|Home | About | Journals | Submit | Contact Us | Français|
Lyme disease, the most common vector-borne zoonosis in North America, is caused by the spirochetal pathogen Borrelia burgdorferi. The telomere resolvase encoded by this organism (ResT) promotes the formation of covalently closed hairpin ends on the linear DNA molecules of B. burgdorferi through a two-step transesterification. ResT is essential for survival and is therefore an attractive target for the development of highly specific antiborrelial drugs. To identify ResT inhibitors, a novel fluorescence-based high-throughput assay was developed and used to screen a library of 27,520 small-molecule drug-like compounds. Six confirmed inhibitors of ResT, with 50% inhibitory concentrations between 2 and 10 μM, were identified. The inhibitors were characterized further and were grouped into three distinct classes based on their inhibitory features. The high-throughput screening assay developed in this paper, along with the six inhibitory compounds identified, provides a starting point for the future development of novel antiborrelial drugs as well as small-molecule inhibitors that will be helpful for the further dissection of the reaction mechanism.
Lyme disease, caused by the bacterium Borrelia burgdorferi sensu lato, is the most common zoonosis in North America and is a problem in many parts of the world (31, 32). The disease is transmitted by hard-body ticks of the genus Ixodes, which usually acquire the infection from small rodents, such as the white-footed mouse (Peromyscus leucopus). Infected ticks subsequently transmit the spirochete to naïve mice, larger mammals such as deer, or an inadvertent human host. Environmental factors such as climate change (24, 25), reforestation, and increasing deer populations (30) are probably responsible for an increasing range of infected ticks and Lyme disease. Since the OspA (outer surface protein A)-based Lyme disease vaccine (33) was discontinued for use in humans, prevention of the spread of Lyme disease has relied solely on personal protection from ticks, prophylactic antibiotic treatment, and tick control (28).
B. burgdorferi has a unique genome composed of a linear chromosome and more than 20 extrachromosomal elements or plasmids, most of which are linear (9, 15). Each of the linear plasmids is terminated by covalently closed hairpin ends or telomeres (3, 8, 16, 35). Replication of the linear plasmids initiates at a central origin of replication and proceeds bidirectionally (4, 27). The resulting dimer junctions are recognized and processed by ResT, a telomere resolvase, which catalyzes a two-step transesterification reaction known as telomere resolution (10, 11, 20). This reaction generates the covalently closed hairpin telomeres found on the ends of all linear replicons in Borrelia species (see reference 10 for a recent review).
The B. burgdorferi telomere resolvase has been studied extensively (2, 12, 18-20, 34-37), and the reaction is catalyzed through a nucleophilic attack by tyrosine 335, which forms a 3′ phosphotyrosyl-enzyme intermediate. Subsequently, a conformational change occurs in order to position the 5′-OH terminus of the opposite DNA strand for nucleophilic attack of the protein-DNA linkage and sealing of the DNA backbone. The catalytic residues of ResT are similar to those of some other DNA breakage and reunion enzymes, specifically type IB topoisomerases and tyrosine recombinases, and ResT possesses a hairpin binding module that may be similar to that found in cut-and-paste transposases (1).
Since ResT is essential for the survival of B. burgdorferi (7, 34), and a telomere resolvase has been reported for only one other bacterium, ResT is a promising target for the development of highly specific antiborrelial agents for the prevention and treatment of Lyme disease. Drugs that target ResT will be useful for mechanistic studies of the enzyme and may provide more-specific and safer treatment alternatives than antimicrobial therapy with currently available drugs (17), which can result in the spread of antibiotic resistance. ResT-specific drugs might also be useful on an environmental scale to eradicate B. burgdorferi from mammalian reservoirs such as the white-footed mouse, thus preventing the environmental spread of infected ticks and reducing transmission to humans (13, 38).
In this study we report the development of a high-throughput screening assay to identify ResT inhibitors. The fluorescence-based assay was used to screen a library of 27,520 small molecules. We report the finding of six inhibitors of ResT with 50% inhibitory concentrations (IC50s) between 2 and 10 μM.
All chemicals used were of analytical grade and did not require further purification. Restriction enzymes were from New England Biolabs (Ipswich, MA). Unmodified oligonucleotides were synthesized and purified by the University of Calgary Core DNA Services (Calgary, Alberta, Canada). The synthesis and use of oligonucleotides containing a 5′-bridging phosphorothiolate (OPS) have been described previously (6, 12, 18).
