HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 4, 2823-2829, January 25, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/4/2823    most recent M108888200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Kim, Y. Articles by Hol, W. G. J. Search for Related Content PubMed PubMed Citation Articles by Kim, Y. Articles by Hol, W. G. J. Social Bookmarking                      What's this?

Precursor Structure of Cephalosporin Acylase

INSIGHTS INTO AUTOPROTEOLYTIC ACTIVATION IN A NEW N-TERMINAL HYDROLASE FAMILY^*

Youngsoo Kim^§¶, Sanggu Kim, Thomas N. Earnest, and Wim G. J. Hol^§**

From the  School of Chemical Engineering, Yeungnam University, Dae-Dong, Kyungsan 712-749, Korea,  Berkeley Center for Structural Biology, Physical Biosciences Division, Lawrence Berkeley National Laboratory, Berkeley, California 94720, and the ^§ Department of Biochemistry, Biomolecular Structure Center and ^** Howard Hughes Medical Institute, University of Washington, Seattle, Washington 98195-7742

Received for publication, September 14, 2001

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Autocatalytic proteolytic cleavage is a frequently observed post-translational modification in proteins. Cephalosporin acylase (CA) is a recently identified member of the N-terminal hydrolase family that is activated from an inactive precursor by autoproteolytic processing, generating a new N-terminal residue, which is either a Ser or a Thr. The N-terminal Ser or Thr becomes a nucleophilic catalytic center for intramolecular and intermolecular amide cleavages. The gene structure of the open reading frame of CAs generally consists of a signal peptide followed by the -subunit, a spacer sequence, and the -subunit, which are all translated into a single polypeptide chain, the CA precursor. The precursor is post-translationally modified into an active heterodimeric enzyme with - and -subunits, first by intramolecular cleavage and second by intermolecular cleavage. We solved the first CA precursor structure (code 1KEH) from a class I CA from Pseudomonas diminuta at a 2.5-Å resolution that provides insight into the mechanism of intramolecular cleavage. A conserved water molecule, stabilized by four hydrogen bonds in unusual pseudotetrahedral geometry, plays a key role to assist the OG atom of Ser¹ to generate a strong nucleophile. In addition, the site of the secondary intermolecular cleavage of CA is proposed to be the carbonyl carbon of Gly¹⁵⁸ (Kim, S., and Kim, Y., (2001) J. Biol. Chem., 276, 48376-48381), which is different from the situation in two other class I CAs.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Autoproteolytic peptide cleavage is a frequent form of post-translational modification. Many protein systems are modified by intramolecular and/or intermolecular cleavages from precursors to be active enzymes. N-terminal (Ntn)¹ hydrolases form a novel class of hydrolytic enzymes that are activated from an inactive precursor polypeptide by autoproteolytic processing, generating a new N-terminal residue (1). The newly generated Ntn residue of the processed chain carries out two catalytic functions, i.e. it is both a nucleophile and a proton donor. The Ntn hydrolase superfamily was defined by SCOP (structural classification of proteins) as containing four layers of -helices and -sheets in an fashion (2). There were five known families of the Ntn hydrolase superfamily: class II glutamine amidotransferases, penicillin G acylase, penicillin V acylase, proteasome subunit, and glycosylasparaginase (GA) (2). Recently, it was reported by identifying the motif in cephalosporin acylase that cephalosporin acylase also belongs to one family of Ntn hydrolase superfamily (3).

Several Ntn hydrolases are activated through autoproteolytic processing by Ser, Thr, or Cys residues. The autoproteolytic processing was observed to occur in an intramolecular manner in several Ntn hydrolases including glutamine 5-phosphoribosyl-1-pyrophosphate amidotransferase, penicillin G acylase, proteasome b subunit (PS), GA, and cephalosporin acylase (CA) (4-6). The structures of the precursors of GA and of PS were determined, and autoproteolytic mechanisms for intramolecular cleavage by Thr residue are proposed (6, 7). Interestingly, those two Ntn hydrolases, GA and PS, proceed to autoproteolysis along different paths even if the same amino acid, Thr, is used as a key nucleophile for autoproteolysis: a water enhances the nucleophilicity of the Thr O at the +1 position in PS, whereas the carboxylate of Asp¹⁵¹ at the 1 position promotes the nucleophilicity of the Thr¹⁵² at the +1 position in GA. It is known that the intramolecular proteolysis proceeds to an NO or NS acyl shift (8-10) and leads to an ester or thioester that is subsequently hydrolyzed to carboxyl and amino groups. The resulting N-terminal Ser, Thr, or Cys residue becomes exposed to solvent to play the role of the nucleophilic catalytic center for Ntn hydrolases (1, 4, 7).

