CD8+ T-cell differentiation during chronic viral infection is supported by metabolic reprogramming to meet distinct bioenergetic demands. Early effector CD8+ T-cell differentiation and function are supported by the PI3K-Akt-mTOR pathway, while the differentiation of late exhausted CD8+ T cells remains incompletely understood. We first characterized the metabolic heterogeneity of the progenitor, effector, and exhausted CD8+ T-cell subsets in chronic infection by utilizing the Compass algorithm, which provides metabolic state predictions based on single-cell RNA sequencing data and flux-based analysis. Our analysis revealed metabolic programs distinct to each subset of virus-specific CD8+ T cells. In addition, it is known that the differentiation of progenitor to effector CD8+ T cells depends on IL-21-producing CD4+ T cells. We found that PIM1 kinase, a known regulator of cellular energy metabolism that functions downstream of IL-21 signaling, displays high gene expression in the effector CD8 T-cell subset. Using the lymphocytic choriomeningitis virus clone 13 model of chronic viral infection, we showed that CD8+ T cell-specific deletion of PIM1 kinase impairs the differentiation and cytolytic function of late effector CD8+ T cells. Furthermore, deficiency in PIM1 kinase reduced oxidative and glycolytic metabolism, potentially contributing to the diminished effector differentiation and function. Overall, these data reveal not only the metabolic heterogeneity of exhausted CD8+ T cells, but also how metabolic regulation through the IL-21-PIM1 axis impacts CD8+ T-cell differentiation.

Supplementary Figure S3 shows the HAllA analysis illustrating associations between OTU expression and metabolite levels in unaffected lobes.

Supplementary Figure S11 shows the effects of stearic acid treatment on viability of cancerous (H2122, H358), non-cancerous (BEAS2B), and monocytic (THP1) cells.

Supplementary Figure S2 shows the agar layer setup used in the anchorage-independent colony formation assay.

Supplementary Figure S10 shows association plots between cytokines and metabolites in unaffected lobes.

L-2-Hydroxyglutarate (L-2-HG) is a low-abundance metabolite in mammals because the mitochondrial enzyme L-2-HG dehydrogenase (L2HGDH) oxidizes L-2-HG to 2-oxoglutarate (2-OG) to prevent its accumulation1. In humans, a lack of L2HGDH activity leads to L-2-HG accumulation and causes L-2-hydroxyglutaric aciduria2. Thus, L-2-HG is often classified as a toxic metabolite2-5. However, whether L-2-HG has any physiological function is unclear. Here we investigate whether L-2-HG qualifies as a physiological signalling metabolite by testing three criteria: regulated levels, defined molecular targets and a measurable physiological function. We report that an increase in mitochondrial NADH/NAD+ ratio drives malate dehydrogenase 2 (MDH2) to reduce 2-OG into L-2-HG. Moreover, L2HGDH oxidizes L-2-HG back to 2-OG in the mitochondrial matrix without requiring a functional electron transport chain. Through proteome integral solubility alteration assays, we show that the KDM4 family of H3K9 demethylases are L-2-HG-responsive targets. L-2-HG represses the nascent transcription of specific genes in mouse embryonic stem cells and increases H3K9me3 (a repressive histone mark) at these loci. In vivo, early embryonic L2HGDH overexpression in mice systemically reduces L-2-HG levels, impairs postnatal growth, causes mortality and produces selective functional and histological renal vulnerabilities. In postnatal kidneys, this reduction in L-2-HG causes H3K9me3 loss at L1MdTf retrotransposons and their derepression, which coincides with the activation of the integrated stress response and inflammation pathways. Our findings establish mitochondrial L-2-HG as a physiological signalling metabolite and indicate that metabolites previously regarded as toxic may also have crucial physiological functions.

Supplementary Figure S4 shows the correlations between microbial abundance and metabolite presence using Spearman’s rank correlation in affected and unaffected lobes.

Supplementary Table S3. Cytokines included in Isoplexis assays

Supplementary Figure S9 shows association plots between microbial taxa (OTUs) and cytokines in unaffected lobes.

Supplementary Figure S8 shows association plots between microbial taxa (OTUs) and metabolites in unaffected lobes.

Supplementary Figure S5 shows principal component analyses indicating separation between tumor and unaffected lobes, including elbow plots, PCA training, and loading distributions for PC1 and PC2.

The spleen contains diverse macrophage subsets that remove aged erythrocytes, prevent the dissemination of circulating pathogens, and shape the adaptive immune response1-3. The mouse spleen hosts red pulp macrophages (RPM), marginal zone macrophages (MZM), marginal zone metallophilic macrophages (MMM), and tingible body macrophages (TBM). However, their transcriptomic identity, ontogeny, and dynamics during aging are unknown. Furthermore, it is not known whether homologous populations of macrophages exist in the human spleen. We find that in mice, MZM and MMM are tissue-resident macrophages that maintain their population via local proliferation, while TBM are slowly replaced by circulating monocytes. Lineage tracing shows that MMM maintain the MZM pool, and that after MMM depletion, circulating monocytes restore MMM. We show that a decrease in MMM abundance in aging precedes changes in other cellular populations and splenic niches. In human spleen, we identify TBM and perifollicular zone macrophages (PFZM) as a single macrophage population homologous to MMM and MZM in mice. We show that in both mouse and human TBM become more abundant during aging. Our results suggest age-related changes in the splenic microenvironment drive changes in tissue-resident splenic macrophage populations with potential importance for the loss of immunologic function in older individuals.

