Note: Descriptions are shown in the official language in which they were submitted.
CA 02454837 2003-10-21
WO 03/093471 PCT/US02/14971
NOVEL 88 PHAGE VECTORS
BACKGROUND
1. Technical Field
S This disclosure relates to phage vectors useful for generating phage display
libraries.
More specifically this disclosure relates to vectors useful for display of
antibodies on phage
particles.
2. Background of Related Art
Filamentous bacteriophage consist of a circular, single-stranded DNA molecule
surrounded by a cylinder of coat proteins. There are about 2,700 molecules of
the major coat
protein pVIII that envelope the phage. At one end of the phage particle, there
are five copies
each of gene III and VI proteins (pIII and pVI) that are involved in host-cell
binding and in
the termination of the assembly process. The other end contains five copies
each of pVII and
pIX that are required for the initiation of assembly and for maintenance of
virion stability.
I S In recent years, vectors have been developed that allow the display of
foreign peptides
on the surface of a filamentous phage particle. By insertion of specific
oligonucleotides or
entire protein coding regions into genes encoding specific phage capsid
proteins, chimeric
proteins can be produced which are able to be assembled into phage panicles.
This results in
the display of the foreign protein or peptide on the surface of the phage
particle.
The display of peptides and proteins on the surface of bacteriophage
represents a
powerful methodology for selection of rare members in a complex library and
for carrying
out molecular evolution in the laboratory. The ability to construct libraries
of enormous
molecular diversity and to select for molecules with predetermined properties
has made this
technology applicable to a wide range of problems.
CA 02454837 2003-10-21
WO 03/093471 PCT/US02/14971
A few of the many applications of such technology are: i) phage display of
natural
peptides including, mapping epitopes of monoclonal and polyclonal antibodies
and
generating immunogens; ii) phage display of random peptides, including mapping
epitopes
of monoclonal and polyclonal antibodies, identifying peptide ligands, and
mapping substrate
sites for proteases and kinases; and iii) phage display of protein and protein
domains,
including directed evolution of proteins, isolation of antibodies, and cDNA
expression
screening.
One important application of phage display has been to construct combinatorial
peptide libraries. Synthetic oligonucleotides, fixed in length but with
unspecified codons, can
be cloned as fusions to genes III or VIII of phage where they are expressed as
a plurality of
peptide:capsid fusion proteins. The libraries, often referred to as random
peptide libraries,
can then be tested for binding to target molecules of interest. This is most
often done using a
form of affinity selection known as "biopanning" or simply "panning".
A variety of commonly used display vectors, with their name, site of
expression,
restriction site used, marker carried on the vector, and reference, are
provided in Phage
Display of Peptides and Proteins, A Laboratory Manual, ed. Kay et al.,
Academic Press,
1996, page 38 and reproduced in the following table:
2
CA 02454837 2003-10-21
WO 03/093471 PCT/US02/14971
Vector Gene Rest. Sites) Marker
(USES III Bgll S-Bgll tetR
fAFF 1 III BstXl _S-BstXItetR
fd-CATI III Pstl _S XholtetR
M663 III Xhol _S-XballacZ+
fdtetDOG III ApaLl S-NotltetR
33 III Sfil S-Notl
88 VIII
Phagemid III ampR
pHENI III Sfil S-Notl ampR
pComb3 III
pComb8 VIII
pCANTAB SE III Sfil SNotl ampR
p8V5 VIII BstXl S-BstX1amp't
~.SurfZap III Notl-S-Spel ampR
A variety of phage and phagemid vectors have been constructed and utilized for
phage
display. Each of the existing vectors has its advantages and disadvantages. By
convention,
vectors that fuse a gene of interest whose protein product is to be displayed
to gene VIII have
been categorized as either type 8, type 8+8 or type 88. Type 8 vectors are
phage vectors
where all copies of gene VIII are fused to a gene of interest for display.
With approximately
2700 copies of pVIII on the surface of the phage particle, there is little
tolerance for large
inserts to be displayed on the phage surface. In addition, strong avidity
effects due to
multivalent display reduce selective pressure for high affinity that is
commonly desired and
may be taken into account.
The 8+8 vectors are phagemid vectors. In the phagemid system, helper phage are
required to package the phagemid genome into a phagemid particle that is
extruded out of the
cell. 1n 8+8 vectors, the gene of interest is fused to a copy of gene VIII on
the plasmid, while
the helper phage retains a wildtype, unfused copy of gene VIII. Hence, the
coat of the
phagemid particle is made up of both wildtype and pVIII fusion proteins
leading to more
stability and a loss of some avidity effects. However, since both helper phage
and phagemid
particles are produced from the same cell, both helper phage and phagemid
viral particles will
3
CA 02454837 2003-10-21
WO 03/093471 PCT/US02/14971
have fusion proteins on the surface leading to a loss of the corresponding
genetic information
from helper phage particles that inadvertently display selected proteins.
