Kinetic Studies of NS3 and NS5B from Hepatitis C Virus

" #$% " &'(#&#%#'%%#% )*) ) )#&((( ! " ##$%$& $ ' ( ( ) * ! ) ) #* *+,* - . !*+

*/+' (+0 )1" 12) +3 ) ! +$ + 42+42 + +(56(6227677267+ ,* )* - ! )* 6 " 2 81" 129 ) ** 8 :9 * * ) - !! :+ ! 1" -* ) -* 1" * 8;22<$24, 45:9-) * * - )) !* 6 + 2'4 2'5 ** )*) =- *)) * - ! +,* )) * * - !*** ) * + ,* - *- * )* = - * + ! )> = >* * * ) !)6 !*1") =+ 3 * -* ! -* - * !**-) ) - )* - ! * + ) * - 1" ) 17$ * - ! !+,* ) * * -)) 17$ - =6

! * + $ ! - 80 ? ))9

* 809 8.2 9- + ,* *- ) !* - 12 6 * ! ) )** ! ! !* * ! +

,? !** * - ) - -* -** 1" 12+,* )! *!! ) - ! ! ) : ) +

1"12 =? * !

! " # !" $%&'! !()&%*+, !

@/A *' (

314264'7 31(56(6227677267 6(55458* BB +?+B C D 6(55459 To Rebecca

List of Papers

ID ahl, G. , Sandström, A., Åkerblom E. and Danielson U.H. Resistance profiling of hepatitis C virus protease inhibitors using full-length NS3. Antivir. Ther. 2007; 12: 733-740.

II Dahl, G., Sandström, A., Åkerblom E. and Danielson U.H. Effects of protease inhibition by modification of helicase residues in hepatitis C virus nonstructural protein 3. FEBS J. 2007; 274: 5979-5986.

III Nurbo, J., Peterson, S.D., Dahl, G., Danielson U.H., Karlén, A. and Sandström, A. Beta-amino acid substitutions and structure-based CoMFA modeling of hepatitis C virus NS3 protease inhibitors. Bioorg. Med. Chem. 2008; 16: 5590-5605.

IV Dahl, G., Gutiérrez-Arenas, O., and Danielson U.H. Hepatitis C virus NS3 protease is activated by low concentrations of protease inhibitors. Submitted.

V Geitmann, M., Dahl, G., Lohmann, V., Friebe, P., Paeshuyse, J., Herdewjin, P., Puerstinger, G., Bartenschlager, R., Neyts, J. and Danielson U.H. Kinetic, mechanistic and chemodynamic characterisation of non-nucleoside hepatitis C virus NS5B polymerase inhibitors using SPR biosensor technology. Submitted.

VI Geitmann, M., Dahl, G. and Danielson U.H. Kinetic characterization of HCV NS3 interactions with NS4A and inhibitors. Submitted.

Contents

Introduction ...... 11 Proteins, enzymes, proteases and polymerases ...... 11 Hepatitis C ...... 12 The disease ...... 12 HCV Genotypes ...... 12 HCV life cycle ...... 13 The viral proteins ...... 14 HCV Model systems ...... 16 Combating HCV ...... 17 Current treatment ...... 17 HCV drug discovery ...... 18 Drug resistance ...... 20 Immune escape ...... 20 HCV NS3 ...... 22 Structure and function ...... 22 NS3 as a drug target ...... 23 Model systems of NS3 ...... 25 HCV NS5B ...... 26 Structure and function ...... 26 NS5B as a drug target ...... 27 Aims ...... 30 Experimental procedures ...... 31 Cloning, expression and purification of NS3 ...... 31 Activity and inhibition measurements of NS3 ...... 31 Biosensor assays for NS5B and NS3 ...... 32 Results ...... 33 Improving expression and purification of NS3 ...... 33 NS3 protease inhibitor resistance (Paper I) ...... 34 Helicase can influence protease inhibition (Paper II) ...... 37 Novel NS3 inhibitor designs (Paper III) ...... 38 Inhibitor activation of NS3 (Paper IV) ...... 38 Studies of interaction between NS5B and inhibitors (Paper V) ...... 40 Characterization of NS3 inhibitors using a biosensor (Paper VI) ...... 41 Perspectives ...... 44 Summary ...... 44 The future ...... 44 Sammanfattning ...... 46 Hepatit C ...... 46 Proteiner, enzymer och läkemedelsutveckling ...... 46 Att slå ut hepatit C viruset ...... 47 Modellsystem ...... 47 Min forskning ...... 48 Acknowledgements ...... 50 References ...... 52 Abbreviations

ADME Adsorption distribution metabolism excretion ATP Adenosine tri-phosphate Cardif CARD-containing adaptor protein CHAPS 3-[(3-Cholamidopropyl)dimethylammonio]-1-propane- sulfonate CoMFA Comparative molecular field analysis DABCYL 4-((4-(dimethylamino)phenyl)azo)benzoic acid DMSO Dimethyl sulfoxide DTT Dithiothreitol E1 and E2 Envelope protein 1 and Envelope protein 2 EDC 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide hydro chloride EDANS 5-((2-Aminoethyl)amino)naphthalene-1-sulfonic acid EMCV Encephalomyocarditis virus ER Endoplasmatic reticulum FRET Förster resonance energy transfer F-protein Frameshift protein HCV Hepatitis C virus HEPES 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid HIV Human immunodeficiency virus Huh-7 Human hepatoma cell line 7 IMAC Immobilized metal ion affinity chromatography IPS-1 IFN- promoter stimulator 1 IRES Internal ribosome entry site JFH-1 Japanese fulminant hepatitis 1 MAVS Mitochondrial antiviral signaling protein MuLV Murine leukemia virus NA Nucleoside analogue Neo Neomycin NHS N-hydroxysuccinimide NNA Non-nucleoside analogue NS Non-structural protein NTP Nucleotide triphosphate PCR Polymerase chain reaction PEG Polyethylene glycol PKR Protein kinase R RNA/DNA Ribonucleic acid/Dehydroxy ribonucleic acid SPR Surface plasmon resonance TLR3 Toll-like receptor 3 TRIF Toll/Interleukin-1-receptor-domain-containing adaptor-inducing interFeron- VISA Virus-induced signaling adaptor Introduction

Biochemistry teaches us about the chemistry of life; How does life work? The more you dig into the mountain of knowledge amassed through the his- tory of natural sciences and learn how things were, are and relate to one another, one cannot avoid being amazed of the complex chemical processes that happen all around us at all times. However, the more you learn, the more you realize how little is actually known. The task for a scientist is to continue the search for knowledge, giving his or her contribution, and by doing so make the world a better place. Even though the practical use for scientific work is sometimes hard to see, not even the wisest can predict the usage of the knowledge obtained. When my grandfather Olle Dahl defended his thesis in 1958 (Dahl 1958), the applications and benefits that mankind have today, just 50 years later, thanks to biochemistry and biotechnology, was unimaginable.

Proteins, enzymes, proteases and polymerases Proteins are composed of long chains of amino acids. Shorter chains of amino acids are called peptides. There are 20 natural amino acids with different physico-chemical properties and since proteins can contain from a few to several thousand amino acids, the variation of functionality in proteins is tre- mendous. Proteins exist in all forms of life and are responsible for essentially all processes taking place; be it the collagen holding your body together or the enzymes responsible for digesting that cinnamon roll you ate yesterday. Enzymes are a subgroup of proteins; namely proteins that can catalyze chemical reactions. These biocatalysts are able to catalyze complicated and energy demanding reactions at normal physiological conditions and at high rates. Proteases are a subgroup of enzymes, categorized by their function; they catalyze the reaction in which proteins are cut into smaller pieces. There are proteases present everywhere, helping us digest meat, fight infections and clean up inside cells. A polymerase is an enzyme which catalyzes the copying of genetic material, be it either ribonucleic acid (RNA) or dehydroxy ribonucleic acid (DNA). The polymerase function is fundamental in life since life would cease to exist without replication of our genes.

11 Proteases and polymerases are two groups of enzymes, among many, which are crucial for life. But they can also extinguish it; errors in proteases or polymerases can be fatal and disease-causing parasites, fungi, bacteria and viruses all have their versions of these enzymes. Obtaining knowledge about the enzymes in entities responsible for suffer- ing and disease in order to find a cure is an area of extensive research. In this thesis, a protease and a polymerase from hepatitis C virus, have been studied and the results obtained may aid in the finding of a future cure for this wide- spread infection.

Hepatitis C The disease Hepatitis C virus, the causative agent of hepatitis C, is a RNA-virus belong- ing to the Flaviviridae family of viruses (Weiner et al. 1991). Over 120 million people are currently infected worldwide (Shepard et al. 2005). It is transmitted through blood-to-blood contact. This makes the re-usage of shots, needle sharing between injecting drug users and usage of non-sterile needles for piercings and tattoos, means for spreading the infection. Since the hepatitis C virus was not discovered until 1989 (Choo et al. 1989), blood transfusions performed prior to adequate testing could be made, was a major spreader of the virus. The initial symptoms of the infection are often absent or, if they occur, diffuse, with abdominal pain and influenza-like symptoms. However, in some cases the infection does cause jaundice in the initial phase, resulting in a correct diagnosis. But in most cases the infection remains undiagnosed (Hoofnagle et al. 1997). Around 85% of all people infected with hepatitis C virus develop a chronic infection and 20% of them develop cirrhosis within 20 years (Seeff et al. 2001). Of the cirrhotic patients, 20-25% will later suffer liver failure or hepatocellular carcinoma and, using the numbers above, that means around 5 million people worldwide. Still, an HCV infection is cur- able, in contrast to, for example, an human immunodeficiency virus (HIV) infection.

HCV Genotypes There are six major genotypes of hepatitis C virus (1 to 6) which are further divided into subtypes (a, b, etc.) (Bukh et al. 1995). The genotypes are somewhat separated by geography with 1, 2 and 3a being most common in the western world, 1, 2, 4 and 5 in Africa and 3 and 6 in Asia (Simmonds et al. 2005). The sequence similarity between genotypes is only around 65%, and between subtypes 75%. Consequentially, you face not one adversary but sev-

12 eral when trying to find a cure for hepatitis C virus infection, and treatment has to be adapted to the genotype in question.

HCV life cycle The life cycle of hepatitis C virus has been generally deciphered (Figure 1). After docking to the liver cell using several cellular receptors (Agnello et al. 1999)(Pileri et al. 1998)(Lozach et al. 2003)(Saunier et al. 2003)(Evans et al. 2007), the virus merges with the cytoplasm through endocytosis (Blanchar et al. 2006). Once inside, the viral particle is uncoated, releasing its RNA. This RNA, which is of positive sense, contains an internal ribosome entry site (IRES) in its 5'-end (Tsukiyama-Kohara et al. 1991)(Wang et al. 1995), which mimics the cellular mRNA 5'-cap, fooling cellular ribosomes to trans- late it into a polyprotein. This polyprotein is associated with the endoplasmatic reticulum (ER) of the cell and processed by cellular and viral proteases into native proteins (Hijikata et al. 1991A)(Shimotohno et al. 1995)(Grakoui et al. 1993A)(Tomei et al. 1993). Viral proteins form a replication machinery, which create RNA copies in a part of the ER called the membranous web (Gosert et al 2003). The new RNA-molecules are then enclosed and assembled into new viral particles, which mature before leaving the cell through exocytosis.

Figure 1: Left: Schematic representation of a hepatitis C virus particle. The core protein forms the RNA capsid, which in turn is encased by a membrane, containing envelope proteins 1 and 2. Right: The life cycle of HCV; After entry (a), the virus is uncoated, releasing the viral RNA (b). The RNA is translated into a polyprotein that is processed into native proteins (c). New viral RNA is synthesized by the viral replication machinery in a department of the cell called the membranous web (d). The new viral RNA is packaged and assembled into new viruses (e) which mature and are released from the cell (f). Right picture taken from (Moradpour et al. 2007) with permission.

13 The viral proteins The translation of the HCV genome and the processing of the polyprotein results in ten native proteins: Core (C), Envelope protein 1 (E1), Envelope protein 2 (E2), p7, non-structural (NS) proteins 2, 3, 4A, 4B, 5A and 5B (Figure 2). These proteins have several functions, but their role in the viral life cycle is as follows: The core protein forms the RNA-binding capsid (Santolini et al. 1994)(Hwang et al. 1995). The envelope proteins, which are highly glycosylated, form a part of the virus outer surface (Hijikata et al. 1991B)(Goffard et al. 2003). These proteins are referred to as the structural proteins since they are part of the viral particle. p7 is a small protein with an ion channel activity and is crucial for HCV proliferation (Carrère-Kremer et al. 2002)(Pavlovic et al. 2003)(Griffin et al. 2003)(Sakai et al. 2003)(Stein- mann et al. 2007). Its exact role is however still unclear. The non-structural proteins form the replication apparatus, responsible for making new copies of the viral RNA. NS2 is a homodimeric cystein protease of a unique fold, responsible for the cleavage of the NS2-NS3 polyprotein junction (Grakoui et al. 1993B)(Lorenz et al. 2006). NS3 is a multi- functional enzyme with protease and helicase/NTPase activity (Gallinari et al. 1998). The protease is responsible for cleaving all junctions between the non-structural proteins NS3 to NS5B (Bartenschlager et al. 1994), and the helicase/NTPase is responsible for separating double stranded RNA into single stranded (Gwack et al. 1996)(NS3 will be described in greater detail later on). NS4A acts as a cofactor to NS3, activating the protease and an- choring it to the ER (Tanji et al. 1995A)(Koch et al. 1996). NS4B is a trans- membrane protein able to alter membrane properties (Hüggle et al. 2002) (Egger et al 2002). NS5A is a phosphoprotein which can undergo hyper phosphorylation. Though essential, its exact role in the HCV life cycle is still unclear (Kaneko et al. 1994)(Tanji et al. 1995B). NS5B is a RNA-dependent RNA-polymerase (RdRp) responsible for the synthesis of new RNA-molecules (Behrens et al. 1996)(This protein will also be described in greater detail later on). The HCV genome can also encode another protein, called F- protein (frame-shift protein), through a +1 ribosomal frame-shift of the RNA encoding the core protein (Xu et al. 2001). This protein and antibodies against it has been found in infected humans (Walewski et al. 2001), but if the F-protein has any role in the viral life cycle or pathogenesis of HCV is still unclear.

14 Figure 2: Illustration of IRES-mediated translation of the HCV genome, as well as processing, folding and membrane association of the HCV proteins. The IRES of the 5'-NCR enables direct translation of the 9.6 kb viral genome by cellular ribosomes into a polyprotein. This polyprotein is processed by ER signal peptidases (solid dia- monds, scissors) and the core protein is further cleaved by signal peptide peptidase (open diamond). The ten viral proteins fold and associate with the ER. The known protein structures are shown in ribbon representations and unknown structures as geometrical shapes. The green trans-membrane segment of the core protein indicates the part that is removed by signal peptide peptidase. E1 and E2 are envelope proteins which are heavily glycosylated as indicated by the yellow dots. p7 form two trans- membrane alpha helices with shown ion channel activity. NS2 is a homodimeric cysteine protease responsible for cleaving the NS2-NS3 junction (circular arrow). NS3 is a protease responsible for cleaving all remaining polyprotein junctions (indicated by arrows) and a helicase responsible for separating double stranded RNA. NS3 is shown with protease and helicase domains separated, and the protease is associated to NS4A, which functions both as an NS3 protease activator and a membrane anchor. NS4B is a trans-membrane protein, capable of modifying membrane properties. NS5A is a phosphoprotein with an unclear function. NS5B is an RNA- dependent RNA-polymerase, responsible for synthesizing new RNA molecules, with a C-terminal membrane anchor. Pictures taken from (Moradpour et al. 2007) with permission.

15 HCV Model systems Studies of the various viral proteins cannot be performed directly. We cannot put infected patients or their livers under the microscope and hope to see much. Instead, various model systems have to be used. These model systems involve studies of individual proteins, parts of the replication machinery, viral entry, the whole virus in cells or infected test animals and patients. The model system chosen depends on the scientific question one hopes to answer. Experiments with isolated enzymes began shortly after the discovery of HCV in 1989. Early studies of the life cycle of the virus was hampered due to very poor viral replication in cell culture (Kato et al. 1996), making it very difficult to study. Furthermore, the only available animal model was the chimpanzee, making in vivo studies of the infection ethically challenging and costly (Prince et al. 1994). Different mice strains were developed to monitor in vivo conditions (Lerat et al. 2002)(Mercer et al. 2001), but the details of the viral replication inside cells was still unknown. It was not until the development of the HCV replicon system in 1999 that parts of the sub-cellular viral life cycle could be studied (Figure 3) (Lohmann et al. 1999). The replicon system is a cell culture-based model system where the genes coding for the structural HCV proteins have been removed. Translation directed by the HCV IRES results in the production of aminoglycoside phosphotransferase, giving the cell resistance to neomycin (Neo), and enables selection of transfected cells. The IRES from another virus, encephalomyocarditis virus (EMCV), directs another translation which results in synthesis of the HCV non-structural proteins. These form the replication machinery responsible for the synthesis of new replicon RNA. However, in order to work efficiently, several adaptive mutations in the RNA encoding NS3, NS5A and NS5B are required (Lohmann et al. 2001, Krieger et al. 2001). With the introduction of the replicon system it became possible to study all non-structural proteins in a cellular environment. Originally developed from genotype 1b, there are now replicons available for viral strains 1a and 2a as well (Blight et al. 2003)(Kato et al. 2003).

