Patent Application
20110251085
The present invention relates to the field of virology. More precisely, the invention provides a method of determining the ability of a test compound to modulate the biological activity of a variant of a target protein, wherein said test compound is previously known to modulate the biological activity of said protein. This invention is useful to determine whether a drug candidate, such as anti-virals, active against a target protein is active against a variant of said protein.
Hepatitis C is a global health problem with 170 million carriers worldwide and 3 to 4 million new cases each year. The virus responsible for this post transfusion non A non B Hepatitis was identified in 1989 as a single stranded RNA virus belonging to the Flaviviridae (Choo et al., 1989). Potential natural histories of HCV include progression from acute infection to long term infection (for 75 to 85% of patients), from long term disease to cirrhosis (for 17 to 20%) and from cirrhosis to decompensation or hepato-cellular carcinoma (for 1 to 5%) (Mc Hutchison, 2004; Walker, 2002). Currently, the standard care consists in a combination between Interferon, a cytokine with immuno-modulatory and antiviral activity (Moussalli et al., 1998) and Ribavirin, a synthetic guanosine nucleoside analogue (Hugle et al., 2003). For patients infected with HCV genotype 1a/1b (the predominant one in USA, Japan and Europe), the sustained viral response (loss of serum HCV RNA following 24 weeks of antiviral therapy) is at best 42-46% (Walker et al., 2002, Gordon et al., 2005; Lake-Bakaar et al., 2003). Besides its relative inefficacy, this combination therapy yields significant side effects (Fried Michael, 2002). New treatment regimens are needed and investigators have focused on the identification of agents that inhibit specific steps in the virus' lifecycle.
Upon interaction and fusion of viral and cellular membranes, RNA genome is released into the cytoplasm of a newly infected cell and serves as template for RNA replication. Translation of HCV genome depends on an internal ribosome entry site and produces a large polyprotein which is proteolytically cleaved to produce 10 viral proteins. The amino terminal one third of the polyprotein encodes the structural proteins: core protein glycoproteins E1+E2. After the structural region, comes a small integral protein, P7, which seems to function as an ion chemical. The remainder of the genome encodes the non structural proteins NS2, N3, NS4A, NS4B, NS5A & NS5B which coordinate the intracellular processes of the virus life cycle (Lindenbach et al., 2005). The development of anti-virals that block essential viral enzymes represents a straight forward approach to developing new anti HCV agents. Although all HCV enzymes are, in theory, equally appropriate for therapeutic intervention, the NS5B RNA polymerase and NS3-4A serine protease are respectively important for genome replication and polyprotein processing and were the most studied. A variety of biochemical in vitro assays have been developed and used in screening campaigns (Behrens et al., 1996; Luo et al., 2000; Oh et al., 1999; Lohmann et al., 1997; Yamashita et al., 1998; Mc Kercher et al., 2004). Although useful, these assays could not predict the enzyme's activity and interaction with inhibitors in a physiological intracellular context. Besides, they don't provide information about the molecule's bioavailability and toxicity.
Until 1999, all cell-based screening efforts for HCV drug discovery relied on surrogate viral systems such as bovine viral diarrhea virus. In 1999, a significant breakthrough in studying HCV RNA replication occurred when the laboratory of R. Bartenschlager developed the HCV replicon system (Lohmann et al., 1999). It became possible to test the effect of inhibitors of traditional targets such as NS3 protease, helicase and NS5B polymerase in an in vitro HCV RNA replication system consisting in a dicistronic, selectable subgenomic HCV replicon (Bartenschlager et al., 2002). The first generation of selectable and reporter-selectable replicons have not been conducive for carrying out high throughput screening due to labor-intensive quantitative RT-PCR methods used in screening and low signal to noise ratio, respectively. In addition, this system requires cell cultures adaptative mutations and still needs improvements of signal window (Blight et al., 2000; Krieger et al., 2001). Recent publication reported substantial improvements (Hao et al., 2006) to replicon assay limitations as screening platform (Bourne et al., 2005; Zuck et al., 2004; O'Boyle et al., 2005) and a few compounds issued from screening on replicon are in clinical phase of development. However, it remains that replicon assay is not a single-target assay.