The type 88 vectors are phage vectors where both a fused and unfused copy of
gene
VIII are present on the phage vector. The phage vector system is less complex
in that helper
phage are not required. Additionally, there is no loss of selected clones that
result from
inadvertent display on the helper phage surface. However, the presently known
88 vectors
are derivatives of fd-tet, where an insert conferring tetracycline resistance
was introduced at a
convenient restriction site. Unfortunately, the insert disrupts the minus
strand origin of
replication, leading to a defect in minus strand synthesis. As a result, these
vectors have a
very low intracellular RF copy number, making vector production for cloning as
well as
library amplification difficult. In addition, the size of the insert
conferring tetracycline
resistance is approximately 2.6 kb. This large insert, in addition to
insertions into the phage
for protein of interest display (including promoter, ribosomal binding sites,
signal sequences,
stuffer fragments in the case of the cloning vectors, and antibody genes in
the case of
antibody display) yield a large phage genome that is not packaged as
efficiently as smaller
phage genomes. The fd-tet vector has served as the starting point of
construction of a variety
of phage vectors including the fUSE vectors of G. Smith (Scott and Smith,
Science, Vol. 249,
pages 386-390, 1990), fd-CAT1 (McCafferty et al., Nature (London), Vol. 348,
pages 552-
554, 1990) and fdtetDOG of Hoogenboom et al., Nucleic Acid Res., Vol. 19,
pages 4133-
4137, 1991.
S UMMARY
This disclosure describes novel phage vectors useful for generating phage
display
libraries. The novel vectors described herein are produced as the result of
modif canon of a
phage genome at an artificially created cloning site not employed in previous
phage vector
constructions.
4
CA 02454837 2003-10-21
WO 03/093471 PCT/US02/14971
Specifically, a phage genome is engineered in accordance with this disclosure
to
include a restriction site at one of two different positions. In a first
embodiment, a restriction
site is inserted into the phage genome between the end of gene IV and the MOS
hairpin
which serves as a phage packaging signal for newly synthesized single strands
of phage
DNA. In a second embodiment, a restriction site is inserted into the phage
genome after the
MOS hairpin and prior to the minus strand origin.
Once the phage genome is modified to contain the new restriction site, cloning
sites
for receiving one or more genes can be inserted into the phage vector in
accordance with this
disclosure. Preferably, the vector is engineered to be a "88" vector by
inserting at the new
restriction site a nucleotide sequence encoding at least a functional domain
of pVIII and at
least a first cloning site for receiving a gene encoding a polypeptide to be
displayed. In an
alternative embodiment, the 88 vector is engineered to cause display of a
dimeric (e.g.,
heterodimeric) species by inserting first cloning site for receiving first
gene encoding a
polypeptide to be displayed and a second cloning site for receiving a second
gene encoding a
polypeptide capable of dimerizing with the polypeptide to be displayed,
thereby resulting in
display of a dimeric polypeptide or protein. The first and second cloning
sites, if desired and
practical, can be inserted with the nucleotide sequence encoding at least a
functional domain
of pVIII as part of a single cassette referred to herein as a display
cassette.
In particularly useful embodiments, the novel vectors are engineered to
produce phage
particles that display antibodies. After creation of the novel restriction
site and insertion of
the display cassette within a phage genome, a first gene encoding an antibody
heavy chain Fd
is inserted adjacent the nucleotide sequence encoding at least a functional
domain of pVIII to
produce a pVIII fused with a heavy chain Fd. A second gene encoding an
antibody light
chain is also inserted into the vector.
CA 02454837 2003-10-21
WO 03/093471 PCT/US02/14971
BRIEF DESCRIPTION OF THE DRAWINGS
Fig. 1 is a flow chart illustrating the strategy for making a vector based on
modification of the fl genome bet veen gene IV and the MOS hair pin;
Fig. 2 is a flow chart illustrating the strategy for making a vector based on
modification of the fl genome between the MOS hairpin and the minus strand
origin;
Fig. 3 is a map of the vector produced in Example 1;
Fig 4a is the sequence (Seq. ID No. 2) of cassette la used in Example l;
Fig. 4b is the sequence (Seq. ID No. 7) of cassette 2 used in Examples l and
2;
Fig. 4c is the sequence (Seq. ID No. 12) of cassette 3 used in Examples 1 and
2;
Fig. 4d is the sequence (Seq. ID No. 22) of an alternative display cassette
useful in
making an 88 vector in accordance with this disclosure;
Fig. 5a is a map of the pAX131 vector;
Figs. Sb-a show the sequence (Seq. ID No.l3) of the pAX131 vector
Fig. 6 is the sequence (Seq. ID No. 14) of the synthetic gVIII portion of
cassette 3.
Fig. 7 shows the alignment of the oligos for preparation of the synthetic
gVIII; and
Fig. 8 shows a map of the vector pAXl31-gVIII;
Fig. 9 is the sequence (Seq. ID No. 23) of the final inserted construct
resulting from
the insertion of cassettes la, 2 and 3 in Example l;
Fig. 10 is a map of the vector produced in Example 2;
Fig. 11 is the sequence (Seq. ID No. 25) for cassette 1b used in Example 2;
and
Fig. 12 is the sequence (Seq. ID No. 30) of the final construct resulting from
the
insertion of cassettes 1b, 2 and 3 as described in Example 2.
