OPTIMIZATION AND PURIFICATION OF α-GLUCOSIDASE INHIBITOR FROM BACILLUS SUBTILIS YT20

α-glucosidase inhibitor has been drawing the researcher’s interest world-wide because of its safe and convenient mechanism in the treatments of type 2 diabetes. However, as recent therapeutic molecules have side effects, alternative sources have been studied to limit the side effects and provide more options for the treatment of type 2 diabetes. In this study, after selecting microorganism strains capable of biosynthesis of α-glucosidase inhibitor from the collection of Bacillus subtilis, we have selected B. subtilis YT20 as the host to produce secondary metabolites. After optimization, the extracellular extract inhibited 91.5 % which is 1.34fold higher than pre-optimization (68.1 %). After purification, the α-glucosidase inhibitory molecule of the B. subtilis YT20 has the molecular weight of 163 Da. The recovery efficiency at the final step reached 16.1 %. The combinational spectral data of MS showed that the compound isolated from the B. subtilis YT20 was 1-deoxynojirimycin.


INTRODUCTION
Diabetes is a recognized epidemic disease in Vietnam and over the world, due to the exponential increase of cases, the serious consequences of the disease on health and longevity, and the burdens of the treatment cost for patients and society. According to International Diabetes Federation, in 2015, 415 million people aged 20 to 79 years had diabetes in the world which is equivalent to 1 in 11 people diagnosed with this metabolic disorder. These numbers will go up to 642 million in 2040. In other words, there is 1 in 10 normal people who will have diabetes. In the treatment of type 2 diabetes, the therapy taking advantage of α-glucosidase inhibitor has been drawing the researcher's interests because of its safety and simplicity as a neutral substance. For optimization and limitation of side effects, more studies are needed to find out alternative α-glucosidase inhibitor from various sources.
Domestic and international researches focus on selection, optimization and purification to discover new α-glucosidase inhibitors including plants [1,2], and microorganism [3]. Alternatively, 1-deoxynomirycin (DNJ) is an iminiosugar that inhibits α-glucosidse enzyme to prevent the formation of sugar in the digestive tract [4]. The substance originated mostly from mulberry leaves in a small number. Thus, different sources have been exploited to produce DNJ via Bacillus sp. In vivo streptozotocin -induced diabetic white mice model, the crude extract of the B. subtilis S10 containing DNJ reduced the blood glucose level by 209.1 mg/dL (59.1%), lower than 510 mg/dL of the control group [5]. Alternatively, the α-glucosidase inhibitor was isolated from B. lentimorbus which is collected from the culture media containing 0.25% natri glutamate and 0.5% glucose; pH 8.0; at 30C in 48 hrs [6]. Later, an α-glucosidase inhibitor was purified from the B. subtilis B2 whose structure was determined as 1-deoxynojirimycin of 163 Da by MS and NMR [7]. In 2013, Lee et al purified 1-deoxynojirimycin from the B. subtilis MORI which could enhance adiponectin level and its receptor in differentiated 3T3-L1 fat cells. The evidences indicated that this active substance proves effective in reducing blood glucose level which showed no sign of toxicity to differentiated 3T3-L1 cells at the concentration of 5 mM [8]. These are the basic foundations for the search of new compounds possessing α-glucosidase inhibitory activity used in the treatment of type 2 diabetes from the selected microorganisms in Vietnam. In this study we isolated and identified 1-deoxynojirimycin from the B. subtilis YT 20 in Vietnam oriented for the treatment of diabetes.

Microorganism strain and culture media
B. subtilis YT 20 was provided by the National key lab of gene technology at Institute of Biotechnology, Vietnam Academy of Science and Technology. The B. subtilis YT 20 was grown in culture media (pH 7.5), which comprised corn starch (10 g/L), soybean meal (5 g/L), and yeast extract (5 g/L), KH 2 PO 4 (0.5 g/L), (NH 4 ) 2 SO 4 (0.5 g/L).

Incubation and harvesting active ingredients
The B. subtilis YT20 was incubated for 5 days with shaking-speed of 200 rpm at 37 C. The culture media consisted of: 1 % corn starch, 0.5 % soybean meal, 0.5 % yeast extract, 0.05 % KH 2 PO 4 ; 0.05 % (NH 4 ) 2 SO 4 ; pH value of 7.5. After incubation, the supernatant was centrifuged at 12000 rpm for 15 mins, and 10 µL of supernatant was evaluated for the α-glucosidase inhibitory activity.
There are several factors which influence the optimal condition of the production of secondary metabolites. Therefore, we conducted optimal experiments to evaluate the effect of each factor by determining its α-glucosidase inhibitory activity following these categories. Effect of incubation times: 10 µL of extracellular extract was evaluated at 48; 72; 96; 120 and 144 hrs. The same volume of sample was used to analyze the effect of pH (4; 7; 7.5; 8 and 10), temperature (28C; 30C and 37C), 1% media of different carbon sources (corn starch, glucose, maltose, galactose, sucrose and lactose), 0.5% media of different nitrogen sources (yeast extract, soybean powder, peptone, malt extract and beef extract).

