In situ localization of T cell receptor beta chain in the murine thymus: changes in the intrathymic distribution of thymocytes expressing beta chain during fetal development

The expression of T cell receptor beta chain in the developing thymus was examined at the light and electron microscopic levels using the monoclonal antibody F23.1. Cells expressing cytoplasmic forms of beta chain were first observed at Day 16 of gestation, while thymocytes expressing cell surface beta chain were detected about a day later. Clustering of cortical F23.1+ cells was more pronounced in fetal thymus when compared to adult. The density of F23.1+ cells in the subcapsular areas of the thymus was initially lower than that in the rest of the cortex or the medulla. Within the subcapsular and cortical areas of the thymus there was an inverse relationship between the density of F23.1+ cells and cells labeled with the lectin from Dolichos bifloris, which binds to terminal alpha-linked N-acetylgalactosamine residues preferentially expressed by L3T4-/Lyt2- thymocytes. Although this pattern was less pronounced with increasing gestational age, it was still apparent at birth.     

Proliferation and production of IL-2 and B cell stimulatory factor 1/IL-4 in early fetal thymocytes by activation through Thy-1 and CD3

To examine which cell surface molecules can operate as transducers of activation signals to early fetal thymocytes, we analyzed the ability of mAb to CD3 and Thy-1 to induce fetal thymocyte activation. Both proliferation and lymphokine secretion were used as measures of activation. We show that anti-CD3 antibodies induce activation of fetal thymocytes as early as day 13 of fetal thymus development, 2 days before CD3 can be detected by flow cytometry. In addition, an alternative activation signal can be delivered to fetal thymocytes through the Thy-1 molecule as early as it is expressed, i.e., day 13. Both CD3- and Thy-1-mediated activation of day 15 fetal thymocytes results in expansion of cells expressing a CD3-gamma delta receptor complex; no CD3-alpha beta receptor complex could be detected. IL-2 production induced by CD3- and Thy-1-induced activation of fetal thymocytes is evident at the 13th day of gestation. Finally, an additional lymphokine B cell stimulatory factor-1 (BSF-1)/IL-4 (so far known only to be produced by mature CD3- cells), is also produced by fetal thymocytes. The results demonstrate that at least two cell surface molecules, Thy-1 and CD3, can function as pathways of activation in fetal thymocytes, and that at least two lymphokines, IL-2 and BSF-1/IL-4, are produced upon activation. These findings may well reflect a role for the early appearance of CD3- cells in thymus ontogeny.     

CD4-CD8- thymocytes from MRL-lpr/lpr mice exhibit abnormal proportions of alpha beta- and gamma delta-TCR+ cells and demonstrate defective responsiveness when activated through the TCR.

MRL-lpr/lpr (lpr) mice develop profound lymphadenopathy resulting from the accumulation of CD4-CD8- (double-negative, DN) cells in the peripheral lymphoid organs. Earlier studies from our laboratory demonstrated an increased proportion of DN cells in the thymus of lpr mice with age. Inasmuch as the DN thymocytes constitute a heterogenous population of cells, in the present study, we investigated the TCR phenotype of DN thymocytes and their responsiveness to activation through the TCR. The DN thymocytes of young (1 month of age) lpr mice contained approximately 65% CD3+ cells of which approximately 60% were alpha beta-TCR+ and approximately 39% were gamma delta-TCR+ as detected by using pan anti-TCR mAbs. In old (4-6 months of age) or young MRL-(+/+) mice, similar proportions of CD3+, alpha beta- or gamma delta-TCR+ DN thymocytes were detected. Interestingly, however, in old (4-6 months of age) lpr mice, the CD3+ T cells increased to approximately 86% and the majority of these (approximately 81%) were alpha beta-TCR+ and only approximately 3% were gamma delta-TCR+. Also, in old lpr mice, there was a 10-fold increase in the absolute number of alpha beta-TCR+ DN cells in the thymus, whereas, the absolute number of gamma delta-TCR+ DN cells in the thymus did not alter significantly. Furthermore, a majority (approximately 84%) of the old lpr DN thymocytes expressed CD45R, similar to the peripheral DN T cells. In contrast, only a small number (approximately 1%) of DN thymocytes from young lpr or MRL-(+/+) mice expressed CD45R. The DN thymocytes from young lpr or MRL-(+/+) mice demonstrated strong and similar proliferative responsiveness to stimulation with PMA + calcium ionophore or PMA + IL-2, or to immobilized mAb directed against the TCRs (CD3, alpha beta and gamma delta). In contrast, the DN thymocytes and the DN peripheral T cells from old lpr mice demonstrated marked defect in responding to the above stimuli. The present study suggests that with the onset of lymphadenopathy, the DN cells in the thymus of old lpr mice are increasingly skewed toward the alpha beta-TCR repertoire, the majority of which express CD45R and respond poorly to mitogenic stimuli or when activated through the TCR. It is suggested that migration of such cells continuously to the periphery may result in severe lymphadenopathy seen in old MRL-lpr/lpr mice.     

