Plant Physiology and Biochemistry 83 (2014) 217e224

Contents lists available at ScienceDirect

Plant Physiology and Biochemistry journal homepage: www.elsevier.com/locate/plaphy

Research article

Identification of active site residues of Fenugreek b-amylase: Chemical modification and in silico approach Garima Srivastava, Vinay K. Singh, Arvind M. Kayastha* School of Biotechnology, Faculty of Science, Banaras Hindu University, Varanasi, India

a r t i c l e i n f o

a b s t r a c t

Article history: Received 27 May 2014 Accepted 7 August 2014 Available online 17 August 2014

The amino acid sequence of Fenugreek b-amylase is not available in protein data bank. Therefore, an attempt has been made to identify the catalytic amino acid residues of enzyme by employing studies of pH dependence of enzyme catalysis, chemical modification and bioinformatics. Treatment of purified Fenugreek b-amylase with EDAC in presence of glycine methyl ester and sulfhydryl group specific reagents (IAA, NEM and p-CMB), followed a pseudo first-order kinetics and resulted in effective inactivation of enzyme. The reaction with EDAC in presence of NTEE (3-nitro-L-tyrosine ethylester) resulted into modification of two carboxyl groups per molecule of enzyme and presence of one accessible sulfhydryl group at the active site, per molecule of enzyme was ascertained by titration with DTNB. The above results were supported by the prevention of inactivation of enzyme in presence of substrate. Based on MALDI-TOF analysis of purified Fenugreek b-amylase and MASCOT search, b-amylase of Medicago sativa was found to be the best match. To further confirm the amino acid involved in catalysis, homology modelling of b-amylase of M. sativa was performed. The sequence alignment, superimposition of template and target models, along with study of interactions involved in docking of sucrose and maltose at the active site, led to identification of Glu187, Glu381 and Cys344 as active site residues. © 2014 Elsevier Masson SAS. All rights reserved.

Keywords: b-Amylase Active site Chemical modification Docking Homology modelling Trigonella foenum-graecum

1. Introduction

b-Amylase or exoamylase hydrolyzes the a-1,4 glucosidic linkage of starch liberating successive b-maltose units from nonreducing ends b-Amylase (EC 3.2.1.2) is a member of 14th family of glycosyl hydrolases (Henrissat, 1991). The enzyme follows a multiple-attack mechanism with a-inversion, releasing multiple bmaltose units from a maltooligosaccharide chain before the chain is released. b-Amylase is a simple protein, with no evidence for requirement of cofactors, metals or non-protein active groups. bAmylases are well characterized in higher plants, and have also been reported in microorganisms (Higashihara and Okada, 1974; Shinke et al., 1975; Murao et al., 1979). The saccharifying activity of the enzyme and production of maltose as end product, finds commercial value in food, beverage and pharmaceutical industries. The amino acid sequences for the plant b-amylases are highly conserved, with similarity ranging from 60% to 96% (Gana et al., 1998). The X-ray crystallographic study of soybean b-amylase had shown the

* Corresponding author. Tel.: þ91 542 2368331; fax: þ91 542 2368693. E-mail addresses: [email protected] (G. Srivastava), officerbiotech.1@ gmail.com (V.K. Singh), [email protected], [email protected] (A.M. Kayastha). http://dx.doi.org/10.1016/j.plaphy.2014.08.005 0981-9428/© 2014 Elsevier Masson SAS. All rights reserved.

presence of core (a/b)8 barrel domain having three loops (L3, L4 and L5) forming deep active cleft and an additional C-terminal loop surrounding the barrel helices. Two side chains of Glu residues (Glu186 and Glu380) have been concluded to act as an acid and base pair in the catalytic process based on the study of soybean b-amylase and maltose complex. The carboxyl group of Glu186 is located on the hydrophilic surface of the glucose, and donates a proton to the glycosidic oxygen. The carboxyl group of Glu380 lies on the hydrophobic face of the glucose residue at the subsite1 and activates an attacking water molecule. The maltose produced by enzymatic hydrolysis contains the reducing glucose in the b-form. Cys residues had also been shown to play an important role in catalytic step by assisting the open-close movement of the flexible loop (residues 96e103 in soybean bamylase), that forms part of the active site. Cys95 residue was found to be located on the flexible loop and Cys345 is in the vicinity of subsites 3 and 4 (Mikami et al., 1994; Kang et al., 2004). The amino acid residues involved in the catalysis and substrate binding are well-conserved between plant and bacteria. The residues present in the core domain of Bacillus cereus and soybean b-amylase are 24% identical. In B. cereus, Glu172 and Glu367 act as the general acid and base catalyst, respectively. A common target site of eSH reagents with the plant enzyme was found to be present at Cys331. Cys91 forms the disulfide bridge and exists near the flexible loop,

