Effects of Sodium Butyrate and Its Synthetic Amide Derivative on Liver Inflammation and Glucose Tolerance in an Animal Model of Steatosis Induced by High Fat Diet

Background & Aims Nonalcoholic fatty liver disease (NAFLD) is the most common form of chronic liver disease. Insulin resistance (IR) appears to be critical in its pathogenesis. We evaluated the effects of sodium butyrate (butyrate) and its synthetic derivative N-(1-carbamoyl-2-phenyl-ethyl) butyramide (FBA) in a rat model of insulin resistance and steatosis induced by high-fat diet (HFD). Methods After weaning, young male Sprague-Dawley rats were divided into 4 groups receiving different diets for 6 weeks: 1. control group (standard diet); 2. HFD; 3. HFD plus butyrate (20 mg/kg/die) and 4. HFD plus FBA (42.5 mg/Kg/die, the equimolecular dose of butyrate). Liver tissues of the rats were analyzed by Western blot and real-time PCR. Insulin resistance, liver inflammation and Toll-like pattern modifications were determined. Results Evaluation of these two preparations of butyrate showed a reduction of liver steatosis and inflammation in HFD fed animals. The compounds showed a similar potency in the normalisation of several variables, such as transaminases, homeostasis model assessment for insulin resistance index, and glucose tolerance. Both treatments significantly reduced hepatic TNF-α expression and restored GLUTs and PPARs, either in liver or adipose tissue. Finally, FBA showed a higher potency in reducing pro-inflammatory parameters in the liver, via suppression of Toll-like receptors and NF-κB activation. Conclusions Our results demonstrated a protective effect of butyrate in limiting molecular events underlying the onset of IR and NAFLD, suggesting a potential clinical relevance for this substance. In particular, its derivative, FBA, could represent an alternative therapeutic option to sodium butyrate, sharing a comparable efficacy, but a better palatability and compliance.


Introduction
A close association between non-alcoholic fatty liver disease (NAFLD) and several findings indicative of insulin resistance (IR) and metabolic disorders has long been reported. The liver produces and is exposed to various types of lipids, such as fatty acids, cholesterol and triglycerides via the portal vein from the diet and visceral adipose tissues. The liver and adipose tissue jointly participate in maintaining glucose and lipid homeostasis through the secretion of several humoral factors and/or neural networks. Perturbation in the inter-tissue communications may be involved in the development of IR and diabetes. An excessive free fatty acids (FFAs) flux into the liver via the portal vein may cause fatty liver disease and hepatic IR. However, the initial events triggering the development of IR and its causal relations with dysregulation of glucose and fatty acids metabolism remain unclear.
It has been suggested that the blood glucose-and lipidlowering effects of soluble dietary fibres may be related in part to short chain fatty acids (SCFAs) generated during anaerobic microbial fermentation [1,2]. Among SCFAs, butyrate constitutes one of the major products derived from intestinal fermentation of undigested dietary carbohydrates, specifically resistant starches and dietary fibres, but also in a minor part by dietary and endogenous proteins. After butyrate uptake by the colon, it is metabolized in part by the colonocytes, and the remaining fraction reaches the liver via the portal vein. The colonocytes absorb butyrate through different mechanisms of apical membrane uptake, including non-ionic diffusion, SCFA/HCO 3 -exchange, and active transport by monocarboxylic acid transporter (i.e. MCT1) [3]. In particular, butyrate is able to exert a powerful pro-absorptive stimulus on intestinal NaCl transport and an anti-secretory effect towards Clsecretion [4,5].
The effects exerted by butyrate are multiple and involve several distinct mechanisms of action. Its well-known epigenetic mechanism, is the hyperacetylation of histones by inhibiting class I and class II histone deacetylases (HDAC), that results in the regulation of gene expression and in the control of cell fate [6,7]. HDAC regulates gene transcription through modification of chromatin structure by acetylation of proteins, including not only histone proteins, but also transcription factors (i.e. NF-κB, p53 and NFAT) [8]. Butyrate also acts as signal molecules, targeting their G protein-coupled receptors Free Fatty Acid Receptor 2 (FFAR2, GPR43) and FFAR3 (GPR41) [9]. At intestinal level, butyrate exerts multiple effects, such as the prevention and inhibition of colonic carcinogenesis, the improvement of inflammation, oxidative status, epithelial defense barrier, and the modulation of visceral sensitivity and intestinal motility. At the extraintestinal level, potential fields of application for butyrate seem to be the treatment of different pathologies, including metabolic diseases, such as hypercholesterolemia, obesity, IR, and ischemic stroke [10].
Recently, it has been demonstrated that dietary supplementation of butyrate can prevent or treat diet-induced IR in mice [11]. The mechanism of butyrate action was related to promotion of energy expenditure and induction of mitochondrial function through stimulation of peroxisome proliferator activated-receptor (PPAR) γ coactivator (PGC)-1α. Moreover, activation of AMP kinase (AMPK) and inhibition of HDAC could contribute to PGC-1α regulation by butyrate. More recently Li et al [12] demonstrated that fibrobloblast growth factor (FGF) 21, which plays an important role in lipid metabolism [13], is induced by butrate and involved in the stimulation of fatty acid β-oxidation in liver.
Some butyrate-based products are marketed but their spread is still very limited and greatly understaffed in view of the wide spectrum of possible indications especially in chronic diseases where it is possible to predict a lasting use of these compounds. The unpleasant taste and odour make extremely difficult the oral administration of butyrate, these difficulties are even more remarkable in children where the administration is complicated. Thus, new formulations of butyrate with a better palatability, which can be easily administered orally, are needed. The purpose of this study is to investigate the efficacy of sodium butyrate (butyrate) and of its more palatable derivative, the N-(1-carbamoyl-2-phenyl-ethyl) butiramide (FBA), in a rat model of NAFLD induced by high fat diet. We hypothesized that orally administered butyrate compounds, could attenuate steatosis and liver injury, with reduction of inflammatory responses via suppression of Toll-like receptors (TLRs) through downregulation of NF-κB activation.

