Screening of serine protease inhibitors with

View Article Online View Journal

Analyst Accepted Manuscript This article can be cited before page numbers have been issued, to do this please use: S. M. Mandal, W. F. Porto, D. De, A. Phule, S. Korpole, A. K. Ghosh, S. Roy and Prof. O. L. Franco, Analyst, 2013, DOI: 10.1039/C3AN01132A.

This is an Accepted Manuscript, which has been through the RSC Publishing peer review process and has been accepted for publication.

Analyst Interdisciplinary detection science www.rsc.org/analyst

Volume 135 | Number 1 | January 2010 | Pages 1–196

Accepted Manuscripts are published online shortly after acceptance, which is prior to technical editing, formatting and proof reading. This free service from RSC Publishing allows authors to make their results available to the community, in citable form, before publication of the edited article. This Accepted Manuscript will be replaced by the edited and formatted Advance Article as soon as this is available. To cite this manuscript please use its permanent Digital Object Identifier (DOI®), which is identical for all formats of publication. More information about Accepted Manuscripts can be found in the Information for Authors.

ISSN 0003-2654

COMMUNICATION R. Graham Cooks et al. Molecular imaging of adrenal gland by desorption electrospray ionization mass spectrometry

HOT ARTICLE David C. Muddiman et al. Improving limits of detection for B-type natriuretic peptide using PC-IDMS: an application of the ALiPHAT strategy

0003-2654(2010)135:1;1-O

Please note that technical editing may introduce minor changes to the text and/or graphics contained in the manuscript submitted by the author(s) which may alter content, and that the standard Terms & Conditions and the ethical guidelines that apply to the journal are still applicable. In no event shall the RSC be held responsible for any errors or omissions in these Accepted Manuscript manuscripts or any consequences arising from the use of any information contained in them.

www.rsc.org/analyst Registered Charity Number 207890

Page 1 of 10

Analyst View Article Online

DOI: 10.1039/C3AN01132A

Analyst Accepted Manuscript

Published on 05 November 2013. Downloaded by Memorial University of Newfoundland on 10/11/2013 18:42:43.

A novel nanoapproach associated to in silico techniques allowed a rapid selection of antimicrobial serine proteinase inhibitors.

Analyst

Analyst

Page 2 of 10 Dynamic Article Links► View Article Online

DOI: 10.1039/C3AN01132A

Cite this: DOI: 10.1039/c0xx00000x

ARTICLE

Screening of serine protease inhibitors with antimicrobial activity by iron oxide nanoparticles functionalized with dextran conjugated trypsin and in silico analyses of bacterial serine protease inhibition Santi M. Mandala, William F. Portob, Debasis Dec, Ajit Phulec, Suresh Korpoled, Ananta K. Ghosha, Sanat c b 5 K. Roy , Octavio L. Franco * Received (in XXX, XXX) XthXXXXXXXXX 20XX, Accepted Xth XXXXXXXXX 20XX DOI: 10.1039/b000000x

10

15

20

Plants produce a variety of proteins and peptides which are involved in their defense against pathogens. Serine protease inhibitors are a well-established class correlated with plant defense. Increased levels of protease inhibitors delay cell damage and expand the cell’s life-span. Recently, the rapid emergence of antibiotic-resistant microbial pathogens has prompted an immense interest in purifying novel antimicrobial proteins or peptides from plant sources. Usually, the purification of protease inhibitors is accomplished by salt-extraction, ultrafiltration and affinity chromatography. Here, we developed a novel approach based on iron oxide nanoparticles conjugated to dextran functionalized with trypsin beads that accelerate the quick screening and purification of antimicrobial peptides with serine protease inhibitor activity. The method described here also works for screening other inhibitors using particular protein kinases, and it is therefore a novel tool for use as the leading method in the development of novel antimicrobial agents with protease inhibitory activity. Finally, and no less important, molecular modelling and dynamics studies of a homologous inhibitor here studied with Escherichia coli trypsin and chymotrypsin are provided in order to shed some light on inhibitor-enzyme interactions.

Introduction

25

30

35

40

45

Proteases are a group of proteolytic enzymes that hydrolyze the specific peptide bonds in target proteins, and they have been classified on the basis of catalytic site amino acid residues, into groups of cysteine proteases, serine proteases, aspartic proteases, threonine proteases and metalloproteases1. Serine proteases are the largest group among them and associated with controlling a number of biochemical pathways, including fibrinolysis, fertilization, and inflammation. They are also involved in causing many diseases and are often the targets of therapeutic interventions2. In addition, proteases are involved in enzymatic modification of food proteins, leading to food spoilage3. Protease inhibitors (PIs) are mostly used to control the unwanted degradation of proteins.Plant serine protease inhibitors belong to groups of small molecular proteins and peptides, such as storage or tissue-specific proteins. They have received attention due to their potential defense function, which involves inhibiting digestive proteases of herbivorous insects and pathogens4. The food industry is also interested in PIs as a way to add value to achieve higher profits, since enzymatic modification of food proteins has an important role in delaying food spoilage3. Protease inhibitors in food and food ingredients can reduce the absorption of free amino acids and prevent protein hydrolysis in the industrial process. In addition, several PIs have shown This journal is © The Royal Society of Chemistry [year]

50

55

antimicrobial activity5,6, which could improve the industrial process, since in addition to avoid protein degradation; these PIs could reduce the chance of bacterial contamination. Given the importance of plant PIs, several methods have been developed to identify plant serine protease inhibitors7. The present paper proposes a novel method for screening serine protease inhibitors with antimicrobial activity from plants. The method was applied to identify a Bowman-Birk protease inhibitor with antimicrobial activity from Pisumsativum seeds (Ps-BBI). In addition, molecular modelling and dynamics were used to shed some light over the interactions of a Ps-BBI homologue with two bacterial proteinases.

