Experimental autoimmune neuritis (EAN) is an animal model of the Guillain-Barr6 syndrome. It permits study of cellular and humoral interactions occurring during autoimmune responses in the PNS. Lymphokines are important mediators of intercellular communication in the immune system. We examined the effects of recombinant interferon-gamma (r-IFN) on EAN. Lewis rats immunized with bovine myelin received daily injections of 100.00 U of homologous rIFN either during the induction phase of the disease (days 1-10 p.i.), or the amplification/effector stage (days 11-19 p.i.). Disease activity was monitored clinically, electrophysiologically, and after perfusion of rats on day 21 p.i. histologically. Early administration of r-IFN markedly augmented disease severity while treatment during the amplification/effector phase slightly improved clinical status and nerve conduction when compared with sham-treated animals. Histology closely correlated with these findings. Early administration of r-IFN could enhance the immune response by induction of MHC class II antigen expression on macrophages and Schwann cells. Late application of r-IFN may ameliorate EAN by inducing the release of immunosuppressive molecules such as prostaglandin E or apolipoprotein E. Our findings implicate IFN-gamma in the pathogenesis of autoimmune diseases of the peripheral nervous system.
The effects of thalidomide and, for reference, dexamethasone on T- and B-cell functions were assayed in vitro in Lewis rats with experimental allergic neuritis induced by active immunization with bovine peripheral nerve myelin (BPM) and complete Freund's adjuvant. Thalidomide and dexamethasone at the concentration ranges 10(-5)-10(-7) g/ml and 4 x 10(-5)-4 x 10(-9) g/ml, respectively, both inhibited phytohemagglutinin (PHA)- and BPM-induced T-cell proliferation as well as levels of PHA- and BPM-reactive interferon (IFN)-gamma-secreting cells, reflecting the suppression of Th1-like cells. The effect of dexamethasone was most pronounced on PHA-induced T-cell proliferation and IFN-gamma secretion, whereas the effect of thalidomide was most pronounced on BPM-induced T-cell proliferation and IFN-gamma secretion. Thalidomide reduced the B-cell responses to both BPM and Mycobacterium tuberculosis purified protein derivative, but to a lesser extent than dexamethasone. The in vitro design described could be useful to evaluate compounds with putative immunomodulatory activities. The inhibitory effects of thalidomide on autoantigen-induced Th1-cell functions may warrant the use of this substance in T-cell-mediated autoimmune diseases.
This study explores nasal administration of myelin basic protein (MBP) as a potential means of inducing tolerance to relapsing experimental autoimmune encephalomyelitis (PR-EAE), an experimental multiple sclerosis (MS) model that was induced in DA rats by immunization with rat spinal cord homogenate and incomplete Freund's adjuvant. DA rats received a total dosage of 0, 6, 60, 600 μg/rat of bovine MBP on ten consecutive days prior to immunization. EAE with typical course was observed in control rats receiving only PBS nasally, and in rats receiving 6 μg/rat of MBP. Rats receiving 60 μg/rat of MBP developed acute EAE but no relapse during 60 days of observation post immunization (p.i.). Only one of eight rats receiving 600 μg/rat of MBP developed slight, transient EAE. This protection was confirmed at the histology level and was associated with decreased levels of MBP-reactive IFN-γ secreting Th1-like spleen cells on day 13 and 60 p.i. Rats receiving 60 and 600 μg/rat of MBP showed decreased serum anti-MBP IgG2b antibody levels on day 60 p.i., and rats receiving 600 μg/rat of MBP had marginally increased anti-MBP IgG1 antibody levels in serum compared to control EAE rats. Cytokine mRNA profiles in central nervous system (CNS) and spleen mononuclear cells were evaluated. Dose-dependent reduction of TNF-α mRNA expression were observed both in CNS and in splenocytes. Increased IL-4 and TGF-β mRNA expression were observed in CNS of low (6 μg/rat) and median (60 μg/rat) dose of MBP tolerized rats and in splenocytes of rats tolerized with 600 μg/rat of MBP. We conclude that nasal administration of MBP in DA rat prevents EAE induced by immunization with whole rat spinal cord homogenate that, besides MBP, contains multiple antigenic myelin proteins. A mechanism involving MBP-reactive regulatory cells expressing IL-4 and TGF-β mRNA acts as part in the induction of this tolerance.
