Catalytic and structural effects of flexible loop deletionin organophosphorus hydrolase enzyme: A thermostability

improvement mechanism

GHOLAMREZA FARNOOSH1, KHOSRO KHAJEH2, MOZAFAR MOHAMMADI

1,KAZEM HASSANPOUR

3, ALI MOHAMMAD LATIFI1* and HOSSEIN AGHAMOLLAEI

4

1Applied Biotechnology Research Centre, Baqiyatallah University of Medical Sciences, Tehran, Iran2Department of Biochemistry, Faculty of Biologic Science, Tarbiat Modares University, Tehran, Iran

3Medical School, Sabzevar University of Medical Sciences, Sabzevar, Iran4Chemical Injuries Research Center, Systems Biology and Poisonings Institute, Baqiyatallah

University of Medical Sciences, Tehran, Iran

*Corresponding author (Email, amlatifi290@gmail.com)

MS received 25 January 2019; accepted 11 March 2020

Thermostability improvement of enzymes used industrially or commercially would develop their capacity andcommercial potential due to increased enzymatic competence and cost-effectiveness. Several stabilizing factorshave been suggested to be the base of thermal stability, like proline replacements, disulfide bonds, surface looptruncation and ionic pair networks creation. This research evaluated the mechanism of increasing the rigidity oforganophosphorus hydrolase enzyme by flexible loop truncation. Bioinformatics analysis revealed that themutated protein retains its stability after loop truncation (five amino acids deleted). The thermostability of thewild-type (OPH-wt) and mutated (OPH-D5) enzymes were investigated by half-life, DGi, and fluorescence andfar-UV CD analysis. Results demonstrated an increase half-life and DGi in OPH-D5 compared to OPH-wt.These results were confirmed by extrinsic fluorescence and circular dichroism (CD) spectrometry experiments,therefore, as rigidity increased in OPHD5 after loop truncation, half-life and DGi also increased. Based onthese findings, a strong case is presented for thermostability improvement of OPH enzyme by flexible looptruncation after bioinformatics analysis.

Keywords. Bioinformatics design; flexibility; loop deletion; organophosphorus compounds; organophos-phorus hydrolase; thermostability improvement

1. Introduction

Organophosphorus hydrolase (OPH) is a homodimericmetalloenzyme whose monomer consists of an eight-stranded parallel b-pleated sheet and two divalent metalions at the active site. This enzyme has been used in thedegradation of organophosphorus compounds, as oneof the most commonly used decontamination proce-dures (Farnoosh and Latifi 2014). The OPH has a goodpotential for industrialization in varied applications(Paliwal 2008). Although the overall conformationalstability of OPH is large (* 40 kcal/mol for the Zn2?

form), the energy required to inactivate the enzyme is

very small (* 4 kcal/mol). The structural changesdestabilizing the enzyme must be better recognized bydetermining unfolding profile and employing rationaldesign (Armstrong 2007).Increased thermostability in enzymes used industri-

ally or commercially would develop their capacity andcommercial potential, due to improved enzymaticcompetence and cost-effectiveness. Until now, there arenumerous procedures to improve enzyme thermosta-bility, by decreasing the entropy of the unfolded state.In fact, modifications improving the entropy of thefolded state can also lead to stabilization (Dagan et al.2013). Effective factors on the thermostability of an

http://www.ias.ac.in/jbiosci

J Biosci (2020) 45:54 � Indian Academy of SciencesDOI: 10.1007/s12038-020-00026-5 (0123456789().,-volV)(0123456789().,-volV)

enzyme have been documented by comparing ther-mozyme and mesozyme structures and sequences(Dagan et al. 2013).Elongation of loop length leads to a decrease in

folding and gives rise to an increase in unfoldingstates of the enzymes. Past studies on flexible loopareas revealed that the entropic effect of ordering aloop upon folding is mainly associated with theenergetic outcomes from modifying loop length(Dagan et al. 2013).Both theoretical and experimental methods have

been documented that the deletion of exposed loopregions can improve stability. Loop deletion has beenan important factor observed in some studies compar-ing crystal structures from thermophilic and mesophilicsources. By comparison of structures from mesophilicor thermophilic, scientists have found an inverse cor-relation between loop length and stability, an increasein stability by loop shortening (Thompson and Eisen-berg 1999).Studies on loop deletions to explain their effect on

the thermostability indicate that removing some por-tions of the loops usually increases the thermostabilityof the mutants (Yipp et al. 2012). This is documentedwith previous bioinformatics analyses and structuralstudies in which shorter loop areas are a communalstrategy for the enzymes to increase the stability (Yippet al. 2012). Conformational changes of enzyme can bestudied by using molecular dynamics simulations(MDS). The root mean square fluctuation (RMSF) androot mean square deviation (RMSD), calculation valuesfor backbone atoms, are useful to determine confor-mationally flexible and thermally sensitive regions(Tian et al. 2010).In previous study, we improved OPH thermostability

by designing and introducing disulfide bridges, as twonew mutated OPH enzymes were more ther-mostable than wild-type (Farnoosh et al. 2016). Basedon the above, this study is aimed to evaluate the ther-mostability of OPH after the truncation of one of theexposed loops using bioinformatics tools.

