Chemopreventive effects of Strobilanthes crispus leaf extract on azoxymethane-induced aberrant crypt foci in rat colon

In this work, microscopic and histological studies suggest that Strobilanthes crispus ethanol extract reduce azoxymethane (AOM)-induced colonic aberrant crypt foci (ACF) in rats. S. crispus is considered a traditional medicine and used as an antioxidant. Its leaf contains a large amount of phenolic compounds to which its radical scavenging role is attributed and enhance its ability to eradicate oxidative stress reactions. The study was designed to determine the chemopreventive effect of S. crispus ethanol extract in vivo and in vitro by elucidating the effect of the extract on intermediate biomarkers which can be used as effective predictors of colon cancer. S. crispus was analyzed for DPPH free radical scavenging, nitric oxide (NO) and ferric acid reduction. The results indicated that S. crispus oral administration significantly inhibited colorectal carcinogenesis induced by AOM as revealed by the reduction in the number of ACF. S. crispus down-regulated the expression of PCNA, Bcl2 and β-catenin. Additionally, it exerted a pronounced inhibitory effect on MDA and NO levels and stimulatory effect on CAT and GPx activities. These results demonstrate that S. crispus is a chemopreventive agent for colorectal cancer through the suppression of early and intermediate carcinogenic phases that may be related to its flavonoid content.

radicals react with protein tyrosine and kinase residues and cause a second wave of lipid peroxidation. Lipid peroxidation activates a programmed death cascade and has been identified as a source of erroneous DNA replication, leading to gene mutation, mitochondrial membrane leakage, cell cycle transition and apoptosis (programmed cell death) 5 .
People are increasingly becoming aware that healthy nutrients cleanse their bodies from free radicals and improve their well-being. Nutrients with calcium, folate, fiber, omega-3 fatty acids, vitamin D, and vitamin A are highly appreciated by practitioners and healers as chemopreventive and chemotherapeutic mediators 6,7 . Dietary phytochemicals are proving to be a relatively simple and practical approach to a healthy life, and the daily consumption of flavonoid-rich vegetables or fruits has been recommended to reduce cancer risk. The wide range of chemical components that are naturally present in these dietary products has been associated with oxidative radical scavenging effects 8,9 . Flavonoid-rich phytochemicals from green tea are emerging as therapeutic agents for cancer cases in preclinical and clinical studies 10 . Their pharmacological effects relevant to cancer control have also been reported in the American Journal of Clinical Nutrition in which Lorenz 2013 11 highlighted the active constituents of plant ethanol extracts and pinpointed the relevant mechanism of action. This research has been conducted in the context of chemotherapeutic treatments 12 , and the ethanol extracts have not consistently been shown to reduce the mortality rate of colorectal cancer cases, although tumor size control has been suggested. For instance, the five-year survival rate is only 10% in patients with metastatic colorectal cancer, even with the heavy use of the most common chemotherapy, fluoropyrimidine 5-fluorouracil (5-FU). Nevertheless, fluoropyrimidine 5-fluorouracil (5-FU) has some side effects and has shown cytotoxicity to normal cells 13 , as it irreversibly inhibits thymidylate synthase (TS).
In the present study, we wanted to extend the potential benefits of diets based on flavonoids for cancer prevention. In particular, we evaluated the chemopreventive effects of Strobilanthes crispus, with an emphasis on an azoxymethane (AOM)-induced ACF model. S. crispus (Family of Acanthaceae) is native to countries such as Malaysia, Madagascar and Indonesia. S. crispus is commonly known as "pecah beling" or "jin batu" in the Malay language. The consumption of nutraceutical S. crispus herbal tea is increasing due to the additional nutrients and antioxidants it provides. The recommended daily level of consumption of the leafy ethanol extract is 5 g. S. crispus herbal tea is known for its diuretic, laxative and antidiabetic properties 14 . It contains highly bioactive compounds 15 , which are reported to be responsible for its antioxidant 16 and antibacterial effects 17 and wound healing 18 ; its anticancer 19 and specifically anti-breast and anti-prostate cancer properties have also been assessed 20 . The phenolic components have been identified as one of the main S. crispus bioactive compounds, in addition to the catechins, alkaloids, caffeine, and tannins, which further contribute to the total defense mechanism. Wasman and Mahmood reported the protective effect of S. crispus on gastric ulcers 21,22 , but the potential effect of S. crispus for preventing colorectal cancer formation has not been studied.

