Human Dipeptidyl Peptidase III : the Role of Asn 406 in Ligand Binding and Hydrolysis †

Human dipeptidyl peptidase III (DPP III) is a member of the zinc-metallopeptidase family M49 with a role in intracellular protein catabolism. Ligand-free crystal structure of yeast and human orthologues provided insight into the enzyme's active center and enabled some assumptions about the enzyme substrate interactions. The molecular modeling performed for human and yeast DPP III indicated involvement of Asn406 and Asn415, respectively in ligand binding. To investigate further the role of this asparagine residue, conserved in all eukaryotic M49 peptidases, a site-directed mutagenesis of human DPP III was carried out. Replacement of Asn406 with Gln, but not with Ser, resulted in a large decrease in the enzyme's binding affinity for competitive peptide inhibitor and in catalytic efficiency. Molecular dynamics simulations performed for enzyme-substrate complexes, revealed changed coordination of the active-site zinc ion and the orientation of the catalytically important His568, in N406Q, but not in the N406S complexes.(doi: 10.5562/cca1808)


INTRODUCTION
Dipeptidyl peptidases (DPPs) cleave dipeptides from the (poly)peptides amino termini.In mammalian tissues originally four types of dipeptidyl peptidases, DPP I to IV have been recognized through hydrolysis of distinct synthetic substrates (unsubstituted dipeptidyl 2naphthylamides) and extensively characterized. 1 Dipeptidyl peptidase III (DPP III; EC 3.4.14.4) is a widelydistributed cytosolic enzyme which selectively hydrolyses arginyl-arginyl-2-naphthylamide (Arg 2 -2NA). 2 It was at first isolated and biochemically characterized from several human (placenta, erythrocytes, seminal plasma) and animal (bovine pituitary gland and adrenal medulla, porcine spleen, rat and guinea pig brain) tissues 3−11 as monomeric acidic metalloexopeptidase (M r ≈ 80 000, pI ≈ 4.5) with a broad specificity whose optimally sized substrates are tetra-to octapeptides. 5,10The same type of enzyme was partially purified and described from two lower eukaryotes, the slime mold Dictyostelium discoideum and the yeast Saccharomyces cerevisiae. 12,13The physiological substrates of DPP III are mostly unknown, although it hydrolyzes a number of biologically active peptides in vitro.Besides its contribution in the final steps of intracellular protein catabolism, a regulatory role for DPP III was suggested. 5,9igh concentrations of DPP III in the rat spinal cord and its high affinity towards opioid peptides enkephalins and endomorphins imply its role in the mammalian pain-modulatory system. 14,15Enhanced level of human DPP III activity and protein was found in endometrial carcinomas and in ovarian malignant neoplasms, 16 moreover correlation of DPP III activity and aggressiveness of ovarian primary carcinomas was shown. 17It was proposed that DPP III, with several other endogenous peptidases in the lens plays an important role in cataractogenesis by inducing lens opacification through the intracellular turnover of lens proteins. 18Recently, upregulated DPP III was identified as a proteomic biomarker for prenatal bisphenol A-exposure in mouse Croat.Chem.Acta 84 (2011) 259.
immune organs. 19The first DPP III amino acid sequences determined have been rat (738 amino acids) and human (737 amino acids), both possessing the characteristic HELLGH motif that resembles the HExxH one present in many metallopeptidases. 2,20Fukasawa et al. demonstrated by atomic absorption spectrometry that both the recombinant rat DPP III and the natural human placental enzyme contain about one Zn 2+ ion per molecule. 20The flow of data on complete genome sequences in the last decade enabled recognition of dipeptidyl peptidase III family (peptidase family M49) as a distinct group of metallopeptidases, based on their sequence relationships and the unique hexapeptide motif HExxGH. 21,22Mutational analysis of rat DPP III has confirmed the involvement of two histidines from the HELLGH motif (residues 450−455) and Glu508 in coordination of the active-site zinc, while for Glu451 it was shown to be crucial for the catalytic activity. 23To reveal yet unknown constituents of DPP III active site, we used bioinformatics to find the amino acid residues which are conserved in all M49 family members, and site-directed mutagenesis of human DPP III.By such combined approach, a functional role for Tyr318 and Trp300 was demonstrated in human DPP III. 24,25n 2008, the gene encoding yeast Saccharomyces cerevisiae DPP III was cloned and heterologously expressed, and the crystal structure of the yeast orthologue was solved revealing a novel protein fold -a prototype for the whole M49 family. 26DPP III molecule is divided into two domains by a cleft containing the catalytic zinc ion.The human DPP III 3-D structure recently solved as well (PDB code: 3fvy) closely resembles that of the yeast protein (PDB code: 3csk).
Although the crystal structures of yeast and human DPPs III (both ligand-free) provided an insight into the possible catalytic mechanism, the molecular details on substrate recognition by this type of enzyme were missing.
In our previous study we investigated ligand binding to human DPP III by combining molecular docking and molecular dynamics (MD) simulations, with sitedirected mutagenesis and biochemical experiments. 27e found that evolutionary conserved His568 is crucial for the catalytic efficiency of human DPP III: its substitution with Asn resulted 1300-fold decrease of k cat for the Arg 2 -2NA hydrolysis in respect to the wild-type enzyme, and significantly reduced the enzyme affinity towards the competitive inhibitor Tyr-Phe-NHOH.Mo-lecular modeling has indicated that His568 is a constituent of the S1' subsite (with 7 other amino acid residues) and a possible proton donor during hydrolysis.The same modeling study pointed out Asn406 as one out of 3 residues forming the S2 subsite which interacts with a ligand P2 residue.In order to check importance of Asn406 for the enzymatic activity of the human DPP III, we substituted this residue by site-directed mutagenesis with Gln and Ser, overexpressed and purified the mutated forms and examined their catalytic properties.

