Specific binding of K- and I-region products of the H-2 complex to activated thymus-derived (T) cells belonging to different Ly subclasses.

Responder cells [C57BL/6J X A.TL)F1 lymph node cells depleted of bursa equivalent-derived (B) cells by filtration through nylon wool columns] were activated against incompatible K-region and I-region products together under conditions where these antigens are presented on separate stimulator cells. The resulting T blasts were stained with different concentrations of antisera directed against incompatible stimulator K-region or I-region products, or both. We obtained results that strongly suggest that in these cultures each activated responder blast stains with antiserum directed against either K-region or I-region products, but not both. Responder blasts from the same cultures were treated with antiserum and complement (C) directed against either Ly-1.2 or Ly-2.2 T-cell-specific surface antigens. Anti-Ly-1.2 serum and C specifically eliminates virtually all responder blasts staining with antiserum directed against stimulator I-region products; whereas anti-Ly-2.2 serum reduces to background levels the proportion of cells staining with antiserum against stimulator K-region products. The results obtained suggest that T cells binding stimulator K-region and I-region products, respectively, belong to two different subclasses distinguishable by their Ly phenotypes. Possible explanations for this association of T- cell subclass and specificity are discussed.

Rosette-forming ability of thymus-derived lymphocytes in humoral and cell-mediated immunity. II. Helper cell activity.

The technique of velocity sedimentation at unit gravity was used to determine the size and rosette-forming ability of helper cells in nonimmune and immune C57BL/6 mice. Helper T cells were assayed by the ability to cooperate with bone barrow (B) cells in the antibody response to sheep erythrocytes (SRBC) in vivo. 19S helper cells in nonimmune animals were nonrosette-forming small lymphocytes; after immunization with SRBC in complete Freund's adjuvant, 19S helper cells were nonrosette-forming medium lymphocytes. 7S helper cells in immune animals were small lymphocytes which did not form rosettes. No SRBC binding by helper cells was observed under the conditions of rosette formation used.

The receptor specificity of alloreactive T cells. Distinction between stimulator K, I, and D region products and degeneracy of third-party H-2 recognition by low-affinity T cells.

The specificity of binding of stimulator-derived H-2 antigens by mixed lymphocyte culture (MLC)-activated T blasts was investigated under conditions of antigen excess. We have shown that the detectable proportion of alloantigen-binding blasts from primary MLC is a function of antigen concentration, and can represent up to more than 90 percent of total blasts, when the antigen is presented in the appropriate form (on mitomycin-treated viable stimulator cells, or membrane vesicles prepared from lipopolysaccharide blasts), and at nonlimiting concentration. Thus stimulator alloantigen-binding directly parallels the proliferative response and is not restricted to a subpopulation of T blasts. However, the marked dependence of the binding on antigen concentration indicates that cells with a wide range of receptor affinities for the stimulating determinants are involved. In view of this possibility, the specificity of binding by these cells was studied. We have demonstrated that stimulator K, I, and D region products are bound by nonoverlapping subpopulations of blasts, the sum of which may represent 93 percent of total blasts. Thus, specific distinction by these cells between different H-2 region products is not affected by the putative heterogeneity in terms of receptor affinities. However, specificity with respect to unrelated H-2 haplotypes is strictly dependent on antigen concentration. A preferential binding of stimulator membrane vesicles occurs at limiting concentrations; whereas the majority of blasts bind stimulator and third- party vesicles equally well at high vesicle concentrations. The binding of both vesicle types is specific in that it can be inhibited with the relevant anti-H-2 sera. Furthermore, stimulator and third-party vesicles seem to compete for binding sites on the same cells, as shown by cold antigen inhibition. From these results, we propose that there is an imperfect distinction between stimulator and third-party H-2 antigens by the majority of primary MLC blasts. In contrast, highly selected long-term MLC blasts do not bind third-party H-2 antigens at any concentration, and seem to have high affinity for the stimulating antigens. We conclude that large numbers of clones with low-affinity (cross- reactive) receptors are generated in primary MLC, most of which become eliminated during long-term selection. This implies that the frequency of cells strictly specific for nonshared stimulating determinants must be minute.

Rosette-forming ability of thymus-derived lymphocytes in cell-mediated immunity. I. Delayed hypersensitivity and in vitro cytotoxicity.

