Eur. J. Immunol. 1991. 21: 1725-1731
Marc Monestier Garden State Cancer Center and Center for Molecular Medicine and Immunology, Newark
Anti-histone V genes
1725
Variable region genes of anti-histone autoantibodies from a MRL/Mp-@d&rmouse* The variable region nucleotide sequences of seven (five IgM and two IgG) anti-histone monoclonal antibodies from a single MFWMp-Zprllpr mouse have been determined. These antibodies are not clonally related and used diverseV, D and J genes. However, six of the seven antibodies have VH segments encoded by genes from the 3558 family, two of these (an IgM and an IgG) share an identical VHgene.The isoelectric points of MRA3 and MRA12, the two IgG antibodies of the panel, range from 6.3 to 7.0 and from 6.0 to 6.3, respectively. The second conplementarity-determiningregion (CDR) of theVH gene of MRA12 (the most acidic and the most strongly histone-reactive antibody) includes only two positively charged but five negatively charged amino acid residues. This feature is unusual since the equivalent CDR in most v ~ J 5 5 8genes are not comprised predominantly of acidic residues and suggests that such negatively charged residues are important for antibody binding to histones.
1 Introduction Histones are major targets of self-reactive antibodies in several autoimmune syndromes. For instance, in humans, anti-histone antibodies are highly prevalent in spontaneous and drug-induced lupus, juvenile arthritis, primary biliary cirrhosis and other autoimmune diseases [l-41. Antihistone antibodies are also detected in the sera of mice prone to develop spontaneous autoimmune syndromes such as MRLMp-lpdlpr, NZB and (NZB x NZW)F1 15, 61. As with other nuclear autoantigens, understanding the mechanisms leading to the production of anti-histone antibodies is an essential question concerning the pathogenesis of autoimmune diseases. The identification of the sequences of the V region genes of antinuclear antibodies such as anti-DNA antibodies has brought important information on the molecular basis of the DNA-antibody interactions and on the mechanisms leading to the expansion of autoreactive clones [7]. To explore these autoimmune phenomena further, we recently obtained a panel of seven anti-histone mAb from a single autoimmune MRLMp-ZpdZpr mouse [8]. In this communication, the V region gene sequences of these anti-histone antibodies are reported.
2 Materials and methods 2.1 mAb
(IgG2,,x).The hybridomas were obtained from the somatic fusion of the splenocytes of an MRLIMp-Zprllprmousewith the SP210 myeloma. The five mAb of the IgM isotype bind to histones H1 and/or H3, while the two IgGZa,xantibodies are specific for histone H1. The fine specificities of these mAb have been previously described [S].
2.2 mRNA sequencing Poly(A)+ RNA was isolated from hybridoma cells using oligo(dT) cellulose according to a previously described method [9]. The VH regions were sequenced directly from the poly(A)+ RNA template by the dideoxy chain termination method using isotype-specific oligonucleotide primers. The J, gene used by these antibodies were identified by synthesizing cDNA from poly(A)+ RNA using four different primers specificfor each of the four functional mouse J, gene segments. Since none of the antibodies was found to use J,2, their L chainV regions were also sequenced by the dideoxy termination method using the appropriate J,specific primer (J,2 is used in the non-functional transcript originating from the SP210 fusion partner; [lo]). The sequences of the oligonucleotide primers and the sequencing protocol were as previously reported [7, 11, 121. The nuclei acid sequences were compared with Ig sequences in the GenBank data library [13].TheVregion sequences were assigned to known gene families by comparison to previously published compilations [14-16].
The following antibodies were utilized in this study, MRAl (IgM,x), MRA3 (IgG2a,x), MRA5 (IgM,x), h~fRA7 2.3 IEF (IgM,x), MRAlO (IgM,x) M R A l l (IgM,x), and MRA12 This was carried out at 8°C in a flatbed apparatus (LKB Multiphor 11, Bromma, Sweden). The anolyte was 0.5 M [I 93771 acetic acid and the catholyte was 1M sodium hydroxide. Approximatively 10 pg of each antibody was applied onto * This work was supportedin part by NIHgrant A1 26665 to M. M. an Isogel agarose gel, pH 3-10 (FMC,Rockland, ME). and grants CA 06927, GM 20964, CA 09035 and an appropria- Focusing was performed at a constant power of 1W for the tion from the Commonwealthof Pennsylvaniato the Fox-Chase first 12 min and 25 W-1000 V for the remainder of the run. Cancer Center. Focusing was considered complete when the current had stopped decreasing significantly(less than 1 mA in 10 min). Correspondence: Marc Monestier, Center for Molecular Medicine The gel was then fixed and stained with Coomassie blue as and Immunology, One Bruce Street, Newark, NJ 07103, USA recommended by the manufacturer. A calibration curve was obtained by measuring the distance from the cathode Abbreviation: CDR: Complementarity-determiningregion 0 VCH Verlagsgesellschaft mbH, D-6940 Weinheim, 1991
+
0014-2980/91/0707-1725$3S O .25/0
1726
Eur. J. Immunol. 1991.21: 1725-1731
M. Monestier
of Isogel PI markers (FMC) and the migration distance of the antibodies was plotted against this curve.
gene segment. TheV gene use for the L chain is even more diverse than for the H chain: all seven light chains are encoded by separate V genes belonging to seven different V,families (Table 1).
