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

J. Biol. Chem., Vol. 275, Issue 30, 22824-22831, July 28, 2000

This Article Abstract Full Text (PDF) All Versions of this Article: 275/30/22824    most recent M000133200v1 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 Duewel, H. S. Articles by Woodard, R. W. Search for Related Content PubMed PubMed Citation Articles by Duewel, H. S. Articles by Woodard, R. W. Social Bookmarking                      What's this?

A Metal Bridge between Two Enzyme Families

3-DEOXY-D-MANNO-OCTULOSONATE-8-PHOSPHATE SYNTHASE FROM AQUIFEX AEOLICUS REQUIRES A DIVALENT METAL FOR ACTIVITY^*

Henry S. Duewel and Ronald W. Woodard

From the Interdepartmental Program in Medicinal Chemistry, University of Michigan, Ann Arbor, Michigan 48109-1065

Received for publication, January 7, 2000, and in revised form, May 12, 2000

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The enzymes 3-deoxy-D-manno-octulosonic acid-8-phosphate synthase (KDO8PS) and 3-deoxy-D-arabino-heptulosonic acid-7-phosphate synthase (DAHPS) catalyze analogous condensation reactions between phosphoenolpyruvate and D-arabinose 5-phosphate or D-erythrose 4-phosphate, respectively. While several similarities exist between the two enzymatic reactions, classic studies on the Escherichia coli enzymes have established that DAHPS is a metalloenzyme, whereas KDO8PS has no metal requirement. Here, we demonstrate that KDO8PS from Aquifex aeolicus, representing only the second member of the KDO8PS family to be characterized in detail, is a metalloenzyme. The recombinant KDO8PS, as isolated, displays an absorption band at 505 nm and contains approximately 0.4 and 0.2-0.3 eq of zinc and iron, respectively, per enzyme subunit. EDTA inactivates the enzyme in a time- and concentration-dependent manner and eliminates the absorption at 505 nm. The addition of Cu²⁺ to KDO8PS produces an intense absorption at 375 nm, while neither Co²⁺ nor Ni²⁺ produce such an effect. The EDTA-treated enzyme is reactivated by a wide range of divalent metal ions including Ca²⁺, Cd²⁺, Co²⁺, Cu²⁺, Fe²⁺, Mg²⁺, Mn²⁺, Ni²⁺, and Zn²⁺ and is reversibly inhibited by higher concentrations (>1 mM) of certain metals. Analysis of several metal forms of the enzyme by plasma mass spectrometry suggests that the enzyme preferentially binds one, two, or four metal ions per tetramer. These observations strongly suggest that A. aeolicus KDO8PS is a metalloenzyme in vivo and point to a previously unrecognized relationship between the KDO8PS and DAHPS families.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Over the past several years, our understanding of a unique class of enzymes responsible for the incorporation of the 3-carbon skeleton of phosphoenolpyruvate (PEP)¹ into pivotal biosynthetic intermediates has dramatically increased. Common to this particular enzyme class, which can be divided into two broad groups, is cleavage of the C-O bond of PEP concurrent with catalysis, as opposed to the more conventional cleavage of the P-O bond (1-7). Members of the first group, 5-enolpyruvylshikimate-3-phosphate synthase and UDP-N-acetylglucosamine enolpyruvate transferase, promote transfer of the intact carboxyvinyl portion of PEP to a cosubstrate alcohol with the formation of an enolic ether linkage to the C-2 of PEP. The mechanisms of enolpyruvylshikimate-3-phosphate synthase (8-12) and UDP-N-acetylglucosamine enolpyruvate transferase (13-15) have been thoroughly characterized.

The second group of enzymes associated with C-O bond cleavage reactions presently includes 3-deoxy-D-manno-octulosonic acid-8-phosphate (KDO8P) synthase (KDO8PS) (16, 17) and 3-deoxy-D-arabino-heptulosonic acid-7-phosphate (DAHP) synthase (DAHPS) (18, 19). Both enzymes catalyze the condensation of PEP with a phosphorylated monosaccharide via coupling of the C-3 of PEP to the C-1 of an aldose cosubstrate to produce a 3-deoxy-2-keto sugar acid three carbons longer. KDO8PS occupies an essential position in the biosynthesis of lipopolysaccharide in Gram-negative microorganisms (20, 21), using D-arabinose 5-phosphate (A5P) in its condensation reaction to yield KDO8P and inorganic phosphate. KDO8P is the precursor to 3-deoxy-D-manno-octulosonic acid (KDO), an unusual octulose found in the inner core of lipopolysaccharide. KDO assumes an important role in the overall assembly of lipopolysaccharide, and its production and utilization have proven central to cellular viability and homeostasis (22-25). In the reaction catalyzed by DAHPS, D-erythrose 4-phosphate (E4P) serves as the aldose cosubstrate yielding the heptulose, DAHP. Constituting the first product of the shikimic acid pathway, DAHP ultimately leads to the biosynthesis of the aromatic amino acids and other aromatic metabolites in microorganisms and plants (19, 26). Bacteria and fungi typically encode multiple forms of DAHPS that are sensitive to a particular feedback effector. In Escherichia coli, three isozymes exist, each specifically inhibited by one of the three aromatic amino acids (27, 28).

Despite numerous studies of the reactions catalyzed by KDO8PS (2-4, 29-35) and DAHPS (7, 36-39), which have centered almost exclusively on the enzymes from E. coli, details pertaining to the mechanism of these enzymes and the nature of reaction intermediates continue to remain uncertain. However, independent investigations have clearly established that both enzymes catalyze a reaction in which a carbanion equivalent generated at C-3 on the si face of PEP is directed to react with the re face of the electrophilic carbonyl of the respective monosaccharide (33, 34, 38, 40).