LEW rats were treated intravenously with recombinant rat interferon-gamma (IFN-gamma) for 3 days to achieve intravascular accumulation, proliferation, and activation of monocytes. Monocytes, defined by their expression of the ED1, ED9, and Ox41 antigens, were recovered from the vasculature by perfusion with PBS/EDTA, subsequently depleted of erythrocytes and granulocytes by Percoll density gradient centrifugation, and analyzed by flow cytometry and immunocytology. In untreated and control-infused specified pathogen-free (SPF) rats, lymphocytes and monocytes formed overlapping cell populations with respect to size and internal granularity. At least two intravascular monocyte subsets, probably central and marginating cells, were distinguished by their size and differential expression of CD43, CD4, CD11a, CD18, and L-selectin. It is interesting to note that a fraction of the monocytes in normal and control-infused animals carried the NKR-P1A molecule. IFN-gamma treatment provoked a duplication of monocyte size and granularity. Both the number of positive monocytes and the level of expression of NKR-P1A strongly increased after IFN-gamma infusion, whereas CD43 (leukosialin) and CD4 were impressively down-regulated. NKR-P1A+ L-selectin+ CD43low CD4- monocytes also occur in the vasculature of rats during immune reactions in vivo. We speculate that these cells are involved in organ damage and that their number is controlled by activation-induced cell death within the vessels.
Monocytes and macrophages perform vital functions ranging from defense against certain bacteria and parasites to tissue repair and antigen presentation. There have been numerous reports about macrophages in cell culture, but the phenotype and function of monocytes and their responses to cytokines in vivo are almost unknown. We have thus investigated the phenotype of monocytes recovered from the vasculature of untreated LEW rats and from rats that had received a continuous 3-day infusion of recombinant rat interferon (IFN)-gamma or a control infusion.
Age-related alterations in the immune system, including changes in lymphocyte subset composition, result in changes of cytokine patterns and might thereby influence the incidence and severity of autoimmune diseases. To investigate the age-related resistance to EAMG, an animal model for human MG, young (4-week-old) and adult (8-10-week-old) female Lewis rats were immunized with Torpedo acetylcholine receptor (AChR) and Freund's complete adjuvant (FCA). Adult Lewis rats showed severe weight loss and progressive muscular weakness after immunization, while young rats developed minor clinical signs of EAMG after a prolonged interval post-immunization. By comparison with adult rats, the young had lower AChR-specific T and B cells responses, and less muscle AChR loss. In situ hybridization performed on mononuclear cells (MNC) from lymph nodes revealed that young rats had lower levels of AChR-specific IFN-gamma, IL-4 and IL-10 mRNA-expressing cells compared with adult rats. Since IFN-gamma, IL-4 and IL-10 promote the development of EAMG, the low expression of these cytokines might contribute to EAMG resistance in young Lewis rats.
Chemo-immunotherapy for the treatment of cancer, whilst promising from a preclinical and clinical perspective, remains limited by a lack of clear understanding of the in vivo antitumour mechanisms of this multi modality strategy. There is now strong evidence that systemic immunological parameters do not correlate with therapeutic activity. In contrast, information on therapy related immunological change at the tumour site is scarce. Having previously demonstrated that the therapeutic activity of doxorubicin chemotherapy can be significantly augmented by the co-administration of two cytokines, interleukin-2 (IL-2) and interferon-gamma (IFN-gamma), the objective of the present study was to investigate the mechanism of action of this enhanced therapeutic activity by characterising the effect of single, double and triple agent therapy upon local tumour immune parameters. Twenty-four hours after the administration of treatment to WAG rats bearing solid tumour implants of a colonic adenocarcinoma, the extent of tumour infiltration in response to the therapy was assessed in haematoxylin and eosin stained tumour sections. Treatment with doxorubicin/IL-2/IFN-gamma or IL-2/IFN-gamma was associated with a marked augmentation of the size of the tumour infiltrate (P < 0.001), as compared to untreated tumours or to those treated with any other single or double agent combination. Phenotypic evaluation of the tumour infiltrate using immunoperoxidase stained tumour sections revealed that a considerable proportion of the infiltrating cells were T cells and macrophages, whilst B cells were not detected in significant numbers. Although this phenotypic profile was not qualitatively influenced by therapy, marked quantitative differences were observed. Most notably, tumours treated with either doxorubicin/IL-2/IFN-gamma or IL-2/IFN-gamma exhibited a significant increase in the numbers of CD25+ infiltrating cells (P < 0.001). These changes in the tumour immunological response closely paralleled the therapeutic responses described previously. Thus, the enhanced therapeutic activity of the triple agent regimen may result from a profound augmentation of the size of the tumour infiltrate, together with a similar increase in the numbers of activated infiltrating cells. This study supports the concept that the immune response within the tumour is the appropriate site for investigations into the immunological antitumour mechanisms of immunotherapy and chemo-immunotherapy.