Effects of butyrate and FBA on liver steatosis and serum parameters
Liver sections from rats fed with HFD demonstrated significant hepatic damage in comparison with standard diet (STD) fed animals. As depicted in Figure 1 A, HFD livers showed foci of mixed inflammatory cell infiltration, evidenced by arrows, and hepatocyte necrosis or apoptosis throughout the lobule. Scattered inflammation and occasionally apoptotic nuclei were observed. No alterations were shown in liver of the rats fed with the STD. Furthermore, HDF rats showed grade 3 hepatic steatosis with a histological pattern characterized by microvesicular steatosis. The hepatocytes showed the cytoplasm filled with small vacuoles which were uniform in size and smaller than the centrally located nucleous. Steatosis affected most of the hepatocytes (Figure 1 B). In animals treated with equimolecular doses of butyrate (sodium butyrate or FBA), liver inflammatory damage appears reduced and steatosis was graded as grade 1 with a microvesicular pattern of lipid accumulation distributed in perivenular and periportal region. This effect was also associated with a reduction of triglycerides content which was significantly enhanced by HFD (Figure 1 C).
In HFD-fed animals TNF-α, a cytokine involved in the development of IR, was significantly increased, butyrate and in particular FBA reduced hepatic expression of this cytokines (Figure 1 D). Similarly the mRNA concentration of chemokine MCP-1 and a specific marker of mature macrophage f4/80 were increased in HFD rats and normalized by butyrate treatments (Figure E and F). Moreover, HFD induced the expression TGF-β1, an early marker of the fibrotic process ( Figure 1 G), which was significantly down-regulated in the butyrate-treated group (P < 0.01) and more significantly by FBA (P < 0.001). We also evaluated α-SMA and pro-collagen type 1 transcript by real time-PCR but it did not show a significant modification after 6 weeks of HFD. Biochemical serum parameters are reported in table 1. Circulating levels of AST, ALT and cholesterol resulted increased at 6 weeks of HFD, while LDL and triglycerides showed a trend of increase. All these parameters were reduced by butyrate and FBA. Compared with animals fed with STD diet, HFD rats showed a marked increase in fasting glucose without change in insulin levels. Both butyrate and FBA prevented glucose alteration. Homeostasis model assessment for insulin resistance (HOMA-IR) was 25% and 30% lower in butyrate and FBA groups, respectively.
No difference in body weight was shown among all groups, weight gain of HFD fed animals did not change after 6 weeks neither modified by butyrate and FBA treatments (data not shown). strongly up-regulated by activated macrophages or other liver cell types (including Kupffer cells and hepatic stellate cells), participating in liver injury. Beyond TNF-α, also IL-6 is involved in metabolic impairment, initiating the pathogenesis of hepatic IR. As shown in Figure 2 A and B, HFD induced a significant increase in hepatic IL-1β and IL-6 mRNAs, and both treatments significantly prevented the transcription of these genes. Moreover, the high-fat diet determined an increase in liver proinflammatory enzymes, COX-2 and iNOS (Figure 2 C and D) and butyrate and FBA reduced the expression of these proteins. In agreement both treatments significantly prevented the increase of nuclear content of p50 NF-κB related to a reduction of the inhibitory cytosolic protein IκBα, showing the inhibition of NF-κB activation (Figure 2 E and F).

