Design and synthesis of novel pyrazole-benzofuran hybrids: in vitro α-glucosidase inhibitory activity, kinetic and molecular modeling study

In this work, new derivatives of biphenyl pyrazole-benzofuran hybrids designed, synthesized and evaluated in vitro through enzymatic assay for inhibitory effect against α-glucosidase activity. Newly identi�ed inhibitors were found to be four to eighteen folds more active with IC 50 values in the range of 40.6 ± 0.2–164.3 ± 1.8 µM, as compared to the standard drug acarbose (IC 50 = 750.0 ± 10.0 µM). Limited Structure-activity relationship was established. A kinetic binding study indicated that most active compound 8e acted as the competitive inhibitors of α-glucosidase with Ki = 38 µM. Molecular docking has also been performed to �nd the interaction modes responsible for the desired inhibitory activity. As expected, all pharmacophoric features used in the design of the hybrid, are involved in the interaction with the active site of the enzyme. In addition, molecular dynamic simulations showed compound 8e oriented vertically into the active site from mouth to the bottom and stabilized the enzyme domains by interacting with the interface of domain A and domain B and the back side of the active site while acarbose formed non-binding interaction with the residue belong to the domain A of the enzyme.


Introduction
α-glucosidases (EC 3.2.1.20)is a membrane-bound enzyme in the brush border of the intestine which hydrolyses oligosaccharides and polysaccharide to D-glucose, as only monosaccharides can be absorbed from the intestinal lumen 1 .Inhibition of α-glucosidase can retard the release of glucose from complex carbohydrates and therefore can be an important strategy to control hyperglycemia in type-2 diabetes.In addition, the role of cellular α-glucosidase in carbohydrate processing caused that the inhibitors of α-glucosidase also regarded as a convincing therapeutic target for the development of novel drugs for the treatment of numerous diseases including cancer and viral infections [2][3][4] .Accordingly, acarbose (Glucobay), miglitol (Glyset), and voglibose (Volix, Basen) are the commercially available αglucosidase inhibitors and are recommended as rst-line drugs for the treatment of type-2 diabetes 5 .Due to the sugar-based structure of mentioned α-glucosidase inhibitors, their synthesis needs complicated multistep procedures 6 .Unfortunately, administration of these inhibitors is associated with undesirable side effects including diarrhea, abdominal discomfort, and atulence 7,8 .Thus, it would be essential to develop novel, safe, and e cient α-glucosidase inhibitors as an effective lead candidate for future antidiabetic drug discovery initiatives.
Pyrazole and its derivatives are one of the most important heterocyclic framework in medicinal chemistry possessing a broad spectrum of pharmacological activities like anti-in ammatory, anti-tubercular, antitumor, antibacterial, anti-Alzheimer's, antileishmanial, and antihypertensive activities 9 .Over the last years, several well-known drugs that possess pyrazole scaffold like Celecoxib, Viagra, Fipronil, etc., are in clinical use as therapeutic agents [10][11][12] .Therefore pyrazole nucleus served as a valuable candidate for the exploration of lead molecules.Historically, different derivative of substituted pyrazole has been introduced as anti hypoglycemic agent 13 .Also recently, several pyrazole-containing agents have been reported as potent antidiabetic [14][15][16] and hypoglycemic [17][18][19] agents.For instance, Munawar et al.
reported a new series of imidazolylpyrazole derivatives as potent α-glucosidase inhibitor (Fig. 1 compound A) 20 and Xiong et al. discovered the pyrazole-containing derivatives as highly potent and selective glucagon receptor antagonist (Fig. 1 compound B) 21 .It is noteworthy that "Teneligliptin", antidiabetic drug containing pyrazole, was approved for the treatment of type-II diabetes 22 .In addition to the success of this scaffold in the chemical class of anti-diabetic reagents, signi cant metabolic stability and pharmacological e ciency of pyrazole-based antidiabetic agents encouraged us to further study pyrazole scaffolds to develop a new agent 23 .
On the other hand, interesting physiological and chemotherapeutic properties of benzofuran scaffolds make them promising pharmacophore to design and develop new potentially useful therapeutic agents 24 .Moreover, several benzofuran derivatives with α-glucosidase inhibitory activity have been reported.Mphahlele et al. design and synthesis 2-arylbenzofuran-selanadiazole hybrids and evaluated them in vitro through enzymatic assays for inhibitory effect against α-glucosidase (Fig. 1 compound C) 25 .Dimer structures of nature-mimicking hydroxybenzofuran methyl ketones have been evaluated for antidiabetic activity through inhibition of protein tyrosine phosphatase-1B (PTP-1B), which is a legitimate target for the treatment of Type 2 diabetes (Fig. 1 compound D) 26 .Spasov et al. have reported 2-acylbenzofurans as potent α-glucosidase inhibitors (Fig. 1 compound E) 27 .
Over the years, molecular hybridization has been extensively used as the most e cient strategy for the design of novel α-glucosidase inhibitors.It is expected that pharmacophoric hybridization facilitate the development of new compounds with improved a nity and e cacy.On the other hand, previous studies have shown that p-stacking and hydrophobic effect play key roles in promoting inhibitory activity of the new compounds against α-glucosidase enzyme 28,29 .Therefore, the pyrazole ring with two attached phenyl groups was incorporated in the design to provide favorable hydrophobic interactions.Hence, prompted by above observations and in continuation to our attempt in development of α-glucosidase inhibitors, we designed the skeleton of biphenyl pyrazole-benzofuran and evaluation their α-glucosidase inhibition potential [30][31][32][33][34] (Fig. 1).
Apart from in vitro assessment of target compounds, the mechanism underlying enzymatic inhibition of the most potent compound was further explored using kinetic analysis.Docking studies and molecular dynamic simulation have been performed to determine plausible protein-ligand interactions.