2. Materials and methods

2.1 Molecular dynamic simulation

The coordinates of OPH-wild-type (OPH-wt) wasretrieved from the protein data bank (PDB code:1HZY) (Benning et al. 2001). Molecular dynamicsimulation was performed on each PDB structure (wildtype and mutated) and were embedded in a box with

dimensions of 8 nm. Subsequently, the system wassolvated with spc216 water model. Energy minimiza-tion was carried out in order to equilibrate the wholesystem. During the minimization step, steepest descentintegrator with a maximum of 50,000 steps and 1000, 0(kJ/mol/nm) were taken. In the two short runs of MD,NVT and NPT, ensembles with restrains on proteinbackbone were taken. The main run of MD was finallyperformed after removing all restrains from the system.In the case of MD, Leap-frog integrator with LINCSconstraint algorithm and particle mesh Ewald Coulombtype were used. The more accurate Nose-Hooverthermostat was used together with Parrinello-Rahmanpressure coupling method. Molecular dynamics simu-lations were performed using the program packageGROMACS 5.1.4 (Abraham et al. 2015) along withOPLS-AA force field (Jorgensen et al. 1996). Calcu-lation of RMSD, RMSF and Rg (radius of gyration)plots were done through 5 ns.

2.2 Construction of the OPH-wt and mutant

Research results have demonstrated that the removal of29 amino acids leader sequence leads to increase inOPH thermostability (Pinkerton 2005). For that reason,the leader sequence was removed from opd(AY766084.1) N-terminal domain. The OPH-wt(1008 bp) and OPH-D5 gene (993 bp) were designedand optimized for E.coli Rosseta Gami by JCat (Groteet al. 2005), Gen Script’s (http://www.genscript.com/cgi-bin/tools/rare_codon_analysis) and Mfold (Zuker2003) softwares, and were chemically synthesized andwere finally cloned into NdeI/HindIII restriction sitesof pET21a vector (Biomatik, Canada) with His tagfused at N-terminal. All the constructs were confirmedby DNA sequencing.

2.3 Expression and purification

The Escherichia coli Rosseta Gami cells were trans-formed by recombinant plasmids using Heat ShockMethod (Froger and Hall 2007). The transformantswere cultured and expressed according to Farnooshet al. method (Farnoosh et al. 2016).In spite of using different conditions of temperature,

time and IPTG concentration, the majority of enzymewas in the insoluble fractions, hence, denaturationrefolding was done. Purification and refolding stepswere done according to Farnoosh et al. method (Far-noosh et al. 2016). Evaluation of protein concentration

54 Page 2 of 10 Gholamreza Farnoosh et al.

was performed using Bradford method (Tian et al.2010, 2013; Yang et al. 2003).

2.4 Activity assay

OPH-wt and OPH-D5 activities were evaluated bymonitoring the production of the p-nitrophenol (PNP),from paraoxon as the specific substrate (SIGMA-ALDRICH). A 40 ll of purified OPH was added into140 ll of the reaction solution (50 mM Tris-HCl buffercontaining 10 mM ZnCl2, pH 8.5). The reaction wasinitialized by adding 20 ll of 20 mM paraoxon (in20% deionized water). After the samples were incu-bated for 10 min at 37 �C, 100 ll of the each reactionwas added in a 96-deep-well plate and the producedPNP was calculated by determining the absorbance at405 nm where PNP absorbs at ambient temperatureusing microplate reader device. A unit of enzymeactivity was described as one micromole of paraoxondegraded to PNP per milliliter, per minute at 37 �C(Briseno-Roa et al. 2011) (Tian et al. 2010).Enzyme activity was calculated by the following

equation:Enzyme activity = Absorbance at OD405 nm 9 Re-

action volume 9 Dilution coefficient/extinction coef-ficient 9 Enzyme volume 9 Time (Abdel-Razek et al.2013).Extinction coefficient for p-Nitrophenol is

17,000 M-1 cm-1.

2.5 Kinetic parameters determination

Calculation of Michaelis Menten kinetic parameterswas done according to the above enzyme assayoptimized with substrate varying in concentrationfrom 0 to 700 lM paraoxon. The final concentrationof enzyme in each assay was kept constant at 200 lg.Negative control reactions were performed withoutany enzyme. Reactions were performed in triplicate.The Vmax, Km and Kcat values were determined bynonlinear regression using Graphpad prism 6.0software(GraphPad Software Inc., La Jolla, CA,USA) (Briseno-Roa et al. 2011; Ely et al. 2010; Leet al. 2012; Tian et al. 2010).

