An ELISA for RNA Helicase Activity: Application as an Assay of the NS3 Helicase of Hepatitis C Virus

An ELISA for RNA Helicase Activity: Application as an Assay of the NS3 Helicase of Hepatitis C Virus

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS ARTICLE NO. 253, 594 –599 (1998) RC989813 An ELISA for RNA Helicase Activity: Application as an...

322KB Sizes 4 Downloads 33 Views

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS ARTICLE NO.

253, 594 –599 (1998)

RC989813

An ELISA for RNA Helicase Activity: Application as an Assay of the NS3 Helicase of Hepatitis C Virus Charles C. Hsu,*,1 Lih-Hwa Hwang,*,†,1 Yu-Wen Huang,† Wei-Kuang Chi,‡ Yi-Ding Chu,‡ and Ding-Shinn Chen*,2 *Hepatitis Research Center, National Taiwan University Hospital and †Graduate Institute of Microbiology, National Taiwan University College of Medicine; and ‡Development Center for Biotechnology, Taipei, Taiwan

Received October 23, 1998

A convenient enzyme-linked immunosorbent assay (ELISA) for RNA helicase activity was developed with principles similar to the standard assay. The helicase ELISA utilizes a non-radioactive double-stranded substrate with a biotin-labeled template (long) strand hybridized to a digoxigenin (DIG)-labeled release (short) strand. The template strand binds to the wells of streptavidin-coated microtiter plates (SA-MTP) where the helicase catalyzes the unwinding reaction. Substrate not unwound retains the DIG-labeled release strand and is detected using anti-DIG coupled to horseradish peroxidase. Chromogenic detection follows. Absorbance measurement allows determination of unwinding efficiency of reactions. To demonstrate effectiveness, the ELISA-based assay was used to study the unwinding activity of the hepatitis C virus (HCV) NS3 helicase. Using a known inhibitor of NS3 helicase activity and two mutant HCV helicases, the ability of the assay to screen potential anti-helicase drugs and putative helicases is illustrated. The helicase ELISA is more convenient than the standard helicase assay and is especially suited for the testing of large numbers of samples. © 1998 Academic Press

Viral RNA helicases have been implicated in viral replication, transcription, and protein translation, and thus appear to play a key role in viral propagation (for review, see 1). RNA stimulated NTPase and helicase activities have been demonstrated for putative helicases of several DNA and RNA viruses, including the simian virus 40 (SV40) large T antigen (2), the vaccinia virus nucleotide triphosphate phosphohydrolase II 1

C. C. Hsu and L. H. Hwang contributed equally to this work. To whom correspondence should be addressed: Dr. Ding-Shinn Chen, Hepatitis Research Center, National Taiwan University Hospital, 7, Chung-Shan S. Road, Taipei, 100, Taiwan. E-mail: [email protected] 2

0006-291X/98 $25.00 Copyright © 1998 by Academic Press All rights of reproduction in any form reserved.

(NPH-II) protein (3), the bovine viral diarrhea pestivirus (BVDV) NS3 protein (4), and also the hepatitis C virus (HCV) NS3 protein (5,6). The NS3 protein (72 kDa) of HCV contains a bipartite structure consisting of an N-terminal serine protease (7,8) and a C-terminal helicase belonging to the DEAD (Asp-Glu-Ala-Asp) box containing NTPase-helicase family (5,6). The DEAD box RNA helicase family contains several conserved motifs (9,10). The first motif (GxGKx), referred to as “Walker motif A”, binds the terminal phosphate groups of the NTP cofactor; the second motif (DEAD), referred to as “Walker motif B”, is responsible for chelating the Mg21 of the Mg-NTP complex (11,12). Corresponding amino acid (aa) sequences of the HCV NS3 protein are G207SGKS (Gly-Ser-Gly-Lys-Ser) for motif A and D290ECH (Asp-Glu-Cys-His) for motif B. To assess unwinding activity, the standard helicase assay involves the incubation of the putative helicase with a duplex RNA substrate in the presence of NTP and a divalent cation, such as Mg21 or Mn21 (1,5,6). The in vitro transcribed double-stranded substrate consists of two annealed complementary RNA strands, one of which is radiolabeled, usually with a-32P. Visualization of substrate unwinding is through resolution of the reaction products using polyacrylamide gel electrophoresis, or PAGE. However, the standard assay is laborious and inefficient. Preparation of the radiolabeled substrate is especially tedious, requiring two to three days and involves PAGE and repeated extraction steps (6). Length of viability of the substrate is limited by the half-life of its radioisotope. Long-term storage is not feasible, necessitating frequent and repeated preparation. Additionally, there are hazards involved with the use of radioisotopes. Visualization with PAGE necessarily limits the number of samples run, making it not amenable to the study of numerous samples, a serious liability for screening putative helicases or inhibitors of helicase activity.

