addition to the drug
resistant gene the
conditional shuttle vector can also contain a counter-selection gene such as a
gene that
confers sensitivity to galactose, for example.
In the Example below, the E.coli Kl2 recA gene (l.3kb) is inserted into the
BamHI site of a
pMB096 vector. In this case the vector already carried a gene that bestows
tetracycline
25 resistance, and in addition contains a pSC 101 temperature sensitive origin
of replication,
which allows the plasmid to replicate at 30 degrees but not at 43 degrees.
The RecA-like protein of a conditional replication shuttle vector can be
controlled by either
an inducible promoter or a constitutive promoter. In one particular embodiment
the transient
expression of the RecA-like protein is achieved by the transient induction of
the inducible
CA 02294619 1999-12-21
WO 98/59060 PCT/US98/12966
26
promoter in a host cell. In another embodiment, the constitutive promoter is
the endogenous
E. toll recA promoter.
The conditional replication shuttle vector should also contain at least one
unique cloning site.
When a building vector is used for the construction of the recombination
cassette as
described below, one unique site is reserved for transferring the
recombination cassette
containing the specific nucleic acid from the building vector to the
conditional replication
shuttle vector. For example a polylinker can be inserted between two specific
restriction
sites to create additional restriction sites that allow cloning of the
recombination cassette into
the conditional replication shuttle vector. In any case the conditional
replication shuttle
vector created should minimally contain a recombination cassette comprising
the specific
nucleic acid, (e.g., containing a point mutation, deletion or a marker gene)
flanked at both the
5' and 3' ends by genomic fragments containing about 350 basepairs (e.g. 250
basepairs to
600 basepairs though less may be sufficient) or more of the gene of interest
of the
independent origin based cloning vector.
1 S In certain cases a building vector is used to construct the recombination
cassette. Two small
genomic fragments, each containing about 350 basepairs (e.g. 250 basepairs to
600 basepairs
though less may be sufficient) or more of the gene of interest are cloned into
the building
vector (e.g., pBVl) in appropriate order and orientation to generate the
flanking regions of
the recombination cassette. DNA containing a promoter sequence 5' to the
specific nucleic
acid, which in turn is 5' to a polyadenine addition signal sequence, is
inserted between the
two genomic fragments in the proper orientation. The recombination cassette is
then
transferred into the conditional replication shuttle vector (e.g., pSV
l.RecA). The
recombination cassette, the RecA-like protein gene, and the drug resistant
gene are linked
together on the conditional replication shuttle vector such that when the
specific nucleic acid
integrates into the particular nucleotide sequence, the RecA-like protein gene
and the drug
resistant gene remain linked together, and neither the RecA-like protein gene
nor the drug
resistant gene remain linked to the integrated specific nucleic acid. In a
preferred
embodiment the conditional replication shuttle vector is a TSSV and the TSSV
is pSVl.RecA
having the ATCC no. 97968.
According to the methods of the present invention the conditional replication
shuttle vector is
transformed into a RecA- host cell containing the independent origin based
cloning vector.
CA 02294619 1999-12-21
WO 98/59060 PCT/US98/12966
27
The independent origin based cloning vector can also contain a gene which
bestows
resistance to a host cell against a corresponding toxic agent/drug such as an
antibiotic or in a
specific embodiment, chloramphenicol. The cells are grown under the conditions
in which
the conditional replication shuttle vector can replicate (e.g., when the
conditional replication
shuttle vector is a TSSV which replicates at 30° but not at 43
°, the host cell is grown at
30°C) and the transformants can be selected via the specific drug
resistant gene (or first drug
resistant gene) carried by conditional replication shuttle vector, and the
second drug resistant
gene carried by the independent origin based cloning vector. Since the
conditional
replication shuttle vector also carries the RecA-like protein gene, homologous
recombination
can occur between the conditional replication shuttle vector and the
independent origin based
cloning vector to form co-integrates through the sequence homology at either
the 5' or the 3'
flanking regions of the recombination cassette. The co-integrates then can be
selected by
growing the cells on plates containing the first and second drugs at non-
permissive
conditions (e.g. for the TSSV above, at 43 °C) so that the non-
integrated, free conditional
replication shuttle vectors are lost. This results in the selection for host
cells carrying the
integrated conditional replication shuttle vectors, (which co-integrate either
into the
independent origin based cloning vector or into the host chromosome). Correct
independent
origin based cloning vector co-integrates can be identified by PCR or more
preferably with
Southern blot analyses.
The co-integrates can then be re-streaked onto plates containing the second
drug, (i.e., the
drug which the gene initially carried by the independent origin based cloning
vector protects
against) and grown under non-permissive conditions overnight. A fraction of
the co-
integrates undergo a second recombination event (defined as resolution),
through sequence
homology at either the 5' or the 3' flanking regions of the recombination
cassette. The
resolved independent origin based cloning vector automatically loses both the
first drug
resistant gene (i.e., the specific drug resistant gene contained by the
conditional replication
shuttle vector) and the RecA-like protein gene due to the linkage arrangement
of the
RecA-like protein gene, the drug resistant gene and the specific nucleic acid
on the
conditional replication shuttle vector, described above. In addition, the
excised conditional
replication shuttle vector cannot replicate under the non-permissive
conditions and is
therefore diluted out.
CA 02294619 1999-12-21
WO 98/59060 PCT/US98/12966
28
The resolved independent origin based cloning vectors can be further selected
for by growing
the host cells (e.g., at 37°C) on plates containing the second drug and
an agent that
counterselects against cells containing the gene resistant to the first drug
(e.g., a gene
conferring tetracycline resistance may be counter-selected against with
fusaric acid). The
resolved independent origin based cloning vector will be either the original
independent
origin based cloning vector or the precisely modified independent origin based
cloning
vector. One method to identify the correctly resolved BAC is to choose 5-10
colonies and
prepare a miniprep DNA. The DNA can then be analyzed using Southern blots to
detect the
correct targeting events. Alternatively, the desired clones can be identified
by colony
hybridization using a labeled probe for the specific nucleic acid contained by
the
recombination cassette. Such probes are well known in the art, and include
labeled
nucleotides probes that hybridize to the nucleic acid sequence. Alternatively,
a marker
nucleic acid can be included in the recombination cassette and constructed so
as to remain
with the specific nucleic acid upon integration into the independent origin
based cloning
vector.
The marker can be a marker gene or marker nucleic acid that encodes a marker
protein that
confers a specific drug resistance to the host cell, as exemplified above,
against drugs such as
antibiotics, e.g., ampicillin, chloramphenicol, and tetracycline, a protein
that confers a
particular physical characteristic to the cells, such as a green fluorescent
protein or a
modified green fluorescent protein as described in U.S. Patent 5,625,048,
Issued 4/29/97 and
WO 97/26333 Published 7/24/97 hereby incorporated by reference in their
entireties, or an
enzyme such as luciferase. Alternatively, it can be another marker protein
including e.g., ~i-
galactosidase.
The methods of homologous recombination of the present invention are
selective, and
nonspecific nucleotide sequence rearrangements either do not occur, or are
essentially
undetectable by one or more conventional methods of analysis. One such method
includes
pulsed field gel mapping of the modified independent origin based cloning
vector and the
unmodified independent origin based cloning vector to determine whether any
unexpected
deletions, or insertions or rearrangement were generated during the
modification procedure.
In one particular embodiment, the same filter can be probed separately with a
probe for the
whole independent origin based cloning vector, with a probe for the specific
nucleic acid, and
a probe for a region of the gene of interest that has not been modified. A
restriction enzyme
CA 02294619 1999-12-21
WO 98/59060 PCT/U898/12966
29
digestion can reveal a finger print of the modified independent origin based
cloning vectors
indicating whether the fragments are preserved. Such a restriction enzyme
digestion is
exemplified below. Restriction enzyme digestions can be repeated with one or
more
additional restriction enzymes selected with respect to the restriction site
map of the
independent origin based cloning vector.
In an alternative method, the modified independent origin based cloning vector
and the
unmodified independent origin based cloning vector can be assayed with both a
probe
specific for any region of the DNA contained by the recombination cassette
predicted to be
inserted into the independent origin based cloning vector (e.g., the promoter
sequence, the
specific nucleic acid, and a polyadenine addition signal sequence) and a probe
specific for a
region outside of the modification region (e.g., near the promoter region but
outside of the
modification region).
