THE JOURNAL OF BIOLOGICAL CHEMISTRY VOL. 289, NO. 39, pp. 27034 –27045, September 26, 2014 © 2014 by The American Society for Biochemistry and Molecular Biology, Inc. Published in the U.S.A.

Allosteric Regulation of a Protein Acetyltransferase in Micromonospora aurantiaca by the Amino Acids Cysteine and Arginine* Received for publication, May 7, 2014, and in revised form, August 11, 2014 Published, JBC Papers in Press, August 14, 2014, DOI 10.1074/jbc.M114.579078

Jun-Yu Xu, Di You, Pei-Qiang Leng, and Bang-Ce Ye1 From the Lab of Biosystems and Microanalysis, State Key Laboratory of Bioreactor Engineering, East China University of Science and Technology, Shanghai 200237, China Background: Reversible lysine acetylation of proteins is ubiquitous in actinomycetales. Results: Arginine and cysteine allosterically regulate the protein lysine acetyltransferase Micau_1670 in Micromonospora aurantiaca. Conclusion: The amino acid-binding domain is fused to GCN5-related acetyltransferases, conferring amino acid-induced allosteric regulation to these enzymes. Significance: Activities mediated by amino acids in these acetyltransferases directly link amino acid metabolism to cellular acetylation of proteins. ACT domains (amino acid-binding domains) are linked to a wide range of metabolic enzymes that are regulated by amino acid concentration. Seventy proteins with ACT-GCN5-related N-acetyltransferase (GNAT) domain organization were found in actinomycetales. In this study, we investigate the ACT-containing GNAT acetyltransferase, Micau_1670 (MaKat), from Micromonospora aurantiaca ATCC 27029. Arginine and cysteine were identified as ligands by monitoring the conformational changes that occur upon amino acids binding to the ACT domain in the MaKat protein using FRET assay. It was found that MaKat is an amino acid-regulated protein acetyltransferase, whereas arginine and cysteine stimulated the activity of MaKat with regard to acetylation of acetyl-CoA synthetase (Micau_0428). Our research reveals the biochemical characterization of a protein acetyltransferase that contains a fusion of a GNAT domain with an ACT domain and provides a novel signaling pathway for regulating cellular protein acetylation. These findings indicate that acetylation of proteins and acetyltransferase activity may be tightly linked to cellular concentrations of some amino acids in actinomycetales.

The dynamic and reversible N⑀-lysine acetylation of proteins is now recognized as a ubiquitous and conserved post-translational modification from prokaryotes to eukaryotes. Recent studies have identified over 3,000 acetylated proteins, ranging from transcriptional factors and ribosomal proteins to many metabolic enzymes that are related to glycolysis, gluconeogenesis, the TCA cycles, and fatty acid, as well as nitrogen metab-

* This

work was supported by China National Science Foundation Grant 21276079, Chinese Ministry of Education Grant Specialized Research Fund for the Doctoral Program 20120074110009, National Key Technologies R&D Program Grants 2014AA021502 and 2007AA02Z331, and grants from the Fundamental Research Funds for the Central Universities. 1 To whom correspondence should be addressed: Lab of Biosystems and Microanalysis, State Key Laboratory of Bioreactor Engineering, East China University of Science and Technology, Shanghai 200237, China. Tel./Fax: 8621-6425-2094; E-mail: [email protected].

27034 JOURNAL OF BIOLOGICAL CHEMISTRY

olism. This kind of post-translational modification has been emerging as an important metabolic regulatory mechanism in bacteria since the discovery of acetylation of the Salmonella enterica acetyl-CoA synthetase in 2002 (1). In the last decade, lysine acetylation of proteins has been reported in other microorganisms, including Escherichia coli, Bacillus subtilis, and Mycobacterium tuberculosis (2, 3). All acetylation/deacetylation systems in prokaryotes consist of protein acetyltransferases and deacetylases. Acetyltransferases catalyze the transfer of the acetyl group from the acetylcoenzyme A (Ac-CoA)2 donor to a primary amine of small molecules and proteins that are involved in a wide variety of cellular processes. Protein acetyltransferases specifically control acetylation of different proteins under various physiological conditions. Protein deacetylases play a role in globally removing the acetyl group from acetylated proteins that respond to changes of cellular energy status via promptly sensing the level of intracellular NAD⫹:NADH ratio. Despite our growing understanding of reversible lysine acetylation, it remains unclear how many protein acetyltransferases and deacetylases are involved in this post-translational modification, how signals regulate the activity of the acetylation/deacetylation system, and how a limited number of acetyltransferases and deacetylases control the acetylation of so many metabolic enzymes. It is likely that the protein acetyltransferase enzymes are carefully regulated at the transcriptional and posttranslational levels in response to changes of the intracellular signals that control the acetylation of specific proteins, which in turn mold the metabolic network. This hypothesis is supported by previous research. In the enteric bacteria E. coli, the transcription of protein acetyltransferase PatZ is controlled at the transcriptional level in response to the intracellular cAMP signal. The catabolite activator protein-cAMP complex binds to two sites in the patZ promoter

2

The abbreviations used are: Ac-CoA, acetyl-coenzyme A; GNAT, GCN5-related N-acetyltransferase; CFP, cyan fluorescence protein.

VOLUME 289 • NUMBER 39 • SEPTEMBER 26, 2014

Amino Acid-mediated Protein Lysine Acetyltransferase and induces the expression of genes that increase the level of acetylated proteins (4). The activity of protein acetyltransferases is also controlled by allosteric effects. In M. tuberculosis and Mycobacterium smegmatis, cAMP directly activates the protein acetyltransferases MtKat (Rv0998) and MsKat (MSMEG_5458) by binding to a cyclic nucleotide-binding domain that is fused to the N terminus of the catalytic GNAT domain (5, 6). It was previously known that the protein acetyltransferases have two different domain organizations: multidomain and single-domain. The multidomain acetyltransferases can be regulated at the transcriptional and post-translational levels; however, the single-domain acetyltransferases are generally controlled at the transcriptional level. For example, the regulation of expression of the B. subtilis acuA gene is under the control of the global regulatory protein CcpA, which is affected by the quality of the carbon source that is available to the cell (7). The ACT domain is recognized as a structurally conserved motif that consists of four ␤ strands and two ␣ helices that are arranged in a ␤␣␤␤␣␤ fold. The amino acid-binding domains are mostly found in a variety of proteins that are directly or indirectly involved in amino acid and purine metabolism (8). The name “ACT” originates from three of the proteins that contain the ACT domain: aspartate kinase, chorismate mutase, and TyrA (prephenate dehydrogenase). In the Pfam database (v.27.0), 20,885 sequences containing ACT domains (PF01842) with 130 domain architectures were recently collected. In the InterPro database (v.47), a total of 107,722 ACT domain-containing proteins (IPR002912) were matched, most of which (95%) were discovered in bacteria. This domain is fused to a variety of different functional domains, thus producing 289 domain organizations. According to the hypothesis proposed by Aravind and Koonin (9), the ACT domain, as the conserved, evolutionarily mobile module, was suggested to be a provider of allosteric effect. The ACT domain was independently fused to a variety of enzymes, conferring ligand-induced allosteric regulation to these enzymes. The majority of these enzymes have not been studied in detail. In this study, we found that 70 proteins contain the ACTGNAT domain organization. These proteins have all been discovered in actinomycetales, indicating that this domain organization is unique to actinomycetales. It is perhaps expected that the binding of amino acids allosterically regulates the activity of the acetyltransferases with ACT-GNAT domain organization. We further investigated the first ACT-containing GCN5-related acetyltransferase, Micau_1670, from Micromonospora aurantiaca ATCC 27029. Micromonospora species are ubiquitous in soils, sediments, and aquatic environments and in the rumen of cattle and the guts of termites; moreover, they have been recognized as important sources of antibiotics (e.g. aminoglycoside antibiotics, gentamicin, netamicin, lomaiviticins A and B, tetrocarcin A, and the LL-E33288 complex) (10). Arginine and cysteine were identified as ligands by monitoring the conformational changes that occur upon amino acids binding to the ACT domain in the Micau_1670 protein using a FRET biosensor. We observed that the two amino acids stimulated the activity of protein lysine acetyltransferase Micau_1670 on acetylation of acetyl-CoA synthetase (Micau_0428), MaAcs. SEPTEMBER 26, 2014 • VOLUME 289 • NUMBER 39