2.6 Structural analysis

To evaluate the structural stability of mutant enzymesafter loop deletion and to compare with OPH-wt,

Fluorescence measurement and circular dichroismspectrometry experiments were proposed.

2.6.1 Fluorescence measurements: Fluorescence mea-surement studies were performed on a Perkin ElmerFluorescence Spectrometer at 25 �C. The enzymeswere incubated at 65 �C for 10, 20, 30, 40, 50 and60 min and were carried out rapidly on FluorescenceSpectrometer. Fluorescence evaluation was performedusing 100 lg/ml of each enzyme at an excitationwavelength of 295 nm. Emission spectra were calcu-lated between wavelengths of 300 and 400 nm. Themechanism of thermal denaturation can occur throughthe changes in the fluorescence intensity at 350 nm(Amini-Bayat et al. 2012; Haghani et al. 2008).

2.6.2 Circular dichroism spectrometry: Recording ofFar-CD spectra of the OPH-wt and mutant were done ona CD spectrophotometer (Jasco J 715, Japan) with aprotein concentration of 300 lg/ml in in 50 mM Trisbuffer, pH 8. The CD spectra were recorded by scanningthe samples from 195 to 260 nm at ambient temperatureusing spectropolarimeter (Model J-810, Jasco, Japan).bandwidth of 1 nm, slit width of 0.02 mm and speed of50 nm s-1 were constant for each scan (Kachooei et al.2014). Extracted data were documented as molar ellip-ticity (deg cm2 dmol-1), according to a mean amino acidresidue weight (MRW) of 37000 g/mol for OPH. Cal-culation of the molar ellipticity was done as k = (h 9

100MRW)/(cl), where c is the protein concentration inmg/ml, l is the light path length in centimeters, and h isthe determined ellipticity in degrees at a wavelength k.Determination of secondary structure was done usingSSE-338 software, which can deconvolute far-UV CDspectra (Haghani et al. 2008).

2.7 Stability measurements

At this stage for thermostability evaluation, OPH-wtand mutant were incubated at 65�C for 0, 10, 20, 30,40, 50 and 60 min, then the reactions were placedrapidly on ice for 3 min and then the thermostabilitywas evaluated by using the above mentioned assayprotocol. The remaining activity was calculated as aratio of the primary activity (Amini-Bayat et al. 2012;Chu et al. 2010; Tian et al. 2010). Calculation of kivalues (rate constant of thermal inactivation) of theOPH-wt and OPH-D5 enzymes were done by the slopeof a ln (Ei/E0) = - ki t linear plot [Ei= initial activity ofinactivated form of enzyme, E0= initial activity of theenzyme before thermal stage, t = time (min)]. The t1/2

Flexible loop deletion in OPH enzyme Page 3 of 10 54

(half-life) of each enzyme is the time necessary for50% inactivation of its activity and was determinedfrom the ln (Ei/E0) = - ki t linear plot by introducingE0 = 2Ei (Kazan et al. 1997).The DG inactivation (DGi) values of the OPH-wt and

OPH-D5 at 65�C were documented by the thermody-namic equation as follows (Pawlowski and Zie-lenkiewicz 2013):DGi = - RT ln(kih/kBT), where ki is inactivation

rate constant (min), kB the Boltzmann constant(1.3805 9 10-23 J K-1), h the Planck’s constant(6.6256 9 10-34 J s) and R the gas constant(8.314 J mol-1 K-1).

3. Results

3.1 Bioinformatics analysis of the OPHand design of the deletion mutant

The atom-positional RMSFs for Ca-atoms were com-puted through 5 ns for OPH-WT protein (figure 1) to findthe most flexible region regarding OPH-WT enzyme thatmay cause instability of the molecule. As seen in figure 1,the largest fluctuations are observed in the loop regionincluding aa237-241, indicating that the latter region isrelatively unstable. Therefore, the 5 amino acids (OPH-D5) related to the loop region were deleted. MD simu-lation was also performed on the mutated enzyme.The stability of the OPH enzyme variants were

analyzed according to two factors, RMSD and Rg.Divergences in the RMSDs of the Ca atoms relative tothe starting structures through the MD simulations areshown in figure 2. A minimum deviation in the averageRMSD values of mutated enzyme, in comparison withwild-type enzyme, from the 1.5-th ns to the end ofsimulation was an indication of the conformationalstability of OPH-D5 against OPH-WT. The radius ofgyration (Rg) of a protein is a measure of its packing.The conformational compactness of protein residues isan important aspect of protein stability (Lobanov et al.2008). For an unstable protein, Rg will drift over time.The reasonably invariant Rg values are seen for OPH-D5 rather than OPH-WT (figure 3), indicating that themutated protein, in its compact form over the period of5 ns at 310 K, remains very stable.

