The Gly385(388)Arg Polymorphism of the FGFR4 Receptor Regulates Hepatic Lipogenesis Under Healthy Diet
Stefan Z Lutz  Anita M Hennige  Andreas Peter  Marketa Kovarova Charisis Totsikas  Jürgen Machann  Stefan M Kröber  Bianca Sperl Erwin Schleicher  Fritz Schick  ... Show more
The Journal of Clinical Endocrinology & Metabolism, Volume 104, Issue 6, June 2019, Pages 2041–2053, https://doi.org/10.1210/jc.2018-01573
Published: 12 December 2018  Article history
 pdfPDF  Split View Cite
Permissions Icon Permissions
 Share
Abstract
Context
The effect of a lifestyle intervention to reduce liver fat content in nonalcoholic fatty liver disease in humans is influenced by genetics. We hypothesized that the amino acid exchange in human Gly388Arg (mouse homolog: Gly385Arg) in fibroblast growth factor receptor 4 (FGFR4), which regulates bile acid, lipid, and glucose metabolism, could determine hepatic lipid accumulation and insulin sensitivity. Mechanisms of this substitution were studied in mice under normal chow and high-fat diets.

Design
In humans, the Gly388Arg polymorphism was studied for its relationship with changes in liver fat content and insulin sensitivity during 9 months of a lifestyle intervention. We also studied a knock-in mouse strain with an Arg385 allele introduced into the murine FGFR4 gene under normal chow and high-fat diets.

Results
In humans, the FGFR4 Arg388 allele was not associated with liver fat content or insulin sensitivity in subjects who were overweight and obese before lifestyle intervention. However, it was associated with less decrease in liver fat content and less increase in insulin sensitivity during the intervention. In mice receiving normal chow, the FGFR4 Arg385 allele was associated with elevated hepatic triglyceride content, altered hepatic lipid composition, and increased hepatic expression of genes inducing de novo lipogenesis and glycolysis. Body fat mass and distribution, glucose tolerance, and insulin sensitivity were unaltered. The FGFR4 Arg385 allele had no effect on glucose or lipid metabolism under the high-fat diet.

Conclusion
Our data indicate that the FGFR4 Arg388(385) allele affects hepatic lipid and glucose metabolism specifically during healthy caloric intake.

Issue Section: Growth, Growth Hormone, and Growth Factors
Nonalcoholic fatty liver disease (NAFLD) is strongly associated with insulin resistance and is an important risk factor for the development of type 2 diabetes mellitus and cardiovascular disease (1–6). Strategies to prevent and treat NAFLD are not fully understood. Lifestyle intervention (LI) in general is an effective and safe strategy to treat NAFLD; however, not all participants benefit to the same extent (7), indicating that the success of NAFLD treatment may depend on genetic variation. In humans, we found that variations in the peroxisome proliferator-activated receptor (PPAR)-δ gene, which encodes a nuclear transcription factor controlling fatty acid oxidation and energy uncoupling, determined changes in hepatic fat content during an LI (8). Furthermore, we found that cardiorespiratory fitness was an independent predictor of reduction in liver fat content during an LI (9). In addition, presence of the -8503 A and -1927 C alleles in the gene encoding the adiponectin receptor 1 (ADIPOR1) predicted changes in liver fat content and insulin sensitivity during an LI (10).

Because lack of the fibroblast growth factor receptor 4 (FGFR4) in mice resulted in features of the metabolic syndrome, such as hyperlipidemia, glucose intolerance, and insulin resistance, but unexpectedly protected from hepatic steatosis (11), we hypothesized that FGFR4 may be another candidate gene of fatty liver affecting the dynamics of liver fat accumulation in humans. FGFR4 belongs to a family of transmembrane receptor protein tyrosine kinases that includes five members: FGFR1 to FGFR5 (12, 13). FGFRs 1 to 4 are single-pass transmembrane tyrosine kinase receptors consisting of a ligand-binding extracellular domain, a single transmembrane domain, and an intracellular kinase domain (14, 15), whereas the recently identified FGFR5 lacks the intracellular tyrosine kinase domain (13). FGFRs are activated by fibroblast growth factors, which are involved in a variety of biological responses such as embryonic development, cell proliferation and migration, angiogenesis, and glucose and lipid homeostasis (16). FGFR4 plays an essential role in suppressing bile acid synthesis and promotes hepatic steatosis under a high-fat diet (HFD). This involves its ligands, the human FGF19, and its mouse ortholog FGF15, which display hormonelike effects (11, 17, 18). Most recently, an FGF19 analogue was shown to reduce liver fat content in patients with nonalcoholic steatohepatitis (19), a finding indicating that FGFR4 may also be important for the regulation of liver fat content in humans.

The polymorphism in codon 388 in the human FGFR4 (mouse homolog FGFR4 codon 385) results in the amino acid exchange of glycine (Gly388) to arginine (Arg388) in the transmembrane domain of the tyrosine kinase and is associated with cancers, such as cancer of the breast, prostate, colon, lung, and skin (20–22). The global minor allele frequency of this common variant is 0.30. However, among patients with cancer, its prevalence is increased up to 50% (20, 23). We previously found that the FGFR4 Gly388Arg substitution induced hypersecretion of insulin in mouse islets and was associated with lower glucose levels in humans (24). Furthermore, the Gly388Arg substitution was associated with activation of the signal transducer and activator of transcription-3 (STAT3) in mice (25) which, when activated, suppresses hepatic gluconeogenic gene expression (26) and induces hepatic lipogenic gene expression (27), pointing to a gain-of-function effect of this amino acid exchange.

Therefore, in the current study, we investigated the role of the human FGFR4 Gly388Arg (mice: Gly385Arg) polymorphism in the regulation of hepatic lipid metabolism and glycemic traits before and after an LI in humans and studied mechanisms of action of this amino acid substitution under normal chow and an HFD in mice.

Materials and Methods
Human data
Subjects
A total of 170 subjects without diabetes from Southern Germany were studied. These individuals participated in a 9-month LI study to reduce adiposity and prevent type 2 diabetes mellitus and had measurements of liver fat content before and after 9 months of the LI (3, 28). For inclusion in the study, at least one of the following criteria had to be fulfilled by the participants: a family history of type 2 diabetes mellitus, previous diagnosis of impaired glucose tolerance or gestational diabetes, or a body mass index (BMI) >27 kg/m2. Moreover, after a standard questionnaire, physical examination, and routine laboratory tests, the participants had to be ascertained as healthy, especially without any history of liver disease.

As assessed by means of a standard questionnaire, participants did not consume more than two alcoholic drinks per day. After the baseline measurements, individuals underwent a 9-month LI composed of dietary counseling with up to 10 sessions with a dietician and at least 3 hours of aerobic endurance exercise per week, as described in detail in (28). Counseling was aimed at reducing body weight by ≥5%. Furthermore, participants were advised to reduce energy intake from fat to <30% of total energy consumed and to increase dietary fiber intake to at least 15 g/4185 kJ (1000 kcal). They were also encouraged to reduce the intake of saturated fatty acids to <10% of total fat. During each visit, participants presented a 3-day food diary and discussed it with the dietitians. Recommended aerobic endurance exercise was walking or swimming with an only moderate increase in heart rate. Informed written consent was also obtained from these participants, and the Ethics Committee of the University of Tübingen approved the protocol. Some of the results from these analyses was also presented in a doctoral thesis of the coauthor (C.T.; https://publikationen.uni-tuebingen.de/xmlui/handle/10900/45745).

Anthropometrics, liver fat content, total body fat mass, and body fat distribution
BMI was calculated as weight divided by the square of height (kg/m2). Waist circumference was measured at the midpoint between the lateral iliac crest and lowest rib. Liver fat content was measured by localized proton magnetic resonance spectroscopy as previously described (29). Total body and visceral fat mass were measured by magnetic resonance tomography using an axial T1-weighed fast spin-echo technique.