DETAILED DESCRIPT10N Or PREFERRED EMBODIMENTS
The novel vectors described herein are prepared by modifying a phage genome.
While the following description is provided with respect to the fl genome as
the starting
G
CA 02454837 2003-10-21
WO 03/093471 PCT/US02/14971
material, it should be understood that other phage genomes (e.g., M13, fd,
etc.) can be used as
the starting material. Additionally, when the following description refers to
"pVIII", it should
be understood that either full pVIII or a truncated version or fragment
thereof is contemplated
(unless the context indicates otherwise) provided the display function of the
protein is
maintained.
In one embodiment, the present vectors are the result of modification of the
fl
genome between gene IV and the hairpin which serves as a packaging signal
(MOS). First,
the phage genome is engineered to contain a novel restriction site at this
location. Then at
least a first cloning site and a nucleotide sequence encoding at least a
functional domain of
pVIII are inserted at the newly formed restriction site. A first gene encoding
a polypeptide to
be displayed can be inserted at the first cloning site. Because the first
cloning site is adjacent
the nucleotide sequence encoding at least a functional domain of pVIII, once
the first gene is
inserted, the vector effectively encodes a fusion protein of pVIII and a
polypeptide to be
displayed by the phage particle. Any polypeptide that can be displayed by
phage can be fused
to pVIII. Non-limiting examples of polypeptides that can be displayed include
naturally
occurring and synthetic enzymes, hormones, antibodies, antigens, toxins and
cytokines. For a
nonlimiting list of proteins and protein domains that can be displayed, see
Phage Display of
Peptides and Proteins, A Laboratory Manual, Kay et al., ed., Academic Press,
1996.
Optionally, a second cloning site is also inserted at the novel restriction
site. The
second cloning site is adapted to receive a second gene that encodes a
polypeptide that can
dimerize with the polypeptide fused to pVLII. 1n this mamler, display of a
dimeric species
(e.g., a heterodimeric species) can be achieved. Where monomeric display of a
single
polypeptide or protein is intended, the second gene can be eliminated.
In a particularly preferred embodiment, the polypeptide fused to pVIII is an
antibody
heavy chain Fd and the modiFcation to the fl genome also involves inserting a
site for
7
CA 02454837 2003-10-21
WO 03/093471 PCT/US02/14971
cloning into the vector a second gene encoding an antibody light chain. In
this manner, the
vector can be used to make phage particles that display antibody libraries.
Figure 1 is a flow chart showing the steps involved in a particularly useful
method for
producing a phage vector capable of generating phage display of polypeptides
(e.g., libraries
of antibodies) in accordance with this disclosure. In the first step, a
restriction site is
introduced into the into the fl genome between the end of gene IV and the
hairpin which
serves as a packaging signal (MOS). The restriction site can be any known
restriction site.
Suitable restriction sites for insertion include, but are not limited to Nhe
I, Hind III, Nco I,
Xma I, Bgl II, Bst I, Pvu I, etc. It should be understood that if a
restriction site selected for
insertion is present in the native genome, it may be desirable to remove or
disable the native
restriction site to avoid unwanted digestion during further processing. The
restriction site can
be inserted using any technique known to those skilled in the art. In a
particularly useful
embodiment, overlap PCR is used to generate a restriction fragment containing
the desired
restriction site. This fragment is then cloned into the phage genome at
suitable sites.
In the next step, the replicative form (RF) DNA is opened by digestion and a
first
cassette containing a terminator and multiple cloning sites is added.
Depending on the
particular restriction site inserted in the first step, specific methods for
opening the RF DNA
are known to and readily selected by those skilled in the art. Preferably the
first cassette is
engineered to include overhangs which, when combined with the ends of the DNA
formed by
the digestive opening thereof at the inserted restriction site will create a
hybrid site that will
no longer be recognized as the inserted restriction site. In this manner,
subsequent cloning
steps advantageously occur at the cloning sites within the first cassette. if
desired, one of the
cloning sites within the first cassette can be the same as the restriction
site inserted in the first
step to decrease the number of different enzymes employed in the process.
8
CA 02454837 2003-10-21
WO 03/093471 PCT/US02/14971
Methods of preparing suitable cassettes for this and subsequent steps are
within the
purview of those skilled in the art. For example, suitable cassettes can be
created using
overlapping oligonucleotides ("oligos") in a PCR fill in reaction. As another
example,
cassettes can be created using long complementary oligos which can form a
double stranded
DNA cassette. The oligos are mixed in a 1:1 ratio, heat denatured and slowly
cooled to allow
the duplexed cassette to form. Other suitable techniques for creating
cassettes will be evident
to those skilled in the art.
Next, the process shown in Figure 1 involves again opening the RF DNA at one
of
the cloning sites within the first cassette and inserting a second cassette
that includes a
promoter. Any promoter recognized by a host cell can be employed. Suitable
promoters
include, but are not limited to, ara, lac and trc promoters. The promoter
drives expression of
other sequences inserted into the vector, such as, for example expression of
the pVIII fusion
protein and any polypeptides intended to dimerize therewith.
