Immunomodulation by low-dose methotrexate. I. Methotrexate selectively inhibits Lyt-2+ cells in murine acute graft-versus-host reactions

We have studied the effect of methotrexate in murine acute graft vs host (GvH) disease at concentrations analogous to those used in human rheumatoid arthritis. The GvH reaction was induced by i.v. injection of parental spleen cells into a normal F1 recipient. The acute suppression of T cell function in GvH mice was prevented by methotrexate given orally for 10 days at 1.0 or 0.5 mg/kg but not at 0.25 mg/kg. T cell mitogen response and IL-2 secretion that were inhibited in GvH mice were restored by methotrexate. Protection from immunosuppression in drug-treated GvH mice lasted at least 3 wk after drug dosing was stopped. The mechanism of the protective effect appears to be a preferential inhibition of donor and host Lyt-2+ Ts cell proliferation. In mixing experiments we found that methotrexate inhibited Ts function in GvH mice. By dual fluorescence labeling we showed that the engraftment of donor Lyt-2+ cells was prevented by drug treatment. This was not true of donor L3T4+ cells which were clearly present in the spleens of GvH mice after methotrexate treatment. These donor L3T4 cells were functional in that they induced the production of anti-DNA autoantibodies in the methotrexate-treated GvH mice.     

Enlargement of Lyt-2-positive T cells is associated with functional impairment and autoimmune hemolytic anemia in New Zealand Black mice

We investigated the relationship between the increased cell diameter of Lyt-2+ T cells and the development of autoimmune disease in aging NZB and NZB X NZW F1 hybrid (BW) mice. Individual animals were analyzed for Lyt-2+ T cell size (by narrow-angle forward light scatter), anti-erythrocyte autoantibodies, anemia, proteinuria, and splenomegaly. The peak light scatter of the Lyt-2+ T cells correlated with the level of anti-erythrocyte autoantibodies and severity of hemolytic anemia, but not with proteinuria or splenomegaly. The cell size of this T cell subset did not increase in old BW or in NZB mice homozygous for the xid gene (NZB.xid). The in vivo administration of bacterial lipopolysaccharide to young NZB mice did not stimulate the enlargement of Lyt-2+ T cells. Ly-2+ T cells from old NZB mice could be stimulated by concanavalin A (Con A) to express interleukin 2 (IL 2) receptors and to synthesize DNA in vitro. However, in vivo administration of Con A to old NZB mice did not induce the expression of IL 2 receptors on Lyt-2+ T cells. Further, in vivo T suppressor function was impaired in old NZB mice with enlarged Lyt-2+ T cells. Thus, the enlargement of Lyt-2+ T cells in old NZB mice appears related to impaired T cell function in vivo and is associated with the development of anti-erythrocyte autoantibodies and autoimmune hemolytic anemia.     

Induction of graft versus host-associated immunodeficiency by CD4+ T cell clones

Previous studies have shown that the injection of parental T cells into MHC class II mismatched F1 recipient mice can lead to graft-vs-host (GvH) reaction that manifests itself by multiple symptoms. The objective of our study was to analyze GvH reactivity induced by a single T cell clone specific for host I-A or I-E alloantigen. The T cell clones tested for GvH potential were CD4+, with or without cytolytic activity in vitro and with a lymphokine pattern that classifies them as Th1 cells. The inoculation of a single T cell clone induced a severe, but transient immunodeficiency in the host that was independent of its cytolytic activity, as demonstrated by the failure to generate a CTL response to third party allogeneic cells in vitro. Induction of immunodeficiency in the recipients required preactivation of the clones in vitro by rIl-2 and the presence of the stimulator class II alloantigen in the host. Spleen cells from these mice lacked suppressor cells, they were deficient in Il-2 secretion and exhibited a decrease in the number of CD4+ T cells. In addition, I-E expression was reduced, however, without any changes in the macrophage population and an increase in surface Ig and the B cell marker B220. Simultaneous to the immunodeficiency, the clone-injected mice produced elevated antibody titers to ssDNA.     

