CA2192662A1 - Method of detecting compounds utilizing genetically modified lambdoid bacteriophage - Google Patents

Abstract

Disclosed is an effective lambdoid bacteriophage which includes a protein construct comprising a genetically modified major tail protein truncated at its carboxy terminus, and a target molecule peptide bonded to the carboxy terminus of the tail protein. Also disclosed are nucleic acids encoding the construct and methods of detecting a molecule-of-interest in a solution and of detecting a cell which produces a molecule-of-interest.

Description

W095/34683 PCT~S94/06543 ~ ., METHOD OF DETECTING COMPOUNDS UTI~IZING
GENETIÇABBY MODIFIED LAMBDOID BACTERIOPHAGE

FTT'Tn OF THE INVENTION

This inventlon relates to the detection of compounds, and more 3pecifically to methods for detecting and assaying for a molecule-of-interest and for cells producing such a molecule-of-interest llt;l;7;ng a genetically modified lambdoid bacteriophage.

BAÇKGROUND OF THE INVENTION

Bacteriophages have been used in strategies for detecting molecules-of-interest. For example, a method employing the bacteriophage M13 has been used to assay for various proteins of interest. In this method, M13 phage displaying peptides fused to pIII, a minor M13 coat protein, have been used to screen for protein binding molecules and ~nt;ho~;es (Scott et al. (1990) Science 249:386; Devlin et al. (1990) Science 249:404). Special M13-derived systems have been used to express ~nt;hn~;es as fusion proteins on the surface of the phage, and techni~ues have been developed to enrich the pornl~t;nn for phage expressing antibodies with desired affinities for an antigen (Garrard et al. (1991) Bio~Technol. 9:1373;
Barbas et al. (l991) Proc. NaTI. A~. Sci. (USA) 88:7978). However, the use of M13 in assay methods is limited because M13 infection is not ; ~ t~ly ~ ascertainable. This is because infection by M13 W095/34683 ~CT~S94/06543 2 i 9 2 6 6 2 does not provide the cell with compounds re~uired for growth and is not lytic Like M13, T4 has been used in assays for various proteins such as nerve growth factor (NGF) (Olger et al. (1974) Proc. Natl. Acad. Sci. (USA) 71:1554-1558). In this assay, T4 was chemically coupled to NGF using glutaraldehyde. The phage was then rendered non-iniective by treatment with 2nt;hn~;es against NGF. When unbound~NGF was added to the medium, NGF-linkea phage w~as displaced from the antibody and became free to infect Escherichia coli (E.
coli) . Bacteriophage T4 has also been:used to detect antibodies again8t a wide range of compounds. For example, Becker et al. (r -' . (1970) 7:741) used a T4 bacteriophage to detect antibodies against p-azobenzenearsonate. Hurwitz et al. (Eur. J. Biochem.
(1970) 17:273) used a T4 bacteriophage to detect and estimate levels of angiotensin-II-beta-amide and its antibodies. Gurari et al. (Eur. J. Biochem. (19~) 26:247) used bacteriophage T4 in the detection of 2nt;hn~;eg to nucleic acids. These detection methods involve the chemical modification of the T4 phage resulting in the non-specific exposure on the phage surface of a compound to which the 2nt;ho~;P2 to be assayed are targeted. Such antibodies render the bacteriophage non-infective, thus ~n~hl;ng the decrease in pla~ue formation to be used as a measure of the level of antibody present. The T4 system has also been used to measure hapten concentrations (see, e.g., Hurwitz et al. (1970) Eur. J. Biochem.
17:273-277) In this system, T4 is chemically WO9s/34683 ~ ; PCT~994/06543 ~ ;'."' 2~92662 modified such that it exposes the desired hapten non-specifically on its surface. The addition of anti-hapten antibody destroys the infectivity of the phage. Infectivity is restored in the presence of hapten.

Although both the M13 and T4 phage systems can be usea to detect the presence of a compound by their ability to become infectious in the presence of that compound, ;nf~r~i~n by M13 is normally not immediately ascertainable, and T4 infection is lethal. Thus, these systems cannot be used where a quick screening or selection method based on the survival of the infected bacterial cell is desired, such as where a particular cell type is being selected, or when the object of phage infection is to restore the ability of an auxotrophic bacterial cell to survive on its own under a given set of growth conditions. Special M13-derived phagemid systems carry genes which could endow an infected cell with a selective growth advantage (Barbas et al. (1991) Proc.Natl.AcadSci. (USA) 88:7978). However, these systems have not been used to detect a molecule-of-interest or cells producing such compounds. Furthermore, because gpIII, the M13 protein to which the target molecules are fused, nr~ 1 ateg on the inner membrane facing the periplasm, there are limitations on the nature of the protein fusion. Fusions that are not able to cross the membrane:will not be assembled into M13.
In ~ n~ in all M13 systems where fusion proteins have been used to display proteins on the Wog5/34683 ~ i 9~ 6 6 2 PCT~594/06543 .. .. . ..
outer surface, the displayed protein (or peptide) itself has been the molecule-of-~nterest.

Thus, what is needed are methods for assaying for molecules-of-interest and for cells producing such molecules which are efficlent, accurate, and fast. What are also needed are assay methods which do not have to result in bacterial cell death.
Additionally, assay methodsl]t;~i7;ng bacteriophage infection are needed for non-proteinous molecules of interest and for cells which c~nt;n-l~usly produce these molecules-of-interest.

SUM~ARY OF THE INVENTION

It has been previously determined that removal of up to one third of the gpV protein of the bacteriophage lambda does not affect the assembly or infectivity of the phage (Katsura (1981) J. Mol. Biol.
146:493-512). Fur~h~ ~ it has been discovered that lambdoid bacteriophage having a target molecule peptide linked to one of its ~ ts, the gpV
protein, can be successfully assembled in vivo such that the target molecule is displayed on the outer surface of the phage. In addition, the genetically modified lambdoid bacteriophage ~e;nr~;nc its ability to infect E. coli. These findings have been exploited to develop the present invention, namely, methods of detecting a molecule-of-interest in a solution and of detecting a cell which produces such a molecule-of-interest, utilizing a genetically modified lambdoid bacteriophage.

w095l34683 ,~ 2 1 q 2 6 6 2 PCT~S94106~3 As used -~erein, the term ~'lambdoid bacteriophage" is meant to ~nt -RR all lambda-related phages and all derivatives, genetically ~ng;n~ed derivatives, and hybrids thereof, such as, but not limited to, 080, ~81, phages 21, 82, 424, 432, Aimm434, Aimm21, phagemids, A~3~, and Agt.

In this method, a protein construct is provided which includes a genetically modified gpV protein truncated at its carboxy terminus and a target molecule peptide bonded to the carboxy terminus of the truncated gpV protein. As used herein the term ~gpV protein" is meant to ~nl -AS any major tail protein found in the lambdoid bacteriophages. This includes but is not limited to lambda gpV protein, gpV-related proteins and equivalents of lambda gpV
protein in the tails of other lambdoid viruses. In preferred ~ ',o~; t~ of the invention, the target molecule is a protein such as an enzyme, enzyme substrate, ; ~globulin, or binding fragment thereon, hormone, ligand, toxin, growth factor, cytokine, receptor, or a fragment or analog of any such protein.

In some embodiments the protein construct further ;n~ at least an antigenic portion of a third protein, or fragment thereof, to which ~nt;ho~;es have been raised. A preferred third protein i8 a marker protein such as B-galactosidase, ~hll _h~n;col acetyltransferase, or ~lk~l;n~
phosphatase. As used herein, the term "marker W095/34683 ~i 9 ~ 6 6 2 PCT~S94/06543 I

protein~ refers to the protein or fragment thereof to which an antibody i8 available.

In one aspect of the invention, the protein construct is provided by transforming a bacterial cell with a nucleic acid encoding the protein construct. This bacterial cell has been preinfected with a lambdoid bacteriophage assembly mutant that has defective or substantially no gpV protein~ The transformed cell is induced to express 1: ~ a~; d components and the protein construct, and then to assemble a lambdoid phage therefrom, the phage having the target protein on=its outer surface. The bacteriophage are then isolated from the cell.

In another ~ i the lambdoid bacteriophage is provided for use in the method of the invention as follows. A bacterial cell is infected with a lambdoid bacteriophage assembly mutant having defective or absent gpV protein. This bacterial cell has been pre-transformed with a nucleic acid encoding the protein construct. The cell is induced to express the viral c ~ ~n~R and protein construct and to assemble a lambdoid phage therefrom. The lambdoid phage thus formed has the target protein on its outer surface.

The target molecule on the bacteriophage is then processed such that the phage is rendered reversibly non-infective or inactive, (i.e., with further treatment the non-infective phage can become infective again). In some aspects of the invention, inactivation is accomplished by treating the WO 95134683 .~ PCT)US94JO6543 ~ ~ qZ662 bacteriophage with a molecule that binds the target molecule. The binding of the target molecule renders the phage non-infective. Preferably, the binding molecule is an immunoglobulin, or binding portion thereof, specific for~ an antigenic determinant on the target molecule, a receptor specific for a ligand-type target molecule, or an ~ ;7ed ligand which binds to a receptor-type target molecule. In other aspects, the binding molecule i8 a matrix to which the bacteriophage-linked target molecule i6 immobilized.
Immobilization renders the phage non-infective because it cannot bind to the lambda cell receptor.

