Fibroblast growth factor (FGF) 21, a hormone produced by the liver, improves glucose and lipid metabolism. We recently demonstrated that the FGF21 gene (Fgf21) underwent DNA demethylation in the mouse liver via peroxisome proliferator-activated receptor (PPAR) α during the fetal to lactation periods. Furthermore, we found that the DNA methylation state of Fgf21 was involved in obesity in adult animals. In the present study, we analyzed the DNA methylation state of the FGF21 gene (FGF21) in obese patients using genomic DNA extracted from human monocytes and macrophages and investigated the pathophysiological significance of the FGF21 expression response to pemafibrate (PM), a PPARα ligand. We examined 67 patients with obesity stratified into in- and outpatient cohorts. A positive correlation was observed between serum FGF21 levels and triglyceride (TG) levels before PM administration. However, changes in serum FGF21 levels following PM administration did not correlate with the FGF21 DNA methylation rate, except at one CpG site. The body mass index (BMI) and serum TG levels positively correlated with the FGF21 DNA methylation rate, particularly at different CpG positions. A negative correlation was observed between absolute changes in serum FGF21 levels and the ratio of change in serum TG levels after PM administration. Collectively, these results indicate the potential of FGF21 DNA methylation as a surrogate indicator of BMI and serum TG levels, while absolute changes in serum FGF21 levels after PM administration may offer prognostic insights into the efficacy of reducing serum TG levels through PM administration.

Fibroblast growth factor (FGF) 21, a hormone that is primarily synthesized in the liver, plays a pivotal role in orchestrating glucose and lipid metabolism within hepatic tissues [1]. We previously reported the peroxisome proliferator-activated receptor (PPAR) α-dependent demethylation of the mouse FGF21 gene (Fgf21) promoter, which occurred during the embryonic and infantile stages in the liver. The Fgf21 DNA methylation patterns established during the suckling period persisted into adulthood [2]. Furthermore, the DNA demethylation of Fgf21 was promoted by the pharmacological activation of PPARα [3]. We also demonstrated that a reduction in Fgf21 DNA methylation correlated with the enhanced induction of hepatic Fgf21 expression upon the activation of PPARα. This finding provides insights into the mitigation of obesity caused by the consumption of a high-fat diet in adulthood. In human subjects, the DNA methylation status of the FGF21 gene (FGF21) follows a similar trajectory, with patterns being established during the embryonic and infantile phases and subsequently being maintained into adulthood within the liver [2].

Based on these findings, we postulated that the DNA methylation status of FGF21 may be a factor contributing to obesity and, thus, attempted to elucidate the relationship between FGF21 DNA methylation and obesity in humans. However, conducting liver biopsies for research purposes poses ethical challenges. Since PPARα is expressed in human monocytes and macrophages [4, 5], we posited that the FGF21 DNA methylation status within genomic DNA extracted from human monocytes and macrophages may reflect that of the liver. Therefore, in the present study, we obtained genomic DNA from the peripheral blood of obese subjects, examined the FGF21 DNA methylation status, and investigated its relationships with serum FGF21 levels and various metabolic parameters.

Pemafibrate (PM), a selective PPARα modulator, specifically binds to the PPARα ligand-binding domain [6]. Once tethered to PPARα, PM up-regulates the expression of PPARα-target genes, chiefly within human hepatocytes [7]. Consequently, PM is regarded as a potent activator of PPARα. Since FGF21 is a PPARα target gene, the expression of which is positively modulated by PPARα, the administration of PM may up-regulate the expression of FGF21. Since each individual harbors a distinct FGF21 DNA methylation status, the extent of the increase in FGF21 expression may vary among individuals. To intensify FGF21 expression in response to PPARα activation cues, we administered PM to obese subjects, an approach termed the “PM loading test.” We then assessed the extent of the increase in serum FGF21 levels and its relationships with the FGF21 DNA methylation status and several metabolic parameters.

