PEDS Advance Access published online on May 30, 2008
Protein Engineering Design and Selection, doi:10.1093/protein/gzn031
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Structural principles of the broad substrate specificity of Thermoactinomyces vulgaris carboxypeptidase T—role of amino acid residues at positions 260 and 262
The V.M. Stepanov Laboratory of Protein Chemistry, Institute of Genetics and Selection of Industrial Microorganisms 1, 1-ii Dorozhnyi Proezd, 117545 Moscow, Russian Federation
1 To whom correspondence should be addressed. E-mail: vhar{at}ostrov.net
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionFundingAcknowledgementsReferences |
|---|
An influence of residues at positions 260 and 262 on a broad substrate specificity of Thermoactinomyces vulgaris carboxypeptidase T (CPT) has been studied by means of site-directed mutagenesis. The structure of the S1'-site of CPT is similar to those of pancreatic carboxypeptidases A (CPA) and B (CPB); however, the enzyme is capable of cleaving off C-terminal hydrophobic (like CPA), C-terminal positively charged (like CPB), and negatively charged residues. The spatial alteration of the S1' site hydrophobic area in CPT by an insertion of one residue in the active site loop with Tyr255 by analogy with CPA and CPB did not change the enzyme specificity. The introduction of Ile262 (CPT D260G/T262I) led to a statistically significant reduction in activity towards charged substrates. The removal of a negative (CPT D260G) and placement of a positive charge (CPT D260G/T262K and CPT D260G/T262R) in the S1' site shifted the specificity of the variants towards substrates with C-terminal Glu. The selectivity profile was 64:1.7:1 for wild-type CPT, 815:115:1 for CPT D260G, 3270:1060:1 for CPT D260G/T262K and 1:2.4:0 for CPT D260G/T262R for substrates with C-terminal Leu, Glu and Arg, respectively. The obtained results confirm the important role of the amino acid residues at positions 260 and 262 in determination of the CPT substrate specificity.
Keywords: carboxypeptidase T/protein engineering/rational redesign/site-directed mutagenesis/substrate specificity
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFundingAcknowledgementsReferences |
|---|
Metallocarboxypeptidases (MCPs) are zinc-dependent exopeptidases that cleave off protein and peptide C-terminal amino acid residues and participate in food protein digestion, regulation of embryonic development, inflammation, fibrinolysis and neuropeptide processing (Arolas et al., 2007
and references therein). Classical objects of enzymology, such as mammalian carboxypeptidases A (CPA) (EC 3.4.17.1
[EC]
) (Auld, 1997) and B (CPB) (EC 3.4.17.2
[EC]
) (Aviles and Vendrell, 1997), as well as corn earworm carboxypeptidases A (CPAHa) (Estebanez-Perpina et al., 2001; Bayes et al., 2003) and glutamate-specific carboxypeptidase (HaCA42) (Bown and Gatehouse, 2004), and Thermoactinomyces vulgaris carboxypeptidase T (CPT) (EC 3.4.17.18
[EC]
) (Stepanov, 1995) are the members of this enzymatic family.
It is widely acknowledged (Reeck et al., 1971; Gardell et al., 1988; Osterman et al., 1992; Teplyakov et al., 1992) that the interactions of substrate C-terminal amino acid side chain with residues of the S1' site of enzyme binding zone [Nomenclature is in accordance with Schechter and Berger (Schechter and Berger, 1967)] (so called primary specificity pocket) completely determine the MCPs substrate specificity, as was postulated by Gardell (Gardell et al., 1988) upon visual analysis of CPA structure with dipeptide Gly-Tyr. The theory was reshaped after a comparison of kinetic parameters of eight different MCPs and their orthologs with the nature of S1' site amino acid residues carried out in this laboratory (Osterman et al., 1992). As a result, four of nine residues that were suggested by Gardell as determining the substrate specificity were rejected, and one was added. At present, it is believed that the nature of amino acid residues at positions 203, 207, 243, 250, 253, 255 and 268 (CPA numeration) define the MCPs substrate specificity. The position 255 is considered to be the major substrate specificity determinant (Reeck et al., 1971).
