Characterization and evaluation of mycosterol secreted from endophytic strain of Gymnema sylvestre for inhibition of α-glucosidase activity

Endophytic fungi produce various types of chemicals for establishment of niche within the host plant. Due to symbiotic association, they secrete pharmaceutically important bioactive compounds and enzyme inhibitors. In this research article, we have explored the potent α-glucosidse inhibitor (AGI) produced from Fusarium equiseti recovered from the leaf of Gymnema sylvestre through bioassay-guided fraction. This study investigated the biodiversity, phylogeny, antioxidant activity and α-glucosidse inhibition of endophytic fungi isolated from Gymnema sylvestre. A total of 32 isolates obtained were grouped into 16 genera, according to their morphology of colony and spores. A high biodiversity of endophytic fungi were observed in G. sylvestre with diversity indices. Endophytic fungal strain Fusarium equiseti was identified through DNA sequencing and the sequence was deposited in GenBank database (https://ncbi.nim.nih.gov) with acession number: MF403109. The characterization of potent compound was done by FTIR, LC-ESI-MS and NMR spectroscopic analysis with IUPAC name 17-(5-ethyl-6-methylheptan-2-yl)-10,13-dimethyl-2,3,4,7,8,9,10,11,12,13,14,15,16,17-tetradecahydro-1H-cyclopenta[a] phenanthren-3-ol. The isolated bioactive compound showed significant α-amylase and α-glucosidase inhibition activity with IC50 values, 4.22 ± 0.0005 µg/mL and 69.72 ± 0.001 µg/mL while IC50 values of acarbose was 5.75 ± 0.007 and 55.29 ± 0.0005 µg/mL respectively. This result is higher in comparison to other previous study. The enzyme kinetics study revealed that bioactive compound was competitive inhibitor for α-amylase and α-glucosidase. In-silico study showed that bioactive compound binds to the binding site of α-amylase, similar to that of acarbose but with higher affinity. The study highlights the importance of endophytic fungi as an alternative source of AGI (α-glucosidase inhibition) to control the diabetic condition in vitro.


Results
Biodiversity of endophytes isolated from Gymnema sylvestre. Different parts of a medicinal herb G. sylvestre were explored for fungal endophytes and total 16 fungal groups were isolated. A total of 32 fungal isolates of which 16 isolated from leaf, 11 from stem and 5 were isolated from the root of G. sylvestre. Strains were identified on the basis of culture characteristic, microscopic studies and spore morphology. A relative frequency of Fusarium sp. was found to be highest while three groups Xylaria sp., Glomastis sp., Aspergillus sp. were found to be in moderate range and remaining were in low frequency (Fig. 1A). Species richness was found to be highest in leaves in comparison to other parts of the plant. The biodiversity of fungal endophytes isolated from different tissues were evaluated by various diversity indices such as Simpson's diversity index (1-D), Simpson's dominance index, Species richness (Dmn), Shannon-Wiener index (H) and Evenness. The highest tissue-specific fungal dominance was found in the root (0.2778) then in stem (0.1074) and least in leaf (0.0859). Fusarium sp. was most dominant which was isolated from leaves and stems. The Shannon and Simpson's indices, respectively, indicated consistency and a high certainty of endophytic fungal species in the root (1.33). Species richness indicates highly diverse and taxonomically rich fungal endophytes i.e. in leaves (13). Species evenness is uniform in leaves and roots while it is slightly higher (0.96) in stems. These diversity indexes represent the significant of endophytes within and between the different tissues of G. sylvestre (Fig. 1B).

Screening of endophytes for antidiabetic activity and bioactivity guided fractionation. After
isolation of fungal endophytes from G. Sylvestre, we initiated comprehensive screening to find the potent fungal endophytes having antidiabetic activity such as α-amylase and α-glucosidase inhibitors. Broth cultures of different fungal isolates were evaluated in different organic solvent to find the antidiabetic activity and suitable organic solvent for further study.
