Effects and Toxicity of Eugenia punicifolia. Extracts in Streptozotocin-Diabetic Rats

Abstract

Extracts and decoctions of Eugenia jambolana. Lam., Eugenia uniflora. L., and Eugenia punicifolia. (Humb., Bonpl. & Kunt) DC. are used in traditional medicine to treat diabetes mellitus. Although there have been reports that Eugenia jambolana. and Eugenia uniflora. have antidiabetic effects, no study has yet been made on Eugenia punicifolia.. We investigated the effects of aqueous, butanol, and methanol extracts of Eugenia punicifolia. leaves administered by gavage to streptozotocin-diabetic rats for 26 to 29 days. Body weight, food and fluid intake, urine volume, and urinary glucose and urea were evaluated every 7 days. At the end of the experiment, we measured serum cholesterol, high-density lipoprotein (HDL)-cholesterol, triglycerides and bilirubin, hepatic glycogen and serum marker-enzymes (alanine and aspartate aminotransferases, alkaline phosphatase, γ.-glutamyltransferase, L-lactate dehydrogenase, creatine kinase, α.-amylase, and angiotensin I converting enzyme). We found that in rats treated with the aqueous extracts, food and liquid intake, urinary volume, and body weight were all reduced, while for rats treated with the methanol extract, not only were liquid intake, urinary volume and body weight reduced, but urinary glucose and urea also decreased. Rats treated with the butanol extract showed no significant alterations in any of the parameters measured. Chronic treatment with extracts had no effect on the marker enzymes nor on serum bilirubin levels. The results indicate that aqueous extracts of Eugenia punicifolia. leaves produced an anorexic effect and that methanol extracts had a beneficial effect on the diabetic state by improving carbohydrate and protein metabolism without provoking hepatobiliary, microvascular, muscular, or pancreatic toxic effects.

Keywords:

• Antidiabetic activity
• chronic treatment
• Eugenia punicifolia
• marker enzymes
• toxicity

Introduction

Plants have traditionally been a very important form of treatment for diabetes mellitus in various regions of the world, and this continues due not only to low purchasing power but also to the increasing tendency of individuals to choose natural products. The genus Eugenia. (Myrtaceae) comprises more than 500 species, which occur mainly in tropical and subtropical America and tropical Asia, with a few species occurring in Australia and Africa (Hillebrand, 1888). Members of the genus Eugenia. have been shown to have various medicinal properties including analgesic, anti-inflammatory, and antipyretic (Karla et al., 1994) effects along with antifungal (Rahhal, 1997) and antidiabetic actions. In Brazil, Eugenia jambolana. Lam., Eugenia uniflora. L., and Eugenia punicifolia. (Humb., Bonpl. & Kunt) DC. are among the Eugenia. species more highly used in traditional medicine for treating diabetes mellitus. Several researchers have revealed the antidiabetic properties of Eugenia jambolana. for specimens collected in India (Rathi et al., 2002; Kar et al., 2003; Ravi et al., 2004a2004b) but not for aqueous extracts or decoctions of Eugenia jambolana. leaves and fruits collected in Brazil (Teixeira et al., 19901997, 2000; Pepato et al., 20012005). The lack of antidiabetic effects of Eugenia jambolana. leaves collected in Brazil could be due either to different cultivation conditions or the pharmaceutical form (aqueous extracts or decoctions) used in the Brazilian experiments.

To the best of our knowledge, there have been no previous reports in the literature to evaluate the antidiabetic potential of Eugenia punicifolia.. In the study described in this paper, we used streptozotocin-diabetic rats to determine the effects of chronic treatment with aqueous, butanol, and methanol extracts of Eugenia punicifolia. leaves on metabolic and physiological parameters that are classically altered in diabetes. We also used serum marker-enzymes to assess possible toxic effects of the Eugenia punicifolia. extracts.

Materials and Methods

Plant material

Leaves of Eugenia punicifolia. were collected in northern Brazil during September at the Mini-Campus of the University of the Amazon and identified as Eugenia punicifolia. (Kunth) DC. by Dr. Gilberto Dolejal Zanetti. Authenticated material was deposited in the Herbarium of the Department of Industrial Pharmacy, Federal University of Santa Maria, Rio Grande do Sul, Brazil, under accession no. HDFI 241.

