Analysis of Ly49 gene transcripts in mature NK cells supports a role for the Pro1 element in gene activation, not gene expression.

The variegated expression of murine Ly49 loci has been associated with the probabilistic behavior of an upstream promoter active in immature cells, the Pro1 element. However, recent data suggest that Pro1 may be active in mature natural killer (NK) cells and function as an enhancer element. To assess directly if Pro1 transcripts are present in mature Ly49-expressing NK cells, RNA-sequencing of the total transcript pool was performed on freshly isolated splenic NK cells sorted for expression of either Ly49G or Ly49I. No Pro1 transcripts were detected from the Ly49a, Ly49c or Ly49i genes in mature Ly49(+) NK cells that contained high levels of Pro2 transcripts. Low levels of Ly49g Pro1 transcripts were found in both Ly49G(+) and Ly49G(-) populations, consistent with the presence of a small population of mature NK cells undergoing Ly49g gene activation, as previously demonstrated by culture of splenic NK cells in interleukin-2. Ly49 gene reporter constructs containing Pro1 failed to show any enhancer activity of Pro1 on Pro2 in a mature Ly49-expressing cell line. Taken together, the results are consistent with Pro1 transcription having a role in gene activation in developing NK, and argue against a role for Pro1 in Ly49 gene transcription by mature NK cells.

Interaction of a dengue virus NS1-derived peptide with the inhibitory receptor KIR3DL1 on natural killer cells.

Killer immunoglobulin-like receptors (KIRs) interact with human leucocyte antigen (HLA) class I ligands and play a key role in the regulation and activation of NK cells. The functional importance of KIR-HLA interactions has been demonstrated for a number of chronic viral infections, but to date only a few studies have been performed in the context of acute self-limited viral infections. During our investigation of CD8(+) T cell responses to a conserved HLA-B57-restricted epitope derived from dengue virus (DENV) non-structural protein-1 (NS1), we observed substantial binding of the tetrameric complex to non-T/non-B lymphocytes in peripheral blood mononuclear cells (PBMC) from a long-standing clinical cohort in Thailand. We confirmed binding of the NS1 tetramer to CD56(dim) NK cells, which are known to express KIRs. Using depletion studies and KIR-transfected cell lines, we demonstrated further that the NS1 tetramer bound the inhibitory receptor KIR3DL1. Phenotypical analysis of PBMC from HLA-B57(+) subjects with acute DENV infection revealed marked activation of NS1 tetramer-binding natural killer (NK) cells around the time of defervescence in subjects with severe dengue disease. Collectively, our findings indicate that subsets of NK cells are activated relatively late in the course of acute DENV illness and reveal a possible role for specific KIR-HLA interactions in the modulation of disease outcomes.

Identification of RANTES receptors on human monocytic cells: competition for binding and desensitization by homologous chemotactic cytokines.

RANTES (regulated on activation, normal T expressed and secreted) is a member of the chemotactic cytokine (chemokine) beta subfamily. High affinity receptors for RANTES have been identified on a human monocytic leukemia cell line THP-1, which responded to RANTES in chemotaxis and calcium mobilization assays. Steady-state binding data analyses revealed approximately 700 binding sites/cell on THP-1 cells with a Kd value of 400 pM, comparable to that expressed on human peripheral blood monocytes. The RANTES binding to monocytic cells was competed for by monocyte chemotactic and activating factor (MCAF) and macrophage inflammatory protein 1 (MIP-1) alpha, two other chemokine beta cytokines. Although MCAF and MIP-1 alpha competed for RANTES binding to monocytes with apparent lower affinity (with estimated Kd of 6 and 1.6, nM respectively) both of these cytokines effectively desensitized the calcium mobilization induced by RANTES. The chemotactic response of THP-1 cells to RANTES was also markedly inhibited by preincubation with MCAF or MIP-1 alpha. In contrast, RANTES did not desensitize the THP-1 calcium mobilization and chemotaxis in response to MCAF or MIP-1 alpha. These results, together with our previous observations that RANTES did not compete for MCAF or MIP-1 alpha binding on monocytic cells, indicate the expression of promiscuous receptors on monocytes that recognize one or more cytokines within the chemokine beta family.

Tec family kinases modulate thresholds for thymocyte development and selection.

