Regulation of in vitro cytotoxic T lymphocyte generation. I. Evidence that killer cell precursors differentiate to effector cells in two steps.

The differentiation of cytotoxic T lymphocyte precursor cells (CTL-P) into CTL effector cells is a two-step process. In the first step, naïve CTL-P (CTL-PN) become activated (CTL-PA) but do not yet have the capacity to kill target cells. CTL-PA can be distinguished from CTL-PN because the former are far less sensitive than the latter to the effects of in vitro-generated suppressor cells. Thus, the addition of suppressor T cells (Ts) to a fresh MLC can totally inhibit the production of CTL from CTL-PN, whereas the same Ts only minimally affect the generation of CTL from CTL-PA. It is not known whether these Ts act directly on CTL-PN or on a helper cell needed for activation to CTL-PA. The production of CTL-PA can take place in allogeneic mixed leukocyte cultures (MLC) treated with the drug pyrilamine, or when heat-inactivated stimulator cells are used. Each of these treatments inhibits the differentiation of CTL-PA to CTL. However, if pyrilamine is removed, a nonspecific MLC-derived signal can induce these CTL-PA to become CTL, even in the presence of significant numbers of Ts. This two step process of differentiation of CTL-P to CTL may be analogous to the way naïve B cells become antibody-producing cells.

Intermediary role of macrophages in the passage of suppressor signals between T-cell subsets.

We have examined the ability of macrophages (Mphi) to transmit T-cell derived suppressor signals to other T cells. The suppressor signal studied is an antigen-specific factor which suppresses the ability of adoptively transferred, sensitized lymphocytes to express contact hypersensitivity in normal recipients. We have found that this factor binds to peritoneal exudate Mphi via cell surface structures which can be blocked with heat-aggregated gamma globulin. Dead (HK) Mphi bind the factor but fail to present it in a functional way to assay (immune) T cells, whereas live (L) Mphi perform both functions. Further, L Mphi can retrieve the factor in an active form from the surfaces of HK Mphi. Based on these and other findings (1-5), we discuss the possibility that Mphi may play as important a role in presenting T-cell communication signals to the cells of the immune system as they do in presenting antigen.

Feedback induction of suppressor T-cell activity.

Adoptively transferred carrier immune T cells interact with nonimmune T cells in recipients in a fashion which generates specific immunosuppression although both the immune and normal cells function quite well as helper cells when not admixed.

Mechanisms of suppression in the transfer of contact sensitivity. Analysis of an I-J+ molecule required for Ly2 suppressor cell activity.

The passive transfer of contact sensitivity (CS) by immune cells can be inhibited with an antigen-specific T suppressor factor. This factor is composed of two subfactors: an antigen-specific subfactor made by an Ly1+ cell (PC1-F) and a antigen nonspecific subfactor made by an Ly2+ T cell (TNBSA-F). The suppressive activity of the complete factor can be eliminated by depleting the assay population of Ly2+ cells, even though it is the Ly1+ cell in the population that transfers the adoptive immunity. This suggests that the Ly2+ cell in the assay population is needed to transduce the suppressive signal to the Ly1+ effector cell of DTH. We found that an Ly2+ cell from immune animals could be induced to produce a cell free subfactor that overcame the requirement for this Ttrans cell in the suppression of CS by TsF. The induction required only PC1-F, TNP-coupled spleen cells, and resulted in the production of an antigen-nonspecific I-J+ subfactor by immune Ly2+, I-J+ cells. The need for the Ly2+ transducer cell could also be overcome by addition of an I-J+ molecule secreted by Ly1 T cells hyperimmunized to SRBC. A suppressor complex made from mixing the I-J+ molecule with TNBSA-F could directly suppress the functional activity of immune T cells not only to transfer CS, but also to deliver help to B cells in an in vitro PFC response. This suppressive complex is antigen-nonspecific and does not require Ly2+ T cells in the assay population for suppressive activity. These results indicate that effector factors of the suppressor circuit require two molecules; one that contains the functional suppressor material and one that serves as a "schlepper," a molecule needed to deliver the suppression to the appropriate target cell. The ability to construct a functional suppressor complex from two subfactors raised against different antigens, using different immunization procedures, which were isolated from factors exhibiting different functional activities suggests that certain cells of the immune system may play a universal role in "transducing" the suppressive signal.

