Gene set enrichment analysis of α-amylase and α-glucosidase inhibitors of Cassia glauca

Abstract

Background

The present study aimed to evaluate in vitro α-amylase and α-glucosidase inhibitory activity of various extracts of Cassia glauca, predict the binding affinity of multiple phytoconstituents with both enzymes via in silico molecular docking and identify the probably modulated pathways by the lead hit.

Methods

Different extracts of Cassia glauca i.e. acetone, ethanol, and aqueous extracts were evaluated for α-amylase and α-glucosidase inhibitory activity using in vitro method in which starch and 4-Nitrophenyl β-D-glucopyranoside were used as substrate respectively. Similarly, the docking study was performed using autodock4 to predict the binding affinity of phytoconstituents with α-amylase and α-glucosidase. After docking, ten different poses were obtained for the ligand molecule. Among them, the pose of ligand molecule with the lowest binding energy was visualized in Discovery Studio 2019.

Results and conclusion

Among the multiple extracts, the aqueous extract showed the highest α-amylase (IC₅₀:652.10 ± 20.09) and α-glucosidase (IC₅₀:482.46 ± 8.70) inhibitory activity. Similarly, cassiaoccidentalin B was predicted to have the highest binding affinity with both enzymes. The potency of aqueous extract to inhibit α-amylase and α-glucosidase could be due to multiple water-soluble compounds like saponins, flavonoids, and glycosides.

Introduction

Diabetes mellitus is a metabolic disorder due to impaired metabolism of carbohydrate, protein, and fat which is characterized by hyperglycemia; mainly due to impaired insulin production or action or combination of both. Type 2 diabetes mellitus is a polygenic condition and is the imbalance between sugar intake/metabolism and insulin function [1]. Further, the choice of current therapeutic agents for type 2 diabetes mellitus like biguanides, sulfonylureas, meglitinides, thiazolidinediones, dipeptidyl peptidase IV inhibitors, and α-glucosidase inhibitors also possess the multiple side effects over long term use [2].

Over this, gut enzymes primarily, α-amylase and α-glucosidase also play an important role in the conversion of polysaccharides into monosaccharides. Although, the multiple molecules are identified to inhibit these enzymes they are also associated with numerous side effects i.e. hepatotoxicity and nephrotoxicity [3]. Hence, it is essential to identify the new therapeutic agent in the management of diabetes mellitus including gut enzymes inhibitors with higher therapeutic value and minimal side effects.

Traditional folk medicines can be alternative medicines in the management of endocrinal disorders including diabetes mellitus. Multiple investigations have been made for the utilization of traditional folk medicines and their beneficial effects [4]. Likewise, investigations have been made to report the anti-diabetic action of natural products based formulations in experimentally induced diabetes in animal models [5, 6]. Among them, Cassia glauca belonging to the Fabaceae is identified to have multiple medicinal values including in managing diabetes mellitus [7]. Although numerous investigations have been made for Cassia glauca against diabetes mellitus [7, 8]; minimal data is available to inhibit α-amylase and α-glucosidase enzyme to manage postprandial hyperglycemia. Further, there are limited reports available to predict the binding affinity of multiple phytoconstituents from Cassia glauca with these two enzymes. Hence, the present study aims to investigate the α-glucosidase and α-amylase inhibitory activity of different extracts of Cassia glauca and predict the binding affinity of multiple phytoconstituents for both enzymes.

Materials and methods

Chemicals

α-amylase, α-glucosidase, 4-Nitrophenyl β-D-glucopyranoside (p-NPG) and dinitro salicylic acid (DNS) were purchased from Sigma-Aldrich. Sodium carbonate, sodium dihydrogen phosphate, disodium hydrogen phosphate were purchased from Hi-Media, Mumbai.

Plant collection and extraction

Cassia glauca was collected from local areas of Belagavi and was authenticated by a botanist at the Indian Council of Medical Research- National Institute of Traditional Medicine (ICMR-NITM), Belagavi and the herbarium was deposited for the same (accession number: RMRC-1328). The collected barks were shade dried, turned to a coarse powder and three extracts were prepared i.e. soxhlet extraction using 95% v/v, maceration (macerated using distilled water with suitable care to avoid fungal contamination for seven days) and acetone (soxhlet). After, extraction, the filtrate was collected, concentrated under a rotary evaporator at low temperature and lyophilized. The yield was 18% w/w, 27% w/w, and 12% w/w for ethanol, aqueous and acetone respectively. After extraction the extracts were stored in −20 °C freezer for future use.

