If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Hepatitis C virus (HCV) is a major cause of liver-related diseases and hepatocellular carcinoma. The helicase domain of one of the nonstructural proteins of HCV, NS3 (nonstructural protein 3), is essential for viral replication; however, its specific biological role is still under investigation. Here, we set out to determine the interaction between a purified recombinant full length NS3 and synthetic guanine-rich substrates that represent the conserved G-quadruplex (G4)-forming sequences in the HCV-positive and HCV-negative strands. We performed fluorescence anisotropy binding, G4 reporter duplex unwinding, and G4RNA trapping assays to determine the binding and G4 unfolding activity of NS3. Our data suggest that NS3 can unfold the conserved G4 structures present within the genome and the negative strand of HCV. Additionally, we found the activity of NS3 on a G4RNA was reduced significantly in the presence of a G4 ligand. The ability of NS3 to unfold HCV G4RNA could imply a novel biological role of the viral helicase in replication.
). The possible roles of NS3 helicase include the following: (1) unwinding the duplex formed by the positive and negative strand during replication, (2) unfolding secondary structures within the HCV genome, and (3) displacing proteins that could potentially deter the viral replication process (
). G4 structures are formed when tandem repeats of guanine bases of an RNA or a DNA molecule interact with one another through Hoogsteen hydrogen bonding to form a planer structure known as a quartet or a tetrad (
). As a result, viruses need to control the folding and unfolding of these nucleic acid structures in a regulated manner. There are several helicases reported to have the ability to unfold RNA or/and DNA G4s (
)”. It is likely that NS3 is the only helicase inside the vesical because, according to the model proposed by Quinkert et al., only molecules of a specific size (not bigger than 16 kDa) can enter through the “neck’ of the membranous web into the vesicle where replication takes place (
). Therefore, it is likely that HCV uses its own helicase (NS3) to unfold the conserved G4 structures within its genome to allow effective replication. In our study, we investigated the activity of NS3 on G4RNA-forming sequences derived from the positive or the negative strand of HCV. Our data suggests that NS3 unfolds G4RNA. However, its G4RNA-unfolding activity is significantly reduced when the G4 is inherently stable or stabilized with a G4 ligand.
NS3 binds to HCV G4-forming substrates with or without a 3′ overhang
To determine whether and how NS3 interacts with G4RNA, we first design three different G-rich sequences: HCVG4, NEGG4, and NONHCVG4 (see Table S1 for sequences). HCVG4 contains the G-rich sequences from the positive strand of genotype 3a (
) and has the potential to form a two-tetrad G4 structure. Unlike HCVG4 and NEGG4, NONHCVG4 does not represent a sequence from HCV; however, it is a useful model substrate because it can potentially fold into a highly stable three-tetrad G4 since it has only one nucleotide loop length. For G4RNA structures containing relatively shorter loops, loop length has an inverse relationship with G4 stability (
). Therefore, the shorter loop length results in more stable G4. NONHCVG4 was included in this study to understand how NS3 interacts with an RNA substrate that is likely to fold into an inherently stable G4.
The helicase domain of NS3 interacts with oligonucleotides in an orientation-specific manner (
). Therefore, a 3′ overhang has a significant effect on the affinity and activity of NS3 on a partial duplex. With the aim of understanding the effect of a 3′ overhang on the interaction between NS3 and G4RNA, 20 nucleotide (nt) adenosine monophosphates (A20) were added to each of the above three substrates to form HCVG4-A20, NEGG4-A20, and NONHCVG4-A20.
CD spectra of all of the G-rich substrates mentioned above indicate that the sequences fold into a parallel G4 structures in 100 mM KCl (Fig. 1). A parallel G4 possesses a distinct CD spectrum with a peak around 265 nm and a valley around 240 nm (
). We conclude that the presence of 20-nt adenosine monophosphates at the 3′ end of HCVG4, NEGG4, or NONHCVG4 does not hinder the formation of a parallel G4.
