Site-directed mutagenesis analysis of Vibrio alginolyticus pepD

Chapter 3 Results

3.5 Site-directed mutagenesis analysis of Vibrio alginolyticus pepD

is analysis was

was notable that the conserved residue Asp119 of PepD was equal to the adjacent resi

In order to identify the putative active site of PepD, site-directed mutagenes

performed to investigate the essential amino acids. Recent studies reported that the active site residues of enzymes in M20 family were almost conserved.^{55, 58} Pep V, the first dinuclear dipeptidase with carnosine-hydrolyzing enzymatic activity protein in M20 family has been crystallized in 2002. PepD showed 20.9% identity and 34.3% similarity with PepV based on the sequence alignment employing. Sequence alignment between PepD and PepV also revealed the proposed active site. Surprisingly, the active site residues of PepV were almost conserved in PepD. These residues, including His80, Asp119, Glu150, Asp173, and His461 were expected for the metal binding, whereas Asp82 and Glu149 were expected for the catalysis (Fig. 8).

due Asp 120 of PepV that is the adjacent residue next to its metal binding residue Asp119.

It is common that aspartic acid residue exhibited the metal binding role extensively at the active site of the enzymes in M20 family. The PepV peptide group between the bridging Asp119 and the adjacent residue Asp120 exhibited a cis-conformation to affect the binding of the metal, whereas the Asp 119 from PepD might be considered to associate with two zinc ions simultaneously. The cis-conformation is thought to be necessary to force the important zinc bridging carboxylate into the correct geometry as described in 1.7. Therefore, the residue Asp119 of PepD may play a more important role on metal binding hence involve in the enzymatic activity. In the study Asp119 and another proposed catalytic residue Glu149 were initially investigated by site-directed mutagenesis.

Fig. 8 Multiple sequence alignment with PepD and PepV

are marked in black and gray,

p119,Glu150, Asp173, His461

The desired mutants were generated by QuickChange site-directed mutagenesis kit as described in 2.9 and the mutant plasm

Identical and conserved amino acids between the sequences

respectively. Dashed lines indicate the gaps introduced for better alignment.

▼ Proposed active site residues : Asp82 ,Glu149

＊ Proposed metal ion binding residues : His80, As

ids were transformed into E. coli. BL21(DE3)pLysS to express the mutant proteins. Following the same purification experimental procedure of V.

alginolyticus wild-type PepD, the mutant PepD proteins were carried out with 20 mM Tris-HCl pH 6.8 buffer containing 150 mM imidazole by Ni-NTA column chromtography.

The purified wild-type and mutant PepD proteins showed the same molecular weight about 55 kDa on SDS-PAGE (Fig. 9~10.).

-type and mutant proteins of D119.

ild type; Lane 2:PepD D119E mutant; Lane 19L mutant; Lane 5:PepD D119I mutant; Lane 19F mutant; Lane 8:PepD D119A; Lane 9:PepD Lane 11:PepD D119C mutant; Lane 12:PepD

Fig. 10 SDS-PAGE of E149.

Lane M:LMW protein marker; Lane 1:PepD wild type (WT); Lane 2:PepD E149A mutant;

Lane 3:PepD E149D mutant; Lane 4:PepD E149G mutant; Lane 5:PepD E149S mutant; Lane 6:PepD E149Q mutant; Lane 7:PepD E149H mutant; Lane 8:PepD E149R mutant

Fig. 9 SDS-PAGE (12%) of purified wild Lane M:LMW protein marker; Lane 1:PepD w 3:PepD D119M mutant; Lane 4:PepD D1 6:PepD D119 R mutant; Lane 7:PepD D1 D119S mutant; Lane 10:PepD D119T mutant;

D119P mutant; Lane 13:PepD D119N mutant

(12%) of purified wild-type and mutant proteins

M 1 2 3 4 5 6 7 M 8 9 10 11 12 13

30 97 66 45

M 1 2 3 4 5 6 7 8 97

66 45

In E. coli, PepD was indicated as a homodimer in native sate.²⁹ To observe the native form of PepD . alginolyticus, the wild-type and mutant PepD derivatives were analyzed with 7.5% Native-PAGE analysis (Fig. 11) A major band with molecular mass near 66 kDa of the protein marker was observed. Interestingly, several weak bands were also examined on the Native-PAGE of the wild-type and mutant PepD. Therefore, it was proposed that PepD in V. alginolyticus might exist in many forms in the native state. The major band with molecular weight about 66 kDa might be the monomer that PepD tended to form in its native state. The r form. To

estern blotting (Fig. 13). However, the calculated molecular weight from sedimentation coef

in V

minor weak band with molecular weight near 140 kDa might be the homodime

ensure the native form of PepD in V. alginolyticus, the Westeewrn blotting, that examining a clear band on the film was carried out using anti-PepD mAb (Fig. 12). Besides, Analytical Ultracentrifugation (AUC) was also used to confirm the result of Native-PAGE analysis and W

ficient (s) indicated that PepD preferred to form homodimer in its native state. The calculated molecular weight of denatured PepD protein was as a control comparing to the molecular weight of wild-type PepD protein (Fig. 14).

