HIF Prolyl 4-Hydroxylase-2 Inhibition Improves Glucose and Lipid Metabolism and Protects Against Obesity and Metabolic Dysfunction

Hypoxia-inducible factor (HIF) regulates the expression of numerous hypoxia-regulated genes (1–3). The HIF-α subunit isoforms HIF-1α and HIF-2α are synthesized constitutively, and hydroxylation of two critical prolines generates 4-hydroxyproline residues that target HIF-α for degradation in normoxia. In hypoxia, this hydroxylation is inhibited so that HIF-α evades degradation and forms a functional dimer with HIF-β (1–3).

The hydroxylation of HIF-α is catalyzed by HIF prolyl 4-hydroxylase (P4H) isoenzymes 1–3 (also known as PHDs 1–3 and EglNs 2, 1, and 3) (1–6) and a transmembrane, P4H-TM (3,7), with HIF-P4H-2 being the main oxygen sensor in the HIF pathway (1–3). Hif-p4h-2 null mice die during embryonic development, whereas Hif-p4h-1 and Hif-p4h-3 null mice are viable (8). Broad-spectrum conditional Hif-p4h-2 inactivation leads to severe erythrocytosis, hyperactive angiogenesis, and dilated cardiomyopathy (3,9,10). We have generated Hif-p4h-2 hypomorphic mice (Hif-p4h-2^gt/gt) that express decreased amounts of wild-type Hif-p4h-2 mRNA and show stabilization of Hif-αs (11). These mice appear healthy and have a normal life span. They have no increased erythrocytosis and show no signs of hyperactive angiogenesis or dilated cardiomyopathy; instead, they are protected against myocardial infarction and ischemia-reperfusion injury (11,12).

Many studies have demonstrated that hypoxia reduces body weight (13–15). The only obvious abnormality found in the Hif-p4h-2^gt/gt mice was that their weights were 85–90% of those of the wild type (11). We focused initially on this difference and found that these mice have less adipose tissue than their littermates. This finding prompted us to study their lipid and glucose metabolism, especially because obesity is a major public health problem and increases the risk of metabolic syndrome, type 2 diabetes, and cardiovascular diseases. The data show that the Hif-p4h-2^gt/gt mice, whether fed normal chow or a high-fat diet (HFD), have major alterations in their adipose tissues and metabolism, including improved glucose tolerance and insulin sensitivity, reduced serum cholesterol levels, and protection against hepatic steatosis.

Small-molecule HIF-P4H inhibitors have been developed for the treatment of anemias and ischemic diseases, for example (3). Our Hif-p4h-2^gt/gt mouse data suggest that pharmacological HIF-P4H-2 inhibition could also be beneficial for the treatment of obesity and metabolic syndrome. We therefore administered FG-4497 to wild-type mice with a metabolic dysfunction, which stabilizes HIF-α in cultured cells and in vivo and increases erythropoiesis in animals with no apparent toxicity (3,16,17). The data demonstrate that FG-4497 administration phenocopied the beneficial effects of genetic HIF-P4H-2 deficiency on adipose tissues and metabolism.

Generation of C57BL/6 Hif-p4h-2^gt/gt mice has been described (11). All experiments were performed according to protocols approved by the Provincial State Office of Southern Finland. All data obtained with the Hif-p4h-2^gt/gt mice were compared with those of their wild-type littermates. The mice were fed a standard rodent diet or HFD (18% and 42% kcal fat, respectively) (Teklad T.2018C.12 and TD88137; Harlan Laboratories). FG-4497 was dissolved in 0.5% sodium carboxymethyl cellulose (Spectrum) and 0.1% polysorbate 80 (Fluka), and the solvent was also used as a vehicle. Both were administered orally.

Five-micrometer sections of formaldehyde-fixed paraffin-embedded tissue samples were stained with hematoxylin-eosin (H-E) and viewed and photographed with a Leica DM LB2 microscope and Leica DFC320 camera or Nikon Eclipse 50i microscope and DS-5M-L2 camera. Representative pictures (five to eight per mouse) were taken, and areas of 100 adipocytes were quantified with Nikon NIS-Elements BR 2.30 imaging software. Macrophage infiltration was analyzed by an anti-CD68 antibody (ab955; Abcam) and EnVision Detection System (Dako). The number of macrophage aggregates was calculated from five to eight fields per sample. Hepatic steatosis was scored (0 to ++++) from H-E–stained sections.

