Cytokines in the induction and circumvention of peripheral tolerance.

Injection of deaggregated (monomeric) human gamma globulin (DHGG) into mice induces a state of immunological tolerance to subsequent challenge with immunogenic forms of HGG. Tolerance was shown to be induced in both the Th1 and Th2 CD4+ subsets. These mice fail to demonstrate antibody production, T cell proliferation, cytokine release, or T cell helper function. On the other hand, simultaneous injection of lipopolysaccharide (LPS) as well as interleukin-1 (IL-1) was capable of interfering with the induction of tolerance to DHGG. The purpose of the present study is to extend these investigations to determination of the cellular mechanisms responsible for the interference of tolerance induction in both CD4+ T cell subsets. It was demonstrated that LPS and IL-1 have differential effects on the induction of tolerance in the CD4+ subsets. As evidenced by immunoglobulin G (IgG) subclass, T helper cell function, and lymphokine secretion, coinjection of LPS with tolerogen interfered with the induction of tolerance in both subsets, whereas IL-1 interfered with the induction of tolerance exclusively in the Th1 subset. LPS and IL-1 had differential effects on the interference of the induction of tolerance in LPS-resistant mice where IL-1, but not LPS, was effective. In contrast to IL-1, IL-12 injected along with DHGG failed to interfere with the induction of tolerance in either Th1 or Th2 cells. Previous studies that demonstrated tolerance in the DHGG models induced in both the Th1 and Th2 subsets was further suggested by the demonstrations in the present study that dose-response curves for the induction of tolerance are identical in both subsets. The above findings taken together are compatible with the suggestion that tolerance induction results from the lack of cytokine production, which then prevents the expansion of Th1 and Th2 subsets following activation of the CD4+ precursor T cell. Furthermore, they support the general opinion that the expansion of these two subsets involve different cytokine pathways and that LPS and IL-1 most likely act through different cell receptors.

CD4+ T-cell subsets and cytokines involved in peripheral tolerance.

Peripheral tolerance is induced under conditions that avoid activation of antigen-presenting cells (APCs) to release cytokines. Such tolerance occurs in both CD4+ T helper (Th)-cell subsets (Th1 and Th2), probably because it is induced in precursor cells. By contrast, activation of APCs to release cytokines by immunization or infection activates either both subsets or predominantly one of them. A model for CD4+ T-cell tolerization and subset expansion is presented here.

Induction of tolerance to human gamma-globulin in FcR gamma- and Fc gammaRII-deficient mice.

FcR gamma-deficient mice were used to examine the role of Fc gamma receptors in the induction of peripheral tolerance to human gamma-globulin (HGG). FcR gamma-deficient mice injected with HGG in adjuvant demonstrated a CD4+ T cell response to in vitro challenge with HGG, as assayed by proliferation, cytokine secretion, and Ag-specific help for B cell Ab production. In vitro kinetics of Ag-specific proliferation were similar in both conventional and knockout mice. Peripheral tolerance could be established in these mice with a single dose of deaggregated protein, despite the lack of functional Fc gammaRI, the high affinity receptor for monomeric IgG. Establishment of unresponsiveness was observed at both the T and B cell levels. T cell tolerance was manifested in the reduction of T cell helper function and Ag-induced release of Th1- and Th2-like cytokines, as well as decreased proliferation to Ag-specific stimulation. B cell tolerance was demonstrated in knockout and normal mice by failure to detect HGG-specific Ab production using an immunization protocol for Ab production that bypasses the need for Ag-specific T cells. These results demonstrate that induction of tolerance in CD4+ cells to HGG does not require transduction of a signal through Fc gammaRI. Furthermore, the ability to induce tolerance to HGG in B cells in Fc gammaRII-deficient mice suggests that down-regulation of Ag-specific B cells through Fc gammaRII is not the mechanism by which B cell tolerance is induced. However, Fc gammaRII plays a role in regulating the immune response since the Ab response to immunogenic HGG in Fc gammaRII-deficient mice is markedly enhanced.

CD4+ T cell activation and tolerance induction in B cell knockout mice.

