Differential effects of Mediterranean vs. Western diets on coronary atherosclerosis and peripheral artery transcriptomics

Abstract

Western diets and social subordination are associated with increased risk of cardiovascular disease. In this study, we investigated the impact of Western versus Mediterranean diets and social status on atherogenesis and arterial transcriptional profiles in a 30-month randomized study in middle-aged, cynomolgus monkeys (Macaca fascicularis). Atherosclerosis (intimal area) in the left anterior descending (LAD) coronary artery was higher in the Western diet group compared to the Mediterranean diet group (F = 5.25, p = 0.03). There was no effect of diet on intimal lesion size in the iliac and carotid arteries (p > 0.05). Diet altered the transcriptome in iliac arteries; at an FDR threshold of 0.05, seven transcripts were upregulated (WDR62, PKDCC, SLC29A2, MARS1, RAD21L1, MAMDC4, and ENSMFAG00000052859), and 13 transcripts were downregulated (PIK3R1, PABPC1, PAQR8, ZNF667, FGGY, EIF4B, ALDH3A2, ANP32A, KDM3B, XPO7, RPS20, TOMM20, and CHCHD7) in the Western compared to the Mediterranean diet cohort. These genes are associated with endothelial dysfunction, smooth muscle proliferation and migration, angiogenesis, and abnormal extracellular matrix (ECM) dynamics. In addition, two transcripts (ENSMFAG00000064154 [LncRNA] and ENSMFAG00000057515 [small nucleolar RNA U13]) were downregulated in subordinate monkeys relative to their dominant counterparts (FDR < 0.05). There was no effect of diet on the carotid artery transcriptome, but we did identify significant social status effects: Eleven transcripts were upregulated (KCNQ4, STIM1, TNKS1BP1, CSNK1D, INPPL1, PNPLA7, F10, RAD9A, KCNIP3, ENSMFAG00000059809 [LncRNA], and ENSMFAG00000053865 [secreted protein A0A7N9CS45]), and seven transcripts were downregulated (IRAK1BP1, KIAA0513, SMIM15, PSMD14, TOPORS, ARPC2, and ENSMFAG00000050714 [LncRNA]) in subordinate relative to dominant monkeys. These alterations were associated with dysregulated vascular tone and smooth muscle contractility, apoptosis, and abnormal ECM dynamics. These findings demonstrate differential effects of diet composition and social status depending on arterial sites. The effects of Western diet were observed primarily in the coronary and iliac arteries, whereas social status differences were observed primarily in the carotid arteries. Our results demonstrate that Western diets and social subordination have adverse, yet distinct and tissue-specific impacts on arterial atherogenesis and transcriptional profiles, highlighting the interplay between diet, social hierarchy, and vascular health.

Keywords: Mediterranean diet; Western diet; arterial transcriptome; atherosclerosis; carotid artery; iliac artery; nonhuman primates; social status.

Figure 1

Western diet promoted intimal lesion formation in the proximal left anterior descending coronary compared to the Mediterranean diet. Tissue sections were stained with Verhoeff-van Gieson’s stain, and cross-sectional area of the arterial intima was measured. Two-way ANOVA showed a significant main effect of diet (Western 0.04 ± 0.03 mm² vs. Mediterranean 0.02 ± 0.008 mm²; F_{(1, 36)} = 5.25, p = 0.03).

Figure 2

Transcript levels (TMM normalized) differentially expressed by diet in the iliac artery (FDR < 0.05).

Figure 3

Diet was associated with differential transcriptional profiles in iliac arteries. (A) Hierarchal clustering of normalized transcript levels of the top DETs (n = 174, FDR < 0.2) in response to diet. Calculated residual transcript level values were converted into z-scores before clustering. (B) Canonical pathways identified by Ingenuity Pathway Analysis from 174 DETs with FDRs < 0.2 in iliac arteries. Ratio indicates the number of target genes in the dataset divided by the total number of genes in the pathway. (C) Predicted activity of cardiovascular function- related pathways in response to the expression state of DETs (FDR < 0.2) using IPA. Arrows indicate directional relationships. Orange indicates predicted activation and blue indicates predicted inhibition of the biological pathway based on the expression levels of the shown genes in the Western cohort relative to the Mediterranean cohort. Red indicates upregulation and green indicates downregulation of the labeled gene in the Western cohort relative to the Mediterranean cohort.

Figure 4

Transcript levels (TMM normalized) differentially expressed by social status in the carotid artery (FDR < 0.05).

Figure 5

Social status is associated with a differential transcriptional profile in carotid arteries. (A) Hierarchal clustering of normalized transcript levels of the top DETs (n = 216, FDR < 0.2) in response to social status. Calculated residual transcript level values were converted into z-scores before clustering. (B) Canonical pathways identified by Ingenuity Pathway Analysis from 216 DETs with FDRs < 0.2 in carotid arteries. Ratio indicates the number of target genes in the dataset divided by the total number of genes in the pathway. (C) Predicted activity of cardiovascular function- related pathways in response to the expression state of DETs (FDR < 0.02) using IPA. Arrows indicate directional relationships. Orange indicates predicted activation and blue indicates predicted inhibition of the biological pathway based on the expression levels of the shown genes in the subordinate cohort relative to the dominant cohort. Red indicates upregulation and green indicates downregulation of the labeled gene in the subordinate cohort relative to the dominant cohort.

Figure 6

Association of normalized transcript levels of differentially expressed genes in iliac and carotid arteries with physical, behavioral, and physiological phenotypes. (A) Diet adjusted Spearman’s correlation between DETs (FDR < 0.05) in iliac arteries with collected phenotypes. (B) Relative rank adjusted Spearman’s correlation between DETs (FDR < 0.05) in carotid arteries with collected phenotypes. Spearman’s test was conducted to identify significant correlations. Squares with colored circles inside indicate significant correlations (FDR < 0.05) between the indicated gene and the respective phenotype. Blue indicates positive correlation and red indicates negative correlation. RelRank: continuous value of social status, EstrogenLevels: plasma estrogen levels in pg./mL measured at necropsy, PulseWaveVelocity: pulse wave velocity measured by tail high definition oscillometry, DayHFHROscilation: percentage of high frequency (HF) oscillations during the day at 1600–1800 h, DayLF. HFHROscilation: ratio of LF/HF oscillations during the day at 1600–1800 h, NightLF. HFHROscilation: ratio of LF/HF oscillations during the night at 0100–0300 h, NightLFHROscilation: percentage of low frequency (LF) oscillations during the night at 0100–0300 h, NightHFHROscilation: percentage of high frequency (HF) oscillations during the night at 0100–0300 h, MeanCoronaryLesionArea: mean lesion area of the three coronary arteries collected, LADLesionArea: lesion area of the left anterior descending coronary artery, MeanCarotidLesionArea: mean lesion area of the three left common carotid arteries collected, MenstrualCycleLength: average length of cycles during experimental phase, measured in days, PerTimeDepressed: percent of time spent in depressive behavior.