Chemistry
The synthetic pathway for the preparation of iphenyl pyrazole-benzofuran hybrid 8a-o is outlined in Scheme 1.The hydrazones 8a-o were prepared by the condensation reaction of the acetophenone derivatives 1 with the phenylhydrazine or 4-methyl phenylhydrazine hydrochloride 2 under re ux condition in absolute ethanol and in the presence of a catalytic amount of sulfuric acid.Intermediate hydrazone derivatives 3a-o were converted into 4-formyl pyrazole derivatives 4a-o by applying Vilsmeier-Haack reaction with POCl 3 -DMF.On the other hand, benzofuran-2-carbohydrazide 7 was prepared by following reaction scheme.Initially, a mixture of salicylaldehyde 5 with ethyl bromoacetate heated at re ux in acetonitrile in the presence of potassium carbonate to afford ethyl benzofuran-2-carboxylate 6.Then, re uxing the ethanolic solution of the ethyl ester 6 and hydrazine hydrate yielded benzofuran-2carbohydrazide 7. Finally, desired products were obtained by coupling the hydrazide 7 with key aldehydes 4a-o in absolute ethanol and in the presence of glacial acetic acid.The structures of newly synthesized compounds were con rmed by their IR, 1 H NMR, 13 C NMR, and elemental analysis 25,35 .