M 1 2 3 4 5 6

Fig. 11 Nativ-PAGE (10%) analysis of purified PepD wild-type and mutant proteins Lane M:HMW Native protein marker; Lane1: PepD wild-type(20 μL); Lane 2:PepD D119E mutant (20 μL); Lane 3:PepD E149D mutant (20 μL); Lane 4:PepD wild-type (40 μL); Lane

5:PepD D119E mutant (40 μL); Lane 6:PepD E149D mutant (40 μL) 232

140

M 1 2 3 4 5 6

0 100000 200000 300000 400000 500000 600000 700000 800000 0.000000

Fig. 12 Western blotting analysis of purified PepD wild-type and mutant proteins

Analytical ultracentrifugation determination of PepD protein

ecular 440

232 140

Lane M:HMW Native protein marker; Lane 1:PepD wild-type; Lane 2:PepD D119E mutant;

Lane 3:PepD E149D mutant; Lane 4:PepD wild-type; Lane 5:PepD D119E mutant; Lane 6:PepD E149D mutant

Fig. 13

The molecular weight of V. alginolyticus PepD is 53548.8 g/mol. The calculated mol weight from sedimentation coefficient (s) is about 96817.977 g/mol.

0 200000 400000 600000 800000 1000000 1200000 0.000000

0.000001 0.000002 0.000003 0.000004 0.000005 0.000006 0.000007

Data: Data1_B Model: Gauss

Chi^2/DoF = 4.1744E-15 R^2 = 0.99639

y0 0 ±0

xc 3783 4

5.70551 ±69.20846 w 190 9.24454 ±136.99134 A 0.1428 ±0.0009

PepD protein

The calculated molecular weight of denatured PepD protein from sedimentation coefficient (s) is about 37835.71 g/mol.

Fig. 14 Analytical ultracentrifugation determination of denatured

3.6 Enzyme kinetic of the mutant PepD

The enzyme activity of the mutant PepD was performed with hydrolyzing L-carnosine in e same experimental process of the wild-type PepD, as described in 2.6. Compared to the ild-type PepD, no apparent activity could be detected among almost all mutanted-PepD xcept the E149D mutant (Fig. 15).

WT E149D E149I E149G E149H E149A E149S E149W

Relative activi

Fig. 15 Enzymatic activities of wild-type and mutant PepD on L-carnosine. Purified wild-type and mutant PepD proteins were subjected to the activity assay on L-carnosine as a substrate. The wild-type activity was defined as 100%

Moreover, the enzyme kinetic of the mutants to determine the Vmax, Km and Kcat values, and compared to that of the wild-type PepD. The Vmax and Km values of the E149D mutant (2 μg, 0.186 μM) for L-carnosine calculated from the respective Lineweaver-Burk plot were 1.1 μM/min and 0.53 mM (Fig. 16).

hich catalyzed the hydrolysis of L-carnosine in 50 mM Tris-HCl, pH 6.8 at 37 ºC. (b) Lineweaver-Burk plot calculated from the respective Michaelis-Menten plot.

Fig. 16 Enzyme kinetics of wild-type and mutant PepD (a) Michaelis-Menten plot for wild-type and E149D mutant w

Therefore, the turnover number (kcat, kcat = Vmax/[E]T) of E149D for L-carnosine in 50 mM Tris-HCl, pH 6.8 at 37 ºC is 5.9 min^-1 and the catalytic efficiency (kcat/Km) is 0.186 mM^-1s^-1. Moreover, no activity at all could be detected for the D119E mutant (Table 6).