While KDO8PS and DAHPS catalyze seemingly identical reactions, the enzymes have a different requirement for a metal cofactor. The three E. coli DAHPS isozymes are fully dependent on a divalent metal for activity, a property apparently shared by other bacterial and eukaryotic DAHP synthases (41, 42). In contrast, and stemming largely from early observations on the enzyme from E. coli (16), it has been generally accepted that KDO8PS has no metal cofactor requirement. Although the identification of other bacterial (42-45) and plant (46, 47) KDO8P synthases has been reported, the enzymes from these organisms have yet to receive adequate characterization in order to establish their individual enzymatic properties. Therefore, a paucity of studies dealing with detailed investigations of KDO8PS from a diverse set of organisms currently exists.

In our pursuit to expand the current understanding of this distinctive enzyme family, we have recently reported on a novel KDO8PS from one of the earliest known diverging eubacteria (48). The kdsA gene from the hyperthermophile Aquifex aeolicus (49) was cloned and expressed in E. coli, and the function of the recombinant enzyme as a KDO8P synthase has been confirmed (48). Consistent with the thermophilic phenotype of A. aeolicus, the recombinant KDO8PS demonstrates exceptional thermostability and is maximally active at 95 °C (48). In the present study, we provide strong evidence that KDO8PS from A. aeolicus requires a divalent metal for activity. Our findings on the first example of a metal-dependent KDO8PS establish a novel relationship between the KDO8PS and DAHPS families.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Materials-- Phosphoenolpyruvate mono(cyclohexylammonium) salt, D-arabinose 5-phosphate disodium salt, 1,10-phenanthroline, Q Sepharose Fast Flow, and reagent grade CaCl₂, CdCl₂, CuCl₂, FeSO₄, and ZnCl₂ were used as provided by Sigma. Puratronic grade ZnSO₄, CuSO₄, MgCl₂, and MnCl₂ (99.999% metal basis), CoCl₂ (99.998% metal basis) and NiCl₂ (99.995% metal basis) were purchased from Johnson Matthey (Ward Hill, MA). Analytical reagent grade EDTA disodium salt was from Mallinckrodt (Paris, KY). All other chemicals were used as supplied by Fischer (Springfield, NJ). High grade Spectra/Por® 7 dialysis tubing (10,000 molecular weight cut-off), obtained from VWR Scientific (Chicago, IL), was boiled in 1 mM EDTA and then extensively washed prior to use. Mono Q (HR 10/10), phenyl-Superose (HR 10/10), and FAST Desalting (HR 10/10) chromatography columns were from Amersham Pharmacia Biotech. Thinwall polymerase chain reaction tubes from United Laboratory Plastics (St. Louis, MO) served as the reaction vessel for enzymatic reactions, and temperature control was achieved with use of a heating block unit from VWR Scientific.

Purification of KDO8PS-- The recombinant A. aeolicus KDO8PS used in these studies was isolated from E. coli BL21(DE3) cells harboring plasmid pAakdsA. The construction of pAakdsA, growth of cultures, and preparation of the crude extract has been described previously (48). All chromatography was performed using an FPLC system (Amersham Pharmacia Biotech) at 23 °C with detection at 280 nm. Solid sodium chloride was added to the cell extract (obtained from a 2.5-liter culture; 65 ml; 330 mg of total protein) to a final concentration of 0.1 M, and the solution was heated in a boiling water bath for 1.5 min and then at 80 °C for 10 min with continuous swirling. The suspension was allowed to cool slowly to room temperature and placed on ice for 15 min, and then precipitated protein was removed by centrifugation (30,000 × g, 20 min, 4 °C). The supernatant (64 ml; 165 mg of total protein) was diluted 2-fold with buffer A (20 mM Tris, pH 7.5) and applied to a Q Sepharose Fast Flow column (1.5 × 30 cm) equilibrated with buffer A. The column was developed using a linear gradient to 0.7 M KCl in buffer A over 60 min at a flow rate of 2 ml/min. The fraction collected between 41 and 51 min after initiating the gradient was dialyzed against buffer A. The retentate (22 ml; 120 mg of total protein) was filtered (0.22 µm), and portions (40 mg of total protein) were fractionated over a Mono Q column as described previously (48). The active fractions from successive runs were pooled (19 ml; 80 mg of total protein), and solid KCl was added to a final concentration of 1.0 M. The sample was filtered (0.22 µm) and applied in three separate runs to a phenyl-Superose column equilibrated with 1.0 M KCl in buffer A. The column was developed first with a linear gradient from 1.0 to 0.6 M KCl in buffer A over 15 min and then with a step gradient to buffer A at a flow rate of 1.0 ml/min. KDO8PS eluted between 4 and 9 min after development with buffer A. Fractions from successive runs were pooled, dialyzed against 2 mM Tris (pH 7.5), and then frozen in liquid nitrogen and stored at 80 °C. A total of 70 mg of protein was obtained, representing a recovery of 28 mg of recombinant protein per liter of culture. Portions of the enzyme solution were concentrated by lyophilization. The specific activity of KDO8PS determined prior to lyophilization and following reconstitution of the lyophilized enzyme is equivalent.

Preparation of Metal-depleted KDO8PS-- KDO8PS (5-6 mg/ml) was treated with 10 mM EDTA, 50 mM Tris (pH 7.5) for 2.5 h at 23 °C and then dialyzed extensively against 1 mM EDTA, 20 mM Tris (pH 7.5) at 4 °C. Aliquots were frozen in liquid nitrogen in plastic microcentrifuge tubes and stored at 80 °C. EDTA-treated enzyme solution prepared for the current study consisted of 168 µM KDO8PS and was diluted 672-fold into assays yielding a final concentration of 0.25 µM with respect to enzyme and 1.5 µM with respect to EDTA. The activity of the EDTA-treated enzyme under standard assay conditions is 0.2 units/mg, approximately 5% of the untreated enzyme (see below).