Preferential proliferation of T cell receptor V gamma 3-positive cells in IL-2-stimulated fetal thymocytes

Thymocyte cell suspensions, prepared from mice at different ages, were cultured in vitro with human rIL-2. This stimulation resulted in a cell population that contained almost 50% TCR-gamma delta-positive cells if thymocytes were taken from fetal day 17 until just after birth. Analysis of the variable (V gamma) region used by the TCR-gamma delta cells revealed that 90% of them expressed TCR-V gamma 3, and less than 5% expressed TCR-V gamma 2. Cells positive for TCR-alpha beta were barely detectable. If fetal day 18 organ cultured thymus lobes, instead of a cell suspension, were stimulated with IL-2, no rise in the number of TCR-V gamma 3+ or TCR-delta+ cells was observed, whereas a partial outgrowth of TCR-alpha beta+ cells occurred. From day 1 after birth, the number of TCR-gamma delta cells recovered from an IL-2-stimulated thymocyte cell suspension dropped to reach a plateau of 15% of the total cell number, whereas TCR-V gamma 3+ cells became undetectable in older animals. TCR-alpha beta+ cells, on the other hand, quickly rose in cell number after birth. Kinetic analysis showed that the preferential outgrowth of TCR-V gamma 3+ cells in IL-2-stimulated fetal day 18 thymocyte cell suspensions was present from the onset of the culture; a significant proliferation of CD4 or CD8 single positive TCR-alpha beta cells was never observed. This lack of proliferation of TCR-alpha beta cells was not due to inhibition by the activated TCR-V gamma 3+ cells. Throughout the IL-2 culture, one-fourth of the TCR-V gamma 3+ thymocytes was positive for CD8. Analysis of the DNA content and the IL-2 receptor (IL-2R) p55 expression showed that during the first days of culture the TCR-V gamma 3+ cells had a much higher proliferation rate than the TCR-V gamma 3- cells, although TCR-V gamma 3+ IL2R p55+ cells could not be detected. From day 3 to 4 of culture, the proliferation rate of TCR-V gamma 3+ cells equaled that of the rest of the cells and less than 20% of the TCR-V gamma 3+ cells expressed the IL-2R p55. The biologic significance of our findings is discussed.     

Systematic development of distinct T cell receptor-gamma delta T cell subsets during fetal ontogeny