The non-infective bacteriophage is then treated with a solution which rrnt~;nc a molecule-of-interest. In some preferred Pmho~;r c the solution is a cell lysate, cell culture medium, or a biological sample such as blood, urine, saliva, serum, semen, or lacrimal secretions.

The term "molecule-of-interest" is meant to encompass any molecule whose activity or presence is desired, and which can render the non-infective bacteriophage infective again. Useful molecules-of-interest are proteins, peptides, l1~LI S~, nucleic acids, carbohydrates, lipids,~ glycoproteins, glycolipids, proteolipids, lipoproteins, lipopolysaccharides, vitamins, toxins, terpenes, antibiotics, and cofactors.

In some r~ho~;r-ntq, the molecule-of-interest is a protein such as an enzyme which cleaves the WO9S/34683 2 PCT~594/06543 1 q2662 ~ ~,q '~
target molecule, an enzyme substrate. Cleaving o~=
the binding molecule-li~ked target molecule liberates the bacteriophage from the binding molecule, thereby renderlng it infective once again.

In other embodiments, the molecule-of-interest is unbound target molecule. Unbound target molecules present in the solution-to-be-tested displace the binding molecule on the phage-linked target molecule and bind with the binding molecule, thereby liberating the phage and rendering it infective once again. In another aspect of the invention, the molecule-of-interest i5 different than the target molecule but yet is capable of binding to the binding molecule, thus displacing the target molecule.

In one preferred : ~Im~; , the target molecule and the molecule-of-interest are the same and are ligands, and the binding molecule i8 a receptor specific for that ligand. In another embodiment, the target molecule and the molecule-of-interest are the same and are receptors, and the binding molecule is a ligand that binds that receptor. In yet another ~ ; , the target molecule and the desired molecule (or molecule-of-interest) contain the same antigenic det~rm;n~nt~
and the binding molecule is an immunogiobulin, or portion thereof, that bind5 to that antigenic determinant. In still another embodiment, the target molecule and the molecule-of-interest are the same and are immunoglobulins, or binding portions 3 . ,, ! . PCT/US94/06543 2 ~ 9 2 6 6 2 thereo~, and the bindlng molecule .~nnr~;nR an antigenic determinant bound by that immunoglobulin.

In the method of the invention, a bacterial cell such as an E. col~ cell, is contacted with the treated bacteriophage for a time sufficient for the bacteriophage to infect the cell. The infected cells are then detected, infection being indicative of the presence of the molecule-of-interest in the solution which has rendered the bacteriophage infective.

In some embodiments, detection is accomplished by observing cell death in the form of cell lysis or plaque formation. Lysis results when the nucleic acid of the phage successfully enters the cytoplasm of the cell, directs the cell to produce viral , nnrntR at the expense of ~.~llnl~r C _ ~ntR and to assemble them into phage particles, and causes the cell to rupture or lyse such that the assembled viral particles are released. Plaques result when multiple nei~Ajhhnr;ng cells plated on solid culture dishes lyse in this way, leaving clear or empty spots on the otherwise cloudy culture lawn.

In other aspects of the invention, detection of infection is a~~ 1; Rh~ by observing bacterial cell survival and/or growth at or below 32~C where the bacterial cell infected by the phage is an auxotrophic mutant requiring a gene supplied by the phage for survival and growth and where the phage is a temperate, temperature sensitive phage. In this aspect, the phage, once rendered infective again, _9_ Wogsl34683 PCT~S94/06543 } l '~' 2 1 92662 infects a bacterial cell by injecting it nucleic acid into the host cell.

As used herein, the term "temperate phage"
refers to a phage that can be lytic or ly~ogenic.
When lysogenic, the phage integrates its nucleic acid into the host cell genome and remains ~uiescent, replicating only when the host genome replicates. In its lytic or vegetative multiplication phage, the phage nucleic acid excises itself from the hogt genome, or does not integrate itself into the host cell genome, but rather takes over the protein synthetic r-rh;n~ry of the cell at the expenEe of cellular~ -nt~ and causes phage progeny to be assembled. New phage are released from the cell when the cell lyses. A temperate phage may contain a mutation conferring temperature sensitivity, i.e., it is lysogenic only at low growth temperatures (e.g., at or below about 32~C) and i9 lytic at high growth temperatures (e.g., at about 37~C and above, such as at about 42~C). Thus, at lower growth temperatures, the lysogenic phage DNA integrates into the bacterial cell genome, providing the genome with a gene which the auxotrophic cell requires to 3urvive. Preferably, such a gene encodes a needed protein.

In another : 'a~; t, detection of infection is also accompli3hed by observing bacterial cell survival and/or growth i~ those : '--~;r--nt~ of the invention where the phage, which is temperature sensitive as described in the above paragraph, carries a gene ~n~o~;ng antibiotic resistance.

WO 95134683 ,~ PCT/US94J06543 2 ~ q 2 6 6 2 Infection ~f ~ co~ by this phage will permit the former to survive/grow on media ~nnt~;n;ng the antibiotic whose resistance is encoded by the gene carried by the phage.

In some embodiments, cells that secrete/excrete the molecule-of-interest can be selected from a generally non-secreting population In these embodiments, bacterial cell growth is ;n~;n~t;Ye of phage infection, and hence, of the secretion/excr~tion of the molecule-of-interest.
The bacterial cell to be infected is an auxotroph which itself produces and secretes the molecule-of-interest, which is the same as the target molecule and thus is capable of displacing the target molecule from the binding molecule. In this method the phage carries a bacterial gene encoding a protein required hy the auxotrophic bacterial cell for survival. The phage is inactivated by nn~;hn~;e5 directed to the target molecule, and then is cnnt~~t~ with the solution-to-be-tested which may be medium in which the mutant bacterial cell had been growing and/or with the bacterial cell, itself.
If the medium cnnt~;n~ unbound molecule-of-interest, or if the cell iB producing and secreting it, antibody bound to the phage linked target molecule is displaced and instead binds to the unbound molecule-of-interest in the solution. The liberated phage then infects the bacterial cell, and at lower growth temperatures (e.g., at or below about 32~C), provides the cell with the b~t~r; ~1 gene it needs for growth.

W095/34683 , PCT~S94/06543 ~ ~ ~ l q2662 The invention also includes the protein construct descrlbed above, nucleic acids or gene fusions encoding those protein constructs, and g~n~tirAlly modified, infective lambdoid bacteriophage displaying t~e target ~olecule on their outer surface. -WO 95134683 r~ ..~ ,S-, t P PCT/US94/06543 2 1 q 2 6 6 2 ~RTFF DESCRIPTION OE THE DRAWINGS

The foregoing and other objects of the present invention, the various features thereof, as well as the invention itself may be more fully understood from the following descriptio~, when read together with the ~Co~nying drawings in which:

FIG. lA is a diagrammatic representation of the bacteriophage lambda;

FIG. 18 is a diagrammatic representation of the genetically modified bacteriophage lambda of the in~ention;

FIG 2 is a schematic reprP~n~;on of the nucleic acid sequence and corresponaing amino acid sequence of the gpV protein;

FIG. 3 is a schematic representation of the strategy for constructing the truncated V gene with a multiple cloning site at its carboxy terminus;

FIG. 4 is a schematic repres~nt~tinn of the 3' and 5' primere used to provide the PCR LL _ ~nnt~;n;ng the full length, modified V gene in plasmid p5YM1;

FIG. 5A is a schematic representation of the pSYM1 plasmid ~nn~;n;ng the PCR fragment of FIG. 4;

W095/34683 PCT~S94/06543 S 2 ~ q 2 6 6 ~ --FIG. 5B is a schematic representation of plasmid pSYM2 rnnts~n~nr a--truncated V gene with multiple clon'ng sites;

FIG. 5C -i8 a schematic representation of plasmid pSYM3 crnt~1ning a-truncated V gene and a gene encoding a marker protein;

FIG. 6A i8 a diagrammatic illustration~of one ~mhr~;r t of the method of=the invention;

FIG. 6B i~ a diagrammatic illustration of another embodiment of the method of the invention;

FIG. 6C is a dia~L t ;c illustration of another :embodiment of the method of the invention;

FIG. 6D is a dia~L t; C illustration of another embodiment of the method of the invention;

FIG. 7A is a diagrammatic illustration of another embodiment of the method of the invention;

FIG. 7B is a diagrammatic illustration of another embodiment of the method of the invention;
and FIG. 8 is a diagrammatic illustration of yet another embodiment of the method of the invention.