An example for the bidirectional exchange of activating signals between a pathogen and immunocompetent cells in the host is presented. Trypanosoma brucei, which include subspecies that cause African sleeping sickness, secrete a molecule that triggers lymphocytes to produce interferon (IFN)‐γ. We now report that proliferation of T. brucei is stimulated in axenic cultures by IFN‐γ. The growth‐enhancing effect on the pathogen is inhibited by anti‐IFN‐γ receptor (R) antibodies and does not occur after exposure to other cytokines, i.e. IFN‐α, IFN‐β and tumor necrosis factor (TNF)‐α. While rodent‐pathogenic T. brucei strains are stimulated by rat IFN‐γ, human pathogenic strains are more potently stimulated by human IFN‐γ. Rat and human IFN‐γ can partially block each others effects. Mice with disrupted IFN‐γ genes have reduced parasitemia and prolonged survival, while the outcome is reversed in mice that lack the IFN‐γR gene.
The mechanism of action underlying the beneficial effect of IFN-beta in multiple sclerosis (MS) is not understood. To date, little information is available on the effects of IFN-beta in experimental autoimmune encephalomyelitis (EAE), the animal correlate of the human disease MS. Therefore, we investigated the effects of recombinant rat IFN-beta (rrIFN-beta) on EAE in Lewis rats with emphasis on a treatment regimen during the paralytic phase of the disease. The results indicated that rrIFN-beta dose-dependently inhibited disease activity with complete prevention at a s.c. dose of 300,000 U/day, provided that treatment was continued for 3 wk. Discontinuation of treatment on day 17 postimmunization resulted in a protracted and relapsing disease course with strongly enhanced clinical severity. Detailed immunohistology of central nervous system (CNS) tissue of protected animals revealed an almost complete absence of CNS lesions and a >90% reduction in the number of infiltrating leukocytes. Accordingly, isolation of mononuclear cells from spinal cord tissue of successfully treated EAE rats revealed a reduction of approximately 95% in the number of cells that produce IFN-gamma in response to the encephalitogenic peptide MBP63-88. Furthermore, rrIFN-beta significantly enhanced serum corticosterone levels, which showed an inverse relationship with disease activity. We show that rrIFN-beta can have both beneficial and detrimental effects on disease activity dependent on the timing and the duration of treatment. Beneficial effects on EAE are associated with inhibition of the extravasation of blood-derived mononuclear cells in the CNS.
IL-4 is the critical regulatory cytokine that preferentially promotes a Th2 type of immune response. In certain models of organ-specific autoimmune diseases in which Th1 cells are implicated in the disease process, treatment with IL-4 has been shown to confer protection by deviating the immune response toward a Th2 type. In this study, we addressed the role of IL-4 in experimental autoimmune uveoretinitis, a prototypic Th1-dependent disease induced in susceptible animals following immunization with soluble retinal Ag. Interestingly, treatment of Lewis rats with IL-4 exacerbated experimental autoimmune uveoretinitis, and simultaneous treatment with neutralizing anti-IL-4 Abs attenuated this increase in the severity of the disease. Ex vivo analysis of cytokines produced in response to the immunizing Ag showed an enhancement in the levels of IFN-gamma, TNF-alpha, and nitric oxide following IL-4 treatment. In vitro, IL-4 augmented the production of IFN-gamma by Con A-stimulated splenocytes in a dose-dependent manner. At low concentrations of IL-4, IFN-gamma production was enhanced, while at higher concentrations this production was inhibited. The specificity of the induction of IFN-gamma by IL-4 was confirmed by neutralizing the activity of IL-4 with anti-IL-4. Taken together, the results herein reported demonstrate that IL-4 can induce the production of IFN-gamma and of inflammatory cytokines under certain conditions, and indicate that IL-4 can exert a dose-dependent differential effect on the induction of immune responses and on autoimmunity.
