Research Paper

FGF21 Has a Sex-Specific Role in Calorie-Restriction-Induced Beiging of White Adipose Tissue in Mice

Authors

Mariah F. Calubag,^1,2,3 Ismail Ademi,^1,2 Chung-Yang Yeh,^1,2 Reji Babygirija,^1,2,3 Heidi H. Pak,^1,2,4 Alyssa M. Bhoopat,^1,2 Ildiko Kasza,⁵ Cara L. Green,^1,2 Michelle M. Sonsalla,^1,2,6 and Dudley W. Lamming^{1,2,3,4,5,6,*}

¹Department of Medicine, University of Wisconsin-Madison, Madison, WI, USA

²William S. Middleton Memorial Veterans Hospital, Madison, WI, USA

³Graduate Program in Cellular and Molecular Biology, University of Wisconsin-Madison, Madison, WI, USA

⁴Interdepartmental Graduate Program in Nutritional Sciences, University of Wisconsin-Madison, Madison, WI, USA

⁵University of Wisconsin Carbone Cancer Center, Madison, WI, USA

⁶Comparative Biomedical Sciences Graduate Program, University of Wisconsin-Madison, Madison, WI, USA

*Corresponding author: dlamming@medicine.wisc.edu

DOI:https://doi.org/10.59368/agingbio.20230002

Abstract |  PDF |  Editorial |  Supplementary 

Abstract

Calorie restriction (CR) promotes healthspan and extends the lifespan of diverse organisms, including mice, and there is intense interest in understanding the molecular mechanisms by which CR functions. Some studies have demonstrated that CR induces fibroblast growth factor 21 (FGF21), a hormone that regulates energy balance and that when overexpressed, promotes metabolic health and longevity in mice, but the role of FGF21 in the response to CR has not been fully investigated. We directly examined the role of FGF21 in the physiological and metabolic response to a CR diet by feeding Fgf21^-/- and wild-type control mice either an ad libitum diet or a 30% CR diet for 15 weeks. Here, we find that FGF21 is largely dispensable for CR-induced improvements in body composition and energy balance, but that the lack of Fgf21 blunts CR-induced changes in glucose regulation and insulin sensitivity in females. Surprisingly, despite not affecting CR-induced changes in energy expenditure, loss of Fgf21 significantly blunts CR-induced beiging of white adipose tissue (WAT) in male but not female mice. Our results shed new light on the molecular mechanisms involved in the beneficial effects of a CR diet, clarify that FGF21 is largely dispensable for the metabolic effects of a CR diet, and highlight a sex-dependent role for FGF21 in the molecular adaptation of WAT to CR.

Introduction

Calorie restriction (CR), defined as a dietary regimen in which calories are reduced without malnutrition, promotes healthspan and increases lifespan in diverse organisms ranging from yeast to mice and nonhuman primates^1–4. CR promotes metabolic health in mammals, reducing adiposity and improving blood sugar control^3,5–7. Despite almost a century of effort, the physiological and molecular mechanisms by which CR functions are still not totally understood, stymieing efforts to develop CR mimetics which could promote health in the rapidly aging global populace.

As CR works to slow aging in all tissues, it has been suggested that CR may work in part through endocrine factors. The roles of hormones, including growth hormone, insulin, insulin-like growth factor 1 (IGF-1), and adiponectin, have been explored by numerous groups, but evidence that one of these hormones is responsible for the beneficial effects of CR remains elusive^7–9. One endocrine factor that could play a role in CR that has not yet been fully explored is fibroblast growth factor 21 (FGF21).

FGF21 is a hormone produced by multiple tissues including the liver and white adipose tissue (WAT) in response to a diverse array of nutrient stresses, including fasting, protein restriction (PR), and restriction of specific dietary amino acids^10–13. FGF21 promotes insulin sensitivity and regulates energy balance by promoting food consumption, activating brown adipose tissue (BAT) and stimulating the beiging of inguinal white adipose tissue (iWAT). FGF21 mediates the adaptive starvation response to induce ketogenesis, gluconeogenesis, lipolysis, and lipid β-oxidation^14–17. In addition to mimicking many of the beneficial metabolic effects of a CR diet, overexpression of FGF21 significantly extends lifespan¹⁸. Intriguingly, several studies have found that FGF21 is induced by CR in mice and rats^19,20, though to different extents; we also have observed an effect of CR on blood levels of FGF21.

Here, we directly examine the role of FGF21 in the metabolic response to CR by examining the effects of CR in wild-type (WT) and Fgf21 ^-/- (KO) mice of both sexes. We find that FGF21 is largely dispensable for the metabolic benefits of a CR diet in both males and females. Both WT and KO mice placed on CR become lean, glucose tolerant, and insulin sensitive, with a minor effect of Fgf21 deletion on fasting blood glucose (FBG) levels in female mice. Similarly, we find that FGF21 is not required for the effects of CR on energy balance. Surprisingly, we find that loss of Fgf21 blunts the CR-induced reprogramming of WAT metabolism in male mice. We conclude that despite the similarity in FGF21-induced and CR-induced phenotypes, and our discovery of a role for FGF21 in the CR-induced reprogramming of WAT, FGF21 is largely dispensable for the effects of CR on metabolic health.

Materials and Methods

Animal care, housing, and diet

All procedures were performed in conformance with institutional guidelines and were approved by the Institutional Animal Care and Use Committee of the William S. Middleton Memorial Veterans Hospital. Male and female WT and KO mice were generated by crossing CMV (cytomegalovirus)-Cre mice²¹ from The Jackson Laboratory (006054) with mice expressing a floxed allele of Fgf21²² from The Jackson Laboratory (022361), and then crossed with C57BL/6J mice to remove CMV-Cre. All mice were acclimated to the animal research facility for at least 1 week before entering studies. All animals were housed in static microisolator cages in a specific pathogen-free mouse facility with a 12:12-hour light–dark cycle, maintained at approximately 22°C.