The present study examined patients of both sexes with obesity (BMI >25), aged between 20 and 75 years, who were not receiving any fibrate drugs (specifically PM, fenofibrate, or bezafibrate). Between 2020 and 2022, 67 obese Japanese patients, comprising 28 inpatients mainly admitted for diabetes management and 29 outpatients, were recruited for this investigation. Ethical approval was granted by the Ethical Committee for Human Research at the Dokkyo Medical University Saitama Medical Center (approval number: #1984), and all procedures were conducted in accordance with the principles outlined in the Declaration of Helsinki. Written informed consent was duly obtained from all participants.

Peripheral blood samples (7 mL) were collected in EDTA-2Na tubes and genomic DNA was extracted using the Puregene Blood Kit (Cat#158023, Qiagen, Hilden, Germany). Following DNA extraction, a bisulfite treatment was performed using the EpiTect Fast DNA Bisulfite Kit (#59824, Qiagen, Hilden, Germany), followed by pyrosequencing on a PyroMark Q24 Advanced system (Qiagen, Hilden, Germany).

After the bisulfite treatment, PCR for pyrosequencing was conducted using the PyroMark PCR Kit (#978703, Qiagen, Hilden, Germany). The PCR profile consisted of 45 cycles at 94°C for 30 s, 62°C for 30 s, and 72°C for 30 s for FGF21 gene and at 94°C for 30 s, 58°C for 30 s, and 72°C for 30 s for NFATC2IP gene. The specific PCR primers for the CpG site position (Pos.) 1–4 (length 235 bp) in FGF21 gene were as follows:

• Forward primer: 5'-GTATTTGAGTAGGGATAGATGAGG-3'

• Reverse primer: 5'-CCACCTCTTAAACCATCTATACAAATCA-3'

• Sequencing primer: 5'-GAGGTTGAGGTTGGT-3'

Primer sequences for the CpG site Pos. 16–20 (length 159 bp) in FGF21 gene were:

• Forward primer: 5'-TGGATAATTGGAATTTGGTATTAATT-3'

• Reverse primer: 5'-TCCCAACAAAAAACCAACCAACA-3'

• Sequencing primer: 5'-TTTATATTTAGGAGATTATTTGAGG-3'

For a CpG site at cg26663590 (length 131bp) of NFATC2IP gene, the primer sequences were:

• Forward primer: 5'-GGGTTTTTAGTTATTTTAGGTAGTGG-3'

• Reverse primer: 5'-TTTAGTTATTTTAGGTAGTGGGAAAATTT-3'

• Sequencing primer: 5'-TTTGGGTTTTTAGTTATTTTAGGTAGTG-3'

The Quantikine ELISA Human FGF-21 Immunoassay kit (#DF2100, R&D Systems, Minneapolis, MN, USA) was employed to quantify serum FGF21 levels.

PM (0.2 mg) was orally administered to patients following their evening meal. Serum FGF21 levels were measured the next morning while fasting.

Continuous and parametric data are expressed as the mean ± standard deviation, while non-parametric data are presented as medians (min-max). To assess differences in the sex distribution between groups, the Student’s two-tailed t-test was applied. The t-test was used to evaluate disparities in BW, HDL-C, and DNA methylation positions, excluding Position 1. The Mann-Whitney test was utilized for other group comparisons. ANOVA was employed to scrutinize variations between Pos. 1–4 and Pos. 16–20, respectively. Correlation analyses were conducted using Pearson’s correlation test or Spearman’s correlation test, contingent on the normal distribution of data. To establish normality, the Shapiro-Wilk test was performed for continuous variables. Significance was ascribed at p < 0.05. All statistical analyses were performed using SPSS 28.0 for Macintosh (SPSS Inc., Chicago, IL, USA) and GraphPad Prism Version 10.2.1 for macOS (GraphPad Software, San Diego, CA, USA).

