Brain serotonin and serotonin transporter expression in male and female postnatal rat offspring in response to perturbed early life dietary exposures

Abstract

Introduction: Serotonin (5-HT) is critical for neurodevelopment and the serotonin transporter (SERT) modulates serotonin levels. Perturbed prenatal and postnatal dietary exposures affect the developing offspring predisposing to neurobehavioral disorders in the adult. We hypothesized that the postnatal brain 5-HT-SERT imbalance associated with gut dysbiosis forms the contributing gut-brain axis dependent mechanism responsible for such ultimate phenotypes.

Methods: Employing maternal diet restricted (IUGR, n=8) and high fat+high fructose (HFhf, n=6) dietary modifications, rodent brain serotonin was assessed temporally by ELISA and SERT by quantitative Western blot analysis. Simultaneously, colonic microbiome studies were performed.

Results: At early postnatal (P) day 2 no changes in the IUGR, but a ~24% reduction in serotonin (p = 0.00005) in the HFhf group occurred, particularly in the males (p = 0.000007) revealing a male versus female difference (p = 0.006). No such changes in SERT concentrations emerged. At late P21 the IUGR group reared on HFhf (IUGR/HFhf, (n = 4) diet revealed increased serotonin by ~53% in males (p = 0.0001) and 36% in females (p = 0.023). While only females demonstrated a ~40% decrease in serotonin (p = 0.010), the males only trended lower without a significant change within the HFhf group (p = 0.146). SERT on the other hand was no different in HFhf or IUGR/RC, with only the female IUGR/HFhf revealing a 28% decrease (p = 0.036). In colonic microbiome studies, serotonin-producing Bacteriodes increased with decreased Lactobacillus at P2, while the serotonin-producing Streptococcus species increased in IUGR/HFhf at P21. Sex-specific changes emerged in association with brain serotonin or SERT in the case of Alistipase, Anaeroplasma, Blautia, Doria, Lactococcus, Proteus, and Roseburia genera.

Discussion: We conclude that an imbalanced 5-HT-SERT axis during postnatal brain development is sex-specific and induced by maternal dietary modifications related to postnatal gut dysbiosis. We speculate that these early changes albeit transient may permanently alter critical neural maturational processes affecting circuitry formation, thereby perturbing the neuropsychiatric equipoise.

Keywords: caloric restriction; high fat diet; intrauterine growth restriction; microbiome; serotonin; serotonin transporter.

Figure 1

Scheme demonstrating the experimental design. (A) Early postnatal rats: Three groups were created. (1) Control (CON) mothers reared on regular chow diet, (2) High fat and high fructose (HFhf) mothers received HFhf diet with fructose in their drinking water from gestational and embryonic day 11 (E11) to P2, (3) The intrauterine growth restriction (IUGR) group mothers received reduced regular chow diet from gestational and embryonic day 11 to P2. (B) Late postnatal rats: Soon after birth, CON and HFhf groups were given the control diet or the HFhf diet, respectively, until postnatal day 21 (P21). On the other hand, the IUGR group was given either the regular control diet (RC) (IUGR/RC) or the HFhf diet (IUGR/HFhf) until postnatal day 21 (P21).

