FGF21 mediating the Sex-dependent Response to Dietary Macronutrients

Abstract

Sex is key variable influencing body composition and substrate utilization. At rest, females maintain greater adiposity than males and resist the mobilization of fat. Males maintain greater lean muscle mass and mobilize fat readily. Determining the mechanisms that direct these sex-dependent effects is important for both reproductive and metabolic health. Here, we highlight the fundamental importance of sex in shaping metabolic physiology and assess growing evidence that the hepatokine fibroblast growth factor-21 (FGF21) plays a mechanistic role to facilitate sex-dependent responses to a changing nutritional environment. First, we examine the importance of sex in modulating body composition and substrate utilization. We summarize new data that point toward sex-biased effects of pharmacologic FGF21 administration on these endpoints. When energy is not limited, metabolic responses to FGF21 mirror broader sex differences; FGF21-treated males conserve lean mass at the expense of increased lipid catabolism, whereas FGF21-treated females conserve fat mass at the expense of reduced lean mass. Next, we examine the importance of sex in modulating the endogenous secretion of FGF21 in response to changing macronutrient and energy availability. During the resting state when energy is not limited, macronutrient imbalance increases the secretion of FGF21 more so in males than females. When energy is limited, the effect of sex on both the secretion of FGF21 and its metabolic actions may be reversed. Altogether, we argue that a growing literature supports FGF21 as a plausible mechanism contributing to the sex-dependent mobilization vs preservation of lipid storage and highlight the need for further research.

Because of differing biologic demands, many aspects of systemic metabolism are controlled in a sex-dependent manner (1). Successful reproduction is more energetically costly for females compared to males, especially during gestation and lactation, and female fertility is closely tied to the maintenance of sufficient body energy storage (1, 2). Consequently, sex is key variable influencing body composition and substrate utilization in response to metabolic challenges. Females tend to maintain a greater adiposity than males, for example, and resist the mobilization of fat storage by reducing energy expenditure during dietary restriction (3). Males tend to maintain a greater lean muscle mass than females and mobilize fat storage more readily (1, 3-5). Physiologic mechanisms directing the sex-dependent control of substrate metabolism are not fully understood but may have implications for both reproductive and metabolic health.

Fibroblast growth factor 21 (FGF21) is an intercellular signaling molecule produced in metabolic organs and tissues and secreted into circulation by the liver in response to diverse nutritional and metabolic stressors (6-10). FGF21 signals via a receptor complex that includes the FGF-receptor 1 (Fgfr1), together with an obligate co-receptor β-Klotho (Klb), to alter body composition and substrate utilization (6, 7, 11). Individuals treated with FGF21 or its analogues, especially males, have reduced body fat and show significant improvements in hepatic steatosis (12), generating considerable interest in its clinical applications for the treatment of fatty liver and other metabolic diseases (13). Recent findings support that both the secretion of FGF21 and the metabolic response to its pharmacologic delivery are sex dependent (10, 14-22).

The purpose of this mini-review is to highlight the fundamental importance of sex as a biological variable shaping metabolic physiology, and to assess accumulating evidence that FGF21 mechanistically links sex-dependent substrate storage and utilization to a changing nutritional environment. First, we examine the importance of sex as a variable modulating substrate storage and utilization, with a focus on adipose, liver, and skeletal muscle. We summarize new data that point toward sex-biased effects of pharmacologic FGF21 administration on these endpoints. Next, we examine the importance of sex as a variable modulating the endogenous secretion of FGF21 in response to changing macronutrient and energy availability. For a more comprehensive discussion of the importance of sex as a variable in metabolic physiology, please see these excellent in-depth reviews (1, 4, 23-25). For a more comprehensive discussion of FGF21 physiology and pharmacology, please see these excellent in-depth reviews (6, 13, 26).

 

Sex and FGF21 Modulate Substrate Storage and Utilization

Adipose tissue and lipid homeostasis

Male and female mammals differ in the provisioning and utilization of lipids fuels. Women have greater adiposity than men, on average, and preferentially carry excess fat in subcutaneous depots in the gluteal-femoral region. Men are more likely to carry excess fat in visceral depots in the abdominal compartment (5, 25, 27). These different anatomical fat depots are thought to have specialized functions (25). Adipocytes derived from subcutaneous fat tissue exhibit less lipid turnover compared to visceral adipocytes, having both a lesser responsiveness to adrenergic stimulation (28) and a lower rate of uptake of free fatty acids (25, 29-31). Depot-related differences in lipid turnover result from differences in the expression of adrenergic receptors, and of intracellular mediators of lipolysis, lipogenesis, and fatty acid oxidation. Adipocytes derived from the subcutaneous depots have greater relative expression of G_i-coupled adrenergic receptors (α2AR) vs G_s-coupled adrenergic receptors (β2 and/or β3), compared to adipocytes derived from the visceral depots, facilitating a lesser lipolytic tone. Subcutaneous adipocytes also have a lower expression and activity of proteins involved in fatty acid uptake and esterification, including CD36, acetyl-coenzyme A synthase, and diglyceride acyltransferase (25, 27, 32).