The gene structure of the open reading frame of CAs varies with each enzyme but generally consists of a signal peptide followed by an -subunit, a spacer sequence, and a -subunit. The genes of CAs are translated into an inactive single precursor peptide that is post-translationally modified into an active enzyme with one -subunit and one -subunit (11). All known CAs use either Ser or Thr as a key nucleophile to carry out both autoproteolysis and enzymatic deacylation. It was suggested that the nascent polypeptide of cephalosporin acylase is autoproteolytically activated through a two-step autocatalytic process upon folding (5, 11). The first step is an intramolecular cleavage of the precursor at the start of -subunit, resulting in an -subunit, a spacer peptide attached to the C terminus of the -subunit, and a -subunit. The second step is an intermolecular event, which may be mediated by the newly generated N-terminal Ser or Thr of -subunit by intramolecular cleavage. The second event results in a further cleavage at the second scissile bond and finally releases the spacer peptide (11, 12). Presently, the mechanism of the intramolecular cleavage of CA is not yet understood, and furthermore, it is not even clear which residue of the spacer peptide is cleaved during the secondary intermolecular cleavage (11, 12).

CA has so far only been used to a limited extent to produce 7-aminocephalosporanic acid (7-ACA), which is a backbone chemical in synthesizing semisynthetic cephalosporins. Because of the poor efficiency of CAs in generating 7-ACA, the majority of 7-ACA is obtained in industry by an expensive chemical process from cephalosporin C using expensive toxic compounds, which causes environmental safety problems. Therefore, CA has been a target for protein engineering to modify its substrate specificity so as to carry out a more efficient enzymatic conversion of cephalosporin C to 7-ACA with cephalosporin C as substrate (13, 14).

A class I CA from Pseudomonas diminuta KAC-1 (CAD) is an Ntn hydrolase, and the Ser¹ of CAD,² generated by intramolecular cleavage, plays a key role as a nucleophile in both the intramolecular autoproteolysis of precursor CAD and the enzymatic deacylation of mature CAD (15).³ The site of the secondary intermolecular cleavage is not yet clearly defined.

In the current report, we describe the first precursor structure of CA, which explains the mechanism of the intramolecular autoproteolysis of the protein. This structure reveals that one conserved water, present both in precursor CAD and CAD, plays a key role in assisting the OG atom of Ser¹ so that the nucleophilic OG atom can carry out an attack on the scissile carbonyl carbon of Gly^169s. The conserved water is hydrogen bonding with four surrounding residues in an unusual pseudotetrahedral geometry. On the other hand, sites of secondary intermolecular cleavage were proposed in the other two class I CAs (5, 11). However, our data propose that the secondary intermolecular cleavage may occur at the carbonyl carbon of Gly¹⁵⁸, which is different from the report on the other known two class I CAs.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Crystallization-- The Ser¹ to Ala (S1A) mutant CAD, S1A precursor CAD (17), from P. diminuta KAC-1 (15, 16) was prepared by site-directed mutagenesis and subcloned into Escherichia coli BL21(DE3) with the overexpression vector pET24d(+) and was purified using phenyl-Sepharose and sephacryl-200 gel filtration column chromatography as will be described elsewhere. The S1A precursor CAD concentration was 10 mg ml¹ in a storage buffer (50 mM sodium phosphate, pH 7.0, and 150 mM NaCl). The S1A precursor CAD crystals grew overnight at 21 °C from hanging drops containing 3 µl of protein solution (10 mg ml¹ S1A precursor CAD, 50 mM sodium phosphate, pH 7.0, and 150 mM NaCl) and 3 µl of reservoir solution (20% (w/v) polyethylene glycol 8000 (PEG 8000), 10 mM dithiothreitol, 200 mM magnesium acetate, and 100 mM sodium cacodylate, pH 6.5) by the vapor diffusion method against a 500-µl reservoir solution. The S1A precursor CAD crystals grew under very similar conditions as the native CAD crystals (3), and they belong to the same space group P4₁2₁2 with almost identical unit cell dimensions. There is a single chain precursor of 77 kDa/asymmetric unit and 57% solvent content. Crystals were transferred to a cryo-solution containing 23% (w/v) PEG 8000, 10% (w/v) glycerol, 200 mM magnesium acetate, 10 mM dithiothreitol, 100 mM sodium cacodylate for 3 days before flash-cooling in a 100 K gaseous nitrogen stream. The long soaking time of the S1A precursor CAD crystal in the cryo-solution improved crystal quality tremendously by reducing mosaicity.