Extraction and fractionation

Extracts were prepared from authenticated plant material at the Chemistry of Natural Products Laboratory, Biotechnology Unit, University of Ribeirão Preto, Ribeirão Preto, Brazil. The aqueous extract was obtained from dried and powdered leaves of Eugenia punicifolia.. The plant material (200 g) was extracted by percolation in boiling water (1.5 l) for 10–15 min, cooled, filtered, and freeze-dried yielding 14 g. Dried and powdered leaves of Eugenia punicifolia. (348.43 g) were macerated in chloroform and methanol sequentially. The extracts were filtered and evaporated under vacuum resulting in the chloroform (10.95 g) and methanol extracts (48.39 g), respectively. All extracts were stored in a desiccator at −20°C. The methanol extract was dissolved in methanol/water (2:8) and partitioned with hexane, followed by dichloromethane, ethyl acetate, and n.-butanol to afford hexane (1.47 g), dichoromethane (0.26 g), ethyl acetate (6.22 g), and n.-butanol (23.76 g) fractions.

Animals and induction of diabetes

Male rats (n = 25 for each treatment) were adapted to metabolic cages for 2 to 3 days and then fasted for 14–16 h, after which they were anesthetized with ethyl ether and injected with 50 mg kg⁻¹ body weight of streptozotocin (STZ) dissolved in 0.01 M citrate buffer (pH 4.5) via the jugular vein. After STZ treatment, all rats were returned to their cages and were given free access to food and water. The average weight of the rats was 144.7 ± 1.37 g for the group treated with aqueous extract, 146.8 ± 3.35 g for the group treated with methanol extract, and 124.95 ± 1.80 g for the group treated with butanol extract.

Regarding feeding and housing conditions, all rats were fed a normal laboratory chow diet containing (w/w) 16% protein, 66% carbohydrate, and 8% fat and were housed under a 12:12 h light/dark cycle at 22–25°C. The experimental protocols were conducted in accordance with internationally accepted principles for laboratory animal use and care as set by the World Health Organization.

Treatments

Three different treatments were applied to three different groups (aqueous-extract, methanol-extract, and butanol-extract) each containing 20 rats. The following procedures were carried out irrespective of the group. Three to 5 days after STZ injection, 20 rats were separated into 10 sets of two rats each matched in terms of body weight, food and water intake, urinary volume, urinary glucose and urea, and plasma glucose (assessed using tail-tip blood) to ensure that each pair of rats had diabetes of a similar level of severity. After matching, the 20 rats were assembled into two initially similar subgroups of 10 rats by randomly allocating one member of each pair to a treated subgroup destined to be treated with extract and the other member of the pair to a control subgroup, which received no extract.

For the aqueous-extract group, treatment started on the seventh day after STZ injection, each rat in the treated subgroup being given 1 ml of a saline solution containing 0.0550 g of aqueous extract by gavage at 8 a.m. The methanol extract group was treated in a similar way except that the saline solution administered to the rats in the treated subgroup contained 0.0273 g of methanol extract. For the butanol extract group, treatment started on the fifth day after STZ injection, each rat in the treated subgroup being given 1 ml of a saline solution containing 0.0160 g of butanol extract by gavage at 8 a.m. The difference between the doses was due solely to the limited availability of the plant. All the rats in the control subgroups (aqueous, methanol, and butanol) received 1 ml of saline only.

Analyses

Every 7 or 8 days (4 sets of measurements in all), for all groups, we measured body weight, daily intake of food and liquid, urine volume, urinary glucose and urea. For the aqueous-extract and methanol-extract groups, we started measuring on the seventh day after STZ injection to the 34th day after STZ injection (29th day of treatment). We did not measure plasma glucose levels each week because daily urinary glucose is a sufficient measure of carbohydrate metabolism. At the end of the experiment, the rats were sacrificed by decapitation and samples of free-running blood collected to determine the levels of serum cholesterol, high-density lipoprotein (HDL)-cholesterol, triglycerides, bilirubins, and marker enzymes for hepatobiliary damage (alanine aminotransferase, EC 2.6.1.2, ALT; aspartate aminotransferase, EC 2.6.1.1, AST; γ.-glutamyltransferase, EC 2.3.2.2, γ.-GT; alkaline phosphatase, EC 3.1.3.1, ALP); and muscle (creatine kinase, EC 2.7.3.2, CK; L-lactate dehydrogenase, EC 1.1.1.27, LD) and pancreatic (α.-amylase, EC 3.2.1.1, AMS) damage and microangiopathy (angiotensin I converting enzyme, EC 3.4.15.1, ACE) (Vranes et al., 1995a1995b; Cooper et al., 1996). Serum and/or urine were stored at −20°C whenever they could not be analyzed on the day of collection. The epididymal and retroperitoneal fat pads and the soleus and extensor digitorum longus (EDL) muscles were dissected out of the sacrificed rats and weighed and the liver removed to measure its glycogen content. For the butanol-extract group, a similar methodology was applied except that the treatment was started on the fifth day after STZ injection and the serum marker-enzymes were determined on the third day after STZ injection (i.e., before the start of treatment with the extract) and the 24th day after STZ injection (19th day of treatment) and on the 31st day after STZ injection (26th day of treatment) when the animals were sacrificed.

