Recombinant granulocyte-macrophage colony-stimulating factor (GM-CSF) or tumour necrosis factor-alpha (TNF-alpha) activate human alveolar macrophages to inhibit growth of Mycobacterium avium complex.

We investigated the effects of certain macrophage-active cytokines on the phagocytosis and growth inhibition of Mycobacterium avium complex (MAC) by human alveolar macrophages (AM). We also evaluated the effects of pretreatment with each cytokine on the superoxide anion release (O2-) from AM. The cytokines that we used were recombinant GM-CSF, natural type TNF-alpha, recombinant interferon-gamma (IFN-gamma), and recombinant IL-2. We found that phagocytosis by the various cytokine-stimulated AM was similar to that of unstimulated AM. On the other hand, significant growth inhibition of MAC was observed in the macrophages treated with GM-CSF or TNF-alpha, while no growth inhibition of MAC was observed in the macrophages treated with IFN-gamma or IL-2. Pretreatment with all cytokines tested enhanced the O2- release from AM, but there was no correlation between the enhancement of O2- release and the growth inhibition of MAC. Thus, we concluded that GM-CSF or TNF-alpha could activate AM to inhibit growth of MAC, probably not through the enhanced production of reactive oxygen intermediates.     

Increased Mycobacterium tuberculosis growth in HIV-1-infected human macrophages: role of tumour necrosis factor-alpha

Synergism between Mycobacterium tuberculosis (M. tuberculosis) and HIV-1 infections was demonstrated in several in vitro models and clinical studies. Here, we investigated their reciprocal effects on growth in chronically HIV-1-infected promonocytic U1 cells and in acutely infected monocyte-derived macrophages (MDM). Phagocytosis of M. tuberculosis induced HIV-1 expression in U1 cells, together with increased TNF-alpha production. M. tuberculosis growth, evaluated by competitive PCR, was greater in HIV-1-infected MDM compared to uninfected cells. M. tuberculosis phagocytosis induced greater TNF-alpha and IL-10 production in HIV-1-infected MDM than in uninfected cells. In uninfected MDM, addition of TNF-alpha and IFN-gamma decreased, whereas IL-10 increased M. tuberculosis growth. On the contrary, in HIV-1-infected MDM, addition of TNF-alpha and IFN-gamma increased, whereas IL-10 has no effect on M. tuberculosis growth. TNF-alpha seems to play a pivotal role in the enhanced M. tuberculosis growth observed in HIV-1-infected MDM, being unable to exert its physiological antimycobacterial activity. Here, for the first time we demonstrated an enhanced M. tuberculosis growth in HIV-1-infected MDM, in line with the observed clinical synergism between the two infections.     

The role of tumour necrosis factor-alpha in combination with interferon-gamma or interleukin-1 in the induction of immunosuppressive macrophages because of Mycobacterium avium complex infection.

The role of some cytokines including tumour necrosis factor-alpha (TNF-alpha), interleukin-1 alpha (IL-1 alpha), interferon-gamma (IFN-gamma), transforming growth factor-beta (TGF-beta) and interleukin-6 (IL-6) in the generation of immunosuppressive macrophages (M phi s) in host spleen cells of Mycobacterium avium complex (MAC)-infected mice was studied. M phi populations with potent suppressor activity against concanavalin A (Con A)-induced mitogenesis of splenocytes (SPCs) were elicited not only in euthymic but also in athymic nude mice during MAC infection. The suppressor M phi s are, therefore, inducible not only through a T-cell-dependent mechanism but also through T-cell-independent mechanism. However, MAC-induced M phi s of athymic mice displayed about four times lower suppressor activity than those of euthymic mice, indicating that mature T cells are important for M phi activation to the highly immunosuppressive state. Anti-TNF, anti-IFN-gamma, and anti-TGF-beta antibodies (Abs) but not anti-IL-6 Ab inhibited in vivo generation of MAC-induced immunosuppressive M phi s, and the neutralizing efficacy was in the order of anti-IFN-gamma Ab > anti-TNF Ab > anti-TGF-beta Ab. The effects of TNF-alpha, IL-1 alpha, IL-6, and IFN-gamma alone or combinations of them upon the acquisition of the suppressor activity by cultured splenic M phi s were studied. When normal splenic M phi s were treated with each cytokine for 3 days, TNF-alpha, IFN-gamma, and IL-1 alpha alone caused a slight elevation of their suppressive activity. Treatment of the normal M phi s with the combination of either TNF-alpha+IL-1 alpha or TNF-alpha+IFN-gamma yielded a marked increase in the suppressor activity, followed by IL-1 alpha+IFN-gamma. These findings indicate the important roles of TNF-alpha, IFN-gamma, and IL-1 alpha in the generation of MAC-induced suppressor M phi s.     

