International Journal of Plant Studies

Terminalia superba root bark is traditionally used for the treatment of diverse diseases including diabetes in the South Western part of Nigeria and part of Africa. This study evaluates the anti-diabetic activity of the extract and partitioned fractions of the root bark of T. superba for scientific validation of the ethno-medicinal claim.

The methanolic extract (A) of T. superba was partitioned with n-hexane, dichloromethane and ethyl acetate to give four fractions (B₁ – B₁). A was screened for its glucose lowering activity using oral glucose tolerance test (OGTT) and alloxan-induced diabetic rat models. B₁ – B₁ were screened for their glucose lowering activity using oral glucose tolerance test (OGTT) model. In the OGTT model, three doses (100, 200 and 400 mg/kg) of the extract were administered to 24 h fasted induced hyperglycaemic rats. The most active dose was used in the alloxan-induced diabetic rat model for A and in the OGTT model for the B₁ – B₁ to monitor the antihyperglucaemic activity. The mother liquor (B₁) being the most active fraction was purified and was further subjected to various chromatographic separations which afforded methyl 3,4,5-trihydroxybenzoate (methyl gallate).

The result showed that 200 mg/kg of the extract was the most active dose in the OGTT model with 33 % blood glucose level reduction that was comparable (p > 0.05) to the 39 % given by glibenclamide (5 mg/kg) at 4 h. In the alloxan-induced diabetic rat model, the extract at 200 mg/ kg gave a significantly higher blood glucose reduction than the glibenclamide (5 mg/kg) throughout the 14-day period of treatment. The 31 and 40 % blood glucose level reduction given by ethyl acetate fraction (B₃) and the mother liquor (B₁) respectively at 4 h were comparable to the 39% activity elicited by glibenclamide (5 mg/kg) at the same hour. The structural elucidation of the methyl gallate isolated from the mother liquor (B₁) was carried out using 1D and 2D NMR spectra and the NMR data compared well with literature.

The study confirmed the ethno-medicinal use of the root bark of the plant in the treatment of diabetes. The methyl gallate isolated from the mother liquor (B₁) could be one of the constituents responsible for the activity observed in this fraction.

Keywords: Terminalia superba; root bark; anti-diabetes; chromatography; methyl gallate;

Diabetes mellitus is a multifactorial metabolic disorder characterized by prolonged hyperglycaemic state, bringing about impairment in carbohydrate, protein and lipid metabolism, resulting from increased hepatic glucose production and diminished insulin secretion or impaired insulin action or both [1]. Globally, an estimated 422 million adults are living with diabetes as the prevalence is increasing rapidly and the number is projected to double by 2030 especially in urban populations of the developing countries due to trend of urbanization and lifestyle changes, such as increased sedentary life styles and perhaps most importantly, a “Western-lifestyle” [2].

Terminalia superba Engl. & Diels. (Combretaceae), commonly known as “Afara” among the Yorubas in Nigeria is a deciduous tree widely spread in West and Central Africa and also grown in the East and South Western part of Nigeria [3,4]. It is used ethnomedically as an anti-diuretic, anti-malarial, in the treatment of conjunctivitis, bronchitis and in ovarian troubles [3]. In Cameroon it is used in the treatment of various ailments, including diabetes mellitus [5]. Its antimicrobial, anti-diarrheal, antifungal, antiulcer activities have been reported [6, 7, 8]. The anti-diabetic activity of the leaves and stem bark has been reported [9, 10, 11]. Ellagic acid derivatives; 3, 4-di-O-methylellagic acid 3-O-b -D-xylopyranoside, 4-O-galloy-3,3-di-O-methylellagic acid 4-Ob -D-xylopyranoside, 3,3-di-O-methylellagic acid and 3,3-di-O methylellagic acid 4-O-b -D-xylopyranoside have been isolated from the plant [12]. The present study was designed to investigate the possible glucose lowering effect of the root bark extract of the plant and isolate the principle(s) responsible for the effect.