Since HCV demonstrated a high mutation rate in vivo, it was not surprising that some of the molecules under clinical evaluation already induced enzyme specific mutations that lead to decreased sensitivity to the inhibitors used (Sarrazin et al., 2007). Thus, it appears that the resistance profile's study of anti-HCV drugs is essential for their optimization and long term clinical success. Efforts are being made to obtain recombinant NS5B polymerases from several different viral isolates but face problems of physiological relevance. New replicons from clinically relevant viral isolates are under investigation but are still limited to genotypes 1 and 2 and to some cell clones. The observed mutation's implication in the resistance phenotype often needs further investigation (Trozzi et al., 2003; Nguyen et al., 2003). Knowing these techniques' limitations, a cellular screening platform for anti-HCV activity evaluation and resistance profiling is urgently needed. This is the aim of the instant invention.
The inventors now demonstrated that the cell based assay developed by M. Mehtali (WO 2006/046134) is able to detect conformational changes in the target HCV polymerase and protease upon binding to inhibitors. This cell based assay, named 3D-Screen platform, is based on the identification of short peptides that bind specifically the un-liganded target using the two-hybrid system and allow the screening for molecules that dissociate the interaction between the target and the 3D-Sensor peptide. The inventors also demonstrate that this HTS compatible assay enabled the rapid identification of HCV specific anti-NS5B molecules, but in addition is sensitive enough to perform resistance mutants profiling.
A recent study reported constrained peptides that act as inhibitors of NS5B (Amin et al., 2003) but were not used as tools for drug discovery. Several studies used phage affinity selection to identify peptides sensitive to estrogen receptor's conformation (Paige et al., 1999). Another patent WO 02/004956 exploited several peptides' binding to a liganded form of the Estradiol receptor (Fowlkes et al., 2002) to predict the biological activity of a substance by comparing its peptide “fingerprint” to that of a known reference ligand. A very recent publication reported FRET-hybrid interaction methods to screen for peptide ligands capable of recognizing target receptors (You et al., 2006). There was, though, to inventors'knowledges, no report that used peptides as 3D-sensors to screen for drugs directed against the “sensed” HCV polymerase and protease target proteins and their variants thereof.
The instant invention provides an in vitro method of determining the ability of a test compound to modulate the biological activity of a variant of a target protein, wherein a ligand is previously known to modulate the biological activity of said target protein, said method comprising the steps of:
Step A) Selecting at least one binding peptide which binds to said target protein and to said variant of the target protein, said method comprising the steps of:
Preferably the test compounds are small organic molecules, e.g., molecules with a molecular weight of less than 2000 daltons, preferably less than 1000 daltons, more preferably less than 800 daltons, and more preferably less than 500 daltons, which are pharmaceutically acceptable. Test compounds are preferably of chemical classes amenable to synthesis as a combinatorial library. The test compounds that are selected by the method of the invention includes substances which are similar or “ligand-like” to the substances already identified as having the ability to specifically bind the target protein such as potential pharmacological agonists, antagonists, co-activators and co-inhibitors for the protein in question. Such “ligand-like” test compounds may have a close mechanism of action and a close biological activity to the known-protein ligand, with the potential side effects of the known-protein ligands (agonist or antagonist). More importantly, the test compounds that are selected by the method of the invention includes substances which are not similar to the substances already identified as having the ability to specifically bind the target protein, because the method of the invention identified binding peptides in their native conformation in opposition to the methods of the prior art. Said latter compounds, which are not ligand-like molecule, are expected to have different biological and pharmacological properties compared to the known protein ligand. Alternatively the test compounds may be the binding peptides or a variant thereof, or chemical molecules that mimic said binding peptides, identified by the method of the invention. The test compound of the invention may have been previously identified in a library of chemical molecules by different in vitro assays known by the man skilled in the art, such as for example the cell based assay described in the patent document WO 2006/046134 published on 4 may 2006, particularly in examples 4 and 5. The present invention also relates to the test compounds selected by the method of the invention.