Among 32 isolates, one Fusarium sp. extracted in ethyl acetate and chloroform was found as active inhibitor of porcine pancreas α-amylase (EC 3.2.1.1) and α-glucosidase (EC 3.2.1.20) from Saccharomyces cerevisiae. Those isolate having antidiabetic activity were selected to identify and characterize the bioactive compound. A total of 32 fungal isolates of G. sylvestre of which one isolate of Fusarium sp. isolated from leaf tissue of G. sylvestre was recorded as an incidental rare strain (1/32 isolates).
The chloroform soluble fraction obtained through silica gel vacuum liquid chromatography was more active than ethyl acetate extract. Potent fraction of chloroform extract of F. equiseti was refractionated through HPTLC. Five separate fractions were obtained of which one sub fraction exhibited high α-amylase and α-glucosidase inhibition with IC 50 values, 4.22 ± 0.0005 and 69.72 ± 0.001 µg/mL respectively. While IC 50 values of acarbose against α-amylase and α-glucosidase were 5.75 ± 0.007 and 55.29 ± 0.0005 µg/mL respectively. Identification and chacterization of potent antidiabetic endophytic strain. The morphological identification was done by microscopic studies, culture characteristics and spore morphology ( Fig. 2A,B). The molecular identification was done by DNA sequencing. The obtained fungal sequence was deposited in GeneBank database (https://ncbi.nim.nih.gov) with accession number MF 403109. The phylogenetic analysis involved 70 nucleotide sequences of Fusarium sp., phylogenetic tree was constructed using NJ based ITS sequences with more than 92% similarity. The maximum likelihood estimate of gamma parameter for site rates was done with MEGA6. A high degree of genetic diversity among Fusarium sp. was observed in phylogenetic analysis (Fig. 3). A potent fungal strain Fusarium equiseti 'SKS01' was isolated from the leaf of G. sylvestre. chemical characterization of α-glucosidse inhibitor (AGi). In IR spectrum, peak showed O-H Stretching vibrations at 3621.2 cm −1 which represent alcoholic group however, three peaks were obtained in hydrogen stretching region (3703.42, 3419.83 and 3338.28 cm −1 ). Three medium to strong peaks were obtained at 2956.27, 2922.27 and 2853.09 cm −1 were due to aliphatic C-H vibrations, fall in between region 2925 and 2850 cm −1 (Fig. 4A). The double bond region (1950-1550 cm −1 ) -C=O stretching vibration is characterized by the absorption at 1711.57 cm −1 , indicating the presence of double bond in cyclohexane. C-C stretching vibrations occurred at 1634.64 while peak 1461.30 indicate aliphatic structure with bending vibration, assigned as alkane while peak at 1176.93 cm −1 indicate amine/ester/tertiary alcohol and vinyl group (Fig. 4B,C).  1 H NMR (500 MHz, CDCL 3 Bruker, Switzerland), δH: 1.378-0.901 (-CH 3 ), δH: 3.334 (-CH 2 ), δH: 1.582 (alcoholic -OH) δ = 7.285-7.231 (Ar-H); δ = 3.559-3.514 (aliphatic CH) were detected and the structure was further confirmed by 13 C NMR spectroscopy.
Enzyme kinetics of active fraction of G. sylvestre and mycosterol. The mode of inhibition of α-amylase and α-glucosidase was determined using the extract of G. sylvestre and mycosterol from F. equiseti using Lineweaver-Burk plot as shown in Figures 6 and 7. The present study revealed mycosterol derived from bioactive subfraction of F. equiseti as competitive inhibitor of α-amylase which binds strongly to active site of enzyme while uncompetitive inhibitor of αglucosidase which is likely to inhibit through binding enzyme-substrate complex. The extracts of G. sylvestre also showed uncompetitive inhibition of α-amylase but appeared as non-competitive for α-glucosidase. The non-competitive inhibitor is likely to show inhibition by binding to the allosteric site of enzyme other than the active site. Moreover, acarbose was found to be the competitive inhibitor in case of both the enzymes and was used as a reference standard. The kinetics of enzyme (Vmax, www.nature.com/scientificreports www.nature.com/scientificreports/ Km), IC 50 value for the inhibitor and their mode of inhibition was studied using Lineweaver Burk pot. The Km value for starch as the substrate was found to be 0.85 (mM) and Vmax 590.40 (µMmin −1 ) in case of α-amylase. In addition, Km value for p-NPG as the substrate was found to be 2.82 (mM) and Vmax 150.62 (µMmin −1 ) in case of α-glucosidase.
