Quantitative methods to study helicase, DNA polymerase, and exonuclease coupling during DNA replication

Abstract

Genome replication is accomplished by highly regulated activities of enzymes in a multi-protein complex called the replisome. Two major enzymes, DNA polymerase and helicase, catalyze continuous DNA synthesis on the leading strand of the parental DNA duplex while the lagging strand is synthesized discontinuously. The helicase and DNA polymerase on their own are catalytically inefficient and weak motors for unwinding/replicating double-stranded DNA. However, when a helicase and DNA polymerase are functionally and physically coupled, they catalyze fast and highly processive leading strand DNA synthesis. DNA polymerase has a 3’-5’ exonuclease activity, which removes nucleotides misincorporated in the nascent DNA. DNA synthesis kinetics, processivity, and accuracy are governed by the interplay of the helicase, DNA polymerase, and exonuclease activities within the replisome. This chapter describes quantitative biochemical and biophysical methods to study the coupling of these three critical activities during DNA replication. The methods include real-time quantitation of kinetics of DNA unwinding-synthesis by a coupled helicase-DNA polymerase complex, a 2-aminopurine fluorescence-based assay to map the precise positions of helicase and DNA polymerase with respect to the replication fork junction, and a radiometric assay to study the coupling of DNA polymerase, exonuclease, and helicase activities during processive leading strand DNA synthesis. These methods are presented here with bacteriophage T7 replication proteins as an example but can be applied to other systems with appropriate modifications.

1. Introduction

The general mechanism of DNA replication is conserved in prokaryotes and eukaryotes (Li & O’Donnell, 2019; Yang, Seidman, Rupp, & Gao, 2019). Minimally, DNA replication requires four basic enzymatic activities: a DNA helicase to catalyze the unwinding of the parental double-stranded (ds) DNA into single-stranded (ss) DNA, a primase to make RNA primers de novo from nucleoside triphosphate (NTP) substrates, DNA polymerase to elongate the primer and copy the parental DNA strands in the 5’–3’ direction using dNTPs as substrates, and a 3’–5’ exonuclease to proofread mistakes made by the polymerase (Benkovic, Valentine, & Salinas, 2001; Burgers & Kunkel, 2017; Hamdan & Richardson, 2009). Due to the anti-parallel arrangement of dsDNA strands and the fixed directionality of DNA synthesis by the polymerase, one of the parental strands, called the leading strand, is copied continuously by the action of the helicase and leading strand polymerase (Ogawa & Okazaki, 1980). The complementary lagging strand is copied as Okazaki fragments discontinuously through frequent priming events by the primase and elongation by the lagging strand polymerase (Duderstadt et al., 2016; Ogawa & Okazaki, 1980; Okazaki, Okazaki, Sakabe, Sugimoto, & Sugino, 1968; Pandey et al., 2009). The hexameric ring-shaped helicases are the central organizers of the DNA replication machinery (Donmez & Patel, 2006; O’Donnell & Li, 2018; Yang et al., 2019). The replicative helicase interacts with the leading and lagging strand DNA polymerases, primase, and accessory proteins such as processivity factor and single-strand DNA binding protein (SSB) (Benkovic & Spiering, 2017; Burgers & Kunkel, 2017; Hamdan & Richardson, 2009; Kose, Xie, Cameron, Strycharska, & Yardimci, 2020; Li & O’Donnell, 2019; Manosas, Spiering, Ding, Croquette, & Benkovic, 2012; Nandakumar, Pandey, & Patel, 2015). Functional and physical interactions with the partner proteins are essential in regulating the helicase and the interacting proteins. This chapter discusses ensemble biochemical and biophysical methods to study the functional coupling between replicative helicase, polymerase, and exonuclease activities during leading strand DNA synthesis.

As motor proteins, helicases utilize NTP hydrolysis to translocate unidirectionally on nucleic acid substrates (Abdelhaleem, 2010; Bourgeois, Mortreux, & Auboeuf, 2016; Patel & Picha, 2000). Their ability to translocate along ssDNA is critical for assisting the DNA polymerase in catalyzing strand displacement leading strand DNA synthesis. Nearly all replicative helicases are poor at unwinding duplex DNA on their own. The rates and processivities of the replicative helicases in isolation are often lower than those observed within the replisomes (Delagoutte & von Hippel, 2002, 2003; Patel, Pandey, & Nandakumar, 2011). Biochemical studies show that bacteriophage T7 and T4 helicases are mostly passive unwindases, and they rely on thermal melting of the base-pairs at the fork junction to separate the strands of the duplex DNA (Johnson, Bai, Smith, Patel, & Wang, 2007; Lionnet, Spiering, Benkovic, Bensimon, & Croquette, 2007; Manosas, Xi, Bensimon, & Croquette, 2010). Fluorescence, optical, and magnetic tweezers-based single-molecule techniques reveal frequent helicase “slippage” and rezipping events limiting their rates and processivity (Manosas, Spiering, Ding, Croquette, & Benkovic, 2012; Schlierf, Wang, Chen, & Ha, 2019; Sun et al., 2011, 2018). Replicative DNA polymerases are efficient at catalyzing DNA synthesis on ssDNA templates but cannot catalyze processive strand displacement synthesis on dsDNA templates (Manosas et al., 2012; Nandakumar et al., 2015; Pandey & Patel, 2014; Yuan & McHenry, 2009). How do the slow and inefficient helicase and DNA polymerase accomplish the task of catalyzing fast and highly processive leading strand synthesis during replication? The answer lies in the physical and functional coupling of the helicase and DNA polymerase activities, which mutually benefits the two proteins.

