IL-2 and IL-7 differentially induce CD4-CD8- alpha beta TCR+NK1.1+ large granular lymphocytes and IL-4-producing cells from CD4-CD8- alpha beta TCR+NK1.1- cells: implications for the regulation of Th1- and Th2- type responses

Effector functions of CD4-CD8- double negative (DN) alpha beta TCR+ cells were examined. Among mouse DN alpha beta TCR+ thymocytes, NK1.1+ cells expressing a canonical V alpha 14/J alpha 281 TCR but not NK1.1- cells produce IL-4 upon TCR cross-linking and IFN-gamma upon cross- linking of NK1.1 as well as TCR. Production of IL-4 but not IFN-gamma from DN alpha beta TCR+NK1.1+ cells was markedly suppressed by IL-2. Whereas V alpha 14/J alpha 281 TCR+ cells express NK1.1+, these cells are not the precursor of DN alpha beta TCR+NK1.1+CD16+B220+ large granular lymphocytes (LGL). IL-2 induces rapid proliferation and generation of NK1.1+ LGL from DN alpha beta TCR+NK1.1- but not from DN alpha beta TCR+NK1.1+ cells. LGL cells exhibit NK activity and produce IFN-gamma but not IL-4 upon cross-linking of surface TCR or NK1.1 molecules. In contrast to IL-2, IL-7 does not induce LGL cells or NK activity from DN alpha beta TCR+NK1.1- cells but induces the ability to produce high levels of IL-4 upon TCR cross-linking. Our results show that DN alpha beta TCR+ T cells have several distinct subpopulations, and that IL-2 and IL-7 differentially regulate the functions of DN alpha beta TCR+ T cells by inducing different types of effector cells.      

Inhibition of skin xenograft rejection by depleting T-cell receptor alpha beta-bearing cells without T-cell receptor gamma delta-bearing cells or natural killer cells by monoclonal antibody.

We compared the effects of in vivo administration of the anti-T-cell receptor (TCR) alpha beta monoclonal antibody (mAb) (H57-597) to those of the anti-CD3 mAb (145-2C11), with or without anti-NK1.1 mAb (PK136), on xenogeneic skin graft survival in mice. In anti-TCR alpha beta mAb-treated B6 mice, F344 rat skin grafts survived for about 54 days, whereas in anti-CD3 mAb-treated B6 mice with or without anti-NK1.1 mAb treatment grafts survived about 25 days. In anti-TCR alpha beta mAb-treated B6 mice, TCR alpha beta-bearing T-lymphocyte function was completely abrogated, although TCR gamma delta-bearing T-lymphocyte function was still intact on day 9. In the anti-CD3 mAb-treated mice, the functions of both types of T lymphocytes were completely abrogated. On day 32, when most of the skin xenografts had been rejected in the anti-CD3 mAb-treated mice, the functions of both T lymphocytes had recovered considerably, and could actually respond to F344 antigens. In contrast, the function of TCR alpha beta-bearing cells had only partially recovered in the anti-TCR alpha beta mAb-treated mice. Finally, natural killer (NK) activity in the anti-TCR alpha beta mAb-treated mice was intact on day 32, when rat skin grafts still survived. In contrast, NK activity in the anti-CD3 mAb plus anti-NK1.1 mAb-treated mice did not recover on day 32, when skin xenografts had already been rejected. These results suggest that TCR gamma delta-bearing T cells and NK cells by themselves, at least in the absence of TCR alpha beta-bearing T cells, do not mediate xenogeneic skin graft rejection in mouse/rat combinations.     

Massive production of Th2 cytokines by human CD4+ effector T cells transiently expressing the natural killer cell marker CD57/HNK1.

