A novel non-radioactive primase–pyrophosphatase activity assay and its application to the discovery of inhibitors of Mycobacterium tuberculosis primase DnaG

Cloning and purification of Mtb DnaG

The primase gene dnaG (locus tag: Rv2343c) was amplified by polymerase chain reaction from Mtb H37Rv genomic DNA (BEI Resources, NIAID, NR-14865) by using primers (5′-AGTTAGCACATATGTCCGGCCGGATCTCCG-3′) and (5′-CCGCTCGAGTCACGCGGTGAGATCG-3′) and cloned between NdeI and XhoI sites of a modified pET19b vector (35), encoding an N-terminal decahistidine tag separated from the recombinant protein by a PreScission protease (GE Healthcare, Piscataway, NJ, USA) cleavage site. The construct expressing Mtb DnaG E268Q was generated by mutagenesis of the aforementioned construct with a QuikChange Kit (Qiagen, Valencia, CA, USA) by using primers 5′-CATCAGGCCGTCGTCGTCCAGGGCTACACCGATGTCATG-3′ and 5′-CA TGACATCGGTGTAGCCCTGGACGACGACGGCCTGATG-3′. The wild-type and the mutant proteins were expressed and purified analogously. Protein expression was carried out in BL21 (DE3) cells cultured in LB broth. The culture was induced with 0.5 mM of IPTG at an attenuance of 0.2 and then incubated for 16 h at 18°C. All purification steps were carried out at 4°C. The cell pellet from a 4 l culture was suspended in 100 ml of buffer A [40 mM Tris–HCl pH 8.0, 600 mM of NaCl, 10% of glycerol, 1 mM of PMSF, 0.5 mM of adenosine triphosphate (ATP), 2 mM of MgCl₂ and 2 mM of β-mercaptoethanol] containing two tablets of complete ethylenediaminetetraacetic acid (EDTA)-free protease inhibitor cocktail (Roche Applied Science, Indianapolis, IN, USA). The cells were disrupted by sonication on ice and clarified by centrifugation at 38 000g for 60 min. The supernatant was filtered through a 0.45 µm Millex-HV PVDF filter (Millipore, Billerica, MA, USA) and applied to a 5 ml Ni-IMAC HisTrap FF column (GE Healthcare) equilibrated with buffer A. The column was washed with 120 ml of buffer A containing 50 mM imidazole, and then protein was eluted with 11 ml of buffer A containing 500 mM imidazole. The eluate was adjusted to NaCl of 200 mM, with buffer A prepared without NaCl and digested overnight at 4°C with PreScission protease. The digested protein was purified on a size-exclusion S-200 column (GE Healthcare) equilibrated in buffer B (40 mM Tris–HCl pH 8.0, 600 mM of NaCl, 10% of glycerol, 0.25 mM of ATP, 2 mM of MgCl₂ and 2 mM of β-mercaptoethanol), and the protein-containing fractions were pooled and concentrated using an Amicon Ultra-15 centrifugal filter device (Millipore) to 5 mg/ml. The enzyme was stored on ice for all assays. The recombinant untagged DnaG was highly homogeneous (Coomassie blue staining indicated >95% purity) and remained stable and active, while on ice, for 2 weeks. One hour before the assays, the storage buffer was exchanged by MicroSpin G-25 columns (GE Healthcare) into buffer C (40 mM of HEPES–NaOH pH 7.5, 600 mM of NaCl, 10% of glycerol, 2 mM of MgCl₂, 2 mM of β-mercaptoethanol and 0.1 of mM ATP). The presence of ATP in the protein buffer results in 10 µM ATP in the reaction mixture. This low concentration of ATP is negligible, as it is below K_m for ATPases and other ATP-interacting enzymes, including Mtb DnaG. Partial degradation of such a small amount of ATP, even if it occurred, would be below or at the detection limit of our assay.

Purification of Mtb PPiase

The pJ411 vector (DNA2.0, Menlo Park, CA, USA) encoding Mtb PPiase bearing an N-terminal hexahistidine-tag was a generous gift from Dr Luiz Pedro Carvalho. Mtb PPiase was expressed similarly to Mtb DnaG. All purification steps were carried out at 4°C. The cell pellet was resuspended in 20 mM of TEA buffer pH 7.8. The cells were disrupted sonication and clarified by centrifugation at 25 000g for 30 min. The supernatant was applied to a 5 ml Ni-IMAC HisTrap FF column (GE Healthcare) equilibrated with buffer A (20 mM TEA pH 7.8, 50 mM of imidazole and 300 mM of NaCl). The column was washed with 220 ml of buffer A, and protein was eluted with a linear gradient of 50–500 mM of imidazole in buffer A (570 ml). The fractions containing Mtb PPiase were pooled, concentrated and passed through a size-exclusion S-200 column (GE Healthcare) equilibrated in 40 mM of Tris–HCl pH 8.0, 100 mM of NaCl. The fractions containing protein were pooled and concentrated using an Amicon Ultra-15 centrifugal filter device (Millipore) to 750 µM. The recombinant His-tagged Mtb PPiase was highly homogeneous (Coomassie blue staining indicated >98% purity) and stable at 4°C for 6 months.