NMR spectra were measured on Bruker DPX Avance 500 instrument using methanol-d4 as solvent and internal standard. For TLC, pre-coated silica gel 60F254 plates were used and compounds were detected under ultra-violet (UV) lamp (254 nm) and further visualized by spraying with vanillin-sulphuric acid solution. Accu-Chek Glucometer with Accu-Chek test strips. alloxan monohydrate (SIGMA chemicals, USA).

The root bark of T. superba was collected from the wild on Obafemi Awolowo University Campus, Ile-Ife, Osun State, Nigeria. It was authenticated by Mr. Ademoriyo of the Department of Botany, OAU, Ile-Ife and its voucher specimen, IFE–17575 was prepared and deposited at the herbarium.

The root bark of T. superba was air-dried and powdered with 1.5 kg of the powdered root bark extracted exhaustively with methanol. The methanolic solution was concentrated in-vacuo to obtain the crude extract (306.26 g; 20.4 %). The crude extract was re-taken in aqueous methanol (1:1, 1 L) and successively partitioned with n-hexane (5×400 mL), dichloromethane (4×400 mL) and ethyl acetate (16×400 mL) and the solutions were concentrated in vacuo to obtain their corresponding n-hexane (7.22 g), dichloromethane (4.23 g), ethyl acetate (81.82 g) and mother liquor (212.36 g) fractions respectively. The mother liquor was retaken in 1 L of distilled water and further partitioned with butanol (10×400 mL) to obtain butanolic fraction (152.54 g).

The butanolic fraction (50.0 g) was adsorbed on 50 g of silica gel and was left to dry overnight to avoid solvent interference. The adsorbed fraction was loaded on glass column dry and gradiently eluted using n-hexane, ethyl acetate and methanol. The eluent (100 mL) was collected from the column into each conical flask and 84 column fractions were obtained. These were bulked into ten column fractions according to their TLC profile. A brown solid (18.21 g) obtained at EtOAc/MeOH (92:8), was further subjected to repeated column chromatography on silica gel and Sephadex LH20 to yield a white crystalline solid (172.00 mg).

White crystalline solid; IR (KBr): υ^max (cm⁻¹) 3570 (OH), 1682 (C=O), 1601 and 1528 (C=C of the benzene ring), 1210 (C-O-C of ester). 1H and 13C NMR, Table 3.

Healthy Wistar rats of either sex (150 g, average weight) that were used for the experiments, were bred under standard conditions (temp. 27±3°C, relative humidity 65%, natural 12 h day–night) and housed in different cages in the animal house, Department of Pharmacology, Faculty of Pharmacy, O.A.U., Ile- Ife, Nigeria. The rats were fed on a standard pellet diet (Bendel Feeds, Benin, Nigeria), with water given ad libitum. They were acclimatized for at least 5 days before commencement of the experiments. Five groups of rats with five rats in each group were fasted for 24 h before administration of either glucose, extract, fractions, drugs or vehicle [13]. All animal experiments conformed to the Guide for the Care and Use of Laboratory Animals published by the National Academies Press [14].

Glucose (10 g/kg, p.o.) was administered to 24-hour fasted normal rats and those that were hyperglycaemic [blood glucose level ≥ 7.0 mmol/L (126 mg/dL)] after 0.5 hour (To) were selected and randomized into groups of five with five rats in each group. They were administered (p.o.) with extract (100, 200, 400 mg/kg), or 1 % Tween 80 in normal saline (negative control), or glibenclamide (5 mg/kg, positive control). At 0.0, 0.5, 1.0, 2.0 and 4.0 hours after administration of the test agents, a drop of blood that was taken from the tip of the tail of each rat was dropped onto a glucometer strip and the blood glucose level was directly read off the glucometer. The blood glucose levels at 0.0 h (To) were taken as 100 %, while those at other times were expressed as percentages of these values. Also, glucose lowering activity of the partition fractions were similarly assayed at 200 mg/kg, the highest active dose of the extract [15].

