The Role of α-Amino Group of the N-terminal Serine of β Subunit for Enzyme Catalysis and Autoproteolytic Activation of Glutaryl 7-Aminocephalosporanic Acid Acylase*

Glutaryl 7-aminocephalosporanic acid (GL-7-ACA) acylase of Pseudomonas sp. strain GK16 catalyzes the cleavage of the amide bond in the GL-7-ACA side chain to produce glutaric acid and 7-aminocephalosporanic acid (7-ACA). The active enzyme is an (αβ)2 heterotetramer of two non-identical subunits that are cleaved autoproteolytically from an enzymatically inactive precursor polypeptide. In this study, we prepared and characterized a chemically modified enzyme, and also examined an effect of the modification on enzyme catalysis and autocatalytic processing of the enzyme precursor. We found that treatment of the enzyme with cyanate ion led to a significant loss of the enzyme activity. Structural and functional analyses of the modified enzyme showed that carbamylation of the free α-amino group of the N-terminal Ser-199 of the β subunit resulted in the loss of the enzyme activity. The pH dependence of the kinetic parameters indicates that a single ionizing group is involved in enzyme catalysis with pK a = 6.0, which could be attributed to the α-amino group of the N-terminal Ser-199. The carbamylation also inhibited the secondary processing of the enzyme precursor, suggesting a possible role of the α-amino group for the reaction. Mutagenesis of the invariant N-terminal residue Ser-199 confirmed the key function of its side chain hydroxyl group in both enzyme catalysis and autoproteolytic activation. Partial activity and correct processing of a mutant S199T were in agreement with the general mechanism of N-terminal nucleophile hydrolases. Our results indicate that GL-7-ACA acylase utilizes as a nucleophile Ser-199 in both enzyme activity and autocatalytic processing and most importantly its own α-amino group of the Ser-199 as a general base catalyst for the activation of the hydroxyl group both in enzyme catalysis and in the secondary cleavage of the enzyme precursor. All of the data also imply that GL-7-ACA acylase is a member of a novel class of N-terminal nucleophile hydrolases that have a single catalytic center for enzyme catalysis.

The nascent polypeptide of the enzyme is synthesized as a 74-kDa polypeptide containing a signal peptide at its N terminus. After the removal of the signal peptide, an enzymatically inactive 70-kDa precursor polypeptide is activated by proteolytic cleavages into two subunits, 16-kDa ␣ and 54-kDa ␤ subunits, in the periplasm (3,4). In a previous study (4), we have proposed that the enzyme is activated through a two-step autocatalytic processing upon folding: the first step is an intramolecular cleavage of the precursor between Gly-198 and Ser-199 for generation of the inactive ␣ subunit containing a spacer peptide of 9 amino acids and the ␤ subunit; the second is an intermolecular event, which may be mediated by the N-terminal Ser (Ser-199) of the ␤ subunit and results in a further cleavage of the peptide bond between Gly-189 and Asp-190 and the removal of the spacer peptide. Our previous data have also shown that Ser-199, the N-terminal residue of the ␤ subunit, is essential for the catalytic activity and autoproteolytic activation of the enzyme (4).
N-terminal nucleophile (Ntn) hydrolases form a novel class of hydrolytic enzymes. They are activated from an enzymatically inactive precursor polypeptide by proteolytic processing, producing a new N-terminal residue. Several amidohydrolases belonging to the Ntn hydrolase family have been described with very different substrate specificities and functions (5). Glutamine 5-phosphoribosyl-1-pyrophosphate amidotransferase (6), penicillin acylase (7), proteasome ␤ subunit (8), and aspartylglucosaminidase (9) all have a similar central four-layer sandwich of ␣ helices and ␤ sheets (␣␤␤␣) as a catalytic domain and an N-terminal nucleophile responsible for catalyzing the hydrolysis of an amide bond.
