Hydroimidazolone Modification of the Conserved Arg12 in Small Heat Shock Proteins: Studies on the Structure and Chaperone Function Using Mutant Mimics

Methylglyoxal (MGO) is an α-dicarbonyl compound present ubiquitously in the human body. MGO reacts with arginine residues in proteins and forms adducts such as hydroimidazolone and argpyrimidine in vivo. Previously, we showed that MGO-mediated modification of αA-crystallin increased its chaperone function. We identified MGO-modified arginine residues in αA-crystallin and found that replacing such arginine residues with alanine residues mimicked the effects of MGO on the chaperone function. Arginine 12 (R12) is a conserved amino acid residue in Hsp27 as well as αA- and αB-crystallin. When treated with MGO at or near physiological concentrations (2–10 µM), R12 was modified to hydroimidazolone in all three small heat shock proteins. In this study, we determined the effect of arginine substitution with alanine at position 12 (R12A to mimic MGO modification) on the structure and chaperone function of these proteins. Among the three proteins, the R12A mutation improved the chaperone function of only αA-crystallin. This enhancement in the chaperone function was accompanied by subtle changes in the tertiary structure, which increased the thermodynamic stability of αA-crystallin. This mutation induced the exposure of additional client protein binding sites on αA-crystallin. Altogether, our data suggest that MGO-modification of the conserved R12 in αA-crystallin to hydroimidazolone may play an important role in reducing protein aggregation in the lens during aging and cataract formation.


Introduction
Small heat shock proteins are a family of stress proteins. a-Crystallin and Hsp27 are the major small heat shock proteins in humans. These proteins are beneficial in preventing cellular damage for various diseases [1,2,3].
a-Crystallin is a major protein of vertebrate eye lenses, although its presence in other organs such as the brain, heart, kidney, spleen and thymus has also been recognized [4,5]. a-Crystallin consists of two highly homologous subunits, aAand aB-crystallin, and each subunit has a molecular weight of ,20 kDa [4]. In the lens aAcrystallin and aB-crystallin subunits combine in a 3:1 ratio to form an ,40 mer a-crystallin oligomer [6]. The aB-crystallin gene has a heat shock promoter element and is induced by various stress conditions [4,5,7]. aB-Crystallin has been implicated in a number of neurological disorders, such as Alzheimer's disease and Parkinson's disease [1,8]. Both aA-crystallin and aB-crystallin can confer cellular thermo-resistance [9]. Both proteins can act as molecular chaperones, and this chaperoning ability is believed to play a crucial role in maintaining the transparency of the eye lens [10]. As a molecular chaperone, a-crystallin not only prevents the aggregation of unfolded proteins, but it also helps in the refolding of denatured client proteins [11,12]. Because protein turnover is virtually absent in the lens, many post-translational modifications accumulate in lens proteins during aging. Several studies have shown that these post-translational modifications decrease the chaperone function of a-crystallin, which might be one reason for lens aging and age-related cataract formation [13,14,15,16,17].
A large number of advanced glycation end products (AGEs) can be found in the aged human lens [18], which suggests that glycation is a major mechanism for post-translational modification in the aging lens. Glycation is the non-enzymatic reaction that adds carbohydrates, especially glucose, to proteins. First, glucose and other sugars react with the amino groups of proteins to form an unstable Schiff's base that slowly undergoes rearrangement to form a relatively stable Amadori product. Through a series of parallel and sequential reactions (often termed the Maillard reaction), these Amadori products form many AGEs, some of which are fluorescent and colored [19,20].
The lens contains relatively high levels of methylglyoxal (MGO). The reported levels are 1-2 mM [21]. MGO is an a-dicarbonyl compound that reacts with lysine, arginine and histidine residues in proteins [22,23] to form AGEs, such as hydroimidazolone [24], argpyrimidine [25] and methylglyoxal lysine dimer (MOLD) [26,27] (Fig. 1). In addition to our own previous findings, others have reported that the aged and cataractous human lenses contain more of these MGO-derived AGEs than the normal lens [25,27,28,29]. Because MGO reacts rapidly with proteins and the lens proteins have long half-lives, it is reasonable to assume that cumulative modification by MGO over many decades of life could be quantitatively significant in the lens proteins.
In general, it is believed that AGE formation is a cause for lens protein aging and cataract formation. However, we and others have observed that MGO-AGE formation in aA-crystallin makes it a better chaperone [30,31]. AGE formation from MGO occurs predominantly in arginine residues of proteins. Examples of arginine-derived AGEs caused by MGO glycation are argpyrimidine and hydroimidazolone [18]. As a result of these modifications, arginine residues lose their positive charge and become neutral. In a previous study, we demonstrated that the loss of the positive charge was the cause for an increase in the chaperone function of aA-crystallin. In that study, we replaced discrete MGO-modifiable arginine residues with a neutral amino acid, alanine, and showed an improvement in the chaperone function of the mutant proteins [32]. In addition, the chemical conversion of lysine residues to homoarginine residues followed by a reaction with MGO also led to an enhancement in the chaperone function of aA-crystallin [33].
Unlike a-crystallin, Hsp27 is ubiquitously expressed throughout the human body. We have shown that Hsp27 is particularly vulnerable to MGO modification in kidney mesangial cells [34]. Others have shown a similar vulnerability of Hsp27 in other cell types [35,36]. Furthermore, we showed that the chaperone and anti-apoptotic functions of Hsp27 were improved after its modification by MGO [37]. Thus, Hsp27 appears to be a prime target for MGO modification, and consequently, its function could be altered in cells.
Altogether, it is clear now that MGO modification of the small heat shock proteins results in an improvement in their key functions. Whether the improvement in the chaperone function of small heat shock proteins occurs via modification of a conserved arginine residue and whether physiological levels of MGO could improve the chaperone function through a hydroimidazolone modification is not known. In this study, we modified human Hsp27 and aAand aB-crystallin with 2-10 mM MGO and identified hydroimidazolone AGEs using mass spectrometry. Interestingly, the only conserved arginine residue that was modified to hydroimidazolone by MGO was R12 in all three proteins. To determine if the hydroimidazolone modification of this arginine residue is responsible for the improvement of the chaperone function, we replaced R12 with alanine (to mimic the hydroimidazolone modification) and explored the effect of this mutation on the structure and chaperone function of Hsp27 and aAand aB-crystallin.

