Immune evasion of human cytomegalovirus : studies of UL18 and US2 function

Abstract

The β-herpes virus cytomegalovirus (HCMV) infects human populations at a high frequency worldwide. Primary infection with HCMV, usually asymptomatic, is followed by lifelong latency. During the long co-evolution of virus and host, a fine balance has developed between viral immune evasion strategies and defence mechanisms of the immune system. Yet for individuals with a defective, immature, or compromised immune system, HCMV becomes a serious threat. The general aim of this thesis was to investigate functional and molecular aspects of the HCMV-derived immune-evasion proteins UL18 and US2.

One sophisticated immune evasion strategy of HCMV is interference with MHC class I presentation of viral peptides, using several unique short (US) proteins. We dissected the mechanisms underlying the allele specificity of US2 and demonstrated that a single arginine residue at position 181 (Arg181) was critical for US2-mediated inhibition of HLA-A2 cell surface expression. Binding of US2 to HLA-A2 resulted in a unique, large conformational change of the side chain of Arg181. Even though a prerequisite for the interaction of US2 with HLA-A2, the presence of this residue is not sufficient to guarantee binding to other MHC class I alleles.

Another suggested immune modulator with ambiguous function is the protein UL18, a viral MHC class I homologue that associates with β2 microglobulin (β2m). UL18 binds with high affinity to the leukocyte immunoglobulin-like receptor-1 (LIR-1). We tested the impact of several substitutions in UL18 proteins for binding affinity to LIR-1. Our results revealed that residues localized both in the α1 and α3 domain are important for LIR-1 binding, and demonstrated β2m dependency of the UL18/LIR-1 interaction. Finally, two disulfide bridges, one of them unique for UL18, were essential for complex formation of UL18 with β2m.

Since LIR-1 is widely expressed on immune cells, we investigated if UL18 could affect dendritic cells (DCs). We demonstrate that UL18 proteins specifically up- regulated CD83, while not influencing other maturation markers. UL18 also induced IL-10 production and to some extent other cytokines such as TNFα and IL-12. The presence of UL18 during DC maturation via CD40L inhibited DC migration and impaired subsequent T cell responses. We concluded that UL18 can alter phenotype and function of monocyte-derived DC.

The expression of LIR-1 particularly on HCMV specific T cells as well as on NK and T cells in lung-transplanted patients prior to development of CMV caused pneumonia supports the hypothesis that the UL18-LIR-1 interaction may be relevant during natural infection. Therefore, we focused on the role of LIR-1 in the immune response to HCMV and potential effects of UL18 on NK and T cells. Cells infected with a virus lacking the gene for UL18 induced less cytokine responses compared to parental virus, proposing an activating function for UL18. In contrast, isolated UL18 proteins inhibited LIR-1+ T cells, which unlikely played a role in response to infected cells.

In summary, this thesis provides further insights into the mechanism underlying the binding of US2 to HLA-A2 and demonstrates that UL18 interacts with LIR-1 in a manner different from MHC class I ligands. Furthermore, our results regarding the effect of UL18 on several immune cells contributes to a better understanding of the complexity of viral evasion mechanisms.