Caloric restriction (CR) and endurance exercise elicit wide-ranging health benefits including reduced risk of select cancers. In addition, diet composition influences oncogenesis, although its interactions with exercise and CR are not well understood. Therefore, to investigate the potential interactions between diet and lifestyle interventions on liver tumorigenesis, the carcinogen diethylnitrosamine was administered to 72 male C57Bl/6 mice that were subsequently fed diets enriched with lard (CTL) or olive oil and were further stratified to voluntary wheel running (Ex) or 30% CR for 49 weeks. Although Ex and diet composition did not influence liver oncogenesis, CR prevented hepatic tumor formation. In addition, CR reduced steatosis, hepatocyte ballooning, inflammation, and immune cell infiltration, all of which are hallmarks in the progression of nonalcoholic fatty liver disease to liver tumorigenesis. RNA sequencing of nontransformed liver tissues from CR mice revealed changes in metabolic pathways and reduced inflammation, cytokine production, stellate cell activation and migration, and genes associated with liver injury and oncogenesis. These data demonstrate that CR protects against steatosis, liver inflammation, and liver injury and is a robust deterrent of carcinogen-induced hepatic oncogenesis. Cancer Prev Res; 10(11); 660–70. ©2017 AACR.

Alterations in energy metabolism resulting from overconsumption of calories are well known to increase the risk of obesity, which accounts for nearly 20% of all cancers in the United States (1). In addition, prospective epidemiologic studies have shown that obesity increases the risk of mortality from nearly every major form of cancer including liver cancer (2). The mortality rate for nonalcoholic fatty liver disease (NAFLD)-derived hepatocellular carcinoma (HCC), the most common liver cancer, is higher than other etiologic factors such as hepatitis viral infection with mortality rates as high as 61% within 1 year of diagnosis (3). Although epidemiologic studies have linked obesity to increased cancer risk and mortality, the molecular mechanisms underlying this relationship have not been fully elucidated.

Adherence to the Mediterranean diet (MD), considered the gold standard for diets, is associated with a lower incidence of cancer and has been shown to reverse the effects of NAFLD in humans (4). Olive oil (OO), which is rich in monounsaturated fatty acids (MUFA), is a food regularly consumed by adherents of the MD has been reported to provide numerous health benefits (5). Studies in mice and humans have shown consumption of high MUFA containing oils, such as OO, to be protective against NAFLD development, even with diets high in fat content (6).

Sedentary lifestyles are well-established contributors to obesity, metabolic disease, and tumorigenesis (1). Exercise, however, has been shown to reduce mortality and protecting against numerous lethal diseases such as cancer (7). Of the two major forms of exercise, endurance and resistance, the former has proven more beneficial to cardiovascular health (8) and obesity (9). Endurance exercise is also an effective therapeutic tool in reversing NAFLD, a major risk factor for liver tumorigenesis (10). Given these benefits, it is perhaps not surprising that numerous studies have demonstrated endurance exercise to be protective against liver tumorigenesis (11). Importantly, physical activity by adherents of the MD lowers the risk of morbidity and mortality more than their inactive counterparts (12), suggesting a cooperative role between the MD and lifestyle interventions.

Unlike increased morbidity and mortality associated with sedentary lifestyles, caloric restriction (CR) has a long-standing history of extending disease-free longevity (13). CR has also been shown to prevent or delay mammary (14), prostate (15), brain (16), intestinal (17), and pancreatic cancers (18). Yet, despite CR-proven effectiveness as a therapy in reversing and preventing NAFLD (19), no hepatic CR oncogenesis studies have investigated exposures and pathologies relevant to Western societies, such as through NAFLD. In addition, although CR and exercise lower the risk for numerous cancers, the molecular pathways involved in elucidating these effects are not fully understood. Therefore, to investigate any collaborative link between dietary fat composition, akin to a Western versus a Mediterranean diet, and either regular aerobic exercise or CR, a long-term liver carcinogenesis study using diethylnitrosamine (DEN) was used. Results show dietary composition and exercise had no effect on tumor development, but CR reduced liver inflammation and injury and abrogated tumorigenesis.

Animals and chemicals

Four-week-old male C57BL/6 mice were purchased from Charles River Laboratories. All mice were maintained at the University of Minnesota Animal Facilities in accordance with the Institutional Animal Care Guidelines and all experimental procedures were approved by the Institutional Animal Care and Use Committee at the University of Minnesota. Mice were individually housed under a 12:12 light/dark cycle with free access to water. DEN and glyceryl triochtonoate were purchased from Sigma-Aldrich. The final injection mixture was prepared by suspending DEN in glyceryl triochtonoate to a concentration of 2.5 mg/mL just prior to injection. At 4 weeks of age, all mice were given a single intraperitoneal injection of DEN mixture (25 mg/kg body weight) to induce liver tumorigenesis. Upon injection, mice were individually housed in cages containing 1/8” irradiated corncob bedding for the duration of the study. A subset of mice (n = 20) were housed in modified cages containing a running wheel, which was retrofitted with a bicycle computer to track weekly distance (km) and wheel revolutions as a means of tracking activity levels. Daily food consumption was determined by weighing CTL and OO diet intakes on consecutive days every week. To achieve CR, daily aliquots of CR diets were prepared by weighing 70% of the average daily food weight consumed by sedentary (Sed) animals; aliquots were prepared weekly and stored at 4°C. Food consumption and body weight were measured weekly. Forty-nine weeks after injection, animals were sacrificed for tissue and serum collection after a 4-hour fast. After anesthesia, livers were harvested by surgical resection and surface nodules were counted.

Diets

To examine differences between dietary fat typically consumed by Western and Mediterranean inhabitants in regards to tumorigenesis, Sed (n = 32) and Ex mice (n = 20) were fed purified diets (AIN-93G) purchased from Harlan Teklad. Sed and Ex mice were stratified to receive diets where 25% of the kcal were derived from fat [15% of kcal of fat was derived from either lard (TD.150656) or OO (TD.150657) and soybean oil contributed the remaining 10%; Supplementary Table S1]. CTL CR or OO CR diets (TD.150658 and TD.150659, respectively) were made to contain 30% less macronutrient content compared with ad libitum diets (10/10/10 CHO/FAT/PRO). CR diets were then supplemented with added vitamin and mineral mix (AIN-93G-MX and AIN93-VX, respectively) to ensure animals remained nutrient sufficient. If initiated at a young age, CR has been shown to have deleterious developmental effects (20). Therefore, we gradually restricted CR mouse food in 10% increments over 3 weeks beginning at 5 weeks of age until a 30% reduction was achieved (Supplementary Fig. S1A).

NAFLD and NASH scoring

Hepatocellular steatosis, inflammation, Mallory body and Kupffer cell infiltration, and ballooning were scored by a pathologist (J.C. Manivel; VA Medical Center, St. Paul, MN) blinded to all treatment groupings.

IHC and histologic analysis

A medial section of liver tissue was excised and fixed in 10% buffered formalin and subsequently embedded in paraffin blocks or placed in optimal cutting media and slowly frozen in liquid nitrogen. Paraffin-embedded sections were prepared for histopathologic and IHC examinations as previously described (21). Cryo blocks were stored at −80°C until preparation. Cryosections of hepatic tissue were stained with Oil-Red-O as described previously (21) and analyzed using ImageJ to determine lipid area per field of five randomly selected views per animal from respective treatment groups.

