The Change Of CD100Expression During The Course Of Hemorrhagic Fever With Renal Syndrome And Its Correlation With The Outcome Of The Disease

Hantaan virus (HTNV), which belongs to the genus Hantavirus of the familyBunyaviridae, could cause a severe lethal hemorrhagic fever with renal syndrome (HFRS)in human. More than100,000cases of HFRS, over90%of which were documented inmainland of China, occurred annually worldwide with a mortality rate of2-10%. Peoplewith HFRS are clinically characterized by sudden fever, hemorrhage, thrombocytopenia,and acute renal failure, leading from an asymptomatic to a severe, life-threatening illness.Typically, the course of HFRS undergoes five sequential stages: febrile, hypotensive,oliguric, diuretic, and convalescent. Although the importance of immune responses afterHTNV infection has been widely recognized including immune complexes, complementactivation, B cell response, T cell response, and HTNV-induced cytokine production, notonly the pathogenesis of HFRS but also the protective immunity dediated by cytotoxic Tlymphocyte (CTL) during the cause of the diease are considerably far from beingcompletely understood. The150kDa transmembrane protein CD100/Sema4D belongs to group IV of thesemaphorin family, which is the first known semaphorin identified in the immune system,and is involved in several aspects of both humoral and cellular immunity. It exists in bothmembrane-bound and soluble forms. The membrane CD100is preferentially expressed onT cells and weakly on B cells and APC. Cellular activation can cause the release ofsCD100. sCD100is demonstrated to retain biological activities such as acting as acostimulator for CD40-induced B-cell proliferation and Ig production, affectingpro-inflammatory cytokines production by APCs. There are two types of receptors thatCD100used to bind: Plexin-B1mainly expressed in nonlymphoid tissues, and CD72inthe immune system. Accumulating evidence indicates that CD100plays a relevant role inphysiological and pathological immune responses. CD100-/-mice are viable, but showdefective T cell priming and B cell responses, whereas adaptive immune responses aresignificantly enhanced in CD100transgenic mice. Furthermore, CD100is believed to beinvolved in some clinical diseases. In the spinal cords of patients with human T-celllymphotropic virus type1-associated myelopathy, the presence of sCD100in the spinalcord suggested the potential pathological effect of sCD100in the central nervous system.In the systemic sclerosis patients, the frequently detectable levels of sCD100in sera andthe dysregulations of CD100expression on lymphocytes were observed, suggesting therole of sCD100in the systemic sclerosis development and/or maintenance. Recently,Eriksson et al investigated the consequence of HIV-1infection on CD100expression by Tcells, and observed a subset of CD8~+T cell lacking of membrane CD100with decreasedfunctional capacity, which suggested that loss of CD100expression would probably playan important role in dysfunctional immunity in HIV-1infection. However, there is stilllimited information on the functional role of CD100in infectious disease. Whether thispathogenetic role of CD100or involvement of CD100expression in CTL subset andfunction could extend to other acute infectious diseases mediated by immune responses isalso unclear.Since the important role of CD100in immune response, we focused on two aspects:whether sCD100release after HTNV infection exist and changed level of plasma sCD100 correlate with the outcome of HFRS, and whether the changes of mCD100expression onCTL are related to CTL subsets and function.Plasma and peripheral blood mononuclear cell (PBMC) samples from a large cohortof HFRS patients and health controls were collected. First, the plasma sCD100levels indifferent disease stages or severities of HFRS patients were quantified by enzyme-linkedimmunosorbent assay (ELISA) which has been established successfully in this study. Thecorrelations between sCD100and disease course, disease severity-indicating parameterswere also analyzed. Second, the changes of mCD100on CD4+T cells, CD8~+T cells andother population in PBMC were investigated by multi-color staining and flow cytometry(FCM) analysis. Particularly, we payed more attention on a novel CD8lowCD100-subsetand detected its phenotype, function as well as the relationship between the distribution ofCD8lowCD100-subset in PBMC and plasma HTNV RNA load or disease severity duringthe course of HFRS.According to the clinical records and diagnostic criteria,17,25,29, and28patientswere diagnosed as mild, moderate, severe, and critical HFRS, respectively. The median(IQR) of sCD100for febrile/hypotensive, oliguric, diuretic and convalescent stage