Cloning and Site-directed Mutagenesis
To obtain DPP III protein with affinity C-terminal sixhistidine tag, full cDNA for DPP III (2.214 kb, contained originally in the clone IRALp962E0242Q2), 24 was cloned into pET-21b vector between NheI and XhoI restriction sites according to procedure described earlier by Špoljarić et al. 25 Thus obtained wild-type expression vector was used for further mutagenesis and protein expressions.

Heterologous Expression and Purifications of Recombinant DPP III Proteins
Heterologous expression was performed as for human DPP III without His 6 -tag, 25 with one modification: after induction, cells were grown overnight at 18 °C.His 6 -tagged DPP III proteins were purified by affinity chromatography on nickel-nitrilotriacetic acid agarose (Ni-NTA Agarose, Qiagen) as already described. 28ll fractions of high purity (according to the SDS-PAGE analysis and the enzymatic activity assayed using Arg 2 -2NA as substrate) were pooled and desalted on PD-10 columns (GE Healthcare) equilibrated with 20 mmol dm −3 Tris-HCl buffer (pH = 7.4).Purified protein preparations were stored in 15.6 mmol dm −3 Tris-HCl buffer (pH = 7.4) containing 22 % glycerol at −10 °C.
Protein purity was confirmed by gel electrophoresis under denaturing conditions, SDS-PAGE, carried out according to the method of Laemmli. 29rotein concentrations were determined using the protein-dye binding assay, with bovine serum albumin as a standard. 30

Enzyme Activity Assay and Determination of Kinetic Parameters
The enzymatic activities of wild-type and mutant DPP III were determined by a standard assay at 37 o C with Arg 2 -2NA as a substrate. 31Kinetic parameters were determined at 25 o C and at pH = 8.6, in the presence of 50 μmol dm −3 CoCl 2 by initial rate measurements. 25ixon plots were used to determine the K i values for hydroxamate inhibitor at pH = 8.0.