594

Vol. 253, No. 3, 1998

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

Inspired by the need for a simple, sensitive, and efficient helicase assay, an RNA helicase enzymelinked immunosorbent assay (ELISA) using a similar strategy to the standard assay was designed and tested with the HCV NS3 helicase domain. With this procedure, radioisotope is avoided, the RNA substrates can be stored long term, and a larger number of samples can be reasonably analyzed. Quantification is reliably done with an ELISA reader. Moreover, the entire assay requires less than one day. The RNA helicase ELISA is presented here to demonstrate its effectiveness, sensitivity, and reliability with comparison to the standard assay. Using KCl, a proven inhibitor of HCV NS3 NTPase (13) and helicase activities (6), the usefulness of the assay for screening anti-helicase agents is proven. In addition, the ability of the helicase ELISA to sensitively screen putative helicases is illustrated through the assaying of several mutant HCV NS3 proteins. MATERIALS AND METHODS Construction, Expression, and Purification of NS3 Helicase Protein Construction, expression, and purification of the helicase domain of the HCV NS3 protein, aa 1175-1657 (designated as clone 59-1175), were carried out essentially as previously described (6). Proteins were dialyzed in a TNE buffer (10 mM Tris [pH 8.0], 100 mM NaCl, 1 mM EDTA), concentrated with Amicon Centriprep-10, and stored at 280°C until use. Two mutant proteins, each with a single point mutation in either motif I (GxGKS) or motif II (DExH), were created by using the Transformer Site-Directed Mutagenesis Kit (Clontech, Palo Alto, CA). The lysine residue of motif I in HCV NS3 (aa sequence GSGK210S) was changed to asparagine (K210N) and resulted in the mutant having the aa sequence GSGNS. The histidine residue of motif II (aa sequence DECH293) was transformed to an alanine residue, with the H293A mutant having an aa sequence of DECA. The wild-type recombinant plasmid of clone 59-1175 was utilized, and in vitro mutagenesis was performed according to the manufacturer’s manual. The mutated sequences of the plasmid DNA encoding the point-mutated proteins were verified by dideoxy DNA sequencing (14). Expression of the mutant proteins was induced with IPTG, and each protein was purified as described previously (6).

Standard Helicase Assay Preparation of the [a-32P]-CTP labeled double-stranded RNA (dsRNA) substrate and procedures for the standard HCV NS3 helicase assay were performed as previously described (6).

RNA Helicase ELISA An overview of the RNA helicase ELISA is provided in Fig. 1. (i) Preparation of substrate-coated SA-MTP wells. The substrate is composed of two annealed complementary RNA strands in vitro transcribed from the multiple cloning sequences (MCS) of pGEM vectors (Promega). Using SP6 RNA polymerase, the template strand (104 nt) was transcribed from PvuII digested pGEM3, and the release strand (41 nt) was transcribed from XbaI digested pGEM4. The template strand was labeled with biotin-16-UTP, and the release strand was labeled with DIG-UTP (Boehringer). 34 nt are comple-

