Germinated brown rice combined with Lactobacillus acidophilus and Bifidobacterium animalis subsp. lactis inhibits colorectal carcinogenesis in rats

Abstract Colorectal cancer is a common cancer strongly associated with diet. Certain probiotics and prebiotics possess an inhibitory activity against colorectal cancer, while synbiotics may be more effective in preventing this cancer than either prebiotics or probiotics alone. Germinated brown rice (GBR) is considered as a candidate prebiotics with anticancer potential. However, the effect of GBR combined with probiotics on colorectal cancer is not clear. The present study investigated the preventive effect of combination of GBR and Lactobacillus acidophilus, Bifidobacterium animalis subsp. lactis, or both on colorectal carcinogenesis and the possible mechanism in rats treated with 1,2‐dimethylhydrazine (DMH) and dextran sulfate sodium (DSS). DMH/DSS treatment induced preneoplastic aberrant crypt foci (ACF) and mucin‐depleted foci (MDF), reduced superoxide dismutase (SOD) activity, increased anti‐apoptotic Bcl‐2 expression, and decreased the expression of pro‐apoptotic p53, Bax, and caspase‐3 in the colon. Germinated brown rice alone or combined with probiotics inhibited the formation of MDF in the middle colon, enhanced the colonic expression of p53 and Bax, and increased the ratio of Bax/Bcl‐2. Combined treatment of GBR and probiotics inhibited the formation of ACF‐producing sialomucin (SIM‐ACF) and recovered the activity of SOD in the colon. Combination of GBR and L. acidophilus further increased caspase‐3 expression and decreased Bcl‐2 expression. These findings suggest that GBR combined with L. acidophilus and/or B. animalis subsp. lactis may inhibit colorectal carcinogenesis by enhancing antioxidative capacity and inducing apoptosis. This synbiotics may be a potential functional food or chemopreventive agent for controlling colorectal cancer.

are defined as substrates that are selectively utilized by host microorganisms conferring a health benefit (Gibson et al., 2017).
Fructooligosaccharide is one of the most reported prebiotics and is generally recognized as safe. Other prebiotics include galactooligosaccharide, xylooligosaccharide, resistant starch, and so on (Raman et al., 2013). However, only limited studies have investigated the prebiotic potential of complex grains and their products.
Synbiotics mean the combination of probiotics and prebiotics.
Emerging data suggest synbiotics as a more effective strategy in the prevention of colorectal cancer than either probiotics or prebiotics alone (Chong, 2014). Common synbiotic combinations for the prevention of colorectal cancer include "L. acidophilus plus resistant starch" (Le Leu, Hu, Brown, Woodman, & Young, 2009;Le Leu et al., 2005), "B. lactis plus oligofructose/inulin" (Dias et al., 2010), and "Lactobacillus rhamnosus GG (LGG) and B. lactis Bb12 plus inulin" (Verma & Shukla, 2014). These synbiotics inhibit colorectal carcinogenesis by attenuating oxidative stress, reducing cell proliferation, or inducing apoptosis. However, certain food ingredients with assumed prebiotic potential seem not to be good fermentation substrates for the growth of certain probiotics. For example, some probiotic Lactobacillus strains without starch-degrading activity were not apparently sustained by resistant starch, so they might show a low pro-apoptotic activity (Le Leu et al., 2005). It suggests the need for development of a novel prebiotic component for the better growth of certain probiotics and their synergistically chemopreventive effect against colorectal cancer.
Whole grains and germinated grains contain soluble dietary fiber, non digestible oligosaccharides, and resistant starch and thus have been suggested to fulfill the prebiotic concept and to be candidate prebiotics (Bindels, Delzenne, Cani, & Walter, 2015;Broekaert et al., 2011;Hubner & Arendt, 2013) . In recent years, the health-promoting activity of germinated grains is gaining high interest as a functional food applied to reduce the risk of some chronic diseases, including colorectal cancer (Nelson, Stojanovska, Vasiljevic, & Mathai, 2013).
The preliminary study in our laboratory demonstrated that germinated brown rice (GBR) inhibited the development of preneoplastic lesions of colorectal cancer in a carcinogen-induced animal model (A.-C. Kuo, C.-K. Shih, unpublished data). However, the detailed mechanism of action remained to be elucidated. The present study was designed to investigate the preventive effect of combination of GBR with L. acidophilus and/or B. animalis subsp. lactis, two wellknown probiotics with anticancer activity, on colorectal carcinogenesis and the associated mechanism.

