Evaluation of the glycemic effect of Ceratonia siliqua pods (Carob) on a streptozotocin-nicotinamide induced diabetic rat model

Materials and chemicals

Streptozotocin (Merck Millipore, Temecula, CA, USA), nicotinamide, sodium citrate monohydrate, phosphate-buffered saline (PBS) and glucose anhydrous (Sigma-Aldrich, St. Louis, MO, USA). Citric acid (Merck, Kenilworth, NJ, USA), glibenclamide (Sigma-Aldrich, USA), formaldehyde (HmbG chemicals, Hamburg, Germany), one-touch glucometer (Accu-Chek Performa, Roche, Mannheim, Germany). Rat Insulin ELISA Kit (ER1113, Wuhan Fine Biological Technology Co., Ltd., China). 2,4,6-Tri (2- pyridyl)-s-triazine, Folin-Ciocalteu reagent and ferric chloride (Sigma-Aldrich, USA). Cellulose extraction thimble (Tokyo Roshi Kaisha, Ltd, Tokyo, Japan), ferrous sulphate heptahydrate (Essex, UK). Amylase and glucosidase enzymes, 3, 5-dinitrosalicylic acid (DNSA) and 4-Nitrophenyl (3-D-glucopyranoside (_P-NPG) substrate, monobasic potassium phosphate and potassium tartrate tetrahydrate (Sigma-Aldrich, USA), α-acarbose 95% (Acros organic, USA). Potato starch (Sigma-Aldrich, USA).

Collection of plant material

Totally, two kilograms of unripe pods of carob were collected from Bani Yousef region, 70 km from Taiz governorate, Yemen. The pods were collected at 8 am, 15th May 2014. The shoots were wrapped in aluminum foil sheets and put in ice-filled box to isolate it from sunlight and external temperature. The unripe carob samples were transported directly to the Laboratory of Pharmacognosy Department, Faculty of Pharmacy, Sana’a University, Yemen. The plant was verified and authenticated by a taxonomist and given a voucher specimen No. (CSL/2014/8/1), which was deposited at the Laboratory of Pharmacognosy, Faculty of Pharmacy, Sana’a University, Yemen.

Extraction

The unripe carob pods were washed with purified water to eliminate debris and dried with a fan. Then, the pods were chopped to small pieces and left to dry at room temperature (28.0 ± 2 °C) for three weeks. Next, chopped pieces were ground with a grinder to get fine powder at department of Pharmacy, Faculty of Medicine, University of Malaya. One kilogram of carob powder was extracted using Soxhlet method as previously described with some modification (Jadhav et al., 2009). Extraction process was performed by methanol using Soxhlet extractor under 50 °C. Filter paper (Whatman No. 1) was used to filter the liquid extract. Rotavapor evaporator was used to separate the solutes from the solvents at 40°C and the filtrate was allowed to dry in dark place. Finally, the dried filtered extract was kept under −20 °C for later work.

In vivo and in vitro experimental model design of study

Fig. 1 Flow chart of the experimental design

Quantification of total phenolic and flavonoid content

Total phenolic content (TPC) of the methanolic extract of carob was determined using Folin-Ciocalteu assay (Chan et al., 2009). Gallic acid was used as a reference. Totally, 20 µL of the extract and gallic acid (1 mg mL⁻¹) were pipetted into 96-well microplate followed by addition of 50 µL of 10% Folin-Ciocalteu reagent to be incubated for 3.0 min. Then 100 µL of sodium carbonate (10%) was added followed by incubation for one hour. The absorbance was measured at 765 nm spectrophotometrically (Infinite M 200; Tecan, Männedorf, Switzerland). The phenolic content was represented as mg of gallic acid equivalent/g extract.

