AN ANALYSIS OF THE SEQUENCES OF THE VARIABLE REGIONS OF BENCE JONES PROTEINS AND MYELOMA LIGHT CHAINS AND THEIR IMPLICATIONS FOR ANTI- BODY COMPLEMENTARITY* BY TAI TE WU, P~.D., ~u~D ELVIN A. KABAT, PI~.D (From the Departments of Microbiology, Neurology, and Human Genetics and Develop- ment, College of Physicians and Surgeons, Columbia University, and the Neuro- logical Institute, Presbyterian Hospital, New York 10032; the Biomathematics Division, Graduate School of Medical Sciences, Cornell University and the Sloan-Kettering Institute, New York 10021) (Received for publication 26 March 1970) The extraordinary versatility of the antibody-forming mechanism in producing an almost limitless number of specific receptor sites complementary for almost any molec- ular conformation of matter within a size range (1-3) represented by a hexa- or hepta- saccharide as an upper and a mono- or disaccharide as a lower limit, is almost certainly related to the unique structural features of immunoglobulins and differentiates them from all other known proteins. These antibody-combining sites are formed as a con- sequence of the interaction of two polypeptide chains, a light and a heavy chain (2, 4, 5). The antibodies usually formed to various antigens often represent heterogeneous populations of immunoglobulin molecules of different classes, subclasses, and genetic variants and also show specificities tgward different antigenic determinants (1, 2, 6, 7). In some instances, however, relatively homogeneous populations of antibodies with respect to many of these properties have been obtained. Among these have been human antibodies to dextran and levan (8, 9) and rabbit antibodies to the group-specific carbo- hydrate of streptococcus (10-12), antibodies to the Type III-specific capsular poly- saccharide of pneumococcus (13, 14), rabbit antihapten (15), and specimens of anti- bodies and of Fab' fragments which crystallized (Nisonoff et al., in references 16, 17), but sequence data on these are not yet available. The large body of sequence data related to immunoglobulin structure comes from the analysis of urinary Bence Jones proteins and from the monoclonal immunoglobu- lins found in large amounts in the sera of patients with multiple myeloma and Walden- strtim macroglobuUnemia (16, 18). While a substantial body of evidence was available relating these proteins to immunoglobulins, the recent demonstration that many myeloma globulins have specific ligand-binding properties like those of many anti- bodies provides increasing confidence that myeloma globulins represent homogeneous populations of antibody molecules (16, 18-27). The ability to produce in BALB/c * Aided by grants from the National Science Foundation (GB-8341) and the National Cancer Institute (CA-08748), and a general Research Support Grant of the U. S. Public Health Service. 211
`
`PETITIONER'S EXHIBITS
`
`Exhibit 1053 Page 1 of 40
`
`

`
`212 BENCE JONES PROTEINS AND MYELOMA LIGHT CI-IAINS mice myelomas and macroglobulinemias (28) which produce myeloma globulins and Bence Jones proteins like those in the human, provides a source of data from which important evolutionary trends can be inferred. Thus the extensive sequence data on Bence Jones proteins, which are considered to be light chains of myeloma globulins and Waldenstriim macroglobulins (29), and on various light and heavy chains, provide information clearly pertinent to the problem of the elucidation of the structure of antibody-combining sites. The unique finding that distinguishes the immunoglobulins from all other proteins is that the N-terminal half of the light chains and the N-terminal quarter of the heavy chains vary in sequence in samples obtained from individual monoclonal immuno- globulins and that indeed no two such variable regions of any chain and no two mye- loma immunoglobulins or Bence Jones proteins have thus far been found to be identical in sequence (30). The constant region, however, is essentially no different from other proteins in that the variation in the amino acids found at any position is ascribable to species and class variations or to genetic variants such as Inv factors. By com- parison of sequence data on the variable and constant regions of Bence Jones proteins with amino acid composition of purified human antibodies, it could be shown that most of the compositional variation could only originate in the variable region (see Kabat in reference 18). From sequence data, a variety of hypotheses have been advanced (7, 31-35) to explain the structural basis of antibody complementarity. All of these are selective theories, i.e. they consider that the information for complementarity is essentially built into the primary sequence of each chain and that a given antigen only triggers the biosynthesis of those antibody molecules having complementary receptor sites. There are two types of selective theories: germ line theories (36) and somatic mutation theories (37-39). At present no hypothesis is generally accepted. Excellent reviews (see above) are available. The present communication is an extension of earlier efforts from this labora- tory (18, p. 87, and 40-43) to locate more precisely those portions of the vari- able region which are directly responsible for antibody complementarity, that is which make direct contact with the antigenic determinant, and to explain the unique capacity of these proteins to have so many complementary regions. As in the earlier studies, all human K, human X, and mouse K Bence Jones protein and light chain sequences are aligned for maximum homology (44) and all variable regions are considered as a unit and compared with the con- stant regions. These earlier studies had called attention to the following: (a) The variable regions had few if any species-specific positions while the constant regions of the human and mouse proteins had 36 species-specific amino acid substitutions per 107 residues (40, 45). A species-specific position is defined as one at which the amino acid residues in the mouse proteins differ from those in the human proteins. (b) When the invariant residues of these two regions were compared, the latest tabulation (45) showed the variable regions to have 10 invariant and almost invariant glycines and no invariant alanines, leucines, valines, histi-
`
`PETITIONER'S EXHIBITS
`
`Exhibit 1053 Page 2 of 40
`
`

