Gijs den Besten,1,2 Aycha Bleeker,1,3 Albert Gerding,4 Karen van Eunen,1,2,3

Rick Havinga,1 Theo H. van Dijk,4 Maaike H. Oosterveer,1 Johan W. Jonker,1

Albert K. Groen,1,2,3,4 Dirk-Jan Reijngoud,1,2,4 and Barbara M. Bakker1,2

Short-Chain Fatty Acids ProtectAgainst High-Fat Diet–Induced Obesityvia a PPARg-Dependent Switch FromLipogenesis to Fat OxidationDiabetes 2015;64:2398–2408 | DOI: 10.2337/db14-1213

Short-chain fatty acids (SCFAs) are the main products ofdietary fiber fermentation and are believed to drive thefiber-related prevention of the metabolic syndrome.Here we show that dietary SCFAs induce a peroxisomeproliferator–activated receptor-g (PPARg)–dependentswitch from lipid synthesis to utilization. Dietary SCFAsupplementation prevented and reversed high-fat diet–induced metabolic abnormalities in mice by decreasingPPARg expression and activity. This increased the ex-pression of mitochondrial uncoupling protein 2 andraised the AMP-to-ATP ratio, thereby stimulating oxida-tive metabolism in liver and adipose tissue via AMPK.The SCFA-induced reduction in body weight and stimu-lation of insulin sensitivity were absent in mice withadipose-specific disruption of PPARg. Similarly, SCFA-induced reduction of hepatic steatosis was absent inmice lacking hepatic PPARg. These results demonstratethat adipose and hepatic PPARg are critical mediatorsof the beneficial effects of SCFAs on the metabolic syn-drome, with clearly distinct and complementary roles.Our findings indicate that SCFAs may be used therapeu-tically as cheap and selective PPARg modulators.

The shift in Western and developing countries froma traditional high-fiber, low-fat, low-calorie diet towarda low-fiber, high-fat, high-calorie diet is accompanied by agrowing prevalence of obesity and insulin resistance (1,2).Dietary fiber supplementation, on the other hand, has beenshown to reduce body weight, insulin resistance, and

dyslipidemia (3–5). The main products of intestinal fermen-tation of dietary fibers are short-chain fatty acids (SCFAs), ofwhich acetate, propionate, and butyrate are the most abun-dant (6). Although these three SCFAs are rapidly assimilatedinto host carbohydrates and lipids—providing ;10% of ourdaily energy requirements (7)—there are clear differences inthe way each is metabolized: propionate is primarily a pre-cursor for gluconeogenesis, whereas acetate and butyrateare rather incorporated into fatty acids and cholesterol(8). Besides serving as an energy source, SCFAs also reg-ulate metabolism by inhibition of histone deacetylasesand chain-length–dependent activation of the endogenousG-protein–coupled receptor (GPR) 41 and 43 (9–11). Thelonger butyrate is more selective for GPR41, the shorteracetate is more selective for GPR43, and propionate bindsto both receptors (12,13). The SCFAs propionate and bu-tyrate were recently shown to increase intestinal gluco-neogenesis (IGN), resulting in beneficial effects on glucoseand energy homeostasis (14). The mechanism, however,was different for the different SCFAs. Butyrate actedthrough a cAMP-dependent mechanism, whereas propio-nate, itself a substrate of IGN, activated IGN gene expres-sion via a gut-brain neural circuit involving GPR41.

Despite the differences between the metabolism andGPR affinities of the three SCFAs, they ameliorate high-fat diet (HFD)–induced obesity and insulin resistance to asimilar extent when given as a dietary supplement (15–17).This suggests that there is a common, GPR-independentmolecular mechanism for the beneficial effects of acetate,

1Center for Liver, Digestive and Metabolic Diseases, Department of Pediatricsand Systems Biology, Center for Energy Metabolism and Ageing, University ofGroningen, University Medical Center Groningen, Groningen, the Netherlands2Netherlands Consortium for Systems Biology, Amsterdam, the Netherlands3Top Institute Food and Nutrition, Wageningen, the Netherlands4Department of Laboratory Medicine, University of Groningen, University MedicalCenter Groningen, Groningen, the Netherlands

Corresponding author: Barbara M. Bakker, b.m.bakker01@umcg.nl.

Received 4 August 2014 and accepted 9 February 2015.

This article contains Supplementary Data online at http://diabetes.diabetesjournals.org/lookup/suppl/doi:10.2337/db14-1213/-/DC1.

© 2015 by the American Diabetes Association. Readers may use this article aslong as the work is properly cited, the use is educational and not for profit, andthe work is not altered.

2398 Diabetes Volume 64, July 2015

METABOLISM

propionate, and butyrate. A GPR-independent mechanismis in line with recent findings that GPR41-deficient miceare still sensitive to the effects of SCFAs on body weightand insulin sensitivity (17). It is likely that a commonmechanism may involve AMPK because acetate and buty-rate have been shown to increase energy expenditure byactivating AMPK in liver and muscle tissue (15,18),whereas propionate activates AMPK in colon cancer cellsand reduces lipid synthesis in isolated rat hepatocytes(19,20). The mechanism by which SCFAs activate AMPKis, however, unknown.

In this study we identify a unifying, AMPK-dependentmechanism by which the three SCFAs mediate the beneficialeffects of dietary fiber on the metabolic syndrome. We reveala cascade of events that starts with downregulation ofperoxisome proliferator–activated receptor-g (PPARg) activ-ity and in which liver and adipose PPARg play distinct,complementary roles.

RESEARCH DESIGN AND METHODS

Animals and Experimental DesignMale C57Bl/6J mice (Charles River, L’Arbresle, France),2 months of age, were housed in a light- and temperature-controlled facility (lights on from 6:30 A.M. to 6:30 P.M., 21°C)with free access to water and food. The experimental groupswere fed a semisynthetic HFD (D1245; Research Diet Ser-vices, Wijk bij Duurstede, the Netherlands) in which 45% ofcalories were from palm oil fat. For the SCFA diets, sodiumacetate (S2889; Sigma-Aldrich), sodium propionate (P1880;Sigma-Aldrich), or sodium butyrate (303410; Sigma-Aldrich)was incorporated into the diet at 5% (w/w). A normal-fatcontrol group received a chow diet (RMH-B; Hope Farms,Woerden, the Netherlands). Mice in which exons 1 and 2 ofthe PPARg gene were loxP-flanked (PPARg f/f) were pro-vided by Ronald M. Evans (Salk Institute) and have beendescribed previously (21). PPARg lox/lox mice were crossedwith C57Bl/6J transgenic mice expressing Cre recombinaseunder the control of the albumin promoter, which isexpressed in liver (L-KO), or the aP2 promoter, which tar-gets adipose tissue (A-KO). Experimental procedures wereapproved by the University of Groningen Ethics Committeefor Animal Experiments.

