Materials
The resistance of water used was more than 18 MΩ, which was purified by reverse osmosis followed by passage through a Millipore Reagent Water System (Millipore).
Candidate substrates for hydantoinase, such as barbituric acid, allantoin, dihydroorotate, uracil, pyromellitic diimide, phthalic anhydride, 5-bromouracil, 2,3-naphthalene-dicarboximide, 1,8-naphthalimide, 3,4,5,6-tetrachlorophthalimide, 4-amino-1,8-naphthalimide, 2H-1,3-benzoxazine-2,4(3H)dione, biuret, bemegride, 5,5-dimethyl-hydantoin, uridine, 5-ethyl-methyl-hydantoin, parabanic acid, rhodanine, 2,4-thiazolidinedione, N-methylmaleimide, N-cyclohexylmaleimide, and oxindole, were obtained from Aldrich (U.S.A.). DEAE Sepharose, chelating Sephacel (fast flow), and HiTrapTM desalting column were purchased from GE Healthcare Bio-Sciences (U.S.A.).
Bis-Tris propane, sodium acetate, PMSF, 8-HQSA and EDTA were purchased from Sigma (U.S.A.). Sodium chloride, sodium hydroxide, potassium phosphate, glycine, sodium dodecyl sulfate (SDS) and zinc acetate were obtained from J. T. Baker (U.S.A.). All other reagents were of the highest grades commercially available.
Enzyme assay
A rapid spectrophotometric assay [4] was used as the standard assay. Briefly, the decrease in absorbency at 298 nm was measured upon hydrolysis of phthalimide as the substrate at 25 °C. To start the reaction, the purified hydantoinase (5-100 µg) was added into a 1 ml solution, containing 1 mM phthalimide and 100 mM Tris-HCl at pH 7.9. Under these conditions, a change in A298 of 2.26 represents the hydrolysis of 1 µmol of the substrate. The hydrolysis of the substrate was monitored with a UV/Vis spectrophotometer (Hitachi U 3300).
The hydrolysis of a number of candidate substrates was assayed spectrophotometrically in an entirely similar manner except for the wavelength used; necessary modifications of the assay were needed for specific compounds. The extinction coefficient of each substrate was determined experimentally by direct measurement with a spectrophotometer.
Protein concentration
The protein concentrations of enzyme solution were determined by BCA protein assay (Pierce, USA) using bovine serum albumin as a standard.
Expression and purification of recombinant hydantoinase
The procedure for the purification of recombinant Agrobacterium radiobacter hydantoinase and its mutant proteins was modified according to a previous report [18]. Total hydantoinase activities in cell cultures grown in zinc, cobalt, or manganese ions (range from 0 to 5 mM) were determined and the maximal activities were obtained at 1 mM metal ion concentration. Cell growth was inhibited by these metal ions at concentration above 2 mM.
Thus, metal ions at 1 mM concentration were amended for cell culture and for routine purification of hydantoinase.
