T cell receptor V-segment frequencies in peripheral blood T cells correlate with human leukocyte antigen type.

We compared T cell receptor (TCR) V-segment frequencies in human leukocyte antigen (HLA) identical siblings to sibling pairs who differ at one or both HLA haplotypes using four V beta-specific and one V alpha-specific monoclonal antibody. In every one of nine families HLA-identical sibs had the most similar patterns of V-segment frequencies in their peripheral blood, whereas totally mismatched sibs were, in general, the most dissimilar; HLA haploidentical sibs tended to be intermediate between the two groups. The degree of similarity among HLA-identical sibs was comparable to that observed among three pairs of identical twins suggesting that HLA is the major genetic component influencing TCR V-segment frequency. Consistent with this observation, it was found that the frequency of T cells expressing particular V beta segments was skewed towards either CD4+ or CD8+ cells indicating that T cells expressing some V beta genes may be positively selected primarily by class I or class II major histocompatibility complex proteins. Finally, it was observed that individuals who express the HLA class I specificity, B38, tend to express high levels of V alpha 2.3+ cells among their CD8+ T cells. These observations represent definitive proof that human V-segment frequencies are profoundly influenced by the HLA complex.

Selective expansion of specific T cell receptors in the inflamed colon of Crohn's disease.

To identify disease-specific T cell changes that occur in Crohn's disease (CD), the T cell receptor BV repertoires of lamina propria lymphocytes (LPL) isolated from both the inflamed and "disease-inactive" colons of seven CD patients were compared by the quantitative PCR and DNA sequence analysis. It was observed that the BV repertoires of LPL isolated from the disease-active and disease-inactive parts of the colon from the same individual were very different. Furthermore, nearly all of the differences occurred in CD4+ LPL, with very few differences in the CD8+ population of LPL. Although the pattern of BV segments that was increased in disease-active tissue relative to disease-inactive tissue was different for all seven CD patients, there were several BV segments that increased uniformly in the disease-active tissue of all seven individuals. CDR3 length analysis and DNA sequencing of these BV segments revealed that in six of the seven CD patients there was a striking degree of oligoclonality that was absent from disease-inactive tissue of the same individual. These observations suggest that at least some of the inflammation in CD is the result of responses by CD4+ T cells to specific antigens. The isolation of such inflammation-specific CD4+ T cells may make it possible to identify the antigens that are responsible for the inflammatory process in CD and provide a better understanding of its pathogenesis.

CD4+ cell oligoclonality in Crohn's disease: evidence for an antigen-specific response.

To identify disease-specific T cell changes that occur in Crohn's disease (CD) the T-cell receptor (TCR) BV repertoires of lamina propria lymphocytes (LPL) from both disease-active and disease-inactive colonic tissue of three CD patients were compared by a quantitative polymerase chain reaction (qPCR) and CDR3 length analysis. It was observed that the BV repertoires of LPL isolated from the disease-active and disease-inactive parts of the colon of the same individual were different, and most of the differences occurred in CD4+ LPL with very few differences in the CD8+ populations of LPL. Although the pattern of BV segments that was increased in disease-active relative to disease-inactive tissue was different for all three CD patients, there was an increase in the levels of BV11, 13S2, 15, 16, and 17 segments in the disease-active tissue of all three patients. Standard CDR3 length analysis of BV11, 13S2, 15, 16, and 17 segments revealed that in two of the three CD patients there was a striking degree of TCR oligoclonality in the disease-active tissue that was absent from disease-inactive tissue of the same individual. Additional differences between the disease-active and disease-inactive tissues were observed using a more refined method of CDR3 length analysis, which employs BV- and BJ-specific primers. These observations suggest that at least some of the inflammation in CD is the result of responses by CD4+ T cells to specific antigens.

HLA class II alleles associated with susceptibility and resistance to Crohn's disease in the Jewish population.

