|
|
Biochem J. 2004 August 15; 382(Pt 1): 279–291. Published online 2004 August 10. Prepublished online 2004 June 1. doi: 10.1042/BJ20031809. | PMCID: PMC1133941 |
A core catalytic domain of the TyrA protein family: arogenate dehydrogenase from Synechocystis Carol A. Bonner,* Roy A. Jensen,*†‡ John E. Gander,* and Nemat O. Keyhani*1 *Department of Microbiology and Cell Science, Bldg 981, PO Box 110700, University of Florida, Gainesville, FL 32611, U.S.A. †Biosciences Division, Los Alamos National Laboratory, Los Alamos, NM 87544, U.S.A. ‡Department of Chemistry, City College of New York, New York, NY 10031, U.S.A. Received November 25, 2003; Revised May 11, 2004; Accepted June 1, 2004. The TyrA protein family includes prephenate dehydrogenases, cyclohexadienyl dehydrogenases and TyrAas (arogenate dehydrogenases). tyrAa from Synechocystis sp. PCC 6803, encoding a 30 kDa TyrAa protein, was cloned into an overexpression vector in Escherichia coli. TyrAa was then purified to apparent homogeneity and characterized. This protein is a model structure for a catalytic core domain in the TyrA superfamily, uncomplicated by allosteric or fused domains. Competitive inhibitors acting at the catalytic core of TyrA proteins are analogues of any accepted cyclohexadienyl substrate. The homodimeric enzyme was specific for L-arogenate (Km=331 μM) and NADP+ (Km=38 μM), being unable to substitute prephenate or NAD+ respectively. L-Tyrosine was a potent inhibitor of the enzyme (Ki=70 μM). NADPH had no detectable ability to inhibit the reaction. Although the mechanism is probably steady-state random order, properties of 2′,5′-ADP as an inhibitor suggest a high preference for L-arogenate binding first. Comparative enzymology established that both of the arogenate-pathway enzymes, prephenate aminotransferase and TyrAa, were present in many diverse cyanobacteria and in a variety of eukaryotic red and green algae. Keywords: arogenate dehydrogenase, enzyme specificity, prephenate, Synechocystis, TyrA, tyrosine Abbreviations: AGN, L-arogenate; DTT, dithiothreitol; LB, Luria–Bertani; PHE, L-phenylalanine; TYR, L-tyrosine; TyrAa, arogenate dehydrogenase; for brevity, single-letter code has been used for amino acids; for example, H197 stands for His-197 Dehydrogenases dedicated to TYR ( L-tyrosine) biosynthesis comprise the TyrA protein family, an assemblage of homologues having different substrate specificities. A given TyrA protein may be specific for AGN ( L-arogenate) for prephenate or may be able to accept either of these cyclohexadienyl substrates. This dehydrogenase protein family also exhibits comparable diversity with respect to its nicotinamide nucleotide co-substrate. Thus a given TyrA enzyme having any of the aforementioned specificities for cyclohexadienyl substrate may be specific for NAD +, for NADP + or may utilize either. This harmonizes with recent reports [ 1, 2] that substrate specificity often varies across a given protein family, even though the basic reaction chemistry deployed is usually maintained throughout the family. In instances of broad substrate specificity, there is variation in that alternative substrates may in some cases be accepted equally well, but in other cases one substrate may be preferred by an order of magnitude or more. Table 1 provides a key to the nomenclature used to describe the various possible cyclohexadienyl substrate combinations exhibited by TyrA proteins. | Table 1Nomenclature used to distinguish substrate specificities |
TyrA a (arogenate dehydrogenase) was first discovered in several species of cyanobacteria [ 3]. Until that time the only known route to TYR biosynthesis was via the coupled activities of prephenate dehydrogenase and aromatic aminotransferase. Although prephenate dehydrogenase activity was not detected in the cyanobacterium studied by Stenmark et al. [ 3], prephenate was consumed in the presence of an amino acid. This suggested that transamination at the side-chain position might precede the oxidative decarboxylation event that yields the aromatic ring of TYR. Both the transaminase (prephenate aminotransferase) and the dehydrogenase (TyrA a) were partially purified, and the new intermediate compound was named pretyrosine. Since pretyrosine was later shown in some organisms to also act as a precursor of PHE ( L-phenylalanine) [ 4], it was renamed L-arogenate (meaning ‘giving rise to aromatics’) [ 5]. A composite diagram of the dual biochemical flow routes to TYR is shown in Figure . This series of reactions is preceded upstream by the catalytic condensation of erythrose 4-phosphate and phosphoenolpyruvate by 3-deoxy- D- arabino-heptulosonate 7-phosphate synthase, followed by six enzymic steps to form chorismate, the central branch-point intermediate of the pathway. L-Tryptophan is formed from chorismate in five overall reaction steps (results not shown), involving seven catalytic domains (see [ 6] and references therein). Prephenate, formed from chorismate by the action of chorismate mutase, is the common precursor that feeds into the two divergent pathways that ultimately yield PHE (results not shown) and TYR. Since the discovery of AGN, combinatorial variations of the dual pathways leading to PHE and TYR synthesis have been documented. For instance, Brevibacterium flavum utilizes AGN as the sole precursor of TYR [ 7, 8], but uses phenylpyruvate as the sole precursor of PHE [ 9]. In opposite symmetry, Pseudomonas diminuta uses AGN to form PHE, but uses 4-hydroxyphenylpyruvate as the precursor of TYR [ 10]. Ps. aeruginosa deploys both of the foregoing routes to PHE and to TYR [ 4]. A single broad-specificity cyclohexadienyl dehydrogenase (TyrA c) accounts for the ability of Ps. aeruginosa to use prephenate or AGN as alternative substrates for TYR biosynthesis [ 11]. On the other hand, the presence of two distinct and spatially separated enzyme systems accounts for the Ps. aeruginosa pathway duality to PHE, i.e. (i) a bifunctional cytoplasmic protein AroQ•PheA that contains fused catalytic domains for chorismate mutase (AroQ) and prephenate dehydratase (PheA) [ 12] and (ii) a periplasmic cyclohexadienyl dehydratase (PheC) [ 13]. PheC has dual-substrate specificity and can either convert prephenate into phenylpyruvate or AGN into PHE. The ‘prephenate dehydrogenase’ of enteric bacteria is technically a cyclohexadienyl dehydrogenase, since it has a documented capability to utilize AGN, although activity with prephenate is superior by an order of magnitude [ 14]. We denote such cyclohexadienyl dehydrogenases as TyrA (p) (Table 1). In plants, the AGN route seems to be the major, if not the only, pathway leading to both PHE and TYR syntheses [ 15– 17]. Cyanobacteria comprise a cohesive but internally diverse lineage and, since the earliest studies [ 3], it has eventually become apparent that various species exhibit individuality in the profile of substrates utilized by their TyrA enzymes [ 18]. Genomic results are now revealing the extent to which this can be attributed to the presence of a single broad-specificity TyrA c on the one hand, or to multiple tyrA genes encoding enzymes of different substrate specificity on the other. TyrA a from the original cyanobacterial strain that was studied most, Agmenellum quadruplicatum BG1 (renamed Synechococcus sp. ATCC 29404), differs from the TyrA a of Synechocystis sp. PCC 6803 studied here, in being insensitive to TYR inhibition and in being able to use NAD + almost as well as NADP + [ 3]. Previously, cloned microbial TYR-pathway dehydrogenases have been limited to those specific for prephenate, or those with broad-substrate specificity, all examples so far being about those that utilize NAD + as the cofactor (see [ 19] and references therein). In the present study, we demonstrate that the single gene present in Synechocystis PCC 6803 encodes a TyrA protein that is absolutely specific for both AGN and NADP +. TYR, the sole inhibitor molecule recognized, is quite effective. TyrA proteins vary from those that possess only a basic catalytic core compared with others that possess allosteric domains