We have reported previously that uncommitted human CD4+ CD45RO- T cells default to the T-helper type 1 (Th1) pathway, if they are costimulated by anti-CD3 plus anti-CD28 monoclonal antibodies (mAb). In contrast, 5% of the uncommitted T cells differentiate into Th2 cells, if they are stimulated by anti-CD28 plus interleukin-2 (IL-2) in the absence of T-cell receptor (TCR) signals. The anti-CD28/IL-2-induced proliferation (and the resulting Th2 commitment) was not affected by neutralizing anti-IL-4 mAb, suggesting a non-conventional IL-4-independent Th2 differentiation pathway. Here we report that the respective CD4+ Th2 cells (but not the Th1 cells) coexpressed the natural killer (NK) cell marker HNK1/CD57. Expression of CD57 on Th2 cells required CD28 stimulation, and was suppressed by CD3/TCR signals. However, Th2 effector cells displayed a TCR V beta-chain usage comparable to that of committed Th1 cells (with V beta 8 dominating). Our data suggest that expression of CD57 on human CD4 T cells may be associated with defined stages of Th2 cell activation/differentiation, and may not necessarily characterize a separate T-cell lineage. The induction of cytokine production and B-cell helper function in both Th1 and Th2 populations required CD3/TCR signalling in costimulation with anti-CD28 or IL-2. Importantly, anti-CD28/IL-2-primed Th2 cells readily secreted IL-4 and induced IgE production by surface IgE- B cells in response to the first TCR signal and independent of previous contact with IL-4. Therefore, CD4+ CD57+ T cells responded comparably to murine CD4+ NK1.1+ T cells, which are critical for the development of Th2/IgE immune responses in vivo. The possible role of human CD4+ CD57/HNK1+ Th2-like cells in cancer, infection and allergy is discussed.     

Signalling through NK1.1 triggers NK cells to die but induces NK T cells to produce interleukin-4.

In vivo inoculation of specific antibody is an accepted protocol for elimination of specific cell populations. Except for anti-CD3 and anti-CD4, it is not known if the depleted cells are eliminated by signalling through the target molecule or through a more non-specific mechanism. C57BL/6 mice were inoculated with anti-natural killer (NK1.1) monoclonal antibody (mAb). Thereafter spleen cells were harvested, stained for both surface and intracellular markers, and analysed by flow cytometry. As early as 2 hr post inoculation, NK cells were signalled to become apoptotic while signalling through the NK1.1 molecule activated NK1.1+ T-cell receptor (TCR)+ (NK T) cells to increase in number, and produce interleukin-4 (IL-4). Anti NK1.1 mAb was less efficient at signalling apoptosis in NK cells when NK T-cell deficient [beta 2-microglobulin beta 2m-deficient] mice were used compared with wild type mice. Efficient apoptotic signalling was restored when beta 2m-deficient mice were reconstituted with NK T cells. NK-specific antibody best signals the apoptotic process in susceptible NK cells when resistant NK T cells are present, activated, and secrete IL-4.     

Cytotoxic activity against tumour cells mediated by intermediate TCR cells in the liver and spleen.

Morphological and phenotypic characterization in previous studies has indicated that intermediate (int) T-cell receptor (TCR) cells or T natural killer (TNK) cells may stand at an intermediate position between NK cells and high TCR cells of thymic origin in phylogenetic development. In this study, a functional study on cytotoxic activity against various tumour targets was performed in each purified subset. When a negative selection method entailing in vivo injection of anti-asialo GM, antibody or anti-interleukin (IL)-2R beta monoclonal antibody (mAb) was applied, IL-2R beta 1 CD3 NK cells were found to have the highest NK activity while IL-2R beta 1 int CD3 (or TCR) cells had a lower level of the NK activity. High CD3 cells (freshly isolated) did not have any such activity. Sorting experiments further revealed that the NK function mediated by int CD3 cells was augmented when they were exposed to anti-CD3 mAb. anti-TCR alpha beta, or anti-TCR-delta mAb. This phenomenon was not observed in NK cells and high CD3 cells. More importantly, when anti-CD3 mAb (or anti-TCR mAb) was added to the assay culture, int CD3 cells became cytotoxic against even NK-resistant tumour (Fc gamma R-. Fas+) targets. Liver mononuclear cells or int CD3 cells exposed to anti-CD3 mAb for 6 hr showed an elevated level of perforin in their cytoplasms. The present results suggest that int CD3 cells are usually non-cytotoxic against various tumours but become functional after being stimulated via the TCR CD3 complex.     

