Aloe vera skin exhibited the highest concentrations of polyphenols and flavonoids. The lowest polyphenol concentration was observed in the Gel-C extract, whereas the Gel-Et extract exhibited the lowest flavonoid content (Table 1). Among the different Aloe vera extracts tested, the lowest scavenging activity was observed for the Gel-Et extract, whereas the highest activity was observed for the Skin-Et extract. The highest FRAP value was also measured for the Skin-Et extract, whereas the Gel-Et extract was observed to have the lowest FRAP value (Table 1).

There was no significant difference in the fasting serum glucose (FSG) levels of the rats between the five experimental groups at baseline (Figure 1). After the oral administration of the five different treatments over 28 days, the FSG levels decreased in the rats of all of the groups except those of the WC group. However, this decrease was not statistically significant. Type 2 diabetic rats that were treated with the ethanolic extract of Aloe vera gel exhibited a significant (p = 0.047) decrease in their FSG levels on days 14 and 28 of the experiment compared with the baseline values. As expected, glibenclamide also resulted in significant (p = 0.026) amelioration of the FSG levels on day 28. However, on day 28, the FSG levels increased significantly (p = 0.022) at the conclusion of the treatment.

There was a 20% increase in the serum insulin levels on day 28 compared with the baseline values for the rats that received Gel-Et. The serum insulin levels remained almost unchanged on day 28 for the glibenclamide-treated group compared with the baseline values. However, for the other three groups, a reduction was observed in the serum insulin levels. No change was observed in the hepatic glycogen contents among the tested groups on day 28.

On day 28, the rats in the Gel-C, Gel-Et and Skin-Et groups exhibited decreases in their serum cholesterol levels by 11%, 17% and 25%, respectively (Figure 2), whereas the rats that received glibenclamide exhibited only a 1% decrease. In contrast, the rats in the WC group exhibited an increase of 15% in their total serum cholesterol levels. In the case of serum triglycerides, type 2 model rats from the WC and Skin-Et groups exhibited increased levels (by approximately 4% and 15%, respectively). The rest of the groups experienced a decrease in triglycerides (10% for the glibenclamide-treated rats, 5% for the Gel-C-treated rats and 12% for the Gel-Et-treated rats).

The glibenclamide-treated rats exhibited decreases in their serum HDL and LDL levels by 3% and 9%, respectively (Figure 3). The HDL cholesterol levels increased by 14%, 5% and 5% in the Gel-C, Gel-Et and Skin-Et groups, respectively. A reduction of approximately 45%, 3% and 69% in the LDL levels were observed when the rats received Gel-C, Gel-Et and Skin-Et treatments, respectively, indicating that the Aloe vera preparations are able to reduce atherogenic LDL cholesterol levels. Furthermore, the protective effects of Aloe vera were more pronounced on the atherogenic LDL cholesterol levels than on the HDL cholesterol levels in the type 2 diabetes model rats.

On day 28, the erythrocyte MDA levels were lower in the rats of the glibenclamide, Gel-C and Gel-Et groups compared with those of the rats in the WC group (Figure 4). However, the rats treated with Skin-Et exhibited higher MDA levels. In terms of the reduced GSH levels, the observed levels were all lower than that of the control group.

Several significant correlations were observed between the biochemical and antioxidant parameters obtained in our study. Strong correlations were found among the levels of polyphenols, flavonoids, DPPH and FRAP, indicating that these parameters are good indicators of the antioxidant activities of Aloe vera (Table 2). The correlations between the phenolic compounds and DPPH and FRAP were r = 0.986 and 0.962, respectively. The most significant correlation was observed between the flavonoid content and the DPPH values, for which r = 0.997. Moreover, the in vivo experimental antioxidant parameter MDA is also strongly correlated with the in vitro antioxidant parameters (levels of flavonoids and polyphenols) and biochemical assay results (DPPH and FRAP), indicating the antioxidant potency of the investigated Aloe vera.

In June 2009, Aloe vera leaves (20 kg) were collected from commercially available sources in Dhaka city, Bangladesh. The botanical name of the investigated Aloe vera species is Aloe barbadensis Miller. The fleshy leaves were washed several times with distilled water and dried at room temperature so the outer skin of the Aloe vera was devoid of water droplets. The stiff and spiny edges from both sides of the Aloe vera skin were then removed using knives. The green outer skin and the gel (i.e., the pulpy portion inside the leaves) were carefully separated and the respective portions were individually weighed (Scientech SA 210, Boulder, CO, USA) before being chopped and blended using a commercially available blender.

