Conventional CD4+ T cells regulate IL-22-producing intestinal innate lymphoid cells.

The innate and adaptive immune systems in the intestine cooperate to maintain the integrity of the intestinal barrier and to regulate the composition of the resident microbiota. However, little is known about the crosstalk between the innate and adaptive immune systems that contribute to this homeostasis. We find that CD4+ T cells regulate the number and function of barrier-protective innate lymphoid cells (ILCs), as well as production of antimicrobial peptides (AMPs), Reg3γ and Reg3ß. RAG1-/- mice lacking T and B cells had elevated ILC numbers, interleukin-22 (IL-22) production, and AMP expression, which were corrected by replacement of CD4+ T cells. Major histocompatibility class II-/- (MHCII-/-) mice lacking CD4+ T cells also had increased ILCs, IL-22, and AMPs, suggesting that negative regulation by CD4+ T cells occurs at steady state. We utilized transfers and genetically modified mice to show that reduction of IL-22 is mediated by conventional CD4+ T cells and is T-cell receptor dependent. The IL-22-AMP axis responds to commensal bacteria; however, neither the bacterial repertoire nor the gross localization of commensal bacteria differed between MHCII+/- and MHCII-/- littermates. These data define a novel ability of CD4+ T cells to regulate intestinal IL-22-producing ILCs and AMPs.

Recipient MHC class II expression is required to achieve long-term survival of murine cardiac allografts after costimulatory blockade.

To study the role of the direct and indirect pathways in achieving tolerance, we used genetically altered mouse strains in two ways: 1) MHC class II-deficient mice were used as donors of skin and cardiac grafts to eliminate the direct CD4(+) T cell response, and 2) B6 II(-)4(+) mice, which are MHC class II-deficient mice expressing an MHC class II transgene only on thymic epithelium, were used as recipients of normal grafts. These mice cannot mount an indirect response. Eliminating the indirect pathway actually made it more difficult to achieve prolonged allograft survival when we used costimulatory blockade than when both pathways were available. Costimulatory blockade was ineffective even when CD4(+) T cells from normal animals were transferred into recipients that lacked MHC class II molecules. These results suggest that an active CD4(+) response through the indirect pathway is necessary for costimulatory blockade to be effective in prolonging allograft survival.

Major histocompatibility complex class II-positive cortical epithelium mediates the selection of CD4(+)25(+) immunoregulatory T cells.

CD4(+)25(+) T cells are a unique population of immunoregulatory T cells which are critical for the prevention of autoimmunity. To address the thymic selection of these cells we have used two models of attenuated thymic deletion. In K14-A(beta)(b) mice, major histocompatibility complex (MHC) class II I-A(b) expression is limited to thymic cortical epithelium and deletion by hematopoietic antigen-presenting cells does not occur. In H2-DMalpha-deficient mice, MHC class II molecules contain a limited array of self-peptides resulting in inefficient clonal deletion. We find that CD4(+)25(+) T cells are present in the thymus and periphery of K14-A(beta)(b) and H2-DMalpha-deficient mice and, like their wild-type counterparts, suppress the proliferation of cocultured CD4(+)25(-) effector T cells. In contrast, CD4(+)25(+) T cells from MHC class II-deficient mice do not suppress responder CD4(+) T cells in vitro or in vivo. Thus, development of regulatory CD4(+)25(+) T cells is dependent on MHC class II-positive thymic cortical epithelium. Furthermore, analysis of the specificities of CD4(+)25(+) T cells in K14-A(beta)(b) and H2-DMalpha-deficient mice suggests that a subset of CD4(+)25(+) T cells is subject to negative selection on hematopoietic antigen-presenting cells.

A minimal level of MHC class II expression is sufficient to abrogate autoreactivity.