Total flavonoid content (TFC) was estimated using AlCl₃ method (Nabavi et al., 2008) in which 10 µL of the extract and quercetin were transferred into 96-well plate, followed by addition of 60 µL of DMSO. Then, 10 µL of 10% (w/v) aluminium chloride, 10 µL of 1 M potassium acetate and 120 µL of deionized water were added to the mixture, which was incubated for 30 min at room temperature. Serial dilutions of quercetin (500–31.25 µg mL⁻¹) were prepared to generate the calibration curve. The absorbance was measured at 415 nm spectrophotometrically (Infinite M 200; Tecan, Switzerland). The flavonoid content was presented as mg quercetin equivalent/g extract.

Evaluation of antioxidant properties of carob methanolic extract

Ferric reducing antioxidant power assay (FRAP)

Measuring of ferric reducing antioxidant activity of the methanolic extract was performed according to (Benzie & Strain, 1996). Working FRAP reagent was prepared through mixing 300 mM acetate buffer (pH 3.6), 10 mM 2,4,6-tripyridyl-s-triazine (TPTZ) in 40 mM HCL, and 20 mM ferric chloride (FeCl₃.6H₂O). Calibration curve was plotted using FeSO₄⋅7H₂O serial concentrations between 100–1,000 mg mL⁻¹. Freshly prepared FRAP working reagent was warmed at 37 °C for 5 min, then 30 µL of the extract, Trolox and quercetin (1.0 mg dissolved in 1.0 mL DMSO) were transferred into 96-well microplate and mixed with 300 µL of FRAP reagent. Then, absorbance was measured at 593 nm against the blank (DMSO). The values were represented as µmoles of ferrous sulphate equivalent/mg extract.

In vitro inhibitory effect of carob against α-amylase and α-glucosidase

The α-amylase inhibition capacity of the extract was determined according to Loizzo et al. (2007). In brief, 20 µL of each test sample (1 mg mL⁻¹extract) dissolved in DMSO or standard Acarbose 95% (Acros Organic, Bridgewater, NJ, USA) and 50 µL 2U mL⁻¹ of α-amylase enzyme (Porcine Pancreas Amylase; 2 mg/10 mL; 10 Units/mg; Sigma-Aldrich, USA) dissolved in cold deionized water were added into each well. After incubation for 10 min at 28 °C, 100 µL of starch solution (0.5% w/v (Sigma-Aldrich, USA) prepared by mixing 250 mg potato starch in previously prepared 50 mL phosphate buffer pH 6.9 prepared by mixing 20 mM monobasic sodium phosphate and 6.7 mM of sodium chloride and heated at 60 °C for 15 min) was added to the reacting mixture. Thereafter, the reaction mixture was re-incubated for 10 min at 28 °C and 100 µL of colorimetric DNSA reagent 96 mM (Prepared by mixing 12 g of sodium potassium tartrate tetrahydrate in 8 mL of 2 M NaOH and 0.5 mg 3,5-dinitrosalicylic acid solution) was added with mixing the contents well. The reaction was terminated by incubating the mixture in a water bath 86 °C for 15 min and later was cooled to room temperature. The blank was conducted in a similar method, with the replaced α-amylase enzyme by deionized water. All the reagents involved the α-amylase enzyme with the exception of the test extracts was pointed as a reference sample. The absorbance of reaction was measured by using UV-Visible spectrophotometry (Infinite M 200 Tecan, Switzerland) at 540 nm and the α-amylase inhibitory activity was expressed as percentage inhibition. The assay was carried out in 96-well microplates (Solid clear F-bottom, Greiner Bio One, Austria). The percentage of inhibition was obtained using the following formula: % inhibition = [A reference–(A sample – A blank)/A reference)] ×100.

