|Home | About | Journals | Submit | Contact Us | Français|
Viral inhibitors of host programmed cell death (PCD) are widely believed to promote viral replication by preventing or delaying host cell death. Viral FLIPs (Fas-linked ICE-like protease [FLICE; caspase-8]-like inhibitor proteins) are potent inhibitors of death receptor-induced apoptosis and programmed necrosis. Surprisingly, transgenic expression of the viral FLIP MC159 from molluscum contagiosum virus (MCV) in mice enhanced rather than inhibited the innate immune control of vaccinia virus (VV) replication. This effect of MC159 was specifically manifested in peripheral tissues such as the visceral fat pad, but not in the spleen. VV-infected MC159 transgenic mice mounted an enhanced innate inflammatory reaction characterized by increased expression of the chemokine CCL-2/MCP-1 and infiltration of γδ T cells into peripheral tissues. Radiation chimeras revealed that MC159 expression in the parenchyma, but not in the hematopoietic compartment, is responsible for the enhanced innate inflammatory responses. The increased inflammation in peripheral tissues was not due to resistance of lymphocytes to cell death. Rather, we found that MC159 facilitated Toll-like receptor 4 (TLR4)- and tumor necrosis factor (TNF)-induced NF-κB activation. The increased NF-κB responses were mediated in part through increased binding of RIP1 to TNFRSF1A-associated via death domain (TRADD), two crucial signal adaptors for NF-κB activation. These results show that MC159 is a dual-function immune modulator that regulates host cell death as well as NF-κB responses by innate immune signaling receptors.
Successful immunity against pathogenic challenges is central to the survival of all organisms. Metazoans employ a wide array of innate and adaptive immune responses to control various pathogens. In response, pathogens have developed various strategies to evade detection and elimination by the immune system. Programmed cell death (PCD) plays an important role in host defense against pathogens by directly eliminating infected cells to limit the viral factory. A role for host cell death in antiviral responses is highlighted by the identification of viral inhibitors of apoptosis (3). In addition to apoptosis, nonapoptotic PCD pathways, such as necrosis and autophagy have recently been shown to participate in host defense against pathogens (31, 44). For instance, we recently showed that genetic ablation of an essential necrosis mediator, RIP3, resulted in severely impaired innate immune responses against vaccinia virus (VV) infection characterized by the lack of virus-induced tissue necrosis and inflammation (11). In addition, certain vFLIPs (viral Fas-linked ICE-like protease [FLICE; caspase-8]-like inhibitor proteins) are potent inhibitors of programmed necrosis (6, 8). These results indicate that host PCD machineries play important roles in controlling the viral factory and dissemination of the virus within the infected host.
Despite the widely accepted view that inhibition of host cell death is an important viral immune evasion strategy, relatively few in vivo studies have been performed to directly test this hypothesis. This is due partly to the lack of suitable animal models in which specific components of host apoptotic machinery are inhibited. For instance, germ line inactivation of many of the components of the PCD machinery, such as Fas-associated via death domain (FADD) and caspase-8, resulted in embryonic lethality (50, 55, 57), thus preventing in vivo virus infection studies from using these animal models. Another approach that was widely used was transgenic expression of viral apoptosis inhibitors, such as poxvirus CrmA, baculovirus p35, and vFLIPs. However, since expression of these inhibitors was restricted mostly to the lymphoid compartment (26, 28, 34, 46, 51, 54, 58), they do not permit evaluation of the role of host cell death in the parenchyma in antiviral responses. Cell death in the stromal compartment could impact the innate inflammatory reaction, cross-priming of antigens, and viral dissemination to other tissues. Because cells in the parenchyma are the primary targets for many virus infections, it is important to determine the contribution of cell death in the parenchyma in antiviral responses.