α-amylase inhibitory activity

α-amylase inhibitory activity of different extracts was performed as explained by the previous method [9]. Multiple samples of 20 μl extracts (200, 400, 600, 800, and 1000 μg/ml) or acarbose (10, 20, 40, 80, 160, and 320) was incubated with 10 μl α-amylase (2 U/ml) for 20 min at 37 °C followed by adding of 1%w/v starch further incubated to 30 min. Then 100 μl of DNS reagent was added and boiled for 10 min. Then the absorbance was measured at 540 nm in the multi-plate reader. Measurement was also taken for controls without containing test samples for each test. All the experiments were performed in triplicates. IC₅₀ was measured using the following formula.

$$\text{Inhibitory activity}\left(\%\right)=\left(1-\frac{\mathit{As}}{\mathit{Ac}}\right)X100$$

Where, As is the absorbance of extracts/acarbose and Ac is the absorbance without extract/acarbose.

α-glucosidase inhibitory activity

α-glucosidase inhibitory activity of different extracts was performed as explained by the previous method [9]. Briefly, 20 μl of extracts (200, 400, 600, 800, and 1000 μg/ml) and acarbose (10, 20, 40, 80, 160, and 320) were incubated with 10 μl α-glucosidase (1 U/ml), at 37 °C for 15 min followed by addition of 20 μl P-NPG (5 mM) to incubate at 37 °C for 20 min. Then, 50 μl Na₂CO₃ was added and the absorbance was measured at 405 nm using the multi-plate reader. Controls were taken without test samples and all the experiments were performed in triplicates. The percent inhibition of enzyme was measured using the following formula.

$$\text{activity}\left(\%\right)=\left(1-\frac{\mathit{As}}{\mathit{Ac}}\right)X100$$ Inhibitory$$ \mathrm{Inhibitory} $$

Where, As is the absorbance of extracts/acarbose and Ac is the absorbance without extract/acarbose.

In-silico molecular docking

Phytoconstituents present in Cassia glauca were identified from published literature and multiple open-source databases. All the compounds were downloaded from the PubChem (https://pubchem.ncbi.nlm.nih.gov/) database in .sdf format and converted into .pdb format using Discovery studio 2019 and minimized using uff forcefield. α-amylase protein (PDB: 5VA9) was retrieved from Research Collaboratory for Structural Bioinformatics (RCSB) protein bank. The α-glucosidase protein was modeled using a query sequence (accession number: ABI53718.1) and PDB: 5KZW in Modeller 9.10. The hetero-atoms and water molecules present in the proteins were removed using Discovery studio 2019. Docking was carried using autodock4.0. After docking the pose scoring lowest binding energy was chosen to visualize ligand-protein interaction using Discovery studio 2019.

Enrichment and network analysis

SMILES of cassiaoccidentalin B was queried in DIGEP-Pred (http://www.way2drug.com/GE/) at the probable activity (Pa) > 0.5 to identify the probably regulated protein/genes. The identified targets were queried in STRING (https://string-db.org/) to evaluate the protein-protein interaction and the pathways were identified concerning the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway. The network was constructed among the proteins, compounds, and pathways using Cytoscape version 3.5.1 (https://cytoscape.org/).

Statistical analysis

All the in vitro tests were performed in triplicates and values were expressed in mean ± SD and IC₅₀ was calculated using GraphPad Prism 5 version 5.01 (Graph pad software, Inc., La Jolla, CA, USA.) statistical software.

Results

α-amylase and α-glucosidase inhibitory activity

Among the three extracts, the aqueous extract showed the highest α-amylase inhibitory activity (IC₅₀:652.10 ± 20.09) and α-glucosidase inhibitory activity (IC₅₀:482.46 ± 8.70). The inhibitory activity of each extract at individual concentrations is summarized in Fig. 1. Similarly, Table 1 summarized the IC₅₀ of each tested extracts.