T-cell-independent granuloma formation in response to Mycobacterium avium: role of tumour necrosis factor-alpha and interferon-gamma.

We used Mycobacterium avium infection in severe combined immunodeficiency (SCID) mice to examine T-cell-independent mechanisms of inflammatory cell recruitment. SCID mice infected with a virulent strain of M. avium (TMC724) were able to recruit macrophages to sites of mycobacterial replication and formed organized and coherent granulomas in the absence of functional T cells. Phagocyte recruitment was almost totally ablated by neutralization of either tumour necrosis factor-alpha (TNF-alpha) or interferon-gamma (IFN-gamma) in vivo demonstrating that granuloma formation was dependent on the presence of these cytokines. This was concomitant with a reduction in the in situ cytokine mRNA levels otherwise induced in infected mice, for chemokines, pro-inflammatory and regulatory cytokines, including TNF-alpha, IFN-gamma, macrophage inflammatory protein-1 alpha, interleukin-1 beta (IL-1 beta) and IL-10. Furthermore, in vivo treatment of infected mice with anti-asialo GM-1 antisera, which depletes natural killer (NK) cells, prevented recruitment of inflammatory cells. In vitro studies confirmed that M. avium was able to elicit IFN-gamma from SCID spleen in a dose-dependent manner. These data show for the first time that secretion of IFN-gamma from NK cells can mediate a T-cell-independent pathway of granuloma formation and cellular infiltration in response to mycobacteria.     

Predicting death from tumour necrosis factor-alpha and interleukin-6 in 80-year-old people

Ageing is associated with low-grade inflammation and markers such as IL-6 possess prognostic value. Tumour necrosis-alpha (TNF-alpha) initiates the inflammatory cascade and has been linked to several age-associated disorders. It remains, however, unknown if TNF-alpha is associated with mortality in old populations. The aim of the present study was to investigate if serum levels of TNF-alpha were associated with all-cause mortality independently of interleukin (IL)-6 in a prospective study of 333 relatively healthy 80-year-old people. A Cox regression model was used to explore effects of TNF-alpha and IL-6 on survival in the following 6 years. A total of 133 participants died during this follow-up period. TNF-alpha was associated with mortality in men, but not in women, whereas low-grade elevations in IL-6 were associated strongly with mortality in both sexes. TNF-alpha explained only 7% of the variability in IL-6 and effects of the two cytokines were independent of each other as well as of other traditional risk factors for death [smoking, blood pressure, physical exercise, total cholesterol, co-morbidity, body mass index (BMI) and intake of anti-inflammatory drugs]. These findings indicate that at least in old populations chronic elevated levels of TNF-alpha and IL-6 have different biological functions that trigger age-associated pathology and cause mortality.     

Evidence that vesicles containing living, virulent Mycobacterium tuberculosis or Mycobacterium avium in cultured human macrophages are not acidic.

Mycobacterium tuberculosis and Mycobacterium avium multiply in cultured human macrophages (MP) within membrane-enclosed vesicles. These vesicles are generally assumed to be acidic. The evidence most frequently cited for this assumption is that pyrazinamide, which requires an acid pH to be effective, is effective and streptomycin, which loses most of its activity at a low pH, is poorly effective against tubercle bacilli. This assumption was tested by using the two weak bases chloroquine and NH4Cl to raise the pH of acidic vesicles in MP experimentally infected with M. tuberculosis or M. avium. An immunocytochemical locator of acidic regions in the MP was used to monitor the association of intracellular bacilli with acidity. MP were infected with M. tuberculosis or M. avium and incubated with various combinations of the drugs and the weak bases. Replication of the bacteria in the MP was measured by culture counts. Intracellular associations of the mycobacteria with acidity were assessed by electron micrographs and by using the weak base 3-(2,4-dinitroanilino)-3'-amino-N-methyl dipropylamine, which was detected with colloidal gold-labeled antibodies. It was confirmed by immunocytochemistry that both chloroquine and NH4Cl raise the pH of acidic vesicles in the infected MP. However, neither caused any pH-related change in the antimycobacterial activities of pyrazinamide or streptomycin or of the pH-independent drug isoniazid. Immunochemical analyses showed acidity to be associated with killed but not living mycobacteria in the MP. These findings suggest that living M. tuberculosis and M. avium are located in human MP in vesicles which are not acidic.     