Results and Discussion
MGO is derived mostly from triose phosphate intermediates of glycolysis by non-enzymatic mechanisms in vivo [38]. It is a major precursor of AGEs in tissue proteins [39]. In previous studies, we have shown that MGO modifications of small heat shock proteins, such as aA-crystallin and Hsp27, enhanced their chaperone function [30,37]. In this study, our primary goal was to determine whether a similar increase in the chaperone function occurred with physiological levels of MGO and to determine whether a modification of the conserved R12 ( Fig. 2A) to hydroimidazolone contributed to the increased chaperone function.
We first determined the ''first hit'' arginine residues for modification to hydroimidazolone. To accomplish this, we modified the proteins with 2, 5 and 10 mM of MGO. With 2 mM MGO, we found that 6, 6 and 8 arginine residues were modified to hydroimidazolone in aAand aB-crystallin and Hsp27, respectively (Table 1). With 10 mM of MGO, this modification reached 10, 8 and 10 arginine residues in the three respective proteins. R12 was the only common residues among the three proteins converted to hydroimidazolone with 2 mM MGO, which suggested that in small heat shock proteins, R12 is the most susceptible for modification to hydroimidazolone by MGO. Notably, a previous study detected a modification of R12 in human lens aA-crystallin that had a molecular weight identical to hydroimidazolone [40].
The modification of arginine residues to hydroimidazolone converts the positive charge on arginine to a neutral charge. Previously, we reported that the substitution of MGO-modifiable arginine residues with neutral alanine residues enhanced the chaperone function of aA-crystallin, similarly to MGO-modification [32]. Because R12 is the most susceptible arginine for MGO modification, we sought to determine if the chaperone function would be improved if it was replaced with alanine. To accomplish this, we cloned and expressed the wild-type (Wt) proteins and the Hsp27 R12A , aA-crystallin R12A (aA R12A ) and aB-crystallin R12A (aB R12A ) mutant proteins in E. coli BL21(DE3). We then purified the proteins by sequential chromatographic methods (gel filtration and ion-exchange chromatography), as previously described [41]. SDS-PAGE analysis showed a single protein band with the correct molecular weight for all proteins (Fig. 2B).
The chaperone function for the small heat shock proteins was evaluated using three different client proteins. The aA R12A mutant showed a 61%, 15% and 10% increase in the chaperone function compared to the Wt protein with CS, c-crystallin and LDH, respectively, as client proteins (Fig. 3). Although, aB R12A showed better protection against thermal aggregation of CS than its Wt variant (,70% better protective ability), it showed a slight reduction in the chaperone function against the other two client proteins tested ( Fig. 3B & C). The R12A mutation had no effect on the chaperone function of Hsp27 ( Fig. 3A-C). Previously, Oya-Ito et al. [37] showed that MGO modification made Hsp27 a better chaperone. The findings in this study that theR12A mutation did not improve the chaperone function of Hsp27 suggest that in addition to hydroimidazolone modification, argpyrimidine modification may be necessary for the improvement of the chaperone function inHsp27.
To understand the molecular basis behind the enhancement in the chaperone function of aA R12A , we determined the structural changes in the protein. Numerous studies have suggested a strong correlation between the chaperone function and the surface hydrophobicity of small heat shock proteins [42,43,44,45,46,47], but others have failed to find such a correlation [48,49]. The surface hydrophobicity of aA R12A and aB R12A (determined by TNS binding) were nearly identical to that of their Wt protein counterparts. Hsp27 R12A had a decreased (,65%) surface hydrophobicity compared to its Wt counterpart (Fig. 4). These results suggested a lack of correlation between surface hydrophobicity and the chaperone function in the three small heat shock proteins, which is similar to previous reports [48,49].
To assess whether the binding sites of these heat shock proteins for client proteins were altered by the R12A mutation, we performed an equilibrium binding study using c-crystallin as the client protein. We incubated the three heat shock proteins (12.5 mM each) for 1 hr at 60uC with various concentrations of c-crystallin (2-18 mM). The unbound (S) and bound c-crystallin were determined by filtration, as described in Methods and Materials section. We determined the dissociation constant (K d ) using the Scatchard equation:ṽ whereṽ v is the number of moles of the substrate bound per mole of chaperone, and n is the number of binding site and K d is the dissociation constant. The stoichiometry of n and K d obtained from the Scatchard plot (Fig. 5) is 3.61 per subunit of aA-crystallin and 8.44 mM, respectively (Table 2). We noted that the number of binding sites (n) per subunit of aA R12A increased from 3.61 in Wt to 4.68, and the association constant increased from 0.118 mM 21 (K d = 8.44 mM) in Wt to 0.124 mM 21 (K d = 8.07 mM) in aA R12A . The n and K d values decreased ,25% and ,13%, respectively, in aB R12A . In contrast, Hsp27 R12A showed no changes in either of these parameters when compared to its Wt counterpart (Table. 