Interleukin-1 and interleukin-2 defects associated with murine graft-versus-host-induced immunodeficiency

Prostaglandin E2 production by chronic graft-versus-host disease dermal fibroblasts

Chronic graft versus host disease (cGVHD) across minor histocompatibility barriers is associated with the development of cutaneous fibrosis, disappearance of mast cells and immunosuppression. The idea, which has been the basis of our previous and present studies, is that fibroblasts are not only a target for modulation in cGVHD, but also have effector roles in this condition. In the present study we investigated the production of prostaglandin E2 (PGE2) and of collagen by cultured dermal fibroblasts obtained from cGVHD and control mice. Early in the development of the disease (Day 8) cGVHD fibroblasts generated constitutively more PGE2 (3-fold) than did control fibroblasts. Thereafter, PGE2 production declined to near normal levels by Day 20 post cGVHD induction. On the other hand, at this time point cGVHD fibroblasts displayed an enhanced synthesis of collagen as compared to the control fibroblasts and to earlier time points. Therefore, PGE2 synthesis appears to inversely correlate with collagen synthesis by cGVHD fibroblasts. We propose that fibroblasts may contribute to the development of immunosuppression, which characterizes the early phase of cGVHD.

Macrophage synthesis of nitric oxide in the mouse mixed leucocyte reaction

The production of nitric oxide (.N = O) in the splenocyte mixed leucocyte reaction (MLR) results in inhibition of allospecific lymphocyte effector function. In order to more clearly define the circumstances which promote .N = O synthesis in the MLR, responder accessory cell depleted spleen cells (ACDSC) were co-cultured with allogeneic macrophage cell lines or peritoneal macrophages. .N = O synthesis and C57BL/6 (H-2b) ACDSC proliferation were concurrently monitored in cultures comparing RAW 264.7 (H-2d, a high .N = O producer), P388D1 (H-2d, a low .N = O and BALB/c (H-2d) peritoneal macrophages as allogeneic antigen presenting cells (APC). A concentration-dependent increase in lymphocyte proliferation was observed in the presence of 1 x 10(4) to 1 x 10(5) P388D1. In contrast, addition of NG-monomethyl-L-arginine (NMA), a competitive inhibitor of .N = O synthase, was necessary in order to observe lymphocyte proliferation in the presence of increasing numbers of RAW 264.7 and BALB/c peritoneal macrophages. The addition of both anti-IL-2 and anti-IFN gamma (interferon-gamma) monoclonal antibodies inhibited .N = O synthesis in alloantigen-stimulated cultures. The IFN gamma induced expression of class II antigen, as well as the constitutive expression of class I antigen, on RAW 264.7 was similar in the presence or absence of NMA, indicating that induction of .N = O synthesis by IFN gamma does not inhibit H-2 antigen expression. Thus, cytokines produced as a result of alloimmune interaction initiate macrophage .N = O synthesis. However, allogeneic APC function, as assessed by H-2 antigen expression and subsequent stimulatory capacity of MLR, is not affected by initiation of the .N = O pathway.

Thymic selection and thymic major histocompatibility complex class II expression are abnormal in mice undergoing graft-versus-host reactions

The graft-vs.-host reaction (GVHR) results in damage to the epithelial and lymphoid compartments of the thymus and thus in abnormal maturation and function of thymocytes in mice undergoing GVHR. In this report, the effects of GVHR on thymic T cell receptor (TCR) expression and usage have been investigated. GVHR was induced in unirradiated F1 hybrid mice by the intravenous transfer of parental lymphoid cells. Expression of the CD3/TCR complex on thymocyte subsets defined by CD4 and CD8 was studied by three-color flow cytometry. The level of CD3/TCR was decreased on CD4+CD8-, but not CD4-CD8+, mature thymocytes. The lack of upregulation of CD3/TCR on CD4 single-positive thymocytes, but not on their CD8+ counterparts, suggested an abnormality of class II major histocompatibility complex (MHC) expression in the thymuses of mice undergoing GVHR. Immunofluorescence staining of thymic frozen sections revealed that MHC class II expression was dramatically decreased in GVH-reactive mice. GVHR-induced changes in positive and negative selection were evaluated by determining the incidence of specific V beta TCR segment usage in the thymus. In normal mice, thymocyte usage of any given V beta segment was highly consistent between individuals of the same strain and age; however, a marked divergence in the incidence of TCR V beta 6hi and V beta 8hi cells between GVH-reactive littermate mice was observed, suggesting that thymic positive selection had become disregulated in these animals. Furthermore, negative selection was defective; the incidence of phenotypically self-reactive V beta 6hi T cells was significantly greater in the thymuses of GVH-reactive mice bearing the endogenous superantigen Mls-1a than in untreated controls. Thus, mice undergoing GVHR showed defective TCR upregulation on CD4+CD8- thymocytes and changes in TCR usage reflecting aberrant thymic selection, in conjunction with decreased expression of MHC class II. Most abnormalities of TCR expression and usage on CD4+ thymocytes observed in GVH-reactive mice were analogous to those of class II knockout mice.

