An Unusual Cation-Binding Site and Distinct Domain–Domain

Article pubs.acs.org/biochemistry

An Unusual Cation-Binding Site and Distinct Domain−Domain Interactions Distinguish Class II Enolpyruvylshikimate-3-phosphate Synthases Samuel H. Light,† Sankar N. Krishna,† George Minasov,† and Wayne F. Anderson*,† †

Center for Structural Genomics of Infectious Diseases and Department of Biochemistry and Molecular Genetics, Feinberg School of Medicine, Northwestern University, 303 East Chicago Avenue, Chicago, Illinois 60611, United States S Supporting Information *

ABSTRACT: Enolpyruvylshikimate-3-phosphate synthase (EPSPS) catalyzes a critical step in the biosynthesis of a number of aromatic metabolites. An essential prokaryotic enzyme and the molecular target of the herbicide glyphosate, EPSPSs are the subject of both pharmaceutical and commercial interest. Two EPSPS classes that exhibit low sequence homology, differing substrate/glyphosate affinities, and distinct cation activation properties have previously been described. Here, we report structural studies of the monovalent cation-binding class II Coxiella burnetii EPSPS (cbEPSPS). Three cbEPSPS crystal structures reveal that the enzyme undergoes substantial conformational changes that alter the electrostatic potential of the active site. A complex with shikimate-3-phosphate, inorganic phosphate (Pi), and K+ reveals that ligand induced domain closure produces an unusual cation-binding site bordered on three sides by the N-terminal domain, C-terminal domain, and the product Pi. A crystal structure of the class I Vibrio cholerae EPSPS (vcEPSPS) clarifies the basis of differential class I and class II cation responsiveness, showing that in class I EPSPSs a lysine side chain occupies the would-be cation-binding site. Finally, we identify distinct patterns of sequence conservation at the domain−domain interface and propose that the two EPSPS classes have evolved to differently optimize domain opening−closing dynamics.

T

EPSPSs, Agrobacterium sp. CP4, Bacillus subtilis, Pseudomonas aeruginosa, Staphylococcus aureus, and S. pneumoniae class II EPSPSs are activated by monovalent cations (most significantly by NH4+, K+, and Rb+).3,8−11 In addition to enhancing catalytic rates, cations also increase the affinity for both substrates and glyphosate, leading to the proposal that cations act as allosteric activators.12 We used phylogenetic, structural, and biophysical approaches to address functional differences in the EPSPS classes. Here, we show that the class II cbEPSPS undergoes extensive, functionally important, conformational changes and possesses an unusual cation-binding site formed by domain closure and substrate binding. We propose a novel phylogenetic classification scheme that recognizes multiple EPSPS subclasses, identify multiple subclass specific features, and show that the identity of a single active site residue systematically differs between class I and class II EPSPSs and accounts for differential cation-binding properties.