In vitro α-glucosidase inhibitory activity
All of the newly synthesized compounds 8a-o were screened for α-glucosidase inhibition activity.α-Glucosidase enzyme from (Saccharomyces cerevisiae, EC3.2.1.20)was used to evaluate the αglucosidase inhibitory activity.Compared to human α-glucosidase, they shared similarity in substrate speci city, pH optimum, catalytic residues in active site and inhibitor sensitivity 36 .Therefore, αglucosidase from yeast is extensively used for the preliminary screening of α-glucosidase inhibitors 37,38 .Acarbose, a commercially available α-glucosidase inhibitor, was used as the reference drug and the results are expressed as IC 50 values (Table 1).Albeit, all pharmacophoric groups in the designed hybrid seem to have involved in the inhibitory potential but to further elucidate the role of substituents on the aryl rings connected to pyrazole, a wide variety of compounds 8 were synthesized and structure-activity relationship (SAR) has been established on the basis of varying substituents in R 1 and R 2 .Accordingly, SAR was investigated in two categories of 8a-i (Table 1, R 2 = H) and 8j-o (Table 1, R 2 = CH 3 ).It is worth mentioning that all synthetic compounds showed signi cant activity with IC 50 values in the range of 40.6 ± 0.2-164.3± 1.8 µM, when compared to acarbose (750.0 ± 10.0 µM).Among them compound 8e (IC 50 = 40.6 ± 0.2 µM), having a nitro group in R 1 position, was found to be the most potent inhibitor.This is approximately eighteen-fold more potent than the standard acarbose.The introduction of other electronwithdrawing groups, such as halogen atoms and tri uoromethyl (8f-i), led to the reduction of inhibitory activity.Halogenated counterparts of compound 8e in the rst category of compounds, exhibited inhibitory activity generally depending on the size of halogen substitution.Compound 8f (IC 50 = 79.7 ± 0.5 µM) bearing bromo substituent was found to be the second most active analog in this category and ninefold more potent than standard.Compounds 8g having a relatively large chlorine substituent and 8h having a small uorine group showed lower activities with IC 50 values of 125.3 ± 1.0 and 164.3 ± 1.8 µM, respectively.It should be noted that change the uoro group with tri uoromethyl moiety as in the case of compound 8i (IC 50 = 101.7 ± 0.7 µM), led to a signi cant improvement in the inhibition potential.Finally, the order of inhibition for compounds 8a-d was 8c > 8b > 8d > 8a (IC 50s = 100.3± 0.7, 108.3 ± 0.8, 133.6 ± 1.2, and 141.2 ± 1.4 µM, respectively), which illustrated that inhibitory activity almost affected by the electron-donating property of substituent.
For further investigation the SAR, second category of compounds 8j-o possessing methyl group in R 2 were also examined.Compound 8l (IC 50 = 54.7 ± 0.3 µM) containing methoxy group was found to be the most active compound in this series.Also, this compound is the second most potent among the screened compounds and > 13 folds more active than the standard.Replacement of methoxy group in 8l with methyl (8k, IC 50 = 115.6 ± 0.9 µM) or hydroxyl group (8m, IC 50 = 127.0± 1.0 µM) resulted in a remarkable decrease the biological activity.The bromo substituted derivative 8n was found as the second most potent molecule among this series of compounds and exhibited approximately eight-fold enhanced activity as compared to the standard acarbose.So, as mentioned in the previous sections, the size of the bromine may be responsible for the higher activity of brominated compounds than other halogens.
It can be concluded that the in vitro α-glucosidase inhibitory activity mainly depends upon the substituents on R 1 and the nature of substituents at this position affected the e cacy of methyl group on R 2 .In this regard, the α-glucosidase inhibitory activity of bromine (8f, 8n) and chlorine (8g, 8o) substituted analogs in two series of mentioned compounds revealed that the introduction of methyl groups on R 2 led to lower activity (8f > 8n and 8g > 8o).Similarly, the lack of methyl group in the case of compound 8k resulted in a slight decreased in inhibitory activity as compound 8b.In a different manner, compound 8l was more active than compound 8c.Both compounds have a methoxy group at the R 1 position but the presence of the methyl group on R 2 in compound 8l led to a two-fold increase in activity.The same is the case of compounds 8d and 8m having hydroxyl group, the methyl substituent on R 2 position was found to confer an increase in inhibitory activity.

Kinetic study
To gain further insight into the mechanism of action of the synthesized compounds against αglucosidase, kinetics analysis was performed on the most potent compound 8e.In different concentrations of test compound (0, 10, 25, and 40 µM) and with the incremental concentration (2-10 mM) of substrate, the rate of the enzyme activity was calculated.The type of inhibition and experimental inhibition constant (K i ) value were determined by employing Lineweaver-Burk plots and secondary replot of these plots, respectively.The Lineweaver-Burk plot showed that with varying concentrations of compound 8e, V max of enzyme gradually increased without affecting the K m of enzyme (Fig. 2a).This pattern indicates a competitive type of inhibition.The plot of the slope of lines in the Lineweaver-Burk plots (k m ) against the inhibitor concentration gave an approximation of the inhibition constant, K i of 38 µM for compounds 8e (Fig. 2b).