Furthermore, the important residues implicated in binding of Cys or Arg in the ACT domain of Micau_1670 were investigated. These findings indicate that acetylation of proteins and acetyltransferase activity may be tightly linked to cellular concentrations of some amino acids in actinomycetales.

EXPERIMENTAL PROCEDURES Cloning, Overexpression, and Purification of Proteins—The Micau_1670 and Micau_0428 genes were amplified by PCR from the genomic DNA of M. aurantiaca ATCC 27029 using two primer pairs (5⬘-TAAGAATTCATGAGCGAG GCATTGGCCAACT/TAAAAGCTTTCAGTCCTCGTCGGACTTCCCG-3⬘ and 5⬘-TAAGAATTCATGGCGCTCTGGCGGATC/TAAAAGCTTACTCGACCGCTGTGCCGG-3⬘). After restriction digestion with HindIII and EcoRI, the genes coding for Micau_1670 and Micau_0428 were cloned into pET-28a to generate PET28a-Micau_1670 and PET28a-Micau_0428. The clones were confirmed by sequencing. The proteins were expressed using the E. coli BL21 (DE3) strain. A single colony was selected to begin a 3-ml overnight culture, which was then used for inoculation in 50 ml of Luria Bertani medium that was supplemented with 1% kanamycin. The cells were grown at 37 °C and then induced with 0.7 mM isopropyl-␤-D-thiogalactoside at 20 °C overnight. Cells were harvested by centrifugation and resuspended in PBS buffer (137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4,1.8 mM KH2PO4, pH 7.4) and incubated on ice for 15 min. The cells were sonicated in PBS buffer, and cell debris was removed by centrifugation at 8,000 rpm for 20 min. The supernatant was purified with a nickel-nitrilotriacetic acidagarose column (Merck), and bound protein was pre-equilibrated with the binding buffer. After discarding the flow through, the column was washed with 10 ml of wash buffer (50 mM NaH2PO4, 300 mM NaCl, and 20 mM imidazole, pH 8.0) to remove the hybrid proteins, and bound proteins were eluted using a linear gradient from 20 to 250 mM imidazole in 50 mM NaH2PO4 and 300 mM NaCl, pH 8.0. The fractions were analyzed by SDS-PAGE, and those containing the desired protein were pooled and dialyzed against buffer P (37 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, 1.8 mM KH2PO4, 5% glycerol, pH 7.9). The His tag of Micau_0428 was digested by thrombin (4 °C overnight). The protein was then concentrated using an Amicon Ultra-4 30,000 cutoff centrifugal device (Millipore, Billerica, MA). Protein concentration was monitored by the BCA method using buffer P as the control, and the amount of protein after concentration was also analyzed by SDS-PAGE. Generating the Constructs Used for FRET Analysis—To construct FRET biosensors, the genes coding for Micau_1670 and Micau_1670(⌬ACT) were all cloned into PET28a-YFP-CFP that was digested with EcoRI and HindIII to generate recombinant plasmid PET28a-YFP-Micau_1670-CFP and PET28aYFP-Micau_1670(⌬ACT)-CFP. Micau_1670(⌬ACT) represented the region that encompasses residues 151–362 of Micau_1670. The constructs were confirmed by DNA sequencing. The proteins are expressed in the same way as described above. The FRET biosensors were constructed according to the procedure that was previously reported by our laboratory (11). JOURNAL OF BIOLOGICAL CHEMISTRY

27035

Amino Acid-mediated Protein Lysine Acetyltransferase Site-directed Mutagenesis and Purification of Micau_0428 and Micau_1670 —The D11A, L16A, and D45A mutations were separately introduced into the recombinant plasmid PET28a-Micau_1670, as well as PET28a-YFP-Micau_1670CFP, using a QuikChange mutagenesis kit (TransGen, Beijing, China). The K619Q mutation for PET28a-Micau_0428 was introduced in the same way. All mutations were confirmed by DNA sequencing. The expression and purification of the mutants were performed following the same procedure described above. In Vitro Protein Acetylation Assays—The protein concentrations of the samples were determined using the BCA protein assay kit (Pierce) with BSA as the standard. Assays were carried out in a 100-␮l total reaction volume. MaAcs (Micau_0428) (1.5 ␮M) was incubated in the presence of Micau_1670 (0.2 ␮M), acetyl-CoA (60 ␮M), and Cys or Arg (1 mM) at 37 °C for 60 min. The reaction contained 0.05 M HEPES buffer (pH 7.5). After reaction, the acetylated MaAcs was isolated from the reaction mixture by gel filtration (adding nickel beads to the reaction and the MaKat binds to the beads) and ultrafiltration (removing micromolecules such as acetyl-CoA), which can be used in the measurement of the kinetic parameters. Acetylation of MaAcs by MaKat was monitored by SDS-PAGE, Western blot, and activity analysis of the acetyl-CoA synthetase. In Vitro Steady-state Kinetic Assays—To continuously monitor the acetylation reaction, we utilized a coupled enzymatic assay to measure the kinetic parameters of MaKat, in which CoA, a product of the acetylation reaction, was used by pyruvate dehydrogenase to convert NAD⫹ to NADH, resulting in an absorbance increase at 340 nm (12, 13). This enzyme-coupled assay condition was optimized for each enzyme and substrate to avoid other rate-limiting factors. Finally, the standard reaction mixture contained 0.2 mM NAD, 0.2 mM thiamine pyrophosphate, 5 mM MgCl2, 1 mM DTT, 2.4 mM pyruvate, 50 ␮M acetyl-CoA, 50 ␮M MaAcs, 0.03 units of pyruvate dehydrogenase, 0.65 ␮M MaKat with or without amino acids, 100 mM sodium acetate, 50 mM Bis-Tris, and 50 mM Tris, pH 7.5, in a total volume of 300 ␮l. All assay components except MaKat were incubated at 25 °C for 5 min, and the reaction was initiated by the addition of MaKat. The rates were analyzed continuously for 2 min by measuring NADH production at 340 nm. The initial velocities were measured at varying concentrations of one substrate while maintaining a fixed saturating concentration of the other one. Each data point was the average of three independent assays. The data were then fitted into the Michaelis-Menten equation using GraphPad Prism 5 to obtain Km and kcat. Western Blotting Assays—Samples were electrophoresed on SDS-polyacrylamide gels (12% acrylamide, 1.2% bis-acrylamide) and transferred to a PVDF membrane for 60 – 80 min at 100 V. The membrane was blocked in BSA blocking buffer for 120 min. Acetyl-lysine antibody (catalog number ICP0380) from ImmuneChem Pharmaceuticals Inc. (Burnaby, Canada) was used at a dilution of 1:15,000. After incubation at 4 °C overnight, the blot was washed with TBST buffer (20 mM Tris-HCl, pH 7.5, 150 mM NaCl, and 0.1% Tween 20) three times (once every 10 min), and bound antibody was detected by enhanced chemiluminescence.