WO 95/34683 ~ ~, . PCT/US94/06543 ; s ;~ ' 2 1 q 2 6 6 2 DES~TPTION OF T~E ~K~KK~ n,,, Mr ~L~

t has been discovered that a protein construct formed from a lambdoid bacteriophage gpV protein truncated at its carboxy terminus and peptide linked to a target molecule may successfully be assembled in vivo into an infective lambdoid bacteriophage having the target molecule displayed on its outer surface. Furthermore, a phage modified in this manner still retains its ability to infect bacteria.
Ut;l;7;ng such a phage a method of detecting a molecule-of-interest has been developed. In this method, either the death or growth of certain bacterial strains results from the presence of a molecule-of-interest in the solution-to-be-tested ~p~n~;ng on the nature of the infecting 1; ''~;~
bacteriophage genome and any specific needs of the infected bacteria. This method has also been adapted to select or screen for cell lines that r~nt;nll~usly produce a molecule-of-intereat.

One type of lambdoid bacteriophage, the bacteriophage lambda, consists of a l~o~h~ral head or capsid with a radius of 30 nm and a flexible tail 150 nm long ending in a tapered basal part and a single tail fiber (FIG. lA). The genome of the bacteriophage is linear DNA. This DNA is found in the capsid head and has cohesive ends, the right one of which (as defined by the genetic map) protrudes into the upper third of the tail. The tail consists mainly of a tube of 32 disks each consisting of six gpV proteins, the products of the V gene.

W095/34683 PCT~S94/06543 5 ~ 3 ~ 2 1 9 2 6 ~o 2 In the present i~vention, a 7 '~
bacteriophage is genetically modified so as to expose a target molecule on the outer surface of its tail (FIG. 1~). This is accomplished by providing a truncated gene which encodes at least the amino terminal two-thirds of a lambdoid major tail protein such as, but not limited to, the gpV protein, or other major lambdoid tail protein, and linking this gene r, ~, t. to a gene encPding a target protein, thereby forming a gene fusion. The protein product of the gene fusion, i.e., a protein construct, may be expressed in a bacterial cell where it, along with the other phage ~ ~nPntS, is assembled into a lambdoid bacteriophage if genes encoding the other viral ( ~n~nt.~ and enzymes re~uired for phage assembly are present.

The gene fusion may be prepared as follows.
The nucleic acid sequence of the V gene is known ~Sanger et al. ~1982~ J.Mol.Biol. 162:729). This gene is simultaneously cloned and modified by PCR methods ~Scarf, ~Cloning with PCR" in PCR ProtocoLc. A Guide to Methodc and A~ ntinnc ~Innis et al., eds.) Academic Pres6, San Diego, CA ~1990) pp.84-91), resulting in a full length U gene with its carboxy terminal Ser246 codon replaced with a Cys codon TGT. The sequence for the modified V gene is set forth in the Se~uence Listing as SEQ ID N0:2. The modified gpV has been cloned into an expression vector ~pKK223-3, Pharmacia, Piscataway, N~) resulting in the pSYM1 plasmid shown in FIG. 4. This plasmid is used to transform E. coli. Cf course, other p1~ c may be used as well. The transformed strain is induced WO 95~34683 , PCT/IJS94~06543 ? ~ ;~ 21 q2662 such as with isopropylt~io-~-D-galactoside (IPTG) (Sam~rook et al. in Molecular Cloning: A Lohoratory Manual (1989) p.17.13. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY), and then lysed as a source of modified gpV protein. If n~r~RR~ry~ the gpV protein can be purified further as described by ~atsura et al. (Virology (1977) 76:129). When pSYM1 is digested with PstI (New England Biolabs, Beverly, MA) and religated using T4 DNA ligase (New England Biolabs, Beverly, MA), the pSYM2 plasmid shown in FIG. 5B is obtained. ~his digestion results in the 1088 of nucleic acid onro~;ng the C-terminal 2L amino acids of the gpV protein and its r~r' ~c t by nucleic acid encoding the hexapeptide Ser-Phe-Cys-Phe-Gly-Gly (set forth in the Sequence Listing as SEQ ID
N0:7), as depicted in FIG. 3_ Of course, the plasmid may be designed such that it may be digested with other restriction ~n~nnllrleases in the alternative or as well, resulting in the 1088 of other gpV protein amino acids.

Plasmid pSYM2 has a unique PstI cleavage site near the 3' t~r~;mlR of the truncated V gene. The target molecule ~nrn~; ng gene to be fused with the V gene is isolated using the PCR strategy employed for the cloning of the V gene. In this strategy, PCR primers that contain PstI restriction sites are employed to obtain a PstI fragment cnn~-n;ng the gene-to-be-fused. This fragment is then ligated to the PstI site in pSYM2 using T4 DNA ligase. This approach requires that there be no PstI site present ~ in the gene-to-be-fused.

WOg~/34683 ~1 9 ~ PCT~S94/06543 ~ 6 6 2 When the target molecule gene c~nt~1nc a PstI
site, a restriction enzyme that produces blunt ends is used instead of PstI, provided its recognition sequence is not present in the coding r~gion_ Such enzymes include BsaAI, BstllO7I, DraI, Ec1136II, Eco47III, EcoRV, and EheI.

After digesting the pSYM2 plasmid with PstI, the PstI site is converted to a hlunt end using T4 DNA polymerase. (Maniatis et al., Molecular Cloning: A
Laborato~ Manual Cold Spring ~arbor Laboratory, Cold Spring ~arbor,-NY (1982) p.395). Of course, other restriction on~nll~lea5e5 with single recognition sites may be used as well. The gene for the target molecule is then ligated into the plasmid using T4 DNA ligase.

A fl _ ~ of another gene~n~ ng at least an antigenic ~portion of a third protein can be incorporated into the plasmid in a posltion such as, but not limited to, a position distal to the V gene, resulting in a plasmid such as pSYM3 ~FIG. 5C). The third protein may be a marker protein such as ~-galactosidase (encoded by the lacZ gene), chl~L h~n;col acetyltransferase, or ~lk~l;n~
phosphatase, among others. Inclusion of the third protein alleviates the need oi o~taining ~nt;h~;es to the target molecule since ~nt;ho~;es to the third protein can be used to inactivate the phage. Thus, using the same PCR methodology de8cribe above, a V/target/laCZ gene fusion can be prepared which encodes a gpV/target molecule/~-g~ t~ e or protein construct.
-lB-Wogs/34683 ~ i' 2 t 9 2 6 6 2 PCT~S94/06543 The plasmid pSYM3 (FIG. 5C~ is an example of such a plasmid where the gene Pnro~;ng the target molecule is closed between the V gene and a fragment of lacZ. Transcription and translation of this unit results in the production of a gpV/target/~-galactosidase fusion protein. The plasmid pSYM3 ia constructed from pSYM2 as described above where the aforementioned gene-to-be-fu3ed is a 500 bp fragment of lacZ beginning at the ATG start codon of the gene.
The primer used to anneal to the 5' end of the lacZ
gene fragment r~nt~;nc a PstI recognition site, and the primer used for the 3' end r~nt~;n R a HindIII
restriction site. This results in the formation of a V gene~lacZ gene fusion whic~ still rrnt~;nr a unique PstI restriction site into which the gene encoding the target molecule can be cloned.

The target molecule can be any protein, polypeptide, or peptide which is translated from a known nucleic acid sequence and which can be peptide bonded to the carboxy terminal end of the truncated gpV protein without abolishing virus assembly or infectivity. Such target molecules include, but are not limited to proteins such as enzymes (e.g., beta-lactamase, triose phosphate isomerase, or hP~r,k;n~re) enzyme substrates (e.g., pre-interleukin-l, proinsulin, preproinsulin, or erythropoietin) immunoglobulins, orportions thereof (e.g., Fv, Fab, or (Fab')2), receptors or portions thereof (e.g., the estrogen receptor or the insulin receptor), ligands ~e.g., ciliary neuronotrophic factor or lutP;n;~;ng hormone)~, cytokines (e.g., macrophage migration inhibition factor or the WO95/34683 ~ ~ iq 2 r PCT~S94/06543 5a~ 6~

interleukins), growth factors ~e g., fibroblast growth factor or granulocyte colony stimuiating factor) and toxins ~e.g., pertussis toxin or botulinum toxin).

5When constructing the ge~e fusion, the process often results in or requires the inclusion of extraneous DNA sequences that, when transcribed and subsequently translated, result in the inclusion of extraneous additional amino acids in the gene fusion 10product. These additional amino acids may be located between any of the component genes of the construct.

To obtain the modified lambdoid bacteriophage of the invention, the gene fusion encoding the 15protein construct is provided in a plasmid which is used to transform a bacterial cell such as E. coli that can be infected by the bacteriophage.
Alternatively, this cell may be preinfected with bacteriophage having nonfunctional gpV protein prior 20to transformation with the plasmid. The cell is then induced to produce modified phage by rhe~; r~l 8~ e.g., with IPTG) and temperature shifting to a high growth temperature (e.g., about 42~C) 25The assembled phage are purified from the bacterial cell lysate and then rendered non-infective. This may be accomplished by the binding of a molecule to the target molecule on the bacteriophage. Binding stops the phage from being 30able to infect a cell. Useful binding molecules WO 95134683 . ~ ~ ~ ~ ' PCT/IJS~>4/06~;43 2 I q2662 include antibodies-or binding portions thereof such as Fv, Fab, or ~Fab~ fragments. The production of such antibodies and biochemically or genetically produced fragments is well known in the art (see, e g., Antibodies: A Loboratory Manual (Harlow and Lane, eds.) Cold Spring ~arbor Laboratory, Cold Spring Harbor, New York 1988).

