Raspberry consumption: identification of distinct immune-metabolic response profiles by whole blood transcriptome profiling

• DOI: 10.1016/j.jnutbio.2022.108946
• Publication Date: 2022-01-10T12:59:05Z
• Authors: Maximilien Franck; Juan de Toro-Martín; Thibault V. Varin; Véronique Garneau; Geneviève Pilon; Denis Roy; Patrick Couture; Charles Couillard; André Marette; Marie-Claude Vohl
• Keywords: Adult; Male; 0301 basic medicine; 2. Zero hunger; Gene Expression Profiling; Immunity; Lymphocyte Activation; 16. Peace & justice; Lipids; Diet; Gastrointestinal Microbiome; 3. Good health; Feces; 03 medical and health sciences; C-Reactive Protein; Biological Variation, Population; Fruit; Lipidomics; Metabolome; Cytokines; Humans; Female; Rubus; Transcriptome

Numerous studies have reported that diets rich in phenolic compounds are beneficial to immune-metabolic health, yet these effects are heterogeneous and the underlying mechanisms are poorly understood. To investigate the inter-individual variability of the immune-metabolic response to raspberry consumption, whole-blood RNAseq data from 24 participants receiving 280 g/d of raspberries for 8 weeks were used for the identification of responsiveness subgroups by using partial least squares-discriminant analysis (PLSDA) and hierarchical clustering. Transcriptomic-based clustering regrouped participants into two distinct subgroups of 13 and 11 participants, so-called responders and non-responders, respectively. Following raspberry consumption, a significant decrease in triglycerides, cholesterol and C-reactive protein levels were found in responders, as compared to non-responders. Two major gene expression components of 100 and 220 genes were identified by sparse PLSDA as those better discriminating responders from non-responders, and functional analysis identified pathways related to cytokine production, leukocyte activation and immune response as significantly enriched with most discriminant genes. As compared to non-responders, the plasma lipidomic profile of responders was characterized by a significant decrease in triglycerides and an increase in phosphatidylcholines following raspberry consumption. Prior to the intervention, a distinct metagenomic profile was identified by PLSDA between responsiveness subgroups, and the Firmicutes-to-Bacteroidota ratio was found significantly lower in responders, as compared to non-responders. Findings point to this transcriptomic-based clustering approach as a suitable tool to identify distinct responsiveness subgroups to raspberry consumption. This approach represents a promising framework to tackle the issue of inter-individual variability in the understanding of the impact of foods on immune-metabolic health.