Construction, Affinity Maturation, and Biological Characterization of an Anti-tumor-associated Glycoprotein-72 Humanized Antibody

The tumor-associated glycoprotein (TAG)-72 is expressed in the majority of human adenocarcinomas but is rarely expressed in most normal tissues, which makes it a potential target for the diagnosis and therapy of a variety of human cancers. Here we describe the construction, affinity maturation, and biological characterization of an anti-TAG-72 humanized antibody with minimum potential immunogenicity. The humanized antibody was constructed by grafting only the specificity-determining residues (SDRs) within the complementarity-determining regions (CDRs) onto homologous human immunoglobulin germ line segments while retaining two mouse heavy chain framework residues that support the conformation of the CDRs. The resulting humanized antibody (AKA) showed only about 2-fold lower affinity compared with the original murine monoclonal antibody CC49 and 27-fold lower reactivity to patient serum compared with the humanized antibody HuCC49 that was constructed by CDR grafting. The affinity of AKA was improved by random mutagenesis of the heavy chain CDR3 (HCDR3). The highest affinity variant (3E8) showed 22-fold higher affinity compared with AKA and retained the original epitope specificity. Mutational analysis of the HCDR3 residues revealed that the replacement of Asn97 by isoleucine or valine was critical for the affinity maturation. The 3E8 labeled with 125I or 131I showed efficient tumor targeting or therapeutic effects, respectively, in athymic mice with human colon carcinoma xenografts, suggesting that 3E8 may be beneficial for the diagnosis and therapy of tumors expressing TAG-72. The tumor-associated glycoprotein (TAG)-72 is expressed in the majority of human adenocarcinomas but is rarely expressed in most normal tissues, which makes it a potential target for the diagnosis and therapy of a variety of human cancers. Here we describe the construction, affinity maturation, and biological characterization of an anti-TAG-72 humanized antibody with minimum potential immunogenicity. The humanized antibody was constructed by grafting only the specificity-determining residues (SDRs) within the complementarity-determining regions (CDRs) onto homologous human immunoglobulin germ line segments while retaining two mouse heavy chain framework residues that support the conformation of the CDRs. The resulting humanized antibody (AKA) showed only about 2-fold lower affinity compared with the original murine monoclonal antibody CC49 and 27-fold lower reactivity to patient serum compared with the humanized antibody HuCC49 that was constructed by CDR grafting. The affinity of AKA was improved by random mutagenesis of the heavy chain CDR3 (HCDR3). The highest affinity variant (3E8) showed 22-fold higher affinity compared with AKA and retained the original epitope specificity. Mutational analysis of the HCDR3 residues revealed that the replacement of Asn97 by isoleucine or valine was critical for the affinity maturation. The 3E8 labeled with 125I or 131I showed efficient tumor targeting or therapeutic effects, respectively, in athymic mice with human colon carcinoma xenografts, suggesting that 3E8 may be beneficial for the diagnosis and therapy of tumors expressing TAG-72. Monoclonal antibodies (mAbs) 4The abbreviations used are: mAb, monoclonal antibody; TAG-72, tumor-associated glycoprotein-72; SDR, specificity-determining residue; CDR, complementarity-determining region; HCDR, heavy chain CDR; FR, framework region; LCDR, light chain CDR; VH, heavy chain variable domain; VL, light chain variable domain; JH, heavy chain joining region; JK, κ light chain joining region; Cκ, κ light chain constant region; Cγ, γ heavy chain constant region; CHO, Chinese hamster ovary; ELISA, enzyme-linked immunosorbent assay; Fab, antigen-binding fragment.4The abbreviations used are: mAb, monoclonal antibody; TAG-72, tumor-associated glycoprotein-72; SDR, specificity-determining residue; CDR, complementarity-determining region; HCDR, heavy chain CDR; FR, framework region; LCDR, light chain CDR; VH, heavy chain variable domain; VL, light chain