Effector cells in delayed hypersensitivity and in vitro cytotoxicity were studied in lymph node cells from animals immunized with sheep erythrocytes (SRBC) in complete Freund's adjuvant. Delayed hypersensitivity response (DHR) was assayed by the increase in foot pad swelling after the intrafoot pad injection of immune cells plus antigen. Cell-mediated cytotoxicity against SRBC was assayed by a microcytotoxicity test with sheep fibroblasts as target cells. Effector cells were antigen specific, sensitive to anti-theta serum plus complement (C), and insensitive to anti-Ig serum plus C. A nonrosette-forming (non-RFC) small lymphocyte effector T cell and a rosette-forming medium lymphocyte effector T cell were isolated by velocity sedimentation. The small lymphocyte non-RFC required a longer time than the medium lymphocyte RFC effector cell to produce maximum activity. Buoyant density failed to distinguish medium lymphocyte effector cells in DHR and in vitro cytotoxicity.

Thymus-derived rosettes are not "helper" cells.

Rosettes against SRBC were made from normal spleen cells. Although T rosettes tend to dissociate, they could be stabilized with 0.05% sodium azide. A clear separation of nonrosettes, T rosettes, and B rosettes was obtained by subjecting the suspension of splenic rosettes to velocity sedimentation at unit gravity. Each fraction was injected with either normal bone marrow cells or normal thymus cells with antigen into 650-R-irradiated hosts. Direct plaque-forming cells (PFC) were assayed in the spleens 7 days later. Synergism with thymus cells occurred only in the B-rosette fraction; PFC precursors therefore sedimented as B rosettes. Synergism with bone marrow cells occurred only in the nonrosette small lymphocyte fraction; helper cells therefore did not bind detectable numbers of sheep red blood cells (SRBC). Thus T rosettes are not helper cells in the direct PFC response of bone marrow B cells to SRBC.

Specific binding of alloantigens to T cells activated in the mixed lymphocyte reaction.

Immunoglobulin (Ig) is present on a large fraction of T cells from unfractionated lymphocytes activated by in vitro stimulation with H-2-incompatible cells (mixed lymphocyte reaction [MLR]). Removal of bursa equivalent-derived (B) cells from the responder cell population before mixed culture, by filtration through nylon wool columns, reduces the percentage of Ig-bearing responder T blasts to background levels. Thus, Ig on the T blast is probably of B cell origin. A large fraction of T blasts activated against the stimulator cells. This staining occurs with "early" and hyperimmune alloantisera, including the 7S fraction of the latter. B-depleted responder cells were activated against a mixture of two different stimulator cells and the resulting T blasts stained with different concentrations of sera directed either against one or both stimulator cells. We obtained results which strongly suggest that most or all responder T blasts stain with only one antistimulator serum. When antisera directed against different segments of the H-2 complex of the stimulator cells were used, it seemed that most responder T cells only bound antibody directed against a single segment. We propose that T cells activated in MLR carry stimulator alloantigens on their surface, and that this is due to specific antigen binding, not requiring the presence of B-cell-derived antibody. These histocompatibility antigen-binding T blasts can be detected by appropriate antistimulator alloantibodies.

Classification of thymus-derived and marrow-derived lymphocytes by demonstration of their antigen-binding characteristics.

Antigen-binding cells of T and B origin can readily be determined by quantitating the number of sheep erythrocytes per rosette after glutaraldehyde fixation. The T(1) and T(2) populations have low antigen-binding properties and are very unstable without fixation. The B(1) and B(2) populations are stable and correlate with precursor and secretory cells. Fixation of rosettes permits a sensitive test for studying differentiation of T and B cells.

Src-dependent ezrin phosphorylation in adhesion-mediated signaling.

In addition to providing a regulated linkage between the membrane and the actin cytoskeleton, ezrin participates in signal transduction pathways. Here we describe that expression of the ezrin Y145F mutant delays epithelial cell spreading on fibronectin by inhibiting events leading to FAK activation. The defect in spreading was rescued by the overexpression of catalytically functional Src. We demonstrate that ezrin Y145 is phosphorylated in A431 cells stimulated with epidermal growth factor (EGF) and in v-Src-transformed cells. Moreover in cells devoid of Src, SYF-/- fibroblasts, ezrin Y145 phosphorylation could only be detected upon the introduction of an active form of Src. The phosphorylation of ezrin at Y145 required prior binding of the Src SH2 domain to ezrin. Our results further show that Src activity influences its binding to ezrin and a positive feedback mechanism for Src-mediated Y145 phosphorylation is implied. Interestingly, cells expressing ezrin Y145F did not proliferate when cultured in a 3D collagen gel. Collectively, our results demonstrate a key signaling input of Src-dependent ezrin phosphorylation in adhesion-mediated events in epithelial cells.

