A structural feature of Dda helicase which enhances displacement of streptavidin and trp repressor from DNA

Abstract

Helicases are molecular motors with many activities. They use the energy from ATP hydrolysis to unwind double‐stranded nucleic acids while translocating on the single‐stranded DNA. In addition to unwinding, many helicases are able to remove proteins from nucleic acids. Bacteriophage T4 Dda is able to displace a variety of DNA binding proteins and streptavidin bound to biotinylated oligonucleotides. We have identified a subdomain of Dda that when deleted, results in a protein variant that has nearly wild type activity for unwinding double‐stranded DNA but exhibits greatly reduced streptavidin displacement activity. Interestingly, this domain has little effect on displacement of either gp32 or BamHI bound to DNA but does affect displacement of trp repressor from DNA. With this variant, we have identified residues which enhance displacement of some proteins from DNA.

1. INTRODUCTION

Cellular DNA is not a naked molecule but is coated with proteins including histones and nonhistone DNA binding proteins. These proteins can be an obstacle to processes which require access to the DNA such as replication and transcription. In addition to unwinding duplex DNA or RNA, helicases remove proteins from the nucleic acid or reposition proteins such as histones on the DNA. ¹ This protein displacement activity not only allows access to the DNA by complexes such as the replisome, it also allows helicases to regulate the activity of DNA binding proteins. Saccharomyces cerevisiae Pif1 removes telomerase from both telomeres and double‐stranded DNA breaks ² , ³ , ⁴ , ⁵ and the related Rrm3 is required for replication through non‐nucleosomal protein–DNA complexes. ⁶ , ⁷ Similarly, E. coli Rep displaces protein complexes in the path of the replication fork. ⁸ Human RECQL5 displaces RAD51, and human FANCJ and RECQ1 remove TRF1 and TRF2 from telomeric DNA. ⁹ , ¹⁰

Dda is a superfamily 1B (SF1B) helicase from bacteriophage T4 ¹¹ which is efficient at displacing proteins from DNA. Dda is monomeric in solution ¹² and unwinds DNA as a monomer ¹³ ; however, as the length of the single‐stranded DNA (ssDNA) region of the substrate increases, multiple Dda monomers can function together resulting in increased processivity for unwinding of the duplex through functional cooperativity. ¹⁴ Dda displaces the E. coli RNA polymerase allowing T4 replisome to pass. ¹⁵ Unwinding of duplex DNA by Dda is not inhibited by the E. coli replication terminator protein (Tus) bound to Ter sites in either the permissive or nonpermissive orientation ¹⁶ or bound lac repressor. ¹⁷ Unwinding of double‐stranded DNA (dsDNA) with bound E. coli trp repressor is limited by the rate of trp repressor displacement. ¹⁸ Displacement of streptavidin from biotinylated oligonucleotides is frequently used as a model for protein displacement reactions by helicases. ¹⁹ , ²⁰ , ²¹ , ²² , ²³ , ²⁴ , ²⁵ , ²⁶ , ²⁷ , ²⁸ , ²⁹ Monomeric Dda can displace streptavidin, but the reaction increases by more than an order of magnitude when two Dda molecules can bind to the substrate and by 1,000‐fold when five or more Dda molecules are bound. ²⁰

S. cerevisiae Pif1 is a SF1B helicase closely related to Dda. ³⁰ Like Dda, Pif1 displaces proteins bound to DNA. Pif1 displaces a dCas9‐R‐loop block to allow CMG‐Polε to continue replication. ³¹ Similarly, Pif1 alleviates inhibition of Polδ by the telomeric binding protein Rap1, the transcription factor Tbf1, and a nucleosome by removing the bound protein. ³² , ³³ However, streptavidin displacement by Pif1 is much less efficient and rapid than streptavidin displacement by Dda. ²⁰ , ²⁴ The structure of Dda contains a β‐hairpin in the SH3 domain (Figure 1, cyan) referred to as the hook (Figure 1, purple) which points toward the incoming DNA ³⁴ that is absent in Pif1 ³⁵ and other related helicases (Figure S1). The location of the hook suggests that it could be involved in both strand separation and protein displacement. We investigated the role of this region directly by removing six amino acids at the tip of the hook (Tyr279–Lys284) to decrease the length of the hairpin.

FIGURE 1.

Structures of Dda and Pif1 bound to ssDNA. (a) Dda (PDB code 3UPU) ³⁴ structure is shown with the RecA‐like domains in green (domain 1A) and gray (domain 2A) and the SH3 domain (domain 2A) in cyan. The pin (domain 1B) is shown in red and DNA is in yellow. The hook region of the SH3 domain is highlighted in purple. (b) S. cerevisiae Pif1 (PDB code 5O6B) ³⁵ structure is shown with domains colored as in (a). The N‐terminal domain is orange

2. RESULTS

2.1. Deletion of the hook does not affect affinity for DNA or the rate of ATP hydrolysis

The structure of Dda reveals two prominent β‐hairpins near the DNA binding site: domain 1B, also called the pin (red, Figure 1) and one within domain 2B (cyan, Figure 1) referred to as the hook (purple and cyan, Figure 1). The pin has already been shown to be necessary for DNA unwinding. ³⁴ The function of the hook is unknown, although its position in the vicinity of the DNA entering the active site suggests it could be involved in DNA unwinding or displacement of proteins bound to the DNA. Deletion of the tip of the hook (amino acids 279–284, purple in Figure 1) does not affect the ability of Dda to bind to DNA (Figure 2a). The maximal rate of ATP hydrolysis is also unchanged, although the concentration of DNA required to achieve the maximal ATPase rate increased twofold with deletion of the hook (Figure 2b). Δ279–284 Dda retained full DNA binding and ATPase activities.

