AU1855999A - Anti-EGFR single-chain Fvs and anti-EGFR antibodies - Google Patents

Description

OUr ReCf: 720495 10 II Regzulation 3:2) At7~STRAUiA

ORIGINAL

COMPLETE SPECI FICATI ON STANDARD

PATENT

Applicant~s): Address for Service: invention Title: Merck Patent GmbH- Frankfurter Strasse 250 D-64293 Darmstadt

GERMANY

DAVIES COLLI SON CAVE Patent Trade Mark Attorneys Level 10, !0 Barrack Street SYDNEY NSW 2000 Anti-EGFR single-chain Fvs and anti-U jFR anti bodies The following statement is a full description of this invention. includin-u the best method of performing it known to me:- Anti-EGFR Single-Chain Fvs and Anti-EGFR Antibodies TECHNICAL FIELD OF THE INVENTION This invention relates to new anti-EGFR antibodies and antibody fragments, preferrably single-chain Fvs (scFvs) which can be obtained from phage-antibody libraries constructed from cells of an immunized 15 mammalian, preferably a mouse. The antibody fragments isolated from the phage-antibody libraries can be engineered to create partially humanized whole antibody molecules. These chimeric anti-EGFR antibodies contain constant regions of human immunoglobulins, and 20 can be.used as well as their fragments as agents for the diagnosis and therapy of human tumors.

Furthermore, the invention demonstrates that phage-antibody libraries are an alternative, and more versatile, method for isolating antibodies from immur,;-ed mammalians in comparison with the standard hybridoma technology.

I

The invention relates, moreover, to pharmaceutical compositions comprising said antibodies or fragments for the purposes of treating tumors like melanoma, giioma or carcinoma. The said antibodies or fragments can be used also for diagnostic applications regarding locating and assessing the said tumors in vitro or in vivo.

,j The specification relates to several technical terms which are herewith defined as follows: "FRs" (framework regions) mean the four subregions of the light or heavy chain variable regions that support the three CDRs.

"CDRs" (complementarity determining regions) mean the three subregions of the light or heavy chain variable regions which have hypervariable sequences and form loop structures that are primarily 10 responsible for making direct contact with antigen.

"Chimeric" or partially humanized antibodies mean antibodies comprising constant regions deriving from human sources and variable ,regions (CDRs included) deriving from non-human sources, e.g. from the mouse.

"Humanized" or fully humanized antibodies mean antibodies comprising constant regions and FRs deriving from human sources whereas the CDRs derive from non-human sources.

"EGF" and "EGFR" mean the epidermal growth factor ant its receptor.

"PCR" means the polymerase chain reaction.

"scFv" means single-chain Fv which is an antibody fragment- "V means light chain variable region.

"Vk" means kappa light chain variable region.

"V means heavy chain variable region.

PB

E means phosphate buffered saline FCS means fetal calf serum HBSS means Hanks balanced salt solution FITC means fluoresceineisothiocyanate MTC means mixed cell culture -3- BACKROUND OF THE INVENTION Epidermal growth factor (EGF) is a polypeptide hormone which is mitogenic for epidermal and epithelial cells. When EGF interacts with sensitive cells, it binds to membrane receptors (EGFR). The EGFR is a transmembrane glycoprotein of about 170 kD and is a gene pro-duct of the c-erb-B proto-oncogene.

MAb 425 is a murine monoclonal antibody raised against the well known human A431 carcinoma cell line (ATCC CRL 1555), binds to a 15 polypeptide epitope of the extemal domain of the human EGFR, and inhibits the binding of EGF. MAb 425 (ATCC HB 9629) was found to mediate tumor cytotoxicity in vitro and to suppress tumor cell growth of S'epidermoid and colorectal carcinoma-derived cell lines in vitro (Rodeck et al., Cancer Res. 1987. 47: 3692). Humanized and chimeric versions of MAb 425 have been disclosed in WO 92/15683.

Over the last few years, methods have been described (Skerra and Plickthun, Science 1988. 240: 1038; Better et al., Science 1988. 240: 1041) with which functional antibody fragments can be produced in eukaryotic host cells, such as E. coli. These include the Fv fragment and the Fab fragment, whereby the Fv fragment is of special interest.

Single-chai,, Fvs (wherein the VL and the VHchain are linked together) have been also described (Bird et al., Science 1988. 242: 423; Huston et al., Proc. Natl. Acad. Sci. USA 1988. 85: 5879).

Phage-antibody libraries offer an alternative technology to hybridoma technology in the isolation of antibodies from immunized animals.

Hybridoma technology works by immortalizing the cells that produce the antibodies. Phage-antibody technology works by immortalizing the genes that code for the antibodies (Winter, G. and Milstein, Nature 1991. 349: 293). In phage-antibody technology, the antibody heavy S chain variable region and light chain variable region (VL) genes are PCR-amplified, the variable regions are randomly combined and expressed as antibody fragments on the surface of phage particles, and libraries of phage antibodies are screened for antibodies that bind 0 to antigens of interest.

4 S Hybridoma technology has been very successful at isolating mouse monoclonal antibodies when it has been possible to generate a strong 15 immune response in the spleens of the animals. For example, mouse MAbs against human epidermal growth factor receptor (EGFR) have been isolated from the spleens of mice immunized intraperitoneally with human A431 tumor cells (Murthy et al., Arch. Biochem. Biophys.

1987. 252: 549). The potential advantage of phage-antibody technology over hybridoma technology is that virtually any source of antibody-expressing cells can be used as starting material and that large numbers of different antibodies can be rapidly screened. Another advantage of the phage-antibody technology is that the genes coding for the variable regions of the antibodies of interest have already been cloned and are immediately available for further genetic engineering.

In one report, an anti-tetanus toxoid Fab fragment isolated from a phage-antibody library was converted into a whole antibody molecule (Bender et al., Hum. Antibod. Hybridomas 1993. 4: 74).

During the last ten years, in vitro immunization has been used as an alternative technique to active immunization to generate monoclonal antibodies (mAbs) against a wide variety of antigens from both human and murine systems Vaux, Helenius, A. and Mellman, I.; Nature, 1988. 336: 36; Gathuru, J.K et al.; J. Immunol. Methods, 1991.

137: 95; Borrebaeck, Immunol. Today, 1988.9:355). Advantages of this approach are that only small amounts of antigen are required and that the method is applicable for generating human hybridomas.

However, the generation of poor affinity IgM antibodies and the difficulty of immortalizing human lymphocytes after in vitro immunization have become persistent problems associated with this technology.

10 A new way of obtaining antibodies is by PCR amplification of repertoires of heavy (VH) and light (VL) chain variable region genes which are then randomly recombined and expressed as phage display libraries 15 Antibody variable-region genes were cloned and fused to the minor coat protein (gene 3) as a single chain Fv fragment (scFv) The phage particle displays on its suface the antibody fragment and can be selected by panning using the antibody's binding properties. This technology has 2 the advantage that random recombination of V genes may produce novel pairings with new specificities and affinities which could not be selected by natural processes. Moreover, such an approach makes possible the use of naive or in vitro immunized lymphocytes from murine or human sources.

Previous attempts to obtain mAbs against EGFR by murine B cells in vitro immunization and hybridoma technology rendered low-affinity, crossreacting antibodies. in order to overcome such handicaps, the combination of in vitro immunization followed by PCR cloning technology was carried out.

Therefore, it is an object of the invention to develop antibodies and antibody fragments which have a high affinity to the EGF-receptor and antibody fragments which have a high affinity to the EGF-receptor and -6which can be obtained by the advantageous procedure described above and below.

SUMMARY OF THE INVENTION 10 This invention compares mouse anti-EGFR antibodies isolated from three different phage-antibody libraries with a mouse MAb (425) isolated by standard hybridoma technology (Murthy et al., Arch.

Biochem. Biophys. 1987. 252: 549; Kettleborough et al., Protein Eng.

15 1991. 4: 773). Libraries were prepared, not only from the spleen of an immunized mouse, but also from the draining lymph node of an immunized mouse and from in vitro immunized mouse cells. Two of the single-chain Fvs (scFvs) that were isolated from the libraries were engineered to create chimeric whole antibody molecules with the mouse variable regions joined to human constant regions.

In detail, the present invention relates to an anti-EGFR single-chain Fv obtainable from phage-antibody libraries constructed from cells, preferrably of the spleen or the draining lymph node of an immunized mammalian, preferably a mouse, or from in vitro immunized cells. In principal, the invention is not restricted to scFvs but extends also to other anti-EGFR antibody fragments such as Fab or F(ab')2.

Some of the scFvs according to the invention have well-defined DNA and amino acid sequences. Therefore, it is another object of the invention to provide a single-chain Fv fragment, wherein the varia invention to provide a single-chain Fv fragment, wherein the variable 1 1 I I .t it, -7regions of the heavy and light chain comprise a DNA and/or a amino acid sequence selected from one of the heavy and light chain sequences given in Sequence Id. Nos. 1 32, preferably in Fig. 5 8.

Because in many cases ony completely functioning, whole antibodies can be used for diagnostic and therapeutic purposes it is in the interest of the invention to link the variable regions from single-chain Fvs with the constant regions of human immunoglobulins forming whole, partially or humanized anti-EGFR antibodies.

Therefore, it is an object of this invention to provide a whole anti-EGFR antibody constructed from DNA sequences deriving from antibody fragments as defined above, below or as defined in the claims, and from DNA sequences deriving from constant regions of human immunogobulins, wherein, as a preferred embodiment, the heavy chain comprises the amino acid sequence of a gamma-1 chain, and the light S 20 chain comprises the amino acid sequence of a kappa chain.

According to the present invention the anti-EGFR scFvs, are isolated by using the phage-antibody library technology. Therefore the invention relates to a process for the preparation of an anti-EGFR single-chain Fv comprising the following steps: isolating RNA from immunized mammalian cells, preferably S.mouse cells, S 0 (ii) synthesizing first-strand cDNA, (iii) amplifying the VH and Vk genes in cDNAs from the immunized cells, (iv) cloning said genes together with suitable restriction sites into a phagemid vector, transforming prokaryotic cells with the ligation mixtures, (vi) screening the phage libraries for phage antibodies directed to EGFR using purified EGFR, and (vii) producing the desired single-chain Fv in prokaryotic host ceils, preferably E. coli.

Additionally, it is an object of the invention to disclose a process for the preparation of a whole anti-EGFR antibody by cloning the DNA coding for the variable regions of anti-EGFR antibody fragments produced as indicated above or as defined in the claims into at least one eukaryotic expression vector containing genomic DNA which codes for the constant regions of human immunoglobulins, transforming eukaryotic cells with said vector(s) and expressing and isolating the antibody.

The anti-EGFR scFvs, and above all, the whole anti-EGFR antibodies can be used in diagnosis and therapy of human tumors. Thus, the 20 i invention relates to a pharmaceutical composition comprising an anti- EGFR single chain Fv or a whole anti-EGFR antibody as defined above or as defined in the claims.

The results and advantages of the present invention can be summarized as follows: Novel mouse anti-EGFR antibodies were isolated from phage-antibody libraries. The new antibodies represented at least four different vH subgroups and four different VKsubgroups (Kabat et al., Sequences of proteins of immunological interest. 5th Eds., U.S. Dept. of Health and Human Services, Bethesda 1991). They showed different pairings and sequences from those used by a mouse MAb isolated using hybridoma technology. Mouse 425 MAb has a vH2b and Vk4 pairing that was not observed in the phage antibodies. The VH of scFv L3 11D had the highest percent identity to 425 w The majority of the differences were in the CDRs. The Vk of scFv S4 2D had the highest percent identity to 425 Vk Again the majority of differences were in the CDRs, particularly CDR3. In this invention, a variety of new anti-EGFR antibodies were isolated from the phage-antibody libraries and these antibodies all differ from 425 MAb with at least two of the scFvs recognizing a different epitope on EGFR from that recognized by 425 MAb. This is in contrast to a previous report where the antibodies isolated from combinatorial libraries were reported to be very similar to those isolated by hybridoma technology (Caton and Koprowski, Proc.

Natl. Acad. Sci. USA 1990. 87: 6450).

Of the three phage-antibody libraries, the best library in terms of the *e number of selection steps required to get high affinity antibodies and in terms of the diversity of high affinity antibodies isolated was the library 20 generated from the draining lymph node. Lymph nodes were selected as a source of RNA for the construction of phage-antibody libraries for two reasons. First, previous work had demonstrated that a higher proportion of B cells producing high affinity IgG antibodies was obtained from the popliteal lymph nodes following immunization via the footpad than from spleens following immunization via the peritoneum (Venn and Dresser, J. Immunol. Methods 1987. 102: 95). Second, draining lymph nodes are considered to be a good source for the 30. isolation of human anti-tumor antibodies. Thus, the isolation of mouse anti-EGFR antibodies from the popliteal lymph node of a footpadimmunized mouse was a model for the isolation of human anti-EGFR antibodies from the axillary lymph nodes of a breast cancer patient.

The feasibility of preparing good size libraries from small amounts of li I I 10 lymph node material and of then isolating high affinity antibodies from the libraries was demonstrated.

Although mouse anti-EGFR antibodies were isolated from all three phage-antibody libraries, it is not clear that any of the newly isolated antibodies have higher affinities than the mouse 425 MAb isolated using hybridoma technology. In the first analyses, the phage-antibody derived scFvs appeared to bind to EGFR better than the scFv constructed from 425 MAb (Fig. In other experiments with chimeric whole antibody molecules, one of the chimeric antibodies (S4 2D) showed an affinity for EGFR that equalled that of the chimeric 425 antibody. The second chimeric antibody (L3 11 D) had an affinity that was four-fold lower than that of chimeric 425 antibody (Fig. Binding data obtained using scFvs was misleading probably because preparations of scFvs can contain mixtures of monomers and dimers (Griffiths et al., EMBO J. 1993. 12: 725). In contrast, chimeric IgG 20 antibodies are not expected to form dimers and the chimeric L3 11D and S4 ?D antibodies were demonstrated to be the size expected for bivalent, monomeric chimeric IgG antibodies. Analysis of affinitypurified preparations of 425, L3 11D, and S4 2D scFvs, however, showed that these preparations of scFvs did contain monomeric, dimeric, and other multimeric forms. In addition, the relative proportions of monomeric and multimeric forms varied for each scFv.

The 425 scFv had the lowest percentage of dimeric forms. As predicted, the dimeric and particularly the larger multimeric forms showed stronger binding to purified EGFR than the monomeric form. It appears that 425 scFv has a weaker tendency to dimerize than some of the newly isolated scFvs.

Although the expression of antibody fragments on the surface of phage particles forms the basis of a powerful method for rapidly selecting for

-II

-11antibodies with the desired specificities, neither phage antibodies nor the antibody fragments themselves (scFvs or Fabs) are likely to be the desired end product. Further it is demonstrated how the mouse scFvs isolated from phage libraries can be readily converted into whole antibody molecules. In this case, the mouse variable regions were joined to human constant regions to create partially humanized chimeric antibodies.

These results show that it is possible to use phage-antibody technology to isolate a variety of anti-EGFR antibody fragments from immunized mice. Whole antibody molecules with the desired constant regions can then be constructed from the antibody fragments. In some cases, hybridoma technology may still be the method of choice for isolating monoclonal antibodies from mice. If a highly immunogenic S.i. antigen is available and if a few hybridoma cell lines producing one or a few different anti-antigen antibodies are adequate, then there is probably little reason to consider phage-antibody technology. If, however, special immunization protocols such as footpad injections would be advantageous in generating high affinity antibodies, or if a Slarge number of antibodies against a variety of epitopes on the antigen Sare required, or if antibodies directed against a very discreet, and io". 25 possibly less immunogenic, epitope are required, then phage-antibody technology may be the method of choice. Also, if further genetic engineering of the antibodies is anticipated, then the phage-antibody technology is advantageous in that the antibody genes have already been cloned.

The present approach of combining in vitro immunization with a particulate antigen and PCR-cloning technology has generated scFv *4 35 fragments which reacted with EGFR and did not cross-react with other antigens. The immunization protocol reported here depends on the 12antigen presentation, which is not soluble but is a membrane vesicle preparation, and on the culture medium itself, which is devoid of FCS.

Both methodologies have been reported as a means of increasing the efficiency of in vitro immunization by making the antigen available to the antigen-presenting cells Brams, P. et al.; J. Immunol. Methods, 1987.

98:11).

The results obtained with MTC are in agree with previous papers g.

Borrebaeck, C.A.K and Mkler, SA; J. Immunol., 1986. 136: 3710; .Moller, S.A. and Borrebaeck CAK, in Borrebaeck, CAK In Vitro Immunization in Hybridoma Technology, Elsevier Science Pubishers B.V., Amsterdam 1988, p. whch propose the use of MTC supematants as a source of lymphokines for improving the in vitro immunization process.

The membrane vesicle preparation should be envisaged as a polyantigen since many different antigenic determinants are present in such S. vesicles. For this reason, it would appear that they induce a certain level S of polyclonal activation. We have ruled this out because the anti-EGFR S specific response was clearly different from the response obtained after a standard polyconal activator.

Instead of immortalizing the B-cells after in vitro immunizations, we have 2 used the molecular strategy of immortalizing the antibody VH and VL genes. These monoclonal antibody fragments were expressed and produced in bacteria. The phage display system is a powerful method to isolate antibody fragments against specific antigens. The presence of a 3. .a 3 stop codon between the antibody fragment and the g3p coat protein permits the switch between surface display and secretion as a soluble scFv fragment using suppressor or iron-suppresor strains (Hoogenboom et al., Nucl. Acids Res. 1991. 19: 4133).

Due to the increase of specific response and mRNA levels in in vitro antigen stimulated B-cells, in vitro immunization contributes to the Irp M tl i(i il )l 13 isolation of antibody fragments with high specificities to the antigen. After two rounds of selection, 100% of the clones were positive for binding EGFR. In contrast, clones derived from in vivo immunization processes were 100 positive only after four rounds of selection (Kettleborough, et al., EP 94104160 and Eur. J. Immunol. 1994. 24: 952).

The use of phage display libraries from naive antibody genes might allow specific human antibody fragments to be made without immunization or after in vitro immunization. Antibody fragments can be directly produced in bacteria, thus in a simple, fast and economic way.

BIOLOGICAL MATERIALS AND GENERAL METHODS Microorganisms, cell lines, plasmids, phagemids, promoters, resistance markers, replication origins or other fragments of vectors which are mentioned in this application are commercially or otherwise generally available. Provided that no other information in the application is given, they are used only as examples and are not essential according to the invention and can be replaced by other suitable tools and biological materials, respectively.

