Male Wistar rats each weighing between 80 and 100 g were acquired from the Veterinary Medicine Animal Holding, Department of Veterinary, University of Ibadan, Nigeria. They were conditioned for 2 weeks in Department of Biochemistry Animal Care Unit in the same institution. Water and rat chow were given ad libitum to the animals in a conducive environmental situations of temperature and 12-hour bright/gloom phase.

Srinivasan et al. (2005) showed that T2DM could be induced in rat model by placing them on high fat diet (HFD) for 2 weeks and thereafter administered STZ. The HFD composition was presented in Table 1. Fifty rats in this study were exposed to HFD for 28 days after which a single dose STZ (35 mg/kg) dissolved in cold citrate (0.1M) buffer adjusted to pH 4.5, was administered intraperitoneally to induce diabetes mellitus (Detaille et al., 2005). After 3 days the rats underwent an overnight fast and their blood glucose was determined using glucometer (On-Call Plus^R). Animals having blood glucose status >250 mg/dL were confirmed diabetic and ascertained suitable for further experiment. Thirty-one rats were discovered to have blood glucose level of 250 mg/dL and above. The success rate was 60%. However, ten out of the thirty-one diabetic rats died before the commencement of the treatment, with only twenty-one diabetic rats left for the experiment.

The normal control (fed normally with rat chow) and diabetic rats were grouped (n = 7) and treated orally once daily as follows:

Group1: Normal Control (NC) (Received corn oil)

Approval for this research was obtained from Animal Care and Use Research Ethics Committee with reference number UI-ACUREC/19/0065.

The rats were sacrificed by cervical dislocation, the blood sample was collected into anticoagulant-free tube and centrifuged at 3,500 rpm for 5 minutes to obtain the serum. The abdominal cavity was opened up and the pancreas was excised into 10% formalin for histological studies.

Reduced ionic strength mitochondria were separated using method designed by Younes and Schneider (1984), and Johnson and Lardy (1967). The rats were sacrificed by cervical dislocation and the abdominal cavity was opened up and liver samples were removed into ice-cold beaker. The blood stain was rinsed from the liver with isolation buffer that contains 0.21M mannitol, 0.079M sucrose, 0.005M HEPES-KOH and 0.001M EGTA (pH 7.4, Sigma). The samples were weighed and chopped into small pieces using a pair of scissors. This was then homogenized using Teflon and homogenizer (DELFLEX^R) in a 10% suspension in isolation buffer. The entire process was carried out at 4°C to ensure viability of the mitochondrial membrane. The homogenate was subjected to five differential centrifugation steps in a cold centrifuge MSE. The first two-5 minutes each was used to separate the nuclear debris as the pellet at 2,300 rpm. The supernatant was then discarded, and the pellet containing the mitochondria was resuspended in a wash buffer (0.21M mannitol, 0.079M sucrose, 0.005M HEPES-KOH, pH 7.4, and 0.5% BSA, Sigma). The re-suspended mitochondria were centrifuged twice for 10 min at 12,000 rpm to wash away any artifacts. The pellet was re-suspended in suspension buffer (0.21M mannitol, 0.079M sucrose, 0.005M HEPES-KOH, pH 7.4, Sigma) and kept on ice for immediate use.

Mitochondrial protein concentration was determined according to the method of Lowry et al (1951) using Bovine Serum Albumin as standard. Mitochondria (10 µL) were suspended in 990 µL of distil water in test tubes in triplicate. Then, 3 mL mixture of 100:1:1 of 2 g Na₂CO₃, 0.1M NaOH and 1% CuSO₄.5H₂O respectively, was added to the protein suspension, thoroughly mixed and left standing for 10 min. Thereafter, 300 µL of 2N Folin-Ciocalteau diluted in four-fold was added into the mixture, followed by energetic shaking and incubation for 30 min. At the expiration of the time, absorbance was determined spectrophotometrically at 750 nm.

