Understanding Helicases As a Means of Virus Control

Home , ATP synthase, Closterovirus, DnaB helicase, Geminiviridae, Hordeivirus, NS3 (HCV)

Current Pharmaceutical Design, 2006, 12, 1315-1338 1315 Understanding Helicases as a Means of Virus Control

D. N. Frick* and A. M. I. Lam

Department of Biochemistry & Molecular Biology, New York Medical College, Valhalla, NY 10595, USA

Abstract: Helicases are promising antiviral drug targets because their enzymatic activities are essential for viral genome replication, transcription, and translation. Numerous potent inhibitors of helicases encoded by herpes simplex virus, severe acute respiratory syndrome coronavirus, hepatitis C virus, Japanese encephalitis virus, West Nile virus, and human papillomavirus have been recently reported in the scientific literature. Some inhibitors have also been shown to decrease viral replication in cell culture and animal models. This review discusses recent progress in understanding the structure and function of viral helicases to help clarify how these potential antiviral compounds function and to facilitate the design of better inhibitors. The above helicases and all related viral proteins are classified here based on their evolutionary and functional similarities, and the key mechanistic features of each group are noted. All helicases share a common motor function fueled by ATP hydrolysis, but differ in exactly how the motor moves the protein and its cargo on a nucleic acid chain. The helicase inhibitors discussed here influence rates of helicase-catalyzed DNA (or RNA) unwinding by preventing ATP hydrolysis, nucleic acid binding, nucleic acid release, or by disrupting the interaction of a helicase with a required cofactor. Key Words: ATPase, DExD/H-box proteins, motor proteins, enzyme mechanism.

INTRODUCTION translation, disruption of RNA-protein complexes, and packaging of nucleic acids into virions. Most importantly, their Viruses evolve rapidly to evade not only the human im- validity as antiviral drug targets was recently confirmed mune system but also antiviral drugs. Although it is some- when compounds that inhibit a helicase encoded by herpes times possible to use only a single antiviral drug to control simplex virus (HSV) were shown to block viral replication DNA viruses that proofread and repair their genomes, single and disease progression in animal models [1, 2]. agents are rarely effective for viruses that evolve more rapidly. The most successful method to control such a virus is to Most of the best-studied viral helicases are encoded by simultaneously target multiple loci with several agents. Such large DNA viruses or bacteriophage. In fact, some of the first combination therapies successfully control human immu- helicases identified over two decades ago were purified from nodeficiency virus (HIV) replication to delay the develop- these organisms [3-5]. Similar proteins are encoded by sev- ment of AIDS, and are presently the standard of care for HIV eral important human pathogens like HSV and human pa- patients. Similar approaches could be used to combat other pillomavirus (HPV). About 10 years after the discovery of viruses, but the technique is limited because most present DNA helicases, similar proteins were found to be made by antiviral drugs function by inhibiting either viral proteases, RNA viruses [6]. Most species of positive sense single- which are needed to process viral proteins, or nucleic acid stranded RNA ((+)ssRNA) viruses encode a helicase, and polymerases, which are required to copy the viral genome. several of these proteins have now been extensively studied. Although a few other antiviral drugs function by blocking The most detailed structural and mechanistic studies have viral entry or by modulating the host immune system, more been performed using the RNA helicase encoded by hepatitis drug targets are clearly needed. This review focuses on the C virus (HCV). A few other helicases from pathogenic hu- development of another critical protein synthesized by many man RNA viruses have also been studied in detail including viruses, an enzyme called helicase, as a potential antiviral those from severe acute respiratory syndrome coronavirus drug target. (SARS-CoV), Dengue fever virus (DFV), Japanese encephalitis virus (JEV), and West Nile virus (WNV). Curi- Helicases are motor proteins that use energy derived ously, no retroviruses or negative sense single-stranded RNA from ATP hydrolysis to separate nucleic acid strands. Such ((-)ssRNA) viruses have been reported to encode the synthe- an activity is necessary at several points during genome rep- sis of a helicase. Since the members of these classes replicate lication. In viruses with duplex nucleic acid genomes, the in the nucleus, they might simply utilize helicases encoded double helix must be separated for copying. Conversely, by the host instead of their own proteins. Indeed, HIV repli- viruses with single-stranded genomes must separate duplexes cation has recently been shown to require the human DDX3 that form after genome replication. Helicases have also been DEAD-box RNA helicase [7]. Large DNA viruses that in- shown to be required in transcription of viral mRNAs, vade the nucleus, like HSV, frequently bring with them a full complement of replication proteins, including at least one *Address correspondence to this author at the Department of Biochemistry helicase. Viruses that replicate outside the nucleus almost and Molecular Biology, New York Medical College, Valhalla, NY 10595, USA; Tel: 914-594-4190; Fax: 914-594-4058; always encode a helicase, without which they cannot sur- E-mail: [email protected] vive.

Developing non-toxic helicase inhibitors is considerably This review will summarize the various viral helicases more difficult than developing drugs designed to inhibit that have been characterized to date, their evolutionary rela- other viral enzymes. Problems arise from several areas. First, tionships, mechanisms of action, and any inhibitors that have compared with proteases and polymerases, the mechanisms been reported in the scientific literature. The helicase litera- of the reactions catalyzed by helicases are still not as clearly ture is rapidly expanding but fortunately the field is fre- understood. Second, the helicase ATP-binding site is con- quently reviewed, and the reader will be directed to relevant served not only in all classes of helicases, but also in motor reviews in appropriate sections. It should also be noted at proteins, small GTPases, kinases, the AAA+ family (AT- this point that Delagoutte & von Hippel have recently re- Pases associated with various cellular activities), and even viewed the entire helicase field in an extensive two-part re- the mitochondrial ATP synthase (F1 ATPase) (Fig. (1)). view that is highly recommended [9, 10]. Thus, compounds that inhibit helicases via their ATP- binding sites could be quite toxic. Third, the role of helicases VIRAL HELICASE CLASSIFICATION in the viral lifecycle is still not well-defined. Although, in The best way to understand the bewildering amount of vitro all helicases are able to separate a nucleic acid strand from its complement, their movements could also rearrange viral helicase information is to realize that all helicases, from secondary structures or dislodge nucleic acid binding pro- both viruses and cellular organisms, share many common teins. Finally, the traditional assays measuring helicase- properties. Understanding these features will provide the catalyzed unwinding are tedious, making inhibitor screening basis for understanding the mechanism of action of these time-consuming. In the past few years, considerable progress complex enzymes. It should also be recognized that nature has been made in the area of assay development (see refer- has used the basic building blocks shared by all helicases to ence [8] for a review) and it is now possible to identify po- manipulate nucleic acids in many different ways and for tent helicase inhibitors using high throughput screening. The many different purposes. As a result, different helicase fami- challenge now is to understand how these compounds inter- lies have evolved that share little resemblance, at least super- act with helicases so that they can be developed into actual ficially. Thus, to really understand viral helicases, one must drugs. understand the common properties shared by all helicases, and the distinctive properties that characterize the various

Fig. (1). Evolutionary relationship of viral helicases. All known viral helicases belong either to one of three helicase superfamilies or to the RecA/F1-ATPase superfamily. The five families of viral helicases from DNA viruses and three families of viral helicases from RNA viruses are highlighted with grey bars. Prototypes of each family are listed in parentheses. SF3 and DnaB-like helicases are fundamentally different from SF1 and SF2 helicases in that they contain only one RecA-like domain per subunit and must form rings and/or filaments to catalyze ATP hydrolysis. All helicases are in the ASCE subdivision of P-loop NTPases and share many basic features with the numerous other protein families listed. The diagram is based on information found in references [234, 235]. Understanding Helicases as a Means of Virus Control Current Pharmaceutical Design, 2006, Vol. 12, No. 11 1317 helicase families. If a new viral pathogen is discovered, its discussed based on their functional similarities rather than genome sequence can be used to predict not only if the virus their phylogenic relationships. We begin with the DnaB-like encodes a helicase, but also exactly which helicase family in helicase family, then discuss SF1 DNA helicases, SF1 RNA which the putative helicase belongs. If the properties of that helicases, SF2 DNA helicases, SF2 RNA helicases, SF3 particular family are understood, it is likely that the helicase DNA helicases, and finally SF3 RNA helicases. The discus- of interest shares many of the same features. sion is focused on helicases from model organisms and those The evolutionary relationship of all known viral helicases encoded by important human pathogens, which are highlighted with subheadings. is outlined in Fig. (1). Based only on protein sequence analysis, Koonin and his colleagues have shown that all helicases Two of the best studied viral helicases are the DnaB-like can be placed in one of several genetic families [11]. All but proteins needed for replication of E. coli phages T7 and T4. two of the helicase families can be grouped into one of three These helicases are believed to unwind DNA by encircling larger “superfamilies, ” designated as superfamily 1 (SF1), one strand while displacing the other. While unwinding the superfamily 2 (SF2) [12], and superfamily 3 (SF3) [13]. The double helix, DnaB-like helicases also coordinate leading remaining 2 families are more similar to the RecA protein and lagging strand DNA synthesis. Replication fork coordi- and the F1 ATP synthase than helicases in the three helicase nation is accomplished by interactions with the primase that superfamilies. One family is similar to the DnaB helicase of synthesizes the RNA oligonucleotides needed to initiate new E. coli [14] and the other is similar to the E. coli Rho heli- Okazaki fragments (for a review see reference [18]). case that is used in transcriptional termination [15]. Of these Phage T4 is still perhaps the best model system to study two “odd” families, only the DnaB-like family contains viral the replication of large DNA viruses. The 168,930 base pair ( bacteriophage) proteins. The DnaB-like helicase family i.e. T4 genome with its approximately 300 genes (of which 62 is sometimes called family 4 (F4) [16], some authors refer to are essential) encodes analogs of almost all proteins needed it as superfamily 4 (SF4), and others mistakenly combine the for DNA replication [19]. The T4 helicase responsible for DnaB-like helicases with those in SF3. Sequence analysis coordinating leading and lagging strand replication is the reveals the DnaB-like family and the SF3 families are evo- product of T4 gene 41 (gp41). The gene 41 protein forms a lutionarily distinct [14], and as we will see below, the SF3 hexamer [20] that moves in a 5' to 3' direction along the lag- and DnaB-like helicases appear to function somewhat differ- ging strand DNA template. The gene 59 protein loads the ently. The various helicase families are distinguished by the helicase onto the DNA [21], and once loaded onto DNA, presence of specific conserved signature sequences, some of gp41 forms a tight complex with the T4 gene 61 primase which are conserved in all helicase families, some of which [22]. T4 encodes two additional proteins that have also been are slightly different between families, and some of which shown to possess helicase activity but which are not abso- are unique (see reference [16] for a review of these helicase lutely essential for viral replication under ideal conditions. motifs). Both proteins apparently unwind DNA as monomers not as Helicases are also classified based on their quaternary rings like gp41. One of these is the SF1 helicase Dda [23], structures and their direction of movement. Some helicases and the second is the SF2 helicase UvsW [24]. oligomerize to form rings (usually hexamers) that appear in The T7 DNA replication system is simpler than that of electron micrographs to encircle one nucleic acid strand (see T4 in that the primase and helicase are covalently tethered as reference [17] for a review). Other helicases do not form part of the same polypeptide, which is encoded by phage such rings. All helicases can also be classified according to gene 4. T7 gene 4 encodes two different proteins from two their movement relative to the nucleic acid strand to which separate in-frame start codons. The larger, 63-kDa gene 4 they are primarily associated. In most present models for protein, sometimes called gene 4A protein or gp4A, pos- helicase action, the helicase is depicted as moving along one sesses a zinc finger domain that is missing from the shorter strand of a double helix while displacing the other strand. 56-kDa protein, which is also referred to as the 4B protein or Since the strands in a double helix are oriented in an anti- gp4B. Both gp4A and gp4B combine to form a hexamer that parallel configuration, such movement could either occur in possesses both helicase and primase activities. The N- a 5'-3' direction or in a 3'-5' direction. The direction of heli- terminal portion of the gene 4 protein provides primase case movement can be easily diagnosed by analyzing the function [25], and the C-terminal portion provides helicase ability of a helicase to unwind duplex substrates possessing function [26]. The zinc finger domain missing from the 56- ssDNA (or ssRNA) tails. 5'-3' helicases only unwind sub- kDa gp4B helps guide T7 primase to certain sequences on strates with a 5'-tail, while 3'-5' helicases need a 3'-tail to the lagging strand template where primers are synthesized initiate unwinding. Thus, a helicase can be classified based [27, 28]. As a result, hexamers comprised of only the 56-kDa on each of the three above schemes. For example the heli- gene 4 protein lack primase activity but retain helicase activ- case encoded by HCV is a SF2, non-ring, 3'-5' RNA heli- ity [29]. case. HPV helicase is a SF3, ring, 3'-5' DNA helicase. There are no known viruses that infect animals (or plants) KNOWN VIRAL HELICASES AND THEIR PROPER- that express a helicase in the DnaB-like family. Neverthe- TIES less, other viral helicases are functionally similar. For example, as discussed below, the SF3 DNA helicases form rings Viruses encode helicases that are members of SF1, SF2, like those made by the DnaB-like proteins. Similarly, one of SF3, and the DnaB-like family. In Table 1, the known viral the helicases produced by HSV forms a tight complex with helicases are grouped according to virus genome structure, that virus’ primase and is thought to coordinate activities at order, family, and species. Below, key viral helicases are the DNA replication fork like the DnaB-like proteins. It is 1318 Current Pharmaceutical Design, 2006, Vol. 12, No. 11 Frick and Lam

Table 1. Viral Helicases

Genome Host Family (genus) Species Protein Activity [Reference] Family

dsDNA bacteria Myoviridae (T4-like) Enterobacteria phage T4 gp41 5'-3' DNA helicase [208] DnaB-like

Dda 5'-3' DNA helicase [209] SF1

UvsW RNA-DNA helicase [210] SF2

(P2-like) Bacteriophage P4 Gene a 3'-5' DNA helicase [211] DnaB-like

Bacteriophage P1 Ban DNA helicase [212] DnaB-like

Gene 4A, Podoviridae Enterobacteria phage T7 5'-3' DNA helicase [5, 213] DnaB-like 4B protein

Siphoviridae (l -like) Bacteriophage SPP1 G40P 5'-3' DNA helicase [214] DnaB-like

Autographa californica nuclear animals Baculoviridae p143 DNA binding [215] SF1 polyhedrosis virus

Herpesviridae Human herpesvirus I UL5 5'-3' DNA helicase [32] SF1

UL9 3'-5' DNA helicase [40, 41] SF2

Polyomaviridae Simian virus 40 Tag 3'-5' DNA/RNA helicase [216] SF3

Papillomaviridae Human papillomavirus E1 DNA helicase [217] SF3

Poxviridae Vaccinia virus NPH-I DNA-dependent ATPase [81] SF2

NPH-II 3'-5' RNA helicase [85, 218] SF2

A18R 3'-5' DNA helicase [83] SF2

VETF ATPase, DNA binding [84] SF2

ssDNA plants Geminiviridae Coconut foliar decay virus Rep ATPase, DNA binding [219] SF3

animals Parvoviridae Adeno-associated virus 2 Rep 3'-5' DNA helicase [105] SF3

Minute Virus of Mice NS-1 DNA helicase [220] SF3

dsRNA Viruses bacteria Cystoviridae Bacteriophage f 6 P4 5'-3' RNA helicase [221] SF3

animals Reoviridae Bluetongue virus VP6 5'-3' and 3'-5' RNA helicase [222] SF2

Bromoviridae (+)sense RNA plants Brome mosaic virus 1a Predicted [223] SF1 (Bromovirus)

Closteroviridae beet yellows closterovirus hel RNA binding [224] SF1 (Closterovirus)

Flexiviridae bamboo mosaic virus Orf1 RTPase [55] SF1 (Potexvirus)

Potato virus X TGNBp1 5'-3' and 3'-5' RNA helicase [225] SF1

Potyviridae plum pox virus C1 RNA helicase [6] SF2 (Potyvirus)

Tamarillo mosaic virus C1 RNA helicase [226] SF2

Tymoviridae turnip yellow mosaic virus p206 NTPase, RNA binding [227] SF1 (Tymovirus)

No assigned family poa semilatent virus TGNBp1 5'-3' and 3'-5' RNA helicase [225] SF1 (Hordeivirus)

barley stripe mosaic virus TGNBp1 5'-3' and 3'-5' RNA helicase [225] SF1

(Tobamovirus) Tobacco mosaic virus Rep RNA helicase [159] SF1 Understanding Helicases as a Means of Virus Control Current Pharmaceutical Design, 2006, Vol. 12, No. 11 1319

(Table 1) contd….