At approximately 9 weeks of age, all animals were singly housed and placed on 2018 Teklad Global 18% Protein Rodent Diet for 1 week before randomization. At 10 weeks of age, mice were randomized to either an ad libitum (AL) diet or a 30% CR diet, in which animals were fed once per day during the beginning of the light period, at approximately 7 a.m. A stepwise reduction in food intake starting at 20% was carried out for mice in the CR group in week 1 before maintenance of a 30% restriction. The caloric intake of the mice in the AL group was calculated weekly to determine the appropriate number of calories to feed the mice in the CR groups. This feeding schedule has been widely used and extends the lifespan of mice^5,7,23.

Metabolic phenotyping

Glucose, insulin, and pyruvate tolerance tests (PTTs) were performed by fasting all mice for 7 hours or overnight (∼21–22 hours) and then injecting either glucose (1 g/kg), insulin (0.5 U/kg), or pyruvate (2 g/kg) intraperitoneally^7,24. Glucose measurements were taken using a Bayer Contour blood glucose meter and test strips. Mouse body composition was determined using an EchoMRI Body Composition Analyzer. For assay of multiple metabolic parameters (O₂, CO₂, food consumption, and activity tracking), mice were acclimatized to housing in a Columbus Instruments Oxymax/CLAMS-HC metabolic chamber system for approximately 24 hours, and data from a continuous 24-hour period were then recorded and analyzed. AL-fed animals had AL access to diet; CR groups were fed once per day at the beginning of the light cycle.

Collection of tissues for molecular and histological analysis

For analysis of plasma FGF21 levels in Figure 1A, mice were sacrificed at 7 a.m. prior to the regularly scheduled morning feeding time of the CR-fed animals. For all other experiments, mice were euthanized in the fed state after approximately 15 weeks. Mice euthanized in the fed state had all food removed starting at 3 p.m. the day prior to sacrifice, fed at 7 a.m. the day of sacrifice, and then euthanized 3 hours later. Following blood collection via submandibular bleeding, mice were euthanized by cervical dislocation, and most tissues were rapidly collected, weighed, and then snap frozen in liquid nitrogen. A portion of iWAT was fixed in 10% formalin for 48 hours before being transferred to 70% ethanol, sectioned, and hematoxylin and eosin stained by the University of Wisconsin Carbone Cancer Center Experimental Pathology Laboratory. Images of the iWAT were taken using an EVOS microscope (Thermo Fisher Scientific Inc., Waltham, MA, USA) at a magnification of 40× as previously described^13,25. Scale bars were inserted automatically or manually by the investigator. For quantification of adipocyte size, six independent fields were obtained for each tissue from each mouse and quantified using ImageJ (NIH, Bethesda, MD, USA). For quantification of multilocular cells, a blinded participant counted the number of multilocular cells per image for each mouse using ImageJ (NIH, Bethesda, MD, USA).

Figure 1.

Quantitative real-time Polymerase Chain Reaction (PCR)

RNA was extracted from iWAT using TRIzol Reagent according to the manufacturer’s protocol (Thermo Fisher Scientific). The concentration and purity of RNA were determined by absorbance at 260/280 nm using Nanodrop (Thermo Fisher Scientific). 1 μg of RNA was used to generate cDNA (Superscript III; Invitrogen, Carlsbad, CA, USA). Oligo dT primers and primers for real-time PCR were obtained from Integrated DNA Technologies (Coralville, IA, USA); sequences are in Table S1. Reactions were run on a StepOne Plus machine (Applied Biosystems, Foster City, CA, USA) with SYBR Green PCR Master Mix (Invitrogen). Actin was used to normalize the results from gene-specific reactions.

Enzyme-linked immunosorbent assay (ELISA) assays and kits

Blood plasma for insulin was taken the morning of euthanasia from fasted mice prior to refeeding as well as from the terminal submandibular bleeding. Blood plasma for FGF21 was obtained from the terminal submandibular bleeding. Plasma insulin was quantified using an ultrasensitive mouse insulin ELISA kit (90080), from Crystal Chem (Elk Grove Village, IL, USA). Blood FGF21 levels were assayed by a mouse/rat FGF21 quantikine ELISA kit (MF2100) from R&D Systems (Minneapolis, MN, USA).

Statistics

Data are presented as the mean ± standard error of the mean (SEM) unless otherwise specified. Statistical analyses were performed using one-way or two-way analysis of variance (ANOVA) followed by Sidak or Tukey–Kramer post hoc tests, as specified in the figure legends. Other statistical details are available in the figure legends. Energy expenditure differences were detected using analysis of covariance (ANCOVA). ANCOVA assumes a linear relationship between the variables and their covariates. If the slope is equal between groups, then the regression lines are parallel, and elevation is then tested to determine any differences (i.e., if slopes are statistically significantly different, elevation will not be determined). In all figures, n represents the number of biologically independent animals. Sample sizes were chosen based on our previously published experimental results with the effects of dietary interventions^7,13,25–27. Data distribution was assumed to be normal, but this was not formally tested. Phenotypic data (i.e., body weight, body composition, metabolic health) analyses for principal component analysis (PCA) plots were performed in R (v. 4.0.3) using gplots. PCA plots were produced by imputing missing data and scaling the data using the R package “missMDA” using the PCA analysis, and plots were generated using the R packages “factoextra” and “factoMine”^28,29. For the heatmap, gene expression data were logged to the base 2 before the plot was generated³⁰.

Randomization

All studies were performed on animals or on tissues collected from animals. Animals of each sex and strain were randomized into groups of equivalent weight before the beginning of the in vivo studies.

