Characterization of Flavonol Inhibition of DnaB Helicase: Real-Time Monitoring, Structural Modeling, and Proposed Mechanism

DnaB helicases are motor proteins essential for DNA replication, repair, and recombination and may be a promising target for developing new drugs for antibiotic-resistant bacteria. Previously, we established that flavonols significantly decreased the binding ability of Klebsiella pneumoniae DnaB helicase (KpDnaB) to dNTP. Here, we further investigated the effect of flavonols on the inhibition of the ssDNA binding, ATPase activity, and dsDNA-unwinding activity of KpDnaB. The ssDNA-stimulated ATPase activity of KpDnaB was decreased to 59%, 75%, 65%, and 57%, in the presence of myricetin, quercetin, kaempferol, and galangin, respectively. The ssDNA-binding activity of KpDnaB was only slightly decreased by flavonols. We used a continuous fluorescence assay, based on fluorescence resonance energy transfer (FRET), for real-time monitoring of KpDnaB helicase activity in the absence and presence of flavonols. Using this assay, the flavonol-mediated inhibition of the dsDNA-unwinding activity of KpDnaB was observed. Modeled structures of bound and unbound DNA showed flavonols binding to KpDnaB with distinct poses. In addition, these structural models indicated that L214 is a key residue in binding any flavonol. On the basis of these results, we proposed mechanisms for flavonol inhibition of DNA helicase. The resulting information may be useful in designing compounds that target K. pneumoniae and other bacterial DnaB helicases.


Introduction
DNA helicases are motor proteins that play essential roles in DNA replication, repair, and recombination [1,2]. In the replicative hexameric helicase, the fundamental reaction is the unwinding of double-stranded DNA (dsDNA) into single-stranded DNA (ssDNA) intermediates to provide ssDNA templates for DNA polymerases at the replication fork [3]. The most widely studied replicative helicase is Escherichia coli DnaB helicase (EcDnaB) [4]. EcDnaB participates in the initiation of DNA replication once the oriC region of E. coli is bound by the DnaA initiator protein [5,6]. It continues to act during the priming and elongation phases of DNA replication, and it catalyzes ATP hydrolysis and migrates on the DNA template with a strict 5 -3 direction [5]. ATP hydrolysis may drive the movement of the helicase toward the 3 end of the lagging strand [7].
Currently, infections occur that are resistant to all antibacterial options [8]. Few therapies are effective against the six antibiotic-resistant ESKAPE pathogens (Enterococcus faecium, Staphylococcus aureus, Klebsiella pneumoniae, Acinetobacter baumannii, Pseudomonas aeruginosa, and Enterobacter species) [9,10]. K. pneumoniae (Kp) is a ubiquitous opportunistic pathogen that causes severe diseases such as septicemia, pneumonia, urinary tract infections, and soft tissue infections [11]. Since DnaB helicase is required for DNA replication, blocking the activity of DnaB helicase would be detrimental to bacterial survival [12]. In addition,  because the structure and function between eukaryotic and prokaryotic DnaB-like helicases are different [5,12], the K. pneumoniae DnaB helicase (KpDnaB) and other bacterial DnaB-like proteins may be a promising target in developing antibiotics. Previously, we determined the three-dimensional structures of the Geobacillus kaustophilus DnaB-family protein (GkDnaB) and its complex with ssDNA [7]. Although the overall structure of the apo and the complex forms of GkDnaB is similar, the largest difference is found in their loop I region, which undergoes a strong conformational change during the helicase's action [7]. These structural and functional analyses are useful in helping our understanding of the mechanism of DNA translocation in replication forks, and the resulting information may be useful in designing compounds that target bacterial DnaB helicases. Recently, we established that the 4 flavonols ( Figure 1) myricetin (Myr), quercetin (Que), kaempferol (Kae), and galangin (Gal) can interact with KpDnaB and prevent dNTP binding [13]. Flavonoids are the most common group of plant polyphenols, and are responsible for much of the flavor and color of fruits and vegetables [14]. Over 5000 different flavonoids have been described; many of these compounds display structure-dependent biological and pharmacological activities [15]. The 6 major subclasses of flavonoids are flavonols, flavones, flavanones, flavanols, anthocyanidins, and isoflavones [14]. Flavonols, composed of 2 aromatic rings linked by a heterocyclic pyran-4-one ring, are known to have antioxidant [16], antiradical [17], and antibacterial activities [18]. In the present study, we further investigated the effect of flavonols on the inhibition of the ssDNA binding, ATPase activity, and dsDNA-unwinding activity of KpDnaB. A new assay that enables the real-time measurement of DNA helicase activity was also developed. This assay may be useful for screening helicase inhibitors at high throughput.

