Food preservative sorbic acid deregulates hepatic fatty acid metabolism.

Sorbic acid (SA) is one of the most commonly used food preservatives worldwide. Despite SA having no hepatotoxicity at legal dosages, its effect on hepatic lipid metabolism is still unclear. We investigated the effect of SA on hepatic lipid metabolism and its mechanism of action in C57BL/6 mice. Daily treatment with SA (1 g/kg in diet) for 4 weeks did not alter the body weight, organ weight, and blood lipids in mice. However, hepatic lipid accumulation, particularly that of triglycerides, fatty acids, and glycerol, but not cholesteryl ester and free cholesterol, was increased with SA treatment. Mechanistically, SA decreased the expression of proteins related to de novo fatty acid lipogenesis, fatty acid internalization, and very low-density lipoprotein (VLDL) secretion-related pathways, including sterol regulatory element-binding proteins, acetyl-coA carboxylase, fatty acid synthase, liver fatty acid-binding protein, CD36, and apolipoprotein E. In contrast, SA increased the expression of diacylglycerol O-acyltransferase 2, the key enzyme for triacylglycerol synthesis. Moreover, SA downregulated the protein expression of autophagy-related and β-oxidation-related pathways, the two major metabolic pathways for lipid metabolism, including LC-3, beclin-1, autophagy related protein 5 (ATG-5) and ATG-7, acyl-CoA synthetase long chain family member 1, carnitine palmitoyltransferase Iα, peroxisome proliferator-activated receptor α (PPARα), PPARγ, and PPARγ coactivator-1. Collectively, SA deregulates de novo lipogenesis and fatty acid internalization, VLDL secretion, autophagy, and β-oxidation in the liver, leading to impaired lipid clearance and ultimately, resulting in lipid accumulation in the liver.

1. Introduction

Sorbic acid (SA), also known as 2,4-hexadienoic acid, is a natural organic compound. It is derived from the berries of the mountain rowan tree. Since the 1930s, SA has been used as a preservative for food and wine due to its ability to prevent spoilage caused by yeast, fungi and molds, as well as some bacteria. To cater to huge demand, SA is produced by a chemical synthesis method. Currently, preservatives are being widely used with the purpose to extend the expiration date of foods [1]. In Taiwan, SA can be used legally in 34 specific food items and regulations limit the concentration to 0.5–2 g/kg of food [2]. For instance, SA cannot exceed 2 g/kg in meat products, in soy sauce, miso, jam, and ketchup to a maximum concentration of 1 g/kg and in drinks to a maximum concentration of 0.5 [2]. Ling et al. reported that the use of SA at the current dosage in the Taiwanese diet does not pose a risk for public health and safety [2]. However, an elegant study by Luo et al. demonstrated that diets supplemented with SA (2 g/kg) for 35 days increased plasma levels of albumin, triglycerides, high-density lipoprotein (HDL), and low-density lipoprotein (LDL) in pigs [3].

The liver is the major organ for controlling lipid homeostasis in the human body including that of cholesterol and triglycerides, with fluxes in de novo lipid synthesis in the liver, and from dietary, circulating, and peripheral local lipid pools [4,5]. It is well-established that de novo lipogenesis in the liver and lipids from dietary fats are the main sources for maintaining physiological function [4,5]. Under physiological conditions, about 70–80% of total daily cholesterol and fatty acids are produced in the liver [6,7]. Hepatic lipid homeostasis is dynamically regulated by integrating several aspects of metabolic pathways including de novo lipogenesis, fatty acid uptake, triglyceride synthesis, β-oxidation of fatty acids, secretion of very-low-density lipoprotein (VLDL), and lipoprotein internalization (chylomicron remnants, VLDL, intermediate-density lipoprotein, LDL, and HDL) from circulation [8-12]. It has been well-documented that the balance between de novo lipogenesis and lipid metabolism is a crucial factor for maintaining whole-body lipid homeostasis [8,12]. For example, increased hepatic lipogenesis and decreased β-oxidation of fatty acids may result in excessive lipid accumulation in the liver and hyperlipidemia, which are important risk factors in the development of fatty liver and related metabolic diseases [13,14]. During the past decades, adequate evidence suggests that the deregulation of hepatic lipid metabolism by genetic factors, clinical drugs, and environmental pollutants leads to hyperlipidemia, fatty liver, and metabolic disorders [15-18]. However, less is known about the effect of SA on hepatic lipid metabolism. To this end, investigations delineating the effect and molecular mechanisms of SA in hepatic lipid metabolism and related diseases are warranted.