variable domain; JH, heavy chain joining region; JK, κ light chain joining region; Cκ, κ light chain constant region; Cγ, γ heavy chain constant region; CHO, Chinese hamster ovary; ELISA, enzyme-linked immunosorbent assay; Fab, antigen-binding fragment. are increasingly being used as therapeutic agents for cancer and other diseases. Murine mAbs have limited use as therapeutic agents because of a short half-life, an inability to trigger human effector functions, and the induction of a human anti-mouse antibody response (1Seccamani E. Tattanelli M. Mariani M. Spranzi E. Scassellati G.A. Siccardi A.G. Int. J. Rad. Appl. Instrum. B. 1989; 16: 167-170Abstract Full Text PDF PubMed Scopus (66) Google Scholar, 2Colcher D. Milenic D.E. Ferroni P. Carrasquillo J.A. Reynolds J.C. Roselli M. Larson S.M. Schlom J. J. Nucl. Med. 1990; 31: 1133-1142PubMed Google Scholar). To reduce the immunogenicity of murine antibodies in humans, chimeric antibodies with mouse variable regions and human constant regions were initially constructed (3Morrison S.L. Johnson M.J. Herzenberg L.A. Oi V.T. Proc. Nat1 Acad. Sci. U. S. A. 1984; 21: 6851-6855Crossref Scopus (800) Google Scholar). Although chimeric antibodies proved to be less immunogenic than murine mAbs, human anti-chimeric antibody responses have been observed (4Bell M. Kamm M. Aliment Pharmacol. Ther. 2000; 14: 501-514Crossref PubMed Scopus (67) Google Scholar). To further reduce the immunogenicity of the mouse variable regions, a humanized antibody has been constructed by grafting the complementarity-determining regions (CDRs) of a murine mAb onto the human framework regions (FRs) by a procedure commonly referred to as CDR grafting (5Jones P.T. Dear P.H. Foote J. Neuberger M.S. Winter G. Nature. 1986; 321: 522-525Crossref PubMed Scopus (1101) Google Scholar). Simple grafting CDRs, however, often decreased the affinity, because some FR residues directly contact the antigen or support the conformation of the CDR loops (6Queen C. Schneider W.P. Selick H.E. Payne P.W. Landolfi N.F. Duncan J.F. Avdalovic N.M. Levitt M. Junghans R.P. Waldmann T.A. Proc. Natl. Acad. Sci. U. S. A. 1989; 86: 10029-10033Crossref PubMed Scopus (557) Google Scholar, 7Al-Lazikani B. Lesk A.M. Chothia C. J. Mol. Biol. 1997; 273: 927-948Crossref PubMed Scopus (578) Google Scholar). Therefore, humanized antibody is currently constructed primarily by CDR grafting, while retaining those rodent FR residues that influence antigen-binding activity (8Winter G. Harris W.J. Immunol. Today. 1993; 14: 243-246Abstract Full Text PDF PubMed Scopus (143) Google Scholar). Since the FR residues often differ from one humanized antibody to another, the identification of key rodent FR residues is a crucial part of the humanization (9Reichert J.M. Nat. Biotechnol. 2001; 19: 819-822Crossref PubMed Scopus (153) Google Scholar). Clinical studies have indicated that such humanized antibodies are, in general, less immunogenic than murine or chimeric antibodies and are tolerated by humans (10Stephens S. Emtage S. Vetterlein O. Chaplin L. Bebbington C. Nesbitt A. Sopwith M. Athwal D. Novak C. Bodmer M. Immunology. 1995; 85: 668-674PubMed Google Scholar, 11Mateo C. Moreno E. Amour K. Lombardero J. Harris W. Perez R. Immunotechnology. 1997; 3: 71-81Crossref PubMed Scopus (167) Google Scholar). Unfortunately, the non-human CDRs of humanized antibodies can induce human anti-humanized antibody responses in patients (12Hwang W.Y.K. Foote J. Methods. 2005; 36: 3-10Crossref PubMed Scopus (454) Google Scholar). Comprehensive analyses of the three-dimensional structures of the antibody-combining sites have revealed that only 20–33% of the CDR residues participate in antigen-binding (13Glaser S.M. Vasquez M. Payne P.W. Schneider P.W. J. Immunol. 1992; 149: 2607-2614PubMed Google Scholar, 14Padlan E.A. Mol. Immunol. 