Materials and Method
2.1. Materials. All restriction enzymes and DNA-modifying enzymes were purchased from New England Biolabs (Ipswich, MA, USA) unless explicitly stated otherwise. All custom oligonucleotide primers were obtained from Invitrogen Corporation (Carlsbad, CA, USA). All chemicals were purchased from Sigma-Aldrich (St. Louis, MO, USA) unless explicitly stated otherwise.

Protein Expression and
Purification. The encoding region of KpDnaB was put on pET21e expression vector and expressed with a hexahistidine tag at the C-terminus of the recombinant protein. The pET21e vector [19] was engineered from the pET21b vector (Novagen Inc., Madison, WI, USA), to avoid having the N-terminal T7 tag fused with the gene product. Details of the construction of pET21e-KpDnaB expression vector and protein purification have been described previously [13]. The purified protein was dialyzed against Buffer B (20 mM HEPES, 100 mM NaCl; pH 7.0), and concentrated to 5 mg/mL. Protein purity remained greater than 95% as determined by Coomassie-stained SDS-PAGE ( Figure 2).

Gel Shifts.
EMSA for KpDnaB was carried out using the same protocol as described previously for SSB proteins [20][21][22], with a minor modification. ssDNA oligonucleotides were custom synthesized by MdBio, Inc., Frederick, MD, USA. Radiolabeling was carried out with [γ 32 P]ATP (6000 Ci/mmol; PerkinElmer Life Sciences) and T4 polynucleotide kinase (Promega, Madison, WI, USA). KpDnaB (0, 90, 170, 340, 680, 1360, 2730, and 5450 nM) was incubated for 30 min at 25 • C with 1.7 nM ssDNA substrate (dT30) and 16.7 μM flavonol in a total volume of 10 μL in 20 mM HEPES (pH 7.0) and 100 mM NaCl. Aliquots (5 μL) were removed from each reaction solution and added to 2 μL of gel-loading solution (0.25% bromophenol blue and 40% sucrose, w/v). The resulting samples were resolved on a native 8% polyacrylamide gel (8.3 × 7.3 cm) at 4 • C in TBE buffer (89 mM Tris borate and 1 mM EDTA) for 1 h at 100 V and visualized by phosphorimaging. The phosphor storage plate was scanned, and the data for complexed and free DNA bands were digitized for quantitative analysis. The ssDNA binding ability (K d,app value) for the protein was estimated from the protein concentration that binds 50% of the input DNA [23][24][25]. Each K d value is calculated as the average of at least three measurements ± S.D.

FRET-Based dsDNA Unwinding Activity Assay.
To monitor the 5 -3 DNA helicase activity of KpDnaB in real-time, we developed an assay on the basis of fluorescence resonance energy transfer (FRET) method. The dsDNA substrate was prepared with the fluorescent strand, 5 -TAGTACCGCCACCCTCAGAACC-3 with Alexa Fluor 488 (maximum excitation/emission = 495/519 nm) coupled to the 5 -end, and the complementary quencher strand, 3 -ATCATGGCGGTGGGAGTCTTGGTTTTTTTTTTTTTT-5 with BHQ1 coupled to the 3 -end, at a 1 : 1.2 concentration ratio. The dsDNA substrate was formed in 20 mM HEPES (pH 7.0) and 100 mM NaCl, by brief heating at 95 • C for 5 min and then followed by slow cooling to room temperature overnight. The fluorescence helicase assay was performed in 20 mM HEPES (pH 7.0), 100 mM NaCl, 3 mM MgCl 2 , 50 nM dsDNA substrate, and 5 mM ATP in 2.0 mL of reaction volume. The unwinding reaction was started by adding KpDnaB (200 nM) and was carried out at 37 • C for 60 min using a spectrofluorimeter (Hitachi F-2700; Hitachi High-Technologies, Tokyo, Japan). The fluorescence intensity was recorded every 5 s. The DNA helicase activity was calculated as the initial reaction velocity from the linear part of the progress curve using the linear regression method.