Effect of tilorone hydrochloride on graft-versus-host (GvH) reaction in mice.

The effects of tilorone hydrochloride (TH), a powerful interferonogenic agent exhibiting also immune modulatory properties, on GvH reaction was studied, using the popliteal lymph node assay in mice. Administration of TH at ten days 0 or +2 relative to cell transfer to recipient mice led to a significant dose-dependent reduction of GvH reaction, whereas treatment of prospective donor mice at day -4 or -2 induced an enhanced GvH re activity of donor spleen cells. This effect was found not to be due to an altered proportion in the spleen cell inoculum of B and T lymphocytes, which latter are responsible for induction for GvH reaction. However, since normal parental lymphocytes are prepared for an enhanced GvH reactivity by addition of TH-treated macrophages, a stimulatory effect of the latter cells via macrophage-derived mediators, induced by TH, is suggested.     

Cytotoxic T cell response and thymic hormonal dysfunction in graft-vs-host mice

As an approach to dissect complex mechanisms that lead to graft-vs-host (GvH)-associated immune disorders, we have compared the splenic cytotoxic T lymphocyte (CTL) response and thymic hormonal function in nonirradiated F1 hybrid mice injected with parental spleen cells. Thymic secretory function was studied by the determination of serum thymulin levels, and the number of thymic epithelial cells containing thymulin as assessed by indirect immunofluorescence with the use of an anti-thymulin monoclonal antibody. In addition, the epithelial cell network was analyzed with an anti-keratin serum, and the general histology pattern was studied by conventional histologic methods. An initial analysis was performed on day 15, which was characterized by CTL suppression mediated by parental suppressor T cells. No thymic abnormalities were detected at this time. By day 45 after GvH induction, active CTL suppression had decreased, and GvH was associated with a progressive decline in thymic hormonal function. Finally, by day 60 and thereafter, F1 GvH mice recovered normal in vitro CTL responsiveness, which contrasted with profound alterations of the epithelial cell network and severely reduced serum thymulin levels. This hormonal dysfunction was shown to be directly associated with a reduction in the number of thymulin-containing cells. Moreover, no anti-thymulin auto-antibodies could be detected. The results are discussed with respect to the role of thymic hormonal dysfunction in the modulation of F1 CTL responses observed during the course of a GvH reaction, and the additional analogy of this GvH model with human immunodeficiency.     

Age restriction in antigen-specific immunosuppression

Age-related alterations of antigen-specific T cell-mediated suppression have been examined in the 4-hydroxy-3-nitrophenyl acetyl (NP) system. Inducer suppressor T cells (Tsi) were activated in mice at the age of 3 mo (young) or 18 mo (old) by i.v. injection of NP-conjugated syngeneic spleen cells (SC). Spleen cells from the NP-SC-injected mice were subcultured in vitro with spleen cells from normal young or old mice to generate transducer suppressor T cells (Tst). Four days later subcultured cells were added to responder cell cultures 1 day before the PFC assays to trigger effector suppressor T cells (Tse). Responder cell cultures, containing NP-conjugated horse red blood cells (HRBC) and spleen cells from HRBC-primed young or old mice, were assayed on day 4 for anti-NP and anti-HRBC PFC. Suppression was found to be antigen specific and age restricted. NP-specific suppressor cells are easily induced in subculture if the Tsi and Tst cell populations are both derived from young or old mice. Conversely, if Tsi cells from young or old mice are subcultured with Tst cells from mice of a different age, suppression of the anti-NP PFC response is hardly observed. Age restriction was also found to operate in the interactions between subcultured and responder cell populations, indicating that age-matching is required for effective triggering of Tse cells by Tst cells. These results altogether suggest that aging may affect the recognition repertoire expressed in suppressor T cell subsets. Moreover, the finding that suppression is less efficient when exerted on responder spleen cells from old than from young mice provides an explanation for the increased frequency of autoimmune disorders in aging.     

