Skip to main content
NCBI home page
The EMBO Journal logoLink to The EMBO Journal
. 2008 May 22;27(12):1718–1726. doi: 10.1038/emboj.2008.100

Coupling of DNA unwinding to nucleotide hydrolysis in a ring-shaped helicase

Ilker Donmez 1, Smita S Patel 1,a
PMCID: PMC2435122  PMID: 18497749

Abstract

The ring-shaped T7 helicase uses the energy of dTTP hydrolysis to perform the mechanical work of translocation and base pair (bp) separation. We have shown that the unwinding rate of T7 helicase decreases with increasing DNA stability. Here, we show that the dTTPase rate also decreases with increasing DNA stability, which indicates close linkage between chemical transition steps and translocation steps of unwinding. We find that the force-producing step during unwinding is not associated with dTTP binding, but dTTP hydrolysis or Pi release. We determine that T7 helicase extracts ∼3.7 kcal/mol energy from dTTPase to carry out the work of strand separation. Using this energy, T7 helicase unwinds ∼4 bp of AT-rich DNA or 1–2 bp of GC-rich DNA. T7 helicase therefore adjusts both its speed and coupling ratio (bp/dTTP) to match the work of DNA unwinding. We discuss the mechanistic implications of the variable bp/dTTP that indicates T7 helicase either undergoes backward movements/futile hydrolysis or unwinds DNA with a variable bp-step size; ‘long and fast' steps on AT-rich and ‘short and slow' steps on GC-rich DNA.

Keywords: coupling ratio, helicase, motor, stepsize

Introduction

Helicases are key components of the biological machinery that catalyse the numerous DNA and RNA metabolic processes in the cell (Matson et al, 1994; Lohman and Bjornson, 1996; Patel and Picha, 2000; Singleton et al, 2007). Helicases are classified as nucleic acid motor proteins because of their unique ability to convert the chemical energy of NTP hydrolysis into mechanical work of directional movement and nucleic acid strand separation. The unidirectional translocation activity of helicases drives processes such as DNA recombination, nucleosome remodelling, packaging of genomes, DNA restriction, and protein dissociation, whereas the nucleic acid base-pair (bp) separation activity of helicases is critical for genome replication and repair. Very little is known about the energetics and the chemo-mechanical coupling during nucleic acid unwinding by helicases. We investigate here the mechanism of a hexameric ring-shaped helicase, the bacteriophage T7 helicase (T7 gp4A′ protein), which has been a model protein to understand DNA replication as well as the mechanisms of DNA unwinding (Donmez and Patel, 2006).

Movement along nucleic acids is an important property of helicases. It is assumed that processive helicases move and unwind nucleic acids by taking discrete elementary steps; each step being fuelled by the energy from NTP hydrolysis. T7 helicase uses the chemical energy from dTTP hydrolysis to move along ssDNA in a highly processive manner and to unwind dsDNA (Matson and Richardson, 1983; Matson et al, 1983). The cooperativity among the six subunits of T7 helicase and the topological association of the hexameric ring with the DNA are likely responsible for the high processivity of T7 helicase (Egelman et al, 1995). The six subunits of T7 helicase exert an effect cooperatively and fire dTTPs in a sequential manner, which facilitates sequential hands-off of DNA to neighbouring subunits providing a mechanism for stepwise and processive DNA movement (Liao et al, 2005; Crampton et al, 2006).

T7 helicase unwinds dsDNA using the strand exclusion mechanism, whereby it threads and moves unidirectionally along one strand of the dsDNA while excluding the complementary strand from its central channel (Ahnert and Patel, 1997; Hacker and Johnson, 1997; Jezewska et al, 1998; Patel and Picha, 2000; Kaplan and O'Donnell, 2002). Recent studies have shown that T7 helicase unwinds DNA by a partially active mechanism (Donmez et al, 2007; Johnson et al, 2007). That is, it destabilizes the bp at the junction, but the interaction energy of T7 helicase with the ss/ds junction is not sufficient to allow the helicase to move through duplex regions as fast as it does along ssDNA. Therefore, the dsDNA poses a barrier to helicase's movement and decreases its translocation rate during DNA unwinding (Donmez et al, 2007; Johnson et al, 2007).

The unidirectional translocation of T7 helicase has been measured by ensemble methods and more recently by single-molecule force measurements (Kim et al, 2002; Johnson et al, 2007). T7 helicase moves efficiently along ssDNA and hydrolyses one dTTP on an average to move ∼3 nt of ssDNA (Kim et al, 2002). When T7 helicase is stalled at a lesion on ssDNA, its dTTP hydrolysis is inhibited and the helicase–dTTP complex remains bound to the DNA at the site of modification (Brown and Romano, 1989, 1991). The data indicate that the dTTPase activity of T7 helicase is tightly coupled to movement along ssDNA, and therefore the experimentally measured ∼3 nt/dTTP coupling ratio is likely to be close to the elementary nt step-size of T7 helicase translocating along ssDNA (nt step-size is the average distance moved per dTTP hydrolysed). As compared with non-ring-shaped helicases such as PcrA (Dillingham et al, 2000) and UvrD (Tomko et al, 2007) that move on an average 1 nt for every ATP hydrolysed, the ring-shaped T7 helicase is three times more fuel efficient.

The hydrolysis of a single NTP molecule can provide enough energy to unwind 4–6 bp of dsDNA (Lohman and Bjornson, 1996). This information can be obtained from the determination of the bp/NTP coupling ratio, which is not known for any ring-shaped helicase. Similarly, it is not known how much energy is extracted from NTP hydrolysis by the helicase to carry out the work of strand separation (thermodynamic efficiency). As it takes more energy to separate GC-rich DNA regions than AT-rich DNA regions, it is of interest to determine how the thermodynamic efficiency changes as the helicase unwinds DNA regions of different stabilities. It also remains unknown whether the coupling ratio (bp unwound/NTP hydrolysed) remains constant or changes as the helicase faces an increasing opposing force from more stable DNA regions.