and/or fusion domains [ 19]. The fusion of the enteric tyrA (p) gene with aroQ (encoding chorismate mutase) promotes an interesting additional specificity determinant in that the dehydrogenase domain will be preferentially exposed to prephenate in its catalytic microenvironment due to the proximity of the fused AroQ domain (which generates prephenate). Analysis of TyrA a in Synechocystis, which possesses only the basic catalytic module, allows the study of properties of this core domain, uncomplicated by any effect of allosteric or other domains. Materials Buffers, reagents and cell culture media were purchased from commercial sources. AGN and barium prephenate (converted into the potassium salt before the enzyme assay) were >90% pure and prepared as described in [ 20]. Reagents for molecular biology were obtained from New England Biolabs (Beverly, MA, U.S.A.), Stratagene (La Jolla, CA, U.S.A.) and Promega (Madison, WI, U.S.A.). Escherichia coli strain BL21(DE3) (Novagen; Madison, WI, U.S.A.), harbouring designated plasmid constructs where indicated, were stored as frozen cultures in LB (Luria–Bertani) broth supplemented with 10% (v/v) glycerol. Molecular-mass standards for gel filtration were purchased from Sigma–Aldrich (St. Louis, MO, U.S.A.). Frozen cell pellets of Synechocystis sp. (PCC 6902), Fisherella sp. (ATCC 29539), Anabaena sp. (PCC 7119), Synechococcus sp. (PCC 6301), Porphyridium cruentum and Prochlorothrix hollandica were generously provided by Geraldine Hall (Elmira College, Elmira, NY, U.S.A.). The cosmid containing the genomic clone CS01241 was kindly provided by CyanoBase (Genome Database for Synechocystis sp. PCC 6803, Kazusa DNA Research Institute, Chiba, Japan). Construction of pET:tyrAa, a TyrAa overexpression vector DNA preparations, restriction enzyme digests, ligations and transformations were performed using standard techniques. The cosmid clone CS01241, which spans the Synechocystis genome between 1532800 and 1570740 bp, was transformed and maintained in Epicurian Coli™ strain XL1-Blue MR (CyanoBase) [ 21]. The tyrAa gene was amplified by PCR using appropriate primers. The 5′-PCR primer was designed to include an NdeI restriction site (in boldface type below) to facilitate cloning into the ATG start site of pET24b(+) (Novagen), just downstream of the T7 promoter in the overexpression vector. To facilitate directional cloning of the tyrAa gene, an XhoI restriction site (underlined) was designed into the 3′-PCR primer. The primers used to construct the overexpression vector were 5′-GATAAA CATATGAAAATTGGTGTTGTTGGT-3′ and 5′-GATAAA CTCGAGTTATTCAACATACTTGTCCCGATC-3′. The amplified PCR product (861 kb) was doubly digested with NdeI and XhoI, ligated directly into an equivalently digested sample of the pET24b(+) vector and transformed into the T7 polymerase-inducible host strain BL21(DE3). The isolated clone in pET24b(+) was confirmed by sequencing the entire insert. Nucleotide and deduced amino acid sequence analyses were performed using web-based software tools. TyrAa assays Three assay methods were used. Two methods involved measurement of the rate of NADPH (or NADH) formation in continuous assays using either fluorimetric or spectroscopic detection systems. Continuous spectroscopic measurements were performed at 340 nm using a Beckman spectrophotometer. Continuous fluorimetric detection of NADPH at an excitation wavelength of 340 nm and an emission wavelength of 460 nm was performed with a Shimadzu spectrophotofluorimeter. Based on standard-curve values of authentic NADPH, a conversion factor of 23 FU (fluorescent units)=1 nmol/min NADPH was used in calculations of enzyme activity. The highly sensitive fluorimetric assay was used for kinetic studies of purified protein to ensure that initial rates as accurate as possible could be obtained at low substrate concentrations. For kinetic studies, initial rates were measured using the combinations of substrate specified in Figures and . The assays for each set of concentrations were performed in triplicate. Initial velocities were determined from the slopes of the progress curve within 90 s of elapsed reaction time, and linear portions of progress curves were used to determine reaction rates. Each data point is the average of three determinations; S.E.M. 2.6% (n=71). The third method utilized HPLC (Beckman, Schaumburg, IL, U.S.A.) for direct measurement of TYR formation. A standard reaction mix consisted of 50 mM Epps [4-(2-hydroxyethyl)piperazine-1-propanesulphonic acid] buffer (pH 8.6), 0.5 mM NADP +, 0.08 mM AGN and enzyme at room temperature (25 °C), unless otherwise stated in the text. TYR production was directly confirmed by HPLC assay after the reactions were stopped by the addition of NaOH; the samples were derivatized with o-phthalaldehyde for fluorimetric detection of peak area, and then injected into a C-18 reverse-phase column (Altech) as described in [ 22]. HPLC peak areas of authentic TYR were used to construct a standard curve. Aminotransferase assay Aminotransferases were assayed by o-phthalaldehyde derivatization of amino acids and fluorimetric detection by HPLC [ 23]. The substrates for prephenate aminotransferase, prephenate and L-glutamate were transaminated to AGN and α-oxoglutarate respectively. Since L-glutamate and AGN overlap on the HPLC elution profile, samples were acidified, resulting in the quantitative conversion of AGN to PHE. PHE can be readily quantified, since it is eluted well away from both L-glutamate and AGN. For the assay of aromatic aminotransferase, the phenylpyruvate/ L-glutamate co-substrates were transaminated to PHE and α-ketoglutarate. When crude extracts were assayed for prephenate aminotransferase, they were incubated for 15 min at 65 °C to inactivate prephenate dehydratase. Prephenate aminotransferase is stable to this heat treatment as commonly found [ 23], and this procedure is generally successful in avoiding interference of prephenate dehydratase with the assays. Aminotransferase assays consisted of 50 mM Epps buffer (pH 8.6), 5 mM amino acid donor, from 1 to 5 mM oxo-acid acceptor, 0.1 mM pyridoxal 5′-phosphate and enzyme. Determination of protein concentrations Protein concentrations were estimated by the Bradford Bio-Rad protein assay [ 24], using BSA as a standard. Overexpression and purification of TyrAa Step 1: crude extracts A single colony of E. coli BL21(DE3) harbouring pET:tyrAa was inoculated into 10 ml of LB medium, supplemented with 30 μg/ml kanamycin and grown overnight at 32 °C with aeration. A 200-ml volume of LB growth medium containing kanamycin was inoculated with 10 ml of the overnight culture. After growth at 32 °C to an absorbance A600 0.5, 1 mM isopropyl β-D-thiogalactoside was added, and the cells were incubated further for 2 h before harvest by centrifugation at 6000 g for 10 min at 4 °C. The resultant cell pellet (wet weight=3.44 g) was resuspended in 50 mM Epps buffer (pH 8.6), containing 20% glycerol and 1 mM DTT (dithiothreitol). The cells were disrupted by sonication (Ultratip Labsonic System), using 3×30-s pulses with 2 min of inter-pulse cooling, and cell debris was removed by ultracentrifugation (150000 g, 1 h at 4 °C). The high-speed supernatant (18 ml) was desalted by passing through a DG30 column (Bio-Rad) equilibrated with 50 mM Epps buffer (pH 8.6), containing 20% glycerol. The final total volume of the desalted crude extract of 24 ml was used for assay and for purification with an FPLC system (Amersham Biosciences). Step 2: preparative FPLC anion-exchange (Mono-Q) chromatography Crude extract from step 1 (2×10-ml aliquots) was injected into an ice-jacketed Mono-Q HR 10/10 column (bed volume, 8 ml; Amersham Biosciences), equilibrated in 50 mM potassium phosphate buffer (pH 7.5), containing 20% glycerol. The column was washed with 3 column volumes of the same buffer, and a gradient (220 ml) from 0 to 0.5 M KCl in the same buffer was applied to the column. Fractions containing activity eluted at approx. 