Scurfy, the Foxp3 locus, and the molecular basis of peripheral tolerance.

The ability to rapidly and efficiently recognize and eliminate pathogens while sparing normal self tissue is a hallmark of the mammalian immune system. When it fails, however, autoimmune disease results. The genetic and environmental factors that control the process of making such distinctions, not to mention the specific targeted tissues, are extraordinarily complex in the human population; only now are we characterizing the candidate genes responsible for these responses to pathogens. The examination of specific traits in murine models of disease has led to the identification of many of the candidate genes for human disease. The study of mouse mutations (both induced and spontaneous) has also greatly advanced our understanding of the immune responses and autoimmune disease. Here, we describe the use of classical mouse genetics to identify one gene centrally involved in the control of immune responses. Furthermore, although mutations in the orthologous human gene result in a virtually identical phenotype to that seen in the mouse, it is unlikely that studying the human disease populations alone would have successfully identified this gene. Thus, despite the complete sequencing of the human and mouse genomes, the examination of murine mutations remains a powerful and unbiased tool to connect genotype and phenotype.

The amount of scurfin protein determines peripheral T cell number and responsiveness.

In the absence of the recently identified putative transcription factor scurfin, mice develop a lymphoproliferative disorder resulting in death by 3 wk of age from a pathology that resembles TGF-beta or CTLA-4 knockout mice. In this report, we characterize mice that overexpress the scurfin protein and demonstrate that these animals have a dramatically depressed immune system. Mice transgenic for the Foxp3 gene (which encodes the scurfin protein) have fewer T cells than their littermate controls, and those T cells that remain have poor proliferative and cytolytic responses and make little IL-2 after stimulation through the TCR. Although thymic development appears normal in these mice, peripheral lymphoid organs, particularly lymph nodes, are relatively acellular. In a separate transgenic line, forced expression of the gene specifically in the thymus can alter thymic development; however, this does not appear to affect peripheral T cells and is unable to prevent disease in mice lacking a functional Foxp3 gene, indicating that the scurfin protein acts on peripheral T cells. The data indicate a critical role for the Foxp3 gene product in the function of the immune system, with both the number and functionality of peripheral T cells under the aegis of the scurfin protein.

Cellular and molecular characterization of the scurfy mouse mutant.

Mice hemizygous (Xsf/Y) for the X-linked mutation scurfy (sf) develop a severe and rapidly fatal lymphoproliferative disease mediated by CD4+CD8- T lymphocytes. We have undertaken phenotypic and functional studies to more accurately identify the immunologic pathway(s) affected by this important mutation. Flow cytometric analyses of lymphoid cell populations reveal that scurfy syndrome is characterized by changes in several phenotypic parameters, including an increase in Mac-1+ cells and a decrease in B220+ cells, changes that may result from the production of extremely high levels of the cytokine granulocyte-macrophage CSF by scurfy T cells. Scurfy T cells also exhibit strong up-regulation of cell surface Ags indicative of in vivo activation, including CD69, CD25, CD80, and CD86. Both scurfy and normal T cells are responsive to two distinct signals provided by the TCR and by ligation of CD28; scurfy cells, however, are hyperresponsive to TCR ligation and exhibit a decreased requirement for costimulation through CD28 relative to normal controls. This hypersensitivity may result, in part, from increased costimulation through B7-1 and B7-2, whose expression is up-regulated on scurfy T cells. Although the specific defect leading to this hyperactivation has not been identified, we also demonstrate that scurfy T cells are less sensitive than normal controls to inhibitors of tyrosine kinases such as genistein and herbimycin A, and the immunosuppressant cyclosporin A. One interpretation of our data would suggest that the scurfy mutation results in a defect, which interferes with the normal down-regulation of T cell activation.

Impaired expansion of mouse B cell progenitors lacking Btk.

Mutations in the gene encoding the protein tyrosine kinase Btk are associated with the human B cell immunodeficiency X-linked agammaglobulinemia (XLA). In the mouse, a point mutation in the Btk pleckstrin homology domain segregates with a milder X-linked immunodeficiency (xid). To assess the importance of Btk function in murine lymphopoiesis, we generated multiple embryonic stem cell clones bearing a targeted disruption of the btk gene and examined their potential to produce lymphocytes in both C57BL/6 and RAG2-/- host chimeric animals. These mice provide a complementary set of in vivo competition assays that formally establish the genetic basis for the xid phenotype. Although the null mutation yields a phenotype quite similar to that of xid, it also compromises expansion of B cell precursors. Our results suggest that the murine and human consequences of Btk deficiency differ only quantitatively, and represent the same disease process.

Involvement of p59fynT in interleukin-5 receptor signaling.