Lipogenesis and b-OxidationIn vivo lipogenesis was determined by incorporation of[1-13C]acetate into palmitate by providing 2% (w/v)[1-13C]acetate in drinking water for 24 h, as describedpreviously (22). Fatty acid b-oxidation capacity was deter-mined in fresh liver and adipose homogenates accordingto Hirschey et al. (23). Briefly, tissue was homogenized insucrose/Tris/EDTA buffer, incubated for 30 min in thereaction mixture (pH 8.0) containing [1-14C]palmiticacid, and trapped [14C]CO2 was measured.

Insulin Tolerance and SensitivityIntraperitoneal glucose tolerance was tested after anintraperitoneal injection of glucose (2 g/kg body weight)

after an overnight fast. Intraperitoneal insulin tolerancewas tested after an intraperitoneal injection of insulin(NovoRapid; 0.75 units/kg body weight) after a 4-h fast.Hyperinsulinemic-euglycemic clamp studies were per-formed as previously described (24).

Plasma and Tissue SamplingThe mice were fasted from 6:00 to 10:00 A.M. Blood glu-cose concentrations were measured using a EuroFlashmeter (Lifescan Benelux, Beerse, Belgium). Mice weresubsequently killed by cardiac puncture under isofluraneanesthesia. Liver and epididymal fat pads were weighed,snap-frozen in liquid nitrogen, and stored at 280°C. Partof the fat tissue was fixed in 4% paraformaldehyde in PBSand embedded in paraffin. For adipocyte histology, paraf-fin sections (3 mm) were stained with hematoxylin andeosin and analyzed at original magnification 320. Bloodwas centrifuged at 4,000g for 10 min at 4°C, and plasmawas stored at 220°C. Plasma nonesterified fatty acid(NEFA) concentrations were determined using a commer-cially available kit (Roche Diagnostics, Mannheim, Ger-many). Plasma leptin and insulin levels were determinedusing ELISA (ALPCO Diagnostics, Salem, NH). Hepatictriglyceride content was determined using a commerciallyavailable kit (Roche) after lipid extraction (25).

Hepatic malonyl-CoA and adenosine concentrationswere determined by high-performance liquid chromatog-raphy according to Demoz et al. (26) and Miller et al. (27),respectively. Carnitine palmitoyltransferase I (CPT-1) ac-tivity was determined in liver homogenates at differentdilutions (20, 10, 5, 2, and 1 ng/mL) according to vanVlies et al. (28) with 50 mmol/L palmitoyl-CoA and2 mmol/L L-carnitine as substrates. The reaction wasquenched at 0, 2, 5, and 10 min.

Indirect CalorimetryOxygen consumption, energy expenditure, respiratoryexchange ratio (RER), food intake, and activity patternswere measured simultaneously for each mouse usinga Comprehensive Laboratory Animal Monitoring System(TSE Systems GmbH, Bad Homburg, Germany). Theenergy balance was determined by measuring the energycontent of diet and that of dried, homogenized fecesusing a bomb calorimeter (CBB 330; standard benzoic acid6,320 cal/g, BCS-CRM no. 90N).

Oxygen Consumption Rates in Liver MitochondriaMitochondria were isolated from fresh liver tissue accord-ing to Mildaziene et al. (29). The rates of oxygen con-sumption in isolated liver mitochondria were measuredat 37°C using a two-channel high-resolution Oxygraph-2k (Oroboros, Innsbruck, Austria) in mitochondrial respi-ration medium (30) with palmitoyl-CoA as substrate.Maximal ADP-stimulated oxygen consumption (i.e., state3) was achieved by adding 4.8 units/mL hexokinase,12.5 mmol/L glucose, and 1 mmol/L ATP. The restingstate (i.e., state 4) oxygen consumption rate was deter-mined after blocking ADP phosphorylation with 1.25

diabetes.diabetesjournals.org den Besten and Associates 2399

mmol/L carboxyatractyloside. The respiratory control ra-tio was calculated by dividing oxygen consumption rate instate 3 by oxygen consumption rate in state 4.

HepG2, 3T3-L1, and C2C12 experimentsHepG2, 3T3-L1, and C2C12 cells, purchased from Ameri-can Type Culture Collection (Manassas, VA), were main-tained at 37°C in 5% CO2 in DMEM with 10% FBS and1% penicillin/streptomycin. For HepG2 experiments, cellswere plated in DMEM with 10% FBS and 1% penicillin/streptomycin in six-well plates and incubated with 0.1, 1,or 3 mmol/L sodium acetate, sodium propionate, or sodiumbutyrate for 24 h with or without 100 nmol/L rosiglitazone(Sigma-Aldrich) or 10 mmol/L GW9662 (Sigma-Aldrich) asindicated. For 3T3-L1 experiments, cells were differenti-ated in six-well plates according to Zebisch et al. (31) andincubated with 0.1, 1, or 3 mmol/L sodium acetate, sodiumpropionate, or sodium butyrate for 24 h, with or without100 nmol/L rosiglitazone or 10 mmol/L GW9662, as in-dicated. For C2C12 experiments, cells were differentiatedwith 2% horse serum in six-well plates according to Fujitaet al. (32) and incubated with 3 mmol/L sodium acetate,sodium propionate, or sodium butyrate for 24 h.

Gene Expression Levels and Immunoblot AnalysisRNA was extracted from liver and adipose tissue using Trireagent (Sigma-Aldrich, St. Louis, MO) and converted intocDNA by a reverse transcription procedure using Moloneymurine leukemia virus and random primers according tothe manufacturer’s protocol (Sigma-Aldrich). For quanti-tative PCR, cDNA was amplified using the appropriateprimers and probes. The sequence of the other primerscan be found in Supplementary Table 1. mRNA levels werecalculated relative to 36b4 expression and normalized forexpression levels of mice fed an HFD.