Tec family kinases are implicated in T cell receptor (TCR) signaling, and combined mutation of inducible T cell kinase (Itk) and resting lymphocyte kinase (Rlk)/Txk in mice dramatically impairs mature T cell function. Nonetheless, mutation of these kinases still permits T cell development. While itk(-)(/)- mice exhibit mild reductions in T cells with decreased CD4/CD8 cell ratios, rlk(-)(/)-itk(-)(/)- mice have improved total T cell numbers yet maintain decreased CD4/CD8 ratios. Using TCR transgenics and an in vitro thymocyte deletion model, we demonstrate that mutation of Tec kinases causes graded defects in thymocyte selection, leading to a switch from negative to positive selection in rlk(-)(/)-itk(-)(/)- animals. The reduction in both positive and negative selection and decreased CD4/CD8 ratios correlates with decreased biochemical parameters of TCR signaling, specifically defects in capacitive Ca(2+) influx and activation of the mitogen-activated kinases extracellular signal-regulated kinase 1 and 2. Thus, Tec kinases influence cell fate determination by modulating TCR signaling, leading to altered thresholds for thymocyte selection. These results provide support for a quantitative model for thymic development and provide evidence that defects in negative selection can substantially alter thymic cellularity.

Interleukin 12 (IL-12) induces tyrosine phosphorylation of JAK2 and TYK2: differential use of Janus family tyrosine kinases by IL-2 and IL-12.

Interleukin (IL-12) has many effects on the function of natural killer and T cells, and is important in the control of cell-mediated immunity. IL-2 and IL-12 display many similar activities, yet each also induces a distinct set of responses. A human IL-12 receptor subunit has recently been cloned and, like the IL-2R beta and IL-2R gamma, is a member of the hematopoietic receptor superfamily; however, the molecular mechanisms of IL-12 action are unknown. In this report we show that IL-12 and IL-2 induce tyrosine phosphorylation of distinct members of the Janus (JAK) family of protein tyrosine kinases in human T lymphocytes. IL-12, but not IL-2, stimulates the tyrosine phosphorylation of TYK2 and JAK2, whereas JAK1 and JAK3, which are phosphorylated in response to IL-2, are not phosphorylated after IL-12 treatment. The use of distinct but related JAK family tyrosine kinases by IL-12 and IL-2 may provide a biochemical basis for their different biological activities.

Regulation of JAK3 expression in human monocytes: phosphorylation in response to interleukins 2, 4, and 7.

The Janus family of kinases (JAKs) has been shown to be involved in the signal transduction of a number of cytokine receptors. Recently, we have cloned a novel JAK family member, JAK3, that is expressed in natural killer and activated T cells and is coupled functionally and physically to the interleukin 2 (IL-2) receptor in these cells. Here we report that JAK3 was expressed at low but detectable levels in human monocytes. In contrast, JAK3 expression was strongly induced during activation by interferon gamma (IFN-gamma) or lipopolysaccharide. Moreover, JAK3 became tyrosine phosphorylated in response to IL-2, IL-4, and IL-7 but not response to IFN-gamma or granulocyte/macrophage colony-stimulating factor. Together, these findings suggest that JAK3 is functionally important in activated monocytes and cells of the myeloid lineage and is involved in signaling responses of cytokines that use the common gamma-chain of the IL-2 receptor.

IL-4 and IL-13 induce Lsk, a Csk-like tyrosine kinase, in human monocytes.

Lsk is a protein tyrosine kinase with homology to the COOH-terminal Src kinase (Csk). Unlike Csk that is ubiquitously expressed, Lsk has limited tissue distribution. Here we have examined the expression and regulation of Lsk and Csk in peripheral human monocytes. We have found that Lsk mRNA and protein were not expressed in resting monocytes but were induced by treatment with interleukin 4 (IL-4) or IL-13 but not by interferon gamma (IFN-gamma) or IL-2. In fact, IFN-gamma, but not IL-2, efficiently blocked Lsk induction by IL-4 or IL-13. In contrast, Csk was constitutively present in human monocytes and was upregulated by IFN-gamma but not by IL-4 or IL-13. These results suggest that despite their structural similarities, Lsk and Csk may play distinct regulatory roles in monocyte functions elicited by cytokines, with Lsk functioning specifically within the context of a Th2-type immune response.

Alterations in signal transduction molecules in T lymphocytes from tumor-bearing mice.

