Expression of T cell receptor V gamma 5 in the adult thymus of irradiated mice after transplantation with fetal liver cells

T cell receptor (TcR) gamma/delta displays limited diversity and its diversity is distinct in different stages of ontogeny and in different anatomical sites. The V gamma 5 and V delta 1 gene products are preferentially expressed on the early fetal thymocytes and on Thy-1+ dendritic epidermal cells, whereas the V gamma 4 and V delta 5 gene products are abundantly expressed on the adult thymocytes. To elucidate whether the developmentally ordered appearance of thymocytes expressing TcR gamma/delta is dependent on the source of T cell precursors or is controlled by the thymic environment where T cells develop, we compared the expression of V gamma 5 on the early-appearing thymocytes between irradiated mice after transplantation with fetal liver (FL) cells and those after transplantation with bone marrow (BM) cells. Sequential appearance of thymocyte subpopulations was observed in the thymus of radiation FL chimeras similar to that seen in radiation BM chimeras. A substantial number of thymocytes bearing V gamma 5 appeared in the thymus at the early stage of radiation FL chimeras, whereas few, if any, of such V gamma 5-bearing thymocytes were detected in the thymus at any stage of radiation BM chimeras. These results suggested that the ordered expression of V gamma repertoire may depend on the origin of the T cell precursors but not on the thymic environment.     

Environmental and allele-specific influences on T cell receptor gene rearrangement: skewed alpha, delta, and gamma gene rearrangement patterns in chimeric mice.

This study was performed as a means to examine the effects of environment and stem cell origin on the molding of T cell receptor gene rearrangements during T cell development. Lethally irradiated adult mice were first reconstituted with (C57BL/6 x DBA/2)F1 stem cells, which derived either from a fetal liver or adult bone marrow source. The gene rearrangements in the irradiated thymi were then compared to those in normal fetal or adult thymi by hybridoma analysis. Results showed a general absence of gene rearrangements typical of the normal fetal thymus in both sets of chimeras. alpha and gamma gene rearrangements in chimeric mice matched normal adult patterns. However, chimeric thymocytes were unique in an unusually low frequency of rearrangements at the delta locus. The use of the bone marrow vs. fetal liver as stem cell sources in chimeric mice did not affect the patterns, indicating that environmental factors played a major role in the molding of gene rearrangement at multiple T cell receptor loci. Interestingly, the chromosomal sequence added an additional influence as homologous chromosomes showed allele-specific rearrangement patterns. We now question whether the unique patterns of rearrangement in irradiated adult thymi may substantially affect developing T cell populations in chimeric animals, particularly with regard to the gamma/delta T cell subset. Further analyses are warranted, both in experimental and clinical settings, regarding ultimate potentials for T cell diversity and function in transplantation recipients.     

Developmentally regulated fetal thymic and extrathymic T-cell receptor gamma delta gene expression

The gamma delta T-cell receptor (TCR) is the first TCR to be expressed in ontogeny in all vertebrates in which it has been examined thoroughly. Murine gamma delta cell-surface protein is detected by the fourteenth day of gestation. In this work, the activation of gamma delta RNA has been studied. Data indicate that the first TCR protein to appear in the thymus is encoded by gamma genes that are activated after cells colonize the thymus. However, the sequential appearance of different gamma delta TCR proteins during thymic ontogeny cannot be readily explained by differential temporal activation of V gamma genes in the thymus. There are distinct patterns of gamma and delta gene expression during fetal liver development and in the fetal gut (or tissue associated with it). Cells apparent in the liver of mice at birth express gamma delta cell-surface protein, but they disappear from the liver very soon afterward. One V gamma gene is rearranged and expressed prethymically. In addition, gamma gene expression is detectable in the livers of newborn athymic mice. Together, these observations indicate a thymic-independent pathway of activation of TCR genes.     

Rearrangements of T-cell antigen receptor gamma and delta chain genes are detected in the long-term cultured bone marrow cells of athymic nude mice but not in those of euthymic mice.