Homology modeling and molecular docking study
Molecular docking studies were also performed to rationalize the results of biological assays and gain structural insight into the binding of the synthetic derivatives against α-glucosidase.Due to the unavailability of the crystallographic structure of α-glucosidase from S. cerevisiae, the 3D structure of αglucosidase was modeled using MODELER inbuilt in Discovery Studio (DS) package and synthetic compounds were docked against established homology model.For this purpose, the FASTA format of the primary sequence was downloaded from Uniprot (Access code P53341) and submitted to NCBI BLAST to get a template with a suitable identity for sequence alignment 39 .Isomaltase from S. cerevisiae (PDB ID: 3A4A) with 71.4% identity and 86.7 % similarity was selected as the template for modeling 40 .The best model was selected based on the lowest PDF Total Energy (3270.6404) and DOPE Scores (-73110.257813) and evaluated for further validation.The PROCHECK program was applied to assess the stereochemical quality of the model.The phi/psi Ramachandran plot distributions indicated that 99.6% residues are in the favored and allowed regions and only 0.2% residues lies in the outlier region (Fig. 3).
The superposed structure of acarbose (standard inhibitor) and the predicted top-scored conformation of the most potent compound 8e in the active site of a homology model of α-glucosidase was shown in Fig. 4 (left).The detailed binding mode of acarbose showed that it formed hydrogen bonding interactions with residues Asp349, His239, Asp68, Pro309, Glu304, Arg439, Arg212, Glu276 and one hydrophobic interaction with Phe157 (Fig. 4, right).
The theoretical binding mode of most active compounds 8e, 8l, 8f and 8n were also shown in Figs. 5.
Binding mode analysis showed that the following interactions are common among these compounds: (1) Pyrazole moiety interacts with Glu276 through π-anion binding.(2) Phenyl groups attached to the pyrazole moiety, on the other hand, are involved in hydrophobic interactions with Arg439, Ala278 and Tyr71.
(3) The carbonyl oxygen of the amide group formed hydrogen bonds with the hydroxyl group of Tyr313, and (4) the planar benzofuran scaffold interacted with Arg312 via π-alkyl interaction.As expected, all pharmacophoric features used in the design of the hybrid, are involved in the interaction with the active site of the enzyme.The most potent compound 8e establishes more interactions with the residues in the binding pocket.The nitro substituent of this compound created a hydrogen bond with Arg439 and also two electrostatic interactions with Asp68 and Tyr71.Besides, the N1-phenyl ring of pyrazole moiety and benzofuran also formed hydrophobic interactions with Leu218 and Phe157, respectively, which leads to a snug t at the binding site (Fig. 5a).When the interaction mode of compound 8l as the second most potent compound is compared to that of compound 8e, only one hydrogen bond with Asp68 stabilizes diphenyl pyrazole moiety in 8l while in compound 8e the electrostatic interactions play an important role in binding of this moiety to the enzyme.Methyl substituent in compound 8l is located in a hydrophobic pocket formed by residues Leu218, His245, His279, Phe300 and Ala278 (Fig. 5b).In the case of compounds 8f and 8n, both have bromine substitution at the R 1 position, the presence of methyl group in R 2 may cause their different orientations in the active site and then the difference in inhibitory activity.The predicted binding mode of compound 8f shows that the NH proton of the amide group forms hydrogen bond interaction with the amide group of Gln350.The hydrogen bond belonging to the carbonyl oxygen of the amide group was not seen in this compound.Moreover, the CH imine group is forming a hydrogen contact with carboxyl oxygen of Asp349 and thus leads to a better t of this compound in the enzyme's active pocket (Fig. 5c-d).The calculated GOLD Fitness Scores for compounds 8e (70.4927), 8l (65.7091), 8f (64.84.4) and 8n (61.1567) were in good agreement with those results obtained in in vitro assay.