Cytotoxicity assay. The cytotoxicity study of bioactive fraction of plant extract and mycosterol of fungus was done using L929 cells through MTT assay to evaluate the toxicity. The cytotoxicity assay showed that the bioactive fraction of plant extract and mycosterol of fungus had very low cytotoxicity over a concentration range of 0 to 400 μg/mL. Cellular viability was minimally affected even at high concentration (70% cellular viability at 100 μg/mL for bioactive fraction of plant extract and 78.15% at 50 μg/mL for mycosterol of fungus) (Fig. 8).

Discussion
Endophytes are 'the tiny factories of nature' which produce various bioactive secondary metabolites and has a capability to encode similar type of metabolites as synthesized by their host plant 21 . These bioactive compounds have a broad spectrum of biological activities such as antimicrobial, antioxidant, antidiabetic and anticancer. These bioactive compounds are classified into many categories such as alkaloids, phenol, steroids, terpenoids and lignin 8 . Some steroid derivative showed hypoglycaemic activity by decreasing the blood glucose levels in diabetic rat 22 .
The richness of endophytes depends upon various factors of environment and habitat. In previous study, 11 fungal groups were isolated in summer session 3 . In winter session, out of 16 fungal groups, one potent fungal strain F. equiseti (MF 403109) was deposited in NCBI Genebank and published in Pubmed. Phylogenetic www.nature.com/scientificreports www.nature.com/scientificreports/ analysis indicates that potent strain is ancient in origin with less mutation acorss the evolutionary lineage. FTIR, LC-ESI-MS and NMR spectroscopic analysis characterized this purified compound as mycosterol. By comparing with the previous reports, compound was identified as mycosterol [23][24][25] . The MS/MS spectra and the proposed fragmentation pattern of mycosterol is shown in Fig. 5A. The IC 50 value of sub fraction of chloroform of G. sylvestre against α-amylase is 10.47 ± 0.0005 µg/mL and it is better as compared to extract of leaves of Petalostigma banksii and P. pubescens (IC 50 value of 166.50 ± 5.50 μg/mL and 160.20 ± 27.92 μg/mL, respectively) 26 and also from methanolic extract of Phyllanthus virgatus (IC 50 value of 33.20 ± 0.556 μg/mL) 27 .
The bioactive compound of F. equiseti exhibited significant inhibitory activity against α-amylase with IC 50 value 4.22 ± 0.0005 µg/mL and it is more potent than Alternaria longipes strain VITN14G isolated from Avicennia officinalis (IC 50 value of 27.05 µg/mL) 28 . The IC 50 value of sub fraction of chloroform of G. sylvestre against α-glucosidase is 85.73 ± 0.001 µg/mL and it is better as compared to extract of leaves of Petalostigma pubescens (IC 50 value of 167.83 ± 23.82 μg/mL) 26 and ethyl acetate fraction of Cornus capitata (IC 50 value of 50 μg/mL) 29 . The bioactive sub fraction derivative of F. equiseti exhibited significant inhibitory activity against α-glucosidase with IC 50 value 69.72 ± 0.001 µg/mL and is better than ethyl acetate fraction of Phlomis tuberose (IC 50 value of 100 μg/mL) 30 . Result of MTT assay also showed that potent compound is non-cytotoxic.
In-silico study showed that mycosterol binds to the binding site of α-amylase and α-glucosidase similar to that of acarbose but with high affinity. The prominent feature of α-amylase is the presence of three extremely    www.nature.com/scientificreports www.nature.com/scientificreports/ conserved and catalytic residues such as Asp197, Asp300 and Glu233 in the active site pocket. In previous study, similar types of hydrophobic interactions and hydrogen bonding have been found in crystal structures of α-amylase with different inhibitor molecules. In 2018, Sohretoglu et al. suggested that hydroxyl group of inhibitor increases the inhibition activity 31 . In this study, the isolated compound possess a hydroxyl group at the C-3 position of first carbon ring which may increase the activity of α-amylase and α-glucosidase inhibition. Computational docking analysis also revealed that compound is competitive inhibitor for α-amylase. Its binding conformation was similar to that of acarbose. The synthetic steroidal drug tibolone is transformed by fungus Fusarium lini, which inhibit α-glucosidase 32 . Similarly this F. equiseti may also be transformed to synthesize potent antidiabetic bioactive sub fraction derivative on large scale in lesser span of a time. Additionally, recently two glibenclamide-pregnenolone derivatives with hypoglycaemic activity were prepared 33 .