Oral glucose tolerance test
The 170 individuals underwent a 2-hour 75-g oral glucose tolerance test (OGTT). We obtained venous plasma samples at 0, 30, 60, 90, and 120 minutes for determination of plasma glucose and insulin levels. Whole-body insulin sensitivity was calculated from glucose and insulin values obtained at five time points (0, 30, 60, 90, and 120 minutes) during a 75-g OGTT, as proposed by Matsuda and DeFronzo (30) (10,000 / √ (mean insulin × mean glucose) × (fasting insulin × fasting glucose). Furthermore, adipose tissue insulin resistance (fasting insulin × fasting free fatty acids) was calculated (31, 32).

Genotyping
For genotyping, DNA was isolated from whole blood using a commercial DNA isolation kit (NucleoSpin; Macherey & Nagel, Düren, Germany). The single-nucleotide polymorphism rs351855, a G to A exchange leading to substitution of Gly by Arg at position 388 of FGFR4, was genotyped using TaqMan assay (Applied Biosystems, Foster City, CA) (24). The overall genotyping success rate was 99.1%. Rescreening of 3.1% of the subjects gave 100% identical results.

Experimental animals
The generation of mice carrying the FGFR4 Arg385 allele was described previously (33). In brief, to replace the Gly codon in position 385 with an Arg codon in the murine FGFR4 gene, a knock-in mouse strain was generated in the genetic background of SV/129 mice. Twelve-week-old male FGFR4 Gly/Arg385 or FGFR4 Arg/Arg385 mice were compared with litter-matched FGFR4 Gly/Gly385 control mice. Mice (4 weeks old) were kept on a 12-hour/12-hour light/dark cycle (lights on at 6:00 AM) and were allowed access to regular chow ad libitum (Diet #1310; Altromin, Lage, Germany) or to an HFD (D12451; Research Diets, New Brunswick, NJ) for 8 weeks. The HFD contained 45% kcal as fat. All animal procedures were performed in accordance with the guidelines of laboratory animal care and were approved by the local governmental commission for animal research.

Western blotting
For the analysis of insulin signal transduction, after an overnight fast, anesthetized mice were injected in the inferior vena cava with 2 units of human insulin. For Western blotting, after 7 minutes liver and skeletal muscle tissues were lysed at 4°C in lysis buffer containing 2 mM EDTA, 137 mM NaCl, 1% NP-40, 10% glycerol, 12 mM β-glycerol phosphate, 1 mM phenylmethylsulfonyl fluoride, and 10 mg/µL leupeptin and aprotinin for 30 minutes and precleared by centrifugation at 12,000g for 20 minutes at 4°C. Western blot analysis of tissue lysates was done with antibodies against phospho-Akt [(34); Ser173; #9271; Cell Signaling Technology, Beverly, MA] and glyceraldehyde-3-phosphate dehydrogenase [(35); #2118; Cell Signaling Technology]. Signals were visualized with an enhanced chemiluminescence system (Amersham Biosciences, Buckinghamshire, UK).

Analysis of body fat mass and liver fat content in mice
For MRI of fat mass, mice were anesthetized and measured on a 3T whole-body magnetic resonance analyzer (Magnetom Trio; Siemens Healthcare, Erlangen, Germany) applying a T1-weighted fast spin-echo technique. Mice were placed in prone position in the wrist coil of the system. Images were recorded with an in-plane spatial resolution of 0.25 mm and a slice thickness of 2 mm. Postprocessing was done as described (36). For the histological detection of liver fat content, liver tissue was fixed in 4% paraformaldehyde in PBS. Hematoxylin and eosin staining of paraffin liver sections was carried out for analysis of fat content by visual inspection of vesicles (37). To determine the liver fat content by clinical chemical methods, triacylglycerides (TAGs) were quantified in the homogenate using the ADVIA 1800 Clinical Chemistry System analyzer (Siemens Healthcare Diagnostics, Eschborn, Germany) and calculated as mg/100 mg tissue.

Analysis of fatty acid composition
For determination of the fatty acid composition of mouse hepatic TAGs, snap frozen tissue samples were homogenized in PBS containing 1% Triton X-100 with a TissueLyser (Qiagen, Hilden, Germany). The TAG fraction was isolated from the liver homogenate extracts using thin-layer chromatography. Transesterification of the fatty acids and quantification by gas chromatography with flame ionization detection were performed as previously described (37).

Analysis of mouse blood samples
Blood glucose was measured from tail bleeds using the Glucometer Elite (Bayer, Elkhart, IN). For insulin measurements, a radioimmunoassay (Linco Research, St. Charles, MO) was used. Nonesterified fatty acids (NEFAs) were analyzed in the EDTA-plasma collected after decapitation. NEFA concentrations were detected using an enzymatic method (WAKO Chemicals, Neuss, Germany) on the clinical chemistry analyzer ADVIA 1800 (Siemens Healthcare Diagnostics).

Analysis of glucose homeostasis and insulin responsiveness in mice
A glucose tolerance test was performed in mice that were fasted overnight. Mice were injected intraperitoneally (i.p.) with 2 g/kg body weight α-D-glucose. Blood was collected immediately before the application and 15, 30, 60, and 120 minutes after injection. Plasma glucose was determined as described previously. To measure glucose-stimulated insulin release, mice were injected i.p. after an overnight fast with 3 g/kg body weight α-D-glucose. Blood was collected immediately before the application and at 2, 5, 10, and 30 minutes after injection. To determine insulin sensitivity, an insulin tolerance test was performed, with mice receiving 1 U/kg body weight insulin i.p. (Actrapid, Novo Nordisk, Denmark). Blood was collected immediately before the application and at 15, 30 and 60 minutes after injection.

Gene expression analysis
For quantification of mRNA expression in mouse liver, tissues were frozen in liquid nitrogen. Total RNA was extracted with the AllPrep Mini Kit (Qiagen) according to the manufacturer’s instructions. After treatment with RNase-free DNase I, total RNA was transcribed into cDNA using the First-Strand cDNA Kit from Roche Diagnostics (Mannheim, Germany). Quantitative PCR was performed on a LightCycler 480 (Roche Diagnostics) using Probes Master and fluorescent probes from the Universal Probe Library (Roche Diagnostics). Primers were obtained from TIB MOLBIOL (Berlin, Germany). Primer sequences can be provided upon request. Measurements were performed at least in duplicate. RNA content was normalized for the housekeeping gene RPS13 using the ΔΔCt method.

Statistical analysis
Data are expressed as mean ± SEM, and the number of independent experiments or mice is indicated in the respective figure legends. Groups were compared by the Cochran-Armitage test for trend based on the linear regression model, where the significance of linear trends across the three genotypes, Gly385(388)Gly, Gly385(388)Arg, and Arg385(388)Arg, was tested using a dose-response effect (here, increasing numbers of the Arg allele), assuming a codominant and a dominant mode of inheritance or two-tailed unpaired t tests, as appropriate. P values <0.05 were considered statistically significant. The statistical software package JMP 13.0 (SAS Institute Inc, Cary, NC) was used.

Results
Humans carrying the FGFR4 Arg385 allele were more resistant to reduction of liver fat content and improvement in insulin sensitivity during the LI
The demographic and metabolic characteristics of the population at baseline and at follow-up are presented in Table 1. During a follow-up period of 9 months, there were decreases in body weight, total body fat, intramyocellular fat, and visceral fat. The largest change was found for liver fat content (−29%). Moreover, fasting as well as 2-hour glycemia and insulinemia and adipose tissue insulin resistance decreased, and whole-body insulin sensitivity increased. At baseline, the FGFR4 Gly388Arg polymorphism was not associated with liver fat content, body fat mass, body fat distribution, or insulin sensitivity. However, during the LI, individuals carrying the minor 388Arg allele had both less decrease in fasting and 2-hour insulinemia and less decrease in 2-hour glycemia and adipose tissue insulin resistance (Table 2) than individuals who were homozygous for the 388Gly allele. Furthermore, subjects carrying the 388Arg allele had a similar decrease in total body fat mass (Fig. 1A), but less reduction in liver fat content and less augmentation of insulin sensitivity compared with homozygous carriers of the 388Gly allele (Fig. 1B and 1C).





