GASTROENTEROLOGY 2001;121:15–23 Restricted VH Gene Usage in Lamina Propria B Cells Producing Anticolon Antibody From Patients With Ulcerative Colitis NAGAMU INOUE,* MAMORU WATANABE,‡ TOSHIRO SATO,* AKIRA OKAZAWA,* MOTOMI YAMAZAKI,‡ TAKANORI KANAI,‡ HARUHIKO OGATA,* YASUSHI IWAO,* HIROMASA ISHII,* and TOSHIFUMI HIBI‡ *Department of Internal Medicine, School of Medicine, and ‡Keio Cancer Center, Keio University, Tokyo, Japan Background & Aims: Autoimmune responses against colonic epithelium may play a role in the development of colonic inflammation associated with ulcerative colitis (UC). In this study, we established and characterized B-cell lines and clones that produced anticolon antibody from inflamed colonic mucosa of UC subjects. Methods: B-cell lines were generated through Epstein–Barr virus transformation of lamina propria lymphocytes (LPLs) from colonic mucosa and peripheral blood lymphocytes, and these lines were screened for the production of anticolon antibodies. B-cell lines were then cloned by limiting dilution culture, and messenger RNA expression of immunoglobulin heavy-chain variable region (VH) was assessed. Results: VH gene families used in B-cell lines established from LPLs of normal controls were diverse, and B-cell lines from UC LPLs expressed a restricted VH3 family usage. All 15 clones from UC used a restricted VH3 gene family, whereas diverse VH gene families were used by 24 clones from normal controls. The analysis of nucleotide sequences indicated that these clones were derived from various germline gene segments. Conclusions: The restricted VH gene usage in anticolon autoantibodies producing B-cell clones suggests that a particular antigenic stimulus contributes to the pathogenesis of UC. U lcerative colitis (UC) is a chronic inflammatory disease of the colon and rectum of unknown origin, but there is increasing evidence that autoimmune mechanisms play a role in the pathogenesis of this disorder.1,2 T cell–mediated immune responses, such as cytotoxic T cells that react with autoantigens in colonic epithelium, may participate in the initiation and/or maintenance of colonic inflammation.3,4 Autoantibodies against colonic epithelial cells are frequently found in the serum of individuals with UC.5–7 Although numerous investigations of the autoantibodies found in UC have been performed, the corresponding antigens of these autoantibodies remain obscure. Recently, Geng et al.7 investigated autoantibodies against tropomyosins in UC and showed that those autoantibodies recognized distinct isoforms of tropomyosins from smooth muscle of intestinal epithelial cells. In addition, we have previously demonstrated a high incidence of anticolon antibodies in the serum of individuals with UC, which can react with colonic epithelial cells and colonic mucin produced by goblet cells.8 –11 These autoantibodies may contribute to the destruction of colonic mucosa through antibody-dependent cell-mediated cytotoxicity mechanisms against colonic epithelial cells.8,10 Therefore, the characterization of circulating autoantibodies may be very important for understanding the pathophysiology involved in UC. The genetic composition of antibodies reacting to the various kinds of endogenous or exogenous antigens has been investigated extensively in mice and humans. A diverse repertoire of antibodies is achieved through the combinational rearrangement of a certain number of gene segments, i.e., variable (VH), diversity (D), and joining (JH) regions of immunoglobulin heavy chain gene, as well as subsequent somatic mutation.12 In humans, there are an estimated 100 –200 VH germline genes that can be divided into 6 families (VH1–VH6) based on amino acid or nucleotide sequence similarity.13 Members of a given family are more than 80% identical at the nucleotide level. B lymphocytes responding to particular antigens, including autoantigens, have been found to use a restricted number of VH gene segments, followed by a process of somatic hypermutation and then antigendriven clonal selection.14 Moreover, sequences of immunoglobulin VH genes are of particular interest from a number of perspectives; for instance, they can provide a foundation for studies into the molecular basis of antibody interactions, and they may also be useful in studies Abbreviations used in this paper: EBV, Epstein–Barr virus; ELISA, enzyme-linked immunosorbent assay; HBSS, Hank’s balanced salt solution; Ig, immunoglobulin; LPL, lamina propria lymphocyte; mRNA, messenger RNA; PBL, peripheral blood lymphocyte; PCR, polymerase chain reaction. © 2001 by the American Gastroenterological Association 0016-5085/01/$35.00 doi:10.1053/gast.2001.25477 16 INOUE ET AL. on V gene usage in a variety of pathologic conditions such as B-cell malignancies or autoimmune diseases. In