To test whether NS3 binds to the G4RNA structures, a fluorescence anisotropy–binding assay was performed in 100 mM KCl binding buffer containing fluorescently labeled HCVG4, NEGG4, or NONHCVG4 and increasing concentration of NS3. NS3 bound tightly to the ssRNA, A20-3′FAM, with similar affinity (Kd = 6 ± 1 nM) to what was reported in the past for NS3 binding to a ssRNA (
)), we also fit the data points to a two-sites binding model and determined the Kd1 and Kd2 to be 4.9 nM and 860 nM, respectively (Fig. S1). NS3 bound to HCVG4, NEGG4, or NONHCVG4 tightly with Kd values in the nano-molar range (Kd = 17 ± 4 nM, 19 ± 5 nM, and 23 ± 7 nM, respectively) (Fig. 2, A,B, and D). Similarly, NS3 binds tightly to the 3′ overhang–containing substrates, HCVG4-A20, NEGG4-A20, or NONHCVG4-A20 (Kd= 11 ± 3 nM, 6 ± 2 nM, and 12 ± 3 nM, respectively) (Fig. 2, A,C, and D). The binding affinity of NS3 to the 3′ overhang containing substrates is within two-fold of the single-stranded substrate. However, the substrates that lack a 3′ overhang bind to NS3 with a dissociation constant as high as four-fold than the ssRNA substrate. Nonetheless, the absence of a single-strand protein-loading zone does not prevent the NS3 from binding to the G4-forming substrates.
NS3 is active in a buffer that supports G4RNA folding
Unfolding of a G4 structure requires a reaction buffer containing enough cations to maintain the folding of the G4 during the span of the reaction. Therefore, it is necessary to first verify the activity of the helicase in a buffer in which the G-rich oligonucleotides fold to form a stable G4. Several previous studies that examined the interaction of a helicase with a G4-forming oligonucleotides reported of using a buffer containing 100 mM KCl (
). Therefore, we first determined the ATPase activity of a recombinant NS3 protein in 100 mM KCl buffer by conducting enzyme-coupled spectrophotometric ATPase assay that links ATP hydrolysis with the oxidation of NADH (Fig. 3A). In saturating conditions, which is the concentration of Poly (U) at which the specific activity is maximum, NS3 has ATPase activity of 52.5 ± 1.8 s−1 in low salt (50 mM KCl) buffer and 49.8 ± 5.0 s−1 in high salt (100 mM KCl) buffer, which is not significantly different (Fig. 3B). Therefore, the purified NS3 is as active at high salt concentration as it is at low salt concentration.
The ability of NS3 to unfold G4 could be inferred from enzyme-catalyzed G4 reporter duplex unwinding assay
To test whether NS3 can unfold G4-forming substrates to which it binds, we conducted a G4 reporter duplex unwinding assay. A radiolabeled–G4 reporter duplex substrate that contains the G-rich sequences from HCVG4, NEGG4, or NONHCVG4 oligonucleotides flanked by a 25 base pair duplex region and A20 3′ overhang was made as described in the Experimental procedures and illustrated in Fig. S2. Three nucleotides were added as a linker between the duplex region and the G4-forming sequences to allow proper folding of the G4 structure. The reporter substrate was incubated with the enzyme and the reaction was initiated with the addition of ATP (Fig. 4, A and B). In the reporter assay, the unfolding of the G4 structure by the enzyme is reported by the unwinding of the duplex region (Fig. 4B).
Initially, we evaluated the amount of NS3 needed for the G4 reporter assay to saturate the substrate. We carried out the G4 reporter assay with varying concentrations of NS3 (400 nM, 600 nM, 800 nM, or 1000 nM) and 2 nM HCVG4-A20 reporter duplex substrate. The product formed from 400 nM NS3 was significantly lower than the product formed from the other NS3 concentrations; however, the amount of products obtained from 600 nM, 800 nM, and 1000 nM NS3 are not significantly different (Figs. 4, C, D and S3), suggesting 600 nM NS3 is enough to saturate the substrate. Consequently, 600 nM NS3 is the concentration used for subsequent dsRNA or G4 reporter duplex unwinding reactions.