Prokaryotic and eukaryotic replicative helicases have an opposite polarity of translocation on the DNA, giving rise to different leading strand replisome architectures (Fig. 1). The RecA-type prokaryotic (bacteria and bacteriophages) and mitochondrial replicative helicases translocate and unwind DNA in the 5’ to 3’ direction. This class of helicases bind on the opposite lagging strand to the DNA polymerase bound on the leading strand. In contrast, AAA⁺-type eukaryotic replicative helicases (and eukaryotic viruses) unwind DNA in the 3’ to 5’ direction and bind on the same leading strand as the DNA polymerase. Despite these differences, recent structures of phage T7 and eukaryotic replisomes show conserved mechanisms of interactions between the helicase and leading strand DNA polymerase. In both cases, the C-terminal domain of the helicase, harboring the NTPase and helicase activities, are involved in interactions with the leading strand DNA polymerase (Li & O’Donnell, 2019; Yang et al., 2019). In the case of the T7 replisome, the helicase subunits are physically coupled to the leading strand DNA polymerase via their C-terminal tail regions (Foster et al., 2019; Gao et al., 2019). This physical coupling provides stable positioning of the helicase and polymerase at the fork junction, which is critical for efficient strand displacement synthesis of the leading strand without slippages and fork reannealing. Often replication hurdles such as reduced nucleotide levels, lesions on the DNA, protein blocks, or secondary structures uncouple helicase-polymerase, prompting slippage and junction base-pair reannealing. These events result in the fraying of the primer-end and excessive excision of the nascent DNA by the exonuclease activity (Singh et al., 2020). Measurements of the polymerase to exonuclease activity ratio provide a way to assess helicase-polymerase coupling.

Fig. 1.

Architectures of prokaryotic (A) and eukaryotic (B) replisomes show similar modular structures with C tier of replicative helicase interacting with leading strand DNA polymerase and N tier with lagging strand DNA polymerase or primase. The polarity of the helicase rings is opposite in the two cases.

The poor unwinding and synthesis activities of the replicative helicase and DNA polymerase, respectively, in isolation, and the mutual dependency ensure that inappropriate unwinding or copying of DNA regions does not occur in an unregulated manner. The partnering of the replicative helicase with the leading strand DNA polymerase increases the replication rate and processivity, eliminates nucleotide sequence dependency, and prevents rezipping of the unwound DNA (Delagoutte & von Hippel, 2002; Georgescu et al., 2014; Kim, Dallmann, McHenry, & Marians, 1996; Manosas, Spiering, Ding, Bensimon, et al., 2012; Manosas, Spiering, Ding, Croquette, & Benkovic, 2012; Pandey & Patel, 2014; Stano et al., 2005; Sun et al., 2018). Here, we describe three in vitro ensemble methods to study the functional coupling between helicase, DNA polymerase, and exonuclease activities on synthetic replication fork substrates mimicking leading strand DNA synthesis. The methods developed using bacteriophage T7 proteins apply to other systems with appropriate modifications.

A fluorescence-based method to quantify the single turnover kinetics of DNA unwinding-synthesis by the coupled T7 helicase-DNA polymerase complex is detailed (Nandakumar et al., 2015). As the isolated helicase is poor at unwinding the fork DNA substrate, and the T7 DNA polymerase (1:1 complex of T7 gp5 and its processivity factor, Escherichia coli thioredoxin) cannot synthesize beyond 5–7 base-pairs on its own, the separation of the two strands by the coupled helicase-DNA polymerase complex essentially provides the rate of unwinding-synthesis. We also describe a 2-aminopurine (2-AP) fluorescence-based method that provides structural information on helicase and DNA polymerase positions with respect to the replication fork junction. This is a powerful technique that can be applied to quantify the contributions made by individual proteins in the unwinding of the fork junction (Nandakumar et al., 2015). Finally, we describe a radiometric, thin layer chromatography (TLC)-based method to study DNA polymerase and exonuclease coupling during processive leading strand synthesis on a minicircle DNA substrate (Singh et al., 2020). This method provides valuable information on the polymerase-exonuclease active-site switching and its dependence on helicase-DNA polymerase coupling as it allows simultaneous measurement of polymerase and exonuclease activities in the same reaction. The methods described here are fast, less tedious than gel-based, and can be adapted to design high-throughput assays.

2. Real-time quantitation of unwinding-synthesis by coupled helicase-DNA polymerase

A fork DNA substrate is designed with 5’ and 3’ single-stranded tails and a 40 bp duplex DNA region, mimicking the DNA replication fork (Fig. 2A). Based on several biochemical trials, we chose 35 nucleotide long 5’ tail for optimal assembly and unwinding of the fork DNA by T7 helicase. The trials consisted of changing the length of the 5’ tail and measuring the unwinding yield and rate. Best yields/rates were obtained when the 5’ tail was > 25 nt. Similarly, on the 3’ tail of the fork DNA, a 24-mer primer was annealed to assemble the T7 DNA polymerase. The crystal structure of T7 DNA polymerase on the DNA shows interactions with ~21 bp of dsDNA (Doublie, Tabor, Long, Richardson, & Ellenberger, 1998). At the blunt end of the fork DNA duplex, a fluorophore-quencher pair was positioned; fluorescein was attached at the 3’ end and BHQ-1 quencher at the 5’ end (Fig. 2A). The proximity of the quencher to fluorescein keeps the fluorescence intensity low. The fluorophore-quencher typically works well, and FRET pairs are not necessary.

Fig. 2.

Real-time measurement of unwinding-synthesis by coupled helicase-DNA polymerase complex (Nandakumar et al., 2015). (A) Schematic representation of the experimental design used to determine rates of unwinding-synthesis by T7 helicase and DNA polymerase complex. The leading strand complex is assembled on a primed fork DNA substrate with “L” bp long DNA duplex. Fluorescence signal by the fluorescein probe increases in response to the separation of the two strands of the duplex DNA. (B) A schematic representation of the experiment in the stopped-flow rapid mixing device (KinTek Corp). (C) A representative time course shows the increase in fluorescence from leading strand synthesis reaction by T7 helicase and DNA polymerase. The initial lag (L_t) reflects the time the helicase-DNA polymerase complex takes to reach the blunt-end of the fork DNA. The time corresponding to the mid-point of fluorescence transition, t_m, can be used to obtain the rate of unwinding-synthesis reaction. (D) Fluorescence time traces from leading strand synthesis reactions by T7 helicase and DNA polymerase conducted at different dTTP concentrations and fixed 50μM dVTPs. (E) Rate constants of fork DNA (50% GC) unwinding by T7 helicase at increasing dTTP concentrations. The solid line shows the fit to a hyperbola. (F) Dependence of the helicase unwinding rate constant on the GC-content of the dsDNA region of the fork DNA substrate at a fixed 1 mM dTTP concentration. (G) Rate constants of fork DNA (50% GC) unwinding-synthesis reactions by DNA polymerase and E. coli SSB as a function of dNTP concentration. The solid line shows the fit to a hyperbola. (H) Dependence of unwinding-synthesis rates by DNA polymerase and SSB complex on the GC-content of the dsDNA region of the fork DNA substrate. All reactions were performed with 5 μM dNTPs. (I) Rate constants of fork DNA (50% GC) unwinding-synthesis reactions by the coupled helicase-DNA polymerase complex as a function of dVTP concentration at a fixed dTTP concentration of 1 mM. The solid line shows the fit to a hyperbola. (J) Rate constants of unwinding-synthesis reactions by coupled helicase-DNA polymerase complex on fork DNA substrates with varying GC content conducted with 1 mM dTTP and 5 μM dVTP.

