Ursolic Acid and Oleanolic Acid Suppress Preneoplastic Lesions Induced by 1,2-Dimethylhydrazine in Rat Colon

Abstract

Ursolic acid (UA) and oleanolic acid (OA) are pentacyclic triterpenoid compounds found in plants used in the human diet and in medicinal herbs, in the form of aglycones or as the free acid. These compounds are known for their hepatoprotective, anti-inflammatory, antimicrobial, hypoglycemic, antimutagenic, antioxidant, and antifertility activities. In the present study, we evaluated the effects of UA and OA on the formation of 1,2-dimethyl-hydrazine (DMH)–induced aberrant crypt foci (ACF) in the colon of the male Wistar rat. The animals received subcutaneous (sc) injections of DMH (40 mg/kg body weight) twice a week for two weeks to induce ACF. UA, OA and a mixture of UA and OA were administered to the rats five times a week for four weeks by gavage at doses of 25 mg/kg body weight/day each, during and after DMH treatment. All animals were sacrificed in week 5 for the evaluation of ACF. The results showed a significant reduction in the frequency of ACF in the group treated with the triterpenoid compounds plus DMH when compared to those treated with DMH alone, suggesting that UA and OA suppress the formation of ACF and have a protective effect against colon carcinogenesis.

Introduction

Ursolic acid (3β-hydroxy-urs-12-en-28-oic-acid) (UA) and its isomer, oleanolic acid (3β-hydroxy-olea-12-en-28-oic-acid) (OA), are triterpenoid compounds found in plants used in the human diet and in medicinal herbs, in the form of the free acid or as aglycones of triterpenoid saponins. UA and OA share many common pharmacological properties. These triterpenoids have shown hepatoprotective (Liu, 1995), antiallergic (Banno et al., 2004), anti-ulcer (Ovesná et al., 2004), cardioprotective (Senthil et al., 2007), antimicrobial (Ngouela et al., 2005), anti-inflammatory, and analgesic (Vasconcelos et al., 2006) activities. These compounds also present an antiparasitic activity against Leishmania species (Torres-Santos et al., 2004) and Trypanosoma sp (Cunha et al., 2006) and an antiprotozoal potential against Plasmodium falciparum (Van Baren et al., 2006). UA and OA have also been used in the treatment of kidney diseases (Yoshimura et al., 2003) and hypertension (Somova et al., 2004). In addition, protective effects against periodontal pathogens, antiviral activity against HIV (Ovesná et al., 2004), and an anti-tubercular potential against Mycobacterium tuberculosis (Gua et al., 2004) have been reported, in addition to immunomodulatory, hypoglycemic (Zhang et al., 2006) and antifertility activities (Chattopadhyay et al., 2005).

Over the past few years, interest has been increasing on the part of researchers in the chemopreventive, antioxidant, antiproliferative, cytotoxic, and anti-invasive properties of triterpenoids. UA and OA can act on various stages of tumor development, including the inhibition of tumor initiation and promotion, as well as inducing tumor cell differentiation and apoptosis. UA and OA are effective in inhibiting angiogenesis, invasion of tumor cells, and metastases (Liu, 2005; Ovesná et al., 2006).

Pharmacological studies indicate that UA and OA have many beneficial effects; some are attributed, in part, to their action against free radicals. However, many aspects of their biological activities are not completely understood. Therefore, the purpose of the present study was to investigate the possible chemopreventive activity of UA and OA on the formation of the carcinogen 1,2-dimethylhydrazine (DMH)–induced aberrant crypt foci (ACF), which are considered to be an important early step in colorectal carcinogenesis and tumors of the rat colon.