After insertion of the second cassette, the RF DNA is again opened at one of
the other
cloning sites contained in the first cassette, and a display cassette is
added. As noted above,
the display cassette contains at least a nucleotide sequence encoding at least
a functional
domain of pVIII and a f rst cloning site adapted to receive a gene encoding a
polypeptide to
be displayed. The nucleotide sequence encoding at least a functional domain of
pVIII can be
natural or synthetic. Preferably, the display cassette contains a synthetic
gene VIII to avoid
having identical native gene VIII sequences at two different locations within
the vector. The
nucleotide sequence can encode a tnmcated pVIII provided the display function
of the protein
is maintained.
The display cassette contains at least a first cloning site for receiving a
first gene
encoding a polypeptide to be displayed. The cloning site is a region of the
nucleic acid
between two restriction sites, typically with a nonessential region of
nucleotide sequence
9
CA 02454837 2003-10-21
WO 03/093471 PCT/US02/14971
(commonly referred to as a "stuffer" sequence) positioned therebetween. In the
flow chart of
Fig. 1, the first cloning site is defined by XhoI and SpeI restriction sites
adjacent to the
synthetic gene VIII. As those skilled in the art will appreciate, a
suppressible stop codon
could be positioned between the first gene and the nucleotide sequence
encoding at least a
functional domain of pVIII such that fusion display is obtained in a
suppressing host (as long
as the first gene is inserted in-frame) and a secreted protein without pVIII
is obtained in a
non-suppressing host.
The display cassette optionally also contains a second cloning region for
receiving a
second gene encoding a polypeptide that can dimerize with the polypeptide to
be displayed.
For example, where the vector expresses a heavy chain Fd fused to pVIII, the
second gene
preferably encodes an antibody light chain. As with the first cloning site,
the second cloning
site is a region of the vector between two restriction sites, typically with a
stuffer positioned
therebetween. In the flow chart of Fig. l, the second cloning site is defined
by SacI and XbaI
restriction sites. It should of course be understood that where a polypeptide
other than an
1 S antibody is to be displayed (such as, for example, where monomeric display
of a single
polypeptide or protein is intended) a second gene need not be cloned into the
vector. In such
cases the second cloning site can either remain unused, or eliminated
entirely. As those
skilled in the art will also appreciate, where a single chain antibody is
encoded by the first
gene, there is no need to insert a second gene into the vector at the second
cloning site.
Thus, the phage vector produced by the process illustrated in Fig. 1 will be a
modified fl genome that contains, after the native gene IV but before the MOS
hairpin, a
terminator, a promoter, a cloning region for receiving a gene encoding an
antibody light
chain, a cloning region for receiving a gene encoding an antibody heavy chain
Fd to be
displayed and a synthetic gene VIII.
CA 02454837 2003-10-21
WO 03/093471 PCT/US02/14971
In another embodiment, the present vectors are the result of modification of
the fl
genome between the hairpin which serves as a packaging signal (MOS) and the
minus strand
origin. After engineering a novel restriction site at this location, the
vector has inserted at this
site at least a nucleotide sequence encoding a pVIII and a cloning site for
receiving a first
gene encoding a polypeptide to be fused to pVIII and thus displayed by the
phage particle.
Suitable polypeptides to be displayed are those described above in connection
with the
previous embodiment. Optionally, a cloning site for receiving a second gene is
also inserted
at this site. The second gene preferably encodes a polypeptide that can
dimerize with the
polypeptide fused to pVIII. In this manner, display o.f a dimeric species
(e.g., a heterodimeric
species) can be achieved. Where monomeric display of a single polypeptide or
protein is
intended, the second gene can be eliminated.
In a particularly preferred embodiment, the polypeptide fused to pVIII is a
heavy
chain Fd and the modification to the fl genome also involves inserting a site
for cloning into
the vector a second gene encoding an antibody light chain. In this manner, the
vector can be
1 S used to make phage particles that display antibody libraries.
An example of a method of this alternative embodiment of forming a phage
vector
for generating phage display libraries of antibodies in accordance with this
disclosure is
shown in the flow chart of Figure 2. In this embodiment, the first step
involves introducing a
restriction site into the fl genome between the hairpin which serves as a
packaging signal
(MOS) and the minus strand origin. The restriction site can be any known
restriction site.
Suitable restriction sites for insertion include Nhe I, Hind III, Nco I, Xma
I, Bgl II, Bst I, Pvu
I, etc. It should be understood that if a restriction site selected for
insertion is present in the
native genome, it may be desirable to remove or disable the native restriction
site to avoid
unwanted digestion during further processing. The restriction site can be
inserted using any
technique known to those skilled in the art. Tn a particularly useful
embodiment, overlap
11
CA 02454837 2003-10-21
WO 03/093471 PCT/US02/14971
PCR is used to generate a restriction fragment containing the desired
restriction site. This
fragment is then cloned into the phage genome at suitable sites.
In the next step, the replicative form (RF) DNA is opened by digestion and a
first
cassette containing multiple cloning sites and a terminator is added.