The N-terminal residue, which is exposed upon the proteolytic cleavage of the precursor polypeptide, has been shown to be essential for the enzymatic activity in the Ntn hydrolases. This residue is threonine in aspartylglucosaminidase and proteasome ␤ subunit, serine in penicillin acylase, and cysteine in glutamine 5-phosphoribosyl-1-pyrophosphate amidotransferase. All of these residues can function as a catalytic nucleophile and are located at the beginning of a ␤ strand. In the case of aspartylglucosaminidase and penicillin acylase activation, the precursor polypeptide chain is cleaved into two polypeptide subunits of the active protein, whereas in proteasome and glutamine 5-phosphoribosyl-1-pyrophosphate amidotransferase, the activation results in the removal of a propeptide. Recently, a number of studies have suggested that the breaking of this critical peptide bond occurs autocatalytically and that the autoprocessing is initiated by a proximal N-O or N-S acyl shift which occurs by nucleophilic attack by Ser, Cys, or Thr (10 -16). The suggested catalytic mechanisms of these four enzymes contain another interesting feature in common that has only been described for them. It has been suggested that the ␣-amino group of the N-terminal catalytic amino acid could function as a base that increases the nucleophilicity of the hydroxyl/thiol group (5,7,9,17).
Here, we report the first experimental study that probes the function of the free ␣-amino group of the N-terminal catalytic amino acid within the family of Ntn hydrolases. Our results suggest that the ␣-amino group of GL-7-ACA acylase is an essential catalytic group, acting as a general base catalyst by deprotonating the hydroxyl group of the N-terminal Ser-199 of the ␤ subunit in enzyme catalysis and possibly in secondary processing of the enzyme precursor. Mutagenesis of the invariant N-terminal Ser-199 confirmed the key function of its side chain hydroxyl group as a putative nucleophile in both enzyme catalysis and autoproteolytic activation. Our data would also imply that GL-7-ACA acylase is a member of a novel family of structurally similar hydrolytic enzymes with an analogous catalytic mechanism resembling that of serine proteases and in which the N-terminal amino acid of a polypeptide chain acts both as the nucleophile and the base.

EXPERIMENTAL PROCEDURES
Materials-The sources of materials used were as follows: potassium cyanate was from Fluka; Bio-Gel P-6 DG desalting gel from Bio-Rad; Sephacel C18 column from Amersham Pharmacia Biotech; Lys-C endopeptidase of sequencing grade from Roche Molecular Biochemicals; GL-7-ACA from ChongKunDang Co. All other chemicals were of analytical grade or the highest quality commercially available.
Protein Purification-GL-7-ACA acylase with the C-terminal (His) 6 tag was purified using Talon TM metal-charged affinity columns as described previously (4).
Chemical Modification of GL-7-ACA Acylase with KCNO-Purified GL-7-ACA acylase (12 g) was titrated with increasing concentrations of KCNO (10 -160 mM) in 20 mM Tris-HCl, pH 8.0. The reaction mixtures were incubated at 37°C for 1 h. After incubation, each of the mixtures was gel filtrated on Bio-Gel P-6 DG spin columns to terminate the modification reaction. The residual GL-7-ACA acylase activity was measured as described later.
Purification and Lys-C Digestion of ␤ Subunit-GL-7-ACA acylase (600 g) was incubated with 1 M KCNO in 500 l of 20 mM Tris-HCl, pH 8.0, at 37°C for 2 h. This treatment resulted in a 95% reduction of the enzyme activity. Modified and unmodified enzymes were separately dialyzed against 20 mM Tris-HCl, pH 8.0, 100 mM NaCl, and 6 M guanidine HCl at 4°C. The ␤ subunits from modified and unmodified enzymes were purified as described previously (4), and then dialyzed extensively against 25 mM Tris-HCl, pH 8.0. Each of the purified ␤ subunits (77 g) was digested with the protease Lys-C in an enzyme to substrate ratio of 1:100 (w/w) at 37°C for 24 h. The reaction mixtures were then applied to a Sephacel C18 reverse-phase column (2.1 ϫ 100 mm) in SMART system (Amersham Pharmacia Biotech) and eluted at a flow rate of 0.2 ml/min with a linear gradient of acetonitrile containing 0.1% trifluoroacetic acid. The column eluate was monitored at 214 nm.
Mass Spectrometry and Amino-terminal Sequence Analysis-Each of the entire Lys-C digests of modified and unmodified ␤ subunits was desalted using ZipTip TM (Millipore), and then mixed together with an equal volume of ␣-cyano-4-hydroxycinnamic acid solution. The sample/ matrix solution was dropped onto a sample plate for matrix-assisted laser desorption ionization time of flight mass spectrometry, and then dried under ambient conditions. A mass spectrum was obtained on a Kratos Kompact MALDI II instrument (Kratos). N-terminal amino acid sequence analysis was performed on an Applied Biosystems 491 protein sequencer fitted with a high-pressure liquid chromatography on-line system.
