Mutagenesis study and enzyme kinetics of V. alginolyticus PepD

3.3-1 Mutational analysis on metal-binding and catalytic residues of V. alginolytic PepD

Previously, His⁸⁰, Asp¹¹⁹, Glu¹⁵⁰, Asp¹⁷³, and His⁴⁶¹ were described as being putatively involved in metal binding in PepD⁷⁶. We individually mutated each of these residues using an alanine-scanning mutagenesis strategy, and characterized the expressed proteins with CD spectrometry (Fig. 3.10). Each of the mutated proteins was produced in similar quantities from the expression system and exhibited homologous CD spectra, indicating that the overall structure of the mutated enzymes was not affected by the manipulation of the amino acid sequence.

56

Nevertheless, we could detect no activity for any of the mutants. This suggested that each of these residues played an essential role in PepD enzymatic activity (Tables 3.1 and 3.2). In a parallel experimental procedure, Asp¹¹⁹ was substituted with Glu, Met, Leu, Ile, Arg, Phe, Ala, Ser, Thr, Cys, Pro or Asn, Glu¹⁵⁰ was replaced with Arg or His, and Asp¹⁷³ was mutated to a Glu residue. As expected, substitution of Asp¹¹⁹ with other proteinogenic amino acid residues completely abolished the enzymatic activity. On the other hand, substitution of Glu¹⁵⁰ with Asp led to the retention of ~60% of the maximal hydrolytic activity of the wild-type enzyme, whereas substitution of Glu¹⁵⁰ with Arg or His completely abolished enzymatic activity. Substitution of Asp¹⁷³ with Glu also completely abolished the enzymatic activity.

Figure 3.10. CD spectra of V. alginolyticus PepD wild-type and various mutants.

57

Next, we subjected the Asp⁸² and Glu¹⁴⁹ residues to site-directed mutagenesis, to evaluate their putative roles in PepD catalysis. Asp⁸² was substituted for either Gly, Val, Phe, Tyr, His, or Glu, whereas the Glu¹⁴⁹ was replaced with either Gly, Ala, Ile, Ser, His, Trp, or Asp. Again, no activity was detected for any of the Asp⁸² mutants. Substituting Glu¹⁴⁹ with Gly, Ala, Ile, Ser, His, or Trp also resulted in the abolishment of enzymatic activity, with the exception of the Asp mutant being able to retain ~55% of the wild-type activity (Tables 3.1 and 3.2). It is interesting to note that replacement of Glu¹⁴⁹ or Glu¹⁵⁰ with aspartic acid led to partial retention of enzymatic activity, despite the residue being only one carbon shorter and having the same negative charge. We speculate that shortening the amino acid side chain in this particular position may allow for its acidic group to move away from an optimum position, consequently promoting activation of the catalytic water molecule, or perhaps the replacement of Glu with Asp at this position may partially affect the metal ligand-binding affinity and impair subsequent activation of the catalytic water for substrate-enzyme tetrahedral intermediate formation. Either or these processes may have ultimately resulted in partial loss of the enzymatic activity.

To investigate whether Asn¹²⁰ of the cis-peptide is involved in catalysis or protein folding/stabilization, Asn¹²⁰ was substituted with Ala and its enzymatic activity was examined. As expected, no activity was detected for the PepD^N120A mutant. In addition, the CD spectra of the PepD wild-type and PepD^N120A mutant proteins presented almost the same profile in the range of 198–250 nm (Fig. 3.10) implying that the PepD^N120A mutant protein was not perturbed in either its stability or folding properties. These results indicate that Asn¹²⁰ plays an essential role in the enzyme reaction.

58

Table 3.1: Enzyme kinetics of V. alginolyticus PepD wild-type and mutant proteins.

Kinetic parameters are for the hydrolysis of L-carnosine at 37 °C and pH 7.0.

a

ND, not detected

Table 3.2: Enzymatic studies of V. alginolyticus PepD wild-type and mutant proteins. Residual activity was determined for the hydrolysis of L-carnosine at 37 °C and pH 7.0.

a

ND, not detected

59

3.3-2 Mutational analysis on probable substrate C-terminal binding residues within the lid domain of V. alginolyticus PepD

Jozic et al. have previously identified three residues, Asn²¹⁷, His²⁶⁹, and Arg³⁵⁰, within the lid domain of PepV that are putatively involved in the substrate C-terminal and/or transition state binding through hydrogen bonding²⁶. Due to the different topology of the β-sheet order, a simple primary sequence alignment was not able to identify the corresponding residues in the lid domain of PepD, except for Arg³⁶⁹, which aligned with Arg³⁵⁰ of PepV. This residue also superimposed with Arg³²⁴, Arg²⁸⁰, and Arg²⁷⁶ in the small domains of the dimeric CPG2, PepT, and hACy1, respectively. We then used structure-based sequence alignment to identify the other equivalent residues in PepD. A structure superimposition but inversed sequence order, in which the Asn²¹⁷ and His²⁶⁹ residues of PepV superimposed with the Asn²⁶⁰ and His²¹⁹ residues of PepD, was noticed. Asn²⁶⁰ is conserved among PepV, CPG2, βAS, and PepT, but is substituted by Thr in human CN1 and mouse CN2 as well as by Tyr in PepT. Remarkably, the His²¹⁹, Asn²⁶⁰, and Arg³⁶⁹ residues are located on the same side of the lid domain for both PepD dimers, but the corresponding residues are located on the opposite side of the lid domain of the same monomer for CPG2 and related dimeric proteins (Fig. 3.7). Therefore, in CPG2, the Arg²⁸⁸ from the lid domain of one monomer interacts with the His²²⁹ and Asn²⁷⁵ from the lid domain of the other monomer; in contrast, the Arg³⁶⁹ from the lid domain of the PepD monomer interacts with the Asn²⁶⁰ and His²¹⁹ from the lid domain of the same monomer.

