Cerebrospinal fluid cell-free mitochondrial DNA is associated with HIV replication, iron transport, and mild HIV-associated neurocognitive impairment

BACKGROUND: Mitochondria are abundant organelles critical for energy metabolism and brain function. Mitochondrial DNA (mtDNA), released during cellular injury and as part of the innate immune response to viral pathogens, contains CpG motifs that act as TLR-9 ligands. We investigated relationships between cerebrospinal fluid (CSF) cell-free mtDNA levels and HIV viral load (VL), biomarkers of inflammation and iron transport, and neurocognitive (NC) function in the CNS HIV Antiretroviral Therapy Effects Research (CHARTER) cohort. METHODS: We quantified cell-free mtDNA in CSF by droplet digital PCR in 332 CHARTER participants who underwent comprehensive neuropsychiatric evaluation. NC performance was assessed using the global deficit score (GDS) as either a continuous or a binary measure (GDS ≥ 0.5, impaired vs. GDS < 0.5, unimpaired). CSF, clinical, and biomarker data from the earliest available time point were analyzed. Cell-free mtDNA associations with CSF inflammation and iron-related biomarkers [CXCL10, IL-6, IL-8, TNF-a, transferrin (TF), ceruloplasmin (CP), and vascular endothelial growth factor (VEGF)], VL, and GDS were evaluated by multivariable regression. RESULTS: CSF cell-free mtDNA levels were significantly lower in participants with undetectable (vs. detectable) VL in either plasma (p < 0.001) or CSF (p < 0.001) and in those on antiretroviral therapy (ART; p < 0.001). Participants on ART with undetectable VL in both CSF and plasma had lower mtDNA levels than those with detectable VL in both compartments (p = 0.001). Higher mtDNA levels were observed in participants in the highest vs. lowest tertile (T3 vs. T1) of CSF CXCL10 (T3 vs. T1, p < 0.001) and TNF-a (T3 vs. T1, p < 0.05) in unadjusted analyses. MtDNA levels also correlated with CSF leukocyte count. After adjusting for CSF leukocyte count and VL, mtDNA levels were also associated with other inflammation- and iron-related biomarkers in CSF, including TF (T3 vs. T1, p < 0.05) and CP (T3 vs. T1, p < 0.05). With additional correction for ART use, mtDNA was also negatively associated with CSF VEGF (p < 0.05) and IL-6 (p = 0.05). We observed no associations of CSF mtDNA levels with age or GDS-defined NC impairment. CONCLUSIONS: CSF cell-free mtDNA levels were associated with HIV RNA and ART status, as well as with biomarkers of iron transport and VEGF, a growth factor with known effects on mitochondrial integrity and autophagy. CSF mtDNA may be a biomarker of iron dysregulation and/or neuroinflammation during HIV infection. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (doi:10.1186/s12974-017-0848-z) contains supplementary material, which is available to authorized users.