To address the questions raised above, we have measured the rates of DNA unwinding and dTTP hydrolysis using transient state kinetic methods for T7 helicase, and more importantly carried out these studies with a range of dsDNA substrates with defined DNA stability. The DNA stability exerts an effect as a load or an opposing force to the forward movements of the translocating helicase. Studying helicase parameters as a function of dsDNA stability provides not only the energetics of unwinding but can also provide the location of the mechanical step of movement within the dTTPase cycle. Through studies of unwinding and dTTPase, we explore the possibility that the rate-limiting step for DNA unwinding is one of the chemical transition steps. Our results show that the coupling ratio bp/dTTP of T7 helicase is not a constant, but decreases with increasing dsDNA stability. The results provide new mechanistic insights into the mechanisms of DNA unwinding and reveal possibilities of backward movements or variable bp-step sizes of T7 helicase during DNA unwinding.

Results

Real-time unwinding of DNA substrates

To create DNA substrates with predefined stabilities, we introduced different percentage of GC bps, from 5 to 80% uniformly distributed in a 40-bp duplex DNA region (sequences in Supplementary data). The average bp stability of the DNA substrates was calculated using the nearest neighbour analysis (Breslauer et al, 1986), and it ranged from 1.1 to 2.1 kcal/mol/bp. Each DNA substrate contained a 5′-dT35 tail, which is of sufficient length to load a single T7 helicase hexamer on the DNA before starting the unwinding reaction, and a 3′-dT15 tail for providing strand exclusion during unwinding (Ahnert and Patel, 1997). To measure the average DNA-unwinding rate, we used the all or none unwinding assay (Ali and Lohman, 1997), which is commonly carried out in a discontinuous manner whereby radioactively labelled DNA substrates and unwound products are resolved using gel electrophoresis after the reactions are stopped with EDTA and SDS. The discontinuous nature of the assay, however, creates an uncertainty in the actual length of the dsDNA unwound. This is because partially unwound duplexes, especially those intermediates that have a very short stretch of dsDNA remaining to be unwound, can spontaneously fall apart in the time period between reaction quenching and product analysis by gel electrophoresis. Thus, the actual length of dsDNA that is unwound by the helicase could be shorter and needs to be defined. The length of the duplex region that spontaneously falls apart (Lm) has been measured experimentally to be around 10 bp (Galletto et al, 2004; Jeong et al, 2004; Eoff and Raney, 2006). This is not desirable because the 10 bp Lm is almost 25% of the 40-bp length of the dsDNA substrates used here. To circumvent the Lm problem, we have developed a fluorescence-based unwinding assay that measures dsDNA unwinding in real time.

In this fluorescence-based assay, a fluorescein dye was covalently attached to the 5′-end of the excluded strand, and four consecutive guanines (G) were placed at the 3′-end of the loading strand (three of the G's constituting an overhang) (Figure 1A). The G nucleotides being in close proximity to the fluorescein dye quench the dye fluorescence only when the DNA is duplexed (Supplementary Figure 1a) (Torimura et al, 2001; Noble et al, 2005). When the dsDNA gets completely unwound and the dye-labelled strand moves away from the G-containing complementary strand, the fluorescein fluorescence increases. Neither the binding nor the translocation of T7 helicase on the dye-labelled strand changed the fluorescein fluorescence (Supplementary Figures 1c and d). The fluorescence-based assay is an all or none assay (two signal levels corresponding to ss and dsDNA), but it has the advantage of monitoring DNA unwinding in real time.

Figure 1.

Figure 1

Open in a new tab

Determination of Lm using the real-time unwinding assay. (A) Fluorescent DNA-unwinding substrate used for the real-time unwinding assays. The DNA contained two ssDNA overhangs, -GGG overhang at the 3′-end of the loading strand, and a fluorescein dye (F) at the 5′-end of the displaced strand. The ring-shaped helicase (blue ring) travels along the 5′-ssDNA to unwind the dsDNA region. (B) T7 helicase hexamer (40 nM) was preincubated with the fluorescent DNA substrates (5 nM) with dTTP (1 mM) and EDTA at 18°C in one syringe of the stopped-flow apparatus and unwinding reaction was initiated by addition of Mg(II) and SSB. Four DNA substrates with the same GC content (5%), but varying lengths of duplex regions (40, 50, 60, and 70 bp in black, red, green, and orange, respectively) were assayed for unwinding. The signal results from the separation of the fluorescein label opposite to several guanines at the end of the duplex. The resulting time traces of unwinding (Supplementary Figure 3a) were normalized between the values of 1 and 2. The time traces were globally fit to Equation 1) in which stepping rate (k), step size m=(L−Lm)/n (where L is the length of the dsDNA region, n is the number of steps required to unwind the dsDNA), and the length of the spontaneously unwound region at end (Lm) were globally constrained. The best fit results (solid lines) provided k of 42.03 (36.6, 51.5) bp steps/s, m of 1.34 (1.09, 1.56) bp, and Lm of 1.5 (0, 3.37) bp. The average unwinding rate calculated from these parameters (k × m) is 56.2 bp/s. The upper and lower limits for the parameters given in parentheses were obtained as described (Straume and Johnson, 1992).

To test the effectiveness of this assay, we first determined the Lm value of our DNA substrates by measuring the unwinding kinetics as a function of dsDNA length for substrates with an average GC content of 5%. We used DNA substrates with the lowest GC content to minimize processivity complications and because it is expected to provide the greatest Lm being the least stable of all the DNAs used. Four duplexes with the same 5% GC content, but varying lengths of duplex regions (40, 50, 60, and 70 bp) were used as unwinding substrates. T7 helicase was preincubated with the dye-labelled DNA in the presence of dTTP and EDTA (which prevents dTTP hydrolysis and unwinding, but allows T7 helicase to form a stable hexamer on the DNA) (Hingorani and Patel, 1993, 1996; Picha and Patel, 1998). The unwinding reactions were initiated by the addition of Mg(II) and Escherichia coli single-stranded binding (SSB) in a stopped-flow apparatus. The SSB protein acts as a DNA trap and stops unwinding beyond a single turnover (Donmez et al, 2007). The SSB was used as a trap instead of a ssDNA to keep the reaction conditions as similar as possible between measurement of DNA-unwinding and dTTP hydrolysis rates (see below). The unwinding reactions were carried out at 18°C in a stopped-flow instrument and monitored by recording the fluorescein fluorescence in real time. All the DNA substrates were unwound with lag kinetics consistent with the all-or-none type of assay used and the observed lag time increased progressively with increasing dsDNA lengths (Figure 1B and Supplementary Figure 3a). The time courses were fit globally to the stepping model (Equation (1)), where three parameters were shared: Lm, the step size, m=(L−Lm)/n, and the stepping rate, k (where L is the length of the dsDNA and n is the apparent number steps to unwind the dsDNA). The best global fit provided Lm of 1.5 bp, k of 42 bp step/s, and m of 1.4 bp. Note that the Lm of 1.5 bp obtained when using the real-time unwinding assay is much smaller than the ∼12 bp reported for ∼30% GC-rich DNA whose unwinding was measured using the gel assays (Jeong et al, 2004). We expect the Lm for the DNA substrates with a higher GC content to be even smaller, and hence the unwinding rates measured by the real-time assay are accurate and not significantly affected by the Lm.