27036 JOURNAL OF BIOLOGICAL CHEMISTRY

Circular Dichroism Spectroscopy—Circular dichroism spectra were obtained on a CD spectrophotometer (Applied Photophysics, Leatherhead, UK) using a 10-mm cuvette that was equilibrated at 25 °C. All samples were desalted, and the buffer was exchanged with PBS buffer (pH 8.0) through a ZebaTM desalt spin column (Thermo Scientific, Pierce). Sample concentrations were diluted to 0.035 ␮g/␮l. The CD spectra were the average of three scans that were obtained by collecting data between 195 and 280 nm. Buffer contributions were subtracted from all spectra.

RESULTS MaKat Containing ACT-GNAT Domain Organization Is a Protein Acetyltransferase—In the InterPro database (v.47), 70 proteins containing the domain organization of ACT-GNAT were discovered in actinomycetales. The N-terminal ACT domain is fused to a catalytic GNAT domain (Fig. 1A). A phylogenetic analysis of the full-length sequences of all ACTGNAT proteins is shown in Fig. 2. Additionally, multiple alignment of eight proteins was performed using ClustalX (Fig. 1B) and revealed that these proteins contain the conserved ␤␣␤␤␣␤ fold of the ACT domain. In particular, ␤1, ␣1, ␤2, and ␤3, as the core of the ACT domain, are the most highly conserved motifs that are important for amino acid recognition and binding. Four ACT-GNAT proteins were found in Micromonospora (Micau_1670, MCAG_04541, ML5_1930, and MILUP08_41772). The high conservation among ACT domains of Micromonospora proteins indicates that they would have similar ligand binding. In this work, we studied the acetyltransferase Micau_1670 as a representative model of the ACT-GNAT enzymes. The ACT domain of Micau_1670 was also compared with the representative ACT domains, including those whose structures have been determined, such as AHAS/ILVN, ASPK, PGDH, Glyc, PHE, PURU, GLND, and SPOT (Fig. 1C). The multiple alignments revealed that the ACT domains have low conservation in the sequence. However, all have the conserved fold of secondary structures, which are sequentially labeled ␤1, ␣1, ␤2, ␤3, ␣2, and ␤4. The crystal structure studies of ACT domains with ligands illustrated that the amino acids tend to bind at the interfaces between ACT domains, and there appears to be a correlation of the ligand binding sites with specific glycine residues that is located in loops between the first strand, ␤1, and the first helix, ␣1 (14). As shown in Fig. 1B, ACT-GNAT acetyltransferases indeed contain a conserved glycine residue (Gly-14 of Micau_1670) at this location. In prokaryotes, some acetyltransferases for protein acetylation have been investigated, including SePat in S. enterica (15), EcPatZ in E. coli (4), RpPat in Rhodopseudomonas palustris (16), MtKat in M. uberculosis, MsKat in M. smegmatis (6), and SlPatA in Streptomyces lividans (17). These protein acetyltransferases containing a GNAT domain can acetylate acetyl-CoA synthetases. Comparison of the amino acid sequence of the GNAT domain in Micau_1670 (named for MaKat) with other representative GNAT domains in various proteins is shown in Fig. 3A. The results of this study reveal that MaKat contains some conserved pivotal residues in motif A of the GNAT domain. VOLUME 289 • NUMBER 39 • SEPTEMBER 26, 2014

Amino Acid-mediated Protein Lysine Acetyltransferase

FIGURE 1. Sequence alignment of ACT domains. A, domain organization of Micau_1670. B, amino acid sequence alignment of ACT domains of eight proteins: Micau_1670 from M. aurantiaca ATCC 27029, MCAG_04541 from Micromonospora sp. ATCC 39149, ML5_1930 from Micromonospora sp. strain L5, VAB18032_12965 from Verrucosispora maris, ACPL_7045 from Actinoplanes sp. strain ATCC 31044, AMIS_67970 from Actinoplanes missouriensis, MILUP08_41772 from Micromonospora lupini strain Lupac 08, and L083_6630 from Actinoplanes sp. N902–109. C, ACT domain of Micau_1670 was also compared with the representative ACT domains including those whose structures have been determined such as AHAS/ILVN (acetolactate synthase I/III small subunit), ASPK (aspartate kinase), PGDH (1,3-phosphoglycerate dehydrogenase), Glyc (glycine cleavage system transcription regulator), PHE (prephenate dehydratases), PURU (formyltetrahydrofolate deformylases), GLND (uridylyl transferases), and SPOT (guanosine polyphosphate 3-pyrophosphohydrolases).

Motif A, as the core of the GNAT domain, is the most highly conserved motif and generally has an (R/Q)XXGX(G/A) sequence that is important for acetyl-CoA recognition and binding. The QRRGLG sequence is indeed found in motif A of MaKat. Interestingly, GNAT domain contains a nearly invariant Glu residue (Glu-269 in MaKat, Glu-235 in MtKat, or Glu234 in MsKat) in the cleft between motifs A and D. This most conserved amino acid provides an active site that serves as a key base to deprotonate the lysine residue of the acetyl acceptor. Furthermore, a phylogenetic analysis with the GNAT domain sequence of these acetyltransferases showed MaKat clusters with MsKat, MtKat, and SlPat (Fig. 3B), thus indicating that MaKat is able to acetylate acetyl-CoA synthetase in M. aurantiaca such as MsKat, MtKat, and SlPat that occur in other actinomycete species. Moreover, we confirmed that MaKat directly acetylates acetyl-CoA synthetase MaAcs (Micau_0428) in vitro by incubating purified MaKat with Ac-CoA and recombinant Micau_0428 of M. aurantiaca. As shown in Fig. 3C, this result strongly indicates that MaKat has protein lysine acetyltransferase activity for acetylation of acetyl-CoA synthetase. To test the effect of acetylation on enzyme activity, MaAcs was incubated with MaKat in the presence or absence of acetyl-CoA for 2 h. In the presence of both acetyl-CoA and MaKat, MaAcs activity was reduced, indicating that MaKat lysine acetylation effectively decreases MaAcs activity (Fig. 3D). Time-dependent inactivation of MaAcs by MaKat acetylation was also investigated. In Fig. 3E, it is observed that MaAcs gradually lost its activity during acetylation by the MaKat enzyme. Conformational Changes of MaKat Induced by Cys and Arg Are Monitored by FRET—It was known that the ACT domain is involved in binding of specific amino acids and potentially provides allosteric regulation via transmission of finely tuned conformational changes, leading to the change of the activity of the SEPTEMBER 26, 2014 • VOLUME 289 • NUMBER 39