Coupled colorimetric primase–pyrophosphatase assay (optimization of conditions)

In all coupled primase–pyrophosphatase assays that were carried out during the optimization phase to determine DNA template (sequence and length), divalent metal type and concentration, buffer type and pH and in time course experiments, we mixed DNA (Integrated DNA Technologies, Coralville, IA, USA), buffer, NTP (Promega, Madison, WI, USA) and pyrophosphatase on ice and then added Mtb DnaG, so that the reagent mixture is at twice the intended concentration of all reagents. The mixture was then diluted two-fold with water to give the final concentration of Mtb DnaG (0.7 µM or as specified) DNA (1.25 µM or as specified), NTP (110 µM or as specified), NaCl (50 mM), KGlu (150 mM), buffer (20 mM of CAPS pH 8.8 or as specified) and divalent metals (2 mM of Mn²⁺ and 1 mM of Mg²⁺ or as specified). Mtb DnaG alone is prone to severe aggregation and rapidly becomes inactive at salt concentrations <275 mM. However, in the presence of DNA and nucleotides, Mtb DnaG maintains its solubility and activity. The reaction mixture was incubated at room temperature (22°C) for 30 min. In the assay optimization experiments, inorganic pyrophosphatase from Saccharomyces cerevisiae (Sigma-Aldrich, St. Louis, MO, USA) was used in a three-fold excess of the minimal concentration needed for complete PP_i cleavage (as determined experimentally for each enzyme stock). The yeast enzyme was replaced by Mtb PPiase in the HTS and in the mode of inhibition studies. The P_i formed as a consequence of breakdown of the released PP_i was quantitated using procedures adapted from earlier studies (36). In brief, 150 µl of the malachite green reagent (MGR; a 2:1:1:2 mixture of 0.0812% malachite green, 2.32% w/v polyvinyl alcohol, 5.72% in 6 M of HCl ammonium molybdate and water) was added to 30 µl reactions and followed by addition of 15 µl of 34% sodium citrate after 1 min. After allowing colour to develop for 30 min, absorbance at 620 nm was measured with a Biotek Synergy H1 plate reader (Biotek, Winooski, VT, USA). In all coupled primase–pyrophosphatase assays, we calibrated the absorbance at 620 nm to the concentration of P_i (twice that of PP_i) by using sodium phosphate and verified the linearity of the signal in the experimental range (the signal was linear up to Abs = 1.5). For the assay optimization and inhibitor characterization, we used flat-bottom 96-well clear polystyrene plates (Fisher Scientific, Pittsburgh, PA, USA).

Primer synthesis assay

Primer synthesis reactions to observe the radiolabelled RNA products were carried out under conditions similar to those determined in the course of the optimization of conditions by the coupled primase–pyrophosphatase assay. More specifically, the reactions were carried out in buffer containing DNA (1 µM; the sequences are specified in the text), NTP (100 µM) in 20 mM of CAPS pH 8.8, 150 mM of KGlu containing divalents (Mn²⁺ or Mg²⁺), DTT (1 mM) and 110 nM of α–³²P-GTP. The products were extracted with phenol and run in a 12% urea–polyacrylamide gel. The gels were exposed to phosphorimager plates and read in a Typhoon phosphorimager system (GE Healthcare).

High-throughput screening assay

For its performance in HTS, we adapted the assay so that it can sustain the manipulations at room temperature for 4–6 h without any significant deleterious effects on the protein or other reagents. Fifty nanomolars of Mtb PPiase, instead of yeast PPiase, was used in the HTS and inhibitor characterization. To minimize the number of injections and error in dispensing reagents using liquid handling robotics, we revised the assay protocols. Small molecule compounds were dispensed into 384-well flat bottom, non-binding surface, polystyrene plates (Corning Incorporated, Corning, NY, USA) from the compound library plates with a Biomek FX robot (Beckman Coulter, Brea, CA, USA) with 200 nl of inhibitors (to obtain the final concentration in the reaction of 13 µM). Twenty microlitres of the NTP mix (with 2 mM of β-mercaptoethanol) in water was then added to all wells. EDTA at the final concentration of 30 mM was added to the last two columns of wells as a positive inhibition control. Finally, a 10 µl mixture of DNA, Mtb PPiase and Mtb DnaG in the appropriate buffer (at three-fold higher concentrations of all reaction and buffer components than their final concentrations) was added. After 30 min incubation at room temperature, 30 µl of the MGR was added followed by 10 µl of 10% sodium citrate, and plates were shaken briefly. Absorbance at 620 nm was read in 30 min with a PHERAstar plate reader (BMG Labtech, Cary, NC, USA). This assay was used to screen a library of 2556 small molecules comprising 2000 food and drug administration-approved drugs (Spectrum Chemicals, New Brunswick, NJ, USA), 450 molecules that have been used in human therapy (NIH Clinical Collection) and 106 kinase inhibitors (Enzo Life Sciences, Farmingdale, NY, USA).

The Z′ score was calculated for the pilot 384-well assay where DMSO was delivered without the small molecules, by using the following equation (37):

where Av(pos) and SD(pos) are the signal (absorbance) and its standard deviation for the positive controls (EDTA-containing reactions), and Av(neg) and SD(neg) are the respective values for negative controls (DMSO-containing reactions).

Dose–response assays

For the mode of inhibition studies, the reactions were carried out under conditions similar to those used in the HTS, except the volumes of the colour reagents were slightly modified to adapt it to the 96-well plate and to further increase the stability and consistency of the colour formation. In these assays, 90 µl of MGR was added to a 30 µl reaction mixture followed by 30 µl of 10% sodium citrate.