Mechanochemistry of the DNA-unwinding reaction

Very little is known about the interdependence of the chemical reaction of dTTP hydrolysis and DNA movement or the chemo-mechanical coupling. It is expected that one or more of the elementary steps in the dTTPase pathway are associated with the DNA movement step. By studying the kinetics of unwinding as a function of [dTTP] and dsDNA stability, we can gain insights into the location of the step associated with DNA movement (Bustamante et al, 2004). This is because the stability of the dsDNA is effectively a load or an opposing force that the helicase experiences when it is unwinding the dsDNA; thus, the dependence of unwinding rate on dsDNA stability can provide insights into the mechanical step.

The DNA-unwinding kinetics were measured using the fluorescence-based assay for the 40 bp dsDNA substrates, 5–80% GC rich at 500 μM dTTP. T7 helicase (40 nM) was preincubated with the DNA substrate (5 nM) in the presence of EDTA and dTTP, and reactions were initiated with Mg(II) and SSB in a stopped-flow instrument. A typical set of unwinding time traces at [dTTP] of 500 μM (Figure 2A and Supplementary Figure 3b) shows that the unwinding lag time increases as the dsDNA stability increases. The kinetic traces were fit to the stepping model and the average rate of DNA unwinding was determined from k/n parameters (Supplementary Table 1). The average unwinding rates were plotted against dsDNA stability (Figure 2B), which shows a linear decrease in rate from ∼60 to 9 bp/s over the range of 5–80% GC content in the DNA. Thus, the decrease in unwinding rate of T7 helicase with increasing dsDNA stability is nearly seven-fold.

Figure 2.

Figure 2

Open in a new tab

Dependence of DNA-unwinding rate on dsDNA stability. (A) The unwinding kinetics of the 40-bp DNA substrates with GC content from 5 to 80% were measured as described in Figure 1 at 500 μM dTTP. Experimental data (normalized between 1 and 2 for visual purposes, see Supplementary Figure 3b for unnormalized traces) and the fits to the simplest model (Equation (1), Materials and methods) in solid lines are shown. The fitted parameters with their confidence intervals are shown in Supplementary Table 1 from which the average unwinding rates (40 × k/n) were calculated. (B) The average unwinding rates are plotted against average bp stability. The error bars shown here represent standard deviations of 6–8 independent determinations.

The unwinding kinetics were measured at various [dTTP] ranging from 20 μM to 1.28 mM for each of the 40 bp DNA substrates (Supplementary Figures 3c and d), and the average unwinding rates derived from fitting to the stepping model were plotted as a function of dsDNA stability (Figure 3A). At all [dTTP], the unwinding rate decreases almost linearly with increasing dsDNA stability. At high [dTTP], the decrease in the unwinding rate with increasing dsDNA stability is steeper as compared to when the [dTTP] was low.

Figure 3.

Figure 3

Open in a new tab

Dependence of DNA-unwinding rate on dsDNA stability and [dTTP] (A). The average DNA-unwinding rates, determined as shown in Figure 2, at various [dTTP] when plotted against bp stability show linear decreases at all [dTTP]. Individual time traces for 5 and 80% GC DNAs at various [dTTP] are shown in Supplementary Figures 3c and d. (B) The same data in (A) are replotted as average DNA-unwinding rates against [dTTP] for each of the DNA substrates ranging in GC content from 5 to 80%. The solid lines are the fits of the rate versus [dTTP] to a hyperbola (observed unwinding rate=kcat × [dTTP]/Km+[dTTP]). (C) The maximal rate of DNA unwinding (kcat) and dTTP Km are plotted against average bp stability. (D) The ratio kcat/Km shows a slight decrease with increasing bp stability. The error bars are standard errors from computer fits and from combined errors of measurements (Fersht, 1999).

The data in Figure 3A were replotted as unwinding rates against [dTTP] (Figure 3B). The plots show that the unwinding rates increase in a hyperbolic manner with increasing [dTTP] and hence were fit to the Michaelis–Menten equation. The derived kcat and Km were then plotted against dsDNA stability (Figure 3C). Examination of the dependence of these kinetic parameters on dsDNA stability allowed us to investigate which step/s in the dTTPase cycle is associated with movement. As the GC content increased from 5 to 80% in the DNA substrates, the maximal unwinding rate or kcat decreased 8.5-fold and the Km decreased 6.6-fold. The ratio kcat/Km, however, showed a relatively small decrease of about 25% and thus remained essentially independent of dsDNA stability (Figure 3D). The observation that both kcat and Km decrease with increasing dsDNA stability, but the kcat/Km ratio, which is the apparent second order rate constant for dTTP binding, remains independent of dsDNA stability indicates that the dTTP-binding steps are not the force-producing steps for dsDNA unwinding (Bustamante et al, 2004).

dTTP hydrolysis is coupled to DNA unwinding

The above experiments indicate the presence of a mechanically sensitive rate constant in the cycle of dTTP hydrolysis. To explore this finding further and to determine the bp/dTTP coupling ratio of T7 helicase, we designed experiments to measure the dTTP hydrolysis rate during DNA unwinding. For the determination of the bp/dTTP coupling, we made sure that the dTTP hydrolysis rate was measured under solution conditions (temperature, buffer composition, [dTTP] and T7 helicase) similar to DNA unwinding (Martinez-Senac and Webb, 2005). The reactions were also carried out under conditions where T7 helicase is a stable hexamer (500 μM dTTP and 20 nM helicase hexamer) (Hingorani and Patel, 1993, 1996; Patel and Hingorani, 1993; Picha and Patel, 1998). To optimize the signal, the dTTPase reaction was carried out at [DNA]>[helicase], and to make the reactions single turnover E. coli SSB was added, which acts as a DNA trap and stops both unwinding and DNA-stimulated dTTP hydrolysis beyond a single turnover (Donmez et al, 2007). The ssDNA is a commonly used protein trap that works well in the unwinding reaction, but it stimulates the helicase's dTTPase activity and hence cannot be used as a trap to measure the dTTPase kinetics. The dTTPase rate is also sensitive to the active [helicase–DNA] and processivity. The active [helicase–DNA] was experimentally determined from its Kd measured using a functional unwinding assay (Supplementary Figure 1e). The unwinding processivity of T7 helicase has been experimentally determined (Donmez et al, 2007) and was taken into account when determining the dTTPase rate from model fittings (Equations (2) and (3)).