Regulation of an in vitro anti-DNA antibody response by thymocytes and T cells from NZB/W and normal mice

The spontaneous in vitro anti-DNA antibody response generated by preautoimmune and many normal mouse spleen cells was suppressed by the addition of syngeneic thymocytes or splenic T cells. Suppressive activity was found in normal mice (DBA/2J) and to an equivalent degree in the autoimmune (New Zealand Black X New Zealand White)F1 (B/W) strain. The suppressor cells were cortisone-resistant, radiosensitive and carried Lyt 1 and Lyt 2 markers. Nonspecific suppression was not involved since the primary and primed in vitro anti-sheep erythrocyte (anti-SRBC) responses were unaffected. Both spontaneous and lipopolysaccharide-stimulated anti-DNA antibody responses could be suppressed. There was no difference in the suppressive activity of cells from young or old, normal or autoimmune mice. These T cells may therefore play a role in preventing the anti-DNA antibody response in normal and young B/W mice, but evidently fail to influence the development of in vivo anti-DNA autoimmune responses in the old B/W mice.     

CBA/N X-linked defect delays expression of the Y-linked accelerated autoimmune disease in BXSB mice

BXSB male mice spontaneously develop progressive autoimmune disease characterized by high serum immunoglobulins, including anti-nuclear antibodies (ANA), enlarged spleen and lymph nodes, and diffuse proliferative glomerulonephritis. Females develop symptoms at a much slower rate. The mechanisms underlying the autoimmune disease and the nature of the Y-linked accelerating factor have not yet been elucidated. We found that the male progeny of the cross between the non-autoimmune strain CBA/Ca and BXSB (CBA/Ca X BXSB)F1 showed progressing signs of autoimmunity starting at 6 to 7 mo. In contrast, the male progeny that resulted from BXSB males crossed with immune-defective CBA/N females (Xid) were devoid of splenic B colonies, were nonresponsive to TNP-Ficoll, and were free of autoimmune disease for at least 10 mo. At 18 mo, some of the (CBA/N X BXSB)F1 mice developed weak antinuclear antibodies, but no spleen or lymph node enlargement was seen. The same mice had low anti-TNP Ficoll responses but did not produce B colonies in vitro. The role of the X chromosome in regulating expression of autoimmunity in young and old BXSB mice is discussed.     

Age-dependent onset of liver-specific IGF-I gene deficiency and its persistence in old age: implications for postnatal growth and insulin resistance in LID mice

To explore the limitations of the liver-specific IGF-I gene-deficient (LID) model and to further evaluate the role of endocrine IGF-I in early postnatal life and old age, we have studied these mice during the prepubertal period (from birth to 3 wk of age) and when they are 2 yr old. During the first 2 wk of life, IGF-I gene deficiency and the resulting reduction in serum IGF-I levels in LID mice did not reach sufficiently low levels when mice experience the most rapid and growth hormone (GH)-independent growth. It suggests that the role of liver-derived IGF-I in prepubertal, GH-independent postnatal growth cannot be established. From our previous studies, liver IGF-I mRNA level was abolished in adult LID mice, which causes elevated GH level, insulin resistance, pancreatic islet enlargement, and hyperinsulinemia. Interestingly in 2-yr-old LID mice, although liver IGF-I mRNA and serum IGF-I levels were still suppressed, serum insulin and GH levels had returned to normal. Compared with same-sex control littermates, aged male LID mice had significantly reduced body weight and fat mass and exhibited normal insulin sensitivity. On the other hand, aged female LID mice exhibited normal weight and marginal resistance to insulin actions. The pancreatic islet percentage (reflecting islet cell mass) was also restored to normal levels in aged LID mice. Thus, although the IGF-I gene deficiency is well maintained into old age, the insulin sensitivity, islet enlargement, and hyperinsulinemia that occurred in young adult mice have been mostly restored to normal levels, further supporting the age-dependent and sexual dimorphic features of the LID mice.     