Data Collection-- Data set for a S1A precursor CAD crystal was collected to a resolution of 2.5 Å at the Advanced Light Source (ALS) 5.0.2 beamline from the frozen crystals using a wavelength of 0.91 Å. All data were indexed and integrated using DENZO and scaled by SCALEPACK (18).

Refinement-- The S1A precursor CAD structure yielded an excellent initial crystallographic model based on the CAD structure, and only minor adjustments are needed in the course of refinement. The F_o  F_c difference Fourier maps provided an excellent guide to locate spacer residues (see Fig. 1). Only the spacer residues were built using XtalView/Xfit (19) and O (20) onto the CAD structure. All crystallographic refinements were carried out using CNS (21) with maximum likelihood refinement. Model geometry was checked by PROCHECK (22). In the S1A precursor CAD structure, 86.6, 12.6, and 0.5% of the main chain dihedral angles are in the most favorable, the additionally allowed, and generously allowed regions, respectively. Only one residue, Phe¹⁷⁷, which is positioned in the active site, is located in a disallowed region, in agreement with the observation of Herzberg and Moult (23) that residues with unfavorable , combinations tend to occur near functionally important residues of proteins. The CAD structure is almost identical to the native structure except for the spacer residues. Data and refinement statistics are shown in Table I. Figures are generated by MOLSCRIPT (24), BOBSCRIPT (25), RASTER3D (26), and GRASP (27).

MALDI Mass Spectroscopy of the -Subunit of Wild-type Cephalosporin Acylase-- Protein spots of the -subunit of CAD were excised from SDS-PAGE gel and in-gel digested with trypsin as described previously in detail (28). Microcrystalline matrix surfaces with the trypsin-digested protein sample were prepared on the probe tips of the mass spectrometer according to the sample preparation procedure described previously (28).

All mass spectra were obtained on a Voyager TOF-MS mass spectrometer (PerSeptive Biosystems, MA). The original data acquisition, data transfer, subsequent averaging of time-of-flight data, as well as all further data processing were carried out using the PerSeptive-GRAMS/32 processing software provided by the manufacturer, operated on Microsoft Windows 98. MALDI peptide spectra were calibrated using several matrix ion peaks as internal standards.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Precursor Structure of Cephalosporin Acylase-- Precursor CAD spontaneously autoproteolyzes the scissile peptide bond between Gly^169s and Ser¹ by intramolecular cleavage upon folding after protein synthesis (5). It was previously reported that the S1A mutant protein completely lost intramolecular autoproteolytic activity (17). To obtain a non-processed CAD precursor for structure determination, the S1A CAD mutant protein was prepared by site-directed mutagenesis, and the size of non-cleaved precursor CAD was confirmed by SDS-PAGE from cell extract of the S1A mutant of precursor CAD (data not shown). A 77-kDa S1A protein, which contains an -subunit, a spacer, and a -subunit, appeared as a single band that corresponds to non-processed CAD (data not shown).

The structure of the S1A precursor CAD was determined to a 2.5-Å resolution (Table I). In the course of the structure refinement, the F_o  F_c difference Fourier map, with all the spacer residues omitted for phase calculation, showed clear positive density for spacer sequence prior to adding any information of these backbone residues in the unbiased map (Fig. 1). The 11 residues of the spacer from 159s to 169s (EGDPPDLADQG) were easily built into the F_o  F_c difference map using O (20) and were refined using CNS (21) to an R_cryst of 20.3% and an R_free of 23.7% (Table I). (The sequence of spacer peptide and the determination of the secondary intermolecular cleavage site will be discussed below.)

View larger version (49K): [in this window] [in a new window]   Fig. 1.   Stereo view of Fo  Fc difference Fourier map. Fo  Fc difference Fourier map at the region of intramolecular cleavage including several spacer residues and neighboring amino acids. Coefficients for difference Fourier maps are Fobs, precursor  Fcalc, precursor with the shown residues omitted, using calc, the shown residues omitted. The maps are contoured at 3.4 . The mutation of Ser1 to Ala is labeled as S1A in the S1A precursor CAD. The conserved waters (WAT1 and WAT2) are labeled and will be discussed. The difference electron density is superimposed upon the refined coordinates.

