AU665440B2 - Rabbit single domain antibody and use thereof - Google Patents

Description

RABBIT SINGLE DOMAIN ANTIBODY AND USE THEREOF

BACKGROUND OF THE INVENTION

The advent of hybridoma technology has made it possible to prepare a single species of an antibody [monoclonal antibody, also referred to herein as "mAb"] with the desired specificity to antigen [Kohler, G. and Milstein, C, Nature , 256:52-52 (1975) ] . However, hybridoma cultures typically have a low production capacity and are considered to be unstable.

To circumvent these problems, techniques have been recently developed to bypass hybridomas, thus allowing antibody genes to be cloned directly from lymphocytes of immunized animals and expressed in bacteria. For example, a complete antigen-binding molecule has been expressed and secreted from pro aryotes using a two-phage system. In this system, each phage was able to express VDJ or VL gene segments isolated from mRNA of B cells obtained from immunized mice. The two phage systems can be recombined to co-express VDJ and VL gene segments in E. col± , resulting in a complete antigen-binding molecule [Huse, et al., Science , 246:1275-1281 (1989)]. In another example, mouse immunoglobulin (Ig) heavy chain, single domain antibodies, have been expressed in E. coli were found sufficient to bind antigen. [Ward, et al., Nature , 341:544-546 (1989)].

Antibodies from different species recognize different antigenic epitopes present on one antigen [S. Twining, et al., Biochem . , 191:681-697 (1980)]. Rabbit has been the animal of choice for the production of polyclonal antibodies of high affinity and specificity directed against native or denatured antigens. However, mAb from this species has not been available.

Enzyme-linked immunosorbent assay (ELISA) is a solid phase immuno-assay which employs antibodies and has become a versatile and practicle tool in biology and medicine. [Endwall, et al., Biochem . Biophys . Acta . , 251:427 (1971)]. These assays are commonly used for rapid diagnosis of disease, and are of great commercial importance in both research and human health care. The major principle of these assays is the direct immobilization of the antibody to plastic [Catt and Tregaer, Science , 158:1570-1572 (1967)]. Subsequently, simple washing of the plastic solid phase permits separation of bound from non-bound reactants that are added step wise in the process of completing the assay.

Monoclonal antibodies are commonly used in in vitro diagnosis, such as the sandwich ELISA discussed above. Unfortunately, it has been shown that monoclonal antibodies have lower antigen affinity or antigen capture capacity when absorbed directly on the plastic commonly used as the solid phase in the ELISA assay [Butler, et al.. Molecular JJΠTΠUJI., 23:971-982 (1986)]. One explanation for the apparent low affinity of MoAbs absorbed hydrophobically to plastic could be alteration of the antigen-binding site due to absorbtion-induced denaturation [Suter and Butler, Immun . Lett. , 13:313-316 (1986)]. As a result of the absorbtion- induced loss of antigen capture capacity of capture antibodies used in sandwich ELISAs, the assay loses valuable sensitivity. In order to avoid the direct interaction of the capture antibody with the solid phase, a method has been developed which introduces a biotinylated carrier protein adsorbed on plastic followed by streptavidin to allow the biotinylated MoAb to be firmly bound to the solid phase. This system developed by Suter and Butler is referred to as the protein-avidin-biotin capture (PABC) system. [Suter and Butler, Jnuπun. Lett . , supra ; Suter, et al.. Molecular Immun ., 26:221-230 (1989)]. However, use of the PABC system can add additional complexity and cost to the assay.

Accordingly, there is a desire in the art to produce antibodies having both suitable affinity and specificity, while also having the ability to be bound directly to the solid phase in a sandwich ELISA without losing these characteristics. This would allow the development of diagnostic assays having increased sensitivity while keeping complexity and costs down.

SUMMARY OF THE INVENTION

In accordance with the present invention, there is provided a recombinant rabbit heavy chain variable domain, or single domain antibody having the ability to bind to a solid phase, for example, plastic, without a decrease in antigen capture capacity. The present invention further provides an improved immunological sandwich assay, such as an ELISA, in which the rabbit single domain antibody of the present invention is directly bound to the solid phase as the capture antibody. The resulting assay is less complex than systems currently utilized without any significant decrease in sensitivity.