KDO8PS Activity Assay-- A typical assay, in a final volume of 100 µl, consisted of Tris acetate (100 mM, pH 7.5), PEP and A5P (3 mM each), and KDO8PS (0.25-0.33 µM). The concentrations of additional additives (metal salts or chelating agents), when included, are as indicated in the figure and table legends. Stock solutions of FeSO₄ were freshly prepared in 0.01 N HCl (using Trace Metal Grade HCl; Fisher) and were diluted >10-fold into reactions immediately before assays. Reactions were generally initiated by the addition of 5 µl of enzyme solution to the assay mixtures preincubated at 80 °C for 2 min. In some cases, EDTA-treated KDO8PS was preincubated with PEP and the metal salt at 80 °C (5-20 min) prior to initiating reactions with 5 µl of A5P solution. Reactions were incubated at 80 °C for 4 min and then quenched with the addition of either ice-cold 10% (100 µl) or 40% (15 µl) trichloroacetic acid and immediately placed on ice. The mixture was centrifuged for 2 min, and a portion (20-150 µl) of the supernatant was used for the determination of KDO8P by the periodate-thiobarbituric acid (TBA) assay. In all cases examined under these conditions, PEP and A5P remained at saturating concentrations over the 4-min incubation. One unit of activity is defined as the production of 1 µmol of KDO8P per min at 80 °C.

Periodate-TBA Assay-- The determination of KDO8P followed the general method for detection of a 3-deoxy-2-keto sugar acid and was performed essentially as described previously (50). The procedure involves oxidative cleavage of KDO8P by NaIO₄ to -formylpyruvic acid and the reaction of the latter with 2 mol eq of TBA to yield the chromophore observed at 549 nm (51-53). Briefly, the quenched reaction mixture containing KDO8P (20-150 µl) was treated with NaIO₄ (25 mM in 0.125 N H₂SO₄; 0.2 ml) for 10 min, NaAsO₂ solution (2% in 0.5 N HCl; 0.4 ml) was added to reduce excess oxidant, and then TBA solution (0.36%, pH 9.0; 2.0 ml) was added, and the samples were heated at 110 °C for 10 min. The samples were allowed to rest at ambient temperature for 5 min, and the absorbance at 549 nm was recorded (₅₄₉ = 1.03 × 10⁵ M¹ cm¹).

In general, the volume of trichloroacetic acid used to terminate reactions and the portion of the quenched solution subjected to analysis were selected in order to maintain the quantity of KDO8P in the latter between approximately 0.3 and 30 nmol. Although the sensitivity of the periodate-TBA assay, in our hands, permits a lower limit of detection of 25 pmol of KDO8P, 300 pmol of KDO8P was chosen to represent the absolute lower limit of observation to ensure accurate measurement of low enzyme activity. Under the standard activity assay conditions described above (0.25 µM KDO8PS, 4-min reaction), this amount of product would be generated by enzyme with a specific activity of ~0.1 unit/mg and would yield an A₅₄₉ of ~0.01 following development by the periodate-TBA assay.

The effect of various assay components (e.g. metal salts, chelating agents) on the periodate-TBA assay was determined as follows. A solution of KDO8P (0.21 mM) was prepared enzymatically under standard assay conditions. Portions of the quenched reaction were mixed with several concentrations of each additive and subjected to the periodate-TBA assay. The absorbance of the developed sample was compared with that of a similarly prepared control in which no additives were included. With the exception of CuSO₄, CuCl₂ and FeSO₄, no assay components were found to interfere with KDO8P determination.

Metal Analysis-- The quantitation of individual metals in enzyme and buffer samples was determined by high resolution inductively coupled plasma-mass spectrometry on a Finnigan MAT ELEMENT instrument at the W. M. Keck Elemental Geochemistry Laboratory (Department of Geology, University of Michigan). Enzyme concentrations in samples are given in the table legends. The detection limits in parts per trillion are as follows: calcium, 30; cadmium, 0.8; cobalt, 0.1; copper, 10; iron, 10; manganese, 2; nickel, 2; zinc, 10.

Samples for metal analysis were prepared by the addition of a metal salt (16 µl of a 14 mM stock solution) or water (16 µl) to EDTA-treated KDO8PS (134 µl consisting of 168 µM enzyme, 1 mM EDTA, 20 mM Tris), resulting in a solution containing 150 µM enzyme, 0.9 mM EDTA, and, when included, 1.5 mM metal salt. As such, the concentration of the metal salt in excess of EDTA was approximately 0.6 mM, representing a 4-fold molar excess of metal to enzyme monomer. Samples were incubated at 23 °C for 1 h and centrifuged, and then 100 µl was applied to a FAST Desalting Column to remove excess metal salt and EDTA from the protein-metal complex. The column was equilibrated with 10 mM Tris (pH 7.5) and developed at a flow rate of 0.5 ml/min. The fraction eluting between 3.0 and 4.5 min post-sample injection containing the entire protein fraction was collected. Complete separation of the protein fraction from excess salts is achieved under these conditions. The protein concentration of the fraction was determined, and samples were subjected to metal analysis as described above. In a separate experiment, KDO8PS (i.e. without EDTA treatment; 150 µM) was incubated in 20 mM Tris (pH 7.5) in the presence of 1 mol eq each of a mixture of metal salts (as indicated in Table III) and subjected to similar manipulations as described above.

To determine the metal content of KDO8PS directly following purification, samples of enzyme (~20 µM) were dialyzed extensively against 5 mM Tris, pH 7.5, at 4 °C. Following dialysis, the concentration of the dialyzed enzyme was determined and subjected to metal analysis as described above. A sample of the dialysate was also analyzed in order to determine the metal content of the dialysis buffer.

Analytical Methods-- Protein concentrations were estimated using the Bio-Rad Protein Assay Dye Reagent with bovine serum albumin (Sigma) serving as the calibration standard. Protein concentrations cited refer to monomer concentration and were calculated using a molecular mass of 29,734 Da for recombinant KDO8PS (48). Optical spectroscopy was performed on a Cary 3 Bio UV-Visible Spectrophotometer (Varian Associates; Sugarland, TX). When required for spectroscopic analyses, EDTA-treated KDO8PS was desalted as described above before reconstitution with metals. Protein sequence comparisons and alignments were produced by the CLUSTAL W multiple sequence alignment program (version 1.7) (54).