Effect of butyrate and FBA on hepatic Toll-like receptors pattern
The activation of TLRs family, especially TLR4, by inflammatory cytokines or increased NEFA could modulate insulin sensitivity [14]. Recently, it has been hypothesized that FFA-related high-mobility group box 1 (HMGB1) release mediates the activation of TLR4 signaling in hepatocytes and plays an essential part in the early stage of NAFLD induced by  Effects of butyrate and its derivative in NAFLD PLOS ONE | www.plosone.org HFD [15]. Here, HMGB-1 mRNAs were strongly up-regulated by HFD feeding while both butyrate, and in particular FBA, determined a significant reduction of these levels (Figure 3 A). As shown in Figure 3B and C, HFD induced the mRNA increase of TLR4 and its co-receptor CD14 in liver. FBA and, to a lesser extent, butyrate inhibited these effects. Notably, TLR2 and TLR9, which are able to detect lipoproteins and unmetilated CpG-containing DNA, respectively, were also upregulated by HFD and reduced by two treatments, with a great effectiveness of the FBA ( Figure 3D and E).

Effect of butyrate and FBA on glucose homeostasis and insulin resistance
To confirm the effects of butyrate and its derivative on insulin resistance in HFD-fed rats, glucose tolerance and insulin signaling were analyzed. In oral glucose tolerance test, blood samples were collected sequentially before, and 30, 60, 90, and 120 min after the glucose load, HFD induced an altered response to glucose at all time points paralleled with ST ( Figure   FBA groups exhibited improved response to glucose at all time points, in particular FBA raised a significant reduction of glycemia 90 min after glucose load (Figure 4 A).
The insulin receptor is acted on an unique group of regulatory proteins, including protein tyrosine phosphatase 1B (PTP1B), which is considered as a major negative regulator of insulin receptor signaling. As depicted in Figure 4C the HFD feeding determined an over-expression of this enzyme while both treatments showed a significant reduction of this protein in hepatic tissue. Moreover, we showed an increase in SOCS3 protein expression in hepatic tissues from HFD rats, that was significantly inhibited by butyrate and FBA (P<0.05, Figure 4 D) confirming the restoration of insulin signaling.
According with the inflammatory alterations, other modifications demonstrated metabolic impairment and insulin tissue resistance in HFD fed rats. The basal level of PPAR-α, which regulates fatty acid β-oxidation and catabolism, was detected in liver homogenates from STD fed animals. PPAR-α decreased in liver from HFD rats (P<0.01) and restored by butyrate or FBA (P<0.01 and P<0.05, respectively, Figure 5A). Very recently, it has been demonstrated that FGF21, a cytokine/hormone that plays an important role in the regulation of lipid and carbohydrate metabolism, was induced by butyrate [12]. In liver extracts, FGF21 was significantly reduced by HFD and its decreased expression was slightly prevented by butyrate and significantly by FBA ( Figure 5B). Moreover, GLUT2, a glucose-sensitive gene in liver cells [16], was markedly reduced by HFD and this effect was significantly prevented by butyrate and FBA (and Figure 5C).
It is well known that adipose tissue, behind its reserve and secretory role, represents an important insulin-sensitive target tissue contributing to glucose homeostasis. Here, the expression of PPAR-γ, a ligand-dependent transcription factor, was lower in HFD compared with STD group, whereas both treatments significantly increased it ( Figure 6A). Consistently, the PPAR-γ transcription coactivator, (PGC)-1α, was significantly reduced in HFD group, while both treatments prevented this effect ( Figure 6B). It is well known that this  coactivator controls energy metabolism interacting with several transcription factors, including PPAR-α, and its reduction is associated with mitochondrial dysfunction, reduction in fatty acid oxidation and insulin resistance [17]. Moreover, in adipose tissue butyrate and FBA were also able to restore GLUT-4 protein expression, that was deeply reduced by HFD (P<0.05, respectively, Figure 6C).

