ANTI-TNF ALPHA ANTIBODIES WITH PH-DEPENDEND ANTIGEN BINDING FOR IMPROVED TARGET CLEARENCE

Abstract

The invention relates to anti-TNFa antibodies which are engineered to exhibit a pH-sensitive antigen binding. The invention is preferably directed to anti-TNFa antibody adalimumab (Humira®) or biologically active variants and fragments thereof, wherein the original adalimumab antibody or variant or fragment thereof is engineered by modifications of amino acid sequence within the variable regions. Specifically, the invention relates to adalimumab or biologically active variants or fragments thereof, wherein the CDR domains are modified by replacing one or more amino acid residues by histidine residues.

Claims

1. A human antibody or an antigen binding fragment thereof with a pH dependent antigen binding, comprising: light and heavy chain variable regions of human antibody adalimumab or a variant thereof with same or similar biological activity, wherein at least one of the CDR domains of the light and/or the heavy chain variable regions is mutated by replacement of one or more amino acids within said CDR domains by a histidine residue, thereby generating a mutated adalimumab or adalimumab variant eliciting a pH dependent antigen binding with an antigen dissociation rate (K.sub.dis) ratio pH 6/pH 7 measured by biolayer interferometry which is at least 5 fold higher compared to a respective K.sub.dis rate ratio of non-mutated adalimumab.

2. The human antibody or the antigen binding fragment thereof of claim 1, wherein the mutated antibody or antigen binding fragment thereof has a reduced antigen binding affinity, which is at least 1% of the binding affinity of the non-mutated adalimumab.

3. The human antibody or the antigen binding fragment thereof of claim 1, comprising a CDR3 heavy chain amino acid sequence selected from the group consisting of: TABLE-US-00027 (SEQ ID NO: 12) VSYHSTASSLDY, (SEQ ID NO: 13) VSYLSTAHHLDY, (SEQ ID NO: 14) VSYHSTAHHLDY, and VHYHSTASSLDY (SEQ ID NO: 31).

4. The human antibody or the antigen binding fragment thereof of claim 1, comprising a CDR1 light chain amino acid sequence selected from the group consisting of: TABLE-US-00028 (SEQ ID NO: 15) RASQGIRNHLA [[,]] and (SEQ ID NO: 16) RASQGIRNHHA.

5. The human antibody or the antigen binding fragment thereof of claim 1, comprising the CDR2 light chain amino acid sequence of: TABLE-US-00029 (SEQ ID NO: 32) AAHTLQS.

6. The human antibody or the antigen binding fragment thereof of claim 1, comprising a CDR3 light chain amino acid sequence selected from the group consisting of: TABLE-US-00030 (SEQ ID NO: 17) HHYHRAPYT, (SEQ ID NO: 18) QHYHRAPYH and (SEQ ID NO: 38) QRHNRAPYT.

7. The human antibody or the antigen binding fragment thereof of claim 1, comprising: a CDR3 heavy chain amino acid sequence of claim 3 selected from the group consisting of TABLE-US-00031 (SEQ ID NO: 12) VSYHSTASSLDY, (SEQ ID NO: 13) VSYLSTAHHLDY, (SEQ ID NO: 14) VSYHSTAHHLDY,  and (SEQ ID NO: 31) VHYHSTASSLDY, and a CDR1 light chain amino acid sequence selected from the group consisting of TABLE-US-00032 (SEQ ID NO: 15) RASQGIRNHLA and (SEQ ID NO: 16) RASQGIRNHHA.

8. The human antibody or the antigen binding fragment thereof of claim 1, comprising: a CDR3 heavy chain amino acid sequence selected from the group consisting of TABLE-US-00033 (SEQ ID NO: 12) VSYHSTASSLDY, (SEQ ID NO: 13) VSYLSTAHHLDY, (SEQ ID NO: 14) VSYHSTAHHLDY,  and (SEQ ID NO: 31) VHYHSTASSLDY, and a CDR3 light chain amino acid sequence selected from the group consisting of TABLE-US-00034 (SEQ ID NO: 17) HHYHRAPYT, (SEQ ID NO: 18) QHYHRAPYH,  and (SEQ ID NO: 38) QRHNRAPYT.

9. The human antibody or the antigen binding fragment thereof of claim 1, comprising: a CDR3 heavy chain amino acid sequence selected from the group consisting of TABLE-US-00035 (SEQ ID NO: 12) VSYHSTASSLDY, (SEQ ID NO: 13) VSYLSTAHHLDY, (SEQ ID NO: 14) VSYHSTAHHLDY,  and (SEQ ID NO: 31) VHYHSTASSLDY, the CDR2 light chain amino acid sequence AAHTLQS (SEQ ID NO: 32), and a CDR3 light chain amino acid sequence selected from the group consisting of TABLE-US-00036 (SEQ ID NO: 17) HHYHRAPYT, (SEQ ID NO: 18) QHYHRAPYH,  and (SEQ ID NO: 38) QRHNRAPYT.