The DNA unwinding-synthesis reactions on the short fork DNA substrates occur in the milliseconds time range; hence, reactions are carried out in a rapid mixing stopped-flow device (Fig. 2B). A stopped-flow apparatus consists of two motor-powered syringes, a rapid mixing chamber, a flow cell connected to a fluorescence detector, and a stop-syringe for controlling the flow of the reaction in the flow cell. The temperature of the syringes, mixing chamber, and flow cell are precisely controlled with the help of a circulating water bath. A pre-assembled leading strand complex containing the fork DNA, helicase, dTTP, and DNA polymerase is filled in one syringe. T7 helicase binds stably to the fork substrate as a hexamer only in the presence of dTTP, and we leave out Mg²⁺ from the buffer to prevent the unwinding of the DNA fork (Ahnert & Patel, 1997; Picha & Patel, 1998). Reactions are initiated with the addition of Mg²⁺, 100-fold excess of dT₉₀ ssDNA over the fork DNA substrate, and the rest of the three dNTPs (dCTP, dGTP and dATP, referred to as dVTPs hereafter) from the other syringe. The dT₉₀ ssDNA traps the free or dissociated helicase molecules from rebinding to the fork DNA, ensuring single turnover conditions for rate measurements.

The unwinding-synthesis reaction is measured in real-time on the stopped-flow apparatus by following the increase in fluorescence intensity of fluorescein with time (1–2ms onward). A typical time course shows an initial lag phase, representing the helicase-DNA polymerase complex’s time to reach the blunt-end of the fork DNA. The lag phase is followed by a steep, exponential increase in fluorescence signal and a plateau representing the complete separation of the two strands (Fig. 2C). The transient dip in fluorescence intensity at the reaction end is due to the helicase going over the fluorescein probe. The lag time before the steep fluorescence increase should directly correlate with the length of the DNA duplex used. With minor modifications, the method can be used to measure rates of DNA unwinding and strand-displacement synthesis by isolated helicases and DNA polymerases, respectively (Nandakumar et al., 2015; Ozes, Feoktistova, Avanzino, Baldwin, & Fraser, 2014).

The desired parameter is the rate of DNA unwinding-synthesis, which is expressed in base-pairs unwound-synthesized per second (bp/s). The average unwinding-synthesis rate can be obtained by fitting the fluorescence time traces into a stepping-mechanism model using a publicly available gfit program, described previously (Levin, Hingorani, Holmes, Patel, & Carson, 2009; Pandey, Levin, & Patel, 2010), or using other software such as the KinTek Explorer (Chib, Byrd, & Raney, 2016; Johnson, 2009; Johnson, Simpson, & Blom, 2009). The average rate of unwinding-synthesis can also be estimated from the time required to reach the midpoint of fluorescence transition (t_m) and multiplying 1/t_m and dsDNA length (Fig. 2C). One can measure the rate of unwinding-synthesis at single-nucleotide resolution by conducting leading strand synthesis reactions using a fork DNA substrate with a fluorescently labeled or radiolabeled DNA primer in a quenched-flow rapid mixing apparatus and resolving the DNA products by sequencing PAGE (Pandey & Patel, 2014). Although the technique provides high-esolution information, it is not high-throughput and requires substantial time and effort compared to the stopped-flow method described above.

Factors such as dTTP concentration, dVTP concentration, and GC content of the duplex region of DNA substrates will affect the observed rate of unwinding-synthesis reactions. For example, the kinetics of unwinding-synthesis by T7 helicase-DNA polymerase complex depends on dTTP concentration, requiring at least 500 μM dTTP to produce optimum unwinding-synthesis (Fig. 2D). Measuring the rates of helicase unwinding, DNA polymerase strand displacement synthesis, and unwinding-synthesis by the helicase-DNA polymerase complex as a function of these parameters provides mechanistic insights into the coupling of the two proteins. T7 helicase in isolation has poor unwinding activity (Fig. 2E), and the rates depend on the GC content of the duplex DNA (rate of unwinding decreases from 62 bp/s for 5% GC fork to 8.3 bp/s for 100% GC fork) (Fig. 2F). Likewise, T7 DNA polymerase on its own is unable to extend the DNA primer beyond a few base-pairs on a fork DNA substrate. E. coli SSB or T7 gp2.5 (T7 SSB) supports strand displacement synthesis (Fig. 2G), but the rate depends on the GC content (Fig. 2H) (Nandakumar et al., 2015). The GC dependency indicates that helicase and DNA polymerase on their own are primarily passive motors. When T7 helicase and DNA polymerase are combined, they synergistically enhance each other’s activity to catalyze unwinding-synthesis at a fast rate of ~140 bp/s (Fig. 2I). Moreover, the coupled action of the two proteins eliminates the dependency of the rate of unwinding-synthesis on GC content (Fig. 2J), which is crucial for the accurate replication of various regions of the genome. Although SSB supports DNA polymerase’s strand displacement synthesis, high dNTP concentrations must be added to reach the maximal rate of ~140 bp/s (dNTP K_m with SSB is 124 μM) (Fig. 2G). On the other hand, the helicase-coupled strand displacement synthesis occurs at a much lower dNTP concentration (dVTP K_m with helicase is ~4 μM), which appears to be more physiologically relevant (Fig. 2I). These assays are highly convenient for measuring the dNTP K_m values, which provide physiologically meaningful mechanistic information.