Regulation of graft-versus-host-reaction by Mlsa-reactive donor T cells

Injection of A/J splenocytes (H-2Dd, Mlsc) into unirradiated (A/J x CBA)F1 (BAF1) host mice (H-2Dd/k, Mlsd) results in an acute suppressive graft-vs.-host reaction (GVHR), characterized by immune dysfunction and appreciable donor cell engraftment; injection of the CBA/J parent (H-2Dk, Mlsd), which recognizes no Mls disparity in the host, results in little or no GVHR. Furthermore, the Mlsa-reactive V beta 6 and V beta 8.1 T cell subsets in A/J T cells expand significantly in the GVHR host. Finally, depletion of V beta 6+ and V beta 8.1+ from the A/J population abrogates the proliferative response to BAF1 in vitro and the development of GVHR in vivo. Thus, the response to Mls determinants can contribute to the generation of a GVHR.

Antitumor effects of endotoxin against solid murine Meth A tumors of different ages. Quantitative histology of the tumors and regional lymph nodes

Mice with 3-, 6-, 9- and 15-day Meth A tumors in the skin were injected intravenously with endotoxin to study age-dependent induction of tumor necrosis and reactive changes in draining lymph nodes. By 24 h after treatment with endotoxin, macroscopic necrosis was seen in 9-day, and to an even greater extent in 15-day tumors; microscopy showed extensive necrosis in 9-day tumors and some necrosis in 6-day tumors. The necrosis was predominantly coagulative, but the 9- and 15-day tumors showed a rim of hemorrhagic necrosis near the skin with surviving tumor tissue located at the lateral and basal margins. All the tumors showed about equal hyperemia by 4 h and mitotic arrest by 4 and 24 h. Depletion of mast cells, which were most numerous in the 9-day tumors, was seen on the dermal aspects by 24 h. Endotoxin did not change the morphology of the lymph nodes and disseminated tumor cells within 24 h. Endotoxin is known to cure 9-day Meth A tumors, and the extent of the rapid necrosis induced is clearly the crucial factor. Necrosis, however, is not a direct consequence of early hyperemia or mitotic arrest and other factors related to the age and site of the tumor, apparently affect whether or not necrosis ensues.

Association between the degree of thymic dysplasia and the kinetics of thymic NK cell activity during the graft-versus-host reaction

In this study, by employing different cell doses and parent into F1 hybrid combinations, we have investigated the relationship between the severity of thymic medullary dysplasia and the kinetics of thymic natural killer (NK) cell activity after the induction of graft-versus-host (GVH) reactions. GVH reactions were induced by injecting different doses (30, 20, 10 X 10(6] of C57BL/6 (B6) of A parental lymphoid cells (PLC) into non-X-irradiated adult B6xAF1 (B6AF1) mice. On different days after the induction of GVH reactions, the thymuses were examined histologically and thymocyte NK cell activity was tested by using YAC targets. Our results show that, depending upon the genotype and dose of PLC injected, various degrees of thymic medullary dysplasia (mild, moderate, or severe) can be induced. Furthermore, severe to moderate thymic medullary dysplasia is observed only in those groups of GVH-reactive mice in which thymic NK cell activity occurs early and increases rapidly. In contrast, when mild thymic medullary dysplasia or no thymic alterations was observed, thymic NK cell activity peaked later and was of lower intensity than that of the groups with moderate to severe lesions. These results suggest an association between the degree of thymic medullary dysplasia and the kinetics of NK cell activity in the thymus. Furthermore, the different degrees of thymic medullary dysplasia as described here may serve as a powerful tool to study the role of thymic medullary dysplasia in determining the duration of T-cell immunodeficiency associated with the GVH reactions.