Reactions were carried out with [Mtb DnaG] = 0.5 µM, [DNA] = 1.25 µM, [NTP] = 110 µM at different concentrations of inhibitors, as specified in the text. Relative activities were calculated by dividing the concentrations of PP_i released in the priming reactions in the presence of inhibitors by that released in its absence. The following dose–response equation was used to fit these data with SigmaPlot 9.0 (SysStat Software, San Jose, CA, USA), with IC₅₀ and the Hill coefficient n as the fitting parameters:

Analysis of mode of inhibition assays

The assays were performed with Mtb DnaG (0.5 µM) and Mtb PPiase (50 nM) in the same reaction buffer as that used in the HTS assays. The steady-state rate of PP_i release was calculated based on measurements of the concentration of released PP_i in 20-min reactions, as described earlier in the text. Two sets of reactions were run: (i) at constant [DNA] = 1.25 µM at different concentrations of NTP (31.25, 62.5, 125, 250 and 500 µM) and (ii) at constant [NTP] = 110 µM at different concentrations of DNA (0.19, 0.38, 0.75 and 1.5 µM). We observed inhibition of P_i generation by NTP and DNA at higher NTP and DNA concentrations beyond the aforementioned ranges; therefore, they were not used in this analysis. For each pair of DNA and NTP concentrations, different concentrations of the inhibitors (suramin and doxorubicin) were used (0, 1.25, 2.5, 5, 10, 20 and 40 µM). As a simplifying approximation, to analyse the data, we used an ordered sequential rapid equilibrium mechanism of RNA synthesis by DnaG, in which DNA binds free DnaG first, followed by binding of the first NTP and subsequent binding and incorporation of an average of m nucleotides (releasing m PPi molecules) and dissociation of DnaG from DNA:

Here, K_DNA and K_NTP are equilibrium constants for binding primed DNA and NTP, respectively, both defined for the dissociation direction. This mechanism assumes that the concentration of NTP affects only initiation of primer synthesis. If the first NTP incorporation is slow (primer nucleation) relative to the m-1 subsequent incorporation steps (elongation), then k_cat is the catalytic rate constant of the slow nucleation step.

Because the data exhibit competitive inhibition with respect to both DNA and NTP and the inhibitor:target stoichiometry higher than 1:1 (model-independent observations), we considered the simplest competitive inhibition mechanism, in which an inhibitor (I) can bind non-cooperatively to n binding sites on Mtb DnaG:

We designate as K_i an intrinsic equilibrium constant for binding of an inhibitor molecule to a site on DnaG, defined for the dissociation direction. This mechanism yields the following equation for the steady-state rate of PP_i release (V) in large excess of NTP and inhibitor, but not DNA, over Mtb DnaG:

(1)

Because one nucleation event leads to release of m PPi molecules, k_cat,app = mk_cat. The concentrations of free Mtb DnaG, [DnaG] and all DNA species not bound to DnaG, [DNA], are obtained from the conservation of material equations for DnaG and DNA and from the aforementioned equilibria as:

(2)

where

(3)

(4)

(5)

(6)

Here, [DnaG]_T and [DNA]_T are the total concentrations of Mtb DnaG and DNA in the reaction, respectively. This rate law was globally fit to all the data collected for each inhibitor by using non-linear regression with SigmaPlot 9.0, with k_cat,app, K_DNA, K_NTP and K_i as fitting parameters for different fixed integer values of n. A satisfactory fit was not obtained for n = 1 or n = 2 for either suramin or doxorubicin, consistent with the model-independent dose–response analysis. For suramin, the data were well fit for n = 4, whereas for doxorubicin, an equally good fit was produced for n = 3, yielding the best fit parameter values and the theoretical curves generated by using these as presented in the ‘Results’ section. The data could be fit equally well by a more complex model, in which the inhibitors bind Mtb DnaG with high cooperativity, with n = 3 and n = 2 for suramin and doxorubicin, respectively. However, because the cooperative binding model contains too many fitting parameters to be resolved by the data, and because it lacks support by other experimental evidence, it is not presented here in further detail.

DNA-binding assay

DNA affinity of Mtb DnaG for ssDNA was measured at the conditions of the priming activity assays, either in the absence of NTPs or in the presence of all four NTPs, at [NTP] = 110 µM. We used a single-stranded 28-mer ssDNA: 5′-CCGAATCAGTCCGACGACGCATCAGCAC-3′, labelled at the 5′-end with 6-carboxyfluorescein (6FAM; Integrated DNA Technologies, Coralville, IA, USA). Titrations were set-up at the constant concentration of ssDNA of 20 nM at different concentrations of Mtb DnaG in separate 50 µl reaction mixtures and were equilibrated for 20 min at 22°C. Fluorescence anisotropy was monitored by a SpectraMax M5 plate reader (Molecular Devices, Sunnyvale, CA, USA), set at the excitation and emission wavelengths of 495 and 520 nm, respectively. To obtain equilibrium protein-DNA binding constant K_d, non-linear least-squares regression data fitting to the equilibrium 1:1 binding isotherm was performed with SigmaPlot 9.0 as previously described (38). No significant difference was observed between titrations in the presence or absence of NTPs; therefore, these data were averaged together.

Purification of Mtb DnaG

Expression and purification of Mtb DnaG have not been previously reported. Initially, we tested a variety of expression and purification protocols previously reported for other DnaG proteins; these efforts yielded inactive aggregated protein. During the extensive optimization of the purification procedure, we found that ATP (0.25 mM), NaCl (600 mM) and glycerol (∼10%) all need to be present during all purification steps to prevent Mtb DnaG from aggregation. As a result, we obtained pure and active Mtb DnaG. The purified protein (Figure 1A) is homogeneous and monomeric as judged by S-200 size-exclusion chromatography (Figure 1B).

Development of a novel non-radioactive coupled primase–phosphatase assay

Primase activity has been historically monitored by visualization and quantification of radiolabelled RNA primers on a denaturing gel. We sought to develop a novel activity assay that would be robust (even at room temperature), would not require the use of radioactivity and would not require accessory factors, such as ssDNA-binding protein or DnaB helicase, which have been used previously to enhance primase activity. Such an assay would be highly suitable for high-throughput applications.

