Effects of Atorvastatin and Pravastatin on Immune Activation and T-Cell Function in ART-suppressed HIV-1 Infected Patients

Abstract

This retrospective study was designed to assess statin effects on T-cell activation from HIV-infected individuals. PBMC from ART-suppressed HIV-infected individuals receiving atorvastatin or pravastatin were evaluated for T-cell activation, exhaustion and function. Atorvastatin was associated with a significant reduction in CD8 T-cell activation (HLA-DR, CD38/HLA-DR) and exhaustion (TIM-3, TIM-3/PD-1) while pravastatin had no effect. In contrast, pravastatin increased antigen specific IFN-γ production. These results suggest a differential effect of statins on immune activation and function.

RESEARCH LETTER

Immune activation contributes to HIV pathogenesis [1-10]. Even with suppressive antiretroviral therapy (ART), immune activation persists and remains elevated compared to HIV-uninfected individuals [11-15]. While statins primarily serve as lipid-lowering therapy [16, 17], they also reduce T-cell surface signaling molecules and decrease T-cell activation [18-21]. Recent studies demonstrate that statins reduce T-cell activation and are associated with decreased morbidity and mortality among HIV-infected persons [22-24]. Specifically, atorvastatin significantly reduced T-cell activation in viremic HIV-infected individuals [25]. Here, we evaluated the effects of pravastatin and atorvastatin on T- cell activation and exhaustion in ART-suppressed individuals.

We used cryopreserved samples from 21 HIV-infected persons who were virologically suppressed (<50 copies/ml) with ART for >12 months (range, 20-137) and remained suppressed throughout follow-up [26]. Seventeen received adjunctive statin therapy for >6 months (range, 3-32): seven subjects received 10mg atorvastatin; 10 received pravastatin (5 with 20 mg; 5 with 40 mg). A median viability of 97% and recovery at 87% was observed for PBMC samples used in this study. As a control group, we identified four ART-suppressed individuals (median CD4 T-cell count: 498 c/mm³; median duration of ART: 119 weeks) not receiving statins. The study was approved by the IRB at UAB.

To assess statin effects on T-cell proliferation, we performed in-vitro assays as previously described [27] using cryopreserved PBMC from four statin-naïve individuals. We compared invitro effects of three statins on T-cell proliferation in response to SEB using cells from four ART-suppressed controls (no statin). Compared to pravastatin, both atorvastatin and rosuvastatin significantly suppressed SEB mediated CD4 and CD8 T-cell proliferation which was partially restored with exogenous mevalonate (Figure 1A-B).

Figure 1. A-B. In-vitro effects of statins on T-cell proliferation.

For in-vitro experiments, PBMC from 4 HIV infected individuals with undetectable viral load on ART were CFSE labelled and stimulated with SEB for 5 days in the presence of the indicated statins, with or without mevalonic acid (MA, +/-). Experiments were performed in triplicate and data are expressed as a % of SEB only stimulated proliferation in CD4 (A) and CD8 (B) T-cell subsets. Proliferation in the presence of atorvastatin and rosuvastatin (without mevalonic acid) was significantly different from the pravastatin group in a Wilcoxon Rank Sum test (**p<0.01). C-F. In-vivo effects of statins on markers of immune activation and exhaustion in T-cells. The in-vivo effect of atorvastatin and pravastatin on markers of T cell activation and immune exhaustion are shown in panels C-D. The expression of CD38 and HLA-DR (C) and TIM3 and PD1 (D) is shown as percentage change in marker expression relative to pre-statin i.e. baseline (mean+ SEM). The two statin groups were compared by the Mann-Whitney U test (*p≤0.05).The proportion of the four CD8 T cell subsets i.e. naive (CD27+CD45RO+), intermediate (CD27+CD45RO-), late (CD27-CD45RO+), and terminally (CD27-CD45RO-) differentiated subsets before (pre) and during (on) pravastatin and atorvastatin treatment are shown as pie charts (E and F respectively).