Results
S. crispus exhibited DPPH free radical scavenging activity. The ability of S. crispus to reduce the free radical scavenging activity was tested by assaying DPPH. The DPPH absorption is attributed to the donation of an electron or hydrogen to a stable free radical that is observed in vitro as fading of the purple coloration of the absorption band. Our results demonstrated a significant difference between the obtained values of S. crispus-treated samples (approximately 86% at 25 μ g/ml) and control samples. S. crispus exhibited dose-dependent free radical scavenging activity ( Figure S1). S. crispus showed an IC 50 value of 5.44 ± 1.76 μ mol/l, which was closer to that of vitamin C (3.882 ± 0.628 μ mol/l) ( Figure S1, Table 1). S. crispus ethanol extract eradicated the overproduction of oxygen free radicals and significantly recovered the DPPH free radical scavenging activity.
S. crispus reduced ferric oxidation and decreased the level of nitric oxide radicals. Nitric oxide and ferric acids are very unstable radical species that produce stable products i.e. nitrate, nitrite, ferrous when reacting with oxygen molecules. The oxidation effect of S. crispus was tested in vitro under specific emission and excitation range and relative to the reported IC 50 value. Figures S2,S3, and Table 1 illustrate the variation of the scavenging activity with maximum effect in samples treated with S. crispus compared to gallic acid (GA), whereas vitamin C exhibited less effect at the same concentration. The ethanol extract of S. crispus exhibited a scavenging activity against nitric oxide radicals that expressed an IC 50 of 0.39 ± 0.11 μ mol/l compared to vitamin C and GA (IC 50 of 0.529 ± 0.009 and 2.78 ± 0.13 μ mol/l, respectively). Supplementation with the ethanolic extract of S. crispus decreased the reactive oxygen species profiles by decreasing the nitric oxide parameters. A two-fold higher effect on the reduction of iron from the ferric (Fe 3+ ) to ferrous (Fe 2+ ) form was observed with S. crispus-treated samples relative to BHT and vitamin C (12085.7 ± 0.006, 5228.6 ± 0.01, and 1181.4 ± 0.006 mmol/g, respectively) ( Figure S2, Table 1). The cytotoxic effect of reactive oxygen species and the intercellular liberation of nitric oxide directly or indirectly affected the balance between the ferric (Fe 3+ ) and ferrous (Fe 2+ ) derivatives. Thus, in the S. crispus-treated model, the reduction of the ferric (Fe 3+ ) to ferrous (Fe 2+ ) form was remarkably increased compared to that observed with other conventional antioxidant agents.
Phenol and flavonoid contents were identified from the plant ethanolic extract. S. crispus ethanol extract was evaluated for its total phenolic and total flavonoid content. The TPC was 737.67 ± 0.024 mg ( Figure S4-a) and was expressed as gallic acid equivalent in mg per g of plant ethanol extract. The TFC was 262.86 ± 0.005 mg ( Figure S4-b) and was expressed as quercetin equivalent in mg per g of plant ethanol extract (Table 1).
Sublethal dose of S. crispus showed no toxicity when orally administered. S. crispus at a dose of 2500 mg/kg was nontoxic. No mortality or visible manifestations of hepatotoxic and nephrotoxic effects were observed within the 14 days of observations. Furthermore, there were no significant differences in blood parameters (urea, creatinine, total protein, albumin, glucose and enzymes (ALT, AST, ALP)) between the control and treated groups. Oral administration to rats was therefore considered safe. This gives a wide safety margin for the dosage of S. crispus extracts in the tested animals as there were no notable side effects associated to higher doses.