We developed a colorimetric coupled primase–pyrophosphatase assay satisfying these requirements (Figure 2). Inorganic pyrophosphatase (PPiase, type I) is used to cleave the pyrophosphate (PP_i) liberated in the priming reaction into phosphate (P_i). To measure the P_i concentration, we use the MGR (36), whose absorbance at 620 nm increases with increasing concentration of P_i, changing its colour from yellow to green. PPiase cleaves exclusively PP_i and does not exhibit any cleavage activity on NTP (29), a property that we also confirmed for yeast and Mtb PPiases. The concentration of PPiase was chosen to be three-fold higher than the minimum concentration required for cleavage of all released PP_i.

By monitoring synthesis of radiolabelled primers, we directly established that at the minimal conditions of our novel coupled assay, Mtb DnaG displays robust Mn²⁺-dependent primer synthesis activity and has no detectable activity in Mg²⁺ (Figure 3). This activity is not highly processive, as evidenced by several short (<10 nt) products of different lengths, which is also observed with other primases (21). Because Mtb DnaG is an intrinsically inefficient polymerase, in developing the new assay, it was important to minimize any unwanted phosphate-borne signal from spurious ATPase activity that may be present in the protein preparations. Therefore, to avoid possible interference from ATPases, we designed a new DNA template (Figure 4, top inset). This template does not contain thymidine nucleotides in its 5′-half, so that ATP is not needed for the synthesis of complementary RNA; as a result, we could omit ATP in the reaction. Indeed, we obtained a robust DNA-dependent signal from the primer synthesis activity by Mtb DnaG in this reaction set-up with only CTP, GTP and UTP (Figure 4). An Mtb DnaG point mutant E268Q of a conserved divalent metal-coordinating residue (39) that is absolutely required for catalysis (21,40) was used as a control to ensure that the observed signal resulted specifically from the primer synthesis activity of Mtb DnaG. No signal was observed without the inorganic pyrophosphatase. We then performed the assay with a variety of divalent metal salts (Figure 5). The results confirmed the lack of detectable activity with Mg²⁺ and demonstrated that Mn²⁺ was the only divalent metal that supported robust activity among the seven metals tested. Mn²⁺ was previously reported to increase DnaG activity without significantly affecting other properties of DnaG (21). In a recently determined crystal structure of the catalytic core of Staphylococcus aureus DnaG in complex with nucleotides, a catalytic Mn²⁺ ion was shown to be interchangeable with Mg²⁺ (39).

Optimization of assay conditions

The colorimetric primase–pyrophosphatase assay provides a quantitative measure of kinetics of NTP incorporation into growing RNA by measuring PP_i release (Figure 6A). At these conditions ([Mtb DnaG] = 0.7 µM, [DNA] = 1.25 µM, [NTP (CTP, GTP, UTP)] = 110 µM each), the linear steady-state accumulation of PP_i persists until ∼30 min from the start of the reaction, after which the signal starts to reach the plateau, likely because of accumulation of stable RNA–DNA duplexes or because of protein aggregation. Therefore, to obtain a robust signal in the steady-state measurements, a reaction time of 20–30 min (as specified) was chosen for further measurements. To find conditions that maximize the primer synthesis activity in our assay, we tested a variety of solution parameters. When measured as a function of the concentration of the divalent metal of choice, Mn²⁺, Mtb DnaG exhibited a maximum activity near 2 mM of Mn²⁺ (Figure 6B). Mg²⁺ (1 mM) is used in the reaction to ensure robust activity of PPiase, which is not highly active in Mn²⁺ alone (34). Potassium glutamate (KGlu) was chosen for the assay, as it is the major electrolyte in vivo (41), and it significantly enhances protein–DNA interactions in vitro (42). The activity was optimal in the physiological KGlu concentration range, 150–200 mM (Figure 6C). The primase was active in a variety of buffers and with pH values in the range 6.0–9.0 (Figure 6D); 20 mM of CAPS pH 8.8 was chosen, as it supported the highest activity of Mtb DnaG among the tested buffers and pHs.

The observed steady-state rate of PP_i release measured as a function of DNA concentration at [NTP] = 110 µM displayed a saturating behaviour indicative of tight binding of Mtb DnaG to DNA (a nearly linear increase of activity until half-saturation), with K_m,obs < 0.3 µM and V_max,obs = 1.16 ± 0.06 µM/min (Figure 6E); the subscript ‘obs’ is used here to reflect the dependence of these parameters on the concentration of NTP. Only an upper bound on K_m,obs (<0.3 µM) can be determined from this titration, because [Mtb DnaG] > K_m. Based on this dependence, we used [DNA] = 1.25 µM for HTS and inhibitor characterization, to maximize the signal and to ensure the sensitivity of the assay to potential inhibition. We observed a hyperbolic (Michaelis–Menten) saturation behaviour of the rate of PP_i release as a function of NTP concentration (at [DNA] = 1.25 µM; Figure 6F), yielding K_m,obs = 177 ± 7 µM and V_max,obs = 2.85 ± 0.04 µM/min. Note that the V_max,obs value obtained in this assay is higher than that in the previous assay because [DNA] = 1.25 µM is saturating, whereas [NTP] = 110 µM is not. At [NTP] near 100 µM, a robust signal was achieved. Because less than a two-fold increase in signal required using five-fold higher concentrations of expensive high-purity NTPs in material-consuming HTS, we chose [NTP] = 110 µM for HTS.

High-throughput screening of 2560 compounds to discover inhibitors of Mtb DnaG and Mtb PPiase by the primase–pyrophosphatase assay

We adapted the assay to the 384-well format for its use in HTS of libraries of small molecules. To evaluate the suitability of the assay for HTS, we performed a control assay in a 384-well plate where DMSO solution (without small molecule compounds) was delivered into the reactions in all but 32 wells, and EDTA (at final concentration of 30 mM) was added in the remaining 32 wells as a positive inhibition control (Figure 7A). The inhibition by EDTA (manifested by the absence of the green colour of the MGR) was clearly distinguishable by the naked eye. The absorbance measurements in this control assay yielded a Z′ score of 0.7, indicating that the assay is robust and suitable for HTS (37).