RNA extraction, library preparation, and next-generation sequencing

Six samples of nontransformed, snap-frozen hepatic tissue were selected using a random number generator from the CTL Sed and CTL CR groups for RNA sequencing. RNA was extracted from snap-frozen liver tissue using the RNEasy RNA Mini Kit purchased from Qiagen. A total of 12 RNA samples [6 samples per group × 2 groups (CTL Sed and CTL CR)] were sent to University of Minnesota Genomics Core (UMNGC) for quality check, library preparation, and sequencing. Eukaryotic RNA isolates were quantified using a fluorimetric RiboGreen assay and total RNA integrity was assessed using capillary electrophoresis (e.g., Agilent BioAnalyzer 2100). Only samples higher than 1 μg with a RIN of 8 or greater proceeded to sequencing. Total RNA samples were converted to Illumina sequencing libraries using Illumina's Truseq RNA Sample Preparation Kit (catalog no. RS-122-2001 or RS-122-2002) or stranded mRNA Sample Preparation Kit (catalog no. RS-122-2101). One microgram of total RNA was oligo-dT purified using oligo-dT–coated magnetic beads, fragmented, and then reverse transcribed into cDNA. The cDNA was fragmented, blunt-ended, and ligated to indexed (barcoded) adaptors and amplified using 15 cycles of PCR. Final library size distribution was validated using capillary electrophoresis and quantified using fluorimetry (PicoGreen) and via qPCR. Indexed libraries were then normalized, pooled, and size selected to 320 bp ± 5% using Caliper XT instrument. Truseq libraries were hybridized to a single read flow cell and individual fragments were clonally amplified by bridge amplification on the Illumina cBot. Once complete, the flow cell was loaded on the HiSeq 2500 and sequenced. Upon completion of read 1, an 8-bp forward and 8-bp reverse (i7 and i5) index read was performed. Base call files for each cycle of sequencing were generated by Illumina Real Time Analysis (RTA) software. Primary analysis and demultiplexing were performed using Illumina bcl2fatstq software version 2.17.1.14.

For the RNA sequencing analysis, 50-bp FastQ Reads (n = 12 million per sample) were trimmed using Trimmomatic (v 0.33) enabled with the optional “-q” option; 3-bp sliding-window trimming from 3′ end requiring minimum Q30. Quality control checks on raw sequence data for each sample were performed with FastQC. Read mapping was performed via Bowtie (v2.2.4.0) using the UCSC mouse genome (mm10) as reference. Gene quantification was done via Cuffquant for FPKM values and Feature Counts for raw read counts. Differentially expressed genes were identified using the edgeR (negative bionomial) feature in CLCGWB (Qiagen) using raw read counts. The generated list was filtered on the basis of a minimum 2× Abs Fold Change and Bonferroni corrected P < 0.05. These filtered genes were then imported to Ingenuity Pathway Analysis (IPA) Software (Qiagen) for pathway identification.

Ingenuity pathway analysis

Isoforms that exhibited a log2-fold change greater than 1 and a FDR less than 0.05 were subjected to IPA (IPA 4.0, Ingenuity Systems, www.Ingenuity.com). The input isoforms were mapped to IPA's knowledge bases, and the relevant biological functions, networks, and pathways related to the treatment were identified.

Reactive oxygen species

Reactive oxygen species were detected from 50 μg of nontransformed hepatic tissue homogenate using the OxiSelect In Vitro Reactive Oxygen Species Kit (Cell Biolabs) per manufacturer's instructions.

Serum analyses

Serum was isolated from whole blood samples taken via cardiac puncture at the time of animal sacrifice and kept on ice until centrifugation. Samples were spun at 5,000 × g for 10 minutes and supernatant was aliquoted and stored at −80°C. Nonesterified fatty acids (NEFA) were analyzed using a kit purchased from Stanbio labs. Total ketone bodies and serum insulin were analyzed using the total ketone body isolation and insulin kits, respectively, purchased from Wako.

Statistical analysis

Statistical analysis was performed using GraphPad Prism7. Data are expressed as mean ± SEM. Statistical analyses were performed using Student t test or ANOVA where appropriate. χ2 test was used for tumor percentages by grouping. Statistical analysis for RNAseq is described in detail under the respective methods. Differences were considered significant at P < 0.05.

Phenotypic effects of diet, Ex, and CR

To determine the independent and potentially synergistic effects of exercise, CR, and diet lipid composition on liver tumorigenesis, 4-week-old male C57BL/6 mice were given a single intraperitoneal of DEN. Diets consisting of higher fat content are well known to contribute to obesity and obesity related liver pathologies. However, little evidence exists examining the role varying dietary fat content plays in tumorigenesis in the setting of obesity. Therefore, we stratified mice to receive diets with fat content being enriched in either lard (to mimic fat content typical of a Western diet) or OO (to mimic fat content typical of a Mediterranean diet; Supplementary Table S1). Mice were further subcategorized to either remain sedentary (CTL Sed or CTL OO) or given unlimited access to running wheels (Ex CTL or Ex OO) or incremental CR (CTL CR or OO CR; Supplementary Fig. S1A). Sed and Ex mice were fed ad libitum, whereas CR mice were fed 70% of calories of their respective sedentary controls. Mice with access to running wheels ran over 45 km during the first week of exposure, but running wheel use gradually declined especially during the first 6 months (Fig. 1A and B; Supplementary Fig. S1B). Within the Ex groups, mice fed the OO diet ran significantly more than those fed the CTL diet over the duration of the study (Fig. 1B). However, as with total running wheel use, the increased running of mice fed the OO diet was only evident during the first 6 months of the study (Supplementary Fig. S1B). OO significantly reduced food intake in Sed mice, but not in those in the Ex group (Fig. 1C and D). As expected, CR reduced body weight gain, which was unaffected by dietary lipid composition (Fig. 1E and F). Consistent with reduced body weight, both CR groups had smaller inguinal fat pads (Supplementary Fig. S1C) although the reduction in epididymal fat pads was only observed in the OO CR mice (Supplementary Fig. S1D). Over the course of the study, both Ex groups gained more body weight than CTL Sed (Fig. 1F) perhaps due to increased muscle mass, which commonly occurs with exercise training (22). There were no treatment effects on food efficiency defined as body weight gain/food intake (Supplementary Fig. S1E).

Figure 1.

Phenotypic responses to treatment regiments. Average weekly running distance for each month of the study (A) and collective average running distance of OO- and CTL-fed mice (B). Average daily food intake by month (C) and average daily food consumption over the course of the study (D). Monthly body weight change (E) and average total body weight change (F). Data are presented as ± SEM. *, P < 0.05 compared to CTL; #, P < 0.05 compared with Sed.

Figure 1.

Phenotypic responses to treatment regiments. Average weekly running distance for each month of the study (A) and collective average running distance of OO- and CTL-fed mice (B). Average daily food intake by month (C) and average daily food consumption over the course of the study (D). Monthly body weight change (E) and average total body weight change (F). Data are presented as ± SEM. *, P < 0.05 compared to CTL; #, P < 0.05 compared with Sed.

Close modal

Caloric restriction prevents hepatic tumorigenesis

Upon termination of the study, 49 weeks after DEN administration, liver surface nodules were quantified. Tumors were present in approximately half of mice in the Sed and Ex groups irrespective of diet (Fig. 2A). However, no surface nodules were observed in any mice in the CR groups (Fig. 2A and B); tumor volume did not differ among any of the Sed or Ex mice (data not shown). Clinical diagnosis of tumors was confirmed by a blinded pathologist from H&E-stained slides and demonstrated a diversity of hepatic tumors including HCCs, hepatoblastomas, adenomas, and cholangiocarcinomas. Pearson correlation analysis demonstrated a significant positive correlation between body weight change or food intake and tumor number (Fig. 2C and D), whereas activity level was not correlated to surface nodules (Fig. 2E).

Figure 2.

CR prevents hepatic tumor formation. A, Scatter plot of quantified visible surface nodules per mouse [n = 10 per group except OO Ex (n = 8) and OO CR (n = 9)] and representative images of livers from mice from each group; horizontal bars represent group mean. B, Groupings of mice by percentages of tumor number. C–E, Pearson correlation analysis of average food intake, average fold body weight change, and average weekly running wheel activity with surface nodule count. F, Representative 10× and 40× images from five randomly selected views per animal (n = 4) from nontransformed tissue of CTL Sed and CTL CR mice and quantification of positively stained nuclei. Data are presented as ± SEM. *, P < 0.05 compared with CTL; #, P < 0.05 compared with Sed.