We also performed site-directed mutagenesis experiments to test the roles of these equivalent lid domain residues. The mutated PepD proteins were produced in a procedure similar to that of the wild-type PepD. All mutants exhibited similar purification characteristics and the same electrophoretic mobility as the wild-type

60

enzyme in SDS-PAGE. Although each of the mutations produced similar quantities of the protein, the Arg³⁶⁹ to Ala mutation resulted in complete loss of the enzymatic activity for hydrolyzing L-carnosine, whereas the Asn²⁶⁰ to Ala mutation decreased the catalytic activity to almost half. Interestingly, the His²¹⁹ to Ala mutation did not affect the enzymatic activity significantly, yielding only a slight increase in activity of ~10%

as compared with the wild-type PepD. In PepV, the Arg³⁵⁰ was located near the C terminus of the bound inhibitor (2.7 Å) but appeared to be too far away from the zinc ions (~8 Å), indicating a role in substrate binding but not in catalysis. The replacement of Arg with Ala might disrupt the hydrogen bond network between the Arg³⁶⁹ side chain and Asn²⁶⁰ Nδ with the carboxylate group of the substrate. In the case of PepV, Jozic et al. have argued that domain flexibility is required to allow substrate access. Moreover, the bad diffraction and high mosaicity observed in the inhibitor-free PepV crystal have been attributed to conformational variability between open and closed states. A significant opening of the protein conformation would clearly benefit access of the peptides to the active site cavity. It is conceivable that even the whole lid domain might move away from its site to allow for easier substrate access and product egress.

Therefore, although the Arg³⁶⁹ guanidinium side chain and the Asn²⁶⁰ Nδ within the active site of PepD are both ~16 Å away from the zinc ion, a conformational change between the open and closed states might have contributed to the movement of both Arg³⁶⁹ and Asn²⁶⁰ upon substrate binding and subsequent transition state stabilization.

Furthermore, binding of the His²¹⁹ in PepD to the substrate likely persists during the conformational change between the open and closed states and contributes to transition-state stabilization through an electrostatic interaction between His²¹⁹ and the free carboxyl group of the ligand, as shown in the PepV-inhibitor complex (Fig. 3.9).

61

3.3-3 Mutational analysis on dimeric interface of V. alginolyticus PepD

Based on the crystal structure, PepD and PepV revealed similar architecture, except that PepD is a dimer and PepV is a monomer. From the structural illustration of PepD, the dimeric interface appears to be formed by hydrogen bonds through Ser³⁷⁴ and Ser³⁸⁵ of one subunit to Glu²⁹⁴ and Asn³²⁹ of the other subunit, respectively (Fig. 3.8).

We then performed site-directed mutagenesis experiments on these residues to investigate the putative dimeric interface interactions. According to the results obtained from analytical sedimentation velocity ultracentrifugation, the molecular mass of the PepDS374A/S385A double mutant was determined to be 54.4 ± 0.02 kDa, whereas the PepD wild-type was 96.8 ± 0.11 kDa. Size exclusion chromatography of PepD^WT and the PepDS374A/S385A double mutant also revealed the corresponding sizes of ~103.7 and 50.6 kDa, respectively (Fig. 3.11). These results suggested that the PepDS374A/S385A mutant existed as a monomer in solution. Interestingly, the PepDS374A/S385A mutant exhibited

~130% activity of the wild-type, indicating independent function for the monomer (Table. 3.2). However, the exact reason for forming the dimeric structure remains unclear, but a physiochemical or regulatory function of PepD may be involved.

62

Figure 3.11. Molecular masses determination of PepD^WT and PepDS374A/S385A mutant.

Analytical ultracentrifugation determination of (A) PepD^WT and (B) PepDS374A/S385A

mutant, showing the calculated molecular weight from sedimentation coefficient (s) of approximately 96817.977 ± 105.3 g/mol and 54452.62 ± 16.5 g/mol, respectively. (C) Chromatographic separation and calibration curve for the standard proteins, PepD^WT and PepDS374A/S385A on Superdex 200 10/300 GL column (GE Healthcare^TM). In the calibration curve, the molecular mass of PepD^WT (green dot) and PepDS174A/S385A (red dot) were determined to be 103.7 and 50.6 kDa, respectively.

63