CPA with the hydrophobic S1' site that is partially formed by Leu203, Ile243, Ala250 and Ile255 (Table I and Fig. 1) preferably cleaves off C-terminal hydrophobic amino acid residues (Lipscomb et al., 1968; Reeck et al., 1971). CPB has the negative charge at the bottom of the substrate specificity pocket (Asp255) and hydrolyses substrates with C-terminal Arg and Lys (Wolff et al., 1962; Schmid and Herriott, 1976). When Asp255 in CPB was replaced with Arg or Lys, the obtained mutants lost their CPB-like specificity and gained a capability to cleave off dicarboxylic amino acid residues (Edge et al., 1998). Corn earworm glutamate-specific carboxypeptidase that hydrolyses C-terminal Glu and Asp also possesses Arg255 (Bown and Gatehouse, 2004). CPAHa has both hydrophobic Ile243 and polar Ser255, and as a result, in addition to hydrophobic C-terminal residues, it can remove positively and negatively charged amino acids (Estebanez-Perpina et al., 2001; Bayes et al., 2003).
|
|
T. vulgaris CPT cleaves off C-terminal positively charged residues with catalytic efficacy 6 x 103 and 61 x 103 (M–1s–1) for DnpAAK and DnpAAR (Stepanov, 1995
), which is in all probability the consequence of the Asp260 negative charge (Osterman et al., 1984; Teplyakov et al., 1992). Moreover, it catalyses the removal of C-terminal hydrophobic residues more efficiently than C-terminal positively charged amino acids with kcat/Km value being 5 x 105 (M–1s–1) for CbzAAL substrate that is only six times worse than that in CPA (3 x 106 (M–1s–1)) (Stepanov, 1995). However, unlike CPA that possesses four hydrophobic residues in its specificity pocket (Table I and Fig. 1), CPT has only two, Ala251 and Leu211, the latter being conservatively represented in both CPA and CPB.
Thus, the CPT substrate specificity differs significantly from that of CPA and CPB, and this discrepancy can not be explained readily from the positions of the presently existing theory. Both introduction of Asp at position 262 and reconstruction of CPB primary specificity pocket in CPT failed to restrict the broad substrate specificity of the enzyme (Trachuk et al., 2002; Akparov et al., 2007).
Comparison of the CPB and CPT S1' site structures performed in this laboratory showed the difference in the size of their hydrophobic area that is formed by conservative residues Tyr248(255) and Leu203(211) and also one of homologous residues Ile247 in CPB and Leu254 in CPT (Fig. 2). This difference is a putative consequence of shortening in CPT active site loop that contains Tyr255 by one residue as compared to CPB (Akparov et al., 2007). Whether or not the capability of CPT of efficiently cleaving off C-terminal hydrophobic residues is due to the difference in the loop organization was verified by lengthening of the latter by one residue (variant CPT6).
|
To elicit the role of positions 260 and 262 in the broad substrate specificity of CPT, the negatively charged Asp260 was replaced by neutral Gly (CPT D260G). Then Ile was placed at position 262 in the same way as in CPA (CPT D260G/T262I). Finally, in order to reverse the CPT specificity towards negatively charged substrates, CPT variants with Arg and Lys at the key 262 position (CPT D260G/T262K and CPT D260G/T262R) similarly with the structure of glutamate-specific carboxypeptidase (Bown and Gatehouse, 2004
) and CPB variants with the selectivity towards C-terminal dicarboxylic amino acid residues (Edge et al., 1998) were obtained. |
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFundingAcknowledgementsReferences |
|---|
Materials
DNA endonucleases, T4 DNA ligase, Pfu DNA polymerase, T7 promoter primer were purchased from MBI Fermentas (Lithuania). T7 terminator primer, pET23a vector and E.coli BL21(DE3)pLysS strain were from Novagen (USA). Primers bearing the desired mutations were produced by Sintol (Russia). The initial construction pET-CPTwt encodes the wild-type CPT (CPTwt) in the pET23a vector. The open reading frame of a chimerical gene contains 17 N-terminal residues of T7 Tag followed by the sequence of 72 residues of prepropeptide and 326 residues of mature CPTwt. The first amino acid of CPTwt in the construction corresponds to the 21st amino acid residue in T. vulgaris cpT gene (Smulevitch et al., 1991) whose sequence was deposited in EBI/EMBL Gene Bank (accession number X56901
[GenBank]
). Construction pET-CPT5 containing gene cpT5 with mutations that lead to amino acid replacements G215S/A251G/T257A/D260G/T262D (Table I) was prepared as described in the previous publication (Akparov et al., 2007). A Superdex TM 75 HR 10/30 column for molecular exclusion chromatography and cation-exchange resin SP-Sephadex C-25 were received from Pharmacia LKB Biotechnology (Sweden). PM10 membranes and stirred ultrafiltration cells were from Amicon (USA). Affinity resin [N-(
-aminocaproyl)-p-aminobenzyl]succinyl-Sepharose 4B (CABS-Sepharose) was prepared in this laboratory according to Cueni et al. (Cueni et al., 1980), as well as peptide substrates for the kinetic assays CbzAALpNA (Liublinskaia et al., 1977), DnpAAR (Iusupova et al., 1995), CbzAAL and CbzAAE (Voiushina et al., 1987).
