A Lymphotoxin-IFN- Axis Essential for Lymphocyte Survival Revealed during Cytomegalovirus Infection1
http://www.100md.com
免疫学杂志 2005年第11期
Abstract
The importance of lymphotoxin (LT) R (LTR) as a regulator of lymphoid organogenesis is well established, but its role in host defense has yet to be fully defined. In this study, we report that mice deficient in LTR signaling were highly susceptible to infection with murine CMV (MCMV) and early during infection exhibited a catastrophic loss of T and B lymphocytes, although the majority of lymphocytes were themselves not directly infected. Moreover, bone marrow chimeras revealed that lymphocyte survival required LT expression by hemopoietic cells, independent of developmental defects in lymphoid tissue, whereas LTR expression by both stromal and hemopoietic cells was needed to prevent apoptosis. The induction of IFN- was also severely impaired in MCMV-infected LT–/– mice, but immunotherapy with an agonist LTR Ab restored IFN- levels, prevented lymphocyte death, and enhanced the survival of these mice. IFN-R–/– mice were also found to exhibit profound lymphocyte death during MCMV infection, thus providing a potential mechanistic link between type 1 IFN induction and lymphocyte survival through a LT-dependent pathway important for MCMV host defense.
Introduction
Herpes viruses and their vertebrate hosts share a long evolutionary history that is revealed by the multitude of tactics used by the virus to evade both innate and adaptive immune responses of the host. The result of this evolutionary tango is the ability of herpesviruses to establish a persistent infection without overt pathogenicity. CMV, a -herpesvirus, provides an insightful model of host-virus interactions, exemplified by its extensive array of evasion strategies, from disruption of Ag-processing pathways to the modulation of cytokines, all most likely contributing to the success of CMV in establishing coexistence with its host (1). Control of murine CMV (MCMV)6 requires a robust innate immune response, mediated predominately by IFNs and NK cells, to limit viral replication in the spleen and liver (2, 3, 4). Adaptive cellular responses, characterized by CD8+ and CD4+ T cells, mediate viral clearance and control virion shedding from the salivary gland (5, 6, 7). Together these defenses sufficiently restrict CMV to a latent/persistent infection without palpable disease in the immunocompetent host, whereas suppression of the immune system invariably leads to viral reactivation and disseminated disease (8).
The cell death and survival activities of the TNF-related cytokines (9) may provide strong selective pressure for viruses, such as CMV, to evolve immune evasion strategies that promote host-virus coexistence (10). Not unexpectedly, several cytokines and receptors belonging to the TNF superfamily are specifically targeted by herpesviridae (11). For example, recent evidence indicates that the lymphotoxin (LT) -LTR signaling pathway may function as an antiviral effector system in the host’s defense against CMV. In vitro studies have shown that LTR signaling induces the expression of IFN- in fibroblasts infected with human CMV, resulting in viral stasis in which viral replication can be curtailed without the concomitant destruction of virus-infected cells (12). Mice genetically deficient in LT were found to be highly susceptible to MCMV, suggesting a potentially conserved role for this cytokine/receptor family in CMV host defense (12). However, even though the correlation between this in vitro and in vivo data is striking, the evolutionary divergence between human and mouse CMV is substantial, thus precluding any direct conclusion regarding the mechanistic role of LT in a physiologic context.
Genetic models in mice have established that the LT-LTR pathway is crucial for the complex processes involved in the development (13, 14, 15, 16, 17, 18) and homeostasis of lymphoid tissues (19, 20, 21, 22, 23) and for the differentiation of NK and NK-T cells (24, 25, 26, 27, 28, 29, 30), key effectors of innate defenses. As a result, LT-deficient mice (i.e., LT–/–, LT–/–, LTR–/–, and the double-knockout LT/LIGHT–/–), in which LT is unable to effectively activate the LTR, all demonstrate a complex, developmentally fixed phenotype characterized by a lack of secondary lymphoid organs, multiple defects in splenic architecture, deficiencies in the number and function of NK/NK T cells, and decreased levels of certain chemokines. However, in contrast to their innate response deficiencies, the adaptive immune system in LT-deficient mice appears largely intact, with normal T and B lymphocyte development, although impaired dendritic cell (DC) migration has been suggested (29, 31). In addition to LT, the LTR can also be activated by a second ligand LIGHT, which, in turn, is able to engage yet another receptor, the herpesvirus entry mediator (HVEM) (32). In contrast to LT-deficient mice, LIGHT-deficient mice possess a full complement of lymphoid organs and normal levels of all lymphocyte subsets (18, 33, 34), whereas constitutive transgenic expression of LIGHT has been shown to induce destructive T cell-mediated inflammatory processes and autoimmunity (35, 36). Although the role of the LIGHT-HVEM system in host defense has only recently begun to be investigated, initial reports indicate that HVEM-deficient mice demonstrate normal lymphoid tissue development and lymphocyte differentiation (61).
The shared usage of the LTR by at least two different cytokines, LT and LIGHT, as well as other potentially unknown abnormalities associated with the developmentally fixed LT-deficient phenotype, adds to the complexity of delineating the contributions of this cytokine-receptor system to host defense. In the present investigation, we identify LT as an essential effector system that contributes substantially to the induction of the IFN- system early during infection by MCMV. The results reveal the previously unrecognized involvement of the LT-LTR and IFN- pathways in promoting the survival of the adaptive immune system to this viral pathogen. Pharmacological and genetic approaches demonstrate that the LTR pathway is a critical effector pathway that links the pleiotropic IFN- response to the survival of the adaptive immune response. Thus, by counteracting the virulence of MCMV for the lymphoid compartment, this LT-IFN axis is likely to play an important role in the establishment of host-virus coexistence.
Materials and Methods
Flow cytometry
Cell suspensions were prepared from spleens or livers; RBC were removed by lysis in hypotonic buffer; and mononuclear cells were isolated from perfused livers on a Percoll gradient. Cells were preincubated with anti-FcRII/III blocking reagent (2.4G2; BD Pharmingen), stained with direct fluorochrome-conjugated mAbs at 1 μg/ml per 5 x 105 cells, and then incubated on ice for 30 min before washing the cells three times in PBS containing 0.1% BSA and 0.01% sodium azide. The following Abs were used for lymphocyte subset detection: CyChrome RM4-4 (anti-CD4), allophycocyanin 53-6.7 (anti-CD8), CyChrome RA3-6B2 (anti-B220), PE HL3 (anti-CD11c), allophycocyanin M1/70 (anti-CD11b), PE MR5-2 (anti-V8.1/8.2), and PE RR3-15 (anti-V11) (BD Pharmingen). Cell death was assessed by annexin V staining (annexin V FITC; BD Pharmingen) and nuclear DNA staining by propidium iodide for subdiploid peak analysis, as described previously (43). Stained cells were analyzed on a FACSCalibur flow cytometer using CellQuest research software (BD Biosciences).