10. The human antibody or the antigen binding fragment thereof of claim 1, comprising: a CDR3 heavy chain amino acid sequence selected from the group consisting of TABLE-US-00037 (SEQ ID NO: 12) VSYHSTASSLDY, (SEQ ID NO: 13) VSYLSTAHHLDY, (SEQ ID NO: 14) VSYHSTAHHLDY,  and (SEQ ID NO: 31) VHYHSTASSLDY, a CDR1 light chain amino acid sequence selected from the group consisting of TABLE-US-00038 (SEQ ID NO: 15) RASQGIRNHLA and (SEQ ID NO: 16) RASQGIRNHHA, and a CDR3 light chain amino acid sequence selected from the group consisting of TABLE-US-00039 (SEQ ID NO: 17) HHYHRAPYT, (SEQ ID NO: 18) QHYHRAPYH,  and (SEQ ID NO: 38) QRHNRAPYT.

11. The human antibody or the antigen binding fragment thereof of claim 1, comprising a variable light chain amino acid sequence selected from the group consisting of: TABLE-US-00040 (i) (SEQ ID NO: 28) DIQMTQSPSS LSASVGDRVT ITCRASQGIR NHHAWYQQKP GKAPKLLIYA ASTLQSGVPS RFSGSGSGTD FTLTISSLQP EDVATYYCQR YNRAPYTFGQ GTKVEIK, (ii) (SEQ ID NO: 29) DIQMTQSPSS LSASVGDRVT ITCRASQGIR NYLAWYQQKP GKAPKLLIYA ASTLQSGVPS RFSGSGSGTD FTLTISSLQP EDVATYYCHH YHRAPYTFGQ GTKVEIK [[.]], (iii) (SEQ ID NO: 30) DIQMTQSPSS LSASVGDRVT ITCRASQGIR NYLAWYQQKP GKAPKLLIYA ASTLQSGVPS RFSGSGSGTD FTLTISSLQP EDVATYYCQH YHRAPYHFGQ GTKVEIK [[.]], (iv) (SEQ ID NO: 33) DIQMTQSPSS LSASVGDRVT ITCRASQGIR NYLAWYQQKP GKAPKLLIYA ASTLQSGVPS RFSGSGSGTD FTLTISSLQP EDVATYYCQH YHRAPYTFGQ GTKVEIK [[.]], (v) (SEQ ID NO: 34) DIQMTQSPSS LSASVGDRVT ITCRASQGIR NHLAWYQQKP GKAPKLLIYA ASTLQSGVPS RFSGSGSGTD FTLTISSLQP EDVATYYCQR YNHAPYTFGQ GTKVEIK, (vi) (SEQ ID NO: 35) DIQMTQSPSS LSASVGDRVT ITCRASQGIR NHLAWYQQKP GKAPKLLIYA ASTLQSGVPS RFSGSGSGTD FTLTISSLQP EDVATYYCQR YNRAPYTFGQ GTKVEIK, (vii) (SEQ ID NO: 36) DIQMTQSPSS LSASVGDRVT ITCRASQGIR NYLAWYQQKP GKAPKLLIYA ASTLQSGVPS RFSGSGSGTD FTLTISSLQP EDVATYYCQR HNRAPYTFGQ GTKVEIK, and (viii) (SEQ ID NO: 37) DIQMTQSPSS LSASVGDRVT ITCRASQGIR NHLAWYQQKP GKAPKLLIYA AHTLQSGVPS RFSGSGSGTD FTLTISSLQP EDVATYYCQR YNRAPYTFGQ GTKVEIK.

12. The human antibody or the antigen binding fragment thereof of claim 1, comprising a variable heavy chain amino acid sequence selected from the group consisting of: TABLE-US-00041 (i) (SEQ ID NO: 25) EVQLVESGGG LVQPGRSLRL SCAASGFTFD DYAMHWVRQA PGKGLEWVSA ITWNSGHIDY ADSVEGRFTI SRDNAKNSLY  LQMNSLRAED TAVYYCAKVS YHSTASSLDY WGQGTLVTVS S; (ii) (SEQ ID NO: 26) EVQLVESGGG LVQPGRSLRL SCAASGFTFD DYAMHWVRQA PGKGLEWVSA ITWNSGHIDY ADSVEGRFTI SRDNAKNSLY  LQMNSLRAED TAVYYCAKVS YLSTAHHLDY WGQGTLVTVS S;  and (iii) (SEQ ID NO: 39) EVQLVESGGG LVQPGRSLRL SCAASGFTFD DYAMHWVRQA PGKGLEWVSA ITWNSGHIDY ADSVEGRFTI SRDNAKNSLY  LQMNSLRAED TAVYYCAKVH YHSTASSLDY WGQGTLVTVS S.