Other useful binding molecules include receptors which if necessary may be presented in lipid or detergent micelles or l~iposomes or on cell surfaces to keep their configuration. Such receptor-nnnt~;n;nr; liposomes and micelles can be prepared using any number of methods known in the art ~see, e.g., Georgoussi et al. (1990) Biochem.
Biophys.Acfa 1055:69). When the target molecule is a receptor ligand, the receptor will serve as the immobilizing agent. Receptors which can be presented to the phage in this way include nicotinic acetylcholine receptor ~Chak et al. (1992) Meth.
Enzymol. 207:546), inositol 1,4,5-triphosphate receptor (Kamata et al. (1992~ J. Biochem. 111:546), hepatic vasopressin receptor (Georgoussi, ~d. ), and the rat ovarian receptor for lut~;n;7;ns hormone (Kusuda et al. (1986) ~. Biol. Chem. 261:16161).

Yet other useful binding molecules include all molecules capable of binding to the target molecule in a competitive fashion. When ligands are used as the bindi~g molecule, they must be ; '-;1;7ed as described in the following paragraph.

wos5/34683 ~ PCT~894/06543 )~C~ ~ ! !' L i 9266~ ~!

Alternatively, the=phage can be rendered non-~
infec~ive by binding it via its target molecule to a matrix. Such matrices include, but are not limited to, commercially available materials such as a gel consisting of dextran cross-linked with epichlorohydrin ~e.g., Sçphadex~), a special gel prepared from agarose (e.g., Sepharose~), and agarose. When the phage is immobilized to a matrix it i8 unable to bind to and infect a cell. In this method the phage is immobilized tQ the matrix and thus is unable to enter and infect a cell T h;l;7ation to the matrix may be ac ~1; RhP~ by chemical linkage or by various chemical cross-linking methods (see, e.g. U.S. Patent No.
5,112,615, herein incorporated by reference, and Wilchek et al. (1984) Meth. Enzmol. 104:3~. One type of useful cross-linking reagent is a bifunctional reagent such as ~-maleimidopropionic acid N-hydroxysuc~;n;m;~P ester which can be employed according to the method described in Laooratory Techniques in Bi.- L ~~y and Molecular Biolo~y (Elsevier Science pl~hl;qh;ng Co., Amsterdam, (1988), vol. 19).

The method of the invention has been designed such that the inactivated phage i8 released or liberated from the matrix or binding molecule by the molecule-of-interest. Thus, if the molecule-of-interest is an enzyme, it can be used to liberate non-infective phage by cleaving target molecule bound to antibodies (FIG. 6~), matrices (FIG. 6B), ligands (FIG. 6C), or receptors (FIG. ÇD). In this way, the presence of the molecule-of-interest can be -22- ~ ~

W095/34683 ~ 2 1 9 2 6 6 ~ PCT~S94106543 determined and '~quantztated by the relative infectivity of the phage.

For example, to detect a molecule-of-interest which is=an enzyme capable of cleaving the target molecule ~an enzyme substrate), the method of the invention is performed as follows. Expression of the V gene-~enzyme substrate gene fusion protein i3 induced in E. coli, carrying either pSYM2 (FIG. 5B) or pSYM3 (FIG. 5C), or another similar V gene-enzyme substrate gene fusion-carrying plasmid; by the addition of 1 m~M IPTG (Sambrook et al., in Molecular Clonlng: A Laboratory Manual, Cold Spring Harbor ~aboratories, Cold Spring Harbor, NY (1989) p.
17.13). The bacteria are then infected with a non-lysogenic lambdoid bacteriophage such as Avir (Arber et al., in Lambda~ (Hendrix, ed.) Cold Spring Harbor ~aboratories, Cold Spring Harbor, ~Y (19a3) p. 438).
In this case, successful infection results in the production of non-lysogenic lambdoid bacteriophage c~nt~;r;ng modified gpV protein. The modified bacteriophage are then purified using any purification method known in the art (e.g., Helms et al. (1987) Meth. En~mol. 153:69-82). The modified bacteriophage are then rendered reversibly non-infective nt;1;7;ng ~nt;ho~;es directed against either the enzyme substrate (when pSYM2 is employed), or against a marker protein such as ~-galactosidase (Boehringer M~nnhelm, Tn~;~nArQlis, IN) (w~en pSYM3 is employed). Plasmid pSYM3, is preferred because ~nt;ho~;~c directed against the marker protein can then be used to inactivate the W095l34683 ~ ~ 2 1 ~ ~ 6 6 2 PCT~S94/06543 bacteriophage regardless of the ide~tity of the target molecule. The desired enzyme present in a solution will cleave the antibody-bound phage-linked enzyme substrate, thereby releasing the phage. The released phage are in$ective, and thus can be detected by their ability to lyse a cell.

If the molecule-of-interest is a ligand, the method of the invention can be carried out as follows. In this embodiment, expression of the modified V gene-target gene $usion protein is induced in E. coli which carries either pSYM2 (FIG.
5B), pSYM3 (FIG. 5C), or some similar V gene-target gene fusion-r~nt~;n;ng plasmid. Induction can be~
accomplished by the addition o$ 1 mM IPTG, as described above, which stimulates the tac promoter found in these pl~rm;~. The bacterial cells are then infected with a non-lysogenic 1~ ''-id bacteriophage such as ~vir (Arber et al., in Lom~da Il (Hendrlx, ed.), Cold Spring Harbor ~aboratories, Cold Spring Harbor, NY (1983) p. 438). Infection in this case results in cell lysis and the production of non-lysogenic lambdoid bacteriophage c~nt~i n i ng modified gpV protein. The modified bacteriophage are then isolated as described above and rendered non-infective. This can be ~cr ~ hPd by employing ~ntiho~ies or binding portions thereo$, directed against the target molecule on the outside surface of =~the h~rt~riophage. For example, ~n t;h~ieg, when incubated with bacteriophage lambda:
under the conditions described by ~urwitz et al.

Wo95/34G83 ~ 2 ~ q 2 ~ 6 2 PCT~594/06543 ~Eur. J. Biochem. (1972) 20:247-250), cross-link the phages as a result of their divalent nature.

The modified phage may also be rendered non-infective by employing a receptor which binds phage-linked ligand. However, receptor8 may have to be incorporated into micelles or liposomes as previously noted or presented on the surface of a cell to m~int~;n their configuration for binding ligand (see FIG. 8). Binding of the receptor to the phage-linked ligand adheres the phage to the surface of the micelle, liposome, or cell, thus sterically hindering the ability of the phage to attach to and infect a cell. If the ligand-of-interest is present in the solution-to-be-tested the antibodies (FIG.
7A), or receptor (FIG. 7B) bound to the phage-linked ligand may release the phage in favor of the unbound ligand, thus rendering the phage infective again.
Infectivity is measured by screening for cell lysis.

The method of the invention may al80 be used to detect a cell excreting or secreting a desired ligand, which is the molecule-of-interest (FIG. 8).
In this method, a cell that produces the desired ligand ~hereafter designated PopAl) is selected from a population ~herein designated PopA), that doe8 not produce the ligand. The cells of PopA must be capable of being infected by bacteriophage lambda and must re~uire, for growth, a gene to be supplied by the bacteriophage. For example, a strain of bacteriophage lambda, such as ~ hpE CIts857, which - 30 carries both the temperature sensitive repressor CIts857, and a selectable marker gene, npE

W095/34683 , , ~ 2 ~ q2 PCT~S94/06543 ~ ?~ ?; ~ 6 6 2 ., ., ~

~Erischauf et al. (1983) J. Mol. Biol. 170:827-842), may be employed to infect a bacterial strain carrying the modified gpV protein.

To construct A trpE CIts857, both AEM8L3 DNA and ACIts857 DNA were dige5ted with NheI and the large LL~. t from AEM~3~3 and the small LL~ t from ACIts857 were isolated by electrophoresis in agarose (Maniatis et al. (1982) Molecular Cloning: A Loboratory Manual. Cold Spri~g Harbor ~aboratory, Cold Spring ~arbor, NY. pp. 150-170). The isolated fragments were ligated using T4 DNA ligase and the resulting DNA was packaged in vitro. The resulting phage were used to infect E. coli and a phage stock was prepared from the infected E. coli (Davis et al. (1980) Advanced Bacterial Genehcs, Cold Spring ~arbor ~aboratories, Cold Spring, NY pp.74-77).

