Apo-10'-Lycopenoic Acid, a Lycopene Metabolite, Increases Sirtuin 1 mRNA and Protein Levels and Decreases Hepatic Fat Accumulation in ob/ob Mice

Abstract

Lycopene has been shown to be beneficial in protecting against high-fat diet-induced fatty liver. The recent demonstration that lycopene can be converted by carotene 9′,10’-oxygenase into a biologically active metabolite, ALA, led us to propose that the function of lycopene can be mediated by ALA. In the present study, male ob/ob mice were fed a liquid high-fat diet (60% energy from fat) with ALA supplementation (ALA group, 240 μg · kg body weight⁻¹ · d⁻¹) or without ALA supplementation as the control (C group) for 16 wk. Steatosis, SIRT1 expression and activity, genes involved in lipid metabolism, and ALA concentrations in the livers of mice were examined. The results showed that ALA supplementation resulted in a significant accumulation of ALA in the liver and markedly decreased the steatosis in the ALA group without altering body and liver weights compared to the C group. The mRNA and protein levels of hepatic SIRT1 were higher in the ALA group compared to the C group. SIRT1 activity also was higher in the ALA group, as indicated by the lower levels of acetylated forkhead box class O1 protein levels. In addition, the mRNA level of acetyl CoA carboxylase 1 was significantly lower in the ALA group than in the C group. Because SIRT1 plays a key role in lipid homeostasis, the present study suggests that the lycopene metabolite, ALA, protects against the development of steatosis in ob/ob mice by upregulating SIRT1 gene expression and activity.

Introduction

NAFLD⁷, which is commonly associated with obesity, is one of the major causes of chronic liver diseases (1) and a serious health problem in many countries. In the early stage of NAFLD, TG accumulate in the cytoplasm of hepatocytes, leading to hepatic steatosis or fatty liver. Although generally asymptomatic and reversible, the fatty liver may not be benign and can prime the liver to progress to more serious pathological lesions, including steatohepatitis, fibrosis, cirrhosis, and even hepatocellular carcinoma (2). We recently demonstrated that hepatic fat accumulation exacerbates liver injury due to carcinogen and alcohol exposure (3, 4). Currently, there are no effective agents available to treat NAFLD. Thus, the investigation of effective substances from dietary sources that are able to mitigate the development of hepatic fat accumulation is warranted.

Lycopene, one of the major carotenoids in Western diets, is the most predominant carotenoid in human plasma. A high intake of lycopene or lycopene-rich foods is associated with reduced risk of developing chronic diseases. In particular, serum and liver lycopene concentrations were significantly decreased in patients with NAFLD (5, 6). Many experimental studies have shown that lycopene supplementation effectively suppresses the development of various liver injuries (7–10). Recently, we and others have shown that lycopene supplementation protects against high-fat diet-induced steatohepatitis and resultant hepatic carcinogenesis (11, 12). Although the exact mechanism of this protective effect is unclear, there is evidence from both mice and cell culture studies that lycopene has multifaceted biological actions (13). In particular, the recent characterization and study of carotene 9′,10’-oxygenase in mammalian tissues has demonstrated that this enzyme can catalyze the cleavage of lycopene to form apo-10’-lycopenal, apo-10-lycopenol, and ALA from lycopene (14). Several reports, including our own, suggest that the biological activities of lycopene may be mediated in part by lycopene metabolites in vitro and in vivo (15–18), which have displayed unique biological activities. A series of apo-lycopenals, including apo-10’-lycopenal, has recently been identified in human plasma (19). These studies led us to propose that lycopene metabolites may have an important role in the modulation of hepatic lipid metabolism. Clearly, this hypothesis needs supporting evidence.