Figure 2.

CR prevents hepatic tumor formation. A, Scatter plot of quantified visible surface nodules per mouse [n = 10 per group except OO Ex (n = 8) and OO CR (n = 9)] and representative images of livers from mice from each group; horizontal bars represent group mean. B, Groupings of mice by percentages of tumor number. C–E, Pearson correlation analysis of average food intake, average fold body weight change, and average weekly running wheel activity with surface nodule count. F, Representative 10× and 40× images from five randomly selected views per animal (n = 4) from nontransformed tissue of CTL Sed and CTL CR mice and quantification of positively stained nuclei. Data are presented as ± SEM. *, P < 0.05 compared with CTL; #, P < 0.05 compared with Sed.

Close modal

Because no differences in tumor formation were observed between dietary or exercise interventions, subsequent analyses were largely focused on the differences between Sed and CR groups fed the CTL diet. Consistent with resistance to tumor formation, nontransformed livers of CR mice showed an approximately 85% reduction in staining of the proliferative marker Ki-67 (Fig. 2F). Collectively, these data demonstrate that CR robustly protects mice from hepatic tumorigenesis initiated by DEN and promoted through ad libitum caloric intake.

Caloric restriction prevents pathologies associated with NAFLD-driven liver tumorigenesis

A recent model describing the progression of NAFLD to liver cancer suggests that steatosis, inflammation, and cellular damage may act simultaneously rather than in succession in the development of liver tumorigenesis (23). In this context, CR has been shown to improve hepatic lipid metabolism and reduce hepatic inflammation (24). Indeed, compared with Sed mice, liver weights were significantly lower in CR mice from both dietary groups, but higher in the Ex mice (Fig. 3A). The OO diet increased liver weights within the Ex group, but lowered them in response to CR (Fig. 3A). Livers from CTL CR mice showed reduced lipid droplet accumulation as visualized with H&E staining, which was confirmed with Oil Red O staining (Fig. 3C and D). These data were further confirmed via steatosis scoring of H&E slides performed by a blinded pathologist (Fig. 3E), demonstrating that CR prevented steatotic burden. In addition, serum ketone body concentrations were significantly increased in CTL CR mice (Fig. 3B) in the absence of changed serum NEFAs (Supplementary Fig. S1F) and reduced insulin (Supplementary Fig. S1G), suggesting reduced steatosis in these mice is partially explained by increased oxidation of fatty acids. Pearson's correlation analysis showed a strong positive relationship between steatosis with tumor burden (Fig. 3F) supporting the established link between NAFLD and hepatic tumorigenesis (25–27).

Figure 3.

CR prevents hepatic steatosis. Liver weight to body weight ratio (A) of mice (n = 57). B, Serum ketone body concentrations analyzed from CTL Sed and CTL CR mice (n = 7 per group). C, Representative H&E images of livers from CTL Sed and CTL CR mice (n = 10 per group). D, Oil Red O staining and quantification from representative 10× and 40× images from five randomly selected views per animal (n = 4) from nontransformed tissue of CTL Sed and CTL CR mice and quantification of average lipid-stained area per field. E, steatosis scoring assigned to each representative case as performed by a pathologist (J.C. Manivel.) blinded to treatments; horizontal bars represent group means (n = 10 per group except OO Ex n = 8 and OO CR n = 9). F, Pearson correlation of steatosis score with surface nodules (n = 57). E, Larger data points represent multiple mice falling with the same nodule/score intersection. Data are presented as ± SEM. *, P < 0.05 compared with CTL; #, P < 0.05 compared with Sed.

Figure 3.

CR prevents hepatic steatosis. Liver weight to body weight ratio (A) of mice (n = 57). B, Serum ketone body concentrations analyzed from CTL Sed and CTL CR mice (n = 7 per group). C, Representative H&E images of livers from CTL Sed and CTL CR mice (n = 10 per group). D, Oil Red O staining and quantification from representative 10× and 40× images from five randomly selected views per animal (n = 4) from nontransformed tissue of CTL Sed and CTL CR mice and quantification of average lipid-stained area per field. E, steatosis scoring assigned to each representative case as performed by a pathologist (J.C. Manivel.) blinded to treatments; horizontal bars represent group means (n = 10 per group except OO Ex n = 8 and OO CR n = 9). F, Pearson correlation of steatosis score with surface nodules (n = 57). E, Larger data points represent multiple mice falling with the same nodule/score intersection. Data are presented as ± SEM. *, P < 0.05 compared with CTL; #, P < 0.05 compared with Sed.

Close modal

In addition to abrogating steatotic burden, CR also reduced markers of immune cell infiltration. Staining of neutrophils (Fig. 4A) and CD3 T cells (Fig. 4B) and the presence of Kupffer cells (Supplementary Fig. S2A) was reduced in livers of CR mice. Consistent with reduced immune cells, CR significantly attenuated lobular inflammation (Fig. 4C), hepatocyte ballooning (Fig. 4E), and the presence of Mallory bodies (Supplementary Fig. S2B). The pathologic scores associated with lobular inflammation and ballooning, but not Mallory bodies, correlated with tumor burden (Fig. 4D and F; Supplementary Fig. S2C). Despite the frequent association of inflammation with ROS, there were no differences in hydrogen peroxide or total reactive oxygen species in liver tissues of CTL CR mice (Supplementary Fig. S2D). Collectively, this data show that CR reduces hepatic steatosis, inflammation, cellular damage, and immune cell infiltration, markers associated with NAFLD progression that correlate to hepatic tumorigenesis.

Figure 4.

CR abrogates pathologies associated with NAFLD progression. A, IHC staining of neutrophils from representative 40× images from five randomly selected views per animal (n = 4) from nontransformed tissue of CTL and CR mice and quantification of positively stained nuclei. B, IHC staining of CD3+ T cells from representative 40× images from five randomly selected views per animal (n = 4) from nontransformed tissue of CTL Sed and CTL CR mice and quantification of positively stained nuclei. C, Lobular inflammation from H&E staining of representative 40× images from five randomly selected views per animal and blinded pathologist scoring of lobular inflammation [n = 10 per group except OO Ex (n = 8) and OO CR (n = 9)]. D, Pearson correlation of lobular inflammation to surface nodules (n = 57). E, Hepatocellular ballooning from H&E staining of representative 40× images from five randomly selected views per animal and blinded pathologist scoring of lobular inflammation (n = 10 per group except OO Ex n = 8 and OO CR n = 9). F, Pearson's correlation of ballooning to surface nodules (n = 57). D, E, Larger data points represent multiple mice falling with the same nodule/score intersection. Data are presented as ±SEM. *, P < 0.05 compared with CTL; #, P < 0.05 compared with Sed.

Figure 4.

CR abrogates pathologies associated with NAFLD progression. A, IHC staining of neutrophils from representative 40× images from five randomly selected views per animal (n = 4) from nontransformed tissue of CTL and CR mice and quantification of positively stained nuclei. B, IHC staining of CD3+ T cells from representative 40× images from five randomly selected views per animal (n = 4) from nontransformed tissue of CTL Sed and CTL CR mice and quantification of positively stained nuclei. C, Lobular inflammation from H&E staining of representative 40× images from five randomly selected views per animal and blinded pathologist scoring of lobular inflammation [n = 10 per group except OO Ex (n = 8) and OO CR (n = 9)]. D, Pearson correlation of lobular inflammation to surface nodules (n = 57). E, Hepatocellular ballooning from H&E staining of representative 40× images from five randomly selected views per animal and blinded pathologist scoring of lobular inflammation (n = 10 per group except OO Ex n = 8 and OO CR n = 9). F, Pearson's correlation of ballooning to surface nodules (n = 57). D, E, Larger data points represent multiple mice falling with the same nodule/score intersection. Data are presented as ±SEM. *, P < 0.05 compared with CTL; #, P < 0.05 compared with Sed.