BRIEF DESCRIPTION OF THE FIGURES

Figure 1 shows a 2% agarose gel comparing PCR amplified VDJ genes from mRNA isolated from rabbit PBL (lane 2) or amplified from a cloned VDJ gene (clone 4K7, lane 3). Lane 1: 123 bp standard and its multimers.

Figure 2 shows the results of an inhibition experiment using colony plaque lifts. The binding of labelled protein C to nitrocellulose bound gene products expressed by clones 17.1 (lanes 1-3) and 21.1 (lanes 4-5) upon inhibition by various molar amounts (B) of native protein C (lanes 2-5) or BSA(A) (lane 1) .

Figure 3 is a graph showing the specificity of antigen binding shown by ELISA. Bacterial supernatant produced by clones 21.1 (A), 17.1 (B) , 16.1 (C) and 13.2 (D) was tested for binding to protein C. The resulting OD is shown as a function of time after induction with IPTG.

DETAILED DESCRIPTION OF THE INVENTION

In accordance with the present invention, there is provided a recombinant rabbit single domain antibody having the ability to bind to a solid phase, for example, plastic, without any significant decrease in antigen capture capacity. More specifically, the present invention directed to a recombinant rabbit single domai antibody, its production and use. The gene encoding, the rearranged gene for the rabbit heavy chain variable domain or single domain antibody may be produced, for instance, by cloning or gene synthesis, and placing it into a suitable expression vector. The expression vector is then used to transform a compatible host cell which is then cultured to allow the single domain antibody to be expressed and, preferably secreted.

One method for preparing such gene segments combines the following elements [Huse, et al., Science , 246:1275-1281 (1989)]:

1. isolating nucleic acids sequences or DNA containing a substantial portion of the rabbit immunological repertoire;

2. preparing polynucleotide primers for cloning polynucleotide segments containing immunoglobulin V_H region genes;

3. preparing a gene library containing a plurality of different V_H genes from the re^pertoire. Thereafter, one can express the V_H polynucleotides in a suitable host; and

4. screen the expressed polypeptide for a preselected activity (e.g., affinity binding to a protein f interest.

Nucleic acid sequence containing a substantial portion of the rabbit immunological gene repertoire may be isolated from a heterogeneous population of antibody-producing cells, i.e., B-lymphocytes (B cells) , preferably rearranged B cells such as those found in the circulation or spleen of a rabbit which has been infected, immunized or partially immunized with the antigen of interest. Such lymphocytes include, for example, peripheral blood lymphocytes (PBLs) .

The immunization can be carried out conventionally. Antibody titer in the animal can be monitored to determine the stage of immunization desired, which stage corresponds to the amount of enrichment or biasing of the repertoire desired. Partially immunized animals typically receive only one immunization and cells are collected therefrom shortly after a response is detected. Fully immunized animals display a peak titer, which is achieved with one or more repeated injections of the antigen into the host mammal, normally at 2 to 3 week intervals. Usually three to five days after the last challenge, the spleen is removed and the genetic repertoire of the spleenocytes, about 90% of which are rearranged B cells, is isolated using standard procedures. [See, Ausubel, et al., "Current Protocols in Molecular Biology", eds., John Wiley & Sons, NY.] Any strain of rabbit can be used depending on the allotype desired.