he shikimate pathway comprises the enzymes that catalyze the first 7 committed steps in the biosynthesis of a variety of aromatic metabolites, including the aromatic amino acids.1 Enolpyruvylshikimate-3-phosphate synthase (EPSPS) catalyzes the conversion of phosphoenolpyruvate (PEP) and shikimate3-phosphate (S3P) to 5-enolpyruvylshikimate-3-phosphate and inorganic phosphate (Pi), the sixth step in the shikimate pathway (Scheme 1). EPSPS is the molecular target of the herbicide glyphosate, which binds competitively with PEP and uncompetitively with S3P, forming a ternary complex that resembles the EPSPS transition state.2 Crystal structures of EPSPSs from Agrobacterium sp. CP4, Escherichia coli, Mycobacterium tuberculosis, and Streptococcus pneumoniae reveal that the enzyme comprises two similar domains that are characterized by a 3xβαβαββ-folding motif.3−6 These structures demonstrate that the two domains are splayed outward in the unliganded state but come together and form the active site at their interface in the presence of ligands.3−6 Probably in part due to the extensive conformational change that ligand binding induces, PEP and S3P bind synergistically to EPSPS. Two EPSPS classes (class I and II) have been identified on the basis of biochemical (class II enzymes are glyphosate resistant and have higher PEP affinity) and phylogenetic (class I and II EPSPSs share 30,000× reduction from wild-type activity in Asp315 and Glu343 mutants.22,23 These striking conformational changes at the electrostatic sandwich appear to mask Asp315 and Glu343, effectively preventing them from repulsively interacting with the substrates, which, at neutral pH, are anticipated to assume a net −6 charge (Scheme 1). In the liganded state, the C-terminal domain presents a uniformly negative face, but in the two unliganded structures a positive strip runs across the portion of the C-terminal domain to which the substrates interact (Figure S2). The positively charged surface in the unliganded structures results partly from the unmasking of Arg346 and Arg387 that results from the departure of Asp315 and Glu343 from the active site. The addition of Arg340 to the active site further contributes to the positively charged unliganded state. Arg340 is part of the β18-α13 connecting loop in the unliganded structures but twists out of the active site, moving 10−15 Å, into the α12.5-helix unique to the liganded structures (Figure S2). These observations suggest that EPSPS initially presents a positively charged surface to the substrates and only at a later step in catalysis undergoes substantial conformational changes that replace the positively charged Arg340 with the negatively charged members of the catalytic sandwich. cbEPSPS S3P-Pi-K+ Complex Clarifies the Mechanism of Class II EPSPS Cation Activation. To identify the cationbinding site of the class II cbEPSPS, we soaked a crystal grown in the presence of S3P and glyphosate in 0.2 M KPO4. The resulting crystal structure reveals the domains adopt the liganded-closed conformational state and electron density consistent with S3P, K+, and Pi is present at the active site (Table S1 and Figure 3a and b). A superposition with the

state adopted among unliganded EPSPS structures (Figure S1, Table S2). A third cbEPSPS crystal structure, in complex with glyphosate and S3P, was determined in the P21 space group at 1.70 Å and contains a single molecule in the asymmetric unit (Table S1). Like previously characterized EPSPSs, the domains adopt a fully closed conformational state when ligand is bound (Figure 1c). A comparison of the three EPSPS structures reveals that the hinge region comprises the two domain connecting stretches (Pro18-Ile23 and Pro231-Ile234). A comparison of cbEPSPS with unliganded Agrobactrerium CP4, M. tuberculosis, and S. pneumoniae EPSPS crystal structures reveal the relative domain position differs within each of the five structures (Figure S1 and Table S2). This observation suggests that either a stable open conformational state is not conserved across EPSPSs or that the domains assume an ensemble of low occupancy positions in the unliganded state, in which case the state observed in a particular crystal structure must result from packing forces unique to the crystal form. In addition to the discrete domain conformations, a comparison of the three cbEPSPS crystal structures revealed substantial differences in secondary structure. Most strikingly, two helices (α11 and α13) in the unliganded-open structure exhibit distinct conformational states in the other two structures (Figure 2a, b, and c). In the unliganded-closed structure, the C-terminal portion of α11 has unwound and the remaining N-terminal portion has shifted ∼5 Å from its unliganded-open conformation (Figure 2d). In the liganded state, the C-terminal portion of α11 has unwound, and the Nterminal portion of α13 has unwound and rewound to form a distinct α-helix (α12.5) that runs perpendicular to α13 (Figure 2c and d). Notably, in the unliganded-open structure Asp315 on α11 hydrogen bonds with Glu343 on α13, but with these C

DOI: 10.1021/acs.biochem.5b00553 Biochemistry XXXX, XXX, XXX−XXX

Article

Biochemistry

Lysine Side Chain in Class I EPSPSs Recapitulates the Class II Cation-Binding Site. Having identified the K+ binding site, we sought to address the basis of differential EPSPS cation responsiveness. A 1.02 Å crystal structure of the class I Vibrio cholerae EPSPS (vcEPSPS) was determined in complex with glyphosate and S3P (Table S1 and Figure 4a). Superimposing the vcEPSPS (or, alternatively, the class I E. coli EPSPS structure, PDB 1G6S) to the K+ bound cbEPSPS reveals that the Nε atom from the vcEPSPS Lys412 side chain assumes the precise position adopted by K+ in the cbEPSPS complex (Figure 4b). Comparing the structures reveals that Lys412 in vcEPSPS corresponds with Thr413 in cbEPSPS and that other residues surrounding the cation-binding site are conserved across the two enzymes. This suggests that substituting the smaller threonine side chain for Lys412 eliminates the active site lysine Nε atom but opens a cavity to which cations can bind with similar effect.