0.2 M KCl, and these were pooled (44.5 ml) and dialysed against 50 mM Epps buffer (pH 8.6), containing 20% glycerol. A final volume of 45 ml was obtained. Step 3: 2′,5′-ADP–Sepharose 4B affinity chromatography Aliquots (1 ml) of the pooled dialysed sample from step 2 were injected on to an ice-jacketed 2′,5′-ADP–Sepharose 4B HR10/10 FPLC column (bed volume, 8.25 ml; Amersham Biosciences), equilibrated in 50 mM Epps buffer (pH 8.6), containing 20% glycerol. The column was washed with 2 column volumes of the same buffer, and a simultaneous gradient from 0 to 0.4 M KCl and 0 to 0.30 mM NADP+ in the same buffer was applied to the column. Appropriate fractions were pooled, and purity was estimated by SDS/PAGE. Molecular-mass determination by MS A sample of pure enzyme was subjected to MALDI–TOF (matrix-assisted laser-desorption ionization–time-of-flight) analysis (Voyager-DE PRO, Applied Biosystems) at the Protein Chemistry Core Laboratory, University of Florida. Native molecular-mass estimation by size-exclusion chromatography An FPLC Superdex 75 HR10/30 column (Amersham Biosciences) was used to estimate the native molecular mass of TyrAa. The column was equilibrated in 50 mM Epps buffer (pH 8.6), containing 20% glycerol and 0.15 M KCl. Purified protein was applied on to the column (200 μl aliquots) and eluted with the same buffer using a flow rate of 0.5 ml/min. Protein standards, including Blue Dextran (2000 kDa), albumin (66 kDa), carbonic anhydrase (29 kDa) and cytochrome c (12.4 kDa) were prepared in the same buffer and were applied individually to the column. SDS/PAGE and N-terminal amino acid determination Protein samples were denatured by SDS and subjected to SDS/PAGE to estimate monomeric mass following the method of Laemmli [ 25]. Purified TyrA a protein was electroblotted from SDS/PAGE to a PVDF membrane (Fisher Scientific, Fair Lawn, NJ, U.S.A.). After transfer, the blot was lightly stained with Coomassie Blue. The N-terminal amino acid sequence of the protein was determined at the Protein Chemistry Core Facilities (University of Florida Biotechnology CORE Center). Preparation of crude extracts from photosynthetic bacteria and algae Frozen cell pellets from species of Synechocystis, Synechococcus, Fisherella, Anabaena, P. cruentum, P. hollandica and Chlorella sorokiniana were dissolved in 50 mM potassium phosphate buffer (pH 7.5), containing 20% glycerol, 1 mM pyridoxal 5′-phosphate, 1 mM DTT and 1 mM PMSF. Samples were sonicated, and cell debris was removed by ultracentrifugation at 150000 g for 1 h at 4 °C. The resulting supernatant was dialysed against 50 mM potassium phosphate buffer (pH 7.5), containing 20% glycerol and 1 mM pyridoxal 5′-phosphate, and used as crude extract for enzyme activity assays. Molecular cloning of tyrAa Homology searching led to the identification of a single tyrA gene from Synechocystis sp. PCC 6803 (incorrectly annotated as a prephenate dehydrogenase). The nucleotide sequence of tyrA displayed a 48.6% GC content, consistent with the 48.7% GC genomic average for Synechocystis sp. ( www.kazusa.or.jp/codon/). The open reading frame corresponding to tyrAa encodes a 279-residue protein with the following predicted parameters: molecular mass of 30.21595 kDa, pI of 5.51 and a molar absorption coefficient ε 280 of 26330 M −1·cm −1. No signal peptide was found. Primers were designed for cloning tyrAa into the pET24b(+) overexpression vector. Cosmid clone CS01241, containing a 37.9 kb insert of the Synechocystis genome, was used as template for amplification of tyrAa in PCR mixtures as described in the Materials and methods section. The resulting construct, designated as pET: tyrAa, was used for further characterization. The presence of the pET: tyrAa insert was confirmed by nucleotide sequencing. The nucleotide sequence upstream of the start site contains two closely spaced stop codons and is generally A/T-rich; however, an obvious ribosome-binding site with the conventional spacing was not apparent. The nucleotide and deduced amino acid sequence corresponding to the Synechocystis sp. PCC 6803 cloned tyrAa gene has been deposited in GenBank® database and given the accession number AF482689. Purification and properties of the recombinant TyrAa The enzyme was purified from recombinant E. coli BL21(DE3) cells harbouring pET: tyrAa as described in the Materials and methods section, yielding an apparently homogeneous protein (Figure ). Little or no protein was obtained in the pellet fraction as inclusion bodies. The enzyme was purified approx. 10-fold from the crude extracts with a 16% yield (Table 2). N-terminal amino acid sequencing of the purified recombinant protein resulted in the sequence MKIGVVGLGLIGASL, in complete agreement with the N-terminal sequence deduced from the nucleotide sequence. | Table 2Purification of recombinant Synechocystis TyrAa |
MS (matrix-assisted laser-desorption ionization–time-of-flight analysis) resulted in a monomer peak at a molecular mass of 30210.56, a value almost identical with that predicted from the amino acid sequence data. The native molecular mass of TyrA a was estimated by gel-filtration chromatography as described in the Materials and methods section. A native molecular mass was calculated in the range 57–65 kDa, consistent with the dimeric structure of other TyrA proteins (monomer, 30.2 kDa). Reports in the literature of higher native molecular masses of TyrA proteins from Corynebacterium [ 8] and Brevibacterium [ 8] may indicate that the oligomer species is variable. The higher molecular masses previously reported for Acinetobacter [ 26] and Ps. aeruginosa [ 11] reflect the then unrecognized fusion of tyrAc with another gene of aromatic biosynthesis. Crude extracts of E. coli BL21 (harbouring a plasmid without insert) were used as controls for measuring the extent to which E. coli TyrA (p) might contribute to TyrA a activity. It has been previously reported that AGN is a poor substrate for the E. coli TyrA (p) enzyme, which exhibits an absolute requirement for NAD + as co-substrate [ 14]. As expected, none of the TyrA a activity measured in the presence of NADP + could be attributed to the TyrA (p) of E. coli BL21. The purified recombinant Synechocystis TyrA a showed no activity with prephenate in combination with either NADP + or NAD + as cofactor. TyrA a activity (AGN/NADP +) was proportional to elapsed time and protein concentration at saturating concentrations of substrates. TyrA a activity was not detected with AGN in combination with NAD +. TyrAa activity was tested to determine the optimal pH by measuring initial rates of catalysis over a pH range 6.5–10.0 at 25 °C in a variety of buffers as described in the Materials and methods section. No measurements were feasible below a pH of 6.5 due to the acid lability of AGN. The optimum pH of the purified recombinant enzyme was between 8.25 and 8.75. Epps buffer at pH 8.5 was found to be optimal. Lower activities were observed in other buffers at this pH, including Tris/HCl, sodium hydroxide/borate and glycine. The purified TyrAa enzyme was less stable during storage at 4 °C than when maintained frozen at −20 °C or at −70 °C. Full activity was retained in repeated freeze−thaw cycles, provided that concentrations of 10 μg of protein/ml or more were maintained. The enzyme displayed an optimum temperature of 28−30 °C, with activity decreasing almost 50% above 42 °C (results not shown). DTT (up to 0.5 mM) and EDTA/EGTA (up to 5 mM) did not affect enzyme activity. Additionally, no effect on enzymic activity was observed after the addition to standard reaction mixtures of various bivalent metal ions (up to 0.5 mM) including Mg2+, Mn2+, Fe2+ or Ca2+. Confirmation of TYR as the product of TyrAa-mediated catalysis Figure demonstrates the NADP+-dependent conversion of AGN into TYR. C-18 reverse-phase HPLC was used to analyse reaction mixtures at zero elapsed time (left) and after 30 min of reaction (middle). After the reaction, a fraction of AGN disappeared compared with the zero-time control, and a corresponding amount of TYR product appeared. Unchanged AGN was non-enzymically converted into PHE in the presence of acid ([H+]), thus confirming the