By using this assay, we carried out HTS of a library of 2556 diverse molecules, which comprised known drugs used in human therapy, natural products with undetermined biological activities, compounds in agricultural use and kinase inhibitors. The assay yielded nine different hits that displayed inhibition levels larger than three standard deviations (>3σ) of the negative controls on the same plate (Figure 7B). Three compounds (disopyramide phosphate, oleandomycin phosphate and fosfosal) displayed increase in absorbance. This was not because of enzyme activation, but because the first two compounds were phosphate salts and because phosphate was released on rapid hydrolysis of fosfosal at the acidic pH of the MGR (43). We performed dose–response measurements with three strong hits, suramin (7.2σ), doxorubicin (3.7σ) and ellagic acid (5.1σ) (Figure 8). These data yielded IC₅₀ values of 6.3 ± 0.2 µM and 7.7 ± 0.5 µM for suramin and doxorubicin, respectively, indicating their potent inhibitory effects. Ellagic acid displayed approximately three-fold weaker inhibition (IC₅₀ = 22 ± 3 µM) and was not characterized further. None of the three compounds inhibited Mtb PPiase as assayed separately with inorganic pyrophosphate as a substrate, indicating that these compounds are bona fide Mtb DnaG inhibitors (the other six hits remain to be tested for DnaG and PPiase inhibition). The dose–response curves for suramin and doxorubicin could be well modelled only by using the Hill coefficients of 2.9 ± 0.1 and 2.0 ± 0.2, respectively (Figure 8), which indicates that the target contains multiple inhibitor-binding sites (44).

Mode of inhibition of Mtb DnaG by suramin and doxorubicin

We first confirmed directly that suramin and doxorubicin are potent Mtb DnaG inhibitors by observing their effect on the synthesis of radioactively labelled RNA products on the same DNA template at conditions similar to those of the colorimetric assay (Figure 9) and in assays with a different template DNA and with all four NTPs present in the reaction (Figure 3). To investigate the inhibitory mechanisms of suramin and doxorubicin in more detail, we measured the effect of these compounds on the steady-state accumulation of PP_i in the primer synthesis reaction. To determine whether suramin and doxorubicin interfere with interactions of Mtb DnaG with DNA, NTP or both, we performed measurements of the steady-state rate of PP_i release as a function of inhibitor concentration for different concentrations of DNA and NTP. One set of measurements was performed at a fixed concentration of DNA ([DNA] = 1.25 µM) with different concentrations of NTP (Figure 10A and C) and the other set at a fixed concentration of NTP ([NTP] = 110 µM) with different concentrations of DNA (Figure 10B and D). The common y-axis intercepts in the Lineweaver–Burk plots of these data (insets in Figure 10) show that both suramin (Figure 10A and B) and doxorubicin (Figure 10B and C) are competitive inhibitors with respect to both NTP and DNA. These data effectively contain multiple dose–response curves, which, similarly to the dose–response curves presented in the previous section, indicate the presence of multiple binding sites for both suramin and doxorubicin. To obtain the equilibrium binding constant K_i and the number of sites on Mtb DnaG for each inhibitor, we performed a global non-linear regression analysis of these data. We used a simple assumption of multiple non-cooperative inhibitor binding sites. In addition, we used a simple approximation of an ordered sequential mechanism of priming, where an ssDNA molecule binds free Mtb DnaG first followed by binding of an NTP, slow primer nucleation and its rapid elongation, incorporating m NTP molecules per DNA-binding event. This minimal kinetic mechanism is consistent with the observed first-order kinetics of the steady-state activity of Mtb DnaG with respect to NTP concentration (Figure 6F) and with other data (Figure 10); the ordered binding of DNA and NTP was previously proposed for a eukaryotic primase (45). This analysis yields the values of the mechanistic parameters for Mtb DnaG: the catalytic rate constant k_cat,app = 5.7 ± 0.2 min⁻¹, the effective equilibrium constants for binding of ssDNA to Mtb DnaG K_DNA = 24 ± 10 nM and for binding of an NTP to the primed DNA-Mtb DnaG complex K_NTP = 170 ± 10 µM. To obtain an independent measurement of Mtb DnaG affinity for ssDNA, we performed a fluorimetric assay, in which increase in fluorescence polarization of a carboxyfluorescein label on ssDNA was monitored as a function of concentration of Mtb DnaG (Supplementary Figure S1). This assay yields equilibrium binding constant K_d = 36 ± 16 nM, in good agreement with the K_DNA value obtained from the kinetic assays.

The aforementioned value of k_cat,app is equal to the intrinsic k_cat only if the primase is completely non-processive, dissociating from DNA after incorporation of one NTP, that is, when m = 1. Because primases are at least somewhat processive (46,47), this value overestimates k_cat. In the other extreme, if Mtb DnaG rapidly incorporates on average 30 nt per DNA binding event, we obtain k_cat = k_cat,app/30 = 0.19 ± 0.01 min⁻¹, an underestimate of k_cat. The processivity of Mtb DnaG and the kinetics of individual NTP incorporation steps will be investigated in detail by future studies. For suramin inhibition data, an excellent fit was obtained for a minimum of n = 4 binding sites on Mtb DnaG (binding to one of which is sufficient for inhibition), with an intrinsic equilibrium binding constant for each site of K_i = 3.2 ± 0.5 µM (Figure 10A and B). The analogous analysis of doxorubicin inhibition yielded best-fit values of n = 3 and K_i = 2.5 ± 0.5 µM and also an excellent agreement with the model mechanism (Figure 10C and D). The data can also be fit well to a more complex model, where three molecules of suramin or two molecules of doxorubicin bind Mtb DnaG with high cooperativity (see ‘Materials and Methods’ section). Future studies are aimed at analysing the mechanism of inhibition of Mtb DnaG by these inhibitors in kinetic and structural detail.