Groups of rats were injected (i.p.) with 150 mg/kg alloxan monohydrate dissolved in normal saline [16]. For six days, they were fed and water was given ad libitum. The rats with blood glucose (BG) levels ≥ 11.0 mmol/L (200 mg/dL) were considered diabetic. The diabetic rats were randomized into groups of three with five rats in each group. They were administered daily for fourteen days with 1 % Tween 80 in normal saline (negative control) or extract (200 mg/kg) or glibenclamide (5 mg/kg) (positive control) dissolved in the vehicle. Their BG levels were determined and recorded on 1, 4, 7, 10 and 14 days after administration of test agents, as reported by Ojezele and Abatan, 2011.

Data obtained from this study were expressed as the mean ± SEM for the number (N) of animals in the group. One way analysis of variance (ANOVA) was first used followed by Student–Neuman– Keuls’ test to determine the source of significant differences for all determinations and p < 0.05 was considered to be statistically significant.

There was significant time dependent reductions in the blood glucose levels up to the fourth hour in glucose-induced hyperglycaemic rats administered with 1 % tween 80 in normal saline (negative control) (Table 1) that was caused by homeostatic regulatory mechanism in the normal animals. This observation confirmed that the pancreases of the rats were functioning well [17]. The T. superba extract at 100 mg/kg lacked activity at 0.5-2 h but gave 20 % activity at 4 h. However, at 200 and 400 mg/kg, its activity was time dependent up to the fourth hour with 200 mg/ kg showing the highest activity at 4 h. The anti-hyperglycaemic activity of the extract at 200 mg/kg was comparable (p > 0.05) to glibenclamide (5 mg/kg) at all-time points (Table 1) suggesting similar minor exrapancreatic and major insulin stimulation mechanism of action of glibenclamide [18]. Furthermore, based on similar results, extracts of Uvaria afzelii, Chrystophyllum albidum, Xylopia aethiopica and Parquetina nigrescens have been reported to have insulin stimulation as their mechanism of action [15, 19, 20].

The alloxan-induced diabetic rats that were given 1 % tween 80 in normal saline were consistently hyperglycaemic throughout the 14 days of treatment indicating that they were permanently diabetic. The anti-hyperglycaemic activity of the extract at 200 mg/kg, the most effective anti-hyperglycaemic dose in glucoseloaded rats (Table 1), daily administered for 14 days to the diabetic rats was significantly higher than that of glibenclamide (5 mg/kg) at all times (Figure 1). Also, the time-dependent effect shown by the extract, similar to glibenclamide, may suggest insulin release as its major mechanism of action, which was earlier suggested from its anti-hyperglycaemic action using the glucose-loaded model (Table 1). This suggested the usefulness of the extract of T. superba in the management of prolonged hyperglycaemic conditions typified by the alloxan-induced rats. Similar to glibenclamide, significant anti-hyperglycaemic activity in glucose-loaded and drug induced diabetic rats have been reported for the methanolic extracts of Uvaria afzelii, S. cayennensis, J. tanjorensis and B. monandra [21, 22, 23].

Apart from the dichloromethane partition fraction (B2) of T. superba, other fractions gave similar profile of activities to glibenclamide indicating that they contained the various concentrations of the insulintropic constituents of the extract. The extract, its ethyl acetate fraction (B₃) and glibenclamide (5 mg/ kg) elicited comparable blood glucose reduction at all time points in the glucose-loaded rats. However, the aqueous fraction (B₁) showed significantly higher (p < 0.05) activity than glibenclamide at 0.5-1 h and comparable activity (p > 0.05) at 2-4 h indicating an additional extra-pancreatic activity to its insulin stimulating effect (Table 2). The overall order of anti-hyperglycaemic effect of the extract and its partition fractions at 4 h therefore is, B₁ > Gli > A > B₃ > B₁ > B2. This showed B₁ as the most active fraction and thus its choice for further purification (Table 2).