Regarding α-glucosidase inhibitory activity of carob methanolic extract, it was assessed spectrophotometrically according to Oboh et al. (2012) with few modifications. Briefly, 100 µL phosphate buffer solution of 2.5 U mL⁻¹ α-glucosidase enzyme (Saccharomyces cerevisiae; 2.5 mg 10 mL⁻¹; 10 U mg⁻¹; Sigma-Aldrich. St Louis, USA) and 50 µL of each test sample solution (1 mg mL⁻¹extract) dissolved in 30% DMSO were added in each well and incubated at 25 °C for 10 min. Then, 50 µL of 4-Nitrophenyl (3-D-glucopyranoside (_P-NPG) substrate solution (5 mM of _P-NPG (Sigma-Aldrich, St Louis, USA) 7.53 mg/5mL dissolved in pH 6.9 phosphate buffer which prepared by mixing with 100 mM monobasic potassium phosphate); was added to the reacting mixture and re-incubated for 5 min at 25°C. Thereafter, 80 µL of 0.1M sodium carbonate solution was added to terminate the catalytic reaction. Acarbose 95% (Acros Organic, USA) dissolved in 30% DMSO was used as a positive control. The reference sample contained the same reaction mixture except the same amount of test sample solution and positive control that were replaced by phosphate buffer solution. The blank was included all reagents and test samples with the exception of α-glucosidase enzyme. All measurements were performed in triplicate. The absorbance of the reaction was recorded on UV-Visible spectrophotometry (Infinite M 200; Tecan, Switzerland) at 405 nm. The assay was carried out in 96-well microplates (Solid clear F-bottom, Greiner Bio One, Austria). The inhibitory rate of samples on α-glucosidase enzyme was obtained using the following formula: % inhibition = [A reference – (A sample – A blank)/A reference)] ×100.

In vitro cytotoxicity of the methanolic extract of carob

Cellular viability assay

The inhibitory effect of methanolic extract of carob was determined by MTT assay. The reduction of MTT (3-[4, 5-dimethylthiazol-2-yl]-2, 5-diphenyltetrazolium bromide) by the mitochondrial dehydrogenase in cells and producing purple formazan is the principle of the experiment. In which 5 ×10³ of WRL 68 and RIN-5F per well were seeded in triplicate in 96-well microplates and incubated for 24 h to be attached at 37 °C with 5% CO₂ saturation. After 24 h of incubation, decreasing concentrations of carob methanolic extract (such as 1,000, 500, 250, 125, 60, 30, 15 and 7 µmol) were prepared (0.025% DMSO) and transferred to the cells with incubating for 24, 48,72 h at 37 °C and 5% CO₂. Subsequently, 20 µL of MTT solution (5 mg mL⁻¹) was added to the treated cells in a dark place, covered with aluminum foil and incubated for 4 h. All media was discharged and a total of 100 µL of DMSO was poured into each well until the purple formazan crystals dissolved (Jayash et al., 2016; Jayash et al., 2017). The blank was performed in a similar manner but the test samples were replaced by media (0.025% DMSO). The plates were measured using a microplate reader UV-Visible spectrophotometry (Infinite M 200; Tecan, Switzerland) at absorbance 570 nm. The tests were carried out in 96-well microplates (Solid clear F-bottom; Greiner Bio One, Kremsmuenster, Austria). All experiments were conducted in triplicate to evaluate the IC₅₀ (The concentration of the methanolic carob extract that was able to inhibit 50% of cells viability). The percentage of cell viability was calculated using the following equation: Cell viability% = A treated cells/A blank ×100.

According to Jayash et al. (2017), the cytotoxic effect of each serial dilution of the methanolic extract of carob was evaluated.

Acute oral toxicity of the methanolic extract of carob

Acute oral toxicity was conducted according to the guidelines of OECD (OECD, 2005) and ethical approval (2014-07-01/PHARM/MAQA) that was issued by the Institutional Ethics Committee, University of Malaya, Malaysia. The animals were kept under the standard conditions of housing as mentioned above. Totally, 36 adult SD-rats (18 males and 18 females) were randomly dispersed into normal control (n = 6), low dose methanolic carob extract (CS2000) (n = 6) and high dose methanolic carob extract (CS5000) (n = 6), which received an oral single dose of 5 mL kg⁻¹ normal saline, 2,000 mg kg⁻¹ carob and 5,000 mg kg⁻¹ carob during the 14 days of the acute oral toxicity time course, respectively. Experimental rats were fasted overnight before dosing. Immediately after dosing, rats were observed at half an hour, 2, 4, 24 and 48 h for any toxicological signs. By the 14th day, rats were fasted overnight, anaesthetized, scarified to collect blood samples via cardiac puncture. Blood samples were centrifuged at 1,500 rpm to obtain serum to measure liver function tests, kidney function tests and total protein. as well as liver and kidney were harvested and their sections were processed to be stained with H&E stain for histopathology evaluation.