The mitochondria were first investigated to determine their suitability for this experiment. Isolated mitochondria protein (0.4 mg/mL) from normal control group were pre-incubated with 8 µM rotenone for 3.5 min. Subsequently, 5 mM succinate was added to energize the reaction and change in absorbance was read using UV-752 spectrophotometer at 540 nm for 12 min at 30 s interval. Similarly, assessment of the inductive effect of calcium was carried out as follows: mitochondria of the same protein concentration (0.4 mg/mL) were pre-incubated with 8 µM rotenone for 3 min, followed by the addition of exogenous 3 µM calcium. Thirty seconds later, 5 µM succinate was added and the absorbance was measured. The reversal of the calcium-induced opening was assessed by pre-incubating the same mitochondria protein with 8 µM rotenone and 4 mM spermine. Exogenous calcium was then added immediately after 3 min pre-incubation; succinate was added after 30 s and absorbance was read. Corresponding mitochondria protein from the treatment groups were investigated for permeability transition under similar condition without addition of exogenous calcium.

Isotonic solution (0.25M sucrose) was used to isolate mitochondria of viable integrity from rat liver in this experiment. The isolation followed the same process with that of mPT (Section 2.6) except for the buffer employed.

Mitochondrial ATPase activity was determined as described by Lardy and Wellman (1953) with minor modification. Each test tube (in triplicate) contained 25 mM sucrose, 65 mM Tris-HCl (pH 7.4) and 0.5 mM KCl in 1 mL final reaction volume. The ATP (1 mM) was added to the set of tubes labelled ATP only, uncoupler, zero time, test groups and control with the exception of the tube labelled mitochondria only. These were incubated at 27 °C in a shaking water bath. Mitochondria (0.4 mg/mL protein) from the test groups were dispensed into labelled tubes other than mitochondria only, uncoupler and zero time tubes which contained mitochondria isolated from the normal control group. While uncoupler (25 μM, 2, 4-dinitrophenol) was added to the uncoupler tubes instantly after mitochondria were added, 1 mL of 10% sodium dodecylsulphate (SDS) was immediately added to the zero-time test tubes. The suspensions were incubated for 30 min and the reaction was terminated by adding 1 mL SDS except in the zero time test tube which had been stopped before. One ml suspension was withdrawn from each test tube and diluted with 4 mL distilled water. Thereafter, 1 mL of 1.25% ammonium molybdate (prepared in 6.5% H₂SO₄) and 1 mL of 9% ascorbic acid newly prepared were added successively and the absorbance was read at λ_{660 nm} in a UV-752 spectrophotometer. Ammonium molybdate (1 mM) was also treated as the sample and used as the standard from which the absorbance of the unknown could be extrapolated. Inorganic phosphate released was quantified using a phosphate standard curve.

Liver samples (0.25 g) from each rat were weighed and homogenized with 5 mL TET (5 mM Tris-hydroxymethyl-aminomethane, 20 mM EDTA and 2 mL of Triton X-100, adjusted to pH 8.0, Sigma) buffer and spinned using ultracentrifuge at 27,000 rpm for one-third hour. Supernatant was removed and the pellet was reconstituted to concentration of 1:1 (v/v) using TE (5 mM Tris-HCl and 20 mM EDTA pH 8.0) buffer. Exactly 0.5 mL of the supernatant and reconstituted pellet were withdrawn and 1.5 mL of 9 mM diphenyl amine solution (prepared in amber bottle because of photosensitivity) was added to each mixture. The mixture in each test tube was allowed to incubate at physiologic temperature for 16 h to 1 day for colour development to be observed and absorbance taken at 620 nm via UV-752 spectrophotometer. Percentage    DNA  fragmentation = A A + B X   100 Where A is the absorbance of supernatant; B is the absorbance of the pellet.