NS3 unfolding of G4 RNA is affected by substrate sequence and reaction conditions
We used these reaction conditions as our standard condition: 600 nM NS3 incubated with 2 nM-radiolabeled HCVG4-A20 reporter duplex substrate in a buffer containing 100 mM KCl (Fig. 5A). Under standard conditions, NS3 unwinds HCVG4-A20 reporter duplex with an amplitude of product formation of 50 ± 1% at 20 min. The product formation is drastically reduced to 11 ± 1% when 0.25 μM PDS, a known G4-stabilizing ligand, is added to the reaction (Fig. 5, B and C). The same concentration of PDS has less effect on the activity of NS3 on a partial duplex substrate with the same length and sequence as the HCVG4-A20 reporter duplex, except that the guanines within the G-rich sequence are mutated to adenine (MUTHCVG4-A20; Table S1 and Fig. S4). For the HCVG4-A20 reporter duplex unwinding, there is a small (<5%) but measurable amount of product observed in the absence of ATP during the course of the reaction. In contrast, virtually no product is measured in the absence of NS3, suggesting that the product observed under the standard reaction conditions is driven by the enzyme. Interestingly, some product is measured from the zero second-time point (Fig. 5B). It is possible that this product is formed during preincubation in an ATP-independent manner. Reynolds et al. reported NS3 unwinding of 3′- or 5′-tailed short duplexes in an ATP-independent manner and proposed a noncanonical duplex melting mechanism by NS3, which involved local strand separation utilizing binding energy (
). Therefore, the product we observe from the G4 reporter duplex unwinding reaction in the absence of ATP could be due to NS3 binding and unwinding of the duplex region adjacent to the G4 structure in an ATP-independent manner.
A single-turnover reaction involves product formation by the actions of NS3 molecules that are bound to the substrate at the initiation of the reaction. To create single-turnover reaction conditions, a protein trap is added in the reaction buffer to keep the dissociated NS3 molecules from rebinding to the substrate. The amount of product formed from a single-turnover reaction indicates the level of processivity of NS3 with the substrate studied. Under single-turnover conditions, 11 ± 1% product is formed from HCVG4-A20 reporter duplex, which is a 78% reduction from the product measured under a multi-turnover reaction condition in which NS3 molecules are allowed to bind, unwind, dissociate, and rebind for another cycle of enzymatic action (Fig. 6, A and B). The low product formation from the single-turnover reaction indicates NS3 might need to undergo a series of association and dissociation events from the substrate in order to unfold the G4 and then unwind the adjacent duplex region.
Unlike KCl and NaCl, LiCl does not support the formation of a stable G4 (
). To test the effect of not having a stable G4 on the unfolding activity of NS3, we allowed the enzyme to unfold the same substrate, HCVG4-A20 reporter duplex, in 100 mM LiCl reaction buffer, which resulted in 91 ± 2% product formation (Fig. 6, C and D). The product obtained in 100 mM KCl is significantly lower than in 100 mM LiCl, implying that a stable G4 could be a formidable barrier, keeping the NS3 from unwinding the adjoining duplex region of the G4 reporter duplex substrate.
In addition to the external factors discussed above, the length of the 3′ overhang could also regulate the G4-unfolding activity of NS3. Efficient partial duplex unwinding by NS3 is highly dependent on the length of the 3′ overhang; the longer the tail, the higher the unwinding amplitude (
). In order to investigate the effect of the length of a 3′ overhang on G4-unfolding activity of NS3, a G4 reporter duplex unwinding reaction was conducted with HCVG4 reporter substrate containing 20-nt, 9-nt, or 2-nt 3′ overhangs. We observed 50 ± 1%, 40 ± 3%, and 20 ± 2% product at 20 min for substrates with 20-nt, 9-nt, and 2-nt of 3′ overhangs, respectively, in a multi-turnover reaction (Fig. 7). The result strongly implies that the length of the single-strand extension is important for efficient unfolding of a G4 by NS3.
