Protein kinase C independent activation of inducible nitric oxide synthase by tumor necrosis factor-alpha in TM4 Sertoli cells

To investigate the nitric oxide (NO) production and its signalling mechanism in TM4 Sertoli cells, the cells were treated with recombinant tumor necrosis factor-alpha (rTNF-alpha), recombinant interleukin-1 alpha (rIL-1alpha), or lipopolysaccharide (LPS), either alone or in combination with recombinant interferon-gamma (rIFN-gamma), and NO production was measured by using the Griess method. TM4 Sertoli cells produced a small amount of NO upon treatment with rIFN-gamma. The effect of rIFN-gamma was drastically increased by cotreatment with rTNF-alpha in a dose-dependent manner. However, combination of rIL-1alpha or LPS with rIFN-gamma did not synergize to activate cells. RIFN-gamma in combination with rTNF-alpha showed marked increase of the expression of iNOS protein. Protein kinase C inhibitors did not inhibit the production of NO induced by rIFN-gamma plus rTNF-alpha. These results suggest that the role of TNF-alpha is to provide TM4 Sertoli cells with the active cofactor for NO production and TNF-alpha-induced signaling for induction of NO synthesis is not dependent on protein kinase C activation.     

Effects of biological response modifiers on lung natural killer activity.

Natural killer (NK) activity plays an important role in host defense against tumors, especially once augmented by immunomodulators. It is likely that the modulation of NK cells is a reflection of the environment in which they reside. The current study was undertaken to characterize the response profile of lung interstitial lymphocyte natural killer (LLNK) activity to various biological response modifiers (BRM) in vitro after short term incubation (18h). The presented data show that treatment of lung lymphocytes with human recombinant interleukin 2 (rIL-2), purified rat interferon alpha/beta (IFN-alpha/beta), or murine recombinant tumor necrosis factor alpha (rTNF-alpha) resulted in a dose-dependent increase in LLNK activity. The maximum stimulation was similar for rIL-2 and IFN-alpha/beta, although a much higher concentration of IFN-alpha/beta was required to reach this level of stimulation. The maximum response to rTNF-alpha treatment was about half that seen with rIL-2 or IFN-alpha/beta and it, too, required a high concentration. By contrast, rat recombinant interferon gamma (rIFN-gamma) or murine recombinant interleukin 1 (rIL-1) failed to alter LLNK activity significantly when used alone. Furthermore, doses of IFN-alpha/beta and rTNF-alpha that had little enhancing effect were able to synergize with a suboptimal dose of rIL-2, whereas rIL-1 and rIFN-gamma failed to do so. These data demonstrate the response of lung NK activity to BRM treatment, which is important for the responsible and effective use of BRM. However the spectrum of lung NK cell response to BRM is smaller than that previously reported for NK cells from other anatomic compartments.     

Human alveolar macrophage and blood monocyte interleukin-6 production

Interleukin-6 (IL-6) modulates a number of processes relevant to host immunity and inflammation. We investigated the capacity of the human alveolar macrophage to elaborate IL-6 in response to lipopolysaccharide (LPS), recombinant interleukin-1 (rIL-1), and recombinant tumor necrosis factor (rTNF), and compared macrophage IL-6 production to that of blood monocytes and lung fibroblasts. Unstimulated and TNF-stimulated alveolar macrophages and monocytes produced little or no detectable IL-6. In contrast, macrophages and monocytes produced large amounts of IL-6 in response to LPS and monocytes produced lesser but readily detectable amounts in response to rIL-1. Monocytes and alveolar macrophages differed significantly in their capacity to produce IL-6, with macrophages making more IL-6 in response to LPS and less IL-6 in response to rIL-1 than autologous blood monocytes. Monocytes aged in vitro produced little detectable IL-6 in response to LPS or rIL-1, suggesting that differences in cell maturity may account for the diminished capacity of the alveolar macrophage to produce IL-6 in response to IL-1 but not its enhanced capacity to produce IL-6 in response to LPS. Mononuclear phagocytes and lung fibroblasts also differed in their ability to produce IL-6. Lung fibroblasts produced more IL-6 in response to rIL-1 and less IL-6 in response to LPS than monocytes and macrophages. In addition, monocytes and macrophages elaborated electrophoretically identical IL-6 moieties that differed from those produced by lung fibroblasts. These differences could be at least partially attributed to differences in sialylation and/or glycosylation.(ABSTRACT TRUNCATED AT 250 WORDS)     

Modulation and function of intercellular adhesion molecule-1 (CD54) on human retinal pigment epithelial cells.