A cohort of 67 obese Japanese patients, consisting of 41 males and 26 females, with a mean age of 51.5 years and a median BMI of 39.0 kg/m² were examined in the present study (Table 1). Median HbA1c and TG levels were 7.8% and 135 mg/dL, respectively. When stratified into in- and outpatients, HbA1c levels were significantly higher in inpatients than in outpatients (10.5% vs. 6.6%, p < 0.001). On the other hand, systolic blood pressure (SBP) was significantly higher in outpatients than in inpatients (119 mmHg vs. 127 mmHg, p < 0.001) (Table 2). However, no significant differences were observed in age, sex, BW, BMI, diastolic blood pressure (DBP), TG, HDL-C, LDL-C, AST, or ALT between the two groups. Furthermore, there were no marked differences in baseline serum FGF21 levels between the two groups. Serum FGF21 levels positively correlated with serum TG levels at the baseline (r = 0.391, p = 0.001, Fig. 1A). However, neither serum FGF21 nor TG levels correlated with BMI at the baseline (Fig. 1B, C).

Fig. 1

Correlation analysis of serum FGF21 levels, TG levels, and BMI at baseline

A: Correlation between serum FGF21 levels and TG levels.

B: Correlation between serum FGF21 levels and BMI.

C: Correlation between serum TG levels and BMI.

Asterisks indicate a significant difference (** p < 0.01).

Since we used whole blood to extract genomic DNA, we evaluated the monocyte counts of in- and out-patients. The median monocyte count was 346.5 (min-max: 238.5–872.0) for the in-patients (n = 21) and 335.0 (226.2–1,296.0) for the out-patients (n = 24). No significant differences in the monocyte counts were found between the two groups (p = 0.43). There were no significant correlations between monocyte counts and the FGF21 DNA methylation rate at all positions in all the patients for whom monocyte count data are available (n = 45) (Supplementary Table 1).

The average DNA methylation rates of CpG sites located up- (Fig. 2A; Pos. 1–4) and downstream (Fig. 2A; Pos. 16–20) of the transcription start site of the FGF21 promoter were 81.5% (Fig. 2B) and 93.0% (Fig. 2B), respectively. The former exhibited a significantly lower methylation rate than the latter. Additionally, the DNA methylation rate of Pos. 3 was the highest among Pos. 1–4 (Fig. 2B), and Pos. 18 displayed the lowest methylation rate among Pos. 16–20 (Fig. 2B). There were no significant differences in the average DNA methylation rate of Pos. 1–4 between in- and outpatients (Fig. 2C). However, the average DNA methylation rate of Pos. 16–20 was significantly lower in inpatients than in outpatients (Fig. 2D).

Fig. 2

DNA methylation analysis of FGF21

A: Schematic representation of the promoter region of FGF21. Open circles and gray boxes indicate CpG sites and PPAR response elements (PPREs), respectively. TSS: Transcription Start Site. bp: base pairs. Each number above the circles indicates the position. B: DNA methylation rates at Position (Pos.) 1–4 and 16–20.

DNA methylation rates of in- and outpatients at Pos. 1–4 (C) and Pos. 16–20 (D).

In the box plot, each box represents the interquartile range (IQR) of the data distribution, with the median indicated by a horizontal line inside the box. Whiskers extend to the minimum and maximum values within 1.5 times the IQR from the lower and upper quartiles, and any data points beyond the whiskers are shown as individual points (outliers). Asterisks indicate a significant difference (*** p < 0.001).

N.S.: not significant. The significance of differences between the two groups of the entire set of CpG sites as indicated by bars was evaluated with the Mann-Whitney U-test (# <0.05, ### <0.001).

Serum FGF21 levels did not correlate with the FGF21 DNA methylation rate at any position, except for Pos. 19, in all patients (Fig. 3A, B). However, BMI positively correlated with the FGF21 DNA methylation rate at Pos. 1–4, particularly at Pos. 2 and 4 (Fig. 4A, B). SBP also positively correlated with the FGF21 DNA methylation rate at Pos. 1–4, particularly at Pos. 4, and negatively correlated at Pos. 18 (Supplementary Table 2). No correlations were noted between DBP and the FGF21 DNA methylation rate at any position (Supplementary Table 3). Furthermore, baseline serum TG levels positively correlated with the FGF21 DNA methylation rate at Pos. 16–20, particularly at Pos. 18 (Fig. 5A, B).

