Preventive effects of combinative natural foods produced by elite crop varieties rich in anticancer effects on N‐nitrosodiethylamine‐induced hepatocellular carcinoma in rats

Abstract The World Cancer Research Fund International has released 32 anticancer effects (ACEs) that targeted every stage of cancer processes. Thus, we designed two formulas of natural food combination Diet I and Diet II, mainly produced by elite crop varieties rich in ACEs with different mixture ratios, and evaluated their cancer preventive effects on N‐nitrosodiethylamine (NDEA)‐induced hepatocarcinogenesis. After 20 weeks of dietary intervention, Diet I and Diet II reduced incidence, size, and number of hepatic nodules (p < 0.01) and prevented hepatic tumor formation in NDEA‐induced hepatocarcinogenesis rats. Low‐grade hepatic dysplasia incidence was 20% for Diet II and 40% for Diet I, and apparent hepatocellular carcinomas (HCC) rates were both 0, while 90% HCC in control diet treatment group (p < 0.01). Diet I and Diet II ameliorated abnormal liver function enzymes, reduced serum alpha fetal protein, tumor‐specific growth factor, dickkopf‐related protein 1, tumor necrosis factor‐alpha and interleukin‐6 levels, regulated hepatic phase I and II xenobiotic‐metabolizing enzymes, enhanced antioxidant capacity, suppressed NDEA‐initiated oxidative DNA damage, and induced apoptosis coupled to down‐regulation of proinflammatory, invasion, and angiogenesis markers. Daily intake of combination diet produced from ACEs‐rich elite crop varieties can effectively prevent or delay occurrence and development of NDEA‐induced hepatocarcinogenesis in rats.


| INTRODUC TI ON
Cancer prevention has been drawing more and more attention in modern society. At the current stage of research, products like synthetic chemical compounds, single natural plant chemicals, and single natural plants are frequently used for cancer prevention.
However, most of these chemopreventive agents only target cancer at certain stages of its occurrence and development to suppress tumor development (Kweon, Adhami, Lee, & Mukhtar, 2006;Liu et al., 2017;Stagos et al., 2012;Zhou et al., 2016). This often renders their preventive effects unsatisfactory. For example, for the breast cancer patients who were positive for estrogen receptor, the postoperative recurrence rate decreased by 47% among those who took tamoxifen continuously for 5 years, but the likelihood of those patients developing early endometrial cancer seemed to increase (Buzdar et al., 2006), suggesting that tamoxifen has a serious side effect. According to a report published by Chen, Wallig, and Jeffery (2016), feeding 10% more broccoli (Brassica oleracea L. var. Green Magic) significantly decreased the levels of hepatic triacylglycerols and tumor necrosis factor, and the rates of nonalcoholic fatty liver disease induced by diethylnitrosamine and diet rich in refined carbohydrate in B6C3F1 mice, but carcinogenesis was not blocked. Yoxall et al. (2005) showed that sulforaphane (SFN) at a typical dietary dose stimulated NAD (P) H quinone dehydrogenase 1 in a dosedependent fashion but did not influence glutathione-S-transferase (GST), epoxide hydrolase, or uridine diphosphate-glucuronosyl transferase activities in rat livers exposed to SFN in their drinking water for 10 days at equivalent daily doses of 3 and 12 mg/kg.

The 2007 Second Expert Report of the World Cancer Research
Fund International (WCRF) has shown that the occurrence and development of cancer are multi-step processes, and each step can be interrupted by a number of anticancer effects (ACEs) according to Figure 2.5, Chapter 2, such as organic sulfur compounds, epigallocatechin gallate (EGCG), vitamin A, and resveratrol, in total of 32 ACEs (WCRF, 2007) (Supporting Information   Table S1).
The concentration of ACEs is crucial to the cancer prevention (Misaka, Miyazaki, Fukushima, Yamada, & Kimura, 2013). In many cases, the effects of chemopreventive agents in cultured cells or tissues are only achievable at supraphysiological concentrations; such concentrations might not be reached when the phytochemicals are administered as part of an organism's diet (Tan, Shi, Tang, Han, & Spivack, 2010). For most people, eating the right foods and drinks is more likely to prevent cancer than dietary supplements, according to the summary of the Third Expert Report of World Cancer Research Fund/American institute for Cancer Research (2018). In our previous work, we have bred and excavated elite crop varieties with high contents of ACEs (Cheng, Xu, Yang, Chen, & Zheng, 2016;Cheng et al., 2014;Zheng, 2004Zheng, , 2014Zheng et al., 2012)  The contents of these substances were determined by fresh sample.
In this study, we designed two formulas of natural food com- bination Diet I and Diet II, which were mainly produced by ACEsrich elite crop varieties with different mixture ratios (Supporting   Information Table S2) and evaluated the cancer prevention of the two formulas in N-nitrosodiethylamine (NEDA)-induced hepatocellular carcinoma (HCC) rats. The modulatory effects of Diet I and Diet II on xenobiotic-metabolizing enzymes, antioxidant activity, and markers of cell proliferation, inflammation, apoptosis, invasion, and angiogenesis during NDEA-induced rat hepatocarcinogenesis are also explored.