As part of the blood-retina barrier, the neuroectodermally-derived retinal pigment epithelial (RPE) monolayer is strategically positioned to interact with circulating leukocytes and regulate their access to the retina. We, therefore, studied whether human RPE cells express intercellular adhesion molecule-1 (ICAM-1), a specialized cell surface glycoprotein that binds the leukocyte function antigen-1 receptor present on all leukocytes. Using specific monoclonal antibody to ICAM-1, immunohistochemical staining of freshly-isolated primary and fourth passaged human RPE cells resulted in delicate reaction product that increased dramatically upon exposure to human recombinant (r) interferon-gamma (rIFN-gamma), interleukin-1-beta (rIL-1 beta), or tumor necrosis factor-alpha (rTNF-alpha). Fluorescence-activated cell sorting analysis demonstrated 2-fold increases in constitutive RPE ICAM-1 expression within 6 hours of exposure to physiologic concentrations of rIFN-gamma, rIL-1 beta, or rTNF-alpha. In standardized leukocyte adherence assays, cultured RPE cells showed avid binding of neutrophils that increased significantly after stimulation with rIFN-gamma, rIL-1 beta, or rTNF-alpha (p less than 0.001). In parallel assays, monoclonal antibody to either ICAM-1 on RPE cells, or subunits of leukocyte function antigen-1 receptors on leukocytes significantly blocked leukocyte binding to unstimulated (p less than 0.001) or rIFN-gamma-stimulated RPE cells (p less than 0.001). To demonstrate RPE ICAM-1 expression in intact human tissue, fresh uveoretinal explants were exposed to rIFN-gamma, rIL-1 beta, or rTNF-alpha and stained using mAb to ICAM-1. Tissue sections of cytokine-stimulated explants revealed dramatic increases in RPE ICAM-1 immunoreactivity over the low levels observed in unstimulated uveoretinal tissue. Our results indicate that: (a) ICAM-1 is expressed at low levels on unstimulated RPE cells, (b) RPE ICAM-1 may be augmented by inflammatory cytokines, and (c) RPE ICAM-1 is a functional receptor mediating leukocyte binding. ICAM-1 on RPE cells at the blood-retina barrier may regulate leukocytic infiltration in ocular diseases in which leukocytes are important pathogenetically and may be important to the generation of ocular immune responses.     

IL-8 causes in vivo neutrophil migration by a cell-dependent mechanism.

A dose-dependent neutrophil migration was observed following the injection of recombinant interleukin-8 (rIL-8) into rat peritoneal cavities. This finding contrasts with the inability of rIL-8 to induce neutrophil emigration into subcutaneous air-pouches. Pre-treatment of the animals with dexamethasone (0.5 mg/kg, s.c.) or depletion of the peritoneal resident cell population abolished the neutrophil migration induced by rIL-8 and by recombinant interleukin-1 beta (rIL-1 beta). Different from that which occurs with rIL-1 beta, neutrophil migration induced by rIL-8 was not enhanced by an increase in the peritoneal macrophage population. Transference of homologous total resident peritoneal cells to the air-pouch rendered this cavity responsive to the chemotactic effect of rIL-8 and potentiated the neutrophil migration induced by rIL-1. Our results show that both rIL-8 and rIL-1 beta are able to induce in vivo neutrophil migration by an indirect mechanism, dependent on resident cells. Neither macrophages nor lymphocytes seem to be involved in the rIL-8 chemotactic effect. However, peritoneal resident mast cells may be implicated in this mechanism. These cells, when stimulated in vitro by rIL-8, released a factor that when injected into peritoneal and air-pouch cavities induced neutrophil migration.     

Characterization of the adherence of human monocytes to cytokine-stimulated human macrovascular endothelial cells.