Discussion
Metabolic and anti-inflammatory effects of butyrate and its derivative, phenylalanine-butyramide (FBA), were examined in this study in a HFD-induced rat model of hepatic steatosis and insulin resistance. Sodium butyrate shows considerable drawbacks: it has fairly strong hygroscopicity and butirric nauseating smell, associated to a poor compliance related to extremely unpleasant taste and epigastric discomfort after oral assumption. On the other hand, the conjugate of butyrate to phenylalanine, FBA, has allowed to obtain a molecule with chemical-physical characteristics suitable for an easier oral administration compared to butyrate.
The most important observation is that both treatments prevented the impairment of glucose homeostasis and the development of insulin resistance. In our model there are two major hallmarks of insulin resistance: hepatic inflammatory process and alteration of glucose tolerance related to an impairment of insulin signaling. The improvement in insulin sensitivity may be, at least in part, a consequence of antiinflammatory effects of these compounds in this nutritional model. Insulin controls whole body glucose homeostasis with several mechanisms, such as the promotion of glucose disposal in sensitive target tissues, such as liver and fat.
Our study shows that both treatments are able to prevent liver inflammation and damage, steatosis, the onset of IR, and imbalance of TLRs pattern in the early stage of NAFLD. In particular, we showed a reduction in hepatic lipid accumulation mainly significant in FBA treated rats, together with a reduction of the inflammatory infiltrates. The evaluation of the fibrotic process revealed no appreciable modification in HFD fed rats, and consequently no modification by both butyric treatments (data not shown), even if TGF-β, an early pro-fibrogenic marker, was up-regulated in liver from rats on HFD, and markedly reduced by both treatments. TGF-β is considered the most powerful mediator of hepatic stellate cell activation and kupffer cells are a main source of TGF-β in the liver, promoting, collectively to inflammatory cytokines and oxidative stress, later fibrosis [18].
Moreover, both treatments normalize the hepatic markers of steatosis, and preserve glucose tolerance, reducing fasting glucose and modulating HOMA-IR. It is now clear that TNF-α and IL-6 represent crucial effectors of IR, that link liver inflammatory process to hormonal and metabolic alterations [19,20]. In our experimental model, both compounds reduce the above reported cytokines in parallel with a lower expression of PTP1B and SOCS3 inhibitory proteins, suggesting two molecular mechanisms of insulin sensitization. In accordance, recently it has been highlighted how the prototypic phoshotyrosine-specific phoshatase PTP1B dephosphorylates the insulin receptor and downstream IRS-1/2 proteins. Its inhibition enhances insulin signaling and attenuates insulin resistance both in conventional and non conventional insulinresponsive tissues [21], indicating PTP1B as a target for the development of novel therapeutics for diabetes and obesity. On the other hand, SOCS family, including SOCS3, associates with the insulin receptor and inhibits its signaling through ubiquitin-induced degradation of IRS-1. In particular, the induction of SOCS3 in liver may be an important mechanism of IL-6-mediated insulin resistance [19]. Moreover, both treatments reduce the activation of NF-κB pathway induced by HFD, reducing IκBα degradation and inhibiting p50, with a subsequent reduction not only in cytokine transcription (TNFα, IL-6 and IL-1β), but also in inflammatory enzymes (COX-2 and iNOS). Furthermore, we found that HFD increases macrophage infiltration markers, MCP-1 and F4/80, and both butyrate treatments are able to reduce these parameters, in particular reducing F4/80-positive macrophage, indicative of M1state, associated with inflammation and tissue injury [22].
Here, we demonstrate that both treatments reduce inflammation and its mediators inhibiting NF-κB activation arguably through the down-regulation of TLR pattern in the liver, which are involved in bacterial sensing and are crucial for "liver tolerance". We demonstrate that FBA and, to a lesser extent, butyrate are able to limit the increased transcription and expression of TLRs induced by HFD. The increased expression of hepatic TLRs confirms a greater exposure of the liver to ligands for these receptors deriving from the intestine (i.e. pathogen-associated molecular patterns or PAMPs and endogenous damage-associated molecular patters or DAMPs), including the widely expressed nucleus protein HMGB1. Recently, it has been hypothesized that FFA-related HMGB1 release mediates the activation of TLR4 signaling in hepatocytes and plays an essential part in the early stage of NAFLD induced by HFD [15]. Here, hepatic HMGB1 transcription was strongly up-regulated by HFD feeding while both butyrate, and in particular, FBA determined a significant reduction of its levels. In parallel, they also normalized the expression of TLR4 and its co-receptor CD14 in liver. Accordingly, we evidenced the same profile