E. coli BL21 (DE3) cells were transformed with plasmid pHDT200 and grown in LB medium supplemented with 50 µg of ampicillin per ml plus metal ion (1 mM zinc chloride, cobalt chloride, manganese chloride, nickel chloride, or cadmium chloride) at 37 °C with rapid shaking. When the cultures (500 ml) reached an A600 of 0.6, IPTG was added to 50 µM, the temperature of flasks was shifted to 25 °C, and growth continued for an additional 25 hours. Cells were chilled on ice, harvested by centrifugation, and frozen at -80 °C. After thawing, cells were resuspended in 20 mM Tris-HCl buffer (pH 7.9) and disrupted by sonication (the power was set at 3.5; each pulse lasted 2 secs, waited 2 secs between pulses and continued for 5 minutes; the complete pulse sequence was repeated three times) with ice
cooling between pulses. The disrupted cell suspensions were centrifuged at 50,000g for 30 min. The supernatant solutions were chromatographed with DEAE Sepharose HR 10/16 column (GE Healthcare Bio-Sciences). After being washed with 20 mM Tris-HCl (pH 7.9), hydantoinase was eluted with a 0 to 0.3 M NaCl gradient in 20 mM Tris-HCl (pH 7.9) in a total volume of 500 ml. Fractions were examined for the presence of hydantoinase protein by SDS-PAGE analysis, and those containing enzyme of sufficient purity were pooled and directly applied to a column (1.6-cm diameter by 10 cm; GE Healthcare Bio-Sciences) of chelating Sephacel (fast flow) that was treated with five times gel volume containing 0.2 M zinc acetate and then equilibrated with buffer A (20 mM Tris and 0.5 M NaCl, pH 7.0). The loaded column was washed with 100 ml buffer A. The enzyme was then eluted with a linear glycine gradient from 0 to 1 M with buffer A and buffer A plus 1 M glycine in 500 ml. The fractions were examined by SDS-PAGE analysis, and those containing greater than 95% pure hydantoinase were pooled, dialyzed against the dialysis buffer (20 mM Tris-HCl, pH 8.0) with two buffer changes, and concentrated to ~10 mg/ml. The mutant proteins could be purified to homogeneity in a manner similar to native hydantoinase. In general, about 10 mg purified hydantoinase (or mutant protein) was obtained per 0.5 liter cell culture.
Preparation of apo-hydantoinase
The preparation of apo-form of the hydantoinase was according to the published procedure [19]. Briefly, purified enzyme (10 mg/ml; 10-15 ml) was dialyzed against a chelating buffer (pH 6.5) that contained MES (50 mM) and 8-HQSA (15 mM) at room temperature for 2 days. The enzyme solution was dialyzed against HEPES buffer (pH 7.0, 10 mM) at 4 °C for 4 hours, and then passed through a HiTrapTM desalting column and eluted with HEPES buffer (pH 7.0, 10 mM) at 4 °C. The activity of the resultant enzymes (apo-hydantoinase) detected by the standard assay was less than 0.4% of that of Co-reconstituted enzyme and the metal content was about 0.01 to 0.09 mol of zinc, manganese, cobalt and iron per mol of the enzyme monomer according the Inductively Coupled Plasma-Mass Spectrometry (ICP-MS) (Elan 5000, Perkin Elmer, USA). The apo-hydantoinase was stored at 4 °C and was stable for a minimum of a month.
Preparation of metal-reconstituted hydantoinase
The reconstitution of apo-enzymes with metal ion was according to the published procedure [19]. The apo-enzyme (10 mg/ml in 4 ml) was dialyzed at 4°C for 2 days in HEPES buffer (4L, 10 mM at pH 7) plus one of the following metal ions (1 mM): Co2+, Mn2+, or Zn2+. The approximate time for recovery of the maximum enzymatic activity under the condition was about 40 ± 5, 30 ± 5, and 10 ± 2 min for the Co2+, Mn2+, and Zn2+, respectively. ICP-MS was used to measure the concentration of each cation in the protein samples. Prior to the performance of ICP-MS analysis, the protein solution was passed through a HiTrapTM desalting column eluted with HEPES buffer (pH 7.0, 10 mM) at 4 °C to remove excess metal ions in the protein solution. Measurements for each sample were repeated five times and the standard deviation was calculated. Normally, two or more samples were used for each determination by ICP-MS. Less than 1 ml enzyme sample (0.05-0.2 mg/ml) was used for each determination.
The pH profiles
The Km and kcat of the enzymatic reaction was determined over the pH range of 5.5-9.5.
The reaction was buffered with 0.1 M sodium acetate (pKa = 4.8) and 0.1 M Bis-Tris propane (pKas = 6.8 and 9.0) in desired pH, and appropriate amounts of the enzyme (about 5-400 µg) were added to start the reaction. The pH of buffers was determined at 25 °C. Km and kcat were obtained by nonlinear regression (Enzyme Kinetics module of SigmaPlot) using 10-15
measurements determined at different substrate concentrations. Reaction rates obtained at each pH value were fitted to the following equation using Sigma-Plot, (Vmax)H=VmaxKa/Ka+[H+], then pKa value was given, where (Vmax)H is the maximal degradative reaction rate at a particular pH; Vmax is the maximal rate when all the enzyme sites are in the appropriate ionic form, and is used to represent the kcat; Ka is the acid dissociation constant for a catalytic residue at the active site [20]. A plot of kcat/Km or Km as a function of pH was also fitted to this equation in a similar manner.
