Role of TNF-alpha in the induction of fungicidal activity of mouse peritoneal exudate cells against Cryptococcus neoformans by IL-12 and IL-18

We have recently demonstrated that two IFN-gamma-inducing cytokines, interleukin (IL)-12 and IL-18, synergistically induced the fungicidal activity of mouse peritoneal exudate cells (PEC) against Cryptococcus neoformans through NK cell production of interferon (IFN)-gamma and nitric oxide (NO) synthesis. In the present study, we further dissected these effects by examining the involvement of tumor necrosis factor (TNF)-alpha in the induction of IL-12/IL-18-stimulated PEC fungicidal activity. The addition of neutralizing anti-TNF-alpha mAb significantly suppressed IL-12/IL-18-stimulated PEC anticryptococcal activity. This effect was ascribed to the inhibition of macrophage NO synthesis, but not of IFN-gamma production by NK cells, because the same treatment inhibited the former response, but not the latter one. On the other hand, combined treatment with IL-12 and IL-18 synergistically induced the production of TNF-alpha by PEC and this effect was almost completely abrogated by neutralizing anti-IFN-gamma mAb. The cell type producing TNF-alpha among PEC was mostly macrophage. TNF-alpha significantly promoted macrophage NO production and anticryptococcal activity induced by IFN-gamma, and furthermore anti-TNF-alpha mAb partially inhibited these responses. Considered together, our results indicated that TNF-alpha contributed to the potentiation of IL-12/IL-18-induced PEC fungicidal activity against C. neoformans through enhancement of IFN-gamma-induced production of NO by macrophages, but not through increased production of IFN-gamma by NK cells.     

Activity of natural killer cells of the spleen of mice exposed to low-intensity of extremely high frequency electromagnetic radiation

The dose dependence of natural killer (NK) cell activity from mouse spleen upon action of low-intensity millimeter waves in the exposure range from 5 to 96 hours was studied. It has found an increase of NK activity by 24 hours posttreatment that returned to normal level in a day after the cessation of the irradiation. Also the stimulation of isolated NK cell activity after millimeter waves treatment within 1 hour was revealed.     

Oxidant activity in skeletal muscle fibers is influenced by temperature, CO2 level, and muscle-derived nitric oxide

Free radicals are produced continuously by skeletal muscle fibers. Extracellular release of reactive oxygen species (ROS) and nitric oxide (NO) derivatives has been demonstrated, but little is known about intracellular oxidant regulation. We used a fluorescent oxidant probe, 2',7'-dichlorofluorescin (DCFH), to assess net oxidant activity in passive muscle fiber bundles isolated from mouse diaphragm and studied in vitro. We tested the following three hypotheses. 1) Net oxidant activity is decreased by muscle cooling. 2) CO(2) exposure depresses intracellular oxidant activity. 3) Muscle-derived ROS and NO both contribute to overall oxidant activity. Our results indicate that DCFH oxidation was diminished by cooling muscle fibers from 37 degrees C to 23 degrees C (P < 0.001). The rate of DCFH oxidation correlated positively with CO(2) exposure (0-10%; P < 0.05) and negatively with concurrent changes in pH (7.0-8.5; P < 0.05). Separate exposures to anti-ROS enzymes (superoxide dismutase, 1 kU/ml; catalase, 1 kU/ml), a glutathione peroxidase mimetic (ebselen, 30 microM), NO synthase inhibitors (N(omega)-nitro-l-arginine methyl ester, 1 mM; N(omega)-monomethyl-l-arginine, 1 mM), or an NO scavenger (hemoglobin, 1 microM) each inhibited DCFH oxidation (P < 0.05). Oxidation was increased by hydrogen peroxide, 100 microM, an NO donor (NOC-22, 400 microM), or the substrate for NO synthase (l-arginine, 5 mM). We conclude that net oxidant activity in resting muscle fibers is 1) decreased at subphysiological temperatures, 2) increased by CO(2) exposure, and 3) influenced by muscle-derived ROS and NO derivatives to similar degrees.     