3.2 Protein expression and purification

SDS-PAGE investigation of OPH-wt and OPH-D5fractions confirmed distinct bands of the anticipated

monomeric molecular weight (*35 kDa). Accord-ingly, majority of the enzymes were observed ininsoluble fractions (figure 4), therefore, denaturationrefolding protocol was employed to solubilize theenzyme. The purification results showed single bandson SDS-PAGE with a molecular mass approximate to35 kDa for both OPH-wt and OPH-D5 (figure 5).

3.3 Kinetic analysis

Both enzymes of OPH-wt and OPH-D5 were capableof degrading paraoxon to PNP. To compare the kineticproperties after loop deletion, Km, Vmax and kcat werecalculated as mentioned in the materials and methodssection. Vmax and kcat calculated in OPH-d5 showed aslight decrease compared to OPH-wt, whereas Kmincreased (table 1).The loop truncation in OPH-D5 creates new inter-

actions which may cause changes in enzyme catalytic

Figure 1. Atom-positional RMSFs of the Ca-atoms ofOPH-WT.

54 Page 4 of 10 Gholamreza Farnoosh et al.

activity and affinity (D’Amico et al. 2003). The ther-mostability improvement in protein engineering hasoften been documented to be accompanied withdecrease in their catalytic activity and affinity, probablyas a result of the overall protein inflexibility (Dansonet al. 1996).

3.4 Structural studies

To evaluate the structural changes of OPH-D5 com-pared to the OPH-wt after loop truncation, intrinsicfluorescence and circular dichroism (CD) spectrometryexperiments were applied.The fluorescence measurement was used as a pre-

ferred method to investigate folding modifications ofthe enzymes in solutions. Intrinsic fluorescence

emissions of a folded protein are predominantly relatedto the excitation of tryptophan residues and, to a lesserextent, due to phenylalanine and tyrosine residues(Ghisaidoobe and Chung 2014). Two Trp residues(Trp69 and Trp131) are in both OPH-wt and OPH-D5that present the conformational changes in the tertiarystructure. The results showed a significant increase influorescence intensity of OPH-D5 in comparison withOPH-wt (figure 6), showing that unfolding occursmore slowly in OPH-D5 than the OPH-wt.Circular dichroism (CD) is a valuable technique to

investigate the secondary structure of proteins aftermutation (van Mierlo and Steensma 2000). The far-UVCD spectra of OPH-wt and OPH-D5 show changes inthe secondary structure (figure 7). In contrast to theOPH-wt, a slight increase in the molar ellipticity ofOPH-D5 was observed.

Figure 2. RMSDs of the backbone atoms (N, Ca, and C) of OPH-WT (red) and OPH-D5 (green).

Figure 3. The radius of gyration plot of OPH-WT (red) and OPH-D5 (green) proteins.

Flexible loop deletion in OPH enzyme Page 5 of 10 54

These results confirmed the previous research inthermostability studies, demonstrating that an increasein thermostability decreases structural flexibility.

Obtained data were also in line with previous studieson the conformational changes, indicating a decrease inthe protein unfolding process after loop truncation(Dagan et al. 2013; Damnjanovic et al. 2014; Li et al.2012; Sakurai et al. 1996a).

Figure 4. Native SDS-PAGE analysis of OPH-wt andOPH-D5. M: molecular weight marker (#SM0761); Lane1: insoluble fraction of OPH-wt, Lane 2: insoluble fractionof OPH-D5, Lane 3: soluble fraction of OPH-wt and Lane 4:soluble fraction of OPH-D5.

Figure 5. SDS-PAGE analysis of purified OPH-wt andOPH-D5. M: molecular weight Marker (#SM0761); Lane 1:negative control, Lane 2: OPH-wt, Lane 3: OPH-D5.

Table 1. Kinetic parameters of OPH-wt and OPH-D5

Protein Vmax (lmol/min) Km (mM) kcat(s-1)

OPH-wt 3068 0.175 176.32OPH-D5 2980 0.261 171.26

Figure 6. Fluorescence emission spectra of OPH-wt andOPH-D5. (A) Spectra were taken at 25 �C in Tris buffer(50 mM, pH 8), the excitation wavelength was 280 nm.(B) Relative fluorescence after incubation at 65 �C for 0, 10,20, 30, 40, 50, 60 min, the excitation wavelength was280 nm and emission wavelength was 340 nm. The samplescontained 80 lM proteins in Tris buffer (50 mM, pH 8). 1:OPH-D5, 2: OPH-wt.