Figure 3: Schematic view of a HCV replicon genome. The HCV IRES controls the expression of the Neo-gene, giving the cell resistance to neomycin, while the EMCV IRES controls the expression of the HCV non-structural proteins which in turn are responsible for the synthesis of new replicon RNA.

16 In a similar way, viral entry has been possible to monitor by using HCV pseudo particles (Bartosch et al. 2003). Here it is the non-structural proteins that have been removed and the genome expresses HCV E1 and E2 protein together with murine leukemia virus (MuLV)- or HIV core protein. Since the structural proteins are omitted in the replicons and the non-structural proteins are omitted in the pseudo particles, new viruses with which to infect new cells cannot be formed. Much effort was invested into finding a cell culture system for the entire virus, but adapted changes in replicon RNA that improved RNA replication in vitro resulted in reduced infectivity in vivo (Bukh et al. 2002). A milestone in HCV research was reached in 2005; a HCV isolate from a Japanese patient, lacking any adaptive mutations, replicated efficiently in cultured cells (Wakita et al. 2005). This strain of HCV, called Japanese fulminant hepatits 1 (JFH-1) and belongs to genotype 2a, could replicate inside a human hepatoma cell line (Huh-7) and produce viruses which were able to infect new cells. With this new cell based model system, the final questions regarding the viral life cycle will hopefully be answered. All these model systems have been used to elucidate the life cycle of HCV and the role of the individual proteins.

Combating HCV Current treatment There is still no vaccine for hepatitis C virus and current treatment is limited to combination therapy using interferon and ribavirin. Interferons are protein cytokines, a form of signaling substances which activate the immune system (Isaacs et al. 1957). Ribavirin is a nucleoside analogue, similar to guanosine, that functions as a general anti-viral drug (Sidwell et al. 1972). Both these compounds are indirect antiviral drugs. Interferon must be administered intravenously and the stability in blood is improved by using poly-ethylene glycol (PEG) fused to interferon (Abuchowski et al. 1977). Thus injections can be made once a week whereas ribavirin is administered orally on a daily basis. The side effects from this combination therapy is influenza like symptoms (fever, head ache and joint- and muscle pain), hemolytic anemia and depressions. The length of treatment and the outcome depends greatly on the genotype of the virus with up to 90% cure rates for genotypes 2 and 3, but only around 50% for genotype 1 and 4 (Manns et al. 2001)(Fried et al. 2002). Treatment times can be as long as 1.5 years, and this in combination with the side effects show the urgent need for new and more efficient drugs.

17 HCV drug discovery Even though much effort is directed into finding a vaccine for HCV and the eradication of this disease will probably involve a prophylactic vaccine in one form or another, how this is done will not be described here. Nor will this section be devoted to other indirect anti-viral drugs that are in development such as different improvements of interferon (Osborn et al. 2002)(Chan et al. 2007)(Musch et al. 2004)(Okuse et al. 2005)(De Leede et al. 2008) or ribavirin (Lin et al. 2003)(Huang et al. 2005), glucosidase inhibitors (like Celgosivir) (Whitby et al. 2004), or cyclophilin inhibitors (like DEBIO-025) (Paeshuyse et al. 2006). There are drugs being developed that target NS4A (like ACH-806)(Huang et al. 2007) and NS5A (like A-831/AZD-2836)(Ar- row Therapeutics product pipeline, 2008)(Schmitz et al. 2008), but they have only been studied using the replicon system and exactly how they work on the protein level is unclear. Therefore NS4A and NS5A inhibition will not be further discussed here. This section will focus on the development of anti- HCV drugs that target the viral enzymes NS3 and NS5B. The research focus in this field has been devoted to finding “the third drug”, a compound that can be administered synergistically with interferon and ribavirin. Implementing this new and improved therapy will hopefully increase the cure rates for genotypes 1 and 4, with a next step to find a sub- stitute for ribavirin and interferon altogether. The vision is to enable HCV patients to take just one pill a day for a limited time and thereafter be cured. The most important questions in any drug discovery endeavor are; does it work and is it safe? When trying to find a new drug you must decide where to attack; what is a suitable drug target? For viruses it is relatively simple to validate an enzyme drug target. You take viral RNA, introduce a mutation which will result in a disrupted enzyme activity, and then you infect a test animal with this RNA. If the RNA fails to produce viruses which can replicate inside the host, the disrupted activity was most likely essential and inhibitors targeting this enzyme activity will probably make good drugs. This ap- proach was used to validate both NS3 protease activity and NS5B polymerase activity as good targets (Kolykhalov et al. 2000). The methodology is not fail safe however, since this study also pointed out the NS2-NS3 protease as being an equally good drug target. Since a function for this protease other than cleaving the NS2-NS3 junction have not been found, it is currently not believed to be a good drug target even though this may change in the future. After the target has been validated, the first step is to obtain fundamental information about the enzyme. This involves obtaining it in a pure form and determining basic enzymatic parameters. It usually takes years before such assays have been optimized and sometimes, as for NS3, the choice of model system used can be debated (see Model systems section). Once completed, the first real pre-clinical steps aimed at drug discovery are initialized. When it was found that NS3 was inhibited by its own products, the first peptide-

18 based inhibitors started to emerge (Steinkühler et al. 1998). Similarly for NS5B, the first inhibitors were modified nucleotides (Lohmann et al. 1998). These first generations of inhibitors were however useless as drugs; they were not potent enough, they were too big, they had poor selectivity, they were poorly absorbed and they were too unstable. New and better inhibitors are made by modifying structural elements of the compounds, using knowledge of enzyme, substrate and previous inhibitors. Gradually the inhibitors will have less of the unfavorable characteristics described above and be more drug-like. After several years of testing, potent inhibitors with desirable characteristics may be found. These compounds will then undergo further testing using cell-culture systems (replicon system, HCV cell culture system) and test animals to monitor basic toxicity and tolerability and if it passes all controls without remarks, it can enter clinical trials. The final structure of the compound that enters clinical trials may be very different from the initial inhibitors (Figure 4).

N N H O OH N OH S O S O O O O H H H N N N H N N N OH H O 2 N N O O O O SH OH O O N HO O HO O H O Figure 4: Illustration of how an NS3 protease inhibitor is changed, from the peptide DEMEEC (left), to the drug candidate, BILN 2061 (right) during the optimization process. Not many structural elements remain in the final inhibitor design.

Clinical trials is a long process composed of several phases (0 to IV) aimed at testing drug candidates in humans. Studies of tolerability, pharmacokinetics, dose range, drug administration, drug efficacy and side effects are performed involving, from a few up to several thousand participants. Clinical trials are so designed that if a drug candidate fails at some point, further studies are halted and the compound is re-evaluated. Sometimes only minor changes are needed to circumvent the problem, but in the worst case it can lead to a restart of the whole process with another compound, rendering years of clinical trials more or less useless. This was what happened with the first NS3 protease inhibitor to reach clinical trials; BILN 2061 (Ciluprevir), which showed to induce cardiac toxicity in monkeys (Reiser et al. 2005). Be- ing canceled during clinical trials is common, a fate many other potential anti-HCV drugs also have suffered (Georgopapadakou 2007)(Ryder 2007) (Ryder 2009). For drug discovery in general, only 1% of all compound found during pre-clinical trials make it all the way to become drugs and only around 10%

19 of the compounds entering clinical trials result in a new drug. This makes the average development time and cost for a drug 15 years and 800 million US dollars, respectively. As if this was not enough, even after an anti-viral drug has reached the market, the battle is not won. Viruses exists in many genotypes, so the drug efficacy is different from strain to strain. Also, viruses in general, and HCV in particular, are highly variable and mutations leading to drug resistance are major problems.

Drug resistance A virus is constantly changing and it is an important part of their survival strategy. HCV produces an incredible 1012 new viruses per day inside a human and the HCV NS5B RNA polymerase, which lacks proof-reading activities, introduces about 1 error in every 10,000 bases made (Neumann et al. 1998). As a comparison, the human DNA polymerase, which is responsible for copying our DNA upon cell division, introduces about 1 error in 10 000 000 000 bases made and other error-preventing systems reduces this rate even further. Doing some combinatorial math, assuming an even distribution of mutations in the HCV genome, an infected person can potentially carry a virus with every single nucleotide mutation and 10% of all double mutations at any one time (Lenz 2008). This variability makes it difficult to design drugs against HCV since the drug can be very effective against its target, reducing viral loads, only to see the forces of evolution select another, resistant quasi-species of the virus towards which the drug is ineffective. The only way to combat drug resistance is to use synergistic combination therapy. The threat that resistance poses forces pharmaceutical companies to closely monitor their drugs even after they have reached the market.

Immune escape A striking feature of HCV is its ability to persist in the host. It is very un- common for an RNA virus to be able to prevail in 20+ years in humans. One study has identified a viral genome integrated into host DNA but neither the mechanism behind such an event, nor how common this is, has so far been elucidated (Zemer et al. 2008). Even though hepatocytes are the predominant target for HCV, the virus has also been found in lymphoid nodes and brain tissue (Laskus et al. 1998)(Radkowski et al. 2002), indicating “reservoirs” of virus able to reinfect the liver, posing a threat to HCV patients that undergo liver transplantation. The main reason for persistence is however considered to be the virus ability to elude the immune system, called immune escape. HCV has essentially two ways of avoiding immune detection; mutations and host interference.

20 Mutations occur rapidly in HCV, as described above, but they are not evenly distributed. The viral genes encoding the envelope proteins contain so-called hyper-variable regions with a higher sequence variation (Weiner et al. 1991)(Hijikata et al. 1991). The envelope proteins form the major part of the outer surface of a virus that antibodies must recognize and bind to in order to trigger an immune response. With a high mutation rate and the sheer amount of new viruses formed each day, the virus can simply work faster than the immune system; once an antibody has been made, its target virus is rapidly replaced with another version of the virus that the antibody fails to recognize. This strategy is also successful for evading adaptive, T-cell mediated immune responses since viral epitopes constantly change, leading to re- cognition failure of infected cells (Timm et al. 2004). Host interference is the ability for HCV to disrupt immune signaling inside the infected hepatocyte. Viral genomes and proteins are like swiss army knives; many functions in a limited area. Several HCV proteins are involved in tricking the immune system, especially in order to evade the interferon signaling system. The core protein can affect the transcription of cellular genes, in particular genes responding to interferon stimuli, which encodes anti-viral proteins (Ray et al. 1995)(Melén et al. 2004)(de Lucas et al. 2005). HCV IRES and E2 protein can inhibit cellular protein kinase R (PKR), a protein that responds to interferon signaling and down-regulates protein synthesis (Vyas et al. 2003)(Taylor et al. 1999). NS5A has been proposed to inhibit PKR as well (Gale et al. 1998), but this is not universal (François et al. 2000)(Podevin et al. 2001), the NS5A protein does however affect the cellular interferon response (Enomoto et al. 1995). NS3 protease cleaves the IPS-1/MAVS/VISA/Cardif- (Kawai et al. 2005) (Seth et al. 2005)(Xu et al. 2005)(Meylan et al. 2005) and TRIF (Li et al. 2005) proteins, playing a key role in efficiently disrupting both the classical intracellular- and the toll-like receptor 3 (TLR3)-driven endosomal pathways for interferon regulatory gene induction. Many of these functions are probably also the reason why HCV causes hepatocellular carcinoma in humans since one of many ways of recognizing cancer cells is that they become un- responsive to outer stimuli. NS5B polymerase and the NS3 protease are considered the most important drug targets in the HCV field due to their crucial role in the viral life cycle and in evading the immune system.

21 HCV NS3 Structure and function The HCV NS3 protein consists of around 630 amino acids. The enzyme is composed of two domains; the N-terminal third of the protein is a protease and the remainder is a helicase/ATPase (Figure 5) (Yao et al. 1999). The helicase has been classified into super-family 2, DEXH-family, of helicases (Korolev et al. 1998). Its function is to separate double stranded RNA using adenosine tri-phospahte (ATP) as an energy source. The exact mechanism for how the helicase works is still not clear, some studies have shown that NS3 helicase works as dimers, rolling around the RNA molecule (Khu et al. 2001), whereas other studies have shown a monomeric helicase moving like an “inchworm” across the RNA (Preugschat et al. 1996). Recently, indica- tions that the helicase can function both as a dimer and a monomer were presented (Sikora et al. 2008)(Jennings et al. 2008). The NS3 protease is a chymotrypsin-like serine protease (Hahm et al. 1995). The catalytic triad consists of His-57, Asp-81 and Ser-139. The substrate specificity of NS3 is rather broad, it cleaves decapeptides with a consensus sequence of DXXXXC-SXXX (Grakoui et al. 1993A). NS3 contains a structural zinc ion which is of importance for the NS2-3 auto-protease and the NS3 protease domain (Love et al. 1996)(Tedbury et al. 2007). This zinc

Figure 5: Left: A surface and backbone trace representation of full-length NS3 (pdb: 1CU1). The helicase domain (blue) is situated in close proximity to the protease domain (red). A part of the NS4A cofactor (green) has been engineered to be covalently attached to the N-terminus of NS3 through a small GSGS-linker (yellow). Just above this loop structure, the structural zinc atom is seen as a gray sphere. Located in the helicase domain is a phosphate group, bound to the NTPase site. The active site of the protease is indicated by the box. Right: A zoomed in view of the protease active site with the catalytic triad His 57, Asp 81 and Ser 139 shown in ball-and- stick representation. The lower part of the helicase is seen at the top of the picture and the NS3 C-terminus is visible as the blue beta strand arrow.

22 ion is coordinated by residues Cys-97, Cys-99, Cys-145 and, via a water molecule, His-149 of NS3. The protease domain binds the viral protein NS4A as a cofactor (Satoh et al. 1995). NS4A is a 54 amino acid protein with two roles; it stabilizes the structure of NS3 resulting in activation of the protease, and it also anchors NS3 to membranes (Urbani et al. 1999)(Tanji et al. 1995A). NS3 interacts with NS4A through NS3 residues 4-10, 13-21 and 33-37, and NS4A consti- tutes a missing beta-strand in the enzyme structure (Kim et al. 1996). Studies have shown that only the small, predominantly hydrophobic 13 amino acid long fragment of NS4A that binds directly to NS3, is required for full activation of the protease in vitro (Tomei et al. 1996).

NS3 as a drug target The first attempts to find a protease inhibitor for HCV NS3 was hampered by the fact that NS3 has a fairly broad substrate specificity and that the truncated NS3 protease domain, with the shallow and featureless substrate binding site, was used for the studies. It was not until the discovery of product inhibition of NS3 and optimization of these compounds that the first somewhat potent inhibitors were found (Steinkühler et al. 1998)(Ingallinella et al. 1998)(Llinàs-Brunet et al. 1998). The first NS3 protease inhibitor to show a proof-of-concept in vivo was reported in 2003 with BILN-2061 (Ciluprevir), a macrocyclic protease inhibitor developed by Boehringer Ingelheim (Lamarre et al. 2003) (Figure 6). In

N N H O F O N N N S N O S O

H O H O O O N H O O N N O O N O N N O OH S O S N N H O O H O O N O N N H H O

BILN 2061 (Ciluprevir) ITMN 191 TMC435350

N H O N H O N O N O H H N O O N H N O O NH N N N H 2 O H O N O

VX 950 (Telaprevir) SCH 503034 (Boceprevir) Figure 6: Structure of some protease inhibitors that are or have been in clinical trials. BILN 2061 (Ciluprevir) was later withdrawn due to cardiac toxicity. ITMN 191 is in phase Ib clinical trials and TMC435350 in phase IIa clinical trials. VX 950 (Telapre- vir) and SCH 503034 (Boceprevir) are both in phase III clinical trials.