Fc alpha receptors mediate release of tumour necrosis factor-alpha and interleukin-6 by human monocytes following receptor aggregation.

The functional capacity of the human monocyte receptor for the Fc portion of IgA (Fc alpha R) in mediating signal transduction was evaluated by cytokine release. F(ab')2 fragments of anti-Fc alpha R monoclonal antibodies (mAb) were used as specific probes to induce release of tumour necrosis factor-alpha (TNF-alpha) and interleukin-6 (IL-6). Multivalent cross-linking by a secondary anti-mouse antibody [F(ab')2 fragments] induced a significant release of TNF-alpha and IL-6 by human blood mononuclear cells, indicating requirements for Fc alpha R aggregation on the cell surface to transmit signals. Both cytokines were released exclusively by adherent cells, identifying monocytes as the responding cells within the mononuclear cell population. This cytokine release could not be due to contaminating endotoxins, because it was not abolished by polymyxin B, a lipopolysaccharide (LPS) inhibitor. Moreover, purified recombinant soluble Fc alpha R inhibited the anti-Fc alpha R mAb-mediated cytokine release from blood monocytes, demonstrating that TNF-alpha and IL-6 were released in a receptor-specific manner. Our data suggest that Fc alpha R, through its capacity to mediate secretion of IL-6, may play an important role in B-cell proliferation and immunoglobulin production. On the other hand, release of TNF-alpha following stimulation of Fc alpha R molecules directly implicates these receptors in amplification and regulation of the inflammatory process occurring during IgA-mediated host defence.     

Production of tumor necrosis factor-alpha and interleukin-6 by human alveolar macrophages exposed in vitro to coal mine dust.

Following our previous demonstration of cytokine secretion by alveolar macrophages (AM) from coal miners and from patients with coal workers' pneumoconiosis, we investigated the effect of in vitro exposure to coal dust and to its silica content on tumor necrosis factor-alpha (TNF), interleukin (IL)-1 beta, and IL-6 production by normal human AM. TNF and IL-1 beta concentrations were estimated by a specific radioimmunoassay, while IL-6 levels were evaluated by the proliferation of 7TD1 cells. After 24-h culture, coal dust triggered a significant release of TNF and IL-6 at the dose of 0.1 mg/ml and more obviously at 1 mg/ml in comparison with titanium dioxide (TiO2), used as a biologically inert control dust (with 1 mg/ml of dust: 3,526 +/- 3,509 versus 330 +/- 138 pg TNF/ml and 224 +/- 74 versus 72 +/- 34 U IL-6/ml, respectively; P less than 0.01 in both cases). After 3-h culture, a significant TNF secretion as well as an increased TNF mRNA expression were also detected for AM stimulated by coal dust at variance with TiO2. In contrast, no modification of IL-1 beta concentration could be evidenced in AM exposed to coal dust, although we detected an increased expression of specific mRNA expression. In order to define the role of silica among the main components of coal dust in AM activation, we evaluated the effect of silica (alpha-quartz, 30 micrograms/ml, which is the concentration and the type of silica present in our coal dust) alone or mixed with TiO2 (1 mg/ml) on monokine production.(ABSTRACT TRUNCATED AT 250 WORDS)     

Dose-dependent increase in plasma interleukin-6 after recombinant tumour necrosis factor infusion in humans.

Several studies have shown that the cytokine interleukin-6 (IL-6) is produced in response to tumour necrosis factor (TNF) in vitro. This study examines the in vivo relation between these two cytokines with assays of plasma IL-6 and TNF levels in subjects with chronic hepatitis B undergoing immunomodulatory therapy with recombinant TNF (rTNF). Plasma IL-6 was detected from 20 min after rTNF infusion with levels peaking after 2-3 h and levels correlated with the dose of rTNF administered (r = 0.67, P = 0.004). Peak levels of IL-6 (mean 295, range 266-297 ng/l) were lower than those seen in certain disease states despite the very high peak levels of rTNF (mean 11,750, range 5623-18,620 ng/l). These findings suggest that the very high levels of IL-6 found in certain disease states are not purely the result of circulating TNF. Other factors such as endotoxin or other cytokines may also play a role in determining levels of plasma IL-6.     