A modified independent origin based cloning vector of the present invention
can be purified
by gel filtration, e.g. a column filled with SEPHAROSE CL-4B yielded intact
linear BAC
1 S DNA. The column can be pre-equilibrated in an appropriate buffer, as
described in the
Example below. The purified DNA can be directly visualized with ultraviolet
light after
ethidium bromide staining, for example. Columns such as the SEPHAROSE CL-4B
column
also can efficiently separate degraded DNA from the pure linear DNA.
The present invention also provides methods of using the modified independent
origin based
cloning vectors of the present invention. Such modified independent origin
based cloning
vectors contain a nucleic acid that can be inserted into an animal to make a
transgenic
animal. The modified independent origin based cloning vectors of the present
invention can
be introduced into the desired host cells by methods known in the art, e.g.,
transfection,
electroporation, microinjection, transduction, cell fusion, DEAF dextran,
calcium phosphate
precipitation, lipofection (lysosome fusion), use of a gene gun, or a DNA
vector transporter
(see, e.g., Wu et al., 1992, J. Biol. Chem. 267:963-967; Wu and Wu, 1988, J.
Biol. Chem.
263:14621-14624; Hartmut et al., Canadian Patent Application No. 2,012,311,
filed March
15, 1990).
Constitutive expression of any selected gene, even if at low levels is
contemplated by the
present invention. Various therapeutic heterologous genes can be inserted into
an
CA 02294619 1999-12-21
WO 98/59060 PCT/US98/12966
independent origin based cloning vector of the invention such as but not
limited to adenosine
deaminase (ADA) to treat severe combined immunode6ciency (SCID); marker genes
or
lymphokine genes into tumor infiltrating (TIL) T cells [Kasis et al., Proc.
Natl. Acad. Sci.
U.S.A. 87:473 (1990); Culver et al., ibid. 88:3155 (1991)]; genes for clotting
factors such as
5 Factor VIII and Factor IX for treating hemophilia [Dwarki et al. Proc. Natl.
Acad. Sci. USA,
92:1023-1027 (19950); Thompson, Thromb. and Haemostatis, 66:119-122 (1991)];
and
various other well known therapeutic genes such as, but not limited to, (3-
globin, dystrophin,
insulin, erythropoietin, growth hormone, glucocerebrosidase, [3-glucuronidase,
a-antitrypsin,
phenylalanine hydroxylase, tyrosine hydroxylase, ornithine transcarbamylase,
10 apolipoproteins, and the like. In general, see U.S. Patent No. 5,399,346 to
Anderson et al.
One particular method comprises the pronuclear injection of the modified
independent origin
based cloning vector into a fertilized animal zygote. Such a method is
exemplified below
with the modified independent origin based cloning vector being a BAC which
has been
linearized, and the animal zygote being a mouse zygote. 2 pl of 0.6 pglml of
BAC DNA was
15 inj ected.
The presence of both ends of the modified independent origin based cloning
vector can be
assayed for in the transgenic animal to determine if the intact nucleic acid
insert of the
IOBCV has been integrated into the genome. Since both ends of the nucleic acid
insert
contain some vector sequence, PCR primers specific to the vector sequence can
be generated
20 and used to amplify the transgenic DNA. The amplified products can then be
probed with a
third labeled oligonucleotide probe within the amplified region.
The transgenic animals that are formed give rise to germline transmission
after appropriate
breeding (B6/CBA mice were used in the Example). The ratio of transgenic
animals to wild
type animals should follow Mendelian genetics.
25 The expression of the specific nucleic acid and/or the gene of interest
inserted into the
transgenic animal can be determined by a variety of methods well known in the
art which
depend on the nature of the insert. For example, enzymes can be appropriately
assayed for
activity, in the case of ~i-galactosidase, whole mount staining can be
performed, in situ
hybridization can be used to detect the corresponding mRNA, and specific
antibodies can be
30 used to identify the expression of a corresponding protein. In preferred
embodiments such
CA 02294619 1999-12-21
WO 98/59060 PCT/US98/12966
31
exrrPssion will be evident only in cells in which the endogenous gene of
interest is
expressed. In the Example, in which the gene of interest was the murine zinc
forger RU49,
and the specific nucleic acid inserted therein was the IacZ marker gene,
analyses of the
expression of the IacZ marker gene in the entire cerebellum of postnatal day 6
transgenic
mice closely resembled the corresponding endogenous RU49 expression pattern.
The present invention also provides the use of targeted BBPAC modification to
obtain a
high rate of gene targeting in vertebrates. The BBPAC contains a nucleic acid
insert
comprising the gene targeting construct. The circular BBPAC can be used, or
preferably
the linearized nucleic acid insert is used. In either case, the BBPAC or
linearized nucleic
acid insert can be purified by gel filtration as described herein.
In one aspect of the invention the gene targeting is performed in ES cells
using a BBPAC
gene targeting construct that is greater than 100kb. In a general sense, the
BBPAC gene
targeting construct is similar to the conventional positive selection gene
targeting construct
(Figure 7): it contains two regions of homology, a long arm ( > 80kb) and a
short arrn (10-
20kb), with the neo cassette (pgk-neo polyA) introduced into the middle of the
BBPAC.
Two targeted BBPAC modifications are used to make this construct. The first
modification
is to introduce the neo gene to disrupt the gene of interest in the BBPAC. A
second
modification is to create the short arm (10-20kb). The reason for the second
modification is
enable the use of an endogenous probe flanking the short arm (KO probe) to
detect a
polymorphism between the targeted allele and the wild type allele in screening
ES cell
clones (Fig 7; Gene Targeting, a practical approach, supra).
A preferred version of the BBPAC gene targeting methodology of the present
invention also
includes negative selection. The conventional negative selection cassettes,
such as the use
of the herpes thymidine kinase cassette or the diphtheria toxin gene cassette,
may not always
work with BBPAC constructs since BBPAC DNA tends to exist in transfected
mammalian
cells as episomal DNA for a long period of time [Baker et al., NAR 25:1950-
1956]. In one
example, the EGFPI cassette can be used as a negative screening cassette. In
this case, in
the second step of modification to generate the short arm, the CMV promoter
driven green
fluorescent protein (EGFP-1) and the polyA signal can be introduced. Unlike
other
negative selection cassettes, GFP is not toxic to the cells but serves as a
fluorescent marker
protein. When gene targeting occurs, the EGFP-1 cassette will be lost and the
cell will not
CA 02294619 1999-12-21
WO 98/59060 PCT/US98/12966
32
exhibit a green fluorescence under UV tight. On the other hand, when the BBPAC
integrates non-homologously, the EGFP-1 cassette also integrates, and the
cells will
therefore exhibit the green fluorescence under UV. For the definitive Southern
blot
analyses only those neo resistant cell lines which do not exhibit a green
fluorescence under
UV light are chosen.
The process of generating the targeted ES cells with a BBPAC targeting
construct is
essentially the same as with the conventional protocols (Gene Targeting, A
Practical
Approach, supra), except for the following steps. First the linearized intact
BBPAC nucleic
acid insert (for example) is purified using the gel filtration procedure
described herein.
Next, the transfection of ES cells with the Iinearized intact BBPAC nucliec
acid insert is
performed as described by Baker [NAR, 25:1950-1956 (1997)], using psoralen-
inactivated
adenovirus as carriers, for example.
The method enables transfection efficiency in mammalian cells with linear
BBPAC DNA to
be similar to the transfection efficiency of a conventional DNA construct. On
the other
hand, the BBPAC targeting construct can potentially provide 10-100 fold higher
targeting
frequency than the conventional targeting construct, thereby making gene
targeting in
mouse ES cells easier and cheaper, since only a few dozen colonies need to be
isolated and
screened to obtain the targeted clones.
The present invention further provides a method of performing gene targeting
in fertilized
vertebrate zygotes by the injection of a BBPAC targeting construct, or
preferably the
linearized intact BBPAC nucleic acid insert containing the targeting construct
to generate a
transgenic knock-out animal (TKO). A large targeting construct ( > 100kb) can
provide a
very high targeting rate (predicted by mathematical modeling described above)
and gene
targeting can be directly performed with a fertilized vertebrate zygote via
pronuclear
injection of the modified BBPAC targeting construct. TKO methodology has
previously
been attempted by Brinster et al. [PNAS, 86:7087-91 (1989)] with a small DNA
construct
(2.6-8.9kb) but those workers only obtained a relatively low targeting rate
(0.2 % ). The
large homology DNA in the BBPAC ( > 100kb) of the present invention increases
the
targeting rate to a favorable range of 2% to 10% .