FIGURE 2.

DNA binding and ATPase activity are not affected by deletion of the hook. (a) Fluorescence anisotropy of 1 nM 5′‐FAM‐T₇16bp in the presence of varying concentrations of wtDda (red) and Δ279–284Dda (blue). Data were fit to the quadratic equation to obtain the K _Ds of 41 ± 3 nM and 44 ± 5 nM for wtDda and Δ279–284Dda, respectively. Error bars are the SD of three independent experiments. (b) ATP hydrolysis rates of wtDda (red) and Δ279–284Dda (blue) were measured in the presence of increasing concentrations of calf thymus DNA. Data were fit to a hyperbola to determine K _act and k _cat where K _act is the concentration of calf thymus DNA required to obtain half the maximal ATPase rate (k _cat). K _act values were 0.35 ± 0.03 μg/ml and 0.83 ± 0.12 μg/ml for wtDda and Δ279–284Dda, respectively, and k _cat values were 249 ± 3 s⁻¹ and 239 ± 6 s⁻¹ for wtDda and Δ279–284Dda, respectively. Errors are the SE of the fit

2.2. Hook deletion affects the processivity of unwinding

The ability of Δ279–284 Dda to unwind duplex DNA was measured using a partial duplex substrate that can accommodate two Dda monomers. ¹⁴ Although the processivity of the hook deletion mutant is decreased relative to wild type Dda as seen by the drop in the amplitude of the product formation plot and an increase in the dissociation rate constant, the rate constant for unwinding is similar (Figure 3b,c). Unwinding of a shorter duplex by a single Δ279–284 Dda monomer also occurs at a similar rate as wild type Dda, although the processivity is again decreased and the dissociation rate constant is increased (Figure 3d). The rate constants for unwinding are 1.2–1.3‐fold lower for the Δ279–284 Dda, but the dissociation rate constants are increased 1.7–2‐fold. The increase in the dissociation rate constant was confirmed by monitoring the change in Dda tryptophan fluorescence upon dissociation from the DNA (Figure S2). This indicates that the hook plays a minor role in unwinding.

FIGURE 3.

Unwinding of duplex DNA is moderately affected by deletion of the hook. (a) Scheme describing an n‐step sequential mechanism for DNA unwinding. At the beginning of the reaction, enzyme (E) is bound to substrate to form the enzyme substrate complex (ES). The substrate is unwound in n sequential steps with the rate constant k _u through the intermediates I ₁ … I _n−1 to form ssDNA (P). At each step, the enzyme can dissociate with the rate constant k _d . (b) Unwinding reaction products of 5 nM T₁₄20bp by 100 nM Dda are separated by PAGE. (c) Unwinding of 5 nM T₁₄20bp by 100 nM wtDda (red) or Δ279–284Dda (blue). Data were fit to a four‐step sequential mechanism (a) using KinTek Explorer to obtain the unwinding rate constants of 68.3 ± 9.7 s⁻¹ and 54.4 ± 0.9 s⁻¹ and dissociation rate constants of 6.4 ± 1.8 and 10.8 ± 2.9 s⁻¹ for wtDda and Δ279–284Dda, respectively. (d) Unwinding of 5 nM T₇16bp by 100 nM wtDda (red) or Δ279–284Dda (blue). Data were fit to a three‐step sequential mechanism (a) using KinTek Explorer ³⁶ to obtain the unwinding rate constants of 84.6 ± 5.3 s⁻¹ and 67.6 ± 2.4 s⁻¹ and dissociation rate constants of 17.0 ± 2.2 and 33.4 ± 1.0 s⁻¹ for wtDda and Δ279–284Dda, respectively. Errors are the SD of three independent experiments

2.3. Streptavidin displacement is inhibited by deletion of the hook subdomain

Displacement of streptavidin from 3′‐biotinylated oligonucleotides by Dda has been well studied so this was used as a model protein displacement reaction to determine the effect of deletion of the hook (Figure 4a). Comparison of the rates of streptavidin displacement by wild type and hook deletion Dda (Figure 4) reveals a 19‐fold reduction in the rate constant for streptavidin displacement from a 3′‐biotinylated 30mer to which multiple Dda molecules can bind (Figure 4b,c). Interestingly, the rate constant is reduced only sevenfold for the Δ279–284 Dda for displacement of streptavidin from a 3′‐biotinylated 9mer which can accommodate only a single Dda molecule (Figure 4d).

FIGURE 4.