The establishment of CD4(+) T cell tolerance requires that self-reactive thymocytes are negatively selected during thymic development. A threshold of antigen concentration appears to exist for both MHC class I- and class II-mediated negative selection, below which clonal deletion of a self-reactive transgenic TCR does not occur. Similarly, both the specificity and thymic concentration of MHC molecules affect the efficiency with which autoreactive thymocytes are deleted. However, this threshold for MHC class II concentration has not been well established. Here, we show that this threshold must be extraordinarily low. We have used the human lysozyme promoter to re-express an A(beta)(b) cDNA on macrophages and other phagocytic myelomonocytic cells of class II-deficient A(beta)(b) -/- mice. Surface expression of I-A(b) could be detected on mature peritoneal macrophages and, minimally, on thymic dendritic cells; however, this level of expression was not sufficient for antigen-specific T cell activation. Nevertheless, when backcrossed onto an autoreactive K14 background, this minimal level of class II was sufficient to induce negative selection of a polyclonal self-reactive population. We conclude that provision of extremely low levels of class II to thymic dendritic cells confers on them the ability to mediate clonal deletion of autoreactive T cells.

Self-reactive T cells selected on thymic cortical epithelium are polyclonal and are pathogenic in vivo.

Positive selection of CD4+ T cells requires that the TCR of a developing thymocyte interact with self MHC class II molecules on thymic cortical epithelium. In contrast, clonal deletion is mediated by dendritic cells and medullary epithelium. We previously generated K14 mice expressing MHC class II only on thymic cortical epithelium. K14 CD4+ T cells were positively, but not negatively, selected and had significant in vitro autoreactivity. Here, we examine the function of these autoreactive CD4+ T cells in more detail. Analysis of a series of K14-derived T hybrids demonstrated that the autoreactive population of CD4+ T cells is phenotypically and functionally diverse. Purified K14 CD4+ T cells transferred into lethally irradiated wild-type B6 mice cause acute graft vs host disease with bone marrow failure. Further, these autoreactive CD4+ T cells cause hypergammaglobulinemia and the production of autoantibodies when transferred into unirradiated wild-type hosts. Thus, positive selection by normal thymic cortical epithelial cells, unopposed by negative selection, produces polyclonal CD4+ T cells that are pathologic.

Using thymus anatomy to dissect T cell repertoire selection.

The thymic development of CD4(+) T cells incorporates the opposing processes of positive and negative selection to produce mature lymphocytes which respond to foreign peptides in the context of self-major histocompatibility complex class II molecules. Here, we present a model in which these events occur in two temporally and anatomically distinct steps. We propose that an initial positive selection step is mediated exclusively by thymic cortical epithelium. Subsequently, all those thymocytes which have been positively selected will interact with medullary epithelium and bone marrow-derived cells. Those thymocytes reacting with excess affinity or avidity to these antigen presenting cells will be negatively selected. Although acknowledging the importance of differential signalling to the developing thymocyte, we will emphasize the centrality of the phenotype of the tissues which comprise the thymus.

The strength of cell-mediated xenograft rejection in the mouse is due to the CD4+ indirect response.

Previous studies have shown that CD4+ T cells are responsible for the great strength of cell-mediated xenograft rejection in the mouse. In vitro studies have suggested that this CD4+ response is to xenogeneic antigens that are presented indirectly. The present studies were carried out in order to determine whether the strength of cell-mediated xenograft rejection in vivo is dependent on the CD4+ indirect response. We grafted pig skin onto mice that express class II MHC antigens only on their thymic epithelial cells (II-4+ mice). These mice have normal numbers of functional peripheral CD4+ T cells; however they lack class II MHC expression on their antigen presenting cells and are thus incapable of mounting a CD4+ T cell-mediated indirect response. Xenograft survival was prolonged on these mice. Furthermore, administration of cyclosporine and anti-CD8 monoclonal antibodies to II-4+ recipients prolonged xenograft survival to at least the same extent as allograft survival, demonstrating that the strength of cell-mediated xenograft rejection resides in the CD4+ indirect response. Despite the increased survival time, xenograft rejection still occurred in the absence of the indirect pathway. Depletion of the II-4+ recipients of their CD4+ T cell population prolonged xenograft survival even further, suggesting the presence of a weaker CD4+ direct mechanism which was virtually undetectable in vitro.

