Effects of follicle-stimulating hormone on energy balance and tissue metabolic health after loss of ovarian function

Abstract

Loss of ovarian function imparts increased susceptibility to obesity and metabolic disease. These effects are largely attributed to decreased estradiol (E₂), but the role of increased follicle-stimulating hormone (FSH) in modulating energy balance has not been fully investigated. Previous work that blocked FSH binding to its receptor in mice suggested this hormone may play a part in modulating body weight and energy expenditure after ovariectomy (OVX). We used an alternate approach to isolate the individual and combined contributions of FSH and E₂ in mediating energy imbalance and changes in tissue-level metabolic health. Female Wistar rats were ovariectomized and given the gonadotropin releasing hormone (GnRH) antagonist degarelix to suppress FSH production. E₂ and FSH were then added back individually and in combination for a period of 3 wk. Energy balance, body mass composition, and transcriptomic profiles of individual tissues were obtained. In contrast to previous studies, suppression and replacement of FSH in our paradigm had no effect on body weight, body composition, food intake, or energy expenditure. We did, however, observe organ-specific effects of FSH that produced unique transcriptomic signatures of FSH in retroperitoneal white adipose tissue. These included reductions in biological processes related to lipogenesis and carbohydrate transport. In addition, rats administered FSH had reduced liver triglyceride concentration (P < 0.001), which correlated with FSH-induced changes at the transcriptomic level. Although not appearing to modulate energy balance after loss of ovarian function in rats, FSH may still impart tissue-specific effects in the liver and white adipose tissue that might affect the metabolic health of those organs.NEW & NOTEWORTHY We find no effect of follicle-stimulating hormone (FSH) on energy balance using a novel model in which rats are ovariectomized, subjected to gonadotropin-releasing hormone antagonism, and systematically given back FSH by osmotic pump. However, tissue-specific effects of FSH on adipose tissue and liver were observed in this study. These include unique transcriptomic signatures induced by the hormone and a stark reduction in hepatic triglyceride accumulation.

Keywords: adipose tissue; energy balance; follicle-stimulating hormone; lipids; liver.

Graphical abstract

Figure 1.

Experimental design and groups. A: experimental scheme and timeline for the various phases of the study. Rats were placed on a 6-wk lead-in period on ad libitum (AL) high-fat diet, after which body mass composition is measured, rats were ovariectomized (OVX), given a bolus of the gonadotropin releasing hormone (GnRH) antagonist degarelix, implanted with pumps with either vehicle or follicle-stimulating hormone (FSH), and provided water with or without estradiol (E₂). After a 2-day recovery period, rats were then maintained for 12 days at their pre-OVX weight during the calorically limited (CL) phase while metabolic measurements are performed. Finally, animals were allowed to eat ad libitum (AL) for the final week of the experiment. B: groups derived from experimental treatments. A comparative control consisting of animals that underwent ovariectomies (OVX-only) but no other treatments was used for qualitative comparisons. Four experimental groups were analyzed: 1) degarelix-only (DEG, ↓E↓F), 2) degarelix + E₂ (E2, ↑E↓F), 3) degarelix + FSH (FSH, ↓E↑F), and 4) degarelix + E₂ + FSH (E₂ + FSH, ↑E↑F).

Figure 2.

Follicle-stimulating hormone (FSH) addback does not affect body weight, body mass composition, food intake, or energy expenditure. A and B: plasma levels of FSH and luteinizing hormone (LH) in rats receiving treatments. Target concentrations were achieved for both hormones. C: change in total body mass [pre-ovariectomized (OVX) to final] for all animals. D: change in lean mass (pre-OVX to final) for all groups. E: change in body fat (pre-OVX to final) for animals receiving treatments. F: change in energy intake (pre-OVX to final) over the course of the experiment. G: change in total energy expenditure (pre-OVX to end of the calorically limited phase) for all groups. H: change in nonresting energy expenditure (pre-OVX to end of the calorically limited phase). The OVX-only group (n = 8) is a comparative control and is shown for qualitative purposes. ↓E↓F: n = 11; ↑E↓F: n = 11; ↓E↑F: n = 7; ↑E↑F: n = 7. P values for main effects and statistical interactions are displayed on each graph. Error bars indicate SEM. ↓E↓F, degarelix-only; ↑E↓F, degarelix + E₂; ↓E↑F, degarelix + FSH; ↑E↑F, degarelix + E₂ + FSH.

