Upsala J Med Sci 82: 3 9 4 1 , 1977 Clinical Trial in Patients with Diabetes Mellitus of an Insulin-like Compound Obtained from Plant Source V. S. BALDWA, C. M. BHANDARI,' A. PANGARIA and R . K . GOYAL From the Department of Medicine, S . M . S . Medical College Hospital, Jaipur, India ABSTRACT Clinical study of an insulin-like compound obtained from vegetable source (vegetable insulin) was carried out on nine patients with diabetes mellitus. The active hypoglycaemic principle, purified protein extract, was obtained from fruits as well as from tissue cultures of the plant Mornordica charanlia L. This extract was homologous to insulin obtained from animal pancreas. It showed a consistent hypoglycaemic effect in patients with diabetes mellitus. The average fall in blood sugar level at the peak effect of vegeta- ble insulin was found to be statistically significant. The onset of action was within 30-60 min with the peak effect six hours after the administration of the dose of plant insulin. No hypersensitivity reaction to this extract was observed in the group of patients studied. INTRODUCTION A crude extract obtained from the fruit of a plant known as Momordica charantia L. (bitter gourd), has been shown to possess hypoglycaemic activity when tested in rabbits ( 1 , 2 , 3). However, this ex- tract was also found to have many side effects, including uterine haemorrhage in gravid female rabbits. An intraperitoneal injection of this extract invariably caused the death of the experimental animal. Khanna & Mohan (4) were able to extract an abortifacient factor present in this crude extract and isolated diosgenin from the fruit as well as from the in vitro tissue cultures of this plant. Subse- quently they were able to extract out the active principle in a pure protein from (vegetable insulin, v-insulin) which could be used as a hypoglycaemic agent i n human beings after biological standardiza- tion ( 5 ) . Vegetable insulin (v-insulin) is structurally and pharmacologically comparable in many re- ' Guest research fellow, Department of Pulmonary Dis- eases, University Hospital, Uppsala, Sweden. spects to bovine insulin ( 5 ) . The method of extrac- tion of v-insulin is similar to that of extraction of pure insulin from the pancreas of animals (6). Dur- ing its extraction, traces of zinc are added, resulting in the formation of colourless, needle-like crystals. The crystals of v-insulin are purified by thin-layer chromatography. The electrophoretic pattern also resembles that of bovine insulin. The infrared spectrum of p-insulin is superimposable on that of standard zinc crystalline insulin. Qualitative amino acid analysis by paper chromatography and quantitative analysis by an amino acid analyser showed that p-insulin consisted of 17 amino acids. The three-dimensional structure was found to con- sist of two chains of amino acids, bound together with sulphide bonds. The biological assay of the hypoglycaemic activity of v-insulin has been de- termined in animal experiments (5). Vegetable insulin is available as a suspension which is stable at 4°C and denatured by heat. The compound is suspended in sterile double-distilled water and ultra-violet light and potassium perman- ganate fumigation is used for sterilization. The dose is so standardized that the final concentration is 40 units per ml (1.8mg per 40 units). It can be administered by the subcutaneous or the intra- muscular route. MATERIAL AND METHODS Nine in-patients, eight males and one female, with an age range of 16 to 52 years, from S.M.S. Medical College Hospital, Jaipur, India, were studied after informed con- sent was obtained. All had diabetes mellitus and the dura- tion of their disease ranged between 3 months to 10 years. Diagnosis of diabetes mellitus was confirmed by clinical examination and laboratory investigations. Patients with primary or idiopathic diabetes mellitus were studied. Up.rula J M e d Sci 82 40 V. S. Baldwa e t al. Table I. Clinical data of patients with diabetes mellitus Case Age height weight disease no. Name Sex (Y.) (cm) (kg) (Y.) Type of diabetes 1 JC M 22 168 48.5 6 Juvenile 2 GR M 20 168 45.0 6 Juvenile 3 GS M 16 148 43.0 7 Juvenile 4 PL F 20 152 39.5 8 Juvenile 5 SK M 22 170 58.0 10 Juvenile 6 us M 18 165 48 .O 8 Juvenile 7 HR M 50 165 55.0 0.5 Chemical 8 GC M 52 172 57.0 0.3 Chemical 9 MR M 50 162 60.5 1 Maturity onset Body Body Duration of These patients were placed in two groups, depending on their stage of carbohydrate decompensation, as follows: (i) Overt, clinical diabetes, i.e. patients with elevated fasting and random blood sugar levels (more than llOmg per 100ml). They were further subdivided into juvenile and maturity onset types, depending on the age a t onset of diabetic symptoms. (ii) Asymptomatic diabetes, i.e. patients with normal blood sugar levels (less than 1 lOmg per 100 ml) but with an abnormal intravenous glucose tolerance test. There were 6 patients with juvenile diabetes, one with maturity onset and 2 with asymtomatic diabetes mellitus (Table I). All antidiabetic medication in the form of insulin or oral hypoglycaemic agents was stopped 24 hours prior Table 11. Dose of vegetable insulin, according to blood sugar levels Fasting blood Dose of Seventy of sugar level vegetable diabetes (rng%) insulin Mild Less than 180 10 units Moderate 180-250 20 units Severe 250 and above 30 units to the study. A