The role of H-2 linked genes in helper T-cell function. IV. Importance of T-cell genotype and host environment in I-region and Ir gene expression.

We have studied the properties of helper T cells specific for sheep erythrocytes (SRBC), keyhole limpet hemocyanin (KLH), or poly-L-(Tyr,Glu)-poly-DL-Ala-poly-L-Lys [(T,G)-A--L]. These T cells differentiated and were primed in vivo in irradiation chimeras constructed of various combinations of F1 and parental bone marrow donors and irradiated recipients. Primed T cells were then tested for helper activity in the in vitro response of B cells and macrophages (Mphi) of parental or F1 origin to the hapten trinitrophenol coupled to the priming antigen. When testing either SRBC or KLH-specific T cells of parental H-2 type which had differentiated in F1 hosts, we found that they cooperated equally well with B cells and Mphi of either parental H-2 type. On the other hand, when testing F1 T cells which had differentiated in parental hosts, we found that they cooperated well only with B cells and Mphi which had the K-IA region type of the parental host. In similar experiments we found that (T,G)-A--L-specific T cells of low responder H-2 type which had differentiated in (high responder X low responder) F1 hosts induced high responses in high responder B cells and Mphi (T,G)-A--L-specific F1 T cells which differentiated in high responder but not those which differentiated in low responder hosts induced high responses in high responder B cells and Mphi. Low responder B cells and Mphi yielded low responses in all cases regardless of the source of (T,G)-A--L-specific T cells with what they were tested. Our results support the conclusion that I-region and Ir genes function via their expression in B cells and Mphi and in the host environment during helper T-cell differentiation, but not, at least under the conditions of these experiments, via their expression in the helper T cell itself. These findings place constraints upon models which attempt to explain the apparent dual recognition of antigen and I-region gene products by helper T cells.

The role of H-2-linked genes in helper T-cell function. III. Expression of immune response genes for trinitrophenyl conjugates of poly-L(Tyr, Glu)-poly-D,L-Ala--poly-L-Lys in B cells and macrophages.

Using lymph node T cells from poly-L(Tyr,Glu)-poly-D,L-Ala--poly-L-Lys[(TG)-A--L]-primed animals and B cells from animals primed with trinitrophenylated (TNP) protein or lipopolysaccharide, we have obtained anti-TNP-(TG)-A--L direct plaque-forming responses in vitro. Response to this antigen was shown to be controlled by the H-2 haplotype of the animal studied. The strain distribution of in vitro response was very similar to that previously reported by others for in vivo secondary IgG responses to (TG)-A--L. We investigated the cell types expressing the Ir gene(s) for (TG)-A--L in our cultures. F1, high responder x low responder mice were primed with (TG)-A--L. Their T cells were active in stimulating anti-TNP-(TG)-A--L responses of high responder but not low responder B cells and macrophages (MPHI), even though both preparations of B cells and Mphi were obtained from mice congenic at H-2 with one of the parents of the F1. For three low responder strains tested, of the H-2h2, H-2k, and H-2f haplotypes, the anti-TNP-(TG)-A--L response of low responder B cells and Mphis in the presence of high responder, F1 T cells could not be improved by the addition of high responder, antigen-bearing Mphis to the cultures. In one strain of the H-2a haplotype, it was shown that neither the B cells nor Mphis could be functional in anti-TNP-(TG)-A--L responses. Our results therefore suggested the Ir genes for anti-TNP-(TG)-A--L responses were expressed at least in B cells in all the low responder strains we studied, and, in mice of the H-2a haplotype, in Mphis too.

The role of H-2-linked genes in helper T-cell function. VI. Expression of Ir genes by helper T cells.

We examined the expression of (TG)-A--L specific Ir genes in helper T cells using T cells from low responder leads to (B10, high responder x low responder) F1 chimeric mice. In this paper, the low responder strain studied was B10.M, H-2f. B10.M T cells from these chimeric animals do not help anti-TNP-(TG)-A--L responses, even though they have matured in a high responder thymus and been primed and challenged with antigen on high responder Mphi and B cells. These findings indicate that in the H-2f haplotype an Ir-gene controlling anti-(TG)-A--L activity is expressed in helper T cells. The findings are in contrast to those we have obtained and previously reported with T cells of another low responder haplotype, H-2a. Taken together with our previous findings that (TG)-A--L specific Ir genes are expressed by B cells and Mphi of both the H-2a and H-2f haplotypes, the results indicate two sites of action for Ir genes, and suggest two different gene products acting at different stages of the response, both of which are defective in H-2f cells, and only one of which is defective in H-2a cells.