13. The human antibody or the antigen binding fragment thereof of claim 1, comprising: a variable light chain amino acid sequence selected from the group consisting of: TABLE-US-00042 (i) (SEQ ID NO: 28) DIQMTQSPSS LSASVGDRVT ITCRASQGIR NHHAWYQQKP GKAPKLLIYA ASTLQSGVPS RFSGSGSGTD FTLTISSLQP EDVATYYCQR YNRAPYTFGQ GTKVEIK, (ii) (SEQ ID NO: 29) DIQMTQSPSS LSASVGDRVT ITCRASQGIR NYLAWYQQKP GKAPKLLIYA ASTLQSGVPS RFSGSGSGTD FTLTISSLQP EDVATYYCHH YHRAPYTFGQ GTKVEIK [[.]], (iii) (SEQ ID NO: 30) DIQMTQSPSS LSASVGDRVT ITCRASQGIR NYLAWYQQKP GKAPKLLIYA ASTLQSGVPS RFSGSGSGTD FTLTISSLQP EDVATYYCQH YHRAPYHFGQ GTKVEIK [[.]], (iv) (SEQ ID NO: 33) DIQMTQSPSS LSASVGDRVT ITCRASQGIR NYLAWYQQKP GKAPKLLIYA ASTLQSGVPS RFSGSGSGTD FTLTISSLQP EDVATYYCQH YHRAPYTFGQ GTKVEIK [[.]], (v) (SEQ ID NO: 34) DIQMTQSPSS LSASVGDRVT ITCRASQGIR NHLAWYQQKP GKAPKLLIYA ASTLQSGVPS RFSGSGSGTD FTLTISSLQP EDVATYYCQR YNHAPYTFGQ GTKVEIK, (vi) (SEQ ID NO: 35) DIQMTQSPSS LSASVGDRVT ITCRASQGIR NHLAWYQQKP GKAPKLLIYA ASTLQSGVPS RFSGSGSGTD FTLTISSLQP EDVATYYCQR YNRAPYTFGQ GTKVEIK, (vii) (SEQ ID NO: 36) DIQMTQSPSS LSASVGDRVT ITCRASQGIR NYLAWYQQKP GKAPKLLIYA ASTLQSGVPS RFSGSGSGTD FTLTISSLQP EDVATYYCQR HNRAPYTFGQ GTKVEIK, and (viii) (SEQ ID NO: 37) DIQMTQSPSS LSASVGDRVT ITCRASQGIR NHLAWYQQKP GKAPKLLIYA AHTLQSGVPS RFSGSGSGTD FTLTISSLQP EDVATYYCQR YNRAPYTFGQ GTKVEIK. and a variable heavy chain amino acid sequence selected from the group consisting of: TABLE-US-00043 (i) (SEQ ID NO: 25) EVQLVESGGG LVQPGRSLRL SCAASGFTFD DYAMHWVRQA PGKGLEWVSA ITWNSGHIDY ADSVEGRFTI SRDNAKNSLY  LQMNSLRAED TAVYYCAKVS YHSTASSLDY WGQGTLVTVS S; (ii) (SEQ ID NO: 26) EVQLVESGGG LVQPGRSLRL SCAASGFTFD DYAMHWVRQA PGKGLEWVSA ITWNSGHIDY ADSVEGRFTI SRDNAKNSLY  LQMNSLRAED TAVYYCAKVS YLSTAHHLDY WGOGTLVTVS S;  and (iii)  (SEQ ID NO: 39) EVQLVESGGG LVQPGRSLRL SCAASGFTFD DYAMHWVRQA PGKGLEWVSA ITWNSGHIDY ADSVEGRFTI SRDNAKNSLY  LQMNSLRAED TAVYYCAKVH YHSTASSLDY WGQGTLVTVS S.