Previous studies implicate the nonreceptor protein tyrosine kinase (PTK) p59fyn in the propagation of signals from the B cell antigen receptor. To elucidate the functions of this kinase, we examined B cell responsiveness in mice engineered to lack the hematopoietic isoform of p59fyn. Remarkably, antigen receptor signaling was only modestly defective in fynTnull B cells. In contrast, signaling from the interleukin (IL)-5 receptor which ordinarily provides a comitogenic stimulus with antiimmunoglobulin, was completely blocked. Our results document the importance of p59fynT in IL-5 responses in B cells, and they support a general model for cytokine receptor signal transduction involving the simultaneous recruitment of at least three families of PTK.

Control of lymphopoiesis by p50csk, a regulatory protein tyrosine kinase.

The csk gene encodes a nonreceptor protein tyrosine kinase that acts in part by regulating the activity of src-family protein tyrosine kinases. Since the src-family kinases p56lck and p59fyn play pivotal roles during lymphocyte development, it seemed plausible that p50csk might contribute to these regulatory circuits. Using a gene targeting approach, mouse embryonic stem cell lines lacking functional csk genes were generated. These csknull embryonic stem cells proved capable of contributing to many adult tissues, notably heart and brain. However, although csknull progenitors colonized the developing thymus, T and B cell differentiation were both blocked at very early stages. This represented a relatively selective interdiction of lymphocyte maturation, since csknull hematopoietic progenitors supported the development of normal-appearing MAC-1+ blood leukocytes, and the successful maturation of granulocyte/macrophage-colony-forming units from fetal liver progenitors. We conclude that p50csk regulates normal lymphocyte differentiation, but that it almost certainly does so by acting on targets other than p56lck and p59fyn.

Developmental regulation of TCR-CD3-dependent [Ca2+]i responses of individual normal and pp59fyn-deficient T lymphocytes.

The aim of this study was to ascertain whether different types of T-cell receptor (TCR)-mediated [Ca2+]i signals could begin to explain the different cellular responses of mature and immature T cells to ligation of the TCR-CD3 complex. Using a digital fluorescence imaging system, we measured and compared [Ca2+]i of individual cells from immature and mature murine T-cell populations following application of CD3-epsilon monoclonal antibody (mAb). Our approach revealed distinctions among developmental subsets which were not seen by previous measurements of [Ca2+]i in bulk cell populations. The CD3-mediated [Ca2+]i responses of individual thymocytes were very complex. Latencies to peak [Ca2+]i varied greatly among thymocytes, but the responses of splenic T cells were synchronized, novel evidence that the timing of [Ca2+]i responses may be an important informative parameter for TCR-CD3 signalling. In addition, among cells responding to CD3 mAb, higher peak [Ca2+]i responses correlated with maturity (CD4+ CD8+ thymocytes < single-positive thymocytes < splenic T cells). Examination of cells from pp59fyn-deficient mice showed that pp59fyn deficiency affects the amplitude and probability, but not the latency or synchrony, of CD3-mediated [Ca2+]i responses of CD4+ CD8+ and CD4+ CD8- thymocytes. All subsets showed equivalent receptor-independent mobilization of [Ca2+]i. These developmentally distinct [Ca2+]i features most probably reflect meaningful developmental changes in how the TCR-CD3 complex couples to intracellular signalling machinery including pp59fyn. By clearly showing how [Ca2+]i responses change during development, these results support the hypothesis that distinctive types of [Ca2+]i responses drive thymocyte differentiation.

Involvement of p21ras distinguishes positive and negative selection in thymocytes.

Small molecular weight GTP binding proteins of the ras family have been implicated in signal transduction from the T cell antigen receptor (TCR). To test the importance of p21ras in the control of thymocyte development, we generated mice expressing a dominant-negative p21ras protein (H-rasN17) in T lineage cells under the control of the lck proximal promoter. Proliferation of thymocytes from lck-H-rasN17 mice in response to TCR stimulation was nearly completely blocked, confirming the importance of p21ras in mediating TCR-derived signals in mature CD4+8- or CD8+4- thymocytes. In contrast, some TCR-derived signals proceeded unimpaired in the CD4+8+ thymocytes of mice expressing dominant-negative p21ras. Analysis of thymocyte development in mice made doubly transgenic for the H-Y-specific TCR and lck-H-rasN17 demonstrated that antigen-specific negative selection occurs normally in the presence of p21H-rasN17. Superantigen-induced negative selection in vivo also proceeded unhindered in H-rasN17 thymocytes. In contrast, positive selection of thymocytes in the H-Y mice was severely compromised by the presence of p21H-rasN17. These observations demonstrate that positive and negative selection, two conceptually antithetical consequences of TCR stimulation, are biochemically distinguishable.

Regulation of lymphocyte function by protein phosphorylation.