Under multi-turnover conditions, 82 ± 4% of the MUTHCVG4-A20 control duplex is unwound, which is significantly higher than the product obtained from HCVG4-A20 reporter duplex substrate (50 ± 1%) (Fig. 8, A and B). In addition, the fraction of the duplex substrate unwound under single-turnover conditions was considerably higher than the amount of G4 containing reporter duplex unwound (Fig. 8, A and B). This higher activity of NS3 with the control duplex substrate than with the G4 reporter duplex implies that the helicase is more processive on the former than on the later. It also indicates that G4 is a strong deterrent for helicase progression. Unlike for HCVG4-A20 reporter duplex unwinding, the type of cation in which the unwinding reaction takes place does not make a significant difference for MUTHCVG4-A20, which suggests that at the concentration tested, the KCl does not interfere with the helicase activity of NS3 (Fig. 8, C and D). Therefore, the lower level of product formed from the G4-containing reporter duplex in 100 mM KCl (Fig. 8, B and D) is due to the presence of a G4.
We also deleted the G-rich sequence within the HCVG4-A20 reporter substrate to create another control substrate containing 25-mer duplex and 20-mer 3′ overhang (DELHCVG4-A20; Table S1). The unwinding of DELHCVG4-A20 resulted in more product formation compared to the WT substrate (Fig. S5), supporting the conclusion that the G-rich sequence within HCVG4-A20 reporter duplex is responsible for reducing the unfolding activity of NS3 in 100 mM KCl.
To test the effect of oligonucleotide sequence on the ability of NS3 to disrupt G4, we conducted G4 reporter duplex unwinding assays with three different RNA G4-forming reporter substrates: HCVG4-A20, NEGG4-A20, or NONHCVG4-A20. NS3-catalyzed HCVG4-A20, NEGG4-A20, and NONHCVG4-A20 reporter duplex unwinding reactions in 100 mM KCl resulted in 50 ± 1%, 80 ± 1%, 49 ± 3% product formation, respectively (Fig. 9A). The activity of NS3 on the three-tetrad G4-forming substrates (HCVG4-A20 and NONHCVG4-A20) is significantly lower than on a two-tetrad G4-forming substrate (NEGG4-A20). No significant difference in product formation among the three substrates was observed when the unwinding reaction contained 100 mM LiCl (Fig. 9B), further confirming the conclusion that NS3 unfolding is affected by G4 stability, which in turn is determined by oligonucleotide sequence and the type of monovalent cation used.
The gel images from the unwinding reactions of the above three reporter substrates (Fig. S6) indicate that NONHCVG4-A20 forms a significant amount of higher order structure (intermolecular G4) in 100 mM KCl, but not in 100 mM LiCl buffer (Fig. S6, E and F). NEGG4-A20 also forms intermolecular G4 in 100 mM KCl (Fig. S6, C and D). Therefore, when analyzing the fraction G4 unfolded from NONHCVG40-A20- or NEGG4-A20- unwinding reaction (Fig. 9), we took the intermolecular G4s present at each time points into consideration when quantifying the results. On the other hand, since the amount of intermolecular G4 observed is negligible (Fig. S6,A and B), the analysis of the fraction of G4 unfolded from HCVG4-A20 reporter substrate did not involve intermolecular G4 structures.