Protein structure superimposition and side chain rotation were carried out in Deep View (Guex and Peitsch, 1997). For the structure visualization, VMD (Humphrey et al., 1996) was used. Molecular mechanics was performed using Gromacs 3.3.1 software (Van der Spoel et al., 2005).
CPT and CPB structures used for energy minimization were designed on the basis of the experimentally determined structures [PDB ID 1OBR for CPT (Teplyakov et al., 1992) and 1ZG9 for CPB (Adler et al., 2005)] by removal of the inhibitor and sulphate anion atoms and addition of the substrate atoms. Phe-Phe-Val-Phe substrate position in the active sites of the enzymes was designed similarly with the structure of a complex of CPA with the phosphonate inhibitor Bz-Phe-ValP-(O)-Phe (PDB ID 7CPA) (Kim and Lipscomb, 1991) that simulates the transition state of Phe-Phe-Val-Phe hydrolysis. Position of the C-terminal Phe side chain was manually altered so that it looked into the S1' site hydrophobic area forming unfavourable steric contacts with its walls. Ten such complexes were obtained both for CPT and CPB with systematic rotation about C
-Cβ bond.
Preparation of cpT gene mutants
Extension overlap PCR (Ho et al., 1989) was used to prepare cpT gene mutants with T7 promoter and T7 terminator flanking primers and the forward and reverse mutagenesis primers whose sequence was complementary to that of the gene with the exception of the introduced mutations. The mutant genes were cloned in pET23a vector (Novagen pET System Manual, 1997) according to Sambrook et al. (Sambrook et al., 1989). The cpT6 mutant gene had Thr codon ACC introduced between Asp253 and Leu254 codons in addition to the mutations in cpT5. The cpT D260G mutant gene had Asp260 codon GAT replaced by Gly codon GGT. The cpT D260G/T262I, cpT D260G/T262K and cpT D260G/T262R mutant genes had Thr codon ACC replaced by Ile codon ATC, Lys codon AAA and Arg codon CGC, respectively, in addition to the mutation of cpT D260G.
The presence of desired mutations and the identity of non-targeted portions of the mutant sequences to the CPTwt gene (Smulevitch et al., 1991) were proved by full-length gene sequencing.
Expression of CPT proenzyme in E. coli in the form of inclusion bodies and subsequent renaturation of the enzyme in vitro
The CPT mutant genes were expressed in E. coli strain BL21(DE3)pLysS according to the instructions of Novagen (Novagen pET System Manual, 1997). After IPTG-induced expression, the cells were precipitated and sonicated. Purification of CPT from inclusion bodies was carried out according to Trachuk et al. (Trachuk et al., 2005). Briefly, the inclusion bodies were separated by centrifugation, washed with 0.05% CHAPS (w/v), 2 M NaCl and water, and dissolved in 8 M urea up to the final concentration of 5 mg/ml. Then the protein solution was 10-fold fast diluted with 50 mM Tris–HCl/30% glycerol (v/v, hereinafter the percentage is in volume-to-volume ratio)/0.5 M NaCl/10 mM CaCl2, pH 8.0, and incubated for 16 h at 37°C. The solution was further 2-fold diluted with 50 mM Tris–HCl/0.5 M NaCl/10 mM CaCl2, pH 8.0, and concentrated by ultrafiltration up to the volume of 20 ml. In order to activate the proenzyme, subtilisin 72 was added at the molar ratio CPT:subtilisin 200:1 and the solution was incubated for 30 min at 37°C. Then subtilisin was inactivated with diisopropyl fluorophosphate. After that, the protein solution was concentrated by ultrafiltration up to the volume of 0.5 ml.
The activated CPT was purified by molecular exclusion chromatography in an FPLC system on the Superdex TM 75 HR 10/30 column, pre-equilibrated with 10 mM Tris–HCl/0.5 M NaCl/10 mM CaCl2, pH 7.5. After gel-filtration, pH value of the protein solution was adjusted to 6.0 by 100 mM MES-NaOH buffer, pH 6.0. Then the protein solution was concentrated up to the volume of 2 ml and loaded on the 2 ml CABS-Sepharose column (Cueni et al., 1980). The protein was washed with 4 ml of 10 mM MES-NaOH/0.5 M NaCl/10 mM CaCl2 and eluted with 10 mM Tris–HCl/0.5 M NaCl/10 mM CaCl2, pH 8.0.
The protein concentration was determined by the Bradford assay (Bradford, 1976) and by absorption at 280 nm using 280 = 64 000 M–1cm–1 that was predicted for CPT according to Pace et al. (Pace et al., 1995). SDS–polyacrylamide gel electrophoresis was performed according to Laemmli (Laemmli, 1970).