After IPTG iLduction of the modified gpV
protein, temperature shifting to 42~C results in the production of l~h~nid bacteriophage that carry the gene re~uired for growth by all cells present in PopA. Either antibodies direoted against the target molecule or a cell receptor specific for the ligand are utilized to render the modified bacteriophage non-infective, as de~cribed above. The presence of a ligand-produci~g bacterial cell ~opAl causes the release of phage by provi~ding unbound ligand to~
which the phage-linked ligand-bound antibody or receptor can bind instead of the phage-linked ligand. When the antibody or receptor chooses to bind with the unbound ligand, it releases the phage WO 95134683 ~ ~i} I ~ PCT/US94106543 ''; 2 1 92662 ~n~h~;ng it to infect the nearby cell which ~ecreted the molecule-of-interest. Infection provides the needed gene, and thereby endows the cell with the ability to grow.

When a temperature sensitive derivative of bacteriophage lambda is employed (e.g., CIts), the ratio of gpV protein to modified gpV protein can be regulated to some extent by varying the time between plasmid (and hence modified gpV protein) expression and bacteriophage (hence gpV protein) expression.
Expression of modified gpV is in~n~;h1e by addition of IPTG. CIts derivatives of bacteriophage lambda are also ;n~n~;hle upon temperature shifting (Maniatis et al., in MolecularCloning: ALoboratoryManual, Cold Spring Harbor Laboratories, Cold Spring Harbor, NY (1982) pp. 78-79). IPTG induction followed by a temperature shift upward to 42~C lead~ to cell lysi5 and the release of bacteriophage lambda C~nt~;n;ng modified gpV.

Likewise, the method of the invention may be used to select a bacterial strain that secretes a molecule-of-interest which i8 an enzyme from a population that does not secrete the enzyme (FIG.
8). This bacterial strain is auxotrophic for a bacterial component and 80 will grow only if provided with the component or with a gene capable of correcting the auxotrophy. In this method, a lambdoid bacteriophage that has a temperature sensitive genotype (e.g., CIts 857) and carries a selectable marker gene may be employed to infect a strain carrying a gene fusion encoding gpV protein WO95/34683 ~ ~ 2 1 9 2 6 6 2 PCT~S94/06543 modified with an enzyme ~ub6trate as the target molecule. After IPTG ;n~nrtlnn of modified gpV, temperature shifting to 42~C results in cell lysis:
and the production of bacteriophage lambda carrying the gene re~uired for growth by all cells.
Antibodies directed against either the target molecule (when pSYM2 is em~loyed) or ~-galactosidase (when pSYM3 is employed) are used to render the~
modified bacteriophage non-infective, as described above. Alternatively, the target molecule may be inactivated by immobilization to a matrix or receptor. If an enzyme-pro~ducing ce~ is present, the en~yme produced by the bacterium cleaves the bound, phage-linked target protein, thereby releasing the phage and r~n~in~ it infective again. The released phage then infects this auxotrophic ceLL at low growth~ temperature, providing it with the gene it needs to survive and grow.

The method of the invention offers several advantages over other systems ~ employing bacteriophages such as Ml3 or T4. First, any target molecule that can be linked to the gpV protein can be employed as long as it does not completely interfere with in vivo assembly or the ability of the resulting bacteriophage to infect bacteria.

Second, the method does not have to result in the death of the infected bacteria. Rather, it can be used to isolate cells that excrete/secrete a desired compound, unlike the M13 and T4 systems. 3y using a temperature sensitive strain of W0 95/34683 ;, ~ 2 1 9 2 6 6 2 PCT/US94/06543 bacteriophage lambda and a bacterial cell population that requires for growth a particular gene product supplied by the bacteriophage, those cells that excrete/secrete the desired compound will render infective an inactivated bacteriophage lambda which, in turn, will infect the cell, and at lower temperatures enable the cell to grow. ~ikewise, the method can be used to isolate either mutant bacterium or a genetically engineered bacterium that excretes or secretes a molecule-of-intere8t from a population of ~on-excretors.

Third, this method enables the selective modification of a specific protein, and hence the selective display of a target molecule, unlike the T4 system. With non-specific modifications, a large percentage of the modified phages are rendered p~rr-n~ntly non-infective. For example, when nerve growth factor ~NGF) was coupled to bact~r;oph~ge T4, 76~ of the phage were rendered non-infective (Olger et al. (1974) Proc. N~l. Acod. Sci. (U5A) 71:1554-1558).

Fourth, as an extension of the method described in the previous paragraph, the method can also be used to screen enzyme libraries for clones having the ability to cleave altered substrate.
Immobilization of the bacteriophage via the altered substrate enables isolation of Etrains from a library that contain an enzyme with the altered specificity from the library. This approach differs from M13 systems where fusion proteins have been used to display proteins because those systems W095/34683 ~J ~'\ 2 1 9 2 6 6 2 PCT~S94/06543 display only the molecuIe-o~-interest, and thus are not useful for the detection of such molecules. The approach described herein with the lambdoid system is unique in this respect.

The following examples illustrate the preferred mode of making and practicing the present invention, but is not meant to limit the scope of the invention since alternative methods may be utilized to obtain similar results.

EXA~PLES

1. Cloning and Modification of the V Gene The V gene was simultaneously cloned into the expression vector pkk223-3 (Pharmacia, Piscataway, NJ) and modified using the PCR protocol of Scharf ("Cloning with PCR," in PCP Protoco~. A Guide to Method ond ~,~i;~/;("~ (InniB et al., eds.) Academic Press, San Diego, CA (1990) pp. 84-91). The resulting plasmid is shown in FIG. 5A (pSYM1). The primers used for the procedure are shown in FIG. 4 and are set forth in the Sequence Listing as SEQ ID NOB-3 and 4. The primer that anneals to the 5' end of the V gene (SEQ
ID NO:3) is designed to include an EcoRI restriction ~nnl~cleA~e cleavage site. The primer that anneal3 to the 3' end of the V gene (SEQ ID NO:4) is designed to include ~indIII and PSTI restriction en~nn1lc~e~Re cleavage sites. In addition, this primer ~nntA;n~ a 3ingle ba3e substitution in the W095/34683 ~ 2 1 92 6 6 2 PCT~S94/06543 last codon of the V gene. This substitution results in the conversion o~ Ser'~s to Cysl~6 The cloned modified V gene is digested with EcoRI and ~indIII (New England Biolabs, Beverly, MA) and ligated, using T4 DNA ligase (New England Biolabs, Inc ), into the expression vector pRK223-3 (Pharmacia, Piscataway, NJ) which was digested with EcoRI and HindIII. DNA digestion with the restriction Pn~nn~leases~ EcoRI and HindIII, was accomplished as described in the New England Biolabs Protocols provided with the ~n~nnn~lea5es. The resulting pSYM1, is shown in FIG 5A.

~hen pSYM1 is digested with PstI and religated using T4 DNA ligase, the plasmid pSYM2 is obtained (FIG. 5B) This digestion results in the loss of nucleic acid ~nco~ins the C-t~rm;n~l 24 amino acids of the V protein and its r~pla~ ~ by nucleic acid Pn~;ng the hexapeptide Ser-Phe-Cys-Phe-Gly-Gly (set forth in the Sequence Llsting as SEQ ID NO:7).

The plasmid pSYM3 was formed by r~pl~m;ng the oligonucleotide generated by digesting pSYM2 with PstI and HindIII with a 501 bp fragment of the E
colilacZ gene. The sequence of lacZ is available from GenBank (Los Alamos, NM; ~cc~R~;~n no. J01636). The lacZ L.~ t encodes the first 167 amino acids of the enzyme, ~-galactosidase. The lacZ fragment was isolated from ~gtll (Young et al (1983) Proc. Natl.
Acad. Sci. USA 80:1194) using PCR as described above for the ;q~l~t;~n of the V gene. The primer that W095/34683 ' PCT~594/06543 2 6 ~ 2 anneals to the 5~ end o~ the -gene i~
CCGCTÇCAGGAATGACCATGATTACGGATTC (SEQ ID N0:8), wherein the underline Requence is a PstI recognition site and the double underlined sequence i8 that of the 5' start of the coding sequence of lacZ. The~
primer that anneals to the 3' end is CCGAAGCTTAACGA~ ~l~CCGTAAC ~SEQ ID N0:9), wherein the underlined sequence i9 a HindIII
recognition site and the double ~in~rl;n~ sequence i8 complementary to the 3' end of the lacZ fragment.
Both pSYM2 and the PCR-cloned lacZ rL~ t are digested with PstI and XindIII and ligated together using a five-~old molar excess of the lacZ fragment.
The resulting plasmid, pSYM3, is shown in FIG. 5C.