218

G. Srivastava et al. / Plant Physiology and Biochemistry 83 (2014) 217e224

however in B. cereus it has been thought to play role in stabilization of flexible loop without affecting the open-close movement (Oyama et al., 1999). The term ‘active site’ refers to the amino acid residues involved in the catalysis, along with those involved in interaction with the substrate. There are numerous examples of proteins that show 8e15% sequence similarity but adopt similar structures, contain identical or related amino acid residues in their active site, and have similar catalytic mechanism (Ollis et al., 1992). Also, some examples of enzymes belonging to the same family but showing different alignment of catalytic amino acids are known, e.g. pea b-amylase is not inactivated in presence of sulfhydryl group specific reagents, whereas plant b-amylase are known to require free thiol groups for activity (Lizotte et al., 1990). Therefore, it is necessary to perform experimental analysis to identify active site residues. Simple and rapid methods of selective chemical modification of side chains of proteins provide preliminary data regarding the role of particular amino acids in a given protein. This allows one to predict the possible significance of amino acid residues in the catalytic activity by providing knowledge of functionally important sites of the concerned protein. Thus, it can be useful in guiding experimental and molecular mechanism analysis. Bioinformatics approach based on analysis of protein sequence offers better understanding of the fundamentals of biocatalysts by providing clue regarding the role of individual residues in the protein. Bioinformatics program also allows the calculation of various characteristics of constructed molecule, alignment of protein sequences, creation of homology based structure and investigation of docking of ligand to the active site of enzymes. Homology modelling has been used as a tool to study protein structure by some authors (Singh et al., 2008; Kumari et al., 2013; Singh et al., 2013). Three dimensional alignment based on the comparison of the geometric orientation of amino acid residues in tertiary structure, provide significant clues regarding protein's function and properties. Fenugreek is cultivated as a leafy vegetable, condiment and as a medicinal plant. It has high medicinal value in Indian Ayurvedic and Chinese medicine, where it is used for labour induction, aiding digestion and for improving metabolism and health. Fenugreek seeds have been found to contain many nutrients including protein, carbohydrates, fat in the form of volatile and fixed oil, vitamin and minerals as well as enzymes, fibres, saponins, choline and trigonelline (Basch et al., 2003). In fenugreek seeds, three different types of carbohydrates (galactomannans, soluble sugars including galactosyl sucrose sugars and starch) are utilized during germination and seedling development (Dirk et al.,1999). Starch is not present in seeds and its synthesis starts only after germination. Thus, fenugreek seeds appeared to be an interesting material to study the potential role of bamylase in carbohydrate metabolism (Reid, 1971). b-Amylase has been purified to apparent homogeneity from fenugreek (Trigonella foenum-graecum) seeds and its association with amyloplasts and protophloem in cotyledon was demonstrated during different stages of germination (Srivastava and Kayastha, 2014). The amino acid sequence of b-amylase from T. foenum-graecum is not available in protein database. Therefore, with a view to understand the possible active site residues involved in the catalytic functioning of the Fenugreek b-amylase, in the present study we have employed combined approach of chemical modification method and in silico study of enzymeeligand interactions.

732.59 units mg1 (Srivastava and Kayastha, 2014). Prior to modification reactions, the purified enzyme was dialyzed against the respective reaction buffer for 24 h at 4  C, followed by determination of protein concentration and specific activity. 2.2. Chemicals DEPC (diethylpyrocarbonate), EDAC (1-ethyl-3-(3-dimethyla minopropyl) carbodiimide, NTEE (3-nitro-L-tyrosine ethylester), Glycine methyl ester, DTNB (5, 5'-dithiobis-(2-nitrobenzoic acid), NEM (N-ethylmaleimide), IAA (iodoacetamide), DTT (dithiothreitol), GnHCl (guanidine hydrochloride), BSA (bovine seum albumin) were obtained from Sigma Chemicals Co. (St. Louis, MO, USA). Tris buffer, sodium acetate, p-CMB (p-chloromercuribenzoate) were from Himedia Labs (Mumbai, India). All other chemicals were of analytical grade and all the solutions were prepared in Milli Q water. 2.3. Enzyme and protein assay Bernfeld's method (1955) was used for measurement of enzyme activity. Reaction mixture was prepared by taking 0.5 mL of suitably diluted enzyme and 0.5 mL of 1% starch prepared in 50 mM sodium acetate buffer (pH 5.0). This was incubated at 30  C for 3 min. Reaction was stopped by addition of 3, 5-dinitrosalicylic acid. Test tubes were then placed in boiling water bath for 5 min and were allowed to cool down to room temperature, followed by addition of 10 mL of distilled water. Absorbance was recorded at 540 nm. One unit of b-amylase is defined as the amount required for release of 1 mM of b-maltose per min at 30  C and pH 5.0 under the specified condition. Protein concentration was determined using Lowry's method (1951), using BSA as standard. 2.4. pH dependent studies The effect of pH on Km and Vmax values of Fenugreek b-amylase were observed by varying the starch concentration in the range of 0.5e10 mg mL1 using 50 mM sodium acetate buffer (pH range 4.0e5.5), and phosphate buffer (pH range 6.0e7.5) and determining the enzyme activity by standard assay procedure. The pKa values of the ionizable groups in the enzymeesubstrate complex were obtained from the plots of pKm and log Vmax against pH. 2.5. Modification of histidine residues DEPC was dissolved in ethanol immediately before use. The concentration of DEPC was measured by reacting an aliquot with 10 mM imidazole prepared in 50 mM sodium phosphate buffer (pH 6.8) and monitoring the absorbance increase at 230 nm using an extinction coefficient of 3200 M1 cm1 (Miles, 1977). 1 mM DEPC was used for inhibition studies. To the enzyme (0.2 mg mL1), 50 mL of DEPC solution was added in 50 mM sodium phosphate buffer (pH 6.8) at 30  C. The reaction was monitored at 240 nm (Miles, 1977; Lundblad and Noyes, 1984) and aliquots withdrawn at different time intervals from DEPC-enzyme mixture were assayed for residual activity. 2.6. Modification of carboxylate residues