Depending on the
particular restriction site inserted in the first step, specific methods for
opening the RF DNA
will be known to and readily selected by those skilled in the art. Preferably
the first cassette
is engineered to include overhangs which, when ligated with the ends of the
DNA formed by
digestion at the inserted restriction site will create a hybrid site that will
no longer be
recognized as the inserted restriction site. In this manner, subsequent
cloning steps
advantageously occur at the cloning sites within the first cassette. If
desired, one of the
cloning sites within the first cassette can be the same as the restriction
site inserted in the first
step to decrease the number of different enzymes employed in the process.
Next, the process shown in Figure 2 involves again opening the RF DNA at one
of
the cloning sites within the first cassette and inserting a second cassette
that includes a
1 S promoter. Any promoter recognized by the host cell can be employed.
Suitable promoters
include, but are not limited to, ara, lac and trc promoters.
After insertion of the second cassette, the RF DNA is again opened at one of
the other
cloning sites contained in the first cassette, and a display cassette is
added. As shown in the
flow chart of Fig. 2, the display cassette contains a synthetic gVIII and at
least a first cloning
site for receiving a first gene that encodes a polypeptide to be displayed,
such as, for example,
an antibody heavy chain Fd. The display cassette optionally contains a second
cloning region
for receiving a second gene, such as, for example, a gene encoding an antibody
light chain. It
should of course be understood that where a polypeptide other than an antibody
is to be
displayed (such as, for example, where monomeric display of a single
polypeptide or protein
12
CA 02454837 2003-10-21
WO 03/093471 PCT/US02/14971
or display of a single chain antibody is intended) a second gene need not be
cloned into the
vector.
Thus, the phage vector produced by the process illustrated in Figure 2 will be
a
modified fl genome that contains, after the MOS hairpin but before the minus
strand origin, a
promoter, an antibody cloning region for receiving a gene encoding an antibody
light chain,
an antibody cloning region for receiving a gene encoding an antibody heavy
chain Fd to be
displayed, a synthetic gene VIII and a terminator.
Optionally, a selectable marker can be added to the present vectors. Non-
limiting
examples of suitable markers include tetracycline or kanamycin resistance.
There are
multiple positions within the phage genome where a selectable marker could be
inserted
provided that transcriptional control elements are recreated if necessary so
that phage
particles can still be produced.
The vectors described herein can be transformed into a host cell using known
techniques (e.g., electroporation) and amplified. The vectors described herein
can also be
1 S digested and have a first gene and optionally a second gene ligated
therein in accordance with
this disclosure. The vector so engineered can be transformed into a host cell
using known
techniques and amplified or to effect expression of polypeptides andlor
proteins encoded
thereby to produce phage particles displaying single polypeptides or dimeric
species. Those
skilled in the art will readily envision other uses for the novel vectors
described herein.
The following examples illustrate the present invention without limiting its
scope.
EXAMPLE I
A novel vector was prepared by using the phage fl genome as the starting
material. A
unique Nhe I restriction site was introduced into the fl genome (GenBank
accession
#NC_001397) between the end of gene IV and the MOS hairpin. (See Fig. 3)
Overlap PCR
was used to generate the restriction site between the end of gene IV and the
MOS hairpin (see
13
CA 02454837 2003-10-21
WO 03/093471 PCT/US02/14971
Fig. 3). The following pairs of primers were used to create the Nhe I site,
which is
underlined: 5' CGCGCTTAATGCGCCGCTAGCTACAGGGCGCGTA 3' (Seq. ID No. 1)
was paired with primer 5' GGTTAATTTGCGTGATGGACAGAC 3' (Seq. ID No. 31) to
generate a 213 by fragment. 5' TACGCGCCCTGTAGCTAGCGGCGCATTAAGCG 3'
(Seq. ID No. 32) was paired with primer S' GAAAAGCCCCAAAAACAGGAAGAT 3'
(Seq. ID No. 33) to generate a 502 by fragment. The 213 by fragment and the
502 by
fragments generated, which overlap, were then used in a PCR reaction to create
a 681 by
fragment which contained two Psi I restriction sites that flanked the
introduced Nhe I site.
These Psi I sites were used to clone the final 492 by fragment into Psi I
digested Fl phage
DNA. This causes a new NheI site to be created, which is underlined in the
primer
sequences. The double underline indicates the four additional bases generated
by the creation
of this restriction site. The incorporation of the new restriction site was
verified using the
resulting replicative form (RF) DNA of phage 205-13.1-1 by Nhe I digestion
and/or sequence
analysis. Additionally, the impact on phage assembly appeared minimal as
plaque size of the
wild type was similar to that of the modified phage. Plaque assays were
performed by
allowing dilutions of phage 205-13.1-1 to infect a bacterial host, then the
mixture was plated
in top agar onto an LB-agar plate. The plates were incubated overnight to
allow a bacterial
lawn to form. Circular areas of slower bacterial growth are the result of
phage infection and
were visualized on the plate. If the site of insertion/modification of the fl
genome interfered
with the phage morphogenesis cycle, then the size of the clear circular plaque
for the wild
type would have been bigger and less turbid than that of the modif ed phage,
but this was not
the case.