2). From these data, we concluded that the substitution of alanine for the conserved arginine residue at position 12 of aA-crystallin increased its affinity for denatured client proteins, whereas the same substitution in aB-crystallin lowered its interaction with the denatured substrate protein. Our results also demonstrated that TNS binding sites are different than the client protein binding sites in all three proteins. We speculate that a structural alteration in aA R12A exposed additional client protein binding sites, and thus, aA R12A bound more client proteins and exhibited better chaperone function than its Wt counterpart. We used tryptophan (W) fluorescence along with near-and far-UV CD techniques to determine if there were any changes in the tertiary and secondary structures of the Wt proteins compared to the mutant proteins. The intrinsic fluorescence spectra indicated some differences between Wt and mutant proteins (Fig. 6). The fluorescence intensity of aA R12A , aB R12A and Hsp27 R12A increased ,27%, 8% and 10%, respectively, compared to the corresponding Wt proteins. Moreover, the l max of the tryptophan fluorescence spectra of the wild-type proteins did not alter due to the mutation. The changes in fluorescence intensity may reflect changes in the microenvironment of W9 (in aAand aB-crystallin) and W16 (in Hsp27), which are located close to the mutation sites. The near-UV CD spectra of these three proteins (both wild type and mutant) agreed with our intrinsic fluorescence data (data not shown). However, these changes in tryptophan fluorescence (perturbation in tertiary structure) did not correlate with the changes in the chaperone function. While some studies showed a direct relationship between an increase in tryptophan fluorescence with improved chaperone function, others did not find such a relationship [33,50,51,52]. Therefore, it is unclear whether changes in the microenvironment of tryptophan are determinants of changes in the chaperone function of a-crystallin.
Quantitative analysis of the far-UV CD data using the CONTINLL program showed that Hsp27 and aAand aB- crystallin are major b-sheet proteins ( Table 3). The data showed no significant perturbation in the secondary structure in the three proteins as a result of the R12A mutation. Based on these data, we concluded that the contribution of R12 for the secondary structure in the three small heat shock proteins was minimal.
Multi-angle light scattering experiments determine the polydispersity and the absolute molar mass of proteins. We used this technique to determine whether the subtle changes in tertiary structure altered the quaternary structure (i.e., the oligomeric assembly) of these three small heat shock proteins. From the data in Table 4, it is evident that a perturbation in the tertiary structure had little effect on the molecular mass of the mutant proteins. The hydrodynamic radius (R h ) was slightly increased only with aA R12A (Table 4). Kundu et al. [53] previously reported that the deletion of the first 20 amino acid residues in aA-crystallin had no effect in its oligomeric size, which is analogous to the findings in this study. The relationship between oligomeric size and chaperone function of small heat shock proteins is still unclear. Some studies have    shown that higher oligomeric assembly diminishes the chaperone function of these proteins [52,54], whereas others have demonstrated contrary results [55,56]. Our results failed to find a correlation between oligomeric size and chaperone function of these three small heat shock proteins.
Several studies have revealed that other factors, such as oligomerization and structural perturbation, may also be required for the proper execution of the chaperone function of a-crystallin [15,52,53,54,57,58]. To quantify the perturbation in the structural stability caused by the R12A mutation, we compared the thermodynamic stability of the Wt and mutant Hsp27 and aAand aB-crystallin proteins. We measured the equilibrium unfolding caused by urea by following tryptophan fluorescence of the proteins at various urea concentrations. Tryptophan fluorescence intensities were recorded at 337 and 350 nm, respectively, at various concentrations of urea. The data were then plotted as the ratio of intensities at 337 and 350 nm as a function of the urea concentration (Fig. 7). A crude estimate of the transition midpoint (C 1/2 ) from the sigmoidal analysis of the denaturation profiles revealed that the C 1/2 value increased from 2.34 M of urea for Wt aA-crystallin to 2.72 M of urea for the R12A mutant ( Fig. 7 and Table 5). This increase in the C 1/2 value clearly indicated that the substitution of R12 by alanine increased the thermodynamic stability of aA-crystallin. The C 1/2 value did not change in aB R12A and Hsp27 R12A compared to the respective Wt proteins (Table 5). These results clearly indicated that the R12 residue had a marginal influence on the structural stability of aB-crystallin and Hsp27. To quantify the stability against chemical denaturation, all of the profiles were analyzed using a global three-state fitting procedure, according to the following equation: where F 0 , F 1  , refers to the standard free energy change of unfolding (between the native and the unfolded form) at a urea concentration of zero. The fitted parameters are listed in Table 5. The standard free energy change of aA-crystallin unfolding at 25uC is 20.90 kJ/mol. This value of DG 0 is comparable to that we and others have previously reported [42,59]. The DG 0 value for aA R12A increased to 25.54 kJ/mol, indicating an enhancement in thermodynamic stability by ,4.5 kJ/mol. However, the R12A substitution had no effect on the structural stability of the other two small heat shock proteins ( Table 5). In several previous studies, investigators found that increased chaperone function of acrystallin was often associated with the greater structural stability of this protein [42,58]. Therefore, we can also conclude that structural perturbation of aA-crystallin due to the R12A mutation is a cause for the enhancement of its chaperone function.
In summary, our study showed that the molecular basis behind MGO-induced enhancement in the chaperone function of small heat shock proteins is different. Although mild MGO modification changes the conserved arginine residue (R12) in all three small heat shock proteins, this modification is likely beneficial only for aAcrystallin. Because aA-crystallin is predominantly found in the eye lens, MGO-induced enhancement in the chaperone function of this protein may be important in maintaining the transparency of the lens. Table 3. Percent levels of secondary structure in the wildtype and R12A mutants of human Hsp27 and aAand aBcrystallin using CONTINLL software.