The dTTP hydrolysis kinetics were measured in real time using the phosphate-binding protein (PBP) that acts as a sensitive sensor for fast detection of low concentrations of Pi (Brune et al, 1994). This assay enables an instantaneous monitoring of Pi concentration from the fluorescent signal produced when Pi binds to coumarin dye-labelled PBP. T7 helicase was preincubated with the 40-bp dsDNA substrates of different stabilities and the Pi release kinetics were measured at 500 μM dTTP in a stopped-flow instrument. Two prominent phases of Pi release kinetics were observed (Figure 4A). The fast phase of Pi production (burst phase) occurs in the same time scale as dsDNA unwinding (compare time traces in Figure 2A to Figure 4A) for each of the DNA substrates. The prominent burst phase of Pi production, therefore, provides information on the rate of dTTP hydrolysis during the time T7 helicase is unwinding the dsDNA. This was further supported by a set of Pi release kinetics carried out with DNA substrates of different lengths (40, 50, 60, and 70 bp 5% GC-rich DNAs). Consistent with the analysis above, the burst amplitude of Pi production increased with increasing length of dsDNA (Supplementary Figure 2a). The linear increase in Pi production after the initial burst phase occurs at the same rate as the intrinsic dTTPase rate of T7 helicase free from bound DNA (Jeong et al, 2002). A brief phase at the beginning of the reactions with an even steeper slope was relatively invariant from DNA to DNA (Supplementary Figure 2e). This phase likely results from the fast hydrolysis of pre-bound dTTPs.

Figure 4.

Figure 4

Open in a new tab

Dependence of dTTP hydrolysis rates on dsDNA stability. T7 helicase hexamer (40 nM) was preincubated with the unwinding substrate (200 nM) in the presence of dTTP (1 mM), EDTA, Pi-mop, and MDCC-labelled PBP (as described in the Materials and methods) at 18°C in one syringe of the stopped-flow apparatus and reaction was initiated by addition of Mg(II), Pi-mop, and SSB from a second syringe. (A) Time traces of Pi production (Pi/T7 helicase hexamer, M/M) during T7 helicase unwinding of DNA substrates with different stability. The solid lines are fit to the model for DNA unwinding and dTTP hydrolysis (Equations (2) and (3), Materials and methods). The following were the best fit parameters from model fitting for DNA substrates (5, 20, 35, 50, 65, and 80% GC): b (15.2, 13.4, 11.0, 8.4, 9.5, and 5.7 s−1), c (2.1, 2.2, 2.1, 2.2, 2.7, and 1.6 s−1), k (40.3, 14.85, 7.98, 2.27, 4.4, and 1.2 bp step/s), and n (48.6, 19.1, 15.1, 5.5, 12.9, and 7.0). (B) The rates of dTTP hydrolysis during dsDNA unwinding obtained from model fitting, b, or slopes of the burst phase are plotted against the dsDNA stability. The error bars are standard deviations of average data from 6–8 independent measurements.

Analysis of the Pi release burst amplitudes (Figure 4A) showed that about 15–20 dTTP molecules were hydrolysed during the unwinding time of the 40 bp dsDNAs. In principle, the amplitude can be used to determine the bp/dTTP coupling ratio. However, the amplitude needs to be corrected for the imperfect processivity of the helicase and for the active helicase–DNA complex in the reaction. The dTTP hydrolysis rate associated with DNA unwinding can also be determined from the slope of the burst phase. A more rigorous treatment of the dTTPase data requires correction for the imperfect processivity of the helicase, the active [helicase–DNA], and the intrinsic hydrolysis rate (dTTP hydrolysis rate in the absence of DNA). The Pi release kinetics were fit to the model (Equations (2) and (3), Materials and methods) to obtain the dTTPase rate during DNA unwinding. The dTTP hydrolysis rates (b) derived from model fitting are plotted next to the rates obtained from the simpler treatment of slope determination (Figure 4B). The rates from both treatments are quite similar, and the data clearly show a decrease in the dTTPase activity with increasing dsDNA stability. The decrease in unwinding rate (Figure 2) and the corresponding decrease in the dTTPase rate with increasing dsDNA stability indicate that the chemical transition steps are closely linked with the translocation steps during DNA unwinding. When the helicase slows down due to greater opposing force from dsDNA stability, its dTTPase activity also slows down.

The bp/dTTP coupling ratio decreases with increasing DNA stability

Knowing the unwinding rate and the dTTP hydrolysis rate allowed us to calculate the bp/dTTP coupling ratio during DNA unwinding. The coupling ratio was calculated by dividing the unwinding rate with the dTTPase rate (from data in Figures 2B and 4B) for each of the DNA substrates at [dTTP]=500 μM. Interestingly, we find that the coupling ratio is not a constant, but decreases with increasing dsDNA stability (Figure 5). The coupling ratio for the AT-rich DNAs is 3–4 bp/dTTP, which is very similar to the coupling ratio nt/dTTP of T7 helicase when it is translocating along ssDNA. The coupling ratio for the GC-rich DNAs, on the other hand, is lower and 1–2 bp/dTTP.

Figure 5.

Figure 5

Open in a new tab

Dependence of bp/dTTP coupling ratio and work-step on dsDNA stability. The bp/dTTP coupling ratio was determined by dividing the unwinding rate (from Figure 2B) and the dTTP hydrolysis rate (from Figure 4B, model fitting) at 500 μM dTTP. The work-step was calculated by multiplying the bp/dTTP with the average bp stability of the DNA substrates. The error bars were determined from the combined errors of measurements (Fersht, 1999).