To elucidate the developmental pattern and diversity of murine cluster of differentiation (CD)3-associated TCR-gamma delta heterodimers, adult and fetal thymocytes were examined for cell-surface expression of various gamma- and delta-encoded TCR. Biochemical analysis, using antisera specific for distinct C gamma gene products, revealed the presence of T cells expressing C gamma 1 and/or C gamma 4 heterodimers in adult and fetal CD4- CD8- thymocyte populations. Although CD4-CD8- thymocyte populations express both C gamma 1 and C gamma 4 TCR-gamma delta heterodimers early in fetal thymus development, the relative level of C gamma 4-expressing T cells was significantly lower than previously observed in peripheral lymphoid organs. In addition, biochemical studies revealed the presence of TCR-gamma delta heterodimer(s) expressed during fetal ontogeny which were not detected in adult thymocyte or peripheral lymphoid populations. Studies of N-glycosylation patterns of one of these heterodimers suggested that it contained a rearranged V gamma 3/C gamma 1 gene product. To examine in detail individual TCR-gamma delta heterodimers, a panel of TCR-gamma delta expressing hybridomas was prepared. Biochemical analysis at the clonal level revealed that indeed three distinct TCR-gamma delta heterodimers were present at day 16 of fetal thymus development, with TCR-gamma-chains most likely encoded by V gamma 2/C gamma 1, V gamma 3/C gamma 1, and V gamma/C gamma 4. Together these findings suggest an ordered development of TCR-gamma delta T cells in the thymus and selective expression of distinct TCR-gamma delta subsets in peripheral lymphoid organs such as spleen and lymph nodes.     

Distribution of T cells bearing different forms of the T cell receptor gamma/delta in normal and pathological human tissues

Frozen sections from normal and pathologic human tissues were immunostained by the APAAP technique with three mAb directed against different epitopes of the TCR gamma delta; TCR delta 1 which binds to all cells bearing the TCR gamma delta; BB3 and delta TCS1 which, by immunoprecipitation studies, appear to react respectively with the disulfide-linked and nondisulfide-linked form of the TCR gamma delta. In normal thymus, TCR delta 1+ cells accounted for approximately 2% of the CD3+ thymocytes and were about three times more numerous in the medulla than in the cortex. TCR delta 1+ cells were mostly constituted by the delta TCS1 reactive subset (average ratio delta TCS1/BB3: 3.7). In the tonsil, the TCR delta 1+ cells (about 3% of CD3+ elements) were mainly located in the interfollicular area, where they frequently tended to arrange around high endothelium venules. In most samples, TCR delta 1+ cells were distributed beneath to the tonsil epithelium. Unlike thymus, the majority of TCR delta 1+ cells were usually constituted by the BB3-reactive subset (average BB3/delta TCS1 ratio: 2.0). A similar predominance of BB3+ over delta TCS1+ cells was also observed in normal peripheral blood. The spleen was the organ with the highest concentration of TCR delta 1+ cells that, like in the thymus, were mostly represented by delta TCS1+ elements. Noteworthy, the TCR delta 1+ cells were preferentially located in the splenic sinusoids while TCR alpha beta-bearing lymphocytes mostly occupied the periarteriolar sheaths of penicilliary arteries. The majority of neoplastic T cell proliferations studied lacked to express the TCR gamma delta. Two cases of beta F1-(TCR alpha beta-) T lymphoblastic lymphoma, however, were TCR gamma delta+ (delta TCS1+/BB3-). Both of them showed a stage II cortical phenotype, e.g., CD1+/CD3+/CD4+/CD8+/TCR delta 1+. Among inflammatory conditions, an increase of BB3+ cells was observed in close association with necrotic areas in cases of Kikuchi's and tuberculous lymphadenitis. The significance of this finding is under study.     

Epithelial heterogeneity in the murine thymus: fucose-specific lectins bind medullary epithelial cells

By means of histochemical techniques, two lectins with nominal specificity for L-fucose, Tetragonolobus purpureas agglutinin (TPA), and Ulex europeus agglutinin (UEA) were found to specifically label the medullary area of murine thymuses. Although the binding of both lectins was restricted to the medullary area of the thymus, each staining pattern was unique. Cells binding UEA formed a reticular network throughout the medulla, whereas cells binding TPA occurred as single cells or small clumps of cells and resembled Hassall's corpuscles. The cells binding either lectin were identified as epithelial on the basis of ultrastructural features (tonofilaments, desmosomes, and keratohyalin bodies) and resembled Ia+ medullary epithelial cells described previously. An age-related decline in UEA binding was observed, whereas labeling with TPA remained unchanged. On the basis of the labeling patterns obtained with UEA and TPA and the reported specificities of these two lectins, it is suggested that the majority of the fucose detected is associated with type 1 carbohydrate chains.     

Tolerogenic ability of thymocytes in organ-cultured thymus lobes.