Genome Host Family (genus) Species Protein Activity [Reference] Family

animals Arteriviridae Equine arteritis virus nsp10 5'-3' RNA/DNA helicase [57] SF1

Caliciviridae Southampton virus p41 NTPase [110] SF3 (Norovirus)

5'-3' RNA/DNA helicase [56] Coronaviridae human coronavirus 229E nsp13 SF1 RTPase [52]

5'-3' RNA/DNA helicase [53] SARS coronavirus nsp13 SF1 RTPase [54]

Flaviviridae Yellow fever virus NS3 RNA stimulated NTPase [69] SF2 (Flavivirus)

RTPase [78] West Nile Virus NS3 SF2 3'-5' helicase [72]

3'-5' RNA helicase [70] Dengue fever virus NS3 SF2 RTPase [71, 77]

Japanese encephalitis virus NS3 3'-5' RNA helicase [73] SF2

Powassan virus NS3 RNA-stimulated ATPase [75] SF2

(Pestivirus) Bovine viral diarrhea virus NS3 3'-5' RNA helicase [228, 229] SF2

(Hepacivirus) Hepatitis C virus NS3 3'-5' RNA/DNA helicase[65] SF2

(unclassified) GB virus B NS3 3'-5' RNA helicase [230] SF2

Hepatitis G virus NS3 3'-5' RNA/DNA helicase [231] SF2

Picornaviridae Poliovirus 2C NTPase [106] SF3 (Enterovirus)

Echovirus 9 strain barty 2C NTPase [108] SF3

Togaviridae Semliki Forest virus nsP2 RNA helicase [232] SF1 (Alphavirus)

(Rubivirus) Rubella virus P70 RNA stimulated NTPase [233] SF1 the HSV helicase-primase complex that is the target of the (see references [30] and [31] for reviews). Within the HSV new anti-HSV drug candidates. helicase-primase complex, the NTPase activity is part of the UL5 protein [32, 33]. The UL52 protein contains the primase Herpes Simplex Virus Type UL5:UL52:UL8 Helicase active site that is responsible for synthesizing RNA primers on the lagging strand DNA. The UL8 protein does not ex- There are nine different human herpesviruses (HHV), hibit any enzymatic activity but serves as a co-factor to which can be grouped into three sub-families based on their stimulate both the helicase and the primase activities. UL8 genome sequence and biological characteristics. The type a also facilitates nuclear transport of the helicase-primase includes the herpes simplex viruses HHV-1 and HHV-2, and complex, and coordinates the helicase and primase activities Varicella zoster virus, which cause oral and genital herpes with those of other HSV proteins involved in DNA replica- simplex and chicken pox, respectively. The group includes b tion. human cytomegalovirus (HHV-5), HHV-6A, HHV-6B, and HHV-7, and the g class includes Epstein-Barr virus (HHV-4) The HSV helicase-primase complex unwinds duplex and Karposi’s sarcoma associated herpesvirus (HHV-8). Of DNA with a 5'-3' directionality and requires the presence of these, HHV-1 (also called herpes simplex virus-1 (HSV-1)) ICP8 and ATP, indicating that ATP fuels the process of un- is the best-characterized. The 152,000 base pair HSV-1 ge- winding, and that ssDNA binding proteins assist duplex nome encodes the synthesis of all the key proteins required separation by stabilizing partially unwound products [32, for its replication: an origin binding protein (UL9), a ssDNA 34]. UL5 possesses all the motifs that are conserved among binding protein (ICP8), a DNA polymerase (UL30), a po- SF1 proteins, and site-directed mutagenesis of the SF1 heli- lymerase processivity factor (UL42), and a helicase-primase case motifs leads to a complex that lacks an ability to un- complex consisting of the UL5, UL52, and UL8 proteins wind DNA [35, 36]. When UL52 is separated from UL5, 1320 Current Pharmaceutical Design, 2006, Vol. 12, No. 11 Frick and Lam

ATP hydrolysis is not stimulated by DNA and the protein no structural and nonstructural proteins. The helicase is part of longer unwinds DNA [37], indicating that UL52 is an inte- nonstructural protein 13 (nsp13), which is processed from gral part of the HSV helicase. Site-directed mutagenesis was the C-terminal portion of pp1ab [51]. SARS-CoV helicase is later used to confirm that UL52 interacts directly with DNA localized on the endoplasmic reticulum of SARS-CoV in- through a zinc finger domain [38] and that while UL52 fected cells, where RNA replication is likely to take place mainly contacts the ssDNA tail, UL5 interacts with DNA [52]. closer to the replication fork [39]. Thus, UL5 and UL52 Even though SARS-CoV is an RNA virus with no known work together at the replication fork to unwind the double DNA stage in its lifecycle, SARS-CoV helicase unwinds helix and simultaneously synthesize primers on the lagging both RNA and DNA duplexes in a 5¢ to 3¢ direction. To fuel strand DNA template. strand separation, SARS-CoV helicase hydrolyzes ATP or In addition to the above helicase-primase complex, HSV any of the eight canonical NTPs [53]. SARS-CoV helicase expresses a second helicase that is a member of SF2, not also has the ability to cleave the terminal phosphate from a SF1. UL9 is a 851 amino acid (94-kDa) protein that was triphosphate moiety linked to the 5¢ end of a RNA molecule originally isolated as a protein that specifically binds the [52, 54]. Using this RNA 5¢-triphosphatase activity (RTPase) HSV origin of replication [40]. Later, it was found that UL9 activity, SARS-CoV helicase is able to prepare viral RNA to unwinds DNA with a 3¢ to 5¢ directionality beginning at receive a 5¢ cap structure. The fact that ATP is a competitive these HSV origins [41-43]. inhibitor of the RTPase reaction catalyzed by SARS helicase suggests that the RNA is hydrolyzed at the same active site T4 Dda is another well-characterized SF1 DNA helicase. used to fuel helicase movement [54]. A similar RTPase that Dda was one of the first helicases purified because it can be plays a role in CAP formation was also found in the SF1 relatively easily separated from cellular ATPases using helicase encoded by bamboo mosaic virus [55] chromatography [3]. Dda acts as a monomer [44] and trans- locates in a 5' to 3' direction [45]. Soon after its discovery, it In addition to the SF1 helicase motifs, SARS-CoV nsp13 was noted that Dda allows replication to proceed past a protein, and its relatives from human coronavirus 229E [56] DNA-bound RNA polymerase, suggesting that the Dda heli- and equine arteritis virus [57], all contain a cysteine-rich zinc case unwinds a double helix and can also displace proteins binding domain located at their N-termini. Zinc binding do- bound to DNA [46]. Recently, a kinetic mechanism for such mains have been found linked to other helicases and they are protein displacement by Dda has been proposed based on frequently found in proteins that interact intimately with nu- measurements of the rate of Dda catalyzed displacement of cleic acids. For example, both the T7 and HSV helicases streptavidin from biotin labeled DNA [47]. have zinc binding domains that help facilitate primer synthesis by their attached primases, and the HCV helicase has a Several SF1 helicases have also been cloned from RNA zinc ion bound to its N-terminal protease region. When the viruses. Most of the RNA virus SF1 helicases are encoded by plant viruses, few of which have been studied in great zinc binding domain attached to nidovirus helicases is al- detail. A few SF1 helicases from animal RNA viruses have tered using site-directed mutagenesis, the virus displays de- fective RNA synthesis in infected cells [58] and the helicase been purified and characterized, but in general, less is known about SF1 RNA helicases than the above SF1 DNA heli- is not fully active [59]. cases. Some of the helicases in this family are made by Like the proteins in SF1, SF2 helicases generally do not pathogenic human viruses, the most noteworthy being the form rings. SF2 is almost as large as SF1 and likewise con- virus that causes SARS. tains cellular proteins, and proteins encoded by both DNA and RNA viruses. Many are key proteins for the replication SARS Coronavirus Helicase of RNA viruses, and they unwind duplex RNA structures before and/or after RNA genomes are copied by viral RNA- Severe acute respiratory syndrome (SARS) is a life- dependent RNA polymerases. Other SF2 helicases act during threatening form of pneumonia caused by an RNA virus transcription, translation, or to dislodge RNA binding pro- classified in the genus Coronavirus. The finding that the teins. Several SF2 proteins have been isolated from a variety SARS is a member of this group was surprising be- virus of viral pathogens; the best studied being the helicase en- cause most other coronaviruses cause only relatively mild coded by HCV. Many of the viral RNA helicases in SF2 respiratory illnesses (i.e. the common cold). The genus belong to a large family of evolutionarily related proteins is part of family that, along with Coronavirus Coronaviridae called DExD/H box proteins [60]. This family is named after the families and belongs to the Arteriviridae Roniviridae, a shared Asp, Glu, and Asp (or His) containing sequence in order . contains viruses with the Nidovirales Nidovirales the second conserved SF2 helicase motif (see references [61] largest known RNA genomes. The (+)ssRNA SARS-CoV and [62] for reviews of DExD/H box proteins). genome has about 29,700 ribonucleotides with fourteen open reading frames that encode both structural and replica- Hepatitis C Virus NS3 Helicase tive proteins [48]. The two largest open reading frames (ORF1a and ORF1b) overlap at the 5¢ end of the genome. HCV is the main agent responsible for non-A-non-B viral ORF1a and ORF1b are translated into two large polyproteins hepatitis, and the cause of a world-wide epidemic of chronic (pp): pp1a (~490 kDa) and pp1ab (~790 kDa). The N- liver disease. Current treatments using pegylated interferon terminus of pp1ab is identical to pp1a; the C-terminal half of and ribavirin are costly, produce severe side effects, and pp1ab is generated by ribosomal slippage [49, 50]. The eliminate detectable virus in less than 60% of HCV patients. SARS-CoV polyproteins are then processed by two viral HCV contains a (+)ssRNA genome with one main open proteases into mature viral peptides, which include both reading frame that encodes a long polypeptide about 3000 Understanding Helicases as a Means of Virus Control Current Pharmaceutical Design, 2006, Vol. 12, No. 11 1321 amino acids long. The genome is translated and the polypep- explain the ability of certain nucleoside analogs to inhibit tide is processed into structural and non-structural (NS) pro- flavivirus helicases but not HCV helicase [76]. teins by both cellular and viral proteases. While the struc- The flavivirus NS3 helicases also differ from HCV heli- tural proteins generate the viral capsid and envelope proteins, case in that flavivirus helicases possess RTPase activity, as the NS proteins are responsible for genome replication. RNA has been shown with both DFV [71, 77] and WNV [78]. synthesis is carried out by the NS5B RNA-dependent RNA Like the SARS-CoV helicase, the flavivirus helicases spe- polymerase. HCV helicase is part of the bi-functional NS3 cifically cleave the b-g bond of a triphosphate linked to the 5¢ protein, which possesses helicase, NTPase, and serine prote- end of a RNA in the first step needed to generate a RNA 5¢- ase activities. The two N-terminal NS3 domains provide terminal cap structure [71, 77, 78]. Although the next en- protease function, and the remaining three C-terminal do- zyme in the capping pathway, which transfers a GMP to the mains comprise the helicase activity. The NS3 helicase re- RNA 5' diphosphate has not yet been identified, flaviviruses gion was first shown to have RNA stimulated ATPase [63], encode a RNA cap methyltransferase (the NS5 protein) that and was later shown to bind RNA [64], unwind RNA [65, methylates the G5_ppp-5_N RNA cap at the 7 position of gua- 66] and DNA [67, 68]. nine [79]. Since HCV utilizes a 5¢ internal ribosomal entry There are approximately 12 distinct types of HCV, called site (IRES) instead of cap-dependent RNA translation, its genotypes, which have nucleotide sequences that differ by as RNA likely lacks the 5¢-cap structure present in flavivirus much as 30%. All of the various HCV genotypes are and coronavirus genomes. grouped into the genus , which is one of three Hepacivirus DNA viruses also encode members of SF2. As mentioned genera in . The other two genera are Flaviviridae Flavivirus above, both HSV and phage T4 encode SF2 helicases, the and , both of which have a genome organization Pestivirus UL9 protein and UvsW proteins, respectively (Table 1). In similar to HCV, with the helicase formed by the C-terminus general, SF2 helicases encoded by DNA viruses are not di- of the NS3 protein. The genus includes several Flavivirus rectly involved in coordinating proteins at the DNA replica- important pathogens such as the prototype yellow fever virus tion fork (like the DnaB-like and SF3 helicases). Rather, SF2 (YFV), Dengue fever virus (DFV), West Nile virus (WNV), proteins from DNA viruses are often found to be involved in and Japanese encephalitis virus (JEV). The Pestivirus genus DNA repair, recombination, transcription, or translation. contains viruses infecting non-human animals including Those needed for transcription are most similar to the RNA some that cause serious livestock infections, such as Bovine helicases in the DExD/H-box family. Some of the proteins in viral diarrhea virus (BVDV). this group are required for virus replication, but some are not because host factors can substitute for their functions. Flavivirus Helicases (YFV, JEV, DFV, and WNV Heli- cases) Four different proteins in the SF2 helicase family are encoded by the prototypic poxvirus vaccinia virus [80]. Vac- The mosquito-born YFV is endemic in sub-Saharan Af- cinia virus (VV) has been the choice model system to study rica and tropical South America and, although rarely fatal, transcription in part because the viral RNA polymerase and still poses a threat to unvaccinated travelers. Like YFV, there numerous transcription factors are packaged into each virion. is a vaccine for JEV, which is the leading cause of viral en- Vaccinia virus has a linear double-stranded DNA genome of cephalitis in Asia. Dengue fever, with symptoms that range 192,000 base pairs, which it replicates in the cytoplasm. VV from mild fever to a fatal hemorrhagic syndrome, is more genes are assigned letters and numbers based on a restriction serious because no vaccine is available. Both yellow and map of the virus. The four VV SF2 helicases are the NTP Dengue fever have seen resurgence since mosquito eradica- phosphohydrolase I (NPH-I) encoded by gene D11L [81], tion programs have been scaled back. WNV is a bird virus NTP phosphohydrolase II (NPH-II) encoded by I8R [82], the that is also spread to humans by mosquitoes and has recently A18R gene product [83], and the VV early transcription received much attention since its introduction to North factor (VETF) encoded by D6R [84]. Of the four, NPH-II is America. WNV normally causes only mild flu-like symp- the best characterized. As an RNA helicase, NPH-II can un- toms but compromised WNV patients may suffer severe neu- wind duplex RNA in a processive manner [85], and dislodge rological damage. RNA binding proteins as it moves in a 3'-5' direction [86, Although the NS3 protein from YFV has not yet been 87]. shown to unwind RNA, it does possess RNA stimulated Unlike the SF1 and SF2 helicases, SF3 helicases gener- NTPase activity [69]. Many related viruses including DFV ally form rings like the DnaB-like helicases. SF3 contains [70, 71], WNV [72], and JEV [73] have each been shown to many viral proteins. In fact, SF3 contains only viral proteins, unwind RNA in a 3¢ to 5¢ direction. Although nucleic acid which are derived from both DNA and RNA viruses. Al- stimulates ATP hydrolysis catalyzed by flavivirus, pestivi- though the best studied SF3 helicase is a protein synthesized rus, and hepacivirus helicases, different nucleic acids stimu- by SV40, which causes no known disease in humans, SF3 late the enzymes differently. Each enzyme possesses a dis- also contains proteins from key human pathogens, such as tinct nucleic acid stimulation profile, suggesting that they HPV and poliovirus. bind RNA (or DNA) in a sequence specific manner or unwind certain sequences better than others [63, 74]. Some The best studied SF3 helicases are those encoded by flavivirus helicases also contain an additional conserved nine small non-enveloped viruses with short, circular double- amino acid sequence upstream from the SF2 helicase motif 1 stranded DNA genomes. These viruses are grouped into the that is not present in HCV NS3 [75]. Called the “Q motif, ” families Polyomaviridae and Papillomaviridae, which until this region likely interacts with the adenine base and could recently were subfamilies in a family called Papovavirdae. 1322 Current Pharmaceutical Design, 2006, Vol. 12, No. 11 Frick and Lam