Table 6. Kinetic Parameters for the hydrolysis of L-carnosine at 37 ºC and pH 6.8 of wild-type and mutant V. alginolyticus PepD

PepD variant kcat (min^-1) Km (mM) kcat/ Km (mM^-1s^-1)

Wild-type 8.6 0.36 0.398

E149D 5.9 0.58 0.186

D11 ND ND ND

3.7 The secondary structure of Vibrio alginolyticus PepD

There is a common problem that the mutants created by using site-directed mutagenesis techniques might cause global conformational changes that inactivate the protein. Circular dichroism (CD) spectrum analysis can give information on the secondary structure content of protein. The α-helix, β-sheet, and γ-turn are three main types of secondary structure in proteins. The different types of regular secondary structure in proteins would give rise to

pD spectra were similar (Fig. 17).

characteristic CD spectra in the far UV.⁶³ CD spectroscopy was performed on the purified PepD WT and mutant proteins to prove that the loss or decreasing activity of the mutants were not due to the change of secondary structure of protein. The far-UV CD spectra of the wild-type and mutant Pe

-1500 -1000

(10 -500

mdeg

500

[θ]MRE2 l ¹⁰⁰⁰

1500 e-1 ) cm2 dmo

WT D119E E149D

180 200 220 240 260 280 300

Wavelength (nm)

V. alginolyticus PepD

α-helix (%) β-sheet (%) random coil (%)

WT 31 11 58

D119E 31 10 58

E149D 30 14 57

(http://www.embl-heidelberg.de/~andrade/k2d/) Fig. 17 The CD spectra of Vibrio alginolyticus PepD wild-type and mutant proteins.

The predict percentages of secondary structure from CD spectra also suggested that the secondary structure of wild-type and mutant PepD were almost the same (Table 7).

Accordingly, PepD mutants did not produce the structural misfolding dramatically and the influence on its activity might come from the mutated-inducing enzymatic activity change.

Table 7. The secondary structure content of wild-type and mutant

3.8 Structure features of Vibrio alginolyticus PepD

To predict the active site residues of V. alginolyticus PepD, the computational analysis was also carried out through homology modeling analysis. The PepD model was obtained with L. delbrueckii PepV as the template, and the generated PepD structure is quite similar to PepV structure (Fig. 18).

. The PepD catalytic domain has a fold similar ons at llow) including e quite similar.

ology modeling, seven residues including His80, Asp1

ought to be involved in the active site of V. alginolyticus PepD crucial for associating with inc ions , and Asp82 and Glu149, two catalytic residues. Moreover, the superimposition of Fig. 18 Three-dimensional ribbon of the crystal structure of PepV(right) and the generated PepD(left) model based on PepV

to the dinuclear carnosine-hydrolyzing enzyme PepV. The structure predicted two zinc i the catalytic pocket center which were held by five metal binding residues (ye

the adjacent residue Asp119 (light green). The residues for catalytic (blue) wer

Apparently, based on the results of sequence alignment, activity assay, and hom 19, Glu150, Asp119, Asp173, His461 were th

PepV active site residues with PepD revealed that the predicted active site residues of PepD were almost equivalent to that of PepV (Fig. 19).

H461 (PepD)

ig. 19 Stereo view of PepD (blue) superimposed with the active site residues of PepV.

The

gray sticks are residues of PepV, and the orange sticks (metal binding) and the blue sticks (catalytic) are residues of PepD. The predicted metal binding residues of PepD are almost equivalent to that of PepV. The yellow stick is the phosphinic inhibitor AspΨ [PO2CH2]AlaOH of PepV.

It was considered that the active site residues of enzymes in M20 family were almost conserved. Carboxypeptidase G2 (CPG2), the first dinuclear dipeptidase with available crystal structure in M20 family, was proven to cleave the the C-terminal Glu from folic acid and folate analogs.⁵⁸ The structure homology models showed that the catalytic domain of PepV has a similar folding to the CPG2 and the active sites residues of PepV were fully conserved in CPG2.⁵⁵ In the other hand, the sequence of PepD showed 19.3% identity and 34.2%

similarity with CPG2. To determine the putative catalytic domain of PepD, the sequence alignment and superimposition of CPG2 active site residues with PepD was also performed.

The result revealed that the predicted active site residues of PepD were almost equivalent to that of PepV and CPG2 (Fig. 20, 21), only the residue Asp119 of PepD displayed a notable difference in this model. However, loss of enzymatic activity of PepD D119 mutant indicated that this residue might play an important role for the PepD mediated cleavaging reaction.

D119(PepD)

Fig. 20 Local view of PepD superimposed with the active site residues of PepV and CPG2. The residues were respectively colored in gray, yellow and blue for PepD, PepV and CPG2. The active site residues were almost equivalent. The zinc binding residue Asp173 in PepD and equivalent aspartic acid in PepV was substituted by glutamic acid in CPG2.