The vFLIPs were first identified as inhibitors of caspase-dependent apoptosis. They share homology with caspase-8 and caspase-10 in the tandem death effector domains (DEDs) at the amino termini. However, vFLIPs lack the caspase enzyme domain at the carboxyl termini. Thus, binding of vFLIPs to FADD and caspase-8/-10 via DED-mediated homotypic interaction led to inhibition of FasL-, tumor necrosis factor (TNF)-, and TNF-related apoptosis-inducing ligand (TRAIL)-induced apoptosis (18, 19). Importantly, certain vFLIPs, including MC159 and E8, are also potent inhibitors of programmed necrosis induced by TNF-like death cytokines (8). These results suggest that viral inhibitors could inhibit multiple host PCD pathways to avoid elimination by the host immune system.
In order to determine the effect of vFLIPs on host responses against viral infections, we generated transgenic mice expressing vFLIP MC159 under the control of the ubiquitous H2-Kb promoter (53). We previously showed that transgenic expression of MC159 did not alter lymphocyte functions and development, but rather caused a mild form of lymphoproliferation that resembled that seen with the lpr mice with Fas/CD95/Apo-1 mutations (53). Here, we show that MC159 transgenic mice exhibited enhanced innate immune responses to VV infections, which led to enhanced viral clearance in peripheral tissues. Surprisingly, the enhanced control of VV production was not due to enhanced lymphocyte survival. Rather, VV-induced expression of the chemokine CCL-2/MCP-1 was highly elevated in MC159 transgenic mice and was accompanied by enhanced recruitment of γδ T cells to peripheral tissues. MC159 promotes the binding between TRADD and RIP1, two crucial signal adaptors for NF-κB activation. Consequently, MC159 transgenic fibroblasts exhibited enhanced NF-κB activation to TNF and Toll-like receptor 4 (TLR4) stimulation. These results reveal a previously unappreciated effect of MC159 on NF-κB activation and indicate that viral cell death inhibitors could impact innate immune responses through mechanisms beyond cell death regulation.
The MC159 transgenic mice have been described (53). For all experiments, unless otherwise stated, mice between 12 to 14 weeks of age were used. All experiments were performed according to protocols approved by the University of Massachusetts Medical School animal care and use committee. For VV infections, one million PFU of the Western Reserve (WR) strain were used to infect mice via the intraperitoneal route. Viral titers were determined by plaque assays using Vero cells and crystal violet staining as described before (8). Briefly, the whole organ/tissue was ground up in 1 ml of RPMI 1640 medium. Viral titers were determined by serial dilutions of the tissue supernatants. The plaque counts obtained were used to determine the viral load for the whole organ by using the formula N = Y/(vx), where N is the final titer per ml, Y is the number of plaques at the dilution used, v is the volume plated in ml, and x is the dilution factor. For lipopolysaccharide (LPS)-induced septic shock, mice were injected with 50 μg/kg of LPS (Sigma) and 1,000 mg/kg of d-galactosamine (Sigma) via the intraperitoneal route. Sera and liver tissues were collected 5 h after LPS administration for alanine aminotransferase (ALT) and caspase assays. The number of apoptotic nuclei was scored in a double-blind manner by averaging the number of condensed nuclei in six different fields. For bone marrow chimeras, 4- to 5-week-old mice were irradiated (950 rads) and reconstituted with 5 × 106 bone marrow cells depleted of B and T cells. Four months later, the chimeras were challenged with VV as described above and analyzed for lymphocyte infiltration 24 h postinfection and viral titers 4 days postinfection.
One hundred micrograms of liver cell extracts were diluted in 200 μl of iTFB buffer (10% sucrose, 30 mM HEPES [pH 7.4], 10 mM CaCl2 and 5 mM dithiothreitol [DTT]) containing 25 μM N-acetyl-Asp-Glu-Val-Asp-7-amino-4-methylcoumarin (Ac-DEVD-AMC) (Biomol). The release of fluorescent AMC was measured with a Fluostar multiwell plate reader at 340 nm for excitation and 450 nm for emission over 3 h. The rate of conversion was calculated by including 25 μM free AMC as the control.