of activity of butyric treatments on TLR2 mRNA. Consistently with our data, Ehses et al [23] have reported that TLR2 deficient mice are protected from IR and β cell dysfunction induced by HFD, linking TLR2 to the increased dietary lipid and the alteration of glucose homeostasis. Finally, FBA significantly inhibits the HFD-related increase in TLR9 synthesis. Intracellular TLR9 activates innate immune defenses against viral and bacterial infection and plays a role in the pathogenesis of NASH [24]. We previously demonstrated that HFD feeding is associated with the reduction of PPAR-α expression in liver [25] and the administration of a PPAR-α agonist or probiotics restores PPAR-α and improves hepatic steatosis [25,26]. Here, we evidence that butyrate and FBA are also able to do this.
Our data confirm recent in vitro and in vivo findings that identify butyrate as a new inducer of FGF21 [12]. In that study, butyrate injection increased FGF21 serum concentration and protein levels in the liver of obese mice through activation of PPAR-α which was dependent on HDAC3 inhibition by butyrate. Despite the differences in the two experimental models (6 weeks vs 20 weeks of diet, an sub-chronic treatment vs an acute one), in our experiments, butyrate and its derivative are still able to restore hormone levels in the early stage of NAFLD, in liver of non obese rats assuming the involvement of HDAC3 inhibition. FGF21 is a metabolic hormone predominantly produced by the liver, but also expressed in adipocytes and pancreas, where it regulates glucose and lipid metabolism through pleiotropic actions [27]. We studied PPAR-α and PPAR-γ expression in tissues where they are more abundant, liver and adipose tissue, respectively [28]. PPAR-γ promotes fatty acid uptake and increases insulin sensitivity by upregulating GLUT-4, an insulin dependent glucose transporter in adipose tissue and striated muscle [29] and attenuating the induction of SOCS3 [30]. In the current report, we provide evidence that butyrate and FBA not only prevent, in adipose tissue, the HFD-induced reduction of PPAR-γ but also positively modulate PPAR-γ-coactivator PGC-1α [31]. This transcription coactivator controls energy metabolism, interacting with several transcription factors, including PPAR-α and PPAR-δ, that regulate gene transcription for mitochondrial biogenesis and respiration [32]. In fact, the reduction in PGC-1α function is associated with mitochondrial dysfunction, reduction in fatty acid oxidation, and risk for IR or type 2 diabetes [33,34]. Dietary intervention of PGC-1α activity holds promise in the prevention and treatment of metabolic syndrome [11,35]. Our data support the concept that the stimulation of PGC-1α activity may be a molecular mechanism of butyrate activity, in agreement with previous data demonstrating that the inhibition of histone deacetylases and activation of AMPK may contribute to the PGC-1α regulation [11].
Consistently, butyrate and FBA are able to normalize the facilitative hexose transporter, GLUT-2 and insulin-stimulated glucose transport GLUT-4 expression in liver and adipose tissue, respectively, supporting systemic effects of both compounds. Our data are in agreement with previous findings demonstrating that butyrate up-regulates GLUT2 mRNA abundance in other cell types, such as Caco 2 -BBe monolayers, by activating specific regions within the human GLUT2 promoter [36]. GLUT-2 is located in the plasma membrane of hepatocytes and pancreatic beta cells, in contrast with GLUT-1 and GLUT-3, has a low affinity for glucose. Their high Km (15-20mM) allows for glucose sensing; rate of glucose entry is proportional to blood glucose levels. GLUT4 transporters are insulin sensitive, and are found mainly in muscle and adipose tissue. As muscle is a principle storage site for glucose and adipose tissue for triglyceride (into which glucose can be converted for storage), GLUT4 is important in post-prandial uptake of excess glucose from the bloodstream or in other conditions of over-nutrition (i.e. HFD). Moreover, recent findings have reported an improvement of liver glycogen storage by acute butyrate supply that was explained by the competition between butyrate and glucose oxidation, and by a likely reduced glycogenolysis from the newly synthesized glycogen [37].
To our knowledge, our work is the first to propose concomitant biochemical mechanisms in the liver and adipose tissue to better understand how butyrate and its derivative may regulate glucose metabolism and improve insulin sensitivity.
Our results show a protective effect of butyrate to limit early molecular events underlying IR linked to steatosis, suggesting a potential clinical utility as innovative, preventive and therapeutic strategy for NAFLD. In fact, these treatments prevent the transition from steatosis toward steatohepatitis, dampening the onset of several hits responsible for the shift and the progression of the disease. Since FBA does not have the characteristic odor of rancid butyrate, this derivative may represent a viable therapeutic alternative to butyrate, favoring a better compliance and a greater effectiveness.