Structural Comparison between Precursor Cephalosporin Acylase and Cephalosporin Acylase-- The structure of S1A precursor CAD is shown in ribbon diagrams (Fig. 2, A and B). The red spacer structure adopts a loop conformation (P-loop) and is attached at two ends: one end is linked to the -helical end of an -subunit (Gly¹⁵⁸), and the other is attached to the start of -strand of a -subunit (Ser¹). The overall conformation of the precursor CAD structure is very similar to CAD structure (3) such that a root mean square (r.m.s.) deviation of 0.18 Å is obtained after superimposing 672 common C atoms. The Cs of the nine active site residues of CAD, which interact directly with the substrate (16), glutaryl-7-ACA (GL-7-ACA), superimpose onto the corresponding C atoms of precursor CAD with an r.m.s. deviation of 0.22 Å. This indicates that conformational change is negligible in the active site before and after autoproteolysis. Even the crucial Ser¹ hardly changes position after activation, with an r.m.s. deviation of all common Ala/Ser¹ atoms in the two structures of 0.06 Å and a shift in position of the C atom of 0.11 Å (Fig. 3). Apparently, the conformation of the active site is virtually fully established in precursor CAD.

View larger version (72K): [in this window] [in a new window]   Fig. 2.   A, stereo view of the S1A precursor CAD. The precursor structure is a single chain protein, consisting of the  subunit of active CAD in green, a red P-loop (spacer structure in loop form), and the -subunit of CAD in yellow. The view is looking at the side from the active site cleft. The side-chain pocket for binding the substrate GL-7-ACA is represented by a ball-and-stick model in the center (16). The first six residues of S1A precursor CAD are disordered, and thus the label N in the -subunit indicates the locations of Gln7. B, an orthogonal view of panel A.

View larger version (19K): [in this window] [in a new window]   Fig. 3.   Superposition of CAD and S1A precursor CAD structures. Nine interacting residues with the substrate (16), GL-7-ACA, are superimposed onto each other. Active site residues are shown in ball-and-stick models. The S1A precursor CAD is shown in gray, and the CAD structure is shown in black. The mutation of Ser1 to Ala is labeled as S1A in the precursor CAD.

The key difference between precursor and mature enzyme is that the P-loop of the precursor proceeds through the substrate-binding site (16) so that it blocks binding of GL-7-ACA (Fig. 4). The side chain of the precursor residue, Gln^168s, occupies the binding site of the four-membered -lactam ring that is fused to a six-membered ring of GL-7-ACA, but the binding pocket for the glutaryl side chain of GL-7-ACA remains unoccupied. In addition, the carboxylic side chain of the precursor residue, Asp^154s, is located at the same position a carboxyl group that is attached to a six-membered ring of GL-7-ACA occupies when the substrate is placed into the tentative binding site of precursor CAD (Fig. 4). Once the spacer is cleaved at one end by intramolecular cleavage, the spacer peptide becomes free to move its C-terminal end elsewhere. After the first intramolecular cleavage step, the key nucleophile for enzymatic catalysis (Ser¹) is liberated to be functional for enzymatic reaction. As a result, the substrate-binding site is available for substrate (GL-7-ACA), and also probably for intermolecular cleavage of the spacer peptide at the secondary cleavage site.

View larger version (16K): [in this window] [in a new window]   Fig. 4.   Stereo view of the spacer residues occupying the binding site of substrate. Nine active site residues of S1A precursor CAD are in yellow. They are supposedly interacting with the substrate GL-7-ACA once precursor CAD becomes CAD. The spacer residues from Asn164s to Gln168s are in red. The GL-7-ACA is in purple. The Asn164s and Gln168s collide with GL-7-ACA, showing that the spacer residues block binding of substrate before intramolecular cleavage occurs. The mutation of Ser1 to Ala is labeled as S1A in precursor CAD.

