BackgroundApproximately 10–40% of people with human immunodeficiency virus (HIV) infection are unable to obtain successful improvements in immune function after antiretroviral therapy (ART). These patients are at greater risk of developing non-acquired immunodeficiency syndrome (AIDS)-related conditions, with the accompanying increased morbidity and mortality. Discovering predictive biomarkers can help to identify patients with a poor immune response earlier and provide new insights into the mechanisms of this condition.MethodsA total of 307 people with HIV were enrolled, including 110 immune non-responders (INRs) and 197 immune responders (IRs). Plasma samples were taken before ART, and quantities of plasma microRNAs (miRNAs) were determined using reverse transcriptase-quantitative polymerase chain reaction (RT-qPCR). Candidate biomarkers were established through four phases: discovery, training, validation, and blinded test. Binary logistic regression was used to analyze the combined predictive capacity of the identified miRNAs. The effect of one miRNA, miR-16-5p, on T cell function was assessed in vitro.ResultsExpression of five miRNAs (miR-580, miR-627, miR-138-5p, miR-16-5p, and miR-323-3p) was upregulated in the plasma of INRs compared with that in IRs. Expression of these miRNAs was negatively correlated with both CD4+ T cell counts and the increase in the proportion of CD4+ T cells after one year of ART. These five miRNAs were combined in a predictive model, which could effectively identify INRs or IRs. Furthermore, we found that miR-16-5p inhibits CD4+ T cell proliferation by regulating calcium flux.ConclusionWe established a five-miRNA panel in plasma that accurately predicts poor immune response after ART, which could inform strategies to reduce the incidence of this phenomenon and improve the clinical management of these patients.

The complex mechanism of immune-system damage in HIV infection is incompletely understood. HIV-infected “rapid progressors” (RPs) have severe damage to the immune system early in HIV infection, which provides a “magnified” opportunity to study the interaction between HIV and the immune system. In this study, forty-four early HIV-infected patients (documented HIV acquisition within the previous 6 months) were enrolled. By study the plasma of 23 RPs (CD4+ T-cell count < 350 cells/µl within 1 year of infection) and 21 “normal progressors” (NPs; CD4+ T-cell count > 500 cells/μl after 1 year of infection), eleven lipid metabolites were identified that could distinguish most of the RPs from NPs using an unsupervised clustering method. Among them, the long chain fatty acid eicosenoate significantly inhibited the proliferation and secretion of cytokines and induced TIM-3 expression in CD4+ and CD8+ T cells. Eicosenoate also increased levels of reactive oxygen species (ROS) and decreased oxygen consumption rate (OCR) and mitochondrial mass in T cells, indicating impairment in mitochondrial function. In addition, we found that eicosenoate induced p53 expression in T cells, and inhibition of p53 effectively decreased mitochondrial ROS in T cells. More importantly, treatment of T cells with the mitochondrial-targeting antioxidant mito-TEMPO restored eicosenoate-induced T-cell functional impairment. These data suggest that the lipid metabolite eicosenoate inhibits immune T-cell function by increasing mitochondrial ROS by inducing p53 transcription. Our results provide a new mechanism of metabolite regulation of effector T-cell function and provides a potential therapeutic target for restoring T-cell function during HIV infection.