1994; 31: 169-217Crossref PubMed Scopus (783) Google Scholar, 15Padlan E.A. Abergel C. Tipper J.P. FASEB J. 1995; 9: 133-139Crossref PubMed Scopus (123) Google Scholar). Most of the variable positions of CDRs are directly involved in the interaction with antigen (i.e. specificity-determining residues (SDRs)), whereas most of the conserved residues mainly serve to stabilize the structure of the combining site (15Padlan E.A. Abergel C. Tipper J.P. FASEB J. 1995; 9: 133-139Crossref PubMed Scopus (123) Google Scholar). The SDRs are observed primarily in the C-terminal part of light chain CDR1 (LCDR1), the first and sometimes also the middle positions in LCDR2, the middle portion of LCDR3, most of the heavy chain CDR1 (HCDR1), the N-terminal and middle parts of HCDR2, and most of HCDR3 except the terminal position. The SDRs form the center of the antibody-combining site. Therefore, one possible way to minimize the human anti-humanized antibody response is to graft only the SDRs of a murine antibody onto human FRs while maintaining the conformations of the CDRs of the murine antibody. Tumor-associated glycoprotein (TAG)-72 is expressed by the majority of human adenocarcinomas in the colon, ovary, pancreas, breast, prostate, and lung but not in most normal tissues, except the endometrium in the secretory phase (16Thor A. Ohuchi N. Szpak C.A. Johnston W.W. Schlom J. Cancer Res. 1986; 46: 3118-3124PubMed Google Scholar, 17Thor A. Viglione M.J. Muraro R. Ohuchi N. Schlom J. Gorstein F. Int. J. Gynecol. Pathol. 1987; 6: 235-247Crossref PubMed Scopus (130) Google Scholar). A murine mAb B72.3 (16Thor A. Ohuchi N. Szpak C.A. Johnston W.W. Schlom J. Cancer Res. 1986; 46: 3118-3124PubMed Google Scholar) that specifically binds to TAG-72 is approved for in vivo imaging in patients with ovarian and colorectal cancers. A second generation antibody to B72.3, CC49 (18Muraro R. Kuroki M. Wunderlich D. Poole D.J. Colcher D. Thor A. Greiner J.W. Simpson J.F. Molinolo A. Noguchi P. Schlom J. Cancer Res. 1988; 48: 4588-4596PubMed Google Scholar, 19Colcher D. Minelli F.M. Roselli M. Muraro R. Simpson-Milenic D. Schlom J. Cancer Res. 1988; 48: 4597-4603PubMed Google Scholar), which has higher affinity for TAG-72 than B72.3 does, has shown efficient targeting to various carcinomas in A.M. S. N. C. P. F. J.M. Cancer Res. 1995; Google Scholar, A.M. K. S. K. N. F. A.M. Schlom J. Larson S.M. Nucl. Med. Biol. 1994; 21: PubMed Scopus Google Scholar, Carrasquillo J.A. Milenic D.E. Schlom J. C. P. Reynolds J. G. W. W. R. D. K. J. Cancer Res. 1995; Google Scholar). The epitope for B72.3 or CC49 has been as (i.e. to S. Cancer Res. 1988; 48: Google Scholar). CC49 human anti-mouse antibody responses in patients A.M. L. P. K. S. N. D. L. J. Nucl. Med. 1995; 36: Google Scholar). To a humanized CC49 antibody has been constructed by CDR grafting while retaining the FR residues that the structure of the antibody and those involved in interaction or antigen L. E.A. Milenic D.E. Schlom J. P.H. 1995; 14: PubMed Scopus Google Scholar). to reduce the immunogenicity of the mouse CDR and FR residues have been with human and the antigen-binding activity and potential immunogenicity of variant have been M. Milenic D.E. E.A. R. Schlom J. Mol. Immunol. 36: PubMed Scopus Google Scholar, M. Milenic D.E. M. E. Schlom J. J. Immunol. 2000; Scopus Google Scholar, E.A. R. P. Schlom J. Mol. Immunol. PubMed Scopus Google Scholar, R. E.A. P. Schlom J. Cancer Res. 9: Google Scholar). the humanized antibody was constructed by grafting only the SDRs onto a human germ line with CDRs of the structures as those of the murine antibody while retaining two key murine FR The resulting humanized antibody AKA showed only about 2-fold lower affinity compared with antibody affinity was by random mutagenesis of HCDR3 residues by affinity The resulting humanized antibody (3E8) showed 27-fold lower reactivity compared with HuCC49 and higher affinity compared with original The antibody showed tumor targeting and therapeutic in athymic mice human expressing TAG-72. of human FRs for grafting, the and of CC49 were compared with those of human variable and joining germ line was that and were the most homologous to the and of To the of the humanized light chain only in the of CC49 was onto human The E.A. C. of of of and of Scholar). The of was by a humanized P. C. PubMed Scopus Google Scholar), which is also the as a and and and The were with and and to the the humanized was to human in by The was with and and the sites of to To a humanized VH, the the SDRs in