CA 02294619 1999-12-21
WO 98/59060 PCT/US98/12966
33
In one such embodiment, the design of the gene targeting construct is similar
to the ES cell
targeting construct except that instead of the neo gene, an IRES-GFP cassette
or an IRES-
lacZ cassette is fused to an exon of the gene of interest to disrupt the gene
(Figure 7). As
described above, two consecutive steps of BBPAC modifications are involved in
generating
the BBPAC containing the gene targeting construct.
The modified BBPAC TKO construct can be prepared in milligram quantities and
linearized
as described above. The linearized DNA then is introduced into the fertilized
zygote by a
standard protocol, e.g., pronuclear injection (Hogan et al., (1986) supra).
The transgenic
animal is then identified by standard Southern blots. The gene targeting event
can be
further identified by digesting DNA of the transgenic animal with appropriate
enzymes,
such as enzyme X, (Figure 7) and probed with the flanking KO probe (Figure 7).
Mice
with the targeting event will have an additional band of the appropriate size.
Such gene
targeting events can further be confirmed by expression of the GFP or LacZ
marker gene in
the expression pattern of the targeted endogenous gene, since the construct is
designed to
trap the endogenous promoter.
The TKO method has important ramifications in the field of vertebrate
genetics. It enables
gene targeting in many organisms that do not have ES cells, such as zebra
fish, rats and
other mammals. This will help to generate better animal models for human
diseases (e.g. ,
rats and monkeys), or to create genetically targeted animals suitable for
organ transplants
(such as pigs or baboons) or for commercial reasons (e.g., leaner pork or
beef). This
method also has additional advantages, even for gene targeting in mice. For
example, this
method will automatically provide germline transmission, since transgenic
animals are
rarely chimeric. It also enables targeted mice in strains other than the 129
strain to be
obtained, and avoids the expensive and time-consuming out-breeding protocols.
In still another aspect of the present invention, methods of performing gene
targeting in
somatic cells using BBPAC targeting constructs are provided. Since gene
targeting in
somatic cells is also dependent on the length of homology, using large DNA
targeting
construct also improves the targeting rate in somatic cells. The experimental
design in this
case is similar to that with the ES cells described above. Somatic cell gene
targeting is
useful in gene therapy, for example, in-a targeted insertion of a functional
gene in a
CA 02294619 1999-12-21
WO 98/59060 PCT/US98/12966
34
hereditary disease of the hematopoietic system. Such methods are also useful
to generate
targeted cell lines for experimental purposes.
Conditional replication shuttle vectors that encode a RecA-like protein are
also provided by
the present invention. The RecA-like protein can be controlled by either an
inducible
promoter or a constitutive promoter. The conditional replication shuttle
vector is preferably
a temperature sensitive shuttle vector (TSSV). In one such embodiment the TSSV
contains
both a gene that confers tetracycline resistance and a RecA-like protein that
is recA. In a
preferred embodiment, the TSSV is pSVl.RecA having the ATCC no. 97968.
Independent origin based cloning vectors that contain a gene of interest that
has been
mcd:fied by the methods of the present invention are also included in the
present invention.
More particularly such independent origin based cloning vectors have undergone
homologous recombination with a conditional replication shuttle vector in a
RecA- host cell,
wherein the conditional replication shuttle vector encodes a RecA-like
protein. In a preferred
embodiment the independent origin based cloning vector has undergone
homologous
1 S recombination in a RecA- host cell with a temperature sensitive shuttle
vector encoding a
RecA-like protein. In a preferred embodiment the modified independent origin
based cloning
vector is a BAC that has undergone homologous recombination with the
temperature
sensitive shuttle vector pSVLRecA having the ATCC no. 97968.
The present invention may be better understood by reference to the following
non-limiting
Examples, which are provided as exemplary of the invention. The following
examples are
presented in order to more fully illustrate the preferred embodiments of the
invention. They
should in no way be construed, however, as limiting the broad scope of the
invention.
EXAMPLE
HOMOLOGOUS RECOMBINATION BASED MODIFICATION IN E. COLI AND
GERMLINE TRANSMISSION IN TRANSGENIC MICE OF AN 131 KILOBASE
BACTERIAL ARTIFICIAL CHROMOSOME
Introduction
Bacterial based artificial chromosomes, such as Bacterial artificial
chromosomes (BACs)
and P-lderived artificial chromosomes (PACs), are circular bacterial plasmids
that may
CA 02294619 1999-12-21
WO 98/59060 PCT/US98/12966
propogate as large as 300kb of exogenous genomic DNA {Shizuya et al, PNAS, 89,
8794-
97, 1992; Ioannou et al, Nature Genet., 6, 84-90, 1994). For the majority of
BAC and
PAC libraries, the average size of the insert is 130-150 kb. There are several
advantages of
using bacterial based artificial chromosomes for genomic and functional
studies, compared
5 to the yeast based system (i.e. YACs): First, BAC and PAC libraries are much
easier to
construct due to higher cloning efficiency. Second, BACs and PACs are
propagated in
recombination deficient E. coli host cells, so they have high stability and
minimal
chirnerism. No rearrangements have been observed in BACs or PACs after 100
generations of growth. Third, isolation of BAC and PAC DNA is very easy since
they
10 exist as supercoiled circular plasmids that are resistant to shearing.
Conventional bacterial
plasmid DNA isolation methods can be applied to obtain milligrams of intact
BAC or PAC
DNA. Finally, direct DNA sequencing can be applied to BAC or PAC DNA, which is
not
possible for YAC DNA. These advantages have made BACs and PACs important tools
for
genome studies in many species.
15 Although BBPACs are useful for physical mapping in genome studies, no
simple method is
available to modify BBPACs , as is available for the YACs. A simple homologous
recombination based BBPAC modification method is disclosed, termed targeted
BBPAC
modification (See Figure 7 for a schematic representation of the method}. This
method
allows precise modification, such as marker insertion, deletion, point
mutation, at any
20 chosen site within a given BBPAC. This method involves several steps:
isolation of
BBPACs using cDNA or genomic DNA probes, simple mapping and partial sequencing
of
the BBPACs, cloning of the shuttle vector, targeted modifications, pulsed
field gel analyses
of the modified BBPACs, and finally preparation of linearized BBPAC DNA for
functional
studies, such as pronuclear injection to produce BBPAC transgenic mice. Since
the method
25 is simple and reliable, it is reasonable to expect that the entire
procedure, from the step of
screening for a BBPAC with a cDNA or genomic DNA probe to the step of modified
BBPACs ready for functional studies, can be completed within 6-8 weeks.
Using this method, the IRES-LacZ marker gene has been introduced into an 131kb
bacterial
artificial chromosome (BAC) containing the murine zinc finger gene, RU49. No
30 rearrangements or deletions are detected in the modified BACs. Furthermore,
transgenic
mice are generated by pronuclear injection of the modified BAC and germline
transmission
of the intact BAC has been obtained. Proper expression of the IacZ transgene
in the
CA 02294619 1999-12-21
WO 98/59060 PCT/US98/12966
36
cerebellum has been observed, which could not be obtained with conventional
transgenic
constructs. In summary, a novel and efficient method has been developed to
modify BACs,
PACs and P 1 for in vivo studies of gene expression and gene function.
Materials and Methods
1. Isolation and initial mapping of BACs
(I) BAC isolation (3-4 days):
A BAC clone is isolated with either a unique cDNA or genomic DNA probe. BAC
libraries for various species, (in the form of high density BAC colony DNA
membrane) are
available from Research Genetics, Inc. and Genome Research, Inc. The mouse 129
genomic BAC library from Research Genetics has proved to be a good source for
genomic
DNAs. To avoid damage to the membrane, the probe is first tested on a mouse
genomic
Southern blot to ensure that the probe does not contain any repetitive
elements. The library
is screened according to manufacture's direction. The positive clones can be
obtained from
the company within a few days.
(II) Preparation of midiprep BAC DNA by alkaline lysis method (1 day):
Reagents: 1. Solution I: 50mM glucose, 25 mM Tris.HCl (pH 8.0); 10 mM EDTA
(pH 8.0)
10% SDS).
2. Solution II: 0.2N NaOH, 1 % SDS (0.4 g NaOH, 45 ml ddH20, 5m1
3. Solution III: 5M KOAc (60m1), glacial acetic acid (11.5m1), H20 (28.5
ml).
Protocol:
1). Inoculate each BAC containing bacterial to 50m1 LB containing 12.5 ug/ml
chloramphenicol. Grow overnight in 37°C.
2). Spin the overnight culture in a 50m1 Falcon tube for 20 min. at 3500 RPM
at 4
°C. Pour off the supernatant.