Materials and Methods

Extraction and Isolation of Ursolic Acid and Oleanolic Acid

Miconia fallax was collected along the Franca-Claraval highway, São Paulo, Brazil, and identified by Dr. Angela Borges Martins, Institute of Biology, UNICAMP, Campinas, Brazil. Voucher specimens (UEC 65132) were deposited in the herbarium of this institute. The aerial parts of Miconia fallax were dried at 40°C, and the dried material was powdered (M. fallax: 1.3 kg) and extracted by maceration with methylene chloride (51, three days × 3) and ethanol (51, three days × 3) at room temperature. The methylene chloride extract of M. fallax (13.0 g) was chromatographed on 300 g silica gel 60 (0.063–0.200 mm; Merck) by vacuum liquid chromatography and eluted with hexane and mixtures of hexane–ethyl acetate (AcOEt) and AcOEt–ethanol of increasing polarity, resulting in six fractions of 1000 mL each (F1: hexane, F2: hexane–AcOEt [3:1], F3: hexane–AcOEt [1:1], F4: AcOEt, F5: AcOEt–EtOH [3:1], F6: AcOEt/EtOH [1:1]). The ¹H- and ¹³C-NMR data of fractions F4 (1.76 g) and F5 (1.23 g) revealed the presence of a mixture of UA and OA. Part of this pooled fraction (500 mg) was filtered through 60 g of a mixture of Celite and Norit (3:1) and eluted with AcOEt, yielding 350 mg of a mixture containing 65% UA and 35% OA. These compounds were separated by HPLC using an isocratic mobile phase consisting of hexane–isopropyl alcohol (49:1) at a flow rate of 9 mL/min, resulting in UA as a white amorphous solid [(t_R: 5.7 min), +63.4° (c 0.18, MeOH)] and OA as a white amorphous solid [(t_R: 5.0 min), +68.2° (c 0.18, MeOH)].

The chemical structures of the two compounds were established by ¹H and ¹³C NMR data analysis in comparison to those of authentic compounds. Purity of the isolated compounds was determined by HPLC and ¹³C NMR (Cunha et al., 2006) and was considered to be higher than 95% (Figure 1).

Animals

Male Wistar rats (Rattus norvegicus) weighing approximately 120 g were obtained from the animal house of the Faculty of Pharmaceutical Sciences, University of São Paulo, Ribeirão Preto (São Paulo, Brazil) and acclimated for a period of one week before the beginning of the experiment. The animals were kept in plastic boxes in an experimental room under controlled conditions of temperature (22 ± 2°C) and humidity (50 ± 10%) under a twelve-hour light–dark cycle; standard rat chow and water were available ad libitum. This study was conducted in accordance with the internationally accepted principles for laboratory animal use and care, as found in the US guidelines (NIH #85–23, revised in 1985). The study protocol was approved by the Ethics Committee for Animal Care of the University of Franca (process no. 023/05A).

Carcinogen Treatments

The well-known colon carcinogen DMH (Sigma-Aldrich) was dissolved immediately before use in 1 mM EDTA. A total dose of 160 mg/kg body weight (bw) was divided into four subcutaneous (sc) injections of 40 mg/kg bw administered in two doses/week for two weeks (weeks 2 and 3), as described by Takahashi et al. (1992).

Experimental Design

Each experimental group consisted of six animals fed a standard chow throughout the experiment (five weeks). After one week of acclimation, the animals were divided into ten treatment groups: animals treated with UA (25 mg/kg bw); OA (25 mg/kg bw); a mixture of UA and OA (25 mg/kg bw; 65% UA and 35% OA); DMH (160 mg/kg bw); dimethyl sulfoxide solvent (DMSO; Sigma-Aldrich; 0.6 g/kg bw) and DMH; UA and DMH; OA and DMH; UA/OA and DMH; and negative (EDTA) and solvent (DMSO) controls according to the scheme shown in Table 1. In the present investigation, a study period of four weeks was selected on the basis of literature data. This period was considered to be long enough to observe the formation of ACF (Bazo et al., 2002; Rodrigues et al., 2002). The negative and positive controls received EDTA (0.05 mL/10 g bw) or DMH (40 mg/kg bw), respectively, twice a week for two weeks (weeks 2 and 3). The doses of UA and OA were selected on the basis of preliminary experiments. UA, OA, and a mixture of UA (65%) and OA (35%) were administered to the rats five times a week for four weeks by gavage in doses of 25 mg/kg bw/day each, during and after DMH treatment. A solution of 4 mL/kg UA, OA, or a mixture of UA and OA in 14% DMSO in water was administered. The amount of DMSO administered was identical in all treatment groups. Body weight and water consumption were measured five times a week throughout the experimental period. All animals were placed under anesthesia with sodium pentobarbital (45 mg/kg bw, i.p.) and sacrificed five weeks after the beginning of the experiment (i.e., four weeks after the first DMH treatment) by exsanguination.