Materials and methods Preparation of IONPs and dextran-coated IONPs (DIONPs) 60

65

70

Iron oxide nanoparticles (IONPs) were synthesized following the protocol described earlier by Mandalet al.8. In brief, the IONPs used in this experiment were synthesized from a precursor of Fe3+ species dispersed in polyvinyl alcohol (PVA) in water. Aqueous solutions of Fe(NO3)3.9H2O (1M) and PVA (30 g.L-1) were prepared by stirring in hot conditions at 60~70°C. Aqueous sucrose solution (30 g.L-1) was mixed with the PVA solution by stirring at 55-65°C and used as a dispersive medium to obtain a metal ion-polymer precursor (Fe3+-sucrose-PVA). In this process, the nitrate solution was added drop wise into this solution with stirring at 60°C for 30 min. A Fe3+-sucrose-PVA complex thus [journal], [year], [vol], 00–00 |1



www.rsc.org/xxxxxx

Page 3 of 10


DOI: 10.1039/C3AN01132A

10

Phase and microstructure analysis

15

20

25

The phase analysis and crystal structure of IONPs was studied in terms of XRD. The XRD pattern was recorded using a diffractometer from Rigoku Instruments. Particle size and morphology were studied with a high-resolution transmission electron microscope (HRTEM) from JEOL (JEM-2100), with an operating voltage of 200 kV. In a sample preparation for HRTEM studies, a small amount of IONPs and DIONPs was dispersed in acetone and sonicated for 30 min. Part of this dispersion was dropped onto a carbon film supported by a copper grid and dried in a vacuum before imaging.

55

60

HPLC separation

65

70

75

MALDI-TOF-MS analysis

IR spectra of the samples in the transmission mode were measured in the liquid form (samples) against deionized water by FTIR spectrometer (Bruker, Tensor 27). A small drop of sample was put in between two quartz cells (circular plates) and data were recorded over the range of 400-4000 cm-1.

The lyophilized fractionated peptides were re-suspended in 80% (v/v) acetonitrile solution containing 0.1% (v/v) trifluoroacetic acid, and 2 µL microliters of peptide solution was mixed with 4 µL of matrix (α-cyano-4-hydroxy-cinnamic acid, 10 mg.mL-1). One microliter of this mixture solution was spotted onto the MALDI 100-well stainless steel sample plate and allowed to air-dry prior to the MALDI ToF analysis following Mandalet al.11.

80

Absorption spectra of IONP and DIONP samples (filled in the transparent quartz cell) were measured against deionized water by UV-visible spectrometer (Perkin Elmer, Lamda 750) over a range of 200-1000 nm.

85

Antimicrobial assays

CNBr- activation and trypsin coupling 35

40

After the preparation and confirmation of dextran-coated IONPs, the particles were activated using cyanogen halides (CNBr), a process which consists of linking primary amino groups directly to the pre-activated particle. The activated beads were washed several times with PBS (1X) and trypsin (10 mg.mL-1), and were then incubated overnight in a cold room with mild shaking. After incubation, they were washed gently with PBS (1X) buffer and were ready for use.

90

95

Sample application and elution

45

50

The extracted lyophilized proteins and peptides were dissolved in 5% acetic acid solution and fractionated by reverse phaseHPLC (Agilent 1100 series) with a ZORBAX 300-SB18 column (4.6 mm × 250 mm, particle size 5 µm), at a flow rate of 1 mL.min-1 following Mandal et al.10. The solvent was 0.1% aqueous TFA (A) and 80% acetonitrile containing 0.1% TFA (B). A step gradient of solvent B used to run the column was as follows: 0-60% for 0-45 min, 60-80% for 45-55 min and 80100% for 55-60 min. All the solvents of HPLC grade were purchased from Spectrochem, India. The elution from the column was monitored at 215 nm in a diode array detector and all the peaks of the HPLC chromatogram were collected using a fraction collector (GILSON, France) coupled with the system. Collected fractions were concentrated by lyophilization, and antimicrobial and serine protease inhibition activities were screened.

FTIR analysis

UV-Vis spectroscopy analysis 30

washed repeatedly with 100 mMTris-HCl buffer and, finally, the bound inhibitors were eluted with 0.3% aqueous TFA solution. After eluting twice with TFA solution, the beads were immediately neutralized with 100 mMTris-HCl buffer (pH 8.0). The eluents were pooled and lyophilized, to be used later for HPLC separation.

Pisum sativum seeds were ready for germination after overnight soaking in water, and radicles started to emerge after two days of incubation at room temperature. The germinating seeds were frozen in liquid nitrogen and ground well with pestle and mortar. The obtained powder was homogenized with 0.5% aqueous acetic acid solution and incubated overnight in a rotary shaker at 120 rpm. The resulting acetic acid extracted solution was centrifuged (12,000 rpm) repetitively and the filtrate solution was applied to trypsin-activated dextran-coated IONPs in 50 mM Tris-HCl buffer (pH 8.0), and left for 1h incubation in a cold room with mild shaking. After sample incubation, the sample was 2|Journal Name, [year], [vol], 00–00