high purity of the AGN substrate preparation. In a comparable experiment using prephenate and NADP+, prephenate was not utilized. When such mixtures were acidified, prephenate was recovered quantitatively as phenylpyruvate. Thus we could conclude that the prephenate had not been scavenged for some unknown reaction and had therefore remained potentially available as a substrate for TyrAa. Kinetic analysis Kinetic parameters Initial-rate measurements taken when varying the concentration of either substrate in the presence of a fixed concentration of the other revealed saturable Michaelis–Menten kinetics. The results were transformed to Lineweaver–Burk double-reciprocal plots (Figures A and B), useful for displaying results in a form that facilitates derivation of most kinds of equations and that allows visualization of the influence of each substrate on the various slopes or intercepts. Nearly identical Vmax values were obtained in the series with AGN or NADP + as variable substrate, confirming the accuracy of the results. The straight lines obtained indicate that the two catalytic sites of the dimer are independent. The converging linear Lineweaver–Burk plots also indicate a ternarycomplex (sequential) mechanism. A degree of substrate inhibition was observed at very high concentrations of both AGN and NADP +. A commercial program, GraphPad Prism (version 4.0), from GraphPad Software (San Diego, CA, U.S.A.) was used to estimate kinetic constants directly from non-linear regression by global fitting to the initial velocity equation for a sequential mechanism, and these results are summarized in Table 3. The Km for NADP + (38 μM) was almost an order of magnitude smaller than that for AGN (331 μM). The kcat/ Km value, known as the catalytic capture value [ 27], calculated for AGN was 3.90×10 5 M −1·s −1 and the kcat/ Km value for NADP + was 3.39×10 6 M −1·s −1. kcat values were calculated by assuming that the enzyme is a 60.4 kDa homodimer with two active sites. | Table 3Kinetic constants of recombinant Synechocystis TyrAa |
Plots of Figure (A) resulted in lines intersecting at a common point on the abscissa (1/ v0=0), whereas the intersect in Figure (B) is in the second quadrant (1/ v0>0). The intersect positions indicate the relationship between Km and the dissociation constant Kia for the variable substrate [ 28]. Thus the intersect position in Figure (A) on the abscissa implies that Km= Kia for AGN. The intersect position in Figure (B), on the other hand, implies that Kia> Km for NADP +. Inhibitors The immediate product of the TyrA a reaction, TYR, was an effective inhibitor. Double-reciprocal plots (Figures C and D) showed TYR to be a competitive inhibitor with respect to AGN ( Ki=89 μM) and a non-competitive inhibitor ( Ki=78 μM) with respect to NADP +. Prephenate, NAD +, NADPH, PHE and 4-hydroxyphenylpyruvate were tested as inhibitors of the Synechocystis TyrA a activity, but none of these produced detectable inhibition. Total insensitivity to NADPH inhibition seems quite striking in view of NADPH inhibition documented for other TyrA homologues, e.g. E. coli TyrA (p) [ 29] and especially TyrA a from Arabidopsis [ 30]. Although 2′,5′-ADP, as expected, was a competitive inhibitor with respect to NADP+, an unusual relationship with AGN was observed. At high concentrations of AGN, even in combination with low NADP+ concentration, little or no inhibition occurred. When the AGN concentration was lowered to Km, inhibition up to approx. 40% could be obtained at the very lowest concentrations of NADP+ technically feasible. Progressively greater inhibition was observed as AGN concentrations were decreased below Km. Under conditions of sufficiently low AGN concentration, to allow sensitivity to inhibitor, NADP+ could abolish inhibition in a strictly competitive fashion over the tested NADP+ concentration range of 20–75 μM. A possible explanation for the unexpected effect of AGN on sensitivity to the inhibitor is that 2′,5′-ADP can bind to the E•Eagn species (hereafter considered equivalent to the agnE•E species), but not to the agnE•Eagn species. This scenario is depicted in Figure , and the scheme shown presumes a high preference for an order of substrate binding whereby AGN binds first. At 331 μM AGN (Km concentration), the enzyme will partition as follows: 25% AGN-free dimer (E•E), 25% agnE•Eagn dimer and 50% E•Eagn dimer. The proposed sensitive target of inhibitor action (i.e. E•Eagn) would represent 50% of the total AGN-bound active sites and therefore 50% of the total catalytic activity. As the AGN concentration is increased, the relative fraction (Fr) of the enzyme, which is in the agnE•Eagn form, increases to the Fr2 value and that of the E•Eagn form changes to {[AGN]/([AGN]+Km)−Fr2}. With 0.05 mM AGN the abundance of the E•EAGN species will be more than 14-fold greater than the agnE•Eagn species. If 2′,5′-ADP selectively targets the E•Eagn species, one would expect a correlation between abundance of the E•Eagn target and sensitivity of TyrAa to inhibition by 2′,5′-ADP. Figure shows the result of fitting the inhibition data to the E•Eagn-target mechanism proposed. Calculations were performed to relate the expected relative abundance of E•Eagn and agnE•Eagn dimer species that would be generated at a continuum of AGN concentrations (from 0.02 to 1.5 mM). The E•Eagn species has one AGN-bound active site, whereas the agnE•Eagn species has two AGN-bound active sites, and hence each agnE•Eagn species accounts for twice as much catalytic activity as does each E•Eagn species. The upper curve in Figure shows the relative activity that can be attributed to E•Eagn (calculated as the number of AGN-bound active sites on E•Eagn divided by the total number of AGN-bound active sites present on E•Eagn plus agnE•Eagn). When the overall inhibition obtained in the presence of 0.5 mM 2′,5′-ADP was subtracted from the activity attributed to E•Eagn (top line), the resulting data points shown in Figure represent the remaining uninhibited activity attributed to E•Eagn. This is very close to a calculated 50% inhibition of E•Eagn that holds throughout the range of AGN concentrations used in Figure , as shown by the dotted line. Thus, even though overall sensitivity to the inhibitor decreased with increasing AGN concentration, the sensitivity of the calculated portion of the total activity contributed by E•Eagn was constant (approx. 50%) under conditions where NADP+ (0.02 mM) and 2′,5′-ADP (0.5 mM) were fixed. Since 0.5 mM 2′,5′-ADP only causes 50% inhibition of E•Eagn at concentrations of NADP+ less than Km, it seems qualitatively apparent that the Ki value for 2′,5′-ADP must be well over an order of magnitude greater than the Km value for NADP+ (38 μM). Kinetics and mechanism of TyrAa Unlike many dehydrogenases, the reaction catalysed by TyrA a is irreversible. TYR (having a stable aromatic ring) cannot be converted into AGN (having an unstable cyclohexadienyl ring) in the reverse direction (Figure ). The reversible binding reactions of AGN and NADP + are followed by the hydride transfer and irreversible decarboxylation, which we assume to be concerted (rather than stepwise) as is the case for the TyrA (p) homologue of E. coli [ 31]. The irreversibility of the reaction dictates that the three rate constants associated with the oxidative decarboxylation, as well as the rate constants koff associated with release of NADPH and TYR, determine the value of kcat. Since the intercept positions of Figures (A) and (B) indicate that the Km and Kia values for AGN are equal but that the Km value for NADP+ is less than the corresponding dissociation constant, a rapid equilibrium random mechanism can be ruled out since, then, Ka/Kia is expected to equal Kb/Kib. Also, since (as exploited for purification) the enzyme can bind 2′,5′-ADP–Sepharose in the absence of AGN, either a steady-state random mechanism or an ordered mechanism with NADP+ as the first substrate to bind were a priori possibilities. However, the fact that TYR inhibits competitively with respect to AGN (Figure C) and non-competitively with