Similarity Search and Sequence Comparisons
Sequences of human DPP III homologs from other species were obtained by BLAST program search 32 from UniProt Knowledgebase.Multiple sequence alignment was performed using the CLUSTAL W2. 33

COMPUTATIONAL METHODS
System Preparation.3D structure of the mutants were derived from the model based on the crystal structure of the ligand-free human DPP III (PDB code: 3fvy). 27omplexes between the inhibitor Tyr-Phe-NHOH and yeast DPP III were built using the crystal structure of the C130S yeast DPP III variant (PDB code: 3csk) as a template.In the models all Arg and Lys residues are positively charged (+1) and all Glu and Asp residues are negatively charged (−1), as expected at physiological (experimental) conditions.The initial protonation of the protein was determined by the program WHATIF (with a pH of 7.0), and the protonation of histidines was checked according to their ability to form hydrogen bonds with neighboring amino acid residues.
Ligands were accommodated into the wild-type proteins binding pocket using the combined approach consisting of docking by the AutoDock 4 program 34 and the steered molecular dynamics simulations (for details of the procedure see the reference 27).The complexes with the mutated proteins were built from the wild-type ones by replacing Asn406 either by Gln or Ser.
The protein parametrization was performed within the AMBER ff03 force field of Duan et al., 35 while the ligands parameters were derived using Antechamber in the Amber10 program package (http://ambermd.org/).For the zinc cation, Zn 2+ , the parameters derived in our previous work 36 and modified according to a PDB survey were used. 37mulations.Simulations of the ligand-free proteins and protein-ligand complexes were performed in explicit water (TIP3P water type) using periodic boundary conditions.In order to neutralize the system Na + ions were placed at the protein surface, in vicinity of the charged amino acid residues.The resulting systems, consisted of ≈ 80 000 atoms (≈ 23 000 molecules of water).The electrostatic interactions were calculated using the particlemesh Ewald method. 38,39rior to molecular dynamics simulations, the protein geometry was optimized in three cycles with different constraints.In the first cycle (1500 steps), waters and substrate molecule were relaxed, while protein and zinc atom were constrained using a harmonic potential with a force constant of 133.98 kJ mol −1 Å −2 .In the second (2500 steps), and the third cycles (1500 steps), the same constant was applied to the zinc atom while only the protein backbone was constrained with 133.98 and 41.87 kJ mol −1 Å −2 , respectively.The energy minimization procedure, consisting of 470 steps of steepest descent followed by the conjugate gradient algorithm for the remaining optimization steps, was the same in all three cycles.During the first period of equilibration (100 ps of gentle heating from 0 to 300 K), the NVT ensemble was used.The rest of the MD simulations (water density adjustment and production simulations) were performed at constant temperature and pressure (300 K and 1 atm, the NPT ensemble).The temperature was held constant using Langevin dynamics with a collision frequency of 1 ps −1 and a time step 1 and 2 fs in the case of the human, and yeast DPP III, respectively.Bonds involving hydrogen atoms were constrained using the SHAKE algorithm.The longest were the MD simulations for the DPP III-Arg 2 -2NA(RRNA) complex (30 ns), follows simulations for the yeast DPP III-Tyr-Phe-NHOH(INH) complex (2 × 20 ns), N406Q-RRNA (10 ns), N406S-RRNA (5 ns) and N406S-INH (2.5 ns).
Importantly, the octahedral coordination of the Zn cation was retained in all these complexes, except in the N406Q-RRNA complex.
The root-mean square deviation (RMSD) and rootmean square coordinate fluctuation (RMSF) of each residue in the system were calculated using the "ptraj" module of the Amber program.
MM-PBSA Calculations.The binding free energies were calculated using the Molecular Mechanics Poisson-Boltzmann Surface Area (MM-PBSA) approach 40 as implemented in the AMBER10 program. 41,42MM-PBSA calculations were performed on the trajectory sampled during the last 5 or 2 ns of MD simulations.For this purpose snapshots were extracted from the MD trajectory every 10 ps and the binding free energies were calculated using the equation: The Gibbs free energies were approximated with the sum of conformational energy and the free energy of solvation.The polar component of desolvation free energy was calculated using the Poisson-Boltzmann method and the non-polar component was determined by ΔG nonpol = γSASA + β, in which the solventaccessible surface area (SASA) was calculated with the MolSurf program. 43The surface tension γ and the offset β were set to the standard values of 0.02269 kJ mol −1 Å −2 and 3.85204 kJ mol −1 , respectively.