We have previously shown that extrathymic rearrangements of T-cell receptor (TcR) gamma and delta chain genes occur in the peripheral lymphoid tissues of athymic nude mice. To further determine where the TcR gene rearrangements occur in nude mice, we investigated the rearrangement and expression of the TcR genes in the long-term cultured bone marrow (LTBM) cells which were homogenous in developments without mature T cells as assessed by FACS analysis. The LTBM derived from euthymic mice contained TcR gamma and delta chain genes in germline configuration, while gene rearrangements of both locus were detected in the LTBM cells from nude mice. These results suggested that gamma and delta gene rearrangements do occur in the bone marrow cells of nude mice and that the T-cell precursors in bone marrow may be increased in frequency in such animals.     

Escherichia coli infection induces only fetal thymus-derived gamma delta T cells at the infected site

Intraperitoneal infection of mice with Escherichia coli induced activated TCR gamma delta T cells in the peritoneal cavity. We provide evidence that the E. coli-induced gamma delta T cells are derived only from the fetal thymus on the following grounds. The gamma delta T cells were not induced in athymic nude mice and irradiated bone marrow-transferred mice which lack fetal thymus-derived T cells. However, E. coli infection of fetal thymus-grafted nude mice did induce fetal thymus-derived gamma delta T cells. These results suggest that the fetal thymus-derived gamma delta T cells colonize the periphery during early ontogeny, and are maintained until adult age. The E. coli-induced gamma delta T cells express only the Vdelta1 gene. Vgamma6 was predominantly expressed whereas anti-Vgamma1 and anti-Vgamma4 monoclonal antibodies stained less than 3 % of the cells. Direct sequencing of PCR products revealed that Vgamma6 and Vdelta1 genes expressed by the E. coli-induced gamma delta T cells were invariant sequences identical to those expressed in the fetal thymus. The antigen (Ag) specificity of a T cell hybridoma expressing the fetal type Vgamma6 / Vdelta1(+) TCR could not be identified as the cells failed to respond to lipopolysaccharide, E. coli Ag, mycobacterial heat shock protein 65, or isopentenyl pyrophosphate. These results suggest that the Vgamma6 / Vdelta1(+) gamma delta T cells derived from fetal thymus can participate in immune responses against bacterial infection through recognition of a novel class of Ag which is not yet identified.     

Expansion of mycobacterium-reactive gamma delta T cells by a subset of memory helper T cells.

Human gamma delta T cells expressing the V gamma 9/V delta 2 T-cell receptor have been previously found to proliferate in response to certain microorganisms and to expand throughout life, presumably because of extrathymic activation by foreign antigens. In vitro expansion of V gamma 9/V delta 2 cells by mycobacteria has been previously shown to be dependent on accessory cells. In order to gain an insight into the mechanisms involved in the expansion of these cells, we have undertaken to identify the peripheral blood subset of cells on which proliferation of V gamma 9/V delta 2 cells in response to mycobacteria is dependent. Contrary to their role in antigen presentation to alpha beta T cells, professional antigen-presenting cells, such as monocytes, B cells, and dendritic cells, were unable to provide the cellular support for the expansion of V gamma 9/V delta 2 cells. Selective depletion of T-cell subsets, as well as the use of highly purified T-cell populations, indicated that the only subset of peripheral blood cells that could expand V gamma 9/V delta 2 cells were CD4+ CD45RO+ CD7- alpha beta T cells. These cells underwent distinct intracellular signaling events after stimulation with the mycobacterial antigen. Expansion of V gamma 9/V delta 2 cells by alpha beta T cells was dependent on cell-cell contact. This is the first evidence that a small subset of the memory helper T-cell population is exclusively responsible for the peripheral expansion of V gamma 9/V delta 2 cells. These data illustrate a unique aspect of antigen recognition by gamma delta T cells and provide new means to study their immune defense role.     

Contribution of extrathymic gamma delta T cells to the expression of heat-shock protein and to protective immunity in mice infected with Toxoplasma gondii.