A bacterial strain (GCE203) carrying plasmid pYT1 (36) was grown overnight at 30°C with agitation at 250 rpm in a Fernbach flask containing 1 liter of Luria-Bertani (LB) broth and 50 μg/ml kanamycin. Plasmid pYT1 was purified using a Qiafilter Plasmid Mega kit (Qiagen) according to the instructions for low-copy-number DNA. The DNA concentration was determined via absorbance at 260 nm. For use in telomere resolution assays, plasmid DNA was digested using PstI (New England Biolabs) according to the manufacturer's specifications.
To overexpress wild-type ResT, a 5-ml starter culture of GCB195 (20) was grown overnight at 30°C in LB broth containing 1% glucose, 30 μg/ml chloramphenicol, and 100 μg/ml ampicillin. This culture was then added to 1 liter of LB broth containing 1% glucose, 30 μg/ml chloramphenicol, and 100 μg/ml ampicillin and was grown to an optical density of 0.4 at 32°C. Protein expression in Escherichia coli was induced by the addition of isopropyl-β-d-thiogalactopyranoside (IPTG) to 1 mM, and growth continued at 18°C and 250 rpm overnight. Recombinant ResT1-449 was purified as described previously (20).
To express ResT164-449, a 5-ml culture of pYT42 was grown overnight at 37°C and 250 rpm in LB broth containing 100 μg/ml carbenicillin and 25 μg/ml of chloramphenicol. One milliliter of this starter culture was used to inoculate 250 ml of LB broth containing 100 μg/ml carbenicillin and 25 μg/ml of chloramphenicol. This culture was incubated at 37°C and 250 rpm to an optical density of 0.45. IPTG was added to 0.75 mM, and the culture was incubated at 18°C and 250 rpm overnight. ResT164-449 was purified as described previously (37).
ResT activity was measured by a fluorescence-based high-throughput screen (see Fig. Fig.1).1). The reactions were performed, and results were measured, in 384-well black microplates (catalogue number 781076; Greiner Bio-One). The final assay conditions were as follows: 10% dimethyl sulfoxide (DMSO) (dehydrated with molecular sieves [pore size, 3 Å]; Fisher Scientific), 395 pM PstI-linearized pYT1, 25 mM Tris-HCl (pH 8.5), 1 mM EDTA, 100 mM NaCl, 19 μg/ml bovine serum albumin (BSA), and 11 nM ResT in a final reaction volume of 30 μl. Reaction mixtures were assembled by the MultiProbe HT EX liquid handling system by the ordered addition of 3 μl of DMSO; 21 μl of the reaction mixture containing the DNA, Tris buffer, EDTA, NaCl, and BSA; and 6 μl of ResT. Following a 15-min incubation at room temperature, 6 μl Na3PO4 was added to the reaction mixture to 42 mM, and the reaction mixture was incubated for 10 min at room temperature. Subsequently, an equal volume (36 μl) of SYBR green (Molecular Probes) at a 2× concentration in 10 mM Tris-HCl (pH 7.5) plus 1 mM EDTA was added. The final reaction volume was 72 μl. The plates were read on a Fusion Universal microplate analyzer (Perkin-Elmer) at an excitation wavelength of 485 nm and an emission wavelength of 530 nm. The ResT stock was kept on ice between additions.
ResT was screened against a Canadian Chemical Biology Network library of 27,520 compounds in a semiautomated fashion on an eight-pin MultiProbe HT EX liquid handling system. Reaction mixtures contained 10 μM concentrations of the library compounds. Due to mechanical constraints, 96 reactions were performed at a time in 384-well plates (4 groups of reactions per 384-well plate). Each group of 96 reactions included eight positive controls (reaction mixtures with ResT and without a library compound) and eight negative controls (reaction mixtures with buffer instead of ResT and without a library compound). Primary inhibitors were defined as compounds that reduced the average activity of the positive controls by 70% or more. The inhibition of ResT activity was calculated using the following equation: percent inhibition = 100 × [1 − (fluorescence − average fluorescence for negative controls)/(average fluorescence for positive controls − average fluorescence for negative controls)].
The assay quality was determined by calculating the Z′ factor (39) as discussed in Results and Discussion: Z′ = 1 − [(3 standard deviations for the positive control + 3 standard deviations for the negative control)/(mean of the positive-control data − mean of the negative-control data)].