Genetic studies on a tumor growth-inhibitory factor in serum of normal mice and its relationship to NK activity

It has been shown that normal mouse serum contains a tumor growth-inhibitory factor (GIF). and that strain-dependent levels of GIF correlate with mouse NK activity. To further analyze the genetic control of GIF we have studied the growth-inhibitory activity of normal mouse serum from 8 different mouse strains and their F1 hybrids. A sensitive method using a chromogenic substrate for an endogenous lysosomal enzyme was used to measure the inhibitory activity of normal mouse serum on the mouse B16 melanoma. The highest level of GIF was found in old mice, lower activity in serum of young animals and no activity in suckling mice. To compare the genetic control of GIF and NK, spleen NK activity against B16 as well as YAC-1 targets was measured in parallel in the same animals. Confirming previous results we found the H-2k strains CBA and C3H to have high levels of GIF as well as NK activity, while the strain A/Sn and the A congenic strain A.SW had low levels of both activities. Experiments with H-2d and H-2b strains, however, showed that GIF and NK had a different genetic control; thus the DBA/2 and Balb/c strains had considerably higher GIF activity than the C57B1 and Leaden strains, while the reverse was true for NK activity. In F1 hybrid crosses between strains with high and low activity, high activity was inherited as a dominant trait for both GIF and NK. A backcross analysis in (A X CBA) X A backcross mice, segregating for NK and GIF showed that the two activities did not cosegregate. These studies therefore demonstrate that GIF and NK activity are under different genetic control, and do not support any direct or simple relationship between GIF and NK cells.     

Strain difference in mouse natural killer activity augmented by mouse interferon and poly I.C

The level of natural killer (NK) activity was found to vary considerably among several mouse strains. In vivo and in vitro, interferon (IFN) and IFN inducers have been shown to augment mouse NK activity. C3H/He mice showed high NK activity, DDD/1 and A/J mice low NK activity, and C57BL/6, BALB/c and DBA/2 mice intermediate NK activity after injection with polyinosinic polycytidylic acid (poly I. C.). The same NK activity correlation was observed in nontreated mice, but the NK activities were lower compared with the poly I. C.-injected mice. Moreover, the DDD/1 and A/J mice showed almost no augmentation of NK activity on injection with poly I.C. In vivo, C3H/He, BALB/c and C57BL/6 mice injected with IFN showed augmented NK activity, but DDD/1 mice showed no such reaction. In vitro, C3H/He, BALB/c and C57BL/6 mouse spleen cells treated with IFN also showed augmented NK activity, but DDD/1 mouse spleen cells showed almost none. F1 hybrids between high (C3H/He) and low (DDD/1) NK-activity strains showed high NK activity. Thus, activity is dominant over low activity. The segregation of (DDD/1 X C3H/He) Fl X DDD/1 back-cross mice suggested that the strain differences in NK activity are under polygenic control.     

The thiol proteinase inhibitors improve the abnormal rapid down-regulation of protein kinase C and the impaired natural killer cell activity in (Chediak-Higashi syndrome) beige mouse

Protein kinase C (PKC) is essential in intracellular signal transduction for various cell functions including natural killer (NK) cell activity. This enzyme is hydrolysed by calpain, which is Ca2+-dependent thiol proteinase. We showed here that in NK activity-deficient beige (bg/bg) mouse, the model of Chediak-Higashi syndrome, the translocated membrane-bound PKC activity declined rapidly in NK cell-enriched lymphocytes after TPA stimulation. However, the rapid decline was abolished by the pretreatment of cells with leupeptin (a thiol and serine proteinase inhibitor) or E64 (a thiol proteinase inhibitor). Furthermore, these reagents improved the impaired NK cell activity in beige mouse whereas they did not affect NK cell activity in C57BL/6 (+/+) and the heterozygous (+/bg) mice. Meanwhile, TPA stimulation induced only low levels in NK cytotoxic factors (NKCF) release from beige NK cells, but these reagents augmented the lowered NKCF release. These results suggest that the improvement of impaired NK cell activity in beige mouse by the thiol proteinase inhibitors may be due to the elimination of abnormal rapid down-regulation of PKC, resulting in the augmentation of the lowered PKC activity.     