3). Resuspend the pellet in 1 ml cold solution I. Transfer the cell mix to a
15 ml
polybrene centrifugation tube and place on ice for 5 min.
4). Then add 2m1 fresh ( < 2 weeks old) solution II. Mix well by inverting
vigorously a few times.
CA 02294619 1999-12-21
WO 98/59060 PCT/U598/12966
37
5). Immediately add 1 ml cold solution III, mix by inverting gently several
times,
and place on ice for 10 min (this solution may be left overnight).
6). Spin at 10, OOOrpm for 12 min. at 4 °C. Transfer the supernatant to
a new
polybrene tube.
7). Add 4m1 Phenol (pH6.0)/Chloroform (1:1), and mix well by inverting the
tube
several times. Spin again at 10,000 rpm for 12 min. at 4 °C.
8). Transfer the upper layer to a new tube, and add 8m! 100% ethanol to it.
Invert
the tube vigorously several times to mix well. Spin at 10, 000 rpm for 30 min
at 4
°C. It can also be kept in -20 °C for overnight prior to
centrifuging.
9). Wash the pellet with 70% ethanol. Dry by vacuum and resuspend the DNA in
200 ul TE. The BAC midiprep DNA may be stored in 4 °C for months (Do
not
freeze the BAC DNA, since repetitive freezing and thawing will result in
degradations).
(III) BAC maxiprep DNA preparation:
Two methods were used to prepare large quantities of RNA-free BAC maxiprep
DNA.
The first method is the standard cesium chloride banding method (see Maniatis,
supra).
This method was used routinely to obtain > 500ug BAC DNA from 1 liter bacteria
culture.
The second method, uses a commercially available column, the Nucleobond AX-500
{made
by The Nest Group, Southborough, Mass.).The maxiprep DNA are also stored in 4
°C for
long-term storage.
(IV) Mapping the BACs by Pulsed Gel Electrophoresis and Southern blots (3-
Sdays):
To determine the size of each BAC and to confirm that the BAC contains the
gene of
interest, a simple mapping of the BACs is done. The following enzymes are used
to map
each BAC: Not I (to release the BAC insert), Mlu I, NotI/Mlu I (double
digest), PmeI,
PmeI/NotI and XhoI. Digestion is done in a 40u1 total volume, which contains
the
following: 5u1 midiprep DNA, 4u1 digestion buffer, 4u1 lOx BSA(if necessary),
lul 100mM
spermidine( final concentration 2.5mM), 2ul enzyme(10-40units), and ddH20.
Digestion is
done at 37 °C for > Shrs.
The digested BACs are resolved on a pulsed field gel (Bio-Rad's CHEF-DRII).
The gel is
1 % agarose in 0.5 x TBE. The gel is run in 0.5xTBE. The separation condition
is the
following: 6v/cm, 5s to 15s linear ramping for IShrs to l8hrs at 14 °C.
The New England
CA 02294619 1999-12-21
WO 98/59060 PCT/US98112966
38
Biolab's PFGE marker I or II as the high molecular weight marker and lkb DNA
ladder
(Life Technologies Inc.) as the low molecular weight marker are used.
The gel is then stained with ethidium bromide (1 to 5000, or 1 to 10,000
dilution of
lOmg/ml stock) for 30 min prior to taking the photograph. Then the gel is
blotted onto the
nitrocellulose membrane and hybridized to cDNA and genomic DNA probes
according to
standard protocols (Maniatis, supra). To ensure the entire cDNA is included in
the BAC,
probes/or oligonucleotides from both the 5'end and the 3' end of the gene are
used to probe
the blot separately. Those large BACs containing the entire gene are usually
selected for
BAC modification.
2. Construction of the shuttle vector with the recombination cassette
Since targeted BAC modification is a method based on homologous recombination,
homologous sequence from the BAC has to be obtained. Two homologous sequences
of
about SOObp each (namely A and B, Figure 7) is all that is needed to construct
the shuttle
vector for BAC modification. The homologous sequences are chosen such that a
given
modification (i.e. insertion, deletion and point mutation) will be introduced
between A and
B in the BAC. A and B can be obtained by direct sequencing of the BACs. The
sequencing oligonucleotides are designed based on the cDNA sequence.
(I) Direct sequencing of the BAC (2-3days):
1) If maxiprep DNA is used, go directly to step 2. If midiprep DNA is used,
first
add 100u1 ddH20 and 10u1 lOmg/ml RNAse A to 100u1 midiprep BAC DNA, and
incubate at 37 °C for > lhr. (This step is critical, incomplete RNAse
treatment will
result in poor precipitation and sequencing).
2) Add 132 ul PEG mix (2.SM NaCI and 20% PEG 8000) to the treated DNA. Put
on ice for Smin.
3) Spin for 15 min at 4 °C. Discard the supernatant. Spin again for 2
min.
Completely remove the remaining supernatant, which contains the PEG mix.
4) Wash the pellet with 70% ethanol. Dry in Speedvac and resuspend in 20u1
ddH20.
5) Run 2u1 on a agarose gel to estimate the final concentration. Usually use 6-
8 ul
(SOOng-1000ng) DNA for automatic sequencing, also use 150ng sequencing oligos.
CA 02294619 1999-12-21
WO 98/59060 PCT/US98/12966
39
Each sequencing reaction will result in up to a SOObp sequence. Sequence more
than one
BAC for a given primer to compare the sequences. The main purpose for
sequencing is to
design a 20 by PCR primer, which is about 500 by away from the sequencing
oligo (which
usually is the other PCR primer), to enable PCR amplification of this genomic
fragment and
to clone it into the building vector. Therefore, as long as a 20bp sequence
can be identified
which is at the appropriate position, and which is the same in several
independent
sequencing reactions, the goal is achieved. The quality of the DNA sequence in
between is
not very critical.
(II). Vectors used in targeted BAC modification:
A two vector system is designed to construct the shuttle vector for BAC
modification
(Figure 1). The first vector is a pBS.KS based building vector, which is used
to construct
the recombination cassette containing homologous sequence A and homologous
sequence B
and the modification to be introduced between them. The recombination cassette
was not
constructed in the pSVl.RecA shuttle vector was for the following reasons:
first, it is a low
copy plasmid so that it is difficult to obtain high quantity DNA; second, it
is a large piasmid
(llkb), so it is relatively difficult to clone. The building vector contains
the marker gene to
be introduced into the BAC, cloning sites flanking it (usually EcoRI for
cloning the
homology A and XbaI for homology B, and rare restriction sites such as MIuI,
PmeI and
Pac I for mapping of the modified BAC). There are two Sal I sites (or one Sal
I, one XhoI)
flanking the multiple cloning sites. They are used to release the
recombination cassette and
subclone it into the SaI I site of the pSVl.RecA vector, to complete the
shuttle vector
construction. One thing about designing the building vector is that there
should not be any
Not I sites within the recombination cassette, since NotI sites are used in
the end to release
the linear modified BAC for biological experiment (e.g., pronuclear
injection). The map
and utility of various building vectors and the shuttle vector are described
below.
(A) Building Vectors (pBV) All based on pBS.KS (Stratagene)
pBV.IRES.LacZ.PA (Fig. 9) This vector is designed to introduce lacZ marker
gene
into a coding exon or the 3' UTR of a given gene, to study gene expression and
gene regulation in vivo. IRES will enable the translation of the marker gene
independent of the endogenous translation initiation codon.
CA 02294619 1999-12-21
WO 98/59060 PCT/US98/12966
pBV.EGFPl (Fig. 10) This vector is designed to introduce the brighter version
of
the green fluorescent protein, EGFPI (Clontech), into an exon of a given gene
before the endogenous ATG or fused in frame with the endogenous ger_e. The
green fluorescent protein will mark gene expression in living cells and living
5 organisms. Since the marker gene does not contain its own polyA addition
sequence, the endogenous polyA sequence is used.
pBV.IRES.EGFPi (Fig. 11) This vector is used to introduce EGFP1 gene into the
coding region or the 3' UTR of a given gene, with its translation independent
of the
endogenous translation frame.
10 pBV.pGK.Neo.PA (Fig. 12) This vector is designed to introduce a neo
expression
cassette into the BAC, containing the neo gene with the pgk promoter and the
polyA addition signal. Modified BAC can be introduced into tissue culture cell
lines (i.e. ES cells) to obtain stable transfected cells by selecting for
neomycin
resistance. This vector is particularly useful for gene targeting with
modified
15 BACs. Notice that although there are two identical pgkpA sequence at the 3'
end
of the neo gene, it will not interfere with the proper expression of the neo
gene.