Close modal

Nontransformed liver tissue from CR and CTL mice produce significantly distinct gene profiles

Because CR mice developed no visible tumors, transcriptome profiling of nontransformed tissue was performed to elucidate potential mechanisms underlying the tumorigenic process in our model. Initial analysis demonstrated a total of 487 differentially expressed genes (Supplementary Fig. S3A) with the top 20 up-/downregulated genes listed in (Fig. 5A). To better understand the biological relevance of the observed differences in the gene signature profiles of CTL Sed and CTL CR mice, multiple parameters were analyzed using IPA. IPA's initial assessment identified a total of 223 canonical pathways as being significantly changed between CTL CR and CTL Sed groups. Interestingly, of the 10 most significantly impacted pathways, nine were related to hepatic inflammation, immune cell activation, and hepatic fibrosis (Table 1); these data corroborated the histologic analyses (Fig. 4). Further investigation of the identified pathways elucidated cell adhesion/movement (Fig. 5B), toll-like receptor signaling (Supplementary Fig. S4A), lipid/carbohydrate metabolism (Supplementary Fig. S4B), and cell growth and proliferation (Supplementary Fig. S4B) as regulatory networks relevant to transcriptomic alterations. From this information, IPA predicted suppression of several features relevant in the NAFLD progression to liver cancer including hepatic stellate cell activation and proliferation (Fig. 5C) and HCC (Fig. 5D).

Figure 5.

CR produce significantly distinct gene profiles from control mice. A, Heatmap of the top 10 most upregulated and 10 most downregulated genes. B, Cell adhesion/migration pathway was identified as a major signaling pathway by IPA. C, IPA identification of hepatic stellate cell activation and hepatic stellate cell proliferation as top biological functions. D, IPA identification of HCC as the most significant disease associated with gene signatures. B–D, Green targets indicate lower expression in CR compared with CTL and red indicate increased expression; table in B corresponds to B–D.

Figure 5.

CR produce significantly distinct gene profiles from control mice. A, Heatmap of the top 10 most upregulated and 10 most downregulated genes. B, Cell adhesion/migration pathway was identified as a major signaling pathway by IPA. C, IPA identification of hepatic stellate cell activation and hepatic stellate cell proliferation as top biological functions. D, IPA identification of HCC as the most significant disease associated with gene signatures. B–D, Green targets indicate lower expression in CR compared with CTL and red indicate increased expression; table in B corresponds to B–D.

Close modal
Table 1.

The 10 most significant canonical pathways in NT hepatic tissue comparing CTL CR with CTL Sed groups

Ingenuity canonical pathways−Log (P)Ratio (numerical)Genes
Hepatic fibrosis/hepatic stellate cell activation 7.46 17/117 (0.145) CCR5, VCAM1, ICAM1, COL4A1, COL12A1, KLF6, PDGFC, COL1A2, COL1A1, COL6A3, TIMP1,CYP2E1, CD14, SERPINE1, COL27A1, COL3A1, PDGFRB 
Atherosclerosis signaling 7.43 14/78 (0.179) COL1A2, ITGB2, COL1A1, VCAM1, APOA4, ICAM1, IL1RN, CXCR4, CD36, LPL,SERPINA1, PLA2G7, PDGFC, COL3A1 
Altered T cell and B cell signaling in rheumatoid arthritis 7.09 11/48 (0.229) TLR2, HLA-A, IL1RN, TLR1, HLA-DMB, TLR8, HLA-DQA1, TLR7, Tlr13, HLA-DQB1, Tlr12 
Communication between innate and adaptive immune cells 5.54 9/43 (0.209) CXCL10, TLR2, HLA-A, IL1RN, TLR1, TLR8, TLR7, Tlr13, Tlr12 
B-cell development 5.11 5/11 (0.455) PTPRC, HLA-A, HLA-DMB, HLA-DQA1, HLA-DQB1 
Dendritic cell maturation 4.51 12/101 (0.119) TLR2, COL1A2, COL1A1, ICAM1, HLA-A, IL1RN, TYROBP, TREM2, HLA-DMB, HLA-DQA1, HLA-DQB1, COL3A1 
TREM1 signaling 4.09 8/50 (0.16) TLR2, ICAM1, TYROBP, TLR1, TLR8, TLR7, Tlr13, Tlr12 
LPS/IL1-mediated inhibition of RXR function 4.07 15/166 (0.0904) ALDH1B1,SLC10A1, GSTA5, ALAS1,GSTT2/GSTT2B, Cyp2a12/Cyp2a22, CYP3A5, Gstm3, IL1RN,NR1I3, CYP7A1, FABP4, CD14, CYP2A6 (includes others), CYP4A11 
Antigen presentation pathway 4.05 5/17 (0.294) HLA-A, HLA-DMB, HLA-DQA1, HLA-DQB1, CD74 
Ingenuity canonical pathways−Log (P)Ratio (numerical)Genes
Hepatic fibrosis/hepatic stellate cell activation 7.46 17/117 (0.145) CCR5, VCAM1, ICAM1, COL4A1, COL12A1, KLF6, PDGFC, COL1A2, COL1A1, COL6A3, TIMP1,CYP2E1, CD14, SERPINE1, COL27A1, COL3A1, PDGFRB 
Atherosclerosis signaling 7.43 14/78 (0.179) COL1A2, ITGB2, COL1A1, VCAM1, APOA4, ICAM1, IL1RN, CXCR4, CD36, LPL,SERPINA1, PLA2G7, PDGFC, COL3A1 
Altered T cell and B cell signaling in rheumatoid arthritis 7.09 11/48 (0.229) TLR2, HLA-A, IL1RN, TLR1, HLA-DMB, TLR8, HLA-DQA1, TLR7, Tlr13, HLA-DQB1, Tlr12 
Communication between innate and adaptive immune cells 5.54 9/43 (0.209) CXCL10, TLR2, HLA-A, IL1RN, TLR1, TLR8, TLR7, Tlr13, Tlr12 
B-cell development 5.11 5/11 (0.455) PTPRC, HLA-A, HLA-DMB, HLA-DQA1, HLA-DQB1 
Dendritic cell maturation 4.51 12/101 (0.119) TLR2, COL1A2, COL1A1, ICAM1, HLA-A, IL1RN, TYROBP, TREM2, HLA-DMB, HLA-DQA1, HLA-DQB1, COL3A1 
TREM1 signaling 4.09 8/50 (0.16) TLR2, ICAM1, TYROBP, TLR1, TLR8, TLR7, Tlr13, Tlr12 
LPS/IL1-mediated inhibition of RXR function 4.07 15/166 (0.0904) ALDH1B1,SLC10A1, GSTA5, ALAS1,GSTT2/GSTT2B, Cyp2a12/Cyp2a22, CYP3A5, Gstm3, IL1RN,NR1I3, CYP7A1, FABP4, CD14, CYP2A6 (includes others), CYP4A11 
Antigen presentation pathway 4.05 5/17 (0.294) HLA-A, HLA-DMB, HLA-DQA1, HLA-DQB1, CD74 

NOTE: Genes in bold are upregulated.