Streptavidin displacement is severely impaired by deletion of the hook. (a) Streptavidin (blue) is displaced from 3′‐biotinylated DNA by Dda upon addition of ATP, and a biotin trap (pink). ²⁸ (b) Gel image showing displacement of streptavidin from 10 nM 3′‐bioT₃₀ by wtDda or Δ279–284Dda. Data are plotted in (c) with 250 nM Dda in filled shapes and 500 nM in open shapes. Data were fit with a single exponential function. The rate constants were 0.21 ± 0.04 s⁻¹, 0.011 ± 0.003 s⁻¹, 0.20 s⁻¹ and 0.013 s⁻¹ by 250 nM wtDda, 250 nM Δ279–284Dda, 500 nM wtDda, and 500 nM Δ279–284Dda, respectively. Errors represent the SD of three independent experiments. (d) Displacement of streptavidin from 10 nM 3′‐bioT₉ by 250 nM (filled shapes) or 500 nM (open shapes) wtDda or Δ279–284Dda. Data were fit with a single exponential function. The rate constants were 0.13 ± 0.02 min⁻¹, 0.018 ± 0.003 min⁻¹, 0.12 min⁻¹, and 0.015 min⁻¹ by 250 nM wtDda, 250 nM Δ279–284Dda, 500 nM wtDda, and 500 nM Δ279–284Dda, respectively. Errors represent the SD of three independent experiments

Streptavidin displacement by Dda is highly cooperative ²⁰ so the greater effect of deletion of the hook for streptavidin displacement from a 30mer relative to a 9mer suggested that the Δ279–284 Dda variant may have reduced cooperativity relative to wild type Dda. The cooperativity of both wild type and Δ279–284 Dda were measured by determining the rate of streptavidin displacement from a 3′‐biotinylated 15mer with varying concentrations of Dda (Figure 5). The 15mer can accommodate two Dda molecules and cooperativity of Dda on this substrate has been well characterized. ²⁰ Displacement by wild type Dda (Figure 5c) is much faster than displacement by Δ279–284 Dda (Figure 5d). The rate constants for displacement are plotted in Figure 5e and fit with the Hill equation. The same data is plotted on a different scale in Figure 5f so that the sigmoidal nature of the curve for Δ279–284 Dda can be discerned. The Hill coefficients are 2.9 ± 0.5 and 2.4 ± 0.6 for wild type Dda and Δ279–284 Dda, indicating that deletion of the hook does not significantly affect the cooperativity of Dda.

FIGURE 5.

Cooperativity for streptavidin displacement is not affected by deletion of the hook. (a) Streptavidin (blue) is displaced from 3′‐BioT₁₅ by Dda upon addition of ATP and a biotin trap (pink). (b) Gel image showing displacement of streptavidin from 10 nM 3′‐BioT₁₅ by 500 nM wtDda or Δ279–284Dda. Data are plotted in (c) for 40 nM (red circles), 60 nM (blue squares), 80 nM (green diamonds), 120 nM (black triangles), 180 nM (orange open circles), 240 nM (purple open squares), 300 nM (cyan open diamonds), and 500 nM (pink x) wtDda and fit with a single exponential function. Data are plotted in (d) for 40 nM (red circles), 80 nM (blue squares), 120 nM (green diamonds), 150 nM (black triangles), 300 nM (orange open circles), 400 nM (purple open squares), 500 nM (cyan open diamonds), and 650 nM (pink x) Δ279–284Dda and fit with a single exponential function. The rate constants for displacement are replotted in (e) and zoomed in (f). Data are fit with the Hill equation. The Hill coefficients are 2.9 ± 0.5 and 2.4 ± 0.6 for wtDda and Δ279–284Dda, respectively

2.4. Deletion of the hook has varied effects on displacement of DNA binding proteins

Since deletion of the hook resulted in a dramatic reduction in the ability of Dda to displace streptavidin from biotinylated oligonucleotides, this suggested the hook may be important for displacement of proteins bound to DNA. Gp32 is a ssDNA binding protein from bacteriophage T4. Displacement of gp32 labeled with 5‐acetamidofluorescein (gp32‐Fl) can be measured by the increase in fluorescence upon removal of gp32 from the DNA. ³⁷ Displacement of gp32‐Fl from M13 ssDNA by wild type and Δ279–284 Dda was measured (Figure 6), and the rate constants for displacement were nearly identical. This indicates that the hook is not involved in displacement of gp32 from ssDNA by Dda.

FIGURE 6.

Deletion of the hook has little effect on gp32 displacement. (a) Illustration of the reaction. In the presence ATP, Dda can displace gp32 from M13 ssDNA. The fluorescence of gp32‐Fl increases upon dissociation from ssDNA. ³⁷ (b) Displacement of 25 nM gp32‐Fl from 5 μM M13 ssDNA in nucleotides by 25 nM Dda was measured. Data were fit to a single exponential plus a slope. The rate constants were 0.15 ± 0.01 and 0.0064 ± 0.0008 s⁻¹ for wtDda and 0.16 ± 0.01 and 0.0065 ± 0.0013 s⁻¹ for Δ279–284Dda

The effect of deletion of the hook on displacement of a duplex DNA binding protein was determined by measuring displacement of catalytically dead BamHI‐E111A from duplex DNA. BamHI‐E111A has been shown to bind tightly to a duplex containing its recognition site and inhibit DNA unwinding by FANCJ and RECQ1. ¹⁰ The effect of deletion of the hook on unwinding duplex DNA with BamHI‐E111A bound was measured on a substrate that can accommodate a single Dda molecule (Figure 7a–c). The rate constants for displacement by wild type and Δ279–284 Dda were similar, indicating that deletion of the hook does not affect displacement of BamHI‐E111A from double‐stranded DNA (dsDNA).

FIGURE 7.

