INTRODUCTION

Recently, several studies of alpha-amylase inhibition have been conducted with the aim of discovering new potential drugs capable of reducing postprandial hyperglycemia that could be used in the treatment of diabetes mellitus type 2 and obesity. Inhibitors of this enzyme reduce postprandial hyperglycemia by delaying carbohydrate digestion and decrease intestinal glucose absorption (^{Ali et al. 2006}, ^{Funke and Melzig 2006}, ^{Gad et al. 2006}, ^{Kim et al. 2005}, ^{Shu et al. 2009}). Plants continue to play an important role in the treatment of diabetes, particularly in developing countries (^{Marles and Farnsworth 1994}, ^{Oliveira et al. 2005}). Medicinal plants from the Cerrado biome has been previously investigated by bioprospection (de Mesquita et al. 2009, Flausino et al. 2009, ^{Mesquita et al. 2005}, ^{Napolitano et al. 2005}, ^{Rodrigues et al. 2006}). However, few studies have investigated the biological activity of such medicinal plants extracts or their isolated components on the activity of alpha-amylase (^{Silva et al. 2007}, ²⁰⁰⁹).

Inhibition studies of alpha-amylase activity by natural products have been performed in vitro using human pancreatic alpha-amylase (HPA) (^{Ali et al. 2006}, ^{Conforti et al. 2005}, ^{Gad et al. 2006}). Alternatively, as human salivary alpha-amylase (α-1,4-D-glucan-4-glucanohydrolases, EC 3.2.1.1, HSA) is highly homologous to HPA, and both proteins adopt very similar structures (^{Brayer et al. 1995}, ^{Ramasubbu et al. 2003}), the use of HSA as an experimental model for in vitro inhibition studies with natural products has the advantage of saliva being ease and not invasive to collect. Previous studies have shown similar in vitro inhibition of HPA and HSA activities (^{Kim et al. 2005}, ^{McDougall et al. 2005}).

The aims of this study are (1) to screen selected plants extracts from the Brazilian Cerrado for possible HSA inhibitory activity, and (2) to carry out a biological assay in mice to investigate any effects on glycemic and toxicity of the stem bark extract from Pouteria ramiflora (Mart.) Radlk.

MATERIALS AND METHODS

RESULTS AND DISCUSSION

One hundred and nine crude extracts were tested, of which 47 were ethanolic and 52 hexanic. In the initial screening, all extracts were prepared as hexanic and ethanolic crude extracts from different parts of 17 plant species belonging to six botanical families (Flacourtiaceae, Sapindaceae, Sapotaceae, Bignoniaceae, Asteraceae and Apocynaceae) (Table I). The percentages of inhibition of the HSA by the most effective extracts are shown in Table II. Only 16 extracts from six species belonging to the botanical families, Sapotaceae, Flacourtiaceae and Sapindaceae, were found to have an inhibitory effect on HSA activity (greater than 75%), including four ethanolic extracts obtained from root bark, root wood and stem bark of P. ramiflora. The extracts were more effective to inhibit the activity of alpha-amylase in vitro than acarbose in the same concentration (200 µg/mL). We observed that ethanolic extracts presented more effective HSA inhibition activity than hexanic extracts. This result was also reported in other studies of HSA inhibition activity with plant extracts of different families from the Cerrado biome (de Souza et al. 2012, ^{Silva et al. 2009}).

Preliminary qualitative phytochemical analysis of the active extracts was performed to determine the probable type of compounds present in the extracts causing HSA inhibition. We visualize by TLC the components of only few extracts. Flavonoids were found in the ethanolic extract of the P. torta leaves and hexane extract of the P. ramiflora stem bark. Tannins and saponins were found in the ethanolic extract of the Serjania lethalis stem bark and root bark, and ethanolic extracts of the P. ramiflora root wood. Tannins were found in the ethanolic extracts of the P. ramiflora root bark. As we observed, previous studies have reported that the capacity of alpha-amylase inhibition by vegetal extracts is normally associated with polar compounds, as phenolic, tannins and triterpenoids compounds (^{Ali et al. 2006}, ^{Gad et al. 2006}, ^{Kandra et al. 2004}).