IL10 deficiency promotes alveolar enlargement and lymphoid dysmorphogenesis in the aged murine lung

The connection between aging‐related immune dysfunction and the lung manifestations of aging is poorly understood. A detailed characterization of the aging IL10‐deficient murine lung, a model of accelerated aging and frailty, reconciles features of both immunosenescence and lung aging in a coherent model. Airspace enlargement developed in the middle‐aged (12 months old) and aged (20–22 months old) IL10‐deficient lung punctuated by an expansion of macrophages and alveolar cell apoptosis. Compared to wild‐type (WT) controls, the IL10‐deficient lungs from young (4‐month‐old) mice showed increased oxidative stress which was enhanced in both genotypes by aging. Active caspase 3 staining was increased in the alveolar epithelial cells of aged WT and mutant lungs but was greater in the IL10‐deficient milieu. Lung macrophages were increased in the aged IL10‐deficient lungs with exuberant expression of MMP12. IL10 treatment of naïve and M2‐polarized bone marrow‐derived WT macrophages reduced MMP12 expression. Conditioned media studies demonstrated the secretome of aged mutant macrophages harbors reduced AECII prosurvival factors, specifically keratinocyte growth factor (KGF) and hepatocyte growth factor (HGF), promotes cell death, and reduces survival of primary alveolar epithelial cells. Compared to WT controls, aged IL10‐deficient mice have increased parenchymal lymphoid collections comprised of a reduced number of apoptotic cells and B cells. We establish that IL10 is a key modulator of airspace homeostasis and lymphoid morphogenesis in the aging lung enabling macrophage‐mediated alveolar epithelial cell survival and B‐cell survival within tertiary lymphoid structures.     

Interaction between antigen presenting cells and autoreactive T cells derived from BXSB mice with murine lupus

Systemic lupus erythematosus （SLE） is a typical autoimmune disease involving multiple systems and organs. Ample evidence suggests that autoreactive T cells play a pivotal role in the development of this autoimmune disorder. This study was undertaken to investigate the mechanisms of interaction between antigen presenting cells （APCs） and an autoreactive T cell （ATLI） clone obtained from lupus-prone BXSB mice. ATLI cells, either before or after 7-ray irradiation, were able to activate naive B cells, as determined by B cell proliferation assays. Macrophages from BXSB mice were able to stimulate the proliferation of resting ATL 1 cells at a responder/stimulator （R/S） ratio of 1/2.5. Dendritic cells （DCs） were much more powerful stimulators for ATLI cells on a per cell basis. The T cell stimulating ability ofmacrophages and B cells, but not DCs, was sensitive to T-ray irradiation. Monoclonal antibodies against mouse MHC-Ⅱ and CD4 were able to block DC-mediated stimulation of ATL 1 proliferation, indicating cognate recognition between ATL 1 and APCs. Our data suggest that positive feedback loops involving macrophages, B cells and autoreactive T cells may play a pivotal role in keeping the momentum of autoimmune responses leading to autoimmune diseases.     

Transferred B cells from autoimmune NZB/N mice fail to activate T suppressor cells

T-cell-mediated suppression of the antibody response of autoimmune NZB/N mice to Type III pneumococcal polysaccharide (SSS-III) can readily be induced in situ by priming with a subimmunogenic dose of SSS-III; however, the transfer of either "young" (8 weeks old) or "old" (42 weeks old) SSS-III-primed B cells, which activates suppressor T cells in normal BALB/cByJ mice, fails to induce suppression of the antibody response in recipient NZB/N mice, regardless of the number of cells transferred or the time interval between transfer and immunization. Transfer of 51Cr-labeled B cells demonstrated that syngeneic primed B cells home to the spleens of NZB/N mice in somewhat lower numbers than in BALB/cByJ mice, although the differences observed may not be sufficient to explain the complete absence of activation of suppressor T cells. These findings suggest that B cells from autoimmune NZB/N mice are unable to activate T suppressor cells upon transfer; this disorder in a normal regulatory mechanism may be important in the pathogenesis of disease.     

