Biochemical Characterization of WbpA, a UDP-N-acetyl-d-glucosamine 6-Dehydrogenase Involved in O-antigen Biosynthesis in Pseudomonas aeruginosa PAO1*

WbpA (PA3159) is an enzyme involved in the biosynthesis of unusual di-N-acetyl-d-mannosaminuronic acid-derived sugar nucleotides found in the O antigen of Pseudomonas aeruginosa PAO1 (serotype O5). The wbpA gene that encodes this enzyme was cloned into pET-28a, overexpressed as a histidine-tagged fusion protein, and purified by nickel chelation chromatography. Capillary electrophoresis was used to examine substrate conversion by WbpA, and the data revealed that WbpA is a UDP-N-acetyl-d-glucosamine 6-dehydrogenase (EC 1.1.1.136), which uses NAD+ as a coenzyme. The enzyme reaction product was purified by HPLC and analyzed using NMR spectroscopy. Our results showed unequivocally that the product of the WbpA reaction is UDP-N-acetyl-d-glucosaminuronic acid. WbpA requires either NH4+ or K+ for activity and the accompanying anions exert secondary effects on activity consistent with their ranking in the Hofmeister series. Kinetic analysis showed positive cooperativity with respect to UDP-N-acetyl-d-glucosamine binding with a K0.5 of 94 μm, a kcat of 86 min–1, and a Hill coefficient of 1.8. In addition, WbpA has a K0.5 for NAD+ of 220 μm, a kcat of 86 min–1, and a Hill coefficient of 1.1. The oligomerization state of WbpA was analyzed by gel filtration, dynamic light scattering, and analytical ultracentrifugation, with all three techniques indicating that WbpA exists as a trimer in solution. However, tertiary structure predictions suggested a tetramer, which was supported by data from transmission electron microscopy. The electron micrograph of negatively stained WbpA samples revealed structures with 4-fold symmetry.

known as A-band and B-band O antigen. The B-band O antigen is an important virulence factor for evasion of host defenses and confers resistance to phagocytosis and serum-mediated killing (1)(2)(3)(4)(5)(6). It is also required for virulence, as demonstrated in mouse challenge infection models, and it contributes to initial tissue damage and inflammatory responses in the lungs of cystic fibrosis patients (2,7).
WbpA and WbpI of P. aeruginosa are essential for B-band O-antigen biosynthesis, since wbpA and wbpI knockout mutants are devoid of B-band polysaccharide and produce rough LPS (15). We previously proposed that biosynthesis of the mannosaminuronic acid-derived residues of P. aeruginosa would follow a similar pathway to UDP-D-ManNAcA biosynthesis based on ϳ50% amino acid similarity between WecB and WbpI, and between WecC and WbpA (9,15). The current investigation of the activity of WbpA was prompted by the inability of wbpI and wbpA to complement wecB and wecC knockout mutants (15). Interestingly, di-N-acetylated sugars have also been found to be constituents of LPS in other bacterial pathogens, including Bordetella pertussis and Bordetella parapertusis, causative agents of whooping cough, and Bordetella bronchiseptica, which causes upper respiratory tract infections in animals (16). * This research was supported by Grant MOP-14687 (to J. S. L.) from the Canadian Institutes of Health Research. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
‡ Both authors contributed equally to this work and should be considered co-first authors. Their names are listed in alphabetical order.
§ Recipient of a Doctoral Research Grant from the Canadian Institutes of Health Research.
¶ Current address: The Hospital for Sick Children, Toronto, Ontario M5G 1X8, Canada. ** Canada Research Chair in Cystic Fibrosis and Microbial Glycobiology. To whom correspondence should be addressed. Tel.: 519-824-4120 (ext. 53823); Fax: 519-837-1802; E-mail: jlam@uoguelph.ca. 1 The abbreviations used are: LPS, lipopolysaccharide; CE, capillary electrophoresis; CE-IEF, capillary electrophoresis isoelectric focusing; COSY, correlation spectroscopy; HMBC, heteronuclear multiple bond correlation; HMQC, heteronuclear multiple quantum coherence; HSQC, heteronuclear single quantum coherence; IPTG, isopropyl thio-␤-D-ga- Here we show that WbpA (PA3159, Pseudomonas genome data base www.pseudomonas.com) is a 50.3 kDa protein that catalyzes a C-6 dehydrogenation reaction, but its activity differs from that of Cap5O and WecC in that it uses UDP-D-GlcNAc as a substrate instead of UDP-D-ManNAc. Furthermore, we propose that this represents the first step in a common pathway for the biosynthesis of di-N-acetylated hexosaminuronic acid sugar nucleotides.