`
`TAI TE WU AND ELVIN A. KABAT 213 dines, lysines, or serines while the constant regions had 3 each of invariant alanine, leucine, and valine, and 2 invariant histidines, 2 invariant lysines, and 5 invariant serines. It was suggested that the invariant glycines were important in contributing to the flexibility needed by the variable region in accommo- dating the numerous substitutions (41, 43) at the variable positions. It was also suggested that the invariant glycines near the end of the variable region at positions 99 and 101, plus the almost invariant glycine at position 100, provided a pivot upon which the complementarity-determining regions might move to make better contact with the antigenic determinant (43; 18, p. 87) just as the walls of the lysozyme site have been shown to adjust somewhat to accommodate its hexasaccharide substrate (46). The hydrophobic residues in the constant region were hypothesized to be involved in noncovalent bonding to the heavy chain. (c) From an examination of sequences of the ~I, KII, and ~III subgroups (Hood et al. in reference 16) (47, 48) of the human Bence Jones proteins in which many of the proteins in a subgroup had an identical sequence for the first 2(~24 residues, it was postulated that there are two kinds of residues in the variable regions, those making direct contact with the antigenic determi- nant (complementarity determining) and those which are involved only in three-dimensional folding (42). The latter would be expected to have less stringent requirements, and more mutation noise would be permitted than with the complementarity-determining residues. This distinction led to the inspection of the sequences for short stretches showing very high variability and two of these were identified: the most variable beginning at residue 89 and ending at 97, the other running from residue 24 through 34. Each of these two unusually highly variable regions began after an invariant half-cystine and was followed by an invariant phenylalanine (residue 98) and an invariant tryptophane (residue 35) respectively. It is of interest that the two regions are brought close together by the S--S bond I2~-II88 (45). Milstein (47), Milstein and Pink (7), and Fran~k (49) have also called attention to the highly variable positions in these regions and Fran~k (49) has noted an additional highly variable region around residues 52-55. It was hypothesized (45) that these first two regions might represent the complementarity-determining regions and that complementarity might be acquired by the insertion of small linear se- quences into the light and heavy chains by some episomal or other insertion mechanism. It is striking that the differences in chain length seen in the Bence Jones proteins are confined to these two regions of the chain. The remaining portions of each chain would be essentially under the control of structural genes. The inserted sequences would be drawn from a large but finite set and either inserted under the influence of antigen, if antibody-forming cells are multipotent, or individual sequences might be distributed to immunoglobulin-
`
`PETITIONER'S EXHIBITS
`
`Exhibit 1053 Page 3 of 40
`
`

`
`214 BENCE JONES PROTEINS AND MYELOMA LIGHT CHAINS forming cells during differentiation if the capacity of individual cells to synthe- size antibody is restricted. This working hypothesis offers several advantages: (a) It is capable of providing the evolutionary stability and accounts for the universality of the antibody-forming mechanism throughout the verte- brates. Germ line theories (34-36) postulate one gene for each of the thousand or more variable regions (30). This would be expected to result in divergence during evolution since the loss by mutation of any one variable region would only minimally affect the capacity to form antibody and survival; thus indi- viduals and populations lacking certain variable regions would arise. (b) It offers a substantial simplification to the problem of producing a very large number of complementary sites. While it is known that in all proteins with specific receptors the site is formed by residues from widely separated portions of the chain, these sites are all formed by single chains. Thus, form- ing a three-dimensional site must involve residues from various regions. The antibody site being formed by a heavy and a light chain need not necessarily be so restricted. Since much additional data on the light chains and a number of heavy chain sequences have been accumulated, the present communication represents a further attempt at analyzing the unique features of the variable regions of immunoglobulin chains. Among aspects considered are the role of glycine, invariant residues, and hydrophobicity patterns, and the highly variable por- tions, with a view to localizing the regions responsible for complementarity and evaluating various theories in terms of evolutionary origin and perpetua- tion of the antibody-forming mechanism. Sequence Data Employed--Complete and partial sequence data have been published on 77 Bence Jones proteins and immunoglobulin light chains as well as on a number of heavy chains. Data were available on 24 human KI, 4 human KII, 17 human dII, l0 human M, 2 human MI, 6 human MII, 5 human MV, 2 human XV, 2 mouse KI, and 5 mouse KII proteins. 1 The original light chain sequence data may be found in the following references. HBJ 98: Baglioni, C. 1967. Biochem. Biophys. Res. Commun. 26:82. Eu: Cunningham, B. A., P. D. Gottlieb, W. H. Konigsberg, and G. M. Edelman. 1968. Biochemistry. 7:1983. Mil (human dI): Dreyer, W. J., W. R. Gray, and L. Hood. 1967. Cold Spring Harbor Syrup. Quant. Biol. 32:353. Hac, Dob, Pal: Grant, A., and L. Hood. Unpublished work. Roy, Cum: Hilschman, N., and L. C. Craig. 1965. Proc. Nat Acad. Scl. U. S. A. 53"1403; Hilschmann, N. 1967. Hoppe-Seyler's Z. Physiol. Chem. 348:1077; Hilschmann, N., H. U. Barnikol, M. Hess, B. Langer, H. Ponstingl, M. Stelnmetz-Kayne, L. Suter, and S. Watanabe. 1968. Fed. Eur. Biochem. Soc. Syrup., 5th. In press. 1 The World Health Organization has recently changed the notation of subgroups so that human ~II in this paper will become human ~[II and human rIII will become human KII.
`
`PETITIONER'S EXHIBITS
`
`Exhibit 1053 Page 4 of 40
`
`