For fat pad lymphocyte isolation, visceral fat pads were incubated in the enzyme digestion buffer (0.5 mg/ml collagenase type II [Sigma], 100 units/ml type I DNase [Sigma], and 10% fetal calf serum in Hank's buffered saline solution [Invitrogen]) for 45 min at 37°C. For isolation of splenic dendritic cells (DCs), spleens were incubated at 37°C for 30 min in 2 mg/ml collagenase D (Roche) dissolved in 10 mM HEPES (pH 7.4), 150 mM NaCl, 5 mM KCl, 1 mM MgCl2, and 1.8 mM CaCl2. For terminal deoxynucleotidyltransferase-mediated dUTP-biotin nick end labeling (TUNEL) staining, lymphocytes were incubated in a 37°C CO2 incubator for 1 h prior to staining. For annexin V staining, lymphocytes were incubated in a 37°C CO2 incubator for 5 h prior to staining. For annexin V staining of splenic DCs, cells were incubated at 37°C for 60 min before staining.
Lymphocytes were stimulated with 10 μg of the VV-specific peptide B8R, K3L, or A47L (32) in a 37°C CO2 incubator for 5 h. For intracellular cytokine staining, cells were fixed using a BD cytofix/cytoperm kit (BD Pharmingen) per the manufacturer's instructions. Staining antibodies were obtained from BD Pharmingen or eBioscience. Peptides were synthesized by EZBiolab.
Mouse embryonic fibroblasts (MEFs) were generated from day 12 to 14 embryos. BMDCs were obtained by culturing bone marrow cells in 10 ng/ml granulocyte-macrophage colony-stimulating factor (GM-CSF) and 5 ng/ml interleukin-4 (IL-4) for 7 days (43). Bone marrow-derived macrophages (BMMs) were obtained by culturing bone marrow cells in L929-conditioned medium for 7 days (15). Nonadherent cells were carefully transferred to a new culture dish, rested for 24 h, and stimulated with 0.1 μg/ml LPS (for BMDCs and BMMs). Total-cell extracts were prepared by lysis in NP-40 lysis buffer (150 mM NaCl, 10 mM Tris [pH 7.5], 1% Nonidet P-40 and 1× Complete protease inhibitor cocktail [Roche]). Cell extracts were resolved with 4 to 12% NuPAGE gels (Invitrogen), blotted onto nitrocellulose membranes, and probed with antibodies to phosphorylated IκBα (p-IκBα), IκBα (Cell Signaling), or β-actin (BD Pharmingen). The antibody against E3L was a kind gift from J. W. Yewdell (56). For electrophoretic mobility shift assays (EMSA), nuclear extracts binding to 32P-labeled oligonucleotides were performed as described previously (7). Oligonucleotide sequences used were 5′-AGTTGAGGGGACTTTCCCAGGC-3′ and 5′-GCCTGGGAAAGTCCCCTCAACT-3′ (for NF-κB) and 5′-GAGGTGGGTGGAGTTTCGCG-3′ and 5′-CGCGAAACTCCACCCACCTC-3′ (for the GT box).
Transfections with 293T cells were performed using Fugene 6 transfection reagent (Roche) per the manufacturer's instructions. For measuring NF-κB activation, cells were transfected with the plasmids pNF-κB-luc (Stratagene) and pTK-βgal (Promega). Luciferase activity was normalized to β-galactosidase activity by using kits from Promega. For fluorescence-activated cell sorter (FACS) analysis, a cyan fluorescent protein (CFP) reporter was amplified by PCR to replace the luciferase gene in the pNF-κB-luc vector. For immunoprecipitations, cells were lysed in 150 mM NaCl, 20 mM Tris (pH 7.5), and 1% NP-40 supplemented with protease inhibitor cocktails (Roche). Antibodies against TRADD and hemagglutinin (HA) were purchased from Millipore and Covance, respectively.