We demonstrated that gamma delta T cells contribute to protective immunity against Toxoplasma gondii by inducing the expression of a 65,000 MW heat-shock protein (hsp 65) in host macrophages. Here we examined the role of extrathymic and intrathymic gamma delta T cells in protective immunity and hsp 65 expression in mice infected with T. gondii. Intrathymic gamma delta T cells were obtained from severe combined immunodeficiency (SCID) mice grafted with syngeneic fetal thymus (TG-SCID), in which only T cells derived from the donor thymus developed, whereas extrathymic gamma delta T cells were obtained from nude mice that lack thymus. Extrathymic gamma delta T cells from T. gondii-infected nude mice differed from intrathymic gamma delta T cells of infected TG-SCID mice, in terms of Thy1.2 expression and V-region gene usage of T-cell receptor (TCR) gamma delta. Extrathymic gamma delta T cells expressed extremely high levels of Thy1.2, and had V gamma 7 repertoire but lacked V gamma 5,6 and V delta 1,5. On the other hand, intrathymic gamma delta T cells express intermediate and low levels of Thy1,2. These cells possessed V gamma 5,6 and V delta 1,5 but failed to rearrange the V gamma 7 gene. Peritoneal macrophages from infected nude mice contained hsp 65, whereas this protein was scarcely expressed in those of infected TG-SCID mice. Transfer of extrathymic, but not of intrathymic gamma delta T cells to SCID mice enabled their macrophages to express hsp 65. Athymic nude mice were significantly resistant to the infection compared with SCID mice which lack gamma delta T as well as alpha beta T cells. The resistance was dependent upon extrathymic gamma delta T cells, since nude mice depleted of gamma delta T cells using a corresponding monoclonal antibody became extremely susceptible. These results indicated that extrathymic rather than intrathymic gamma delta T cells play some crucial roles in protection against T. gondii and in hsp 65 expression.     

Predominant appearance of gamma/delta T lymphocytes in the liver of mice after birth

gamma/delta T lymphocytes residing in the liver of mice were systematically characterized with respect to their age-related variation, phenotype and V gene segment usage of gamma/delta T cell receptor (TcR). Previous human and murine studies have shown that a high proportion of gamma/delta T cells reside in the liver and that such liver gamma/delta T cells have lymphoblastic morphology and can spontaneously proliferate in vitro. In the present study, a predominant appearance of gamma/delta T cells (up to 23% among CD3+ cells) in the liver was confirmed in 4-week old mice of various strains. gamma/delta T cells in the liver preferentially co-expressed CD8 antigens, whereas the vast majority of gamma/delta T cells in the spleen lacked the CD8 antigens. The identification of gamma/delta T cells in various lymphoid and non-lymphoid organs also revealed the liver to be one of the organs where gamma/delta T cell are most abundant. The level of such liver gamma/delta T cells showed a clear age-related variation. In the fetal stage and just after birth, gamma/delta T cells were not detectable in the liver (less than 0.2%). However, a significantly higher percentage of gamma/delta T cells among both the total population of mononuclear cells and CD3+ cells was detected in the liver of young 2- to 8-week-old mice; this percentage subsequently declined. As the total number of liver mononuclear cells increased in aged mice, the absolute number of liver gamma/delta T cells also increased as a function of age. V gene segment usage analysis by the polymerase chain reaction method demonstrated that V gamma 1 or V gamma 2/V delta 6 were preferentially used by liver gamma/delta T cells. The age-related increase of gamma/delta T cells was more prominent in the liver of athymic nude mice, and such gamma/delta T cells highly co-expressed the CD8 antigens and also utilized the V gamma 1 or V gamma 2/V delta 6 for gamma/delta Tcr. The predominant appearance of unique gamma/delta T cells in the liver, which was inversely related to the existence of the thymus, indicates that these gamma/delta T cells may differentiate extrathymically in the liver.     

The effect of an immunoglobulin mu transgene on B cell maturation.

In mu 17.2.25-transgenic (M54) mice the absolute number of surface IgM (sIgM) B cells in lymphoid organs is drastically reduced compared to normal C57BL/6 mice and a high frequency of B cells express the immunoglobulin (Ig) encoded by the transgene rather than endogenous Ig on the surface. To determine the effect of a mu transgene on B cell development, adoptive cell transfers were performed using mu transgenic (M54) bone marrow and fetal liver cells. The data presented support the following conclusions: (a) adult transgenic bone marrow contains functional B cell precursors able to mature and repopulate the spleen and peritoneum of recipient mice. The relative frequency of transgene (sIgMa) and endogenous (sIgMb) surface sIgM-positive B cells reconstituted by transgenic bone marrow in allotype-matched C57BL/6 recipients is the same as in the M54 donors; (b) serum analysis indicates that transgenic bone marrow donor cells can reconstitute B cells in congenic and severe combined immunodeficient (SCID) recipient mice; (c) transgenic fetal liver cells are not a richer source of precursors for B cells expressing endogeneous Ig; (d) in transgenic mice sIgM+ B cells are not restricted to the CD5+ phenotype, however, the relative frequency of sIgMb B cells that are CD5+ is higher in transgenic than normal mice; and (e) bone marrow cells from adult normal and transgenic mice are able to generate CD5+ B lymphocytes in the spleen and peritoneum of allotype-congenic and neonatal SCID recipient mice. The results indicate that the presence of a complete mu heavy chain transgene does not result in a selective developmental block of "conventional" bone marrow-derived pre-B and B cells.     