the current investigation, we established stable anticolon antibody–producing B-cell lines and clones from both the colonic mucosa and the peripheral blood of subjects with UC through transformation by Epstein– Barr virus (EBV) to elucidate the characteristics of anticolon antibodies in UC. We assessed the messenger RNA (mRNA) expression of immunoglobulin VH gene family usage of these B-cell lines and clone, and performed nucleotide sequence analysis of VH genes to identify potential characteristic features of the colonic antigens recognized by these anticolon antibodies in UC. Materials and Methods Subjects Seven subjects with UC were involved in the establishment of anticolon antibody–producing B-cell lines and clones. UC was diagnosed by typical clinical, radiographic, endoscopic, and pathologic findings, and all cases were of the chronic continuous or relapsing-remitting type and at an active stage of disease. Normal colonic mucosa obtained at colonoscopy in 7 subjects with colon cancer and 3 with colonic polyps served as controls. Histopathologic examination showed no evidence of malignancy and no inflammation in the control subjects. All the sera from the 7 subjects with UC were positive for anticolon antibodies by enzyme-linked immunosorbent assay (ELISA). In contrast, all the sera from the control subjects were negative for anticolon antibodies. Written informed consent was given by all individuals, and all experiments were approved by the Keio University Hospital Committee on Human Subjects. Lymphocyte Preparation Peripheral blood lymphocytes (PBLs) were isolated from heparinized blood samples using a Ficoll-Hypaque density gradient and washed 3 times with Hank’s balanced salt solution (HBSS). Colonic lamina propria lymphocytes (LPLs) were isolated from colonic mucosa by a modification of the method of Bull and Bookman.15 The colonic mucosa was obtained from fresh specimens at surgical resection or colonoscopy. The mucosa was dissected off of the underlying muscular layers and cut into small pieces with iris scissors. The tissue pieces were then washed with calcium- and magnesium-free HBSS and incubated while stirring for 5 minutes at 22°C in calcium- and magnesium-free HBSS containing 0.5 mmol/L dithiothreitol. The tissue was then washed and incubated twice while stirring for 60 minutes at 37°C in calcium- and magnesium-free HBSS containing 0.75 mmol/L EDTA to remove any epithelial cells. The supernatant was decanted, and the remaining tissue was incubated while stirring at 37°C in a 5% CO2 in RPMI 1640 medium containing antibiotics, 10% fetal calf serum, 0.02% collagenase (CLSPA; Worthington Biochemical Corp., Freehold, NJ), and 0.01% deoxyribonuclease I GASTROENTEROLOGY Vol. 121, No. 1 (DP; Worthington) for 18 hours. The crude cell suspension was then filtered through a 50-mm stainless steel mesh and washed. After resuspension in 100% fetal calf serum, the cell suspension was separated by Ficoll-Hypaque density gradient, and the lymphoid cell fraction was washed and resuspended in complete medium. Preparation of EBV EBV was obtained from the culture supernatant of the B95-8 marmoset cell line (American Type Culture Collection [ATCC], Rockville, MD). B95-8 cells were cultured at a concentration of 106 cells/mL and incubated for 10 –14 days at 37°C for virus production. The virus-containing supernatant was filtered through a 0.45-mm Millipore filter (Bedford, MA) and stored at 280°C. EBV Transformation Lymphocytes (PBLs or LPLs; 5 3 106) were suspended in 5 mL of the culture supernatant of B95-8 cell lines and then washed once with complete medium. Cells (1 3 104) exposed to EBV were seeded in each well of a 96-well plate. Plates were incubated in 5% CO2 at 37°C. At weekly intervals, two thirds of the medium was removed and replaced with fresh medium. Three or four weeks after culture was initiated, supernatant from each well was harvested and screened for anticolon antibody secretion by ELISA. To estimate the frequencies of anticolon antibody-secreting cells, the following formula was used: Frequency of Anticolon Antibody–Secreting Cells 5 (Number of ELISA-Positive Wells)/(Number of Cells per Well 3 Number of Wells). Screening for Anticolon Antibodies The human colonic epithelial cells were isolated from the colons of normal subjects by methods described previously.16 Briefly, the epithelial cells were disrupted with a polytron and the homogenate was centrifuged at 700g. The supernatant was sonicated and centrifuged at 10,000g. The supernatant of the sonicated homogenate was used as a colonic epithelial antigen for ELISA measurement. Protein concentration was determined by the Bradford assay. E.I.A. plates (Flow Laboratories, McLean, VA) were coated with the colonic epithelial antigen at 1.5 mg protein/mL