The role of H-2-linked genes in helper T cell function. VII. Expression of I region and immune response genes by B cells in bystander help assays.

The mode of action by bystander helper T cells was investigated by priming (responder X nonresponder) (B6A)F1 T cells with poly-L-(Tyr, Glu)-poly-D,L-Ala--poly-L-Lys [(TG)-A--L] and titrating the ability of these cells to stimulate an anti-sheep red blood cell (SRBC) response of parental B cells and macrophages in the presence of (TG)-A--L. Under limiting T cell conditions, and in the presence of (TG)-A--L, (TG)-A--L-responsive T cells were able to drive anti-SRBC responses of high-responder C57BL/10.SgSn (B10) B cells and macrophages (M0), but not of low-responder (B10.A) B cells and M0. Surprisingly, the (TG)-A--L-driven anti-SRBC response of B10.A B cells was not restored by addition of high-responder acessory cells, in the form of (B6A)F1 peritoneal or irradiated T cell-depleted spleen cells, or in the form of B10 nonirradiated T cell-depleted spleen cells. These results suggested that (TG)-A--L-specific Ir genes expressed by B cells controlled the ability of these cells to be induced to respond to SRBC by (TG)-A--L-responding T cells, implying that direct contact between the SRBC-binding B cell precursor and the (TG)-A--L-responsive helper T cells was required. Analogous results were obtained for keyhold limpet hemocyanin (KLH)-driven bystander help using KLH-primed F1 T cells restricted to interact with cells on only one of the parental haplotypes by maturing them in parental bone marrow chimeras. It was hypothesized that bystander help was mediated by nonspecific uptake of antigen [(TG)-A--L or KLH] by SRBC-specific b cells and subsequent display of the antigen on the B cell surface in association with Ir of I-region gene products, in a fashion similar to the M0, where it was then recognized by helper T cells. Such an explanation was supported by the observation that high concentrations of antigen were required to elicit bystander help. This hypothesis raises the possibility of B cell processing of antigen bound to its immunoglobulin receptor and subsequent presentation of antigen to helper T cells.

The role of H-2-linked genes in helper T-cell function. I. In vitro expression in B cells of immune response genes controlling helper T-cell activity.

The ability of murine helper T cells primed to the antigen, sheep erythrocytes (SRBC) to cross-react with burro erythrocytes (BRBC) in the in vitro anti-trinitrophenol (TNP) response to TNP-RBC was shown to be under genetic control. Although non-H-2 genes were shown to influence the absolute level of helper activity assayed after SRBC priming, the extent of cross-reaction of SRBC-primed helpers with BRBC was shown to be controlled by an H-2-1inked Ir gene(s). H-2 haplotypes were identified which determined high, intermediate, or low response to the cross- reacting determinants and the gene(s) controlling the cross-reaction tentatively mapped to the K through I-E end of the H-2 complex. Helpers primed in F(1) mice of high x intermediate or high x low responder parents were tested for cross-reaction using B cells and macrophages (Mphi) of parental haplotypes. In each case the extent of cross-reaction was predicted by the H-2 haplotype of the B cells and Mphi, establishing the expression of the Ir gene(s) in B cells and/or Mphi a t least, but not ruling out its expression in T cells as well. The low cross-reaction seen when T cells from F(1) mice of high x low responder parents were tested on low responder B cells and Mphi was not increased by the presence of high responder Mphi, indicating the Ir gene(s) is expressed in the B cell a t least although it may be expressed in Mphi as well. These and our previously reported experiments are consistent with the hypothesis that helper T cells recognize antigen bound to the surface of B cells and Mphi in association with the product(s) of Ir gene(s) expressed on the B cell and Mphi.

Regulation of the immune response. 3. Kinetic differences between thymus- and bone marrow- derived lymphocytes in the proliferative response to heterologous erythrocytes.

