NOTE Immunology Construction of Chicken-Mouse Chimeric Antibody and Immunogenicity in Mice

Yoshiko TATEISHI1), Norihisa NISHIMICHI1), Hiroyuki HORIUCHI1), Shuichi FURUSAWA1) and Haruo MATSUDA1)*

1)Laboratory of Immunobiology, Department of Molecular and Applied Bioscience, Graduate School of Biosphere Science, Hiroshima University, 1–4–4 Kagamiyama, Higashi-Hiroshima 739–8528, Japan

(Received 5 October 2007/Accepted 10 December 2007)

ABSTRACT. Chicken monoclonal antibodies are potentially useful for diagnostic research and have clinical applications, as chicken show higher potential for antibody production with mammalian-conserved biological molecules. However, the applications of chicken anti- bodies are limited because of their immunogenicity in mammals. To overcome this problem, we have constructed a chicken-mouse chi- meric antibody containing the chicken variable region and the mouse constant region. This chimeric antibody retained similar binding affinities as the parental chicken antibody. The chimeric antibody was also producible as an ascitic antibody in BALB/c mice. Further- more, when the chimeric antibody was administered to mice, it did not provoke the mouse anti-chicken antibody response. These results indicate that the chimeric antibody is suitable for application to preclinical mouse studies. KEY WORDS: chicken, chimeric antibody, immunogenicity. J. Vet. Med. Sci. 70(4): 397–400, 2008

Mouse monoclonal antibodies (mAbs) are becoming recognize prion protein using cell fusion and phage display more important in various research, diagnostic and clinical techniques [12, 13, 20], and have reported that chicken applications. Clinically, however, mouse mAbs induce anti- mAbs are useful in the diagnosis of BSE and other prion dis- murine antibody response in humans, and thus therapeutic eases [15–17]. effects cannot be expected with mouse mAbs. To reduce the Because chicken antibodies are immunogenic in mam- immunogenicity of mouse antibodies in humans, various mals, their clinical applications are limited. In particular, techniques for chimeric and humanized antibodies have the constant region contributes a significant component to been successively developed. Construction of mouse- immunogenicity [18]. We recently constructed chicken- human chimeric antibodies [2, 3, 4, 11, 18, 19] and human- human chimeric antibody expression systems [22] by ized mouse antibodies [8, 9, 29] has been reported, and these replacing the chicken constant regions with human regions, antibodies show decreased immunogenicity [3, 11, 29] and and developed a simple method for humanizing chicken longer serum half-lives [11, 29]. Numerous mouse-human mAbs [21]. chimeric mAbs (such as edrecolomab, , and basil- Although chicken mAbs have become an attractive tool iximab) and humanized mouse mAbs (such as , in numerous fields, the antibody production levels of palivizumab and ) have been marketed and are chicken hybridomas remain relatively low and unstable now clinically used. when compared with murine hybridomas. Moreover, purifi- Although the rapid progress in the clinical application of cation of chicken IgY is difficult because the IgY does not mouse mAbs is remarkable, it is thought to be difficult to bind with protein A or protein G, and it is more sensitive to produce more favorable therapeutic mouse mAbs against acid or alkaline conditions when compared with mammalian highly conserved biological molecule targets. However, a IgG [27]. To improve chicken IgY purification, recombi- large number of biological molecules are conserved in nant IgY (rIgY) with a histidine tag was recently developed, mammalian species, and therefore provide limited immune and rIgY purification was simplified to a one-step process response in rodents due to immunotolerance. The chicken is using a nickel-affinity column [26]. a useful animal for development of specific antibodies Michael et al. [14] generated chicken-mouse chimeric against mammalian-conserved biological molecules as a antibodies, and these antibodies were functional in quantita- result of the phylogenic differences between chickens and tive ELISA and immunostaining. To establish the basic tech- mammals; consequently, chickens have a higher potential nology for clinical application of chicken antibodies, we for antibody production. Furthermore, the molecular diver- attempted to generate a chicken-mouse chimeric antibody sification of immunoglobulin in the chicken differs from and to determine its immunogenicity in mice. Administra- that in mammals; it is possible to perform RT-PCR of the V- tion of this chimeric antibody confirmed that the chimeric region repertoire with a single pair of primers. We have antibody was less immunogenic than a control chicken rIgY. generated and characterized numerous chicken mAbs that To generate chicken-mouse chimeric antibody, 2 expres- sion vectors, pcChMo/H for the heavy chain and pcChMo/L *CORRESPONDENCE TO: Dr. MATSUDA, H., Laboratory of Immunobi- ology, Department of Molecular and Applied Bioscience, Gradu- for the light chain, were constructed (Fig. 1). Total RNA ate School of Biosphere Science, Hiroshima University, 1–4–4 was prepared from mouse SP2/0-Ag14 (SP2) cells using Kagamiyama, Higashi-Hiroshima 739–8528, Japan. TRIzol reagent (Stratagene, U.S.A.). The first cDNA strand e-mail: [email protected] was primed with Oligo (dT)15 primer (Roche Diagnostics, 398 Y. TATEISHI ET AL.