Conservation of Asn406
Molecular modeling pointed out importance of Asn406 in human DPP III for the substrate binding. 27To investigate the conservation of this asparagine in the metallopeptidase M49 family, we performed multiple alignments of all M49 family members sequences available in the Uni-Prot Knowledgebase.Interestingly, it was found that this Asn is conserved in all 89 eukaryotic sequences, while all of 75 prokaryotic sequences available contained a serine residue at this position.Figure 1 illustrates this finding at reduced dataset consisting of primary structures from 14 selected species.Asparagine 406 was located in the second, starting from the N-terminus, out of 5 highly conserved regions.This result compares well with those reported by us in 2004 when altogether only 14 homologous protein sequences − members of the family M49were deposited in public databases. 21However, up to now, there was no experimental evidence on the function of the conserved region 2 in proteins of DPP III family.

Expression and Purification of Wild-type and Mutated Human DPP III
In addition to conservative substitution by glutamine, asparagine at position 406 in human DPP III was re-placed, using the site-directed mutagenesis, also by the serine, due to the absolute conservation of this amino acid residue in the bacterial sequences (Figure 1).Constructs for the His-tagged DPP III, wild-type and mutants N406Q and N406S were generated and the genes were overexpressed in E. coli BL21(DE3)RIL + as described in the Experimental Methods section.Mutations were confirmed by DNA sequence analysis.The wild-type DPP III and mutated forms were purified by affinity chromatography on Ni-NTA agarose.Figure 2 shows the purification profile for N406S.SDS-PAGE demonstrated that the wild-type and mutant recombinant proteins appeared as bands with similar molecular mass of about 82 kDa.

Substrate Specificity towards Dipeptidyl 2-naphthylamides
Substrate specificity of His-tagged human DPP III and its variants was examined with dipeptidyl 2naphthylamides (Table 2).Apparently the mutants N406Q and N406S similarly to the wild-type enzyme preferred Arg 2 -2NA, while Ala 2 -2NA was hydrolyzed less efficiently (Table 2).This finding is in agreement with the published data on native human erythrocyte and recombinant untagged human DPP III. 25,44

Effect of Metal Ions and Chelating Agent
Since the Zn 2+ ion is natural cofactor of DPPs III, the influence of the zinc salt (zinc acetate) addition on the activity of all three enzyme variants was investigated.Zinc acetate was added in reaction mixture in the concentration range of 0.02 μmol dm −3 to 30 μmol dm −3 .There was no activation observed for any of the enzyme forms (wild-type human DPP III, N406S, N406Q).On the contrary, already 1 μmol dm −3 Zn(Ac) 2 inhibited all three enzymes about 30 %, and in concentration of 30 μmol dm −3 it caused 80 % inhibi-tion of the wild-type DPP III, and about 95 % inhibition of both mutants.
The addition of CoCl 2 (5 to 200 μmol dm −3 ) activated the wild-type DPP III and both enzyme variants, maximal activation (of about 2-fold) being at 50 μmol dm −3 CoCl 2 for all enzyme forms.
Effect of metal chelator ethylenediaminetetraacetic acid (EDTA) was examined by preincubation of the enzyme with EDTA for 20 min at 23 °C and by following the residual activity in function of the chelating agent concentration (Figure 3).A very similar inactivation profile was obtained for all enzyme forms investigated: wild-type human DPP III, wild-type His-tagged human DPP III, His-tagged mutants N406S and N406Q.Preincubation with already 0.5 mmol dm −3 EDTA substantially (40−55 %) decreased the enzyme activity (Figure 3).