Circular dichroism spectroscopy
Protein concentration in the samples was 0.5–1.3 mg/ml. Far-UV CD spectra were recorded using Jasco J-810 spectropolarimeter (Tokyo, Japan) between 190 and 250 nm (20 nm/s) with 1 nm slit at 24°C. A 0.01 cm path-length cell (Suprasil quartz) with a detachable window (Hellma GmbH & Co. KG, Müllheim, Germany) was used. Baseline spectra were recorded from the corresponding buffers and subtracted from the protein spectra. Data were analysed for peptide secondary structure using the latest version of CONTINLL program (http://lamar.colostate.edu/~sreeram/CDPro/) as described (Sreerama and Woody, 2000).
Investigation of enzymatic properties of the CPT mutants
The subtilisin activity in CPT solution was tested with the specific substrate ZAALpNA applied in a concentration of 0.1 mM in the 50 mM Tris–HCl/10 mM CaCl2 buffer, pH 8.5. The reaction mixture was incubated at 37°C until the pale yellow colour appeared. The reaction was stopped by the addition of 0.5 M HCl. The enzyme amount was calculated taking into account 410 (p-nitroaniline molar extinction coefficient) = 8200 M–1cm–1, and the activity of pure enzyme of 10 units/1 mg.
The enzymatic activity of CPT mutants was characterized by the ability to cleave off a Arg residue from DnpAAR, Leu residue from CbzAAL and Glu residue from CbzAAE. Hydrolysis of DnpAAR was performed at 37°C in 50 mM Tris–HCl/10 mM CaCl2, pH 7.5, containing the 0.35–3.5 mM substrate. The reaction was stopped by the addition of 50% CH3COOH. The product DnpAA was separated from the positively charged substrate on SP-Sephadex C-25 resin and its amount was measured spectrophotometrically (360 = 15 000 M–1cm–1). Hydrolysis of CbzAAL and CbzAAE was performed at 25°C in 50 mM Tris–HCl/10 mM CaCl2, pH 7.5, containing 40–400 µM CbzAAL or 0.1–1 mM CbzAAE. The observed optical density decrease at 225 nm was recalculated using 225 = 376 M–1cm–1. The enzyme concentration was chosen individually for each mutant.
The initial rates at four substrate concentrations from the above range were determined. For each substrate concentration, the rates were measured four times. The kinetic data were processed by linear and nonlinear regression in Origin v.6.1 software (www.originlab.com) taking into account all the data on the initial rates.
The procedures of expression, renaturation, activation, purification and kinetic constants measurement were performed in triplicate for all CPT mutants.
Energy minimization of 10 CPT and CPB complexes with Phe-Phe-Val-Phe was carried out in vacuo by mdrun program of Gromacs 3.3.1 molecular modelling software (Van der Spoel et al., 2005) using GROMOS 53A6 force-field parameter set (Oostenbrink et al., 2004) and a mixed algorithm of minimization: 100 step of conjugated gradient and 1 step of steepest descent. Electrostatic and van der Waals interactions were treated with 1.4 cut-off. Minimized structures were obtained after 1000 steps of energy minimization. As a result of the above manipulations, the system energy reached its minimum.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFundingAcknowledgementsReferences |
|---|
Production of CPT variants
The genes of variants were expressed in E. coli and produced as inclusion bodies containing proenzymes with the yield of 50% of total cell protein. The yield of the renaturated, activated and purified enzymes was 1.5% of protein amount taken for renaturation and 
3–5 mg per litre of growth medium both for CPTwt and CPT variants. The final purity of variants was no <95% according to the SDS–electrophoresis data. The amount of subtilisin in the variant preparations as was tested by hydrolysis of chromogenic substrate CbzAALpNA was <0.00015%. Kinetic parameters of CPTwt renaturated according to the described procedure upon the hydrolysis of DnpAAR and CbzAAL were in good agreement with those of the enzyme purified from T. vulgaris growth broth (Stepanov, 1995; Akparov et al., 2007).
The CPT and CPB structures with Phe-Phe-Val-Phe were obtained after energy minimization. The CPT primary specificity pocket possessed enough space at its entrance near to Tyr255 and Leu254 residues to accommodate bulky C-terminal Phe side chain (Fig. 2). The same substrate position was not observed in CPB complexes because Ile247 side chain pushed benzene ring 1.5 Å towards the specificity pocket bottom. This difference is due to different side chain position of homologous Leu254 of CPT and Ile247 of CPB (Fig. 2). Thus the primary specificity pocket hydrophobic area is more spacious and more suitable for binding to large hydrophobic substrates in CPT than in CPB.