Chemical analysis

Plasma glucose, serum cholesterol, HDL-cholesterol, triglycerides, and AST, ALT, ALP, and γ.-GT were determined in a Bayer/Technicon RA-XT autoanalyzer (Dublin, Ireland) using reagents from the Bayer kit. Urinary glucose was measured by the o.-toluidine method of Dubowski (1962) and urea by the urease method (Bolleter et al., 1961; Bergmeyer, 1985). Serum LD (Whitaker, 1969) and AMS (Caraway, 1959) and bilirubin (Malloy & Evelyn, 1937; Sims & Horn, 1958) were assayed colorimetrically and CK (Committee on Enzymes, 1976) and ACE (Groof et al., 1993) activities by kinetic spectrophotometric methods. Hepatic glycogen was extracted with 30% KOH, precipitated with alcohol (Carrol et al., 1956) and the yield determined by the colorimetric anthrone method of Collowick and Kaplan (1957). These enzymic activities and other measurements were carried out using a Hitachi U3000 spectrophotometer (Tokyo, Japan). In all assays, storage times and temperatures were rigorously controlled. All chemical reagents and solvents were of at least analytical grade and obtained from Merck (Clevenot Corp., Darmstadt, Germany), Sigma Chemical (St. Louis, MO, USA) or LabSynth (Diadema, São Paulo, Brazil).

Statistical analysis

Data were evaluated by analysis of variance (ANOVA) using the Newman–Keuls test for subsequent multiple comparisons to analyze physiological parameters and urinary glucose and urea; the protected test to analysis the results for the marker enzymes in terms of type of group (treated or untreated) and treatment time; and the unpaired Student's t.-test to analyze the results for cholesterol, HDL-cholesterol, triglycerides, bilirubin, hepatic glycogen, and weight of tissues. In all cases, the significance level was p < 0.05.

Results

Table 1 shows the results obtained after STZ injection but before treatment with extract; all the physiological and metabolic variables did not differ statistically between the experimental and control groups, and the values for the treated and control subgroups were very close, indicating that the pairing method used was satisfactory.

Table 1 Physiological and metabolic parameters for streptozotocin-diabetic rats before treatment with Eugenia punicifolia. leaf extracts

Group	Body mass (g)	Δw (g)	Food intake (g · 24 h^−1)	Liquid intake (ml · 24 h^−1)	Urinary volume (ml · 24 h^−1)	Urinary urea (mg · 24 h^−1)	Urinary glucose (g · day^−1)	Plasma glucose (mg · dl^−1)
Aqueous extract
Treated subgroup	154.3 ± 6.2	10.8 ± 5.5	18.3 ± 0.7	54.4 ± 6.6	30.3 ± 4.8	639.8 ± 53.4	2.5 ± 0.4	431.7 ± 31.5
Control subgroup	164.7 ± 3.8	19.1 ± 3.4	18.7 ± 0.6	55.5 ± 5.2	29.2 ± 4.9	576.8 ± 62.0	2.6 ± 0.5	430.0 ± 25.6
Methanol extract
Treated subgroup	155.0 ± 2.9	11.4 ± 2.3	21.0 ± 0.5	69.5 ± 5.2	40.9 ± 4.4	814.6 ± 69.5	3.1 ± 0.3	481.6 ± 28.2
Control subgroup	161.2 ± 3.7	13.5 ± 2.8	19.9 ± 1.1	67.5 ± 7.0	41.8 ± 4.6	804.6 ± 77.5	3.1 ± 0.2	424.2 ± 19.5
Butanol extract
Treated subgroup	141.1 ± 3.6	15.6 ± 1.4	18.9 ± 0.8	60.8 ± 4.8	41.2 ± 3.9	788.4 ± 39.8	3.7 ± 0.3	424.4 ± 15.4
Control subgroup	139.2 ± 3.8	15.1 ± 1.5	17.8 ± 0.7	62.0 ± 3.6	41.3 ± 3.1	771.4 ± 43.2	3.5 ± 0.2	436.1 ± 22.1