Fig. 3

Correlation between baseline serum FGF21 levels and DNA methylation rates at Pos. 1–4 (A) and Pos. 16–20 (B) in all patients

The asterisk indicates a significant difference (* p < 0.05). ave: average.

Fig. 4

Correlation between BMI at baseline and DNA methylation rates at Pos. 1–4 (A) and Pos. 16–20 (B) in all patients

Asterisks indicate a significant difference (* p < 0.05). ave: average.

Fig. 5

Correlation between baseline serum TG levels and DNA methylation rates at Pos. 1–4 (A) and Pos. 16–20 (B) in all patients

The asterisk indicates a significant difference (* p < 0.05). ave: average.

DNA methylation at a single CpG locus (cg26663590) of NFATC2IP was recently identified as being causally related to BMI [8]. The average DNA methylation rate at cg26663590 was 2.32% (n = 67). Moreover, we found no significant correlation between the NFATC2IP DNA methylation rate at cg26663590 and BMI in all the patients (Supplementary Fig. 1).

Following the PM loading test, absolute changes in serum FGF21 levels showed a non-normal distribution (Supplementary Fig. 2). Absolute changes, but not the ratio of change, were significantly higher in inpatients than in outpatients (Table 2). A negative correlation was observed between absolute changes in serum FGF21 levels and the ratio of change in serum TG levels after the PM loading test in all patients (r = –0.256, p = 0.036, Fig. 6A) and in outpatients (r = –0.422, p = 0.008, Fig. 6B), but not in inpatients (Fig. 6C). The ratio of change in serum TG levels after the PM loading test did not correlate with FGF21 DNA methylation (Supplementary Fig. 3).

Fig. 6

Correlation between absolute changes in serum FGF21 levels and the ratio of change in serum TG levels (changes in TG) after the PM loading test

A: all patients, B: outpatients, C: inpatients

Asterisks indicate a significant difference (* p < 0.05, ** p < 0.01).

FGF21 is a pivotal regulator of energy metabolism across various organs [9]. It promotes fatty acid β-oxidation, leading to reduced serum TG levels and improved insulin resistance [10, 11]. Contrary to the initial hypothesis that elevated FGF21 levels correspond to reduced serum TG levels, the present results revealed an unexpected outcome. Elevated FGF21 levels were previously reported in patients with obesity and type 2 diabetes [12, 13], a phenomenon attributed to “FGF21 resistance” [14, 15]. FGF21 resistance implies diminished responsiveness to exogenous FGF21 in obese individuals [14]. This is corroborated by the reduced expression of FGF21 receptors, such as FGFR1c, FGFR2, and FGFR3, and the co-receptor β-Klotho (encoded by KLB), in the white adipose tissue of obese individuals [14]. Furthermore, the transcript and protein levels of β-Klotho are reduced in human adipose tissue in obesity [16]. In murine models of mild nonalcoholic fatty liver disease, serum FGF21 levels have been shown to reflect liver fat accumulation and disruptions in metabolic pathways within the liver [17]. These findings are consistent with the positive correlation observed between serum FGF21 levels and serum TG levels in the present study.

However, when the PM loading test was conducted, serum FGF21 levels exhibited a more robust response and were significantly elevated, coinciding with a larger reduction in serum TG levels. At higher pharmacological doses, FGF21 overcomes resistance and effectively signals [18]. The administration of PM stimulates PPARα, resulting in the increased secretion of FGF21. Previous studies reported an increase in serum FGF21 levels and a decrease in serum TG levels with the continuous administration of PM [19-21]. In the present study, a single dose of PM reduced serum TG levels, which correlated with an increase in serum FGF21 levels, suggesting that even a single dose of PM may be effective.