Molecular dynamic investigation
Molecular dynamic simulation performed in order to understand the effect of the compound over the enzyme active site.For this purpose, the structural perturbations incurred by the most potent compound (8e) has been investigated through the study of the RMSD, RMSF and its effect on the active site environment in comparison to acarbose as α-glycosidase standard inhibitor and the apoenzyme.
Root mean square deviation (RMSD) of the enzymes' backbone was analyzed over 20 ns MD simulation in order to study the stability of the protein-ligand complex trajectories (Fig. 6).The RMSD value of the apo α-glycosidase depicts broad uctuations throughout simulation time which is higher than the two enzyme complexes.The RMSD value increased after about 4 ns and steadily increased up to 16 ns and become more stable for the last 4 ns of simulation time with the value of 2.5 Å.The RMSD value of glycosidase complexed with acarbose was stable until 12 ns and slightly increased through the next 4 ns and become steady for the rest of the simulation time with the RMSD value of 2.1 Å. Although, the mentioned value of α-glycosidase complexed with compound 8e is the same as acarbose bounded state for the rst 8 ns it is observed that compound 8e had higher RMSD than acarbose for its higher number of ring and exibility which makes it more deviate from the initial structure for the next 8 ns and nally it decreased and stabilized for the last 2 ns with the same RMSD as acarbose bounded state (2.1 Å).
In summary, the RMSD value of the bounded state enzymes deviate from the initial structure of apoenzyme in the early part of the simulation and obviously decreased as a result of α-glycosidase structural rigidity.Thus, the structures at the last 2 ns of the MD equilibrium state used to investigate the structural speci city of the ligand-protein complexes.
The RMSF, which indicates the exibility of protein structure, refers to the uctuation of the Cα atom from its average position throughout the simulation time.Figure 7a compares the residue RMSF values of αglycosidase bound state and unbound state in which, the apoenzyme (yellow color line) had higher RMSF uctuations compared to the glycosidase bound-states (green and red-colored lines).This observation occurs upon ligand binding to the enzyme, in which residues movement decrease as a result of nonbonding interaction between the ligand and the enzyme.In addition, the structural segments which are affected upon ligand binding have recognized and categorized into four apparent parts including; B domain loop (residues 139-149), the active site lid, A domain and B domain sides of the active site mouth.Comparing RMSF values shows that the residues of the B domain loop, 139-149, would have signi cantly lower RMSF value in glycosidase/ acarbose and compound 8e bound-state rather than apoenzyme (Fig. 7a and 7b).In contrast, the exibility of the active site lid increased in enzyme boundstate and is more pronounced through acarbose binding.In order to investigate the reason for the mentioned observation, as noticed in Fig. 7b (glycosidase/ acarbose complex) acarbose interacted with several residues located into the A and B domain side of the active site mouth (the vertical green line).Although compound 8e interacted with the same regions, it formed fewer interaction with the residues of the A domain.So, it can be proposed that the more ligand interaction with A and B domain sides of the active site mouth, the higher the RMSF of the active site lid.Moreover, Fig. 7c and 2d represent the organization of the α-glycosidase three main domains; A, B, and C and the close-up representation of the active site mouth with the corresponding residues of A and B domain at both sides in which the active side lid and a back-wall helix situated at the front and the back of the mentioned mouth, respectively.Backing to the Fig. 7b, compound 8e provides higher interaction with the back-wall of the active site rather than acarbose.Based on the observed result of RMSF plot, although α-glycosidase/ acarbose complex with higher interaction at the entrance region loop covering the active site (310-315) have lower RMSF value than in α-glycosidase-8e complex, the other lid loop covering the active site consists of residues 230-236 41 shows signi cantly lower RMSF value in α-glycosidase-8e complex rather than αglycosidase/ acarbose complex.
Finally, based on RMSF plot it can be proposed that compound 8e had more interaction over the B domain side and back-wall helix of the active site, while acarbose formed more interaction with the A domain side of the α-glycosidase active site.
Figure 8a and 8b represent the detailed orientation and interactions that occurred more than 30% of the simulation time during the equilibrated phase over α-glycosidase complexed with compound 8e.The interaction analysis depicts compound 8e oriented vertically from the mouth to the bottom of the active site and stabilized the enzyme domains by interacting with Phe311, Tyr313, Arg312 from the A domain side and Phe158, Phe177 and His239 from the B domain side of the active site mouth (Fig. 8a).In addition, polar residues including; Asp214, Asp349 and Arg439 provide polar interactions with compound 8e at the depth part of the active site.In the same way, acarbose disposed vertically and formed nonbinding interactions with the Phe311, Asn241, Arg439, Asp68, His245, Asp349, Asp214 which belong to the domain A of the enzyme (Fig. 8c).
Comparing MD simulation of compound 8e and acarbose proposed the long-lasting non-binding interactions with Asp349, Asp214 and Arg439 have a signi cant role in inhibition activity of the mentioned compounds (Fig. 8b and 8d). Figure 8a and 8b represent two important structural moieties in stabilizing compound 8e at the mouth of the active site.The rst one is the benzofuran ring which interacts with Phe311 and Tyr313 through T-shape π-π hydrophobic interactions for 90% and 55% of the simulation time and the next one, pyrazole ring which interacts with Arg312 through π-cation hydrophobic interaction for 96% of the simulation time.Along with the interactions which stabilized compound 8e in front of the A domain side of the active site entrance, the phenyl substituent interacts with Phe157 and His239 which faced at the B domain side of the entrance for about one-third of the simulation time.In addition, the hydrazide moiety can provide H-bond interaction with Asp408 at the back part of the active site (previously known as the back wall side) for about 35% of simulation time.Finally, the nitrophenyl group as a polar moiety interacts with polar catalytic residues Arg439, Asp349 and Asp215 through ionbridge, H-bond and water-mediated H-bond interactions for a signi cant amount of simulation time that is similar to the behavior of the NH2 group in the acarbose (Fig. 8d).It is obvious from the MD study that H-bond, hydrophobic interactions and ion-bridge interactions have a critical role in stabilizing compound 8e at different sides of the active site during the simulation time and cause α-glycosidase inhibition activity.This observation may propose the contribution to the higher α-glycosidase inhibition activity.