`
`TA1 TE WU AND ELVIN A. KABAT 215 HS 78, HS 92, HS 94, HS 68, HS 70, HS 77, HS 86, HS 24: Hood, L., and D. Ein. 1968. Nature (London). 220:754. HBJ 7, HBJ 11, HBJ 2, HBJ 8: Hood, L., W. R. Gray, and W. J. Dreyer. 1965. J. Mol. Biol. 22:179. MBJ 41, MBJ 70, MBJ 6: Hood, L., W. R. Gray, and W. J. Dreyer. 1966. Proc. Nat'l Acad. Sci. U. S. A. 55:826. HBJ 10, HBJ 1, HBJ 4, HBJ 6, HBJ 5, HS 4, HBJ 12, HS 6, HBJ 15: Hood, L., W. R. Gray, B. G. Sanders, and W. J. Dreyer. 1967. Cold Spring Harbor Symp. Qnant. Biol. 32:133. Ste: Edman, P., and A. G. Cooper. 1968 Fed. Eur. Biochem Soc. Letters. 2:33; Hood, L., and D. W. Talmage. 1969. In Developmental Aspects of Antibody Formation and Structure. Prague. In press. Lay, Mar, Ioc, Wag, How, Koh: Kaplan, A. P. and H. Metzger. 1969. Biochemistry. 10: 3944. New, III, Mil (human XIV): Langer, B., M. Steinmetz-Kayne, and N. Hilschmann. 1968. Hoppe-Seyler's Z. Physiol. Chem. 349:945. BJ, Ker: Milstein, C. 1966. Biochem. J. 101:352. Rad, Fr4: Milstein, C. 1967. Nature (London) 216:330. X: Milstein, C. 1968. Biochem. J. 110:631. Bel, Man, B6: Milstein, C. 1968. Fed. Eur. Biochem. Soc. Symp. on "~-globulin, Prague. Day, MBJ46, Roy: Atlas of Protein Sequence and Structure, M. O. Dayhoff, Editor. 1969. Mz: Milstein, C., B. Frangione, and J. R. L. Pink. 1967. Cold Spring Harbor Symp. Quant. Biol. 32:31. Ale, Car, Dee: Milstein, C., C. P. Milstein, and A. Feinstein. 1969. Nature (London) 221:151. Cra, Pap, Lux, Mon, Con, Tra, Nig, Win, Gra, Cas, Smi: Niall, H., and P. Edman. 1967. Nature (London) 216:262. MOPC 149, AdjPC 9, MOPC 157: Perham, R., E. Appella, and M. Potter. 1966. Science (Washington) 154:391. Kern: Ponstingl, H., M. Hess, and N. Hilschmann. 1968. Hoppe-Seyler's Z. Physiol. Chem. 349:867. Tew: Putnam, F. W. 1969. Science (Washington). 163:633. Ag, Ha, Bo, Sh: Putnam, F. W., K. Titani, M. Wikler, and T. Shinoda. 1967. Cold Spring Harbor Symp. Quant. Biol. 32:9; Titani, K., T. Shinoda, and F. W. Putnam. 1969. J. Biol. Chem. 244:3550. TI: Suter, L., H. U. Barnikol, S. Watanabe, and N. Hilschmann. 1969. Hoppe-Seyler's Z. Physiol. Chem. 1150:275. The accumulation of such large numbers of sequences makes it possible to use statistical criteria in defining the types of residues. Thus in earlier studies, an invariant residue was rigidly defined, e.g., a position at which all samples showed the same amino acid residue sometimes allowing a single exception. The definition of an invariant residue used in this paper is taken as a position at which 88-90 % or more of the samples contain the same amino acid. This may allow potential functions to be recognized despite possible errors or uncertainties in sequence, or occasional substitutions compatible with function. A summary of the sequence data is provided in Table I which lists the amino acids found at any position in any subgroup of human K-, human X-, and mouse K-chains, the number of times each occurs, and the total number of sequences
`
`PETITIONER'S EXHIBITS
`
`Exhibit 1053 Page 5 of 40
`
`

`
`216 BENCE JONES PROTEINS AND MYELOMA LIGHT CHAINS TABLE I Amino Acids Found at each Position in the Variable Region of the Various Subgroups of Human ~-, Human X- and Mouse (cid:127)-Bence Jones Proteins No. of Protein Amino Human Kappa Human Immbda Mouse Kappa Position Total Sequences Acids III IIl III Ill IV V I II Studied 0 74 Glu i 0 i 0 O 0 O O O 0 0 --- 73 23 3 17 9 2 6 4 2 2 5 i 74 Lys i O 0 i 0 0 0 O 0 0 0 PCA 17 0 0 0 8 2 6 0 0 0 i Asp 29 18 4 i 0 0 O 0 0 2 4 Asx 5 5 O 0 0 O O O O 0 0 Glu 15 0 0 14 i 0 0 O O 0 0 Glx i O 0 i 0 O O O O 0 0 --- 6 O 0 0 0 0 0 4 2 0 0 2 73 lle 48 22 3 16 O 0 0 0 0 2 5 Tyr 4 00 0 O O 0 4 0 0 0 Val i I 0 0 0 0 0 0 O 0 0 Met i O 0 i O O 0 0 O O 0 Ser 19 O 0 O 9 2 6 0 2 O 0 3 72 lle i O 0 I 0 0 0 O 0 O O Pro i 0 0 0 0 0 i 0 0 0 0 Leu i i 0 0 0 0 0 0 0 0 0 Val 33 0 3 16 8 0 0 i 0 0 5 Ala 9 0 0 0 1 2 5 i 0 O 0 Asp i 0 0 0 0 0 0 i 0 0 0 Glu 2 0 O 0 O 0 0 i i O O Gln 21 19 0 0 0 0 O 0 0 2 0 Glx 3 3 O O O 0 O O O O 0 4 71 Leu 43 3 i lh 9 2 6 4 i 0 3 Val 2 0 O i 0 O 0 O O 0 i Met 26 20 2 2 0 O O O 0 2 0 5 7o Ala 2 0 00 O 0 2 O O 0 0 The 67 23 3 17 9 2 3 4 1 2 3 Ser i 0 0 O O 0 i 0 O 0 0 6 69 Gln 63 20 3 16 8 1 6 4 1 i 3 Glx 6 30 i i i O 0 0 O O 7 63 Pro 20 O 0 0 9 2 6 3 0 0 0 Thr i 0 i 0 0 0 0 0 0 0 0 Ser 41 22 2 14 0 O 0 0 O i 2 Asp i O O 0 O 0 0 0 i 0 0
`
`PETITIONER'S EXHIBITS
`
`Exhibit 1053 Page 6 of 40
`
`