conclusion
After isolation of fungal endophytes from G. sylvestre, the comprehensive screening to find the potential fungal endophytes having antidiabetic activity as α-amylase and α-glucosidase inhibitior was carried out. In this study, we have isolated and identified potent endophytic fungal species F. equiseti from the leaves of G. sylvestre. The bioactive compound was isolated and identified by FTIR, LC-ESI-MS, 1 H NMR and 13 C NMR. This mycosterol showed the similar characteristic appeared in phytosterol such as β-sitosterol. The known bioactive compound mycosterol was authenticated by comparing their NMR data with those of reported previously. This isolated mycosterol, exhibited significant α-amylase and α-glucosidase inhibition activity. Kinetics study also confirmed the competitive mode of inhibition to α-amylase and uncompetitive mode of inhibition to α-glucosidase. The biopharmaceutical importance of extract of F. equiseti (MF 403109) was further established by the cytotoxic activity against mice fibroblast cell line (L929) ; our MTT results indicated that it is safe even at high doses. In-silico studies revealed that mycosterol is a competitive inhibitor for α-amylase and α-glucosidase and binds to the active site similar to that of acarbose but with high affinity due to the presence of hydroxyl group at the C-3 position of the first carbon ring of the compound. isolation and culture of endophytes. All small pieces (1.0 × 1.0 cm) were washed under running tap water and were dried under aseptic conditions. The margin of small pieces of leaves, roots and stems were cut and initially surface treated with 70% ethanol for 1 min to eliminate the epiphytic microorganisms. Thereafter, sterilization of tissue with aqueous sodium hypochlorite (4% available chlorine) for 3 min and then rinsed in 70% ethanol for nearly 30 seconds. before a final triple rinsing in sterilized double distilled water, later the tissue was air dried. The concentration of 100 µg/mL of streptomycin was added in potato dextrose agar (PDA) media to prevent the bacterial contamination. Each culture plate contain four segments of tissue and plates were sealed with parafilm and was incubated in a BOD incubator in 12-hours light and dark cycle at 27 ± 2 °C for 1 week. After 1 week, actively growing hyphal tips of fungi were then sub cultured into new PDA media. www.nature.com/scientificreports www.nature.com/scientificreports/ Extraction and purification of metabolites of endophytic fungi and G. sylvestre. The pure strain of isolated fungi was grown in 500 ml erlenmeyer flasks containing 200 ml potato dextrose broth and incubated in BOD cum orbital shaker for 2 weeks at 28 ± 2 °C with periodical shaking at 240 rpm. The mycelium of fungal culture was removed by filtration and broth was then extracted gradually with heptanes, chloroform and ethyl acetate in separating funnel for three times. The organic phase was evaporated to dryness under reduced pressure using a rotary evaporator to constitute the crude broth extract. The crude extract was lypholized in lypholizer and was stored at −20 °C.