Data are expressed as means ± SEM. Liquid intake, urinary volume, food intake, urinary glucose and urea are per 100 g of body weight as measured on the fifth day after streptozotocin (STZ) injection for the aqueous extract and methanol extract groups and on the third day after STZ injection for the butanol extract group before starting treatment with the respective Eugenia punicifolia. leaf extracts. The treated subgroup was administered 1 ml of Eugenia punicifolia. leaf extract diluted in saline by gavage while the control subgroup received 1 ml of saline (see text for details). Each group contained 10 rats. Δw = Change in weight relative to the weight on the day of STZ injection.

After 29 days of treatment (four determinations made during the period) with Eugenia punicifolia. aqueous extract, rats in the treated subgroup showed significantly reduced body weight, liquid and food intake, and urinary volume as compared with those in the untreated control subgroup (Table 2). For the group treated with methanol extract, the mean values of all the parameters investigated, except for food intake, were significantly lower for the treated subgroup as compared with the control subgroup (Table 2). At the end of the experiments (26 days for the butanol-extract group, 29 days for the other two groups), there were no significant differences between the treated and control subgroups in regard to the levels of serum lipids or bilirubins or the amount hepatic glycogen (Table 3) or of epididymal or retroperitoneal adipose tissue or soleus or EDL muscle mass (Table 4). The high standard error for the retroperitoneal adipose tissue of the methanol-extract-treated and control subgroups (Table 4) was due to some animals in these groups having no retroperitoneal adipose tissue (i.e., all this tissue was mobilized).

Table 2 Physiological and metabolic parameters for streptozotocin-diabetic rats treated or untreated with Eugenia punicifolia. leaf extracts

Group	Body weight (g)	Δw (g)	Food intake (g · 24 h^−1)	Liquid intake (ml · 24 h^−1)	Urinary volume (ml · 24 h^−1)	Urinary urea (mg · 24 h^−1)	Urinary glucose (g · 24 h^−1)
Aqueous extract
Treated subgroup	183.8 ± 6.2^†	40.3 ± 6.6^†	16.0 ± 0.5^†	48.6 ± 3.4*	28.9 ± 2.7^†	662.0 ± 31.4	2.2 ± 0.2
Control subgroup	215.3 ± 6.6	69.3 ± 7.2	18.1 ± 0.5	61.7 ± 3.7	40.7 ± 3.0	736.3 ± 34.4	2.8 ± 0.2
Methanol extract
Treated subgroup	169.3 ± 3.0^‡	27.5 ± 3.7^‡	19.3 ± 0.7	56.1 ± 4.0^‡	33.4 ± 3.3^‡	704.2 ± 29.5^†	2.9 ± 0.2^†
Control subgroup	200.4 ± 2.7	52.7 ± 3.3	20.4 ± 0.6	78.4 ± 3.6	54.3 ± 3.0	830.6 ± 27.2	3.8 ± 0.2
Butanol extract
Treated subgroup	180.0 ± 6.0	53.0 ± 5.8	18.0 ± 0.5	60.0 ± 3.2	43.0 ± 2.8	768.0 ± 27.0	3.4 ± 0.2
Control subgroup	185.0 ± 5.7	64.0 ± 5.5	17.0 ± 0.5	53.0 ± 3.1	37.0 ± 2.7	737.0 ± 26.0	3.1 ± 0.2

Except for body mass and change in body weight (Δw = change in weight relative to the weight on the day of streptozotocin injection), all values are per 100 g of body weight and represent the means ± SEM of four determinations made during the treatment period (29 days for the aqueous extract and methanol extract groups; 26 days for the butanol extract group). The treated subgroup was administered 1 ml of Eugenia punicifolia. leaf extract diluted in saline by gavage while the control subgroup received 1 ml of saline (see text for details). Each group contained 10 rats (n = 10). Significant at *p < 0.05, ^†p < 0.01, ^‡p < 0.001 compared with the control subgroup.

