High-fat diet, microbiome-gut-brain axis signaling, and anxiety-like behavior in male rats

Abstract

Obesity, associated with the intake of a high-fat diet (HFD), and anxiety are common among those living in modern urban societies. Recent studies suggest a role of microbiome-gut-brain axis signaling, including a role for brain serotonergic systems in the relationship between HFD and anxiety. Evidence suggests the gut microbiome and the serotonergic brain system together may play an important role in this response. Here we conducted a nine-week HFD protocol in male rats, followed by an analysis of the gut microbiome diversity and community composition, brainstem serotonergic gene expression (tph2, htr1a, and slc6a4), and anxiety-related defensive behavioral responses. We show that HFD intake decreased alpha diversity and altered the community composition of the gut microbiome in association with obesity, increased brainstem tph2, htr1a and slc6a4 mRNA expression, including in the caudal part of the dorsomedial dorsal raphe nucleus (cDRD), a subregion previously associated with stress- and anxiety-related behavioral responses, and, finally, increased anxiety-related defensive behavioral responses. The HFD increased the Firmicutes/Bacteroidetes ratio relative to control diet, as well as higher relative abundances of Blautia, and decreases in Prevotella. We found that tph2, htr1a and slc6a4 mRNA expression were increased in subregions of the dorsal raphe nucleus in the HFD, relative to control diet. Specific bacterial taxa were associated with increased serotonergic gene expression in the cDRD. Thus, we propose that HFD-induced obesity is associated with altered microbiome-gut-serotonergic brain axis signaling, leading to increased anxiety-related defensive behavioral responses in rats.

Keywords: tph2; Anxiety; Dorsal raphe nucleus; High-fat diet; Microbiome; Microbiome-gut-brain axis; Obesity; Raphe nuclei; Serotonergic system; Serotonin.

Fig. 1

Experimental timeline. Beginning at about 5 or 6 weeks of age (postnatal day (pnd) 35–42; post-weaning period; 100 g ± 10 g), adolescent male rats were housed in groups of 4 per cage. Animals were fed either a standard control diet (CD) or a 45% high-fat diet (HFD), for 9 weeks, until the young adult age (pnd 98–105). Fecal stool samples were collected from both CD and HFD animals, once a week during the nine weeks of the diet protocol, for analysis of the fecal microbiome. Rats were subjected to the elevated plus-maze (EPM) test on day 63 and the light/dark box (LDB) test on day 64 followed by the open-field (OF) test to assess anxiety-like behavior. Twenty-four hours following the behavioral tests, rats were euthanized, and their fat pad tissues (epididymal, inguinal, and retroperitoneal) were removed for determination of the adiposity index (AI). Further, brains were removed, and flash frozen in liquid isopentane, wrapped in aluminum foil and stored at − 80 ºC in preparation for in situ hybridization histochemistry for analysis of tph2, htr1a, and slc6a4 mRNA expression in the brainstem raphe nuclei. Abbreviations: AI, adiposity index; CD, control diet; EPM, elevated plus-maze; HFD, high-fat diet; LDB, light/dark box; OF, open-field

Fig. 2

Effects of high-fat diet (HFD) on alpha diversity, beta diversity, and community composition of the gut microbiome across mid-adolescence, late adolescence, and adulthood. Alpha diversity as measured by (a) Shannon’s diversity index, (b) Observed OTUs, (c) Faith’s phylogenetic diversity. (d) Beta diversity distance comparison plot with box plots illustrating distances within and between a single treatment group using Bray-Curtis distance, (e) Bray-Curtis principal coordinates (PCoA) plot of dissimilarity matrix (phylum level), and (f) age-arranged Bray-Curtis PCoA plot of dissimilarity matrix (phylum level). Biplot using Bray-Curtis distance for (g) phylum and (h) genus, illustrating microbial community analysis and composition (vectors) using principal component analysis of center log ratio transformed and standardized data from CD and HFD groups. Vectors point in the direction of the greatest increase of values for corresponding phylum and genus. Top ten taxa with highest relative abundances, illustrated by stacked vertical bar charts for (i) phylum and (j) genus. Data are expressed as means ± SEMs, p < 0.05; Kruskal–Wallis H test and PERMANOVA pairwise test. Abbreviations: CD-A, control diet group/adulthood; CD-L, control diet group/late adolescence; CD-M, control diet group/middle adolescence; HFD-A, high-fat diet group/adulthood; HFD-L, high-fat diet group/late adolescence; HF D-M, high-fat diet group/middle adolescence; pnd, postnatal day