`
`TAI TE WU AND ELVIN A. KABAT 217 TABLE I--Contlnued No. of Protein Amino Position Total Sequences Acids Studied 8 64 Pro 98 22 315 Ala 6 0 0 0 9 63 Leu 3 0 30 Ala 7 0 0 4 Thr i i 0 0 Ser 41 21 0 0 Gly i0 0 0 i0 Asx 1 0 0 1 lO 63 Phe 1 1 0 0 Th~ 17 3 014 Ser 25 18 4 0 --- 20 0 0 0 ii 63 Leu 43 22 414 Val 15 0 0 0 Ala 5 0 0 0 12 61 Pro 4 0 40 Ala i 0 0 0 Set 56 21013 13 61 Leu 12 0 012 Val ii 2 4 0 Met i 0 0 1 Ala 23 19 0 0 Oly 14 0 0 0 14 61 Ala 6 0 0 0 Thr 9 o 4 o Ser 46 21 013 19 61 Pro 36 0 413 Leu 4 i 0 0 Val 20 20 0 0 Asx i 0 0 0 16 61 krg i i 0 0 Gly 60 20 4 13 17 61 Asp 23 210 1 Glu 17 0 4 ii Gln 19 0 0 0 Glx 2 0 0 i Human Kappa Human Lambda Mouse Kappa I II IIl I II III IV V I II 920 31 12 OO6 00 0 0 00000 O0 00001 02 000 00 00 926 20 i o 00000 O O 00000 O0 00000 O0 00000 O0 00000 12 9 2 6 2 l 0 0 000 O0 12 6 06 2 1 0 0 32000 00 000 00 00 00000 01 92621 ii 00000 0 0 00021 02 00000 00 30000 i0 626 O0 O0 50001 O0 40000 0 i 02620 ii 826 2o o l 00001 ii 00000 O0 i00 O0 0 0 00000 O0 9 2 6 2 i i 2 00000 01 00100 i0 825 21 0 1 i00 O0 O0
`
`PETITIONER'S EXHIBITS
`
`Exhibit 1053 Page 7 of 40
`
`

`
`218 BENCE JONES PROTEINS AND MYELOMA LIGt~T CHAINS TABLE I--Continued No. of Protein Amino Human Kappa Human L~mbda Mouse Kappa position Sequences Acids Total III Iii III III IV V I II S t udie s 18 61 Pro 4 0 4 0 0 0 0 0 O Lys i 0 0 0 i O 0 0 0 Ala i 0 0 0 i 0 0 0 0 Arg 4O 21 O 13 400 O O Tbm ~ 0 0 0 i O i 2 1 8er 9 0 O O i 2 5 0 O Gly i 00 0 i O 0 O 0 19 53 Ile 7 l 0 0 0 0 6 0 0 Val 28 17 O 0 6 2 0 0 i Ala 18 O 4 i0 i O O 2 O 20 53 Ile i O O O I O O 0 O Val i 0 0 0 O 0 0 i 0 Ala 3 i 0 2 O 0 0 0 O Arg i O 0 0 0 0 O 0 i Thr 40 17 O 8 6 2 6 00 Ser 7 0 4 0 O 0 O I O 21 43 lle 30 14 4 0 ~ 2 0 3 1 Leu 12 i O 9 O O O O 0 VafL i 0 0 i 0 0 0 0 0 22 42 AI~ i l O O O O O O O Thr 19 i~ 0 0 0 0 0 3 1 Ser 22 O 4 9 5 2 0 0 O 23 30 Cys 30 9 3 4 5 2 0 3 i 24 26 Arg ii 1 3 4 O 0 0 O 0 Thr 2 O O O O 2 O O O 8er 6 0 0 0 & 0 O 2 0 Gly i 00 0 O 0 0 i 0 Gln 4 3 0 O O O O O i Glx 2 2 O 0 O O O 0 O 25 25 Ala 13 6 O 4 O O 0 O O 8er 2 0 2 0 0 0 0 0 O oly 3_o o o o 4 2 o 3 z 26 24 Thr 2 O O O O 2 0 O O Ser 16 6 2 4 i 0 0 0 O Gly 2 O O O 200 O O Asp 4 0 O O O 0 0 3 i 0 0 0 0 0 0 i i 0 0 0 i 0 0 0 0 I i 0 l O 0 0 0 0 0 0 0 0 i I i 0 i i i 0 0 O 0 1 0 0 2 i 2 i 2 0 0 0 0 0 0 0 0 0 0 i 2 0 0 O O 0 0 i 2 0 0 0 0
`
`PETITIONER'S EXHIBITS
`
`Exhibit 1053 Page 8 of 40
`
`