Bacterial hosts are preferably used for cloning the scFvs and for producing the scFv proteins. Examples for these hosts are: E. coli or .C Bacillus.

Eukaryotic hosts like COS, CHO or yeasts, for example, are preferred in order to produce the whole anti-EGFR-antibodies according to the invention.

.1l1 I i til III it -14- The techniques which are essential according to the invention are described in detail in the specification. Other techniques which are not described in detail correspond to known standard methods which are well known to a person skilled in the art, or are described more in detail in the cited references and patent applications and in the standard literature.

Brief Description of the Figures: Figure 1. Amino acid sequences of scFvs isolated from phageantibody libraries. scFvs from the lymph node library. scFvs from the spleen library. Complementarity determining regions (CDRs) and framework regions (FRs) are indicated.

Figure 2. Binding of scFvs to EGFR. The concentrations of scFvs S 20 in bacterial supematants were estimated and the scFvs tested by ELISA for binding to purified EGFR. scFvs from the lymph node library. scFvs from the spleen library. P1 (positive control) is the scFv derived from MAb 425. L1 and S1 (negative controls) are nonbinding scFvs from the pre-selected lymph node and spleen libraries.

Figure 3. Intermediate vectors used to reconstruct the variable regions for expression in mammalian cells. VH vector. VK A vector.

Figure 4. Binding of chimeric whole antibodies to EGFR. The concentrations of antibodies in COS cell supematants were determined by ELISA and the antibodies tested by ELISA for binding to purified EGFR.

IJ Figure 5.DNA and amino acid sequence of scFv No. L2 11 C.

Light chain; Heavy chain- Amino acid positions: FR-i: 1 -23, CDR-1: 24 -34, FR-2: 35 49, CDR-2: 50 56, FR-3: 57 88, CDR-3: 89 S7, FR-4: 98 109.

FR-i: 1 30, CDR-1: 31 FR-2: 36 49, CDR-2: 50 66, 67 98, CDR-3: 99 108, FR-4: 109 -119.

Figure 6.DNA and amino acid sequence of scFv No. L2 12B.

Light chain; Heavy chain.

Amino acid positions:- :20 FR-i:- 1 -23, CDR-1: 24 -38, FR-2:- 39 49, CDR-2: 50 56, FR-3: 57 -88, CDR-3: 89 97, FR-4: 98 -109.

FR-i: 1 30, CDR-l: 31 FR-2: 36 49, CDR-2: 50 66, FR-3- 67 98, CDR-3: 99 108, FR-4: 109 -119.

Figure 7.DNA and aminio acid sequence of scFv No. L3 11iD.

Light chain;, Heavy chain.

The amino acid posititions of the FRs and CDRs correspond to those given in Fig. 6- 16- Figure 8. DNA and amino acid sequence of scFv No S4 2D Light chain; Heavy chain.

Amino acid positions: FR-1: 1- 23, CDR-1: 24-35, FR-2: 38- 50, CDR-2: 51-57, FR-3: 58- 89, CDR-3: 90 98, FR-4: 99 110 FR-1: 1-30, CDR-1: 31-35, FR-2: 36- 49, CDR-2: 50-66, FR-3: 67- 98, CDR-3: 99- 107, 5 FR-4: 108 118.

The sequences of Figures 5 8 are also given in the attached Sequence Listing which is part of the disclosure of this invention.

DETAILED DESCRIPTION OF THE INVENTION (1)Construction and screening of phage-antibody libraries Three phage-antibody libraries were constructed, one from the spleen of a mouse immunized with human carcinoma cell line A431 (8.8 x 105 members), one from the popliteal lymph node of a mouse irrmnunized in the footpad with purified EGFR (6.5 x 106 members), and one from mouse lymphocytes immunized in vitro with A431 vesicles (1.1 x 10 members), (details of construction of A431 vesicles and in vitroimmunization are given in Examples 1, Prior to selection, at least 46 clones from each library were analyzed by BstNI fingerprinting -17- (Clackson et al, Nature 1991. 352: 624) to determine the diversity of the repertoires. A wide range of digestion patterns was observed. Also prior to selection, scFvs from 96 clones from each library were tested by ELISA for binding to EGFR. None of the scFvs from the spleen and S 5 lymph node library bound to EGFR. One of the scFvs from the in vitro immrunized library bound to EGFR. After one round of selection using EGFR-coated immunotubes, a clear enrichment for EGFR-binding scFvs was observed with the lymph node library and with the in vitro 10 immunized library. A second round of selection was needed before any EGFR-binding scFvs were detected from the spleen library. By the third round of selection, the majority of the scFvs from the lymph node and in vitro immunized libraries were positive for binding to EGFR.

S* 1 5 After a fourth round of selection with the spleen library, the majority of the scFvs were positive for binding to EGFR (Table 1).

1-1 I In I I" I 18- Percent of EGFR-binding clones after each round of Table 1.

selection.

ii o e r '4 o i o o i i o Lymph Splee In vitro Node n Immuniz Library Libra ed ry Cells Library Pre- 0 0 1 selection First 77 0 84 round Second 86 26 100 round Third 90 77 100 round Fourth not 97 not round tested tested Sequence analysis of EGFR-binding clones After each round of selection, scFv inserts from EGFR-binding clones were analyzed by BstNI fingerprinting (Clackson et al, Nature 1991.

352: 624). It became clear that there was an enrichment for certain digestion patterns. Clones with different BstNI fingerprints were chosen from the second and third rounds of selection of the lymph node library and from the third and fourth rounds of the spleen library for DNA sequencing of the VHs and VKs. Clones from later rounds of selection 19were analyzed because higher affinity antibodies were expected to be in the later rounds (Clackson et al, Nature 1991. 352: 624).

Sixteen clones from the lymph node library were sequenced and six different scFvs were obtained (Figure Five of these were pairings of unique VHS and VKS. The sixth was a variation of a previously occurring VH with six amino acid changes, five of which were in framework region (FR) 1. Two of these changes can be attributed to the use of the degenerate VH1BACKSFI primer (Hoogenboom et al., Nucl. Acids Res. 1991. 19:4133). The others may be a result of PCR errors. The VHS were classified into two subgroups, VH2b and VH3d, while the VKS fell into four subgroups, V, 3

VK

4

V,

5 and V,, 6 (Kabat 15 et al., Sequences of proteins of immunological interest. 5th Eds., U.S.

Dept. of Health and Human Services, Bethesda 1991). Ten individual clones from the spleen library were sequenced and four different scFvs were found. Three of these were pairings of unique VHS and VKS while 20. the fourth was similar to one of the previous pairings with only two 20 amino acid differences in VH, one of which occurred in complementarity determining region (CDR) 2, and two amino acid Sdifferences in VK. Classification into subgroups revealed VHS from subgroups VH2a, VH2c, and VH3d and VKs from subgroups V, 3 and

VK

4 Comparison of the scFvs obtained from the lymph node and spleen libraries revealed only one scFv that was common to both libraries, scFv L3 10AlscFv S4 10H (Figure This clone appeared to 3* bind strongly to EGFR when tested by ELISA. While much care was taken to eliminate any cross-contamination between libraries, it is difficult to rule out minor contamination with a strongly-binding EGFR clone. However, considering the inbred nature of Balb/c mice, it is possible that the same scFv arose independently from two different libraries.

1.11 Analysis of the affinity and specificity of binding to EGFR Based on good binding to antigen and diversity in DNA sequences, several scFvs derived from the lymph node and spleen libraries were chosen for further analysis. These scFvs were analyzed by ELISA for binding to purified EGFR, binding to irrelevant antigens, and binding to tumor cell lines that did or did not express EGFR. As a positive control, scFvs were prepared from mouse 425 MAb As negative controls, o0 scFvs were prepared from phage antibodies isolated from the lymph node and spleen libraries prior to selection (L1 and S1, respectively).

The concentration of scFvs was determined by comparing dilutions of the scFvs to be tested with dilutions of a purified scFv of known 5 concentration in a Western blot.

The scFvs were tested by ELISA for binding to purified EGFR and the results plotted (Figure It was possible to rank the scFvs with respect to their binding to EGFR. These rankings were reproducible between experiments. The scFvs that bound most strongly to EGFR were 12 1C and L3 10A from the lymph node library and S4 1 OH from the spleen library. As described previously, scFvs L3 10A and S4 have the same DNA sequences. A scFv (S4 5A) that was very similar to scFv S4 10H, with two amino acid changes in VH and two in Vk, consistently gave a lower ranking than S4 10H. In contrast, the differences in sequence observed between L2 12B and L3 11D did not appear to have a pronounced effect on the binding. Of the scFvs isolated only two, L2 8C and L2 11C, appeared to bind less well than scFv 425.

The scFvs were tested by ELISA for binding to plastic and to a panel of unrelated proteins (ovalbumin, hen egg lysozyme, cytochrome c, g!yceraldehyde 3-phosphate dehydrogenase, CBA albumin, and BSA).

None of the scFvs gave a signal above background.

-21- The scFvs were tested by ELISA for binding to three tumor cell lines.

Cell lines A431 and MDA MB 468 are EGFR-bearing tumor cells isolated from the vulva and breast, respectively. Cell line SK-MEL-23 is a ganglioside-bearing melanoma cell line and was included as a negative control. Of the ten scFvs tested, only four bound to both purified EGFR and EGFR-bearing tumor cells (L2 128, L3 11D, L2 11 C, and 34 2D, Figures 5 No binding to SK-MEL-23 cells was detected. There are several possible explanations for this surprising result. One may be that the EGFR that was used for immunization, selection, and ELISA was secreted EGFR-related protein (Weber et al., Science 1984. 224: 294). This protein has an additional 17 amino acids at the C-terminus (GUnther et al., J. Biol. Chem. 1990. 265: 22082). The scFvs were tested by ELISA for binding to this 17 amino acid peptide and no binding was observed. It is possible that the secreted EGFR-related protein and the EGFR on the tumor cell surface have differences in conformation or glycosylation.

To further investigate binoing to tumor cells, three scFvs (L2 11A, L3 11D and S4 2D) were purified and analyzed for binding to A431 tumor eo cells by flow cytometry. The 425 scFv was used as a positive control.

Of the three scFvs tested, only L3 11D and S4 2D bound to A431 cells.

These two scFvs had similar binding profiles to scFv 425.

Purified scFvs prepared from two of the isolates that bound to both EGFR and EGFR-bearing tumor cells (L3 11D and S4 2D) were tested in competition binding assays with mouse 425 MAb. While purified scFv 425 was able to inhibit mouse 425 MAb from binding to EGFR over a given concentration range, scFvs L3 11D and S4 2D did not inhibit mouse 425 MAb from binding to EGFR at these concentrations.

These two scFvs appear to recognize an epitope on EGFR that is different from that recognized by mouse 425 MAb.

'A 1" 1. I I I 22 Chimeric whole antibodies derived from scFvs.

Two scFvs (L3 11D and S4 2D) were selected for conversion into whole antibody molecules. DNAs coding for the mouse VHS and VKs were cloned into intermediate vectors containing DNA sequences coding for immunoglobulin leader sequences and splice donor signals (Fig. The positioning of the cloning sites in the VH intermediate vector meant that the first residue of the VH was changed from aspartic add to glutamic acid. From the intermediate vectors, DNA fragments containing the VHs and VKs, now joined to leader and splice donor sequences, were cloned into mammalian cell expression vectors 15 containing DNAs coding for either human gamma-1 constant region or human kappa constant region (Maeda et al., Hum. Antibod. Hybridomas 1991. 2: 124). For each chimeric antibody, the heavy chain and light chain expression vectors were co-transfected into COS cells. As a 20 positive control, cells were also co-transfected with heavy and light chain expression vectors coding for chimeric 425 antibody (Kettleborough et al., Protein Eng. 1991. 4: 773). Medium was S- collected from the cells and analyzed by ELISA to determine the concentration of antibody present and the ability of the antibody to bind to EGFR (Fig. When the antibody concentration required to achieve half-maximum binding to antigen were compared, chimeric S4 2D antibody bound to EGFR equally as well as chimeric 425 antibody.

Chimeric L3 11D antiDody, however, bound to EGFR approximately four-fold less well than chimeric 425 antibody. The affinity of chimeric 425 antibody (Kettleborough et al, Protein Eng. 1991. 4: 773) has been determined by competition binding analysis to be 1.9 x 108 M- 1 These results were surprising because previous data analyzing the scFvs had indicated that scFvs S4 2D and L3 11D both bound to EGFR better 23 than scFv 425 (Fig. Protein A-purified samples of chimeric L3 11D and S4 2D antibodies were analyzed by SDS-PAGE under reducing and non-reducing conditions. Chimeric L3 11D and S4 2D antibodies were also tested by flow cytometry for binding to A431 and SK-MEL-23 cells. Both chimeric antibodies bound well to the EGFR-expressing A431 cells and did not bind to the EGFR-negative SK-MEL-23 cells.

Therapeutic and diagnostic use The antibody fragments and whole antibodies according to the invention can be administered to human patients for therapy.

Therefore, it is an object of the invention to provide a pharmaceutical formulation comprising as active ingredient at least one antibody or antibody fragment as defined above and in the claims, associated with one or more pharmaceutically acceptable carrier, excipient or diluent 2* therefore.

Typically the antibody of this invention will be injected intravenously or parenterally. Generally, the dosage ranges for the administration of the antibodies fragments are large enough to produce the desired tumor suppressing and tumor lysing effect. The dosage will depend on age, condition, sex and extent of the disease in the patient and can vary from 0.1 mg/kg to 200 mg/kg, preferably from 0.1 mg/kg to 100 mglkg/dose in one or more doses administered daily, for one or several days.

Preparations for parenteral administration includes sterile aqueous or non-aqueous solutions, suspensions, and emulsions. Examples of nonaqueous solvents are propylene glycol, polyethylene glycol, vegetable oils such as olive oils, and injectable organic esters such as ethyl oleate and other solvents known in the art which are suitable for these purposes. The antibodies of this invention can be used in a composition comprising a physiologically acceptable carrier. Examples of such suitable carriers are saline, PBS, Ringer's solution, or lactated Ringer's solution. Preservatives and other additives such as antibiotics, antioxidants, and chelating agents may also be present in the pharmaceutical formulations.

The antibody (fragment) can also be conjugated according to known methods to cytokines such as IL-2 in order to support their cytotoxicity.

The pharmaceutical formulations of the present invention are suitable for the treatment of all kinds of tumors, including melanomas, gliomas and carcinomas, as well as tumors of the circulating system and solid tumors.

For diagnostic purposes the antibody can be conjugated, for example, S to a radio-opaque dye or can be radiolabelled. A preferred labelling method is the lodogen method. Preferably the antibody will be administered as F(ab') 2 or scFv fragments for diagnostic purposes.

This provides superior results so that backround substraction is unnecessary.

o EXAMPLE 1: A431 vesicles Shed membrane vesicle preparations were obtained as previously described by (Cohen et al., J. Biol Chem. 1982. 257: 1523; Yeaton et al., J. Biol. Chem. 1983. 258: 9254) with some modifications. Confluent flasks containing A431 cells were washed with PBS containing calcium II;V It *I I 25 and magnesium. Hypotonic PBS was added and flasks were shaken for 15 minutes. Cells were then washed with vesiculatiol buffer (100 mM NaCI, 50 mM Na 2

HPO

4 5 mM KCI, 0.5 mM MgSO? pH Vesiculation buffer was added and flasks were kept in agitation at room temperature and at 37°C. Then, buffer was decanted through metallic screen into 50 mi tubes in ice and centrifuged for 5 minutes at 150 x g at 4°C. The pellet was discarded and the supematant was uitracentrifuged at 39,000 rpm ,Jr 90 minutes. The final pellets were resuspended in 10 mM Hepes buffer (pH To analyze EGFR from vesicles, samples were precipitated with 9 volumes of ethanol resuspended with 0.08 M Tris, pH 6.8, and then SDS-PAGE was carried out with MAb 425 as standard.

The protein content of the preparations was quantitated by a modified Coomassie Plus method using BSA as a standard and read at 595 nm.

To analyze EGFR from vesicles, samples were precipitated with 9 volumes of ethanol (overnight at The pellet was resuspended with 20 Tris (0.08 M, pH 6.8) and then a SDS-PAGE was run stacking gel; Sh, 35 mA; 10% running gel; 2.5 h; 40 mA). Samples and standard were in duplicate. One of them was stained with Coomassie Blue and the other Swas blotted onto nitrocellulose sheets (12 V; 16 h at 40C) and treated with mouse mAb 425 (anti-EGFR) and anti-mouse IgG antibody conjugated to S. alkaline phosphatase.

Three media were used in the in vitro immunizations. Medium-1 (Ml), 30 Medium-2 (M2) and Mixed Thymocyte Culture medium (MTC). M1 consisted of HL1 (Ventrex Laboratories, USA) supplemented with 50 mM 2-mercaptoethanol and 2 mM L-Glutamine (Gibco). M2 consisted of HL1 supplemented with 50 mM 2-mercaptoethanol; 40 U/ml IL-2 (Genzyme); 3 20 mg/ml Adjuvant Peptide (Sigma); 2 mM L-glutamine; 100 U/ml penicillin (Gibco); 100 mglml streptomycin (Gibco). 4% or 20% of FCS I ll" 1, 1 I I 26 (Biological Industries) was added to M2. MTC was prepared as described by Vaux Briefly single cell suspensions of thymuses of three-week-old Balb/c and C57/BL-1 mice were prepared by pressing the thymus glands through a sterile 50-mesh screen. The cell sus-pension was collected, washed twice in HBSS and the number of viable cells was determined by trypan blue exclusion. Thymocytes were then cultured at a density of 2.5x10 6 thymocytes of each strain per ml in HL1 medium 1 containing 4% FCS, 2 mM L-glutamine, 100 U/ml penicillin and 100 mg/ml streptomycin. After 48 hours, the supematant was recovered, filtered through a 0.22 mm filter, and stored at A suspension of sp!enocytes from non-immunized eight-week old BALB/c mice was obtained as described for thymocytes. Viability was determined by trypan blue exclusion.

