Biochemistry 1993, 32, 799-805
799
Escherichia coli Fumarase A Catalyzes the Isomerization of Enol and Keto Oxalacetic Acid? Dennis H. Flint Central Research and Development Department, E. I. du Pont de Nemours and Company, Experimental Station, Wilmington, Delaware I9880 Received June 9, 1992; Revised Manuscript Received September 28, 1992
ABSTRACT: Fumarase A, a product of the f u m A gene of Escherichia coli, has been found to catalyze the isomerization of enol to keto oxalacetic acid (OAA) in addition to catalyzing the fumarase reaction. The kat/& for the isomerization is almost identical to that for the fumarase reaction. The isomerization reaction apparently takes place at the same active site as the fumarase reaction since both reactions show a similar sensitivity to inactivation by 0 2 , both reactions are strongly inhibited by 2-hydroxy-3-nitropropionate, and the isomerization reaction is inhibited by fumarate and malate. The isomerization requires the presence of a [4Fe-4S] or [ 3 F e 4 S ] cluster, perhaps for structural rather than catalytic reasons. Hydration of enol OAA to the gem diol has been ruled out as a possible mechanism of isomerization on the basis of the preservation of the oxygen on carbon 2 and the position of protonation on carbon 3. The isomerization is not stereospecific in the position of protonation at carbon 3 but appears to be stereoselective, with protonation preferentially occurring in the 3-pro-S position. Porcine fumarase, isopropyl malate isomerase, and dihydroxyacid dehydratase do not catalyze this isomerization. Fumarase A and aconitase, two enzymes with 4Fe-4S clusters that bind a linear 4-carbon dicarboxylic acid moiety in the trans conformation during their normal hydro-lyase reaction, do catalyze this isomerization.
It has been known for some time that an enzymatic activity exists in several organisms which catalyzes the conversion of enol to keto oxalacetic acid (herein referred to as OAA) (Annett & Kosicki, 1969; Wesenberg et al., 1976). The Nomenclature Committee of the International Union of Biochemistry has given this activity the name OAA ketoenol-isomerase (herein referred to as OAAKE isomerase) and the numerical designation EC 5.3.2.2. None of the protein(s) responsible for the OAAKE isomerase activity had been identified until recently, when OAAKE isomerase activity from bovine heart mitochondrial matrix was shown to be associated with proteins migrating with molecular masses of 80 and 37 kDa (Belikova et al., 1988). Subsequently, the protein migrating with a mass of 8OkDa hasbeenshowntobeaconitase(Belikovaetal., 1989), but the identity of the 37-kDa protein remains unknown. In addition, an OAAKE isomerase activity from porcine kidney with a molecular mass of 55 kDa has been reported (Johnson et al., 1986). The relation between the 55-kDa porcine kidney OAAKE isomerase and the 80-kDa (aconitase) and 37-kDa beef heart mitochondrial proteins is unknown. The porcine kidney OAAKE isomerase has been shown to lack stereospecificity in the protonation of carbon 3 of enol OAA (Johnson etal., 1986). Thelackofstereospecificityisunusualinenzymecatalyzed reactions. The discoverythat aconitase exhibits an OAAKE isomerase activity was unexpected and is of unknown physiological significance. Aconitase is a member of the hydro-lyase class of enzymes. Native bovine aconitasehas been shown to contain a [ 4 F d S ] cluster which activates the hydroxyl group of the substrate for elimination and then addition in the interconversion of citrate to isocitrate (Emptage, 1988). Aconitase is easily oxidized to a form that readily loses one Fe to give a [3Fe-4S] cluster. This form of aconitase is inactive in t This iscontribution 6180from thecentral Research andDevelopment Department, E. I. Du Pont de Nemours & Co., Wilmington, DE 19880.