ACF Assay

After laparotomy, the colons were excised, flushed with 0.9% saline, cut open along the longitudinal axis, and fixed in 10% phosphate-buffered formalin (pH 6.9–7.1) for twenty-four hours. Immediately before analysis, the colon was stained with 0.02% methylene blue for five minutes, mounted on microscope slides with the mucosal side facing upward, and observed under a light microscope at 100x magnification. Fifty sequential fields of the distal colon were screened for ACF, which were characterized by elongated, slit-shaped lumens surrounded by thickened epithelium that stained more intensely than the surrounding normal crypts (Bird, 1987) (Figure 2). The number of ACF and crypt multiplicity (number of crypts in each focus) were recorded. The multiplicity of ACF is expressed as aberrant crypts (AC)/focus. Each colon specimen was examined by at least three observers in a double-blind manner.

Statistical Analysis

All data were analyzed statistically by analysis of variance for completely randomized experiments, with calculation of the F statistics and respective p values. In cases in which p < .05, treatment means were compared by the Tukey test, and the minimum significant difference was calculated for α = .05.

Results

Table 2 shows the final body weight, body weight gain, and water consumption during the experimental period. No statistically significant differences in these variables were observed between groups.

No ACF were observed in the negative control and in the groups receiving UA, OA, a mixture of UA and OA, or DMSO (data not shown), but only in DMH-treated rats. Table 3 shows that the number of ACF and the number of AC were significantly lower in the groups treated with UA, OA, or a mixture of UA and OA during and after DMH treatment compared to the group treated with only DMH. No significant differences were detected in the number of ACF or AC between the groups treated with UA, OA, or a mixture of UA and OA plus DMH. Animals treated with DMSO plus DMH did not differ significantly from animals of the DMH group. The number of AC and the AC/ACF ratio obtained for all groups treated with DMH showed a higher frequency of foci with one crypt.

Discussion

In the present study, analysis of the frequency of ACF demonstrated the protective effect of the triterpenoid compounds UA and OA against DMH-induced putative preneoplastic foci in the rat colon. In vivo studies have shown that DMH is metabolized to azomethane, azoxymethane, methylazoxymethanol, ethane, and carbon dioxide (Swenberg et al., 1979). Furthermore, DMH has been reported to induce carcinogenesis in rats and mice because of the high production of reactive free radicals (Salim, 1993; Tomasi et al., 1987) that react with DNA, thus demonstrating its genotoxic effect. Although the mechanisms underlying the protective effect against ACF formation are not clearly understood, the inhibitory action of UA and OA might be explained, in part, by their putative antioxidant activity. Recently, Ovesná et al. (2006) showed that UA and OA significantly inhibited the incidence of single-strand DNA breaks induced by H₂O₂, with this effect being attributed to the antioxidant capacity of these compounds.

The inhibitory effects of antioxidants on ACF were also observed in other studies. Morioka et al. (2004) showed that Peucedanum japonicum, a traditional herb in the Ryukyu Islands and an antioxidant, inhibited ACF formation induced by azoxymethane carcinogen. Cloudy apple juice containing mononeric polyphenols, complex polyphenols, or nonpolyphenolic compounds, such as pectin, significantly reduced the formation of DMH-induced ACF in the colon of rats (Barth et al., 2005). The activity of resveratrol antioxidant as an inhibitor of colon carcinogenesis was evaluated in Wistar rats. The results showed that resveratrol markedly reduced ACF incidence, and oxidative imbalance in DMH-treatment was significantly modulated with the supplementation of this compound (Sengottuvelan et al., 2006).

