Immune dysfunction in diabetes-prone BB rats. Interleukin 2 production and other mitogen-induced responses are suppressed by activated macrophages.

Spleen cells of diabetes-prone BB Wistar rats were found to generate excessively low proliferative responses, and interleukin 2 (IL-2) levels in response to T-dependent mitogens. This abnormality was not due solely to abnormal T cell numbers since: (a) addition of BB spleen cells of BB splenic macrophages to normal major histocompatibility complex (MHC)-matched Wistar Furth (WF) spleen cells resulted in severe suppression of concanavalin A (Con A)-, phytohemagglutinin (PHA)-, and pokeweed mitogen (PWM)-mediated proliferation, and IL-2 production; (b) macrophage depletion from BB spleen cells, but not B cell or T cell depletion, removed completely the suppressive effects of BB cells on WF cells; (c) macrophage depletion greatly enhanced the response of BB lymphocytes to T-dependent mitogens. Although suppressor macrophages could also be found in the spleen of WF control rats they were present in much smaller numbers than in the spleen of BB rats. The suppressive effect of BB macrophages was partially reduced by addition of the prostaglandin synthetase inhibitor indomethacin to cultures. Furthermore, indomethacin (but not catalase or PMA) considerably augmented IL-2 secretion of Con A-stimulated BB spleen cells, but had little effect on WF spleen cells. In contrast, prostaglandins E1 and E2 (PGE1 and PGE2) suppressed IL-2 production. While IL-2 secretion was severely depressed in BB rats unstimulated and lipopolysaccharide (LPS)-stimulated IL-1 secretion by splenic macrophages was normal. BB macrophages did not inactivate IL-2. Low IL-2 production and macrophage-mediated suppression were features of all BB rats tested.

Macrophage priming and lipopolysaccharide-triggered release of tumor necrosis factor alpha during graft-versus-host disease.

In this report we have investigated macrophage (M phi) activity and tumor necrosis factor alpha (TNF-alpha) production during graft-vs.-host disease (GVHD). TNF-alpha production by M phi requires two signals: priming of M phi by interferon followed by triggering of TNF-alpha production and release by lipopolysaccharide (LPS). The state of M phi activation was examined in nonirradiated B6AF1 recipient mice injected with either 60 x 10(6) (acute GVHD) or 30 x 10(6) (nonlethal GVHD) parental B6 lymphoid cells. During the early phase of acute GVHD, administration of normally sublethal amounts of LPS-triggered release of significant amounts of TNF-alpha into the serum resulting in death of the animals within 36 h. Normal animals treated with the same dose of LPS neither died nor produced detectable amounts of serum TNF-alpha. In vitro studies demonstrated that M phi were primed during GVHD. The level of M phi priming was greater during acute GVHD than nonlethal GVHD since 100-fold less LPS was required to trigger killing of a TNF-alpha-sensitive cell line by M phi from acute GVHD animals. The amount of TNF-alpha released into the serum after LPS injection increased during the course of the GVHD and was significantly greater in acute GVH-reactive mice. Endogenous LPS was detected in the serum of acute GVH-reactive animals coincident with the onset of mortality. The data provide evidence that during GVHD M phi are primed as a result of the allogeneic reaction and that endogenous LPS therefore triggers M phi production of TNF-alpha resulting in the symptoms characteristic of acute GVHD.

Mammary carcinoma arising in mice undergoing a chronic graft-versus-host reaction.

A graft-versus-host reaction (GVHR) was induced in 30 (CBA X A)F1 mice by the iv injection of 50 X 10(6) parental strain A lymphoid cells. Solid tumors emerged in 5 of 10 experimental animals that survived beyond 14 months after the GVHR was initiated. The neoplasms were judged to be mammary carcinomas by light and electron microscopic examinations. C-type RNA viral structures were observed in some tumor cells. The neoplasms were successfully transplanted into syngeneic animals by either sc or ip injections of tumor cell suspensions. Tumor transfer to syngeneic mice was not possible if only spleen cells from tumor-bearing animals were transferred. No tumors developed in an age-matched control group that received no treatment.

Epstein-Barr virus in a malignant lymphoproliferative disorder of B-cells occurring after thymic epithelial transplantation for combined immunodeficiency.