The target protein of the invention may be a naturally occurring substance, or a subunit or domain thereof, from any natural source, including a virus, a micro-organism (including bacterial, fungi, algae, and protozoa), an invertebrate (including insects and worms), the normal or pathological cells of an animal, especially a vertebrate (especially a mammal, bird or fish and, among mammals, particularly humans, apes, monkeys, cows, pigs, goats, llamas, sheep, rats, mice, rabbits, guinea pigs, cats and dogs), or the normal or pathological cells of a plant. The target proteins may be alternatively a non-naturally occurring protein that have been in vitro created or modified such as a mutated protein, a chimeric protein or a artificial protein. The target protein may be a glyco-, lipo-, phospho-, or metalloprotein. It may be a nuclear, cytoplasmic, membrane, or secreted protein. It is also an object of the present invention that the target protein, instead of being a protein, may be a macromolecular nucleic acid, lipid or carbohydrate. If it is a nucleic acid, it may be a ribo-or a deoxyribonucleic acid, and it may be single or double stranded. The target protein does not need to be a single macromolecule. The target protein may be a homo or hetero-multimer (dimer, trimer, tetramer, . . . ) of macromolecules. Additionally, the target protein may necessitate binding partners, such as proteins, oligo-or polypeptides, nucleic acids, carbohydrates, lipids, or small organic or inorganic molecules or ions. Examples include cofactors, ribosomes, polysomes, and chromatin.
The biological activity of the target protein is not limited to a specific activity such as a receptor or an enzymatic activity. Non limiting examples of target proteins include nuclear receptors, orphan nuclear receptor, tyrosine kinase receptors, G-protein coupled receptors, endothelin, erythropoietin receptor, FAS ligand receptor, protein kinases (protein kinase C, tyrosine kinase, serine kinase, threonine kinase, nucleotide kinase, polynucleotide kinase), protein phosphatases (serine/threonine phosphatase, tyrosine phosphatase, nucleotide phosphatase, acid phosphatase, alkaline phosphatase, pyrophosphatase), cell cycle regulators (cyclin cdk2, CDC2, CDC25, P53, RB), GTPases, Rac, Rho, Rab, Ras, endoprotease, exoprotease, metalloprotease, serine protease, cysteine protease, nuclease, polymerase, reverse transcriptase, integrase, ion channels, chaperonins (i.e. heat shock proteins), deaminases, nucleases (i.e. deoxyribonuclease, ribonuclease, endonuclease, exonuclease), telomerase, primase, helicase, dehydrogenase, transferases (peptidyl transferase, transaminase, glycosyltransferase, ribosyltransferase, acetyl transferase, guanylyltransferase, methyltransferase, . . . ), hydrolases, carboxylases, isomerases, glycosidases, deaminases, lipases, esterases, sulfatases, cellulases, lyases, reductases, ligases, . . . .
The target proteins of the invention may be structural and non-structural proteins selected among viral proteins, bacterial proteins, vegetal proteins, animal proteins and human proteins. In a preferred embodiment, the target protein is a viral protein. Said viral protein is preferably a hepatitis C virus protein, more preferably a hepatitis C virus protein selected among core protein, glycoproteins E1 and E2, NS1, NS2, NS3, NS4A, NS4B, NS5A, NS5B. Preferably hepatitis C virus proteins are NS5B and NS3. Therefore, the present invention relates to an in vitro method of determining the ability of a test compound to modulate the biological activity of a variant of a hepatitis C virus protein selected among core protein, glycoproteins E1 and E2, NS1, NS2, NS3, NS4A, NS4B, NS5A, NS5B, and more preferably among the dimmer NS4A-NS3 full length protein, or NS4A-NS3 protease protein, or NS5B, wherein a ligand is previously known to modulate the biological activity of said target protein, said method comprising the steps of:
Step A) Selecting one binding peptide which binds to said target protein and to said variant of the target protein, said method comprising the steps of:
Said viral protein is preferably an influenza virus protein preferably selected among neuraminidase, protein M2 and haemagglutinin. Said viral protein is preferably a lentiviral protein, such as human immunodeficiency virus (HIV) protein, selected among reverse transcriptase, integrase, protease, TAT, NEF, REV, VIF, Vpu, Vpr. Said viral protein is preferably a human respiratory syncytial virus protein preferably selected among proteins N, F and G. Said viral protein is preferably a hepatitis B virus protein preferably the polymerase.