Ethics Statement
This study was carried out in strict accordance with the Institutional Guidelines and complied with the Italian D.L. no. 116 of January 27, 1992 of Ministero della Salute and associated guidelines in the European Communities Council Directive of November 24, 1986 (86/609/ECC). All animal procedures reported herein were approved by the Institutional Committee on the Ethics of Animal Experiments (CSV) of the University of Naples "Federico II" and by the Ministero della Salute under protocol no. 2008-0099793. Prior to sample collection, animals were euthanized by an intraperitoneal injection of a cocktail of ketamine/xylazine, followed by cervical dislocation to minimize pain. All efforts were made to minimize animal suffering.

Drugs and reagents
Standard and high-fat diet (HFD) were purchased from Laboratorio Dottori Piccioni (Gessate, Milan, Italy). Standard diet had 15% fat, 22% proteins, and 63% carbohydrates, while HFD had 58% of energy derived from fat, 18% from protein, and 24% from carbohydrates. The composition of high fat diet has been previously described [38]. Standard and high fat diets contained 4.06 kcal/g and 5.56 kcal/g, respectively.
Sodium butyrate was purchased by Sigma-Aldrich (Milan Italy) and phenylalanine-butyramide (FBA; Italian patent RM2008A000214; April 21, 2008) was provided by Prof. Calignano. FBA is present in a solid, poorly hygroscopic, easily weighable form, stable to acids and alkalis and capable of releasing butyric acid at small and large bowel level in a constant manner over time. This product has demonstrated a toxicological profile comparable to that of butyrate; it shows physicochemical characteristics distinctly more suitable for extensive clinical use than those of butyrate. A particular aspect of FBA is that it does not present the unpleasant odour of butyrate and is practically tasteless, thus making possible to overcome the main limitation to the use of butyrate in the therapeutic field, namely its very poor palatability. Moreover, the solubility of FBA in water is satisfactory in that it produces clear solutions up to the concentration of 0.1 M and suspensions for higher concentrations.