3 Results 3.2 Gene segment usage in anti-histone antibodies
3.1 V gene usage in anti-histone antibodies
The nucleotide sequences and the deduced protein A great variety of gene segments are also represented in the sequences of the H and L chains of the seven anti-histone panel of anti-histone antibodies. Each of the four JH gene mAb are reported in Figs. 1 to 4. All the anti-histone segments is used at least once among the seven anti-histone antibodies of this panel are encoded by VHgenes belonging antibodies (Table 1). The sequences of the seven JH segto the 5558 family,with the exception of MRA5, encoded by ments are identical to the germ-line genes, with the a member of the S107 family (Table 1). Only two antibod- exception of MRA3, in which a nucleotide substitution ies, MRA3 and MRA11, are encoded by an identical VH results in the replacement of a tyrosine by a cysteine residue (Fig. 1). MRA3 and MRA11, which are encoded by an identical VHgene segment, use different D and JH segments indicating that these two antibodies are not clonally Table 1. Gene segment usage in anti-histone antibodiesa) related.The D segment usage is also diverse since segments from all three families are used including two indirect D-D joinings (Table 1). The L chain segments of these antiAntibody Isotype VH D JH vx Jx histone antibodies were equally heterogeneous since their I&¶ J558 Q52+SPZ-C J"4 VxS Jx5 J, segments originated from three of the four functional J, MRAl IgGz, 5558 J H ~ vx21 Jx4 MRA3 Q52 genes (Table 1). J"2 Vxl IgM S107 MRAS Jxl IgM 5558 SP2+SP2-c J H ~ Vx23 Jx4 MRA7 J H ~vxlo Jx4 IgM 5558 MRAlO 3.3 IEF IgM 5558 FL16 J H ~ Vx9 Jx4 MRAll SP2 J H ~ vx2 Jx4 MRA12 IgG2;, 5558 Previous studies have indicated that IgG autoantibodies to DNA, particularly if they are nephritogenic, are often a) The VH sequences of MRAS and MRAlO did not allow cationic molecules [17-191. The assess whether the electric assignment to a particular D gene. charges of the antibody molecules played a role in the FRI 1
10
20
30
60
40
NRAlH WASH
MRA7H YRAl0H YRAllH MRA12H MUASH
CDRI
___
70
80
98
340
360
I
I
360
--- __-_--
ACC TCA QTC ACC GTC TCC TCA A-T C-- --A
----- --- --- --- --- --_-- --- --- --- --- --- --A-0 --- ___ -__--- ----- --- --- --- --- --- -----
___
A-T C--
--A
--_ --- ---
I
FRII
100
110
CDRII
120
130
140
160
Figurel. Nucleotide sequences of the portions of mRNA encoding VH regions of anti-histone autoantibodies. Sequence identities are indicated by horizontal bars. N signifies undetermined bases. The start of the framework (FR) and CDR regions are offset from preceding sequences by a space and designated by abbreviations above the first codon in each region. Spaces have been introduced to maximize alignement. MRAS antibody, which is encoded by aVH gene belonging to the S107 family, is listed last.
Eur. J. Immunol. 1991. 21: 1725-1731
Anti-histone V genes
1727
MRAl MRA3
MRA7 MRAlO MRAl 1
MRAl2 MRA5
MRAl
Y
S
F
S
T
A
30 T G
Y
CDRl N M
H
W
V
K
Q
40 S
S
L
T
S
E
D
5
H
G
K
S
L
E
W
I
90 S A
V
Y
Y
C
A
R
G
0
Y
I
D
K
L
MRA3
MRA7 MRAlO
MRAll MRA12 MRA5
MRAl MRA3
MRA7 MRAlO MRAll MRAl2 MRA5
MRAl MRA3 MRA7 MRAlO
Y
80 M Q
L
N
MRAll MRAl2 MRA5
binding to histones, which are cationic proteins, the isoelectric points of the two IgG antibodies of the panel were determined (Fig.5). The PI of M U 3 ranges between 6.3 and 7.0 while IvlRA12 has a more restricted PI spectrum, 6.0-6.3. The PI value of MRA12, the most strongly H1-reactive antibody [S], is at the acidic end of the usual range of murine IgG antibodies [20] and it contrasts with the alkaline isoelectric points often observed with IgG anti-DNA antibodies [17-19]. Differences in isoelectric points among Ig result partially from differences in the amino acid contents of their V regions. Examination of the sequence of MRA12 reveals that the second complementarity-determining region (CDR) of the H chain is characterized by the presence of 5 acidic residues (2 glutamic acid and 3 aspartic acid residues) out of a total of 17 residues.To assess the potential significance of electric charges, we examined, in the second CDR, the frequency of the amino acid residues whose side chains are ionizable and can contribute to the overall charge of the molecule [21]. Table 2 shows the frequency of the residues with side chains having alkaline pK, values (arigine, lysine and tyrosine) or acidic pK, values (glutamic and aspartic acids) in the second CDR of the H chains of -12 or of a compilation of 44 V ~ J 5 5 8sequences [15]. The other two amino acid residues
with ionizable side chains were not included in this table since cysteine is infrequently found in this region and histidine (also rarely found) has an almost neutral (6.5-7.0) side chain pK, [21]. This Table shows that the second CDR of V H M R Ahas ~ ~more glutamic and aspartic acid residues Table2. Amino acid residues with ionizable side chains in the and of a compilation of second CDR of the H chains of -12 V~J558antibodiesa)
MRAl2 V~J558Compilation Number of Frequency Number of Frequency residues residues Glutamic acid Aspartic acid
Qmine Arginine Lysine Other residues
2 3 1 0 2 10
0.117 0.176 0.058
0 0.117 0.588
29 36
71 17 91
m
0.038 0.048 0.095 0.022
0.121 0.672
a) The compilation of 44 V ~ J 5 5 8sequences is from Dildrop 1151.
1728
M. Monestier
MRAl MRA3
s s
MRA5 M 7 MRAlO MRAll
MRA12
Eur. J. Immunol. 1991. 21: 1725-1731
CDRl
Q Q Q
V - V E Y Y - L V H S N S D D - L L D S D
T N K S -
I -
E Q
Q
S G G I I I G
30 40 S S Y F H W Y Q Q K P G T - L M Q - - - - - - -
N N K
T D N T
-
L L L L L
N S N
-
A I L V I M - L
A G G G G
T M V V I V
Y -
Y F F -
F L
L L
-
R
S -
S S P K L W Q P - - - L - Q - P - - L H E - - R - L D G T V - - L - K - - - T L - Q - - - R L
80
MRAl MRA3 MRA5 MRA7
MRAlO MRA11 MRA12
S H S N -
- - K
-
-
S P R N
V L - L R -
E -
A E P P Y
E D -
- - -
D -
-
C Q - S Q S T - - W Q G T
Figure 4. V, region amino acid sequences deduced from the nucleotide sequences of Fig. 3.
Eur. J. Immunol. 1991. 21: 1725-1733
Anti-histone V genes
1729
MRA3 and MRAl1 could suggest that certain members of the 5558 family are preferentially used among anti-histone antibodies.
Figure 5. IEF of mAb MRA3 and MRA12.
and less residues with alkaline pKa values than other antibodies of the same VH family (x2=9.43, 0.001 < p < 0.001). A comparison of the MRA12 VHCDR2 sequence with another larger v ~ J 5 5 8sequence compilation [14] resulted in similar conclusions (not shown).