Prevention of experimental allergic encephalomyelitis (EAE) in the SJL/J mouse by whole body ultraviolet irradiation

The cellular requirements for the in vivo induction of experimental allergic encephalomyelitis (EAE) were investigated in the SJL/J mouse. Exposure of mice to whole body ultraviolet (UV) irradiation, a treatment that has been shown in other systems to interfere selectively with antigen-presenting cell function, prevented the development of clinical and pathologic signs of acute EAE. Splenic T cells from UV-treated animals did not adoptively transfer resistance to EAE, making it unlikely that UV irradiation resulted in the generation of a specific suppressor cell population responsible for protection from EAE. UV irradiation was effective in preventing EAE when administered before initial immunization; UV irradiation was ineffective in modifying ongoing EAE or in preventing relapses of EAE induced by reimmunization. In additional experiments, adult thymectomized, lethally x-irradiated mice reconstituted with syngeneic marrow cells depleted of mature T lymphocytes were found to be resistant to the induction of EAE. Susceptibility was restored by the addition of splenic T cells, demonstrating that EAE induction is T cell-dependent in the mouse. The prevention of an experimental autoimmune demyelinating disease by whole body UV irradiation suggests that interference with the function of Ia-bearing accessory cells may represent an approach for immunotherapy in autoimmune disorders.     

lack of suppressor cell activity for natural killer cells in infant, aged and a low responder strain of mice

We have examined the reported role of suppressor cells in the regulation of NK activity in mice with naturally low NK activity (infant and aged (C57 X A)F1 hybrids (CAF1) and low responder strain AKR mice). Possible suppressor activity was assayed by mixing, at a 1 : 1 ratio, spleen cells from low activity mice with spleen effector cells from normally active 8 to 10 wk old CAF1 mice. The lytic activity of the mixture was compared with the activity of effector cells diluted with medium alone or diluted 1 : 1 with "non-suppressor" population which served as a control for nonspecific decreases in lysis. The control or "filler" cells employed were suspensions of adult CAF1 thymus, thymus from adult mice exposed to 2,000 R, and adult CAF1 spleen cells cultured for 24 hours, a procedure that depleted NK activity. In no case was the activity observed in the presumed suppressor-effector mixture significantly lower than that observed in the filler-effector cell mixtures. Thus, in infant (1 to 2 wk) and aged (12 to 18 mo) CAF1 mice and in 8 to 10 wk old AKR mice, we found no evidence for specific cell-mediated suppression of natural cytotoxicity.     

Lasting effects of an impairment of Th1-like immune response in γ-irradiated mice: A resemblance between irradiated mice and aged mice

Although one of the several chronic effects of ionizing radiation is aging, there are no experimental data on radiation-induced immunological aging. The most interesting change in aging was a helper T (Th) 1/Th2 imbalance. We investigated chronic effect on immune responses after ionizing radiation and its effects in irradiated mice were compared with those of aged mice. The 2-month-old mice received a whole-body irradiation of 5 Gy. At 6 months after irradiation, we compared the immune functions of the irradiated mice with those of normal mice of the same age and with those of older. Interferon (IFN)-γ and antigen-specific immunoglobulin (Ig)G2a level were lower in the irradiated mice than in normal mice of same age, showing similar levels to those of old normal mice. In contrast, interleukin (IL)-4 and IL-5 and antigen-specific IgG1 level were increased in irradiated mice when compared with the same aged-normal mice. Next, we investigated the low expression of IL-12p70, IL-12 receptors and IL-18 receptors in irradiated and old mice. Also, the decrease of natural killer cell activity was intensified in the irradiated mice, showing lower than values to those of old mice. Interestingly, in irradiated mice, the absolute numbers and the percentages of natural killer (NK) cells was extremely decreased. But the absolute numbers of Th cells and cytotoxic T (Tc) cells in old mice were significantly decreased. In conclusion, an immunological imbalance by the whole-body irradiation of 5 Gy induces to persist in the long term, resulting in the similar results with aging. Our results suggest that the downregulation of the Th1-like immune response shown in old mice rapidly occurred through exposure of ionizing radiation.     