The only consequence is that during BAC modification, one of the pgkPA
sequence
may be deleted due to homologous recombination.
(B) Temperature sensitive, recombination inducing shuttle vector (pSVl.RecA)
(Fig. 8)
20 This plasmid vector was modified from the pMB096 vector originally
constructed
by O'Connor et al (Science, 1989, Vol 244, pp.1307-1312). The pMB096 vector
was a gift from Dr. Michael O'Connor. The original vector carries tetracycline
resistance, and contains a pSC101 temperature sensitive origin of replication,
which
allows the plasmid to replicate at 30°C but it will cease replication
and is lost at
25 43°C. The E. coli RecA gene was amplified by PCR and sub-cloned into
the Bam
HI site, to create the pSVl.RecA vector. The Sal L-site is used to subclone
the
recombination cassette from the building vector.
(III) Cloning two PCR amplified BAC fragments into the building vector ( 6-8
days ):
30 The first step of targeted BAC modification involves the subcloning of two
small genomic
fragments (A and B) into an appropriate buiiding vector, which includes two
steps of
CA 02294619 1999-12-21
WO 98/59060 PCT/US98/12966
41
conventional sub-cloning. One should pay attention to the following points
when designing
the A and B fragments.
1. Each fragment should be > 500 by (the shortest attempted was 450bp). PCR
amplified fragment with appropriate restriction sites designed at the end of
the PCR
primer is the method of choice. Frequently, an additional restriction site is
designed into one of the two PCR primers to assist in determining the
orientation of
the cloned PCR fragment. The relative imprecision of PCR amplification does
not
appear to affect the BAC modification efficiency.
2. As mentioned before, neither A nor B fragments should containing internal
XbaI, EcoRI and Sal I sites, since these sites will be used for subcloning.
Nor
should they contain NotI sites since NotI is used to linearize the BAC
3. The orientation of the arms must be preserved as in the endogenous loci.
(IV) Subcloning the recombination cassette from the building vector into the
pSVl.RecA
shuttle vector (4 days)
1. Prior to cloning the recombination cassette into the shuttle vector, the
following plates
are usually prepared: the tetracycline ( l0ug/ml) LB agar plates and the
tetracycline(l0ug/ml) + chloramphenicol (12.5 ug/ml) LB agar plates. Plates
are made
according to standard protocol [Sambrook et al. , ( 1989) supra].
2. Prepare pSVl.RecA and building vector midi-prep DNA by the alkaline lysis
method
see above). For the pSV 1.RecA vector, Qiagen columns can also be used to
obtain high
purity DNA, though yield is usually low. This is due to the low copy number of
the pSV 1
plasmid. For preparation of pSVl.RecA DNA, the culture should be grown at
30°C in LB
+ tetracycline ( l0ug/ml). The final midi-prep DNA is usually dissolved in 200
ul TE or
ddH20.
3. Digest 2-5 ug of the pSVl.RecA and pBV with Sal I . For pSVl.RecA, the
reaction is
done in 200 ul volume:
100 ld medi-prep DNA (2-5 ug) or
20 ~,1 of Qiagen midi-prep of pSVl.RecA
20 ~I H buffer (Boehringer Mannheim)
8 ul Rnase (10 mg/ml) (for alkaline lysis preps)
10 ul Sal I (200 units, Boehringer Mannheim)
62 ul ddH20
The reaction is performed at 37 °C for > 6 hours (usually overnight),
then 30 units more Sal
I is added, and the digestion continue for another 1-2 hours. (Optional) A
small sample of
CA 02294619 1999-12-21
WO 98/59060 PCT/US98/12966
42
the digestion (Sul) may be run on a gel to ascertain that a complete digestion
has been
achieved.
4. (Optional)At the end of the digestion, Sal I is inactivated by heating to
65 °C for 15
minutes.
5. The vector is then treated with alkaline phosphatase by adding 20u1 lOx
dephosporylaiton buffer, 4 ul (lunit/ul) calf intestinal alkaline phosphatase
(Boehringer
Mannheim) for 30 minutes at 37 °C. The enzyme is then inactivated by
adding 20 ul 50
mM EDTA (to a final concentration of SmM}, and heating at 75 °C for 15
minutes.
6.The digested pSV 1 vector and pBV with recombination cassette are run on a 1
% low
melting Seaplaque GTG agarose at 75 V for 8-10 hours. The DNA should be run in
a
large well created by taping together several teeth of the comb.
7. An l lkb linearized plasmid band should be visible on the gel for pSV
l.RecA. Cut this
band and also the recombination cassette insert band from the gel. Purity
these DNA
fragments using Geneclean Spin columns (Bio 101, Inc.) according to
manufacture's
1 S direction. Run a small portion of the purified DNA on a gel to estimate
the DNA
concentrarion.
8. Ligation reaction: Each ligation reaction is done in 20u1 total volume
containing: > SOng
pSV l .vector, 100-200ng insert, 2ul lOX ligation buffer (Boehringer-
Mannheim), 2ul IOmM
ATP, lul ligase (Boehringer-Mannheim) and ddH20. Ligation is carried out at 16
°C
overnight.
9. Transformation of DHSa competent cells with pSVl vectors: Half of the
ligation
reaction (10 ul) is used for transformation, by adding to 100 ul of cold,
chemical- induced
DHSa competent cells. Incubate 15 minutes on ice, then heat shock at 37
°C for 2 minutes,
add lml LB to the tube, and shake at 30°C for 30 minutes. The cells are
then centrifugated
at 6000 x g for 4 minutes and the pellet is resuspended in 100 ul LB and
spread onto Tet
( l0ug/ml) LB agar plates. Incubate the plates at 30 °C for > 15 hrs
hours.
11. Pick colonies and do colony hybridization according to standard protocols
[Sambrook
et al., (1989), Supra], probing with a fragment derived from the pBVI, such as
homology
arms (A or B)or the marker gene. Positive clones are further analyzed by
restriction digest,
and if necessary, Southern blots.
CA 02294619 1999-12-21
WO 98/59060 PCT/US98/12966
43
3. Targgted BAC HomoloEous Recombination in Bacteria
(I) Equipment
Bacterial incubator: set either at 30°C or at 43°C.
Shakers: set either at 30°C or at 43°C.
(II) Reagents and Plates
The following reagents and plates should be prepared prior to the targeted
modification
experiment. All the plates can be stored in 4 °C for up to one month.
Detailed methods for
preparation of various antibiotic resistant plates can be found in Maniatis.
1. Tetracycline stock solution (1000X): 10 mg/ml in 50% ethanol, wrapped in
aluminum foil and stored in -20°C for up to one month.
2. Chloramphenicol stock soiution (1000X): 12.5 mg/ml, dissolved in ethanol
( > 50% ), stored in -20 °C.
3. Tetracycline plates (tet plates): LB agar plates containing 10 ug/ml
tetracycline.
Store in 4 °C and wrapped in aluminum foil to avoid the light.
4. Chloramphenicol plates (Chl plates): LB plates contain 12.5 ug/ml
Chloramphenicol.
5. Tetracyline+ Chloramphenicol plates: LB plates contain l0ug/ml tetracycline
and 12.5 ug/ml chloramphenicol.
6. Fusaric acid + Chloramphenicol TB plates (FA+Chl plates): Prepared as
following.
First, make tryptone broth agar, or TB agar:
500m1 TBTB 1 L TB
Tap H20(not distilled H20) 500 ml 1L
Bacto tryptone Sg lOg
Yeast extract 0.5g lg
Glucose 0.5g lg
NaCI 4g 8g
O.1M ZnCl2 0.25 0.5
ml ml
Chlorotetracycline (6.3mg/ml) 4 ml 8 ml
Bacto agar 7.Sg 15g
CA 02294619 1999-12-21
WO 98/59060 PCT/US98/12966
44
Autoclaving the above TB. Also autoclave 500 ml of 1M NaH2P04.H20. After
autoclave, wait till the TB agar
drop to about 60C, then add the
following:
SOOmI TB 1L TB
NaHzP04. H20 ( 1 M) 36 ml 72 ml
Fusaric Acid (2mg/ml, filter ster.)3 ml 6 ml
Chloramphenicol (12.5 mg/ml ) 0.5 ml I ml
Pour the plates and leave the plates outside overnight and then store at 4
°C. There
is no need to avoid the light.
(lll)Making competent BAC containing bacteria (I day)
A chemical method is used to prepare competent cells from BAC containing
bacteria host
(Inoue et al, Gene 96, p23-28, 1990).
