Innate Lymphoid Cell Activation and Sustained Depletion in Blood and Tissue of Children Infected with HIV from Birth Despite Antiretroviral Therapy

N/A

Innate lymphoid cells (ILCs) are important for response to infection and for immune development in early life. HIV infection in adults depletes circulating ILCs, but the impact on children infected from birth remains unknown. We study vertically HIV-infected children from birth to adulthood and find severe and persistent depletion of all circulating ILCs that, unlike CD4+ T cells, are not restored by long-term antiretroviral therapy unless initiated at birth. Remaining ILCs upregulate genes associated with cellular activation and metabolic perturbation. Unlike HIV-infected adults, ILCs are also profoundly depleted in tonsils of vertically infected children. Transcriptional profiling of remaining ILCs reveals ongoing cell-type-specific activity despite antiretroviral therapy. Collectively, these data suggest an important and ongoing role for ILCs in lymphoid tissue of HIV-infected children from birth, where persistent depletion and sustained transcriptional activity are likely to have long-term immune consequences that merit further investigation.

Further information and requests for resources and reagents should be directed to and will be fulfilled by the Lead Contact ([email protected]). This study did not generate new unique reagents. The RNA-seq datasets supporting the current study have not been deposited in a public repository because the subjects from which they were generated are at-risk children. The processed expression matrices are available upon request from the lead contact. Access to the raw data will be considered on a case-by-case basis with supporting IRB approval on the behalf of the requestor. Peripheral blood (PB) samples from children were obtained from the Ithemabalabantu pediatric cohort in Durban, KwaZulu-Natal (KZN), South Africa (Muenchhoff et al., 2016) and from Stanger Hospital, Stanger, KwaZulu-Natal (KZN), South Africa (Roider et al., 2019). PB samples from newborn/infants were obtained from the Ucwaningo Lwabantwana (meaning learning from children) infant cohort from Edendale, Mahatma Gandhi Memorial, Stanger and Queen Nandi Memorial Hospitals in KZN (Adland et al., 2020). Tonsil tissue samples were obtained from pediatric patients undergoing routine tonsillectomy at Stanger Hospital, Stanger, KwaZulu-Natal (KZN), South Africa (Roider et al., 2019). Informed consent was obtained from all adult study participants; and for underage children and adolescents, informed consent was obtained from their guardians. All HIV infected individuals were infected via vertical transmission from maternal HIV infection. For non-adult participants, 4 age groups were defined: 1. Newborns, aged 3–45 h; 2. Infants, aged 2–60 months; Paediatrics, aged 5–18 years. Paediatric slow progressors (PSP) were defined by stable CD4 T cell percentage of total PBMCs > 20% and found to be clinically healthy, while untreated pediatric progressors (PP) were defined by CD4+ T cell percentage of total PBMCs < 20% or otherwise meeting requirements for treatment. The pediatric treated cohort have individuals on antiretroviral therapy (ART). All subjects are from black Sub-Saharan ethnicity. This study was approved by the respective institutional review boards and Biomedical Research Ethics Committee, University of KwaZulu-Natal (UKZN) in Durban, South Africa. Peripheral blood mononuclear cells (PBMCs) were isolated by Histopaque 1077 (Sigma-Aldrich) density gradient centrifugation. Tonsil tissue samples was minced and digested with Collagenase D (0.5 mg/ml; Roche) and DNase I (20 μg/ml; Sigma-Aldrich) for 30 min in 37°C shaking incubator. Digested tissue was passed through 70 μm cell strainers. Lymphocytes from the tonsil were isolated by Histopaque 1077 (Sigma-Aldrich) density gradient centrifugation. For FACS analysis, different antibody panels for phenotype and intracellular cytokine staining (ICS) were used. A complete list of antibodies used with identifier and source information can be found in the Key Resources Table. All samples were surface stained at room temperature for minimum 20 min and near-infrared live/dead cell viability staining kit (Invitrogen). For experiments involving ICS, the cells were stimulated with PMA (5 ng/ml; Sigma) plus Ionomycin (1 μg/ml; Sigma) in the presence of Golgiplug and Golgistop (BD Biosciences) for 4 hr in 37°C incubator. Cells were stained with fluorochrome-conjugated monoclonal antibodies and subsequently fixed, permeabilized, and stained by BD Cytofix/Cytoperm Kit (BD Biosciences). Blocking with 20% goat serum for 20 min was done prior to intracellular antibody staining. After staining cells were washed and fixed in 2% paraformaldehyde before acquisition on a 4 laser, 17 parameter BD FACSAria Fusion flow cytometer within 24 h of staining. Data were analyzed with FlowJo v.9.7.2 (TreeStar). For cell sorting experiments, cells were processed from cryopreserved, surface stained, kept on ice in PBS and sorted