A fatal disseminated polyclonal malignant lymphoproliferative disorder of B-cells (immunoblastic sarcoma) developed shortly after a second thymic epithelial peritoneal implant in a 5-yr-old girl with combined immunodeficiency. The immunodeficiency was characterized by low T-cell numbers and function, very low levels of thymic hormone, dysgammaglobulinemia, and an inability to mount a primary antibody or cell-mediated response to new antigens. At necropsy, the thymus fulfilled morphological criteria for thymic dysplasia. Epstein-Barr virus (EBV) antigen and DNA were identified in neoplastic infiltrates in the lymph nodes and thymus by immunofluorescence for the EBV nuclear antigen and by EBV-specific complementary RNA/DNA hybridization. No antibodies to nuclear antigen, early antigen, or viral capsid antigen of EBV were identified in the serum. The concurrence of these events suggests that the thymic epithelial implant itself may have been instrumental in the pathogenesis of this neoplasm. It is proposed that the thymus may have provided factors which indirectly potentiated the proliferation of EBV-infected B-cells, possibly by induction of nonspecific T-helper cells and perhaps through other thymic humoral factors. It is suggested that some forms of immunoblastic sarcoma, even when polyclonal, and especially those which arise in immunocompromised hosts, may, in some instances, represent an opportunistic form of EBV-induced B-cell neoplasia.

Immunological and metabolic concomitants of cyclosporin prevention of diabetes in BB rats.

The metabolic and immunological effects of cyclosporin given to prevent diabetes in BB rats were examined. Diabetes-prone (BBdp) and normal (BBn) BB rats received either oral cyclosporin (10 mg X kg-1 X day-1 or its vehicle from age 30-150 days. Six of 21 (29%) vehicle-treated rats became glycosuric, with hyperglycemia, weight loss, and unremitting insulin requirements, and showed destruction of islet beta-cells. Five of 24 (21%) cyclosporin-treated rats became glycosuric, but none demonstrated weight loss, all required insulin only intermittently after onset, and all showed persistence of islet beta-cells. Cyclosporin induced hypoinsulinemic glucose intolerance in BBn rats. Cyclosporin inhibited the normal rise with age of peripheral blood lymphocyte cell numbers, identified with monoclonal antibodies. OX19+ (pan-T) and W3/25+ helper T-lymphocytes were affected, and there was an increase in the large W3/13+ OX19- population characteristic of BBdp rats; in addition, this subset appeared in BBn rats. Cyclosporin also caused the appearance and/or increase in both BBdp and BBn rats of W3/25+ OX19- and OX8+ OX19- subsets. Suppressor/cytotoxic (OX8+) T-lymphocytes and Ia+ cells were less affected. The incidence of hyperglycemia and glycosuria was therefore unaltered by cyclosporin, although the diabetic syndrome was milder. BBn rats receiving cyclosporin showed glucose intolerance, suggesting that in BBdp rats, the net effects of immunosuppression on beta-cell destruction may have been counterbalanced by the direct effect on the same cells. The attenuation of diabetes in BBdp rats occurred through further immunosuppression rather than by correction of its preexisting immunodeficiency.

Diabetes prevention in BB rats by frequent blood withdrawal started at a young age.

The BB rat diabetic syndrome has been prevented by various immunosuppressive and reconstitution measures. We observed an effect of multiple blood samplings on diabetes incidence and examined its immunological correlates. Individual litters were divided into two groups; one was sampled and the other was sham sampled as the control group. Sixty-four diabetes-prone and 59 non-diabetes-prone rats were studied. The sampled rats had blood removed at 15 (28% of total blood volume), 30 (30%), 50 (21%), 75 (16%), and 120 days of age. The sham-sampled control rats had blood removed only at 120 days of age. The incidence of diabetes in the sampled group was markedly lower than that of their sham-sampled littermates (22 vs. 78%). This result was associated with a correction of their OX19+ (pan-T-lymphocytes) and W3/25+ (helper/inducer) T-lymphocyte-number defects. An increase in lymphocyte subsets was also seen in the non-diabetes-prone BB rats, significant for all but the OX19+ cells. Islet pathology and pancreatic insulin content were consistent with metabolic outcomes. The effect of blood withdrawal thus has implications for understanding the pathogenesis of both the diabetes syndrome and the lymphopenia of the BB rat. Furthermore, it suggests that a stimulation of lymphopoiesis by blood withdrawal (analogous to that of erythropoiesis) may be a hitherto unrecognized physiological response in normal animals.

Effects of cyclosporine on glucose tolerance in the rat.