catalytic domain. The ACT-GNAT domain organization of MaKat strongly indicates that MaKat has protein acetyltransferase activity with allosteric regulation altering its activity by binding to one or several specific amino acids. First, we tried to utilize genetically encoded FRET-based biosensors to screen the putative amino acid ligands that can change the configuration of MaKat among all 20 L-amino acids that exist in nature. The intact MaKat protein from M. aurantiaca was used as the binding unit, and it was sandwiched directly with enhanced CFP and YFP. Conformational change in the MaKat protein is likely to lead to a change in FRET efficiency between CFP and YFP in this sandwich configuration. Two FRET-based biosensors were constructed by flanking the ACT-GNAT domain (intact MaKat protein) and GNAT domain (MaKat protein without ACT domain, as a control) with CFP and YFP to investigate the conformational change of the ACT domain in response to amino acids (Fig. 4A). Two biosensors were excited at 440 nm, and their emission spectra showed two peaks at 478 and 528 nm, which correspond to CFP and YFP. As shown in Fig. 4B and 4C, the addition of Arg, Cys, and Asn resulted in apparent changes of the 478/528 ratios of the CFP and YFP emission intensities between two biosensors. The changes in the 478/528 ratio induced by amino acids indicate that conformation changes that result from the ligandbinding ACT domain are translated into a change in FRET efficiency. To further investigate whether amino acids regulate the activity of MaKat, Western blotting was conducted to detect the acetylation level of MaAcs with anti-acetyl-lysine antibody in the presence or absence of amino acids. As shown in Fig. 4D, Arg and Cys were able to increase the rate of acetylation of MaKat, which is consistent with the results obtained from the FRET method. However, no similar effect was observed for asparagine. JOURNAL OF BIOLOGICAL CHEMISTRY

27037

Amino Acid-mediated Protein Lysine Acetyltransferase

FIGURE 2. Phylogenetic analysis. All proteins containing the domain organization of ACT-GNAT were discovered in actinomycetales using the InterPro database (v.46). A phylogenetic analysis with the full-length sequences of all ACT-GNAT proteins is presented.

27038 JOURNAL OF BIOLOGICAL CHEMISTRY

VOLUME 289 • NUMBER 39 • SEPTEMBER 26, 2014

Amino Acid-mediated Protein Lysine Acetyltransferase

FIGURE 3. MaKat is a protein acetyltransferase. A, multiple sequence alignment of protein acetyltransferases in both prokaryotes and eukaryotes, including 1Z4RA in human beings, 1QSNA in tetrahymena, 1YGH_A in yeast, MsKat in M. smegmatis, MtKat in M. tuberculosis, SIPat in S. lividans, and MaKat in M. aurantiaca. B, The phylogenetic analysis was conducted with GNAT domains in the above acetyltransferases. C, the purified MaAcs was in vitro incubated with or without MaKat and Ac-CoA at 37 °C for 2 h. After incubation, samples were collected and analyzed by SDS-PAGE, and the acetylation levels were determined by Western blot using specific anti-AcK antibody. D, in vitro acetylation affected the activity of MaAcs. The enzyme activity of MaAcs after incubation with or without Ac-CoA in the presence of MaKat for the indicated amount of time was measured. The MaAcs activity is described as a percentage of the maximum activity determined for MaAcs before acetylation. The data are expressed as means ⫾ S.D. for three identical assays. E, time-dependent inactivation of MaAcs (1.5 ␮M) by acetylation. MaAcs activity was measured at different time intervals during incubation with MaKat. E, 0.2 ␮M MaKat and 60 ␮M Ac-CoA; , 0.2 ␮M MaKat; Œ, 60 ␮M Ac-CoA. Each data point represents the average from three repeated assays.

Furthermore, the changes in the 478/528 ratio of the ACTGNAT biosensor are amino acid concentration-dependent. When the Cys concentration was changed from 10 nM to 100 mM, the emission ratios increased, following sigmoid curves; alternatively, when the Arg concentration changed from 10 nM to 100 mM, the emission ratios decreased, following “Z” curves (Fig. 5, A and B). Cys and Arg showed differential effects on ACT conformation. On the other hand, Cys binds to the ACT domain in MaKat with an affinity of EC50 ⫽ 80 ␮M, whereas Arg shows a much lower affinity of interaction (IC50 ⫽ 210 ␮M). Acetyl-CoA and MaAcs are substrates of MaKat. We also monitored FRET ratio of the ACT-GNAT biosensor in the presence of acetyl-CoA and MaAcs. As shown in Fig. 5 (C and D), no change in FRET ratio was observed in the presence of either substrate; meanwhile, MaAcs did not alter the FRET ratios that are induced by Cys and Arg in the sensor, which demonstrated that substrate binding to the catalytic site in the GNAT domain did not affect the FRET ratio. To confirm that the increase in FRET ratio observed in the presence of Cys was a consequence of Cys binding to the ACT domain, we investigated the effect of two compounds, DTT (reductant) and mercaptoethanol (thiol compound), on the FRET ratio of ACT-GNAT biosensor. No increase in the FRET ratio was observed upon the addition of two compounds (Fig. 5C). The FRET ratios of ACT-GNAT biosensors were tested for serine and lysine, which have similar structures to Cys and Arg, respectively. SEPTEMBER 26, 2014 • VOLUME 289 • NUMBER 39

MaKat Could Function as an Amino Acid-regulated Protein Acetyltransferase—Bioinformatics analysis and Western blotting assay demonstrated that MaKat is a protein lysine acetyltransferase and can acetylate the acetyl-CoA synthetase MaAcs. As seen in Fig. 6A, acetylation of MaAcs was observed when incubated with MaKat and acetyl-CoA, and acetylation increased in the presence of Cys or Arg. The Cys/Arg-binding ACT domain fused to the acetyltransferase domain in MaKat, and an increase in acetylation level of MaAcs in the presence of Cys or Arg indicates that the ACT domain may allosterically regulate the activity of the GNAT domain, and MaKat could function as an amino acid-regulated protein acetyltransferase. Furthermore, in the presence of Cys or Arg, MaKat activity exhibits a ligand concentration-dependent increase for protein acetylation (Fig. 6B). The concentrations of amino acid ligands that would saturate acetylation of protein were in the submicromolar range, which is comparable with the EC50 or IC50 concentrations that were obtained from the FRET assay. The previously reported bacterial acetyltransferases exhibit specificity of the lysine site for protein acetylation. There is a proposed acetylation motif (PXXXXGK) that is found in AMPforming acyl-CoA synthetases (14). GNAT acetyltransferases recognize this motif and acetylate the last lysine residue of PXXXXGK, such as Lys-609 of SeAcs from S. enterica, Lys-606 of RpAcs from R. palustris, Lys-617 of MtAcs from M. tuberculosis, and Lys-549 of BsAcs from B. subtilis (Fig. 7A). JOURNAL OF BIOLOGICAL CHEMISTRY