It is generally believed that macrophages and dendritic cells are the major cell populations that present tolerogenic self antigens to developing thymocytes. However, it is still controversial whether self antigens expressed on thymocytes themselves work as tolerogens in the thymus. To evaluate this possibility, Thy-1 bright cells were sorted out from fetal thymus cells on the 15th gestation day, and were colonized into 2'-deoxyguanosine-treated allogeneic thymus lobes. The repopulated thymus lobes were organ-cultured, and the allo-specific killer activity of thymocytes recovered from the lobes was examined. These cells were tolerant to class I but not to class II-MHC of the donor haplotype, indicating that class I molecules expressed on the thymocytes worked as tolerogen. Tolerogenic ability of Thy-1+ cells was also demonstrated in another system. Upon intimate contact with allogeneic thymus lobes on a polycarbonate filter, thymus lobes fused with each other and Thy-1+ cells co-migrated (Eur. J. Immunol. 19:1525-1530, 1989). In thymus lobes rendered parabiotic from day 5, CTL tolerance was achieved against class I but not to class II MHC. These data indicate that thymocyte-thymocyte interaction is sufficient to induce class I CTL tolerance in developing thymocytes.     

Thymus-dependent and thymus-independent developmental pathways for peripheral T cell receptor-gamma delta-bearing lymphocytes

To elucidate the developmental pathways of T cells that bear TCR gamma delta, we have analyzed the kinetics of expression and biochemical characteristics of gamma delta receptors in the thymus and spleen of normal and athymic (nude) mice, as well as nude mice engrafted with neonatal thymuses. TCR gamma delta-bearing thymocytes and splenocytes have a CD4-8- phenotype, and both populations express products of the C gamma 1 locus. TCR gamma delta-bearing cells develop in the thymus before their appearance in the spleen. Young nude mice have no detectable TCR gamma delta-bearing cells in their spleens. When young nude mice are given thymus grafts, TCR gamma delta-bearing cells of host origin first develop in the engrafted thymus, followed by their appearance in the spleen. In the absence of a thymus graft, the spleens of old nude mice eventually develop small numbers of TCR gamma delta + cells, as well as TCR alpha beta + cells. These results demonstrate that there is a major thymic-dependent pathway for TCR gamma delta expression, as well as a minor thymic-independent pathway seen in older nude mice. The development of TCR gamma delta + cells in the thymus before their appearance in the spleen, both in normal ontogeny as well as in the thymus-engrafted nude mouse model, suggests that thymic TCR gamma delta + cells are precursors of the thymus-dependent population of peripheral TCR gamma delta + cells.     

The expression of T cell receptor-associated proteins during T cell ontogeny in man

The expression of TCR-associated molecules was examined in human fetal and postnatal tissues. From gestational wk 7 onward in the fetal liver, putative prothymocytes have been identified with cytoplasmic CD3 positivity (cCD3+). These immature cells are TdT- and do not express membrane CD3 (mCD3-) or TCR beta identified by beta F1, but show CD7 and CD45 positivity without CD1, CD2, CD5, CD4, CD8, CD10, and class II Ag. Their high proliferative activity is indicated by greater than 85% Ki67 positivity. After the 10th wk, beta F1+, mCD3+ cells also appear in the liver and these are mostly Ki67- but no TCR gamma delta-bearing cells can be identified at such an early stage of extrathymic development. In the mCD3- TdT-fetal thymus (10 1/2 to 18th wk) cCD3+, mCD3- CD1-blasts proliferate (Ki67+) and lack TCR-beta or TCR-gamma delta. The TdT-, CD1+ cortical thymocytes develop into TCR-beta + and WT31-positive (TCR-alpha beta +) cells. Subsequently TdT-positive thymocytes become detectable around 19 to 20 wk, and in such glands the peak of proliferative activity is seen among TdT+, cCD3+ cells which appear to acquire, in a regular sequence, cytoplasmic beta F1 (TCR-beta), mCD3, and TCR-alpha beta (WT31 positivity) together with the loss of TdT and Ki67 positivity. A newly described transitional population of cells is TdT-, beta F1+ but exhibits no detectable WT31 positivity. These cells correspond to the CD1+, mCD3+ thymocytes and are probably the targets of thymic selection. The cells of the TCR-gamma delta lineage, detected by mAb TCR-delta-1 and delta TCS1, are rare (0.02 to 0.5%) among thymocytes from gestational wk 10 1/2 onward through the whole span of thymic development, but these cells include a proportion (18 to 59%) of cells expressing CD1 Ag, suggesting that these TCR-gamma delta cells differentiate in the thymus. Among the CD1+, TCR-gamma delta + thymocytes, no TdT positivity can be detected.     