Ex-vivo analyses of expression levels of activation (CD38, HLA-DR), exhaustion (PD-1, TIM-3), and memory (CD27, CD45RO) markers were performed on cryopreserved PBMC from the 17 subjects receiving adjuvant statins before and during statin therapy and the 4 controls using surface staining with fluorochrome conjugated antibodies for specific cell surface markers [27]. Clinical and demographic features were not significantly different between groups (data not shown). Hyperlipidemia was the indication for statin use in all subjects; none were diabetic. In addition, baseline levels of immune activation and exhaustion were similar between the two groups. CD8 T-cell activation (HLA-DR and HLA-DR/CD38) and exhaustion (TIM-3) markers were significantly reduced following atorvastatin but not pravastatin treatment (Figure 1C-D). A significant reduction of CD4 T-cell exhaustion markers (TIM-3 and PD-1) was also observed with atorvastatin (p<0.05). We did not identify any sex-specific differences in activation or exhaustion marker expression. To evaluate whether ART alone explained these changes, we measured changes in expression levels on PBMC from four virologically suppressed and statin naive individuals. We identified no changes in activation (CD38, HLA-DR) or exhaustion markers (TIM-3, PD1), suggesting that ART therapy alone did not dictate our findings.

We measured CD27 and CD45RO expression to assess changes in memory T-cell subsets: naïve (CD27+CD45RO+), intermediate (CD27+CD45 RO−), late (CD27-CD45RO+) and terminally differentiated (CD27-CD45RO-) phenotypes [28]. A shift to an earlier stage of differentiation and a reduction in terminally differentiated subset was noted with atorvastatin but no change with pravastatin (Figure 1E-F). Atorvastatin related reduction of activation marker expression was restricted to terminally differentiated (CD27-CD45RO-) subset of CD8 T-cells (HLA-DR (44% decline, p=0.063) and CD38/HLA-DR (55% decline, p=0.031); data not shown. These differences were not explained by different proportions of terminally differentiated CD8 T-cells between the two groups at the pre or on statin time points (p=0.13, Figure 1 E-F). By contrast, CD4 T-cell activation (HLA-DR and CD38/HLA-DR) increased with pravastatin in the early differentiated subset (CD27+CD45RO+ ; p≤0.0312), but this change was not reflected in the total CD4 T-cell population (data not shown).

Since atorvastatin was able to reduce markers associated with chronic T cell activation and exhaustion, we next evaluated whether statin treatments improved T-cell functionality. When PBMC were antigenically stimulated with HIV (gag) or CMV (pp65) peptide pools, increased proliferation and IFN-γ production by CMV specific CD4 and CD8 T cell subsets was observed only for the pravastatin group (data not shown). No functional changes for HIV were observed in either statin group.

Our study demonstrates a differential effect of atorvastatin and pravastatin on T-cell activation, exhaustion and function. Atorvastatin significantly reduced markers of activation and exhaustion on T-cell populations; an effect most pronounced on terminally differentiated effector memory CD8 T-cells. Given that this T-cell subset contributes to the detrimental inflammatory state of chronic viral infections, these changes may be very important in the context of HIV infection, even in persons with suppressed viremia [29, 30]. We focused on virologically suppressed individuals given persistent immune activation. Similar to data from viremic subjects, we found reduced CD8+ T cell activation markers with atorvastatin [25]. This appears to be statin-specific, as pravastatin had no effect on these markers.

Atorvastatin, but not pravastatin, reduced markers of exhaustion (TIM-3 and TIM-3/PD1) on CD8 T-cells. PD-1 is upregulated on both CD4 and CD8 T-cells during chronic HIV infection even after ART administration and has been correlated with functional T-cell exhaustion in chronic viral infections [31]. Reducing PD-1 and TIM-3 expression may restore T-cell function [32, 33]. Whether the reduction of these inhibitory pathways by statins will yield improved immune responses needs further evaluation.

Limitations of our study include its small sample, non-randomized approach, and baseline CD4 count differences. Speculatively, since the pravastatin group had higher CD4 pre-and on-statin CD4 count, this group might exhibit lower levels of immune activation prior to statin initiation. However, baseline expression of markers was not different between groups (data not shown). In addition, a comparison of activation markers in patients (CD4 <500c/mm³) showed similar patterns as shown in figure 1C-D which suggests that a higher CD4 count in pravastatin was not responsible for a lack of detectable impact on the activation/exhaustion markers.

In conclusion, atorvastatin reduced markers of T cell activation and exhaustion in virologically suppressed HIV-infected individuals. Given the anti-inflammatory effects of statins and the recent ACC/AHA guidelines suggesting the benefits of statins extend beyond lipid-lowering, future research will assess which statin provides the greatest benefit for HIV-infected persons.