Figure 3. CbEPSPS K+ + Pi + S3P complex. (a) The cbEPSPS K+ + Pi + S3P complex is depicted with ligands and residues at the K+ binding site shown in stick representation. (b) The Fo − Fc difference map, calculated with K+ + Pi + S3P omitted from the model, contoured at the 4.0σ level. (c) The K+ binding site is shown, with dashed lines indicating coordinating groups.

previously determined class I M. tuberculosis EPSPS enolpyruvylshikimate-3-phosphate and Pi complex reveals that the bound Pi in cbEPSPS superimposes with the Pi reaction product, suggesting that the observed K+ coordination sphere is representative of the product bound state (Figure S3a). The K+ ion is positioned directly adjacent to the Pi product where it interacts with two Pi oxygen atoms, Asp48 side chain carboxylate, Asn93 side chain carboxamide, Asn93 main chain carbonyl, and Glu343 side chain carboxylate (Figure 3c). Thus, the K+ ion interacts with the Pi product, the N-terminal domain, and the C-terminal domain. The identification of the cbEPSPS K+ binding site clarifies the mechanism of class II EPSPS cation activation. Situated at the intersection of the product, N-terminal domain, and C-terminal domain, it is obvious why cation binding is not observed when these three components of the binding site are separated in the open/unliganded conformational states. The required domain closure for the establishment of the cation-binding site sets up a complex binding event. In addition to forming the cation site, domain closure is required for establishing PEP (or glyphosate) and S3P binding sites. All three molecules bind synergistically to the enzyme and PEP (or glyphosate) contacts both S3P and cation.11,12 To embed at the enzyme core, all three molecules need to be properly oriented for the extensive conformational changes that establish the closed conformational state. The previous observation that cations enhance the affinity for PEP, glyphosate, and S3P (lower KM/KD) and increase the catalytic rate (kcat) implies that cations effect class II EPSPS catalysis in multiple ways.11,12 The location of the binding site poises the cation to impact catalysis by three distinct mechanisms: (1) increasing PEP and glyphosate affinity through favorable interactions with the Pi oxygen; (2) increasing PEP, glyphosate, and S3P affinity by favoring the closed conformational state to which the compounds bind; and (3) enhancing that catalytic rate by stabilizing the transition state. This final role of cation binding is supported by a consideration of the EPSPS reaction mechanism. The EPSPS reaction mechanism is characterized by a tetrahedral reaction intermediate that the enzyme destabilizes to promote bond breakage (Figure S3b).24,25 This bond breakage generates Pi and a cationic intermediate, which ultimately gives way to EPSP. By providing a cationic counterbalance to Pi, K+ is positioned to destabilize the tetrahedral reaction intermediate and promote this key step in the reaction (Figure S3c).

Figure 4. Structure and comparative analysis of class I vcEPSPS in complex with S3P + glyphosate. (a) Structure of vcEPSPS in complex with S3P + glyphosate (yellow sticks). (b) Superposition of class I vcEPSPS S3P + glyphosate and class II cbEPSPS K+ + Pi + S3P complexes (RMSD = 1.2 Å over 297 Cα atoms). (c) Showing the superposition in b, the K+ binding site is emphasized. The vcEPSPS Lys412, the Nε atom of which superimposes with the K+, and corresponding cbEPSPS residue, Thr413, are shown as sticks.