The thermodynamic efficiency of T7 helicase as a function of dsDNA stability

Knowing the bp/dTTP coupling ratio and the energetics of base pairing in dsDNA, we can determine the average work involved in unwinding AT-rich versus GC-rich DNA. We define the term work-step here as the average work carried out in separating the DNA strands per dTTP hydrolysed. Work-step (kcal/mol)=[(unwinding rate (bp/s)/dTTPase rate (s−1)] × ΔG/bp (kcal/mol/bp). Where ΔG/bp is the average free energy of the bp for a particular substrate calculated from the free energy of the dsDNA using the nearest neighbour analysis (Breslauer et al, 1986). The work-step was calculated for each of the DNA substrates and plotted against dsDNA stability (Figure 5). The results show a relatively small decrease in work-step from 4 to 3 kcal/mol with increasing dsDNA stability. We calculated an average work-step of 3.7 kcal/mol for T7 helicase over the physiological range of DNA stability, which is the free energy that T7 helicase extracts from dTTP hydrolysis to carry out useful work of unwinding dsDNA.

Knowing the work carried out for dsDNA separation allowed us to calculate the thermodynamic efficiency of the helicase, which is the ratio of the work done by the helicase for bp separation to the energy input from dTTP hydrolysis. Assuming that dTTP hydrolysis can generate ∼10 kcal/mol of free energy, the average thermodynamic efficiency of T7 helicase is ∼37% (3.7 kcal/mol divided by 10 kcal/mol). The average thermodynamic efficiency of the T7 helicase motor is very similar to other motor proteins such as φ29 packaging motor, RNA and DNA polymerases, and cytoskeletal motors such as kinesin and myosin (Bustamante et al, 2004). The thermodynamic efficiency of T7 helicase decreases slightly however from 40 to 30% within the physiological range of dsDNA stability.

Discussion

The studies in this paper show that the DNA-unwinding rate of T7 helicase as well as its presteady-state dTTPase rate during DNA unwinding are strongly influenced by dsDNA stability and both decrease as the GC content in the dsDNA increases. In principle, the decrease in the unwinding rate or the translocation speed of the helicase with increasing dsDNA stability can be due to slowed chemical transitions or due to a slower movement of the helicase. The observed decrease in the presteady-state rate of dTTP hydrolysis with increasing dsDNA stability indicates that the decrease in translocation speed is linked to a slowed chemical transition. To investigate which step in the dTTPase pathway is most sensitive to dsDNA stability, the unwinding rates were measured as a function of DNA stability at various [dTTP]. The kinetic parameters obtained from these experiments show that both kcat and Km decrease with increasing DNA stability, but the kcat/Km ratio remains nearly constant. This characteristic behaviour is similar to DNA stability acting as an uncompetitive inhibitor affecting equilibrium between states not associated with dTTP binding. The data indicate that movement is not associated with the dTTP-binding steps. A similar behaviour has been observed for the phi29 packaging motor (Chemla et al, 2005) where ATP-binding step has been eliminated as the force-producing step.

To narrow down the step involved in force production during DNA unwinding by T7 helicase, we examined the kinetic pathway of dTTP hydrolysis that has been studied in detail (Hingorani et al, 1997; Jeong et al, 2002; Liao et al, 2005). To perform an energetic unfavourable movement on DNA, the helicase needs to have a controllable hold on the DNA. The ssDNA does not bind tightly to the T7 helicase–dTDP state (Hingorani and Patel, 1993, 1996); therefore, the possibility of dTDP release being the power stroke step is low. We are left therefore with Pi release or a conformational change associated with dTTP hydrolysis as the power stroke step. It has been argued that the bond cleavage reaction of nucleotide hydrolysis is unlikely to cause major conformational changes that can drive movement (Oster and Wang, 2000). Therefore, a more likely candidate for a power stroke step is Pi release. The Pi release is the rate-limiting step during translocation of T7 helicase along ssDNA (Jeong et al, 2002; Liao et al, 2005). Therefore, it is very likely that translocation is limited by the rate of Pi release during DNA unwinding. This possibility needs to be examined by further studies.

Knowing the rates of dTTP hydrolysis and DNA unwinding allowed us to calculate the bp/dTTP coupling ratio that provides information on how many bp of DNA are unwound on an average per molecule of dTTP hydrolysed. We observed that the average bp/dTTP ratio of T7 helicase is 3–4 when it is unwinding AT-rich DNA (<50% GC content) and 1–2 when it is unwinding GC-rich DNA. Thus, the coupling ratio of T7 helicase is not a constant and it decreases with increasing dsDNA stability. One reason for the decrease in the coupling ratio with an increase in dsDNA stability could be that T7 helicase undergoes futile dTTP hydrolysis cycles during DNA unwinding. The futile dTTP hydrolysis may arise from occasional backward movements of the helicase or continued dTTP hydrolysis during helicase pauses (Figure 6A). Backward movements have been detected in single molecule studies of recBCD helicase (Perkins et al, 2004) and pauses have been observed in single-molecule studies of the HCV helicase (Dumont et al, 2006; Cheng et al, 2007; Myong et al, 2007). Although such data on T7 helicase are not available, based on the observation that T7 helicase does not hydrolyse dTTP when it is stalled on ssDNA (Brown and Romano, 1989, 1991), we argue that futile dTTPase hydrolysis during pauses of the helicase is less likely. Therefore, one explanation for the decrease in coupling ratio with increasing dsDNA stability is that T7 helicase undergoes occasional backward movements that increases in frequency with increasing dsDNA stability. This possibility needs to be tested experimentally.

Figure 6.

Figure 6

Open in a new tab

Two models of DNA unwinding. (A) Helicase unwinds DNA with a constant bp-step size, but has varying probability of achieving a productive step due to occasional backward movements during DNA unwinding. The backward movements shown here are not fuelled by dTTP hydrolysis and are not necessarily of the same size as the forward movements. The backward movements result in futile dTTP hydrolysis cycles because a segment of dsDNA occasionally is unwound by more than one attempt from the helicase. The higher probability of backward movements on GC-rich DNA can explain its lower coupling ratio (bp/dTTP) as compared with AT-rich DNA. (B) Helicase unwinds DNA with variable bp-step sizes. The helicase takes longer steps on AT-rich DNAs and shorter steps on GC-rich DNAs.

