The main finding of the present study is that animals fed a WD (high fat + high sucrose) for only 2 weeks depicted lower gene expression of several key markers of liver cholesterol metabolism including LDL-R, PCSK9, SREBP2, and HMGCoAR compared to animals fed a SD or a HFD. The same response was also observed for PCSK9 measured in plasma. These findings were not observed when rats were fed the HFD. The 2-week WD also resulted in higher plasma total cholesterol levels. The HFD and even more so the WD promptly (2 weeks) induced excessive TG accumulation in the liver. These results indicate that the WD, which does not contain higher levels of cholesterol, disturbs the mechanisms involved in the regulation of cholesterol metabolism in the liver. The current results indicate that the enrichment in sucrose (50% glucose and 50% fructose) and fat of the WD promptly and potently promoted TG accumulation in the liver. Hence, this effect may be put forward as the element responsible for the alterations observed in the regulation of cholesterol metabolism. From a clinical perspective, these results suggest that high plasma cholesterol levels may be linked to lipid metabolism in the liver.
The first observation of the present study is that short-term (2 weeks) exposure to either a HFD (60% fat) or a WD (40% fat + 35% sucrose (17.5 % fructose)) induced important increases in the liver TG levels and this effect remained stable after 6 weeks of treatment. An early accumulation of lipids in the liver in response to high-fat feeding has been previously observed (11, 15, 16). This response has been interpreted as if the liver was acting as a buffer to protect the other organs from a potentially deleterious surge of lipids (11, 17). Accordingly, plasma TG levels were not changed after 2 weeks of HFD or WD. An additional finding of the present study is that liver TG accumulation after 2 weeks was even higher in the WD than in the HFD. Fructose contained in the WD is a strong inducer of hepatic lipogenic enzymes (18). Therefore, the present increases in gene expression of ACC and ChREBP measured in the liver of animals submitted to 2 weeks of WD support the assumption that de novo lipogenesis is associated with the excessive accumulation of fat in this organ.
In addition to resulting in a rapid increase in liver TG accumulation, the short 2-week period of feeding applied in the present study has the advantage of preceding significant increases in body weight gain and the systemic development of metabolic complications associated with longer exposure to obesogenic diets. In support of this, the body weight observed among animals fed the HFD and WD was similar to the body weight observed among rats fed the SD diet after 2 weeks. Body weight was approximately 50 g higher among the animals fed the obesogenic diets after 6 weeks. Moreover, the sum of abdominal and subcutaneous fat accumulation was approximately 3 times higher after 6 weeks of the obesogenic diets than it was after only 2 weeks (7 vs 20 g), all while liver TG levels were similar after 2 and 6 weeks. It is, therefore, our view that the present 2-week feeding experimental design is appropriate to isolate the metabolic effects of an increase in liver fat. For this reason, the present discussion is centered on results derived from the 2-week feeding condition.
Using this experimental design, the present study aimed to gather evidence for the existence of an association between a rapid increase in liver TG among rats submitted to short-term (2 weeks) obesogenic diets and disturbances in gene expression of key regulators of cholesterol metabolism in the liver. The main finding of the present study was that feeding animals with a WD for both 2 and 6 weeks resulted in a decrease in gene expression of three important and related markers of liver cholesterol metabolism: LDL-R, PCSK9, and the transcription factor SREBP2. The combined action of these 3 molecules regulates transport of highly atherogenic LDL particles. With approximately 70% of its expression carried out in the liver, it is clear that LDL-R is crucial for the clearance of LDL-cholesterol particles from circulation (19). Therefore, the increase in plasma total cholesterol observed in rats fed a WD at 2 and 6 weeks could be the result of the decrease in gene expression of liver LDL-R.
Our results also showed that the levels of mRNA and plasma PCSK9 (a protease that down-regulates LDL-R protein expression (20, 21, 22)) are reduced at 2 and 6 weeks after WD feeding. This could have normally contributed to promoting the clearance of plasma cholesterol by increasing LDL-R protein levels. However, because of the decrease in upstream gene expression of SREBP2, a transcription factor that regulates the expression of both LDL-R and PCSK9 (23, 24), the reduction in PCSK9 levels did not positively affect the expression of LDL-R. Consequently, plasma total cholesterol remained high in rats fed the WD at 2 and 6 weeks. Similarly, gene expression of LRP1, responsible for the removal of circulating lipoprotein remnants enriched in cholesterol (25), was reduced in rats fed the WD, thus potentially contributing to the increased plasma cholesterol levels. In addition, cholesterol transport from HDL to the liver is carried out through its interaction with receptor SR-B1 (26) whose gene expression remained unchanged by the WD while being modulated by the HFD. All of these responses may have contributed to the decrease in cholesterol uptake by the liver and, in turn, promoted its increased concentration in the circulation.