Sex-dependent adipose tissue growth, distribution, and function is controlled by the actions of sex chromosomes and sex steroid hormones. Using the “4 core genotypes model” to discriminate the separate effects of sex hormones and sex chromosomes in mice, Reue, Arnold, and colleagues found gonadectomized XX mice had greater lipid deposition in the inguinal subcutaneous fat pads, whereas gonadectomized XY mice had greater lipid deposition in the visceral compartment (33). Estrogen fluctuation during the female reproductive cycle additionally promotes subcutaneous fat expansion. Reducing estrogen action in rodents by ovariectomy, or as occurs naturally in humans after menopause, is associated with a redistribution of body fat toward the visceral depots (23, 24, 34, 35). Estradiol signaling via estrogen receptor alpha (ERα) reduces lipolytic tone in subcutaneous adipocytes, by increasing α2AR expression, but has no effect on adrenergic receptors in visceral adipocytes (23, 36). Testosterone, by contrast, enhances lipolysis and the expression of βAR in male adipocytes and this is more evident in visceral depots (37, 38). Finally, estrogen and testosterone have opposing effects on adipogenesis. Estrogen promotes adipocyte hyperplasia, whereas testosterone is thought to inhibit or have no effect (39, 40). Thus, via the combined influences of sex hormones and sex chromosomes, females at rest benefit from maintaining larger and more stable subcutaneous fat reserves providing continued reproductive capacity in a stochastic nutritional environment. By contrast, males at rest tend to oxidize free fatty acids preferentially derived from more labile visceral depots (1, 23, 41-45).

FGF21 affects the adipose tissue both indirectly via its actions in the nervous system and directly via its expression and signaling in adipocytes. Pharmacologic treatment with recombinant FGF21 and its analogues causes body weight and body fat loss in male rodents, by crossing the blood-brain barrier (46) to increase activation of the sympathetic nervous system and adrenergic signaling in adipose tissue (47-49). This is recapitulated by direct administration of recombinant FGF21 to the brain (50) and requires the expression of Klb in glutamatergic neurons (51). Therefore, the actions of endocrine FGF21 favor adipose lipid catabolism. By contrast, local paracrine signaling by FGF21 in adipocytes is lipogenic. In 3T3-L1 adipocytes and in cultured primary adipocytes, FGF21 administration stimulated the peroxisome proliferator-activated receptor-γ (PPARγ) signaling to promote both adipogenesis and lipogenesis (52). And genetic loss-of-function studies revealed that physiologic FGF21 promotes healthy fat expansion, specifically of subcutaneous fat depots. Both obese Fgf21-null mice (52, 53) and obese adipose-specific Klb-null mice (53) had less subcutaneous adipose tissue compared to controls (53). Likewise in humans, FGF21 affects body fat distribution because both serum FGF21 (54) and a common genetic variant of the Fgf21 allele (55) predict waist-hip ratio. Whereas the indirect effects of FGF21 via the nervous system favor a more male-like resting phenotype characterized by lipid mobilization, the direct effects of FGF21 locally in adipocytes favor a more female-like resting phenotype characterized by lipid storage in subcutaneous adipocytes.

Although most of the preclinical literature has used male subjects only, including the previously referenced studies, recent work highlights the importance of sex as a variable modulating FGF21s effect on fat storage. We recently reported that systemic FGF21 treatment reduced body fat storage in DIO male mice, but not female littermates (14). Likewise Bazhan and colleagues reported that females Ay and Mc4r-null mice are relatively less sensitive to FGF21-induced lipid catabolism compared to male controls (18, 20). Accordingly, FGF21 altered the expression of adrenergic receptors in a sex-dependent manner. FGF21 treatment increased the relative expression of α2AR:β3AR in female inguinal white adipose tissue, compared to saline-treated controls, favoring decreased lipolytic tone. Conversely, in male littermates, FGF21 decreased the relative expression α2AR:β3AR, compared to saline-treated controls, favoring increased lipolytic tone (14). FGF21 similarly altered mRNA expression of lipogenic and lipolytic genes in a sex-dependent manner (18, 20). As a result, the mobilization of lipid stores by FGF21 was diminished in female mice compared to males.