Previous studies have suggested that susceptibility to Crohn's disease (CD) is associated with the histocompatibility complex (HLA) class II alleles DR1, DQ5, and DR13 in the Caucasian population, DR7 in the French and German populations, and DR4 and DQ4 in the Japanese population. However, little is known about the relationship between HLA class II alleles and CD in the Jewish population since these previous studies included few Jewish individuals. In order to determine whether the HLA associations observed with predominantly non-Jewish populations were also present in the Jewish CD population and whether there were any HLA class II alleles uniquely associated with CD in the Jewish population, 132 CD patients, of which 82 were Ashkenazi Jewish, were HLA-typed using serologic and DNA methods. Ethnically matched controls were similarly typed. No association with DR1 or DR13 was observed in the Jewish CD population although an association with DR13 (OR [odds ratio] = 5.3, p = 0.02) was observed in the non-Jewish CD population. However, an association with DR15 (OR = 2.7, p = 0.03), which is normally associated with ulcerative colitis, was observed in the Jewish, but not non-Jewish, CD group. In addition, a strong negative association was observed with DR3, which was especially striking in the Jewish population (OR = 0.35, p = 0.025); similar negative associations with DR3 have been observed by others using non-Jewish populations. Furthermore, a significant negative association with DR7 (OR = 0.45, p = 0.04) was observed in the Jewish, but not non-Jewish, population. Consistent with this was the negative association with DQ2 (OR = 0.38, p = 0.005), which is in strong linkage disequilibrium with both DR3 and DR7, in the Jewish, but not non-Jewish, population. These studies support previous suggestions that susceptibility to CD in Jewish and non-Jewish populations is determined by distinct genes and provide further support to the hypothesis that a gene on the DR3 haplotype may protect against CD. Furthermore, protection is conferred by the same or another gene found on Jewish, but not non-Jewish, DR7 haplotypes.

Marked gamma delta T-cell decrease in peripheral blood of patients with primary biliary cirrhosis (PBC).

PBC is a cholestatic liver disease of unknown etiology with autoimmune features that is often associated with other autoimmune diseases. We analyzed peripheral blood T-cell subsets in patients groups with PBC (n = 11), non-PBC hepatobiliary disease (n = 11) and an age and sex matched control group (n = 11) by two color FACS-analysis. Seven out of eleven PBC patients exhibited markedly lowered and nearly undetectable levels of gamma delta T-cells (< 0.8%). None of the individuals in the non-PBC hepatobiliary disease (HBD) group or the normal control group had gamma delta values below 1%. The other four individuals in the PBC group had gamma delta values within the normal range. Overall, the PBC group had a statistically significant, lowered mean percentage of gamma delta T-cells (1.50%) as compared to the hepatobiliary disease group (3.76%) and the control group (4.22%, p = 0.01). The percentages of CD4+ and CD8+ and alpha beta TCR+ CD4-CD8- double negative cells in PBC patients did not differ from the control group. PBC patients with normal gamma delta cell counts did not differ from the PBC group with low gamma delta values in autoantibody titers, liver tests or treatment of the disease. As a possible cause for the observed decrease of gamma delta T-cells three sera of PBC patients with low gamma delta T-cell counts were screened by single color, indirect immunofluorescence for antibodies to gamma delta T-cell enriched lymphocytes, but no differences to control sera were observed.(ABSTRACT TRUNCATED AT 250 WORDS)

Analysis of the peripheral blood T-cell receptor (TCR) repertoire in monozygotic twins discordant for Crohn's disease.

T cell involvement in the inflammatory process of Crohn's Disease (CD) is evident by an increase in activated T cells and their cytokines in actively inflamed CD tissue. It has been suggested that CD may involve a superantigen based on the observation that a significant proportion of CD patients express elevated levels of V beta 8+ T cells in their peripheral blood compared to normal controls. In order to determine whether a superantigen might play a role in the pathogenesis of CD we have compared the TCR repertoires of four pairs of monozygotic twins discordant for CD. By using monozygotic twins, we could rule out the effects of HLA and other genes on the TCR repertoire. The TCR repertoires were analyzed by using a panel of V-segment-specific mAb and by quantitative polymerase chain reaction (qPCR) using V beta-specific oligonucleotide primers. In all cases the TCR repertoires of the affected and unaffected sibs were strikingly similar. We did not observe any TCR segment that was consistently altered in frequency or expression levels in all of the affected sibs compared to their identical twin. Furthermore, we did not see an increase in V beta 8+T cells in the peripheral blood of the CD sibs relative to their normal counterpart. These studies suggest that the presence of CD does not alter the TCR repertoire of peripheral blood in any obvious way and argue against the role of a superantigen in the etiology of pathogenesis of CD.

Evidence for an altered T-cell receptor repertoire in Crohn's disease.