100

S. epidermidis NCIM2493 and E. coli ATCC8739 were used for antimicrobial bioassays. The bacterial species were cultured in 1.0 mL LB broth for 2 h at 37ºC to prepare the inoculum. Bacterial growth and the period of incubation were carried out according to protocols described by CLSI guidelines12. Each experiment was carried out in triplicate. To determine the minimum inhibitory concentration (MIC) values, purified proteins and peptides were serially diluted from 256 to 1 µg.mL-1 in LB medium at final concentration.MIC was determined where no visible growth was observed. In each well of a 96-well polypropylene plate, 100 µL of each peptide or protein dilution and 10 µL of bacterial inoculum were added (approximately 5X106 cfu.ml-1 is the final concentration in the plate) following the protocol described earlier by Mandalet al.13. Proteinase inhibition assays

105

The inhibitory activity of the purified fractions against trypsin was performed as described by Calderon et al.14. In brief, chromogenic substrates N-benzoyl-L-arginine-p-nitroaniline (BAPNA) were used to determine the inhibition of trypsin activity. Each purified fraction (0.1 mg.mL-1) of 100 µL was incubated with equal volume of trypsin (0.064 mg.mL-1) at 25ºC for 15 min. After incubation, 500 µL of BAPNA (at a final concentration of 0.5 mM in 0.1 M Tris–HCl buffer, pH 8.2) was This journal is © The Royal Society of Chemistry [year]



5

obtained was aged for a few hours at room temperature and then at ice cooled temperature. Next, an ammonia solution was added drop-wise until a gel appeared at pH 12-14. IONPs were obtained after firing a dried gel (at 80-90°C in air) at 300°C in air. To prepare the dextran-coated IONP, first dextran (MW 10000) was reduced with sodium borohydride at room temperature for 12 h and the reaction was stopped by drop wise addition of 6N HCl until pH 7. Then the reduced dextran was dialyzed with membrane (Cut off MW~ 3500 Da) against distilled water. Next to that, the reduced dextran coated IONPs were synthesized according to the method described by Jarrett et al.9.

Analyst

Page 4 of 10 View Article Online

DOI: 10.1039/C3AN01132A

10

60

65

N-terminal amino acid sequencing

15

20

Fraction 2, which showed both antimicrobial and serine protease activity, was run onto Tricine-SDS-PAGE, further transferred to PVDF membrane (Bio-Rad), stained with amido black (Sigma) for 2-3 min, marked and de-stained with repetitive washing with 50% methanol. The membrane was finally rinsed in Milli-Q water. The marked region of the membrane was cut and subjected to N-terminal sequencing. The N-terminal sequence was determined by automatic degradation in Procise 491 CLC protein sequencer (Applied Biosystem, USA).

70

75

In-solution digestion and MS/MS analysis

25

30

35

40

HPLC purified fraction-2 was digested using endoprotinase Asp-N and Glu-C. The lyophilized dried fraction was initially dissolved in 50 mM ammonium bicarbonate (pH 8.0) solution. Then 20 µL of peptide solution (0.26 µg/ µL ) was taken in a tube followed by reduction with 1 µL of 0.1M DTT at 56°C for 30 min. Afterward, alkylation of the solution was done with 1 µL of 0.3 M iodoacetamide at room temperature for 15 min in the dark. After incubation the solution was subjected to digestion with endoproteinase Asp-N and Glu-C were used in combined at 1µL each of 1 mg. mL-1 and incubate at 37°C for overnight. Two microliter of digested sample solution was spotted directly onto a MALDI target plate and 1 µL of CHCA matrix solution was applied on the sample spot and allowed to air dry. The spectra were recorded in same MALDI MS using the post-source decay (PSD) ion mode as average of 100 laser shots with a grid voltage of 75%. The reflector voltage was reduced in 25% steps and guide wire was reduced 0.02-0.01 % with an extraction delay time 100 ns. Reproducibility of the spectrum was checked 3 times from separately spotted samples.

80

85

90

95

In silico analyses and molecular modelling

45

50

55

Initially, the Pisumsativum Bowman-Birk inhibitor (Ps-BBI) N-terminal sequence was used to search for a complete sequence among the Pisumsativum (taxid: 3888) proteins deposited in the NCBI protein databank through PSI-BLAST15. However, no Bowman-Birk inhibitors were found. Therefore, the search was done again against the higher plant (taxid: 3193) proteins from the NCBI protein data bank through PSI-BLAST. The BowmanBirk inhibitor DE-3 (BBI DE-3) from Macrotylomaaxillare (SwssProt ID: P01057)16 was selected as a prototype for Ps-BBI. The sequences of E. coli chymotrypsin (GenBank ID: YP_006100904) and trypsin (GenBank ID: EII35409) were selected as the prototypes for bacterial chymotrypsin and trypsin, respectively. The prototype sequences were submitted to Phobius17 for signal peptide prediction, and then the signal peptides were removed, in order to perform the molecular This journal is © The Royal Society of Chemistry [year]