Analytical Techniques-SDS-PAGE was performed according to the method of Laemmli (17). Gels were stained with Coomassie Brilliant Blue R-250 (Sigma-Aldrich). The concentration of purified WbpA was determined as described by Bradford (18).
Cloning and Overexpression of His 6 -WbpA-wbpA was cloned into the pET-28a expression vector with a N-terminal histidine tag. wbpA was amplified from genomic DNA (P. aeruginosa strain PAO1) by PCR using the following primers: forward primer 5Ј-GTTGGCATATGATA-GATGTTAACAC-3Ј, which incorporates an NdeI restriction digestion site; reverse primer 5Ј-GGGACTCGAGATCAAGCCTTG ATG-3Ј, which incorporates a XhoI restriction digestion site. The PCR reaction consisted of 100 ng of genomic DNA, 0.5 M of each primer, 2.5 mM of each dNTP, 1 mM MgSO 4 , and 1ϫ buffer. A 5-min denaturation was carried out before the addition of 2.5 units of Pwo polymerase (Roche Applied Science), followed by 30 cycles of 30 s at 94°C, 45 s at 60°C and 1 min at 72°C. A final elongation step of 7 min at 72°C was performed. Both the PCR product and the pET-28a vector were digested with NdeI and XhoI, purified using UltraClean TM 15 (Mo Bio Laboratories, Solana Beach, CA), and ligated overnight at 15°C using T4 DNA ligase (New England Biolabs, Beverly, MA). The construct obtained, pCQW14, was verified by restriction digestion and sequencing, and subsequently transformed into the expression strain E. coli BL21(DE3), using kanamycin (50 g ml Ϫ1 ) for selection. For the expression of His 6 -WbpA, 250 ml of LB broth containing kanamycin was inoculated with 6 ml of an overnight culture, and grown at 37°C. When OD 600 nm reached 0.6, IPTG was added to a final concentration of 1 mM, and expression was allowed to proceed for 3 h at 37°C. Cells were harvested by centrifugation at 5,000 ϫ g for 15 min at 4°C, and pellets were stored at Ϫ20°C.
Purification of WbpA-The pellet from 250 ml of induced culture was resuspended in 25 ml of binding buffer (20 mM sodium phosphate, pH 7.4, 500 mM sodium chloride). The cells were disrupted by ultrasonication on ice, and cell debris and membrane fractions were removed by ultracentrifugation at 175,000 ϫ g for 1 h at 4°C. Purification using the HiTrap chelating HP column was performed as recommended by the manufacturer, with nickel as the chelating ion, and binding buffer containing 200 mM imidazole as the eluent. Following purification, the concentration of sodium chloride and imidazole in purified WbpA was reduced to 100 and 5 mM, respectively, using a PD-10 desalting column (Amersham Biosciences, Baie d'Urfè, PQ) according to the protocol recommended by the manufacturer. The purity of WbpA was analyzed by SDS-PAGE, and purified WbpA was stored at Ϫ20°C after the addition of glycerol to a final concentration of 25%.
Determination of pI by Capillary Electrophoresis Isoelectric Focusing (CE-IEF)-CE-IEF analyses were performed using a P/ACE MDQ Glycoprotein System with UV detection (280 nm) and the cIEF 3-10 Kit (Beckman Coulter, Inc., Fullerton, CA). The anode and cathode electrolyte solutions (91 mM phosphoric acid and 20 mM sodium hydroxide, respectively) were prepared as recommended by the manufacturer, and the capillary was a 27-cm eCAP™ neutral capillary with a detector at 17 cm. Each run included 13.2 g of sample and ribonuclease A (pI 9.45), carbonic anhydrase II (pI 5.9), and ␤-lactoglobulin A (pI 5.1) as standards, and was performed according to the standard separation conditions, except that an additional desalting step of 5.4 kV for 2 min was added prior to focusing. Analysis of electropherograms was done using the Beckman 32 Karat TM Software, and experimental pI values were determined in each case from a curve of the three standards.