Circular dichroism spectroscopy
The spectra of all mutants were similar to that of CPTwt (Fig. 3). The major differences were observed at 190–195 nm range where the absorbtion of buffer itself was significant. The calculated secondary structure composition for CPTwt: 35–36% of residues in -helice conformation and 14–19% in β-sheet conformation is in good agreement with X-ray data for this enzyme: 37 and 14% of residues are in - and β-structure, respectively.
|
] is expressed in (deg·cm2·decimole–1) units.Enzyme activity and kinetics
The removal of the negative charge from the binding area of CPT (CPT D260G) reduced 10-fold and increased 6.5-fold the activity towards substrates with C-terminal Arg and Glu, respectively, and did not affect the catalysis of the hydrophobic substrate (Table II). CPT D260G was 7- and 815-fold more active with hydrophobic than with negatively and positively charged substrates, respectively. The changes in activity were at least partially caused by shifts in Km value, which increased for DnpAAR and decreased for ZAAE upon mutation.
|
The introduction of Lys or Arg residue at position 262 (CPT D260G/T262K and CPT D260G/T262R) drastically, 21-fold for CPT D260G/T262K, reduced the activity towards positively charged substrate. Surprisingly, the Km value of the variant was similar to that of wild-type enzyme, while kcat value decreased 360 times. The low activity of CPT D260G/T262R was <5 (M–1s–1) that is the limit of detection assay. The activity towards hydrophobic substrate reduced 5- and 23-fold for CPT D260G/T262K and CPT D260G/T262R, respectively. In the case of CPT D260G/T262K, the drop in activity was cased by changes in both Km and kcat. The lowered activity of CPT D260G/T262R was due only to Km shift. At the same time, the catalytic efficiency of the variants towards negatively charged substrate did not exhibit substantial changes because the augmentation of Km values was compensated by the augmentation of kcat.
The value [kcat/Km]Glu/[kcat/Km]Arg increased 68-fold upon the removal of the negative charge and 620-fold or even more upon the introduction of the positively charged residues in the primary specificity pocket in comparison with the wild-type enzyme which is in good accordance with the results obtained for the other proteases (Wells et al., 1987; Bech et al., 1992; Olesen et al., 1994; Ballinger et al., 1995; Sorensen and Breddam, 1997).
Regardless of the decrease in the activity towards the hydrophobic substrate, the initial value was so high that the introduction of Arg at position 262 only led to the obtaining of a protease 2.4-fold more active towards the negatively charged than hydrophobic substrate. The introduction of Lys at this position only resulted in a 2-fold increase in the selectivity to CbzAAE when compared with CbzAAL.
The replacement of Thr262 with Ile by analogy with CPA (CPT D260G/T262I) reduced the activity towards DnpAAR and CbzAAE by 2.6 and 1.4 times, respectively. The hydrolysis rate of the hydrophobic substrate remained at the same level: kcat/Km*10–3 = 282 (M–1s–1) for CPT D260G and 184 (M–1s–1) for CPT D260G/T262I.
The insertion of one residue (Thr246) in the active site loop of CPT5 by analogy with CPA and CPB (CPT6) reduced the hydrolytic efficacy by one order of magnitude for all the substrates. The specificity profile of CPT6 259:1:3.8 for the substrates with C-terminal Leu, Glu and Arg remained similar with that of CPT5, 172:1:1.5, respectively (Table III). Like both CPT5 and CPTwt, CPT6 hydrolysed the hydrophobic substrate by two orders of magnitude more efficiently, than the positively charged compound.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFundingAcknowledgementsReferences |
|---|
CPT has so far been considered to possess a ‘dual’ substrate specificity being capable of cleaving off C-terminal positively charged (Arg and Lys) and hydrophobic (Trp, Phe, Leu, Ile, Met, Ala) residues (Osterman et al., 1984
). The obtained data strongly suggest that CPT is able to cleave off negatively charged as efficiently as positively charged residues with the substrate specificity profile 64:1.7:1 for substrates with C-terminal Leu, Glu and Arg, respectively (Table III). Thus, CPT appears to be an enzyme with a broad specificity resembling carboxypeptidase CPAHa that is able to cleave off C-terminal hydrophobic (including Ala and Trp) as well as Arg and Glu residues (Bayes et al., 2003). The renaturation yields of CPT variants that are comparable with that of the wild-type enzyme, the capacity of binding to the affinity resin, the resistance to the proteolytic degradation by subtilisin and occurrence of the enzymatic activity similar with that of CPTwt presuppose the correct folding of the variants. CD data corroborate this statement too.
We showed earlier that the replacement of five of seven residues determining the substrate specificity of MCPs (Table I) in CPT by the analogous residues from CPB (variant CPT5) did not practically change the substrate specificity of the enzyme (Table II) (Akparov et al., 2007). Nevertheless, the structural differences in S1' sites of CPB and CPT related to the shortening by one residue in CPT Tyr255-containing active site loop remained (Fig. 2). This loop is involved in the induced fit and spans residues 252–256. As a consequence, according to the data obtained by us using molecular mechanics, the CPT primary specificity pocket proves to be more spacious for binding to large substrates with hydrophobic side chains.