Fig. 1.

a α-glucosidase and b α-amylase inhibitory activity of Cassia glauca bark extracts

Table 1.

IC₅₀ of different extracts of Cassia glauca

Test Samples	Inhibitory concentration (IC50)
	α-glucosidase	α-amylase
Ethanolic extract	698.33 ± 27.919	833.20 ± 24.32
Aqueous extract	482.46 ± 8.70	652.10 ± 20.09
Acetone extract	840.42 ± 42.46	761.80 ± 23.72
Acarbose	143.72 ± 5.59	116.10 ± 1.68

In silico molecular docking

Docking study showed cassiaoccidentalin B to possess the highest binding affinity with α-glucosidase (−9.4 kcal/mol) with four hydrogen bond interactions with VAL544, TYR543, and ALA93 amino acid residues. Although kaempferol 3-β-gentiobioside scored −9 kcal/mol binding energy, it shared the highest number of hydrogen bond interaction with ASP91, LYS96, TRP126, and ILE98. Similarly, cassiaoccidentalin B was predicted to have the highest binding affinity (−9.2 kcal/mol) with α-amylase by via six hydrogen bond interactions with HIS299, ASP197, THR163, and ASP356. Table 2 summarizes the binding affinity of each compound with α-amylase and α-glucosidase along with their hydrogen bond interactions. Figure 2 represents the best pose interaction of cassiaoccidentalin B with α-amylase and α-glucosidase.

Table 2.

Binding affinity of phytoconstituents of Cassia glauca with α-glucosidase and α-amylase

Ligand	α-glucosidase			α-amylase
	Binding energy (kcal/mol)	Number of hydrogen bonds	Hydrogen bond residues	Binding energy (kcal/mol)	Number of Hydrogen Bonds	Hydrogen Bond Residues
3, 5-di-O-caffeoyl quinic acid	−8.6	6	ARG594, VAL867, GLU866, MET363, HIS717, ARG608	−8.6	6	ARG303, ASP356, HIS305, GLU233, ASP197
3-O-methylcalopocarpin	−7.9	–	–	−8	2	SER289, ARG421
Cassiaoccidentalin B	−9.4	4	VAL544, TYR543, ALA93	−9.2	6	HIS299, ASP197, THR163, ASP356
Cassiarin A	−7.2	1	CYS127	−7.3	2	ASP300, GLN63
Cassiarin B	−7.8	3	GLN124, TRP126, TYR110	−7.7	–	–
Cinnamic acid	−6.1	3	ASP645, ARG672, TRP613	−6	1	GLN63
Diphenyl sulfone	−6.9	3	GLU866, VAL867, ARG594	−7	1	THR163
Epiafzelechin	−8.7	2	ASP91, ALA93	−8.9	2	HIS201, ASP197
kaempferol 3-β-gentiobioside	−9	6	ASP91, LYS96, TRP126, ILE98	−8.7	4	GLU233, TRP59, ASP300, HIS305
Methyl 3, 5-di-O-caffeoyl quinate	−9.1	5	ASP95, GLN121, TYR110, CYS127, ASP91	−8.9	4	HIS305, GLN63, GLU233
Sandwicensin	−7.9	3	ARG608, VAL867, LEU868	−8.3	2	ASP197, GLU233

Fig. 2.

Interaction of cassiaoccidentalin B with (a) α-amylase and b α-glucosidase

Enrichment and network analysis

Cassiaoccidentalin B was predicted to downregulate CHEK1 and MMP7 and upregulate PLAU, TNFRSF1A, PLAT, TIMP1, NPPB, CBR1, SMN2, TP73, and AR. Similarly, KEGG identified modulation of complement and coagulation cascades, fluid shear stress and atherosclerosis, NF-kappa B signaling pathway, p53 signaling pathway, prostate cancer and transcriptional misregulation in cancer pathways. Among them, the p53 signaling pathway was identified to be involved in the pathogenesis of diabetes mellitus concerning the KEGG database. The interaction of cassiaoccidentalin B with proteins and regulated pathways is represented in Fig. 3.

Fig. 3.