Deletion of the hook has more effect on trpR displacement than BamHI displacement. (a) Illustration of BamHI displacement reaction. In the presence ATP, Dda can displace catalytically dead BamHI‐E111A from dsDNA containing the BamHI recognition site. (b) Displacement of 200 nM BamHI‐E111A from 5 nM T₇16bp DNA by 100 nM Dda was measured. (c) Data were fit to a single exponential. The rate constants were 2.2 ± 0.6 s⁻¹ for wtDda and 1.0 ± 0.2 s⁻¹ for Δ279–284Dda. (d) Illustration of trpR displacement reaction. Dda can displace trpR from dsDNA containing the E. coli trpEDCBA operator. (e) Displacement of 5 μM trpR from 100 nM T₈30bp DNA by 750 nM Dda was measured. (f) Data were fit to a sum of two exponentials. The rate constants were 3.3 ± 1.2 and 0.31 ± 0.14 s⁻¹ for wtDda and 2.9 ± 1.0 and 0.13 ± 0.12 s⁻¹ for Δ279–284Dda. Errors represent the SD of three independent experiments

Dda has previously been shown to displace E. coli trp repressor (trpR) from duplex DNA. ¹⁸ The effect of deletion of the hook on displacement of trpR was determined on a substrate that can accommodate a single Dda molecule (Figure 7d–f). The rate constants for trpR displacement by wild type Dda was twofold faster than displacement by Δ279–284 Dda. This suggests that deletion of the hook reduces the displacement of trpR from double‐stranded DNA (dsDNA).

3. DISCUSSION

Unlike other SF1B helicases such as RecD2 and Pif1, Dda contains a β‐hairpin (Leu275–Phe291) insertion in the SH3 domain called the hook (Figure 1). ³⁴ , ³⁸ , ³⁹ Here we deleted six amino acids at the tip of the hook (Y279–K284) and measured the effect on a variety of reactions catalyzed by Dda. The Δ279–284 Dda variant had little effect on DNA binding or ATP hydrolysis (Figure 2). The effect on dsDNA unwinding was also minimal with a similar unwinding rate but reduced processivity relative to wild type Dda (Figure 3). The reduction in processivity is due to a small increase in the off‐rate (Figures 3 and S2). The hook does not contain the helicase motifs required for interacting with DNA and deletion of the hook did not affect the affinity for DNA (Figure 2), suggesting that the hook does not directly affect affinity for DNA. However, it is possible that when Dda encounters an obstacle in its translocation path, the off‐rate may be increased when the hook is not present.

Displacement of streptavidin from 3′‐biotinylated oligonucleotides is significantly reduced by Δ279–284, which suggested the hook may function in protein displacement (Figure 4). However, streptavidin displacement by Dda is highly cooperative, ²⁰ and the effect of deleting the tip of the hook is reduced on the 9mer relative to the 30mer (7‐fold vs. 19‐fold). We measured the cooperativity for streptavidin displacement for wild type and Δ279–284 (Figure 5) and found that they are similar. This indicates that the hook enhances streptavidin displacement but has minimal effect on cooperativity. The Hill coefficients for streptavidin displacement by wild type and Δ279–284 Dda are 2.9 and 2.4, respectively (Figure 5). Since only two Dda molecules can bind to the substrate, ²⁰ , ³⁴ this results in a Hill coefficient greater than the number of possible enzymes involved. However, this assay measures enzyme activity, and that increases greater than 10‐fold on a substrate which can bind two Dda molecules (15‐mer) relative to a substrate that can bind one Dda molecule (9‐mer). Since two Dda molecules displace streptavidin much faster than a single Dda molecule, this likely results in a Hill coefficient greater than 2 when increasing the Dda concentration from sub‐saturating to the level that both binding sites are occupied.

Interestingly, removal of the hook has little effect on displacement of two different DNA binding proteins but did reduce displacement of a third DNA binding protein. Displacement of gp32, the T4 single‐stranded binding protein by wild type and Δ279–284 Dda are similar (Figure 6). Due to the specific interaction between Dda and gp32, ⁴⁰ we also compared the ability of wild type and Δ279–284 Dda to displace another DNA binding protein. Catalytically dead BamHI‐E111A is a duplex DNA binding protein which has been characterized previously in protein displacement reactions. ¹⁰ As with gp32, there was a minimal effect of Δ279–284 on displacement of BamHI‐E111A (Figure 7c). However, displacement of trpR, another duplex DNA binding protein that Dda has previously been shown to displace ¹⁸ is reduced with Δ279–284 Dda (Figure 7f). The magnitude of the effect is greatest for displacement of streptavidin from biotinylated oligonucleotides.

Wild type Dda has an extraordinary ability to displace streptavidin from biotinylated oligonucleotides. The maximal streptavidin displacement rates reported at saturating enzyme concentrations are hundreds of times faster than for other helicases for which similar measurements have been recorded (Table 1). Dda displaces streptavidin much faster than both superfamily 1 and 2 helicases such as Pif1, NS3, and XPD and hexameric helicases such as gp41 and T antigen. However, streptavidin displacement by Δ279–284 Dda occurs at a rate that is similar to that of the other helicase for which the streptavidin displacement rate has been recorded.

TABLE 1.

Rate constants for streptavidin displacement from a biotinylated 60mer at saturating enzyme concentration

Helicase	Organism	Rate constant for streptavidin displacement (min−1)	Fold reduction in displacement rate relative to Dda
Dda	Bacteriophage T4	474 28
Δ279–284 Dda	Bacteriophage T4	0.78 a	610
Pif1	S. cerevisiae	0.53 24	890
NS3	Hepatitis C	0.11 29	4,300
XPD	T. acidophilum	1.26 a , 23	380
gp41	Bacteriophage T4	0.17 28	2,800
T antigen	SV40	0.56 29	850

^a Displacement from a biotinylated 30 mer.