Variations in protein phosphorylation provide the predominant means of enzymatic regulation now known in biological systems, especially in the regulation of signal transduction from cell surface receptors. Analysis of these signaling pathways has proceeded especially rapidly in lymphocytes, in part because these cells can be isolated with relative ease and can in many cases be maintained in vitro for prolonged periods as clonal populations. During the past few years, both biochemical and genetic evidence has been adduced indicating that the antigen receptors of T and B lymphocytes associate functionally with nonreceptor protein tyrosine kinases. Similar data implicate protein tyrosine kinases in signaling from the CD4 and CD8 coreceptors and the beta chain of the IL-2 receptor. Protein serine/threonine kinases and several different phosphatases also participate in the intracellular propagation of antigen receptor-derived signals. Here we review the lymphocyte surface receptors that are believed to act by altering protein phosphorylation, the kinases and phosphatases that are believed to regulate signal transduction in lymphocytes, and the implications of these results for the broader study of cell signaling mechanisms.

Defective T cell receptor signaling in mice lacking the thymic isoform of p59fyn.

Considerable evidence supports the hypothesis that the nonreceptor protein tyrosine kinase p59fyn participates in signal transduction from the T cell receptor (TCR). To examine this hypothesis in detail, we have produced mice that lack the thymic isoform of p59fyn but retain expression of the brain isoform of the protein. fynTnull mice exhibit a remarkably specific lymphoid defect: thymocytes are refractile to stimulation through the TCR with mitogen or antigen, while peripheral T cells, following what appears to be a normal maturation sequence, reacquire significant signaling capabilities. These data confirm that p59fynT plays a pivotal role in TCR signal transduction and demonstrate that additional developmentally regulated signaling components also contribute to TCR-induced lymphocyte activation.

In vivo and in vitro effects of v-fos and EJ-Ha-ras oncogene expression in murine epidermal keratinocytes.

Primary neonatal Balb/c keratinocyte (NEK) cultures grown, using 3T3 feeder cell support in high calcium, serum supplemented medium, were transfected with EJ-Ha-ras or v-fos DNA sequences and pSV2 neo. Several neo resistant clones were isolated and several established cell lines expressing the transfected gene products derived. Two of these lines, Ras 8 and Fos 1, have been examined in detail with respect to their self renewal capacity and differentiation potential in vitro and in vivo. In vitro, both lines (when compared to normal NEK) have an extended probably immortal phenotype, enhanced colony forming efficiency (a measure of in vitro self renewal capacity) and a reduction in growth factor and serum dependence. When grafted onto syngeneic recipients neither cell line is tumourigenic. Instead a histologically abnormal epithelium with no stratum corneum and with features specific to the oncogene expressed is formed. The extent of the histological atypia correlates with the in vitro alterations in cytoskeletal peptides as revealed by 2D PAGE. However despite the gross histological abnormality there is no alteration in the in vivo self renewal capacity (measured as the number of grafted cells required for epidermal reformation) between normal NEK and the Ras 8 or Fos 1 lines; in each case a minimum of 10(5) cells/1.14 cm2 is required before a full thickness epithelium forms.

The effect of donor age on the proliferative response of human and mouse keratinocytes to phorbol, 12-myristate, 13-acetate.

Epidermal keratinocytes grow clonally when provided with a 3T3 feeder layer and medium supplemented with 20% foetal bovine serum, hydrocortisone and cholera toxin. In this culture system the proliferative response of freshly isolated human and mouse keratinocytes to a short exposure (24 h) of the tumour promoter phorbol, 12-myristate, 13-acetate (PMA) was dependent on the donor age but was independent of the species or the biopsy site. Human keratinocytes from early (16-18 week) foetal donors displayed a strong proliferative response to PMA as determined by a 5- to 7-fold increase in colony number and an increase in the average colony size. In contrast, adult and juvenile keratinocytes of all ages from both mice and humans displayed an inhibition of colony forming efficiency and a reduction in colony size. When continuously passaged in culture human foetal keratinocytes gradually changed the pattern of their response to PMA so that they were inhibited from growing rather than being stimulated and after 21 days (three passages) their response was quantitatively similar to juvenile keratinocytes. Co-culture of juvenile keratinocytes with irradiated foetal keratinocytes did not alter their response to PMA so that the observed proliferative response of foetal keratinocytes to PMA is not readily explained by the autosecretion of mitogens or other regulatory molecules by these cells. Late foetal (17 days gestation), neonatal and post-neonatal (5-10 days old) mouse keratinocytes were also inhibited from growing by PMA but the magnitude of the effect was directly related to the age of the mouse and was in all cases less than that observed with adult mice. The relationship of these results to the mechanism of action of phorbol esters in epidermis is discussed.