The metabolic state of a cell is intrinsically linked to its immune status. Leukocytes utilize diverse metabolic pathways to regulate immune-specific gene expression at multiple levels (epigenetic, transcriptional, post-transcriptional, and post-translational). In T cells, glycolysis is crucial for effector function and cytokine production, while activation via AKT signaling enhances both glycolysis and oxidative phosphorylation (OXPHOS). Antigen-presenting cells (APCs) require glycolysis, glycogen metabolism, and fatty acid synthesis for immunostimulatory function. Conversely, regulatory T cell (Treg) formation needs fatty acid synthesis, and tolerogenic dendritic cells rely on fatty acid oxidation for suppression. This metabolic flexibility, often controlled by mTOR signaling, is essential for shaping cellular phenotypes and functions. Current technologies for analyzing immune cell metabolism are limited. High-dimensional single-cell analysis techniques like flow cytometry, mass cytometry, and single-cell RNA sequencing (scRNA-seq) excel at phenotypic analysis, but bulk methods are mainly used for metabolic respiration. These methods are often incompatible with analyzing heterogeneous cell populations at the protein level. Existing single-cell metabolic techniques offer limited multi-pathway analysis. This study addresses these limitations by presenting Met-Flow, a novel flow cytometry-based method designed to capture the metabolic state of immune cells at a single-cell level, offering a comprehensive assessment of metabolic pathways within a complex population.

The existing literature strongly supports the critical role of metabolic reprogramming in immune cell function. Studies have shown the importance of glycolysis in T cell activation and cytokine production (Chang et al., 2013), the involvement of AKT and STAT5 in the early metabolic response of naive CD4+ T cells (Jones et al., 2019), and the contribution of glycolysis, glycogen metabolism, and fatty acid synthesis to dendritic cell activation (Everts et al., 2014; Gotoh et al., 2018; Krawczyk et al., 2010; Thwe et al., 2017). The role of lipid metabolism in regulatory T cell differentiation and tolerogenic dendritic cell function has also been established (Berod et al., 2014; Sim et al., 2016; Malinarich et al., 2015). However, a lack of suitable technology has hindered a comprehensive understanding of metabolic heterogeneity within complex immune populations. While scRNA-seq can assess metabolic genes, the temporal disconnect between mRNA abundance and protein levels and the inability to capture post-translational modifications limit its usefulness. Techniques like single-modality analysis of metabolites such as NADPH or lactate offer limited insights into the multi-pathway metabolic state.

Met-Flow utilizes a high-parameter flow cytometry panel targeting 10 critical metabolic proteins representing rate-limiting enzymes and transporters across multiple metabolic pathways (glycolysis, pentose phosphate pathway (PPP), tricarboxylic acid (TCA) cycle, OXPHOS, fatty acid synthesis, fatty acid oxidation, phosphate uptake, glucose uptake, and antioxidant response). These proteins were selected based on their critical roles in their respective pathways. The panel includes antibodies against SLC20A1 (phosphate transporter), ASS1 (arginine metabolism), SLC2A1/GLUT1 (glucose transporter), IDH2 (TCA cycle enzyme), G6PD (oxidative PPP enzyme), ACAC/ACCI (fatty acid synthesis enzyme), PRDX2 (antioxidant enzyme), HK1 (glycolysis enzyme), CPT1A (fatty acid oxidation enzyme), and ATP5A (OXPHOS enzyme). Phenotypic markers were also included to identify 11 major leukocyte subsets. Antibody performance was optimized and validated using fluorescence-minus-one controls. PBMCs were isolated from healthy donors using Ficoll density gradient centrifugation and prepared for flow cytometry staining using standard protocols. Intracellular staining was performed to capture metabolic proteins after cell fixation and permeabilization. Data were acquired using a FACSymphony flow cytometer and analyzed with FlowJo software. FitSNE analysis was used for dimensionality reduction and visualization of metabolic profiles across cellular subsets. To analyze metabolic remodeling during T cell activation, purified T cells were stimulated with anti-CD3/CD28 beads in the presence or absence of 2-FDG (a glucose analog used for glycolytic inhibition). The effect of glycolytic inhibition on T cell activation, maturation, and cytokine production was measured by flow cytometry. Extracellular flux analysis was used to assess glycolytic function and mitochondrial respiration in bulk T cells, corroborating flow cytometry data. In some experiments, a phosphorylation-based approach targeting ribosomal protein S6 (pS6) was incorporated to monitor mTORC1 signaling. Finally, GM-CSF production was measured by incorporating a capture antibody into Met-Flow.