We have compared the frequencies of T cells expressing each of four different T cell receptor (TCR) V beta segments in lamina propria and peripheral blood lymphocytes of 12 Crohn's disease (CD), six ulcerative colitis (UC), and 10 control patients in an attempt to identify disease-specific changes. The frequencies of CD4+ and CD8+ cells reacting with each of four fluoresceinated TCR-specific monoclonal antibodies directed against V beta 5, V beta 6.7a, V beta 8, and V beta 12 were determined by flow cytometry. There was no difference among the groups in the average frequency of any single V beta segment in either the CD4+ or CD8+ subpopulations. However, when the sum of the differences in V beta frequencies (delta score) between peripheral blood lymphocytes (PBL) and lamina propria lymphocytes (LPL) were determined for each individual, significant differences were observed between the CD4+ and CD8+ populations and among the patient groups. In all three patient groups, there were significant individual differences between LPL and PBL in the frequencies of CD8+ and CD4+ cells reacting with the four V beta-specific mAb. In Controls and UC, this difference was, on average, two-fold greater in CD8+ cells than in CD4+. In CD, however, this difference was, on average, the same for CD8+ and CD4+ cells. These observations suggest that (1) the human colonic LPL TCR repertoire is normally different from that of PBL, especially in the CD8+ population and (2) there is an alteration in the LPL TCR repertoire in CD which is not observed in Controls or UC.

Differential patterns of T-cell receptor BV-specific activation of T cells by gp120 from different HIV strains.

Studies by several groups have suggested that HIV infection in vivo results in a BV-specific alteration of the TCR repertoire and that this might play a role in the pathogenesis of AIDS. Our earlier studies demonstrated that both a crude extract of HIV451 as well as purified gp 160 from HIV451 could specifically activate, in vitro, T cells expressing a common set of TCRBV segments (TCRBV3, 12, 14, 15, and sometimes BV17 and 20) in individuals of disparate HLA type. Furthermore, purified gp120 from HIV451 was shown to have a similar ability to activate T cells, although with a slightly different TCRBV-specific pattern. In order to determine whether gp120 from other HIV strains could similarly activate T cells in a TCRBV-specific pattern, PBMC from HIV seronegative individuals of disparate HLA type were stimulated with gp120 from three strains of HIV (451, IIIB, and MN). The authors found that gp120 from all three strains activate T cells bearing TCRBV2 and BV3 in nearly every individual. T cells expressing other BV segments are also activated, but this is more variable and appears to be unique to each individual. Furthermore, gp120(451) and gp120 from HIVIIIB and HIVMN differ in their ability to activate T cells expressing these other TCRBV segments. These observations suggest that variation in the structure of gp120 and in the genetic and/or environmental background of the individual play an important role in determining which TCRBV segments are 'triggered' by gp120. Furthermore, these observations may have important implications for the rate of disease progression in HIV-infected individuals.

The influence of non-HLA genes on the human T-cell receptor repertoire.

We previously demonstrated a central role for HLA genes in determining the T-cell receptor (TCR) repertoire. However, these studies also suggested that other genetic factors might also play a role in the development of this repertoire. In order to assess the role of non-HLA genes in the development of the TCR repertoire, we have analysed and compared the TCR repertoires of individuals in three families consisting of both monozygotic twins as well as an HLA-identical sib. TCR repertoire analysis was performed with both V-segment-specific MoAb and the polymerase chain reaction using TCRBV-segment-specific oligonucleotide primers. We observed that in every case the TCR repertoires of identical twins were more similar to each other than to their HLA-identical sib. Furthermore, in one family we were able to show by genotype analysis that most of the differences in repertoire between the identical twins and their HLA-identical sib were caused by polymorphisms in the TCR genes that influence expression levels. These studies document an important role for non-HLA genes in determining the TCR repertoire in man and raise the possibility that such TCR polymorphisms may play a significant role in determining disease susceptibility.

Influence of HLA genes on T cell receptor V segment frequencies and expression levels in peripheral blood lymphocytes.