B cell knockout mice microMT/microMT were used to examine the requirement for B cell antigen (Ag) presentation in the establishment of CD4+ T cell tolerance. CD4+T cells from microMT mice injected with exogenous protein Ag in adjuvant responded to in vitro challenge by transcription of cytokine mRNA, cytokine secretion, and proliferation. Peripheral tolerance could be established in microMT mice with a single dose of deaggragated protein. This tolerance was manifested by a loss of T cell proliferation and cytokine production (including both T helper cell type 1 [Th1]- and Th2-related cytokines), indicating that B cells are not required for the induction of peripheral T cell tolerance and suggesting that the dual zone tolerance theory is not applicable to all protein Ags and is not mediated through Ag presentation by B cells.

Intervention of CD4+ cell subset shifts and autoimmunity in the BXSB mouse by murine CTLA4Ig.

In the BXSB autoimmune disease-prone mouse strain, male mice develop severe lupus-like symptoms and die early in life (4-6 mo), whereas females do not. We have previously demonstrated that profound phenotypic and functional changes occur with age in CD4+ cells from BXSB males. CD4+ cells from males (4 mo old) were predominantly CD44high, CD45RBlow, and MEL-14low (activated/memory phenotypes), while the reciprocal phenotypes characteristic of naive cells were prevalent in age-matched females and young adult males (2 mo old). CD4+ cells from older males proliferated less and produced less IL-2 and IFN-gamma than cells from either females or young males in response to immobilized anti-CD3 mAb. We tested the effect of CTLA4Ig treatment on the progression of disease in BXSB males. CD4+ cells from CTLA4Ig-treated mice at 4 mo of age were predominantly CD44low, CD45RBhigh, and MEL-14high phenotypes that were identical with those observed in CD4+ cells from young (3-mo-old) females. In contrast, control male mice treated with IgG2a accumulated the CD4+ memory phenotype. CD4+ cells from 4-mo-old male CTLA4Ig-treated mice proliferated and produced IL-2 at levels similar to those of cells from females in response to immobilized anti-CD3 mAb. Furthermore, in contrast to IgG2a-treated mice, female and CTLA4Ig-treated male mice at 4 mo of age produced no anti-chromatin Abs. Three of four male mice injected with CTLA4Ig until 6 mo of age appeared healthy at 8 mo of age, whereas all five of IgG2a-treated control males died by 6 mo of age. These 8-mo-old CTLA4Ig-treated males showed variable resistance to autoimmunity as well as function and phenotype marker expression, and a less striking glomerulonephritis than 4-mo-old untreated males. The results of this study demonstrate that the rampant T cell activation and T cell dysfunction that occur in male BXSB mice by 4 mo of age are abrogated by blocking the CTLA4-dependent costimulatory signal(s). They also show that treatment with CTLA4Ig can suppress the pathogenesis of disease and increase longevity.

In vivo induction of tolerance in murine CD4+ cell subsets.

The induction of tolerance in mice to preparations of deaggregated human gamma globulin (DHGG) results in in vitro antigen-specific unresponsiveness in CD4+ T cells as well as in both the T helper 1 (Th1) and Th2-like subpopulations. Whereas both CD45RB(hi) and CD45RB(lo) cells from lymph nodes of HGG/complete Freund's adjuvant-immunized mice (control) proliferated in vitro to HGG, both subpopulations from mice previously tolerized with DHGG failed to respond. Furthermore, CD4+ T cells from control, but not from DHGG-injected mice, secreted high levels of interleukin 2 (IL-2) after in vitro stimulation with HGG. Although significant levels of IL-4 in supernatants of control CD4+ cells stimulated with HGG were detected in some, but not all, experiments, significant levels of IL-4 were never detected in supernatants of HGG-stimulated tolerant CD4+ cells. The demonstration that serum IgG1 anti-HGG is preferentially produced in a few tolerant mice that exhibit a leaky tolerant state suggests that tolerance induction may be more difficult to induce in IL-4- than in IL-2-producing cells.

The expression of CD45RB on antigen-responsive CD4+ lymphocytes: mouse strain polymorphism and different responses to distinct antigens.