Analysis of the interactions between two molecules that are required for the expression of Ly-2 suppressor cell activity. Three different types of focusing events may be needed to deliver the suppressive signal.

We described a T suppressor factor made by an I-J- Ly-2 T cell (Ly-2 TsF) that expresses biological activity only when its acceptor cell shares H-2-linked polymorphic genes with the cells that made the Ly-2 TsF (or when the producer cell had differentiated in a thymic environment where the gene products of the acceptor cell were expressed). The Ly-2 TsF requires the presence of I-J+ Ly-1 cells in the assay culture to express its suppressive activity, although removal of the I-J+ Ly-1 cells in the assay cultures with an I-J+ soluble factor derived from them. This I-J+ molecule not only fails to bind antigen but is also antigen nonspecific in that it can come from Ly-1 cells making factors of irrelevant specificities. For the I-J+ molecule to replace the activity of the I-J+ Ly-1 cell in the assay population, in restoring suppressive function in cultures depleted of I-J+ Ly-1 cells, it must share genetic polymorphisms linked to the I-J subregion with the Ly-2 TsF and genetic polymorphisms linked to Igh-V with the target cell. These results indicate that an I-J+ antigen-nonspecific molecule combines with an antigen-specific Ly-2 TsF via an I-J- anti-I-J "type" of interaction. The resultant molecular complex is focused on a cell surface receptor of the acceptor cell. This focusing event is controlled by the antigen-nonspecific I-J+ molecule, and the precise interaction with the receptor on the acceptor cell is controlled by Igh-V-linked polymorphic gene products. The antigenic specificity of the interaction is controlled by a receptor for antigen on the I-J- component of the complex. Thus, three focusing events are required for Ly-2 TsF to express biologic activity: (a) the Ly-2 TsF must be focused on an acceptor cell that has the same antigenic specificity (most likely via an antigen bridge); (b) it must also be focused onto an I-J+ antigen-nonspecific molecule that we refer to as a "schlepper" molecule (most likely via an I-J anti-I-J bridge); and (c) the schlepper molecule must focus the molecular complex on an Igh-V-controlled receptor on the antigen-specific target cell.

The response of T cells to histocompatibility-2 antigens. Dose-response kinetics.

The DNA synthetic response of a wide variety of parental thymocyte doses was studied in the lymphoid tissues of lethally irradiated F(1) mice. The response curves of the thymocytes were strikingly similar in shape to those of T cells responding to antigens, such as sheep red cells, which are more labile than the histocompatibility antigens of the F(1) host. The response was characterized by a dose-dependent latent period of 1-3 days, followed by a sharp increase in activity and a significant subsequent shutoff. Larger thymocyte doses tended to shorten the latent period. A comparison of the responses of different cell doses to one another indicated that the response usually simulated a suppressed first-order reaction. However, occasional multiple order reactions were observed. We have interpreted these data to indicate that both positive and negative interactions occur between thymocytes, without the mediation of B cell products such as conventional antibodies.

Potentiation of the T-lymphocyte response to mitogens. I. The responding cell.