As cation activation has previously been suggested to be a defining property of class II EPSPSs, we hypothesized Thr413/ Lys412 identity might systemically differ between the EPSPS classes.7 Maximum-likelihood methods were used to construct a phylogenetic tree from a diverse set of 194 EPSPS amino acid sequences. Consistent with previous analyses, the constructed phylogenetic tree has a dichotomous structure (Figure 5).18,26 Of the two primary branches, one contains all known class I EPSPSs, while the other contains known class II enzymes. A branch within the class I enzymes separates archeal, as well as several bacterial, enzymes from eukaryotic and the majority of bacterial enzymes. Enzyme sequences in the archeal branches differ significantly from well-studied variants in the bacterial/ eukaryotic branch and, to our knowledge, have yet to be functionally characterized. We reasoned that functional distinctions might distinguish the two groups of enzymes and therefore treated them as distinct subclasses (referring to the bacterial/eukaryotic branch as Iα and the archeal branch as Iβ) D

DOI: 10.1021/acs.biochem.5b00553 Biochemistry XXXX, XXX, XXX−XXX

Article

Biochemistry

pocket that functionally replaces the lysine Nε atom. Pyruvate kinases and membrane-integral pyrophosphatases are either responsive or unresponsive to K+.27,28 The Hsc70 ATPase domain is responsive to K+, but the structurally related actin domain is unresponsive to cations.29 Thrombins are responsive or unresponsive to sodium.30 In each of these cases, a lysine in the cation unresponsive variant is converted to a smaller amino acid in the cation responsive variant. Furthermore, structural studies indicate that, in each case, the lysine Nε atom in the cation unresponsive variant adopts the position assumed by the cation in the cation responsive variant.29,31−33 Like EPSPSs, analyses of both the pyruvate kinases and membrane-integral pyrophosphatases show the cation dependent and independent variants correspond to phylogenetically distinct subtypes.27,28 The clear-cut break down in residue identity between the class Iα and II EPSPS classes argues against Lys412 → Thr/ Val/Ile representing a functionally neutral mutation. Rather, the systematic difference suggests that other class specific features may optimize Lys412 for class Iα and Thr/Val/Ile for class II function. This is supported by the observation that swapping out Lys412 in the E.coli class I EPSPS reportedly leaves a completely inactive enzyme, rather than a partially active, cation-activated enzyme.34 As such, we reasoned that other features that differ between the EPSPS classes could clarify the distinct role played by lysine in class I enzymes and Thr/Val/Ile in class II enzymes. To identify such distinctions, we contrasted patterns of sequence conservation in the three phylogenetically defined EPSPS subclasses. With the benefit of structural insight, we were able to identify three additional branch-specific differences: (1) Class II EPSPSs have a short insertion and short deletion relative to class I enzymes. The cbEPSPS insertion (Tyr282-Thr290 on cbEPSPS) extends the β15-β16 β-hairpin, which allows it to contact the N-terminal domain in the closed conformational state (Figure 6a). The cbEPSPS deletion (Lys268-Asn269 on vcEPSPS) removes two loop residues that connect β16−β17, which are positioned next to the extended β15−β16 hairpin (Figure 6a). Because the insertion and deletion both neighbor the domain hinge region and would sterically clash if present in the same protein, the mutual exclusivity of the two is likely required to produce a functional enzyme. (2) Class Iα and II EPSPSs form distinct interactions with the S3P phosphate group. Class Iα EPSPSs have a Ser/ Thr(1)-Ser(2)-Gln motif (Ser170-Gln172 in vcEPSPS), whereas class II EPSPSs have a Ser(1)-Ala(2)-Gln motif (Ser167Gln169 in cbEPSPS) at S3P’s phosphate binding site (Table S3). In the vcEPSPS class I complex, Ser(1) and Ser(2) side chains hydrogen bond with adjacent S3P phosphate oxygens. In the class II cbEPSPS, a single hydrogen bond is formed by Ser(1). Compared to the class I complex, Ser(1) is shifted ∼2 Å and hydrogen bonds with what would be Ser(2)’s phosphate oxygen (Figure 6b). In the cbEPSPS structures, the absence of Ser(2) and related shift in Ser(1) opens the space neighboring the S3P phosphate group. Here, the N-terminal Arg127 and Cterminal Glu338 form a bidentate salt bridge. These residues are 24 Å apart in the unliganded-open cbEPSPS structure and only interact when the domains collapse into the closed conformational state (Figure 6c). The importance of this salt bridge is suggested by the strict conservation of these residues in class II enzymes (Table S3). By contrast, residue identity varies at these positions in class I enzymes, and a pair that could form a similar salt bridge are never paired. (3) Class Iα and II EPSPS hinges establish distinct hydrogen bonding networks. Ser21

