Search

An association between T‐cell lymphopenia and autoimmunity has long been proposed, but it remains to be elucidated whether T‐cell lymphopenia affects B‐cell responses to autoantigens. Human neonatal thymectomy (Tx) results in a decrease in T‐cell numbers and we used this model to study the development of autoreactivity. Two cohorts of neonatally thymectomized individuals were examined, a cohort of young (1–5 years post‐Tx, n = 10–27) and older children (>10 years, n = 26), and compared to healthy age‐matched controls. T‐cell and B‐cell subsets were assessed and autoantibody profiling performed. Early post‐Tx, a decrease in T‐cell numbers (2.75 × 109/L vs. 0.71 × 109/L) and an increased proportion of memory T cells (19.72 vs. 57.43%) were observed. The presence of autoantibodies was correlated with an increased proportion of memory T cells in thymectomized children. No differences were seen in percentages of different B‐cell subsets between the groups. The autoantigen microarray showed a skewed autoantibody response after Tx. In the cohort of older individuals, autoantibodies were present in 62% of the thymectomized children, while they were found in only 33% of the healthy controls. Overall, our data suggest that neonatal Tx skews the autoantibody profile. Preferential expansion and preservation of Treg (regulatory T) cell stability and function, may contribute to preventing autoimmune disease development after Tx.

Antibodies to DNA are important markers of various autoimmune diseases and can be pathogenic; however, their generation is not understood. We previously reported that anti-DNA antibodies could be induced in mice by idiotypic immunization to PAb-421, an antibody to a DNA-binding domain of p53. We now report that two monoclonal antibodies of moderate affinity (K(D) asymptotically equal to 10(-7)), raised from PAb-421-immunized mice, specifically recognized both PAb-421 and DNA. These antibodies feature multiple arginine residues in the antigen-binding site, a unique characteristic of disease-associated anti-DNA antibodies; nevertheless, these anti-DNA antibodies show specific complementarity to PAb-421 by competing with p53 for PAb-421 binding and recognize defined oligonucleotides with a specificity similar to that of p53. To study the structural basis for the cross-recognition of PAb-421 and DNA by the anti-DNA antibodies, we constructed computer models (fine-tuned by protein-protein docking) of PAb-421 and one of the monoclonal anti-DNA antibodies. The modeled structures manifested structural complementarity. Most notably, the modeled structure of PAb-421 resembled the structure of DNA by the positions of negatively charged groups and aromatic side chains. Thus, a protein molecule may mimic the structure of DNA and the elusive generation of anti-DNA antibodies could be explained by idiotypic immunity to a DNA-binding protein, like p53.

Prion diseases are caused by conformational alterations in the prion protein (PrP). The immune system has been assumed to be non-responsive to the self-prion protein, therefore, PrP autoimmunity has not been investigated. Here, we immunized various strains of mice with PrP peptides, some selected to fit the MHC class II-peptide binding motif. We found that specific PrP peptides elicited strong immune responses in NOD, C57BL/6 and A/J mice. To test the functional effect of this immunization, we examined the expression of proteinase-K-resistant PrP by a scrapie-infected tumor transplanted to immunized syngeneic A/J mice. PrP peptide vaccination did not affect the growth of the infected tumor transplant, but significantly reduced the level of protease-resistant PrP. Our results demonstrate that self-PrP peptides are immunogenic in mice and suggest that this immune response might affect PrP-scrapie levels in certain conditions.

The tumor suppressor molecule p53 features a regulatory domain at the C terminus that recognizes damaged DNA. Since damaged DNA might be involved in activating anti-DNA autoantibodies, we tested whether autoimmunity to the C terminus of p53 might mark murine systemic lupus erythematosus (SLE). We now report that MRL / MpJ-Fas(lpr) mice, which spontaneously develop SLE, produce antibodies both to the C terminus of p53 and to a monoclonal antibody (PAb-421) that binds the p53 C terminus. Anti-idiotypic antibodies to PAb-421 (sampled as monoclonal antibodies) could also bind DNA. Thus, the PAb-421 antibody mimics DNA, and the anti-idiotypic antibody to PAb-421 mimics the p53 DNA-binding site. This mimicry was functional; immunization of BALB / c mice to PAb-421 induced anti-DNA antibodies and antibodies to the C terminus of p53, and most of the mice developed an SLE-like disease. Immunization of C57BL / 6 mice to PAb-421 induced antibodies to p53, but not to its C-terminal domain. The C57BL / 6 mice also did not develop anti-DNA antibodies or the SLE-like disease. Thus, network autoimmunity to the domain of p53 that recognizes damaged DNA can be a pathogenic feature in SLE in genetically susceptible strains of mice.