`
`TAI TE WU AND ELVIN A. KABAT 219 TABLE l--Continued No. of Position ~otein ~inOTotalH~nKapp a H~anL~bda MouseK~pa Sequences Acids III IIl III Ill IV V I II Studied 23 ~s i 000 00010 O0 Ser 5 000 31001 O0 Asn i 000 00010 O0 Glu i 000 00010 O0 Gln 13 5 i 4 0 0 0 0 0 1 2 G~ 2 i i 0 0 0 0 0 0 0 0 a 22 Ser 5 0 i 3 0 0 0 0 0 0 i Asn i 010 00000 O0 --- ~ 601 31021 ii b 22 ~u 2 020 000 00 00 Val i 000 00000 Of --- ~ 604 31021 ii c 22 Leu 2 020 000 00 0 0 --- 20 604 31021 12 d 22 T~ i 000 i0000 00 Ser 3 000 21000 00 Asp i 010 00000 O0 --- ~ 614 000 21 i 2 e 22 Asp 2 010 010 00 0 0 Asn 3 000 30000 00 Asx i O00 00000 Of --- ~ 614 000 21 1 1 f 22 Val 2 001 010 O0 0 0 Set 2 O10 00000 Of G~ i 000 i0000 O0 G~ i o i o o o o o o o o --- 16 603 20021 ii 28 22 Ile i 000 i 00 00 0 0 ~u 4 001 00021 O0 Val 2 002 00000 O0 Met i 000 i0000 O0 i 000 10000 O0 Set 3 iii 00000 00 Gly 2 010 010 00 0 0 His i 000 00000 01 Asp 4 400 00000 00 Asx 3 i00 00000 ii
`
`PETITIONER'S EXHIBITS
`
`Exhibit 1053 Page 9 of 40
`
`

`
`220 BENCE JONES PROTEINS AND ]~{YELOMA LIGHT CHAINS TABLE I--Continued Position 3o 31 32 33 34 No. of Protein Amino Total Human Kappa Human Lambda Sequences Acids III IIl I II Ill IV V Studied 21 lie 8 600 00000 ~g 2 001 000 Ol Ser 3 003 00000 G~ 4 000 20010 G~ i 000 00010 Asp i 010 00000 Asx 2 010 010 00 21 lle 2 i00 00000 ~s 2 i00 00010 Set 2 i00 00000 G~ 5 021 00001 Asp i 000 00010 Asn 8 303 20000 Asx I 000 010 O0 20 ~ i 000 00001 Phe i 000 00010 ~s 2 i00 01000 T~ 2 ii0 00000 Ser 7 102 20000 His i i00 00000 Ash 4 102 00000 Asx 2 ii0 00000 20 Trp i i00 000 O0 I0 223 21000 Phe 5 301 00000 Leu 1 000 00000 T~ 1 000 00010 Asp i 000 00001 Ash i 000 00000 Leu Ii 424 00000 Val 4 000 210 i0 Met i 000 00000 Ala i 000 00001 Ser i 000 00000 ~ i 000 i0000 Ala 6 i0~ 00001 Ser 3 000 ii0 i0 Asp i 010 00000 Ash 4 210 000 O0 Asx 3 i00 00000 Mouse Kappa I II i i 0 0 0 0 0 i 0 0 0 0 0 0 0 1 0 0 0 1 i 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 i i 0 0 0 i 0 0 0 0 0 0 0 i i 0 0 0 0 0 0 i 0 i 0 0 0 i 0 0 i 0 0 0 0 0 0 0 0 0 0 i i l
`
`PETITIONER'S EXHIBITS
`
`Exhibit 1053 Page 10 of 40
`
`

`
`TAI TE WU AND ELVIN A. KABAT 221 TABLE 1--Continued No. of Protein Amino Human Kappa Position Total Sequences Acids III III Studied 35 17 Trp 17 42 4 36 17 Tyr 13 424 Phe 2 0 0 0 Leu i 0 O O His i O O O 37 16 Leu 2 020 Gln 12 3 0 4 Glx 2 0 O 0 38 16 His I 000 Gin 12 3 i 4 Glx 3 0 i O 39 16 Leu 2 0 0 0 Lys 9 223 Arg ~ 0 0 i Gly l 1 0 0 His i 0 0 0 Asx I 0 0 0 40 16 Pro 15 3 i 4 Ala i 0 1 0 41 16 I~s i i 0 0 Gly 15 2 2 4 42 16 Lys 3 30 0 /krg i 0 0 0 Thr 2 00 0 Gin 7 0 1 4 Glx 3 0 10 43 16 Pro i 00 0 Ala ii 3 0 4 Thr i O O 0 Ser 2 0 i O Gln i 0 i O 41~ 16 lle i 0 0 0 Pro 15 3 2 4 45 16 Leu 2 0 0 0 Lys 9 3 0 i A~g 3 0 0 3 Ser i 0 i 0 Glx 1 0 1 0 Human Lambda Mouse Kappa I II iII IV V I II 210 ii 1 1 ii001 O0 OOOlO O1 00000 i0 i 0 0 0 0 0 0 O 0 0 O O 0 0 2 i 0 i i 0 O O0000 ii i0000 0 0 ii0 ii O0 00000 ii 200 O0 0 0 000 Ol 0 i OOOlO O O 00000 O0 010 O0 O O 00000 i0 210 ii i i 00000 O0 000 O0 00 210 ii ii 00000 O0 O10 O0 O0 20000 O O O0011 O O 00000 ii 000 O0 0 i 21001 O O O0000 i0 00010 O0 00OO0 O O O00OO i0 210 ii 0 i 00011 O O 21000 i i 00000 O O 00000 O0 000 O0 0 0
`
`PETITIONER'S EXHIBITS
`
`Exhibit 1053 Page 11 of 40
`
`

