Genetic Regulation of Cytokine Responses and Its Relevance to Multiple Sclerosis
Human immune responses differ substantially between individuals, even when immune cells are exposed to the same pathogen or inflammatory stimulus. Such variation can influence susceptibility to infection, the magnitude of inflammation and the development of immune-mediated diseases. In Integration of multi-omics data and deep phenotyping enables prediction of cytokine responses, Bakker and colleagues investigated the biological factors underlying this heterogeneity by integrating genetic, cellular, metabolic, hormonal and transcriptional measurements from healthy individuals. A particularly important implication of the study concerns multiple sclerosis, a chronic immune-mediated disorder of the central nervous system in which genetically influenced abnormalities in immune-cell activation contribute to inflammation and tissue injury. By connecting inherited disease risk with measurable cytokine responses, the study provides a functional framework for examining how predisposition to multiple sclerosis may be expressed before the onset of clinical disease.
Comprehensive Immune Phenotyping of Healthy Individuals
The main analysis was performed within the 500 Human Functional Genomics cohort, which included more than 500 healthy volunteers of predominantly Western-European ancestry. The investigators collected genome-wide single-nucleotide polymorphism data, immune-cell frequencies, circulating metabolites, immunoglobulin concentrations, inflammatory mediators, hormone levels, platelet-activation measurements, demographic variables and seasonal information. Immune responsiveness was assessed by stimulating peripheral blood mononuclear cells, whole blood and PBMC-derived macrophages with pathogens, microbial products and Toll-like-receptor ligands. Production of IL-1β, IL-6, TNF, IFN-γ, IL-17 and IL-22 was then quantified. These cytokines represent distinct functional programs: IL-1β, IL-6 and TNF largely reflect monocyte-associated inflammatory activity, whereas IFN-γ, IL-17 and IL-22 are more closely associated with lymphocyte-mediated immunity. This distinction is particularly relevant to multiple sclerosis because T-cell-derived cytokines, including IL-17 and IFN-γ, are central to inflammatory pathways implicated in central nervous system autoimmunity.
Baseline Immune State as an Interconnected System
The study demonstrated that baseline immunological parameters were strongly intercorrelated rather than biologically independent. Hormonal profiles, platelet characteristics, immune-cell frequencies, metabolites and inflammatory mediators formed structured clusters, indicating that immune competence emerges from coordinated physiological networks. Steroid hormones involved in related biosynthetic pathways were positively correlated, while leptin showed inverse associations with progesterone and testosterone. Platelet p-selectin expression also clustered with fibrinogen-related activation measurements, reflecting shared mechanisms of platelet activation. Considerable variation was observed in immune-cell composition, with effector T-cell populations showing some of the greatest interpersonal variability. This observation has particular relevance to multiple sclerosis, in which the abundance, activation state and differentiation of effector T cells can influence inflammatory cytokine production and migration into the central nervous system. The results therefore suggest that variation in autoimmune susceptibility may partly reflect pre-existing differences in cellular immune architecture.
Genetic Control of Cytokine Responses
Genetic variation emerged as the largest individual contributor to stimulus-induced cytokine production. Across the 91 cytokine–stimulus combinations examined, single-nucleotide polymorphisms explained an average adjusted R2 of approximately 0.18, and the genetic contribution was statistically significant for 59 responses. Immune-cell counts, circulating metabolites, season, hormones and platelet-related measurements contributed additional explanatory power, although their average effects were generally smaller. When these biological layers were integrated, the models explained as much as 67% of the interindividual variation in selected cytokine responses. This finding is highly relevant to multiple sclerosis because the disease is polygenic: numerous common variants each contribute modestly to susceptibility, while their combined effects influence immune regulation. The study shows that such inherited variation is not merely associated with disease probability but can also be reflected in quantifiable differences in cytokine production by immune cells.
Cytokine Regulation by Metabolic and Transcriptional Factors
Several circulating molecules showed stimulus- and cell-type-specific relationships with cytokine production. IL-18-binding protein was negatively associated with lymphocyte-derived IFN-γ, IL-17 and IL-22, suggesting that it may serve as an indicator of reduced lymphocyte activity. Acetate was inversely correlated with influenza-induced IL-1β, IL-6 and TNF production and suppressed selected macrophage responses in experimental validation assays. HDL cholesterol reduced inflammatory cytokine production following Aspergillus fumigatus stimulation, while glutamine was broadly associated with lower cytokine production. The transcriptional analyses were especially informative: gene-expression profiles obtained after Candida albicans stimulation explained substantially more cytokine variation than unstimulated expression profiles, with adjusted R2 values reaching approximately 0.75. For multiple sclerosis research, this finding underscores the importance of studying activated immune cells rather than relying exclusively on resting-state measurements, because disease-relevant regulatory abnormalities may become apparent only after immune stimulation.
Multiple Sclerosis Genetic Risk and Lymphocyte Cytokine Production
The strongest multiple-sclerosis-specific finding arose from the comparison between polygenic disease risk and cytokine responsiveness. Individuals with higher genetic risk scores for multiple sclerosis showed a preferential association with increased lymphocyte-derived cytokine production, particularly involving IL-17, IL-22 and IFN-γ. The distinction between lymphocyte- and monocyte-derived cytokines was highly significant for multiple sclerosis, with a reported P value of approximately 4.85×10 −11. This result indicates that common genetic variants associated with multiple sclerosis collectively influence T-cell-related immune functions in healthy individuals. The observation is biologically plausible because IL-17-producing T helper cells and IFN-γ-producing lymphocytes contribute to inflammatory signaling, leukocyte recruitment and disruption of immune tolerance in multiple sclerosis. Importantly, the study does not establish that increased cytokine production directly causes the disease. Instead, it demonstrates that inherited susceptibility is associated with a measurable functional immune phenotype, thereby helping to connect genome-wide association signals with specific immunological mechanisms.
Implications for Multiple Sclerosis Prediction and Precision Immunology
The authors used regularized predictive models to estimate cytokine responses from genetic and baseline biological data. SNP-based models produced correlations between predicted and measured cytokine levels ranging from approximately 0.28 to 0.89 in internal cross-validation, while the addition of cellular, metabolic, hormonal and immunological measurements produced modest overall improvement. External validation was less successful, demonstrating that these models remain exploratory rather than clinically applicable. Nevertheless, the study has substantial conceptual importance for multiple sclerosis. It suggests that polygenic susceptibility may be translated into functional biomarkers based on lymphocyte cytokine responses, potentially supporting future approaches to risk stratification, disease-subtype classification or treatment-response prediction. Major limitations include the predominantly European ancestry of the cohort, limited sample sizes for certain datasets, the use of ex vivo stimulation assays and the absence of longitudinal clinical outcomes. Future studies should therefore include larger and more diverse populations, repeated immune measurements and direct comparison between healthy individuals, patients with early multiple sclerosis and individuals at elevated genetic risk.
Disclaimer: This blog post is based on the provided research article and is intended for informational purposes only. It is not intended to provide medical advice. Please consult with a healthcare professional for any health concerns.
References:
Bakker, O.B., Aguirre-Gamboa, R., Sanna, S. et al. Integration of multi-omics data and deep phenotyping enables prediction of cytokine responses. Nat Immunol 19, 776–786 (2018). https://doi.org/10.1038/s41590-018-0121-3