The human HLA-DQ8 (A1*0301/B1*0302) allelic product manifests a strong association with insulin-dependent diabetes mellitus (IDDM). Previous biochemical studies of the major histocompatibility complex (MHC) class II I-A(g)7 protein of IDDM-prone non-obese diabetic mice produced controversial results. To better define the biochemical properties of IDDM-associated MHC class II molecules, we analyzed DQ8 proteins, in comparison to other DQ allelic products, by partially denaturing sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). We now report that DQ8 proteins have a normal peptide occupancy and lifespan in cells. Similar to I-A(g)7, DQ8 proteins formed only a minor fraction of SDS-stable complexes with peptides. Although this phenotype was not unique to DQ8, some DQ allelic products such as IDDM-protective DQ6 proteins were SDS resistant. The DQ9 allelic product, differing from DQ8 only at position (P) beta 57, was SDS stable, suggesting that non-Asp residues at beta 57 might decrease the SDS stability of DQ proteins. We identified a single peptide which specifically induced an SDS-stable conformation in DQ8 as well as in I-A(g)7 molecules. The residues at anchor P1 in this peptide were found to influence the SDS stability of both molecules. Together with our previous observation of similar binding motifs of I-A(g)7 and DQ8, these results demonstrate an overall biochemical similarity of mouse and human diabetes-associated MHC class II molecules. This similarity might contribute to a common immunological mechanism of IDDM in both species.

We previously reported that immunity to the p277 peptide of the human 60-kDa heat shock protein (hsp60) was a causal factor in the diabetes of non-obese diabetic (NOD) mice, which are genetically prone to develop spontaneous autoimmune diabetes. The present study was done to test whether immunization with the p277 peptide could cause diabetes in standard strains of mice. We now report that a single administration of the p277 peptide conjugated to carrier molecules such as bovine serum albumin or ovalbumin can induce diabetes in C57BL/6 mice and in other strains not genetically prone to develop diabetes. The diabetes was marked by hyperglycemia, insulitis, insulin autoantibodies, glucose intolerance and low blood levels of insulin. The diabetes could be transferred to naive recipients by anti-p277 T cell lines. Similar to other experimentally induced autoimmune diseases, the autoimmune diabetes remitted spontaneously. After recovery, the mice were found to have acquired resistance to a second induction of diabetes. Susceptibility to induced diabetes in C57BL/6 mice was influenced by sex (males were much more susceptible than were females) and by class II genes in the major histocompatibility complex (B6.H-2bm12 mice with a mutation in the MHC-II molecule were relatively resistant). Other strains of mice susceptible to induced diabetes were C57BL/KSJ, C3HeB/FeJ, and NON/Lt. BALB/c and C3H/HeJ strains were relatively resistant. Immunization to p277-carrier conjugates could also induce transient hyperglycemia in young NOD mice, but upon recovery from the induced diabetes, the NOD mice were found to have acquired resistance to later development of spontaneous diabetes. Thus, T cell immunity to the p277 peptide can suffice to induce diabetes in standard mice, and a short bout of induced diabetes can affect the chronic process that would otherwise lead to spontaneous diabetes in diabetes-prone NOD mice.