2. Preparation of A~tibody Column A column having ~nt;hn~ directed to the target molecule of the V gene protein construct is prepared essentially as described in=~A--'c' A
Loboratory Manual ((Harlow and ~ane, eds.) Cold Spring Harbor Baboratory, Cold Spring Harbor NY, (1988)).
Briefly, specific antibodies are mixed with protein A beads (Sigma Chemical Company, St. Louis, M0) using 2 mg of antibody per milliliter of beads. The bead ~n1nt;nn is mixed gently for l hour at room temperature. The beads are then washed and chemically cross-linked to the ~nt;hQ~;es using a bifunctional cross-linking reagent such as dimethylpimelimidate (Sigma Chemical Company).
Chemical cross-linking is ~c ~ h~ by shaking the antibody-coated beads for 30 minutes in the W095/34683 ~; ~" ~ PCT~594/06543 precence o~ 20 mM dimethylpi-~limi~te. The cross-linking reaction is stopped by washing the beads in 0.2 M eth~nnl ~mi nP followed by a 2 hour incubation at room temperature in 0.2 M eth~nnlAminP~

3. Detection of Ciliary N~uLuLL~hic Factor The gene Pn~o~ing ciliary neurotrophic factor (CNTF) has been cloned, expressed in chinese hamster ovary (CHO) cell5 and sequenced (Negro et al. (1991) ~ur. J. Biochem. 201:289-294). The entire coding sequence for CNTF i9 also available from Genbank (~08 Alamos, NM) (accession no. M29828). This gene does not have any PstI recognition sites. The truncated V gene does contain a Pst site near its 3' terminus: CTGCAG (see SEQ ID NO:1). A PstI fragment ront~;n;ng the CNTF is obtained by PCR using the 5' primer: GTTGCTGCAGGTATGGCTTTCATGGAGCATTCA (SEQ ID
NO:5), wherein the underlined sequence is a PstI
recognition site and the double underlined sequence is that of the 5' start of the coding sequence for CNTF, and the 3' primer: CTGr~.CTACAlllc~ll~lC~llAG:
(5EQ ID NO:6), wherein the underlined seguence is P8tI recognition site and the double underlined sequence is complementary to the 3' end of the coding sequence. Insertion of this PstI LL~ ~
into pSYM2 results in the joining of the truncated V gene to the entire CNTF gene. The GT ~;nnclpntide inserted between the PstI recognition site and the beg;nning of the CNTF coding region is necessary to keep the V gene and CNTF gene in the aame open reading frame 80 that the two genes will be translated into a single polypeptide. This specific WogS/34683 ~ 2 ~ c, - PCTN594/06543 , ~ 6 6 ~ --dinucleotide was cho~-en co ac to not introduce any extraneous amino acids into the gene fusion product.
Competent E. coli SCSl ~Stratagene, La Jolla, CA) i8 transformed with the resulting plasmid as described by Hanahan ( J. Mol . Biol. (1983) 166:557). The=
transformed strain is induced by IPTG and then incubated with Avir for 15 minutes at 37~C. Top agar is added and the mixture is plated. After 6 hours, the plate is overlayed with lambda dilution buffer (10 mM Tris-HCl, pH 8; 2 mM MgCll) and incubated overnight at 4~C. Phage --nt~;ning CNTF
are purified from the resulting lysate by running the lysate over an anti-CNTF antibody column, prepared as described above. The CNTF-modified phage are inactivated using anti-CNTF antibodies obtained commercially or by method~ well known in the art (see, e.g., Antibodies: ALaboratoryManual (Harlow and Lane, eds.) Cold Spring Harbor Laboratory, Cold Spring Harbor, NY (1988)). The appropriate ratio of modified phage to antibody is determined experir-nt~lly as described by Olger et al. (Proc.
Natl. Acad. Sci. (USA) (1974) 71:1554). Inactivated phage are then incubated with media suspected of -~nt~;n;ng CNTF (the solutions-to-be-tested). The infectivity of the phage are~assayed using the plate method of Davis et al. (in Advanced Bacterial Genetics, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY (1980) p.71). The increase in the number of infective phage is directly proportional to the amount of CNTF present in the original sample.

4. Detection of Interleukin-l~ Converting Fnzyme Wogs/34683 ~. ~ , PCT~S94/06543 2 ~ 9 2 ~ G 2 Th~ gene enccding the precursor form of lL-1 (pIL-l~) has been cloned and se~uenced (March et al.
(1985) Nafure 315:641-647). The gene Pn~o~ing pIL-l$
i8 fused, in frame, to the 3~-t~rm;n1lq of the truncated V gene present in pSYM3, as described above, keeping in mind that the 3' primer must be constructed to produce an "in frame" fusion between the 3' terminus of the pIL-l$ fragment and the 5' t~r~innq of the R-galactosidase fragment. Competent E. coli SCSl (Stratagene, La Jolla, CA) are transformed with the resulting plasmid. The transformed strain is induced by IPTG and then infected with ~vir. Phage c~ntAinin~ pIL-l~ are purified from the resulting lysate by running the lysate over the anti-$-galactosidase antibody column, prepared as described above. The pIL-l$-modified phage :are inactivated using anti-$-galactosidase ~rt;ho~;es (Boehringer Mannheim, Tn~i~n~r~1is, IN). The appropriate ratio of modified phage to antibody is determined exper;r ~11y as described by Olger et al. (Proc.
Natl. Acad. Sci. (USA) (1974) 71:1554). Inactivated phage are then incubated with media suspected of cGnt~ininS IL-1$ converting enzyme (ICE), an enzyme which cleaves pIL-l$ to form mature IL-1$. The infectivity of the phage is then assayed by the plate method of Davis et al. (~id. ) . The increase in the number of infective phage is directly proportional to the amount of IOE present in the original sample.

W095/34683 ~ PCTNS94/06543 2 ~ 9 2 ~ 6 2 Selection of Gells Secreting Fibroblast Growth Factor The gene ~nro~;ng human fibroblast growth factor ~FGF) has been cloned, expressed in E. coli, and sequenced ~Zazo et al. (1992) Gene 113:231-238).
This gene is fused, in frame, to the 3'-terminus of the truncated V gene present in pSYM2, as described above. E. coli SCS1 (Stratagene, La Jolla, CA) i9 tran3formed with the resulting plasmid, as described=
above. The transfcrmed strain i9 induced by IPTG
and then infected with AtrpE Clts857 (Stratagene, La Jollal CA). Phage crnt~;ning FGF are purified from the resulting lysate by running the lysate over an anti-FGF antibody column, prepared as described~
above using commercially obtained anti-FGF
antibodies (Sigma Chemical Company, St. Louis, MO).
The FGF-~ ;f;ed phage are inactivated using the same anti-FGF ~n~;ho~;es. The appropriate ratio of modified phage to antibody is determined exper;r ~l1y ag degcribed by Olger et al (Proc.
Natl. Acad. Sci. (USA) (1974) 71:1554). Inactivated phage are incubated with E. coli Sym3 (having the A-, F+ ~trpE recA hflA genotype) that has been transformed with a mouse brain cDNA library that has been cloned into pYEUra3 (~lnnterh Laboratories, Palo Alto, CA), and plated on minimal media (ExperimentsinMolec~larGenetics (Miller, ed.) Cold Spring Harbor Laboratory, Cold Spring Harbor, NY (1972)) lacking tryptophan. Because E. coli Sym3 requires tryptophan for growth, they will grow poorly unless infected by ~trpE Clts857 which carries a gene that WO 95134683 ~ i I . ' ~ 2 ' ~ PCT~US94~06543 , 9~662 restores growth o~ E coli Sym3 on medium lacking tryptophan. Therefore, a cDNA transformant of E. coli Sym3 that secretes FGF releases nearby AE~;3~3 which ~ then infect the cell resulting in a great enhancement of its growth rate relative to other cells on the plate. The infected cell grows into a visible colony. The colony is then streaked onto the same media, and colonies arising from single cells are ~hose that secrete FGF.

Those skilled in the art will recognize, or be able to ascertain, using no more than routine exper'-- t~t1~n~ uu~ equivalents to the specific substances and procedures described herein.
Such equivalents are considered to be within the scope of this invention, and are covered by the following claim8.