23 humans carrying genotype 1 HCV, the viral levels dropped up to a thousandfold after just a few days taking the drug orally. Unfortunately, when the drug was further tested it gave rise to cardiac toxicity and its development was halted (Reiser et al. 2005). Today, several NS3 protease inhibitors are in different phases of clinical trials: VX 950 (Telaprevir) is a linear, slow-binding, mechanism based protease inhibitor (Perni et al. 2006), developed by Vertex Pharmaceuticals and Tibotec. It is currently in phase 3 clinical trials (Vertex Pharmaceuticals press release, 2008) and has shown good results in combination with interferon and ribavirin (Lawitz et al. 2008). SCH 503034 (Boceprevir), is another protease inhibitor similar to VX 950 both in structure and mechanism (Venkatraman et al. 2006) and is developed by Schering-Plough. This compound is also in phase III clinical trials (Schering-Plough press release, 2008), showing good potency and tolerability. ITMN 191 is a macrocyclic inhibitor (Seiwert et al. 2008), similar in structure to BILN 2061, with an ex- tremely high affinity for NS3, developed by Intermune/Roche. It is currently in phase Ib clinical trials in combination with interferon and ribavirin (Inter- mune press release, 2009). TMC435350 (Raboisson et al. 2008), also a macrocyclic protease inhibitor, developed by Tibotec and Medivir is in phase IIa clinical trials (Medivir Press release, 2008). There are also many more compounds in different stages of pre-clinical trials, but all of them cannot be covered here. Some compounds show reduced efficacy for certain genotypes where, for example, BILN 2061 is less potent towards genotype 2 and 3 compared to genotype 1 both in vitro and in vivo (Thibeault et al. 2004)(Reiser et al. 2005). Furthermore, resistant escape mutants have been identified in both replicon cells and in humans where substitutions of certain amino acids in the protease domain of NS3 confer high levels of resistance (Trozzi et al. 2003)(Lin et al. 2004)(Sarrazin et al. 2007). BILN 2061 has a reduced potency towards enzymes carrying substitutions R155Q/K, A156T/V and D168A/V in vitro (Lin et al. 2004)(Lu et al. 2004)(Paper I)(He et al. 2008). VX 950 has impaired efficacy towards A156S/T/V and R155Q/K enzyme in vitro (Lin et al. 2004)(Paper I)(He et al. 2008) and in vivo the A156T/V, V36M/A + R155K/T or V36M/A + A156T/V substitutions has shown to confer a high level of resistance (Sarrazin et al. 2007). SCH 503034 has a mod- erately reduced efficacy towards enzymes with substitutions T54A, A156S and V170A and a large potency reduction towards variant A156T in vitro (Tong et al. 2006). ITMN 191 is less efficient towards V170A, R155K, A156S/T and D168A/V/S/E enzymes in vitro (He et al. 2008). For TM- C435350, no resistance information is currently available. Taken together, it becomes clear that changing some amino acids in NS3 can confer cross-resistance to several protease inhibitors. In particular residues Arg 155, Ala 156 and Asp 168 seem to be involved and prone to substitutions. Thus, a therapy using only protease inhibitors will probably

24 not work as a cure for HCV infection, and it is clear that studies of NS3 has to continue in order to find new protease inhibitors with different resistance profiles that can be used to fight emerging drug resistance.

Model systems of NS3 Even though the two domains of NS3 are functional as separate entities, they are not separated in vivo. However, most studies performed on isolated NS3 are done with truncated protein and not the full-length enzyme. The reasons for this are many; full-length NS3 is more difficult to over-express and purify than truncated forms (Poliakov et al. 2002), structure determinations have revealed that the folding of the separate domains are identical to the folding of the domains in the full-length NS3 (Love et al. 1996)(Kim et al. 1996) (Yao et al. 1999) and the enzymatic profiles of the two domains correlate well with the activity for the full-length enzyme (Howe et al. 1999)(Thibeau- lt et al. 2009). There are however several studies showing the interplay between the two domains, where helicase activity is increased in full-length NS3 compared to using only the helicase domain and likewise the helicase domain have an impact on the activity of the protease domain (Sali et al. 1998)(Johansson et al. 2001)(Kuang et al. 2004)(Frick et al. 2004)(Gu et al. 2005)(Zhang et al. 2005)(Beran et al. 2007)(Beran et al. 2008)(Paper II) (Paper IV). The shallow substrate binding site of the NS3 protease is believed to make the design of inhibitors with high specificity and affinity difficult. This is however not necessarily true. A crystal structure of full-length NS3 positions the helicase domain in close proximity to the protease (Yao et al. 1999), forming a much more enclosed protease active site. Thus protease inhibitors should be able to utilize residues in the helicase domain as anchor points. Still, the positioning of the two domains could be due to the crystallographic forces at work. Furthermore, a protease inhibitor relying on interactions with helicase residue has yet to be seen and no crystal structures showing a protease inhibitor co-crystallized with full-length NS3 have been presented so far. A recent study of how NS3, in the absence of NS4A, binds to intracellular membranes indicates that the helicase domain cannot be positioned close to the protease domain (Brass et al. 2008). Instead, the two domains must be separated to some extent and thus model systems using truncated NS3 protease will not be as far from the presumed in vivo situation. The recently solved structure of full-length NS3 from the related Dengue virus, where helicase and protease domains are in close contact but in a much different configuration than in the HCV structure (Luo et al. 2008), could perhaps shed some light on how the two domains interact. Two recent studies sugges- ted contradicting conclusions of whether truncated protease was as good a

25 model system as full-length enzyme, showing that the matter is not settled (Beran et al. 2008)(Thibeault et al. 2009). To summarize, full-length NS3 can probably adapt several conforma- tions, some with domains in close proximity and some further away, and in solution this transition is less hindered than when the protein is associated to a membrane. This flexibility may be important for the function of the enzyme activities (Paper IV), with conformation changes varying depending on the task at hand. By omitting one of the two domains upon enzymatic analysis, the model system becomes less complex but could also be less rel- evant and the risk of missing important information is evident. For these reasons, full-length NS3 have been used for all studies presented here.

HCV NS5B Structure and function The NS5B RNA-dependent RNA polymerase consists of about 590 amino acids and has three domains; finger, palm and thumb, analogous to a right hand (Figure 7) (Lesburg et al. 1999). The finger and thumb domains, which interact with one another forming a “closed” substrate binding tunnel, binds the RNA template, through electrostatic interactions between enzyme Arg and Lys residues and RNA phosphate groups, directing the substrate 3'-end towards the NS5B palm region (Labonté et al. 2002)(O'Farrell et al. 2003). Here, the two aspartic acid residues 220 and 318 coordinates two magnesium ions (manganese in the crystal structure in Figure 7) which in turn catalyze the formation of the phospodiester bond between nucleotides in a primer-independent manner (Oh et al. 1999). The C-terminal 21 amino acids, position 570 to 590, constitute a membrane anchor domain that binds the enzyme to the ER, and it is crucial for the formation of a functional replication complex in vivo but is not required for full enzymatic activity in vitro (Lee et al. 2004)(Ferrari et al. 1999). The amino acids between position 545 and 562 form a flexible loop capable to inhibit polymerase activity, as has been shown in truncation studies (Adachi et al. 2002). As a consequence, truncated versions of NS5B are commonly used for enzymatic studies where either only the membrane anchor (Δ21) or the anchor plus the flexible loop (Δ51) are omitted. In contrast to NS3, the overall structure of NS5B is not believed to change much upon membrane association.

26 Figure 7: A surface and backbone trace representation of NS5BΔ21 (pdb: 1NB7). The three domains; thumb (green), palm (magenta), fingers (blue) and the C-terminal flexible loop structure (yellow) is shown. The membrane anchor is not present in the structural representation. Located in the palm domain are the two aspartic acid residues 220 and 318, shown in ball-and-stick representation, coordinating two manganese (magnesium in vivo) ions which participate in the catalysis. A small PolyU oligonucleotide, shown in stick representation, is bound to the enzyme. The right figure shows the location of amino acids, as spheres, that are important for the binding of NAs; Ser 96 in purple and Ser 282 in cyan, and NNAs; Pro 495 and Pro 496 in orange, Met 414, Tyr 415 and Tyr 448 in gray and Ser 476 and Tyr 477 in red.

NS5B as a drug target Several facts point to NS5B being an ideal drug target for combating HCV; It has a crucial role in the viral life cycle, humans do not have any RNA-dependent RNA polymerases, and NS5B can be inhibited both directly at the active site but also indirectly via several allosteric sites (Figure 7). Nucleos- ide analogues (NAs) (Figure 8), can function both as competitive substrates in NS5B RNA synthesis, leading to early termination or genetic errors once incorporated, or as competitive inhibitors. One of the first NAs described was 2'-C-metyladenosine which proved to be a potent inhibitor towards NS5B in vitro (Carrol et al. 2003). This concept was also successful for cytidine analogues, leading to the discovery of NM 283 (Valopicitabine) by Idenix Pharmaceuticals, a 2'-C-metylcytidine pro-drug which entered phase II clinical trials in 2006, but appears to no longer be in development (Pierra et al. 2006)(Idenix product pipeline 2009). Another NA, R1626, is under development by Roche (Roberts et al. 2008). R1626, a pro-drug of 4'-azidocytidine (R1479), entered phase IIb clinical trials in fall 2007 (Roche update 2008). However, one general concern with NAs is that they can be antagon- istic in combination with Ribavirin (Coelmont et al. 2008), making them less viable candidates for such a combination therapy. Non-nucleoside analogues (NNAs) (Figure 8) inhibit NS5B by binding to the active site or one of several allosteric sites (Figure 7). Benzimidazole in-

27 hibitors (Tomei et al. 2003), like JT-16 developed by Japanese Tobacco Com- pany, bind to the allosteric site around Pro 495, locking NS5B in an open non- catalytic conformation. Benzothiadiazine inhibitors (Dhanak et al. 2002), like A-848837 developed by Abbott, bind to the interface between thumb and palm domains around residue Met 414 and benzofurans (Gopalsamy et al. 2006), like HCV-796 developed by ViroPharma, binds close to this site as well. Thiophenes (Chan et al. 2004) bind to the allosteric site around residues Ser 476 and Tyr 477 and is believed to disturb the replication complex formation. Diketoacids are pyrophosphate analogues and bind the magnesium ions in the active site (Summa et al. 2004)(Paper V). Currently only HCV-796 is in clinical trials, phase II (HCV-796 clinical trials, 2009).

O NH H 2 HO O N N HO N N O F O HO N N O O CH3 S O O O OH H O N Benzofuran, HCV 796 H4N+ Cl O

2'-C-methylcytidine Cl O S pro-drug, Benzimidazole, JT-16 OH

R NH2 O O N S N Thiphene carboxylic acid O O N S O O O N O O N

N3 O O O OH O O O OH R O 4'-Azidocytidine pro-drug, R1626 Benzothiadiazine, A-848837 Diketoacid Figure 8: Structure of some NS5B polymerase inhibitors including the nucleoside analogues NM 283 and R1626 and the non-nucleoside analogues JT-16, A-848837, HCV 796 as well as general structures of thiophenes and diketoacids.

As for NS3 inhibitors, selection experiments with NS5B inhibitors using replicon cells result in resistant escape mutants. The location of the adaptation is dependent on the inhibitor. For NAs, resistance towards NM 283 is achieved through a S282T substitution, and for R1626 through a S96T substitution but no cross resistance has been reported for these two compounds (Le Pogam et al. 2006A). For NNAs, resistance substitutions occur at the respective binding sites. Benzimodazole inhibitors lose more than a hundred-fold in potency when residue 495 (P495L/S/T/A) or the adjacent residues 496 (P496S) or 499 (V499A) are substituted (Kukolj et al. 2005). These compounds also lose potency towards other selected variants (L419M, M423V/L/T, I482L) (Howe et al. 2006)(Le Pogam et al. 2006B).

28 Benzothiadiazine and benzofuran inhibitors and variants carrying substitutions at position 414 (M414I/V or M414T/L) (Nguyen et al. 2003) or 316 (C316Y/F/S) (Howe et al. 2008) result in up to thousandfold reduced potencies. Thiophene based inhibitors are less efficient towards NS5B enzymes carrying substitutions at position 419 (L419M), 423 (M423T/V/I) and 482 (I482L) (Le Pogam et al. 2006B). Selection with both a thiophene- and a thi- adiazine inhibitor resulted in double resistant variants (M414L + M423T or M414L + M423T + I482L) with reduced susceptibility towards both classes of inhibitors (Le Pogam et al. 2006B). Replicons resistant to both polymerase and protease inhibitors have also been identified, harboring adaptive mutations in the same genome (Flint et al. 2008), showing that monitoring of emerging resistance is crucial for obtaining a cure for HCV infection.

29 Aims

The aims of these studies were to study the kinetics of the NS3 and NS5B enzymes from HCV. Such knowledge provide insights into their enzymatic properties and how drugs targeting them can be designed. In order to perform these studies, several methodologies and strategies had to be developed. The work specifically involved:

Improve expression and purification procedures to increase yields and quality of full-length NS3 enzyme, enabling the use of new analysis methods Investigate the effect that specific amino acid substitutions in full- length NS3 can have on activity and inhibition of the protease. This involved looking at residues contributing to protease inhibitor resistance, R155, A156 and D168, and two, previously unstudied, helicase residues, Q526 and H528 Analyze new NS3 protease inhibitors, providing insights into how a such compounds can be designed and optimized Investigate why, and how, some protease inhibitors at low concentrations activate full-length NS3 protease Study the detailed binding kinetics of inhibitors to NS3 or NS5B by developing SPR-based biosensor assays for these enzymes

30 Experimental procedures

Cloning, expression and purification of NS3 Construction of gene variants encoding amino acid substitutions R155Q, A156T, D168V (Paper I), Q526A, Q526S, H528A and H528S (Paper II) were made using PCR and blunt-end ligation procedures. An expression and purification procedure for full-length NS3 had already been developed and described by Poliakov A. et al. 2002. In short, a pBAD-vector containing the gene for full-length NS3, genotype 1a, was transformed into competent E. coli TOP10 cells and liquid cul- tures were grown and protein expression was induced by addition of L-(+)- arabinose. Cells were then harvested, resuspended, lysed and the cleared ly- aste was loaded onto an IMAC column. After washing, partially purified enzyme was eluted from the column using imidazole. The sample underwent a buffer change and was loaded onto a PolyU-column; washed and pure NS3 was finally eluted using NaCl and stored at -80°C.

Activity and inhibition measurements of NS3 Pure, full-length NS3 was thawed, diluted in assay buffer (50 mM HEPES pH 7.5, 40% (w/v) glycerol, 0.1 % (w/v) n-octyl--D-glucopyranoside and 10 mM DTT) to 6 μg/ml and refrozen in liquid nitrogen. The standard procedure for activity measurements was performed by incubating 1 nM NS3 and 25 μM NS4A-peptide (KKGSVVIVGRIVLSGK) in assay buffer for 10 minutes at 30°C before adding the substrate. When inhibition was monitored, the inhibitor was also included in the incubation step. Modification to this protocol was implemented when looking at the enzyme activation (Pa- per IV), where components were added in different orders. The substrate used acted according to Förster resonance energy transfer (FRET)-principles; a fluorescent group (EDANS) and a quenching group (DABCYL) are covalently attached to the substrate peptide. On the intact substrate the fluorescence signal is efficiently quenched but once the protease has cleaved the substrate, a signal is detected. This processing of substrate was monitored using a fluorescence plate reader and initial rate velocities were determined. For analysis of activity as a function of substrate, the concentrations were varied between 0.125 to 10 μM. The Michaelis-Menten

31 equation was then fitted to the initial rate velocities and Km- and Vmax-values for the enzymes were determined. For determination of Ki-values, different inhibitor concentrations and a constant substrate concentration of 0.5 μM was used. An equation describing competitive inhibition, or tight-binding competitive inhibition, was fitted to the initial rate velocities.

Biosensor assays for NS5B and NS3 NS5B, genotype 2a, was immobilized using amine coupling, by injecting it on a NHS/EDC activated sensor chip surface. Interactions between NS5B and various NS5B non-nucleoside analogues, magnesium or RNA were performed by injecting the ligand and following the signal as a function of time (sensorgrams). Sensorgrams from reference surfaces and blank injections were subtracted from the raw data. Kinetic parameters from the interaction were determined by fitting a suitable equation, representing an interaction model, to the corrected data. NS3 was immobilized using amine coupling and, for studies of NS3/NS4A complexes, NS4A was then injected followed by an deactivation of the surface using ethanolamine or Tris-HCl. Interactions between NS3 and NS3 protease inhibitors or NS4A were performed in the same way as for NS5B. All biosensor experiments were performed on a BIAcore 2000 instrument.