The rabbit immunoglobulin genes can be isolated from either genomic material containing the gene expressing the variable region or the messenger RNA (mRNA) which represents a transcript of the variable region. When using genomic DNA from other than non-rearranged B-lymphocytes care should be taken in positioning the sequences coding for the variable region because the sequences are separated by introns. This can readily be done by the skilled artisan based upon the present disclosure. The DNA fragment(s) containing the proper exons are isolated, the introns excised and the introns then spliced in proper order and in proper orientation. An alternative technique employing rearranged B cells is preferred because the constant (C) , diversity (D) and joining (J) im unoglobulin gene regions have translocated to become adjacent, so that the sequence is continuous (free of introns) for the entire variable region. Where mRNA is utilized the cells must be lysed under RNAse inhibiting conditions. mRNA may be separated from other RNA by oligo-dT chromatography. A complementary strand of cDNA may then be synthesized on the mRNA template, using reverse transcriptase and a suitable primer, to yield an RNA-DNA heteroduplex [Gubler, et al., Gene , 25:263-269 (1983)]. A second strand of DNA can be made one of several ways, for example, by priming with RNA fragments of the mRNA strand (made by incubating RNA-DNA heteroduplex with RNAse H) , and using DNA polymerase, or by priming with a synthetic oligodeoxynucleotide primer which anneals to the 3' end of the first strand and using DNA polymerase.

The isolation of polyA selected mRNA and first strand cDNA synthesis can also be accomplished using a commercially available kit available from Invitrogen, Madison, Wisconsin.

To amplify the V_H-coding DNA homologs using polymerase chain reaction amplification, two primers must be used for each coding strand of nucleic acid to be amplified. The first primer becomes part of the non-sense, (minus or complementary) strand and hybridizes to a nucleotide sequence conserved among V_H (plus) strands within the repertoire. To produce V_H coding DNA homologs, first primers are therefore chosen to hybridize to (i.e., complementary to) conserved regions within the J region, CHI region, hinge region, CH2 region or CH3 region of the immunoglobulin genes. Primers to the J, CHI and hinge regions are preferred. To produce the V_H- coding DNA homologs, second primers are chosen to hybridize with a conserved nucleotide sequence at the 5¹ end of the V_H-coding immunoglobulin gene, such as in that area coding for the leader or first framework region. One or both of the first and second primers can contain a nucleotide sequence defining an endonuclease restriction site. The site can be heterologous to the immunoglobulin gene being amplified, and typically appear at or near the 5' end of the primer. The use of primers with restriction sites has the advantage that the DNA can be cut with at least one restriction enzyme which leaves 5' or 3' overhanging nucleotides. Such DNAs are more readily cloned into the corresponding sites on the vector than blunt-ended fragments taken directly from the method. The double-stranded cDNA produced at the end of the cycles is readily insertable into a cloning vector by using an appropriate restriction enzyme, which can be chosen empirically based upon the present disclosure. Preferably, the choice of restriction sites is such that the double stranded (ds) cDNA is cloned directly into an expression vector. Preferably, the heavy chain primer pair consists of a V_H primer and a J_H primer, each containing convenient restriction sites for cloning. Using for example, the Kabat database on immunoglobulins [Kabat, et al., "Sequences of

Proteins of Immunological Interest", 5th Ed. (U.S. Department of Health and Human Services) (1991)], one can analyze the amino acid and codon distribution found in rabbits of the desired allotype.

In accordance with the present invention, analysis of nucleotide sequences of VH and JH regions cloned from rabbits of VHal allotype revealed extensive homologies [M. Suter, et al., J. Immunol . , 188:1997-2000 (1990); E. Kabat, et al., "Sequences of Proteins of Immunologic Interest", supra ; L.A. DiPetro, et al., J. Immunol . , 20:1401- 1404 (1990); R. Becker, et al., J. Immunol . , 142:1351-1355 (1989); R.S. Becker, et al. , Eur. J. Immmunol. , 20:397:402 (1990)], which data allowed the synthesis of VHal specific oligonucleotide primers (Table 1) . For example, the 39 base pair 5• V_H primer was designed to be degenerate for two different nucleotides at two positions. Similarly, a 24 base pair J_H region oligonucleotide can be designed for reverse priming at the 3• end of the heavy chain variable gene. The primers can then be tested by PCR using cloned genomic DNA containing VDJ gene segments encoding VHal allotypes [M. Suter, et al. , . Immunol., supra] .

The primers do not need to have a sequence exactly complementary to the target sequence to which it is annealed. Differences can arise for instance because of nucleotide variations or because of the introduction of a restriction enzyme site. One can adjust conditions in the annealing mixture to enable the primers to anneal to the ds nucleic acid by making annealing conditions less stringent.

