Francisella tularensis 2-C-Methyl-D-Erythritol 4-Phosphate Cytidylyltransferase: Kinetic Characterization and Phosphoregulation

Deliberate and natural outbreaks of infectious disease, the prevalence of antibiotic resistant strains, and the ease by which antibiotic resistant bacteria can be intentionally engineered all underscore the necessity of effective vaccines and continued development of novel antimicrobial/antiviral therapeutics. Isoprenes, a group of molecules fundamentally involved in a variety of crucial biological functions, are derived from either the mevalonic acid (MVA) or methylerythritol phosphate (MEP) pathway. While mammals utilize the MVA pathway, many bacteria utilize the MEP pathway, highlighting the latter as an attractive target for antibiotic development. In this report we describe the cloning and characterization of Francisella tularensis MEP cytidylyltransferase, a MEP pathway enzyme and potential target for antibiotic development. Size exclusion chromatography indicates the protein exists as a dimer in solution. Enzyme assays produced an apparent , , , , and a . The enzyme exhibits a strict preference for Mg+2 as a divalent cation and CTP as the nucleotide. Titanium dioxide chromatography-tandem mass spectrometry identified Thr141 as a site of phosphorylation. T141D and T141E site-directed mutants are catalytically inactive, suggesting a mechanism for post-translational control of metabolic flux through the F. tularensis MEP pathway. Overall, our study suggests that MEP cytidylyltransferase is an excellent target for the development of novel antibiotics against F. tularensis.


Introduction
Biothreat agents are classified by the US Centers for Disease Control and Prevention (CDC) based upon their morbidity/ mortality rates, ease of dissemination, emergency response procedures, and projected social impact. Category A and B biothreat agents are of greatest concern and include Bacillus anthracis (the causative agent of anthrax), Yersinia pestis (plague), Francisella tularensis (tularemia), Brucella species (brucellosis), Burkholderia mallei (glanders), and Burkholderia pseudomallei (melioidosis). Our vulnerability to both deliberate and natural outbreaks of infectious disease is illustrated by the 2009 H1N1 swine flu pandemic, 2003 SARS outbreak, 2001 anthrax letter attacks, and 1984 Rajneeshee Salmonella attacks, highlighting the necessity of effective vaccines and antimicrobial/antiviral therapeutics. The prevalence of antibiotic resistant strains and the ease by which antibiotic resistance can be deliberately engineered into bacteria further underscores the need for continued development of novel antibiotics against new bacterial targets.
Antibiotic development generally proceeds through five distinguishable phases, including target identification, target validation, lead molecule identification, lead molecule optimization, and preclinical and clinical trials [1]. Comparative genomic investigations, utilizing bioinformatics and complete genome sequences, has greatly empowered the target identification phase of novel antimicrobial drug development [2]. Genome sequences reveal that Francisella, Brucella, Bacillus, Burkholderia, and Yersinia each harbor methylerythritol phosphate (MEP) pathway genes involved in isoprene biosynthesis. Isoprenoids are a large and structurally diverse class of molecules essential to a variety of crucial biological functions, including cell wall biosynthesis (dolichols), the regulation of membrane fluidity (hopanoids and cholesterol), electron transport (quinones), and signal transduction (prenylated proteins). Despite their diversity, all isoprenoids are synthesized from two building blocks, isopentenyl diphosphate (IPP) and dimethylallyl diphosphate (DMAPP), which are produced by the mevalonic acid (MVA) or MEP pathway (Fig. 1). Because the MVA pathway is exclusively utilized by mammals, the bacterial MEP pathway enzymes are promising targets for the development of novel antibiotics (reviewed in [3][4][5]).
Target validation typically involves loss of function assays (e.g. small molecule inhibition and/or genetic knockout) to determine the essentiality of the target, followed by protein characterization to ascertain its amenability to high-throughput screening (HTS) assays (HTS is often used in the lead identification phase, wherein large molecular libraries are screened for inhibitors of the target protein). Growth inhibition by fosmidomycin [6] (a specific inhibitor of MEP synthase) and lethal genetic knockouts of MEP synthase (FTN_1483) and MEP cytidylyltransferase (FTN_0623) [7] illustrate the vulnerability of F. tularensis to MEP pathway inhibition. We recently detailed the enzymatic characterization of F. tularensis MEP synthase and demonstrated its amenability to high-throughput inhibitor screens [6]. We describe here the cloning and enzymatic characterization of F. tularensis MEP cytidylyltransferase and report on its tractability to HTS assays. We detail the apparent Michaelis-Menten kinetic constants, survey the cation and nucleotide specificity of the enzyme, and assess its quaternary structure in solution. Additionally, we use titanium dioxide chromatography-tandem mass spectrometry to identify a site of phosphorylation on the enzyme, and suggest that phosphoinhibition may serve to regulate metabolic flux through the MEP pathway. Overall, our results validate MEP cytidylyltransferase as a promising target for the development of novel antibiotics against F. tularensis.