Proposed active site residues : Asp82 ,Glu149

Proposed metal ion binding residues : His80, Asp119,Glu150, Asp173, His461 Fig. 21 Multiple sequence alignment with CPG , PepV and PepD 2

3.9 Crystallization of Vibrio alginolyticus PepD

The V. alginolyticus PepD crystals were obtained by using hanging drop technique as described in 2.12. Hanging drops were formed by mixing 1 μL of 10 mg/mL enzyme solution with 1 μL crystallization reagent. The crystals were yielded at 20 ºC in two weeks under two condition: Crystallization reagent was composed of (a) 20% PEG-4000, 10% isopropanol, 0.1 M Na-HEPES, pH 7.5 (b) 28% PEG-400, 0.1 M Na-HEPES, 0.2 M CaCl2, pH 7.5 (Fig.

22). The crystals diffraction was 2.7 Å at best, but the quality of this data collection was not useful for phasing. Therefore, further modification of the precipitation condition should be undertaken and hope to obtain the high-resolution crystals for structure determination in the future.

(b) (a)

Fig. 22 The V. alginolyticus PepD crystals were grown at 20 ºC in two condition: (a) 20%

8% PEG-400, 0.1 M a-HEPES, 0.2 M CaCl2, pH 7.5

PEG-4000, 10% isopropanol, 0.1 M Na-HEPES, pH 7.5 (b) 2 N

Chapter 4 Discussion and conclusion

In this study, V. alginolyticus PepD which identified as a member of metallopeptidase M20 family and considered as an aminoacylhistidine dipeptidase was being investigated. The recombinant PepD of V. alginolyticus with a His-tag on N inus was successfully expressed and purified with high purity. Based on the SDS-PAGE analysis, the molecular mass of PepD is about 55 kDa near to the calculated molecular mass of 53.6 kDa. and is also similar to that of previously identified E. coli PepD.²⁹

-term

which was similar to that exhibited by all of the known PepD. It could hydrolyze the unique dipeptide L-carnosine (β-Ala-L-His), which has potential neuroprotective function in brain and could act as an antioxidant or antiglycation agent. The substrate specificity of PepD in V. alginolyticus was also identified in this study. It could degrade a large number of Xaa-His dipeptides besides brain-specific dipeptide like GABA-His. The amino group in α or β position of N-terminus residue did not affect the recognition and hydrolysis of dipeptide. However, with the more tendency for α-Ala-L-His than L-carnosine as substrate was first identified in bacterial PepD, whereas the same result could also be observed in human cytosolic nonspecific dipeptidase (CN2).⁴² Several Xaa-His dipeptides including Val-His, Leu-His, Tyr-His, Ile-His, His-His and Ser-His were superior over carnosine as substrates for degradation and the enzyme also could hydrolyze Gly-His ith good activity but had no apparent activity on β-Asp-L-His. Therefore, we preliminary

part or C-terminal position were not degraded, indicating at V. alginolyticus PepD is a Xaa-His dipeptidase in activity. The study of substrate

ecificity on Xaa-His dipeptides was first carried out in bacterial aminoacylhistidine

The V. alginolyticus PepD enzymatic activity was performed on

L-carnosine-hydrolyzing,

assumed that this enzymatic activity on Xaa-His dipeptides is dependent on the charge of Xaa amino acid. The His-Xaa dipeptides including His-Ile or His-Val, as well as tripeptides containing histidine in the central

th sp

dipeptidase, but had already been identified in mammals.

-1 -1

-1 -1), PepD catalyzes at a relative low efficiency. However, the Km value of V. alginolyticus PepD was lower than PepD of

Escherichia coli ³⁰

alginolyticus

The kinetic values including kcat or kcat/Km, were compared with the mammal aminoacylhistidine dipeptidases while lacking of the related reports for the bacterial PepD and PepV. The Km value and catalytic efficiency (kcat/Km) of V. alginolyticus PepD for

L-carnosine were 0.36 mM and 0.398 mM s , respectively. As compared with that of human carnosinase (CN1) (Km 1.2 mM and kcat/Km 8.6 mM s

K-12 (2 to 5 mM) indicated a relatively higher interaction of V.

PepD with its substrates. Based on the result of higher hydrolysis rate on α-Ala-L-His than β-Ala-L-His as the substrate, PepD could have other more suitable dipeptide substrate for degradation or even another totally different enzymatic activity.