Results

Loss of FGF21 does not affect the response of body weight and composition to a CR diet

The effect of CR on FGF21 levels, as well as the role of FGF21 in the response to CR, is unclear^19,20. We determined FGF21 levels in the blood of both male and female AL-fed mice and mice in which calories were restricted by 30% using Envigo Global 2018 chow, a regimen which we and others have shown extends the lifespan of C57BL/6J mice^5,31. We observed a significant increase of FGF21 level in CR-fed female mice relative to AL-fed controls, but not males, which actually had decreased levels of circulating FGF21 ( Figs. 1A, 2A).

Figure 2.

We recently showed that Fgf21 has sex-dependent roles in the response of C57BL/6J mice to PR, with Fgf21 being vital to many of the metabolic responses to PR in male but not in female mice³². We therefore hypothesized that loss of Fgf21 might have sex-dependent impacts on the response to CR, particularly with respect to metabolic health improvements. To examine this possibility, we placed WT and KO male and female mice on either AL or CR diets for 15 weeks ( Figs. 1B, 2B). Deletion of Fgf21 was confirmed by PCR genotyping, ELISA, and examining Fgf21 gene expression in the iWAT (Fig. S1A–D).

CR is well known to reduce weight gain and adiposity, and we therefore followed the weight and body composition of the mice during the course of the study. Both male WT and male KO mice had reduced weight gain on a CR diet, with a reduction in both lean mass and fat mass gain; the overall effect, as we expected, was one of reduced adiposity ( Fig. 1C–J). While the reduction in weight and lean mass by CR was very similar in both WT and KO mice, there was a greater reduction of fat mass and adiposity in WT mice due to a reduced level of fat and reduced adiposity in AL-fed KO mice. In contrast, in female mice a CR diet reduced weight gain and lean mass gain in both WT and KO mice, but did not have decrease fat mass gain or adiposity in either genotype ( Fig. 2C–J). In accordance with the similar response of both WT and KO mice, food intake was similar between WT and KO mice in both sexes ( Figs. 1K–N, 2K–N).

Loss of Fgf21 has sex-specific impacts on aspects of the response of glucose homeostasis to CR

In mammals fed a CR diet, one of the most striking and broadly conserved effects is improved glucose homeostasis, which manifests in improved glucose tolerance and improved sensitivity to insulin. As administration of FGF21 promotes insulin sensitivity, we hypothesized that mice lacking Fgf21 would have a reduced improvement in glucose homeostasis when fed a CR diet. After 8 weeks, which we and others have found is sufficient for CR to improve glucose homeostasis^5,33, we performed glucose, insulin, and PTTs. We performed glucose tolerance tests and insulin tolerance tests after both a 7- and 21-hour fasting period to better understand their postprandial glucose regulation (7 hours) and glucose regulation after a fasting period in which CR animals are accustomed to (21 hours). Although we understand that typically animals fast for 3–5 hours to prevent drastic changes in liver function, performing these tests at these times may provide better insight into the regulation of glucose homeostasis in a CR-fed mouse.

We observed that both male WT and KO mice had robust improvements in glucose tolerance when fed a CR diet, whether fasted for 7 or 21 hours ( Fig. 3A, B). In agreement with other recent work from our laboratory, while CR-fed mice of both genotypes were extremely insulin sensitive following a prolonged fast, both WT and KO CR-fed mice had a postprandial increase in insulin resistance, which is suggestive of having a tighter regulation of their blood glucose levels ( Fig. 3C). After a longer fast, CR-fed mice regardless of genotype were more insulin sensitive ( Fig. 3D). Fasting glucose and insulin levels were similar in all groups of male mice ( Fig. 3E–H). However, KO CR-fed males had increased FBG compared to WT CR-fed males after a 21-hour fast ( Fig. 3F). In addition, after refeeding, KO AL-fed males had higher insulin levels than mice in any other group ( Fig. 3H).

Figure 3.

Investigating the mechanisms by which CR improves glucose tolerance, we found that CR-fed mice of both genotypes had significantly improved tolerance to pyruvate, indicating improved suppression of hepatic gluconeogenesis ( Fig. 3I). However, CR-fed WT male mice were more pyruvate tolerant than CR-fed KO mice ( Fig. 3I).

In agreement with what we observed in male mice, we observed improved glucose tolerance with CR in both WT and KO females after 7 hours of fasting ( Fig. 4A). However, the effect of CR after 21 hours of fasting was significantly different only in WT mice; CR did not significantly improve glucose tolerance in KO females at this time point ( Fig. 4B). Similar to males, CR-fed females of both genotypes were extremely insulin sensitive following a prolonged fast; however, unlike in males, KO CR-fed mice did not have a postprandial increase in insulin resistance ( Fig. 4C, D). Fasting insulin and glucose levels in KO female mice responded to CR differently than KO male mice ( Fig. 4E–H). KO CR-fed females had increased FBG compared to WT CR-fed females after 16 or 21 hours of fasting, whereas males only showed this effect after 21 hours of fasting ( Fig. 4E, F). With regards to fasting insulin levels, female mice did not show a statistically significant effect of diet, and KO CR-fed females had significantly higher insulin levels following refeeding than their WT CR-fed counterparts ( Fig. 4G, H). As with males, CR-fed females of both genotypes show improved suppression of hepatic gluconeogenesis during a PTT ( Fig. 4I). We conclude that Fgf21 plays a role in the regulation of glucose metabolism in CR-fed female mice, but not in male mice.

Figure 4.

Loss of Fgf21 does not affect CR-induced changes in components of energy balance and fuel utilization

We next observed respiration, energy expenditure, and spontaneous activity using metabolic CLAMS-HC cages. CR has been previously shown to have a very distinct respiratory exchange ratio (RER) curve and reduced energy expenditure⁵. As Fgf21 is a critical regulator of energy balance through modulation of food intake and energy expenditure, we hypothesized that loss of Fgf21 will affect energy balance in CR-fed mice. We, therefore, placed the mice in metabolic chambers, allowing us to determine energy expenditure via indirect calorimetry while also assessing food consumption, activity, and fuel source utilization.