2.1. Equipment

• Stopped Flow instrument equipped with fluorescence detection capabilities and KinTek software (KinTek Corp, Snow Shoe, PA).

• Water baths to maintain the stopped-flow syringes, mixing chamber, flow cell, and lamp at desired temperatures.

2.2. Buffers and reagents

• Buffer A: 50mM Tris–HCl, pH 7.6, 40mM NaCl and 10% glycerol

• Ethylenediaminetetraacetic acid (EDTA)

• Dithiothreitol (DTT)

• Fork DNA substrates prepared by annealing appropriate DNA oligos

• A dT₉₀ or similar ssDNA

• Magnesium chloride solution

• dNTPs

• Purified T7 helicase and T7 DNA polymerase proteins (Donlin, Patel, & Johnson, 1991; Pandey et al., 2009).

2.3. Procedure

1. Set the temperature of the stopped-flow reaction chamber to 18 °C.

2. Set up the experiment with an excitation wavelength of 480nm and fluorescence emission with a long pass 515 nm cut-off filter.

3. Prepare “Mixture A” by mixing Buffer A, 1.5 mM EDTA, 5 mM DTT, 20 nM fork DNA substrate, 40 nM T7 helicase, 40 nM T7 DNA polymerase, and 2000 μM dTTP.

4. Mix MgCl₂, 100 μM dVTPs, and 2 μM dT₉₀ ssDNA in reaction buffer A to prepare “Mixture B.”

5. Pre-incubate Mixtures A and B separately at 18 °C for 10min.

6. Load the Mixtures A and B in separate loading syringes.

7. Trigger the rapid mixing of A and B in the mixing chamber and measure the fluorescence change starting in the millisecond range to up to 5 s. Change the recording time depending upon the time it takes to complete the reaction. Measure at least three time courses for each reaction and use the averaged trace to quantitate the rates.

2.4. Analysis

1. The average rate of unwinding-synthesis (bp/s) can be obtained by fitting the time traces into a stepping-mechanism model using the publicly available gfit program, described previously (Levin et al., 2009; Pandey et al., 2010). The data fitting into a multistep unwinding-synthesis reaction can also be carried out using the KinTek Explorer software to obtain the stepping rate (Chib et al., 2016; Johnson, 2009; Johnson et al., 2009).

2. A rough estimate of the average unwinding-synthesis rate constant (k) on a fork DNA with a duplex length of L bp can be obtained from the time required to reach the midpoint of fluorescence transition (t_m) (Fig. 2C):

   $$k=1∕t_m\times L$$ (1)

   We have previously shown that the rate constants determined by the two methods are comparable (Nandakumar & Patel, 2016).

3. Average rate constants from the reactions are plotted as a function of dNTP concentration and are fitted to hyperbolic trend to obtain dNTP K_m and k_cat using the following equation:

   $$k=k_{\text{cat}}.[\text{dNTP}]∕(K_m+[\text{dNTP}])$$ (2)

2.5. Notes

1. To make the fork DNA substrate, DNA primer, template DNA and fluorescein labeled non-template DNA flap are mixed in a ratio of 1:1:0.9 in an annealing buffer consisting of 1X TE and 100mM NaCl. A slightly lesser concentration of fluorophore-labeled oligonucleotide is used to keep the baseline fluorescence intensity low. DNA annealing is carried out by heating the mixed strands at 95 °C for 2 min on a heat-block and allowing the heat block to gradually cool down to room temperature by turning the heat-block off. Proper annealing of the fork DNA substrate should be confirmed by running the annealed DNA on a native PAGE. The fluorescein-labeled DNA strand and the duplex DNA can be visualized from the fluorescein fluorescence or stained with SYBR Green to view all strands.

2. The 5’ and 3’ tail lengths in the fork DNA should be optimized for the helicase-DNA polymerase complex based on their known footprints on the DNA.

3. Kinetics of nucleotide incorporation by T7 replisome is extremely fast which makes accurate rate measurements difficult at physiologically relevant temperatures. Therefore, 18 °C is used to slow down the reaction rates and to make the kinetics measurable. The measurement temperatures can be changed for other replication systems based on the kinetics of the replication proteins, deadtime of the stopped-flow apparatus, etc.

4. In Fig. 2A, fluorescein was attached at the 3’ end and BHQ-1 quencher at the 5’ end on the opposite strand. A sequence of 3–4 dG residues in place of BHQ-1 also efficiently quenches the fluorescence of fluorescein and is a cost-effective alternative to BHQ-1.

5. Fluorescence increase with gentle slope may suggest problems in replisome assembly, variable rates of unwinding-synthesis by heterogeneous complexes, and interrupted DNA synthesis.

6. Experiments should be carried out with and without the dT trap. We have experimentally confirmed that dT₉₀ ssDNA does not affect the T7 replication complex. However, it can deleteriously affect the kinetic measurements of other replication systems by dissociating the pre-assembled replisomes. Thus, an experimental control without a dT ssDNA trap should be conducted.

3. Aminopurine-based method to study cooperative base-pair melting at replication fork junction

It is essential to know the exact position of the helicase and DNA polymerase relative to each other and the replication fork junction in the assembled leading strand replisome complex. This allows a proper understanding of the contributions made by the helicase and DNA polymerase in unwinding the junction base-pairs. 2-AP is a fluorescent analog of adenine and guanine that base-pairs with thymine and cytosine (Law, Eritja, Goodman, & Breslauer, 1996; Ward, Reich, & Stryer, 1969). The 2-AP fluorescence is quenched when it is base-paired and has stacking interactions with the neighboring bases. This property is used to monitor enzyme-catalyzed DNA unwinding reaction and to study the interaction of helicase with the junction base-pairs in the fork DNA (Jose, Datta, Johnson, & von Hippel, 2009; Reha-Krantz, 2009). By strategically incorporating a 2-AP residue at different positions around the junction base-pair of the fork DNA, one can map the precise position of the helicase and DNA polymerase with respect to the 2-AP base-pair. Furthermore, the method reveals the roles of individual helicase and DNA polymerase proteins and their coupled action in unwinding the junction base-pairs.

