The social life of NK cells.

Natural killer (NK) cells represent a distinct population of lymphocytes originally identified by its ability to kill transformed cell lines in vitro. It is now clear that these cells also play an important role in the innate immune response against a variety of pathogens, such as virus, bacteria and parasites. In the past few years, different protocols have been developed to activate NK cells ex vivo, allowing a detailed molecular analysis of the interaction of these cells with their cellular targets. NK activity is regulated by signals generated by both inhibitory and stimulatory receptors expressed by target cells. Indeed, recent results indicate that, while major histocompatibility complex class I molecules expressed on target cells inhibit NK lytic activity by engaging surface inhibitory receptors, costimulatory molecules, such as B7-1, B7-2 and CD40, are able to actively trigger NK activity. This review discusses the most recent findings on the role of costimulation on NK activation and forsees the possible consequences of the interaction between NK cells and dendritic cells on the development of an adaptive immune response.

Vaccination with mouse mammary adenocarcinoma cells coexpressing B7-1 (CD80) and B7-2 (CD86) discloses the dominant effect of B7-1 in the induction of antitumor immunity.

Nonreplicating TS/A mammary adenocarcinoma cells expressing B7-2 (CD86) (TS/A-2) are more immunogenic than those expressing B7-1 (CD80) (TS/A-1), indicating that B7-1 and B7-2 display nonredundant costimulatory effects in inducing antitumor responses. Whereas transfection of B7-2 cDNA into TS/A-1 cells does not improve their immunogenicity, transfection of B7-1 cDNA into TS/A-2 cells (TS/A-2/1) decreases their immunogenicity in a manner that is directly related to the surface levels of B7-1. Ab blocking of B7-1 on TS/A-2/1 cells before their injection in vivo restores the higher immunogenicity characteristic of single B7-2 transfectants, indicating therefore that B7-1 actively modulates the B7-2-dependent costimulation. The expression of B7-1 also modifies quantitatively the balance of endogenous IFN-gamma and IL-4 induced in vivo by TS/A-2 vaccines. In fact, we find that vaccination with TS/A-2/1 cells results in the production of more IFN-gamma and less IL-4 than TS/A-2 vaccines, a pattern comparable to that induced by TS/A-1 cells. Thus, in the TS/A model of antitumor response, B7-1 modulates B7-2-dependent costimulatory effects in a dominant, noncompetitive way.

NK cell triggering by the human costimulatory molecules CD80 and CD86.

NK cell-mediated effector functions are regulated by a delicate balance between positive and negative signals. Receptors transmitting negative signals upon engagement with target cell MHC class I molecules have been characterized in detail in recent years. In contrast, less information is available about receptor-ligand interactions involved in the transmission of positive or "triggering" signals to NK cells. Recently, it has been described that murine NK cells are triggered by the costimulatory molecules CD80, CD86, and CD40. Using NK cell lines derived from PBMC as effectors, we demonstrate that the human CD80 and CD86 gene products can function as triggering molecules for NK cell-mediated cytotoxicity. Expression of human CD80 or CD86 molecules in murine B16.F1 melanoma cells rendered these significantly more susceptible to lysis by human NK cell lines. Blocking of the transfected gene products with specific mAb reduced lysis levels to that of nontransfected control cell lines. Triggering of human NK cells by CD80 and CD86 appeared to be independent of CD28 and CTLA-4, at least as determined by the reagents used in the present study, because the expression of these molecules could not be detected on the NK cell lines by either flow cytometry or in redirected lysis assays. Thus, human NK cells may use receptors other than CD28 and CTLA-4 in their interactions with CD80 and CD86 molecules. Alternatively, interactions may involve variants of CD28 (and possibly CTLA-4) that are not recognized by certain anti-CD28 mAb.

Natural killing of MHC class I(-) lymphoblasts by NK cells from long-term bone marrow culture requires effector cell expression of Ly49 receptors.