Supplementary Data

• 1.van der Ende A, Baker TA, Ogawa T, Kornberg A. Initiation of enzymatic replication at the origin of the Escherichia coli chromosome: primase as the sole priming enzyme. Proc. Natl Acad. Sci. USA. 1985;82:3954–3958. doi: 10.1073/pnas.82.12.3954. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 2.Kuchta RD, Stengel G. Mechanism and evolution of DNA primases. Biochim. Biophys. Acta. 2010;1804:1180–1189. doi: 10.1016/j.bbapap.2009.06.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 3.Keck JL, Roche DD, Lynch AS, Berger JM. Structure of the RNA polymerase domain of E. coli primase. Science. 2000;287:2482–2486. doi: 10.1126/science.287.5462.2482. [DOI] [PubMed] [Google Scholar]
• 4.Lipps G, Weinzierl AO, von Scheven G, Buchen C, Cramer P. Structure of a bifunctional DNA primase-polymerase. Nat. Struct. Mol. Biol. 2004;11:157–162. doi: 10.1038/nsmb723. [DOI] [PubMed] [Google Scholar]
• 5.Augustin MA, Huber R, Kaiser JT. Crystal structure of a DNA-dependent RNA polymerase (DNA primase) Nat. Struct. Biol. 2001;8:57–61. doi: 10.1038/83060. [DOI] [PubMed] [Google Scholar]
• 6.Ito N, Nureki O, Shirouzu M, Yokoyama S, Hanaoka F. Crystal structure of the Pyrococcus horikoshii DNA primase-UTP complex: implications for the mechanism of primer synthesis. Genes Cells. 2003;8:913–923. doi: 10.1111/j.1365-2443.2003.00693.x. [DOI] [PubMed] [Google Scholar]
• 7.Podobnik M, McInerney P, O’Donnell M, Kuriyan J. A TOPRIM domain in the crystal structure of the catalytic core of Escherichia coli primase confirms a structural link to DNA topoisomerases. J. Mol. Biol. 2000;300:353–362. doi: 10.1006/jmbi.2000.3844. [DOI] [PubMed] [Google Scholar]
• 8.Pan H, Wigley DB. Structure of the zinc-binding domain of Bacillus stearothermophilus DNA primase. Structure. 2000;8:231–239. doi: 10.1016/s0969-2126(00)00101-5. [DOI] [PubMed] [Google Scholar]
• 9.Kato M, Ito T, Wagner G, Richardson CC, Ellenberger T. Modular architecture of the bacteriophage T7 primase couples RNA primer synthesis to DNA synthesis. Mol. Cell. 2003;11:1349–1360. doi: 10.1016/s1097-2765(03)00195-3. [DOI] [PubMed] [Google Scholar]
• 10.Oakley AJ, Loscha KV, Schaeffer PM, Liepinsh E, Pintacuda G, Wilce MC, Otting G, Dixon NE. Crystal and solution structures of the helicase-binding domain of Escherichia coli primase. J. Biol. Chem. 2005;280:11495–11504. doi: 10.1074/jbc.M412645200. [DOI] [PubMed] [Google Scholar]
• 11.Su XC, Schaeffer PM, Loscha KV, Gan PH, Dixon NE, Otting G. Monomeric solution structure of the helicase-binding domain of Escherichia coli DnaG primase. FEBS J. 2006;273:4997–5009. doi: 10.1111/j.1742-4658.2006.05495.x. [DOI] [PubMed] [Google Scholar]
• 12.Bailey S, Eliason WK, Steitz TA. Structure of hexameric DnaB helicase and its complex with a domain of DnaG primase. Science. 2007;318:459–463. doi: 10.1126/science.1147353. [DOI] [PubMed] [Google Scholar]
• 13.Toth EA, Li Y, Sawaya MR, Cheng Y, Ellenberger T. The crystal structure of the bifunctional primase-helicase of bacteriophage T7. Mol. Cell. 2003;12:1113–1123. doi: 10.1016/s1097-2765(03)00442-8. [DOI] [PubMed] [Google Scholar]
• 14.Frick DN, Richardson CC. DNA primases. Annu. Rev. Biochem. 2001;70:39–80. doi: 10.1146/annurev.biochem.70.1.39. [DOI] [PubMed] [Google Scholar]
• 15.Syson K, Thirlway J, Hounslow AM, Soultanas P, Waltho JP. Solution structure of the helicase-interaction domain of the primase DnaG: a model for helicase activation. Structure. 2005;13:609–616. doi: 10.1016/j.str.2005.01.022. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 16.Rowen L, Kornberg A. Primase, the dnaG protein of Escherichia coli. An enzyme which starts DNA chains. J. Biol. Chem. 1978;253:758–764. [PubMed] [Google Scholar]
• 17.Swart JR, Griep MA. Primer synthesis kinetics by Escherichia coli primase on single-stranded DNA templates. Biochemistry. 1995;34:16097–16106. doi: 10.1021/bi00049a025. [DOI] [PubMed] [Google Scholar]
• 18.Mitkova AV, Khopde SM, Biswas SB. Mechanism and stoichiometry of interaction of DnaG primase with DnaB helicase of Escherichia coli in RNA primer synthesis. J. Biol. Chem. 2003;278:52253–52261. doi: 10.1074/jbc.M308956200. [DOI] [PubMed] [Google Scholar]