Sex-dependent effects of FGF21 on adipose lipid metabolism could be due to differential FGF21 action in the nervous system, in the adipose tissue, or both. FGF21 may have greater effects on (or access to) Klb-expressing presympathetic neurons of males compared to females, favoring increased lipolysis. Alternately, FGF21 may have greater effects in adipose tissue of females compared to males, favoring increased adipogenesis and adipose lipogenesis. In our hands, the mRNA expression of Fgfr1 and Klb is similar in male vs female adipose tissue (14), hypothalamus, and brainstem (unpublished observations) under basal conditions. However, we cannot rule out spatiotemporal differences in gene expression, or differences in receptor protein expression and/or stability, or sex-dependent modulation of intracellular signaling cascades. Additional research will be needed to distinguish these possibilities, and to determine the importance of sex hormones and/or sex chromosomes to mediate sex-dependent effects of FGF21 on adipose lipid storage.

Liver and lipid homeostasis

Among all the somatic organs, the liver shows the highest degree of sexual dimorphism (43, 56, 57). The liver plays a dynamic role to uptake, store, and distribute lipids for use throughout the body as fuel, as organic building blocks, and/or as a substrate for steroid hormone synthesis. Genes encoding proteins important for fatty acid, steroid, and lipid-metabolic pathways are more highly expressed in females compared to males. Fluctuations in their expression are thought to coordinate substrate availability with changing need, across the reproductive cycle and during pregnancy and lactation (43, 57, 58). Accordingly, estrogen signaling via ERα plays a key role to direct hepatic lipid metabolism and to differentiate the activities of male and female livers (59-61).

Hepatic lipid accumulation can occur in both physiologic and pathophysiologic states. Transient hepatic steatosis may occur during fasting, when adipose tissue lipolysis leads to a release of free fatty acids that exceeds clearance by nonhepatic tissues (62). Chronic fatty liver disease occurs during pathophysiologic states that are characterized by unrestrained adipose lipolysis, such as diabetes or alcoholism (63, 64). Women are protected from fatty liver relative to men, in part because women tend to store excess energy in the more-stable subcutaneous adipose depots, whereas men tend to store excess energy in more labile visceral depots (65). Though women may develop fatty liver at any age, the risk increases with aging, presumably because of the decreasing influence of estrogen after menopause (65).

Pharmacologic treatment with FGF21 reduces liver triglycerides in both preclinical and clinical studies and holds significant promise for the treatment of fatty liver disease (13, 66). The reduction in hepatic steatosis is an indirect effect; FGF21 is equally effective to reduce liver triglycerides in male liver-specific Klb-null mice and controls (49). The underlying mechanism is independent of body weight loss (14) and is thought to involve increased secretion of adiponectin from white adipose tissue. Adiponectin then acts via its receptors in liver to stimulate fatty acid oxidation and reduce liver triglycerides (66-68) (but see (69)). Importantly, we recently reported that the benefit of FGF21 treatment for improving hepatic steatosis depends on sex. Pharmacologic FGF21 treatment decreased liver triglycerides by approximately 60% in obese male mice but had no effect in obese females. This was associated with sex-dependent effects of FGF21 on adipocytes including sex-dependent adrenergic receptor signaling, intracellular cAMP, and adiponectin secretion via Epac1. The sex-dependent hepatic response did not depend on activational effects of the ovarian hormones because FGF21 remained ineffective to resolve fatty liver in adult ovariectomized and 24-month-old aging females (14).

Skeletal muscle and protein homeostasis

Both sex hormones and sex chromosomes contribute to sex differences in skeletal muscle protein homeostasis and maintenance of lean muscle mass. Women have less muscle mass than men of the same age, even among individuals with the same total body weight (70, 71). This is largely because of the anabolic effects of testosterone. Testosterone increases muscle protein synthesis, resulting in increased muscle length and increased total mass (72, 73). Sex differences in lean muscle mass become more evident after puberty, because of surging testosterone in boys. Nonetheless, body composition differs even between very young boys and girls, supporting a role for sex chromosomes (70). In agreement with this, when Arnold, Reue and colleagues used a sex chromosome trisomy mouse to model human Kleinfelter syndrome, they found XY mice had significantly greater lean mass compared to XXY mice of both sexes (74).