In a second preferred embodiment, said protein is a receptor. The term “receptor” includes both surface and intracellular receptors. Nuclear receptors are of particular interest, especially human nuclear receptor. Nuclear receptors (NR) have been previously described. (NR) are a family of ligand-activated transcriptional activators. These receptors are organized into distinct domains for ligand binding, dimerization, transactivation, and DNA binding. The steroid receptor family is a large family composed of receptors for glucocorticoids, mineralo-corticoids, androgens, progestins, and estrogens. Receptor activation occurs upon ligand binding, which induces conformational changes allowing receptor dimerization and binding of co-activating proteins. These co-activators, in turn, facilitate the binding of the receptors to DNA and subsequent transcriptional activation of target genes. In addition to the recruitment of co-activating proteins, the binding of ligand is also believed to place the receptor in a conformation that either displaces or prevents the binding of proteins that serve as co-repressors of receptor function. If the ligand is a pharmacological agonist, the new conformation is one which interacts with other components of a biological signal transduction pathway, e.g.; transcription factors, to elicit a biological response in the target tissue. If the ligand is a pharmacological antagonist, the new conformation is one in which the receptor cannot be activated by one or more agonists which otherwise could activate that receptor. A non exhaustive list of NR is described in WO 2006/046134 (see pages 14 and 15, and FIG. 1). The NR are preferably selected among an estrogen receptor, an androgen receptor, a glucocorticoid receptor, a retinoic acid receptor alpha (RARα), a retinoic X receptor (RXR), a peroxisome proliferators-activated receptor (PPARs), a liver X receptor alpha (LXRα).
According to a preferred embodiment, the target protein and the variant of target protein of the invention are in a native conformation. Preferably, steps A), B), C and D) of the method of the invention are performed with target protein and the variant of target protein in a native conformation. By native conformation, it is meant that the protein is not liganded to its natural ligand. Ligands are substances which bind to target proteins and thereby encompass both agonists and antagonists. However ligands exist which bind target proteins but which neither agonize nor antagonize the receptor. Natural ligands are those ligands, in nature, without human intervention, responsible for agonizing or antagonizing or binding a natural target protein. The target protein in an un-liganded state has a particular conformation, i.e., a particular 3-D structure. When the protein is complexed to a ligand, the protein's conformation changes.
The term “variant” as used herein encompasses the terms “modified”, “mutated”, “polymorphisms” “genotypes”, quasispecies or “mutant”, terms which are interchangeable. The term “variant” refers to a gene or gene product (i.e., target protein) which displays modifications in sequence and or functional properties (i.e., altered characteristics) when compared to the gene or gene product of reference (i.e., target protein); usually, the gene or gene product of reference is the wild-type gene or gene product. The term “wild-type” refers to a gene or gene product which has the characteristics of that gene or gene product when isolated from a naturally occurring source. A wild-type gene is that which is most frequently observed in a population and is thus arbitrarily designed the “normal” or “wild-type” form of the gene. It is noted that naturally-occurring variants can be isolated; these are identified by the fact that they have altered characteristics when compared to the wild-type gene or gene product.
According to the invention, the variant of a target protein includes mutants of said target protein. Non-limiting examples of mutations are point mutations, deletion, insertion, translocation, missense mutations, etc. Usually, the mutation alters the coding sequence of the gene encoding the target protein, and alters the functional properties of said target protein.
According to another embodiment, the variant of the target protein is a natural polymorphism of said target protein. Usually, the polymorphism is one or several modification(s) in the sequence of the gene encoding for said protein, that do not affect significantly the functional properties of said protein (i.e the polymorphism like the mutation is heritable, but not necessarily harmful to the functional properties of the target protein). Polymorphisms in genes and gene products may arise in the general human population. Therefore the present invention, allows determining whether a drug candidate active against a target protein (i.e human nuclear receptors, . . . ), will be active against a polymorphism of said target protein.