the the and two residues and of CC49 were onto to the of a humanized heavy chain The of the and humanized are shown in The humanized was by a humanized E.A. P. C. PubMed Scopus Google Scholar), which is also human as a and and and The humanized was to human in P. C. PubMed Scopus Google Scholar) by The was with and and to humanized heavy chain was constructed from by as a and was the sites of to and of humanized antibody was expressed in and the was to to antigen-binding activity and the of humanized was a Chinese hamster in in the were for the of antibody by an and were to as PubMed Scopus Google Scholar). the and of the line that expressed the antibody was in and the was to affinity a as E.A. P. C. PubMed Scopus Google Scholar). The and of the antibody were by of the an of was used for a of A.M. W. in Immunology. of Scholar). an ELISA, antibody were to that been with of as M. Milenic D.E. M. E. Schlom J. J. Immunol. 2000; Scopus Google Scholar), and were of were to and for of and were to and for The was by the of of and the was in an To antigen-binding affinity, a was as E.A. P. C. PubMed Scopus Google Scholar). of antibody and various of as a antigen were for and were to that been with of the The antibody was by an The constant of antibody was from a B. L. J. Immunol. Methods. PubMed Scopus Google Scholar). To the epitope of a CC49 and 3E8 was as PubMed Scopus Google Scholar). CC49 or 3E8 was with in for in the of of a The CC49 or 3E8 was by anti-mouse or a of the humanized which specifically binds to the antigen of S. M.S. PubMed Scopus Google Scholar), was potential immunogenicity of AKA in humans was by the reactivity of the humanized antibody to the serum from patient in a phase Carrasquillo J.A. Milenic D.E. Schlom J. C. P. Reynolds J. G. W. W. R. D. K. J. Cancer Res. 1995; Google Scholar). Since the serum from patient variable it was used to the reactivity of AKA with that of the serum also TAG-72 antigen and which may have with the of the humanized antibody to the antibodies in the Therefore, to the serum reactivity TAG-72 antigen and antibodies were from the serum by to murine mAb, that an epitope of TAG-72 that is from the one by CC49 M. Wunderlich D. Colcher D. Simpson J.F. Poole D.J. Schlom J. Cancer Res. 1990; Google Scholar). was to The patient serum was to the and was with The were for and the were and for the serum reactivity To the serum reactivity of a was a as P. Schlom J. J. Immunol. Methods. PubMed Scopus Google Scholar). were the by The of the was by of a of and a of in a of were of were The the were by To the serum reactivity of HuCC49 and AKA or HuCC49 was with the patient serum of TAG-72 and antibodies and the was to the in and a in as a to for to serum the was for the were from the with a of and the were with a of The of antibody was as of of the with of the The for antibody was as the for of the of the serum to of the affinity of the HCDR3 of AKA was and the resulting was to a K. A. F. L. M. P. W. J. Immunol. Methods. PubMed Scopus Google Scholar). To the was constructed by of the from Proc. Natl. Acad. Sci. U. S. A. 1992; PubMed Scopus Google Scholar) and by replacement of the first by the of and the of were the and respectively, of to was used as a for random mutagenesis of the HCDR3 by The first was the and and HCDR3 respectively, or HCDR3 and The HCDR3 was and random the first residues of the HCDR3 of The of the were as HCDR3 HCDR3 The two first were to The was with and and was the sites of to a the of with the were onto a a and were A was with antigen by in for with and with for The was with and in with and was and the was of the the were The was with with for and with for the was by and the the were used to the the original was in and to of as were and the antigen-binding activity of was by an The with affinity were and the of the HCDR3 were of to of and heavy chain E.A. P. C. PubMed Scopus Google Scholar) were by The were with and and were the site of The resulting was and the antibody in the was by to the antigen-binding of was used for the AKA and 3E8 with 125I or 131I The antibody was by a and was by The and were with in the phase and in the athymic mice were from were a of human colon line