0006-2960/93/0432-799$04.00/0
catalyzing the conversion of citrate to isocitrate. It was surprising then to find that not only native aconitase in the [4Fe4S] form but also oxidized aconitase in the [3Fe-4S] form was active in the isomerization of enol to keto OAA (Belikova et al., 1989). This result brings up the question of what involvement, if any, the Fe-S cluster of aconitase has in the OAAKE isomerization reaction since both the [ 4 F e 4S] and [3Fe-4S] cluster forms catalyze the OAAKE isomerase reaction at similar rates. We report here that Escherichia coli fumarase A, another member of the hydro-lyase class which contains a [4Fe-4S] cluster (Flint et al., 1992), also exhibits OAAKE isomerase activity. The details of our investigations on the OAAKE isomerase activity of E. coli fumarase A are described in this paper.
MATERIALS AND METHODS OAA was obtained from Sigma as item number 0 4126 and is identified in their 1990 catalog as cis-hydroxymaleic acid. 2H20 and H P 0 were obtained from MSD Isotopes. 2-[180]-OAA was made by the exchanging the 2-oxygen of 2-[l6O]-0AA in H2**0for 30 min. 2-Hydroxy-3-nitropropionate was prepared by us. All other reagents including porcine fumarase were purchased from Sigma. Enzyme Source. Fumarase A is the gene product of the E. colifumAgene (Woods et al., 1988;Yumoto & Tokushige, 1988) and was isolated from E. coli strain JRGl905 as previously described (Flint et al., 1992). This strain was derived from E. coli strain JH400, a fumAC fumB triple deletion mutant, by addition of a plasmid (pGS57) carrying a copy of the fumA gene. The fumarase A used in these studies was purified and stored under strict anaerobic conditions. It was 8&90% pure as judged by gel electrophoresis. This purified fumarase A rapidly loses activity in air. Except where otherwise noted, all fumarase A samples used in these experiments were with active native enzyme (defined as containing an intact [4Fe-4S] cluster). 0 1993 American Chemical Society
800 Biochemistry, Vol. 32, No. 3, 1993 Enzyme Assays. The assays were done in 10mM phosphate buffer, pH 7.5, because the spontaneous rate of isomerization of OAA is conveniently low in this buffer. Enol OAA solutions (80 mM) were prepared in diethyl ether. The assays were started by the addition of aliquots of the stock enol OAA solution directly into assay buffer. The small amount of diethyl ether added to the assay solution in this way was shown in control experiments to not affect the activity of fumarase A in the fumarase or OAAKE isomerase reaction. The enol to keto isomerization of enol OAA was followed by measuring the change in absorbance at 260 nm when the concentration of OAA was below 0.5 mM and at 290 nm when the concentration was 0.5 mM or above. The difference in extinction coefficients between the enol and keto forms of OAA was found to be 1 X lo4 M-I at 260 nm and 2 X lo3 M-' at 290 nm. Sufficient enzyme was added in each case so the enzymatically catalyzed rate of isomerization was at least 10-fold above the spontaneous background rate. The rate of enzyme-catalyzed isomerization was determined by subtracting the spontaneous background rate from the overall rate. GC Mass Spectral Analysis. GC/MS analysis was used to determine the content of 2Hand l80in samples of malate. Samples of malate were derivatized in acidic methanol to form the dimethyl ester. The dimethyl ester was chromatographed and a high-resolution mass spectrum was obtained using a 30-m DB-Wax column on a Varian Vista 600 gas chromatograph (temperature programmed from 60 to 220 OC at 12 OC/min) coupled to a VG Micromass 7070-HS mass spectrometer. A major fragment in the mass spectrum of the dimethyl ester of malate is generated by the loss of a carboxyl ester. This fragment has a m / e of 103 in the case of the dimethyl ester of [3-2Ho]malate,m / e of 104 in the case of the dimethyl ester of [3-2Hl]malate, and m / e of 105 in the case of the dimethyl ester of [3-2H2] and in the case of [2-'*O]malate. In each case the data obtained from the GC/MS were corrected to account for the natural abundance of I3C, leg,and 2H. Determination of Stereochemical Specificity. Porcine fumarase, which specifically removes the 3-pro-R proton from 2-(S)-malate (Hill & Teipel, 1971), was used to determine the location of 2Hin the 3-position of labeled sample of malate. Oneunit of porcine fumarase was added to half of each labeled sample in 10 mM phosphate buffer prepared with IH20. Incubation was continued long enough to catalyze the conversion of 100 times the amount of malate (to fumarate) present in the samples. These samples were then derivatized and analyzed by GC/MS to determine the 2H content.