Table 1.
Subject Characteristics at Baseline and After 9 Months of Follow-Up in Caucasians

  Baseline Follow-Up P Value
Demographics and body composition
 Sex (male/female) 68/102 —
 Age, y 46 ± 1 47 ± 1 —
 Body weight, kg 85.4 ± 1.2 82.9 ± 1.2 <0.0001
 Body mass index, kg/m2 28.9 ± 0.3 28.1 ± 0.3 <0.0001
 Body fat, kg 25.2 ± 0.7 23.0 ± 0.7 <0.0001
 Visceral fat, kg 3.0 ± 0.1 2.6 ± 0.1 <0.0001
 IMCLtibialis anterior, arbitrary unitsa 4.0 ± 0.1 3.7 ± 0.1 <0.0001
 Liver fat, % 5.1 ± 0.4 3.6 ± 0.3 <0.0001
Metabolic characteristics
 Fasting glucose, mM 5.23 ± 0.04 5.16 ± 0.04 0.01
 2-h glucose, mM 6.87 ± 0.12 6.62 ± 0.12 0.02
 Fasting insulin, pM 59 ± 3 51 ± 2 0.0002
 2-h insulin, pM 480 ± 30 420 ± 33 0.007
 Fasting FFA μM 655 ± 17 611 ± 15 0.02
 2-h FFA, μM 84 ± 6 76 ± 10 0.002
 HOMA-IR index 1.89 ± 0.09 1.60 ± 0.08 <0.0001
 Adipose tissue IR, mmol/l·(µU/mL) 5.46 ± 3.76 4.61 ± 3.37 0.0002
 Whole-body IS, arbitrary units 13.3 ± 0.5 15.1 ± 0.6 0.0001
Values represent mean ± SE. P is for paired differences after log transformation of nonnormally distributed parameters. Bolface indicates P values ≤ 0.05.

Abbreviations: FFA, free fatty acid; HOMA-IR, homeostasis model assessment of insulin resistance; IMCL, intramyocellular lipid; IR, insulin resistance; IS, insulin sensitivity.

a
Available in 149 subjects.

View Large








Table 2.
Associations of Gly388Arg SNP in FGFR4 With Changes in Subject Characteristics During the Lifestyle Intervention

  Gly/Gly
  Gly/Arg
  Arg/Arg
  P Baseline
  P for Change Between Genotypes

Baseline Follow-Up Baseline Follow-Up Baseline Follow-Up Additive Dominant Additive Dominant
Demographics
 Sex (male/female) 29/54 31/39 8/9 0.20a 0.19a
 Age, y 46 ± 1 47 ± 1b 46 ± 1 47 ± 1b 45 ± 4 46 ± 0b 0.53 0.66
 Change in body composition
 Body weight, kg 85.8 ± 1.7 83.1 ± 1.6b 83.8 ± 1.8 81.7 ± 1.8b 89.6 ± 4.3 86.3 ± 4.2b 0.60 0.24 0.74 0.94
 Body mass index, kg/m2 29.0 ± 0.5 28.1 ± 0.5b 28.5 ± 0.5 27.8 ± 0.5b 30.1 ± 1.0 29.0 ± 1.0b 0.71 0.83 0.83 0.95
 Body fat, kg 25.9 ± 1.1 23.7 ± 1.0b 23.7 ± 1.1 21.7 ± 1.1b 28.2 ± 2.2 24.8 ± 1.9c 0.65 0.68 0.42 0.35
 Visceral fat, kg 3.02 ± 0.20 2.61 ± 0.19b 2.84 ± 0.19 2.45 ± 0.19b 3.26 ± 0.51 2.71 ± 0.48b 0.12 0.31 0.61 0.95
 IMCLtibialis anterior, (arbitrary units)d 3.91 ± 0.20 3.78 ± 0.16 4.09 ± 0.20 3.73 ± 0.21e 3.90 ± 0.53 3.39 ± 0.47 0.77 0.36 0.30 0.30
 Liver fat, % 5.11 ± 0.67 3.06 ± 0.37b 5.33 ± 0.65 4.11 ± 0.51b 4.49 ± 0.99 4.10 ± 1.29e 0.90 0.69 0.009 0.007
Change in metabolic characteristics
 Fasting glucose, mM 5.24 ± 0.05 5.17 ± 0.05 5.25 ± 0.06 5.15 ± 0.06 5.11 ± 0.14 5.16 ± 0.15 0.44 0.83 0.58 0.92
 2-h glucose, mM 6.95 ± 0.18 6.45 ± 0.18b 6.93 ± 0.18 6.89 ± 0.19 6.16 ± 0.26 6.39 ± 0.26 0.24 0.71 0.02 0.006
 Fasting insulin, pM 55 ± 4 46 ± 3b 62 ± 5 55 ± 4e 67 ± 8 59 ± 5 0.17 0.10 0.01 0.01
 2-h insulin, pM 465 ± 41 342 ± 34b 506 ± 47 487 ± 60 442 ± 105 524 ± 125 0.74 0.35 0.0005 0.06
 Fasting FFA, μM 674 ± 22 615 ± 20c 618 ± 27 589 ± 26 708 ± 68 675 ± 31 0.68 0.35 0.19 0.56
 2-h FFA, μM 92 ± 11 90 ± 19 73 ± 5 61 ± 4e 92 ± 17 67 ± 6e 0.47 0.30 0.77 0.61
 HOMA-IR index 1.78 ± 0.13 1.46 ± 0.10b 1.97 ± 0.15e 1.70 ± 0.14 2.10 ± 0.28 1.88 ± 0.2 0.25 0.13 0.016 0.017
 Adip. IR, mmol/L·(µU/mL) 5.41 ± 4.00 4.28 ± 3.40b 5.32 ± 3.59 4.70 ± 3.47 6.26 ± 3.45 5.86 ± 2.60 0.36 0.44 0.0047 0.011
 Insulin sensitivityOGTT, arbitrary units 14.08 ± 0.78 16.75 ± 0.94b† 12.52 ± 0.81 13.92 ± 0.87 12.48 ± 1.82 12.38 ± 1.53 0.29 0.13 0.003 0.003
Data represent unadjusted mean ± SE. For statistical analyses, nonnormally distributed parameters were log transformed. The genotype effect at baseline and during the intervention was tested using an additive and a dominant model. Body weight, BMI, and total body fat at baseline were adjusted for age and sex. The other parameters were additionally adjusted for total body fat. For longitudinal analyses, fold-changes in the parameters (follow-up over baseline) were adjusted for the baseline parameters. Body weight, BMI, and body fat were also adjusted for age and sex. The other parameters were additionally adjusted for body fat at baseline and at follow-up. Boldface indicates P values ≤ 0.05.

Abbreviations: Adip. IR, adipose tissue insulin resistance; FFA, free fatty acid; HOMA-IR, homeostasis model assessment of insulin resistance; IMCL, intramyocellular lipid; SNP, single-nucleotide polymorphism.

a
χ2 test.

b
P < 0.001 for paired differences between baseline and follow-up.

c
P < 0.01 for paired differences between baseline and follow-up.

d
Available in 149 subjects (Gly/Gly, n = 74; Gly/Arg, n = 60; Arg/Arg, n = 15).

e
P < 0.05 for paired differences between baseline and follow-up.

View Large
Figure 1.
Percentage of change (indicated as follow-up over baseline) in (A) total body fat, (B) liver fat content, and (C) insulin sensitivity during 9 months of a lifestyle intervention in humans, depending on the respective genotype of the Gly388Arg polymorphism of FGFR4. Change in total body fat was adjusted for age, sex, and total body fat at baseline. Changes in liver fat content and insulin sensitivity were adjusted for age, sex, and body fat at baseline and at follow-up. Statistical significance is given by use of both an additive model and a dominant model for the 388Arg-encoding allele. Values represent means ± SEM.
View largeDownload slide
Percentage of change (indicated as follow-up over baseline) in (A) total body fat, (B) liver fat content, and (C) insulin sensitivity during 9 months of a lifestyle intervention in humans, depending on the respective genotype of the Gly388Arg polymorphism of FGFR4. Change in total body fat was adjusted for age, sex, and total body fat at baseline. Changes in liver fat content and insulin sensitivity were adjusted for age, sex, and body fat at baseline and at follow-up. Statistical significance is given by use of both an additive model and a dominant model for the 388Arg-encoding allele. Values represent means ± SEM.