Effect of in vivo administration of prostaglandins and interferon on natural killer activity and on B-16 melanoma growth in mice

We investigated the effect on in vivo administration of 16,16-dimethyl-prostaglandin E2 (di-M-PGE2) alone and in combined treatment with alpha-beta interferon (IFN) on NK activity in normal, cyclophosphamide (CY)-treated, tumor bearing or irradiated and bone marrow reconstituted mice and on B-16 melanoma growth. Normal mice inoculated with IFN (30,000 U/mouse, 24 hr before testing) showed a significant increase in NK activity, while those treated with di-M-PGE2 (10 micrograms/mouse daily X 4 days) showed a suppressed NK response. No difference was observed between mice treated with di-M-PGE2 alone and those treated with di-M-PGE2 associated with IFN. In immunosuppressed animals single treatments were slightly (di-M-PGE2) or not (IFN) effective, while the combined administration of di-M-PGE2 and IFN caused a marked increase in NK activity. Moreover, di-M-PGE2 was able to accelerate the recovery rate of NK activity in bone marrow-reconstituted murine chimeras, suggesting that the synergistic effect of prostaglandins and IFN could derive from the action of di-M-PGE2 on progenitor pre-NK cells and of IFN on effector cells. Finally, we observed a good correlation between the enhancement of the NK activity and the tumor growth inhibition.     

Natural killer cell activity in murine muscular dystrophy. III. NK-sensitive myoblast cells and lack of NK activity in beige/dystrophic hybrid mice

The NK-susceptibility of dystrophic mouse myoblast cells was investigated. Spleen cells from 8- to 10-week-old normal (+/+) and dystrophic (dy2J/dy2J) male C57BL/6J mice were fractionated on Percoll density gradients and the cells at each density interface were incubated with either 51Cr-labeled YAC-1 or myoblast cells in a 6 hr 51Cr-release assay. Myoblast target cells were obtained from either heterozygous (+/dy2J) or homozygous (dy2J/dy2J) muscle cultures or a transformed tetraploid myoblast line (M14D2). The data indicate that the interface between the 50 and 60% (1.060-1.075 g/ml) Percoll density fractions of spleen cells from either normal or dystrophic mice contains the largest proportion of asialo GM-1 positive and NK-1 positive cells displaying NK activity. Myoblast cells from either heterozygous (phenotypically normal) or homozygous dystrophic mice were not significantly different in susceptibility to NK-mediated lysis by Percoll enriched normal or dystrophic mouse NK cells. However, dystrophic mouse spleen cells had the highest NK activity against both myoblast targets as compared with normal mouse spleen cells. The transformed myoblast cell line, M14D2, was significantly less susceptible to NK-mediated lysis by dystrophic mouse spleen cells when compared with freshly cultured myoblast target cells. Target cell binding studies revealed that conjugate forming cells from the 50% Percoll density interface of dystrophic mouse spleen cells were approximately twofold greater than that of normal mouse spleen cells against either heterozygous or homozygous dystrophic mouse myoblast targets. Cold target inhibition studies revealed that the natural killing of dystrophic mouse myoblast cells was due to a YAC-1 reactive NK cell. Breeding experiments between C57BL/6J homozygous "beige" (bgJ/bgJ) mutant mice and dystrophic (dy2J/dy2J) mice produced beige/dystrophic hybrid mice which displayed clinical symptoms of the dystrophy process by 3 to 4 weeks of age. Spleen cells from these hybrid mice showed no significant differences in NK activity against YAC-1 target cells when compared with homozygous beige mice. Taken together, these results demonstrate the first reported evidence that murine myoblasts are susceptible to NK-mediated lysis. In addition, the data indicate that although dystrophic mouse NK cells recognize myoblast cells as targets, the NK cell studies with the beige/dystrophic hybrid mice do not indicate a direct in vivo role for NK cells in the dystrophy process.     

Interleukin 2 dependence of human natural killer (NK) cell activity

When purified populations of human natural killer (NK) cells were tested for cytotoxic activity in the presence of partially purified preparations of human interleukin 2 (IL 2), a definite, dose-dependent linear increase in reactivity was observed. To determine whether such augmentation by IL 2 might reflect an important aspect of the physiologic regulation of NK activity, we examined the effects of monoclonal antibodies against human IL 2 on spontaneous NK activity. The presence of such antibodies during the 4-hr cytotoxicity assay resulted in significant inhibition of NK activity, and when the NK cells were pretreated for 16 to 20 hr with anti-IL 2, little or no activity remained. These data suggest that the spontaneous cytotoxic activity of NK cells is dependent on their continued exposure to IL 2. The reduction in NK activity resulting from treatment with anti-IL 2 could be at least partially restored by exposure to only low amounts of partially purified IL 2. These data have provided the basis for formulating a novel model of NK cell activation.     