Human and mouse lymphoid cells, stimulated by phytohemagglutinin (PHA) or lipopolysaccharide W (LPS), release supernatant factor(s) which are mitogenic for mouse thymocytes and which potentiate their responses to PHA or concanavalin A (Con A), The term LAF (lymphocyte-activating factor) is proposed for this activity. LAF not only enhances the mitotic responses of the less dense thymus subpopulations (A, B, and C) separable on discontinuous bovine serum albumin (BSA) gradients but also gives substantial responses in the otherwise inert cells of the denser fractions D and P. LAF does not exert a potentiating stimulatory effect on the responses of unfractionated mouse spleen cells, but does act synergistically with PHA on nonadherent spleen cells and on spleen cells of mice of several strains 5 days after irradiation and injection of thymocytes. Similarly LAF, which has no visible effect on unfractionated human peripheral blood cells, strongly potentiates the PHA response of column-purified lymphocytes, when these are cultured at low concentration. We conclude that LAF stimulates both central and peripheral T lymphocytes and enhances their responses to other stimulants.

Identification of a B-cell surface structure involved in antigen-dependent triggering: absence of this structure on B cells from CBA/N mutant mice.

CBA/N mice have an X-linked B-cell maturation defect which is reflected in part in an absence or dysfunction of a subclass of mature B cells. We have immunized the defective male offspring of the mating (CBA/N female X BALB/c male) with BALB/c spleen cells. The resulting antiserum (alphaLyb3) selectively reacts with a component on the surface of a portion of B cells from a panel of H-2 different mouse strains. Binding of alphaLyb3 serum to this B-cell subclass results in substantial (10- to 20-fold) enhancement of the antibody response to low doses of SRBC. Both binding and enhancing activity are removed by absorption with B cells from B6 and BALB/c, but not CBA/N mice. Absorption of the serum with bone marrow cells, T cells, or thymocytes from Lyb3+ strains does not remove activity. Since the enhanced plaque-forming cell (PFC) responses are specific for the immunizing antigen, and since no PFC response is produced by injection of the antiserum alone, this enhancement probably reflects a second signal produced by specific interaction between antibody and the surface Lyb3 component. Moreover, this signal can partially replace the requirement for T cells in the production of antibody to a "thymus-dependent" antigen. These findings (taken in conjunction with the previously described immune defects in CBA/N mice and other studies of B-cell maturation) suggest to us that Lyb3 is a cell surface component expressed selectively on a mature B-cell subclass. This component is important in B-cell triggering by antigen and fails to develop in CBA/N mice, due to a dysfunction of a regulatory gene on the CBA/N X chromosome.

Role of contrasuppression in the adoptive transfer of immunity.

The data presented in this paper show that the population of cells that adoptively transfer contact hypersensitivity are Lyt-1+ 2-, I-J- and nonadherent to V. villosa lectin. However, the adoptive transfer of immunity by this population of cells is successful only when the recipient has been treated in such a way as to impair the host immunosuppression mechanism. This population cannot, on its own, transfer immunity to adult, untreated naive recipients unless an additional population of immunoregulatory cells is present. This immunoregulatory population does not itself adoptively transfer immunity. This latter population is differentiated from the immune cells in that they are Lyt-1+ 2-, I-J+ and are adherent to V. villosa lectin. Both populations are required to adoptively transfer immunity to adult untreated naive recipients.

Genetic control of immunoregulatory circuits. Genes linked to the Ig locus govern communication between regulatory T-cell sets.

Antigen-stimulated Ly1:Qa1+ cells induce a nonimmune set of T-acceptor cells (surface phenotype Ly123+Qa1+) to participate in the generation of specific suppressive activity. The experiments reported here were designed to test the possibility that the interaction between T-inducer and T-acceptor cells might be governed by genes linked to the Ig locus. We find that inducer:acceptor interactions occur only if the inducer and acceptor T-cell sets are obtained from donor that are identical at the Ig locus and are independent of the Ig locus expressed on the B cells used for assay of T-helper activity. In addition, experiments using inducer and acceptor T cells from the congenic recombinant BAB. 14 strain show that T-T interactions are not governed by Ig-CH genes, per se. These data indicate that T-inducer: T-acceptor interactions are governed by Ig-linked genes that may control expression of VH-like structures on T cells, or control expression of as yet unidentified cell-surface molecules.