The extracts that showed the highest inhibition in the screening test at a concentration of 200 µg/mL were subsequently investigated at concentrations of 100, 50 and 20 µg/mL. The S. lethalis stem wood ethanolic extract and stem bark ethanolic extract (20 µg/mL) inhibited almost 40% of the HSA activity. The ethanolic extracts (20 µg/mL) of Matayba guianenses and S. lethalis root bark and P. torta leaves inhibited 70-80% of the HSA activity. Ethanolic extract of the P. ramiflora root bark and hexane extract of the P. ramiflora stem bark (20 µg/mL) showed a 95-100% of inhibition effect, respectively. Studies from the bark of P. caimito reveal the presence of the triterpenes (^{Ardon and Nakano 1973}, ^{Pellicciari et al. 1972}). Seven polyphenolic compounds, gallic acid, (+)-gallocatechin, (+)-catechin, (-)-epicatechin, dihydromyricetin, (+)-catechin-3-O-gallate, and myricitrin, were isolated and identified from the fresh fruits of P. campechiana, P. sapota and P. viridis (^{Ma et al. 2004}). These reports indicate that the capacity of alpha-amylase inhibition by genus Pouteria might be associated with such compounds.

We selected the stem bark aqueous extract of P. ramiflora (PrSBAE) for in vivo studies because it inhibited 92% of the HSA activity when it was diluted 1/10. In the acute toxicity test, mice treated with PrSBAE extract in doses higher than 500 mg/kg showed toxicity signals as hypnoses, dyspnea, writhing. Two animals died. In mice that received 5,000 mg/kg of the extract we observed signals such as hypnosis, dyspnea. One animal died. When mice were treated orally and intraperitoneal with 100 mg/kg of the PrSBAE dose, they did not show any toxicity effect.

Weight and blood glucose level in mice treated with 25, 50 and 100 mg/kg of the PrSBAE extract are shown in figures 1 and 2. All administrated doses have shown a significant effect on weight loss when compared to the control group after eight days of treatment. Animals that received 50 and 100 mg/kg of extracts showed a significant reduction of the levels of blood glucose when compared to the control group. Mice treated with acarbose, a competitive inhibitor of intestinal alpha-glucosidases that slows the breakdown of sucrose and starch (^{Santeusanio and Compagnucci 1994}), were not effective to reduce blood glucose level when compared with polyphenolic enriched crude plant extract after 30 min of treatment (^{Mai and Chuyen 2007}). A previous study of our laboratory (^{Deconte et al. 2011}) conducted in rats treated by 20 days with acarbose did not reveal difference of weight gain and glycemic level of non-diabetic and diabetic groups when compared with the respective controls. ^{Heo et al. (2009)} showed that diphlorethohydroxycarmalol, isolated from Ishige okamurae, had inhibitory effects on alpha-glucosidase and alpha-amylase activities higher than those of acarbose. Therefore, our results showed that the P. ramiflora extracts were more effective to inhibit the activity of alpha-amylase in vitro than acarbose. The in vivo results suggest that PrSBAE reduction of glucose level of treated mice may be due to inhibition of carbohydrate-hydrolyzing enzymes.

Figure 1 The effect of administration of stem bark aqueous extract from P. ramiflora (PrSBAE) for eigth days on body weight loss in mice. Each column represents mean±S.E.M for five mice. ^* P < 0.05, ^** P < 0.01, ^*** P < 0.001 different from control group. 

Figure 2 The effect of administration of stem bark aqueous extract from P. ramiflora (PrSBAE) for eigth days on blood glucose level in mice. Each column represents mean±S.E.M for five mice. ^* P < 0.05, ^** P < 0.01, ^*** P < 0.001 different from control group. 

Several plant extracts have high concentrations of tannins, that are compounds that may exert an anti-nutritional effect by interfering with gut function (^{Carbonaro et al. 2001}) and reduce the glycemic response to carbohydrate foods in humans (^{Gin et al. 1999}). In another study, it was investigated whether blood glucose reduction was due to reduced food intake in mice treated with the Syzygium cumini ethanol extract and tannic acid (^{Oliveira et al. 2005}). In the study by Oliveira et al., it was hypothesized that polyphenolic compounds contribute to the reduction of food intake, body weight and the levels of blood glucose. Hence, we cannot exclude the possibility that the P. ramiflora extract may contain phenolic compounds and that its effects on weight reduction and the decrease in blood glucose are related to the dietary restriction associated with the inhibition of alpha-amylase.

In conclusion, the present study showed that M. guianensis, S. lethalis, P. torta and P. ramiflora inhibit alpha-amylase activity in vitro even at low concentrations. In addition, P. ramiflora bears positive effects on body weight and blood glucose levels.