We examined the contribution of diet and genotype to energy expenditure in male mice over a 24-hour period, correcting for differences in body weight ( Fig. 5A) and lean mass ( Fig. 5B) using ANCOVA. Male mice on CR diets had decreased energy expenditure relative to AL-fed male mice, regardless of genotype ( Fig. 5A, B). CR mice undergo rapid lipogenesis following refeeding, then sustain themselves via the utilization of these stored lipids^7,34. We determined substrate utilization by examining the RER, the ratio of O₂ consumed and CO₂ produced; a value close to 1.0 indicates that carbohydrates are primarily utilized for energy production and for active de novo lipogenesis, whereas a value approaching 0.7 indicates that lipids are the predominant energy source³⁵. Both WT and KO CR-fed male mice displayed a distinct pattern in RER, with a rapid increase in RER indicating lipogenesis following once-per-day feeding ( Fig. 5C–D). We also observed no effect of genotype on spontaneous activity, although there was an overall trend of CR-fed males toward increased activity (p = 0.06) ( Fig. 5E). Female mice behaved similarly to male mice, with strong effects of diet but not genotype on energy expenditure and RER ( Fig. 5F–I). While female CR-fed mice also had an overall trend toward increased activity (p = 0.0589), we observed a trend toward decreased activity of KO females (p = 0.0719) ( Fig. 5J).

Figure 5.

Loss of Fgf21 blunts CR-induced reprogramming of WATs in males but not females

Adipose tissue plays a central role in the response to CR;³⁶ in mice, CR promotes the beiging of iWAT, increasing the expression of thermogenic, lipogenic, and lipolytic genes^7,37. FGF21 plays a key role in adipose tissue, promoting energy expenditure in response to various stimuli, in part through promoting the beiging of iWAT^1,38–42. We therefore, examined the role of FGF21 in CR-induced beiging of iWAT. We examined the expression of thermogenic, lipogenic, and lipolytic genes, increases in which are characteristic of beiging iWAT, and we examined iWAT histology.

As expected, in male CR-fed mice, expression of thermogenic genes Ucp1, Cidea, and Elovl3 was increased ( Fig. 6A–C). For all three genes, there was a diminished effect of CR in the KO mice lacking Fgf21. We also looked at the lipogenic genes Dgat1, Fasn, and Acc1. We found that a CR-diet-induced Acc1 and Dgat1 in both WT males and KO males; however, there was a significant effect of genotype and a significant diet × genotype interaction on the induction of Fasn by CR, with significantly less induction of Fasn in KO than in WT males ( Fig. 6D–F). Finally, we examined the expression of two lipolytic genes, Atgl and Lipe. Similar to Fasn, we found a significant induction of both genes by CR in WT males, but a significant effect of genotype and a significant diet × genotype interaction resulting from a blunting of the induction of these genes in CR-fed KO males ( Fig. 6G–H). Finally, we characterized iWAT histology. We observed a strong visual effect of CR on multilocularity in WT males, indicative of adipose tissue beiging, but a reduced effect in KO mice, which is supported by the quantification of multilocular cells per animal ( Fig. 6I–K). Therefore, loss of Fgf21 seems to slightly blunt CR-induced beiging of male mice.

Figure 6.

In contrast, in female mice we observed a substantially decreased effect of CR on gene expression as compared to males, even in WT mice ( Fig. 7A–H). There was not a significant effect of CR on the expression of either Ucp1 or Lipe. There was a significant overall effect of CR on the expression of the thermogenic genes Cidea and Elovl3, but in contrast to males, a similar effect was seen in both WT and KO females ( Fig. 7B, C). Also, in contrast to males, deletion of Fgf21 did not impact lipogenic or lipolytic gene expression in CR-fed females ( Fig. 7D–H). Histological examination of iWAT suggests that CR increases beiging equally well in both WT and KO female mice ( Fig. 7I–K).

Figure 7.

Discussion

CR improves metabolic health and longevity across diverse species^3,5,43–45. Many of the metabolic effects of CR are at first glance similar to those induced by the energy balance hormone FGF21. FGF21 promotes the browning/beiging of iWAT as well as activating BAT, thereby leading to increased energy expenditure and reduced body weight, several dietary interventions that increase FGF21 extend lifespan, and overexpression of FGF21 is sufficient to extend mouse lifespan^{1,18,38,39,42,46}. Some work has identified a role for FGF21 in specific phenotypes of CR in male mice and rats^19,20, and we found that morning-fed CR mice display a sex-specific induction of FGF21 in their natural feeding state ( Figs. 1A, 2A). While it is plausible that CR promotes metabolic health in part via FGF21, the requirement for FGF21 in the metabolic response to CR has not been comprehensively investigated.

Here, we examined the role of FGF21 in the metabolic response to 30% CR in male and female mice fed once per day in the morning, a paradigm widely used in previous studies. We find that deletion of Fgf21 does not substantially alter the effect of CR on body weight, composition, or food consumption in either males or females ( Fig. 8A). However, when we examined the regulation of glucose homeostasis, we identified several roles for FGF21 in the response to CR. In males, WT mice fed a CR diet had better suppression of glucose production from pyruvate and higher FBG than CR-fed KO males. We observed a similar trend in females, with KO females also showing a blunted improvement in glucose tolerance on a CR diet as compared to WT mice when examined after a prolonged fast, and KO CR-fed females also had increased FBG at two time points. WT female mice, but not KO females, showed a postprandial increase in insulin resistance. Our results are consistent with a minor role for Fgf21 in the suppression of hepatic glucose output during CR in both sexes, as well as literature showing that administration of FGF21 promotes hepatic insulin sensitivity⁴⁷.

Figure 8.