Determination of pH and Temperature Optima-All reactions contained 1 mM UDP-D-GlcNAc and 2.5 mM NAD ϩ . For determination of the pH optimum, reactions contained 2 g of enzyme and 100 mM ammonium sulfate in a reaction volume of 200 l and were performed in 100 mM MES buffer at pH 5, 5.5, 6, and 6.5; and in 100 mM bis-Trispropane buffer at pH 7, 7.5, 8, 8.5, 9, 9.5, and 10. Reactions were incubated at 37°C, and the progress of the reactions was monitored at A 340 in a BMG FLUOstar Optima spectrophotometer (BMG LabTechnologies, Durham, NC). Subsequently, the concentration of NADH produced was determined with the use of a standard curve prepared by measuring A 340 at various concentrations of NADH. For the temperature study, reactions contained 0.35 g of enzyme, 100 mM Tris, pH 7.5, and 100 mM ammonium sulfate in a reaction volume of 35 l and were incubated at 0, 15, 20, 30, 37, 42, 55, or 65°C. Percent substrate conversion for the temperature study was determined by CE.
Analysis of Cofactor Requirements of WbpA-All reactions contained 2 g of enzyme in a reaction volume of 200 l, 1 mM UDP-D-GlcNAc, 2.5 mM NAD ϩ , and 100 mM Tris (pH 7.5). Salt dependence series reactions contained 100 mM of each of the salts listed in Table II, and salt concentration series reactions contained varying concentrations of ammonium sulfate. Reactions were incubated at 37°C, and the concentration of NADH produced was determined spectrophotometrically in the same manner as the pH study described above.
Determination of Kinetic Parameters-All reactions contained 100 mM Tris (pH 7.5), 100 mM ammonium sulfate, and 1.7 g of enzyme in a final volume of 170 l. For determination of kinetic parameters for UDP-D-GlcNAc, reactions contained 3 mM NAD ϩ and concentrations of UDP-D-GlcNAc that varied from 0.025 to 1.5 mM. For determination of kinetic parameters for NAD ϩ , reactions contained 1.5 mM UDP-D-Glc-NAc and concentrations of NAD ϩ that varied from 0.05 to 3 mM. Reactions were incubated at 37°C, and the progress of the reactions was followed by measuring at A 340 in a Varian Cary 100 spectrophotometer (Varian Instruments, Walnut Creek, CA). Subsequently, the concentration of NADH produced was determined from A 340 measurements and ⑀ 340 ϭ 6,220 M Ϫ1 cm Ϫ1 . The data were fitted to the Hill equation Analysis of the Reaction Products by Capillary Electrophoresis-CE analyses were performed using a P/ACE MDQ Glycoprotein System (Beckman Coulter, Inc., Fullerton, CA) with UV detection. The running buffer was 25 mM sodium tetraborate, pH 9.5, and the capillary was bare silica 75 m ϫ 50 cm with a detector at 40 cm. The capillary was conditioned before each run by washing with 0.2 M sodium hydroxide for 2 min, water for 2 min, and running buffer for 2 min. Samples were introduced by pressure injection for 8 s, and the separation was performed at 22 kV and monitored by measuring UV absorbance at 254 nm. Peak integration was done using the Beckman 32 Karat TM Software.
Purification of the WbpA Reaction Product-A preparative scale enzymatic reaction containing 35 mol of UDP-D-GlcNAc, 77 mol NAD ϩ , and 2.5 mg of purified WbpA was incubated for 3 h at 37°C in a buffer containing 100 mM ammonium sulfate and 50 mM bis-Tris-propane (pH 10). Protein was subsequently removed from the completed reaction by ultrafiltration through a Centriplus YM-3 cartridge (Millipore, Bedford, MA). The WbpA reaction product was purified by anion-exchange FPLC, using an Econo-Pac High Q anion exchange column and a linear gradient of 0 -500 mM triethylammonium bicarbonate (pH 8.0). Fractions containing the WbpA reaction product were pooled, and the triethylammonium bicarbonate buffer was removed by addition of AG 50W-X4 resin (Bio-Rad) until the pH reached 4.5, followed by filtration to remove the resin. After adjusting the pH to 7.0 by addition of triethylamine, the pooled fractions were then lyophilized and analyzed by NMR.