S. crispus supplementation affected body and colon weights. The initial body weights of rats
were measured at the start of the experiment and thereafter weekly, spanning the two weeks of induction and eight weeks of treatment. Apart from the weight gain in the animal group treated with 250 mg/kg S. crispus, the average weights of 500 mg/kg S. crispus-treated and 5-FU drug-treated rats were similar to those of the AOM induced group ( Table 2). The colon weights of S. crispus-treated groups (both doses) were significantly increased compared to the AOM induced group. In general, there were no significant differences in the liver, spleen and kidney weights among all groups. S. crispus induced an increase in the colon and body weight, which might have enhanced the stability of the intestinal wall and therefore improved the removal of superoxide radicals. The integrity of the lumen displayed catalase decomposition of hydrogen peroxide to water and oxygen; thus, allowing the intestinal enzymes to modulate the redox state of lysate and plasma.

S. crispus ethanol extract induced a decrease in ACF number in rat colon. Aberrant crypt foci
(ACF) were visualized as slightly elevated connective tissue above the surrounding mucosa and demonstrated characteristic oval or slit-like orifices. The intestinal lumen had irregular luminal, elongated protruded crypts. The ACF was observed predominantly in the middle and distal colon in AOM-treated animals, and to a lesser extent in the proximal colon (Fig. 1). AOM induction in the animal models leads to mitochondrial dysfunction and inhibits the release of protective intestinal acids. This was accompanied by the increase of the reactive oxygen species content in the intestinal lumen, contributing to hypertrophy of the cells and deformity of the glands followed by the ACF formation.
All rat groups treated with AOM developed ACF. Table 3 summarizes the effect of S. crispus on AOM-induced ACF formation. S. crispus produced a marked decrease in the number of ACF (approximately 71-74% of the AOM-induced values) (P < 0.01). S. crispus supplementation attenuated the distribution of ACF throughout the colon segments. ACF was mainly found in the middle and distal parts of the colon in the ethanol extract-treated groups and in the middle and proximal areas of the colon in the AOM-induced and 5-FU-treated groups. S. crispus treatment resulted in restoring the low number of crypts per focus compared to the AOM-induced rats, as shown in (Table 4). The decrease in the number and size of ACF in S. crispus-treated animals indicated that the extract restored the anatomical integrity and decreased the free radical generation, which may be due to the increased antioxidant defense. The average number of ACF in total and per foci decreased remarkably in the group of animals treated with S. crispus. The decline in the crypt formation was dominant in animals given S. crispus compared to positive and negative control. The above findings may constitute another evidence on how S. crispus addition contributes towards the cellular defense mechanisms against free radicals.

S. crispus-treated rats recovered normal intestinal anatomy and histopathological features.
The histology of ACF and normal colon cells was observed using H&E staining. The colon sections showed normal architecture with intact mucosal and submucosal features. Proliferating mucosal glands and the presence of stained cells with elongated stratified nuclei and depletion of mucus were observed in the colon sections of AOM-induced rats (Fig. 2). It may be relevant that the increase in colon weight was consistent with the reaction of S. crispus towards the free radicals, suggesting that S. crispus played a role in intestinal homeostasis. The long term oral administration of S. crispus revealed no histopathological changes of liver and kidney in plant treated groups with respect to the normal (vehicle) group ( Figure S5).