100

105

modelling. Therefore, one hundred molecular models for each complex (BBI-chymotrypsin and BBI-trypsin) were constructed by comparative molecular modelling by MODELLER 9.1018, using the structure of BBI BTCI from Vigna unguiculata in complex with chymotrypsin and trypsin from Bos taurus (PDB ID: 3RU4). The models were constructed using the default methods of automodel and environ classes from MODELLER. The final models were selected according to the discrete optimized protein energy score (DOPE score). This score assesses the energy of the model and indicates the best probable structures. The best models were evaluated through PROSA II19, PROCHECK20. PROCHECK checks the stereochemical quality of a protein structure, through the Ramachandran plot, where good quality models are expected to have more than 90% of amino acid residues in the most favoured and additional allowed regions, while PROSA II indicates the fold quality. Structural visualization was done in PyMOL (The PyMOL Molecular Graphics System, Version 1.4.1, Schrödinger, LLC). Molecular Dynamics The molecular dynamics simulations (MD) of the inhibitorenzyme complexes were carried out in water environment, using the Single Point Charge water model21. The analyses were performed by using the GROMOS96 43A1 force field and computational package GROMACS 422. The dynamics used the three-dimensional models of the enzyme-inhibitor complexes as initial structures, immersed in water molecules in cubic boxes with a minimum distance of 0.7 nm between the complexes and the boxes’ frontiers. Sodium ions were also inserted in the complexes in order to neutralize the system charge. Geometry of water molecules was constrained by using the SETTLE algorithm23. All atom bond lengths were linked by using the LINCS algorithm24. Electrostatic corrections were made by Particle Mesh Ewald algorithm25, with a cut-off radius of 1.4 nm in order to minimize the computational time. The same cut-off radius was also used for van der Waals interactions. The list of neighbours of each atom was updated every 10 simulation steps of 2 fs. The system underwent an energy minimization using 50,000 steps of the steepest descent algorithm. After that, the system temperature was normalized to 300 K during 100 ps, using the velocity rescaling thermostat (NVT ensemble). Then the system pressure was normalized to 1 bar during 100 ps, using the Parrinello-Rahmanbarostat (NPT ensemble). The systems with minimized energy, balanced temperature and pressure were simulated for 30 ns by using the leap-frog algorithm. The trajectories were evaluated through RMSD. In addition, the initial and the final structures were compared through the TM-Score26, where structures with TM-Scores above 0.5 indicate that the structures share the same fold.

Results and Discussion Screening System Preparation

110

To date, several methods have been developed to isolate and purify the serine protease inhibitors from natural sources7. Immobilization of serine proteases (trypsin, chymotrypsin and kallikrein etc.) in Sepharose beads is a common tool for the purification of serine protease inhibitors due to its availability and simplicity. The usefulness of protease immobilization for Journal Name, [year], [vol], 00–00 |3



5

added to reaction tube and allowed to incubate for 10 min. Finally, 100 µl of 30 % acetic acid solution was added to stop the reactions. The relative trypsin activities were evaluated by the release of p-nitroanilide from substrate, which was measured at 410 nm. Inhibition curves were obtained by plotting decreasing residual enzymatic activities against inhibitor/enzyme molar ratio. All assays were performed in triplicate. Differences between fractions were compared by using ANOVA and Tukey Test. Differences were considered significant when p value was less than 0.05.

Page 5 of 10



45

50

55

5

Fig. 1Schematic diagram of the synthesis and development of trypsin-conjugated dextran-coated iron oxide nanoparticles and their uses for high-throughput screening of serine protease inhibitors. proteinase inhibitors purification was shown for the first time by Fritz et al.27. Here, we have tested a novel approach using affinity matrices where trypsin was immobilized on CNBr-activated dextran-coated iron oxide nanoparticles (IONPs) for easy purification of proteinase inhibitors from plants (Fig 1).

60

65

The shape and size of IONPs and dextran coated-IONPs have been investigated using HR-TEM images. The bright field images, as can be seen in Fig. 2, revealed nearly rhombohedral shaped particles of IONPs. Most IONPs were of average size ~40 nm, which is in close agreement with an average D-value 34 nm determined from the XRD peak broadening. The microstructure of dextran-coated IONPs is displayed in Fig. 2b, which clearly shows that a thin layer (~5 nm) of dextran is coated on the surface of IONPs. The halo rings embedded with disordered bright dots appeared in the SAED (selected area electron diffraction) pattern. The halo rings in SAED patterns may stem from dextran, a polymer compound coated on IONPs, and the bright dots may come from the crystallite nanoparticles. A typical XRD pattern, as given in Fig. 2c, characterizes the formation of a single-phase sample of IONPs after heating an Fe3+-sucrose-PVA polymer. For the IDT (Dextran coated IONP with Trypsin in deionised water) sample all corresponding absorptions have been observed (data not shown). FTIR spectra exhibit the characteristic nanoparticles also showed O–H strong absorption bands at around 3442 cm-1 because of physically adsorbed water; this showed that the surface of iron oxide particles had large numbers of hydroxyls. In addition, two broad bands placed at about 583 and 441 cm-1 correspond to Fe–O vibration modes (Fig. 2d).

10 70

15 75

20 80

25 85

30 90

35 95

40

Fig. 2. Characterization of trypsin-conjugated iron oxide nanoparticles. TEM images of chemically synthesized iron oxide nanoparticles (a); TEM images of chemically synthesized iron oxide nanoparticles conjugated with dextran and trypsin (b); XRD pattern of synthesized nanoparticles (c); FTIR overlay 100 spectrum from iron oxide nanoparticle (IONP), from only dextran (D); from trypsin (T) and from IONP conjugated dextran with trypsin (IDT).

4|Journal Name, [year], [vol], 00–00

This journal is © The Royal Society of Chemistry [year]


DOI: 10.1039/C3AN01132A

Analyst


DOI: 10.1039/C3AN01132A

10

15

45

completely elute the inhibitors from trypsin, and the stability of trypsin functionalized beads was also verified in a wide pH range (3-10). Below pH 3.0 and above pH 10, trypsin also released inhibitors (data not shown).

50

55

60

65

70

75

20

Fig . 3. Purification and MS analysis of serine protease inhibitors attached to developed beads. Reversed-phase HPLC chromatogram profile of inhibitors eluted from beads (a); mobile phase and other conditions are described in the text. MALDI TOF mass spectrum of individual fractions collected during HPLC separation and marked with corresponding letter in the picture. The HPLC peaks are corresponding to MS spectra as b (F1); c 25 (F2); d (F3); e (F4) and f (F5).