The inhibitors identified from the primary screen were confirmed by repeating the assay with those compounds. Compounds that retained the ability to inhibit ResT activity by 70% or more were classified as confirmed inhibitors. These compounds were also tested in the presence of 0.01% (wt/vol) BSA, and 1 mM 1,4-dithiothreitol (DTT), each in separate trials. The compounds were also tested for their ability to quench fluorescence by incubating the compounds with the products of completed ResT reactions, followed by the addition of Na3PO4 and SYBR green as noted above.
The 62-bp wild-type replicated telomere substrate contained the annealed oligonucleotides OGCB158 (5′-CCG GCG CCG CGG CAC TCT ATA CTA ATA AAA AAT TAT ATA TAT AAT TTT TTA TTA GTA TAG AGT 3′) and OGCB159 (5′ TAC TCT ATA CTA ATA AAA AAT TAT ATA TAT AAT TTT TTA TTA GTA TAG AGT GCC GCG GCG CCG-3′) and were assembled as previously described (1). The oligonucleotides were 5′ end labeled using T4 polynucleotide kinase (New England Biolabs) and [γ-32P]ATP according to the manufacturer's instructions. Annealing of oligomers was carried out by heating equimolar amounts of each oligonucleotide at 95°C for 5 min. The assembled oligonucleotides were then slow-cooled to 30°C. Assembled oligonucleotides were electrophoresed on a 4% Metaphor gel for 3.5 h at 30 mA to separate the annealed substrate from unannealed oligonucleotides and unincorporated [γ-32P]ATP. Gels were stained using ethidium bromide, and bands corresponding to the assembled substrate were excised. DNA was extracted from the gel slices by freezing the gel slices at −80°C for 20 min and then spinning them through a Spin-X Centrifuge tube filter (0.22-μm-pore size cellulose acetate membrane in a 2.0-ml tube; Costar) for 8 min at 8,000 rpm and 4°C. The OPS substrates were made from oligonucleotides OGCB116 (5′-GAT CCA CTC TAT ACT AAT AAA AAA TTA TAT AT-3′) and OGCB117 (5′ ATA ATT TTT TAT TAG TAT AGA GTG-3′) and were assembled as previously described (1, 12, 18, 19). OGCB116 was 5′ end labeled as described above with [γ-32P]ATP, and OGCB117 was similarly phosphorylated, but with cold ATP. In both reactions, the polynucleotide kinase was heat inactivated at 65°C for 45 min. The oligonucleotides were annealed as described above and were then ligated at 16°C for 2 h. The replicated telomeres were purified as described above. The 140-mer wild-type replicated telomere was composed of annealed oligonucleotides OGCB2 (5′-GAT CCA AGT TAA AGT TAG CAA TTT AAA GGG TAA AGT TTT GAG TCA AAA TAC TCT ATA CTA ATA AAA AAT TAT-3′) and OGCB3 (5′-ATA TAT AAT TTT TTA TTA GTA TAG AGT ATT TTG ACT CAA AAC TTT ACC CTT TAA ATT GCT AAC TTT AAC TTG-3′). OGCB3 was 5′ end labeled as described above using [γ-32P]ATP. Equimolar amounts of OGCB2 and OGCB3 were annealed, assembled, and purified as previously described for OGCB116 and OGCB117 (1, 12, 18, 19).
The IC50 values for the compounds were determined by performing telomere resolution assays using the following reaction conditions: 1.1 nM PstI-linearized pYT1 plasmid DNA, 10 μM each compound (10% DMSO [final concentration]), 110 nM ResT, 25 mM Tris-HCl (pH 8.5), 100 mM NaCl, and 1 mM EDTA in a final reaction volume of 30 μl. The reaction mixtures were incubated at 30°C for 30 min, and reactions were stopped by the addition of sodium dodecyl sulfate (SDS) to a final concentration of 0.1%. Samples were analyzed by electrophoresis on a 0.8% agarose gel for 1.5 h at 75 V. Gel documentation and band intensity quantification were performed using the FluorChem 8900 imager and AlphaEaseFC software (Alpha Innotech). The percentage of telomere resolution was determined by dividing the net area for the product bands by the total area of all of the bands (substrate plus products). The relative percentage of telomere resolution was determined by calculating the percentage telomere resolution exhibited by the reaction mixtures containing the inhibitors relative to the percentage of telomere resolution exhibited by the positive control, defined as 100%. The IC50 values for each compound were obtained by nonlinear-regression curve fits (sigmoidal dose-response curves) using GraphPad Prism software, version 4.