Genomic and Plasmid DNA isolation
Genomic DNA was isolated from F. tularensis subsp. holarctica using a Wizard Genomic DNA Purification Kit (Promega, Madison, WI), per the manufacturer's instructions. A GenElute Plasmid miniprep kit (Sigma-Aldrich, St. Louis, MO) was used to isolate plasmid DNA from E. coli.

Construction of the F. tularensis MEP cytidylyltransferase expression plasmid
The MEP cytidylyltransferase coding region (ispD, FTL_1525) was identified in the F. tularensis subsp. holarctica LVS genome (accession number NC_007880) via a BLAST search using the E. coli K12 homologous sequence as the query. Polymerase chain reaction (PCR) primer pairs, designed to flank ispD (FtIspD-f; 59-CACCGTGTCAAATAAATACGTAATTATTC -39 and FtIsp-D-r; 59-ACCTAACTGCC AATTAGCATATTC -39), were purchased from Sigma-Genosys (The Woodlands, TX) and used to amplify the gene from purified genomic DNA. The primer FtIspD-f contains four additional 59 residues (CACC) to facilitate the unidirectional insertion of the PCR product into plasmid pET101/D-TOPO (Invitrogen). FtIspD-r is designed to eliminate the stop codon in the PCR product to permit the expression of a C-terminal His-tagged MEP cytidylyltransferase protein. PCR was performed with Platinum Pfx polymerase (Invitrogen) and the following parameters: 2 minutes at 94uC followed by 30 cycles of 30 sec at 94uC, 30 sec at 53uC, 1.5 min at 68uC, and a final elongation of 10 min at 68uC. A Qiaquick PCR Purification Kit (Qiagen, Valencia, CA) was used to purify the PCR product, which was then cloned into pET101/D-TOPO to create the plasmid pFtIspD. Restriction mapping and DNA sequencing were used to confirm the fidelity of the PCR and the correct construction of the plasmid. pFtIspD was transformed into chemically competent E. coli BL21 CodonPlus (DE3)-RIL cells to express the protein.

Expression and purification of F. tularensis MEP cytidylyltransferase
Shake flask cultures (1 L) were used for protein expression (37uC, 250 rpm). A 10 mL overnight culture of E. coli BL21 CodonPlus (DE3)-RIL containing pFtIspD was used to inoculate the shake flask and upon the cell density reaching an OD 600 of 1.1, protein expression was induced by the addition of 0.5 mM isopropyl b-D-thiogalactopyranoside (IPTG). After incubation for an additional 18 hours, cells were harvested by centrifugation and stored at 280uC. To purify the His-tagged protein, the cell pellet was thawed then cells were lysed using Lysis Buffer A (100 mM Tris pH 8, .032% lysozyme; 3 mL per mg cell pellet), followed by Lysis Buffer B (0.1 M CaCl 2 , 0.1 M MgCl 2 , 0.1 M NaCl, .020% DNase; 0.3 mL per mg cell pellet). Clarified cell lysate was obtained by centrifugation (48,000 x g, 20 min) then passed through a TALON immobilized metal affinity chromatography column (Clontech Laboratories, Mountain View, CA). The column was washed with 15 column volumes of 1X equilibration buffer (50 mM HEPES pH 7.5, 300 mM NaCl), 10 column volumes of 1X wash buffer (50 mM HEPES pH 7.5, 300 mM NaCl, 10 mM imidazole) and the His-tagged protein was then eluted with 4 column volumes of 1X elution buffer (150 mM imidazole pH 7.0, 300 mM NaCl). Buffer was exchanged by addition of 0.1 M Tris pH 7.5, 1 mM NaCl while concentrating the protein by ultrafiltration. Protein concentration was determined using the Advanced Protein Assay Reagent (Cytoskeleton, Denver, CO) with c-globulins (Sigma-Aldrich) as the standard. The protein was visualized via Coomassie stained SDS-PAGE and a Western blot with an anti-His antibody (Qiagen). The yield of purified MEP cytidylyltransferase averaged 20-30 mg per 1 L LB shake flask.

