Immune activation is a cardinal feature of HIV disease and contributes to pathologic outcomes before and after the initiation of ART. In each instance, it is not clear if immune activation is due directly to viral replication and/or to the host response to such replication. It is clear, however, that the higher the level of activation, the faster the course of disease progression in untreated subjects[13, 20] and the lower the level of CD4+ T cell recovery in those provided ART.
We demonstrate that suppressive ART restores homeostatic levels of monocyte population frequencies (Figure 1B, C). There may be multiple explanations for these findings. For example, diminished levels of CD14+ myeloid cells could arise if CD14 cell surface receptors are actively shed after activation. The CD14 molecule serves as the bacterial lipopolysaccharide (LPS) receptor that is cleaved when engaged by LPS, resulting in circulating soluble CD14. The increase in the frequency of classical monocytes observed after suppressive ART may reflect decreased plasma LPS, allowing persistent expression of CD14 on these cells. Moreover, and as suggested by the current studies, expression of immunomodulatory enzymes such as HO-1 by classical monocytes may contribute to control of immune activation after ART-mediated viral suppression. Cellular phenotyping of monocytes was performed on cryopreserved samples, and previous studies have comparable results upon the freeze-thawing process of peripheral blood mononuclear cells[22, 23].
We hypothesized that frequencies of circulating monocyte subpopulations may not only be differentially altered during the course of HIV disease but may also predict the rate of CD4+ T cell recovery in patients that are suppressed on ART. Our study was designed such that the first time point of analysis occurred after the early months of ART (when there is substantial patient-to-patient variation in the kinetics of suppression of viremia and of T cell redistribution) as well as after the virus load has been effectively suppressed (thereby avoiding the confounding effects of unsuppressed and variable viral loads that would have otherwise been a major driver of most if not all of the measured parameters). In the cohort studied, however, circulating classical or non-classical monocyte percentages were not predictive of CD4+ T cell recovery in early ART-suppressed patients (Figure 4A, B). However, the possibility of missing an association due to low statistical power could be related to the relatively small sample size and/or small effect size of CD4+ T cells (cells/uL) gain per patient.
A similar study looking at a panel of candidate T cell biomarkers within the same patient cohort described in Table 2 (Early ART suppressed) demonstrated that poor levels of CD4+ T cell recovery are predicted by high levels of CD8+ T cells with a senescent phenotype, i.e., increased cell surface expression of CD57 and/or decreased cell surface expression of CD27 and of CD28.
The altered monocyte populations observed during the context of HIV disease have further implications as they may constitute a viral reservoir. It appears that CD16-positive monocytes (5% of monocyte population) are both more susceptible to infection and preferentially harbor the virus long-termin vitro. We have also reported that immunoregulatory enzymes like HO-1 and indoleamine 2, 3-dioxygenase (IDO), may have beneficial effects in HIV-seropositive subjects. While HO-1 expression in CD14+ monocytes was not predictive of CD4+ T cell recovery when measured at time points early after ART-mediated viral suppression, (Figure 4C) suppressive ART did restore homeostatic levels of HMOX1 gene expression (Figure 3B). HO-1 is an important immune modulator, with effects that are anti-proliferative, anti-oxidant, anti-apoptotic, and cytoprotective. (Reviewed in) HO-1 deficiency results in macrophages that produce greater amounts of TNFα, IFNα, IL-6, and IL-2 after LPS stimulation in mouse models. Reciprocally, increased levels of HO-1 are associated with decreased levels of HIV replication in monocytes in vitro. Therefore, PBMC HO-1 levels may be reflective of the overall immune activation state, resulting in parallel decreases in both parameters over time on ART.