WO95/346~3 ~ US~.~ ' 2 t 9 2 6 6 2 PCT~S94/06543 ~U~NC~ ~ISTING

~1) GENERAL INFORMATION:
(i) APPLICANT: Ray, Bryan L.
Lin, Edmund C C.
Crea, Roberto (ii) TITLE OF INVENTION: Method Of Detecting Compound~~
TJtilizing ~.~n~t;r~lly Modified Lamhdoid ~:~rt~.ri rph~ge (iii) NUMBER OF ~Uu~N~:S: 9 (iv) CORRE~u~:N~ ADDRESS:
(A) ~TlllRT'.q.r~R: Lappin & Kusmer (B) STREET: 200 State Street (C) CITY: Boston (D) STATE: MA
(E) COUNTRY: U.S.A.
(F) ZIP: 02109 (V) U,).A~U'l'.t!;K RTtAl)AT~T~ FORM:
(A) MEDIIJM TYPE: Floppy disk (B) COMPIJTER: IBM PC compatible (C) OPERATING SYSTEM: PC-DOS/MS-DOS
(D) SOFTWARE: Paten~n Release#1.0,Ve~ion#1.25 (vi) CURRENT APPLICATION DATA: . .
(A) APPLICATION NUMBER:
(B) FILING DATE:
(C) CLASSIFICATION:
(viii) ATTORNEY/AGENT INFORMATION:
(A) NAME: Kerner, Ann-Loui~e (B) REGISTRATION NUMBER: 33,523 (C) ~ N~/DOCKET NUMBER: SYZZ-OllPCT
(ix) TELECOMMUNICATION INFORMATION-(A) TELEPHONE: 617/330-1300 (B) TELEFAX: 617/330-1311 (2) INFORMATION FOR SEQ ID NO:l:
(i) ~U~N~ ~MARA~TT'RT.qTICS:
(A) LENGT~:--2~Ç amino acids (B) TYPE: amino acid (C) STRANnT~n~ ingle WO9Sl34683 ~?~ 2 1 926 62 PCT~594806543 (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (iii) XYPOTXETICAL: YES
(iv) ANTI-SENSE: NO
- (xi) ~yu~N~ D~SCRIPTION: SEQ ID NO:l:
Met Pro Val Pro Asn Pro Thr Met Pro Val Lys Gly Ala Gly Thr Thr Leu Trp Val Tyr Lys Gly Ser Gly Asp Pro Tyr Ala Asn Pro Leu Ser Asp Val Asp Trp Ser Arg Leu Ala Lys Val Lys Asp Leu Thr Pro Gly Glu Leu Thr Ala Glu Ser Tyr Asp Asp Ser Tyr Leu Asp Asp Glu Asp Ala Asp Trp Thr ~la Thr Gly Gln Gly Gln Lys Ser Ala Gly Asp Thr Ser Phe Thr Leu Ala Trp Met Pro Gly Glu Gln Gly Gln Gln Ala Leu g0 95 Leu Ala Trp Phe Asn Glu Gly Asp Thr Arg Ala Tyr Lys Ile Arg Phe Pro Asn Gly Thr Val Asp Val Phe Arg Gly Trp Val Ser Ser Ile Gly Lys Ala Val Thr Ala Lys Glu Val Ile Thr Arg Thr Val Lys Val Thr Asn Val Gly Arg Pro Ser Met Ala Glu Asp Arg Ser Thr Val Thr Ala Ala Thr Gly Met Thr Val Thr Pro Ala Ser Thr Ser Val Val Lys Gly Gln Ser Thr Thr Leu Thr Val Ala Phe Gln Pro Glu Gly Val Thr Asp Lys Ser Phe Arg Ala Val Ser Ala Asp Dy8 Thr Lys Ala Thr Val Ser Val Ser Gly Met Thr Ile Thr Val Asn Gly Val Ala Ala Gly Lys Val W095/34683 ~ 3 3 ~ 2 1 9 2 6 6 2 PCT~S94/06543 Asn Ile Pro Val Val Ser Gly Asn Gly Glu Phe Ala Ala Val Ala Glu Ile Thr Val Thr Ala CYB

WO 95t34683 ~ ;,.'' ~; '~; j i ~ 1 9 2 6 6 2 PCT/ltS94/OC543 (2) INFORMATION FOR SEQ ID NO:2 (i) SEQ~ENCE CHARACTERISTICS:
(A) LENGTH: 741 base pairs (B) TYPE: nucleic acid (C) sT~ANn~nNE~ single (D) TOPOLOGY: linear (ii) MOLECU~E ~YPE~ DNA tgenomic) (iii) HYPOTHETICAL: YES
(iv) ANTI-SENSE: NO
(Xi) ~UU~N~ DESCRIPTION: SEQ ID NO:2:

ATGCCTGTAC CA~ATCCTAC AATGCCGGTG AAAGGTGCCG GGACCACCCT ~lG~lllAT 60 ~Arrr~rArCG GTGACCCTTA CGCGAATCCG CTTTCAGACG TTGACTGGTC GCGTCTGGCA 120 A~AGTTAAAG ACCTGACGCC rrGr~A~rTG ACCGCTGAGT CCTATGACGA CAGCTATCTC 180 GATGATGAAG ATGCAGACTG GACTGCGACC GGGCAGGGGC AGAAATCTGC rr,rA~.ATArc 240 AGCTTCACGC TGGCGTGGAT r-rr~r.rArAr. CAGGGGCAGC AGGCGCTGCT GGC~L~llL 300 AATGAAGGCG ATACCCGTGC rTATA~AATc CGCTTCCCGA ACGGCACGGT CGATGTGTTC 360 CGTGGCTGGG TCAGCAGTAT CGGTA~GGCG GTGACGGCGA AGGAAGTGAT QCCCGCACG 420 GTGAAAGTCA CCAATGTGGG A~ ~L~ ATr.r,r~r.AAr. ATrr~r~rr~r GGTAACAGCG 480 GCAACCGGCA TGACCGTGAC GCCTGCCAGC A~ lGG Tr.~AAr.GGr~ GAGCACCACG 540 CTGACCGTGG CCTTCCAGCC GGAGGGCGTA Arrr.~r~A~.A G~1LL~1~C G~l~l~l~CG 600 r.~TAA~rAA AAGCCACCGT ~LC~l~AGT GGTATGACCA TCACCGTGAA CGGCGTTGCT 660 G~r~Gr~rr, TCAACATTCC GGTTGTATCC GGTAATGGTG AGTTTGCTGC GGTTGCAGAA 720 ATTACCGTCA CCGCCTGTTA ~ 741 (2) INFORMATION FOR SEQ ID NO:3:
(i) ~uu~ C~ARACTERISTICS:
(A) LENGTH~ 34 base pairs (B) TYPE: nucleic acid WO9S/34683 ~ \' L'' ~ 1 9 2 6 6 2 PCT~S94/06543 ~C) sTR~Nn~nN~..q.q single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA
(iii) HYPOTHETICAL: NO : ~ S~
(iv) ANTI-SENSE: NO
(ix) FEATURE:
(A) NAME/KEY: misc. feature (B) LOCATION:
(D) OTHER INFORMATION: standardname = "5' Primer"
(Xi) ~UU~N~ DESCRIPTION: SEQ ID NO:3:
CGGGAATTCA Al~UUl~lAC r~TrrT~r AATG 34 (2) INFORMATION FOR SEQ ID NO:4:
(i) ~I:;UU~;N~; rlT;~RZ~rT~RT.qTICS:
(A) LENGTX: 47 base pair8 (B) TYPE: nucleic acid (C) sTR~Nn~nN~ss sin~le (D) TOPO~OGY: linear (ii) MOLECULE TYPE: 8ynthetic (iii) ~Y~u~ CAL: NO
(iv) ANTI-SENSE: YES
(ix) FEATURE:
(A) NAME/KEY: misc. feature ~ :
(B) LOCATION:
(D) OTHER INFORMATION: 8tandardname = "3' Primer"
(xi) ~Uu~N~ DESCRIPTION: SEQ ID NO:4:
CCCGAAGCTT CCTGCAGTTA ACAGGCGGTG ACGGTA~TTT CTGCAAC 47 (2) INFORMATION FOR SEQ ID NO:5:
(i) SEQUENCE r~R~rT~RTsTIcs:
(A) LENGTH: 33 base pairs W095134683 ,,~,J~ q?662 PCT/US94/06543 (B) TYPE: n~cleic acid ~C) STR~Nll~n~ single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (iii) HYPOTHETICAL: NO
(iv) ANTI-SENSE: NO
(xi) ~u~ DESCRIPTION: SEQ ID NO:5:

W095/34683 ~ 2~ t~3~ 2i ~ 2 6 6 2 PCT~S94/06543 (2) INFORMATION FOR SEQ ID ~0:6:
U~N~ CHARACTERISTICS:
~A) LENGT~: 29 base pairs (B) TYPE: nucleic acid (C) STRA~nRnNR.c.~ 5 ingle (D) TOPOLOGY: linear (ii) MOLECULE TYPE: genomic DNA :
(iii) ~Y~ CAL: NO :
(iv) ANTI-SENSE: NO
(Xi) ~U~N~ DESCRIPTION: SEQ ID NO:6:
CCGCTGCAGC TACATTTCCT TGTCGTTAG = ~ =29 (2) INFORMATION FOR SEQ ID NO:7:
U~N~ CHARACTERISTICS:
(A) LENGTH: 6 amino acids (B) TYPE: amino acid (C) sTR~NnRnNR~ single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (iii) HYPOTHETICAL: YES
(iv) ANTI-SENSE: NO
(xi) ~UU~'N~'h' DESCRIBTION: SEQ ID NO:7:
Ser Phe Cys Phe Gly Gly l 5 (2) INFORMATION FOR SEQ ID NO:8:
( i ) ~UU~N~ CHARACTERISTICS:
(A) LENGTH: 31 base pairs (B) TYPE: nucleic acid (C) sTRANnRnNR~ single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA __ _ WO 95/34683 ~ i i3 ~ ~ ~ 2 1 9 2 6 6 2 PCT/U594~06543 (iii) ~Y~O~ CAL: NO
(iv) ANTI-SENSE: NO
(ix) FEATURE:
(A) NAME/KEY: misc. feature (B) LOCATION:
(D) OTHER INFORMATION: standardname = ~5' Primer"
(Xi ) ~UU~N~ DESCRIPTION: SEQ ID NO:8:

(2) INFORMATION FOR SEQ ID NO:9:
(i) ~Uu~N~ R~T~RT.~TICS:
(A) LENGTH: 30 ~ase pairs (B) TYPE: nucleic acid (C) sTR~Nn~n~s single (D) TOPODOGY: linear (ii) MOLECU~E TYPE: DNA (genomic) (iii) hY~l~llCAL: NO
(iv) ANTI-SENSE: NO
(ix) FEATURE:
(A) NAME/KEY: misc. feature (B) LOCATION:
(D) OTHER INFORMATION: standardname = "3' Primer"
(Xi ~ U~N~ DESCRIPTION: SEQ ID NO:9:

. -45-

Claims (37)

What is claimed is:

1. A protein construct comprising:

(a) a genetically modified gpV protein truncated at its carboxy terminus; and (b) a target molecule peptide bonded to the carboxy terminus of the modified gpV protein.