Loss of Fgf21 also does not substantially alter the effects of CR on energy balance in either male or female mice. This was rather surprising, as the role of FGF21 in these phenotypes in response to restriction of protein or dietary amino acids has been well established^10,13,48,49. A limitation of the literature is that much of this work has been performed in C57BL/6J male mice, and we have found that many effects of PR depend on sex and strain. This includes the role of FGF21, which we have found plays an important role in the response to PR in C57BL/6J males, but not C57BL/6J females³². Our results are consistent with a model in which FGF21 is not essential for the effects of CR on energy balance.

In light of this conclusion, it is somewhat surprising that loss of Fgf21 blunts the beiging of WAT of male mice placed on a CR diet ( Fig. 8B). We find that CR-fed KO male mice do not have increased expression of the thermogenic gene Ucp1, typically thought of as a main driver for thermogenesis during beiging. However, the expression of two other thermogenic genes, Cidea and Elovl3, are also elevated by CR in males, although the induction of these genes by CR is blunted in the absence of FGF21. Elovl3, which leads to an increase in very long-chain fatty acids, is an adaptive thermogenesis marker found to be increased in BAT in response to cold exposure⁵⁰. Futile cycling between lipogenic and lipolytic gene expression has also been shown to be a characteristic of beiging WAT^13,51. We also found blunted expression of lipogenic and lipolytic genes in KO CR-fed male mice relative to WT CR-fed males. We also found a reduced effect of multilocular adipocytes at the histological level, suggesting that loss of Fgf21 is required for the full induction of beiging by CR in male mice. Intriguingly, these effects are sex-specific, as in CR-fed female mice, beiging is not induced to the same extent as in the males, loss of Fgf21 does not impact thermogenic, lipogenic, or lipolytic gene induction in response to CR in females, and KO female mice on a CR diet have histologically similar iWAT to WT CR-fed female mice. We speculate that these sex differences may be due to the CR regimen itself, such that females may need a different CR regimen to experience the same metabolic benefit as males, as indicated by previous research showing that different sexes maximally benefited from different degrees of CR³¹.

Limitations of this study include that our study lasted less than 4 months; there may be more prominent differences in loss of Fgf21 under a longer CR duration that we did not see in this short time frame, and we did not examine the contribution of FGF21 to the effects of CR on frailty, cognition, or lifespan. We examined the role of FGF21 on the response of iWAT to CR, but we did not examine the molecular response of other adipose depots or other tissues. We collected tissues for molecular examination in the refed state, but due to the distinct eating patterns of CR-fed mice, examining tissues in other feeding states might prove insightful. We fed CR mice in the morning shortly after lights on; while this intervention extends lifespan^5,7, recently it has been suggested that there may be distinct effects of feeding mice in the morning or evening⁵², and examining CR mice fed in the evening might find different results. Finally, Green and colleagues showed that only C57BL/6J males had a robust induction of FGF21 in response to PR, whereas females had a reduced response (p = 0.056), and FGF21 did not increase in response to PR in either sex of DBA/2J or UM-HET3 mice³². Examining the response to CR in multiple strains might reveal more details about the role of FGF21 as well as FGF21-independent mechanisms in the metabolic response to CR.

In summary, here we have tested the hypothesis that the nutrient-responsive hormone FGF21 is a key mediator of the CR-induced improvements in body composition, glucose homeostasis, and energy balance. We have determined that deletion of Fgf21 from the whole body does not impair the ability of CR to improve body composition or energy balance and has relatively minor effects on CR-induced improvements in glucose regulation. In agreement with the literature linking FGF21 to the beiging of iWAT, we find that FGF21 plays a key role in CR-induced beiging of iWAT, but only in males. Here, we disprove the hypothesis that FGF21 mediates the metabolic benefits of CR, bringing us one step closer to identifying the true molecular mediators of the beneficial effects of CR. Notably, we have not excluded a role for FGF21 in the effects of CR on frailty or lifespan. In addition, we have not excluded the role of factors upstream of FGF21, including the general control nonderepressible 2 kinase, activating transcription factor 4, nor factors downstream of FGF21, such as IGF-1 and mechanistic target of rapamycin, in the response to CR.

Acknowledgments

The authors would like to thank all members of the Lamming laboratory, including those who provided any advice during this project. The Lamming laboratory is supported in part by the NIH/National Institute on Aging (NIA) (AG056771, AG062328, and AG061635 to D.W.L.), NIH/National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) (DK125859 to D.W.L.), and startup funds from the University of Wisconsin–Madison School of Medicine and Public Health and Department of Medicine to D.W.L. M.F.C. was supported in part by a Diana Jacobs Kalman/AFAR Scholarships for Research in the Biology of Aging. C.Y.-Y. was supported by a training grant (T32AG000213) and an NIA F32 postdoctoral fellowship (F32AG077916). R.B. is supported by a training grant (T32DK007665). C.L.G. was supported in part by a Glenn/AFAR postdoctoral fellowship. H.H.P. was supported by an NIA F31 predoctoral fellowship (AG066311). Support for this research was provided by the University of Wisconsin–Madison Office of the Vice Chancellor for Research and Graduate Education with funding from the Wisconsin Alumni Research Foundation. The University of Wisconsin Carbone Cancer Center (UWCCC) Experimental Pathology Laboratory is supported by P30 CA014520 from the NIH/National Cancer Institute (NCI). This work was supported in part by the US Department of Veterans Affairs (I01-BX004031), and this work was supported using facilities and resources from the William S. Middleton Memorial Veterans Hospital. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH. This work does not represent the views of the Department of Veterans Affairs or the U.S. Government.

Author Contributions

M.F.C., H.H.P., and D.W.L. conceived of and designed the experiments. M.F.C., I.A., C.Y.Y., R.B., A.M.B., I.K., and M.M.S. performed the experiments. M.F.C., I.K., C.L.G., and D.W.L. analyzed the data and wrote this article.