Determination of α-glucosidase inhibitory activity
The α-glucosidase enzyme hydrolyzes substrate p-nitrophenyl-α-D-glucopyranoside to produce yellow p-nitrophenol that absorb 405 nm wavelength [9]. Each well of 96 wells plate was added 50 µL containing 40 µL of 0.1 M phosphate buffer pH 6.9 and 10 µL of sample, while 50 µL of 0.1 M phosphate buffer was used as control. Next, 100 µL of α-glucosidase 1U/ml in 0.1 M phosphate buffer pH 6.9 was added to previous mixture. The sample was incubated at 25 C in 10 mins, shaking 300 rpm. Then, 50 µL 5 mM p-nitrophenyl-α-Dglucopyranoside in 0.1 M phosphate buffer pH 6.9 was added. The solution was mixed well and incubated at 25 C in 5 mins. All the measurements are triplicated in 96 wells plate.
The αglucosidase inhibitory activity was calculated by the following formula The change in measurement of OD value before and after incubation of control. S A  : The change in measurement of OD value before and after incubation of the sample.

Thin layer chromatography
TLC is an effective method to determine and analyze the number of different groups of substances presented in the collected fractions. The principal bases on the identical polarities and Rf of different compound. The crude extract was centrifuged at 4000 rpm for 10 mins, the upper phase was continuously centrifuged at 12500 rpm for 15 mins. The supernatant was collected and mixed with absolute alcohol by the ratio of 1 sample: 4 alcohol (v/v). After 30 mins, the fermented broth mixture was centrifuged at 12500 rpm in 15 mins, the supernatant was collected and evaluated by the TLC.
The TLC was performed on the Merck 60 F254 thin silica gel plate, with a thickness of 0.25 nm and a solvent system 4 isopropanol : 1 acetic acid : 1 H 2 O (v/v/v). The TLC plate was then dried and colorized by ninhydrin reagent.

Purification of α-glucosidase inhibitor
In order to purify α-glucosidase inhibitor from fermented broth of the B. subtilits YT20, the supernatant was centrifuged at 1200 rpm for 15 mins and precipitated with absolute alcohol at 4C in 1 hr at the ratio of 1 sample: 4 alcohol (v/v) to eliminate protein and polysachharides. The upper phase was then dissolved in deionized H 2 O and passed through the 0.45 µm filter membrane. Next, the solution was filtrated with two membrane pore size 100 Da and 1000 Da, respectively. (Molecular weight cut off; Spectra/Por*, USA). The supernatant was loaded onto the activated charcoal column (Column size: 24 x 0.6 cm; flow speed 1.5 ml/min). Different concentrations of ethanol from 5 -30 % were used to collect the fractions. These elutions later were examined by the TLC and α-glucosidase inhibitory activity. The potential segments were loaded onto sephadex G100 column G100 (24 x 0.6 cm), to collect putative compound. The fractions were eluted with potassium phosphate buffer 50 mM, pH 7.5. The flow speed was at 25 mL/h, the active substance was evaluated for the α-glucosidase inhibitiory activity and elucidated structure by mass spectroscopy.

Statistical analysis
All measurements were carried out in triplicate. The means were presented for averages of experiments.

Alpha-glucosidase inhibitor activity of crude extract of B. subtilis YT20
The crude extract of B. subtilis YT20 showed a high α-glucosidase inhibitory activity at 68.1 %. While standard acarbose reached 75.2 % at the concentration of 1 mg/mL (Sigma). Equally, at the same concentration, the inhibitory activity of DNJ (Sigma) reached 70.1 % (Table 1). The Bacillus sp. in particular, is a potential bacterial source which has been extensively investigated for production of -glucosidase inhibitor, including the B. subtilis S10 [10], B. subtilis B2 [7], and B. mycodes TKU040 with 143 U/mL and IC 50 value of the α-glucosidase inhibitors produced in the culture supernatant being 3 mg/mL [11]. Alternatively, the αglucosidase inhibitory activity from the B. subtilis was recorded at as high as 8 5 % [12].