Given the impact of SA in increasing the level of plasma lipid, we aimed to characterize the effect and molecular mechanisms of SA regarding hepatic lipid metabolism. We first investigated the effect of SA on plasma lipid level and hepatic lipid accumulation in wild-type (WT) mice and then examined whether SA affected the molecular mechanisms in hepatic lipid metabolism. Our findings demonstrated that SA might reduce the production and uptake of fatty acids in the liver and decrease the lipid clearance and metabolism, resulting in hepatic lipid accumulation.

2. Materials and methods

2.1. Reagents

SA and oil red O dyes were obtained from Sigma–Aldrich (St. Louis, USA). Mouse antibody for carnitine palmitoyltransferase 1α (CPT1α), rabbit antibodies for acyl-coA synthetase α (ACCα), acyl-CoA oxidase (ACOX1), apolipoprotein B (ApoB), and peroxisome proliferator-activated receptor γ coactivator 1α (PGC1), alanine aminotransferase (ALT) and aspartate aminotransferase (AST) assay kits were procured from Abcam (Cambridge, UK). Mouse antibodies for sterol regulatory element-binding protein 1 (SREBP1) and sterol regulatory element-binding protein 2 (SREBP2) were procured from BD Bioscience (San Jose, CA, USA). Mouse antibodies for β-actin, peroxisome proliferator-activated receptor γ (PPARγ), rabbit antibodies for acyl-CoA synthetase (ACSL1), beclin-1, diacylglycerol acyltransferase 1 (DGAT1), fatty acid synthase (FAS), 3-hydroxy-3-methylglutaryl-CoA reductase (HMGCR), liver fatty acid-binding protein (L-FABP), and peroxisome proliferator-activated receptor α (PPARα), goat antibodies for diacylglycerol acyltransferase 2 (DGAT2), and microsomal triglyceride transfer protein (MTP) were obtained from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Rabbit antibodies for autophagy related protein 5 (ATG5), autophagy related protein 7 (ATG7), CD36 and low-density lipoprotein receptor (LDLR) were from Novus Biologicals (Littleton, Co, USA). Rabbit antibody for LC3A/B was purchased from Cell Signaling Technology (Danvers, MA, USA). Rabbit antibody for apolipoprotein E (ApoE) was from Epitomics (Burlingame, CA, USA). Cholesterol and triglyceride assay kits were obtained from Randox (Crumlin, Co Antrim, UK). Total cholesterol, free cholesterol, cholesteryl ester, triglyceride fatty acid and glycerol fluorometric assay kits were purchased from BioVision (Milpitas, CA, USA).

2.2. Mice

This study conformed to the Guide for the Care and Use of Laboratory Animals (Institute of Laboratory Animal Resources, eighth edition, 2011), and all animal experiments were approved by the Animal Care and Utilization Committee of National Yang-Ming University. Eight-week old C57BL/6 mice purchased from National Laboratory Animal Center (Taipei, Taiwan) were randomly divided into two groups and treated with either vehicle or SA. Mice were housed in barrier facilities, maintained in a 12-h/12-h light–dark cycle. Temperature (22 °C) and humidity (40–60%) of the vivarium were tightly controlled. Mice were group-housed 5 per cage and fed a regular chow diet containing 4.5% fat by weight (0.02% cholesterol) (Newco Distributors, Redwood, CA, USA). Two-month-old male C57BL/6 mice received daily treatment with SA (1 g/kg in diet) or vehicle for four weeks. At the end of the experiment, mice were euthanized with CO2. The liver, white adipose tissue (WAT), and brown adipose tissue (BAT) were isolated and weighed. These were then subjected to histological analysis or stored at −80 °C. The isolated liver, WAT, and BAT were homogenized, and lysates were subjected to Western blot analysis.