Self antigens bind to MHC class II molecules in vivo. The capacity of class II molecules to bind native self antigen, namely antigen that has not been processed through an endososomal pathway, could increase susceptibility to autoimmune diseases if recognized by autoreactive T lymphocytes. To confirm this prediction we designed experiments to show that: (a) native myelin basic protein (BP) activates encephalitogenic T lymphocyte lines, and (b) these activated lines cause experimental autoimmune encephalomyelitis (EAE) in naive rats. We show that two encephalitogenic T lymphocytes lines, LEW and BN, are activated by native BP in the presence of paraformaldehyde pre-fixed antigen-presenting cells (APC). The degree of activation, as measured by T lymphocyte proliferation, was equal to that obtained in response to BP processed by APC prior to paraformaldehyde fixing. The response to native BP was confirmed by demonstrating insensitivity to the presence of the lysosomotropic agent chloroquine or protease inhibitors. Activation of T lymphocytes by BP required the presence of syngeneic APC. The activated T lymphocyte lines were injected into naive recipient rats that developed EAE within 5 days. Both disease incidence and severity were equal to that observed in rats that were treated with T lymphocytes activated by processed BP. Hence, at least in EAE, the risk of an autoimmune event could be precipitated by a complex of native antigen and class II molecules on cells that do not possess an endososomal pathway for antigen processing and presentation.

The role of Treg in immune regulation is the topic of this Viewpoint series in the European Journal of Immunology (EJI); the question to be discussed in this section is the effector function of Treg in immune regulation. In this manuscript, we take on the following three postulates outlined by Rolf Zinkernagel on the role of Treg in the control of immunity. First, the immune response is regulated primarily by the antigen and not by Treg. Second, immune non-responsiveness results from the deletion of specific receptor-bearing T cells. Third, there is no definitive proof of the existence of specialized Treg that know what is needed for an equilibrated immune response. Herein, we discuss data demonstrating the existence of specialized Treg and therefore arguing against the validity of the first two postulates. However, based on the reactive nature of the immune system, we agree with Rolf's third postulate in that Treg cannot know ahead of time an ideal set-point for immune homeostasis.

Immunization with attenuated activated autoreactive T cell lines and clones induces a response in syngeneic animals which can induce protection or recovery from autoimmune disease. This process has been termed T cell vaccination. The aim of the present study was to investigate the effect of immunization with MHC-reactive T cells on the mixed lymphocyte reaction (MLR). By injecting attenuated activated T cells primed for an alloantigen, we markedly reduced the MLR in both rats and mice. This depression appeared to be mediated by active suppression; lymphoid cells from T cell-vaccinated animals suppressed the MLR responsiveness of T cells from naive animals. Suppression of the MLR was not restricted to the major histocompatibility complex (MHC) alleles used to prime the animals from which the T cell vaccines were prepared; the MLR to other MHC allelic stimulator cells was also suppressed. This MHC-unrestricted suppression could not be attributed to an anti-ergotypic response to non-MHC-linked activation markers on T cells; an anti-ergotypic response augmented rather than suppressed the MLR. We herein propose that T cell vaccination might influence the MLR by suppressing the responses of diverse T cells which bear shared T cell receptor idiotypes.

Behçet's syndrome is a multi-system inflammatory disease affecting mainly the oral and urogenital mucosa and the uveal tract. The etiology and pathogenesis of Behçet's syndrome are unknown, but autoimmune mechanisms are implicated. We initiated this work to identify self-antigens targeted by patients with Behçet's syndrome. We used patient sera to immuno-blot tissue lysates, and we found that some patients manifest antibodies to a 37-kDa band. The 37-kDa band was detected in extracts of skin, tongue, vagina, muscle and heart but not in brain, kidney, lung, liver, intestine and thymus. In-gel digestion and mass spectrometry revealed the band to be alpha-tropomyosin. Autoimmunity to alpha-tropomyosin can be pathogenic; immunized Lewis rats developed lesions in the uveal tract and skin, with features of Behçet's disease.

The expression of major histocompatibility complex (MHC) class II molecules in murine T cells has been controversial. We therefore reexamined the transcription, synthesis and surface expression of MHC class II determinants in rat T cells both in vivo and in vitro. In naive rats, a large proportion of small CD4+8+ and mature CD4+8-/CD4-8+ thymocytes was found to be MHC class II positive. At least some of the MHC class II molecules found on thymocytes were actively synthesized. The synthesis of MHC class II proteins was detected in peripheral T cells activated in vivo during induction of experimental allergic encephalomyelitis (EAE). A proportion of T cells from the inflammatory lesion of EAE exhibited MHC class II on the surface. A panel of helper T cell lines and clones was shown to synthesize MHC class II proteins. In a prototypic clone, a weak constitutive expression of MHC class II was observed. During activation, the rate of endogenous MHC class II synthesis increased and passive absorption of surface MHC class II from other cells occurred. Our data demonstrate the expression of MHC class II molecules in rat T cells in both the thymus and periphery. Since the primary function of MHC class II molecules is the presentation of peptide epitopes to T cells, these results call attention to the possible role of MHC class II molecules in T-T interactions during T cell maturation and activation.

