Increasing evidence suggests that the human gut microbiota and its products are key players in host metabolism, body weight, and insulin sensitivity, thereby contributing to the etiology of obesity and related disorders (1). The gut microbiota can ferment indigestible nutrients into short-chain fatty acids (SCFA), of which acetate, propionate and butyrate are the most abundant (2). Of note, these SCFA can be taken up by the epithelial lining of the gut and released into the blood stream (3). They may thereby act as important signaling molecules between gut microbiota and host physiology, by exerting effects on energy and substrate metabolism such as on adipogenesis and lipolysis in the adipose tissue (4).

Disturbances in adipose tissue function, characterized by a reduced capacity to store lipids, seem to play a major role in the development of insulin resistance and type 2 diabetes mellitus in humans (5). Under normal healthy conditions, the adipose tissue is an important buffering organ for daily postprandial fatty acid (FA) fluxes when endogenous lipolysis is inhibited. The adipose tissue thereby prevents excessive supply of lipids to nonadipose tissues such as liver, skeletal muscle, and pancreas. This buffering action may be impaired under obese insulin-resistant conditions (6, 7), resulting in increased circulating lipids and ectopic fat storage in nonadipose tissues, thereby provoking disturbances in insulin signaling and substrate metabolism (8, 9).

SCFA may affect adipose tissue lipid buffering capacity by affecting intracellular lipolysis, the process of hydrolysis of stored triacylglycerol into one molecule of glycerol and three FA molecules, and may thereby affect circulating lipid concentrations (4). Indeed, already decades ago, a decrease in plasma free fatty acids (FFA) after a single oral acetate ingestion was observed, pointing to an antilipolytic role of acetate (10). The more recent identification of the two pertussis toxin (PTX)-sensitive inhibitory G (Gi) protein-coupled receptors (GPRs) for SCFA, free fatty acid receptor 3 (FFAR3, also known as GPR41), and FFAR2 (also known as GPR43) in human adipose tissue (11), has led to renewed interest in the lipolytic properties of SCFA. In addition, a direct association between the SCFA/FFAR signaling pathway and lipolytic activity in murine adipocytes was recently discovered (12). Treatment of differentiated murine 3T3-L1 adipocytes with acetate and propionate in a range between 0.1 and 0.3 mmol/L exhibited FFAR2 activation and a reduction in intracellular lipolytic activity as assessed by a decreased release of glycerol in the culture medium (12). In contrast, incubation of 3T3-L1 adipocytes with supraphysiological concentrations of propionate (20 mmol/L) or butyrate (5 mmol/L) resulted in enhanced glycerol release (13). However, in murine 3T3-L1 adipocytes only FFAR2, and not FFAR3, is expressed (14–17). Therefore, it remains to be determined, whether these findings extend to human adipocytes, in which both FFAR3 and FFAR2 are expressed (11, 18). Thus, further investigation on the role of SCFA in human adipocyte lipolysis is urgently warranted.

We recently observed that colonic infusions of mixtures of acetate, propionate and butyrate, in ratios and concentrations that can be reached after dietary fiber intake, attenuated whole-body lipolysis in overweight normoglycaemic men (19). Therefore, the aim of the present study was to elucidate whether an altered intracellular adipocyte lipolytic rate is responsible for the antilipolytic effect of SCFA found in vivo, as well as to further investigate underlying mechanisms. Hence, we investigated the in vitro effects of incubation with SCFA mixtures and single SCFA on intracellular lipolysis in a human white adipocyte model, the human multipotent adipose tissue-derived stem (hMADS) cells. To study whether these effects are mediated via Gi-coupled receptors, we investigated the effect of SCFA on lipase activation and performed PTX-mediated inhibition of FFARs.

This study provides new insight in the effects of SCFA on human adipocyte lipolysis. We previously showed that acute colonic administration of three physiological-relevant SCFA mixtures, and subsequent elevated circulating acetate concentrations, reduced circulating glycerol concentration in overweight males, indicative of a reduced whole-body lipolysis (19). However, to further investigate whether the reduction in whole-body lipolysis was related to a putative SCFA effect on white adipocyte intracellular lipolysis, we performed several in vitro experiments using our validated hMADS adipocyte model. Our present in vitro study in hMADS adipocytes demonstrated that mainly acetate had antilipolytic effects, which was accompanied by a reduced phosphorylation of HSL (at SER650). Incubation with the Gi inhibitor PTX prevented the acetate-mediated antilipolytic effect, suggesting that this antilipolytic effect may be mediated through an acetate-FFAR-coupled signaling pathway (for an schematic overview see Figure 6).