80

85

30

35

40

The polysaccharide-based dextran exhibited a good mechanical and chemical stability for the coupling and elution conditions, and minimal nonspecific interaction with proteins; it is therefore widely used as a coupling material33. Cross-linked dextran-IONPs immobilized the protease trypsin by covalent bonding. Interestingly, dextran-conjugated IONPs offer several points of connection for each particle, and part of trypsin enzymes have been immobilized, which reduces the chance of non-specific interaction of proteins and increases the effectiveness of the interaction of the inhibitor with trypsin. Moreover, absence of any porous nature reduced the chance of retaining any non-specific large molecules which might have a chance during Sepharose column flow. After incubation with samples, several washes with 100 mM Tris-HCl buffer completely removed any non-specific proteins or peptides from the functionalized bead. The elution was performed in acidic conditions. Only aqueous 0.3 % TFA solution was optimized to This journal is © The Royal Society of Chemistry [year]

90

95

100

Fig. 4. Three-dimensional models of BBI DE-3 and E. coli trypsin and chymotrypsin. (A) The molecular models from BBI DE-3 from trypsin complex (left panel) and chymotrypsin complex (right panel). (B) The molecular models of trypsin (left panel) and chymotrypsin (right panel). (C) The catalytic triad of trypsin (left panel) and chymotrypsin (right panel); the electrostatic interactions are represented as yellow dashed lines, and the distances are in Ångströms Screening System Application In order to determine the effectiveness of synthesized beads, a routine test of seed extract from P. sativum was applied and eluted with the same protocol as described above. The eluted extract was run into RP-HPLC and fractionated with a 300SBC18 column. It was observed that five major fractions appeared and subsequently concentrated with lyophilization (Fig. 3a). Each fraction was analyzed with MALDI TOF MS to determine its molecular mass; the range between 5 kDa to 16 kDa was observed (Fig. 3b-3f). All the fractions were evaluated for their trypsin inhibition and antimicrobial activity. Maximum trypsin activity was inhibited by fraction 4, whereas only fraction 2 exhibited both enzyme inactivation and significant antimicrobial activity among them, which has been compared by using ANOVA and Tukey Test. Differences were considered

Journal Name, [year], [vol], 00–00 |5



5

The Fe2O3 nanoparticle (IONP) sample was observed to provide absorption in the UV region, which corresponds to the direct charge transfer transitions of O2− 2p→Fe3+ 3d, and the absorption in the visible region is due to the Fe3+ 3d→3d indirect transition28. Trypsin (T) has two absorption peaks, of which the weak one at 274 nm appears because of the three buried aromatic amino acids (Trp, Tyr, and Phe)29. However, there was an obvious decrease in absorbance at around 200 nm, where there is the strong absorption peak mainly due to the transition of π → π* of trypsin’s characteristic polypeptide backbone structure C=O30.complex in air at 300°C for 2 h. All the XRD peaks in this pattern are indexed, assuming the usual rhombohedral crystal structure, with lattice parameters a = 0.5037nm, c = 1.3745 nm and γ = 109.4°, of hematite (α-Fe2O3). Bulk α-Fe2O3 has very similar values, with a = 0.50355 nm, c = 1.37471 nm and γ = 109.4°, confirming that IONPs prepared with a chemical method in this work form the usual α-Fe2O3 crystal lattice. An analysis of the peak broadening in the characteristic XRD peaks as per the Debye-Scherrer relation yields an average 34 nm crystallite size.

Page 7 of 10


10

15

20

significant when p value was less than 0.05 (data not shown). The MIC value of fraction 2 against S. epidermidis and E. coli was the same as 16 µg. mL-1.The purity of fraction 2 has been confirmed with re-chromatogram HPLC analysis Since the classification of serine protease inhibitors take into account the molecular mass, structural similarity, presence of cysteine residues and disulphide content34,35, the fractions here identified could be grouped into the Bowman-Birk subfamily. Bowman-Birk inhibitors are classified on the basis of molecular weight (8-16 kDa). The average mass was measured by the MALDI TOF-MS method, which showed that they were close to 8 kDa. In addition, the Bowman-Birk inhibitors could have one or two reactive sites (single headed or dual headed), which is also used for classification. Analysis of the reaction mechanism for different inhibitors showed that the first reactive site is located in the N-terminal region and it is more conserved than the second one, which is located in the C-terminal region36. Once the fraction 2 showed antimicrobial activity, it was named Ps-BBI and it was selected for amino acid sequencing. However, only partial Nterminal sequence (1DHHVST6) could be determined.

5

35

40

45

Identification of Inhibitor Showed Antimicrobial Activity The HPLC separated fraction-2 showed both protease inhibitory and antimicrobial activity. The molecular weight was found to be 8040 Da (Fig. 3c) by MALDI-TOF-MS analysis and N-terminal sequence was determined the partial sequence as “DHHVST”. Further, attempt was made on MS/MS analysis of fraction-2 to confirm the protein identification, after digestion with endopeptidase Asp-N and Glu-C in combination. The digested fraction produced several peptide ions as 847.7,1161.0, 1791.7, 1971.3, 3531.856, 6448.260, 6822.718 during MS analysis and after MS/MS sequencing in post source decay (PSD) mode of peptide ion, 1971.3 showed homology with N-terminal sequencing which confirm the ion generated from N-terminal part. The obtained sequence from ion 1971.3 was “DHHVSTDEPSESSKPCC” (data is not shown), which are identical to the N-terminal sequence of Bowman-Birk Inhibitor (BBI DE-3).