mentary between the template and release strands. Hybridization buffer and washing buffer were provided from the Northern ELISA kit (Boehringer). For each helicase reaction, 2 ng biotin-labeled template strand was added to 6.4 ng DIG-labeled release strand in 120 ml hybridization buffer per round-bottom well of an uncoated microtiter plate. For the standard curve, 2 ng template was added to varying amounts of release strand (0.2 ng–16 ng) per well. The hybridization plate, covered to prevent evaporation, was incubated for 3 h at 50°C and shaken at 250 rpm. 100 ml of each hybridization mix was transferred to SA-MTP wells and incubated for 10 min at 50°C (250 rpm). To remove unhybridized release strand and unbound substrate, the hybridization mixes were removed, and each well was washed three times with 280 ml washing buffer (Boehringer). (ii) Helicase reaction. To equilibrate for the helicase reaction, each well was washed three times with 280 ml buffer A (20 mM HEPES [pH 7.0], 2 mM DTT, 1.5 mM MnCl2, 0.1 mg BSA per ml). The helicase reaction was performed in 100 ml helicase reaction solution (buffer A plus 2.5 mM ATP, and 10 U of RNasin) using 0.5 mg (unless otherwise specified) of purified NS3 helicase protein. The reaction was carried out at 37°C for 1 h (250 rpm). Unwound release strand and the helicase reaction solutions were then removed and washed six times with buffer A. (iii) Detection of remaining duplex substrate and interpretation of results. Substrate which was not unwound by the helicase was detected with peroxidase-conjugated anti-DIG (Boehringer), then developed with TMB (3,39-5,59-tetramethylbenzidine). Absorbance was measured using an ELISA reader (Molecular Devices) at 450 nm with a reference wavelength of 650 nm (A450 nm-A650 nm). To compensate for procedural variability and to determine the percentage of substrate unwound, a standard curve using dsRNA representative of varying amounts of DIG-labeled release strand was performed each time. The linear equation and R2 value of the linear portion of the standard curve were determined. Using the above derived equation, from the absorbance value one can interpolate the remaining amount (ng) of release strand (simply referred to as RA) bound to the template strand. The RA of each reaction was then compared to the RA of control reactions with no helicase. From this, one can determine the percentage of unwinding for each helicase reaction.

RESULTS The Efficacy of the RNA Helicase ELISA The principle of the RNA helicase ELISA is illustrated in Fig. 1. Procedures for the assay are described in Materials and Methods. Using 2 ng of template strand hybridized to varying amounts of release strand (0.2 ng–16 ng), a standard curve was first established in the absence of helicase to accomplish two objectives: 1) to determine the linear relationship between increasing amounts of release strand and measured absorbance, and 2) to find the maximum annealing capacity of the 2 ng of template strand used per reaction. As shown in Fig. 2A, a linear relationship exists between absorbance and amounts of release strand added below 6 ng. Absorbance plateaus at about 6.4 ng, signifying that template strands have annealed to the maximum amount of complementary release strands. Thus the ratio of 2 ng of template strand to 6.4 ng of release strand (1 mole template: 8 moles release) was used to provide for a maximal amount of duplex substrate bound to the helicase reaction well.

595

Vol. 253, No. 3, 1998

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

demonstrates that with an increase in the amount of helicase, there is a corresponding decrease in the amount of DIG-labeled release strand remaining bound to the template strand in the reaction well.

FIG. 1. (A). The double-stranded RNA substrate. The dsRNA substrate employed in the helicase reaction is displayed. The template strand (thick line) was labeled with biotin and the release strand (thin line) with digoxigenin. (B). Experimental design of the RNA helicase ELISA. The biotin-labeled template strand and the DIG-labeled release strand are in vitro transcribed, hybridized, and then transferred and immobilized on a streptavidin-coated microtiter plate (SA-MTP) which binds the template strand. In the substrate-coated wells, the helicase reaction occurs and the dsRNA is unwound. Remaining dsRNA with DIG-labeled release strands are detected by means of anti-digoxigenin coupled to peroxidase antibody (anti-DIG-POD) and the peroxidase substrate tetramethylbenzidine (TMB). Absorbance measurements are then made which correlate with the amount of remaining duplex substrate.