NK cells from long-term bone marrow culture (LTBMC) were compared with IL-2-activated splenic NK cells [short-term spleen cell culture (STSC)] with regard to expression of inhibitory Ly49 receptors and cytotoxic function. In the LTBMC, the total number of NK cells expressing either one of the Ly49 molecules A, C/I and G2 was strongly reduced (10-15% of NK1.1(+) cells) compared to the STSC (80-90% of NK1.1(+) cells). With regard to cytotoxic function, we confirmed that LTBMC-derived NK cells efficiently killed the prototype NK target YAC-1. However, against other targets, killing was more variable. First, while STSC-derived NK cells clearly distinguished MHC class I(-) from MHC class I(+) tumor cell targets, LTBMC-derived NK cells did not; they either killed both targets equally well or not at all. Secondly, LTBMC-derived NK cells were largely incapable of killing lymphoblast targets deficient in MHC class I expression. To test whether this cytotoxic defect was due to the low number of Ly49(+) NK cells in the LTBMC, we separated Ly49(+) and Ly49(-) NK cells by cell sorting and tested them individually. This experiment showed that only Ly49(+) NK cells in the LTBMC were able to kill MHC class I(-) lymphoblasts (and to distinguish them from MHC class I(+)), despite good cytotoxicity against YAC-1 cells in both populations. These data suggest that certain modes of NK cell triggering are dependent on Ly49 receptor expression. From our results, we speculate that inhibitory receptors are expressed before triggering receptors for normal self cells during NK cell development, which may be an important mechanism to preserve self tolerance during the early stages of NK cell maturation.

Triggering of murine NK cells by CD40 and CD86 (B7-2).

NK cell-mediated cytotoxicity is regulated by both triggering and inhibitory signals. The interaction between MHC class I molecules expressed on target cells and specific MHC class I-binding receptors expressed by NK cells generally leads to inhibition of lysis. We have shown recently that CD80 (B7-1) in mice and CD40 in humans trigger NK cell-mediated cytotoxicity in vitro. In the present study, we show that murine CD40 and CD86 (B7-2) trigger murine NK cell-mediated cytotoxicity in vitro when expressed on tumor cells. Preincubation of the transfected cell lines with anti-CD40 F(ab')2 fragments or cytolytic T lymphocyte-associated Ag-4-Ig (CTLA-4-Ig) before the cytotoxic assay abolished the triggering effect. Furthermore, radiolabeled CD40- and B7-2-expressing cells were rapidly eliminated in vivo in an NK cell-dependent manner. NK cells from CD40 ligand (CD40L)-/- or CD28-/- mice were triggered by tumor cells transfected with CD40 and B7-2, respectively, and these transfectants were rapidly eliminated in vivo when inoculated into CD40L-/- and CD28-/- mice. This suggests that the CD40 and B7-2 molecules can interact with receptors on NK cells other than CD40L and CD28, respectively, and that these may account for some of the reactivities observed in the present study. Collectively, these data demonstrate that 1) costimulatory molecules, other than B7-1, can modulate NK cell responses in vitro, 2) they can also affect NK cell-dependent responses in vivo, and 3) parts of these reactions are independent of CD28 and CD40L.

The immunogenicity of experimental tumors is strongly biased by the expression of dominant viral cytotoxic T-lymphocyte epitopes.

The immunogenic Friend-Moloney-Rauscher (FMR) virus-induced tumors have been used extensively to clarify the cellular and molecular mechanisms responsible for tumor rejection and to develop immunotherapeutic strategies. We characterize here the trimolecular complex MHC class I-antigenic determinant-T cell receptor involved in the induction of a protective CTL response against the RMA thymoma. This complex is mainly composed by the D(b) molecule interacting with a Rauscher virus antigen (Ag) determinant and the Vbeta5+ T cell receptor. We also show that the chemically induced EL-4 thymoma acquires the susceptibility to recognition by anti-RMA CTLs and the ability to elicit a protective anti-RMA CTL response only upon infection by a virus of the FMR family and that RMA and FMR virus infected EL-4 cells share tumor-associated Ag. The data strongly support the hypothesis that the high immunogenicity of virus-induced or infected tumors is determined by the expression of immunodominant virus-encoded Ag. The demonstration of a different outcome in the immune responses elicited in the presence or in the absence of viral Ag further open the contention of the molecular requirements for immunogenicity and should stimulate a more careful revision of unexpected cross-reactivity among tumors.

Rejection of a nonimmunogenic melanoma by vaccination with natural melanoma peptides on engineered antigen-presenting cells.