overnight at 4°C. After being washed, the plates were incubated with 200 mL of blocking buffer (1% bovine serum albumin in 0.1 mol/L sodium bicarbonate, pH 8.2) at 22°C for 4 – 6 hours. The plates were then washed 3 times with phosphate-buffered saline and incubated with 100 mL of culture supernatant at 37°C for 1 hour. Culture supernatant was discarded, and the plates were washed 3 times with phosphate-buffered saline. The plates were then incubated with 100 mL of 1:2500 dilution of goat anti-human immunoglobulin (Ig)G conjugated to alkaline phosphatase (Tago, Burlingame, CA) at 37°C for 1 hour. After being washed again, the plates were incubated with 100 mL of 0.5 mmol/L p-nitrophenyl phosphate. The optical density was read at 409 nm by a Toyo EYT-III ELISA reader (Toyo, Tokyo, Japan). The culture supernatant July 2001 from EBV-transformed cell lines that produced a large amount of IgG or IgM, but not anticolon antibody, was used as a negative control. Pooled sera from patients with UC with high titers of anticolon antibody served as a positive control. Wells with optical densities above the mean 1 2 SD of the negative controls were considered positive for anticolon antibody activity. For measurement of IgG secreted into culture supernatants, a standard sandwich ELISA procedure was used. Cloning by Limiting Dilution Method EBV-transformed cell lines secreting anticolon antibodies were cloned by limiting dilution (0.5 cell per well) on the feeder layers of human PBLs treated with mitomycin C. Three or four weeks after the cell cloning, the wells were screened again for anticolon antibody production by ELISA. Reverse Transcription–Polymerase Chain Reaction Total RNA was extracted from 1 3 108 cells of each clone or bulk culture using RNAzol B (Tel-Test Inc., Friendswood, TX). First-strand DNA was synthesized from 2 mg of total RNA with 1 mg oligo(dT) primer and 400 U/mL Moloney murine leukemia virus reverse transcriptase (Perkin-Elmer Cetus Instruments, Norwalk, CT) by using SuperScript Preamplification System (Gibco BRL, Gaithersberg, MD) in 20 mL of reaction mixture. The 59-sense primers were homologous to conserved regions at the 59 end of the leader sequences for each of the 6 human Ig heavy-chain V gene families (VH1: 59-CCA TGG ACT GGA CCT GGA GG-39; VH2: 59-ATG GAC ATA CTT TGT TCC AC-39; VH3: 59-CCA TGG AGT TTG GGC TGA GC-39; VH4: 59-ATG AAA CAC CTG TGG TTC TT-39; VH5: 59-ATG GGG TCA ACC GCC ATC CT-39; VH6: 59-ATG TCT GTC TCC TTC CTC AT-39). The 39antisense primer is derived from a consensus sequence of 6 heavy-chain joining segments (JH: 59-ACC TGA GGA GAC GGT GAC CAG GGT-39).17 Each primer set yielded a polymerase chain reaction (PCR) product of the predicted size of approximately 430 base pairs. Primers for G3PDH (59 primer: 59-TGA AGG TCG GAG TCA ACG GAT TTG GT-39; 39 primer: 59-CAT GTG GGC CAT GAG GTC CAC CAC-39, Mapping amplimers; Clontech, Palo Alto, CA) were used as controls. For PCR, 2 mL of complementary DNA (cDNA) was amplified in the presence of 0.5 mmol/L each of the 59 and primers and 0.5 U of Taq DNA polymerase (Ampli Taq; Perkin-Elmer Cetus). PCR was performed in a DNA thermal cycler for 30 cycles (94°C for 45 seconds, 60°C for 45 seconds, and 72°C for 90 seconds) followed by a 7-minute extension at 72°C. Nine microliters of the PCR products were subjected to gel electrophoresis using 1.6% agarose gels stained with 0.5 mg/mL ethidium bromide. Nucleotide DNA Sequence Analysis The PCR products amplified by VH3 and JH primers were size-fractionated by gel electrophoresis in 2% low melting point agarose gels and purified by phenol-chloroform VH GENE USAGE IN ULCERATIVE COLITIS 17 extraction. The purified PCR products were directly sequenced by a modification of the dideoxynucleotide chain-termination method and through cycle sequencing using DNA polymerase (Sequencing High Cycle; Toyobo Co., Osaka, Japan). V-region sequences were analyzed by aligning homologous germline gene regions in reference to advanced BLAST searches.18 Statistical Analysis The statistical significance of the data was determined by Student t test. A P value of ,0.01 was considered significant. Results Estimated Frequencies of Anticolon Antibody–Producing B Cells in PBLs and LPLs To assess the VH gene usage in anticolon antibody–producing B cells, we first established B-cell lines that produced anticolon antibodies through the transformation of B cells by EBV, then calculated the frequencies of anticolon antibody–producing EBV-transformed B cells. In PBLs of the UC patients, estimated frequencies of IgG anticolon antibody–producing B cells were 0.4 – 0.5 per 106 PBLs. In contrast, the number of IgG anticolon antibody–producing cells in LPLs from inflamed colonic mucosa of UC were significantly (P , 0.01) higher (1.5–12.5 per 