They are now grouped as separate families because their gins through its own DNA binding domain [99]. By inter- genomes have been found to be clearly different. Polyomavi- acting with E1, E2 loads the E1 helicase on the HPV DNA ruses have smaller genomes (~5,000 base pairs), and papil- origin. When E1 and E2 first interact, E2 blocks the E1 oli- lomaviruses have larger genomes (~8,000). gomerization interface, ensuring the proper assembly of the The prototype polyomavirus is simian virus 40 (SV40). E1 proteins at the HPV DNA origin [100]. E2 association SV40 has been intensely studied since it was found contami- with the E1-DNA origin complex is then destabilized upon nating early polio vaccines. SV40 is pathogenic to monkeys, binding of ATP to the E1 protein [100, 101]. The E1 mono- and causes tumors in hamsters, but fortunately seems to mers then oligomerize, forming a double hexamer around the cause no ill effects to humans who were inadvertently in- DNA substrate [102]. As in the SV40 system, the hexamers move away from each other because they are associated with oculated with SV40. Related polyomaviruses, such as the JC virus and the BK virus, infect the majority of the world’s complementary strands that have opposite polarities, and population causing no apparent harm except to severely im- their movement forms two replication forks at opposite ends muno-compromised patients, in whom JC virus can cause of a replication bubble [102]. the fatal neurological disorder progressive multifocal leu- SF3 DNA helicases are also encoded by viruses with koencephalopathy [88, 89]. The small polyomavirus genome ssDNA genomes, such as the human parvovirus adeno- encodes only six proteins, which are expressed from either associated virus (AAV), a small, apparently harmless virus early or late promoters. Three of the four proteins expressed used for gene therapy [103]. The 4700 nt ssDNA AAV ge- from the late promoter are capsid proteins. The mRNA tran- nome has only two open reading frames, one of which en- scribed from the early promoter is alternatively spliced to codes four Rep proteins that make a SF3 DNA helicase yield either small or large T antigens, so named because they needed for integration into the human genome. The Rep can be detected in animals bearing polyomavirus induced open reading frame encodes 4 different proteins, called tumors. The small t antigen (tag) and large T antigen (Tag) Rep78, Rep68, Rep52, and Rep40, by alternate RNA splic- share the same N-terminus, but differ at their C-termini. Tag ing. All four Rep proteins, including the smallest Rep40, (79-kDa) is much larger than tag, which is only 20-kDa. Tag share a central motor domain needed for integration. The assembles as two hexamers that each surround DNA when largest version, Rep78, forms a hexameric ring when bound they binds the SV40 origin of replication [90-92]. The duel to the AAV origin of replication [104]. Rep68 and Rep40 Tag hexamers both move in a 3' to 5' direction [93] with re- lack a C-terminal zinc finger domain that is present in Rep78 spect to the strand on which they are bound and act to open a and Rep 52. Rep40 and Rep52 lack the 224-amino acid N- replication bubble so that two replication forks can assemble terminal domain common to Rep78 and Rep68. Rep52 re- and proceed in opposite directions. In the process, Tag inter- tains an ability to unwind DNA in a 3' to 5' direction, but acts with and coordinates the action of numerous cellular does not apparently need to form rings to catalyze the reac- DNA replication proteins (for review see [94]). tion [105]. Surprisingly little work has been done with the SF3 heli- Human Papillomaviruses Helicase cases encoded by RNA viruses. One might suspect that they The human papillomaviruses are similar to polyomavi- are similar to SF3 helicases encoded by the small DNA viruses except that they have slightly larger capsids and larger, ruses in form and function, but there is as yet no direct evi- somewhat more complex, genomes. There are over 100 dence to support this contention. RNA viruses that infect known human papillomaviruses, which are numbered as both plants (families Comoviridae and Sequiviridae) and HPV-1 through HPV-96 (over 20 types are still unclassi- animals (families Picornaviridae and Caliciviridae) have fied). All HPVs infect the epithelial tissues, and the vast been predicted to encode SF3 helicases but none of the pro- majority “low risk” types only lead to benign warts or le- teins have been shown to unwind duplex RNA. Nevertheless, sions on the pharynx, esophagus, or genitals. A few “high the poliovirus 2C protein [106, 107], the polio-like virus risk” variants, such as HPV-16, HPV-18, and HPV-31, can echovirus 9 strain barty NS protein 2C [108, 109], and a cause cervical cancer [95]. The ~8,000 bp small closed- Norwalk-like virus 2c-like protein, called p41 [110] have the circular double-stranded DNA HPV genome contains several abilities to hydrolyze NTPs. Perhaps the conserved motor open reading frames that are translated and processed into function in these proteins is used for something other than eight early (E1 to E8) and two late (L1, L2) gene products RNA unwinding. [96, 97]. L1 and L2 form the viral capsid, and the early gene products are responsible for viral replication. The HPV heli- HELICASE STRUCTURE AND FUNCTION case is the E1 protein (for review see [98]). Over 20 different helicases have been studied at an HPV E1 contains a nuclear localization signal at its N- atomic resolution using x-ray crystallography and/or nuclear terminus, a site-specific DNA binding domain in the central magnetic resonance (NMR). Six of these helicases are pro- region, and an SF3 helicase domain within the C-terminal teins from viruses, and they include two non-ring helicases region. While the E1 C-terminal domain alone is sufficient from HCV and phage T4, and four ring helicases from phage for oligomerization, ATP hydrolysis and DNA unwinding, T7, AAV, HPV, and SV40. The structures reveal that even the entire E1 protein is needed for HPV replication in vivo though all helicases share a common fold surrounding the because the N-terminal domains specifically bind origins. ATP-binding site, ring (DnaB-like and SF3) and non-ring Stable formation of an E1 complex at the HPV DNA origin helicases (SF1 and SF2) harness the energy liberated from of replication requires the recruitment of E1 monomers by ATP hydrolysis to unwind DNA differently. ATP hydrolysis the HPV E2 protein, which also binds to the HPV DNA ori- by non-ring helicases causes a shift in two domains cova- Understanding Helicases as a Means of Virus Control Current Pharmaceutical Design, 2006, Vol. 12, No. 11 1323 lently linked on the same polypeptide, while in ring helicases consists of a series of beta-sheet sandwiched between two ATP hydrolysis leads to movement of the various subunits in sets of alpha-helices and is so critical to helicase function the ring relative to each other. The structures of several SF1, that all helicases contain at least two such domain. In the SF2, SF3, and DnaB-like helicases have been reported and non-ring helicases each protein monomer contains two the structures within each superfamily appear quite similar. RecA-like domains covalently attached to one another by a How these structures relate to the function of each class of short linker (Fig. (2a)). In ring helicases each subunit of the helicases is discussed below. Correlating structure and func- ring contains one RecA-like domain (Fig. (2b, 2c)). tion not only clarifies how helicases move along nucleic ac- The configuration of residues at the ATP-binding site is ids but also facilitates the rational design of helicase inhibi- also similar for all helicases. ATP and a required metal ion tors. 2+ cofactor (usually Mg ) bind to a helicase in the cleft that The first viral helicase structures reported were of the separates two adjacent RecA-like domains as diagramed in HCV RNA helicase (Fig. (2a)). The first studies utilized Fig. (3). The ATP-binding cleft is lined by the conserved recombinant proteins containing only the helicase portion of residues from the helicase signature sequences. Although the the HCV NS3 protein, which was isolated from genotype 1a exact composition of the signature sequences varies among (PDB 1HEI) [111] or genotype 1b (PDB 8OHM) [112]. The the helicase families, all helicases share two amino acid pat- HCV genotype 1a NS3 helicase fragment was also crystal- terns that were first described by John Walker [125], which lized with a bound DNA oligonucleotide (PDB 1A1V) [113]. are honorably referred to as the Walker A and Walker B mo- Later, a crystal structure of the full-length NS3 protein with tifs. The Walker A motif forms a phosphate binding loop (P- its NS4 cofactor covalently tethered to the recombinant pro- loop) with a conserved Lys that contacts the g phosphate of tein’s N-terminus was reported (PDB 1CU1) [114]. The ATP. The Walker B motif contains acidic residues that coor- atomic structure of an isolated domain of the 3-domain HCV dinate the positively charged divalent metal cation, which in helicase has also been determined in solution using NMR turn contacts the phosphates of ATP. The ATP-binding site (PDB 1JR6 and 1ONB) [115]. The structure of another SF2 of a helicase is completed by an Arg “finger” and a catalytic viral helicase, the UvsW protein from phage T4 (PDB 1RIF), base, which accepts a proton from the attacking water mole- has also been recently reported [24]. cule. In related proteins this catalytic base has been demonstrated to be a conserved Glu near the Walker B motif [126, The first ring helicase crystallized was the DnaB-like 127]. The Arg-finger points from a second RecA-like do- helicase of phage T7 (Fig. (2b)). The first studies utilized a short fragment of the T7 gene 4 primase/helicase protein main that is adjacent to the domain containing the P-loop and provides a positive charge that stabilizes the transition state lacking the primase and the linker region connecting the [128, 129]. In ring helicases the Arg-finger and P-loop are helicase to the primase. Because the linker region is necessary for hexamer formation, this fragment, called the 4E part of different polypeptide chains in the hexamer [130], explaining why ring helicases need to oligomerize in order to peptide, exists as a monomer in solution and is unable to catalyze a reaction. In non-ring helicases the Arg-finger is unwind DNA [26] but crystallized as a filament, which when part of the C-terminal RecA-like domain while the P-loop is viewed down the helical axis resembles a six-member ring on the N-terminal RecA-like domain. [116]. Structures are available for the T7 4E peptide alone (PDB 1CR0), with dTTP (PDB 1CR1), with dATP (PDB All theories explaining helicase function suggest that 1CR2), and with dTDP (PDB 1CR4) [116]. Later, a structure changes in the orientations of the RecA-like domains occur was reported for a larger fragment of the gene 4 protein, upon ATP binding and hydrolysis, and that these changes called the 4D peptide, that had been previously crystallized subsequently allow the protein to move along a nucleic acid [117]. The 4D fragment contains the primase linker, forms template. The RecA-like domains are thus analogous to a hexamers and unwinds DNA. Structures were determined for molecular motor and are therefore commonly referred to as T7 4D both alone (PDB 1E0K) and with the ATP analog 5'- “motor” domains. The motor domains move the helicase and adenylyl-b-g -imidodiphosphate (ADPNP) (PDB 1E0J) any attached cargo along a track of DNA or RNA. [118]. Most recently, a structure of the full-length 56-kDa gene 4 protein (4B protein) was reported (PDB 1Q57) [119]. Mechanism of Action of Non-Ring Helicases (SF1, SF2 The three SF3 viral helicases that have been examined Helicases) using x-ray crystallography all have similar overall folds that In addition to contacting ATP, the RecA-like domains in are somewhat different from the DnaB-like T7 helicase. The helicases also contact nucleic acids. When the first structures AAV helicase Rep40 has been crystallized without ligands of the HCV helicase were published, the authors speculated (PDB 1S9H) [120] and as a complex with ADP (PDB 1U0J ) that nucleic acid would bind in the cleft that separates the [121]. SV40 Tag (Fig. (2c)) structure has been determined in two RecA-like domains [111, 112]. The N-terminal RecA- the absence of ligands (PDB 1N25 [122] and PDB 1SVO like domain of HCV helicase (domain 1) contains the P-loop [123]), with ATP bound (PDB 1SVM) [123] and with ADP and the C-terminal RecA-like domain (domain 2) contains bound (PDB 1SVL) [123]. The helicase portion of the HPV several conserved arginines that form a basic patch that lines E1 protein (residues 428-629) has also been crystallized the inter-domain cleft. Although one of these arginines is along with the E2 viral protein, which aids its assembly at likely the Arg finger described above (Fig. (3)), they were at the origin (PDB 1TUE) [100]. the time believed to form an RNA binding motif. Because All helicase structures share several common features. the cleft that separates the third HCV helicase domain (do- First, they share a common fold that was first described in main 3) from domains 1 and 2 is lined with negatively the E. coli protein RecA [124]. This “RecA-like” domain charged residues that should repel RNA, molecular modeling 1324 Current Pharmaceutical Design, 2006, Vol. 12, No. 11 Frick and Lam

Fig. (2). Helicase Structures. A. HCV helicase structures [113, 114]. The N-terminal RecA-like domain (domain 1) is colored yellow, the C- terminal RecA-like domain is purple, domain 3 is pink, the protease is green, and the NS3 protease cofactor NS4A is blue. DNA and sulfate ions (which occupy the ATP binding site) are noted as spheres. B. T7 helicase structures [116, 118, 119]. The 4D protein lacks the primase domain of the T7 gene 4B protein (56-kDa gene 4 protein). The 4E protein is like the 4D protein but lacks a linker region necessary for hexamer formation. In the 4D and 4B structures, the different subunits are colored differently. The 4E filament structure is colored based on secondary structure. C. SV40 Tag helicase structures [123]. The different subunits in the SV40 hexamer are noted with different colors. In both B and C nucleotides are displayed as spheres. The noted PDB coordinate files were used to generate the structures using Pymol (DeLano Scientific LLC, San Francisco, CA). Understanding Helicases as a Means of Virus Control Current Pharmaceutical Design, 2006, Vol. 12, No. 11 1325

Walker A

Lys Walker B O 2+ Asp N O O + Mg N NH3 O N O O O O Catalytic E N O P O P O O O O P O H O O Glu O + H NH3 N N

Arginine Finger

Fig. (3). Mechanism of helicase-catalyzed ATP hydrolysis. All known helicases and related proteins coordinate an ATP, Mg2+, and a water molecule using a conserved Lys and Asp in the Walker A and B motifs on one RecA-like domain and an Arg on an adjacent RecA-like domain. A nearby Glu likely acts as a catalytic base by accepting a proton from the attacking water molecule. logically predicted RNA to bind in the other cleft, the one structures appear mechanistically relevant [141, 142], and which separates the RecA-like domains [111, 112]. It was suggest that DNA passes through one end of the helicase therefore somewhat surprising when a co-structure of a short when the protein is in the open conformation and the other oligonucleotide bound to HCV helicase showed the DNA when it is closed. When ATP is bound, the protein is closed bound to the cleft separating domain 3 from the RecA-like and the cleft opens again after hydrolysis and ADP release. domains (Fig (2a)) [113]. Extensive site-directed mutagene- Such movement would permit the protein to move like an sis of HCV helicase has since confirmed the validity of the inchworm. HCV helicase-DNA structure and demonstrated that the ob- The PcrA structures have also provided information served protein-nucleic acid contacts are needed for efficient about the nature of the forces that move the helicase along unwinding [131-136]. Furthermore, a similar DNA binding DNA. The PcrA-DNA structure differs from that of the cleft, which is perpendicular to the ATP binding cleft, has PcrA-DNA-NTP structure significantly in the DNA binding since been seen in co-structures of two other non-ring SF1 site. In both structures, several bases are flipped out of the helicases with DNA [137, 138]. Thus, in both SF1 and SF2 helix to interact with certain amino acids, but this flipping helicases, it appears that the single-stranded nucleic acid is pattern is different in the two structures. Velenkar et al. pro- bound in a cleft that is perpendicular to the ATP-binding pose that sequential ATP hydrolysis events lead to a wave site. like motion in the side chains that acts to move the protein How ATP hydrolysis is translated into helicase move- along ssDNA in a 3' to 5' direction [138]. Soultanas & Wig- ment on nucleic acid is the key question that is presently ley call this the “Mexican wave mechanism” because the being addressed in several laboratories. Many of the leading motion of the side chains is reminiscent of the “waves” that theories propose that movement of the RecA-like domains propagate through stadium crowds, which (though common upon ATP binding and hydrolysis allow the protein to move in the USA for decades) were first recognized internationally like an inchworm on one strand, while simultaneously dis- at the Mexican world cup [143]. This modified inchworm placing the complementary strand or nucleic acid binding mechanism implies that PcrA helicase would move only one proteins. The most convincing evidence for such a model nucleotide for each molecule of ATP hydrolyzed, and this comes from studies of a non-viral SF1 helicase encoded by hypothesis is supported by an analysis of the kinetics of oli- Bacillus stearothermophilus called PcrA. PcrA has been gonucleotide-stimulated PcrA-catalyzed ATP hydrolysis crystallized alone [139], in the presence of an ATP analog [144]. It is likely that SF1 viral helicases, such as the SARS [140], in the presence of DNA, and in the presence of DNA and HSV helicases, utilize such a method to move on DNA and an ATP analog [138]. These structures show movements or RNA. of the non-RecA-like domains upon DNA binding, and a While this wave-like action might work well for non-ring closure of the cleft between the RecA-like domains when SF1 helicases, SF2 helicases likely function somewhat dif- ATP binds, which coincides with a distortion of the DNA ferently. Although similar inchworm models have been pro- duplex. Based on the rigorous analysis of numerous PcrA posed to explain the movement of SF2 helicases, the struc- site-directed mutants, the interactions seen in the PcrA ture of HCV helicase reveals that the protein makes fewer 1326 Current Pharmaceutical Design, 2006, Vol. 12, No. 11 Frick and Lam direct contacts with nucleic acid bases than does PcrA [113]. observed that the binding of nucleic acid to a HCV helicase- Instead, most contacts are made with the phosphodiester ATP complex is pH dependent [149]. Thus, electrostatic backbone. The importance of several residues that contact forces may be as critical for helicase action as the mechani- the backbone in the HCV-DNA structure has been verified cal forces described by the Mexican wave model. Lam et al. using site-directed mutagenesis. Critical nucleic acid back- proposed that since negatively charged nucleic acid is held in bone-binding HCV residues include Thr269 [131], Thr411 a negatively charged protein cleft, the potential energy [131], Thr450 [74], Arg461[132] and Arg393 [136]. Only buildup could be utilized to propel the helicase along a nu- one HCV helicase amino acid side chain, that of Trp501, is cleic acid track. In this “propulsion-by-repulsion” model, known to contact the bases [113]. Trp501 stacks against the ATP binding leads to a conformational change such that the base at the 3'-end of the HCV bound oligonucleotide, and is nucleic acid bases can clear the Trp501 bookend. As dia- not in a RecA-like domain. Site-directed mutants with an gramed in Fig. (4), in the absence of ATP, RNA cannot exit alanine substituted at position 501 unwind RNA poorly [131- the enzyme because it is blocked by Trp501 and clamped in 135]. Because it stacks with a base, Trp501 is believed to the cleft by an Arg “clamp” on domain 2, which has been prevent DNA (or RNA) from sliding through the helicase in shown to be critical for binding and translocation [136]. the absence of ATP. Kim et al. [113] proposed a “ratcheting” When ATP binds, domain 2 rotates bringing with it the posi- inchworm model suggesting that when ATP binds, the cleft tively charged Arg-clamp. The Arg-clamp attracts the nega- between the RecA-like domains closes, and Trp501 slides tively charged phosphodiester backbone so that the RNA past one or two nucleotides. When ATP is hydrolyzed, the moves free from the bookend. The negatively charged RNA cleft opens, and Trp501 acts as a bookend so that the protein is then repelled by the negatively charged binding cleft, so it moves towards the 5'-end of the nucleic acid. Supporting this moves through the protein until ATP is hydrolyzed, and the model are the observations by Porter et al. that only a few protein clamps tightly again. Thus, the step size of the pro- base pairs of fluorescently-labeled DNA are unwound by tein would depend on the nature of the nucleic acid on which HCV helicase in a single turnover event and that the helicase the protein is translocating. For example, since RNA is more binds DNA weaker in the presence of ATP [145]. Others polar than DNA, one would predict a larger step size on have since confirmed that ATP binding weakens the affinity RNA than DNA as has been observed [147, 148]. Also, a of HCV helicase for DNA and RNA [136, 146]. However stronger repulsion of RNA would cause the enzyme to fall there is yet no general agreement on the number of base pairs from that substrate before it would fall from DNA. Such a unwound in a single event (called “step size”). Levin et al. lower processivity for HCV helicase-catalyzed RNA un- [147] have calculated a step size of 9 base pairs using unla- winding than DNA unwinding has been frequently reported beled DNA, and using a long RNA substrate Serebrov & [136, 150, 151]. Pyle have determined that 18 base pairs are unwound by The above inchworm models all predict that a SF1 or HCV helicase in a single step [148]. SF2 helicase can function as a monomer but do not account Insights into the force which propels the nucleic acid for all observed behaviors of non-ring helicases. For exam- through HCV helicase were recently made when Lam et al. ple, with HCV helicase, unwinding rates are not linearly