Modification of proteins by MGO
Small heat shock proteins (1.0 mg/ml Hsp27 and aAand aBcrystallin) were incubated with either 2, 5 or 10 mM MGO (in 100 mM sodium phosphate buffer, pH 7.4) for 3 days at 37uC. The incubated samples were subjected to SDS-PAGE (12% gel) under reducing conditions.

Identification of hydroimidazolone by mass spectroscopy
Gel pieces containing aA-crystallin, aB-crystallin and Hsp27 cut from the SDS-PAGE were first destained with 50% acetonitrile in 100 mM ammonium bicarbonate, and then 100% acetonitrile. Then, the proteins were reduced by 20 mM DTT at room temperature for 60 min, followed by the alkylation of the proteins using 50 mM iodoacetamide for 30 min in the dark. The reaction reagents were removed, and the gel pieces were washed with 100 mM ammonium bicarbonate and dehydrated in acetonitrile. Sequencing grade modified trypsin (Promega, Madison, WI) in 50 mM ammonium bicarbonate was added to the dried gel pieces and incubated at 37uC overnight. Proteolytic peptides extracted from the gel with 50% acetonitrile in 5% formic acid were then dried and dissolved in 0.1% formic acid. Liquid chromatographytandem mass spectrometric analysis of the resulting peptides was performed on a LTQ Orbitrap Velos (Thermo Fisher Scientific, Waltham, MA) equipped with nanoACQUITY UPLC (Waters, Milford, MA) system. The spectra were acquired by data dependent methods consisting of a full scan and MS/MS on the ten most abundant precursor ions at the normalized collision energy of 35%. The data that were obtained were submitted to Mascot Daemon (Matrix Science, Boston, MA) for identification of the hydroimidazolone modification on arginine residues. The modification sites were then verified by manual interpretation of the MS/MS spectra.

