5-Enolpyruvyl shikimate 3-phosphate synthase from Escherichia coli. Identification of Lys-22 as a potential active site residue.

5-Enolpyruvyl shikimate 3-phosphate synthase catalyzes the reversible condensation of phosphoenolpyruvate and shikimate 3-phosphate to yield 5-enolpyruvyl shikimate 3-phosphate and inorganic phosphate. The enzyme is a target for the nonselective herbicide glyphosate (N-phosphonomethylglycine). In order to determine the role of lysine residues in the mechanism of action of this enzyme as well as in its inhibition by glyphosate, chemical modification studies with pyridoxal 5'-phosphate were undertaken. Incubation of the enzyme with the reagent in the absence of light resulted in a time-dependent loss of enzyme activity. The inactivation followed pseudo first-order and saturation kinetics with Kinact of 45 microM and a maximum rate constant of 1.1 min-1. The inactivation rate increased with increase in pH, with a titratable pK of 7.6. Activity of the inactive enzyme was restored by addition of amino thiol compounds. Reaction of enzyme with pyridoxal 5'-phosphate was prevented in the presence of substrates or substrate plus glyphosate, an inhibitor of the enzyme. Upon 90% inactivation, approximately 1 mol of pyridoxal 5'-phosphate was incorporated per mol of enzyme. The azomethine linkage between pyridoxal 5'-phosphate and the enzyme was reduced by NaB3H4. Tryptic digestion followed by reverse phase chromatographic separation resulted in the isolation of a peptide which contained the pyridoxal 5'-phosphate moiety as well as 3H label. By amino acid sequencing of this peptide, the modified residue was identified as Lys-22. The amino acid sequence around Lys-22 is conserved in bacterial, fungal, as well as plant enzymes suggesting that this region may constitute a part of the enzyme's active site.

5-Enolpyruvyl shikimate 3-phosphate synthase catalyzes the reversible condensation of phosphoenolpyruvate and shikimate 3-phosphate to yield 5-enolpyruvyl shikimate 3-phosphate and inorganic phosphate. The enzyme is a target for the nonselective herbicide glyphosate (N-phosphonomethylglycine). In order to determine the role of lysine residues in the mechanism of action of this enzyme as well as in its inhibition by glyphosate, chemical modification studies with pyridoxal 5'-phosphate were undertaken. Incubation of the enzyme with the reagent in the absence of light resulted in a time-dependent loss of enzyme activity. The inactivation followed pseudo first-order and saturation kinetics with Kinact of 45 p~ and a maximum rate constant of 1.1 min". The inactivation rate increased with increase in pH, with a titratable pK of 7.6. Activity of the inactive enzyme was restored by addition of amino thiol compounds. Reaction of enzyme with pyridoxal 5"phosphate was prevented in the presence of substrates or substrate plus glyphosate, an inhibitor of the enzyme. Upon 90% inactivation, approximately 1 mol of pyridoxal 5"phosphate was incorporated per mol of enzyme. The azomethine linkage between pyridoxal 5'-phosphate and the enzyme was reduced by NaB3H4. Tryptic digestion followed by reverse phase chromatographic separation resulted in the isolation of a peptide which contained the pyridoxal B'-phosphate moiety as well as 3H label. By amino acid sequencing of this peptide, the modified residue was identified as Lys-22. The amino acid sequence around Lys-22 is conserved in bacterial, fungal, as well as plant enzymes suggesting that this region may constitute a part of the enzyme's active site.
The enzyme 5-enolpyruvyl shikimate 3-phosphate synthase (EPSP synthase)' (EC 2.5.1.19) catalyzes the reversible addition of the carboxyvinyl moiety of phosphoenolpyruvate to shikimate 3-phosphate (S3P) as shown in Fig. 1. The product, 5-enolpyruvyl shikimate 3-phosphate (EPSP) is an intermediate of the biosynthetic pathway leading to chorismate and thus to aromatic compounds in both microorganisms and * 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.
$ To whom correspondence should be addressed.
plants (1)(2)(3). The enzyme is the primary target of glyphosate (N-phosphonomethylglycine) ( Fig. 1) which inhibits the enzyme by competing with phosphoenolpyruvate for binding at the active site (4-5). Glyphosate is the herbicidal component of Roundup@ which is widely used as a broad-spectrum, nonselective, post-emergence weed killer. In an effort to understand the mechanism of action of this enzyme and its inhibition by glyphosate, the subunit structure, mechanisms, and kinetic properties of the enzyme have been extensively investigated (6-11). The complete DNA sequences of the enzyme from Escherichia coli, Salmonella typhimurium, and Aspergillus nidulans are known (12-14). A mutant form of the enzyme with replacement of Pro-101 to Ser is reported to be glyphosate resistant (13) and determination of the enzyme's threedimensional structure is in progress (15). However, little is yet known about which residues in the active center participate in binding and in catalysis. In view of the anionic nature of the substrates and glyphosate which interact with EPSP synthase, one would expect the presence of arginine (16) and/ or lysine residues at the active site of the enzyme. In this paper, we describe our studies on the chemical modification of the lysine residues of EPSP synthase from E. coli using pyridoxal 5'-phosphate. The results show that pyridoxal 5'phosphate inhibits the enzyme by formation of a Schiff base with the t-amino group of Lys-22 at or near the active center. This lysine residue is conserved in bacterial, fungal, and plant EPSP synthases studied to date (12)(13)(14). A preliminary report of a part of this work has been published (17).