Fig. (4). The electrostatic inchworm model for HCV helicase movement. In the model, ssRNA is held in a negatively charged cleft separating the RecA-like domains from a third domain. A bookend residue (Trp501 in HCV helicase) prevents ssRNA from sliding through this cleft. Upon ATP binding, the RecA-like domains rotate, moving positively charged RNA-bound residues (e.g. Arg393 in HCV helicase), which in turn move the RNA so that it clears the bookend. Charge repulsion between the RNA and the negatively charged cleft causes the protein to slide along the helix. Based on information from references [136, 149, 236]. Understanding Helicases as a Means of Virus Control Current Pharmaceutical Design, 2006, Vol. 12, No. 11 1327 dependent on the amount of protein present in the reaction (Fig. (2b)) the channel is only large enough to accommodate but rather accelerate greatly once a critical protein concen- one DNA strand, suggesting that the other must wind outside tration is reached [136, 151]. Yeast two-hybrid assays also the ring as shown in Fig. (5a) [118]. suggest that HCV helicase forms a dimer [152, 153], and that residues Thr266, Tyr267 and Met288 in domain 1 are critical If DNA does, in fact, pass through the center of T7 helicase, it would make contact with loops that contain residues for dimerization [153]. Kinetic models explaining this coop- that were previously shown using mutagenesis to be in- erativity have recently been presented [147, 154], and it re- volved in DNA binding, like Arg487 [166, 167]. These resi- mains possible that HCV and other non-ring (SF1/2) viral dues are part of the loops that extend from the RecA-like helicases could function as dimers as has been proposed for domains into the central channel. The DNA would therefore other helicases like the E. coli Rep [155] and UvrD helicases [156, 157] (for a review of the “rolling dimer” helicase be perpendicular to the ATP binding cleft, but at an angle mechanism see [158]). There is also the possibility that all somewhat different from that seen in the SF1 and SF2 heli- helicases form rings or filaments on nucleic acid, and we case structures. It is also interesting to note that the subunits have yet to find the proper conditions to observe these rings of the T7 helicase do not form a symmetrical hexamer. Rather, each subunit of each half hexamer is rotated 15º with the SF1 and SF2 helicases discussed above. For exam- relative to each other. This means that each DNA binding ple, a purified recombinant form of a SF1 helicase encoded by tobacco mosaic virus is capable of forming hexameric loop is also rotated relative to the loops in adjacent subunits rings that can be observed using both gel filtration chroma- (Fig. (5c)). Singleton et al. [118] have proposed that this tography and electron microscopy [159]. subunit rotation is modulated by the binding and hydrolysis of ATP. Thus, as the binding loops make sequential contacts with the DNA backbone, they rotate and push the ssDNA Mechanism of Action of Ring Helicases (DnaB-like, SF3 through the hexamer. In support of this model, when NTPs Helicases) are added to T7 helicase crystals, not all the potential sites ATP binding and the mechanism of its hydrolysis by ring are occupied, and the side chain configurations are different helicases is basically identical to that outlined above for non- surrounding each of the three binding site types seen in the ring helicases (Fig. (3)). The main difference between ring asymmetric hexamer [118]. and non-ring helicases is that the Arg finger is on a separate polypeptide chain. Since in ring helicases, there is only one The ring structure of SF3 and DnaB-like helicases raises RecA-like domain, each subunit has both a P-loop and an the question of how ring helicases are loaded onto DNA to Arg-finger. The subunits are arranged in a head to tail man- initiate unwinding. In vitro, ring helicases require a forked ner, with the head being the P-loop, and the Arg-finger being molecule with ssDNA tails to initiate unwinding, but in vivo, the tail. The ATP is bound between the head and tail, and unwinding begins within a duplex at replication origins. there is the same number of binding sites as there are Some viruses express AAA+ proteins that use ATP to load the helicase onto DNA. An example is the gene 59 protein of subunits in the oligomer (Figs. (2) and (5)). Even though T4 [21]. Other helicases, like T7 helicase, appear to be able there can be up to six ATP molecules bound per hexameric helicase ring, fewer sites are often observed. The DnaB-like to break open the ring to bind DNA. For example, not all helicases, like the F ATPase, bind ATP with negative coop- crystal structures show the T7 helicase as a ring. In the first 1 structure the subunits packed like a filament (Fig. ( )) erativity. This means that the binding of the first ligands 2b [116], and such a filament could be the conformation used cause a decrease in the affinity for subsequent ligands. As a by the protein to slip on an off DNA. result, only 2-4 NTPs bind per hexamer [160]. The kinetics of NTP hydrolysis catalyzed by T7 helicase are also re- In the latest structure of the full length T7 gene 4 protein, markably similar to those seen with F1 ATPase [161]. In both the protein is seen as a seven subunit ring instead of a hex- proteins, it appears that ATP is hydrolyzed sequentially by amer (Fig. (2b)) [119]. In the T7 heptamer structure the cen- different subunits. In the case of the F1 ATPase, this leads to tral channel is large enough to accommodate a duplex. A a rotation of the g subunit, which in turn contacts the mem- similar large channel has been reported in SV40 Tag struc- brane embedded Fo proton channel [162]. When protons pass tures (Fig. (2c)) [122, 123]. A larger channel suggests that a through the Fo complex, the reverse rotation leads to ATP double helix could pass through the protein, and such an synthesis [163]. In the T7 helicase, the DNA is proposed to arrangement is diagramed for Tag in Fig (5b). SV40 Tag be analogous to the g subunit of F1 ATPase, and the rotation was recently crystallized alone, with ADP, and with ATP leads to the helix moving through the oligomer. with interesting results that support a fundamentally different The above “rotational” model predicts that one or both model for translocation than the rotational model describing strands of DNA pass through the center of the hexamer, as T7 helicase action. First, all six potential SV40 Tag NTP diagramed in Fig. (5). Although ring helicases have not yet binding sites bind nucleotides. This evidence would support been crystallized in the presence of DNA, this notion is sup- the more traditional Monod-Wyman-Changeux “concerted” ported by electron micrographs and model building. In the model of ligand binding [168]. Second, upon NTP binding, presence of DNA, extra density appears in the central chan- changes in subunit rotation create an “iris-like” motion so nels of hexamers formed by both T7 helicase [164] and that the diameter of the channel changes and six b hairpins, SV40 Tag [165]. Also, electrostatic analysis of hexameric suspected to bind DNA, change orientation within the chan- helicase structures reveals in all cases that the central holes nel. The motion of the hairpins could push DNA through the are lined with positive charges [116, 118, 120, 122, 123], center of the protein as has been proposed with T7 helicase, which would attract negatively charged DNA. In crystal but all binding loops move together rather than in pairs. structures of the T7 helicase portion of the gene 4 protein Thus, SF3 helicases could be fundamentally different from 1328 Current Pharmaceutical Design, 2006, Vol. 12, No. 11 Frick and Lam

DnaB-like helicases in that ATP hydrolysis occurs via a con- act with other proteins. Helicases are usually intimately as- certed rather than a sequential mechanism. Both the sequen- sociated with other proteins that travel with them along a tial model (Fig. (5a)) and concerted model (Fig. (5b)) predict nucleic acid chain. Many helicases form a tight complex that DNA movements are controlled by a binding loop (or with one or more other proteins and sometime the helicase is hairpin) in the center of the ring that remains connected to covalently tethered to a separate independent functional do- the same point on the DNA backbone during a single cycle main with an entirely different biological activity. For exam- of ATP binding and hydrolysis. As diagramed in Fig. (5c), as ple in many RNA viruses, including HCV and SARS-CoV, the hairpin moves, so does the protein relative to the DNA. the helicase is part of a protein that also possesses a protease function (Fig. (2a)). No relationship between helicase and protease has been clearly defined, but the helicase might, for example, allow the protease to travel with cellular translation machinery to help coordinate viral polyprotein processing. In a similar example, helicases that coordinate semi-disconti- nuous DNA replication, such as T7 and HSV helicases, are found coupled to a primase that synthesizes primers on the lagging strand (Fig. (2b)).

POTENTIAL ANTIVIRAL AGENTS SF1 Helicase Inhibitors In addition to the HSV helicase-primase inhibitors that are being developed as antiviral drugs [1, 2], inhibitors of other SF1 RNA helicases such as the SARS-CoV helicase, have been reported [169]. Both classes of compounds could serve as templates to develop inhibitors of any SF1 helicase (Fig. (6)). Current HSV therapies using nucleoside analogs such as acyclovir are quite effective but frequently lead to the evolution of resistant viruses. Since chain terminating nucleoside analogs must be activated before incorporation by HSV polymerase, the virus can develop resistance by not expressing a functional kinase or by evolving a polymerase with a higher fidelity. Because the new helicase-primase inhibitors do not require metabolic activation, the virus has fewer pathways through which it can develop resistance. The new series of HSV helicase-primase inhibitors, currently in pre- clinical development at Boehringer Ingelheim [1] and Bayer [2], offer a new option for treating acyclovir-resistant infections, and appear more effective than current drugs for treating latent HSV infections. Representatives of the two series of compounds are shown in Fig. (6a). The Boehringer Ingelheim series (e.g. BILS 179 BS) was first identified by screening inhibitors of DNA unwinding, while the Bayer series (e.g. BAY 57-1293) was discovered using a cell-based viral replication assay. Fig. (5). The sequential and concerted models for ring helicase The aminothiazolylphenyl compound BILS 179 BS was action. A. In the sequential “rotational” model for T7 helicase identified at Boehringer Ingelheim as a HSV UL5/UL52/ movement, the various subunits hydrolyze ATP in turn to cause the UL8 inhibitor using a high throughput helicase assay [1]. A protein to rotate along one strand of DNA while displacing the later derivative called BILS 45 BS, which is absorbed more complementary strand [118]. B. In the concerted model for SV40 rapidly into the blood, differs only by the lack of a methyl helicase action, all subunits simultaneously bind and hydrolyze group [170]. In the presence of the inhibitor, HSV primase ATP [123]. C. Both models predict that ATP hydrolysis leads to a activity is inhibited with an IC50 of 0.15 mM, while the DNA rotation of DNA binding loops that contact DNA in the center of unwinding activity of the HSV helicase is inhibited with an the hexamer. If the binding loops remain attached to the same point IC50 of 1.3 mM. BILS 179 BS also inhibits DNA-stimulated on DNA during an entire ATP hydrolysis cycle, then the protein ATP hydrolysis (IC50 ~0.43 mM) but does not affect the ATP will move along DNA as diagramed. hydrolysis in the absence of DNA (IC50 >100 mM). DNA binds more tightly to an UL5/UL52/UL8 complex in the Outside the RecA-like domains helicase structures differ presence of BILS 179 BS. Thus, BILS 179 BS likely inhibits considerably. The non-RecA-like domains act to confer HSV helicase-primase activity by preventing DNA release unique properties to the helicase, which might involve addi- during translocation [1]. tional nucleic acid binding motifs, or motifs needed to inter- Understanding Helicases as a Means of Virus Control Current Pharmaceutical Design, 2006, Vol. 12, No. 11 1329