Figure 5. Phylogenetic topology of EPSPSs. Branches are colored based upon designated Iα, Iβ, and II EPSPS subclasses. Uniprot accession numbers are shown for each sequence. Values for nodes supported in >60% of 1000 random bootstrap replicate trees are shown. Green diamonds highlight the position of cbEPSPS and vcEPSPS. Blue and red circles denote experimentally verified class I and class II EPSPSs, respectively.

in subsequent analyses. The class II branch is composed exclusively of bacterial EPSPSs. Having constructed a phylogenetic tree that clearly delineated EPSPS classes, we used it to contrast Lys412/ Thr413 identity. This analysis reveals Lys412 is one of 24 residues that are invariant across the 77 aligned class Iα enzyme sequences (Table S3). By contrast, Thr, Val, or Ile is found at this position in the 90 aligned class II enzyme sequences (not shown). In the 27 class Iβ enzyme sequences, Lys, Ile, and Val are represented at this position (not shown). This finding suggests a systematic difference in Lys412/Thr413 (or Ile/Val) identity between class Iα and class II but not class Iβ EPSPSs. The strict conservation of Lys412 across the class Iα EPSPSs suggests that these enzymes are characteristically cation independent. As Thr413 does not directly interact with K+ in the cbEPSPS structure, substituting sterically similar Val or Ile is unlikely to compromise cation binding and, therefore, cationbinding properties are likely retained across the class II enzymes. Given that the patterns of conservation observed for Iα and II EPSPSs do not hold up for the Iβ enzymes, the cation dependency and kinetic parameters in general of these enzymes remain unexplored. Phylogenetic and Structural Analyses Suggest Cation Binding and Other Class Specific Features Differently Tune the Dynamics of Class I and Class II Domain Closure. In addition to establishing the relationship between EPSPS class and cation activation, the phylogenetic analysis also provides insight into the evolution of EPSPS cation responsiveness. Notably, the significantly broader distribution of class I EPSPSs (Figure 5) suggests that the monovalentunresponsive class Iα EPSPSs evolutionarily predate the monovalent-responsive class II EPSPSs. Interestingly, this would not be the first example of the substitution of a smaller amino acid side chain for a lysine opening up a cation-binding E

DOI: 10.1021/acs.biochem.5b00553 Biochemistry XXXX, XXX, XXX−XXX

Article

Biochemistry

conclude that class I and II EPSPSs provide an example of two phylogenetically related enzyme subtypes acquiring a distinct combination of features that, nevertheless, yield comparable relative domain conformational stability to result in similar catalytic properties.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.biochem.5b00553. Crystallographic data collection and refinement statistics, rotations associated with EPSPS domain closure, and conserved residues in each EPSPS subclass; conformational heterogeneity in unliganded EPSPS structures, changes in the electrostatic potential of the cbEPSPS active site, and a comparison of cbEPSPS and mtEPSPS reaction complexes (PDF)

■

Figure 6. Distinct featuers in EPSPS class. (a) Superimposed class I vcEPSPS (gray) and class II cbEPSPS (pink) S3P + glyphosate (yellow sticks) complexes from two perspectives. The class II insertion is colored blue on the cbEPSPS structure and the deletion is colored green on the vcEPSPS structure. Residues that exhibit different patterns of conservation in class Iα and II EPSPSs are shown as sticks and dashed lines denote polar interactions. (b) Different perspective of (a), with cbEPSPS (smaller font) and vcEPSPS (asterisks/larger font) residues labeled. (c) Unliganded-open (left panel) and liganded-closed (right panel) cbEPSPS structures with the distance between conserved class II bidentate salt bridge-forming residues shown.