Figure 3.

Organ weights at euthanasia. Organ weights for liver (A), left gastrocnemius muscle (B), left retroperitoneal white adipose tissue (C), mesenteric white adipose tissue (D), and subcutaneous white adipose tissue (E). The ovariectomized (OVX)-only group (n = 8) is a comparative control and is shown for qualitative purposes. ↓E↓F: n = 11; ↑E↓F: n = 11; ↓E↑F: n = 7, ↑E↑F: n = 7. P values for main effects and statistical interactions are displayed on each graph. Statistics were calculated using two-way ANOVA, and P ≤ 0.05 was considered to be statistically significant. Error bars indicate SEM. ↓E↓F, degarelix-only; ↑E↓F, degarelix + E₂; ↓E↑F, degarelix + FSH; ↑E↑F, degarelix + E₂ + FSH.

Figure 4.

Follicle-stimulating hormone (FSH) addback may exert effects on circulating glucose, leptin, and cholesterol. A: blood glucose at the time of euthanasia. B–G: plasma insulin, leptin, triglycerides, cholesterol, nonesterified free fatty acids, and insulin-like growth factor-1 (IGF-1) at the time of euthanasia. The ovariectomized (OVX)-only group (n = 8) is a comparative control and is shown for qualitative purposes. ↓E↓F: n = 11, ↑E↓F: n = 11, ↓E↑F: n = 7, ↑E↑F: n = 7. P values for main effects and statistical interactions are displayed on each graph. Error bars indicate SEM. ↓E↓F, degarelix-only; ↑E↓F, degarelix + E₂; ↓E↑F, degarelix + FSH; ↑E↑F, degarelix + E₂ + FSH.

Figure 5.

Follicle-stimulating hormone (FSH) supplementation may impart a transcriptomic signature on retroperitoneal white adipose tissue. A: heatmap showing all differentially expressed genes in the FSH group compared with degarelix-only (DEG) animals. A total 334 genes are downregulated, whereas 296 are increased in FSH compared with DEG. The false discovery rate (FDR) threshold was set at 0.05. B and C: downregulated and upregulated gene ontology (GO) enrichment biological processes in FSH animals compared with the DEG group. The FDR threshold was set at 0.05. D: differentially expressed retroperitoneal white adipose genes in select gene ontology (GO) biological processes downregulated or upregulated in FSH vs. degarelix-only including genes involved with lipid biosynthesis (downregulated in FSH), carbohydrate transport (downregulated in FSH), and angiogenesis (upregulated in FSH). DEG (↓E↓F) n = 6; FSH (↓E↑F) n = 5. ↓E↓F, degarelix-only; ↑E↓F, degarelix + E₂; ↓E↑F, degarelix + FSH; ↑E↑F, degarelix + E₂ + FSH.

Figure 6.

Follicle-stimulating hormone (FSH) addback after loss of ovarian function may lower hepatic triglycerides and alter the liver transcriptome. A: liver triglycerides measured in experimental groups. Both estradiol (E2) and FSH (as well as the combination of the two) lower triglyceride (TG) compared with the degarelix-only control. Ovariectomized (OVX)-only is shown as a qualitative comparative control. B: hepatic cholesterol measurement. For hepatic triglycerides and cholesterol, OVX-only n = 5, ↓E↓F: n = 6; ↑E↓F: n = 5, ↓E↑F: n = 5, ↑E↑F: n = 6. Error bars indicate SEM. C: heatmap showing all differentially expressed genes in the FSH group compared with degarelix-only (DEG) animals. A total 45 genes are downregulated, whereas 117 are increased in FSH compared with DEG. A false discovery rate of 0.05 and minimum fold-change of 2 were used. D: gene ontology (GO) biological process enrichment analysis showing processes upregulated in FSH compared with DEG. No enrichment was observed for downregulated genes in FSH. E: differentially expressed retroperitoneal white adipose genes in select gene ontology (GO) biological processes upregulated FSH vs. degarelix-only including genes involved with lipid oxidation, fatty acid metabolic processes, and small molecule metabolic processes. For RNAseq, DEG (↓E↓F) n = 5; FSH (↓E↑F) n = 4. ↓E↓F, degarelix-only; ↑E↓F, degarelix + E₂; ↓E↑F, degarelix + FSH; ↑E↑F, degarelix + E₂ + FSH.