Introduction
Nonalcoholic fatty liver disease (NAFLD), characterized by excessive accumulation of triglycerides in hepatocytes, is the most common chronic liver disorder in the western world, and its prevalence is rapidly increasing globally [
1,
2]. NAFLD can progress to inflammatory steatosis, known as nonalcoholic steatohepatitis (NASH), which can eventually lead to fibrosis, cirrhosis, and hepatocellular carcinoma [
2,
3]. Steatosis of the liver is often accompanied by insulin resistance and obesity [
4]. NAFLD is recognized as an early independent risk factor for the development of atherosclerosis; therefore, the identification of compounds that can inhibit liver steatosis may be an important step in the design of potential therapies for atherosclerosis and its complications [
2,
5].
The excessive accumulation of triglycerides in the cytoplasm of hepatocytes was shown to be caused by an imbalance between cellular uptake/de novo synthesis of fatty acids and fatty acid oxidation (FAO), the latter marked by a reduction in the level of PPAR-α—the key transcription factor for liver FAO [
6]. Glucose metabolism tightly regulated by insulin. Hepatic insulin resistance is marked by a failure of insulin to suppress hepatic glucose production, while at the same time stimulating increased lipogenesis. Elevated levels of circulating free fatty acids (FFAs), in part related to decreased insulin suppression of adipose tissue lipolysis, result in increased delivery of FFAs to the liver [
7]. Therefore, the presence of liver steatosis is associated with a constellation of alterations in glucose, fatty acid, and lipoprotein metabolism, as well as mitochondrial dysfunction, which has been shown to aggravate the development of NAFLD [
4].
In recent years, much attention has been focused on the action of free fatty acid receptors (FFARs), which have been shown to be expressed in hepatocytes and to impact metabolism [
1,
8]. The type 4 receptor (FFAR4), with pleiotropic effects, is of particular interest [
1,
9,
10]. FFAR4, also known as GPR120, acts as a receptor for long-chain polyunsaturated free fatty acids (PUFAs). It plays a key role in various physiological mechanisms including adipogenesis, appetite regulation, and regulation of food preferences [
8]. Ichimura et al. showed that FFAR4-deficient mice fed a high-fat diet (HFD) developed obesity, glucose intolerance, and fatty liver, with increased hepatic lipogenesis [
11,
12]. Oh et al. showed that activation of FFAR4 dampened inflammatory activation of macrophages [
13]. The combination of the metabolic effects and the inhibition of inflammatory reactions have made FFAR4 stimulation an interesting area of investigation in the search for new drugs for the treatment/prevention of metabolic and circulatory disorders, but few data are available from animal models using FFAR4 agonists.
Recently, we demonstrated that activation of FFAR4 by its agonist, TUG-891, inhibited atherosclerosis in mice with knockout of apolipoprotein E (apoE
−/−mice), which was associated with a beneficial phenotypical change between M1/M2 macrophages in atherosclerotic plaques [
5]. ApoE
−/− mice are a well-known and widely used model for research on atherosclerosis and liver steatosis [
14‐
16]. Unlike wild-type mice, they develop atherosclerosis on a chow diet, while liver steatosis still requires the use of a diet with higher fat content. However, the diet that causes homogeneous, moderate liver steatosis in apoE
−/− mice does not have to contain very high levels of fat (more than 20%) or cholesterol (ca. 1.25%), resulting in better model stability and homogeneity of results.
In this study we investigated the effects of TUG-891 on the development of hepatic steatosis in apoE−/− mice fed an HFD, and identified possible pathways for this action using sequential window acquisition of all theoretical mass spectra (SWATH-MS) quantitative proteomics and molecular methods.
Discussion
In this study, we demonstrated that TUG-891 significantly inhibited the development of liver steatosis in apoE
−/− mice fed a western diet. To our knowledge, this is the first study to report that pharmacological stimulation of FFAR4 may effectively inhibit hepatic steatosis in vivo. We found that TUG-891 treatment resulted in a 20% reduction in the number of hepatocytes with signs of steatosis and a significant reduction in liver TG levels. Such effects were accompanied by hepatoprotective action, as evidenced by the decrease in plasma AST levels. These results represent an important continuation of our previous study showing that TUG-891 significantly inhibited the development of atherosclerosis in the very same experimental setting, in which it did not induce significant changes in the blood lipid levels or weight of the animals, and no toxic effects of the compound were observed [
5].