The levels of CD45RB expression by HGG-specific CD4+ cells residing in the Ag-draining lymph nodes of HGG-primed CBA/CaJ mice were analyzed. When sorted populations of CD4+, CD45RBhi, and CD4+, CD45RBlo cells were cultured with HGG and Ag-presenting cells, the majority of the proliferative response was found in the CD45RBlo fraction early after in vivo priming (Day 6), and this pattern remained stable through 12 days postpriming. To determine whether this segregation of responsiveness was consistent in other mouse strains, HGG-primed C57BL/6J mice were similarly analyzed. In contrast to findings with the CBA/CaJ strain, the CD4+, CD45RBhi cell fraction obtained from C57BL/6J mice was the predominant responding population early after in vivo priming (Day 6); however, there was a parallel increase in responsiveness of CD4+, CD45RBhi, and CD4+, CD45RBlo cells by Day 12. Thus, there was not a decrease in CD45RBhi expression with a concommitant increase in CD45RBlo expression in CD4+ cells proliferating to HGG. Despite the heterogeneity in CD45RB expression by the primed CD4+ cells of the two strains, the entire proliferative response to HGG early after priming resided in the fraction bearing high levels of membrane CD44, thus arguing for the existence of CD45RBhi, CD44hi and CD45RBlo, CD44hi cells during the early phase of the response. In both mouse strains the CD4+, CD45RBhi subset of primed lymph node cells produced significant levels of IL-2 in response to HGG and APC, whereas no significant IL-2 or IL-4 production was detectable in HGG-stimulated CD45RBlo cells of either strain. The CD4+, CD45RBhi subset also proliferated more vigorously in response to polyclonal activation than the CD4+ CD45RBlo fraction. To examine whether the patterns of CD45RB expression on HGG-primed cells from C57BL/6J mice were common to other antigens, the response profiles were examined after in vivo priming with a second antigen, KLH. In contrast to studies with HGG as the Ag, the proliferative response to KLH in C57BL/6J mice was evenly divided among the CD45RBhi and CD45RBlo fractions on Day 8 after priming, but shifted markedly to the CD45RBlo fraction by Day 12 after priming. Taken together, these data show that the patterns of CD45RB expression on primed populations of CD4+ cells can exhibit mouse strain polymorphism and can differ depending on the choice of antigen for immunization.

The effect of lipopolysaccharide desensitization on the regulation of in vivo induction of immunologic tolerance and antibody production and in vitro release of IL-1.

As previously reported, LPS and 8-derivatized guanosine (both generators of IL-1 release), as well as IL-1 itself interfere with the in vivo induction of tolerance to DHGG in A/J mice. In the present studies it was demonstrated that desensitization of either A/J or CBA/CaJ mice with LPS aborts the ability of LPS to interfere with the induction of tolerance to DHGG. The abrogation of the ability of LPS to interfere with tolerance by LPS desensitization is not the result of neutralization of LPS by antibody produced to LPS during desensitization. Desensitization with LPS also aborts the interference with tolerance induction by 7-methyl-8-oxoguanosine. LPS desensitization inhibits the ability of LPS and/or 7-methyl-8-oxoguanosine to both convert a tolerogenic signal to an immunogenic signal and interfere with the induction of a tolerant state to a subsequent injection of Ag. The effects resulting from desensitization may be in part attributed to the depletion of IL-1. LPS desensitization also modulates the antibody response to injection of the AG, AHGG. Desensitization with LPS markedly suppresses the antibody response to a subsequent injection of AHGG in CBA/CaJ mice. Desensitization with LPS also inhibits the anti-HGG antibody response in A/J mice, but in this strain its effect is dependent on the route of injection of AHGG. In an experiment directly comparing the responses of normal and desensitized A/J mice to either intravenous or intraperitoneal injection of AHGG, desensitization only suppressed the response in mice injected with AHGG i.p.. Desensitization with LPS also inhibits the ability of LPS to act as an adjuvant in a subsequent antibody response to AHGG. Not only does desensitization interfere with the primary antibody response to AHGG, but it also interferes with the secondary response, suggesting that the primary injection after desensitization induces a state of immunologic tolerance.