IL-4-producing NK T cells are biased towards IFN-γ production by IL-12. Influence of the microenvironment on the functional capacities of NK T cells

NK T cells are an unusual T lymphocyte subset capable of promptly producing several cytokines after stimulation, in particular IL-4, thus suggesting their influence in Th2 lineage commitment. In this study we demonstrate that, according to the cytokines present in the micro environment, NK T lymphocytes can preferentially produce either IL-4 or IFN-γ. In agreement with our previous reports showing that their IL-4-producing capacity is strikingly dependent on IL-7, CD4⁻ CD8⁻ TCRα β⁺ NK T lymphocytes, obtained after expansion with IL-1 plus granulocyte-macrophage colony-stimulating factor, produced almost undetectable amounts of IL-4 or IFN-γ in response to TCR/CD3 cross-linking. However, the capacity of these T cells to produce IFN-γ is strikingly enhanced when IL-12 is added either during their expansion or the anti-CD3 stimulation, while IL-4 secretion is always absent. A similar effect of IL-12 on IFN-γ production was observed when NK T lymphocytes were obtained after expansion with IL-7. It is noteworthy that whatever cytokines are used for their expansion, IL-12 stimulation, in the absence of TCR/CD3 cross-linking, promotes consistent IFN-γ secretion by NK T cells without detectable IL-4 production. Experiments in vivo demonstrated a significant up-regulation of the capacity of NK T cells to produce IFN-γ after anti-CD3 mAb injection when mice were previously treated with IL-12. In conclusion, we provide evidence that the functional capacities of NK T cells, which ultimately will determine their physiological roles, are strikingly dependent on the cytokines present in their microenvironment.     

In vivo and in vitro death of mature T cells induced by separate signals to CD4 and alpha beta TCR.

To investigate whether a clonal deletion mechanism is responsible for the mature T cell tolerance that may be induced in vivo by TCR signal to anti-CD4 (H129.19 mAb) coated cells, we analyzed the T cell repertoire in anti-CD4 mAb treated BALB/c mice by flow cytometry following TCR signals through anti-alpha beta TCR mAb or SEB superantigen. Lymph nodes showed a strong reduction in the CD4+/CD8+ cell ratio, and a selective clonal loss of CD4+ V beta 8+ cells 4d following anti-alpha beta TCR or SEB injection, respectively. Following lymph node cell activation in a short-term in vitro assay with SEB or anti-V beta 8 mAb, a selective elimination of CD4+ V beta 8+ cells was again detected, and DNA fragmentation analysis disclosed a cell death by apoptosis. These findings suggest that TCR triggering transduces an apoptotic signal into CD4+ mAb saturated cells that in turn leads to specific holes in the mature T cell repertoire.     

In vivo administration of IL-18 can induce IgE production through Th2 cytokine induction and up-regulation of CD40 ligand (CD154) expression on CD4+ T cells

IL-18 is considered to be a strong cofactor for CD4+ T helper 1 (Th1) cell induction. We have recently reported that IL-18 can induce IL-13 production in both NK cells and T cells in synergy with IL-2 but not IL-12, suggesting IL-18 can induce Th1 and Th2 cytokines when accompanied by the appropriate first signals for T cells. We have now found that IL-18 can act as a cofactor to induce IL-4, IL-10 and IL-13 as well as IFN-gamma production in T cells in the presence of anti-CD3 monoclonal antibodies (mAb). IL-18 can rapidly induce CD40 ligand (CD154) mRNA and surface expression on CD4+ but not CD8+ T cells. The administration of IL-18 alone in vivo significantly increased serum IgE levels in C57BL/6 (B6) and B6 IL-4 knockout mice. Furthermore, the administration of IL-18 plus IL-2 induced approximately 70-fold and 10-fold higher serum levels of IgE and IgG1 than seen in control B6 mice, respectively. IgE and IgG1 induction in B6 mice by administration of IL-18 plus IL-2 was eliminated by the pretreatment of mice with anti-CD4 or anti-CD154, but not anti-CD8 or anti-NK1.1 mAb. These results suggest that IL-18 can induce Th2 cytokines and CD154 expression, and can contribute to CD4+ T cell-dependent, IL-4-independent IgE production.     

Human CD3+4-8-WT31- T lymphocyte populations expressing the putative T cell receptor gamma-gene product. A limiting dilution and clonal analysis