As measured by plaque formation, BILS 179 BS de- with HSV-1 [2, 173]. In addition, BAY 57-1293 reduces creases HSV DNA replication in BHK cells infected with healing time and is effective even when treatment is delayed, either wild type HSV strains or acyclovir-resistant mutants as opposed to the nucleoside analogs which require early (EC50 ~100 nM). HSV infected mice fed BILS 179 BS, or its application upon HSV infection [2]. derivative BILS 45 BS, have fewer and smaller cutaneous A cell-based assay similar to that used to discover the and genital lesions than mice fed a placebo. HSV viral titers BAY series of HSV helicase inhibitors was also used to find also show a dose dependent reduction following administra- compounds that inhibit cell death induced by SARS-CoV. tion of the BILS compounds [1, 170]. Mice infected with Kao et al. found 104 compounds that protect cells from acyclovir-resistant HSV also respond to oral administration SARS-CoV induced cytopathic effects [169]. Of these, seven of BILS 45 BS [170]. Thus, these aminothiazolphenyl-based were potent inhibitors of recombinant purified SARS-CoV molecules are a promising class of compounds for the treat- helicase-catalyzed DNA unwinding and ATP hydrolysis. ment of nucleoside-resistant HSV disease in humans. The structure of one, which also suppresses viral plaque In order to determine which protein component of the formation in SARS-infected tissue culture cells with an EC50 HSV helicase-primase complex is targeted by the BILS of 6 mM, is shown in Fig. (6b)). HE602 inhibits nucleic acid- compounds, HSV was incubated with cells in the presence of stimulated ATP hydrolysis by SARS-CoV helicase with an BILS 22 BS (an analog of BILS 45 BS). Three mutant vi- IC50 of 6.9 mM, but has no effect on its basal rate of ATP ruses were found to have mutations in their UL5 gene [171]. hydrolysis. One had Gly352 changed to Val, one had Gly352 changed to Cys, and the third had Lys356 changed to Asn, indicating A H that the UL5 subunit of the HSV helicase-primase is likely N O N the target of the BILS inhibitors. Gly352 and Lys356 are H2N N near SF1 conserved motif IV, with Gly352 only 2 amino O acids downstream from SF1 motif IV. DNA-stimulated ATP S CH N hydrolysis catalyzed by purified UL5/UL52/UL8 complexes 3 with these mutations is less sensitive to the inhibitors. Both Gly352 and Lys356 are completely conserved in all human herpesviruses [172], but the mutant viruses are viable and BILS 179 BS still cause disease [171]. The frequency with which BILS resistant mutants arise is significantly less than that seen for acyclovir-resistant mutants [171]. CH3 O Unlike the Boehringer Ingelheim compounds, which N O N were identified using high throughput enzyme assays, the S NH2 N S development of BAY 57-1293 (Fig. (6a)), was initiated us- O ing a cell-based high throughput assay [2]. In this assay, CH3 which measures HSV induced cytopathogenicity, BAY 57- BAY 57-1293 1293 (IC50 = 20 nM) is 50-fold more potent than acyclovir (IC50 = 1 mM). In addition, BAY 57-1293 inhibits acyclovir- resistant HSV-induced cell lysis with the same potency as B wild type HSV-induced cell damage. O H The first indication that the HSV helicase-primase was N N the target for the BAY compounds came when resistant mu- N tants were sequenced. Seven amino acid substitutions were Cl O O identified, six of them clustered near the N-terminal of the UL5 gene and one mutant had a substitution in the UL52 HE602 gene [2]. As with the BILS compounds, BAY 57-1293- resistant mutants contain mutations (G352V, M355T, Fig. (6). SF1 helicase Inhibitors. A. HSV helicase-primase inhibi- K356Q) near SF1 motif IV. Remarkably, two of these are at tors [1, 173]. B. SARS-CoV helicase inhibitor [169]. the same positions as the BILS-resistant mutants. All three positions are completely conserved in all human herpesvi- SF2 Helicase Inhibitors ruses. BAY 57-1293 resistance occurs about 10 times less frequently than acyclovir resistance [2]. Numerous compounds have been reported that inhibit the HCV, WNV, and JEV helicases. These inhibitors can be Biochemical assays using purified HSV helicase-primase classified into three basic groups: (1) nucleoside analogs and also show that BAY 57-1293 inhibits the DNA-stimulated related small molecules that modulate ATP hydrolysis and/or ATPase activity in a dose-dependent manner with an IC of 50 nucleic acid unwinding, (2) nucleic acids and modified de- ~30 nM. Orally administered BAY 57-1293 has potent anti- rivatives that specifically target the RNA-binding site, and viral efficiency in mice, rats and guinea pigs infected with (3) antibodies. HSV-1 or HSV-2. BAY 57-1293 is 40 times more effective than acyclovir in treating mice infected with cutaneous HSV- Small molecule inhibitors of HCV helicase and related 1, suppresses intravaginal lesions in HSV-2 infected guinea helicases were recently reviewed by Borowski and his col- pigs, and exhibits a potent antiviral effect in rats infected leagues in references [174] and [175]. One of the most 1330 Current Pharmaceutical Design, 2006, Vol. 12, No. 11 Frick and Lam promising of these early inhibitors is 1-(2'-O-methyl-beta-D- inhibit ATP hydrolysis, but would modulate ATP hydrolysis ribofuranosyl)imidazo[4, 5-d]pyridazine-4, 7(5H, 6H)-dione and/or unwinding by binding an allosteric site. Several such (HMC-HO4) (Fig. (7)), which inhibits WNV helicase and compounds recently have been reported [186-188]. Repre- produces a potent antiviral effect (IC50 = 25-30 mM) in sentatives are shown along with HMC-HO4 in Fig. (7). WNV-infected cells [176]. Interestingly, a lower concentra- A series of halogenated benzimidazoles and benzotria- tion of HMC-HO4 is needed to inhibit viral RNA synthesis zoles, which were originally reported as HCV inhibitors in in cells than is needed to inhibit DNA (and presumably patents by Viropharma Inc., were recently examined for their RNA) unwinding by purified WNV helicase. At the lower ability to inhibit HCV, WNV, and JEV helicases [186]. Di- HMC-HO4 concentrations that produce an antiviral effect chloro(ribofuranosyl)benzotriazole (DRBT) and di- ( below 20 M), DNA unwinding rates catalyzed by the i.e. m chloro(ribofuranosyl)benzimidazole (DRB) are nucleoside purified enzyme are actually stimulated. At higher HMC- analogs, while tetrachlorobenzotriazole (TCBT) and tetrab- HO4 concentrations, unwinding is inhibited as is viral RNA romobenzotriazole (TBBT) are base analogs. DRBT, DRB, synthesis. At all concentrations of HMC-HO4, ATP hydroly- TCBT, and TBBT inhibit WNV helicase catalyzed RNA sis is stimulated, suggesting that the inhibitor somehow un- unwinding with IC ’s in the low micromolar range, but only couples the ATPase and helicase functions [176]. 50 TBBT (Fig. (7)) inhibits RNA unwinding by HCV NS3 heli- Another explanation for the effect of HMC-HO4 on case (IC50 ~60 mM). Interestingly, none of these inhibit the WNV helicase is that the compound binds to a second NTP JEV helicase, even though it is closely related to WNV heli- binding site other than the one made by the cleft between the case. Under the conditions used, none of the compounds RecA-like domains. Some evidence for such a site has come inhibited helicase-catalyzed ATP hydrolysis, suggesting that from studies of the related HCV helicase. The HCV helicase they may bind to another NTP-binding site or interact with hydrolyzes all eight canonical nucleoside triphosphates [74, the nucleic acid substrates [186]. 177, 178], but some reports suggest that not all NTPs fuel equal rates of unwinding. For example, Locatelli et al. found O that only some NTPs fuel unwinding with an efficiency N comparable to that seen with ATP whereas other (d)NTPs, NH Br particularly dATP, were found to be poor substrates and po- Br tent inhibitors of unwinding [179]. Although other studies N NH N O have not confirmed their observations [178, 180], Locatelli HO N O N et al.’s report [179] suggests that NTPs could bind at two Br H sites, one used to fuel helicase movement, and a second that Br allosterically modulates the enzyme’s activity. HO OMe Other evidence for a second site arises from product in- HMC-HO4 TBBT hibition studies. The products of ATP hydrolysis, ADP and O Pi, do not inhibit the HCV NS3 helicase. However, in the H (CH2)13CH3 presence of NaF and PolyU RNA, ADP inhibits ATP hy- N N N drolysis with about two moles of ADP binding per protein H monomer [181]. When beryllium fluoride is added to the N reaction, ADP inhibits ATP hydrolysis more potently with a N Ki of about 8.5 mM [136, 146]. In structures of other en- O O zymes, BeF3 is often found to occupy a position analogous to 2+ OH the g phosphate of ATP when bound with ADP and Mg HO [182]. Thus, ADP(BeF3) is likely a non-hydrolyzable ATP OH REN Compound 17 analog, but unlike other non-hydrolyzable ATP analogs such as ADPNP [177, 183], ADP(BeF3) binds HCV helicase Fig. (7). SF2 helicase inhibitors. A. imidazo[4, 5-d]pyridazine nu- tightly. ADP(BeF3) binds with a stoichiometry of one nucleotide per protein monomer and mimics ATP in that it re- cleoside inhibitor of WNV helicase [176]. B. Halogenated benzotri- duces the affinity of the helicase for nucleic acids in the azole inhibitor of WNV helicase and HCV helicase [186]. C. Ring- expanded nucleoside inhibitor of WNV helicase [187]. same manner as ATP [136, 146, 149, 184]. ADP(BeF3) also binds other P-loop ATPases as a ground state ATP analog [182], explaining the cellular toxicity of beryllium fluoride. Several ring-expanded “fat” nucleosides (RENs), which ADP fluoride complexes, on the other hand, likely do not are currently being investigated as antiviral and anticancer resemble the substrate as closely, so they may bind both ac- agents (see ref. [189] for a review), have also recently been tive and allosteric sites. Other evidence of a second ATP examined as potential inhibitors of the WNV, JEV, and HCV binding site was presented in a second report by Locatelli et helicases [187, 188]. At least five different RENs inhibit al. that demonstrated that nucleotides bind HCV helicase WNV, one inhibits JEV, but none inhibit HCV helicase cooperatively [185]. catalyzed DNA or RNA unwinding. Many of the RENs, such as compound 17 (Fig. (7)), are more potent inhibitors of ei- Thus, it is possible that the interaction between HCV ther WNV helicase-catalyzed RNA or DNA unwinding, sug- helicase and nucleotides could be effectively exploited for gesting that they exert their effects by binding the major or drug design by finding compounds that bind the second pu- minor groove of the duplex nucleic acid. Only one REN was tative binding site. Such compounds would not competitively found to inhibit both DNA and RNA unwinding by WNV Understanding Helicases as a Means of Virus Control Current Pharmaceutical Design, 2006, Vol. 12, No. 11 1331 helicase. Again, as seen with HMC-HO4 and TBBT, inhibi- Although directly using RNA as an antiviral drug will be tion of the unwinding activity by RENs does not correlate challenging because of its cellular instability, the information with a reduction in ATP hydrolysis (unless REN triphos- derived from aptamer studies could be used to make more phates are used in the assays). To see any inhibition of ATP stable derivatives. For example, the same aptamer sequences hydrolysis ATP must be used at a concentration well below could be used to generate peptide nucleic acids (PNA), its apparent Km, and under such conditions only two RENs which are modified nucleic acids composed of natural purine show any inhibition. At high concentrations (> 0.5 mM), all and pyrimidine bases linked together via an uncharged 2- of the RENs enhanced ATP hydrolysis. None of the RENs aminoethyl-glycine backbone. Both HCV helicase [197] and were reportedly tested in cell-based assays for a true antiviral HSV UL9 helicase [198] have been shown to be inhibited by effect although antiviral effects for RENs have been reported PNA. PNA most likely inhibits helicase activity by forming for other viruses [189]. a duplex that is more stable than those seen with natural DNA or RNA: The PNA-DNA substrates are unwound more Another method to inhibit a helicase is to directly target slowly (up to 80-fold) when compared to DNA duplexes the nucleic acid-binding site rather than the NTP-binding containing the same sequence, and only minimal PNA-RNA site. The HCV NS3 helicase nucleic acid-binding site is lo- duplex is unwound by the HCV NS3 helicase [197]. cated in the cleft that separates domains 1 and 2 from domain 3 (Fig. (2)), and represents a target site for antiviral Delivery of RNA aptamers to infected cells also presents drugs that compete with HCV helicase for its viral RNA sub- a challenging problem. One logical way to introduce an ap- strate. tamer is to introduce a gene, encoding its transcription, into cells using an appropriate vector. Using such a gene therapy SELEX (systematic evolution of ligands by exponential approach, the aptamer would be expressed inside cells (for amplification) has been used to find RNA sequences that review see [199]). tightly bind HCV helicase. Such sequences, called “aptamers, ” are oligonucleotides that exhibit molecular recognition, Gene therapy can also be used to introduce protein mole- and could conceivably be directly used as antivirals, or could cules into cells. The most practical proteins to introduce are provide templates for the synthesis of antiviral drugs. In the antibodies, which could conceivably act inside a cell to in- SELEX procedure (see [190] for a review), an RNA library hibit any enzyme. Antibodies are the first line of defense that is screened for sequences that bind a macromolecule using a the body has against viruses, and have long been recognized binding assay. Only those sequences that bind tightly are as powerful antiviral agents. Natural biological diversity amplified to create a new library, and the selection process is allows for the selection of antibodies that tightly bind virtu- repeated with the new library. Nishikawa’s group first used ally any epitope of any enzyme. Once antibodies are isolated SELEX to uncover a series of aptamers with specific affinity with a high avidity for a desired antigen, a gene encoding to the NS3 protease region. The protease-binding aptamers that antibody can be cloned and introduced into infected all share the conserved sequence GA(A/U)UGGGAC [191]. cells. This process, called “cellular immunization, ” has been These aptamers bind to the NS3 protease with high affinity used successfully with HIV [200], and such an approach has (Kd ~10 nM) and are effective at inhibiting the HCV NS3 been used to target the HCV helicase [201-203]. protease [191, 192]. The conserved part of the RNA aptam- The first anti-HCV helicase antibodies designed for in- ers likely interacts with a cluster of basic amino acid (resi- tracellular immunization were single chain fragment (ScF) dues Arg130, Arg161 and Lys165), which is located outside antibodies. A ScF is composed of the immunoglobulin vari- of the serine protease catalytic pocket in the region that links able domains, of both the heavy and light chains, connected the protease to the helicase [193]. When a 14-mer uridine tail by a polypeptide linker, and can be constructed using PCR. is added to one of the most effective HCV protease-binding Several high affinity ScFs that specifically interact with RNA aptamers, the new, longer aptamer interacts with both HCV NS3 helicase have been cloned [201, 204] and shown the protease and helicase domains of the full-length NS3 to inhibit HCV helicase catalyzed DNA unwinding [202, protein [194]. This dual-function RNA aptamer binds to the 204]. One ScF has been expressed in HCV infected hepato- HCV NS3 helicase domain with high affinity (Kd ~4 nM) cytes and shown to reduce HCV RNA synthesis [204]. This [194] and also inhibits the NS3 helicase activity with an particular ScF consists of the variable regions of the human EC50 of ~500 nM. monoclonal antibody CM3.B6, which recognizes an epitope An aptamer that bind HCV helicase have also been di- that spans conserved SF2 motifs IV and V of the NS3 heli- rectly selected using the helicase domain of the NS3 protein case [205]. Immunoblots show that an intracellularly ex- as the bait in the SELEX procedure [195]. This aptamer pressed ScF derived from CM3.B6 interacts with the HCV (called SE RNA) folds to form four stem loops with GC NS3 protein, and HCV RNA synthesis within primary he- pairs that are similar to the stem loop located at the 3'- patocytes infected with HCV is reduced by 10-fold when the terminal of the negative strand HCV RNA. Since previous cells contain an adenoviral vector carrying the ScF gene studies have shown that HCV helicase specifically binds to [204]. this region of the HCV genome [196], the SE aptamer might Another approach to achieve intracellular immunization compete for this binding site. SE RNA binds the HCV heli- uses an antibody fragment (Fab). A Fab is larger and usually case with a high affinity (Kd ~990 pM), efficiently competes more stable than a ScF because in addition to regions present with poly(U), stimulates ATP hydrolysis, but severely inhib- in ScFs, it contains the complete light chain and the first its RNA unwinding (IC50 ~12.5 nM). Most importantly, hu- constant domain of the heavy chain. Prabhu et al. has iso- man liver cells (Huh 7) infected with HCV replicons show lated a human Fab, termed HFab-aNS3, with effective HCV less HCV RNA synthesis in the presence of SE RNA. antiviral activity [203]. Derived from a HCV-infected pa- 1332 Current Pharmaceutical Design, 2006, Vol. 12, No. 11 Frick and Lam tient, HFab-aNS3 binds HCV helicase and recognizes an that DnaB-like and SF3 helicases encircle DNA as rings. epitope that spans motifs I to V of the protein. HCV NS3 ATP binding to ring helicases causes the movement of loops helicase pre-incubated with the HFab-aNS3 fails to unwind implicated in DNA binding that could push the protein either DNA duplexes. Intracellular expression of HFab-aNS3 by moving in a sequential or concerted manner. The active within replicon-transfected Huh 7 cells suppresses NS3 pro- sites appear to work in concert in SF3 helicases, and sequen- tein expression and significantly inhibits positive strand tially in DnaB-like helicases. There is little evidence that HCV RNA synthesis of both subgenomic and full-length SF1 and SF2 helicases function as rings. Rather, most evi- HCV replicons [203]. dence points to an inchworm mechanism in which the functional unit is a monomer. In non-ring helicases, ATP binds at SF3 Helicase Inhibitors an interface between two RecA-like domains in a cleft that closes upon ATP binding. The main difference between SF1 Among the viral helicases from SF3, the E1 helicase and SF2 proteins appears to be that the SF1 helicases move from HPV is currently the target of antiviral molecules for by interacting with the nucleic acid bases whereas SF2 pro- the treatment of HPV infection. Through screening a com- teins interact primarily with the phosphodiester backbone. pound library, two classes of small molecules have been identified and are being developed at Boehringer Ingelheim. The first group involves agents that disrupt the interaction A Br between the HPV E1 helicase and its co-activator HPV E2 O Br protein. Because HPV DNA replication is dependent on proper assembly of E1 proteins at the HPV origin of replica- O tion, which requires initial complex formation between E1 and E2, compounds that interrupt this interaction can poten- O N tially serve as promising antiviral drugs. The second group - O consists of inhibitors that specifically target HPV E1 helicase O O DNA unwinding. N Using a high throughput assay that analyzes the binding O between E1 and E2 proteins, a class of small molecules E1-E2 Inhibitor 3 (HPV E1-E2 inhibitors 1 to 3) that disrupt the E1-E2 complex has been identified [206]. The most potent, inhibitor 3, B is shown in Fig. (8). These inhibitors do not significantly affect E1 helicase activity, nor do they block DNA binding H to either E1 or E2. These molecules only target the E2 pro- N HOOC tein [206]. These molecules bind to the region of E2 that S interacts with E1 and prevent it from associating with the E1 O helicase. In Chinese hamster ovary cells transfected with O O plasmids expressing HPV E1 and E2 together with a plasmid containing the HPV DNA origin of replication, HPV DNA Compound 6g replication is inhibited in a dose-dependent manner in the Fig. (8). SF3 helicase inhibitors. A. HPV E1-E2 complex inhibitor presence of inhibitor 3 (EC50 ~1 mM). Currently, this class of inhibitors is effective against low risk HPV (types 6 and 11), [206]. B. HPV E1 inhibitor [207]. but unfortunately not the high risk types. Potent inhibitors have been isolated for members of each Inhibitors that are specific for the E1 helicase are also superfamily of viral helicases. The best characterized are being developed [207]. High throughput screening of a li- those that inhibit the SF1 HSV helicase-primase complex. brary of compounds has identified a lead molecule bearing a The anti-HSV compounds function by binding near con- biphenyl-4-sulfonylacetic acid moiety, which specifically served SF1 motif IV, preventing dissociation of the protein inhibits the ATPase activity of the HPV-6 E1 helicase. from viral DNA, slowing helicase movement, and conse- Among the derivatives of this parent molecule, compound 6g quently the activity of the associated primase necessary for demonstrates potent inhibition against the ATPase activity of viral genome synthesis. The inhibitors of HCV helicase and the E1 protein (IC50 ~4.3 nM) (Fig. (8b)). However, it has related SF2 proteins that have been isolated are small mole- limited in vivo application because of poor cell permeability cules, RNA aptamers, or antibodies. Developing the HCV and is relatively unstable. These compounds are currently inhibitors will likely be slow but several helicase inhibitors being improved in their chemical structures so they can be have shown clear efficacy in treating West Nile virus infec- suitable for further cell-based evaluation [207]. tion. SF3 helicase inhibitors targeting HPV also have been isolated and work by either directly reducing HPV helicase- CONCLUSION catalyzed ATP hydrolysis or its interaction with a protein All viral helicases can be classified as DnaB-like, SF1, that loads the helicase on replication origins. SF2, or SF3 helicases. All helicases share a common mecha- Clearly, helicase inhibitors are in an early stage of devel- nism of ATP hydrolysis to produce conformational changes opment and much more work will need to be done until these allowing the protein to move on DNA (or RNA). Although compounds will be regularly used as antiviral agents. Nev- precisely how chemical energy is converted to mechanical ertheless, the critical first steps have been taken, and if new movements is unknown at this time, there is solid evidence insights into helicase mechanisms continue at the current Understanding Helicases as a Means of Virus Control Current Pharmaceutical Design, 2006, Vol. 12, No. 11 1333 pace, helicase inhibitors could one day be prescribed as TCBT = Tetrachlorobenzotriazole commonly as protease and polymerase inhibitors. ss = Single-stranded ACKNOWLEDGEMENT VV = Vaccinia virus This work was supported by National Institutes of Health YFV = Yellow fever virus grant AI052395. WNV = West Nile virus