SUMMARYThe subepithelial immune deposits of Dorus Zadel Black (DZB) rats with mercury-induced membranous nephropathy consist of autoantibodies directed to laminin P1 and of complement. The animals develop massive proteinuria within 10–14 days which is associated with obliteration of foot processes of glomerular visceral epithelial cells (GVEC), or podocytes. Previous studies indicate that these autoantibodies are probably not the sole mediator of proteinuria and GVEC damage. In this study we investigated whether circulating or macrophage-derived cytokines can contribute to the GVEC changes as detected in vivo. In vivo at the height of the proteinuria. increased intraglomerular IFN-γ immunoreactivity was found. In diseased rats a five-fold increase in intraglomerular macrophages was found, but we could not detect intraglomerular IFN-α, IFN-β, IL-1β or tumour necrosis factor-alpha (TNF-α) by using immunohistology. Subsequently, we exposed cultured GVEC to these cytokines to investigate their cytotoxic effects on several physiological and structural parameters. IFN-γ and IL-4 were the only cytokines that exerted toxic effects, resulting in a rapidly decreased transepithelial resistance of confluent monolayers. which was closely associated with altered immunoreactivity of the light junction protein ZO-I. IL-4 also affected vimentin and laminin immunoreactivity. IFN-γ and IL-4 only interfered with monolayer integrity when added to the basolateral side of the GVEC, indicating specific (receptor-mediated) effects. Only IL-4 decreased the viability of the cells, and treated monolayers demonstrated an increased passage of the 44-kD protein horseradish peroxidase. From our experiments we concluded that IFN-γ subtly affected monolayer integrity at the level of the tight junctions, and that IL-4 additionally induced cell death. We hypothesize that the toxic effects of the cytokines IFN-γ and IL-4 as seen with cultured podocytes are necessary together with the autoantibodies. for the ultimate induction of proteinuria in mercury nephropathy in the DZB rat.
Cellular immune responses to filarial antigens were examined in persons before and 1 year after beginning treatment with diethylcarbamazine (DEC). The subjects (17 microfilaremics, 13 asymptomatic amcirofilaremics, and 13 with elephantiasis) had not responded to Brugia malayi adult worm antigen (BmA) before chemotherapy. T cell proliferative responses to BmA improved significantly after therapy in the 3 clinical groups (P < .05) but was highest in the elephantiasis patients and asymptomatic amicrofilaremics. Cytokine release profiles after stimulation with parasite antigen were analyzed, Production of interferon (IFN)-gamma by BmA-stimulated mono-nuclear cells increased significantly after DEC treatment (geometric mean, 39.6-55.7 U/mL; P < .05), largely due to improved responses in elephantiasis patients and asymptomatic amicrofilaremics, In contrast, BmA-induced interleukin (IL)-4 release did not change significantly in these same patients after treatment. Thus, both microfilaremic and amicrofilaremic infections with B. malayi are associated with similar down-regulation of proliferative T cell function and IFN-gamma release.
Recently we have shown that the generation of IFN-γ-producing cells (IFN-γ pc) were severely suppressed in cultures of Con A-stimulated splenocytes obtained from HgCl2-exposed BN but not Lewis rats. Since BN rats develop a TH2-biased polyclonal autoimmune syndrome upon exposure to HgCl2, whereas Lewis rats exhibit a resistant phenotype, possibly by generating a protective TH1 response, downregulation of IFN-γ production may be a critical denominator in the disease process. In the present study it was found that this suppression of IFN-γ production was almost completely antagonized by NG-monomethyl-L-arginine (NMMA), a competitive inhibitor of nitric oxide (NO) synthesis, and potentiated by the depletion of splenic erythrocytes, a cell type known to effectively scavenge NO. Determination of nitrite (NO-2), the stable metabolite of NO, revealed that splenocytes obtained from HgCl2-exposed BN rats produced two- to threefold higher levels of NO-2 than those from HgCl2-exposed Lewis rats and approximately two times more than saline-injected controls. Upon depletion of erythrocytes, splenocytes from normal BN and Lewis rats produced similar amounts of nitrite which was enhanced approximately twofold after in vivo exposure to HgCl2 up to 5 days after initiation of HgCl2 exposure. At Day 7, NO release by Lewis splenocytes returned to baseline levels, whereas NO release by BN splenocytes remained high up to the effector phase of the autoimmune disease. Flow cytometric analysis and spleen cell counts revealed marked differences in the splenic subset composition between Lewis and BN rats with an anomalous low frequency of erythrocytes and CD8+ T cells and a strikingly high number of B cells in the BN spleen. Our findings are in confirmation with a model in which HgCl2 triggers the upregulation of NO synthesis in the spleen. The enhanced production of NO is insufficiently scavenged by erythrocytes in the BN spleen resulting in a suppressive effect on the production of IFN-γ. The sustained inhibitory effect of NO on IFN-γ production may promote the development of a TH2-biased autoimmune syndrome in BN rats.