respect to NADP+ (Figure D) eliminated the possibility of an ordered mechanism in which NADP+ is the leading substrate. Although a steady-state random-order mechanism is thus indicated, we favour the possibility that there exists a distinct preference for the binding of AGN first to explain the 2′,5′-ADP inhibition results. TyrAa and prephenate aminotransferase activities in photosynthetic bacteria and algae TyrA a specific activities were identified in crude extracts of P. hollandica, P. cruentum (a red alga), Chlorella sorokiniana (a green alga) and in four species of cyanobacteria (Table 4). In these organisms, TyrA a activity was strictly dependent on NADP + and no prephenate dehydrogenase activity (NAD +- or NADP +-dependent) was detected, exactly the features typical of the Synechocystis PCC 6803 system. Crude extracts were also assayed for prephenate aminotransferase activity, since the enzyme catalyses the penultimate step in TYR biosynthesis (Figure ) and yields the substrate for TyrA a. Prephenate aminotransferase activity was found in all the crude extracts tested (Table 4). | Table 4Specific activities of TyrAa and prephenate aminotransferase (PAT) in cyanobacteria, Chlorophyta, Prochlorophyta and Rhodophyta |
Two subtypes of TyrA proteins in heterocystous cyanobacteria A search of the SWISS-PROT and GenBank® databases identified nine cyanobacterial homologues of Synechocystis sp. PCC 6803 tyrAa. Two different tyrA genes were identified in the genomes of Anabaena and in Nostoc, and one each from Synechocystis (PCC 6803), Synechococcus W8102, Synechococcus 7002, Gloeobacter violaceus, Prochlorococcus marinus CCMP1378 and P. marinus MED4. When each sequence was used as a query against the BLAST database, it was qualitatively apparent that the ten sequences fell into two subgroups. An alignment of the amino acid sequences of these proteins is shown in Figure . The upper subgroup of eight protein sequences (one from each organism) is predicted to comprise NADP +-dependent TyrA a proteins, since our characterized Ssp TyrA a protein falls into this group. These eight proteins cluster tightly on a protein tree (results not shown) in which the position of this TyrA subgroup enjoys strong (100%) bootstrap support. The bottom two cyanobacterial sequences shown in Figure are quite distinct from the upper ones and belong to a large and cohesive protein subgroup (results not shown) that is dominated by TyrA (p) proteins from the enteric lineage (defined as all organisms that have diverged as far from E. coli as Shewanella putrefaciens on the 16 S rRNA tree). Residues that are invariant within this latter TyrA (p) subgroup are shown as grey shading in Figure . The middle group in Figure consists of experimentally established arogenate-specific dehydrogenases from Corynebacterium efficiens (C. A. Bonner and R. A. Jensen, unpublished work), Corynebacterium glutamicum [ 7, 8], Nitrosomonas europaea [ 32] and higher plants [ 15, 17, 33]. The TyrA a proteins from coryneform bacteria, N. europaea, and higher plants do not cluster closely with one another on the phylogenetic tree (not shown), but we included them in the multiple alignment of Figure in an attempt to visualize a signature motif that might correspond to AGN specificity. Figure indicates the position of the Wierenga ‘fingerprint’ [ 34] for the N-terminal ADP-binding βαβ fold, which is commonly recognized for nicotinamide nucleotide binding. Ssp TyrA a has the minimal variable loop of two amino acids compared with the maximal five amino acids for the Npu TyrA (p) and Asp TyrA (p) (and E. coli TyrA (p)). Ssp TyrA a matches the Wierenga ‘fingerprint’ residues at positions 1, 2, 4, 6, 8, 11, 15, 18 and 28, but not at positions 30 and 32. It would not be expected to obey the ‘fingerprint’ at position 32 (D or E), since this negatively charged residue is crucial for hydrogen-bonding to the diol group of the ribose near the adenine moiety in NAD +-specific enzymes. It is well known that NADP +-specific dehydrogenases cannot tolerate a negatively charged residue at position 32 [ 35, 36]. The general presence of positively charged residues in this region for the TyrA a proteins, but not for the TyrA (p) or TyrA c proteins might be significant. Figure illustrates the anchor residues that are invariant throughout the TyrA superfamily. These residues probably dictate the conserved scaffold responsible for the common chemistry of the separate reactions within the TyrA superfamily. Six additional residues (boxed lightly) are invariant for the currently established TyrA a proteins (which includes, in addition to cyanobacteria, those from C. efficiens, C. glutamicum, N. europaea and two selected higher-plant sequences). Some of these residues may eventually prove to be tied to narrow specificity for AGN. However, many of them exhibit relatively high conservation throughout the TyrA superfamily, and the apparent invariance quite probably reflects the current small size of the TyrA a sample. No convincing motif having predictive value for AGN specificity is apparent. One of the conserved arginine residues near the C-termini of the cyanobacterial proteins probably contributes the positively charged side chain that is known [ 37] to contribute the single most conserved structural feature associated with stabilization of the NADP +–protein complex. Distribution of TyrAas in nature Four classes of cyclohexadienyl substrate specificity are known within the TyrA superfamily of homologues. These include prephenate-specific (TyrA p), AGN-specific (TyrA a) and the broad-specificity cyclohexadienyl (TyrA c) dehydrogenases. A fourth class is represented by an enzyme of antibiotic biosynthesis (PapC) that converts 4-amino-4-deoxy-prephenate into 4-amino-phenylpyruvate [ 38]. Representatives of each specificity class have been studied at the molecular-genetic level. Recently, a plant tyrAa has been cloned and characterized from Arabidopsis thaliana [ 33]. Interestingly, the latter consists of two near-identical domains that are fused. The gene encoding this 68 kDa protein co-exists in the genome with a single-domain gene [ 30] that encodes a predicted 37 kDa protein, somewhat larger than the core catalytic domain of TyrA a from Synechocystis. The well-studied E. coli TyrA (p) differs from Synechocystis TyrA a not only in its substrate specificity, but also in possession of a fused AroQ domain and two allosteric sites, at one of which prephenate acts weakly [ 39]. The latter differences may not be so surprising, but it is striking how many differences distinguish Synechocystis TyrA a from higher-plant TyrA a in view of the prevailing hypothesis of endosymbiotic origin. Although both have high specificity for AGN and NADP +, the Arabidopsis TyrA a has a Km for AGN of 70 μM (compared with 331 μM). It is inhibited by NADPH ( Ki=54 μM) and an E–NADPH–AGN dead-end complex has been proposed. One of the two paralogues has a weak ability to utilize prephenate and positive co-operativity for AGN is observed [ 30]. In photosynthetic eukaryotes, TyrA a is ubiquitous. In prokaryotes, the TyrA a class is less widespread and is currently represented by three widely spaced lineages: cyanobacteria, coryneform bacteria and N. europaea. This observation is consistent with an evolutionary scenario whereby the ancestral dehydrogenase was a broad-specificity TyrA c and in which narrowing of substrate specificity (to yield TyrA p or TyrA a) has occurred independently on multiple occasions in modern lineages. TyrA a in higher-plant chloroplasts [ 17] may have originated from cyanobacteria via endosymbiosis. Assignment of the substrate specificity of experimentally uncharacterized TyrA homologues in silico is uncertain unless they exhibit very high amino acid identity with known TyrA a proteins. For example, the high identities of TyrA sequences from Mycobacterium tuberculosis, Bifidobacterium ( Thermomonospora) and Streptomyces species with that of C. glutamicum suggests a reasonable possibility that actinomycete bacteria as a group will prove to possess the TyrA a specificity. N. europaea currently has no close genome relatives that have been sequenced. The first BLAST hit returned from an N. europaea TyrA a query is the protein from Ralstonia solanacearum which is known to differ from that of N. europaea in specificity for both of its substrates [ 32]. The core catalytic domain of TyrA proteins A core catalytic domain can be identified that is common to all TyrA proteins [ 19]. Some TyrA proteins have a C-terminal extension that may be an allosteric domain. The simplest set of proteins belonging to the TyrA family exhibit only a core catalytic domain (approx. 