CMV can replicate in a wide variety of cells and tissues in the host, including fibroblasts, myeloid-derived cells, endothelial cells, and a variety of epithelial cells. In the past few years we have discovered that vector tropism plays a major role in CMV vector immunogenicity and efficacy. We found that the 68-1 RhCMV/SIV vectors that are associated with stringent protection against highly pathogenic SIV challenge elicit SIV-specific CD8+ T cells that were highly unusual in their epitope targeting. Rather than recognize SIV peptides in the context of classical MHC-Ia molecules, these CD8+ T cells broadly targeted SIV epitopes presented in the context of either MHC-II or non-classical MHC-E molecules. This unusual response phenotype was found to be due to the loss of the Rh157.5 (UL128) and Rh157.4 (UL130) genes in the 68-1 RhCMV vector, as repair of these genes in the 68-1.2 RhCMV vectors reverted the vector-elicited CD8+ T cell responses back to conventional MHC-Ia restriction. Remarkably, in 2 independent experiments, these Rh157.5/Rh157.4-repaired 68-1.2 RCMV vectors, though equivalently immunogenic in terms of response magnitude and phenotype as their ?Rh157.5/Rh157.4 68-1 RhCMV vector counterparts, were found not to protect against SIV challenge, strongly suggesting that the unconventionally restricted CD8+ T cells associated with 68-1 vectors are required for efficacy. Rh157.5 and Rh157.4 are 2 essential components of the pentameric receptor complex (PRC) involved in facilitating CMV infection of non-fibroblasts. Further work demonstrated that specific loss of PRC function partially recapitulates the unconventional CD8+ T cell responses of 68-1 RhCMV vectors, resulting in vectors that elicit a mixture of MHC-II- and MHC-Ia-restricted responses. This phenotype was mimicked by PRC-independent modification of the tropism of PRC?intact RhCMV vectors using microRNA (miR)-142 restriction, which prevents vector infection of myeloid-derived cells, suggesting that absence of PRC function in RhCMV vectors (and the resultant less efficient vector infection of myeloid cells) promotes the induction of MHC-II-restricted CD8+ T cell responses. Interestingly, MHC-E-restricted CD8+ T cell responses appear to require infection of myeloid-derived cells for their generation, as MiR-142 restriction of the strain 68-1 RhCMV vector completely prevents MHC-E-restricted response priming, leaving MHC-II response priming intact. In this project, we seek to define the role of vector tropism in CMV vector CD8+ T cell response programming, and to exploit this biology to focus vector immunogenicity on the most efficacious response type. We will 1) determine how in vivo tropism of CMV vectors that differentially induce conventionally (MHC-Ia) restricted vs. unconventionally (MHC-E- and MHC-II-) restricted CD8+ T cell responses differ in infected tissues, 2) define the target cell tropism requirements for CMV vector-elicited MHC-E- and MHC-II-restricted CD8+ T cell responses using cell type-specific miRs, and 3) develop CMV vectors that preferentially or exclusively elicit MHC-E- and MHC-II-restricted CD8+ T cell responses by strategic modification of vector tropism.
|Walters, Lucy C; Harlos, Karl; Brackenridge, Simon et al. (2018) Pathogen-derived HLA-E bound epitopes reveal broad primary anchor pocket tolerability and conformationally malleable peptide binding. Nat Commun 9:3137