Preparation of concentrated Aloe vera gel: After blending, the Aloe vera gel was concentrated under a reduced pressure using a rotary evaporator, freeze-dried (at −55 °C) and transferred to a reagent bottle to be stored in a freezer (at −8 °C).

Preparation of ethanol (80%) Aloe vera gel and skin: The blended Aloe vera gel and skin were extracted using ethanol (80%) as a solvent. Following the extraction, both the gel and the outer green skin extract were concentrated under reduced pressure using a rotary evaporator. The semi-dried ethanolic extract was then freeze-dried (at −55 °C) and transferred to a reagent bottle to be stored in a freezer (at −8 °C).

For the in vitro experiments, the Gel-C (10 g), Gel-Et (10 g) and Skin-et samples were dissolved in 80% ethanol at a 1:1 (w/v) ratio. The mixture was sonicated for 10 min and shaken using a Bigger Bill-120 Shaker (Thermolyne, Dubuque, IA, USA) for 1 h. The solvent was collected after centrifugation at 3,000 × g for 10 min. This process was repeated twice to ensure the total extraction of all compounds from the lyophilized extracts. The supernatants were pooled and evaporated to dryness under reduced pressure using a rotary evaporator. The residue was subsequently stored in dry sterilized containers at −20 °C until further use.

1,1-Diphenyl-2-picrylhydrazyl (DPPH), 2,4,6-tripyridyl-s-triazine (TPTZ), gallic acid and Folin and Ciocalteu’s phenol reagent were purchased from Sigma-Aldrich (St. Louis, MO, USA). Sodium carbonate (Na₂CO₃), aluminum chloride (AlCl₃), sodium nitrite (NaNO₂), ferrous sulfate heptahydrate (FeSO₄·7H₂O), ferric chloride hexahydrate (FeCl₃·6H₂O) and sodium hydroxide (NaOH) were purchased from Merck (Darmstadt, Germany).

Adult Long-Evans female rats (180–220 g) were used in this study. The animals were bred at the BIRDEM animal house in Dhaka, Bangladesh, at a constant room temperature of 22 ± 5 °C at a humidity between 40% and 70%. The animals received a natural 12 h day-night cycle. The rats were provided with a standard laboratory pellet diet and water ad libitum. Type 2 diabetes was induced by a single intraperitoneal injection of STZ in citrate buffer (10 mL) at a dose of 90 mg/kg to the rat pups (48 hours old, average weight 7 g) as described by Bonner-Weir et al. [49]. The experiments were performed three months after the STZ injection. The type 2 diabetic model rats were selected for the experiment after conducting an oral glucose tolerance test (OGTT) and only the diabetic model rats with blood glucose levels of 8–12 mmol/L under fasting conditions were selected for the experiments. Thirty-four female rats were divided into the following five groups: group 1 (n = 7), water-treated control rats; group 2 (n = 7), glibenclamide-treated rats; group 3 (n = 6), Gel-Et-treated rats; group 4 (n = 7), Gel-C-treated rats; and group 5 (n = 7), Skin-Et-treated rats. All treatments were given for 28 consecutive days.

To evaluate the anti-diabetic activity of each extract, the extracts were administered orally for 28 days at a dose of 1.25 g/kg. Only a single daily dose was administered to the rats. For all of the pharmacological studies, the drug glibenclamide was administered orally to the rats at a dose of 5 mg/10 mL (9.9 mL H₂O + 0.1 mL Twin 20) per kg of Type 2 model rat [50]. The animals in the control groups received 10 mL water per kg of body weight.

Blood samples (1 mL) were collected by cutting the tail tips of the rats at baseline as well as on day 14. At the end of day 28, the rats were sacrificed and blood samples (5 mL) were collected again to measure several different parameters. These parameters included the fasting blood glucose levels at baseline and on day 14 and the glucose, insulin, creatinine, triglycerides (TG), total cholesterol (TC), low-density lipoprotein (LDL), high-density lipoprotein (HDL), malondialdehyde (MDA) and reduced glutathione (GSH) levels at baseline and on day 28. The glucose, insulin, creatinine, TG, TC, LDL and HDL levels were measured using the serum, whereas the MDA and GSH levels were measured from the RBC homolysate. The body weights of the individual rats were also measured in seven-day intervals.