S. crispus ethanol extract decreased the level of MDA and NO, and increased the GPx and CAT activities.
Radical species play a crucial role in oxidative stress and their excessive generation is believed to react at the mitochondrial system and produce up-and down-regulation of key proteins that have toxic effects in the biological tissues. We tested the indirect suppressive effect of S. crispus on the radical species and the secondary expression of critical cellular proteins. For example, MDA is an aldehydic product of lipid peroxidation that reacts quickly with biomolecules, such as proteins, lipids and nucleic acids and leads to cellular dysfunction. On the other hand, the enhanced interaction of superoxide with NO in the oxidant environment increases the formation of peroxynitrite, which in turn oxidizes tetrahydrobiopterin (BH4, a cofactor for eNOS), leading to eNOS uncoupling. The uncoupling of eNOS, increase in nitric oxide synthase (iNOS) expression and diminished NO bioavailability play major roles in the onset of colon cancer formation. S. crispus and 5-FU caused a remarkable decrease in the level of MDA and lipid peroxidation. The group treated with S. crispus supplementation was found to normalize the formation of oxygen-derived free radicals. The levels of nitric oxide were significantly decreased in S. crispus-treated groups compared with the AOM induced group and the 5-FU reference group. Elevation of GPx and CAT activities was observed in both the AOM induced group and reference 5-FU treated group more than the S. crispus-treated groups ( Table 5). The decline in MDA and NO activities accompanied by a significant increase in GPx and CAT and the overall change of the antioxidant markers indicate a positive effect of S. crispus against the oxidative stress that may also occur in all rat tissues.    Suppression of beta catenin staining in the S. crispus-treated groups. Beta catenin is a multifunctional protein complex that is a component of cell-to-cell adhesion and part of the WNT signaling pathway. Beta catenin is identified also as one protein of DNA polymerase and its high expression suggests either cellular proliferation (turn over) or DNA mutation. Accumulation of beta-catenin in the nucleus was detected in almost all the AOM-treated groups (Fig. 5). Beta-catenin is a key factor in colorectal carcinogenesis and it is mostly expressed in the membrane at an early stage of the cancer, while the nucleus translocation occurs at an advanced stage illustrating the stages of genetic mutations. The labeling index of beta catenin in the colonic mucosa is presented in Fig. 5. Qualitative analysis of the stained sections following AOM-treatment showed a clear elevation and translocation of the protein to the colonic cancer cells compared with groups treated with 5 FU and S. crispus. Microscopic examination of the colonic tissue sections stained for beta catenin clearly showed a faint or no staining in S. crispus-induced AOM-treated groups.

Discussion
We investigated the chemopreventive effect of S. crispus ethanol extracts on AOM-induced colorectal lesions by in vivo and the scavenging activity for free radicals by in vitro assays. AOM has been commonly used as a carcinogenic method to determine the chemopreventive effects of food ingredients such as indigestible sugars, red meat, and green tea in rodent models 23 . We used the AOM experimental approach 24 to initiate colon cell outgrowth that mimics the cryptal foci lesions in cancers. The AOM carcinogenic effect is largely reported in the literature by other groups of researchers 25 , as well as in our work 9 . In fact, we had already validated this approach from histopathological and biochemical perspectives in a previous study 26 . The dose safety and efficacy had been evaluated previously 18 and in the present study and in accordance with the highest recommended dose, and the animals were healthy and showed no hepatotoxicity and/or nephrotoxicity. Moreover, because an objective of the present work was to study the pharmacological effects of S. crispus, a simple acute model was useful. S. crispus ethanol extract reduced the expected aberrant cryptal foci histological deficit, restoring the anatomical appearance of the wall as well as the physiological performance of mucus production. Histologically, the AOM-induced lesions showed proliferating mucosal glands with ACF characterized by  Table 5. Effect of S. crispus on GPx and CAT activities, NO and MDA levels in colon homogenate 60 days after treatment. All values are expressed as mean ± SEM. Significant difference at *P < 0.01, **P < 0.001, ***P < 0.0001 (ANOVA, Tukey's post hoc). 