Examination of NS3 unfolding of G4RNA by G4-trapping assay
Because of the G4 reporter duplex substrate consists of a duplex region linked to a G4, unfolding of the G4 precedes the unwinding of the duplex. Consequently, the overall rate of the G4 reporter unwinding reaction is the sum of the rates of G4 unfolding and duplex unwinding. There is a possibility that G4 reporter assay could overestimate the actual G4 unfolding in situations where the duplex region is unwound by direct binding of NS3 without translocating along the G-rich sequences. It could also underestimate the amount of G4 unfolded in cases where the duplex unwinding is much slower than the G4 unfolding. To overcome such scenarios, we decided to further investigate NS3 G4RNA unfolding by a G4-trapping method where NS3 interacts with a G4-containing substrate and the unfolded product is captured by a complementary strand (Q-trap) (Fig. 10, A and B). The duplex product and the G4 substrate can then be separated by gel electrophoresis due to the difference in their migration speed (Fig. 11A). A second trap, C-trap, is present in the quench buffer to prevent the Q-trap from unfolding the substrate. We conducted the G4-trapping assay at saturating conditions (Fig. S7) with respect to the substrate. The amplitude of product formation in the presence of NS3 from HCVG4-A20, NEGG4-A20, and NONHCVG4-A20 are 35 ± 3%, 92 ± 4%, and 38 ± 4%, respectively (Fig. 10, C and D). The products formed by the Q-trap alone, on the other hand, are 12 ± 1%, 24 ± 2%, and 8 ± 2%, respectively (Fig. 10, C and D). For each substrate, NS3-dependent unfolding is drastically higher than unfolding by Q-trap, which indicates that the complementary strand alone has only a minimal effect on G4RNA unfolding. Similar to the G4 reporter–unwinding reactions, the most product is formed from HCVNEGG4-A20 unfolding, most likely due to its lower stability. However, there is no significant difference between the products obtained from NONHCVG4-A20 and HCVG4-A20. The gel image for NONHCVG4-A20–unfolding experiment shows three major bands (Fig. 11A). The bands represent product, intramolecular G4 species (migrates faster than the product), and intermolecular G4 species or higher order G4 structure (migrates slower than the product). Interestingly, the NS3 is completely inactive on the intramolecular G4 subpopulation (Fig. 11, B and D). However, the intermolecular G4 is disrupted by NS3 more so than by the complementary strand (Fig. 11, B, C, and E). Therefore, the 38 ± 4% product formed from NONHCVG4-A20 is due to the disruption of the intermolecular G4 species not the intramolecular G4 species, which are too stable to be unfolded by NS3 or the complementary strand.
The binding experiment shows that NS3 binds with high affinity to G4-containing substrates with or without a 3′ overhang (Fig. 2). Zhou et al. reported that the affinity of NS3 full length from genotype 1b for RNA duplex was drastically reduced with shorter 3′ overhang length (
), signifying the need for having a 3′ overhang for NS3 to efficiently bind to a partial duplex. Similarly, NS3 binds with higher affinity to G4-containing substrates with a 3′ overhang than without (Fig. 2), which suggests the importance of a 3′ overhang for NS3-dependent unfolding of G4 structures. Unlike certain RNA helicases, NS3 does not show preferential binding to G4RNA compared to ssRNA. By performing EMSAs, Liu et al. demonstrated DHX36’s ability to bind to RNA G4 of C9orf72 repeat with higher affinity than substrates that form antiparallel G4, hairpins, or non-G4–forming sequences (
). The preference of NS3 for ssRNA to G4 structure–containing RNA substrate could be an indication of the potential mechanism it possesses to effectively unfold a G4 structure. NS3 might first bind to the single-stranded region and form a functional complex with the substrate. Then, it could translocate in the 3′ to 5′ direction toward the G4 structure using the energy from ATP hydrolysis. Other helicases are also thought to unfold G4 structures by translocation-based mechanism (
); therefore, it is likely that more than one molecule of NS3 is binding to the same ssRNA. The fact that the change in anisotropy for substrates lacking a 3′ overhang is smaller could be an indicative of the binding of fewer NS3 molecules than the ssRNA. It was surprising to us that the change in anisotropy of the G4-containing substrates with a 3′ overhang is significantly smaller than that measured for the ssRNA (Fig. 2). It is possible that the presence of a G4 structure in these substrates could affect the single strand extension, reducing the size of the loading site and allowing the binding of fewer NS3 molecules.