Mice carrying the FGFR4 Arg385 allele exhibited increased liver fat content and altered hepatic lipid composition without altering body fat mass and body fat distribution under a chow diet
To investigate the role of the FGFR4 Gly388Arg (in mice: Gly385Arg) substitution in the regulation of lipid metabolism, we used a knock-in mouse model in which the Gly codon in position 385 of the murine FGFR4 gene was replaced with an Arg codon. We found no association between the FGFR4 Arg385 allele and body weight (38), in line with total or visceral adipose tissue mass under a chow diet, as measured by MRI (Fig. 2A). Moreover, the FGFR4 Arg385 allele was not associated with NEFA or total cholesterol levels (38). The serum TAGs tended to be higher in the FGFR4 Arg385 mice; however, this difference was not statistically significant (P for trend = 0.165) (Fig. 2B).

Figure 2.
Fat distribution in FGFR4 Gly/Gly385 mice (black bars), FGFR4 Gly/Arg385 mice (white bars), and FGFR4 Arg/Arg385 mice (gray bars) that were fed a chow diet. (A) MRI analysis of visceral and subcutaneous fat deposits (n = 3 per group). (B) Plasma triacylglycerides (n = 9 to 12). (C) Liver triacylglyceride content (n = 7 to 9). (D) Hematoxylin and eosin staining of liver sections. Images are representative of 7 to 9 stained sections per group. Data are expressed as mean ± SEM. TAT, total adipose tissue; VAT, visceral adipose tissue.
View largeDownload slide
Fat distribution in FGFR4 Gly/Gly385 mice (black bars), FGFR4 Gly/Arg385 mice (white bars), and FGFR4 Arg/Arg385 mice (gray bars) that were fed a chow diet. (A) MRI analysis of visceral and subcutaneous fat deposits (n = 3 per group). (B) Plasma triacylglycerides (n = 9 to 12). (C) Liver triacylglyceride content (n = 7 to 9). (D) Hematoxylin and eosin staining of liver sections. Images are representative of 7 to 9 stained sections per group. Data are expressed as mean ± SEM. TAT, total adipose tissue; VAT, visceral adipose tissue.


However, the FGFR4 Arg385 allele was associated with an elevated hepatic TAG content (P for trend = 0.005) (Fig. 2C). As expected, histological analyses confirmed the morphological differences in terms of increases in the number and size of lipid droplets in the liver of FGFR4 Arg/Arg385 mice compared with that of FGFR4 Gly/Gly385 mice (Fig. 2D).

When we analyzed the fatty acid composition of hepatic TAGs, we detected distinct alterations in the livers of FGFR4 Arg385 mice on a chow diet (Table 3). Despite the absolute increase in liver fat content, we detected a relative reduction in saturated fatty acids of 14:0 and 18:0, as well as in polyunsaturated fatty acids of the n-3 series (22:6) and the n-6 series (20:4 and 22:6) in carriers of the FGFR4 Arg385 allele. In contrast, the relative amount of the n-3 series precursor (18:3) was increased. The stearoyl-CoA-desaturase activity index of fatty acid desaturation was significantly increased for the ratio of 18:1/18:0, indicating increased fatty acid synthesis.



Table 3.
Relative Fatty Acid Composition in Percentage of the Triacylglyceride Fraction in the Liver of the Mice

  FGFR4
 
 

  GG GR RR Padd Padjusted for TAG
TAG, % 4.28 ± 0.120 4.32 ± 0.170 5.55 ± 0.305 0.0116
14:0 1.12 ± 0.049 0.980 ± 0.033 0.965 ± 0.052 0.0497 0.2883
16:0 20.66 ± 0.636 21.68 ± 0.410 21.69 ± 0.919 0.3025 0.3399
18:0 1.55 ± 0.255 1.08 ± 0.057 1.03 ± 0.109 0.0426 0.278
20:0 0.058 ± 0.012 0.04 ± 0.008 0.056 ± 0.006 0.9592 0.4741
22:0 0.018 ± 0.004 0.010 ± 0.003 0.010 ± 0.002 0.1652 0.2630
24:0 0.089 ± 0.007 0.106 ± 0.011 0.111 ± 0.016 0.2217 0.0747
t-16:1 n-7 0.014 ± 0.0007 0.016 ± 0.002 0.016 ± 0.006 0.7934 0.2424
c-16:1 n-7 4.82 ± 0.719 5.97 ± 0.329 6.49 ± 0.591 0.0770 0.4495
18:1 n-9 22.23 ± 0.269 21.26 ± 0.567 21.65 ± 0.514 0.4296 0.0513
18:1 n-7 1.87 ± 0.107 1.78 ± 0.037 1.85 ± 0.106 0.8430 0.3243
24:1 n-9 0.071 ± 0.018 0.040 ± 0.011 0.022 ± 0.003 0.0139 0.2064
18:3 n-3 1.25 ± 0.107 1.69 ± 0.106 1.63 ± 0.057 0.0421 0.0544
18:4 n-3 1.25 ± 0.331 1.17 ± 0.068 1.15 ± 0.011 0.9736 0.8183
20:4 n-3 0.032 ± 0.003 0.042 ± 0.007 0.039 ± 0.003 0.2417 0.6107
20:5 n-3 0.220 ± 0.014 0.242 ± 0.014 0.239 ± 0.019 0.3802 0.5083
22:5 n-3 0.198 ± 0.029 0.239 ± 0.010 0.218 ± 0.024 0.5262 0.9345
24:5 n-3 0.030 ± 0.006 0.014 ± 0.004 0.015 ± 0.002 0.1317 0.0921
24:6 n-3 0.047 ± 0.014 0.031 ± 0.003 0.037 ± 0.005 0.6539 0.1318
22:6 n-3 0.838 ± 0.029 0.756 ± 0.016 0.627 ± 0.012 0.0001 0.0051
18:2 n-6 34.24 ± 0.567 35.17 ± 0.645 34.66 ± 0.959 0.7043 0.1292
18:3 n-6 1.57 ± 0.303 1.25 ± 0.027 1.19 ± 0.032 0.1390 0.2034
20:3 n-6 0.482 ± 0.042 0.451 ± 0.037 0.419 ± 0.039 0.2494 0.1957
20:4 n-6 2.58 ± 0.26 2.12 ± 0.047 1.89 ± 0.057 0.0099 0.0895
22:4 n-6 0.357 ± 0.040 0.251 ± 0.028 0.222 ± 0.012 0.0144 0.0203
24:4 n-6 0.013 ± 0.003 0.018 ± 0.007 0.011 ± 0.002 0.6008 0.6092
24:5 n-6 0.014 ± 0.003 0.012 ± 0.003 0.012 ± 0.003 0.6841 0.6846
22:5 n-6 0.180 ± 0.092 0.110 ± 0.001 0.093 ± 0.005 0.6145 0.5204
The genotype-specific fatty acid composition of hepatic triglyceride fraction from n = 3 mice per group a standard diet is displayed (mean ± SE). Total liver fat (triacylglyceride) content is displayed in the first lane; stearoyl-CoA-desaturase 1 activity indices as well as the de novo lipogenesis index are calculated and displayed in the lower lanes. P values are given for additive inheritance model. P values ≤ 0.05 were indicated as boldface text. GG, Gly/Gly; GR, Gly/Arg; RR, Arg/Arg.

View Large
The FGFR4 Gly385Arg polymorphism altered glycolysis and liver de novo lipogenesis without affecting insulin sensitivity or glucose tolerance in mice under a chow diet
To determine whether the altered hepatic lipid metabolism was associated with changes in glucose metabolism in the FGFR4 Arg385 mice, we examined plasma glucose and insulin levels. The FGFR4 Arg385 allele was not associated with circulating levels of blood glucose or insulin in the fed state (38) or in the fasted state (38), although there was a trend for higher insulin levels in mutant mice in the fed state (P for trend = 0.09). When subjected to a glucose-stimulated insulin release test by administration of 3 g of glucose per kilogram of body weight, FGFR4 Arg/Arg385 mice exhibited significantly elevated insulin levels after 2 minutes compared with control FGFR4 Gly/Gly385 mice (38).