180 amino acids). These include the well-characterized TyrA c enzymes from Neisseria gonorrhoeae [ 32] and Zymomonas mobilis [ 40], as well as TyrA a from the cyanobacteria (the present study). These proteins do not cluster together on the TyrA protein tree. In addition, the core catalytic domain from Ps. stutzeri (having a tyrAc• aroF fusion) has been engineered for study [ 19]. Xie et al. [ 19] suggested that the foregoing four TyrA groupings, although divergent from one another, define a common catalytic domain whereby inhibitors bind at the catalytic site and exhibit classical competitive inhibition with respect to the cyclohexadienyl substrates used [ 19]. In this model, one would expect that the specificity for the side chains of substrates utilized would parallel the specificity for side chains of any inhibitors. Synechocystis sp. and A. thaliana TyrAa proteins recognize an alanyl side chain in AGN, which in fact is the only cyclohexadienyl substrate that they accept. In line with this, the latter TyrAa proteins can recognize TYR (alanyl side chain) but not hydroxyphenylpyruvate (pyruvyl side chain) as an inhibitor. The ring-carboxylate moiety of AGN is not essential for binding at the catalytic site since TYR lacks this substituent. On the other hand, since N. europaea TyrAa and Z. mobilis TyrAc are not inhibited by TYR, a 1-carboxy substituent is, probably, necessary for successful binding at the catalytic site. Finally, the N. gonorrhoeae TyrA(p) exhibits an overwhelming preference for prephenate (pyruvyl side chain), and, consistent with the above discussion, is subject to inhibition by 4-hydroxyphenylpyruvate (pyruvyl side chain) but not by TYR (alanyl side chain). The tyrAc/trp supraoperon of Nostoc/Anabaena All of the cyanobacterial organisms evaluated in the present study possess a highly conserved tyrAa gene, as well as a complete suite of tryptophan-pathway genes that are dispersed (unlinked) in the genome. Curiously, one divergent cyanobacterial lineage of large-genome organisms ( Nostoc and Anabaena) also possesses a trp/ aro supraoperon consisting of a number of seemingly redundant genes [ 6, 41]. These include a second tyrA gene, additional trp-pathway genes (all except trpC) and genes encoding the first two general steps of aromatic amino acid biosynthesis. All of these linked genes are represented elsewhere in the genomes of Nostoc and Anabaena at scattered loci. The smallgenome cyanobacteria possess single copies of the above genes, all of them at dispersed genomic locations. The closest BLAST hits for the cyanobacterial TyrA (p) proteins are not the TyrA a homologues in these same organisms, but are the TyrA (p) domains of the AroQ•TyrA (p) fusions in the enteric lineage. Since the enteric proteins are NAD +-specific and strongly prefer prephenate, it is quite possible that the ‘extra’ cyanobacterial proteins possess a similar specificity pattern. Indeed, this would be consistent with biochemical evidence provided in the literature for both Nostoc and Anabaena [ 18]. Specificity for the cyclohexadienyl substrate within the TyrA superfamily Knowledge of the atomic details of interaction of TyrA proteins with their substrates is limited since X-ray crystallography results are not yet available. Other detailed information is limited to E. coli. In retrospect, one can see that E. coli TyrA (p) is not the simplest model for studying the basic properties of the catalytic core region because of the aroQ fusion. Site-directed mutagenesis has established that H197 (highly conserved amongst all TyrA proteins) of E. coli TyrA (p) is an essential catalytic residue [ 42], and that it specifically interacts with the 4-hydroxy moiety of prephenate. One exception to the invariance of H197 is PapC from Streptomyces pristinaespiralis, and this is fully expected since its substrate (4-amino-4-deoxy-prephenate) lacks the 4-hydroxy moiety. In this region, the E. coli sequence is 197HPMFG 201, compared with NPMFA in Streptomyces pristinaespiralis PapC. A second exception is the Sco_2 protein in Streptomyces coelicolor, a paralogue of TyrA that resides in the large calcium-dependent antibiotic gene cluster [ 43]. This predicts that the substrate for Sco_2 TyrA is neither prephenate nor AGN. The alignment match corresponding to the E. coli 197HPMFG 201 motif is APVVG in Sco_2 TyrA. H197 and G201 are otherwise invariant in the TyrA superfamily. Evidence has also been obtained that R294 interacts electrostatically with the ring carboxylate of prephenate in E. coli. Although we suggest (based on sensitivity to TYR inhibition) that TyrA a from Synechocystis sp. PCC 6803, TyrA c from Ps. stutzeri and TyrA (p) from N. gonorrhoeae exemplify instances where binding at the catalytic site does not absolutely require a carboxylate ring, R294 is conserved throughout the TyrA superfamily, with the exception of higher plants. Important residues that co-ordinate with the pyruvyl side chain of prephenate (for TyrA p), the alanyl side chain of AGN (for TyrA a) or both (for TyrA c) are unknown, although Christendat and Turnbull [ 44] have asserted that, in E. coli, residues K178, R286 and R294 can at least be eliminated as ones which interact with the pyruvyl side chain of prephenate. Residues that dictate AGN specificity may be differently influenced by a complex interacting relationship with residues that influence acceptance of NAD+ and/or NADP+. For example, C. glutamicum and N. europaea both recognize AGN in a very specific way, but they differ in that the Corynebacterium TyrAa is broadly specific for the pyridine nucleotide cofactor whereas the Nitrosomonas TyrAa is NADP+-specific. Another complexity is that AGN-specific TyrA proteins presumably differ from one another in that the 1-C group must be recognized for AGN-binding in some cases (as for the Nitrosomonas TyrAa, since TYR is not an inhibitor), whereas it must not be required for binding in other cases (as for Synechocystis TyrAa, since TYR is a potent inhibitor). Specificity for nicotinamide nucleotide cofactor within the TyrA superfamily A general axiom of biochemistry holds that dehydrogenases participating in reductive biosynthetic steps utilize NADPH, whereas oxidative catabolic steps utilize NAD +. Dehydrogenases of oxidative biosynthetic steps (such as the one catalysed by TyrA) belong to neither of the foregoing categories, and some use NAD + and others use NADP + as the redox cofactor. In the vast majority of cases, a strong preference exists within a given protein family for either NAD + or NADP +. One exception is the glutamate dehydrogenase family of enzymes, which subdivides into groups exhibiting strict specificity for NAD +, for NADP + or those that are broadly specific and can accommodate both [ 45, 46]. There are reports of contemporary TyrA proteins that can use either NAD + or NADP + [ 47]. Admittedly, results acquired from survey results using crude extracts should be viewed with caution. For example, in crude extracts, NADP + can readily be converted into NAD + by various phosphatases, or prephenate can unexpectedly be a substrate for a separate dehydrogenase, e.g. a broad-specificity lactate dehydrogenase of unknown significance that converts prephenate into prephenyl-lactate [ 48]. In most cases, where rigorously purified TyrA proteins have been characterized, they have been specific for NAD + or for NADP +. C. glutamicum TyrA a does exemplify, however, a well-documented case where either cofactor