The effect of aging on the induction of experimental autoimmune thyroiditis.

The effect of age on the ability to elicit the various immune functions comprising experimental autoimmune thyroiditis in mice has been examined. Compared with young mice (2 to 3 mo), CBA/CaJ and A/J aged mice (20 to 30 mo) show a drastic reduction in their ability to develop circulating antibody after injection of mouse thyroglobulin (MTg) or mouse thyroid extract (MTE) in complete Freund's adjuvant (CFA). Delayed-type hypersensitivity responses were also depressed, as well as the ability of aged lymph node cells to proliferate in vitro to antigen and the ability of aged splenic T cells to function as helper cells for in vitro antibody production. However, after injection of these thyroid antigens in CFA, aged mice developed thyroid lesions either comparable to or only slightly less intense than those observed in young mice. The disparity between the levels of immune responses and thyroid lesions observed in aged mice can be explained by the greater susceptibility of aged thyroids to tissue damage, since transfer of identical numbers of Con A-activated MTE-primed young splenocytes to young and aged recipients results in a more severe thyroiditis in the aged recipients. Priming mice to MTE in CFA at 9 mo of age, at which time mice are responsive to MTE, did not enhance either T or B cell responsiveness to injection of MTE in CFA at 24 mo of age. Lymphocytes from MTE-injected aged mice also failed to transfer thyroiditis to young recipients after in vitro activation of the lymphocytes with Con A.

Transfer of experimental autoimmune thyroiditis with T cell clones.

We have investigated three T lymphocyte clones isolated from CBA/CaJ mice primed with mouse thyroid extract (MTE) in adjuvant. All three clones are L3T4+, Ig-, and Lyt2- and proliferate to MTE, mouse thyroglobulin (MTG) and rat thyroid extract. Clones A7 and B7 transfer thyroiditis to irradiated (475 rad) syngeneic mice, but not to normal recipients. The thyroid lesion induced by the B7 clone is characterized by the infiltration of both mononuclear and polymorphonuclear cells. The thyroiditis is transient in that lesions are apparent 7 and 14 days after transfer, but thyroids return to normal by day 21. Clone B7 showed helper activity for trinitrophenyl-keyhole limpet hemocyanin-primed B cells in vitro when stimulated with trinitrophenyl-MTG and also stimulated the production of anti-MTG antibody in recipient mice. Clone A7 induced thyroid lesions characterized by infiltration of the thyroid with mononuclear cells, with virtually no polymorphonuclear cell infiltration. This clone has shown no helper activity following stimulation with trinitrophenyl-MTG. The third clone (D2) proliferates to and shows helper activity to MTG, but fails to transfer thyroiditis to syngeneic, irradiated mice. On continuous culture, clone B7 lost its surface Thy. The loss of Thy appears unrelated to the ability to transfer thyroiditis since subclones of B7 with markedly different percentages of Thy+ cells transferred disease equally well.

T cell competence to heterologous and homologous thyroglobulins during the induction of experimental autoimmune thyroiditis.

The ability of T cells to respond to homologous vs. heterologous thyroglobulins (Tg) has been evaluated using different models for the induction of experimental autoimmune thyroiditis. Substantial levels of T cell activation could be demonstrated to heterologous Tg following immunization with heterologous Tg in complete Freund's adjuvant, whereas only minimal levels of T cell activation to homologous Tg could be obtained following immunization with homologous Tg in complete Freund's adjuvant. Using this immunization protocol, heterologous and homologous Tg induced equivalent levels of serum antibody to the immunizing Tg. However, when injected in incomplete Freund's adjuvant, homologous Tg induced less antibody than heterologous Tg. Even greater differences in serum antibody levels to heterologous and homologous Tg, were apparent following immunization with soluble Tg. These thyroiditis differences are attributed to the presence of only a minimal level of T cell competence for homologous Tg, which is capable of inducing experimental autoimmune thyroiditis with stringent immunization protocols, but not with weaker immunization regimens.

A radioimmunoassay for murine thyroglobulin in serum: age-related increase of serum thyroglobulin levels in AKR/J mice.