Wild type Dda is also a powerful motor for displacement of DNA binding proteins. Dda displaces E. coli RNA polymerase bound to a promoter, allowing the T4 replication complex to proceed, but also destabilizes transcription complexes in the absence of replication. ¹⁵ , ⁴¹ Dda is able to displace the E. coli replication terminator protein, Tus, from Ter sites in both orientations. ¹⁶ The other helicases which were studied, T antigen, E. coli helicase II (UvrD) and E. coli DnaB were only able to remove Tus in one orientation which is more expected based on the ability of a replisome that approaches a Tus‐Ter complex from the permissive face to pass but one that approaches from the non‐permissive face is blocked. Notably, wild type Dda displaces streptavidin 850‐fold faster than T antigen (Table 1).

Deletion of the hook increases the off‐rate of the enzyme (Figures 3 and S2). This could result in reduced streptavidin displacement for the hook deletion variant compared to the wild type enzyme. Only ~10% of the streptavidin is displaced from a 15‐mer in a single binding event by wild type Dda. ²⁰ Thus, multiple binding and dissociation events are likely involved in displacing streptavidin from the 15‐mer. Since the hook deletion variant dissociates from the substrate faster than the wild type enzyme, it will have a reduced opportunity to displace streptavidin in each binding event. For unwinding, which is rapid, this results in a decrease in the quantity of product formed in a single turnover. For streptavidin displacement, which is much slower, the increased off rate increases the chance that the hook deletion variant dissociates from the substrate before displacing streptavidin. Thus, the rate of streptavidin displacement is significantly reduced.

The difference in effect of Δ279–284 on streptavidin displacement relative to displacement of DNA binding proteins may be due to the difference in the interaction of the proteins with the DNA. The 3′‐biotin label to which streptavidin binds has an extended linker that results in streptavidin bound near the DNA. However, gp32, BamHI, and trpR bind directly to the DNA. BamHI binds as a dimer with one subunit bound to each strand of the DNA duplex (Figure 8a) while trpR binds with both subunits on one side of the DNA (Figure 8b). ⁴² , ⁴³ Since BamHI wraps around the DNA strands, it may occlude access to the DNA by the hook on Dda. TrpR rests on top of the DNA instead of wrapping around it so it may be easier for the hook to access the DNA. In the case of streptavidin, the DNA may be more accessible to the hook since the biotin is attached to the DNA with a long linker. This could allow the hook to slide underneath the protein and help to pry it off as Dda translocates, functioning like a molecular snowplow. It may be more difficult for the hook to access the DNA in the case of proteins which wrap around the DNA like BamHI and access may be more intermediate for proteins like trpR that directly contact with DNA but do not encircle the DNA. The structure of gp32 does not have well resolved DNA, but the ssDNA binds in a cleft (Figure 8c) ⁴⁴ that would likely make it difficult for the hook to act as a wedge to remove gp32 from the DNA. The minimal effect of the Dda hook deletion variant on gp32 displacement is consistent with this idea.

FIGURE 8.

DNA bound by BamHI and gp32 is less accessible than DNA bound by trpR. (a) A dimer of Bacillus amyloliquefaciens BamHI (green) wraps around the duplex DNA (yellow and orange). PDB ID: 1BHM ⁴² (b) A dimer of E. coli trp repressor (trpR; blue) sits on top of the duplex DNA (yellow and orange). PDB ID: 1TRO. ⁴³ (c) A monomer of bacteriophage T4 gp32 (purple) with the DNA binding site marked. PDB ID: 1GPC ⁴⁴

Although the processes of displacement of streptavidin vs. DNA binding proteins show dramatic differences in the importance of the hook, other parts of the protein displacement mechanisms are more similar. A monomer of wild type Dda is unable to displace either streptavidin or trp repressor under single turnover conditions. At least two Dda molecules are required for both displacement reactions. Dda also exhibits cooperativity for displacement of streptavidin ²⁰ and E. coli trp repressor ¹⁸ with dramatic increases in product formation as more Dda molecules bind to the same substrate. Trp repressor displacement rates do not change as increasing numbers of Dda molecules bind. However, for streptavidin displacement, both the rate and quantity of product formation increase as the number of Dda molecules bound increases. This cooperativity is observed for both the wild type and Δ279–284 Dda. Therefore, although the hook appears to be much more important for streptavidin displacement than for displacement of DNA binding proteins, other aspects of Dda's protein displacement mechanism such as functional cooperativity are important for both displacement of streptavidin and DNA binding proteins.

4. MATERIAL AND METHODS

4.1. Oligonucleotides

Oligonucleotides were purchased from Integrated DNA Technologies and purified by denaturing polyacrylamide gel electrophoresis. ¹² The sequences are in Table S1.