27039

Amino Acid-mediated Protein Lysine Acetyltransferase

FIGURE 4. Screening of MaKat ligands. A, the whole sequence of Micau_1670 (amino acids 1–361) and the truncated sequence of Micau_1670 (amino acids 150 –361) were sandwiched in the FRET pair YFP/CFP. B, the basal FRET ratio was normalized, and the ratio was calculated to evaluate the specific amino acid response to two sensors. C, comparison of the emission ratio between two sensors in the presence of Cys and Arg. D, the effect of amino acids on enzyme activity of MaKat. The reaction involves MaAcs (1.5 ␮M), MaKat (0.2 ␮M), acetyl-CoA (60 ␮M), and one of the 20 amino acids. Basal mean was observed by adding PBS buffer.

To confirm the site of acetylation, we created substitution mutations at the Lys-619 position to generate K619Q variant of MaAcs. Glutamine abolishes the positive charge and serves as a structural mimic for acetyl-lysine. MaAcsK619Q and MaAcsWT were incubated with the MaKat enzyme and Ac-CoA. Western blotting was conducted to detect the acetylation level of two MaAcs enzymes with anti-acetyl-lysine antibody. As shown in Fig. 7B, acetylation of only MaAcsWT was observed. No acetylation was observed at MaAcsK619Q, indicating that MaKat modified only conserved the lysine residue Lys-619 of the active site in MaAcs. To test the effect of Cys/Arg on the MaKat enzyme activity, we determined the initial rate of acetylation of MaAcs in the

27040 JOURNAL OF BIOLOGICAL CHEMISTRY

absence and presence of Cys/Arg. The acetylation reaction of MaKat was monitored by a coupled enzymatic assay where the amount of CoA that is liberated following acetylation is measured by formation of reduced NADH from NAD⫹ by pyruvate dehydrogenase (12, 13). As shown in Fig. 8, MaKat could acetylate MaAcs in the absence of amino acid ligand (4.6 ⫾ 0.9 nmol of NADH formed/min/ml), and the rate of acetylation was increased 208 and 150% in the presence of Cys (9.6 ⫾ 1.5 nmol of NADH formed/min/ml) and Arg (6.9 ⫾ 1.2 nmol of NADH formed/min/ml), respectively. We further determined the kinetic parameters of the MaKat catalyzed acetylation reaction of MaAcs in the absence and presence of Cys/Arg. We used a coupled enzymatic assay to VOLUME 289 • NUMBER 39 • SEPTEMBER 26, 2014

Amino Acid-mediated Protein Lysine Acetyltransferase

FIGURE 5. Allosteric regulation of MaKat conformation by Cys and Arg. A and B, binding curve of Cys or Arg with the FRET biosensor. The emission ratio of 478/528 nm was determined at various amino acid concentrations. The data were fitted with a dose-response equation, and EC50 or IC50 were acquired using GraphPad Prism 5. C, the FRET ratios were measured in the presence of Cys (1 mM), acetyl-CoA (60 ␮M), Acs (5 ␮M), DTT (1 mM), and C2H6OS (1 mM). D, the FRET ratios were measured in the presence of Arg (1 mM), acetyl-CoA (60 ␮M), and MaAcs (5 ␮M). Basal mean was observed by adding PBS buffer.

FIGURE 6. Allosteric regulation of MaKat activity for protein acetylation by Cys and Arg. A, acetylation of MaAcs (Micau_0428) by MaKat. MaAcs (10 ␮g) was incubated alone or in the presence of MaKat (0.2 ␮M), acetyl-CoA (60 ␮M), and Cys or Arg (1 mM) in a volume of 100 ␮l at 37 °C for 60 min, followed by SDS-PAGE analysis. The level of acetylation was determined by Western blot with acetyl-lysine antibodies (upper panel). At the same time, another protein electrophoresis gel was stained with Coomassie Brilliant Blue (lower panel). B, acetylation levels of MaAcs were measured using a Western blot assay. With fixed concentrations of MaAcs (1.5 ␮M), MaKat (0.2 ␮M), and acetyl-CoA (60 ␮M), varying concentrations of Cys or Arg were added to the reaction system. The reaction lasted 60 min at 37 °C, followed by SDS-PAGE analysis.

continuously monitor the acetylation reaction. The resulting data were fitted using the Michaelis-Menten kinetics model. The results are shown in Table 1. The Km and kcat values of MaKat for Ac-CoA are 4.2 ␮M and 0.02 s⫺1, respectively. Similarly, the Km and kcat values for MaAcs are 21.6 ␮M and 0.03 s⫺1. MaKat exhibited a Km value of 4.2 ␮M for Ac-CoA substrate, which was slightly higher compared with that of yeast GCN5 HAT (Km of 2.5 ␮M) (18), although it was 5-fold lower than that of B. subtilis AcuA (Km of 22 ␮M) (19). The kcat value of MaKat for Ac-CoA was only 0.02 s⫺1, which is 85-fold lower than the kcat value (1.7 s⫺1) of yeast GCN5 HAT and 15-fold lower than the kcat value (0.3 s⫺1) of B. subtilis AcuA. As seen in Table 1, Cys or Arg was able to increase the kcat value of MaKat SEPTEMBER 26, 2014 • VOLUME 289 • NUMBER 39

for Ac-CoA and MaAcs by 2–2.5-fold and showed no significant effect on Km values. Identification of Residues in the ACT Domain That Are Associated with Cys and Arg Binding—The ACT domain is an amino acid-binding domain, which is involved in the allosteric regulation of prokaryotic amino acid metabolism. In a recent study, it was found that most of the ACT domains bind to only one amino acid. However, some ACT domains can bind to more than one amino acid. In Arabidopsis thaliana, AK-HSDH I and AK-HSDH II are inhibited by Thr (AK-HSDH II is also inhibited by Leu) and activated by five amino acids (Ala, Cys, Ile, Ser, and Val) (20 –22). AK1 is synergistically inhibited by lysine and S-adenosylmethionine (23, 24). In E. coli, formyl FH4 hydrolase JOURNAL OF BIOLOGICAL CHEMISTRY

27041

Amino Acid-mediated Protein Lysine Acetyltransferase

FIGURE 7. MaKat acetylates Lys-619 of the MaAcs enzyme. A, sequence alignment of acetyl-CoA synthetases (Acs). A conserved motif called PKTRSGK in Acs was found in MaAcs (Micau_0428). B, Western blot analysis of MaAcs acetylated by MaKat. Wild type MaAcs or MaAcsK619Q mutant was incubated with various components as indicated at 37 °C for 60 min, followed by SDS-PAGE analysis. The reaction was performed in the presence or absence of Cys (1 mM), as well as Arg (1 mM).