is accepted, although NADP + is favoured by almost an order of magnitude [ 8]. In view of the finding that replacement of T175 by asparagine in NADP-specific aldehyde dehydrogenase of Vibrio harveyi resulted in a highly increased utilization of NAD + without loss of ability to use NADP + [ 49], it is suggestive that the C. glutamicum TyrAa residue homologous with the crucial aspartate of NAD +-specific TyrA proteins is asparagine. These two dehydrogenases are further similar in that each still possesses a distinct preference for NADP +. So far, all of the rigorously characterized TyrAp (from B. subtilis) and TyrAc (from Z. mobilis, Ps. stutzeri and Ps. aeruginosa) proteins are NAD+-specific. Those cyclohexadienyl dehydrogenases that prefer prephenate over AGN by well over an order of magnitude (denoted TyrA(p)) are also NAD+-specific. These include the TyrA domains of AroQ•TyrA(p) proteins of the enteric lineage, and TyrA(p) from N. gonorrhoeae. There is a distinct tendency for prephenate-specific enzymes to prefer NAD+ and for AGN-specific enzymes to prefer NADP+. Perhaps, there is a structural relationship that favours interaction between the greater positive charge of AGN and the greater negative charge of NADP+ (relative to the prephenate/NAD+ couple). However, note that in the pseudomonad clade marked by the tyrA•aroF fusion, the Acinetobacter sp. TyrA is NADP+-specific, whereas the sister subclade Pseudomonas/Azotobacter exhibits NAD+ specificity. Thus the entire clade shares approximately the same profile of cyclohexadienyl substrate preference, even though cofactor specificity has been narrowed in opposite directions. Physiological ramifications of substrate specificity A striking feature of oxygenic photosynthetic prokaryotes and eukaryotes is that they have consistently proven to favour the AGN/NADP + pattern of specificity for TYR biosynthesis, regardless of their fundamental divergence with respect to peripheral antenna proteins and pigments utilized in the photosynthetic process [ 50, 51]. This includes cyanobacteria, Prochlorophyta, Rhodophyta, unicellular Chlorophyta [ 52] (Table 4), Euglenophyta [ 53] and multicellular Chlorophyta [ 15, 23, 33]. Since NADP + is the crucial electron acceptor during photosynthesis and since TYR biosynthesis and maximal growth take place in the light, the favoured utilization of NADP + may have evolved in response to the mechanisms that enhance the abundance of this cofactor during photosynthesis. Indeed, recent data have been reported to show that the intracellular levels of NADP + exceed those of NAD + by more than an order of magnitude in Synechocystis sp. strain PCC 6803 [ 54]. This is quite striking (a factor difference of approx. 30 or more) in comparison with the ‘typical’ NAD + to NADP + ratio of approx. 3–5 that is frequently cited in biochemical textbooks [ 55]. Pyrococcus furiosus has recently been shown [ 56] to possess an NADP + to NAD + ratio that is approx. 4-fold higher than in E. coli, and some enzymes that are generally NAD +-dependent are NADP +-dependent in P. furiosus. Thus it appears that the relative pool sizes of these redox cofactors may vary more in nature than previously considered. A potential physiological advantage of the insensitivity of Synechocystis TyrA a to inhibition by NADPH is that even if intracellular levels of NADPH are high as a result of redox flux, TyrA a is invulnerable to NADPH inhibition. For example, under conditions of high light and low CO 2, the NADPH produced in the photosynthetic light reaction can exceed its utilization with a concomitant relative increase in the reductive state [ 57]. Insensitivity of TyrA a to NADPH would allow TYR biosynthesis to be independent of such NADPH flux variations. A mutant lacking type I NADPH dehydrogenase has been isolated in Synechocystis sp. PCC 6803, and has been reported [ 54] essentially to lack oxidized NADP +. This mutant is not auxotrophic for TYR (W. Vermaas, personal communication), presumably because NADP + gets regenerated very quickly by such processes as CO 2 fixation. Nevertheless, it would be interesting to know whether the mutant might be bradytrophic or hypersensitive to TYR analogue inhibitors. This is Florida Agriculture Experimental Station Journal series number R-09164. 1. Todd A. E., Orengo C. A., Thornton J. M. Evolution of function in protein superfamilies, from a structural perspective. J. Mol. Biol. 2001;307:1113–1143. [PubMed] 2. Teichmann S. A., Rison S. C. G., Thornton J. M., Riley M., Gough J., Clothia C. The evolution and structural anatomy of the small molecule metabolic pathways in Escherichia coli. J. Mol. Biol. 2001;311:693–708. [PubMed] 3. Stenmark S. L., Pierson D. L., Glover G. I., Jensen R. A. Blue–green bacteria synthesize L-tyrosine by the pretyrosine pathway. Nature (London). 1974;247:290–292. [PubMed] 4. Patel N., Pierson D. L., Jensen R. A. Dual enzymatic routes to L-tyrosine and L-phenylalanine via pretyrosine in Pseudomonas aeruginosa. J. Biol. Chem. 1977;252:5839–5846. [PubMed] 5. Zamir L. O., Jensen R. A., Arison B., Douglas A., Albers-Schonberg G., Bowen J. R. Structure of arogenate (pretyrosine), an amino acid intermediate of aromatic biosynthesis. J. Am. Chem. Soc. 1980;102:4499–4504. 6. Xie G., Keyhani N. O., Bonner C. A., Jensen R. A. The ancient origin of the tryptophan operon and tracking the subsequent dynamics of evolutionary change. Microbiol. Mol. Biol. Rev. 2003;67:303–342. [PubMed] 7. Fazel A. M., Bowen J. R., Jensen R. A. Arogenate (pretyrosine) is an obligatory intermediate of L-tyrosine biosynthesis: confirmation in a microbial mutant. Proc. Natl. Acad. Sci. U.S.A. 1980;77:1270–1273. [PubMed] 8. Fazel A. M., Jensen R. A. Obligatory biosynthesis of L-tyrosine via the pretyrosine branchlet in coryneform bacteria. J. Bacteriol. 1979;138:805–815. [PubMed] 9. Fazel A. M., Jensen R. A. Regulation of prephenate dehydratase in coryneform species of bacteria by L-phenylalanine and by remote effectors. Arch. Biochem. Biophys. 1980;200:165–176. [PubMed] 10. Whitaker R. J., Byng G. S., Gherna R. L., Jensen R. A. Diverse enzymological patterns of phenylalanine biosynthesis in pseudomonads are conserved in parallel with deoxyribonucleic acid homology groupings. J. Bacteriol. 1981;147:526–534. [PubMed] 11. Xia T., Jensen R. A. A single cyclohexadienyl dehydrogenase specifies the prephenate dehydrogenase and arogenate dehydrogenase components of the dual pathways to L-tyrosine in Pseudomonas aeruginosa. J. Biol. Chem. 1990;265:20033–20036. [PubMed] 12. Calhoun D. H., Bonner C. A., Gu W., Xie G., Jensen R. A. The emerging periplasm-localized subclass of AroQ chorismate mutases, exemplified by those from Salmonella typhimurium and Pseudomonas aeruginosa. Genome Biol. 2001;2:0030.1–0030.16. 13. Zhao G., Xia T., Fischer R. S., Jensen R. A. Cyclohexadienyl dehydratase from Pseudomonas aeruginosa: molecular cloning of the gene and characterization of the gene product. J. Biol. Chem. 1992;267:2487–2493. [PubMed] 14. Ahmad S., Jensen R. A. The prephenate dehydrogenase component of the bifunctional T-protein in enteric bacteria can utilize L-arogenate. FEBS Lett. 1987;216:133–139. [PubMed] 15. Gaines C. G., Byng G. S., Whitaker R. J., Jensen R. A. L-tyrosine regulation and biosynthesis via arogenate dehydrogenase in suspension-cultured cells of Nicotiana silvestris (Speg. et Comes). Planta. 1982;156:233–240. 16. Connelly J. A., Conn E. E. Tyrosine biosynthesis in Sorghum bicolor: isolation and regulatory properties of arogenate dehydrogenase. Z. Naturforsch. 1986;41c:69–78. 17. Jensen R. A. Tyrosine and phenylalanine biosynthesis: relationship between alternative pathways, regulation and subcellular location. Recent Adv. Phytochem. 1986;20:57–82. 18. Hall G. C., Flick M. B., Gherna R. L., Jensen R. A. Biochemical diversity for biosynthesis of aromatic amino acids among the cyanobacteria. J. Bacteriol. 1982;149:65–78. [PubMed] 19. Xie G., Bonner C. A., Jensen R. A. Cyclohexadienyl dehydrogenase from Pseudomonas stutzeri exemplifies a widespread type of tyrosine-pathway dehydrogenase in the TyrA protein family. Comp. Biochem. Physiol. C. 2000;125:65–83. 