Total RNA was isolated from visceral fat pads by using an RNeasy kit from Qiagen. cDNA was generated using an Omniscript reverse transcription (RT) kit (Stratagene). Real-time quantitative PCR was performed using SYBR green (Bio-Rad). Signals were normalized to 18S RNA. Primer sequences used were as follows: CCL-2, 5′-TGCTACTCATTCACCAGCAA-3′ and 5′-GTCTGGACCCATTCCTTCTT-3′; IL-6, 5′-CGGAGAGGAGACTTCACAGA-3′ and 5′-CCAGTTTGGTAGCATCCATC-3′; and 18S RNA, 5′-TGGTGGAGGGATTTGTCTGG-3′ and 5′-TCAATCTCGGGTGGCTGAAC-3′.
Statistically analyses were performed using unpaired Student's t tests. One-way analysis of variance was used for Fig. 6B to D.
We have previously shown that transgenic expression of the viral cell death inhibitor MC159 under the ubiquitously expressed H2-Kb promoter results in protection of lymphocytes from cell death induced by TNF-like cytokines (53). Since MC159 was also expressed in nonhematopoietic tissues (53), we asked whether parenchymal tissues in the MC159 transgenic mice were also protected from cytokine-induced PCD. To this end, we used an LPS-induced sepsis model, because TNF plays a central role in liver cell injury in this model (16, 37, 41). Transgenic expression of MC159 protected hepatocytes from LPS-induced liver injury as determined by a reduced serum ALT level (Fig. (Fig.1A)1A) and reduced caspase-3 activity in liver cell extracts (Fig. (Fig.1B)1B) compared with results for wild-type littermates. Moreover, apoptosis marked by nuclear condensation was completely absent in the MC159 transgenic hepatocytes (Fig. (Fig.1C,1C, compare panels c and d, and Fig. Fig.1D).1D). However, LPS-induced mortality in MC159 transgenic mice was comparable to that of their wild-type counterparts (Fig. (Fig.1E).1E). This is consistent with the requirement for other cytokines, including IL-1β, in LPS-induced systemic organ failure (52). These results show that transgenic expression of MC159 inhibited death cytokine-induced PCD in the liver but did not inhibit signaling by other cytokines.
Having established that MC159 expression protected cells in the parenchyma against death cytokine-induced PCD, we examined whether antiviral responses are affected in the MC159 transgenic mice. We recently showed that TNF-mediated programmed necrosis plays a crucial role in the innate immune protection against VV infections (11). Since MC159 is a potent inhibitor of death receptor-induced apoptosis and programmed necrosis (8), we expected that MC159 expression would result in defective control of VV production. Surprisingly, viral titers of the visceral fat pads (Fig. (Fig.2A)2A) and livers (Fig. (Fig.2B)2B) of transgenic mice were significantly reduced compared with those of wild-type mice. The reduction in virus production in the transgenic mice was observed 4 days postinfection, prior to the onset of acute CD8+ T-cell responses. Importantly, virus production in the spleen was unaffected in the transgenic mice (Fig. (Fig.2C),2C), indicating that the enhanced control of viral production by the MC159 transgene was restricted to nonlymphoid tissues. The reduction in viral titers was not due to lack of productive infection, since the VV antigen E3L was readily detected in the fat pads (Fig. (Fig.2D,2D, lanes 1 to 6), livers (Fig. (Fig.2D,2D, lanes 7 to 10), and spleens (Fig. (Fig.2D,2D, lanes 11 to 14) of wild-type and transgenic mice 24 h postinfection. Furthermore, reduction of viral titers in the transgenic fat pad was also observed 5 days postinfection (Fig. (Fig.2E).2E). Thus, MC159 transgenic mice exhibited enhanced control of VV infection.
