Distinct cellular immune properties in cerebrospinal fluid are associated with cognition in HIV-infected individuals initiating antiretroviral therapy

The identification of intact HIV proviral DNA from human cerebrospinal fluid

We evaluated the HIV-1 DNA reservoir in peripheral blood mononuclear cells (PBMC) and cerebrospinal fluid (CSF) in people with HIV (PWH) and associations to cognitive dysfunction. Using the intact proviral DNA assay (IPDA), an emerging technique to identify provirus that may be the source of viral rebound, we assessed HIV DNA in CSF and PBMC in PWH regardless of antiretroviral therapy (ART). CSF was used as a sampling surrogate for the central nervous system (CNS) as opposed to tissue. IDPA results (3' defective, 5' defective, and intact HIV DNA) were analyzed by compartment (Wilcoxon signed rank; matched and unmatched pairs). Cognitive performance, measured via a battery of nine neuropsychological (NP) tests, were analyzed for correlation to HIV DNA (Spearman's rho). 11 CSF and 8 PBMC samples from PWH were evaluated both unmatched and matched. Total CSF HIV DNA was detectable in all participants and was significantly higher than in matched PBMCs (p ​= ​0.0039). Intact CSF HIV DNA was detected in 7/11 participants and correlated closely with those in PBMCs but tended to be higher in CSF than in PBMC. CSF HIV DNA did not correlate with global NP performance, but higher values did correlate with worse executive function (p ​= ​0.0440). Intact HIV DNA is frequently present in the CSF of PWH regardless of ART. This further supports the presence of an HIV CNS reservoir and provides a method to study CNS reservoirs during HIV cure studies. Larger studies are needed to evaluate relationships with CNS clinical outcomes.

Cerebrospinal fluid CD14++CD16+ monocytes in HIV-1 subtype C compared with subtype B

CD14++CD16+ monocytes are susceptible to HIV-1 infection, and cross the blood-brain barrier. HIV-1 subtype C (HIV-1C) shows reduced Tat protein chemoattractant activity compared to HIV-1B, which might influence monocyte trafficking into the CNS. We hypothesized that the proportion of monocytes in CSF in HIV-1C is lower than HIV-1B group. We sought to assess differences in monocyte proportions in cerebrospinal fluid (CSF) and peripheral blood (PB) between people with HIV (PWH) and without HIV (PWoH), and by HIV-1B and -C subtypes. Immunophenotyping was performed by flow cytometry, monocytes were analyzed within CD45 + and CD64 + gated regions and classified in classical (CD14++CD16-), intermediate (CD14++CD16+), and non-classical (CD14lowCD16+). Among PWH, the median [IQR] CD4 nadir was 219 [32-531] cell/mm3; plasma HIV RNA (log10) was 1.60 [1.60-3.21], and 68% were on antiretroviral therapy (ART). Participants with HIV-1C and -B were comparable in terms of age, duration of infection, CD4 nadir, plasma HIV RNA, and ART. The proportion of CSF CD14++CD16+ monocytes was higher in participants with HIV-1C than those with HIV-1B [2.00(0.00-2.80) vs. 0.00(0.00-0.60) respectively, p = 0.03 after BH correction p = 0.10]. Despite viral suppression, the proportion of total monocytes in PB increased in PWH, due to the increase in CD14++CD16+ and CD14lowCD16+ monocytes. The HIV-1C Tat substitution (C30S31) did not interfere with the migration of CD14++CD16+ monocytes to the CNS. This is the first study to evaluate these monocytes in the CSF and PB and compare their proportions according to HIV subtype.

High-throughput sequencing unravels the cell heterogeneity of cerebrospinal fluid in the bacterial meningitis of children

