Generation, maintenance and tissue distribution of T cell responses to human cytomegalovirus in lytic and latent infection

Abstract

Understanding how the T cell memory response directed towards human cytomegalovirus (HCMV) develops and changes over time while the virus persists is important. Whilst HCMV primary infection and periodic reactivation is well controlled by T cell responses in healthy people, when the immune system is compromised such as post-transplantation, during pregnancy, or underdeveloped such as in new-born infants and children, CMV disease can be a significant problem. In older people, HCMV infection is associated with increased risk of mortality and despite overt disease rarely being seen there are increases in HCMV-DNA in urine of older people suggesting that there is a change in the efficacy of the T cell response following lifelong infection. Therefore, understanding whether phenomenon such as "memory inflation" of the immune response is occurring in humans and if this is detrimental to the overall health of individuals would enable the development of appropriate treatment strategies for the future. In this review, we present the evidence available from human studies regarding the development and maintenance of memory CD8 + and CD4 + T cell responses to HCMV. We conclude that there is only limited evidence supportive of "memory inflation" occurring in humans and that future studies need to investigate immune cells from a broad range of human tissue sites to fully understand the nature of HCMV T cell memory responses to lytic and latent infection.

Keywords: Human cytomegalovirus (HCMV); Inflation; Latency; T cell memory.

Fig. 1

CD4 + and CD8 + T cell IFN-γ responses to pp65, gB, IE1 and IE2 HCMV proteins. The T cell IFN-γ responses measured by FluoroSpot versus donor age from the ARIA study [23] to four lytic expressed HCMV proteins pp65 (green), gB (light blue), IE1 (dark blue) and IE2 (purple) are shown. There is no significant correlation (Spearman rank correlation test) between the magnitude of the IFN-γ response with donor age for any of the four proteins shown

Fig. 2

CMV IgG response related to age and latent CMV load in CD14 + monocytes. Serum HCMV IgG levels [immune system ratio (ISR)] from the ARIA study [23] related to donor age (a); there is no significant correlation (Spearman rank correlation test). Latent CMV load in CD14 + monocytes measured by droplet digital PCR as previously described [23] is shown related to low (< 3.5 ISR), medium (3.5–4.99 ISR) and high (> 5 ISR) CMV IgG levels (b); there is no significant difference in the magnitude of the latent viral load between the three groups (Kruskal–Wallis ANOVA test and Dunn’s multiple comparison post-test)

Fig. 3

CD4 + T cell IL-10 responses to UL138, LUNA, US28, vIL-10, US3 and pp71 HCMV proteins. The CD4 + T cell IL-10 responses measured by FluoroSpot versus donor age from the ARIA study [23] to six HCMV proteins UL138 (turquoise circles), LUNA (turquoise squares), US28 (blue triangles), vIL-10 (blue inverted triangles), US3 (green diamonds) and pp71 (orange hexagons) are shown. There is no significant correlation (Spearman rank correlation test) between the magnitude of the IL-10 response with donor age for any of the six proteins shown

Fig. 4

CD8 + T cell IFN-γ responses and CD4 + T cell IFN-γ and IL-10 responses in paired peripheral blood and bone marrow samples. IFN-γ and IL-10 T cell responses to 11 HCMV proteins were measured by FluoroSpot as previously described [23] in peripheral blood and bone marrow sample from the same donor. The CD8 + T cell IFN-γ (top graph—pink bars), the CD4 + T cell IFN-γ (middle graph—blue bars) and the CD4 + T cell IL-10 (bottom graph—turquoise bars) responses in the two compartments are shown

Fig. 5

Illustration of the viral dissemination assay model system. A schematic of the generation of autologous primary dermal fibroblasts from a 2-mm punch skin biopsy sample and a microscope image (×10 magnification—bright field) of the fibroblasts growing from a portion of the skin biopsy are shown (a). A pictorial representation of the experimental protocols of the viral dissemination assay (b) and the myeloid cell-based modified viral dissemination assay (c) are shown