Conclusion
With aim of development of novel class of α-glucosidase inhibitors, new series of biphenyl pyrazolebenzofuran hybrids derivatives were designed, synthesized and evaluated for their α-glucosidase inhibition.All screened compounds displayed multifold enhanced inhibitory strength in the range of 40.6 ± 0.2-164.3± 1.8 µM when compared to acarbose (IC 50 = 750.0± 10.0 µM).Among them compound 8e, having a nitro group in R 1 position, was found to be the most potent inhibitor.This is approximately eighteen-fold more potent than the standard acarbose.Also, the kinetic analysis revealed that compound 8e compete with the substrate for binding to the binding site of the enzyme.Limited SAR studies indicated that the in vitro α-glucosidase inhibitory activity mainly depends upon the substituents on R 1 and the nature of substituents at this position affected the e cacy of methyl group on R 2 .Binding mode analysis showed that almost all structural features such as pyrazole ring, Phenyl groups attached to the pyrazole moiety, amide linkage and benzofuran scaffold are contributing to binding a nity through hydrogen bonding, hydrophobic and electrostatic interactions.In addition, MD simulations showed compound 8e oriented vertically into the active site from mouth to the bottom and stabilized the enzyme domains by interacting with the interface of domain A and domain B and the back side of the active site while acarbose formed non-binding interaction with the residue belong to the domain A of the enzyme.
Moreover, carbonyl hydrazide linker, pyrazole and its related substituents provide such a strategic point with the ability to interact with various parts of the active site which has the binding and catalytic role for α-glycosidase activity.
Taken together, the above results suggest that newly synthesized hybrids could be promising hits for the further development of α-glucosidase inhibitors for the treatment of diabetes patients.

Experimental
All reagents and organic solvents were purchased from Sigma Chemical Co.(St.Louis, USA) and used without further puri cation.Thin-layer chromatography (TLC) was carried out on pre-coated silica gel aluminum plates (Merck silica gel 60, F254).Melting points of target compound 8a-o were measured on a Ko er hot stage apparatus and were uncorrected. 1H NMR and 13 C NMR spectra were recorded on Bruker FT-500 spectrometer (Bruker, Rheinstetten, Germany) in DMSO-d 6 with tetramethylsilane (TMS) as the internal standard.IR spectra were recorded on Nicolet Magna FTIR 550 spectrophotometer (resolution 2 cm − 1 ) in KBr pellets.Elemental analysis was carried out with an Elemental Analyzer system GmbH VarioEL CHNS mode (Germany).