4 Discussion Analyses of autoantibodies and of antibodies directed against foreign antigens have shown that their V region gene segments are rearranged according to the same mechanisms (reviewed in [22]). In addition, autoantibody Vgenes accumulate somatic mutations during the course of the disease as do conventional antibodies during the maturation of the immune response [23,24]. Our panel of anti-histone antibodies does not show any exclusive gene segment use that could be associated with histone-binding activity.The only gene use restriction observed was with the V ~ J 5 5 8family to which six of the seven VH genes belong. However, this might not necessarily be significant since 5558 represents approximately 50% of the V, gene repertoire and it is widely used among autoantibodies of various specificities [22,25, 261. The total number of V, gene families is not yet known [16,27], but it is noteworthy that, despite their diversity, all antibodies were assigned to one of the knownV, gene families. Since this first report of Vgene sequences in anti-histone antibodies does not indicate the existence of any new V, family, it supports the view that most of these families are known and that previous estimates of approximately 16 to 18V, families are likely to be correct [16,27]. This absence of gene use restriction among autoantibodies, even if they are directed against a limited set of autoepitopes, has also been observed by other authors [28, 291; a rare exception being antibodies directed against bromelainated erythrocytes that are nearly always encoded by genes belonging to the V ~ l family l [30]. Nevertheless, the sharing of a VH gene segment between
The relatively acidic (6.0-6.3) isoelectric point of MRA12, an IgG2a reacting strongly with cationic histone H1, parallels observations with IgG anti-DNA antibodies showing that such antibodies, directed against an anionic antigen, are often basic molecules [17-191. Moreover, this cationic fraction of IgG anti-DNA antibodies is preferentially found in kidney deposits in lupus syndromes [17]. It will be important to assess whether the electric charges of the anti-histone antibodies correlate with a potential pathogenic role. The second CDR or MRA12 is characterized by the presence of five negatively charged residues and only two positively charged residues. In a compilation of 112 CDR2 sequences belonging to the 5558 family [14], none has more than three negative residues. Four of the five negative residues in the CDR2 of MRA12 are clustered within six amino acid residues between positions 52 and 56 (Fig. 1). Nevertheless, this acidic stretch, albeit rare, is not unique to MRA12 since a search of the GenBank data library has indicated that it is also found in three antinitrophenyl antibodies ([31]; cf. below). More V region sequences need to be determined to assess whether acidic residues are a common feature among IgG anti-histone antibodies, but there is additional evidence t o suggest that such residues are important for histone-binding. To the author’s knowledge, the sequence of only one bona fide anti-histone IgG autoantibody has been previously reported [32]. This antibody, MRL-Histone7, is also encoded by a member of the 5558 family and has three negative residues and one positive residue in the second CDR. More striking is the presence of two contiguous aspartic acid residues at positions 31 and 32 in the first CDR. The importance of acidic residues is further supported by the fact that the putative histone-binding domains of the high-mobility group proteins (HMG 1,2,14 and 17; the main histone binding proteins present in the nucleus) are also composed of short stretches of acidic residues [33,34]. Another recently described H1-binding protein, nucleolin, is characterized by a long domain of glutamic and aspartic acid residues [35]. However, the anti-histone specificity of -12 cannot be merely explained by electrostatic interactions due to the presence of acidic residues since this IgG antibody is specific for H1 and does not react significantlywith the core histones which are also very basic proteins [S]. Clearly, other factors must be important for antibody binding to histones since, in addition, the IgM antibodies of the panel do not contain stretches of acidic residues in their V regions.
A role for acidic residues in IgG anti-histone antibodies would again mirror observations made with IgG anti-DNA antibodies, which have an elevated content of arginine residues [7,36-381, since arginine residues are a feature of several nuclear DNA-binding proteins [39-411. These arginine residues in the V regions accumulate mostly through somatic mutations and frame-shifts of the D segment [7,37,38], but the mechanisms that could lead to an enrichment in acidic residues can only be partially assessed from this study. There is no evidence that many somatic mutations have occurred in either of the two IgGz, anti-H1 antibodies, MRA3 and -12. The JHMRA~ segment differs by only one nucleotide from the germ-line
1730
M. Monestier
sequence while JHMRA12 is identical to the germ-line gene.The germ-line genes for either MRA3 or MRA12 VH segments are not known, but V H M R Ais ~ 100% identical to VHMRAll (an IgM antibody), suggesting that this particular segment is germ-line encoded. The closest match for vHM&?d2 in the GenBank library is V~P3.6.5,an IgGl anti-nitrophenyl antibody [31].Their sequences differ by 10 nucleotides resulting in 6 amino acid residue changes (not shown). Four of these amino acid differences in the framework or the first CDR do not affect the relative electric charges of either antibody (not shown), but two differences in the second CDR confirm the acidic nature of MRA12: a trytophan residue at position 50 and an aspartic acid residue at position 65 in V H M R A correspond ~~ to an arginine and a valine residue, respectively,inV~P3.6.5.It is important to note that the acidic stretch from residues 52 to 56 is identical in VHMRA12, V~P3.6.5and two other anti-nitrophenyl antibodies belonging to the same specificity subgroup (V), suggesting that this stretch is germ-line encoded [31]. This last observation suggests that, if this acidic segment is important for binding to histone H1, it has been selected from the germ-line repertoire and not acquired through somatic mutations. The possibility that anti-nitrophenyl antibodies sharing this acidic stretch are cross-reactive with histones is currently being explored. There is also little evidence of somatic mutations in the L chains of MRA3 and MRA12 since they are closely related to the L chains of other autoantibodies: V,MRA3 differs by only three nucleotides fromV,AM12, an IgG2b rheumatoid factor L chain gene [23], and V,MRA12 is 100% identical to V,BxWDNA14, the L chain gene of an IgM anti-DNA antibody from an (NZB x NZW)Fl mouse [42].These last observations indicate that autoantibodies of different specificities can be encoded by similar or identical V region genes. In view of several recent studies supporting the role of direct autoantigen stimulation in the activation of autoreactive clones [7,19,23,24,29,36,38], the sharing of genes between various subsets of autoantibodies suggests that certain V genes are important for the autoreactive specificity of these antibodies. It is conceivable that the use of these V genes is required for binding to the native stimulatory autoantigen (such as chromatin or nucleosomes), while the other diversity mechanisms (opposite chainvgenes, D and J gene segments, somatic mutations) will be responsible for the fine specificity of the autoantibody. The author wishes to express his gratitude to Dr. Martin Weigertfor his support of the project and for providing him access to the sequencing facility at the Fox Chase Cancer Center and acknowledges the cooperation of allmembers of Dr. Weigert’s laboratory and particularly the technical expertise of Ms. Anita Cywinski. The author also thanks Ms. Nancy Blazejewski for her assistance with the isoelectric focusing. Received March 10, 1991.
5 References 1 Rubin, R. L. and Waga, S., J. Rheurnatol. 1987.14 (Suppl. 13): 118. 2 Muller, S. and Van Regenmortel, M. H.V., Int. J. Immunopathol. Pharmacol. 1988. 1: 139.