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Purification and Metal Content of KDO8PS-- In our initial report on KDO8PS from A. aeolicus, purification of the recombinant enzyme was achieved through a two-step protocol involving a very effective heat treatment step of the cell extract followed by anion exchange chromatography over Mono Q (48). A four-step protocol involving heat treatment of the cell extract followed by sequential chromatography over Q Sepharose, Mono Q, and phenyl-Superose was used to purify the recombinant enzyme for the current study. This last procedure gave rise to KDO8PS that appeared homogenous on SDS-polyacrylamide gel electophoresis gels (visualized with Coomassie staining) and eliminated the minor contaminants that co-purified with KDO8PS using the two-step protocol. Fractions obtained from the different stages of purification as well as concentrated solutions of purified KDO8PS (>3 mg/ml) had a pinkish-red coloration. Metal analysis of KDO8PS samples purified by the current (sample A) and the previous (sample B) method found each to contain approximately 0.4 mol eq of zinc per subunit, lesser amounts of iron, and trace amounts of various other metals (Table I). The specific activities at 80 °C of both enzyme samples are essentially equivalent (Table I), the difference possibly reflecting the small variation in metal content or the improvement in purity of the enzyme used for this study. Thus, the metal content and activity of enzyme purified by either the two- or four-step method are similar.

Influence of Metal Chelators on KDO8PS Activity-- To investigate whether the metals associated with purified KDO8PS (Table I) are important for catalytic activity, two metal chelating agents, EDTA and 1,10-phenanthroline, were examined for their abilities to inactivate KDO8PS. Dilution of KDO8PS into assays containing various concentrations of either chelator, a situation in which enzyme and chelator are concurrently exposed for a period equal to the time of assay (4 min), is accompanied by a concentration-dependent reduction in KDO8PS activity (Fig. 1A). EDTA inactivates KDO8PS more effectively than 1,10-phenanthroline at each concentration examined. Time-dependent inactivation by EDTA is also observed and appears to follow biphasic kinetics in which an initial fast phase is followed by a relatively slower phase of inactivation (Fig. 1B).

View larger version (19K): [in this window] [in a new window]   Fig. 1.   Inhibition of KDO8PS activity by metal chelators. KDO8PS activity (0.33 µM enzyme) was measured at 3 mM PEP and A5P in 100 mM Tris acetate (pH 7.5) for 4 min at 80 °C in the presence of various concentrations of either EDTA () or 1,10-phenanthroline () (A) or after incubation of KDO8PS (6.6 µM) with either no EDTA () or 0.01 mM (), 0.1 mM (), or 1.0 mM () EDTA in 50 mM Tris (pH 7.5) at 23 °C for various times (5-60 min) prior to dilution of the enzyme 20-fold into an assay containing the same concentration of EDTA as in the preincubation (B). Activities are expressed as a percentage of the activity in the absence of added chelating agent.

The effectiveness of EDTA in inactivating KDO8PS was exploited in order to secure metal-free enzyme for subsequent studies. To this end, concentrated enzyme (~0.2 mM) was treated with 10 mM EDTA followed with dialysis against buffer containing 1 mM EDTA. KDO8PS activity is typically reduced to <0.5 units/mg when treated in this manner. For these studies, EDTA-treated enzyme activity is 0.2 units/mg, an activity roughly 5% that of the untreated enzyme and approximately 0.4% that of the maximum activity observed in the presence of various divalent metals (see below; Fig. 2 and Table II). When assayed as described, an activity of 0.2 units/mg is twice the level set as the lower limit of observation and is reliably determined using the periodate-TBA assay (see "Experimental Procedures"). At present, it is unclear whether the residual activity of the EDTA-treated enzyme is a result of trace metals present in assays (contributed by the assay components and plastic ware), the inability of EDTA treatment to render the enzyme 100% metal-free, or true residual activity expressed by the apoenzyme.

View larger version (26K): [in this window] [in a new window]   Fig. 2.   Concentration dependence of KDO8PS activation by divalent metals. The activity of EDTA-treated KDO8PS (0.25 µM) was determined at 3 mM PEP and A5P, in 100 mM Tris acetate (pH 7.5) for 4 min at 80 °C in the presence of various concentrations of CdCl2 (), MnCl2 (), CoCl2 (), NiCl2 (), or CaCl2 (×) (A) or CuSO4 (), CuCl2 (), ZnSO4 (), ZnCl2 (), or MgCl2 () (B). The activity determined in the absence of added metal was 0.2 ± 0.1 units/mg. All reactions were initiated with enzyme, and activities were determined in triplicate. Error bars indicate S.D. values. U, units

View this table: [in this window] [in a new window]   Table II Activation of KDO8PS following preincubation with metals The activity of EDTA-treated KDO8PS was assayed as described in the legend of Fig. 2, except KDO8PS, PEP, and metal salt were incubated for 10-15 min at 80 °C before initiating reactions with A5P. +, activities determined following preincubation of enzyme with metal; , activities determined without preincubation.

Dependence of KDO8PS Activity on Divalent Metals-- KDO8PS activity is readily restored to the EDTA-treated enzyme by supplementing reaction mixtures with divalent metal salts, including CaCl₂, CdCl₂, CoCl₂, CuCl₂, CuSO₄, FeSO₄, MgCl₂, MnCl₂, NiCl₂, ZnCl₂, and ZnSO₄. To provide an initial assessment of the metal activation, activity was determined under conditions in which EDTA-treated KDO8PS was used to initiate reactions containing various concentrations (0.001-10 mM) of the individual metals. Results from this study (Fig. 2) indicate that metal activation is markedly dependent on both the type and concentration of metal examined. Here, the maximum activity supported by each metal is as follows (in units/mg): Zn²⁺, 4.5; Mg²⁺, 4.6; Cu²⁺, 8.0; Ca²⁺, 15; Co²⁺, 27; Ni²⁺, 28; Cd²⁺, 50; Mn²⁺, 51 (Fig. 2).