Figure 7. Far-UV CD spectra of OPH-wt and OPH-D5.The concentration of proteins was 0.3 mg/ml. The sampleswere equilibrated in 50 mM Tris buffer, pH 8 at 25 �C. 1:OPH-D5, 2: OPH-wt.

54 Page 6 of 10 Gholamreza Farnoosh et al.

3.5 Thermal tolerance assay

By calculating the percentage of the remaining activityof OPH-wt and OPH-D5 after the heat treatment at65 �C for 60 min during the time interval of 10 min,the result of temperature on the stability was evaluated.The residual activity of OPH-wt after 60 min at 65 �Cwas 31% whereas for OPH-D5 it was 46%. Accordingto the results shown in figure 8, OPH-D5 shows morethermostability than the OPH-wt. The calculation ofthermal parameters, including half-life (t1/2), rate ofthermal inactivation (ki) and DGi, and comparing themcan be beneficial to have more information about thethermostability after loop deletion.The evaluation results showed an increase in both

DGi and half-life in OPH-D5 compared to OPH-wt thatclearly indicates the increasing effect of loop deletion(table 2). The higher t1/2 and DGi values at OPH-D5show that this mutation causes an increase in thethermostability, consequently, the conformation ofenzyme would not simply be unfolded after incubationat 65 �C.The increase in DGi and half-life parameters, as

important factors in stability analysis, confirmed thatthe loop truncation reduces the susceptibility ofenzyme to thermal inactivation. The acquired resultsare in line with those of previous studies indicating thatthis method leads to thermal tolerance in enzyme

(Dagan et al. 2013; Damnjanovic et al. 2014; Imaniet al. 2010; Le et al. 2012).

4. Discussion

The energy needed to inactivate the OPH is muchlower compared to its overall conformational stability,and therefore, it would be necessary to understand howthe OPH unfolded into its inactive form to reconstruct astable folding (Armstrong 2007).Previous research studies have suggested that the

stability of thermophilic proteins arises from theirrigidity compared to their mesophilic homologs. It isdocumented that deleted sites in thermophilic proteinsare related to the exposed loops, and therefore, deletingthe exposed loops is regarded as a regular mechanismto improve thermostability (Boone et al. 2015b; Kumarand Nussinov 2001; Thompson and Eisenberg 1999).Previous studies on engineered enzymes revealed

that the truncation of flexible loops can protect theenzyme against unfolding at high temperatures. Inthese studies, the mentioned mechanism was used toimprove the thermostability in several enzymes suchas: phosphatidylinositol-synthesizing streptomycesphospholipase D, serine protease subtilisin and Photi-nus Pyralis firefly luciferase (Almog et al. 2002;Asgeirsson et al. 2007; Damnjanovic et al. 2014;Halliwell 2015).Human carbonic anhydrase II has a long loop region

between residues 230 and 240 that was a candidate fortruncation, in which the findings on loop deletionshowed that the modification could have positive effecton the thermal stability (Boone et al. 2015b).In previous study, the improvement of OPH ther-

mostability was done by designing and introducingdisulfide bonds as two new mutated OPH enzymes(Farnoosh et al. 2016). Although loop truncation hasbeen found as an important approach for protein ther-mostability, this study evaluates this mechanism inOPH enzyme.The structural analysis of OPH showed the presence

of some flexible areas with high temperature factors.Bioinformatics data (figure 1) showed a higher tem-perature factor in a loop region from residues 237 to241 located between the a-helix and beta sheet (fig-ure 9) compared to other areas, indicating that thisregion might be very flexible and might play a key rolein thermal stability.According to these data, D5 loop is a flexible and

hotspot region in OPH that is located at the mostexposed portion of loop.

Figure 8. Residual activity (%) graphs of OPH-wt andOPH-D5 after incubation at 65 �C for 0, 10, 20, 30, 40, 50,60 min. 1: OPH-D5, 2: OPH-wt.

Table 2. The values of half-life (t1/2) and denaturationenergy change (DGi) of OPH-wt and OPH-D5 at 65 �C

DGi (kcal/mol) t1/2 (min-1)