At sites of inflammation, interactions between monocytes and vascular endothelium play an important role in the margination and extravasation of monocytes. The aim of this study was to investigate the relative contributions of the CD11/CD18 family of leucocyte adhesion molecules on monocytes and ICAM-1 and ELAM-1 molecules on endothelial cells (EC) to the binding of monocytes to EC stimulated with recombinant interleukin-1 alpha (rIL-1 alpha), rIL-6, recombinant tumour necrosis factor-alpha (rTFN-alpha) or recombinant interferon-gamma (rIFN-gamma). The adhesiveness of EC for monocytes increased 1.8-2.3-fold after incubation of monolayers of venous or arterial EC with rIL-1 alpha or rTNF-alpha for 4 hr, and 1.6-2.0-fold after stimulation of both types of EC with rIL-1 alpha, rTNF-alpha or rIFN-gamma for 24 hr. Incubation with rIL-6 was without effect. The monoclonal antibodies (mAb) against CD11a, b, c and CD18 on monocytes did not inhibit the increase in the number of monocytes bound to rIL-1 alpha-, rTNF-alpha-, or rIFN-gamma-stimulated EC. However, mAb against ELAM-1 expressed on the surface of 4 hr rIL-1 alpha-stimulated EC slightly inhibited (15-21%) the enhanced monocyte binding. ICAM-1, which exhibited marked expression on 24 hr rIL-1 alpha-, rTNF-alpha- or rIFN-gamma-stimulated EC, did not contribute to the enhanced monocyte binding. The percentage of EC-bound monocytes which had stretched out over the surface of cytokine-stimulated venous or arterial EC was significantly increased compared to the percentage found for non-stimulated EC. It was observed that mild fixation of EC as well as treatment of EC with cytochalasin B or mAb against ICAM-1 did not affect the number of monocytes that were bound to EC, but considerably reduced the percentage of EC-bound monocytes with a stretched morphology. It is concluded that the binding of monocytes to cytokine-stimulated EC is dependent on the type of cytokine and the duration of cytokine stimulation. The increase in the binding of monocytes to cytokine-stimulated EC occurred as a result of CD11/CD18- and ICAM-1-independent factors. The subsequent morphological changes, i.e. stretching of monocytes over the surface of EC, required viable EC and ICAM-1.     

Endogenous tumor necrosis factor alpha is required for enhanced antimicrobial activity against Toxoplasma gondii and Listeria monocytogenes in recombinant gamma interferon-treated mice.

In vitro studies have shown that macrophages stimulated with recombinant gamma interferon (rIFN-gamma) produce tumor necrosis factor alpha (TNF-alpha), which in an autocrine fashion activates these cells. The aim of the present study was to determine whether endogenously formed TNF-alpha also is required for rIFN-gamma-induced macrophage activation and enhanced antimicrobial activity in vivo. After an intraperitoneal injection of rIFN-gamma into CBA/J mice, their peritoneal macrophages released enhanced amounts of NO2- and inhibited the intracellular proliferation of Toxoplasma gondii. Injection of neutralizing antibodies against TNF-alpha simultaneously with the rIFN-gamma completely inhibited both the release of NO2- by macrophages and their toxoplasmastatic activity. Similar results were observed after intraperitoneal injection of a competitive inhibitor of L-arginine, NG-monomethyl-L-arginine, together with rIFN-gamma, demonstrating that in vivo L-arginine-derived reactive nitrogen intermediates are essential for the induction of toxoplasmastatic activity. Intravenous injection of rIFN-gamma inhibited the growth of Listeria monocytogenes in the livers and spleens of mice; this effect was abrogated by antibodies against TNF-alpha. Intravenous injection of a large dose of rTNF-alpha resulted in a decrease in the number of bacteria in the liver and spleen, but an injection of rIFN-gamma and rTNF-alpha did not result in enhanced inhibition of the proliferation of L. monocytogenes. Together, the results of the present study are the first to demonstrate that endogenous TNF-alpha is required in vivo for the expression of macrophage activation with respect to the release of reactive nitrogen intermediates and toxoplasmastatic activity and for enhanced listericidal activity in the livers and spleens of mice stimulated with rIFN-gamma.     

Effects of gamma interferon, tumor necrosis factor alpha, and interleukin-2 on infection and proliferation of Theileria parva-infected bovine lymphoblasts and production of interferon by parasitized cells.