32 Results

Improving expression and purification of NS3 The protocol used for expression and purification of NS3 (Poliakov et al. 2002), is cumbersome, demanding, and produced very low yields (around 0.1 mg enzyme per liter cell culture). The reasons for this could be many; NS3 is expressed without NS4A leading to an unstable protein and since NS3 is membrane-associated on its own, lack of intracellular membranes in E. coli could prevent efficient over-expression. Furthermore, NS3 is rather “sticky” with a habit of binding to other proteins and cell contents, resulting in loss of material during the purification. Analysis of lysed cells shows NS3 activity in the pellet after centrifugation, so some material is lost in the insol- uble fraction. However, this was the best system found since expression trials using other E. coli strains and vectors did not result in any NS3 at all. Puri- fication using denaturing conditions and refolding have been described (Li et al. 2003), but with low yields of refolded protein contra unfolded protein and no means for separating the two, this method has not been used further. In the hope of increasing NS3 yields and stability, a new gene construct was developed. It included a small part of NS4A to be co-expressed with NS3, a construct which would result in an enzyme similar to the one depic- ted in Figure 5 (Yao et al. 1999). The difference was that instead of using just a small GSGS-loop structure to link the NS4A-segment to NS3, a poly- histidine sequence was inserted. This served the purpose of acting both as an affinity tag for IMAC-purification and a way to know where the tag was situated in the construct, preventing it from interfering with enzymatic functions. The initial experiments showed promising results; lysates from induced cells contained active NS3. Furthermore, there was no need to add any NS4A, indicating a correct fold of NS3 and association to NS4A. However, the expression levels were still low and purification of this new protein construct was unsuccessful despite much effort and time invested. Initial attempts of using P. pa st o ri s instead of E. coli for this gene construct were performed but canceled at an early stage. Focus was instead directed at improving the existing protocol as much as possible. In the original protocol, the steps from harvesting the cells to after the IMAC purification must be performed non-stop, something that takes between 12 to 18 hours. Attempts to pause at different time points, like freeze cell pellets or lysates were unsuccessful and resulted in total enzyme

33 inactivation. As a consequence, two persons were normally required to man- age the NS3 purification. However, NS3 was found to be quite stable inside bacteria and changing induction conditions from 3h at 30°C to 14h at 20°C gave similar final yields. This was implemented, increasing the total time needed for the entire purification, but the overall procedure was simplified and manageable for one person. However, for reasons unknown, this procedure ceased to work at some point during 2007. Attempts to use it resulted in too low enzyme yields, so low that the chromatographic purification was unsuccessful. Much time and effort was invested into trying to figure out the reasons for it, looking at all steps from cell growth to PolyU-column material but to no avail. Finally the old, demanding method to produce NS3 had to be reinstated. A change in cell lysation method from sonication to a cell-disruptor resulted in a more efficient lysation with a higher reproducibility. By omitting a NaCl-wash during the IMAC purification, time was saved without any loss of yield or purity. This procedure was used for producing all enzyme for the studies presented in Paper I, II, III and IV. The amount of NS3 obtained using this method is enough for enzymatic studies, but in the lower range for what is needed for the direct-binding assay described in Paper VI. In an attempt to solve the problem once and for all, cloning of several NS3 gene constructs into insect cells using the Baculo-virus system was initialized. This procedure for producing NS3 has been described previously (Sali et al. 1998). Initial results were encouraging and lysates from infected cells showed clear NS3 activity. Moreover, both cells and lysates could be frozen without any significant enzyme inactivation. This work is still in progress but this system will hopefully enable the production of more enzyme. Access of sufficient and pure enzyme is fundamental for progress in this, and perhaps any, project. Currently many methodologies and experiments are out of reach simply because they require too much material.

NS3 protease inhibitor resistance (Paper I) The study of protease inhibitor resistance was initialized when the first re- ports of replicons resistant to protease inhibitors BILN 2061 and VX 950 were presented (Lin et al. 2004)(Lu et al. 2004). These studies showed that substitutions R155Q, A156T and D168V (Figure 9) were the culprits. Trun- cated protease variants of these enzymes had previously been analyzed and showed a high level of resistance towards the two compounds. The aim in our work was to look both at how these substitutions affected the characteristics of full-length NS3 and also how the inhibition by other types of inhibitors was affected by these residue changes.

34 Figure 9: Illustration of the residues involved in the resistance towards some NS3 protease inhibitors; Arg 155, Ala 156 and Asp 168 (purple), and the catalytic triad of NS3, Ser 139, Asp 81, and His 57 (white). The substitutions were introduced into the gene encoding NS3 using PCR. Expression and purification resulted in similar yields for both A156T and D168V enzymes as for the wild type, but a lower yield was obtained for R155Q. Analysis of the recombinant enzyme variants revealed that the A156T and D168V variants had the same catalytical efficiency as wild type enzyme, but R155Q had ten-fold lower efficiency. These results would indic- ate that the isolated R155Q variant is less viable but apparently it can sustain RNA replication since it emerged under the selection pressure of certain inhibitors. This mean that there has to be an even larger reduction in enzymatic activity before the variant is unfit or that other factors govern RNA replication inside cells. A small set of inhibitors used in our previous studies, representing a vari- ety of potencies, charge distribution and size as well as BILN 2061 and VX 950, were chosen for this study (Figure 10). Inhibition measurements revealed that these substitutions did indeed result in resistance. Up to 3,000 fold higher Ki-values were observed for some inhibitors. The most potent inhibitors were the ones who lost most of their effect whereas moderate or weak inhibitors were unaffected by these substitutions. Usage of vitality values was a good way to combine loss of enzymatic activity with gain in resistance, something clearly seen when looking at the R155Q variant. An interesting outcome of this study was that VX 950 was more potent towards the

35 O O OH OH

O O O O O O O O O O H H H H H H H N N N HO N N HO N N N N N OH N N OH N N OH H H H H H H H O O O O O O SH O O O O SH SH HO O O OH HO O HO 1. 2. 3.

O O O O O O H H H H N HO N N HO N N N N OH N N OH H H H H O O O O O O O SH SH

O HO 5. O O O O H H 4. O N N S O N N O H H O O

N N 6. O O O

N N H O H O N N O N O O O N O O N O O H OH H N S O O H O

7. 8. N H O N O O O O N O O H N S O H N N H N 9. O S

N H O H O N O O N N OH N O O H H H N H O N N O N O N O O

10. BILN 2061 11. VX 950 Figure 10: Structures of the different protease inhibitors used in Paper I and II. Their size and characteristics vary from very peptide like (1 to 5), to compounds with large P2 side chains (6 to 10), C-terminal carboxylic acid (1 to 4 and 8) or acyl sulf- onamide (6, 7 and 9) and various side chains in P1 up to P6. The compound which were used in the selection of the resistant replicons, BILN 2061 and VX 950, were also tested.

D168V enzyme variant than towards the wild type. This was seen in corresponding replicons but not when using truncated D168V NS3 protease, indicating that use of full-length NS3 gives a better correlation with replicon data. Virus with the A156T substitution is cross-resistant to BILN 2061, VX 950 and other protease inhibitors currently in clinical trials (Lin et al. 2004)(Tong et al. 2006). The variant has also been found in humans after mono-therapy, which means that the virus can become resistant to such protease inhibitor based drugs through this escape substitution (Sarrazin et al. 2007). Thus it was encouraging to see that the A156T substitution had no effect on some of the tested compounds. Even though these specific compounds are not very potent it should be possible to find protease inhibitors with a non-overlapping resistance profiles. Monitoring resistance is of outmost importance when dealing with antiviral drug discovery and these enzyme variants now constitute a tool which will be used in future resistance profiling of new NS3 protease inhibitors.

36 Helicase can influence protease inhibition (Paper II) Studies of HCV NS3 have predominantly been performed on truncated protease or helicase respectively, for reasons already described. Our studies have been performed on full-length NS3 despite all difficulties. Even though it has been shown that the protease domain influences the helicase domain in full-length NS3, the influence the helicase domain has on the protease domain has been less clear. The aim of this study was to investigate how small changes, substitutions of two amino acids in the helicase domain, could influence protease activity and inhibition. The targeted residues, Gln 526 and His 528 (Figure 11), were chosen since previous modeling studies had shown them to be potential interaction partners for protease inhibitors (Jo- hansson et al. 2002). By substituting His 528 with either alanine or serine, it was also possible to see how a different amino acid in the same position influenced the enzyme.

Figure 11: Illustration of the residues in the helicase domain, Arg 526 and His 528 (orange) believed to be involved in the binding of NS3 protease inhibitors, and the catalytic triad of NS3, Ser 139, Asp 81, and His 57 (white).

The same set of inhibitors as in Paper I (Figure 10) were chosen since they represent differences in their structure where the presumed interaction with the helicase takes place. Also, use of the same compounds enabled us to make direct comparisons between the enzyme variants from the two studies.

The substitutions in the helicase domain had no effect on the Km-values but resulted in a reduction in kcat. The overall effect catalytic efficiency was however small. The inhibition studies revealed that the helicase substitutions

37 did affect protease inhibition, but the effect was neither large, nor easy to in- terpret. The Q526A variant showed only minor differences as compared to the wild type enzyme, despite its proposed importance for inhibitor binding (Johansson et al. 2002). As for the H528 substitutions, the most interesting outcome was that BILN 2061 had a 10-fold difference in Ki between H528A and H528S. This shows that not only the position of the substitution, but also the nature of the introduced amino acid influenced the inhibitor potency. Furthermore, some inhibitors became more potent and, taken together, this study show that the helicase can influence protease inhibition. Thus, the use of full-length NS3 for finding protease inhibitors may be more appropriate than the use of truncated NS3 and the helicase can serve as an anchor point in future drug designs.

Novel NS3 inhibitor designs (Paper III) Investigating the ADME (adsorption distribution metabolism excretion) as- pects of a potential drug in the early stages of discovery is becoming more and more important. Peptide-based protease inhibitors are unfit as drugs for a number of reasons, one being that they are quickly degraded. Incoporating beta-amino acids in peptide-mimetic inhibitors has successfully been used in other studies where proteolytic degradation of the compound was reduced but potency retained (Aguilar et al. 2007). Here, beta-amino acids in different positions of a tripeptide inhibitor to NS3 was investigated. Unfortunately, the potency of these inhibitors was lower compared to their alpha counterparts. Substitutions at position P2 or P3 of the inhibitor was generally more tolerated whereas substitutions in position P1 resulted in a larger decrease in potency. Modeling studies revealed that the reduction in potency was mainly due to unfavorable hydrogen bond formations between the inhibitor backbone and residues Arg 155 and Ala 157 of NS3, and possibly a poorer interaction with the oxyanion hole. Results from these compounds and several others were used to make a predictive CoMFA (Compar- ative molecular field analysis) model which can be used to predict binding modes and potencies of future compounds.

Inhibitor activation of NS3 (Paper IV) Many compounds inhibiting NS3 have been synthesized and tested over the years. A puzzling phenomenon that we have been encountering from time to time was that some inhibitors activated the enzyme at low inhibitor concentrations (Figure 12).

38 Figure 12: Illustration of inhibitor activation. Normally the activity of the enzyme (A) decreases with increasing inhibitor concentration ([I]), as shown by the blue line. For inhibitor activation, an increase in activity at low inhibitor concentrations is observed, as shown by the red line. Disregarded as an obscure artifact, the activation remained uninvestig- ated. However, during the experiments leading to Papers I and II, some compounds produced a high level of activation and it was decided that the mechanism behind the activation should be elucidated once and for all. The reason for using a non-normalized equation for tight binding inhibition in Paper I and II, not commented at the time, was because of this activation effect. Early on in the study it was found that no trivial errors or non-enzymatic processes caused the activation. Activation was only seen when the inhibitor was incubated with NS3 and NS4A for some time prior to adding substrate. Interestingly, all inhibitors studied, except one (compound 8 in Figure 10), resulted in activation. The only non-activating inhibitor was somewhat shorter than the rest, but was otherwise very similar to activating compounds. Even VX 950, which is supposed to be a slow-binding, mechanism based inhibitor, showed signs of activation. This indicated that activation is a general phenomenon, not dependent on potency, structure or mechanism of the inhibitor. The forces behind the process were further investigated and it was revealed that the activation was reduced upon major changes in assay conditions (high ionic strength, low pH, no glycerol), but all at the expense of an overall reduction in NS3 activity. The exception was when changing deter- gent from n-octyl-glucopyranoside to CHAPS, which reduced the activation without affecting the enzyme activity. Different enzymatic explanations for the activation were sought. The most plausible explanation was concluded to be an enzyme memory process, where the inhibitor induces or stabilizes a more active conformation of the enzyme. After the inhibitor has dissociated, the enzyme slowly relaxes to a less active form (Ricard et al. 1974). All full-length NS3 variants previously studied in this thesis; R155Q, A156T, D168V, Q526A, H528A and H528S (Paper I and II) showed similar activation patterns as the wild type, indicating that these changes in the pro-

39 tease do not affect the activation. However, truncated NS3 protease was not activated by any inhibitor, indicating that the helicase domain is somehow involved in this memory process. This study indicates that full-length NS3 is able to adopt different con- formations, supporting the idea of a flexible enzyme with different catalytic properties. If and how this freedom of movement is affected when the enzyme is membrane associated remains to be investigated. Still, this study once more shows how the helicase domain can affect the protease and the importance of using full-length NS3 as a model enzyme.

Studies of interaction between NS5B and inhibitors (Paper V) NS5B is perhaps an even better drug target than NS3 because it can be inhibited both at the active site and at several allosteric sites. Despite its attract- iveness, the mechanisms behind its inhibition is still largely unknown. In order to study the detailed interaction kinetics between NS5B and five different classes of non-nucleoside inhibitors (Figure 13), a biosensor based assay was developed in this study. The binding of the inhibitors to NS5B was monitored using SPR, surface plasmon resonance, biosensor technology. When mass density changes close to a surface occurs, they give rise to a plasmon which in turn change the diffractive index of polarized light. This change can be monitored and visualized in a sensorgram (Figure 14). From these sensorgrams, rate constants from the association (kon) and the dissociation (koff) as well as the equilibrium dissociation constant (Kd) can be obtained. The interaction between the inhibitors and the allosteric sites of NS5B was weak, whereas the diketo acid, which interacts with the active site magnesium ions, showed a stronger interaction. This interaction was enhanced by including phosphate ions, indicating a phosphate ion mediated interaction. Including magnesium ions in the buffer resulted in a weaker interaction between the diketo acid and NS5B, which was to be expected given the metal ion chelating property of the inhibitor. The diketo acid also showed a different binding stoichiometry compared to the other compounds, where two inhibitors bound to one enzyme molecule which could correspond to one inhibitor per magnesium ion in the active site. Despite the fact that the inhibitors interacted with different binding sites on NS5B, the only cooperativity observed was between the diketo acid and the other compounds and not between compounds binding to the different allosteric sites.

40 Signal (RU) koff K=d kon O H O N Association HO Dissociation N F HO N O N O S O Injection O H Benzofuran, HCV 796 N start Injection O O S stop OH Cl Time (s) Benzimidazole, JT-16 N O Figure 14: A schematic SPR bio- O O S sensor sensorgram. Upon injection OH HN Thiophene carboxylic acid N of a ligand to the sensor, an in-

N O Cl creased signal is observed. From N O OH this, the association rate constant, Benzothiadiazine O OH Cl kon, can be derived. After the end of O Diketoacid the injection, the ligand leaves the Figure 13: Structure of the different non-nuc- surface and the rate of the dissoci- leoside polymerase inhibitors used in the bio- ation, koff, can be determined. Using sensor study of NS5B. these rate constants, the equilibrium dissociation constant, KD, can easily be determined. The presence of RNA reduced the binding of all the inhibitors to NS5B, suggesting that RNA can prevent inhibitor binding, and have a direct effect on the binding affinity. Taken together, this study provided detailed information of NS5B inhibitor binding and of the factors that affect the affinities. These results and the use of a biosensor based method will be valuable for finding and optimizing NS5B inhibitors.

Characterization of NS3 inhibitors using a biosensor (Paper VI) Even though measurements of enzyme inhibition are crucial for drug discovery, the information content is limited. Moreover, as illustrated in Paper IV, the activity assay for full-length NS3 can give rise to an activation phenom- ena resulting in difficulties when interpreting data due to the system complexity. A complementary method is a direct binding assay, for example using biosensor technology. By omitting the substrate, the system is simpler and detailed information of the interaction can be obtained.

41 O O O N N N

O O O

N H O N O N H O O N O N H N O O O O O O N O N O H N S O O H O H N S N O O H O H OH

1. 2. 3.

O O O

N N N

O O O

N H N O H O N H O N O N O O N O N O O N O O O H OH H N CN N O O O O O H N S H H

4. 5. 6.

N H N N O S

H O N N N O O H O OH N H H H O N N N N H O N O O O O O BILN 2061 VX 950 Figure 15: Inhibitors used in the biosensor study with full-length NS3.

In this study, full-length NS3 was immobilized and its interaction with a series of inhibitors was tested (Figure 15). Since we express and purify NS3 without NS4A, and the binding assay is not dependent on a catalytically active enzyme, inhibitor interaction with NS3 could be analyzed in the absence of NS4A. By using three surfaces; one with NS3 alone, NS3 with NS4A, and NS4A alone, the kinetics of the interaction between NS3 and protease inhibitors could be determined. NS3 showed a strong interaction with NS4A, with a Kd of 30 nM, but the affinity was reduced more than 600-fold by increasing the ionic strength. The protease inhibitors all had a higher affinity, up to 40- fold (Figure 16), for NS3 co-immobilized with NS4A than for NS3 alone. However, a surface with only NS3 was almost as good, indicating that NS4A can be excluded which further reduces the system complexity.