fasting blood sugar sample was taken a t 7 a.m. before giving v-insulin. The dose of v-insulin was varied according to the severity of the disease and was de- cided upon arbitrarily according to the fasting sugar level (Table 11). Vegetable insulin was administered subcutane- ously and samples for blood sugar determination were taken at regular intervals. The first three samples were drawn at half-hourly intervals (7.30 a m . , 8.00 a.m. and 8.30 a.m.) to establish the onset of action and the subse- quent samples were collected a t 11.00 am., 1.00 p m . , 3.00 p m . and 7.00 p.m. to determine the peak effect and duration of action of v-insulin. All samples were collected intravenously. Blood sugar estimation was done on whole blood by the method described by King & Wooton (7). All the subjects were kept fasting during the interval for col- lection of samples and only plain lemon water was given if desired by the patients. A provision was kept for the administration of glucose in the event of the development of hypoglycaemic symptoms. Five healthy volunteers (control group I) and 5 patients with overt diabetes mellitus (control groups 11) served as controls. A placebo injection was used. The control sub- jects were kept fasting and blood samples were collected and analysed in a similar way. The administration of v- insulin to control subjects was avoided due to its inherent hypoglycaemic properties, as evaluated in animal experi- ments. Table 111. Effect of vegetable insulin and placebo on blood sugar levels Results given in percentage of fall in blood sugar levels. Statistical values are shown ~~~ ~~ Fasting % of fall in blood sugar levels No. of values Clinical sub- (mean) 7.30 8.00 8.30 11.00 1 .00 3 .oo 7.00 group jects mg% a.m. a.m. a.m. a.m. p.m. p.m. p.m. Healthy 5 75 5.0 5.0 5.6 5.4 5.6 5.5 5.4 controls I f 7.4 f 1 . 7 21.7 k1.4 1 1 . 8 f 1 . 6 21.4 k l . 5 Diabetic 5 2 10 4.6 4.5 5.0 5.8 5.4 5.8 5.7 controls I1 k11.8 k2.0 f 2 . 6 k2.1 11.8 k1.6 f1.8 f 2 . 0 Diabetes 9 295 21.5 24.8 30.2 49.2 40.3 35.9 28.8 mellitus f 1 5 . 7 k 8.9 211.0 f12.1 fY3.7 f13.4 f 1 0 . 3 f11.4 L'pscrlri J Med S c i 82 RESULTS The placebo injection in the control groups did not produce any appreciable reduction in blood sugar levels at different intervals. A definite hypoglycaemic effect of v-insulin was observed in the patient group in this study. The onset of vegeta- ble insulin effect was observed within +-1 hours, with the peak effect after 4 hours in 6 juvenile diabetics, after 6 hours in 2 patients with chemical diabetes mellitus, and after 12 hours in one patient with maturity onset of diabetes mellitus. All values in this study were analysed statistically by applying the paired t-test and the calculated f was 3.3 and the tabulated t was 2.3 at d.f. 8”, which was highly significant for the diabetic patient group compared with the healthy and diabetic controls during peak hours. The hypersensitivity reactions were conspicu- ously absent after administration of vegetable insu- lin and there was no local reaction at the site of injection. DISCUSSION The present investigation revealed that vegetable insulin has a consistent hypoglycaemic effect in pa- tients with diabetes mellitus. The onset of action is similar to that of standard zinc crystalline insulin (30-60 min). However, the peak effect of vegetable insulin was seen after 4-12 hours as compared with, for 2-3 hours regular insulin. The greatest fall in blood sugar levels observed in the patient group was found to be statistically significant. There were no anaphylactic reactions to vegetable insulin how- ever, as regards its long-term use, further studies are required in order to evaluate its antigenic prop- erties. The availability of vegetable insulin should open new horizons in the treatment of diabetes mellitus, especially where it is taboo to use animal products. Since the active principles are derived from a veg- etable source, it can be obtained in abundance. Further clinical trials are needed in order to establish its duration of action, assay, antigenicity and various effects on intermediary metabolism in human beings. ACKNOWLEDGEMENT The authors are grateful to Dr P. Khanna, Department of Botany, University of Rajasthan, Jaipur, for providing the vegetable insulin used in this study. Clinicat trial of an insulin-like compound 4 1 REFERENCES 1. 2. 3 . 4. 5. 6. 7. Rivera, G . : Preliminary chemical and pharmacological studies on “Cundea Mor” and “Charantia L”. Am J Pharm I1 3: 28 1 , 194 1. Rivera, G . : Charantia L , 11. Am J Pharm 114: 72, 1942. Sharma, V. N . , Sogani, R. K., Arora, R. B.: Some observations on hypoglycaemic activity of Momordica Charantia. Ind J Med Res 48: 471, 1%0. Khanna, P. & Mohan, S.: Isolation and identification of diosgenin and sterols from fruits and in vitro cultures of Momordica Charantia Linn. Ind J Exp Biol 1 1 : 5 8 , 1973. Khanna, P., Nag, T. N. & Jain, S . C.: Extraction of insulin from plant cultures in vitro. Third International Congress of Plant Tissue and Cell Culture, held at Leicester, England, July, 1974. Vestling, C. S.: Insulin. Biochem Preps6:28, 1958. King, E. J . & Wooton, I.D.P.: Microanalysis in Medi- cal Biochemistry, 3rd ed. J & A Churchill Ltd, London, 1956. Received September 20, 1976 Address for reprints: Dr V . S. Baldwa, M.D. 12-Viveka Nand Marg, C-Scheem, Jaipur, India Upsula J M e d Sci 82