5-FU: 5-Fluorouracil; ACF: aberrant crypt foci; AOM: azoxymethane.  PCNA and Bcl2 antibodies are commonly used for grading neoplasms 27,28 . It is well known that AOM produces loss of mucosal integrity and decreases the power of releasing mucin, accompanied by deficits in the submucosal glands 29 . Introduction of S. crispus ethanol extracts before AOM seems beneficial in retarding colorectal cancer formation by preventing cell proliferation and inhibiting the activation of PCNA and Bcl2. In conclusion, the integrity of the intraluminal wall structure is fundamental from both a protection and absorption prospective, as it contributes to the encoding of physiological and anatomical balance, hinders the transmission of harmful chemical signals and is essential for maintaining the flow in the mucous layer on the intraluminal surface. Therefore, it is not surprising that S. crispus supplementation increased the colon and total animal weights compared to other treatment groups. Colorectal cancer, among other serious cancers, is correlated with oxidative insults. In vitro evidence shows that AOM administration generates a toxic cascade that mediates the apoptotic pathway in culture models 30 . Our data indicated the scavenging effect of S. crispus in the treated samples. S. crispus inhibits oxidative stress via attenuation of apoptotic signaling pathways in the cell. The high potent activities of S. crispus were due to the deactivation of the metabolic redox correlated directly with its suppressing effect on mitochondrial and nuclear levels. The S. crispus ethanol extract was shown to attenuate the toxic action of oxidative free radicals induced by the demonstration of AOM. The effect was higher than that of the anti-oxidative agents (i.e., vitamin C, gallic acid, quercetin and butylated hydroxytoluen) as assessed by DPPH, ferric reducing and nitric oxide scavenging assays. AOM supplementation induced the accumulation of free radicals and increased the intracellular level of apoptotic-sensitive proteins. The free radical scavenging mechanism involves oxidative-mediated apoptosis, which requires ferric acid reduction; therefore, the production of NO and iNOS is regulated 31 . S. crispus at an IC50 of 5.44 + 1.76 effectively inhibited DPPH free radicals, nitric oxide and ferric reduction sequences. These findings are supported by other studies that have shown decreased MDA 31 and NO levels 32 in rats treated with Strobilanthes asperrimus. Moreover, the measurements of serum biomarkers following oxidative stress showed that the levels were within the normal range and suggested a possible role of S. crispus to normalize the level of serum biomarkers (i.e., urea, creatinine, total protein, albumin, glucose and enzymes (ALT, AST, and ALP)). The apoptotic signaling pathway of AOM induced a deleterious increase in the apoptosis-sensitive proteins (specifically MDA, GPx and CAT, among others) in the tissue, which were attenuated by S. crispus. The intestinal microenvironment after chemical erosions by AOM seems to produce the second messenger nitric oxide (NO) following the excessive release of inducible nitric oxide synthase (iNOS). The production of iNOS is Ca 2+ -independent and known to participate in and inhibit the initiation of certain diseases. This process altered specific genes expression as well as a wide range of proteins and signaling cascades 31,32 . For instance, ROS release causes lipid peroxidation of polyunsaturated fatty acids and damage to specific mitochondrial proteins and transport systems by direct inhibition of enzymes resulting in loss of mitochondrial integrity. Malondialdehyde (MDA), one of the final products of polyunsaturated fatty acids peroxidation, is commonly known as a marker of oxidative stress and reflects the antioxidant status. In the present study, the data showed that MDA levels significantly decreased, indicating that S. crispus stopped the lipid peroxidation that is the prominent player in the membranes damage. Previous studies have demonstrated that the primary antioxidant enzymes which include SOD, GPx and CAT generate the antioxidant defense systems. The antioxidant enzymes provide the first line of cellular defense against ROS. In the present study, the data showed that SOD, GPx, and CAT significantly increased leading to reduction of the free radical damage. Additionally, PCNA and Bc12 genes are produced by enzymatic cleavage and considered as a biomarker of oxidative DNA damage. Increased levels of DNA damage can cause the synthesis of a variety of incorrect proteins and therefore mutations and cancer formation.