For immunoblot analysis, whole-cell lysate was preparedin lysis buffer, and the protein concentrations were de-termined using the BCA Protein Assay kit (Pierce). In-dividual samples were mixed with loading buffer, heated for5 min at 96°C, and subjected to SDS-PAGE. Antibodies andtheir sources were as follows: AMPK (no. 2532; Cell Signal-ing), phosphorylated AMPK (pAMPK Thr172, no. 2531; CellSignaling), acetyl-CoA carboxylase (ACC, no. 45174; Abcam),phosphorylated ACC (pACC S79, no. 31931; Abcam), uncou-pling protein 2 (UCP2, no. 6525; Santa Cruz Biotechnology),PPARg (no. 2435; Cell Signaling), and fatty acid synthase(no. 3180; Cell Signaling). As loading control, b-actin (no.2066; Sigma-Aldrich) was used for liver and adipose tissue,and TOM20 (no. 11415; Santa Cruz Biotechnology) wasused for isolated liver mitochondria. Finally, horseradishperoxidase–conjugated anti-rabbit from donkey (AmershamPharmacia Bioscience) or horseradish peroxidase–conjugatedanti-goat from donkey (Dako, Glostrup, Denmark) andSuperSignal West Pico Chemiluminescent SubstrateSystem (Pierce) were used. The immunoblots were ana-lyzed by densitometry using Image Laboratory software(Bio-Rad).

Statistical analysisData are presented as mean values 6 SEM. Statisticalanalysis was assessed by one-way ANOVA using the Tukeytest for post hoc analysis. Statistical significance wasreached at a P value of ,0.05.

RESULTS

Prevention and Treatment of HFD-Induced Obesity andInsulin Resistance by SCFAsTo examine the effects of SCFAs on the development ofobesity and insulin resistance, wild-type (WT) C57Bl/6Jmice were fed an HFD with acetate, propionate, orbutyrate (5% w/w) for 12 weeks. This provided theanimals with a physiological amount of SCFA because thecontribution of SCFA to the total energy intake has beenestimated at ;10% (7). The substantial raise in bodyweight that was observed in controls, fed an HFD only,was attenuated by all three SCFAs to a similar extent(Fig. 1A). This coincided with reductions in white adi-pose tissue (WAT) mass, adipose cell size, and plasmaleptin concentrations (Supplementary Fig. 1A–C). Thelower body weight in the SCFA groups was not due toalterations in food intake or physical activity becauseneither of these was significantly affected by SCFA sup-plementation (Supplementary Fig. 1D and E). We did,however, observe enhanced energy expenditure in theSCFA-treated mice (Fig. 1B and Supplementary Fig. 1F)as well as a shift from carbohydrate to fatty acid oxida-tion, as indicated by lower RER (Fig. 1C). This suggeststhat SCFAs reduce HFD-induced body weight gain byenhancing energy expenditure through increased lipidoxidation.

SCFA-fed and control-fed mice had similar fastingblood glucose levels and glucose tolerance (Supplemen-tary Fig. 1G–I) but much lower fasting insulin levels (Fig.1D). Together with enhanced disposal of glucose uponinsulin injection (Supplementary Fig. 1J), this points tohigher insulin sensitivity in these mice. We thereforeperformed hyperinsulinemic-euglycemic clamp studiesunder matched insulin exposure. The glucose infusionrate required for maintaining euglycemia (a measure ofwhole-body insulin sensitivity) in SCFA-fed mice was;1.5-fold higher than that in control mice (Fig. 1E).Although the hepatic glucose production rate duringthe clamp was similar in all groups (Ra in Fig. 1F), thedegree to which insulin stimulated the rate of glucoseuptake by peripheral tissues (primarily muscle and adi-pose tissue) was much higher in SCFA-fed mice (Rd inFig. 1F). This indicates improved peripheral insulin sen-sitivity in these mice. Collectively, our observations dem-onstrate that supplementation of any of the three SCFAsenhances the insulin sensitivity of HFD-fed mice toa similar extent.

SCFA supplementation completely prevented HFD-induced obesity and insulin resistance, as indicated bysimilar body weight, WAT mass, plasma leptin and insulinlevels, and peripheral insulin sensitivity compared with

2400 SCFAs Prevent Metabolic Disorders via PPARg Diabetes Volume 64, July 2015

mice fed the normal-fat chow diet (Supplementary Fig.2A–G). The HFD did not significantly affect the basalglucose level, which was similar in mice fed chow andthe HFD (Supplementary Fig. 2D).

Finally, we wondered whether SCFAs could also be usedto treat existing obesity and insulin resistance. To this end,we first fed mice an HFD for 12 weeks to induce obesityand then supplemented the HFD with SCFAs for 6 weeks.Indeed, the mice supplemented with SCFAs showedsignificantly lower body weight and WAT mass after thistreatment than controls as well as a shift toward fatty acidoxidation and enhanced insulin sensitivity, without affect-ing the food intake (Supplementary Fig. 3A–I). After 18weeks, the HFD control mice had the same basal glucoselevel as the HFD and chow controls at 12 weeks (comparedwith Supplementary Figs. 3E and 2D).

SCFAs Stimulate Mitochondrial Fatty Acid Oxidation byActivation of the UCP2-AMPK-ACC PathwayControl mice fed the HFD had high plasma concentrationsof NEFAs and high liver concentrations of triglycerides(hepatic steatosis). These aspects of the metabolic syn-drome were reduced when SCFAs were supplemented fromthe beginning (Supplementary Fig. 4A and B) or after the

mice were already obese (Supplementary Fig. 3H and I).This prompted us to study hepatic fatty acid synthesisand oxidation. SCFA-fed mice had lower transcript levelsof genes involved in hepatic lipogenesis and a lower con-centration of hepatic fatty acid synthase protein (Supple-mentary Fig. 4C and D). In agreement, these mice hada twofold reduction in in vivo hepatic lipid synthesis (Fig.2A). The capacity for hepatic lipid oxidation in SCFA-fedmice was twofold higher than that of controls (Fig. 2B).Clearly, hepatic lipid metabolism was shifted towarda more oxidative state.