was42.8(26.5-55),34.8(21.3-52.6),12.1(8.5-18.2), and15.9(8.6-21.4) ng/ml, respectively.The elevated plasma sCD100level of HFRS patients was observed in acute phase(including febrile, hypotensive, or oliguric stage) when compared with healthy controlsand decreased, but still higher than that of healthy controls in convalescent phase(including diuretic or convalescent stage). Patients with different disease severity showedthe same tendency of the plasma sCD100change, but more dramatic decline in patients ofsevere/critical group. When plasma sCD100concentrations in mild/moderate group werecompared with those in severe/critical group, only7cases whose plasma sCD100levelswere over50ng/ml measured among38cases of mild/moderate group (18.4%), while26cases whose plasma sCD100levels were over50ng/ml in61cases of severe/critical groupcould be detected (42.6%,2.32fold high vs. mild/moderate group). In addition, wedetected the plasma sCD100by Western Blot, which showed that the sCD100level washigher in acute phase than in convalescent phase apparently. The molecular weight of the sCD100in plasma from both patients and normal controls is120kDa in the reducedcondition which is consistent with the size of extracellular region of CD100. Furthermore,the relationships between plasma sCD100levels at febrile or hypotensive stages (the timeof admission generally about3-7days after the onset of disease) and four clinicalparameters that could represent the severity of the disease were analyzed. The resultrevealed a significant negative correlation between plasma sCD100levels and the plateletcount (r=-0.50, P=0.0001) and significant positive correlations between plasma sCD100levels and the white blood cell count (r=0.54, P <0.0001), or the level of serumcreatinine (r=0.49, P=0.0001), or the level of blood urea nitrogen (r=0.56, P=0.0002).The membrane CD100on PBMCs was also investigated. We found the expressionof membrane CD100on PBMCs including CD4+T cells, CD8~+T cells, B cells, naturalkiller cells and monocytes all decreased to different extents in the acute phase of HFRScompared with that of the normal controls and recovered in the convalescent phase.For the first time, we revealed a novel functional subpopulation in CD8~+T cells inHFRS patients characterized by the phenotype of CD8lowCD100-. The average distributionof CD8lowCD100-subpopulation in PBMC was10%in acute phase and2.6%inconvalescent phase, whereas this subset did not exist in health controls. Furthermulti-color staining and flow cytomety analysis demonstrated that CD8lowCD100-subsetexpressed higher level of CD38+HLA-DR+Ki-67+than the other two subsets did, highlyexpressed cytolytic effector molecules including perforin and granzyme B, displayed thephenotype CD45RA-CD127int/highCD27lowCD62L-, which were consistent with the majorfeature of effector CD8~+T cells. This observation was supported by the investigations ofthe relationship between the distribution of CD8lowCD100-subset in PBMC and plasmahantaan virus load of HFRS patients, and by cytokine production of different CD8~+T cellsubsets when stimulated by specific HTNV-NP derived9mer peptide pool. Thepercentage of CD8lowCD100-subset in PBMC was negatively correlated with plasmaHTNV virus load (P<0.0001) in acute phase of HFRS. In addition, the percentage ofCD8lowCD100-subset in PBMC of mild group was higher than that of moderate, severe orcritical gruops (P<0.05). TNF-and IFN-γ are key effector cytokines for effector CD8~+ cells. In vitro CD8lowCD100-subset produce little TNF-and IFN-γ without thestimulation. However, when stimulated by the HTNV-NP9mer peptide pool, theCD8lowCD100-cells produced a large amount of TNF-and IFN-γ identified byintracellular cytokine staining and FCM, whereas other two subsets of CD8~+T cells(CD8highCD100+and CD8lowCD100+) produced small amount of TNF-and IFN-γ. Withthe clearance of HTNV virus and the number of CTL returned to normal level, and thenumber of CD8lowCD100-T cells decreased dramatically. This kinetic change ofCD8lowCD100-T cells may be due to clonal contraction of effecter T cells via PD-1pathway and reflects the homeostasis pathway of immune system after virus infection.In summary, the elevated sCD100in plasma seemed to be a characteristic in HFRSpatients especially in the acute phase of the disease indicating a possible associationbetween increased release of sCD100and different disease severity in HFRS. Forthermore,a novel functional CD8lowCD100-subset in PBMC from HFRS patients was negativelycorrelated with plasma HTNV virus load and severity of the disease severity. These resultsare useful for understanding the pathogenesis and CD8~+T cell mediated immunity as wellas immune homeostasis after HTNV infetion in human.