| GBR preparation
Germinated brown rice used in the present study was obtained from Asia Rice Biotech, Inc. (Taipei, Taiwan). Brown rice (Oryza sativa, Taikeng No. 9) was germinated at 37 °C for 22 hr to obtain GBR. Raw GBR was mixed with water and cooked. After cooling treatment, the cooked GBR was frozen-dried, ground, and screened.

| Experimental design
This protocol for animal study was approved by the Institutional Animal Care and Use Committee of Taipei Medical University. Sixtysix male F344 rats were from the National Laboratory Animal Center (Taipei, Taiwan) and housed in plastic cages in a room maintained at 21 °C with a 12-hr light-dark cycle. They were divided into six groups. Rats in groups B and D were fed AIN-93G diet. Others were fed modified AIN-93G diets containing 10% GBR alone (group G) or combined with L. acidophilus (5 × 10 7 c.f.u./g, group GA), B. animalis subsp. lactis (5 × 10 7 c.f.u./g, group GB), and both strains (2.5 × 10 7 c.f.u./g for each strain, group GAB), respectively. One week after the beginning of experimental diet, all rats except those in group B received DMH (40 mg/kg body weight, i.p.) three times during a week and DSS (2% in drinking water) for 1 week to induce colorectal carcinogenesis. Body weight and food intake were recorded weekly. All rats were sacrificed after 10 weeks of feeding, and colons were collected for analysis.
Briefly, each colon was cut into three equal-length sections and fixed between filter papers in formalin solution for 1 days. The fixed colons were stained with methylene blue solution and examined for ACF using a light microscope (Nikon Corp., Tokyo, Japan). The area of the colon was calculated by NIS-Elements microscope imaging software (Nikon Corp.). The location of each ACF, the number of ACF in each colon section, and the number of aberrant crypt (AC) in each ACF were recorded. Data of ACF and AC were presented as numbers/cm 2 .

| Assay of mucin-producing ACF and mucindepleted foci (MDF)
Mucin-producing ACF and MDF were identified according to the established methods of Jenab, Chen and Thompson (2001)) and Caderni et al. (Caderni et al., 2003), respectively, as described in our previous study (Li et al., 2011). Briefly, the methylene blue-stained colon was faded with 70% ethanol and then stained using high-iron diamine alcian blue (HIDAB) method. Firstly, each colon section was stained with high-iron diamine solution for 30 min and rinsed in distilled water. Secondly, the colon section was stained with 1% alcian blue solution (in 3% acetic acid) for 15 min, rinsed in 80% ethanol followed by distilled water, and finally examined under a light microscope (Nikon Corp.). Brown and blue staining by HIDAB indicated sulfomucin (SUM) and SIM secretion, respectively. SUM-ACF and SIM-ACF were defined as ACF with more than 85% SUM-and SIMstained area, respectively. ACF stained with a smaller percentage of these two mucins were defined as mixed-type ACF (MIX-ACF).
Furthermore, those with very little or no production of mucins were defined as MDF. The area of the colon was calculated using TA B L E 1 Effects of germinated brown rice and synbiotics on DMH/DSS-induced ACF (number/cm 2 ) according to the various size of crypts in the colon of male F344 rats a,b

Group c
Incidence d

| Measurement of antioxidative enzyme activity
The assay of superoxide dismutase (SOD) activity was performed using Superoxide Dismutase Assay Kit (Cayman Chemical, Ann Arbor, MI, USA). Colonic mucosa was homogenized in five volumes of buffer (1 mM EDTA, 210 mM mannitol, 70 mM sucrose, and 50 mM phosphate buffer, at pH 7.4). The commercial kit was used according to the manufacturer's protocol.

| Statistical analysis
All data are expressed as means and standard deviations (SD). The difference between experimental groups was assessed by oneway analysis of variance (ANOVA) followed by Duncan's multiple range test using Statistics Analysis System (SAS Institute, Cary, NC, USA). The p values lower than 0.05 were considered statistically significant.