The reduction in viral replication by 4 days postinfection suggests that innate immune responses against VV were enhanced in the MC159 transgenic mice. Indeed, within 24 h postinfection, the infected livers of transgenic mice exhibited a conspicuous increase in the number of inflammatory foci compared with that of wild-type littermate controls (Fig. (Fig.3A,3A, compare panels b and d with panels c and e). A similar increase in inflammation was detected in the visceral fat pads of the transgenic mice (Fig. (Fig.3B,3B, compare panels b and d with panels c and e). The chemokine CCL-2/MCP-1 and inflammatory cytokine IL-6 were highly upregulated in the visceral fat pads of VV-infected mice (Fig. 3C and D). Strikingly, the expression of the chemokine CCL-2/MCP-1 in the transgenic mice further increased about 3-fold compared with that in wild-type mice (Fig. (Fig.3C).3C). In contrast, IL-6 production was comparable in wild-type and MC159 transgenic mouse fat pads (Fig. (Fig.3D).3D). CCL-2/MCP-1 is crucial for the recruitment of innate immune effectors, including γδ T cells (13, 36). A flow cytometric analysis shows that a higher percentage of CD3+ T cells was detected in the visceral fat pads of MC159 transgenic mice (Fig. (Fig.3E,3E, compare panel a [4.23% in wild-type mice] with panel b [9.9% in transgenic mice]). Importantly, 65% of the transgenic CD3+ T cells recovered from the visceral fat pads expressed the γδ T-cell receptor (TCR) (Fig. (Fig.3F,3F, panel b). This is in contrast to CD3+ lymphocytes isolated from the fat pads of wild-type mice, which only contained 27.8% of γδ+ T cells (Fig. (Fig.3F,3F, panel a). These results indicate that γδ T-cell infiltration accounts for the enhanced inflammation detected in the peripheral tissues of the transgenic mice. The increased γδ T-cell infiltration correlated with increased killing of virus-infected cells, since cell death in the transgenic liver as measured by the serum ALT level (Fig. (Fig.3G)3G) and by caspase-3 activity (Fig. (Fig.3H)3H) was elevated compared with that in wild-type controls. These results are consistent with the crucial role of γδ T cells in the innate immune control of VV replication (45).
Since MC159 protects cells from the cytotoxic effects of TNF-like death cytokines, we next examined whether the increased γδ T-cell infiltration in peripheral tissues was attributed to their enhanced survival. Surprisingly, γδ T cells isolated from the fat pads of wild-type and transgenic mice exhibited similar levels of TUNEL staining when they were analyzed after 1 h of incubation ex vivo (Fig. (Fig.3I,3I, compare panels a and b). The low level of TUNEL staining is likely due to the efficient clearance of dying cells by professional phagocytes (1). To further evaluate if there were any differences in cell deaths in the transgenic lymphocytes, we incubated lymphocytes isolated from the fat pads of infected mice at 37°C for 5 h to allow the expression of the apoptotic marker annexin V in cells that were committed to the death program (1). Under these conditions, lymphocyte cell death levels were still indistinguishable between wild-type and transgenic mice (Fig. (Fig.3J).3J). Since MC159 inhibits only extrinsic PCD pathways, the lack of a difference between TUNEL and annexin V staining suggests that the intrinsic PCD pathway might contribute to lymphocyte cell death during VV infection. Taken together, these results suggest that the enhanced inflammation and control of viral replication in the MC159 transgenic mice were caused by increased γδ T-cell infiltration into the peripheral tissues rather than enhanced lymphocyte survival.