In a study of prevention of spontaneous diabetes in BB rats by therapeutic doses of cyclosporine (10 mg/kg/day), the male control non-diabetes-prone rats showed glucose intolerance after a 0.25 g/kg glucose load by gavage, at 90 and 130 days of treatment. Non-BB male Wistar rats treated similarly showed glucose intolerance at 1 wk of treatment, with progressive worsening for 5 wk, then sustained up to 12 wk of treatment. Fasting euglycemia was maintained, but both pre- and postchallenge plasma insulin levels were significantly lower with cyclosporine at several time points. Total pancreatic insulin was decreased to one-third that of control after 5 wk. After withdrawal of cyclosporine, glucose tolerance returned to normal in 2 wk. Sprague-Dawley rats responded similarly and in both strains, an increase in the cyclosporine dose to 15 mg/kg/day augmented the glucose intolerance. These results demonstrate that therapeutic doses of this agent induce reversible glucose intolerance due, in part, to inhibition of insulin secretion and also possibly inhibition of synthesis, though a peripheral effect is not excluded. This hyperglycemic effect of cyclosporine has implications for its potential use in type I diabetes mellitus, transplantation, and other autoimmune disease.

Association of susceptibility to spontaneous diabetes in rat with genes of major histocompatibility complex.

This study was designed to map the diabetes susceptibility gene(s) associated with the rat major histocompatibility complex (MHC) RT1. We have crossed spontaneously diabetic male rats bearing the recombinant RT1r8 haplotype with female rats of the AC1.1r4 congenic strain. Three diabetic rats were determined to be homozygous for the r4 haplotype by serotyping. The absence of recombination within the MHC was confirmed by inspection of restriction-fragment-length patterns of the diabetic animals and the parental strains. In conjunction with previous breeding studies, this study maps the diabetes susceptibility gene to the right of the RT1-A locus and to the left of the RT1-C locus. A low incidence of diabetes in the F2 (4.5%) emphasizes the multifactorial nature of the susceptibility. The presence of depressed responsiveness of peripheral blood lymphocytes to concanavalin A stimulation increases the prevalence of the overt disease. An unusual feature of the diabetic syndrome in this study is the sparse or absent pancreatic lymphocytic inflammatory response, with true insulitis being a rare finding.

IDDM in BB rats. Enhanced MHC class I heavy-chain gene expression in pancreatic islets.

Modulation in major histocompatibility complex (MHC) gene expression correlates with the inflammatory reactions that occur during graft rejection and autoimmune disease. We analyzed the expression of class I and II MHC genes in the pancreatic islets of prediabetic and newly diabetic BB rats by immunohistochemistry of tissue sections and Northern blotting of RNA extracted from isolated islets. We show that enhanced levels of MHC class I heavy-chain RNA are present in pancreatic islets before overt inflammation and the onset of insulin-dependent diabetes mellitus (IDDM) in the spontaneously diabetic BB rat. Immunohistochemical analysis revealed enhanced class I antigen expression throughout the pancreatic islets of newly diabetic animals but no induction of class II antigen on endocrine cells within the islet. Varying degrees of inflammatory infiltrate were observed in the sections exhibiting enhanced class I antigen expression or in nearby serial sections. Southern blot analysis revealed no restriction-fragment-length polymorphism or amplification of the endogenous class I heavy-chain genes compared with those of seroidentical disease-resistant Wistar-Furth rats. I-A alpha and I-E alpha hybridizing RNA appeared de novo before overt diabetes, although concomitantly with T-lymphocyte-receptor beta-chain and interferon-gamma gene hybridizing RNA and after MHC class I heavy-chain RNA enhancement was observed. These data indicate the possibility that enhanced class I heavy-chain gene expression plays a role in the progression of IDDM.

Immunologic and genetic studies of diabetes in the BB rat.

The spontaneous development of diabetes in the Bio-Breeding (BB) rat is an excellent model of human insulin-dependent diabetes mellitus (IDDM). Disease expression is dependent on several genetically determined abnormalities, including specific major histocompatibility complex (MHC) genes. At least one MHC class II locus of the U haplotype is a necessary, but not sufficient, condition for disease expression. The immune system of BB rats is markedly abnormal. There is a striking reduction in the number and function of mature cytotoxic/suppressor T cells, a poor proliferative response to mitogens and in mixed lymphocyte culture, poor interleukin-2 production, and a reduced ability to reject skin allografts. While these immune system abnormalities are closely related to the development of diabetes, the immune recognition and effector mechanisms resulting in islet cell destruction are still poorly understood. The hypothesis that MHC class II induction on pancreatic beta cells serves to target these lymphokines, natural killer (NK) cells, macrophages, etc.) have been implicated in islet cell killing. The incidence of IDDM is reduced by immunosuppressive therapy in both rats and humans, further supporting the role of immune mechanisms in this disease.