NMR Analysis of the WbpA Reaction Product-All spectra were acquired using a Varian Inova 500 MHz spectrometer equipped with a Z-gradient 3 mm triple resonance ( 1 H, 13 C, 31 P) probe. The lyophilized sugar nucleotide sample was dissolved in 140 l of D 2 O. The experiments were performed at 25°C with suppression of the HOD (deuterated water) signal at 4.78 ppm. The methyl resonance of acetone was used as an internal reference at ␦ H 2.225 ppm and ␦ C 31.07 ppm. External 85% phosphoric acid was used for the 31 P chemical shift reference (0 ppm). Standard sequences from Varian, COSY, HSQC, HMBC, and 31 P HMQC and selective 1D-TOCSY experiments were used for assignments (19 -21).
Size-Exclusion HPLC Analysis-The molecular weight of WbpA in aqueous solution was determined with the use of a Superose 6 HR 10/30 gel filtration column (Amersham Biosciences, Baie d'Urfè, PQ) with a solvent of 50 mM sodium phosphate, pH 7.4, and 150 mM sodium chloride. The column was calibrated with carbonic anhydrase (29 kDa), bovine serum albumin (66 kDa), yeast alcohol dehydrogenase (150 kDa), and ␤-amylase (200 kDa) as standards. A calibration curve was plotted as the log (MW) versus [Ϫlog(K av )] 1/2 , giving correlation values between 0.96 and 0.99. After calibration and equilibration of the column with solvent, 100 l of WbpA (1.1 mg ml Ϫ1 ) was injected. For each of the standards and WbpA samples, the average distribution coefficient, K av , was calculated from the elution volume, . V 0 is the column void volume, and V t is the column total bed volume. The molecular mass of WbpA was subsequently determined from the calibration curve.
Dynamic Light Scattering-The hydrodynamic radius of WbpA was measured by dynamic light scattering in a DynaPro instrument with Dynamics software (Protein Solutions, Lakewood, NJ). WbpA was measured at a concentration of 73 g ml Ϫ1 at room temperature in buffer containing 20 mM sodium phosphate, 150 mM sodium chloride, and 25% glycerol (v/v).
Transmission Electron Microscopy (TEM) of WbpA-TEM experiments were performed as described previously (22) with the following modifications. The copper grids were floated on a drop of 10 g ml Ϫ1 WbpA, 20 mM sodium phosphate, pH 7.4 without pre-wetting the grid, and then floated on 1.0% (w/v) uranyl acetate (pH 4.5; Fisher Scientific Ltd.) for 20 s. Samples were then examined with a Phillips EM 400 transmission electron microscope operating at an accelerating voltage of 100 kV.

RESULTS
Expression and Purification of WbpA-WbpA was expressed to high levels using the pET expression system and E. coli BL21(DE3) expression host cells, and ϳ80% of expressed WbpA was soluble (Fig. 1). Typically, 7.7 mg of highly purified WbpA (Ͼ98% pure, as estimated from the SDS-PAGE gel) could be obtained from 250 ml of culture. The apparent molecular mass of WbpA (50.4 kDa) as determined from SDS-PAGE correlates well with the predicted molecular mass of 50.3 kDa. CE-IEF analysis of WbpA performed without the desalting step showed several species with pI values ranging from 5.18 to 7.71 (Fig.  2B). When the desalting step was added, WbpA appeared as one species with an apparent pI of 5.5 ( Fig. 2A), which is slightly lower than the predicted pI of 5.8.
