Increased permeability of human endothelial cell line EA.hy926 induced by hantavirus-specific cytotoxic T lymphocytes

Abstract

Hantavirus infection causes two human diseases, hemorrhagic fever with renal syndrome and hantavirus pulmonary syndrome. The typical feature of these diseases is increased permeability in microvascular beds in the kidneys and the lungs, respectively. The mechanism of capillary leakage, however, is not understood. Some evidence suggests that hantavirus disease pathogenesis is immunologically mediated by cytotoxic T lymphocytes and other immune cells in target organs producing inflammatory cytokines. In this study we examined the roles of virus-specific cytotoxic T lymphocytes in increased permeability of human endothelial cells infected with hantavirus. We used a human CD8⁺ hantavirus-specific cytotoxic T lymphocyte line, 1A-E2, specific for the HLA-A24-restricted epitope in Sin Nombre and Puumala virus G2 protein, and the human endothelial cell line, EA.hy926 that expresses HLA-A24 molecule. The cytotoxic T lymphocyte line recognized and lysed target cells infected with Sin Nombre virus, and in transwell permeability assays increased permeability of EA.hy926 cell monolayer infected with Sin Nombre virus or recombinant adenovirus expressing the Sin Nombre virus G2 protein. These results suggest that cytotoxic T lymphocyte activity contribute to capillary leakage observed in patients with hantavirus pulmonary syndrome or hemorrhagic fever with renal syndrome.

Introduction

Hantaviruses, belonging to the family Bunyaviridae, are distributed worldwide. Two forms of zoonotic human diseases are caused by hantavirus species (Enria et al., 2001, Linderholm and Elgh, 2001, Schmaljohn and Hjelle, 1997). Hemorrhagic fever with renal syndrome (HFRS) and its mild form nephropathia epidemica (NE) are caused by Old World (Europe and Asia) hantaviruses, such as Hantaan, Seoul, Dobrava and Puumala viruses. Hantavirus pulmonary syndrome (HPS) is caused by New World (North and South America) hantaviruses, such as Sin Nombre and Andes viruses (Khaiboullina and St Jeor, 2002, Zeier et al., 2005). The hantavirus genome consists of three RNA segments, large, medium and small segments, encoding RNA-dependent RNA polymerase, envelope glycoproteins (G1 and G2), and nucleocapsid (N) protein, respectively (Jonsson and Schmaljohn, 2001). Hantaviruses persistently infect their natural rodent reservoirs without apparent diseases (Meyer and Schmaljohn, 2000). Humans are infected with hantaviruses by direct contact with infected rodents or through the inhalation of excreted viral aerosols (Linderholm and Elgh, 2001).

Human hantavirus diseases are characterized by an increased permeability in microvascular beds of the kidneys in HFRS and the lungs in HPS, and endothelial cells are considered to be the primary targets of hantavirus infection (Hautala et al., 2002, Khaiboullina and St Jeor, 2002, Mustonen et al., 1994, Temonen et al., 1996, Zaki et al., 1995). In vitro hantavirus infection alone, however, did not induce visible cytopathic effects in cultured human endothelial cells (Geimonen et al., 2002, Niikura et al., 2004, Pensiero et al., 1992) nor did it increase capillary permeability of an infected endothelial cell monolayer (Khaiboullina et al., 2000, Niikura et al., 2004, Sundstrom et al., 2001). Increased levels of cytokines including tumor necrosis factor (TNF) -α, interleukin (IL) -2, IL-6 and interferon (IFN) -γ have been detected in sera of HFRS and HPS patients (Khaiboullina and St Jeor, 2002, Vapalahti et al., 2001). An expanded leukocyte population including monocytes, T and B cells and an increase of CD8⁺/CD4⁺ ratio have been observed in HFRS patients (Chen and Yang, 1990, Huang et al., 1994, Lewis et al., 1991, Markotic et al., 1999). We found significantly higher frequencies of Sin Nombre virus (SNV)-specific T cells in patients with severe HPS requiring mechanical ventilation (up to 44.2% of CD8⁺ T cells) than in moderately ill HPS patients hospitalized but not requiring mechanical ventilation (up to 9.8% of CD8⁺ T cells). These results suggest that virus-specific CD8⁺ T cells contribute to HPS disease outcome (Kilpatrick et al., 2004). Increased numbers of CD8⁺ T cells are found in the kidneys of HFRS (Temonen et al., 1996) and in the lungs of HPS patients (Zaki et al., 1995). There is an abundance of immune cells expressing a variety of cytokines in the lungs of HPS cases (Mori et al., 1999, Zaki et al., 1995) and in the kidneys of NE cases (Temonen et al., 1996). In addition, preliminary evidence suggests that HLA-B*3501 is associated with severe HPS in SNV infection, implying involvement of CD8⁺ cytotoxic T lymphocytes (CTLs) (Kilpatrick et al., 2004) (Koster et al., 2001). A similar linkage between disease severity and MHC haplotype was observed between Puumala virus infection and the HLA-B8-DR3 extended haplotype (severe outcome) or the HLA-B27 (milder disease) (Makela et al., 2002, Mustonen et al., 1998). These reports suggest that host immune responses against hantavirus, especially virus-specific CTLs and inflammatory cytokines produced by virus-specific T cells may contribute to disease pathogenesis of HPS and HFRS (Terajima et al., 2004).

We have established panels of CTL lines from the PBMC of NE and HPS patients (Ennis et al., 1997, Kilpatrick et al., 2004, Terajima et al., 2002, Terajima et al., 2004, Van Epps et al., 1999, Van Epps et al., 2002). These cell lines were able to recognize and lyse autologous B lymphoblastoid cell lines pulsed with the epitope peptide or infected with recombinant vaccinia viruses expressing the hantaviral protein containing the epitope. In this report, we tested one of these CTL lines against human endothelial cell line, EA.hy926, infected with SNV in vitro. Primary human umbilical vein endothelial cells (HUVECs) or human lung microvascular endothelial cells (HMVEC-Ls) have been used to examine the effect of hantavirus infection on endothelial cell function (Gavrilovskaya et al., 2002, Khaiboullina et al., 2000, Niikura et al., 2004, Sundstrom et al., 2001). Although ideally the interaction between hantavirus-specific CTL and hantavirus-infected endothelial cells should also be analyzed using HUVECs or HMVEC-Ls, it is difficult to obtain endothelial cells that express the MHC class I molecules by which our CTL lines were restricted. Therefore, we used the immortalized human endothelial cell line EA.hy926 (Edgell et al., 1983), which we found to express the HLA-A24 allele (in this report), and one of our hantavirus-specific CTL line, 1A-E2, which was established from convalescent PBMC from the patient infected with Puumala virus and was restricted by HLA-A24 (Terajima et al., 2002). 1A-E2 recognized the epitope, HWMDATFNL, encoded by Puumala virus G2 protein, and was cross-reactive to the corresponding peptide, HWMDGTFNI, in SNV G2 protein (Terajima et al., 2002). EA.hy926 cells are the most similar to HUVEC among the available immortalized human endothelial cell lines and have been used to study the endothelial cell/leukocyte interactions (Lidington et al., 1999).

In this study, we first tested the infectivity of SNV to EA.hy926 cells. Next, we showed that the hantavirus-specific CD8⁺ T cell line 1A-E2 could recognize and lyse EA.hy926 cells infected with SNV or presenting SNV antigen. Finally, we demonstrated that 1A-E2 enhanced the permeability of EA.hy926 cells infected with SNV. These results suggest that virus-specific CTLs contribute to the capillary leakage.