Morphological identification of the endophytic fungal Isolates. The endophytic fungi were identified according to their macroscopic characteristics such as colony morphology and spore morphology under Light microscope and also by Scanning Electron Microscopy (SEM). Each identified endophytic fungus was assigned with specific code numbers and maintained in cryo-vials on Potato Dextrose Agar media layered with glycerol (15%,v/v) and also in a lyophilized form. All Fungal strains were stored at −20 °C for future use. isolation of total genomic DnA. The culture plate was washed with 1X PBS, 1 mL of RiboZol per 10 cm 2 of culture disc area was added further. The cells were lyzed by continuous pipetting and all substance were transferred into a nuclease free tube. The cells were homogenized and were incubated for 5-10 minutes at RT. 200 µL of chloroform per 1 mL RiboZol was added, tube was further shaken vigorously for 15 seconds to mix the sample, later it was incubated for 2-3 minutes at RT. Sample was centrifuged at 12,000 rpm for 15 minute at 4 °C. www.nature.com/scientificreports www.nature.com/scientificreports/ Three separated phases were observed. Aqueous phase was removed and 0.3 mL ethanol per mL of RiboZol was added, samples were mixed well and were incubated for 3 minutes at 15-30 °C. Mix was centrifuged at 12,000 rpm for 5 minutes at 4 °C, later the supernatantwas discarded. The bottom of the tube contained DNA which was further washed by adding 0.1 M sodium citrate/10% ethanol (1 mL per ml RiboZol) to the pellet. The resulting mix was further incubated for 30 minutes at 15-30 °C; after that it was centrifuged at 12,000 rpm for 5 minutes at 4 °C. DNA washing step was repated, pellet was resuspended in 75% ethanol, incubated for 10-20 minutes at 15-30 °C with intermittent mixing. Mix was further centrifuged at 12,000 rpm for 5 minutes at 4 °C, pellet was air dried for 5-10 minutes. Pellet was further dissolved in 1X TAE buffer solution and the mix was stored for further experiments. In-silico study for sequencing alignment and phylogenetic analysis. The in-silico BLAST tool was used to compare the specificity of the selected PCR primers to amplify Fusarium equiseti among other species. The ITS sequences of Fusarium sp. of 70 strains were downloaded from the NCBI GenBank database and these were used as reference sequences in the phylogenetic analyses. All these sequences were aligned with the program MUSCLE3.7. The resulting aligned sequences were cured using the program G blocks 0.91b. Finally, the program PhyML3.0 aLRT was used for tree building. The program Tree Dyn198.3 was used for tree rendering. The DNA sequences thus obtained were submitted to the ribosomal gene database (https://ncbi.nim.nih.gov) and the sequences were aligned to identify the fungus. This multiple-alignment file was used for phylogenetic analysis which was performed using Mega 6 with Neighbor-Joining method.
fourier transform infra red spectroscopy (ftiR). Chemical characterization of potent fraction was done by FTIR, NMR and ESI mass. FTIR is a well-known tool for detection of presence of functional groups. The bioactive subfraction (100 mg) was subjected to FTIR analysis (JASCO 1400, JASCO, Tokyo, Japan). In this analysis, sample was prepared in potassium bromide discs and scanned within the range of 500-4000 cm −1 . The absorbances of molecular vibrations under IR radiation are proportional to the abundance of the functional groups. nuclear magnetic resonance (nMR) spectroscopy. NMR spectrum of compound was obtained on a Bruker (500 MHz) NMR (Bruker, Switzerland) at a constant temperature, controlled and adjusted to room temperature. The chemical shifts were shown in δ values (ppm) with tetra-methylsilane as an internal standard. The deuterated chloroform was used as the solvent for recording of 1 H and 13 C NMR spectra. α-amylase inhibition assay. The activity of α-amylase was carried out using starch as a substrate as described by Khare and Prakash 34 . Reaction mixture (1.0 ml) containing starch was dissolved in sodium phosphate buffer (100 mM, pH 6.9), suitably diluted porcine pancreatic α-amylase was incubated at 27 °C for 3 min. The reaction was terminated by adding 1.0 ml of 3,5-dinitrosalicylic acid solution followed by heating the reaction mixture in a boiling water bath for 5 min and subsequently cooling down to room temperature. Thereafter, 10 ml of double distilled water was added and the amount of reducing sugar (maltose) produced was quantitated using spectrophotometer at 540 nm. For α-amylase inhibition assay, 250 μl of enzyme dissolved in sodium phosphate buffer (100 