The small peripheral blood CD3+ T cell population lacking both CD4 and CD8 surface antigens has been analyzed in the present study. Enriched CD3+4-8- populations were obtained by depletion with anti-CD4 or anti-CD8 monoclonal antibodies (mAb) and complement. The resulting populations contained greater than 99% CD2+ cells, whereas CD3+ represented approximately 50%. Virtually all of the cells were CD4-8- and did not react with the WT31 mAb, specific for a framework determinant of the alpha/beta T cell receptor (TCR). In order to analyze the molecular nature of CD3-associated molecules in CD3+WT31- populations, cells were stimulated with 0.5% phytohemagglutinin (PHA) for 24 h and expanded for an additional 7-14 days in interleukin 2 (IL 2). The resulting cells were greater than 95% CD3+ and expressed neither CD4/CD8 nor WT31 antigen. Cell surface iodination followed by cross-linking and immunoprecipitation with anti-CD3 mAb showed that CD3-associated molecules consisted of a major 45-kDa band and a minor band of 43 kDa. Thus, whereas CD3-associated molecules isolated from polyclonal CD3+WT31+ populations (expanded in IL 2 under the same culture conditions) appeared as diffuse bands, CD3-associated molecules isolated from CD3+WT31- populations displayed a homogeneous molecular mass. Northern blot analysis revealed the presence of mRNA for the TCR gamma chain whereas the mRNA for the alpha chain was mostly represented by a truncated (1.2 kb) form. Also small amounts of a nonproductive mRNA for the beta chain were detected. Freshly isolated CD3+WT31--enriched populations proliferated in response to PHA and concanavalin A, moreover, IL 2 was detected in the culture supernatants after cell stimulation. By applying culture conditions which allow virtually all T cells to undergo clonal expansion, approximately 1/3 CD3+WT31- were clonogenic. In addition, the large majority of proliferating microcultures lysed the K562 cell line and about half the natural killer (NK)-resistant fresh melanoma target cells. A large number of clones derived from CD3+WT31- enriched populations by limiting dilution has been further analyzed. More than 95% of the clones were CD3+4-8-WT31-; 12/15 clones analyzed in more detail displayed NK activity and 6/15 lysed melanoma cells; in addition, all lysed P815 target cells in the presence of PHA, thus indicating that all the clonogenic CD3+WT31- cells have a cytolytic potential.     

Expression of perforin and cytolytic potential of human peripheral blood lymphocyte subpopulations.

To verify the physiological role of the pore-forming protein perforin in vivo, its expression in subpopulations of human peripheral blood lymphocytes was examined by immunocytochemical staining and their cytolytic potentials compared. In addition to NK cells and gamma delta T cells, which uniformly expressed abundant perforin in their cytoplasmic granules, only a small subpopulation of CD8+ alpha beta T cells contained perforin, namely the CD11b+ subset. However, in vitro activation with an anti-CD3 antibody and IL-2 induced perforin expression in approximately 50% of the CD8+CD11b- T cells and also in a small subset of CD4+ T cells. A distribution of perforin in CD8+ and CD4+ T cells, similar to in vitro activated T cells, was observed in fresh peripheral blood lymphocytes from infectious mononucleosis patients. In all instances, the expression of perforin correlated with the cytolytic potential of these subpopulations. The results strongly suggest that perforin plays a role in the manifestation of cytotoxic activity in vivo.     

Oral antigen induces antigen-specific activation of intraepithelial CD4+ lymphocytes but suppresses their activation in spleen

Intraepithelial lymphocytes (IELs) are considered to drive immune surveillance of the epithelial layer to the mucosa, which is initially exposed to exogenous antigens. However, how IELs are activated by orally administered antigens remains unclear. To clarify this mechanism, we fed ovalbumin (OVA) to T cell receptor transgenic (TCR-Tg) mice with OVA-specific MHC class II-restricted TCR and found that the cytotoxic activity of IELs was increased against both NK and LAK target cells, but notably reduced after depleting CD8 + IELs. Cytoplasmic staining showed that the production of IFN-gamma and IL-2 was increased in mice fed with OVA both in the supernatant of cultured IELs with immobilized anti-CD3 mAb and in fresh CD4+ IELs. In contrast, the cytotoxic activity against NK and LAK target cells and the production of IL-2 and IFN-gamma was decreased in splenic T cells from mice fed with OVA. However, when the splenic T cells from these mice were cultured with OVA and IL-2, IFN-gamma production recovered. The decreased response demonstrated the clonal anergy of T cells. Furthermore, tumor growth was enhanced in TCR-Tg mice carrying an OVA-transfected counterpart A20 B cell lymphoma (OVA-A20) and fed with OVA. These results indicate that the oral administration of soluble antigens can activate CD4+ IELs in an antigen-specific manner but induces hyporesponsiveness in the spleen. In addition, Th1-type cytokines produced by activated CD4+ IEL might provide a bystander effect on the cytotoxic activity of IELs.     