Kinetic Characterization of Asparaginyl Mutants
Catalytic properties of His-tagged wild-type and asparaginyl mutants of human DPP III were examined using favorable synthetic substrate Arg 2 -2NA, and competitive inhibitor Tyr-Phe-NHOH.Substrate hydrolysis was measured at pH 8.6 in the presence of 50 μmol dm −3 CoCl 2 .
The Asn406 with Ser substitution did not change significantly K m and k cat for the Arg 2 -2NA hydrolysis,  neither did K i for Tyr-Phe-NHOH, compared to the wild-type (Table 3).However, large decrease in both the binding affinity of the competitive inhibitor (156-fold higher K i value) and in the catalytic efficiency towards Arg 2 -2NA (100-fold smaller ratio k cat /K m , mostly due to the decrease in k cat ) was observed for the N406Q mutant (Table 3).The dissociation constant of Tyr-Phe-NHOH, K i , determined for the wild-type human DPP III-His 6 is in very good agreement with that determined for the wild-type human DPP III without His 6 -tag. 25The k cat value for the hydrolysis of Arg 2 -2NA of His-tagged human DPP III does not differ significantly from the k cat of its untagged counterpart (21.8 ± 2.8 s −1 ), while K m is somewhat lower (3.37 μmol dm −3 versus 11.6 μmol dm −3 ). 25 Apparently, impaired catalytic efficiency of N406Q mutant and its dramatically lowered inhibitor binding affinity present experimental evidences of the Asn406 importance in human DPP III ligand binding and hydrolysis.Since the K m is not a simple enzymesubstrate binding constant, but represents the dissociation constants for all enzyme-bound substrate species, it is not surprising that the change of K m value does not correlate with the inhibitor binding affinity, K i .Interestingly, the replacement of Asn406 with Ser did not influence kinetic properties of human DPP III (Table 3).
Based on the results of the kinetic investigations it is evident that the substitution of conserved Asn with either Gln or Ser did not change the enzyme preference for the synthetic substrates dipeptidyl 2-naphthylamides (Table 2).
The catalytic efficiency reduction induced by the N406Q mutation is comparable with that induced by the W300L mutation (100-fold versus 78-fold decrease in k cat /K m ) as shown by us recently. 25However, contribution of Asn406 to the competitive inhibitor (Tyr-Phe-NHOH) binding seems to be much more pronounced than that of the conserved Trp300, since for W300L the K i was "only" 29-fold higher in comparison with the value determined for the wild-type. 25zymatic hydrolysis of Arg 2 -2NA was measured also without the addition of metal activator (CoCl 2 ) to the reaction mixture.Again, large decrease in k cat value was observed for the N406 mutant only, compared to the wild-type His-tagged human DPP III (0.06 s −1 versus 6.0 s −1 ), while the K m value was not changed significantly (not shown).

