Date Published: July 6, 2017
Publisher: Public Library of Science
Author(s): Adriana Tovar-Salazar, Adriana Weinberg, Frederic Rieux-Laucat.
Cytomegalovirus (CMV) infection is associated with immune-suppression in immune-compromised hosts and old adults. We previously showed that ex vivo CMV restimulation of peripheral blood mononuclear cells (PBMC) of CMV-seropositive volunteers expanded CD4+CD27-CD28- regulatory T cells (Tregs). Here we evaluate the phenotype and function of circulating CD4+CD27-CD28- T cells of CMV-seropositive adults. Compared with CMV-seronegative, CMV-seropositive adults had 10-fold higher CD4+CD27-CD28-% T cells in PBMC. Circulating CD4+CD27-CD28- T cells from both CMV-seropositive and seronegative donors expressed higher levels of TGFβ, granzyme B, CD39, CD147 and IL-35, and lower levels of CD127, compared with their parent circulating CD4+ T cells. However, only CMV-seropositive circulating CD4+CD27-CD28- had increased FOXP3 expression. CD4+CD27-CD28- sorted from the PBMC of CMV-seropositive donors expanded ex vivo in the presence of rhIL2 and inhibited ex vivo proliferation of autologous PBMC restimulated with CMV, varicella-zoster virus or C. albicans antigens. CD4+CD27-CD28- sorted from CMV-seronegative PBMC did not expand in the presence of rhIL2 and did not inhibit autologous PBMC proliferation. CD3+CD27-CD28- circulating T cells (≥80% CD8+) from CMV-seropositive HIV-infected donors also inhibited ex vivo proliferation of autologous PBMC restimulated with CMV or HIV. These data indicate that CMV-seropositive individuals have circulating Tregs that inhibit cell-mediated immune responses to CMV and other antigens and may be contribute to an immune-suppressive effect of CMV infection. Moreover, the phenotypic similarity between circulating CD4+CD27-CD28- Tregs with differentiated effector T cells suggests that the two T-cell subsets might evolve in parallel or in sequence from the same progenitor cells in response to CMV stimulation during reactivations.
CMV infection is associated with immune-suppression. The mechanism underlying this effect is unknown. We determined that peripheral blood CD4+CD27-CD28- T cells, which generally represent effector T cells, of CMV-seropositive adults have regulatory function, which may explain the association of CMV seropositivity, high CD4+CD28- T cell frequencies and immune-suppression. Notably, we did not find regulatory T cells among peripheral blood CD4+CD27-CD28- cells of CMV-seronegative individuals. In HIV-infected CMV-seropositive individuals with low numbers of CD4+ cells, CD3+CD27-CD28- T cells (mostly CD8+CD27-CD28-) have regulatory T-cell function. Immune-suppression of CMV-seropositive old adults has also been associated with high CD4+CD28- or CD8+CD28- T-cell numbers. We propose that the regulatory CD4+CD27-CD28- and CD8+CD27-CD28- T cells of CMV-seropositive individuals contribute to the immune suppression associated with CMV infection. The data suggest that CMV effector and regulatory T cells may evolve together and that decreasing CMV T cell stimulation might limit the generation of regulatory T cells.
Cytomegalovirus (CMV)-infected individuals with cell-mediated immune disorders, such as transplant recipients and human immunodeficiency virus (HIV)-infected individuals, or at the extremes of age without any age-independent immune disorders have higher morbidity and mortality than uninfected individuals [1–7]. Most reports found an association of CMV infection with increased HIV disease progression and death in HIV-infected individuals, increased bacterial and fungal superinfections in transplant recipients and decreased immune responses to vaccines and increased respiratory infection in older adults [7–9], although there have also been some exceptions [10, 11]. Overall, the data suggest that CMV has an immune suppressive effect on the host.
Our data demonstrate that circulating CD4+CD28-CD27- T cells of CMV-pos donors have functional and phenotypic Treg characteristics, whereas circulating CD4+CD27-CD28- T cells of CMV-neg donors share selected phenotypic but not functional characteristics with those of CMV-pos donors. The phenotypic characterization used a broad panel of Treg markers, including FOXP3, which is a transcription factor that activates the regulatory program in T cells and is therefore considered the Treg hallmark; CD39, IL-35 and granzyme B, which are functional mediators of the Treg inhibitory activity; and TGFβ, which has both Treg activating and effector functions. Another Treg phenotypic characteristic found in the CD4+CD27-CD28- peripheral blood Treg was the low expression of CD127. Although many of these markers were shared by CMV-pos and CMV-neg donors, FOXP3 was found only in the circulating CD4+CD27-CD28- of CMV-pos donors. Moreover, only peripheral blood CD4+CD27-CD28- T cells from CMV-pos individuals proliferated in culture in the presence of rhIL2 and inhibited ex vivo proliferation of autologous PBMC. Taken together, these data indicate that the CD4+CD27-CD28- T cells of CMV-pos donors include Tregs, but those of CMV-neg individuals do not.