EXAMPLE 2: In vitro immunization and screening Three media were used in the in vitro immunizations. Medium-1 (M1), S" Medium-2 (M2) and Mixed Thymocyte Culture medium (MTC). M1 S consisted of HL1 (Ventrex Laboratories, USA) supplemented with 50 mM 2-mercaptoethanol and 2 mM L-Glutamine (Gibco). M2 consisted of HL1 S supplemented with 50 mM 2-mercaptoethanol; 40 U/ml IL-2 (Genzyme); 20 mg/ml Adjuvant Peptide (Sigma); 2 mM L-glutamine; 100 U/ml penicillin (Gibco); 100 mg/ml streptomycin (Gibco). 4% or 20% of FCS (Biological Industries) was added to M2. MTC was prepared as described by Vaux Briefly single cell suspensions of thymuses of three-week-old Balb/c and C57/BL-1 mice were prepared by pressing the thymus glands through a sterile 50-mesh screen. The cell sus-pension was collected, washed twice in HBSS and the number of viable cells was determined by trypan blue exclusion. Thymocytes were then cultured at a density of 2.5x10 6 thymocytes of each strain per ml in HL1 medium t ,b I -27containing 4% FCS, 2 mM L-glutamine, 100 U/mi penicillin and 100 mg/ml streptomycin. After 48 hours, the supematant was recovered, filtered through a 0.22 mm filter, and stored at A suspension of splenocytes from non-immunized eight-week old BALB/c mice was obtained as described for thymocytes. Viability was determined by trypan blue exclusion.

Single cell suspensions from thymuses of three-week-old Balb/c and C57/BL-1 mice were obtained by pressing the thymus glands through a sterile 50-mesh screen. The cell suspension was collected, washed with HBSS and the number of viable cells was determined by trypan blue exclusion. Thymocytes were then cultured at a density of 2.5x106 thymocytes of each strain per ml in HLI-medium containing 4% FCS, 2 mM L-glutamine, 100 U/ml penicillin and 100 pg/ml streptomycin.

After 48 hours the supematant was recovered, filtered and stored. A suspension of splenocytes from non-immunized eight-week-old 20 BALB/c mice was obtained as described for thymocytes. Viability was determined by trypan blue exclusion.

In vitro immunizations were performed in 6-well plates (Costar). Wells containing 107 splenocytes in 3.5 ml of Mi-medium (consisting of HL1-medium, Ventrex Laboratories, USA, supplemented with 50 pM 2mercaptoethanol and 2 mM L-glutamine (Gibco)) were incubated (37 0 C, 5% CO 2 with vesicles bearing EGFR at the desired concentration. Vesicles from cells not expressing EGFR or PBS were 30 added in control wells. After some hours, 3.5ml of M2-medium (consisting of HL1 supplemented with 50 pM 2-mercaptoethanol, U/ml IL-2 (Cenzyme), 20 pg/ml adjuvant peptide (SIGMA), 2 mM Lglutamine, 100 U/ml penicillin (Gibco), 100 pg/ml streptomycin (Gibco)) containing 4% or 10% FCS (Biological Industries) was added to each well. In some experiments M2 was replaced by MTC-medium I

I

28 (mixed thymocyte culture medium (Vaux et al., Nature 1988. 336: 36) supplemented with adjuvant peptide (20pg'ml) and IL-2 (40 U/ml) (Note that the final concentration of FCS, IL2 and adjuvant peptide in culture is 50% reduced). Cells were incubated for 72, 96, 120 or 144 h in the same conditions and, finally, the cells were tested for the presence of specific immunoglobulin or processed for RNA isolation.

Screening was carried out with purified antigens or A431 fixed cells.

The procedure was essentially as previously described (Carroll et al., Hybridoma 1990. 9: 81) with some modifications. Briefly, sterile 96well plates (Nunc, Maxisorb) were coated overnight with purified EGFR pg/ml), GD 3 ganglioside (2pg/ml), or RNase (10pg/ml) in PBS.

When A431 cells were used as antigen, cells were cultured in 96-well plates until confluent and fixed with 0.1 glutaraldehyde. In vitro immunized iymphocytes were washed and resuspended in HL1 medium supplemented with 2 FCS and 2 mM of L-glutamine at 20 5x10 5 cells/ml and 1x10 5 cells were added to each well and incubated (370C, 5% CO2) for 48 h. Sixteen duplicates of each group were done. Lymphocytes were then removed by washing 5 times in PBS containing 0.1 Tween-20. Specific immunoglobulins were 25 detected using peroxidase labelled rabbit anti-mouse immunoglobulin (Dako) (1 hour, 370C). 2,2'-Azino-bis(3-ethylbenz-thiazoline-6- S sulfonicacid)-diammonium salt (ABTS) (Sigma) in citrate-phosphate buffer (0.55 mg/ml) was used as substrate.

S 30 EXAMPLE 3: Library construction Three libraries were constructed from RNA prepared from the spleen of a mouse immunized intraperitoneally with A431 cells (Murthy et al., Arch. Biochem. Biophys. 1987. 252: 549) from the popliteal lymph node of a mouse immunized in the footpad with purified EGFR, and from mouse cells immunized in vitro with A431 vesicles. First-strand -29cDNA was synthesized. The VH and VK genes were PCR-amplified and assembled (Clackson et al., Nature 1991. 352: 624). Using PCR, Notl and Sfil restriction sites were appended and the scFvs cloned into the phagemid vector pHEN1 (Hoogenboom et al., Nucl. Acids Res.

1991.19: 4133). The ligation mixtures were electroporated into E. coli cells and the resulting colonies scraped into medium to generate library stocks (Marks et al., J. Mol. Biol. 1991. 222: 581).

EXAMPLE 4: Library screening Phage antibodies were rescued from ihe libraries using M13K07 helper phage (Promega, Madison, WI) (Marks et al., J. Mol. Biol. 1991.

S222: 581). Immunotubes (Nunc, Life Sciences, Paisley, UK) were coated with 4 ml of 2.5 igg/ml EGFR in PBS overnight. After three washes with PBS, tubes were incubated at 370C for at least 1 h in PBS 20 containing 2% milk powder (PBSM). The phage (1012 to 1013) were 20 resuspended in 4 ml PBSM and incubated in the EGFR-coated tube for 1 h at room temperature. The tube was washed 20 times with PBS, 0.1% Tween and 20 times with PBS. Bound phage were eluted after a min incubation in 1 ml of 0.1 M triethylamine with end-over-end mixing. The eluted phage were neutralized by the addition of 0.5 ml of 1 M Tris-HCI, pH 7.5 and used to infect log-phase E coliTG1 cells.

Infected cells were plated and individual colonies picked for small- *,scale induction of scFvs. The remaining colonies were scraped into medium and an aliquot used to prepare phage for the next round of screening.

EXAMPLE 5: Production and analysis of scFvs %1 1)1'IJ Soluble scFvs were produced in E. coli HB2151 as previously described Kettleborough et al., i. The scFv concentrations in the bacterial supematants were estimated using a purified scFv preparation of known concentration as a standard. Supematants were filtered and sodium azide added to Serial dilutions of the supematants and of the standard were spotted onto Immobilon-PVDF filters (Millipore, Watford, UK) using a 96-well manifold. The filters were treated as for a Western blot (Towbin et al., Proc. Natl. Acad. Sci.

USA 1979. 76: 4350). The scFvs were detected using an antibody (9E10) directed against the C-terminal tag (Munro and Pelham, Cell 1986. 46: 291) followed by a peroxidase-conjugated goat anti-mouse IgG and IgM antibody (Jackson ImmunoResearch Lab Inc., West Grove, PA). The reactions were developed using the ECL system (Amersham, Aylesbury, UK). Pre-flashed autoradiographs were *i scanned using a densitometer. A standard curve was prepared and 20 used to estimate the scFv concentrations in the supematants.

Antigen-binding ELISAs were carried out with EGFR-coated plates pg/ml). Supematants containing scFvs were diluted in PBSM and added to the plates. Bound scFvs were detected using 9E10 antibody Sas described above. Supematants were also tested for binding to a panel of unrelated proteins and plastic. ELISA plates were coated overnight at 100 gg/ml with ovalbumin, hen egg lysozyme, cytochrome c, glyceraldehyde 3-phosphate dehydrogenase, murine albumin (CBA strain), and BSA. Undiluted supematants containing 2% milk powder were added in duplicate to the coated plates and bound scFvs detected as described above.

Cell-binding ELISAs were carried out using tumor cell lines, A431 (ATCC CRL 1555), MDA MB 468 (ATCC HTB 132), and SK-MEL-23 (negative control). Cells were grown to confluency in poly-D-lysine- -31treated 96 well tissue culture trays (Nunc). Cells were washed with DMEM and blocked at 370C for 2 h with PBS containing 2.5% BSA.

After aspiration, supematants were added to each well together with an equal volume of 2xYT media containing 4% milk powder and incubated at 4 0 C for 1 h. Bound scFvs were detected as described above.

A competition-based ELISA was carried out by pre-incubating EGFRcoated ELISA plates with 50 pl of purified scFv (100 pg/ml) for 10 min.

Mouse MAb 425 (50 0) was then added to give concentrations of 3.13 to 200 ng/ml. Following incubation and washing, bound mouse MAb 425 was detected using peroxidasa-conjugated goat anti-mouse IgG 15 and IgM antibody.

EXAMPLE 6: DNA analysis For BstNI fingerprinting, the scFv inserts from individual clones were amplified by PCR and the products digested with BstNI (Clackson et al., Nature 1991. 352: 624). DNA was sequenced using a Sequenase Se kit (United States Biochemical, Cleveland, OH).

EXAMPLE 7: Purification of scFvs Bacterial supematants were clarified by centrifugation and filtration through 0.2pm filters before loading onto a 1 ml column of purified EGFR (5 mg) coupled to cyanogen bromide-activated Sepharose 4B (Pharmacia, Uppsala, Sweden). The column was washed with 30 ml of PBS followed by 5 ml 0.2M glycine, pH 5.0. The scFvs were eluted I 'II LI -32with 0.2M glycine/HCI, pH 2.8. The eluate was neutralized with PBS. Protein-containing fractions were pooled and the buffer changed by ultrafiltration (Amicon, Stonehouse, UK) to PBS containing 1% BSA and 0.05% sodium azide.

EXAMPLE 8: FACS analysis of purified scFvs A431 cells were trypsinized and incubated in DMEM containing FCS. Cells were washed twice with cold DMEM and filtered through a pm screen. Cells (106) were incubated on ice for 30 min in 50 pi PBS, 1% BSA, with purified scFvs. After two washes with cold PBS, bound scFvs were detected using 50 pJ FITC-conjugated 9E10 antibody (100 jpg/ml). After 30 min on ice, cells were washed once with PBS, fixed in PBS containing 1% formaldehyde, and analyzed using a FACSCAN (Becton-Dikinson, Cowley, UK).

SEXAMPLE 9: Construction, analysis and expression of whole chimeric antibodies Using Pstl and BstEII sites, DNAs coding for the VHS of the selected scFvs were subcloned into an intermediate VH vector containing a eukaryotic leader sequence derived from human antibody HG3 CL (Rechavi et al., Proc. Natl. Acad. Sci. USA 1983. 80: 855) and a splice donor site (Fig. The DNAs coding for the VkS were adapted for insertion into an intermediate Vk vector using PCR primers to incorporate Xhol and Sstl sites at the and ends (VkFor: 5'-CCG TTT CAG CTC GAG CTT GGT CCC-3', VkBack: 5'-GAC ATT GAG CTC ACC CAG TCT CCA-3').

The Sstl-Xhol fragments were cloned into the intermediate Vk vector containing a eukaryotic leader sequence derived from reshaped I t: 1 -33 human CAMPATH-1 light chain (Riechmann et al., Nature 1988. 332: 21) and a splice donor site (Fig. The DNAs coding for the variable regions plus eukaryotic flanking regions were cloned as Hindlll-BamHI fragments into mammalian cell expression vectors containing genomic DNAs coding for human gamma-1 constant region or human kappa constant region (Maeda et al., Hum. Anfibod. Hybridomas 1991. 2: 124). The heavy and light chain expression vectors were electroporated into COS cells. After 72 h, medium was collected and the chimeric anti-EGFR antibodies analyzed by ELISA (Kettleborough et al., Protein Eng. 1991. 4: 773).

15 EXAMPLE 10: Production of sc Fvs derived from in vitro immunized cells.

The methods disclosed below are slight modifications of the methods described above. Immunization, library construction and screening are 20 given in Examples 1 4. The following steps are described in detail below: After screening the primary library and the clones derived from the three rounds of panning, some single ampicillin-resistant colonies were selected. Phagemid DNA was prepared by alkaline lysis and used to transfect E.coliHB2151, a non-supressor strain, by heat shock. Colonies were inoculated into 2xTY-Amp-Glu and grown overnight at 30 0 C. A 5 ml aliquot was used to inoculate 50 ml of 2xTY broth containing 100 mg ampicillin/ml and 0.1% glucose and grown with shaking at 30°C for 1 h (until log-phase). Cells were harvested and expression of soluble scFv was induced by the addition of isopropyl i-D-thiogalactopyranoside (IPTG) to a final concentration of 1 mM (De Bellis, D. and Schwartz, I.; Nucleic Acids Res.; 1990. 18: 1311). Cultures were grown overnight at -34 with shaking. Supematants containing scFv were taken, clarified by centrifugation and filtration through 0.22 mm filters and tested. Bacterial supematants were tested for binding to EGFR by ELISA, as described (Kettleborough, et al., EP 94104160 and Eur. J. Immunol. 1994. 24: 952).

The specificity of selected scFv fragments was checked by ELISA using plates coated with various proteins related and non-related to EGFR, as well as other antigens and plastic. The antigens used were: RNase, BSA, OVA, GD 3 ganglioside, vitronectin receptor (VNR), platelet glycoprotein Ilbllla (GPlIbllla), and disialyl-lacto-N-tetraose (DSLNT). Coating was done overight at the optimum concentration for each antigen. Coated ELISA plates were blocked for 1 h at 37 0 C with 1.5% skimmed milk in PBS After washing, 100 ml of scFv supematants were added to the microtiter wells and incubated for 2 h at 37 0 C. Bound scFv were detected using the anti-c-myc antibody 9E10 (spent culture media from Myc 1- 9E10.2 hybrid) and an alkaline phosphatase-conjugated rabbit anti-mouse antibody (Dako).

Three EGFR-bearing tumor cell lines, A431, MDA MB 231 human breast adenocarcinoma (ATCC, HTB 26), and HT29 human colon adenocarcinoma (ATCC, HTB 38), and one non-expressing EGFR cell line, WM164, were used to test the ability of scFv to bind to EGFR on cells by mean FACS analysis and immunofluorescence with unfixed cells.

For the indirect immunofluorescence analysis, cells were plated into Terasaki plates (2x10 4 cells/well) and cultured for 24 h. Cells were them incubated with 20 ml of crude bacterial supematant containing the scFv fragments for 90 min at room temperature. Incubations with primary antibody (anti-c-myc) and secondary antibody were carried out for 60 min at room temperature. The secondary antibody, FICT-conjugated rabbit anti-mouse antibody (Dako) was diluted 1:20.

l*11r 1 For FACS analysis, 5x10 5 cells were washed with PBS with 1% BSA and 0.1% sodium azide (PBS-BSA) and incubated at 4 0 C for 20 min with ml of crude bacterial supematant. After two washes with cold PBS-BSA, bound scFv was detected using anti-c-myc antibody and FITC-conjugated goat anti-mouse antibody (Becton-Dickinson) diluted 1:25 in PBS-BSA.

Propidium iodide (PI) was added at a final concentration of 5 mg/ml. Flow cytometry analysis were performed in a EPICS Profile II equipped with an t air-cooled argon laser. The 488 nm line (15 mV) was used for the excitation. A 530 nm band pass filter was used to collect FITC emission and a 625 nm band pass filter was used to collect PI emission. Living cells were selected by setting a bitmap on forward and side scatter and by exclusion of PI-stained cells.

The diversity of the primary and selected libraries was determined by SPCR amplification of cloned fragments (Gossow, D. Clackson, T; Nucleic Acids Res. 1989. 17: 4000) and analysis of the BstNI digestion pattern 20 Some clones were sequenced using a Sequenase kit (USB) by the dideoxy chain termination method (Sanger, F et al.; Proc. Nat.Acad. Sci., U.S.A. 1977. 74: 5463).

bacterial supematants (10 ml) were subjected to SDS-PAGE using a 12.5% gel. Western blotting was performed essentially as described by Towbin (Towbin et al. J. Proc.Nat.Acad.Sci, U.S.A. 1979. 76: 4350). Proteins were transferred by electroblotting to Immobilon-P (Millipore) or nitrocellulose (Bio-Rad). The blot was blocked with PBS containing 2% skimmed milk scFv fragments were detected using anti-c-myc antibody (9E10), peroxidase-conjugated anti-mouse antibody (Jackson), and an enhanced chemiluminiscence system (ECL, Amersham).

The quantitative analysis of the shed membrane vesicles revealed a total protein concentration of 2.5 mg/ml, of which only 10-14% corresponded to EGFR (Sato et al.; J. Natl. Cancer Inst. 1989. 21: 1601; Yeaton, R et al., J. Biol. Chem., 198o. 258: 9254), 250 to 350 ng/ml. Electrophoretic analysis using PAGE-SDS followed by Coomassie-blue staining showed that the vesicles contained a rather complex mixture of proteins. No S protein degradation was detected. Western blot analysis revealed that under our experimental conditions complete molecules of EGF receptor were present in the membrane vesicle preparation.

In order to determine the requirements for FCS and limphokines MTC and M2 containing 20% or 4% FCS were compared. Vesicles bearing EGFR and PBS were used as antigen and control respectively. Splenocytes were incubated in six well plates with or without antigen for 3 h in M1 (serum-free). MTC or M2 was then added and, after 72, 96, 120 or 144 h, i. 15 screening was carried out using A431 fixed cells. In all experiments, the number of viable cells recovered was between 20 and 40% in agreement with published results (Gavilondo-Cowley, J. et al.; In Vitro Immunization in Hybridoma Technology, Elsevier Science Publishers Amsterdam 1988, p.

131). The maximum specific response was obtained on day four with MTC; whereas, M2 at 4% or 20% FCS or 10% final concentration) delayed the maximum response until day six (Table However MTC Sand 10% FCS triggered a non-specific response, probably by polyclonal activation, as could be seen when the results were expressed as the ratio of specific non-specific response. For further assays we decide to use M2 supplemented with 4% FCS and 6 d of culture.

The presence of EGFR in the suface of vesicles strongly enhanced the response to this antigen. Insimilar protocols as described above, vesicles from expressing and non-expressing EGFR cell lines were compared.