Similarly, the method of the invention allows to determine whether a drug candidate active against a target protein of one virus strain (for example: neuraminidase of influenza virus strain H1N1 or NS5B polymerase of HCV genotype 1), will be active against the homologous protein of another virus strain (i.e neuraminidase of influenza virus strain H3N2 or NS5B polymerase of HCV genotype 2). The present invention will be particularly useful to determine whether a drug candidate against HCV NS5B polymerase or HCV NS3 protease/helicase identified from one HCV genotype, will be also active against, respectively, HCV NS5B polymerase or HCV NS3 protease/helicase from other HCV genotypes.
According to another embodiment, the variant of the target protein is a protein having sequence homology or structural homology with said target protein. Here is a non-limiting example: if the target protein is a protease from HCV, the variant of said target protein may be a human cellular protease having sequence homology or structural homology with said HCV protease. Therefore, the present invention will be particularly useful to determine whether a drug candidate against a viral protease, will be also active against, a human cellular protease; this will be particularly important to evaluate potential toxicity and side effect of a drug candidate. Therefore the method of the invention is useful to predict the toxicity of a drug candidate.
According to a preferred embodiment, the target protein of the invention is the wild-type sequence of hepatitis C virus NS5B protein (genotype 1b) encoded by SEQ ID No 1 or provided in SEQ ID No 51. Non-limiting examples of variants of said target protein (i.e., Hepatitis C virus NS5B protein) are M414T, S368A, C316Y, Y448H, P495L, S419M, M423T, S282T, S96T, N142T, G152E, P156L (the second amino acid of SEQ ID No 51 is considered as the first position for these referenced mutations). According to a second preferred embodiment, the target protein of the invention is the wild-type sequence of hepatitis C virus NS3 protein encoded by SEQ ID No 2 or provided in SEQ ID No 52. Non-limiting examples of variants of hepatitis C virus NS3 protein are A156V/S/T, D168A/V, V107A, R155K, T54A, V36M, V36M/A156T, V36A/A156, V36M/R155K, V36T/T54, V36M/T54A, A156V/R155K (the third amino acid of SEQ ID No 52 is considered as the first position for these referenced mutations). Non-limiting examples of variants of respiratory syncytial virus N protein (see SEQ ID No 53 and 54) are N105D, I129L, L139I, and I129L with L139I. Non-limiting examples of variants of influenza virus neuraminidase are R292K, D151E, R152K, R371K, R224K, E119G, and D198N (see SEQ ID No 55 and 56). Non-limiting examples of variants of HIV protease are L10V, I13V, E35G, M36I and H69K (see SEQ ID No 57 and 58). Non-limiting examples of variants of HIV reverse transcriptase are V179E, T69S, and M184V (see GenBank A.N. AAO84275; GI:37934320; accession AY237815.1 for an example of reference sequence).
The nomenclature of the above mentioned drug resistant mutations mentions the original deduced amino acid (as mentioned in the consensus sequence) followed by the codon number relative to its position and then the amino acid derived by the mutation (and that results in reduced susceptibility to the antiviral agent). The original deduced amino acid could be different from the one mentioned in the consensus sequence if the sequence is derived from patient's serum; without affecting significantly the target's activities and susceptibilities to antivirals.
The techniques of introducing mutations in proteins are well known to the man skilled in the art and can be done either by biological, physical or chemical means.
A known protein ligand is a substance known to be a ligand for the target protein. The known protein ligand of the invention is able to alter the conformation of the protein upon binding to said protein. Usually, said known protein ligand is a pharmacological agonist or antagonist of a target protein in one or more target tissues of the organism from which the target protein is coming from. However, a known protein ligand may be neither an agonist nor an antagonist of the target proteins. For example, it may be the substrate of the target protein if this latter is an enzyme. The known protein ligand may be, but need not be, a natural ligand of the protein. Said known-protein ligand of the invention is selected among known-target protein agonist and known-target protein antagonist. The known ligand is either endogenously or preferably exogenously added to the cells of step B.