Human immunodeficiency virus type 1 enhances intracellular growth of Mycobacterium avium in human macrophages.

Disseminated Mycobacterium avium infections are common in patients with AIDS and result in a reduced life expectancy. Human monocytes/macrophages are important target cells for both human immunodeficiency virus (HIV) and M. avium. We have studied the interaction in vitro of M. avium and HIV type 1 (HIV-1) in human macrophages. Human monocytes isolated from the peripheral blood of healthy individuals were infected with HIV-1, M. avium, or both. The intracellular growth of M. avium and the replication of HIV-1 were monitored for up to 5 weeks. Intracellular mycobacterial growth was seen in all M. avium infected cell cultures and was paralleled by increased production of interleukin 1 alpha and beta. Preinfection of the macrophages with HIV-1 reduced the interleukin 1 production and accelerated the intracellular growth of M. avium. These findings may explain in part the impaired control of mycobacterial infections seen with patients with AIDS.     

Interleukin-10 inhibits polymethylmethacrylate particle induced interleukin-6 and tumor necrosis factor-alpha release by human monocyte/macrophages in vitro.

Periprosthetic membranes commonly observed at sites of total joint implant loosening exhibit abundant macrophages and particulate debris. Macrophages phagocytose orthopedic debris and release the pro-inflammatory mediators interleukin-1, interleukin-6, tumor necrosis factor-alpha, and prostaglandin E2. Populations of activated lymphocytes are often seen in periprosthetic membranes. These lymphocytes may modulate the monocyte/macrophage response to particulate debris and influence aseptic loosening. In addition, other immunologic agents, such as interleukin-10, are present in tissues harvested from the bone-implant interface of failed total joint arthroplasties. The present study examined the effects of interleukin-10 on polymethylmethacrylate (PMMA) particle challenged human monocyte/macrophages in vitro. Human monocyte/macrophages isolated from buffy coats of five healthy individuals were exposed to 1-10 microm PMMA particles. Interleukin-10 was added to the monocyte/macrophages with and without the addition of PMMA particles. Interleukin-10-induced alterations in monocyte/macrophage metabolism were determined measuring interleukin-6 and tumor necrosis factor-alpha release by the cells following exposure to PMMA particles. Exposure of the monocyte/macrophages to PMMA particles resulted in a dose-dependent release of interleukin-6 and tumor necrosis factor-alpha at 48 h. Interleukin-10 reduced the levels of interleukin-6 and tumor necrosis factor-alpha release by macrophages in response to PMMA particles in a dose-dependent manner. At 48 h, particle-induced interleukin-6 release was inhibited by 60 and 90% with 1.0 and 10.0 ng/ml treatments of interleukin-10, respectively. At 48 h, particle-induced tumor necrosis factor-alpha release was inhibited by 58 and 88% with 1.0 and 10.0 ng/ml treatments of interleukin-10, respectively. Interleukin-10 challenge alone did not significantly alter basal interleukin-6 or tumor necrosis factor-alpha release relative to control cultures. The data presented in this study demonstrate that the anti-inflammatory cytokine, interleukin-10, inhibits monocyte/macrophage release of the pro-inflammatory cytokines interleukin-6 and tumor necrosis factor-alpha in response to PMMA particle challenge in vitro.     

Growth within macrophages increases the efficiency of Mycobacterium avium in invading other macrophages by a complement receptor-independent pathway.

Infections caused by organisms of the Mycobacterium avium complex occur in approximately 50 to 60% of patients with AIDS. M. avium is an intracellular pathogen that survives and multiplies within mononuclear phagocytes. In this study, we investigated the uptake of M. avium grown within macrophages (intracellular growth M. avium [IG]) by a second macrophage compared with M. avium cultured in broth (extracellular growth M. avium [EG]). The results showed that IG was six- to eightfold more efficient than EG in entering macrophages. In addition, while an anti-CR3 antibody was able to inhibit approximately 60% of EG uptake by macrophages, it failed to inhibit the entry of IG. In contrast to EG, IG uptake into macrophages was significantly inhibited in the presence of anti-beta1-integrin and anti-transferrin receptor antibodies. Entry into macrophages by alternate receptors was associated with resistance to tumor necrosis factor alpha (TNF-alpha) stimulation. While stimulation with TNF-alpha resulted in inhibition of the growth of EG, it was not associated with inhibition of intracellular growth of IG. Investigation of the reason why M. avium is able to sense the changes in the intracellular environment triggering a change to the invasive phenotype suggests a direct relationship with macrophage apoptosis. These results suggest that intracellular growth is associated with novel mechanisms of M. avium uptake of macrophages and that those mechanisms appear to offer advantages to the bacteria in escaping the host defense.     