The kinetics of the in vivo response to SRBC was studied in mouse spleen at both the B cell and T cell levels. The B cell response was assayed by following the appearance of antibody-secreting cells in the spleen using the hemolytic plaque assay. The T cell response was monitored by following the increase in or "priming" of helper activity in the spleen using a quantitative in vitro assay. The role of cellular proliferation in both responses was established with the inhibitor of mitosis, vinblastine. The results show that, although the development of T cell activity precedes that of anti-SRBC PFC by as much as 1 day, T cells lag at least 1 day behind B cells in the onset of cellular proliferation. The evidence suggests either that the helper T cell which proliferates in response to SRBC does so after helping in the initiation of the primary B cell response or that the proliferative T cell response and the initiation of the primary B cell response involve two different subpopulations of T cells.

Regulation of the immune response. II. Qualitative and quantitative differences between thymus- and bone marrow-derived lymphocytes in the recognition of antigen.

The specificity of antigen recognition by thymus-derived helper cells (T cells) and antibody was examined in mice, heterologous erythrocyte antigens from sheep (SRBC), goat (GRBC), burro (BRBC), chicken (CRBC), and toad (TRBC) being used. Antibody specificity was tested by a number of functional assays: hemagglutination, hemolysis, and immune suppression. The specificity of T cells was determined by titrating their ability to help the in vitro antitrinitrophenol (TNP) responses of mouse spleen cultures immunized with the hapten coupled to the various test erythrocytes as carrier. Anti-SRBC antibody cross-reacted with GRBC, but not with BRBC, CRBC, or TRBC. In contrast, SRBC-primed helper T cells cross-reacted with both GRBC and BRBC, but not with CRBC or TRBC, indicating a difference in the specificity of antigen recognition between the cellular and the humoral immune responses.

Analysis of V beta 17a expression in new mouse strains bearing the V beta a haplotype.

A set of new mouse strains were produced that carry the V beta a haplotype of the TCR-alpha/beta and any of a number of different H-2 haplotypes on backgrounds derived from related C57BL, C57L, and C57BR mice. Study of V beta 17a expression in these mice confirms the association between the presence of IE and the deletion of V beta 17a+ T cells. A second H-2 gene causing deletion of V beta 17a+ T cells was mapped in these mice to the K end of H-2k, and H-2 influences on the level of selection of CD4+ V beta 17a+ T cells were indicated.

Transcytosis of staphylococcal superantigen toxins.

Staphylococcus aureus produces a set of proteins (e.g., staphylococcal enterotoxin A [SEA], SEB, toxic shock syndrome toxin 1 [TSST-1]) which act both as superantigens (SAgs) and toxins. Although their mode of action as SAgs is well understood, little is known about how they enter the body via the intestine and cause food poisoning. To examine this problem we used an in vitro culture system to study the capacity of class II MHC-negative human intestinal epithelial cells (Caco-2) to transcytose several staphylococcal toxins. We found that Caco-2 cells are capable of dose-dependent, facilitated transcytosis of SEB and TSST-1, but not SEA. We extended these studies in vivo in mice by showing that ingested SEB appears in the blood more efficiently than SEA. Our data suggest that these toxins can cross the epithelium in an immunologically intact form. These results may have important implications for the pathogenesis of food poisoning.

B cell helper factors. I. Requirement for both interleukin 2 and another 40,000 mol wt factor.

A helper factor(s) distinct from interleukin 2 (IL-2) was shown to be present in the concanavalin A-stimulated supernatant of normal mouse spleen cells (normal Con A Sn). Spleen cells thoroughly depleted of T cells required both IL-2 and this factor to produce antibody-secreting cells in response to sheep erythrocytes, although in the presence of IL-2 and a few T cells the requirement for the factor was less apparent. The factor had an apparent approximately 40,000 mol wt. The factor was found in normal Con A Sn that had been depleted of IL-2 by absorption with IL-2-dependent T cells and was absent from Con A-stimulated supernatants of the IL-2-producing T cell hybridoma, FS6-14.13. These results indicate that multiple helper factors control the B cell response to antigen and that IL-2, in addition to its T cell growth promoting activity, plays a direct role in B cell responses.