A modified TBARS (thiobarbituric acid reactive substances) method was used for determining the level of peroxidation in mitochondrial membrane lipids (Ruberto et al., 2000). For this assay, exactly 2 mL of a 10% mitochondrial suspension was dispensed into a test tube and diluted with distilled water to a final volume of 4 mL. The mixture was allowed to stand at room temperature for 30 min. Subsequently, 6 mL of 20% acetic acid and 6,000 µL of 0.75% TBA in 1.1% SDS were added to the tube. The resulting solution was then subjected to steam heating for 60 min. After cooling, 5 mL of butanol was added to the solution, leading to the formation of organic and aqueous layers. The malondialdehyde (MDA), a product of lipid peroxidation, was selectively extracted into the butanol phase. The mixture was centrifuged at 3,000 rpm, resulting in a clear demarcation between the butanol and aqueous phases.

The butanol phase, containing the extracted MDA, was measured spectrophotometrically at a wavelength of 532 nm to quantify the level of peroxidation in the mitochondrial membrane.

To further purify the chloroform fraction of O. gratissimum leaf extract, a graded solvent system was employed using vacuum liquid chromatography, resulting in the isolation of the methanol/chloroform subfraction (1:1; v/v). Preparative thin layer chromatography plates were utilized for the analysis. Various solvent systems were employed to facilitate the elution of the samples on the plate.

To identify specific phytochemicals, chromogenic agents were applied to the separated metabolites on the plate, which was subsequently visualized under UV light (both at 254 nm and 366 nm). The samples dissolved in suitable solvents, were carefully spotted on the plate using capillary tubes. After drying, the plate was gently positioned in a chromatographic jar containing the appropriate solvent system. The jar was sealed, allowing the samples to migrate with the mobile phase. The experiment was concluded when the solvent front reached a point about 1 cm before reaching the top of the plate. The plate was then removed, air-dried, and examined under the UV light. Later, the spots on the plate were separately scraped, dissolved in the appropriate solvent and spun in the centrifuge to sediment the gel. Dissolved isolated metabolites were aspirated from the centrifuge tubes and concentrated to dryness. The samples were subjected to analysis using a nuclear magnetic resonance system (Bruker Avance^III 400 MH₃).

Calibrators and samples (25 µL each) were added into the right micro-wells along with 100 μL enzyme conjugate 1X deposited in each well and shaken in a shaker at normal environmental temperature for 2 h to allow for incubation to occur. It was washed repeatedly six rounds with washing buffer 1X solution and the left-over reaction content was disposed off by inversion of the microplate. Afterwards, wash solution (WS) (350 µL) was introduced and later removed using absorbent material. The removal of WS was performed in five successive intervals. Incubation at room temperature was allowed to proceed for 15 min after addition of substrate, TMB (200 µL) to the wells. The reaction was thereafter halted with “stop solution” (50 µL) and thoroughly mixed using shaker for just 5 s. Spectrometry method via microplate reader was then employed to take the optical density reading at λ450 nm not later than duration of half-an hour.

These were calculated using the relationship between serum insulin and blood glucose level (Matthews et al., 1985). HOMA − IR = Insulin µ U / ml x   Glucose mg / dl 405 HOMA − β = 360   x   Insulin µ U / ml Glucose   mg / dl   − 63

Wax was removed using xylene for about 15 min and passed through absolute, 95% and 70% alcohol successively. This would be followed by rinsing the section in water and staining with Harris hematoxylin for 300s. Before differentiating quickly in 1% acid alcohol, it was dipped in water again. After this, it was put under running tap for 10 min and counterstained with 1% Eosin stain for 180s, then rinsed with water. It was dried in rising grades of alcohol and cleared in xylene subsequent to mounting in DPX (Avwioro, 2010).

Statistical analysis was carried out using Graph pad prism (version 8.0) for one-way ANOVA and Turkey’s multiple comparison test was used to compare the mean among the groups. Level of significance was set at p < 0.05. All the results were expressed as mean ± standard deviation (SD). Representative profiles of absorbance of mitochondria were used for the mitochondrial permeability transition pore opening assays. Each assay was repeated three times (in each case) and representative of similar kinetic assay for the mPT in each group was used.