2.3. Histological examination

The liver, white adipose tissue (WAT), and brown adipose tissue (BAT) blocks were cut into 8-μm sections and subjected to histological examination. Cryosections were subjected to oil red O staining and deparaffinized sections were stained with hematoxylin and eosin (H&E), and then viewed under a Motic TYPE 102 M microscope (Motic Images, China).

2.4. Western blotting analysis

The liver samples were lysed using immunoprecipitation lysis buffer (50 mmol/L Tris, pH 7.5, 5 mmol/L EDTA, 300 mmol/L NaCl, 1% Triton X-100, 1 mmol/L phenylmethylsulfonyl fluoride, 10 μg/mL leupeptin, and 10 μg/mL aprotinin). Aliquots of lysates were separated by SDS-PAGE and trans-blotted onto an Immobilon-P membrane, blocked with 5% skim milk for 1 h at room temperature, and then incubated with primary antibodies overnight and then with corresponding secondary antibody for 1 h. The protein bands were detected by using an enhanced chemiluminescence kit (PerkinElmer, Boston, MA, USA) and quantified using Image-Quant 5.2 (Healthcare Bio-Sciences, PA, USA).

2.5. Serum lipid profile analysis

Blood was collected by cardiac puncture. After clotting and centrifugation, serum was isolated and levels of triglycerides, total cholesterol, HDL-cholesterol (HDL-c), and non-HDL-c in serum were measured using Spotchem EZ SP 4430 (ARKRAY, Inc., Kyoto, Japan).

2.6. Determination of hepatic lipids and glycerol

The levels of total cholesterol, free cholesterol, cholesteryl ester, triglycerides, fatty acids, and glycerol were measured using fluorescence assay kits (BioVision, Milpitas, CA, USA).

2.7. Statistical analysis

Data are presented as mean ± SEM. Data from mice were evaluated by parametric tests. The unpaired t-test was used to compare two independent groups. SPSS software v20.0 (SPSS Inc, Chicago, IL, USA) was used for analysis. Differences were considered statistically significant at P < 0.05.

3. Results

3.1. SA increases lipid accumulation in the liver without changing body weight, organ weight and blood lipid profile of mice

The liver is the major organ for lipid metabolism in humans and animals [6,7]. We examined the effect of SA on hepatic lipid metabolism and its potential mechanism(s) of action. Results of H&E and oil red O staining showed that daily treatment of mice with SA for 4 weeks increased lipid accumulation (Fig. 1A). In addition, treatment with SA increased hepatic triglyceride levels, free fatty acid, and glycerol but not total cholesterol, free cholesterol, and cholesteryl ester (Fig. 1B–G). These results suggest that increased lipid accumulation in the liver due to SA treatment might be attributed to the elevation of triglycerides, free fatty acid, and glycerol.

Fig. 1

SA increases lipid accumulation in the liver. Eight-week old C57BL/6 mice were treated with SA (1 g/kg diet/day) or vehicle (oil) for four weeks. (A) The liver appearance, representative H&E and oil red O staining of liver tissue. (B–G) Lipid profile in liver was assessed by fluorometric assay kits. (B–D) Hepatic levels of (B) triglycerides, (C) fatty acids, and (D) glycerol, (E) total cholesterol, (F) free cholesterol and (G) cholesteryl ester. Data are represented as mean ± SEM from five mice. *P < 0.05 vs. vehicle.

Intriguingly, we demonstrated that daily treatment of mice with SA for four weeks did not alter their appearance (Supplementary Fig. S1A), body weight, the liver weight, and the ratio of the liver weight to body weight as compared to vehicle treatment (Supplementary Fig. S1B, C and F). SA treatment showed an increasing trend in WAT weight, BAT weight, and their ratios with respect to body weight; however, there was no statistical significance (Supplementary Fig. S1D, E, G, and H). Results of H&E sating also showed that SA treatment did not affect the histology of WAT and BAT (Supplementary Fig. S1I and J). In addition, SA did not affect the circulating levels of triglycerides, total cholesterol, HDL-c, and non-HDL-c in C57BL/6 mice (Supplementary Fig. S2A–D).