| Animals, diets, and water
The study was performed in accordance with the Chinese national guidelines for the care of laboratory animals and was approved by  Table S2), which were provided by the Agricultural Product Quality Institute, Fujian Agriculture and Forestry University. The others were purchased from the common market. Diet I is a food combination of nine pure natural foods of plant origin from the elite crop varieties rich in ACEs. 100 g Diet I contains 12 g black rice, 28 g job's-tears, 10 g sweet potato, 5 g broccoli, 2 g carrot, 1 g alfalfa, 5 g mulberry, and 1.0 g reishi mushroom and is supplemented with 7 g corn, 24 g bean cake, 1 g dried yeast, 2 g bran, 1.0 g oleum morrhuae and then blended to mixture. The mixture was then grain shaped, dried at 60°C and sterilized by Co 60 . Diet II is composed of primarily by foods of nine pure natural foods of plant origin from the elite crop varieties rich in ACEs and subsidiary by three light meat. 100 g Diet II contains 11.5 g black rice, 26.9 g job's-tears, 7.7 g sweet potato, 4.8 g broccoli, 1.9 g carrot, 0.5 g alfalfa, 4.8 g mulberry, and 1.4 g reishi mushroom, 3.8 g trepang, 1.0 g abalone, 1.4 g clam and is supplemented with 5.3 g corn, 16.3 g bean cake, 1.4 g dried yeast, 1.9 g bran, 1.0 oleum morrhuae, 1.9 g grifola fron-

| Experimental design
After 1-week adaptive feeding on control diet, forty rats were randomly divided into four groups (n = 10/group): normal control group Diet II+NDEA group were with 25 mg/kg NDEA (twice a week) as described earlier (Liao, Liu, Xu, & Zheng, 2018). Rats in the ConD group received intraperitoneal injection of the same volume of saline. Body weights, feed and tea, or water intake were measured weekly. After 20 weeks, the rats were anesthetized by intraperitoneal injection of pentobarbital sodium (40 mg/kg) for collection of blood samples. Immediately after that, animals were sacrificed by bleeding from the abdominal aorta, and their organs were collected.
The size and number of liver nodules were measured as described earlier . Liver tissue was collected for biochemical, histopathological, and ultrastructural analyses.

| Histological analysis of the liver tissue
Small blocks of liver from median lobe were fixed in 10% neutral buffered formalin, embedded in paraffin, sectioned (4 μm), and stained with hematoxylin and eosin (H&E) for histopathological examination. The stained slides were analyzed by assessing the morphological changes under the OPTIPHOT-2 light microscope by an experienced investigator that was unaware of the experimental conditions. Five microscopy-stained slides per animal were examined.
The extent of cancerization was assessed according to Edmondson classification method (Edmondson & Steiner, 1954

| Biochemical assays
The Microsomal fractions were prepared from liver tissues as described previously (Velayutham et al., 2007). The concentration of microsomal protein was determined using the BCA kit. The contents of cytochrome b5 (Cyt b5) and cytochrome P450 (Cyt P450) were assayed using the method published by Omura and Sato (1964). The activity of NADPH-cytochrome b5 reductase (Cyt b5R) was assayed using the methods published by Philips and Langdon (1962). Cytochrome C (Cyt c) was used as the substrate to determine NADH-cytochrome P450 reductase (Cyt P450R) and DT-diaphorase (DTD) activity using the methods reported by Cummings, Parker, and Lash (2000) and Smitskamp-Wilms, Giaccone, Pinedo, vander, and Peters (1995), respectively. Aryl hydrocarbon hydroxylase (AHH) and aniline hydroxylase CYP2E1 (ANH) activity were determined using double-beam ultraviolet spectrophotometry (Nebert & Gelboin, 1968). The activities of GST and UDP-glucuronosyltransferase (UGT) in the liver homogenates were detected by specific ELISA kits (Shanghai Jiang Lai Biotechnology Co., Shanghai, China). The epoxide hydrolyzyme (EPT) activity in liver microsome was determined using the Fabian method (Fabian et al., 2016).
The malondialdehyde (MDA) content and the activities of superoxide dismutase (SOD), catalase (CAT), and glutathione peroxidase (GSH-Px) were assayed within 12 hr using standard commercially available kits according to the manufacturer's instructions (Nanjing Jiancheng Bioengineering Engineering Institute, Nanjing, China).