Because gene signatures, canonical pathways, and downstream networks/signaling pathways corroborated previous data linking NAFLD and subsequent pathologies, such as inflammation and cellular damage, leading to nonalcoholic steatohepatitis (NASH), IPA was used to elucidate downstream effects of pathway and network influences. Initial analysis identified 188 biological functions, diseases, and toxologic outcomes related to our identified differentially expressed gene profiles. The most significant diseases and biological functions were categorized by steatosis, liver damage, liver necrosis, hepatocellular adhesion and fibrosis, hepatic hypertrophy and hyperplasia, and liver cancer; livers from CTL CR mice demonstrated robustly altered gene profiles, attenuating these pathologies in comparison with CTL Sed mice (Table 2). Importantly, these gene signatures matched the biochemical and histologic profiles previously identified (Figs. 4 and 5). Collectively, these results suggest CR leads to unique transcriptomic alterations, which are relevant to numerous canonical pathways and gene networks associated with NAFLD progression to liver tumorigenesis.

Table 2.

The 12 most significant categories of disease/toxologic function in NT hepatic tissue comparing CTL CR to CTL Sed groups

CategoriesDisease of functions annotationPGenesNumber of molecules
Liver damage Injury of liver 9.35E−08 Ccl2, CCR5, CD14, CDKN1A, CXCL10,CYP2E1, CYP7A1, HLA-A, IL1RN, ITGB2, KRT8, LIF, NT5E,SERPINA1, SERPINE1,SLC10A1, TLR2 17 
Liver damage Damage of liver 4.63E−07 Ccl2, CCR5, CD14, CD44, CDKN1A, CXCL10,CYP2E1, CYP7A1, HLA-A, IL1RN, ITGB2, KRT8, LIF, NR1I3, NT5E,SERPINA1, SERPINE1,SLC10A1, TLR2 19 
Liver necrosis/cell death Necrosis of liver 5.21E−05 CD14, CDKN1A, CXCL10,CYP2E1, CYP7A1, E2F1, GADD45B, IL1RN, ITGB2, KRT8, LGALS3, LIF,NR1I3, PTPRC, SERPINE1, TIMP1, TLR7, TYROBP 18 
Liver adhesion Adhesion of hepatocytes 2.27E−04 ICAM1, KRT8, VCAM1 
Liver fibrosis Fibrosis of liver 2.57E−04 COL1A1, COL1A2, HSPB1, IL2RG, LGALS3, PDGFC, PDGFRB, SERPINE1, TIMP1, Ccl2, CXCL10, LGALS3, PDGFC, PDGFRB, TIMP1 15 
Liver cholestasis Progressive familial intrahepatic cholestasis type 2 4.45E−04 CYP7A1, SLC10A1, SLC10A2 
Liver inflammation/hepatitis Inflammation of liver  CD14, CD44, CYP2E1, HLA-A, IL1RN, ITGB2, LGALS3, PTPRC, SERPINE1,SLC10A1, SPI1, TLR7 12 
Liver necrosis/cell death Cell death of liver cells 1.22E−03 CDKN1A, CXCL10,CYP2E1, CYP7A1, E2F1, GADD45B, IL1RN, ITGB2, KRT8, NR1I3, PTPRC, TIMP1, TLR7 13 
Liver cholestasis Hepatic cholestasis 1.62E−03 CD68, CDKN1A, CYP7A1, RDH16, SLC10A1, SLC10A2,Slco1a1 
Liver hypertrophy Hypertrophy of liver 1.94E−03 CYP1A2, NR1I3 
Liver steatosis Hepatic steatosis 4.79E−03 Ccl2, CCR5, CD14,CYP2E1, INSIG2,NR1I3, PDGFC, TLR2 
Liver hyperplasia/hyperproliferation Liver cancer 8.13E−03 B3GALT1, CCDC80, CD36, CD44, CD5L, CDKN1A, CIB3, CLCN5, COL12A1, COL1A1, COL1A2,COL27A1, COL4A1, COL6A3, COTL1, CREB3L2, CSF1R, CSF2RB, CX3CR1, CXCL10, CYP2A6 (includes others), CYP3A5,CYP4A11, CYP7A1, CYP7B1, DDC, DYRK3, E2F1, EHD4, ELMO1, ENDOD1, EPS8L2, GAS6, GCK, GSTA5, HAUS8, HSPA5, ICAM1,LAMA3, LGALS3,LIFR, LOXL1, MAP3K13, MICAL2, MUSTN1, MYO1F, MYOF, NAT8, NID1, NPR2, NR1I3, NT5E, NTRK2, NUPR1, PDGFC, PDIA4, PEX26, PITPNM1, PLS1, POSTN,PPP1R1B, PTPRC, ROBO1,RORC, SDCBP2, SIGLEC1, SLA, SLC10A2, SLC11A1, SLC39A4, SLC7A8, SMPD3, SOCS2, ST5, SYBU, SYK, SYVN1, TGM2, THEMIS, TIMP1, TLR2, TLR7,TOMM40L, TOR3A, TREM2, TRIM2, UCP2,UPP2, VCAM1 89 
CategoriesDisease of functions annotationPGenesNumber of molecules
Liver damage Injury of liver 9.35E−08 Ccl2, CCR5, CD14, CDKN1A, CXCL10,CYP2E1, CYP7A1, HLA-A, IL1RN, ITGB2, KRT8, LIF, NT5E,SERPINA1, SERPINE1,SLC10A1, TLR2 17 
Liver damage Damage of liver 4.63E−07 Ccl2, CCR5, CD14, CD44, CDKN1A, CXCL10,CYP2E1, CYP7A1, HLA-A, IL1RN, ITGB2, KRT8, LIF, NR1I3, NT5E,SERPINA1, SERPINE1,SLC10A1, TLR2 19 
Liver necrosis/cell death Necrosis of liver 5.21E−05 CD14, CDKN1A, CXCL10,CYP2E1, CYP7A1, E2F1, GADD45B, IL1RN, ITGB2, KRT8, LGALS3, LIF,NR1I3, PTPRC, SERPINE1, TIMP1, TLR7, TYROBP 18 
Liver adhesion Adhesion of hepatocytes 2.27E−04 ICAM1, KRT8, VCAM1 
Liver fibrosis Fibrosis of liver 2.57E−04 COL1A1, COL1A2, HSPB1, IL2RG, LGALS3, PDGFC, PDGFRB, SERPINE1, TIMP1, Ccl2, CXCL10, LGALS3, PDGFC, PDGFRB, TIMP1 15 
Liver cholestasis Progressive familial intrahepatic cholestasis type 2 4.45E−04 CYP7A1, SLC10A1, SLC10A2 
Liver inflammation/hepatitis Inflammation of liver  CD14, CD44, CYP2E1, HLA-A, IL1RN, ITGB2, LGALS3, PTPRC, SERPINE1,SLC10A1, SPI1, TLR7 12 
Liver necrosis/cell death Cell death of liver cells 1.22E−03 CDKN1A, CXCL10,CYP2E1, CYP7A1, E2F1, GADD45B, IL1RN, ITGB2, KRT8, NR1I3, PTPRC, TIMP1, TLR7 13 
Liver cholestasis Hepatic cholestasis 1.62E−03 CD68, CDKN1A, CYP7A1, RDH16, SLC10A1, SLC10A2,Slco1a1 
Liver hypertrophy Hypertrophy of liver 1.94E−03 CYP1A2, NR1I3 
Liver steatosis Hepatic steatosis 4.79E−03 Ccl2, CCR5, CD14,CYP2E1, INSIG2,NR1I3, PDGFC, TLR2 
Liver hyperplasia/hyperproliferation Liver cancer 8.13E−03 B3GALT1, CCDC80, CD36, CD44, CD5L, CDKN1A, CIB3, CLCN5, COL12A1, COL1A1, COL1A2,COL27A1, COL4A1, COL6A3, COTL1, CREB3L2, CSF1R, CSF2RB, CX3CR1, CXCL10, CYP2A6 (includes others), CYP3A5,CYP4A11, CYP7A1, CYP7B1, DDC, DYRK3, E2F1, EHD4, ELMO1, ENDOD1, EPS8L2, GAS6, GCK, GSTA5, HAUS8, HSPA5, ICAM1,LAMA3, LGALS3,LIFR, LOXL1, MAP3K13, MICAL2, MUSTN1, MYO1F, MYOF, NAT8, NID1, NPR2, NR1I3, NT5E, NTRK2, NUPR1, PDGFC, PDIA4, PEX26, PITPNM1, PLS1, POSTN,PPP1R1B, PTPRC, ROBO1,RORC, SDCBP2, SIGLEC1, SLA, SLC10A2, SLC11A1, SLC39A4, SLC7A8, SMPD3, SOCS2, ST5, SYBU, SYK, SYVN1, TGM2, THEMIS, TIMP1, TLR2, TLR7,TOMM40L, TOR3A, TREM2, TRIM2, UCP2,UPP2, VCAM1 89 

NOTE: Genes in bold are upregulated.