2. Materials and methods 2.1. Enzyme

b-Amylase was purified to apparent homogeneity from fenugreek seeds with specific activity of the purified enzyme of

Aliquots of purified Fenugreek b-amylase (0.2 mg mL1) in 50 mM sodium acetate buffer (pH 4.5), were incubated with 10 mM EDAC and 0.1 M glycine methyl ester. Small aliquots were withdrawn from reaction mixture, followed by determination of the residual activity by standard assay method. In a separate set of

G. Srivastava et al. / Plant Physiology and Biochemistry 83 (2014) 217e224

experiment, enzyme was incubated with 10 mM EDAC and 5 mM NTEE, at 30  C for 10 min. Aliquots were withdrawn at different time intervals and the reaction was terminated by addition of 100 mL of 1 M sodium acetate buffer (pH 5.5), followed by addition of 10% trichloroacetic acid (1 mL). The mixture was left at 4  C for 30 min to precipitate EDAC/NTEE enzyme complex. The precipitate was collected by centrifugation (7000 g, 10 min), washed extensively with chilled acetone and was air dried followed by reconstitution in 0.5 mL of 0.1 M NaOH. The number of nitrotyrosyl group incorporated was determined spectrophotometrically at 430 nm, using a molar absorption coefficient of 4600 M1 cm1. Enzyme samples incubated in the absence of EDAC/NTEE served as control. 2.7. Effect of metal ion on enzyme activity Stock solution of metal ions (Cu2þ, Hg2þ and Agþ) were prepared in 50 mM TriseHCl buffer (pH 7.2), and were suitably diluted for the experiments. The activity of Fenugreek b-amylase was determined in the presence of varying concentrations of these inhibitors. Inhibition pattern and inhibition constants (Ki) were calculated by the Dixon method (1953). 2.8. Sulfhydryl group assay and inactivation studies using DTNB Free and total cysteine residues present in Fenugreek b-amylase were determined using the DTNB method of Ellman (1959). The enzyme was reduced with 0.05 M DTT for free cysteine content determination, while for the estimation of total number of cysteine residues, the enzyme was denatured in 6 M GnHCl followed by treatment with 0.05 M DTT. After reduction the excess reducing agent was removed by dialysis with extraction buffer (50 mM TriseHCl buffer (pH 7.2)) followed by addition of 0.2 mL (0.2 mg mL1) aliquot to the solution containing DTNB (0.1 mM) and 3.1 mL, 0.1 M sodium phosphate buffer (pH 8.0). The liberated TNB (2-nitro-5-thiobenzoic acid) anion after reaction of sulfhydryl group with DTNB was monitored spectrophotometrically at 412 nm. The number of sulfhydryl residues modified were calculated by considering molar absorption coefficient of 1.32  104 M1 cm1 at pH 8.0. 2.9. Inactivation with various sulfhydryl group specific reagents

219

molecular mass information was matched with b-amylase from other plant sources using MASCOT search (Mascot is a search engine which uses mass spectrometry data to identify proteins from peptide sequence databases). The most significant match was found with b-amylase of Medicago sativa (alfalfa) (Accession no. T09300) with a score of 82 and E-value of 0.0015. The peptides matched are shown in bold red in Fig. 1 (Srivastava and Kayastha, 2014). Further, protein sequence of b-amylase from M. sativa (PDB ID: AAD04188.1) was retrieved from the NCBI database as MALDI based peptide mass fingerprint of Fenugreek b-amylase (SLFQLVQK, TAIEIYSDYMKSF RENMSDLLK, DGYRPIAK, HHAILNFTCLEMR, ENIEVAGENALSRYDAT AYNQIILNARPQGVNK and MYGVTYLR) were found to be conserved. Basic local alignment search tool (BLAST) analysis was performed against the protein data bank (PDB) database to search template based on similarities and identities for homology modelling. Crystal structure of soybean b-amylase complexed with maltose at a resolution of 1.86 Å (PDB ID: 1Q6C) showing high score and low E-value was selected as the appropriate template (Hirata et al., 2004). The sequence alignment between the target and template was calculated using Bl2seq and homology modelling of target protein was performed by the ESyPred3D (http://www.unamur.be/sciences/ biologie/urbm/bioinfo/esypred/) server (Lambert et al., 2002). To check the geometric restraints of the b-amylase from M. sativa, energy minimization was accomplished by DS CHARMm of Discovery Studio 3.0 program (Accelrys Software Inc, San Diego, CA, USA) (Kumari et al., 2013). Ramachandran plot for investigation of the backbone conformation of the predicted model and analysis of structural motif was developed by the RAMPAGE server (http:// mordred.bioc.cam.ac.uk/~rapper/rampage.php) (Lovell et al., 2002) and PDBsum server (http://www.ebi.ac.uk/pdbsum). The visualization of the structure and superimposition of the modelled protein with template was performed using Discovery Studio 3.0. The calculation of the interactions between different atom types for structural verification was carried out by ERRAT server (http:// nihserver.mbi.ucla.edu/ERRAT) (Colovos and Yeates, 1993). The ProSA server (https://prosa.services.came.sbg.ac.at/prosa.php) (Wiederstein and Sippl, 2007) was applied for recognizing the errors in three-dimensional structures between target and template protein models and VADAR server was used for quantitative evaluation of predicted model. Final verified model was submitted to PMDB database.