The modified fl phage 205-13.1-1 was digested with the restriction
endonuclease Nlle
I and cassette 1 a (Seq. ID No. 2) (see Figure 4a), which contains a
terminator (Krebber, A.,
Bunnester, J., and Pluckthun, A., Gene (1996) 178, pp71-4) and multiple
cloning sites was
14
CA 02454837 2003-10-21
WO 03/093471 PCT/US02/14971
ligated into that position. Cassette la was created by making use of long
complimentary
oligos which formed the double-stranded DNA cassette. The two oligos were
mixed together
at a 1:1 molar ratio, heat denatured and slowly cooled to allow the duplexed
insert to form.
The annealing of the oligos was such that single stranded DNA overhangs were
present at
each end (underlined). These overhangs are compatible with the Nhe I overhangs
remaining
after Nhe I digestion of the vector. However, the ends do not regenerate
functional Nhe I
sites after annealing. The oligos used for this construction method were:
Cas. 1 a-F2:
S'CTAGAGTACCCGATAAAAGCGGCTTCCTGACAGGAGGCCGTTTTGTTTTGCAGC
CCACCTGCTAGCATGAATTCGTGGTACCT 3' (Seq. ID No. S)
Cas. 1 a-B2:
5'CTAGAGGTACCACGAATTCATGCTAGCAGGTGGGCTGCAAAACA_AAACGGCCT
CCTGTCAGGAAGCCGCTTTTATCGGGTACT 3' (Seq. ID No. 6).
These two oligos contain Kpn I, Eco RI, and Nhe I sites (double underlined).
As described
above, the constructs were verified by both sequence analysis of the ItF DNA
and by
analyzing plaque size, which remained unaltered. Insertion of this cassette
generated phage
205-63.1.
The TZF DNA of 205-63.1 was digested with Nhe I and Eco RI and cassette 2,
(Seq.
ID No. 7, see Fig.4b) which contains the trc promotor, was added (see for
example
Invitrogen's pTrcHis A promotor sequence). Cassette 2 was generated by making
use of long
complementary oligos which formed the double stranded DNA cassette. The two
oligos were
mixed together at a 1:1 molar ratio, heat denatured and slowly cooled to allow
the duplexed
insert to form. The annealing of the oligos was such that single stranded DNA
overhangs
were present at each end, which were compatible with the NheI/ EcoR 1 digested
vector. The
oligos used .for construction were:
CA 02454837 2003-10-21
WO 03/093471 PCT/US02/14971
Cas. 2-F2
5' CT AGC tgt tga caa tta ate ate egg cte gta taa tgt gtg gaa ttg tga gcg gat
aac aat tG 3' (Seq.
ID No. 10)
Cas.2-B2
5' AAT TCa att gtt ate cgc tea caa ttc cac aca tta tae gag ccg gat gat taa ttg
tea aca G 3' (Seq.
ID No. 11)
The Eco RI overhang is underlined, the Nhe I overhang is double underlined.
Following verification of the resulting phage 205-87.4-3 by sequence analysis,
a third cassette
was inserted between the EcoRI and Kpn I sites of the modified fl . Cassette 3
(Seq. ID No.
12, see Fig. 4c) is the display cassette and contains the first and second
cloning regions and a
synthetic gene VIII from the vector pAX131-gene VIII (205-91, see below).
PAX131 is a phagemid vector prepared by modifying Bluescript II. Fig. 5a is a
map
of pAX131. Figs. Sb-a show the nucleic acid sequence (Seq. ID No. 13) for
pAXl3l. The
preparation of pAX131 is described more fully in commonly owned pending
application
entitled PHAGEMID VECTORS filed on even date herewith under Express Mail Label
No.
EL820507456US (U.S. Provisional Application Serial No. 60/287,355, filed April
27, 2001),
the disclosure of which is incorporated herein in its entirety by this
reference.
Preparation of Synthetic gene VIII and The Display Cassette
The cloning region of pAX131 was constructed using an overlapping oligo
approach
(synthetically generated region). The area of interest includes a ribosomal
binding site
followed by an optimized (for E. coli expression) ompA leader sequence, an Sfi
I, then Sac .I
and Xba I cloning sites for antibody light chains, another ribosomal binding
sequence, an
optimized pel B leader sequence, Xho I and Spe I heavy chain cloning sites
followed by a
downstream Sfi I. The portion ofpAXl3l, v~hich was replaced in the creation of
cassette 3
includes the sequence for a gene lII of fl phage. See Figure Sa.