Results
LT–/– mice were shown to be highly susceptible to MCMV, requiring 100-fold less virus than wild-type B6 mice to induce a lethal infection (12). Similarly, BALB/c mice expressing LTR-Fc as a transgene (38) also demonstrated an increased susceptibility to MCMV infection; thus, the contribution of LT to host resistance appears to be independent of the mouse strain used. The susceptibility of LT-deficient mice to MCMV was also evident by the increased virus production at both early times (day 3) in the spleen and liver and at later times (day 12) in the salivary gland (Fig. 1). Virus titers in the spleen and liver were increased by 100-fold or more in LT–/–, LT–/–, and LTR–/– mice, and in double-knockout LT/LIGHT–/– mice (p < 0.0001), and by 5-fold in LIGHT–/– mice (p < 0.05) as compared with wild-type B6 mice (Fig. 1, a and b). However, HVEM –/– mice did not show increased virus production in any of the organs examined. Similarly, virus titers in the salivary glands were also increased in all of the LT-deficient mice strains and to a lesser extent in LIGHT–/– mice (Fig. 1c). Together, these results indicate that while the LT-LTR pathway is clearly the dominant interaction critical for controlling replication of MCMV, LIGHT also contributes to MCMV resistance. The discordance of phenotype between LIGHT- and HVEM-deficient mice for virus production suggests that LIGHT is acting via the LTR.
The number of hepatic DC defined by costaining with Abs to CD11c and CD8 was also decreased (>10-fold difference) in MCMV-infected LT–/– mice as compared with wild-type B6 mice (Fig. 4d). The loss of hepatic CD11c+CD8+ DC is consistent with the ability of this DC subset in the spleen to support lytic replication of MCMV (47). Additionally, a loss of hepatic CD8+CD11c– T cells was also seen in MCMV-infected LT–/– mice, as compared with similarly infected wild-type B6 mice, indicating lymphocyte death was not restricted to the spleen (also observed in the thymus; data not shown).
The cell death observed in MCMV-infected, LT-deficient mice also prompted an examination of lymphocyte viability in BALB/c mice challenged with MCMV. BALB/c mice are more susceptible to MCMV than B6 mice because they lack the Cmv1-encoded Ly-49H NK cell-activating receptor required for controlling virus replication in the spleen (48, 49, 50). MCMV infection of BALB/c mice caused a specific increase in annexin V staining of CD8+ T cells, whereas CD4+ T cells appeared unchanged and B220+ B cells showed only a fractional increase in apoptosis (Fig. 4e). In contrast, BALB/c mice expressing soluble LTR-Fc as a transgene showed a striking increase in annexin V staining in all lymphocyte subsets examined (CD8, CD4, and B220) similar to LT-deficient mice, suggesting that LT and LIGHT contribute to lymphocyte viability independently of Cmv1.
LTR-deficient mice infected with HSV-1 (-herpesvirus, McKrae strain) or with lymphocytic choriomeningitis virus (LCMV; arenavirus, Armstrong strain) did not show the apoptotic demise of lymphocytes observed during MCMV infection (data not shown). In addition, the proliferation and contraction of the V8+ TCR CD4+ and CD8+ subsets of T cells in response to staphylococcus enterotoxin B occurred normally in LT–/– mice, as did the proliferation and viability of B cells and macrophages to LPS (data not shown). Thus, the loss of lymphocyte viability observed in LT-deficient mice during MCMV infection does not appear to be a global defect in response to all pathogens or inflammatory stimuli.
To help identify the subpopulations of cells infected with MCMV, a recombinant virus expressing EGFP under control of the IE1 promoter was used to identify infected subpopulations of cells in the spleen (40). A significant increase in EGFP+ cells (27% of the cells; 3-fold) was observed in the CD11b+ cell population from LTR–/– mice by 2 days following infection (Fig. 5a, top panel). Moreover, EGFP fluorescence was detected in 50% of the CD11c+ DC population in LT–/– mice when compared with 20% of DC in B6 mice (Fig. 5b), consistent with previous observations that MCMV can infect macrophages and DC (40, 47). By contrast, EGFP fluorescence in either CD4 or CD8 T cell subpopulations was limited to a minor fraction of these lymphocytes from infected LTR–/– or B6 mice (Fig. 5a, lower panels), although a significant increase in EGFP expression occurred in 8% of the B220+ cells in LTR–/– mice. The increase in splenocyte infection is associated with significantly increased levels of CMV replication in the absence of LT signaling.
To determine whether direct viral infection contributed to lymphocyte apoptosis, gates were established to analyze the level of EGFP expression in the annexin V+ CD4/8 and B220 subsets undergoing death. The results indicated that both annexin V+ T and B cell subsets were >90% EGFP negative at day 3 following infection (Fig. 5c). This result indicates that the degree of apoptosis (>80%) is not proportional to virus-derived EGFP expression, indicating that the observed lymphocyte apoptosis does not appear to result from direct lytic infection with MCMV.
Adoptive transfers of BM from wild-type B6 (LT-sufficient) or LT-deficient donor mice into lethally irradiated LT-sufficient or LT-deficient recipients and their subsequent infection with MCMV revealed that the apoptotic phenotype was not observed when wild-type B6 BM cells were transferred into LT–/– recipients. This indicates that hemopoietic cell expression of LT is required for the protection of lymphocytes from apoptosis (Table I). Furthermore, because the LT–/– recipients lack secondary lymphoid organs and have defects in lymphoid tissue architecture, this result also indicates that the developmentally fixed lymphoid tissue defects do not contribute to the apoptotic phenotype. In contrast, only a partial restoration of lymphocyte viability was observed following the transfer of wild-type BM into irradiated LTR–/– recipient mice, as was the case when LTR–/– BM was transplanted into wild-type B6 recipient mice (Table I). This partial, nonreciprocal reconstitution indicates that LTR expression is required on both stromal and hemopoietic cells and suggests that these compartments must cooperate to fully protect lymphocytes from apoptosis during MCMV infection.