Site-directed Mutagenesis-Point mutations to form mutants S199T and K208Q were introduced into pSH, a plasmid that contains a sequence coding for mature wild-type GL-7-ACA acylase with a C-terminal (His) 6 tag downstream from a T7 RNA polymerase promoter site (4), using overlap extension polymerase chain reaction method (18). All the mutations were verified by automated DNA sequencing to ensure that only the desired mutation and no other changes were inserted into the sequence.
GL-7-ACA Acylase Activity Assay-The assay for GL-7-ACA acylase activity was based on the colorimetric measurement of 7-ACA released from the substrate GL-7-ACA (4). For pH profiles of enzyme activity, purified wild-type and S199T mutant enzymes were diluted into the buffers with different pH values. For pH 4 -6, 20 mM citric acid/ Na 2 HPO 4 buffers were used. For pH 7-8 and 9 -10, 20 mM potassium phosphate and 20 mM bis-Tris-propane/HCl buffers were used, respectively. Determination of the K m and V max values for wild-type and S199T mutant enzymes was performed at varying substrate concentrations at pH 7.0.
pH Dependence of Kinetic Parameters-The following pH range was explored: citric acid/Na 2 HPO 4 , pH 5-6.5, sodium phosphate, pH 7-8, bis-Tris-propane/HCl, pH 8.5-9.5. All the buffers were 20 mM, and the ionic strength of the solution was kept at 0.1 M by adding NaCl. The kinetic parameters were obtained by determining initial rates at various concentrations of the substrate at different pH values. The apparent pK a value governing the pH dependence of steady-state kinetic parameters was extracted from y versus pH profiles fitted to the following general equation, where y lim represents the maximum pH-independent value.

RESULTS
Effect of KCNO Treatment of GL-7-ACA Acylase on Enzyme Activity-The recently described Ntn hydrolases suggest that the ␣-amino group of the N-terminal catalytic amino acid of the enzymes may play an essential role as a general base to enhance the nucleophilicity of the hydroxyl/thiol group (5,7,9,17). As the penicillin acylase of these enzymes is functionally homologous to GL-7-ACA acylase, the ␣-amino group of the N-terminal Ser-199 of the ␤ subunit of GL-7-ACA acylase may be critical for the enzyme activity. To examine this possibility, we attempted to carbamylate the ␣-amino group of the Nterminal Ser-199 by treating the enzyme with KCNO under various conditions, and the chemically modified derivatives of the enzyme were examined for their enzyme activity. As shown in Fig. 2, the incubation of GL-7-ACA acylase with KCNO led to a progressive loss of enzyme activity in a concentration-dependent manner, suggesting the importance of one or more amino group-containing residues.
Determination of the Site of Modification Responsible for the Loss of Enzyme Activity of GL-7-ACA Acylase-We first examined which carbamylated subunit of GL-7-ACA acylase was responsible for the loss of enzyme activity. Since enzyme activity can be regained by enzyme reconstitution with inactive ␣ and ␤ subunits (4), the ␣ and ␤ subunits were purified from unmodified and modified GL-7-ACA acylases, and mixed in various combinations for enzyme reconstitution. Table I shows that the modification in the ␤ subunit results in enzyme inactivation.