Conformation of Active Site in Precursor Cephalosporin Acylase-- In the CAD structure (3), the nucleophilic OG atom of Ser¹ is making a hydrogen bond with a water molecule (WAT1 in Fig. 5A), which in its turn makes two additional hydrogen bonds, one with the main chain NH of Val⁷⁰ and another with the OD1 atom of Asn²⁴⁴. The latter residue, which plays a key role in the chemical catalysis of CAs (3), is conserved among four classes of CAs (CA I-IV), including penicillin G acylase, that go through the same path of chemical catalysis (29). The Ala¹ residue of the S1A precursor CAD can be modeled as Ser¹, based on the structural comparison of active site conformations between S1A precursor CAD and mature CAD as represented in Fig. 3. The OG atom was added to the residue Ala¹ of the S1A precursor CAD, resembling the torsional geometry of the OG atom of Ser¹ residue in CAD structure as shown in Fig. 5A, thereby resulting in a model of Ser¹ in the S1A precursor CAD (Fig. 5B). Interestingly, water WAT1 of CAD (Fig. 5A) is also present at the same location in precursor CAD (Fig. 5B), but in the latter, it is forming four hydrogen bonds in pseudotetrahedral geometry with the main chain NH of Val⁷⁰, the OD1 of Asn²⁴⁴, the carbonyl oxygen of Gln^168s, and the OG of S1A. The conserved water (WAT1) of S1A precursor CAD shown in Fig. 5B functions similarly to a member of the catalytic triad of serine proteases (30). During catalysis by serine proteases, the OG atom of Ser is polarized to become a catalytic nucleophile by the proton acceptors such as His and Asp (30). In precursor CAD, the conserved water donates its two hydrogens to two hydrogen acceptors (the carbonyl oxygen of Gln^168s and OD1 of Asn²⁴⁴) and accepts two hydrogens from the hydroxyl group of Ser¹ and the main chain NH of Val⁷⁰, thereby assisting in carrying out the nucleophilic attack by the hydroxyl group of the serine in intramolecular cleavage.

View larger version (23K): [in this window] [in a new window]   Fig. 5.   Hydrogen bonding of a conserved water in precursor CAD and CAD. The dotted lines represent hydrogen bonds ranging from 2.8 to 2.9 Å. A, the conserved water of CAD. The conserved WAT1 is hydrogen-bonded to two proton acceptors and one donor, which are also involved in chemical catalysis of CAD (3). The residues of CAD are in green. B, the conserved water of precursor CAD. The OG atom of Ser1 in the S1A precursor CAD was modeled after the torsional geometry of the OG atom of Ser1 residue of CAD structure. The modeled OG atom of the S1A precursor CAD is shown in green. The modeled residue is represented as S1A. In the Ser1 model of the residue S1A, the conserved WAT1 is hydrogen-bonded to two proton acceptors and two donors in pseudotetrahedral geometry. The residues of S1A precursor CAD are shown in yellow in the ball-and-stick model.

Model of Autocatalytic Mechanism of Precursor CA-- Given the closely similar conformations of Ser¹ in CAD and Ala¹ in S1A precursor CAD, we are assuming in the following discussion that the OG of Ser¹ in S1A precursor CAD adopts a position, which is close to that observed in the Ser¹ of CAD (Fig. 5, A and B). On the basis of the conformation of the conserved water WAT1 in Fig. 5, A and B, we would propose a plausible mechanism for the intramolecular cleavage of precursor CA (Fig. 6, A and B).

View larger version (26K): [in this window] [in a new window]   Fig. 6.   A proposed mechanism of the intramolecular cleavage for CAD. The arrows represent nucleophilic attack. The boxed residue represents each amino acid. WAT1 and WAT2 represent the conserved waters as also shown in Figs. 1 and 6. The numbers near the lines represent distances between two atoms. The Ser1 is the key nucleophilic center for intramolecular cleavage. A, the first nucleophilic attack on the scissile carbonyl carbon of Gly169. The dotted lines represent hydrogen bonds from the structure of S1A precursor CAD. The OG atom of Ser1 is assisted by WAT1 that is forming four hydrogen bonds in pseudotetrahedral geometry. The WAT2 is 3.9 Å away from the carbonyl oxygen of Gly169 before the nucleophilic attack occurs. The carbonyl oxygen of Gly169s is hydrogen-bonded to the main chain NH of residue His23, and the WAT2 is also hydrogen-bonded to the side chain of the residue His23. B, the second nucleophilic attack to hydrolyze the intermediate ester linkage. In this hypothetical model on the basis of the proposed mechanism, the carbonyl oxygen of the intermediate ester and the intermediate oxyanion may be stabilized by the oxyanion hole consisting of the conserved WAT2 and the main chain NH of His23. The oxyanion hole may be shaped by conformational change, resulting from the ester bond formation due to the NO acyl shift. The dotted lines represent hypothetical hydrogen bonds.