Human umbilical cord blood effect on sod mice (amyotrophic lateral sclerosis)

In previous studies we observed that human umbilical cord blood (HUCB) could have a protective effect on the onset of disease and time of death in MRL Lpr/Lpr mice which have an autoimmune disease that may be considered similar to human lupus. We believed a temporary xenograph may have occurred in these animals with the disease process delayed and the life span markedly increased. When HUCB is stored at 4 degrees C in gas permeable bags, there is a decrease of the cell reaction in mixed lymphocyte cultures. The blood, however, maintains a significant number of cells capable of producing replatable colonies. This study attempted to determine the effect of HUCB on SOD1 mice (transgenic B6SJL-TgN(SOD1-G93A)1GUR), which have a mutation of the human transgene, (CuZn superoxide dismutase gene SOD1) that has been associated with amyotrophic lateral sclerosis. We previously developed evidence that the survival of lethally irradiated mice was related to the number of human mononuclear cells administered. In the present study, we decided to investigate the effect of a relatively large dose of human mononuclear cord blood cells on SOD1 mice subjected to a sublethal dose of irradiation preceded by antikiller sera (rabbit anti-asialo). The SOD1 mice show evidence of paralysis at 4 to 5 months. The average expected lifetime of these mice is reported to be 130 days (Jackson Laboratory). In this experiment, there were 23 mice. Two mice died before the onset of paralysis. The remainder were divided into three groups: group I: control group of 4 untreated mice; group II: an experimental group of 6 mice treated with antikiller sera, 800 cGy irradiation plus 5 x 10(6) congenic bone marrow mononuclear cells; group III: another experimental group of 11 mice treated with antikiller sera, 800 cGy irradiation plus 34.2-35.6 x 10(6) HUCB mononuclear cells, previously stored for 17-20 days at 4 degrees C in gas permeable bags. The results were as follows: the average age at death was: (I) 127 days for the untreated control group, (II) 138 days for the group that received 800 cGy of irradiation and congenic bone marrow (BM) and (III) 148 days for the group that received irradiation and HUCB. (P < 0.001 HUCB vs control, p < 0.01 HUCB vs BM). The longest surviving mouse in each group was 131, 153, and 182 days old respectively. In summary, large doses of HUCB mononuclear cells produced considerable delay in the onset of symptoms and death of SOD1 mice. These preliminary results may not only indicate that amyotrophic lateral sclerosis is an autoimmune disease, but may also indicate a possible treatment for a devastating disease and possibly others.     

Selective alteration at the growth-hormone- releasing-hormone nerve terminals during aging in GHRH-green fluorescent protein mice