According to a preferred embodiment, the target protein is a hepatitis C viral protein selected among:
Concerning these above-referenced HCV protein ligands, their structure are well known by the skilled person. For example:
can be also cited (see Beaulieu et al., 2007, page 616, FIG. 4);
also depicted in the document Manns et al., 2007 (Table 1, page 996, Ref 85); and
According to another preferred embodiment, the target protein is an influenza virus viral protein selected among:
According to another preferred embodiment, the target protein is an HIV viral protein selected among:
According to a preferred embodiment, the target protein is a nuclear receptor selected among:
The main limitation in the method of selecting binding peptides of the invention is the availability of known ligands (i.e drugs) to identify conformation sensitive binding peptides. In the virus field for example, it is often the case that no antiviral drug is known to bind to a viral target protein (i.e respiratory syncythial virus). However, since viruses are intracellular parasites which are using the cellular machinery for their replication, viral proteins are interacting with host cell proteins or molecules (ATP, tRNA, ribosomes, cell enzymes, . . . ). Interactions of numerous virus proteins with those of the host cells have been reported in the scientific literature (Tellinghuisen & Rice Curr Opin Microbiol., 2002, 5:419-427; HIV-1, Human proteins interaction database, National Institute of Allergy & Infectious Diseases; Lo et al., J. Virol., 1996 Aug. 70(8):5177-82). According to another embodiment of the invention, when no ligand (i.e. pharmacological agonists or antagonists) is known to bind to the target proteins, either cellular or viral proteins interacting with the target protein might be used instead to identify conformation sensitive binding peptides. Once peptides interacting with the target protein are identified, an expression vector encoding the protein interacting with the target protein is introduced in the yeast already modified with an expression vector encoding the target protein and an expression vector encoding the peptide library. Peptides that will be blocked by the protein/protein interaction, either due to the conformation modification of the target or due to the inaccessibility of their recognition site on the target protein due to a steric hindrance, will be identified. Therefore, the invention also relates to a method of selecting binding peptide(s) (steps A) and B) of the present invention) which binds a target protein in a native conformation, said method comprising the steps of:
If proteins interacting with the target protein are not known, the man of the art knows how to identify proteins that are able to interact with a target protein. The yeast two-hybrid is a validated technology to identify protein-protein interactions. For example, cDNA clones encoding proteins interacting with the target protein can be isolated with a yeast two-hybrid screen of a cDNA library from human cells of interest. Human cDNA library could be commercially available or will have to be constructed.
By the terms “bind” or “binding”, it is meant herein that the peptide and the target protein interact, attach, join or associate to one another sufficiently that the intended assay can be conducted. By the terms “specific” or “specifically”, it is meant herein that the peptide and the target protein bind selectively to each other and not generally to other components unintended for binding to the subject components. The parameters required to achieve specific interactions can be determined routinely, e.g. using conventional methods in the art.
The term “library” generally refers to a collection of chemical or biological entities which are related in origin, structure, and/or function, and which can be screened simultaneously for a property of interest. The term “combinatorial library” refers to a library in which the individual members are either systematic or random combinations of a limited set of basic elements, the properties of each member being dependent on the choice and location of the elements incorporated into it. Typically, the members of the library are at least capable of being screened simultaneously. Randomization may be complete or partial; some positions may be randomized and others predetermined, and at random positions, the choices may be limited in a predetermined manner. The members of a combinatorial library may be oligomers or polymers of some kind, in which the variation occurs through the choice of monomeric building block at one or more positions of the oligomer or polymer, and possibly in terms of the connecting linkage, or the length of the oligomer or polymer, too. Or the members may be non-oligomeric molecules with a standard core structure with the variation being introduced by the choice of substituents at particular variable sites on the core structure. Or the members may be non-oligomeric molecules assembled like a jigsaw puzzle, but wherein each piece has both one or more variable moieties (contributing to library diversity) and one or more constant moieties (providing the functionalities for coupling the piece in question to other pieces). In a “simple combinatorial library”, all of the members belong to the same class of compounds (e.g., peptides) and can be synthesized simultaneously. A “composite combinatorial library” is a mixture of two or more simple libraries, e.g., DNAs and peptides, or chemical compound. Preferably, a combinatorial library will have a diversity of at least 100, more preferably at least 1,000, still more preferably at least 10,000, even more preferably at least 100,000, most preferably at least 1,000,000, different molecules. The peptide library of the invention is a combinatorial library. The members of this library are peptides having at least three amino acids connected via peptide bonds. Preferably, they are at least four, five, six, seven, eight, nine, ten, eleven, twelve, thirteen, fourteen, fifteen, sixteen, seventeen, eighteen, nineteen, twenty, twenty-one, twenty-two, twenty-three, twenty-four, twenty-five, twenty-six, twenty-seven, twenty-eight, twenty-nine, or thirty amino acids in length. Preferably, they are composed of less than 50, preferably less than 25 amino acids and more preferably less than 15. The peptides may be linear, branched, or cyclic, and may include non-peptidyl moieties. The amino acids are not limited to the naturally occurring amino acids.