At least two models for genetic alterations can be characterized in the development of colorectal cancer. In the presence of β -catenin staining, the scenario can be that the transcriptional effects were initiated by Rac1-dependent nuclear translocation of β -catenin followed by mutation of the Wnt ligand, and deleted in colorectal cancer (DCC). The WNT signal transduction can also transcriptionally activate both NOS2 and COX2 by TCF4-β-catenin binding to their promoters in a regulatory feedback loop. Under these conditions and secondary to the release of iNOS, K1 or 2 kinases and GSK3beta phosphorylated β-catenin and activated threonine and serine residues in the N-terminal domain. Conversely, the NO-related DNA damage leads to p53 accumulation and p53-mediated apoptosis. NO can activate the WNT pathway by enhancing the enzymatic activity of cyclo-oxygenase 2 (COX2 and the release of prostaglandins 2E (PGE2)) 25 . Beside these two pathways there is a possibility that less characterized co-activators/co-repressors displayed the transcription with a high degree of overlap among the main pathways.
Finally, recent studies have opened a new perspective for the possible use of S. crispus in medicine 18,22 . In fact, experimental data clearly indicated a significant protective role of S. crispus in breast cancer 20 . These findings are consistent with previous studies reporting the phenolic content and antioxidant activity of this plant 15,16,33 . Moreover, a recent study demonstrated the phenolic content and cytotoxic potential role of S. crispus 16,34 . Accordingly and consistent with this research, our laboratory recently reported, in vivo, using rats, a significant inhibition of colorectal cancer formation after similar treatment with medicinal plants (Andrographis paniculata and Gynura procumbens) 9,25 . Our results demonstrated the chemoprotective efficiency of S. crispus ethanol extracts, which can be attributed to its anti-proliferative, antioxidant and anti-radical scavenging properties. Accordingly, this natural agent might provide a potential template for the design of drug targets for treating colorectal cancers.

Materials and Method
Chemicals. Azoxymethane (Sigma Chemical Co., St Louis, MO, USA) was dissolved in sterile 0.9% normal saline and was given subcutaneously at 15 mg/kg body weight once a week for 2 weeks to induce ACF 35 . Fluorouracil (5-FU) (Sigma Chemical Co., St. Louis, MO, USA) was dissolved in 0.9% normal saline and was administered at a concentration of 35 mg/kg body weight intraperitoneally twice a week 36 .
Plant materials. The fresh plant leaves were obtained from Ethno Resources Sdn. Bhd., Selangor, Malaysiy, and identified by a voucher specimen placed at the Herbarium of Rimba Ilmu, Institute of Biological Sciences, University of Malaya, Kuala Lumpur. The leaves were washed, shade-dried and grounded. The powdered plant material (100 g) was extracted with ethanol 95% (1000 ml) at room temperature three times, three days per each crop with occasional stirring and filtered. Three extracts were combined and evaporated in vacuo to give 11.5 g of crude ethanol extract 37 which was kept dry at 45 °C, and the clear semisolid ethanol extract was then dissolved in the vehicle solution.
In vitro study of the antioxidant properties of S. crispus ethanol extract. The 2, 2-diphenyl-1-picryl-hydrazyl free radical scavenging assay. The free radical scavenging activity of the plant ethanol extract was determined spectrophotometrically using 2, 2-diphenyl-1-picryl-hydrazyl (DPPH, Sigma-Aldrich, UK). DPPH, which is violet, produces a yellow complex when it is reduced in the presence of antioxidants 38 . The DPPH working solution was prepared using 95% ethanol at a concentration of 3.94 mg/100 ml, and ascorbic acid was used as the standard. The sample and DPPH reagent were mixed well and incubated before the absorbance was measured using an ELISA plate reader at 517 nm in triplicate 39 . The scavenging activity of the ethanol extract was calculated based on the percentage of scavenged DPPH. (BIO-TEK instruments, Winooski, VT, USA).
Nitric oxide scavenging activity. Nitrite (NO 2, ) is the stable product of nitric oxide conversion as assessed using Griess reagent (Sigma-Aldrich, UK). A series of dilutions were prepared from a 1 mg/ml stock of S. crispus ethanol extract. For every concentration, 50 μ l of the ethanol extract was transferred into a 96-well microplate, and an equal volume of sodium nitroprusside was added prior to the addition of Griess reagent (100 μ l). After 5 minutes, the absorbance was read using a microplate reader at 550 nm under visible polychromatic light 41 using sodium nitrite as the standard. The percentage of inhibition was obtained by calculating the ability of the plant ethanol extracts to inhibit nitric oxide formation relative to 0% inhibition of the control 42 . This ability was defined as (Ao − As/Ao)*100, where Ao is the absorption of control, and As is the absorption of the test samples.