The activities represented in Fig. 2 include any time dependence in metal binding. Additional experiments indicate that the reaction rate is not linear in all cases, increasing over the duration of the assay before becoming constant. When observed, the magnitude of this lag period demonstrates a strong inverse relationship to metal concentration (not shown). These observations suggest that the metal ions differ with respect to their rates of association (k_on) to the enzyme. However, if EDTA-treated enzyme is preincubated with metal and PEP for 5-15 min at 80 °C prior to initiating reactions with A5P, the rate of product formation appears constant throughout the course of the reaction. The reaction rate of KDO8PS as isolated remains constant with time regardless of whether the enzyme is preincubated with PEP.

A comparison of metal activation determined under conditions of preincubation with that in which enzyme is not preincubated with metal (Table II) illustrates that in general the activities measured under both conditions are similar. The most significant differences are observed for activation by Ca²⁺, Mg²⁺, and Ni²⁺ at each concentration examined. Preincubation has less of an effect with Co²⁺, Cu²⁺, Mn²⁺, and Zn²⁺, especially at 0.1 and 1 mM metal ion. Activation by Cd²⁺ appears to be independent of preincubation (Table II).

Due to inherent limitations associated with the discontinuous assay at 80 °C, it remains difficult to assess the reaction rate and linearity at the very onset of reactions (<30 s). Although continuous spectrophotometric assays for PEP consumption (55) or inorganic phosphate liberation (56) are available and are more suited to measurement of initial velocities, the elevated temperatures required for KDO8PS activity restrict their use. The Arrhenius plot of ln(k) versus 1/T, k representing the specific activity of the purified enzyme at temperatures (T) between 30-100 °C (48), demonstrates two linear regions between 30 and 60 °C and between 70 and 95 °C (not shown). The energy of activation values calculated from the respective slopes are approximately 16 kcal/mol (30-60 °C) and 10 kcal/mol (70-95 °C). These observations suggest a change in the enzyme mechanism/structure with temperature. As such, and considering the growth temperature optima of 80-90 °C for A. aeolicus (49), enzyme reactions were performed at 80 °C.

The assessment of activation by Fe(II)SO₄ is complicated by two factors. First, the inherent instability of ferrous iron in aqueous solution (57) makes uncertain the effective concentration of Fe(II) present in reactions and available to the enzyme. Second, increasing concentrations of FeSO₄ in samples of KDO8P subjected to the periodate-TBA assay result in a progressive, nonlinear decrease in the A₅₄₉ of the developed samples (not shown). This interference² requires application of concentration-specific, empirically derived correction factors to the measured values in order to provide reasonable estimates of actual enzymatic activity. Notwithstanding these complications, preliminary results clearly indicate that FeSO₄ stimulates enzyme activity. Although only a qualitative assessment of FeSO₄ activation is warranted at this time for the reasons given above, the response of EDTA-treated KDO8PS to this cation appears unique with respect to the other metals. When the concentration of FeSO₄ added to assays is between 0.01 and 1 mM, enzyme activity varies between ~3 and 6 units/mg in an apparent multimodal manner (not shown). However, inclusion of higher concentrations of FeSO₄ produced an approximate 10-fold increase in activity, reaching an apparent maximum of ~50 units/mg at concentrations between 3-6 mM. The latter activity is similar to the maximum activity observed for both CdCl₂ and MnCl₂. Additional studies designed to measure activity under reducing or anaerobic conditions at 70-80 °C will be required to fully evaluate the effect of Fe(II) on enzyme activation.

Increasing concentrations (>1 mM) of certain metals lead to subsequent inhibition (Fig. 2). The activity of KDO8PS in the presence of 10 mM Cd²⁺, Cu²⁺, Mn²⁺, or Zn²⁺ is approximately 15-30% that of the respective maximum observed in the presence of each (Fig. 2). Preincubation of metal with enzyme produced a slight increase in the inhibition (not shown). Similar inhibition is not observed with Ca²⁺, Co²⁺, Mg²⁺, or Ni²⁺ under the conditions examined. Consequently, additional experiments intended to identify the basis for the observed inhibition were performed. Preincubation of A5P and PEP in the presence of 10 mM CdCl₂, MnCl₂, ZnSO₄, or CuSO₄ at 80 °C for various times (2-15 min) prior to initiating the reaction with enzyme did not affect the observed activity (not shown). KDO8PS preincubated with a 10 mM concentration of the inhibitory metals for 15 min at 80 °C retains full activity when diluted 10-fold into metal-free assays. In addition, the preincubated enzymes are indistinguishable from nontreated enzyme by SDS-polyacrylamide gel electophoresis analysis (not shown). Inhibition is observed whether CuCl₂ or CuSO₄ and ZnCl₂ or ZnSO₄ serve as the activating salt (Fig. 2); furthermore, the inclusion of 20 mM NaCl in assays with either 1 or 10 mM metal (each metal salt was examined) either did not alter activity or resulted in a nominal increase in the measured activity (~10% for 1 mM Co²⁺, Ni²⁺; ~5% for 1 mM Ca²⁺, Cd²⁺, Mn²⁺, and 10 mM Co²⁺, Ni²⁺). Ionic strength effects or possibly the influence of chloride ion behaving as a chaotrope could impart this increase in activity for a variety of reasons. From these observations, it would appear that certain cations at high concentrations inhibit KDO8PS activity, although the basis for this phenomenon remains unclear.