The overall schematic summary shown in Fig. 6A is that the nucleophilic attack on the carbonyl carbon of Gly^169s by the OG atom of Ser¹ results in an ester formation between the carbonyl carbon of Gly^169s and the OG atom of Ser¹. Initially, the hydroxyl of Ser¹ is assisted by the conserved water (WAT1 in Fig. 6A), which is stabilized by four hydrogen bonds in pseudotetrahedral geometry and may accept a proton from the hydroxyl group. As a result, the hydroxyl of Ser¹ is precisely positioned for nucleophilic attack. After the OG atom of Ser¹ carries out the nucleophilic attack on the main chain carbonyl carbon of Gly^169s, a tetrahedral intermediate is formed, and the resulting oxyanion may move toward WAT2 so as to be stabilized by hydrogen bonds from the oxyanion hole consisting of WAT2 and the main chain NH of His²³. In the S1A precursor CAD structure, WAT2 is located 3.9 Å away from the carbonyl oxygen of Gly^169s before the nucleophilic attack (Fig. 6A). Therefore, the oxyanion needs to move by about 1 Å to form a hydrogen bond with the WAT2. The oxyanion intermediate may collapse and result in shifting the linkage of the amide bond to an ester bond (NO acyl shift) (5, 31). The WAT1 charged with a proton from the hydroxyl of Ser¹ will promote this step by donating a proton to the main chain NH of Ser¹. Subsequently, the conserved water (WAT1) may carry out a second nucleophilic attack on the carbonyl carbon of the newly formed ester intermediate to result in another tetrahedral transition intermediate, which is stabilized in the same manner as before by the oxyanion hole consisting of WAT2 and the backbone NH of His²³. The ester bond will be broken, resulting in a free N-terminal Ser¹ and the carboxylate of Gly^169s (Fig. 6B). The space peptide is now only linked at one end to the -chain and can presumably move quite freely around. Once the mobile end of the spacer peptide is moving out of the catalytic site, it can be used to accept its low molecular weight substrates (for example, GL-7-ACA) but can probably also carry out releasing the spacer peptide for intermolceular catalysis.

The Intermolecular Cleavage Site of Cephalosporin Acylase-- The secondary intermolecular cleavage site of CAD was confirmed by MALDI-TOF mass spectroscopy with the trypsin-digested peptide fragments of wild-type -subunit (Table II). The mixture of the digested fragments was subject to MALDI-TOF mass spectroscopy, and each peak in the mass spectrum is annotated to a digested fragment by comparing molecular weight of the mass spectrum peaks with calculated molecular weight. Identification of peaks is also aided by knowing the cleavage sites of trypsin (Arg and Lys) in the -subunit of CAD. As shown in Table II, the mass spectrum of the peptide fragments agreed well with the calculated molecular weights. The presence of fragment 144-158, which is the last peptide fragment of the -subunit in the trypsin-digestion mixture of active CAD, clearly indicates that the last residue of the -subunit is Gly¹⁵⁸. This can be accompanied with the structure of wild-type CAD, which was recently solved at 2.0 Å by multiwavelength anomalous dispersion method (3). The C terminus of the -subunit in the superb experimental electron density map obtained by the CAD structure determination (3) is shown in Fig. 7. The electron density at the C terminus of CAD matches Gly¹⁵⁸ clearly and accommodates well a carboxylate moiety. The residue, Gly¹⁵⁸, is positioned at the end of -helix, and the main chain NH of Gly¹⁵⁸ is hydrogen-bonded to the main chain O of Gly¹⁵⁴ at a 2.6-Å distance, which may make the electron density at the C terminus visible. The x-ray map can obviously not be taken as proof of the absence of residues beyond Gly¹⁵⁸, but it is reassuring that the x-ray map and the mass spectrometry data are in excellent agreement. Therefore, it appears that the spacer residues, ranging from 159 to 169 with sequence EGDPPDLADQG, are first cleaved at the scissile peptide bond between Gly^169s and Ser¹ by intramolecular cleavage. Subsequently, the mobile spacer is cut off at the secondary scissile amide bond between Gly¹⁵⁸ and Glu^159s by intermolecular cleavage.