was digested with BamHI and PinAI and subcloned into the expression vector pcDNA4/myc-His A (Invitrogen, U.S.A.) (pcChCH). The chicken heavy chain leader sequence and variable region were amplified by PCR using the primers HUC2-HF4 and VH-leader-R for the leader sequence, VH- bivalent-F and MVH-bivalent-R for the variable region, and MVH-CHF and MIgG-CH1R for the heavy chain CH1 region. As PCR templates, pcCK1 [26] was used for the leader sequence, and chicken anti-DNP scFv antibody iso- lated from a chicken naïve phage display library [15] was used for the variable region. Heavy chain leader sequence, variable region and CH1 PCR fragments were amplified by overlapping PCR and were digested with Hind III and BamH I and ligated into pcChCH (pcChMo/H). Light chain genes were similarly amplified by overlapping PCR using the amplified PCR fragments using the primers HUC2- VLF3 and VL-leader-R for the leader sequence, VL-biva- lent-F and MVL-bivalent-R for the variable region, and MIgG-CκF and MIgG-CκR for the Cκ sequence, and were then digested with Hind III and Xba I and subcloned into the pcDNA3.1/myc-His A (Invitrogen, U.S.A.) (pcChMo/L). All PCR primers used in this study are listed in Table 1. The pcChMo/H vector is about 6.7 kbp and contains the mouse IgG1 constant region gene, as well as the zeocin resistance gene as a selection marker, whereas the pcChMo/L vector is about 6.3 kbp and contains the mouse Cκ gene, as well as the neomycin resistance gene as a selection marker. Chicken anti-DNP antibody was used as a control antibody Fig. 1. Structure of chicken-mouse chimeric antibody expression and was isolated from a chicken naïve phage display library vectors. (A) Heavy chain expression vector, pcChMo/H. (B) [15]. rIgY was generated as described previously [17]. As Light chain expression vector, pcChMo/L. VH, heavy chain variable region gene; CH1/CH2/CH3, heavy chain constant PCR templates, pcChMo/L was used for the light chain vari- region gene; VL, light chain variable region gene; Cκ, light able region, and pcChMo/H was used for the heavy chain chain constant region gene; Amp, gene for ampicillin resistance; variable region. Zeo, gene for zeocin resistance; Neo, gene for neomycin resis- For stable transfection, SP2 cells were co-transfected tance; CMV, cytomegalovirus promoter; 6xHis, histidine repeat with two expression vectors by electroporation as described sequence; Amber, amber stop codon. by Schoonjans et al. [24]. To determine the structure of the chimeric antibody in culture supernatants, culture superna- Switzerland) and synthesized using Super Script III Reverse tants before selection in antibiotics were analyzed by West- Transcriptase (Stratagene, U.S.A.). For cloning of the ern blotting (Fig. 2). The chimeric antibody was detectable mouse IgG1 heavy chain gene, the constant region gene as a single band at about 180 kDa by using anti-mouse IgG (CH1, CH2 and CH3) was amplified from cDNA by PCR antibody (SouthernBiotech, U.S.A.). Moreover, antigen- with KOD Plus DNA polymerase (Toyobo, Japan) using the binding affinity of the chimeric antibody was measured by primers MIgG-CHF and MIgG-CHR. Purified PCR product surface plasmon resonance on a BIAcore 2000 (Biacore AB,