Our results suggest that the inhibitory effect of TUG-891 on the development of fatty liver is associated with inhibition of the de novo lipogenesis (DNL) pathway, as evidenced by a significant reduction in the expression of
Srebp-1c mRNA and acetyl-CoA carboxylase (
Acc) mRNA/protein. Several studies point to Srebp-1 as a key regulator of DNL [
34]. The liver triglyceride level was shown to be higher in transgenic mice [
35]. Furthermore, SREBP-1 is elevated in the liver of patients with NAFLD. There is also a correlation between SREBP-1c expression and the severity of insulin resistance and obesity in patients with metabolic syndrome [
2]. ACC is the first enzyme involved in the DNL pathway, and its cytosolic isoform ACC1 is involved in the rate-limiting step in DNL (ACC2, a mitochondrial membrane-associated enzyme, is involved in FAO). DNL has been shown to be inhibited in the liver of ACC1-knockout mice [
36]. In this context, further studies of the molecular mechanisms responsible for the TUG-891-dependent decrease in liver expression of
Srebp-1c and
Acc1 appear to be a very interesting prospect.
In our model, the influence of TUG-891 on other important factors related to DNL consistently indicated its inhibitory effect on this metabolic pathway in the liver: increased
Cpt1a and decreased
Ppar-γ levels. The carnitine palmitoyltransferase (CPT) system is responsible for the transport of long-chain fatty acids from the cytoplasm to mitochondria, where the fatty acids undergo β-oxidation. Decreased CPT1A expression was shown to be involved in the development of NAFLD in a mouse model with NAFLD [
37]. The nuclear receptor PPAR-γ, mainly due to its role in the processes of differentiation and modulation of inflammatory responses of various cells, also has a modulatory effect on DNL in the liver [
38]. PPAR-γ deletion in mouse hepatocytes has been shown to protect against the development of steatosis in mice with liver steatosis and diabetes. Therefore, the increase in CPT1A and the decrease in PPAR-γ induced by TUG-891 seem to suggest its broad inhibitory effect on liver DNL [
36].
In our setting, TUG-891 treatment resulted in a decrease in the expression of genes related to fatty acid uptake and oxidation—
Cd36 and
Fabp1, and
Ehhadh and
Acox1, respectively [
39]. As shown, CD36-knockout mice were protected against HFD-induced liver steatosis and reduced expression of FABP1 resulting in decreased intracellular lipid-binding capacity and reduced lipotoxicity. The influence of TUG-891 on factors associated with these pathways may also play a mechanistic role in reducing liver steatosis, but this requires further research.
The development of fatty liver has been shown to be associated with abnormal glucose metabolism in hepatocytes [
40]. In our model, TUG-891 treatment resulted in changes in liver expression of factors related to glucose metabolism in the liver. Glucose transporter 2 (GLUT2, coded for by the
Slc2a2 gene),
Pdk4, and
Pklr were decreased and
G6pdx mRNA increased after treatment with TUG-891. The interpretation of the influence of TUG on the development of NAFLD in the context of its influence on factors related to glucose metabolism is ambiguous. Our results on the expression of GLUT-2 seem to contradict the observations so far: Plasma levels of FFAs, as well as their uptake into hepatocytes, were reported to increase in GLUT2-deficient mice [
41]. On the other hand, our results for PKLR and PDK4 appear to indicate a beneficial effect of TUG-891 on hepatic glucose metabolism in the context of the development of steatosis, as PKLR and PDK4 were shown to aggravate NAFLD [
42,
43]. Therefore, the mechanistic links between the influence of TUG-891 on factors related to glucose metabolism and its potential importance for the antisteatotic action of TUG-891 in the liver are unclear.
In general, many mechanisms (that is, related to inflammatory stimulation, mitochondrial dysfunction, reactive oxygen species [ROS] production, and endoplasmic reticulum [ER] stress) at the cellular and intracellular levels are believed to play a role in the development of NAFLD [
44]. Consequently, the action of FFAR4 appears to be pleotropic and largely cell-dependent [
8].
Mitochondria are considered key organelles and are at the crossroads of many pathways that play a role in the development of the fatty liver. Mitochondria are involved not only in cellular respiration, but also in gluconeogenesis and other key biosynthetic activities, as well as contributing to the oxidative stress and inflammatory reactions observed in NAFLD. Our objective was to thoroughly detect possible changes in mitochondrial/cytosolic pathways in hepatocytes after treatment with TUG-891 using a quantitative proteomics method, SWATH-MS proteomics. To maximize the sensitivity and proteome coverage of the measurements, we employed a subcellular fractionation of liver tissue prior to SWATH-MS analysis, as sample fractionation at the peptide level is not recommended in data-independent acquisition (DIA) measurements. One of the main advantages of SWATH (DIA) data sets is high reproducibility with low CV values. The sample fractionation process at the peptide level introduces additional variability into the sample preparation protocol, which decreases the quality of the SWATH data set. In principle, increased coverage in a SWATH analysis is achieved by optimizing the chromatographic separation and mass spectrometry acquisition methods and building a better spectral library. To ensure the highest quantitative data output, we prepared separate spectral libraries for both fractions instead of one for the entire tissue lysate. In this way, we were able to assess the quantitative changes in protein induced by TUG-891 administration in approximately 6500 unique protein groups. Furthermore, we were able to focus our quantitative protein analysis on the cytosolic and mitochondrial compartments. The latter was especially important given that mitochondrial dysfunction is a well-described characteristic of NAFLD. As a result, we revealed some interesting clues for further research on the mechanisms of action of TUG-891. Furthermore, we validated several observations from mRNA measurements at the protein level. Quantitative data were of high quality (Supplementary Fig.