Vascular and stromal changes in irradiated and recovering rat thymus.

To analyze the mechanisms responsible for thymocyte proliferation, maturation and migration in the thymus, the rat thymus just after, and recovering from irradiation was studied morphologically. The vascular structures of the rat thymus after a radiation dose of 6 Gy were found to be destroyed on day 3, but had recovered to almost normal by day 7, suggesting that the abrupt recovery of thymus structure after irradiation was due primarily to this change in vascular structure. Furthermore, the epithelial tissues in the thymic cortex appeared to contribute to this abrupt proliferation, and possibly to the abrupt maturation of thymocytes, while medullary epithelial tissues remained sparse and appeared inactive for a relatively long period. These findings are considered important for understanding the interrelationship between thymic epithelial cells and thymocytes with respect to thymocyte proliferation, maturation and migration.     

Exogenous control of I-A expression in fetal thymus explants

With the use of a system in which 14 day fetal thymus lobes were cultured in vitro with bone marrow or other lymphoid cells, evidence was obtained that entry of macrophage/dendritic cells (M phi/DC) into the thymus causes a marked rise in the density of endogenous I-A molecules expressed in the cortex, presumably on epithelial cells. High cortical I-A expression also occurred when thymus lobes were cultured with supernatants containing IFN-gamma; addition of anti-IFN-gamma antibody blocked I-A induction. The working hypothesis for these findings is that cellular interactions occurring in the medullary region between M phi/DC and a subset of thymocytes leads to production of IFN-gamma, which then diffuses into the cortex and promotes epithelial I-A expression. The possible relevance of this scheme to the process of thymic "education" is discussed.     

Intrathymic T cell differentiation in radiation bone marrow chimeras and its role in T cell emigration to the spleen. An immunohistochemical study

Immunohistochemical studies were made on the regeneration of T cells of host- and donor-type in the thymus and spleen of radiation bone marrow chimeras by using B10- and B10.BR-Thy-1 congenic mice. Both the thymic cortex and the medulla were first repopulated with thymocytes of irradiated host origin, restoring the normal histologic appearance by days 11 to 14, regardless of the H-2 compatibility between the donor and the host. In Thy-1 congenic chimeras, thymocytes of donor bone marrow origin, less than 100 cells in one thymic lobe, were first recognized at day 7, when the thymus involuted to the smallest size after the irradiation. The thymocytes of donor-type then proliferated exponentially, showing a slightly faster rate when higher doses of bone marrow cells were used for reconstitution, reaching a level of 100 million by day 17 and completely replacing the cortical thymocytes of host origin by day 21. The replacement of cortical thymocytes started from the subcapsular layer in a sporadic manner. The replacement of medullary thymocytes from host- to donor-type occurred gradually between days 21 and 35, after the replacement in the cortex was completed. In the spleen, about 1 million survived cells were recovered at day 3 after the irradiation, and approximately 60% of them were shown to be host-type T cells that were observed in the white pulp areas. The host-type T cells in the spleen increased gradually after day 10, due to the influx of host-type T cells from the regenerating thymus. Thus a pronounced increase of T cells of host-type was immunohistochemically observed in the splenic white pulp between days 21 and 28, when thymocytes of host-type were present mainly in the thymic medulla. These host-type T cells were shown to persist in the spleen for a long time, as long as 420 days after the treatment. Phenotypically, they were predominantly Lyt-1+2+ when examined at day 28, but 5 mo later, they were about 50% Lyt-1+2+ and 50% Lyt-1+2-. Donor-type T cells in the spleen began to appear at about day 14 in chimeras that were transplanted with a larger dose of bone marrow cells, whereas this was slightly delayed in those grafted with a smaller dose of bone marrow cells, starting at about day 28.(ABSTRACT TRUNCATED AT 400 WORDS)     