NE-PER extraction reagents (Thermo Fisher Scientific) were used to prepare nuclear fractions. Samples of 30–100 μg were resolved by SDS-PAGE, blotted, and probed with the following primary antibodies: Hif-1α (NB100-479; Novus Biologicals), Hif-2α (ab199; Abcam), and β-actin (NB600-501; Novus Biologicals).

Total RNA from tissues was isolated with E.Z.N.A. Total RNA Kit II (Omega Bio-Tek) or TriPure Isolation Reagent (Roche Applied Science) and reverse transcripted with an iScript cDNA Synthesis Kit (Bio-Rad). Quantitative PCR was performed with iTaq SYBR Green Supermix with ROX (Bio-Rad) in a Stratagene Mx3005 thermocycler or C1000 Touch Thermal Cycler and CFX96 Touch Real-Time PCR Detection System (Bio-Rad) with the primers shown in Supplementary Table 1.

Tissue fat content was determined using the 3-point Dixon method and a 3-T clinical scanner (Skyra; Siemens, Erlangen, the Netherlands). The sequence used was T1 turbo spin echo Dixon (field of view 320 × 320; repetition time 720; echo time 20, 24 slices). The torso of the animal was scanned, and the region of interest was drawn to separate the regional volume of subcutaneous fat for possible correlation with other variables.

Glucose tolerance test (GTT) and insulin tolerance test (ITT) were performed on mice fasted for 6–12 h and anesthetized with fentanyl/fluanisone and midazolam. For GTT, mice were injected intraperitoneally with 1 mg/g glucose, and blood glucose concentrations were monitored with a glucometer. Serum insulin values were determined with Rat/Mouse Insulin ELISA kit (EZRMI-13K; Millipore), and HOMA-IR scores were calculated from the glucose and insulin values. For ITT, mice were injected intraperitoneally with 1 IU/kg insulin (Humulin Regular; Lilly), and blood glucose concentrations were determined as for GTT. For deoxyglucose uptake test, mice were injected intraperitoneally with 0.6 μCi/g ¹⁴C-deoxyglucose (PerkinElmer) combined with 1 mg/g glucose and killed 60 min later. Fifty-microgram pieces of tissues and 50-μL blood samples were homogenized in 1:1 (or in the case of white adipose tissue [WAT], 2:1) chloroform:methanol and centrifuged. The pellet was re-extracted, and the supernatants were combined and scintillated for ¹⁴C activity.

Serum total cholesterol, HDL cholesterol, and triglyceride levels were determined by an enzymatic method (Roche Diagnostics). The Friedewald equation (18) was used to calculate LDL + VLDL cholesterol concentrations.

Snap-frozen tissues were homogenized in precooled 8% perchloric acid in 40% (volume for volume) ethanol and centrifuged. The supernatant was neutralized, and the amount of acetyl-CoA was measured with a PicoProbe Acetyl CoA Assay Kit (ab87546; Abcam).

Mice were terminally anesthetized with fentanyl/fluanisone and midazolam and ∼30-mg pieces of WAT and liver were excised and incubated for 90 min at 37°C with 40 μCi [¹⁴C]acetate (PerkinElmer) in DMEM pregassed with 95% O₂ and 5% CO₂. Lipids were extracted after saponification and scintillated for ¹⁴C activity (19).

Student two-tailed t test was used for comparisons between groups, and Student t test for paired samples was used for pairwise testing. Fisher exact test was used to calculate significance for the difference between genotypes in liver steatosis analyses. Areas under the curve were calculated by the summary measures method. All data are mean ± SEM. P < 0.05 was considered statistically significant.