Interferon-alpha (IFN alpha), -beta, and -gamma are well known for their antiviral, antiproliferative, and immunoregulatory activities. Although several studies suggest an involvement of IFNs in the spermatogenic process, nothing is known about the possible production of these molecules within the testis. Moreover, the antiviral capabilities of testicular cells have not yet been explored despite their importance in the context of sexually transmissible diseases. Using reverse transcription-polymerase chain reaction, a cytopathic inhibition micromethod assay, and an enzyme-linked immunosorbent assay, the present study demonstrates for the first time that IFN alpha and -gamma are produced by testicular cells. IFN alpha protein and corresponding messenger RNA are expressed by peritubular, Sertoli, and germ cells. In vitro, IFN alpha production by Sertoli cells, peritubular cells, and early spermatids was inducible by the Sendai virus, whereas pachytene spermatocyte IFN alpha production was not triggered by this virus. Of all the testicular cell types tested, Sertoli cells by far produced the highest concentrations of IFN alpha/beta, followed by peritubular cells. Both IFN gamma messenger RNA and IFN gamma protein were found in early spermatids, but, in contrast, were not produced by peritubular cells, Sertoli cells, or pachytene spermatocytes. In conclusion, our study establishes the cellular distribution of IFNs within the seminiferous tubules and provides the basis for research into the possible involvement of IFNs in regulation of the spermatogenic process. To the best of our knowledge, our results afford the first insight on how the testicular antiviral defense system is organized.
The repeated administration of low-dose HgCl2 to brown Norway (BN) rats induces an autoimmune syndrome which is characterized by polyclonal B cell activation, high-level synthesis of IgE and IgG1, and massive proteinuria. Data have been presented suggesting that during disease development there is a preferential expansion of CD4+ T cells belonging to the TH2 subset. In the present study it was found that aged BN rats are far less susceptible to the immunopathological effects of HgCl2 compared to their younger counterparts. Whereas rats at 10 weeks of age develop high-level proteinuria upon three repeated injections with HgCl2, animals at 18 to 24 months of age do not release urinary protein under these conditions and develop low-level proteinuria with a delayed onset after five repeated injections with HgCl2. FACScan analysis of splenocytes from old and young rats revealed a defined increase in the frequency of CD45RB(RC)+/CD4+ T cells in the splenocyte population of older rats, suggesting an age-related shift to a more TH1-like phenotype. Moreover, splenocytes of aged rats generated a threefold higher number of IFN-γ-producing cells than those of young rats upon polyclonal activation in vitro. The administration of neutralizing anti-rat IFN-γ mono- and/or polyclonal antibodies to aged BN rats just prior to HgCl2 exposure significantly augmented IgE and IgG1 serum levels and exerted a small but significant stimulatory effect on proteinuria in the initial stage but not in the more advanced stages of the renal disease. When antibodies were given 7 days after the beginning of HgCl2 exposure no stimulatory effect on both IgE/IgG1 levels and proteinuria was observed. The data indicate that splenic T cells of aged BN rats possess a higher capacity to release IFN-γ than those of young rats and that this cytokine functions to downregulate IgG1 and IgE synthesis in HgCl2-exposed BN rats. The findings further suggest that IFN-γ plays a regulatory role in the development of glomerulonephritis.
We investigated whether enumeration of lymphokine-secreting T cells can be used as a quantitative measure to determine the immunogenicity of foreign proteins in rhesus monkeys. In addition, it was assessed whether this approach can supplement and/or substitute for the well-established lymphoproliferation assay. Two candidate vaccine proteins (e.g., HIV-1 gp120 and HSV-2gD) were used as model antigens for immunization, PBMCs from immunized animals were antigenically stimulated and evaluated on their proliferative capacity and lymphokine release at the single cell level. The experiments showed a close quantitative correlation between antigen-triggered proliferative responses and the antigen-induced generation of Il-2 and IFN-gamma producing cells (pc). Il-4pc were found to appear relatively late after the initiation of antigen exposure. The data indicate that ELISPOT assays provide valuable tools for the assessment of the antigenicity of foreign proteins in vivo.