Biological activity of mutants of human tumour necrosis factor-alpha.

Point mutations in different regions of the tumour necrosis factor-alpha (TNF-alpha) molecule influence anti-tumour cytotoxic/cytostatic activities as well as haemorrhagic tumour necrosis, tumour regression and lethal toxicity in mice. Mutations in the C-terminal region in positions 150 and 155 markedly decrease cytotoxicity for murine L929 fibroblasts and human MCF7 mammary carcinoma cells. Competitive binding experiments with 125I-labelled TNF-alpha revealed that the loss of cytotoxicity is caused by a loss of target cell binding. In contrast to the reduced activity against L929 and MCF7 cells, neither binding to nor cytostatic activity against the human myeloid leukaemia cell lines HL60 and U937 are affected. This target cell type-dependent behaviour is probably due to the fact that L929 and MCF7 cells express different types of TNF receptor compared with myeloid leukaemia cells. While a mutation in position 127 decreases the overall activity of TNF-alpha, a deletion of four N-terminal amino acids does not reduce biological activity. In vivo the TNF mutants differed in their anti-tumour effects and lethal toxicity, but a segregation of anti-tumour activity and toxicity was not observed.     

IL-8 as a circulating cytokine: induction by recombinant tumour necrosis factor-alpha.

Tumour necrosis factor-alpha (TNF-alpha) is a pivotal cytokine at the centre of a cascade of cytokines and inflammatory mediators which modulate the host response to infection and trauma, and in particular the metabolic changes resulting in shock and subsequent multi-organ failure. The cytokine IL-8--predominantly an activator and chemotactic factor for circulating polymorphonuclear neutrophil leucocytes--is produced in response to TNF-alpha in vitro, and high circulating levels of IL-8 are found in septic primates. We have studied the release of IL-8 into the circulation of subjects with chronic hepatitis B undergoing a 10 week pilot trial of recombinant TNF-alpha (rTNF-alpha) therapy in doses of 15-100 micrograms/m2. A marked dose-dependent increase in plasma IL-8 levels was seen commencing at 30-60 min after the start of rTNF-alpha infusion and peaking between 2 and 3 h (mean peak level 4300 ng/l). The temporal pattern of IL-8 production exactly echoed that of IL-6, another component of the cytokine cascade, but peak plasma levels of IL-8 were up to 17 times higher than those of IL-6. This study confirms in vitro data suggesting that IL-8 is a component of the acute circulating cytokine cascade with a potential role in the modulation of the acute immune and metabolic response to infection and trauma.     

African swine fever: Expression of interleukin-1 alpha and tumour necrosis factor-alpha by pulmonary intravascular macrophages

To determine, in the acute form of African swine fever (ASF), the relationship between the appearance of pulmonary oedema and viral replication and expression of cytokines by pulmonary intravascular macrophages (PIMs), 14 pigs were inoculated intramuscularly with ASF virus (strain Espa?a'70) and killed in pairs on days 1-7 post-inoculation. Samples of lung were examined immunohistochemically and ultrastructurally. The immunohistochemical study was carried out with antibodies against interleukin-1 alpha (IL-1alpha), tumour necrosis factor-alpha (TNF-alpha), viral antigen of ASF (Vp73) and a myeloid marker (SWC3). Viral replication was observed mainly in PIMs, which at the same time showed intense activation, accompanied by the expression of IL-1alpha and TNF-alpha. The occurrence of interstitial oedema, neutrophil sequestration and fibrin microthrombi in septal capillaries coincided with high degrees of cytokine expression by infected PIMs. Alveolar macrophages did not show a significant change in cytokine expression as a result of ASF infection, and viral replication was detected in only a low percentage of these cells.     

Macrophage migration inhibitory factor reduces the growth of virulent Mycobacterium tuberculosis in human macrophages

Macrophage migration inhibitory factor (MIF) is a key mediator of the innate immune system and plays a crucial role in the host response to bacterial infections. Its role in immunity to intracellular pathogens has not been well studied. Here, we show that MIF released by infected human macrophages inhibits the growth of virulent Mycobacterium tuberculosis.     

Effects of fibronectin and group B streptococci on tumour necrosis factor-alpha production by human culture-derived macrophages.

