The collected fruits were examined carefully and old, infected and damaged fruits were removed. Initially, the pods were separated the arils were isolated manually from brown peel and black seed. The fruit peels were washed with tap water and then with distilled water to remove any debris or dust particles, healthy fruit peels were spread out and dried at room temperature for about 15 - 20 days and pulverized by a mechanical grinder and passed through a 40-mesh sieve to get a fine powder and stored in an airtight container [12] .

The air dried powdered fruit peel material of Pithecellobium dulce was successively extracted with pet ether (60˚ - 80˚) and methanol for 16 hours (thrice) using Sohxlet extractor. The methanol extract was concentrated by rotary vacuum evaporator and then dried.

Dried and ground fruit peel was extracted with methanol for 16 hrs in a Soxhlet apparatus. The solvents were concentrated in rotary evaporator at reduced pressure below 40˚C. The crude extract was used for isolation. The crude extract was added to 40 grams of silica gel (60 - 120 mesh) to make admixture. 2.4 diameter columns were packed with the admixture mixed with hexane. The column was eluted with increasing solvent polarity from hexane to ethyl acetate. Column of 2.4 cm, column bed height 20 cm was used for the isolation procedure. Hexane (100% - broad fraction 1), hexane: ethyl acetate (90:10 - broad fraction-2)—compound 1, hexane: ethyl acetate (80:20 - broad fraction3), hexane: ethyl acetate (70:30 - broad fraction 4), hexane: ethylacetate (60:40 - broad fraction 5)—compound 2, hexane: ethyl acetate (50:50 - broad fraction 6), hexane: ethyl acetate (40:60 - broad fraction 7), hexane: ethyl acetate (30:70 - broad fraction 8), ethyl acetate (100% - broad band 9). Fractions with similar spots were pooled together and concentrated at reduced pressure and temperature. Compounds 1 and 2 answered Molisch test for sugar and glycoside test.

The blood was collected from a healthy human volunteer and transferred into a blood bottle containing an anticoagulant. Haemolysate was prepared based on the principle of hypotonic lysis [13] . The red blood collected was washed thrice with 0.14 M NaCl solution and one volume of red blood cells suspension was lysed with two volumes of 0.01 M phosphate buffer, pH 7.4 and 0.5 volume of CCl₄. The haemolysate was then freed from the debris by centrifugation at 2300 rpm for 15 min at room temperature. The haemoglobin rich fraction i.e., the upper layer was separated and dispensed into sample bottle for storage and refrigerated until required for use [14] .

Antidiabetic activity of compounds isolated from leaf, fruit and fruit peel of Pithecellobium dulce were investigated by estimating degree of non-enzymatic haemoglobin glycosylation, measured colorimetrically at 520 nm. Glucose (2%), haemoglobin (0.06%) and gentamycin (0.02%) solutions were prepared in phosphate buffer 0.01 M, pH 7.4. 1 ml each of above solution was mixed. 0.2 to 1 mg/ml of compounds was added to above mixture. Mixture was incubated in a dark room temperature for 72 hrs. The degree of glycosylation of haemoglobin was measured colorimetric ally at 520nm. Alpha-Tocopherol (Trolax) was used as a standard drug for assay % inhibition was calculated as the earlier methods [15] .

A starch solution (0.1% w/v) was obtained by stirring 0.1 g of potato starch in 100 ml of 16 mM of sodium acetate buffer. The enzyme solution was prepared by mixing 27.5 mg of alpha amylase in 100 ml of distilled water. The colorimetric reagent is prepared by mixing sodium potassium tartarate solution and 3, 5 dinitro salicylic acid solution 96 mM. Both control and plant compound were added with starch solution and left to react with alpha-amylase solution under alkaline conditions at 25˚C. The reaction was measured over 3 minutes. The generation of maltose was quantified by the reduction of 3, 5 dinitro salicylic acid to 3-amino-5-nitro salicylic acid. This reaction is detectable at 540 nm [13] .

The concentration of the plant compound required to scavenge 50% of the radicals (IC₅₀) was calculated by using the percentage scavenging activities at five different concentrations of the compounds. Percentage inhibition (I %) was calculated by

$I\%=\left(Ac-As\right)/Ac\times 100$.

Ac is the absorbance of the control;

As is the absorbance of the sample.

Diabetes mellitus is a metabolic disorder with increasing incidence throughout the world. Insulin is a key player in the control of glucose homeostasis. Lack of insulin affects carbohydrate, fat and protein metabolism [16] . Management of diabetes without side effects are still challenge to the medical community. It was proposed that inhibition of the activity of such alpha-amylase would delay the degradation of carbohydrate, which would in turn cause a decrease in the absorption of glucose; as a result there is a reduction of postprandial blood glucose level. In vitro refers to the technique of performing a given procedure in a controlled environment outside the living organism. The purpose of in vitro testing is to demonstrate the hypoglycemic activity of compounds isolated from the fruit peel of pithecellobium dulce. The report was tabulated in ( Table 1 and Table 2 ). The present research has been carried out to isolate the new compounds present in the fruit peel of Pithecellobium dulce and their effect in inhibiting glycosylation of haemoglobin and alpha-amylase. At the concentration of 0.2 ml, compound 1 & compound 2 showed a percentage inhibition of 26.1% and 7.2% for non-enzymatic glycosylation of haemoglobin assay. In the case of 1.0 ml concentration inhibition level was 73.7% & 53.9% ( Table 1 ) ( Figure 1 ) for non-enzymatic glycosylation of haemoglobin assay. The compounds isolated from pithecellobium dulce revealed a significant inhibitory enzyme action. The percentage inhibition at 0.2 - 1.0 ml concentrations of Pithecellobium dulce compound showed a dose dependent increase in percentage inhibition. The percentage inhibition for compound 1 and compound 2 varied from 11.8% to 80.9% and 7.1% - 56.5% from the highest concentration to the lowest concentration for alpha-amylase inhibitory activity. ( Table 2 ) ( Figure 2 ).

The present finding reveals that the compound 1 efficiently inhibits both alpha amylase and alpha-glucosidase enzymes in a dose dependent manner compared to compound 2. Manikandan et al. (2013) [17] investigated the phytochemical bioactive compounds of the methanol extract of Psidium guajava leaves, its in vitro anti-diabetic activity. It was proposed by Rhabaso and Chiasson, (2004) [16] that inhibition of the activity of alpha-amylase and alpha-glucosidase would delay the degradation of carbohydrate, which would in turn cause a decrease in the absorption of glucose, as result the reduction of postprandial blood glucose level elevation. The assay results suggested that the presence of bioactive compounds, could be responsible for the versatile medicinal properties of this plant, including diabetes. The antidiabetic action of compounds isolated from Pithecellobium dulce fruit peel can also be attributed due to the presence of polyphenolic compounds in Pithecellobium dulce may have a potentially important role in managing diabetes via the inhibition of α-amylase and non-enzymatic glycosylation of haemoglobin. In this present study, we evaluated in vitro non-enzymatic glycosylation of haemoglobin and alpha amylase. It was suggested that further studies are required to elucidate the mechanism of antidiabetic potential.