106 LPLs) than those in PBLs from the same UC patients. VH Gene Expression of Anticolon Antibody– Producing B-Cell Lines Next, mRNA expression of VH genes was assessed in these bulk B-cell lines by reverse-transcription PCR using VH gene family–specific primers. As shown in Figure 1, VH genes expressed in the bulk B-cell lines from normal PBLs and LPLs were diverse. In contrast, the bulk B-cell lines from UC LPLs expressed mainly Figure 1. VH gene expressions of B-cell lines from patients with UC and normal controls. VH genes used in bulk B-cell lines from UC LPLs were mainly VH3, although those in B-cell lines from PBLs were diverse. In contrast, VH genes used in bulk B-cell lines from normal PBLs and LPLs were diverse. 18 INOUE ET AL. Figure 2. VH gene expressions of B-cell lines from UC PBLs that produced anticolon antibodies. VH genes used in bulk B-cell lines that produced anticolon antibodies were mainly VH3, and those in B-cell lines that did not produce anticolon antibodies were diverse. VH3 family, and the VH gene usage in the UC PBLs was diverse, similarly to those from normal PBLs and LPLs. To further clarify the relationship between anticolon antibody production and VH gene usage, bulk B-cell lines that produced anticolon antibodies were compared with those that did not produce anticolon antibodies as previously established from UC PBLs. As shown in Figure 2, VH gene expression in B-cell lines that produced anticolon antibody were primarily VH3, whereas those in B-cell lines that did not produce anticolon antibodies were far more diverse in their VH gene usage. VH Gene Expression of Anticolon Antibody– Producing B-Cell Clones The strength of the VH bands of bulk B-cell lines could not be quantitatively compared with each other because of the nature of this kind of experiment using different primers, even though the amounts of cDNA added in the PCR reaction were equal within a given individual. To evaluate the expression of VH gene usage in each B cell, we established anticolon antibody–producing B-cell clones through limiting dilution culture techniques and assessed mRNA VH gene expression of these B-cell clones by reverse-transcription PCR. As shown in Figure 3, 3 representative B-cell clones from normal PBLs 3A5–3B, 3A5– 4D, and A1–A3 used VH1, VH3, and VH5 genes, respectively. In contrast, VH genes used in 3 representative anticolon antibody–producing B-cell clones from UC were all VH3 family (Figure 3). We could establish 15 anticolon antibody–producing B-cell clones from 7 UC patients, i.e., 2 or 3 clones were GASTROENTEROLOGY Vol. 121, No. 1 Figure 3. VH gene expressions of B-cell clones from patients with UC and normal controls. VH genes used in all 3 representative B-cell clones obtained from UC PBLs and LPLs—IC5, Y-F3, and A5H11— were VH3. In contrast, 3 representative B-cell clones obtained from normal PBLs—3A5-3B, 3A5-4D, and A1-A3— used VH1, VH3, and VH5 genes, respectively. derived from each patient. The analysis of these 15 clones showed that a restricted VH gene, the VH3 family, was used by all anticolon antibody–producing clones from both UC PBLs and UC LPLs. In contrast, diverse VH gene families were used by 24 normal control PBL clones that did not produce anticolon antibodies, i.e., 6 clones used VH1, 1 clone used VH2, 9 clones used VH3, 7 clones used VH4, 1 clone used VH5, and none used VH6. DNA Sequence Analysis of VH Gene These data imply that all anticolon antibody– producing clones from UC express a restricted VH3 family gene usage, giving rise to the question of whether these clones use identical germline VH gene segments and undergo similar somatic mutation. In addition, another question was raised as to whether these anticolon antibodies recognized similar antigenic determinant of the colonic epithelial cell. Therefore, we proceeded to analyze the VH gene segments and nucleotide sequences of VH gene segments of these anticolon antibody–producing B-cell clones. Eight anticolon antibody–producing clones, 1 clone from LPLs (we sequenced 2 clones, A5H10 and A5H11; however, these 2 clones were identical), and 7 clones from PBLs showed similar nucleotide sequences to each other in their framework regions but showed diversity in their complementarity-determining regions. Database searches of the obtained nucleotide sequences indicated that all these clones showed fairly high homology to several germline VH gene segments in ‹ Figure 4. Nucleotide sequence of VH3 family genes used by anticolon antibody–producing B-cell clones and corresponding germline gene. The figure shows the whole sequences between leader sequences and JH segments containing complementarity-determinant region 1 (CDR1) and CDR2. The upper sequence, i.e., A5H10 and A5H11, IC5, SA9, SH8, TH3, TH7, YE5, or YF3, in each