FIGURE 8. Initial rates of the acetyltransferase activity of MaKat. The acetyltransferase activity of MaKat was measured using a coupled enzymatic assay. In the presence or the absence of two amino acids, the initial rate of formation of NADH is shown.

TABLE 1 Kinetic analysis of MaKat on MaAcs Enzyme MaKat MaKat (Cys) MaKat (Arg)

Substrate Ac-CoA MaAcs Ac-CoA MaAcs Ac-CoA MaAcs

Km

kcat

␮M

s⫺1

4.2 ⫾ 0.3 21.6 ⫾ 0.9 4.1 ⫾ 0.7 14.8 ⫾ 1.3 4.3 ⫾ 0.6 17.3 ⫾ 1.5

0.02 ⫾ 0.003 0.03 ⫾ 0.003 0.04 ⫾ 0.01 0.05 ⫾ 0.01 0.05 ⫾ 0.01 0.05 ⫾ 0.01

kcat/Km ⫺1 ⫺1 M s

(4.7 ⫾ 0.8) ⫻ 103 (1.4 ⫾ 0.3) ⫻ 103 (9.8 ⫾ 0.1) ⫻ 103 (3.4 ⫾ 0.5) ⫻ 103 (1.2 ⫾ 0.1) ⫻ 104 (2.9 ⫾ 0.5) ⫻ 103

is positively regulated by Met and negatively regulated by Gly (25). The first ACT domain determined was E. coli D-3-phosphoglycerate dehydrogenase, which folds with a ferredoxin-like ␤␣␤␤␣␤ topology, and the most conserved portion of the ACT domain is the region at the interface between the first strand (␤1) and the first helix (␣1) (9). According to previous studies (21, 22, 26), the important residues may be found at the end of the first ␤ strand ␤1 of the conserved ␤␣␤␤␣␤ fold. Comparison of the conserved amino acid residues among the ACT domains and predication of the secondary structure (Fig. 1B) showed that MaKat also contained the conserved topology. It has been suggested that the hydrogen-bonding network may

27042 JOURNAL OF BIOLOGICAL CHEMISTRY

play an important role in the binding of specific amino acids. To investigate the important residues implicated in binding of Cys or Arg in MaKat, the structure of the ACT domain of MaKat was modeled using the ACT domain of formyltetrahydrofolate hydrolase (PURU) from Thermus thermophilus as a model (RCSB identifier rcsb095976 and Protein Data Bank code 3W7B). As shown in Fig. 9 (A and B), the residues implicated in hydrogen bonding with Cys (Asp-11, Arg-12, Gly-14, Tyr-15, Leu-16, and Asp-45) and Arg (Asp-11, Leu-16, Ala-40, and Asp-45) are mostly located in the first strand (␤1) and the first helix (␣1) of the MaKat ACT domain. Gly-14 is the most conserved residue in the loop and has an important role in maintaining the stabilization of the topology. To investigate the effect of residues on binding of amino acids, we mutated residues Asp-11, Leu-16, and Asp-45 to Ala. Circular dichroism assays showed that these mutants do not perturb the ACT structure of MaKat (data not shown). The conformational changes of the wild type and mutant proteins induced by Cys and Arg were monitored by FRET biosensor assay (Fig. 9C). Mutate aspartic acid 11 to alanine reduced the conformational change induced by Arg, whereas L16A reduced the conformational change that is induced by Arg and Cys. The VOLUME 289 • NUMBER 39 • SEPTEMBER 26, 2014

Amino Acid-mediated Protein Lysine Acetyltransferase

FIGURE 9. Amino acid binding sites of MaKat. A and B, stereo drawings of the ACT domain of MaKat show that the possible residues of MaKat interacted with Cys or Arg through hydrogen bonds. C, the changes of FRET ratios were presented after mutation in three residues (Asp-11, Ile-16, and Asp-45) emerged in the amino acid-binding region of MaKat. D–F, Western blot analysis and initial rate of NADH formation were shown for the MaKat mutants (D11A, L16A, and D45A, respectively). Insets, MaAcs was acetylated by MaKat D11A, MaKat L16A, and MaKat D45A in the presence and absence of Arg or Cys. The level of acetylation was determined by Western blot with acetyl-lysine antibodies (upper panel). The band intensities were quantified by densitometry using ImageJ software. At the same time, another protein electrophoresis gel was stained with Coomassie Brilliant Blue (lower panel).

acetylation activity of the wild type and mutant proteins in the presence of Cys or Arg was also determined using a coupled acetyltransferase enzyme assay and Western blot analysis. As shown in Fig. 9D, no increase in the rate of acetylation of MaAcs in the presence of Arg was observed in the D11A mutant protein. This is consistent with the FRET ratio that revealed that the smallest conformational change is induced by Arg. The presence of two amino acids did not result in an increase in the initial rate of the L16A mutant (Fig. 9E), which can be correlated with the lower change in FRET of that was induced by Arg and Cys. Two amino acids were able to enhance acetylation of MaAcs by the D45A mutant to a similar extent, as seen in the wild type (Fig. 9F). In agreement with this, the FRET ratios of the D45A mutant demonstrated similar conformational changes as the wild type in the presence of Cys and Arg. These results revealed that these amino acids are critical for binding in the region between ␤1 and ␣1. SEPTEMBER 26, 2014 • VOLUME 289 • NUMBER 39

DISCUSSION Growing evidence has indicated that metabolic pathways are coordinated through reversible acylation of metabolic enzymes in response to the nutritional status of cells to maintain homeostasis. Approximately 78 acetyltransferases containing putative GNAT domains are found in M. aurantiaca, but no acetyltransferases have been characterized. In this work, we report an amino acid-regulated protein acetyltransferase (MaKat, Micau_ 1670) for acetylation of acetate scavenging acetyl-CoA synthetase (MaAcs, Micau_0428) in M. aurantiaca. Our research reveals the first biochemical characterization of a protein acetyltransferase that contains a fusion of a GNAT domain with an ACT domain and provides a novel signaling pathway for regulating cellular protein acetylation in actinomycetales, in which all of the proteins containing the domain organization of ACT-GNAT have been discovered. It was previously known that the activities of protein acetyltransferases and deacetylases are carefully regulated in JOURNAL OF BIOLOGICAL CHEMISTRY