A DNA fork substrate can be designed with a single 2-AP residue on the lagging strand base-paired to a dT on the leading strand at the fork junction (junction 2-AP) (Fig. 3A). The position of the primer-template junction from the fork junction can be changed by shortening the primer and creating 1–3 nt gaps between the primer-end and the junction base-pair (Fig. 3A). This reveals the position of the DNA polymerase active site with respect to the junction base-pair. A different type of DNA fork can be designed with an internal 2-AP probe at a position downstream to the junction base-pair (internal 2-AP) (Fig. 3A). Such probe position reveals the ability of helicase and DNA polymerase to melt the internal base-pairs upon binding to the fork DNA. Fork DNA substrates are prepared by annealing appropriate oligonucleotides.

Fig. 3.

2-AP fluorescence assay to determine the relative contribution of the T7 helicase and T7 DNA polymerase to the unwinding of fork DNA duplex region (Nandakumar et al., 2015). (A) Schematic of 2-AP labeled fork DNA substrates. A single 2-AP probe (red) is incorporated in the lagging strand at the fork junction base-pair (Junction 2-AP) or the base-pair at 1-nt downstream of the fork junction (Internal 2-AP). The DNA primer length is changed to introduce a 0 to 3-nt gap between the primer-end and the fork junction. (B) Fluorescence intensities (in arbitrary units) of 2-AP containing replication fork DNA with and without T7 DNA polymerase. (C) Representative 2-AP fluorescence signals measured at 370 nm in the presence and absence of T7 DNA polymerase. (D) Fold change in “junction” 2-AP fluorescence upon binding of T7 DNA polymerase to the fork DNA relative to free DNA (shown as black line). A significant increase in 2-AP fluorescence is only observed when the DNA fork has no gap and 1-nt gap between primer-end and junction base-pair. (E) Fold change in “junction” 2-AP fluorescence in response to binding of T7 helicase to the fork DNA. The increase in fluorescence signal is independent of the gap size. (F) Fold change in “internal” 2-AP fluorescence in response to binding of T7 helicase to the fork DNA. No increase in 2-AP fluorescence is observed. (G) Fold change in “junction” 2-AP fluorescence in response to T7 helicase and DNA polymerase binding to the fork DNA. The enhanced 2-AP fluorescence is observed with no gap and 1-nt gap fork DNA substrates. (H) Fold change in “internal” 2-AP fluorescence in response to binding of T7 DNA polymerase alone or the T7 helicase-DNA polymerase complex to the fork DNA. Enhanced 2-AP fluorescence was observed with no gap and 1-nt gap fork DNA substrates.

Two fluorescence intensity measurements are made, one with fork DNA alone and the other with fork DNA incubated with DNA polymerase, helicase, or their combination (Fig. 3B and C). A parallel experiment is carried out with an unlabeled replication fork substrate to correct for the protein fluorescence changes resulting from the DNA binding. Optimal protein concentration that produces the maximum fluorescence intensity change must be determined empirically by carrying out experiments are various protein concentrations. The 2-AP fluorescence change in response to protein binding is expressed as “fold change.” When a “no gap” fork DNA is used, addition of T7 DNA polymerase to the fork DNA increases the fluorescence intensity by twofold over 2-AP fork DNA alone, demonstrating that the DNA polymerase unwinds the junction base-pair to accommodate itself at the primer-end (Fig. 3D). Interestingly, the increase in fluorescence intensity in response to DNA polymerase binding is also observed with the “1-nt gap” fork DNA, slightly with 2-nt gap but not on 3-nt gap substrate (Fig. 3D). This indicates that the DNA polymerase unwinds two base-pairs at the junction to create a 2-nt gap between the primer-end and junction base-pair to optimally accommodate itself on the replication fork.

Similar experiments with T7 helicase were conducted with dTMP-PCP, a non-hydrolysable analog of dTTP, needed to stabilize the helicase hexamer on the fork DNA without unwinding the fork. Other analogs such as dTDP.BeF₃ or dTDP.AlF₃ can also be used. T7 helicase also increases the fluorescence intensity of 2-AP at the junction base-pair upon binding to the fork DNA, like the DNA polymerase (Fig. 3E). However, this increase is observed in all gap-sized forks but only when the 2-AP probe is at the junction base-pair and not at an internal position (Fig. 3E and F) (Nandakumar et al., 2015). This indicates that T7 helicase binds at the fork junction and can melt only the junction base-pair.

We observe the greatest increase in 2-AP fluorescence when both helicase and DNA polymerase are added to the fork DNA (Fig. 3G). The enhanced melting of the 2-AP at the junction base-pair is observed in the no-gap substrate, and the change is much less when the gap size is longer than 1-nt (Fig. 3G). Enhanced unwinding by helicase-DNA polymerase is apparent when the 2-AP is placed internally one nucleotide downstream of the fork junction on the no-gap and 1-nt gap substrate and not seen in the 2-nt gap substrate (Fig. 3H). These experiments indicate that helicase and DNA polymerase unwind one to two base-pairs of the fork junction to accommodate them at the replication fork of the leading strand. Such studies could be instrumental in designing fork DNA scaffolds for structural studies of the replisome. For example, the replisome architecture was confirmed by the cryoEM structure of the T7 leading strand complex (Gao et al., 2019). The agreement between the cryoEM and the biochemical models showcases the effectiveness of 2-AP based mapping method in deciphering the architecture of the helicase-DNA polymerase complex at the leading strand fork.

3.1. Equipment

• FluoroMax-4 (Horiba, Ltd., Kyoto, Japan) or equivalent spectrofluorometer with temperature control

• Heat block to keep the reaction mixture in microcentrifuge tubes at the desired temperature

3.2. Buffers and reagents

• Buffer B: 50mM Tris–HCl, pH 7.6, 40mM NaCl, 10mM MgCl₂, 5mM DTT.