OPH-wt 94 40OPH-D5 97 65

Flexible loop deletion in OPH enzyme Page 7 of 10 54

Therefore, this region might be susceptible tounstable agents, such as heat. The results imply that thedeletion of 5 residues in this loop has a critical con-sequence in its stability. In the present study, the role ofthe flexible loop in OPH thermostability is described.Through loop truncation, the mutant enzyme

reserved its catalytic activity. The investigation of thekinetic parameters of OPH-wt and OPH-D5 showed aminimal decrease in Kcat and an increase in Km forOPH-D5 compared to OPH-wt (table 1). These datashowed that the conformation of OPH-D5 might bechanged after loop truncation which is accompaniedwith decrease in catalytic activity and affinity(D’Amico et al. 2003; Han et al. 2009; Kim et al.2015; Ugarova and Koksharov 2012). The increase inKm of OPH-D5 compared to OPH-wt is possiblycaused by an alteration in the competence of universalgeometry for binding substrates and reduced flexibilityin the binding site. Protein stabilization by engineeringmethods has often been accompanied with a decreasein their activity and affinity (D’Amico et al. 2003;Danson et al. 1996; Yin et al. 2015).As shown in figure 9, Loop D5 is located near ‘large

packet’ of binding site (Met317, His267, His254 andLeu271) that affects the geometry of this pocket andresults in decreased enzyme affinity to specific sub-strate. The study performed on the deletion of flexibleNC-loop in lipase (FClip1) confirms our observationsin the present study, where it was described that theNC-loop is positioned adjacent to the substrate-bindingpocket and may play roles in enzyme-substrate contact,

therefore the deletion of the region can affect enzy-matic activity and substrate affinity (Li et al. 2012).Based on thermostability results, DGi and t1/2 param-

eters were improved in OPH-D5 compared to OPH-wtthat were correlated to structural studies by extrinsic flu-orescence and circular dichroism (CD) spectrometryexperiments. These data prove that the loop truncationresults in the reduction of flexibility in OPH-D5. Struc-tural analysis and the investigation of dynamic andenergetic states of the mutant enzyme show that the loopdeletion results in changes in the construction, stabilityand activity (Boone et al. 2015a; Farnoosh et al. 2016; Liet al. 2012; Sakurai et al. 1996b).The loop D5, which was selected for the truncation,

links a-helix and b-strand of the enzyme (figure 10). Itis predicted that the deletion of five residues at LoopD5 decreases the flexibility of loop and stabilizes the b-strand that may be involved in the enzyme unfolding inhigh temperatures.Our findings are also consistent with a study conducted

to compare the sequence and the structure of OPH fromPseudomonas diminuta with homolog thermophileenzymes including OPH identified from the bacteriumGeobacillus stearothermophilus (GsP), Deinococcusradiodurans (Dr-OPH) and S. solfataricus (SsoPox), inwhich D5 loop is not observed (Hawwa et al. 2009).Briefly, rational method design by flexible loop trunca-

tion showed a valuable approach to improve thermosta-bility with no obvious negative effects on the function andthe folding of the mutant enzyme. This improvement alsoshowed the accuracy in selecting the residues to delete.Consequently, the results not only assist us for betteridentification of the flexible regions in OPH by bioinfor-matics method, but also promote our knowledge in engi-neering enzymes for industrial applications.

Figure 9. The three-dimensional structure of Loop D5OPH, which describes the location of loop truncation andbinding site (large packet). The yellow region is Loop D5.The conserved His254, His257, Leu271 and Met317residues are binding sites (PDB code: 1HZY).

Figure 10. The three-dimensional structure of Loop D5 inOPH, describes the location of loop truncation between a-helix and the b-strand. The green region is Loop D5 (PDBcode: 1HZY).

54 Page 8 of 10 Gholamreza Farnoosh et al.

Acknowledgements

The authors would like to thank the manager and col-leagues of Applied Biotechnology Research Centre,Baqiyatallah University of Medical Sciences, Tehran,Iran, for their useful discussions and critical reading ofthe manuscript. It should be noted that this paper wasextracted from dissertation of the research project ofApplied Biotechnology Research Center, BaqiyatallahUniversity of Medical Sciences, Tehran, Iran.

References

Abdel-Razek MA-RS, Folch-Mallol JL, Perezgasga-Cisco-mani L, Sanchez-Salinas E, Castrejon-Godınez ML andOrtiz-Hernandez ML 2013 Optimization of methylparathion biodegradation and detoxification by cells insuspension or immobilized on tezontle expressing the opdgene. J. Environ. Sci. Health 48 449–461

Abraham MJ, Murtola T, Schulz R, Pall S, Smith JC, Hess Band Lindahl E 2015 GROMACS: High performancemolecular simulations through multi-level parallelismfrom laptops to supercomputers. SoftwareX 1–2 19–25

Almog O, Gallagher DT, Ladner JE, Strausberg S, Alexan-der P, Bryan P and Gilliland GL 2002 Structural basis ofthermostability analysis of stabilizing mutations in sub-tilisin BPN. J. Biol. Chem. 277 27553–27558

Amini-Bayat Z, Hosseinkhani S, Jafari R and Khajeh K 2012Relationship between stability and flexibility in the mostflexible region of Photinus pyralis luciferase. Biochim.Biophys. Acta Proteins Proteomics 1824 350–358

Armstrong CD 2007 Elucidating the chemical and thermalunfolding profiles of organophosphorus hydrolase andincreasing its operational stability (Texas A&M University)