Impaired immune responses occur frequently in cancer patients or in tumor-bearing mice, but the mechanisms of the tumor-induced immune defects remain poorly understood. In an in vivo murine colon carcinoma model (MCA-38), animals bearing a tumor longer than 26 days develop CD8+ T cells with impaired cytotoxic function, decreased expression of the tumor necrosis factor-alpha and granzyme B genes, and decreased ability to mediate an antitumor response in vivo. T lymphocytes from tumor-bearing mice expressed T cell antigen receptors that contained low amounts of CD3 gamma and completely lacked CD3 zeta, which was replaced by the Fc epsilon gamma-chain. Expression of the tyrosine kinases p56lck and p59fyn was also reduced. These changes could be the basis of immune defects in tumor-bearing hosts.

Interaction of IL-2R beta and gamma c chains with Jak1 and Jak3: implications for XSCID and XCID.

Interleukin-2 (IL-2) signaling requires the dimerization of the IL-2 receptor beta.(IL-2R beta) and common gamma (gamma c) chains. Mutations of gamma c can result in X-linked severe combined immunodeficiency (XSCID). IL-2, IL-4, IL-7 (whose receptors are known to contain gamma c), and IL-9 (whose receptor is shown here to contain gamma c) induced the tyrosine phosphorylation and activation of the Janus family tyrosine kinases Jak1 and Jak3. Jak1 and Jak3 associated with IL-2R beta and gamma c, respectively; IL-2 induced Jak3-IL-2R beta and increased Jak3-gamma c associations. Truncations of gamma c, and a gamma c, point mutation causing moderate X-linked combined immunodeficiency (XCID), decreased gamma c-Jak3 association. Thus, gamma c mutations in at least some XSCID and XCID patients prevent normal Jak3 activation, suggesting that mutations of Jak3 may result in an XSCID-like phenotype.

Molecular cloning of lsk, a carboxyl-terminal src kinase (csk) related gene, expressed in leukocytes.

Regulation of the activity of src-family kinases is thought to occur, in part, through the phosphorylation of conserved carboxyl-terminal tyrosine residues. Although the src-family includes several molecules with tissue or cell-type restricted expression, the only kinase implicated in the regulatory phosphorylation of these enzymes is p50csk. Herein we report the molecular cloning of a tissue specific p50csk-related gene. Like p50csk, the deduced protein sequence of this novel cDNA includes a tyrosine kinase catalytic domain, SH2 and SH3 domains, a short amino terminus, and no autophosphorylation or carboxyl-terminal tyrosine residues. Additionally, neither this novel kinase nor p50csk contain the amino-terminal myristoylation site characteristic of the src-family. However, whereas csk is ubiquitously expressed, mRNA corresponding to this novel gene is expressed in brain, natural killer (NK) cells, and activated T cells but not in a variety of other tissues and cell lines. In agreement with the mRNA expression pattern, antiserum reactive with the predicted carboxyl-terminus of the cDNA recognizes a 57 kDa polypeptide in immunoblots of NK cells and PHA-activated T cells. Because of its limited expression and high homology to p50csk, we named this gene lsk; leukocyte carboxyl-terminal src kinase related gene. Identification of a molecule like lsk suggests the existence of tissue specific src-regulatory pathways that function in activated lymphocytes.

Intracellular signaling by the killer immunoglobulin-like receptors and Ly49.

Once thought to be promiscuous killers, it is now known that natural killer (NK) cells possess an elaborate array of receptors that regulate NK cytotoxic and secretory functions upon interaction with target cell MHC class I proteins. These receptors, known as killer cell immunoglobulin-like receptors (KIRs) in humans, and Ly49 receptors in the mouse, have become the focus of intense study in an effort to discern the underlying biology of these large receptor families. These receptor families include both inhibitory and activating receptors. Interrogation of a target expressing KIR ligands leads to coengagement of the inhibitory receptor with as-yet poorly defined activation receptors. Kinases activated during engagement mediate the phosphorylation of the KIR or Ly49 cytoplasmic immunoreceptor tyrosine-based inhibitory motifs (ITIMs). The phosphorylated ITIMs serve as efficient recruitment points for the cytosolic protein tyrosine phosphatases, SHP-1 and SHP-2, resulting in the dephosphorylation of substrates critical for cellular activation. In contrast, some KIRs and Ly49s lack the ITIM and possess a charged residue in their transmembrane domains that mediates interaction with the DAP12 signal transduction chain. DAP12 uses its cytoplasmic immunoreceptor tyrosine-based activation motif (ITAM) to mediate cellular activation. Engagement of a DAP12 coupled KIR or Ly49 results in phosphorylation of DAP12, and other key substrates, including the Syk tryosine kinase, phospholipase C, and c-Cbl. DAP12 activation then leads to the Mapk cascade and ultimately to enhanced degranulation, and production of cytokines and chemokines. Although the context in which inhibitory and activating KIR and Ly49s function is not yet known, the dissection of the activating and inhibitory signal transduction pathways should shed light on their method of integration into the activation sequela of NK cells. Ultimately, this work will lead to concrete understanding of the immunobiology of these seemingly antagonistic receptor systems.