Functional Characterization of WbpA Using CE-CE analysis of WbpA-catalyzed reactions containing UDP-D-GlcNAc and NAD ϩ revealed two reaction products that eluted at 11 and 13 min, respectively (Fig. 3). The product peak at 11 min was identified to be NADH by comparison to an NADH standard, while the product peak at 13 min was subsequently identified as UDP-N-acetyl-D-glucosaminuronic acid (UDP-D-GlcNAcA) by NMR analysis (see below). Further analysis showed that 2 mol of NAD ϩ were consumed for every mole of UDP-D-GlcNAc catalyzed, and 2 mol of NADH were produced for every mole of UDP-D-GlcNAcA produced. Reactions carried out at pH 7.5 revealed 84% conversion of UDP-D-GlcNAc while reactions carried out at pH 8.5 revealed Ն 98% conversion. Reactions carried out at pH 7.5 with UDP-D-glucose, UDP-N-acetyl-D-galactosamine, or UDP-D-galactose as substrates failed to show any conversion, demonstrating the preference for UDP-D-GlcNAc as its substrate.
Identification of Reaction Product by NMR-The 1 H spectrum is presented in Fig. 4A. Proton assignments were made using COSY and TOCSY experiments. From the one-dimensional selective TOCSY experiment on the anomeric sugar resonance and the proton spectrum, accurate proton coupling constants could be obtained (Fig. 4B). A J 1,2 value of 3.2 Hz and large J 2,3 , J 3,4 , and J 4,5 values of 10.5, 9.1, and 10.1 Hz, respectively, indicated an ␣-glucopyranose configuration. From the HSQC spectrum (Fig. 4D) the protonated carbons were assigned. From the HMBC spectrum, the quaternary carbon at C-6 and the NAc-CϭO resonances were assigned. The 1 H and 13 C chemical shifts are given in Table I. The 31 P HMQC spec- trum (Fig. 4C) indicated the correlations to the pyrophosphate group with 31 P chemical shifts at Ϫ10.5 (P ␣ ) and Ϫ12.1 ppm (P ␤ ). From the 1 H spectrum J P␤, H1 was found to be 7.6 Hz, and J P␤, H2 was 2.8 Hz. The 1 H, 13 C chemical shifts for the UDP moiety were similar to those previously reported (19). These results provided the evidence that the structure of the WbpA enzyme-substrate reaction product is UDP-D-GlcNAcA.
Determination of Physical Parameters by CE-WbpA is able to catalyze the conversion of UDP-D-GlcNAc over a wide pH range, with an optimum pH between 8.0 and 9.0 (data not shown). WbpA is active at temperatures ranging from near 0 to 65°C, with maximum activity between 33 and 37°C (data not shown). The activity of WbpA can be maintained by storage in 25% glycerol at Ϫ20°C for more than 3 months.
Analysis of Cofactor Requirements of WbpA-WbpA is able to catalyze the conversion of UDP-D-GlcNAc in the presence of ammonium and potassium salts, but no significant conversion is observed in the presence of tetramethylammonium, dimethylammonium, or sodium salts (Table II). The initial rate values are highest when the accompanying anion is kosmotropic, and lowest when the accompanying anion is chaotropic, with a strong correlation between initial rate and the placement of the anion within the Hofmeister series (27). The concentration of ammonium sulfate has a different effect on initial rate than on overall substrate conversion. An increase in ammonium sulfate concentration correlates with an increase in the initial rate (Fig. 5). Similarly, overall substrate conversion (as measured at 250 min) increases as ammonium sulfate concentration increases, up to a level of 50 mM. Above this level, however, as the concentration of ammonium sulfate increases, the overall substrate conversion decreases.
Determination of Kinetic Parameters of WbpA-While the data from experiments with UDP-D-GlcNAc as the variable substrate and NAD ϩ as the fixed substrate fit reasonably well to the Michaelis-Menten model when analyzed by non-linear curve fitting, Eadie-Scatchard plots of the same data showed a convex curve indicative of positive cooperativity (data not shown). As a result, the data were fitted to the Hill equation (Fig. 6A), and the steady-state kinetic parameters are listed in Table III. Subsequently, Hill plots of the UDP-D-GlcNAc rate  data revealed a Hill coefficient of 1.8 (Fig. 6B). Kinetic analysis of experiments with NAD ϩ as the variable substrate (and UDP-D-GlcNAc fixed) showed near-Michaelis-Menten behavior, with a Hill coefficient of 1.1. In all cases, kinetic parameters were determined by linear regression (R 2 Ն 0.99) from Lineweaver-Burk plots. The data were also analyzed from Eadie-Scatchard and Hanes plots with similar results.