Met-Flow successfully distinguished major leukocyte subsets based solely on their metabolic protein profiles, demonstrating the unique metabolic signatures of different immune cell types. Significant metabolic heterogeneity was observed within and between leukocyte populations, with monocytes showing the highest expression of all metabolic proteins. T cell activation induced a significant increase in the expression of multiple metabolic proteins, highlighting extensive metabolic reprogramming. The glucose transporter GLUT1 was particularly strongly upregulated and showed a positive correlation with CD25 expression, a marker of T cell activation. Glycolytic inhibition using 2-FDG significantly attenuated T cell activation, altering the expression of metabolic proteins and cytokine production. This effect was not uniform across all activation markers and cytokines, demonstrating selective dependence on glycolysis. Analysis of T cell memory subsets revealed distinct metabolic profiles, particularly in central memory (TCM) and effector memory (TEM) cells. Glycolytic inhibition led to the selective expansion of a TCM subset with a unique metabolic profile characterized by high expression of glycolytic proteins, fatty acid synthesis enzymes, OXPHOS proteins, and antioxidant enzymes. Interestingly, this TCM population displayed high GM-CSF production even under glycolytic inhibition, suggesting metabolic independence. Extracellular flux analysis validated the metabolic reprogramming observed by Met-Flow, showing increased glycolytic function and mitochondrial respiration upon T cell activation. Inhibition of glycolysis reduced activation-induced increases in glycolytic parameters but did not significantly impact overall mitochondrial respiration, indicating reliance on alternative carbon sources. Incorporation of pS6 into Met-Flow revealed that the expanded TCM subset maintained high mTORC1 signaling even under glycolytic inhibition.

Met-Flow provides a powerful approach for studying cellular metabolism at the single-cell level, overcoming the limitations of existing bulk and single-cell methods. The findings highlight the significant metabolic heterogeneity within immune cell populations, the extensive metabolic reprogramming during T cell activation, and the differential dependence of various immune processes on glycolysis. The discovery of a glucose-independent, GM-CSF-producing TCM subset with a unique metabolic profile is particularly noteworthy, suggesting a potential mechanism for persistent inflammation. This subset's high expression of GLUT1, ACAC, PRDX2, ATP5A, ASS1, and HK1 points towards a distinct metabolic strategy that may be therapeutically targetable. The study's findings confirm previously reported metabolic changes associated with T cell activation and memory differentiation, providing new insights into the role of metabolic heterogeneity in immune responses. The results underscore the importance of single-cell analysis in understanding complex cellular processes and offer new avenues for investigating metabolic regulation in health and disease.

Met-Flow, a novel high-dimensional flow cytometry technique, enables simultaneous analysis of metabolic pathways, cellular lineages, and activation markers at the single-cell level. This technology revealed significant metabolic heterogeneity in immune populations, demonstrating the distinct metabolic signatures of different immune cell subsets. The study identified a unique, glycolysis-independent, GM-CSF-producing TCM subset, suggesting a new therapeutic target for inflammatory diseases. Future research could expand Met-Flow to incorporate additional metabolic pathways and explore its application in various disease contexts, particularly in immunotherapeutic settings.

The study was performed using PBMCs from healthy donors, limiting the generalizability of the findings to other populations (e.g., patients with immune disorders). The use of 2-FDG as a glycolytic inhibitor may have off-target effects, although these were considered in the discussion. While Met-Flow provides a measure of metabolic capacity, it doesn't directly measure metabolic flux. Further investigations employing techniques such as metabolomics would help to strengthen these findings.