ABBREVIATIONS REFERENCES (+)ssRNA = Positive sense single-stranded RNA References 237-239 are related articles recently published in (-)ssRNA = Negative sense single-stranded RNA Current Pharmaceutical Design. AAA+ = ATPases associated with various cellular [1] Crute JJ, Grygon CA, Hargrave KD, Simoneau B, Faucher AM, Bolger G, et al. Herpes simplex virus helicase-primase inhibitors activities are active in animal models of human disease. Nat Med 2002; 8: AAV = Adeno-associated virus 386-91. [2] Kleymann G, Fischer R, Betz UA, Hendrix M, Bender W, ADPNP = 5'-adenylyl-b-g -imidodiphosphate Schneider U, et al. New helicase-primase inhibitors as drug candidates for the treatment of herpes simplex disease. Nat Med 2002; 8: BVDV = Bovine viral diarrhea virus 392-8. [3] Krell H, Durwald H, Hoffmann-Berling H. A DNA-unwinding CoV = Coronavirus enzyme induced in bacteriophage-T4-infected Escherichia coli cells. Eur J Biochem 1979; 93: 387-95. ds = Double-stranded [4] Tjian R, Robbins A. Enzymatic activities associated with a purified DFV = Dengue fever virus simian virus 40 T antigen-related protein. Proc Natl Acad Sci USA 1979; 76: 610-4. DRB = Dichloro(ribofuranosyl)benzimidazole [5] Matson SW, Tabor S, Richardson CC. The gene 4 protein of bacteriophage T7. Characterization of helicase activity. J Biol Chem DRBT = Dichloro(ribofuranosyl)benzotriazole 1983; 258: 14017-24. [6] Lain S, Riechmann JL, Garcia JA. RNA helicase: a novel activity IRES = Internal ribosome entry site associated with a protein encoded by a positive strand RNA virus. HCV = Hepatitis C virus Nucleic Acids Res 1990; 18: 7003-6. [7] Yedavalli VS, Neuveut C, Chi YH, Kleiman L, Jeang KT. Re- HHV = Human herpesviruses quirement of DDX3 DEAD box RNA helicase for HIV-1 Rev-RRE export function. Cell 2004; 119: 381-92. HIV = Human immunodeficiency virus [8] Frick DN. Helicases as antiviral drug targets. Drug News Perspect 2003; 16: 355-62. HSV = Herpes simplex virus [9] Delagoutte E, von Hippel PH. Helicase mechanisms and the coupling of helicases within macromolecular machines. Part I: Struc- HPV = Human papillomavirus tures and properties of isolated helicases. Q Rev Biophys 2002; 35: JEV = Japanese encephalitis virus 431-78. [10] Delagoutte E, von Hippel PH. Helicase mechanisms and the cou- NMR = Nuclear magnetic resonance pling of helicases within macromolecular machines. Part II: Inte- gration of helicases into cellular processes. Q Rev Biophys 2003; NPH = NTP phosphohydrolase 36: 1-69. [11] Gorbalenya AE, Koonin EV. Helicases: amino acid sequence com- NS = Nonstructural parisons and structure-function relationships. Curr Opin Struct Biol nsp = Nonstructural protein 1993; 3: 419-429. [12] Gorbalenya AE, Koonin EV, Donchenko AP, Blinov VM. Two NTP = Nucleoside triphosphate related superfamilies of putative helicases involved in replication, recombination, repair and expression of DNA and RNA genomes. ORF = Open reading frame Nucleic Acids Res 1989; 17: 4713-30. [13] Gorbalenya AE, Koonin EV, Wolf YI. A new superfamily of puta- PNA = Peptide nucleic acids tive NTP-binding domains encoded by genomes of small DNA and RNA viruses. FEBS Lett 1990; 262: 145-8. pp = Polyprotein [14] Ilyina TV, Gorbalenya AE, Koonin EV. Organization and evolu- RTPase = RNA 5¢-triphosphatase activity tion of bacterial and bacteriophage primase-helicase systems. J Mol Evol 1992; 34: 351-7. RENs = Ring-expanded “fat” nucleosides [15] Richardson JP. Loading Rho to terminate transcription. Cell 2003; 114: 157-9. SARS = Severe acute respiratory syndrome [16] Hall MC, Matson SW. Helicase motifs: the engine that powers DNA unwinding. Mol Microbiol 1999; 34: 867-77. tag = Small T (tumor) antigen [17] Patel SS, Picha KM. Structure and function of hexameric helicases. Tag = Large T (tumor) antigen Annu Rev Biochem 2000; 69: 651-97. [18] Frick DN, Richardson CC. DNA primases. Annu Rev Biochem SELEX = Systematic evolution of ligands by exponential 2001; 70: 39-80. amplification [19] Miller ES, Kutter E, Mosig G, Arisaka F, Kunisawa T, Ruger W. Bacteriophage T4 genome. Microbiol Mol Biol Rev 2003; 67: 86- SF = Superfamily 156, table of contents. [20] Dong F, Gogol EP, von Hippel PH. The phage T4-coded DNA SV40 = Simian virus 40 replication helicase (gp41) forms a hexamer upon activation by nucleoside triphosphate. J Biol Chem 1995; 270: 7462-73. TBBT = Tetrabromobenzotriazole 1334 Current Pharmaceutical Design, 2006, Vol. 12, No. 11 Frick and Lam

[21] Morrical SW, Hempstead K, Morrical MD. The gene 59 protein of [43] Makhov AM, Lee SS, Lehman IR, Griffith JD. Origin-specific bacteriophage T4 modulates the intrinsic and single-stranded DNA- unwinding of herpes simplex virus 1 DNA by the viral UL9 and stimulated ATPase activities of gene 41 protein, the T4 replicative ICP8 proteins: Visualization of a specific preunwinding complex. DNA helicase. J Biol Chem 1994; 269: 33069-81. Proc Natl Acad Sci USA 2003; 24: 24. [22] Dong F, von Hippel PH. The ATP-activated hexameric helicase of [44] Nanduri B, Byrd AK, Eoff RL, Tackett AJ, Raney KD. Pre-steady- bacteriophage T4 (gp41) forms a stable primosome with a single state DNA unwinding by bacteriophage T4 Dda helicase reveals a subunit of T4-coded primase (gp61). J Biol Chem 1996; 271: monomeric molecular motor. Proc Natl Acad Sci USA 2002; 99: 19625-31. 14722-7. [23] Jongeneel CV, Formosa T, Alberts BM. Purification and charac- [45] Morris PD, Raney KD. DNA helicases displace streptavidin from terization of the bacteriophage T4 dda protein. A DNA helicase biotin-labeled oligonucleotides. Biochemistry 1999; 38: 5164-71. that associates with the viral helix-destabilizing protein. J Biol [46] Bedinger P, Hochstrasser M, Jongeneel CV, Alberts BM. Proper- Chem 1984; 259: 12925-32. ties of the T4 bacteriophage DNA replication apparatus: the T4 dda [24] Sickmier EA, Kreuzer KN, White SW. The crystal structure of the DNA helicase is required to pass a bound RNA polymerase mole- UvsW helicase from bacteriophage T4. Structure (Camb) 2004; 12: cule. Cell 1983; 34: 115-23. 583-92. [47] Byrd AK, Raney KD. Protein displacement by an assembly of [25] Frick DN, Baradaran K, Richardson CC. An N-terminal fragment helicase molecules aligned along single-stranded DNA. Nat Struct of the gene 4 helicase/primase of bacteriophage T7 retains primase Mol Biol 2004; 11: 531-8. activity in the absence of helicase activity. Proc Natl Acad Sci USA [48] Ziebuhr J. Molecular biology of severe acute respiratory syndrome 1998; 95: 7957-62. coronavirus. Curr Opin Microbiol 2004; 7: 412-9. [26] Guo S, Tabor S, Richardson CC. The linker region between the [49] Marra MA, Jones SJ, Astell CR, Holt RA, Brooks-Wilson A, But- helicase and primase domains of the bacteriophage T7 gene 4 pro- terfield YS, et al. The Genome sequence of the SARS-associated tein is critical for hexamer formation. J Biol Chem 1999; 274: coronavirus. Science 2003; 300: 1399-404. 30303-9. [50] Rota PA, Oberste MS, Monroe SS, Nix WA, Campagnoli R, [27] Kusakabe T, Hine AV, Hyberts SG, Richardson CC. The Cys4 zinc Icenogle JP, et al. Characterization of a novel coronavirus associ- finger of bacteriophage T7 primase in sequence-specific single- ated with severe acute respiratory syndrome. Science 2003; 300: stranded DNA recognition. Proc Natl Acad Sci USA 1999; 96: 1394-9. 4295-300. [51] Thiel V, Ivanov KA, Putics A, Hertzig T, Schelle B, Bayer S, et al. [28] Frick DN, Richardson CC. Interaction of bacteriophage T7 gene 4 Mechanisms and enzymes involved in SARS coronavirus genome primase with its template recognition site. J Biol Chem 1999; 274: expression. J Gen Virol 2003; 84: 2305-15. 35889-98. [52] Ivanov KA, Ziebuhr J. Human coronavirus 229E nonstructural [29] Bernstein JA, Richardson CC. A 7-kDa region of the bacteriophage protein 13: characterization of duplex-unwinding, nucleoside T7 gene 4 protein is required for primase but not for helicase activ- triphosphatase, and RNA 5'-triphosphatase activities. J Virol 2004; ity. Proc Natl Acad Sci USA 1988; 85: 396-400. 78: 7833-8. [30] Boehmer PE, Lehman IR. Herpes simplex virus DNA replication. [53] Tanner JA, Watt RM, Chai YB, Lu LY, Lin MC, Peiris JS, et al. Annu Rev Biochem 1997; 66: 347-84. The severe acute respiratory syndrome (SARS) coronavirus [31] Marintcheva B, Weller SK. A tale of two HSV-1 helicases: roles of NTPase/helicase belongs to a distinct class of 5' to 3' viral heli- phage and animal virus helicases in DNA replication and recombi- cases. J Biol Chem 2003; 278: 39578-82. nation. Prog Nucleic Acid Res Mol Biol 2001; 70: 77-118. [54] Ivanov KA, Thiel V, Dobbe JC, van der Meer Y, Snijder EJ, Zie- [32] Crute JJ, Mocarski ES, Lehman IR. A DNA helicase induced by buhr J. Multiple enzymatic activities associated with severe acute herpes simplex virus type 1. Nucleic Acids Res 1988; 16: 6585-96. respiratory syndrome coronavirus helicase. J Virol 2004; 78: 5619- [33] Dodson MS, Crute JJ, Bruckner RC, Lehman IR. Overexpression 32. and assembly of the herpes simplex virus type 1 helicase-primase [55] Li YI, Shih TW, Hsu YH, Han YT, Huang YL, Meng M. The heli- in insect cells. J Biol Chem 1989; 264: 20835-8. case-like domain of plant potexvirus replicase participates in for- [34] Crute JJ, Lehman IR. Herpes simplex virus-1 helicase-primase. mation of RNA 5' cap structure by exhibiting RNA 5'- Physical and catalytic properties. J Biol Chem 1991; 266: 4484-8. triphosphatase activity. J Virol 2001; 75: 12114-20. [35] Zhu LA, Weller SK. The six conserved helicase motifs of the UL5 [56] Seybert A, Hegyi A, Siddell SG, Ziebuhr J. The human coronavirus gene product, a component of the herpes simplex virus type 1 heli- 229E superfamily 1 helicase has RNA and DNA duplex-unwinding case-primase, are essential for its function. J Virol 1992; 66: 469- activities with 5'-to-3' polarity. RNA 2000; 6: 1056-68. 79. [57] Seybert A, van Dinten LC, Snijder EJ, Ziebuhr J. Biochemical [36] Graves-Woodward KL, Gottlieb J, Challberg MD, Weller SK. characterization of the equine arteritis virus helicase suggests a Biochemical analyses of mutations in the HSV-1 helicase-primase close functional relationship between arterivirus and coronavirus that alter ATP hydrolysis, DNA unwinding, and coupling between helicases. J Virol 2000; 74: 9586-93. hydrolysis and unwinding. J Biol Chem 1997; 272: 4623-30. [58] van Dinten LC, van Tol H, Gorbalenya AE, Snijder EJ. The pre- [37] Dodson MS, Lehman IR. Association of DNA helicase and primase dicted metal-binding region of the arterivirus helicase protein is in- activities with a subassembly of the herpes simplex virus 1 heli- volved in subgenomic mRNA synthesis, genome replication, and case-primase composed of the UL5 and UL52 gene products. Proc virion biogenesis. J Virol 2000; 74: 5213-23. Natl Acad Sci USA 1991; 88: 1105-9. [59] Seybert A, Posthuma CC, van Dinten LC, Snijder EJ, Gorbalenya [38] Biswas N, Weller SK. A mutation in the C-terminal putative Zn2+ AE, Ziebuhr J. A complex zinc finger controls the enzymatic ac- finger motif of UL52 severely affects the biochemical activities of tivities of nidovirus helicases. J Virol 2005; 79: 696-704. the HSV-1 helicase- primase subcomplex. J Biol Chem 1999; 274: [60] Bork P, Koonin EV. An expanding family of helicases within the 8068-76. 'DEAD/H' superfamily. Nucleic Acids Res 1993; 21: 751-2. [39] Biswas N, Weller SK. The UL5 and UL52 subunits of the herpes [61] Tanner NK, Linder P. DExD/H box RNA helicases: from generic simplex virus type 1 helicase-primase subcomplex exhibit a com- motors to specific dissociation functions. Mol Cell 2001; 8: 251-62. plex interdependence for DNA binding. J Biol Chem 2001; 276: [62] Silverman E, Edwalds-Gilbert G, Lin RJ. DExD/H-box proteins 17610-9. and their partners: helping RNA helicases unwind. Gene 2003; 312: [40] Fierer DS, Challberg MD. Purification and characterization of 1-16. UL9, the herpes simplex virus type 1 origin-binding protein. J Virol [63] Suzich JA, Tamura JK, Palmer-Hill F, Warrener P, Grakoui A, 1992; 66: 3986-95. Rice CM, et al. Hepatitis C virus NS3 protein polynucleotide- [41] Boehmer PE, Dodson MS, Lehman IR. The herpes simplex virus stimulated nucleoside triphosphatase and comparison with the re- type-1 origin binding protein. DNA helicase activity. J Biol Chem lated pestivirus and flavivirus enzymes. J Virol 1993; 67: 6152-8. 1993; 268: 1220-5. [64] Kanai A, Tanabe K, Kohara M. Poly(U) binding activity of hepati- [42] He X, Lehman IR. Unwinding of a herpes simplex virus type 1 tis C virus NS3 protein, a putative RNA helicase. FEBS Lett 1995; origin of replication (Ori(S)) by a complex of the viral origin 376: 221-4. binding protein and the single-stranded DNA binding protein. J Vi- [65] Kim DW, Gwack Y, Han JH, Choe J. C-terminal domain of the rol 2000; 74: 5726-8. hepatitis C virus NS3 protein contains an RNA helicase activity. Biochem Biophys Res Commun 1995; 215: 160-6. Understanding Helicases as a Means of Virus Control Current Pharmaceutical Design, 2006, Vol. 12, No. 11 1335