Declaration of Interests

D.W.L. has received funding from, and is a scientific advisory board member of, Aeovian Pharmaceuticals, which seeks to develop novel, selective mechanistic target of rapamycin (mTOR) inhibitors for the treatment of various diseases. The remaining authors declare no competing interests.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Supplemental Information

Supplemental information can be found online at Supplementary.

References

1. Green C.L. , Lamming D.W. , & Fontana L. (2022). Molecular mechanisms of dietary restriction promoting health and longevity. Nat. Rev. Mol. Cell. Biol. 23(1), 56–73. PMID: 34518687; doi: 10.1038/s41580-021-00411-4.

2. Lin S.J. , Defossez P.A. , & Guarente L. (2000). Requirement of NAD and SIR2 for life-span extension by calorie restriction in Saccharomyces cerevisiae . Science 289(5487), 2126–2128. PMID: 11000115; doi: 10.1126/science.289.5487.2126.

3. Mattison J.A. , Colman R.J. , Beasley T.M. , … de Cabo R. , & Anderson R.M. (2017). Caloric restriction improves health and survival of rhesus monkeys. Nat. Commun. 8, 14063. PMID: 28094793; doi: 10.1038/ncomms14063.

4. McCay C.M. , Crowell M.F. , & Maynard L.A. (1935). The effect of retarded growth upon the length of life span and upon the ultimate body size. J. Nutr. 5, 155–171. PMID: 2520283; doi: 10.1093/jn/10.1.63.

5. Pak H.H. , Haws S.A. , Green C.L. , … Denu J.M. , & Lamming D.W. (2021). Fasting drives the metabolic, molecular and geroprotective effects of a calorie-restricted diet in mice. Nat. Metab. 3(10), 1327–1341. PMID: 34663973; doi: 10.1038/s42255-021-00466-9.

6. Wei S. , Zhao J. , Bai M. , Li C. , Zhang L. , & Chen Y. (2019). Comparison of glycemic improvement between intermittent calorie restriction and continuous calorie restriction in diabetic mice. Nutr. Metab. (Lond.) 16, 60. PMID: 31485253; doi: 10.1186/s12986-019-0388-x.

7. Yu D. , Tomasiewicz J.L. , Yang S.E. , … Syed F.A. , & Lamming D.W. (2019). Calorie-restriction-induced insulin sensitivity is mediated by adipose mTORC2 and not required for lifespan extension. Cell Rep. 29(1), 236–248.e233. PMID: 31577953; doi: 10.1016/j.celrep.2019.08.084.

8. Balasubramanian P. , Schaar A.E. , Gustafson G.E. , … Diffee G.M. , & Anderson R.M. (2021). Adiponectin receptor agonist AdipoRon improves skeletal muscle function in aged mice. Elife 11, e71282. PMID: 35297761; doi: 10.7554/eLife.71282.

9. Bonkowski M.S. , Rocha J.S. , Masternak M.M. , Al Regaiey K.A. , & Bartke A. (2006). Targeted disruption of growth hormone receptor interferes with the beneficial actions of calorie restriction. Proc. Natl. Acad. Sci. U. S. A. 103(20), 7901–7905. PMID: 16682650; doi: 10.1073/pnas.0600161103.

10. Laeger T. , Henagan T.M. , Albarado D.C. , … Schwartz M.W. , & Morrison C.D. (2014). FGF21 is an endocrine signal of protein restriction. J. Clin. Invest. 124(9), 3913–3922. PMID: 25133427; doi: 10.1172/JCI74915.

11. Nishimura T. , Nakatake Y. , Konishi M. , & Itoh N. (2000). Identification of a novel FGF, FGF-21, preferentially expressed in the liver. Biochim. Biophys. Acta. 1492(1), 203–206. PMID: 10858549; doi: 10.1016/s0167-4781(00)00067-1.

12. Yap Y.W. , Rusu P.M. , Chan A.Y. , … Müller O.J. , & Rose A.J. (2020). Restriction of essential amino acids dictates the systemic metabolic response to dietary protein dilution. Nat. Commun. 11(1), 2894. PMID: 32518324; doi: 10.1038/s41467-020-16568-z.

13. Yu D. , Richardson N.E. , Green C.L. , … Malecki K.C. , & Lamming D.W. (2021). The adverse metabolic effects of branched-chain amino acids are mediated by isoleucine and valine. Cell Metab. 33(5), 905–922.e906. PMID: 33887198; doi: 10.1016/j.cmet.2021.03.025.

14. Coskun T. , Bina H.A. , Schneider M.A. , … Moller D.E. , & Kharitonenkov A. (2008). Fibroblast growth factor 21 corrects obesity in mice. Endocrinology 149(12), 6018–6027. PMID: 18687777; doi: 10.1210/en.2008-0816.

15. Inagaki T. , Dutchak P. , Zhao G. , … Mangelsdorf D.J. , & Kliewer S.A. (2007). Endocrine regulation of the fasting response by PPARalpha-mediated induction of fibroblast growth factor 21. Cell Metab. 5(6), 415–425. PMID: 17550777; doi: 10.1016/j.cmet.2007.05.003.

16. Izumiya Y. , Bina H.A. , Ouchi N. , Akasaki Y. , Kharitonenkov A. , & Walsh K. (2008). FGF21 is an Akt-regulated myokine. FEBS Lett. 582(27), 3805–3810. PMID: 18948104; doi: 10.1016/j.febslet.2008.10.021.

17. Xu J. , Stanislaus S. , Chinookoswong N. , … Lindberg R. , & Veniant M.M. (2009). Acute glucose-lowering and insulin-sensitizing action of FGF21 in insulin-resistant mouse models--association with liver and adipose tissue effects. Am. J. Physiol. Endocrinol. Metab. 297(5), E1105–1114. PMID: 19706786; doi: 10.1152/ajpendo.00348.2009.