It was shown previously that experimental autoimmune diseases could be prevented or treated specifically by administering suitably attenuated autoimmune T lymphocytes to animals, a process termed T cell vaccination (Cohen, I. R., Sci. American 1988. 256: 52). We now report that T cell vaccination is an effective way of inducing tolerance to contact sensitivity to simple chemical haptens. Vaccines were prepared from populations of lymph node cells from specifically sensitized mice by activating the T cells with the T cell mitogen concanavalin A and then treating the T cell blasts with glutaraldehyde. The vaccinated mice showed decreased delayed sensitivity responses to the specific sensitizing antigen and developed significant delayed sensitivity responses to the T cells of the same specificity as those used for vaccination. Thus, T cell vaccination against contact sensitivity reactions appears to function similarly to T cell vaccination against autoimmune disease.

Previously we reported that activated T lymphocytes express a heparanase enzyme that degrades the heparan sulfate moiety of the proteoglycan of the extracellular matrix (ECM). Expression of the heparanase enzyme was found to be associated with the ability of activated T lymphocytes to penetrate blood vessel walls and accumulate in target organs. We recently found that relatively low doses of heparin administered to mice or rats inhibited T cell-mediated immune reactions. In the present study we investigated the effects in vitro and in vivo of the heparanase inhibitor, heparin, on the expression of T lymphocyte heparanase and on the ability of T lymphocytes to mediate a delayed-type hypersensitivity (DTH) reaction. We found that heparanase was induced by immunizing mice with antigen in vivo or by activating T lymphocytes with concanavalin A in vitro. Relatively low doses of heparin administered once daily in vivo (5 micrograms) or present in vitro (0.1 microgram/ml) inhibited the expression of heparanase induced by immunization or by concanavalin A incubation. Higher or lower doses of heparin did not have these effects. The same doses of heparin that inhibited expression of heparanase also inhibited the ability of the lymph node cells to migrate to a site of antigen and adoptively produce a DTH reaction. These findings suggest that modulation of cell-mediated immune reactions may be achieved by relatively low doses of heparin which inhibit expression of T lymphocyte heparanase.

The restriction of the pathogenesis of experimental autoimmune thyroiditis (EAT) by H-2 gene products was investigated. EAT was induced by injecting thyroglobulin extract plus adjuvant into F1, hybrid mice that had been implanted under the kidney capsules with thyroid glands originating from either the EAT-susceptible or -resistant parental strain mice. We found relative H-2 restriction of thyroid damage to those glands originating from the H-2-susceptible parental strain. H-2 restriction of damage at the level of the target thyroid gland implicates cytotoxic effector T lymphocytes as a pathogenic agent of EAT.

We previously found that mouse T lymphocytes sensitized in vitro against allo- or syngeneic fibroblasts, upon injection into syngeneic recipients, do not themselves differentiate into effector cells, but recruit effector T lymphocytes within the draining lymph nodes. Asaresult of sensitization, these initiator lymphocytes acquire a trypsin-sensitive membrane property which is necessary for recruitment. We now report studies on the in vivo migratory behavior of initiator lymphocytes following sensitization. We injected 51Cr-labeled initiator lymphocytes into recipient footpads and found significantly increased migration of sensitized cells to the draining popliteal lymph node (PLN) during the first day. By amputation ofthe foot at various times, we showed that migration during the first 12—24 hours was critical for subsequent recruitment. Trypsin treatment of initiator lymphocytes abolished this accelerated migration. Lymphocytes triggered nonspecifically by Con A migrated to the PLN like antigen-sensitized cells. We also compared the migration of injected lymphocytes from the footpad to the PLN in graftversus-host and host-versus-graft reactions, and found these reactions to differ both from each other and from recruitment in terms of lymphocyte migration. These findings are discussed in terms of the physiology of the cell-mediated immune response and the notion of peripheral sensitization.