Effects of microwave exposure on the hamster immune system. I. Natural killer cell activity

Hamsters were exposed to repeated or single doses of microwave energy and monitored for changes in core body temperature, circulating leukocyte profiles, serum corticosteroid levels, and natural killer (NK) cell activity in various tissues. NK cytotoxicity was measured in a 51Cr-release assay employing baby hamster kidney (BHK) targets or BHK infected with herpes simplex virus. Repeated exposure of hamsters at 15 mW/cm2 for 60 min/day had no significant effect on natural levels of spleen-cell NK activity against BHK targets. Similarly, repeated exposure at 15 mW/cm2 over a 5-day period had no demonstrable effect on the induction of spleen NK activity by vaccinia virus immunization, that is, comparable levels of NK were induced in untreated and microwave-treated animals. In contrast, treatment of hamsters with a single 60-min microwave exposure at 25 mW/cm2 caused a significant suppression in induced spleen NK activity. A similar but less marked decrease in spleen NK activity was observed in sham-exposed animals. Moreover, the sham effects on NK activity were not predictable and appeared to represent large individual animal variations in the response to stress factors. Depressed spleen NK activity was evident as early as 4 h postmicrowave treatment and returned to normal levels by 8 h. Hamsters exposed at 25 mW/cm2 showed an elevated temperature of 3.0-3.5 degrees C that returned to normal within 60 min after termination of microwave exposure. These animals also showed a marked lymphopenia and neutrophilia by 1 h posttreatment that returned to normal by 8-10 h. Serum glucocorticosteroids were elevated between 1 aNd 8 h after microwave treatment. Sham-exposed animals did not demonstrate significant changes in core body temperature, peripheral blood leukocyte (PBL) profile, or glucocorticosteroid levels as compared to minimum-handling controls.     

Inhibition of postbinding target cell lysis and of lymphokine-induced enhancement of human natural killer cell activity by in vitro exposure to ultraviolet B radiation

In vitro exposure of human peripheral blood mononuclear cells (PBMC) to ultraviolet B (uvB) radiation has been shown to inhibit natural killer (NK) cell-mediated cytotoxicity in a dose-dependent fashion. The purpose of this study was to examine the manner by which uvB produced these deleterious effects. Inhibition of NK activity was not due to lethal injury to NK cells since the viability of cell populations enriched for NK activity was greater than 90% with the uvB doses employed. uvB appeared to directly affect NK cells since procedures which removed suppressor mechanisms, such as removal of monocytes and pharmacologic inhibition of the cyclooxygenase pathway, failed to reverse the response. Furthermore, no suppression of activity of unirradiated NK cells could be produced by coincubation of unirradiated NK cells with uv-irradiated NK cells. When the single cell assay for binding and killing was employed to determine at which stage in the lytic sequence inhibition occurred, it was found that binding was normal but lysis of bound targets and the recycling capacity of active NK cells were markedly reduced. At uvB doses above 50 J/m2, both interferon alpha (IFN-alpha) and interleukin 2 (IL-2) were ineffective in augmenting NK cell-mediated cytotoxic reactions after cells had been irradiated with uvB. Furthermore, incubation of NK cells with IFN-alpha prior to irradiation failed to protect against the inhibitory effects. These studies provide evidence that in vitro exposure of NK cells to uvB radiation inhibits their function by a direct nonlethal effect and that this inhibition occurs selectively at the postbinding stage of target cell lysis.     

Modulation of natural cytotoxicity by alloantibodies. V. The mechanism of a selective augmenting effect of anti-H-2 antisera on the natural killer activity of mouse spleen cells