Determination of Oligomerization State of WbpA-The results from analytical gel filtration chromatography showed that WbpA has an apparent MW of 162,000. A comparison of this value to the predicted molecular weight of 50,318 for monomeric WbpA suggests that WbpA monomers associate to form a trimer in solution. Similar results were obtained when WbpA was analyzed by dynamic light scattering, which showed WbpA to have a hydrodynamic radius in solution of 5.11 Ϯ 0.16 nm and a calculated molecular weight of 153,000. However, when WbpA was analyzed by TEM and negative staining, the resulting micrographs revealed oligomers that have two axes of symmetry, one with 2-fold symmetry and one with 4-fold symmetry (Fig. 7).
Structure Predictions for WbpA-Three-dimensional models of WbpA, including side chains, were generated by 3D-PSSM and SWISS-MODEL. Interestingly, the algorithm of each program selected GDP-mannose 6-dehydrogenase from P. aeruginosa (AlgD; 1mfz.pdb, 1muu.pdb, 1mv8.pdb) as the most appropriate template. The model of WbpA generated by SWISS-MODEL contains no gaps and, based on superimposition of this model onto AlgD (1muu.pdb), 324 of 436 total amino acid residues of WbpA are in structurally equivalent positions and 90 of these are identical. WbpA shares 58.7% amino acid sequence similarity and 24.1% amino acid identity to AlgD. It is intriguing that despite this low sequence identity, WbpA has identical residues in 16 of 19 positions in AlgD that form hydrogen bond contacts with NADH, 4 of 6 that form contacts to the pyranose ring, 4 of 6 that form contacts to pyrophosphate, and 3 of 9 that form contacts to the GDP/ribose region (data not shown).

DISCUSSION
WbpA is essential for B-band O-antigen biosynthesis in P. aeruginosa PAO1 (15), and in this report we show that it is a UDP-D-GlcNAc 6-dehydrogenase. Although UDP-D-GlcNAc 6-dehydrogenase activity was previously observed in both Achromobacter georgiopolitanum and Micrococcus luteus (28,29), the enzyme itself was never identified. Therefore, this is the first report of identification, purification, and in-depth biochemical characterization of this enzyme, including detailed analysis of cofactor requirements.
CE analysis of WbpA reactions demonstrated that it is able WbpA reactions using UDP-D-GlcNAc and NAD ϩ as substrates All reactions contained 2 g of enzyme in a reaction volume of 200 l, 1 mM UDP-D-GlcNAc, 2.5 mM NAD ϩ , 100 mM Tris (pH 7.5), and 100 mM of the appropriate salt, and were incubated at 37°C. The data shown are averages from experiments performed in triplicate.   to use UDP-D-GlcNAc as a substrate, and based on the inability of wbpA to complement a wecC mutant (15), WbpA is unable to use UDP-D-ManNAc as a substrate. Our results show, then, that biosynthesis of the di-N-acetylated mannosaminuronic acid-derived sugar nucleotides of P. aeruginosa differs from that of the UDP-D-ManNAc-derived sugars of E. coli and S. aureus. Biosynthesis of the former begins with a C-6 dehydrogenation reaction whereas biosynthesis of the latter first proceeds via an epimerization at C-2 before C-6 dehydrogenation. This is also the case with biosynthesis of UDP-N-acetyl-D-galactosaminuronic acid (UDP-D-GalNAcA), which begins with an epimerization at C-4 before C-6 dehydrogenation (Fig. 8) (30, 31). Furthermore, due to the presence of WbpA homologues in P. aeruginosa strains that contain di-N-acetylated glucosaminuronic acid and di-N-acetylated gulosaminuronic acid, we predict that the WbpA reaction represents a common first step in the biosynthesis of di-N-acetylated hexosaminuronic acid sugars. Data from our CE analysis also showed that WbpA requires 2 mol of NAD ϩ for every mole of UDP-D-GlcNAc and produces 2 mol of NADH for every mole of product, which was shown by NMR analysis unequivocally to be UDP-D-GlcNAcA. Based on these results, and the known mechanism of UDPglucose dehydrogenase (32), we conclude that WbpA catalyzes the following shown in Reaction 1.