Eur. J. Immunol. 1991.21: 1725-1731 3 Shoenfeld,Y. and Segol, O., Clin. Exp. Rheumatol. 1989. 7: 265. 4 Monestier, M., Losman, J. A., Fasy, T. M., Debbas, M. E., Massa, M., Albani, S., Bohm, L. and Martini, A., Arthritis Rheum. 1990. 33: 1836. 5 Gioud, M., Kotzin, B. L., Rubin, R. L., J o s h , F. G. and Tan, E. M., J. Immunol. 1983. 131: 269. 6 Monestier, M., Fasy,T. M., Debbas, M. E. and Bohm, L., Clin. Exp. Immunol. 1990. 81: 39. 7 Shlomchik, M., Mascelli, M., Shan, H., Radic, M. Z., Pisetsky, D., Marshak-Rothstein, A. and Weigert, M., J. Exp. Med. 1990. 171: 265. 8 Monestier, M., Fasy,T. M. and Bohm, L., Mol. Immunol. 1989. 26: 749. 9 Badley, J. E., Bishop, G. A., St. John, T. and Frelinger, J. A., Biotechniques 1988. 6: 114. 10 Strohal, R., Kroemer, G., Wick, G. and Kofler, R., Nucleic Acids Res. 1987. 15: 2771. 11 Radic, M. Z., Mascelli, M. A., Erikson, J., Shan, H., Shlomchik, M. and Weigert, M. G., Cold Spring Harbor Symp. Quant. Biol. 1990. 54: 933. 12 Geliebter, J., Focus 1987. 9: 5. 13 Devereux, J., Haeberli, l? and Smithies, O., Nucleic Acids Res. 1984. 12: 387. 14 Kabat, E. A., Wu, T. T., Reid-Miller, M., Perry, H. M. and Gottesman, K. S., Sequences of proteins of immunological interest, U.S. Government Printing Ofice, Bethesda 1987. 15 Dildrop, R., Immunol. %day 1984. 5: 84. 16 Potter, M., Newell, J. B., Rudikoff, S. and Haber, E., Mol. Immunol. 1982. 19: 1619. 17 Ebling, F. and Hahn, B. H., Arthritis Rheum. 1980. 23: 392. 18 Madaio, M. l?, Carlson, J., Cataldo, J., Ucci, A., Migliorini, €! and Pankewycz, O., J. Immunol. 1987. 138: 2883. 19 OKeefe,T. L., Bandyopadhyay, S., Datta, S. K. and ImanishiKari, T., J. Immunol. 1990. 144: 4275. 20 Kreth, H.W. and Williamson, A. R., Eur. J. Immunol. 1973.3: 141. 21 Cantor, C. R. and Schimmel, l? R., Biophysical Chemistry, W. H. Freeman and Co., San Francisco 1980, p. 41. 22 Kofler, R., Dixon, F. J. and Theoflopoulos, A. N., Immunol. Today 1987. 8: 374. 23 Shlomchik, M. J., Marshak-Rothstein, A., Wolfowicz, C. B., Rothstein, T. L. and Weigert, M. G., Nature 1987. 328: 805. 24 Marion,T. M., Bothwell, A. L. M., Briles, D. E. and Janeway, C. A., Jr., J. Immunol. 1989. 142: 4269. 25 Monestier, M., Manheimer-Lory, A., Bellon, B., Painter, C., Dang, H., Talal, N., Zanetti, M., Schwartz, R., Pisetsky, D., Kuppers, R., Rose, N., Brochier, J., Klareskog, L., Holmdahl, R., Erlanger, B., Alt, F. and Bona, C. A., J. Clin. Invest. 1986. 78: 753. 26 Komisar, J. L. ,Yeung, K.Y., Crawley, R. C. ,Tala],N. and Teale, J. M., J. Immunol. 1989. 143: 340. 27 Strohal, R., Helmberg, A., Kroemer, G. and Kofler, R., Immunogenetics 1989. 30: 475. 28 Gleason, S. L., Gearhardt, €!, Rose, N. R. and Kuppers, R. C., J. Immunol. 1990. 145: 1768. 29 Reininger, L., Shibata,T., Ozaki, S., Shirai,T., Jaton, J.-C. and Izui, S., Eur. J. Immunol. 1990. 20: 771. 30 Reininger, L., Kaushik, A., Izui, S. and Jaton, J.-C., Eur. J. Immunol. 1988. 18: 1521. 31 Boersch-Supan, M. E., Aganval, S.,White-Scharf, M. E. and Imanishi-Kari, T., J. Exp. Med. 1985. 161: 1272. 32 Kofler, R., Noonan, D. J., Strohal, R., Balderas, R. S., Moller, N. P. H., Dixon, F. J. and Theofilopoulos, A. N., Eur. J. Imrnunol. 1987. 17: 91. 33 Reeck, G. R. and Teller, D. C., in Bekhor, I. (Ed.), Progress in nonhistone protein research, vol II, CRC Press, Boca Raton 1985, p. 1. 34 Bustin, M., Lehn, D. A. and Landsmann, D., Biochim. Biophys. Acta 1990. 1049: 231. 35 Erard, M. S., Belenguer, P, Caizergues-Ferrer, M., Pantaloni, A. and Amalric, F., Eur. J. Biochem. 1988. 175: 525.
Eur. J. Immunol. 1991. 21: 1725-1731 36 Shlomchik, M . , Aucoin, A. H., Pisetsky, D. S. and Weigert, M. G., Proc. Natl. Acad. Sci. USA 1987. 84: 150. 37 Eilat, D.,Webster, D. M. and Rees, A. R., J. Immunol. 1988. 141: 1745. 38 Marion,T. M. ,Tillman, D. M. and Jou, N.-T., J. Immunol. 1990. 145: 2322. 39 Seeman, N. C., Rosenberg, J. M. and Rich, A., Proc. Natl. Acad. Sci. USA 1976. 73: 804.
Anti-histone V genes
1731
40 McClarin, J. A., Frederick, C. A.,Wang, B., Greene, I?, Boyer, H-W., Grable, J. and Rosenberg, J. M., Science 1986. 234: 1526. 41 Wu, R. S., Panusz, H.T., Hatch, C. L. and Bonner,W. L., CRC Crit. Rev. Biochern. 1987. 20: 201. 42 Kofler, R., Strohal, R., Balderas, R. S., Johnson, M. E., Noonan, D. J., Duchosal, M. A., Dixon, F. J. and Theofilopoulos, A. N., J. Clin. Invest. 1988. 82: 852.