Mutagenesis
The Thr 141 to Asp 141 and Thr 141 to Glu 141 mutants of MEP cytidylyltransferase were engineered by GenScript Corporation (Piscataway, NJ) using the pFtIspD plasmid as the template for mutagenesis. The sequence of both mutant constructs were confirmed by nucleotide sequencing. The mutant genes (in pET101/D-TOPO) were transformed into chemically competent E. coli BL21 CodonPlus (DE3)-RIL cells to express the C-terminal His-tagged protein. The T141D and T141E mutants were purified as described for the wildtype His-tagged MEP cytidylyltransferase.

Size-Exclusion Chromatography
The molecular mass of MEP cytidylyltransferase and the mutant derivatives were estimated by loading 1 mg of protein onto a Sephacryl 200HR (Sigma Aldrich, St. Louis, MO) sizeexclusion chromatography column equilibrated with 0.1 M Tris pH 7.5, 1 mM NaCl (flow rate of 2 mL/min) and calibrated with a gel filtration standard kit purchased from Bio-Rad (Hercules, CA). Blue dextran was used to determine the void volume of the column.

Fluorescence spectroscopy
Fluorescence spectra of wildtype and mutant MEP cytidylyltransferase were measured using a Fluoromax-3 fluorometer (Horiba Jobin Yvon) at an excitation wavelength of 290 nm using cuvettes with an optical path length of 1 cm. The emission spectra of protein samples with a concentration of 5 mM in 0.1 M Tris pH 7.5, 1 mM NaCl were measured from 310 to 400 nm (excitation and emission slit width = 5 nm). The temperature was maintained at 30uC. All fluorescence spectra were corrected for background with pure buffer.

Enzyme Assays
MEP cytidylyltransferase activity was evaluated using a coupledenzyme colorimetric assay, derived from that originally described by Bernal et al [8]. To determine the apparent K M for 2-C-methyl-D-erythritol 4-phosphate (MEP), the assay (200 mL) was performed using 2.5 mM IspD, 100 mM Tris pH 8.0, 1 mM MgCl 2 , 1 mM DTT, 0.2 mM CTP, 100 mU/mL inorganic pyrophosphatase, and varying concentrations of MEP (Echelon Biosciences, Salt Lake City, UT). The assay mixture was incubated at 37uC while 40 mL aliquots were removed at 30 sec intervals and combined with 120 mL water and 40 mL of an acidic malachite green solution (2.4 M sulfuric acid containing 1.5% ammonium molybdate, 1.6% Tween 20, and 9.5% malachite green). After a 10 minute incubation, sodium citrate was added to 3.4%. After an additional 30 min incubation the A 660 was spectrophotometrically determined. A pyrophosphate standard curve was used to calculate the rate of the enzyme. To determine the apparent K M for CTP, the assay was performed as described but with a fixed concentration of MEP (400 mM) and a variable CTP concentration. To determine cation specificity, assays were performed with 0.2 mM MEP, 0.2 mM CTP, and 1 mM MgCl 2 , CaCl 2 , CoCl 2 , CuCl 2 , MnCl 2 , or ZnCl 2 . Nonlinear regression fitting to the Michaelis-Menten equation was used to determine the kinetic constants. High-throughput assays were performed using 96-well plates with assay volumes adjusted to 50 mL. The Z-factor was calculated as described by Zhang et al [9] with calcium as the inhibitor control.

Mass spectrometry method for phosphopeptide identification
To obtain MEP cytidylyltransferase for phosphopeptide analysis, protein expression and purification was essentially as described above, with the exception of using 0.01 mM IPTG for induction. Purified MEP cytidylyltransferase was reduced with 10 mM DTT), alkylated by iodoacetamide (50 mM), and then digested by trypsin (Promega) in buffer containing ammonium bicarbonate (50 mM, pH 9) and urea (2 M). The digestion mixture was then desalted by a SepPak column (Waters, Milford, MA). Phosphopeptides were enriched from the tryptic peptides by a TiO 2 column as previously described [6]. The purified phosphopeptides were analyzed by reversed-phase liquid chromatography nanospray tandem mass spectrometry (LC-MS/MS) using an LTQ-Orbitrap mass spectrometer (ThermoFisher) using previously described methods [10]. Tandem mass spectra were searched using the program SEQUEST (Bioworks software, Thermo) with full tryptic cleavage constraints, static cysteine alkylation by iodoacetamide, and variable phosphorylation of Ser/ Thr/Tyr. Phosphopeptide identification was determined using database match scoring criteria filters followed by manual evaluation of the raw data, as described [10].