Supplementary Figure S13 shows the percentage distribution of cytokine-producing cells in untreated and stearic acid–stimulated THP1-derived macrophages.

BACKGROUND:Despite the high morbidity and mortality of heart failure with preserved ejection fraction (HFpEF), treatment options remain limited. The HFpEF syndrome is associated with a high comorbidity burden, including high prevalence of obesity and hypertension. Although inflammation is implicated to play a key role in HFpEF pathophysiology, underlying causal mechanisms remain unclear. METHODS:Comparing patient samples and animal models, we defined the innate immune response during HFpEF in situ and through flow cytometry and single-cell RNA sequencing. After identifying transcriptional and cell signatures, we implemented a high-fat diet and hypertensive model of HFpEF and tested roles for myeloid and hematopoietic stem cells during HFpEF. Contributions of macrophage metabolism were also evaluated, including through mass spectrometry and carbon labeling. Primary macrophages were studied ex vivo to gain insight into complementary cell-intrinsic mechanisms. RESULTS:Here we report evidence that patients with cardiometabolic HFpEF exhibit elevated peripheral blood hematopoietic stem cells. This phenotype was conserved across species in a murine mode of high-fat diet and hypertension. Hematopoietic stem cell proliferation was coupled to striking remodeling of the peripheral hematopoietic stem cell niche and expression of the macrophage adhesion molecule Vcam1. This could be partially inhibited by sodium-glucose cotransporter-2 inhibitors and explained by elevated fatty acid metabolism in macrophage mitochondria, which in turn remodeled the Vcam1 promoter to enhance its expression. CONCLUSIONS:These findings identify a significant new stem cell signature of cardiometabolic HFpEF and support a role for myeloid maladaptive fatty acid metabolism in the promotion of systemic inflammation and cardiac diastolic dysfunction.

Supplementary Figure S6 shows association plots between microbial taxa (OTUs) and metabolites in tumor lobes.

Lung cancer is the leading cause of cancer-related deaths. The human microbiome plays an important role in regulating response to cancer therapeutics, outcomes, and biological processes. However, little is known about the interplay between the lung microbiome and other biological processes in cancer. In an exploratory pilot study, we collected bronchoalveolar lavage fluid and brushings from 20 patients with early-stage lung cancer and performed microbial sequencing, untargeted metabolomics, and cytokine analysis. In addition, we employed computational and machine learning approaches to identify integrated microbial-immunometabolic pathways. Finally, we performed preliminary mechanistic studies to confirm our findings. Previously, we published that upper airway microbiota were selectively enriched in tumor-affected lobes. In the present study, we demonstrate that enrichment of protumorigenic cytokines and specific fatty acids is associated with tumor-affected lobes. Finally, we find that long-chain fatty acid stimulation of macrophages leads to neoplastic transformation of lung epithelial cells. Therefore, the findings of this study identify a perturbed fatty acid-macrophage axis that is a potential biomarker of early-stage lung cancer and will lead to the development of novel therapeutic agents. PREVENTION RELEVANCE:This study identifies a lung microbiome-driven immunometabolic axis involving stearic acid and MIP1β in tumor-affected lobes of patients with early-stage lung cancer. These localized microbial and cytokine-metabolite signatures may serve as biomarkers for early detection and provide targets for preventive strategies in high-risk individuals undergoing lung cancer screening.

Supplementary Table S1. Forward and reverse primer sequence for CCL3/ MIP1α, CCL4/MIP1β, and β-Actin. Primer sequences were curated from PrimerBank

Supplementary Figure S12 shows qPCR analysis of MIP1α and MIP1β expression in cancerous and non-cancerous cell lines under vehicle and stearic acid treatment.

Supplementary Figure S7 shows association plots between cytokines and metabolites in tumor lobes.

V-domain immunoglobulin suppressor of T cell activation (VISTA) is an immune checkpoint protein that impairs antitumor T cell responses. While broadly expressed on myeloid cells and T cells, the specific contribution of T cell-intrinsic VISTA to antitumor immunity remains undefined. This study investigated the phenotypic and functional consequences of T cell-specific VISTA deletion in tumor-specific CD8+ T cells. Single-cell transcriptomic analysis, TCR repertoire profiling, and flow cytometry revealed that loss of T cell-intrinsic VISTA enhanced early priming and short-term expansion of CD8+ T cells, yet this initial advantage failed to confer durable tumor control. Persistent dysfunction in VISTA-deficient T cells was in part driven by trans-VISTA on myeloid cells, while CTLA-4 upregulation further constrained T cell responses. T cell-intrinsic VISTA deficiency cooperated with CTLA-4 blockade to improve T cell survival and broaden TCR repertoire diversity, resulting in more robust tumor regression than CTLA-4 inhibition alone. A transcriptional signature enriched in VISTA-deficient cytotoxic T cells correlated with favorable outcomes in cancer patients treated with existing immune checkpoint inhibitors. These findings collectively define T cell-intrinsic mechanisms by which VISTA enforces T cell dysfunction and underscore its potential as both a therapeutic target and a biomarker of resistance to current immunotherapies.