Lymphocytes were cultured with vesicles in Ml for 3 h. Afterwards M2 containing 4% FCS was added. After 6 d, lymphocytes from each group were cultured for 48 h in 96 well plates coated with EGFR, A431-fixed cells, RNase or G3. As expected, the results of these assays showed a cells, RNase or GD3. As expected, the results of these assays showed a I I. fIII 37 multispecific patter of response (Table The reactivity against EGFR was clearly increased in terms of optical density when EGFR-expressing vesicles were used as antigen.

Taken together, these results suggest that, although immature, there was a measurable antigen-dependent response after in vitro immunization which generated several pools of immunized lymphocytes against EGFR suitable for PCR-cloning of variable regions.

A library of 1.1x10 5 clones was obtained after cloning scFv fragments derived from in vitro immunization into the pHEN1 phagemid. This library was generated in parallel with two more libraries providing of in vivo immunization. The construction of these phages libraries has been

S

i 15 described previously (Kettleborough, et al., EP 94104160 and Eur. J.

Immunol. 1994.24: 952).

To select the scFv fragments binding to EGFR, phage were panned using EGFR-coated immunotubes. Eluted phage were used to reinfect a SupE 20 strain of Ecoli. In total, three rounds of selection were carried out. In each round, a tube without antigen was tested in paralel to calculate the background. In the first panning, 1.5x1010 phage particles were applied to the immunotube and 6.6x104 were eluted from the coated immunotube; whereas, only 200 colonies were obtained from the background population. After the third panning, lx1011 phages were applied and 5.6x10 1 0 were eluted.

To further characterize the scFv fragments, we selected 22 clones from the phage populations, before selection and after each round of selection.

The diversity of the library was analyzed by the BstNI digestion patterns of the cloned fragments. Prior to selection the library appeared to be extremely diverse. Fingerprinting of binding clones derived after the first round of selection indicated the presence of several groups with the same restriction pattem.

Clones were selected from different rounds of selection based on their digestion patterns. DNA sequencing revealed the presence of diferent sequences in most of the selected clones. The length and composition of complementarity determining regions (CDRs) of clones 10D2, 503, 10E2, 1 B3, 4B3 and 5E2 were different. The most variation was observed in the CDR3s of VH and VL sequences. Clones 5D3 and 1 E3 were derived from the third round of selection. They bound strongly to EGFR as analyzed by ELISA and flow cytometry and had the same sequence.

Soluble scFv fragments were obtained by growth of the non-suppressor 15 Ecolistrain HB 2151 in presence of IPTG.

To verify scFv production, bacterial medium from individual clones, was .analyzed by gel electrophoresis. Western blot analysis revealed a clear band around 35,000 kD.

Clones with binding activities to EGFR were identified by ELISA- To examine the cross-reactivity of selected clones, ELISA assays using different antigens were carried out. The antigens (EGFR, RNase, BSA, KLH, OVA, GD 3 ganglioside, vitronectin receptor, platelet glycoprotein lbllla, and disialyl-lacto-N-tetraose) were coated into ELISA plates at the S. optimum concentration (Table No binding to non-EGFR antigens was detected. The scFvs were also tested for binding to three EGFR-bearing tumor cell lines (human epidermoid carcinoma A431. human breast 3 adenocarcionma MDA MB 231 and human colon adenocarcinoma HT 29). WM 164 a human melanoma non-expressing EGFR was used as a negative control. Those that bound to tumor cell lines was tested by indirect immunofluorescence using unfixed cells and quantified by FACS analysis. The use of unfixed cells ensures the natural conformation of the membrane receptors. Positive clones showed a clear fluorescence using -39- A431 cells. Fluorescence with the others EGFR-bearing tumor cell lines was weak. No binding to the negative cell line was detected. The results were confirmed by flow cytometry. Seventeen positive clones and three negative clones were analyzed for binding to A431, MDA MB 231 and HT 29 cells by flow cytometry. WM 164 was used as the negative cell line.

The 425 scFv (P1 clone) was used as a positive control and the cloning J vector (HEN) as a negative cont-ol. The results are summarized in Table 0 5. Two clones, 4B2 and 5E2, were positive for binding to EGFR, as analyzed by ELISA, but negative for binding to EGFR-expressing tumor cell lines.

F

I t V I I I I I 40 Table 2. Effect of different media on in vitro irmunization.

Day of screening against of A431I 3th day 4th day 5th day 6th day Amsy Antigen O.D.P 1 Ratiod) O.D. Ratio O.D. Ratio 0,D. Ratio I Vesicles 0.393 2.11 0.801 3.76 0.784 3.90 0.951 10.3 PBS 0.186 0.213 0.201 0.092 2 Vesicles 0.527 2.50 0.852 1.76 0.863 2.75 1.168 3.94 0.210 0.482 0.3 13 0.296 3 Vesicles 0.763 1.48 1.169 2.01 1.089 2.07 1-115 1.91 PBS 0.513 0.581 0.525 0.581 ::':Assay 1: Ml plus M2. FCS (Final FCS: 2%) *Assay 2: M1 plus M2, 20% FCS (Final FCS: Assay 3: A-medium plus MTC, 4% FCS (Final FCS: 2%) a) BALB/c mouse spleen cells (I0~ were incubated in 3.5 ml of M1 with 20 vesicles from A431 cells or PBS for 3 h in wells of 6 well plates. Afterwards ml of MTC or M2 containing 4% or 20% FOS were added and the plates incubated. At 3, 4, 5 or 6 days in vitro immunized lymphocytes were removed from culture medium, washed in HBSS to remove vesicles and seeded in 96 well plates coated with fixed A431 cells, and incubated for 48 h (see Methods).

b) Final concentration of FCS in culture medium.

c) O.D- Optical densitiy read at 405nm. It represents the mean of sixteen wells.

d) Ratio of specific response (vesicles as antigen) unspecific response (PBS as antigen).

M111-16. )I I D( A, -41- Table 3. Multi-specificity of the resp~onse after in vitro immuizationl EGFR 0.427 0.070 0.123 0.304 Assay 2 EGFR 1.430 0.730 0.233 0.670 EGFR- 0.789 0.195 0.118 0.561 a) Lymphocytes "ere in vitm imrmunized using either EGFIR-expressiflg vesicles (EGFR+) or non-EGFR expressing vesicles (EGFR-). After six days of incubaition, cells were removed from culture and screened against the above mentioned antigens.

b) Response is expressed as optical density (405 nm).

Table 4. Cross-reactivity of selected scFv fragments against several antigens

ANTIGEN

3 COATIG [mghnl

BEIULT

EGFR a t:RNase to *BSA KLH **OVA

GD

3 ganglioside 2 VNR 1 GPl~bMh I DSLNT a) ELISA assays were perfonned as describedl b) Vitronectin receptor (VNR) platelet glycoprotein UMi~ (GPI~blia);- disialyl-lacto- N-tetraose

(DSLNT).

.k -I ;1t 1U IIIK- 42 Table 5. Reactivity of scFv clunes against EGFR. Comparative results bet-ween an ELISA method with purified soluble antigen and cytometric analysis of cellt lines.

CLONES CYTONEETRIC ANALYSIS OF TUMOR CELL LINES ELISA (mma of tiridtqi fhiorescence units) Postiv WJ164 A431 kDAAMB231 H{129 EGFR 7111 1.5 112.9 16.4 2.6, 1.2 4112 1.2 5.3 412 0.6 2 10D2 1.5 145.3 36.3 4.8 2 12D2 1.8 129.5 29.3 5.7 2 5E2 I4 2.5 7.1 0.5 1.8 15 8E2 1.5 134.5 47.7 5.1 1.9 5F2 1.3 146.3 4.0.6 5.7 19 21.9 152.2 25.3 2 1.9 1113 0-6 105.1 36.4 5.2 >2 4B3 0.5 78 15.8 2.3 2 3D3 1.2 94.3 25.1 4.8 1.9 *5D3 0.5 112 22.2 5.5 >2 4 F3 0.4 110.3 32.3 6.2 >2 4G3 0.4 76.5 20.4 2 >2 I F-3 0.4 118.3 33.8 5.1 2 3H3 0.6 76.5 33.7 4.2 >2 Negative 5F1 2.4 2.3 3.6 1.8 0.2 7G1 1.4 10.2 4 2.8 C12 1H1 0.5 5 4 0.75 0.2 43 Table 5: continuation Conitrols Ib HEN 0.4 4.1 3.7 1 0.2 PI 0-6 85.5 21.3 2.5 19 a) Three EGFR-bearing cell lines (A431, MDAAMB231 and HT29) and one non-exressing cell line (WM164) were used to assay the ability of scFv to bind to tumor cells lines by cytometric analysis as described.

b) Vector without fragment (HEN) and scFv fi-agnent from 425 mAb (PI) were used as negative and positive controls, respectively.

C

a a .4 a. a.

*SO~

IU 1 III), -44- SEQUENCE LISTING GENERAL INFORMATION:

APPLICANT:

NAME: Merck Patent GmbH STREET: Frankfurter Str. 255 CITY: Darmstadt COUJNTRY: Germany IF) POSTAL CODE (ZIP): 64271 TELEPHOKE: 49-6151-727022 TELEFAX: 49-6151-727191 (ii) TITLE OF INVENTION: Anti-EGFR single-Chain Fvs and Anti-EGFR Antibodies (iii) NUMBER OF SEQUENCES: 32 P. (iv) COMPUTER READA.BLE FORM: MEDIUM TYPE: Floppy disk 15 COMPUTER: IBM PC compatible OPERATING SYSTEM: PC-DOS/MS-DOS so 13 EO SOFTWARE: Patentln Release #1l.0, Verso 13 (EO INFORMATION FOR SEQ ID NO: 1: SEQUENCE CHARACTERISTICS: LENGTH: 327 base pairs 20 TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE:, cDNA (iii) HYPOTHETICAL:

NO

(iv) ANTI-SENSE:

NO

FRAGMENT TYPE:, N-terminal (vi) ORIGINAL SOURCE: ORGANISM: mouse STRAIN: Balb/c DEVELOPMENTAL STAGE: adult TISSUE TYPE: Lymph node 30 (vii) IMMEDIATE SOURCE: CLONE: L2 11C (Liyht chain) (ix) FEATURE: NAME/KEY: CDS LOCATION:l. .327 (xL) SEQUENCE DESCRIPTION: SEQ ID NO: 1: GAC ATT GAG CTC ACC CAG TCT CCA GCC TCC CYG GCT GCA TCT GTG GGA 48 Asp Ile Glu Leu Thr Gin Set Pro Ala Ser Leu Ala Ala Ser Val Gly 1 5 10 is I %X11;4 11 111? I'll, 45 GAA ACT GTC ACC ATC ACA TGT OGA GCA AGT GAG AAC ATT Glu Thr Val Thr Ile Thr Cys Arg Ala Ser Giu Asn Ile 25 TTA GCA TGG TAT CAG CAG AAG CAA GGG AAA TOT COT CAG Leu Ala Trp Tyr Gin Gin Lye Gin Gly Lys Ser Pro Gin 35 40 45 TAT AGT GCA AGO GCC TTG GAA GAT GOT GTO CCA TOG AGO Tyr Ser Ala Ser Ala Leu Glu Asp Gly Val Pro Ser Arg 55 .60 AGT GGA TOT COO ACA CAG TAT TOT TTA AAG ATC AAC AAC Ser Gly Ser Gly Thr Gin Tyr Ser Leu Lys Ile Aan Asn 6S 70 75 GAA GAT ACC GOT ACT TAC TTC TGT AAA CAG ACT TAT GAO Giu Asp Tbr Ala Thr Tyr Phe Cys Lye Gin Thr Tyr Asp 90 AOG TTC GGT GGA CC-G ACC AAG CTG GAA ATA AAA CGO CC Thr Phe Gly Cly Gly Thr Lys Leu Glu Ile Lye Arg Aia 100 105 TAC TAT AGT Tyr Tyr Ser CTC OTC ATO Leu Leu Ile TTC ACT GCO Phe Ser Glv ATG CAG CCT Met Gin Pro GTT COG TG Val Pro Trp 96 144 192 240 288

V

INFORMATION FOR SEQ ID FO: 2: SEQUENCE CHARACTERISTICS: LENGTH: 109 amino acids TYPE: amino acid TOPOLOGY: linear (ii) MOLECULD TYPE: protein (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 2: Asp Ile Glu Leu Thr Gin Ser Pro Ala Ser Leu 1 5 10 Giu Thr Val Thr Ile Thr Cys Arg Ala Ser Glu 25 Leu Ala Trp, Tyr Gin Gin Lye Gin Gly Lye Ser 1 25354 Tyr Ser Ala Ser Ala Leu Giu Asp Gly Val Pro 55 Ser Gly ser Gly Thr Gin Tyr Ser Leu Lye Ile 1 70 75 Glu Asp Thr Ala Thr Tyr Phe Cys Lye Gin Thr 1 85 90 Thr Phe Gly Gly Gly Thr Lye Leu Clu Ile Lye A 100 105 INFORMATION FOR SEQ ID NO: 3: SEQUENCE CHARACTERISTICS: LENGTH: 357 base pairs TYPE: nucleic acid STRANDEDNESS: singie TOPOLOGY: linear Ala Ser Val Gly Ile Tyr Tyr Ser Gin Leu Leu Ile Arg Phe Ser Cly Asn Met Gln Pro Asp Val Pro Trp Ala 46 1 (ii) MOLECULE TYPE: CDNA (iii) H~YPOTHETICAL: NO (iv) ANTI-SENSE: NO FRAGMENT TYPE: N-terminal ORIGINAL SOURCE: ORGANISM: mouse STRAIN: Balb/c DEVELOPMENTAL STAGE: adult TISSUE TYPE: Lymph Node (vii) IMMEDIATE SOURCE: CLONE: L2 11C (heavy chain) (ix) FEATURE: NAME/KEY: CDS LOCATION:1. .357 (xi) SEQUEN4CE DESCRIPTION: SEQ ID NO: 3: CAG GTG CAA CTG CAG GAG TCA GGG CCT GAG CTG Gin Val Gin Leu Gin Glu Ser Gly Pro Glu Leu 110 115 120 TCA GTG AAG ATG TCC TGC AAG GCT TCA GGC TAT Ser Val Lys Met Ser Cys Lys Ala Ser Gly Tyr 130 135 TGG ATA CAC TGG ATG AAA CAG AGG CO-T GGA CAA Trp Ile His Trp Met Lys Gin Arg Pro Gly Gin 145 150 20 GGC ATG ATT GAT CCT TCC AAT AGT GAA ACT AGG Gly Met Ile Asp Pro Ser Asn Ser Glu Thr Arg 160 165 AGG GAC AAG GCC ACA TTG AGT GTA GAO AAA TCC Arg Asp Lys Ala Thr Leu Ser Val Asp Lys Ser 175 180 ATG CAG CTIC AGC AGC CTG ACA TCT GAG GAO TCT Met Gln Leu Ser Ser Lou Thr Ser Clu Asp Ser 190 195 200 GCA AGA TGG GAC TAC GG;T AGT GGC CAC TTY GAC Ala Arg Trp Asp Tyr Gly Ser Gly His Phe Asp 210 215 ACC ACG GTC ACC GTC TCC TCA Thr Thr Val Thr Val Ser Ser 225 INFORMATION FOR SEQ ID NO: 4: SEQUENCE CHARACTERISTICS: LENGTH: 119 amino acids TYPE: amino acid TOPOLOGY: linear (ii) MOLECULE TYPE: protein (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 4:

GTG

Val

ACC

Thr

GOC

Gly

TTA

Leu

TCC

Ser 185

GCA

Ala

TAC

Tyr AGG CCT Arg Pro TTC ACT Phe Thr OTT CAG Leu Gin 155 AAT CAG Asn Gin 170 AAT AAA Aen Lye ATC TAT Ile Tyr TGG GGC Trp Gly GGG GCT Gly Ala 125 ACC TAC Thr Tyr 140 TGO ATT Trp Ile AAT TTC Asn Phe GCC TAO Ala Tyr TAC TOT Ty r Cys 205 CAA 000 Gln Gly 220 48 96 144 192 240 288 336 M ill'* V I 41 47 G1i Se Trj Gic Arc 6: Mel 1o Al.