| Immunohistochemical analysis
Paraffin section (4 μm) was dewaxed and rehydrated through a gradual decrease in ethanol concentration. The slides were incubated in sodium citrate buffer for two cycles of 5 min at 37°C for antigen retrieval. After washing with phosphate buffer solution (PBS), the slides were then treated with 3% hydrogen peroxide to remove any endogenous peroxidases. Then the sample was washed with PBS for three cycles of 5 min and blocked with normal goat serum in a humidified chamber for several hours at 4°C. The sections were then incubated overnight at 4°C in a humidified chamber using the appropriate primary antibodies: 8-OH-dG, NF-κB, COX-2, TNFα, p53, Bcl-2, PCNA, and VEGF with recommended dilution. The slides were washed with TBS and then incubated with HRP-labeled sheep antirabbit secondary antibody at room temperature for 1 hr followed by streptavidin-biotin-peroxidase at room temperature for 30 min.
The slides were washed with PBS, and the immunoprecipitation was visualized by treating with 3, 3′-diaminobenzidine for color development for 25 min. Then slides were counterstained with hematoxylin, and the brown color signifying the presence of antigen bound to antibody was detected by light microscopy. For the negative control, tris-buffered saline (TBS) was used instead of a primary antibody. From ten randomly selected sections of each slide, 500 cells were counted. The percentage of positive cells for each group was calculated using the Image-Pro Plus 6.0 image analysis system.

| Quantitative real-time PCR assay
Total RNA was isolated from liver tissue using TRIzol (Invitrogen

| Western blot analysis
The protein was extracted from liver tissue with RIPA buffer containing protease inhibitor cocktail. The protein extraction was separated using centrifugation (15,000 × g; 15 min; 4°C), and the supernatant was collected and then quantified using the BCA kit. Equal amounts of proteins were separated through 10%-12% SDS-polyacrylamide electrophoresis gels and transferred to polyvinylidene fluoride membranes. After blocked with 5% skim milk, the membranes were incubated with the appropriate primary antibodies respectively, and β-actin (CUSABIO Biotech Co., Ltd., Wuhan, China) with recommended dilution at 4°C overnight, and then incubated with goat antirabbit IgG horse radish peroxidase (HRP) secondary antibodies at room temperature for 1 hr. The films were developed using an ECL Plus chemiluminescence reagent kit (Millipore, MA, USA) and visualized using ChemiDocXRS (Bio-Rad Laboratory, CA, USA).
Densitometries were analyzed using Quantity One software and normalized to β-actin.

| Statistical analysis
Statistical analysis was performed using SPSS19.0 software. All results are expressed as mean ± standard deviation (SD) and were analyzed using a one-way, 2-or 3-factor ANOVA, followed by LSD test or Tukey's test when the differences were indicated. A p value of <0.05 was here considered significant in this study.

| Body weight, relative liver weight, feed and tea/water intake were not changed by long-term Diet I and Diet II consumption
In NDEA administered rats, the food intake, water consumption, and body weight gain were significantly (p < 0.05) decreased (Supporting Information Table S4) and the relative liver weight was significantly (p < 0.01) increased in NDEA+ConD group rats relative to ConD group rats (Supporting Information Table S4). All the changes induced by NDEA intoxication were significantly (p < 0.01 or p < 0.05) reduced except body weight when rats treated with Diet I and Diet II (Supporting Information Table S4), and there were no significant changes in hematological indicators (Supporting Information Table   S5).