For the assessment of telomere resolution using the oligonucleotide substrates, the following reaction conditions were used: 5.4 nM oligonucleotide substrate (OGCB158+OGCB159), 165 nM ResT, 10 μM each compound (10% DMSO [final concentration]), 25 mM Tris-HCl (pH 8.5), 100 mM NaCl, and 1 mM EDTA in a total reaction volume of 30 μl. In order to assess telomere cleavage, the following reaction conditions were used: 5.4 nM oligonucleotide DNA (OGCB116+OGCB117), 165 nM ResT, 10 mM each compound (10% DMSO [final concentration]), 25 mM Tris-HCl (pH 8.5), 100 mM NaCl, 1 mM EDTA, and 25% glycerol. All reaction mixtures were incubated at 30°C for 30 min, and reactions were stopped using 0.1% SDS. Samples were loaded onto 5% polyacrylamide-SDS gels and run in 1× E buffer (11) supplemented with 0.1% SDS at 30 mA for 2 h at 4°C. The C-terminal ResT protein was added to binding reaction mixtures in a final concentration of 245 nM in a 30-μl reaction volume. The reaction mixtures contained 10 μM each inhibitor, 5.4 nM substrate DNA (OGCB2+OGCB3), 25 mM Tris-HCl, 100 mM NaCl, and 0.5 μg/ml salmon sperm DNA. Reaction mixtures were incubated for 20 min at 0°C. Samples were loaded onto a 5% native polyacrylamide gel in E buffer and were run at 30 mA for 2 h. All polyacrylamide gels were vacuum dried, and radioactive signals were detected using a Packard Cyclone phosphorimager. Band intensities were quantified using ImageQuant software (Molecular Dynamics). The percentage of telomere resolution, DNA cleavage (as evidenced by the formation of a covalent protein-DNA complex [CPD]), or DNA binding was determined by dividing the net counts corresponding to the end products (product minus background) by the total counts (substrate plus net products). The relative percentage of resolution/cleavage/shift was determined by calculating the percentage of telomere resolution, DNA cleavage (CPD formation), or DNA binding in the reaction mixtures containing inhibitors relative to those for positive controls (containing ResT and DMSO), which were considered to be 100%.
The compounds were analyzed via electrospray ionization-time-of-flight mass spectrometry (by the Southern Alberta Mass Spectrometry Centre for Proteomics, a Core Facility of the University of Calgary, Calgary, Alberta, Canada).
A schematic describing the fluorescence-based high-throughput screen developed for ResT, based on its unique activity, is shown in Fig. Fig.1A.1A. A linearized plasmid substrate (pYT1) containing the 50-bp replicated telomere from the left end of the B. burgdorferi linear plasmid 17 (lp17) was treated with ResT, which performs a two-step transesterification to cleave the two indicated phosphodiester bonds and generate two DNA molecules terminated with covalently closed hairpin ends. Under alkaline conditions (pH ~12), unreacted DNA substrate was irreversibly denatured and did not form appreciable regions of base-paired secondary structure (23), as indicated by low levels of fluorescence emitted by a sensitive and specific double-stranded DNA-binding dye, SYBR green. However, hairpin-containing products “snapped back” to re-form double-stranded DNA molecules, which were detected by the addition of SYBR green. Upon excitation at 497 nm, SYBR green bound to double-stranded DNA emits a signal at 520 nm, while the single-stranded substrate emits only a very low background signal. In this assay, the inhibition of ResT by small molecules resulted in a depletion of the fluorescence signal.
The high-throughput assay described here differs from typical DNA-based high-throughput assays in its use of a 4.6-kb plasmid substrate. The use of a plasmid rather than an oligonucleotide substrate provided enhanced sensitivity; the fluorescence yield for SYBR green bound to 2.0- and 2.6-kb DNA fragments is almost 2 orders of magnitude higher on a molar basis than that for oligonucleotide products of 25 to 30 bp. The amount of plasmid substrate (36 ng/11.85 fmol) used per 30-μl reaction mixture provided a robust signal and required only about 1.2 mg of DNA substrate for 33,000 reactions. The low substrate concentration (395 pM) also minimized the concentration of enzyme required (11 nM) to perform a large number of screening reactions.