This study demonstrated that SCFA mixtures at physiological (1 µmol/L) and more supraphysiological (1 mmol/L) concentrations attenuate intracellular lipolysis in human adipocytes. Furthermore, by subsequently incubating human adipocytes with single SCFA, we demonstrated that the intestinally and systemically most abundant SCFA acetate seems to be the main driver of this antilipolytic effect. Acetate is the most abundant circulating SCFA and is found in serum and plasma mean concentrations varying from 5 up to 220 µmol/L, depending on the nutritional status (3, 22–24). Propionate and butyrate are found at much lower maximal mean concentrations of 13 and 12 µmol/L, respectively (3, 22, 23). However, no human data are available reflecting SCFA concentrations that reach the adipose tissue via their capillaries. Based on the scarcely available data on circulating SCFA concentrations, an acetate to propionate to butyrate ratio of 80:10:10 and 60:20:20 could resemble physiologically circulating concentrations and the SCFA concentrations used in this study might be in a physiological (1 µmol/L) to supraphysiological (1 mmol/L) range. Interestingly, the most pronounced effects on lipolysis were found with an acetate concentration of 1 µmol/L, which thus seems to be lower than circulating concentrations. Therefore it would be of major interest to measure actual acetate concentrations in adipose tissue capillaries or interstitial fluids. To the knowledge of the authors, no data on this are available, which would be very interesting to detect via for example microdialysis techniques.

In addition, we observed that the antilipolytic effect of acetate was accompanied by a reduced phosphorylation of HSL at the serine 650, a major protein kinase A (PKA) regulatory site. In accordance with this observation, Aberdein et al. (25) indicated that treatment of murine 3T3-L1 adipocytes with supraphysiological concentrations (4 mmol/L) of sodium acetate reduced the β-adrenergic receptor stimulated non-esterified FA release and decreased HSL phosphorylation at another PKA regulatory site (serine 563) (25). Ge et al. (12) showed that treatment of 3T3-L1 adipocytes with acetate and propionate in a range between 0.1 and 0.3 mmol/L reduced the basal and β-adrenergic receptor stimulated intracellular lipolytic activity as assessed by a decreased release of glycerol in the culture medium (12). In contrast to the antilipolytic effect of acetate, we observed a slightly increased basal lipolytic response after butyrate treatment in hMADS adipocytes. Comparable results were reported by Rumberger et al. (13) showing an increased basal lipolytic response (glycerol release) following incubation of murine 3T3-L1 adipocytes with 5 mmol/L butyrate (13). However, the underlying mechanism of this lipolytic effect of butyrate needs further investigation in hMADS cells. Together, these results suggest that acetate is the main driver for the antilipolytic effects of SCFA, which is accompanied by an attenuated HSL phosphorylation in both, murine and human adipocytes.

Finally, we observed that the antilipolytic effect of acetate might be FFAR dependent. We first showed in accordance to other reports in human adipocyte models (18) that both FFAR3 and FFAR2 transcripts and protein are expressed in our hMADS adipocyte model, and that both increased during adipogenic differentiation. Furthermore, we showed that the effects of acetate are abrogated with co-incubation of PTX. PTX is a well-known FFAR inhibitor and irreversible inactivates Gi proteins. Thereby, these data suggest that the acetate effects were mediated via a Gi protein receptor-PKA pathway. In line, a previous study in murine 3T3-L1 adipocytes has shown that the lipolytic effect of acetate was mediated by activation of FFAR2 (12). However, further investigations are warranted to elucidate whether SCFA effects on human intracellular adipocyte lipolysis are mediated mainly via FFAR2 and/or FFAR3, using specific human knockdown models. In particular the role of FFAR3/2 protein should be further investigated via the use of knockdown and overexpression in human adipocyte models.