in the or were the of the athymic mice a of mice was to and and was in a The of the of was of was in the athymic mice the tumor were the of athymic mice or of 3E8 was mouse the first The and tumor were for The tumor a was expressed to the tumors of the mice with or the tumor was as the to a tumor of The tumor was as in mice with in mice with in mice with the were the of the athymic or of 3E8 was mouse the first The of the mice was for and were for of as a of were to the of the of mice in and of a humanized the of the CC49 and were compared with those of human germ line The showed that and were most homologous to CC49 and VL, The the CC49 and revealed that only positions and in the one in the LCDR2, and one in the were whereas positions in the FRs were CDR have structure of and B. Lesk A.M. Chothia C. J. Mol. Biol. 1997; 273: 927-948Crossref PubMed Scopus (578) Google Scholar). grafting, only one in the or and in the and in the residues of CC49 were onto human and the resulting humanized were to human to humanized light or The FR residues of CC49 were not The of the and humanized light chain are shown in To a humanized VH, the of CC49 was compared with those of human and The CC49 and revealed that positions in the and positions in the were whereas positions in the FRs were and have and B. Lesk A.M. Chothia C. J. Mol. Biol. 1997; 273: 927-948Crossref PubMed Scopus (578) Google Scholar). The HCDR3 is not by human germ line and has structure B. Lesk A.M. Chothia C. J. Mol. Biol. 1997; 273: 927-948Crossref PubMed Scopus (578) Google Scholar). To a humanized VH, the SDRs residues in the in the HCDR2, and in the were onto while retaining two residues and that were to influence the conformations of and The resulting humanized was to the human to the humanized heavy chain The of and humanized are shown in The humanized heavy and light were expressed in and the humanized antibody (AKA) in the was for antigen-binding affinity by a The constant of AKA was To the of two mouse residues and retained in or was with or respectively, of human or residues were with the human residues The of and were and that and are in antigen-binding activity and to be retained in humanized antibody. The affinity of the humanized antibody that of the humanized heavy chain and the humanized light chain was which was higher than that of that the two residues and may not be The affinity of the antibody that of chimeric CC49 heavy chain and humanized light chain was which is to that of AKA Therefore, it is that the humanized heavy chain is for The affinity of AKA was only lower than that of a chimeric antibody that of chimeric CC49 heavy and light The affinity of the are in The humanized antibody AKA the of mouse residues was for further affinity of various humanized CC49 affinity of antibody was from of as of antibody was from of as in a of potential AKA was from a line and as The potential immunogenicity of AKA to HuCC49 was by reactivity to the serum from the patient in a phase Carrasquillo J.A. Milenic D.E. Schlom J. C. P. Reynolds J. G. W. W. R. D. K. J. Cancer Res. 1995; Google Scholar). The serum antibodies to CC49 and also TAG-72 antigen which with the of HuCC49 or AKA to the serum Therefore, to the serum reactivity TAG-72 antigen and antibodies were from the serum by to murine mAb, which an epitope of TAG-72 that is from the one by CC49 M. Wunderlich D. Colcher D. Simpson J.F. Poole D.J. Schlom J. Cancer Res. 1990; Google Scholar), as The serum reactivity of AKA and HuCC49 to antibodies was a the AKA or HuCC49 was with the patient serum that been of TAG-72 and the was to the and the was the of as a of the (i.e. the of the antibody for of the of the serum to was A higher a decreased reactivity to the antibodies to The for AKA was whereas that for HuCC49 was to a 27-fold in the reactivity of compared with to the patient suggesting that AKA may be less immunogenic than HuCC49 in of AKA and of the the affinity of AKA was in the The SDRs of the HCDR3 were and to as were and by for antigen-binding were to and with higher than that of AKA the of affinity, the was and to a shown in the of the and were and respectively, which to and in the affinity, respectively, compared