The Fenugreek b-amylase (0.2 mg mL1) was incubated separately with the desired concentration of specified reagents, such as p-CMB (10 mM) in 50 mM TriseHCl buffer (pH 8.5), NEM (0.1 mM, 0.5 mM) and IAA (1 mM) in 50 mM TriseHCl buffer (pH 7.2), at 37  C. Aliquots withdrawn at different time intervals were assayed for residual activity. For reactivation studies, enzyme (0.2 mg mL1) was incubated with p-CMB (10 mM) for 28 min in 50 mM TriseHCl buffer (pH 8.5) at 37  C followed by addition of excess cysteine (200 mM) The recovery of activity was monitored at different time intervals.

Bl2seq (www.ncbi.nlm.nih.gov/blast/bl2seq/wblast2.cgi) was used for comparative alignment to investigate the similar active site residues in b-amylase from M. sativa as compared to the soybean bamylase. Probable binding sites were searched with PDBsum server (http://www.ebi.ac.uk/pdbsum).

2.10. Substrate protection and kinetic studies

Docking studies were conducted to evaluate the predictive ability of the Fenugreek b-amylase (based on the homology model

2.12. Active site identification

2.13. Molecular docking

The protective effect of substrate on modification was studied by incubating enzyme with excess amount of starch, prior to treatment with various modifying reagents. The residual activity of enzyme was assayed periodically by standard assay method. 2.11. Homology modelling The amino acid sequence of b-amylase from T. foenum-graecum is not available in protein database. Therefore, the peptide map obtained by MALDI-TOF analysis of purified protein along with its

Fig. 1. Amino acid sequence of b-amylase from Medicago sativa. The peptides matched on the basis of MALDI-TOF analysis of purified Fenugreek b-amylase are shown in bold red. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

220

G. Srivastava et al. / Plant Physiology and Biochemistry 83 (2014) 217e224

constructed of M. sativa) for understanding the molecular interactions between ligands and protein receptor. The structure of sucrose (CID 5988) and maltose (CID 6255) were obtained from the Pubchem database. The DockingServer (http://www.dockingserver. com) was employed for molecular docking calculation. The ligandeprotein complex was analyzed and visualized with Discovery Studio 3.0. 3. Result and discussion The pKa values (ionization constant) of the active site residues in an enzyme are of importance to the functionality of the catalytic mechanism of the enzyme. The pKa value of a residue is influenced by the immediate microenvironment, while the long range interactions are less important. The bends obtained in the plot of log Vmax against pH provide information regarding the pKa of the ionizing group which affect the activity in the enzymeesubstrate complex. The pH dependence of pKm and log Vmax for starch hydrolysis is shown in Fig. 2(a) and Fig. 2(b), respectively. Bell shaped pH profile of pKm and log Vmax suggest involvement of two ionizable groups with pKa 5.4 and 6.6, functionally at the active site. 3.1. Modification of histidine residues The plots of log Vmax and pKm against pH indicated involvement of amino acid with pKa of 6.6 in catalysis, which is close to the pKa value for histidine. Therefore, modification of histidine residues by DEPC was attempted so as to verify its role in catalysis. Enzyme incubated for 20 min with 1 mM DEPC did not show significant loss in activity, indicating absence of histidine at or near active site. In view of this observation the role of above residue at the active site of Fenugreek b-amylase was not investigated further. The pKa values represent the variation occurring due to the effect of microenvironment in which amino acid is located. Therefore, the value of pKa is only a rough guide to the nature of the groups at the active site. Hoschke et al. (1980) reported insignificant loss in substrate binding affinity and no significant change in activity following the ethoxycarbonylation of the histidine residues of sweet potato bamylase. In soybean b-amylase two histidine residues were found to be present at the border of the active site (His93 interacts with Glc 1 and His300 with Glc 4), but their imidazole groups were too remote from the reaction centre to be significantly involved in the catalytic chemistry. Therefore, a probable role of secondary assistance in transition-state stabilization was suggested (Mikami et al., 1994).

3.2. Modification of carboxylate groups Participation of carboxylate residues in hydrolysis of starch by Fenugreek b-amylase was identified by the pKa value of 5.4 obtained from the plots of log Vmax and pKm against pH. EDAC was used as specific reagent for carboxylate group modification. The reaction taking place in between carboxylic acid and carbodiimide in water leads to formation of o-acylisourea, which may then rearrange to give the N-acylurea. This N-acylurea on reacting with a nucleophile such as glycine methyl ester leads to formation of acyl glycine ester (Hoare and Koshland, 1967). Modification of purified Fenugreek b-amylase with 10 mM EDAC at pH 4.5 and at 30  C resulted in the loss of 80% of enzyme activity in 30 min (Fig. 3a). The inactivation was concentration dependent and followed pseudo first-order kinetics. The loss of activity was further correlated with the number of carboxylic acid groups modified by estimating the number of nitrotyrosyl groups incorporated in the reaction. The plot of % residual activity versus the number of residues modified showed stoichiometric relationship between the extent of inactivation and extent of modification so that complete loss of the activity resulted from the modification of two essential carboxylic acid groups per monomer of Fenugreek bamylase (Fig. 3b). EDAC mediated inactivation of enzyme was prevented to a great extent by pre-incubating the enzyme with substrate (Fig. 3a). This observation indicated that the modification of carboxylate residues resulted in the activity loss and supported the involvement and significance of carboxylate group at active site of enzyme. The microenvironment in which the groups are located affect the pKa values observed for various amino acid groups. The interactions of certain amino acid side chain affect its accessibility and reactivity, resulting in variation by several orders of magnitude in reactivity of protein. In some enzyme in which Asp and Glu residues act as proton donors and catalytic nucleophiles, an upward shift in the pKa value of proton donor was observed. Higher pKa value of the acidic amino acid residue participating as the acid catalyst in carbohydrases has been observed (Bray and Clarke, 1994). In catalysis by sweet potato b-amylase, Thoma and Koshland (1960) and Thoma et al. (1965) found two ionizing groups whose pKa constants were close to a carboxyl and a protonated imidazole group. Zherebtsov (1968) also described the same two catalytic groups in barley b-amylase based on heats of ionization. Hoschke et al. (1980) noted that a protonated carboxyl group rather than a protonated imidazole, is the catalytic group responsible for the higher pKa value observed by Thoma and Koshland (1960). Depending on above observations it can be