A novel activating function of c-Src and Stat3 on HGF transcription in mammary carcinoma cells.

In the normal breast, hepatocyte growth factor (HGF) is primarily expressed by stromal cells, and stimulates in a paracrine manner epithelial cells expressing the HGF receptor (Met). In invasive human breast carcinomas, HGF and Met are frequently overexpressed, possibly establishing an autocrine HGF/Met loop that promotes tumour cell invasion. However, the mechanisms leading to autocrine HGF expression in carcinoma cells are not known. We previously demonstrated a cooperative effect between c-Src and Stat3 in the activation of HGF transcription in mammary carcinoma cells. The present report defines a novel Stat3 consensus site at nt -95 in the HGF promoter that is highly conserved in human and mouse, and is required for c-Src and Stat3 to activate HGF transcription in breast epithelial cells. DNA-protein binding studies demonstrated high affinity binding of a Stat3-containing complex to the nt -95 site. Endogenous Stat3 binding to this region of the HGF promoter in carcinoma cells expressing HGF was demonstrated using a chromatin immunoprecipitation assay. In addition, coexpression of Stat3 and activated c-Src caused increased expression of endogenous HGF mRNA and protein and marked cell scattering in breast epithelial cells. Our results delineate a novel c-Src/Stat3-dependent mechanism that regulates HGF promoter activity, and is linked to transformation of mammary epithelial cells.

Failure of expression of class I major histocompatibility antigens to alter tumor immunogenicity of a spontaneous murine carcinoma.

We have shown previously that clonal immunogenic variants of murine mammary adenocarcinoma 10.1 can be isolated after treatment in vitro with the DNA-hypomethylating agent 5-azacytidine (5-aza). Such immunogenic variants frequently express elevated class I major histocompatibility complex antigens relative to the level of expression in the parent tumor and are rejected in syngeneic mice by a T-cell-dependent process. To ascertain whether elevated immunogenicity is a function of increased class I antigen expression, we isolated high class I antigen expressors from 5-aza-treated 10.1 cells by using the fluorescence-activated cell sorter. Clonal variants displaying any increase in class I antigen expression were more efficient stimulators of allo-class I antigen-specific cytolytic T-cell precursors. However, these variants displayed unaltered tumorigenicity in immunocompetent syngeneic mice. Thus, phenotypic changes other than, or in addition to, elevated class I antigen expression cause the reduced tumorigenicity of immunogenic clones of 10.1 cells isolated after 5-aza treatment.

Enhanced expression of class I major histocompatibility complex gene (Dk) products on immunogenic variants of a spontaneous murine carcinoma.

Both immunogenic and nonimmunogenic variant clones were isolated from a recently obtained spontaneous murine adenocarcinoma after treatment (xenogenization) with either the mutagen ethyl methanesulfonate or the DNA hypomethylating agent, and "gene activator," 5-azacytidine. Clonal analysis of the untreated tumor population confirmed that immunogenic variants arose as a consequence of the xenogenization protocol. At a dose of 10(6) cells per mouse, nonimmunogenic variants, like the parental tumor line, grew progressively in normal syngeneic recipients. In contrast, immunogenic variants were rejected in normal syngeneic mice and grew progressively only in T-cell-deficient nude mice. Serologic analysis of the respective clonal variants revealed that immunogenic variants expressed substantially elevated (fourfold to tenfold) levels of class I H-2Dk antigen relative to parental or nonimmunogenic cell lines. Two variants exhibiting marginal immunogenicity expressed high and low levels of major histocompatibility complex (MHC) antigen, respectively suggesting that elevated MHC expression, although possibly a contributing factor, did not account for the immunogenic phenotype in all cases. Finally, the immunogenic phenotype of two variants decayed with time in culture. Clones in the process of reversion lost their elevated Dk gene expression and became progressively more tumorigenic in normal syngeneic mice. Together, these data are consistent with a hypothesis that elevated MHC expression can contribute to the immunogenic phenotype of originally low MHC-expressing tumors and that the reduced level of MHC observed in certain clinical cancers may have significant implications with regard to immunologic aspects of the tumor-host relationship.

Heritable high frequency modulation of antigen expression in neoplastic cells exposed to 5-aza-2'-deoxycytidine or hydroxyurea: analysis and implications.