Cloning and purification of proteins
The previously described constructs for Wt aAand aBcrystallin were used as templates [32,58]. Hsp27 cDNA from Thermo Scientific Open Biosystems, Huntsville, AL was used as a template. Wild-type heat shock proteinsaA-, aB-crystallin and Hsp27 were amplified by PCR using the following primers.
aA FP: 5-GGCCATATGGACGTGACCATCCAGCAC aARP: 3-CCCAAGCTTGGACGAGGGAGCCGAGGTG aB FP: 5-GGCCATATGGACATCGCCATCCACCAC aB RP: 3-CCCAAGCTTTTTCTTGGGGGCTGCGGTGAC Hsp27 FP: 5-GGCCATATGACCGAGCGCCGCGTCC-CCTTCTCG Hsp27 RP: 3-CCCAAGCTTTTACTTGGCGGCAGTCT-CATCGGATTT The amplified PCR product was cloned into the pET23a vector using NdeI and HindIII restriction sites. The R12A mutants were generated by site-directed mutagenesis using the respective Wt cDNA as the template. The following primers were used. aA R12A FP:5-TGGTTCAAGGCCACCCTGGGG aA R12A RP:3-CCCCAGGGTGGCCTTGAACAA aB R12A FP:5-TGGATCCGCGCCCCCTTCTTT aB R12A RP:3-AAAGAAGGGGGCGCGGATCCA Hsp27 R12A FP:5-TCGCTCCTGCGGGGCCCCAGC Hsp27 R12A RP:3-GCTGGGGCCCCGCAGGAGCGA The resulting PCR product was digested with DpnI and then transformed into E.coli DH5alpha cells. Plasmids from the resulting colonies were sequenced to confirm the presence of the mutation. The recombinant proteins were overexpressed in E.coli BL21(DE3) by induction with 250 mM IPTG when the OD 600 nm of the culture in LB broth reached ,0.6. The bacterial pellet obtained after centrifugation at 10,000 g was suspended in 50 mM Tris, pH 8.0 containing 50 mM NaCl, 2 mM EDTA and 10 ml/ ml of a protease inhibitor cocktail (Sigma). Lysozyme was added at 0.3 mg/ml to the cell suspension and incubated for 10 min at 37uC, followed by sonication on ice at 40 duty cycles at 30% amplitude. Benzonase nuclease (1.0 ml) was then added to the resulting cell lysate and incubated at 37uC in a shaker for 20 min, which was followed by the addition of sodium deoxycholate at 1.0 mg/ml and a subsequent incubation for 10 min at 37uC. DTT was then added to the lysate at a 5 mM concentration and incubated for 10 min at 37uC. The cell lysate was then centrifuged at 20,000 g for 30 min at 4uC. DNA in the lysate was precipitated by adding 0.2% polyethyleneimine followed by centrifugation at 20,000 g for 15 min. Ammonium sulfate was added to the lysate to reach 70% saturation, and the suspension was then left at 4uC overnight and then centrifuged at 20,000 g for 5 min. The resulting pellet was suspended in 50 mM sodium phosphate buffer (pH 7.4), which contained 150 mM NaCl and 5 mM DTT, and was then centrifuged at 20,000 g for 5 min. The supernatant was filtered with a 0.45 mm filter and applied onto a Superdex-200 prep grade (GE Healthcare, WI) gel filtration column that was preequilibrated with 50 mM sodium phosphate buffer pH 7.4. Fractions of 2.0 ml were collected and their OD 280 nm was recorded. The peak fractions were pooled and dialyzed overnight at 4uCin 20 mM Tris, pH 8.0 that contained 0.1 mM EDTA. The dialyzed sample was applied onto a Q-Sepharose (GE Healthcare, WI) anion exchange column equilibrated with 20 mM Tris, pH 8.0 with 0.1 mM EDTA. The bound protein was eluted with a 0-1 M NaCl gradient. The protein peak fractions were pooled and dialyzed in PBS containing 0.1 mM EDTA and stored in aliquots at 280uC.