(1) Media and plates:
LB +Ampicilin (50 ug/ml) agar plates;
TB media (IOmM Pipes, SSmM MnCl2, lSmM CaCl2 and 250mM KCl), all
the components except for MnClz are mixed and the pH is adjusted to 6.7
with KOH. Then, MnCl2 was dissolved, the solution was sterilized by
filtration through a 0.45u filter unit and stored at 4 °C. All salts
were added
as solids.
(2) Frozen stock of BAC containing DH10B cells were taken by a metal loop and
inoculated into 3 ml of LB+chloramphenicol (l2.Sug/ml). Grow the culture with
rigorous
shaking in 37 °C for overnight.
(3) Take O.SmI overnight culture, add to SOmI LB+chloram. (l2.Sug/ml) and grow
at 37
°C with rigorous shaking till an optical density at 600 nm of about 0.6
is achieved.
(4) Place the flask on ice for 10 min. Then transfer to a SOrnI falcon tube
and centrifuge at
3000 rpm for 10 min at 4 °C.
(5) Pour the supernatant. Resuspend the pellet in 16m1 ice-cold TB. Incubate
on ice for
l0min, then spin again as above.
(6) The cell pellet was gently resuspend in 4 ml of TB supplemented with 7 %
DMSO.
Incubate on ice for lOmin, then dispense O.SmI aliquot and immediately frozen
by
immersion into liquid nitrogen. The tubes are stored in -80 °C for
further use.
CA 02294619 1999-12-21
WO 98/59060 PCT/US98/12966
(I~ Co-integrate formation and identification through Southern blot analyses
(4 days):
1. Transform the competent BAC cells with the Ts shuttle vector, using 10 ul
of the
midiprep DNA and 200u1 BAC containing competent cells. Transformation is done
as in
(IV) of part II. Plate 1/10 of the transformed cells onto Tet+Chl plates, and
grow
5 overnight at 30°C.
2. To generate co-integrates, single colonies (up to 6 in total) is picked up
with a sterilized
metal loop and diluted each into lml LB. Vortex to disperse the bacteria in
LB. Plate
100u1 LB+Bacteria on to two Tet+Chl plates. Incubate one at 43 °C
incubator, and
incubate the other at 30 °C overnight.
10 3. A thick lawn of bacteria will grow on the plates incubated in 30
°C. For the plates
incubated in 43 °C, only dozens of individual colonies will grow on top
of a hazy
background of very small satellite colonies. Pick 20 of these large colonies,
inoculate each
colony to 2 ml LB supplemented with tet (l0ug/ml) and chloramphenicol (12.5
ug/ml), and
streak the same colony onto a tet+chl plates. Grow the miniculture with
rigorous shaking
15 at 43 °C overnight. Incubate the master plate at 43 °C
incubator overnight and stored in 4
°C for further use.
4. Make miniprep DNA from a 1.5 ml miniculture using standard alkaline lysis
methods.
Dissolve the DNA in a 30,1 TE and use 5-10 ~,1 of the DNA for restriction
enzyme
analysis.
20 5. Restriction digest with appropriate enzymes and analysis of the co-
integrate by Southern
blot. Due to the high efficiency of co-integrate formation even with S00bp
homology
( > 10 % ), I usually only analyze co-integration on one homology side (either
A or B). For
example, to analyze co-integrate on A side, use fragment A as a probe and
digest the BAC
DNA with an enzyme that will detect the co-integrate formation on A side (such
as EcoRI).
25 Standard southern blots are done to reveal the co-integrates. As controls,
the original BAC
and the shuttle vector should be included in this analysis. The reason to use
the homology
arms as Southern blot probes is that it will hybridize to two bands of
appropriate size in the
co-integrate BAC. As controls, the original BAC and the shuttle vector should
be included
in this analysis.
CA 02294619 1999-12-21
WO 98/59060 PCT/US98/12966
46
(I~ Resolution and Southern blot analyses of correctly resolved BACs (6days):
1. Once the co-integrates are identified, a purified single colony of the co-
integrate from
the Tet+Chl plates grown at 43 °C is picked and streaked onto a
Chlrorampenicol plate
(l2.Sug/ml)) to grow single colonies.
S 2. Incubate the Chl plate at 43°C overnight, to allow some bacteria
to resolve and to lose
the temperature sensitive pSVl plasmid, and hence lose the tet resistance
gene.
3. To select for tet sensitivity in the resolved BAC, 8 to 16 single colonies
from the Chl
plate are picked, and streaked onto Fusaric acid + Chloramphenicol plate (2 to
8 individual
colonies can be streaked onto each plate). Two controls can be done to test
the
effectiveness of antibiotic selection of the FA+Chl plates: one is streaking a
Tet-resistant
colony (from the Tet+Chl plate), and the second is a tet-sensitive colony
(from the plate
growing the original BAC). Another control can be done is to streak the co-
integrate
colonies on just Chl plate (without fusaric acid).
4. Incubate the FA+Chl plates at 37 °C for 2-3 days. A long incubation
time is necessary
since the resolved colonies grow very slowly due to the presence of the
fusaric acid. Tet
containing colonies should not grow even in 48 hrs incubation. Therefore,
there should be
much fewer colonies on the Chl+Fusaric acid plates than on the Ch1 plates.
These colonies
are the resolved colonies.
5. A) Two alternative methods can be used to identify the correctly resolved
BACs. If
both A and B homology are about the same length, one can just pick 10-20
colonies,
prepare miniprep DNA by alkaline lysis and do Southern blot to analyze the
targeting
events. About half of the resolved BACs should contain the correctly targeted
marker
genes. B) If the two homology arms are not the same length ( > 500 by
difference), one
should use the colony hybridization to select the correctly resolved BACs.
Pick 50-100
individual colonies from FA+Chl plates, streak them onto ChI plates and also
onto the
Tet+Chl plates, as a control for Fusaric acid selection. Each plate can
accommodate 50
test colonies and two positive control colonies, which are the co-integrate
colonies from the
Chl plate. Grow the colonies overnight at 37 °C. Abundant colonies
should grow on the
Chl plate, and none on the Tet+ Chl plate, except the positive co-integrate
controls. The
selection for tet sensitivity at step 4 is very stringent and has essentially
no background.
Therefore, all the colonies that grow on FA+Chl plates have been found to
contain
resolved colonies. Colony hybridizations is performed, according to the
standard protocols
[Sambrook et al., (1989) supra], to select for the colonies that are resolved
and resulted in
CA 02294619 1999-12-21
WO 98/59060 PCT/US98/12966
47
targeted modification. The colony hybridization probe should be part of the
recombination
cassette excluding the arms, such as IacZ, Neo, GFP or polyA sequences.
6. Midi-prep DNA are prepared for the positive clones by the alkaline lysis
method as
described above. Restriction digests and Southern blots are performed to
confirm targeting
event on both homology side (A and B).
7. Pulse field gel analyses should be done to confirm the modification event
and to
determine if there are any rearrangements in the modified BACs. Since there
are two Not I
site flanking the BAC insert (Research Genetics), digestion with Not I should
reveal the size
of the modified BAC. Generally MIuI, PacI and PmeI sites are included in the
recombination cassette. Digestion with these enzymes will confirm the
targeting events.
Double digestion with these enzymes and with Not I will help to determine the
integration
site of the recombination cassette in the BAC. XhoI is usually used to
fingerprint the
modified BAC, since it has a wide distribution of fragment sizes. Comparing
the Xho
digestion pattern of the modified BAC with the original BAC will reveal any
gross
rearrangements in the modified BAC. Other enzymes, such as BamHI and AvrII can
also
be used for this purpose. Targeted BAC modification has been found not to
introduce any
unwanted rearrangements into the BACs. Probes used to hybridized to the PFGE
blots
inch~de: insert specific probes (s.a. lacZ, PolyA, GFP and Neo) and whole BAC
probe (to
reveal all the digested bands from the BAC). Once the modified BACs are
confirmed to
have the specific targeted modification events and the lack of rearrangements,
these BACs
are ready to be used for the biological experiments, such as producing
transgenic mice or
transfecting cells.
4. Preparation of large~uant'~ty hieh quality linearized BAC DNA for
nronuclear
injection
(I) Maxiprep BAC DNA preparation(1 day):
See the isolation and initiai mapping of BACs section above.