There is, however, an alternative explanation for the decrease in coupling ratio of T7 helicase. If T7 helicase does not undergo backward movements during DNA unwinding or futile dTTPase, then the coupling ratio should be equal to the elementary bp-step size, which is the average distance moved per dTTP hydrolysed. If this were the case, then our data would indicate that the bp-step size of T7 helicase is longer for AT-rich DNA and shorter for GC-rich DNA (Figure 6B). A possible rationale for this behaviour could be that by decreasing the bp-step size the motor is able to keep moving forward at a reasonable speed under increasing load from dsDNA stability (similar to changing gears). Such behaviour has been observed for the cytoplasmic dynein, which is a tightly coupled motor; that is, it takes a forward step for every ATP hydrolysed. The step size of dynein, however, depends on the load and decreases with increasing opposing load (Mallik et al, 2004).

One way that T7 helicase can achieve variable bp-step size is if the helicase's stepping process was dictated by the availability of ssDNA at the fork junction. Such a molecular mechanism for variable step size is illustrated in Supplementary Figure 4. A variable bp-step size can be achieved if the ssDNA-binding loop of T7 helicase in the central channel is flexible and can move 1–3 nt of ssDNA. On AT-rich DNA, the size of the frayed DNA segment could be longer such that ∼3 nt of ssDNA are available to the DNA-binding loop for interactions during the lifetime of a bound dTTP. Therefore, the bp-step size of T7 helicase on AT-rich DNA is ∼3 bp. On GC-rich DNA, fewer bps are frayed, and thus the bp-step size is shorter (1–2 bp) (Supplementary Figure 4).

Further studies will be required to distinguish between the two models. Such studies could include direct step size measurements by single-molecule methods and investigation of backward movements. T7 helicase, as many helicases, works in concert with other proteins to catalyse DNA unwinding during DNA replication. T7 helicase–primase is a part of the replication machinery that includes proteins such as the DNA polymerase and SSB proteins. Our studies here were carried out with the E. coli SSB protein, but in the absence of the DNA polymerase. It has been shown that T7 DNA polymerase increases the speed of T7 helicase unwinding (Stano et al, 2005). The DNA polymerase could accelerate the helicase by dampening helicase's backward motions or by destabilizing the junction and trapping the unwound bp through DNA synthesis. Studies of T7 helicase with the DNA polymerase will be very important in understanding not only the mechanism of DNA replication but also the possible pathways of DNA unwinding. It will be interesting to investigate how the DNA polymerase would modulate the mechanochemical coupling in T7 helicase during DNA unwinding.

Materials and methods

DNA-unwinding assays

All the reactions were performed at 18°C in a stopped-flow instrument (Kintek Corp). An equal volume of a mixture of fluorescein-labelled DNA-unwinding substrate, T7 helicase, and dTTP from one syringe was mixed with Mg(II) and E. coli SSB protein to initiate the reaction. Final reaction buffer was 50 mM Tris–Cl pH 7.6, 50 mM NaCl, 10% glycerol, 2 μM BSA. Final concentrations unless indicated otherwise were dTTP 500 μM, DNA substrate 2.5 nM, T7 helicase hexamer 20 nM, MgCl2 4 mM, and SSB 1 μM. The reaction solution was excited at a wavelength of 480 nm with a slit width of 2 mm (∼8 nm spectral bandwidth) and fluorescence was detected by a photomultiplier tube (Hamamatsu) filtered through a long-pass filter with a cutoff of 515 nm.

dTTP hydrolysis kinetics

The dTTP hydrolysis kinetics was measured using the same buffer conditions and protein concentrations, except that the unwinding substrate was unlabelled and its final concentration was 100 nM. To eliminate contaminant inorganic phosphate, both syringes of the stopped-flow instrument were treated with the Pi-mop (PNPase 1 U/ml and 7 mG, 0.5 mM) for 10 min before starting the reactions. Syringe A contained the Pi-mop (PNPase 0.1 U/ml and 7 mG, 0.5 mM), PBP (4 μM), T7 helicase, DNA-unwinding substrate, and dTTP, and syringe B contained Pi-mop, Mg(II), and SSB. The mixed solution was excited with 425 nm with 1 mm slits (∼4 nm spectral bandwidth) and the emission was detected by PMT with a 450 nm long-pass filter (Corion).

For each set of experiment, a Pi calibration curve was created as follows: standard Pi solutions (Sigma) were mixed with the PBP protein and 6–8 time traces were collected at each [Pi]. The time courses were corrected for the experimentally determined dead time of the stopped-flow instrument (4 ms) and fit to a first degree polynomial to determine the amplitudes, which were corrected for no Pi control, and plotted against [Pi] to obtain the slope.

Data analysis

The unwinding kinetics were fit to Equation (1) (Lucius et al, 2003; Jeong et al, 2004; Donmez et al, 2007). Inline graphic where Fss is the background fluorescence signal, Fds is the maximum fluorescence signal after completion of reaction, n the apparent number of steps to unwind the dsDNA, and ko the apparent unwinding rate per step (k+kd) (k being the stepping rate and kd the dissociation rate). This was the simplest model that takes into account only two states (ds and ssDNA signal levels) and ignores the slight peaking due to fast SSB binding to the newly generated ssDNA. We used a number of more sophisticated models to take care of the third mode that is manifested just before the trace is completed; however, the final normalized unwinding rates were not significantly different, so we present the fits for the simplest model above.

Typically, 6–8 Pi release reaction time courses were averaged, corrected for the instrument dead time, and normalized. The fluorescence signal was converted to [Pi] using the Pi standard curve. To use as initial values in model fitting, we estimated the Pi release rates during different phases by taking the derivative of the time courses with respect to time.