Theileria parva is a protozoan parasite that infects bovine B cells and alpha beta and gamma delta T cells and transforms them into continually proliferating cells. CD4+ T. parva-antigen-specific immune T cells have been shown to produce cytokines in response to stimulation with parasitized cells, and T. parva-infected lymphocytes produce and consume T-cell growth factors and interleukin-2 (IL-2). To ascertain the role of T-cell cytokines on T. parva infections, we evaluated recombinant gamma interferon (rIFN-gamma), rIL-2, and tumor necrosis factor alpha (rTNF-alpha) for their effects on establishment, proliferation, and survival of parasitized cells. The results indicate that neither rIFN-gamma nor rTNF-alpha had an enhancing or inhibitory effect on the growth of established T. parva-infected T-cell clones, whereas bovine rIL-2 increased the proliferation of infected B-cell and alpha beta T-cell clones but not that of gamma delta T-cell clones. To evaluate the effects of the cytokines on establishment of parasitized cell lines, peripheral blood mononuclear cells were cultured in their presence immediately following infection with T. parva sporozoites. Neither rIFN-gamma nor rIL-2 altered the proportion of cells initially developing schizonts, but both enhanced establishment of infected cell lines by about twofold. In contrast, rTNF-alpha resulted in about a 33% decrease in the proportion of schizont-infected cells. Inhibitory effects on establishment of parasitized cell lines by rTNF-alpha were no longer apparent by 12 days following infection. Tests conducted during this study indicated that T. parva-infected lymphocytes also spontaneously produce IFN that is neutralized by acidic pH treatment. In conclusion, we speculate that none of these T-cell cytokines are likely to have a profound inhibitory effect in vivo on T. parva infections. Instead, IFN-gamma and IL-2 may facilitate the establishment of infection by T. parva.     

A Comparative Study on the Effects of rIL-4, rIL-2, rIFN-γ, and rTNF-α on Specific T-Cell Non-Responsiveness to Mycobacterial Antigens in Lepromatous Leprosy Patients in Vitro

We have studied lepromatous leprosy (LL) as a human model disease for T-cell non-responsiveness to specific mycobacterial antigens and studied the effect of rIL-4, rIL-2, rIFN-gamma and rTNF-alpha thereon. T-cell non-responsiveness to Mycobacterium bovis bacillus Calmette-Guerin (BCG) or purified protein derivative of M. tuberculosis (PPD) antigens could be overcome in 5 out of 8 non-responder patients by rIL-2 and in 2 out of 8 by rIL-4. The ability of rIL-4 to overcome BCG/PPD non-responsiveness was strongly dose-dependent. When rIL-2 and rIL-4 were added simultaneously, they seemed to synergize in their effect. T-cell non-responsiveness to M. leprae could be overcome only in 2 out of 18 non-responders by rIL-2 but not by rIL-4 alone. The ability of rIL-2 to overcome T-cell non-responsiveness to M. leprae antigens became particularly marked when the recombinant 65-kDa heat shock antigen of M. leprae was used instead of whole bacilli. Exogenously added rIL-4, and to a lesser extent rIL-2, strongly enhanced existing T-cell responses to BCG or M. leprae in the majority (8 out of 11) of responders. These findings may have implications for the in vivo manipulation of the immune response by recombinant lymphokines and vaccines.     

Augmentation of the human monocyte/macrophage chemiluminescence response during short-term exposure to interferon-gamma and tumour necrosis factor-alpha.

The effects of short-term (30 min) pre-incubation of human monocytes and macrophages (3-day cultured monocytes) with leucocyte-derived human interferon-gamma (IFN-gamma) and recombinant human tumour necrosis factor-alpha (rTNF-alpha) were examined. Pre-incubation of either monocytes or macrophages with rTNF-alpha or IFN-gamma (100 U/5 x 10(5) cells) augmented their respiratory burst to formyl-L-methionyl-L-leucyl-L-phenylalanine (fMLP), measured by the luminol- and lucigenin-dependent chemiluminescence assay. In addition, both cell types showed a burst of respiratory activity in the presence of rTNF-alpha or IFN-gamma only. The effects of IFN-gamma were removed by adsorption with an anti-IFN-gamma monoclonal antibody and those of rTNF-alpha were abolished by heating at 100 degrees C, or by the addition of anti-TNF-alpha monoclonal antibody. The results demonstrate that both IFN-gamma and rTNF-alpha are stimulators of monocytes and macrophages, and rapidly alter the capacity of the cells to respond to fMLP, which binds to cell surface receptors.     

Effects of recombinant human interleukin-1 beta on accumulation of inflammatory peritoneal macrophages in mice treated with pertussis toxin.

In this study we report that treatment with recombinant human interleukin-1 beta (rIL-1 beta) (10 U per mouse, intraperitoneally) significantly increased the number of inflammatory macrophages in the peritoneal cavity of mice treated with pertussis toxin (PT) (1 micrograms per mouse, intravenously). The administration of rIL-1 beta in a single intraperitoneal dose (10 U per mouse) 1 or 2 days before challenge with PT did not prevent the decrease in the number of inflammatory macrophages in the peritoneal cavity of mice. On the other hand, the simultaneous administration of rIL-1 beta and PT, as well as the administration of rIL-1 beta 24 h after injection of PT, significantly counteracted the inhibitory effect of PT on inflammatory peritoneal macrophages.     