In order to determine the reliability of the screening assay, replicates of positive and negative controls were prepared manually. The Z′ factor, a parameter calculated using control data, is indicative of the assay quality (39) and was found to be 0.84 for the manual assay (data not shown). A Z′ factor of 0.5 or greater is indicative of an excellent screen with a robust signal window (39). Following optimization of the manually prepared reaction conditions, an eight-pin MultiProbe HT EX liquid handling system (Perkin-Elmer) was programmed to run the assay semiautomatically in a 384-well plate setting. One hundred ninety-two replicates of both positive- and negative-control assays were prepared using the MultiProbe liquid handling system, yielding a Z′ factor of 0.54 (data not shown).
The compound library that was used to screen for ResT inhibitors was obtained from the Canadian Chemical Biology Network (http://www.ccbn-rcbc.ca/?q=profile) and contained 27,520 synthetic and natural small molecules (http://www.ccbn-rcbc.ca/?q=canadiancollection). A schema of the screening method is given in Fig. Fig.2.2. The primary and secondary screens were run in a semiautomated fashion as described in Materials and Methods. The Z′ factor for the primary screen was 0.47 (Fig. (Fig.1B).1B). The threshold for inhibitors in the primary screen was set at 70% inhibition of ResT activity in the presence of the compound. This threshold resulted in 147 confirmed primary-screen inhibitors.
Secondary screening involved experimental steps aimed at eliminating nonspecific inhibitors of ResT. Promiscuous inhibitors can inhibit by binding or reacting nonspecifically with the target enzyme. The effect of inhibition detected by promiscuous inhibitors will decrease in the presence of another protein, such as BSA (22). Sulfur-containing compounds also have the potential to react with the two cysteine residues of ResT leading to inhibition of the enzyme (14, 21). Compounds that inhibited the ResT reaction less than 70% upon the addition of either BSA or DTT were eliminated. The remaining 46 compounds were tested for the ability to quench the fluorescent signal of SYBR green. Compounds that reduced fluorescence by 70% or more when added to ResT reaction mixtures after complete telomere resolution were labeled as fluorescence quenchers and were also eliminated. Upon completion of the secondary screen, 25 compounds still exhibited 70% inhibition of ResT activity in the presence of BSA and DTT and did not quench the fluorescent signal of SYBR green. The remaining 25 compounds were tested for the ability to inhibit ResT-mediated telomere resolution by use of an in vitro gel-based assay (Fig. (Fig.3A).3A). Compounds were screened at a concentration of 10 μM under standard ResT reaction conditions previously used for mechanistic studies of the enzyme (1, 2, 12, 18-20, 36, 37). Compounds that did not inhibit ResT activity at a level of 70% or more under standard assay conditions were also eliminated.
The remaining 16 compounds were tested for the ability to bind to double-stranded DNA, a feature that would likely result in the ability to inhibit a large number of DNA-metabolizing enzymes indirectly through DNA binding. Nonspecific DNA binding was tested by examining the ability of compounds to inhibit the restriction endonucleases EcoRI and ClaI. In addition, the DNA binding activities of the compounds were probed by testing their abilities to displace the fluorescent DNA-binding compound PicoGreen when bound to DNA. None of the remaining 16 compounds affected the restriction activity of EcoRI or ClaI, and the binding ability of the DNA-binding dye PicoGreen was not affected (data not shown).
Finally, the remaining 16 compounds were reordered from their respective suppliers, and the IC50 of each was determined using the gel-based telomere resolution assay (Fig. (Fig.3).3). Compounds that did not exhibit reproducible IC50 values of ≤10 μM upon reordering were eliminated, leaving six inhibitors of ResT with IC50s of <10 μM. The identities of these six compounds were verified by mass spectroscopy; their structures and corresponding information are shown in Fig. Fig.44.
Compound 5 is an aromatic carboxylic acid containing a pharmacophore that could mimic the interaction of a nucleotide with an enzyme. Compound 5 shares structural similarities with compound 8, which is also an aromatic carboxylic acid. Compounds 15 and 22 are both aliphatic carboxylic acids with large and bulky aliphatic groups. Compound 15 has a sterol scaffold, whereas compound 22 has a fatty acid scaffold. Compounds 5, 8, 15, and 22 are all large hydrophobic molecules with a hydrogen bond-capable pharmacophore (carboxylate moiety). Compound 14 is a polyphenolic compound composed of several pharmacophores, many of which are capable of hydrogen bonding. Compound 11 is also a polyphenolic compound with its own distinctive arrangement of pharmacophores, many capable of forming hydrogen bonds.