| Body weight and food intake
There were no significant differences in initial body weight, weight gain, food intake, and food efficiency among groups (data not shown).

| ACF in the colon
As shown in Table 1, all rats treated with DMH/DSS developed preneoplastic ACF in the colon. Group GA had a significantly lower number of ACF containing one crypt than did group D (p < 0.05).
There were no significant differences in the numbers of ACF containing more crypts and total ACF among groups.

| Mucin secretion by ACF
The majority of ACF in the colon secreted SUM ( Table 2). There were no significant differences in the numbers of ACF producing either SUM or mixed SUM and SIM among groups. However, groups GA, GB, and GAB had significantly lower numbers of ACF producing SIM than did group D (p < 0.05).

| MDF in the colon
The incidence of MDF and their distribution in the colon is shown in Table 3. The highest incidence rate (66%) of MDF was observed in group D, although there was no significant difference among groups.
MDF mainly developed in the distal colon, while no MDF appeared in the proximal colon. Groups G, GA, GB, and GAB had significantly lower numbers of MDF in the middle colon than did group D (p < 0.05).

| Colonic SOD activity
As shown in Figure 1, DMH/DSS treatment decreased the SOD activity in colonic mucosa. Group D had significantly lower SOD activity than did group B (p < 0.05). The SOD activities were significantly elevated in groups GA, GB, and GAB compared with group D (p < 0.05).

| Apoptosis-related proteins in the colon
The expressions of apoptosis-related proteins in colonic mucosa are shown in Figure 2. DMH/DSS treatment decreased the expression of pro-apoptotic protein and increased the expression of anti-apoptotic protein. Group D showed significantly lower expressions of p53, Bax, and caspase-3 (p < 0.05) and a significantly higher expression of Bcl-2 (p < 0.05) than did group B. The Bax/Bcl-2 ratio was also significantly lower in group D than that in group B (p < 0.05). In contrast, consumption of GBR and probiotics improved the abnormal expression of apoptosis-related proteins. The expressions of p53 and Bax as well as Bax/Bcl-2 ratio were significantly increased in groups G, GA, GB, and GAB compared with group D (p < 0.05). Group GA showed significantly higher (p < 0.05) caspase-3 expression and significantly lower (p < 0.05) Bcl-2 expression than did group D.