AUTHOR INFORMATION

Corresponding Author

*Phone: 312-503-1697. Fax: 312-503-5349. E-mail: [email protected]. Funding

The Center for Structural Genomics of Infectious Diseases has been funded in whole or in part with federal funds from the National Institute of Allergy and Infectious Diseases, National Institutes of Health, Department of Health and Human Services, under Contract Nos. HHSN272200700058C and HHSN272201200026C (to W.F.A). Use of the Advanced Photon Source was supported by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences, under Contract No. DE-AC02-06CH11357. Use of the LSCAT Sector 21 was supported by the Michigan Economic Development Corporation and the Michigan Technology TriCorridor for the support of this research program (Grant 085P1000817).

on the N-terminal domain, Glu241 on the domain linker, and Gln272 on the C-terminal domain are conserved in the class Iα branch. Corresponding residues Asp20, Pro231, and Arg263 are invariant in the class II enzymes (Table S3). A structural analysis reveals that these residues, which localize to the domain hinge region, are separated in the open state but form a distinct network of interactions in the closed conformational state (Figure 6b). In the class I closed conformational state, Glu241 hydrogen bonds with Gln272, whereas in the class II closed conformational state, Asp20 forms a bidentate salt bridge with Arg263 (Figure 6). While a connection between the unique class properties cannot be definitively established, it is interesting to note that each of the identified differences is poised to impact domain opening/closing dynamics. The class II EPSPS insertion, deletion, and coordinated amino acid changes all localize to the hinge (Figure 6a), a region that must be important for determining the relative stability of open versus closed conformational states. Furthermore, the coordinated change in hinge amino acid identity coupled with the unique conservation of Arg127 and Glu338 results in the formation of two bidentate salt bridges specific to the class II EPSPS closed conformational state (Figure 6b). Positioned at the domain interface, the cation-binding site is also well situated to influence the relative stability of open versus closed conformational states. As the EPSPS domains must open for substrate binding and product release but close for catalysis, clearly, domain conformational stability must be fine-tuned to optimize catalysis. Collectively, the identified distinctions in the hinge region and domain interface, including the cation-binding site, suggest that class I and II EPSPSs have differently optimized domain opening/closing dynamics. It thus seems reasonable to

Notes

The authors declare no competing financial interest.

■

ABBREVIATIONS EPSPS, enolpyruvylshikimate-3-phosphate synthase; Pi, inorganic phosphate; PEP, phosphoenolpyruvate; S3P, shikimate-3phosphate; cbEPSPS, Coxiella burnetii EPSPS; vcEPSPS, Vibrio cholerae EPSPS

■

REFERENCES

(1) Bentley, R., and Haslam, E. (1990) The shikimate pathway–a metabolic tree with many branches. Crit. Rev. Biochem. Mol. Biol. 25, 307−384. (2) Steinrucken, H. C., and Amrhein, N. (1984) 5-Enolpyruvylshikimate-3-phosphate synthase of Klebsiella pneumoniae 2. Inhibition by glyphosate [N-(phosphonomethyl)glycine]. Eur. J. Biochem. 143, 351−357. (3) Funke, T., Han, H., Healy-Fried, M. L., Fischer, M., and Schonbrunn, E. (2006) Molecular basis for the herbicide resistance of Roundup Ready crops. Proc. Natl. Acad. Sci. U. S. A. 103, 13010− 13015. (4) Stallings, W. C., Abdel-Meguid, S. S., Lim, L. W., Shieh, H. S., Dayringer, H. E., Leimgruber, N. K., Stegeman, R. A., Anderson, K. S., Sikorski, J. A., Padgette, S. R., and Kishore, G. M. (1991) Structure and topological symmetry of the glyphosate target 5-enolpyruvylshikimateF