In spite of unaltered glucose tolerance and insulin sensitivity (38), the FGFR4 Arg385 allele was associated with higher mRNA expression of hepatic glucokinase (Gck), indicating enhanced glycolysis (P for trend = 0.022) (Fig. 3A). Moreover, mRNA expression of the rate-limiting enzyme in lipogenesis, fatty acid synthase (Fasn), was also higher in carriers of the FGFR4 Arg385 allele (P for trend = 0.002) (Fig. 3B), whereas the mRNA expression of acetyl-coenzyme A (acyl-CoA) carboxylase 1 (Acc1) and sterol regulatory element binding transcription factor 1 (Srebp1) were unchanged (Fig. 3C and 3D, respectively). Because the enzyme acyl-CoA:diacylglycerol acyltransferase-2 (DGAT2), which catalyzes the final step of TAG synthesis, was found to be involved in the dissociation between fatty liver and insulin resistance in mice and in humans (39, 40), we also measured Dgat2 mRNA expression in the liver. In accordance with unchanged parameters of insulin resistance and glucose tolerance, the FGFR4 Arg385 allele was associated with elevated Dgat2 mRNA levels, in parallel with enhanced liver fat content in FGFR4 Arg385 mice (P for trend = 0.005) (Fig, 3E).

Figure 3.
Gene expression analysis of FGFR4 Gly/Gly385 mice (black bars), FGFR4 Gly/Arg385 mice (white bars), and FGFR4 Arg/Arg385 mice (gray bars) on a chow diet. (A) Glucokinase (n = 7 to 9). (B) Fasn (n = 7 to 9). (C) Acc1 (n = 6 to 9). (D) Srebp1 (n = 7 to 9). (E) Dgat2 (n = 7 to 9). Data are expressed as mean ± SEM.
View largeDownload slide
Gene expression analysis of FGFR4 Gly/Gly385 mice (black bars), FGFR4 Gly/Arg385 mice (white bars), and FGFR4 Arg/Arg385 mice (gray bars) on a chow diet. (A) Glucokinase (n = 7 to 9). (B) Fasn (n = 7 to 9). (C) Acc1 (n = 6 to 9). (D) Srebp1 (n = 7 to 9). (E) Dgat2 (n = 7 to 9). Data are expressed as mean ± SEM.


The FGFR4 Gly385Arg mutation was not associated with mRNA expression of the two major enzymes of hepatic gluconeogenesis, phosphoenolpyruvate-carboxykinase (Pepck) and glucose-6-phosphatase (G6Pase) (38).

Elevated liver fat content was not associated with liver cytokine gene expression or reduced insulin signaling in liver and skeletal muscles in FGFR4 Arg385 mice under a chow diet
In parallel with findings of unaltered glucose and insulin tolerance, analysis of the transcriptional regulation of the proinflammatory genes interleukin 6 (Il-6) and interleukin 1 beta (Il-1β) revealed no association in liver tissue with the FGFR4 Arg385 allele (38). Together with the concomitant finding that gene expression of the Il-6 mediator suppressor of cytokine signaling 3 (Socs3) was unaltered independent of the presence of the FGFR4 Arg385 allele, these results indicate that FGFR4 Arg385 mice did not suffer from liver inflammation (38).

In addition, to exclude the effect of the mutation on cellular insulin signal transduction, we evaluated the insulin-stimulated serine phosphorylation of Akt in liver and skeletal muscle. We found no significant differences in basal or insulin-stimulated serine phosphorylation of Akt in liver and skeletal muscle, indicating normal insulin signaling in the FGFR4 Arg385 mice (38).

Unaltered liver fat content as well as glucose and lipid metabolism in FGFR4 Arg385 mice under an HFD
To dissect whether the described phenotype could be aggravated with an HFD, we compared lipid and glucose metabolism of FGFR4 Arg/Arg385, FGFR4 Gly/Arg385, and FGFR4 Gly/Gly385 mice under HFD conditions. Body weight (38), fasting blood glucose levels (38), and glucose and insulin tolerance (38) were comparable between the mice independent of the presence or absence of the FGFR4 Arg385 allele. Interestingly, no significant differences were noted between the genotypes in regard to liver TAGs (38). Furthermore, unlike in mice under a chow diet, hepatic gene expression analysis revealed no association of the FGFR4 Arg385 allele with the mRNA expression of Fasn, Gck, or Dgat2 (38). The expression levels of Srebp1 and Acc1 were also comparable between the genotypes (38). Finally, an HFD resulted in elevated hepatic Fgfr4 and β-Klotho mRNA expression compared with a normal chow diet (38).

Discussion
The role of FGFR4 in the regulation of lipid metabolism was first suggested on the basis of observations in FGFR4-deficient mice (17). Despite showing features of the metabolic syndrome under normal dietary conditions, these mice were protected from HFD-induced fatty liver (11). Furthermore, in another study, improved insulin resistance and glucose metabolism were found in FGFR4-deficient mice; however, consistent with the previous study, protection from HFD-induced fatty liver was observed (41). In addition, targeting of FGF19, which signals through FGFR4, is currently being discussed as a tool to treat hepatic steatosis (19, 42).

In the current study, we focused on the human FGFR4 Gly388Arg (mice: Gly385Arg) polymorphism, as a gain-of-function effect of the FGFR4 Arg388 allele in humans has been demonstrated (25). The FGFR4 Arg388 allele is associated with enhanced STAT3 tyrosine phosphorylation (25), which in turn is involved in hepatic gluconeogenesis and hepatic lipogenesis (26, 27). Here, we specifically investigated whether this amino acid exchange influences hepatic lipid and glucose metabolism.

Performing studies with this amino acid substitution in both humans and mice, we found an association of the FGFR4 Arg388 allele (mice: FGFR4 Arg385) with elevated hepatic liver fat content under healthy dietary conditions. Interestingly, this mutation did not affect body fat mass or body fat distribution. Furthermore, we found that under normal dietary conditions, mice carrying the FGFR4 Arg385 allele had (i) altered liver de novo lipogenesis and hepatic lipid composition, indicating increased fatty acid synthesis, and (ii) enhanced glycolysis and insulin release, without changes in insulin sensitivity, glucose tolerance, or insulin signaling in liver and skeletal muscle. Furthermore, no metabolic alterations were found in these mice under an HFD compared with HFD-fed controls.

The elevated liver TAG accumulation in mice carrying the FGFR4 Arg385 allele under a chow diet is in agreement with our data in humans, where we found carriers of the minor 388Arg allele to have a lower decrease in liver fat content during an LI. An important function of the liver is quick adaptation to changes in nutrient availability, maintaining a well-controlled balance in the fluxes of carbohydrates and lipids between the circulation and itself. Ulaganathan et al. (25) recently found that the substitution of the Gly 388 residue to Arg in FGFR4 resulted in an increased tyrosine phosphorylation and activation of the transcription factor STAT3, which is known to regulate both carbohydrate metabolism in vivo by suppressing the expression of the gluconeogenic genes Pepck and G6Pase in the liver (26) and lipid metabolism, increasing plasma levels of TAGs and total cholesterol by enhanced expression of the key hepatic lipogenic enzymes Fasn and Acc (27). Although circulating plasma lipid levels were not associated with the FGFR4 Arg385 allele in our study, we found enhanced expression of Fasn in livers of these mice under normal chow conditions, demonstrating an elevated hepatic lipogenesis in carriers of the mutation. The alterations in lipogenesis were reported to be independent of the SREBP1 pathway (27), which was also the case in the current study. However, we could not detect changes in Il-6 gene expression or in its downstream target, Socs3, possibly because the transcriptional regulation of Il-6 in our mice was lower than that in previously studied mice with genetically forced expression of STAT3. Thus, together with the elevated transcription of Gck as a key glycolytic enzyme and unaltered fasting glucose levels and gluconeogenesis, our results indicate enhanced glucose flux toward lipogenesis in mice carrying the FGFR4 Arg385 allele, possibly to maintain stable plasma glucose levels. Certainly, other mechanisms by which the FGFR4 Arg385 allele may also regulate hepatic fat accumulation cannot be ruled out and remain to be clarified.