In vivo glycaemic activity of the methanolic extract of carob

Induction of type 2 diabetic animal model

According to the protocol of (Arya et al., 2012a) , the rats were fasted overnight and injected with a single intraperitoneal dose of 55 mg kg⁻¹ of freshly prepared streptozotocin (STZ) in 0.1 M citrate buffer (pH 4.5). Immediately after 15 mins, rats were injected intraperitoneally with 210 mg kg⁻¹ nicotinamide (NAD) which exert an antioxidant capacity that minimize the cytotoxic actions of STZ and protects pancreatic β-cell against the destructive effect of STZ (Furman, 2015). After 96 h, rats were fasted overnight except free access for tap water to measure fasting blood sugar using Accu-Chek glucometer. The injected rats were monitored for 10 days and those which attained fasting blood sugar between 10-14 mmol L⁻¹ were considered type 2 diabetic.

Dose rationale of carob

In this study, the doses of the methanolic carob extract were selected to be 500 mg kg⁻¹ as a low dose (CS500) and 1,000 mg kg⁻¹ as a high dose (CS1000) according to the pervious study by Rtibi et al. (2015b).

The experimental design

The type2 induced rats were randomly assigned into four groups (n = 6); namely, diabetic untreated control (D₀), diabetic glibenclamide-treated (GD), diabetic low dose carob-treated (CS500) and diabetic high dose carob-treated (CS1000), which received a single oral daily dose of 5 mL kg⁻¹ normal saline, 5 mg kg⁻¹ glibenclamide (Hafizur et al., 2015), 500 mg kg⁻¹ carob and 1,000 mg kg⁻¹ carob for four weeks respectively. An extra untreated non-diabetic control (NC) (n = 6) was added and given oral daily single dose of 5 mL kg⁻¹ of normal saline for four weeks. Blood glucose level was monitored weekly from the tail vein using Accu-Chek glucometer. Body weight was recorded at the day of starting (initial body weight) and cession of treatment (final body weight). By the 28th days of treatment, rats were anaesthetized using ketamine (50 mg kg⁻¹) and xylazine (5.0 mg kg⁻¹). Then, 5 mL of blood samples were collected for blood analysis through cardiac puncture. Blood samples were centrifuged at 2,000 rpm for 10 min and serum samples were separated and stored at −80 for the future use. Subsequently, rats were scarified by increasing anesthetic dose (ketamine 150 mg kg⁻¹& xylazine 15.0 mg kg⁻¹) to obtain pancreas for gross pathology examination.

Oral glucose tolerance test

Oral glucose tolerance test (OGTT) evaluated the glucose challenge ability of each animal one week prior to animal sacrifice (Arya et al., 2012b). After fasting rats overnight, tip tail fasting blood glucose was measured at 0 time. Twenty minutes after rats had received their own treatment, rats were given a single oral dose of glucose solution (3 g/kg), and blood glucose was measured at 30, 60, 90 and 120 min.

Evaluation of insulin resistance

After measuring fasting serum insulin, homeostatic model assessment (HOMA-IR, HOMA-β and HOMA-S) was estimated (Mather, 2009). Serum insulin was quantified by the enzyme-linked immunosorbent assay (ELISA) using an ultrasensitive rat insulin ELISA kit (ER1113, Wuhan Fine Biological Technology Co., Ltd., China) and read at 450 nm with a microplate reader (Hydroflex Elisa; Chemo pharm, Vienna, Austria) according to the manufacturer’s instructions. HOMA-IR, HOMA- β and HOMA-S were estimated in term of fasting insulin (µU mL⁻¹) and fasting blood glucose (mmol L⁻¹) using HOMA 2 calculator software (Diabetes Trial Unit UoO, UK, 2004).

