CM1 intervention alleviated hyperlipidemia
LDLR(+/−) hamsters have an autosomal inherited hypercholesterolemia [15, 24]. The body weight of the hamsters in the NC, HFD, ezetimibe, and CM1 group increased approximately 29.1, 40.4, 41.0, and 32.1%, respectively, after feeding for 5 months (Fig. 1C, P < 0.01). Although the average food intake had no obvious differences among groups (Fig. 1D), CM1 intervention significantly decreased the final body weight of the hamsters (Fig. 1C, P < 0.05). Of note, HFD dramatically increased the average plasma TC (Fig. 2A, 128.8 vs 393.9 mg/dL) and TG (Fig. 2B, 104.6 vs 238.5 mg/dL) levels of the LDLR(+/−) hamsters (P < 0.01). In line with previous studies [4, 24], ezetimibe also significantly decreased the elevated plasma TC level by ~ 72% and TG level by ~ 49% (P < 0.01). CM1 intervention notably decreased the plasma TC level by ~ 28% (Fig. 2A, P < 0.05, 283.2 vs 393.9 mg/dL) and TG level by ~ 16% (Fig. 2B, 201.2 vs 238.5 mg/dL). Furthermore, CM1 intervention significantly decreased the low-density lipoprotein (LDL) and high-density lipoprotein (HDL) cholesterol levels (Fig. 2C). It also reduced TG level in the very low-density lipoprotein (VLDL) fractions (Fig. 2D).
In line with HDL cholesterol, HFD dramatically enhanced the plasma apoAI level (Fig. 2E, P < 0.01). Of note, ezetimibe and CM1 intervention decreased the elevated plasma apoAI level by approximately 29% and ~ 36%, respectively (Fig. 2E, P < 0.05). Hamsters have both apoB100 and apoB48 in the plasma, which are mainly carried by non-HDL particles [15, 25, 26]. Ezetimibe reduced the elevated plasma apoB100 and apoB48 levels by approximately 54 and 69%, respectively (Fig. 2F, P < 0.01). Although CM1 intervention had no effect on the plasma level of apo100, this molecule significantly reduced the plasma level of apoB48 by 27% (Fig. 2F, P < 0.05). The alteration of plasma apoB48 was consistent with the plasma levels of TG. HFD also increased the plasma LPL level by 119% (Fig. 2G, P < 0.01). Of note, ezetimibe intervention significantly increased the level of plasma LPL protein by nearly 38% (P < 0.05). However, CM1 intervention did not affect the plasma level of LPL (Fig. 2G). In addition, HFD significantly increased plasma LPL activity (Fig. 2H, P < 0.01). In this study, neither ezetimibe nor CM1 intervention affected the plasma LPL activity (Fig. 2H).
CM1 intervention modulated the liver genes
SREBP-2 modulate the expression of several genes, such as PCSK9 and LDLR, which are involved in cholesterol metabolism at the transcriptional level [16, 18, 27]. HFD dramatically reduced the gene expression of SREBP-2 and PCSK9 by approximately 26 and 78%, respectively (Fig. 3A and B, P < 0.01). Ezetimibe increased the gene expression of SREBP-2 by ~ 2.8-fold (Fig. 3A, P < 0.01) and PCSK9 by ~ 3.8-fold (Fig. 3B, P < 0.01) in comparison with the HFD group. Of note, CM1 intervention reduced the mRNA levels of SREBP-2 and PCSK9 by approximately 88 and 80%, respectively (Fig. 3A and B, P < 0.01). Furthermore, CM1 intervention also dramatically decreased the expression of these genes compared to the ezetimibe treatment (Fig. 3A and B, P < 0.01). Therefore, CM1 may decrease cholesterol synthesis at the transcriptional level.
LXRα is an important modulator of cholesterol metabolism [28]. HFD did not affect the gene expression of LXRα in this study (Fig. 3C). However, HFD dramatically increased the mRNA level of SREBP-1c by around 32-fold (Fig. 3D, P < 0.001). Ezetimibe treatment notably decreased the gene expression of LXRα by 28% (P < 0.05) and SREBP-1c by 80% (Fig. 3C and D, P < 0.01). CM1 also reduced the gene expression of SREBP-1c by approximately 61% (Fig. 3D, P < 0.05), but not LXRα (Fig. 3C). In this study, the Ct numbers of LDLR, CYP7A1, and ABCG8 were greater than 30, suggesting the undetectable of these three genes.