20. Bonner C. A., Fischer R. S., Ahmad S., Jensen R. A. Remnants of an ancient pathway to L-phenylalanine and L-tyrosine in enteric bacteria: evolutionary implications and biotechnological impact. Appl. Environ. Microbiol. 1990;56:3741–3747. [PubMed] 21. Kaneko T., Sato S., Kotani H., Tanaka A., Asamizu E., Nakamura Y., Miyajima N., Hirosawa M., Sugiura M., Sasamoto S., et al. Sequence analysis of the genome of the unicellular cyanobacterium Synechocystis sp. strain PCC6803. II. Sequence determination of the entire genome and assignment of potential protein-coding regions. DNA Res. 1996;3:109–136. [PubMed] 22. Bonner C. A., Jensen R. A. Arogenate dehydrogenase. Methods Enzymol. 1987;142:488–494. [PubMed] 23. Bonner C. A., Jensen R. A. Prephenate aminotransferase. Methods Enzymol. 1987;142:479–487. [PubMed] 24. Bradford M. M. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein–dye binding. Anal. Biochem. 1976;72:680–685. 25. Laemmli U. K. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (London). 1970;227:680–685. [PubMed] 26. Byng G. S., Berry A., Jensen R. A. Evolutionary implications of features of aromatic amino acid biosynthesis in the genus Acinetobacter. Arch. Microbiol. 1985;143:122–129. [PubMed] 27. Northrup D. B. Rethinking fundamentals of enzyme action. Adv. Enzymol. Rel. Areas Mol. Biol. 1999;73:15–55. 28. Rudolph F. B., Fromn H. J. Plotting methods for analyzing enzyme rate data. Methods Enzymol. 1979;63:138–159. [PubMed] 29. Sampathkumar P., Morrison J. F. Chorismate mutase-prephenate dehydrogenase from Escherichia coli. Kinetic mechanism of the prephenate dehydrogenase reaction. Biochim. Biophys. Acta. 1982;702:212–219. [PubMed] 30. Rippert P., Matringe M. Purification and kinetic analysis of the two recombinant arogenate dehydrogenase isoforms of Arabidopsis thaliana. Eur. J. Biochem. 2002;269:4753–4761. [PubMed] 31. Hermes J. D., Tipton P. A., Fisher M. A., O'Leary M. H., Morrison J. F., Cleland W. W. Mechanisms of enzymatic and acid-catalyzed decarboxylations of prephenate. Biochemistry. 1984;23:6263–6275. [PubMed] 32. Subramaniam P., Bhatnagar R., Hooper A., Jensen R. A. The dynamic progression of evolved character states for aromatic amino acid biosynthesis in Gram-negative bacteria. Microbiology. 1994;140:3431–3440. [PubMed] 33. Rippert P., Matringe M. Molecular and biochemical characterization of an Arabidopsis thaliana arogenate dehydrogenase with two highly similar and active protein domains. Plant Mol. Biol. 2002;48:361–368. [PubMed] 34. Wierenga R. K., Terpstra P., Hol W. G. J. Prediction of the occurrence of the ADP-binding βαβ-fold in proteins, using an amino-acid sequence fingerprint. J. Mol. Biol. 1986;187:101–107. [PubMed] 35. Fan F., Lorenzen J. A., Plapp B. V. An aspartate residue in yeast alcohol dehydrogenase I determines the specificity for coenzyme. Biochemistry. 1991;30:6397–6401. [PubMed] 36. Feeney R., Clarke A. R., Holbrook J. J. A single amino acid substitution in lactate dehydrogenase improves the catalytic efficiency with an alternative coenzyme. Biochem. Biophys. Res. Commun. 1990;166:667–672. [PubMed] 37. Carugo O., Argos P. NADP-dependent enzymes. I: conserved stereochemistry of cofactor binding. Proteins: Struct. Funct. Genet. 1997;28:10–28. [PubMed] 38. Blanc V., Gil P., Bamasjacques N., Lorenzon S., Zagorec M., Schleuniger J. Identification and analysis of genes from Streptomyces pristinaespiralis encoding enzymes involved in the biosynthesis of the 4-dimethylamino-L-phenylalanine precursor of pristinamycin I. Mol. Microbiol. 1997;23:191–202. [PubMed] 39. Turnbull J. L., Morrison J. F., Cleland W. W. Kinetic studies on chorismate mutase-prephenate dehydrogenase from Escherichia coli: models for the feedback inhibition of prephenate dehydrogenase by L-tyrosine. Biochemistry. 1991;30:7783–7788. [PubMed] 40. Zhao G., Xia T., Ingram L., Jensen R. A. An allosterically insensitive class of cyclohexadienyl dehydrogenase from Zymomonas mobilis. Eur. J. Biochem. 1993;212:157–165. [PubMed] 41. Xie G., Bonner C. A., Brettin T. T., Gottardo R., Keyhani N., Jensen R. A. Lateral gene transfer and ancient paralogy of operons containing redundant copies of tryptophan-pathway genes in Xylella species and heterocystous cyanobacteria. Genome Biol. 2003;4:R14. [PubMed] 42. Christendat D., Saridakis V. C., Turnbull J. L. Use of site-directed mutagenesis to identify residues specific for each reaction catalyzed by chorismate mutase-prephenate dehydrogenase from Escherichia coli. Biochemistry. 1998;37:15703–15712. [PubMed] 43. Ryding N. J., Anderson T. B., Champness W. C. Regulation of the Streptomyces coelicolor calcium-dependent antibiotic by absA, encoding a cluster-linked two-component system. J. Bacteriol. 2002;184:794–805. [PubMed] 44. Christendat D., Turnbull J. L. Identifying groups involved in the binding of prephenate to prephenate dehydrogenase from Escherichia coli. Biochemistry. 1999;38:4782–4793. [PubMed] 45. Baker P. J., Britton K. L., Engel P. C., Farrants G. W., Lilley K. S., Rice D. W., Stillman T. S. Subunit assembly and active site location in the structure of glutamate dehydrogenase. Proteins. 1999;12:75–86. [PubMed] 46. Miñambres B., Olivera E. R., Jensen R. A., Luengo J. M. A new class of glutamate dehydrogenases (GDH). Biochemical and genetic characterization of the first member, the AMP-requiring NAD-specific GDH of Streptomyces clavuligerus. J. Biol. Chem. 2000;275:39529–39542. [PubMed] 47. Byng G. S., Whitaker R. J., Gherna R. L., Jensen R. A. Variable enzymological patterning in tyrosine biosynthesis as a means of determining natural relatedness among the Pseudomonadaceae. J. Bacteriol. 1980;144:247–257. [PubMed] 48. Zamir L. O., Tiberio R., Devor K. A., Sauriol F., Ahmad S., Jensen R. A. Structure of D-prephenyllactate. A carboxycyclohexadienyl metabolite from Neurospora crassa. J. Biol. Chem. 1988;263:17284–17290. [PubMed] 49. Zhang L., Ahvazi B., Szittner R., Vrielink A., Meighen E. Change of nucleotide specificity and enhancement of catalytic efficiency in single point mutants of Vibrio harveyi aldehyde dehydrogenase. Biochemistry. 1999;38:11440–11447. [PubMed] 50. Tomitani A., Okada K., Miyashita H., Matthijs H. C. P., Ohno T., Tanaka A. Chlorophyll b and phycobilins in the common ancestor of cyanobacteria and chloroplasts. Nature (London). 1999;400:159–162. [PubMed] 51. Grabowski B., Cunningham J. F. X., Gantt E. Chlorophyll and carotenoid binding in a simple red algal light-harvesting complex crosses phylogenetic lines. Proc. Natl. Acad. Sci. U.S.A. 2001;98:2911–2916. [PubMed] 52. Bonner C. A., Fischer R. S., Schmidt R. R., Miller P. W., Jensen R. A. Distinctive enzymes of aromatic amino acid biosynthesis that are highly conserved in land plants are also present in the chlorophyte alga Chlorella sorokiniana. Plant Cell Physiol. 1995;36:1013–1022. 53. Byng G. S., Whitaker R. J., Shapiro C. L., Jensen R. A. The aromatic amino acid pathway branches at L-arogenate in Euglena gracilis. Mol. Cell. Biol. 1981;1:426–438. [PubMed] 54. Cooley J. W., Vermaas W. F. J. Succinate dehydrogenase and other respiratory pathways in thylakoid membranes of Synechocystis sp. strain PCC 6803: capacity comparisons and physiological function. J. Bacteriol. 2001;183:4251–4258. [PubMed] 55. Kaplan N. O. The pyridine coenzymes. In: Boyer P. D., Lardy H., Myrback K., editors. The Enzymes, vol. 3. 2nd edn. New York: Academic Press; 1960. pp. 105–169. 56. Pan G., Verhagen M. F. J. M., Adams M. W. W. Characterization of pyridine nucleotide coenzymes in the hyperthermophilic archaeon Pyrococcus furiosus. Extremophiles. 2001;5:393–398. [PubMed] 57. Heber U., Bligny R., Streb P., Douce R. Photorespiration is essential for the protection of the photosynthetic apparatus of C3 plants against photoinactivation under sunlight. Bot. Acta. 1996;109:307–315. 58. Jensen R. A., Gu W. Evolutionary recruitment of biochemically specialized subdivisions of family I within the protein superfamily of aminotransferases. J. Bacteriol. 1996;178:2161–2171. [PubMed]
|
This article has been 