Possible regulation of the synthesis of murine thyroglobulin (mTg) and a possible role for the major histocompatibility complex in this process were evaluated in 18 inbred strains of mice. A double antibody RIA was first developed to measure mTg in serum. The isolated mTG was labeled with 131I since 125I rapidly degraded the mTg molecule. The sensitivity of this mTg assay allowed detection of 15-20 ng/ml. Specificity was demonstrated by the minimal cross-reactivity with rat Tg (0.008%) and the total lack of cross-reactivity with human Tg. There was no correlation between H-2 haplotype and serum mTg levels. In five strains of mice with the H-2k haplotype, mTg levels varied from 30 +/- 2 to 48 +/- 11 ng/ml (mean +/- SD); however, only aged AKR/J mice (H-2k) exceeded this range (196 +/- 27 ng/ml). Strains with other haplotypes (a, b, d, g, q, v) demonstrated a similar range of mTg levels, but none had this age-related increase in mTg levels. The high levels of mTg were not caused by a decrease in the half-life of this protein and probably not caused by virus-induced alterations in the thyroid economy. Thyroids from the AKR/J mice, however, had larger follicles and flatter epithelia compared to thyroids from other strains. These studies suggest that AKR/J mice may represent a useful animal model for the study of goitrogenesis.

T and B cell reactivity in experimental autoimmune thyroiditis.

Using several immunization protocols, T cells specific for self Tg were shown to be activated to only a limited degree compared with the ability to activate T cells to a foreign Tg. Thus, a high degree of tolerance exists in T cells to self Tg, while tolerance is not maintained in B cells. Whether the small degree of T cell activation that is induced under stringent immunization procedures plays a role in the induction of the spontaneous disease is unknown. Preliminary results are also presented showing that experimental autoimmune thyroiditis (EAT) can be more readily induced in aged than in young adult mice and these results are discussed in terms of the generalized decrease in T cell reactivity in aged mice.

Role of C3 in the regulation of a splenic PFC response in rabbits.

The effects of in vivo C3 depletion on the immune response were examined in rabbits by assaying for splenic PFC after immunizing normal or cobra venom factor-treated animals with aggregated human gamma-globulin. The response to this T-dependent antigen has previously been shown to be regulated such that several cycles of PFC appear following a single intravenous injection of antigen. C3 depletion had no effect on the first peak of PFC (appearing 5 days after injection), but resulted in depression of the second peak of PFC (day 13). In rabbits depleted of C3, antigen localization in splenic germinal centers was markedly decreased. Delaying C3 depletion until after antigen localization had occurred resulted in no depression of the second peak of PFC. These results suggest that one mechanism by which C3 affects immune responses in vivo is via its role in influencing the persistence of antigen. In the absence of C3, no significant localization of antigen occurs, resulting in interference with the cyclical production of antibody.

Modulation of regulatory mechanisms operative in the cyclical production of antibody.

Modulation of the cyclical response in rabbits to aggregated human gamma globulin (AHuIgG) was investigated in order to study some of the parameters involved in self-regulation of the immune response. Several mitogens (lipopolysaccharide [LPS], phytohemagglutinin [PHA], and concanavalin A [Con A]), when injected simultaneously with antigen, have been shown to modulate the normal splenic plaque-forming cell (PFC) response in rabbits to a single intravenous injection of AHuIgG. This response to AHuIgG has previously been characterized by the initial appearance of PFC in the spleen 3 days later, with a peak of PFC at 5 days after injection. The number of PFC in the spleen then decreases and remains at a low level until a second increase begins on day 10, peaking on day 13. The 8-day cycle between peak PFC repeats, with a third peak appearing on day 21. In the present studies, injection of LPS with AHuIgG was shown to affect the PFC response by enhancing only the initial peak of PFC, PHA was shown to enhance both the initial and secondary peaks of PFC, while injection of Con A with AHuIgG resulted in a prolonged increase in PFC with no apparent cycling. Irradiation 24 h after injection of antigen resulted in PFC kinetics similar to those observed with PHA, although the increase in PFC was more marked with irradiation. Thus, although LPS, PHA, Con A, and irradiation markedly affected the immune response to AHuIgG, Con A was the only substance which altered the cyclical appearance of PFC to HuIgG. The cyclical nature of the PFC kinetics was shown to occur with either intravenous or intraperitoneal injection of antigen and in both primary and secondary responses, provided that the rabbits were primed with a low dose of antigen. Data were obtained that suggest that the response in distal lymph nodes may be regulated by immunological events occurring in the spleen. Cycling of PFC was not observed in the draining node after subcutaneous injection of AHuIgG in the hind foot. However, if the antigen was also injected intravenously at the same time as the subcutaneous injection, the response in the node became cyclical.