2) and all proteins of interest were specifically enriched in one of the fractions. Therefore, our SWATH data set represents robust and comprehensive proteomic insight into TUG-891-induced changes in the liver of apoE
−/− mice. In addition, taking into account the longitudinal nature of the experimental design of the present study and the relative homogeneity of liver tissue, the quantitative changes observed in transcripts and proteins constitute the molecular fingerprint of the prolonged action of the FFAR4/GPR120 agonist.
Analysis of the liver proteome revealed the effect of TUG-891 on the expression of many proteins related to lipid/xenobiotic metabolism, homoeostasis, and immunity.
Myo-inositol was reported to lower serum lipid levels and reduce the risk of fatty liver, and inositol synthesis may contribute to NAFLD [
45]; therefore, the increase in inositol-3-phosphate synthase 1 (ISYNA1), involved in inositol synthesis, could potentially play a role in the TUG-891-dependent inhibition of liver steatosis.
Increasing evidence suggests that some members of the CYP450 family, mainly due to the aggravation of oxidative stress, may contribute to the pathogenesis of NAFLD and NASH [
46]. Interestingly, TUG-891 downregulated the 17-β hydroxysteroid dehydrogenase 13 (HSD17B13), an important regulator of the biogenesis, growth, and degradation of lipid droplets in the liver [
46]. Notably, several loss-of-function mutations in the human HSD17B13 gene have been shown to confer strong anti-inflammatory and antifibrotic effects in the liver, and a few single-nucleotide polymorphisms in the HSD17B13 gene have also been associated with the development of NAFLD in humans [
47]. This interesting clue on the potential mechanism of antisteatotic action of TUG-891 in the liver needs to be verified.
Somewhat surprisingly, massive changes in proteins that regulate oxidative stress did not change with TUG-891. The TUG-891-dependent upregulation of STEAP4 may be interesting in this regard, since metalloprotease STEAP4, a member of the family of six transmembrane proteins, has been recognized as a modulator of inflammation and nutrient metabolism [
48]. The loss of STEAP4 in mice leads to increased inflammatory cytokine production in visceral white adipose tissue and systemic insulin resistance. Furthermore, STEAP4 overexpression significantly increased SOD2 expression and mitochondrial antioxidant capacity, ultimately reducing cellular oxidative stress. In the context of the protective effect on mitochondria, the mechanism linking FFAR4 and STEAP4 may be interesting and worthy of further research. Additional research is also warranted on results showing that TUG-891 increases the production of the main urinary proteins (MUP-6, MUP-11, MUP-18) in the liver, which, as studies increasingly indicate, may have interesting functions in the regulation of metabolism [
49].
Some of our proteomics results indicated increased liver production of proteins involved in coagulation (coagulation factor VII and IX, platelet factor 4, heavy chain 3 and 4, and fibrinogen alpha/beta/gamma chain) and inflammatory processes of innate immunity (e.g., orosomucoid 1, selected complement components, neutrophil gelatinase-associated lipocalin, myeloperoxidase) in the liver after treatment with TUG-891. Such results may come as a surprise in the context of reports of the mutual involvement of inflammatory reactions and thrombosis (often referred to by the common term thromboinflammation) in many cardiovascular pathologies [
50]. In particular, in our setting, we did not observe any thrombotic complications or increased plasma levels of biochemical markers of inflammation in blood serum [
5]. Certainly, these aspects of the action of TUG-891 in vivo require further research.
Our work indicates for the first time the steatosis-inhibiting effect of the FFAR4/GPR120 agonist, but we cannot ignore the limitations of this study. The difficulties in interpreting proteomics results and the need for their further verification in studies of cause-and-effect relationships constitute the most obvious limitation of our work. An important addition to our research would be the metabolomic data that confirm the effect of TUG-891 on the lipogenesis and lipolysis pathways in the liver. Despite the well-documented selectivity of TUG-891 for FFAR4/GPR120, its possible off-target effects in vivo could be verified by the concomitant administration of the FFAR4/GPR120 antagonist (e.g., AH-7614). The experiment should also be repeated on other NAFLD models and supported by histological methods that will show even more accurately the effects of TUG-891 on steatosis and possible fibrosis (e.g., Oil Red O and Masson’s trichrome staining). Furthermore, it would be very interesting to compare the potency of TUG-891 with known substances that inhibit steatosis (for example, metformin) in the same model. Thus, it is clear that great caution must be exercised when translating the results from animal models to human pathology.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.