UA and OA are natural, relatively nontoxic compounds that find an application not only as preventive agents but also as new therapeutic agents (Liu, 1995). One of the important properties of these compounds is their antioxidant capacity. Martin-Aragón et al. (2001) investigated possible protective effects of UA against carbon-tetrachloride–induced alterations in antioxidant defense enzymes in vivo and in vitro. The authors suggested that UA may prevent the initiation and propagation of the lipid peroxidation process by scavenging free radicals through conjugation with glutathione, with the consequent control of oxidative damage and tissue protection. Inhibition of lipid peroxidation and an increase in glutathione mediated by OA have been proposed to play a role in the prevention of carbon-tetrachloride–induced liver injury (Liu, 1995; Visen et al., 1998). Balanehru and Nagarajan (1992) demonstrated the free radical scavenging potential of these compounds against adriamycin-induced lipid peroxidation in both liver and heart microsomes in vitro, with OA being identified as a strong protector and UA as a mild protector.

The elimination of free radicals by Nepeta sibthorpii, a plant commonly found in Greece, was studied by Miceli et al. (2005), who used a spectrophotometric method based on the reduction of the stable radical 1,1-diphenyl-2-picrylhydrazil. In that study, UA was found to be one of the main active compounds, again demonstrating its antioxidant capacity. Saravanan et al. (2006) investigated the hepatoprotective activity of UA and showed that, in addition to reducing lipid peroxidation markers in plasma, UA increased the levels of circulatory antioxidants such as reduced glutathione, ascorbic acid, and α-tocopherol, thus demonstrating that the protective effect of this agent is probably related to its antioxidant capacity.

Another possible mechanism of UA and OA responsible for the reduction of ACF induced by DMH might be their action on cytochrome P450 enzymes. DMH is metabolized by cytochrome P450 and this metabolism is required for its carcinogenicity. Albano et al. (1989) observed that the formation of free radicals from DMH in rat liver microsomes and rat hepatocytes was inhibited by cytochrome P450 inhibitors. In this respect, UA and OA have been shown to reduce the activity and expression of several cytochrome P450 enzymes in human liver microsomes (Kim et al., 2004). Therefore, one target of the chemopreventive effect of UA and OA might be the inhibition of the metabolizing activity of cytochrome P450.

Wargovich et al. (2000) observed that UA administered in the feed had no effect on formation of azoxymethane (AOM)-induced ACF in rat colon. However, Andersson et al. (2008), using chow containing a different dosage of UA, obtained similar results to the present study. UA administration to rats during the cancer initiation phase significantly reduced the incidence of ACF containing three or more crypts, but it had no significant effect in the promotion/outgrowth phase. This finding might indicate that once the first mutation in AOM-induced carcinogenesis has occurred, the preventive effects of UA are significantly reduced. These authors have also suggested that UA has chemopreventive effects in the initiation phase of colon cancer associated with changes in sphingomyelin metabolim.

The present results show that treatment with a mixture of UA (65%) and OA (35%) significantly reduced the frequency of ACF and the number of AC induced by DMH, although there was no synergistic effect since this reduction was similar to that observed in treatments involving each compound alone plus DMH. Resende et al. (2006) also found that a mixture of UA and OA was no more effective than each compound alone in the protection against chromosomal damage caused by doxorubicin. Other experiments using a combination of chemopreventive agents also reported no synergistic effect (Antunes et al., 2000; Tavares et al., 1998). One possible explanation for the absence of a synergistic effect of the compounds evaluated might be related to the site of action of these antioxidants, with both substances acting on the same site.

In conclusion, the results of the present study demonstrated that UA, OA, and a mixture of UA and OA significantly reduced DMH-induced ACF and AC in rat colon. In addition, the results indicated UA and OA as promising candidates in the prevention of various types of cancer and other diseases. Further investigations elucidating the mechanisms of action of these triterpenoid compounds are desirable before they can be used as chemopreventive agents in humans.