• 19.Biswas SB, Wydra E, Biswas EE. Mechanisms of DNA binding and regulation of Bacillus anthracis DNA primase. Biochemistry. 2009;48:7373–7382. doi: 10.1021/bi900086z. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 20.Sun W, Schoneich J, Godson GN. A mutant Escherichia coli primase defective in elongation of primer RNA chains. J. Bacteriol. 1999;181:3761–3767. doi: 10.1128/jb.181.12.3761-3767.1999. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 21.Rodina A, Godson GN. Role of conserved amino acids in the catalytic activity of Escherichia coli primase. J. Bacteriol. 2006;188:3614–3621. doi: 10.1128/JB.188.10.3614-3621.2006. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 22.Swart JR, Griep MA. Primase from Escherichia coli primes single-stranded templates in the absence of single-stranded DNA-binding protein or other auxiliary proteins. Template sequence requirements based on the bacteriophage G4 complementary strand origin and Okazaki fragment initiation sites. J. Biol. Chem. 1993;268:12970–12976. [PubMed] [Google Scholar]
• 23.Mendelman LV, Richardson CC. Requirements for primer synthesis by bacteriophage T7 63-kDa gene 4 protein. Roles of template sequence and T7 56-kDa gene 4 protein. J. Biol. Chem. 1991;266:23240–23250. [PubMed] [Google Scholar]
• 24.Scherzinger E, Lanka E, Hillenbrand G. Role of bacteriophage T7 DNA primase in the initiation of DNA strand synthesis. Nucleic Acids Res. 1977;4:4151–4163. doi: 10.1093/nar/4.12.4151. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 25.Zhang Y, Yang F, Kao YC, Kurilla MG, Pompliano DL, Dicker IB. Homogenous assays for Escherichia coli DnaB-stimulated DnaG primase and DnaB helicase and their use in screening for chemical inhibitors. Anal. Biochem. 2002;304:174–179. doi: 10.1006/abio.2002.5627. [DOI] [PubMed] [Google Scholar]
• 26.Koepsell SA, Hanson S, Hinrichs SH, Griep MA. Fluorometric assay for bacterial primases. Anal. Biochem. 2005;339:353–355. doi: 10.1016/j.ab.2004.12.004. [DOI] [PubMed] [Google Scholar]
• 27.Chen J, Brevet A, Fromant M, Leveque F, Schmitter JM, Blanquet S, Plateau P. Pyrophosphatase is essential for growth of Escherichia coli. J. Bacteriol. 1990;172:5686–5689. doi: 10.1128/jb.172.10.5686-5689.1990. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 28.Bailey K, Webb EC. Purification and properties of yeast pyrophosphatase. Biochem. J. 1944;38:394–398. doi: 10.1042/bj0380394. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 29.Josse J. Constitutive inorganic pyrophosphatase of Escherichia coli. 1. Purification and catalytic properties. J. Biol. Chem. 1966;241:1938–1947. [PubMed] [Google Scholar]
• 30.Pais JE, Bowers KE, Stoddard AK, Fierke CA. A continuous fluorescent assay for protein prenyltransferases measuring diphosphate release. Anal. Biochem. 2005;345:302–311. doi: 10.1016/j.ab.2005.07.040. [DOI] [PubMed] [Google Scholar]
• 31.Lawrence AJ, Coote JG, Kazi YF, Lawrence PD, MacDonald-Fyall J, Orr BM, Parton R, Riehle M, Sinclair J, Young J, et al. A direct pyrophosphatase-coupled assay provides new insights into the activation of the secreted adenylate cyclase from Bordetella pertussis by calmodulin. J. Biol. Chem. 2002;277:22289–22296. doi: 10.1074/jbc.M106701200. [DOI] [PubMed] [Google Scholar]
• 32.Upson RH, Haugland RP, Malekzadeh MN. A spectrophotometric method to measure enzymatic activity in reactions that generate inorganic pyrophosphate. Anal. Biochem. 1996;243:41–45. doi: 10.1006/abio.1996.0479. [DOI] [PubMed] [Google Scholar]
• 33.Lloyd AJ, Thomann HU, Ibba M, Soll D. A broadly applicable continuous spectrophotometric assay for measuring aminoacyl-tRNA synthetase activity. Nucleic Acids Res. 1995;23:2886–2892. doi: 10.1093/nar/23.15.2886. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 34.Tammenkoski M, Benini S, Magretova NN, Baykov AA, Lahti R. An unusual, His-dependent family I pyrophosphatase from Mycobacterium tuberculosis. J. Biol. Chem. 2005;280:41819–41826. doi: 10.1074/jbc.M509489200. [DOI] [PubMed] [Google Scholar]
• 35.Biswas T, Tsodikov OV. Hexameric ring structure of the N-terminal domain of Mycobacterium tuberculosis DnaB helicase. FEBS J. 2008;275:3064–3071. doi: 10.1111/j.1742-4658.2008.06460.x. [DOI] [PubMed] [Google Scholar]
• 36.Biswas T, Pero JM, Joseph CG, Tsodikov OV. DNA-dependent ATPase activity of bacterial XPB helicases. Biochemistry. 2009;48:2839–2848. doi: 10.1021/bi8022416. [DOI] [PubMed] [Google Scholar]
• 37.Zhang JH, Chung TD, Oldenburg KR. A simple statistical parameter for use in evaluation and validation of high throughput screening assays. J. Biomol. Screen. 1999;4:67–73. doi: 10.1177/108705719900400206. [DOI] [PubMed] [Google Scholar]