Epidemiologic and experimental data have long supported CR as a preventative intervention against numerous forms of cancer. CR studies have both focused on increasing longevity and delaying tumorigenesis in aged models (28), in improving disease-free longevity (29), and limiting tumor progression (30). This study demonstrates CR serves as a potent deterrent of hepatic carcinogenic promotion and progression. The results from this study both confirm previous reports demonstrating the longevity and anticarcinogenic properties of CR and expand upon them, characterizing pathologic and transcriptome alterations induced by CR, thus providing mechanistic insights as to how CR elicits its oncogenic protection.

Etiologies of liver tumorigenesis vary greatly between developed and developing countries. In developed regions of the world, obesity and subsequent NAFLD progression, has been well described to contribute to increase the risk of HCC and other liver tumors. In this study, CR significantly reduced steatotic burden, which strongly correlated with tumor risk independent of dietary fat composition. CTL CR mice also demonstrated increased levels of circulating ketone bodies compared with CTL Sed mice (Fig. 3B). This is of interest as previous studies have demonstrated sustained ketosis is sufficient in preventing tumorigenesis (31, 32). Therefore, despite our diets not being designed to induce ketosis, we cannot rule out CR's induction of ketosis as a partial explanation for its protective effects on hepatic tumorigenesis. Collectively, these data support previous studies showing that CR can reverse steatosis in obese rats (33).

However, simple steatosis is not in and of itself pathologic. Instead, it is the subsequent damage resultant of continued neutral lipid accumulation in combination with derangements, which often accompany severe obesity, that play a pathologic role in hepatic tumorigenic risk. NASH, which is characterized by increased inflammation and cell damage greatly increases the risk in development of liver tumorigenesis (33, 34). In fact, population estimates from clinical studies suggest NASH patients are twice as likely to develop cirrhosis when compared with hepatitis C patients (35). Demonstrating the importance of limiting progression of NAFLD, CR has been shown to curb the development of NASH (35). Corroborating this feature, CTL CR mice had reduced lobular inflammation and hepatocyte ballooning. Resident hepatic macrophages known as Kupffer cells play an important role in mediating the response to liver injury (36). In addition, the innate immune response plays an important role in mediating acute inflammation and hepatocyte damage (37). However, in NAFLD and NASH, chronic inflammation results in immune cell activation and infiltration, which promotes liver injury (38). Most notably, this aberrant response sensitizes toll-like receptor signaling (39), which among other effects, increases production of inflammatory cytokines such as TNFα and IL12 (40). Histologic and IHC analysis of CR mice exhibited reduced Kupffer cells, CD3+ T cells, and neutrophils. In addition, RNA sequencing data showed that CR mice had reduced expression of genes encoding transcription factors, cytokines, and receptors involved in inflammation, liver damage, necrosis, and immune cell activation and infiltration. Thus, CR robustly prevents alterations of gene signatures, effectors, signaling networks, and subsequent disease development involved in the progression of NAFLD to liver tumorigenesis.

A major consequence of progressive NASH, and risk factor for cirrhosis, is the development of fibrosis (35). Fibrosis is most notably accomplished through Kupffer cell and inflammatory cytokine activation of hepatic stellate cells. Located in the space of Disse, hepatic stellate cells serve as the primary source of extracellular matrix proteins that contribute to fibrosis (41). TLR4 signaling initiates hepatic stellate cells to produce chemokines (CCL family), leading to recruitment of Kupffer cells, and increased TGFβ signaling (42). At the same time, TLR9 signaling initiates hepatic stellate cell collagen production, further progressing fibrosis (43). Hepatic stellate cell activation and proliferation were two major biological functions found to be downregulated in CTL CR mice through IPA of RNAseq data. In addition, TLR4 and TLR9 were both downregulated as were numerous CCL chemokines in CTL CR mice, but only TLR9 and CCL1, 2, 20, and 22 met our cut-off parameters for statistical significance. These data support previous reports investigating the effects of long-term CR on NF-κB signaling (44) and inflammatory cytokines (45). The development of fibrosis and cirrhosis are established risk factors for the development of liver cancers including HCC (46). In accordance with this, IPA identified several genes related to the development of HCC with most downregulated genes in this network being positively associated with HCC (47), whereas those that were upregulated in CTL CR mice are identified liver tumor suppressors (48). Within the hepatic tumor microenvironment, cell adhesion and migration in part driven by CD44 plays a critical role in the invasiveness of tumors (49). IPA determined CD44-mediated migration of proinflammatory molecules was universally repressed in CTL CR mice, identifying CR as a mitigator of cell damage leading to tumorigenesis. Taken together, these data highlight CR as a robust deterrent of signaling cascades and downstream biologic consequences associated with hepatic tumorigenic etiologies relevant to developed countries.

As a staple of the Mediterranean Diet, extra virgin OO elicits numerous health benefits. Previous studies have shown that the polyphenolic compounds in OO have anti-inflammatory and antioncogenic effects (50). However, these compounds are greatly enriched in extra virgin OO and variable among other oils. Analyses of these polyphenolic compounds in our diets revealed undetectable levels, suggesting poor phenolic content of our OO diet could potentially explain a lack of dietary effect on tumorigenesis.

Independent evidence has shown that exercise and the MD reduce liver tumorigenic risk, yet no studies have compared the synergistic effects of exercise and specific dietary constituents of the MD. Because mice have been documented to readily utilize running wheels (51), we used voluntary wheel running to test the effects of exercise on liver oncogenesis. Although mice initially utilized running wheels, after approximately 5 months activity levels did not exceed reported ambulatory movement (52). Lack of activity is further apparent when assessing body weight, which increased as running wheel utilization diminished. Given that phenotypic changes of Ex mice mirrored that of Sed mice, it is not surprising there were no differences in tumorigenesis. Future studies of this nature should use controlled exercise interventions to account for individual variability in activity level and to ensure adherence of treatment for the duration of the study.

Overall, these findings demonstrate CR as a robust deterrent of liver tumorigenesis and provide insight to possible mechanisms related to the pathology of liver tumorigenesis. Liver cancer is among the fastest growing cancers, especially in men, in developed countries and it remains among the world's deadliest cancers with mortality rates as high as 61% within 1 year of diagnosis. Given the limited treatment options, results here provide promise in describing CR or related dietary regiments as a therapeutic tool in tumorigenesis prevention and advance our understanding into the molecular mechanisms underlying the antioncogenic effects of CR.

No potential conflicts of interest were disclosed.

Conception and design: J.M. Ploeger, D.G. Mashek

Development of methodology: J.M. Ploeger, D.G. Mashek

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): J.M. Ploeger, J.C. Manivel, L.N. Boatner, D.G. Mashek

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): J.M. Ploeger, J.C. Manivel, D.G. Mashek

Writing, review, and/or revision of the manuscript: J.M. Ploeger, D.G. Mashek

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): J.M. Ploeger, L.N. Boatner, D.G. Mashek

Study supervision: J.M. Ploeger, D.G. Mashek

We would like to thank Dr. Dawn Lowe for graciously lending the running wheels used for our exercise intervention, Colleen Forester for her expertise and assistance with histologic and IHC staining, the University of Minnesota Genomics Core for their expertise in performing RNA sequencing; the University of Minnesota Informatics Institute (Juan E. Abrahante Lloréns) for RNAseq data processing support, Christina Dailey and Hosam Alkhatib for their dedicated efforts throughout the course of this study, the University of Minnesota Animal Care Facility staff (Meri Durand and Nicholas Burrows) for their overall excellence and assistance of animal care during the course of this study, and Mara Mashek for her assistance with initial animal injections.