RESULTS Determination of Isomerase Kinetic Constants and Commonality of Active Site. OAA in the crystalline form and in solution in solvents of low polarity exists predominantly in the enol form, whereas in solvents of high polarity, OAA exists predominantly in the keto form. The forms can distinguished from each other by their UV absorbance spectra (Gruber et al., 1956; Banks, 1961; Hess & Reed, 1972). Addition of OAA to water from the solid form or from a solution of OAA in a solvent of low polarity is followed by the spontaneous isomerization of the enol to the keto form. The progress of this isomerization can be monitored by the changes in the UV. The rate of isomerization of enol to keto OAA has been reported to be acid/base catalyzed (Emly & Leussing, 1981) and in our hands is considerably faster at pH 7.5 in 100 mM than in 10mM phosphate buffer. In 10mM phosphate buffer, the spontaneous isomerization took several minutes to reach equilibrium.
Flint
1 0
IO
5
15
20
1/[S] (mMolar)
FIGURE 1: Velocity-I vs [enol OAAI-l for the isomerase activity of a fumarase A preparation.
OAAKE Isomerase Activity
0.6 -
0.4
-
0.2
-
Fumarase Activity
I
0
5
10
15
20
Time (min) FIGURE 2: Inactivation of the isomerase and fumarase activity of fumarase A when exposed to air. The addition of aliquots of a purified fumarase A preparation greatly accelerated the rate isomerization as monitored in the UV; however, the final UV absorbance spectrum of the OAA solution after equilibrium was reached was not affected (except for a very small effect due to addition of protein). This demonstrates a component in the fumarase A preparation can act as a catalyst for the isomerization of enol to keto OAA. The isomerization of enol to keto OAA catalyzed by the fumarase A preparation follows saturation kinetics typical of enzymatic reactions. A double-reciprocal plot of the inverse rate of isomerization vs inverse substrate concentration is shown in Figure 1. Since the fumarase A preparation used in these experiments was 80-90% pure, we attempted to determine if the OAAKE isomerase activity was associated with fumarase A or some other constituent in the sample. Three lines of evidence indicated that the fumarase and isomerase reactions are catalyzed by the same enzyme. First and most definitive is the very similar kinetics of inactivation of both the fumarase and isomerase activity of fumarase A preparation in air (see Figure 2 and associated discussion). The fumarase activity of fumarase A is known to be air-sensitive due to the oxidation of the F e S cluster by 0 2 (Flint et al., 1992), and the isomerase activity shows a similar sensitivity. Second, we have found that the possible transition-state analog, 2-hydroxyl-3-nitropropionic acid dianion, is a strong inhibitor of the fumarase activity of fumarase A preparations (Flint, unpublished work), and we have also found it is a strong inhibitor of the isomerase reaction. For example, in the presence of 10 pM 2-hydroxy3-nitropropionic acid dianion the initial rate of enzymecatalyzed isomerization with 400 pM enol OAA as substrate is inhibited 96%. Third, in the presence of an equilibrium mixture of fumarate and malate in which the concentration of these two species adds up to 10 mM, the rate of enzyme-
Biochemistry, Vol. 32, No. 3, 1993 801
Fumarase A Isomerizes Enol to Keto Oxalacetic Acid Table I: Kinetic Constants for Hydro-Lyase and Isomerase Activities of Fumarase A, Porcine Fumarase, and Aconitase
Scheme I
E. coli porcine bovine aconitase" fumarase A fumarase" (for fumarate) (for fumarate) (for cis-aconitate) Hydro-Lyase Activity Km (MI
6X
lo4
kcat (8-I) 3 x 103 kca,/Km(M-I s-I) 5 X lo6
5x10-6
3 x 102 6X
lo7
8 X 10" 4 x 101 5 x 106
Isomerase Activity 1 x 10-4 2 x 10-4 Km (MI 3 x 102