T cell-dependent mast cell degranulation and release of serotonin in murine delayed-type hypersensitivity.

We have previously suggested that the release of serotonin (5-hydroxytryptamine) (5-HT) by local tissue mast cells is required for the elicitation of delayed-type hypersensitivity (DTH) in mice. In the current study, light microscopic radioautographs from animals treated with [3H]5-HT indicated that local mast cells released 5-HT between 6 and 18 h during the evolution of DTH. Ultrastructural examination of mast cells revealed surface activation, indicated by extension of surface filopodia, and degranulation by fusion and exocytosis. Light and electron microscopic studies of the endothelium of postcapillary venules at sites of DTH revealed the development of gaps between adjacent cells. The development of gaps permitted extravasation of tracers that was abolished by depletion or antagonism of 5-HT. Thus mast cells degranulated and released 5-HT in DTH, and this 5-HT acted on local vessels. Recipients of nonadherent, non-immunoglobulin-bearing sensitized lymphocytes also demonstrated similar mast cell degranulation and the formation of endothelial gaps. This indicated that mast cell degranulation and 5-HT release in murine DTH were probably T cell dependent.

Cell interactions in the suppression of in vitro antibody responses.

Normal T and immune B lymphocytes interact in a fashion that leads to suppression of the immune response. Normal spleen cells added to cultures of primed spleen cells specifically suppressed both the IgM and IgG secondary antibody response of the primed cells to less than 30% of the response of the immune cells cultured alone. Cell crowding as a possible in vitro artifact was ruled out. The suppression was specific for the priming antigen, even when the specific and nonspecific antigens were included in the same cultures. Suppression required both normal T and immune B cells to be present in culture. We suggest that the immune population produces a signal that can induce normal T cells to become specific suppressor cells. This form of interaction may represent an important regulatory (homeostatic) mechanism in the immune system.

Suppressor cells: dependence on assay conditions for functional activity.

Spleen cells educated in vitro with sheep red blood cells (SRBC) suppressed the plaque-forming cell response of Mishell-Dutton assay cultures challenged with optimal doses of SRBC. Changing conditions in the assay cultures changed the effect educated cells had on the assay culture responses. For example, educated cells helped rather than suppressed assay cultures of suboptimal numbers of spleen cells. Similarly, augmentation resulted upon addition of educated cells to assay cultures challenged with suboptimal doses of SRBC. Such a reversal of regulatory effects was not observed when assay cultures were challenged with supraoptimal antigen doses. Educated cells helped assay cultures of B spleen cells, and the addition of normal T cells reinstated suppression. Furthermore, maintenance of assay cultures under stationary rather than the usual rocking conditions allowed educated cells to help rather than suppress the antibody response of assay cultures. These results show that when the response of the target population (assay cultures) is low, the regulator (educated) cells augment the response, and vice versa, supporting the hypothesis that the effect regulator cells produce depends on the activity of the cells they regulate.

Cell-mediated immunity: delayed-type hypersensitivity and cytotoxic responses are mediated by different T-cell subclasses.

Cell-mediated immunity includes both the generation of cytotoxic cells and initiation of delayed-type hypersensitivity (DTH). The resting T-cell population, before stimulation by antigen, already contains cells of the Lyl subclass that are programmed to initiate DTH (and helper function) but not cytotoxic responses, as well as Ly23 cells which can generate killer activity (and suppressive function) but not DTH. The central implication of these findings is that the broad division between humoral and cell-mediated immune responses does not precisely correspond to the division of labor among T-cell subclasses. The relative contribution of DTH-competent Lyl cells and cytotoxic Ly23 cells to the classical homograft response remains to be determined.

Requirement for vasoactive amines for production of delayed-type hypersensitvity skin reactions.