Spectroscopic Properties of KDO8PS-- The UV-visible electronic absorption spectrum of KDO8PS as isolated following purification reveals a broad band centered at 505 nm (Fig. 3). The addition of increasing concentrations of either NiCl₂ or CoCl₂ to the enzyme attenuates the A₅₀₅, reducing the intensity by 30 and 85% after the addition of 5 eq of NiCl₂ and CoCl₂, respectively, without any additional changes to the spectra (not shown). In contrast, the addition of CuSO₄ to the purified enzyme eliminates the A₅₀₅ with the concomitant appearance of an intense peak at 375 nm (₃₇₅  3000 M¹ cm¹) and a broad band of lesser intensity centered at approximately 630 nm (₆₃₀  150 M¹ cm¹) (Fig. 3). Titration of EDTA-treated KDO8PS with CuSO₄ increases the A₃₇₅ up to the addition of 1 mol eq, after which further titration with up to 5 mol eq of CuSO₄ does not affect the intensity of the peak (Fig. 3, insert). The lack of distinguishing spectral features of the EDTA-treated enzyme is consistent with the removal of bound metals (Fig. 3).

View larger version (29K): [in this window] [in a new window]   Fig. 3.   Absorption spectra of KDO8PS. Spectra of KDO8PS (75 µM) as purified (thick curve) and after the addition of 375 µM CuSO4 (dashed curve) are shown. The thin curve is that of EDTA-treated KDO8PS (25 µM) after removal of excess EDTA by gel filtration. Inset, titration of desalted EDTA-treated KDO8PS (15 µM) with various molar equivalents (as indicated) of CuSO4. The spectrum taken in the presence of 1 mol eq of CuSO4 is indicated by the thick curve. All spectra were acquired at 35 °C in 10 mM Tris (pH 7.5), and those with CuSO4 were acquired subsequent to a 15-min incubation with enzyme.

Evidence for Stable Enzyme-Metal Complexes-- In order to determine whether metal activation involves formation of a stable complex with KDO8PS, the binding stoichiometry for several metals was investigated. For analysis of metal binding, EDTA-treated KDO8PS was incubated with an excess of metal salt and then separated from free metal by gel filtration, and the protein fraction was analyzed for metal content (Table III).

In the case of Zn²⁺, Cu²⁺, and Cd²⁺, metals producing maximal activation at equally low concentrations (0.1 mM; Fig. 2), KDO8PS is found to bind approximately 0.5 mol eq of metal per enzyme subunit (Table III). Under the same experimental conditions, KDO8PS binds approximately 0.25 mol eq of Mn²⁺, suggesting a weaker association of Mn²⁺ with enzyme than observed for Zn²⁺, Cu²⁺, and Cd²⁺. Relative to the latter, higher concentrations of Mn²⁺ are required to achieve maximum activation of KDO8PS (~1 mM; Fig. 2). The highest metal binding stoichiometry is observed for KDO8PS incubated for 1 h with FeSO₄, the enzyme found to contain 1.21 mol eq of iron per enzyme subunit subsequent to gel filtration (Table III). However, the activity of the iron-enzyme complex (Table III) is significantly lower than the level of Fe(II) activation discussed earlier. These observations may reflect (partial) oxidation of bound Fe(II) to some form of enzyme-coordinated Fe(III) over the time interval of the experiment, the latter assumed ineffective in enzyme activation but not necessarily in binding to the enzyme. Other studies show that Fe(II) activation decays relatively quickly with time (<30 min) to ~2 units/mg if EDTA-treated enzyme and FeSO₄ are preincubated at 23 °C prior to initiating the reaction with substrates (not shown).

In the cases above, EDTA-treated KDO8PS was incubated in the presence of a 4-fold molar excess of metal to enzyme subunit. When KDO8PS (as isolated) is challenged at once with 1 eq each of the above metals, the enzyme is recovered containing primarily 0.5, 0.3 and 0.1 mol eq of Cd²⁺, iron, and Zn²⁺, respectively, per subunit (Table III). Under these conditions, it appears that KDO8PS preferentially binds Cd²⁺. The activity of KDO8PS and the stoichiometry of Cd²⁺ binding are essentially equal to those observed when the enzyme is incubated exclusively with CdCl₂ (Table III).

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

This investigation has identified for the first time the existence of a metallo-KDO8PS, that from A. aeolicus. Our initial characterization of A. aeolicus KDO8PS has shown that the recombinant enzyme demonstrates thermal properties (including temperature optimum and thermostability) representative of other enzymes isolated directly from thermophilic microorganisms (48). In the current study, several lines of evidence are presented substantiating the metal requirement of KDO8PS including the following. (i) The active enzyme is isolated with bound metal; (ii) metal chelating agents inactivate the purified enzyme; (iii) divalent metal ions restore a broad range of activities to enzyme rendered inactive with EDTA; (iv) metal ions form stable complexes with KDO8PS; and (v) certain metals form a chromophoric complex with KDO8PS that displays characteristic spectral properties.

Purified preparations of KDO8PS isolated using slightly different methodologies contain similar and substoichiometric amounts of bound zinc and iron (Table I). The foremost presence of these two metals is probably a consequence of the metal availability in the culture medium. Heterologous expression of KDO8PS is achieved from cultures maintained in 2× YT broth, an enriched medium similar in composition to Luria-Bertani broth (58), which has been shown to contain significant concentrations of iron and zinc (~13 µM) and lesser amounts (<0.4 µM) of other metals (41). A comparable metal content between the two media would account for the predominance of zinc and iron isolated with the enzyme. Other observations suggest that these metals remain associated with the enzyme throughout the process of purification. Accordingly, EDTA-treated enzyme reconstituted with zinc or iron at concentrations comparable with that suspected in the culture medium demonstrates an activity akin to the enzyme as isolated.