3.2. Effects of SA on de novo hepatic lipogenesis-and fatty acid metabolism-related protein expression

The lipid homeostasis in the liver is tightly controlled by several metabolic pathways including de novo lipogenesis, fatty acid uptake, triglyceride synthesis, secretion of VLDL, and lipoprotein internalization [8-12]. We next examined the mechanisms underlying the detrimental effect of SA on lipid accumulation in the liver by evaluating the protein expression of hepatic lipid metabolic pathways. As compared to vehicle treatment, daily SA treatment decreased the protein expression of SREBP-1 and its downstream proteins ACCα and FAS, the key regulators for fatty acid synthesis [8,9], without affecting the expression of SREBP-2 and its downstream proteins LDLR and HMGCR, the key regulators for cholesterol synthesis (Fig. 2A–C). Furthermore, treatment with SA increased the protein expression of DGAT2, but not DGAT1 (Fig. 3A–C), the key regulators of triglyceride synthesis in the liver [19]. On the other hand, treatment with SA decreased the protein expression of L-FABP and CD36 (Fig. 3A–C), two crucial hepatic regulators in the internalization of fatty acids from circulation [20,21]. Moreover, SA treatment decreased the protein expression of apoE, but not apoB and MTP (Fig. 3A–C), key players in the process of VLDL assembly and secretion [22,23]. These results suggest that increased triglyceride synthesis by upregulation of DGAT-2 and impaired VLDL secretion by downregulation of apoE might result in the disturbance of hepatic lipid metabolism and thereby lead to lipid accumulation in the liver.

Fig. 2

Effect of SA on de novo fatty acid and cholesterol synthesis-related protein expression in the liver. Eight-week old C57BL/6 mice were treated with SA (1 g/kg diet/day) or vehicle (oil) for four weeks. (A and B) Western blot analysis of protein levels of SREBP1, ACCα, FAS, SREBP2, LDLR, HMGCR, and β-actin, in the liver. (C) Summary for the effect of SA on hepatic de novo lipogenesis. Data are represented as mean ± SEM from five mice. *P < 0.05 vs. vehicle.

Fig. 3

Effect of SA on fatty acid metabolism-related protein expression in the liver. Eight-week old C57BL/6 mice were treated with SA (1 g/kg diet/day) or vehicle (oil) for four weeks. (A and B) Western blot analysis of protein levels of DGAT1, DGAT2, CD36, L-FABP, MTP, apoB, apoE, and β-actin, in the liver. (C) Summary of the effect of SA on fatty acid metabolism in the liver. Data are represented as mean ± SEM from five mice. *P < 0.05 vs. vehicle.

3.3. SA deregulates catabolic metabolism of hepatic lipid droplets in mice

Autophagy, lipolysis, and β-oxidation of fatty acids are key metabolic processes in regulating intracellular lipid droplets [24,25]. Thus, we next evaluated the effects of SA on the expression of proteins related to catabolic metabolism of hepatic lipid droplets in the liver. We demonstrated that SA markedly decreased the expression of proteins involved in autophagy flux, including beclin-1, SQSTM1, LC3, ATG5, and ATG7 without altering the protein expression of ATGL and HSL, the two major regulators in the process of lipolysis (Fig. 4A–C). Moreover, treatment with SA significantly downregulated the protein expression of ACSL1, CPT1α, PPARα, PPARγ, and PGC1α, all of which are crucial regulators involved in β-oxidation of fatty acids and energy release (Fig. 5A–C). These findings suggest that deregulation of catabolic metabolism of lipid droplets by decreasing autophagy and β-oxidation is the key factor for SA-induced lipid accumulation in the liver.

Fig. 4

SA decreases the protein expression of indispensable regulators involved in autophagy and lipolysis. Eight-week old C57BL/6 mice were treated with SA (1 g/kg diet/day) or vehicle (oil) for four weeks. (A and B) Western blot analysis of protein levels of beclin-1, SQSTM1, LC3, ATG5, ATG7, ATGL, HSL and β-actin, in the liver. (C) Summary of the effect of SA on autophagy and lipolysis in the liver. Data are represented as mean ± SEM from five mice. *P < 0.05 vs. vehicle.