Network interaction of Cassiaoccidentalin B with targets and respective pathways

Discussion

The present study was aimed to investigate the α-amylase and α-glucosidase inhibitory activity of different extracts of Cassia glauca via in vitro methods. Further, we also predicted the binding affinity of multiple phytoconstituents from the Cassia glauca with both enzymes using in silico molecular docking studies.

Currently, multiple herbal medicinal formulations/plants are also used to manage various metabolic disorders including diabetes mellitus. The ethnopharmacological study reports more than 1200 plants for anti-diabetic activity which are being utilized in the management of diabetes mellitus reflecting the importance of traditional folk medicines [10]. Among them, many plants are also α-amylase and α-glucosidase inhibitors. Enzymes α-amylase and α-glucosidase are responsible for postprandial hyperglycemia via the cleavage of 1,4-glycosidic linkages converting the polysaccharides into mono-saccharides [11, 12]. The present study identifies the multiple extracts; primarily aqueous extracts to inhibit α-glucosidase and α-amylase enzyme which could be due to the presence of multiple water-soluble phytoconstituents including flavonoids and glycosides.

In silico molecular docking utilizes the computational tools which are robust and are helpful to predict the binding affinity of multiple phytoconstituents with respective targets [13–15]. Further, it helps to identify the lead hit molecule which can be further investigated via experimental studies. The previous study also identifies the lead phytoconstituents from various plants as potential lead molecules in multiple diseases via the utilization of in silico tools [16–19]. In the present study, we docked eleven different phytoconstituents which were retrieved from the open-source database and published literature. Among them, cassiaoccidentalin B was predicted to possess the highest binding affinity with both enzymes which can be further investigated for gut enzyme inhibitory activities. Further, the potency of aqueous extract to inhibit α-glucosidase and α-amylase inhibitory activity could be due to cassiaoccidentalin B and related compounds that need to be further investigated.

The previous report suggests that α-glucosidase inhibitors from natural sources may not be limited within the gastrointestinal tract but may get absorbed into the systemic circulation and modulate the multiple proteins involved in the progression of diabetes pathogenesis [19]. Although the present study reports the α-glucosidase inhibitory activity of Cassia glauca, the phytoconstituents from extract may get absorbed from the intestine and modulate other metabolic pathways involved in the diabetes mellitus which needs to be further investigated via network pharmacology approach [20].

Since cassiaoccidentalin B was predicted to have the highest binding affinity over both enzymes, we investigated its probable ability to modulate other proteins and pathways; identified one of the pathways to be involved in the pathogenesis of diabetes mellitus i.e. p53 signaling pathway which is involved in maintaining the pancreatic β- cell mass [21] and is also responsible in regulating glycolysis, gluconeogenesis, pentose phosphate pathway, and autophagy which are the main contributors in the diabetes mellitus [22]. In the present study, we identified one of the probable lead molecules to inhibit α-glucosidase enzyme i.e. cassiaoccidentalin B to modulate the p53 signaling pathway in diabetes mellitus. However, plant extracts compose the multiple phytoconstituents which can interact with multiple protein molecules to regulate the numerous protein molecules; can be evaluated via the network pharmacology approach as described previously [20, 23]; can help to propose the probable molecular mechanism of anti-diabetic action of aqueous extract of Cassia glauca.

Conclusion

The α-amylase and α-glucosidase inhibitory activity of Cassia glauca extracts; primarily aqueous extract could be due to the presence of cassiaoccidentalin B which needs to confirm via experimental protocols. Although the study identified the lead hit i.e. cassiaoccidentalin B as α-glucosidase and α-amylase inhibitor and aqueous extract as a potential inhibitor of both enzymes; the mode of inhibition needs to be still investigated which is the future scope of the study.

Authors’ contributions

SGT performed the work and drafted the manuscript, MBP and IP designed and reviewed the manuscript and PK reviewed and finalized the manuscript draft.

Funding

This work did not receive any funds from national and international agencies.

Data availability

Data for this work are freely available from Dr. Ismail Pasha upon request.

Compliance with ethical standards

Conflict of interest

The authors declared that they have no conflict of interest

Consent for publication

Not Applicable

Ethics approval and consent to participate

Not Applicable