Site-directed mutagenesis
Each hydantoinase mutant was generated according to the Stratagene QuickChange mutagenesis protocol (Stratagene, La Jolla, CA) using the plasmid pHDT2000 as the template [18]. Each mutant was confirmed by nucleotide sequencing. Expression and purification of the mutant proteins were the same as described for wild-type hydantoinase protein. The oligonucleotide primers for the preparation of mutants were
GGCATCGACGTTGCTACGCATGTCGAG and
CTCGACATGCGTAGCAACGTCGATGCC for H57A;
ATCGACGGTCATACGGCTGTCGAGACGGTC and
GACCGTCTCGACAGCCGTATGAACGTCGAT for H59A;
AGCCATGAAGACCGCGAAGGAGGTGATGCC and
GGCATCACCTCCTTCGCGGTCTTCATGGCTTA for K148A;
GCGATAAGCCATGAAGACACAGAAGGAGGTG and
GGCATCACCTCCTTCTGTGTCTTCATGGCTTA for K148C;
GCGATAAGCCATGAAGACATCGAAGGAGGTG and
GGCATCACCTCCTTCGATGTCTTCATGGCTTA for K148D;
GCGATAAGCCATGAAGACCGAGAAGGAGGTG and
GGCATCACCTCCTTCTCGGTCTTCATGGCTTA for K148S;
CTCGTCATGGTGGCCGCGGAGAACGGC and
GCCGTTCTCCGCGGCCACCATGACGAG for H181A;
GCCCCGATCTACATCGTGGCTCTGACCTGCG and
TTCTTCGCAGGTCAGAGCCACGATGTAGAT for H237A;
CTCGAAACGGTCTCCTCGGCCCATTGTTCCT and
GAGCCAGGAACAATGGGCCGAGGAGACCGT for D313A;
CTCGAAACGGTCTCCTCGGAGCATTGTTCCT and
GAGCCAGGAACAATGCTCCGAGGAGACCGT for D313E. The underlined sequences denote the mutated amino acids.
RESULTS
Metal contents and activities of recombinant hydantoinase
The addition of divalent metal ions to the standard growth medium was found to substantially increase the hydantoinase activity. Metal contents and activities of hydantoinase purified from unsupplemented zinc-, cobalt-, and manganese-amended cultures and are shown in Table 1. Cadmium and nickel ions were also tested as supplements but the resulting hydantoinase activity was similar to that from unsupplemented medium.
Mixtures of metals were found in a purified hydantoinase as shown in Table 1.
Significant amounts of iron, zinc and manganese were observed in recombinant hydantoinase obtained from cell culture without supplementing additional metal ions. It was also determined that the metal contents for Zn, Fe, Mn, Co and Ni were 12.7±0.06, 6.66±0.19, 0.34±0.004, 0.19±0.004 and 0.02±0.003 µM, respectively, in unsupplemented cell culture.
The total metal contents (for Zn, Co, Mn and Fe) in recombinant hydantoinase were near to 2 and 1.6 per enzyme subunit for purified hydantoinase obtained from metal-amended and unsupplemented media (Table 1), respectively. Interestingly, zinc, presumably the native metal ion in hydantoinase, was neither the major metal found in recombinant hydantoinase from unsupplemented medium nor the metal to give the highest hydantoinase activity (Table 1). The results about metal contents in recombinant hydantoinase shown in Table 1 indicate that preparation of apo-hydantoinase and reconstitution of the apo-hydantoinase with metals were needed for the proper understanding of the metal effect on catalytic properties of recombinant hydantoinase.
Reactivation of apo-hydantoinase
A titration curve for the reactivation of apo-hydantoinase with cobalt is shown in Fig. 2a.