Effect of incubation time
The fermented conditions for production of the α-glucosidase inhibitors from the B. subtilis YT20 in a basic culture condition are 28 C with shaking speed of 200 rpm, and pH 7.5. After 96 hrs of culture, the α-glucosidase inhibition activity of the extracellular extract from the B. subtilis YT20 reached the highest peak of 80 %, followed by a slight decrease after 120 hrs, and almost remained unchanged at 75 % after 144 hrs of fermentation. After 96 hrs, the nutrition in the culture media slowly depleted, and led to some exotic stress on microorganism so other secondary metabolites was secreted to media and slightly reduced the inhibitory activity. Our results are in agreements with other studies in which the secondary metabolites showed the highest α-glucosidase inhibition activity after 96 hrs of incubation from the B. subtilis B2 [7] and the B. mycodes TKU040 [11]. However, the α-glucosidase inhibitory activity from the B. subtilis was over 85 % after incubation for 9 days [12]. The α-glucosidase inhibitory activity of the B. subtilis B2 fermentation was increased slightly after 6 days [7] and the B. subtilis S10 (24 hrs) [10]. Thus, the different strains have the ability to produce the α-glucosidase inhibitors whose activity is varied depend on the fermentation medium.  Five pH value at 4; 7; 7.5; 8 and 10 were evaluated for the biosynthesis of α-glucosidase inhibitors from the B. subtilis YT20. We found that the optimal pH is 7.0 and 7.5 possessing the highest activity of up to 82%. At lower or higher pH 7.0, it exerted negative effects on the production of the α-glucosidase inhibitors. This would suggest the active compounds were affected by pH of culture media which is agreeable with previous studies of the Bacillus sp. [13].

Effect of temperature
The B. subtilis YT20 strain was incubated in the basic fermentation media supplementing with corn starch, soybean powder, yeast extract and minerals at pH 7.0 with shaking-speed of 200 rpm, and incubated for 4 days. The results indicated the production of α-glucosidase inhibitory activity reached the highest point at 80 %.

Effect of carbon source
Lactose producing media showed the highest activity of 86.5 %, followed by other media supplementing with galactose (81.8 %); corn starch (72.2 %). However, glucose and sucrose containing media showed lowest inhibition activity.

Effect of nitrogen
Culture media supplementing with different nitrogen sources exhibited different αglucosidase inhibitory activity (Table 3). Peptone and yeast extract adding media enhance the inhibition activity of the crude extract by 91.5 % and 76.5 %, respectively. In contrast, soybean powder and beef extract supplementing media exhibited a lower inhibition activity of 73.2 % and 73.5 %, respectively, and the lowest one is malt extract containing media. In conclusion, from the above data of optimizing the culture conditions and compositions, we found out the favorite media for enhancement of α-glucosidase inhibitor, which includes: lactose 1 %; peptone 0.5 %; pH 7.0; temperature 37 C; shaking speed of 200 rpm, and 4 days of incubation. At the optimal conditions, the inhibition activity rose to the highest point of 91.5 % which is 1.34-fold higher than pre-optimization. Previously, the production of DNJ increased by 3.3 folds when fermentation media was supplemented with sodium citrate (0 h, 5 g/L), sorbose (0 h, 1 g/L), iodoacetic acid (20 h, 50 mg/L), and glucose (26h, 7 g/L) [14].
The purified efficacy of 1-deoxynojirimycin was listed in the table below The result from table 4 demonstrated the purified efficacy reached 16 %. These fractions were determined on TLC (Figure 5 A). There is an appearance of one band constrained at Rf 0.34 which corresponds to standard DNJ. In 2008, DNJ was isolated from the B. subtilis S10 yielding 0.75 g/L. In the next 2 years, Zhu and his team isolated DNJ from the B. subtilis B2 by loading the fermented broth on active charcoal column and chromatography CM-sepharose column [7].
The MS spectrum data showed that the purified α -glucosidase inhibitor was 1deoxynojirimycin with a molecular weight of 163 Da (Figure 5 B). From the collected evidence, we can confirm that the putative compound was DNJ-alpha glucosidase inhibitor purified from the B. subtilis YT20 strain with molecular weight of 163 Da which is similar to the previous report [7].

CONCLUSION
We have determined the optimal culture media for fermentation of the α-glucosidase inhibitior from the B. subtilis YT20 composed of (g/L): Lactose 1 %; Peptone 0.5 %; KH 2 PO 4 0.05 %; (NH 4 ) 2 SO 4 0.05 %; pH 7.0. The optimal incubated conditions for the biosynthesis of 1deoxynojirimycin was 96 hrs, 37C and shaking speed of 200 rpm. Combination of all the optimal conditions, the α-glucosidase inhibition activity exhibited the highest point at 91 %. The putative 1-deoxynojirimycin was purified by continuous steps including ethanol sedimentation, condensation, activated charcoal column chromatography, sephadex G100 chromatography and TLC. 1-deoxynojirimycin was purified from the B. subtilis YT20 with the yield of 16% (w/w) and confined at Rf 0.34 corresponding to standard 1-deoxynojirimycin (Sigma). This proves that the 1-deoxynojirimycin can be mass produced in Vietnam by the native B. subtilis YT20.