Hif-p4h-2^gt/gt mice have their Hif-p4h-2 gene disrupted by a gene trap insertion cassette. Small amounts of wild-type Hif-p4h-2 mRNA were nevertheless generated from the gene-trapped alleles (11), its smallest relative amounts in the Hif-p4h-2^gt/gt tissues studied being found in the heart and skeletal muscle, whereas the highest proportion was found in the liver (11). We analyzed these values in the skeletal muscle and liver of the mice in the present study and found that the relative proportions of wild-type Hif-p4h-2 mRNA in these Hif-p4h-2^gt/gt tissues were ∼25% and ∼60%, respectively (Supplementary Fig. 1A). We found previously that the amount of Hif-p4h-2 protein in Hif-p4h-2^gt/gt skeletal muscle, heart, and kidney corresponds to that of the wild-type mRNA (11). Hif-1α was well stabilized in the heart and weakly stabilized in the skeletal muscle and kidney of the Hif-p4h-2^gt/gt mice, whereas Hif-2α was weakly stabilized in the heart and skeletal muscle (11). We also found a weak stabilization of Hif-2α, but not of Hif-1α, in the Hif-p4h-2^gt/gt liver with a more sensitive antibody (Fig. 1A).

The weight of 5-week-old Hif-p4h-2^gt/gt mice fed normal chow (18% kcal fat) was lower than that of their wild-type littermates, and the Hif-p4h-2^gt/gt mice gained less weight during the subsequent 10 weeks (Fig. 1B). The weight difference persisted later, with the weight of 1-year-old Hif-p4h-2^gt/gt mice being 80–85% of that of wild type (Fig. 1B). No difference was found between the genotypes in tibia length or liver weight relative to tibia length (Supplementary Fig. 1B).

No difference was found in the amount of food intake between the Hif-p4h-2^gt/gt and wild-type mice in metabolic home cage analyses whether expressed per mouse or body weight, and no difference was found in the physical activity between these mice (Supplementary Fig. 2). There was a trend for increased O₂ consumption and CO₂ production when expressed per body weight in the Hif-p4h-2^gt/gt mice, but this reached statistical significance only in the case of O₂ consumption in the dark, with no difference in respiratory exchange ratio between the genotypes (Supplementary Fig. 2).

The amount of gonadal WAT in the Hif-p4h-2^gt/gt mice, expressed relative to tibia length to correct for body weight differences, was one-half of that in the wild-type littermates (Fig. 1C). Furthermore, the adipocytes in the Hif-p4h-2^gt/gt WAT were smaller (Fig. 1D) and the amount of wild-type Hif-p4h-2 mRNA in the Hif-p4h-2^gt/gt WAT was ∼50% of that in wild-type tissue (Supplementary Fig. 1A). The amount of Hif-p4h-2 protein in the Hif-p4h-2^gt/gt WAT was correspondingly reduced (Supplementary Fig. 1C), and its Hif-1α but not Hif-2α was stabilized to a low extent (Fig. 1A). The weight of the brown adipose tissue (BAT) was likewise reduced (Fig. 1E), and the wild-type Hif-p4h-2 mRNA level in the Hif-p4h-2^gt/gt BAT was only ∼10% of that in the wild-type tissue (Supplementary Fig. 1A). No difference in the mRNA level of uncoupling protein 1 (Ucp1) was found between the Hif-p4h-2^gt/gt and wild-type BAT (Supplementary Fig. 1D). MRI analyses of live anesthetized mice indicated that the amount of subcutaneous fat was also reduced in the Hif-p4h-2^gt/gt mice (Fig. 1F).

Obesity is associated with chronic adipose tissue inflammation and formation of macrophage aggregates around adipocytes (20,21). Chronic 21-day hypoxia decreases the number of such aggregates in 1-year-old mice (15). We found in the present study that their number is significantly lower in the WAT of 1-year-old Hif-p4h-2^gt/gt mice than in wild-type mice (Fig. 1G). As in the mice subjected to chronic hypoxia (15), increased adiponectin mRNA level and decreased leptin and chemochine (C-C motif) ligand 2 mRNA levels were found in the Hif-p4h-2^gt/gt WAT (Supplementary Fig. 3A), whereas no increase was seen in serum adiponectin level, but serum leptin level was decreased (Supplementary Fig. 3B).