Growth hormone (GH) secretion decreases spontaneously during lifespan, and the resulting GH deficiency participates in aging-related morbidity. This deficiency appears to involve a defect in the activity of hypothalamic GH-releasing hormone (GHRH) neurons. Here, we investigated this hypothesis, as well as the underlying mechanisms, in identified GHRH neurons from adult ( approximately 13 weeks old) and aged ( approximately 100 weeks old) transgenic GHRH-green fluorescent protein mice, using morphological, biochemical and electrophysiological methods. Surprisingly, the spontaneous action potential frequency was similar in adult and aged GHRH neurons studied in brain slices. This was explained by a lack of change in the intrinsic excitability, and simultaneous increases in both stimulatory glutamatergic- and inhibitory GABAergic-synaptic currents of aged GHRH neurons. Aging did not decrease GHRH and enhanced green fluorescent protein contents, GHRH neuronal number or GHRH-fibre distribution, but we found a striking enlargement of GHRH-positive axons, suggesting neuropeptide accumulation. Unlike in adults, autophagic vacuoles were evident in aged GHRH-axonal profiles using electron microscopy. Thus, GHRH neurons are involved in aging of the GH axis. Aging had a subtle effect at the nerve terminal level in GHRH neurons, contrasting with the view that neuronal aging is accompanied by more widespread damage.     

Expansion of the population of double negative CD4-8- T alpha beta-cells in the liver is a common feature of autoimmune mice.

There have been several reports that double negative (DN) CD4-8- T alpha beta-cells might be responsible for the onset of autoimmune diseases in humans and mice. We previously revealed that such DN T alpha beta-cells are generated in the liver of autoimmune MRL-lpr/lpr mice. In the present study, we further characterize the histology of the liver in these mice by light and electron microscopic studies. An intensive accumulation of mononuclear cells in the liver was demonstrated and a significant proportion of these mononuclear lymphocytes was found to intimately interact with Kupffer cells or endothelial cells of the hepatic sinusoids. The majority of such lymphocytes were TcR+CD4-8-Pgp-1+ alpha beta-cells. Identification of DN T alpha beta-cells was then performed in various autoimmune model mice. Interestingly, all autoimmune mice tested (i.e., MRL-lpr/lpr, C3H/HeJ-gld/gld, BXSB, NOD, MRL(-)+/+ and NZB/W F1 mice), showed an increased proportion of DN T alpha beta-cells (greater than 11% among all MNC) in the liver when they became old and diseased. On the other hand, young and old normal mice and young autoimmune mice before the onset of disease did not have such a high proportion of DN T alpha beta-cells (less than 10%) in the liver. Among autoimmune mice, MRL-lpr/lpr and C3H/HeJ-gld/gld mice had lymphadenopathy, which consisted of DN T alpha beta-cells (greater than 25%), after the onset of disease. Autoimmune mice of the other strains had neither lymphadenopathy nor DN T alpha beta-cells in the periphery, even when they were diseased. These results suggest that the expansion of the DN T alpha beta-cell population in the liver is a common feature of autoimmune mice, irrespective of the information of lymphadenopathy.     

Influence of T-lymphocyte deprivation on the antileukemic effect of bacillus calmette-guérin in vivo

Summary The effect in AKR mice of T-lymphocyte deprivation in vivo, obtained by adult thymectomy plus/minus whole-body irradiation and bone-marrow reconstitution, was studied in the growth of grafted leukemia cells obtained from spontaneous AKR leukemia. Both thymectomized mice and mice subjected to thymectomy, whole-body irradiation, and bone-marrow reconstitution (B) had a lower take-frequency of graft leukemia than conventional mice. Growth of graft leukemia was inhibited by BCG treatment both in thymectomized mice and in B mice. Concomitant with the increased inhibition of leukemia growth, an increased incidence of wasting-like disease was observed. In vitro cytotoxicity studies revealed that spleen lymphoid cells from nonleukemic mice were cytotoxic to AKR leukemia cells, to nonmalignant AKR fibroblasts, and to other nonmalignant cells with H-2^k haplotype. The activity of this self-directed cytotoxicity was most marked in AKR mice with wasting-like disease. The presence of autocytotoxic cells was frequently associated with a positive direct Coombs' test. Immunofluorescence studies showed, further, that the cytotoxic activity was independent of retrovirus antigens as tested by indirect immunofluorescence with anti-MuLV antibodies. Adult thymectomy of AKR mice confers an increased antitumoral immune potential, but also an increased risk of development of serious autoimmune disease.