DISCUSSION
Both the substrates, S3P and phosphoenolpyruvate, contain a carboxyl and a phosphate group which probably bind to lysine and arginine residues (16) of EPSP synthase. Since glyphosate inhibits the enzyme competitively with phosphoenolpyruvate (4, 5), its carboxyl or phosphate binding sites may overlap with that of phosphoenolpyruvate. In the absence of x-ray crystal structure data, identification of the amino acid residues at the active site is important for an understanding of the mechanism of action of EPSP synthase and its inhibition by glyphosate. In this study, we provide the evidence for the presence of an essential lysine residue within or near the active site. Among many chemical modification reagents for lysine, PALP was used since: ( a ) it has an aldehydic group which can form with the t-amino group of lysine and a phosphate group which can interact with lysine * Portions of this paper (including "Experimental Procedures," "Results,"  are presented in miniprint at the end of this paper. Miniprint is easily read with the aid of a standard magnifying glass. Full size photocopies are included in the microfilm edition of the Journal that is available from Waverly Press.  The inactivation appears to result from modification of an essential lysine residue since: ( a ) the inactivation follows pseudo first-order and saturation kinetics (Fig. 2, Miniprint); ( b ) PALP stoichiometrically binds to the enzyme (Fig. 3, Miniprint); ( c ) S3P, EPSP, and S3P plus glyphosate protect the enzyme from inactivation (Fig. 4, Miniprint); and ( d ) the activity of the inactivated enzyme is restored by addition of amino thiol compounds (Fig. 5,Miniprint). In order to determine the position of the modified lysine residue, the inactivated enzyme was reduced with NaB"H4, carboxymethylated with iodoacetic acid, and digested with trypsin. Fractionation of the tryptic digest on high performance liquid chromatography yielded a radioactive peptide (Fig. 8, Miniprint) which after acid hydrolysis, showed the presence of N-c-pyridoxyllysine ( Table I, Miniprint). Sequence analysis of the labeled peptide (Table 11, Miniprint) showed that the modified residue is Lys-22. This residue is conserved in all four known amino acid sequences of EPSP synthase from E. coli (121, S. typhimurium (13), A. nidulans (14), and plant (Petunia hybrid^)^ (Fig. 9, Miniprint).
Although the above results indicate that Lys-22 is located in or close to the enzyme active site, its role in enzymatic activity is an open question. However, since the Is0 for glyphosate and the K,,, for the substrates of the modified enzyme are much higher than those of the unmodified enzyme (Table  111, Miniprint), the modified lysine might be located close to the binding sites of glyphosate and substrates. Recent developments in site-directed mutagenesis have permitted studies on the essentiality of residues identified by chemical modifi-

PEP EPSP
-oH2c I -0,P C", H cation (18) or by x-ray crystal structure data (19); we have therefore replaced Lys-22 with 3 other amino acids: alanine, arginine, and glutamic acid. Of these, only the Lys-22 to Arg mutant enzyme was active, the other two replacements (Ala and Glu) led to inactivation of the e n~y m e .~ This suggests that the positively charged t-amino group of Lys-22 may play a critical role in substrate binding. Additional studies are in progress with the mutant enzymes to verify this hypothesis.