Thymocytes can become mature T cells without passing through the CD4+ CD8+, double-positive stage.

T cells bearing the class II-restricted, DO-T cell receptor (TCR) are CD4+ if their thymocyte precursors are positively selected on the class II protein, IAd, but they are almost all CD4- after positive selection on a class II for which they have higher avidity, IAb. DO-TCR+ T cells mature in H-2b mice lacking CD4. CD4- DO-TCR+ T cells appear in H-2b mice at the same rate as their CD4+ counterparts appear in H-2d animals, suggesting that the CD4- cells are not the product of some minor pathway of thymocyte development and selection. In H-2b CD4 knock out mice expressing human CD2 under the control of the mouse CD4 promoter, mature DO-TCR+ cells did not express human CD2. These results suggest that the CD4-CD8-, DO-TCR+ mature T cells have developed without ever passing through the equivalent of a CD4+,CD8+ stage. The early expression of alpha/beta receptors (TCRs) on thymocytes in TCR transgenic mice may allow maturation of this type. Passage through the equivalent of the CD4+ CD8+, double-positive stage is not essential for differentiation of thymocytes into mature T cells.

Functional analysis of T cells expressing Ia antigens. I. Demonstration of helper T-cell heterogeneity.

We have examined the expression of I-region antigens on functional subpopulations of murine T cells. A.TH anti-A.TL (anti-Ik, Sk, Gk) alloantiserum was raised by immunization of recipients with concanavalin A (Con A) stimulated thymic and peripheral T-cell blasts. In contrast to similar antisera made by conventional methods, the anti-Ia blast serum was highly cytotoxic for purified T lymphocytes. Moreover, it reacted in a specific fashion with T cells having particular functions. Treatment of keyhole limpet hemocyanin (KLH)-primed B10.A (H-2 alpha) T cells with this antiserum plus complement resulted in the elimination of helper activity for B-cell responses to trinitrophenyl-KLH. Inhibition was shown to be a result of the selective killing of one type of helper T cell whose activity could be replaced by a factor(s) found in the supernate of Con A-activated spleen cells. A second type of helper cell required for responses to protein-bound antigens appeared to be Ia-. By absorption and analysis on H-2 recombinants, at least two specificities were detectable on helper T cells; one mapping in the I-A subregion and a second in a region(s) to the right of I-J. In addition, the helper T cell(s) involved in the generation of alloreactive cytotoxic lymphocytes was shown to be Ia+, whereas cytotoxic effector cells and their precursors were Ia- with this antiserum. These results provide strong evidence for the selective expression of I-region determinants on T-cell subsets and suggest that T-cell-associated Ia antigens may play an important role in T-lymphocyte function.

Helper signals in the plaque-forming cell response to protein-bound haptens.

We have demonstrated the ability of a series of murine T cell hybridomas to deliver an antigen-specific, B cell I-region-restricted helper signal in the generation of specific PFC responses to protein-bound haptens. With some hybridomas the elicitation of optimal PFC responses required the addition of nonspecific factors provided by culture supernatants of concanavalin A-stimulated (Con A SN) spleen cells. Using hapten-primed B cells depleted of both T cells and macrophages (Mphi) we have now demonstrated a requirement for three nonspecific factor preparations to substitute for spleen Con A SN in the elicitation of optimal PFC responses. The first preparation was the interleukin 1 containing culture supernatant of the Mphi tumor cell line P388D1, the second the interleukin 2 (IL-2) and B cell growth factor containing Con A SN of the T cell hybridoma FS6-14.13, and the third, the gamma interferon containing Con A SN of the T cell hybridoma FS7-20.6.18. The P388D1 and FS6-14.13 factor preparations were most effective when added at the initiation of culture, while the FS7-20.6.18 factor preparation was most effective when added at 24 h of culture. The activity of FS6-14.13 Con A SN was depleted by incubation with the IL-2-dependent T cell line HT-2. The activity of FS7-20.6.18 Con A SN was abrogated by incubation at pH 2. The results suggest that the generation of PFC responses to protein-bound haptens require at least three nonspecific factors in addition to an antigen/Ia specific helper signal.