Immunomodulatory role of interleukin-10 in visceral leishmaniasis: defective activation of protein kinase C-mediated signal transduction events

Leishmania donovani, an intracellular protozoan parasite, challenges host defense mechanisms by impairing the signal transduction of macrophages. In this study we investigated whether interleukin-10 (IL-10)-mediated alteration of signaling events in a murine model of visceral leishmaniasis is associated with macrophage deactivation. Primary in vitro cultures of macrophages infected with leishmanial parasites markedly elevated the endogenous release of IL-10. Treatment with either L. donovani or recombinant IL-10 (rIL-10) inhibited both the activity and expression of the Ca2+-dependent protein kinase C (PKC) isoform. However, preincubation with neutralizing anti-IL-10 monoclonal antibody (MAb) restored the PKC activity in the parasitized macrophage. Furthermore, we observed that coincubation of macrophages with rIL-10 and L. donovani increased the intracellular parasite burden, which was abrogated by anti-IL-10 MAb. Consistent with these observations, generation of superoxide (O2-) and nitric oxide and the release of murine tumor necrosis factor-alpha were attenuated in response to L. donovani or rIL-10 treatment. On the other hand, preincubation of the infected macrophages with neutralizing anti-IL-10 MAb significantly blocked the inhibition of nitric oxide and murine tumor necrosis factor-alpha release by the infected macrophages. These findings imply that infection with L. donovani induces endogenous secretion of murine IL-10, which in turn facilitates the intracellular survival of the protozoan and orchestrates several immunomodulatory roles via selective impairment of PKC-mediated signal transduction.     

Interleukin-one induced arachidonic acid turnover in macrophages

Recombinant Interleukin-1 beta (rIL-1 beta) induced dose-dependent synthesis of PGE2 and activation of both cell-associated and soluble phospholipase A2 (PLA2) in cultured mouse peritoneal macrophages. Kinetic studies showed that rIL-1 beta was able to induce a 2-fold increase in membrane bound PLA2 activity in the cells within 1 hour. Elevated levels of PGE2 were however observed only after 4 hours stimulation of the cells while significant levels of soluble PLA2 activity were observed only after 16 hours stimulation of the cells. Stimulation with rIL-1 beta of macrophage cultures pre-labelled with [14C]arachidonic acid resulted in a loss of label from both phosphatidyl-choline and phosphatidylinositol in only 10 min, which was not observed in control (unstimulated) cultures. These results are indicative that the two phospholipids are the principal source of arachidonic acid for IL-1 induced prostaglandin synthesis.     

Human corneal interleukin-8. IL-1 and TNF-induced gene expression and secretion.

Corneal leukocytic infiltration is an important component of numerous ocular diseases, but specific corneal-derived leukocyte chemotaxins have not been identified. In this study, the authors identified interleukin-8 (IL-8), a known neutrophil and lymphocyte chemotaxin, to be an important chemotaxin produced by human corneal tissue. In situ hybridization and immunohistochemistry of corneas exposed to human recombinant (r) interleukin-1-beta (rIL-1 beta) or tumor necrosis factor-alpha (rTNF-alpha) revealed significant increases in corneal endothelial and stromal cell IL-8 mRNA (P less than 0.001) and marked increases in cell-associated immunoreactive IL-8 compared with unstimulated controls. ELISA assays revealed four- to eight-fold increases in corneal IL-8 secretion after 24-hour exposures to either cytokine over that obtained with unstimulated corneas (P = 0.01). In neutrophil chemotactic bioassays, significant increases in functional IL-8 were detected in media conditioned by corneas exposed to rIL-1 beta or rTNF-alpha for 24 hours (P less than 0.001). Preincubation of these corneal media with anti-IL-8 antibody significantly reduced neutrophil chemotaxis by more than 80%. These results suggest that the cornea is an active participant in ocular inflammation and raise the possibility that agents used in experimental corneal pocket models may produce indirect effects by inducing corneal secretion of other factors, such as IL-8.     