Previous research on mice and humans revealed that the DNA methylation rate of the FGF21 gene in the liver was high at Pos. 1–4, but low at Pos. 16–20 [22]. However, the opposite results were obtained in the peripheral blood of human subjects in the present study, indicating differences in DNA methylation between the liver and peripheral blood [23]. Moreover, the present study primarily analyzed samples from obese adult patients with a BMI ≥25, which may have contributed to the highly methylated state observed. Previous studies reported a negative correlation between the DNA methylation rates of Fgf21 in the mouse liver and hepatic FGF21 levels [2, 24]. In contrast, the present study showed a positive correlation between DNA methylation at Pos. 19 and serum FGF21 levels, but not at any other positions. The positive correlation at Pos. 19 suggests that higher DNA methylation levels correspond to elevated FGF21 levels, possibly reflecting the concept of FGF21 resistance. Nevertheless, if a DNA methylation analysis of the FGF21 gene had been performed using peripheral blood-derived genomes in a previous study on mice [25], the findings obtained may have been similar to those using human peripheral blood-derived genomes.

The positive correlations identified in the present study between FGF21 DNA methylation levels, BMI, and serum TG levels are consistent with previous findings. Observational studies that examined the DNA methylation of genes related to obesity in human peripheral blood reported negative correlations between DNA methylation rates and BMI [26]. For example, the genotype of CPT1A has been implicated in the relationship between dietary fat intake and adiposity variables, and a negative correlation between CPT1A DNA methylation and BMI has been reported, indicating that lower DNA methylation corresponds to the increased accumulation of adipose tissue [27]. Conversely, some genes that promote adiposity have shown positive correlations between DNA methylation rates and fat accumulation in DNA methylation studies using human peripheral blood (e.g., ABCG1, which encodes the ATP-binding cassette, subfamily G, member 1, is involved in cholesterol and phospholipid transport in macrophages) [28, 29]. These findings suggest that genes related to obesity and its regulation exhibit phenotypic expression that induces epigenetic dysregulation, reflected in DNA methylation variability. Since DNA methylation in peripheral blood is unlikely to directly contribute to obesity, the causality of the relationship between DNA methylation levels and BMI/TG remains unclear in the context of this study [30].

Recently, Wahl et al. found that DNA methylation at the cg26663590 site of NFATC2IP affects BMI [8]. However, no significant correlation was found between the DNA methylation rate at the site and BMI in the patients enrolled in this study. This may be because all patients enrolled in this study were obese with a BMI of 25 or higher. On the other hand, the DNA methylation rate at the cg26663590 site was much lower than that of FGF21, perhaps supporting that DNA methylation levels at the cg26663590 site decrease with higher BMI [31].

In addition, blood pressure positively correlated with BMI [32] and FGF21 was also found to be associated with blood pressure [33]. In the present study, BMI and SBP both positively correlated with the FGF21 DNA methylation rate at Pos. 1–4, particularly at Pos. 4, suggesting that the DNA methylation rate at these CpG sites in FGF21 is closely related to obesity and hypertension.

However, FGF21 DNA methylation, except at Pos. 16, in peripheral blood did not correlate with absolute changes in serum FGF21 levels after the PM loading test (Supplementary Fig. 4). Since PM primarily acts on the liver [1], the lack of a correlation may be attributed to the examination of FGF21 DNA methylation in peripheral blood in the present study.

There are several limitations that need to be addressed. The present study included patients with obesity and diabetes who were hospitalized for the management of high serum glucose levels. Therefore, systemic chronic inflammation may have been present, which may contribute to FGF21 resistance [34], potentially explaining the lack of correlations between serum FGF21 levels and some metabolic parameters. Furthermore, the present study focused exclusively on patients with a BMI ≥25, while a positive correlation between FGF21 levels and BMI has been reported [35, 36]. A previous study on FGF21 analogs suggested their potential for reducing BMI [37]. However, this study did not assess changes in BW, and inherent biases were present among outpatients regarding adherence to weight reduction measures. Moreover, the present study did not comprehensively analyze FGF21 DNA methylation across all positions. Therefore, some unexamined positions may exhibit correlations with metabolic parameters or serum FGF21 levels.