Inhibition of antisuppression by the acute murine graft-versus-host reaction

Profound suppression of both humoral and cell-mediated immunity is a significant systemic effect of graft-versus-host reactions. Although no complete explanation has been advanced for this immunosuppression suppressor cells have been implicated. The data presented in this paper indicate that acute GVH reactions in (C57BL/6J X A/J) F1-hybrid mice induced by the injection of A/J cells severely disrupts the function of the antisuppressor T-cell pathway at both its induction and effector stages. Results show that within 3 weeks of induction of the reaction, Ly1+-T antisuppressor inducer cells lose their ability to generate the serum factor that mediates antisuppression. This factor is normally taken up by and activates Ly2+ T cells which then inhibit suppressor T-cell function. The data also reveal that Ly2+ T cells collected 2 weeks after induction lose their ability to be activated by the antisuppressor factor produced in normal mice. These cells are thus unable to function as antisuppressor effector cells. The uptake of the antisuppressor factor by Ly2+ T cells depends on the expression of Ia antigens on the surface of these cells. Experiments have shown that these antigens are absent from the surface of T cells derived from mice with GVH reactions. This finding may provide an explanation for the inability of these cells to function as antisuppressor effectors. Antisuppression is an important T-cell pathway that is intimately associated with the regulation of immune function. It is possible that the immunosuppression arising in mice with GVH reactions may stem, in part, from unopposed suppressor T-cell activity that results from widespread interference by the reaction with a pathway that normally inhibits suppressor cell activity.

Enhanced interferon-alpha/beta (IFN-alpha/beta) and defective IFN-gamma production in chronic graft versus host disease: a potential mechanism for immunosuppression

Immunosuppression is a well-characterized consequence of chronic graft-versus-host disease (GVHD). We have previously shown that interferon (IFN) is produced in high levels during acute GVHD. Our objective in this study was to determine if IFN, as a cytokine with known immunosuppressive qualities, could be detected in mice experiencing chronic GVHD-induced immunosuppression. Two different experimental models were used to induce chronic GVHD. The first model involved the injection of parental strain spleen cells into adult F1 hybrids (AJ----B6AF1), while the second model utilized GVHD induced across minor histocompatibility barriers (B10.D2----BALB/c). Results indicated that significant levels of serum IFN-alpha/beta are present in mice undergoing chronic GVHD. Spleen cells from chronic GVHD mice were also shown to produce significant levels of IFN-alpha/beta upon in vitro culture in medium only. This IFN-alpha/beta production was greatly increased when GVHD spleen cells were cultured with either concanavalin A (Con A) or IL-2. In contrast, IFN-gamma production was undetectable in these Con A- or IL-2-containing cultures. Additionally, these same spleen cells which produced high levels of IFN-alpha/beta were immunosuppressed as measured by mitogen-induced cell proliferation. These results suggest that IFN-gamma production is defective in GVHD spleen cells, and that the presence of high IFN-alpha/beta production by GVHD mice may contribute to the immunosuppression associated with chronic GVHD.

The functional and histological basis for graft-versus-host-induced immunosuppression

The GvH reaction resulting from the injection of parental strain cells into adult F1 hybrids suppresses both cell-mediated and humoral immune responses and is dependent on the donor-host combination and the number of parental cells used to induce the GvH reaction. The early suppression is due, at least in part, to the increased number of macrophages and the activation of suppressor macrophages which act directly on the T-helper cell and perhaps the B-cell as well. The macrophage suppression is associated with an increase in PGE production. The long-term T-cell immunodeficiency is mediated by GvH-induced thymic dysplasia resulting in a block or an arrest in T-cell differentiation and deficient IL-2 production. The B-cell immunodeficiency is associated with both a decrease in B-cell production from lymphoid progenitors and a decrease in CFU-s production. The GvH reaction induces 2 types of thymic lesion, a stress-related effect causing atrophy of the thymic cortex and a cytolytic process causing severe-to-moderate lesions in the thymic medulla as a consequence of injury to medullary epithelial cells and a loss of Hassall's corpuscles (thymic dysplasia). By employing the NK-cell-deficient beige mutation, it was shown that the severe-to-moderate thymic medullary lesions occurred in F1 mice only in those transplant situations in which the donor inoculum was of the +/bg genotype, regardless of the genotype of the recipient. It is proposed that activation of parental T cells may contribute to the early immunosuppressive events; however, the relatively permanent immunosuppression appears to be associated with NK-like effector cells which are capable of causing injury to lymphoid and epithelial tissue, especially epithelium of the thymic medulla. These studies raise the possibility that the GvH reaction may contribute to some T- and B-cell immunodeficiencies observed in the SCID and AIDS syndromes, as well as in patients following bone marrow transplantation.