Fig. 5

SA downregulates the protein expression of key regulators involved in β-oxidation pathway. Eight-week old C57BL/6 mice were treated with SA (1 g/kg diet/day) or vehicle (oil) for four weeks. (A and B) Western blot analysis of protein levels of ACSL1, CPT1α, Acox1, PPARα, PPARγ, PGC1α, and β-actin in the liver. (C) Summary of the effect of SA on β-oxidation of fatty acids in the liver. Data are represented as mean ± SEM from five mice. *P < 0.05 vs. vehicle.

4. Discussion

Based on this study, we provided new insights into the effect of SA on hepatic lipid metabolism and the underlying molecular mechanisms. We used a mouse model to investigate the effect of SA on hepatic lipid metabolism and found that chronic treatment with SA (1 g/kg diet/day) for four weeks increased the glycerol and the lipid accumulation, particularly that of triglycerides and free fatty acids in the liver. While the severity of hepatic lipid accumulation by SA did not reach the degree of fatty liver; the plasma levels of ALT and AST were not different with that in control group. However, our knowledge about the molecular mechanism behind the detrimental effect of SA on hepatic lipid metabolism is limited. We then explored the effect of SA on de novo lipogenesis, fatty acid internalization, autophagy, and β-oxidation in this C57BL/6 mouse model. We found that SA decreased the protein expression related to de novo lipogenesis and fatty acid internalization and VLDL secretion in the liver. Moreover, SA downregulated the expression of proteins involved in autophagy and β-oxidation pathways, which in turn limited lipid clearance. These key events might work in concert to promote lipid accumulation in the liver (Fig. 6). Intriguingly, we demonstrated that treatment with SA did not affect the body weight, organ weight, and blood lipid profile of wild type C57BL/6 mice, which, however, was inconsistent with a previous finding reporting that treatment with SA (2 or 4 g/kg diet/day) for 35 days increased the body weight and circulating levels of total cholesterol or triglycerides in piglets [3]. Although the exact mechanism is unclear, the reason for the discrepancy between our study and the study by Luo et al. could be the difference in the dosage of SA and the species of experimental animals. On the other hand, Itagaki et al. reported that the degree of fatty liver and steatohepatitis induced by methionine- and choline-deficient diet was returned to normal condition when switched back to the chow diet for 16 weeks [26]. Therefore, we believe that the effect of SA on hepatic lipid deregulation is reversible once the mice are switched back on the regular chow diet.

Fig. 6

Proposed model of mechanisms by which SA accelerates lipid accumulation in the liver. SA deregulates de novo lipogenesis and fatty acid internalization and VLDL secretion and interferes with autophagy and β-oxidation in the liver, leading to impaired lipid clearance and ultimately resulting in lipid accumulation in the liver.

Regarding with the concentration of SA used in this study, the legal dosages (0.5–2.0 g/kg) of SA used in food have been established [2]. Therefore, we simply adopted the average dosage (1 g/kg) and tried to address the effect of SA at the legal concentration on lipid metabolism. We demonstrated that there was no difference in food intake daily among the vehicle group and SA group (vehicle group: SA group = 4.03 ± 0.32 g: 3.98 ± 0.28 g/mouse [20 g]). Therefore, the consumption amount of SA daily is about 4 mg/mouse. After calculating, the concentration of SA used in our study is about 200 mg/kg body weight for mice. By Meeh-Rubner equation, this concentration of 200 mg/kg for mice could be comparable to the concentration of 4–8 mg/kg for human consumers. The consumption amount in our study is lower than the average intake level of SA at 37 mg/kg for human consumers, suggesting the detrimental effect of SA at legal concentration on lipid metabolism could be an important issue for human health.