In conclusion, DNA methylation modifications to the FGF21 gene in peripheral blood were associated with BMI and lipid metabolism in obese patients. Although no direct link was observed between these modifications and changes in the metabolic status during the PM loading test, individuals with a favorable response in serum FGF21 levels also exhibited a significant reduction in TG levels following the administration of a single dose of PM (Graphical Abstract). These results suggest the potential of absolute changes in serum FGF21 levels after the PM loading test to predict the effectiveness of administering PM in order to reduce serum TG levels. To further investigate the weight-reducing effects of PM, future studies may divide participants into two groups based on high and low baseline serum FGF21 levels. Each group may undergo weight reduction through dietary and exercise therapy or the daily administration of PM, enabling a comparison of the weight-reducing effects of these two methods.

Graphical Abstract

In this study, genomic DNA extracted from peripheral whole blood of patients with obesity was used to analyze the DNA methylation status of FGF21 (lower panel). Baseline BMI was positively correlated with DNA methylation rates of Pos. 1–4 (especially Pos. 2 and 4). Baseline serum TG and FGF21 levels were positively correlated, the former with DNA methylation rates in Pos. 16–20 (especially Pos. 18) and the latter with DNA methylation rates in Pos. 19. Circled are CpG sites; Pos: position

K.H. conceived the project, recruited subjects, conducted the PM loading test, evaluated the data, wrote the manuscript, and supervised the entire project. H.S. recruited subjects, evaluated the data, wrote the manuscript, and prepared figures and tables. S.K. and M.G-K. analyzed the data, wrote the manuscript, and prepared figures and tables. A.T., K.Tsu, and K.F. contributed to the DNA methylation analysis and measurements of serum FGF21 levels. M.Y., K. Hara, T.T., and K.Ta. recruited subjects, conducted the PM loading test, and evaluated the data. All authors discussed the results, provided comments on the manuscript, and gave their final approval for the version to be submitted.

The authors declare no competing interests.

The authors have nothing to disclose. K.H. is a member of the Endocrine Journal’s Editorial Board.

This work was supported in part by a Grant-in-Aid for Scientific Research (KAKENHI) from the Japan Society for the Promotion of Science (JSPS) [grant numbers 23K08015 and 23K10720] (to K.H., K. Hara, respectively) and Dokkyo International Medical Education Research Foundation (to K.H.).