Table 1. PCR primers used in this study Primer Sequence MIgG-CHF 5’-GCCAAAACGACACCCCCATC MIgG-CHR 5’-TACCGGTTTTACCAGGAGAGTGGGAGAGG MVH-bivalent-R 5’-GGAGGAGACGATGACTTCGGT MVH-CHF 5’-GTCATCGTCTCCTCCGCCAAAACGACACCCCC MIgG-CH1R 5’-TCTTGTCCACCTTGGTGCTGC MVL-bivalent-R 5’-GCCGAGGACGGTCAGGGTT MIgG-CκF 5’-GACCGTCCTCGGCGCTGATGCTGCACCAACTG MIgG-CκR 5’-ATCTAGAACACTCATTCCTGTTGAAGCTC Restriction sites are underlined. HUC2-HF4, VH-leader-R, VH-bivalen-F, HUC2- VLF3, VL-leader-R and VL-bivalent-F primers were as described by Shimamoto et al. [26] and Miyamoto et al. [17]. IMMUNOGENICITY OF CHIMERIC CHICKEN ANTIBODY 399

Fig. 2. Western blot analysis for specificity of chimeric anti- body. Culture supernatants from SP2 cells were run on 12.5% SDS-PAGE gels under non-reducing conditions, and were then transferred to a PVDF-membrane and probed. (A) Reacted with peroxidase-labeled anti-chicken IgG-Fab fragment. (B) Reacted with peroxidase-labeled anti-mouse IgG. Lane 1, cul- ture supernatants from SP2 cells co-transfected with expression vectors pcChMo/H and pcChMo/L; lane 2, culture supernatants from SP2 cells transfected with empty vector.

Sweden). Each data set was fit globally to a simple 1:1 Fig. 3. Analysis of chicken-mouse chimeric antibody in ascites. (A) Western blot analysis of chimeric Langmuir binding model using BIAevaluation software ver. antibody in ascites. Ascitic fluid was run under non- 3.0 (Biacore AB). The KD value for the chimeric antibody reducing conditions, and was transferred and reacted was approximately 99.5 nM, while that of the parental with peroxidase-labeled anti-chicken IgG-Fab frag- chicken antibody was 94.4 nM. These results demonstrate ment antibody. (B) Comparison of chimeric anti- that the chimeric antibody retains similar binding affinity as body titers in ascitic fluid and supernatants of the parental chicken antibody. transfected HEK 293-F cells. 1/1 and 1/400 dilu- For large-scale production of chimeric antibody, we tions of chimeric antibody in ascitic fluid ( ) and supernatant of transfected HEK 293-F cells ( ). attempted to grow the cells as ascites in BALB/c mice. After selection in G418 and zeocin, SP2 transfectant clones were screened for antibody production by specific quantita- their genetic distance [6, 7, 10, 12, 13, 20, 28]. In a previous tive ELISA using anti-mouse IgG (Fc) antibody (OEM Con- study, various chicken mAbs against prion protein, a con- cepts, U.S.A.) for capture and peroxidase-labeled anti served mammalian protein, were found to react with human, mouse Cκ antibody (Kirkegaard and Perry Laboratories, mouse, sheep and bovine prion protein [20]. Furthermore, U.S.A.) for detection. Purified mouse IgG was used as a the chicken has approximately 100-fold more B cells than standard. The highest producer was selected. The chimeric the mouse, and thus has a larger initial cell repertoire from antibody was productive at a concentration level of approx- which to select reactive B cells after in vivo priming [23]. imately 1.3 µg/ml. In the ascites of BALB/c mice trans- If chicken mAbs are to be used in clinical applications, planted with 2 × 106 SP2 transfectant cells, chimeric the mAbs must first be examined in mouse models. There- antibody was detectable at concentration level of approxi- fore, they are required to recognize both human antigens and mately 200 µg/ml (data not shown). Although the antibody mouse homologues, and this is a vital property for the devel- production levels in ascitic fluid in this study were not high, opment of therapeutic antibodies. However, the use of production may be improved by using transfected clones chicken mAbs in mice has been restricted by the mouse anti- producing large amounts of antibody. chicken antibody response. Similarly, in humans, the use of Chimeric antibody was also detected by Western blotting mouse mAbs for therapy results in an anti-mouse antibody using the anti-chicken IgG-Fab fragment (Fig. 3A). More- response [25]. To overcome this problem, mouse-human over, ascitic fluid showed a higher antibody titer against chimeric antibodies comprising mouse variable regions and DNP-BSA than chimeric antibody purified from superna- human constant regions have been generated. In clinical tri- tants of transfected HEK 293-F cells (Fig. 3B). als in humans, the degree of human anti-mouse antibody The chicken has various advantages with regard to the response to these chimeric antibodies was lower than that production of antibodies against conserved mammalian pro- induced by the equivalent mouse mAbs [1]. There have teins, as such molecules are immunogenic in chickens due to been several successful reports of reduced immunogenic 400 Y. TATEISHI ET AL.