Table 3 Metabolic parameters of streptozotocin-diabetic rats treated or untreated with Eugenia punicifolia. leaf extracts

Group	Cholesterol (mg · dl^−1)	HDL-cholesterol (mg · dl^−1)	Triglycerides (mg · dl^−1)	Direct bilirubin (mg · dl^−1)	Indirect bilirubin (mg · dl^−1)	Total bilirubin (mg · dl^−1)	Hepatic glycogen (mg %)
Aqueous extract
Treated subgroup	67.3 ± 2.4	24.8 ± 2.1	94.3 ± 9.7	0.05 ± 0.01	0.27 ± 0.02	0.32 ± 0.02	2.3 ± 0.6
Control subgroup	77.0 ± 3.8	29.8 ± 2.0	167.2 ± 43.9	0.05 ± 0.01	0.32 ± 0.03	0.36 ± 0.03	2.1 ± 0.2
Methanol extract
Treated subgroup	71.2 ± 7.3	57.0 ± 5.4	169.0 ± 62.8	0.06 ± 0.02	0.28 ± 0.01	0.34 ± 0.03	2.6 ± 0.6
Control subgroup	76.6 ± 6.0	56.0 ± 5.0	197.0 ± 47.0	0.08 ± 0.02	0.28 ± 0.07	0.36 ± 0.08	2.3 ± 0.2
Butanol extract
Treated subgroup	72.3 ± 3.8	35.7 ± 3.8	229.6 ± 55.6	0.10 ± 0.20	0.45 ± 0.09	0.55 ± 0.08	1.7 ± 0.4
Control subgroup	64.7 ± 2.9	38.1 ± 2.7	123.8 ± 17.0	0.10 ± 0.02	0.37 ± 0.07	0.46 ± 0.06	1.8 ± 0.2

Data are expressed as means ± SEM. As measured for sacrificed rats at the end of the experiments on the 26th day of day of treatment for the butanol extract group and the 29th day of treatment for the aqueous extract or methanol extract groups. The treated subgroup was administered 1 ml of Eugenia punicifolia. leaf extract diluted in saline by gavage (see text for details) while the control subgroup received 1 ml of saline. Each group contained 10 rats. There is no significant difference between the control and treated subgroups analyzed. HDL, high-density lipoprotein.

Table 4 Weight of epididymal and retroperitoneal adipose tissue and soleus and extensor digitorum longus (EDL) muscles of streptozotocin-diabetic rats treated and untreated with Eugenia punicifolia. leaf extracts

	Weight of tissue per 100 g of body mass
Group	Epididymal adipose tissue	Retroperitoneal adipose tissue	Soleus muscle	EDL muscle
Aqueous extract
Treated subgroup	0.3010 ± 0.0886	0.1036 ± 0.0503	0.0847 ± 0.0033	0.0688 ± 0.0059
Control subgroup	0.3509 ± 0.0551	0.1228 ± 0.0356	0.0786 ± 0.0031	0.0717 ± 0.0054
Methanol extract
Treated subgroup	0.2327 ± 0.0612	0.0239 ± 0.0239	0.0425 ± 0.0008	0.0361 ± 0.0015
Control subgroup	0.1794 ± 0.0096	0.0079 ± 0.0079	0.0404 ± 0.0016	0.0368 ± 0.0008
Butanol extract
Treated subgroup	0.1754 ± 0.0717	0.0360 ± 0.0235	0.0450 ± 0.0029	0.0357 ± 0.0043
Control subgroup	0.2646 ± 0.0495	0.0946 ± 0.0363	0.0424 ± 0.0010	0.0396 ± 0.0017

Data are expressed as means ± SEM, as measured for sacrificed rats at the end of the experiments on the 26th day of treatment for the butanol extract group and the 29th day of treatment for the aqueous extract or methanol extract groups. The treated subgroup was administered 1 ml of Eugenia punicifolia. leaf extract diluted in saline by gavage (see text for details) while the control subgroup received 1 ml of saline. Each group contained 10 rats. There is no significant difference between the control and treated subgroups analyzed.