Activation of LTR induces IFN- and prevents lymphocyte death during MCMV infection
If there is a critical effector function for LTR signaling in host defense during MCMV infection, we reasoned that the administration of an agonist LTR Ab to LT–/– mice should restore induction of IFN- mRNA and protect lymphocytes from apoptosis. As predicted, the IFN- response in LT–/– mice was fully restored to wild-type levels with anti-LTR Ab treatment and in a similar time course (Fig. 6a). However, the anti-LTR Ab did not induce IFN- in the absence of virus infection, nor did it augment induction in LT-sufficient wild-type B6 mice (Fig. 6b). Accordingly, MCMV-infected LT–/– mice treated with anti-LTR Ab revealed a dramatic increase in total live splenic lymphoid cells as compared with mice in the other experimental groups (Fig. 6c, upper panel). Specifically, the percentages of apoptotic T and B cells (annexin V+) decreased by 3- to 6-fold in the anti-LTR Ab-treated LT–/– mice, but not in LTR–/– mice (Fig. 6c, lower panels), which establishes the specificity of the anti-LTR Ab. In addition, H&E-stained spleen sections offered visual evidence that the anti-LTR Ab treatment of MCMV-infected LT–/– mice rescued splenic cellularity, a finding that was not observed in similarly treated MCMV-infected LTR–/– mice (Fig. 6d). Furthermore, virus challenge experiments demonstrated that anti-LTR Ab-treated LT–/– mice survived significantly longer (mean survival of 10.5 days) in response to a lethal dose of MCMV (2 x 105 PFU) than isotype control-treated LT–/– mice (6.5 days) or anti-LTR Ab-treated LTR–/– mice (6.0 days) (Fig. 6e).
Discussion
The results presented in this work establish the LT-LTR signaling pathway as an essential effector pathway for host defense against the -herpesvirus MCMV. Mice deficient in LTR signaling were found to be highly susceptible to MCMV infection, as evidenced by increased mortality, enhanced virus production in target organs, impaired IFN- induction, and an apoptotic collapse of the adaptive immune system affecting T and B lymphocytes and DC. This striking apoptotic phenotype, in which the majority of lymphocyte death occurs without direct MCMV infection, led us to predict that the loss of a key survival factor(s) may offer a mechanistic explanation for the observed cell death. Our results, which revealed that administering an agonist LTR Ab to MCMV-infected LT-deficient mice could restore induced IFN- to wild-type levels, prevent lymphocyte apoptosis, and extend the survival of these mice, strongly implicate type 1 IFN as key survival factors for lymphocytes during MCMV infection. That MCMV-infected IFN-R–/– mice also exhibited the apoptotic phenotype provides further genetic evidence that lymphocyte survival requires LT-dependent activation of IFN- gene expression, revealing an LT-IFN axis crucial for host defense to MCMV.
In the absence of LTR signaling, the host’s immune system is unable to control MCMV infection, exposing the virulent potential of MCMV. Indeed, increased MCMV susceptibility appears rather selective for the LT-LTR cytokine signaling system because genetic deficiencies in the related receptors Fas or TNFR1 do not result in increased susceptibility to MCMV (51). With regard to the role of LT in host defense against other herpesviruses, LT-deficient mice demonstrate increased susceptibility and functionally impaired CD8+ T cell responses to HSV-1 (52), but without the significant amount of lymphocyte apoptosis observed during MCMV infection. In contrast, LT-deficient mice do not show increased susceptibility to murine -herpesvirus-68 and are effective, albeit with slightly delayed kinetics, in clearing a productive infection and controlling latency (53). These results indicate that individual pathogen-specific characteristics must figure prominently in the susceptibility profiles demonstrated by LT-deficient mice infected not only with different viruses, but with other nonviral pathogens as well. MCMV is a virus in which both innate and adaptive immune responses are needed to effectively control this infection, and the fact that both arms of the immune response appear to be compromised in LT-deficient mice most likely contributes significantly to their susceptibility to MCMV.
Increased susceptibility in LT-deficient mice to certain viral pathogens has also been attributed to developmental defects, resulting in abnormal lymphoid tissue architecture (52, 54). For example, adoptive transfer experiments demonstrated that T cell-specific responses to LCMV were impaired when wild-type splenocytes were transferred into LT–/– recipients, which lack organized microenvironments, as compared with the reciprocal transfer (LT–/– cells into wild-type recipients), indicating that the development of LCMV-specific CD8+ T cell responses requires intact lymphoid tissue (54). In contrast, evidence presented in this work indicates that an LTR effector system, independent of lymphoid tissue architecture, can contribute significantly to MCMV host defense. For example, BM from wild-type donor mice when transferred into LT–/– recipients, which possess abnormal lymphoid architecture, prevented lymphocyte apoptosis following MCMV infection, whereas the reciprocal transfer of LT–/– BM into wild-type recipients did not (see Table I), indicating that the underlying developmentally dependent lymphoid tissue defects present in LT–/– mice do not appear responsible for the apoptotic phenotype.
In the absence of LTR signaling, MCMV infection induced the apoptotic collapse of the adaptive immune system, revealing a phenotype not previously associated with the LT signaling pathway. All of the LT-deficient mice examined exhibited significant apoptosis of T cells (both CD4+ and CD8+), B cells, and DC, whereas increased apoptosis in LIGHT–/– mice was limited to CD4+ T cells only. Interestingly, a selective loss of CD8+ T cells was observed in BALB/c mice, which are defective in Ly-49H NK cells. These results suggest that the survival of lymphocyte subsets may depend on signaling by different subpopulations of LT/LIGHT-expressing cells. However, apoptosis was observed in all lymphocyte subsets when LTR-Fc was genetically introduced into BALB/c mice, underscoring the potential virulence of MCMV in the absence of an LT-mediated protective mechanism for cells of the adaptive immune system. By comparison, the level of lymphocyte apoptosis in MCMV-infected normal B6 mice was usually less than 20% at the virus inoculum used, although very high doses can induce splenic necrosis in B6 mice, suggesting that deficiency in LT lowers this threshold. In LT-deficient mice, most of the apoptosis in lymphocytes appeared to occur as a bystander effect of MCMV infection because very little virus promoter-driven EGFP expression was detected in lymphocytes that had already entered the apoptotic process, as determined by annexin V staining. This conclusion is restricted by the possibility that an abortive infection without significant viral gene expression cannot be excluded. We were also unable to block lymphocyte death by treatment with Fas-Fc fusion protein in vivo (data not shown), indicating that this death receptor system does not appear to be prominently involved during MCMV infection.
The results presented in this work identify type 1 IFN as a primary survival system for T and B lymphocytes and DC in response to MCMV. Type 1 IFN are necessary for resistance to MCMV (55), and increased virus replication in the spleens of IFN-R–/– mice as compared with B6 controls supports this conclusion. In response to MCMV infection, IFN- was poorly induced in LT–/– mice, and IFN- mRNA accumulation was also decreased in the spleens of LT–/– mice early after infection (K. Schneider and C. Benedict, unpublished observations), suggesting a generalized failure in the type 1 IFN response. However, IFN- mRNA was restored to wild-type levels following treatment with an agonist LTR Ab, indicating that the cellular elements involved in IFN- production are present, but lack the appropriate stimulus. Moreover, treatment with the agonist anti-LTR was able to restore lymphocyte viability in MCMV-infected LT–/– mice consistent with the idea that LTR signaling, by inducing type 1 IFN during MCMV infection, is essential for lymphocyte viability. The observation that IFN- can enhance the survival of activated T cells in certain scenarios supports this concept (56, 57, 58, 59). However, most compelling is the observation that IFN-R–/– mice also demonstrate a profound apoptotic phenotype in response to MCMV, thus providing genetic evidence that IFN-R signaling is critical for lymphocyte survival during MCMV infection.