DOI: 10.1021/acs.biochem.5b00553 Biochemistry XXXX, XXX, XXX−XXX

Article

Biochemistry 3-phosphate synthase: a distinctive protein fold. Proc. Natl. Acad. Sci. U. S. A. 88, 5046−5050. (5) Schonbrunn, E., Eschenburg, S., Shuttleworth, W. A., Schloss, J. V., Amrhein, N., Evans, J. N., and Kabsch, W. (2001) Interaction of the herbicide glyphosate with its target enzyme 5-enolpyruvylshikimate 3phosphate synthase in atomic detail. Proc. Natl. Acad. Sci. U. S. A. 98, 1376−1380. (6) Park, H., Hilsenbeck, J. L., Kim, H. J., Shuttleworth, W. A., Park, Y. H., Evans, J. N., and Kang, C. (2004) Structural studies of Streptococcus pneumoniae EPSP synthase in unliganded state, tetrahedral intermediate-bound state and S3P-GLP-bound state. Mol. Microbiol. 51, 963−971. (7) Barry, G. F., Kishore, G. M., and Padgette, S. R. (1992) Glyphosate Tolerant 5-Enolpyruvylshikimate-3-phosphate Synthases, World Intellectual Property Organization International Publication Number WO 92/04449. (8) Fischer, R. S., Rubin, J. L., Gaines, C. G., and Jensen, R. A. (1987) Glyphosate sensitivity of 5-enol-pyruvylshikimate-3-phosphate synthase from Bacillus subtilis depends upon state of activation induced by monovalent cations. Arch. Biochem. Biophys. 256, 325−334. (9) Fischer, R. S., Berry, A., Gaines, C. G., and Jensen, R. A. (1986) Comparative action of glyphosate as a trigger of energy drain in eubacteria. J. Bacteriol. 168, 1147−1154. (10) Priestman, M. A., Funke, T., Singh, I. M., Crupper, S. S., and Schonbrunn, E. (2005) 5-Enolpyruvylshikimate-3-phosphate synthase from Staphylococcus aureus is insensitive to glyphosate. FEBS Lett. 579, 728−732. (11) Du, W., Wallis, N. G., Mazzulla, M. J., Chalker, A. F., Zhang, L., Liu, W. S., Kallender, H., and Payne, D. J. (2000) Characterization of Streptococcus pneumoniae 5-enolpyruvylshikimate 3-phosphate synthase and its activation by univalent cations. Eur. J. Biochem. 267, 222− 227. (12) Du, W., Liu, W. S., Payne, D. J., and Doyle, M. L. (2000) Synergistic inhibitor binding to Streptococcus pneumoniae 5enolpyruvylshikimate-3-phosphate synthase with both monovalent cations and substrate. Biochemistry 39, 10140−10146. (13) Otwinowski, Z., and Minor, W. (1997) Processing of X-ray diffraction data collected in oscillation mode. Methods Enzymol. 276, 307−326. (14) McCoy, A. J., Grosse-Kunstleve, R. W., Storoni, L. C., and Read, R. J. (2005) Likelihood-enhanced fast translation functions. Acta Crystallogr., Sect. D: Biol. Crystallogr. 61, 458−464. (15) Winn, M. D., Ballard, C. C., Cowtan, K. D., Dodson, E. J., Emsley, P., Evans, P. R., Keegan, R. M., Krissinel, E. B., Leslie, A. G., McCoy, A., McNicholas, S. J., Murshudov, G. N., Pannu, N. S., Potterton, E. A., Powell, H. R., Read, R. J., Vagin, A., and Wilson, K. S. (2011) Overview of the CCP4 suite and current developments. Acta Crystallogr., Sect. D: Biol. Crystallogr. 67, 235−242. (16) Murshudov, G. N., Vagin, A. A., and Dodson, E. J. (1997) Refinement of macromolecular structures by the maximum-likelihood method. Acta Crystallogr., Sect. D: Biol. Crystallogr. 53, 240−255. (17) Emsley, P., and Cowtan, K. (2004) Coot: model-building tools for molecular graphics. Acta Crystallogr., Sect. D: Biol. Crystallogr. 60, 2126−2132. (18) Richards, T. A., Dacks, J. B., Campbell, S. A., Blanchard, J. L., Foster, P. G., McLeod, R., and Roberts, C. W. (2006) Evolutionary origins of the eukaryotic shikimate pathway: gene fusions, horizontal gene transfer, and endosymbiotic replacements. Eukaryotic Cell 5, 1517−1531. (19) Edgar, R. C. (2004) MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic acids research 32, 1792−1797. (20) Jones, D. T., Taylor, W. R., and Thornton, J. M. (1992) The rapid generation of mutation data matrices from protein sequences. Bioinformatics 8, 275−282. (21) Tamura, K., Peterson, D., Peterson, N., Stecher, G., Nei, M., and Kumar, S. (2011) MEGA5: molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Mol. Biol. Evol. 28, 2731−2739.