General procedure for the preparation of 1,3disubstituted-4-pyrazole carbaldehydes 4a-n
To a solution of 4-substituted phenylhydrazine hydrochloride 2 (20 mmol) in ethanol (15 mL), substituted acetophenone 1 (20 mmol) and catalytic amounts of sulfuric acid was added and then re ux the mixture for 8-12 h.After reaction completion, the mixture was cooled to room temperature and poured on crushed ice to afford hydrazone intermediate 3a-n.The resulting solid was ltered and recrystallized from ethanol.A solution of hydrazone 3a-n (20 mmol) in DMF (5 mL) was added drop-wise to an ice-cold solution of DMF (15 mL) and phosphorus oxychloride (60 mmol) and the resulting mixture re uxed at 60-70 ºC for 5-8 h.After the completion of the reaction, the reaction mixture allowed to cool, poured into ice-cold water and then neutralized with saturated aqueous sodium hydroxide solution.Further, the solid precipitated was ltered, washed with excess cold water and recrystallized from ethanol to afford aldehydes 4a-n.4.2.General procedure for the preparation of ethyl benzofuran-2-carboxylate 6 A mixture of salicylaldehyde (20 mmol), ethyl bromoacetate (20 mmol) and K 2 CO 3 (40 mmol) in acetonitrile (10 mL) heated under re ux for 4h.After completion of the reaction (monitored by TLC), the reaction mixture was allowed to cool to room temperature and poured into crushed ice.After extracting the product with ethyl acetate (2 × 25 mL), the organic layer was washed using brine solution (2 ×20 mL) and dried over anhydrous MgSO 4 .The solvent was evaporated under vacuum to afford the product as an oil.
4.3.General procedure for the preparation of benzofuran-2carbohydrazide 7 benzofuran-2-carboxylate 6 (20 mmol) and hydrazine hydrate (30 mmol) in EtOH (10 mL) heat under re ux overnight.Upon cooling, the product precipitated was ltered, washed with cold water and recrystallized in methanol to afford the pure product.

General procedure for the preparation of pyrazolebenzofuran hybrids 8a-n
To a mixture of the appropriate pyrazole aldehydes 4 (1 mmol) and catalytic amount of glacial acetic acid (3-4 drops) in absolute ethanol (10 mL), were added hydrazide 7 (1.1 mmol) and re uxed for 12-18 h.As the reaction was completed, the mixture was allowed to cool to room temperature, the precipitate was ltered off and crystallized from ethanol to give the pure nal derivatives 8a-n.

In vitro α-glucosidase inhibition assay
Stock solutions of the test compounds and acarbose were prepared in DMSO and eventually diluted with phosphate buffer saline to attain the desired concentration.A mixture containing various concentrations of test compound and enzyme (Saccharomyces cerevisiae, EC3.2.1.20,0.2 U/mL) in phosphate buffer saline was added in the 96-well plate and incubated at 37 ºC for 10 min.Then, the enzyme-catalyzed reaction was commenced by the addition of p-nitrophenyl-α-glucopyranoside as substrate (25 µL, 4 mM) and the absorbance was measured spectrophotometrically at 400 nm after 20 min incubation at 37°C.DMSO (10% nal concentration) and acarbose were used as the control and standard inhibitor, respectively.The percentage of enzyme inhibition for each entry was calculated using the following formula: % Inhibition = [(Abs control -Abs sample)/Abs control] ×100.
The concentrations of the compounds that inhibited 50% of α-glucosidase activity (IC 50 values) were calculated from non-linear regression curve using the Logit method.

study
The kinetics of enzyme inhibition was performed according to the reaction conditions in 2.5 by preparing a series of test solutions in which the concentration of the substrate (PNPG) was varied (2-10 mM) in the absence and presence of compound 8e at different concentrations (0, 10, 25, and 40 µM).A Lineweaver-Burk plot was generated to identify the type of inhibition and the Michaelis-Menten constant (K m ) value was determined from the plot between reciprocal of the substrate concentration (1/[S]) and reciprocal of enzyme rate (1/V) over various inhibitor concentrations.Experimental inhibitor constant (K i ) value was constructed by secondary plots of the inhibitor concentration [I] versus K m .

Homology modeling and docking study
The primary sequence of Saccharomyces cerevisiae α-glucosidase downloaded from UniProtKB database (Uni-ProtKB, http://www.uniprot.org/)with accession number P53341.Hence, a search was executed to identify a protein with a high sequence similarity using NCBI BLAST server (https:// blast.ncbi.nlm.nih.gov/Blast.cgi).The crystallographic structure of Saccharomyces cerevisiae (PDB ID: 3A4A) was chosen as a template and subjected to sequence alignment using Align sequence to template protocol in Discovery Studio 4.1 (Accelrys, San Diego, CA) (DS).The 3D structure of α-glucosidase for S cerevisiae was predicted using the Build Homology Models protocol implemented in DS software.The number of models was set to 10, and the optimization level was changed to high.DOPE score and pro le-3D for preliminary evaluation of developed model were carried out using verify protein protocol in DS.
Ramachandran plot with PROCHECK program (http://servicesn.mbi.ucla.edu/PROCHECK)was also applied to verify the quality of the obtained homology model 42,43 .With the modeled structure, the docking of selected compounds was carried out.

Figures
Figures