Asgeirsson B, Adalbjornsson BV and Gylfason GA 2007Engineered disulfide bonds increase active-site localstability and reduce catalytic activity of a cold-adaptedalkaline phosphatase. Biochim. Biophys. Acta ProteinsProteomics 1774 679–687

Benning MM, Shim H, Raushel FM and Holden HM 2001High resolution X-ray structures of different metal-substituted forms of phosphotriesterase from Pseu-domonas diminuta. Biochemistry 40 2712–2722

Boone CD, Rasi V, Tu C and McKenna R 2015a Structuraland catalytic effects of proline substitution and surfaceloop deletion in the extended active site of humancarbonic anhydrase II. FEBS J. 282 1445–1457

Briseno-Roa L, Oliynyk Z, Timperley CM, Griffiths AD andFersht AR 2011 Highest paraoxonase turnover rate foundin a bacterial phosphotriesterase variant. Protein Eng.Design Selection 24 209–211

Chu X-y, Tian J, Wu N-f and Fan Y-l 2010 An intramolec-ular disulfide bond is required for the thermostability of

methyl parathion hydrolase, OPHC2. Appl. Microbiol.Biotechnol. 88 125–131

D’Amico S, Gerday C and Feller G 2003 Temperatureadaptation of proteins: engineering mesophilic-like activ-ity and stability in a cold-adapted alpha-amylase. J. Mol.Biol. 332 981–988

Dagan S, Hagai T, Gavrilov Y, Kapon R, Levy Y, Reich Z2013 Stabilization of a protein conferred by an increase infolded state entropy. Proc. Nat. Acad. Sci. 110 10628–10633

Damnjanovic J, Nakano H and Iwasaki Y 2014 Deletion of adynamic surface loop improves stability and changeskinetic behavior of phosphatidylinositol-synthesizingStreptomyces phospholipase D. Biotechnol. Bioeng. 111674–682

Danson MJ, Hough DW, Russell RJ, Taylor GL and Pearl L1996 Enzyme thermostability and thermoactivity. ProteinEng. 9 629–630

Ely F, Hadler K, Gahan L, Guddat L, Ollis D and Schenk G2010 The organophosphate-degrading enzyme fromAgrobacterium radiobacter displays mechanistic flexibil-ity for catalysis. Biochem. J. 432 565–573

Farnoosh G and Latifi AM 2014 A review on engineering oforganophosphorus hydrolase (OPH) enzyme. J. Appl.Biotechnol. Rep. 1 1–10

Farnoosh G, Khajeh K, Latifi AM and Aghamollaei H 2016Engineering and introduction of de novo disulphidebridges in organophosphorus hydrolase enzyme forthermostability improvement. J. Biosci. 41 577–588

Froger A and Hall JE 2007 Transformation of plasmid DNAinto E. coli using the heat shock method. JoVE 6 253

Ghisaidoobe AB and Chung SJ 2014 Intrinsic tryptophanfluorescence in the detection and analysis of proteins: afocus on Forster resonance energy transfer techniques. Int.J. Mol. Sci. 15 22518–22538

Grote A, Hiller K, Scheer M, Munch R, Nortemann B,Hempel DC and Jahn D 2005 JCat: a novel tool to adaptcodon usage of a target gene to its potential expressionhost. Nucleic Acids Res. 33 W526–W531

Haghani K, Salmanian AH, Ranjbar B, Zakikhan-Alang Kand Khajeh K 2008 Comparative studies of wild typeEscherichia coli 5-enolpyruvylshikimate 3-phosphatesynthase with three glyphosate-insensitive mutated forms:Activity, stability and structural characterization. Biochim.Biophys. Acta Proteins Proteomics 1784 1167–1175

Halliwell LM 2015 Protein engineering utilising singleamino acid deletions within Photinus pyralis fireflyluciferase (Cardiff University)

Han Z-l, Han S-Y, Zheng S-P and Lin Y 2009 Enhancingthermostability of a Rhizomucor miehei lipase by engi-neering a disulfide bond and displaying on the yeast cellsurface. Appl. Microbiol. Biotechnol. 85 117–126

Hawwa R, Aikens J, Turner RJ, Santarsiero BD and MesecarAD 2009 Structural basis for thermostability revealedthrough the identification and characterization of a highlythermostable phosphotriesterase-like lactonase from

Flexible loop deletion in OPH enzyme Page 9 of 10 54

Geobacillus stearothermophilus. Arch. Biochem. Biophys.488 109–120

Imani M, Hosseinkhani S, Ahmadian S and Nazari M 2010Design and introduction of a disulfide bridge in fireflyluciferase: increase of thermostability and decrease of pHsensitivity. Photochem. Photobiol. Sci. 9 1167–1177