In addition, FGFR4 is known to be expressed in adult pancreatic islets (43, 44). Our data are in agreement with previous findings showing that the FGFR4 Arg385 allele promotes pancreatic insulin secretion in both mice and humans (24). Elevated insulin levels in FGFR4 Arg/Arg385 mice 2 minutes after the glucose load may contribute, in part, to the maintenance of normal glucose levels in these animals.

Nevertheless, in contrast to the animal data, in which no changes in insulin sensitivity were observed based on the presence or absence of the FGFR4 Arg385 allele, humans carrying the minor 388Arg allele displayed less improvement in whole-body and adipose tissue insulin resistance under an LI with a lower caloric load. So far, we have no explanation for this discrepancy between the human and animal data. It may well be that the FGFR4 Gly388Arg polymorphism has a larger effect on glucose metabolism in humans than in our mouse model. Furthermore, because our measurement of insulin resistance in humans is a function of high glucose and high insulin levels during the OGTT, a FGFR4 Gly388Arg‒mediated hypersecretion of insulin may result in a higher estimate of insulin resistance in humans. In addition, it should also be noted that the previously reported changes in insulin resistance and glucose intolerance in mice (7, 31) were found in FGFR4-deficient animals. This may have resulted from a more severe genetic modification than in our transgenic mice.

Interestingly, although the change in body fat mass was not statistically significant among the genotypes, subjects carrying the Gly/Gly alleles had the lowest decrease in body fat mass. One can only hypothesize about mechanisms explaining this difference. However, because insulin resistance in adipose tissue was also most strongly decreased in carriers of this genotype, improved suppression of insulin-mediated lipolysis in carriers of the Gly/Gly genotype may have resulted in less decrease of fat mass in these subjects.

Under conditions of overnutrition in mice and obesity before the intervention in humans, the FGFR4 Arg385(388) allele did not associate with liver fat content or insulin sensitivity. To further clarify this point, we first confirmed the finding of a previous report demonstrating marked upregulation of Fgfr4 expression in liver under HFD conditions (45). Together with established knowledge about the effect of FGFR4 signaling on liver fat content under both chow and HFD conditions, our data indicate that the FGFR4 Arg385 allele regulated hepatic lipogenesis predominantly under conditions of healthy nutrition. Thus, under caloric overload, the comparatively modest modulatory effect of the minor Arg385 allele seems to be overridden by forced FGFR4 expression and a large amount of liver fat accumulation.

In our mouse model, we found a lack of hepatic inflammation in FGFR4 Arg385 mice. The finding that transcriptional regulation of key hepatic proinflammatory genes, such as Il-6, Socs3, and Il-1β, was not altered in the presence of the Arg385 allele—together with normal insulin signaling—points to a metabolically healthy fatty liver (46–48). Regarding other mechanisms of a metabolically healthy fatty liver, further evidence for a genetically determined dissociation of hepatic steatosis from insulin resistance is provided by the single-nucleotide polymorphism (rs738409) in PNPLA3. The amino acid exchange Ile148Met causes the largest known effect on hepatic steatosis in humans, without insulin resistance or dyslipidemia (49–53), but a different quality of stored hepatic TAGs, with distinct alterations in fatty acid composition (35). Of note, when we analyzed the hepatic TAG profile of FGFR4 Arg385 mice under a chow diet, we detected somewhat similar changes, including a reduction in stearate (18:0) and an increase in linolenic acid (18:3), as well as a reduction in some major polyunsaturated fatty acids. This altered hepatic lipid composition may contribute to healthy metabolic effects of the stored hepatic lipids in FGFR4 Arg385 mice.

Another model of a metabolically healthy fatty liver is the liver-specific Dgat2 transgenic mouse (39). In accordance with these mouse data, we found that genetic variability in DGAT2 is also a mediator of the dissociation between fatty liver and insulin resistance in humans (40). There is evidence that in addition to liver inflammation, the mode of storing hepatic lipids plays a key role in the determination of metabolically healthy and unhealthy fatty liver. According to this hypothesis, the accumulation of hepatic fatty acyl-CoAs is detrimental for glucose metabolism, whereas TAG storage provides protection from their deleterious effects (54, 55). DGAT2 catalyzes the final step of TAG synthesis and is predominantly expressed in the liver (56). In our study, the FGFR4 Arg385 allele was associated with elevated Dgat2 expression in liver tissue under a chow diet, pointing to a possible mechanistic explanation for the metabolically benign TAG accumulation in these mice. Thus, our results underline a main regulatory function of hepatic FGFR4 in lipid homeostasis, without affecting systemic glucose homeostasis.

In conclusion, we provide evidence for an association between the FGFR4 Arg388 allele and elevated liver fat content in humans consuming a healthy diet, but not under nutritional overload. This finding was in agreement with our data in mice who carried the FGFR4 Arg385 allele. Given that hepatic lipid accumulation was not accompanied by metabolic disorders as a result of lipotoxicity in mice carrying the FGFR4 Arg385 allele under a healthy diet, hepatic lipid storage may lead to a metabolically healthy fatty liver, possibly because of an effective detoxification of fatty acyl-CoAs. Thus, FGFR4 may be a candidate gene for not only progression of several cancers but also hepatic steatosis in humans.

Abbreviations:

acyl-CoA
acetyl-coenzyme A


Arg
arginine


BMI
body mass index


DGAT2
diacylglycerol acyltransferase 2


Fasn
fatty acid synthase


FGFR4
fibroblast growth factor receptor 4


Gck
glucokinase


Gly
glycine


HFD
high-fat diet


IL-6
interleukin-6


i.p.
intraperitoneal


LI
lifestyle intervention


NAFLD
non-alcoholic fatty liver disease


NEFA
non-esterified fatty acid


OGTT
oral glucose tolerance test


Socs3
suppressor of cytokine signaling-3


Srebp1
sterol regulatory element-binding protein 1


STAT3
signal transducer and activator of transcription-3


TAG
triacylglyceride

Acknowledgments
The authors thank all study participants for their cooperation. We gratefully acknowledge the excellent assistance of Alke Guirguis, Dorothee Neuscheler, and Anja Dessecker.

Financial Support: This study was supported in part by a grant from the German Federal Ministry of Education and Research to the German Center for Diabetes Research (to H.-U.H.).

Author Contributions: S.Z.L. and N.S. designed the study, performed the analyses, and drafted the manuscript. M.K., J.M., S.M.K., and B.S. performed laboratory measurements and contributed to the discussion. C.T. was involved in the collection of the human data and contributed to the discussion. A.M.H., A.P., E.S., F.S., M.H., and A.U. contributed to the discussion. H.-U.H. and N.S. supervised the project and contributed to the discussion. All authors approved the final version of the manuscript prior to submission. H.-U.H. is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Disclosure Summary: The authors have nothing to disclose.