Association of spontaneous thyroiditis with the major histocompatibility complex of the rat.

Overt insulin-dependent diabetes mellitus in the rat is associated with the u haplotype of the rat major histocompatibility complex (MHC), RT1. Thyroiditis of sufficient severity to result in elevation of TSH levels is seen in Buffalo rats (RT1b). In order to examine the association of autoimmune thyroid disease with MHC gene products, we have crossed inbred Buffalo rats with diabetic BB rats and examined the RT1 genotype, the histology of thyroid and pancreatic tissue, and two indices of thyroid function. The data indicate that animals having pancreatic lymphocytic infiltration and insulinopenic overt diabetes mellitus had at least one RT1u haplotype. All but one animal having severe histological thyroid lymphocytic infiltration had at least one RT1b haplotype. Rats with severe thyroiditis had higher mean TSH levels than rats with normal histology or rats with mild thyroiditis. We conclude that gene products of the rat MHC affect the severity of spontaneous organ-specific autoimmune disease in terms of clinically apparent as well as tissue inflammatory disease.

Elastosis in benign sclerosing ductal proliferation of the female breast.

Elastosis, an abundance of elastic tissue, is commonly present in breast carcinoma. However, its diagnostic significance remaines an unsettled issue. This study documents 17 cases of elastosis occurring in a distinctive benign sclerosing ductal lesion of the female breast (Fenoglio and Lattes: Cancer 33: 691-700, 1974). Elastosis was characterized by staining reactions and, in several instances, by elastase digestion and electron microscopy. Yellow streaks and flecks may be apparent grossly and probably reflect the increased elastic tissue. Histologically, the lesion is generally stellate with central sclerosis and marked peripheral intraductal and ductular hyperplasia which is often papillary. Elastosis, which may be marked, is a constant finding and is predominantly periductal in location. It is emphasized that the gross and histologic features of the lesion may mimic carcinoma and that elastosis may be found in benign ductal lesions of the breast.

Desmoplastic fibroma of bone. An ultrastructural study.

The ultrastructure of three cases of desmoplastic fibroma of bone is presented. The lesion is principally characterized by myofibroblasts admixed with lesser numbers of fibroblasts and primitive mesenchymal cells. Thus, the cellular composition is similar to that described in a variety of nonneoplastic proliferative processes of soft tissue. It is postulated that the myofibroblastic proliferation develops in response to unknown factors acting on marrow fibroblasts or primitive mesenchymal cells.

Myofibroblasts in the stroma of invasive and metastatic carcinoma: a possible host response to neoplasia.

After observing the presence of numerous stromal myofibroblasts in scirrhous mammary carcinomas, a series of invasive and metastatic carcinomas from diverse sites was examined by electron microscopy to determine whether myofibroblasts might also be present in their stroma. Myofibroblasts were identified in each instance and were most abundant in neoplasms which were hard, sclerotic, and retracted. This finding suggests that myofibroblasts represent a component of the stromal reaction to many carcinomas and contribute to the desmoplasia and retraction which characterize many of these neoplasms. The host commands several responses to neoplasia. As a result of the expression of tumor-associated antigens, the immune system contributes lymphocytes, macrophages, and antibodies, a reflection of immunologic surveillance against neoplasia. In contrast to experimental systems tumor neoantigens are poorly expressed or even lacking in many human neoplasms; thus, the immune system may be weakly stimulated or not activated at all. Tumor neovascularization induced by a tumor-angiogenesis factor represents a second host response, possibly deleterious, for it may facilitate tumor dissemination. The stromal myofibroblast reaction to many invasive and metastatic carcinomas may constitute a third, albeit more primitive response. The density of collagen produced and contractile state of such tissue may signify an attempt by the host stroma to contain the neoplasm and impede vascular invasion. If so, myofibroblast induction may complement immune surveillance or constitute a separate mechanism of response to invasive neoplasia in man.

Pleomorphic soft tissue myogenic sarcomas of adulthood. A reappraisal in the mid-1990s.