Furthermore, evidence is increasing that metabolic phenotype should be considered in future lipolysis studies. Present literature provides evidence that obesity-related metabolic disturbances, such as insulin resistance, are linked to differences in circulating acetate levels and acetate-induced metabolic responses. For example, in our acute studies (19, 24) we included overweight and obese individuals with average fasting acetate concentrations of approximately 20–50 µmol/L, whereas in another study of our group with insulin resistant obese individuals markedly higher acetate concentrations of approximately 70–90 µmol/L have been found (26). In addition, a kinetic study showed that the acetate clearance rate is lower and the half-life is longer in type 2 diabetic patients when compared with healthy normoglycaemic controls (27). This suggests a disturbed uptake and/or metabolism of acetate, which might be relevant to elicit acetate-induced metabolic effects and cell signaling in peripheral tissues. Furthermore, there are indications that overweight insulin resistant compared to normoglycaemic individuals have lower acetate-induced antilipolytic responses on a whole-body level (28). An acute study demonstrated that intravenously administered acetate resulted in a greater FFA fall and rebound in healthy adults compared with hyperinsulinaemic individuals (28). Therefore, comparing SCFA-mediated inhibition of the lipolytic response and intracellular signaling mechanism in adipocytes derived from normoglycaemic, insulin sensitive versus metabolically more compromised donors is of major interest. Here, we included cells from human adults with a wide range of BMI and glucometabolic status, therefore we did not distinguish between metabolic phenotypes, which is as limitation.

Nevertheless, this study has clinical implications. If the observed results can be translated into long-term in vivo metabolic effects, increased systemic acetate availability might improve human white adipose tissue lipid buffering capacity and reduce adipose tissue lipid spillover. This could ultimately result in attenuated ectopic fat accumulation and improved insulin action in insulin sensitive tissues such as skeletal muscle, pancreas and liver, preventing insulin resistance. The present study showed that acetate induces a partial inhibition of intracellular lipolysis during basal conditions. Interestingly, combined data derived from rodents and humans demonstrated that a comparable partial inhibition of intracellular lipolysis has beneficial effects on insulin sensitivity without affecting adipose tissue mass in the longer term (29, 30). In addition to elevated basal lipolysis, β-adrenergic receptor agonist sensitivity is blunted in obese insulin resistant individuals (31–33), which poses the question whether a further decrease in β-adrenergically mediated lipolysis by SCFA is positive with respect to metabolic health. Thus further research including isoprenaline concentration–response curves are needed to ascertain SCFA-induced changes in efficiency or potency of β-adrenergic receptor agonists in different metabolic phenotypes.

With the present in vitro study using our human adipocyte model, we explored a mechanism that might explain the previously in vivo observed antilipolytic effect of physiologically relevant SCFA mixtures (19). However, other mechanisms, which might also contribute to the SCFA-induced antilipolytic effect on whole-body level, could not be excluded here. For example, in parallel to acetate concentrations, circulating peptide YY (PYY) concentrations were increased after the colonic infusions of SCFA mixtures in our in vivo experiment (19). Indeed, PYY was previously recognized for its antilipolytic property in human adipocytes (34). In addition, an intriguing study in rodents showed that SCFA can influence energy homeostasis including lipolysis via dorsal sympathetic ganglions and spinal pathways (35).

In conclusion, we demonstrated that in particular the colonic and peripheral most abundant SCFA acetate plays an important role in the regulation of human adipose tissue lipolysis. We showed that the luminal and systemically most abundant SCFA acetate was mainly responsible for the antilipolytic response, via FFAR-mediated attenuation of HSL phosphorylation in human adipocytes. Indicating that the modulation of colonic and systemic acetate might be a target to prevent or improve insulin resistance in human. Therefore, future studies should focus on increasing nutritional strategies to enhance circulating acetate availability, for example via supplementation of specific acetogenic fibers, to improve human lipid metabolism.

hMADS cells, a validated human white adipocyte model to study lipid metabolism, were obtained from human subcutaneous adipose tissue biopsies from male donors. The male donors participated in two different clinical trials (http://ClinicalTrials.gov, NCT02241421 and NCT02598544). The study protocols were approved by the Medical Ethical Committee Jessa hospital, Hasselt and Hasselt University, Belgium, and by the Medical Ethical Committee of Maastricht University Medical Center, Maastricht, The Netherlands. All procedures were according to the declaration of Helsinki (revised version, October 2008).

JE, EB, and EC were responsible for the study concept and design, analysis and interpretation of the data, and critical revision of the manuscript for important intellectual content. MH, YE, and NH generated data. JE and EC acquired all data, completed statistical analysis, and drafted the manuscript. CB critically revised the manuscript. EB obtained funding and supervised the study.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.