Parameters involved in the induction and abrogation of the lethal graft-versus-host reaction directed against non-H-2 antigens

The grafting of cells from donors incompatible for non-H-2 antigens alone can lead to GvHR mortality in up to 100% of lethally irradiated adult recipients. GvHR severity correlates with the number of mature immunocompetent cells present in the bone marrow inoculum. Histologic and clinical manifestations of GvHR observed in these mice differ from those seen when GvHR is induced across an H-2 barrier. The number of non-H-2 genes capable of influencing GvHR mortality is probably great, and their effects may vary as a function of sex. The non-H-2 genes influence GvHR mortality mainly via their interactions, the consequences of which are complex and can result in either cumulative or suppressive effects. GvHR mortality is considerably reduced by donor immunization, shortly before grafting, against host-specific non-H-2 antigens; and it is virtually abrogated by an additional immunization of the donors against nonspecific (foreign) H-2 antigens. Three weeks after grafting, these "protected" mice are easily distinguishable from those undergoing lethal GvHR, as assessed by both clinical appearance and histologic examination; in contrast, they are nearly indistinguishable from control mice grafted with syngeneic cells. However, depending upon the conditions used for the immunization, an additional immunization against nonspecific H-2 antigens can lead to acceleration rather than suppression of GvHR mortality; this phenomenon is not seen, under the same experimental conditions, after immunization against specific non-H-2 antigens alone. It is therefore suggested that a "second signal" provided by an additional nonspecific stimulus can potentiate either the establishment of specific suppression or the activation of a secondary ("positive") response. Suppressive effects of the specific and nonspecific immunizations are cumulative, and both treatments activate suppressor cells. The intensity of suppression induced by both specific and nonspecific immunizations is antigen dose-dependent. At equivalent antigen doses the specific immunization is considerably more effective than the nonspecific immunization, and is detectable after injection of as few as 2.5 X 10(5) cells. In both cases, irradiation of the immunizing cells abolishes the suppression induced by the lower cell doses tested, while it merely decreases the intensity of the suppression induced by the higher cell doses tested. The impairment of suppression after irradiation of the immunizing cells is not attributable to a modification of their homing pattern, but to the fact that proliferation of the immunizing cells, which leads to an augmentation of the antigen dose, is abolished by irradiation.(ABSTRACT TRUNCATED AT 400 WORDS)

Cellular interactions in graft-versus-host-induced T cell immune deficiency

The present review summarizes some of this laboratory's recent findings and interpretations concerning the cellular interaction involved in graft-versus-host-induced immune deficiency. Our results indicate that GvHID induced by parental T cell recognition of allogeneic F1 class I plus class II MHC determinants abrogates CTL responses involving both self + X and H-2 alloantigen recognition. In contrast, GvHID induced by parental T cell recognition of only allogeneic F1 class II antigens reduces self + X CTL responses, but does not reduce allogeneic CTL activity. Parental T cell recognition of only class I H-2 antigens does not affect CTL responses. The GvHID is associated with a loss of ability of spleen cells from GvH mice to produce IL 2 and to express IL 2 receptors after stimulation with T cell mitogens. This reduction in IL 2 potential is associated with a reduction in T cells that survive mitogenic stimulation in culture. Freshly explanted spleen cells from GvHID mice exhibit normal numbers of T cells. Our findings are discussed in terms of there being distinct helper T cell pathways for the generation of CTL to self + X and alloantigens (Singer et al. 1984). The model proposed for selective loss of self + X T cell immunity in class II-induced GvHID is discussed with respect to the parallel selective loss of self + X CTL activity in the developmental stages of acquired immune deficiency syndrome (AIDS).