Fig. 2. pH dependence of (a) Km and (b) Vmax for the hydrolysis of starch by Fenugreek b-amylase.

G. Srivastava et al. / Plant Physiology and Biochemistry 83 (2014) 217e224

221

Fig. 3. (a) Effect of EDAC on inhibition of Fenugreek b-amylase in presence and absence of starch (b) Correlation between inactivation of Fenugreek b-amylase and carboxyl residues modified.

2.0

100 1.8

80

% Residual Activity

log % Residual Activity

1.6

1.4

1.2

1.0 Enzyme+ 10 µM pCMB Enzyme+ 1 mM IAA Enzyme+ 1 mM NEM Enzyme+ Starch+ 10 µM pCMB

0.8

0

10

20

40

20

0.6

(a)

60

30

40

50

Time (minutes)

(b)

0 0.0

0.2

0.4

0.6

0.8

1.0

-SH group blocked

Fig. 4. (a) Enzyme inhibition in presence of sulfhydryl group specific reagents in presence and absence of starch (b) Relationship between loss of activity and blocking of eSH group with 10 mM p-CMB.

concluded in the present case that the pKa value of 6.6 indicates towards involvement of protonated carboxyl group and imply involvement of two carboxyl groups in hydrolysis of starch by Fenugreek b-amylase. 3.3. Modification of sulfhydryl groups The inhibition of enzyme activity by Cu2þ, Hg2þ and Agþ at a Ki of 27.5 mmoles, 28 nmoles and 5.62 nmoles, respectively indicated towards requirement of thiol group for enzyme activity. The detailed study of inhibition of eSH groups and the relationship between the accessible eSH groups and the catalytic activity of Fenugreek b-amylase was investigated by incubating enzyme with reagents specific to thiol groups (IAA, NEM and p-CMB). The aliquots were withdrawn at different time intervals and were assayed for residual activity as well as for the residual accessible eSH groups by titrating with DTNB. More than 80% loss in activity was observed when Fenugreek b-amylase was allowed to react with sulfhydryl specific reagents (Fig. 4a). On being incubated with the above reagents, single exponential decay (pseudo first-order kinetics) of enzyme activity was observed. A plot of % residual activity against the number of accessible eSH groups blocked due to modification by p-CMB, is shown in Fig. 4(b). It has been shown that

the inactivation of Fenugreek b-amylase is directly proportional to the blocking of its accessible eSH groups and it was found from the plot that one accessible eSH group needs to be blocked for complete inactivation of enzyme. Titration of native and GnHCl denatured Fenugreek b-amylase with excess DTNB revealed the presence of one and four accessible eSH groups, respectively per enzyme molecule. Presumably, IAA, NEM and p-CMB, follow a common mechanism, of blocking the accessible eSH groups to bring about inactivation of this enzyme. Inactivation of enzyme by p-CMB was observed to be largely reversed on addition of excess of cysteine, resulting in gain of 62% of enzyme activity. It may be due to unblocking of eSH groups by mercaptide exchange reaction. Such studies involving the inactivation of eSH groups of enzymes by p-CMB and subsequent reversal of inactivation by the addition of cysteine has been reported earlier (Srivastava and Kayastha, 2000; Kumar and Kayastha, 2010). The enzyme was protected against time-dependent inactivation by incubating with starch before addition of sulfhydryl modifying reagents (Fig. 4a). This result suggests involvement of sulfhydryl group at the active site of Fenugreek b-amylase. Role of cysteine residues in the catalysis of plant b-amylase is wellstudied. In soybean and sweet potato b-amylase, Cys95 and Cys96, respectively were found to be located on flexible loop, which forms