MD Simulations of Human DPP III-substrate Complexes
In order to explain the observed differences in the catalytic properties of two asparaginyl mutants and to understand better the Asn406 role in ligand binding and in the substrate catalysis we performed molecular dynamics simulations for N406S and N406Q mutants ligated with Arg-Arg-2NA (RRNA) substrate.Besides a short simulation was performed for the complex between the N406Q mutant and the Tyr-Phe-NHOH inhibitor.Complexes between RRNA and N406Q and N406S were simulated for 10 and 5 ns, respectively, while the N406Q -Tyr-Phe-NHOH complex was simulated for 2.5 ns at room temperature.The simulations resulted with an interesting finding on the zinc cation coordination in the complex with the less active mutant (N406Q).While in the N406S mutant, whose activity was unchanged compared to the wild-type (Table 3), the octahedral coordination of the zinc ion, determined as stable in the wild-type DPP III and its complexes, 27 was preserved during the MD simulations, in the N406Q-RRNA complex zinc was during MD simulations mostly hepta-coordinated (see Figure 4).
In addition, two hydrogen bonds, between RRNA and Glu508 and Ser504, were lost in the N406Qsubstrate complex (Figure 4b).In the N406Q − inhibitor complex, the weak van der Waals (v.d.W.) interaction with Ser504 was present only during the last 0.5 ns of MD simulations, while the CH-π interaction between the inhibitor and Ser504 has not been established (Fig- ure 4d).K i values were determined with Arg-Arg-2NA as substrate at pH = 8.0, using Dixon plots.
Analysis of the overall proteins flexibility showed the largest discrepancies in the calculated delta RMSF values for residues constituting the protein "lower" domain.In the complexes between RRNA and the mutated proteins the "lower" domain region, situated just below the enzyme active site, a part of which is mutated Asn406, showed increased flexibility, while the rest of the domain seemed to be less flexible.In the N406Q-RRNA complex the beta sheet from the "lower" domain (residues 388−408), just below the zinc binding site, slightly moved in the direction of the upper domain.In addition, delta RMSF values revealed increased relative mobility of the helix connecting the lower and upper domain in mutated proteins, as well as increased rigidity of the "upper" domain in the N406Q mutant protein (Figure 5).
Another interesting matter noticed in the complexes with the N406Q mutant is orientation of the catalytically important His568.Actually, during MD simulations His568 is mostly in position to form a hydrogen bond with Glu512 and away from Glu508 (not shown).Apparently its suggested role in the peptide hydrolysis seems to be hindered in the mutant, and this is in agreement with the experimental findings.On the other hand, according to the MM-PBSA calculations, the substrate (RRNA) binding affinities towards the wild-type protein and mutants are practically the same, with energy difference being within the error of calculations.

MD Simulation of Yeast DPP III-inhibitor Complexes
Asn415 in the yeast DPP III corresponds to Asn406 in the human DPP III.In order to investigate if this structurally equivalent residue plays a role in the ligand binding to the yeast orthologue, we performed MD simulations for the yeast DPP III Tyr-Phe-NHOH complexes.
The inhibitor, Tyr-Phe-NHOH, was accommodated into the active site of the yeast DPP III in two different orientations.In one (complex 1) it was bound to the Zn ion by the carbonyl oxygen of the first peptide bond from the N terminus, and in the other (complex 2) by the carbonyl oxygen of the second peptide bond from the N terminus.
In the case of complex 1, Asn415 has v.d.W. interactions with the substrate N terminus, and electrostatic and CH-π interactions with its Tyr (Table 4), while in the complex 2, Asn415 interacts with both amino acid residues of the inhibitor, Tyr and Phe via v.d.W and electrostatic interactions (Table 5).
The binding free energy evaluated using MM-PBSA method was lower for complex 1 (−102.8± 14.1 kJ mol −1 ), than for complex 2 (−46.1 ± 14.3 kJ mol −1 ).For the inhibitor binding to the human DPP III, the evaluated binding free energy was −118.7 ± 21.6 kJ mol −1 .In this complex inhibitor is bound to the zinc cation by the carbonyl oxygen of the second peptide bond from the N terminus. 27nterestingly, MD simulations of human enzymeligand complexes indicated the involvement of Asn406 in the binding of the residue in P2 position (first Arg and Tyr in RRNA and Tyr-Phe-NHOH, respectively) only. 27Considering relatively low homology between the human and yeast DPP III protein (36 % sequence The delta RMSF values of the protein backbone (RMSF − <RMSF>) in wild-type and mutant's complexes with synthetic substrate for the last 5 ns and 2 ns of MD simulations, respectively.Flexible loop and α-helix connecting "lower" and "upper" domain are made by residues 337 to 374.Amino acids 424 to 666 constitute the enzyme's "upper" domain, while rest constitute the "lower" domain.identity), the different role of Asn415 in the ligand stabilization observed in the yeast complex is not surprising.Different catalytic efficiency (k cat /K m ) in the RRNA hydrolysis, the human enzyme is two orders of magnitude more efficient than the yeast DPP III, also suggests that their active sites are somewhat different.