NF-κB is a proinflammatory transcription factor that plays crucial roles in innate inflammatory responses. Since inhibition of cell death does not appear to contribute to the enhanced inflammation observed for the MC159 transgenic mice, we explored whether MC159 might promote innate inflammatory responses through NF-κB. TNF induced a biphasic NF-κB activation with an early wave of signal within minutes followed by a more sustained, second wave of activation hours after TNF stimulation (10). This biphasic pattern of NF-κB activation was revealed by IκBα phosphorylation and IκBα degradation in wild-type and transgenic primary mouse embryonic fibroblasts (MEFs) (Fig. (Fig.4A).4A). Although early NF-κB activation in wild-type MEFs was similar to that in transgenic MEFs, transgenic MEFs exhibited a more sustained IκBα phosphorylation and a higher level of IκBα expression late during stimulation (Fig. (Fig.4A,4A, compare lanes 7 to 9 and lanes 16 to 18). Since IκBα is a transcriptional target of NF-κB (27), the higher IκBα expression level and sustained IκBα phosphorylation indicate that transgenic MEFs exhibited stronger NF-κB activation.
Enhanced NF-κB activation in the form of sustained IκBα phosphorylation was also observed when transgenic MEFs were stimulated with the TLR4 agonist LPS (Fig. (Fig.4B,4B, compare lanes 2 to 5 with lanes 7 and 8). However, unlike TNF, TLR4 stimulation causes production of cytokines, including TNF, which causes further IκBα degradation and contributed to the reduction in IκBα expression in the LPS-treated transgenic MEFs (Fig. (Fig.4B,4B, lanes 6 to 10). A similar reduction in IκBα expression was observed with LPS-treated bone marrow-derived dendritic cells (BMDCs) (Fig. (Fig.4C,4C, compare lanes 7 and 8 with lanes 15 and 16). Enhanced NF-κB activation in response to LPS stimulation in transgenic bone marrow-derived macrophages (BMMs) was further revealed by EMSA. As has been reported previously (15), untreated BMMs often exhibited strong basal NF-κB DNA binding activity (Fig. (Fig.4D).4D). While wild-type BMMs exhibited little to no NF-κB DNA binding activity 8 h after LPS treatment, strong nuclear NF-κB DNA binding activity was detected with transgenic BMMs (Fig. (Fig.4D,4D, compare lanes 4 and 9). These results show that in addition to inhibiting host cell death, MC159 could modulate host immune responses by promoting NF-κB activation.
We next sought to understand the mechanism by which MC159 promotes NF-κB activation. To this end, we first validated the NF-κB-promoting effect of MC159 by transient expression of MC159 in Jurkat cells using an NF-κB-driven CFP reporter. Using this system, we found that MC159 expression alone was sufficient to induce a modest amount of NF-κB activity (Fig. (Fig.5A,5A, compare panels a and c). Consistent with the observations for transgenic MEFs, BMDCs, and BMMs, MC159 significantly increased the TNF-induced NF-κB activity in Jurkat cells (Fig. (Fig.5A,5A, compare panels b and d). The adaptor proteins TRADD (9, 17, 38) and RIP1 (14, 23, 30) are crucial mediators for NF-κB activation for multiple receptors, including TNF receptor 1 (TNFR-1) and TLR4. We therefore tested whether MC159 might enhance NF-κB activation through TRADD and RIP1. Strikingly, we found that MC159 significantly increased the binding of RIP1 to TRADD (Fig. (Fig.5B,5B, compare lanes 6 and 7). In contrast, the binding of TRADD to another TNFR-1 signal adaptor, TNFR-associated factor 2 (TRAF2), was not affected by the presence of MC159 (Fig. (Fig.5B).5B). These results are consistent with a model in which MC159 facilitates NF-κB activation by enhancing RIP1 recruitment to TRADD and the TNFR-1 complex.
