Gene expression signatures and cardiometabolic outcomes following 8-week mango consumption in individuals with overweight/obesity

Abstract

Background: Little is known about the impact of mango consumption on metabolic pathways assessed by changes in gene expression.

Methods: In this single-arm clinical trial, cardiometabolic outcomes and gene expression levels in whole blood samples from 26 men and women were examined at baseline and after 8 weeks of mango consumption and differential gene expression changes were determined. Based on changes in gene expression profiles, partial least squares discriminant analysis followed by hierarchical clustering were used to classify participants into subgroups of response and differences in gene expression changes and in cardiometabolic clinical outcomes following the intervention were tested.

Results: Two subgroups of participants were separated based on the resemblance of gene expression profiles in response to the intervention and as responders (n = 8) and non-responders (n = 18). A total of 280 transcripts were significantly up-regulated and 603 transcripts down-regulated following the intervention in responders, as compared to non-responders. Several metabolic pathways, mainly related to oxygen and carbon dioxide transport as well as oxidative stress, were found to be significantly enriched with differentially expressed genes. In addition, significantly beneficial changes in hip and waist circumference, c-reactive protein, HOMA-IR and QUICKI indices were observed in responders vs. non-responders, following the intervention.

Conclusion: The impact of mango consumption on cardiometabolic health appears to largely rely on interindividual variability. The novel transcriptomic-based clustering analysis used herein can provide insights for future research focused on unveiling the origins of heterogeneous responses to dietary interventions.

Clinical trial registration: [clinicaltrials.gov], identifier [NCT03825276].

Keywords: Mangifera; cardiometabolic risk factors; mango; precision nutrition; transcriptomics.

Conflict of interest statement

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Copyright © 2022 Keathley, de Toro-Martín, Kearney, Garneau, Pilon, Couture, Marette, Vohl and Couillard.

Figures

FIGURE 1

Discriminant analysis and hierarchical clustering. (A) Projection of pre- and post-intervention visits (black and orange dots, respectively) of the ungrouped cohort onto the space, spanned by the two principal components derived from partial least-squares discriminant analysis (PLS-DA). The two first components of the model along with their corresponding explained variance are shown on x- and y-axis, respectively. Ellipses represent 95% confidence intervals around the centroid of each visit group. Blue dashed circles encompass participants into the newly sub-group of responders identified by hierarchical clustering analysis (HCA). (B) Dendrogram resulting from applying HCA (Ward’s method) on the two first components of PLS-DA. The Euclidean distance on the y-axis measures the dissimilarity between each pair of observations. Blue dashed squares encompass matched pre- (black dots) and post-mango consumption visits (orange dots) of participants from the so-called responder sub-group. Numbers represent approximately unbiased p-values (AU) for each cluster. (C) Results from multilevel sPLS-DA show the complete discrimination between pre- (black dots) and post- visits (orange dots) in the responder (R) group. In the non-responder group (NR), pre- (gray dots) and post- visits (blue dots) are mixed. The two first components of the sPLS-DA accounted for 55% and 16%, and were composed of 230 and 460 genes, respectively. (D) The heatmap shows group classification (26 matched-pairs in rows) based on the two main sPLS-DA components (genes in columns). Within the group of responders, pre- (black dots) and post-mango consumption visits (orange dots) are clearly separated on the top and bottom rows of the heatmap.

FIGURE 2

Top differentially expressed transcripts in the responder sub-group following the mango consumption. Individual gene expression change between pre- and post-mango consumption visits is shown for responder and non-responder groups. (A) The two transcripts showing the most significant over-expression in the group of responders (RNF19A and SIMC1) are shown on the two left columns. (B) The two transcripts showing the most significant under-expression (ABR and ARHGAP30) are shown on the two right columns. Box and whisker plots show median, first, and third quartiles, and maximum and minimum values for the 26 matched participants before (Pre) and after (Post) the mango consumption. Green and red lines stand for increasing or decreasing gene expression levels between pre- and post-mango consumption visits within individual paired samples.