`
`222 BENCE JONES PROTEINS AND iVfYELOMA LIGHT CHAINS TABLE I---Continual Position ~6 47 48 49 9o 51 52 53 54 No. of P~'otein Amino Total Sequences Acids Studied 14 Ile i Leu 12 Arg i 14 Izau ii Val 3 14 lle 12 Met 2 14 Tyr 13 Phe I 14 Leu i Val i Lys i Ala 2 Arg i Thr i Set i ol~ 3 Asp 2 Glu 1 14 Leu 1 Val 2 Ala 5 Arg i Thr i Gly i Hi s 1 Asp 2 14 Ser i0 Asp 1 Asn 2 Glu i ~s 3 Set 4 Asn 4 Glu i Glx 1 lb. Leu 4 irg 9 Gln 1 Human Kappa Human Lambda Mouse Kappa I II IIl III III IV V I II i00 00000 00 222 21011 01 000 00000 i0 322 20000 ii 000 01011 00 221 210 ii i i I01 000 00 0 0 322 20011 ii 000 01000 O0 010 000 O0 0 0 001 00000 00 i00 00000 0 0 000 00000 i I 000 lO000 O0 010 00000 O0 000 00010 O0 001 01001 O0 200 00000 00 000 100 00 0 0 010 000 O0 0 0 001 01000 O0 301 00000 Of 000 00001 O0 000 00000 i0 OlO 000 O0 O0 000 00010 O0 000 20000 O0 322 01000 ii 000 i0000 O0 000 i0001 O0 000 000 i0 0 0 010 00000 O0 ZOO 20000 O0 102 00000 i0 llO O0001 O1 000 01000 O0 000 000 i0 O0 300 00000 l O 022 21011 O0 000 000 O0 O 1
`
`PETITIONER'S EXHIBITS
`
`Exhibit 1053 Page 12 of 40
`
`

`
`TAI TE WU AND ELVIN A. KABAT 223 TABLE I--Continued No. of Protein Amino Position Total Sequences Acids Studied 5~ 16 Pro 5 Ala 6 Gly i Asx i GIu 3 56 16 Ala i Thr 5 Ser i0 57 ~6 Thr l GIF 15 58 16 lle 6 Val 9 Thr 1 59 16 Pro 16 60 16 Val 1 Lys i Ala 1 Ser 3 Asp 8 Asn i GIu i 61 16 Arg 16 62 16 lle i Phe 15 63 16 lle I Ser 15 64 16 Ala i Oly 15 65 16 Thr i Set 15 66 16 Lys 3 Arg i 8er 2 Gly i0 67 16 Phe i Set 15 Human Kappa Human Lambda Mouse Kappa I II Ill I II llI IV V I II 000 21011 O0 024 00000 O0 000 00000 O l 000 00000 i0 3OO 00000 O0 i00 00000 O0 104 00000 O0 120 210 ii i i 000 00010 O0 324 21001 ii 003 i00 ii 0 0 320 ii000 ii 001 00000 0 0 24 210 I 1 i l OO1 00000 O0 000 00000 i0 000 00000 O1 300 00000 O0 013 21001 O0 010 00000 O0 000 00010 O0 324 21011 Ii 000 i0000 00 324 ii011 ii i00 00000 O0 224 21011 l l 000 i00 O0 O0 324 ii0 ii i i i00 000 00 0 0 224 21011 l l 000 21000 0 0 000 00000 i0 000 O0011 O0 324 00000 Of lO0 00000 O0 224 21011 l l
`
`PETITIONER'S EXHIBITS
`
`Exhibit 1053 Page 13 of 40
`
`

`
`224 BENCE ~ONES PROTEINS AND MYELOMA LIGHT CHAINS TABLE I--Continue~ NO. of Protein Amino Human Kappa Position Total Sequences Acids III IIl Studied 68 16 Gly 15 324 Asn I O O 0 69 16 Ala 2 001 Thr ii 323 Sel i 0 0 0 His i 0 0 0 Asp i 0 0 0 70 16 Thr 3 0 0 0 Set 2 0 0 0 Asp 9 2 i 4 Asx i O i O GIu i i O 0 71 16 Tyr i O O 0 Phe i0 3 2 4 Ala 5 O O O 72 16 Thr Ii 3 2 Ser 5 O 0 0 73 16 Phe 2 200 Leu 14 i 2 )4 74 16 Lys i O i 0 Ala i 0 0 0 Thr 12 31 Oly . i 000 Asn 1 0 0 0 75 16 Ile 15 32 4 Val 1 0 0 0 76 16 Thr 2 000 Ser 13 3 2 4 His 1 0 0 0 77 17 Pro i 0 0 0 Arg 6 O 2 )4 Ser )4 3 0 0 Gly 6 i O O 78 17 Leu 13 )4 1 /~ Val i 0 i 0 Met i O 0 0 Ala 2 0 0 O Human Lambda Mouse Kappa ! II III IV V I II 20011 ii OlO00 O0 00010 O0 20000 Of 00000 i0 00001 O0 O1000 O0 010 ii O0 20000 O O 00OOO ii OOO00 O0 000 O0 0 O O0000 i0 000 O0 0 i 21011 O0 lOOlO O0 llO O1 l l OO0 OO O0 210 ii i i 00000 O0 i0000 O0 01011 i0 i0000 O0 00000 O l 20011 ii 01OO0 O0 i00 Ol O0 ii0 i0 1 0 00000 Of 00000 O l O0000 O0 00000 i0 21011 O0 21000 i0 000 O0 0 0 0OOO0 O i 000 ii O O
`
`PETITIONER'S EXHIBITS
`
`Exhibit 1053 Page 14 of 40
`
`