Naturally processed peptides, obtained by acid extraction of tumor cells, contain Ags able to activate specific CTL in vitro. We recently reported that the nonprofessional APC, RMA-S, expressing the B7.1 molecule (RMA-S/B7), pulsed with naturally processed peptides from the nonimmunogenic B16F1 melanoma (B16F1a.e.) primed syngenic CD8+ T cells against the tumor in vitro. Here, we show the rejection of B16F1 melanoma by C57BL/6 mice after immunization with RMA-S/B7 cells pulsed with B16F1a.e. This response is critically dependent on both CD4+ and CD8+ cells, but not on NK cells. However, only CD8+ T cells exert anti-B16F1 cytolitic activity in vitro. Moreover, RMA-S/B7 cells pulsed with B16F1a.e. can be used to prevent the growth of 24-h preestablished melanomas. These results may have important implications for the clinical use of natural peptide fractions of tumor cells as therapeutic cancer vaccines.

Heterogeneous effects of B7-1 and B7-2 in the induction of both protective and therapeutic anti-tumor immunity against different mouse tumors.

Experimental mouse tumors are classified as intrinsically immunogenic when, after a single injection into syngeneic mice as nonreplicating cell vaccines, they elicit a protective immune response against a subsequent lethal challenge. Tumors that do not retain this residual immunogenicity are defined as poorly immunogenic or nonimmunogenic. The expression of the B7-1 co-stimulatory molecule on immunogenic tumors can further increase their capacity to induce a T cell-dependent anti-tumor immunity, whereas it has limited effects on nonimmunogenic tumors. Recently, B7-2, a second molecule with an apparently similar co-stimulatory activity, has been cloned. In this report, we compare the efficiency of nonreplicating cells from one immunogenic and two nonimmunogenic mouse tumors transfected with B7-1 or B7-2 in the induction of protective and curative anti-tumor immunity. Immunogenic lymphoma cells expressing B7-1 or B7-2 are equally effective in both protecting against a subsequent challenge and curing established tumors. By contrast, nonimmunogenic adenocarcinoma and melanoma cells expressing B7-2 provide superior protective immunity, and only B7-2+ adenocarcinoma cells induce an efficient curative immunity. CD8+ and polymorphonuclear cells, but not CD4+ T cells, are critically involved in the rejection of the adenocarcinoma elicited by both B7-1+ and B7-2+ vaccines. These data indicate that B7-1 and B7-2 are not redundant co-stimulatory molecules and that, in these experimental models, B7-2 is superior to B7-1 in the induction of an efficient immunity when the immunogenicity of a tumor is a limiting factor.

Co-expression of B7-1 and ICAM-1 on tumors is required for rejection and the establishment of a memory response.

Although the transfection of B7-1 cDNA into a few mouse tumor cell lines can induce anti-tumor T cell immunity, its expression alone is ineffective in many other tumor cell lines tested. We were interested to study what factors limit B7-1 co-stimulatory activity, and decided to investigate whether B7-1 requires the cooperation of ICAM-1 to provide the minimal co-stimulatory signal for establishing an efficient anti-tumor immunity. We show that the transfection of B7-1 cDNA into three ICAM-1+ (plasmocytoma J558L, T lymphomas EL-4 and RMA), but not into two ICAM-1- tumors cell lines (adenocarcinoma TS/A and melanoma B16.F1), is sufficient to induce their complete rejection in syngeneic mice. The expression of ICAM-1 is necessary for the rejection of the B7 expressing tumors, since the primary response elicited by B7-1+ EL-4 and RMA clones expressing reduced levels of ICAM-1 is severely reduced. Furthermore, super-transfection of ICAM-1 cDNA into B7-1+ adenocarcinoma and melanoma clones optimizes their primary rejection. Histologic examination of transfected tumors reveals that B7-1 and ICAM-1 exert a potent pro-inflammatory activity. The intra-tumor infiltration is composed of both eosinophils and lymphomonocytes, and is already massive 5 days after the tumor challenge. The primary rejection of the B7-1+ ICAM-1+ tumors depends critically on CD8+ T cells, natural killer cells and granulocytes, but is independent of CD4+ T cells. Remarkably, in addition to its effects on the early phases of the immune response, the co-expression of ICAM-1 and B7-1 on tumors is also necessary for the efficient induction of a memory response. In fact, only the primary challenge with B7-1+, ICAM-1+ tumor cells protects the majority of the mice from a second injection of parental tumor cells. Collectively, our findings indicate that B7-1 and ICAM-1 are fundamental components for triggering the primary rejection of tumors and establishing a protective memory response. These findings may help to define new strategies for the rational application of co-stimulation in tumor immunotherapy.

Role of thymic stromal cells in thymocyte education: a comparitive analysis of different models.