A study was undertaken to determine the ability of the DNA hypomethylating drug 5-aza-2'-deoxycytidine (5-azadCyd) to induce antigen expression in a high proportion of treated tumor cells and to evaluate if this effect could be mimicked by the drug hydroxyurea (HU) which is a genomic DNA hypermethylating agent. Induction of heritable changes in gene expression by 5-azacytidine or the 2'-deoxy analogue (5-azadCyd), at least in some cases, may not be necessarily due to their hypomethylating properties, but to some other induced high frequency genetic change which occurs when DNA synthesis or repair is perturbed. A comparison of 5-azadCyd and HU effects on human and murine tumors was chosen for this study. The phenotypic properties examined with the above treatments were (a) induction of a Mr 110,000 antigen, detected with M111 antibody, in a variant subpopulation (SMeLus7) of human melanoma cells which fail to maximally express this antigen (M111). The parent cell line, MeWo, and other MeWo-derived variant cell cell lines do not demonstrate a similarly inducible phenotype; and (b) induction of class I major histocompatibility complex antigens in a population of mouse mammary adenocarcinoma cells which normally fail to express these antigens. The results showed that 5-azadCyd was effective in inducing antigen expression in both systems whereas HU was effective (and equally so) only in the human melanoma cell line system. In these treatments 5-azadCyd was demonstrated to transiently hypomethylate the same human melanoma cell line whereas HU hypermethylated genomic cytosines. The results suggest that some of the reported effects of 5-azacytidine or 5-azadCyd in inducing very high frequency heritable phenotypic alterations may not necessarily be related to the drugs' ability to cause DNA hypomethylation. The implications of these results are discussed in terms of the use of 5-azacytidine or 5-azadCyd and the effects of chemotherapy on tumor progression and metastasis.

Expression of epithelial-like markers and class I major histocompatibility antigens by a murine carcinoma growing in the mammary gland and in metastases: orthotopic site effects.

A spontaneous murine mammary carcinoma, designated SP1, grew more aggressively in the mammary gland than in the subcutis exhibiting a 10-fold lower 50% lethal tumor dose and the ability to metastasize spontaneously from the orthotopic mammary gland site. The appearance of metastasis could be abrogated by resection of the primary tumor up to 21 days postinjection, arguing against the possibility that metastasis occurred due to trauma of the injection and/or healing processes. In addition, tumor cells recovered from lung metastases exhibited an increased ability to metastasize when reinjected into either the s.c. or mammary sites. Tumor cells from lung metastases showed low levels of Class I major histocompatibility (MHC) antigens, like the parental SP1 cells, but were found to express differentiation markers typical of normal basal and luminal mammary epithelium. SP1 tumors expressed increased Class I MHC antigens, as well as high levels of basal and luminal breast epithelial markers, within 7 days of implantation into the mammary gland. On the other hand, SP1 tumors growing in the subcutis never expressed increased Class I MHC levels and expressed the epithelial marker antigens at lower levels and not until at least 21 days of growth. Removal of host epithelium by cauterization of the mammary bud at 3 weeks had no effect on the increased growth, metastasis and acquired heterogeneity of MHC and epithelial associated antigens, suggesting that the mammary gland stroma was responsible for the observed phenomenon. These findings suggest that the mammary gland either selects distinct tumor subpopulations, or induces a phenotypic change leading to tumor progression and the generation of metastatic subpopulations.

In situ augmentation of class I major histocompatibility antigen expression on immunogenic variants of a spontaneous murine mammary carcinoma.

The relationship between expression of cell surface glycoproteins encoded by the major histocompatibility complex (MHC) and immunogenicity of a recently obtained spontaneous murine mammary adenocarcinoma (designated CBA.SP1) was examined. Immunogenic and nonimmunogenic variant clones were isolated from a subclone of the parent tumor after treatment with the mutagen N-methyl-N'-nitro-N-nitrosoguanidine (MNNG) or the DNA hypomethylating agent and "gene activator," 5-aza-2'-deoxycytidine (5-aza-dCyd). All clones from the untreated tumor population were tumorigenic in normal syngeneic recipients. In contrast, immunogenic variant clones, isolated at high frequencies after drug treatment [ranging from 5% (5-aza-dCyd treated) to greater than 90% (MNNG treated)], were rejected in normal syngeneic mice but grew progressively in T-cell deficient nude mice. Consistent with our previous report (J. Natl. Cancer Inst., 75: 291, 1985), all 5-aza-dCyd induced immunogenic clones expressed elevated levels of class I (particularly Dk) MHC antigens. However some (three out of nine) nonimmunogenic clones also showed enhanced class I MHC expression, implying that not all high MHC expressors were immunogenic. In contrast to 5-aza-dCyd induced variants, only 50% of MNNG induced immunogenic variants showed elevated levels of Dk or Dk and Kk antigens in vitro. Strong augmentation of class I MHC antigens in situ was observed on all immunogenic, but not nonimmunogenic, clones following transplant into syngeneic mice; no increase in MHC expression on variants during progressive growth in athymic nude mice occurred. Although no class II (Ak or Ek) antigens were detected on the parent line or any of the immunogenic variants, a strong infiltration of host I-A bearing cells occurred during immune rejection of SP1 variants. These results are consistent with the hypothesis that induction of class I MHC antigen expression on certain low MHC expressing tumors, although not the sole requirement for immunogenicity, can facilitate immune rejection of the SP1 tumor and, conversely, that the reduced level of MHC observed in certain clinical cancers may significantly affect the immunological aspects of the tumor-host relationship.