Regulation of NK cells through the 80-kDa TNFR (CD120b).

By using monoclonal antibody specific for tumor necrosis factor receptor80 (TNFR80) (CD120b) and TNFR60 (CD120a), we determined which receptor transduces the signals involved in activating natural killer (NK) cells. Purified CD56+CD3- large lymphocytes express TNFR80 but not TNFR60 and interleukin-2 (IL-2) up-regulates TNFR80 expression, consistent with NK cells being activated in vivo. Treatment of NK cells with anti-TNFR80 for 18 h enhanced the NK activity detected on K562 target cells mimicking the effect of TNF. In combination with IL-2, TNF enhanced the development of lymphokine-activated killing. However, only anti-TNFR80 abrogated IL-2 induction of lymphokine-activated killer cell activity. The activity of TNF or anti-TNFR80 was selective for NK cytotoxic function because they did not directly mimic IL-2 activation or induce significant proliferation, expression of cell surface activation antigens (CD25 or HLA-DR), or interferon-gamma secretion. These results indicate that TNFR80 is an important signal transducing receptor for the differentiation of NK cells induced by TNF and IL-2.

T cell activation via the disialoganglioside GD3: analysis of signal transduction.

The monoclonal antibody (mAb) R24 is a murine immunoglobulin G3 (IgG3) that reacts with the GD3 disialoganglioside present on melanoma cells as well as a subset of T cells. R24 mAb has induced antitumor responses both alone and in combination with interleukin-2 (IL-2) in clinical trials. We have reported T cell activation via GD3 as measured by the induction of tyrosine phosphorylation. In this study a more detailed analysis of signal transduction after ligation of GD3 was performed in an attempt to understand the mechanism of in vivo therapeutic benefits observed. Analysis of subsequent events indicated that GD3 engagement resulted in phospholipase C(gamma) phosphorylation and calcium flux. When ras-associated events were examined, GD3 signaling resulted in ras activation as determined by GDP/GTP conversion as well as dose-and time-dependent IP3 activation. In addition, the majority of the IP3 activation by GD3 was inhibited by herbimycin A pretreatment. Elucidation of the nature and potential role of this moiety in GD3 signal transduction should be useful. Collectively, these data suggest a novel mechanism of T cell activation via a single, non-protein, surface moiety. This novel form of T cell-mediated activation may permit the delivery and local activation of effector cells at the tumor resulting in site-specific activation of the immune system.

Induction of DAP12 phosphorylation, calcium mobilization, and cytokine secretion by Ly49H.

The ability of several Ly49 family members to inhibit natural killer (NK) cell functions through recruitment of SHP-1 phosphatase has been reported. In contrast, the mechanisms underlying the activating signal generated by Ly49D are poorly understood. A homodimeric phosphoprotein (pp16) that physically and functionally associates with Ly49D has been described. In this study, a rabbit anti-mouse pp16 antiserum was generated and used to demonstrate that pp16 corresponds to the recently described DAP12 molecule. In addition, we show that a second Ly49 family member that lacks an immunoreceptor tyrosine-based inhibitory motif and contains a charged residue in the transmembrane domain, Ly49H, also associates with DAP12. Furthermore, we show that engagement of the Ly49H/DAP12 complex results in phosphorylation of DAP12, intracellular calcium mobilization, and tumor necrosis factor secretion in transfected cells. These results thus provide evidence that Ly49H is an activating receptor that associates with DAP12, previously described as a pp16 component of the Ly49D receptor complex.

Cloning and expression of a second human natural killer cell granule tryptase, HNK-Tryp-2/granzyme 3.