27043

Amino Acid-mediated Protein Lysine Acetyltransferase response to the change of intracellular signals (such as level of acetyl-CoA and NAD) that control the acetylation of specific proteins, which in turn affects the metabolic network. AcetylCoA and NAD are key indicators of cellular energy status and demonstrate that protein lysine acetylation serves as a link that connects cellular energy levels with protein acetylation/ deacetylation activity. In mycobacteria, cAMP directly activates MsKat and MtKat by binding of cAMP to the cyclic nucleotide-binding domain of two protein acetyltransferases, indicating that the levels of protein acetylation can also be modulated by response to intracellular cAMP levels (27). This study demonstrated that ACT domains are linked to GNAT acetyltransferases, which confirms amino acid-induced allosteric regulation of these enzymes. Protein acetyltransferases with ACT-GNAT domain organization could be a novel mechanism for connecting cellular metabolic status (amino acid metabolism) with levels of protein acetylation in actinomycetales. The allosteric stimulation of MaKat acetyltransferase activity by Cys and Arg exhibited mixed activation kinetics in which kcat values increased; Km values for MaAcs decreased. Our FRET and kinetic analyses indicate that IC50 values of MaKat for Cys and Arg are in the submicromolar ranges (Cys, 80 ␮M; Arg, 210 ␮M). This is comparable with the intracellular levels reported for Cys and Arg, which can reach cytoplasmic concentrations of 100 –200 ␮M in growing E. coli cultures (no data for M. aurantiaca) (28, 29). The normal intracellular level of two amino acids is evidently sufficient to effectively contribute to allosteric regulation of MaKat acetyltransferase. Recently, advancements in mass spectrometry and high affinity purification of acetylated peptides allow identification of thousands of lysine acetylation sites (acetylproteome) in prokaryotic and eukaryotic cells. It was found that many enzymes involved in the amino acid metabolism were lysine-acetylated in E. coli (30, 31), S. enterica (32), B. subtilis (33), Geobacillus kaustophilus (34), Thermus thermophiles (35), Saccharopolyspora erythraea (36), and Saccharomyces cerevisiae (37). On the other hand, most of the ACT domain-containing proteins appear to interact with amino acids and are involved in some aspect of regulation of amino acid metabolism in E. coli, including serine binding to 3-phosphoglycerate dehydrogenase, which catalyzes the first step in the biosynthesis of serine; phenylalanine binding to chorismate mutase, which catalyzes the first two steps in the biosynthesis of phenylalanine; valine binding to acetohydroxyacid synthase, which is involved in biosynthesis of branched chain amino acids; lysine binding to aspartokinase, which catalyzes the first step in the biosynthesis of methionine, lysine, and threonine; isoleucine and valine binding to threonine deaminase involved in biosynthesis of branched chain amino acids; threonine binding to homoserine dehydrogenase, which catalyzes the third step in the aspartate pathway; leucine binding to isopropylmalate synthase, which catalyzes the first committed step in the leucine biosynthetic pathway; and histidine binding to ATP phosphoribosyltransferase, which catalyzes the first and committed step in histidine biosynthesis (8, 9). Taken together, these data reveal that MaKat perhaps acetylates some enzymes that are involved in Cys and Arg metabolism and regulates these pathways in response to intracellular

27044 JOURNAL OF BIOLOGICAL CHEMISTRY

levels of Cys or Arg. It was previously known that Cys and Arg metabolism are regulated at the transcriptional and enzymatic levels. Arg is synthesized from glutamate in eight enzymatic steps. The expression of genes coding for all enzymes is subject to repression by Arg and mediated by the repressor ArgR. In addition, the first or secondary enzyme of the pathway, N-acetylglutamate synthase or N-acetylglutamate kinase, respectively, is subject to feedback inhibition by arginine (29). Cys is the metabolic source of sulfur for all thiol-containing compounds in cells and has a central role in sulfur metabolism for assimilation and utilization of sulfur. CysB or CymR controls the genes involved in cysteine synthesis and transport in response to N-acetylserine or O-acetylserine, precursor of cysteine in E. coli and B. subtilis. Meanwhile, the first or second enzyme of the pathway of cysteine biosynthesis from serine, serine acetyltransferase, or OAS-sulfhydrylase is inhibited by Cys. The challenge ahead is to identify all substrates of MaKat and the structure and mechanism of MaKat (38) and to elucidate the signaling link that connects intracellular levels of Cys and Arg with the regulation of metabolic enzyme activity at the post-translational modification level. Acknowledgment—We thank Sandhya S. Visweswariah for critically reading this manuscript. REFERENCES 1. Starai, V. J., Celic, I., Cole, R. N., Boeke, J. D., and Escalante-Semerena, J. C. (2002) Sir2-dependent activation of acetyl-CoA synthetase by deacetylation of active lysine. Science 298, 2390 –2392 2. Schwer, B., Bunkenborg, J., Verdin, R. O., Andersen, J. S., and Verdin, E. (2006) Reversible lysine acetylation controls the activity of the mitochondrial enzyme acetyl-CoA synthetase. Proc. Natl. Acad. Sci. U.S.A. 103, 10224 –10229 3. Hallows, W. C., Lee, S., and Denu, J. M. (2006) Sirtuins deacetylate and activate mammalian acetyl-CoA synthetases. Proc. Natl. Acad. Sci. U.S.A. 103, 10230 –10235 4. Castaño-Cerezo, S., Bernal, V., Blanco-Catalá, J., Iborra, J. L., and Cánovas, M. (2011) cAMP-CRP co-ordinates the expression of the protein acetylation pathway with central metabolism in Escherichia coli. Mol. Microbiol. 82, 1110 –1128 5. Nambi, S., Basu, N., and Visweswariah, S. S. (2010) cAMP-regulated protein lysine acetylases in mycobacteria. J. Biol. Chem. 285, 24313–24323 6. Xu, H., Hegde, S. S., and Blanchard, J. S. (2011) Reversible acetylation and inactivation of Mycobacterium tuberculosis acetyl-CoA synthetase is dependent on cAMP. Biochemistry 50, 5883–5892 7. Grundy, F. J., Turinsky, A. J., and Henkin, T. M. (1994) Catabolite regulation of Bacillus subtilis acetate and acetoin utilization genes by CcpA. J. Bacteriol. 176, 4527– 4533 8. Grant, G. A. (2006) The ACT Domain: A small molecule binding domain and its role as a common regulatory element. J. Biol. Chem. 281, 33825–33829 9. Aravind, L., and Koonin, E. V. (1999) Gleaning non-trivial structural, functional and evolutionary information about proteins by iterative database search. J. Mol. Biol. 287, 1023–1040 10. Hirsch, A. M., and Valdés, M. (2010) Micromonospora: an important microbe for biomedicine and potentially for biocontrol and biofuels. Soil Biol. Biochem. 42, 536 –542 11. Zhang, C., Wei, Z. H., and Ye, B. C. (2013) Quantitative monitoring of 2-oxoglutarate in Escherichia coli cells by a fluorescence resonance energy transfer-based biosensor. Appl. Microbiol. Biotechnol. 97, 8307– 8316 12. Berndsen, C. E., and Denu, J. M. (2005) Assays for mechanistic investigations of protein/histone acetyltransferases. Methods 36, 321–331 13. Kim, Y., Tanner, K. G., and Denu, J. M. (2000) A continuous, nonradioac-