222

G. Srivastava et al. / Plant Physiology and Biochemistry 83 (2014) 217e224

part of active site and Cys345 and Cys346, were present in the vicinity of subsites 3 and 4. Cysteine residues of sweet potato b-amylase were inert to DTNB whereas Cys95 of soybean b-amylase showed reactivity towards DTNB (Mikami et al., 1993; Tanaka et al., 2002). 3.4. Homology modelling of b-amylase from M. sativa

b-Amylase from M. sativa was used for analysis, as MALDI based peptide mass fingerprint of Fenugreek b-amylase were found to be conserved. For homology modelling crystal structure of soybean bamylase was taken as template using NCBI BLAST search against PDB database. The identity and positivity between target and template were 85% and 93%, respectively with 99% query coverage. Homology modelling was performed by ESyPred3D and energy minimization was performed by CHARMm force field, an application of Discovery studio 3.0. ESypred3D tool and server led to the relaxation of unfavourable contacts and resulted in stable structure having electrostatic energy of 17,892.83 kcal/mol. Further, three dimensional alignment was performed by structurally superimposing Ca trace of the predicted model and the template protein domain. Root Mean Square Deviation (RMSD) value, which is a measure of distance between the atoms of superimposed proteins was calculated by Discovery Studio 3.0, through the simulation using first frame as reference. RMSD value of 0.13 was obtained for Ca superimposition and a value of 0.176 was obtained for side chain superimposition. This result implies that the three dimensional structure of predicted model is similar to template and the active site was conserved between the target and template. 3.5. Structural assessment and verification After energy minimization of predicted model, structural assessment and verification was done using different tools and servers (RAMPAGE, PDBsum, ERRAT, ProSAand VADAR). The geometry of the final defined enzyme model was evaluated using Ramachandran plot computed with RAMPAGE and PDBsum server. The result obtained from both the servers showed presence of more than 90% residues in the allowed region (Supplementary Fig. 1), indicating the stereochemical quality of the predicted model to be satisfactory. Both the servers showed Asp495 as the only residue present in outlier/disallowed region. The calculation of non-bonded interactions between different atom types was executed by ERRAT server to validate the overall quality of the enzyme model. The quality factor of the predicted model obtained from ERRAT server was 85.89, which implied presence of low error in the structure. The two characteristics of Z-score and a plot of energies of the input structure as a function of amino acid sequence position, was determined by ProSA. The Z-score indicates overall model quality and measures the deviation of the total energy of the structure with respect to an energy distribution derived from random conformations. The Z-score value obtained from ProSA for target (12.51) and for template (12.84) showed that both have nearly same structural relation and quality of the model to be acceptable. Supplementary Fig. 2(a) and (b), display comparable energy plots for target and template structure, respectively. VADAR statistics yielded presence of 38% helices, 17% sheets, 43% coil and 24% turn in the predicted model. Mean hydrogen bond distance, energy and residues with hydrogen bonds were found similar to the expected one. The above investigations validated the reliability of the predicted model. The final modelled structure (Fig. 5a) was successfully submitted to PMDB database and was assigned PMDB ID PM0079409.

3.6. Active site residues identification Active site residues were analyzed using Bl2seq tool of NCBI database. The sequence alignment and superimposition of predicted model with that of template showed highly conserved active site residues in b-amylase of M. sativa. This identification has assisted in locating the exact binding site in the homology model. In b-amylase from M. sativa, following key residues were identified to be involved in interaction and stability of the protein-ligand complex: Arg327, Lys331, Thr343, Cys344, Met347, Glu381, Asn382, Ala383, Leu420 and Arg421, which were also found to be present in peptide mass fingerprint of Fenugreek b-amylase. PDBsum was also used for identification of energetically favourable binding active sites based on protein electrostatics and spatial proximity. The PDBsum analysis suggested 10 possible clefts. These clefts were compared to the active site of the template for the determination of residues forming the binding pocket and to find their role in catalysis. The first cleft was found to be prominent and responsible for interaction with compound. 3.7. Molecular docking Molecular docking was performed for predicting the binding affinity of the ligand to the protein based on the complex geometry. Sucrose and maltose were selected for docking study with the modelled structure of b-amylase from M. sativa, on the basis of our previous study where we had found that sucrose acts as a competitive inhibitor of Fenugreek b-amylase whereas maltose does not (Srivastava and Kayastha, 2014). Sucrose docked at the active site of b-amylase from M. sativa showed a frequency of 70%. Glu187 and Glu381 were found to be present at a distance of 2.32 Å and 2.76 Å, respectively (Fig. 5b). Glu187 forms H-bond with sucrose involving hydrogen atom (H2) of sucrose and oxygen atom of its side chain (OE1, OE2). Glu381 also being present in the vicinity takes part in polar interactions with oxygen atom of its side chain (OE1, OE2) and hydrogen atoms of sucrose (H7, H8). Interactions other than H-bond and polar interactions, involving oxygen atom of side chain of Glu381 (OE2) and carbon atom (C11) of sucrose was also observed. Similar interactions were also found to take place between carbon (CD, CG) and oxygen atoms of side chain (OE1, OE2) of Glu187 and hydrogen (H2), carbon (C2, C3) and oxygen atom (O4) of sucrose. Maltose docked at the same active site showed a frequency of 30%. Maltose gets involved in H-bond formation with Asp54, Ala185, Glu381 and Ala383 (Fig. 5c). The Glu187 was found to be present at a distance of 3.22 Å from maltose. No polar interaction between Glu187 and maltose was observed within a cut-off radius of 3.5 Å. This residue was found to get involved in interactions other than H-bond and polar interactions involving carbon atoms of side chain (CD, CG) and hydrogen atoms (H7, H8) of maltose. The value of electrostatic energy (0.06 kcal/mol for sucrose and 0.52 kcal/mol for maltose), total intermolecular energy (4.91 kcal/mol for sucrose and 7.20 kcal/mol for maltose) and the combined value for van der Waals force, H-bond and desolvation energy (4.85 kcal/mol for sucrose and 6.67 kcal/mol for maltose, respectively), was lower in case of sucrose docked at the active site, thus indicated better stability of this structure compared to the maltose docked at the same site. Also, the analysis of residue proximities in docked structure of sucrose and maltose revealed that the Glu187 is involved in interaction with sucrose and not in the case of maltose. Glu186 was reported to initiate the reaction by donating a proton to the glycosidic oxygen of the substrate in soybean b-amylase, followed by attack of the water molecule activated by Glu380 on the carbon atom of cleaved product from above, resulting in an inversion of product (Kang et al., 2004). Nitta et al. (1989) had identified in soybean b-amylase the carboxyl group