The levels of serum total cholesterol and HDL and LDL + VLDL cholesterol were lower in the Hif-p4h-2^gt/gt mice than in the wild-type mice, with the HDL/LDL + VLDL ratio being increased, whereas the triglyceride level was not altered significantly (Fig. 2A). Livers of 1-year-old wild-type mice had steatosis, even though these mice were not fed an HFD, whereas the Hif-p4h-2^gt/gt mice were protected against its development (Fig. 2B).

The glucose tolerance of 1-year-old Hif-p4h-2^gt/gt mice was markedly better than that of their littermates, and even the fasting blood glucose levels (at 0 min) were significantly lower than in the wild type (Fig. 3A and Supplementary Fig. 4A). The fasting serum insulin levels and HOMA-IR scores were likewise lower, indicating that the lower blood glucose levels were due to increased insulin sensitivity (Fig. 3B and Supplementary Fig. 4B). A similar, though less marked, difference in glucose tolerance was seen between the genotypes in 4–5-month-old mice (Fig. 3A). Correspondingly, 3–4-month-old Hif-p4h-2^gt/gt mice had lower blood glucose levels than wild-type mice in an ITT (Fig. 3C). To learn which Hif-p4h-2^gt/gt tissues were responsible for the increased glucose intake, we studied the uptake of ¹⁴C-deoxyglucose in fasting mice and found an increased uptake relative to the wild type in the skeletal muscle (Fig. 3D).

A comparison of data between 1-year-old and 4–5-month-old wild-type mice indicated that the former had larger adipocytes, a larger number of adipose tissue macrophage aggregates, increased fasting blood glucose and serum insulin levels, and higher HOMA-IR scores, whereas none of these were significantly different between 1-year-old and 4–5-month-old Hif-p4h-2^gt/gt mice (Fig. 4A–E). The 1-year-old wild-type mice thus had relative insulin resistance and metabolic dysfunction compared with the younger mice, whereas the Hif-p4h-2^gt/gt mice were protected against their development.

The mRNA levels of the HIF-1α targets Glut1 and several enzymes of glycolysis (22) were increased in the Hif-p4h-2^gt/gt skeletal muscle and WAT (Fig. 5A and B) but not in Hif-p4h-2^gt/gt liver (Supplementary Fig. 5A). The mRNA level of the glucose-regulated Glut2 (23) was lower in the liver of the Hif-p4h-2^gt/gt mice than in wild-type mice, presumably because of their improved glucose tolerance, whereas the mRNA level of the insulin-regulated Glut4 (24) was higher in the skeletal muscle and WAT of the Hif-p4h-2^gt/gt mice, probably because of their increased insulin sensitivity (Fig. 5A and B). The mRNA level of the HIF-1α target Pdk1, which inhibits pyruvate dehydrogenase activity (22), was increased in the Hif-p4h-2^gt/gt skeletal muscle and WAT (Fig. 5A). The Pparγ mRNA level was likewise increased in the Hif-p4h-2^gt/gt skeletal muscle and WAT (Fig. 5A), this change being similar to that seen in the WAT of mice with adipocyte-specific Hif-p4h-2 deletion (25), whereas the Pparα mRNA level was slightly decreased in the Hif-p4h-2^gt/gt liver (Fig. 5B). The mRNA levels of the lipolysis markers Lipe and Pnpla2 were increased in the Hif-p4h-2^gt/gt WAT (Fig. 5A), suggesting that an increased lipolysis may have contributed to the decreased amount of WAT in these mice. The mRNA levels of Srebp1c, which regulates lipogenesis and fatty acid synthesis, and its targets Accα and Fas, enzymes of fatty acid synthesis, were lower in the Hif-p4h-2^gt/gt liver, whereas the mRNA level of the Ldl receptor was similar in the Hif-p4h-2^gt/gt and wild-type livers (Fig. 5B). The mRNA level of the HIF-2α target Irs2 (26), which regulates Srepb1c and hepatic lipid accumulation (27), was increased in the Hif-p4h-2^gt/gt liver, whereas the mRNA level of Irs1 was not altered (Fig. 5B). To study whether the increased mRNA levels led to increased protein levels, we analyzed Glut4, Gadph, and Pdk1 in WAT by Western blotting and found increased levels of all three proteins in the Hif-p4h-2^gt/gt WAT (Supplementary Fig. 5B).