In relation to marker enzymes, the results indicate that for all 3 types of extract there were no differences at any time in the 26–29 days of the experiment between the treated and control subgroups (inter-subgroup) or between different times in any subgroup (intra-subgroup) in levels of marker enzymes for hepatobiliary damage (Table 5) or muscle or pancreatic damage or microangiopathy (Table 6).

Table 5 Serum levels of marker enzymes (AST, ALT, γ.-GT, and ALP) indicating hepatobiliary injury

	Marker enzymes (Units · l^−1)
Group	AST	ALT	γ.-GT	ALP
Before starting treatment (t0)
Butanol extract treated subgroup	222.1 ± 20.8	125.9 ± 9.2	19.1 ± 2.7	1857.0 ± 155.2
Butanol extract control subgroup	205.4 ± 16.9	116.1 ± 5.0	23.4 ± 3.6	1794.0 ± 238.2
After 19 days treatment (tM, halfway through treatment)
Butanol extract treated subgroup	230.0 ± 22.1	188.2 ± 18.9	25.1 ± 4.1	2085.9 ± 179.8
Butanol extract control subgroup	259.9 ± 28.3	179.3 ± 15.6	18.3 ± 1.6	1662.4 ± 246.9
After 26 days of treatment (tF, end of treatment)
Butanol extract treated subgroup	240.2 ± 19.7	240.7 ± 57.0	24.3 ± 3.6	1977.4 ± 280.7
Butanol extract control subgroup	216.6 ± 22.2	185.7 ± 25.6	24.0 ± 3.5	1727.6 ± 271.2
After 29 days of treatment (tF, end of treatment)
Aqueous extract treated subgroup	368.2 ± 87.3	271.3 ± 66.2	13.2 ± 2.7	1782.7 ± 399.4
Aqueous extract control subgroup	383.6 ± 139.5	287.8 ± 92.6	15.2 ± 3.2	1766.6 ± 323.6
Methanol extract treated subgroup	299.5 ± 133.2	229.2 ± 116.3	15.7 ± 4.5	2151.2 ± 636.8
Methanol extract control subgroup	285.6 ± 49.9	217.8 ± 38.0	9.2 ± 0.8	2703.6 ± 411.1

Data are expressed as means±SEM. The treated subgroup was administered 1 ml of Eugenia punicifolia. leaf extract diluted in saline by gavage (see text for details) while the control subgroup received 1 ml of saline. Each group contained 10 rats. Treatment ended (t_F) after 26 days administration of extract for the butanol extract group and after 29 days administration of extract for the aqueous and methanol extract groups. The t₀ and t_M data are for live rats and the t_F data for sacrificed rats. There is no significant difference at any time between control and treated subgroups or between different times for any subgroup. AST, aspartate aminotransferase; ALT, alanine aminotransferase; γ.-GT; γ.-glutamyltransferase; ALP, alkaline phosphatase.

Table 6 Serum levels of marker enzymes indicating muscle (CK and LD) and pancreatic (AMS) damage and microangiopathy (ACE)

	Marker enzymes (Units · l^−1)
Group	CK	LD	AMS	ACE
Before starting treatment (t0)
Butanol extract treated subgroup	583.6 ± 29.2	2882.9 ± 235.7	1219.8 ± 85.7	943.0 ± 137.5
Butanol extract control subgroup	617.0 ± 20.3	3009.8 ± 310.2	1264.1 ± 63.4	1053.0 ± 163.0
After 19 days treatment (tM, halfway through treatment)
Butanol extract treated subgroup	698.2 ± 52.7	2448.7 ± 229.2	1333.1 ± 105.8	902.2 ± 91.0
Butanol extract control subgroup	705.4 ± 101.0	2648.3 ± 327.7	1329.6 ± 111.4	771.0 ± 63.1
After 26 days of treatment (tF, end of treatment)
Butanol extract treated subgroup	707.5 ± 85.8	2200.5 ± 153.1	1381.6 ± 89.6	1024.3 ± 176.8
Butanol extract control subgroup	751.5 ± 130.0	2529.7 ± 121.8	1408.0 ± 91.2	779.9 ± 95.8
After 29 days of treatment (tF, end of treatment)
Aqueous extract treated subgroup	1782.7 ± 399.4	1242.5 ± 126.0	1999.7 ± 930.9	754.4 ± 168.5
Aqueous extract control subgroup	1766.6 ± 323.6	1285.0 ± 123.3	1483.6 ± 208.7	847.0 ± 108.7
Methanol extract treated subgroup	579.7 ± 108.5	1292.5 ± 193.8	694.2 ± 24.7	614.4 ± 122.8
Methanol extract control subgroup	1028.2 ± 158.7	1407.2 ± 228.2	695.4 ± 13.9	760.5 ± 83.5