The skin sites of the mouse where delayed-type hypersensitivity (DTH) reactions are most easily elicited (foot pads and ears) are particularly rich in 5-hydroxytryptamine (5-HT)-containing mast cells. Since mice are deficient in circulating basophils, which play a role in at least some DTH reactions, we investigated the possibility that the mast cells were playing an important role in the evolution of the skin reactions of DTH in mice. We found that reserpine, a drug which depletes mast cells of 5-HT, abolished the ability of the mouse to make DTH reactions in the skin. The suppressive effect of reserpine could be partially blocked by monoamine oxidase inhibitors which prevent the degradation of 5-HT in the cytosol of the mast cell. Spleen cells of immune, reserpine-treated mice transferred DTH reactions to nonimmune mice normally, indicating that the reserpine treatment did not affect immune T cells. DTH reactions could not be transferred into reserpine-treated mice. We suggest that T cells are continually emigrating from the blood, through postcapillary venule endothelium, by a mechanism which does not depend on vasoactive amines. If they are appropriately immune and meet the homologous antigen in the tissue, they induce mast cells to release vasoactive amines which cause postcapillary venule endothelial cells to separate, allowing the egress from the blood of cells which ordinarily do not recirculate. The secondarily arriving vasoactive amine-dependent cells are responsible for the micro- and macroscopic lesions of DTH reactions. Chemotactic factors may also be involved in bringing cells to the DTH reaction sites but we propose that T-cell regulation of vasoactive amine-containing cells allows the effector cells to pass through the endothelial gates after they are called.

Augmentation of delayed-type hypersensitivity by doses of cyclophosphamide which do not affect antibody responses.

Mice immunized with more SRBC than are required to produce optimal delayed-type hypersensitivity reactions, developed good antibody responses and poor delayed foot pad reactions. Cyclophosphamide treatment in low doses (20 mg/kg) before immunization, augmented the delayed-type hypersensitivity without affecting antibody responses. Cyclophosphamide did not augment delayed responses to optimal doses of SRBC (0.01%), but did augment the delayed hypersensitivity response of mice immunized with a suboptimal antigen dose (0.001%); which produced no detectable antibody response with or without cyclophosphamide pretreatment. These results suggest that antibody feedback is not the sole regulator of delayed reactions; the possibility that suppressor T cells may also be involved is discussed.

Homologies between cell interaction molecular controlled by major histocompatibility complex- and Igh-V-linked genes that T cells use for communication. Tandem "adaptive" differentiation of producer and acceptor cells.

We have asked the question: how do partner cells in immunoregulatory interactions between T cell subsets acquire the ability to recognize and react appropriately with one another? In particular, we have asked whether these communication events are completely determined by the cell's genetic constitution, or whether the recognition events can be learned during ontogeny. We have found that the T cells of parent into F1 chimeras and homozygous nude mice with F1 thymus grafts not only learn to react with genetically disparate acceptor cells, but that the receptors on the acceptor cells themselves learn to react with genetically disparate producer cells. This learning process can overcome both major histocompatibility complex- and immunoglobulin heavy chain variable region-linked restricted communication between T cell subsets. We interpret these findings to indicate that thymic elements can start a cascade of differential events. The thymic elements, whether endogenous or passively acquired, select from a pool of producer cells those that will react appropriately with the thymic selecting cells, and these cells become expanded. Then, the private markers (idiotype) on the expanded pool of producer cells act as selecting and expanding elements that choose from a pool of acceptor cells those that recognize the producer cells idiotype as self. This second differentiational event, although apparently thymus evidence that this type of acceptor cell differentiation could also take place during the course of an immune response.

Role of antigen-presenting cells in the development and persistence of contact hypersensitivity.

