TL;DR
This study identifies three distinct immune profiles within people with HIV (PWH) on antiretroviral therapy (ART), driven by NF-κB signaling and characterized by varying levels of inflammatory cytokines like IL-1β and TNF-α. These profiles are linked to immune recovery markers like the CD4:CD8 ratio, suggesting they represent different inflammatory states that could inform personalized treatment strategies.
What Problem Does This Paper Address?
Despite effective antiretroviral therapy (ART), people with HIV (PWH) on ART continue to experience higher rates of morbidity and mortality compared to HIV-negative individuals. This is often attributed to persistent chronic inflammation and immune dysfunction that doesn't fully resolve even with treatment. The underlying mechanisms driving this persistent inflammation and its impact on immune recovery (e.g., CD4:CD8 T cell ratio) are not fully understood. This paper aims to identify distinct immune phenotypes within PWH on ART and understand how these phenotypes relate to immune markers and inflammatory pathways, potentially uncovering new targets for managing HIV-associated immune dysfunction.
Methods
The researchers employed a multi-omics approach, combining bulk RNA sequencing (RNA-seq) of CD4+ T cells with plasma immune marker quantification. A key methodological innovation was the use of Pairwise Controlled Manifold Approximation (PaCMAP), a novel dimension reduction technique, to identify distinct subgroups (clusters) of PWH based on their transcriptomic profiles. This was complemented by differential gene expression analysis (using DESeq2) to identify genes driving these clusters and pathway enrichment analysis (using MSigDB and GO terms) to understand the biological basis of the observed differences. Plasma cytokine levels were measured using standard assays.
Datasets
The study utilized a dataset from a cohort of 154 PWH on ART. This dataset was derived from previously published and publicly available data [11], combined with additional unpublished data on ART initiation timing and clinical measurements. The data included:
- Bulk RNA-seq of peripheral CD4+ T cells
- Plasma levels of seven immune markers (IL-10, IL-1β, Pentraxin, TNF-α, IP-10, G-CSF, and sTLR4)
- HIV reservoir measures (total HIV DNA, intact HIV DNA, and viral RNA expression)
- Clinical data including ART initiation timing, pre-ART viral load, and nadir CD4+ T cell counts.
Key Insights and Ideas
- Distinct Immune Subgroups Emerge: The application of PaCMAP revealed three distinct transcriptomic clusters within the PWH cohort, suggesting that PWH on ART are not a monolithic group immunologically.
- NF-κB as a Central Driver: These clusters were largely defined by differential expression of genes regulated by the NF-κB transcription factor, highlighting its critical role in shaping immune responses in PWH on ART.
- Inflammatory Cytokines Link Transcriptomics to Clinical Markers: Specific plasma cytokines, particularly IL-1β and TNF-α, were strongly associated with these transcriptomic clusters and, importantly, with the CD4:CD8 ratio, a key indicator of immune recovery and a predictor of mortality.
- CD4:CD8 Ratio as a Prognostic Marker: The study reinforces the significance of the CD4:CD8 ratio, showing its association with distinct transcriptomic profiles and, in one cluster, a lower ratio was linked to lower pre-ART viral load and nadir CD4+ T cell counts.
- HIV Reservoir Size Not Directly Linked to Transcriptomic Clusters: Interestingly, the clustering was not associated with the size of the total or intact HIV reservoir, suggesting that the observed transcriptomic differences are more related to ongoing immune responses rather than the latent viral burden itself.
Key Results
- Three Clusters Identified: PaCMAP analysis successfully identified three distinct transcriptomic clusters of PWH, comprising 94 individuals (Cluster 1), 51 individuals (Cluster 2), and 9 individuals (Cluster 3).
- NF-κB Pathway Dominance: Differential gene expression analysis revealed significant enrichment of NF-κB pathway-associated genes and inflammatory response genes in clusters 2 and 3 compared to cluster 1.
- Plasma Cytokine Associations:
- IL-1β levels were significantly higher in clusters 2 and 3 compared to cluster 1 (41 fg/mL vs. 323 fg/mL and 233 fg/mL, respectively).
- TNF-α levels were significantly higher in cluster 3 compared to clusters 1 and 2 (median 10869 fg/mL in cluster 3 vs. 186 fg/mL in cluster 1).
- G-CSF was modestly higher in cluster 1 compared to cluster 2.
- CD4:CD8 Ratio Association: Cluster 2 exhibited a significantly lower CD4:CD8 T cell ratio (median 0.78) compared to cluster 1 (median 1.04).
- ART Timing Impact: While early and late ART initiation groups were found across all clusters, early ART initiation showed a slightly higher representation in cluster 1, and a lower representation in cluster 2 compared to late ART.
Limitations and Future Work
- Cross-Sectional Design: The study is cross-sectional, limiting the ability to infer causality or track longitudinal dynamics of the observed phenomena.
- Small Cluster Size: Cluster 3 is very small (N=9), which limits the statistical power for comparisons involving this group, particularly for parameters like the CD4:CD8 ratio.
- Unmeasured Confounders: The study acknowledges that unmeasured confounders such as genetics, lifestyle factors (e.g., drug use), and secondary infections could contribute to the observed clustering patterns.
- Cohort Demographics: The cohort was predominantly male (96%), which may limit the generalizability of findings to female PWH.
- Bulk RNA-seq Limitations: The study used bulk RNA-seq of CD4 T cells. Future work could benefit from single-cell RNA-seq for higher resolution and investigation of immune cells and the HIV reservoir in tissues.
- Clinical Significance: The full clinical significance of these PWH clusters is currently unknown. Future research should focus on correlating cluster membership with actual clinical comorbidities (e.g., vascular events, neurocognitive decline) to establish predictive power.
- HIV Reservoir Dynamics: While no direct association was found between cluster membership and reservoir size, further analysis with larger cohorts could clarify the relationship between the intact reservoir and transcriptomic clusters.
- NF-κB and Viral RNA Paradox: The observation of lower HIV RNA levels in clusters with elevated NF-κB expression warrants further investigation, as NF-κB is typically a positive regulator of HIV expression. The authors speculate that elevated NF-κB activity might accelerate the negative selection of transcriptionally active proviruses.
Summary updated