Thi 1.5 25 (2) n Val Gin Leu Gin C-lu Ser Gly Pro Glu r Val Lys Met Ser Cys Lys Ala Ser Gly 25 p Ile His Trp Met Lys Gin Arg Pro Gly 40 yr Met Ile Asp Pro Ser Asn Ser Glu Thr 55 g Asp-Lys Ala Thr Leu Ser Val Asp Lys 5 70 tGin Leu. Ser Ser Leu Thr Ser Glu Asp 85 90 i Arg Trp Asp Tyr Gly Ser Gly HIis Phe I 100 105 Thr Val Thr Val Ser Ser 115 INFORMATION FOR SEQ ID NO: SEQUENCE CHARACTERISTICS: LENGTH: 339 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: CDNA (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO FRAGMENT TYPE: N-terminal (vi) ORIGINAL SOURCE: ORGANISM: mouse STRAIN: Balb/c DEVELOPMENTAL STAGE: adult TISSUE TYPE: Lymph node (vii) IMMEDIATE SOURCE: CLONE: L2 123 (light chain) (ix) FEATURE: NAME/KEY: CDS LOCATION:1. .339 Leu Val Tyr Thr Gin Gly Arg Leu Ser Ser 75 Ser Ala ~Sp Tyr Arq Pro Phe Thr Leu Gin Asn G Ln Asn Lys Ile Tyr Trp Gly 110 Gly Ala Thr Tyr Trp Ile Asn Phe Ala Tyr Tyr Cys Gln Gly

GAC

Asp 120

CAG

Gin (xi) SEQUENCE DESCRIPTION: SEQ ID NO: ATT GAG 010 ACC CAG TCT CCA GCT TOT TTG GCT GIG TCT CIA GGG Ile Glu Leu Thr Gin Ser Pro Ala Ser Leu Ala Val Sec Leu Gly 125 130 135 AGG GCC ACC ATO TCC ICC AGA GCC AGO GAA AGT GTT CAT AAT TTT Arg Ala Thr Ile Ser Cys Arg Ala Ser GLu Ser Val Asp Asn Phe 140 145 150 1 1. ?11. 1. 1 114 48 *1 GOC FiTT AGT TTT ATG Gly Ile Ser Phe Met 155 AAA CTC CTC ATC TAT Lys Leu Leu Ile Tyr 170 AGG TTT AGT GGC AGT Ara Phe Ser c-iy Ser 185 CCT CTG GAG GAG GAT Pro Leu. Glu Giu Asp 200 GAG GTT CCG CTC ACG Glu Val Pro Leu Thr 220

GCG

Ala AAC TGG TTC CAA CAG AAA CCA OGA CAG CCA CCC Asn Trp Phe Gin Gin Lys Pro Gly Gin Pro Pro 160 165 GGT GCA TCC AAC CAA GIGA TCC GGG GTC CCT GC Gly Ala Ser Asn Gin Gly Ser Gly Val Pro Aia 175 180 COG TCT COG ACA GAC TTC AGC CTC AAC ATC CAT Gly Ser Gly Thr App Phe Ser Lea An Ile His 190 195 GAT ACT GCA ATG TAT TTC TGT CAG CAA AGT AAG Asp Thr Ala Met Tyr Phe Cys Gin Gin Ser Lys 205 210 215 TTC GOT GOT GGG ACC AAG CTG GAA ATA AAA OGG Phe Gly Ala Gly Th~r Lys Lau Gu Ile Lys Arg 225 230 339 (2) Asp Gin Gly Lys Arg Pro Giu INFORMATION FOR SEQ ID NO: 6: SEQUENCE CHARACTERISTICS: LENGTH: 113 amino acids TYPE: amino acid TOPOLOGY: linear (ii) MOLECULE TYPE: protein (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 6: Ile Giu. Lea Thr Gin Ser Pro Ala Ser Leu Ala Val 5 10 Arg Ala Thr Ile Ser Cys Arq Ala Ser Glu Ser Val 25 Ile Ser Phe Met Asn Trp Phe Gin Gin Lys Pro Gly 40 45 Leu Leu Ile Tyr Gly Ala Ser Asn Gin Gly Ser Gly 50 55 60 Phe Sec Gly Ser Gly Ser Gly Thr Asp Phe Ser Leu 70 75 Leu 'Glu Giu Asp Asp Thr Ala Met Tyr Phe Cys Gin 90 Val Pro Leu Thr Phe Gly Ala Gly Thr Lys Lea Gbu 100 105 Leu. Gly Asn Phe Pro Pro Pro Ala Ile His Ser Lys Lys Arg Ala INFORMATION FOR SEQ ID NO: 7: SEQIUENCE CHARACTERISTICS: LENGTH: 357 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear 49 (ii) (vi) (vii) (ix)

CAG

TTA

Leu 130

TGG

Trp

GGA

Gly

AAG

Lys

ATG

Met

GCA

Ala 210

ACC

Thr (xi)

GTG

Val 115

GTG

Val

ATG

Met

GAG

Glu

GGC

Gly

CAG

Gin 195

AGA

Arg

ACG

Thr MOLECULE TYPE: CDNA HYPOTHETICAL: NO ANTI-SENSE: NO FRAGMENT TYPE: N-terminal ORIGINAL SOURCE: ORGANISM: mouse STRAIN: Baib/c DEVELOPMENTAL STAGE: adult TISSUE TYPE: Lymph node IM1MEDIATE SOURCE: (13) CLONE! L2 12B (heavy chain)

FEATURE:

NAMEIKEY:

CDS

LOCATION:..35 7 S4QUENCE DESCRIPTION: SEQ ID NO: 7: CAG CTG CAG GAG TCT GGA CCT GAG CTG GTG Gin Leu Gin Gin Ser Gly Pro Glu Leu Val 120 125 AAG ATA TCC TGC AAG GOT TOT GGT TAC ACC Lys Ile Ser Cys Lys Ala Ser Gly Tyr Thr 135 140 CAC TGG GTG AAG CAG AGG CCT GGA CAA G43C His Trp Val Lys Gin Arg Pro Gly Gin Gly 150 155 ATT GAT COT TCT GAT AGT TAT ACT AAC TAC Ile Asp Pro Ser Asp Ser Tyr Thr Asn Tyr 165 170 AAG GCC ACA TTG ACT GTA GAC AAA TCC TCC Lys Ala Thr Leu Thr Val Asp Lye Ser Ser 180 185 GIG AGC AGC GIG ACA TCT GAG GAG TCT GCG Leu Ser Ser Len Thr Ser Glu Asp Ser Ala 200 205 TCG GAC TAG GGT AGI AGC CAC TTT GAC TAG Ser Asp Tyr Giy Ser Ser His Phe Asp Tyr 21S 220 GTC ACC GTC TCC TCA Val Thr Val Ser Ser 230 CCT GGG OCT Pro Gly Ala ACC AGC TAG Thr Ser Tyr 145 GAG TOG ATC Giu Trp Ile 160 CAA JAAG TIC Gin Lys Phe 175 ACA GCC TAC Thr Ala Tyr TAT TAC TGT Tyr Tyr Cys GGC CAA G Gly Gin Gly 225 INFORM4ATION FOR SEQ ID NO: 8: SEQUENCE CHARACTERISTICS: LENGTH: 119 amino acids TYPE: amino acid TOPOLOGY: linear 11: I 50 (ii) MOLECULE TYPE: protein (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 8: Gin Val Gin Leu Gin G1u Ser Gly Pro Glu Leu Val Lys Pro Gly Al 1 5 10 15 Lau V~al Lys Tie Ser Cys Lys Ala Ser Gly Tyr Thr Phe Thr Ser Ty 20 25 30 Trp Met His Trp Val Lys GIn Arg Pro Gly Gin Gly Leu Glu Trp Ii 40 Gly Glu Ile Asp Pro Ser Asp Ser Tyr Thr Asn Tyr Aan Gin Lys Ph 55 Lys Gly Lys Ala Thr Leu Thr Val Asp Lys Ser Sar AEn Thr Ala Ty to 65 70 75 8a Met Gin Leu Ser Ser Leu Thr Ser Glu Asp Ser Ala Val Tyr Tyr Cy 90 Ala Arg Ser Asp Tyr Gly Ser Ser His Phe Asp Tyr Trp Gly Gin G1I 100 105 110 Thr Thr Val Thr Val Ser Ser 115 INFORMATION FOR SEQ ID NO: 9: SEQUENCE CHARACTERISTICS: LENGTH: 339 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA (iii) HYPOTHETICAL: NO (1v) ANTI-SENSE: NO FRAGMENT TYPE: N-terminal 25 (vi) ORIGINAL SOURCE: ORGANISM: mouse STRAIN: Balb/c DEVELOPMENTAL STAGE: adult TISSUE TYPE: Lymph node (vii) IMMEDIATE SOURCE: CLONE: L3 110 (light chain) 30 (ix) FEATURE: NAME/KEY:

CDS

LOCATiON:l. .339 (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 9: GAC ATT GAG CTC ACC CAG TCT CCA GCT TOT TTG GCT GIG TOT CT?. GGG Asp Ile Glu Leu Thr Gin Ser Pro Ala Ser Leu Ala VJal Ser Leu Gly 120 125 130 135 CAG AGG GCC ACC ATC TCC TGC 00?. 0CC AGO GA?. AGT GTT GAT A-AT TTT Gin Arg Ala Thr Ile Ser Cys Arg Ala Ser Giu Ser Val Asp Asn Phe 140 14S 150 a r e e r -51- GGC ATT ACT TTT ATG AAC TG, TTC CAA CAG AAA CCA GGA CAG CCA CCC 144 GLy Ile Ser Phe Met Asn Trp Phe Gin Gin Lys Pro Gly Gin Pro Pro 155 160 165 AAA CTC CTC ATC TAT GGT GCA TCC AAC CAA GCGA TCC GCG GTC CCT GCC 192 Lye Leu Leu Ile Tyr Gly Ala Ser Aen Gin Gly Ser Gly Val Pro Ala 170 175 180 AGG TTT AGT GGC AGT GCG TCT GGC ACA GAC TTC AGC CTC AAC ATC CAT 240 Arg Phe Ser Gly Ser Gly Ser Gly Thr Asp Phe Ser Leu Asn Ile His 185 190 195 CCT TTG GAG GAG GAT GAT ACT GCA ATG TAT TIC TGT CAG CAA ACT AAG 288 Pro Leu Glu Glu Asp Asp Thr Ala Met Tyr Phe Cye Gin Gin Ser Lye 200 205 210 215 GAG GTT CCG CTC ACG TIC OCT GCT GGG ACC AAG CTC GAG CTG AAA COG 336 Glu Val Pro Leu Thr Phe Gly Ala Gly Thr Lys Leu Glu Leu Lye Arg 220 225 230 GCG 339 Ala INFORMATION FOR SEQ ID NO: SEQUENCE CHARACTERISTICS: LENGTH: 113 amino acids TYPE: amino acid TOPOLOGY: linear (ii) MOLECULE TYPE: protein (xi) SEQUENCE DESCRIPTION: SEQ ID NO: Asp Ile Giu Leu Thr Gin Ser Pro Ala Ser Leu Ala Val Ser Leu Cly 1 5 10 Gin Arg Aia Thr lie Ser Cys Arg Ala Ser Giu Ser Val Asp Asn Phe 25 Gly Ile Ser Phe Met Aen Trp Phe Gin Gin Lys Pro Gly Gin Pro Pro 35 Lys Leu Leu lie Tyr Gly Ala Ser Asn Gin Gly Ser Gly Val Pro Ala so 55 60 Arg Phe Ser Gly Ser Gly Ser Giy Thr Asp Phe Ser Leu Asn Ile His 70 75 Pro Leu Giu Giu Asp Asp Thr Ala Met Tyr Phe Cys Gin Gin Ser Lys 85 90 Glu Val Pro Leu Thr Phe Gly Ala Gly Thr Lys Leu Giu Leu Lys Arg 100 105 110 Ala INFORMATION FOR SEQ ID NO: 11: SE~QUENCE CHARACTERISTICS: LENGTH: 357 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear MOLECULE TYPE: cDNA (iii) HYPOTHETICAL; NO (iv) ANTI-SENSE: NO FRAGMENT TYPE, N--terminal (vi) ORIGINAL SOURCE: ORGANISM: mouse STRAIN: Balb/c DEVELOPIMENTAL STAGE, adult TISSUE TYPE: Lymph node (vii) IMMEDIATE SOURCE: CLONE: L3 11D (heavy chain) (i)FEATURE: NAME/KEY: CDS LOCATION:l. .357 (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 11: GAG OTO Glu Val 115 TCA GTG Ser Val 130 CAG CTG CZAG CAC TCA CCC GCT G-AG CTT Gln Leu Gin Gln Ser Gly Ala Glu Leu 120 GTG AAG Val Lys 125 CCT GGG CCT Pro Gly Ala AAG OTG TOO Lys Leu Ser TGC PACG GCT TCT GGC TAG ACC TTC ACC AGC TAG Cys Lys Ala Ser Gly Tyr Thr Phe Thr Ser Tyr 135 140 145 TOG ATG CAC TGG GTG AAG CAC AGG cCT GGA CAA CCC CTT GAG TGG ATO Trp Met His Trp Val Lye Gln Arg Pro GlY Gin Gly Leu Glu Trp, Ile 150 155 160 OGA. GAG ATT Gly C~u Ile OCT TOT GAT AGT Pro Ser Asp Ser ACT AAC TAC AAT Thr Asn Tyr Aen CAA AO TTC Gli, Lys Phe 175 liOA GCC TAO W;_hr Ala Tyr AAG GGC AAG CCC ACA ITT ACT GIA Lys GiY Lys Ala Thr Leu Thr Val 185 GAO AAA ICC ICC Asp Lys Ser Ser ATG CAG Met Gin -3195 CTO AGO AGO CTG Leu Ser Ser Leu.

TCT GAG GAO TOT Ser Glu Asp Ser GTC TAT TAC IT Val Tyr Tyr Cys GCA AGA TCG GAC TACGGOT ACT AGC CAC ITT GAG TAO TGG CCC CAA G Ale Arg Ser Asp Tyr Oly Ser Ser His Phe Asp Tyr Trp Cly Gln Gly 210 215 220 225 ACC ACG GTC ACC Thr Thr Val Thr GTG TOO TCA 357 Val Ser Ser 230 -53 INFORMATION FOR SEQ ID NO: 12: SEQUENCE CHARACTERISTICS: LENGTH: 119 amino acids TYPE: amino acid TOPOLOGY: linear (ii) MOLECULE TYPE: protein (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 12: Glu Val Gln Leu Gin Gin Ser Gly Ala Glu Leu Val Lys Pro Gly Ala 1 5 10 Ser Val Lys Leu Ser Cys Lys Ala Ser Gly Tyr Thr Phe Thr Ser Tyr 25 Trp Met His Trp Val Lys Gin Arg Pro Gly Gin Gly Leu Glu Trp Ile 35 40 Gly Glu Ile Asp Pro Ser Asp Ser Tyr Thr Asn Tyr Asn Gin Lys Phe 55 Lys Gly Lys Ala Thr Leu Thr Val Asp Lys Ser Ser Ser Thr Ala Tyr 70 75 Met Gin Leu Ser Ser Leu Thr Ser Glu Asp Ser Ala Val Tyr Tyr Cys 1 85 90 Ala Arg Ser Asp Tyr Gly Ser Ser His Phe Asp Tyr Trp Gly Gin Gly 100 105 110 Thr Thr Val Thr Val Ser Ser 115 20 INFORMATION FOR SEQ ID NO: 13: SEQUENCE CHARACTERISTICS: LENGTH: 327 base pairs TYPE: nucleic acid S* STRANDEDNESS: single TOPOLOGY: linear S(ii) MOLECULE TYPE: cDNA 25 (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE:

NO

FRAGMENT TYPE: N-terminal (vi) ORIGINAL SOURCE: ORGANISM: mouse STRAIN: Balb/c DEVELOPMENTAL STAGE: adult TISSUE TYPE: Lymph node (vii) IMMEDIATE

SOURCE:

CLONE: S4 20 (light chain) (ix) FEATURE: NAME/KEY: CDS 4 LOCATION:1. .327 MIN 11 54 a

C

C

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 13: GAO ATT GAG cT0 AiCC CAG TCT CCA ACC0 ACC ATG GOT GCA TOT Asp lie Glu Leu Thr Gin Ser Pro Thr Thr Met Ala Ala Ser 120 125 130 GAG ARC ATC ACT ATC ACC IGC AGT GCC AGC TCA AGT ATA ACT Giu Lys Ile Th r lie Thr Cys Ser Ala Ser Ser Ser Ile Ser 140 145 TAC TIGCOAT TGG TAT CAG CAG AAG CCA GGA TTC TCC CCT AAA Tyr Leu His Trp Tyr Gin Gin Lys Pro Giy Phe Ser Pro Lys 155 160 165 ATT TAT AGG ACA TOO AAT CTC GOT TOT GGP. GTO OCA GOT 0CC Ile Tyr Ary Thr Ser Asn Leu Ala Ser Cly Val Pro Ala Arg 170 175 180 GGC AGI GGC TOT CCC ACC TCT TAO TOT 010 ACA ATT GGC ACC Gly Ser Cly Ser Gly Ihr Ser Tyr Ser Leu Thr Ile Gly Thr 185 190 195 GCT GAA GAT CTT CO ACT TAO TAC IC CAG CAG GGT AGT ACT Ala Gilu Asp Val Ala Thr Tyr Tyr Cys Gin Gin CGly Ser Ser 15 100 205 210 CGC AOG ITO GGA CCC CCC ACC AC CTC GAP. ATC AAA CGG Arg Thr Phe Cly Gly Gly Thr Lys Leu Ciu Ile Lys Arg 220 225 INFORMATION FOR SEQ ID NO: 14: SEQUENCE

CHARACTERISTICS:

LENGTH: 109 amino acid~s TYPE: amino acid TOPOLOGY: linear (ii) MOLECULE TYPE: protein (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 14: Asp Ile Glu Leu Thr Gin Ser Pro Thr Thr Met Ala Ala Scr 1 5 Ciu Lys Ile Thr Ile Ihr Cys Ser Ala Ser Ser Ser Ile Ser 20 25 Tyr Leu His Trp Tyr Gin Gin Lys Pro Gly Phe Ser Pro Lys 40 Ile Tyr Arg Thr Ser Asn Leu Ala Ser Gly Val Pro Ala Arg so55 Gly Ser Gly Sec GIy Thr Ser Tyr Ser Leu Thr Ile Gly Thr 70 Aia Giu Asp Val Ala Thr Tyr Tyr Cys Gin Gin Gly Ser Ser 85 Arg Thr Phe Gly Gly Gly Thr Lys Leu Glu l Te Lye Ara 100 105 CC0 GGG Pro Gly 13S TOO ART Ser Asn IS0 010 TIC Leu Leu TITC ACT Phe Ser ATG GAG Met Glu ATA OCA Ile Pro 215 Pro Gly 15 Ser Asn Leu Leu Phe Sec Met Giu Ile Pro M 71 11 )1 I 55 INFORMATION FOR SEQ ID NO: SEQUENCE CHARACTERISTICS: LENGTH: 354 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear MOLECULE TYPE: cDHA (iii) HYPOTHETICAL: NO (1v) ANITI-SENSE: NO FRAGMENT TYPE: N-terminal (vi) ORIGINAL SOURCE: ORGANISM: mouse STRAIN: Balb/c DEVELOPMENTAL STAGE: adult (Fv TISSUE TYPE: Lymph node (vii) IMMEDIATE SOURCE: CLONE: S4 2D (heavy chain) 15 (ix) FEATURE: NAME/KEY: CDS LOCATION:1. .354 (xi) SEQUENCE DESCRIPTION: SEQ ID NO: GAG GTC AAG CTG GAG CAG TCA GGA COT GAG CTG CTA Giu Val Lys Leu Gin Gin Ser Cly Pro Glu Leu Val -n 110 115 120 a.