To evaluate effectiveness of the RNA helicase ELISA in comparison to the standard helicase assay, varying amounts (0.125 mg to 5 mg) of the recombinant NS3 helicase protein were incubated with substrate in both assays. With the RNA helicase ELISA, the resulting absorbance of each reaction inversely correlated with the amount of helicase protein added (Fig. 2B); this

FIG. 2. (A). The standard curve. 2 ng of template strand was hybridized to varying amounts of release strand (0.2 ng–16 ng). The duplex RNA was directly transferred to the SA-MTP, unbound RNA was removed, and the absorbance was determined (see Materials and Methods). The inset shows the linear region of the standard curve, its linear equation, and R2 value. (B). Absorbance measurements of the helicase ELISA to determine the RNA helicase activity. 0.125 mg to 5 mg of the HCV NS3 helicase were added to substrate coated SA-MTP for the unwinding reaction. The resulting absorbance (A450nm-A650nm ) of each reaction inversely correlated with the amount of helicase protein added. (C). Using the linear equation of the standard curve shown in (A), the absorbance values were translated into the percentages of unwinding for each reaction.

596

Vol. 253, No. 3, 1998

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

From the linear portion of the standard curve of this reaction, the linear equation (y 5 0.6023x 1 0.1101; y 5 absorbance, x 5 ng of release strand) can be determined (Fig. 2A), which is used to translate the absorbances into the amounts of release strand remaining bound and thus the percentages of unwinding. As Fig. 2C illustrates, the percentages of unwinding corresponded to increasing amounts of helicase protein and plateaued at approximately 85-90%. With the helicase ELISA, 0.5 mg of NS3 helicase protein unwound 72% of the dsRNA and 1 mg protein unwound 85%. It was then determined that 0.5 mg of protein was used per reaction afterwards because the corresponding percentage of unwinding had not plateaued and would enable the testing of both helicase inhibitors and enhancers. With the traditional helicase assay (6), using 0.5 mg and 1.0 mg of helicase protein resulted in percentages of unwinding of 90% and 97% of dsRNA, respectively (Fig. 3). These results indicate the ability of the RNA helicase ELISA to measure helicase activity is comparable to that of the standard assay. Use of the RNA Helicase ELISA to Screen Inhibitors It had been shown that increasing concentrations of KCl inhibits the NTPase activity (13) and the helicase activity (6) of the HCV NS3 helicase. To demonstrate the potential of the helicase ELISA to detect inhibition of helicase activity, recombinant enzyme was incubated with varying concentrations of KCl. As Fig. 4A illustrates, increasing concentrations of potassium ion significantly increase absorbance measurements; this reflects increasing amounts of DIG-release strand remaining bound to the template strand. These increases in absorbance correspond to decreases in the percentage of unwinding (Fig. 4B). At 250 mM KCl, approximately 80% of the original helicase activity has been inhibited. These results are consistent with our previous observations using the standard helicase assay (6). Other potential inhibitors of helicase activity were also tested, including ribavirin, a nucleoside analogue and antiviral agent currently used in HCV therapy; however, no inhibitory effects were demonstrated for any of the candidates (data not shown). Use of the RNA Helicase ELISA to Assay Helicase Mutants Point mutational studies of the NS3 helicase domain of HCV have been reported (15, 16). In all mutational analyses of motif I (GSGK210S) of the HCV NS3 helicase, replacement of the lysine residue resulted in little or no helicase activity in the mutant protein (15, 16). Using both the standard helicase assay and the helicase ELISA, our mutant K210N (and a BSA negative control in the helicase ELISA) indeed demonstrated no detectable helicase activity (see Fig. 5A). Concerning

FIG. 3. Standard assay for HCV RNA helicase activity. Doublestranded RNA substrate employed in the helicase reactions is displayed at the top of the figure. The release strand (thin line) is radiolabeled with [a-32P]-CTP. The standard helicase assay was performed as previously described (6). The unwound products are then visualized using PAGE. Lane 1 (-E), the reaction did not contain the enzyme and the substrate was left native; lane 2 (D), the reaction did not contain enzyme and the substrates were heat denatured; lane 3, the reaction mixture contained 0.5 mg of recombinant HCV NS3 helicase protein and 90% of the substrate was unwound; lane 4, the reaction contained 1.0 mg of protein 59-1175 and 97% of the duplex substrate was unwound. The intermediate bands (marked with *) between the double-stranded substrate band and single-stranded product band are probably release strands that may have formed intermediate secondary structures.