Determination of molecular mass by multi-angle light scattering
The molar mass and the hydrodynamic radius of wild-type and R12A mutants of a-crystallin and Hsp27 were estimated by multiangle light scattering measurements as previously described [32,58]. The molar mass (M w ) and the hydrodynamic radius

Determination of secondary and tertiary structure by CD spectroscopy
The far-UV CD spectra were measured at 25uC using a Jasco 815 spectropolarimeter (Jasco, Inc., Japan). The spectra were collected from 250 to 200 nm using a cylindrical quartz cell of 2 mm path length. Proteins (0.2 mg/ml) were dissolved in 10 mM phosphate buffer (pH 7.2). The resultant spectra after five scans were analyzed for secondary structure by the curve-fitting program CONTINLL [61].
The near-UV CD spectra were measured at 25uC using the same spectropolarimeter as stated above. The spectra were measured with a 0.5 mg/ml protein solution in 50 mM phosphate buffer (pH 7.2). The reported spectra were the average of 5 scans.

Tryptophan fluorescence measurements
The intrinsic tryptophan fluorescence spectra of proteins (0.1 mg/ml) in 50 mM phosphate buffer (pH 7.2) at 25uC were recorded using a Fluoromax-4P spectrofluorometer (Horiba Jobin Mayer, USA). The excitation wavelength was set to 295 nm, and the emission spectra were recorded between 310 and 400 nm. Data were collected at a 0.5 nm wavelength resolution.

Estimation of Surface hydrophobicity
The surface hydrophobicity of the different protein solutions (0.1 mg/ml of the wild-type and R12A mutants) was measured using a hydrophobic probe, TNS (emission: 350-520 nm, excitation: 320 nm), as previously described [33]. The concentration of TNS that was used was 100 mM.

Equilibrium binding study
The chaperone-substrate binding study was performed by a membrane filtration method that we recently described [58]. Briefly, wild-type or mutant Hsp27 and aAand aB-crystallin (12.5 mM) were incubated at 60uC for 1 hr with 2-18 mM ccrystallin in 50 mM phosphate buffer containing 100 mM NaCl (pH 7.2). After equilibration, the incubation mixture was spun through a Microcon centrifugal device (4,000 g) fitted with a 100-kDa cut off membrane filter to separate the unbound substrate. The number of binding sites (n) and dissociation constant (K d ) were determined by a similar procedure [33,42,58].

Determination of structural stability of proteins
The structural stability of Wt and mutant proteins was determined by equilibrium chemical denaturation experiment. Wt and mutant proteins (0.05 mg/ml in 50 mM phosphate buffer, pH 7.5) were incubated separately with various urea concentrations (0-7 M) for 18 hrs at 25uC. Tryptophan fluorescence spectra of all samples were taken in the 300-400 nm region using 295 nm as the excitation wavelength. The equilibrium unfolding profile was fitted according to a three state model [42,53,58].