Origins of intestinal intraepithelial lymphocytes: direct evidence for a thymus-derived gamma delta T cell component

The contribution of T cell precursors from the thymus and the bone marrow to the pool of intestinal intraepithelial lymphocvtes (IELs) has been studied in a system using donor cells from enhanced-green fluorescent protein (EGFP+ ) transgenic mice a doptively transferred into EGFP- recipient mice. Consistent with previous studies, regeneration of gamma delta and alpha beta T cell populations in the intestinal epithelium occurred within 2-3 weeks of bone marrow transfer into irradiatiated EGFP- animals and prior to T cell repopulation of the spleen, of interest, however, although transfer of whole adult EGFP+ thymocytes to non-irradiated EGFP- congenitally-athymic nude mice produced alpha beta T cells in both the spleen and intestine. Gamma delta T cells in significant number were detected only in the intestine of recipient mice. In contrast, transfer of CD3-, CD4-, CD8- immature thymocytes resulted in no detectable T cells in either the intestine or the spleen of nude mice up to twelve weeks post-cell transfer, suggesting that intestinal IELs generated from thymocytes arose from differentiated lineage-committed cells rather than from immature thymocytes. These findings provide direct evidence for both thymus-independent and thymus-dependent sources of intestinal gamma delta T cells, and they suggest that murine IELs consist of diverse groups of T cells with distinct developmental origins.     

Functionally distinct subsets of human gamma/delta T cells.

To determine if effector subsets exist among human gamma/delta T cells, we examined the cytokine production and cytotoxic activity of gamma/delta T cell clones with different accessory molecule phenotypes, V delta and V gamma gene expression, and J gamma rearrangements. T cell clones bearing gamma/delta T cell receptor produce an array of cytokines like alpha/beta T cell clones. Individual gamma/delta T cell clones produced a characteristic array of cytokines without correlation with V delta or V gamma gene expression. However, when phenotypic subsets were considered, CD4+ gamma/delta clones produced significantly higher levels of interleukin 2 and granulocyte-monocyte colony-stimulating factor compared with CD4-CD8- and CD8+ gamma/delta clones. Similarly, when cytotoxic potential was assessed, CD4+ gamma/delta clones exhibited minimal activity when compared with CD4-CD8- and CD8+ adult peripheral blood gamma/delta clones. We conclude that functionally distinct gamma/delta T cell subsets exist and suggest that these subsets may correlate with expression of the CD4 accessory molecule.     

Thymus-dependent modulation of Ly49 inhibitory receptor expression on NK1.1+gamma/delta T cells

Major histocompatibility complex (MHC) class I-specific inhibitory receptors are expressed not only on natural killer (NK) cells but also on some subsets of T cells. We here show Ly49 expression on gamma/delta T cells in the thymus and liver of beta2-microglobulin-deficient (beta2m-/-) and C57BL/6 (beta2m+/+) mice. Ly49C/I or Ly49A receptor was expressed on NK1.1+gamma/delta T cells but not on NK1.1-gamma/delta T cells. The numbers of NK1.1+gamma/delta T cells were significantly smaller in beta2m+/+ mice than in beta2m-/- mice with the same H-2b genetic background. Among NK1.1+gamma/delta T cells, the proportions of Ly49C/I+ cells but not of Ly49A+ cells, were decreased in beta2m+/+ mice, suggesting that cognate interaction between Ly49C/I and H-2Kb is involved in the reduction of the number of Ly49C/I+ gamma/delta T cells in beta2m+/+ mice. The frequency of Ly49C/I+ cells in NK1.1+gamma/delta T cells was lower in both lethally irradiated beta2m+/+ mice transplanted with bone marrow (BM) from beta2m-/- mice and lethally irradiated beta2m-/- mice transplanted with BM from beta2m+/+ mice than those in adult thymectomized BM-transplanted chimera mice. These results suggest that reduction of Ly49C/I+ NK1.1+gamma/delta T cells in beta2m+/+ mice is at least partly due to the down-modulation by MHC class I molecules on BM-derived haematopoietic cells or radioresistant cells in the thymus.     