5. Bruce, C. J., Howard, C. J., Thomas, L. H., Tempest, P. R. and Taylor, G. 1999. Vet. Immunol. Immunopathol. 71: 215–231. 6. Gassmann, M., Thömmes, P., Weiser, T. and Hübscher, U. 1990. FASEB J. 4: 2528–2532. 7. Goueli, S. A., Hanten, J., Davis, A. and Ahmed, K. 1990. Bio- chem. Int. 21: 685–694. 8. Graves, S. S., Goshorn, S. C., Stone, D. M., Axworthy, D. B., Reno, J. M., Bottino, B., Searle, S., Henry, A., Pedersen, J., Rees, A. R. and Libby, R. T. 1999. Clin Cancer Res. 5: 899–908. 9. Jones, P. T., Dear, P. H., Foote, J., Neuberger, M. S. and Win- ter, G. 1986. Nature 321: 522–525. 10. Lemamy, G-J., Roger, P., Mani, J-C., Robert, M., Rochefort, H. and Brouillet, J-P. 1999. Int. J. Cancer 80: 896–902. 11. LoBuglio, A. F., Wheeler, R. H., Trang, J., Haynes, A., Rogers, K., Harvey, E. B., Sun, L., Ghrayeb, J. and Khazaeli, M. B. 1989. Proc. Natl. Acad. Sci. U. S. A. 86: 4220–4224. Fig. 4. Mouse anti-chicken antibody response. Mice were 12. Matsuda, H., Mitsuda, H., Nakamura, N., Furusawa, S., Mohri, injected with 10 µg (i.p.) of chimeric antibody or rIgY. S. and Kitamoto, T. 1999. FEMS Immunol. Med. Microbiol. 23: Antibody titer was measured by ELISA in mice treated 189–194. with chicken-mouse chimeric antibody ( ) or rIgY ( ). 13. Matsushita, K., Horiuchi, H., Furusawa, S., Horiuchi, M., Shi- nagawa, M. and Matsuda, H. 1998. J. Vet. Med. Sci. 60: 777– responses to mouse mAbs in other mammals through the use 779. of chimeric antibodies. A mouse-bovine chimeric antibody 14. Michael, N., Accavitti, M. A., Masteller, E. and Thompson, C. provoked a bovine anti-mouse response about one-tenth of B. 1998. Proc. Natl. Acad. Sci. U. S. A. 95: 1166–1171. that seen with the mouse antibody [5]. Repeated administra- 15. Miyamoto, K., Kimura, S., Nakamura, N., Yokoyama, T., Hori- tion of mouse-cat chimeric antibodies to cats did not induce uchi, H., Furusawa, S. and Matsuda, H. 2007. Biologicals 35: anaphylactic reactions, whereas repeated administration of 303–308. 16. Miyamoto, K., Nakamura, N., Aosasa, M., Nishida, N., mouse resulted in anaphylaxis [30]. Yokoyama, T., Horiuchi, H., Furusawa, S. and Matsuda, H. In the present study, we examined a chicken-mouse chi- 2005. Biochem. Biophys. Res. Commun. 16: 197–204. meric antibody composed of chicken heavy and light chain 17. Miyamoto, K., Shimamoto, T., Aosasa, M., Kimura, S., Naka- variable regions linked to mouse heavy and Cκ light chain mura, N., Okubo, Y., Yokoyama, T., Horiuchi, H., Furusawa, S. constant regions, respectively. Eight-week-old BALB/c and Matsuda, H. 2007. Biologicals 35: 31–34. mice were injected with 10 µg (i.p.) of chimeric antibody 18. Morrison, S. L., Johnson, M. J., Herzenberg, L. A. and Oi, V. T. (n=8) or rIgY (control, n=8) purified from culture superna- 1984. Proc. Natl. Acad. Sci. U. S. A. 81: 6851–6855. tants of transfected HEK 293-F cells, and sera were col- 19. Mount, P. F., Sutton, V.R., Li, W., Burgess, J., McKEnzie, I.F., lected weekly. Antibody titer was determined by ELISA Pietersz, G.A. and Trapani, J.A. 1994. Cancer Res. 54: 6160– using plates coated with rIgY. Antibody titers (mean ± SD) 6166. 