Development of CD8 alpha/beta + TCR alpha beta intestinal intraepithelial lymphocytes in athymic nu/nu mice and participation in regional immune responses.

On the basis of the CD8 coreceptor expression, T-cell receptor (TCR)alpha beta-bearing intestinal intraepithelial lymphocytes (i-IEL) segregate into two populations. The CD8 alpha alpha + TCR alpha beta i-IEL develop thymus independently, whereas the CD8 alpha beta + TCR alpha beta i-IEL are generally considered to be thymus dependent. Flow cytometry analysis revealed a distinct population of CD8 alpha beta + TCR alpha beta i-IEL in individual athymic nu/nu mice. The i-IEL encompassing CD8 alpha beta + TCR alpha beta cells expressed potent cytolytic and interferon-gamma-producing activities. These findings demonstrate that CD8 alpha beta + TCR alpha beta i-IEL can develop in nu/nu mice independently from a functional thymus and suggest that these cells, directly or indirectly, perform biological functions in the gut.     

Human T cell receptor-mediated recognition of HLA-E

The HLA-E class Ib molecule presents hydrophobic peptides derived from the leader sequences of other class I molecules, constituting the ligands for CD94/NKG2 lectin-like receptors. Along the course of our studies on human CD94+ T cells, we characterized an alpha beta CD8+CD94/NKG2C+ CTL clone (K14). In cytolytic assays against the murine TAP-deficient RMA-S cells transfected with human beta2 microglobulin and HLA-E (RMA-S/HLA-E), loaded with different synthetic peptides, K14 displayed a pattern of specific recognition distinct to that observed in CD94/NKG2C+ NK clones tested in parallel. RMA-S/HLA-E cells loaded with some but not all HLA class I leader sequence peptides were efficiently recognized by K14 but not by CD94/NKG2C clones, andvice versa. Remarkably, K14 also reacted with HLA-E loaded with a peptide derived from the BZLF-1 Epstein-Barr virus protein. Anti-CD94 mAb did not prevent K14 cytotoxicity against RMA-S/HLA-E cells, whereas incubation with anti-clonotypic mAb specific for the K14 TCR markedly inhibited lysis. Soluble HLA-E tetramers refolded with different peptides (i.e. VMAPRTVLL, VMAPRTLIL, VMAPRTLFL) specifically stained K14 cells. HLA-E tetramer binding was minimally reduced by pretreatment with anti-CD94 mAb alone, but was completely prevented in combination with anti-clonotypic mAb. Altogether, the data unequivocally imply the generation of human T cells potentially recognizing through the alpha beta TCR HLA-E molecules that bind to class I- and virus-derived peptides.     

Occurrence of interleukin-5 production by CD4- CD8- (double-negative) T cells in lungs of both normal and congenitally athymic nude mice infected with Toxocara canis.

We studied cells in the lungs of BALB/c and BALB/c-nu/nu (nude) mice infected with Toxocara canis, which produced interleukin-5 (IL-5) in in vitro culture with larval excretory-secretory antigen (ESAg). The proportion of CD4+/CD8+/CD4- CD8- cells in lungs of both BALB/c and nude mice was unchanged before and after infection with T. canis. Panning and complement-mediated lysis using monoclonal antibody (mAb) to CD4 showed that CD4+ cells in the lung from both mice produced IL-5. Anti-CD4 mAb suppressed ESAg-stimulated IL-5 production in vitro. In vitro depletion or inhibition of CD8+ cells reduced IL-5 production significantly in some cases, suggesting involvement with IL-5 production. Anti-CD3 mAb enhanced IL-5 production when incubated with or without ESAg. Production of IL-5 was reduced by in vivo depletion of CD4+ cells only and both CD4+ and CD8+ T cells, by intraperitoneal injection with appropriate mAb; IL-5 production was stimulated by anti-CD3 mAb. In contrast, IL-5 production by lung cells of BALB/c mice decreased by more than 90% after simultaneous injection with anti-CD4, anti-CD8 and anti-CD3 mAb, and was not enhanced by anti-CD3 mAb. Similar results were obtained in nude mice. These results suggest that CD4- CD8- T cells, as well as CD4+ T cells, produce IL-5.     

IL-4 is able to reverse the CD2-mediated negative apoptotic signal to CD4-CD8- alpha beta and/or gamma delta T lymphocytes.