Results and Discussion
Cloning, expression, and purification of F. tularensis MEP cytidylyltransferase   Table 1 cytidylyltransferase sequence from subsp. tularensis (FTT_0711), subsp. novicida (FTN_0623), and subsp. mediasiatica (FTM_1371), respectively. PCR was used to amplify subsp. holarctica ispD and the resulting product was cloned into an expression plasmid designed to express a C-terminal His-tagged protein in E. coli. The recombinant protein was purified by immobilized metal affinity chromatography and the protein was visualized by SDS-PAGE and Western blotting using an anti-His antibody (Fig. 2). Sizeexclusion chromatography using a calibrated column indicates that the F. tularensis MEP cytidylyltransferase exists in solution as a dimer (Fig. S1), similar to the MEP cytidylyltransferase from Mycobacterium tuberculosis [13] and E. coli [14,15].

Kinetic characterization of F. tularensis MEP cytidylyltransferase
The kinetic activity of purified MEP cytidylyltransferase was spectrophotometrically evaluated by monitoring the substrate dependent enzyme catalyzed production of pyrophosphate (Fig. 1B). Nonlinear regression fitting of enzyme velocity versus substrate concentration was used to determine the apparent kinetic constants (Fig. 3 and Table 1 (Fig. 4), unlike the E. coli and Mycobacterium orthologs which also accommodate Mn +2 or Co +2 [13,14]. Additionally, the F. tularensis MEP cytidylyltransferase displays a clear preference for CTP over alternative nucleotides (Fig. 5).

Phosphorylation of F. tularensis MEP cytidylyltransferase
Multifarious control mechanisms converge to collectively regulate the activity of HMG-CoA reductase, thereby controlling metabolic flux through the MVA pathway (reviewed in [16]). In contrast, engineered expression studies suggest that several pathway enzymes share control of regulating flux through the MEP pathway (reviewed in [17]). We recently described the phosphorylation of F. tularensis MEP synthase, illustrating its potential for metabolic flux control and highlighting a region of the protein that may be amenable to rational drug design [6]. To ascertain if the F. tularensis MEP cytidylyltransferase might also be phosphoregulated, we expressed the recombinant enzyme in E. coli (induced with 10 mM IPTG), purified the His-tagged protein, subjected the protein to trypsinization, and isolated and identified phosphopeptides via titanium dioxide chromatography-tandem mass spectrometry. Phosphothreonine 141 (equivalent to threonine 140 in the E. coli homolog) was identified (Fig. 6a). The resolved crystal structure of E. coli MEP cytidylyltransferase in complex with CDP-ME (Fig. 6b) shows Thr140 positioned in the substrate binding site, with the sidechain hydroxyl and backbone amide groups hydrogen bonded to the C3 and C1 hydroxyl oxygen atoms of ME, respectively [15,18]. Threonine 141 in the F. tularensis homolog presumably has a similar role.
In light of the above observations, we hypothesized that phosphorylation of Thr141 would disrupt the enzyme-substrate interaction, negatively affecting enzyme activity. To test this hypothesis, two mutants of F. tularensis MEP cytidylyltransferase were created, T141D and T141E, wherein Thr141 was changed to an aspartate or glutamate, respectively, which serve to mimic a phosphothreonine. Each mutant was expressed in E. coli and purified to near homogeneity via a C-terminal His-tag. Relative to the wildtype enzyme, little difference was observed in the intrinsic fluorescence maximum (Em l max ) for either of the mutants (Fig. 7a), indicating that the introduction of the Asp or Glu has little structural influence on the protein globular fold. Sizeexclusion chromatography indicates that both mutants exist as dimers in solution, as was observed with the wildtype isoform (Fig.  S1). Enzyme assays using the purified proteins reveal that 75% of wildtype activity is lost with the T141D mutation, whereas the T141E mutant is completely inactive (Fig. 7b). Collectively, these results suggest that phosphorylation of Thr141 in F. tularensis MEP cytidylyltransferase inhibits the enzyme, possibly serving as a control mechanism for regulating metabolic flux through the MEP pathway (analogous to the phosphoregulation of HMG-CoA reductase [19], a kinase/phosphatase cycle may control metabolic flux by phosphoregulating the activity of MEP synthase and MEP cytidylyltransferase). Additionally, these results highlight Thr141 for the rational design of small molecule inhibitors of MEP cytidylyltransferase.