The telomere resolvase ResT performs a two-step transesterification reaction with several distinct steps along the reaction pathway (see reference 10). It was, therefore, of interest to determine the stage at which each of the inhibitory compounds was affecting the reaction. The first chemical step in the reaction results in the formation of a transient 3′-phosphotyrosyl-ResT intermediate (CPD) that is rapidly converted either into the product or back into the substrate through reaction reversal. Evaluation of the DNA cleavage step in isolation, therefore, requires stabilization of the covalently linked protein-DNA reaction intermediate. This can be accomplished through the use of a synthetic oligonucleotide substrate carrying an OPS substitution at the scissile phosphate (6, 12, 18). In an OPS substrate, the 5′-bridging oxygen, which is responsible for the nucleophilic attack of the phosphotyrosine linkage in the second transesterification, is replaced with sulfur (Fig. (Fig.5A).5A). Sulfur is a good leaving group, resulting in efficient substrate cleavage, but a poor nucleophile in phosphate chemistry; thus, the phosphotyrosine linkage between ResT and the DNA substrate cannot be transesterified, leaving ResT permanently attached to the substrate.
ResT reactions with the various inhibitory compounds were performed using an OPS oligonucleotide substrate, and the cleavage step was monitored by assessing the amount of CPD generated by use of a denaturing (SDS) gel assay (Fig. (Fig.5A).5A). The level of DNA cleavage in the presence (10 μM) of each inhibitor relative to that in the absence of inhibitor is shown in Fig. Fig.5B.5B. The amount of DNA cleavage by ResT in the presence of compounds 5, 8, and 14 (65%, 43%, and 32%, respectively) was significantly different (>95%) from that for the positive control by Student's t test. The relative levels of DNA cleavage by ResT in the presence of compounds 15 and 22 were 79% and 75%, respectively. Although these numbers are not statistically significant, the compounds still exhibited a reproducible 20 to 25% reduction in the cleavage of DNA by ResT. Only a 6% reduction in DNA cleavage by ResT was observed in the presence of compound 11, making this the compound that seemed to least affect the ability of ResT to cleave DNA.
To compare the effect that each inhibitor had on the DNA cleavage step relative to the second transesterification step (hairpin formation), the effect of each compound on the full telomere resolution reaction using an unmodified oligonucleotide substrate was determined. All of the molecules demonstrated 70% or more inhibition of telomere resolution by ResT on the oligonucleotide substrates (Fig. (Fig.5C),5C), and all gave results significantly different (>95%) from that for the positive control by a two-tailed Student t test. For all of the compounds, although cleavage was affected to various extents, the effect on the final product-forming step of the reaction was more pronounced. Since the ligation step was assessed indirectly here, we do not know whether the greater effect of the inhibitors on this step stems from direct inhibition of the reaction chemistry or from some other mechanism, such as inhibition of a conformational change required for the second chemical step. Similarly, a defect at the cleavage step might have resulted from a direct effect on the reaction chemistry or from the perturbation of a previous step, such as DNA binding and/or subsequent dimerization by ResT.
DNA binding by ResT is an important feature of the telomere resolution reaction. The region of the protein that recognizes and binds to a specific sequence in the replicated telomere substrate (TAGTA, or box 3) resides in the C-terminal catalytic domain, ResT164-449 (36, 37). Specific binding can be assessed using an electrophoretic mobility shift assay with purified ResT164-449 (but not with the wild-type protein, due to aggregation problems). The effect of each inhibitor on DNA binding was investigated using ResT164-449. As shown in Fig. Fig.5D,5D, inhibitors 5, 8, 14, 15, and 22 significantly reduced the ability of ResT164-449 to specifically bind telomeric DNA and induce a mobility shift, giving values of 16, 33, 27, 14, and 5% binding, respectively, relative to that for the positive control. In contrast, the specific DNA binding ability of ResT164-449 was increased over that for the positive control (+ResT164-449) by compound 11.