(22) Berti, P. J., and Chindemi, P. (2009) Catalytic residues and an electrostatic sandwich that promote enolpyruvyl shikimate 3phosphate synthase (AroA) catalysis. Biochemistry 48, 3699−3707. (23) Lou, M., Burger, S. K., Gilpin, M. E., Gawuga, V., Capretta, A., and Berti, P. J. (2012) Transition State Analysis of Enolpyruvylshikimate 3-Phosphate (EPSP) Synthase (AroA)-Catalyzed EPSP Hydrolysis. J. Am. Chem. Soc. 134, 12958. (24) Anderson, K. S., Sikorski, J. A., and Johnson, K. A. (1988) A tetrahedral intermediate in the EPSP synthase reaction observed by rapid quench kinetics. Biochemistry 27, 7395−7406. (25) Lou, M., Burger, S. K., Gilpin, M. E., Gawuga, V., Capretta, A., and Berti, P. J. (2012) Transition state analysis of enolpyruvylshikimate 3-phosphate (EPSP) synthase (AroA)-catalyzed EPSP hydrolysis. J. Am. Chem. Soc. 134, 12958−12969. (26) Zhi, X. Y., Yao, J. C., Li, H. W., Huang, Y., and Li, W. J. (2014) Genome-wide identification, domain architectures and phylogenetic analysis provide new insights into the early evolution of shikimate pathway in prokaryotes. Mol. Phylogenet. Evol. 75, 154−164. (27) Oria-Hernandez, J., Riveros-Rosas, H., and Ramirez-Silva, L. (2006) Dichotomic phylogenetic tree of the pyruvate kinase family: K + -dependent and -independent enzymes. J. Biol. Chem. 281, 30717− 30724. (28) Belogurov, G. A., and Lahti, R. (2002) A lysine substitute for K +. A460K mutation eliminates K+ dependence in H+-pyrophosphatase of Carboxydothermus hydrogenoformans. J. Biol. Chem. 277, 49651− 49654. (29) Wilbanks, S. M., and McKay, D. B. (1998) Structural replacement of active site monovalent cations by the epsilon-amino group of lysine in the ATPase fragment of bovine Hsc70. Biochemistry 37, 7456−7462. (30) Bush, L. A., Nelson, R. W., and Di Cera, E. (2006) Murine thrombin lacks Na+ activation but retains high catalytic activity. J. Biol. Chem. 281, 7183−7188. (31) Marino, F., Chen, Z. W., Ergenekan, C. E., Bush-Pelc, L. A., Mathews, F. S., and Di Cera, E. (2007) Structural basis of Na+ activation mimicry in murine thrombin. J. Biol. Chem. 282, 16355− 16361. (32) Laughlin, L. T., and Reed, G. H. (1997) The monovalent cation requirement of rabbit muscle pyruvate kinase is eliminated by substitution of lysine for glutamate 117. Arch. Biochem. Biophys. 348, 262−267. (33) Kellosalo, J., Kajander, T., Kogan, K., Pokharel, K., and Goldman, A. (2012) The structure and catalytic cycle of a sodiumpumping pyrophosphatase. Science 337, 473−476. (34) Shuttleworth, W. A., Pohl, M. E., Helms, G. L., Jakeman, G. L., and Evans, J. N. S. (1999) Site-directed mutagenesis of putative active site residues of 5-enolpyruvylshikimate-3-phosphate synthase. Biochemistry 38, 296−302.

G

DOI: 10.1021/acs.biochem.5b00553 Biochemistry XXXX, XXX, XXX−XXX