42 20 20

15 15

10 10

5 5 Response (RU) Response (RU)

0 0

-5 -5 -50 0 50 100 150 -50 0 50 100 150 Time (s) Time (s) Figure 16: Comparison of the interaction of an inhibitor, compound 3 from figure 15, interacts with a surface with only NS3 (left) and NS3 co-immobilized with NS4A (right). The surface with both NS3 and NS4A also have free NS3. As a consequence, two Kd values are obtained from the right sensorgrams, one exactly the same as in the left and one 40-times lower, indicating that NS4A improves the affinity for this compound fifty times.

The binding of VX 950 was monitored, revealing a bi-phasic dissociation phase which is symptomatic with a slow-binding inhibitor (Figure 17), all in agreement with the proposed slow-binding mechanism for this compound. This shows that not only the detailed binding of inhibitors to NS3 but specific binding mechanisms could be visualized. Comparing interaction kinetic data with inhibition data (Ki) or replicon data (EC50) revealed a good correlation. Thus the data obtained using a biosensor assay provide information not easily obtained otherwise and is a good complement to inhibition measurements for understanding details in binding kinetics. These results provide a better understanding of this enzyme which will aid in the development of more potent protease inhibitors. 25

Response (RU) 5

-5 -200 0 200 400 600 Time (s) Figure 17: Sensorgrams from injecting VX 950 for different time periods. The dissociation is affected by increased injection times, leading to a base-line shift (double- headed arrow). This is in agreement with a time-dependent binding, the proposed mechanism for this inhibitor.

43 Perspectives

Summary The results presented herein can be divided into three parts: Firstly, a general understanding and information of the dynamics of full- length NS3 have been obtained. Modifications in the helicase domain had an effect on protease inhibition and it was shown that protease inhibitors could induce an activation of the full-length enzyme, most likely mediated through the helicase domain. These results suggest the usage of full-length NS3 for enzymatic studies. Secondly, this work have provided new tools with which to study both NS3 and NS5B. The development of the biosensor assays are great tools for the further studies of these enzymes. Furthermore, with the construction of variants of full-length NS3 which carry amino acid substitutions, detailed studies of how these changes affect enzyme activity and inhibition can be performed. These variants can be used both for mechanistic studies of NS3 and for resistance profiling of future inhibitors. Thirdly, these results are of use from a drug discovery point of view. The detailed binding kinetics determined for several inhibitors targeting these enzymes can be used to aid in the design of better inhibitors. Also, the resistance profiling of NS3 inhibitors revealed that multiple cross-resistance can possibly be avoided. Taken together, this work has provided much information on NS3 and NS5B, with both implications and applications for drug discovery.

The future The work of a scientist is never over, many new questions arises for every answer provided. During the time I have been working with HCV, there have been major progresses and setbacks, both within my project and in the field in general. The future will be interesting where the first direct anti-HCV drug will probably reach the market soon. Our studies of HCV enzymes will continue and the work presented here have provided both new results and tools to use. With the possibility to perform resistance profiling, and measure detailed binding kinetics, of inhibitors the future of the project looks prom-

44 ising. There are some things that should be aimed for in the extension of these projects:

Improve the expression and purification of full-length NS3. This point cannot be stressed enough since many experiments and methods are currently out of reach because of the amount of enzyme they require. Develop an activity assay for full-length NS3 associated with full- length NS4A and embedded in a membrane of correct composition. This would enable the investigation of how membrane association affects the enzymatic activity. Expand the biosensor assays to comprise full-length NS5B, full-length NS3 and full-length NS4A inserted into a reconstituted membrane of correct composition on the sensor surface. The ambition should be to build up the entire HCV replicon on a chip, enabling advanced interaction studies. Update the collection of enzyme variants at our disposal when new resistance information becomes available. This will enable the use of correct references for resistance profiling of new inhibitors.

45 Sammanfattning

Hepatit C Hepatit C är en leverinflammation som orsakas av hepatit C viruset (HCV), ett virus som sprids genom blodkontakt. Det finns för närvarande inget vaccin mot detta virus och de läkemedel som används idag, interferon och ribavirin, botar bara ungefär 50% av patienterna. Dessa läkemedel har dessutom många obehagliga biverkningar, såsom psykoser, blodbrist och influensaliknande symptom. Man måste dessutom ta dem under lång tid, upp till 1,5 år, för att bli botad. Med nya och bättre läkemedel mot hepatit C skulle man kunna bota fler, och minska lidandet hos drabbade patienter.

Proteiner, enzymer och läkemedelsutveckling Proteiner består av långa kedjor av aminosyror. Det finns 20 naturligt förekommande aminosyror och proteiner kan bestå av upp till flera tusen aminosyror. Alla proteiner har en unik sekvens av aminosyror som är definierad av den gen som kodar för proteinet och detta gör att proteiner har en stor variation både vad gäller storlek och funktion. Vissa proteiner är enzymer, en sorts molekylära maskiner som kan påskynda (katalysera) kemiska reaktioner. De nödvändiga för det mesta som sker i våra celler. Enzymer delas upp i grupper beroende på vilken typ av reaktion de katalyserar. Läkemedel är ofta inhibitorer, molekyler som förhindrar enzymer från att arbeta. En utmaning när man utvecklar läkemedel är att hitta inhibitorer som inte påverkar de mänskliga enzymerna utan bara enzymerna hos det som orsakar sjukdomen. Detta gör att man måste testa inhibitorerna, först genom att se hur de påverkar enskilda enzymer för att sedan undersöka om de har effekt i cellkultur eller försökdjur, system som mer liknar människan. Alla dessa försök gör att den genomsnittliga tiden för att utveckla ett läkemedel är 15 år. Dessutom måste man följa upp läkemedlet även efter det att man börjat använda det för att behandla patienter. Dels för att det kan uppkomma biverkningar först efter lång tid av användning, något som inte alltid kan ses under utvecklingsfasen. Dels för att om läkemedlet är riktat mot bakterier eller virus så kan dessa utveckla resistens vilket gör läkemedlet verkningslöst på sikt.

46 Att slå ut hepatit C viruset Hos HCV finns det två enzymer som man vet är bra mål för läkemedel; ett polymeras som har beteckningen NS5B, och ett proteas som har beteckningen NS3. NS är en förkortning för “non-structural protein”, det vill säga icke-strukturellt protein. Enzymgruppen polymeraser är ansvarig för att kopiera organismers arvsmassa, till exempel RNA eller DNA. Enzymgruppen proteaser är ansvarig för att klippa sönder andra proteiner vilket bland annat är viktigt för att bryta ner protein i mat och för att städa upp inuti celler. Genom att slå ut dessa två enzymer med inhibitorer hoppas man kunna bota HCV. För närvarande finns det ett antal olika inhibitorer mot dessa enzym som verkar lovande och dessa håller nu på att testas i människor. Dock har man redan sett begynnande läkemedelsresistens och det står klart att mycket arbete återstår innan man helt kan bota HCV.

Modellsystem För att kunna hitta inhibitorer mot ett enzym måste man studera enzymet i detalj. Tyvärr är det nästintill omöjligt att studera enzymer där de finns naturligt; de är till exempel för små för att synas i ett mikroskop. Istället måste man använda sig av olika modellsystem och mätmetoder. Man kan använda bakterien E. coli för att tillverka nästan vilket enzym som helst. Det är bra eftersom dessa bakterier är lätta att odla vilket gör att man, teoretiskt sett, kan få fram mycket av sitt enzym. Efter att bakterierna producerat enzymet slår man sönder bakterierna och renar fram enzymet. Man kan sedan studera enzymet isolerat så att man vet exakt vad det är man tittar på. En teknik går ut på att man mäter hur snabbt enzymet arbetar, man bestämmer hur aktivt det är. Om man då tillsätter en inhibitor kan man se hur mycket enzymet hindras och få fram information om hur bra inhibitorn är. Med denna information kan man sedan gå vidare och till exempel försöka göra hämmaren ännu bättre eller göra så att den lättare tas upp av kroppen. En annan teknik för att studera hur bra en inhibitor är att använda en biosensor. Med denna teknik tittar man inte på hur enzymet arbetar utan på hur molekyler binder till enzymet. Man fäster enzymet på en yta och låter sedan de molekyler, vars bindning till enzymet man vill studera, flöda över denna yta. Om molekylerna binder så får man en signal som säger hur snabbt det gick. När man sedan stoppar flödet av molekyler kan man se hur snabbt molekylerna lossnar igen. För att en inhibitor skall vara bra måste den binda hårt till enzymet och med en biosensor kan man få fram sådan information som kan vara till stor hjälp när man försöker göra förbättringar i inhibitorn. Man kan dessutom använda en biosensor för att titta på inbindningen av andra molekyler, såsom RNA eller andra proteiner.

47 Min forskning Jag har studerat de två enzymerna NS3 och NS5B från HCV. Man hade tidigare sett att resistens mot NS3-inhibitorer berodde på att vissa aminosyror i NS3 blivit utbytta, men man visste inte så mycket om hur dessa förändringar i NS3 påverkade andra inhibitorer. Genom att införa de här förändringarna i den variant av NS3 vi arbetar med kunde jag studera just detta (Artikel I). Vad jag såg var att dessa förändringar gjorde många inhibitorer mycket sämre, men det fanns vissa inhibitorer som inte påverkades. Således finns det hopp om att kunna hitta inhibitorer som viruset får svårt att utveckla resistens mot. NS3 enzymet består av två delar (domäner), ett proteas och ett helikas. Helikaser är ytterligare en grupp av enzymer som har till uppgift att separera dubbelsträngad arvsmassa, som DNA eller RNA. Jag har inte studerat helikaset som sådant, men har ändå haft det med eftersom vi misstänkt att det är av vikt även för proteaset. Trots detta väljer många att bara studera den domän av NS3 som de är intresserade av. För att se om helikaset kunde påverka proteaset testade jag att byta ut två aminosyror i helikaset och mätte sedan på proteaset för att se om det gav någon skillnad (Artikel II). Jag såg små skillnader, men ändå så pass stora att man kunde sluta sig till att det är bättre att använda hela enzymet, även om man bara är intresserad av en av domänerna, när man studerar NS3. En intressant sak som jag upptäckte när jag testade några av inhibitorerna var att de märkligt nog inte bara hämmade enzymet utan även aktiverade det (Artikel IV). Detta är anmärkningsvärt eftersom det är precis tvärtemot vad man förväntar sig. När jag studerade detta vidare visade det sig att det bara var när man lät enzymet och inhibitorn befinna sig tillsammans under en längre tid som man fick en aktivering och det var dessutom bara vid väldigt låga koncentrationer av inhibitorn. Den slutsats jag drog var att det måste handla om en form av “enzymminne”. Vid låga koncentrationer när ”inhibitorn” binder till NS3 så ändras NS3 form och denna nya form av enzymet har en högre aktivitet. När ”inhibitorn” sedan lossnar så tar det ett tag för enzymet att återfå sin gamla, mindre aktiva form; enzymet “minns” vad det senast var i kontakt med och man får en aktiveringseffekt. När jag undersökte NS3 utan helikas såg jag ingen aktivering, vilket betyder att helikaset är involverad i denna process. Detta är ytterligare ett exempel på att de två domänerna hos NS3 påverkar varandra och att man bör använda hela enzymer när man studerar det. När man försöker utveckla läkemedel får man inte glömma bort att de ska fungera inuti en människa. De flesta inhibitorer som finns mot NS3 liknar väldigt korta proteiner, så kallade peptider. Detta gör att de inte fungerar som läkemedel då de lätt bryts ner av de proteaser som finns i matsmältnings- systemet. Ett sätt att göra inhibitorer stabilare är att bygga upp dem med modifierade aminosyror istället för vanliga aminosyror (Artikel III). Man

48 har tidigare sett att det går att göra detta och få inhibitorerna mycket mer stabila utan att de blir sämre. Tyvärr visade det sig att de inhibitorer som testades blev mycket sämre på att inhibera NS3 än motsvarande inhibitorer gjorda av vanliga aminosyror. Den här studien visar ändå på att det kan vara bra att ta hänsyn till att läkemedlet ska fungera i en människa, tidigt i läkemedelsutvecklingen. Samtliga studier som beskrivits ovan gjordes genom att studera enzymets aktivitet, men den information man får från sådana experiment är ibland inte tillräcklig. Därför utvecklade vi en biosensorbaserad mätmetod för NS3 och NS5B där man i detalj kan studera hur olika molekyler binder till dessa enzymer (Artikel V och VI). Vi kunde med denna metod mäta bindningen av inhibitorer till enzymerna men också hur andra molekyler, som till exempel RNA, band in. Med hjälp av den detaljerade information man får fram går det att se hur bindningen påverkas av faktorer såsom salthalt och pH. För NS3 gjorde vi en jämförelse mellan hur hårt inhibitorerna band på biosensorn och hur mycket de inhiberade enzymet. Vi fick fram att de två metoderna stämde mycket väl överens och att biosensorn därmed är ett bra hjälpmedel för studier av NS3. Biosensormetoderna kommer troligtvis att få stor användning i framtiden och underlätta sökandet efter nya läkemedel.

Sammanfattningsvis kan man säga att min forskning bidragit till ny och viktig information om de studerade enzymerna från HCV. Detta, i kombination med utvecklandet av ny metodik, kommer att underlätta framtida studier och förhoppningsvis bidra till utvecklingen av nya, bra läkemedel mot Hepatit C.

49 Acknowledgements

Helena Danielson: Thank you for being a great supervisor and inspirator. It has been a priv- ilege.

Gun Stenberg: Thank you for being an excellent co-supervisor and a constant help with my cloning issues.

My co-workers at Organic Pharmaceutical Chemistry: Anja, Eva, Pernilla, Johanna, Shane and Anna, thank you for a fruitful collaboration!

My past and present co-workers in the lab: Cynthia, thank you for all encouragement. Maria, thank you for always providing an optimistic atmosphere. Anton, thank you for introducing me to the NS3 project. Thomas, thank you for sharing the burden of working with full-length NS3. Dan, thank you for being a good friend and discussion partner. Matthis, thank you for the great collaboration when developing the biosensor assays for NS3 and NS5B. Omar, thank you for all insights concerning theoretical enzymology and experimental design. Malin, thank you for being a great person and bundle of energy. Helena N, thank you for all help during my first dabbling biosensor experiments. Angelica, thank you proceeding with the NS3 project after me. Sofia, thank you for allowing yourself to get involved in the HCV field despite its difficulties. Johan, thank you for continuing the work with NS5B. Tony, thank you for our discussions on biosensor membranes.

Past and present colleagues at the Biochemistry section of the department: Thank you all for providing a pleasant and stimulating environment.

50 New colleagues at the Organic chemistry section of the department: I don't know half of you half as much as I would like, but I like all of you as much as you deserve!

My families: Rebecca, I love you and you know it! Margaretha and Ingvar, thanks for the genes, environment and love making all this possible. Magnus, Håkan, Christina for all the fun times we shared and will share.

Relatives and friends: Thank you for showing me the world outside the lab. Special thanks to my old role-playing buddies Jon, Henrik and Peter as well as my WoW- buddies Andreas and Alexander.