Flow cytometric analysis of specific binding of soluble Ia by I-region restricted alloactivated T cells.

An antigen binding assay has been developed for quantitation by flow cytometry of vesicular and soluble Ia binding by alloactivated T cells. Binding of stimulator membrane vesicles was detected by anti-Ly-6.2 or anti-Ia monoclonal antibodies coupled to fluorescent latex beads. Vesicle binding by an I-Ak specific A.TH anti-A.TL T cell line occurred via I-Ak molecules, in that (a) vesicles expressing I-Ak molecules bound much more effectively than vesicles of H-2b,q strains, and (b) inhibition of H-2k vesicle binding occurred with anti-I-Ak, but not anti-Kk, anti-Ek, or anti-Dk antibodies. T cell receptor/Ia interactions were directly studied by inhibition of H-2k vesicle binding by T cells with partially purified Ia glycoproteins. Inhibition of binding occurred via Ia molecules since (a) affinity column partially purified allogeneic I-Ak molecules inhibited binding much more effectively than syngeneic I-As molecules and (b) depletion of I-Ak but not Ek molecules in Iak containing glycoprotein fractions abrogated the inhibitory effect. The ability of this method to detect specific binding of soluble Ia with antigen activated T cells makes it a useful tool for studying interaction of membrane free major histocompatibility complex (MHC) products with native T cell receptor.

An enzyme-linked immunosorbent assay (ELISA) for detergent solubilized Ia glycoproteins using nitrocellulose membrane discs.

An enzyme-linked immunosorbent assay has been developed for the quantitative detection of detergent solubilized murine Ia. Nitrocellulose membrane discs were used to bind membrane glycoproteins applied in solutions containing detergent. The bound antigen was detected by monoclonal antibodies and horseradish-peroxidase-coupled anti-IgG. The assay produced a linear response with respect to antigen concentration, and could readily detect partially purified Ia derived from 10(3) to 10(4) mitogen stimulated spleen cells. Nitrocellulose discs efficiently bound protein in the presence of deoxycholate, taurocholate, and octylglucoside. Less binding occurred in the presence of Triton X-100 or Tween 80, but 90% binding efficiency was obtained in 0.01% solutions of these detergents. The association of protein with the discs was stable under normal conditions for antigen detection, but could be further stabilized by briefly fixing with glutaraldehyde for more rigorous procedures. The ability of this method to detect antigen in detergent solutions makes it useful in monitoring fractions during the purification of cell membrane proteins.

Detection of Fc receptors.

There are an almost bewildering array of assays one can use to detect Fc-receptors, but we have focused, for the most part, on relatively rapid and convenient methods--those that have become standard, e.g., EA rosetting procedures or uptake of soluble antigen-antibody complexes. For most types of Fc-receptor studies these methods are quite adequate. We have also tried to stress the newer methodology of detecting Fc receptors, i.e., the use of anti Fc-receptor antibodies, including monoclonal antibodies. The advent of monoclonal antibody technology, described in detail by Galfrè and Milstein, is changing the approach of many immunological procedures, and detection of Fc receptors is no exception. There are, it should be noted, many newer Fc-receptor methods, e.g., a solid-phase radioimmunoassay and a quantitative fluorometric assay, and methods such as these may be the most appropriate ones to choose in a particular experimental situation. In addition, we have not stressed "functional" Fc-receptor detecting assays, for example, antibody-dependent cell-mediated cytotoxicity (ADCC), which can be used to monitor the activity of Fc receptors in a whole cell population, rather than in single cells.