4.2. Expression and purification of recombinant proteins

4.2.1. Dda purification

Wild type and Δ279–284 Bacteriophage T4 Dda were expressed from pET28b with an N‐terminal 6x His‐tag. Cells were grown to an OD₆₀₀ of 1.0 at 37°C. IPTG was added to 0.5 mM and cells were allowed to grow for 18 hr at 16°C. Cells were harvested by centrifugation at 18,000g for 10 min. The cell pellet was resuspended in lysis buffer (20 mM sodium phosphate buffer pH 7.5, 500 mM NaCl, 5 mM β‐mercaptoethanol, 10% glycerol, 20 mM imidazole, 4 mM phenylmethane sulfonyl fluoride, 4 μg/ml pepstatin A) with 2.25 mg lysozyme per gram of cells, and passed through a microfluidizer to lyse the cells. The lysate was centrifuged at 100,000g for 1.5 hr before loading on a Ni‐NTA Agarose column. After washing with five bed volumes of buffer containing 20 mM imidazole, bound proteins were eluted with buffer containing 500 mM imidazole. Protein was further purified on a Ni‐sepharose column, washed with five bed volumes of buffer, and eluted with a gradient from 20 mM to 500 mM imidazole in buffer. Fractions containing Dda were loaded onto a ssDNA cellulose column in 25 mM Tris, pH 7.5, 20 mM NaCl, 1 mM EDTA, 5 mM β‐mercaptoethanol, and 10% glycerol and washed with five bed volumes of the same buffer. Proteins were eluted with a gradient from 50 mM to 2 M NaCl in the same buffer. Fractions containing pure Dda were concentrated using Amicon centrifugal filter units. Protein was stored at −80°C in 25 mM HEPES, pH 7.5, 50 mM NaCl, 1 mM EDTA, 2 mM β‐mercaptoethanol, 20% glycerol. Protein concentration was determined by UV absorbance at 280 nm using 64,000 M⁻¹ cm⁻¹ as an extinction coefficient and with Coomassie Plus Protein assay using BSA as a standard. A gel image of the purified proteins is in Figure S3.

4.2.2. BamHI purification

Codon optimized Bacillus amyloliquefaciens H BamHI‐E111A was ordered from Genscript and cloned into pSUMO at the BsaI restriction site using Gibson Assembly to produce a 6× His‐SUMO tagged BamHI‐E111A. LOBSTR cells ⁴⁵ containing the plasmid were grown to an OD₆₀₀ of 0.9 at 37°C. IPTG was added to 1 mM and cells were allowed to grow for 20 hr at 18°C. Cells were harvested by centrifugation at 18,000g for 10 min. The cell pellet was resuspended in lysis buffer (50 mM sodium phosphate buffer pH 7.5, 300 mM NaCl, 5 mM β‐mercaptoethanol, 10% glycerol) with EDTA free protease inhibitor tablets (Pierce) and passed through a microfluidizer to lyse. The lysate was centrifuged at 174,000g at 4°C for 1 hr before loading on Ni‐NTA column. The column was washed with 20% elution buffer (50 mM sodium phosphate buffer pH 7.5, 300 mM NaCl, 5 mM β‐mercaptoethanol, 10% glycerol, 300 mM imidazole), and proteins were eluted with 100% elution buffer. The His‐SUMO tag was cleaved with His‐tagged Ulp1, and the sample was reloaded on the Ni‐NTA column and eluted with elution buffer. Untagged BamHI‐E111A eluted in the flow through and was further purified on a 120 ml Superdex 75 size exclusion column. Protein concentration was determined by with Coomassie Plus Protein assay with BSA as a standard. A gel image of the purified protein is in Figure S3.

4.2.3. Ulp1 purification

The catalytic domain of S. cerevisiae Ulp1 (amino acids 403–621) with a C‐terminal His‐tag was expressed from pET24d in Rosetta2(DE3) cells. ⁴⁶ Cells were grown to an OD₆₀₀ of 0.9 at 37°C. IPTG was added to 1 mM and cells were allowed to grow for 20 hr at 18°C. Cells were harvested by centrifugation at 18,000g for 10 min. The cell pellet was resuspended in lysis buffer (50 mM Sodium Phosphate buffer pH 7.5, 300 mM NaCl, 5 mM β‐mercaptoethanol, 10% glycerol) with EDTA free protease inhibitor tablets (Pierce) and passed through a microfluidizer to lyse. The lysate was centrifuged at 174,000g at 4°C for 1 hr before loading on Ni‐NTA column. Proteins were eluted with elution buffer (50 mM Sodium Phosphate buffer pH 7.5, 300 mM NaCl, 5 mM β‐mercaptoethanol, 10% glycerol, 300 mM imidazole). Fractions containing Ulp1 were combined and buffer was exchanged for storage buffer (50 mM Sodium Phosphate buffer pH 7.5, 300 mM NaCl, 5 mM β‐mercaptoethanol, 20% glycerol) using Amicon Centrifugal Filter units. Protein concentration was determined with Coomassie Plus Protein assay with BSA as a standard.