The inhibition of telomere resolution exhibited by five of the six inhibitory compounds may be due to the ability of the compounds to interfere with specific binding to the telomere substrate by the C-terminal region of the protein. However, it should be noted that the DNA binding assays were performed using the C-terminal domain ResT164-449, because the full-length protein is subject to aggregation and cannot be used in sequence-specific DNA binding assays. The abilities of the compounds to inhibit sequence-specific DNA binding by ResT164-449 may, therefore, not accurately reflect the binding properties of the full-length protein. Nonetheless, the ability of five of the six compounds to perturb substrate binding may be an important feature of the mechanism of inhibition. Moreover, assessment of the effects of the inhibitors on sequence-specific DNA binding by ResT164-449 has revealed an important difference between compound 11 and the other inhibitory compounds, as discussed below.
A comparison of the effects of the six inhibitors on telomere cleavage, resolution, and sequence-specific DNA binding (Fig. (Fig.6)6) showed both similarities and differences. In all cases, a more profound effect on telomere resolution than on cleavage was observed. This was not surprising, since the same active site promotes both chemical steps. Therefore, a compound that either binds directly to the active site or affects other steps in the reaction would be expected to display a cumulative effect through several steps of the reaction pathway. In contrast, comparison of the DNA-binding data with that of the chemical steps allowed organization of the six inhibitors into three distinct classes.
Class I contains compounds 8 and 14 (Fig. (Fig.6,6, top left). The level of sequence-specific DNA binding by ResT164-449 in the presence of either of these compounds roughly correlates with the level of telomere cleavage. This suggests that the level of inhibition in terms of telomere cleavage may be due to the inability of ResT to bind DNA.
Class II consists of compound 11 (Fig. (Fig.6,6, upper right), which exhibited near-control levels of telomere cleavage and higher-than-control levels of ResT164-449 DNA binding. Compound 11 may be targeting any number of steps after, but not before, telomere cleavage.
Finally, compounds 5, 15, and 22 were assigned to Class III (Fig. (Fig.6,6, bottom). This class of compounds is unique in that telomere binding by ResT164-449 was much more severely affected than telomere cleavage. One explanation for this could be that the greater inhibition of DNA binding observed with ResT164-449 may be due to the compound's ability to gain better access, or to bind to the C-terminal domain of ResT more strongly, in the absence of the N-terminal domain of the protein.
Lyme disease is the most common vector-borne disease in North America, with approximately 20,000 new cases reported every year (http://www.cdc.gov/ncidod/dvbid/lyme/ld_rptdLymeCasesbyState.htm). In the absence of a commercially available vaccine for human use, the prevention of Lyme disease relies solely on the avoidance of tick-infested areas, self-examination for ticks, and the use of tick repellents or acaricides. The telomere resolvase, ResT, is a promising target for the development of new, highly specific antiborrelial compounds for use in both the treatment and the prevention of Lyme disease. In this study we have developed a novel high-throughput assay to screen large libraries of compounds for inhibitors of ResT. From a library of 27,250 small molecules, we recovered six ResT inhibitors with IC50 values of 2 to 10 μM, and we investigated the points in the reaction pathway where inhibition occurred. The six inhibitors that we recovered were also tested by whole-cell assays at 10 μM but did not display any inhibition of the growth of infectious B. burgdorferi B31 clone 5A4 (29) or the high-passage-number laboratory strain B31-A (5) (data not shown). This is perhaps not surprising, considering the small and limited chemical library used in the screening process and the general difficulty of obtaining effective antibacterial drugs (26). The most likely explanation for the lack of efficacy in the whole-cell assay is that our inhibitors are not effectively taken up by B. burgdorferi or that they are actively exported from the cell. Options for further development of antiborrelial ResT inhibitors include further analysis and modification of the six inhibitory compounds through medicinal chemistry approaches or the screening of substantially larger small-molecule libraries. In view of the structural heterogeneity of the six compounds and their chemical characteristics, the latter approach would appear to hold the greatest promise. In summary, the high-throughput screening assay developed in this paper provides a starting point in the quest for future development of novel antiborrelial drugs as well as small-molecule inhibitors that will be helpful for further dissection of the reaction mechanism.
We thank Michael Surette for generously providing access to his liquid handling system and Eric Brown and Samir Roy for helpful advice and discussions.
This research was supported by the Canadian Institutes of Health Research (MOP-53086), the Canada Research Chairs Program, and the Alberta Heritage Fund for Medical Research. G.L. was supported by a Studentship award from the Canadian Institutes of Health Research and the Alberta Heritage Foundation for Medical Research. G.C. was supported by a Scientist award from the Alberta Heritage Fund for Medical Research and holds a Canada Research Chair in the Molecular Biology of Lyme Disease.
Published ahead of print on 13 July 2009.