The thymus is the privileged lymphoid organ for T-cell differentiation, including clonal selection. Precursor thymocytes are selected on the basis of their ability to recognize self-major histocompatibility complex molecules (MHC) that are expressed on thymic stromal cells, including epithelial cells, macrophages, and interdigitating cells/dendritic cells (IDCs/DCs). Epithelial cells have been associated with positive selection, and dendritic cells with negative selection, but the exact role played by these stromal elements is still unresolved. This review focuses on the different in vitro and in vivo approaches that have been used to elucidate the role of thymic stromal cell types in thymic function, and on a critical evaluation of these approaches.

In vitro effects of deoxyguanosine (dGuo) on thymic stromal cells. A discussion.

Different in vitro approaches have been used to enrich thymic epithelial cells for the analysis of their role in inducing tolerance. One widely used approach consists of the organ culture of mouse fetal thymic lobes (FTOCs) in the presence of 2'-deoxyguanosine (dGuo), that is thought to selectively eliminate both thymocytes and interdigitating cells/dendritic cells (IDCs/DCs) from the thymic cultured tissue. The dGuo-induced effects on rat thymic stromal cells appear to differ from those described in mouse. In this paper we present evidence supporting the presence of rat DCs in both thymic cultured fragments and thymic stromal cell primary cultures after dGuo treatment. Possible explanations for these contradictory results are discussed.

The neuro-endocrine component of the rat thymus: studies on cultured thymic fragments before and after transplantation in congenitally athymic and euthymic rats.

An immunohistochemical study was done for the presence of tyrosine hydroxylase (noradrenergic innervation), neuron-specific protein PGP9.5, and anterior pituitary hormones (beta-subunit of follicle-stimulating hormone, growth hormone, beta-subunit of luteinizing hormone, prolactin, and beta-subunit of thyroid-stimulating hormone) in cultured thymic fragments before and after transplantation in congenitally athymic and euthymic rats. The cultured thymic fragments consisted of epithelial cells and were depleted of lymphocytes. After implantation in syngeneic and allogeneic athymic recipients and in syngeneic euthymic recipients, a recovery of the original architecture was found within 6 weeks; rejection occurred within 3 weeks for allogeneic transplantation in euthymic rats. During culture nerve-like profiles almost disappeared from the tissue, and reappeared simultaneously with the influx of host-derived cells and the restoration of the original thymic architecture. A high immunoreactivity for hormones and PGP9.5 was found in epithelial cells after culture and in the first phase after transplantation. These epithelial cells may represent precursor-epithelial cells, based on their unusual ultrastructure and combined expression of markers that in the normal thymus occur only on subcapsular/medullary epithelium or on cortex epithelium. These data indicate a potential role of the neuroendocrine function of the thymus during restoration of the thymus architecture starting from precursor-like epithelial cells.

Transplantation of cultured thymic fragments in congenitally athymic and euthymic rats. Culture with deoxyguanosine or cyclosporin A does not influence the histologic characteristics and outcome after transplantation in syngeneic and allogeneic combinations.

Cultured thymic fragments (CTF) from WAG/CPB (RT1u) and DA/01a (RT1a) rats were prepared in the presence or absence of 2'deoxyguanosine or cyclosporin A, and subsequently transplanted under the kidney capsule of congenitally athymic and euthymic WAG/CPB recipients. The rationale of the culture supplements was that these may affect the disappearance of medullary dendritic cells, with subsequent induction of allotolerance. However, the immunohistology of the CTF showed more RT1 class II-positive cells than keratin-positive cells, indicative of the maintenance of dendritic cells. Grafts in athymic animals showed the recovery of the original thymic architecture within 6 weeks after transplantation. The influx of host-derived lymphocytes was accompanied by an influx of dendritic cells in the medulla-like area and macrophages in the cortex. A similar recovery was observed for syngeneic CTF in euthymic recipients. In addition lymphocytic infiltration was seen in the connective tissue surrounding the epithelial areas. Allogeneic grafts in euthymic animals were rejected within 3 weeks after transplantation. This outcome of the transplanted CTF under different conditions was not affected by the supplementation of the thymic culture before transplantation with 2'deoxyguanosine or cyclosporin A. We conclude that there is no tolerance induction after transplantation in euthymic allogeneic rats of CTF prepared in the presence of 2'deoxyguanosine. This conclusion is in contrast to data in the mouse, which may be explained by the maintenance of dendritic cells during culture. A chimaeric state of donor-derived epithelium and host-derived dendritic cells is obtained by transplantation of allografts in athymic rats.