4.2.4. gp32 and trp repressor purification

Bacteriophage T4 gp32 was expressed, purified, and fluorescently labeled with 5‐acetamidofluorescein as described. ³⁷ E. coli trp repressor (trpR) was expressed and purified as described. ¹⁸

4.3. DNA binding

Fluorescence polarization was measured in assay buffer (25 mM HEPES pH 7.5, 10 mM KOAc, 0.1 mM EDTA, 2 mM β‐mercaptoethanol, 0.1 mg/ml BSA, and 10 mM Mg[OAc]₂). Fluorescently labeled DNA (1 nM 5′‐FAM‐T₇16bp) was incubated with Dda in assay buffer for 30 min at room temperature before measuring the polarization on a 1420 Victor3V plate reader (PerkinElmer Life Sciences) with excitation at 485 nm and emission after a 535 nm cut‐on filter. Anisotropy was calculated and data were fit to the quadratic equation (Equation (1)) to obtain the dissociation constant (K _D) where E _T is the total enzyme concentration, and D _T is the total DNA concentration.

$$y=0.5\left(K_{\mathrm{D}}+E_{\mathrm{T}}+D_{\mathrm{T}}\right)-0.5\sqrt{{\left(K_{\mathrm{D}}+E_{\mathrm{T}}+D_{\mathrm{T}}\right)}^2-4E_{\mathrm{T}}{D}_{\mathrm{T}}}$$ (1)

4.4. ATP hydrolysis

ATPase activity was measured using an assay in which the hydrolysis of ATP is coupled to NADH oxidation through pyruvate kinase (PK) and lactate dehydrogenase (LDH). The decrease in absorbance at 380 nm due to conversion of NADH to NAD⁺ was measured and directly correlated to ATP hydrolysis in assay buffer (25 mM HEPES pH 7.5, 10 mM KOAc, 0.1 mM EDTA, 2 mM β‐mercaptoethanol, 0.1 mg/ml BSA, 10 mM Mg(OAc)₂, 5 mM ATP, 0.8 mM phosphoenol pyruvate, 17.05 U/ml PK, 24.75 U/ml LDH, and 0.7 mg/ml NADH). The decrease in absorbance at 380 nm by 25 nM Dda upon addition of calf thymus ssDNA (Sigma) was measured using an Ultrospec 2100 pro (Amersham Biosciences). The quantity of ATP hydrolyzed was calculated using the extinction coefficient of NADH at 380 nm of 1,200 M⁻¹ cm⁻¹. Data were fit to a hyperbolic function (Equation (2)) where K _act is the concentration of calf thymus DNA (D) required to obtain half the maximal ATPase rate (k _cat).

$$y=\frac{k_{\mathrm{cat}}\times D}{K_{\mathrm{act}}+D}$$ (2)

4.5. Duplex DNA unwinding

All concentrations are final. Dda (100 nM) was incubated with 5 nM ³²P‐labeled T₁₄20bp or T₇16bp in assay buffer (25 mM HEPES pH 7.5, 10 mM KOAc, 0.1 mM EDTA, 2 mM β‐mercaptoethanol, 0.1 mg/ml BSA) for 5 min at 25°C before initiating the reaction. Enzyme‐DNA complex was mixed at 25°C in a rapid quench flow (KinTek) with 5 mM ATP, 10 mM Mg(Ac)₂, and 125 nM 20mer or 16mer complement (to trap the displaced strand after unwinding) in assay buffer. To ensure single cycle conditions with respect to the DNA, 75 μM (in nucleotides) poly(dT) was added with the ATP. Samples were quenched with 400 mM EDTA. Bromophenol blue and xylene cyanol were added to 0.5 mg/ml each, and glycerol was added to 10%. Samples were separated by 20% native PAGE, visualized with a Typhoon Trio phosphorimager (GE Healthcare), and quantitated using ImageQuant software. Data were fit to the mechanism in Figure 3a with KinTek Explorer ³⁶ to obtain the rate constants for unwinding (k _u) and dissociation (k _d ). Dda has been previously shown to unwind the T₁₄20bp substrate in 4 steps and the T₇16bp substrate in three steps. ⁴⁷

4.6. Dissociation from DNA

Dissociation rate constants were measured by pre‐incubating 100 nM Dda with 1 μM T₇16bp DNA in reaction buffer (25 mM HEPES pH 7.5, 10 mM KOAc, 0.1 mM EDTA, 2 mM β‐mercaptoethanol, and 10 mM Mg(OAc)₂). The reactions were initiated at 25°C by mixing with 12 mg/ml dextran sulfate in an SX.18MV stopped‐flow reaction analyzer (Applied Photophysics). The change in tryptophan fluorescence was monitored after a 320‐nm cut‐on filter (Newport Optical Filter #FSQ‐WG320) with excitation from a 280 nm LED.

4.7. Streptavidin displacement

All concentrations listed are final. ³²P‐labeled 3′‐biotinylated DNA (10 nM) was pre‐incubated with 1 μM streptavidin in assay buffer (25 mM HEPES pH 7.5, 10 mM KOAc, 0.1 mM EDTA, 2 mM β‐mercaptoethanol, 0.1 mg/ml BSA) for 5 min at 25°C before addition of Dda (250 nM, unless otherwise noted) for 5 min before initiating the reaction. The DNA mixture was mixed at 25°C with 5 mM ATP, 10 mM Mg(Ac)₂, and 10 μM biotin to trap displaced streptavidin. Phosphoenol pyruvate (4 mM) and pyruvate kinase/lactate dehydrogenase (10 U/ml each) were added with the ATP to regenerate ATP consumed during the reaction. Samples were quenched with 400 mM EDTA, 0.5 mg/ml bromophenol blue, 0.5 mg/ml each xylene cyanol, and 10% glycerol. Samples were separated by 15% native PAGE, visualized with a Typhoon Trio phosphorimager (GE Healthcare), and quantitated using ImageQuant software. Data were fit to a single exponential function (Equation (3)) where A is the amplitude, k is the rate constant, and t is time.