• 1   Fisher  FM,  Maratos-Flier  E (2016) Understanding the physiology of FGF21. Annu Rev Physiol 78: 223–241.
• 2   Yuan  X,  Tsujimoto  K,  Hashimoto  K,  Kawahori  K,  Hanzawa  N, et al. (2018) Epigenetic modulation of Fgf21 in the perinatal mouse liver ameliorates diet-induced obesity in adulthood. Nat Commun 9: 636.
• 3   Ehara  T,  Kamei  Y,  Yuan  X,  Takahashi  M,  Kanai  S, et al. (2015) Ligand-activated PPARalpha-dependent DNA demethylation regulates the fatty acid beta-oxidation genes in the postnatal liver. Diabetes 64: 775–784.
• 4   Duan  SZ,  Usher  MG,  Mortensen  RM (2009) PPARs: the vasculature, inflammation and hypertension. Curr Opin Nephrol Hypertens 18: 128–133.
• 5   Varga  T,  Czimmerer  Z,  Nagy  L (2011) PPARs are a unique set of fatty acid regulated transcription factors controlling both lipid metabolism and inflammation. Biochim Biophys Acta 1812: 1007–1022.
• 6   Yamamoto  Y,  Takei  K,  Arulmozhiraja  S,  Sladek  V,  Matsuo  N, et al. (2018) Molecular association model of PPARalpha and its new specific and efficient ligand, pemafibrate: structural basis for SPPARMalpha. Biochem Biophys Res Commun 499: 239–245.
• 7   Raza-Iqbal  S,  Tanaka  T,  Anai  M,  Inagaki  T,  Matsumura  Y, et al. (2015) Transcriptome analysis of K-877 (a Novel Selective PPARalpha Modulator (SPPARMalpha))-regulated genes in primary human hepatocytes and the mouse liver. J Atheroscler Thromb 22: 754–772.
• 8   Wahl  S,  Drong  A,  Lehne  B,  Loh  M,  Scott  WR, et al. (2017) Epigenome-wide association study of body mass index, and the adverse outcomes of adiposity. Nature 541: 81–86.
• 9   Potthoff  MJ,  Kliewer  SA,  Mangelsdorf  DJ (2012) Endocrine fibroblast growth factors 15/19 and 21: from feast to famine. Genes Dev 26: 312–324.
• 10   Kharitonenkov  A,  Shiyanova  TL,  Koester  A,  Ford  AM,  Micanovic  R, et al. (2005) FGF-21 as a novel metabolic regulator. J Clin Invest 115: 1627–1635.
• 11   Talukdar  S,  Zhou  Y,  Li  D,  Rossulek  M,  Dong  J, et al. (2016) A long-acting FGF21 molecule, PF-05231023, decreases body weight and improves lipid profile in non-human primates and type 2 diabetic subjects. Cell Metab 23: 427–440.
• 12   Coskun  T,  Bina  HA,  Schneider  MA,  Dunbar  JD,  Hu  CC, et al. (2008) Fibroblast growth factor 21 corrects obesity in mice. Endocrinology 149: 6018–6027.
• 13   Xu  J,  Lloyd  DJ,  Hale  C,  Stanislaus  S,  Chen  M, et al. (2009) Fibroblast growth factor 21 reverses hepatic steatosis, increases energy expenditure, and improves insulin sensitivity in diet-induced obese mice. Diabetes 58: 250–259.
• 14   Fisher  FM,  Chui  PC,  Antonellis  PJ,  Bina  HA,  Kharitonenkov  A, et al. (2010) Obesity is a fibroblast growth factor 21 (FGF21)-resistant state. Diabetes 59: 2781–2789.
• 15   Ge  X,  Chen  C,  Hui  X,  Wang  Y,  Lam  KS, et al. (2011) Fibroblast growth factor 21 induces glucose transporter-1 expression through activation of the serum response factor/Ets-like protein-1 in adipocytes. J Biol Chem 286: 34533–34541.
• 16   Gallego-Escuredo  JM,  Gomez-Ambrosi  J,  Catalan  V,  Domingo  P,  Giralt  M, et al. (2015) Opposite alterations in FGF21 and FGF19 levels and disturbed expression of the receptor machinery for endocrine FGFs in obese patients. Int J Obes (Lond) 39: 121–129.
• 17   Rusli  F,  Deelen  J,  Andriyani  E,  Boekschoten  MV,  Lute  C, et al. (2016) Fibroblast growth factor 21 reflects liver fat accumulation and dysregulation of signalling pathways in the liver of C57BL/6J mice. Sci Rep 6: 30484.
• 18   Hale  C,  Chen  MM,  Stanislaus  S,  Chinookoswong  N,  Hager  T, et al. (2012) Lack of overt FGF21 resistance in two mouse models of obesity and insulin resistance. Endocrinology 153: 69–80.
• 19   Zhang  Z,  Diao  P,  Zhang  X,  Nakajima  T,  Kimura  T, et al. (2022) Clinically relevant dose of pemafibrate, a novel selective peroxisome proliferator-activated receptor alpha modulator (SPPARMalpha), lowers serum triglyceride levels by targeting hepatic PPARalpha in mice. Biomedicines 10: 1667.