Although skeletal muscle is the largest reservoir for amino acids in the body, accessing amino acids from skeletal muscle is costly because it requires catabolism of structural and functional proteins. Consequently, systemic protein homeostasis is maintained primarily by adjusting feeding behavior and macronutrient selection in response to proteostatic need (75). During the absorptive state, the rate of muscle protein synthesis varies acutely with the protein content of a recent meal (76). During the postabsorptive state, amino acids are used sparingly as fuel and carbohydrate and lipid stores are preferentially catabolized. Following a prolonged fast or starvation, muscle proteins can be accessed for use as organic building blocks and as a substrate for hepatic gluconeogenesis. This metabolic adaptation to starvation is largely accomplished by the actions of the hypothalamic–pituitary–adrenal (HPA) axis via glucocorticoid receptor signaling in skeletal muscle, leading to the loss of lean muscle mass (77, 78).

We recently reported that a short course of systemic FGF21 treatment reduced skeletal muscle mass and muscle fiber cross-sectional area in mice (79). FGF21 acts via the nervous system to activate the HPA axis (80-82), suggesting a potential mechanism. Indeed, we found FGF21 treatment increased hypothalamic Crh mRNA, plasma corticosterone, and adrenal weight, and increased expression of glucocorticoid receptor target genes known to reduce muscle protein synthesis and/or promote degradation. All these outcomes were more apparent among females compared to male littermates. In agreement with sex-dependent effects on muscle mass, the most enriched metabolic pathways in plasma collected from FGF21-treated females were related to amino acid metabolism, and the relative abundance of plasma proteinogenic amino acids was increased up to 3-fold. By contrast, and in agreement with sex-dependent effects on lipid metabolism discussed previously, the topmost enriched metabolic pathways in plasma collected from FGF21-treated males were related to lipid metabolism (79). In the resting state, therefore, responses to pharmacologic FGF21 mirror broader sex differences in substrate storage and utilization: males conserve lean mass at the expense of increased lipid catabolism, whereas females conserve fat mass at the expense of reduced lean mass (14, 79).

The effect of FGF21 on substrate storage and utilization is context dependent

New findings highlight the importance of the nutritional environment to modulate metabolic responses to FGF21. Solon-Biet, Simpson, and colleagues found pharmacologic FGF21 significantly decreased body weight and body fat mass, as expected, among ad libitum fed male mice eating a standard AIN93G-based diet. By contrast, among mice eating a very low-energy version of the same AIN93-based diet, FGF21 administration significantly reduced energy expenditure and fat mass was preserved at the expense of lean mass (83). Thus, the effect of FGF21 on body composition depends on energy status, with opposing effects in energy surfeit vs energy deficit in males (84).

Sex differences in resting-state lipid storage and utilization, discussed in detail previously, are thought to favor more effective protein sparing during starvation, in women vs men (85). Compared to males, females accumulate and defend lipid reserves in the resting state (including moderate caloric restriction (45)), which are preferentially mobilized during prolonged starvation. Starved female pigs and rats, for example, maintained a 2-fold higher ratio of lipid to protein loss compared to males and had greater survival (41, 85). Similarly, women are more likely to survive extreme famine compared to men (85, 86). Unfortunately, the Solon-Biet study discussed previously did not include both sexes, but it will be interesting to know from future research whether the effect of FGF21 on female body composition also depends on energy status. If so, we expect FGF21 favors lipid storage during energy surfeit and favors lipid catabolism during energy deficit, facilitating the greater survival of females during starvation.

Sex and the Nutritional Environment Modulate FGF21 Secretion

Sex, FGF21, and starvation

FGF21 was initially described as a “starvation hormone.” Plasma concentrations rise substantially during 24 to 48 hours of complete fasting in rodents (87, 88). As glycogen stores become depleted, substrate utilization shifts away from glucose and toward fatty acids and ketone bodies. FGF21 was thought to be as a key hormonal mechanism contributing to this metabolic shift (87, 89). In this model, free fatty acids released from adipose tissue triglycerides circulate to the liver and prompt FGF21 transcription by binding to and activating PPARα (88, 90). FGF21 is secreted into circulation by the liver (69) and acts, via its receptors in preautonomic neurons, to facilitate continued lipid catabolism (7, 48). Indeed hepatic Fgf21 mRNA expression is elevated after a prolonged fast, which is abrogated in PPARα-null mice (87, 91). Treatment with PPARα agonists also increases liver Fgf21 expression in mice (88, 92) and humans (93). The importance of FGF21 as a fasting hormone in people has been questioned, however, because circulating FGF21 is not reliably increased in human plasma until after a full week of fasting, perhaps because of species differences in basal metabolic rate (94, 95). Circulating FGF21 is thought to be derived almost entirely from liver (6, 69), but it is also made locally in skeletal muscle and adipose tissue, where it acts in a paracrine manner to promote protein catabolism (96) and lipogenesis (52), respectively. Accordingly, whereas Fgf21 mRNA is increased by fasting in liver (89) and muscle (96), adipose Fgf21 mRNA expression is suppressed during periods of food deprivation (52).