Both EDTA and 1,10-phenanthroline are effective in reducing KDO8PS activity (Fig. 1A). Although the possibility remains that nonchelating properties of EDTA and 1,10-phenanthroline contribute to the observed inhibition, additional evidence presented herein indicates that the loss of activity is probably a direct consequence of metal chelation. This type of inhibition can occur when the chelating agent either reacts with the metal as it dissociates from the enzyme or forms a ternary enzyme-metal-chelator complex that is itself inactive or collapses to apoenzyme and the chelated metal (59, 60). In the case of the latter, the chelating agent catalyzes the removal of the metal per se, and inhibition is often rapid. EDTA is generally considered to rapidly sequester free metals following spontaneous dissociation of the metalloenzyme (59). The time dependence of EDTA inhibition (Fig. 1B) is consistent with this type of mechanism but does not preclude the slow dissociation of an active ternary complex. Considering that the stability constant for the zinc chelate of EDTA is greater than that of 1,10-phenanthroline (59), the superior inhibition afforded by EDTA is consistent with removal of zinc from KDO8PS (Table I).

Prolonged incubation of KDO8PS with 10 mM EDTA followed by dialysis against 1 mM EDTA typically reduces enzyme activity to less than 10% of the initial activity which corresponds to less than 1% of the maximum activity (50 unit/mg). In comparison with the purified enzyme, which demonstrates an absorption at 505 nm, the spectrum of the EDTA-treated enzyme is not distinctive (Fig. 3). Metal analysis of the EDTA-treated enzyme confirms removal of the zinc and iron from the enzyme (Table III). In addition, the oligomeric state of EDTA-treated KDO8PS and the enzyme as isolated, as evidenced by analytical gel filtration, are identical and are unchanged by the presence or absence of EDTA in the chromatographic buffer (not shown). Collectively, the above observations correlate the loss of enzyme activity following EDTA treatment of KDO8PS with removal of enzyme-bound metal ions.

The spectral properties of KDO8PS lend additional evidence in support of an enzyme-coordinated metal. The broad absorption band at 505 nm displayed by the enzyme as isolated (Fig. 3) is suggestive of a ligand-to-metal charge transfer electronic transition of Fe(III) (61). Assuming 0.3 mol eq of iron (Table I), the extinction coefficient at 505 nm (  1300-1400 M¹ cm¹) is also characteristic of this type of ligand-to-ferric iron transition (61). The decrease in A₅₀₅ following the addition of excess Ni²⁺ or Co²⁺ suggests displacement of Fe(III) from the enzyme by the former. Several enzymes shown to contain Co²⁺ bound in a tetrahedral coordinated geometry often display peaks between 500 and 800 nm (  300 M¹ cm¹) (62, 63). The absence of new absorbance features in the visible region subsequent to Co²⁺ addition could indicate octahedral coordination of Co²⁺. The relative intensities and absorbance maxima resulting from the addition of Cu²⁺ to KDO8PS are suggestive of a ligand-to-Cu²⁺ charge transfer and ligand field transition for the 375- and 630-nm peaks, respectively (64, 65).

The metal cofactor requirement of A. aeolicus KDO8PS is clearly demonstrated by the ability of a variety of divalent cations to restore activity to the EDTA-inactivated enzyme (Fig. 2; Table II). Although the mechanism of metal activation has not been examined in detail, some inferences concerning the activation of KDO8PS can be derived from the observations recorded in this study. Using the concentration of free metal producing half-maximal activation, the apparent relative affinity of KDO8PS for the metals is Zn²⁺  Cd²⁺  Cu²⁺ > Mn²⁺  Co²⁺ > Ni²⁺/Ca²⁺/Mg²⁺. In addition to metal affinity, the 10-fold range in maximum activity seen with the individual metals indicates that KDO8PS activity is determined, in part, by the specific properties of the activating metal. To a first approximation, the larger the ionic radius of the divalent cation the greater the activation, assuming octahedral geometry for the metal (65). It is noteworthy that the highest activation (~50 units/mg) is achieved with both Mn²⁺ and Cd²⁺ (under appropriate conditions, iron also appears to activate KDO8PS to a similar level), suggesting that at this maximal rate, a step independent of metal identity becomes rate-limiting. In the case of E. coli KDO8PS, product release has been shown to be rate-limiting (2).

Incubation of EDTA-treated KDO8PS with several metals under identical conditions produces stable complexes of different yet related stoichiometries (Table III). To a first approximation, binding of 0.25 (manganese), 0.5 (zinc, copper, cadmium), or 1 (iron) mol eq of metal per subunit is observed in the complexes. Several observations provide convincing evidence that A. aeolicus KDO8PS exists as a tetramer in solution: the structure of E. coli KDO8PS determined for two crystal forms reveals a tightly packed homotetramer with four individual active sites (66, 67); the oligomeric structure of A. aeolicus KDO8PS is equivalent to the E. coli enzyme in solution (48); and the sequence identity between the two enzymes (46%; see below) is significant. Therefore, it is reasonable that each subunit of the A. aeolicus KDO8PS tetramer will itself contain a lone active site that binds a single metal ion. In relation to the KDO8PS tetramer, the stoichiometries noted above translate to specific binding of one, two, or four metal ions. This apparent all or none relationship regarding metal binding to individual subunits predicts that the binding affinities of the individual subunits are not equal. More detailed kinetic and thermodynamic analyses will be required to address any correlations between KDO8PS activity and metal binding stoichiometry.

As highlighted in the Introduction, the available mechanistic information along with the clear parallelism in the overall reactions catalyzed by KDO8PS and DAHPS strongly suggests that they follow a closely related reaction pathway. In essence, the foremost differences in the two reactions, at least in the case of the E. coli enzymes, are the metal cofactor requirement and the length of the phosphorylated monosaccharide substrate (E4P versus A5P). Recent observations on the E. coli phenylalanine-sensitive isoform of DAHPS have extended its kinship to KDO8PS. In addition to E4P, DAHPS(Phe) can also catalyze the metal-dependent condensation of PEP with A5P to form KDO8P (40). This raises the question as to whether the A. aeolicus enzyme under current investigation is itself a DAHPS capable of condensing A5P with PEP. However, incubation of KDO8PS with PEP and E4P under a wide variety of conditions fails to provide any evidence that E4P serves as a substrate and indicates that the A. aeolicus enzyme functions as a true, metal-dependent KDO8PS. Therefore, from a mechanistic position, our current understanding of the correspondent reactions appears insufficient to account for the necessity of a metal cofactor in the reaction catalyzed by E. coli DAHPS and A. aeolicus KDO8PS and the lack of such a requirement in the case of E. coli KDO8PS. More recent studies have also indicated the lack of a metal ion requirement for recombinant KDO8PS from Salmonella typhimurium³ and Neisseria gonorrhoeae (69).