This research was supported by a grant from the University of Minnesota Healthy Foods Healthy Lives Institute grant.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Wolin
KY
,
Carson
K
,
Colditz
GA
. 
Obesity and cancer
.
Oncologist
2010
;
15
:
556
65
.
2.
Calle
EE
,
Rodriguez
C
,
Walker-Thurmond
K
,
Thun
MJ
. 
Overweight, obesity, and mortality from cancer in a prospectively studied cohort of U.S. adults
.
N Engl J Med
2003
;
348
:
1625
38
.
3.
Younossi
ZM
,
Otgonsuren
M
,
Henry
L
,
Venkatesan
C
,
Mishra
A
,
Erario
M
, et al
Association of nonalcoholic fatty liver disease (NAFLD) with hepatocellular carcinoma (HCC) in the United States from 2004 to 2009
.
Hepatology
2015
;
62
:
1723
30
.
4.
Ryan
MC
,
Itsiopoulos
C
,
Thodis
T
,
Ward
G
,
Trost
N
,
Hofferberth
S
, et al
The Mediterranean diet improves hepatic steatosis and insulin sensitivity in individuals with non-alcoholic fatty liver disease
.
J Hepatol
2013
;
59
:
138
43
.
5.
Parkinson
L
,
Cicerale
S
. 
The health benefiting mechanisms of virgin olive oil phenolic compounds
.
Molecules
2016
;
21
:
pii: E1734
.
6.
Jurado-Ruiz
E
,
Varela
LM
,
Luque
A
,
Berna
G
,
Cahuana
G
,
Martinez-Force
E
, et al
An extra virgin olive oil rich diet intervention ameliorates the nonalcoholic steatohepatitis induced by a high-fat “Western-type” diet in mice
.
Mol Nutr Food Res
2017
;
61
(
Epub ahead of print; DOI: 10.1002/mnfr.201600549
).
7.
Lee
IM
,
Hsieh
CC
,
Paffenbarger
RSJ
. 
Exercise intensity and longevity in men. The Harvard Alumni Health Study
.
JAMA
1995
;
273
:
1179
84
.
8.
Meka
N
,
Katragadda
S
,
Cherian
B
,
Arora
RR
. 
Endurance exercise and resistance training in cardiovascular disease
.
Ther Adv Cardiovasc Dis
2008
;
2
:
115
21
.
9.
Willis
LH
,
Slentz
CA
,
Bateman
LA
,
Shields
AT
,
Piner
LW
,
Bales
CW
, et al
Effects of aerobic and/or resistance training on body mass and fat mass in overweight or obese adults
.
J Appl Physiol
2012
;
113
:
1831
7
.
10.
Lee
IM
,
Paffenbarger
RSJ
. 
Associations of light, moderate, and vigorous intensity physical activity with longevity. The Harvard Alumni Health Study
.
Am J Epidemiol
2000
;
151
:
293
9
.
11.
Piguet
A-C
,
Saran
U
,
Simillion
C
,
Keller
I
,
Terracciano
L
,
Reeves
HL
, et al
Regular exercise decreases liver tumors development in hepatocyte-specific PTEN-deficient mice independently of steatosis
.
J. Hepatol
2015
;
62
:
1296
303
.
12.
Belahsen
R
,
Rguibi
M
. 
Population health and mediterranean diet in southern mediterranean countries
.
Public Health Nutr
2006
;
9
:
1130
5
.
13.
Rocha
NS
,
Barbisan
LF
,
de Oliveira
MLC
,
de Camargo
JLV
. 
Effects of fasting and intermittent fasting on rat hepatocarcinogenesis induced by diethylnitrosamine
.
Teratog Carcinog Mutagen
2002
;
22
:
129
38
.
14.
Dunlap
SM
,
Chiao
LJ
,
Nogueira
L
,
Usary
J
,
Perou
CM
,
Varticovski
L
, et al
Dietary energy balance modulates epithelial-to-mesenchymal transition and tumor progression in murine claudin-low and basal-like mammary tumor models
.
Cancer Prev Res
2012
;
5
:
930
42
.
15.
Galet
C
,
Gray
A
,
Said
JW
,
Castor
B
,
Wan
J
,
Beltran
PJ
, et al
Effects of calorie restriction and IGF-1 receptor blockade on the progression of 22Rv1 prostate cancer xenografts
.
Int J Mol Sci
2013
;
14
:
13782
95
.
16.
Jiang
Y-S
,
Wang
F-R
. 
Caloric restriction reduces edema and prolongs survival in a mouse glioma model
.
J Neurooncol
2013
;
114
:
25
32
.
17.
Harvey
AE
,
Lashinger
LM
,
Otto
G
,
Nunez
NP
,
Hursting
SD
. 
Decreased systemic IGF-1 in response to calorie restriction modulates murine tumor cell growth, nuclear factor-kappaB activation, and inflammation-related gene expression
.
Mol Carcinog
2013
;
52
:
997
1006
.
18.
Lanza-Jacoby
S
,
Yan
G
,
Radice
G
,
LePhong
C
,
Baliff
J
,
Hess
R
. 
Calorie restriction delays the progression of lesions to pancreatic cancer in the LSL-KrasG12D; Pdx-1/Cre mouse model of pancreatic cancer
.
Exp Biol Med
2013
;
238
:
787
97
.
19.
Kim
KE
,
Jung
Y
,
Min
S
,
Nam
M
,
Heo
RW
,
Jeon
BT
, et al
Caloric restriction of db/db mice reverts hepatic steatosis and body weight with divergent hepatic metabolism
.
Sci Rep
2016
;
6
:
30111
.
20.
McCay
CM
,
Crowell
MF
,
Maynard
LA
. 
The effect of retarded growth upon the length of life span and upon the ultimate body size. 1935
.
Nutrition
1989
;
5
:
155
71
.
21.
Ong
KT
,
Mashek
MT
,
Bu
SY
,
Greenberg
AS
,
Mashek
DG
. 
Adipose triglyceride lipase is a major hepatic lipase that regulates triacylglycerol turnover and fatty acid signaling and partitioning
.
Hepatology
2016
;
53
:
116
26
.
22.
Vandbakk
K
,
Welde
B
,
Kruken
AH
,
Baumgart
J
,
Ettema
G
,
Karlsen
T
, et al
Effects of upper-body sprint-interval training on strength and endurance capacities in female cross-country skiers
.
PLoS One
2017
;
12
:
e0172706
.
23.
Nakagawa
H
. 
Recent advances in mouse models of obesity- and nonalcoholic steatohepatitis-associated hepatocarcinogenesis
.
World J Hepatol
2015
;
7
:
2110
8
.
24.
Gong
Z
,
Tas
E
,
Yakar
S
,
Muzumdar
R
. 
Hepatic lipid metabolism and non-alcoholic fatty liver disease in aging
.
Mol Cell Endocrinol
2016 Dec 23
. [Epub ahead of print].
25.
Guan
L
,
Zhang
L
,
Gong
Z
,
Hou
X
,
Xu
Y
,
Feng
X
, et al
FoxO3 inactivation promotes human cholangiocarcinoma tumorigenesis and chemoresistance through Keap1-Nrf2 signaling
.
Hepatology
2016
;
63
:
1914
27
.
26.
McCommis
KS
,
Hodges
WT
,
Brunt
EM
,
Nalbantoglu
I
,
McDonald
WG
,
Holley
C