Sex is a key variable modulating the metabolic response to starvation, as discussed previously. The energy partitioning strategy of females favors lipid storage during the resting state, and thereby allows for greater reliance on lipid oxidation during periods of food scarcity compared to males (41, 85, 86). Regarding FGF21, starved female mice had higher serum FGF21 compared to males. The sex difference depended on an intact female reproductive system and was abrogated by ovariectomy. Evidence supports an important role for estrogen and ERα signaling because treatment with estradiol or the ERα agonist PPT increased liver and serum FGF21. These effects were blunted in liver-specific ERα-null mice (19). It is tempting to speculate that the greater FGF21 secretion by starved females contributes to their greater reliance on lipid oxidation, compared to males, during the starved state. Importantly, such a mechanism would require status-dependent control of lipid metabolism in females (as occurs in males (83)) because FGF21 has minimal effects on female lipid catabolism in the resting/fed state (14). Whether starvation induced FGF21 secretion is sex dependent in humans has not been tested.

Sex, FGF21, and macronutrient imbalance

Ketogenic diets

FGF21 secretion is induced in nutritional environments characterized by macronutrient imbalance, even in the face of energy surplus. High-fat, low-carbohydrate ketogenic diet (KD), for example, increase liver and plasma FGF21; accordingly, KD causes body weight and body fat loss and increases fatty-acid oxidation and ketogenesis downstream of liver PPARα in males (87, 97). PPARα-null mice maintained, ad libitum, on a standard rodent KD have markedly less Fgf21 expression, and lose less fat, compared to controls maintained on KD. Yet KD-induced FGF21 is not completely absent in the PPARα-null model (87), suggesting other transcription factors contribute to the rise in FGF21. As with fasting, the clinical translation of KD to promote FGF21 secretion is uncertain. Pharmacological activation of PPARα robustly induces FGF21 in people, but the effect of KD in activating PPARα and increasing circulating FGF21 in humans is inconsistent at best (93, 95, 98).

Although there are limited data on the topic, at least 2 recent studies find the metabolic response to KD is sex dependent. In a human study, male gender predicted greater weight loss among individuals eating KD for 45 days (99). Similarly, male mice eating KD for up to 6 weeks lost body weight and body fat but female mice did not. KD boosted FGF21 secretion in both sexes, but the magnitude and relative time course was sex dependent (100).

Sex, FGF21, and amino acid restriction/dilution

Key recent studies now further define plasma FGF21 as a specific signal of dietary protein and/or amino acid restriction, rather than starvation or ketogenesis per se. Both hepatic Fgf21 mRNA and circulating FGF21 protein were increased by 24 hours of starvation in rats, but 12 hours of refeeding with either pure fat or pure carbohydrate did not restore this. Rather, Fgf21 was further elevated (101). A later study (102) reported 48 hours of starvation doubled the plasma FGF21 in rats, and refeeding with a high-carbohydrate, low-protein diet exaggerated this response, but refeeding with a high-protein, low-carbohydrate diet reduced FGF21 to resting state. And Solon-Biet, Simpson and colleagues used a geometric framework to show FGF21 was elevated during low protein intake and maximally when low protein was coupled with high carbohydrate (103). Finally, dietary protein restriction appears to explain the FGF21 response to rodent KD. Rodent KDs have very little protein because rats and mice more efficiently use amino acids for gluconeogenesis compared to humans. Accordingly, the FGF21 response to rodent KD is abrogated by the addition of dietary protein (102, 104). Finally, dietary protein restriction increases circulating FGF21 in men, resolving the clinical-translational inconsistencies regarding KD and FGF21 discussed previously (102, 105). Mechanistically, the FGF21 response to protein restriction—or more accurately, protein “dilution” because diets are provided ad libitum (105)—occurs downstream of the canonical GCN2→ eIF2α → ATF4 intracellular amino acid response pathway. Yet it requires hepatic PPARα for its maintenance (102, 106). Notably, such an arrangement may provide a mechanism for the preferential use of lipid fuels so long as total energy, and especially fatty acids, are abundant.