2. The protein construct of claim 1 wherein the target molecule is a protein selected from the group consisting of an enzyme, enzyme substrate, immunoglobulin, toxin, growth factor, cytokine, hormone, ligand, and receptor.

3. The protein construct of claim 1 further comprising at least an antigenic portion of a second protein to which antibodies have been raised.

4. The protein construct of claim 3 wherein the second protein is a marker protein.

5. The protein construct of claim 4 wherein the third protein is a marker protein selected from the group consisting of chloramphenicol acetyltransferase, alkaline phosphatase, and .beta.-galactosidase.

6. The protein construct of claim 3 wherein the second protein is peptide bonded to the carboxy terminus of the target molecule.

7. A nucleic acid encoding the protein construct of claim 1.

8. A nucleic acid encoding the protein construct of claim 3.

9. A plasmid comprising the nucleic acid of claim 7.

10. A plasmid comprising the nucleic acid of claim 8.

11. An infective lambdoid bacteriophage comprising the protein construct of claim 1, the target molecule being displayed on the outer surface of the bacteriophage.

12. An infective lambdoid bacteriophage comprising the protein construct of claim 3, the target molecule being displayed on the outer surface of the bacteriophage.

13. A method of detecting a molecule-of-interest in a solution comprising the steps of:

(a) providing an infective lambdoid bacteriophage including a protein construct, the protein construct comprising:

(i) a genetically modified gpV protein truncated at its carboxy terminus; and (ii) a target protein peptide bonded to the carboxy terminus of the gpV protein;
(b) processing the target protein such that the bacteriophage is rendered reversibly non-infective;

(c) treating the non-infective bacteriophage with a solution-to-be-tested, the non-infective bacteriophage being rendered infective if the solution contains the molecule-of-interest;

(d) contacting a bacterial cell susceptible to lambdoid bacteriophage infection with the treated bacteriophage for a time sufficient to enable the bacteriophage to infect the cell; and (e) detecting bacteriophage infection of the cell, infection being indicative of the presence of the molecule-of-interest in the solution.

14. The method of claim 13 wherein providing step (a) comprises providing an infective lambdoid bacteriophage having a target molecule comprising a protein selected from the group consisting of an enzyme, enzyme substrate, immunoglobulin, receptor, ligand, growth factor, toxin, cytokine, and hormone.

15. The method of claim 13 wherein providing step (a) comprises providing an infective lambdoid bacteriophage including a protein construct, the construct further comprising a peptide linker that is peptide bonded to the carboxy terminus of the gpV protein and to the amino terminus of the target molecule.

16. The method of claim 13 wherein providing step (a) comprises providing an infective lambdoid bacteriophage including a protein construct, the construct further comprising at least a second protein reactive with antibodies.

17. The method of claim 16 wherein providing step (a) comprises providing an infective lambdoid bacteriophage including a protein construct, the construct comprising at least an antigenic portion of a second protein peptide bonded to the carboxy terminus of the target molecule.

18. The method of claim 13 wherein the providing step (a) comprises:

(i) providing a nucleic acid encoding the protein construct;

(ii) transforming a bacterial cell with the nucleic acid, the cell being pre-infected with a lambdoid bacteriophage assembly mutant having a defective or substantially absent gpV protein;

(iii) inducing the transformed cell to express lambdoid components and to assemble a lambdoid bacteriophage therefrom, the bacteriophage having the target molecule on its outer surface; and (iv) isolating the lambdoid bacteriophage from the cell.

19. The method of claim 13 wherein the providing step (a) comprises:

(i) providing a lambdoid bacteriophage assembly mutant having a defective or substantially absent gpV
protein;

(ii) infecting a bacterial cell with the bacteriophage, the cell being pre-transformed with a nucleic acid encoding the protein construct;

(iii) inducing the infected cell to express lambdoid components and to assemble a lambdoid bacteriophage therefrom having the target molecule on its outer surface; and (iv) isolating the bacteriophage from the cell.

20. The method of claim 13 wherein processing step (b) comprises treating the bacteriophage with a binding molecule that binds the target molecule, the binding of the target molecule rendering the bacteriophage reversibly non-infective.

21. The method of claim 20 wherein processing step (b) comprises treating the bacteriophage with a binding molecule selected from the group consisting of an enzyme, enzyme substrate, immunoglobulin, receptor, ligand, and matrix.

22. The method of claim 21 wherein processing step (b) comprises immobilizing the bacteriophage-linked target molecule to a matrix.

23. The method of claim 13 wherein treating step (c) comprises treating the non-infective bacteriophage with solution-to-be-tested selected from the group consisting of a culture medium, cell lysate, blood, serum, saliva, semen, and lacrimal secretions.

24. The method of claim 13 wherein treating step (c) comprises treating the non-infective bacteriophage with a molecule-of-interest selected from the group consisting of proteins, peptides, hormones, nucleic acids, carbohydrates, lipids, glycoproteins, glycolipids, proteolipids, lipoproteins, lipopolysaccharides, vitamins, toxins, terpenes, antibiotics, and cofactors.

25. The method of claim 13 wherein treating step (c) comprises treating the non-infective bacteriophage with a solution-to-be-tested, the solution-to-be-tested containing a molecule-of-interest which is an enzyme which cleaves the target molecule.

26. The method of claim 13 wherein treating step (c) comprises treating the non-infective bacteriophage with a solution-to-be-tested, the solution-to-be-tested containing a molecule-of-interest selected from the group consisting of an unbound target molecule, and an analog, agonist, and antagonist thereof.

27. The method of claim 13 wherein the target molecule and the molecule-of-interest are the same and are ligands, and the binding molecule is a receptor specific for the ligands.

28. The method of claim 13 wherein the target molecule and the molecule-of-interest are the same and are receptors, and the binding molecule is a ligand that binds the receptors.

29. The method of claim 13 wherein the target molecule and the molecule-of-interest contain the same antigenic determinant, and the binding molecule is an immunoglobulin that binds the determinant.

30. The method of claim 13 wherein the target molecule and the molecule-of-interest are the same and are immunoglobulins, and the binding molecule contains an antigenic determinant to which the immunoglobulins bind.

31. The method of claim 13 wherein detecting step (e) comprises detecting cell death, cell death being indicative of the presence in the solution of the molecule-of-interest which has rendered the bacteriophage infective.

32. The method of claim 13 wherein contacting step (d) comprises infecting an auxotrophic bacterial cell with a temperature sensitive lambdoid bacteriophage at or below about 32°C, the bacteriophage carrying a gene capable of alleviating the auxotrophy; and detecting step (f) comprises detecting bacterial cell survival and growth, survival and growth being indicative of the presence of the molecule-of-interest in the solution.

33 The method of claim 32 wherein contacting step (d) comprises infecting a bacterial cell incapable of sustained growth with an infective lambdoid bacteriophage lambda carrying a gene capable of restoring sustained growth to the cell; and detecting step (f) comprises detecting bacterial cell survival and growth, survival and growth being indicative of the presence of the molecule-of-interest in the solution.

34. A method of selecting a cell expressing a molecule-of-interest comprising the steps of:

(a) providing an infective lambdoid bacteriophage including a protein construct, the protein construct comprising:

(i) a genetically modified gpV protein truncated at its carboxy terminus; and (ii) a target protein peptide bonded to the carboxy terminus of the gpV protein;

(b) processing the target protein such that the bacteriophage is rendered reversibly non-infective;

(c) contacting a bacterial cell with the bacteriophage for a time sufficient to enable the molecule-of-interest produced by the cell to render the non-infective bacteriophage infective and for the inactive bacteriophage to infect the cell; and (d) detecting bacteriophage infection of the cell, infection being indicative of the presence of the molecule-of-interest in the solution.

35. The method of claim 34 wherein providing step (a) comprises providing a temperature sensitive lambdoid bacteriophage that carries a gene required by the cell to be contacted in step (c); contacting step (c) comprises infecting the cell at or below about 32°C, the cell requiring the gene carried by the bacteriophage for survival and growth; and detecting step (d) comprises detecting the growth of the infected bacteria cell, growth being indicative of the molecule-of-interest in the solution.

36. The method of claim 34 wherein the molecule-of-interest and the target molecule are the same.

37. The method of claim 34 wherein providing step (a) comprises providing a temperature sensitive lambdoid bacteriophage that carries a gene required by the cell to be contacted in step (c) for survival, the gene being selected from the group consisting of a gene required for cell biosynthesis, and a gene conferring drug resistance