Baculovirus encoding BACH1-WT, BACH1-P47A, BACH1-M299I, or BACH1-K52R with a C-terminal FLAG tag was used to infect High Five insect cells and the recombinant BACH1 protein was purified as previously described (9). At least three independent protein purifications of the purified wild-type and mutant recombinant BACH1 proteins were performed, yielding very similar results in catalytic ATPase and helicase assays, indicating uniformity in the determined specific activities of these proteins.

PAGE-purified oligonucleotides used for the preparation of DNA substrates were purchased from either Loftstrand Labs (Gaithersburg, MD) or Midland Certified Reagent Co. (Midland, TX) and are listed in Table 1. DNA duplex substrates were 5′- ³²P-end labeled and prepared as described previously (14). DNA substrates are shown in Table 2.

Helicase assay reaction mixtures (20 μl) contained 40 mM Tris–HCl (pH 7.6), 25 mM KCl, 5 mM MgCl₂, 2 mM dithiothreitol, 2% glycerol, 100 ng/μl BSA, 2 mM ATP, 10 fmol of the specified duplex DNA substrate (0.5 nM DNA substrate concentration), and the indicated concentrations of BACH1 helicase. Helicase reactions were initiated by the addition of BACH1 and then incubated at 30°C for 15 min unless otherwise indicated. Reactions were quenched in the presence of a 10-fold excess of unlabeled oligonucleotide with the same sequence as the labeled strand to prevent reannealing and products resolved on nondenaturing 12% (19:1 acrylamide: bisacrylamide) polyacrylamide gels and quantitated as previously described (13).

For helicase sequestration studies, BACH1 (4.8 nM) was preincubated with the indicated concentrations (0–25 nM) of the unlabeled forked duplex DNA molecule (competitor) in standard helicase reaction buffer as described above along with 2 mM ATP in a total reaction mixture volume of 20 μl for 3 min at 30°C. The reactions were initiated with 10 fmol of radiolabeled forked duplex molecules (Substrate 1) and incubated for 15 min at 30°C. Reactions were quenched and resolved on native polyacrylamide gels. Helicase data (% control) is expressed relative to the control reactions lacking the competitor DNA.

Protein–DNA binding mixtures (20 μl) contained the indicated concentrations of BACH1 and 0.5 nM of the specified ³²P-end-labeled DNA substrate in the same reaction buffer as that used for helicase assays (see above) containing 2 mM ATPγS. The binding mixtures were incubated at 24°C for 30 min after the addition of BACH1 and analyzed as previously described (13).

For k_cat determinations, ATPase assay reaction mixtures (25 μl) contained 40 mM Tris–HCl (pH 7.6), 25 mM KCl, 5 mM MgCl₂, 2 mM dithiothreitol, 2% glycerol, 100 ng/μl BSA, 82 nM forked duplex or single-stranded 45mer DNA effector concentration, 0.8 mM [³H]ATP, and 77 nM BACH-WT or 77 nM BACH1-M299I. Reactions were initiated by the addition of BACH1 and incubated at 30°C. Aliquots (5 μl) were removed at 15 min intervals (0–45 min) and evaluated by thin layer chromatography as described previously (15). Less than 20% of the substrate ATP was consumed in the reaction over the entire time course (45 min) of the experiments. For K_eff determinations, ATPase reaction mixtures (20 μl) contained the same concentrations of components as described above except forked duplex DNA effector concentrations ranged from 0 to 75 nM and the concentration of BACH1-WT or BACH1-M299I was 50 nM. K_eff values were determined using linear regression analysis from double reciprocal plots of ATP (pmol) hydrolyzed versus forked-duplex DNA effector concentration.