[66] Gwack Y, Kim DW, Han JH, Choe J. Characterization of RNA [90] Valle M, Gruss C, Halmer L, Carazo JM, Donate LE. Large T- binding activity and RNA helicase activity of the hepatitis C virus antigen double hexamers imaged at the simian virus 40 origin of NS3 protein. Biochem Biophys Res Commun 1996; 225: 654-9. replication. Mol Cell Biol 2000; 20: 34-41. [67] Tai CL, Chi WK, Chen DS, Hwang LH. The helicase activity asso- [91] Alexandrov AI, Botchan MR, Cozzarelli NR. Characterization of ciated with hepatitis C virus nonstructural protein 3 (NS3). J Virol simian virus 40 T-antigen double hexamers bound to a replication 1996; 70: 8477-84. fork. The active form of the helicase. J Biol Chem 2002; 277: [68] Gwack Y, Kim DW, Han JH, Choe J. DNA helicase activity of the 44886-97. hepatitis C virus nonstructural protein 3. Eur J Biochem 1997; 250: [92] Gomez-Lorenzo MG, Valle M, Frank J, Gruss C, Sorzano CO, 47-54. Chen XS, et al. Large T antigen on the simian virus 40 origin of [69] Warrener P, Tamura JK, Collett MS. RNA-stimulated NTPase replication: a 3D snapshot prior to DNA replication. Embo J 2003; activity associated with yellow fever virus NS3 protein expressed 22: 6205-13. in bacteria. J Virol 1993; 67: 989-96. [93] Morris PD, Byrd AK, Tackett AJ, Cameron CE, Tanega P, Ott R, et [70] Li H, Clum S, You S, Ebner KE, Padmanabhan R. The serine pro- al. Hepatitis C virus NS3 and simian virus 40 T antigen helicases tease and RNA-stimulated nucleoside triphosphatase and RNA displace streptavidin from 5'-biotinylated oligonucleotides but not helicase functional domains of dengue virus type 2 NS3 converge from 3'- biotinylated oligonucleotides: evidence for directional bias within a region of 20 amino acids. J Virol 1999; 73: 3108-16. in translocation on single-stranded DNA. Biochemistry 2002; 41: [71] Benarroch D, Selisko B, Locatelli GA, Maga G, Romette JL, Ca- 2372-8. nard B. The RNA helicase, nucleotide 5'-triphosphatase, and RNA [94] Fanning E, Knippers R. Structure and function of simian virus 40 5'-triphosphatase activities of Dengue virus protein NS3 are Mg2+- large tumor antigen. Annu Rev Biochem 1992; 61: 55-85. dependent and require a functional Walker B motif in the helicase [95] Walboomers JM, Jacobs MV, Manos MM, Bosch FX, Kummer JA, catalytic core. Virology 2004; 328: 208-18. Shah KV, et al. Human papillomavirus is a necessary cause of in- [72] Borowski P, Niebuhr A, Mueller O, Bretner M, Felczak K, vasive cervical cancer worldwide. J Pathol 1999; 189: 12-9. Kulikowski T, et al. Purification and characterization of West Nile [96] Chen EY, Howley PM, Levinson AD, Seeburg PH. The primary virus nucleoside triphosphatase (NTPase)/helicase: evidence for structure and genetic organization of the bovine papillomavirus dissociation of the NTPase and helicase activities of the enzyme. J type 1 genome. Nature 1982; 299: 529-34. Virol 2001; 75: 3220-9. [97] Danos O, Katinka M, Yaniv M. Human papillomavirus 1a com- [73] Utama A, Shimizu H, Morikawa S, Hasebe F, Morita K, Igarashi plete DNA sequence: a novel type of genome organization among A, et al. Identification and characterization of the RNA helicase papovaviridae. Embo J 1982; 1: 231-6. activity of Japanese encephalitis virus NS3 protein. FEBS Lett [98] Wilson VG, West M, Woytek K, Rangasamy D. Papillomavirus E1 2000; 465: 74-8. proteins: form, function, and features. Virus Genes 2002; 24: 275- [74] Lam AMI, Keeney D, Eckert PQ, Frick DN. Hepatitis C virus NS3 90. ATPases/helicases from different genotypes exhibit variations in [99] Frattini MG, Laimins LA. Binding of the human papillomavirus E1 enzymatic properties. J Virol 2003; 77: 3950-3961. origin-recognition protein is regulated through complex formation [75] Gallivan JP, McGarvey MJ. The importance of the Q motif in the with the E2 enhancer-binding protein. Proc Natl Acad Sci USA ATPase activity of a viral helicase. FEBS Lett 2003; 554: 485-8. 1994; 91: 12398-402. [76] Tanner NK, Cordin O, Banroques J, Doere M, Linder P. The Q [100] Abbate EA, Berger JM, Botchan MR. The X-ray structure of the motif: a newly identified motif in DEAD box helicases may regu- papillomavirus helicase in complex with its molecular matchmaker late ATP binding and hydrolysis. Mol Cell 2003; 11: 127-38. E2. Genes Dev 2004; 18: 1981-96. [77] Bartelma G, Padmanabhan R. Expression, purification, and char- [101] White PW, Pelletier A, Brault K, Titolo S, Welchner E, Thauvette acterization of the RNA 5'-triphosphatase activity of dengue virus L, et al. Characterization of recombinant HPV6 and 11 E1 heli- type 2 nonstructural protein 3. Virology 2002; 299: 122-32. cases: effect of ATP on the interaction of E1 with E2 and mapping [78] Wengler G. The NS 3 nonstructural protein of flaviviruses contains of a minimal helicase domain. J Biol Chem 2001; 276: 22426-38. an RNA triphosphatase activity. Virology 1993; 197: 265-73. [102] Fouts ET, Yu X, Egelman EH, Botchan MR. Biochemical and [79] Egloff MP, Benarroch D, Selisko B, Romette JL, Canard B. An electron microscopic image analysis of the hexameric E1 helicase. RNA cap (nucleoside-2'-O-)-methyltransferase in the flavivirus J Biol Chem 1999; 274: 4447-58. RNA polymerase NS5: crystal structure and functional characteri- [103] McCarty DM, Young Jr SM, Samulski RJ. Integration of Adeno- zation. Embo J 2002; 21: 2757-68. Associated Virus (AAV) and Recombinant AAV Vectors. Annu [80] Koonin EV, Senkevich TG. Vaccinia virus encodes four putative Rev Genet 2004. DNA and/or RNA helicases distantly related to each other. J Gen [104] Smith RH, Spano AJ, Kotin RM. The Rep78 gene product of Virol 1992; 73 ( Pt 4): 989-93. adeno-associated virus (AAV) self-associates to form a hexameric [81] Deng L, Shuman S. Vaccinia NPH-I, a DExH-box ATPase, is the complex in the presence of AAV ori sequences. J Virol 1997; 71: energy coupling factor for mRNA transcription termination. Genes 4461-71. Dev 1998; 12: 538-46. [105] Smith RH, Kotin RM. The Rep52 gene product of adeno-associated [82] Shuman S. Vaccinia virus RNA helicase: an essential enzyme virus is a DNA helicase with 3'-to-5' polarity. J Virol 1998; 72: related to the DE-H family of RNA-dependent NTPases. Proc Natl 4874-81. Acad Sci USA 1992; 89: 10935-9. [106] Rodriguez PL, Carrasco L. Poliovirus protein 2C has ATPase and [83] Simpson DA, Condit RC. Vaccinia virus gene A18R encodes an GTPase activities. J Biol Chem 1993; 268: 8105-10. essential DNA helicase. J Virol 1995; 69: 6131-9. [107] Mirzayan C, Wimmer E. Biochemical studies on poliovirus poly- [84] Li J, Broyles SS. The DNA binding domain of the vaccinia virus peptide 2C: evidence for ATPase activity. Virology 1994; 199: early transcription factor small subunit is an extended helicase-like 176-87. motif. Nucleic Acids Res 1995; 23: 1590-6. [108] Klein M, Eggers HJ, Nelsen-Salz B. Echovirus 9 strain barty non- [85] Jankowsky E, Gross CH, Shuman S, Pyle AM. The DExH protein structural protein 2C has NTPase activity. Virus Res 1999; 65: 155- NPH-II is a processive and directional motor for unwinding RNA. 60. Nature 2000; 403: 447-51. [109] Klein M, Eggers HJ, Nelsen-Salz B. Echovirus-9 protein 2C binds [86] Jankowsky E, Gross CH, Shuman S, Pyle AM. Active Disruption single-stranded RNA unspecifically. J Gen Virol 2000; 81: 2481-4. of an RNA-Protein Interaction by a DExH/D RNA Helicase. Sci- [110] Pfister T, Wimmer E. Polypeptide p41 of a Norwalk-like virus is a ence 2001; 291: 121-125. nucleic acid-independent nucleoside triphosphatase. J Virol 2001; [87] Kawaoka J, Jankowsky E, Pyle AM. Backbone tracking by the SF2 75: 1611-9. helicase NPH-II. Nat Struct Mol Biol 2004; 11: 526-30. [111] Yao N, Hesson T, Cable M, Hong Z, Kwong AD, Le HV, et al. [88] Shitrit D, Lev N, Bar-Gil-Shitrit A, Kramer MR. Progressive multi- Structure of the hepatitis C virus RNA helicase domain. Nat Struct focal leukoencephalopathy in transplant recipients. Transpl Int Biol 1997; 4: 463-7. 2004. [112] Cho HS, Ha NC, Kang LW, Chung KM, Back SH, Jang SK, et al. [89] Berger JR, Pall L, Lanska D, Whiteman M. Progressive multifocal Crystal structure of RNA helicase from genotype 1b hepatitis C vi- leukoencephalopathy in patients with HIV infection. J Neurovirol rus. A feasible mechanism of unwinding duplex RNA. J Biol Chem 1998; 4: 59-68. 1998; 273: 15045-52. 1336 Current Pharmaceutical Design, 2006, Vol. 12, No. 11 Frick and Lam

[113] Kim JL, Morgenstern KA, Griffith JP, Dwyer MD, Thomson JA, [136] Lam AM, Keeney D, Frick DN. Two novel conserved motifs in the Murcko MA, et al. Hepatitis C virus NS3 RNA helicase domain hepatitis C virus NS3 protein critical for helicase action. J Biol with a bound oligonucleotide: the crystal structure provides insights Chem 2003; 278: 44514-24. into the mode of unwinding. Structure 1998; 6: 89-100. [137] Korolev S, Hsieh J, Gauss GH, Lohman TM, Waksman G. Major [114] Yao N, Reichert P, Taremi SS, Prosise WW, Weber PC. Molecular domain swiveling revealed by the crystal structures of complexes views of viral polyprotein processing revealed by the crystal struc- of E. coli Rep helicase bound to single-stranded DNA and ADP. ture of the hepatitis C virus bifunctional protease-helicase. Struc- Cell 1997; 90: 635-47. ture Fold Des 1999; 7: 1353-63. [138] Velankar SS, Soultanas P, Dillingham MS, Subramanya HS, Wig- [115] Liu D, Wang YS, Gesell JJ, Wyss DF. Solution structure and back- ley DB. Crystal structures of complexes of PcrA DNA helicase bone dynamics of an engineered arginine-rich subdomain 2 of the with a DNA substrate indicate an inchworm mechanism. Cell 1999; hepatitis C virus NS3 RNA helicase. J Mol Biol 2001; 314: 543-61. 97: 75-84. [116] Sawaya MR, Guo S, Tabor S, Richardson CC, Ellenberger T. [139] Subramanya HS, Bird LE, Brannigan JA, Wigley DB. Crystal Crystal structure of the helicase domain from the replicative heli- structure of a DExx box DNA helicase. Nature 1996; 384: 379-83. case- primase of bacteriophage T7. Cell 1999; 99: 167-77. [140] Soultanas P, Dillingham MS, Velankar SS, Wigley DB. DNA [117] Bird LE, Hakansson K, Pan H, Wigley DB. Characterization and binding mediates conformational changes and metal ion coordina- crystallization of the helicase domain of bacteriophage T7 gene 4 tion in the active site of PcrA helicase. J Mol Biol 1999; 290: 137- protein. Nucleic Acids Res 1997; 25: 2620-6. 48. [118] Singleton MR, Sawaya MR, Ellenberger T, Wigley DB. Crystal [141] Soultanas P, Dillingham MS, Wiley P, Webb MR, Wigley DB. structure of T7 gene 4 ring helicase indicates a mechanism for se- Uncoupling DNA translocation and helicase activity in PcrA: direct quential hydrolysis of nucleotides. Cell 2000; 101: 589-600. evidence for an active mechanism. Embo J 2000; 19: 3799-810. [119] Toth EA, Li Y, Sawaya MR, Cheng Y, Ellenberger T. The crystal [142] Dillingham MS, Soultanas P, Wiley P, Webb MR, Wigley DB. structure of the bifunctional primase-helicase of bacteriophage T7. Defining the roles of individual residues in the single-stranded Mol Cell 2003; 12: 1113-23. DNA binding site of PcrA helicase. Proc Natl Acad Sci USA 2001; [120] James JA, Escalante CR, Yoon-Robarts M, Edwards TA, Linden 98: 8381-7. RM, Aggarwal AK. Crystal structure of the SF3 helicase from [143] Soultanas P, Wigley DB. Unwinding the 'Gordian knot' of helicase adeno-associated virus type 2. Structure (Camb) 2003; 11: 1025-35. action. Trends Biochem Sci 2001; 26: 47-54. [121] James JA, Aggarwal AK, Linden RM, Escalante CR. Structure of [144] Dillingham MS, Wigley DB, Webb MR. Demonstration of unidi- adeno-associated virus type 2 Rep40-ADP complex: insight into rectional single-stranded DNA translocation by PcrA helicase: nucleotide recognition and catalysis by superfamily 3 helicases. measurement of step size and translocation speed. Biochemistry Proc Natl Acad Sci USA 2004; 101: 12455-60. 2000; 39: 205-12. [122] Li D, Zhao R, Lilyestrom W, Gai D, Zhang R, DeCaprio JA, et al. [145] Porter DJ, Short SA, Hanlon MH, Preugschat F, Wilson JE, Structure of the replicative helicase of the oncoprotein SV40 large Willard DH, Jr., et al. Product release is the major contributor to tumour antigen. Nature 2003; 423: 512-8. kcat for the hepatitis C virus helicase-catalyzed strand separation of [123] Gai D, Zhao R, Li D, Finkielstein CV, Chen XS. Mechanisms of short duplex DNA. J Biol Chem 1998; 273: 18906-14. conformational change for a replicative hexameric helicase of [146] Levin MK, Gurjar MM, Patel SS. ATP binding modulates the SV40 large tumor antigen. Cell 2004; 119: 47-60. nucleic acid affinity of hepatitis C virus helicase. J Biol Chem [124] Story RM, Steitz TA. Structure of the recA protein-ADP complex. 2003; 278: 23311-23316. Nature 1992; 355: 374-6. [147] Levin MK, Wang YH, Patel SS. The functional interaction of the [125] Walker JE, Saraste M, Runswick MJ, Gay NJ. Distantly related hepatitis C virus helicase molecules is responsible for unwinding sequences in the alpha- and beta-subunits of ATP synthase, my- processivity. J Biol Chem 2004; 279: 26005-12. osin, kinases and other ATP-requiring enzymes and a common nu- [148] Serebrov V, Pyle AM. Periodic cycles of RNA unwinding and cleotide binding fold. Embo J 1982; 1: 945-51. pausing by hepatitis C virus NS3 helicase. Nature 2004; 430: 476- [126] Orelle C, Dalmas O, Gros P, Di Pietro A, Jault JM. The conserved 80. glutamate residue adjacent to the Walker-B motif is the catalytic [149] Lam AMI, Rypma RS, Frick DN. Enhanced nucleic acid binding to base for ATP hydrolysis in the ATP-binding cassette transporter ATP-bound hepatitis C virus NS3 helicase at low pH activates BmrA. J Biol Chem 2003; 278: 47002-8. RNA unwinding. Nucl. Acids. Res. 2004; 32: 4060-4070. [127] Goetzinger KR, Rao VB. Defining the ATPase center of bacterio- [150] Pang PS, Jankowsky E, Planet PJ, Pyle AM. The hepatitis C viral phage T4 DNA packaging machine: requirement for a catalytic NS3 protein is a processive DNA helicase with cofactor enhanced glutamate residue in the large terminase protein gp17. J Mol Biol RNA unwinding. Embo J 2002; 21: 1168-1176. 2003; 331: 139-54. [151] Frick DN, Rypma RS, Lam AM, Gu B. The nonstructural protein 3 [128] Ahmadian MR, Stege P, Scheffzek K, Wittinghofer A. Confirma- protease/helicase requires an intact protease domain to efficiently tion of the arginine-finger hypothesis for the GAP-stimulated GTP- unwind duplex RNA. J Biol Chem 2004; 279: 1269-1280. hydrolysis reaction of Ras. Nat Struct Biol 1997; 4: 686-9. [152] Flajolet M, Rotondo G, Daviet L, Bergametti F, Inchauspe G, [129] Nadanaciva S, Weber J, Wilke-Mounts S, Senior AE. Importance Tiollais P, et al. A genomic approach of the hepatitis C virus gen- of F1-ATPase residue alpha-Arg-376 for catalytic transition state erates a protein interaction map. Gene 2000; 242: 369-79. stabilization. Biochemistry 1999; 38: 15493-9. [153] Khu YL, Koh E, Lim SP, Tan YH, Brenner S, Lim SG, et al. Mu- [130] Crampton DJ, Guo S, Johnson DE, Richardson CC. The arginine tations that affect dimer formation and helicase activity of the finger of bacteriophage T7 gene 4 helicase: role in energy coupling. hepatitis C virus helicase. J Virol 2001; 75: 205-14. Proc Natl Acad Sci USA 2004; 101: 4373-8. [154] Tackett AJ, Chen Y, Cameron CE, Raney KD. Multiple full-length [131] Lin C, Kim JL. Structure-based mutagenesis study of hepatitis C NS3 molecules are required for optimal unwinding of oligonucleo- virus NS3 helicase. J Virol 1999; 73: 8798-807. tide DNA in vitro. J Biol Chem 2005. [132] Tai CL, Pan WC, Liaw SH, Yang UC, Hwang LH, Chen DS. [155] Wong I, Lohman TM. Allosteric effects of nucleotide cofactors on Structure-based mutational analysis of the hepatitis c virus ns3 Escherichia coli Rep helicase-DNA binding. Science 1992; 256: helicase. J Virol 2001; 75: 8289-97. 350-5. [133] Preugschat F, Danger DP, Carter LH, 3rd, Davis RG, Porter DJ. [156] Ali JA, Lohman TM. Kinetic measurement of the step size of DNA Kinetic analysis of the effects of mutagenesis of W501 and V432 unwinding by Escherichia coli UvrD helicase. Science 1997; 275: of the hepatitis C virus NS3 helicase domain on ATPase and 377-80. strand-separating activity. Biochemistry 2000; 39: 5174-83. [157] Maluf NK, Fischer CJ, Lohman TM. A Dimer of Escherichia coli [134] Paolini C, Lahm A, De Francesco R, Gallinari P. Mutational analy- UvrD is the Active Form of the Helicase In vitro. J Mol Biol 2003; sis of hepatitis C virus NS3-associated helicase. J Gen Virol 2000; 325: 913-35. 81 Pt 7: 1649-58. [158] Lohman TM, Bjornson KP. Mechanisms of helicase-catalyzed [135] Kim JW, Seo MY, Shelat A, Kim CS, Kwon TW, Lu HH, et al. DNA unwinding. Annu Rev Biochem 1996; 65: 169-214. Structurally Conserved Amino Acid W501 Is Required for RNA [159] Goregaoker SP, Culver JN. Oligomerization and activity of the Helicase Activity but Is Not Essential for DNA Helicase Activity helicase domain of the tobacco mosaic virus 126- and 183- of Hepatitis C Virus NS3 Protein. J Virol 2003; 77: 571-582. kilodalton replicase proteins. J Virol 2003; 77: 3549-56. Understanding Helicases as a Means of Virus Control Current Pharmaceutical Design, 2006, Vol. 12, No. 11 1337