• 20   Yokote  K,  Yamashita  S,  Arai  H,  Araki  E,  Matsushita  M, et al. (2021) Effects of pemafibrate on glucose metabolism markers and liver function tests in patients with hypertriglyceridemia: a pooled analysis of six phase 2 and phase 3 randomized double-blind placebo-controlled clinical trials. Cardiovasc Diabetol 20: 96.
• 21   Araki  M,  Nakagawa  Y,  Oishi  A,  Han  SI,  Wang  Y, et al. (2018) The peroxisome proliferator-activated receptor alpha (PPARalpha) agonist pemafibrate protects against diet-induced obesity in mice. Int J Mol Sci 19: 2148.
• 22   Kawahori  K,  Kondo  Y,  Yuan  X,  Kawasaki  Y,  Hanzawa  N, et al. (2020) Ascorbic acid during the suckling period is required for proper DNA demethylation in the liver. Sci Rep 10: 21228.
• 23   Olsson Lindvall  M,  Angerfors  A,  Andersson  B,  Nilsson  S,  Davila Lopez  M, et al. (2021) Comparison of DNA methylation profiles of hemostatic genes between liver tissue and peripheral blood within individuals. Thromb Haemost 121: 573–583.
• 24   Geissler  C,  Krause  C,  Neumann  AM,  Britsemmer  JH,  Taege  N, et al. (2022) Dietary induction of obesity and insulin resistance is associated with changes in Fgf21 DNA methylation in liver of mice. J Nutr Biochem 100: 108907.
• 25   Yuan  X,  Tsujimoto  K,  Hashimoto  K,  Kawahori  K,  Hanzawa  N, et al. (2018) Epigenetic modulation of Fgf21 in the perinatal mouse liver ameliorates diet-induced obesity in adulthood. Nat Commun 9: 636.
• 26   Aslibekyan  S,  Demerath  EW,  Mendelson  M,  Zhi  D,  Guan  W, et al. (2015) Epigenome-wide study identifies novel methylation loci associated with body mass index and waist circumference. Obesity (Silver Spring) 23: 1493–1501.
• 27   Robitaille  J,  Houde  A,  Lemieux  S,  Perusse  L,  Gaudet  D, et al. (2007) Variants within the muscle and liver isoforms of the carnitine palmitoyltransferase I (CPT1) gene interact with fat intake to modulate indices of obesity in French-Canadians. J Mol Med (Berl) 85: 129–137.
• 28   Dhana  K,  Braun  KVE,  Nano  J,  Voortman  T,  Demerath  EW, et al. (2018) An epigenome-wide association study of obesity-related traits. Am J Epidemiol 187: 1662–1669.
• 29   Klucken  J,  Buchler  C,  Orso  E,  Kaminski  WE,  Porsch-Ozcurumez  M, et al. (2000) ABCG1 (ABC8), the human homolog of the Drosophila white gene, is a regulator of macrophage cholesterol and phospholipid transport. Proc Natl Acad Sci U S A 97: 817–822.
• 30   Ling  C,  Ronn  T (2019) Epigenetics in human obesity and type 2 diabetes. Cell Metab 29: 1028–1044.
• 31   Ma  J,  Joehanes  R,  Liu  C,  Keshawarz  A,  Hwang  SJ, et al. (2022) Elucidating the genetic architecture of DNA methylation to identify promising molecular mechanisms of disease. Sci Rep 12: 19564.
• 32   Kotsis  V,  Stabouli  S,  Bouldin  M,  Low  A,  Toumanidis  S, et al. (2005) Impact of obesity on 24-hour ambulatory blood pressure and hypertension. Hypertension 45: 602–607.
• 33   Greenhill  C (2018) Link between FGF21 and blood pressure. Nat Rev Endocrinol 14: 380.
• 34   Diaz-Delfin  J,  Hondares  E,  Iglesias  R,  Giralt  M,  Caelles  C, et al. (2012) TNF-alpha represses beta-Klotho expression and impairs FGF21 action in adipose cells: involvement of JNK1 in the FGF21 pathway. Endocrinology 153: 4238–4245.
• 35   Zhang  X,  Yeung  DC,  Karpisek  M,  Stejskal  D,  Zhou  ZG, et al. (2008) Serum FGF21 levels are increased in obesity and are independently associated with the metabolic syndrome in humans. Diabetes 57: 1246–1253.
• 36   Kralisch  S,  Tonjes  A,  Krause  K,  Richter  J,  Lossner  U, et al. (2013) Fibroblast growth factor-21 serum concentrations are associated with metabolic and hepatic markers in humans. J Endocrinol 216: 135–143.
• 37   Gaich  G,  Chien  JY,  Fu  H,  Glass  LC,  Deeg  MA, et al. (2013) The effects of LY2405319, an FGF21 analog, in obese human subjects with type 2 diabetes. Cell Metab 18: 333–340.