Moreover, the inhibitiory effect of bestatin on L-carnosine hydrolysis by PepD will be investigated.

According to the result of sequence alignment between PepV and PepD, the active site residues were almost conserved. We predicted that there are five putative metal binding residues, His80, Asp119, Glu 150, Asp173, and His461, and two catalytic residues, Asp82 and Glu149 in the enzymatic active site cavity. The conservation of the active site residues suggests that the hydrolytic mechanism of PepD and PepV might be closely related. The PepD mutants created by site-directed mutagenesis on putative metal binding residue Asp119 resulting in losing the activities but without changing the secondary structure with CD spectra analysis. It revealed that this residue might be involved in the metal binding for the dramatically affecting enzymatic activity. There were no detectable zinc ions in the D119E mutant crystal by X-ray diffraction, preliminary confirming our hypothesis. We also investigated the residue of catalytic Glu149 for site-directed mutagenesis. Surprisingly,

almost the mutants lost the activities besides E149D that retained partial enzymatic activity.

The similar spectra and the predict percentages of secondary structure from CD spectra compared with the wild-type PepD via CD spectra analysis confirmed that the similar secondary structure was existed in the mutant E149D. Based on this result, we suggested that changing of the putative catalytic residue glutamic acid to the aspartic acid with the same negative charge would keep the enzymatic ability to interact with the substrate.

To confirm our suggestion, the orientation relationship between the putative active site residues and zinc ions were investigated by molecular modeling. The carbonyl oxygen of the mutant D119E was far away from zinc ion to involve in the metal binding ability for the dramatically affecting enzymatic activity (Fig. 23).

Fig. 23 Stereo view of orientation relationship between zinc ion and the putative metal binding residues Asp119.

The catalytic mechanism proposed that the bridging catalytic water attacks the carbonyl carbon of the scissile peptide bond to form a sp³-orbital substrate-enzyme tetrahedral intermediate (Fig.24). The distant from the catalytic water to the carbonyl carbon of the mutant E149D was too far to from the substrate-enzyme tetrahedral intermediate that further involved in substrate binding ability and caused the mutant losing partial enzyme activity (Fig. 25).

Fig. 24 Proposed catalytic mechanism for the hydrolysis of N-terminal amino acid residues. Proposed general mechanism for the hydrolysis of a peptide, catalyzed by a

inal carboxylate.

metallopeptidase with a co-catalytic active site where R1, R2, R3 are substrate side chains and R is an N-terminal amine or a C-term

Fig. 25 Stereo view of orientation relationship between catalytic water and the putative catalytic residues Glu149.

The homology modeling structure of PepD was obtained from the L. delbrueckii PepV crystal structure with dizinc nuclear was quite similar. The putative residues for catalytic Asp82 and Glu149 of PepD were primarily superimposed on PepV Asp89 and Glu153 residues. Asp82 was conserved in all of the active enzymes from clan MH and considered to

p the imidazolium ring of His80. Glu149 served as a general base in catalysis, whereas

water molecule. Moreover, the position of the metal binding residues of PepD, PepV and CPG2 were almost clam

the water molecule was bridged by two zinc ion acting as the attacking hydroxyl ion nucleophile.⁵⁸ These two zinc ions, as described by Jozic et al.,⁵⁵ were considered to play two different roles for hydrolyzing substrates: for stabilization of the substrate-enzyme tetrahedral intermediate as well as for activation of the catalytic

superim

indistinguishable except for Asp119 (Asp141) which was considered as the bridging residue held both two zinc ions in PepV (CPG2). The orientation of Asp119 in PepD seems with less association for two zincs. However, mutants with losing activity provide some evidences for Asp119 involving in the catalytic reaction undoubtedly. Thus, we assumed that the active site pocket of PepD and PepV were similar and the hydrolytic mechanism might also be closely related but with slight difference.

Enzymes with the known crystal structures in M20 family such as PepV was identified as monomer whereas CPG2 as homodimer in their native state. The native form of V.

alginolyticus PepD was analyzed on the Native-PACE as well as western blotting analysis.

Based on the Native-PAGE and the film, we assumed that both wild-type and mutant PepD existed in several forms while monomer form was formed in dominant. However, the result of analytical sedimentation velocity ultracentrifugation indicated that there were only

formed and non-covalent interaction was apparently weak. The imerized PepD might be separated by electricity through electrophoresis analysis.

在文檔中溶藻弧菌胺醯組胺酸雙胜肽酶之生化特性分析及其功能性胺基酸之研究 (頁 43-0)