Fluorescence stopped-flow kinetic experiments were performed using an Applied Photophysics SX.18MV stopped-flow reaction analyzer. Hoechst 33258 dye interacted into the dsDNA substrate was excited at a wavelength of 344 nm, and the fluorescence emission was monitored at wavelengths >400 nm with a 400 nm long pass filter from Edmund Optics (Barrington, NJ). Experiments were carried out in the two-syringe mode where BACH1 and ATP were preincubated at 30°C in one syringe for 1 min, while the DNA substrate was preincubated with Hoechst 33258 dye (100 nM, Molecular Probes) at 30°C in the second syringe. Each syringe contained 40 mM Tris–HCl (pH 7.6), 25 mM KCl, 5 mM MgCl₂, 2 mM dithiothreitol, 2% glycerol, and 100 ng/μl BSA. Final concentrations of BACH1 (4.8 nM), ATP (2 mM), Hoechst 33258 dye (100 nM) and DNA fork substrate (0.8 nM) in the syringe were double that of the final concentration in the reaction. Equal volumes (60 μl) of sample from both syringes were mixed to initiate the reaction, which took place at 30°C. One thousand data points were collected from monitoring 20 μl of each kinetic time course reaction. For converting the output data from volts to percent fluorescence decrease, a time course with the same set up was performed except with only the unannealed oligonucleotides instead of the forked duplex substrate. Data were then normalized by defining the voltage obtained with the unannealed oligonucleotide as 100% fluorescence decrease. In control reactions, no fluorescence change was detected in reaction mixtures containing BACH1-WT helicase and DNA substrate with intercalated Hoechst 33258 dye but lacking ATP. In addition, no fluorescence change was observed when the BACH1-K52R ATPase-helicase dead mutant protein was incubated with ATP and the DNA substrate, indicating that the signal detected was a consequence of BACH1 helicase activity that is dependent on ATP hydrolysis.

To evaluate the biochemical effects of two helicase domain BACH1 missense polymorphisms (P47A, M299I) on enzymatic function, we compared their unwinding activities with wild-type BACH1 on a preferred substrate, a forked duplex DNA structure (Table 2, Substrate 1). BACH1-P47A weakly unwound the forked duplex in an ATP- and protein concentration-dependent manner (Figure 1A, lanes 2 and 3; Figure 1B), whereas BACH1-K52R that has an amino acid substitution of the invariant lysine in motif I completely failed to unwind the forked duplex substrate (Figure 1B). In stark contrast, BACH1-M299I efficiently unwound the DNA substrate (Figure 1A, lanes 4 and 5, Figure 1B). Quantitative analyses demonstrated that BACH1-M299I helicase activity was ∼1.8-fold greater than BACH1-WT at an enzyme concentration of 0.3 nM (Figure 1B). At least three independent protein purifications of the purified wild-type and mutant recombinant BACH1 proteins were performed, and very similar results in the helicase measurements were obtained, indicating that the differences in helicase activity observed were reproducible.

Preferential unwinding of the forked duplex by BACH1 reflects its ability to bind the DNA substrate more efficiently compared to a simple 5′ tailed duplex (13). Gel-shift assays demonstrated that a greater fraction of the forked duplex is bound by BACH1-M299I compared to BACH1-WT throughout the protein concentration range used for the helicase assays (Supplementary Figure 1). The maximal difference in DNA binding between the mutant and wild-type proteins was observed at the highest protein concentration tested, 4.8 nM, in which ∼40% of the forked duplex was bound by BACH1-M299I compared to 15% substrate bound by BACH1-WT. BACH1-P47A, however, displayed similar binding compared to BACH1-WT (data not shown). The wild-type and variant enzymes were tested for DNA-dependent ATPase activity using the forked duplex as the DNA effector at a concentration of ATP (0.8 mM) in which <20% of the nucleotide was hydrolyzed during the entire time course of the reaction. Both BACH1-WT and BACH1-M299I displayed a similar dependence on the forked duplex for ATPase activity as evidenced by their K_eff values in the same range (Table 3); however, the turnover rate constant (k_cat) was 6.7-fold greater for BACH1-M299I, indicating that the variant hydrolyzed ATP substantially better than the wild-type enzyme (Table 3). BACH1-P47A catalyzed ATP hydrolysis poorly as evidenced by its low k_cat, 12-fold less than BACH1-WT (Table 3). Similar results from ATPase assays for the wild-type and mutant (P47A, M299I) BACH1 proteins were obtained at an ATP concentration of 2 mM and with several different preparations of the mutant and wild-type BACH1 proteins (data not shown). The k_cat values for ATP hydrolysis using the single-stranded 45mer oligonucleotide DC26 (Table 1) were 28.4 ± 4.8 per minute and 218 ± 2.1 per minute for BACH1-WT and BACH1-M299I, respectively, indicating that the differences in ATPase activity between BACH1-WT and BACH1-M299I proteins were observed for both forked duplex and ssDNA molecules. These results suggest that the increased helicase activity of BACH1-M299I relative to BACH1-WT is accompanied by better binding to the DNA substrate and significantly improved efficiency for ATP hydrolysis in the presence of the forked duplex. Although BACH1-M299I unwound the forked duplex better than BACH1-WT, the difference was not nearly as great as the observed effects on ATP hydrolysis. BACH1-P47A is severely compromised in both its ATPase and helicase activity; however, a low residual ATP-dependent unwinding activity exists.