`
`TAI TF_, WU AND ELVIN A. KABAT 225 TABLE I--Continued No. of Protein Amino Human Kappa Human Lambda Mouse Kappa Position Total Sequences Acids III III III III IV V I II Studied 79 17 Arg 3 0 O 0 2 i 0 0 0 Glu 5 0 i 3 0 O 0 0 0 Gln 6 4 0 0 0 0 0 i i Glx 3 0 1 1 0 0 O 0 0 80 i 7 Pro 9 15 i 4 0 0 0 0 0 Ala 3 0 i 0 0 i 0 0 i Thr i 0 0 0 i 0 0 0 0 Set 3 0 0 0 I 0 0 i 0 Glx 1 0 0 0 0 0 0 0 0 81 17 V&l 1 0 0 0 0 0 0 1 0 Ala 1 1 0 0 0 0 0 0 0 Gly 1 0 0 0 1 0 0 0 0 Asp 1 1 0 0 0 0 0 0 0 Asx i 0 0 0 0 0 0 0 0 G!u i0 2 i 4 I 0 0 0 i Glx 2 0 1 0 0 1 0 0 0 82 17 Asp 14 4 l 4 2 0 0 l l Asx 3 0 i 0 0 i 0 0 0 83 18 Ile 2 2 O 0 0 0 0 0 0 Phe 8 2 i 4 0 0 0 0 0 Val i 0 i 0 0 0 0 0 0 Thr i 0 0 0 0 0 0 0 0 Glu 5 0 0 0 2 0 0 2 i Glx I 0 0 0 0 i 0 0 0 84 18 Val 1 0 0 0 0 0 0 0 0 Aim 16 4 I 4 2 i 0 2 i Gly i 0 1 0 0 0 0 0 0 85 18 Val 6 0 2 4 0 0 0 0 0 Met i 0 0 0 0 0 0 0 0 Thr 4 4 0 0 0 0 0 0 0 His i 0 O 0 i 0 0 0 0 Asp 5 0 0 0 i 0 0 2 i Asx i 0 0 0 0 I 0 0 0 86 20 Tyr 20 6 2 4 2 i 0 2 i 87 20 Tyr 16 6 2 3 2 1 0 i l Phe 3 0 0 1 0 0 0 i 0 His i 0 0 0 i 0 0 0 0 88 21 Cys 21 6 2 4 3 i 0 2 i 0 0 i 0 0 0 0 3. 0 0 0 0 0 0 i 0 0 i 0 0 0 0 0 0 0 0 0 i 3_ 0 0 0 i 0 0 i 0 0 i 0 0 0 0 1 0 0 0 0 i 0 0 i 0 0 0 0 0 i 0 0 0 O i 0 0 0 i i 0 0 0 i 0 0 i 1
`
`PETITIONER'S EXHIBITS
`
`Exhibit 1053 Page 15 of 40
`
`

`
`226 BENCE JONES PROTEINS AND MYELOMA LIGHT CHAINS TABLE I--Contlnued position 89 9o 91 92 93 94 No. of Protein Amino Total Human Kappa Sequences Acids III llI Studied Human Lambda Mouse Kappa I II Ill IV V I II 22 Leu I 000 00000 i0 Met i 010 00000 00 Ala 2 000 20000 00 8er 1 000 01000 00 Asn I 000 00001 00 Gin ~ 624 10020 O0 Glx i O00 00000 01 22 Met i 010 OOOO0 O0 Ala 3 000 200 lO O0 T~ 2 000 i0010 O0 Set 2 000 01001 00 Gin 13 622 00000 iO Glx 1 000 00000 01 22 Trp 5 000 30020 O0 ~r 12 51~ 0"i000 i0 Phe i i00 00000 00 Ala 1 010 00000 00 Arg 2 010 00001 O0 Set I 000 00000 O l 21 Leu 2 020 00000 O0 Val i 000 O10 O0 00 ~s i 000 00000 Of Ala i 000 00000 i0 G~ 3 00 3 O 00 O O O 0 Asp 9 400 20021 O0 ASh i i00 OO000 O O GIu 3 Iii 00000 00 21 ~ 1 000 i0000 00 T~ 4 111 00010 00 Ser 7 102 i0011 i0 Gly i 001 00000 O0 His i L00 00000 00 Asp i i00 00000 00 Ash 2 200 00000 O O Asx i 000 O1000 O0 GIu 2 010 00000 01 Gin i 010 00000 O0 21 Ile 2 010 00010 O0 Leu 5 500 00000 00 Val i 000 00000 Of Met i OOO O0010 00 Ala i 010 00000 O0 ~g i 000 i0000 O0 Ser 8 014 i0001 iO Asp i iO0 00OOO O0 Asx i O00 01OO0 O0
`
`PETITIONER'S EXHIBITS
`
`Exhibit 1053 Page 16 of 40
`
`