cluster shows each anticolon antibody–producing B-cell clone. The lower sequence in each cluster is used for germline comparison. Identities are indicated by dashes. July 2001 VH GENE USAGE IN ULCERATIVE COLITIS 19 20 INOUE ET AL. GASTROENTEROLOGY Vol. 121, No. 1 Table 1. VH, D, and JH Segment of Anticolon Antibody–Producing Clones VH segment Clone Closest germline gene Number of mutations Homology D segment JH segment VH26 VH26 VH26 DP31 VH26 1.9III COS-8 22-2B 14/294 3/294 27/294 12/294 12/294 10/294 9/294 17/294 95.2% 99.0% 90.8% 95.9% 95.9% 96.6% 97.6% 94.2% D1–26 D3–9 D3–10 D5–5 D2–8 D1–26 D3–10 D3–22 JH4b JH4b JH4b JH4b JH4b JH4b JH4b JH2 A5H11a IC5 SA9 SH8 TH3 TH7 YE5 YF3 aA5H10, another clone derived from LPLs, was identical to A5H11. their variable regions (Figure 4 and Table 1). UC LPL clone, A5H11, and 3 UC PBL clones from 3 different patients, IC5, SA9, and TH3, were derived from 1 VH3 germline gene segment, VH26. However, 4 other UC PBL clones established from 3 different patients, SH8, TH7, YE5, and YF3, were derived from DP31, 1.9III, COS-8, and 22-2B, respectively, and were also members of the VH3 family. The numbers of somatic mutations of these clones were 3–27 in VH regions, and the extent of homology to germline gene segments was 90.8%–99.0% (Table 1). The sequences of these clones were not identical or derived from the same VH germline gene, even in clones derived from the same patient (SA9 and SH8, TH3 and TH7, YE5 and YF3). Collectively, all of the anticolon antibody–producing B-cell clones from subjects with UC that we analyzed expressed the VH3 gene family. However, VH genes of each of these clones were derived from different germline VH gene segments in each individual. Discussion It has been previously reported that the production of antibodies against intestinal luminal antigens or autoantigens in the epithelium is increased in the lamina propria of inflamed colonic mucosa.19 In addition, various forms of autoantibodies have been demonstrated in the sera of individuals with UC.2 However, it remains unclear whether these autoantibodies are involved in the disease process. UC shows a familial predisposition, and we and other investigators have shown that particular HLA antigens or genes were dominant in UC subjects.20 –22 More recently, Folwaczny et al.23 showed a higher prevalence of anti– goblet cell autoantibodies in first-degree relatives of individuals with UC. Furthermore, we have demonstrated that anticolon antibodies in the sera from UC subjects can react with surface antigens in colonic epithelial cells or colonic mucin in goblet cells. These antibodies may contribute to the destruction of colonic mucosa through antibody-dependent cell–mediated cytotoxicity mechanisms against colonic epithelial cells,8,10 and subsequent investigation showed that autoantibodies against purified colonic mucin were frequently found in the refractory type of UC.24 In the present study, we established and characterized anticolon antibody–producing B-cell lines and clones from colonic mucosa and peripheral blood of individuals with UC. The importance of this investigation was to begin to characterize the features of colonic antigens recognized by anticolon antibodies in this disease. Initially, we found evidence of an increased number of IgG anticolon antibody–producing cells in inflamed colonic mucosa compared with those in the peripheral blood. In autoimmune processes, it has been shown that B lymphocytes respond to autoantigens with a limited number of VH gene segments. To fully assess the VH gene usage of a specific B-cell population, this population must be expanded in vitro to provide enough mRNA for analysis. We used EBV transformation as a means of establishing anticolon antibody–producing lines and clones. Although it might be possible that EBV transformation biases the repertoire of VH gene usage, Logtenberg et al.25 previously have shown that this process does not change the VH gene repertoire. So far, there has been no reported structural analysis of VH gene segments of anticolon antibody–producing B cells. McCabe et al.26 have reported on the VH gene usage in human intestinal B lymphocytes from subjects with inflammatory bowel disease, and higher frequencies of VH4 have been detected in Crohn’s disease. However, their investigation assessed mRNA of whole cells from biopsy materials and therefore did not analyze the VH gene usage in antibody-producing cells alone. DunnWalters et al.27 have reported the biased JH6 and DXP91 usage of naive B cells in inflammatory bowel diseases. In July 2001 the present study, we demonstrated that VH gene usage in B-cell lines and clones that produced anticolon antibodies were biased toward a restricted VH3 gene family compared with the broad array of VH gene