To identify the carbamylated residue, we used ␤ subunits purified from intact and carbamylated GL-7-ACA acylases. The ␤ subunits were subjected to Lys-C endopeptidase digestion, after which the Lys-C digests were separated by reverse-phase high performance liquid chromatography. The differences observed between the two elution profiles were a single shift of an arrow-labeled peak which only appears as a minor shoulder in the HPLC chromatogram of carbamylated sample (Fig. 3, A  and B). Because the peak of overlapping peptides including the carbamylated peptide seen in Fig. 3B was not further separated by means of a shallow water/acetonitrile gradient, we analyzed the entire Lys-C digests of intact and carbamylated ␤ subunits by matrix-assisted laser desorption ionization time of flight. The mass spectrometric analyses clearly allowed us to find a new peptide having an additional 43 mass units, not found among peptides from the Lys-C digest of intact ␤ subunit (Table II). The single difference of molecular masses between the Lys-C digests of intact and carbamylated ␤ subunits corresponds to the mass of a carbamyl adduct, suggesting that the peptide having an additional 43 mass units is a peptide corresponding to residues from N-terminal Ser-199 to Lys-208 of the ␤ subunit bearing a single carbamylated amino group. Furthermore, the SNSWAVAPGK sequence corresponding to residues from N-terminal Ser-199 to Lys-208 of the ␤ subunit was determined in the N-terminal sequence analysis on the arrowlabeled peak that appears from uncarbamylated sample. For the N-terminal sequence analysis of carbamylated ␤ subunit itself, no phenylthiohydantoin-derivative of amino acid was detected. Also, an active K208Q mutant was still sensitive to enzyme inactivation with KCNO, and treatment of wild-type enzyme with 1,3,5-trinitrobenzenesulfonate, a specific lysine modifier, had no effect on enzyme activity (data not shown). All the results indicate that the single site modification of an ␣-amino group of the N-terminal Ser-199 occurred on the ␤ subunit and led to the inactivation of GL-7-ACA acylase.
Effect of the Carbamylated N-terminal Ser of ␤ Subunit on the Secondary Processing of GL-7-ACA Acylase Precursor-By enzyme reconstitution experiments, we have previously revealed that the GL-7-ACA acylase precursor undergoes the secondary processing catalyzed by the N-terminal Ser-199 of the ␤ subunit, which is generated from the primary processing of the precursor (4). To examine whether the carbamylation of the N-terminal Ser-199 of the ␤ subunit affects the secondary processing, each of the intact and carbamylated ␤ subunits was mixed and reconstituted with ␣ subunit containing a spacer peptide from a S199C mutant, which undergoes only the primary processing of its precursor (4). As shown in Fig. 4, the carbamylated N-terminal Ser of the ␤ subunit was not able to catalyze the secondary processing, i.e. the removal of the spacer peptide, suggesting that the ␣-amino group of the N-terminal Ser-199 of the ␤ subunit also plays an important role in the secondary processing of the precursor as well as enzyme catalysis.
Effects of a S199T Mutation on Enzyme Activity and Autoproteolytic Activation of GL-7-ACA Acylase-Previous substitutions of the ␤-subunit N-terminal Ser-199 by Cys or Ala suggested that the N-terminal Ser is essential both for enzyme catalysis and for complete processing of the precursor (4). To investigate whether a hydroxyl-containing residue at the position is critical for the autoproteolytic activation and enzyme activity of GL-7-ACA acylase, Ser-199 was replaced with Thr using site-directed mutagenesis. The conservative replacement of Ser-199 with Thr had no effect on the correct cleavage of the enzyme precursor into ␣ and ␤ subunits but greatly affected the enzyme activity, which was reduced to only 13% that of wild Each of ␣ and ␤ subunits was purified from intact and carbamylated GL-7-ACA acylases, which were denatured with 6 M urea. For enzyme reconstitution, the denatured ␣ and ␤ subunits were mixed in various combinations at a molar ratio of 1:1, and diluted 15-fold with renaturation buffer (20 mM Tris-HCl, pH 8.0) at 25°C for 12 h so that the enzymes could be reconstituted. The activity of a reconstituted enzyme with ␣ and ␤ subunits purified from intact GL-7-ACA acylase was normalized to 100. Ϫ, from intact enzyme; ϩ, from carbamylated enzyme.

Components
Enzyme activity FIG. 2. Effect of KCNO treatment of GL-7-ACA acylase on enzyme activity. GL-7-ACA acylase (2 M) was incubated with 10 -160 mM KCNO in 20 mM Tris-HCl, pH 8.0, at 37°C for 60 min. After incubation, each of the samples was gel filtrated on spin columns to terminate the modification reaction. The residual activity was measured as described under "Experimental Procedures." The enzyme activity of intact GL-7-ACA acylase was taken as 100%.
type. In combination with the inability of Cys to replace the N-terminal Ser of the ␤ subunit for the enzyme activity and secondary processing of the precursor, these results indicate that Ser-199 is critical not only for two steps of processing but also for catalysis as a nucleophile. Relative to wild-type enzyme, both K m and k cat values of the S199T mutant were reduced about 25-fold, but k cat /K m values were almost the same for both enzymes (Table III).
