Germ cell-induced immune suppression in mice. Effect of inoculation of syngeneic spermatozoa on cell-mediated immune responses.

Spleen cells from mice injected intravenously with syngeneic male germ cells exhibited reduced immune functions as determined by natural killer cell activity, mixed lymphocyte reactivity and cytotoxic lymphocyte (CTL) function. The decrease in CTL responses to trinitrophenyl-modified self (TNP-self) was detected as early as 4 d after sperm injection and was observed to H-2 alloantigens 3 wk after injection. Radiosensitive suppressor T cells were found to suppress the CTL response to TNP-self. Suppression lasted for a period of at least 7 wk after a single inoculation of the germ cells. Some variability in immune suppression capability was observed using different preparations of germ cells which are not yet completely understood. Sperm were more effective in inducing suppression than testicular cells derived from the seminiferous tubules. Furthermore, sperm from older animals were more effective than those from younger mice. These findings are discussed with respect to possible regulatory influences of germ cells on the immune system when the blood-testes barrier is broken.

Induction of tolerance in vitro by autologous murine testicular cells.

We have investigated the regulation of self tolerance in mice by examining lymphocyte reactivity in vitro against two subpopulations of autologous testicular cells: germ cells that were derived from the seminiferous tubules, and interstitial somatic cells. In the presence of germ cells, lymphocyte proliferation was strongly reduced. In contrast, somatic interstitial cells stimulated lymphocyte proliferation. In both cases, reactive lymphocytes were mostly T cells. Suppressor T cells activated by autologous germ cells were nonspecific and capable of inhibiting lymphocyte proliferation against autologous and allogeneic somatic testicular cells as well as against allogeneic spleen cells. Suppression was abrogated after treatment of the responder lymphocytes with anti-Ly-2.2 serum plus complement. Lymphocyte proliferation by autologous interstitial cells was considerably reduced, but not completely abolished, by complement-dependent lysis with anti-Thy-1.2 serum. This may indicate the participation in proliferation of a lymphoid cell population other than T cells.

Protection against graft vs. host-associated immunosuppression in F1 mice. I. Activation of F1 regulatory cells by host-specific anti-major histocompatibility complex antibodies.

Injection of parental spleen cells into unirradiated F1 hybrid mice results in suppression of the potential to generate cytotoxic T lymphocyte (CTL) responses in vitro. In an attempt to protect the F1 mice from immunosuppression, the recipients were injected with antibodies specific for major histocompatibility complex (MHC)-encoded antigens of the F1 mice 24 h before inoculation of the parental spleen cells. 8-14 d later, the generation of CTL responses in vitro against H-2 alloantigens was tested. Alloantiserum directed against either parental haplotype of the F1 strain markedly diminished the suppression of CTL activity. Furthermore, monoclonal antibodies recognizing H-2 or Ia antigens protected the F2 mice from parental spleen cell-induced suppression. Although this study has been limited to reagents that recognize host H-2 determinants, these findings do not necessarily imply that protection against graft vs. host (GvH) can be achieved only with anti-MHC antibodies. However, protection was observed only by antibodies reactive with F1 antigens, and small amounts of the alloantibodies were sufficient to diminish CTL suppression. Adoptive transfer of spleen cells from syngeneic F1 mice treated with anti-h-2a alloantiserum 24 h previously provided protection equal to that of injection of the recipients with alloantibodies. The cells necessary for this effect were shown to be T cells and to be radiosensitive to 2000 rad. This cell population is induced by antisera against F1 cell surface antigens and effectively counteracts GvH-associated immuno-suppression.

Prevention of autoimmune diabetes in non-obese diabetic mice by treatment with a class II major histocompatibility complex-blocking peptide.