1G
CA 02454837 2003-10-21
WO 03/093471 PCT/US02/14971
A synthetic gene VIII was generated with a nucleotide sequence optimized for
bacterial expression (Seq. ID No. 14, see Fig. 6). The gene was assembled
using overlapping
phosphorylated oligonucleotides ligated together and cloned into a PCR script
vector. The
assembled gene was cut from this vector using the flanking Spe I and Not I
sites and cloned
into pAX131 at the same sites. The sequences of the overlapping oligos were:
G8-lf: 5'CT AGT GGC CAG GCC GGC Ctg GCT GAA GGC GAC GAC CCG
GCT AAA GCT GCT TTC GAC TCC CTG CAG GCT TCC GCT ACC GAA TAC ATC
GGC TAC 3' (Seq. ID No. 1 S)
G8-2f: 5'GCT TGG GCT ATG GTG GTG GTG ATC GTG GGC GCT ACC ATC
GGC ATC AAA CTG TTC AAA AAA TTC ACC TCC AAA GCT TCC taa GGT ACC GC
3' (Seq. ID No. 16)
G8-3b: 5'GGC CGC GGT ACC tta GGA AGC TTT GGA GGT GAA TTT TTT
GAA CAG TTT GAT GCC GAT GGT 3' (Seq. ID No. 17)
G8-4b: 5' AGC GCC CAC GAT CAC CAC CAC CAT AGC CCA AGC GTA GCC
GAT GTA TTC GGT AGC GGA AGC CTG 3' (Seq. ID No. 18)
G8-Sb: 5' CAG GGA GTC GAA AGC AGC TTT AGC CGG GTC GTC GCC TTC
AGC caG GCC GGC CTG GCC A 3' (Seq. ID No. 19)
Spe I and Not I are shown underlined, and Kpn I is shown with a double
underline.
Figure 7 shows the alignment of the oligos for the preparation of the
synthetic gVIII. All
oligo sequences are shown in the sense orientation, meaning reverse oligos G8-
3b, G8-4b,
and G8-Sb are shown in Figure 7 as their reverse complement in order to see
the alignment
with the forward oligos G8-1 f and G8-2f. Constmction of the synthetic ge- ne
was actually
done by contract with Aptagen. The resulting vector 205-91 (see Fig. 8) was
digested by
restriction enzymes EcoR I and Kpn I to create the display cassette (Seq. ID
No. 12, see Fig.
4c). This display cassette was then inserted into Eco RI and Kpn I digested
205-87 to create
17
CA 02454837 2003-10-21
WO 03/093471 PCT/US02/14971
the final vector, 228-49.14
The sequence of the final inserted construct, 228-49.14, (Seq. ID. No. 23)
resulting
from the insertion of cassettes 1 a, 2 and 3 is shown in Figure 9.
Verification of the final
construct includes sequence analysis of the resulting RF DNA and phage plaque
size as
S described above. Additionally, a tetanus toxoid control antibody was cloned
into the phage
using the Sac I/Xba I sites for the light chain and Xho I/ Spe I sites for the
heavy chain Fd to
create 241-15.29. Western blots of phage virion preps of 241-15.29 indicated
that the Fab
was expressed as a fusion protein with the synthetic gene VIII and
incorporated into virions.
A test panning experiment will also be performed to ensure that the Fab-fusion
is presented
on the phage surface and available for antigen selection. A phage mixture at a
ratio of 1
specific phage/antibody into 106 or more non-specific phage/antibody was used
as the starting
sample. Following 3 to 4 rounds of panning, the specific antibody was selected
and therefore
present at a much higher ratio than the starting ratio. Solid phase panning
was also performed
by adding 10'°-10'z phage to an antigen coated microtiter well for 1-2
hours at 37°. Non-
specific phage were washed off with 0.5% Tween/PBS. Specific phage were eluted
with low
pH (such as O.1M HCI, pH 2.2 with glycine) for 10 minutes at room temperature.
Eluted
phage were neutralized (with 2M Tris Base) and then added to bacterial cells
to allow
infection 15 minutes at room temperature. All cell/phage were plated in top
agar on LB-agar
plates and incubated overnight at 37°. The next day, phage were
recovered from bacterial
plaques by adding S mls media to each large petri dish and scrapping the top
agar into 50 ml
conical tubes. Agar debris was removed by centrifugation. Phage stock was used
directly but
can be concentrated by PEG precipitation if necessay : 4% PEG 8000 + O.SM NaCI
on ice .for
minutes followed by centrifugation at 12,000 x g for 20 minutes at 4°.
The enriched phage
were reselected in additional rounds ofpanning, typically 3-4 rounds total.
18
CA 02454837 2003-10-21
WO 03/093471 PCT/US02/14971
EXAMPLE 2
The overall scheme for modification of fl at an alternate site is similar to
that
described above in Example 1. However, the insertion site for this Example is
between the
MOS hairpin and the minus origin. (See Fig. 10). As above, overlap PCR
was used to generate the restriction site between the MOS hairpin and the
minus origin (see
Fig. 10). The following pairs of primers were used to create the Nhe I site,
which is
underlined: 5' GAACGTGGCGAGAAAGCTAGCGAAGAAAGCGAAAGG 3' (Seq. ID
No. 24) was paired with primer S' GGTTAATTTGCGTGATGGACAGAC 3' (Seq. ID No.
31) to generate a 305 by fragment. 5' CTTCGCTAGCTTTCTCGCCACGTTCGCC 3' (Seq.
ID No. 34) was paired with primer 5' GAAAAGCCCCAAAAACAGGAAGAT 3' (Seq. ID
No. 33) to generate a 397 by fragment. The 305 by fragment and the 397 by
fragments
generated, which overlap, were then used in a PCR reaction to create a 677 by
fragment
which contained two Psi I restriction sites that flanked the introduced Nhe I
site. These Psi I
sites were used to clone the final 488 by fragment into Psi I digested Fl
phage DNA. The
double underlines indicate the mutations introduced in order to create the Nhe
I site. This
procedure generated construct 205-13.2-1, which was sequenced for
verification. Plaque size
did not appear to be significantly altered by this mutation.