Our results differ from those in an earlier report that shows that MC159 could inhibit TNF-induced NF-κB activation (33). The discrepant results could be attributed to the different expression levels of MC159 achieved in this early report. To test whether the expression level of MC159 might influence its effect on NF-κB activation, we coexpressed RIP1 with increasing doses of MC159 in HEK 293T cells. Similar to the results for Jurkat cells (Fig. (Fig.5A),5A), expression of MC159 alone was sufficient to induce a low level of NF-κB-driven luciferase activity (Fig. (Fig.5C,5C, compare lanes 1 and 2). We found that low doses of MC159 synergized with RIP1 to promote NF-κB activation (Fig. (Fig.5C,5C, compare lanes 3 and 4). However, at higher doses of expression, MC159 suppressed NF-κB induction (Fig. (Fig.5C,5C, lanes 5 to 8). Thus, we conclude that the expression level of MC159 determines whether it enhances or inhibits NF-κB activation.
Since cells of both hematopoietic and nonhematopoietic origins exhibited increased NF-κB activation, we created radiation chimeras to determine the contribution of these compartments to the enhanced anti-VV responses observed with MC159 transgenic mice. Successful reconstitution of the hematopoietic compartment of irradiated wild-type mice with transgenic bone marrow and the reciprocal chimeras was confirmed by flow cytometry (Fig. (Fig.6A6A and data not shown). Interestingly, transgenic hosts, but not wild-type hosts, reconstituted with transgenic bone marrow exhibited enhanced control of virus production in the visceral fat pads (Fig. (Fig.6B).6B). Moreover, transgenic hosts reconstituted with wild-type bone marrow exhibited control of virus production that was similar in effectiveness to that of hosts that received transgenic bone marrow (Fig. (Fig.6B),6B), suggesting that expression of MC159 in the stromal parenchyma, but not in the hematopoietic compartment, is crucial for the enhanced innate immune control of VV infections. Furthermore, the enhanced γδ T-cell recruitment to the visceral fat pad also tracks with expression of MC159 in the nonhematopoietic parenchyma (Fig. 6C and D). Taken together, these results show that although MC159 enhanced TNF- and TLR4-induced NF-κB activation in hematopoietic and stromal cells, its expression in the parenchyma was necessary and sufficient to drive the enhanced innate inflammatory responses against VV infections.
In this report, we show that transgenic mice expressing the viral PCD inhibitor MC159 exhibited enhanced control of VV production in peripheral tissues such as the visceral fat pad and liver, but not in the spleen. The enhanced clearance of VV correlated with increased cell death in the liver and was marked by the increased expression of the chemokine CCL-2/MCP-1. This is accompanied by an early infiltration of γδ T cells, which is consistent with the role of γδ T cells in the innate immune control of VV replication (5, 45, 47). The differential protection by MC159 in peripheral tissues might reflect the requirement to recruit innate immune effector cells to peripheral sites. In contrast, the abundance of immune effectors might mitigate the requirement for chemokine-driven recruitment of γδ T cells to the infected spleen. Although MC159 is a potent inhibitor of death cytokine-induced apoptosis and programmed necrosis (4, 8, 24, 48), perforin/granzyme-mediated cell death could bypass inhibition by MC159 to eliminate virus-infected cells in the transgenic mice. This notion is supported by the normal responses of MC159 transgenic mice to lymphocytic choriomeningitis virus (LCMV) (53), whose clearance requires the coordinated function of perforin/granzyme and Fas (40). Perforin/granzyme-mediated target cell killing might also explain why MC159 transgenic hepatocytes were protected from LPS-induced liver injury mediated by TNF but remained sensitive to cell-mediated cytotoxicity during VV infections.
MC159 is known predominantly as an inhibitor of death cytokine-induced apoptosis and programmed necrosis (8). However, MC159 was unable to inhibit apoptosis induced by staurosporine (49) and by overexpression of caspase-8/FLICE (24), which bypass the death receptors. Similarly, MC159 transgenic lymphocytes exhibited normal cell death markers, such as annexin V and TUNEL, during VV infection. These results are reminiscent of the lack of inhibition of CD8+ T-cell death by pan-caspase inhibitors during VV infection (35). Although we cannot rule out the possibility that the transgenic lymphocytes could still undergo programmed necrosis, our results do indicate that the enhanced γδ T-cell infiltration and innate immune control in MC159 transgenic mice during VV infection could not be attributed to inhibition of apoptosis. Since our previous work indicates that lymphocyte activation to various stimuli, including that during LCMV infection, was normal in MC159 transgenic mice (53), lymphocyte-intrinsic effects are unlikely to account for the enhanced clearance of VV in the MC159 transgenic mice.