CM1 intervention improved the levels of CYP7A1 and ABCG5
HFD notably increased the level of LDLR protein by 1.6-fold and PCSK9 protein by 53%, but not that of SR-BI and SREBP-2 (Fig. 4A-D). However, ezetimibe or CM1 administration had no effect on SR-B1 (Fig. 4A). Of note, ezetimibe decreased the protein expression of LDLR by 64% (Fig. 4B, P < 0.01). Compared to ezetimibe, CM1 notably increased the amount of LDLR protein (Fig. 4B, P < 0.05). As shown in Fig. 4D, CM1 dramatically decreased the expression of PCSK9 (38%, P < 0.05), a protein that can promote LDLR degradation [16, 29]. The changes of LDLR protein in ezetimibe and CM1 intervention groups were consistent with the alteration of PCSK9. Additionally, the expression of SREBP-2 had no significant difference in ezetimibe or CM1 intervention group (Fig. 4C).
CYP7A1 is a key enzyme for bile acid synthesis [30]. CM1 treatment, but not ezetimibe, significantly enhanced the amount of CYP7A1 protein (Fig. 4E, P < 0.05). A proportion of cholesterol metabolites in the liver are transported to the gall bladder for excretion [28, 31]. In the present study, HFD and CM1 had no effect on the ABCG8 and LXRα proteins (Fig. 4F and H). However, CM1 treatment notably elevated the amount of hepatic ABCG5 protein in comparison with the HFD or ezetimibe treatment group (Fig. 4G, P < 0.05). In the liver, CM1 intervention did not affect the level of NPC1L1 protein (Fig. 5A), which mediates the reabsorption of biliary cholesterol [31, 32].
CM1 intervention modulated TG metabolism-related proteins in the liver of LDLR(+/−) hamsters
HFD enhanced the level of PPARα protein compared to the NC group (Fig. 5C, P < 0.05). Ezetimibe had no effect on SREBP-1c and PPARα proteins in comparison with the HFD group (Fig. 5B and C). However, ezetimibe decreased the expression of PPARβ by approximately 40% (Fig. 5D, P < 0.05) and increased the levels of PPARγ and LPL protein by 35 and 43%, respectively (Fig. 5E and F, P < 0.05). Of note, CM1 intervention enhanced the level of PPARα protein by approximately 43% (Fig. 5C, P < 0.05), but not that of SREBP-1c, PPARβ, or PPARγ (Fig. 5B, D and E). Furthermore, CM1 intervention notably enhanced the expression of PPARβ protein in comparison with the ezetimibe intervention (Fig. 5D, P < 0.01). Additionally, CM1 also increased the expression of LPL protein by 70% (Fig. 5F, P < 0.01) as that of ezetimibe.
CM1 intervention inhibited the protein expression of NPC1L1 and SREBP-2 and enhanced the LXRα/ABCG8 in the gut
In this study, HFD increased the mRNA expression of NPC1L1 by approximately 74% (Fig. 6A, P < 0.05). However, ezetimibe or CM1 intervention had no effect on the mRNA expression of NPC1L1. Furthermore, HFD notably reduced the mRNA expression of LXRα and ABCG8 by 92 and 41.5%, respectively (Fig. 6B and C). Ezetimibe reduced the mRNA level of ABCG8 by 73.8%, but not LXRα, compared to the high-fat diet group (Fig. 6 C). On the contrary, CM1 intervention increased the mRNA expression of LXRα by 15.8-fold and ABCG8 by 1.6-fold (Fig. 6B and C, P < 0.01) compared to the HFD group. In contrast to ezetimibe, CM1 intervention also enhanced the mRNA expression of LXRα and ABCG8 (P < 0.01, Fig. 6B and C). Additionally, SREBP-2 was undetectable in the small intestine due to the Ct number was greater than 30.