families in normal individuals. VH gene families used in bulk B-cell lines from UC LPLs were mainly VH3, and those in bulk B-cell lines from UC PBLs were more diverse. The strength of the VH bands of bulk B-cell lines cannot be compared quantitatively because of the nature of this kind of experiment using different primers, even though the amounts of cDNA added in the PCR reaction were equal within a given individual. Therefore, we established the B-cell clones and analyzed the VH gene usage of each clone in this study. The restricted VH3 gene family was used by all 15 clones from both UC PBLs and LPLs that produced anticolon antibody. In contrast, diverse VH gene families were used by 24 clones from normal PBLs. Although the order of VH genes in the germline was VH3 (49%) . VH4 (22%) . VH1 (18%) . VH2 (7%) . VH5 (2%) and VH6 (2%),28 the frequency of VH3 gene family usage in healthy individuals is approximately 20%–55% in the peripheral blood B-cell repertoire.29,30 This variation in VH3 family usage can be explained by the large differences in VH gene usage among the different individuals and the method used for analysis of VH gene usage. Our finding that 9 of 24 clones (37.5%) from normal PBLs used the VH3 gene family is consistent with the results of other studies. Thus, it is notable that the VH3 gene family was used by all 15 B-cell clones from UC PBLs and LPLs that produced anticolon antibodies. This finding suggests that an autoimmune process may be involved in the pathogenesis of UC, and it also implies that a unique antigenic stimulus might contribute to the production of anticolon autoantibodies in UC. The characteristics of the corresponding antigens of the autoantibodies associated with anticolon antibodies remain unknown. Analysis of the nucleotide sequences of the VH gene segment compared with those of the germline genes may provide us with information about or characteristic of the antigens. Our results of nucleotide sequences from 8 anticolon antibody–producing clones show that they were derived from various germline gene segments. The number of somatic mutations in our study was 3–27 mutations in the VH region, which is not a high frequency of mutations. Although IgVH genes that have switched to IgG are more mutated than those used by IgM,31 Dunn-Walters et al.27 showed that the numbers of mutations in VH gene from the lamina propria cells of non–inflammatory bowel disease control and UC patients were 4 –37 (mean, 17) and 0 –27 (mean, 12), VH GENE USAGE IN ULCERATIVE COLITIS 21 respectively, and Boursier et al.32 reported that the number of mutations of IgG VH gene was 12–51 (mean, 26). LPL clone A5H11 and 3 PBL clones, IC5, SA9, and TH3, showed the greatest homology with the VH3 germline gene segment use of VH26. The germline gene segment VH26 has been used in monospecific IgM rheumatoid factor33 and polyspecific IgM anti-DNA antibody.34 The clones from PBL, SH8, TH7, YE5, and YF3 showed the greatest homology with other VH3 germline gene segments, DP31, 1.9III, COS-8, and 22-2B, respectively. The DP31 and COS-8 genes were described by Tomlinson et al.35 as members of VH3 germline gene segments. The DP31 gene was then reported to express in anti-SS-A/Ro autoantibodies in Sjögren syndrome36; the relationship of COS-8 gene to the production of autoantibodies has not been reported. The 1.9III gene is expressed in IgM anti-dsDNA antibody37 and specific antibodies induced by vaccination with Haemophilus influenzae type b capsular polysaccharide.38 The 22-2B gene is commonly expressed in the human B-cell repertoire to encode various specificities of antibodies such as monospecific IgM rheumatoid factor,14 IgM insulin-induced anti-insulin antibody,14 and IgA high-affinity anti–rabies virus antibody.39 Recently, Claeys et al.40 showed that an autoimmune reaction against gastric H1,K1-ATPase, which is closely related to classical autoimmune gastritis, occurs in a substantial group of Helicobacter pylori–infected patients. Enteric bacteria are increasingly believed to play important roles in triggering the inflammation associated with UC; therefore, it is possible that mechanisms similar to those in H. pylori–associated gastritis may be involved in the pathogenesis of UC. Although VH26, DP31, 1.9III, COS-8, and 22-2B gene segments were all in the VH3 family, they are not closely related to each other. Moreover, in view of our data on the homology compared with germline gene segments, we cannot exclude that anticolon antibody– producing clones use novel VH genes that are yet to be characterized. Furthermore, the fact that VH genes of anticolon antibody–producing B-cell clones were derived from these various germline gene segments suggests that the colonic antigens recognized by autoantibodies in UC may be heterogeneous or that these autoantibodies recognize the different epitopes of the colonic antigens among the study population. Also, the number of clones examined thus far may not be sufficient to provide a complete analysis of all the VH gene usage. More precise analysis of VH genes may elucidate the nature of the autoantigen(s) involved in UC. 22 INOUE ET AL. References 1. Brandtzaeg P. Autoimmunity and ulcerative colitis: can two enigmas make sense together? Gastroenterology 1995;109:307– 312. 