Since little is known about the BACH1 helicase mechanism, we investigated the importance of structural and chemical features of the DNA duplex for unwinding. The forked duplex tested for BACH1 helicase activity contained either a polyglycol modification to the sugar phosphate backbone or a set of three adjacent abasic sites positioned 16 bp from the fork entry site on either the translocating (top) or non-translocating (bottom) strand that the enzyme is presumed to translocate based on its demonstrated 5′ to 3′ directionality of unwinding (9,13). These two types of structural perturbations to the DNA double helix enabled us to assess the relative importance of backbone integrity or base contacts for BACH1 as it tracks through the duplex and catalytically separates the two strands.

BACH1-WT was profoundly sensitive to the polyglycol backbone modification in either the translocating or non-translocating strands of the duplex. No unwinding of the backbone modified DNA substrates by BACH1-WT (up to 20 nM) could be detected in the 15 min reaction (Figure 2) or 1 h incubation (data not shown). Thus BACH1-WT helicase activity does not display strand-specific inhibition with respect to the strand that it translocates on as observed for other helicases [vaccina virus NPH-II (16), Escherichia coli Dda, bacteriophage (17,18), T7 gene 4 (19)], suggesting that BACH1 has a unique ability to sense both strands of the DNA duplex during the unwinding reaction. Indeed, we found that unwinding of the same forked duplex substrates by human RECQ1, a 3′ to 5′ helicase, is inhibited in a strand-specific manner such that the polyglycol modification in only the translocating (bottom) strand inhibited unwinding (manuscript in preparation). It should be pointed out, however, that studies of E.coli DnaB (SF4) (20) and hepatitis C virus NS3 (SF2) (21) helicases have suggested that transient interactions of the helicase with the displaced strand are important as well.

We next examined the ability of BACH1-M299I to unwind the backbone modified DNA substrates. For the non-translocating strand modification, BACH1-M299I unwinding was reduced compared to the helicase activity on the control (unmodified) substrate throughout the protein titration range; however, the M299I mutant retained the ability to unwind the DNA substrate with the non-translocating strand modification in a protein concentration dependent manner, exhibiting up to 40% substrate unwound at 19 nM M299I helicase (Figure 2). The translocating strand modification in the backbone further inhibited BACH1-M299I helicase; only 5% of the backbone modified substrate was unwound at 2.5 nM helicase concentration compared to ∼50% of the control substrate. At the highest helicase concentration tested (19 nM), BACH1-M299I unwound ∼10% of the DNA substrate with the backbone modification in the translocating strand. Nearly complete inhibition was observed for the substrate with backbone modifications in both strands for either BACH1-WT or BACH1-M299I at all protein concentrations tested (data not shown). These results indicate that both the BACH1-WT and BACH1-M299I proteins are sensitive to backbone modifications in either the non-translocating or translocating strands of the DNA substrate; however, the M299I mutant, unlike BACH1-WT, retains the ability to unwind the backbone modified DNA substrates to different extents, depending on which strand the modification is placed, throughout the helicase titration range.

We also tested the ability of BACH1 to unwind a DNA substrate containing three adjacent abasic sites in either the top or bottom strand of the duplex at the same position as the polyglycol linkage. BACH1-WT retained the ability to unwind the forked duplex substrate with the abasic sites in the non-translocating strand similar to the control substrate (Figure 3). The presence of the abasic sites in the translocating strand of the DNA substrate resulted in a 1.6-fold reduction of BACH1-WT helicase activity. The presence of abasic sites in both strands resulted in almost complete inhibition of BACH1-WT helicase activity. Thus BACH1-WT is significantly more sensitive to a perturbation of sugar phosphate backbone continuity than the absence of bases in the duplex region, suggesting that electrostatic interactions of BACH1 helicase with the sugar phosphate backbone are more important than base-stacking interactions. In contrast to wild-type BACH1, the M299I variant was not inhibited to any appreciable extent by the presence of abasic sites in the non-translocating or translocating strands of the DNA substrate. However, the presence of abasic sites in both strands reduced BACH1-M299I helicase activity 2-fold compared to the control substrate. Collectively, BACH1-M299I has a greater tolerance for abasic sites or backbone modifications nested within the duplex in either the translocating or non-translocating strands of the DNA substrate compared to BACH1-WT.