This plot shows a slow increase in hydantoinase activity at low metal ion content (less than 1 cobalt per enzyme subunit) followed by a sharp increase in hydantoinase activity that reached maximal activity when the ratio of metal to protein was equal to 2. A kinetic model (Fig. 2b) for metal-activation of apo-hydantoinase was proposed for the non-linear increase in catalytic activity with an increasing ratio of cobalt to protein as shown in Fig. 2a. It was suggested that in this model only protein with a fully assembled binuclear metal center (M1M2-hydantoinase) was catalytically active and the first metal (M1) was preferentially bound to the binuclear center. It was unlikely that metal-binding of apo-hydantoinase occurred in cooperative manner.
In such a case, the metal would have been preferentially bound in pairs and a plot with a linear increase in catalytic activity would have been observed. Thus, the binding of the first metal ion to apo-hydantoinase was proposed to be tighter than that of the second metal ion.
The reconstitution of apo-hydantoinase with different divalent metal ions resulted in a differential recovery of catalytic activity as shown in Table 1. The metal-reconstituted hydantoinase gave significantly higher specific activities than those of enzyme purified directly from metal-amended cultures. Incubation of apo-hydantoinase with Cd2+ and Ni2+ did not significantly enhance hydantoinase activity. The metal content per subunit enzyme was 2.6±0.1, 2.5±0.1, and 2.8±0.1 for Co-, Zn-, and Mn-reconstituted hydantoinase, respectively.
These results indicate that the hydantoinase might have more than two metal binding sites;
two of them are needed for maximal activity (Fig. 2a).
Effect of metals on the substrate specificity of hydantoinase
The kinetic constants for the metal-reconstituted hydantoinase for several representative substrates are shown in Table 2. Significant changes of Km, kcat and kcat /Km were found for each substrate with different metals in hydantoinase, which indicate that the substrate specificity of hydantoinase was metal dependent. However, the variations were all within one order of magnitude.
Effect of metals on the pH profiles of hydantoinase
The pH dependence of enzyme activity and catalytic efficiency (kcat, and kcat/Km) of the metal-reconstituted hydantoinase was determined (figures of the pH profiles are given as supporting information). Metal ions appeared to play a significant role on enzyme function.
Both kcat and kcat/Km were notably influenced by the variation in metal ions in the optimal pH range. According to the pH profiles, a general-base catalyzed reaction of the hydantoinase was expected, and a group within the active site of hydantoinase must be ionized for catalytic activity. The kinetic pKas of the Co-, Zn-, and Mn-reconstituted hydantoinase derived from pH profiles were 6.3±0.1, 6.0±0.1, and 6.6±0.1, respectively, for kcat; 7.0±0.1, 6.7±0.1, and 6.8±0.1, respectively, for kcat/Km. These results reflect that ionization of the active site functional group was dependent on the specific metal ion bound to the binuclear metal center.
Mutational analysis for the metal binding site
The role of conserved residues (as shown in Fig. 1) for metal binding of hydantoinase was probed by site-directed mutagenesis, and the metal content and specific activity of the hydantoinase mutants are listed in Table 3. As expected, the catalytic activities for these mutant proteins were severely impaired. Only D313E was found to be active, and its specific activity was about 20-fold less than that of wild-type hydantoinase. D313E also contained about 2 metals per enzyme subunit (Table 3). K148 mutants (K148A, K148C, K148D, K148S), unable to possess a carboxylated side-chain, were found to have 0.5-0.9 metal per enzyme subunit, less than those of other mutants. Other mutant proteins (H57, H59, H181 and H237 shown in Table 3) were found to possess 1.2 to 1.4 metals per subunit. Results shown in Table 3 indicate that the binuclear metal center was essential for the catalytic activity of hydantoinase. These data were in agreement with the titration curve and proposed model presented in Fig. 2 that only the M1M2-hydantoinase form was active.