205-13.2-1 was digested with Nhe I. Cassette 1b was created for insertion into
205-
13.2-I by making use of long complementary oligos which form the double
stranded DNA
cassette. The two oligos were mixed together at a 1:1 molar ratio, heat
denatured and slowly
cooled to allow the duplexed insert to form. The almealing of the oligos was
such that ends
of single stranded DNA overhangs were present at each end, which were
complementary to
the Nhe I digested vector. Oligos used for this method were:
Cas.l b-F2: 5' CT AGA GCT AGC at GAA TTC gt GGT ACCgta ccc gat aaa age ggc ttc
ctg
aca gga ggc cgt ttt gtt ttg cag ccc ace t1' 3' (Seq. ID No. 28)
19
CA 02454837 2003-10-21
WO 03/093471 PCT/US02/14971
Cas.lb-B2: S' CTA GAa ggt ggg ctg caa aac aaa acg gcc tce tgt cag gaa gcc get
ttt atc ggg
tac GGT ACC ac GAA TTC at GCT AGC T 3' (Seq. ID No. 29)
The overlapping ends are underlined, the Nhe I, Eco RI, and Kpn I sites are
double
underlined. The sequence is such that the ends do not regenerate a functional
Nhe I site.
Insertion of Cassette 1b into 205-13.2-1 generated 205-83.1. This was digested
with
Nhe I and Eco RI and cassette 2, (Seq. ID No. 7, see Fig. 4b) was generated
and added as
described above in Example 1 to create construct 205-93.12. This was verified
by sequence
analysis, digested with Eco RI and Kpn I, and cassette 3 was inserted as
described above in
Example 1 to create the f nal construct 228-88.5. This was verified by
sequence analysis and
analysis of plaque size, which again did not appear to be significantly
affected.
The sequence of the final inserted construct (Seq. ID No. 30) resulting from
the
insertion of cassettes 1b, 2 and 3 is presented in Figure 12. Likewise a
tetanus toxoid test
antibody was inserted into the phage vector 228-88.5 to create phage 241-30.7
and this was
analyzed for expression by Western Blot. This indicated that Fab attached to
gene VIII
expressed well and was incorporated into virions. A test panning experiment
can also be
performed as described above.
It is contemplated that the present novel vector can be used in connection
with the
production and screening of libraries made in accordance with conventional
phage display
technologies. Both natural and synthetic antibody repertoires have been
generated as phage
displayed libraries. Natural antibodies can be cloned from B-cell mRNA
isolated from
peripheral blood lymphocytes, bone marrow, spleen, or other lymphatic tissue
of a human or
non-human donor. Donors with an immune response to the antigens) of interest
can be used
to create imrrrune antibody libraries. Alternatively, non-immune libraries may
be generated
from donors by isolating naive antibody B cell genes. PCR using antibody
specific primers
CA 02454837 2003-10-21
WO 03/093471 PCT/US02/14971
on the 1s' strand cDNA allows the isolation of light chain and heavy chain
antibody fragments
which can then be cloned into the display vector.
Synthetic antibodies or antibody libraries can be made up in part or entirely
with
regions of synthetically derived sequence. Library diversity can be engineered
within
variable regions, particularly within CDRs, through the use of degenerate
oligonucleotides.
For example, a single Fab gene may be modified at the heavy chain CDR3
position to contain
random nucleotide sequences. The random sequence can be introduced into the
heavy chain
gene using an oligonucleotide which contains the degenerate coding region in
an overlap
PCR approach. Alternatively, degenerate oligo cassettes can be cloned into
restriction sites
that flank the CDR(s) to create diversity. The resulting library generated by
such approaches
can then be cloned into a display vector in accordance with this disclosure.
Upon introduction of the display into bacteria, phage particles will be
generated that
have antibody displayed on the surface. The resulting collection of phage-
displayed
antibodies can be selected for those with the ability to bind to the antigen
of interest using
techniques known to those skilled in the art. Antibodies identified by this
system can be used
therapeutically, as diagnostic reagents, or as research tools.
It is contemplated that single and double stranded versions of the vectors
described
herein are within the scope of the present invention. It is well within the
purview of those
skilled in the art to prepare double stranded vectors from the single stranded
nucleic acids
described herein.
It will be understood that various modifications may be made to the
embodiments
described herein. For example, as those skilled in the art will appreciate, a
(first gene
encoding a fusion protein having an antibody light chain to be fused to and
displayed by
pVIII and a second gene encoding a heavy chain Fd can be inserted into the
vector at the
newly created restriction site to provide effective antibody display.
Therefore, the above
21
CA 02454837 2003-10-21
WO 03/093471 PCT/US02/14971
description should not be construed as limiting, but merely as
exemplifications of preferred
embodiments. Those skilled in the art will envision other modifications within
the scope and
spirit of the claims appended hereto.
22