SIRT1 is a NAD⁺-dependent protein deactylase that catalyzes the deacetylation of many nonhistone proteins, including p53, NF-κB, FOXO1, and LXR (20), and has been implicated as a key regulator of lipid metabolism (21). The acute downregulation of SIRT1 in mouse livers, using adenoviral delivery, suppressed the expression of genes involved in fatty acid β-oxidation while upregulating the expression of lipogenic genes (22). Similarly, the hepatocyte-specific deletion of SIRT1 in mice upregulated lipogenic enzymes such as ACACA, resulting in severe fat accumulation in liver tissue after high-fat diet feeding (23). On the other hand, the overexpression of SIRT1 prevented the development of fatty livers in mice fed a high-fat diet (24), further supporting the protective role of SIRT1 against hepatic fat accumulation. There is no information on whether SIRT1 can be regulated by lycopene metabolites. Therefore, in the present study, we investigated whether or not ALA supplementation could modulate the hepatic expression of SIRT1 and its associated genes, thereby preventing development of fatty liver in ob/ob mice.

Materials and Methods

Mice and diet.

Male ob/ob mice with the C57BL/6J background (6 wk old, The Jackson Laboratory) were individually housed in plastic cages for a 1-wk acclimation period. Mice were then divided by weight-matching into 2 groups (n = 12 in each group) with ALA supplementation at a dose of 40 μg/g diet (ALA group) or without ALA supplementation as the control (C) group. The rationale for the selection of this ALA dose was based on multiple factors: 1) this dose of ALA significantly increased plasma ALA concentrations (~7 nmol/L) and inhibited lung tumorigenesis in the A/J mouse model in our previous study (16); 2) the dose calculation using an established equation (25) indicated that the dose of ALA at 40 μg/g diet (0.24 mg·kg body weight⁻¹ · d⁻¹) is approximately equivalent to 14.4 mg lycopene/d in a 60-kg adult man; and 3) it has been well documented that rodents absorb much less carotene than humans. Because the ALA concentration in plasma achieved by this dose of ALA supplementation was much lower than reported plasma lycopene concentrations in humans (0.1–1 μmol/L) who consumed 10–30 mg/d of lycopene, the potential function of ALA can be investigated at a “physiologically relevant” condition. The Lieber-DeCarli high-fat liquid diet (60% of total energy from fat, 22% from carbohydrate, and 18% from protein) was purchased from Dyets, Inc. (no. 712037). Mice were group pair-fed for the period of the study. The amount of diet fed to the C group was based on the mean intake of diet by the ALA group from the preceding day. The diets were prepared twice per week and were stored at 4°C in opaque bottles to prevent degradation of ALA. The ALA that was used in this study was chemically synthesized by Dr. Hansgeorg Ernst (BASF) with a level of purity > 99% and added directly into the liquid diet. All animal protocols were approved by the Institutional Animal Care and Use Committee at the USDA Human Nutrition Research Center on Aging at Tufts University.

HPLC analysis.

Liver tissues were homogenized in a mixture of saline and ethanol (1:2, v:v), extracted with a mixture of hexane and ether (1:1, v:v), and reconstituted in 100 μL ethanol and ether (1:2, v:v). A gradient reverse-phase HPLC for ALA analysis was used as previously described with minor modifications (14) (Supplemental Materials and Methods).

Histological examination and TG quantitation.

Formalin-fixed and paraffin-embedded liver tissue was routinely processed for hematoxylin and eosin staining. Liver histology was examined and graded according to the magnitude of steatosis, as described earlier (26, 27). Briefly, the degree of steatosis was graded 0–3 based on the average percentage of fat-accumulated hepatocytes per field at 100× magnification under hematoxylin and eosin staining (0: <6%, 1: 6–33%, 2: 33–66%, and 3: >66%). The sections were photographed and examined by investigators who were not aware of the treatment groups. The total TG contents were determined as previously described (28).

Western-blot analysis.

Western blotting was performed with whole cell homogenates of liver tissues by using the previously described method (29). All primary antibodies were purchased from Santa Cruz Biotechnology, except for the antibody against SIRT1 (Millipore). Blots were developed using ECL Western Blotting system (Amersham) and analyzed with a densitometer (GS-710 calibrated imaging densitometer, Bio-Rad).