References
1. Byrne CD, Targher G. NAFLD: a multisystem disease. J Hepatol . 2015;62(1, Suppl):S47–S64.
Google ScholarCrossrefPubMed
2. Gastaldelli A. Insulin resistance and reduced metabolic flexibility: cause or consequence of NAFLD? Clin Sci (Lond) . 2017;131(22):2701–2704.
3. Stefan N, Fritsche A, Schick F, Häring HU. Phenotypes of prediabetes and stratification of cardiometabolic risk. Lancet Diabetes Endocrinol . 2016;4(9):789–798.
Google ScholarCrossrefPubMed
4. Eslam M, Valenti L, Romeo S. Genetics and epigenetics of NAFLD and NASH: clinical impact. J Hepatol . 2018;68(2):268–279.
Google ScholarCrossrefPubMed
5. Anstee QM, Targher G, Day CP. Progression of NAFLD to diabetes mellitus, cardiovascular disease or cirrhosis. Nat Rev Gastroenterol Hepatol . 2013;10(6):330–344.
Google ScholarCrossrefPubMed
6. Bril F, Cusi K. Management of nonalcoholic fatty liver disease in patients with type 2 diabetes: a call to action. Diabetes Care . 2017;40(3):419–430.
Google ScholarCrossrefPubMed
7. Chalasani N, Younossi Z, Lavine JE, Diehl AM, Brunt EM, Cusi K, Charlton M, Sanyal AJ; American Gastroenterological Association; American Association for the Study of Liver Diseases; American College of Gastroenterology. The diagnosis and management of non-alcoholic fatty liver disease: practice guideline by the American Gastroenterological Association, American Association for the Study of Liver Diseases, and American College of Gastroenterology. Gastroenterology . 2012;142(7):1592–1609.
Google ScholarCrossrefPubMed
8. Thamer C, Machann J, Stefan N, Schäfer SA, Machicao F, Staiger H, Laakso M, Böttcher M, Claussen C, Schick F, Fritsche A, Haring HU. Variations in PPARD determine the change in body composition during lifestyle intervention: a whole-body magnetic resonance study. J Clin Endocrinol Metab . 2008;93(4):1497–1500.
Google ScholarCrossrefPubMed
9. Kantartzis K, Thamer C, Peter A, Machann J, Schick F, Schraml C, Königsrainer A, Königsrainer I, Kröber S, Niess A, Fritsche A, Häring HU, Stefan N. High cardiorespiratory fitness is an independent predictor of the reduction in liver fat during a lifestyle intervention in non-alcoholic fatty liver disease. Gut . 2009;58(9):1281–1288.
Google ScholarCrossrefPubMed
10. Stefan N, Machicao F, Staiger H, Machann J, Schick F, Tschritter O, Spieth C, Weigert C, Fritsche A, Stumvoll M, Häring HU. Polymorphisms in the gene encoding adiponectin receptor 1 are associated with insulin resistance and high liver fat. Diabetologia . 2005;48(11):2282–2291.
Google ScholarCrossrefPubMed
11. Huang X, Yang C, Luo Y, Jin C, Wang F, McKeehan WL. FGFR4 prevents hyperlipidemia and insulin resistance but underlies high-fat diet induced fatty liver. Diabetes . 2007;56(10):2501–2510.
Google ScholarCrossrefPubMed
12. Eswarakumar VP, Lax I, Schlessinger J. Cellular signaling by fibroblast growth factor receptors. Cytokine Growth Factor Rev . 2005;16(2):139–149.
Google ScholarCrossrefPubMed
13. Turner N, Grose R. Fibroblast growth factor signalling: from development to cancer. Nat Rev Cancer . 2010;10(2):116–129.
Google ScholarCrossrefPubMed
14. Schlessinger J, Ullrich A. Growth factor signaling by receptor tyrosine kinases. Neuron . 1992;9(3):383–391.
Google ScholarCrossrefPubMed
15. Schlessinger J, Plotnikov AN, Ibrahimi OA, Eliseenkova AV, Yeh BK, Yayon A, Linhardt RJ, Mohammadi M. Crystal structure of a ternary FGF-FGFR-heparin complex reveals a dual role for heparin in FGFR binding and dimerization. Mol Cell . 2000;6(3):743–750.
Google ScholarCrossrefPubMed
16. Beenken A, Mohammadi M. The structural biology of the FGF19 subfamily. Adv Exp Med Biol . 2012;728:1–24.
Google ScholarCrossrefPubMed
17. Yu C, Wang F, Kan M, Jin C, Jones RB, Weinstein M, Deng CX, McKeehan WL. Elevated cholesterol metabolism and bile acid synthesis in mice lacking membrane tyrosine kinase receptor FGFR4. J Biol Chem . 2000;275(20):15482–15489.
Google ScholarCrossrefPubMed
18. Marra F, Svegliati-Baroni G. Lipotoxicity and the gut-liver axis in NASH pathogenesis. J Hepatol . 2018;68(2):280–295.
Google ScholarCrossrefPubMed
19. Harrison SA, Rinella ME, Abdelmalek MF, Trotter JF, Paredes AH, Arnold HL, Kugelmas M, Bashir MR, Jaros MJ, Ling L, Rossi SJ, DePaoli AM, Loomba R. NGM282 for treatment of non-alcoholic steatohepatitis: a multicentre, randomised, double-blind, placebo-controlled, phase 2 trial. Lancet . 2018;391(10126):1174–1185.
Google ScholarCrossrefPubMed
20. Bange J, Prechtl D, Cheburkin Y, Specht K, Harbeck N, Schmitt M, Knyazeva T, Müller S, Gärtner S, Sures I, Wang H, Imyanitov E, Häring HU, Knayzev P, Iacobelli S, Höfler H, Ullrich A. Cancer progression and tumor cell motility are associated with the FGFR4 Arg(388) allele. Cancer Res . 2002;62(3):840–847.
Google ScholarPubMed
21. Thussbas C, Nahrig J, Streit S, Bange J, Kriner M, Kates R, Ulm K, Kiechle M, Hoefler H, Ullrich A, Harbeck N. FGFR4 Arg388 allele is associated with resistance to adjuvant therapy in primary breast cancer. J Clin Oncol . 2006;24(23):3747–3755.
Google ScholarCrossrefPubMed
22. Wang J, Stockton DW, Ittmann M. The fibroblast growth factor receptor-4 Arg388 allele is associated with prostate cancer initiation and progression. Clin Cancer Res. 2004;10(18 Pt 1):6169–6178.
23. Abecasis GR, Auton A, Brooks LD, DePristo MA, Durbin RM, Handsaker RE, Kang HM, Marth GT, McVean GA; 1000 Genomes Project Consortium. An integrated map of genetic variation from 1,092 human genomes. Nature . 2012;491(7422):56–65.
Google ScholarCrossrefPubMed
24. Ezzat S, Zheng L, Florez JC, Stefan N, Mayr T, Hliang MM, Jablonski K, Harden M, Stančáková A, Laakso M, Haring HU, Ullrich A, Asa SL. The cancer-associated FGFR4-G388R polymorphism enhances pancreatic insulin secretion and modifies the risk of diabetes. Cell Metab . 2013;17(6):929–940.
Google ScholarCrossrefPubMed
25. Ulaganathan VK, Sperl B, Rapp UR, Ullrich A. Germline variant FGFR4 p.G388R exposes a membrane-proximal STAT3 binding site. Nature . 2015;528(7583):570–574.
Google ScholarCrossrefPubMed
26. Inoue H, Ogawa W, Ozaki M, Haga S, Matsumoto M, Furukawa K, Hashimoto N, Kido Y, Mori T, Sakaue H, Teshigawara K, Jin S, Iguchi H, Hiramatsu R, LeRoith D, Takeda K, Akira S, Kasuga M. Role of STAT-3 in regulation of hepatic gluconeogenic genes and carbohydrate metabolism in vivo. Nat Med . 2004;10(2):168–174.
Google ScholarCrossrefPubMed
27. Kinoshita S, Ogawa W, Okamoto Y, Takashima M, Inoue H, Matsuki Y, Watanabe E, Hiramatsu R, Kasuga M. Role of hepatic STAT3 in the regulation of lipid metabolism. Kobe J Med Sci . 2008;54(4):E200–E208.
Google ScholarPubMed
28. Lehmann R, Franken H, Dammeier S, Rosenbaum L, Kantartzis K, Peter A, Zell A, Adam P, Li J, Xu G, Königsrainer A, Machann J, Schick F, Hrabé de Angelis M, Schwab M, Staiger H, Schleicher E, Gastaldelli A, Fritsche A, Häring HU, Stefan N. Circulating lysophosphatidylcholines are markers of a metabolically benign nonalcoholic fatty liver. Diabetes Care . 2013;36(8):2331–2338.
Google ScholarCrossrefPubMed