Cytotoxic lymphocytes possess a number of serine proteases (granzymes) usually localized in cytoplasmic granules. To date, the DNA sequences of four human granzymes have been reported. A fifth human granzyme (granzyme 3) has been biochemically purified and its N-terminal amino acid sequence has been reported. This enzyme was described as possessing tryptase activity, cleaving synthetic substrates after arginine or lysine. We recently cloned a rat granzyme tryptase (RNK-Tryp-2), and used this cDNA to screen human cDNA libraries. Isolation of cDNA fragments of a human gene could be overlapped to provide a complete cDNA sequence, which we designated HNK-Tryp-2. The N-terminal amino acid sequence deduced from HNK-Tryp-2 was identical to that reported for granzyme 3. This gene appears to be a single copy gene that is expressed in isolated natural killer cells and T cells as well as in tissues containing these cells.

Intralesional infusion of lymphokine-activated killer (LAK) cells and recombinant interleukin-2 (rIL-2) for the treatment of patients with malignant brain tumor.

Twenty patients with supratentorial, intracerebral lesions defined by computed tomographic scan or magnetic resonance imaging were treated by surgery and adoptive immunotherapy with lymphokine-activated killer (LAK) cells and recombinant Interleukin-2 (rIL-2, Cetus). Seventeen patients had glioblastoma, two had high-grade oligodendroglioma, and one patient had two metastatic sarcoma lesions. LAK cells were produced from blood mononuclear cells (MNC) obtained by 2 to 3 leukapheresis procedures and cultured (2.5 x 10(6) MNC/ml) 3 to 5 days with 1000 units rIL-2/ml. Although LAK cells could be produced from MNC of all patients, those taking steroids or with a low Karnofsky functional status generated, on average, suboptimal LAK cell activity. Age, sex, and serum anticonvulsant levels do not seem to influence a patient's ability to produce LAK cells in vitro. For therapy, cultured MNC (1-15 x 10(9] containing LAK cells were suspended in saline containing 10(6) units rIL-2 and injected into tissue surrounding the tumor cavity during craniotomy. For 3 days after their operations, patients received 10(6) units rIL-2 into the tumor cavity through an Ommaya reservoir. The treatment protocol was tolerated well by all patients, although they all experienced some degree of headache, fever, or lethargy that cleared within a few days of the last rIL-2 injection. When computed tomographic (CT) scans were obtained soon after treatment, areas of low density suggested a greater-than-normal extent of edema around the operative site. At the present time, CT scans indicate that the tumors of seven patients have recurred with an average disease-free interval of 25 +/- 6 weeks.(ABSTRACT TRUNCATED AT 250 WORDS)

In vitro analysis of the proliferative potential of T cells from patients with brain tumor: glioma-associated immunosuppression unrelated to intrinsic cellular defect.

Patients harboring a malignant brain tumor have been described as being highly immunosuppressed, as evidenced by reduced numbers of T cells and the decreased ability of their lymphocytes to produce interleukin-2 (IL-2). In order to determine whether an intrinsic abnormality exists in the T lymphocytes of glioma patients and to evaluate what role corticosteroids may play in glioma-associated immunosuppression, in vitro T cell proliferative function in the presence of recombinant IL-2 (rIL-2) was examined in age-matched groups of normal control subjects, steroid-free patients with glial tumors, steroid-dependent patients with glial tumors, and steroid-dependent patients with nonglial cerebral tumors. The results demonstrated that, when enriched T cell populations of all brain-tumor patients were stimulated with rIL-2 and phytohemagglutinin (PHA), there were no statistically significant differences between any groups. In contrast, when T cell populations were stimulated with mitogenic combinations of phorbol ester, calcium ionophore, and rIL-2, those from steroid-dependent patients with glial tumors had a significantly lower response than those from normal control subjects, suggesting that a population of T cells capable of responding to phorbol ester/ionomycin and not PHA stimulation is inhibited by corticosteroid therapy in glioma patients. In addition, T cells of four brain-tumor patient/age-matched control subject pairs were stimulated with either phorbol ester/ionomycin or PHA for 24 hours; three of the four patients expressed low-affinity IL-2 receptor levels as high or higher than their respective control subjects, suggesting that IL-2 receptor expression in these patients may be quantitatively normal once the T cell number is corrected. Taken together, these results show that the decreased PHA responsiveness that has been previously reported in lymphocytes of glioma patients is not due to a cellular abnormality within the potentially responsive cells, but rather reflects the reduced proportion of T cells within their peripheral blood which, as a consequence, reduces the level of IL-2 production attained upon activation.