The 1H-NMR (MeOD, 500 MHz) spectrum showed two signals for aromatic protons at δ7.00 (s, 2H) and non-aromatic protons at δ3.82 (s, 3H). The 13C-NMR (125 MHz) showed signals at 169.8, 146.7, 139.8, 121.5, 109.9 and 52.2. The analysis of the carbon-13 and the DEPT-135 gave molecular formula of C8H8O2. In the HMQC, the protons at 3.82 showed correlations with carbon-13 at 52.2 and protons at 7.00 showed correlations with carbon-13 at 109.9. In the HMBC, proton at 3.82 showed correlations with carbon-13 at 169.8 and protons at 7.00 showed correlations with carbon-13 at 169.8, 146.7, 139.8 and 121.5 as shown in (Table 3). The NMR data compared well with published data [24]. Thus the compound was identified as methyl 3,4,5-trihydroxybenzoate, (Figure 2).

The anti-hyperglycemic effect exhibited by the extract of root bark of T. superba both in glucose-loaded rats and alloxan-induced hyperglycaemic rats in this study justified its ethno-medicinal anti-diabetic usage and methyl gallate that was isolated from the aqueous methanol, most active fraction of the extract is being proposed as the constituent or one of the principles responsible for the activity observed.

Authors are very grateful to Mr Nwamadi Mutshiyalo of the Department of Spectrau, University of Johannesburg, South Africa for his assistance in the acquisition of the NMR spectra.