The role of antigen presentation as a possible mechanism underlying major histocompatibility complex (MHC) association of autoimmune disease has been studied in non-obese diabetic (NOD) mice. By screening for inhibition of antigen presentation to NOD T cell hybridoma, we have selected a synthetic peptide, yTYTVHAAHAYTYt (small letters denote D amino acids), that efficiently blocks antigen presentation by the NOD class II MHC molecule A alpha g7A beta g7 (Ag7) in vitro. The inhibition is MHC selective, in that it does not affect antigen presentation by the E(d) and E(k) molecules, and has only a marginal effect on presentation by the A(d) molecule. This peptide also inhibits the priming for Ag7-restricted T cell responses in vivo, and prevents the spontaneous development of diabetes in female NOD mice, when administered chronically from 3 wk of age on. Chronic treatment with a control peptide, KMKMVHAAHAKMKM, that fails to bind to Ag7 has no effect on the disease. These data indicate that antigen presentation by the Ag7 molecule plays a pivotal role in the induction of autoimmune diabetes. Furthermore, the results demonstrate that interference with antigen presentation by a class II molecule can prevent the onset of spontaneous autoimmune disease associated with the same molecule.

Graft-versus-host resistance induced by class II major histocompatibility complex-specific T cell clones.

Possible mechanisms of graft-vs.-host (GVH) resistance have been studied using a panel of seven class II major histocompatibility complex-specific T cell clones for elicitation and challenge. One clone recognized I-Ak,d,f, and expressed V beta 8.3 together with J beta 1.5. The remaining six clones were I-Ek specific and expressed V beta 15 rearranged to J beta 1.1 or J beta 1.3. The I-Ek-specific clones were also homologous to each other and different from the I-A-reactive one in the D and N regions. Four of the seven clones exhibited I-Ek-specific cytolytic activity. Each clone, when injected in sublethal numbers into appropriate recipients, could induce resistance to a subsequent lethal dose of any other clone in the panel. The resistance did not require sharing of either T cell receptor beta chains or antigen specificity, or MHC molecules by the eliciting and challenging clone. Cytolytic and noncytolytic clones were equally efficient in inducing GVH resistance. A prerequisite of resistance induction was the activation of eliciting clone subsequent to recognition of class II molecules in the host. Clones preactivated with high concentrations of recombinant interleukin 2, in vitro, could induce GVH resistance also in syngeneic hosts, suggesting that resistance induction was associated with the activated state of clone, rather than antigen recognition per se. In all instances of resistance, the challenging clones failed to induce vascular leakage, which was the cause of death in susceptible recipients (Lehmann, P. V., G. Schumm, D. Moon, U. Hurtenbach, F. Falcioni, S. Muller, and Z. A. Nagy. 1990. J. Exp. Med. 171:1485). Lipopolysaccharide (LPS) induced resistance to vascular leakage did not provide crossresistance to GVH and vice versa, suggesting that interleukin 1 alpha and tumor necrosis factor alpha implicated in LPS resistance are not involved in GVH resistance. Although the mechanism remains unclear, the most likely explanation for GVH resistance in this system is either the downregulation of permeability increasing effect in the challenging clone, or an induced refractoriness of blood vessels to this effect.

Acute lethal graft-versus-host reaction induced by major histocompatibility complex class II-reactive T helper cell clones.

T cell clones isolated from class II MHC-disparate MLR combinations, and specific for I-Ak and I-Ek molecules, respectively, are shown to induce acute lethal graft-vs-host disease in unirradiated recipients. Cytolytic and noncytolytic clones are equally efficient in this respect. The lethal disease is dependent on recognition of the stimulatory class II molecules in the host. The clones home to lungs and liver, and become activated in these organs as demonstrated by an in vivo thymidine incorporation assay. After activation, a severe vascular leak syndrome develops causing death of the recipients within 5 d after the injection of 5 x 10(6) to 10(7) cloned cells. The disease develops without the participation of secondary host-derived inflammatory mechanisms, such as mast cell degranulation, complement activation, and the release of prostaglandins, oxygen radicals, or proteolytic enzymes. The results raise the possibility that Th cells can directly influence vascular permeability, and control, thereby, the acute inflammatory reaction of blood vessels.

Resistance of young gelatinase B-deficient mice to experimental autoimmune encephalomyelitis and necrotizing tail lesions.