CD8+ effector cells responding to residual class I antigens, with help from CD4+ cells stimulated indirectly, cause rejection of "major histocompatibility complex-deficient" skin grafts.

BACKGROUND: Skin grafts from mice that are deficient in the expression of both class I and class II major histocompatibility complex (MHC) antigens are rejected rapidly by normal recipients. METHODS: To determine the mechanism of this rejection, MHC-deficient skin grafts were placed on recipients with different degrees of antigenic disparity and on recipients depleted of selected T cell subpopulations. In addition, the recipient's T cells were examined in vitro for their responses before and after graft rejection. RESULTS: The results indicate that (1) CD4+ cells provide help for this rejection by recognizing donor antigens presented by recipient class II antigens, and (2) CD8+ cells can participate as effector cells, recognizing residual class I antigens expressed by the MHC-deficient grafts. CONCLUSIONS: The primary conclusion from these studies is that the supposedly MHC-deficient mice actually do have sufficient class I antigen expression to cause skin graft rejection. This finding prevents the use of these mice to answer definitively the question of whether grafts entirely lacking MHC antigens would be rejected. However, these studies do illustrate two important (although previously recognized) features of allogeneic skin graft rejection: (1) that rejection can be initiated by help provided entirely through the indirect pathway, and (2) that help provided through the indirect pathway is available for effector T cells sensitized directly by donor cells. However, the results from these and other studies suggest that indirect effector mechanisms would probably be able to destroy truly MHC-deficient grafts under some circumstances.

Thymic stromal cell specialization and the T-cell receptor repertoire.

Ten years ago, we proposed a model for thymus function in which thymic epithelial cells are primarily responsible for imprinting major histocompatibility complex (MHC)-restricted specificity, and bone marrow-derived macrophages or dendritic cells are responsible for the induction of self-tolerance. Since then, transgenic and knockout models have allowed for a dissection of thymic stromal components in vivo, leading to a new understanding of their specialized functions. We have determined that with regard to class II-restricted CD4 T-cell development, two distinct subsets of thymic epithelium help shape the repertoire: Cortical epithelium appears solely responsible for positive selection, whereas a fucose-bearing subset of medullary epithelium is specialized for negative selection. This absolute separation of positive and negative selection into two distinct spatial and temporal compartments leads to a much simpler view of the process of repertoire selection. Finally, a novel view of the function of the thymic medulla is discussed.

Single amino acid mutations in the murine MHC class II A beta cytoplasmic domain abrogate antigen presentation.

Class II MHC molecules are heterodimeric transmembrane glycoproteins that function in the presentation of Ag to CD4+ T cells. Deletion of the cytoplasmic domains of the murine class II A alpha- and A beta-chains has previously been shown to diminish Ag presentation and abrogate rejection of class II-transfected tumor cells. To examine the contributions of individual amino acid residues of the A beta cytoplasmic domain to Ag presentation and tumor rejection, we have produced a series of cell lines expressing A beta class II molecules with site-directed mutations. An A beta(k) cDNA was constructed with mutations in the five conserved amino acid residues, Q224, K225, L235, L236, and Q237 (delta5). In addition, cDNA were produced in which alanine was individually substituted for A beta(k) cytoplasmic domain residues 224 through 237 or doubly substituted at residues G226 and P227 or L235 and L236. These mutant cDNAs were individually cotransfected with wild-type A alpha cDNA into the class II-negative M12.C3 B lymphoma and Sal sarcoma cell lines. As was previously reported for transfectants lacking the entire A beta(k) cytoplasmic domain, the delta5 M12.C3 transfectant could not effectively present Ag to an autoreactive Ak-restricted T cell hybrid, and the delta5 Sal transfectant was not rejected when inoculated into syngeneic hosts. A finer analysis revealed that alteration of the individual residue Q224 or the two residues G226 and P227 abrogated Ag presentation in vitro, while mutation of G226 diminished tumor rejection in vivo. Thus, the function of the A beta cytoplasmic domain in Ag presentation both in vitro and in vivo can be disturbed by mutation of single amino acid residues.