Primary sequence comparisons also fail to provide any distinguishing correlation between enzymatic reaction and metal requirement. Sequence alignment of A. aeolicus and E. coli KDO8P synthases produces an overall identity of 46% with extensive stretches of similarity (approximately 70% total homology), the latter predominantly reflecting conservative replacements in hydrophobic residues. Nonconservative substitutions include a higher proportion of charged residues in A. aeolicus KDO8PS, which may contribute to the enzyme's relative thermostability (not shown). In contrast, neither the metallo- nor nonmetallo-KDO8PS shows any significant sequence identity with any of the three E. coli DAHPS isozymes (~15-20% for each pairwise alignment). Less than 10% identity is observed for a multiple alignment between A. aeolicus KDO8PS, E. coli KDO8PS, and E. coli DAHPS(Phe) (not shown). The weak homology between representative members of the KDO8P and DAHP synthase families has been previously recognized (42).

Additional observations further confound the discrepancy of a metal cofactor. Despite the poor level of sequence identity between KDO8PS and DAHPS, the crystal structures of E. coli KDO8PS (66) and DAHPS(Phe) (28) bear a striking resemblance in both the overall core structures of the two enzymes and in the composition and architecture of their active sites. The structure of the complex of DAHPS(Phe) with PEP and Pb²⁺ clearly locates an active site position for the metal ion in close proximity to PEP and indicates that Cys⁶¹, His²⁶⁸, Glu³⁰², and Asp³²⁶ serve as ligands for the Pb²⁺ ion (28). These residues are virtually superimposable with Asn²⁶, His²⁰², Glu²³⁹, and Asp²⁵⁰ in the active site of E. coli KDO8PS, which are themselves homologous to Cys¹¹, His¹⁸⁵, Glu²²², and Asp²³³ in the alignment with A. aeolicus KDO8PS. Nevertheless, close examination and comparison of the crystal structures provides no additional insight into the role of the metal. Additional similarities exist between A. aeolicus KDO8PS and E. coli DAHPS regarding the metal cofactor in solution. EDTA treatment of the DAHPS isozymes is equally effective in rendering these enzymes inactive, and the hierarchy of activities supported by various metals (41) is very similar to the current observations on A. aeolicus KDO8PS. In addition, the spectroscopic features of the metal-enzymes bear a strong resemblance, particularly in the case of Cu²⁺ (41, 68). As reported herein, A. aeolicus KDO8PS absorbs at 375 nm in the presence of Cu²⁺. A similar absorption maximum, but at approximately 350 nm, is observed for DAHPS reconstituted with Cu²⁺. As suggested by the crystal structures, this analogy indicates a comparable ligation environment around Cu²⁺ in solution.

In conclusion, the finding that A. aeolicus KDO8PS is a metalloenzyme is of considerable significance. The genus Aquifex is representative of the deepest branching bacterial lineage in existence. The strong sequence homology with E. coli KDO8PS yet the metal cofactor characteristics shared in common with DAHPS indicate that A. aeolicus KDO8PS may be related to both, suggesting that the two enzyme families arose from a common ancestor. In addition, the identification of a metallo-KDO8PS provides, for the first time, evidence pointing to the possible existence of two KDO8PS classes, a metal-independent class and a metal-dependent class. This information is important in interpreting not only the role of the metal but also the overall reactions catalyzed by KDO8PS and DAHPS. Further investigation of A. aeolicus KDO8PS in addition to the study of KDO8P synthases from other diverse organisms should provide key information leading to an even greater understanding of this enzyme family.

	ACKNOWLEDGEMENTS

We thank Dr. Ted Huston (Department of Geology, University of Michigan) for performing the metal analysis studies.

	FOOTNOTES

^* This work was supported by National Institutes of Health Grant GM 53069 (to R. W. W) and a Natural Sciences and Engineering Research Council of Canada postdoctoral fellowship (to H. S. D).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.

To whom correspondence should be addressed: College of Pharmacy, 428 Church St., Ann Arbor, MI 48109-1065. Tel.: 734-764-7366; Fax: 734-763-5633; E-mail: rww@umich.edu.

Published, JBC Papers in Press, May 15, 2000, DOI 10.1074/jbc.M000133200

² A similar approach as taken with FeSO₄ verified that none of the other metal salts, EDTA, 1,10-phenanthroline, NaCl, Na₂SO₄, PEP, A5P, or buffers have any affect on the determination of KDO8P. KDO8P samples containing CuSO₄ or CuCl₂ accumulated a precipitant proportional to the concentration of Cu²⁺ following development by the periodate-TBA assay, increasing the A₅₄₉ due to scattering. This complication could be averted by centrifugation of the sample prior to absorbance measurements.

³ W. P. Taylor, G. Sheflyan, and R. W. Woodard, submitted for publication.

	ABBREVIATIONS

The abbreviations used are: PEP, phosphoenolpyruvate; KDO8P, 3-deoxy-D-manno-octulosonic acid 8-phosphate; KDO8PS, 3-deoxy-D-manno-octulosonic acid-8-phosphate synthase; DAHP, 3-deoxy-D-arabino-heptulosonic acid 7-phosphate; DAHPS, 3-deoxy-D-arabino-heptulosonic acid-7-phosphate synthase; A5P, D-arabinose 5-phosphate; TBA, thiobarbituric acid; KDO, 3-deoxy-D-manno-octulosonic acid; E4P, D-erythrose 4-phosphate.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg