The Role Of B And T Lymphocyte Attenuator In MHV-3Induced Murine Fulminant Hepatitis

Fulminant viral hepatitis (FH) is a clinical syndrome characterized by massivehepatocyte necrosis, severe impairment of liver function and hepatic encephalopathy duringvirus infection. In China and many other Asian countries where viruses such as hepatitis Bvirus (HBV) are epidemic and FH is one of the most common diseases requiring immediatehospitalization. Despite rapid advances in treatment options, including liver transplantation,the mortality due to FH remains very high. More pervasive and efficient treatment of FHrelies on well understand of the immunological mechanisms of FH. At present, studiesusing a FH model induced by murine hepatitis virus-3(MHV-3) infection have providedsignificant insights into the mechanisms underlying the pathogenesis of FH and thedevelopment of novel therapeutics to treat the disease. MHV-3is a single-stranded,positive-sense RNA virus that belongs to the coronavirus family. The virus causeshepatocellular necrosis and mortality in susceptible BALB/cJ and C57BL/6mice, whereasA/J mice are completely resistant to infection. Previous study indicated that thefibrinogenlike protein-2(FGL2)prothrombinase plays an important role in MHV-3-inducedfulminant hepatitis.B and T lymphocyte attenuator (BTLA), a recently identified member of CD28superfamily, is a negative regulator of T cell activatioin and proliferation. BTLA caninhibit CMV-specific CD8+T cell proliferation and IL-2secretion through ligation with itsligand HVEM, a member of the tumor necrosis factor receptor (TNFR) superfamily.Previous study demonstrated that BTLA play an important role in various disease, such asexperimental autoimmune encephalomyelitis (EAE), allergic airway inflammation,autoimmune hepatitis and MHC-mismatched allograft rejection. BTLA seems to be anegative regulator in infectious disease models. For example, it has been observed thatactivated cytomegalovirus (CMV)-specific CD8+T cells express BTLA and blockade ofBTLA signaling augments the proliferation of these T cells, indicating that BTLA inhibits the virus specific CTL function. There is also evidence that the interruption of BTLAsignaling promotes the early clearance of Listeria bacterial infection. These studiesgenerally indicate that BTLA acts to attenuate immune responses. We have previouslyshowed that the co-inhibitory signals supplied by programmed death-1(PD-1) can reducetissue damage. Therefore, PD-1-deficient (PD-1-/-) mice exhibit a significantly higher rateof mortality than their WT littermates following MHV-3infection. This finding implies thatCD28superfamily proteins participate in the regulation of FH pathogenesis. Nevertheless,the biological significance of BTLA in the pathogenesis of FH remains uninvestigated.In this study, we examined the role of BTLA in MHV-3-mediated FH. We firstinfected the WT and BTLA-/-mice with30PFU of MHV-3viruses. Surprisingly, wefound that all of the WT mice died within9days post-infection, whereas in the BTLA-/-group, over87.5%of the mice survived after20days post infection. Severe necrosis wasobserved by H&E staining in the livers and spleens of WT mice, but the morphology ofthese organs from the infected BTLA-/-mice was mostly normal under the same conditionsof infection. In agreement with the observation that fewer hepatotyes were undergoingapoptosis, BTLA-/-mice exhibited lower alanine aminotransferase (ALT) and aspartateaminotransferase (AST) compared to the WT mice. FGL2is a critical molecule thatmediates hepatocellular necrosis. We therefore measured the expression of FGL2in theliver tissues of infected mice. Compared with WT mice, there were much lower fgl2mRNAin BTLA-/-mice liver and much lower FGL2in BTLA-/-mice serum and liver after MHV-3infection. Immunofluorescence showed that FGL2was produced by CK-18+epithelial cells,CD68+macrophages, CD11c+DCs, CD31+endothelial cells and, to a lesser extent, CD3+Tcells. Immunohistochemistry demonstrated that following MHV-3infection, there was amassive distribution of FGL2+cells accompanied by severe tissue damage in WT mice,whereas there were only a few