The DNA polymerase used in this method may be any DNA polymerase known in the art, for example any of those commercially available such as Taq polymerase, or Vent® polymerase (New England Biolabs, Inc.). The conditions to be used for each polymerase are well known. The polymerase reaction will be carried out in the presence of the four nucleoside triphosphates. These and the polymerase enzyme may already be present in the sample or may be provided afresh for each cycle.

Denaturing of the DNA strands may be carried out by any known method, for instance, by heating the sample. When heating is used to control the method, a suitable cycle of heating comprises denaturing at about 95°C for about one minute, annealing at from 30°C to 65°C for about one minute, and primer extension at about 75°C for about two minutes. To insure the elongation and renaturation is complete, the mixture after the final cycle is preferably held at about 72°C for about five minutes when Vent® polymerase is used.

The product double-stranded cDNA may be separated from the mixture, for instance, by gel- electrophoresis using agarose gels. However, if desired, double-stranded cDNA may be used in unpurified form and inserted directly into a suitable ci „ing or expression vector by conventional methods. This will be particularly easy to accomplish if the primers include restriction enzyme recognition sites.

For the construction of a VDJ gene library, the combinatorial bacteriophage λ vector [D. Huse, et al., Science, supra] can be used. The system is commercially available from Stracyte, California, USA. Briefly, amplified DNA is purified using, for example, anionic exchange columns (Quiagen, Kontron, CH) and ethanol precipitated. The purified and ethanol precipitated DNA is digested with the appropriate restriction enzymes at 2-3 fold excess and repurified as described above. For ligation into vector arms of, for example, λHC2 [D. Huse, et al.. Science, supra] , the amount of insert and vector DNA is estimated using a DNA-indicator system (Invitrogen) . Different insert/vector ratios approaching equimolar amounts are test ligated and aliquots packaged using Gigapack Plus II kits (Stratagene) . The resulting libraries are titrated. A large size library can be constructed using the insert/vector ratios which had resulted in the highest titer in the test ligation. The resulting library can be assayed for the expression of allotype specific determinants using allotype specific antisera. Affinity purified and biotinylated antibody [M. Suter, et al., J. Immunol . , supra] can be used to screen expressed VDJ genes bound to nitrocellulos* filters (Schleicher and Schuell, Zurich. CH) after plating (see below) , followed by ¹²⁵I-labelled Streptavidin [M. Suter, et al., Mol . Immunol . , 26:221-230 (1989)]. Briefly, E. coli XL-1 Blue cells are infected with approximately 10 phage particles and plated at a density to obtain 30 - 35 x 10³ pfu/150 mm petri dish. Rescreening is done using biotinylated protein C and ¹²⁵I-labelled Streptavidin [M. Suter, Mol . Immunol . , supra] .

The expression vector pSWl-VHpolyTagl can also be used to express rabbit VDJ genes [Ward, et al., Nature , 381:544-546 (1989)]. To facilitate detection of expressed proteins by this plasmid in solid phase assays, a fusion protein using a single Fc binding portion of staphylococcal protein A (SpA, Pharmacia, Uppsala, S) can be constructed. The coding region of the Fc binding part is amplified by PCR and cloned in frame into the PstI site resulting in plasmid pSWl-VHSpApolyTagl. Fusion proteins expressed by pSWl-VHpSpApolyTagl are detected by ELISA using purified rabbit Ig coated to 96 well microtiter wells as described above to immobilize the unique Fc binding portion of SpA, followed by a suitable detection system.