ATPase Activity Assay.
The protocol for measuring the ATPase activity of purified KpDnaB is based on the colorimetric determination of inorganic phosphate (Pi) released by the hydrolysis of ATP [26]. The ATPase activity assay for KpDnaB (20-50 ng) was performed in 20 mM HEPES (pH 7.0), 100 pmol of φX-ssDNA, 10 μM flavonol, 100 mM NaCl, 3 mM MgCl 2 , and 5 mM ATP in 100 μL of reaction volume. Liberated Pi was detected using ammonium molybdate and malachite green solutions, and absorbance was measured at 610 nm using a UV/vis spectrophotometer (Thermo Scientific Helios Omega; Thermo Fisher Scientific Inc., Waltham, MA, USA). The ATPase activity of KpDnaB was calibrated with sodium phosphate (NaH 2 PO 4 ) of known concentrations.

Sequence Analysis.
The gene KPN04439, encoding K. pneumoniae DnaB helicase, was initially found using a database search through the National Center for Biotechnology Information (NCBI). Based on the known nucleotide sequence, the predicted KpDnaB monomer protein has a length of 471 amino acid residues and a molecular mass of 52.5 kDa, with a pI of 4.93. Analysis of the primary structure of KpDnaB revealed the presence of the putative Walker A motif (aa 232-238), Walker B motif (aa 340-343), ATP binding sites (aa 237-238, 343, 384, 420, 442, and 453), and DNA binding sites (348, 349, 356-358, 383-391, 419-420, and 432-433); these are common in all known DnaB helicases. Figure 3 shows an alignment of the amino acid sequences of K. pneumoniae [13], E. coli [31], Geobacillus stearothermophilus [32], and Geobacillus kaustophilus DnaB helicases [7]; their ATP (boxed) and DNA binding sites (shaded in gray) are highly conserved.

ssDNA-Dependent ATPase Activity of KpDnaB. The
ATPase activity of purified KpDnaB is based on the vanadatesensitive colorimetric determination of inorganic phosphate released by the hydrolysis of ATP [26]. Figure 4(a) shows a standard linear regression curve generated by plotting the optical density (OD) at 610 nm against the concentrations of sodium phosphate (NaH 2 PO 4 ). After some preliminary experiments to establish a suitable range of concentrations for the reagents (including KpDnaB) used for the assay, 3 mM MgCl 2 , 5 mM ATP, 10 μM flavonol, and 20-50 ng of KpDnaB were chosen for analysis of inhibition of KpDnaB ATPase activity by the flavonol. In the absence of the flavonol and φX-ssDNA, the specific activity of KpDnaB was 0.42 ± 0.09 μmol·min −1 ·mg −1 . In the presence of φX-ssDNA, the specific activity of KpDnaB was increased to 5.5 ± 0.6 μmol·min −1 ·mg −1 . These data are similar to reported values for EcDnaB; the activity of EcDnaB is 3.3 × 10 5 (without M13 ssDNA) and 6.4 × 10 6 pmol·min −1 ·mg −1 (with M13 ssDNA), respectively, [33]. Thus, the ATPase activity of KpDnaB was stimulated by ssDNA, a property common to all DnaB helicases.

ATPase Activity of KpDnaB Is Inhibited by Flavonols.
Previously, we showed that the binding of KpDnaB to dNTP is inhibited by the flavonols Myr, Kae, Gal, and Que [13]. Here, as shown in Figure 4(b), the ATP hydrolysis activity of KpDnaB was also inhibited by the flavonols. In the presence of Myr, Que, Kae, or Gal, the specific activity of KpDnaB was decreased to 59%, 75%, 65%, and 57%, respectively, meaning that Gal exhibited the strongest inhibitory effect on the ATP hydrolysis of KpDnaB in the presence of φX-ssDNA.