Genetic control has been studied of the response to the encephalitogenic nonapeptide (NP) determinant of myelin basic protein (BP) in inbred guinea pigs of strains resistant or susceptible to induction of experimental autoimmune encephalomyelitis (EAE). By studying bone marrow-reconstituted animals, we found that susceptibility to induction of EAE was a function of the genotype of the cells of the lymphohematopoietic system and not of the physiological environment or target organ. Analysis of the T cell response showed that susceptible strains 13 or (2 X 13)F1 hybrid guinea pigs recognized the NP determinant when injected with whole BP in adjuvant. Resistant strain 2 guinea pigs responded to undefined determinants on BP, but not to the NP moiety. We investigated the cells involved in regulating the response to the NP determinant by injecting susceptible F1 hybrids with BP-pulse macrophages of either parental strain. Susceptible strain 13 macrophages triggered a response to the NP determinant and induced clinical EAE. In contrast, F1 animals injected with resistant strain 2 macrophages failed to respond to the NP determinant, although the macrophages were capable of presenting other undefined determinants present on whole BP. Therefore, genetic control of the immune response to the NP determinant appears to be exerted at the level of antigen presentation by macrophages to T lymphocytes.

Lines of rat T lymphocytes responsive to the basic protein of myelin (BP) or to the purified protein derivative of Mycobacterium tuberculosis (PPD) were inoculated i.v. into recipient rats. As reported previously, the anti-BP line cells, but not the anti-PPD line cells spontaneously accumulated in the central nervous system and caused encephalomyelitis. However, the anti-PPD line cells could be induced to enter the brain and cause bystander encephalitis by intracerebral inoculation of PPD. Anti-PPD or anti-BP line cells could mediate delayed-type hypersensitivity skin reactions or bystander arthritis elicited by specific antigen. The lines did not cause specific cytolysis in vitro. Susceptibility to delayed-type hypersensitivity or bystander disease was long lasting in rats inoculated with anti-PPD line cells, while rats inoculated with anti-BP line cells were susceptible for only a few days. Thus, lines of T lymphocytes can mediate a variety of pathological reactions directed by the presence of specific antigen, self or foreign.

The cell-mediated immune response of animals to a lethal syngeneic tumor was investigated by inoculating C57BL mice with Lewis lung carcinoma (3LL) cells T lymphocytes, obtained from the enlarged spleens of the tumor-bearing mice were found to be cytotoxic to 3LL target cells in vitro. However, we found that such spleen cells enhanced tumor growth in vivo when mice were injected with a mixture of spleen cells and tumor cells. Removal of T lymphocytes by treatment of the spleen cells with anti-Θ serum plus complement reduced the enhancement of tumor growth. Hence, the tumor enhancing cells, like the cytotoxic cells, appeared to be T lymphocytes. Removal of T lymphocytes from normal mice by adult thymectomy before tumor inoculation led to a reduction in the number of tumor metastases. Thus, enhancing T lymphocytes appear to exist in normal as well as in tumor-bearing mice. Investigation of this mechanism of tumor enhancement suggested that the enhancing T lymphocytes act as suppressor T cells inhibiting natural immune resistance to tumor growth.

To analyze immunologic mechanisms involved in experimental autoimmune encephalomyelitis (EAE), a system for studying the cytotoxic reaction of guinea pig lymphocytes against syngeneic macrophages pulsed with myelin basic protein (BP) was developed. It was found that both EAE-susceptible strain 13 and EAE-resistant strain 2 animals developed cytotoxicity directed against whole BP. However, only strain 13 guinea pigs developed such cells in response to immunization with the encephalitogenic nonapeptide determinant of BP. Strain 2 macrophages appeared able to present the nonapeptide determinant to strain 13 lymphocytes. Anti-BP cytotoxic lymphocytes were found to be recruited by anti-BP initiator lymphocytes.