Marc Monestier Garden State Cancer Center and Center for Molecular Medicine and Immunology, Newark
Anti-histone V genes
1725
Variable region genes of anti-histone autoantibodies from a MRL/Mp-@d&rmouse* The variable region nucleotide sequences of seven (five IgM and two IgG) anti-histone monoclonal antibodies from a single MFWMp-Zprllpr mouse have been determined. These antibodies are not clonally related and used diverseV, D and J genes. However, six of the seven antibodies have VH segments encoded by genes from the 3558 family, two of these (an IgM and an IgG) share an identical VHgene.The isoelectric points of MRA3 and MRA12, the two IgG antibodies of the panel, range from 6.3 to 7.0 and from 6.0 to 6.3, respectively. The second conplementarity-determiningregion (CDR) of theVH gene of MRA12 (the most acidic and the most strongly histone-reactive antibody) includes only two positively charged but five negatively charged amino acid residues. This feature is unusual since the equivalent CDR in most v ~ J 5 5 8genes are not comprised predominantly of acidic residues and suggests that such negatively charged residues are important for antibody binding to histones.
1 Introduction Histones are major targets of self-reactive antibodies in several autoimmune syndromes. For instance, in humans, anti-histone antibodies are highly prevalent in spontaneous and drug-induced lupus, juvenile arthritis, primary biliary cirrhosis and other autoimmune diseases [l-41. Antihistone antibodies are also detected in the sera of mice prone to develop spontaneous autoimmune syndromes such as MRLMp-lpdlpr, NZB and (NZB x NZW)F1 15, 61. As with other nuclear autoantigens, understanding the mechanisms leading to the production of anti-histone antibodies is an essential question concerning the pathogenesis of autoimmune diseases. The identification of the sequences of the V region genes of antinuclear antibodies such as anti-DNA antibodies has brought important information on the molecular basis of the DNA-antibody interactions and on the mechanisms leading to the expansion of autoreactive clones [7]. To explore these autoimmune phenomena further, we recently obtained a panel of seven anti-histone mAb from a single autoimmune MRLMp-ZpdZpr mouse [8]. In this communication, the V region gene sequences of these anti-histone antibodies are reported.
2 Materials and methods 2.1 mAb
(IgG2,,x).The hybridomas were obtained from the somatic fusion of the splenocytes of an MRLIMp-Zprllprmousewith the SP210 myeloma. The five mAb of the IgM isotype bind to histones H1 and/or H3, while the two IgGZa,xantibodies are specific for histone H1. The fine specificities of these mAb have been previously described [S].
2.2 mRNA sequencing Poly(A)+ RNA was isolated from hybridoma cells using oligo(dT) cellulose according to a previously described method [9]. The VH regions were sequenced directly from the poly(A)+ RNA template by the dideoxy chain termination method using isotype-specific oligonucleotide primers. The J, gene used by these antibodies were identified by synthesizing cDNA from poly(A)+ RNA using four different primers specificfor each of the four functional mouse J, gene segments. Since none of the antibodies was found to use J,2, their L chainV regions were also sequenced by the dideoxy termination method using the appropriate J,specific primer (J,2 is used in the non-functional transcript originating from the SP210 fusion partner; [lo]). The sequences of the oligonucleotide primers and the sequencing protocol were as previously reported [7, 11, 121. The nuclei acid sequences were compared with Ig sequences in the GenBank data library [13].TheVregion sequences were assigned to known gene families by comparison to previously published compilations [14-16].
The following antibodies were utilized in this study, MRAl (IgM,x), MRA3 (IgG2a,x), MRA5 (IgM,x), h~fRA7 2.3 IEF (IgM,x), MRAlO (IgM,x) M R A l l (IgM,x), and MRA12 This was carried out at 8°C in a flatbed apparatus (LKB Multiphor 11, Bromma, Sweden). The anolyte was 0.5 M [I 93771 acetic acid and the catholyte was 1M sodium hydroxide. Approximatively 10 pg of each antibody was applied onto * This work was supportedin part by NIHgrant A1 26665 to M. M. an Isogel agarose gel, pH 3-10 (FMC,Rockland, ME). and grants CA 06927, GM 20964, CA 09035 and an appropria- Focusing was performed at a constant power of 1W for the tion from the Commonwealthof Pennsylvaniato the Fox-Chase first 12 min and 25 W-1000 V for the remainder of the run. Cancer Center. Focusing was considered complete when the current had stopped decreasing significantly(less than 1 mA in 10 min). Correspondence: Marc Monestier, Center for Molecular Medicine The gel was then fixed and stained with Coomassie blue as and Immunology, One Bruce Street, Newark, NJ 07103, USA recommended by the manufacturer. A calibration curve was obtained by measuring the distance from the cathode Abbreviation: CDR: Complementarity-determiningregion 0 VCH Verlagsgesellschaft mbH, D-6940 Weinheim, 1991
+
0014-2980/91/0707-1725$3S O .25/0
1726
Eur. J. Immunol. 1991.21: 1725-1731
M. Monestier
of Isogel PI markers (FMC) and the migration distance of the antibodies was plotted against this curve.
gene segment. TheV gene use for the L chain is even more diverse than for the H chain: all seven light chains are encoded by separate V genes belonging to seven different V,families (Table 1).