• 38.Tsodikov OV, Enzlin JH, Scharer OD, Ellenberger T. Crystal structure and DNA binding functions of ERCC1, a subunit of the DNA structure-specific endonuclease XPF-ERCC1. Proc. Natl Acad. Sci. USA. 2005;102:11236–11241. doi: 10.1073/pnas.0504341102. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 39.Rymer RU, Solorio FA, Tehranchi AK, Chu C, Corn JE, Keck JL, Wang JD, Berger JM. Binding mechanism of metal-NTP substrates and stringent-response alarmones to bacterial DnaG-type primases. Structure. 2012;20:1478–1489. doi: 10.1016/j.str.2012.05.017. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 40.Lee SJ, Richardson CC. Acidic residues in the nucleotide-binding site of the bacteriophage T7 DNA primase. J. Biol. Chem. 2005;280:26984–26991. doi: 10.1074/jbc.M504817200. [DOI] [PubMed] [Google Scholar]
• 41.Measures JC. Role of amino acids in osmoregulation of non-halophilic bacteria. Nature. 1975;257:398–400. doi: 10.1038/257398a0. [DOI] [PubMed] [Google Scholar]
• 42.Leirmo S, Harrison C, Cayley DS, Burgess RR, Record MT., Jr Replacement of potassium chloride by potassium glutamate dramatically enhances protein-DNA interactions in vitro. Biochemistry. 1987;26:2095–2101. doi: 10.1021/bi00382a006. [DOI] [PubMed] [Google Scholar]
• 43.Chanley JD, Gindler EM, Sobotka H. The mechanism of the hydrolysis of salicyl phosphate.I1. J. Am. Chem. Soc. 1952;74:4347–4352. [Google Scholar]
• 44.Prinz H. Hill coefficients, dose-response curves and allosteric mechanisms. J. Chem. Biol. 2010;3:37–44. doi: 10.1007/s12154-009-0029-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 45.Sheaff RJ, Kuchta RD. Mechanism of calf thymus DNA primase: slow initiation, rapid polymerization, and intelligent termination. Biochemistry. 1993;32:3027–3037. doi: 10.1021/bi00063a014. [DOI] [PubMed] [Google Scholar]
• 46.Sun W, Tormo J, Steitz TA, Godson GN. Domains of Escherichia coli primase: functional activity of a 47-kDa N-terminal proteolytic fragment. Proc. Natl. Acad. Sci. USA. 1994;91:11462–11466. doi: 10.1073/pnas.91.24.11462. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 47.Kuchta RD, Reid B, Chang LM. DNA primase. Processivity and the primase to polymerase alpha activity switch. J. Biol. Chem. 1990;265:16158–16165. [PubMed] [Google Scholar]
• 48.Wickner S, Hurwitz J. Conversion of phiX174 viral DNA to double-stranded form by purified Escherichia coli proteins. Proc. Natl Acad. Sci. USA. 1974;71:4120–4124. doi: 10.1073/pnas.71.10.4120. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 49.Schekman R, Weiner A, Kornberg A. Multienzyme systems of DNA replication. Science. 1974;186:987–993. doi: 10.1126/science.186.4168.987. [DOI] [PubMed] [Google Scholar]
• 50.Koepsell SA, Larson MA, Frey CA, Hinrichs SH, Griep MA. Staphylococcus aureus primase has higher initiation specificity, interacts with single-stranded DNA stronger, but is less stimulated by its helicase than Escherichia coli primase. Mol. Microbiol. 2008;68:1570–1582. doi: 10.1111/j.1365-2958.2008.06255.x. [DOI] [PubMed] [Google Scholar]
• 51.Ono K, Nakane H, Fukushima M. Differential inhibition of various deoxyribonucleic and ribonucleic-acid polymerases by suramin. Eur. J. Biochem. 1988;172:349–353. doi: 10.1111/j.1432-1033.1988.tb13893.x. [DOI] [PubMed] [Google Scholar]
• 52.Frick DN, Richardson CC. Interaction of bacteriophage T7 gene 4 primase with its template recognition site. J. Biol. Chem. 1999;274:35889–35898. doi: 10.1074/jbc.274.50.35889. [DOI] [PubMed] [Google Scholar]
• 53.Mustaev AA, Godson GN. Studies of the functional topography of the catalytic center of Escherichia coli primase. J. Biol. Chem. 1995;270:15711–15718. doi: 10.1074/jbc.270.26.15711. [DOI] [PubMed] [Google Scholar]
• 54.Khopde S, Biswas EE, Biswas SB. Affinity and sequence specificity of DNA binding and site selection for primer synthesis by Escherichia coli primase. Biochemistry. 2002;41:14820–14830. doi: 10.1021/bi026711m. [DOI] [PubMed] [Google Scholar]
• 55.Urban M, Joubert N, Purse BW, Hocek M, Kuchta RD. Mechanisms by which human DNA primase chooses to polymerize a nucleoside triphosphate. Biochemistry. 2010;49:727–735. doi: 10.1021/bi9019516. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 56.Frick DN, Kumar S, Richardson CC. Interaction of ribonucleoside triphosphates with the gene 4 primase of bacteriophage T7. J. Biol. Chem. 1999;274:35899–35907. doi: 10.1074/jbc.274.50.35899. [DOI] [PubMed] [Google Scholar]
• 57.Record MT, Jr, Lohman ML, De Haseth P. Ion effects on ligand-nucleic acid interactions. J. Mol. Biol. 1976;107:145–158. doi: 10.1016/s0022-2836(76)80023-x. [DOI] [PubMed] [Google Scholar]
• 58.Ha JH, Capp MW, Hohenwalter MD, Baskerville M, Record MT., Jr Thermodynamic stoichiometries of participation of water, cations and anions in specific and non-specific binding of lac repressor to DNA. Possible thermodynamic origins of the “glutamate effect” on protein-DNA interactions. J. Mol. Biol. 1992;228:252–264. doi: 10.1016/0022-2836(92)90504-d. [DOI] [PubMed] [Google Scholar]

Supplementary Materials