We wondered whether the metabolic effects of SCFAsmight be mediated through activation of AMPK, which isknown to shift metabolism from lipid synthesis tooxidation (33). Indeed, we observed increased phosphor-ylation of AMPK and its downstream target ACC (Fig. 2C),without affecting the total hepatic AMPK and ACC levels(Supplementary Fig. 4E). Phosphorylation inactivatesACC and should lead to lower concentrations of itsproduct malonyl-CoA, an endogenous inhibitor of CPT-1, the first enzyme in the b-oxidation of fatty acids (34).Consistently, SCFA treatment reduced hepatic malonyl-CoA concentrations (Supplementary Fig. 4F) and in-creased the enzyme capacity (Vmax) of CPT-1 in diluted

Figure 1—SCFAs protect against HFD-induced obesity and insulin resistance. Eight-week-old male C57Bl/6J mice were fed an HFDsupplemented with acetate, propionate, or butyrate (5% w/w). A: Body weight was monitored for 12 weeks on the indicated diets. *P< 0.05acetate vs. control; #P < 0.05 propionate vs. control; $P < 0.05 butyrate vs. control. Energy expenditure (B) and RER (C) were evaluatedusing indirect calorimetry data in animals after 10 weeks on the indicated diets. D: Blood insulin levels were measured in animals after12 weeks on the indicated diets and after a 4-h fast. E and F: Hyperinsulinemic-euglycemic clamp studies were performed in animals fedthe indicated diets for 10 weeks. Glucose infusion rate (GIR), glucose production rate (Ra), and glucose uptake rate (Rd) were calculatedafter the test. Values are means 6 SEM for n = 7–8. *P < 0.05, **P < 0.01, ***P < 0.001 vs. control.

diabetes.diabetesjournals.org den Besten and Associates 2401

liver homogenates (Fig. 2D). This implies that CPT-1is stimulated by SCFAs via a dual mechanism: by rais-ing its Vmax and by reducing the concentration of itsinhibitor.

SCFA-fed mice had decreased hepatic ATP concentra-tions and increased AMP-to-ATP ratios (SupplementaryFig. 4G and H). The latter is a sensitive reflection of theenergetic state of the cell and a direct activator of AMPK(35). Reduced ATP concentrations can be a result of in-creased mitochondrial proton leakage, leading to mitochon-drial uncoupling and subsequently reduced ATP synthesis(36). Therefore, we examined oxygen consumption by iso-lated liver mitochondria using palmitoyl-CoA as a respira-tory substrate, both in the presence of ADP (state 3) and inthe presence of an inhibitor of ATP production (state 4).We observed lower respiratory control ratios (the rate ofstate 3 divided by state 4 respiration) in liver mito-chondria from SCFA-fed mice. This could be attributedto an increased state 4 respiration rate (Fig. 2E), dem-onstrating that there is intrinsic uncoupling of mito-chondrial oxidative phosphorylation in the livers ofthese mice. In line with this, SCFA feeding led to in-creased expression of UCP2 (Fig. 2F), suggesting thatproton leak via UCP2 may be responsible for the ob-served uncoupling (36).

Activation of the UCP2-AMPK-ACC Pathway by SCFAsDepends on PPARg

Next, we studied how SCFAs activate the UCP2-AMPK-ACC pathway. Possible candidates were PPARa and g,which are known regulators of UCP2 expression, fattyacid oxidation, and whole-body lipid metabolism (37–40). SCFAs did not significantly affect the expression ofPPARa, whereas expression of target genes involved infatty acid oxidation was decreased rather than increased(Supplementary Fig. 5A–C). This suggests that PPARa isnot responsible for the enhanced fatty acid oxidation ca-pacity of SCFA-treated mice. In contrast, SCFAs did re-duce expression of PPARg—and its target genes Cd36,Lpl, Fabp4, and Pltp—in liver and adipose tissue, butnot in muscle (Supplementary Fig. 5D–G). A reduced ex-pression of PPARg expression or activity stimulates UCP2expression and fatty acid oxidation and reduces lipogen-esis and hepatic triglyceride levels (41,42), suggesting thatPPARg may well be the mediating factor between SCFAsand the UCP2-AMPK-ACC pathway. PGC-1a mRNA ex-pression in liver, adipose, and muscle tissue and UCP-1mRNA expression in brown adipose tissue did not changeafter SCFA feeding (Supplementary Fig. 5H and I), sug-gesting that neither PGC-1a nor UCP-1 play a role in theSCFA-induced effect.

Figure 2—SCFAs enhance oxidative metabolism. A: Hepatic lipogenesis was determined in vivo by incorporation of [1-13C]acetatedissolved in drinking water after 12 weeks of the HFD with or without SCFAs. B: Hepatic b-oxidation was determined ex vivo in liverhomogenates by trapping 14C-labeled CO2 produced during incubation with [1-14C]palmitic acid. C: Hepatic pAMPK and pACC proteinlevels were analyzed by Western blot of tissue lysates from mice after 12 weeks on the diet. D: CPT-1 activity was analyzed by measure-ment of palmitoyl-carnitine produced in total liver homogenates during incubation with palmitoyl-CoA and L-carnitine as substrates. E: Livermitochondria were isolated and maximal ADP-stimulated oxygen consumption (i.e., state 3) and oxygen consumption in the presence ofoligomycin inhibition of ATP synthesis (i.e., state 4) was determined. RCR, respiratory control ratio. F: Mitochondrial UCP2 protein levelswere analyzed by Western blot of tissue lysates from mice after 12 weeks on the indicated diets. Values are means6 SEM for n = 7–8. *P<0.05, **P < 0.01, ***P < 0.001 vs. control.

2402 SCFAs Prevent Metabolic Disorders via PPARg Diabetes Volume 64, July 2015

To find out whether PPARgmay be causally involved inthe induction by SCFAs of the UCP2-AMPK-ACC signalingpathway, we first investigated this in vitro in liver cells(HepG2), differentiated adipose cells (3T3-L1), and mus-cle cells (C2C12). Treating cells for 24 h with 3 mmol/L ofany of the SCFAs reduced mRNA and protein levels ofPPARg and its target genes in HepG2 and 3T3-L1 cellsbut not in C2C12 cells (Fig. 3A–C). SCFAs also enhancedthe activity of the UCP2-AMPK-ACC signaling pathway inHepG2 and 3T3L1 cells but not in C2C12 cells (Fig. 3D–F), in line with our in vivo results. The induction of theUCP2-AMPK-ACC pathway by SCFAs was abolished whenthe partial repression of PPARg expression and activitywas compensated by the PPARg agonist rosiglitazone(Supplementary Fig. 6A). However, complete inhibitionof the activity of PPARg by the PPARg antagonistGW9662 did not affect the SCFA-induced reduction inPPARg expression but did abolish the accompanying

increase in the activity of the UCP2-AMPK-ACC pathway(Supplementary Fig. 6B). Apparently, activation or inhibi-tion of the activity of PPARg abolished the SCFA-inducedincrease of the activity of the UCP2-pAMPK-pACCpathway.