14. A human antibody or an antigen binding fragment thereof with a pH dependent antigen binding, comprising: light and heavy chain variable regions of human antibody adalimumab a variant thereof with the same or similar biological activity, wherein at least one of the CDR domains of the light and/or the heavy chain variable regions is mutated by replacing one or more amino acids within said CDR domains by a histidine residue, thereby generating a mutated adalimumab eliciting a significant pH dependent antigen binding, said mutated adalimumab comprising: the variable heavy chain amino acid sequence TABLE-US-00044 (SEQ ID NO: 25) EVQLVESGGG LVQPGRSLRL SCAASGFTFD DYAMHWVRQA PGKGLEWVSA ITWNSGHIDY ADSVEGRFTI SRDNAKNSLY  LQMNSLRAED TAVYYCAKVS YHSTASSLDY WGQGTLVTVSS and the variable light chain amino acid sequence TABLE-US-00045 (SEQ ID NO: 27) DIQMTQSPSS LSASVGDRVT ITCRASQGIR NHHAWYQQKP GKAPKLLIYA ASTLQSGVPS RFSGSGSGTD FTLTISSLQP EDVATYYCQR YNRAPYTFGQ GTKVEIK .

15. A human antibody or an antigen binding fragment thereof with a pH dependent antigen binding, comprising: light and heavy chain variable regions of human antibody adalimumab a variant thereof with the same or similar biological activity, wherein at least one of the CDR domains of the light and/or the heavy chain variable regions is mutated by replacing one or more amino acids within said CDR domains by a histidine residue, thereby generating a mutated adalimumab eliciting a pH dependent antigen binding, said mutated adalimumab comprising: the variable heavy chain amino acid sequence TABLE-US-00046 (SEQ ID NO: 26) EVQLVESGGG LVQPGRSLRL SCAASGFTFD DYAMHWVRQA PGKGLEWVSA ITWNSGHIDY ADSVEGRFTI SRDNAKNSLY LQMNSLRAED TAVYYCAKVS YLSTAHHLDY WGQGTLVTVS S and the variable light chain amino acid sequence TABLE-US-00047 (SEQ ID NO: 28) DIQMTQSPSS LSASVGDRVT ITCRASQGIR NYLAWYQQKP GKAPKLLIYA ASTLQSGVPS RFSGSGSGTD FTLTISSLQP EDVATYYCHH YHRAPYTFGQ GTKVEIK, or the variable light chain amino acid sequence TABLE-US-00048 (SEQ ID NO: 29) DIQMTQSPSS LSASVGDRVT ITCRASQGIR NYLAWYQQKP GKAPKLLIYA ASTLQSGVPS RFSGSGSGTD FTLTISSLQP EDVATYYCQH YHRAPYHFGQ GTKVEIK .

16. The human antibody or the antigen binding fragment thereof of claim 1 comprising the human heavy chain IgG1 constant region of SEQ ID NO: 11.

17. The human antibody or the antigen binding fragment thereof of claim 16, wherein a Fc portion of said IgG1 constant region is mutated at one or more amino acid positions resulting in an respective antibody with modified FcRn binding.

18. The human antibody or the antigen binding fragment thereof of claim 16, comprising a human kappa constant region.

19. An antibody-drug conjugate comprising the human antibody or an antibody fragment thereof of claim 1 linked directly or indirectly to a cytotoxic chemical drug or recombinantly fused to a cytokine.

20. A pharmaceutical composition suitable for the treatment of inflammatory, autoimmune or cancer diseases comprising: the human antibody or the antigen binding fragment thereof of claim 1, or an antibody-drug conjugate comprising the antibody or the antibody fragment thereof of claim 1 linked directly or indirectly to a cytotoxic chemical drug or recombinantly fused to a cytokine together with a pharmaceutically acceptable carrier, diluent or excipient.

21. A method of treating TNFa induced inflammatory, autoimmune or cancer diseases, the method comprising: administering an effective amount of the human antibody, or the antigen binding fragment thereof of claim 1 or an antibody-drug conjugate comprising the human antibody or the antibody binding fragment thereof of claim 1 linked directly or indirectly to a cytotoxic chemical drug or recombinantly fused to a cytokine, to a subject in need thereof.

22. A method for manufacture of a medicament for treating TNFa induced inflammatory, autoimmune or cancer diseases, the method comprising: mutating at least one of CDR domains of light and/or the heavy chain variable regions of human antibody adalimumab or a variant thereof having the same or similar biological activity by replacement at least one amino acid within the at least one of CDR domains by a histidine residue, thereby generating a mutated adalimumab or adalimumab variant eliciting a pH dependent antigen binding with an antigen dissociation rate (K.sub.dis) ratio pH 6/pH 7 measured by biolayer interferometry which is at least 5 fold higher compared to a respective K.sub.dis rate ratio of non-mutated adalimumab.

Description

SHORT DESCRIPTION OF THE FIGURES

[0095] FIG. 1: Proposed differences between a) conventional and b) pH-dependent antibody on soluble antigen binding (Modified from: Igawa et al., 2010).