18. Zhang Y. , Xie Y. , Berglund E.D. , … Kliewer S.A. , & Mangelsdorf D.J. (2012). The starvation hormone, fibroblast growth factor-21, extends lifespan in mice. Elife 1, e00065. PMID: 23066506; doi: 10.7554/eLife.00065.

19. Fujii N. , Uta S. , Kobayashi M. , Sato T. , Okita N. , & Higami Y. (2019). Impact of aging and caloric restriction on fibroblast growth factor 21 signaling in rat white adipose tissue. Exp. Gerontol. 118, 55–64. PMID: 30620889; doi: 10.1016/j.exger.2019.01.001.

20. Thompson A.C. , Bruss M.D. , Nag N. , Kharitonenkov A. , Adams A.C. , & Hellerstein M.K. (2014). Fibroblast growth factor 21 is not required for the reductions in circulating insulin-like growth factor-1 or global cell proliferation rates in response to moderate calorie restriction in adult mice. PLoS ONE 9(11), e111418. PMID: 25369265; doi: 10.1371/journal.pone.0111418.

21. Schwenk F. , Baron U. , & Rajewsky K. (1995). A cre-transgenic mouse strain for the ubiquitous deletion of loxP-flanked gene segments including deletion in germ cells. Nucleic Acids. Res. 23(24), 5080–5081. PMID: 8559668; doi: 10.1093/nar/23.24.5080.

22. Potthoff M.J. , Inagaki T. , Satapati S. , … Kliewer S.A. , & Burgess S.C. (2009). FGF21 induces PGC-1alpha and regulates carbohydrate and fatty acid metabolism during the adaptive starvation response. Proc. Natl. Acad. Sci. U. S. A. 106(26), 10853–10858. PMID: 19541642; doi: 10.1073/pnas.0904187106.

23. Turturro A. , Witt W.W. , Lewis S. , Hass B.S. , Lipman R.D. , & Hart R.W. (1999). Growth curves and survival characteristics of the animals used in the Biomarkers of Aging Program. J. Gerontol. A Biol. Sci. Med. Sci. 54(11), B492–501. PMID: 10619312; doi: 10.1093/gerona/54.11.b492.

24. Bellantuono I. , de Cabo R. , Ehninger D. , … Trejo J.L. , & Lamming D.W. (2020). A toolbox for the longitudinal assessment of healthspan in aging mice. Nat. Protoc. 15(2), 540–574. PMID: 31915391; doi: 10.1038/s41596-019-0256-1.

25. Cummings N.E. , Williams E.M. , Kasza I. , … Merrins M.J. , & Lamming D.W. (2018). Restoration of metabolic health by decreased consumption of branched-chain amino acids. J. Physiol. 596(4), 623–645. PMID: 29266268; doi: 10.1113/JP275075.

26. Fontana L. , Cummings N.E. , Arriola Apelo S.I. , … Kimple M.E. , & Lamming D.W. (2016). Decreased Consumption of Branched-Chain Amino Acids Improves Metabolic Health. Cell Rep. 16(2), 520–530. PMID: 27346343; doi: 10.1016/j.celrep.2016.05.092.

27. Yu D. , Yang S.E. , Miller B.R. , … Cryns V.L. , & Lamming D.W. (2018). Short-term methionine deprivation improves metabolic health via sexually dimorphic, mTORC1-independent mechanisms. FASEB J. 32(6), 3471–3482. PMID: 29401631; doi: 10.1096/fj.201701211R.

28. Husson F. , Josse J. , & Lê S. (2008). FactoMineR: An R package for multivariate analysis. J. Stat. Software 25; doi: 10.18637/jss.v025.i01.

29. Kassambara A. , & Mundt F. (2020). factoextra: Extract and visualize the results of multivariate data analyses. R package version 1.0.7. https://CRAN.R-project.org/package=factoextra.

30. Warnes G. , Bolker B. , Bonebakker L. , … Schwartz M. , & Venables B.A. (2005). gplots: Various R programming tools for plotting data (Vol. 2).

31. Mitchell S.J. , Madrigal-Matute J. , Scheibye-Knudsen M. , … Bernier M. , & de Cabo R. (2016). Effects of sex, strain, and energy intake on hallmarks of aging in mice. Cell Metab. 23(6), 1093–1112. PMID: 27304509; doi: 10.1016/j.cmet.2016.05.027.

32. Green C.L. , Pak H.H. , Richardson N.E. , … Simcox J. , & Lamming D.W. (2022). Sex and genetic background define the metabolic, physiologic, and molecular response to protein restriction. Cell Metab. 34(2), 209–226.e205. PMID: 35108511; doi: 10.1016/j.cmet.2021.12.018.

33. Solon-Biet S.M. , Mitchell S.J. , Coogan S.C. , … Simpson S.J. , & Le Couteur D.G. (2015). Dietary protein to carbohydrate ratio and caloric restriction: Comparing metabolic outcomes in mice. Cell Rep. 11(10), 1529–1534. PMID: 26027933; doi: 10.1016/j.celrep.2015.05.007.

34. Bruss M.D. , Khambatta C.F. , Ruby M.A. , Aggarwal I. , & Hellerstein M.K. (2010). Calorie restriction increases fatty acid synthesis and whole body fat oxidation rates. Am. J. Physiol. Endocrinol. Metab. 298(1), E108–116. PMID: 19887594; doi: 10.1152/ajpendo.00524.2009.

35. Hasek B.E. , Stewart L.K. , Henagan T.M. , … Orentreich N. , & Gettys T.W. (2010). Dietary methionine restriction enhances metabolic flexibility and increases uncoupled respiration in both fed and fasted states. Am. J. Physiol. Regul. Integr. Comp. Physiol. 299(3), R728–739. PMID: 20538896; doi: 10.1152/ajpregu.00837.2009.