Functional association of FcepsilonRIgamma with arginine(632) of paired immunoglobulin-like receptor (PIR)-A3 in murine macrophages.

Paired immunoglobulin-like receptors (PIR) are expressed on B cells and macrophages and include inhibitory and putative activating receptors referred to as PIR-B and PIR-A, respectively. Although PIR-B's inhibitory pathway has been described, it is unknown whether PIR-A receptors can deliver activation signals to macrophages, and if so, through what mechanism. Here we use chimeric receptors to address the mechanisms of PIR-A signaling. Cotransfection of chimeric receptors comprised of the extracellular region of human CD4 and the transmembrane and cytoplasmic domains of murine PIR-A3 showed the ability of PIR-A3 to physically interact with the FcepsilonRIgamma chain in 293T cells. This interaction is dependent on Arg(632) within the PIR-A3 transmembrane domain. We also demonstrate PIR-A3 interaction with the endogenous FcepsilonRIgamma of the ANA-1 macrophage cell line, again in an Arg(632)-dependent manner. Furthermore, we show that crosslinking of these chimeric receptors synergizes with IFN-gamma in the production of nitric oxide. Our data are the first to show the potential of PIR-A3 to deliver activation signals to macrophages and establish its dependence on Arg(632). These findings suggest that further study of the PIR-A receptors should be aggressively pursued toward a complete understanding of the intricate regulation of macrophage biology.

Paired inhibitory and activating receptor signals.

The immunological literature has become inundated with reports regarding paired inhibitory receptors. Paired inhibitory receptor systems are highly conserved families that contain receptors involved in either cellular inhibition or activation. In most cases the paired putative biochemical antagonists are co-expressed on a given cell and thought to bind similar, if not identical, ligands making their biological role difficult to understand. Examples of these systems include immunoglobulin (Ig)-like receptors (Killer Ig Receptors, Immunoglobulin-like Transcripts/Leukocyte Ig-like Receptors/Monocyte Macrophage Ig Receptors, and Paired Ig-like Receptors), and type II lectin-like receptor systems (NKG2 and Ly49). General characteristics of these inhibitory receptors include a cytoplasmic immunoreceptor tyrosine-based inhibitory motif (ITIM). The ITIM is phosphorylated upon engagement and recruits protein tyrosine phosphatases that dephosphorylate cellular substrates that would otherwise mediate activation. In contrast, the activating receptors of these pairs use charged residues within their transmembrane domains to associate with various signal transduction chains including the gamma chain of the receptor for the Fc portion of IgE, DAP12 or DAP10. Once phosphorylated, these chains direct the signal transduction cascade resulting in cellular activation. Here we review the signaling of several paired systems and present the current models for their signal transduction cascades.

Corticosteroids inhibit the delivery of short-term activational pulses of phorbol ester and calcium ionophore to human peripheral T cells.

Although there is evidence that corticosteroids inhibit receptor-ligand-induced phospholipid hydrolysis, the immunosuppressive effects of these agents downstream of protein kinase C (PK-C) activation and cytosolic Ca2+ mobilization is unclear. Previous studies indicated that T cell proliferative activation could be achieved with simultaneous short-term (e.g., 15-120 min) exposure to agents activating PK-C and elevating cytosolic Ca2+. In the studies reported here, similar procedures were utilized for determining whether corticosteroids alter T cell activation signals downstream of second messenger events. Dexamethasone interfered with T cell activation induced by short-term exposure to phorbol 12,13-dibutyrate (PDBu) and the calcium ionophore, ionomycin. The inhibitory effect was evident with as little as 15 min of exposure to dexamethasone and T cell activating agents, making mechanisms involving de novo protein synthesis unlikely. Dexamethasone's effects in this system were blocked by the steroid receptor antagonist RU-486, indicating that the inhibition was mediated through the glucocorticoid receptor. The inclusion of recombinant interleukin-2 (IL-2) only partially overcame the dexamethasone inhibitory effect. Long-term (i.e., 48 hr) direct stimulation of PK-C with either PDBu or the non-tumor-promoting PK-C activator, bryostatin 1, also substantially overcame dexamethasone's effects, resulting in a recovery of IL-2 production and significant restoration of the T-cell proliferative response. These observations suggest that treatment with a PK-C-activating agent such as bryostatin 1 could reduce glucocorticosteroid-induced immunosuppression.