[0096] FIG. 2: Adalimumab amino acid sequences of the variable regions. The complementary determining regions are highlighted with red boxes. a) Variable region of the heavy chain (VH). b) Variable region of the light chain (VL).

[0097] FIG. 3: Aligned protein sequences of seven unique VH variants with the parental VH. Unique sequences were isolated from the heavy chain library approach after three rounds of screening. Parental VH sequence is shown on top and residues that vary from the parental VH are highlighted for every variant.

[0098] FIG. 4A: First part of protein sequence alignment (1-3 parts) of the parental VL and 38 unique VL variants that were isolated from the light chain library after three rounds of screening. Parental VL sequence is shown on top and residues that vary from the parental VL are highlighted for every variant.

[0099] FIG. 4B: Second part of protein sequence alignment (1-3 parts) of the parental VL and 38 unique VL variants that were isolated from the light chain library after three rounds of screening. Parental VL sequence is shown on top and residues that vary from the parental VL are highlighted for every variant.

[0100] FIG. 4C: Third part of protein sequence alignment (1-3 parts) of the parental VL and 38 unique VL variants that were isolated from the light chain library after three rounds of screening. Parental VL sequence is shown on top and residues that vary from the parental VL are highlighted for every variant.

[0101] FIG. 5: Octet Red sensorgrams of adalimumab and three variants with pH-dependent binding to rhTNFα. Association was done with rhTNFα concentrations ranging between 0.26 nM-2 nM at pH 7.4 and dissociation was carried out at pH 7.4. Kinetics binding constants were determined through global fitting using Octet 8.0 Software.

[0102] FIG. 6: Octet Red sensorgrams of adalimumab and three variants with pH-dependent binding. Association was done with rhTNFα concentrations ranging between 0.26 nM-2 nM at pH 7.4 and dissociation was carried out at pH 6. Off-rates were determined for PSV#1, PSV#2 and PSV#3 through local partial fitting using Octet 8.0 Software. Off-rates for adalimumab were generated by using global fitting.

[0103] FIG. 7: Octet Red sensorgrams of adalimumab and three variants with pH-dependent binding (PSV#1, PSV#2, PSV#3). Adalimumab shows fast association of rhTNFα and maintains tight binding during the dissociation step at pH 6. In contrast, pH-dependent binding variants show reversible rhTNFα binding at pH 7.4 after fast release of rhTNFα during the dissociation step at pH 6. Association to 13 nM rhTNFα was measured for 200 sec and dissociation carried out for 400 sec. After two binding cycles the last dissociation step was done at pH 7.4, showing slow release of rhTNFα.