Structural analysis of a mouse mammary tumor virus superantigen.

It has recently been shown that the minor lymphocyte stimulating-like products expressed by some mice are actually encoded by open reading frames in the 3' long terminal repeats of mouse mammary tumor viruses. These products act as viral superantigens (vSAGs). That is, they stimulate most T cells bearing particular V beta s almost regardless of the rest of the variable components of the T cell receptors expressed by those cells. To find out more about the structure of these vSAGs, a set of truncated vSAG genes was used in transfection and in vitro translation experiments to show that the functional vSAG is a type II integral membrane protein with a large glycosylated extracellular COOH-terminal domain and a small, nonessential, intracellular NH2-terminal cytoplasmic domain. These results are consistent with the fact that the vSAGs must be expressed on the cell surface in order to interact with T cells and class II major histocompatibility complex proteins. They also account for the finding that much of the V beta specificity of the vSAGs is controlled by amino acids at the COOH-terminal end of the vSAG proteins, amino acids that will be extracellular in type II proteins.

Antigen-specific, H-2-restricted helper T cell hybridomas.

We have examined the carrier-specific helper activity of a number of antigen-specific, I region-restricted T cell hybridomas prepared in our laboratory. The hybridomas were assayed for helper activity in the presence or absence of exogenously added nonspecific factors found in the concanavalin A-activated supernatants of normal mouse spleen cells. Of six hybridomas tested, all six could stimulate the IgM anti-hapten response of hapten-primed B cells in the presence of the appropriate hapten-carrier conjugates. At low or moderate carrier doses, the response was dependent upon hapten-carrier linkage and the ability of the hybridoma cells to interact with carrier in association with H-2 products of the responding B cells themselves. Plaque-forming cell responses stimulated by some of the hybridomas were absolutely dependent upon the addition of nonspecific factors, suggesting that anti hapten-protein responses require both an antigen specific I region restricted signal from the T cell hybridomas and nonspecific helper factors, made either by the T cell hybridomas or added exogenously. Under two sets of circumstances, B cells were stimulated in the absence of a simultaneous signal delivered through their immunoglobulin receptor. This occurred either when hapten-primed B cells were stimulated with an ovalbumin/I-Ak-specific hybridoma in the presence of very high concentrations of ovalbumin, or when H-2b B cells were incubated with a hybridoma specific for I-Ab alone. This was interpreted to mean that B cells can be stimulated by reaction of T cells with surface I molecules.

Regulation of the immune response. I. Differential effect of passively administered antibody on the thymus-derived and bone marrow-derived lymphocytes.

The effect of passively transfered antiserum against sheep erythrocytes (SRBC) on the antigen stimulated increase of SRBC-specific plaque-forming cells (anti-SRBC-PFC) and SRBC-specific thymus-derived lymphocytes (SRBC-specific T-cells) in the mouse spleen was examined. A dose of antiserum which severely suppressed the development of anti-SRBC-PFC did not prevent the increase in SRBC-specific T-cells, as measured by their ability to cooperate in the in vitro response to trinitrophenylated (TNP) SRBC. It was shown that the insensitivity of these T-cells to antiserum could not be explained by their low antigen requirement as compared to that of PFC. In the in vivo response of mice to TNP-SRBC, antibody specific for TNP suppressed the appearance of both anti-TNP- and anti-SRBC-PFC. The presence of free SRBC specifically prevented the suppression of the anti-SRBC-PFC. These observations are consistent with opsonization by phagocytic cells as the primary means of the observed suppression of PFC development by antibody.

Analysis of the response of B cells from CBA/N-defective mice to nonspecific T cell help.

We have investigated the induction of antibody responses to erythrocyte (RBC)-bound antigens in the (CBA/N x B10)F1 mouse. Male B cells, which express the CBA/N defect, were shown to be unresponsive to RBC antigens when the delivered T cell helper activity was solely nonspecific. Thus we demonstrated that defective B cells did not respond to concanavalin A supernatants or bystander helper activity, in spite of the fact that CBA/N-defective mice could produce these T cell activities. The defective B cell did not respond to RBC-bound antigen in the presence of RBC-primed T cells, although the magnitude of this response was usually twofold less than normal controls. The insensitivity of CBA/N defective B cells to nonspecific T cell helper activities seemed to involve at least the inability of RBC antigens to activate defective B cells in the absence of antigen-specific T cell help.