We also measured the relative enzyme activity of wild-type and S199T mutant enzymes at different pH values to detect variations in the pH optimum (Fig. 5). The S199T mutant enzyme showed some alterations in its pH optimum profiles. The activity of the wild type increased as the pH rose, having a broad optimum range between pH 6 and 10, while the S199T mutant had a relatively narrow range of pH optimum from pH 6 to 8, after which the activity declined sharply. However, it was questionable whether the decline in enzyme activity of S199T mutant at pH Ͼ 8.0 is due to an irreversible inactivation of the enzyme. To answer this issue, the enzyme was preincubated at various pH values in the range of 5-10 for the usual assay time, and the activity was then measured after buffer exchange into pH 7.0 buffer by spin column chromatography. Whereas the enzyme preincubated between pH 5.0 and 8.0 regained full activity at pH 7.0, irreversible inactivation for the enzyme increased rapidly above pH 8.0 (data not shown), indicating that the decline of the enzyme activity in the pH profile of S199T mutant enzyme must result from the irreversible inactivation of the enzyme.
pH Dependence of the Kinetic Parameters-If the ␣-amino group of the N-terminal Ser-199 of the ␤ subunit of GL-7-ACA acylase is responsible for deprotonating the hydroxyl group of Ser-199, then the enzyme would be active only when this acidic group is unprotonated in the enzyme. The pH dependence of the reaction rate should indicate the presence of an ionizable group that must be unprotonated for enzyme catalysis. Thus, we generated pH profiles of the steady-state kinetic parameters, k cat and k cat /K m . The effect of pH on the enzyme catalysis of wild-type and S199T mutant enzymes was determined over the pH range of 5.0 -9.5 or 5.5-8.0, respectively. The pH profile

FIG. 3. High performance liquid chromatography separation of Lys-C peptides of intact (A) and carbamylated (B) ␤ subunits.
The ␤ subunits from intact and KCNO-treated GL-7-ACA acylases were purified and dialyzed as described under "Experimental Procedures," and then digested with Lys-C endopeptidase in an enzyme to substrate ratio of 1:100 (w/w) at 37°C for 24 h. The Lys-C digests were applied to a Sephacel C18 reverse-phase column (2.1 ϫ 100 mm) and eluted at a flow rate of 0.2 ml/min with a linear gradient of acetonitrile containing 0.1% trifluoroacetic acid. The column eluate was monitored at 214 nm (solid lines) and the broken line shows the percentage of acetonitrile in the elution medium.

TABLE II
Experimental and calculated masses of Lys-C peptides from intact and carbamylated ␤ subunits Although the molecular mass of a ␤ subunit is determined to be 59 kDa by MALDI-TOF, the molecular mass of the ␤ subunit is given as 54 kDa in this paper as mentioned by Matsuda 4. Effect of the carbamylated N-terminal Ser of the ␤ subunit on the secondary processing. The ␣ subunit containing a spacer peptide and the ␤ subunits were purified from a S199C mutant protein (4) and intact and carbamylated GL-7-ACA acylases, respectively, all of which were denatured with 6 M urea. For enzyme reconstitution, each of the intact and carbamylated ␤ subunits was mixed with the ␣ subunit containing the spacer peptide at a molar ratio of 1:1, and diluted 15-fold with renaturation buffer (20 mM Tris-HCl, pH 8.0) at 25°C for 12 h so that the enzymes could be reconstituted. The reconstituted enzyme mixtures were then analyzed by SDS-PAGE (12% gel) followed by Coomassie Blue staining. I and C correspond to the intact and carbamylated ␤ subunits, respectively, involved in each of the reconstituted enzyme mixtures. STD and pro stand for protein standards and the spacer peptide, respectively. of both k cat /K m and k cat revealed a critical ionization that must be unprotonated for catalysis (Fig. 6). As the pH increases, there is a corresponding increase in catalytic activity that plateaus above the critical ionization with pK a values of 5.6 Ϯ 0.1 for k cat and 6.0 Ϯ 0.1 for k cat /K m . The S199T mutant showed similar results of pK a values of 5.7 Ϯ 0.1 for k cat and 6.0 Ϯ 0.1 for k cat /K m . These data indicate the catalytic requirement of an unprotonated active site group in the enzyme. The slight difference of pK a values for k cat and k cat /K m also imply little change in the states of ionization of the catalytic center group on formation of the acyl-enzyme. DISCUSSION