Total phenolic and total flavonoid content of the plant ethanol extract. Total phenolic content (TPC) was determined according to the Folin-Ciocalteu (Thermo Fisher Scientific, USA) spectrophotometric method with slight modifications using a microplate reader. The measurement was compared to a standard curve of GA solution, and the value was expressed as milligrams of GA equivalent (GAE) per gram of dry plant ethanol extract (mg GAE/g db) 43 . The total flavonoid (TFC) content was measured using the aluminum chloride colorimetric method. The measurement was expressed as a quercetin equivalent in mg (QE)/g of ethanol extract 44 . The assays were performed in triplicate. Experimental animals. Healthy adult SD female rats were used for the acute toxicity study, whereas male rats were used for the chemopreventive study. Rats (6-8 weeks old), weighing between 150-180 g, were placed individually in separate cages. The animals were maintained on a standard diet and tap water ad libitum, under controlled conditions of humidity (50-60%), lighting (12-h light/dark cycle) and temperature (22-24 °C), and they were weighed weekly.

In vivo
Acute toxicity study. A total of 16 healthy SD female rats were obtained from the Experimental Animal House, Faculty of Medicine, University of Malaya. The animals were divided equally into two groups: a) treated with vehicle (10% Tween-20, Thermo Fisher Scientific, USA) and treated with 2500 mg/kg ethanol extract in vehicle, respectively. The animals were fasted overnight (food but not water) prior to dosing. Food was withdrawn for a further 3 to 4 hours, and the animals were observed for 30 min and at 2, 4, 24 and 48 h for the onset of clinical or toxicological symptoms. Mortality, if any, was noted over a period of 2 weeks; then, the animals were sacrificed on the 15th day. The colon, liver and spleen weights were reported. Serum biochemical and histological (liver and kidney) parameters were determined following standard methods 45 .
Experimental design. Following the safety/toxicity test and within the range of the chosen safe dose, we divided the adult male SD rats randomly into the following five groups:-Group 1: (Saline group) injected with 0.9% sterile normal saline once a week for two weeks and fed daily with 10% Tween-20 (Thermo Fisher Scientific, USA) (5 ml/kg) for 60 days.
Group 2: (AOM group) injected with (15 mg/kg/ml) of AOM once a week for two weeks and fed daily with 10% Tween-20 (5 ml/kg) for 60 days.
Group 3: (5-FU group) injected with (15 mg/kg/ml) of AOM once a week for two weeks and injected intraperitoneally with (35 mg/kg/ml) of 5-FU twice weekly eight weeks. Groups 4 & 5: (Plant treated groups) injected with (15 mg/kg/ml) of AOM once a week for two weeks and fed daily with 250 mg/kg and 500 mg/kg (5 ml/kg) of ethanol extract for 60 days 36,46 .
Histopathological examination. All rats were sacrificed by cervical dislocation after receiving an overdose of ketamine and xylazine (50 and 5 mg/kg body weight, respectively). Colons were removed and washed with cold phosphate buffered saline (PBS, Thermo Fisher Scientific, USA). They were then dissected longitudinally and cut into three equal segments (proximal, middle, and distal). The segments were fixed flat between filter paper in 10% formalin for 5-10 minutes and stained with methylene blue (Sigma-Aldrich, UK) (0.2% in PBS solution) for 15-20 minutes to visualize the internal surface of the lumen furnished with the crypts. Each segment was further sliced into 2 cm sections and placed on microscope slides with the mucosa facing upwards. The aberrant crypts were recognized by their abnormal sizes and scored under the light microscope at 4× magnification. The number of ACF per colon, the number of aberrant crypts (ACs) in each optical focus, and the location of each focus were recorded 46,47 . The paraffin blocks were processed (Leica Microsystem, Nussloch, Germany) and sectioned at 4 μ m thickness using a microtome and stained with hematoxylin-eosin (H&E). Sections were examined and imaged with a Nikon camera connected to an Olympus light microscope (Tokyo, Japan).