REFERENCES

[0104] 1. Aggarwal B. B., Signalling pathways of the TNF superfamily: a double-edged sword. Nat Rev Immunol. 2003 September; 3(9):745-56. Review [0105] 2. Aggarwal R. S., What's fueling the biotech engine-2012 to 2013.Nat Biotechnol. 1 (2014) 32-9. [0106] 3. Atmanene C., Wagner-Rousset E., Malissard M., Chol B., Robert A., Corvaïa N., Van Dorsselaer A., Beck A., Sanglier-Cianférani S., Extending mass spectrometry contribution to therapeutic monoclonal antibody lead optimization: characterization of immune complexes using noncovalent ESI-MS. Anal Chem. 15 (2009) 16364-73. [0107] 4. Boder E. T., Wittrup K. D., Yeast surface display for screening combinatorial polypeptide libraries, Nat. Biotechnol. 6 (1997) 553-7 [0108] 5. Carter P. J., Introduction to current and future protein therapeutics: a protein engineering perspective. Exp Cell Res. 9 (2011) 1261-9 [0109] 6. Chaparro-Riggers J., Liang H., DeVay R. M., Bai L., Sutton J. E., Chen W., Geng T., Lindquist K., Casas M. G., Boustany L. M., Brown C. L., Chabot J., Gomes B., Garzone P., Rossi A., Strop P., Shelton D., Pons J., Rajpal A., Increasing serum half-life and extending cholesterol lowering in vivo by engineering antibody with pH-sensitive binding to PCSK9. J Biol Chem. 14 (2012) 11090-7. [0110] 7. Choy, E. H. S., Panayi, G. S., Cytokine pathways and joint inflammation in rheumatoid arthritis. N Engl J Med. 12 (2001) 907-16 [0111] 8. Dall'Acqua W. F., Kiener P. A., Wu H., Properties of human IgG1s engineered for enhanced binding to the neonatal Fc receptor (FcRn). J Biol Chem 281 (2006) 23514-24. [0112] 9. Feldmann, M., Development of anti-TNF therapy for rheumatoid arthritis. Nat Rev Immunol 2 (2002), 364-371 [0113] 10. Finkelman F. D., Madden K. B., Morris S. C., Holmes J. M., Boiani N., Katona I. M., Maliszewski C. R., Anti-cytokine antibodies as carrier proteins. Prolongation of in vivo effects of exogenous cytokines by injection of cytokine-anti-cytokine antibody complexes. J Immunol. 3 (1993) 1235-44 [0114] 11. Gera N., Hill A. B., White D. P., Carbonell R. G, Rao B. M., Design of pH sensitive binding proteins from the hyperthermophilic Sso7d scaffold. PLoS One. 7 (2012) [0115] 12. Humira (adalimumab) prescribing information. Abbott Laboratories. September 2013. Available at: http://www.rxabbott.com/pdf/humira.pdf. Accessed Apr. 21, 2014. [0116] 13. Horiuchi T., Mitoma H, Harashima S, Tsukamoto H, Shimoda T. Transmembrane TNF-alpha: structure, function and interaction with anti-TNF agents. Rheumatology (Oxford). 7 (2010) 1215-28 [0117] 14. Igawa T., Ishii S., Tachibana T., Maeda A., Higuchi Y., Shimaoka S., Moriyama C., Watanabe T., Takubo R., Doi Y., Wakabayashi T., Hayasaka A., Kadono S., Miyazaki T., Haraya K., Sekimori Y., Kojima T., Nabuchi Y., Aso Y., Kawabe Y., Hattori K., Antibody recycling by engineered pH-dependent antigen binding improves the duration of antigen neutralization. Nat Biotechnol. 11 (2010) 1203-7 [0118] 15. Igawa T., Maeda A., Haraya K., Tachibana T., Iwayanagi Y., Mimoto F., Higuchi Y., Ishii S., Tamba S., Hironiwa N., Nagano K., Wakabayashi T., Tsunoda H., Hattori K., Engineered monoclonal antibody with novel antigen-sweeping activity in vivo. PLoS One. 8 (2013) [0119] 16. Ito W., Sakato N., Fujio H., Yutani K., Arata Y., Kurosawa Y., The His-probe method: effects of histidine residues introduced into the complementarity-determining regions of antibodies on antigen-antibody interactions at different pH values. FEBS Lett. 1 (1992) 85-8. [0120] 17. Kaymakcalan Z., Sakorafas P., Bose S., Scesney S., Xiong L., Hanzatian D. K., Salfeld J., Sasso E. H., Comparisons of affinities, avidities, and complement activation of adalimumab, infliximab, and etanercept in binding to soluble and membrane tumor necrosis factor. Clin Immunol. 2 (2009) 308-16 [0121] 18. Kuo T. T., Aveson V. G., Neonatal Fc receptor and IgG-based therapeutics. MAbs. 5 (2011) 422-30 [0122] 19. Murtaugh M. L., Fanning S. W., Sharma T. M., Terry A. M., Horn J. R., A combinatorial histidine scanning library approach to engineer highly pH-dependent protein switches. Protein Sci. 9 (2011) 1619-31 [0123] 20. Roopenian D. C., Akilesh S., FcRn: the neonatal Fc receptor comes of age. Nat Rev Immunol. 9 (2007) 715-25 [0124] 21. Sarkar C. A., Lowenhaupt K., Horan T., Boone T. C., Tidor B., Lauffenburger D. A., Rational cytokine design for increased lifetime and enhanced potency using pH-activated histidine switching. Nat Biotechnol. 9 (2002) 908-13 [0125] 22. Schabert V. F., Watson C., Joseph G. J., Iversen P., Burudpakdee C., Harrison D. J., Costs of tumor necrosis factor blockers per treated patient using real-world drug data in a managed care population. J Manag Care Pharm. 8 (2013) 621-30. [0126] 23. Schottelius, A. J. G., Moldawer, L. L., Dinarello, C. A., Asadullah, K., Sterry, W., & Edwards, C. K. Biology of tumor necrosis factor-alphaimplications for psoriasis. Exp Dermatol 13 (2004) 193-222. [0127] 24. Tracey D, Klareskog L, Sasso E H, Salfeld J G, Tak P P. Tumor necrosis factor antagonist mechanisms of action: a comprehensive review. Pharmacol Ther. 2 (2008) 244-79. [0128] 25. Wajant H, Pfizenmaier K, Scheurich P. Tumor necrosis factor signaling. Cell Death Differ. 1 (2003) 45-65.