To explore the possibility of general base catalysis by the free ␣-amino group of the N-terminal Ser-199 of the ␤ subunit of GL-7-ACA acylase, we treated the enzyme with cyanate ions under controlled conditions, and identified a single amino group essential for the catalytic activity of the enzyme. We also examined the pH dependence of the kinetic parameters of wild-type and S199T mutant enzymes. We found that carbamylation of the ␣-amino group of the N-terminal Ser-199 of GL-7-ACA acylase led to a significant loss of its enzyme activity, indicating that the ␣-amino group is essential for the catalytic activity of the enzyme. Penicillin acylase (7), aspartylglucosaminidase (9), proteasome ␤ subunit (8), and glutamine 5-phosphoribosyl-1-pyrophosphate amidotransferase (6) have been described and suggested to form a novel class of Ntn hydrolases (5). All these hydrolases have N-terminal catalytic amino acids of Ser, Thr, or Cys. The suggested mechanisms of these enzymes for catalysis contain an interesting feature in common that has only been described for them. Because no adjacent histidine equivalent to that found in the serine proteases is in close enough proximity to the N-terminal catalytic amino acid in the crystal structures of Ntn hydrolases, the proposal was made that the ␣-amino group of the N-terminal catalytic amino acid functions as a base that increases the nucleophilicity of the hydroxyl/thiol group. The ␣-amino group of the N-terminal catalytic amino acid is suitably positioned to mediate proton transfers directly or via a bridging water molecule (5,7,9,17). If the ␣-amino group of GL-7-ACA acylase has a same role for enzyme catalysis as those of the Ntn hydrolases, it is speculated that the carbamylation of the Nterminal Ser-199 of the ␤ subunit of GL-7-ACA acylase must lead to a reduction of basicity of the N␣ nitrogen of the ␣-amino group, which probably makes it unable to act as a general base to enhance the nucleophilicity of the hydroxyl group, thereby abolishing the enzymatic activity.
An active site base is required to catalyze proton transfers in overall catalysis, and we thus determined the pH dependence of k cat and k cat /K m of wild-type and S199T mutant enzymes. The k cat /K m pH profile of the wild type indicates that a single ionizable group with a pK a of 6.0 Ϯ 0.1 has to be unprotonated for activity. The k cat pH profile is very similar, except that the pK a is shifted to 5.6 Ϯ 0.1. It is likely that the acidic group revealed in the k cat /K m and k cat pH profiles corresponds to the same ionizable group. Typically, the k cat /K m pH profile represents pK a values of residues in the free enzyme or free substrate and the k cat pH profile represents pK a values in the enzyme-substrate complex (19). The pK a value from the k cat /K m pH profile most likely represents an ionizable group in free enzyme, since the free substrate does not have an ionizable group in the pH range tested. The pK a values from the k cat /K m and k cat pH profiles of the S199T mutant enzyme were similar to those of the wild type, suggesting that the same ionizable group as the wild type is required for the catalytic activity of the S199T mutant. Therefore, the pH dependence of the kinetic parameters in GL-7-ACA acylase indicates that a group with an apparent pK a of 6.0 in the free enzyme is critical for activity. Although the value of apparent pK a is somewhat less than a pK a value of 6.8 -7.9 expected for an ␣-amino group, a decreased value is plausible in a protein environment that is less polar than that of the aqueous solvent. A similar result is seen in penicillin acylase, a functional homologue of GL-7-ACA acylase and a member of the Ntn hydrolase family, in which the pK a value of the free ␣-amino group of the N-terminal catalytic Ser was assigned to 6.1 (20). Since histidine's first pK a value is 6.2, however, we could not exclude a possibility that a His residue may function as a general base in the enzyme catalysis. Based on a multiple amino acid sequence alignment of cephalosporin acylases and a comparative modeling study with the FIG. 5. The pH optimum profiles of wild-type and S199T mutant GL-7-ACA acylases. The enzyme activity was measured with wild-type and S199T mutant enzymes at various pH values as described under "Experimental Procedures." Results are shown for wild type (open circles) and S199T mutant (closed circles).