T cell receptor repertoire of gamma delta cells generated from the 14-day embryonic mouse thymus.

Whole, undisrupted 14 day mouse fetal thymus lobes cultured in the presence of 10 U/ml IL-2 generate a heterogeneous population of gamma delta-expressing T cells. Phenotypic analysis has shown that the majority of gamma delta T cells in such cultures stain with the anti-V gamma 3-specific mAb 536. To investigate the V gamma T-cell receptor diversity of cultured fetal thymocytes, cDNA was prepared and amplified using the polymerase chain reaction. The DNA fragments obtained were subsequently cloned and sequenced and compared with those obtained from fresh and organ-cultured 14 day fetal thymus lobes. Results obtained tend to support a positive selection model of gamma delta T cell differentiation.     

The T-cell receptor repertoire is strikingly similar in older nude mice compared to normal adult mice

Due to the absence of a normal thymus, nude mice are characterized by low numbers of mature T cells. It has been shown in older nude mice that appreciable numbers of Thy-1+, CD8+ and, to a much lesser extent, CD4+ T cells can be found. A second T-cell receptor (TCR), gamma delta-TCR, has been shown to be expressed by a large number of T cells in young and old nude mice. Recent data suggest that certain gamma and delta chain variable (V), diversity (D), joining (J) and constant (C) region genes are associated with both temporal and spatial organization of the gamma delta-T-cell repertoire in normal mice. In this study we observed the predominant use of V gamma 2J gamma 1C gamma 1 and V delta 5J delta 1C delta in the spleen of older nude mice. In the skin, a near exclusive expression of V gamma 3J gamma 1C gamma 1 and V delta 1J delta 2C delta gene segments was observed. In addition, a dendritic epidermal cell (DEC) clone was isolated from the skin of a nude mouse and expressed a gamma delta receptor consisting of V gamma 3J gamma 1C gamma 1 and V delta 1J delta 2C delta. Preliminary results show that this clone kills only YAC (and NK target) and not other tumor targets. The killing can be redirected by anti-CD3 monoclonal antibody which suggests that the receptor is potentially functional. These results demonstrate that without normal thymus development nude mice not only produce gamma delta-T cells with a lytic potential, but also manage to express gamma delta-TCR's that are similar in spatial arrangement to normal mice. This suggests a possible difference in the requirements for gamma delta-TCR ontogeny.     

Origin of thymic and peritoneal Ly-1 B cells.

Murine Ly-1 B cells were first found in the peritoneal cavity (Per C). It has been thought that Ly-1 B cells are responsible for the development of autoimmune diseases, since they spontaneously produce autoantibodies. However, we have found that such Ly-1 B cells are present in the thymus of normal mice, and that they play a crucial role in negative selection in the Mls system. In addition, thymic Ly-1 B cells and Per C Ly-1 B cells have been found to differ in function, ontogeny and location. In this communication, we report on the different origins of these cells. C57BL/6 mice are lethally irradiated and reconstituted with fetal (day 14) liver cells or bone marrow cells from 4-, 6- or 12-week-old mice. Fetal liver cells are shown to have the capacity to reconstitute both thymic and Per CLy-1 B cells, whereas bone marrow cells from 6- or 12-week-old mice have the capacity to reconstitute thymic Ly-1 B cells, but not Per C Ly-1 B cells, indicating that fetal hemopoietic stem cells (HSC) differ from HSC in the bone marrow of adult mice, and that thymic and Per C Ly-1B cells have different origins.     

Abnormal rearrangements of T-cell receptor genes occur in long-term cultured bone marrow cells of lpr/lpr mice.