Biochemical markers. Blood samples were collected in gel activating tubes and centrifuged at 3000 rpm for 10 minutes. The serum was separated and sent for determination of creatinine, total protein, glucose, albumin, alkaline phosphatase (ALP), alanine aminotransferase (ALT), aspartate aminotransferase (AST) and urea levels. Biochemical parameters were determined using the standard automated analyzer at the Central Diagnostic Laboratory, University of Malaya Medical Center.
Determination of antioxidants in colon homogenate. The colon tissue was washed with ice-cold PBS solution, pH 7.4, to remove other cells or debris. The colon tissue was homogenized on ice and centrifuged at 4500 rpm for 15 minutes at 4 °C. Tissue homogenate supernatant was used to measure the antioxidants 48,49 . Protease, a mixture of protease inhibitors with broad specificity for the inhibition of serine, cysteine, aspartic proteases and amino peptidases, was used to preserve enzymes from being degraded. Malonaldehyde (MDA) (cat #10009055) and nitric oxide (NO) (cat #D2NO-100) levels were measured. Glutathione peroxidase (GPx) (cat #703102) and catalase (CAT) (cat #707002) activities were also measured. All measurements were performed using commercial kits (Cayman Chemical Company, U.S.A) on a microplate reader.
Immunohistochemical staining. Colon sections were mounted on poly-lysine coated slides, dried overnight in an oven at 60 °C and then deparaffinized and rehydrated in graded alcohols. For retrieval of the antigen target, a retrieval solution containing 10 mM sodium citrate buffer was used.
Immunohistochemical staining was done according to the manufacturer's protocol (Dako ARK ™ USA), based on avidin, biotin and peroxidase methodologies. Biotinylated primary antibody of diluted mouse PCNA and Bcl2 monoclonal antibody were applied at 1:100 and 1:50 dilution, respectively, followed by applying streptavidin-peroxidase (streptavidin conjugated to horseradish peroxidase in PBS containing the anti-microbial agent). Proliferating cell nuclear antigen (PCNA) is a well-known indicator of proliferation, and Bcl2 is a well-known anti-apoptotic protein indicator. Staining was completed by a 5-minute incubation with 3, 3'-diaminobenzidine (DAB)-substrate-chromogen. Negative control sections were processed similarly but with the omission of the primary antibodies. Slides were rinsed in a bath of distilled or deionized water for 2-5 min, mounted and cover-slipped with an aqueous-based mounting medium or a non-aqueous permanent mounting.
Beta catenin immunostaining. Briefly, sections were deparaffinized and rehydrated through graded alcohol. Endogenous peroxidase activity was blocked with 0.5% hydrogen peroxide in methanol for 20 minutes. Antigen retrieval was used by microwave treatment; dilution was 1:100 for beta catenin (Novocastra, United Kingdom). Biotinylated universal antiserum (Starr Trek HRP Universal, Biocare Medical, CA) was used as the secondary antibody. After washing, the slides were incubated for 20 minutes at room temperature with Trekavidin HRP label (Starr Trek HRP Universal Biomedical CA) and developed for 10 minutes using 3-3-diaminobenzidine as chromogen. After rinsing in water, the sections were counterstained with Meyer haematoxylin, dehydrated, and mounted. Appropriate positive and negative controls were included in each run of immunohistochemistry.

Statistical analysis.
Values were expressed as the mean ± SEM. Variation between groups was investigated employing one-way ANOVA followed by Tukey's post-hoc test using SPSS version 20 (SPSS Inc. Chicago, IL, USA). P values less than 0.05 were statistically significant when compared to the control group.