Real-time PCR analysis.

Total RNA was extracted using Tri-Pure reagent (Roche Applied Science). cDNA was generated with Moloney murine leukemia virus RT (Invitrogen) as indicated in the manufacturer's manual. Real-time PCR was performed on an Applied Biosystems 7000 sequence detection system, using a Platinum SYBR Green qPCR kit (Roche Applied Science) according to the manufacturer's instructions. Primer sequences and real-time PCR conditions are provided in Supplemental Table 1 and in Supplemental Materials and Methods, respectively. Product purity was confirmed by dissociation curve analysis. The mRNA levels, measured relative to GAPDH mRNA, were determined using the 2^−ΔΔCt method and expressed as fold of the C group.

Statistical analysis.

Results were expressed as means ± SEM. Data from the 2 groups were compared using Student's t test or Mantel-Haenszel chi-square test. P < 0.05 was considered significant.

Results

Body and liver weights and hepatic ALA concentrations.

Body weights of the C (33.5 ± 0.9 g) and ALA (33.6 ± 0.8 g) groups did not differ at the beginning of the study or after 16 wk of treatment, when they were 68.1 ± 0.7 g and 66.6 ± 1.3 g, respectively. Similarly, final liver weights did not differ between the C (2.8 ± 0.1 g) and ALA (2.6 ± 0.2 g) groups. The hepatic ALA concentrations were 17.7 ± 1.4 pmol/g liver in the ALA group, whereas none was detected in the C group (<0.2 pmol/g).

Hepatic steatosis and other histopathology.

The effects of ALA supplementation on hepatic steatosis in ob/ob mice were examined by hematoxylin and eosin staining. As expected, there was severe hepatic steatosis development in the C group (Fig. 1A); 9 of 11 mice had hepatic steatosis with grade 2 or 3 and no mice had grade 0 in the C group (Fig. 1C). ALA supplementation markedly improved hepatic steatosis (Fig. 1B). No mice had hepatic steatosis with grade 3, whereas 2 of 12 mice had grade 0 in the ALA group (Fig. 1B,C). Hepatic steatosis in the ALA group was lower compared to the C group (P < 0.05). However, hepatic TG concentrations were not significantly different between the C group (0.54 ± 0.14 mmol/g protein) and the ALA group (0.49 ± 0.12 mmol/g protein).

FIGURE 1.

Liver steatosis in ob/ob mice fed a C (A) or ALA (B) diet for 16 wk and the distribution of steatosis scores (C). Representative hematoxylin and eosin-stained sections of liver are shown. Original magnifications were 100×. The degree of steatosis was graded 0–3 based on the average percent of fat-accumulated hepatocytes per field at 100× magnification under hematoxylin and eosin staining (0: <6%, 1: 6–33%, 2: 33–66%, and 3: >66%). *Distribution differed from that of the C group, P < 0.05. ALA, apo-10’-lycopenoic acid; C, control.

Hepatic SIRT1 expression and activity.

Hepatic levels of SIRT1 mRNA (Fig. 2A) and protein (Fig. 2B) were higher in the ALA group compared with the C group (P < 0.05). The levels of acetylated FOXO1, a well-known target of SIRT1, were significantly lower in livers of the ALA group compared to those in the C group, indicating that hepatic SIRT1 deacetylase activity was greatly induced by ALA supplementation (Fig. 2C).

FIGURE 2.

Hepatic SIRT1 mRNA (A), SIRT1 protein (B), and acetylation of FOXO1 (C) in ob/ob mice fed a C or ALA diet for 16 wk. Values are mean ± SEM, n = 11–12. *Different from C, P < 0.05. ALA, apo-10’-lycopenoic acid; C, control; FOXO1, forkhead box class O1; SIRT1, sirtuin 1.