These results are consistent with the idea that LT and IFN- function in a common pathway for lymphocyte survival, adding support to the biochemical evidence that LTR signaling regulates IFN- induction via NF-B-dependent signaling (12). In the context of microbial infection, IFN- is a highly pleiotrophic cytokine whose signaling pathways inhibit viral replication, block virus spread to neighboring cells, regulate the differentiation of NK cells, promote DC maturation, and modulate immunity (reviewed in Ref.60). Thus, a compromised IFN- response in LT-deficient mice may have several distinct impacts on the overall susceptibility to MCMV. In this regard, MCMV infection of DC in LT-deficient mice may block or modulate the induction of IFN, contributing to the increased apoptosis of lymphocytes and heightened susceptibility observed in these mice. Recent studies by Biron and colleagues (46) indicate that the CD11c+CD8+ DC subset is a primary target for MCMV infection in 129SvEv mice. We found that the corresponding hepatic DC subset was also specifically depleted in MCMV-infected LT–/– mice, as well as an increase in the percentage of infected splenic DC, suggesting a protective role for LT in the DC compartment. The failure to activate IFN- expression in LT-deficient mice may also halt DC differentiation, causing a loss of the costimulatory signals required for sustained activation of T and B cells. The recruitment of DC to the spleen occurs during MCMV infection, and the plasmacytoid DC subset, a major IFN--producing cell, requires IFN- receptor expression for this recruitment (46). Thus, a poor IFN response may affect DC recruitment, which is known to be already impaired in LT-deficient mice, due, in part, to the decreased production of CCR7-binding chemokines by stromal cells (19, 29). In addition, the expression of LTR on stromal cells and DC (K. Potter and C. Ware, unpublished observations) suggests that the LTR could provide direct antiviral signaling to these cells via IFN- production. Thus, the blockade or loss of IFN- and other lymphocyte survival factors normally induced by LTR signaling during the innate response to MCMV infection may significantly compromise adaptive immunity.
BM transfer experiments revealed that LTR expression by both radioresistant stroma and hemopoietic cells was needed to prevent lymphocyte apoptosis, demonstrating that both compartments are compromised by MCMV. These results suggest that both the stromal and hemopoietic compartments may play an important role in the production of type I IFN in response to MCMV. Interestingly, BM transfers between wild-type B6 and IFN-R-deficient mice revealed that expression of IFN-R in either compartment (stromal or hemopoietic) was sufficient to prevent lymphocyte death during MCMV infection. One interpretation of these results is that type I IFN-mediated protection acts directly on the hemopoeitic compartment and indirectly on the stroma, possibly through the production of secondary survival factors downstream of IFN-R signaling.
The survival function of LT and LIGHT revealed by MCMV resembles the functional activities of their genetically linked paralogs on chromosome 19 (CD27L, 41BBL), 1 (Fas ligand, GITRL, Ox40L), and 9 (TL1A, CD30L), which provide cooperative signaling during T cell activation. It is tempting to speculate that the selective pressure responsible for the diversification of these ligands and their cognate receptors may have been a primordial herpesvirus.
Although mouse and human CMV share a common evolutionary ancestor, they have diverged with their respective host species, as evidenced by the differences in the details of their immune subversion mechanisms. The results presented in this work establish that both mouse and human CMV are controlled at least, in part, by the regulation of the type 1 IFN system through LT signaling. Thus, further delineation of the molecular mechanisms controlling this LT-IFN axis should assist in designing strategies for targeted antiviral therapies.
Acknowledgments
We thank all members of the Molecular Immunology Laboratory at La Jolla Institute for Allergy and Immunology for their assistance with this project. We also thank R. Rickert, A. Angulo, and P. Ghazal for many helpful discussions; K. Vroom for assistance with histology; S. Santee-Copper for generating the agonist LTR Ab; and Karin Mink for help in establishing the HVEM-deficient mouse line. Special thanks to L. Hall for expert graphics and manuscript preparation and to the vivarium staff at La Jolla Institute for Allergy and Immunology for excellent animal care.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was supported in part by National Institutes of Health Grants AI47644 (to T.A.B.), AI33068, CA69381, and AI48073 (to C.F.W.); American Heart Association Grant 63264-00-351 (to C.A.B.); and Grants Pf 259/2-6 and Pf 259/3-1 from the Deutsche Forchungsgemeinshaft (to K.P.). This is manuscript 582 from the La Jolla Institute for Allergy and Immunology.
2 T.A.B., S.R., and C.A.B. contributed equally to this study.
3 Current address: Gemini Sciences, San Diego, CA 92121.
4 Current address: Division of Rheumatology, University Hospital Gent, Gent, Belgium.
5 Address correspondence and reprint requests to Dr. Carl F. Ware, La Jolla Institute for Allergy and Immunology, 10355 Science Center Drive, San Diego, CA 92121. E-mail address: cware{at}liai.org
6 Abbreviations used in this paper: MCMV, murine CMV; BM, bone marrow; DC, dendritic cell; EGFP, enhanced GFP; HVEM, herpesvirus entry mediator; LCMV, lymphocytic choriomeningitis virus; LIGHT, TNF superfamily member 14; LT, lymphotoxin; p.i., postinfection.
Received for publication November 22, 2004. Accepted for publication March 16, 2005.
References
Tortorella, D., B. E. Gewurz, M. H. Furman, D. J. Schust, H. L. Ploegh. 2000. Viral subversion of the immune system. Annu. Rev. Immunol. 18: 861-926.
Tay, C. H., R. M. Welsh. 1997. Distinct organ-dependent mechanisms for the control of murine cytomegalovirus infection by natural killer cells. J. Virol. 71: 267-275.
Orange, J. S., B. Wang, C. Terhorst, C. A. Biron. 1995. Requirement for natural killer cell-produced interferon in defense against murine cytomegalovirus infection and enhancement of this defense pathway by interleukin 12 administration. J. Exp. Med. 182: 1045-1056.
Orange, J. S., C. A. Biron. 1996. Characterization of early IL-12, IFN-, and TNF effects on antiviral state and NK cell responses during murine cytomegalovirus infection. J. Immunol. 156: 4746-4756.
Jonjic, S., I. Pavic, P. Lucin, D. Rukavina, U. H. Koszinowski. 1990. Efficacious control of cytomegalovirus infection after long-term depletion of CD8+ T lymphocytes. J. Virol. 64: 5457-5464.