Inhibitory Effects of Flavonols on ssDNA Binding of KpDnaB.
To investigate whether the ssDNA-binding ability of KpDnaB is inhibited by the flavonol, we used EMSA to study the binding of KpDnaB to dT30 ( Figure 5) when mixed with Myr, Que, Kae, or Gal. In the absence of any flavonol, KpDnaB formed a stable complex with dT30, as shown by electrophoresis ( Figure 5(a)). In the presence of 16.7 μM Gal ( Figure 5

Real-Time Monitoring of KpDnaB Helicase Activity Based on FRET.
To investigate whether the helicase activity of KpDnaB is also inhibited by flavonols, the fluorescence helicase assay was carried out in the absence or presence of Myr or Gal. This assay, modified from that for HCV NS3 3 -5 RNA helicase [34], was found to be useful for KpDnaB. The dsDNA substrate was prepared by annealing 2 oligonucleotides, a 5 fluorophore-labeled (Alexa Fluor 488) 22-nucleotide donor and a 3 quencher-labeled (BHQ1) 36-nucleotide quencher (Figure 6(a)). When the dsDNA substrate is unwound by the helicase, the fluorophore (F) emits upon its release from the quencher (Q). The fluorescence quenching efficiency (signal-to-background ratio) for this substrate was >90%, suggesting high sensitivity for monitoring helicase activity, similar to that for HCV NS3 RNA helicase [34]. Because of its high inhibitory effects on the ATPase activity of KpDnaB (Figure 4(b)), Myr and Gal were selected for this assay. When KpDnaB was added, fluorescence was continuously emitted, whereas no increase in fluorescence occurred in the absence of KpDnaB (as the negative control). These results indicate that the observed fluorescence emission arose from the unwinding of the dsDNA substrate by purified KpDnaB (Figure 6(b)). When assaying the helicase activity of KpDnaB in the presence of Myr or Gal, a linear increase in fluorescence emission was also observed. In addition, the initial velocity for KpDnaB helicase activity was nearly identical with or without flavonol (∼1 × 10 −3 fluorescence intensity/s). In contrast, unlike fluorescence emission of KpDnaB that continued to increase, the fluorescence emission of KpDnaB assayed in the presence of the flavonol Myr or Gal soon reached its maximal point. The magnitude of fluorescence activity for KpDnaB was in the following order: no flavonol added > Myr present > Gal present.

Homology
Modeling. The most thoroughly studied replicative DnaB helicase from bacteria is that of E. coli. Although KpDnaB shows a high degree of sequence identity with EcDnaB, its crystal structure is not yet reported. To deeply understand the structure-function relationship of the flavonol-KpDnaB-ssDNA complex(es), we decided to model its three-dimensional structure by homology modeling. The model of KpDnaB was built from the coordinates of 2R6D (crystal structure of GsDnaB), 2VYF (crystal structure of GkDnaB) and 2VYE (crystal structure of GkDnaB in complex with ssDNA) by using SWISS-MODEL, http://swissmodel.expasy.org/ [28]. KpDnaB and GkDnaB share 47% identity and 69%, similarity, and KpDnaB and GsDnaB share 46% identity and 69% similarity, respectively, in the amino acid sequences level. The three-dimensional model of the hexameric KpDnaBs all forms a ring structure (Figure 7(a)), of which the loop I position in GkDnaB-ssDNA complex based model was different. We propose that, as in the GkDnaB complexes [7], loop I of KpDnaB undergoes a conformational change, and thereby playing an important role in stabilizing DNA binding. It should be noted that the NTP-binding site, Walker A and B motifs, is adjacent to the DNA interaction site loop I (Figure 7(b)).