A novel gamma delta T cell receptor for antigen adds limited diversity to the gamma delta repertoire in adult thymus

The surface expression of CD3-associated TCR chains on hybridoma cell lines derived from adult gamma delta thymocytes was analyzed. These cell lines were unusual, in that a) they expressed a surface heterodimer consisting of a 40- and a 42-kDa chain, i.e., comprised of chains different from any previously reported gamma delta-TCR all of which express C gamma 1- or C gamma 2-encoded gamma-chains; b) their CD3-associated TCR could not be categorized as alpha beta-TCR dimers, despite the similarities in m.w. of the TCR chains, because full size 1.3-kb beta-chain mRNA capable of encoding a functional beta-chain could not be detected in these cells; c) neither of the receptor chains could be precipitated with anti-C gamma 1C gamma 2-peptide antisera. Biochemical analysis demonstrated that the 42-kDa delta-chain is a novel chain, which differs from any reported delta-chains in size, charge and number of glycosylation sites. Collectively, the data on analysis of the 40-kDa chain strongly suggest that it represents a gamma-chain encoded for by the C gamma 4 locus, protein products of which have not yet been reported in the thymus. This gamma-chain was also unique, in that its isoelectric point was much lower than that of other gamma-chains. The gamma- and delta-chains on these C gamma 4-expressing hybridomas were indistinguishable from one another in size and charge (as determined by nonequivalent pH gradient electrophoresis/SDS-PAGE analysis and analysis after endoglycosidase treatment). Because the cell lines were randomly chosen from large panels of hybridomas, these results may well imply strikingly nonrandom pairing of thymocyte-derived C gamma 4 chains and the delta-chains reported here. Thus, only limited additional gamma delta repertoire diversity may be generated by availability of this gamma delta-TCR in the thymus.     

Tissue localization and CD8 accessory molecule expression of T gamma delta cells in humans

In this study, we used TCR isotype-specific antibodies to examine the frequency, phenotype, and histologic localization pattern of T gamma delta cells in humans. The TCR delta 1+ cells comprised an average of 15% of the splenic CD3+ cells and 7% of circulating T cells. The T gamma delta cells in these human tissues, like their avian counterparts, were often not "double-negative" for the CD4 and CD8 accessory molecules. Approximately 50% of the splenic delta+ cells expressed CD8, and 30% of the delta+ cells in blood were CD8+. T cells of both gamma delta and alpha beta TCR isotypes were exceedingly rare in the skin. The T gamma delta cells exhibited preferential homing to the sinusoidal areas (red pulp) of the spleen and into the epithelial layer of the intestine in humans, as had been previously noted in chickens. Although 80% of the T gamma delta cells in the human intestinal mucosa were localized in the epithelial layer, these cells represented only 5 to 10% of all the CD3+ T cells in this microenvironment. We conclude that T gamma delta cells represent a sizeable subpopulation of the T cells in human peripheral tissues. The phylogenetic conservation of the CD8 expression by peripheral T gamma delta cells and of their preferential homing pattern suggests a special role in bodily defense for this T cell subpopulation.     

Effect of cyclosporin A on the ontogeny of different T cell sublineages in chickens