F. tularensis MEP cytidylyltransferase and drug discovery
Lead molecule identification by high-throughput inhibition assays typically involves the screening of large molecular libraries using multi-well plate assays, a process that is highly dependent on the quality and robustness of the assay, often described in terms of the Z-factor [9]. An excellent high-throughput assay has a Z-factor score between 0.5 and 1.0. To determine the Z-factor for the F. tularensis MEP cytidylyltransferase assay, we adjusted the assay volume to accommodate a 96-well plate, fixed the substrate concentration at the K m (178 mM), used a saturating concentration of CTP (200 mM), and evaluated three separate lots of purified MEP cytidylyltransferase in a series of assays performed over three consecutive days. Calcium, an inhibitor of pyrophosphatase, was used as a positive control for inhibition. The Z-factor with F. tularensis MEP cytidylyltransferase was found to be 0.8, indicative of an assay (and enzyme) well suited for use in a high-throughput screen.

Conclusions
The target validation phase of novel antibiotic drug development typically involves genetic knockout assays to assess lethality as well as protein purification and characterization to evaluate amenability to HTS and/or rational drug design. While transposon mutagenesis demonstrated the essentiality of MEP cytidylyltransferase to F. tularensis viability, the protein remained to be characterized. We demonstrate here that purified recombinant F. tularensis MEP cytidylyltransferase is kinetically active and highly amenable to use in HTS. Additionally, our investigation is the first to show that MEP cytidylyltransferase may be phosphoregulated at Thr141, further implicating a phosphoregulatory mechanism of controlling metabolic flux through the F. tularensis MEP pathway, as well as highlighting the residue for the rational design of enzyme inhibitors. Figure 6. Structural features of the MEP cytidylyltransferase. A) Predicted structure of the F. tularensis MEP cytidylyltransferase, homologymodeled using templates selected by I-TASSER's threading alignment algorithm. A cartoon representation of the dimeric enzyme is shown, with alpha helices colored red, beta sheets colored yellow, and coiled regions colored gray. ProQRes was used to evaluate the quality of the model (Fig.  S2), providing scores ranging from 0 (unreliable) to 1 (reliable). Regions of the model scoring ,0.5 are colored light blue. Residues comprising the substrate binding site (colored dark blue with backbone and sidechain residues shown) were identified via primary sequence alignment and the resolved structure of E. coli MEP cytidylyltransferase [15]. Threonine141 (colored green with backbone and sidechain residues shown) is equivalent to E. coli Thr140. B) Active site of the E. coli MEP cytidylyltransferase co-crystallized with CDP-ME (PDB accession code 1INI). Thr140 (arrow) hydrogen bonds to the C1 and C3 hydroxyl oxygen atoms of CDP-ME [15,18]. doi:10.1371/journal.pone.0020884.g006

Supporting Information
Figure S1 Molecular weight determination by sizeexclusion chromatography. Protein standards (&) were used to calibrate the column. Linear regression fitting (R 2 is indicated) generated the standard curve, which was used to determine the molecular weight of wildtype MEP cytidylyltransferase (IspD) and the two mutants (T141D and T141E). (TIF) Figure S2 ProQRes evaluation of the F. tularensis MEP cytidylyltransferase structural model generated by I-TASSER. ProQRes uses atom-atom contacts, residue-residue contacts, solvent accessibility, and secondary structure information to score the model over a sliding window of 9 residues [12]. Scores range from 0 (unreliable) to 1 (reliable). (TIF) Figure 7. Regulation of F. tularensis MEP cytidylyltransferase. A) Intrinsic fluorescence spectra of MEP cytidylyltransferase and two mutant derivatives. Wildtype and mutant (T141D and T141E) proteins were adjusted to 5 mM in 0.1 M Tris pH 7.5, 1 mM NaCl and analyzed using an excitation wavelength of 290 nm. The emission spectra was measured from 310 to 400 nm. The Em l max of all three proteins was detected at 320 nm. Identical spectra indicate little structural change in the protein globular fold accompanies the introduction of either Asp or Glu. B) The relative activity of wildtype MEP cytidylyltransferase and the T141D and T141E mutants. Assays were performed in duplicate with 200 mM MEP and 200 mM CTP. doi:10.1371/journal.pone.0020884.g007