Docking Study Using
PatchDock. The flavonol, found in DrugBank, was individually computationally docked into the three-dimensional models of KpDnaB by using Patch-Dock, http://bioinfo3d.cs.tau.ac.il/PatchDock/ [30]. After uploading the coordinate and topology file of the flavonol and KpDnaB, the docking was automatically performed. The docking results for KpDnaB (GkDnaB based model) interacting with Myr or Que are shown in Figure 8 and Table 1. Myr (Figure 8(a)) and Que (Figure 8(b)) were found to be docked in the ATP-binding pocket of KpDnaB; however, there were also a few binding poses docked outside the pocket (Figure 8(b)). Despite a similar structure, these flavonols were found to be docked with distinct binding poses in the ATP-binding pocket of both DNA-unbound and DNA-bound models of KpDnaB. For example, the binding pose of KpDnaB to Myr is different from that of Que (Figure 9(a)). Que, the compound with 2 hydroxyl groups on the B ring, interacts with L214 and N215, whereas Myr (3 hydroxyl groups) interacts with L214, N247, and F460. In addition, the same compound is also found to be docked in different positions of the ATP-binding pocket of both DNAunbound and DNA-bound models of KpDnaB. For instance, Myr interacts with L214, N247, and F460 in the DNAunbound model, and interacts with L214 and N462 in the DNA-bound model of KpDnaB, respectively (Figure 9(b)). Generally, L214 is a key residue in all DNA-unbound and DNA-bound modeled structures regardless of which flavonol was used.