Treatment of mouse spleen cells with specific anti-H-2 antisera augments their natural killer (NK) activity against K562 cells but not against YAC target tumor cells. The same population of natural killer cells was found to lyse K562 as well as YAC target cells, since (a) depletion of YAC reactive NK cells by absorption on YAC monolayers resulted in a concomitant depletion of anti-K562 NK activity of mouse spleen cells, and (b) both K562 and YAC cells could inhibit their own as well as each others lysis in a cross-competition assay. Anti-H-2 antiserum could not induce anti-K562 NK activity in spleen cells previously depleted of NK cells by absorption on YAC monolayers, indicating that alloantiserum does not act by recruiting otherwise nonreactive cells to become cytotoxic toward K562 target cells. In a target-binding assay, K562 binding of NK cells (T-cell-, B-cell-, and macrophage-depleted spleen cells) increased five- to eightfold in the presence of anti-H-2 antiserum whereas YAC cells binding of NK cells was not increased. H-2 antigens per se did not appear to be involved in the alloantisera effect since anti-NK antiserum directed against a non-H-2 antigen selectively expressed on NK cells, showed a similar selective NK enhancing effect. Protein A, a reagent which binds to the Fc region of immunoglobulin molecules, completely blocked the alloantiserum induced augmentation of anti-K562 NK activity, but did not alter basal NK activity. Moreover, the F(ab)₂ fraction of alloantibodies failed to enhance anti-K562 cytotoxic activity of mouse spleen cells, indicating a crucial role for the Fc portion of the alloantibodies attached to the NK cells, in NK augmentation. Utilization of several target cell lines with or without membrane Fc receptors (FcR) revealed that alloantiserum enhanced the lysis of only FcR⁺ target cells. It is proposed that alloantibody-coated NK cells, as a result of a secondary interaction between attached alloantibody and Fc receptors on target cells, interact more readily with the target cells and thereby cause a higher level of lytic activity.     

Changes in number and density of large granular lymphocytes upon in vivo augmentation of mouse natural killer activity

NK activity of mice as well as humans and rats has been clearly associated with large granular lymphocytes (LGL). To better understand the effects of interferon (IFN) and IFN inducers on natural killer (NK) cells, we have compared the LGL in the spleens of normal and boosted mice. Cells were fractionated by centrifugation on discontinuous Percoll density gradients, and each fraction was tested for NK activity against YAC-1 targets and for the presence of LGL. In vivo treatment with C. parvum (0.7 mg/mouse, i.p., day-3), MVE-2 (25 mg/kg, i.p., day-3), poly I:C (4 mg/kg, i.p., day-3), or IFN (10(5) U/mouse, i.p., day-1) resulted in a marked augmentation and a change of distribution of cytotoxic activity. Most of the NK activity of boosted spleen cells was associated with lower density fractions 1 and 2, whereas active normal spleen cells had somewhat higher density (fractions 2 and 3). In parallel to their increased reactivity, the boosted spleens had a marked increase in the percentage of LGL, particularly in fractions 1 and 2. The augmented activity appeared to be mediated by the LGL, because treatment with anti-asialo GM1 or anti-Thy-1.2 plus complement reduced NK as well as the number of LGL. These results indicate that IFN-mediated boosting of NK activity in the spleen is due to an increase in the lower density LGL, as well as to an increase in the function of preexisting NK cells.     

The NK-1.1(-) mouse: a model to study differentiation of murine NK cells

The NK-1.1(-) mouse was constructed by weekly injections of monoclonal anti-NK-1.1 antibody from birth through adulthood. Spleen cells from these mice have decreased NK-1.1+ cells and null (Thy-1- and B220-) cells. Their splenic NK activity to YAC targets was low and was not enhanced by IFN-alpha or IFN-beta. Bone marrow (BM) of these NK-1.1(-) mice have normal precursors to NK cells: 1) NK activity could be generated from NK-1.1(-) BM cells cultured in rIL 2 for 5 to 6 days. These cultured BM cells expressed Qa-5, Thy-1, AsGm-1, and NK-1.1 antigens. The precursor cells of these BM cytotoxic cells are NK-1.1-; 2) transfer of BM cells from the NK-1.1(-) mice reconstituted the NK activity of irradiated, NK-depleted recipients. Lymphokine-activated killer cells could also be generated from spleens of these NK-1.1(-) mice. Therefore, the NK-1.1(-) mice were specifically depleted of mature cytotoxic NK cells, but not the NK-1.1- precursors of NK cells. This mouse model is valuable to study ontogeny and physiologic relevance of NK cells.     