The SCFA concentrations of 3 mmol/L used hererepresent typical intestinal concentrations rather thanportal blood concentrations (43), although even portalSCFA concentrations may exceed 2 mmol/L postprandiallyafter a fiber-rich diet is digested (44–46). Therefore, wealso incubated HepG2 and differentiated 3T3-L1 cells atdifferent concentrations of SCFAs. In both HepG2 and3T3-L1 cells, there was only a significant effect on PPARgexpression after 24 h exposure to 3 mmol/L SCFAs,whereas 1 and 0.1 mmol/L showed no reduction comparedwith the control (Supplementary Fig. 6C and D).

Altogether, these results indicate that the activation ofthe UCP2-pAMPK-pACC pathway by SCFAs in HepG2 and

Figure 3—Activation of the UCP2-AMPK-ACC pathway by SCFAs depends on PPARg. HepG2 cells, differentiated 3T3-L1 cells, anddifferentiated C2C12 cells were incubated with 3 mmol/L SCFAs for 24 h in the presence of 100 nmol/L rosiglitazone or 10 mmol/L GW9662as indicated. mRNA expression of PPARg and target genes was assessed via quantitative PCR in HepG2 (A), differentiated 3T3-L1 (B), andC2C12 (C ) cells after 24h incubation with SCFAs. PPARg, UCP2, pAMPK, and pACC protein levels were assessed by Western blot inHepG2 (D), differentiated 3T3-L1 (E), and C2C12 (F) cells after 24-h incubation with SCFAs. Values are means 6 SEM for n = 6. *P < 0.05 vs.control.

diabetes.diabetesjournals.org den Besten and Associates 2403

3T3-L1 cells was due to the observed reduction of PPARgexpression and activity, whereas PPARa had no role inthese SCFA-induced effects.

Hepatic PPARg Mediates the SCFA-InducedReduction in Hepatic Steatosis, Whereas AdiposePPARg Mediates SCFA Effects on HFD-InducedObesity and Insulin ResistanceTo distinguish between the role of PPARg in the liver andadipose tissue, L-KO mice or A-KO of PPARg mice werefed an HFD, with or without SCFA supplementation. TheSCFA-induced effects in these KO mice were comparedwith those of the WT above to analyze whether theywere mediated by liver or adipose PPARg.

Like the WT mice, the L-KO mice also showed a re-duction of body weight, WAT mass, and insulin levels inresponse to SCFA supplementation, demonstrating thatthese effects were not mediated by hepatic PPARg (Fig.4A and B and Supplementary Fig. 7A–C). This is in line

with our previous observation that SCFAs increased pe-ripheral but not hepatic insulin sensitivity (Fig. 1F).Whereas in WT mice the RER values and plasma NEFAconcentrations were reduced by SCFAs, they were in-creased by SCFAs in L-KO mice (Supplementary Fig. 7Dand E). In addition, the SCFA-induced increase in hepaticlipid oxidation capacity and the concomitant reduction inhepatic triglycerides that was observed in WT mice wereabolished in PPARg L-KO mice (Fig. 4C and D). Finally, wewere no longer able to detect any differences in hepaticprotein levels of UCP2, pAMPK, and pACC between theSCFA-fed and non-SCFA-fed PPARg L-KO mice (Fig. 4E),nor did we see any effects on the target genes of PPARg(Supplementary Fig. 7F). These results suggest that thebeneficial effect of SCFAs on liver lipid metabolism wasdirectly mediated by hepatic PPARg.

Our results with the PPARg A-KO mice were almostopposite to those with the L-KO mice. Disruption of

Figure 4—Hepatic PPARg deficiency impairs the SCFA-induced reduction in hepatic steatosis. Eight-week-old male L-KO PPARg micewere fed an HFD supplemented with acetate, propionate, or butyrate (5% w/w). A: Body weight was measured for 10 weeks. *P < 0.05acetate vs. control; #P< 0.05 propionate vs. control; $P< 0.05 butyrate vs. control. B: Insulin tolerance tests were performed on mice after9 weeks on the indicated diets and after a 4-h fast. C: Liver triglycerides in mice after 10 weeks on the indicated diets. D: Fatty acidb-oxidation in liver and WAT was measured in mice after 10 weeks on the indicated diets. E: PPARg, UCP2, pAMPK, and pACC proteinlevels were assessed by Western blot in liver and WAT lysates. Values are means 6 SEM for n = 6–8. *P < 0.05 vs. control.

2404 SCFAs Prevent Metabolic Disorders via PPARg Diabetes Volume 64, July 2015

adipose PPARg abolished the effects of SCFAs on bodyweight gain, WAT mass, and insulin levels (Fig. 5A and Band Supplementary Fig. 8A–C). In contrast to our obser-vations in the PPARg L-KO mice, SCFAs had no effect onRER and plasma NEFA concentrations in the PPARg A-KOmice (Supplementary Fig. 8D and E), whereas the strongSCFA-induced reduction in liver triglycerides was pre-served in these mice (Fig. 5C). Finally, no increase oc-curred in adipose lipid oxidation capacity or in adiposeprotein levels of UCP2, pAMPK, and pACC when SCFA-fed and non-SCFA-fed PPARg A-KO mice were compared(Fig. 5D and E), nor did we see any effects of SCFAs onthe target genes of PPARg (Supplementary Fig. 8F). Fur-ther reflecting the differences between the L-KO and theA-KO mice was that the adipose tissue of the L-KO micepreserved the SCFA-induced effects on lipid oxidationcapacity, PPARg protein levels, transcript levels of thePPARg target genes and the concomitant increase of

UCP2, pAMPK, and pACC protein levels (Fig. 4D and Eand Supplementary Fig. 7G), whereas the A-KO mice pre-served these characteristics in liver tissue (Fig. 5D and Eand Supplementary Fig. 8G).

Taken together, our results indicated that SCFA-induced protection against HFD-induced obesity andinsulin resistance is impaired by adipose PPARg defi-ciency, whereas hepatic PPARg deficiency impairs theSCFA-induced reduction in hepatic steatosis.