Activation of immature thymocytes or transformed T lymphocytes via T-cell receptor (TCR)/CD3 signalling can induce programmed cell death (apoptosis). Recent data indicate that anti-CD3/TCR monoclonal antibodies (mAb) also trigger apoptosis in activated (but not resting) mature peripheral blood T lymphocytes. Here we report that triggering of resting CD4-CD8-TCR alpha beta+ and/or TCR gamma delta+ via the alternative CD2-dependent activation pathway is able to induce programmed cell death. A pair of mitogenic anti-CD2 mAb provoked a dramatic rise in [Ca2+]i that was almost entirely sustained by extracellular fluxes, and the inhibition of membrane [Ca2+/Mg2+] ATPase. The resulting endonuclease activation was able to induce DNA fragmentation, as revealed by propidium iodide staining and gel electrophoresis. Induction of apoptosis was prevented by the presence of interleukin-4 (IL-4) as well as by endonuclease inactivation with 100 microM ZnCl2, but enhanced by the contemporary block of protein kinase C. Thus it seems that in resting T lymphocytes the strong calcium signal delivered by the alternative CD2 activation pathway may act as a negative apoptotic signal in both alpha beta and gamma delta T cells with low (non-major histocompatibility complex restricted) antigenic affinity, so limiting the extension of polyclonal T-cell growth.     

A lymphocyte differentiation and activation antigen, CZ-1, that distinguishes between CD8+ and unstimulated CD4+ T lymphocytes.

We report the generation and cellular reactivity of a novel rat IgM monoclonal antibody (mAb), CZ-1, made against mouse natural killer (NK) cells activated in vivo. mAb CZ-1 recognizes a molecule whose properties are consistent with that of a trypsin-sensitive, non-phosphatidyl inositol-linked sialoglycoprotein. The expression of the antigen recognized by mAb CZ-1 is restricted mostly to cells of the lymphoid lineage. The antigen is expressed on 10%-25% of bone marrow cells and 3%-5% of thymocytes. Analysis of thymocyte subpopulations indicates expression of the CZ-1 antigen on 100% of the NK1.1+, 27% of the CD4-CD8-, 1.1% of the CD4+CD8+, 1.1% of the CD4+CD8-, and 33% of the CD4-CD8+ cells. In the spleen, the CZ-1 antigen is expressed on B lymphocytes, NK cells, and virtually all CD8+ T lymphocytes. Most unstimulated CD4+ splenic T lymphocytes, monocytes and polymorphonuclear cells, with the notable exception of basophils, do not react with mAb CZ-1. CD4+ T cells activated in vivo by virus infection or in vitro by anti-CD3 and interleukin-2 express the CZ-1 antigen. These results indicate that mAb CZ-1 identifies a novel inducible lymphocyte activation/differentiation antigen that distinguishes between thymic and unstimulated splenic CD4+ and CD8+ T lymphocytes. This mAb will be a useful tool in the identification of lymphocyte subpopulations and in the study of the ontogeny and activation of these cells.     

Prolongation of allograft survival by administration of mAb specific for the three subunits of IL-2 receptor

The IL-2 receptor (IL-2R) gamma chain, the so-called common gamma (gamma(c)) chain, which is shared with multiple cytokine receptors, plays important roles in the immune system. Here we assessed the immunosuppressive ability of mAb specific for the gamma(c) chain in induction of cytotoxic T lymphocytes (CTL) and allograft rejection in combination with mAb specific for the alpha and beta chains of IL-2R. CBA/N (H-2k) mice were injected i.p. with allogeneic splenocytes from BALB/c (H-2d) mice, and then administered with combinations of anti-IL- 2R alpha, anti-IL-2R beta and anti-gamma(c) mAb or a control mAb. Addition of anti-gamma(c) mAb together with anti-IL-2R alpha and anti- IL-2R beta mAb induced a complete inhibition of CTL response. The numbers and populations of CD4+ CD8- and CD4- CD8+ T cells were not significantly affected by administration of the three anti-IL-2R mAb, whereas NK cells were completely depleted in spleens of mice treated with the anti-IL-2R mAb. Furthermore, skin allograft survival was also significantly prolonged by administration of the three anti-IL-2R mAb. These results suggest that the anti-gamma(c) mAb in combination with anti-IL-2R alpha and anti-IL-2R beta mAb is capable of suppressing induction of CTL and NK cells, resulting in prolongation of skin allograft survival.