To study whether the presumed decreased conversion of pyruvate to acetyl-CoA as a result of an increased Pdk1 expression actually decreased the amount of acetyl-CoA, we measured its amount in skeletal muscle, WAT, and liver and found a decreased concentration in all three Hif-p4h-2^gt/gt tissues (Fig. 5C). We also studied whether the decreased Accα and Fas mRNA and acetyl-CoA levels decreased de novo lipogenesis by incubating fresh tissue slices with [¹⁴C]acetate and measuring the incorporation of radioactivity into extractable lipids and found decreased lipogenesis in the Hif-p4h-2^gt/gt WAT (P = 0.03) and liver (P = 0.06) (Fig. 5D).

To study whether Hif-p4h-2^gt/gt mice are protected against obesity-induced changes in glucose metabolism, 6-month-old mice were fed an HFD (42% kcal fat) for 6 weeks. The weight gain of the Hif-p4h-2^gt/gt and wild-type mice during the HFD treatment was similar, the Hif-p4h-2^gt/gt mice thus remaining lighter than their littermates (Fig. 6A). The adipocytes were smaller and the number of macrophage aggregates lower in the Hif-p4h-2^gt/gt WAT than in the wild-type WAT (Fig. 6B and C and Supplementary Fig. 6A). The glucose tolerance of the HFD-fed Hif-p4h-2^gt/gt mice was better than that of the HFD-fed wild-type mice, and their fasting blood glucose levels (at 0 min) were likewise lower (Fig. 6D), whereas the lower serum insulin values (by 15%) and HOMA-IR scores (by 36%) in the HFD-fed Hif-p4h-2^gt/gt mice were not statistically significant (Supplementary Fig. 6B). Livers of all 6-month-old HFD-treated wild-type mice had steatosis (Fig. 6E), with four of the nine having a steatosis score of ++++, whereas only four of the eight Hif-p4h-2^gt/gt mice had steatosis (P = 0.03), with only one of these having a score of +++ and none having ++++. Thus, the Hif-p4h-2^gt/gt mice were protected against the development of HFD-induced hepatic steatosis.

FG-4497 inhibits all three HIF-P4Hs competitively with respect to 2-oxoglutarate, with similar IC₅₀ values (7). We studied whether its oral administration can be used to reverse metabolic dysfunction in two models: 1) 1-year-old wild-type mice fed normal chow that were shown to have metabolic dysfunction (Fig. 4) and 2) 3.5-month-old wild-type mice fed HFD for 6 weeks before the administration of 60 mg/kg FG-4497 on days 1, 3, and 5 of each week was begun (7). This FG-4497 dose stabilizes Hif-1α and Hif-2α in mouse kidney and liver and increases serum erythropoietin concentration about sixfold (7).

After a 1-week adjustment period, FG-4497 administration to 1-year-old mice fed normal chow reduced the weight of these mice during the subsequent 5 weeks by ∼1.3 g, whereas the vehicle-treated mice gained ∼0.6 g (Fig. 7A). The adipocytes were smaller and the number of macrophage aggregates lower in the WAT of the FG-4497–treated than those of the vehicle-treated mice (Fig. 7B and C). The serum total cholesterol level and the HDL and LDL + VLDL cholesterol levels of the FG-4497–treated mice were significantly decreased, whereas their HDL/LDL + VLDL ratio was increased (Fig. 7D). The fasting blood glucose levels of the FG-4497–treated mice were likewise decreased, whereas the decreases in the serum insulin level by ∼25% and HOMA-IR score by ∼75% (Fig. 7E) were not statistically significant (P = 0.11 in both cases).