In the present investigation, the mutant CPT6 that is free of the above discrepancy was obtained. However, contrary to expectations, the hydrophobic specificity of CPT6 failed to be reduced. The general decrease in the hydrolytic efficacy was only observed that might be a result of less favourable position of Tyr255 in the ‘closed’ complex for catalysis or increase in the binding energy that is necessary for the enzyme distortion. Probably, the structural determinants of the narrow specificity in CPB are located beyond the S1' site, like those in trypsin (Perona and Craik, 1995).
At the same time, the analysis of kinetics of variants with replacements at position 260 and 262 [the latter being the key position (Reeck et al., 1971) according to the theory of substrate specificity for MCPs (Gardell et al., 1988; Osterman et al., 1992)], shows the correctness of the theory application to CPT. Indeed, after the negative charge of Asp260 was removed and the positively charged residues were introduced at the key position 262, the specificity of the variants changed in accordance with the theory (Table III).
The removal of the negative charge of Asp260 (CPT D260G) decreased the activity towards DnpAAR and enhanced the catalytic efficacy towards CbzAAE testifying to the fact that this residue binds to the positive and repulses the negative charge of arginine and glutamate substrates. Thus Asp260 functions as both positive and negative substrate specificity determinant in CPT. The CPT D260G activity towards both charged substrates implies the presence of a sufficiently polar area in the specificity pocket that interacts with guanidine or carboxylic group of the substrate C-terminal residue side chain. This area, as was confirmed by the X-ray structure (Teplyakov et al., 1992), is formed by the polar OH-groups of Thr257, Thr275, CO- and NH-groups of the polypeptide backbone and structurally bound water molecules.
The introduction of the positive charge at position 262 (CPT D260G/T262K and CPT D260G/T262R) should have caused a strong electrostatic repulsion with guanidine group of substrate C-terminal arginine. Indeed, the enzyme hydrolytic efficacy towards positively charged substrate dropped 20-fold or more (Table II). The variant CPT D260G/T262R hydrolysed the negatively charged substrate better, than CPT D260G/T262K, which was putatively caused by the ability of guanidine group to form more hydrogen bonds with the substrate than amino group. The observed an order-of-magnitude reduction of the activity towards hydrophobic substrates may be a result of shrinkage of the primary specificity pocket or energetically unfavourable desolvation of a charged amino or guanidine group of the enzyme by a substrate hydrophobic side chain. Thus, we have succeeded to qualitatively reproduce the analogous experiment on the conversion of CPB into glutamate-specific carboxypeptidase (Edge et al., 1998). However, the different substrates used in the two experiments prevented the opportunity of their quantitative comparison.
The introduction of hydrophobic Ile260 instead of Thr (CPT D260G/T262I) reduced the catalysis rate of charged substrates. However, this variant with its greater resemblance in primary structure with CPA, than the wild-type enzyme, unlike CPA (Slobin and Carpenter, 1966), hydrolysed negatively charged substrates by only an order of magnitude weaker than hydrophobic substrates. It is therefore possible that Ile243 is the negative determinant for binding to charged substrates in CPA that lacks in CPT D260G/T262I with Ala at homologous position 251.
The replacements D260G and T262I did not influence the activity towards hydrophobic substrates, which implies residues 260 and 262 in CPT not to form a direct van der Waals contact with a substrate hydrophobic side chain. The given hypothesis was put forward to explain a broad substrate specificity of CPAHa. The enzyme residues Ile243, Ala250 and Leu247 participate in binding to hydrophobic substrates, whereas Ser255 (instead of Asp255 in CPB and Ile255 in CPA) makes the primary specificity pocket more polar and spacious thus facilitating the binding to charged substrates (Estebanez-Perpina et al., 2001).
At the same time, not all of the observed results agree with the simple scheme. For example, the introduction of positively charged residue at position 262 (CPT D260G/T262K and CPT D260G/T262R) had been expected to improve the binding of negatively charged substrates via electrostatic interactions when compared with CPT D260G that has no charge at this position. However, in practice, the little shifts of the hydrolytic efficiency towards the glutamate substrate were observed. Why does CPTwt prefer substrates with C-terminal Glu to substrates with C-terminal Arg is an open question. This evidence occurs regardless of the presence of the negative charge at the bottom of the specificity pocket and a longer aliphatic chain in the arginine compared with the glutamate that should provide a tighter binding to the CPT S1' site hydrophobic area. Moreover, the simple additive model presupposes arginine substrates to bind stronger than hydrophobic substrates due to the capacity of formation a salt bridge with Asp260 in addition to hydrophobic interactions. Nevertheless, the CPT substrate specificity is mainly hydrophobic. These discrepancies suggest a more complex mechanism of CPT substrate selectivity determination than it is believed to be.