50 25

55

60

65

70

75

Fig. 5. The RMSD variation throughout the 30 ns of simulation. (A) trypsin complex and (B) chymotrypsin complex. The red lines indicate the enzymes RMSD and the black ones the BBI DE-3 RMSD.

80

85

30


In silico analyses of BBI DE-3 It was hypothesized that Ps-BBI antimicrobial activity could be related by inhibiting nitrogen absorption from protein source degradation, which would occurs by secretory bacterial proteinase inhibition. In order to clarify this issue, the molecular modelling and further dynamics were performed to verify the inhibition of bacterial proteases by the Ps-BBI homologue, BBI DE-3.The partial N-terminal sequence PS-BBI was used as a seed to PSI-BLAST search into plant proteins from SwissProt. Five hits with Bowman-Birk protease inhibitor sequences were retrieved (data not shown). Of the five hits, the BBI DE-3 sequence, obtained from M. axillare, was selected as a prototype, since it was experimentally validated, being able to inhibit both trypsin and chymotrypsin through two independent sites: Lys24Ser25 and Phe51-Ser52, respectively16. Two E. coli proteases were selected as prototypes for bacterial proteases. The two sequences have a signal peptide predicted according to Phobius. The trypsin has the first 17 residues predicted as the signal peptide, while the chymotrypsin has the first 21 residues predicted. The mature sequences of bacterial proteases have 17% and 21% of identity to bovine trypsin and chymotrypsin from structure 3RU4, respectively. In addition, BBI BTCI shares 84% of identity with BBI DE-3. The structure 3RU4 was selected as a template for molecular modeling for two reasons. First, BBI BTCI is able to inhibit trypsin and chymotrypsin, as well as BBI DE-3, which characterizes the double activity of the inhibitor here evaluated. The second reason is that its structure was solved in a complex with the bovine enzymes, trypsin and chymotrypsin, which allows molecular modeling of the complex to be carried out without docking experiments, as previously described by Porto et al.37. In order to minimize the computational costs, the two complexes with bacterial proteases were modeled separately. The three-dimensional model of BBI DE-3 was composed of two β-sheets with two β-strands, one for each inhibition site, and it was stabilized by seven disulphide bonds. Because the template used for BBI DE-3 modeling did not cover the complete




DOI: 10.1039/C3AN01132A

Analyst


DOI: 10.1039/C3AN01132A

10

15

20

25

30

35

40

45

50

55

Fig. 7. Variation of the number of hydrogen bonds between the BBI DE-3 and the enzymes (A) trypsin and (B) chymotrypsin. 60

65

90

95

70

75

80

85

Fig. 6. Interactions of BBI DE-3 with the enzymes’ active site. (A) Interaction with trypsin, showing the separation of the catalytic triad. (B) Interaction with chymotrypsin, showing that the residue Ser52 from BBI DE-3 interacts with the catalytic triad, taking the place of Ser202 from chymotrypsin.

the same fold. For enzyme structures, a similar situation was observed. The RMSD variations of chymotrypsin and trypsin reached values above 4Å, in both cases (Fig. 6). Since the chymotrypsin three-dimensional model has two disulphide bonds, a higher structural stability was expected in comparison to trypsin, which has no disulphide bonds. Despite the low sequence identity among the bovine enzymes and the E. coli enzymes, the relaxation of the molecular models indicates that the enzymes share the same fold. This is supported by the TM-scores, with This journal is © The Royal Society of Chemistry [year]

100

105

110

values of 0.7243 and 0.7845 for trypsin and chymotrypsin, also indicated that the enzymes share the same fold after the simulation. Protease inhibition occurs in a similar way for trypsin and chymotrypsin. As a competitive inhibitor, BBI DE-3 competes with the enzyme substrates, resulting in a non-covalent complex, which renders the enzyme inactive. In both simulations, the catalytic triad disarticulation was observed as a result of BBI DE3 interaction, where the catalytic serine moves away from the histidine and aspartic acid (Fig 7). In the case of chymotrypsin, the residue Phe51 from BBI DE-3 occupies the catalytic pocket, being stabilized by hydrophobic interactions with the residues Pro199 and Ala223 from chymotrypsin, while the residue Ser52 from BBI DE-3 takes the place of Ser202 from chymotrypsin, inhibiting the formation of the catalytic triad (Fig. 7). For trypsin, the main factor responsible for inhibition by BBI DE-3 seems to be the inhibitor's Lys24, which occupies the catalytic pocket, being stabilized by the backbones of Val173 and Gln185. In contrast to chymotrypsin, Ser25 did not take the place of catalytic serine (Ser174). In this case, Ser174 is distant from His49 and Asp83 without another residue taking its place (Fig. 7). In addition, the BBI DE-3 also interacts with trypsin and chymotrypsin through other residues outside the catalytic triad and pocket by hydrogen bonds, which contribute to the non-covalent complex stabilization (Fig 7).