• Annealed fork DNA substrates.

• dTMP-PCP or dTDP.AlF₃.

• Purified T7 helicase and DNA polymerase proteins.

3.3. Procedure

1. Prepare the samples by mixing 100 nM of fork DNA substrate, 200 nM T7 helicase (where present), 10 μM dTMP-PCP, and 200nM T7 DNA polymerase (where present) in buffer B. Pre-incubate the samples at 25 °C for 10min before starting the fluorescence measurements.

2. Transfer the sample in a quartz cuvette placed in a cuvette holder of spectrofluorometer set at 25 °C.

3. Carry out the equilibrium fluorescence measurements by exciting the sample at 315 nm (at 2 mm slit width) and measuring the emission at 370nm (at 6mm slit width).

4. Make corrections for the observed fluorescence for the buffer background and the protein bound to the unlabeled replication fork substrate. Check for the inner filter effect and correct it if needed (see the following section on data analysis).

3.4. Analysis of the 2-AP fluorescence data

Unstacking of the 2-AP base by the bound DNA polymerase or helicase results in an increase in fluorescence. The observed fluorescence intensity is corrected for background from the buffer, fluorescence changes due to protein binding to the DNA independent of the fluorescence change brought by 2-AP unstacking, and volume and inner filter volume effects. The fluorescence changes caused by protein binding to the DNA can be determined by performing equilibrium fluorescence measurements with mock DNA substrates lacking the 2-AP probe. Fluorescence data are corrected for inner filter and volume effects using the following equation:

$$F_c=Fx{V}_f∕V_o\times {10}^{0.5(A_x+A_m)}$$ (3)

where F_c is 2-AP fluorescence intensity corrected for volume and inner filter effects, F is observed 2-AP fluorescence intensity, V_f and V_o are total and original volume of the reaction mixture, respectively, A_x is the absorbance of the mixture at the excitation wavelength, and A_m is the absorbance at the emission wavelength.

Correction for protein binding is performed as follows:

$$F_{pc}=F_c-F_{mc}$$ (4)

Where, F_pc is the fluorescence intensity of 2AP DNA corrected for protein fluorescence change, inner filter effect, and volume, and F_mc is the fluorescence intensity obtained with mock DNA without the 2AP probe. The fluorescence change of 2AP DNA in response to protein binding is expressed as “fold change” and can be quantified (Eq. 5). Here F_{2AP DNA} is the fluorescence intensity of 2AP DNA in the absence of protein.

$$\text{Fold Change}=F_{pc}∕F_{2\text{AP DNA}}$$ (5)

3.5. Notes

1. The samples should be maintained at a preset temperature during the measurement to minimize the effect of temperature fluctuations.

2. The extent of quenching of 2-AP fluorescence by neighboring bases is dependent on sequence context in which 2-AP is incorporated (Jean & Hall, 2001; Jones & Neely, 2015; Larsen et al., 2004). This information can be used to strategically design the DNA substrates.

3. The 2-AP fluorescence intensity increase can be due to neighboring base-pair unstacking or specific amino acids interacting with the 2-AP base. Proper control experiments should be conducted to eliminate these possibilities.

4. The proteins and any nucleotides present in the reaction may contribute to the inner filter effect, and this must be checked for and corrected when necessary.

4. Simultaneous measurement of polymerase and exonuclease activities using minicircle fork DNA

Replicative DNA polymerases misincorporate with a frequency of 10⁻⁴ to 10⁻⁶ (Kunkel, 2004; Kunkel & Bebenek, 2000). Therefore, replicative DNA polymerases have a proofreading exonuclease activity that excises the 3’-end mismatched nucleotide in the nascent DNA to increase the accuracy of DNA synthesis. The exonuclease activity is either part of the same polypeptide as the DNA polymerase, like in T7 DNA polymerase, or a separate subunit, like in E. coli DNA polymerase III (Scheuermann & Echols, 1984). The polymerase and exonuclease active sites are typically 30–40Å apart (Beese, Derbyshire, & Steitz, 1993; Doublie et al., 1998). Therefore, 3–4 base-pairs at the primer-template junction must melt to partition the primer-end into the exonuclease site, which happens during misincorporation when DNA synthesis is inhibited. However, the probability of spontaneous melting/partitioning of a correctly matched primer-end is expected to be low, estimated to excise less than 0.1% of correctly incorporated nucleotides (Donlin et al., 1991). However, studies show that replicative DNA polymerases, including T7 DNA polymerase, excise close to 7% of correctly incorporated nucleotides into dNMPs during leading strand synthesis. This implies that primer-end shuttling between the exonuclease and polymerase sites happens more frequently than expected during DNA synthesis (Fersht, Knill-Jones, & Tsui, 1982; Singh et al., 2020). The recent study showed that high excision events are partly due to translocation hurdles encountered by the helicase-DNA polymerase during DNA replication (Singh et al., 2020). Therefore, measuring the polymerase to exonuclease ratio (Pol/Exo) provides a convenient way to monitor replication hurdles during DNA synthesis.

We describe a facile way to measure the kinetics of nucleotide incorporation and excision during leading strand DNA synthesis (Singh et al., 2020). There are many ways to measure the kinetics of single nucleotide incorporation during DNA synthesis, including gel-based methods and single-molecule tweezers and nanopore methods (Lieberman et al., 2010; Manosas, Spiering, Ding, Bensimon, et al., 2012; Pandey & Patel, 2014; Sun et al., 2015). However, these methods cannot measure the excision of single nucleotides during processive DNA synthesis. We use a 70bp minicircle fork DNA (Table 1) that supports efficient rolling circle DNA synthesis of kb-sized DNA products (Pandey et al., 2009). T7 helicase and DNA polymerase are assembled on the minicircle fork with a 40-nt long 5’ ssDNA tail (Fig. 4A). Reactions are started by adding a mixture of dVTPs and α-³²P-dGTP, Mg²⁺, and T7 gp2.5 and quenched with formic acid at pre-decided time-points. The radiolabeled dGMP incorporated into the nascent DNA, dGMP excised by the DNA polymerase’s exonucleolytic activity, dGDP produced from helicase’s dNTP hydrolysis activity, and the unused dGTP are simultaneously resolved on the PEI-cellulose TLC sheet (Fig. 4B). The spot intensities of dGTP, dGDP, incorporated dGMP, and excised dGMP are quantified using ImageQuant TL software to determine the amounts of nucleotide added to the synthesized DNA and excised during each reaction period. The time courses provide the rates of DNA synthesis and exonuclease reactions to obtain the Pol/Exo ratio (Fig. 4C and D).