Chemical rescue of K148A by short-chain carboxylic acids
In the presence of a high concentrations of cobalt (5 mM), the K148 mutants, K148A, K148C, K148S, and K148D, were activated to values of 1.4, 4.9, 1.6, and 0.7 x10-3 µmol/min/mg, respectively. Addition of a short-chain carboxylic acid, such as acetic acid, propionic acid, or butyric acid, into the reaction mixtures further improved the activity of K148A by nearly two orders of magnitude as shown in Fig. 3. High concentrations of short chain carboxylic acids, 45, 40 and 35 mM for acetic acid, propionic acid and butyric acid, respectively, were needed to give the maximal activity of K148A. Taken together, the studies of mutational analysis and chemical rescue for K148 indicate that the posttranslational modification of carboxylated Lys148 was essential for binuclear metal coordination and assembly of active site of hydantoinase.
Dual Roles of D313 in metal binding and catalysis
In contrast to D313A, D313E maintained 2 metals per subunit and exhibited enzyme activity (Table 3). The pH dependence of D313E (Fig. 4) was found to differ from those of wild-type hydantoinase. The kinetic pKas of D313E derived from Fig. 4 were 6.7 ± 0.1 and 7.4 ± 0.1 for kcat and kcat/Km, respectively, which were significantly higher (0.4 and 0.7, respectively) than those of wild-type hydantoinase. These observations from site-directed mutagenesis and associated kinetic data indicated that D313 had dual roles in both metal binding and in catalysis.
DISCUSSION
It is interesting to observe that iron appeared to be the most favorable metal ion of recombinant Agrobacterium radiobacter hydantoinase expressed in Escherichia coli.
Significant amounts of iron were found in the recombinant hydantoinase even when a large excess of Zn, Co or Mn metal ion was added in the cell culture (Table 1). However, unlike other iron enzymes, such as cytosine deaminase [22], methionine aminopeptidase [23], and deformylase [24], iron did not activate apo-hydantoinase activity. It has been observed that recombinant hydantoinase exhibits no enzyme activity without supplementing with a large excess of active metal ions, and metal-reconstituted hydantoinase exhibits higher enzyme activity [19]. At that time, the occupation of metal binding site by iron was not suspected. The results shown in Table 1 clearly explained why metal supplementation in the cell culture was needed to obtain active recombinant hydantoinase and why metal-reconstituted hydantoinase exhibited higher enzymatic activity. Both zinc and iron are the most abundant and important metal nutrients for growth of bacteria [25]. It is unclear why iron is the preferred metal incorporated into recombinant hydantoinase and why the highest activity of hydantoinase is observed when the cell cultures are supplemented with cobalt. It is unlikely that cobalt has a biological role in vivo because of its low bioavailability in the cell [25].
Both the metal titration (Fig. 2) and metal contents (Tables 1 and 3) for bacterial hydantoinase suggest that there are two metal ions within the active site needed for hydantoinase activity. However, significantly higher metal content (about 2.5 metals per enzyme subunit) were determined in highly active metal-reconstituted hydantoinase (Table 3).
It was not surprising that additional metal binding sites on the protein surface were available because hydantoinase could efficiently bind to a chelating column (described in MATERIALS and METHODS section for protein purification). Dihydroorotase has been found to have nearly three metals per monomer [26]. The recent crystal structure of Bacillus sp.
hydantoinase reveals that there is a third metal binding site near the binuclear active site of the enzyme [17]. The significant amounts of metals found in K148 mutants (their carboxylated sites were eliminated) and more than two metals per subunit found in wild-type hydantoinase (Table 3) indicate the presence of extra metal binding sites in addition to the binuclear active site. The metal center in the amidohydrolase superfamily is vital for enzyme activity [27] and a mononuclear or binuclear metal center is the structural landmark for hydrolytic enzymes with the TIM-barrel structural fold (named after triosephosphateisomerase consisting of eight α-helices and eight parallel β-strands that alternate along the peptide backbone) [28].