Our data suggest that efficient unfolding of G4RNA by NS3 depends on factors such as reaction conditions, substrate sequence, and length of the 3′ overhang adjacent to the G4 structure. NS3 exhibits low activity on G4-containing substrates under conditions that support stable G4 formation (Figs. 5 and 6). G-rich substrates that fold into a two-tetrad rather than a three-tetrad G4 structure are better substrates for NS3 (Fig. 9). Our results strongly suggest that even if NS3 can unfold G4RNA, doing so requires multiple rounds of ATP hydrolysis over a relatively long period of time when the G4 is further stabilized by small molecules (Fig. 5) or encounters inherently stable intramolecular G4 (Fig. 11). The virus could overcome this low G4-unfolding activity of NS3 by engaging other viral proteins in the process. It is possible that accessory proteins could assist NS3 with G4 unfolding. Rajagopal and Patel reported single-stranded–binding proteins could enhance the processivity of the helicase domain of NS3 on a DNA substrate (
). Similarly, G4 unfolding by NS3 could be affected by NS5A or/and NS5B. In addition, our data indicate that longer 3′ extensions enhance the efficiency of G4RNA unfolding by NS3 (Fig. 7). The 3′ overhang could allow the productive assembly of multiple molecules of NS3, which could lead to cooperative unfolding of a G4. This proportional relationship between enzyme activity and the length of a 3′ overhang is also observed with other helicases. For example, partial duplex DNA-unwinding efficiency of the helicase domain of NS3 (NS3h) (
). One possible evolutionary advantage of viral G4 structures is to allow the virus to avoid the host immune detection by maintaining viral protein production at a level that is sufficient for persistent infection but below the threshold for initiation of the host immune response (
). No matter their function, viruses need to regulate the folding and unfolding of G4 structures in a controlled manner. There are several reasons to believe that NS3 could unfold HCV G4 structures in vivo. One reason could be its high affinity for G4-containing substrates (Kd values in the low nM range; Fig. 2). Second, there are excess copies of NS3 for every positive or negative strand RNA present within an active HCV replicase complex (
). Having an excess of NS3 within the replication complex increases the likelihood of NS3 interacting with HCV G4 structures. Moreover, it is likely that NS3 is the only helicase protein in the membranous web (site of replication) which keeps large molecules from entering (>16 KDa) into the vesicle (
). G4RA unfolding by NS3 could be a novel role for the viral helicase in the life cycle of HCV.
There have been attempts to target viral G4s and suppress replication. For example, the HIV nef gene contains conserved G4 structures which, when stabilized by TMPyP4 (a G4 stabilizer), reduces nef-GFP protein expression in a GFP-based reporter gene assay and nef-dependent HIV-1 infectivity (
). In the present study, the addition of PDS significantly reduced the ability of NS3 to unfold HCV G4 (Fig. 5), suggesting the potential role of G4-stabilizing ligands as anti-HCV therapeutic agents. There have been many attempts to develop NS3 helicase inhibitors as anti-HCV therapeutic agents (
). We suggest that a combination drug in the form of NS3 inhibitor and HCV G4 stabilizer might be more effective in reducing viral replication as opposed to inhibition of a single target.