Humoral and cell-mediated immunity in experimental progressive thyroiditis in rabbits.

Rabbits immunized over a long period of time with serial injections of aqueous preparations of either bovine thyroglobulin or chemically altered rabbit thyroglobulin develop progressive thyroiditis. As is short-term thyroiditis in rabbits and mice, this thyroiditis is characterized by lesions and cellular infiltration similar to that observed in Arthus reactions. Once the progressive thyroiditis is established, the rabbits respond readily to subsequent injections of native rabbit thyroglobulin. No significant reduction of lesions or circulating antibody is observed when injections of native rabbit thyroglobulin are substituted for the preparations used to induce the disease. Cell-mediated hypersensitivity to rabbit thyroglobulin, as evidenced by MIF activity, develops in rabbits after prolonged immunization with altered or cross-reacting thyroglobulin. It is suggested that this activity develops as a result of a loss in the unresponsive state in T lymphocytes. The data indicate that it is the persistence of circulating antibody to autologous thyroglobulin which sequesters autologous thyroglobulin from peripheral lymphoid tissue, and thus, results in the loss of the unresponsive state in lymphocytes of these tissues. It is suggested that similar events may be involved in the development of cell-mediated hypersensitivity in thyroiditis in humans.

The enhancing effect of mitogens on the in vivo immune response of rabbits.

The effect of Con A and PHA on the primary humoral immune response in rabbits was investigated and compared with the effect of LPS and irradiation. Both Con A and PHA enhanced PFC levels and precipitating antibody titers to HuIgG when injected simultaneously with AHuIgG. Comparable enhancement of the response to HuIgG was obtained by simultaneous injection of LPS with AHuIgG or by irradiating rabbits (500 R) 24 hr after injection of the antigen. Further studies with Con A showed that its adjuvant effect was not restricted to a particular antigen since it also enhanced antibody responses to both SRBC and Bov Tg when injected simultaneously with either of these antigens. Although optimal enhancement of the response to HuIgG occurred when antigen and mitogen were injected simultaneously, some enhancement was obtained if Con A was injected 2 days after antigen, whereas the response was decreased if Con A was injected 2 days before antigen. Splenic PFC in A/J mice injected with doses of AHuIgG and Con A equivalent on a body weight basis to those used in rabbits were also enhanced by simultaneous injection of Con A with the antigen.

A cyclical appearance of antibody-producing cells after a single injection of serum protein antigen.

After a single intravenous injection of rabbits with aggregated HuIgG, IgM- and IgG-plaque-forming cells (PFC) in both the spleens and peripheral blood of rabbits peaked 5, 13, and 21 days after injection, while almost no PFC could be detected on days 8 and 16. The available data suggest that the secondary peaks of PFC (days 13 and 21) resulted from stimulation of memory cells by persisting antigen that was localized in the germinal centers in the spleen. No such persistence of antigen occurred in the lymph nodes, and these lymphoid tissues did not exhibit secondary peaks of PFC. The identical kinetic patterns for IgM- and IgG-PFC indicate that the major portion of IgG-PFC did not result from IgM-secreting cells switching to IgG synthesis and secretion. The present data suggest that the antibody produced and present at the site of interaction between committed cells and antigen is responsible for the regulation of antibody synthesis to persisting antigens. Possible cellular events involved in both the regulation and an apparent synchronous appearance of antibody producing cells in the spleens of rabbits were presented.