Mutation of each of the amino acids of the binuclear metal center (Fig. 1) partially eliminated a metal binding site as presented in Table 3. The above observations regarding metal contents might also reconcile the seemingly contradictory data reported previously [29] for the possibility of metal contamination in recombinant hydantoinase. Metal quantitation indicates that only a mutation at lysine could eliminate the binding of nearly two metals (K148A) as compared to wild-type hydantoinase. This finding is consistent with the crystal structure shown in Fig. 1 and indicates that posttranslational lysine caboxylation is necessary for binuclear metal binding.
Hydantoinase of bacterial origin and its mammalian counterpart, dihydropyrimidinase, exhibit a broad substrate spectrum [4]. As shown in Table 2, the Km’s of the three representative substrates are high in the mM ranges, especially for those of dihydrouracil.
Variation of the active-site metal center kept the Km’s within the same order of magnitude, but the Km’s of mammalian dihydropyrimidinase are one order of magnitude lower than those of bacterial enzymes using dihydropyrimidine and 6-methyldihydropyrimidine as substrate [4], which suggests that the mammalian dihydropyrimidinase has its important biological role in the reductive degradation of pyrimidine degradation. The specific biological function of hydantoinase in bacteria is unclear [5].
The sizes of the metal have been proposed as the factors that affect the activity of hydantoinase [19]. In Table 2, the effect of metal on the substrate specificity of hydantoinase was examined for three representative substrates, one six-membered ring dihydropyrimidine, one bicyclic phthalimide and a five-membered ring hydantoin. The results indicate that the metal ion not only affected the activity of hydantoinase but also significantly affected its substrate selectivity. The differences in catalytic efficiency (kcat/Km) between the best and worst substrates tested were 10-, 52- and 178-old for Zn, Co and Mn, respectively, in the metal-reconstituted hydantoinases. This indicates that Zn-hydantoinase gave less discrimination between dihydrouracil and 5-leucinyl-hydantoin, while Mn-hydantoinase saw dihydrouracil and 5-leucinyl-hydantoin as two very distinct substrates.
Activity of K148A could be chemically rescued by small organic acids in the presence of high concentration of metal ions (Fig. 3), and smaller carboxylic acids lead to better hydantoinase activity (acetate > propionate > butyrate) (Fig. 3). The variation may be due to the accessibility of the carboxylic acids to the active site. The maximal activity of K148A was still 10-fold less than that of wild-type hydantoinase (Table 3 and Fig. 3). One of the reasons for the difference in hydantoinase activity might be because carbamate and carboxylate are different in chemical properties as shown in Figs. 5a and 5b. A resonance structure of carbamate gives both oxygen atoms of a carbamate formal negative charges (Fig. 5a) perhaps promoting the bimetal binding (31,32). A partially positive charge, which is not available for carboxylate (Fig. 5b), is also formed in carbamate resulting from lysine carboxylation. This structural feature for the active sites of phosphotriesterase [13], urease [12], dihydroorotase [11], and isoaspartyl dipeptidase [30] is very similar to that of hydantoinase. A combination of site-directed mutagenesis and chemical rescue studied for phosphotriesterase [33] gives
Activity of K148A could be chemically rescued by small organic acids in the presence of high concentration of metal ions (Fig. 3), and smaller carboxylic acids lead to better hydantoinase activity (acetate > propionate > butyrate) (Fig. 3). The variation may be due to the accessibility of the carboxylic acids to the active site. The maximal activity of K148A was still 10-fold less than that of wild-type hydantoinase (Table 3 and Fig. 3). One of the reasons for the difference in hydantoinase activity might be because carbamate and carboxylate are different in chemical properties as shown in Figs. 5a and 5b. A resonance structure of carbamate gives both oxygen atoms of a carbamate formal negative charges (Fig. 5a) perhaps promoting the bimetal binding (31,32). A partially positive charge, which is not available for carboxylate (Fig. 5b), is also formed in carbamate resulting from lysine carboxylation. This structural feature for the active sites of phosphotriesterase [13], urease [12], dihydroorotase [11], and isoaspartyl dipeptidase [30] is very similar to that of hydantoinase. A combination of site-directed mutagenesis and chemical rescue studied for phosphotriesterase [33] gives