Data are expressed as means ± SEM. The treated subgroup was administered 1 ml of Eugenia punicifolia. leaf extract diluted in saline by gavage (see text for details) while the control subgroup received 1 ml of saline. Each group contained 10 rats. Treatment ended (t_F) after 26 days administration of extract for the butanol extract group and after 29 days administration for the water and methanol extract groups. The t₀ and t_M data are for live rats and the t_F data for sacrificed rats. There is no significant difference at any time between control and treated subgroups or between different times for any subgroup. CK, creatine kinase; LD, L-lactate dehydrogenase; AMS, α.-amylase; ACE, angiotensin I converting enzyme.

Discussion

When treatment with aqueous, methanol, and butanol extracts was started, both the extract-treated rats and their respective controls showed the same degree of STZ-induced diabetes (Table 1), and their physiological and metabolic parameters were typical of those present in classic diabetes and were similar to those reported by Mori et al. (2003).

Of the three types of Eugenia punicifolia. leaf extracts, only the methanol extract presented antidiabetic activity, as shown by a reduction in physiological and metabolic parameters caused by an improvement in carbohydrate and protein metabolism in the subgroup treated with methanol extract as compared with the untreated matched control (Table 2). The reduction in the level of 24 h urinary glucose seen in the subgroup treated with methanol extract, considered a practical parameter because it refers to the homeostasis of glucose over a day, did not occur because of greater synthesis or less degradation of hepatic glycogen (Table 3). Different results were obtained with Eugenia jambolana. by Grover et al. (2000), who reported that 10 days of oral administration of an aqueous extract of powdered Eugenia jambolana. fruit partially restored altered skeletal glycogen content. In our case, the combination of reduced levels of urinary glucose and urea in the methanol-extract-treated subgroup as compared with the untreated matched control may explain the reduction in liquid intake and urinary volume (Table 2). It is probable that glucose and amino acid use by the rats was not enough to lead to a lower food requirement and a higher weight gain (Table 2). The fact that the food intake of rats in the subgroup treated with methanol extract was not significantly different from that of the rats in the matched control subgroup supports the notion that the reduction in physiological and metabolic parameters seen in these rats were not the result of anorexia. It is possible that a larger dose of methanol extract, more frequent administration of the extract, a different route of administration, better extraction procedures, or a less severe degree of diabetes in the STZ-diabetic rats could increase the antidiabetic effect of the methanol extract, leading to better use of nutrients and lower food consumption and a higher gain in mass.

Although the active ingredients of aqueous or organic solvent extracts of Eugenia punicifolia. are still unknown, the results in Table 2 suggest that aqueous extract contains constituents that have an anorexic effect in that food intake was reduced in those rats in the subgroup treated with aqueous extract. It is possible that the reduction in food intake was responsible for the decreased liquid intake and urinary volume as well as for the lower weight gain (Table 2). When water was used as the extracting solvent, the extract would have contained more polar constituents and it may be these that were responsible for the anorexic effect seen with the aqueous extract because no anorexic effects were seen with the solvent-based extracts that would have extracted nonpolar components of the Eugenia punicifolia. leaves. Krikorian-Manoukian and Ratsimamanga (1967) reported that Eugenia jambolana. also has an anorexic effect. In our research, the butanol extract showed no effect on any of the physiological or metabolic parameters measured (Table 2).

Although there are no previous studies on the physiological or metabolic effects of aqueous or organic solvent extracts of Eugenia punicifolia., there have been studies on other species of this genus that have shown hypoglycemic effects. Aqueous and alcoholic extracts of the seeds and leaves of Eugenia jambolana. administered orally to experimental animals and human adults at variable dose levels were found to have antidiabetic activity, reducing plasma and urinary glucose and polyuria (Anon, 1968; Kohli & Singh, 1985; Achrekar et al., 1991; Grover et al., 20012002; Vikrant et al., 2001; Kar et al., 2003; Sharma et al., 2003; Ravi et al., 2004a) and there have also been reports that Eugenia uniflora. leaf extracts improved hyperglycemia in mice (Mamose et al., 1999).