3 Results 3.2 Gene segment usage in anti-histone antibodies
3.1 V gene usage in anti-histone antibodies
The nucleotide sequences and the deduced protein A great variety of gene segments are also represented in the sequences of the H and L chains of the seven anti-histone panel of anti-histone antibodies. Each of the four JH gene mAb are reported in Figs. 1 to 4. All the anti-histone segments is used at least once among the seven anti-histone antibodies of this panel are encoded by VHgenes belonging antibodies (Table 1). The sequences of the seven JH segto the 5558 family,with the exception of MRA5, encoded by ments are identical to the germ-line genes, with the a member of the S107 family (Table 1). Only two antibod- exception of MRA3, in which a nucleotide substitution ies, MRA3 and MRA11, are encoded by an identical VH results in the replacement of a tyrosine by a cysteine residue (Fig. 1). MRA3 and MRA11, which are encoded by an identical VHgene segment, use different D and JH segments indicating that these two antibodies are not clonally Table 1. Gene segment usage in anti-histone antibodiesa) related.The D segment usage is also diverse since segments from all three families are used including two indirect D-D joinings (Table 1). The L chain segments of these antiAntibody Isotype VH D JH vx Jx histone antibodies were equally heterogeneous since their I&¶ J558 Q52+SPZ-C J"4 VxS Jx5 J, segments originated from three of the four functional J, MRAl IgGz, 5558 J H ~ vx21 Jx4 MRA3 Q52 genes (Table 1). J"2 Vxl IgM S107 MRAS Jxl IgM 5558 SP2+SP2-c J H ~ Vx23 Jx4 MRA7 J H ~vxlo Jx4 IgM 5558 MRAlO 3.3 IEF IgM 5558 FL16 J H ~ Vx9 Jx4 MRAll SP2 J H ~ vx2 Jx4 MRA12 IgG2;, 5558 Previous studies have indicated that IgG autoantibodies to DNA, particularly if they are nephritogenic, are often a) The VH sequences of MRAS and MRAlO did not allow cationic molecules [17-191. The assess whether the electric assignment to a particular D gene. charges of the antibody molecules played a role in the FRI 1
10
20
30
60
40
NRAlH WASH
MRA7H YRAl0H YRAllH MRA12H MUASH
CDRI
___
70
80
98
340
360
I
I
360
--- __-_--
ACC TCA QTC ACC GTC TCC TCA A-T C-- --A
----- --- --- --- --- --_-- --- --- --- --- --- --A-0 --- ___ -__--- ----- --- --- --- --- --- -----
___
A-T C--
--A
--_ --- ---
I
FRII
100
110
CDRII
120
130
140
160
Figurel. Nucleotide sequences of the portions of mRNA encoding VH regions of anti-histone autoantibodies. Sequence identities are indicated by horizontal bars. N signifies undetermined bases. The start of the framework (FR) and CDR regions are offset from preceding sequences by a space and designated by abbreviations above the first codon in each region. Spaces have been introduced to maximize alignement. MRAS antibody, which is encoded by aVH gene belonging to the S107 family, is listed last.
Eur. J. Immunol. 1991. 21: 1725-1731
Anti-histone V genes
1727
MRAl MRA3
MRA7 MRAlO MRAl 1
MRAl2 MRA5
MRAl
Y
S
F
S
T
A
30 T G
Y
CDRl N M
H
W
V
K
Q
40 S
S
L
T
S
E
D
5
H
G
K
S
L
E
W
I
90 S A
V
Y
Y
C
A
R
G
0
Y
I
D
K
L
MRA3
MRA7 MRAlO
MRAll MRA12 MRA5
MRAl MRA3
MRA7 MRAlO MRAll MRAl2 MRA5
MRAl MRA3 MRA7 MRAlO
Y
80 M Q
L
N
MRAll MRAl2 MRA5
binding to histones, which are cationic proteins, the isoelectric points of the two IgG antibodies of the panel were determined (Fig.5). The PI of M U 3 ranges between 6.3 and 7.0 while IvlRA12 has a more restricted PI spectrum, 6.0-6.3. The PI value of MRA12, the most strongly H1-reactive antibody [S], is at the acidic end of the usual range of murine IgG antibodies [20] and it contrasts with the alkaline isoelectric points often observed with IgG anti-DNA antibodies [17-19]. Differences in isoelectric points among Ig result partially from differences in the amino acid contents of their V regions. Examination of the sequence of MRA12 reveals that the second complementarity-determining region (CDR) of the H chain is characterized by the presence of 5 acidic residues (2 glutamic acid and 3 aspartic acid residues) out of a total of 17 residues.To assess the potential significance of electric charges, we examined, in the second CDR, the frequency of the amino acid residues whose side chains are ionizable and can contribute to the overall charge of the molecule [21]. Table 2 shows the frequency of the residues with side chains having alkaline pK, values (arigine, lysine and tyrosine) or acidic pK, values (glutamic and aspartic acids) in the second CDR of the H chains of -12 or of a compilation of 44 V ~ J 5 5 8sequences [15]. The other two amino acid residues
with ionizable side chains were not included in this table since cysteine is infrequently found in this region and histidine (also rarely found) has an almost neutral (6.5-7.0) side chain pK, [21]. This Table shows that the second CDR of V H M R Ahas ~ ~more glutamic and aspartic acid residues Table2. Amino acid residues with ionizable side chains in the and of a compilation of second CDR of the H chains of -12 V~J558antibodiesa)
MRAl2 V~J558Compilation Number of Frequency Number of Frequency residues residues Glutamic acid Aspartic acid
Qmine Arginine Lysine Other residues
2 3 1 0 2 10
0.117 0.176 0.058
0 0.117 0.588
29 36
71 17 91
m
0.038 0.048 0.095 0.022
0.121 0.672
a) The compilation of 44 V ~ J 5 5 8sequences is from Dildrop 1151.
1728
M. Monestier
MRAl MRA3
s s
MRA5 M 7 MRAlO MRAll
MRA12
Eur. J. Immunol. 1991. 21: 1725-1731
CDRl
Q Q Q
V - V E Y Y - L V H S N S D D - L L D S D
T N K S -
I -
E Q
Q
S G G I I I G
30 40 S S Y F H W Y Q Q K P G T - L M Q - - - - - - -
N N K
T D N T
-
L L L L L
N S N
-
A I L V I M - L
A G G G G
T M V V I V
Y -
Y F F -
F L
L L
-
R
S -
S S P K L W Q P - - - L - Q - P - - L H E - - R - L D G T V - - L - K - - - T L - Q - - - R L
80
MRAl MRA3 MRA5 MRA7
MRAlO MRA11 MRA12
S H S N -
- - K
-
-
S P R N
V L - L R -
E -
A E P P Y
E D -
- - -
D -
-
C Q - S Q S T - - W Q G T
Figure 4. V, region amino acid sequences deduced from the nucleotide sequences of Fig. 3.
Eur. J. Immunol. 1991. 21: 1725-1733
Anti-histone V genes
1729
MRA3 and MRAl1 could suggest that certain members of the 5558 family are preferentially used among anti-histone antibodies.
Figure 5. IEF of mAb MRA3 and MRA12.
and less residues with alkaline pKa values than other antibodies of the same VH family (x2=9.43, 0.001 < p < 0.001). A comparison of the MRA12 VHCDR2 sequence with another larger v ~ J 5 5 8sequence compilation [14] resulted in similar conclusions (not shown).