, et al
Targeting the mitochondrial pyruvate carrier attenuates fibrosis in a mouse model of nonalcoholic steatohepatitis
.
Hepatology
2016
;
65
:
1543
56
.
27.
Wang
Q
,
Yu
W-N
,
Chen
X
,
Peng
X
,
Jeon
S-M
,
Birnbaum
MJ
, et al
Spontaneous hepatocellular carcinoma after the combined deletion of Akt isoforms
.
Cancer Cell
2016
;
29
:
523
35
.
28.
Spindler
SR
. 
Rapid and reversible induction of the longevity, anticancer and genomic effects of caloric restriction
.
Mech Ageing Dev
2005
;
126
:
960
6
.
29.
Colman
RJ
,
Anderson
RM
,
Johnson
SC
,
Kastman
EK
,
Kosmatka
KJ
,
Beasley
TM
, et al
Caloric restriction delays disease onset and mortality in rhesus monkeys
.
Science
2009
;
325
:
201
4
.
30.
Lv
M
,
Zhu
X
,
Wang
H
,
Wang
F
,
Guan
W
. 
Roles of caloric restriction, ketogenic diet and intermittent fasting during initiation, progression and metastasis of cancer in animal models: a systematic review and meta-analysis
.
PLoS One
2014
;
9
:
e115147
.
31.
Nebeling
LC
,
Miraldi
F
,
Shurin
SB
,
Lerner
E
. 
Effects of a ketogenic diet on tumor metabolism and nutritional status in pediatric oncology patients: two case reports
.
J Am Coll Nutr
1995
;
14
:
202
8
.
32.
Stemmer
K
,
Zani
F
,
Habegger
KM
,
Neff
C
,
Kotzbeck
P
,
Bauer
M
, et al
FGF21 is not required for glucose homeostasis, ketosis or tumour suppression associated with ketogenic diets in mice
.
Diabetologia
2015
;
58
:
2414
23
.
33.
Rector
RS
,
Uptergrove
GM
,
Morris
EM
,
Borengasser
SJ
,
Laughlin
MH
,
Booth
FW
, et al
Daily exercise vs. caloric restriction for prevention of nonalcoholic fatty liver disease in the OLETF rat model
.
Am J Physiol Gastrointest Liver Physiol
2011
;
300
:
G874
83
.
34.
Nakagawa
S
,
Wei
L
,
Song
WM
,
Higashi
T
,
Ghoshal
S
,
Kim
RS
, et al
Molecular liver cancer prevention in cirrhosis by organ transcriptome analysis and lysophosphatidic acid pathway inhibition
.
Cancer Cell
2016
;
30
:
879
90
.
35.
Holmberg
SD
,
Spradling
PR
,
Moorman
AC
,
Denniston
MM
. 
Hepatitis C in the United States
.
N Engl J Med
2013
;
368
:
1859
61
.
36.
Han
J
,
Ulevitch
RJ
. 
Limiting inflammatory responses during activation of innate immunity
.
Nat Immunol
2005
;
6
:
1198
205
.
37.
Meyer-Bahlburg
A
,
Khim
S
,
Rawlings
DJ
. 
B cell intrinsic TLR signals amplify but are not required for humoral immunity
.
J Exp Med
2007
;
204
:
3095
101
.
38.
Borlak
J
,
Langer
F
,
Spanel
R
,
Schondorfer
G
,
Dittrich
C
. 
Immune-mediated liver injury of the cancer therapeutic antibody catumaxomab targeting EpCAM, CD3 and Fcgamma receptors
.
Oncotarget
2016
;
7
:
28059
74
.
39.
Dalpke
AH
,
Lehner
MD
,
Hartung
T
,
Heeg
K
. 
Differential effects of CpG-DNA in Toll-like receptor-2/-4/-9 tolerance and cross-tolerance
.
Immunology
2005
;
116
:
203
12
.
40.
Alexopoulou
L
,
Holt
AC
,
Medzhitov
R
,
Flavell
RA
. 
Recognition of double-stranded RNA and activation of NF-kappaB by Toll-like receptor 3
.
Nature
2001
;
413
:
732
8
.
41.
Seki
E
,
De Minicis
S
,
Osterreicher
CH
,
Kluwe
J
,
Osawa
Y
,
Brenner
DA
, et al
TLR4 enhances TGF-beta signaling and hepatic fibrosis
.
Nat Med
2007
;
13
:
1324
32
.
42.
Seki
E
,
Brenner
DA
. 
Toll-like receptors and adaptor molecules in liver disease: update
.
Hepatology
2008
;
48
:
322
35
.
43.
Watanabe
A
,
Hashmi
A
,
Gomes
DA
,
Town
T
,
Badou
A
,
Flavell
RA
, et al
Apoptotic hepatocyte DNA inhibits hepatic stellate cell chemotaxis via toll-like receptor 9
.
Hepatology
2007
;
46
:
1509
18
.
44.
Xu
X
,
Ning
Y-C
,
Wang
W
,
Liu
J
,
Bai
X
,
Sun
X
, et al
Anti-inflamm-aging effects of long-term caloric restriction via overexpression of SIGIRR to inhibit NF-kappaB signaling pathway
.
Cell Physiol Biochem
2015
;
37
:
1257
70
.
45.
Lee
IS
,
Shin
G
,
Choue
R
. 
A 12-week regimen of caloric restriction improves levels of adipokines and pro-inflammatory cytokines in Korean women with BMIs greater than 23 kg/m2
.
Inflamm Res
2010
;
59
:
399
405
.
46.
Fuchs
BC
,
Hoshida
Y
,
Fujii
T
,
Wei
L
,
Yamada
S
,
Lauwers
GY
, et al
Epidermal growth factor receptor inhibition attenuates liver fibrosis and development of hepatocellular carcinoma
.
Hepatology
2014
;
59
:
1577
90
.
47.
Zhang
Y
,
Xu
N
,
Xu
J
,
Kong
B
,
Copple
B
,
Guo
GL
, et al
E2F1 is a novel fibrogenic gene that regulates cholestatic liver fibrosis through the Egr-1/SHP/EID1 network
.
Hepatology
2014
;
60
:
919
30
.
48.
Okamura
Y
,
Nomoto
S
,
Kanda
M
,
Li
Q
,
Nishikawa
Y
,
Sugimoto
H
, et al
Leukemia inhibitory factor receptor (LIFR) is detected as a novel suppressor gene of hepatocellular carcinoma using double-combination array
.
Cancer Lett
2010
;
289
:
170
7
.
49.
Yang
ZF
,
Ho
DW
,
Ng
MN
,
Lau
CK
,
Yu
WC
,
Ngai
P
, et al
Significance of CD90+ cancer stem cells in human liver cancer
.
Cancer Cell
2008
;
13
:
153
66
.
50.
Lama
A
,
Pirozzi
C
,
Mollica
MP
,
Trinchese
G
,
Di Guida
F
,
Cavaliere
G
, et al
Polyphenol-rich virgin olive oil reduces insulin resistance and liver inflammation and improves mitochondrial dysfunction in high-fat diet fed rats
.
Mol Nutr Food Res
2017
;
61
(
Epub ahead of print; DOI: 10.1002/mnfr.201600418
).
51.
Meijer
JH
,
Robbers
Y
. 
Wheel running in the wild
.
Proceedings Biol Sci
2014
;
281
:
20140210
.
52.
Morton
GJ
,
Kaiyala
KJ
,
Fisher
JD
,
Ogimoto
K
,
Schwartz
MW
,
Wisse
BE
. 
Identification of a physiological role for leptin in the regulation of ambulatory activity and wheel running in mice
.
Am J Physiol Endocrinol Metab
2011
;
300
:
E392
401
.