Jonjic, S., W. Mutter, F. Weiland, M. J. Reddehase, U. H. Koszinowski. 1989. Site-restricted persistent cytomegalovirus infection after selective long-term depletion of CD4+ T lymphocytes. J. Exp. Med. 169: 1199-1212.(Theresa A. Banks2,*,, San)
The importance of lymphotoxin (LT) R (LTR) as a regulator of lymphoid organogenesis is well established, but its role in host defense has yet to be fully defined. In this study, we report that mice deficient in LTR signaling were highly susceptible to infection with murine CMV (MCMV) and early during infection exhibited a catastrophic loss of T and B lymphocytes, although the majority of lymphocytes were themselves not directly infected. Moreover, bone marrow chimeras revealed that lymphocyte survival required LT expression by hemopoietic cells, independent of developmental defects in lymphoid tissue, whereas LTR expression by both stromal and hemopoietic cells was needed to prevent apoptosis. The induction of IFN- was also severely impaired in MCMV-infected LT–/– mice, but immunotherapy with an agonist LTR Ab restored IFN- levels, prevented lymphocyte death, and enhanced the survival of these mice. IFN-R–/– mice were also found to exhibit profound lymphocyte death during MCMV infection, thus providing a potential mechanistic link between type 1 IFN induction and lymphocyte survival through a LT-dependent pathway important for MCMV host defense.
Introduction
Herpes viruses and their vertebrate hosts share a long evolutionary history that is revealed by the multitude of tactics used by the virus to evade both innate and adaptive immune responses of the host. The result of this evolutionary tango is the ability of herpesviruses to establish a persistent infection without overt pathogenicity. CMV, a -herpesvirus, provides an insightful model of host-virus interactions, exemplified by its extensive array of evasion strategies, from disruption of Ag-processing pathways to the modulation of cytokines, all most likely contributing to the success of CMV in establishing coexistence with its host (1). Control of murine CMV (MCMV)6 requires a robust innate immune response, mediated predominately by IFNs and NK cells, to limit viral replication in the spleen and liver (2, 3, 4). Adaptive cellular responses, characterized by CD8+ and CD4+ T cells, mediate viral clearance and control virion shedding from the salivary gland (5, 6, 7). Together these defenses sufficiently restrict CMV to a latent/persistent infection without palpable disease in the immunocompetent host, whereas suppression of the immune system invariably leads to viral reactivation and disseminated disease (8).
The cell death and survival activities of the TNF-related cytokines (9) may provide strong selective pressure for viruses, such as CMV, to evolve immune evasion strategies that promote host-virus coexistence (10). Not unexpectedly, several cytokines and receptors belonging to the TNF superfamily are specifically targeted by herpesviridae (11). For example, recent evidence indicates that the lymphotoxin (LT) -LTR signaling pathway may function as an antiviral effector system in the host’s defense against CMV. In vitro studies have shown that LTR signaling induces the expression of IFN- in fibroblasts infected with human CMV, resulting in viral stasis in which viral replication can be curtailed without the concomitant destruction of virus-infected cells (12). Mice genetically deficient in LT were found to be highly susceptible to MCMV, suggesting a potentially conserved role for this cytokine/receptor family in CMV host defense (12). However, even though the correlation between this in vitro and in vivo data is striking, the evolutionary divergence between human and mouse CMV is substantial, thus precluding any direct conclusion regarding the mechanistic role of LT in a physiologic context.
Genetic models in mice have established that the LT-LTR pathway is crucial for the complex processes involved in the development (13, 14, 15, 16, 17, 18) and homeostasis of lymphoid tissues (19, 20, 21, 22, 23) and for the differentiation of NK and NK-T cells (24, 25, 26, 27, 28, 29, 30), key effectors of innate defenses. As a result, LT-deficient mice (i.e., LT–/–, LT–/–, LTR–/–, and the double-knockout LT/LIGHT–/–), in which LT is unable to effectively activate the LTR, all demonstrate a complex, developmentally fixed phenotype characterized by a lack of secondary lymphoid organs, multiple defects in splenic architecture, deficiencies in the number and function of NK/NK T cells, and decreased levels of certain chemokines. However, in contrast to their innate response deficiencies, the adaptive immune system in LT-deficient mice appears largely intact, with normal T and B lymphocyte development, although impaired dendritic cell (DC) migration has been suggested (29, 31). In addition to LT, the LTR can also be activated by a second ligand LIGHT, which, in turn, is able to engage yet another receptor, the herpesvirus entry mediator (HVEM) (32). In contrast to LT-deficient mice, LIGHT-deficient mice possess a full complement of lymphoid organs and normal levels of all lymphocyte subsets (18, 33, 34), whereas constitutive transgenic expression of LIGHT has been shown to induce destructive T cell-mediated inflammatory processes and autoimmunity (35, 36). Although the role of the LIGHT-HVEM system in host defense has only recently begun to be investigated, initial reports indicate that HVEM-deficient mice demonstrate normal lymphoid tissue development and lymphocyte differentiation (61).
The shared usage of the LTR by at least two different cytokines, LT and LIGHT, as well as other potentially unknown abnormalities associated with the developmentally fixed LT-deficient phenotype, adds to the complexity of delineating the contributions of this cytokine-receptor system to host defense. In the present investigation, we identify LT as an essential effector system that contributes substantially to the induction of the IFN- system early during infection by MCMV. The results reveal the previously unrecognized involvement of the LT-LTR and IFN- pathways in promoting the survival of the adaptive immune system to this viral pathogen. Pharmacological and genetic approaches demonstrate that the LTR pathway is a critical effector pathway that links the pleiotropic IFN- response to the survival of the adaptive immune response. Thus, by counteracting the virulence of MCMV for the lymphoid compartment, this LT-IFN axis is likely to play an important role in the establishment of host-virus coexistence.
Materials and Methods
Flow cytometry
Cell suspensions were prepared from spleens or livers; RBC were removed by lysis in hypotonic buffer; and mononuclear cells were isolated from perfused livers on a Percoll gradient. Cells were preincubated with anti-FcRII/III blocking reagent (2.4G2; BD Pharmingen), stained with direct fluorochrome-conjugated mAbs at 1 μg/ml per 5 x 105 cells, and then incubated on ice for 30 min before washing the cells three times in PBS containing 0.1% BSA and 0.01% sodium azide. The following Abs were used for lymphocyte subset detection: CyChrome RM4-4 (anti-CD4), allophycocyanin 53-6.7 (anti-CD8), CyChrome RA3-6B2 (anti-B220), PE HL3 (anti-CD11c), allophycocyanin M1/70 (anti-CD11b), PE MR5-2 (anti-V8.1/8.2), and PE RR3-15 (anti-V11) (BD Pharmingen). Cell death was assessed by annexin V staining (annexin V FITC; BD Pharmingen) and nuclear DNA staining by propidium iodide for subdiploid peak analysis, as described previously (43). Stained cells were analyzed on a FACSCalibur flow cytometer using CellQuest research software (BD Biosciences).