`
`TAI TE WU AND ELVIN A. KABAT 227 TABLE I--Continued NO. Of Protein Amino Human Kappa Position Total Sequences Acids III III Studied 95 21 Pro 14 5 3 4 Leu 2 O 0 O Thr i 0 0 0 Ser 2 i 0 0 Gly I 0 00 Asx i 0 O O 96 21 Trp 2 0 O O Ile i O i O Tyr 2 i 10 Phe 2 i 0 i Pro I i 0 0 Leu 2 i i O Lys 2 i 00 Arg i i 0 O Thr i 0 0 1 Set 2 0 O i Ash i O 0 0 Asx I 0 0 0 Gin i O 0 i --- 2 0 0 0 97 20 Phe l O 0 0 Pro i 0 0 i Met 1 1 0 0 Ala 2 0 0 O Thr 12 4 3 3 His i 0 0 0 --- 2 O 0 O a 19 Val 4 O 0 O Ala i 0 0 O --- 14 5 3 4 b 20 lle i O O 0 Leu i O O 0 Val 4 O 0 0 --- i4 5 3 4. 98 20 Phe 20 5 3 4 99 20 Gly 20 5 3 4 i00 20 Pro i i O O Gly ii 2 10 Gln 8 2 2 4 iO1 19 Gly 19 4 3 4 Human L~mbda Mouse Kappa III llI IV V I II OOO00 ii 20000 O O 000 iO 0 O O00 iO 0 0 00001 O O O10 O0 0 O 00000 i i O0000 O0 00000 O0 0OO00 O0 0OOOO OO O0000 O O 00001 O0 00000 O O 00000 O0 i0000 O O iOOO0 O0 010 OO 0 0 000OO O0 00020 0 0 01000 00 00000 00 00000 00 20000 00 00000 l l 00001 00 00020 O0 20011 00 000 l0 0 0 000 00 i i 000 lO 0 0 00001 00 210 l0 0 0 00000 l l 2 1 0 2 1 1 1 2 i 0 2 1 i i 00000 O0 21021 i i 00000 O0 2 i 0 2 i 1 1
`
`PETITIONER'S EXHIBITS
`
`Exhibit 1053 Page 17 of 40
`
`

`
`228 BENCE JONES PROTEINS AND MYELOMA LIGHT CHAINS TABLE I--Concluded No ° Of Protein Amino Htunan Kappa Human Lambda Mouse Kappa Position Total Sequences Acids III III III III IV V I II Studied 102 19 Thr 18 h 3 3 2 1 0 2 i i i Ser i 0 0 i 0 0 0 0 0 0 0 103 19 l~ys 14 4 i 3 i i 0 i i i i Arg 3 0 I i 0 0 0 i 0 0 0 Asn i 0 i 0 0 0 0 0 0 0 O Gln 1 0 0 0 i 0 O 0 0 0 0 104 P_2 Leu 15 3 2 3 i i 0 2 i I i Val 7 3 2 i i 0 0 0 0 0 0 105 22 Thr 6 0 0 0 2 i 0 2 i 0 0 Asp 4 3 0 i 0 0 0 0 0 0 0 Glu 12 3 4 3 0 0 0 0 0 i i 106 22 lle 12 3 4 3 0 0 0 0 0 i i Phe i i 0 0 0 0 0 0 0 0 0 Leu 2 i 0 i 0 0 0 0 0 0 0 Val 7 i 0 0 2 i 0 2 i 0 0 a 22 Leu 6 0 0 0 2 i 0 2 i 0 0 --- 16 6 4 4 o o o o o l l 107 22 Lys 15 6 3 4 0 0 0 0 0 i i Arg 3 0 i 0 i i 0 0 0 0 0 Set 2 0 0 0 0 0 0 2 0 0 0 Gly 2 0 0 0 i 0 0 0 i 0 0 studied at the given position. Only data for which the sequence has been clearly assigned by the various authors have been included. The Role of Glydne--It has been suggested that glycine plays a unique role in the structure of the variable region of immunoglobulin light chains (18, p. 87; 41-43, 45). Jukes (50) and Welscher (51) have generally agreed with this. A further careful analysis becomes essential for the understanding of the function of the glycines in the over-all structure and in relation to antibody- combining sites. The basic property that differentiates glycine from all other amino acids structurally is the absence of a side chain. As a result, glycine can have many sterically allowable configurations. This has been verified experimentally in the case of lysozyme (46, 52) and tosyl-o~-chymotrypsin (53). The two angles, and ~ (54), which specify the conformation of the backbone of an amino acid have been calculated for each of the amino acids from the known tertiary structures of lysozyme (46, 52) and of tosyl-o~-chymotrypsin (53). A typical plot of the permissible angles of the glycine as compared with the alanine residues is shown in Fig. 1. The allowable configurations of alanine are mostly
`
`PETITIONER'S EXHIBITS
`
`Exhibit 1053 Page 18 of 40
`
`

`
`TAI TE WU AND ELGIN A. KABAT 229 I I x x o x x × o o 8 × × 8 3 o o xX o × x o co ! I o o o o o x × × "-'b- x~ x o I I I oo r~ g~ 0 ;> v 0 e~ g=
`
`PETITIONER'S EXHIBITS
`
`Exhibit 1053 Page 19 of 40
`
`

`
`230 BENCE JONES PROTEINS AND MYELOMA LIGHT CHAINS clustered near the a-helical region of lysozyme (Fig. 1 a and reference 55), while those of glycine are widely distributed. Comparison with similar maps for other amino acids in lysozyme also shows them to be more restricted. The data for tosyl-a-chymotrypsin also show that glycine may have many more con- formations (Fig. 1 b). This unique property of glycine thus may permit relative motion of the chains attached to the two ends of the molecule. With immuno- TABLE II Frequencies o/Glycine Residues at Various Positions in the Variable Region of Light Chains Position Human Kappa Human Lambda Mouse Kappa Total I II III I II III IV V I II 9 10/15 10/63 16 13 6/9 2/2 6/6 14/61 23 16 20/21 4/4 13/13 9/9 2/2 6/6 2/2 3-/1 1/1 2/2 60/61 99 24 i/3 1/26 4 25 4/4 2/2 3/3 i/I 10/25 40 26 2/5 2/24 8 27f i/3 1/22 28 i/2 i/1 2/22 9 29 2/2 1/2 1/2 ~/21 19 3o 2/2 1/~ 1/1 1/1 5/21 24 39 V3 V16 6 41 ~3 2/2 ,+/4 2/2 1/J. 1/1 1/~. 1/1 1/1 15/16 94 ~o 1/2 3./1 ]/1 3/J-~ 21 51 1/2 1/14 7 ~5 1/1 1/1