Histopathological examinations

For histological analysis, pancreatic tissues were collected directly after sacrificing animals and fixed with formaldehyde 10% (v/v). Histopathological examinations were accomplished by an experienced pathologist. The pancreases sections were stained with hematoxylin and eosin (H&E). Histological features including morphological configuration, size of islets and number of β-cells were examined and differentiated for each group.

Statistical analysis

The statistical analysis was determined using a statistical software package (SPSS for Windows, version 23, IBM Corporation, Armonk, NY, USA) and analysis of variance (ANOVA ), followed by Tukey’s-SHD; multiple range post-hoc test. The IC₅₀ was calculated from the linear regression analysis. Pearson correlation coefficient analysis was done to evaluate the correlation between phenolic and flavonoid contents against antioxidant activities. Values were considered statistically significant at P < 0.05. Results were expressed as either means ± SD or means ± SE.

Antioxidant activity

Table 1 demonstrated that the phenolic content (y = 0.0075x – 0.0123, R² = 0.9817) was 127.02 ± 7.18 mg GAE g⁻¹, while the flavonoid content (y = 0.0023x – 0.0165, R² = 0.9979) was 49.74 ± 0.88 mg QE g⁻¹, which means that flavonoid content approximately accounted for 39.2% of the total phenolic content. According to IC₅₀ values, the scavenging activity of the methanolic extract against the free radicals of DPPH (y = 13.339 ln (x) + 17.685, R² = 0.9727) was close to that of Trolox, but it was double that of quercetin. Similarly, FRAP value (y = 0.0002x + 0.002, R² = 0.9649) of the methanolic extract was approximately half that of either quercetin or Trolox.

Pearson correlations analysis displayed that the total phenolic content of carob methanolic extract had a strong positive correlation with FARP assay (r = 0.918, P < 0.01) and a strong negative with DPPH radical (r =  − 0.910, P < 0.01).

Likewise, total flavonoid content of carob methanolic extract reflects a strong positive correlation with FRAP assay (r = 0.886, P < 0.01) and a strong negative correlation with DPPH radical (r =  − 0.867, P < 0.01), (Table 2).

Toxicity assessment of the methanolic extract of carob

In vivo acute oral toxicity

Neither morbidity nor mortality in male or female rats was observed. Similarly, no abnormal signs or behavioral changes were observed during the course of acute oral toxicity. Table 3 showed that no significant difference was observed in body weight of male and female rats. Like body weight, absolute liver, kidney, heart and pancreas weights, as well as relative liver, kidney, heart and pancreas weights of male and female rats were non-significantly different. Table 4 showed that only alkaline phosphatase and urea of high dose carob-treated (CS5000) were significantly increased as compared to normal control (NC) (P ≤ 0.05 and P ≤ 0.01, respectively). In female rats, only serum creatinine showed significant increased as compared to normal control (P ≤ 0.01).

Histopathology of acute oral toxicity

The liver histological sections of normal control (G and J), low dose carob-treated CS2000 (H and K) and high dose carob-treated CS5000 (I and L) of both male and female were of normal architecture of hepatocytes at the level of their cell membrane, nuclei or even cytoplasm. In addition, no congestion in portal veins, no necrotic lesions, no inflammatory signs or fatty infiltrations could be observed. Similarly, kidney histological sections of normal control (A and D), CS2000 (B and E) and CS5000 (C and F) of both male and female were of normal architecture of tubules, Bowman’s capsule, Malpighian corpuscles (Fig. 4).

Glycemic effect of the methanolic extract of carob

In vitro glycemic effect

The IC₅₀ of carob methanolic extract (92.99 ± 0.22 µg mL⁻¹) was higher than that of α-acarbose (23.33 ± 0.73μg mL⁻¹) against α-amylase (Fig. 5). Similarly, the IC₅₀ of the methanolic extract (97.13 ± 4.11μg mL⁻¹) against α-glucosidase was higher than that of α-acarbose (27.05 ± 0.99μg mL⁻¹) (Fig. 6).