0e a AAG CCT GGG OCT Lys Pro Gly Ala 125 TCA GTC AAG ATG Ser Val Lys Met TCC TGC Ser Cys 130 AAC GCT TCT GGA Lys Ala Ser Gly 13S TAC GCA TTC ATA ACT TTT Tyr Ala Phe Ile 5cr Phe 140 GTT ATG CAC TOG GTO AAG CAG AAG OCT GGO Val Met His Trp Val Lys Gin Lys Pro Oly 145 150 CAG 0CC CTT Gin Gly Leu GAG TGG ATT Giu Trp Ile 155 GAG AAG TTC Clu Lys Phe GGA TTT ATT Cly Phe Ile 160 AAT CCT TP.C AAT Asn Pro Tyr An COT ACT AAG Cly Thr Lys TAC AAT Tyr An 170 AAA GAC Lys Asp 17S AAG CCC ACA CTG Lys Ala Thr Leu TCA GAO AAA TCC Ser Asp Lys Ser AGC ACA CCC TAC Ser Thr Ala Tyr

ATG

Met 190 GAG CTC AGO AGO Glu Leu Ser Ser ACC TCT GAG GAO Thr Ser Glu Asp C GTC TAT TAC Ala Val Tyr Tyr CCA ACT GCC Ala Ser Gly CAT TAC Asp Tyr 210 GAO AGO GCT ATG Asp Arg Ala Met CAC TAC TGG CCC CAA COG ACC Asp Tyr Trp Cly Cmn Cly Thr 215 220 ACG OTO ACC Thr Val Thr GTC TOO TCA Val Scr Fcr 225 1 101) I V t It 't 0 I 56 INFORMATION FOR SEQ ID NO: 16: SEQUENCE CHRACTERISTICS: LENGTH: 118 amino acids TYPE: amino acid TOPOLOGY: linear (ii) MOLECULE TYPE: protein SEQUENCE DESCRIPTION: SEQ ID NO: 16: Glu Val Lys Leu Gin Gin Ser Gly Pro Glu Leu Val Lys 1 5 10 Ser Val Lys Met Ser Cys Lys Ala Ser Gly Tyr Ala Phe 25 Val Met His Trp, Val Lye Gin Lys Pro Gly Gin Gly Leu ~35405 Gly Phe Ile Asn Pro Tyr Asn Asp Gly Thr Lys Tyr Asn 5E60 Lye Asp Lys Ala Thr Leu. Thr Ser Asp Lys Ser Ser Ser 70 75 Pro Gly Ala Ile Ser Phe Glu Trp Ile Glu Lys Phe Thr Ala Tyr Met Glu Leu Ser Ser Leu Thr Ser Glu Asp Ser Ala Val Tyr Tyr Cys 85 90 Ala Ser Gly Asp Tyr Asp Arg Ala Met Asp Tyr Trp Gly Gin Gly Thr 100 105 110 Thr Val Thr Val Ser Ser 115 INFORMATION FOR SEQ ID NO: 17: SEQUENCE CHARACTERISTICS: LENGTH: 717 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY, linear (ii) MOLECULE TYPE: cDNA (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO FRAGMENT TYPE: N-terminal (vi) ORIGINAL SOURCE: ORGANISM: mouse STRAIN: Balbfc DEVELOPMENTAL STAGE: adult TISSUE TYPE: splenocytes (vii) IMMEDIATE SOURCE: CLONE: 4 B 2 (ix) FEATURE: NAME/KEY: CDS LOCATION:1. .717 57 (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 17: GAG GTG AAG CTG CAG GAG TCT GC GGA GAC TTA GIG AAG CCT GCA GGG 48 Glu Val Lys Leu GIn Glu Ser Gly Gly Asp Leu Val Lys Pro Gly Gly 120 125 130 TCC CTG AAA CTC TCC TOT GCA GCC TOT GGA TTC ACT TIC AGT AGC TAT 96 Ser Leu Lys Leu Ser Cys Ala Ala Ser Gly Phe Thr Phe Ser Ser Tyr 135 140 145

ISO

GGC ATG TOT TOG OTT COG CAG ACT CCA GAC AAG AGO 010 GAG TCT GTO 144 Gly Met Ser Trp, Val Arg Gin Thr Pro Asp Lys Arg Leu Giu Ser Val 155 160 165 GCA ACC ATT AGT AGT GOT GOT OCT TAO ATC TAC TAT CCA GAC AGT GTG 192 Ala Thr Ile Ser Ser Gly Gly Ala Tyr Ile Tyr Tyr Pro Asp Ser Val 170 175 180 AAG 000 COA TIC ACC ATC TCO AGA GAO AAI GCC AAG MOC ACO 010 TAO 240 Lys Gly Arg Phe Thr Ile Ser Arg Asp Asn Ala Lys Asn Thr Leu Tyr 185 190 195 CTG CAA ATG AGO AGT CTG AMG TOT GAG GAO ACA GCC ATG TAT TAO TOT 288 Leu Gin Met Ser Ser Leu Lys Ser Giu Asp Thr Ala Met Tyr Tyr Cys *..200 205 210 15 G CA AGA OTT OAA ACC G GAC TAT OCT TTG GAC TAO TOG GCC CAA GOG 336 Ala Ara Leu Glu Thr Gly Asp Tyr Ala Leu Asp Tyr Trp Gly Gin Gly 215 220 225 230 .ACC ACO OTO ACC GTO TOO TOA GOT GGC GOT GGC TOO 000 GOT GOT GGG 384 Thr Thr Val Thr Val Ser Ser Gly Gly Gly Oly Ser Gly Gly Oly Gly 235 240 245 TCO GOT 000 000 GOA TOT GAO ATT GAG 010 ACC CAG TOT CCA OCT TOT 432 Ser Gly Gly Gly Oly Ser Asp Ile Giu Leu Thr Gin Ser Pro Ala Ser 250 255 260 TTO GOT GTO TOT OTA 000 CAG AGO GCC ACC ATA ITO TGC AAG GAO AGO 480 Leu Ala Val Ser Leu Gly Gin Arg Ala Thr Ile Phe Cys Lys Asp Ser o265 270 275 CAA AGT OTT OAT TAT OAT CT OAT AGT TAT ATO MOC 100 TAO CAA CAG 528 Gin Ser Val Asp Tyr Asp Gly Asp Ser Tyr Met Aan Trp Tyr Gin Gin 280 285 290 *AAA CCA GGA CAG OCA CCO AAA CTC CTC ATO TAT GOT CGA ICC MAT CTA 576 Lys Pro Gly Gin Pro Pro Lye Leu Leu Ile Tyr Ala Arg Ser Asn Leu 295 300 305 310 GMA TOT 000 010 OCT 000 AGO ITT AGT GGC ACT GGG TCT GGG ACA GAO 624 Olu Ser Gly Vai Pro Ala Arg Phe Ser Gly Ser Gly Ser Gly Thr Asp 315 320 325 TIC AGO OTC AAC ATC CAT OCT OTG GAG GAG OAT GAT ATT OCA ATO TAT 672 Phe Ser Leu Asn Ile His Pro Val Glu Giu Asp Asp Ile Ala Met Tyr 330 335 340 TTC TOT CAG CAA AC-I AGO AAG OTT COG TOG TOG TIC GGT GGA COG 717 Phe Oys Gin Gin Ser Arg Lye Val Pro Trp Ser Phe Gly Gly Oly 345 350 355 -58- INFORMATION FOR SEQ ID NO: 18: SEQUENCE CHARACTERISTICS: LENGTH: 239 amino acids TYPE: amino acid TOPOLOGY: linear o o (ii) MOLECULE TYPE: protein (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 18: Glu Val Lys Leu Gin Glu Ser Gly Gly Asp Leu Val 1 5 10 Ser Leu Lys Leu Ser Cys Ala Ala Ser Gly Phe Thr 25 SGly Met Ser Trp Val Arg Gin Thr Pro Asp Lys Arg 35 Ala Thr Ile Ser Ser Gly Gly Ala Tyr Ile Tyr Tyr 55 Lys Gly Arg Phe Thr Ile Ser Arg Asp Asn Ala Lys 70 75 SLeu Gin Met Ser Ser Leu Lys Ser Glu Asp Thr Ala 85 90 Ala Arg Leu Glu Thr Gly Asp Tyr Ala Leu Asp Tyr 100 105 Thr Thr Val Thr Val Ser Ser Gly Gly Gly Gly Ser 115 120 Ser Gly Gly Gly Gly Ser Asp Ile Glu Leu Thr Gin 130 135 140 Leu Ala Val Ser Leu Gly Gin Arg Ala Thr Ile Phe 145 150 155 Gin Ser Val Asp Tyr Asp Gly Asp Ser Tyr Met Asn 165 170 Lys Pro Gly Gin Pro Pro Lye Leu Leu Ile Tyr Ala 180 185 Glu Ser Gly Val Pro Ala Arg Phe Ser Gly Ser Gly 195 200 Phe Ser Leu Asn Ile His Pro Val Glu Glu Asp Asp 210 215 220 Phe Cys Gin Gin Ser Arg Lys Val Pro Trp Ser Phe I 225 230 235 INFORMATION FOR SEQ ID NO: 19: SEQUENCE CHARACTERISTICS: LENGTH: 732 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA Pro Gly Gly Ser Ser Tyr Glu Ser Val Asp Ser Val Thr Leu Tyr Tyr Tyr Cys Gly Gin Gly 110 Gly Gly Gly Pro Ala Ser Lys Asp Ser 160 Tyr Gin Gin 175 Ser Asn Leu 190 Gly Thr Asp Ala Met Tyr Gly Gly I~ 1 4 59- (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO (vi) ORIGINAL SOURCE: ORGANISM; mouse STRAIN: Balb/c TISSUE TYPE: splenocytea (vii) IMMEDIATE SOURCE: CLONE: 10 D 2 (single-chain Fv, heavy and light chain plus linker) (ix) FEATURE: NAME/KEY: GUS LOCATION:1. .732 (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 19: GAG GTG GAG Glu Val Gin 240 TCA GTG AAG Ser Val Lys TGG ATG GAC Trp Met His GGA GAG ITT Gly Glu Phe 290 AAG AGC AAG Lys Ser Lys 305 ATG CAA CTC Met Gln Leu 320 0CC ACT G Ala Ser Arg CAA COG ACC Gin Cly Thr COT COG TCG Gly Gly Ser 370 OCA ATC ATG Ala Ile Met Leu Gin Gin Ser 245 TTG TCC TGC AAG Lou Ser Cys Lyn 260 TOG GTG AAG GAG Trp Val Lys Gin 275 AAT CCC AGC AAC Ann Pro Ser Ann CCC ACA CTG ACT Ala Thr Leu Thr 310 AGC AGO CTO ACA Ser Ser Leu Thr 325 GAG TAT GAT TAC Asp Tyr Asp Tyr 340 ACG GTC ACC GTC Thr Val Thr Val 355 CGT CCC COC GGA Gly Gly Gly Gly TCT GCA TCT GCA Ser Ala Ser Pro 390 Cly Ala Glu Leu 250 GCT TCC CCC TAG Ala Ser Cly Tyr 265 AGO GCT GGA CAA Arg Ala OWy Gin 280 CCC COT ACT AAC Gly Arg Thr An 29S GTA GAG AAA TCG Val Asp Lys Ser TCT GAG GAG TCT Ser Clu Asp Ser 330 CAC COA COG TAG Asp Cly Arg Tyr 345 TCC TCA COT GC Ser Ser Cly Oly 360 TCT GAG ATT GAG Ser Asp Ile Glu 375 GGG GAG AAG GTG Cly Ciu Lys Val ATG TAC Tf'C TAC Met Tyr Trp Tyr 410 Val Lys Pro ACC TTC ACC Thr Phe Thr GGC CTT GAG Gly Leu Glu 285 TAG AAT GAG Tyr Asn Glu 300 TCC AGC ACA Ser Ser Thr 315 CC GTC TAT Ala Val Tyr TTT GAC TAC Phe Asp Tyr COT GGC TCG Gly Gly Ser 365 cTc ACC GAG Leu Thr Gin 380 ACC ATG ACC Thr Met Thr 395 CAG GAG AAA Gin Gin Lys CTG CAG GAG TCT GGG GCT GAA CTG GTG ALAG, CCT COG GCT Oly Ala 255 AGC CAC Ser His 270 TGG ATC Trp Ile AAA TTC Lys Plie CCC TAG Ala Tyr TAG TOT Tyr Cys 335 TOG GC Trp Cly 350 GCC COT Oly Gly TCT GCA Ser Pro TCC ACT Gys Ser CCA CGA Pro Cly 415 CCC AC Ala Ser 400 TCA AGT GTA ACT TAG Ser Ser Val Ser l'yr 405 I'll NI I C I 1 ;1 111 1, 60

TCC

Ser

GTC

Val

ACA

Thr

CAG

Gln

CTG

Leu 480 (2) TCC CCC AGA CTC CTG ATT TAT Ser Pro Arg Leu Leu Ile Tyr 420 CCT GTT CGC TTC AGT CGC AGT Pro Val Arg Phe Ser Gly Ser 43S ATC AGC CGA ATG GAG GCT GAA Ile Ser Arg Met Giu Ala Glu 450 455 TGG AGT AGT TAC CCA CCC ATG Trp Ser Ser Tyr Pro Pro Met 465 470 GAA ATA AAA Glu Ile Lys GAC ACA TCC AAC CTG GCT TCT GGA Asp Thr Ser Asn Leu Ala Ser Gly 425 430 GGG TCT GCG ACC TCT TAC TCT CTC Gly Ser Gly Thr Ser Tyr Ser Leu 440 445 CAT GCT GCC ACT TAT TAC TGC CAG Asp Ala Ala Thr Tyr Tyr Cys Gin 460 TAC AcG TTC CGA CCC CCC ACC AAG Tyr Thr Phe Gly Gly Gly Thr Lys 475 INFORMATION FOR SEQ ID NO: SEQUENCE CHARACTERISTICS: LENGTH: 244 amino acids TYPE: amino acid TOPOLOGY: linear (ii) MOLECULE TYPE: protein (xi) SEQUENCE DlESCRIPTION: SEQ ID NO: Glu Val Gin Leu Gin Gln Ser Giy Ala Glu Lau Val 1 Ser Val Lys Leu Trp Met His Trp Gly Glu Phe Asn Lys Ser Lys Ala Met Gin Leu Ser Ala Ser Arg Asp 100 Gin Giy Thr Thr 115 Giy Gly Ser Gly 130 Ala Ile Met Ser 145 Ala Ser Ser Ser 10 Aia Ser Giy 25 Arg Aia Cly Gly Arg Thr Val Asp Lys Ser Giu Asp 90 Asp Cly Arg 105 Ser Ser Cly 120 Ser Asp Ile Gly Glu Lys Met Tyr Trp 170 Pro Gly Thr Ser Giu Trp, Giu Lys Thr Ala Tyr Tyr Tyr Trp 110 Ser GLy Gin Ser Thr Cys Ser Tyr Tyr Gin Gin Lys Pro Gly 175

I,

-61 Ser Ser Pro Arg Leu Leu Ile Tyr Asp Thr Ser PAsn Leu Ala Ser Gly 180 185 190 Val Pro Val Arg Phe Ser Gly Ser Gly Ser Gly Thr Ser Tyr Ser Leu 195 200 205 Thr Ile Ser Arg Met Glu Ala Glu Asp Ala Ala Thr Tyr Tyr Cys Gin 5210 215 220 Gin Trp Ser Ser Tyr Pro Pro Met Tyr Tbr Phe Gly Gly Gly Thr Lys 225 230 2.35 240 Leu. Gu Ile Lys INFORMATION FOR SEQ ID NO: 21: t0 SEQUENCE CHARACTERISTICS: LENGTH: 732 base pairs TYPE: nucleic acid STRANDEDNESS: eincile TOPOLO)GY: linear (ii) MOLECULE TYPE: CDNA (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO FRAGMENT TYPE: N-terminal (vi) ORIGINAL SOURCE: ORGANISM: mouse STRAIN: Balb/c TISSUE TYPE: splenocytes (vii) IMMEDIATE SOURCE: CLONE: 3 D 3 (single-chain Fv, heavy and light chain plus linker) (ix) FEATURE: NAME/KEY: CDS LOCATION:1. .732 (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 21: GAG GTC CAA CTG CAG GAG TCA G OCT GAA.CTG GIG AAG CCT GGG GGT Giu Val Gln Leu Gin Gin Ser Gly Ala Glu Leu Val Lys Pro Gly Ala 245 250 255 260 TCA GTG AAG TTG TCG TGC AAG GCT ICC GGC TAG ACC TTC ACC AGC CAG Ser Val Lys Leu Ser Cys Lys Ala Ser Gly Tyr Thr Phe Thr Ser His 265 270 275 TGG ATC GAG ICC- GTG AAG GAG AGG GCT GGA CAA GGC CTT GAG TOG ATG Trp Met His Trp Val Lys Gin Arg Ala Gly Gin Gly Leu Glu Trp Ile 280 285 290 GGA GAG ITT AAT CCC AG AACGCG CGT ACT AAC TAG AAT GAG AAA ATC I Gly Glu Phe Asn Pro Ser Asn Gly Axg Thr Aen Tyr Asn Glu Lye Ile 295 300 305 ALAG AGC AAG 0CC ACA GIG ACT GTA GAC AAA TGC TCC AGC ACA OCG TAG 2 Lye Ser Lye Ala Thr Leu Thr Val Asp Lys Sec Ser Ser Thr Ala Tyr 310 315 320 92 U' -62- ATG CAA CTC AGC AGO CTG ACRA TCT GAG Met Gin Leu Ser Ser Leu Thr Ser Glu 325 330 CCC ACT C

GAC

Ala

CAA

Gin

GGT

Gly A CA Thr

GAC

Aep 405

TCC

Ser

GTC

ValI

ACA

Thr

CAG

Gin

CTG

Leu 485 Ser Arg Asp GGG ACC A4CG Gly Thr Thr 360 GGG TOG COT Gly Ser Gly 375 ATC ATG T(T Ile Met Ser 390 AGC TCA AGT ser Sew' gar TC CCC AGA Ser Pro Arg CCT GTT CGC Pro Val Arg 440 ATC AGO CA Ile Ser Arq 455 TGG AGT AGT Trp Ser Ser 470 GAA ATA AAA Glu Ile Lys