Synthesis and characterization of butyric acid derivative FBA
Briefly, 0.01M of phenylalanine carboximide and 0.01M butyroyl chloride were dissolved in 50 ml of chloroform and the Effects of butyrate and its derivative in NAFLD PLOS ONE | www.plosone.org resulting mixture were left to react at room temperature for 24h. The mixture, evaporated in vacuo, yields a solid white-coloured residue, that was washed with a 1% sodium bicarbonate solution. The aqueous bicarbonate solution was extracted twice with an equal volume of ethyl acetate to recover an additional fraction of the mixture of derivatives. To isolate the single components, the mixture was treated and processed chromatographically on a silica gel column, using dichloromethane as eluent. The compound was re-crystallised with a mixture of chloroform/n-hexane 1:1 v: v, obtaining a final yield equal to or greater than 50%.

Animals and treatments
After weaning, young male Sprague-Dawley rats (average body weight 113.0±2.2 g), purchased from Harlan, Italy, were randomly divided into four groups (at least 6 animals for each group) as follows: 1) a control group receiving STD and vehicle per os by gavage; 2) HFD-fed group receiving vehicle; 3) HFDfed group treated by gavage with sodium butyrate (HFD +butyrate, 20 mg/kg/die) or 4) with N-(1-carbamoyl-2-phenylethyl) butyramide (HFD+FBA 42.5 mg/Kg/die, the equimolecular dose of butyrate). The treatments started together with the HFD and continued for 6 weeks.
We used a nutritional model of IR in non-genetically modified animals [39] that after 6 weeks, induced the early events of NAFLD due to fat overnutrition in young animals, excluding age and gender influences.
Blood sample was collected by cardiac puncture and serum obtained. Liver and white adipose tissue were excised and immediately frozen.

Histological Analysis of Liver Tissue and triglycerides content
Liver sections were stained with hematoxylin and eosin or Oil Red O. Steatosis was graded on a scale of 0 (absence of steatosis), 1 (mild), 2 (moderate) and 3 (extensive). Liver triglycerides were determined as previously described [40].

Oral Glucose Tolerance Test (OGTT) and Insulin Resistance Assessment
Five weeks after the beginning of the experiment, all rats were fasted for 18 h and then underwent a glucose tolerance. Glucose was orally administered (2 g/kg body weight). Blood samples were collected sequentially from the tail vein before, and 30, 60, 90, and 120 min after the glucose load. The area under the glucose concentration versus time curve (AUC) was calculated from time zero, as the integrated and cumulative measure of glycemia up to 120 min for all animals. To compare the course of the glucose concentration among groups, statistical analysis of AUC mean values was performed using the GraphPad Prism software (GraphPad Software, Inc., San Diego, CA).

Real-time semi-quantitative PCR
Total RNA, isolated from liver and adipose tissue, was extracted using TRIzol Reagent (Invitrogen Biotechnologies), according to the manufacturer's instructions. cDNA was synthesized using a reverse transcription kit (Maxima First Strand cDNA Synthesized Kit, Fermentas, Ontario, Canada) from 2 µg total RNA. PCRs were performed with a Bio-Rad CFX96 Connect Real-time PCR System instrument and software (Bio-Rad Laboratories). The primer sequences are reported in Table 2. The PCR conditions were 10 min at 95°C followed by 40 cycles of two-step PCR denaturation at 95°C for 15 s and annealing extension at 60°C for 60 s. Each sample contained 1-100 ng cDNA in 2X Power SYBRGreen PCR Master Mix (Applied Biosystem) and 200 nmol/l of each primer (Eurofins MWG Operon, Huntsville, AL) in a final volume of 25 µl. The relative amount of each studied mRNA was normalized to GAPDH as housekeeping gene, and the data were analyzed according to the 2 -ΔΔCT method.

Statistical analysis
All data were presented as mean ± SEM. Statistical analysis was performed by ANOVA test followed by Bonferroni's test for multiple comparisons. Statistical significance was set at P<0.05.