TABLE III
Catalytic parameters of wild-type and S199T mutant GL-7-ACA acylases Assays of purified wild-type and S199T mutant enzymes were carried out with 6.5 g of each enzyme added in 100 l of 0.1 M potassium phosphate, pH 7.0, containing varying substrate (GL-7-ACA) concentrations at 37°C. The assay for GL-7-ACA acylase activity was based on the colorimetric measurement of 7-ACA released from the substrate. The results shown are the average of three independent assays. The values are in the form of mean Ϯ S.D. crystal structures of penicillin acylase, we mutagenized two completely conserved His residues on the ␤ subunit which seem to reside near the catalytic Ser of GL-7-ACA acylase to Ser or Ala. As these mutant proteins were exclusively expressed as inclusion bodies, we examined the effect of the mutations on the enzyme activity by in vitro refolding and enzyme reconstitution with the ␣ subunit purified from the wild type and the mutant ␤ subunits separately expressed. The mutants had slightly reduced or comparable enzyme activities to that of the wild type in the same conditions. 2 Accordingly, it is evident that the apparent pK a of 6.0 could be ascribed to the ionization of the free ␣-amino group of the N-terminal Ser-199 of the ␤ subunit of GL-7-ACA acylase. The neutral ␣-amino group of an N-terminal amino acid could function as a base in a way similar to histidine, due to their similar pK a values: the ␣-amino group of an N-terminal amino acid has a pK a value of 6.8 -7.9, and histidine's first pK a is 6.2 (21). Like Ntn hydrolases, the ␣amino group of the N-terminal Ser of the ␤ subunit of GL-7-ACA acylase could be mainly uncharged and act as the catalytic base for accepting a proton from the hydroxyl group of the Ser-199. In this study, although it could not be clearly explained why the S199T mutant has reduced K m value and is unstable at high pH as compared with the wild type, Thr did offer somewhat an acceptable replacement for the enzyme activity. This suggests that either hydroxyl group of serine or threonine can function as a nucleophile in the enzyme catalysis. This result is consistent with the general mechanism of Ntn hydrolases in which either Ser, Cys, or Thr functions as the nucleophile.
Carbamylation of the N-terminal Ser-199 of the ␤ subunit completely inhibited the secondary processing (Fig. 4), which is catalyzed by the N-terminal Ser-199 generated from the primary processing of the enzyme precursor and results in the removal of a spacer peptide (4). Moreover, in vitro processing of the GL-7-ACA acylase precursor is maximal between pH 7.0 and 10.0 (data not shown). It is noteworthy that this behavior is similar to the pH-dependent profile for GL-7-ACA acylase activity. Accordingly, the inhibition of the secondary processing by carbamylation could be explained by speculation that carbamylation of the ␣-amino group of the N-terminal Ser makes it unable to act as a general base to activate the hydroxyl group of the N-terminal Ser, since precursor and mature enzyme may share a similar structure (16). In addition, the replacement of Ser-199 with Thr did not affect the correct processing into ␣ and ␤ subunits, suggesting that Thr could offer an acceptable replacement for autocatalytic processing of the enzyme precursor. Overall, these results suggest that a hydroxyl side chain (Ser or Thr) at position 199 is required for the complete processing reaction of the GL-7-ACA acylase precursor. According to the data, the side chain of Ser-199 would play an active role as a nucleophile in the autoproteolytic activation of GL-7-ACA acylase.
The present data allow us to conclude that the single Nterminal amino acid (Ser-199) of GL-7-ACA acylase can function as a nucleophile and a base in a way similar to that of serine and histidine residues in the active sites of serine proteases. Also, our data suggest that GL-7-ACA acylase utilizes Ser-199 as a nucleophile for both the primary and secondary processing and that the free ␣-amino group exposed from the primary processing could act as a general base for the secondary processing reaction. Taking all available experimental data together, we propose that GL-7-ACA acylase from Pseudomonas sp. strain GK16 is a member of a recently described novel class of Ntn hydrolases. X-ray three-dimensional structure analysis on the GL-7-ACA acylase that is underway by us (22) will be able to reveal the detailed catalytic mechanism of the enzyme and to encourage the studies on the catalytic and autoproteolytic activation mechanisms of the enzyme. Further work is required to characterize and understand all biochemical aspects of GL-7-ACA acylase, from autoproteolytic activation to catalysis.