| D ISCUSS I ON
To our knowledge, this is the first study investigating the preventive effect of combination of GBR and probiotics (L. acidophilus and B. animalis subsp. lactis) on colorectal carcinogenesis. Our data indicate that combination of GBR and probiotics may inhibit preneoplastic lesions and regulate antioxidative enzyme and apoptosis-related proteins in the colon. These findings suggest that such synbiotic combination may have the potential to serve as a novel chemopreventive agent for colorectal cancer.
Aberrant crypt foci are the early lesions observed in the colon of animals treated with carcinogen and of humans at cancer risk.
They are commonly used as a biomarker in chemically induced animal models for studies of colorectal carcinogenesis and its prevention or therapy (Corpet & Taché, 2002). The present study showed  (Milosevic et al., 2015). The alteration of mucins is presented primarily by excessive SIM secretion in patients with colorectal cancer and a significant reduction in SUM (Milosevic et al., 2015). It has been shown that SUM secretion is negatively correlated to proliferative index, while SIM production is positively correlated to proliferative activity (Milosevic et al., 2015). An increase in SIM secretion and a decrease in SUM secretion have been observed during colorectal carcinogenesis in both humans and animals. In rats, normal colonic crypts and hyperplastic ACF are stained for SUM, whereas dysplastic ACF are stained for SIM, and the most dysplastic crypts (MDF) do not show any mucin staining (Kim et al., 2016). The above findings confirm that SIM-ACF and MDF are advanced preneoplastic lesions during colorectal carcinogenesis.
In the present study, though the total numbers of ACF were not  (Zhao et al., 2017). In addition, the antioxidative property of exopolysaccharides (EPS) has been assessed in several probiotics such as Lactococcus lactis and Lactobacillus plantarum which have been reported to promote the production of antioxidative enzymes such as SOD (Deepak et al., 2016). Bifidobacteria strains also show an antioxidative capacity related to EPS. The EPS synthesized by B. longum W11 acted as an antioxidant against H 2 O 2induced reactive oxygen species in an in vitro cell model (Inturri et al., 2017).
The present study found that all combinations of GBR and probiotics (groups GA, GB, and GAB) significantly improved the decreased colonic SOD activity induced by DMH/DSS treatment and maintained the activity similar to that of normal rats (group B).
Interestingly, the modulatory effect of GBR and probiotics on colonic mucin alteration showed a similar phenomenon that all combinations of GBR and probiotics (groups GA, GB, and GAB) significantly inhibited the formation of SIM-ACF induced by DMH/DSS. These results suggest that synbiotic combination of GBR and probiotics are more effective in antioxidative activity than GBR given individually. Also, these findings reveal the possibility that enhancement of antioxidative status or reduction of oxidative stress is associated with the suppression of malignant mucin transition in the colon.
The acute apoptotic response to a genotoxic carcinogen (such as DMH) could regulate mutation in the colon and eliminate cells with abnormal DNA that might progress to tumors, so this response might exert a protective effect at the early stage during colorectal carcinogenesis in rats (Le Leu et al., 2005). 1,2-Dimethylhydrazine treatment led to changes in the protein expressions of certain genes involved in the intrinsic apoptosis pathway such as p53, Bax, Bcl-2, and caspase-3 (Walia et al., 2018). A synbiotic combination of B. lactis and RS significantly induced the apoptotic response to a genotoxic carcinogen in the distal colon of rats in a short-term study (Le Leu et al., 2005). RS induced alterations of intestinal microenvironment including acidification of digests, production of SCFAs, and changes in the balance of microbial species, which might create a situation that enabled B. lactis to exert a pro-apoptotic effect (Le Leu et al., 2005). However, the same synbiotic combination used in a long-term (26 weeks) study showed no significant difference in spontaneous apoptotic cells in treatment groups, although this synbiotics significantly protected against the development of colorectal tumors (Le Leu et al., 2009). These findings suggest that apoptosis is an early event during colorectal carcinogenesis and that anticancer agents with apoptosis-inducing ability may tend to be effective at the early stage of colorectal carcinogenesis. The present study was There was a lower Bcl-2 expression in CT-26 cells isolated from mice pre-inoculated with L. acidophilus compared with those from untreated mice. Moreover, the expressions of caspase-3 and caspase-9 were higher in L. acidophilus-treated cells compared with those from untreated mice (Chen et al., 2012). B. lactis Bb12 induced apoptosis through the mitochondrial pathway including Bax translocation, cytochrome c release, and cleavages of caspase-9 and caspase-3 in human colonic carcinoma cells (Altonsy et al., 2010). The impact of synbiotics on apoptosis is a combined effect of both probiotics and prebiotics and usually attributed to the fermentation-mediated production of luminal SCFAs, especially butyrate (Borowicki et al., 2011). An interaction between butyrate produced via fermentation of prebiotics and immunomodulating properties of probiotics may cells (Deepak et al., 2016). A novel EPS purified from L. acidophilus exerted an antiproliferative effect on Caco-2 cells via the apoptotic mechanism in addition to inactivating the inflammatory pathway and stimulating the immune response (El-Deeb et al., 2018). This EPS increased the ratio of apoptotic cells in sub-G0/G1 cell cycle phase and upregulated the expression of IκBα, P53, and TGF genes (El-Deeb et al., 2018). The EPS produced by probiotics may explain, at least partly, the inhibitory effect of probiotics and synbiotics against colorectal carcinogenesis, although further studies are needed to elucidate the exact mechanism.

| CON CLUS ION
In summary, the present study shows that combination of GBR and probiotics L. acidophilus and/or B. animalis subsp. lactis may suppress DMH/DSS-induced colonic preneoplastic lesions via regulating antioxidation and apoptosis in rats. These findings suggest that such synbiotic combination may be a potential functional food or chemopreventive agent for control of colorectal cancer.

ACK N OWLED G M ENTS
This study was supported by the grants from the Ministry of Science

CO N FLI C T O F I NTE R E S T
The authors declare that there are no conflict of interests.

E TH I C A L A PPROVA L
The animal care and experimental procedures were approved by