Alternatively activated macrophages express the IL-27 receptor alpha chain WSX-1

The interleukin (IL)-27 receptor-alpha WSX-1 is one component of the heterodimeric IL-27 receptor that is expressed on various cell types including macrophages. We previously demonstrated that IL-27 induces STAT-3 and is able to inhibit the production of pro-inflammatory cytokines in activated macrophages suggesting a novel feed-back mechanism by which IL-27 can modulate excessive inflammation. Because IL-4 receptor-alpha (IL-4Ralpha)-induced alternatively activated macrophages have also been described to attenuate pathological inflammatory immune responses, we analyzed the contribution of IL-27 in alternative macrophage activation. In the present study, like IL-10 and IL-4, IL-27 was found to suppress IL-12/23p40 production in activated bone marrow-derived macrophages. Whereas IL-10 induced the upregulation of the IL-4Ralpha on macrophages, receptor expression was not triggered by IL-27. In contrast to IL-4, IL-27 did not induce alternative macrophage activation but IL-4 strongly upregulated the expression of WSX-1 on macrophages and alternative macrophage activation enhanced IL-27-mediated signalling. We therefore conclude from our study that IL-10, IL-4 and IL-27 collaborate in modulating macrophage activation by successive upregulation of the IL-4Ralpha and WSX-1 on alternatively activated macrophages.     

Characterization of the immunoregulatory properties of thymosin alpha 1 on interleukin-2 production and interleukin-2 receptor expression in normal human lymphocytes

Thymosin alpha 1 (T alpha 1) and thymosin fraction 5 (TF5) have been shown to induce lymphocyte maturation and differentiation as well as to modulate mature immune responses to antigens and mitogens. The present study focused on the characterization of the mechanisms involved in T alpha 1 and TF5 enhancement of phytohemagglutinin (PHA)-induced interleukin-2 (IL-2) secretion and interleukin-2 receptor (IL-2R) expression in human mononuclear cells. We provide evidence that TF5 and T alpha 1 modulate an early event(s) during lymphocyte activation by mitogens. A short preincubation period (30 min) of non-adherent cells with thymosins, followed by extensive washing and subsequent exposure to PHA, was sufficient to enhance the production of IL-2 and the expression of IL-2R induced by the mitogen. Furthermore, the concomitant addition of PHA and thymosin during the preincubation period is not necessary for the enhancing effects to occur. We have also studied the role of macrophages on thymosin modulation of these responses. Results presented here indicate that macrophages are not essential for the interaction of thymosins with T-cells. However, macrophages are an absolute requirement during the exposure to the mitogen after preincubation with thymosins for the manifestation of TF5- and T alpha 1-mediated enhancing effects on IL-2 production and IL-2R expression. Human recombinant interleukin-1 beta (rIL-1 beta) was able to replace this macrophage requirement, indicating that production of IL-1 by these cells is a critical event in thymosin modulation of the IL-2 system. Two-color flow cytometric analysis and experiments involving the use of highly purified helper/inducer (Th, CD4+) and cytotoxic/suppressor (Tc, CD8+) T-cell populations indicated that both, Th and Tc cell populations are targets of thymosin activity. These studies provide additional evidence that thymosins play an important role in the modulation of the normal immune response and begin to define the mechanisms underlying T alpha 1 immunoregulatory properties.     

Regulatory effect of interleukin-4 (IL-4) on the expression and function of lymphocyte adhesion receptors involved in IL-2-induced cell aggregation.

Human recombinant interleukin-4 (rIL-4) was studied for its capacity to inhibit rIL-2-induced lymphoid cell aggregation. In contrast to rIL-2, rIL-4 was unable to induce cluster formation by itself. However, when added simultaneously with rIL-2 to cultures of freshly isolated peripheral blood lymphocytes (PBL), rIL-4 inhibited cell aggregation in a dose-dependent way. In contrast, PBL, preactivated by a 4-day culture in the presence of 500 U/ml rIL-2, were not inhibited in their adhesive capacity by rIL-4. Inhibition of cell aggregation was most prominent at 24 hr and virtually lost after 72 hr of culture. Phenotypical analysis revealed that rIL-4, with similar kinetics, decreased the rIL-2-mediated up-regulation of the CD2, CD54 and CD49e adhesion molecules. In addition, it was observed that up-regulation of the activation epitope on CD11a recognized by the mAb NKI-L16, was prevented. During 24hr of culture rIL-4 itself did not alter the expression of these antigens. Blocking experiments with mAb directed against adhesion structures did not reveal a direct role for CD49e, but obviously demonstrated involvement of CD11a/CD18-CD54 and CD2-CD58 interactions in the rIL-2-induced adhesion. Therefore, rIL-4 appears to inhibit the early phase of rIL-2-induced aggregation by preventing the up-regulation of CD54 and CD2 antigens and by inhibiting the generation of the activated state of the CD11a/CD18 receptor.