51 References

Abuchowski A, McCoy JR, Palczuk NC, van Es T, Davis FF. Effect of cova- lent at- tachment of polyethylene glycol on immunogenicity and circulating life of bovine liver catalase. J Biol Chem. 1977 Jun 10;252(11):3582-6. Adachi T, Ago H, Habuka N, Okuda K, Komatsu M, Ikeda S, Yatsunami K. The essential role of C-terminal residues in regulating the activity of hepatitis C virus RNA-dependent RNA polymerase. Biochim Biophys Acta. 2002 Nov 19;1601(1):38-48. Agnello V, Abel G, Elfahal M, Knight GB, Zhang QX. Hepatitis C virus and other flaviviridae viruses enter cells via low density lipoprotein receptor. Proc Natl Acad Sci U S A. 1999 Oct 26;96(22):12766-71. Aguilar MI, Purcell AW, Devi R, Lew R, Rossjohn J, Smith AI, Perlmutter P. Beta- amino acid-containing hybrid peptides--new opportunities in peptidomimetics. Org Biomol Chem. 2007 Sep 21;5(18):2884-90. Arrow Therapeutics product pipeline. http://www.arrowt.co.uk/product-hcv.asp. 2008. Bartenschlager R, Ahlborn-Laake L, Mous J, Jacobsen H. Kinetic and structural analyses of hepatitis C virus polyprotein processing. J Virol. 1994 Aug;68(8):5045- 55. Bartosch B, Dubuisson J, Cosset FL. Infectious hepatitis C virus pseudo-particles containing functional E1-E2 envelope protein complexes. J Exp Med. 2003 Mar 3;197(5):633-42. Behrens SE, Tomei L, De Francesco R. Identification and properties of the RNA-dependent RNA polymerase of hepatitis C virus. EMBO J. 1996 Jan 2;15(1):12-22. Beran RK, Serebrov V, Pyle AM. The serine protease domain of hepatitis C viral NS3 activates RNA helicase activity by promoting the binding of RNA substrate. J Biol Chem. 2007 Nov 30;282(48):34913-20. Beran RK, Pyle AM. Hepatitis C viral NS3-4A protease activity is enhanced by the NS3 helicase. J Biol Chem. 2008 Oct 31;283(44):29929-37. Blanchard E, Belouzard S, Goueslain L, Wakita T, Dubuisson J, Wychowski C, Rouillé Y. Hepatitis C virus entry depends on clathrin-mediated endocytosis. J Virol. 2006 Jul;80(14):6964-72. Blight KJ, McKeating JA, Marcotrigiano J, Rice CM. Efficient replication of hepatitis C virus genotype 1a RNAs in cell culture. J Virol. 2003 Mar;77(5):3181-90. Brass V, Berke JM, Montserret R, Blum HE, Penin F, Moradpour D. Structural determinants for membrane association and dynamic organization of the hepatitis C virus NS3-4A complex. Proc Natl Acad Sci U S A. 2008 Sep 23;105(38):14545-50. Bukh J, Miller RH, Purcell RH. Genetic heterogeneity of hepatitis C virus: quasispe- cies and genotypes. Semin Liver Dis. 1995 Feb;15(1):41-63.

52 Bukh J, Pietschmann T, Lohmann V, Krieger N, Faulk K, Engle RE, Govindarajan S, Shapiro M, St Claire M, Bartenschlager R. Mutations that permit efficient replication of hepatitis C virus RNA in Huh-7 cells prevent productive replication in chimpanzees. Proc Natl Acad Sci U S A. 2002 Oct 29;99(22):14416-21. Carrère-Kremer S, Montpellier-Pala C, Cocquerel L, Wychowski C, Penin F, Dubuisson J. Subcellular localization and topology of the p7 polypeptide of hepatitis C virus. J Virol. 2002 Apr;76(8):3720-30. Carroll SS, Tomassini JE, Bosserman M, Getty K, Stahlhut MW, Eldrup AB, Bhat B, Hall D, Simcoe AL, LaFemina R, Rutkowski CA, Wolanski B, Yang Z, Migliac- cio G, De Francesco R, Kuo LC, MacCoss M, Olsen DB. Inhibition of hepatitis C virus RNA replication by 2'-modified nucleoside analogs. J Biol Chem. 2003 Apr 4;278(14):11979-84. Chan L, Das SK, Reddy TJ, Poisson C, Proulx M, Pereira O, Courchesne M, Roy C, Wang W, Siddiqui A, Yannopoulos CG, Nguyen-Ba N, Labrecque D, Bethell R, Hamel M, Courtemanche-Asselin P, L'Heureux L, David M, Nicolas O, Brunette S, Bilimoria D, Bédard J. Discovery of thiophene-2- carboxylic acids as potent inhibitors of HCV NS5B polymerase and HCV subgenomic RNA replication. Part 1: Sulfonamides. Bioorg Med Chem Lett. 2004 Feb 9;14(3):793-6. Chan YP, Meyrueix R, Kravtzoff R, Nicolas F, Lundstrom K. Review on Medusa: a polymer-based sustained release technology for protein and peptide drugs. Ex- pert Opin Drug Deliv. 2007 Jul;4(4):441-51. Choo QL, Kuo G, Weiner AJ, Overby LR, Bradley DW, Houghton M. Isolation of a cDNA clone derived from a blood-borne non-A, non-B viral hepatitis genome. Science. 1989 Apr 21;244(4902):359-62. Coelmont L, Paeshuyse J, Windisch MP, De Clercq E, Bartenschlager R, Neyts J. Ribavirin antagonizes the in vitro anti-hepatitis C virus activity of 2'-C-methylcytidine, the active component of valopicitabine. Antimi- crob Agents Chemoth- er. 2006 Oct;50(10):3444-6. Dahl, O. Chemistry of animal depot fats in view of recent investigations, Svensk Kemisk Tidskrift. 1958 May;70(2): 44-54. De Leede LG, Humphries JE, Bechet AC, Van Hoogdalem EJ, Verrijk R, Spencer DG. Novel controlled-release Lemna-derived IFN-alpha2b (Loc- teron): pharmacokinetics, pharmacodynamics, and tolerability in a phase I clinical trial. J In- terferon Cytokine Res. 2008 Feb;28(2):113-22. De Lucas S, Bartolome J, Carreno V. Hepatitis C virus core protein down-regulates transcription of interferon-induced antiviral genes. J Infect Dis. 2005 Jan 1;191(1):93-9. Dhanak D, Duffy KJ, Johnston VK, Lin-Goerke J, Darcy M, Shaw AN, Gu B, Sil- verman C, Gates AT, Nonnemacher MR, Earnshaw DL, Casper DJ, Kaura A, Baker A, Greenwood C, Gutshall LL, Maley D, DelVecchio A, Macarron R, Hofmann GA, Alnoah Z, Cheng HY, Chan G, Khandekar S, Keenan RM, Sar- isky RT. Identification and biological characterization of heterocyclic inhibitors of the hepatitis C virus RNA-dependent RNA polymerase. J Biol Chem. 2002 Oct 11;277(41):38322-7. Egger D, Wölk B, Gosert R, Bianchi L, Blum HE, Moradpour D, Bienz K. Expres- sion of hepatitis C virus proteins induces distinct membrane alterations including a candidate viral replication complex. J Virol. 2002 Jun;76(12):5974-84. Enomoto N, Sakuma I, Asahina Y, Kurosaki M, Murakami T, Yamamoto C, Izumi N, Marumo F, Sato C. Comparison of full-length sequences of interferon-sensitive and resistant hepatitis C virus 1b. Sensitivity to inter feron is conferred by amino acid substitutions in the NS5A region. J Clin Invest. 1995 Jul;96(1):224-30.

53 Evans MJ, von Hahn T, Tscherne DM, Syder AJ, Panis M, Wölk B, Hatzi- ioannou T, McKeating JA, Bieniasz PD, Rice CM. Claudin-1 is a hepatitis C virus co-receptor required for a late step in entry. Nature. 2007 Apr 12;446(7137):801-5. Ferrari E, Wright-Minogue J, Fang JW, Baroudy BM, Lau JY, Hong Z. Characteriza- tion of soluble hepatitis C virus RNA-dependent RNA polymerase expressed in Escherichia coli. J Virol. 1999 Feb;73(2):1649-54. Flint M, Mullen S, Deatly AM, Chen W, Miller LZ, Ralston R, Broom C, Emini E, Howe AY. Selection and Characterization of Hepatitis C Virus Replicons Dually Resistant to the Polymerase and Protease Inhibitors, HCV-796 and Boceprevir (SCH 503034). Antimicrob Agents Chemother. 2008 Oct 20. [Epub ahead of print]. François C, Duverlie G, Rebouillat D, Khorsi H, Castelain S, Blum HE, Gatignol A, Wychowski C, Moradpour D, Meurs EF. Expression of hepatitis C virus proteins interferes with the antiviral action of interferon in- dependently of PKR-mediated control of protein synthesis. J Virol. 2000 Jun;74(12):5587-96. Frick DN, Rypma RS, Lam AM, Gu B. The non-structural protein 3 protease/ helicase requires an intact protease domain to unwind duplex RNA efficiently. J Biol Chem. 2004 Jan 9;279(2):1269-80. Fried MW, Shiffman ML, Reddy KR, Smith C, Marinos G, Gonçales FL Jr, Häus- singer D, Diago M, Carosi G, Dhumeaux D, Craxi A, Lin A, Hoffman J, Yu J. Peginterferon alfa-2a plus ribavirin for chronic hepatitis C virus infection. N Engl J Med. 2002 Sep 26;347(13):975-82. Gale M Jr, Blakely CM, Kwieciszewski B, Tan SL, Dossett M, Tang NM, Korth MJ, Polyak SJ, Gretch DR, Katze MG. Control of PKR protein kinase by hepatitis C virus nonstructural 5A protein: molecular mechanisms of kinase regulation. Mol Cell Biol. 1998 Sep;18(9):5208-18. Gallinari P, Brennan D, Nardi C, Brunetti M, Tomei L, Steinkühler C, De Francesco R. Multiple enzymatic activities associated with recombinant NS3 protein of hepatitis C virus. J Virol. 1998 Aug;72(8):6758-69. Georgopapadakou N. Discontinued drugs in 2005: anti-infectives. Expert Opin In- vestig Drugs. 2007 Jan;16(1):1-10. Goffard A, Dubuisson J. Glycosylation of hepatitis C virus envelope proteins. Biochimie. 2003 Mar-Apr;85(3-4):295-301. Gopalsamy A, Aplasca A, Ciszewski G, Park K, Ellingboe JW, Orlowski M, Feld B, Howe AY. Design and synthesis of 3,4-dihydro-1H-[1]-benzothieno[2,3-c]pyran and 3,4-dihydro-1H-pyrano[3,4-b]benzofuran derivatives as non-nucleoside inhibitors of HCV NS5B RNA dependent RNA polymerase. Bioorg Med Chem Lett. 2006 Jan 15;16(2):457-60. Gosert R, Egger D, Lohmann V, Bartenschlager R, Blum HE, Bienz K, Moradpour D. Identification of the hepatitis C virus RNA replication complex in Huh-7 cells harboring subgenomic replicons. J Virol. 2003 May;77(9):5487-92. Grakoui A, McCourt DW, Wychowski C, Feinstone SM, Rice CM. Characterization of the hepatitis C virus-encoded serine proteinase: determination of proteinase- dependent polyprotein cleavage sites. J Virol. 1993 May;67(5):2832-43. Grakoui A, McCourt DW, Wychowski C, Feinstone SM, Rice CM. A second hepatitis C virus-encoded proteinase. Proc Natl Acad Sci U S A. 1993 Nov 15;90(22):10583-7. Griffin SD, Beales LP, Clarke DS, Worsfold O, Evans SD, Jaeger J, Harris MP, Row- lands DJ. The p7 protein of hepatitis C virus forms an ion channel that is blocked by the antiviral drug, Amantadine. FEBS Lett. 2003 Jan 30;535(1-3):34-8.

54 Gu B, Pruss CM, Gates AT, Khandekar SS. The RNA-unwinding activity of hepatitis C virus non-structural protein 3 (NS3) is positively modulated by its protease domain. Protein Pept Lett. 2005 May;12(4):315-21. Gwack Y, Kim DW, Han JH, Choe J. Characterization of RNA binding activity and RNA helicase activity of the hepatitis C virus NS3 protein. Bio chem Biophys Res Commun. 1996 Aug 14;225(2):654-9. Hahm B, Han DS, Back SH, Song OK, Cho MJ, Kim CJ, Shimotohno K, Jang SK. NS3-4A of hepatitis C virus is a chymotrypsin-like protease. J Virol. 1995 Apr;69(4):2534-9. HCV-796 clinical trials. A Study Evaluating the Safety and Clinical Activity of HCV-796 in Treatment-Naive and Non-Responder Subjects. http://clinicaltri- als.gov/ct2/show/NCT00367887. 2009 Jan. He Y, King MS, Kempf DJ, Lu L, Lim HB, Krishnan P, Kati W, Middleton T, Molla A. Relative replication capacity and selective advantage profiles of protease inhibitor-resistant hepatitis C virus (HCV) NS3 protease mu tants in the HCV genotype 1b replicon system. Antimicrob Agents Chemother. 2008 Mar;52(3):1101- 10. Hijikata M, Kato N, Ootsuyama Y, Nakagawa M, Shimotohno K. Gene mapping of the putative structural region of the hepatitis C virus genome by in vitro processing analysis. Proc Natl Acad Sci U S A. 1991 Jul 1;88(13):5547-51. Hijikata M, Kato N, Ootsuyama Y, Nakagawa M, Ohkoshi S, Shimotohno K. Hyper- variable regions in the putative glycoprotein of hepatitis C virus. Biochem Bio- phys Res Commun. 1991 Feb 28;175(1):220-8. Hoofnagle JH. Hepatitis C: The Clinical Spectrum of Disease. Hepatology. 1997 Sep;26(3 Suppl 1):15S-20S. Howe AY, Chase R, Taremi SS, Risano C, Beyer B, Malcolm B, Lau JY. A novel recombinant single-chain hepatitis C virus NS3-NS4A protein with Improved helicase activity. Protein Sci. 1999 Jun;8(6):1332-41. Howe AY, Cheng H, Johann S, Mullen S, Chunduru SK, Young DC, Bard J, Chopra R, Krishnamurthy G, Mansour T, O'Connell J. Molecular mechanism of hepatitis C virus replicon variants with reduced susceptibility to a benzofuran inhibitor, HCV-796. Antimicrob Agents Chemother. 2008 Sep;52(9):3327-38. Howe AY, Cheng H, Thompson I, Chunduru SK, Herrmann S, O'Connell J, Agarwal A, Chopra R, Del Vecchio AM. Molecular mechanism of a thumb domain hepatitis C virus nonnucleoside RNA-dependent RNA polymerase inhibitor. Antimi- crob Agents Chemother. 2006 Dec;50(12):4103-13. Huang, M., Y. Sun, W. Yang, X. Hou, J. Fabrycki, X. Nie, A. Sanchez, Y. Zhao, A. Phadke, M. Deshpande. ACH-806: a potent inhibitor of HCV replication with a novel mechanism of action. J. Hepatol. 2007;46:S221. Huang Y, Ostrowitzki S, Hill G, Navarro M, Berger N, Kopeck P, Mau CI, Alfredson T, Lal R. Single- and multiple-dose pharmacokinetics of levovirin valinate hy- drochloride (R1518) in healthy volunteers. J Clin Phar- macol. 2005 May;45(5):578-88. Hügle T, Fehrmann F, Bieck E, Kohara M, Kräusslich HG, Rice CM, Blum HE, Moradpour D. Hügle T, Fehrmann F, Bieck E, Kohara M, Kräusslich HG, Rice CM, Blum HE, Moradpour D. The hepatitis C virus nonstructural protein 4B is an integral endoplasmic reticulum membrane protein. Virology. 2001 May 25;284(1):70-81. Hwang SB, Lo SY, Ou JH, Lai MM. Detection of Cellular Proteins and Viral Core Protein Interacting with the 5' Untranslated Region of Hepatitis C Virus RNA. J Biomed Sci. 1995 Aug;2(3):227-236.