Results
LT–/– mice were shown to be highly susceptible to MCMV, requiring 100-fold less virus than wild-type B6 mice to induce a lethal infection (12). Similarly, BALB/c mice expressing LTR-Fc as a transgene (38) also demonstrated an increased susceptibility to MCMV infection; thus, the contribution of LT to host resistance appears to be independent of the mouse strain used. The susceptibility of LT-deficient mice to MCMV was also evident by the increased virus production at both early times (day 3) in the spleen and liver and at later times (day 12) in the salivary gland (Fig. 1). Virus titers in the spleen and liver were increased by 100-fold or more in LT–/–, LT–/–, and LTR–/– mice, and in double-knockout LT/LIGHT–/– mice (p < 0.0001), and by 5-fold in LIGHT–/– mice (p < 0.05) as compared with wild-type B6 mice (Fig. 1, a and b). However, HVEM –/– mice did not show increased virus production in any of the organs examined. Similarly, virus titers in the salivary glands were also increased in all of the LT-deficient mice strains and to a lesser extent in LIGHT–/– mice (Fig. 1c). Together, these results indicate that while the LT-LTR pathway is clearly the dominant interaction critical for controlling replication of MCMV, LIGHT also contributes to MCMV resistance. The discordance of phenotype between LIGHT- and HVEM-deficient mice for virus production suggests that LIGHT is acting via the LTR.
The number of hepatic DC defined by costaining with Abs to CD11c and CD8 was also decreased (>10-fold difference) in MCMV-infected LT–/– mice as compared with wild-type B6 mice (Fig. 4d). The loss of hepatic CD11c+CD8+ DC is consistent with the ability of this DC subset in the spleen to support lytic replication of MCMV (47). Additionally, a loss of hepatic CD8+CD11c– T cells was also seen in MCMV-infected LT–/– mice, as compared with similarly infected wild-type B6 mice, indicating lymphocyte death was not restricted to the spleen (also observed in the thymus; data not shown).
The cell death observed in MCMV-infected, LT-deficient mice also prompted an examination of lymphocyte viability in BALB/c mice challenged with MCMV. BALB/c mice are more susceptible to MCMV than B6 mice because they lack the Cmv1-encoded Ly-49H NK cell-activating receptor required for controlling virus replication in the spleen (48, 49, 50). MCMV infection of BALB/c mice caused a specific increase in annexin V staining of CD8+ T cells, whereas CD4+ T cells appeared unchanged and B220+ B cells showed only a fractional increase in apoptosis (Fig. 4e). In contrast, BALB/c mice expressing soluble LTR-Fc as a transgene showed a striking increase in annexin V staining in all lymphocyte subsets examined (CD8, CD4, and B220) similar to LT-deficient mice, suggesting that LT and LIGHT contribute to lymphocyte viability independently of Cmv1.
LTR-deficient mice infected with HSV-1 (-herpesvirus, McKrae strain) or with lymphocytic choriomeningitis virus (LCMV; arenavirus, Armstrong strain) did not show the apoptotic demise of lymphocytes observed during MCMV infection (data not shown). In addition, the proliferation and contraction of the V8+ TCR CD4+ and CD8+ subsets of T cells in response to staphylococcus enterotoxin B occurred normally in LT–/– mice, as did the proliferation and viability of B cells and macrophages to LPS (data not shown). Thus, the loss of lymphocyte viability observed in LT-deficient mice during MCMV infection does not appear to be a global defect in response to all pathogens or inflammatory stimuli.
To help identify the subpopulations of cells infected with MCMV, a recombinant virus expressing EGFP under control of the IE1 promoter was used to identify infected subpopulations of cells in the spleen (40). A significant increase in EGFP+ cells (27% of the cells; 3-fold) was observed in the CD11b+ cell population from LTR–/– mice by 2 days following infection (Fig. 5a, top panel). Moreover, EGFP fluorescence was detected in 50% of the CD11c+ DC population in LT–/– mice when compared with 20% of DC in B6 mice (Fig. 5b), consistent with previous observations that MCMV can infect macrophages and DC (40, 47). By contrast, EGFP fluorescence in either CD4 or CD8 T cell subpopulations was limited to a minor fraction of these lymphocytes from infected LTR–/– or B6 mice (Fig. 5a, lower panels), although a significant increase in EGFP expression occurred in 8% of the B220+ cells in LTR–/– mice. The increase in splenocyte infection is associated with significantly increased levels of CMV replication in the absence of LT signaling.
To determine whether direct viral infection contributed to lymphocyte apoptosis, gates were established to analyze the level of EGFP expression in the annexin V+ CD4/8 and B220 subsets undergoing death. The results indicated that both annexin V+ T and B cell subsets were >90% EGFP negative at day 3 following infection (Fig. 5c). This result indicates that the degree of apoptosis (>80%) is not proportional to virus-derived EGFP expression, indicating that the observed lymphocyte apoptosis does not appear to result from direct lytic infection with MCMV.
Adoptive transfers of BM from wild-type B6 (LT-sufficient) or LT-deficient donor mice into lethally irradiated LT-sufficient or LT-deficient recipients and their subsequent infection with MCMV revealed that the apoptotic phenotype was not observed when wild-type B6 BM cells were transferred into LT–/– recipients. This indicates that hemopoietic cell expression of LT is required for the protection of lymphocytes from apoptosis (Table I). Furthermore, because the LT–/– recipients lack secondary lymphoid organs and have defects in lymphoid tissue architecture, this result also indicates that the developmentally fixed lymphoid tissue defects do not contribute to the apoptotic phenotype. In contrast, only a partial restoration of lymphocyte viability was observed following the transfer of wild-type BM into irradiated LTR–/– recipient mice, as was the case when LTR–/– BM was transplanted into wild-type B6 recipient mice (Table I). This partial, nonreciprocal reconstitution indicates that LTR expression is required on both stromal and hemopoietic cells and suggests that these compartments must cooperate to fully protect lymphocytes from apoptosis during MCMV infection.
Activation of LTR induces IFN- and prevents lymphocyte death during MCMV infection
If there is a critical effector function for LTR signaling in host defense during MCMV infection, we reasoned that the administration of an agonist LTR Ab to LT–/– mice should restore induction of IFN- mRNA and protect lymphocytes from apoptosis. As predicted, the IFN- response in LT–/– mice was fully restored to wild-type levels with anti-LTR Ab treatment and in a similar time course (Fig. 6a). However, the anti-LTR Ab did not induce IFN- in the absence of virus infection, nor did it augment induction in LT-sufficient wild-type B6 mice (Fig. 6b). Accordingly, MCMV-infected LT–/– mice treated with anti-LTR Ab revealed a dramatic increase in total live splenic lymphoid cells as compared with mice in the other experimental groups (Fig. 6c, upper panel). Specifically, the percentages of apoptotic T and B cells (annexin V+) decreased by 3- to 6-fold in the anti-LTR Ab-treated LT–/– mice, but not in LTR–/– mice (Fig. 6c, lower panels), which establishes the specificity of the anti-LTR Ab. In addition, H&E-stained spleen sections offered visual evidence that the anti-LTR Ab treatment of MCMV-infected LT–/– mice rescued splenic cellularity, a finding that was not observed in similarly treated MCMV-infected LTR–/– mice (Fig. 6d). Furthermore, virus challenge experiments demonstrated that anti-LTR Ab-treated LT–/– mice survived significantly longer (mean survival of 10.5 days) in response to a lethal dose of MCMV (2 x 105 PFU) than isotype control-treated LT–/– mice (6.5 days) or anti-LTR Ab-treated LTR–/– mice (6.0 days) (Fig. 6e).