To search the abnormality in prethymic T-cell precursors in lpr/lpr(lpr) mice, rearrangement and expression of T-cell receptor (TcR) genes were investigated in long-term cultured bone marrow (LTBM) cells of lpr mice, in which the developmental steps of T-cell precursors may be better synchronized than those in bone marrow (BM) cells. Neither rearrangment nor expression of TCR gamma and delta genes were detected in the LTBM cells from +/+ control mice, whereas some gamma gene rearrangements were detected in those derived from lpr mice, irrespective of the genetic background. When BM cells or LTBM cells from lpr mice were transplanted into supralethally irradiated +/+ mice the lpr-derived BM cells appeared earlier in the thymus of the recipient mice than +/+-derived BM cells and the recipients suffered from lethal wasting syndrome. In addition, the lpr-derived BM cells showed higher activity in colony-forming unit spleen (CFUs) than the +/+-derived BM cells. These results suggest that the T-cell progenitors in the BM of lpr mice may be different not only in quantity but also in quality from those of +/+ mice.     

Development of cytotoxic T lymphocyte precursors in the absence of the thymus

Reconstitution of adult thymectomized and lethally irradiated C3H or (C 3 H × DBA/2) F₁ mice with syngeneic adult bone marrow cells or 18 to 20-day-old fetal liver cells resulted almost regularly in the generation of large numbers of cytotoxic T lymphocyte precursors (CTLP) against trinitrophenylated C3H cells and three types of allogeneic stimulator and target cells. The responses against 2,4,6- trinitrophenylated cells were restricted and in so far typical for cytotoxic T lymphocytes. The failure of bone marrow cells from athymic C 3 H mice to reconstitute significant cytotoxic activity indicated that the CTLP had developed from “cytotoxic T lymphocyte preprogenitors” (CTLPP) which were derived from the donor and which had already been under thymic influence. The stem cell preparations in themselves contained frequencies of mature CTLP far too low to account for the reactivity and CTLP frequencies in the reconstituted mice. This fact implied that the CTLPP proliferated extensively and/or matured in the recipient in the absence of the thymus. Individual mice failed to respond to one or more of these stimulator cells. The failure of individual mice to respond was randomly expressed for the four specificities even when the mice were reconstituted with the same stem cell preparation suggesting that CTLPP with defined specificity and low frequency (about 1/10⁸ in fetal liver) were contained in the stem cell preparation and transferred randomly into the irradiated recipients. This would imply that one CTLPP gives rise to 5 × 10³–20 × 10³ CTLP within 5 weeks. The data suggest that the peripheral pools, at least of certain T cell subclasses, may be maintained largely by post-thymic proliferation of CTLPP.     

Origin of CD57+ T cells which increase at tumour sites in patients with colorectal cancer.

Human T cells carrying natural killer (NK) markers, CD57 or CD56 antigens, appear to be distinguishable from other T cell subsets in terms of their granular lymphocyte morphology and their numerical increase in patients with AIDS and in recipients of bone marrow transplantation. At the beginning of this study, we observed that CD57+ T cells as well as CD56+ T cells were abundant at tumour sites in many patients with colorectal cancer. Since all these findings for CD57+ T cells are quite similar to those of extrathymic T cells seen in mice, we investigated how CD57+ T cells are distributed to various immune organs in humans. They were found to be present mainly in the bone marrow and liver, but to be completely absent in the thymus. Similar to the case of extrathymic T cells in mice, they were observed to consist of double-negative CD4-8- subsets as well as single-positive subsets (preponderance of CD8+ cells), and to contain a considerable proportion of gamma delta T cells. These features are striking when compared with those of CD57- T cells, which are characterized by an abundance of CD4+ subsets and alpha beta T cells. Not only at tumour sites but also in the peripheral blood, some patients with colorectal cancer displayed elevated levels of CD57+ cells. These results suggest that CD57+ T cells may be of extrathymic origin, possibly originating in the bone marrow and liver, and may be associated with tumour immunity, similar to another extrathymic population of CD56+ T cells in humans.     

Expression of a fetal gamma delta T-cell receptor in adult mice triggers a non-MHC-linked form of selective depletion.

Day 14 fetal thymocytes and adult dendritic epidermal T cells (dEC) of all mouse strains express a characteristic non-polymorphic gamma delta T-cell receptor which is rarely found in the adult thymus or lymph nodes. We have made transgenic mice expressing this particular set of receptors on T cells in C3H and C57BL/6 mice. In adult mice of the latter strain, a dramatic depletion of transgene expressing T cells occurs and this effect is primarily mediated by thymic radiosensitive cells. The depletion is genetically dominant but not MHC-linked with major factor(s) mapping to chromosome 18. Taken together, our results show that strain-specific developmental changes in the thymic environment may play a role in shaping the gamma delta TCR repertoire.