The effect of the HLA complex on the TCR repertoire in human peripheral blood was assessed by using nine V beta- and V alpha-specific mAb and the quantitative polymerase chain reaction specific for 22 V beta segments. Studies in randomly selected and unrelated individuals failed to show any influence of the HLA complex on the TCR repertoire. In contrast, studies in large families with multiple siblings showed a strong influence on the TCR repertoire by the HLA complex. In pairwise comparisons, HLA-identical sibs had more similar patterns of V segment frequencies, as measured with the nine V segment-specific mAb, as well as more similar expression levels of V beta-specific RNA, as measured by quantitative polymerase chain reaction, than totally mismatched or haplo-identical sibs. When the amount of V beta-specific RNA expressed in CD4+ and CD8+ T cells was compared, it was found that V beta 2, 5.1, 9, and 20 were skewed toward CD4+ T cells; on the other hand, V beta 7 and 14 showed a bias in expression for CD8+ T cells, suggesting that the former were positively selected predominantly by HLA class II gene products whereas the latter V beta segments were positively selected predominantly by HLA class I gene products. These studies unequivocally document the effects of HLA genes on TCR V segment frequencies and expression levels in peripheral blood T lymphocytes.

Unique phenotype and distinct TCR V beta repertoire in human peripheral blood alpha beta TCR+, CD4-, and CD8- double negative T cells.

TCR-alpha beta+ CD4- CD8- double-negative (DN) T cells represent a small, poorly defined T cell subset in human peripheral blood that has been postulated to be potentially autoreactive. To define some of the characteristics of this subset of T cells, DN cells and CD4+ and CD8+ single-positive (SP) cells were purified from the peripheral blood of six unrelated individuals by a combination of positive selection and depletion using mAb conjugated to immunomagnetic beads. Purified DN cells were found to be enriched in cells expressing HLA-DR and the NK cell marker, CD56, when compared to the SP population. Furthermore, in contrast to SP cells that express the adhesion marker CD44, DN cells were found to express very little, if any, CD44. When the V beta TCR repertoires of DN and SP (CD4+ and CD8+) cells, determined by quantitative (q) PCR, were compared all three populations were found to be considerably different. Furthermore, several V beta segments (V beta 11 and V beta 19) were consistently expressed at higher levels on DN cells than on SP cells. The TCR repertoires of both DN and SP cells were frequently characterized by dominance of one or more V beta segments that could in some instances be shown to be restricted to the CD45RO+ ("memory") population. However, differences in TCR repertoire between DN and SP cells were observed even when CD45RO+ cells were removed before qPCR analysis. These studies suggest that the TCR repertoires of DN and SP cells are determined by different selection mechanisms and that DN and SP cells are directed against different Ag.

Comparisons of T cell receptor (TCR) V beta repertoires of lamina propria and peripheral blood lymphocytes with respect to frequency and oligoclonality.

The TCR repertoires of CD4+ and CD8+ lamina propria and peripheral blood lymphocytes (PBL) were compared with respect to V beta frequencies and oligoclonality. Disease-free colon specimens and paired peripheral blood samples were obtained from eight adult patients undergoing surgical resections for colorectal carcinoma. Mononuclear cell were isolated from the lamina propria and peripheral blood and separated into CD4+ and CD8+ cells by immunomagnetic adsorbtion. Analysis of V beta frequencies by qPCR revealed that PBL and lamina propria lymphocytes (LPL) from the same individual have very different repertoires, especially within the CD8+ population. Furthermore, CD8+, but not CD4+, LPL display extensive oligoclonality, similar to that which has previously been reported for CD8+ PBL. However, the oligoclonal receptors observed in CD8+ LPL are, in general, distinct from those observed in CD8+ PBL, and differ for each individual. These observations indicate that the TCR repertoires of LPL are as diverse as PBL, and suggest that LPL and PBL are normally exposed to different sets of antigens. In addition, these observations provide a baseline for examining the effects of various disease states and environmental insults on the LPL repertoire.

Crohn's disease is accompanied by changes in the CD4+, but not CD8+, T cell receptor BV repertoire of lamina propria lymphocytes.

To identify disease-specific T cell changes that occur in Crohn's disease (CD), the T cell receptor (TCR) BV repertoires of lamina propria lymphocytes (LPL) isolated from the diseased colon of seven CD patients and eight controls were determined by semiquantitative polymerase chain reaction (qPCR). As an internal control for the effects of HLA and other genes on the TCR repertoire, the BV repertoires of peripheral blood lymphocytes (PBL) from the same individuals were similarly determined and used for comparison. It was observed that the BV repertoires of LPL and PBL within the same individual were very different in both the CD and control groups. However, the CD4+, but not CD8+, repertoires of LPL and PBL differed to a much greater extent in the CD group than in the control group. Furthermore, in each CD patient there was a unique pattern of BV segments which were increased in the CD4+ LPL repertoire relative to that in PBL. These observations suggest that the inflammatory process in CD involves responses by specific CD4+ T cells to specific antigens. The isolation of such inflammation-specific CD4+ T cells may make it possible to identify the antigens which are responsible for the inflammatory process in CD and provide a better understanding of its pathogenesis.