2. Fiocchi C. Inflammatory bowel disease: etiology and pathogenesis. Gastroenterology 1998;115:182–205. 3. Okazaki K, Morita M, Nishimori I, Sano S, Toyonaga M, Nakazawa Y, Yamamoto Y, Yamamoto Y. Major histocompatibility antigenrestricted cytotoxicity in inflammatory bowel disease. Gastroenterology 1993;104:384 –391. 4. Yonamine Y, Watanabe M, Kinjo F, Hibi T. Generation of MHC class I-restricted cytotoxic T cell lines and clones against colonic epithelial cells from ulcerative colitis. J Clin Immunol 1999;19: 77– 85. 5. Fiocchi C, Roche JK, Michener WM. High prevalence of antibodies to intestinal epithelial antigens in patients with inflammatory bowel disease and their relatives. Ann Intern Med 1989;110: 786 –794. 6. Duerr RH, Targan SR, Landers CJ, Sutherland LR, Shanahan F. Antineutrophil cytoplasmic antibodies in ulcerative colitis: comparison with other colitis/diarrheal illness. Gastroenterology 1991;100:1590 –1596. 7. Geng X, Biancone L, Dai HH, Lin JJ, Yoshizaki N, Dasgupta A, Pallone F, Das KM. Tropomyosin isoforms in intestinal mucosa: production of autoantibodies to tropomyosin isoforms in ulcerative colitis. Gastroenterology 1998;114:912–922. 8. Hibi T, Aiso S, Yoshida T, Watanabe M, Asakura H, Tsuru S, Tsuchiya M. Anticolon antibody and lymphocytophilic antibody in ulcerative colitis. Clin Exp Immunol 1982;49:75– 80. 9. Hibi T, Ohara M, Toda K, Hara A, Ogata H, Iwao Y, Watanabe N, Watanabe M, Hamada Y, Kobayashi K, Aiso S, Tsuchiya M. In vitro anticolon antibody production by mucosal or peripheral blood lymphocytes from patients with ulcerative colitis. Gut 1990;31:1371–1376. 10. Hibi T, Ohara M, Watanabe M, Kanai T, Takaishi H, Hayashi A, Hosoda Y, Ogata H, Iwao Y, Aiso S, Watanabe N, Toda K, Tsuchiya M. Interleukin 2 and interferon-g augment anticolon antibody dependent cellular cytotoxicity in ulcerative colitis. Gut 1993;34:788 –793. 11. Hibi T, Ohara M, Kobayashi K, Brown WR, Toda K, Takaishi H, Hosoda Y, Hayashi A, Iwao Y, Watanabe M, Aiso S, Kawai Y, Tsuchiya M. Enzyme linked immunosorbent assay (ELISA) and immunoprecipitation studies on anti-goblet cell antibody using a mucin producing cell line in patients with inflammatory bowel disease. Gut 1994;35:224 –230. 12. Tonegawa S. Somatic generation of antibody diversity. Nature 1983;302:575–581. 13. Kabat EA, Wu TT, Reid-Miller M, Perry HM, Gottesman KS. Sequence of proteins of immunological interest (4th ed). Washington, DC: Department of Health and Human Services, 1987. 14. Pascual V, Capra JD. VH4-21, a human VH gene segment overrepresented in the autoimmune repertoire. Arthritis Rheum 1992;35:11–18. 15. Watanabe M, Ueno Y, Yajima T, Iwao Y, Tsuchiya M, Ishikawa H, Aiso S, Hibi T, Ishii H. Interleukin 7 is produced by human intestinal epithelial cells and regulates the proliferation of intestinal mucosal lymphocytes. J Clin Invest 1995;95:2945– 2953. 16. Nakazawa A, Watanabe M, Kanai T, Yajima T, Yamazaki M, Ogata H, Ishii H, Azuma M, Hibi T. Functional expression of a costimulatory molecule, CD86, on epithelial cells in the inflamed colonic mucosa. Gastroenterology 1999;117:536 –545. 17. Campbell MJ, Zelenetz AD, Levy S, Levy R. Use of family specific leader region primers for PCR amplification of the human heavy chain variable region repertoire. Mol Immunol 1992;29:193– 203. GASTROENTEROLOGY Vol. 121, No. 1 18. Altschul SF, Madden TL, SchSffer AA, Zhang J, Zhang Z, Miller W, Lipman DJ. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res 1997;25: 3389 –3402. 19. MacDermott RP, Nash GS, Bertovich MJ, Seiden MV, Bragdon MJ, Beale MG. Alterations of IgM, IgG, and IgA synthesis and secretion by peripheral blood and intestinal mononuclear cells from patients with ulcerative colitis and Crohn’s disease. Gastroenterology 1981;81:844 – 852. 20. Asakura H, Tsuchiya M, Aiso S, Watanabe M, Kobayashi K, Hibi T, Ando K, Takata H, Sekiguchi S. Association of the human lymphocyte-DR2 antigen with Japanese ulcerative colitis. Gastroenterology 1982;82:413– 418. 21. Roussomoustakaki M, Satsangi J, Welsh K, Louis E, Fanning G, Targan S, Landers C, Jewell DP. Genetic markers may predict disease behavior in patients with ulcerative colitis. Gastroenterology 1997;112:1845–1853. 