Compared to the NC group, HFD increased the expression of NPC1L1 protein by approximately 4.6-fold (Fig. 6D, P < 0.01). Ezetimibe intervention showed no effect on the NPC1L1 protein in the present study. Mechanistically, ezetimibe prevents sterol-induced internalization of NPC1L1 [33, 34]. Of note, CM1 administration decreased the elevated NPC1L1 protein by around 39.5% (Fig. 6D, P < 0.05). HFD also increased the level of ABCG8 protein (P < 0.01) and decreased the LXRα protein (P < 0.05) in the small intestine (Fig. 6E and F). Compared to the HFD group, CM1 intervention dramatically increased the level of ABCG8 protein (Fig. 6F, P < 0.05), but not LXRα, in the small intestine. Furthermore, HFD intervention increased the level of SREBP-2 protein by 48% (P < 0.05) compared to the NC group (Fig. 6G). Ezetimibe significantly decreased the elevated SREBP-2 by 42% (Fig. 6G, P < 0.05). Similarly, CM1 intervention reduced the expression of SREBP-2 by 64% (Fig. 6G, P < 0.01) when compared with the HFD group. The inhibitory effect of CM1 on SREBP-2 was greater than that of ezetimibe (38% reduction, Fig. 6G, P < 0.05).
CM1 modulated the lipid metabolism in the Epididymal fat
In this study, HFD increased the fat pad index of the LDLR(+/−) hamsters by approximately 74% (Fig. 7A, P < 0.01). CM1 administration, but not ezetimibe, significantly decreased the elevated fat pad index by around 39% (P < 0.05). HFD also increased the diameter of the adipocyte by 28.3% (P < 0.01), whereas CM1 treatment decreased the elevated diameter of the adipocyte by 34.9% (Fig. 7B and C, P < 0.01). Moreover, HFD increased the expression of PPARα and SREBP-1c proteins by approximately 48 and 38%, respectively, in the epididymal fat (Fig. 7D and F, P < 0.05). Of note, ezetimibe administration significantly decreased the expression of SREBP-1c (Fig. 7F) by 37%, and enhanced the expression of PPARα by 58% and PPARγ by 75% (Fig. 7D and E). Similarly, CM1 intervention reduced the expression of SREBP-1c by 49% (P < 0.05) and increased the level of PPARα by 46% (P < 0.05) when compared with the HFD group. Furthermore, CM1 intervention decreased the level of PPARγ protein by approximately 67% (Fig. 7E, P < 0.01). In the adipose tissue, ATGL promotes the hydrolysis of TGs and the production of fatty acids, thereby playing an important role in energy homeostasis [35]. As shown in Fig. 7G, HFD significantly decreased ATGL protein by approximately 38% (P < 0.01) compared to the regular chow diet. Ezetimibe or CM1 intervention enhanced the level of ATGL protein by 50 and 65%, respectively, compared to the HFD group (Fig. 7G, P < 0.05).
CM1 decreased the lipid droplet formation in vitro
As shown in Fig. 8A, insulin successfully induced the formation of lipid droplet in 3T3-L1 cells, whereas CM1 intervention obviously decreased the lipid droplet formation. In this study, lipid droplet formation was not observed in the blank group. Therefore, the effect of CM1 intervention was only compared to the differentiated group. Statistically, CM1 intervention decreased the average number lipid droplets by 54.2% (P < 0.01). It also reduced the average diameter of lipid droplets by 29.7% (Fig. 8C, P < 0.01). In the differentiated group, the mRNA expression of PPARγ increased by 13.7-fold (Fig. 8E, P < 0.01). Furthermore, the levels of stearoyl-CoA desaturase 1 (SCD1), diacylglycerol acyltransferase (DGAT) 1 and 2 enhanced 1.4-fold (P < 0.01), 55.8% (P < 0.05), and 1.4-fold (P < 0.01), respectively, compared with the blank group (Fig. 8H, I, and J). PPARα siRNA dramatically reduced the mRNA expression of PPARα by approximately 59% compared to the scrambled siRNA group (P < 0.05), while CM1 intervention increased the mRNA expression of PPARα by approximately 72% compared to the single PPARα siRNA treatment group (P < 0.05, Fig. 8D). The gene expression of fatty acid synthase (FAS) and acetyl-CoA carboxylase 1 (ACC1) decreased by approximately 29% in the differentiated group compared to the blank group (Fig. 8F and G, P < 0.05). The above results further demonstrated that PPARγ is important for adipocyte differentiation. It is worth noted that CM1 intervention decreased the mRNA expression of PPARγ, DGAT1, and DGAT2 by 83.8% (Fig. 8E, P < 0.01), 43.8% (Fig. 8I, P < 0.05), and 74.7% (Fig. 8J, P < 0.01), respectively. Furthermore, CM1 intervention did not affect the mRNA expression of PPARα, SCD1, FAS, and ACC1 when compared with the differentiated group.