(II) Prepare intact linearized BACDNA for pronuclear injection (1 day):
1. Digest 50ug cesium banded BAC maxiprep DNA overnight in SOOuI total volume
containing:
50 ~cg DNA
50 ld lOX NotI buffer or Buffer 3 (NEB)
50 ~,1 IOXBSA
12.5 ~,1 100mM Spermidine (final concentration 2.5mM)
CA 02294619 1999-12-21
WO 98/59060 PCT/US98/12966
48
25 ~1 (250units) Not I (NEB)
ddH20 to SOOuI total volume
Digestion is carried out at 37 °C for > lOhrs.
2. Preparation of the CL,4b Column (performed at room temperature):
Take a Sml plastic pipette, air-blow the cotton to the tip and clamp the
pipette on a stand.
Shake the CL4b sepharose (Phamacia) well, and gradually add the sepharose into
the plastic
pipette. Add until the packed sepharose to almost the top (with about lml
space to spare).
Ne ~ cr let the column dry.
3.Once the column is ready, use a lOml syringe to set a reservoir on top of
the column
(buffer is added to the reservoir). Then equilibrate the column with 30m1 of
the injection
buffer (IOmM Tris.HCI,pH7.5, O.ImM EDTA and 100mM NaCI). This takes about 2-3
hours.
4. Now add Sul lOX DNA dye into the O.SmI digested BAC DNA. Take the reservoir
out
and gently add the DNA(+dye) onto the top of the column with a pasteur
pipette. Wait
until the DNA+dye just goes into the column, gently add O.SmI of injection
buffer on top
of the column.
S.Once the injection buffer almost goes in, the reservoir is put back with 10
ml of injection
buffer in it. Now start collecting O.SmI fraction with a 24 well plate.
Generally about 12
fractions are collected (or until the blue dye is almost at the bottom of the
column).
6. Run SOuI of each fraction on a pulse field gel to identify the appropriate
fractions. The
bands should be visible after ethidium bromide staining. A Southern blot is
performed in
order to choose the fractions with highest yield, and the least degradations.
7.Purified DNA is stored at 4 °C. It is stable for weeks (e.g., no
degradation was detected
after 3 weeks).
Results
BACs are useful as tools for studying the regulation of gene expression in
vivo. In one
particular example, a BAC can include the marine brain specific zinc finger
gene, RU49
[Yang et al., Development 122:555 (1996)]. RU49 has beenshown by in situ
hybridization
to be expressed in the granule cell population of the marine cerebellum, the
dentate gyrus
and the olfactory bulb in the brain. However, proper expression of the lacZ
marker gene
could not be obtained in the cerebellum with a 10 kb RU 49 promoter-lacZ
construct in
transgenic mice, e.g., only one out of ten lines showed partial expression in
the cerebellum.
To overcome this problem, an homologous recombination based method for
inserting an
CA 02294619 1999-12-21
WO 98/59060 PCT/US98/12966
49
IRES-lacZ marker gene into the BAC containing RU49 was developed. The germIine
transmission in transgenic mice of an intact modified BAC and proper
expression of the IacZ
transgene in the cerebellum is demonstrated.
To modify BACs in E.coli, a temperature sensitive shuttle vector based system
for
homologous recombination was employed [O'Connor et al., Science 244:1307-1312
(1989);
Hamilton et al., J. Bacteriol. 171:4617 (1989)]. This temperature sensitive
plasmid will
replicate in cells growing at the permissive temperature (30°C), but
will be lost in cells
growing at the restrictive temperature (42-44°C) because its origin of
replication can not
function at the restrictive temperature [Hashimoto-Gotoh et al., J. Bacteriol.
131:405-412
(1977)]. To overcome the recombination deficiency of the BAC host i.e., a RecA-
host cell,
the L.coli recA gene was introduced into the temperature sensitive shuttle
vector. When
transformed with the temperature sensitive shuttle vector (carrying a
recombination cassette
containing the recA gene) the host strain becomes conditionally competent to
perform
homologous recombination allowing in vivo modification of the resident BAC.
The general strategy for targeted BAC modification is shown in Fig. l, which
illustrates the
steps involved in inserting a marker gene, e.g., IRES IacZ pGKpolyA (ILPA),
into the BAC.
First, two small genomic fragments, e.g., A and B, each containing greater
than 500 basepairs
of a gene of interest are cloned into the building vector (pBVl) in
appropriate order and
orientation to generate the recombination cassette. The recombination cassette
is then
transferred into the temperature sensitive shuttle vector (e.g., pSVl.RecA).
The reason the
recombination cassette is not built directly in the shuttle vector is due to
the relative
difficulty in manipulating its DNA, due to low copy number [Bochner et al., J.
Bacteriol.
143:926 (1980); Maloy et al., Bacteriol. 145:1110 (1981)] and large vector
size (l lkb).
This shuttle vector is then transformed into E.coli containing the BAC. The
transformants
can be selected by tetracycline resistance (carried by pSYl.RecA) and
chloramphenicol
resistance (carried by the BACs) at 30°C. Since the shuttle vector also
carries the recA gene,
homologous recombination can occur between the shuttle vector and the BAC,
through either
homology at A or B to form co-integrates. The co-integrates are selected by
growth on
tetracycline and chloramphenicol plates at 43 °C. This temperature is
non-permissive for
shuttle vector replication, so that the non-integrated, free shuttle vectors
are lost, resulting in
the selection for bacteria carrying the integrated shuttle vectors, (either
into the BACs or into
CA 02294619 1999-12-21
WO 98/59060 PCT/US98/12966
SO
the bacterial chromosomes). Correct BAC co-integrates can be identified by
Southern blot
analyses.
The co-integrates are then restreaked onto the chloramphenicol plates and
grown at 43 °C
overnight. A fraction of the co-integrates will undergo a second recombination
event
(resolution), through either homology at A or B. The resolved BACs will
automatically lose
the tet and the recA genes, since the excised shuttle vector plasmids cannot
replicate at the
non-permissive temperature. The resolved BACs can be selected by growing on
chloramphenicol and fusaric acid plates at 37°C, as growth on fusaric
acid plates selects for
the loss of tetracycline resistance, i.e., counterselecting against BACs that
are resistant to
tetracycline. As illustrated in Figure 1, depending on which pair of
homologous fragments
undergo the second recombination event, the resolved BAC can be either the
original BAC or
the precisely modified BAC. The desired clones can be identified by colony
hybridization
using a labeled probe for the inserted marker. One important aspect of the
method is that the
recA gene is only temporally introduced into the bacterial host. Once the
modification is
finished, the bacteria will automatically lose the recA gene, returning to the
recombination
deficient state suitable for stable maintenance of the modified BACs.
This strategy termed targeted modification of BACs, was tested by introducing
the IRES-
lacZ- polyA (ILPA) marker into the 131 kb murine BAC169 containing the RU49
locus (Fig.
2A). In this case, the marker gene to the first coding exon of the RU49 gene
was targeted
with homology fragments being 1 kb and 1.6 kb respectively (Fig. 2B). Placing
the IRES
sequence before the IacZ gene ensures the translation of the marker gene even
when lacZ
gene is placed after the translation start site [Pelletier et al., Nature
334:320 (1988)]. The
pSVl.RecA temperature sensitive shuttle vector containing the recombination
cassette was
transformed into the DH10 E.coli strain containing the BAC169 and selected by
growth at
either 30°C or 43 °C on plates containing chloramphenicol and
tetracycline. In contrast to
growth at 30°C, which produced a thick lawn of transformed cells,
growth at 43 °C resulted
in growth of individual colonies. Twenty of these were picked and tested by
Southern blots
for co-integration of the shuttle plasmid into BAC169. As shown in Fig. 3B,
analysis of
twenty clones using the B1 fragment of the RU49 homology cassette resulted in
the
identification of two clones containing the appropriate 4 and 8 kb EcoRl bands
( 10%),
indicating that these clones carry co-integrates that have occurred through
this region of
homology.
CA 02294619 1999-12-21
WO 98/59060 PCT/US98/12966
51
The co-integrates are then resolved as described above by growing the cells
first on
chloramphenicol plates at 43°C and then on chloramphenicol and fusaric
acid plates at 37°C.
Fusaric acid provides a strong counterselection against bacteria containing
the tetracycline
resistance gene. Indeed, 200 colonies picked from these plates were all tet
sensitive,
indicating the stringency of the selection. Duplicated colonies growing on the
chloramphenicol plates were used for colony hybridization with the pgkpolyA
probe. Eight
out of 200 colonies were positive (4 %). Southern blot analyses using either
homology at A1
or B 1 as the probe showed that all these clones contained correctly resolved
BACs (Fig. 3C
and 3D). Three BACs (lanes 4,5 and 8) also contained wild type bands, which
may represent
either contamination from other clones, or a BAC containing two copies of co-
integrates that
resolved through two different homologous regions.