Antigen-inducible, H-2-restricted, interleukin-2-producing T cell hybridomas. Lack of independent antigen and H-2 recognition.

We developed a method for production of antigen-specific, H-2-restricted T cell hybrids. The tumor cell partner in the fusions was itself a T cell hybrid, FS6-14.13.AG2 (or its derivatives), which could be induced to produce the growth factor, interleukin-2 (IL-2), in response to a challenge with concanavalin A, but had no known antigen specificity. The normal T cell partner in the fusions was a population of lymph node T cell blasts that had been highly enriched in antigen-specific, H-2-restricted T cells by in vivo immunization, followed by in vitro challenge with antigen and clonal expansion in IL-2-containing medium. These fusions produced hybrids that grew constitutively in culture. A sizable proportion of the hybrids demonstrated the ability to produce IL-2 in response to a challenge with specific antigen presented by irradiated spleen cells of the appropriate H-2 type. Four cloned antigen/H-2-specific hybrid lines were produced. AO-40.10 responded to chicken ovalbumin (OVA) when presented by I-A(k)-bearing cells. DC1.18.3 responded to the apo form of beef cytochrome c when presented with I-A(d). AODK-10.4 responded to keyhole limpet hemocyanin (KLH) presented with I-A (d). AODK-1.16 also responded to KLH presented by a product of the I region of H-2(d), but the data were consistent with either a product of the I-J-I-E(d) region or a combinatorial molecule with elements from both I-A(d) and I-E(d)/I-C(d). Coincidentally, AO-40.10 was shown to have an unexpected alloreactivity with a product of H-2(b) mapping to the K-I-A region. These hybrids should prove invaluable as sources of monoclonal material for the study of the receptor(s) on T cells with H-2-restricted antigen specificities. We also generated T cell hybrids with two antigen/H-2 specificities by fusing an azaguanine-resistant clone of AO-40.10 to normal T cells with a different antigen/H-2 specificity. Many of the hybrids retained reactivity to OVA plus H-2(a) and to the second antigen/H-2 combination. None reacted to either OVA plus the second H-2 type or to the second antigen plus H-2(a). One of these hybrids was successfully cloned to produce the line AOFK- 11.11.1. It retained the ability to recognize OVA plus I-A(k) inherited from one parent, and KLH plus IA(f) inherited from the other. It did not recognize OVA plus IA(f) or KLH plus I-A(k). These results have some bearing on models describing the nature of T cell receptors for antigen recognized in association with H-2 products. They do not support models in which antigen and H-2 are recognized separately by two independent T cell receptors.

Antigen recognition properties of mutant V beta 3+ T cell receptors are consistent with an immunoglobulin-like structure for the receptor.

We examined the effect of mutations in the V beta portion of a pigeon cytochrome c (cyto c)-specific V beta 3+/V alpha 11+ T cell receptor on its ability to recognize cyto c/IEk and various superantigens. The results were consistent with an immunoglobulin-like structure for the receptor V beta domain and with separate interaction sites on V beta for conventional antigen and superantigens. An amino acid predicted to lie in CDR1 was critical for cyto c/IEk but not superantigen recognition, while several amino acids predicted to lie in the hypervariable region 4 loop were critical for superantigen but not cyto c/IEk recognition.

Mutations defining functional regions of the superantigen staphylococcal enterotoxin B.

Staphylococcal enterotoxin B (SEB) is both a superantigen and toxin. As a superantigen, SEB can bind to major histocompatibility complex (MHC) class II molecules to form a ligand for alpha/beta T cell receptors bearing particular V beta elements. As a toxin, SEB causes rapid weight loss in mice sometimes leading to death. We show here that both of these functions map to the NH2-terminal portion of the toxin. Three regions were identified: one important in MHC class II binding, one in T cell recognition, and one in both functions. These results support the conclusion that the toxicity of SEB is related to massive T cell stimulation and release of cytokine mediators and show that the residues interacting with MHC and the T cell receptor are intertwined.