Interferon opens up: HIV-induced inflammation reconfigures 3D chromatin conformation and affects where HIV integrates

Abstract

Plaza-Jennings et al. applied single-nucleus RNA-seq, sorted neuronal and microglia cells for HiC, and found that chronic HIV infection in the brain induces interferon stimulation in microglia, drives chromatin reconfiguration into a transcriptionally active environment, and changes HIV integration landscape.

The 3 billion base pairs of the human genome are efficiently packaged and organized as chromatin in the three-dimensional (3D) nuclear space. These structures are organized into topologically associated domains (TADs), bringing distant coregulated regions into close spatial proximity and physical contact. CCCTC-binding factor (CTCF) binds to ~55,000 sites in the human genome, mediates chromatin looping, and defines TAD boundaries. Within this organization, efficient regulation of gene expression is achieved by enhancer-promoter interactions bridged by transcription factors. In the nucleus, chromosomes are compartmentalized in active regions (A compartments, containing more actively transcribed genes and active epigenetic marks) and inactive regions (B compartments, containing more transcriptionally repressed genes and repressive epigenetic marks).¹ Chromatin interactions, such as contacts between two distant cis-regulatory elements, can be measured by chromatin conformation capture (3C)² and HiC.¹ By focusing on chromatin loops associated with active chromatin marks (such as H3K27ac HiChIP),³ active enhancer regions and their respective target genes can be identified. In this issue of Molecular Cell, Plaza-Jennings et al.⁴ examine whether HIV-infected cells, particularly microglia in the brain, have a different 3D chromatin landscape (Figure 1).

Figure 1. Chronic HIV infection induces IFN responses in the brain and shapes 3D chromatin conformation in microglia.

The majority of HIV⁺ cells in HIV encephalitic brain are microglia. Chronic HIV infection (without suppressive ART) and HIV-induced IFN responses shift part of the 3D genome into active chromatin compartments, elicit new chromatin-interacting loops, and increase interferon response gene expressions. Moreover, HIV integration sites in microglia are enriched in highly expressed genes in active and permissive chromatin regions (defined by HiC, H3K4me3 chromatin immunoprecipitation sequencing (ChIPseq), H3K27ac ChIPseq, and ATAC-seq).

Does HIV integration change host 3D chromatin interactions? RNA viruses such as Influenza virus⁵ and SARSCoV-2⁶ do not integrate into the human genome but may change host 3D chromatin interactions through indirect mechanisms. HIV, as a provirus integrated into the human genome, could potentially influence 3D chromatin interactions directly or indirectly. It brings new transcription-factor binding sites into the chromosome, which could change 3D chromatin interactions. However, unlike another retrovirus, human T cell leukemia virus type 1 (HTLV-1), which harbors a CTCF-binding site in the proviral genome that alters host chromatin looping,⁷ HIV does not have a CTCF binding site in its genome and therefore may not change 3D chromatin structure. Another possibility is that HIV-induced inflammation may elicit cellular responses that change 3D chromatin structure in cells affected by inflammation. To understand whether HIV encephalitis changes 3D chromatin structure, particularly in microglia in the brain, the authors used interferon regulator factor 8 (IRF8) staining to isolate microglia and neuronal nuclei (NeuN) protein staining to isolate neurons and evaluated 3D chromatin interactions using HiC. Type I IFN responses persist in chronic untreated HIV infection and HIV encephalitis shifted parts of the microglia genome into the active chromatin compartment, particularly at type I interferon (IFN) response genes, such as ISG15, IFI6, IFI27, IFI44, IFIH1, STAT1, and STAT2. These chromatin changes are specific to microglia, consistent with the fact that microglia are the major IFN-responding cells.

What is the major cell type in the brain that harbors HIV? Advancements in single-cell transcriptomic methods allow capture of HIV⁺ cells by mapping the transcriptome to the HIV RNA genome.⁸ The critical caveat is how to distinguish authentic versus false discovery of HIV⁺ cells because of index hopping in single-cell sequencing methods. The authors judiciously performed control experiments to examine whether their HIV mapping was specific in mouse-tissue mixture and established >5 HIV reads per cell as their HIV⁺ cell defining threshold. The authors examined the single-nucleus transcriptional profile of neuronal and non-neuronal cells in the frontal cortex (and white matter) of the brain from three uninfected individuals, three HIV⁺ individuals, and three HIV⁺ individuals with HIV encephalitis (HIVE). The authors found that microglia (133 out of 2,756 cells) and immune oligodendroglia (10 out of 29 cells) accounted for the majority of HIV RNA⁺ cells in the HIVE brain. Of note, these HIV⁺ individuals had chronic HIV infection and ongoing replication without antiretroviral therapy (ART).

Does the 3D chromatin environment affect the HIV integration site? Presumably, Cleavage And Polyadenylation Specific Factor 6 (CPSF6) directs HIV integration into transcriptionally active chromatin while lens-epithelium-derived growth factor p75 (LEDGF/p75) directs HIV integration within genes.⁹ However, transcriptionally active chromatin regions are distinct between CD4⁺ T cells in the peripheral blood and microglia in the brain. Furthermore, the authors found that LEDGF and CPSF6 expression levels are low in microglia. While the HIV integration landscape has been widely studied in peripheral blood CD4⁺ T cells, little is known about HIV integration sites in microglia in the brain. The authors mapped the HIV integration site landscape in the brain during chronic HIV infection (without suppressive ART) and found that HIV integration sites are enriched in highly expressed genes from active chromatin in microglia. Separately, the authors profiled chromatin activity on multiple axes, including active histone marks (H3K4me3 and H3K27ac) and chromatin accessibility (by assay for transposase accessible chromatin sequencing, ATAC-seq). Indeed, HIV integration sites were favored near open chromatin regions with active histone marks. Of note, the HIV integration site landscape in this study is different from that in the peripheral blood from virally suppressed individuals and the authors did not identify integration at highly enriched genes (such as BACH2 or MKL2) or centromere locations.¹⁰ This may be because of the difference in the viral infection state (chronic infection without treatment versus suppressive ART), anatomical location (brain as an anatomical and immune sanctuary site versus peripheral blood), or cell type (microglia versus CD4⁺ T cells).

Brain samples, and all archived tissues from people living with HIV, are invaluable resources for understanding HIV pathogenesis in vivo. However, they are extremely hard to obtain. Furthermore, archived tissues have suboptimal RNA quality for single-cell transcriptomic profiling. This study provides a standard method using archived frozen tissues for single-nucleus transcriptome profiling and enables cell-type identification and cellular pathway analysis, thereby advancing our understanding of HIV pathogenesis in the body beyond our limited view in peripheral blood studies. Importantly, this study demonstrates the feasibility of cell-type-specific nuclei isolation by flow cytometric sorting and enriches for specific cellular populations. The caveat is the transcriptome difference between single-nucleus RNA-seq and single-cell RNA-seq. In the context of HIV, given that spliced HIV RNA is rapidly exported from the nucleus to the cytoplasm by HIV-1 Rev protein and that Rev-defective HIV RNA is retained in the nucleus, whether these HIV⁺ cells contain intact versus defective HIV remains to be determined.

Overall, this study identifies HIV-induced IFN responses that change host chromatin conformation and HIV integration landscape in microglia in the brain. It presents a feasible and rigorous method to advance our knowledge of HIV persistence and HIV-induced inflammation from peripheral blood to tissues.