The next step in our analysis was extensive mapping of the modified BACs to
determine
whether any unexpected deletions or insertions were generated during the
modification
procedure. Fig. 4 shows pulsed field gel mapping of the modified BAC L1 and L2
and the
original BAC 169. The same filter was probed separately with the whole BAC169
probe,
with a probe from the inserted marker gene (pgkpolyA) and a probe from the 5'
non-modified
region of the RU49 gene (A2). BAC169 probe (left panel) hybridizes with all
the restriction
fragments for each BAC. Thus, Xhol digestion reveals a finger print of the
modified BACs
showing that essentially all fragments are preserved. The only difference is
that the fragment
containing the ILPA insert is slightly smaller than the corresponding wild
type fragment due
to the replacement of the 7 kb RU49 fragment with the 4 kb marker gene (Fig.
2B).
Digestion with Notl, which releases the entire BAC insert, also reveals a
slightly smaller
DNA insert in modified BACs for the same reason. Since the marker gene was
engineered to
carry an additional Pmel site {Fig.2), digestion of the BAC L1 and L2 DNAs
with this
enzyme results in the generation of two fragments, in contrast to the single
fragment seen in
the original BAC169. The sizes of these fragments allow the determination that
these BACs
contain approximately 75 kb 5' to the Pmel site, and 53 kb 3' to it (Fig. 2).
No apparent
rearrangements have occurred during the modification procedure.
To confirm this conclusion, the modified BACs and BAC169 were probed with both
a
marker specific probe (pgkpolyA) and a probe near the promoter region and
outside the
mouification region (A2). Consistently, both modified BACs contained a single
band
homologous to the marker gene probe which is not present in BAC169. When the
A2 probe
CA 02294619 1999-12-21
WO 98/59060 PCT/US98/12966
52
was used, a single band of expected size appeared in all three BACs.
Additional
fingerprinting of all eight modified BACs with HindIIl, EcoRl and Avrll
digests showed that
no detectable rearrangements or deletion existed in these BACs. Thus, the
temporary
introduction of the recA gene into the BAC host strain does not introduce any
rearrangements
or deletions.
To test the reproducibility and reliability of the targeted BAC modification,
the BAC L1 was
further modified by replacing the IRES-IacZ sequence with pgk-neo sequence. In
this case,
homologous fragments of about 500 by each were used. The modified BACs were
also
efficiently obtained and shown not to have any rearrangements or deletions.
Therefore,
targeted BAC modification is a simple method to precisely modify BACs without
introducing any unwanted changes in the BACs.
To demonstrate the possibility of using the modified BACs for in vivo studies
for gene
expression and gene function, transgenic mice carrying the modified BAC169
with the IRES
LacZ insertion were generated. To purify the 128 kb BAC insert for pronuclear
injection,
several established methods for purifying large YAC DNA were attempted, and
resulted in
considerable amount of DNA fragmentation. In contrast, when a simple gel
filtration column
filled with SEPHAROSE CL-4B was tried, very pure fractions of intact linear
BAC DNA
insert were obtained in an appropriate injection buffer, e.g., 100 mM NaCI, 10
mM Tris.HCl,
pH 7.5 and 0.1 mM EDTA (Fig. SA). Unlike YAC DNA purification which typically
results
in a low DNA yield, the purified fractions using the SEPHAROSE CL-4B column
contained
a large quantity of high concentration linear DNA (e.g., 0.5 mls of 3 ~g/ml
DNA or more).
The purified DNA could be directly visualized with ultraviolet light after
ethidium bromide
staining. The SEPHAROSE CL-4B column could also efficiently separate the
degraded
DNA (in this case in fractions 3-6) from the pure linear DNA (fractions 7-9)
(Fig.SA).
Fraction 8 contained 3 ug/ml DNA and was used directly for pronuclear
injection.
Pronuclear injection into the fertilized C57BL/6 mouse zygote is performed
according to a
standard protocol [Hogan et al., in Manipulating the Mouse Embryo (Cold Spring
Harbor
Laboratory Press, New York, 1986)]. Two different concentrations of fraction 8
BAC DNA
(obtained as described above) were used: 3 ug/ml and 0.6 ~cg/ml. No newborns
were
obtained with the high concentration DNA, suggesting that the high
concentrations may be
toxic to the zygote. However, with the lower concentration of pure linear DNA,
15 newborn
CA 02294619 1999-12-21
WO 98/59060 PCT/US98/12966
53
mice were obtained and two of them (13%), Y7 and Y9, contained the lacZ marker
gene as
demonstrated on a Southern blot (Fig. SB). The intensity of the bands allows
an estimate of
2-3 transgene copies for Y7 and one copy for Y9.
To determine if the intact BACs have, been integrated into the genome, the
presence of both
ends of the BAC ends was assayed for in the transgenic mice. Since both BAC
ends contain
some vector sequence, PCR primers specific to the vector sequence were
generated and used
to amplify the transgenic DNA. The amplified products were then probed with a
third
labeled oligonucleotide probe within the amplified region. As shown in Fig. SC
and Fig. SD:
Y3, Y7 and Y9 have both ends present, while the negative controls do not.
Since Y7 and Y9
also have the IacZ gene, they are likely to contain intact BAC transgenes. For
Y3, whereas it
has both ends it does not contain the IacZ gene. This may be due to either a
rearrangement or
fragmentation during the injection prior to integration.
The Y7 transgenic mice also gave rise to germline transmission after breeding
with B6/CBA
mice. In two litters having a total of eight pups, three pups carried the LacZ
transgene (Fig
SE). Further analysis demonstrated that the transgene was transmitted in a
Mendelian
distribution to more than fifty Y7 offspring.
Next the expression of IacZ gene in the cerebellum of the Y7 transgenic mice
was
determined by whole mount lacZ staining. RU49 is normally expressed in the
granule cells
of the cerebellum, the dentate gyros and the olfactory bulb (including the
subventricular
zone, the rostral migratory stream, and the olfactory bulb proper) [Yang et
al., Development,
122:555-566 (1996)J. In previous studies, RU49 promoter lacZ transgenic mice
with 10 kb
promoter had been generated. However, all of the transgenic lines showed
strong positional
effects: either they did not express in the brain at all, or they were
ectopically expressed in
the cortex, but not the cerebellum. One particular l0kb-lacZ transgenic line
did show
restricted expression in the cerebellum, however, the expression was
restricted to the caudal
half of the cerebellum. With 128 kb of R U49 endogenous sequence surrounding
the lacZ
gene in the Y7 line, at postnatal day 6, the transgenic mice showed a lacZ
expression pattern
closely resembling the endogenous expression pattern (Fig. 6). In the
cerebellum, the marker
gene is expressed throughout the cerebellum (Fig. 6A) and no expression is
seen in five
control littermates (Fig. 6B). Further analysis showed that the transgene is
expressed at high
level in the EGL and lower level in the IGL. The lacZ marker gene is also
expressed in the
CA 02294619 1999-12-21
WO 98/59060 PCT/US98/1296b
54
dentate gyros and the rostral migratory stream and the olfactory bulb (Fig. 6C
and 6D). The
pattern of the BAC transgene expression closely resembles the endogenous RU49
expression
pattern in the brain. It is evident that the large genomic DNA in the BAC
transgene can
overcome the positional effects and confer the proper expression of RU49 in
vivo, in contrast
to our results using conventional transgenic constructs.
As taught herein, bacterial based artificial chromosomes (BACs and PACs) are
ideal for
constructing large DNA for gene targeting. As demonstrated herein with the
targeted BAC
modification method, BACs and PACs can be readily modified to introduce
selection genes,
marker genes, and deletions. Making a BBPAC gene targeting construct will take
about the
same time as making a conventional targeting construct (1-3 months). Moreover,
BBPAC
targeting construct DNA can be easily isolated in milligram quantity and high
quality. This
is advantageous over the YAC system, since it is difficult to purify large
quantities of high
quality YAC DNA.
The present invention is not to be limited in scope by the specific
embodiments describe
herein. Indeed, various modifications of the invention in addition to those
described herein
will become apparent to those skilled in the art from the foregoing
description and the
accompanying figures. Such modifications are intended to fall within the scope
of the
appended claims.
It is further to be understood that all base sizes or amino acid sizes, and
all molecular weight
or molecular mass values, given for nucleic acids or polypeptides are
approximate, and are
provided for description.
Various publications are cited herein, the disclosures of which are
incorporated by reference
in their entireties.