GCC

Gly

GCA

Ala

GTA

Val

CTC

Lau 425

TTC

Phe

ATG

Met

TAC

Tyr

TAT

Tyr 345

GTC

Val CAT TAC GAO GGA Asp Tyr Asp Gly ACC GTC TCC TCA Thr Val Ser Ser 365 CCC GGA TOT GAC Gly Gly Ser Asp 380 TCT CCA CCC GAG Ser Pro Cly Giu 395 ACT TAC ATG TAC Ser Tyr Met Tyr 410 CTG ATT TAT GAC Lau Ile Tyr Asp ACT GGC AGT GCC Ser Giy Ser Gly 445 GAG GCT GAA GAT Glu Ala Giu Asp 460 CCA COO ATG TAO Pro Pro Met Tyr 475 GAO TOT CG GTC TAT TAC Asp Ser Ala Val Tyr Tyr 335 CGG TAO TTT GAO TAO TG Arg Tyr Phe Asp Tyr Trp 350 355 COT GGC CGT GGC TCG GGC Cly Cly Giy Cly Ser Giy 370 ATT GAG Ile Glu AAC CTC Lye Val TCG TAO Trp Tyr 415 ACA TC Thr Ser 430 TOT GGG Ser Giy CT GCC Ala Ala AOG TTC Thr Phe OTC ACC CAG TOT Leu Thr Gin Ser 385 ACC ATG ACC TOC Thr Met Thr Cy8 400 CAG CAG AAG ACA Gin Gln Lye Thr AAC CTG GCT TOT Asn Lau Ala Ser 435 ACC TCT TAO TOT Thr ser Tyr Ser 450 ACT TAT TAC TGO 'Th, Tyr Tyr Cys 465 GGA COG GGG ACO Gly Gly Gly Thr 480 INFORMATION FOR SEQ ID NO: 22: SEQUENCE CHARACTERISTICS: LENGTH: 244 amino acids TYPE: amino acid TOPOLOGY: linear (Li) MOLECULE TYPE: protein -3(xi) SEQUENCE DESCRIPTION1: SEQ ID NO: 22: Glu Val Gin Leu Gin Gln Ser Gly Ala Glu Leu Vai Lys Pro Gly Ala 1 5 .10 Sc~r Val Lvs Leu Ser Cys Lys Ala Ser Gly Tyr Thr Phe Thr Ser His 20 25 Trp Met His Trpj Val Lys Gin Arg Ala Gly Gin Cly Leu Giu Trp Ile 35 40 -63 Gly Glu 50O Lys Ser Met Gin Ala Ser Gin Gly Gly Gly 130 Thr Ile 145 Asp Ser Ser Ser Val Pro Thr Ile 210 Gln Trp 20 225 Asn Pro Ala Thr Set Ser Asp Tyr 100 Thr Val Gly Gly Ser Ala Ser Val 165 Arg Leu 180 Arg Phe Arg Met Ser Tyr Ser Asn 55 Leu Thr 70 Leu Thr Asp Tyr Thr Val Gly Gly 135 Ser Pro 150 Ser Tyr Leu Ile Ser Gly Glu Ala 215 Pro Pro 230 Gly Arg Val Asp Ser Glu Asp Gly 105 Ser Ser 120 Ser Asp Gly Glu Met Tyr Tyr Asp 185 Ser Gly 200 Glu Asp Met Tyr Asn Tyr Asn Ser Ser Ser Ser Ala Val Tyr Phe Asp Gly Gly Gly 125 Glu Leu Thr 140 Val Thr Met 155 Tyr Gin Gin Ser Asn Leu Gly Thr Ser 205 Ala Thr Tyr 220 Phe Gly Gly 235 -o r i Leu Glu Ile Lys INFORMATION FOR SEQ ID NO: 23: SEQUENCE CHARACTERISTICS: LENGTH: 738 base pairs 25 TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA (iii) "YPOTHETICAL: NO (iv) ANTI-SENSE: NO FRAGMENT TYPE: N-terminal (vi) ORIGINAL SOURCE: ORGANISM: mouse STRAIN: Balb/c TISSUE TYPE: splenocytes (vii) IMMEDIATE SOURCE: CLONE: 1 E 3 (single-chain Fv, heavy and light chain plus linker) I I, 1. 11 1. 11 1 64 (ix] FEATURE: NAMEfKEY: GDS LOCATIOl:l. .738 (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 23: GAG GTG CAG CTG GAG GAG TCT GGG GCT GAA CTG GTG AAG CCT GGG GCT Glu Val Gin Leu Gin Gin Ser Gly Ala Glu Leu Val Lys Pro Gly Ala 245 250 255 260 TCA GTG AJAG TTG Ser Val Lys Leu TCG TGC AAG GCT TCC GGC TAC ACC TTC ACC AGC CAC Ser Cys Lys Ala Ser Gly Tyr Thr Phe Thr Ser His 265 270 275 TOG ATG CAC TGG GTG AAG GAG Trp Met His Trp Val Lys Gin AGG OCT GGA CAA GGC CTT Arg Ala 285 GGC GT Gly Arg 300 Gly Gin Gly Leu ACT TAG TAG AAT Thr Asn Tyr Asn 305 GGA GAG TTT Gly Glu Phe 295 AAT CCC AGC AAC Asn Pro Ser Agn GAG TGG ATC Giu Trp Ile 290 GAG ARA TTC Giu Lys Phe AGA GCT TAG Thr Ala Tyr AAG AGC AAG GCIC ACA CTG ACT GTA GAC AAA TCC Lys Ser Lys Ala Thr LeU Thr Val Asp Lys Ser 310 315 TGC AGC Ser Ser 320 CAA CTG AGC Gin Leu Ser AGG GTG Ser Leti 330 TAT GAT Tyr Asp 345 ACA TGT GAG GAG Thr Ser Giu Asp GOG GTG TAT TAC Ala Val Tyr Tyr GGC AGT GGG GAG Ala Ser Arg Asp TAC GAG GGA GG TAG TTT GAG TAG Tyr Asp Gly Arq Tyr Phe Asp Tyr 350 TOG GG Trp Giy 355 CAA GGG ACC Gin C-iy Thr GTG AGC GTC TGG Val Thr Val Ser TGA GUT Ser Giy 365 GGG GGT GG Gly Gly Gly GUT GGG TGG GGT GG GGC GGA TGT GGA TCT GAC ATT GAG Gly Gly Ser Gly Gly Gly Gly Ser Gly Ser Asp le Giu 375 380 385 TOG GC GUT Ser Gly Gly 370 GTC ACG GAG Leu Thi Gin ACC ATU ACC Thr Met Thr TCT CCA Ser Pro 390 ACA ATG ATG Thr Ile Met '7V OC-A Ser Ala 395 ACT GTA Ser Val 410D TCT CGA GGGAG Ser Pro Gly Giu

GTG

Val1 ACT GAG AGGC TGA Ser Asp Ser Ser AGT TAG ATG Ser Tyr Met TGG TAG GAG GAG Trp Tyr Gin Gin GGA GGA TCC TCC Pro Gly Ser Ser AGA GIG GIG ATT Ara Leu Leu Tie TAT GAG Tyr Asp 430 AGA TCG AAG Thr Ser Asn 070 GCT Leu Ala 435 TGT GGA GTC Ser Gly Val TGT GTC ACA Ser Leu Thr 455 OTT CGC TTC AGT Val Arg Phe Ser ACT GGG TCT GG Ser Gly Ser Gly ATC AG CGA ATG Ile Ser Arg Met GGT GAA GAT GGT Ala Giu Asp Ala ACC TCT TAG Thr Ser Tyr 450 ACT TAT TAG Thr Tyr Tyr GGA COG GG Gly Gly Gly TGC CAG Cys Gin 470 GAG TGG ACT ACT Gin Trp Ser Ser GGA CCG ATG TAG Pro Pro Met Tyr AGG TTG Thr Phe 480 ACC AAG CTG GAA ATA AAA 738 Thr Lys Leu Glu Ile Lys 485 490 INFORMATION FOR SEQ ID NO: 24: SEQUENCE CHARACTERISTICS: LENGTH: 246 amino acids TYPE: amino acid TOPOLOGY: linear (ii) MOLECULE TYPE: protein (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 24: Val Gin Leu Gin Gin Ser Gly Ala Giu Leu Val Lys Pro Gly k.ia 15 10 Ser Val Lys Leu Set Cys Lys Ala Ser GIy Tyr Thr Phe Thr Ser His 25 Trp Met His Trp, Val Lys Gin Arg Ala Gly Gin Gly Leu Giu Trp Ile 40 Gly Glu Phe Asn Pro Ser Asn Gly Arg Thr Asn Tyr Asn Glu Lys Phe *Lys Ser Lys Ala Thr Leu Thr Val Asp Lye Set Ser Set Thr Ala Tyr 70 75 *Met Gin Leu Ser Ser Leu Thr Ser Glu Asp Ser Ala Val Tyr Tyr Cys 90 0 Ala Ser Arg Asp Tyr Asp Tyr Asp Gly Arg Tyr Phe Asp Tyr Trp Gly 2100 105 110 Gin Gly Thr Tiir Val Thr Val Set Set Gly Gly Gly Gly Ser Gly Gly *Gy 115 120 125 GyGly Ser Gly Gly Gly Gly Set Gly Set Asp Ile Glu Leu Thr Gin 130 135 140 25 Set Pro Thr Ile met Set Ala Set Pro Gly Giu Lys Val Thr Met Thr 145 150 155 160 Cys Set Asp Ser Set Set Val Set Tyr Met Tyr Trp Tyr Gin Gin Lys 165 170 175 pro Gly Set Set Pro Arg Leu Leu Ile Tyr Asp Thr Set Asn Leu Ala 180 185 190 set Gly Val Pro Val A-rg Phe Set Gly Ser Gly Ser Gly flit Set Tyr 195 200 205 ser Leu Thr Ie Set Arg Met Glu Ala Glu~ Asp Ala Ala Thr Tyr Tyr 210 215 220 Cys Gin Gin Trp Set Ser Tyr Pro Pro Met Tyr Thr Phe Gly Gly Gly 225 230 235 240 Thr Lys Leu G;lu Ilie Lys 245 INFORMATIONJ FOR SEQ ID NO: SEQUENCE CHARACTERISTICS: LENGTH: 726 base Fairs TYPE: nucleic acid STRA14DEDUESS: single TOPOLOGY: linear (ii) M4OLECUJLE TYPE: cDNA (iii) HYPOTHETICAL:

NO

ANTI-SENSE: NO FRAGMENT TYPE: N-terminal (vi) ORIGINAL SOURCE: ORGANISM: mouse STRAIN: Balb/c TISSUE TYPE: splenocy-tes (vii) IMMEDIATE SOURCE: CLONE: 5 F I (single-chain Fv, heavy, light chain, linker) (ix) FEATURE: NAMEjKEY: CDS LOCATION:l1 .726 SEQUENCE DESCRIPTION: SEQ ID NO: CAG c-TG AAA CTG CAG G-AG TCT GGC GCT GAA CTG GTG AAG CCT C-C c-CT Gin Val Lys Leu Gin c-lu Ser Gly Ala Glu Leu Val Lys Pro GIy Ala 250 255 260 TCA c-TO AAG Ser Val Lys 265 TTG TCC TGC AAG Leu Ser Cys Lys GCT TCC Ala Ser 270 c-CC TAC ACC TTC ACC Ac-C CAC c-by Tyr Thr Phe Thr Ser His 275 TGG ATG 2 5 Trp Met 280 CAC TOG GTG AAG His Trp Val Lys CAG Ac-C OCT c-CA CAA CCC CTT c-Ac- TOG ATC G-in Arq Ala c-by c-in c-Ly Leo Glu Trp Ile 285 290 GAG ATT AAT CCC c-lu Ile Asn Pro ACG c-Cc- CCT Thr Ala Pro ACT AAC TAC Thr Asn Tyr 305 AAT GAG AAA Asn Glu Lys 192 AAG AC-C AAG GCC ACA CTG ACT c-TA c-AC AAA TCC TCC Ac-C ACA c-CC TAC Lye Ser Lys ALa Thr Leo Thr Val Asp Lys Ser Ser Ser Thr Ala Tyr 315 320 32S ATG CAA CTC Ac-C Ac-C CTG ACA TCT c-Ac- c-AC TCT c-Cc- GTC TAT TAC TGT met c-in Leu Ser Ser Leo Thr Ser c-lu Asp Ser Ala Val Tyr Tyr CyS 330 335 340c-CC AGT COG c-AC TAT G-AT TAC c-AC GGA CGG TAC TIT c-AC Ala Ser Arg Asp Tyr Asp Tyr Asp Gly Arg Tyr Phe Asp 34S 350 355 TAC TGG GGC Tyr Trp Oly 336 CAA c-Cc c-in Gly 360 ACA ACG c-TC ACC Thr Thr Val Thr c-TO TOC ICA GY cGOC VaL Ser Ser Gly Gly 365 GGT c-cC TCG c-CC c-cc-by Gly Ser Gly Gly 370 I Nil,*[ li'l 4 67 CGT CGG TCG Gly Gly Ser 375 ACA ATC ATG Thr Ile Met GAC AGC TCA Asp Ser Ser TCC TCC CCC Ser Set Pro 425 GTC OCT GTT Val Pro Val 440 ACA ATO AGC Thr Ile Ser 455 CAG TGG AGT, Gin Trp Ser ATA AIA Ile Lys GGT GGC Gly Gly TCT GCA Ser Ala 395 ACT GTA Ser Val 410 AGA CTC Arg Leu CC TTC Arg Phe CGA ATG Arg Met AGT TAC Ser Tyr 475 GGC GOA TCT GAC ATT GAG Gly Gly Ser Asp Ile Glu 380 385 TCT CCA GGG GAG AAG GTC Ser Pro Gly Glu Lys Val 400 ACT TAO ACG TAC TGG TAC Set Tyr Thr Tyr Trp Tyr 415 CTG ATT TAT GAC ACA TCC Leu Ile Tyr Asp Thr Ser 430 ACT GGC ACT GGG TOT GCC Ser Gly Ser Gly Ser Gly 445 GAG GCT GAP. GAT GCT CC Giu Ala Giu Amp Ala Ala 460 465 CCG CTC ACG TTC GGT GCT Pro Lau Thr Phe Gly Ala 480 CTC ACC CAG Leu Thr Gin ACC ATG ACC Thr Met Thr CAG CAG AAG Gin Gin Lys 420 AAC CTG GOT Asn Leu Ala 435 ACC TCT TAC Thr Ser Tyr 450 ACT TAT TAC Thr Tyr Tyr GGG ACC AAC 1ly Thr Lys TOT CCA Ser Pro 390 TCC ACT Cys Ser 405 ACA GGA Thr Gly TOT CGA Ser Gly TCT CTC Ser Leu TGC CAC CyB Gin 470 CTG GAA Leu. Glu 485 432 480 528 576 624 672 720 INFORMATION FOR SEQ ID NO: 26: SEQUENCE CHARACTERISTICS: LENGTH: 242 amino acids TYPE: amino acid TOPOLOGY: linear (ii) MOLECULE TYPE: protein (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 26: Gin Val Lys Leu Gin Glu Ser Gly Ala Ciu Leu Val Lys 1 S 10 Ser Val Lys Leu Ser Cys Lys Ala Set Gly Tyr Thr Phe -25 Trp Het His Trp Val Lys Gin Arg Ala Gly Gin Gly Leu 40 Gly Giu Ile Asn Pro Arg Thr Ala Pro Thr Asn Tyr Asn 55 Lys Ser Lys Ala Thr Leu Thr Val Asp-Lys Ser Ser Ser 70 75 Met Gin Leu Set Set Leu Thr Set Glu Asp Set Ala Val 90 Ala Set Arg Asp Tyr Asp Tyr Asp Gly Arg Tyr Phe Asp 100 105 Pro Cly Ala Thr Ser His Giu Trp Ile C-lu Lys Phe Thr Ala Tyr Tyr Tyr Cys Tyr Trp Cly 68 Gin Gly Thr Thr Val Thr Val Ser Ser Gly Gly Gly Gly Set Gly Gly 115 120 125 Gly Gly Ser Gly Gly Gly Gly Ser Asp Ile Giu Leu Thr Gin Ser Pro 130 135 140 Thr Ile Met Ser Ala Ser Pro Gly Giu Lys Val Thr Met Thr Cys Ser 145 150 155 160 Asp Ser Ser Ser Val Ser Tyr Thr Tyr Trp Tyr Gin Gin Lys Thr Gly 165 170 175 Ser Ser Pro Arg Lou Leu Ile Tyr Asp Thr Ser Asn Leu Ala Ser Gly 180 185 190 Val Pro Val Arg Phe Ser Gly Ser Gly Ser Gly Thr Ser Tyr Ser Leu .195 200 205 T le Ser Arg Met Giu Ala Clu. Asp Ala Ala Thr Tyr Tyr Cys Gin 210 215 220 Gin Trp Ser Ser Tyr Pro Leu Thr Phe Gly Ala Gly Thr Lye Leu Glu 225 230 235 240 Ile Lys INFORMATION FOR SEQ ID NO: 27: SEQUENCE CHARACTERISTICS: LENGTH: 726 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (0t) MOLECULE TYPE: CDNA (iii) HYPOTHETICAL:

NO

0 (iv) ANTI-SENSE: NO FRAGMENT TYPE: N-terminal 25 (vi) ORIGINAL SOURCE: ORGANISM: wouse STRAIN: Balb/c TISSUE TYPE; splenocytes (vii) IMMEDIATE SOURCE: LIBRARY, 7 G 1 (single-chain F-j, heavy, light chain, linker) (iJx) FEATURE: NAME/KEY: CDS LOCATION:1. .726 (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 27: GAG GTC AAG CTG CAC CAG TCA GGG GCT GAA CTG GTG AAG CCT COG GCT 48 Glu Val Lye Leu Gin Gin Ser Gly Ala Glu Leu Val Lye Pro Gly Ala 245 250 255 3 5 TCA GTG AAG TTG TCC TGC AAG GCT TCC GGC TAC ACC TIC ACC AGC CAC 96 Ser Val Ly's Leu Ser Cys Lys Ala Ser Gly Tyr Thr Phe Ihr Ser His 260 265 270 I. II )t 69 TTG GAT CAC 7CC GTG AAG GAG AGG Leu Asp His Trp Val Lye Gin Arg 275 280 CGC TGG CAA GGC CTT GAG TGG ATC Gly Trp Gin Gly Leu Glu Trp Ile 285 GGA CAG TTT AAT Gly Gin Phe Asn AGO AAC GGC CGT Ser Asn Gly Arg ACT AAC Thr Asn 300 TAG AAT GAG Tyr Asn Glu AAA TTC Lys Phe 305 AAG AGC AAG GCC ACA CTG ACT GTA GAC AAA TCC TCC AGC ACA GCC TAC Lys Ser Lys Ala Thr Leu Thr Val Asp Lys Sec Sec Ser Thr Ala Tyr 310 315 320 ATC GAA CTC Ile Ciu Leu 325 CCC ACT CCC Ala Ser Arg 340