Figure 2

Brain serotonin and SERT protein concentrations in early postnatal day 2 (P2) rats. (A) Brain serotonin concentrations in combined males and females demonstrate a decrease by 24% in the high fat and high fructose (HFhf) diet group (n = 6) versus Control (n = 14), (One Way ANOVA F statistic, [df:2,22] = 15.84, p value = 0.00005, and Tukey’s post-hoc test *p = 0.00005). (B) Brain serotonin concentrations in separated males and females, demonstrate in male brain a decrease of 32% in HFhf (n = 3) vs. CON (n = 6), (Two Way ANOVA F statistic [df:2,22] = 15.84, p value = 0.0001, and Tukey’s post-hoc test *p = 0.000007), with no difference in females. There is no difference in females between HFhf and CON (Tukey’s post-hoc test p = 0.718). A significant difference was observed in serotonin concentrations between male and female HFhf groups, with females showing ~24% higher concentrations than males (Tukey’s post-hoc test **p = 0.006). (C) SERT expression when males and females were combined trended to a decrease by 28% in HFhf (n = 7) vs. CON (n = 20), (One Way ANOVA F statistic [df:2,40] = 2.351, p = 0.108). (D) When the sexes were separated, while again only a trend toward a decline of ~45% in SERT concentrations occurred in the male HFhf (n = 4) vs. the male CON (n = 10), no significant difference was found (Two Way ANOVA F statistic [df:2,37] = 2.38, p value = 0.106). No inter-group differences were seen in the females, and no male versus female differences were noted either (Two way ANOVA F statistic [df:1,37] = 0.08, p value = 0.771), nor group X sex interaction differences (Two way ANOVA F statistic [df:2,37] = 1.71, p value = 0.193). Further, there is no difference in brain serotonin and SERT concentrations in the IUGR group versus control whether the sexes were combined or separated. Data are shown as mean ± SEM.

Figure 3

Brain serotonin and SERT protein concentrations in late postnatal day 21 (P21) rats. (A) Brain serotonin concentration in combined males and females decreased by 30% in HFhf (n = 8), (One Way ANOVA F statistic [df:3,43] = 34.44, p value <0.001, and Tukey’s post-hoc test *p = 0.0005) but increased by 30% in IUGR/RC (regular chow diet) (n = 8). (Tukey’s post-hoc test **p = 0.002) and by 53% in IUGR/HFhf groups (n = 8) vs. CON (n = 23), (Tukey’s post-hoc test ***p < 0.0001). (B) When the sexes were separated, serotonin trended to decrease by 32% in male HFhf (n = 4), (Two Way ANOVA F statistics [df:3,32] = 29.48, p value <0.00001, and Tukey’s post-hoc test p = 0.146) while trending to increase by 34% in male IUGR/RC (n = 4; Tukey’s post-hoc test p = 0.095) and increasing by 51% in male IUGR/HFhf (n = 4) vs. CON (n = 8; Tukey’s post-hoc test *p = 0.0005). In the female, brain serotonin decreased by 40% in HFhf (n = 4; Tukey’s post-hoc test **p = 0.010), while increasing by 36% in IUGR/HFhf (n = 4; Tukey’s post-hoc test ***p = 0.02259) vs. the sex-matched CON (n = 8). (C) When the sexes are combined, no difference in SERT expression was seen in all groups. (D) Upon separation of males from females, while no differences were seen in the male groups, SERT concentration only decreased by 28% in the female (D) IUGR/HFhf (n = 4) vs. CON (n = 8; Two Way ANOVA F statistic [df:3,24] = 3.54, p value = 0.03, and Tukey’s post-hoc test *p = 0.036) in IUGR. No male versus female differences were seen even within the IUGR/HFhf group (Tukey’s post-hoc test p = 0.11) nor any differences in the group X sex interaction. Data are shown as mean ± SEM.

Figure 4

Association between gut bacterial genera and brain serotonin or SERT in late postnatal day 21 rats. (A) Heatmaps showing differential abundance of gut bacterial genera among the various dietary experimental groups (HFhf, IUGR/HFhf, and IUGR/RC versus CON) stratified by sex (male = M, female = F) (left panel) and their relation to brain serotonin or SERT (right panel). A color gradient demonstrates the significant shifts in the abundance of gut microbial genera, with red indicating enrichment and blue indicating depletion. MaAsLin2 was employed adjusting for the effects of diet and sex with default parameters and the CON dietary group served as a reference for associations with the other three different dietary intervention groups. (B, C) Scatter plots depicting associations between individual gut bacterial genera and brain Serotonin (B) and SERT (C) concentrations. This association was calculated using a MaAsLin2 model without adjusting the effects of diet and sex. Coefficients depict the strength and direction of the observed associations. FDR represents the corrected significance and “N” indicates the total sample size employed in the analyses.