| Diet I and Diet II exerted a powerful protective or delaying effect on hepatic tumorigenesis
When treated with NDEA (ConD+NDEA group), 100% rats developed nodules in the liver (for the two rats died of severe hepatic tumor pathogenesis at week 16 and 19 respectively, and the   remaining eight rats, the nodule formation was observed. While administration of Diet I (Diet I+NDEA group) and Diet II (Diet II+NDEA group) was found to be associated with marked decreases in the number and multiplicity of the nodules relative to ConD +NDEA group rats (p < 0.01, Table 2). The incidence of nodule growth was reduced to 30% in the Diet I+NDEA group and 50% in the Diet II+NDEA group, respectively (Table 2). No hepatic nodules were observed in the ConD group animals.
The results of H&E staining ( Figure 1B and Table 2)

| Diet I and Diet II ameliorated hepatocyte damage and decreased serum tumor markers
Detection of blood biochemical indicators showed that the levels of liver function markers, such as AST, ALT, GGT, TBIL, ALP, and albumin in serum of ConD+NDEA group rats were significantly higher than in the ConD group (Table 3), indicating that the stimulation of NDEA and its metabolites in vivo on hepatocytes caused severe injury and liver dysfunction in the ConD+NDEA group. In the presence of Diet I and Diet II administration, the activities of the above serum marker enzymes in the serum of Diet I+NDEA and Diet II+NDEA group rats were significantly lower than in the ConD+NDEA group (Table 3), which was similar to the normal control group. The levels of AFP, TSGF, CEA, and DKK1 were much higher in the ConD+NDEA group than in the normal control group (p < 0.01); and Diet I and Diet II significantly inhibited the elevation of the serum tumor markers, AFP, TSGF, and DKK1 (p < 0.01, Table 3).

| Diet I and Diet II suppressed the activity of phase I enzymes and enhanced that of phase II enzymes
Carcinogens are primarily activated by phase I enzymes. The body can be protected from carcinogens by induction of phase II enzymes that lead to detoxification and accelerated excretion of carcinogens (Chakraborty et al., 2007;Na & Surh, 2008). As shown in

| Diet I and Diet II reduced NDEA-induced oxidative stress
Oxidative stress is one of the major instigators of the pathogenesis of environmental cancer (Grace et al., 2016). NDEA is converted to reactive oxygen species in liver cells through phase I enzyme metabolism, causing oxidation and inducing oxidative damage to DNA, ultimately leading to pathological changes and initiation of hepatogenesis (Marra et al., 2011). Many chemical carcinogens are associated with free radicals or reactive oxygen species, such as O 2− , H 2 O 2 , and ROOH, and scavenging this free radical or reactive oxygen is mainly dependent on SOD, CAT, GSH-Px, and GSH (Bishayee et al., 2011). As shown in Table 5, single intraperitoneal administration of NDEA significantly elevated serum and liver MDA content The 8-OH-dG, as a biomarker of oxidative DNA damage, was detected using immunohistochemistry to evaluate oxidative DNA damage in the liver. As shown in Figure 2a, the production of 8-OH-dG increased significantly in the liver tissue of ConD+NDEA rats (p < 0.01), indicating severe DNA damage in hepatocytes. However, the production of 8-OH-dG in liver tissue of rats fed with Diet I or Diet II was significantly lower than the level of production in the ConD+NDEA group (p < 0.01). These results showed that Diet I and Diet II can reduce NDEA-induced oxidative stress and protect the body from oxidative DNA damage.

| Diet I and Diet II can reduce NDEA-induced inflammatory response
NDEA may cause severe chronic inflammatory damage to the liver, thus triggering the development and progression of HCC (Raghunandhakumar et al., 2013). Inflammatory factors (TNFα, IL-6, etc.) play important roles in the development of inflammation and tumorigenesis (Sivaramakrishnan & Niranjali, 2009;Sui et al., 2014;Wang et al., 2017;Wojcik et al., 2012). The interaction between TNFα, IL-6, and NF-κB initiates a vicious cycle in the cytokine network, and TNFα and IL-6 may activate NF-κB, further amplifying the inflammatory response. Activated NF-κB may further induce or activate COX-2 (Wang et al., 2017). The mRNA and protein expression of NF-κB, COX-2, and TNFα are presented in Figures 2,3. As shown in Figure 3a, the levels of TNFα and IL-6 in the serum of the model group were significantly higher than in the normal control group (p < 0.01), as well as the number of white blood cells, neutrophils, and lymphocytes (p < 0.01 or p < 0.05).
The mRNA and protein expression of NF-κB, COX-2, and TNFα in the liver of ConD+NDEA group were also significantly increased (p < 0.01 or p < 0.05). These results suggest that NDEA-induced