Homeostasis model assessment is typically used in type 2 diabetes mellitus as index to measure the sensitivity of the insulin to its receptor and as well as viability and availability of beta cells that are responsible for synthesizing insulin. Figure 2 illustrates the influence of CF of O. gratissimum (L) leaf on insulin resistance status in STZ-induced diabetic rats. There was significant elevation of this parameter in the diabetic control relative to normal control. Conversely, no significant difference exists between the remaining treated groups compared to normal control.

A highly significant reduction of β-cell function was observed in the untreated group in comparison with normal control. However, there was drastic increase in beta cell status in the CFOG and glibenclamide treated groups compared to the untreated diabetic control (Figure 3).

The status of mPT pore of the normal rats is crucial to knowing whether the pore opening in diabetic untreated rats was actually due to the administered diabetogenic agent (STZ) or not. Therefore, it is necessary to investigate the mitochondria intactness of normal control rats (non-diabetic). Figure 4 presents the evaluation of calcium induction and spermine inhibition of mPT in normal control rats. The results showed that large amplitude swelling was observed when intact mitochondria were challenged with exogenous calcium (6.3 folds), the triggering agent. However, spermine (standard inhibitor) subsequently reversed calcium-induced opening by 82%.

Furthermore, a significant pore induction (8.5 folds) was discovered in STZ-induced diabetic rats (Figure 5). However, CFOG and glibenclamide were able to preclude the pore opening observed in STZ-induced diabetic rats by 90% and 72% successively.

One of the diabetogenic mechanisms of STZ is generation of free radicals which could lead to lipid peroxidation measured as malondialdehyde (MDA) in this study. The impact of CFOG on lipid peroxidation in STZ-induced diabetic rats was displayed in Figure 6. It was observed that only the diabetic control showed highly significant difference in MDA level relative to normal control. In contrast, MDA produced in other groups were similar to normal control group.

Table 3 represents the effect of CFOG on mitochondrial ATPase function in STZ-triggered diabetic rats. Diabetic control was observed to display highly significant difference in the enzyme activity compared to normal rats. However, 400 mg/kg CFOG and 5 mg/kg glibenclamide drastically reduced the enzyme activity relative to diabetic control.

Figure 7 is an illustration of modulation of chromosomal DNA segmentation by CFOG in STZ-induced diabetic rats. Significantly high DNA fragmentation was observed in the diabetic control relative to normal control. Conversely, there was no significant difference in test groups, 400 mg/kg CFOG and 5 mg/kg glibenclamide in comparison with normal control.

Figure 8A shows that the pancreatic architecture was normal with exocrine acini (blue arrow) abundant and standard. The inter- and intra-lobular ducts were intact. Figure 8B depicts normal pancreas structure. Interlobular duct shows necrosis (black slender arrow) and haemorrhagic abrasion (red arrow). Figure 8C illustrates that exocrine acini filled with zymogen but the intralobular duct (black arrow) looked swelling and hyperemia (red) were noticed. In Figure 8D, undistorted exocrine acini (blue arrow), but intralobular distension and congestion (yellow arrow) were observed.

Figure 9A illustrates that abundant number of islets was normally distributed within the parenchyma cells. Figure 9B shows that some islets of Langerhans appear atrophic (white arrow). The results in Figures 9C,D indicate normal islets of Langerhans (white arrow) consisting of round to oval collections of endocrine cells. Conversely, treatment with CFOG shows that it possesses phytometabolites that can repair damage islet cells.

GSH level, glutathione-S-transferase and catalase activities were displayed in Figure 10. The results showed that similar trends of significant decline were observed in the antioxidant status of diabetic control relative to normal control. Whereas, in the group treated with 400 mg/kg CFOG, there was no significant difference observed in the antioxidant enzymes activities and GSH level in comparison with the normal control.

Figure 11 shows the structural elucidation of the bioactive principle which could be responsible for anti-diabetic agent. Lupanol was characterized from O. gratissimum leaf extract.