The development of T lymphocyte immunocompetence appears to depend upon the function of the thymus gland. To study this function in greater detail, we developed a method for preparing monolayer cultures of thymus reticulum cells. We found that populations of lymphocytes from thymus-deprived mice acquired the ability to produce T lymphocyte reactions when incubated with reticulum cells in vitro. These results provide evidence about the role of the thymus reticulum in inducing immunocompetence of T lymphocytes.

In earlier studies we found that initiator T lymphocytes sensitized in vitro against fibroblasts were able to recruit immunospecific effector T lymphocytes in vivo in syngeneic mice. In the present study, to investigate the possible function in effector cell recruitment of sensitizing antigens passively adsorbed onto the initiator cell membrane, we treated the initiator lymphocytes with trypsin. We found that trypsin inhibited recruitment. However, the initiator cells spontaneously recovered their recruiting ability after incubation in vitro for 4 h in the absence of contact with sensitizing antigens. Recovery could be blocked by cycloheximide. Hence, it appears that recruitment depends on the intrinsic metabolic function of the initiator lymphocytes and is not due merely to passively-adsorbed antigen carried over into the recipient mouse.

We have investigated the specific activation by soluble antigen of rat or mouse long-term T helper cell lines using antigen-presenting cell (APC)-free culture conditions. Some T cell lines specific for avidin or myelin basic protein responded to native antigen in the absence of added APC. Responses in the absence of APC were substantial and specific although, as would be expected, lower than in the presence of APC. Proliferation could not be inhibited by culture with anti-Ia antibodies and the ability of lines to respond to antigen in the absence of APC did not correlate with the endogeneous surface Ia expression of the lines. Furthermore, irradiated T cells were unable to act as presenting cells for lines cells of the same or a different specificity. This suggests that the T cells did not present antigen to each other, and demonstrates, along with other data shown, that activation cannot be attributed to undetected APC remaining in the cultures. Anti-avidin T cell lines differed markedly in their ability to respond to avidin in the absence of added APC.S2, an anti-avidin line of H-2s genotype consistently responded well to avidin seen in the absence of added APC; K2, an H-2k anti-avidin line, responded moderately and B3, and H-2b anti-avidin line, although the most prolific responder in the presence of APC, never responded to antigen in their absence. Z1a, a Lewis rat-derived T cell line specific for myelin basic protein, proliferated well in response to the antigen in the absence of added APC. The present findings demonstrate that some T cells can recognize and respond to native antigens without the mediation of specialized APC.

We have investigated the sequential interaction of macrophages, initiator T lymphocytes (ITL) and recruited T lymphocytes (RTL) in the development of a T cell-mediated response to soluble antigens. Macrophages were pulsed with soluble antigens and used to sensitize ITL in vitro. The ITL were irradiated to prevent their proliferation and then injected into the footpads of syngeneic mice. Sensitized ITL were found to recruit immunospecific RTL in the draining lymph nodes, as determined by a thymidine uptake assay of the lymph node cells in vitro. The richest source of lymphocytes with ITL activity was the thymus, and progressively less activity was detectable among spleen or lymph node lymphocytes. The magnitude of the subsequent RTL response could be modified by genetic differences between the ITL and the antigen-pulsed macrophages that were used to sensitize them. Thus, ITL conveyed an immunogenic signal to RTL whose magnitude reflected the genotype of the macrophages, but whose specificity was directed by determinants of the soluble antigen.

H-2 gene control of the response of T lymphocytes to antigenic determinants of ungulate insulins has been studied. We injected H-2-congenic mice with different insulins emulsified in complete Freund’s adjuvant and measured the proliferative response in vitro of the draining lymph node cells to various related insulins. Two groups of determinants were antigenic: the A chain loop determinant present on bovine and sheep insulins, but absent on pig insulin, and a pig-associated determinant(s) also shared by bovine and sheep insulins. The responses to these two sets of determinants were controlled by H-2 genes as follows: (a) The pig-associated determinant(s) was antigenic for H-2d but not for H-2’ or I-I-2k mice. (b) The A chain loop determinant was immunodominant over the pig-associated determinant(s) for H-2d mice. In the presence of the A chain loop determinant, H-2d mice did not respond to the pig-associated determinant(s) on the same molecule. (c) H-2d mice did not distinguish between the bovine and sheep variants of the A chain loop that differed by substitution of one amino acid. (d) In contrast, H-2k mice responded only to the sheep variant, while H-2b mice responded primarily to the bovine variant of the A chain loop determinant. (e) However, injection of H-2b mice with “forbidden” sheep insulin primed them for an in vitro response to the “permitted” bovine variant of the A chain loop determinant. Similarly, some H-2k mice injected with forbidden bovine insulin responded in vitro to the permitted sheep variant of the A chain loop determinant. Thus, H-2 gene products decided whether the immune response would express an affinity for the specific immunizing antigen itself and/or for a closely related variant.