VOLUME 289 • NUMBER 39 • SEPTEMBER 26, 2014

Amino Acid-mediated Protein Lysine Acetyltransferase tive assay for histone acetyltransferases. Anal. Biochem. 280, 308 –314 14. Liberles, J. S., Thórólfsson, M., and Martínez, A. (2005) Allosteric mechanisms in ACT domain containing enzymes involved in amino acid metabolism. Amino Acids 28, 1–12 15. Starai, V. J., and Escalante-Semerena, J. C. (2004) Identification of the protein acetyltransferase (Pat) enzyme that acetylates acetyl-CoA synthetase in Salmonella enterica. J. Mol. Biol. 340, 1005–1012 16. Crosby, H. A., Pelletier, D. A., Hurst, G. B., and Escalante-Semerena, J. C. (2012) System-wide studies of N-lysine acetylation in Rhodopseudomonas palustris reveal substrate specificity of protein acetyltransferases. J. Biol. Chem. 287, 15590 –15601 17. Tucker, A. C., and Escalante-Semerena, J. C. (2013) Acetoacetyl-CoA synthetase activity is controlled by a protein acetyltransferase with unique domain organization in Streptomyces lividans. Mol. Microbiol. 87, 152–167 18. Tanner, K. G., Langer, M. R., Kim, Y., and Denu, J. M. (2000) Kinetic mechanism of the histone acetyltransferase GCN5 from yeast. J. Biol. Chem. 275, 22048 –22055 19. Gardner, J. G., and Escalante-Semerena, J. C. (2008) Biochemical and mutational analyses of AcuA, the acetyltransferase enzyme that controls the activity of the acetyl coenzyme a synthetase (AcsA) in Bacillus subtilis. J. Bacteriol. 190, 5132–5136 20. Paris, S., Wessel, P. M., and Dumas, R. (2002) Overproduction, purification, and characterization of recombinant bifunctional threonine-sensitive aspartate kinase-homoserine dehydrogenase from Arabidopsis thaliana. Protein Expression Purif. 24, 105–110 21. Paris, S., Viemon, C., Curien, G., and Dumas, R. (2003) Mechanism of control of Arabidopsis thaliana aspartate kinase-homoserine dehydrogenase by threonine. J. Biol. Chem. 278, 5361–5366 22. Curien, G., Ravanel, S., Robert, M., and Dumas, R. (2005) Identification of six novel allosteric effectors of Arabidopsis thaliana aspartate kinase-homoserine dehydrogenase isoforms. J. Biol. Chem. 280, 41178 – 41183 23. Mas-Droux, C., Curien, G., Robert-Genthon, M., Laurencin, M., Ferrer, J.-L., and Dumas, R. (2006) A novel organization of ACT domains in allosteric enzymes revealed by the crystal structure of Arabidopsis aspartate kinase. Plant Cell 18, 1681–1692 24. Curien, G., Laurencin, M., Robert-Genthon, M., and Dumas, R. (2007) Allosteric monofunctional aspartate kinases from Arabidopsis. FEBS J. 274, 164 –176 25. Nagy, P. L., Marolewski, A., Benkovic, S. J., and Zalkin, H. (1995) Formyltetrahydrofolate hydrolase, a regulatory enzyme that functions to balance pools of tetrahydrofolate and one-carbon tetrahydrofolate adducts in Escherichia coli. J. Bacteriol. 177, 1292–1298 26. Costrejean, J. M., and Truffa-Bachi, P. (1977) Threonine-sensitive homoserine dehydrogenase and aspartokinase activities of Escherichia coli K12

SEPTEMBER 26, 2014 • VOLUME 289 • NUMBER 39

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

kinetic and spectroscopic effects upon binding of serine and threonine. J. Biol. Chem. 252, 5332–5336 Nambi, S., Gupta, K., Bhattacharyya, M., Ramakrishnan, P., Ravikumar, V., Siddiqui, N., Thomas, A. T., and Visweswariah, S. S. (2013) Cyclic AMPdependent protein lysine acylation in mycobacteria regulates fatty acid and propionate metabolism. J. Biol. Chem. 288, 14114 –14124 Park, S., and Imlay, J. A. (2003) High levels of intracellular cysteine promote oxidative DNA damage by driving the fenton reaction. J. Bacteriol. 185, 1942–1950 Caldara, M., Dupont, G., Leroy, F., Goldbeter, A., De Vuyst, L., and Cunin, R. (2008) Arginine biosynthesis in Escherichia coli: experimental perturbation and mathematical modeling. J. Biol. Chem. 283, 6347– 6358 Zhang, J., Sprung, R., Pei, J., Tan, X., Kim, S., Zhu, H., Liu, C. F., Grishin, N. V., and Zhao, Y. (2009) Lysine acetylation is a highly abundant and evolutionarily conserved modification in Escherichia coli. Mol. Cell. Proteomics 8, 215–225 Zhang, K., Zheng, S., Yang, J. S., Chen, Y., and Cheng, Z. (2013) Comprehensive profiling of protein lysine acetylation in Escherichia coli. J. Proteome Res. 12, 844 – 851 Wang, Q., Zhang, Y., Yang, C., Xiong, H., Lin, Y., Yao, J., Li, H., Xie, L., Zhao, W., Yao, Y., Ning, Z. B., Zeng, R., Xiong, Y., Guan, K. L., Zhao, S., and Zhao, G. P. (2010) Acetylation of metabolic enzymes coordinates carbon source utilization and metabolic flux. Science 327, 1004 –1007 Kim, D., Yu, B. J., Kim, J. A., Lee, Y. J., Choi, S. G., Kang, S., and Pan, J. G. (2013) The acetylproteome of Gram-positive model bacterium Bacillus subtilis. Proteomics 13, 1726 –1736 Lee, D. W., Kim, D., Lee, Y. J., Kim, J. A., Choi, J. Y., Kang, S. H., and Pan, J. G. (2013) Proteomic analysis of acetylation in thermophilic Geobacillus kaustophilus. Proteomics 13, 2278 –2282 Okanishi, H., Kim, K., Masui, R., and Kuramitsu, S. (2013) Acetylome with structure mapping reveals the significance of lysine acetylation in Thermus thermophiles. J. Proteome Res. 12, 3952–3968 You, D., Yao, L. L., Huang, D., Escalante-Semerena, J. C., and Ye, B. C. (2014) Acetyl coenzyme A synthetase is acetylated on multiple lysine residues by a protein acetyltransferase with a single Gcn5-type N-acetyltransferase (GNAT) domain in Saccharopolyspora erythraea. J. Bacteriol. 196, 3169 –3178 Henriksen, P., Wagner, S. A., Weinert, B. T., Sharma, S., Bacinskaja, G., Rehman, M., Juffer, A. H., Walther, T. C., Lisby, M., and Choudhary, C. (2012) Proteome-wide analysis of lysine acetylation suggests its broad regulatory scope in Saccharomyces cerevisiae. Mol. Cell. Proteomics 11, 1510 –1522 Friedmann, D. R., and Marmorstein, R. (2013) Structure and mechanism of non-histone protein acetyltransferase enzymes. FEBS J. 280, 5570 –5581

JOURNAL OF BIOLOGICAL CHEMISTRY

27045