We used Mathematica to obtain the model equations for fitting the Pi release kinetics. We assume that the helicase traverses the DNA by the n-step sequential model. Inline graphic Equation (2) describes the time evolution of the free helicase concentration. Inline graphic where H(t) is the fraction of free helicase at given time t, and H0 is the bound helicase fraction at the start of the reaction. Inorganic phosphate will be generated from these two states of the helicase (DNA bound and free): Inline graphic Equation (3) describes the time evolution of [Pi]. Inline graphic where, processivity [k/(k+kd)] is denoted by R, P is the Pi concentration, P0 is Pi concentration at the start of the reaction, b is the dTTP hydrolysis rate during dsDNA unwinding, and c is the dTTP hydrolysis of the helicase free from DNA. To simplify regression analysis, we fixed as many parameters as possible: P0 was 0, because we could estimate it by extrapolation, H0 was 0.85 from the known Kd of the T7 helicase–DNA complex. We fixed kd in an indirect way by fixing the processivity (kd would be constrained dynamically) to the value from previous measurements (Donmez et al, 2007). Processivities for 5–80% GC DNAs are 0.9982, 0.9972, 0.9957, 0.9934, 0.9889, and 0.9803. At each step of iteration, the value of kd=k × R(−L/n)−1, where k is the forward stepping rate, L is total length of dsDNA, and n is the number of steps.

Supplementary Material

Supplementary Material

emboj2008100s1.doc (3.8MB, doc)

Acknowledgments

We thank Gayatri Patel for PBP preparation, Daniel L Purich for suggesting the term ‘work-step', G Oster for information and notes on mechanochemistry of molecular motors, and CM Drain for useful discussions. This study was supported by the National Institutes of Health (GM55310).

