Fresh stem barks of Protorhus longifolia (Benrh.) Engl. were collected in March 2012 from Hlabisa, Kwa-Zulu Natal, South Africa. The plant (voucher specimen number RA01UZ) was authenticated by Dr. N.R. Ntuli, Department of Botany, University of Zululand. The plant material was thoroughly washed with tap water and then air dried. The air dried plant material was ground into powder (2 mm mesh) and stored in sterile brown bottles until use.
Extraction and isolation
The method of extraction and isolation of the triterpenes from the stem bark of P. longifolia has been previously described . Briefly, the powdered plant material was first defatted with n-hexane and then extracted (1:5 w/v) with chloroform. The compounds were isolated from the chloroform extract (13 g) using silica gel column chromatography (24 × 700 mm; Silica gel 60; 0.063 - 0.2 mm; 70–230 mesh ASTM, Merck, Darmstadt, Germany). The column was eluted stepwise with a mixture of n-hexane and ethyl acetate (9:1–3:7) and 20 ml fractions were serially collected. Thin layer chromatography (TLC) (silica gel 60 TLC aluminium sheets 20 cm × 20 cm, F254, Darmstadt, Germany) was used to analyse the fractions. The combined fraction 14 was further purified in ethyl acetate to afford the compound (KE1, 1.15 g). Melting point of the compound was determined using Stuart SMP 11 melting point apparatus (Shalom Instruments supplies, Durban, South Africa). Spectroscopic data analysis, NMR (1H-1H, 13C-13C, in DMSO, Bruker 600 MHz), HRMS (in DCM, Waters Synapt G2) and infrared (IR) (Perkin-Elmer 100 FTIR) techniques were used to establish and confirm structure. Chemical shifts were expressed in δ (ppm).
This study was approved by University of Zululand Research Animal Ethics Committee (UZREC 171110–030 Dept. 2013/23). Sprague–Dawley rats (180-220 g) were collected from animal house in the Department of Biochemistry, University of Zululand, South Africa. Experimental procedures were conducted following the guideline for care and supervision of experiments on animals. The animals were housed in standard cages and maintained at room temperature with 12:12-h light: dark cycle. All rats had free access to drinking water and standard rat feed in the experimental environment, for 1 week, before the experiment was conducted. Once the animals had adapted to the environment, forty-two (42) rats were divided into two groups; normal group (ND) consisting of 12 rats and the high fat diet (HFD) group consisting of 30 rats. After 21 days on their respective diets, 6 rats per group were sacrificed and the remaining rats of the HFD group were randomly divided into a total of four (4) groups of six rats per group.
Group 1: normal diet and vehicle throughout the study
Group 2: high fat diet and vehicle throughout the study
Group 3: was subdivided into two groups (A, B) and received high fat diet and compound (100 and 200 mg/kg body weight, respectively), dissolved in 2% Tween 20
Group 4: high fat diet and lovastatin (10 mg/kg body weight), dissolved in 2% Tween 20
Induction of dyslipidemia in rats
The method previously described by Hor et al. was followed to evaluate the anti-hyperlipidemic activity of the triterpene. Rats were made hyperlipidemia by feeding a high fat diet [commercial rat chow (97.3%), Sunflower oil (15%), bile salt (0.5%), cholesterol (5%), Thirmecil (0.2%)]. This HFD preparation was pelleted (about 3 g each) and fed daily to the rats for 36 days to induce hyperlipidemia.
Measurement of body weight and food intake
where wt- weight
Collection of blood samples and Liver for Lipid profile determination
At the end of the experimental periods, the rats were fasted for 8 hours, and then sacrificed by a blow to the head and blood samples were collected by cardiac puncture. The collected blood samples were centrifuged at 3500 rpm for 10 minutes and the serum collected for biochemical studies. The liver was excised, weighed and stored in formalin for histological studies.
Histology studies were carried out at the Vet Diagnostix Laboratories (Pietermaritzburg, SA) by qualified pathologist having no prior knowledge to which group they belonged. The liver tissues were stained with haematoxylin and eosin (H & E). This method allowed for unbiased description of the histological lesions which were present or absent in the samples.
The serum samples were used for the estimation of total cholesterol (TC), total triglyceride (TG), HDL-cholesterol (HDL-c), AST, ALT and Alkaline phosphatase. Analysis was done using the Cobas c 111 analyzer.
LDL-cholesterol (LDL-c) was estimated using Friedwald’s equation 
LDL-c = [TC-(HDL-c + (TG/5)]
Other lipids parameters such as VLDL- cholesterol, coronary risk index and atherogenic index (AI) were calculated  as follows:
LDL-c = [TC-(HDL-c + (TG/5)]
VLDL-c = [TG – (HDL-c + LDL-C)]
Atherogenic index (AI) = LDL-c/HDL-c (mg/dl)
Coronary risk index (CRI) = TC/HDL-c (mg/dl)
Data was analyzed using one-way analysis of variance (ANOVA) followed my Tukey-Kramer multiple comparison test using GraphPad InStat® version 3. The results are presented as mean ± standard error of the mean (SEM). Values of p < 0.05 were considered significant.