Hepatic expressions of genes involved in lipid metabolism.

The hepatic levels of ACACA mRNA were lower (~40%) in the ALA group compared to the C group (P < 0.05). The mRNA level for CPT1a did not differ between the C (1.0 ± 0.2) and ALA (0.8 ± 0.1) groups, nor did those of medium-chain acyl CoA dehydrogenase, which were 1.0 ± 0.3 and 1.2 ± 0.1, respectively. The levels of mRNA for SREBF-1 and SREBF-1 target genes, including SCD1, GPAM, or FASN, also were not affected by ALA treatment (data not shown).

Discussion

In agreement with previous studies (30, 31), all of the ob/ob mice fed a high-fat diet developed hepatic steatosis in this study. Strikingly, 4 of 11 ob/ob mice fed the high-fat diet reached a steatosis grading of level 3, which occupied >66% of the liver per field. Although there was no effect on weight gain, ALA supplementation for 16 wk effectively improved the fatty liver in the ob/ob mice fed the high-fat diet. None of the ALA-supplemented ob/ob mice had a steatosis grading of level 3, whereas steatosis occupied <6% of the liver per field in 2 of 12 mice in the ALA group. These data clearly indicate an important role for the lycopene metabolite ALA in preventing fatty liver development in vivo. It should be mentioned that, unlike histopathologic grading data, hepatic TG concentrations were not significantly different between the 2 groups, likely due to high variability within each group. This discrepancy is likely due to both the regional distributions of steatosis in the liver and the small portion (~25 μg protein) of liver tissue that was used for biochemical analysis. We think that the histopathologic grading of steatosis for the overall area of the liver better reflects the hepatic lipid content.

The present finding is well corroborated by a recent study, which showed that carotene 9′,10’-oxygenase-deficient mice developed severe hepatic steatosis after carotenoid supplementation (32). We recently demonstrated that non-provitamin A carotenoids, such as lycopene, lutein, and zeaxanthin, are preferentially cleaved over pro-vitamin A carotenoids by ferret carotene 9′,10’-oxygenase to form apo-10’-lycopenoids and apo-10’-carotenoids (14, 33). Therefore, the induction of steatosis in carotene 9′,10’-oxygenase–deficient mice could be due to impaired production of apo-10’-lycopenoids. This notion was supported by the present study, which shows that ALA supplementation prevented high-fat diet-induced steatosis in ob/ob mice. The average American diet provides ~9.4 mg/d of lycopene (34). We tested a dose of ALA at 40 μg/g diet, which is approximately equivalent to 14.4 mg/d lycopene in a 60-kg adult man. Therefore, the experimental dose in our study was similar to the usual intake and was much less than those doses that are currently present in dietary supplements (30–60 mg/d of lycopene) and are being tested in prostate cancer clinical trials. Furthermore, the hepatic ALA concentrations that we detected (17 pmol/g) in the mice were much less than the range normally seen in humans (0.1–20.7 nmol/g) (35, 36), indicating that the effect of ALA on hepatic biomarkers takes place at physiologically relevant tissue concentrations of lycopene. Additionally, our preliminary study with a higher dose of ALA (120 μg/g diet) supplementation also resulted in significantly less steatosis in ob/ob mice that were fed a high-fat diet for 16 wk (data not shown).