Effects of nitrogen dioxide exposure on prostacyclin synthesis in lung and thromboxane A2 synthesis in platelets in rats

Effects of nitrogen dioxide (NO2) exposure on prostacyclin (PGI2) synthesis in the rat lung and thromboxane A2 (TXA2) synthesis in the platelets were studied. Male Wistar rats were exposed to 10 ppm NO2 for 1, 3, 5, 7 and 14 days. PGI2 synthesizing activity of homogenized lung decreased. The damage of PGI2 synthesizing activity reaches its maximum at 3 days. At 14 days, PGI2 synthesizing activity returned to the normal level. The activity of PGI2 synthetase decreased significantly. The formation of lipid peroxides due to NO2 exposure may cause the depression of PGI2 synthesizing activity of lung. On the other hand, platelet TXA2 synthesizing activity increased. This increased TXA2 synthesizing activity lasted at least till 3 days. Then, it returned to the normal level. The counts of platelet were decreased significantly by 1, 3, 5 and 7 days NO2 exposure. Then the decreased counts of platelet returned to the normal level at 14 days NO2 exposure. These results indicate that the depression of PGI2 synthesizing activity of lung by NO2 exposure cause an increase in TXA2 synthesizing activity of platelets. It may contribute to induce platelet aggregation and to the observed decrease in the number of platelets during NO2 exposure.     

Augmentation of mouse natural killer cell activity by muramyl dipeptide and its analogs

The ability of muramyl dipeptide (MDP) and its structural analogs (des-MDP, abu-MDP, and des-abu-MDP) to influence mouse natural killer (NK) cells in two different strains of mice was examined. In CBA/J mice, administration of MDP by both intraperitoneal (ip) and intravenous (iv) routes enhanced splenic NK cell activity. Maximum augmentation of NK cell activity was observed 3 days after MDP treatment. NK cell activity was also stimulated upon in vitro culture of CBA/J mouse spleen cells with MDP. Only iv inoculation of MDP to C57BL/6 mice 7 days previously enhanced NK cell activity of spleen cells. Peritoneal NK cell activity was not affected in either strain of mice, regardless of the route of inoculation of MDP. Two structural analogs of MDP, abu-MDP and des-abu-MDP, enhanced peritoneal NK cell activity, whereas des-MDP had no effect when tested 3 days after ip treatment of CBA/J mice with these compounds. Peritoneal NK cell activity of C57BL/6 mice was not modulated by des-MDP, abu-MDP, or des-abu-MDP. A synergistic effect on peritoneal NK cell activity was observed in both CBA/J and C57BL/6 mice treated first with MDP and then with lipopolysaccharide (LPS) or Bacillus Calmette-Guerin (BCG).     

Induction of nitric oxide production by natural killer cells: its role in tumor cell death.

It has been recognized that natural killer (NK) cells destroy AK-5 tumor cells, largely by cytolysis and apoptosis. The objective of this study was to elucidate the existence and the role of nitric oxide (NO) during this killing. The target cell killing ability of NK cells was associated with an increased production of NO with higher expression of inducible nitric oxide synthase. In part, the production of NO was confirmed by significant increase in cell lysis in the presence of l-arginine and attenuation of cell lysis, DNA fragmentation, and apoptosis by N(omega)-nitro-l-arginine methyl ester (L-NAME). An increased oxidation of intracellularly trapped dichlorofluorescein was observed in NK cells, which was effectively prevented by L-NAME. Exposure of AK-5 cells to chemically generated NO also induced DNA fragmentation in AK-5 cells. Further evidence for the involvement of NO in apoptosis was provided by the inhibition of specific cleavage of PARP and activation of CPP32 by L-NAME. Increased production of NO with simultaneous enhancement of the cytotoxic activity of NK cells from sc tumor-transplanted animals has been implicated in tumor regression when compared to the ip tumor-bearing animals. Overall, these observations suggest an important role for NO during NK cell-mediated apoptosis and lysis of AK-5 cells.     

Inhibition of natural killer activity in emotional stress and elimination of this phenomenon by interferon inducer

The authors provide the data on the effect of emotional painful stress (EPS) of varying duration on the activity of natural killers (NK) of August rats' spleen 1 to 9 days after exposure to stress is discontinued. The activity of NK was tested against target cells K 562 labeled with 51Cr in a 14-hour test. The effect of stress was found to depend on the duration of the stressor reaction. 1 1/2- and 3-hour stress activated NK, while prolonged 6-hour stress suppressed the activity of natural killers within the first 2 days, followed by the recovery by day 7-9 of the activity of lymphocytes having natural cytotoxicity. Tiloron, a synthetic inducer of endogenous interferon, corrected the post-stressor depression of NK activity.