4 Discussion Analyses of autoantibodies and of antibodies directed against foreign antigens have shown that their V region gene segments are rearranged according to the same mechanisms (reviewed in [22]). In addition, autoantibody Vgenes accumulate somatic mutations during the course of the disease as do conventional antibodies during the maturation of the immune response [23,24]. Our panel of anti-histone antibodies does not show any exclusive gene segment use that could be associated with histone-binding activity.The only gene use restriction observed was with the V ~ J 5 5 8family to which six of the seven VH genes belong. However, this might not necessarily be significant since 5558 represents approximately 50% of the V, gene repertoire and it is widely used among autoantibodies of various specificities [22,25, 261. The total number of V, gene families is not yet known [16,27], but it is noteworthy that, despite their diversity, all antibodies were assigned to one of the knownV, gene families. Since this first report of Vgene sequences in anti-histone antibodies does not indicate the existence of any new V, family, it supports the view that most of these families are known and that previous estimates of approximately 16 to 18V, families are likely to be correct [16,27]. This absence of gene use restriction among autoantibodies, even if they are directed against a limited set of autoepitopes, has also been observed by other authors [28, 291; a rare exception being antibodies directed against bromelainated erythrocytes that are nearly always encoded by genes belonging to the V ~ l family l [30]. Nevertheless, the sharing of a VH gene segment between
The relatively acidic (6.0-6.3) isoelectric point of MRA12, an IgG2a reacting strongly with cationic histone H1, parallels observations with IgG anti-DNA antibodies showing that such antibodies, directed against an anionic antigen, are often basic molecules [17-191. Moreover, this cationic fraction of IgG anti-DNA antibodies is preferentially found in kidney deposits in lupus syndromes [17]. It will be important to assess whether the electric charges of the anti-histone antibodies correlate with a potential pathogenic role. The second CDR or MRA12 is characterized by the presence of five negatively charged residues and only two positively charged residues. In a compilation of 112 CDR2 sequences belonging to the 5558 family [14], none has more than three negative residues. Four of the five negative residues in the CDR2 of MRA12 are clustered within six amino acid residues between positions 52 and 56 (Fig. 1). Nevertheless, this acidic stretch, albeit rare, is not unique to MRA12 since a search of the GenBank data library has indicated that it is also found in three antinitrophenyl antibodies ([31]; cf. below). More V region sequences need to be determined to assess whether acidic residues are a common feature among IgG anti-histone antibodies, but there is additional evidence t o suggest that such residues are important for histone-binding. To the author’s knowledge, the sequence of only one bona fide anti-histone IgG autoantibody has been previously reported [32]. This antibody, MRL-Histone7, is also encoded by a member of the 5558 family and has three negative residues and one positive residue in the second CDR. More striking is the presence of two contiguous aspartic acid residues at positions 31 and 32 in the first CDR. The importance of acidic residues is further supported by the fact that the putative histone-binding domains of the high-mobility group proteins (HMG 1,2,14 and 17; the main histone binding proteins present in the nucleus) are also composed of short stretches of acidic residues [33,34]. Another recently described H1-binding protein, nucleolin, is characterized by a long domain of glutamic and aspartic acid residues [35]. However, the anti-histone specificity of -12 cannot be merely explained by electrostatic interactions due to the presence of acidic residues since this IgG antibody is specific for H1 and does not react significantlywith the core histones which are also very basic proteins [S]. Clearly, other factors must be important for antibody binding to histones since, in addition, the IgM antibodies of the panel do not contain stretches of acidic residues in their V regions.
A role for acidic residues in IgG anti-histone antibodies would again mirror observations made with IgG anti-DNA antibodies, which have an elevated content of arginine residues [7,36-381, since arginine residues are a feature of several nuclear DNA-binding proteins [39-411. These arginine residues in the V regions accumulate mostly through somatic mutations and frame-shifts of the D segment [7,37,38], but the mechanisms that could lead to an enrichment in acidic residues can only be partially assessed from this study. There is no evidence that many somatic mutations have occurred in either of the two IgGz, anti-H1 antibodies, MRA3 and -12. The JHMRA~ segment differs by only one nucleotide from the germ-line
1730
M. Monestier
sequence while JHMRA12 is identical to the germ-line gene.The germ-line genes for either MRA3 or MRA12 VH segments are not known, but V H M R Ais ~ 100% identical to VHMRAll (an IgM antibody), suggesting that this particular segment is germ-line encoded. The closest match for vHM&?d2 in the GenBank library is V~P3.6.5,an IgGl anti-nitrophenyl antibody [31].Their sequences differ by 10 nucleotides resulting in 6 amino acid residue changes (not shown). Four of these amino acid differences in the framework or the first CDR do not affect the relative electric charges of either antibody (not shown), but two differences in the second CDR confirm the acidic nature of MRA12: a trytophan residue at position 50 and an aspartic acid residue at position 65 in V H M R A correspond ~~ to an arginine and a valine residue, respectively,inV~P3.6.5.It is important to note that the acidic stretch from residues 52 to 56 is identical in VHMRA12, V~P3.6.5and two other anti-nitrophenyl antibodies belonging to the same specificity subgroup (V), suggesting that this stretch is germ-line encoded [31]. This last observation suggests that, if this acidic segment is important for binding to histone H1, it has been selected from the germ-line repertoire and not acquired through somatic mutations. The possibility that anti-nitrophenyl antibodies sharing this acidic stretch are cross-reactive with histones is currently being explored. There is also little evidence of somatic mutations in the L chains of MRA3 and MRA12 since they are closely related to the L chains of other autoantibodies: V,MRA3 differs by only three nucleotides fromV,AM12, an IgG2b rheumatoid factor L chain gene [23], and V,MRA12 is 100% identical to V,BxWDNA14, the L chain gene of an IgM anti-DNA antibody from an (NZB x NZW)Fl mouse [42].These last observations indicate that autoantibodies of different specificities can be encoded by similar or identical V region genes. In view of several recent studies supporting the role of direct autoantigen stimulation in the activation of autoreactive clones [7,19,23,24,29,36,38], the sharing of genes between various subsets of autoantibodies suggests that certain V genes are important for the autoreactive specificity of these antibodies. It is conceivable that the use of these V genes is required for binding to the native stimulatory autoantigen (such as chromatin or nucleosomes), while the other diversity mechanisms (opposite chainvgenes, D and J gene segments, somatic mutations) will be responsible for the fine specificity of the autoantibody. The author wishes to express his gratitude to Dr. Martin Weigertfor his support of the project and for providing him access to the sequencing facility at the Fox Chase Cancer Center and acknowledges the cooperation of allmembers of Dr. Weigert’s laboratory and particularly the technical expertise of Ms. Anita Cywinski. The author also thanks Ms. Nancy Blazejewski for her assistance with the isoelectric focusing. Received March 10, 1991.
5 References 1 Rubin, R. L. and Waga, S., J. Rheurnatol. 1987.14 (Suppl. 13): 118. 2 Muller, S. and Van Regenmortel, M. H.V., Int. J. Immunopathol. Pharmacol. 1988. 1: 139.