5

sequence, the yielded model shows two unstructured regions in its N- and C-terminals (Fig. 4).The three-dimensional model of E. coli trypsin was mainly composed of β-strands, composing three β-sheets (Fig. 4).The three-dimensional model of E. coli chymotrypsin was composed of three α-helices and two β-sheets, comprising six and seven β-strands (Fig. 4). Moreover, chymotrypsin was stabilized by two disulphide bonds. The trypsin catalytic triad was composed of His49, Asp83 and Ser174 (Fig. 4), while chymotrypsin was composed of His63, Asp112 and Ser202 (Fig. 4C). The trypsin complex showed a DOPE score of 21967.73828; and the chymotrypsin, -26903.25391. The BBI DE3 models showed a Z-Score on PROSA of -3.39 (chymotrypsin complex) and -3.82 (trypsin complex), indicating that the quality is similar to structures solved by NMR. For the enzymes, Zscores of -2.14 and -2.93 were observed for chymotrypsin and trypsin, respectively, indicating that the quality is similar to structures solved by X-Ray. In the Ramachandran plot, the trypsin complex showed 71.2% of residues in most favoured regions and 21.6% in additional allowed regions, while the chymotrypsin complex showed 80.6% of residues in most favored regions and 15.8% in additional allowed regions. The complexes were evaluated by molecular dynamics, where the BBI DE-3 accommodation was observed throughout the 30 ns of simulation, Therefore the complexes were maintained during the whole simulation. The RMSD variations for BBI DE-3structure reach values above 6Å in the chymotrypsin complex (Fig. 5); and 7Å in the trypsin complex (Fig. 5). These RMSD variations were related to the unstructured N- and C-terminals from the BBI DE-3 model, which were not covered in the alignment with the template. However, in both cases, the TM scores, with values of 0.6718 and 0.5976 indicated the initial and final structures share

Page 9 of 10


DOI: 10.1039/C3AN01132A

Conclusions

10

70

75

80

Notes and references 15

20

25

a

Central Research Facility, Department of Biotechnology, Indian Institute of Technology Kharagpur, Kharagpur 721302, WB, India b Centro de Análises Proteômicas e Bioquímicas, Pós-Graduação em Ciências Genômicas e Biotecnologia UCB, Brasília-DF, Brazi c Department of Metallurgical and Materials Engineering, Materials Science Centre, Indian Institute of Technology, Kharagpur 721302, WB, Indian. d CSIR-Institute of Microbial Technology, Sector 39A, Chandigarh 160036, India. *Correspondingauthor: Centro de Análises Proteômicas e Bioquímicas, Ciências Genômicas e Biotecnologia-UCB-DF. SGAN 916, módulo B, avenida W5, Brasília – DF, Brasil. +55(61) 3448-7220, e-mails: [email protected]; [email protected]

35

40

45

50

55

60

65

H. Habib, K.M. Fazili, Plant protease inhibitors: a defense strategy in plants, Biotech Mol Biol Rev, 2007, 3, 68-85. 2. G. Opdenakker, J. Van Damme, Cytokines and proteases in invasive processes: molecular similarities between inflammation and cancer, Cytokine, 1992, 251-258. 3. T.C. Baird-Parker, The production of microbiologically safe andstable foods, in: B. M. Lund, T. C. Baird-Parker, G. W.Gould and Gaithersburg (Eds.) In the Microbiological Safety andQuality of Food, Aspen Publishers, 2003, 3-18. 4. F. Jamal, P.K. Pandey, D. Singh, M.Y. Khan, Serine protease inhibitors in plants: nature’s arsenal crafted for insect predators, Phytochem Rev,2013,12, 1-34. 5. J.Y. Kim, S.C. Park, M.H. Kim, H.T. Lim, Y. Park, K.S. Hahm, Antimicrobial activity studies on a trypsin-chymotrypsin protease inhibitor obtained from potato, Biochem Biophys Res Commun, 2005, 330(3), 921-927. 6. J.Y. Kim, S.C. Park, I. Hwang, H. Cheong, J.W. Nah, K.S. Hahm, Y. Park, Protease inhibitors from plants with antimicrobial activity, Int J Mol Sci,2009, 10(6), 2860-2872. 7. A.S. Oliveira, L. Migliolo, R.O. Aquino, J.K.C. Ribeiro, L.L.P. Macedo, et al., Purification and characterization of a trypsin-papain inhibitor from Pithecelobium dumosum seeds and its in vitro effects towards digestive enzymes from insect pests, Plant Physiol Biochem, 2007,45, 858-865. 8. S.M. Mandal, A.K. Ghosh, M. Mandal, Iron oxide nanoparticle assisted purification and mass spectrometry based proteolytic mapping of intact CD4+ T cells from human blood.Prep Biochem Biotechnol,2009,39, 20-31. 9. B.R. Jarrett, M. Frendo, J. Vogan, A.Y. Louie, Size-controlled synthesis of dextran sulfate coated iron oxide nanoparticles for magnetic resonance imaging, Nanotechnology, 2007,18, 035603. 10. S.M. Mandal, L. Migliolo, O.L. Franco, A.K. Ghosh, Identification of an antifungal peptide from Trapa natans fruits with inhibitory effects on Candida tropicalis biofilm formation, Peptides, 2011, 32, 1741-1747. 11. S.M. Mandal, L. Migliolo, S. Das, M. Mandal, O.L. Franco, T.K. Hazra Identification and characterization of a bactericidal and

85

90

95

1.

30


proapoptotic peptide from Cycas revoluta seeds with DNA binding properties, J Cell Biochem2012, 113, 184-93. 12. Clinical and Laboratory Standards Institute, Performance standards for antimicrobial susceptibility testing. Seventeenth informational supplement. Document M100-S17. Wayne, PA: CLSI (2007) 13. T. Samanta, G. Roymahapatra, W.F. Porto, S. Seth, S. Ghorai, et al., N, N′-Olefin functionalized bis-imidazolium gold(I) salt is an efficient candidate to control keratitis-associated eye infection, PLoS ONE, 2013,8, e58346.