DISCUSSION

In this study, we demonstrate that the beneficial metaboliceffects of SCFAs—protection against HFD-induced obesityand improved insulin sensitivity—are mediated by down-regulation of PPARg. Our results are in line with otherstudies in which similar physiological effects of SCFAshave been found (15,17). However, by identifying PPARgas a central regulator of the systemic response, we are the

Figure 5—Adipose PPARg deficiency impairs the SCFA-induced protection against HFD-induced obesity and insulin resistance. Eight-week-old male A-KO PPARg mice were fed an HFD supplemented with acetate, propionate, or butyrate (5% w/w). A: Body weight wasmeasured for 10 weeks on the indicated diets. *P < 0.05 acetate vs. control; #P < 0.05 propionate vs. control; $P < 0.05 butyrate vs.control. B: Insulin tolerance tests were performed on mice after 9 weeks on the indicated diets and after a 4-h fast. C: Liver triglycerides inmice after 10 weeks on the indicated diets. D: Fatty acid b-oxidation in liver and WAT was measured in mice after 10 weeks on the indicateddiets. E: PPARg, UCP2, pAMPK, and pACC protein levels were assessed by Western blot in liver and WAT lysates. Values are means 6SEM for n = 6–8. *P < 0.05, ***P < 0.001 vs. control.

diabetes.diabetesjournals.org den Besten and Associates 2405

first to integrate the role of all three major SCFAs (acetate,propionate, and butyrate) into a single mechanism.

The combined results of our in vivo and in vitroexperiments suggest an SCFA-induced cascade in whichdownregulation of PPARg activates an UCP2-AMPK-ACCnetwork. This cascade shifts metabolism in adipose and livertissue from lipogenesis to fatty acid oxidation. SCFAs acti-vate the supply of fatty acid substrates via CPT1 and thedemand for their oxidation products via uncoupling of therespiratory chain. This is a clear example of what has beendescribed as multisite modulation; that is, the phenomenonthat metabolic fluxes can only be altered by simultaneousmodulation of multiple enzymes in a pathway (47).

The observed mitochondrial uncoupling may be partlyexplained by the increased expression of UCP2. UCP2belongs to a family of mitochondrial UCPs comprisingUCP1, 2, and 3, all three with proton leak activity (48,49).Whereas UCP1 has clear-cut proton-conducting activity,UCP2 catalyzes the proton-driven exchange of phosphatefor small dicarboxylic acids across the mitochondrial innermembrane (50). This allows UCP2 to dissipate the protongradient across the mitochondrial inner membrane in asubstrate-dependent manner.

SCFAs activate the expression of leptin via the SCFAreceptor GPR43 (51). Yet, we found reduced serum levelsof the satiety factor leptin in SCFA-treated animals. Thiscan be explained from the overall decrease in WAT mass.The same observation was made previously in mice thatoverexpress GPR43 (52). Despite their lower leptin levels,the food intake of SCFA-treated animals was not in-creased. This may be due to other satiety factors, whichwe did not investigate further.

In addition to our finding that PPARg is a central regu-lator of the beneficial effects of SCFAs, we also demonstratedthat liver and adipose tissue contribute independently to thismechanism: the beneficial effects on whole-body fat accumu-lation, hepatic steatosis, hypertriglyceridemia, and insulinsensitivity appear to be mediated differently in these twotissues. Whereas an SCFA-induced reduction of PPARg inthe liver reduced hepatic triglyceride concentrations, thesame reduction in adipose tissue reduced body weight andimproved insulin sensitivity. Cultured adipose and liver cellsrespond to SCFAs in the same way, by reducing PPARgand activating the UCP2-AMPK-ACC network, which sug-gests that the organs respond directly to the SCFAs thathave been taken up into the body. We recently showedthat the fluxes of SCFA uptake into the body—but not theSCFA concentrations in the cecum—are correlated withthe physiological effects of dietary fiber (53). This corrob-orates the idea that SCFAs need to be taken up and do notexert their full effect on organ physiology through intes-tinal receptors. In addition, De Vadder et al. (14) recentlyshowed that the intestinal gluconeogenesis pathway isalso an important mediator of SCFA-induced reductionof body weight and insulin resistance as PPARg is inour study. They showed—by capsaicin-induced periportalnervous deafferentation—that propionate exerts its

effect via a gut-brain communication axis. Intestinalknockout of the gluconeogenic gene G6Pc abolished thebenefits of fiber and SCFAs on body weight and glucosehomeostasis, as did our A-KO of PPARg. How this intes-tinal pathway relates to the PPARg pathway in liver andadipose is currently unclear. Similarly, whether and howthe PPARg-dependent pathway interacts with GPR41and/or GPR43 signaling remains to be established.

That we established PPARg as a central mediator of thebeneficial effects of SCFAs has possible consequences forthe treatment of metabolic disorders. A key finding in thePPARg field is that adipose mass increases almost propor-tionally to PPARg activity, whereas inhibition or activationof PPARg sensitizes the body for insulin (54). With regard totreatment, the most suitable molecules are most likely thosethat uncouple these different functions of PPARg such thatthe receptor sensitizes the body for insulin, without anyundesired adipogenesis. Such selective PPARg modulatorshave been proposed by others (55), and our results suggestthat SCFAs act as highly effective endogenous selectivePPARg modulators. The few human studies that havebeen conducted showed that intravenous administrationof acetate or propionate reduced plasma free fatty acids inhumans (56,57), whereas 12 weeks of dietary vinegar (ace-tate) supplementation in obese subjects led to lower bodyweight, body fat mass, and serum triglycerides levels than inthe placebo-controlled group (16). This makes the inexpen-sive SCFAs very attractive compounds to prevent and re-verse HFD-induced obesity and insulin resistance.

Funding. This work was funded by the Netherlands Genomics Initiative via theNetherlands Consortium for Systems Biology (91108004) from Netherlands Or-ganisation for Scientific Research (NWO)/ZonMw. A.B. and K.v.E. currently re-ceive an unrestricted research grant from Top Institute Food and Nutrition. J.W.J.is supported by a European Research Council grant (IRG-277169), by the HumanFrontier Science Program (CDA00013/2011-C), the NWO (VIDI grant 016.126.338),the Dutch Digestive Foundation (grant WO 11-67), and the Dutch DiabetesFoundation (grant 2012.00.1537). M.H.O. and B.M.B. hold a Rosalind FranklinFellowship from the University of Groningen.Duality of Interest. No potential conflicts of interest relevant to this articlewere reported.Author Contributions. G.d.B. conceived the study, designed and per-formed experiments, analyzed data, and wrote the manuscript. A.B., A.G., K.v.E.,R.H., T.H.v.D., and M.H.O. performed experiments, analyzed data, and providedinput into the manuscript. J.W.J., A.K.G., D.-J.R., and B.M.B. conceived thestudy, interpreted data, and edited the manuscript. B.M.B. is the guarantor ofthis work and, as such, had full access to all the data in the study and takesresponsibility for the integrity of the data and the accuracy of the data analysis.