WbpA has a broad pH optimum of 8 -9, which correlates well with the reported pH optima of UDP-D-GlcNAc 6-dehydrogenase from A. georgiopolitanum (pH 9) and from M. luteus (pH 8.8), and of UDP-glucose dehydrogenase from E. coli (pH 9) (28,29,33). This also correlates well with the results of WbpO (pH 8.5), another dehydrogenase that was previously characterized by our group (31). The preference for basic conditions may be due to the involvement of a cysteine residue, since the mechanisms for both AlgD (34) and Ugd (35) involve nucleophilic attack by a cysteine to form a thiosemiacetal intermediate, and a basic pH would help to stabilize the thiolate anion that would be required (typical pK a for a cysteine side chain is ϳ9.0). The most likely candidate, based on amino acid alignment with AlgD, is Cys 273 , which aligns with Cys 268 from the active site of AlgD.
In our initial studies of WbpA we discovered that ammonium sulfate was essential for enzymatic activity. Closer examination revealed that WbpA requires monovalent cations, either NH 4 ϩ or K ϩ in particular, for activity. This observation is in contrast to previous suggestions that ammonium sulfate was required to prevent loss of UDP-D-GlcNAc 6-dehydrogenase activity in extracts from M. luteus (29). Furthermore, our results are consistent with those of other sugar-nucleotide modifying enzymes including MurB, a UDP-N-acetylenoylpyruvylglucosamine reductase from E. coli and Streptococcus pneumoniae that requires K ϩ , NH 4 ϩ , or Rb ϩ (36,37). No significant activity was observed in reactions containing (CH 3 ) 4 N ϩ , (CH 3 ) 2 NH 2 ϩ , or Na ϩ , suggesting the effects observed with NH 4 ϩ or K ϩ do not result from stabilization of WbpA because of the kosmotropic nature of these cations, but rather that NH 4 ϩ and K ϩ are activators of WbpA. NH 4 ϩ and K ϩ have both been identified as activators of fructose-1,6-bisphosphate aldolase from E. coli, and the structure of that enzyme revealed a common binding site in which either of these cations can be pentavalently coordinated (38). The specificity for these cations in the cation binding site of fructose-1,6-bisphosphate aldolase was attributed to the capacity of these cations to participate in hydrogen bonding with atoms 2.7-3.2 Å away, which is significantly longer than the bond distances of all other monovalent and divalent cations (38). WbpA may have a similarly organized cation binding site, since the presence of a similar binding site would provide a rational explanation for the requirement of NH 4 ϩ and K ϩ cations specifically for activity. It is interesting to note that neither AlgD nor Ugd, which perform similar reactions on nucleotide activated hexoses, require NH 4 ϩ or K ϩ for activity, although this does not eliminate the possibility that NH 4 ϩ are K ϩ are directly involved in the active site of WbpA. An example of this is 1-L-myo-inositol-1-phosphate (MIP) synthase from yeast compared with MIP synthase from Archaebacteria fulgidus, both of which catalyze the conversion of D-glucose 6-phosphate to MIP (39). Only the yeast MIP is activated by NH 4 ϩ , and the crystal structure of this enzyme suggests the involvement of NH 4 ϩ in the reaction mechanism of this enzyme (40).