View this table: [in this window] [in a new window]   Table II MALDI-TOF mass spectroscopy of the trypsin-digested fragments for the -subunit of CAD

View larger version (68K): [in this window] [in a new window]   Fig. 7.   Stereo view of the multiwavelength anomalous dispersion experimental electron density map at the C terminus of -subunit. The map is calculated after solvent flipping and contoured at 1.4  (3). The electron density is superimposed upon the refined coordinates from the residue Ala151 to Ala158 of the -subunit of CAD. The electron density clearly ends at the C-terminal carboxylate group, in agreement with the mass spectrometry data (see Table II), indicating that Gly158 is the last residue of the -subunit. The main chain NH of Gly158 is hydrogen-bonded to the main chain O of Gly154. Two waters are represented by WAT.

Park et al. (5) proposed the spacer sequence as DPPDLADQG in a class I CA from Pseudomonas sp. GK16, and Wang et al. (11) reported the spacer as GDPPDLADQG in a class I CA from Pseudomonas sp. 130. These suggested sequences for spacers are two or one amino acids shorter, respectively, than in CAD. Determination of the exact secondary cleavage sites for the two previous cephalosporin acylases may need further investigation, although on the basis of the CAD structure (3), it appears that no collision would occur between the one or two extra residues and the protein or the bound substrate.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Mutagenesis studies showed that mutating Gly^169s to other amino acids, including Ala, which carries the smallest side chain, prohibited the precursor CA from proceeding to intramolecular cleavage (11). To explain those experimental mutagenesis results with reference to the S1A precursor CAD structure, Gly^169s was replaced with Ala in the S1A precursor CAD structure by the graphics program O (20), and its local conformation was optimized accordingly. The methyl group of the replaced Ala contacts the side-chain OD1 of Asn²⁴⁴ in a 2.0-Å distance, which is a component in making four hydrogen bonding networks around WAT1 (Fig. 5B). Therefore, it is evident from the S1A precursor CAD structure that any residue in this position other than Gly^169s will result in breaking the hydrogen bonding network of WAT1, thereby preventing the conserved WAT1 from performing its critical role in the activation process (Fig. 6A).

The detailed pattern of autoproteolysis in S1A precursor CAD is somewhat different from the previously determined precursor structures of Ntn hydrolase families such as GA (7) and PS (6). GA uses a Thr residue as a nucleophile to form an intermediate ester, and the hydroxyl of the nucleophile Thr is assisted by the carboxylic group of Asp to initiate the nucleophilic reaction. The PS also uses a Thr as a nucleophile to form an intermediate ester. The hydroxyl of the Thr residue in PS is assisted by a conserved water molecule, as in the precursor CAD, but the conformation of active site in PS is very different from that of precursor CAD. In PS, the three consecutive residues Thr¹ (a nucleophile residue carrying out autoproteolysis), Gly¹ (a cleaved amino acid at the carbonyl carbon), and Leu² (one upstream from Gly) are forming a tight loop (see Fig. 3 of Ditzel et al. (6)), whereas the corresponding residues of precursor CAD, Ser¹-Gly^169s-Gln^168s, are located on an extended -strand (Fig. 5B). In precursor PS, the OG of Thr¹ is within 3.4 Å from the carbonyl carbon of Leu² (6), whereas in precursor CAD, the equivalent distance between the OG of Ser¹ and the carbonyl carbon of Gln^168s is 5.2 Å, showing important key differences in the active sites of precursors between similar enzymes with the same Ntn hydrolase topology.

In precursor CAD, the planarity ( value, the dihedral angle of peptide bond) of the peptide bond at the scissile peptide (the bond between Gly^169s and Ser¹) is consistently refined close to the ideal value (180°) that deviates by less than 1  from ideality (5.8°) (32) when using the ideal weight between experimental data and empirical energy function during the refinement by CNS (21). In addition, the angles of the N-Ca-C' ( angle) of Gln^168s and Gly^169s are 121.4° and 102.6°, which deviate from the ideal value (112.5° and 111.2°, respectively) by 3  (32). The significance of this value in a 2.5-Å structure is of course limited. In contrast, the  value of the catalytic Thr¹⁵¹ of the 1.9-Å precursor GA structure deviates more than 3  from ideality (180 for trans), and the  angle of Asp¹⁵¹ near the scissile peptide bond of precursor GA deviates from ideality by more than 2 ; thus those two geometric distortions together contributed to raise the energy for each distorted residue (7). Likely, precursor CAD may follow a different pattern from GA in accommodating geometric distortions imposed on the residues near the scissile peptide bond prior to autocatalytic cleavage. In view of these three precursor structures, it seems that Ntn hydrolases, which require auto-activation, may not follow exactly the same path during intramolecular autoproteolysis even though they share the same overall topology and key active site residues.