Thus, guided by the classical concept of structural principles of MCPs substrate specificity (Reeck et al., 1971; Gardell et al., 1988; Osterman et al., 1992), we have succeeded for the first time in substantial shifting of the CPT substrate specificity. The specificity shift towards negatively charged substrates expressed in [kcat/Km]Glu/[kcat/Km]Leu and [kcat/Km]Glu/[kcat/Km]Arg values was 90- and >3000-fold, respectively, for CPT D260G/T262R when compared with CPTwt. A comparison of available data on CPT with kinetic and structural characteristics of homologous enzymes points at the occurrence of other presently unknown determinants of substrate specificity, probably remote from the active site.
|
Funding |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFundingAcknowledgementsReferences |
|---|
Russian Foundation for Basic Research (02-04-48 755); Russian Federal Agency for Science and Innovation (State Contract No: 02.531.11.900 [EC] 3).
|
Footnotes |
|---|
Abbreviations: IPTG, isopropyl-beta-D- thiogalactopyranoside; DnpAAR, 2,4-dinitrophenyl- alanyl-alanyl-arginine; CbzAAL, benzyloxycarbonyl- alanyl-alanyl-leucine; CbzAAE, benzyloxycarbonyl- alanyl-alanyl-glutamic acid; CbzAALpNA, benzyloxycarbonyl- alanyl-alanyl-leucine-p-nitroanilide; Bz-Phe-ValP-(O)- Phe, O-(((1R)-((N-(phenylmethoxycarbonyl)- L-phenylalanyl)amino)isobutyl)hydroxyphosphinyl)-L-3-phenyllactate.
|
Acknowledgements |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFundingAcknowledgementsReferences |
|---|
Authors are grateful to T.L. Voyuishina, M.P. Yusupova, G.S. Konstantinova, A.M. Bushueva, S.P. Sineokii and E. Tikhonova for the technical assistance.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFundingAcknowledgementsReferences |
|---|
Adler M., et al. Biochemistry (2005) 44:9339–9347.[CrossRef][Web of Science][Medline]
Akparov V.Kh., Grishin A.M., Yusupova M.P., Ivanova N.M., Chestukhina G.G. Biochemistry (Mosc) (2007) 72:416–423.[CrossRef][Medline]
Arolas J.L., Vendrell J., Aviles F.X., Fricker L.D. Curr. Pharm. Des. (2007) 13:349–366.[CrossRef][Web of Science][Medline]
Auld D.S. Handbook of Proteolytic Enzymes—Barrett A.J., Rawlings N.D., Woessner J.F., eds. (1997) Academic Press, Inc. 1321–1328.
Aviles F.X., Vendrell J. Handbook of Proteolytic Enzymes—Barrett A.J., Rawlings N.D., Woessner J.F., eds. (1997) Academic Press, Inc. 1333–1335.
Ballinger M.D., Tom J., Wells J.A. Biochemistry (1995) 34:13312–13319.[CrossRef][Web of Science][Medline]
Bayes A., Sonnenschein A., Daura X., Vendrell J., Aviles F.X. Eur. J. Biochem. (2003) 270:3026–3035.[Web of Science][Medline]
Bech L.M., Sorensen S.B., Breddam K. Eur. J. Biochem. (1992) 209:869–874.[Web of Science][Medline]
Bown D.P., Gatehouse J.A. Eur. J. Biochem. (2004) 271:2000–2011.[Web of Science][Medline]
Bradford M.M. Anal. Biochem. (1976) 72:248–254.[CrossRef][Web of Science][Medline]
Cueni L.B., Bazzone T.J., Riordan J.F., Vallee B.L. Anal. Biochem. (1980) 107:341–349.[CrossRef][Web of Science][Medline]
Edge M., et al. Protein Eng. (1998) 11:1229–1234.
Estebanez-Perpina E., Bayes A., Vendrell J., Jongsma M.A., Bown D.P., Gatehouse J.A., Huber R., Bode W., Aviles F.X., Reverter D. J. Mol. Biol. (2001) 313:629–638.[CrossRef][Web of Science][Medline]
Gardell S.J., Craik C.S., Clauser E., Goldsmith E.J., Stewart C.B., Graf M., Rutter W.J. J. Biol. Chem. (1988) 263:17828–17836.