325 diverse sarcomas, 39 rhabdomyosarcomas (RMS), including all histologic variants, and 135 leiomyosarcomas (LMS) were identified. Within these two groups, 18 (46%) of the RMS and 14 (10%) of the LMS represented pleomorphic variants. These neoplasms were studied by morphology (histology and ultrastructure) and by immunohistochemical methods employing antibodies to intermediate filaments (vimentin and desmin) and actin isoforms [alpha-smooth (sm) and alpha-sarcomeric (sr) actins]. Twenty-four pleomorphic malignant fibrous histiocytomas (MFH) and eight pleomorphic liposarcomas (LS) were examined in a similar fashion. By light microscopy, the pleomorphic RMS, LMS, and MFH were indistinguishable, as each was dominated by pleomorphic cells disposed in a haphazard growth pattern; moreover, many featured fascicular, storiform, and sclerotic zones. The distinction between these neoplasms became apparent only following immunohistochemistry and/or ultrastructural study. All pleomorphic RMS disclosed rudimentary sarcomeres and exhibited the following cytoskeletal profile: vimentin (+) (18 of 18), desmin (+) (14 of 18), alpha-sr actin (+) (18 of 18) and alpha-sm actin (+) (five of 18). All the pleomorphic LMS featured smooth-muscle differentiation of variable degrees in the form of cytoplasmic bundles of microfilaments and associated dense bodies; their cytoskeletal profile was vimentin (+) (14 of 14), desmin (+) (seven of 14), alpha-sr actin (+) (none of 14), and alpha-sm actin (+) (eight of 14). The latter was demonstrated in all moderately differentiated, but absent or only focally expressed in poorly differentiated variants. All pleomorphic MFH and LS were devoid of myogenic (skeletal or smooth) ultrastructural features and expressed vimentin solely. This combined morphological and immunohistochemical study illustrates the following: First, these pleomorphic sarcomas are often indistinguishable by histologic growth pattern alone; thus, an accurate diagnosis requires study with all of these techniques. Second, pleomorphic myogenic sarcomas are restricted to adults and are not uncommon neoplasms among pleomorphic sarcomas: RMS (28%), LMS (21%), MFH (38%), and LS (13%). Third, the study defines desmin-negative and alpha-sm actin-positive pleomorphic RMS, and desmin-negative and alpha-sm-actin-negative pleomorphic LMS.

Translocation t(1;3)(p36.3;q25) is a nonrandom aberration in epithelioid hemangioendothelioma.

The cytogenetic findings for two epithelioid hemangioendotheliomas are reported. An identical chromosomal translocation involving chromosomes 1 and 3 [t(1;3)(p36.3;q25)] was detected in both cases of epithelioid hemangioendothelioma, possibly representing a characteristic rearrangement for this histopathologic entity. The presence of clonal karyotypic abnormalities supports a neoplastic origin for the epithelioid variant of hemangioendothelioma. Identification of the 1;3 translocation may be useful diagnostically. Should additional studies confirm these data, this could lead to the identification of the gene(s) central to this neoplastic process.

Adamantinoma-like Ewing's sarcoma: genomic confirmation, phenotypic drift.

Ewing's sarcoma, a highly malignant neoplasm, is characterized by an 11;22 translocation [t(11;22) (q24;q12)], resulting in the fusion of genes FLII and EWS. Adamantinoma of extragnathic bones, a low-grade malignant neoplasm with epithelial features, is not typically considered in the differential diagnosis of Ewing's sarcoma. In this study, three osseous Ewing's sarcomas with histological, immunohistochemical, or ultrastructural epithelial features were subjected to reverse transcription-polymerase chain reaction and sequencing studies for the Ewing's sarcoma molecular rearrangement. (Two of the three cases were originally described as adamantinomas or nontypical Ewing's sarcoma before the availability of genetic characterization.) In addition, traditional cytogenetic analysis and a unique combined interphase molecular cytogenetic/ immunocytochemical approach with bicolor 11;22 translocation breakpoint flanking probes (cosmids) and pancytokeratin antibodies were performed on one neoplasm. At(11;22) (q24;q12) was found in one neoplasm and a type II EWS/FLI-1 fusion transcript was detected in all three neoplasms. The combined genetic/immunocytochemical approach revealed the presence of the 11 ;22 translocation in the nuclei of cytokeratin immunoreactive cells. These genotypic and phenotypic findings delineate a novel Ewing's sarcoma histologic variant, "adamantinoma-like Ewing's sarcoma."

The graft-versus-host reaction and immune function. IV. B cell functional defect associated with a depletion of splenic colony-forming units in marrow of graft-versus-host-reactive mice.