Examples

Example 1: Selection of pH-Sensitive Anti-TNFα Antibodies Derived from Adalimumab

[0129] Selection of anti-TNFα antibody fragments from combinatorial histidine substitution libraries by yeast surface display: Based on the heavy and light chains of adalimumab, two antibody libraries were synthesized by Geneart, Regensburg by using pre-assembled trinucelotides building blocks. During the synthesis either parental or histidine residues were sampled whereby sampling of histidines was restricted to the complementary determining regions (CDRs) of the heavy and light chains. Most adalimumab library members carried three histidine residues that were spread over all three CDRs (FIG. 2) but variants were also synthesized that carried more or less histidine substitutions (ranging between 0 to −20). Theoretical diversities: Heavy chain library ˜10′000 variants, light chain library ˜3000 variants.

[0130] Both libraries were separately subcloned into plasmid vectors by gap-repair cloning in the EBY100 yeast strain that allows covalent yeast surface display of antibody Fab-fragments (Boder and Wittrup, 1997). Corresponding parental chains were paired with the heavy or light chain libraries and the two resulting libraries were separately screened by fluorescence activated cell sorting (FACS). Cells that carried pH-sensitive adalimumab variants were subsequently enriched over three rounds of screening by applying a specific staining & selection strategy (not explained here).

[0131] Fab-fragments were selected that do reversible high affinity (KD within sub nanomolar ranges) binding to recombinant human TNFα (rhTNFα) at pH 7.4, once after rhTNFα has been released within 30 minutes at pH 6. After three rounds of screening for variants that bind to rhTNFα in pH-dependent manner, sequence analysis of isolated single clones revealed variants that carried specific histidine substitutions patterns (shown in FIGS. 3 & 4). One mutational hot-spot was identified within the CDR3 region of the heavy chain sequence data set. Two mutational hot-spots were identified within CDR-L1 and CDR-L3 regions of the light chain sequence data set. Only abundant heavy and light chain variants (occurrence within analyzed sequence set: N>1) were selected for further processing & characterization, resulting in eight light chain and three heavy chain candidates shown in Table 1 and Table 2:

TABLE-US-00024 TABLE 1 Three abundant (N > 1) heavy chain sequences within 38 isolated single clones after three rounds of screening. Number of His substitutions Variant Abundance (Region) code 29/38  1 (CDR H3) VH#1 3/38 2 (CDR H3) VH#2 2/38 2 (CDR H3) VH#5

TABLE-US-00025 TABLE 2 Eight abundant (N > 1) light chain sequences within 98 isolated single clones after three rounds of screening. Number of His substitutions Abundance (Region) Variant code 36/98  1 (CDR L1) VL#3 13/98  2 (CDR L3) VL#5 5/98 3 (CDR L3) VL#9 4/98 3 (CDR L3) VL#16 3/98 2 (CDR L1) VL#14 3/98 1 (CDR L3) VL#6 2/98 1 (CDR L1) VL#1 1 (CDR L3) 2/98 1 (CDR L1) VL#22 1 (CDR L2)

[0132] Variable regions of abundant VH/VL variants as well as parental adalimumab sequences were cloned into vectors that allow expression of full length IgG1 K molecules in mammalian cells (HEK293 & Expi293). All possible combinations of heavy and light chain variants were co-expressed in mammalian cells. Herein 24 heavy chain and light chain variant combinations were expressed as well as 11 IgG species that derived from combinations of the heavy or light chain variants with the corresponding parental chains. Initial Octet Red experiments with immobilized antibodies assessed differential dissociation behavior at pH 6 or pH 7.4 after associating rhTNFα at pH 7.4. Ten variants were selected according to their binding profiles in regard of high affinity binding at pH 7.4 (association/dissociation at pH 7.4) and fast release of the antigen at pH 6 (association at pH 7.4/dissociation at pH 6). For further characterization antibodies were purified via protein-A purification from crude supernatants and finally buffer was exchanged to PBS. Further octet assays revealed differences in binding at pH 7.4 and differences in rhTNFα release at pH 6.

[0133] Subsequently three variants (IDs according to FIGS. 3-4: VH#2+VL#9: PSV#1, VH#2+VL#16: PSV#2 and VH#1+VL#14: PSV#3) were selected for final characterization on Octet Red.

[0134] Binding characteristics of several histidine-mutated adalimumab variants were analyzed in Octet Red experiments. Different histidine mutations in heavy and light chains as well as different heavy and light chain variant combinations have been shown to affect binding affinities at pH 7.4 and the dissociation rates at pH 6.

[0135] A combination of several mutations including Leu98His in the heavy chain and Tyr32His, Leu33His in the light chain generated PSV#3. Light chain mutations GIn89His, Arg90His and Asn92His generated the light chain of PSV#2. The mutations Arg90His, Asn92His and Thr97His generated the light chain of PSV#1. For both, PSV#1 and PSV#2, mutations of Ser100.bHis and Ser100.cHis generated the heavy chain.