55 Idenix product pipeline. http://ir.idenix.com/phoenix.zhtml? c=131556&p=irol-pro- ductpipeline. 2009 Jan. Ingallinella P, Altamura S, Bianchi E, Taliani M, Ingenito R, Cortese R, De Francesco R, Steinkühler C, Pessi A. Potent peptide inhibitors of human hepatitis C virus NS3 protease are obtained by optimizing the cleavage products. Bio- chemistry. 1998 Jun 23;37(25):8906-14. Intermune press release. http://phx.corporate-ir.net/phoenix.zhtml? c=100067&p=irol-newsArticle&ID=1243000&highlight. Jan 2009. Isaacs A, Lindenmann J. Virus interference. I. The interferon. Proc. R. Soc. Lond., B, Biol. Sci. 1957 Sep;147 (927): 258–67. Jennings TA, Mackintosh SG, Harrison MK, Sikora D, Sikora B, Tackett AJ, Camer- on CE, Raney KD. NS3 helicase from the hepatitis C Virus can function as a monomer or oligomer depending on enzyme and substrate concentrations. J Biol Chem. 2008 Dec 16. [Epub ahead of print] Johansson A, Hubatsch I, Akerblom E, Lindeberg G, Winiwarter S, Daniel son UH, Hallberg A. Inhibition of hepatitis C virus NS3 protease activity by product- based peptides is dependent on helicase domain. Bioorg Med Chem Lett. 2001 Jan 22;11(2):203-6. Johansson A, Poliakov A, Akerblom E, Lindeberg G, Winiwarter S, Samuelsson B, Danielson UH, Hallberg A. Tetrapeptides as potent protease inhibitors of Hepat- itis C Virus full-length NS3 (protease-helicase/ NTPase). Bioorg Med Chem. 2002 Dec;10(12):3915-22. Kaneko T, Tanji Y, Satoh S, Hijikata M, Asabe S, Kimura K, Shimotohno K. Produc- tion of two phosphoproteins from the NS5A region of the hepatitis C viral genome. Biochem Biophys Res Commun. 1994 Nov 30;205(1):320-6. Kato N, Ikeda M, Mizutani T, Sugiyama K, Noguchi M, Hirohashi S, Shimotohno K. Replication of hepatitis C virus in cultured non-neoplastic human hepatocytes. Jpn J Cancer Res. 1996 Aug;87(8):787-92. Kato T, Date T, Miyamoto M, Furusaka A, Tokushige K, Mizokami M, Wakita T. Ef- ficient replication of the genotype 2a hepatitis C virus subgenomic replicon. Gastroenterology. 2003 Dec;125(6):1808-17. Kawai T, Takahashi K, Sato S, Coban C, Kumar H, Kato H, Ishii KJ, Takeuchi O, Akira S. IPS-1, an adaptor triggering RIG-I- and Mda5-mediated type I interferon induction. Nat Immunol. 2005 Oct;6(10):981-8. Khu YL, Koh E, Lim SP, Tan YH, Brenner S, Lim SG, Hong WJ, Goh PY. Muta- tions that affect dimer formation and helicase activity of the hepatitis C virus helicase. J Virol. 2001 Jan;75(1):205-14. Kim JL, Morgenstern KA, Lin C, Fox T, Dwyer MD, Landro JA, Chambers SP, Markland W, Lepre CA, O'Malley ET, Harbeson SL, Rice CM, Murcko MA, Caron PR, Thomson JA. Crystal structure of the hepatitis C virus NS3 protease domain complexed with a synthetic NS4A cofactor peptide. Cell. 1996 Oct 18;87(2):343-55. Koch JO, Lohmann V, Herian U, Bartenschlager R. In vitro studies on the activation of the hepatitis C virus NS3 proteinase by the NS4A cofactor. Virology. 1996 Jul 1;221(1):54-66. Kolykhalov AA, Mihalik K, Feinstone SM, Rice CM. Hepatitis C virus-encoded enzymatic activities and conserved RNA elements in the 3' non-translated region are essential for virus replication in vivo. J Virol. 2000 Feb;74(4):2046-51. Korolev S, Yao N, Lohman TM, Weber PC, Waksman G. Comparisons between the structures of HCV and Rep helicases reveal structural similarities between SF1 and SF2 super-families of helicases. Protein Sci. 1998 Mar;7(3):605-10.

56 Krieger N, Lohmann V, Bartenschlager R. Enhancement of hepatitis C virus RNA replication by cell culture-adaptive mutations. J Virol. 2001 May;75(10):4614-24. Kuang WF, Lin YC, Jean F, Huang YW, Tai CL, Chen DS, Chen PJ, Hwang LH. Hepatitis C virus NS3 RNA helicase activity is modulated by the two domains of NS3 and NS4A. Biochem Biophys Res Commun. 2004 Apr 23;317(1):211-7. Kukolj G, McGibbon GA, McKercher G, Marquis M, Lefèbvre S, Thauvette L, Gauthier J, Goulet S, Poupart MA, Beaulieu PL. Binding site characterization and resistance to a class of non-nucleoside inhibitors of the hepatitis C virus NS5B polymerase. J Biol Chem. 2005 Nov 25;280(47):39260-7. Labonté P, Axelrod V, Agarwal A, Aulabaugh A, Amin A, Mak P. Modulation of hepatitis C virus RNA-dependent RNA polymerase activity by structure-based site-directed mutagenesis. J Biol Chem. 2002 Oct 11;277(41):38838-46. Lamarre D, Anderson PC, Bailey M, Beaulieu P, Bolger G, Bonneau P, Bös M, Cameron DR, Cartier M, Cordingley MG, Faucher AM, Goudreau N, Kawai SH, Kukolj G, Lagacé L, LaPlante SR, Narjes H, Poupart MA, Rancourt J, Sen- tjens RE, St George R, Simoneau B, Steinmann G, Thibeault D, Tsantrizos YS, Weldon SM, Yong CL, Llinàs-Brunet M. An NS3 protease inhibitor with antiviral effects in humans infected with hepatitis C virus. Nature. 2003 Nov 13;426(6963):186-9. Epub 2003 Oct 26. Laskus T, Radkowski M, Wang LF, Vargas H, Rakela J. The presence of active hepatitis C virus replication in lymphoid tissue in patients coinfected with human immunodeficiency virus type 1. J Infect Dis. 1998 Oct;178(4):1189-92. Lawitz E, Rodriguez-Torres M, Muir AJ, Kieffer TL, McNair L, Khunvichai A, McHutchison JG. Antiviral effects and safety of telaprevir, peginterferon alfa-2a, and ribavirin for 28 days in hepatitis C patients. J Hepatol. 2008 Aug;49(2):163- 9. Lee KJ, Choi J, Ou JH, Lai MM. The C-terminal transmembrane domain of hepatitis C virus (HCV) RNA polymerase is essential for HCV replication in vivo. J Vir- ol. 2004 Apr;78(7):3797-802. Lenz O. Resistance to HCV protease inhibitors. 2nd Nordic Conference of HIV and Hepatitis Drug Resistance and Treatment Strategies, Uppsala, Sweden. 2008. Lerat H, Honda M, Beard MR, Loesch K, Sun J, Yang Y, Okuda M, Gosert R, Xiao SY, Weinman SA, Lemon SM. Steatosis and liver cancer in transgenic mice ex- pressing the structural and nonstructural proteins of hepatitis C virus. Gastroen- terology. 2002 Feb;122(2):352-65. Lesburg CA, Cable MB, Ferrari E, Hong Z, Mannarino AF, Weber PC. Crystal structure of the RNA-dependent RNA polymerase from hepatitis C virus reveals a fully encircled active site. Nat Struct Biol. 1999 Oct;6(10):937-43. Le Pogam S, Jiang WR, Leveque V, Rajyaguru S, Ma H, Kang H, Jiang S, Singer M, Ali S, Klumpp K, Smith D, Symons J, Cammack N, Nájera I. In vitro selected Con1 subgenomic replicons resistant to 2'-C-methyl-cytidine or to R1479 show lack of cross resistance. Virology. 2006 Aug 1;351(2):349-59. Le Pogam S, Kang H, Harris SF, Leveque V, Giannetti AM, Ali S, Jiang WR, Rajy- aguru S, Tavares G, Oshiro C, Hendricks T, Klumpp K, Symons J, Browner MF, Cammack N, Nájera I. Selection and characterization of replicon variants dually resistant to thumb- and palm-binding nonnucleoside polymerase inhibitors of the hepatitis C virus. J Virol. 2006 Jun;80(12):6146-54. Li K, Foy E, Ferreon JC, Nakamura M, Ferreon AC, Ikeda M, Ray SC, Gale M Jr, Lemon SM. Immune evasion by hepatitis C virus NS3/4A protease-mediated cleavage of the Toll-like receptor 3 adaptor protein TRIF. Proc Natl Acad Sci U S A. 2005 Feb 22;102(8):2992-7.

57 Li M, Poliakov A, Danielson UH, Su Z, Janson JC. Refolding of a recombinant full- length non-structural (NS3) protein from hepatitis C virus by chromatographic procedures. Biotechnol Lett. 2003 Oct;25(20):1729-34. Lin C, Lin K, Luong YP, Rao BG, Wei YY, Brennan DL, Fulghum JR, Hsiao HM, Ma S, Maxwell JP, Cottrell KM, Perni RB, Gates CA, Kwong AD. In vitro resistance studies of hepatitis C virus serine protease inhibitors, VX-950 and BILN 2061: structural analysis indicates different resistance mechanisms. J Biol Chem. 2004 Apr 23;279(17):17508-14. Lin CC, Yeh LT, Vitarella D, Hong Z. Viramidine, a prodrug of ribavirin, shows better liver-targeting properties and safety profiles than ribavirin in animals. Antivir Chem Chemother. 2003 May;14(3):145-52. Llinàs-Brunet M, Bailey M, Fazal G, Goulet S, Halmos T, Laplante S, Mau- rice R, Poirier M, Poupart MA, Thibeault D, Wernic D, Lamarre D. Peptide-based inhibitors of the hepatitis C virus serine protease. Bioorg Med Chem Lett. 1998 Jul 7;8(13):1713-8. Lohmann V, Roos A, Körner F, Koch JO, Bartenschlager R. Biochemical and kinetic analyses of NS5B RNA-dependent RNA polymerase of the hepatitis C virus. Virology. 1998 Sep 15;249(1):108-18. Lohmann V, Körner F, Dobierzewska A, Bartenschlager R. Mutations in hepatitis C virus RNAs conferring cell culture adaptation. J Virol. 2001 Feb;75(3):1437-49. Lohmann V, Körner F, Koch J, Herian U, Theilmann L, Bartenschlager R. Replica- tion of subgenomic hepatitis C virus RNAs in a hepatoma cell line. Science. 1999 Jul 2;285(5424):110-3. Lorenz IC, Marcotrigiano J, Dentzer TG, Rice CM. Structure of the catalytic domain of the hepatitis C virus NS2-3 protease. Nature. 2006 Aug 17;442(7104):831-5. Love RA, Parge HE, Wickersham JA, Hostomsky Z, Habuka N, Moomaw EW, Ada- chi T, Hostomska Z. The crystal structure of hepatitis C virus NS3 proteinase reveals a trypsin-like fold and a structural zinc binding site. Cell. 1996 Oct 18;87(2):331-42. Lozach PY, Lortat-Jacob H, de Lacroix de Lavalette A, Staropoli I, Foung S, Amara A, Houles C, Fieschi F, Schwartz O, Virelizier JL, Arenzana-Seis- dedos F, Alt- meyer R. DC-SIGN and L-SIGN are high affinity binding receptors for hepatitis C virus glycoprotein E2. J Biol Chem. 2003 May 30;278(22):20358-66. Lu L, Pilot-Matias TJ, Stewart KD, Randolph JT, Pithawalla R, He W, Huang PP, Klein LL, Mo H, Molla A. Mutations conferring resistance to a potent hepatitis C virus serine protease inhibitor in vitro. Antimicrob Agents Chemother. 2004 Jun;48(6):2260-6. Luo D, Xu T, Hunke C, Grüber G, Vasudevan SG, Lescar J. Crystal structure of the NS3 protease-helicase from dengue virus. J Virol. 2008 Jan;82(1):173-83. Manns MP, McHutchison JG, Gordon SC, Rustgi VK, Shiffman M, Reindol- lar R, Goodman ZD, Koury K, Ling M, Albrecht JK. Peginterferon alfa-2b plus ribavirin compared with interferon alfa-2b plus ribavirin for initial treatment of chronic hepatitis C: a randomised trial. Lancet. 2001 Sep 22;358(9286):958-65. Medivir press release. http://www.medivir.se/v3/se/ir_media/pressrelease.cfm? year=2008&releaseid=390816. Nov 2008. Melén K, Fagerlund R, Nyqvist M, Keskinen P, Julkunen I. Expression of hepatitis C virus core protein inhibits interferon-induced nuclear import of STATs. J Med Virol. 2004 Aug;73(4):536-47. Mercer DF, Schiller DE, Elliott JF, Douglas DN, Hao C, Rinfret A, Addison WR, Fischer KP, Churchill TA, Lakey JR, Tyrrell DL, Kneteman NM. Hepatitis C virus replication in mice with chimeric human livers. Nat Med. 2001 Aug;7(8):927- 33.

58 Meylan E, Curran J, Hofmann K, Moradpour D, Binder M, Bartenschlager R, Tschopp J. Cardif is an adaptor protein in the RIG-I antiviral pathway and is targeted by hepatitis C virus. Nature. 2005 Oct 20;437(7062):1167-72. Musch E, Malek M, von Eick H, Chrissafidou A. Successful application of highly purified natural interferon alpha (multiferon) in a chronic hepatitis C patient resistant to preceding treatment approaches. Hepatogastroen terology. 2004 Sep- Oct;51(59):1476-9. Nguyen TT, Gates AT, Gutshall LL, Johnston VK, Gu B, Duffy KJ, Sarisky RT. Resist- ance profile of a hepatitis C virus RNA-dependent RNA polymerase benzothiadiazine inhibitor. Antimicrob Agents Chemother. 2003 Nov;47(11):3525-30. Neumann AU, Lam NP, Dahari H, Gretch DR, Wiley TE, Layden TJ, Perel- son AS. Hepatitis C viral dynamics in vivo and the antiviral efficacy of interferon-alpha therapy. Science. 1998 Oct 2;282(5386):103-7. O'Farrell D, Trowbridge R, Rowlands D, Jäger J. Substrate complexes of hepatitis C virus RNA polymerase (HC-J4): structural evidence for nucleotide import and de-novo initiation. J Mol Biol. 2003 Feb 28;326(4):1025-35. Oh JW, Ito T, Lai MM. A recombinant hepatitis C virus RNA-dependent RNA polymerase capable of copying the full-length viral RNA. J Virol. 1999 Sep;73(9):7694-702. Okuse C, Rinaudo JA, Farrar K, Wells F, Korba BE. Enhancement of antiviral activity against hepatitis C virus in vitro by interferon combination therapy. Antiviral Res. 2005 Jan;65(1):23-34. Osborn BL, Olsen HS, Nardelli B, Murray JH, Zhou JX, Garcia A, Moody G, Zarit- skaya LS, Sung C. Pharmacokinetic and pharmacodynamic studies of a human serum albumin-interferon-alpha fusion protein in cynomolgus monkeys. J Phar- macol Exp Ther. 2002 Nov;303(2):540-8. Paeshuyse J, Kaul A, De Clercq E, Rosenwirth B, Dumont JM, Scalfaro P, Bartenschlager R, Neyts J. The non-immunosuppressive cyclosporin DE BIO- 025 is a potent inhibitor of hepatitis C virus replication in vitro. Hepatology. 2006 Apr;43(4):761-70. Pavlovi D, Neville DC, Argaud O, Blumberg B, Dwek RA, Fischer WB, Zitzmann N. The hepatitis C virus p7 protein forms an ion channel that is inhibited by long-alkyl-chain iminosugar derivatives. Proc Natl Acad Sci U S A. 2003 May 13;100(10):6104-8. Perni RB, Almquist SJ, Byrn RA, Chandorkar G, Chaturvedi PR, Courtney LF, Decker CJ, Dinehart K, Gates CA, Harbeson SL, Heiser A, Kalkeri G, Kolaczkowski E, Lin K, Luong YP, Rao BG, Taylor WP, Thomson JA, Tung RD, Wei Y, Kwong AD, Lin C. Preclinical profile of VX-950, a potent, selective, and orally bioavailable inhibitor of hepatitis C virus NS3- 4A serine protease. Antimicrob Agents Chemother. 2006 Mar;50(3):899- 909. Pierra C, Benzaria S, Amador A, Moussa A, Mathieu S, Storer R, Gosselin G. Nm 283, an efficient prodrug of the potent anti-HCV agent 2'-C-methylcytidine. Nucleosides Nucleotides Nucleic Acids. 2005;24(5- 7):767-70. Pileri P, Uematsu Y, Campagnoli S, Galli G, Falugi F, Petracca R, Weiner AJ, Houghton M, Rosa D, Grandi G, Abrignani S. Binding of hepatitis C virus to CD81. Science. 1998 Oct 30;282(5390):938-41. Podevin P, Sabile A, Gajardo R, Delhem N, Abadie A, Lozach PY, Beretta L, Bré- chot C. Expression of hepatitis C virus NS5A natural mutants in a hepatocytic cell line inhibits the antiviral effect of interferon in a PKR-independent manner. Hepatology. 2001 Jun;33(6):1503-11.