Discussion
The results presented in this work establish the LT-LTR signaling pathway as an essential effector pathway for host defense against the -herpesvirus MCMV. Mice deficient in LTR signaling were found to be highly susceptible to MCMV infection, as evidenced by increased mortality, enhanced virus production in target organs, impaired IFN- induction, and an apoptotic collapse of the adaptive immune system affecting T and B lymphocytes and DC. This striking apoptotic phenotype, in which the majority of lymphocyte death occurs without direct MCMV infection, led us to predict that the loss of a key survival factor(s) may offer a mechanistic explanation for the observed cell death. Our results, which revealed that administering an agonist LTR Ab to MCMV-infected LT-deficient mice could restore induced IFN- to wild-type levels, prevent lymphocyte apoptosis, and extend the survival of these mice, strongly implicate type 1 IFN as key survival factors for lymphocytes during MCMV infection. That MCMV-infected IFN-R–/– mice also exhibited the apoptotic phenotype provides further genetic evidence that lymphocyte survival requires LT-dependent activation of IFN- gene expression, revealing an LT-IFN axis crucial for host defense to MCMV.
In the absence of LTR signaling, the host’s immune system is unable to control MCMV infection, exposing the virulent potential of MCMV. Indeed, increased MCMV susceptibility appears rather selective for the LT-LTR cytokine signaling system because genetic deficiencies in the related receptors Fas or TNFR1 do not result in increased susceptibility to MCMV (51). With regard to the role of LT in host defense against other herpesviruses, LT-deficient mice demonstrate increased susceptibility and functionally impaired CD8+ T cell responses to HSV-1 (52), but without the significant amount of lymphocyte apoptosis observed during MCMV infection. In contrast, LT-deficient mice do not show increased susceptibility to murine -herpesvirus-68 and are effective, albeit with slightly delayed kinetics, in clearing a productive infection and controlling latency (53). These results indicate that individual pathogen-specific characteristics must figure prominently in the susceptibility profiles demonstrated by LT-deficient mice infected not only with different viruses, but with other nonviral pathogens as well. MCMV is a virus in which both innate and adaptive immune responses are needed to effectively control this infection, and the fact that both arms of the immune response appear to be compromised in LT-deficient mice most likely contributes significantly to their susceptibility to MCMV.
Increased susceptibility in LT-deficient mice to certain viral pathogens has also been attributed to developmental defects, resulting in abnormal lymphoid tissue architecture (52, 54). For example, adoptive transfer experiments demonstrated that T cell-specific responses to LCMV were impaired when wild-type splenocytes were transferred into LT–/– recipients, which lack organized microenvironments, as compared with the reciprocal transfer (LT–/– cells into wild-type recipients), indicating that the development of LCMV-specific CD8+ T cell responses requires intact lymphoid tissue (54). In contrast, evidence presented in this work indicates that an LTR effector system, independent of lymphoid tissue architecture, can contribute significantly to MCMV host defense. For example, BM from wild-type donor mice when transferred into LT–/– recipients, which possess abnormal lymphoid architecture, prevented lymphocyte apoptosis following MCMV infection, whereas the reciprocal transfer of LT–/– BM into wild-type recipients did not (see Table I), indicating that the underlying developmentally dependent lymphoid tissue defects present in LT–/– mice do not appear responsible for the apoptotic phenotype.
In the absence of LTR signaling, MCMV infection induced the apoptotic collapse of the adaptive immune system, revealing a phenotype not previously associated with the LT signaling pathway. All of the LT-deficient mice examined exhibited significant apoptosis of T cells (both CD4+ and CD8+), B cells, and DC, whereas increased apoptosis in LIGHT–/– mice was limited to CD4+ T cells only. Interestingly, a selective loss of CD8+ T cells was observed in BALB/c mice, which are defective in Ly-49H NK cells. These results suggest that the survival of lymphocyte subsets may depend on signaling by different subpopulations of LT/LIGHT-expressing cells. However, apoptosis was observed in all lymphocyte subsets when LTR-Fc was genetically introduced into BALB/c mice, underscoring the potential virulence of MCMV in the absence of an LT-mediated protective mechanism for cells of the adaptive immune system. By comparison, the level of lymphocyte apoptosis in MCMV-infected normal B6 mice was usually less than 20% at the virus inoculum used, although very high doses can induce splenic necrosis in B6 mice, suggesting that deficiency in LT lowers this threshold. In LT-deficient mice, most of the apoptosis in lymphocytes appeared to occur as a bystander effect of MCMV infection because very little virus promoter-driven EGFP expression was detected in lymphocytes that had already entered the apoptotic process, as determined by annexin V staining. This conclusion is restricted by the possibility that an abortive infection without significant viral gene expression cannot be excluded. We were also unable to block lymphocyte death by treatment with Fas-Fc fusion protein in vivo (data not shown), indicating that this death receptor system does not appear to be prominently involved during MCMV infection.
The results presented in this work identify type 1 IFN as a primary survival system for T and B lymphocytes and DC in response to MCMV. Type 1 IFN are necessary for resistance to MCMV (55), and increased virus replication in the spleens of IFN-R–/– mice as compared with B6 controls supports this conclusion. In response to MCMV infection, IFN- was poorly induced in LT–/– mice, and IFN- mRNA accumulation was also decreased in the spleens of LT–/– mice early after infection (K. Schneider and C. Benedict, unpublished observations), suggesting a generalized failure in the type 1 IFN response. However, IFN- mRNA was restored to wild-type levels following treatment with an agonist LTR Ab, indicating that the cellular elements involved in IFN- production are present, but lack the appropriate stimulus. Moreover, treatment with the agonist anti-LTR was able to restore lymphocyte viability in MCMV-infected LT–/– mice consistent with the idea that LTR signaling, by inducing type 1 IFN during MCMV infection, is essential for lymphocyte viability. The observation that IFN- can enhance the survival of activated T cells in certain scenarios supports this concept (56, 57, 58, 59). However, most compelling is the observation that IFN-R–/– mice also demonstrate a profound apoptotic phenotype in response to MCMV, thus providing genetic evidence that IFN-R signaling is critical for lymphocyte survival during MCMV infection.