We also observed that the anion of the NH 4 ϩ -or K ϩ -containing salt appears to play a significant role in stabilizing the enzyme activity of WbpA. The effects of these anions correlate well with their placement within the Hofmeister series (27),  whereby the initial reaction rate is increased by anions of increasing kosmotropicity and decreased by anions of increasing chaotropicity, as observed in Table II. These results are similar to previous observations that kosmotropes stabilize the native conformation of acidic (negatively charged) proteins, resulting in enhanced enzyme activity (41,42), since WbpA is negatively charged at pH 7.5 (pI 5.5). These results are also consistent with the observed inactivation of phosphofructokinase by chaotropic anions according to their increasing chaotropicity in the Hofmeister series (43). Analysis of the ammonium sulfate concentration data reveals both a concentration-dependent activation effect and, at high concentrations, a salting-out effect (Fig. 5). The activation effect is observed throughout the concentration range, since an increase in the concentration of ammonium sulfate correlates with an increase in initial rate. At high concentrations of ammonium sulfate, however, the overall conversion of substrate to product decreases with an increase in ammonium sulfate concentration, despite an increased initial rate. This effect is consistent with the salting out effect characteristically observed with kosmotropic salts at high concentrations. The salting out phenomenon may also explain why reactions containing F Ϫ have lower initial rates than the equivalent reactions with SO 4 2Ϫ as the anion, despite their ranking in the Hofmeister series.
WbpA has a K 0.5 of 94 M for its specific substrate, UDP-D-GlcNAc, which is lower than those observed for the M. luteus (280 M) or A. geogiopolitanum (500 M) enzymes (28,29) indicating a higher binding efficiency between WbpA and its substrate. This result can be expected since the latter two enzymes were only partially purified when analyzed. The K 0.5 value for NAD ϩ for WbpA (220 M) also showed better binding efficiency than the value obtained for the M. luteus (1,430 M) and A. georgiopolitanum (1,500 M) enzymes, but again these values are likely due to the partial purity of the enzyme preparation used. In addition, the K 0.5 of WbpA for NAD ϩ is also lower than that for WbpO (650 M), but this difference may be due to the difference in the WbpO-preferred substrate (UDP-D-GalNAc rather than UDP-D-GlcNAc). This difference may also in part be caused by the fact that WbpO had to be refolded before kinetic analysis could be carried out (31). For both UDP-D-GlcNAc and NAD ϩ , the V max is 2.9 nmol/min. Interestingly, the k cat values of WbpA (86 min Ϫ1 for both UDP-D-GlcNAc and NAD ϩ ) are higher than the values for WbpO (47.8 min Ϫ1 for UDP-D-GalNAc and 26.8 min Ϫ1 for NAD ϩ ) indicating faster turnover rate for WbpA. This may simply reflect a difference in activity or may be a function of the refolding of WbpO. It is worth noting that the k cat of WbpA for UDP-D-GlcNAc is considerably lower than for WbpM (168 min Ϫ1 ), another enzyme from P. aeruginosa PAO1 O-antigen biosynthesis, which is known to be highly efficient in its catalytic activity (11).
Eadie-Scatchard analysis of WbpA kinetic data for UDP-D-GlcNAc showed the effects of positive cooperativity, an effect that is not immediately apparent from Michaelis-Menten plots of the same data. In addition, these data fit better to the Hill equation (R 2 ϭ 0.99) than to the Michaelis-Menten equation (R 2 ϭ 0.96), and this, combined with a Hill coefficient of 1.8 indicates that WbpA is an allosteric enzyme. These results are particularly interesting since GDP-mannose dehydrogenase (AlgD), which is involved in exopolysaccharide alginate biosynthesis, also shows positive cooperativity (44). Allosteric enzymes often catalyze key regulatory steps in metabolic pathways, and regulation may be in the form of activation or inhibition. Furthermore, regulated enzymes often catalyze a step that represents the first committed step in a particular branch of the metabolic pathway. Since WbpA catalyzes the first committed step in the biosynthesis of di-N-acetylated mannosaminuronic acid residues of the B-band O antigen, the effects of positive cooperativity may be an indication that this is a regulated step.
A Hill coefficient of 1.8 also indicates that WbpA is oligomeric in solution. Subsequent analysis using gel filtration, dynamic light scattering and analytical ultracentrifugation (data not shown) substantiated the interpretation that WbpA is a trimer in solution. These results are surprising since dehydrogenases that have been crystallized to date have been shown as dimeric, tetrameric or hexameric structures (33,34,45,46). However, when visualized by TEM and negative staining, images of WbpA oligomers with 3-fold symmetry could not be discerned. Instead, the images of the protein showed an axis of 2-fold symmetry and an axis of 4-fold symmetry suggestive of a tetrameric structure. This is in agreement with a study of the structure of GDP-mannose dehydrogenase from P. aeruginosa,