Along with LDL-R and PCSK9, the expression of HMGCoAR, the key marker of cholesterol synthesis in the liver, is regulated by the transcription factor SREBP2 (23, 24). Accordingly, SREBP2 and HMGCoAR mRNA levels were highly reduced in rats fed the WD. At the same time, mRNA expression of ACAT-2, which is involved in cholesterol esterification was reduced in rats fed the WD compared to rats submitted to the HFD. Taken together, these results support the hypothesis that in the liver, a short-term WD yielding a large increase in the TG content impairs pathways involved in regulation of cholesterol metabolism.
If liver cholesterol uptake from the circulation is reduced under the present WD, the next question is what can trigger this response? The accumulation of cholesterol in the liver may be an important factor inducing the reduction in gene expression of key markers of the LDL-R pathway and, in turn, the increase in plasma cholesterol levels (23). Liver cholesterol concentrations show a tendency (P < 0. 1) to be higher after 2 weeks of both the HFD and WD, a tendency that became highly significant for the WD after 6 weeks. In this regard, it is noticeable that gene expression of HMGCoAR was lowered in rats fed the HFD and even more so in rats fed the WD for 2 weeks. This response may have been stimulated by a transient rise of cholesterol levels in the liver. In addition, the low mRNA levels of ABCG5/G8 transporters (responsible for the exportation of cholesterol in the bile ducts) in rats fed the WD might have contributed to transient higher liver cholesterol levels in these animals. However, this association does not hold for the HFD-fed animals since ABCG5/G8 mRNA levels were increased in these groups of rats. It is likely that other mechanisms such as fecal excretion of biliary acids (27) or increased synthesis and secretion of VLDL are involved
Other important questions that emerge from the present results are what factors alter the expression of key metabolic markers of hepatic cholesterol homeostasis in response to 2 weeks of WD, and why does the response in animals submitted to the HFD not reach the same extent? It must be taken into account that after 2 weeks, hepatic TG levels were higher in WD- than in HFD- fed rats, and this might have interfered with cholesterol metabolism. It is also possible that the different fat composition of the two diets might have played a role. A key element of the WD diet is its content of fructose. Fructose has been reported to increase protein levels of all de novo lipogenesis enzymes during its conversion into triglycerides (28, 29). The WD-induced lipogenesis necessitates the synthesis of malonyl-CoA, which is not the case for dietary fat. Malonyl-CoA is a potent inhibitor of mitochondrial oxidation (30). By doing so, increased malonyl-CoA levels may have indirectly contributed to the higher level of liver fat accumulation in animals submitted to the WD. A second interaction between fructose and cholesterol metabolism is that lipogenesis and cholesterol synthesis pathways both use acetyl CoA as a molecular precursor and both use the reducing equivalent NADPH as a sole source of energy. Although other factors may be involved, this may have contributed to the lower gene expression of HMGCoAR found in rats treated with the WD compared to rats fed the HFD for 2 weeks. In one of the rare studies evoking the potential link between fructose and hepatic cholesterol metabolism, Ichigo et al (12) reported a decrease in ABCG5/G8 gene expression in rats submitted to 12 days of high fructose compared to rats fed a high glucose diet. Although comparisons between this study and ours remain difficult, these authors did hypothesize that a fructose diet might affect genes involved in regulation of cholesterol metabolism.
Although the precise mechanism linking liver fat accumulation to the disruption of liver cholesterol metabolism is beyond the scope of the present study, our results clearly indicate that a short-term WD disturbs hepatic cholesterol metabolism while promoting a rapid increase in total cholesterol circulating levels. Results derived from previous reports show that cholesterol homeostasis in the liver may be perturbed by activities of other metabolic pathways. For instance, Kemper et al (31) reported that the ingestion of an ester of B-hydroxybutyrate in rats promoted an increase in LDL-R and SREBP2 protein levels. More recently, Rana et al (32) reported that the cell surface protease-activated receptor PAR2 is a factor linking gene expression of several markers of cholesterol metabolism (i.e. LDL-R, SR-B1) to increased TG levels in the liver. This effect could potentially be mediated through the action of the de novo lipogenesis pathway. On a long-term basis, fructose has been highly associated with the development of liver steatosis in humans (33). Several pathways modulating hepatic cholesterol metabolism are dysregulated in response to NAFLD (7). Hence, the difficulty in determining the direct association between liver fat accumulation and impairments in hepatic cholesterol homeostasis is increased by the knowledge that NAFLD is also associated with several metabolic disorders (i.e. inflammation and insulin resistance, etc.), which, in turn, alter cholesterol metabolism in the liver.
Although protein levels were not measured in the present study, mRNA levels of several key markers of the LDL-R pathway display the same pattern of responses. This supports the interpretation that the 2-week WD used as a model of isolated hepatic steatosis most likely leads to a rapid increase in circulating cholesterol levels. It is currently accepted that fructose enrichments of WD are promoting de novo lipogenesis in the liver. Furthermore, the present data indicate that this excessive accumulation of lipids may also interfere with the regulation of cholesterol metabolism in the liver.