1. World Health Organization. Global report on diabetes. World Health Organization. 2016.
2. International Diabetes Federation (IDF), Diabetes Atlas. 5th edition. Brussels, Belgium. 2011.
3.     Burkill HM. The useful plants of west tropical Africa, 2nd edition, Royal Botanic Gardens, Kew, United Kingdom. 1985.
4. Kimpouni V. Terminalia superba Engl & Diels. In: Lemmens, R.H.M.J., Louppe. D. and Oteng-Amoako A.A (Editors) PROTA Wageningen, Netherlands. 1990.
5.     Adjanohoun JE, Aboubakar N, Dramane K, Ebot ME, Ekpere JA, Enow-Orock EG, et al. Traditional Medicine and Pharmacopoeia: Contribution to Ethnopharmacological and Floristic Studies in Cameroon. OAU/STRC, Lagos. Pg.133. 1996.
6. Kuete V, Tabopda TK, Ngameni B, Nana F, Tshikalange TE and Ngadjui BT. Antimycobacterial, antibacterial and antifungal activities of Terminalia superba (Combretaceae). South African Journal of Botany. 2010;76(1):125-131.
7.    Ahon MG, Akapo-Akue JM, Kra MA, Ackah JB, Zirihi GN and Djaman AJ. Antifungal activity of the aqueous and hydro-alcoholic extracts of Terminalia superba Engl. on the in vitro growth of clinical isolates of pathogenic fungi. Agriculture and Biology Journal of North America. 2011;2(2):250-257.
8.     Onyegbulam CM.  Evaluation of anti-ulcer properties of methanol extract of Terminalia Superba Engl. & Diels (Combretaceae) stem bark. (Doctoral Dissertation, Department Of Pharmacognosy and Environmental Medicine, Faculty of Pharmaceutical Sciences, University of Nigeria). 2014.
9.     Kamtchouing P, Kahpui SM, Dzeufiet PDD, Tedong L, Asongalem EA and Dimo, T. Anti-diabetic activity of methanol/methylene chloride stem bark extracts of Terminalia superba and Canarium schweinfurthii on streptozotocin-induced diabetic rats. Journal of Ethnopharmacology. 2006;104(3):306-309.
10. Momo NEC, Oben EJ. In vivo assessment of hypoglycaemic and antioxidant activities of aqueous extract of Terminalia superba in alloxan-diabetic rats. Planta Medica. 2009;75-PH1.
11. Padmashree, Padmavathi PP and Pander S. Antidiabetic properties of methanol/methylene chloride extract of Terminalia superba leaves on streptozotocin induced diabetes in rats. International Journal of Pharm Tech Research, 2010;2(4):2415-2419.
12. Tabopda TK, Ngoupayo J, Tanoli SAK, Mitaine-Offer AC, Ngadjui BT and Ali MS, et al. Antimicrobial pentacyclic triterpenoids fromTerminalia superba. Planta medica. 2009;75(5):522-527.
13. Bamisaye FA, Odutuga AA, Minari JB, Dairo JO, Fagbohunka BS and Oluba OM. Phytochemical constituents and antidiarrhoeal effects of the aqueous extract of Terminalia superba leaves on Wistar rats. African Journal of Pharmacy and Pharmacology. 2013;7(16):848-851.
14. Committee for the Update of the Guide for the Care and Use of Laboratory Animals, Institute for Laboratory Animal Research, Division on Earth and Life Studies, National Research Council of the National Academies. Guide for the Care and Use of Laboratory Animals, 8th edn. The National Academies Press, Washington, DC, 2011.
15. Faloye KO, Ayoola MD, Amos-Tautua BM. and Famuyiwa SO. Anti-diabetic activity of convallatoxin isolated from the root bark of Parquetina nigrescens (Afzel.) Bullock (Asclepiadaceae), European Journal of Medicinal Plants. 2018;25(4):1-9.
16. Ojezele MO, Abatan OM. Hypoglycaemic and coronary risk index lowering effects of Bauhinia thoningii in alloxan induced diabetic rats. African health sciences. 2011;11(1):85–89.
17. Kar A, Choudhary BK, Bandyopadhyay NG. Preliminary studies of the inorganic constituents of some indigenous hypoglycaemic herbs on oral glucose tolerance test. Journal of Ethnopharmacology. 1999;64:179-184.
18. Luzi L, Pozza G. Glibenclamide: an old drug with a novel mechanism of action? Acta diabetologica. 1997;34(4):239-244.
19. Olanudun EA, Ayoola MD, Famuyiwa SO. The isolation and characterization of lupeol-3-acetate from the hyperglycaemia-lowering fraction of the stem bark of Crystophyllum albidum G. DON (Sapotaceae). Ife Journal of Science. 2018;20(2):229-236.
20. Famuyiwa FG, Ayoola MD, Famuyiwa SO and Aladesanmi AJ. Hyperglycaemia lowering effect of kaurane diterpenoids from the fruits of Xylopia aethiopica (A. Dunal) Rich. International Journal of Medicinal Plants and Natural Products. 2018;4(3):11-19.
21. Olayiwola G, Iwalewa EO, Omobuwajo OR, Adebajo AC, Adeniyi AA and verspohl JE. The anti-diabetic potential of Jatropha tanjorensis leaves. Nigerian Journal of Natural Products and Medicine. 2004;8:55-58.
22. Adebajo AC, Olawode EO, Omobuwajo OR, Adesanya SA, Begrow F and Elkhawad A, et al. Hypoglycaemic constituents of Stachytarpheta cayennensis leaf. Planta Medica, 2007;73(3):241-250.
23. Alade GO, Omobuwajo OR, Adebajo CA and Verspohl EJ. Evaluation of the hypoglycaemic activity of Bauhinia monandra leaf in alloxan-diabetic rats and INS-1 insulin cells. Journal of Chemical and Pharmaceutical Research. 2011;3(2):506-521.
24. Ekaprasada MT, Nurdin H, Ibrahim S and Dachriyanus D. Antioxidant activity of methyl gallate isolated from the leaves of Toona sureni. Indonesian Journal of Chemistry. 2010;9(3):457-460.