Fig. 3

Effects of CD or HFD, and stratified life-stages on gut microbiome community composition. Main effect of diet was analyzed using ANCOM-II (significant at FDR 0.05). Main effects of diet were observed for the following taxa. (a) Prevotella, (b) [f_Paraprevotellaceae];g_Prevotella, (c) Clostridiaceae;g_, (d) Lachnospiraceae;g_, (e) f_Lachnospiraceae;g_Dorea, (f) f_Ruminococcaceae;g_Oscillospira, (g) f_Veillonellaceae;g_, (h) f_Veillonellaceae;g_Anaerovibrio, (i) f_Helicobacteraceae;g_, (j) f_Mycoplasmataceae;g_Mycoplasma, and (k) Volcano plot. (l and m) Effect of control diet (CD) and high-fat diet (HFD) on the ratio between the relative abundances of Firmicutes and Bacteroidetes phyla (F/B ratio). (l) and (m) represent Pearson’s correlation coefficient of F/B ratio versus adiposity measures, and r and p values are shown on the panels according to the analysis of both CD and HFD groups x F/B ratio, or separately for CD x FB/ratio, and HFD x F/B ratio. Graphs illustrate correlations between the F/B ratio and (l) Adipose index and (m) total body weight gain. Blue circles represent CD (n = 12) group and red circles represent HFD (n = 12) group. Abbreviations: CD-A, control diet group/adulthood; CD-L, control diet group/late adolescence; CD-M, control diet group/middle adolescence; HFD-A, high-fat diet group/adulthood; HFD-L, high-fat diet group/late adolescence; HFD-M, high-fat diet group/middle adolescence

Fig. 4

Effects of nine weeks of control diet (CD) or high-fat diet (HFD) protocol on tph2 mRNA expression in subdivisions of the dorsal raphe nucleus (DRD, rDRD and cDRD). (a-i) Graphs illustrate (a) tph2, (c) htr1a, and (e) slc6a4 mRNA expression in the dorsal raphe nucleus, dorsal part (DRD), including the rostral (rDRD) and caudal (cDRD) aspects. Expressed as means ± SEMs of gene expression levels at specific rostrocaudal levels or within subregions. (b, e, h) Compiled rostrocaudal levels for DRD, rDRD, and cDRD of (b) tph2, (e) htr1a, and (h) slc6a4. (c, f, i) Photomicrographs for in situ hybridization histochemistry of (c) tph2, (f) htr1a, and (i) slc6a4. White circles represent CD, and black circles represent HFD group. *p < 0.05 versus CD at the same rostrocaudal level: versus CD. Rostrocaudal levels 9 = − 7.412 mm, 8 = − 7.496 mm, 7 = − 7.580 mm, 6 = − 7.664 mm, 5 = − 7.748 mm, 4 = − 7.832 mm, 3 = − 7.916 mm, 2 = − 8.00 mm, 1 = − 8.084 mm, 0 = − 8.168 mm, − 1 = − 8.252 mm, − 2 = − 8.336 mm, − 3 = − 8.420 mm, − 4 = − 8.504 mm, − 5 = − 8.588 mm, and − 6 = − 8.672 mm. Abbreviations: cDRD, caudal aspect of the dorsal raphe nucleus, dorsal part; DRD, dorsal raphe nucleus, dorsal part;; rDRD, rostral aspect of the dorsal raphe nucleus, dorsal part; CD-A, control diet group/adulthood; CD-L, control diet group/late adolescence; CD-M, control diet group/middle adolescence; HFD-A, high-fat diet group/adulthood; HFD-L, high-fat diet group/late adolescence; HF D-M, high-fat diet group/middle adolescence. Scale bar (c, f, i) 1 mm

Fig. 5

(a-b) Behavioral responses as measured by (a) time in the open and enclosed arms of the EPM, (b) number of entries in the open and enclosed arms of the EPM, (c) time in the light and dark part of the LD box, (d) number of entries in the light part of the LD box, (e) time in the center or outer area of the OF arena, (f) OF number of crossed squares. Data are expressed as means ± SEMs or interquartile ranges. g *p < 0.05. EPM, elevated plus-maze, LD, light-dark box; OF, open-field

Fig. 6

(a, b) Compiled significant Pearson’s correlation coefficient (blue and blue arrows, negative correlation; red and red arrows, positive correlation) among cDRD nuclei and relative abundance of different taxa for (a) grouped taxa r value, and (b) grouped taxa p value