SA is a well-known food preservative for preventing microbial growth, thereby extending the shelf life of food. It is widely used in soy sauce, jam, frozen meat products, pickled vegetables, canned foods, and beverages [1,2]. Excessive use of SA may cause hepatotoxicity [27]; therefore, it is necessary to control and use it at legally permissible dosages in preparation and for preservation of food. Indeed, the legally permissible concentration of SA to be used in various food items has been established during the past decades [2]. However, whether the legal concentration of SA used in food disturbs hepatic lipid metabolism and the potential molecular mechanism is largely unknown. The impact of food preservatives on human health and metabolic diseases is still incompletely understood. There have been epidemiological studies reporting that exposure to SA and benzoic acid might be hazardous to human health [2,28]. However, the relationship between the use of food preservatives and human diseases is still unclear. Here, we provided new evidence to support the idea that SA, at legal dosage, increased lipid accumulation in the liver of mice by decreasing autophagy and β-oxidation, two crucial events in fatty acid metabolism. In this study, our findings show that treatment with SA significantly increased the lipid content in the liver, particularly triglycerides, free fatty acids, and glycerol, but not total cholesterol, free cholesterol, and cholesteryl ester, suggesting the unfavorable effect of SA on hepatic lipid metabolism. However, the circulating levels of total cholesterol and triglycerides were not changed. It has been well-documented that dietary fat and de novo lipogenesis in the liver are the main sources of circulating total cholesterol and triglycerides, which are carried in circulation as VLDL [29-31]. In this study, no difference was observed in food intake between vehicle-treated mice and SA-treated mice (data not shown). Therefore, upon SA treatment, elevated levels of triglycerides in the liver might be attributed to increased lipid synthesis, or decreased lipid clearance, or both.

Under physiological conditions, the lipid content in the liver is tightly regulated by the integration of several aspects of metabolic pathways, including de novo lipogenesis, lipoprotein internalization, VLDL synthesis, β-oxidation of fatty acids, and catabolic metabolism of lipid droplets including lipolysis and autophagy [8,10,11,32]. Therefore, we investigated how SA affected hepatic lipid metabolism. Our findings demonstrated that SA-induced accumulation of triglycerides, fatty acid, and glycerol in the liver might not be due to increased de novo lipogenesis or fatty acid uptake as evidenced by the observation that SA decreased the protein expression of SREBP-1, FAS, and ACC, all crucial regulators in the de novo synthesis of hepatic triglycerides [33-35]. In addition, SA decreased the uptake of fatty acids from circulation, as evidenced by the downregulation of L-FABP and CD36, which are key transporters of fatty acids into hepatocytes [36,37]. Moreover, the expression level of the key enzyme DGAT2 for triglyceride synthesis was increased; in contrast, the key regulator apoE for VLDL secretion was decreased by SA treatment. Thus, these deregulations in lipid metabolism with respect to triglyceride synthesis and VLDL secretion might contribute to the elevated levels of triglycerides, free fatty acids, and glycerol in the liver. Moreover, the interaction of the liver-mediated lipid metabolism with peripheral local lipid pools such as WAT and BAT plays a principal role in regulating whole-body lipid homeostasis [30,31], we think that the SA-induced deregulation of lipid metabolism may be not a liver-specific effect. Additionally, SA is a diunsaturated, six carbon-atom fatty acid, can be rapidly broken down quickly and utilized similarly as fatty acids, and ultimately be metabolized to carbon dioxide [1]. Thus, the effects observed on protein expression of different lipid metabolism pathways were unlikely to be direct effects of SA. Despite our unique findings, the detailed mechanisms by which SA modulates lipid metabolism within the liver merit further study.