Guex N., Peitsch M.C. Electrophoresis (1997) 18:2714–2723.[CrossRef][Web of Science][Medline]
Ho S.N., Hunt H.D., Horton R.M., Pullen J.K., Pease L.R. Gene (1989) 77:51–59.[CrossRef][Web of Science][Medline]
Humphrey W., Dalke A., Schulten K. J. Molec. Graphics (1996) 14:33–38.[CrossRef]
Iusupova M.P., Kotlova E.K., Timokhina E.A., Stepanov V.M. Bioorg. Khim. (1995) 21:33–38. (Russian).[Web of Science][Medline]
Kim H., Lipscomb W.N. Biochemistry (1991) 30:8171–8180.[CrossRef][Web of Science][Medline]
Laemmli U.K. Nature (1970) 227:680–685.[CrossRef][Medline]
Lipscomb W.N., Hartsuck J.A., Reeke G.N. Jr, Quiocho F.A., Bethge P.H., Ludwig M.L., Steitz T.A., Muirhead H., Coppola J.C. Brookhaven Symp. Biol. (1968) 21:24–90.[Medline]
Liublinskaia L.A., Iakushcheva L.D., Stepanov V.M. Bioorg Khim (1977) 3:273–279. (Russian).[Web of Science]
Novagen pET System Manual TB055. (1997) 7th edn. Madison, WI: Novagen.
Olesen K., Mortensen U.H., Aasmul-Olsen S., Kielland-Brandt M.C., Remington S.J., Breddam K. Biochemistry (1994) 33:11121–11126.[CrossRef][Web of Science][Medline]
Oostenbrink C., Villa A., Mark A.E., van Gunsteren W.F. J. Comput. Chem. (2004) 25:1656–1676.[CrossRef][Web of Science][Medline]
Osterman A.L., Stepanov V.M., Rudenskaia G.N., Khodova O.M., Tsaplina I.A. Biokhimiia (1984) 49:292–301. (Russian).[Medline]
Osterman A.L., Grishin N.V., Smulevitch S.V., Matz M.V., Zagnitko O.P., Revina L.P., Stepanov V.M. J. Protein Chem. (1992) 11:561–570.[CrossRef][Web of Science][Medline]
Pace C.N., Vajdos F., Fee L., Grimsley G., Gray T. Protein Sci. (1995) 4:2411–2423.[Web of Science][Medline]
Perona J.J., Craik C.S. Protein Sci. (1995) 4:337–360.[Web of Science][Medline]
Reeck G.R., Walsh K.A., Hermodson M.A., Neurath H. Proc. Natl. Acad. Sci. USA (1971) 68:1226–1230.
Sambrook J., Fritsch E.F., Maniatis T. Molecular Cloning: A Laboratory Manual (1989) 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press.
Schechter I., Berger A. Biochem. Biophys. Res. Commun. (1967) 27:157–162.[CrossRef][Web of Science][Medline]
Schmid M.F., Herriott J.R. J. Mol. Biol (1976) 103:175–190.[CrossRef][Web of Science][Medline]
Sreerama N., Woody W.N. Anal. Biochem. (2000) 282:252–260.
Slobin L.I., Carpenter F.H. Biochemistry (1966) 5:499–508.[CrossRef][Web of Science][Medline]
Smulevitch S.V., Osterman A.L., Galperina O.V., Matz M.V., Zagnitko O.P., Kadyrov R.M., Tsaplina I.A., Grishin N.V., Chestukhina G.G., Stepanov V.M. FEBS Lett. (1991) 291:75–78.[CrossRef][Web of Science][Medline]
Sorensen S.B., Breddam K. Protein Sci. (1997) 6:2227–2232.[Web of Science][Medline]
Stepanov V.M. Methods in Enzymology—Barrett A.J., ed. (1995) 248. Academic Press, Inc. 675–683.[Web of Science][Medline]
Teplyakov A., et al. Eur. J. Biochem. (1992) 208:281–288.[Web of Science][Medline]
Trachuk L.A., Bushueva A.M., Shevelev A.B., Novgorodova S.A., Akparov V.Kh., Chestukhina G.G. Vopr. Med. Khim. (2002) 48:577–585. (Russian).[Web of Science][Medline]
Trachuk L., Letarov A., Kudelina I.A., Yusupova M.P., Chestukhina G.G. Protein Expr. Purif. (2005) 40:51–59.[CrossRef][Web of Science][Medline]
Van Der Spoel D., Lindahl E., Hess B., Groenhof G., Mark A.E., Berendsen H.J. J. Comput. Chem. (2005) 26:1701–1718.[CrossRef][Web of Science][Medline]
Voiushina T.L., Liublinskaia L.A., Timokhina E.A., Stepanov V.M. Bioorg. Khim. (1987) 13:615–622. (Russian).[Web of Science][Medline]
Wells J.A., Powers D.B., Bott R.R., Graycar T.P., Estell D.A. Proc. Natl. Acad. Sci. USA (1987) 84:1219–1223.
Wolff E.C., Schirmer E.W., Folk J.E. J. Biol. Chem. (1962) 237:3094–3099.
Received January 15, 2008; revised April 30, 2008; accepted May 6, 2008.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||