G. Srivastava et al. / Plant Physiology and Biochemistry 83 (2014) 217e224

223

Fig. 5. (a) Homology model of b-amylase from Medicago sativa generated using the soybean b-amylase (PDB ID: 1Q6C) as template, pink colour in the figure represents fingerprint residues of Fenugreek b-amylase and blue colour represents active site residues present in the fingerprint. Docking orientation of the (b) sucrose and (c) maltose in the active binding site of b-amylase from Medicago sativa. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

of Glu186 as the catalytically active residue in a unique peptide sequence labelled by an affinity reagent. In sweet potato b-amylase also Glu187 was identified as a potent candidate for direct involvement in catalysis (Toda et al., 1993). Thus, from the above results it is evident that the sucrose can act as inhibitor of b-amylase from M. sativa. The involvement of Glu381 in interaction with maltose indicates the reactivity of this group at the active site. Also, Glu381 was identified as an active site residue based on sequence alignment and superimposition of template and target model. These observations suggest its probable role in catalysis. Therefore, it can be concluded that Glu187 and Glu381 might play role of acidebase pair in hydrolysis of starch by b-amylase from M. sativa. Glu381 was also found to be conserved in Fenugreek b-amylase, as observed from the MALDI based peptide mass fingerprint therefore, catalytic role can be assigned to it. 4. Conclusion It can be concluded that chemical modification of amino acids by specific reagents and bioinformatics analysis has led to recognition

of one cysteine residue (Cys344) and two glutamic acid residues (Glu187 and Glu381) at the active site of Fenugreek b-amylase. Acknowledgement This work was supported by the Council of Scientific and Industrial Research (CSIR), New Delhi, India, in the form of Junior and Senior research fellowships to Garima Srivastava. AMK would like to thank UGC-UPE Grant (4204) (University Grant CommissionUniversity for Potential Excellence) from Banaras Hindu University for financial assistance. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.plaphy.2014.08.005. Contributions Conceived and designed the experiments: GS, AMK. Performed the experiments: GS, VKS.

224

G. Srivastava et al. / Plant Physiology and Biochemistry 83 (2014) 217e224

Analyzed the data: GS, VKS, AMK. Contributed reagents/materials/analysis tools: AMK. Wrote the paper: GS. References Basch, E., Ulbricht, C., Kuo, G., Szapary, P., Smith, M., 2003. Therapeutic applications of fenugreek. Altern. Med. Rev. 8, 20e27. Bernfeld, P., 1955. Amylases a and b. In: Methods in Enzymology, vol. 1. Academic Press Inc Publishers, New York. Bray, M.R., Clarke, A.J., 1994. Identification of a glutamate residue at the active site of xylanase A from Schizophyllum commune. Eur. J. Biochem. 219, 821e827. Colovos, C., Yeates, T., 1993. Verification of protein structures: patterns of nonbonded atomic interactions. Protein Sci. 2, 1511e1519. Dirk, L.M.A., van der Krol, A.R., Vreugdenhil, D., Hilhors, H.W.M., Bewley, J.D., 1999. Galactomannan, soluble sugar and starch mobilization following germination of Trigonella foenum-graecum seeds. Plant Physiol. Biochem. 37, 41e50. Dixon, M., 1953. The determination of enzyme inhibitor constants. Biochem. J. 55, 170e171. Ellman, G.L., 1959. Tissue sulfhydryl groups. Arch. Biochem. Biophys. 82, 70e77. Gana, J.A., Kalengamaliro, N.E., Cunningham, S.M., Volenec, J.J., 1998. Expression of b-amylase from alfalfa taproots. Plant Physiol. 118, 1495e1506. Henrissat, B., 1991. A classification of glycosyl hydrolases based sequence similarities amino acid. Biochem. J. 280, 309e316. Higashihara, M., Okada, S., 1974. b-Amylase of Bacillus megaterium no. 32. Agric. Biol. Chem., 1023e1029. Hirata, A., Adachi, M., Sekine, A., Kang, Y.N., Utsumi, S., Mikami, B., 2004. Structural and enzymatic analysis of soybean b-amylase mutants with increased pH optimum. J. Biol. Chem. 279, 7287e7295. Hoare, D.G., Koshland, D.E., 1967. A method for the quantitative modification and estimation of carboxylic acid groups in proteins. J. Biol. Chem. 242, 2447e2453. Hoschke, A., Laszlo, E., Hollo, J., 1980. A study of the role of histidine side-chains at the active centre of amylolytic enzymes. Carbohydr. Res. 81, 145e146. Kang, Y.N., Adachi, M., Utsumi, S., Mikami, B., 2004. The roles of Glu186 and Glu380 in the catalytic reaction of soybean b-amylase. J. Mol. Biol. 339, 1129e1140. Kumar, S., Kayastha, A.M., 2010. Soybean (Glycine max) urease: significance of sulfhydryl groups in urea catalysis. Plant Physiol. Biochem. 48, 746e750. Kumari, A., Singh, V., Kayastha, A., 2013. Molecular modeling of a-amylase from germinated soybean (Glycine max) and its functional diversity. Int. J. Genom. Proteomics 4, 64e71. onard, N., De Bolle, X., Depiereux, E., 2002. ESyPred3D: prediction of Lambert, C., Le proteins 3D structures. Bioinformatics 18, 1250e1256. Lizotte, P.A., Henson, C.A., Duke, S.H., 1990. Purification and characterization of pea epicotyl b-amylase. Plant Physiol. 92, 615e621. Lovell, S.C., Davis, I.W., Arendall III, W.B., de Bakker, P.I.W., Word, J.M., Prisant, M.G., Richardson, J.S., Richardson, D.C., 2002. Structure validation by C alpha geometry: phi, psi and C beta deviation. Proteins Struct. Funct. Genet. 50, 437e450. Lowry, O.H., Rosebrough, N.J., Farr, L., Randall, R.J., 1951. Protein measurement with the folin phenol reagent. J. Biol. Chem. 193, 265e275.