mM, pH 6.9) was added to 500 μl of test sample containing plant (10.47 µg/ml) or fungal extract (4.22 µg/ml) concentration respectively, followed by incubation at 37 °C for 30 minutes. To this, 250 μl of starch of varying concentration of 0.25-8 mg/mL dissolved in sodium phosphate buffer (100 mM, pH 6.9) was added. Further, reaction mixture (1 mL) was assayed for enzymatic activity as described under standard assay conditions. In case of control, inhibitor was not added to the reaction mixture and the enzyme assay was carried out accordingly. Acarbose (5.75 µg/ml) was used as the reference standard inhibitor. www.nature.com/scientificreports www.nature.com/scientificreports/ αglucosidase inhibition assay. The α-glucosidase inhibition test was carried out in 96-well microplate using a modified procedure of Bilal et al. 35 . The reaction mixture contained 50 μL of sodium phosphate buffer (100 mM, pH 6.8), 10 μL of α-glucosidase and 20 μL of plant (85.73 µg/ml)/fungal extract (69.72 µg/ml) as test sample; samples were thoroughly mixed in a 96-well microplate and were incubated at 37 °C for 30 minutes. However, in case of control, inhibitor was not added. The reaction was initiated by adding 20 μl of pNPG of varying concentration of 0.25-8 mg/mL, prepared in sodium phosphate buffer (100 mM, pH 6.8) as a substrate and incubated for another 15 minutes at 37 °C. The reaction was terminated by adding 50 μl of sodium carbonate. The absorbance was recorded at 405 nm by a 96-well micro plate reader (Bio-Rad, India). Acarbose (55.29 µg/ml) was used as a reference standard inhibitor.

Liquid Chromatography-Electrospray Ionization-Mass
The percentage of inhibition of both enzymes was calculated using the following equation: where A control is the activity of enzyme without the inhibitor and A sample is the enzyme activity with the test sample solution as inhibitor at different concentrations. The amount of the inhibitor required for inhibiting 50% of the enzyme activity under the standard assay conditions was used to determine IC 50 value.
Kinetics of inhibition against α-glucosidase. The mode of inhibition of the bioactive fractions of plant and fungal extracts against α-amylase and α-glucosidase activity were determined using Lineweaver-Burk plot analysis with increasing concentrations of substrate i.e. starch and p-nitrophenyl-α-D-glucopyranoside in the absence (control) and presence of the inhibitor respectively.
Cytotoxic assay of mycosterol. The L929 cells were cultured in Dulbecco's modified Eagle's medium (DMEM) supplemented with 5% fetal bovine serum (FBS) under a humidified atmosphere of 5% CO 2 and 95% air at 37 °C. The cytotoxicity of the bioactive fractions of plant and fungal extracts was assayed using L929 cells, mouse fibroblast cell line by MTT (3-[4,5-dimethyl thiazol-2-yl]-2,5-diphenyl tetrazolium bromide) method. A fixed volume of 100 μL, with a density of 2 × 10 4 cells/ml cell suspension was seeded into each well of 96-well microplate plates, after 24 hours the old media with cell debris was replaced with fresh media, extracts and pure fractions were added in different concentrations in triplicate, incubated for 48 hours. The negative control for the experiment was untreated cells, while the cells treated with 10% DMSO were used as a positive control. Further, the culture media was replaced with 50 μL of MTT solution (0.5 mg/mL) in each well and was incubated for 4 hour at 37 °C, followed by removalof supernatant and solubilisation of water insoluble violet formazan crystals in 100 μL of DMSO in each well. The absorbance was measured by microplate reader (Bio-Rad, India) at 570 nm 36 , results are shown in Fig. 9A,B.
Statistical analysis. The data represented in the study are the mean values of three replicates (n = 3) and shown as means ± standard deviation (SD). One Way Anova was used for comparison between groups (p < 0.05). The customized module of GraphPad was used for IC 50 values through nonlinear regression, comprising a dose-response inhibition.
In-silico study of bioactive fraction of fungal extract against α-amylase and α-glucosidase enzymes. In molecular docking analysis, the SDF file of potent fraction was converted into PDB (PDB ID: 1HNY and 5NN3) by using the ChemDraw software. The molecular docking study of compound was done by using PatchDock online server. The results obtained by PatchDock were ranked according to geometric shape, surface patch matching and complementarily score after molecular shape representation. For visualisation and determining the mode of interaction between the receptor and ligands, Discovery Studio 4.0 Client was used 37 .