detectable FGL2+cells in BTLA-/-mice. The high levels ofFGL2in the WT mice also caused a higher deposition of fibrinogen in the affected liver andspleen tissues. These results clearly show that BTLA, unexpectedly acting as an enhancerrather than an attenuator of pathology, plays an important role in the development of FHand causes high mortality upon MHV-3infection. Mice lacking BTLA signals exhibitsignificantly reduced FGL2expression in response to MHV-3infection and thus have lessfibrinogen deposition, which reduces tissue damage and mortality in FH. To understand the mechanisms of BTLA-mediated tissue damage and FH mortality, weassessed the expression and anatomical distribution of BTLA during MHV-3infection.Immunofluorescence showed that BTLA+cells present in many organs including the liver,spleen and lymph nodes. Flow cytometry revealed BTLA expression was drasticallyreduced in the periphery T and B cells but significant increase in peritoneal F4/80+macrophages24h post infection. These results suggest that BTLA might exert unidentifiedfunctions in macrophages, particularly during MHV-3infection. Flow cytometry analysesshowed that macrophages from WT and BTLA-/-mice express similar levels ofcostimulatory molecules, apoptosis-associated molecules, such as Fas and FasL, andpro-inflammatory factors including TNF-α and IFN-γ. However, by examining theviability of macrophages derived from BTLA-/-and WT mice, we found that upon MHV-3infection, peritoneal macrophages from BTLA-/-mice underwent a significantly higher rateof apoptosis, resulting in a decline in the numbers of macrophages in the peritoneum. Thereduction in peritoneal macrophages further causes a reduction in the numbers ofmacrophages infiltrating the liver. In agreement with these data, we also observed anaccelerated apoptosis of MHV-3-infected BTLA-/-macrophages ex vivo. These datasuggest that the rapid loss of macrophages in BTLA-/-mice is responsible for the reductionin inflammatory responses and FGL2secretion, all together alleviating the liver damageduring MHV-3infection.To examine whether the early virus-induced loss of macrophages in BTLA-/-miceaccounts for the reduction in inflammatory cytokine production and diminished FGL2synthesis, we found that the mRNA for TNF-α, but not Foxp3, IFN-γ, IL-2, IL-4, IL-10,IL-17or TGF-β, was significantly lower in BTLA-/-mice compared with WT mice. Atsingle cell level, it appears that macrophages from BTLA-/-mice secrete more TNF-α uponMHV-3infection, nevertheless, the serum level of TNF-α in infected BTLA-/-mice appearsto be significantly lower than the WT controls. This drop in total TNF-α production is likelydue to macrophage apoptosis. The treatment which BTLA-/-mice infected with MHV-3andsimultaneously treated them with TNF-α (100ng/g) resulted in a reversion of the protectivephenotype observed in BTLA-/-mice, causing severe liver tissue necrosis, high levels ofALT and FGL2in the serum, and increased mortality. TNF-α-/-and their WT littermatemice infected with MHV-3and the result was similar to BTLA-/-and their WT littermate. These results suggest that TNF-α, a major inflammatory cytokine produced by macrophages,plays a central role in the BTLA-dependent liver damage and mortality in the MHV-3infected mice. Higher rate of macrophage apoptosis in BTLA-/-mice causes a reduction inTNF-α secretion, leading to a diminished inflammatory response upon MHV-3infection.To further demonstrate that the abrogation of BTLA expression in macrophagesaccounts for the resistance of BTLA-/-mice to MHV-3infection, we isolated macrophagesfrom WT mice, and then injected these macrophages into the BTLA-/-mice that weresubjected to MHV-3infection. This transfer resulted in a significant acceleration ofmortality in BTLA-/-mice. Further experiments demonstrated that the in vivo injection of aBTLA-blocking mAb (clone6F7) was able to significantly prolong the survival of thevirus-infected WT mice.Taken together, we conclude that the BTLA signaling pathway plays a decisive role inregulating macrophage viability, and by thus it affects the severity of liver pathology andanimal mortality during MHV-3infection.