Group B streptococci (GBS) are an important cause of sepsis and shock in the new-born. We have previously reported that GBS induce the production of tumour necrosis factor-alpha (TNF-alpha) by human monocytes and culture-derived macrophages. We have also shown that fibronectin (FN) promotes interaction between GBS and human phagocytes. In the present study, we investigated the effect of FN and GBS on the production of TNF-alpha by adult and neonatal culture-derived macrophages. We report that soluble FN alone was a strong stimulus for the production of TNF-alpha by culture-derived macrophages (FN 50 micrograms/ml = 623.33 +/- 47 pg/ml TNF, versus media alone 3 +/- 1.5 pg/ml; P < 0.0001). While GBS also induce the production of TNF-alpha by macrophages, the addition of FN to GBS had more than an additive effect on TNF-alpha levels. FN-mediated TNF-alpha production by macrophages was inhibited by both soluble arginine-glycine-aspartic acid (RGD) peptide (71%; P < 0.0001) and anti-beta 3-integrin monoclonal antibody 7G2 (54%; P < 0.0001). Neonatal culture-derived macrophages produced significantly more TNF-alpha in response to GBS (356.4 pg/ml +/- 27.7) than adult cells did (222.0 pg/ml +/- 21.0; P = 0.037), and dramatically more in response to FN alone (neonatal 1931.0 pg/ml +/- 23.0 versus adult 463.5 43.5 pg/ml; P < 0.0001). FN may contribute to the high levels of TNF-alpha production implicated in the pathophysiology of GBS sepsis and shock.     

The Candida albicans phospholipomannan induces in vitro production of tumour necrosis factor-alpha from human and murine macrophages.

We have previously identified a Candida albicans 14,000-18,000 MW antigen reacting with anti-beta-1,2-linked oligomannosides antibodies as being a phospholipomannan (PLM). Because of the structural similarities between the C. albicans PLM and lipophosphoglycans from various microbial pathogens known to be potent tumour necrosis factor-alpha (TNF-alpha) inducers, we investigated the PLM ability to induce TNF-alpha. Incubation of human monocytic cells THP-1 with PLM led to dose-dependent production of TNF-alpha that was significantly increased by prestimulation of the cells with interferon-gamma (IFN-gamma). Production of TNF-alpha by macrophages under PLM stimulation was confirmed by using macrophages elicited from the mouse peritoneal cavity. In all investigated conditions, PLM-induced TNF-alpha production differed significantly in both kinetics and dose dependence from lipopolysaccharide (LPS) induction used as control. It appears, therefore, that the C. albicans PLM shares functional homologies with microbial lipophosphoglycans identified as pathogenicity factors, although prestimulation of the target cells was required for the PLM-derived opportunistic pathogen to trigger the cytokine network.     

The roles of tumour necrosis factor-alpha, interleukin-1 and interleukin-12 in murine cytomegalovirus infection.

The swine is a useful model for immunobiological studies as it has a highly heterogeneous lymphocyte pool, containing several subsets not easily accessible in humans and rodents. In particular, the CD8-positive (CD8+) cells contain a variety of lymphocyte subsets, such as alpha beta-T cells, gamma delta-T cells, CD4 CD8 double-positive (DP) cells and natural killer (NK) cells. In order to define these subsets further, we have selected four monoclonal antibodies (mAb) with differential reactivity on CD8+ cells. Thus, mAb CD8.1 (PPT20) bound to CD8hi and CD8lo subpopulations in a similar way to the conventional anti-CD8. The mAb CD8.2 (PPT21), though binding to all of the CD8+ cells, reacted preferably with CD8hi. Two other mAb, CD8.3 (PPT22) and CD8.4 (PPT23), were specific for CD8hi alpha beta-T-cell subpopulation. These results, complemented by immunoprecipitation, co-modulation and enzyme-linked immunosorbent assay experiments, suggest that CD8.1 and CD8.2 react putatively with the CD8 alpha-chain and CD8.3 and CD8.4 with the CD8 beta-chain. Tissue distribution studies revealed that CD8+ thymocytes and peripheral CD8hi alpha beta-T cells expressed both putative CD8 alpha- and beta-chains while peripheral CD4+ CD8+ alpha beta-T cells, CD8lo gamma delta-T cells and NK cells expressed only putative CD8 alpha-chain. Functional studies indicated that the CD8hi alpha beta-T and CD8lo gamma delta-T cells were effector cells in the CD3-redirected cytotoxicity.