Furthermore, one can use a system to select for binding of antibody, e.g. , single domain antibodies to a selected antigen, to identify desired antibodies from a starting population of greater than 10⁶ proteins [McCafferty, et al., Nature , 348:552-554 (1990)]. For example, using a system that involves the production of filamentous bacteriophage carrying fusion proteins between sFv and the viral gene III surface protein ("phage antibodies") . The gene III protein is responsible for the attachment of the phage to the F pilus of the target bacterium [Kornberg, "DNA Replication" (Freeman, San Francisco) (1980) ] . However, the gene III protein can tolerate the presence of polypeptides (including sFv fragments of antibodies) inserted near the amino terminus without loss of function [Smith, et al., Science , 228:1315-1317 (1985); Parmley and Smith, Gene , 73:305-318 (1988); Scott and Smith, Science , 249:386-390 (1990); Devlin, et al.. Science , 249:404-406 (1990)]. Thus,, replication-competent phage displaying functional antibody variable regions on their surface can be generated. Phage containing functional antibody fragments with desirable phenotypes can be enriched by multiple rounds of affinity chromatography [McCafferty, et al. , supra] .

Specificity of antigen - binding can be shown by competitive inhibition and by ELISA. The competitive inhibition test is performed as described in D. Huse, et al.. Science, supra . In short, using, for example, single domain antibodies to protein X (the protein of interest) , predetermined amount of biotinylated protein X is inhibited by increasing amounts of native protein X. The remaining amount of biotinylated protein bound to nitrocellulose immobilized single domain antibodies is determined using ¹²⁵I-labelled SA followed by autoradiography. Autoradiography obtained by competitive inhibition tests can be scanned using, for example, an IMCO 1000 V 2.0 image analysis system (Kontron, Munich, FRG) . The system allows to integrate grey values and surface areas of autoradiographs. The values are used as a basis to calculate binding constants of the sDAb [E. Ward, et al., Nature, supra ; D. Huse,.et al., Science, supra] . Phagemids from positive clones can be excised in vivo using phage R408 [Stratagene, La Jolla, California, USA] . Selected clones are grown to OD 600nm = 1 and induced by IPTG (Boeringer, Mannheim, FRG) for various times.

Supernatant containing secreted proteins are directly coated to 96 well Immulon II plates, followed by biotinylated protein X and SA- peroxidase as described above.

VDJ genes encoding proteins with binding activity to the protein of interest can be analyzed by comparing the site of the genes by polyacrylamide gel electrophoresis [T. Maniatis, et al., Molecular Cloning: A Laboratory Manual , Cold Spring Harbor Laboratory, USA (1982)]. The genes can be further analyzed by temperature shift gels (Diagen, Dϋsseldorf, FRG) . This system allows determination of single base pair differences between genes and hence allows identification of duplicate clones [D. Riesner, et al., Electrophoresis , 10:377-389 (1989)].

Rabbit single domain antibodies obtained by recombinant technologies and prokaryotic expression systems in accordance with the present invention produce correctly folded single domain antibodies that can be used in immunological capture assays, for example, sandwich ELISA assays. In this assay, the single domain antibody is first coated to the plastic solid phase. After washing, a sample containing the antigen is incubated with a labeled antibody probe. A detection system is then used to determine the presence or absence of the antigen, as well as its concentration. Alternatively, the single domain antibodies of the present invention could be used for affinity isolation procedures to purify the antigen of interest from a mixture of molecules. The relative low affinity of single domain antibodies may be an advantage in affinity isolation because the release of the antigen from the single domain antibody after washing non- reacted molecules is more effective if low affinity ligand receptor interactions are used.

The following examples of the invention are provided to illustrate the invention and not by way of limitation.

EXAMPLE I

PRODUCTION OF RABBIT SINGLE DOMAIN ANTIBODY SPECIFIC TO PROTEIN C

Immunization of Animals.

Purified human protein C, a serum protease which functions to modulate thrombotic potential by inactivation of factors Villa and Va was used as an antigen. Rabbits homozygous for the VHal allotype [J. Oudin, et al., J. Exp. Med . , 112:107-113 (I960)] were kindly provided by Dr. A. Kelus (Basel Institute for Immunology, Basel, CH) . Approximately one year old rabbits were immunized as described in M. Suter, Vet . Med. , 28:414-420 (1981) . Briefly, each rabbit received 100 μg protein C emulsified in complete Freud's adjuvant injected over both shoulders. The animals were boosted twice in three-week intervals using 50 μg of protein C emulsified in incomplete Freud's adjuvant and given subcutaneously. Ten days after the last injection, rabbits were test bled and sera were assayed for binding to protein C using an enzyme-linked immuno sorbent assay (ELISA) . For this test sera were precipitated with ammonium sulfate [M. Suter, Vet . Med. , supra] and 5 μg/ml of the precipitated protein was bound to 96 well microtiter plates (Im ulon II, Nunc, Denmark) as described in J. Bulter, et al., "Enzyme-Mediated Immunoassay" (T. T. Ngo, H.M. Lenhoff, Eds.) pp.