59 Poliakov A, Hubatsch I, Shuman CF, Stenberg G, Danielson UH. Expression and purification of recombinant full-length NS3 protease-helicase from a new variant of Hepatitis C virus. Protein Expr Purif. 2002 Aug;25(3):363-71. Preugschat F, Averett DR, Clarke BE, Porter DJ. A steady-state and pre- teady-state kinetic analysis of the NTPase activity associated with the hepatitis C virus NS3 helicase domain. J Biol Chem. 1996 Oct 4;271(40):24449-57. Prince AM, Brotman B. The biology of hepatitis C virus infection. Lessons learned from chimpanzees. Curr Stud Hematol Blood Transfus. 1994; (61):195-207. Raboisson P, de Kock H, Rosenquist A, Nilsson M, Salvador-Oden L, Lin TI, Roue N, Ivanov V, Wähling H, Wickström K, Hamelink E, Edlund M, Vrang L, Vendeville S, Van de Vreken W, McGowan D, Tahri A, Hu L, Boutton C, Lenz O, Delouvroy F, Pille G, Surleraux D, Wigerinck P, Samuelsson B, Simmen K. Structure-activity relationship study on a novel series of cyclopentane-containing macrocyclic inhibitors of the hepati tis C virus NS3/4A protease leading to the discovery of TMC435350. Bioorg Med Chem Lett. 2008 Sep 1;18(17):4853- 8. Radkowski M, Wilkinson J, Nowicki M, Adair D, Vargas H, Ingui C, Rakela J, Laskus T. Search for hepatitis C virus negative-strand RNA sequences and analysis of viral sequences in the central nervous system: evidence of replication. J Virol. 2002 Jan;76(2):600-8. Ray RB, Lagging LM, Meyer K, Steele R, Ray R. Transcriptional regulation of cellular and viral promoters by the hepatitis C virus core protein. Virus Res. 1995 Aug;37(3):209-20. Reiser M, Hinrichsen H, Benhamou Y, Reesink HW, Wedemeyer H, Aven dano C, Riba N, Yong CL, Nehmiz G, Steinmann GG. Antiviral efficacy of NS3-serine protease inhibitor BILN-2061 in patients with chronic genotype 2 and 3 hepatitis C. Hepatology. 2005 Apr;41(4):832-5. Ricard J, Meunier JC, Buc J. Regulatory behavior of monomeric enzymes. 1. The mnemonical enzyme concept. Eur J Biochem. 1974 Nov 1;49(1):195-208. Roche update. http://www.roche.com//investors/ir_update/invupdate-2008-04- 28b.htm. 2008 Apr. Roberts SK, Cooksley G, Dore GJ, Robson R, Shaw D, Berns H, Hill G, Klumpp K, Najera I, Washington C. Robust antiviral activity of R1626, a novel nucleoside analog: a randomized, placebo-controlled study in patients with chronic hepatitis C. Hepatology. 2008 Aug;48(2):398-406. Ryder NS. Discontinued drugs in 2006: anti-infectives. Expert Opin Investig Drugs. 2007 Dec;16(12):1867-78. Ryder NS. Discontinued drugs in 2007: anti-infectives. Expert Opin Investig Drugs. 2009 Jan;18(1):1-11. Sakai A, Claire MS, Faulk K, Govindarajan S, Emerson SU, Purcell RH, Bukh J. The p7 polypeptide of hepatitis C virus is critical for infectivity and contains functionally important genotype-specific sequences. Proc Natl Acad Sci U S A. 2003 Sep 30;100(20):11646-51. Sali DL, Ingram R, Wendel M, Gupta D, McNemar C, Tsarbopoulos A, Chen JW, Hong Z, Chase R, Risano C, Zhang R, Yao N, Kwong AD, Raman- athan L, Le HV, Weber PC. Serine protease of hepatitis C virus expressed in insect cells as the NS3/4A complex. Biochemistry. 1998 Mar 10;37(10):3392-401. Santolini E, Migliaccio G, La Monica N. Biosynthesis and biochemical properties of the hepatitis C virus core protein. J Virol. 1994 Jun;68(6):3631-41.

60 Sarrazin C, Kieffer TL, Bartels D, Hanzelka B, Müh U, Welker M, Win cheringer D, Zhou Y, Chu HM, Lin C, Weegink C, Reesink H, Zeuzem S, Kwong AD. Dy- namic hepatitis C virus genotypic and phenotypic changes in patients treated with the protease inhibitor telaprevir. Gast roenterology. 2007 May;132(5):1767- 77. Satoh S, Tanji Y, Hijikata M, Kimura K, Shimotohno K. The N-terminal region of hepatitis C virus nonstructural protein 3 (NS3) is essential for stable complex formation with NS4A. J Virol. 1995 Jul;69(7):4255-60. Saunier B, Triyatni M, Ulianich L, Maruvada P, Yen P, Kohn LD. Role of the asia- loglycoprotein receptor in binding and entry of hepatitis C virus structural proteins in cultured human hepatocytes. J Virol. 2003 Jan;77(1):546-59. Schering-Plough press release. http://www.schering- plough.com/news/news_art- icle.aspx?reqid=1229611. Nov 2008. Schmitz U, Tan SL. NS5A--from obscurity to new target for HCV therapy. Recent Patents Anti-Infect Drug Disc. 2008 Jun;3(2):77-92. Seeff LB, Hollinger FB, Alter HJ, Wright EC, Cain CM, Buskell ZJ, Ishak KG, Iber FL, Toro D, Samanta A, Koretz RL, Perrillo RP, Goodman ZD, Knodell RG, Git- nick G, Morgan TR, Schiff ER, Lasky S, Stevens C, Vlahcevic RZ, Weinshel E, Tanwandee T, Lin HJ, Barbosa L. Long-term mortality and morbidity of transfu- sion-associated non-A, non-B, and type C hepatitis: A National Heart, Lung, and Blood Institute collaborative study. Hepatology. 2001 Feb;33(2):455-63. Seiwert SD, Andrews SW, Jiang Y, Serebryany V, Tan H, Kossen K, Rajagopalan PT, Misialek S, Stevens SK, Stoycheva A, Hong J, Lim SR, Qin X, Rieger R, Condroski KR, Zhang H, Do MG, Lemieux C, Hingo rani GP, Hartley DP, Josey JA, Pan L, Beigelman L, Blatt LM. Preclinical characteristics of the hepatitis C virus NS3/4A protease inhibitor ITMN-191 (R7227). Antimicrob Agents Chemother. 2008 Dec;52(12):4432-41. Seth RB, Sun L, Ea CK, Chen ZJ. Identification and characterization of MAVS, a mitochondrial antiviral signaling protein that activates NF-kappaB and IRF 3. Cell. 2005 Sep 9;122(5):669-82. Shepard CW, Finelli L, Alter MJ. Global epidemiology of hepatitis C virus infection. Lancet Infect Dis. 2005 Sep;5(9):558-67. Shimotohno K, Tanji Y, Hirowatari Y, Komoda Y, Kato N, Hijikata M. Processing of the hepatitis C virus precursor protein. J Hepatol. 1995;22(1 Suppl):87-92. Sidwell RW, Huffman JH, Khare GP, Allen LB, Witkowski JT, Robins RK. Broad- spectrum antiviral activity of Virazole: 1-beta-D- ribofuranosyl- 1,2,4-triazole-3- carboxamide. Science. 1972 Aug 25;177(50):705-6. Sikora B, Chen Y, Lichti CF, Harrison MK, Jennings TA, Tang Y, Tackett AJ, Jordan JB, Sakon J, Cameron CE, Raney KD. Hepatitis C virus NS3 helicase forms oli- gomeric structures that exhibit optimal DNA unwinding activity in vitro. J Biol Chem. 2008 Apr 25;283(17):11516-25. Simmonds P, Bukh J, Combet C, Deléage G, Enomoto N, Feinstone S, Hal fon P, In- chauspé G, Kuiken C, Maertens G, Mizokami M, Murphy DG, Okamoto H, Pawlotsky JM, Penin F, Sablon E, Shin-I T, Stuyver LJ, Thiel HJ, Viazov S, Weiner AJ, Widell A. Consensus proposals for a unified system of nomenclature of hepatitis C virus genotypes. Hepatology. 2005 Oct;42(4):962-73. Steinkühler C, Biasiol G, Brunetti M, Urbani A, Koch U, Cortese R, Pessi A, De Francesco R. Product inhibition of the hepatitis C virus NS3 protease. Biochem- istry. 1998 Jun 23;37(25):8899-905.

61 Steinmann E, Penin F, Kallis S, Patel AH, Bartenschlager R, Pietschmann T. Hepat- itis C virus p7 protein is crucial for assembly and release of infectious virions. PLoS Pathog. 2007 Jul;3(7):e103. Summa V, Petrocchi A, Pace P, Matassa VG, De Francesco R, Altamura S, Tomei L, Koch U, Neuner P. Discovery of alpha,gamma-diketo acids as potent selective and reversible inhibitors of hepatitis C virus NS5b RNA-dependent RNA polymerase. J Med Chem. 2004 Jan 1;47(1):14-7. Tanji Y, Hijikata M, Satoh S, Kaneko T, Shimotohno K. Hepatitis C virus-encoded nonstructural protein NS4A has versatile functions in viral protein processing. J Virol. 1995 Mar;69(3):1575-81. Tanji Y, Kaneko T, Satoh S, Shimotohno K. Phosphorylation of hepatitis C virus-encoded nonstructural protein NS5A. J Virol. 1995 Jul;69(7):3980- 6. Taremi SS, Beyer B, Maher M, Yao N, Prosise W, Weber PC, Malcolm BA. Con- struction, expression, and characterization of a novel fully activated recombinant single-chain hepatitis C virus protease. Protein Sci. 1998 Oct;7(10):2143-9. Taylor DR, Shi ST, Romano PR, Barber GN, Lai MM. Inhibition of the in terferon- inducible protein kinase PKR by HCV E2 protein. Science. 1999 Jul 2;285(5424):107-10. Tedbury PR, Harris M. Characterisation of the role of zinc in the hepatitis C virus NS2/3 auto-cleavage and NS3 protease activities. J Mol Biol. 2007 Mar 9;366(5):1652-60. Epub 2006 Dec 28 Thibeault D, Bousquet C, Gingras R, Lagacé L, Maurice R, White PW, Lamarre D. Sensitivity of NS3 serine proteases from hepatitis C virus genotypes 2 and 3 to the inhibitor BILN 2061. J Virol. 2004 Jul;78(14):7352-9. Thibeault D, Massariol MJ, Zhao S, Welchner E, Goudreau N, Gingras R, Llinàs- Brunet M, White PW. Use of the Fused NS4A Peptide-NS3 Protease Domain To Study the Importance of the Helicase Domain for Protease Inhibitor Binding to Hepatitis C Virus NS3-NS4A. Biochemistry. 2009 Feb 3;48(4):744-53. Timm J, Lauer GM, Kavanagh DG, Sheridan I, Kim AY, Lucas M, Pillay T, Ouchi K, Reyor LL, Schulze zur Wiesch J, Gandhi RT, Chung RT, Bhard- waj N, Klen- erman P, Walker BD, Allen TM. CD8 epitope escape and reversion in acute HCV infection. J Exp Med. 2004 Dec 20;200(12):1593-604. Tomei L, Failla C, Santolini E, De Francesco R, La Monica N. NS3 is a serine protease required for processing of hepatitis C virus polyprotein. J Vi rol. 1993 Jul;67(7):4017-26. Tomei L, Failla C, Vitale RL, Bianchi E, De Francesco R. A central hydrophobic domain of the hepatitis C virus NS4A protein is necessary and sufficient for the activation of the NS3 protease. J Gen Virol. 1996 May;77 ( Pt 5):1065-70. Tomei L, Altamura S, Bartholomew L, Biroccio A, Ceccacci A, Pacini L, Narjes F, Gennari N, Bisbocci M, Incitti I, Orsatti L, Harper S, Stansfield I, Rowley M, De Francesco R, Migliaccio G. Mechanism of action and antiviral activity of benzimidazole-based allosteric inhibitors of the hepatitis C virus RNA-dependent RNA polymerase. J Virol. 2003 Dec;77(24):13225-31. Tong X, Chase R, Skelton A, Chen T, Wright-Minogue J, Malcolm BA. Identifica- tion and analysis of fitness of resistance mutations against the HCV protease inhibitor SCH 503034. Antiviral Res. 2006 Jun;70(2):28-38. Trozzi C, Bartholomew L, Ceccacci A, Biasiol G, Pacini L, Altamura S, Nar- jes F, Muraglia E, Paonessa G, Koch U, De Francesco R, Steinkuhler C, Migliaccio G. In vitro selection and characterization of hepatitis C virus serine protease variants resistant to an active-site peptide inhibitor. J Vi- rol. 2003 Mar;77(6):3669- 79.

62 Tsukiyama-Kohara K, Iizuka N, Kohara M, Nomoto A. Internal ribosome entry site within hepatitis C virus RNA. J Virol. 1992 Mar;66(3):1476- 83. Urbani A, Biasiol G, Brunetti M, Volpari C, Di Marco S, Sollazzo M, Orrú S, Piaz FD, Casbarra A, Pucci P, Nardi C, Gallinari P, De Francesco R, Steinkühler C. Multiple determinants influence complex formation of the hepatitis C virus NS3 protease domain with its NS4A cofactor peptide. Biochemistry. 1999 Apr 20;38(16):5206-15. Venkatraman S, Bogen SL, Arasappan A, Bennett F, Chen K, Jao E, Liu YT, Lovey R, Hendrata S, Huang Y, Pan W, Parekh T, Pinto P, Popov V, Pike R, Ruan S, Santhanam B, Vibulbhan B, Wu W, Yang W, Kong J, Liang X, Wong J, Liu R, Butkiewicz N, Chase R, Hart A, Agrawal S, Ingravallo P, Pichardo J, Kong R, Baroudy B, Malcolm B, Guo Z, Prongay A, Madison V, Broske L, Cui X, Cheng KC, Hsieh Y, Brisson JM, Prelusky D, Korf- macher W, White R, Bogdanowich- Knipp S, Pavlovsky A, Bradley P, Saksena AK, Ganguly A, Piwinski J, Giri- javallabhan V, Njoroge FG. Dis- covery of (1R,5S)-N-[3-amino-1-(cyclobutyl- methyl)-2,3-dioxopropyl]-3-[2(S)-[[[(1,1-dimethylethyl)amino]carbonyl]amino] -3,3-dimethyl-1- oxobutyl]-6,6-dimethyl-3-azabicyclo[3.1.0]hexan-2(S)-carboxamide (SCH 503034), a selective, potent, orally bioavailable hepatitis C virus NS3 protease inhibitor: a potential therapeutic agent for the treatment of hepatitis C infection. J Med Chem. 2006 Oct 5;49(20):6074-86. Vertex Pharmaceuticals press release. http://investors.vrtx.com/releasede tail.cfm?ReleaseID=344629. Nov 2008. Vyas J, Elia A, Clemens MJ. Inhibition of the protein kinase PKR by the internal ribosome entry site of hepatitis C virus genomic RNA. RNA. 2003 Jul;9(7):858- 70. Wang C, Siddiqui A. Structure and function of the hepatitis C virus internal ribosome entry site. Curr Top Microbiol Immunol. 1995;203:99-115. Wakita T, Pietschmann T, Kato T, Date T, Miyamoto M, Zhao Z, Murthy K, Haber- mann A, Kräusslich HG, Mizokami M, Bartenschlager R, Liang TJ. Production of infectious hepatitis C virus in tissue culture from a cloned viral genome. Nat Med. 2005 Jul;11(7):791-6. Walewski JL, Keller TR, Stump DD, Branch AD. Evidence for a new hepatitis C virus antigen encoded in an overlapping reading frame. RNA. 2001 May;7(5):710- 21. Weiner AJ, Brauer MJ, Rosenblatt J, Richman KH, Tung J, Crawford K, Bonino F, Saracco G, Choo QL, Houghton M, et al. Variable and hyper variable domains are found in the regions of HCV corresponding to the flavivirus envelope and NS1 proteins and the pestivirus envelope glycoproteins. Virology. 1991 Feb;180(2):842-8. Whitby K, Taylor D, Patel D, Ahmed P, Tyms AS. Action of celgosivir (6 O- butanoyl castanospermine) against the pestivirus BVDV: implications for the treatment of hepatitis C. Antivir Chem Chemother. 2004 May;15(3):141-51. Xu LG, Wang YY, Han KJ, Li LY, Zhai Z, Shu HB. VISA is an adapter protein required for virus-triggered IFN-beta signaling. Mol Cell. 2005 Sep 16;19(6):727-40. Xu Z, Choi J, Yen TS, Lu W, Strohecker A, Govindarajan S, Chien D, Selby MJ, Ou J. Synthesis of a novel hepatitis C virus protein by ribosomal frameshift. EMBO J. 2001 Jul 16;20(14):3840-8. Yao N, Reichert P, Taremi SS, Prosise WW, Weber PC. Molecular views of viral polyprotein processing revealed by the crystal structure of the hepatitis C virus bifunctional protease-helicase. Structure. 1999 Nov 15;7(11):1353-63.

63 Zemer R, Kitay Cohen Y, Naftaly T, Klein A. Presence of hepatitis C virus DNA sequences in the DNA of infected patients. Eur J Clin Invest. 2008 Nov;38(11):845-8. Zhang C, Cai Z, Kim YC, Kumar R, Yuan F, Shi PY, Kao C, Luo G. Stimulation of hepatitis C virus (HCV) nonstructural protein 3 (NS3) helicase activity by the NS3 protease domain and by HCV RNA-dependent RNA polymerase. J Virol. 2005 Jul;79(14):8687-97.

+ + , - ) ./ / 0 , 1 . /,21

, / 0 , 1 . /,213 + , + 13 ,,1 1 * 4 5 / , 6 7 5 / / ,* 3 5/, / 1 , * ,,1 / 2/ / 2 , / + , / 0 , 1 . /,214 89 : 13 $;;3 / 5 *, / / , </ + , / 0 , 1 . /,21=4>

* ) *, 4 4 )*) ) )#&(((