Regulated expression of matrix metalloproteinases (MMPs) and their inhibitors (TIMPs) plays a role in various physiological processes. To determine in vivo how unbalanced expression of these factors can promote or affect the course of pathologies, we knocked out the mouse gelatinase B gene by replacing the catalytic and zinc-binding domains with an antisense-oriented neomycin resistance gene. Adult gelatinase B-deficient mice and wild-type controls could be induced to develop experimental autoimmune encephalomyelitis (EAE) with similar scores for neurologic disease, blood-brain barrier permeability, and central nervous system histopathology. However, whereas diseased control animals showed necrotizing tail lesions with hyperplasia of osteocartilaginous tissue, adult gelatinase B-deficient mice were resistant to this tail pathology. Gelatinase B-deficient mice younger than 4 weeks of age were significantly less susceptible to the development of EAE than were age matched controls and, even as they aged, they remained resistant to tail lesions. These data illustrate that gelatinase B expression plays a role in the development of the immune system and that, in ontogenesis, the propensity to develop autoimmunity is altered by the absence of this MMP.

Murine cortical brain cells are autoantigenic from a distinct developmental stage onwards.

The expression of autoantigens on murine cortical brain cells and their first appearance during development was studied. Autoreactivity was analyzed by weight increase and lymphocyte proliferation in the popliteal lymph node (PLN). Cortical brain cells or defined plasma membrane preparations were injected s.c. without adjuvant into syngeneic recipients. Weak, but significant T cell-dependent PLN enlargement was triggered with brain cells from adult mice. A stronger reaction could be elicited with one defined fraction of purified plasma membranes. The earliest appearance of the antigenic material in the plasma membrane fraction was observed on day 15 after birth. This time point correlates exactly with the completion of the blood-brain barrier in large parts of the central nervous system.

Modulation of T cell differentiation, antigen expression, and radiosensitivity of suppressor cells during late phases of graft-versus-host reactivity.

Suppressor cells generated by inoculation of (B10 X B10.A)F1 hybrid mice with parental spleen cells were characterized during the late period of graft-versus-host (GVH) reactivity, starting on day 14 after parental spleen cell injection and continuing until loss of suppressor activity. The suppressor cell potential was assessed by the ability of spleen cells from the injected F1 mice to suppress the generation of allogeneic cytotoxic T lymphocyte (CTL) responses in vitro by normal F1 responder cells. GVH-associated suppression could be abrogated by anti-Lyt 1, anti-Lyt 2, and anti-Thy 1 plus complement (C') before day 7 postinoculation, and by anti-Lyt 2 and anti-Thy 1 thereafter. However, from day 14 postinoculation on, suppressor cells were insensitive to either of these antibodies. This phase was associated with strong suppressor potential. Subsequent to this period suppressor cells were again sensitive to anti-Thy 1, but not to anti-Lyt 1 or anti-Lyt 2 plus C'--and at the same time they decreased in suppressor activity. Moreover, during the period of resistance to anti-T-cell antibodies plus C', suppressor cells were not killed by anti-asialo GM1 plus C'; were nonadherent to Sephadex G10, and were enriched by nylon-wool passage. In the course of GVH reactivity, suppressor cells alternated in sensitivity and resistance to 500 rads and once went through a brief period of resistance to 2000 rads that was associated with strong suppression and occasionally with mortality. The kinetics of these changes in cell surface phenotype and radiosensitivity were found to vary, coinciding with the strength of the suppression, but the sequence of these events was always the same. During the entire period of GVH reactivity, suppression was mediated by parental cells.

Limited heterogeneity of T-cell receptor V beta gene expression in the early stage of insulitis in NOD mice.

Non-obese-diabetic (NOD) mice spontaneously develop type I diabetes. The disease starts with T cells infiltrating the islets of Langerhans. We therefore examined the T-cell receptor (TCR) V beta region repertoire in islet infiltrates from individual female NOD mice from 4 to 10 week old by polymerase chain reaction (PCR) using V beta 1-V beta 17 specific oligonucleotides. The study revealed a limited heterogeneity of TCR V beta transcripts with a predominance of V beta 1 at the onset of insulitis, i.e. at 4 weeks of age. The TCR VDJ beta sequences of the V beta 1 PCR fragments were identical in most of the individual mice. Among several different mice, similarities in the beta-junctional regions were detected. In contrast, a large heterogeneity of TCR V beta usage was found in mice with advanced insulitis, i.e., from 6 weeks of age on. Thus, these data suggest a limited heterogeneity of TCR V beta usage with a predominance of V beta 1 at the initiation phase of the disease.

The outer surface lipoprotein A of Borrelia burgdorferi provides direct and indirect augmenting/co-stimulatory signals for the activation of CD4+ and CD8+ T cells.