The concentrations of phenolic compounds from the Aloe vera samples were estimated using a modified spectrophotometric Folin-Ciocalteu method [51]. Briefly, the different extracts of Aloe vera (Gel-C, Gel-Et and Skin-Et) (1 mL) were mixed with Folin and Ciocalteu’s phenol reagent (1 mL). After 3 min, a 10% Na₂CO₃ solution (1 mL) was added to the mixture which was diluted to 10 mL with distilled water. The reaction remained in the dark for 90 min, and then the absorbance was measured at 725 nm using a T 80 UV/VIS spectrophotometer (ChromoTek GmbH, Planegg-Martinsried, Germany). Gallic acid was used to calculate the standard curve (20, 40, 60, 80 and 100 µg/mL, r² = 0.996). The concentration of phenolic compounds was measured in triplicate. The results are reported as the mean ± standard deviation and expressed as milligrams of gallic acid equivalents (GAEs) per kg of Aloe vera extracts.

The total flavonoid content (TF) of each Aloe vera sample was determined using a previously developed colorimetric assay [52]. The Aloe vera extract (1 mL) was mixed with distilled water (4 mL). At baseline, NaNO₂ (0.3 mL, 5% w/v) was added. After 5 min, AlCl₃ (0.3 mL, 10% w/v) was added and this step was followed by the addition of NaOH (2 mL, 1 M) 6 min later. Immediately, the volume was increased to 10 mL by the addition of distilled water (2.4 mL). The mixture was vigorously shaken to ensure adequate mixing and the absorbance was measured at 510 nm. A calibration curve was prepared using a standard solution of catechin (20, 40, 60, 80 and 100 µg/mL, r² = 0.997). The results were also expressed as mg of catechin equivalents (CEQ) per kg of honey.

The antioxidant properties of each Aloe vera extract were also studied by evaluating the free radical-scavenging activity of the DPPH radical. This activity was measured using a previously reported method [53]. Briefly, each Aloe vera extract (1 mL) was mixed with methanolic solution containing DPPH radicals (2.7 mL, 0.024 mg/mL). The mixture was vigorously shaken and left to stand for 60 min in the dark until the absorbance of the solution remained unchanged. The degree of reduction of the DPPH radical was determined by measuring the absorbance of the mixture at 517 nm [54]. The radical-scavenging activity (RSA) was calculated as the percentage of DPPH discoloration using the following equation: % RSA = [(A_DPPH−A_S)/A_DPPH] × 100 where A_S is the absorbance of the solution when the sample extract has been added at a particular level and A_DPPH is the absorbance of the DPPH solution.

The ferric reducing antioxidant activity was assayed using a previously described method [55]. The absorbance of the Aloe vera was measured at 593 nm using a T80 UV/VIS spectrophotometer (ChromoTek GmbH, Planegg-Martinsried, Germany). These determinations were performed in quadruplicate and the results are expressed as FRAP values [µM Fe(II) in the Aloe vera solution (10%)].

The glucose oxidase (GOD-PAP) test was performed using the method published by Trinder [56].

The serum insulin levels were estimated using the rat insulin ELISA (enzyme-linked immunosorbant assay) kit (Crystal Chem Inc., Downers Grove, IL, USA) according to the method described by Kratzsch et al. [57].

The glycogen levels in the rat liver were estimated using the anthrone-sulfuric acid method [58].

The serum cholesterol levels of the rats were determined after enzymatic hydrolysis and oxidation. The indicator quinoneimine was prepared from hydrogen peroxide and 4-aminoantipyrine in the presence of phenol and peroxidase [59].

The serum triglyceride levels of the rats were measured using an enzymatic colorimetric glycerol-3-phosphate oxidase phenol aminophenazone (GPO-PAP) method using an Automatic Analyzer (Hitachi 704, Hitachi Ltd., Tokyo, Japan) and reagents from Randox Laboratories Ltd. (Crumlin, UK) [60].

The data from our experiments were analyzed using the Statistical Package for Social Science (SPSS) Software for Windows version 12 (SPSS Inc., Chicago, IL, USA). All of the data were expressed as the mean ± SD or median (range) whenever appropriate. The statistical analysis of the results was performed using Student’s t-test (paired and unpaired), the Mann Whitney U test and an ANOVA (analysis of variance) followed by the Bonferroni and Dunnett post hoc tests. The limit of significance was set at p < 0.05.