CONCLUSIONS
Recent computational results indicated that asparagine residue in position 406 of human dipeptidyl peptidase III (737 amino acids) is a constituent of its substrate binding subsite S2.
In order to investigate the functional role of this, in all eukaryotic M49 peptidases conserved Asn, experimentally, we prepared recombinant human DPPs III (the wild-type and two mutants, N406S and N406Q), heterologously expressed them with C-terminal His 6tag, purified and kinetically characterized.
Kinetic results obtained with the N406Q mutant support the Asn406 involvement in the ligand (competitive Tyr-Phe-NHOH peptide inhibitor) binding and substrate catalysis.However, the substitution of Asn406 with Ser did not change K m and k cat for the Arg-Arg-2NA hydrolysis, and K i for Tyr-Phe-NHOH.Molecular dynamics simulations performed for complexes between different enzyme forms (wild-type, N406S and N406Q) and the substrate Arg-Arg-2NA revealed the difference in the active-site zinc coordination only in the complex with the N406Q mutant.In this complex two hydrogen bonds stabilizing the substrate in the complex with the wild-type protein were lost, and the orientation of catalytically important residue His568 changed.This could impair ligand binding and catalytic efficiency of N406Q enzyme form.
In yeast DPP III, Asn415 corresponds to Asn406 in human enzyme.The importance of Asn415 for ligand binding was confirmed by MD simulations of yeast DPP III-ligand(inhibitor) complexes.

Figure 3 .
Figure3.Effect of EDTA on the wild-type (His-tagged and untagged) and asparaginyl mutants (His-tagged) of recombinant human DPP III.The enzyme was preincubated with EDTA for 20 min at 23 °C, and the residual activity was determined at 37 °C and at pH 8.6 as described in Experimental Methods, but without the addition of metal salt (CoCl 2 ).The enzyme concentrations were: 5.8 × 10 −10 mol dm −3 wild-type without the His-tag, 6.0 × 10 −10 mol dm −3 wild-type Histagged, 1 × 10 −10 mol dm −3 N406S His-tagged, and 6.7 × 10 −8 mol dm −3 N406Q His-tagged.The wild-type human DPP III without the His-tag was purified as described.25

( a )
The kinetic parameters K m and k cat were determined from the initial reaction rates at 25 °C and at pH 8.6 with the Arg-Arg-2NA concentrations from 2.5 to 60 μmol dm −3 , in the presence of 50 μmol dm −3 CoCl 2, using a Hanes plot.(b)

Figure 4 .
Figure 4. Coordination of the central zinc ion after: 10 ns of MD simulation for RRNA in complex with a) WT and b) N406Q mutated human DPP III, 5 ns of MD simulation for c) N406S-RRNA complex, and 2.5 ns of MD simulation for d) N406Q mutant in complex with Tyr-Phe-NHOH.The zinc cation is shown as a magenta sphere, and ligands are shown in ball and stick representation.Magenta dashed lines represent zinc coordination with electron donors, and dark blue dashed lines H-bonds that these electron donors make.

Figure 5 .
Figure 5.The delta RMSF values of the protein backbone (RMSF − <RMSF>) in wild-type and mutant's complexes with synthetic substrate for the last 5 ns and 2 ns of MD simulations, respectively.Flexible loop and α-helix connecting "lower" and "upper" domain are made by residues 337 to 374.Amino acids 424 to 666 constitute the enzyme's "upper" domain, while rest constitute the "lower" domain.

Table 3 .
Kinetic characterization (means S.E.for three separate determinations) of wild-type human DPP III and Asn406 mutants(a,b)

Table 4 .
Interactions between the wild-type yeast DPP III and inhibitor Tyr-Phe-NHOH (in complex 1)

Table 5 .
Interactions between the wild-type yeast DPP III and inhibitor Tyr-Phe-NHOH (in complex 2)