Naive CD4+ and CD8+ T cells require two distinct signals to proliferate and to express effector functions [1]. One is provided by the antigen receptor on the T cell (TCR) after its encounter with antigenic peptides associated with class I or II major histocompatibility complex (MHC) molecules on antigen-presenting cells (APC) [2]. The second signal, which is not antigen-specific but essential for proliferation and differentiation of naive T cells, is provided by co-stimulatory structures. The major co-stimulatory molecules for CD4+ T cells seem to be B7 [3], B7.2 [4,5], and heat-stable antigen (HSA) [6]. These molecules are expressed on a variety of naive and/or activated APC and bind to CD28 and CTLA-4 and possibly other, as yet undefined, TCRs [3,7]. Optimal T cell activation only occurs when co-stimulatory molecules and ligands for the TCR are expressed on the same APC [8,9]. However, co-stimulation for T cells may also be provided via bystander cells [8,9] or by glycoproteins of the extracellular matrix, like fibronectin [10] and laminin [11]. In this case, T-cell VLA integrins function as signaling molecules [10,11]. This indicates that antigen-specific T-cell activation may also occur in areas where antigens are presented in association with extracellular matrix proteins. The recent finding that the invasion protein of Yersinia spp. delivers co-stimulatory signals to anti-CD3-activated human T cells, most probably through the b1 integrins, suggests that bacterial products can also bind to contribute to the activation of T cells [12].(ABSTRACT TRUNCATED AT 250 WORDS)

Modulation of morphology and plasma-membrane enzyme activities of rat thymocytes induced by specific antisera.

Lewis rat thymocytes were incubated with different ligands: specific rat alloantisera, rabbit xenoantisera against whole-rat thymocytes or against thymocyte plasma-membrane vesicles and the two mitogens: concanavalin A and the Ca++ ionophore A 23187. After treatment, a crude plasma-membrane fraction was repaired, and the activities of two plasma-membrane marker enzymes, alkaline phosphatase and gamma-glutamyl transferase, and a general membrane marker enzyme, lysolecithin acyltransferase, were determined. An increase of all marker enzyme activities was observed only when thymocytes had been incubated with alloantiserum directed against the gene products of their major histocompatibility complex (MHC) or with rabbit antiserum against syngeneic thymocytes. Anti-MHC alloantiserum against a nonrelevant haplotype increased moderately the gamma-glutamyltransferase activity. Alloantiserum directed against the weak histocompatibility antigens had no significant effect as had rabbit antiserum raised against thymocyte plasma-membrane vesicles. The mitogens concanavalin A and A 23187 both increased the activity of the alkaline phosphatase and lysolecithin acyltransferase. Scanning electron microscopy showed that treatment with alloantisera did not alter the cell shape drastically. In contrast, incubation with rabbit xenoantiserum against thymocytes resulted in cell rounding and deformation. Rabbit xenoantiserum against the plasma-membrane vesicles of thymocytes resulted in markedly disturbed or damaged plasma membranes.

Type I diabetes in NOD mice is not associated with insulin-specific, autoreactive T cells.

In the present study the hypothesis was tested that T cells specific for autologous insulin may be involved in beta cell destruction. Lymphoid cell populations from non-immunized, non-obese diabetic (NOD) mice were investigated for spontaneous T-cell reactivity in vitro to rat insulin (identical to mouse insulin) and to porcine insulin (identical to mouse insulin in the immunogenic part of the A chain loop). No significant T-cell proliferation was detected. In vivo priming of non-diabetic NOD mice with rat insulin in CFA, or of diabetic or non-diabetic NOD mice with porcine insulin failed to elicit insulin-specific T-cell responses upon restimulation in vitro. Lymph node cells from NOD mice primed with porcine insulin and treated with anti-Lyt 2 antibodies and C also failed to show insulin-specific reactivity, indicating that suppression by Lyt 2 cells is not involved in the non-responsiveness observed. In addition to porcine and rat insulin, NOD mice were also non-responders to bovine insulin; however, they responded to equine and ovine insulin and to the oxidized chain B of bovine insulin, the latter showing no cross-stimulation in vitro to any of the intact insulin variants. In conclusion, this study indicates that autologous insulin does not serve as autoantigen in the autoimmune destruction of beta cells.

Specific sensitization of Lyt-1+2- T cells to spleen cells modified by the drug D-penicillamine or a stereoisomer.