The complete failure of wild-type BACH1 enzyme to unwind the forked duplex substrates with the backbone modification in either strand of the duplex raised the possibility that the polyglycol moiety perturbed the ability of the protein to interact with or bind the DNA substrate. To address this, BACH1-WT was tested for DNA-dependent ATPase activity using the unmodified or backbone modified DNA molecules as effectors. BACH1-WT hydrolyzed ATP with a similar efficiency for each of the DNA effectors (Table 4). BACH1-M299I was also found to hydrolyze ATP similarly and significantly better than the wild-type enzyme (Table 4). A quantitative comparison of the k_cat values for ATP hydrolysis using the various modified or unmodified DNA effects demonstrated a difference in specificity ATPase activity of 5- to 6.5-fold. BACH1-WT and BACH1-M299I were found to bind the modified duplex molecules (data not shown), consistent with the fact that BACH1 is a DNA-dependent ATPase (9).

Certain helicases have been found to be sequestered by DNA molecules harboring various types of covalent lesions (22). Pre-incubation of BACH1 (4.8 nM) with increasing concentrations of unlabeled forked duplex molecules inhibited unwinding of the radiolabeled tracker DNA substrate that was added subsequently to the reaction mixtures. Pre-incubation of BACH1 (4.8 nM) with forked duplexes containing the backbone modification in either the translocating or non-translocating strands inhibited BACH1 unwinding of the tracker DNA substrate to a significantly greater extent than the unmodified forked duplex (Figure 4). These results suggest that BACH1 was preferentially sequestered by the backbone modification irrespective of strand status.

Since the strand displacement assay conventionally employed to monitor helicase activity does not measure partial unwinding, we employed a dye displacement assay (Figure 5A) to address the hypothesis that BACH1-WT unwound the duplex (16 bp) adjacent to the fork and became trapped at the vicinity of the backbone modification, leaving the remaining duplex (12 bp) intact. Incubation of BACH1-WT (4.6 nM final concentration) with the unmodified forked duplex (0.8 nM final concentration) in the presence of ATP (2 mM) under multiple turnover conditions resulted in a time-dependent fluorescence decrease during the 200 sec interval (Figure 5B). In the absence of ATP, no fluorescence change was detected (data not shown). In addition, no fluorescence change was observed when the BACH1-K52R ATPase dead mutant protein was incubated with ATP and the DNA substrate (data not shown), indicating that the signal detected was a consequence of BACH1 helicase activity that is dependent on ATP hydrolysis. As observed for the unmodified substrate, incubation of BACH1-WT with ATP and the DNA substrate containing a backbone modification in either the translocating or non-translocating strands resulted in a fluorescence change that was dependent on time (Figure 5B), indicating that the fluorescent dye is displaced due to BACH1 helicase activity. However, the percent fluorescence decrease for the assays with the modified substrates was less than that observed for the unmodified forked duplex, indicating that BACH1 partially unwinds the duplex region between the fork entry site and the backbone modification, and becomes trapped by the backbone perturbation.

Previously we reported that BACH1 helicase activity on 5′ tailed or forked duplex substrates increased proportionately as the 5′ ssDNA tail was increased from 15 to 35 nt (13), raising the possibility that features of the DNA substrate that influence enzyme loading may affect the unwinding event. This prompted us to study the effect of the backbone modifications on BACH1 unwinding of DNA substrates in which the 5′ tail was increased to the optimal 35 nt compared to the 19 nt 5′ tail in the original set of substrates. As shown in Figure 6A, backbone modifications to the top or bottom stand of the forked duplex with a 35 nt 5′ tail did not have a significant effect on BACH1 helicase activity compared to the control substrate. Polyglycol modifications to both strands reduced BACH1 helicase activity to ∼ 45% that of the control. Similar observations for this series of backbone modified substrates were observed at lower BACH1 protein concentrations in which the percentage of control DNA substrate unwound was comparable to that observed for the forked duplex with a 19 nt 5′ tail (data not shown). For the DNA substrates with only a 35 nt 5′ tail (Figure 6B), BACH1 helicase activity was also unaffected by the non-translocating strand modification, and only reduced by 20% that of the control for the substrate with the translocating strand modification. Backbone modifications in both strands inhibited BACH1 helicase activity on the 35 nt 5′ tailed substrate to a level ∼40% that of the control. These results indicate that BACH1 helicase inhibition by the polyglycol backbone modification in either strand of the duplex can be successfully overcome by increasing the length of the 5′ ssDNA tail the enzyme used for loading on the partially duplex substrate.