with the AKA The HCDR3 of the are shown in and affinity of AKA in in a To 3E8 and CC49 to the 3E8 was and from a and a with CC49 was shown in the of CC49 to an TAG-72 was in a by of 3E8 but not by an humanized S. M.S. PubMed Scopus Google Scholar). that 3E8 and CC49 may to the The (i.e. the of the antibody (3E8) for of the of CC49 to was less than that of 3E8 to TAG-72, that the affinity of 3E8 is higher than that of A of the the AKA and revealed in the were limited was was or by Asn97 was by or was or by and was by or replacement of Asn97 by or in an in the affinity by or respectively, suggesting that Asn97 in CC49 is for TAG-72 To Asn97 of AKA was by and antigen-binding activity was the activity of the was higher than that of AKA To the of the other the affinity to or of AKA was by or respectively, and was for antigen-binding shown in the the activity of suggesting that the of affinity from AKA to 3E8 may have from the of the and and tumor targeting of 3E8 and AKA were in athymic mice human colon The or was the mouse and was and The of the of of tumor and normal are shown in The AKA and 3E8 showed tumor with tumor higher with 3E8 than with AKA the in the tumor was decreased in a whereas the in the tumor was the tumor of and and respectively, of that of is to the affinity of 3E8 compared with The indicated that the showed higher tumor than with of was by the because the use of of antibody is a for tumor and higher of to be than with therapy J.A. Cancer Res. 1997; Google Scholar, D. S. K. M. R. Cancer Res. 1995; Scholar). To the therapeutic of the 3E8 was for and in the athymic mice human colon was by of tumor or of the The from the tumor showed that the in the mice was than that in the mice with The was to be The of mice to of the mice and the mice showed a of to of the not the the from in the mice to in the mice that the of the mice was that of the The of the mice showed a of to of the not Murine mAbs induce human anti-mouse antibody responses in To murine mAbs have been humanized by CDR grafting. the non-human CDRs of humanized antibody can induce human anti-humanized antibody responses in patients (12Hwang W.Y.K. Foote J. Methods. 2005; 36: 3-10Crossref PubMed Scopus (454) Google Scholar). humanization such as E.A. Mol. Immunol. PubMed Scopus Google Scholar) and S. C. J. A. Med. 9: PubMed Scopus Google Scholar), have been but the or antibodies a higher of mouse residues than the and in we constructed a humanized CC49 antibody by grafting only the SDRs within the CDRs of CC49 onto homologous human germ line segments CDRs have the structures as those of CC49 while retaining only two mouse residues in the heavy chain that support the conformation of CDRs The resulting humanized antibody AKA showed only about 2-fold lower affinity compared with CC49 and 27-fold lower serum reactivity compared with the humanized antibody HuCC49 that was constructed by CDR grafting The that grafting is an of antibody The affinity of AKA was by random mutagenesis of the HCDR3 by affinity The highest affinity showed higher affinity and 2-fold tumor targeting compared with it showed higher affinity compared with CC49 while maintaining the epitope of CC49 The and analysis of the HCDR3 AKA and antibodies revealed that Asn97 of CC49 is for TAG-72 and that or is critical for affinity The variant 3E8 has the of mouse residues and the highest affinity the humanized of CC49 that have been constructed to M. Milenic D.E. E.A. R. Schlom J. Mol. Immunol. 36: PubMed Scopus Google Scholar, M. Milenic D.E. M. E. Schlom J. J. Immunol. 2000; Scopus Google Scholar, E.A. R. P. Schlom J. Mol. Immunol. PubMed Scopus Google Scholar, R. E.A. P. Schlom J. Cancer Res. 9: Google Scholar). The 3E8 labeled with 131I showed an therapeutic in athymic mice human colon The mice showed of and compared with the mice with and the mice showed a of Since is by of J.A. N. D. S. G. Cancer PubMed Scopus Google Scholar), the that the in tumor and in the To is the first to the therapeutic of an anti-TAG-72 humanized antibody with human colon in an in vivo tumor The of such as affinity, efficient tumor and therapeutic it a in the diagnosis and therapy of colon expressing TAG-72.