References1. Galisteo M, Duarte J, Zarzuelo A. Effects of dietary fibers on disturbancesclustered in the metabolic syndrome. J Nutr Biochem 2008;19:71–842. Misra A, Singhal N, Khurana L. Obesity, the metabolic syndrome, and type 2diabetes in developing countries: role of dietary fats and oils. J Am Coll Nutr2010;29(Suppl.):289S–301S3. Papathanasopoulos A, Camilleri M. Dietary fiber supplements: effects inobesity and metabolic syndrome and relationship to gastrointestinal functions.Gastroenterology 2010;138:65–72, e1–e2

2406 SCFAs Prevent Metabolic Disorders via PPARg Diabetes Volume 64, July 2015

4. Dall’Alba V, Silva FM, Antonio JP, et al. Improvement of the metabolicsyndrome profile by soluble fibre - guar gum - in patients with type 2 diabetes:a randomised clinical trial. Br J Nutr 2013;110:1601–16105. Brown L, Rosner B, Willett WW, Sacks FM. Cholesterol-lowering effects ofdietary fiber: a meta-analysis. Am J Clin Nutr 1999;69:30–426. Nicholson JK, Holmes E, Kinross J, et al. Host-gut microbiota metabolicinteractions. Science 2012;336:1262–12677. Bergman EN. Energy contributions of volatile fatty acids from the gastro-intestinal tract in various species. Physiol Rev 1990;70:567–5908. den Besten G, Lange K, Havinga R, et al. Gut-derived short-chain fatty acidsare vividly assimilated into host carbohydrates and lipids. Am J Physiol Gastro-intest Liver Physiol 2013;305:G900–G9109. Xiong Y, Miyamoto N, Shibata K, et al. Short-chain fatty acids stimulateleptin production in adipocytes through the G protein-coupled receptor GPR41.Proc Natl Acad Sci U S A 2004;101:1045–105010. Samuel BS, Shaito A, Motoike T, et al. Effects of the gut microbiota on hostadiposity are modulated by the short-chain fatty-acid binding G protein-coupledreceptor, Gpr41. Proc Natl Acad Sci U S A 2008;105:16767–1677211. Vinolo MA, Rodrigues HG, Hatanaka E, Sato FT, Sampaio SC, Curi R.Suppressive effect of short-chain fatty acids on production of proinflammatorymediators by neutrophils. J Nutr Biochem 2011;22:849–85512. Brown AJ, Goldsworthy SM, Barnes AA, et al. The Orphan G protein-coupledreceptors GPR41 and GPR43 are activated by propionate and other short chaincarboxylic acids. J Biol Chem 2003;278:11312–1131913. Le Poul E, Loison C, Struyf S, et al. Functional characterization of humanreceptors for short chain fatty acids and their role in polymorphonuclear cellactivation. J Biol Chem 2003;278:25481–2548914. De Vadder F, Kovatcheva-Datchary P, Goncalves D, et al. Microbiota-generated metabolites promote metabolic benefits via gut-brain neural circuits.Cell 2014;156:84–9615. Gao Z, Yin J, Zhang J, et al. Butyrate improves insulin sensitivity and in-creases energy expenditure in mice. Diabetes 2009;58:1509–151716. Kondo T, Kishi M, Fushimi T, Ugajin S, Kaga T. Vinegar intake reduces bodyweight, body fat mass, and serum triglyceride levels in obese Japanese subjects.Biosci Biotechnol Biochem 2009;73:1837–184317. Lin HV, Frassetto A, Kowalik EJ Jr, et al. Butyrate and propionate protectagainst diet-induced obesity and regulate gut hormones via free fatty acid re-ceptor 3-independent mechanisms. PLoS One 2012;7:e3524018. Sakakibara S, Yamauchi T, Oshima Y, Tsukamoto Y, Kadowaki T. Aceticacid activates hepatic AMPK and reduces hyperglycemia in diabetic KK-A(y) mice.Biochem Biophys Res Commun 2006;344:597–60419. Demigné C, Morand C, Levrat MA, Besson C, Moundras C, Rémésy C. Effectof propionate on fatty acid and cholesterol synthesis and on acetate metabolismin isolated rat hepatocytes. Br J Nutr 1995;74:209–21920. Tang Y, Chen Y, Jiang H, Nie D. Short-chain fatty acids induced autophagyserves as an adaptive strategy for retarding mitochondria-mediated apoptotic celldeath. Cell Death Differ 2011;18:602–61821. He W, Barak Y, Hevener A, et al. Adipose-specific peroxisome proliferator-activated receptor g knockout causes insulin resistance in fat and liver but not inmuscle. Proc Natl Acad Sci U S A 2003;100:15712–1571722. Oosterveer MH, van Dijk TH, Tietge UJ, et al. High fat feeding induceshepatic fatty acid elongation in mice. PLoS One 2009;4:e606623. Hirschey MD, Shimazu T, Goetzman E, et al. SIRT3 regulates fatty-acidoxidation by reversible enzyme deacetylation. Nature 2010;464:121–12524. Grefhorst A, van Dijk TH, Hammer A, et al. Differential effects ofpharmacological liver X receptor activation on hepatic and peripheral insulinsensitivity in lean and ob/ob mice. Am J Physiol Endocrinol Metab 2005;289:E829–E83825. Bligh EG, Dyer WJ. A rapid method of total lipid extraction and purification.Can J Biochem Physiol 1959;37:911–91726. Demoz A, Garras A, Asiedu DK, Netteland B, Berge RK. Rapid method forthe separation and detection of tissue short-chain coenzyme A esters by