36. Miller K.N. , Burhans M.S. , Clark J.P. , … Ntambi J.M. , & Anderson R.M. (2017). Aging and caloric restriction impact adipose tissue, adiponectin, and circulating lipids. Aging Cell 16(3), 497–507. PMID: 28156058; doi: 10.1111/acel.12575.

37. Sheng Y. , Xia F. , Chen L. , … Liu J. , & Ding G. (2021). Differential responses of white adipose tissue and brown adipose tissue to calorie restriction during aging. J. Gerontol. A Biol. Sci. Med. Sci. 76(3), 393–399. PMID: 32222773; doi: 10.1093/gerona/glaa070.

38. Cuevas-Ramos D. , Mehta R. , & Aguilar-Salinas C.A. (2019). Fibroblast growth factor 21 and browning of white adipose tissue. Front Physiol. 10, 37. PMID: 30804796; doi: 10.3389/fphys.2019.00037.

39. Douris N. , Stevanovic D.M. , Fisher F.M. , … Bartness T.J. , & Maratos-Flier E. (2015). Central fibroblast growth factor 21 browns white fat via sympathetic action in male mice. Endocrinology 156(7), 2470–2481. PMID: 25924103; doi: 10.1210/en.2014-2001.

40. Hill C.M. , Laeger T. , Dehner M. , … Münzberg H. , & Morrison C.D. (2019). FGF21 signals protein status to the brain and adaptively regulates food choice and metabolism. Cell Rep. 27(10), 2934–2947.e2933. PMID: 31167139; doi: 10.1016/j.celrep.2019.05.022.

41. Klein Hazebroek M. , & Keipert S. (2022). Obesity-resistance of UCP1-deficient mice associates with sustained FGF21 sensitivity in inguinal adipose tissue. Front Endocrinol. (Lausanne) 13. PMID: 36034414; doi: 10.3389/fendo.2022.909621.

42. Veniant M.M. , Hale C. , Helmering J. , … Xu J. , & Lloyd D.J. (2012). FGF21 promotes metabolic homeostasis via white adipose and leptin in mice. PLoS ONE 7(7), e40164. PMID: 22792234; doi: 10.1371/journal.pone.0040164.

43. Belsky D.W. , Huffman K.M. , Pieper C.F. , Shalev I. , & Kraus W.E. (2017). Change in the rate of biological aging in response to caloric restriction: CALERIE biobank analysis. J. Gerontol. A Biol. Sci. Med. Sci. 73(1), 4–10. PMID: 28531269; doi: 10.1093/gerona/glx096.

44. Kraus W.E. , Bhapkar M. , Huffman K.M. , … Fontana L. , & Investigators C . (2019). 2 years of calorie restriction and cardiometabolic risk (CALERIE): Exploratory outcomes of a multicentre, phase 2, randomised controlled trial. Lancet Diabetes Endocrinol. 7(9), 673–683. PMID: 31303390; doi: 10.1016/S2213-8587(19)30151-2.

45. Rhoads T.W. , Clark J.P. , Gustafson G.E. , … Colman R.J. , & Anderson R.M. (2020). Molecular and functional networks linked to sarcopenia prevention by caloric restriction in rhesus monkeys. Cell Syst. 10(2), 156–168.e155. PMID: 31982367; doi: 10.1016/j.cels.2019.12.002.

46. Trautman M.E. , Richardson N.E. , & Lamming D.W. (2022). Protein restriction and branched-chain amino acid restriction promote geroprotective shifts in metabolism. Aging Cell 21(6), e13626. PMID: 35526271; doi: 10.1111/acel.13626.

47. Gong Q. , Hu Z. , Zhang F. , … Xu A. , & Li Y. (2016). Fibroblast growth factor 21 improves hepatic insulin sensitivity by inhibiting mammalian target of rapamycin complex 1 in mice. Hepatology 64(2), 425–438. PMID: 26926384; doi: 10.1002/hep.28523.

48. Hill C.M. , Laeger T. , Albarado D.C. , … Munzberg H. , & Morrison C.D. (2017). Low protein-induced increases in FGF21 drive UCP1-dependent metabolic but not thermoregulatory endpoints. Sci. Rep. 7(1), 8209. PMID: 28811495; doi: 10.1038/s41598-017-07498-w.

49. Wanders D. , Forney L.A. , Stone K.P. , Burk D.H. , Pierse A. , & Gettys T.W. (2017). FGF21 mediates the thermogenic and insulin-sensitizing effects of dietary methionine restriction but not its effects on hepatic lipid metabolism. Diabetes 66(4), 858–867. PMID: 28096260; doi: 10.2337/db16-1212.

50. Fischer A.W. , Behrens J. , Sass F. , … Scheja L. , & Heeren J. (2020). Brown adipose tissue lipoprotein and glucose disposal is not determined by thermogenesis in uncoupling protein 1-deficient mice. J. Lipid. Res. 61(11), 1377–1389. PMID: 32769145; doi: 10.1194/jlr.RA119000455.

51. Mottillo E.P. , Balasubramanian P. , Lee Y.H. , Weng C. , Kershaw E.E. , & Granneman J.G. (2014). Coupling of lipolysis and de novo lipogenesis in brown, beige, and white adipose tissues during chronic beta3-adrenergic receptor activation. J. Lipid. Res. 55(11), 2276–2286. PMID: 25193997; doi: 10.1194/jlr.M050005.

52. Acosta-Rodriguez V. , Rijo-Ferreira F. , Izumo M. , … Green C.B. , & Takahashi J.S. (2022). Circadian alignment of early onset caloric restriction promotes longevity in male C57BL/6J mice. Science 376(6598), 1192–1202. PMID: 35511946; doi: 10.1126/science.abk0297.