Notably, we demonstrated that SA downregulated the expression of proteins related to autophagy including beclin-1, SQSTM1, ATG5, and ATG7. Autophagy is a conserved self-eating process that is important for cellular homeostasis under various stress conditions. It allows cells to self-degrade intracellular components including lipid droplets and damaged organelles in the lysosomes for balancing energy sources upon nutrient shortage or various cytotoxic insults [38,39]. Our findings are consistent with studies by Du et al. and Xie et al. who found that inhibition of autophagy disturbs lipid homeostasis and worsens the progression of fatty liver [40,41]. More importantly, SA also decreased the protein expression of ACSL1, CPT1α, PPARα, PPARγ, and PGC-1α, all implicated in the β-oxidation pathway for energy metabolism in mitochondria [42-46], which are in line with previous findings that suggest that inhibition of fatty acid oxidation damages energy metabolism and accelerates the development of non-alcoholic fatty liver [47,48]. These findings suggest that SA impaired the functions of autophagy and β-oxidation pathways in catabolic metabolism of triglyceride-rich lipid droplets, leading to lipid accumulation in the liver. This observation agrees with previous reports that impairment of autophagy or β-oxidation disturbs hepatic lipid metabolism and accelerates the progression of fatty liver [48-51]. In view of the function of these proteins, SA might disturb lipid homeostasis by impairing the catabolic metabolism of lipid droplets and energy metabolism. Nevertheless, the molecular mechanism behind autophagy machinery contributing to SA-mediated deregulation of lipid homeostasis remains incompletely understood. Overall, congruous data have been reported suggesting the alteration of key molecules regulating lipid metabolism, which again demonstrates the detrimental effects of SA on hepatic lipid metabolism. The disturbed hepatic lipid metabolism appears to be the key event in SA-induced lipid accumulation.

Response or adaption to environmental stress such as the change in pH or food preservatives is important for the survival and proliferation of living organisms [52]. For example, upon weak acid exposure, the membrane permeability and anion extrusion in Saccharomyces cerevisiae were decreased and the gene-related to metal metabolism and vacuolar H+-ATPase were upregulated to adapt the environmental changes in pH value [53]. In mammalian cells, autophagy is reported to be protective for gastric cancer cells against the acidic environment [54]. However, there is no study investigating the acid stress response in hepatocytes. Although SA is a weak organic acid, we thought that SA may not induce acid stress response in hepatocytes. Because plasma or tissue fluid has a strong capacity of buffering to maintain the pH value within normal range under physiological or pathological condition, to prove our hypothesis, we additionally performed in vitro study to examine whether SA affects the pH value in the culture system of murine hepatocyte cell line AML12 cells. Our results demonstrated that treatment with SA (1 g/L) did not change the pH value of culture medium at time 0 (vehicle group: SA group = 7.49: 7.49) and 18 h after SA treatment (vehicle group: SA group = 7.34: 7.34). This finding suggests that SA may not induce acid stress response in hepatocytes.

Metabolic diseases including fatty liver disease are complex, progressive disease processes with multiple etiologies [55-57]. Although SA did not have a significant impact on appearance and weight; it indeed disturbed the regulation of hepatic lipid metabolism leading to lipid accumulation in the liver. This observation suggests that SA intake might be a risk factor in the development of metabolic diseases under certain circumstances, in developed countries. In this study, we found that SA decreases lipoprotein secretion, autophagy and β-oxidation. All these events induced by SA could speed up the progression of fatty liver. In addition, our results from a cellular model showed that SA exacerbated oleic acid-induced lipid accumulation in Huh7 hepatoma cells, a fatty liver cell model (data not shown). Nevertheless, our study has several limitations since, for experimental analysis, we did not use hyperlipidemic or fatty liver animal models to mimic the pathogenesis of metabolic diseases. Using specific hyperlipidemic mouse models will be helpful to clarify the effects of SA on lipid homeostasis and disorders under these pathological conditions. In addition, the sample size of this study is rather small (n = 5), the experimental design may not have enough power to detect differences between the 2 groups. Indeed, upon SA treatment for 4 weeks, there is an increasing trend in the WAT weight, BAT weight, and their ratios with respect to body weight, as well as the plasma level of HDL-c; however, the p value didn’t reach the level of statistically significant. Increasing the number of samples may improve such a defect in this study. Moreover, we do not have clinical data to support our observations from animal studies. To this end, further investigations describing the implications of food preservatives in hepatic lipid metabolism and related metabolic diseases are warranted.

5. Conclusion

Collectively, our study provides new evidence that SA has detrimental effects on hepatic lipid metabolism by decreasing de novo lipogenesis, fatty acid internalization, VLDL secretion, autophagy, and β-oxidation, thereby promoting lipid accumulation in the liver. Moreover, we discovered a link between a food preservative and hepatic lipid metabolism, which broadens the biomedical implications of SA in the development of metabolic diseases.