100

105

110

115

120

125

130

14. L.A. Calderon, R.C.L. Teles, J.R.S.A. Leite, C. Bloch Jr, S. AstolfiFilho, S.M. Freitas, Serine protease inhibitors from Amazon Leguminosae seeds: Purification and preliminary characterization of two chymotrypsin inhibitors from Inga umbratica, Prot Pep Letters,2001,8, 485-493. 15. S.F. Altschul, T.L. Madden, A.A. Schäffer, J. Zhang, Z. Zhang, et al., Gapped BLAST and PSI-BLAST: A new generation of protein database search programs, Nucl Acids Res, 1997,25, 3389-3402. 16. F.J. Joubert, H. Kruger, G.S. Townshend, D.P. Botes, Purification, some properties and the complete primary structures of two protease inhibitors (DE-3 and DE-4) from Macrotyloma axillare seed, Eur J Biochem,1979,97, 85-91. 17. L. Käll, A. Krogh, E.L.L. Sonnhammer, Advantages of combined transmembrane topology and signal peptide prediction--the phobius web server, Nucl Acids Res, 2007,35, W429-W432. 18. N. Eswar, M.A. Marti-Renomet, B. Webb, M.S. Madhusudhan, D. Eramian, et al., Comparative protein structure modeling with MODELLER, Curr Protoc Bioinform, 2007, 15, 5.6.1–5.6.30. 19. M. Wiederstein, M.J. Sippl, ProSA-web: interactive web service for the recognition of errors in three-dimensional structures of proteins, Nucl Acids Res,2007,35, W407-W410. 20. R.A. Laskowski, M.W. Macartur, D.S. Moss, J.M. Thornton, PROCHECK: a program to check the stereochemical quality of protein structures, J Appl Cryst,1993,26, 283-291. 21. H.J.C. Berendsen, J.P.M. Postma, W.F. van Gunsteren, J. Hermans, Interaction models for water in relation to protein hydration, in: B. Pullman (Ed.) Intermolecular Force, Dordrecht, Reidel, 1981, 331342. 22. B. Hess, C. Kutzner, D. van der Spoel, E. Lindahl, GROMACS 4: algorithms for highly efficient, load-balanced, and scalable molecular simulation, J Chem Theory Comput, 2008,4, 435-447. 23. S. Miyamoto, P.A. Kollman, SETTLE. An analytical version of the SHAKE and RATTLE algorithm for rigid water models, J Comp Chem,1992,13, 2134-2144. 24. B. Hess, H. Bekker, H.J.C. Berendsen, J.G.E.M. Fraaije, LINCS. A linear constant solver for molecular simulations, J Comp Chem, 1997, 18, 1463-1472. 25. T. Darden, D. York, L. Pedersen, Particle mesh ewald: an n long(n) method for ewald sums in large systems, J Chem Phys, 1993,98 10089-10092. 26. Y. Zhang, J. Skolnick, Scoring function for automated assessment of protein structure template quality, Proteins,2004,57, 702-710. 27. H. Fritz, G. Wunderer, Biochemistry and applications of aprotinin, the kallikrein inhibitor from bovine organs, Arzneimittelforschung, 1983,33, 479-494. 28. Z. Zhang, M.F. Hossain, T. Takahashi, Fabrication of shapecontrolled α-Fe2O3 nanostructures by sonoelectrochemical anodization for visible light photocatalytic application, Mat Let, 2010,64, 435–438. 29. X.C. Zhao, R.T. Liu, Z.X. Chi, Y. Teng, P.F. Qin, New insights into the behavior of bovine serum albumin adsorbed onto carbon nanotubes: comprehensive spectroscopic studies, J Physic Chem B,2010, 114, 5625–5631 30. Z. Li, D. Abramavicius, W. Zhuang, S. Mukamel, Two-dimensional electronic correlation spectroscopy of the nπ∗ and ππ∗ protein backbone transitions: a simulation study, Chem Phys2007,341, 29–36




5

Here, a novel approach based on the iron oxide nanoparticles conjugated with dextran functionalized with trypsin beads was developed. The method was successfully applied to P. sativum seeds and a Bowman-Birk protease inhibitor was identified (PsBBI). Although the complete sequence from Ps-BBI is still unknown, the formation of the non-covalent complex could be demonstrated by molecular dynamics through a prototype sequence, BBI DE-3, which formed non-covalent complexes with E. coli trypsin and chymotrypsin. In summary, a novel protocol for identification of novel protease inhibitors with antimicrobial activities was here described, and it is likely to provide direct applications for agricultural, food and pharmaceutical industries.

Analyst


DOI: 10.1039/C3AN01132A

10

15

35. C.A. Ryan, Protease inhibitors in plants-genes for improving defenses against insects and pathogens, Ann Rev Phytopathol,1990,28, 425-49. 36. B. Prakash, S. Selvaraj, M.R. Murthy, Y.N. Sreerama, D.R. Rao, et al., Analysis of the amino acid sequences of plant Bowman–Birk inhibitors, J. Mol. Evol. 1996,42, 560–569. 37. W.F. Porto, V.A. Souza, D.O. Nolasco, O.L. Franco, In silico identification of novel hevein-like peptide precursors, Peptides, 2012,38, 127-136.

20



5

31. C.W. Jung, Surface properties of superparamagnetic iron oxide MR contrast agents: ferumoxides, ferumoxtran, ferumoxsil, Mag Res Imag,1995,13, 675-691. 32. R.Y. Hong, B. Feng, L.L. Chen, G.H. Liu, H.Z. Li, et al., Synthesis, characterization and MRI application of dextran-coated Fe3O4 magnetic nanoparticles, Biochem Eng J, 2008,42, 290–300. 33. P. Cuatrecasas, Protein purification by affinity chromatography, J Biol Chem,1970,245, 3059-3065 34. C.A. Ryan, Proteolytic enzymes and their inhibitors in plants, Annu Rev Plant Physiol,1973,241, 73-96.