22. Satsangi J, Parkers M, Louis E, Hashimoto L, Kato N, Welsh K, Terwillinger JD, Lathrop GM, Bell JI, Jewell DP. Two stage genome-wide search in inflammatory bowel disease provides evidence for susceptibility loci on chromosomes 3, 7 and 12. Nat Genet 1993;14:199 –202. 23. Folwaczny C, Noehl N, Tschöp K, Endres SP, Heldwein W, Loeschke K, Fricke H. Goblet cell autoantibodies in patients with inflammatory bowel disease and their first-degree relatives. Gastroenterology 1997;113:101–106. 24. Takaishi H, Ohara S, Hotta K, Yajima T, Kanai T, Inoue N, Iwao Y, Watanabe M, Ishii H, Hibi T. Circulating autoantibodies against purified colonic mucin in ulcerative colitis. J Gastroenterol 2000; 35:20 –27. 25. Logtenberg T, Young FM, van Es J, Meyling FHJG, Berman JE, Alt FW. Frequency of VH-gene utilization in human EBV-transformed B-cell lines: the most JH-proximal VH segment encodes autoantibodies. J Autoimmun 1989;2:203–213. 26. McCabe RP, Carroll WL, Egan M, Cohn SM, Peters M. Immunoglobulin variable region usage in human intestinal B lymphocytes. Clin Immunol Immunopathol 1994;71:240 –245. 27. Dunn-Walters DK, Boursier L, Hackett M, Spencer J. Biased JH usage in plasma cell immunoglobulin gene sequences from colonic mucosa in ulcerative colitis but not in Crohn’s disease. Gut 1999;44:382–386. 28. Cook GP, Tomlinson IM, Walter G, Riethman H, Carter NP, Buluwela L, Winter G, Rabbitts TH. A map of the human immunoglobulin VH locus completed by analysis of the telomeric region of chromosome 14q. Nat Genet 1994;7:162–168. 29. Davidkova G, Pettersson S, Holmberg D, Lundkvist I. Selective usage of VH genes in adult human B lymphocyte repertoires. Scand J Immunol 1997;45:62–73. 30. Brezinschek H-P, Dörner T, Monson NL, Brezinschek RI, Lipsky PE. The influence of CD40-CD154 interactions on the expressed human VH repertoire: analysis of VH genes expressed by individual B cells of a patient with X-linked hyper-IgM syndrome. Int Immunol 2000;12:767–775. 31. Fischer M, Kuppers M. Human IgA- and IgM-secreting intestinal plasma cells carry heavily mutated VH region genes. Eur J Immunol 1998;28:2971–2977. 32. Boursier L, Dunn-Walters DK, Spencer J. Characteristics of IgVH genes used by human intestinal plasma cells from childhood. Immunology 1999;97:558 –564. 33. Pascal V, Randen I, Thompson K, Sioud M, Forre O, Natvig J, Capra JD. The complete nucleotide sequence of the heavy chain variable regions of six monospecific rheumatoid factors derived from Epstein-Barr virus-transformed B cells isolated from the synovial tissue of patients with rheumatoid arthritis. J Clin Invest 1990;86:1320 –1328. 34. Dersimonian H, Schwartz RS, Barrett KJ, Stollar BD. Relationship of human variable region heavy chain germ line gene to genes July 2001 35. 36. 37. 38. 39. encoding anti-DNA autoantibodies. J Immunol 1987;139:2496 – 2501. Tomlinson IM, Walter G, Marks JD, Llewelyn MB, Winter G. The repertoire of human germline VH sequences reveals about fifty groups of VH segments with different hypervariable loops. J Mol Biol 1992;227:776 –798. Suzuki H, Takemura H, Suzuki M, Sekine Y, Kashiwagi H. Molecular cloning of anti-SS-A/Ro 60-kDa peptide Fab fragments from infiltrating salivary gland lymphocytes of a patient with Sjögren’s syndrome. Biochem Biophys Res Commun 1997;232:101–106. Cairns EP, Kwong C, Misener V, Ip P, Bell DA, Siminovitch KA. Analysis of variable region genes encoding a human antiDNA antibody of normal origin. J Immunol 1989;143:685– 691. Adderson EE, Shackelford PG, Quinn A, Carrol WL. Restricted Ig H chain V gene usage in the human antibody response to Haemophilus influenzae type b capsular polysaccharide. J Immunol 1991;147:1667–1674. Ikemasu H, Ueki Y, Harindranah N, Notkins AL, Casali P. Clonal analysis of a human antibody response. II. Sequences of the VH VH GENE USAGE IN ULCERATIVE COLITIS 23 genes of human IgM, IgG and IgA to rabies virus reveal preferential utilization of VHIII segments and somatic hypermutation. J Immunol 1993;150:1325–1337. 40. Claeys D, Faller G, Appelmelk BJ, Negrini R, Kirchiner T. The gastric H1,K1-ATPase is a major autoantigen in chronic Helicobacter pylori gastritis with body mucosa atrophy. Gastroenterology 1998;115:340 –347. Received October 12, 2000. Accepted March 14, 2001. Address requests for reprints to: Toshifumi Hibi, M.D., Keio Cancer Center, 35 Shinanomachi, Shinjuku-ku, Tokyo 160-8582, Japan. e-mail: thibi@med.keio.ac.jp; fax: (81) 3-3357-6156. Supported in part by grants-in-aid from the Japanese Ministry of Education, Culture and Science, the Japanese Ministry of Health and Welfare, Keio University Medical Science Fund, Tokyo, Japan. The authors thank Professor H.-Michael Dosch for his valuable comments; Drs. Noriaki Watanabe and Makoto Ohara for their helpful discussion; and Drs. Yasuo Hosoda, Atsushi Hayashi, Yoshitaka Ueno, Hiromasa Takaishi, and Kyoko Toda for their technical assistance.