[160] Hingorani MM, Patel SS. Cooperative interactions of nucleotide case activity with the Scintillation Proximity Assay system. J Bio- ligands are linked to oligomerization and DNA binding in bacterio- chem (Tokyo) 2004; 135: 245-52. phage T7 gene 4 helicases. Biochemistry 1996; 35: 2218-28. [181] Porter DJ. Inhibition of the hepatitis C virus helicase-associated [161] Hingorani MM, Washington MT, Moore KC, Patel SS. The ATPase activity by the combination of ADP, NaF, MgCl2, and dTTPase mechanism of T7 DNA helicase resembles the binding poly(rU). Two ADP binding sites on the enzyme-nucleic acid com- change mechanism of the F1-ATPase. Proc Natl Acad Sci USA plex. J Biol Chem 1998; 273: 7390-6. 1997; 94: 5012-7. [182] Xu YW, Morera S, Janin J, Cherfils J. AlF3 mimics the transition [162] Noji H, Yasuda R, Yoshida M, Kinosita K, Jr. Direct observation state of protein phosphorylation in the crystal structure of nucleo- of the rotation of F1-ATPase. Nature 1997; 386: 299-302. side diphosphate kinase and MgADP. Proc Natl Acad Sci USA [163] Itoh H, Takahashi A, Adachi K, Noji H, Yasuda R, Yoshida M, et 1997; 94: 3579-83. al. Mechanically driven ATP synthesis by F1-ATPase. Nature [183] Levin MK, Patel SS. Helicase from hepatitis C virus, energetics of 2004; 427: 465-8. DNA binding. J Biol Chem 2002; 277: 29377-85. [164] Egelman EH, Yu X, Wild R, Hingorani MM, Patel SS. Bacterio- [184] Porter DJ. A kinetic analysis of the oligonucleotide-modulated phage T7 helicase/primase proteins form rings around single- ATPase activity of the helicase domain of the NS3 protein from stranded DNA that suggest a general structure for hexameric heli- hepatitis C virus. The first cycle of interaction of ATP with the en- cases. Proc Natl Acad Sci USA 1995; 92: 3869-73. zyme is unique. J Biol Chem 1998; 273: 14247-53. [165] VanLoock MS, Alexandrov A, Yu X, Cozzarelli NR, Egelman EH. [185] Locatelli GA, Spadari S, Maga G. Hepatitis C virus NS3 AT- SV40 large T antigen hexamer structure: domain organization and Pase/helicase: an ATP switch regulates the cooperativity among the DNA-induced conformational changes. Curr Biol 2002; 12: 472-6. different substrate binding sites. Biochemistry 2002; 41: 10332-42. [166] Notarnicola SM, Park K, Griffith JD, Richardson CC. A domain of [186] Borowski P, Deinert J, Schalinski S, Bretner M, Ginalski K, the gene 4 helicase/primase of bacteriophage T7 required for the Kulikowski T, et al. Halogenated benzimidazoles and benzotria- formation of an active hexamer. J Biol Chem 1995; 270: 20215-24. zoles as inhibitors of the NTPase/helicase activities of hepatitis C [167] Washington MT, Rosenberg AH, Griffin K, Studier FW, Patel SS. and related viruses. Eur J Biochem 2003; 270: 1645-53. Biochemical analysis of mutant T7 primase/helicase proteins de- [187] Zhang N, Chen HM, Koch V, Schmitz H, Minczuk M, Stepien P, et fective in DNA binding, nucleotide hydrolysis, and the coupling of al. Potent inhibition of NTPase/helicase of the West Nile Virus by hydrolysis with DNA unwinding. J Biol Chem 1996; 271: 26825- ring-expanded ("fat") nucleoside analogues. J Med Chem 2003; 46: 34. 4776-89. [168] Monod J, Wyman J, Changeux JP. On the Nature of Allosteric [188] Zhang N, Chen HM, Koch V, Schmitz H, Liao CL, Bretner M, et Transitions: A Plausible Model. J Mol Biol 1965; 12: 88-118. al. Ring-expanded ("fat") nucleoside and nucleotide analogues ex- [169] Kao RY, Tsui WH, Lee TS, Tanner JA, Watt RM, Huang JD, et al. hibit potent in vitro activity against flaviviridae NTPases/ helicases, Identification of novel small-molecule inhibitors of severe acute including those of the West Nile virus, hepatitis C virus, and Japa- respiratory syndrome-associated coronavirus by chemical genetics. nese encephalitis virus. J Med Chem 2003; 46: 4149-64. Chem Biol 2004; 11: 1293-9. [189] Hosmane RS. Ring-expanded ("Fat") nucleosides as broad- [170] Duan J, Liuzzi M, Paris W, Liard F, Browne A, Dansereau N, et al. spectrum anticancer and antiviral agents. Curr Top Med Chem Oral bioavailability and in vivo efficacy of the helicase-primase in- 2002; 2: 1093-109. hibitor BILS 45 BS against acyclovir-resistant herpes simplex virus [190] Klug SJ, Famulok M. All you wanted to know about SELEX. Mol type 1. Antimicrob Agents Chemother 2003; 47: 1798-804. Biol Rep 1994; 20: 97-107. [171] Liuzzi M, Kibler P, Bousquet C, Harji F, Bolger G, Garneau M, et [191] Fukuda K, Vishnuvardhan D, Sekiya S, Hwang J, Kakiuchi N, al. Isolation and characterization of herpes simplex virus type 1 re- Taira K, et al. Isolation and characterization of RNA aptamers spe- sistant to aminothiazolylphenyl-based inhibitors of the viral heli- cific for the hepatitis C virus nonstructural protein 3 protease. Eur J case-primase. Antiviral Res 2004; 64: 161-70. Biochem 2000; 267: 3685-94. [172] Spector FC, Liang L, Giordano H, Sivaraja M, Peterson MG. Inhi- [192] Nishikawa F, Kakiuchi N, Funaji K, Fukuda K, Sekiya S, Nishi- bition of herpes simplex virus replication by a 2-amino thiazole via kawa S. Inhibition of HCV NS3 protease by RNA aptamers in interactions with the helicase component of the UL5-UL8-UL52 cells. Nucleic Acids Res 2003; 31: 1935-43. complex. J Virol 1998; 72: 6979-87. [193] Hwang J, Fauzi H, Fukuda K, Sekiya S, Kakiuchi N, Shimotohno [173] Betz UA, Fischer R, Kleymann G, Hendrix M, Rubsamen- K, et al. The RNA aptamer-binding site of hepatitis C virus NS3 Waigmann H. Potent in vivo antiviral activity of the herpes simplex protease. Biochem Biophys Res Commun 2000; 279: 557-62. virus primase- helicase inhibitor BAY 57-1293. Antimicrob Agents [194] Fukuda K, Umehara T, Sekiya S, Kunio K, Hasegawa T, Nishi- Chemother 2002; 46: 1766-72. kawa S. An RNA ligand inhibits hepatitis C virus NS3 protease and [174] Borowski P, Schalinski S, Schmitz H. Nucleotide triphospha- helicase activities. Biochem Biophys Res Commun 2004; 325: 670- tase/helicase of hepatitis C virus as a target for antiviral therapy. 675. Antiviral Res 2002; 55: 397-412. [195] Hwang B, Cho JS, Yeo HJ, Kim JH, Chung KM, Han K, et al. [175] Borowski P, Niebuhr A, Schmitz H, Hosmane RS, Bretner M, Isolation of specific and high-affinity RNA aptamers against NS3 Siwecka MA, et al. NTPase/helicase of Flaviviridae: inhibitors and helicase domain of hepatitis C virus. Rna 2004; 10: 1277-90. inhibition of the enzyme. Acta Biochim Pol 2002; 49: 597-614. [196] Banerjee R, Dasgupta A. Specific Interaction of Hepatitis C Virus [176] Borowski P, Lang M, Haag A, Schmitz H, Choe J, Chen HM, et al. Protease/Helicase NS3 with the 3'-Terminal Sequences of Viral Characterization of imidazo[4, 5-d]pyridazine nucleosides as Positive- and Negative-Strand RNA. J. Virol. 2001; 75: 1708-1721. modulators of unwinding reaction mediated by West Nile virus nu- [197] Tackett AJ, Wei L, Cameron CE, Raney KD. Unwinding of nucleic cleoside triphosphatase/helicase: evidence for activity on the level acids by HCV NS3 helicase is sensitive to the structure of the du- of substrate and/or enzyme. Antimicrob Agents Chemother 2002; plex. Nucleic Acids Res 2001; 29: 565-72. 46: 1231-9. [198] Bastide L, Boehmer PE, Villani G, Lebleu B. Inhibition of a DNA- [177] Preugschat F, Averett DR, Clarke BE, Porter DJT. A steady-state helicase by peptide nucleic acids. Nucleic Acids Res 1999; 27: and pre-steady-state kinetic analysis of the NTPase activity associ- 551-4. ated with the hepatitis C virus NS3 helicase domain. J Biol Chem [199] Joshi PJ, Fisher TS, Prasad VR. Anti-HIV inhibitors based on 1996; 271: 24449-57. nucleic acids: emergence of aptamers as potent antivirals. Curr [178] Wardell AD, Errington W, Ciaramella G, Merson J, McGarvey MJ. Drug Targets Infect Disord 2003; 3: 383-400. Characterization and mutational analysis of the helicase and [200] Goncalves J, Silva F, Freitas-Vieira A, Santa-Marta M, Malho R, NTPase activities of hepatitis C virus full-length NS3 protein. J Yang X, et al. Functional neutralization of HIV-1 Vif protein by Gen Virol 1999; 80: 701-9. intracellular immunization inhibits reverse transcription and viral [179] Locatelli GA, Gosselin G, Spadari S, Maga G. Hepatitis C virus replication. J Biol Chem 2002; 277: 32036-45. NS3 NTPase/helicase: different stereoselectivity in nucleoside [201] Tessmann K, Erhardt A, Haussinger D, Heintges T. Cloning and triphosphate utilisation suggests that NTPase and helicase activities molecular characterization of human high affinity antibody frag- are coupled by a nucleotide-dependent rate limiting step. J Mol ments against Hepatitis C virus NS3 helicase. J Virol Methods Biol 2001; 313: 683-94. 2002; 103: 75-88. [180] Kyono K, Miyashiro M, Taguchi I. Expression and purification of a hepatitis C virus NS3/4A complex, and characterization of its heli- 1338 Current Pharmaceutical Design, 2006, Vol. 12, No. 11 Frick and Lam

[202] Artsaenko O, Tessmann K, Sack M, Haussinger D, Heintges T. [221] Kainov DE, Pirttimaa M, Tuma R, Butcher SJ, Thomas GJ, Jr., Abrogation of hepatitis C virus NS3 helicase enzymatic activity by Bamford DH, et al. RNA packaging device of double-stranded recombinant human antibodies. J Gen Virol 2003; 84: 2323-32. RNA bacteriophages, possibly as simple as hexamer of P4 protein. [203] Prabhu R, Khalap N, Burioni R, Clementi M, Garry RF, Dash S. J Biol Chem 2003; 278: 48084-91. Inhibition of hepatitis C virus nonstructural protein, helicase activ- [222] Stauber N, Martinez-Costas J, Sutton G, Monastyrskaya K, Roy P. ity, and viral replication by a recombinant human antibody clone. Bluetongue virus VP6 protein binds ATP and exhibits an RNA- Am J Pathol 2004; 165: 1163-73. dependent ATPase function and a helicase activity that catalyze the [204] Sullivan DE, Mondelli MU, Curiel DT, Krasnykh V, Mikheeva G, unwinding of double-stranded RNA substrates. J Virol 1997; 71: Gaglio P, et al. Construction and characterization of an intracellular 7220-6. single-chain human antibody to hepatitis C virus non-structural 3 [223] Ahola T, den Boon JA, Ahlquist P. Helicase and capping enzyme protein. J Hepatol 2002; 37: 660-8. active site mutations in brome mosaic virus protein 1a cause de- [205] Mondelli MU, Cerino A, Boender P, Oudshoorn P, Middeldorp J, fects in template recruitment, negative-strand RNA synthesis, and Fipaldini C, et al. Significance of the immune response to a major, viral RNA capping. J Virol 2000; 74: 8803-11. conformational B-cell epitope on the hepatitis C virus NS3 region [224] Til'kunova T, Zinovkin RA, Agranovskii AA. RNA-binding prop- defined by a human monoclonal antibody. J Virol 1994; 68: 4829- erties of proteins of the beet yellows closterovirus. Mol Biol 36. (Mosk) 2004; 38: 553-8. [206] White PW, Titolo S, Brault K, Thauvette L, Pelletier A, Welchner [225] Kalinina NO, Rakitina DV, Solovyev AG, Schiemann J, Morozov E, et al. Inhibition of human papillomavirus DNA replication by SY. RNA helicase activity of the plant virus movement proteins small molecule antagonists of the E1-E2 protein interaction. J Biol encoded by the first gene of the triple gene block. Virology 2002; Chem 2003; 278: 26765-72. 296: 321-9. [207] Faucher AM, White PW, Brochu C, Grand-Maitre C, Rancourt J, [226] Eagles RM, Balmori-Melian E, Beck DL, Gardner RC, Forster RL. Fazal G. Discovery of small-molecule inhibitors of the ATPase ac- Characterization of NTPase, RNA-binding and RNA-helicase activity of human papillomavirus E1 helicase. J Med Chem 2004; 47: tivities of the cytoplasmic inclusion protein of tamarillo mosaic 18-21. potyvirus. Eur J Biochem 1994; 224: 677-84. [208] Venkatesan M, Silver LL, Nossal NG. Bacteriophage T4 gene 41 [227] Kadare G, David C, Haenni AL. ATPase, GTPase, and RNA bind- protein, required for the synthesis of RNA primers, is also a DNA ing activities associated with the 206-kilodalton protein of turnip helicase. J Biol Chem 1982; 257: 12426-34. yellow mosaic virus. J Virol 1996; 70: 8169-74. [209] Raney KD, Benkovic SJ. Bacteriophage T4 Dda helicase translo- [228] Warrener P, Collett MS. Pestivirus NS3 (p80) protein possesses cates in a unidirectional fashion on single-stranded DNA. J Biol RNA helicase activity. J Virol 1995; 69: 1720-6. Chem 1995; 270: 22236-42. [229] Gu B, Liu C, Lin-Goerke J, Maley DR, Gutshall LL, Feltenberger [210] Dudas KC, Kreuzer KN. UvsW protein regulates bacteriophage T4 CA, et al. The RNA helicase and nucleotide triphosphatase activi- origin-dependent replication by unwinding R-loops. Mol Cell Biol ties of the bovine viral diarrhea virus NS3 protein are essential for 2001; 21: 2706-15. viral replication. J Virol 2000; 74: 1794-800. [211] Ziegelin G, Scherzinger E, Lurz R, Lanka E. Phage P4 alpha pro- [230] Zhong W, Ingravallo P, Wright-Minogue J, Skelton A, Uss AS, tein is multifunctional with origin recognition, helicase and primase Chase R, et al. Nucleoside triphosphatase and RNA helicase ac- activities. Embo J 1993; 12: 3703-8. tivities associated with GB virus B nonstructural protein 3. Virol- [212] Lemonnier M, Ziegelin G, Reick T, Munoz Gomez A, Diaz-Orejas ogy 1999; 261: 216-26. R, Lanka E. Bacteriophage P1 Ban protein is a hexameric DNA [231] Gwack Y, Yoo H, Song I, Choe J, Han JH. RNA-Stimulated AT- helicase that interacts with and substitutes for Escherichia coli Pase and RNA helicase activities and RNA binding domain of DnaB. Nucleic Acids Res 2003; 31: 3918-28. hepatitis G virus nonstructural protein 3. J Virol 1999; 73: 2909-15. [213] Tabor S, Richardson CC. Template recognition sequence for RNA [232] Gomez de Cedron M, Ehsani N, Mikkola ML, Garcia JA, Kaari- primer synthesis by gene 4 protein of bacteriophage T7. Proc Natl ainen L. RNA helicase activity of Semliki Forest virus replicase Acad Sci USA 1981; 78: 205-9. protein NSP2. FEBS Lett 1999; 448: 19-22. [214] Ayora S, Weise F, Mesa P, Stasiak A, Alonso JC. Bacillus subtilis [233] Gros C, Wengler G. Identification of an RNA-stimulated NTPase bacteriophage SPP1 hexameric DNA helicase, G40P, interacts with in the predicted helicase sequence of the Rubella virus nonstruc- forked DNA. Nucleic Acids Res 2002; 30: 2280-9. tural polyprotein. Virology 1996; 217: 367-72. [215] Laufs S, Lu A, Arrell K, Carstens EB. Autographa californica [234] Iyer LM, Leipe DD, Koonin EV, Aravind L. Evolutionary history nuclear polyhedrosis virus p143 gene product is a DNA-binding and higher order classification of AAA+ ATPases. J Struct Biol protein. Virology 1997; 228: 98-106. 2004; 146: 11-31. [216] Scheffner M, Knippers R, Stahl H. Simian-virus-40 large-T- [235] Leipe DD, Wolf YI, Koonin EV, Aravind L. Classification and antigen-catalyzed DNA and RNA unwinding reactions. Eur J Bio- evolution of P-loop GTPases and related ATPases. J Mol Biol chem 1991; 195: 49-54. 2002; 317: 41-72. [217] Yang L, Mohr I, Fouts E, Lim DA, Nohaile M, Botchan M. The E1 [236] Frick DN, Rypma RS, Lam AM, Frenz CM. Electrostatic analysis protein of bovine papilloma virus 1 is an ATP-dependent DNA of the hepatitis C virus NS3 helicase reveals both active and allos- helicase. Proc Natl Acad Sci USA 1993; 90: 5086-90. teric site locations. Nucleic Acids Res 2004; 32: 5519-28. [218] Shuman S. Vaccinia virus RNA helicase. Directionality and sub- [237] Brunetto MR, Bonino F. Treatment of chronic hepatitis B: from strate specificity. J Biol Chem 1993; 268: 11798-802. research to clinical practice via the consensus conferences. Curr [219] Merits A, Fedorkin ON, Guo D, Kalinina NO, Morozov SY. Ac- Pharm Des 2004; 10(17): 2063-75. tivities associated with the putative replication initiation protein of [238] Camma C, Di Bona D, Craxi A. The impact of antiviral treatments coconut foliar decay virus, a tentative member of the genus Nano- on the course of chronic hepatitis C: an evidence-based approach. virus. J Gen Virol 2000; 81: 3099-106. Curr Pharm Des 2004; 10(17): 2123-30. [220] Jindal HK, Yong CB, Wilson GM, Tam P, Astell CR. Mutations in [239] D'Cruz OJ, Uckun FM. Clinical development of microbicides for the NTP-binding motif of minute virus of mice (MVM) NS-1 pro- the prevention of HIV infection. Curr Pharm Des 2004; 10(3): 315- tein uncouple ATPase and DNA helicase functions. J Biol Chem 36. 1994; 269: 3283-9.