Eur. J. Immunol. 1991.21: 1725-1731 3 Shoenfeld,Y. and Segol, O., Clin. Exp. Rheumatol. 1989. 7: 265. 4 Monestier, M., Losman, J. A., Fasy, T. M., Debbas, M. E., Massa, M., Albani, S., Bohm, L. and Martini, A., Arthritis Rheum. 1990. 33: 1836. 5 Gioud, M., Kotzin, B. L., Rubin, R. L., J o s h , F. G. and Tan, E. M., J. Immunol. 1983. 131: 269. 6 Monestier, M., Fasy,T. M., Debbas, M. E. and Bohm, L., Clin. Exp. Immunol. 1990. 81: 39. 7 Shlomchik, M., Mascelli, M., Shan, H., Radic, M. Z., Pisetsky, D., Marshak-Rothstein, A. and Weigert, M., J. Exp. Med. 1990. 171: 265. 8 Monestier, M., Fasy,T. M. and Bohm, L., Mol. Immunol. 1989. 26: 749. 9 Badley, J. E., Bishop, G. A., St. John, T. and Frelinger, J. A., Biotechniques 1988. 6: 114. 10 Strohal, R., Kroemer, G., Wick, G. and Kofler, R., Nucleic Acids Res. 1987. 15: 2771. 11 Radic, M. Z., Mascelli, M. A., Erikson, J., Shan, H., Shlomchik, M. and Weigert, M. G., Cold Spring Harbor Symp. Quant. Biol. 1990. 54: 933. 12 Geliebter, J., Focus 1987. 9: 5. 13 Devereux, J., Haeberli, l? and Smithies, O., Nucleic Acids Res. 1984. 12: 387. 14 Kabat, E. A., Wu, T. T., Reid-Miller, M., Perry, H. M. and Gottesman, K. S., Sequences of proteins of immunological interest, U.S. Government Printing Ofice, Bethesda 1987. 15 Dildrop, R., Immunol. %day 1984. 5: 84. 16 Potter, M., Newell, J. B., Rudikoff, S. and Haber, E., Mol. Immunol. 1982. 19: 1619. 17 Ebling, F. and Hahn, B. H., Arthritis Rheum. 1980. 23: 392. 18 Madaio, M. l?, Carlson, J., Cataldo, J., Ucci, A., Migliorini, €! and Pankewycz, O., J. Immunol. 1987. 138: 2883. 19 OKeefe,T. L., Bandyopadhyay, S., Datta, S. K. and ImanishiKari, T., J. Immunol. 1990. 144: 4275. 20 Kreth, H.W. and Williamson, A. R., Eur. J. Immunol. 1973.3: 141. 21 Cantor, C. R. and Schimmel, l? R., Biophysical Chemistry, W. H. Freeman and Co., San Francisco 1980, p. 41. 22 Kofler, R., Dixon, F. J. and Theoflopoulos, A. N., Immunol. Today 1987. 8: 374. 23 Shlomchik, M. J., Marshak-Rothstein, A., Wolfowicz, C. B., Rothstein, T. L. and Weigert, M. G., Nature 1987. 328: 805. 24 Marion,T. M., Bothwell, A. L. M., Briles, D. E. and Janeway, C. A., Jr., J. Immunol. 1989. 142: 4269. 25 Monestier, M., Manheimer-Lory, A., Bellon, B., Painter, C., Dang, H., Talal, N., Zanetti, M., Schwartz, R., Pisetsky, D., Kuppers, R., Rose, N., Brochier, J., Klareskog, L., Holmdahl, R., Erlanger, B., Alt, F. and Bona, C. A., J. Clin. Invest. 1986. 78: 753. 26 Komisar, J. L. ,Yeung, K.Y., Crawley, R. C. ,Tala],N. and Teale, J. M., J. Immunol. 1989. 143: 340. 27 Strohal, R., Helmberg, A., Kroemer, G. and Kofler, R., Immunogenetics 1989. 30: 475. 28 Gleason, S. L., Gearhardt, €!, Rose, N. R. and Kuppers, R. C., J. Immunol. 1990. 145: 1768. 29 Reininger, L., Shibata,T., Ozaki, S., Shirai,T., Jaton, J.-C. and Izui, S., Eur. J. Immunol. 1990. 20: 771. 30 Reininger, L., Kaushik, A., Izui, S. and Jaton, J.-C., Eur. J. Immunol. 1988. 18: 1521. 31 Boersch-Supan, M. E., Aganval, S.,White-Scharf, M. E. and Imanishi-Kari, T., J. Exp. Med. 1985. 161: 1272. 32 Kofler, R., Noonan, D. J., Strohal, R., Balderas, R. S., Moller, N. P. H., Dixon, F. J. and Theofilopoulos, A. N., Eur. J. Imrnunol. 1987. 17: 91. 33 Reeck, G. R. and Teller, D. C., in Bekhor, I. (Ed.), Progress in nonhistone protein research, vol II, CRC Press, Boca Raton 1985, p. 1. 34 Bustin, M., Lehn, D. A. and Landsmann, D., Biochim. Biophys. Acta 1990. 1049: 231. 35 Erard, M. S., Belenguer, P, Caizergues-Ferrer, M., Pantaloni, A. and Amalric, F., Eur. J. Biochem. 1988. 175: 525.
Eur. J. Immunol. 1991. 21: 1725-1731 36 Shlomchik, M . , Aucoin, A. H., Pisetsky, D. S. and Weigert, M. G., Proc. Natl. Acad. Sci. USA 1987. 84: 150. 37 Eilat, D.,Webster, D. M. and Rees, A. R., J. Immunol. 1988. 141: 1745. 38 Marion,T. M. ,Tillman, D. M. and Jou, N.-T., J. Immunol. 1990. 145: 2322. 39 Seeman, N. C., Rosenberg, J. M. and Rich, A., Proc. Natl. Acad. Sci. USA 1976. 73: 804.
Anti-histone V genes
1731
40 McClarin, J. A., Frederick, C. A.,Wang, B., Greene, I?, Boyer, H-W., Grable, J. and Rosenberg, J. M., Science 1986. 234: 1526. 41 Wu, R. S., Panusz, H.T., Hatch, C. L. and Bonner,W. L., CRC Crit. Rev. Biochern. 1987. 20: 201. 42 Kofler, R., Strohal, R., Balderas, R. S., Johnson, M. E., Noonan, D. J., Duchosal, M. A., Dixon, F. J. and Theofilopoulos, A. N., J. Clin. Invest. 1988. 82: 852.