2451-276, Plenum Press, New York. Various amounts of purified biotinylated protein C [D.R. Gretch, et al., Analyt . Biochem . , 163:270-277 (1985)] were added to the antibody coated wells. Bound protein C was detected by adding Streptavidin (SA) coupled to peroxidase [M. Suter, J. Immunol . Methods , 84:327-341 (1985)], followed by substrate [M. Suter, et al., Immunol . Lett . , 13:313-316 (1986)]. A rabbit was arbitrarily considered successfully immunized if 10 ng ml^"1 of protein C could be detected by ELISA.

Identification of Primers to Amplify VDJ Gene Segments. Analysis of nucleotide sequences of VH and JH regions cloned from rabbits of VHal allotype revealed extensive ho ologies [M. Suter, et al., J . Immunol . , supra , E. Kabat, et al.. Sequences of Proteins of Immunoloqic Interest.4thEd. United States Department of Health and Human Services, Public Health Service,

National Institutes of Health, Bethesda, Maryland, L.A. DiPetro, et al, J. Immunol , 144:1969-1973 (1990); L.A. DiPetro, et al., Eur. J. Immunol , 20: 1401-1404 (1990); R. Becker, et al., J. Immunol , 142:1351-1355 (1989); R.S. Becker, et al., Eur. J. Immunol, 20:397-402 (1990)]. The data allowed the synthesis of VHal specific oligonucleotide primers (Table 1). The primers were tested by PCR using cloned genomic DNA containing VDJ gene segments encoding VHal allotypes [M. Suter, et al, J. Immunol. , supra]. A 350 bp fragment of defined size was amplified and the molecule was expressed in the SpA modified vector pSWl- VHSpApolyTagl [Ward, Nature, supra]. In ELISA, the expressed molecule (clone 4K7) reacted specifically with VHal specific alloantisera as expected but not with VHa2 specific alloantisera used as control. Hence, the primers could be employed to amplify cloned VDJ genes and the same primers were therefore further used to amplify VDJ gene segments from mRNA obtained from B cells of immunized rabbits.

Isolation of mRNA or DNA and Amplification of VDJ Gene Segments From B Cells: Construction of Phage Libraries.

Ten ml of peripheral blood (PBL) was taken from an immunized rabbit three days after i.v. booster of 100 g protein C in physiological saline. The isolation of poly A selected mRNA and first strand synthesis of cDNA by random priming was done using a commercially available kit from Invitrogen (Invitrogen, Madison, Wisconsin, USA). The detailed protocol provided by the manufacturer was followed.

TABLE 1

10 20 30

(SEQ ID NO:l) CAG TCG GTG GAG GAG TCC GGG GGT CGC CTG

c -

G - -A-

C - -G-

— G-

... c-

— G-

CC CTC GAG ATG C/G A/G VHal

(SEQ ID NO:2) Primer

GGA GAC GGT GAC TAG TGT GCC CCA JH

(SEQ ID NO:3) Primer

Fifty ng of the RNA-cDNA hybrid was amplified in 30 cycles by the polymerase chain reaction (PCR) using GeneAmp amplification reagents (Perking Elmer Cetus, Norwalk, Connecticut, USA) and one unit Taq polymerase (Perking Elmer Cetus) in a thermalcycler (Techne PHC-2,

Brouwer, Luzern, CH). Oligonucleotide primers (100 pmoles) used for the PCR were specific for the framework region I of VH (5* primer) and specific for the JH region (3' primer). The primers are set out in Table 1. The condition for one cycle were: Melting, 94C, 30 s; annealing, 55C, 30 s; extension, 72C for 10 minutes. An aliquot of the reaction mixture was run on a 2% agarose gel. Analysis of the gel indicated that a broad band of approximately 350 bp was amplified (Figure 1).