reversed-phase high-performance liquid chromatography. J Chromatogr B Bi-omed Appl 1995;667:148–15227. Miller RA, Chu Q, Xie J, Foretz M, Viollet B, Birnbaum MJ. Biguanidessuppress hepatic glucagon signalling by decreasing production of cyclic AMP.Nature 2013;494:256–26028. van Vlies N, Ruiter JP, Doolaard M, Wanders RJ, Vaz FM. An improvedenzyme assay for carnitine palmitoyl transferase I in fibroblasts using tandemmass spectrometry. Mol Genet Metab 2007;90:24–2929. Mildaziene V, Nauciene Z, Baniene R, Grigiene J. Multiple effects of2,29,5,59-tetrachlorobiphenyl on oxidative phosphorylation in rat liver mitochon-dria. Toxicol Sci 2002;65:220–22730. Gnaiger E, Kuznetsov AV, Schneeberger S, et al. Mitochondria in the cold. InLife in the Cold. Heldmaier G, Klingenspor M, Eds. Berlin, Springer-Verlag, 2000,p. 431–44231. Zebisch K, Voigt V, Wabitsch M, Brandsch M. Protocol for effective differ-entiation of 3T3-L1 cells to adipocytes. Anal Biochem 2012;425:88–9032. Fujita H, Endo A, Shimizu K, Nagamori E. Evaluation of serum-free differ-entiation conditions for C2C12 myoblast cells assessed as to active tensiongeneration capability. Biotechnol Bioeng 2010;107:894–90133. Lage R, Diéguez C, Vidal-Puig A, López M. AMPK: a metabolic gaugeregulating whole-body energy homeostasis. Trends Mol Med 2008;14:539–54934. McGarry JD, Mills SE, Long CS, Foster DW. Observations on the affinity forcarnitine, and malonyl-CoA sensitivity, of carnitine palmitoyltransferase I in ani-mal and human tissues. Demonstration of the presence of malonyl-CoA in non-hepatic tissues of the rat. Biochem J 1983;214:21–2835. Carling D, Mayer FV, Sanders MJ, Gamblin SJ. AMP-activated protein ki-nase: nature’s energy sensor. Nat Chem Biol 2011;7:512–51836. Rousset S, Alves-Guerra MC, Mozo J, et al. The biology of mitochondrialuncoupling proteins. Diabetes 2004;53(Suppl. 1):S130–S13537. Jones JR, Barrick C, Kim KA, et al. Deletion of PPARgamma in adiposetissues of mice protects against high fat diet-induced obesity and insulin re-sistance. Proc Natl Acad Sci U S A 2005;102:6207–621238. Norris AW, Chen L, Fisher SJ, et al. Muscle-specific PPARgamma-deficientmice develop increased adiposity and insulin resistance but respond to thiazo-lidinediones. J Clin Invest 2003;112:608–61839. Kelly LJ, Vicario PP, Thompson GM, et al. Peroxisome proliferator-activatedreceptors g and a mediate in vivo regulation of uncoupling protein (UCP-1, UCP-2, UCP-3) gene expression. Endocrinology 1998;139:4920–492740. Goto T, Lee JY, Teraminami A, et al. Activation of peroxisome proliferator-activated receptor-alpha stimulates both differentiation and fatty acid oxidation inadipocytes. J Lipid Res 2011;52:873–88441. Yamauchi T, Kamon J, Waki H, et al. The mechanisms by which bothheterozygous peroxisome proliferator-activated receptor gamma (PPARgamma)deficiency and PPARgamma agonist improve insulin resistance. J Biol Chem2001;276:41245–4125442. Yamauchi T, Waki H, Kamon J, et al. Inhibition of RXR and PPARgammaameliorates diet-induced obesity and type 2 diabetes. J Clin Invest 2001;108:1001–101343. den Besten G, van Eunen K, Groen AK, Venema K, Reijngoud DJ, BakkerBM. The role of short-chain fatty acids in the interplay between diet, gut mi-crobiota, and host energy metabolism. J Lipid Res 2013;54:2325–234044. Cheng BO, Trimble RP, Illman RJ, Stone BA, Topping DL. Comparative ef-fects of dietary wheat bran and its morphological components (aleurone andpericarp-seed coat) on volatile fatty acid concentrations in the rat. Br J Nutr 1987;57:69–7645. Murase M, Kimura Y, Nagata Y. Determination of portal short-chain fattyacids in rats fed various dietary fibers by capillary gas chromatography.J Chromatogr B Biomed Appl 1995;664:415–42046. Topping DL, Clifton PM. Short-chain fatty acids and human colonic function:roles of resistant starch and nonstarch polysaccharides. Physiol Rev 2001;81:1031–1064

diabetes.diabetesjournals.org den Besten and Associates 2407

47. Fell DA, Thomas S. Physiological control of metabolic flux: the requirementfor multisite modulation. Biochem J 1995;311:35–3948. Diano S, Horvath TL. Mitochondrial uncoupling protein 2 (UCP2) in glucoseand lipid metabolism. Trends Mol Med 2012;18:52–5849. Brand MD, Esteves TC. Physiological functions of the mitochondrial un-coupling proteins UCP2 and UCP3. Cell Metab 2005;2:85–9350. Vozza A, Parisi G, De Leonardis F, et al. UCP2 transports C4 metabolites outof mitochondria, regulating glucose and glutamine oxidation. Proc Natl Acad SciU S A 2014;111:960–96551. Hong YH, Nishimura Y, Hishikawa D, et al. Acetate and propionate shortchain fatty acids stimulate adipogenesis via GPCR43. Endocrinology 2005;146:5092–509952. Kimura I, Ozawa K, Inoue D, et al. The gut microbiota suppresses insulin-mediated fat accumulation via the short-chain fatty acid receptor GPR43. NatCommun 2013;4:1829

53. den Besten G, Havinga R, Bleeker A, et al. The short-chain fatty aciduptake fluxes by mice on a guar gum supplemented diet associate withamelioration of major biomarkers of the metabolic syndrome. PLoS One 2014;9:e10739254. Cock TA, Houten SM, Auwerx J. Peroxisome proliferator-activated receptor-gamma: too much of a good thing causes harm. EMBO Rep 2004;5:142–14755. Higgins LS, Depaoli AM. Selective peroxisome proliferator-activated re-ceptor gamma (PPARgamma) modulation as a strategy for safer therapeuticPPARgamma activation. Am J Clin Nutr 2010;91:267S–272S56. Suokas A, Kupari M, Heikkilä J, Lindros K, Ylikahri R. Acute cardio-vascular and metabolic effects of acetate in men. Alcohol Clin Exp Res 1988;12:52–5857. Al-Lahham SH, Peppelenbosch MP, Roelofsen H, Vonk RJ, Venema K. Bi-ological effects of propionic acid in humans; metabolism, potential applicationsand underlying mechanisms. Biochim Biophys Acta 2010;1801:1175–1183

2408 SCFAs Prevent Metabolic Disorders via PPARg Diabetes Volume 64, July 2015