This study presents a theoretical analysis of the phenotypic expression of the products of immune response (Ir) genes linked to the major histocompatibility complex. Recent evidence indicates that animals with “low”-responder Ir genotypes can be induced to express adequate immune responses by minor variations in the manner or site of immunization. Hence, it is unlikely that Ir genes code for defective structural products in low-responder animals. A quantitative model is presented of the behavior of Ir genes based on two principal elements: (a) clones of lymphocytes compete for prominence through processes of growth, differentiation and feedback suppression, and (b) Ir gene products influence the affinities of lymphocyte clones for antigenic determinants, without coding for the lymphocyte receptors for antigen. The theoretical analysis of this model leads to the conclusion that small differences in affinity on the level of single cells can be amplified through differential population growth and competition between clones of cells to produce a final response that appears “all-or-none” on a population level. Therefore, high or low responsiveness based on Ir genes can be explained by small shifts in the strength of interaction parameters. These small shifts can result from variation in H-2 alleles, the mode of immunization, or the array of determinants on the immunogen; factors that were observed experimentally. These shifts in interaction parameters account for the observed flexibility of phenotypic expression of Ir genes. The model explains a number of immunological observations by a few basic facts or assumptions, and avoids the problem of postulating the existence of defects to account for low responsiveness. It aims at clarifying the relationship between systemic, macroscopic observations and the underlying cellular parameters. The translation of cellular characteristics into patterns of behavior manifested by populations of cells may not necessarily result in an averaging out of the characteristics, but could also accentuate them.

We studied lines of rat T cells, specifically reactive against myelin basic protein (BP), that were functional in mediating autoimmune encephalomyelitis or in vaccinating rats against induction of active EAE. Herein we report that these functions depended on activation of the cells by incubation with BP or with a T cell mitogen prior to inoculation into recipient rats. Activation was accompanied by the exposure of membrane-binding sites specific for the lectin peanut agglutinin. Accumulation of activated line cells in the central nervous system and thymus gland was observed.

Using an in vitro mouse lymphocyte anti-fibroblast reaction (AFR), we recently reported that the addition of allogeneic stimulator lymphocytes to the sensitization phase of the AFR enhanced sensitization to fibroblast antigens as evidenced by a marked increase in the cytolytic activity of the sensitized lymphocyte population. In the present report we studied the mechanism of this helper effect by testing the capacity of cell-free media derived from 48-h mixed spleen cultures to enhance anti-fibroblast sensitization. We found that such cell-free media could produce a marked helper effect when applied to the sensitization phase of the AFR, but not when added to the cytolytic effector phase. The stimulator cells in the mixed lymphocyte culture (MLC) did not have to be syngeneic with the sensitizing fibroblasts in the AFR in order for the helper effect to be demonstrated. Lymphocytes sensitized to fibroblast antigens in the presence of MLC medium retained their specificity of the effector phase. Our data suggest that the MLC medium acts by enhancing differentiation processes of antigen-triggered lymphocytes. Generation of helper activity by the MLC was abolished by 1000 r irradiation of the responder cells. By using nu/nu and normal spleens, both from two different strains, as cell sources for the MLC, we found that the generation of helper activity depended on T cells capable of proliferation. Furthermore, stimulator lymphocytes differing from responder lymphocytes by non-H-2 alloantigens as well as by point mutation within the H-2 complex were capable of eliciting helper factor(s). Thus, soluble factor(s) produced in a MLC, which are dependent on T lymphocyte proliferation, have the capacity to enhance the sensitization of mouse lymphocytes against antigens present on allogeneic fibroblasts.