The most important observation in the present study was that the significant reduction in hepatic fat accumulation by ALA was associated with the induction of SIRT1 mRNA and protein levels as well as its activity in the liver tissue of ob/ob mice. These data indicate a novel role of the lycopene metabolite ALA in the regulation of SIRT1 and its related hepatic lipid metabolism in vivo. Many studies have suggested that SIRT1 plays a key role in protecting against fat accumulation in the liver (21, 23, 24, 37). The activation of SIRT1 by treatment with SIRT1 activators has been shown to improve mitochondrial function, increase fat oxidation, decrease fat synthesis (38–40), and ameliorate NAFLD in mice (41). In the present study, ALA supplementation significantly suppressed one of the SIRT1 targets, ACACA, which is a rate-limiting enzyme in de novo fatty acid synthesis. Enhanced de novo lipogenesis in the liver has been well documented in ob/ob mice (42). Moreover, patients with NAFLD had significantly increased contributions of de novo lipogenesis in the liver tissue compared to healthy controls (43, 44). Therefore, the downregulation of ACACA expression by ALA could inhibit de novo lipogenesis, thereby preventing steatosis in livers of ob/ob mice. Although ALA supplementation did not affect the expression levels of lipolysis enzyme genes such as CPT1a and medium-chain acyl CoA dehydrogenase in our study, it is likely that the ALA-mediated ACACA suppression indirectly upregulates fatty acid oxidation in the liver tissues due to decreased levels of malonyl CoA, the end product of ACACA and an allosteric inhibitor of CPT1. Further, the direct role of ACACA in the development of hepatic steatosis has been evidenced by studies of liver-specific Acaca knockout mice, which showed that the livers of Acaca knockout mice accumulated ~40–70% less TG compared to wild-type controls (45). Together, these data suggest that ALA protects against the development of steatosis in ob/ob mice by targeting SIRT1 gene expression and activity as well as downregulating ACACA gene expression.

The mechanisms by which ALA regulates SIRT1 are unclear. A recent study by Shen et al. (46) reported that treatment with rosiglitazone, a potent PPARγ agonist, induces SIRT1 expression in the liver tissue of alcohol-fed rats and protects against the development of fatty liver. Lycopene has been shown to activate PPARγ-mediated signaling pathways in mammalian cells (47). We have demonstrated that ALA dose-dependently increased both the mRNA expression and protein levels of PPARγ in THLE-2 liver cells and ob/ob mice treated with a carcinogen (K.Q. Hu, Y. Wang, and X-D. Wang, unpublished data). Further studies are needed to determine how ALA regulates SIRT1 expression in the ob/ob mouse model. In addition, we have examined certain markers of liver injury in plasma, such as inflammatory cytokines, and pathologic lesions in the liver (e.g. inflammatory foci formation and fibrosis). However, we did not find a severe degree of inflammation or fibrosis or any differences on inflammation biomarkers between the two groups of ob/ob mice (data not shown). This result is in agreement with previous reports by others indicating that ob/ob mice are resistant to the development of severe inflammation and fibrosis (31, 48).

In the present study, the ALA-induced SIRT1 significantly decreased the levels of acetylated FOXO1, indicating an increase in SIRT1 enzyme activity by ALA treatment. The deacetylation of FOXO1 by SIRT1 is known to promote the nuclear retention of FOXO1, increasing its transcription activity of many glucogenic genes (49). Also, some studies have reported that FOXO1 can antagonize LXR transcription factors, suppressing the transcription of LXR-targets such as SREBF-1 (50, 51). However, the mRNA level did not differ for G6PC, one of the FOXO1 target glucogenic genes, or for SREBF-1, SCD1, GPAM, or FASN in response to ALA. Similar to our findings, a recent study (52) showed that a polyphenol extract from red grapes decreased hepatic fat accumulation in rats via SIRT1 activation and ACACA inhibition, but there were no differences in the mRNA levels of SREBF-1 and its target genes such as SCD1 and FASN. It seems that the inhibition of steatosis by ALA-induced SIRT1 mainly involved the regulation of ACACA rather than FOXO1/ SREBF-1 and their related target genes. In addition, the inhibition of steatosis by ALA-induced SIRT1 could involve other mechanism(s) as well. Recently, it has been shown that adenovirus-mediated hepatic SIRT1 overexpression leads to a significant reduction in hepatic steatosis in ob/ob mice, largely through the inhibition of the mammalian target of rapamycin complex 1 and endoplasmic reticulum stress (53). We currently are investigating whether ALA is involved in this process.