In the other model, 2-month-old mice were fed normal chow or HFD for 6 weeks, after which the mice fed normal chow were given vehicle, and those fed HFD were given either HFD and vehicle or HFD and FG-4497 for 4 weeks. During the initial 6-week period, the HFD-fed mice gained more weight than those fed normal chow (Fig. 7F). The FG-4497 treatment decreased the weight of the HFD-fed mice by ∼0.6 g, whereas their vehicle-treated controls gained ∼3.0 g (FG-4497 vs. vehicle mice P = 0.02) (Fig. 7F). The WAT weight of the FG-4497–treated HFD mice was lower than that of their controls (Fig. 7G). The glucose tolerance of the FG-4497–treated HFD mice was better than that of their vehicle-treated controls (Fig. 7H), and their fasting serum insulin levels and HOMA-IR scores were significantly decreased (Fig. 7I).

The data indicate that Hif-p4h-2^gt/gt mice fed either a normal chow or an HFD have less adipose tissue, smaller adipocytes, a decreased number of adipose tissue macrophage aggregates, and lower serum cholesterol levels than their littermates. They are also protected against hepatic steatosis and show increased glucose tolerance and insulin sensitivity.

The Hif-p4h-2^gt/gt mice had higher levels of glucose transporters and glycolysis enzymes in their skeletal muscle and WAT, with similar higher levels previously found in their hearts (11). Uptake of deoxyglucose in skeletal muscle was also increased. Furthermore, Pdk1 expression was increased in the Hif-p4h-2^gt/gt skeletal muscle and WAT, as has also been found in the Hif-p4h-2^gt/gt heart (11). A higher Pdk1 level may further increase glycolysis by inhibiting the entry of pyruvate into the citric acid cycle (22). Thus, it appears that glycolysis is increased in several Hif-p4h-2^gt/gt tissues, contributing to the overall improved glucose tolerance (Supplementary Fig. 7). These changes agree with the established consequences of the stabilization of HIF-1α (22). HIF-P4H-1 and -3 have been reported to also have enzyme-specific substrates other than HIF-1α and HIF-2α (1–3), and thus, changes in the levels of those two enzymes may also influence HIF-independent pathways. Such substrates have so far not been identified for HIF-P4H-2; therefore, it seems likely that most, if not all, of the metabolic changes found in the Hif-p4h-2^gt/gt mice were mediated by Hif-α.

Obesity is associated with a chronic low-grade inflammation that predisposes to insulin resistance. Adipose tissue macrophages are believed to play a key role in obesity-induced insulin resistance (20,21). They infiltrate obese adipose tissue and along with the hypertrophied adipocytes, release cytokines and adipokines that contribute to the proinflammatory response (20). Macrophage-derived proinflammatory factors block insulin action in adipocytes by downregulating the expression of the insulin-regulated GLUT4 and impairing insulin-stimulated GLUT4 transport to the plasma membrane (20). Because we found decreased size of adipocytes, a reduced number of adipose tissue macrophages, and increased Glut4 expression in the Hif-p4h-2^gt/gt mice, it seems likely that these changes contribute to increased insulin sensitivity in these mice (Supplementary Fig. 7).

Increased expression of the Hif-2α target Irs2 in the liver of mice with acute hepatic Hif-p4h-3 deletion was accompanied by a decreased Srebp1c expression (26). The present results indicating weak stabilization of Hif-2α, increased expression of Irs2, and decreased expression of Srebp1c and its targets Accα and Fas in the Hif-p4h-2^gt/gt liver agree with those data. These changes are probably responsible for the decreased fatty acid synthesis and de novo lipogenesis found in the Hif-p4h-2^gt/gt liver and WAT. The lack of acetyl-CoA in Hif-p4h-2^gt/gt tissues, which is presumably due to pyruvate dehydrogenase inhibition, is likely to contribute to the decreased lipogenesis (Supplementary Fig. 7).

Extensive liver-specific stabilization of Hif-2α leads to hepatic steatosis (26,28). However, the Hif-p4h-2^gt/gt mice in the present study showed no steatosis but were instead protected against it. Liver-specific stabilization of Hif-2α by acute Hif-p4h-3 deletion likewise did not lead to hepatic steatosis, suggesting that low-level hepatic Hif-2α stabilization, as found in the present Hif-p4h-2^gt/gt mice, has beneficial effects, whereas extensive hepatic Hif-2α stabilization leads to steatosis (26,28).