References

  1. Ahnert P, Patel SS (1997) Asymmetric interactions of hexameric bacteriophage T7 DNA helicase with the 5′- and 3′-tails of the forked DNA substrate. J Biol Chem 272: 32267–32273 [DOI] [PubMed] [Google Scholar]
  2. Ali JA, Lohman TM (1997) Kinetic measurement of the step size of DNA unwinding by Escherichia coli UvrD helicase. Science 275: 377–380 [DOI] [PubMed] [Google Scholar]
  3. Breslauer KJ, Frank R, Blocker H, Marky LA (1986) Predicting DNA duplex stability from the base sequence. Proc Natl Acad Sci USA 83: 3746–3750 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Brown WC, Romano LJ (1989) Benzo[a]pyrene–DNA adducts inhibit translocation by the gene 4 protein of bacteriophage T7. J Biol Chem 264: 6748–6754 [PubMed] [Google Scholar]
  5. Brown WC, Romano LJ (1991) Effects of benzo[a]pyrene–DNA adducts on a reconstituted replication system. Biochemistry 30: 1342–1350 [DOI] [PubMed] [Google Scholar]
  6. Brune M, Hunter JL, Corrie JE, Webb MR (1994) Direct, real-time measurement of rapid inorganic phosphate release using a novel fluorescent probe and its application to actomyosin subfragment 1 ATPase. Biochemistry 33: 8262–8271 [DOI] [PubMed] [Google Scholar]
  7. Bustamante C, Chemla YR, Forde NR, Izhaky D (2004) Mechanical processes in biochemistry. Annu Rev Biochem 73: 705–748 [DOI] [PubMed] [Google Scholar]
  8. Chemla YR, Aathavan K, Michaelis J, Grimes S, Jardine PJ, Anderson DL, Bustamante C (2005) Mechanism of force generation of a viral DNA packaging motor. Cell 122: 683–692 [DOI] [PubMed] [Google Scholar]
  9. Cheng W, Dumont S, Tinoco I Jr, Bustamante C (2007) NS3 helicase actively separates RNA strands and senses sequence barriers ahead of the opening fork. Proc Natl Acad Sci USA 104: 13954–13959 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Crampton DJ, Mukherjee S, Richardson CC (2006) DNA-induced switch from independent to sequential dTTP hydrolysis in the bacteriophage T7 DNA helicase. Mol Cell 21: 165–174 [DOI] [PubMed] [Google Scholar]
  11. Dillingham MS, Wigley DB, Webb MR (2000) Demonstration of unidirectional single-stranded DNA translocation by PcrA helicase: measurement of step size and translocation speed. Biochemistry 39: 205–212 [DOI] [PubMed] [Google Scholar]
  12. Donmez I, Patel SS (2006) Mechanisms of a ring shaped helicase. Nucleic Acids Res 34: 4216–4224 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Donmez I, Rajagopal V, Jeong YJ, Patel SS (2007) Nucleic acid unwinding by hepatitis C virus and bacteriophage t7 helicases is sensitive to base pair stability. J Biol Chem 282: 21116–21123 [DOI] [PubMed] [Google Scholar]
  14. Dumont S, Cheng W, Serebrov V, Beran RK, Tinoco I Jr, Pyle AM, Bustamante C (2006) RNA translocation and unwinding mechanism of HCV NS3 helicase and its coordination by ATP. Nature 439: 105–108 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Egelman EH, Yu X, Wild R, Hingorani MM, Patel SS (1995) Bacteriophage T7 helicase/primase proteins form rings around single-stranded DNA that suggest a general structure for hexameric helicases. Proc Natl Acad Sci USA 92: 3869–3873 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Eoff RL, Raney KD (2006) Intermediates revealed in the kinetic mechanism for DNA unwinding by a monomeric helicase. Nat Struct Mol Biol 13: 242–249 [DOI] [PubMed] [Google Scholar]
  17. Fersht A (1999) Structure and Mechanism in Protein Science. A Guide to Enzyme Catalysis and Protein Folding. New York: WH Freeman and Company [Google Scholar]
  18. Galletto R, Jezewska MJ, Bujalowski W (2004) Unzipping mechanism of the double-stranded DNA unwinding by a hexameric helicase: the effect of the 3′ arm and the stability of the dsDNA on the unwinding activity of the Escherichia coli DnaB helicase. J Mol Biol 343: 101–114 [DOI] [PubMed] [Google Scholar]
  19. Hacker KJ, Johnson KA (1997) A hexameric helicase encircles one DNA strand and excludes the other during DNA unwinding. Biochemistry 36: 14080–14087 [DOI] [PubMed] [Google Scholar]
  20. Hingorani MM, Patel SS (1993) Interactions of bacteriophage T7 DNA primase/helicase protein with single-stranded and double-stranded DNAs. Biochemistry 32: 12478–12487 [DOI] [PubMed] [Google Scholar]
  21. Hingorani MM, Patel SS (1996) Cooperative interactions of nucleotide ligands are linked to oligomerization and DNA binding in bacteriophage T7 gene 4 helicases. Biochemistry 35: 2218–2228 [DOI] [PubMed] [Google Scholar]
  22. Hingorani MM, Washington MT, Moore KC, Patel SS (1997) The dTTPase mechanism of T7 DNA helicase resembles the binding change mechanism of the F1-ATPase. Proc Natl Acad Sci USA 94: 5012–5017 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Jeong YJ, Kim DE, Patel SS (2002) Kinetic pathway of dTTP hydrolysis by hexameric T7 helicase–primase in the absence of DNA. J Biol Chem 277: 43778–43784 [DOI] [PubMed] [Google Scholar]
  24. Jeong YJ, Levin MK, Patel SS (2004) The DNA-unwinding mechanism of the ring helicase of bacteriophage T7. Proc Natl Acad Sci USA 101: 7264–7269 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Jezewska MJ, Rajendran S, Bujalowska D, Bujalowski W (1998) Does single-stranded DNA pass through the inner channel of the protein hexamer in the complex with the Escherichia coli DnaB helicase? Fluorescence energy transfer studies. J Biol Chem 273: 10515–10529 [DOI] [PubMed] [Google Scholar]
  26. Johnson DS, Bai L, Smith BY, Patel SS, Wang MD (2007) Single-molecule studies reveal dynamics of DNA unwinding by the ring-shaped T7 helicase. Cell 129: 1299–1309 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Kaplan DL, O'Donnell M (2002) DnaB drives DNA branch migration and dislodges proteins while encircling two DNA strands. Mol Cell 10: 647–657 [DOI] [PubMed] [Google Scholar]
  28. Kim DE, Narayan M, Patel SS (2002) T7 DNA helicase: a molecular motor that processively and unidirectionally translocates along single-stranded DNA. J Mol Biol 321: 807–819 [DOI] [PubMed] [Google Scholar]
  29. Liao JC, Jeong YJ, Kim DE, Patel SS, Oster G (2005) Mechanochemistry of t7 DNA helicase. J Mol Biol 350: 452–475 [DOI] [PubMed] [Google Scholar]
  30. Lohman TM, Bjornson KP (1996) Mechanisms of helicase-catalyzed DNA unwinding. Annu Rev Biochem 65: 169–214 [DOI] [PubMed] [Google Scholar]
  31. Lucius AL, Maluf NK, Fischer CJ, Lohman TM (2003) General methods for analysis of sequential ‘n-step' kinetic mechanisms: application to single turnover kinetics of helicase-catalyzed DNA unwinding. Biophys J 85: 2224–2239 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Mallik R, Carter BC, Lex SA, King SJ, Gross SP (2004) Cytoplasmic dynein functions as a gear in response to load. Nature 427: 649–652 [DOI] [PubMed] [Google Scholar]
  33. Martinez-Senac MM, Webb MR (2005) Mechanism of translocation and kinetics of DNA unwinding by the helicase RecG. Biochemistry 44: 16967–16976 [DOI] [PubMed] [Google Scholar]
  34. Matson SW, Bean DW, George JW (1994) DNA helicases: enzymes with essential roles in all aspects of DNA metabolism. BioEssays 16: 13–22 [DOI] [PubMed] [Google Scholar]
  35. Matson SW, Richardson CC (1983) DNA-dependent nucleoside 5′-triphosphatase activity of the gene 4 protein of bacteriophage T7. J Biol Chem 258: 14009–14016 [PubMed] [Google Scholar]
  36. Matson SW, Tabor S, Richardson CC (1983) The gene 4 protein of bacteriophage T7. Characterization of helicase activity. J Biol Chem 258: 14017–14024 [PubMed] [Google Scholar]
  37. Myong S, Bruno MM, Pyle AM, Ha T (2007) Spring-loaded mechanism of DNA unwinding by hepatitis C virus NS3 helicase. Science 317: 513–516 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Noble JE, Wang L, Cole KD, Gaigalas AK (2005) The effect of overhanging nucleotides on fluorescence properties of hybridising oligonucleotides labelled with Alexa-488 and FAM fluorophores. Biophys Chem 113: 255–263 [DOI] [PubMed] [Google Scholar]
  39. Oster G, Wang H (2000) Reverse engineering a protein: the mechanochemistry of ATP synthase. Biochim Biophys Acta 1458: 482–510 [DOI] [PubMed] [Google Scholar]
  40. Patel SS, Hingorani MM (1993) Oligomeric structure of bacteriophage T7 DNA primase/helicase proteins. J Biol Chem 268: 10668–10675 [PubMed] [Google Scholar]
  41. Patel SS, Picha KM (2000) Structure and function of hexameric helicases. Annu Rev Biochem 69: 651–697 [DOI] [PubMed] [Google Scholar]
  42. Perkins TT, Li HW, Dalal RV, Gelles J, Block SM (2004) Forward and reverse motion of single RecBCD molecules on DNA. Biophys J 86: 1640–1648 [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Picha KM, Patel SS (1998) Bacteriophage T7 DNA helicase binds dTTP, forms hexamers, and binds DNA in the absence of Mg2+. The presence of dTTP is sufficient for hexamer formation and DNA binding. J Biol Chem 273: 27315–27319 [DOI] [PubMed] [Google Scholar]
  44. Singleton MR, Dillingham MS, Wigley DB (2007) Structure and mechanism of helicases and nucleic acid translocases. Annu Rev Biochem 76: 23–50 [DOI] [PubMed] [Google Scholar]
  45. Stano NM, Jeong YJ, Donmez I, Tummalapalli P, Levin MK, Patel SS (2005) DNA synthesis provides the driving force to accelerate DNA unwinding by a helicase. Nature 435: 370–373 [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Straume M, Johnson ML (1992) Monte Carlo method for determining complete confidence probability distributions of estimated model parameters. Methods Enzymol 210: 117–129 [DOI] [PubMed] [Google Scholar]
  47. Tomko EJ, Fischer CJ, Niedziela-Majka A, Lohman TM (2007) A nonuniform stepping mechanism for E. coli UvrD monomer translocation along single-stranded DNA. Mol Cell 26: 335–347 [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Torimura M, Kurata S, Yamada K, Yokomaku T, Kamagata Y, Kanagawa T, Kurane R (2001) Fluorescence-quenching phenomenon by photoinduced electron transfer between a fluorescent dye and a nucleotide base. Anal Sci 17: 155–160 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Material

emboj2008100s1.doc (3.8MB, doc)

Articles from The EMBO Journal are provided here courtesy of Nature Publishing Group

RESOURCES