[0136] (The sequences were numbered as shown in FIGS. 3 and 4a-c according to kabat numbering. For this purpose, sequences of the variants were aligned together with the parental sequence by using Clustal W and numbering was applied considering the rules for kabat numbering.)

[0137] Representative sensorgrams of the Octet Red measurements are shown in FIGS. 5 & 6 and corresponding mean values (N=3) of calculated kinetic parameters for adalimumab, PSV#1, PSV#2 and PSV#3 are shown in table 4.

TABLE-US-00026 TABLE 3 Binding kinetics of adalimumab and pH-dependent binding variants to rhTNFα at pH 7.4 and pH 6. Association rate (kon), dissociation rate (kdis) and binding affinity (KD) of adalimumab, PSV#1, PSV#2 and PSV#3 at pH 7.4. Dissociation rates were determined also at pH 6. Experiments were done at 25° C. and for every experiment mean values of triplicates are shown (exception: Adalimumab was measured at pH 6 in duplicates). Representative sensorgrams that correspond to the data are shown in FIGS. 7 and 8. pH 6 pH 7.4 kdis ratio kon kdis ratio (pH 6) vs. Antibody KD (M) (M.sup.−1s.sup.−1) kdis (s.sup.−1) kdis (s.sup.−1) pH 6/pH 7.4 adalimumab Adalimumab 0.46E−11 1.32E+06 0.55E−05 .sup. 4.81E−05 9 1 PSV#1 4.63E−11 0.67E+06 3.26E−05  754E−05 231 157 PSV#2 7.73E−11 1.14E+06 9.35E−05  7340E−05 785 1527 PSV#3 11.2E−11 1.93E+06 21.8E−05 11000E−05 505 2293

[0138] Parental adalimumab binds with high affinity to rhTNFα at pH 7.4 that was also shown in Kaymakcalan et al., 2009. For the three variants the increasing kdis-values result in a decrease in binding affinity (10-24fold decrease in KD), however interaction with TNFα with picomolar binding affinities in the three-diget range still represents very tight binding (FIG. 5 and table 3).

[0139] Octet Red measurements were also performed to assess the antibodies' pH-sensitivities. Improved antibody efficacy in context of the FcRn-mediated recycling requires tight binding to TNFα in the circulation at pH 7.4 and its fast release in the acidic endosome. In order to evaluate the release of TNFα within the acidic endosome, dissociation was measured at pH 6 after association of rhTNFα at pH 7.4 (FIG. 6 and table 3). Ratios of dissociation rates at pH 6 and pH 7.4 were determined and all variants showed considerably increased dissociation at pH 6 (59-160fold increased kdis at pH 6, in contrast to adalimumab with a kdis ratio (pH6/pH7.4) of 6 (table 3).

[0140] One additional experiment addressed the ability of PSV#1, PSV#2 and PSV#3 to reversibly bind rhTNFα at pH 7.4, after dissociation at pH 6. To ensure that incubation at pH 6 does not irreversibly change TNFα binding capabilities, two cycles of binding (pH 7.4) and release (pH 6) were performed (FIG. 7). As shown in FIG. 8, PSV#1, PSV#2 and PSV#3 can reversibly bind after TNFα has been released at pH 6. In contrast to that, adalimumab maintains tight binding during incubation at pH 6.

Example 2: In Vivo Characterization of Mutants

[0141] The effect of parental IgG construct and mutants thereof on the PK of human TNFα was investigated in heterozygous transgenic human FcRn mice, line 176, as well as in homozygous line 32 mice. The former line was more suitable to investigate PK differences between administered IgG constructs, while the latter mouse line provided a better FcRn protection, resulting in longer half-lives, closer to what was expected in human. This longer residence of the scavenger allowed better to evaluate the impact of the antibody on the clearance of the TNFα.

[0142] Human TNFα and scavenger were administered by SC route in predefined ratios. Plasma concentration profiles of both total scavenger and total hTNFα was investigated. pH-dependent hTNFα binding was expected to result in increased clearance of the cytokine and decreased clearance of the scavenger.

[0143] Selected tissues were collected from the mice, in order to investigate distribution of the scavenger and the target cytokine. Parental IgG were used as reference compound in this study.

[0144] The in vitro and in vivo data sets were used to establish correlations between: [0145] physico-chemical properties and in vivo pharmacokinetics [0146] FcRn affinity and in in vivo pharmacokinetics and tissue distribution

[0147] The correlations were used to build a physiologically-based pharmacokinetic (PBPK) model capable of characterizing and simulating plasma and tissue pharmacokinetics.