Clonal predominance of T cell receptors within the CD8+ CD45RO+ subset in normal human subjects.

Structural models for the TCR alpha/beta predict that the CDR1, CDR2, and CDR3 loops of both the alpha- and beta-chains contribute to specific interactions with the Ag/MHC complex. The CDR3 loops are constructed by joining events involving the V-(D)-J segments, and thus may vary in both sequence and length. We have developed a polymerase chain reaction assay to assess the length variation of the CDR3 loop in TCR derived from seven V beta segment families (V beta 2, V beta 3, V beta 4, V beta 9, V beta 14, V beta 16, and V beta 17). Peripheral blood T cells from 10 normal adults as well as five cord blood samples were studied. CD4+ and CD8+ T cells were analyzed separately. We observed extreme predominance of particular CDR3 lengths in half of the normal adults. These TCR were shown to be clonal by direct sequence analysis. This clonal dominance was found in the CD8+, CD45RO+ T cell population, and was observed in various V segment families. These patterns of TCR clonality were persistent over many months of observation in some individuals. In one subject, the new appearance of a predominant clone was associated with a booster vaccination for hepatitis B. These studies reveal a surprising degree of oligoclonality in the CD8+ cells of normal subjects which may be due to both environmental and genetic factors; the functional significance of persistent clonal dominance in the CD8 compartment remains to be determined.

Do HLA genes play a prominent role in determining T cell receptor V alpha segment usage in humans?

Previous studies in humans have demonstrated that HLA genes can profoundly influence the TCR V beta repertoire. To similarly assess the influence of HLA genes on the TCR V alpha segment repertoire, the V alpha repertoires of 12 individuals from three unrelated families were determined by quantitative PCR. Each family contained at least one pair of HLA-identical and -nonidentical siblings. Repertoire analysis was performed on purified CD4+ and CD8+ cells by using V alpha-specific primers. We were unable to demonstrate more similar V alpha repertoires between HLA-identical siblings than between HLA-nonidentical siblings. In contrast, when a similar analysis was performed on the same individuals for the V beta repertoire, HLA-identical siblings were found to have significantly more similar repertoires than HLA-nonidentical siblings. Furthermore, both the V alpha and V beta repertoires of monozygotic twins showed striking similarity. Despite our inability to show an influence of HLA genes on the V alpha repertoire, we did observe a very strong skewing in terms of preferential expression on CD4+ or CD8+ cells of several V alpha segments, notably TCRAV1, -2, -5, -6, -7, -11, -12, and -13. These studies suggest that HLA genes play less of a role in determining V alpha segment usage than V beta. Nevertheless, the pronounced skewing of V alpha segment expression in the CD4+ or CD8+ populations suggests some role for HLA genes in determining the V alpha TCR repertoire. Furthermore, the striking similarity of V alpha repertoires of identical twins suggests a major role for non-HLA genes in determining the V alpha repertoire.

Differences in risk of Crohn's disease in offspring of mothers and fathers with inflammatory bowel disease.

OBJECTIVE: To determine whether there are any unusual patterns of transmission of susceptibility to inflammatory bowel disease (IBD) within multiplex families. METHODS: Individuals with IBD were recruited for genome-wide screening of susceptibility genes. The extent of familial aggregation and blood relationships in multiplex families were determined by questionnaires given to participants followed up by confirmation of disease diagnosis by participants' physicians. RESULTS: Of 135 families identified in which both a parent and a child had IBD, 93 involved transmission of susceptibility to disease from mother to child versus 42 examples of transmission from father to child (p = 0.00001, exact two-tailed binomial test). This distortion in transmission on the basis of the sex of the parent was observed only among non-Jewish pairs with Crohn's disease (CD), in which, of 33 parent-child pairs with CD, disease susceptibility was transmitted from the mother 28 times (p = 0.00007). CONCLUSION: Susceptibility to CD in a subset of patients may involve a gene that is imprinted.