AGO

Ser AGC CTG ACA Sec Leu Thr TCT GAG Ser Glu 330 GAO TCC TCC GTC Asp Cys Ser Val 335 TAT TAG TOT Tyr Tyr Cys GAG TAT GAT TAC GAO GGA CCC TAG TTT GAG TAC TCG GCC Asp Tyr Asp Tyr Asp Gly Arg Tyr Phe Asp Tyr Trp, Cly 345 350 COG AGC ACC GTC Cly Thc Thr Val ACC CTC TOG TCA GOT CCC GOT GGC TCG Thr Val Sec Sec Gly Cly Cly Cly Sec 360 365s GGC GOT Gly Cly 370 288 336 384 432 480 52S GGT GGC TOG GT Giy Gly Ser Cly GGC GGC GGA TCT GAG ATT GAG CTC ACC GAG TOT OCA Gly Cly Gly Ser Asp Ile Ciu Leu Thr Gin Ser Pro 375 380 385 ACA ATC ATG TCT GCA TCT GCA COG GAG Thc Ile Met Ser Ala Ser Pro Gly Clu 390 395 ARG CTC AGO ATG Lys Val. Thr Met AGO TGC ACT Thr Cys Ser 400 AAG ACA GGA es Thc Cly GAC AGO TGA Asp Sec Ser 405 ACT GTA ACT Ser Val Sec TAC ATG Tyr Met 410 TAO TGG TAG CAG Tyr Tcp Tyr Gin TCC TOO Sec Sec 420 CCC AGA OTT CTG Pro Arg Leu Leu ATT TAT GAG ACA TCC AAC OTG CT TCT GGA Ile Tyr Asp Thc Ser Asn Leu Ala Ser Gly 425 430 OCT GTT CGC TTC Pro Val. Arg Phe ACT GCC Sec Gly 440 ACT COO TOT Ser Cly Ser ACO TOT TAC Thr Sec Tyr TOT OTO Sec Leu 450 C CGAG Gys Gin 465 ACA ATC AGO CA Thr Ile Sec Arg GAG GOT GAA GAT Giu Ala Giu Asp CCC A2T TAT TAG Ala Thc Tyr Tyr GAG TCC ACT Gin Trp Sec TAO COG OTO ACC TTC Tyr Pro Leu Thr Phe 47S COT GOT COO AGO Cly Ala Gly Thc AAG CTG GAA Lys Leu Ciu 480 ATA AAA Ile Lys INFORMATION FOR SEQ ID NCO: 28: SEQUENCE CHARACTERISTICS: LENGTH: 242 amino acids TYPE: amino acid TOPOLOGY: linear 14, 1 1 1 I 70 (ii) MOLECULE TYPE: protein (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 28: Glu Val Lys Leu Gin Gin Ser Gly Ala Glu Leu Val Lys 1 Ser Val Leu Asp dly Gln Lys Ser 6S lie Glu Ala Ser Gin Gly Gly Gly 130 Thr Ile 145 Asp Ser Ser Ser Val Pro Thr Ile 210 Gin Trp 225 Ile Lys 5 Leu Ser 20 Trp Val Asn Pro Ala Thr Ser Ser Asp Tyr 100 Thr Val Gly Gly Ser Ala Ser Val 165 Arg Leu 180 Arg Phe Arg Met Cys Lys Lys Gin Ser Asn 55 Leu Thr 70 Leu Thr Asp Tyr Thr Val Gly Gly 135 Ser Pro 150 Ser Tyr Leu Ile Ser Gly Giu Ala 215 Tyr Gin Asn Ser 75 Cys Tyr Gly Glu Val 155 Tyr Ser Gly Al a Al a 235 Thr Phe Gly Leu Tyr Asn Ser Ser Ser Val Phe Asp Gly Gly 125 Leu Thr 140 Thr Met Gin Gin Asn Leu Thr Ser 205 Thr Tyr 220 Pro Gly Ala Thr Ser His Glu Trp Ile Glu Lys Phe Thr Ala Tyr Tyr Tyr Cys Tyr Trp Gly 110 Ser Gly Gly Gln Ser Pro Thr Cys Ser 160 Lys Thr Gly 175 Ala Ser Gly 190 Tyr Ser Leu Tyr Cys Gin Ser Ser Tyr Leu Thr Phe Gly Gly Thr Lys Leu INFORMATION FOR SEQ ID NO: 29: SEQUENCE CHARACTERISTICS: LENGTH: 726 base pairs TYPE: nucleic acid STRA14DEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA (iii) HYPOTHETICAL: NO 71 (iv) ANTI-SENSE: NO FRAGMENT TYfPE: N-terminal (vi) ORIGINAL SOURCE: ORGANISM: mouse STRAIN: Balb/c ORV7RLOPMKNTfAL STAGE: adult TISSUE TYPE: splenocytes (vii) IMMEDIATE SOUR1CE% CLONE: 11 H I (single-chain Fv, heavy and light chain plus linker (ix) FEATURE: NAME/KEY: CDS LOCATION:l. .726 (xi) SEQUENCE DESCRIPTION: SEQ ID NO; 29: GAG GTC AAG CTG CAG CAG TCA GGG GGT GAA CTG GTG AAG CCT GGG GCT 48 :.Gl Val Lys Leu Gin Gin Ser Gly Ala Glu Leu Val Lys Pro Gly Ala 245 250 255 15 TCA GTG AAG TTG TCC TG AAG GCT TCC GGC TAG ACC TTC ACC AGG CAC 96 *Ser Val Lys Leu Ser Cys Lys Ala Ser Gly Tyr Thr Phe Thr Ser His 260 265 270 TOG ATG CAC TOG GTC AAG CAG AGG OCT GGA CAA GGC ITO GAG TGO ATC 144 Trp, Met His Trp Val Lys Gin Arg Ala Gly Gin Gly Leu Giu Trp Ile 275 280 285 290 GGA GAG TTT AAT CCC AGO AAC GGC CGT ACT AAC TAG AAT GAG AAA TTC 192 Gly C-lu Phe Asn Pro Ser Asn Gly Arg Thr Asn Tyr Asn C-lu Lys Phe 295 300 305 *AAG AGC AAG GCC ACA CTG ACT GTA GAC AAA TCC TCC AGO ACA GCC TAC 240 Lys Ser Lys Ala Thr Leu Thr Val Asp Lye Ser Ser Ser Thr Ala Tyr 310 315 320 ATG CAA CTC AGC AGC CTG ACA TCT GAG CAC TCT GCG GTC TAT TAG TGT 288 Met Gln Leu Ser Ser Leu Thr Ser Clii Asp Ser Ala Val Tyr Tyr Gys 325 330 335 GCC AGT COG GAG TAT GAT TAG GAC GGA CGG TAG TTT GAG TAG TOG GG 336 Ala Ser Arg Asp Tyr Asp Tyr Asp Cly Arg Tyr Phe Asp Tyr Trp Gly 340 345 350 CAA COG ACC ACG GTC ACC GIG TCC TCA GOT CCC GGT GGC TCC 000 GOT 384 Gin Gly Thr Thr Val Thr Val Ser Ser Gly Cly Gly Gly Ser Cly Gly 355 360 365 370 OCT COO TCG GOT GGC GGC GGA TCT GAC ATT GAG CIC ACC CAG TCT CCA 432 Gly Gly Ser Gly Gly Gly Gly Ser Asp Ile Glii Leu Thr Gin Ser Pro 375 380 385 TCA ATC ATG TCT GCA TCT OCA GGG GAG AAG GTC ACG ATO ACC TCC AGT 480 Ser Ie Met Ser Ala Ser Pro Gly Glu Lys VaL Thr Met Thr Cys Ser 390 395 400 GAG AGC TCA AOT OTA ACT TAG ATG TAG TOG TAC GAG GAG AAG ACA OGA 528 Asp Ser Ser Ser Val Ser Tyr Met Tyr Trp Tyr Gin Gin Lys Thr Oly 405 410 415 -72

TCC

Ser

GTC

Val 435

ACA

Thr

CAG

Gin CCC AGA CTC CTG ATT TAT GAC ACA TCC AAC CTG GCT TCT GGA Pro Arg Leu Leu Ile Tyr Asp Thr Ser Aen Leu Ala Ser Gly 425 430 GTT CGC TTC AGT GGC AGT GGG TCT GGG ACC TCT TAC TCT CTC Val Arg Phe Ser Gly Ser Gly Ser Gly Thr Ser Tyr Ser Leu 440 445 450 AGC CGA ATG GAG GCT GAA GAT GCT GCC ACT TAT TAC TGC CAG Ser Arg Met Giu Ala Glu Asp Ala Ala Thr Tyr Tyr Cys Gin 455 460 465 AGT AGT TAC CCA CAC ACG TTC GGT CCT GGG ACC AAG CTG GAA Ser Ser Tyr Pro His Thr Phe Gly Ala Gly Thr Lys Leu Glu 470 475 480 r r o ATA AAA Ile Lye INFORMATION FOR SEQ ID NO: SEQUENCE CHARACTERISTICS: LENGTH: 242 amino acids TYPE: amino acid TOPOLOGY: linear (ii) MOLECULE TYPE: protein (xi) SEQUENCE DESCRIPTION: SEQ ID NO: Glu Val Lys Leu Gin Gin Ser Gly Ala Glu Leu Val Lys Pro Gly Ala 20 Ser Trp Gly Lye Met Ala Gin Gly Ser 145 Asp Lye Ala Ser Gin Arg Ala 40 Asn Gly Arg Thr Val Asp Thr Ser Glu Tyr Asp Gly 105 Val Ser Ser 120 Gly Ser Asp- 135 Pro Gly Glu Tyr Met Tyr Gly Tyr Gly Gin Thr Asn Lye Ser Asp Ser Arg Tyr Gly Gly lTle Glu Lye Val 155 Trp Tyr 170 Phe Thr Leu Glu Asn Glu Ser Thr Ala Val Phe Asp Gly Gly 125 Leu Thr 140 Thr Met Gin Gin I r Ill. Ili, I I I i Ser ger Pro Arg Leu Le Ile Tyr Asp Thr Ser Asn Len Ala Ser Gly 180 185 190 Val 9 ro Val Arg Phe Ser Gly Ser Gly Ser Gly Thr Ser Tyr Ser Leu 195 200 205 Thr Ile Ser Arg Met Glu. Ala Giu. Asp Ala Ala Thr Tyr Tyr CyB Gin 210 215 220 Gin '7rp Ser Ser Tyr Pro His Thr Phe Gly Ala Giy Thr Lys Leu Glu 225 230 235 240 Ile Lys INFORMATION FOR SEQ ID NO: 31: SEQUENCE CHARACTERISTICS: LENGTH: 732 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: cOMA (iii) HYPOTHETICAL:

NO

fiv) ANTI-SENSE: NO FRAGMENT TYPE: N-terminal (vi) ORIGINAL SOURCE: ORGANISM: mouse STRAIN: Balb/c TISSUE TYPE: splenocytes (Yii) IMMEDIATE SOURCE: CLONE: 1 A 1 (single-chain Fv, heavy and light chain plus linker) (ix) FEATURE: NAME/KEY: CDS LOCATION:l. 732 (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 31: GAG gTG GAG CTG CAG CAG TGI GGG GCT GAA GIG GTG AAG GCT GGG GCT 48 *Glu Val Gin Len Gin Gin Ser Gly Ala Gln Leu. Val Lys Pro Gly Ala 245 250 255 TCA 4TG AAG TTG TCC TGC AAG GCT TCC GGC TAG ACC TIC ACC AG GAC 96 Ser Val Lys Leu Ser Gys Lys Ala Ser Gly Tyr Thr Phe Ihr Ser His 260 265 270 TGG ATG GAG TGG GTG AAG GAG AUG GCT GGA CAA GG GTT GAG TGG ATC 144 Trp M'et His Trp Val Lys Gin Ara Ala Gly Gin Gly Leu Glu Trp Ile 275 280 285 290 GGA GAG TTT AAT CCC AGG AAC GGC CGT ACT AAG TAG AAT GAG AAA TTC 192 Gly Olu Phe Asn Pro Ser Asn Gly Arg Thr Asn Tyr Asn Giu Lys Phe 295 300 305 AAG AGC AAG GGG AGA GIG ACT C-IA GAG AAA 'FCC TCC AGC AGA GGT TAG 240 Lys ter Lys Ala Thr Len Ihr Val Asp Lys Ser Ser Ser Thr Ala Tyr 310 315 320 ATG CAA CTC AGC AGO CTG ACA TCT GAG GAC TCT GCG GTC TAT TAO TGT 288 Met Gin Leu Ser Ser Leu Thr Ser Glu Asp Ser Ala Val. Tyr Tyr Cys 32S 330 335 GCC AGT CGG GAG TAT GAT TAG GACGGOA COG TAC TTT GAO TAC TGC GGC 336 Ala Ser Ary Asp Tyr Asp Tyr Asp Gly Arg Tyr Phe Asp Tyr Trp Gly 30345 350 CAA GGG ACO AG GTC ACC GTC TCC TCA GOT GCC OCT 0CC TCG GO-C GOT 384 Gin Gly Thr Thr Val. Thr Val Ser Ser Gly Gly Gly Gly Ser Gly Gly 355 360 365 370 GCT 0CCG TOO OCT 0CC 0CC OCA TCT GAC ATT GAG CTC ACC CAG TOT CCA 432 Gly Oly Ser Gly Gly Cly Oly Ser Asp Ile Glu Leu Thr Gin Ser Pro 375 380 385 ACA ATC ATG TOT OCA TCT CCA GGC GAG AAG, GTC ACC ATG ACC TGC ACT 480 Thr Ile Met Ser Ala Ser Pro Gly Olu Lys Val Thr Met Thr Cys Ser 390 395 400 GAC AGOC ICA AGT GTh AGT TAO ATOG TAC TGG TAC CAG CG AAG ACA GGA 528 Asp Ser Ser Ser Val Ser Tyr Met Tyr Trp Tyr Gin Gin Lys Thr Gly 405 410 41S TCC TCO CCC AGA OTC CTO ATT TAT GAG ACA TOO AAC CTG OCT TCT OGA 576 Ser Ser Pro Arg Lau Leu Ile Tyr Asp Thr Ser Asn Lau Ala Ser Oly 420 425 430 OTO GOT GTT 000 TTO ACT 000 AOT 0CC TOT 0CC ACC TOT TAO TOT OTC 624 Val. Pro Val Arg Phe Ser Oly Ser Gly Ser Oly Thr Ser Tyr Ser Leu 435 440 445 450 ACA ATO AGO CGA ATO GAG OCT GAA OAT OCT 0CC ACT TAT TAG TGO GAG 672 Thr Ile Ser Arg Met Oiu Ala Olu Asp Ala Ala Thr Tyr Tyr Cys Girl =455 460 465 GAG TOG AGT AGT TAG OCA CCC ATO TAC AG ITO GGA COG CCC ACA ALAG 720 Gin Trp, Ser Ser Tyr Pro Pro Met Tyr Thr Phe Oly Cly Gly Thr Lys *470 475 480 TTG GAA ATA AAA 732 Leu Glu ile Lys S 25 485 INFORMATION FOR SEQ ID NO: 32: SEQUENCE CHARACTERISTICS: LENGTH: 244 amino acids TYPE: amino acid TOPOLOGY: linear (ii) MOLECULE TYPE: protein (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 32: GLu Val Gin Leu Gin Gin Ser Gly Ala Oiu Leu Val Lys Pro Gly Ala 1 5 10 is Ser Val Lys Leu Ser Cys Lys Ala Ser Oly Tyr Thr Phe Thr Ser His 2S Trp met His TrD Va]. Lys Gin Arg Ala Oly Gin Cly Lau Glu Trp ile 35404 II~ ,III' 1 11 013111,I 75 Gly Giu Phe Lys Ser Lys Met Gin Leu Ala Ser Arg Gin Gly Thr 115 Gly Gly Ser 130 Thr Ile Met 145 Asp Ser Ser.

Ser Ser Pro Vai Pro Val 195 Thr Ile Ser 210 Asn Pro Ala Thr Ser Ser as Asp Tyr 100 Thr Val Gly Gly Ser Ala Ser Val 165 Arg Leu 180 Arg Phe Arg Met Thr Asn Lys Ser Aaip Ser Arg Tyr Gly Gly Ile Glii Lys Val 155 Trp Tyr 170 Thr Ser Ser Gly Ala Ala Thr Phe 235 Tyr Asn Ser Ser Ala Val Phe Asp Gly Gly 125 Leu Thr 140 Thr Met Gin Gin Asn Leu Thr Ser 205 Thr Tyr 220 20 225 Leu Glu Ile Ser Tyr Pro 230 Lys Pro Met Tyr Gly Gly Gly Thr

Claims (9)

1. Anti-EGFR single-chain Fvs obtainable from phage-antibody libraries constructed from cells from an immunized mammalian.

2. Antibody fragment according to Claim 1 obtainable from cells of an immunized mouse. 1 5

3. Antibody fragment according to Claims 1 or 2 obtainable 15 from cells of the lymph node, (ii) the spleen, or (iii) in vitro immunized cells.

4. Antibody fragment according to Claims 1 3, wherein the variable regions of the heavy and light chain comprise a DNA and/or an amino 25 acid sequence selected from one of the heavy and light chain sequences given in Sequence Id.

Nos. 1- 32. Anti-EGFR antibody constructed from DNA sequences deriving from antibody fragments according to Claims 1 4 and from DNA sequences deriving from constant regions of human immunogobulins.

6. Antibody according to Claim 5 wherein the heavy constant chain region comprises the amino acid sequence of a human gamma-1 chain, and the light constant chain region comprises the amino acid sequence of a human kappa chain.

7. Process for the preparation of an anti-EGFR single-chain Fv according to one of the Claims 1 4 comprising the following steps: I0 isolating RNA from immunized mammalian cells, preferably mouse cells, (ii) synthesizing first-strand cDNA, (iii) amplifying the VH and Vk genes in cDNAs from the 15 immunized cells, (iv) cloning said genes together with suitable restriction sites into a phagemid vector, transforming prokaryotic cells with the ligation mixtures, 20 (vi) screening the phage libraries for phage antibodies directed to EGFR using purified EGFR, and "(vii) producing the desired single-chain Fv in prokaryotic host cells, preferably E. coli.

8. Process for the preparation of a whole anti-EGFR antibody by cloning the DNA coding for the variable regions of anti-EGFR antibody S fragments produced according to Claim 7 into at least one eukaryotic expression vector containing genomic DNA which codes for the constant regions of human immunoglobulins, transforming eukaryotic cells with said vector(s) and expressing and isolating the antibody. I Ir MV, l> I" -7

9. Pharmaceutical composition comprising an anti-EGFR antibody a fragment according to one of the Claims 1 4 or a whole anti-EGFR antibody according to Claims 5 or 6. of an anti-EGFR antibody fragment according to one of the Claims 1 4, or a whole anti-EGFR antibody according to one of the Claims 5 or 6 for the manufacture of a drug directed to tumors or for the diagnostic location and assessment of tumor growth- *s 2