^"~" Amplified cDNA was cut with Xhόl and Spel and cloned uni- directionally in equimolar ratios into 500 ng of the heavy chain phage vector HC2. A library of 6x105 recombinant phages was obtained of which less than 1 % was background as judged by packaging vector only. Approximately 2x103 pfu were induced by overlaying with filters soaked with IPTG and one half of the nitrocellulose filter bound products were reacted with biotinylated anti-VHal specific allotypic antibodies. The other half of the filter was reacted with biotinylated anti-VHa2 specific allotypic antibodies, respectively. Approximately 30% of the cloned and expressed VH molecules were positive on autoradiography when the blots were reacted with 1251 SA using VHal specific allotypic antisera and less than 1 % when VHa2 specific allotypic antisera were used.

Screening of the Library with!25I Protein C.

The library shown to contain VHal encoding molecules and hence a single domain antibodies of rabbit VDJ origin was screened for binding to the antigen protein C. Approximately 0.5x106 pfu of the amplified library were induced by IPTG and the expressed proteins bound to nitrocellulose filters. Twenty-five filters were hybridized with 100 g of 1251-labelled protein C (specific activity 200 cpm/ng). Thirty protein C reactive clones were found upon autoradiography. Ten clones were rescreened using biotinylated protein C and 1251 SA. Three positive clones (12.1, 17.1 and 16.1) and one negative clone (13.2) were further analyzed by inhibition tests and ELISA.

Specificity of Antigen Binding. Phage from the protein C reactive clones was spotted directly on a bacterial lawn (approximately 200 particles per spot). The plate was then overlaid with an IPTG-soaked filter and incubated for 20 hours at 25C. After blocking, the filters were individually incubated with 10 ng/ml of biotinylated protein C and various amounts of native protein C or bovine serum albumin (BSA) as a control. After two hours incubation and washing, the filter strips were incubated with 1251 SA, followed by autoradiography. The data (Figure 2) indicated specific binding to protein C but not to BSA. Based on this assay, the binding constant was calculated to be 106 to 107 M-l [D. Huse, Science, supra],

Analysis of Single Domain Antibodies by ELISA.

Phagemids of phage from the clones 21.1, 17.1, 16.1 and 13.2 were excised in vivo and their gene product analyzed by ELISA. Single colonies containing excised plasmids from each phage (21.1, 17.1, 16.1 and 13.2) were grown in liquid cultures, induced with IPTG and culture supernatants containing secreted proteins were bound directly to 96 well plates. The washed plates were incubated with 100 ng ml-1 of biotinylated protein C followed by SA - peroxidase. The substrate conversion is shown as a function of induction time (Figure 3). The results confirm the inhibition experiments indicating specific binding of clones 21.1, 17.1,

16.1 to protein C with exception of the negative clone 13.2.

Analysis of VDJ Genes Encoding sDAb.

The in vivo excised VDJ genes encoding sDAb were sized on polyacrylamide gels and all clones were estimated to be approximately 350 bp long. All clones were slightly different in size when compared to each other and to the previously cloned VDJ gene 4K7 obtained from a transgenic rabbit [M. Suter, et al., J. Immunol supra]. The absence of identical clones was further examined by temperature shift gels confirming the isolation of separate clones from the library. EXAMPLE π

SANDWISH ELISA USING RABBIT SINGLE DOMAIN ANTIBODY TO PROTEIN C

100 1 of 0.1 M carbonate buffer (pH 9.6) containing 5 g/ml of the purified rabbit single domain antibody obtained in Example I is added to each well of a 96-well polystyrene microtiter plate. The plate is then incubated at 4C for 12 hours. After incubation, the plate is washed with phosphate buffered saline (PBS) including 0.5% Tween 20 (washing buffer).