Three outcomes pertinent to contact sensitivity (CS) follow immunization with various forms of trinitrophenylated (TNP) substrates: (a) specific immunological unresponsiveness for CS is induced when immunization favors activation of splenic suppressor cells. This state is achieved by intravenous injection of trinitrophenyl-conjugated to various types of cells, such as peritoneal exudate cells (PEC). (b) A short-lived or evanescent form of CS is induced when immunization reduces activation of the suppressor circuit. This can be achieved by subcutaneous immunization with trinitrophenyl conjugated to syngeneic PEC, by pretreatment with cyclophosphamide to diminish suppression before intravenous immunization, or by altering the mode of antigen presentation by using TNP-substrate that has undergone phagocytosis. (c) A long-lived form of CS is induced when trinitrophenyl is presented to the immune system on skin cells either by contact skin painting with reactive trinitrophenyl, or by subcutaneous, or even intravenous injection of trinitrophenyl-conjugated epidermal cells. In fact, trinitrophenyl-conjugated epidermal cells induced CS even when the suppressor circuit was activated by intravenous coadministration of TNP-PEC. This implies that antigen presentation on epidermal cells induces sensitized cells that are relatively resistant to suppression. The cell type(s) in the skin that are primarily responsible for this potent form of antigen presentation are most likely Langerhans cells, because they can be concentrated by virtue of their Fc receptors and they are Ia positive. Thus, both the anatomical site where antigen is first encountered by the immune apparatus, as well as the nature of the cells which present the antigen, determine whether a CS response will ensue, as well as whether it will be evanescent or long-lasting.

Molecular composition of an antigen-specific, Ly-1 T suppressor inducer factor. One molecule binds antigen and is I-J-; another is I-J+, does not bind antigen, and imparts an Igh-variable region-linked restriction.

Immunized Ly-1+2-T cells (Ly-1 cells) make an antigen-specific soluble suppressor product (Lyl-1 TsiF) that will induce Ly-2+ cells to express suppressive activity but only if the Ly-2+ and the Ly-1 producer cell share genetic polymorphisms that are linked to the Igh locus and in particular that part where the Igh-V (or VH) is encoded. Ly-1 TsiF can be separated into entities, one binds antigen and does not express I-J determinants, and the other is I-J+ and does not bind antigen. Neither of these "subfactors" has biological activity, but a 50:50 mixture of them reconstitutes biological activity that expresses the antigen specificity of the antigen-binding molecule. Any of the three heterologous erythrocytes (antigens) studied can be used for immunization to produce the I-J+ nonantigen-binding factor, i.e., the I-J+ moiety makes no contribution to the factor's specificity. It does, however, determine the intact factor's Igh-V linked restriction. Thus, the antigen combining site of the factor is irrelevant to the factor's Igh-V restriction but crucial for its specificity. The I-J+ molecule does not bind antigen nor influence the factor's antigen specificity but expresses the Igh-V polymorphism (or anti-Igh-V polymorphism) that is required for the transmission of an inductive signal to the factor's Ly-2+ acceptor cell.

Nylon adherent antigen-specific rosette-forming T cells.

Shortly after intravenous immunization of mice with heterologous erythrocytes (RBC) antigen-specific Thy 1+ cells which form rosettes with the immunizing RBC (thymic-derived lymphocytes-forming rosettes [T-RFC]) appear in the spleen. These T-RFC are much less stable than Thy 1- RFC (non-thymic-derived [B-RFC]) although most if not all of both classes of RFC adhere to nylon. T-RFC are induced with low doses of antigen (which fail to induce B-RFC) and are inhibited by higher antigen doses which are optimal for induction of B-RFC. Pretreatment of mice with cyclophosphamide prevents the high dose inhibition of T-RFC. Although there are many parallels between the production of T-RFC and delayed-type hypersensitivity (DTH) it is unlikely that the T-RFC are essential for DTH reactions since DTH can be transferred with cells which pass through nylon, and such cells are almost totally depleted of T-RFC. Thus immunization can lead to the production of large numbers of antigen-specific T-RFC whose functional role in the immune response is unknown. However, the characteristics of the T-RFC suggest that they may play an important role in amplification of suppressor cell activity.