Jorgensen WL, Maxwell DS and Tirado-Rives J 1996Development and testing of the OPLS all-atom force fieldon conformational energetics and properties of organicliquids. J. Am. Chem. Soc. 118 11225–11236

Kachooei E, Moosavi-Movahedi AA, Khodagholi F, Moza-farian F, Sedeghi P, Hadi-Alijanvand H, Ghasemi A,Saboury AA, Farhadi M and Sheibani N 2014 Inhibitionstudy on insulin fibrillation and cytotoxicity by paclitaxel.J. Biochem. 155 361–373

Kazan D, Ertan H and Erarslan A 1997 Stabilization ofEscherichia coli penicillin G acylase against thermalinactivation by cross-linking with dextran dialdehydepolymers. Appl. Microbiol. Biotechnol. 48 191–197

Kim J, Kim S, Yoon S, Hong E and Ryu Y 2015 Improvedenantioselectivity of thermostable esterase from Archae-oglobus fulgidus toward (S)-ketoprofen ethyl ester bydirected evolution and characterization of mutantesterases. Appl. Microbiol. Biotechnol. 99 6293–6301

Kumar S and Nussinov R 2001 How do thermophilicproteins deal with heat? Cell. Mol. Life Sci. 58 1216–1233

Le QAT, Joo JC, Yoo YJ and Kim YH 2012 Development ofthermostable Candida antarctica lipase B through novelin silico design of disulfide bridge. Biotechnol. Bioeng.109 867–876

Li B, Yang G, Wu L and Feng Y 2012 Role of the NC-loopin catalytic activity and stability in lipase from Fervi-dobacterium changbaicum. PLoS One 7 e46881

Lobanov MY, Bogatyreva NS and Galzitskaya OV 2008Radius of gyration as an indicator of protein structurecompactness. Mol. Biol. 42 623–628

Paliwal S 2008 Development of enzyme-based biosensors forthe detection of organophosphate neurotoxins (ProQuest)

Pawlowski PH and Zielenkiewicz P 2013 Theoretical modelexplaining the relationship between the molecular massand the activation energy of the enzyme revealed by alarge-scale analysis of bioinformatics data. Acta Biochim.Pol. 60 239–247

Pinkerton TS 2005 The recombinant expression and poten-tial applications of bacterial organophosphate hydrolasein Zea mays L (Texas A&M University)

Sakurai M, Ohzeki M, Miyazaki K, Moriyama H, Sato M,Tanaka N and Oshima T 1996a Structure of a loop-deletedvariant of 3-isopropylmalate dehydrogenase from Ther-mus thermophilus: an internal reprieve tolerance mecha-nism. Acta Crystallogr. Biol. Crystallogr. 52 124–128

Thompson MJ and Eisenberg D 1999 Transproteomicevidence of a loop-deletion mechanism for enhancingprotein thermostability. J. Mol. Biol. 290 595–604

Tian J, Wang P, Gao S, Chu X, Wu N and Fan Y 2010Enhanced thermostability of methyl parathion hydrolasefrom Ochrobactrum sp. M231 by rational engineering of aglycine to proline mutation. FEBS J. 277 4901–4908

Tian J, Wang P, Huang L, Chu X, Wu N and Fan Y 2013Improving the thermostability of methyl parathion hydro-lase from Ochrobactrum sp. M231 using a computation-ally aided method. Appl. Microbiol. Biotechnol. 972997–3006

Ugarova N and Koksharov M 2012 Thermostabilization offirefly luciferases using genetic engineering (INTECHOpen Access Publisher)

van Mierlo CP and Steensma E 2000 Protein folding andstability investigated by fluorescence, circular dichro-ism (CD), and nuclear magnetic resonance (NMR)spectroscopy: the flavodoxin story. J. Biotechnol. 79281–298

Yang H, Carr P, McLoughlin SY, Liu J, Horne I, Qiu X,Jeffries C, Russell R, Oakeshott J and Ollis D 2003Evolution of an organophosphate-degrading enzyme: acomparison of natural and directed evolution. ProteinEng. 16 135–145

Yin X, Hu D, Li J-F, He Y, Zhu T-D and Wu M-C 2015Contribution of disulfide bridges to the thermostability ofa type A feruloyl esterase from Aspergillus usamii. PLoSOne https://doi.org/10.1371/journal.pone.0126864

Yipp BG, Petri B, Salina D, et al. 2012 Infection-inducedNETosis is a dynamic process involving neutrophilmultitasking in vivo. Nat. Med. 18 1386–1393

Zuker M 2003 Mfold web server for nucleic acid folding andhybridization prediction. Nucleic Acids Res. 313406–3415

Corresponding editor: BJ RAO

54 Page 10 of 10 Gholamreza Farnoosh et al.