To each well of the prepared plate is added the sample to be tested for the level of protein C in a PBS containing 0.5 % Tween 20 and 1 % dry non-fat milk powder (dilution buffer). The plate is then allowed to stand at room temperature for 1-2 hours. The plate is washed with wash buffer and 100 1 of a phosphatase-labeled polyclonal or monoclonal antibodies specific to protein C diluted 1/1000 in dilution buffer is added to each plate. The plate is then again kept at room temperature for 1-2 hours. After washing with wash buffer, the enzyme activity of the bound phosphatase-labeled antibody is assayed by adding the appropriate substrate to each well, followed by spectrophotometric determination of the substrate conversion, after one hour. Regression analysis comparing the values obtained from a standard to the unknown allows quantitation of protein C in the sample.

SEQUENCE LISTING

(1) GENERAL INFORMATION:

(i) APPLICANT: AEBY, PIERRE BLASER, KURT CRAMERI, RETO SUTER, MARK

(ii) TITLE OF INVENTION: RABBIT SINGLE DOMAIN ANTIBODY AND USE THEREOF

(iii) NUMBER OF SEQUENCES: 3

(iv) CORRESPONDENCE ADDRESS:

(A) ADDRESSEE: LOUISE S. PEARSON; BAXTER DIAGNOSTICS, INC.

(B) STREET: ONE BAXTER PARKWAY

(C) CITY: DEERFIELD

(D) STATE: ILLINOIS

(E) COUNTRY: US

(F) ZIP: 60015

(v) COMPUTER READABLE FORM:

(A) MEDIUM TYPE: Floppy disk

(B) COMPUTER: IBM PC compatible

(C) OPERATING SYSTEM: PC-DOS/MS-DOS

(D) SOFTWARE: Patentin Release #1.0, Version #1.25

(vi) CURRENT APPLICATION DATA:

(A) APPLICATION NUMBER: US 07/894028

(B) FILING DATE: 05-JUN-1992

(C) CLASSIFICATION:

(viii) ATTORNEY/AGENT INFORMATION:

(A) NAME: RESNICK, DAVID S.

(B) REGISTRATION NUMBER: 34235

(C) REFERENCE/DOCKET NUMBER: DA-4325

(ix) TELECOMMUNICATION INFORMATION:

(A) TELEPHONE: (617) 523-3400

(B) TELEFAX: (617) 523-6440

(C) TELEX: 200291 STRE UR (2) INFORMATION FOR SEQ ID NO: 1:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 30 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: unknown

(D) TOPOLOGY: unknown

(xi) SEQUENCE DESCRIPTION: SEQ ID NQ:1:

CAGTCGGTGG AGGAGTCCGG GGGTCGCCTG 30

(2) INFORMATION FOR SEQ ID NO:2:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 37 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: unknown

(D) TOPOLOGY: unknown

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:2: CCCTCGAGAT GCAGTCGSGA GGAGTCCGGG RCGCCTG 37

(2) INFORMATION FOR SEQ ID NO:3:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 24 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: unknown

(D) TOPOLOGY: unknown

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:3:

GGAGACGGTG ACTAGTGTGC CCCA 24

Claims (7)

What is claimed is:

1. A rabbit single domain antibody comprising at least part of a rabbit immunoglobulin heavy chain variable domain.

2. The rabbit single domain antibody of claim 1, wherein said antibody is capable of binding to a solid phase with no decrease in antigen capture capacity.

3. The rabbit single domain antibody of claim 2, wherein said solid phase is plastic.

4. A rabbit single domain antibody of claim 1, wherein the antibody has a binding constant of 106 to 107 M-l.

5. A method for assaying an antigen in a test solution which comprises:

(i) contacting the test solution with a rabbit single domain antibody having binding specificity to the antigen, said antibody bound directly on a solid phase;

(ii) contacting a labeled antibody having binding specificity to the antigen with the reaction of step (i); and

(iii) measuring activity of a labeling agent on the bound antibody.

6. The method of claim 5, wherein the antigen is protein C.

7. The method of claim 5, wherein the solid phase is plastic.