	ACKNOWLEDGEMENTS

We thank Ki-Hong Yoon and Dae-Won Kim for the molecular biology work, Ethan Merritt and Jungwoo Choe for helpful discussions, Jerry McDermott for technical assistance on the Advanced Light Source (ALS) 5.0.2 beamline, and Dr. Dong Bin Yim and Hyun Soo Kim of Kyungsang National University for MALDI-TOF mass spectrometry.

	FOOTNOTES

^* This work was supported by a major equipment grant from the Murdock Charitable Trust to the Biomolecular Structure Center (to W. G. J. H.). This work was supported by a grant from the Korean Ministry of Commerce, Industry and Energy in 2001 (to Y. K.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

The atomic coordinates and the structure factors (code 1KEH) have been deposited in the Protein Data Bank, Research Collaboratory for Structural Bioinformatics, Rutgers University, New Brunswick, NJ (http://www.rcsb.org/).

^¶ To whom correspondence should be addressed: Yeungnam University, School of Chemical Engineering, Dae-Dong, Kyungsan 712-749, South Korea. Tel.: 82-53-810-2521; Fax: 82-53-814-8790; E-mail: ykim1@yu.ac.kr.

Published, JBC Papers in Press, November 8, 2001, DOI 10.1074/jbc.M108888200

² The notation of amino acid sequence in precursor CAD is identical to CAD up to the residue 158 (3). The -subunit is represented by attaching "" at the end of the residue number, and the -subunit is represented by putting "" at the end of the residue number. Additionally, the spacer residue is represented by putting "s" at the end of a residue number, such as 159s. The spacer sequence of precursor CAD, which is from residue 159s to 169s in this structure, was included as part of -subunit in the previous study (3) because it was not clear at that time where the secondary cleavage occurred to release the spacer peptide from the -subunit. For example, residues 159-169 were described as 159-169 in the previous study (3).

³ K.-H. Yoon, GenBank^TM Accession Number, AF251710.

	ABBREVIATIONS

The abbreviations used are: Ntn, N-terminal; CA, cephalosporin acylase; CAD, a class I CA from Pseudomonas diminuta KAC-1; GA, glycosylasparaginase; 7-ACA, 7-aminocephalosporanic acid; GL-7-ACA, glutaryl-7-ACA; S1A, Ser¹ to Ala mutation; P-loop, spacer structure in loop form; MALDI, matrix-assisted laser desorption/ionization; TOF, time-of-flight.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				Y. Sun and H.-C. Guo Structural constraints on autoprocessing of the human nucleoporin Nup98 Protein Sci., March 1, 2008; 17(3): 494 - 505. [Abstract] [Full Text] [PDF]

				T. Okada, H. Suzuki, K. Wada, H. Kumagai, and K. Fukuyama Crystal Structure of the {gamma}-Glutamyltranspeptidase Precursor Protein from Escherichia coli: STRUCTURAL CHANGES UPON AUTOCATALYTIC PROCESSING AND IMPLICATIONS FOR THE MATURATION MECHANISM J. Biol. Chem., January 26, 2007; 282(4): 2433 - 2439. [Abstract] [Full Text] [PDF]

				J. Yoon, B. Oh, K. Kim, J. Park, D. Han, K. K. Kim, S.-S. Cha, D. Lee, and Y. Kim A Bound Water Molecule Is Crucial in Initiating Autocatalytic Precursor Activation in an N-terminal Hydrolase J. Biol. Chem., January 2, 2004; 279(1): 341 - 347. [Abstract] [Full Text] [PDF]

				H. Suzuki and H. Kumagai Autocatalytic Processing of gamma -Glutamyltranspeptidase J. Biol. Chem., November 1, 2002; 277(45): 43536 - 43543. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 277/4/2823    most recent M108888200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Kim, Y. Articles by Hol, W. G. J. Search for Related Content PubMed PubMed Citation Articles by Kim, Y.</