Table 1.

Oligonucleotide sequences used for the radiometric TLC-based assay.

Oligonucleotide	Sequence
70 nucleotide minicircle DNA	5’ CACCATATCCTCGACCATCCCCAATATGGTCCATCAACCCTTCACCTCACTTCACTCCACTATACCACTC 3’
Primer for the 70 nucleotide minicircle DNA	5’ TTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTGAGTGGTATAGTGGAGTGAAGTGAGGTGAAGGGTTGATGGACCATATTGGGGATGGTCGAGGATATGGTG 3’
Linear 70 nucleotide ssDNA	/5’Phos/CACCATATCCTCATCAACCCTTCACGACCATCCCCAATATGGTCCCTCACTTCACTCCACTATACCACTC 3’
Splint DNA	5’ ATGGTGGAGTGG 3’

Fig. 4.

Thin layer chromatography (TLC)-based assay to measure the coupling between polymerase and exonuclease activities of T7 DNA polymerase during leading strand DNA synthesis (Singh et al., 2020). (A) Experimental design to simultaneously measure polymerase and exonuclease activities during leading strand synthesis. T7 helicase and DNA polymerase were assembled on a primed-minicircle DNA substrate to measure rolling circle leading strand synthesis. Asterisks represent incorporated, excised, and unused radiolabeled nucleotides. (B) Representative TLC image shows the spots from dGMP incorporated into the nascent DNA, dGMP excised by the DNA polymerase’s exonucleolytic activity, dGDP, and the unused dGTP. (C, D) Time courses of polymerase activity (C) and exonuclease activity (D) expressed as moles of nucleotides incorporated and excised per mole of leading strand complex. Experiments were performed at a constant 500 μM dTTP concentration and 150 μM dVTPs. (E) Pol/Exo ratios from leading strand DNA synthesis reactions conducted at a constant 500 μM dTTP concentration and increasing concentrations of dVTPs. (F, G) Moles of nucleotides incorporated (F) and excised (G) per mole of leading strand complex per second during leading strand synthesis is shown for reactions performed with 10 μM dVTPs and different dTTP concentrations. (H) Pol/Exo ratios for leading strand synthesis reactions are shown for reactions performed with different dTTP concentrations, as determined from (F, G). (I) Structure shows the basis for physical coupling between T7 helicase-DNA polymerase through interactions between the acidic C-tail of T7 helicase and the basic front patch of T7 gp5 (PDB entry 6P7E) (Foster et al., 2019). (J) Pol/Exo ratio during leading strand synthesis reactions is shown for reactions performed with WT T7 helicase and its C-tail deletion mutant. Reactions were conducted with 500 μM dTTP and 150 μM dVTPs.

The leading strand synthesis reaction by T7 helicase-polymerase on the minicircle shows a Pol/Exo ratio of ~15 under optimum nucleotide concentration conditions (Fig. 4E). Thus, T7 DNA polymerase excises one out of 15 correctly incorporated nucleotides from the nascent DNA. To understand the mechanistic basis for the high excision rate, one can measure the Pol/Exo ratio at increasing dNTPs. Partitioning of the primer-end from the polymerase site into the exonuclease site occurs from the pre-translocated state (Lieberman, Dahl, & Wang, 2014; Singh et al., 2020), and high concentrations of dNTPs stabilize the post-translocated state. If excision reactions are simply due to sub-saturating levels of dNTPs, then high dNTPs should reduce the exonuclease activity. On the other hand, if excision reactions are due to translocation problems, then Pol/Exo should be insensitive to dNTP concentrations, which we found (Fig. 4E). Therefore, Pol/Exo ratio is an indicator of translocation problems during DNA synthesis.

Translocation problems can be created by replication blocks or occasional uncoupling of the helicase and DNA polymerase activities during leading strand synthesis. Stalling of the replisome may result in transient reannealing of the junction base-pair and primer-end fraying, stimulating the exonuclease activity (Singh et al., 2020). This was tested by deliberating slowing down the helicase with reduced dTTP levels (preferred fuel for T7 helicase). Indeed, reduced dTTP levels decreased the polymerase activity and increased the exonuclease activity, reducing the Pol/Exo ratio (Fig. 4F-H). Uncoupling of helicase and polymerase activities can be caused by occasional disruptions in the physical interactions between the helicase and polymerase. The acidic C-tail of T7 helicase interacts with the basic front patch of the T7 DNA polymerase to physically couple the two proteins in the leading strand replisome (Fig. 4I) (Foster et al., 2019; Hamdan et al., 2007). The ΔC17 helicase mutant showed a lower Pol/Exo ratio of 6 relative to 12 in WT helicase (Fig. 4J). Thus, the Pol/Exo ratio measurement provides a convenient way to monitor how well the helicase and DNA polymerase are physically and functionally coupled during leading strand synthesis. Such studies can help understand replication defects in the disease-causing mutants of mitochondrial helicase Twinkle and DNA polymerase γ (Chan & Copeland, 2009; Peter & Falkenberg, 2020), and provide a better understanding of the ultramutational phenotypes of the exonuclease site mutants of human DNA polymerase ε linked with cancer progression (Kane & Shcherbakova, 2014; Li et al., 2018).

4.1. Equipment

• Cooling block to set the temperature of microcentrifuge tubes.

• TLC chambers.

• Air dryer.

• Typhoon FLA 9500 imaging system (GE Healthcare, Chicago, IL).