We have used a panel of murine mAb against chicken TCR and associated molecules to study the effect of cyclosporin A (CsA) on the ontogeny of the different sublineages of T cells. After injection of CsA (20 mg/kg/day from day 0 to 20) we observed a significant suppression of the normal maturation of the TCR2 (alpha beta TCR) cells in their transition from cortical CD4+CD8+ thymocytes to the mature single positive cells in the thymus medulla. The TCR3 subpopulation, a distinct form of alpha beta-like TCR in chickens, was inhibited from initially developing within the cortex by CsA, indicating that the TCR3 subpopulation is functionally distinct from the TCR2+ cells. In contrast, the maturation and peripheral emigration of TCR1 (gamma delta TCR) cells was unaffected by CsA treatment. Mature splenic T cells sorted for either TCR1+ or TCR2+ subsets were equally sensitive to CsA blockade of Con A-stimulated mitogenesis, indicating that there is no inherent difference in CsA sensitivity between these sublineages. Furthermore, no difference was detected in the expression of class II MHC Ag in thymi of birds treated with olive oil vs CsA. Inasmuch as the mechanism of CsA action appears to involve inhibition of TCR initiated signal transduction for lymphokine synthesis, these data indicate that a similar signaling is involved in thymic repertoire selection for TCR2. The lack of an effect on TCR1 cell maturation suggests that the TCR1 repertoire may not undergo selection in the thymus as do TCR2+ cells.     

Thymocyte subpopulations during early fetal development in sheep

Phenotypic analysis of thymocytes during fetal development may identify subpopulations which are either absent or difficult to detect in postnatal thymus. A panel of monoclonal antibodies specific for sheep lymphocyte antigens (SBU-T1, -T4, -T8, -T6) was used to identify thymocyte subpopulations in postnatal and fetal sheep. Thymuses were analyzed by two-color immunofluorescence and flow cytometry or by immunohistology. Two-color immunofluorescent staining of postnatal sheep thymus with anti-SBU-T4 and anti-SBU-T8 revealed four relatively distinct subpopulations with particular localizations: a) SBU-T4-T8-, predominantly outer cortex (12%); b) SBU-T4+T8+, inner cortex (74%); c) SBU-T4+T8-, medulla (10%), and d) SBU-T4-T8+, medulla (4%). One- and two-color immunofluorescent analysis of cells from early fetal thymuses demonstrated the appearance of SBU-T8+ cells well before SBU-T4+ cells. Immunohistologic staining of fetal sheep thymus at various stages of gestation (term = 150 days) revealed that lymphoid cells and MHC class II-positive dendritic cells first appeared at 35 days, at which stage the thymic epithelium was weakly positive for class I MHC antigens but negative for class II MHC antigens. The earliest lymphocyte antigens detectable on fetal sheep thymocytes were SBU-LCA and SBU-T1. By 40 days, the antigens SBU-T6, SBU-T4, and SBU-T8 were detectable on a small number of thymocytes; SBU-T8 preceded SBU-T4, and the number of SBU-T8+ thymocytes always exceeded the number of SBU-T4+ thymocytes throughout early gestation. At 50 days, a thymic medulla appeared and thereafter grew rapidly in size. Immunoperoxidase staining of serial sections of the fetal neck revealed cortical-type thymocytes outside the thymus from 40 days onward, before the appearance of a thymic medulla. However, by 60 days, only medullary-type thymocytes were observed either extrathymically or within the interlobular septa of the thymus, indicating that only thymocytes with a medullary phenotype leave the thymus from this stage of gestation.     

A population of early fetal thymocytes expressing Fc gamma RII/III contains precursors of T lymphocytes and natural killer cells.

We have identified a dominant fetal thymocyte population at day 14.5 of gestation in the mouse that lacks CD4 and CD8 but expresses Fc gamma RII/III several days prior to acquisition of the T cell receptor (TCR) in vivo. If maintained in a thymic microenvironment, this population of CD4-CD8-TCR-Fc gamma RII/III+ thymocytes differentiates first into CD4+CD8+TCRlowFc gamma RII/III- thymocytes and subsequently CD4+CD8-TCRhighFc gamma RII/III- and CD4-CD8+TCRhighFc gamma RII/III- mature Ti alpha-beta lineage T cells. However, if removed from the thymus, the CD4-CD8-TCR-Fc gamma RII/III+ thymocyte population selectively generates functional natural killer (NK) cells in vivo as well as in vitro. These findings show that a cellular pool of Fc gamma RII/III+ precursors gives rise to T and NK lineages in a microenvironment-dependent manner. Moreover, they suggest a hitherto unrecognized role for Fc receptors on primitive T cells.