Human cytomegalovirus (HCMV) is an endemic betaherpesvirus with 40-90% seroprevalence globally. Primary HCMV infection is largely asymptomatic but, in common with all herpesvirus infections, lifelong latency ensues and reactivation in immunocompromised individuals can result in significant morbidity. This thesis examines the T cell immune response to HCMV in two settings: latent infection in healthy individuals and virus reactivation in immunosuppressed individuals.
The central role of T cell immunity in controlling HCMV infection is well established. CD4+ and CD8+ T cells can recognise peptide epitopes from a broad range of structural and non-structural HCMV antigens. However, these target antigens are expressed only in lytic and not latent infection and it is unclear whether latently infected cells, which are the main reservoir for virus persistence, can be recognised by T cells. The pattern of virus gene expression in latent infection has been progressively defined in recent years and a number of latency-associated transcripts have been identified, a subset of which have been shown to encode protein products. The first part of this work sought to test the hypothesis that the protein products of latency-associated transcripts are processed and presented to T cells. A panel of healthy HCMV seropositive individuals were screened for T cell responses to two proteins: pUL138 and latency-associated nuclear antigen (LUNA). An HLA-B35-restricted pUL138 CD8+ T cell epitope was identified. This epitope is efficiently presented by HCMV-infected cells and provides the first evidence that T cells may have a role in controlling the latent virus pool. Further studies shows that this epitope can be processed by two distinct pathways: the conventional major histocompatibility complex (MHC) class I pathway and a non-classical pathway that is independent of both the proteasome and the transporter associated with antigen processing (TAP) complex. This latter pathway uses macroautophagy to deliver the protein to the lysosome. Protein processing and peptide loading takes place within the autophagolysosome and the peptide-MHC complexes are then transported to the cell surface by the endocytic pathway. This autophagy-mediated pathway is not as efficient as the conventional pathway in healthy cells. However, it may be of particular significance in HCMV-infected cells because it is better positioned to circumvent the multitude of virus-encoded immune evasion strategies that target the conventional MHC class I machinery.
HCMV reactivation is a significant clinical problem in allogeneic haematopoietic stem cell transplantation (HSCT). Patients with poor T cell immune reconstitution are at particular risk and immunological monitoring may enable a more targeted treatment approach. As the role of immunological monitoring becomes better defined, there is a need to move towards methodologies that can be readily performed in a routine diagnostic laboratory. The second part of this work sought to test the hypothesis that a whole blood interferon-γ secretion assay is useful for the risk stratification of allogeneic HSCT recipients. In a prospective observational study, it was found that the early reconstitution of HCMV-specific immunity did not correlate with a lower risk of HCMV reactivation. This was partly because patients with poor global immunity can rapidly and selectively reconstitute their HCMV-specific immunity in response to incipient virus reactivation and as a result, the selective development of HCMV-specific immunity can herald overt virus reactivation. Once HCMV reactivation has occurred, the development of HCMV-specific immunity is important for virus control and is associated with lower viral loads and lower risks of recurrent or prolonged reactivation. Hence, immunological monitoring can identify patients who are at risk of complicated, but not uncomplicated, HCMV reactivation. Immunological testing in the week following HCMV reactivation is likely to be clinically useful and cost effective, and should be further investigated.
The work presented here extends our knowledge of the breadth and kinetics of HCMV-specific T cell immunity and the pathways through which viral antigens are presented. It provides a platform for further studies into the mechanisms underlying the immune control of the latent virus pool, identification of other latency-associated epitopes, and the development of more effective immunotherapeutic and immune monitoring strategies.