4.2. Reagents

• Buffer C: 50mM Tris–HCl, pH 7.5, 40mM NaCl, 10% glycerol and 2mM DTT

• Primed minicircle DNA substrate

• Purified T7 replication proteins

• dNTPs

• Formic acid

• PEI Cellulose F TLC sheets (EMD-Millipore, Burlington, MA)

• Phosphor-imaging screens (GE Healthcare, Chicago, IL)

• Potassium phosphate buffer (pH 3.8)

4.3. Procedure

1. Incubate the minicircle DNA substrate (32 nM) with 400 nM T7 helicase, 500 μM dTTP, and 1.5mM EDTA for 30 min. Add 400 nM T7 DNA polymerase to prepare “Mixture A”. Incubate Mixture A for 1 h at 25 °C.

2. Prepare “Mixture B” consisting of 20 μM T7 gene 2.5 protein (gp2.5), 300 μM each of dVTPs, and 10 mM MgCl₂. Spike Mixture B with a small amount of radiolabeled nucleotide, α-³²P-dGTP. Variations of this experimental strategy can be employed to test the effect of helicase-DNA polymerase uncoupling by limiting the concentration of dTTP (at 10 μM dVTPs and different concentrations of dTTP ranging from 10 to 500 μM) or by using C-tail deletion mutant of T7 helicase.

3. When starting the leading strand DNA synthesis reactions, transfer the tubes to a microcentrifuge tube cooler set at 18 °C and allow the mixtures to arrive at the temperature.

4. Initiate the leading strand DNA synthesis reaction by mixing equal volumes of Mixtures “A” and “B.” Mix the reaction thoroughly with the help of a pipette.

5. Take out 5 μL of the reaction at the stated time-points and add 5 μL of 8 M formic acid to quench DNA synthesis.

6. To prepare reactions for background correction, 5 μL of formic acid was added to 2.5 μL Mixture A before adding 2.5 μL Mixture B.

7. Keep the quenched reactions on ice or store them at −80 °C for later use.

8. Lightly mark a line with a pencil 2 cm from one of the sides of the TLC sheet. With the help of a pipette, spot 1 μL of the quenched reaction on the line, leaving about 1.5 cm space between subsequent spots.

9. Let the reaction spots completely dry at room temperature. Fill the phosphate buffer in the TLC chamber. Place the TLC sheet inside the chamber so that the side with spots directly contacts the buffer. Keep the buffer low enough not to touch the spots directly and run through them when the buffer rises.

10. Let the buffer reach the TLC sheet’s top edge (20 cm).

11. Dry the TLC sheets with an air blow dryer and wrap the sheets with a layer of Saran wrap.

12. Expose the blanked phosphor-imaging screens with the wrapped TLC sheets for 2–8h.

13. Scan the exposed phosphor-imaging screens using Typhoon FLA 9500 imaging system.

14. Dispose of the TLC sheets, reactions, and the pipette tips used for handling radioactive material according to the radioactivity safety guidelines.

4.4. Analysis

1. Quantitate the spot intensities for dGMP incorporated in the synthesized DNA, unused dGTP, dGDP, and dGMP excised (I_inc, I_dGTP, I_dGDP and I_ex, respectively) for reactions quenched at different time-points using ImageQuant TL image analysis software (GE Healthcare, Chicago, IL). Correct for the background by subtracting background spot intensities.

2. Calculate the fractions of dGMP incorporated (F_s) and excised (F_e) as:

   $$\begin{gathered} F_s=I_{\text{inc}}∕(I_{\text{inc}}+I_{\text{dGTP}}+I_{\text{dGDP}}+I_{\text{ex}}) \\ {F}_e=I_{ex}∕(I_{inc}+I_{dGTP}+I_{dGDP}+I_{ex}) \end{gathered}$$ (6)

3. Amounts of dNMP incorporated and excised can be quantified from the factions of labeled dGMP incorporated and excised during leading strand DNA synthesis and are expressed as moles of nucleotides incorporated or excised per mole of the replisome complex.

4. Plot the data obtained from step 3 against time and fit it to a line to obtain slopes (nucleotides incorporated or excised per mole of the replisome complex per second). We assume the amount of replisome complex to be the same as the amount of minicircle substrate DNA used in the assay.

5. Determine the Pol/Exo ratio using the obtained slopes.

4.5. Notes

1. DNA minicircle can be synthesized by ligating two ends of a linear ssDNA using T4 DNA ligase. A single-stranded splint DNA stabilizes the circular conformation bringing the two ends of the ssDNA together to facilitate the ligation reaction. Additionally, a DNA secondary structure strategically positioned in the middle of the linear DNA also helps in increasing the yields of circularized DNA. The 70-nt minicircle is synthesized as described (An et al., 2017). The initial reaction contained 2 μM 5’ phosphorylated linear 70 nt ssDNA, 25 μM splint DNA (see Table 1 for DNA sequences), 10U of T4 DNA ligase and 0.05X T4 DNA ligase buffer (Thermo Scientific, PA, USA) in a 20 μL reaction. Linear DNA is added to the reaction intermittently in small batches to final concentration of 10μM as described (An et al., 2017). The reaction is incubated at 20 °C for 12 h. Circularized DNA is separated from remaining linear DNA and higher-order linear polymers on a 15% urea TBE gel and is electroeluted from the gel to obtain pure DNA minicircle. Using a lower concentration of T4 DNA ligase buffer and the gradual addition of linear DNA in the reaction reduces the formation of higher-order linear products.

2. Use caution while using radioactive nucleotides and follow the institutional guidelines for the usage and disposal of radioactive material.

3. Preincubation time and temperature for preassembly of the leading strand complex should be optimized for replication proteins from other systems.

4. It is essential to keep the temperature constant during the reaction.

5. We considered replacing the radioactive dNTP with a fluorescent dNTP analog. However, replicative DNA polymerases do not incorporate fluorescent dNTP analogs with the same kinetics as normal dNTPs. Hence, results with dNTP analogs are unreliable for testing physiologically relevant models.

Abbreviations

2-AP

2-aminopurine

nt

nucleotide

NTP

nucleoside triphosphate

SSB

single-strand DNA binding protein

TLC

thin layer chromatography