The present study concerns the release of the proinflammatory cytokines interleukin-1 beta (IL-1 beta) and tumor necrosis factor alpha and of the anti-inflammatory cytokine IL-10 by human leukocytes in whole blood during stimulation with Streptococcus pneumoniae and the effects of various xanthine derivates, i.e., pentoxifylline (PTX), caffeine, and theofylline, and of dexamethasone (DXM). All three xanthine derivates and DXM inhibited the release of tumor necrosis factor alpha, PTX being the most effective. PTX, theofylline, and DXM inhibited the release of IL-1 beta, but caffeine did not affect IL-1 beta release. The release of IL-10 was significantly reduced by PTX at 24 h and by caffeine at 48 h, but DXM increased the release of this cytokine. In sum, the results of this study demonstrate that DXM inhibits only the release of proinflammatory cytokines but not of the anti-inflammatory cytokine IL-10 by human leukocytes, while PTX is the most potent inhibitor of both proinflammatory and anti-inflammatory cytokines.
An interferon (IFN)-gamma immunoreactive molecule, localized to small neurons in peripheral sensory ganglia (N-IFN-gamma), has been detected with two mouse monoclonal antibodies (DB1 and DB16) directed against different epitopes of rat IFN-gamma. To define N-IFN-gamma with regard to its protein characteristics and bioactivities, DB1 and DB16 were used to purify N-IFN-gamma from rat trigeminal ganglia in a two-step sequential antibody-affinity procedure. Sodium dodecylsulfate polyacrylamide gel electrophoresis (PAGE) and silver staining of purified N-IFN-gamma displayed three bands with an approximate molecular mass of 66, 62 and 54 kDa. The N-IFN-gamma bioactivity was confined to the protein stained on gel when native material was run on PAGE. Biological effects of pure N-IFN-gamma were examined and compared with those of lymphocyte-derived recombinant IFN-gamma. N-IFN-gamma had antiviral effects in vitro and induced major histocompatibility complex class I and II antigens on macrophages and in cells in skeletal muscle cell cultures. N-IFN-gamma also stimulated myoblast proliferation and affected cholinergic receptor distribution on myotubes similar to recombinant IFN-gamma. Both molecules potently stimulated Trypanosoma brucei brucei growth. These data suggest that, although N-IFN-gamma is a protein distinct from lymphocyte-derived IFN-gamma, the two molecules have enough structural similarities to allow for antibody recognition of at least two epitopes, and action on similar target structures on both parasite and mammalian cells.
The effect of intrathecal (i.t.) injection of the cytokine interferon-gamma (IFN-gamma) on the spinal nociceptive flexor reflex was examined in decerebrate, spinalized, unanesthetized rats. IFN-gamma elicited an initial intense, brief facilitation of the flexor reflex followed by a sustained reflex facilitation lasting 40 +/- 5 min (range 20-65 min). The initial and prolonged reflex facilitations by IFN-gamma were partially and totally blocked, respectively, by i.t. pretreatment with nitro-L-arginine-ester, an inhibitor of nitric oxide synthase, at doses which did not influence spinal cord blood flow. Spinal application of IFN-gamma produced powerful and prolonged facilitation of the flexor reflex, possibly reflecting a hyperalgesic action of this cytokine. The facilitatory effect of IFN-gamma was mediated, at least in part, by the activation of the L-arginine-nitric oxide pathway. Thus, IFN-gamma released in the CNS may participate in eliciting pain and hyperalgesia in infectious or neuroinflammatory diseases where there is increased production of this cytokine.
We investigated the role of tumor necrosis factor alpha (TNF-alpha) in the pathogenesis of rat cytomegalovirus (RCMV) infection. TNF-alpha levels found in the sera of radiation-immunosuppressed rats in the course of infection (>350 pg/ml) correlated with the development of RCMV disease. Administration of anti-TNF-alpha antibodies strongly reduced the severity of pneumonia and led to a reduction in virus titers. In immunocompetent rats, anti-TNF-alpha antibodies also significantly suppressed viral replication. Conversely, administration of TNF-alpha augmented RCMV replication and aggravated the disease signs. In vitro, TNF-alpha enhanced RCMV replication in the macrophage, whereas a reduction of viral replication was observed in fibroblasts, indicating that the effect on viral replication is cell type specific. Besides activation of viral replication and exacerbation of RCMV disease, TNF-alpha also favored lymphoid and hematopoietic tissue reconstitution after irradiation, which may contribute to antiviral resistance and survival. This finding demonstrates the protean nature of TNF-alpha, with both beneficial and adverse effects for the host. Our results suggest that TNF-alpha plays an important role in modulating the pathogenesis of RCMV infection.