Bacterial meningitis (BM) is a common life-threatening infection in children that occurs in the central nervous system (CNS). The cytologic examination of cerebrospinal fluid (CSF) is a key parameter in the diagnosis of BM, but the heterogeneity of cells in the CSF has not been elucidated, which limits the current understanding of BM neuroinflammation. In this study, CSF samples were collected from a number of BM patients who were in different stages of disease progression. Single-cell RNA-sequencing (scRNA-seq), with additional bulk transcriptome sequencing, was conducted to decipher the characteristics of CSF cells in BM progression. A total of 18 immune cell clusters in CSF were identified, including two neutrophils, two monocytes, one macrophage, four myeloid dendritic cells, five T cells, one natural killer cell, one B cell, one plasmacytoid dendritic cell, and one plasma cell subtype. Their population profiles and dynamics in the initial onset, remission, and recovery stages during BM progression were also characterized, which showed decreased proportions of myeloid cells and increased proportions of lymphoid cells with disease progression. One novel neutrophil subtype, FFAR2+TNFAIP6+ neutrophils, and one novel monocyte subtype, THBS1+IL1B+ monocytes, were discovered, and their quantity changes positively correlated with the intensity of the inflammatory response in the CSF during BM. In addition, the CSF of BM patients with unsatisfactory therapeutic responses presented with different cell heterogeneity compared to the CSF of BM patients with satisfactory therapeutic responses, and their CSF featured altered intercellular communications and increased proportions of type II myeloid dendritic cells and plasmacytoid dendritic cells. Moreover, the bulk transcriptome profiles of autologous CSF cells and peripheral blood leukocytes of BM patients showed that the immune cells in these two physiological compartments exhibited distinct immune responses under different onset conditions. In particular, the CSF cells showed a high expression of macrophage characteristic genes and a low expression of platelet characteristic genes compared with peripheral blood leukocytes. Our study conducted an in-depth exploration of the characteristics of CSF cells in BM progression, which provided novel insights into immune cell engagement in acute CNS infection.

Main lymphocyte subpopulations in cerebrospinal fluid and peripheral blood in HIV-1 subtypes C and B

HIV-1 subtype C (HIV-1C) shows reduced Tat protein chemoattractant activity compared with HIV-1B. The impact of HIV-1C Tat on the chemotaxis of the main lymphocyte subpopulations in the cerebrospinal fluid (CSF) and the peripheral blood (PB) is unclear. We hypothesized that there would be a lower frequency of specific lymphocyte subpopulations CD3⁺ or CD19⁺ in CSF in HIV-1C than in HIV-1B. The objectives were to detect the differences in the proportions of main lymphocyte subpopulations in CSF and PB, between people with HIV (PWH) and HIV-1-uninfected volunteers (PWoH) and in HIV-1B and HIV-1C. Lymphocyte immunophenotyping was studied in CSF and paired PB samples of PWH (n = 22) and PWoH (n = 14). Lymphocytes were analyzed within the CD45⁺ gated region. The proportions of CSF CD3⁺CD4⁺, CD3⁺CD8⁺, and CD3^-CD19⁺ lymphocytes in CSF were comparable in HIV-1B and C. There was an increase in the proportion of CD3⁺CD8⁺ cells and a decrease in CD3⁺CD4⁺ T cells (ps = 0.016) in the CSF samples of the PWH compared with the PWoH group. In the PWH group, both CD3⁺CD4⁺ and CD3⁺CD8⁺ lymphocytes were significantly higher in the CSF than in the PB (p = 0.047 and 0.005). The proportion of CD3⁺CD4⁺ was lower and that of CD3⁺CD8⁺ was higher in the CSF samples of the aviremic group than that of HIV-negative control (p = 0.0008 and < 0.0001, respectively). HIV-1C Tat substitution (C30S) did not interfere with the CNS migration of the main lymphocyte subpopulations. This is the first study to evaluate these lymphocytes in CSF and PB of HIV-1C compared with HIV-1B.

Longitudinal telomere length profile does not reflect HIV and childhood trauma impacts on cognitive function in South African women

HIV-associated neurocognitive disorders (HAND) present a challenge in South Africa where the burden of HIV infection is the highest. Identification of biological correlates of HAND is required to improve diagnosis and inform interventions. Telomeres maintain genomic integrity and their shortening is a marker of biological aging sensitive to environmental influences. This study examined relative telomere length (rTL) as a predictor of cognitive function in the context of HIV and childhood trauma (CT), a risk factor for HAND. Two hundred and eighty-six women completed a neurocognitive assessment battery and the Childhood Trauma Questionnaire-Short Form (CTQ). Quantitative polymerase chain reaction for amplification of telomeric repeats and the reference gene human beta-globin was used to calculate rTL. Neurocognitive and rTL assessments were repeated at 1 year in 110 participants. Cross-sectional and longitudinal data were assessed using linear and mixed models, respectively. Participants with HIV (n = 135 in cross-sectional and n = 62 in longitudinal study groups) reported more severe CT and had shorter baseline rTL compared to seronegative controls. Participants without HIV had a greater 1-year decline in rTL. Global cognitive and attention/working memory scores declined in participants with HIV. Our data indicate that baseline rTL in the context of CT and HIV did not predict decline in cognitive scores. HIV-associated pathophysiological processes driving cognitive decline may also engage mechanisms that protect against telomere shortening. The results highlight the importance of examining biological correlates in longitudinal studies.