Abstract P2072: Integrative Network Analysis Of Microbiome-Immune Axis In Metabolic Syndrome Patients During A Fasting Intervention

• Published in: Hypertension (Dallas, Tex. 1979) Tex. 1979), 2019-09, Vol.74 (Suppl_1)
• Main Authors: Avery, Ellen G, Balogh, András, Bartolomaeus, Hendrik, Löber, Ulrike, Steckhan, Nico, Markó, Lajos, Wilck, Nicola, Hamad, Ibrahim, Šušnjar, Urša, Mähler, Anja, Hohmann, Christoph, Lesker, Till R, Strowig, Till, Dechend, Ralf, Bzdok, Danilo, Kleinewietfeld, Markus, Michalsen, Andreas, Müller, Dominik N, Forslund, Sofia K
• Format: Article
• Language: English
• ISSN: 0194-911X; 1524-4563
• DOI: 10.1161/hyp.74.suppl_1.P2072

Abstract only Fasting can prolong survival and reduce disease burden in rodent models, and possibly in humans. The relationship between diet, gut microbiota, immune system and host (patho)physiology has only recently been explored, and information is lacking on how periodic fasting affects the gut microbiome in patients with metabolic syndrome (MetS). We show a 5-day fast (FAST) in humans, followed by a modified DASH diet is more effective than DASH alone (DASH) at reducing systolic blood pressure (SBP change measured by ABPM, 95% CI; FAST: [-7.053,-1.142], DASH: [-5.880,1.477]), need for antihypertensive medication (FAST: n=15 of n=35, DASH: n=6 of n=36 patients), and body-mass index at three months post intervention. Fasting altered the gut microbiome, impacting bacterial taxa and functional gene modules associated with the production of short-chain fatty acids (e.g. Faecalibacterium prausnitzii , Eubacterium rectale, Coprococcus comes ), previously linked to vascular health and immunity. Immunophenotyping and cross-system analyses revealed that SBP changes correlated with circulating Il-2 + TNFα + mucosa-associated invariant T (MAIT) cells (FDR-corr P(q) =0.044, Spearman’s rho=0.44), Il-17 - IFNγ + MAITs (FDR-corr P(q) =0.022, Spearman’s rho=0.49), and effector CD4 + T cells (FDR-corr P(q)=0.047, Spearman’s rho=0.43). By stratifying the fasting group into BP responders and non-responders, we identified a set of 76 microbial and 99 immune responder-specific features. Machine learning algorithms could predict long-term SBP responsiveness from baseline immunome data, identifying changes in effector CD8 + T cells, Th17 cells and Tregs as discriminators (Single-subject prediction: 71%). This is the first high-resolution multi-omics characterization of fasting in MetS. Fasting induced long-term reduction in body weight and SBP, accompanied by changes in microbiome and immune homeostasis. Our data implicate fasting as a promising non-pharmacological intervention in MetS.