Unopposed positive selection and autoreactivity in mice expressing class II MHC only on thymic cortex.

The normal development of T cells in the thymus requires both positive and negative selection. During positive selection, thymocytes mature only if their T-cell receptors react with some specificity to host major histocompatibility complex (MHC) and host peptides. During negative selection, thymocytes die if their T-cell receptors react with too high an affinity to the presenting cell, self MHC, and peptides to which they are exposed. These two processes are important for the development of the T-cell repertoire and the acquisition of self-tolerance, but their precise location and temporal relationship are not known. We have used the keratin 14 (K14) promoter to re-express a class II MHC antigen (I-Ab) in class II-negative mice. The transgenic I-A molecule is expressed only on thymic cortical epithelium; thymic medullary epithelium and bone-marrow-derived cells are I-A negative. CD4+ cells are positively selected in K14 mice, but clonal deletion does not ocur in K14 mice or in relB-negative mice, which lack a thymic medulla. The K14 CD4 cells are autoreactive, as they proliferate extensively to and specifically lyse I-Ab-positive target cells. These autoreactive cells make up 5% of the peripheral CD4 T cells, providing and estimate of the minimal frequency of positively selected cells that must subsequently undergo negative selection for self-tolerance to be preserved. Thus positive and negative selection occur in anatomically distinct sites.

Two levels of help for B cell alloantibody production.

We have examined whether T cell stimulation by direct or indirect pathways contributes to alloantibody production by B cells after major histocompatibility complex (MHC)-disparate skin graft rejection in mice. Experiments were performed using normal mice, MHC class II-deficient mice, MHC class II-deficient mice with an intact peripheral CD4+ cell population (due to expression of class II antigens only on thymic epithelium), mice lacking the cytoplasmic tail of their MHC class II antigens, and mice depleted of CD4+ cells by anti-CD4 monoclonal antibody treatment. Depletion of recipient CD4+ cells reduced alloantibody production to barely detectable levels. Absence of donor MHC class II antigens did not affect the production of either immunoglobulin (Ig)M or IgG antibodies directed at class I alloantigens. Absence of recipient MHC class II antigens, however, led to production of only IgM but not IgG antibodies, even if the recipients had an intact CD4+ cell population. Absence of the cytoplasmic tail of the recipient's MHC class II antigens led to the production of slightly reduced amounts of IgG antibody. These findings indicate that (a) CD4+ cells are essential helper cells for B cell alloantibody production; (b) production of IgM alloantibody can occur with help from CD4+ cells, which recognize either donor class II antigens or modified recipient class II antigens; (c) isotype switching from IgM to IgG alloantibody requires help from CD4+ cells activated by antigens presented by recipient MHC class II molecules; and (d) the cytoplasmic domain of the recipient MHC class II molecules may be involved in the mechanism that leads to isotype switching by B cells. Thus, there are two levels of CD4-mediated help available for B cells responding to alloantigens: one (involving a noncognate interaction) can produce B cell activation, and a second (involving a cognate interaction) is required for differentiation and IgG alloantibody production.

Transgenic mice expressing MHC class II molecules with truncated A beta cytoplasmic domains reveal signaling-independent defects in antigen presentation.