| Diet I and Diet II induced cell apoptosis and inhibited tumor cell proliferation, angiogenesis, and invasion
NDEA causes genomic damage in exposed cells. This can trigger the damaged cells to proliferate, leading to the formation of cancerous cells, which showed increased cell proliferation, angiogenesis, and invasion potential (Hanahan & Weinberg, 2011;Yu et al., 2012).
Apoptosis evasion in NDEA-induced hepatocarcinogenesis is associated with imbalance in proapoptotic and antiapoptotic proteins combined with regulation of caspases (Gupta, Bhatia, Bansal, & Koul, 2016;Subramanian & Arul, 2013). As shown in Figures 4 and 5, Diet I and Diet II significantly reduced the expression of the antiapoptotic gene Bcl-2 at both mRNA and protein levels in rat liver tissues relative to the ConD+NDEA group (p < 0.01 or p < 0.05), and they increased the expression of Bax, p53, Caspase-3, and Caspase-8 at both the mRNA and protein levels (p < 0.01 or p < 0.05). These results suggest that Diet I and Diet II could induce apoptosis of NDEAinduced HCC at the initial stages.
Apoptosis may promote cancer cell proliferation, angiogenesis, invasion, and metastasis [41] . As shown in Figures 4 and 5, the results of immunohistochemistry, qRT-PCR, and Western blot showed that cell proliferation markers PCNA, angiogenic factor VEGF, and matrix metalloproteinases (MMP-2, MMP-9) in the liver tissue of ConD+NDEA group were all significantly higher at both the mRNA and protein levels than in the ConD group (p < 0.01 or p < 0.05). The expression of PCNA, VEGF, MMP-2, and MMP-9 in Diet I+NDEA and Diet II+NDEA groups was significantly lower than in the NDEA group (p < 0.01 or p < 0.05), indicating that Diet I and Diet II could inhibit cancer proliferation and affect tumor neovascularization, invasion, and metastasis.

| D ISCUSS I ON
NDEA is a potent hepatocarcinogenic nitrosamine present in a variety of foods and also a commonly used chemical carcinogen to induce hepatocellular carcinoma in animal models (Ajiboye et al., 2013).
Previous studies have shown that the methods of NDEA-induced HCC model through administration of NDEA by free drinking water, gavage, or intraperitoneal injection can simulate the real situation of HCC in human liver and represents an ideal in vivo model for evaluating prevention agents of HCC (Bishayee et al., 2011;Gupta et al., 2016). In this study, administration of the low-dose NDEA (25 mg/ kg, 2 times per week) by intraperitoneal injection for 12 weeks had succeeded in inducing liver cancer; the tumor incidence and the apparent HCC incidence were 100%, 90%, respectively, and only a low mortality rate (20%) for 20 weeks in the ConD+NDEA group.   results showed that treatment of the Diet I and Diet II had a significant effect on reducing the incidence, size, and number of hepatic nodules, and the incidence of hepatic adenoma or HCC formation in the livers of NDEA treatment rats, the low-grade hepatic dysplasia incidence was 20% for Diet II and 40% for Diet I, the apparent HCC rates were both 0, while the apparent HCC rate of the control diet treatment rats was 90% (p < 0.01). The chemoprevention effects of Diet I and Diet II were superior to those of some single phytochemical or single food studied by previous studies (Bhatia, Gupta, Singh, & Koul, 2015;Gupta et al., 2016;Katayama et al., 2003).
This demonstrates that xenobiotic-metabolizing enzymes, especially phase II enzyme are probably crucial to cancer prevention.
The metabolic activation of NDEA by cytochrome P450 enzymes produces active ethyl radical metabolites that are mainly responsible for initiation of carcinogenesis in the liver (Sadeeshkumar et al., 2017). Subsequently, reactive product of NDEA can be detoxified by phase II enzymes including GST and quinone reductase, etc. (Bishayee et al., 2011;Sindhu, Firdous, Ramnath, & Kuttan, 2013 II act as potential dual-acting agents by suppressing phase I and enhancing phase II enzyme activity, thereby promoting detoxification and excretion. Thus, we believed that the stage "carcinogens and other environmental exposures" in Supporting Information Table S1 is the most important step responsible for these effects. During this stage, the metabolic activation of NDEA is inhibited, and its reactive products are detoxified and excreted. In conclusion, the combinative natural food formulas Diet I and Diet II, which were produced by elite crop varieties rich in ACEs according to WCRF, both exhibited cancer preventive or delaying effects on NDEA-induced hepatocarcinogenesis in rats, with a 0 incidence of apparent HCC (the low-grade hepatic dysplasia incidence was 20% for Diet II and 40% for Diet I) relative to the 90% incidence of apparent HCC in the model group (p < 0.01). Diet I and Diet II also can ameliorate the abnormal changes in liver function enzymes and reduce the increase in tumor markers. The results of highly significant cancer prevention or delay of Diet I and Diet II indicated excellent prospects of daily dietary for cancer prevention and the importance of the discovery and creation of ACEs-rich crop varieties.