Results reported by Teixeira et al. (199019972000) and by authors of the current paper (Pepato et al., 2001) using Eugenia jambolana. leaves collected in southern and southeastern Brazil were different from those reported in the current paper for Eugenia punicifolia., which had been collected in northern Brazil, because Eugenia jambolana. extracts did not show any antidiabetic effect. It may be that Eugenia punicifolia. contains more potent antidiabetic constituents than other Eugenia. species and/or methanol extraction may have extracted higher levels of these constituents, more studies being needed to isolate and identify the therapeutically active components of Eugenia punicifolia.. To the best of the authors' knowledge, the only study on the chemical constituents of Eugenia. species was done on Eugenia uniflora., which was found to contain flavonoids, steroids, triterpenoids, tannins, anthraquinones, phenols, cineol, and essential oils (Consolini & Sarubio, 2002; Auricchio & Bacchi, 2003).

Possible mechanisms for the hypoglycemic action of some Eugenia. species have been discussed in the literature and include increased cathepsin B activity, the enzyme responsible for the proteolytic conversion of pro-insulin to insulin (Bansal et al., 1981), reduced insulinase activity (Achrekar et al., 1991) and gastrointestinal transit (Schapoval et al., 1994), and the inhibition of carbohydrate decomposition in the intestine (Arai et al., 1999).

Our results showing that butanol extracts of Eugenia punicifolia. have no reducing effect on hyperglycemia are in partial accord with those of Arai et al. (1999), who reported that few fractions obtained from ethanol extracts of Eugenia uniflora. leaves on the basis of polarity and molecular size had a positive effect in oral glucose tolerance test in mice. Nevertheless, we cannot exclude the possibility that the absence of any effect in these tests was due to the butanol extract being given to the animals in lower doses than the methanol extract, which did reduce hyperglycemia.

In relation to lipid metabolism, the fact that serum cholesterol, HDL-cholesterol, and triglycerides (Table 3) were not significantly altered by treatment with Eugenia punicifolia. extracts agrees with our observation that there was no significant alteration in lipid mass (Table 4). These observations for Eugenia punicifolia. contrast with what has been observed for Eugenia uniflora. by various workers who have reported a dose-dependent inhibitory effect on lipase activity, which was apparently due to the inhibition of the decomposition of fats in the intestine (Aria et al., 1999) and, in mice, improved hypertriglyceridemia (Mamose et al., 1999). It has also been found that ethanol extracts of Eugenia jambolana. seeds resulted in an improved lipid profile and decreased the activity of a key enzyme [3-hydroxy-3-methylglutaryl coenzyme (HMG-CoA) reductase] involved in cholesterol biosynthesis (Sharma et al., 2003).

The reduction in urinary urea (Table 2) is inconsistent with the fact that there was no alteration in muscle mass (Table 4), although it is possible that muscle weight is not a sensitive enough measure to register such changes.

The results for the marker enzymes (Tables 5 and 6) indicate that neither the anorexic properties of the aqueous extract nor the antidiabetic effects of the methanol extract were accompanied by any hepatobiliary, muscular, or pancreatic toxicity, nor by any vascular damage (Tables 3, 5, and 6).

In summary, our work shows that the chronic administration of Eugenia punicifolia. butanol leaf extract to streptozotocin-diabetic rats did not improve the clinical picture of STZ-diabetic rats, whereas administration of the methanol extract had a beneficial effect on the rats by reducing the symptoms of diabetes, while the aqueous extract had an anorexic effect. It appears that the three extracts had no hepatobiliary, muscular, pancreatic, or microvascular toxicity. The lack of effect of the butanol extract should be explored further, at the same dose as the methanol extract. The findings reported here should stimulate future research on the compositions of the various extracts, which could result in the isolation and eventual synthesis of active components for use in the treatment of the diabetes community.

Acknowledgments

The authors are grateful to the Brazilian agencies FUNDUNESP, FUNDAP, and PADC-FCF–Araraquara-UNESP for financial support. We also thank T.N. Nunes and F.A. Iagame for technical help and Dr. G.D. Zanetti (Industrial Pharmacy Department Herbarium, Federal University of Santa Maria, Rio Grande do Sul, Brazil) for identification and authentication of Eugenia punicifolia..