How might MC159 expression in the parenchyma facilitate innate immune responses? Our results indicate that transgenic fibroblasts exhibited heightened NF-κB activation in response to TNF and TLR4 stimulation. MC159 appears to specifically affect late-phase NF-κB activation while having little effect on early NF-κB induction. Interestingly, CCL-2 is a transcriptional target of NF-κB (21), and TNF and TLR4 signaling have been shown to play crucial roles in the innate immune defense against VV (8, 11, 25). Thus, increased NF-κB signaling to TNF or TLR4 stimulation could explain the increased CCL-2 expression and inflammatory leukocyte infiltration in the transgenic fat pad in response to VV infection. Although enhanced NF-κB activation was also observed for cells of hematopoietic origin (e.g., BMDCs and BMMs), radiation chimeras indicate that enhanced NF-κB activation in hematopoietic cells had a minimal contribution to the enhanced innate immunity against VV infections. Collectively, these results demonstrate an important role for signaling by cells in the parenchyma in innate immune responses to pathogens.
It is noteworthy that another vFLIP, K13 from human herpesvirus 8 (HHV-8), has also been reported to enhance NF-κB activation through an unknown mechanism (12, 29). Interestingly, we found that MC159 enhanced the binding between RIP1 and TRADD, two essential adaptors for NF-κB activation, by multiple TLRs and TNF receptors (9, 17, 38). We propose that enhanced interaction between RIP1 and TRADD might underlie the mechanism by which MC159 promotes NF-κB signaling by innate immune receptors. However, the positive effect of MC159 on NF-κB activation was observed only when it is expressed at a low level. When expressed at higher levels, MC159 inhibited NF-κB activation. The inhibitory effect of MC159 at higher expression levels was reminiscent of several previous reports in which MC159 impaired TNF- and PKR-induced NF-κB activation (20, 33), double-stranded RNA (dsRNA)-mediated interferon regulatory factor 7 (IRF7) signaling (2), and T-cell activation (54). The molecular mechanism by which MC159 inhibits NF-κB activation is unclear at the moment. Based on our results, we propose that at low levels of expression, MC159 promotes RIP1 binding to the TRADD. However, as its expression level increases, MC159 might bind to and inhibit other cellular targets that lead to inhibition of NF-κB and other immune functions.
TNF-TNF receptor signaling is particularly important for the innate immune control of poxvirus infections, since TNFR-1−/− and TNFR-2−/− mice were highly susceptible to VV and ectromelia virus infections (8, 39, 42). In contrast, molluscum contagiosum virus (MCV) is readily controlled in immunocompetent individuals but causes persistent lesions in immunocompromised individuals, such as those infected with HIV (22). Due to the ability of MC159 to promote or inhibit immune functions, it is tempting to speculate that MC159 might cooperate with other MCV-encoded immunoregulatory molecules to differentially modulate host responses in healthy and immunocompromised individuals.
We thank R. Welsh and L. Selin for critical readings of the manuscript and T. McQuade for help with the scoring of histology sections.
F.K.C. is a member of the UMass DERC (DK32520). D.M. is supported by a NIH predoctoral training grant (T32 AI07349). This work was supported by NIH grants AI083497, AI042845, AI017672, and AI042845. Core resources supported by the Diabetes Endocrinology Research Center Grant DK32520 were also used. F.K.C. was a recipient of investigator awards from the Smith Family Foundation and the Cancer Research Institute.
Published ahead of print on 11 August 2010.