mutations of the histidine residue of motif II (DECH293), studies have provided conflicting results. Replacement of the histidine residue with alanine has resulted in mutants with activity ranging from approximately 2% of the wild-type (16) to 60% of the wild-type (15). In our standard assay, wild-type 59-1175 resulted in 84% of substrate being unwound while mutant H293A unwound 64% (Fig. 5A); accordingly, the H293A mutant had 76% of wild-type helicase activity (Fig. 5B). With the helicase ELISA, wild-type 59-1175 unwound 89% of the duplex substrate while the DECA mutant unwound 45% (Fig. 5A); mutant H293A had 50% of the wild-type activity (Fig. 5B). Thus, mutantto-wild-type activities as assayed by both assays were comparable. These results regarding mutant H293A are in agreement with those of Kim et al. whose DECA

597

Vol. 253, No. 3, 1998

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

because that certain limitations to the RNA helicase ELISA exist. First, the reaction volume of the standard assay is 1/5 that of the helicase ELISA, thus there is higher concentration of the substrates and the enzyme in the standard assay. Second, unlike the freely floating substrates used in the standard helicase assay, the helicase ELISA substrates are bound to streptavidin in the reaction wells. It is unknown the extent to which this affects the reaction kinetics; however, it has still been demonstrated that the helicase ELISA sensitively detects helicase activity. Another disadvantage of the helicase ELISA is that the reaction products cannot be visualized as with the standard helicase assay. Visualization of reaction products provides the advantage of recognition of RNase contamination. Nonetheless, relative to the standard helicase assay, the RNA helicase ELISA offers many benefits. Preparation of the substrates is much faster, simpler, and requires no radioactivity. Single-stranded substrates can be aliquotted and can be stored long-term at 280°C. Various single-stranded substrates (DNA or RNA) can be prepared and stored, offering a selection of substrates not limited by the half-life of a radioiso-

FIG. 4. Helicase ELISA detection of potassium ion inhibition of helicase activity. (A). Recombinant HCV NS3 helicase protein (0.5 mg) was incubated with varying concentrations (0 mM–250 mM) of KCl. Increasing concentrations of potassium ion significantly increase absorbance measurements. (B). Using the linear equation of the standard curve, the absorbance values were translated into the percentages of unwinding. The increases in absorbance correspond to decreases in the percentage of unwinding. At 250 mM KCl, approximately 80% of the original helicase activity has been inhibited.

mutant possessed approximately 60% wild-type helicase activity (15). DISCUSSION In this study, we have developed an RNA helicase ELISA which was demonstrated to be an effective assay for helicase activity. In terms of reliability and sensitivity, it has been proven to be comparable to the standard helicase assay. The RNA helicase ELISA can effectively detect slight differences in activity of varying amounts of enzyme (Fig. 2); it is also an assay which can be used for screening mutant proteins and drugs which may inhibit (or enhance) helicase activity. The data produced with the RNA helicase ELISA more or less correlated with results from the standard assay, though a little bit less activity in the former assay was always observed. This is

FIG. 5. Helicase ELISA detection of mutant helicase activity. (A) Helicase activity of mutant and wild-type proteins. Using both the standard helicase assay and the helicase ELISA, the helicase activities of mutants K210N and H293A were determined. A BSA negative control was also performed for the helicase ELISA. Wild-type 59-1175 unwound 84% of the substrate using the standard helicase assay; the helicase ELISA measured an unwinding percentage of 89%. H293A unwound 64% of substrate in the standard assay while unwinding 45% of dsRNA in the helicase ELISA. K210N demonstrated no detectable helicase activity as measured by either assay. (B) Mutant helicase activity relative to wild-type activity. The relative activity of the wild-type 59-1175 was defined as 100%. Using the standard assay, the relative activity of H293A was 76%. H293A relative activity was 50% using the helicase ELISA.