A mouse model for assessing the sensitization of T cells to the drug D-penicillamine (D-Pen) has been established. Mice were primed with D-Pen in vivo and the secondary response of specific T cells was measured in a proliferative assay in vitro. Priming was achieved by injecting either free D-Pen or, more effectively, D-Pen in complete Freund's adjuvant at the base of the tail. After 7 to 9 days, the draining lymph node cells (responder cells) were restimulated in vitro with syngeneic spleen cells that had been preincubated with D-Pen for 20 to 22 hr and washed (D-Pen-SC). The thiol (-SH) group of D-Pen was required for generating effective stimulator cells; other thiol compounds, however, or heavy metals (Hg, Au) were unable to generate cross-reacting stimulators on incubation with spleen cells. Although D-Pen-SC proved to be good stimulator cells, thymocytes and mouse erythrocytes, after having been preincubated with D-Pen, completely failed to stimulate. The proliferating responder cells were Lyt-1+2- T cells that were highly specific for D-Pen-SC and were even capable of distinguishing between the two stereoisomers D-Pen and L-Pen. The splenic T cells of recipient mice were effectively primed by injecting D-Pen-SC, but not free D-Pen, via the i.v. route. These findings indicate that D-Pen can act as a hapten for specific T cells when presented on the surface of appropriate stimulator cells. Conceivably, if D-Pen were able to generate effective stimulator cells in vivo, this could lead to T cell reactions comparable to those in the graft-vs-host reaction (GVHR). We discuss the possibility that the GVH-like adverse immunologic side-effects of D-Pen in patients may have a GVHR-like pathogenesis.

Cyclosporin A inhibits the in vivo production of interleukin-1beta and tumour necrosis factor alpha, but not interleukin-6, by a T-cell-independent mechanism.

The effect of cyclosporin A (CsA) on cytokine production in the tissue chamber model of acute inflammation was investigated. CsA caused a dose-related inhibition of interleukin 1beta(IL-1beta) production in both normal and athymic mice, confirming earlier conclusions that this effect is not T cell dependent (ED50s 40 and 53 mg/kg p.o., respectively). Tumour necrosis factor alpha (TNF-alpha) levels were similarly affected with ED50s of 40 and 58 mg/kg p.o. for normal and athymic mice, respectively. By contrast, CsA inhibited interleukin 6 (IL-6) production only in normal mice (ED50 27 mg/kg p.o.) Differences in the absolute production of the three cytokines in normal and athymic mice were also noted. IL-1beta and IL-6 levels were two-fold higher in athymic mice, while for TNF-alpha, there was no difference between the two groups. The present findings support the authors' original hypothesis, that the inhibitory mechanism of CsA on IL-1beta is not mediated via T cells. The same mechanism also seems responsible for the inhibition of TNF-alpha production, but not for IL-6, where inhibition by CsA appears to require the presence of T cells.

Analysis of murine T lymphocyte markers during the early phases of GvH-associated suppression of cytotoxic T lymphocyte responses.

Graft-vs-host- (GvH) associated immune-suppression was generated by injecting nonirradiated (B6 X C3H)F1 hybrid mice with 40 to 60 X 10(6) B6 or C3H parental spleen cells. Susceptibility to suppression as well as the potential of parental donor lymphocytes to generate suppressor activity was investigated in allogeneic cytotoxic T cell responses by mixing spleen cells from the inoculated F1 recipients with syngeneic responder cells from untreated mice. The susceptibility to suppression in F1 recipients, as well as the potential of parental lymphocytes to induce suppression in F1 mice, varied with the age of the mice. Cellular analysis revealed that GvH-suppression was mediated by 500 R radiosensitive parental T cells during the period tested (from day 5 until day 8 after injection of B6 parental spleen cells). In contrast, selected F1 lymphocytes from recipients injected with parental spleen cells failed to exhibit suppressor cell activity. The sensitivity of parental GvH suppressor cells to anti-Lyt reagents was time dependent. Thus, suppressor cells were sensitive to monoclonal anti-Lyt-1 or anti-Lyt-2 and C (usually until day 7 after injection of parental spleen cells); subsequently, suppression became insensitive to anti-Lyt-1, but remained sensitive to anti-Lyt-2 and C treatment.