In vivo glycemic effect

Weekly fasting blood sugar of STZ-NAD diabetic animal model

Figure 7 showed that normal control (NC) showed a higher significant fasting blood glucose (FBG) than untreated type 2 diabetic rats (D0) at the baseline as well as the end of the 1st, 2nd, 3rd and 4th week of treatment (P ≤ 0.001, P ≤ 0.001, P ≤ 0.001, P ≤ 0.001 and P ≤ 0.001, respectively). In addition, FBG of glibenclamide-treated type 2 diabetic (GD) was significantly lower than untreated type 2 diabetic rats at the end of the 2nd, 3rd, and 4th weeks (P ≤ 0.001, P ≤ 0.001 and P ≤ 0.001, respectively). Moreover, low dose carob-treated (CS500) showed a lower significant FBG than untreated type 2 diabetic rats only at the end of 4th week (P ≤ 0.001) while high dose carob-treated (CS1000) exhibited a lower significant FBG than untreated type 2 diabetic rats at the end of 3rd and 4th weeks (P ≤ 0.01 and P ≤ 0.001, respectively).

Oral glucose tolerance test (OGTT)

Figure 8 illustrates that blood glucose of untreated type 2 diabetic (D0) was significantly higher than that of normal control (NC) at zero (P ≤ 0.01), 30 (P ≤ 0.001), 60 (P ≤ 0.001), 90 (P ≤ 0.001) and 120 (P ≤ 0.001). Conversely, blood glucose of glibenclamide-treated type 2 diabetic (GD) was significantly lower than that of untreated type 2 diabetic at 30 (P ≤ 0.05), 60 (P ≤ 0.001), 90 (P ≤ 0.001) and 120 (P ≤ 0.001). However, blood glucose of low dose carob-treated (CS500) and high dose carob-treated (CS1000) was significantly lower than that of untreated type 2 diabetic at 90 (P ≤ 0.01 and P ≤ 0.001, respectively) and 120 min (P ≤ 0.001 and P ≤ 0.001, respectively).

Biochemistry and hemostatic assessment index of STZ-NAD diabetic animal model

Table 6 showed that total protein of glibenclamide-treated type 2 diabetic (GD) and high dose carob-treated (CS1000) was significantly higher than that of untreated type 2 diabetic (D0) (P ≤ 0.01 and P ≤ 0.01, respectively). On the other hand, serum amylase of untreated type 2 diabetic was significantly lower than that of normal control (NC) (P ≤ 0.001), glibenclamide-treated type 2 diabetic ( P ≤ 0.001), low dose carob-treated (CS500) (P ≤ 0.001) and high dose carob-treated (P ≤ 0.001). Serum FBG of untreated type 2 diabetic was higher than that of normal control (P ≤ 0.001), while serum FBG of glibenclamide-treated type 2 diabetic and high dose carob-treated was significantly lower than that of untreated type 2 diabetic (P ≤ 0.001 and P ≤ 0.001, respectively). However, only insulin of untreated type 2 diabetic was less than that of normal control (P ≤ 0.001). In addition, HOMA-B of untreated type 2 diabetic was significantly less than that of normal control (P ≤ 0.001), glibenclamide-treated type 2 diabetic (P ≤ 0.001), low dose carob-treated (P ≤ 0.05) and high dose carob-treated (P ≤ 0.01), while HOMA-S and HOMA-IR showed a non-significant difference.

Histopathology of STZ-NAD diabetic animal model

Figure 9 showed that histological examination of pancreatic tissues exhibited that normoglycemic untreated rats (A) showed a normal tissue morphological configuration and a larger islet (indicated by red arrow) with high number of β-cells (indicated by yellow arrow), while the untreated type 2 diabetic (B) had the smallest islet and structural malformation with lowest number of β-cells. The high dose carob-treated (E), low dose carob-treated (D) and glibenclamide-treated type 2 diabetic (C) groups had comparatively larger islets with some abnormal structure and higher numbers of β-cells were observed as compared to the untreated type 2 diabetic rats (B).