immediately after staining. All samples were surface stained at room temperature for a minimum 20 mins. Bulk populations were cell sorted to a purity 99% on the BD FACSAria Fusion flow cytometer. CD4+ T cells, ILC2s, ILCPs, NK CD16high and NK CD56high cells from PBMCs and NKp44+ ILC3s, NKp44– ILC3s, NK CD127- and 4 distinct CD4+ T cell subsets from TMCs (100 cell replicates) were FACS sorted directly into 50 μL of RLT Lysis Buffer (QIAGEN) supplemented with 1% v/v 2-mercaptoethanol. Briefly, 50 μL of mixed lysate from each sample was transferred to a skirted 96 well plate. Genetic material was pulled down and purified by mixing the lysate in each well with 2.2x volumes of Agencourt RNAClean XP SPRI beads (Beckman Coulter) and washing 3x with 75 μL of 80% ethanol. After drying, the SPRI beads were re-suspended in 4 μL of pre-reverse transcription (RT) mix, incubated for 3 min at 72°C, and placed on ice. Next, Smart-Seq2 Whole Transcriptome Amplification (WTA) was performed: 7 μL of RT mix was added to each well and RT was carried out; then, 14 μL of PCR mix was added to each well and PCR was performed. Thereafter a cDNA cleanup was performed using 0.6x and 0.8x volumes of Agencourt AMPure XP SPRI beads (Beckman Coulter) which was then quantified using a Qubit dsDNA HS Assay Kit (Life Technologies). Library size and quality were measured by Bioanalyzer using a High Sensitivity DNA Analysis Kit (Agilent Technologies). Sequencing libraries were prepared from WTA product using Nextera XT (Illumina). After library construction, a final AMPure XP SPRI clean-up (0.8 volumes) was conducted. Library concentration and size were measured with the KAPA Library Quantification kit (KAPA Biosystems) and a TapeStation (Agilent Technologies), respectively. Finally, samples were sequenced on a NextSeq500 (30 bp paired-end reads) to an average depth of 5 million reads. Reads were aligned to hg38 (Gencode v21) using RSEM and TopHat (Li and Dewey, 2011) and estimated counts and transcripts per million (TPM) matrices generated. Any samples with fewer than 5×105 or more than 6×106 aligned reads or fewer than 10,000 uniquely expressed genes were removed from subsequent analysis. Differential expression analysis was performed using DESeq2 (v1.18.1) (Love et al., 2014). Expected counts from biological replicates for each cell type and participant were averaged prior to differential expression in order to prevent participant specific genes from generating false positives and reduce spurious heterogeneity from small (100-cell) populations. Small populations may show skewed expression based on the cell composition within; thus this replicate averaging approach is particularly important given our limited access to pediatric tissue sources and low frequency of these immune populations in order to remove further bias from small population sorts. See Tables S1 and S6 for replicate numbers. Tonsil analyses for the ILC3 and NK cell subsets were restricted only to samples from female subjects given insufficient age matched male subjects. To calculate DEGs between cell subsets, we used the design ∼1 + HIV.Status + Gender + Cell.Type in blood and the design ∼1 + HIV.Status + Cell.Type in tonsil. To calculate DEGs between HIV infected children and uninfected controls, we separated the samples by cell subset and used the design ∼Gender + HIV.Status for blood subsets and the design ∼HIV for tonsil subsets. For the CD4+ T cell subsets in the tonsil, we used the design ∼Gender + HIV.Status as sample availability required us to use cells from both female and male participants. Gene set analysis was performed using Ingenuity Pathway Analysis (IPA; Winter 2019 Release, QIAGEN Inc.) and Gene Set Enrichment Analysis (GSEA) using the piano package in R (1.18.1). For IPA, DEGs whose FDR corrected q < 0.1 were used in the “Core” analysis with the log2FC and q values included in the analysis. To implement GSEA on our DESeq2 results, we used the log2FC of all genes whose FDR corrected q < 0.1 as t-value input into the runGSA function with setting the argument geneSetStat = “gsea.” We chose to use the KEGG and GO databases (downloaded from MSigDB v7.0) (Subramanian et al., 2005) for GSEA analysis as these databases are well annotated for metabolic and cellular activation gene sets that are cell-type agnostic. For the IPA enrichment on tonsil population comparisons, only 1-3 of the significantly enriched terms had non-N/A values for each population. Thus, we have omitted the z-scores from the manuscript. Graphs were plotted using Prism 8.4.3 (GraphPad Inc.) Differences between groups were analyzed using Mann Whitney U-test or Dunn’s multiple comparisons test (two-sided) with specific test used stated in the figure legends otherwise. Data are presented as the medians values with boxes representing IQR range and range by error bars, with a p value < 0.05 considered statistically significant. The values of n refers to the number of participants used in study. In the other parts, it refers to the number of dependent experiments.

• Dima Health