Secretion of FGF21 in response to dietary protein dilution is sex dependent (16). We fed adult male and female mice purified, isocaloric diets that were matched for dietary fat content, but differed in the ratio of protein (18% vs 4% kcal):carbohydrate (60% vs 74% kcal) for 30 days. In agreement with the studies discussed, we found plasma FGF21 was increased by up about 8-fold in protein-restricted males. However, the same diet increased FGF21 by only about 3-fold in intact female littermates. The sex difference in FGF21 secretion depended on an intact female reproductive system because ovariectomy restored diet-induced FGF21 secretion to the level of intact male littermate controls. Accordingly, and also in agreement with the sex-dependent response to pharmacologic FGF21 (14), protein dilution caused body weight and body fat loss in and increased markers of increased oxidative metabolism in white adipose tissue in males but not females (16). Lamming and colleagues also later reported that both the FGF21 and metabolic response to low-protein diet depends on sex and on genetic background. Their findings confirmed that dietary protein dilution increases FGF21 more so in male vs female mice and that the effect of protein dilution to improve metabolic endpoints was more apparent in males; this depended on strain (107). Intriguingly, in our hands, protein dilution tended to increase fat mass in females (16). Our ongoing unpublished work establishes this is a robust and repeatable outcome, and it is consistent with the lipogenic effect of FGF21 acting locally in adipose tissue (52, 53). During the resting state when energy is not limited, then sex-dependent secretion and action of FGF21 provide a plausible mechanism contributing to the sex-dependent mobilization vs preservation of lipid storage.

Summary

Because of differing biologic demands, systemic lipid and protein homeostasis are controlled in a sex-dependent manner. At rest, females maintain greater adiposity than males and resist the mobilization of fat storage; males maintain greater lean muscle mass and mobilize fat storage to meet daily needs. Here, we reviewed growing evidence that both the physiologic response to FGF21 and its endogenous secretion depend on sex and nutritional status (Fig. 1). When energy is not limited, responses to pharmacologic FGF21 mirror broader sex differences in resting-state substrate storage and utilization. FGF21-treated males conserve lean mass at the expense of increased lipid catabolism, whereas FGF21-treated females conserve fat mass at the expense of reduced lean mass. Moreover, sex and sex steroid hormones modulate the endogenous secretion of FGF21 in response to changing macronutrient and energy availability. During the resting state when energy is not limited, macronutrient imbalance favors the sex-dependent, male-biased secretion and action of FGF21 to preserve fat storage in females compared to males. In the starved state, preliminary evidence suggests this relationship may be reversed. Altogether, we argue that this growing literature supports FGF21 as a plausible mechanism contributing to the sex-dependent control of substrate storage and utilization. Because altered FGF21 signaling is thought to provide a common mechanism downstream of a range of metabolic effectors, including PPARγ (52) and GLP-1 agonists (108), and mitochondrial stress (109), and given its direct clinical importance (13), attention to its sex-dependent physiology is expected to have broad implications.

Figure 1.

Sex-dependent effects of FGF21 on substrate storage and utilization. During the resting state, FGF21 promotes the loss of adipose mass in males more so than in females and promotes the loss of muscle mass in females more so than in males. Sex-dependent effects of FGF21 on body composition during energy deficit have not yet been determined. Figure was created with BioRender.com.

Methods

To obtain information for this review, we used the following search strategies. First, we searched PubMed and Google Scholar using the terms Sex AND Lipid; Sex AND Fat; Sex AND Protein; FGF21 AND Fat; FGF21 AND Protein; FGF21 and Muscle; and FGF21 AND Sex. We used the returned results as a springboard and followed citations in the bibliographies of these papers as needed for depth and clarity.

Abbreviations

ERα

estrogen receptor alpha

FGF21

fibroblast growth factor-21

HPA

hypothalamic–pituitary–adrenal

KD

ketogenic diet

PPAR

peroxisome proliferator-activated receptor

Funding

Research reported in this publication was supported by the NIDDK of the National Institutes of Health under award R01DK121035 to K.K.R. K.A.S.S. was supported by the NIGMS of the National Institutes of Health T32GM007377.

Disclosures

The authors have nothing to disclose.

Data Availability

Data sharing is not applicable to this article as no data sets were generated or analyzed.