We compared induction of sensitization in the mixed lymphocyte reaction (MLR) and in a mouse anti-fibroblast reaction. Our study demonstrated that induction processes in these two reactions, both known to be T cell-mediated reactions, are governed by different factors. Moreover, these two reactions interact with each other and this interaction is manifested by the ability of a MLR occurring during the sensitization phase of the anti-fibroblast reaction to enhance specific cytolysis measured during the effector phase. This helper effect appears to be immunologically nonspecific, since it can be obtained by the use of third party, antigenically unrelated lymphocytes as stimulator cells in the MLR.

Autoimmune encephalomyelitis (EAE) can be induced in genetically susceptible rats by active immunization against myelin basic protein (BP) or by passive transfer of anti-BP lymphocytes. We have developed in vitro lines of T lymphocytes reactive only against BP that produce EAE upon i.v. inoculation of syngeneic rats. The object of the present study was to learn whether attenuated anti-BP line cells could be used to vaccinate rats against passive as well as active EAE. We inoculated rats i.v. with anti-BP line cells that were attenuated by irradiation or treatment by mitomycin C. A single inoculation was sufficient to protect about 70% of rats from subsequent EAE induced actively. However, even repeated vaccination could not protect against EAE mediated by passive transfer of anti-BP line cells. Thus, attenuated cells of EAE effector lines cannot vaccinate against the autoreactive effects of preformed effector T lymphocytes.

The isolation and propagation of functional antigen-specific lines of T lymphoblasts is described. These lines were found to recognize foreign or self antigens in association with accessory cells of syngeneic major histocompatibility complex genotype. Intravenous inoculation of a T cell reactive only against myelin basic protein led to development of clinical paralysis in syngeneic rats. Thus, it is possible to study biological function as well as antigen specificity using T cell lines.

Thymus (T) lymphocytes autosensitized in vitro were shown in previous studies to produce enlargement of draining popliteal lymph nodes upon injection into the footpads of syngeneic rats. Specific autoreactive effector lymphocytes were found to be recruited within these lymph nodes. In the present study, the cellular basis of lymph node enlargement in mice by autosensitized lymphocytes was compared with that produced in a graft-vs.-host (GvH) reaction. T lymphocytes of C3H mice were autosensitized against syngeneic fibroblasts in vitro for 16 to 18 h in the absence of serum, and 107 lymphocytes were injected into the footpads of syngeneic mice. Control lymphocytes were incubated without fibroblasts. The GvH reaction was produced by injecting 107 C3H T lymphocytes into the footpads of (C3H × C57BL)F1 adult recipients. The index of relative enlargement of the draining popliteal lymph nodes was measured 6 days after injection. Experiments were done to identify the origin of the lymph node cells in these reactions. Irradiation of the donor lymphocytes (1000 r) or the recipient mice (550 r) was used to prevent proliferation of the lymphocytes of either origin. The participation of recipient T lymphocytes in lymph node enlargement was investigated by using thymectomized mice. The following results were obtained. 1) The GvH lymph node enlargement was found to depend on proliferation of the donor T lymphocytes, but did not seem to require the participation of radiosensitive cells within the recipient mice. 2) In contrast, the response of the lymph nodes to autosensitized donor T lymphocytes depended on the function of radiosensitive T lymphocytes within the syngeneic recipients. The autosensitized donor lymphocytes themselves did not have to proliferate to recruit the response of recipient T lymphocytes. 3) It was found that recruitment of recipient lymph node cells could be super-imposed upon a conventional GvH reaction by presensitizing the C3H donor lymphocytes in vitro. Both autosensitization against syngeneic or allosensitization against C57BL fibroblasts augmented the lymph node response of (C3H × C57BL)F1 hybrid recipients. The recruited or donor components of these mixed responses could be selectively abolished by irradiating either the donor lymphocytes or the recipient mice. Hence, the autosensitization response, like the host-vs.-graft transplantation reaction, can be induced by sensitization of lymphocytes peripherally and involves recruitment of lymphocytes within regional lymph nodes. The GvH response manifested in the same popliteal lymph nodes does not appear to require the recruitment of radiosensitive T lymphocytes. These findings suggest that different classes of T lymphocytes function in the autosensitization and GvH responses.