Oligonucleotides and protein
The RNA and DNA oligonucleotides used in this study were purchased from Dharmacon and Integrated DNA Technologies, respectively. Table S1 displays the sequences of RNA oligonucleotides. The oligonucleotides were purified as explained in (
), but with minor modifications. RNA oligonucleotides were gel purified, electroeluted, desalted, and then dried by vacuum centrifugation. The dried oligonucleotides were deprotected as directed by the manufacturer before resuspension in a sodium-free storage buffer containing 10 mM Hepes, pH 7.5 and 1 mM EDTA, pH 8.0. All buffers or solutions in this study were made in diethyl pyrocarbonate (DEPC)- (ACROS Organics) treated water to make them RNase free. 0.1% DEPC in deionized water was prepared and then autoclaved before use. The concentrations of the purified oligonucleotides were determined using calculated extinction coefficients or the extinction coefficient provided by the manufacturer. For the G4 reporter assay, the trap reporter substrate was radiolabeled as reported in (
) with a few changes. Here, the RNA substrate was incubated with [γ -32P] ATP (PerkinElmer) in the presence of T4 polynucleotide kinase (Promega) for 1 h at 37 °C. Then, unlabeled excess [γ -32P] ATP was removed using Illustra MicroSpin G-25 Columns containing Sephadex G-25 DNA grade F resin (Cytiva or GE Healthcare). In order to fold G4 and duplex structures, the labeled displaced strand was mixed with 1.5 times excess unlabeled loading strand in a buffer containing 100 mM KCl or LiCl. The mixture was then heated at 95 °C for 6 min and slowly cooled to room temperature to promote the formation of a duplex as well as a G4 structure. Full length genotype con1b NS3 was purified as explained in (
), except only the first and the second talon metal affinity columns were used. The purified protein was stored in a buffer containing 25 mM Hepes, 100 mM NaCl, 1 mM EDTA, 5 mM β-mercaptoethanol, and 20% glycerol. The protein concentration was calculated by UV absorbance at 280 nm with extinction coefficient value of 64,000 M−1 cm−1.
CD spectroscopy was conducted to confirm the ability of the guanine-rich RNA sequences (Table S1) to fold into G4 structures (
) with minor modifications. Here, 9 μM or 10.2 μM guanine-rich RNA oligonucleotides were suspended in a G4-folding buffer containing 10 mM Hepes, pH 7.5, 1 mM EDTA, pH 8.0, and 100 mM KCl, then heated at 95 °C for 6 min and slowly cooled to room temperature before using a Jasco J-715 spectropolarimeter (JASCO) to measure the CD spectra. The scanning speed of the spectropolarimeter was set at 100 nm/min for a wavelength range of 200 to 320 nm.
). In short, all of the Fluorescein Amidite–labeled G-rich RNA substrates (Table S1) were suspended in a G4-folding buffer containing 100 mM KCl, heated at 95 °C for 6 min and slowly cooled to room temperature. The folded G4 substrates (2 nM final concentration, unless stated otherwise) were preincubated in binding buffer (25 mM MOPS, pH 6.5; 100 mM KCl, 0.1 mM EDTA, 2 mM β-mercaptoethanol, 0.1 mg/ml bovine serum albumin in DEPC-treated water) with increasing concentrations of NS3 in a 96-well plate for 30 min at 37 °C. A PerkinElmer 1420 multilabel counter victor 3V plate reader was used to determine the polarization of NS3-G4 complex with 485 nm and 535 nm excitation and emission wavelengths, respectively. The polarization measurements were converted to fluorescence anisotropy values, which were then fit to a quadratic equation (unless otherwise stated) to acquire dissociation constants (Kd).