29. Stefan N, Ramsauer M, Jordan P, Nowotny B, Kantartzis K, Machann J, Hwang JH, Nowotny P, Kahl S, Harreiter J, Hornemann S, Sanyal AJ, Stewart PM, Pfeiffer AF, Kautzky-Willer A, Roden M, Häring HU, Fürst-Recktenwald S. Inhibition of 11β-HSD1 with RO5093151 for non-alcoholic fatty liver disease: a multicentre, randomised, double-blind, placebo-controlled trial. Lancet Diabetes Endocrinol . 2014;2(5):406–416.
Google ScholarCrossrefPubMed
30. Matsuda M, DeFronzo RA. Insulin sensitivity indices obtained from oral glucose tolerance testing: comparison with the euglycemic insulin clamp. Diabetes Care . 1999;22(9):1462–1470.
Google ScholarCrossrefPubMed
31. Lomonaco R, Ortiz-Lopez C, Orsak B, Webb A, Hardies J, Darland C, Finch J, Gastaldelli A, Harrison S, Tio F, Cusi K. Effect of adipose tissue insulin resistance on metabolic parameters and liver histology in obese patients with nonalcoholic fatty liver disease. Hepatology . 2012;55(5):1389–1397.
Google ScholarCrossrefPubMed
32. Gastaldelli A, Harrison SA, Belfort-Aguilar R, Hardies LJ, Balas B, Schenker S, Cusi K. Importance of changes in adipose tissue insulin resistance to histological response during thiazolidinedione treatment of patients with nonalcoholic steatohepatitis. Hepatology . 2009;50(4):1087–1093.
Google ScholarCrossrefPubMed
33. Seitzer N, Mayr T, Streit S, Ullrich A. A single nucleotide change in the mouse genome accelerates breast cancer progression. Cancer Res . 2010;70(2):802–812.
Google ScholarCrossrefPubMed
34. RRID:AB_329825, https://scicrunch.org/resolver/AB_329825.
35. RRID:AB_561053, https://scicrunch.org/resolver/AB_561053.
36. Machann J, Thamer C, Schnoedt B, Haap M, Haring HU, Claussen CD, Stumvoll M, Fritsche A, Schick F. Standardized assessment of whole body adipose tissue topography by MRI. J Magn Reson Imaging . 2005;21(4):455–462.
Google ScholarCrossrefPubMed
37. Peter A, Kovarova M, Nadalin S, Cermak T, Königsrainer A, Machicao F, Stefan N, Häring HU, Schleicher E. PNPLA3 variant I148M is associated with altered hepatic lipid composition in humans. Diabetologia . 2014;57(10):2103–2107.
Google ScholarCrossrefPubMed
38. Lutz SZ, Hennige AM, Peter A, Kovarova M, Totsikas C, Machann J, Kröber SM, Sperl B, Schleicher E, Schick F, Heni M, Ullrich A, Häring H-U, Stefan N. Data from: The Gly385(388)Arg polymorphism of the FGFR4 receptor regulates hepatic lipogenesis under healthy diet. figshare 2018. Deposited 3 September 2018. https://figshare.com/articles/Supplemental_material_pptx/7039472.
39. Monetti M, Levin MC, Watt MJ, Sajan MP, Marmor S, Hubbard BK, Stevens RD, Bain JR, Newgard CB, Farese RV Sr, Hevener AL, Farese RV Jr. Dissociation of hepatic steatosis and insulin resistance in mice overexpressing DGAT in the liver. Cell Metab . 2007;6(1):69–78.
Google ScholarCrossrefPubMed
40. Kantartzis K, Machicao F, Machann J, Schick F, Fritsche A, Häring HU, Stefan N. The DGAT2 gene is a candidate for the dissociation between fatty liver and insulin resistance in humans. Clin Sci (Lond). 2009;116(6):531–537.
41. Ge H, Zhang J, Gong Y, Gupte J, Ye J, Weiszmann J, Samayoa K, Coberly S, Gardner J, Wang H, Corbin T, Chui D, Baribault H, Li Y. Fibroblast growth factor receptor 4 (FGFR4) deficiency improves insulin resistance and glucose metabolism under diet-induced obesity conditions. J Biol Chem . 2014;289(44):30470–30480.
Google ScholarCrossrefPubMed
42. Chen Q, Jiang Y, An Y, Zhao N, Zhao Y, Yu C. Soluble FGFR4 extracellular domain inhibits FGF19-induced activation of FGFR4 signaling and prevents nonalcoholic fatty liver disease. Biochem Biophys Res Commun . 2011;409(4):651–656.
Google ScholarCrossrefPubMed
43. Le Bras S, Miralles F, Basmaciogullari A, Czernichow P, Scharfmann R. Fibroblast growth factor 2 promotes pancreatic epithelial cell proliferation via functional fibroblast growth factor receptors during embryonic life. Diabetes . 1998;47(8):1236–1242.
Google ScholarCrossrefPubMed
44. Dichmann DS, Miller CP, Jensen J, Scott Heller R, Serup P. Expression and misexpression of members of the FGF and TGFβ families of growth factors in the developing mouse pancreas. Dev Dyn . 2003;226(4):663–674.
Google ScholarCrossrefPubMed
45. Sun B, Yang G, Yang M, Liu H, Boden G, Li L. Long-term high-fat diet links the regulation of the insulin-sensitizing fibroblast growth factor-21 and visfatin. Cytokine . 2012;59(1):131–137.
Google ScholarCrossrefPubMed
46. Stefan N, Häring HU. The role of hepatokines in metabolism. Nat Rev Endocrinol . 2013;9(3):144–152.
Google ScholarCrossrefPubMed
47. Iroz A, Couty JP, Postic C. Hepatokines: unlocking the multi-organ network in metabolic diseases. Diabetologia . 2015;58(8):1699–1703.
Google ScholarCrossrefPubMed
48. Sun Z, Lazar MA. Dissociating fatty liver and diabetes. Trends Endocrinol Metab . 2013;24(1):4–12.
Google ScholarCrossrefPubMed
49. Romeo S, Kozlitina J, Xing C, Pertsemlidis A, Cox D, Pennacchio LA, Boerwinkle E, Cohen JC, Hobbs HH. Genetic variation in PNPLA3 confers susceptibility to nonalcoholic fatty liver disease. Nat Genet . 2008;40(12):1461–1465.
Google ScholarCrossrefPubMed
50. Kantartzis K, Peter A, Machicao F, Machann J, Wagner S, Königsrainer I, Königsrainer A, Schick F, Fritsche A, Häring HU, Stefan N. Dissociation between fatty liver and insulin resistance in humans carrying a variant of the patatin-like phospholipase 3 gene. Diabetes . 2009;58(11):2616–2623.
Google ScholarCrossrefPubMed
51. Kotronen A, Johansson LE, Johansson LM, Roos C, Westerbacka J, Hamsten A, Bergholm R, Arkkila P, Arola J, Kiviluoto T, Fisher RM, Ehrenborg E, Orho-Melander M, Ridderstråle M, Groop L, Yki-Järvinen H. A common variant in PNPLA3, which encodes adiponutrin, is associated with liver fat content in humans. Diabetologia . 2009;52(6):1056–1060.
Google ScholarCrossrefPubMed
52. Kovarova M, Königsrainer I, Königsrainer A, Machicao F, Häring HU, Schleicher E, Peter A. The genetic variant I148M in PNPLA3 is associated with increased hepatic retinyl-palmitate storage in humans. J Clin Endocrinol Metab . 2015;100(12):E1568–E1574.
Google ScholarCrossrefPubMed
53. Li JZ, Huang Y, Karaman R, Ivanova PT, Brown HA, Roddy T, Castro-Perez J, Cohen JC, Hobbs HH. Chronic overexpression of PNPLA3I148M in mouse liver causes hepatic steatosis. J Clin Invest . 2012;122(11):4130–4144.
Google ScholarCrossrefPubMed
54. Jou J, Choi SS, Diehl AM. Mechanisms of disease progression in nonalcoholic fatty liver disease. Semin Liver Dis . 2008;28(4):370–379.
Google ScholarCrossrefPubMed
55. Haemmerle G, Lass A, Zimmermann R, Gorkiewicz G, Meyer C, Rozman J, Heldmaier G, Maier R, Theussl C, Eder S, Kratky D, Wagner EF, Klingenspor M, Hoefler G, Zechner R. Defective lipolysis and altered energy metabolism in mice lacking adipose triglyceride lipase. Science . 2006;312(5774):734–737.
Google ScholarCrossrefPubMed
56. Stone SJ, Myers HM, Watkins SM, Brown BE, Feingold KR, Elias PM, Farese RV Jr. Lipopenia and skin barrier abnormalities in DGAT2-deficient mice. J Biol Chem . 2004;279(12):11767–11776.
Google ScholarCrossrefPubMed
Copyright © 2019 Endocrine Society