20. Nakamura, N., Syuyama, A., Hojyo, S., Shimokawa, M., Miya- in mice treated with chimeric antibody were 30 ± 48.3, moto, K., Kawashima, T., Aosasa, M., Horiuchi, H., Furusawa, while those of the mice treated with rIgY were 866.7 ± S. and Matsuda, H. 2004. J. Vet. Med. Sci. 66: 807–814. 583.1 (Fig. 4). The chicken-mouse chimeric antibody 21. Nishibori, N., Horiuchi, H., Furusawa, S. and Matsuda, H. appeared to be significantly less immunogenic than rIgY. 2006. Mol. Immunol. 43: 634–642. In summary, the present chicken-mouse chimeric anti- 22. Nishibori, N., Shimamoto, T., Nakamura, N., Shimokawa, M., body system may provide a useful tool for diagnostics, as Horiuchi, H., Furusawa, S. and Matsuda, H. 2004. Biologicals well as preclinical mouse studies, by eliminating the mouse 32: 213–218. anti-chicken antibody response. 23. Pink, J. R. 1986. Immunol. Rev. 91: 115–128. 24. Schoonjans, R., Willems, A., Schoonooghe, S., Leoen, J., ACKNOWLEDGMENT. This work was supported in part Grooten, J. and Mertens, N. 2001. Biomol. Eng. 17: 193–202. 25. Shawler, D. L., Bartholomew, R. M., Smith, L. M. and Dillman, by grants from the New Energy and Industrial Technology R. O. 1985. J. Immunol. 35: 1530–1535. Development Organization (NEDO) of Japan. 26. Shimamoto, T., Nishibori, N., Aosasa, M., Horiuchi, H., Furu- sawa, S. and Matsuda, H. 2005. Biologicals 33: 169–174. REFERENCES 27. Shimizu, M., Nagashima, H., Sano, K., Hashimoto, K., Ozeki, M., Tsuda, K. and Hatta, H. 1992. Biosci. Biotechnol. Biochem. 1. Adair, J. R. 1992. Immunol. Rev. 130: 5–40. 56: 270–274. 2. Adams, C. W., Allison, D. E., Flagella, K., Presta, L., Clarke, J., 28. Song, C. S., Yu, J. H., Bai, D. H., Hester, P. Y. and Kim, K. H. Dybdal, N., Mckeever, K. and Sliwkowski, M. X. 2006. Cancer 1985. J. Immunol. 135: 3354–3359. Immunol. Immunother. 55: 717–727. 29. Stephens, S., Emtage, S., Vetterlein, O., Chaplin, L., Bebbing- 3. Bose, B., Khanna, N., Acharya, S. K. and Sinha, S. 2006. J. ton, C., Nesbitt, A., Sopwith, M., Athwal, D., Novak, C. and Gastroenterol. Hepatol. 21: 1439–1447. Bodmer, M. 1995. Immunology 85: 668–674. 4. Boulianne, G. L., Hozumi, N. and Shulman, M. J. 1984. Nature 30. Umehashi, M., Imamura, T., Akiyama, S., Kimachi, K., Tokiy- 312: 643–646. oshi, S. and Mikami, T. 2002. J. Vet. Med. Sci. 64: 1017–1021.