Liver-specific stabilization of Hif-1α and Hif-2α appears to have no effect on hepatic cholesterol synthesis or intestinal cholesterol absorption, but extensive liver-specific Hif-2α stabilization increases hepatic and serum cholesterol levels (28,29) as a result of decreased cholesterol oxidation to bile acids (29). However, the Hif-p4h-2^gt/gt mice with low-level hepatic stabilization of Hif-2α had decreased serum cholesterol levels. The decreased amount of acetyl-CoA is likely to contribute to the low serum cholesterol level in the Hif-p4h-2^gt/gt mice (Supplementary Fig. 7), but other mechanisms may also be involved.

Mice with adipocyte-specific Hif-p4h-2 deletion also have less WAT, smaller adipocytes, a lower number of adipose tissue macrophages, and improved glucose tolerance (25); however, such changes were seen only in those fed an HFD, and no changes were reported in serum cholesterol levels, suggesting that Hif-p4h-2 deficiencies in several tissues play an important role in metabolic changes in Hif-p4h-2^gt/gt mice. Acute hepatic Hif-p4h-3 deletion has also been reported to improve glucose tolerance and insulin sensitivity, but no data were available on its effects on weight gain or serum cholesterol levels (26).

Inhibition of Hif-1α by disruption of its gene in adipocytes (30) or administration of its inhibitor (31) or antisense oligonucleotides (32) attenuates the consequences of an HFD in mice. Currently, no explanation is available for the discrepancy between those data and the beneficial effects of Hif-α stabilization by Hif-p4h-2 deficiency, but the additional stabilization of Hif-2α has been suggested to possibly play an important role (25).

Administration of FG-4497 to mice in two models of metabolic dysfunction led to changes very similar to those seen in the Hif-p4h-2^gt/gt mice, indicating that HIF-P4H-2 inhibition may not only protect against the development of obesity and metabolic dysfunction but also reverse them. FG-4497 inhibits all three HIF-P4Hs, but in view of the changes found in the Hif-p4h-2^gt/gt mice, it would seem possible to obtain a similar effect with a compound that specifically inhibits HIF-P4H-2. Of note, administration of another pan-HIF-P4H-inhibitor, FG-4592, currently in clinical trials for treatment of anemia of chronic kidney disease, also lowers serum cholesterol levels and increases the HDL/LDL ratio (33,34), thus supporting the view that HIF-P4H-2 inhibition may indeed be a useful strategy for the treatment of obesity and its consequences.

Acknowledgments. The authors thank T. Aatsinki, R. Juntunen, E. Lehtimäki, S. Rannikko, and M. Siurua for excellent technical assistance.

Funding. This study was supported by Academy of Finland grants 200471 and 202469 (to J.M.); Center of Excellence 2012–2017 grant 251314 (to J.M.); the S. Jusélius Foundation (to J.M. and P.K.); Academy of Finland grants 120156, 140765, 218129, and 266719 (to P.K.); and the Emil Aaltonen Foundation (to P.K.).

Duality of Interest. G.W. is a senior cell biology director at FibroGen, Inc. K.I.K. is a scientific founder and consultant of FibroGen, Inc., which develops HIF-P4H inhibitors as potential therapeutics. K.I.K. and J.M. own equity in this company, and the company has sponsored research in the laboratory headed by K.I.K. and currently supports research headed by J.M. No other potential conflicts of interest relevant to this article were reported.

Author Contributions. L.R.-K., S.K., and R.S. contributed to the research and data analysis. S.H. contributed expertise in serum lipid analyses. R.B.S. and E.L. contributed to the MRI analyses. K.A.M. and K.-H.H. contributed to the metabolic home cage experiments and analysis. G.W. provided the FG-4497 and made useful suggestions. K.I.K. contributed to generating the Hif-p4h-2^gt/gt mouse line and to the study design, data analysis, and writing of the manuscript. J.M. contributed to generating the Hif-p4h-2^gt/gt mouse line and to the discussions. P.K. contributed to generating the Hif-p4h-2^gt/gt mouse line and study supervision and to the study design, data analysis, and writing of the manuscript. P.K. 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.