Studies were conducted to determine whether a functional B cell defect occurred in the bone marrow of mice experiencing a GVH reaction (GVHBM). GVH reactions were induced in AxCBA F1 adult mice by an injection of A strain lymphoid cells. The GVH reaction was confirmed by immunosuppression and thymus histology. At various intervals after GVH induction, GVHBM was tested for its ability to restore B cell function in adult thymectomized irradiated mice reconstituted with normal thymocytes. GVHBM cells obtained seven days after GVH induction restored but slightly the plaque forming cell (PFC) response to sheep erythrocytes and the mitogen response to lipopolysaccharide (LPS). GVHBM cells obtained 10 days or later failed to reconstitute the PFC or LPS responses. GVHBM cells suppressed neither the T or B cell function of normal spleen cells nor the LPS mitogen response of normal bone marrow cells. In addition, the splenic colony-forming units (CFU-s) in GVHBM were slightly decreased by day 10 after GVH induction and markedly depressed by day 22 after GVH induction. These results suggest that the GVH reaction may affect two different events in B cell differentiation. The early decrease in functional B cells that occurs before there is any change in the CFU-s population suggests a direct effect on B cell production, whereas the later absence of functional B cells could be due to the marked decline in stem cell production (CFU-s).

The graft-versus-host reaction and immune function. II. Recruitment of pre-T-cells in vivo by graft-versus-host-induced dysplastic thymuses following irradiation and bone marrow treatment.

The graft-versus-host (GVH) reaction induces thymic dysplasia and an arrest in T cell differentiation. Studies were performed to test the effect of irradiation and reconstitution with bone marrow on GVH-induced thymic dysplasia and T cell differentiation. GVH reactions were induced in CBAxAF1 adult mice by the injection of A strain lymphoid cells. All GVH-reactive mice were immunosuppressed by day 7 after GVH induction and thymic dysplasia was evident by day 24. Forty days after the induction of the GVH reaction the mice were irradiated (850 rads) and repopulated with 10-15 X 10(6) syngeneic or parental bone marrow cells. Thirty days after irradiation and bone marrow reconstitution, GVH-reactive mice were used for histological and functional studies. These mice displayed near-normal thymus morphology with scattered epithelial cells in the medulla, and normal numbers of Thy-1-positive cells. Donor cells had totally repopulated thymuses of irradiated bone marrow reconstituted mice by day 19 after irradiation. T helper cell function did not recover in the reconstituted mice. These results suggest that (1) the process responsible for GVH-induced thymic dysplasia is radiosensitive, and (2) the thymus has the potential to regenerate a normal structure, but fails to regain normal function.

Prevention of murine graft-versus-host disease by inducing and eliminating ASGM1+ cells of donor origin.

In this study anti-asialo GM1 antibodies (anti-ASGM1) were used to further characterize the effector cells responsible for graft-versus-host (GVH)-induced histopathological lesions: Two different types of ASGM1+ cells were identified: an endogenous ASGM1+ population and an induced ASGM1+ population. Both of the ASGM1+ cell populations exhibited natural killer (NK) cell activity, as assessed by their ability to lyse YAC tumor targets in vitro. Donor C57BL/6 (B6) mice were treated in vivo with anti-ASGM1 to eliminate endogenous ASGM1+ cells. ASGM1+ cells were induced in B6 donor mice by treating the animals with 15 x 10(6) B6 x AF1 (B6AF1) lymphoid cells for 44-48 hr. The induced ASGM1+ cells were eliminated by in vivo treatment with anti-ASGM1. GVH reactions were induced by injecting B6 lymphoid cells into B6AF1 mice. Prior to GVH induction the B6 donor cells were tested for NK cell activity against YAC tumor target cells in vitro and for T and B cell functions by mitogen responses in vitro. GVH reactions were determined by splenomegaly, suppression of the plaque-forming cell (PFC) response to sheep red blood cells (SRBC), suppression of the T and B cell mitogen responses, and the development of GVH-associated histopathological alterations in the thymus, liver, and pancreas. Donor lymphoid cells depleted of endogenous ASGM1+ cells were effective at inducing splenomegaly, severe suppression of immune functions, and histopathological lesions. Donor lymphoid cells depleted of both the endogenous and induced ASGM1+ cells displayed normal T cell mitogen responses and were capable of inducing splenomegaly and partial suppression of the PFC response to SRBC when injected into B6AF1 recipients, however, these lymphoid cells failed to induce both GVH-associated histopathological lesions and severe suppression of T and B cell mitogen responses. These results suggest that semiallogeneic stimulation induces an ASGM1+ population in the donor inoculum that displays NK cell-like function (YAC killing) and that plays a crucial role in inducing GVH-mediated histopathological lesions and severe immunosuppression of both T and B cell responses.