$$y=A\left(1-e^{-\mathit{kt}}\right)$$ (3)

Cooperativity was determined using the Hill equation (Equation (4)) where A is the amplitude, E _T is the total enzyme concentration, E _0.5 is the enzyme concentration at which half of the DNA is bound, and n is the Hill coefficient.

$$y=\frac{A\times E_{\mathrm{T}}^n}{E_{\mathrm{T}}^n+E_{0.5}^n}$$ (4)

4.8. gp32 displacement

All concentrations are final. Gp32‐Fl (375 nM) was incubated with 5 μM M13 ssDNA in assay buffer (25 mM HEPES pH 7.5, 10 mM KOAc, 0.1 mM EDTA, 2 mM β‐mercaptoethanol, 10 mM Mg(OAc)₂) for 5 min at 25°C before initiating the reaction. The reaction was initiated by mixing with 25 nM Dda and 5 mM ATP in assay buffer at 25°C in an SX.18MV stopped flow reaction analyzer (Applied Photophysics). Fluorescence was monitored after a 515 nm cut‐on filter (Newport Optical Filter FSR‐OG515) with excitation at 490 nm through 2 mm slits. The fluorescence data were fit to a single exponential plus a slope (Equation (5)) where A is the amplitude, k is the rate constant, m is the slope, and t is time.

$$y=A\left(1-{\mathrm{e}}^{-\mathit{kt}}\right)+\mathit{mt}$$ (5)

4.9. BamHI displacement

All concentrations listed are final. BamHI‐E111A (200 nM) was preincubated with 1 nM ³²P‐labeled T₇16bp‐BamHI duplex DNA for 15 min at 25°C in assay buffer (25 mM HEPES pH 7.5, 10 mM KOAc, 0.1 mM EDTA, 2 mM β‐mercaptoethanol, 0.1 mg/ml BSA). Dda (200 nM) was added to the DNA mixture and incubated an additional 5 min at 25°C. Enzyme‐DNA complex was mixed at 25°C in a rapid quench flow (KinTek) with 5 mM ATP, 10 mM Mg(OAc)₂, and 150 nM 16mer‐BamHI complement to trap the displaced strand after unwinding. To ensure single cycle conditions with respect to the DNA 4 mM (in nucleotides) T ₅₀ was added with the ATP. Samples were quenched with 400 mM EDTA. Loading buffer was added to 5% glycerol, 0.5 mg/ml bromophenol blue, and 0.5 mg/ml xylene cyanol. Samples were separated by 20% native PAGE, visualized with a Typhoon Trio phosphorimager (GE Healthcare), and quantitated using ImageQuant software. Data were fit to a single exponential function (Equation (3)).

4.10. trp repressor displacement

All concentrations listed are final. E. coli trp repressor (5 μM) was pre‐incubated with 100 nM FAM‐labeled 5′‐F‐T₈30bp‐trpR duplex DNA for 5 min at 25°C in assay buffer with tryptophan (25 mM HEPES pH 7.5, 10 mM KOAc, 0.1 mM EDTA, 2 mM β‐mercaptoethanol, 0.1 mg/ml BSA, 0.5 mM L‐Tryptophan). Dda (750 nM) was added to the DNA mixture and incubated an additional 5 min at 25°C. Enzyme–DNA complex was mixed at 25°C in a rapid quench flow (KinTek) with 5 mM ATP, 10 mM Mg(OAc)₂, and 5 μM 12mer‐trpR complement to trap the displaced strand after unwinding. Samples were quenched with 200 mM EDTA, 0.7% SDS. Loading buffer was added to 5% glycerol, 0.5 mg/ml xylene cyanol. Samples were separated by 20% native PAGE, visualized with an Amersham Typhoon RGB Imager (Cytiva) with a 488 nm laser and 525BP20 emission filter, and quantitated using ImageQuant software. Data were fit to a sum of two exponential functions (Equation (6)).

$$y=A_1\left(1-e^{-k_{1}t}\right)+A_2\left(1-e^{-k_{2}t}\right)$$ (6)

CONFLICT OF INTEREST

The authors declare no conflicts of interest.

AUTHOR CONTRIBUTIONS

Alicia K. Byrd: Conceptualization (equal); formal analysis (lead); funding acquisition (supporting); investigation (lead); methodology (lead); project administration (lead); supervision (lead); writing – original draft (lead). Emory G. Malone: Investigation (equal); writing – review and editing (supporting). Lindsey Hazeslip: Investigation (supporting); writing – review and editing (equal). Maroof Khan Zafar: Investigation (supporting); writing – review and editing (supporting). David K. Harrison: Investigation (supporting); writing – review and editing (supporting). Matthew D. Thompson: Investigation (supporting); writing – review and editing (supporting). Jun Gao: Investigation (supporting); writing – review and editing (supporting). Senthil K. Perumal: Investigation (supporting); writing – review and editing (supporting). John C. Marecki: Investigation (supporting); writing – review and editing (equal). Kevin D Raney: Conceptualization (equal); funding acquisition (lead); writing – review and editing (equal).

Supporting information

Appendix S1: Supporting Information

Byrd AK, Malone EG, Hazeslip L, Zafar MK, Harrison DK, Thompson MD, et al. A structural feature of Dda helicase which enhances displacement of streptavidin and trp repressor from DNA . Protein Science. 2022;31:407–421. 10.1002/pro.4232