The thymic development and peripheral activation of CD4+ T cells are critically dependent upon interactions with MHC class II molecules on the surface of antigen presenting cells (APC). In vitro studies involving transfection of cell lines with mutant MHC molecules have demonstrated that the cytoplasmic domains of class II molecules can be required for efficient antigen presentation. To address the role of class II cytoplasmic domains in physiological, non-transformed APC and in vivo immune responses, we have generated transgenic mice which express only truncated class II A beta molecules lacking the 13 membrane distal residues. In vivo, CD4+ T cell development and immune responses to conventional protein antigens, parasitic infections and skin grafts were indistinguishable between control and transgenic mice. Nevertheless, in vitro, APC from transgenic mice poorly stimulate T cell hybridomas and wild-type in vivo-primed T cells. Neither class II-mediated induction of B7-1 expression nor homotypic aggregation were diminished in transgenic B cells, suggesting that both cAMP and tyrosine kinase signaling pathways remain intact despite truncation of the A beta cytoplasmic domain. Furthermore, chemically-fixed cells from transgenic animals are impaired in their antigen presenting capacity. Thus, in contrast to previous studies with cell lines transfected with truncated class II molecules, these results suggest that signaling-independent mechanisms account for the defective in vitro antigen presenting capacity of physiological APC expressing truncated A beta proteins.

Autoimmune diabetes can be induced in transgenic major histocompatibility complex class II-deficient mice.

Insulin-dependent diabetes mellitus (IDDM) is an autoimmune disease marked by hyperglycemia and mononuclear cell infiltration of insulin-producing beta islet cells. Predisposition to IDDM in humans has been linked to the class II major histocompatibility complex (MHC), and islet cells often become aberrantly class II positive during the course of the disease. We have used two recently described transgenic lines to investigate the role of class II molecules and CD4+ T cells in the onset of autoimmune insulitis. Mice that are class II deficient secondary to a targeted disruption of the A beta b gene were bred to mice carrying a transgene for the lymphocytic choriomenigitis virus (LCMV) glycoprotein (GP) targeted to the endocrine pancreas. Our results indicate that class II-deficient animals with and without the GP transgene produce a normal cytotoxic T lymphocyte response to whole LCMV. After infection with LCMV, GP-transgenic class II-deficient animals develop hyperglycemia as rapidly as their class II-positive littermates. Histologic examination of tissue sections from GP-transgenic class II-deficient animals reveals lymphocytic infiltrates of the pancreatic islets that are distinguishable from those of their class II-positive littermates only by the absence of infiltrating CD4+ T cells. These results suggest that in this model of autoimmune diabetes, CD4+ T cells and MHC class II molecules are not required for the development of disease.

Identification of phorbol ester receptors in T-cell growth factor-producing and -nonproducing EL4 mouse thymoma cells.

Two lines of EL4 mouse thymoma cells, one which responds to phorbol ester tumor promoters with production of T-cell growth factor and inhibition of proliferation and one which does not respond, have been examined for the presence of specific phorbol ester-binding components. Both lines displayed saturable specific binding as determined by using [20-3H]phorbol 12,13-dibutyrate in a whole-cell-binding assay. Specific binding in each line was maximal within 10 min at 37 degrees but rapidly decreased to about 30% within 30 min. Maximal binding at 4 degrees was reached after 50 min and was stable for at least 8 hr; however, this level was only about 30% of the maximum obtained at 37 degrees. Saturation of the specific binding after 3 hr at 4 degrees obtained at a concentration (30 to 50 nM) which is consistent with that yielding maximal T-cell growth factor production in the responding line but greater than that at which inhibition of proliferation was detected in those cells. Scatchard analysis of these data was consistent with the existence of a single class of binding sites with a Kd of 11 +/- 0.6 (S.D.) nM for the T-cell growth factor-producing cells and 18 +/- 1.9 nM for the nonproducing line. The numbers of sites per cell measured in the responding and nonresponding lines were 5.6 +/- 1.3 x 10(4) and 7.1 +/- 0.2 x 10(4), respectively. Competition by a series of phorbol esters for [20-3H]phorbol 12,13-dibutyrate binding in the responding line showed the same order of potency as production of T-cell growth factor and inhibition of proliferation by the analogs. These data support identification of the phorbol ester-binding component in the responding cells as the receptor mediating T-cell growth factor production. The presence of binding components in unresponsive cells in numbers and with characteristics indistinguishable from those of responsive cells suggests that the failure to respond is due to modification of a step which succeeds the initial binding event.