These results are consistent with the idea that LT and IFN- function in a common pathway for lymphocyte survival, adding support to the biochemical evidence that LTR signaling regulates IFN- induction via NF-B-dependent signaling (12). In the context of microbial infection, IFN- is a highly pleiotrophic cytokine whose signaling pathways inhibit viral replication, block virus spread to neighboring cells, regulate the differentiation of NK cells, promote DC maturation, and modulate immunity (reviewed in Ref.60). Thus, a compromised IFN- response in LT-deficient mice may have several distinct impacts on the overall susceptibility to MCMV. In this regard, MCMV infection of DC in LT-deficient mice may block or modulate the induction of IFN, contributing to the increased apoptosis of lymphocytes and heightened susceptibility observed in these mice. Recent studies by Biron and colleagues (46) indicate that the CD11c+CD8+ DC subset is a primary target for MCMV infection in 129SvEv mice. We found that the corresponding hepatic DC subset was also specifically depleted in MCMV-infected LT–/– mice, as well as an increase in the percentage of infected splenic DC, suggesting a protective role for LT in the DC compartment. The failure to activate IFN- expression in LT-deficient mice may also halt DC differentiation, causing a loss of the costimulatory signals required for sustained activation of T and B cells. The recruitment of DC to the spleen occurs during MCMV infection, and the plasmacytoid DC subset, a major IFN--producing cell, requires IFN- receptor expression for this recruitment (46). Thus, a poor IFN response may affect DC recruitment, which is known to be already impaired in LT-deficient mice, due, in part, to the decreased production of CCR7-binding chemokines by stromal cells (19, 29). In addition, the expression of LTR on stromal cells and DC (K. Potter and C. Ware, unpublished observations) suggests that the LTR could provide direct antiviral signaling to these cells via IFN- production. Thus, the blockade or loss of IFN- and other lymphocyte survival factors normally induced by LTR signaling during the innate response to MCMV infection may significantly compromise adaptive immunity.
BM transfer experiments revealed that LTR expression by both radioresistant stroma and hemopoietic cells was needed to prevent lymphocyte apoptosis, demonstrating that both compartments are compromised by MCMV. These results suggest that both the stromal and hemopoietic compartments may play an important role in the production of type I IFN in response to MCMV. Interestingly, BM transfers between wild-type B6 and IFN-R-deficient mice revealed that expression of IFN-R in either compartment (stromal or hemopoietic) was sufficient to prevent lymphocyte death during MCMV infection. One interpretation of these results is that type I IFN-mediated protection acts directly on the hemopoeitic compartment and indirectly on the stroma, possibly through the production of secondary survival factors downstream of IFN-R signaling.
The survival function of LT and LIGHT revealed by MCMV resembles the functional activities of their genetically linked paralogs on chromosome 19 (CD27L, 41BBL), 1 (Fas ligand, GITRL, Ox40L), and 9 (TL1A, CD30L), which provide cooperative signaling during T cell activation. It is tempting to speculate that the selective pressure responsible for the diversification of these ligands and their cognate receptors may have been a primordial herpesvirus.
Although mouse and human CMV share a common evolutionary ancestor, they have diverged with their respective host species, as evidenced by the differences in the details of their immune subversion mechanisms. The results presented in this work establish that both mouse and human CMV are controlled at least, in part, by the regulation of the type 1 IFN system through LT signaling. Thus, further delineation of the molecular mechanisms controlling this LT-IFN axis should assist in designing strategies for targeted antiviral therapies.
Acknowledgments
We thank all members of the Molecular Immunology Laboratory at La Jolla Institute for Allergy and Immunology for their assistance with this project. We also thank R. Rickert, A. Angulo, and P. Ghazal for many helpful discussions; K. Vroom for assistance with histology; S. Santee-Copper for generating the agonist LTR Ab; and Karin Mink for help in establishing the HVEM-deficient mouse line. Special thanks to L. Hall for expert graphics and manuscript preparation and to the vivarium staff at La Jolla Institute for Allergy and Immunology for excellent animal care.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was supported in part by National Institutes of Health Grants AI47644 (to T.A.B.), AI33068, CA69381, and AI48073 (to C.F.W.); American Heart Association Grant 63264-00-351 (to C.A.B.); and Grants Pf 259/2-6 and Pf 259/3-1 from the Deutsche Forchungsgemeinshaft (to K.P.). This is manuscript 582 from the La Jolla Institute for Allergy and Immunology.
2 T.A.B., S.R., and C.A.B. contributed equally to this study.
3 Current address: Gemini Sciences, San Diego, CA 92121.
4 Current address: Division of Rheumatology, University Hospital Gent, Gent, Belgium.
5 Address correspondence and reprint requests to Dr. Carl F. Ware, La Jolla Institute for Allergy and Immunology, 10355 Science Center Drive, San Diego, CA 92121. E-mail address: cware{at}liai.org
6 Abbreviations used in this paper: MCMV, murine CMV; BM, bone marrow; DC, dendritic cell; EGFP, enhanced GFP; HVEM, herpesvirus entry mediator; LCMV, lymphocytic choriomeningitis virus; LIGHT, TNF superfamily member 14; LT, lymphotoxin; p.i., postinfection.
Received for publication November 22, 2004. Accepted for publication March 16, 2005.
References
Tortorella, D., B. E. Gewurz, M. H. Furman, D. J. Schust, H. L. Ploegh. 2000. Viral subversion of the immune system. Annu. Rev. Immunol. 18: 861-926.
Tay, C. H., R. M. Welsh. 1997. Distinct organ-dependent mechanisms for the control of murine cytomegalovirus infection by natural killer cells. J. Virol. 71: 267-275.
Orange, J. S., B. Wang, C. Terhorst, C. A. Biron. 1995. Requirement for natural killer cell-produced interferon in defense against murine cytomegalovirus infection and enhancement of this defense pathway by interleukin 12 administration. J. Exp. Med. 182: 1045-1056.
Orange, J. S., C. A. Biron. 1996. Characterization of early IL-12, IFN-, and TNF effects on antiviral state and NK cell responses during murine cytomegalovirus infection. J. Immunol. 156: 4746-4756.
Jonjic, S., I. Pavic, P. Lucin, D. Rukavina, U. H. Koszinowski. 1990. Efficacious control of cytomegalovirus infection after long-term depletion of CD8+ T lymphocytes. J. Virol. 64: 5457-5464.
Jonjic, S., W. Mutter, F. Weiland, M. J. Reddehase, U. H. Koszinowski. 1989. Site-restricted persistent cytomegalovirus infection after selective long-term depletion of CD4+ T lymphocytes. J. Exp. Med. 169: 1199-1212.(Theresa A. Banks2,*,, San)