Lundblad, R.L., Noyes, C.M., 1984. In: Chemical Reagents for Protein Modification, vol. I. CRC Press, West Palm Beach, Florida. Mikami, B., Degano, M., Hehre, E.J., Sacchettini, J.C., 1994. Crystal structures of soybean b-amylase reacted with b-maltose and maltal: active site components and their apparent roles in catalysis. Biochemistry 33, 7779e7787. Mikami, B., Hehre, E.J., Sato, M., Katsube, Y., Hirose, M., Morita, Y., Sacchettini, J.C., 1993. The 2.0 Å resolution structure of soybean b-amylase complexed with cyclodextrin. Biochemistry 32, 6836e6845. Miles, E.W., 1977. Modification of histidyl residues in proteins by diethylpyrocarbonate. Methods Enzymol. 47, 431e442. Murao, S., Ohyama, K., Arai, M., 1979. b-Amylases from Bacillus polymyxa no. 72. Agric. Biol. Chem. 43, 719e726. Nitta, Y., Isoda, Y., Toda, H., Sakiyama, F., 1989. Identification of glutamic acid 186 affinity-labeled by 20,30-epoxypropyl a-D-glucopyranoside in soybean bamylase. J. Biochem. 105, 573e576. Ollis, D.L., Cheah, E., Cygler, M., Dijkstra, B., Frolow, F., Franken, S.M., Harel, M., Remington, S.J., Silman, I., Schrag, J., Sussman, J.L., Verschueren, K.H.G., Goldman, A., 1992. The a/b- hydrolase fold. Prot. Eng. 5, 197e211. Oyama, T., Kusunoki, M., Kishimoto, Y., Takasaki, Y., Nitta, Y., 1999. Crystal structure of b-amylase from Bacillus cereus var. mycoides at 2.2 Å resolution. J. Biochem. 125, 1120e1130. Reid, J.S.G., 1971. Reserve carbohydrate metabolism in germinating seeds of Trigonella foenum-graecum L. (Leguminosae). Planta 100, 131e142. Shinke, R., Kunimi, Y., Nishira, H., 1975. Production and some properties of b-amylases of Bacillus sp. BQIO. J. Ferment. Technol., 693e697. Singh, B.K., Sarkar, N., Jagannadham, M.V., Dubey, V.K., 2008. Modeled structure of trypanothione reductase of Leishmania infantum. BMB Rep. 41, 444e447. Singh, A.N., Yadav, P., Dubey, V.K., 2013. cDNA cloning and molecular modeling of procerain B, a novel cysteine endopeptidase isolated from Calotropis procera. PLoS One 8, e59806. Srivastava, G., Kayastha, A.M., 2014. b-Amylase from starchless seeds of Trigonella foenum-graecum and its localization in germinating seeds. PLoS One 9, e88697. Srivastava, P., Kayastha, A., 2000. Significance of sulfhydryl groups in the activity of urease from pigeonpea (Cajanus cajan L.) seeds. Plant Sci. 159, 149e158. Tanaka, N., Kajimoto, S., Mitani, D., Kunugi, S., 2002. Effects of guanidine hydrochloride and high pressure on subsite flexibility of b-amylase. Biochim. Biophys. Acta 1596, 318e325. Thoma, J.A., Koshland, D.E., 1960. Three amino acids at the active site of b-amylase. J. Mol. Biol. 2, 169e170. Thoma, J.A., Koshland, D.E., Shinke, R., Ruscica, J., 1965. Influence of sulfhydryl groups on the activity of sweet potato b-amylase. Biochemistry 4, 714e722. Toda, H., Nitta, Y., Asanami, S., Kim, J.P., Sakiyama, F., 1993. Sweet potato b-amylase. Primary structure and identification of the active-site glutamyl residue. Eur. J. Biochem. 216, 25e38. Wiederstein, M., Sippl, M., 2007. ProSA-web: interactive web service for the recognition of errors in three-dimensional structures of proteins. Nucleic Acids Res. 35, W407eW410. Zherebtsov, N.A., 1968. Concerning the catalytic site in b-amylase. Biokhimiya 33, 435e444.