Excess enzyme-catalyzed dsRNA and G4 reporter duplex unwinding assay
The concentrations mentioned for the excess enzyme-catalyzed dsRNA and G4RNA reporter duplex unwinding assay are final. The dsRNA or a G4 reporter substrate (2 nM) in 100 mM KCl or LiCl was preincubated with 600 nM NS3 (unless otherwise specified) for 30 min at 37 °C in a reaction buffer containing 25 mM MOPS, pH 6.5, 100 mM KCl or LiCl, 0.1 mM EDTA, 2 mM β-mercaptoethanol, and 0.1 mg/ml bovine serum albumin. Simultaneously, in another reaction tube, 5 mM ATP, 10 mM MgCl2, and 60 nM trap substrate (Table S1) were preincubated together. For a single cycle reaction, the second reaction tube also contained 0.5 mM T50 in nucleotides as a protein trap. The reaction commenced when the contents of the two reaction tubes were combined. The reaction was terminated at increasing time points by addition of quench (100 mM EDTA and 1% SDS). For a time zero (blank) sample, equal volumes of samples from the NS3-substrate and ATP-Mg+2 tubes were taken out separately and then quenched into the same quench buffer immediately before the reaction was initiated (unless otherwise stated). For a heat control sample, half of the quenched zero-point sample was heated at 95 °C for 6 min in a separate vial and cooled down on ice. The products and the reactants of each of the samples (time-zero, heat control, and reaction samples) were separated by 20% native PAGE. A Typhoon Trio phosphorimager was used to image the samples on the gel, which was then quantified by ImageQuant software (GE Healthcare) for data analysis. Each datum point was normalized to the product measured at zero-time point. Data was fit to a single or a double exponential equation (unless otherwise stated) using KaleidaGraph software (Synergy Software).
Enzyme-coupled spectrophotometric ATPase assay
An ATPase assay was performed as communicated in reference (
). The activity of NS3 was examined in an ATPase assay reaction buffer containing 50 mM MOPS, pH 6.5, 50 or 100 mM KCl, 10 mM MgCl2, 5 mM ATP, 4 mM phosphoenolpyruvate, 0.7 mg/ml NADH, and 10 u/ml pyruvate kinase and lactate dehydrogenase. Poly (U) (5 μM, 100 μM, or 250 μM) was added into the reaction to initiate hydrolysis of ATP by NS3. PK/LDH links the hydrolysis of ATP with the oxidation of NADH. The change in the concentration of NADH was determined by measuring the change in the UV absorbance of NADH at 380 nm. The rate of ATP hydrolysis is equivalent to the rate of change in the concentration of NADH. ATPase activity was calculated by dividing the rate of ATP hydrolysis by the amount of NS3 used in the reaction (50 nM) and is expressed in inverse of a second (s−1).
G4RNA unfolding by trapping
The concentrations mentioned for G4RNA-unfolding trap assay are final. All substrates were mixed in a buffer containing 100 mM KCl, heated at 95 °C for 6 min, and slowly cooled to room temperature before using it in the unfolding reaction. In one reaction tube, 25 nM Fluorescein Amidite–labeled HCVG4A20, NEGG4A20, or NONHCVG4A20 substrate was incubated with 800 nM NS3 at 37 °C for 30 min in a reaction buffer containing 25 mM MOPS, pH 6.5, 100 mM KCl, 0.1 mM EDTA, 2 mM β-mercaptoethanol, and 0.1 mg/ml bovine serum albumin. In a second reaction tube, 5 mM ATP, 10 mM MgCl2, 750 nM Q-trap (complementary to the G-rich substrate) were incubated together at 37 °C for 30 min. The G4RNA-unfolding reaction was triggered by combining equal volumes of the contents of the two reaction tubes. The reaction was quenched at increasing time points by 100 mM EDTA, 1% SDS, and 60 μM C-trap (it sequesters excess Q trap). The zero-time point (blank) sample was collected immediately before the reaction was initiated. The partial duplex product was resolved from the G4RNA substrate with 20% native PAGE. The gel was then visualized using a Typhoon Trio phosphorimaging system and quantified using ImageQuant software. The fraction product formed was determined using the formula mentioned in reference (
The authors declare that they have no conflicts of interest with the contents of this article.
We thank Dr Robert Eoff for proofreading the article and Dr John Marecki for helpful discussions.
B. B., A. K. B., and K. D. R. conceptualization; B. B., J. G., and A. K. B. methodology; B. B. formal analysis; B. B. investigation; B. B. writing–original draft; B. B. visualization; J. G., A. K. B., and K. D. R. writing–review and editing; J. G., A. K. B., and K. D. R. supervision; K. D. R. funding acquisition.
Funding and additional information
This work was supported by National Institute of Health Grant R35GM122601 (K. D. R). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institute of Health.