Discussion
The development of clinically useful small molecule antibiotics has been a seminal event in the world of infectious diseases [35]. DNA replication is one of the most basic biological functions and should be a prime target in antibiotic development. For example, some novel inhibitors were discovered to target topoisomerase [36,37], DNA gyrase [38], RNA polymerase [39], helicase-primase [40], and retroviral reverse transcriptase [41,42]. Since DNA helicases are important components of the cellular replication machinery in all organisms, inhibition of helicase activity would be detrimental to bacterial survival as well. Previously, we have shown that some flavonoid compounds can directly bind to KpDnaB and inhibit its binding ability  to nucleotides, thus interfering with the growth of K. pneumoniae [13]. Flavonoids [14] are the most common group of plant polyphenols with antioxidant [16], antiradical [17], and antibacterial activities [18]. It is now clear that some flavonoids are ATP-inhibiting agents as competitors for ATP-binding proteins. For example, several flavonoid derivatives have been developed as therapeutic agents for cancer [43]. In this study, we used several assays to analyze the effects of 4 flavonols, namely, Myr, Que, Kae, and Galwhich contain different numbers of hydroxyl substituents on the aromatic rings-on the ssDNA binding, ATP hydrolysis, and dsDNA unwinding abilities of KpDnaB. For the first time, our results demonstrated that these flavonols were capable of inhibiting the unwinding activity of the helicase.
DnaB helicase is an ssDNA-dependent ATPase. Previously, we showed that the binding of KpDnaB to dNTP is inhibited by the flavonols Myr, Kae, Gal, and Que. This study further investigated the flavonol-mediated inhibition of KpDnaB binding to dNTP, which may have led to the associated decrease in ATPase activity in the presence of ssDNA (Figure 4). The ATP-hydrolyzing activity of KpDnaB was inhibited by flavonols in the following order of decreasing efficiency: Gal > Myr > Kae > Que. Although it is well established that flavonoids have several hydroxyl groups, and thus have marked potentials to bind any protein, the strength of inhibition of KpDnaB ATP-hydrolyzing activity did not correlate with the number of hydroxyl substituents on the flavonol aromatic rings. In addition, the structural model of KpDnaB shows that these flavonols bind to KpDnaB with distinct binding poses in the ATP-binding pocket (Figure 9). This situation is also found in several ATP-binding proteins that bind to different flavonoids. Crystal structures of PIM1 kinase in complex with quercetagetin and Myr reveal 2 distinct binding poses in the ATP-binding pocket, namely, orientations I and II [44]. Myr, adopting orientation II, has flipped 180 • in PIM1 kinase, in contrast to quercetagetin (orientation I), such that the B ring is oriented toward the entrance of the ATP pocket. These orientations in PIM1 are also found to closely resemble those of Que and Myr in phosphatidylinositol 3-kinase [45]. Although the ATPbinding pocket in these proteins, including KpDnaB, is different, cases of PIM1 and phosphatidylinositol 3-kinase with different binding poses, are very similar to that of KpDnaB. Despite a very similar structure, flavonols may bind to KpDnaB by using different mechanisms, as shown in the structural models (Figures 8 and 9). This may explain why the strength of inhibition in KpDnaB ATPase activity was not correlated with the number of hydroxyl substituents on the flavonol aromatic rings.
Other studies showed that Myr noncompetitively inhibits E. coli DnaB helicase with an IC 50 of approximately 10 μM [46]; however, we have recently discovered that Gal, Kae, Que, and Myr can inhibit dNTP binding to KpDnaB, indicating a specific inhibitory process. Furthermore, these flavonols can be docked into the KpDnaB ATP-binding pocket. It is not known whether this disparity is due to the inherent differences among the species, or an alternative (allosteric) binding site(s) in DnaB helicase. However, the docking results also show that a few binding poses of the flavonol outside the ATP-binding pocket ( Figure 8), suggesting that there may be more than one site for flavonol binding in the DnaB helicase. This may explain why Myr non-competitively inhibits EcDnaB. In addition, we also showed flavonol-mediated inhibition of KpDnaB-ssDNA complex formation, which depended on the [flavonol]/[KpDnaB monomer] ratio; at a ∼6 fold ratio, the inhibitory effect did disappear ( Figure 5). Since these flavonols inhibited not only the ATPase activity of KpDnaB but also the ssDNA-binding ability, we believe that there may be more than one site in KpDnaB for flavonol binding. However, this speculation must be confirmed by further biochemical experiments.
In this study, we describe a new in vitro fluorescence assay for measuring 5 -3 DNA helicase activity by using dsDNA substrate ( Figure 6). Alexa Fluor 488 and BHQ1 were selected as the fluorophore-quencher pair. This assay enables the real-time, high-throughput measurement of DNA helicase activity, and does not require time-consuming procedures like the conventional gel-based assays. For example, on the basis of this assay, we observed that the initial velocity of KpDnaB for the unwinding activity assayed in the absence of the flavonol, or with Myr or Gal were very similar; however, their maximal activities were different. While fluorescence was continuously emitted without the addition of flavonol to the KpDnaB solution, the fluorescence increase stopped at ∼300 s and ∼600 s for Myr and Gal, respectively ( Figure 6(b)). This real-time unwinding kinetics of the DNA helicase cannot be easily observed by the conventional gelbased assays. Our laboratory is currently screening DNA helicase inhibitors using this high-throughput method.
All DNA-unbound and DNA-bound modeled structures showed flavonols binding to KpDnaB with distinct poses. However, these models all displayed a key residue involved in the flavonol binding, namely, L214. The L214 residue in DnaB helicases is highly conserved (Figure 3), but its role has not yet been determined. On the basis of these results, we propose that these flavonols may inhibit KpDnaB in 2 possible ways. First, since DnaB helicase binding to dNTP causes a large conformational change [31,47,48] to become a translocase [7], these flavonols may partially occupy the ATP-binding pocket of the DnaB helicase and inhibit conformational change, thereby causing varying degrees of inhibition. This is a possible inhibition mechanism because the L214 residue of KpDnaB is not involved in ATP binding (Figure 3), but most structural models indicate its importance in binding flavonols ( Figure 9). Second, more than one flavonol binds to KpDnaB; 1 is at the active site, and the other(s) is at an unknown site. The overall inhibition possibly results from the interactions between the flavonols and the enzyme. This may be a reason why flavonols not only bind at the ATP pocket of the DNA helicase, but can also non-competitively inhibit the ATPase activity of the DNA helicase [46]. Our crystal structure of GkDnaB in complex with ssDNA previously suggested that ATP hydrolysis may drive the movement of the helicase toward the 3 end of the lagging strand [7]. In addition, the dNTP-binding site of the helicase at loop I, part of the Walker B motif, is adjacent to the DNA interaction site. In conclusion, the flavonol likely inhibits the DNA helicase (translocase) activity by affecting the ATP binding of KpDnaB, to ultimately shut down and lock the enzyme in the ATP-unbound state.