598

Vol. 253, No. 3, 1998

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

tope. The RNA helicase ELISA is also much faster, requiring less than one day for completion. Since the procedure is ELISA based, assaying and quantifying large numbers of samples is much simpler; the assay is feasible for large-scale screening of samples such as potential inhibitors and putative helicases. The HCV NS3 helicase domain has been demonstrated to have the ability to unwind DNA/DNA, RNA/DNA, and RNA/RNA substrates (6). Due to the ease with which the substrates can be changed, the helicase ELISA can be utilized to further probe the HCV helicase’s interactions with a variety of substrates. Furthermore, by choosing an appropriate substrate and properly modifying the helicase solutions, the principle of the helicase ELISA can be applied to any putative helicase, either DNA or RNA. With the increasing interest and recognized importance of viral helicases (1), the helicase ELISA may be a tool which can facilitate the study of this class of enzymes. In conclusion, the helicase ELISA we developed is an effective and reliable assay which can be used either singularly or in conjunction with the standard helicase assay. There are numerous applications for the helicase ELISA in the study of putative helicases. In light of the recent discoveries regarding the structure of the HCV NS3 helicase domain (11, 12), the design of potential helicase inhibitors may be forthcoming. With the seriousness of the clinical manifestations of HCV infection, it is hoped that this assay can expedite the search for anti-helicase drugs and facilitate the discovery of an effective therapy for HCV infection. ACKNOWLEDGMENTS This work was supported by grants from the National Science Council and the Department of Health, Executive Yuan, Taiwan.

This research opportunity for C.C.H. was made possible with a United States Fulbright Scholarship for Taiwan.

REFERENCES 1. Kadare, G., and Haenni, A. L. (1997) J. Virol. 71, 2583–2590. 2. Scheffner, M., Knippers, R., and Stahl, H. (1989) Cell, 57, 955– 963. 3. Shuman, S. (1992) Proc. Natl. Acad. Sci. USA 89, 10935– 10939. 4. Warrener, P., and Collett, M. S. (1995) J. Virol. 69, 1720 –1726. 5. Kim, D. W., Gwack, Y., Han, J. H., and Choe, J. (1995) Biochem. Biophys. Res. Commun. 215, 160 –166. 6. Tai, C.-L., Chi, W.-K., Chen, D.-S., and Hwang, L.-H. (1996) J. Virol. 70, 8477– 8484. 7. Grakoui, A., McCourt, D. W., Wychowski, C., Feinstone, S. M., and Rice, C. M. (1993) J. Virol. 67, 2832–2843. 8. Bartenschlager, R., Ahlborn-Laake, L., Mous, J., and Jacobsen, H. (1993) J. Virol. 67, 3835–3844. 9. Gorbalenya, A. E., Koonin, E. V., Donchenko, A. P., and Blinov, V. M. (1989) Nucleic Acids Res. 17, 4713– 4729. 10. Koonin, E. V. (1991) Nature 352, 290 (Letter). 11. Yao, N., Hesson, T., Cable, M., Hong, Z., Kwong, A. D., Le, H. V., and Weber, P. C. (1997) Nature Struct. Biol. 4, 463– 467. 12. Kim, J. L., Morgenstern, K. A., Griffith, J. P., Dwyer, M. D., Thomson, J. A., Murcko, M. A., Lin, C., and Caron, P. R. (1998) Structure 6, 89 –100. 13. Suzich, J. A., Tamura, J. K., Palmer-Hill, F., Warrener, P., Grakoui, A., Rice, C. M., Feinstone, S. M., and Collett, M. S. (1993) J. Virol. 67, 6152– 6158. 14. Sanger, F., Nicklen, S., and Coulson, A. R. (1977) Proc. Natl. Acad. Sci. USA 74, 5463–5467. 15. Kim, D. W., Kim, J., Gwack, Y., Han, J. H., and Choe, J. (1997) J. Virol. 71, 9400 –9409. 16. Heilek, G. M., and Peterson, M. G. (1997) J. Virol. 71, 6264 – 6266.

599