The cells of the invention are cells which, naturally or not, functionally express the suitable target protein. Preferably, the cells are eukaryotic cells. The cells may be from a unicellular organism, a multi-cellular organism or an intermediate form (slime mold). If from a multi-cellular organism, the latter may be an invertebrate, a lower vertebrate (reptile, fish, amphibian), a higher vertebrate (bird, mammal) or a plant. According to a preferred embodiment, the cell of steps A), B), C) and D) of the method of the invention is a cell, preferably isolated cell, not integrated into a whole multi-cellular organism or a tissue or organ of an organism.
According to a preferred embodiment, the cells used to select the binding peptides (steps A) and B) of the method of the invention) are non-mammalian eukaryotic cells, preferably a yeast cell. Preferably, the yeast cells are of one of the following genera: Saccharomyces, Schizosaccharomyces, Candida, Hansenula, Pichia, Kluyveromyces, Cryptococcus, Yarrowia and Zygosaccharomyces. More preferably, they are of the species: Saccharomyces cerevisiae. Other non-mammalian cells of interest include plant cells (e.g. arabidopsis), arthropod cells, annelid or nematode cells (e.g., Caenorhabditis elegans; planaria; leeches; earthworms; polychaetus annelids), crustaceans (e.g., daphnia), protozoal cells (e.g. Dictyostelium discoideum), and lower vertebrate (reptiles, amphibians, fish) cells.
According to a preferred embodiment, the cell used to select and test the test compounds (steps C) and D) of the method of the invention) is a mammalian cell of animal or human origin. Non-limiting examples of mammalian cells are human primary cells, human embryonic cells, human cell lines. In a preferred embodiment, human cells are continuous human cell lines; non-limiting examples of human cell lines includes Hela, Huh7, T47D, A549, HEK-293, MCF7. Alternatively, said human cells are embryonic cells, preferably embryonic cells selected among totipotent cells (i.e. Embryonic Stem cells, ES cells), pluripotent, multipotent and unipotent cells. In a specific embodiment, the cell is a human ES cells and the target protein is a key protein involve in the control of a differentiation pathway of ES cells to human differentiated cells. In a second embodiment, said cells are selected among animal cells such as animal primary cells, animal embryonic cells and animal cell lines. Non limiting examples of animal cell lines include VERO, CHO, NSO, COS, MDCK, MDBK and 3T3. When the target protein is an hepatitis C or B viral protein, the cells used to select and test the test compounds (steps C) and D) of the method of the invention) is preferably animal or human hepatocytes, or hepatocyte cell lines. Non limiting examples of hepatocyte cell lines are Huh7, HepG2, and HepaRG. When the target protein is an HIV viral protein, the cells used to select and test the test compounds (steps C) and D) of the method of the invention) is preferably animal or human lymphocytes or lymphoblastoid cell lines, such as HL60, U937, TF-1, K562 and IM-9. When the target protein is an influenza or RSV viral protein, the cells used to select and test the test compounds (steps C) and D) of the method of the invention) is preferably animal or human epithelial cell lines. Non limiting examples of lung cell lines are A549 and HEp-2.
Each cell used to select the binding peptides (steps A) and B) of the method of the invention) cell is co-expressing:
Preferably, said signal generated by the reporter gene expression is a quantitative signal which can be measured.
Preferably, the total quantity of said signal which is measured is proportional to the number of peptide specifically bound to the target protein or to a variant thereof, resulting to the measure of an increased, a conserved, a decrease or an absence of binding (compared to a control or referenced sample).
A signal is produced when said peptide binds to said target protein and to said variant of target protein. Said signal is decreased or absent when said peptide has a decreased ability or is unable to bind to said target protein or to said variant of target protein, liganded to a known ligand to said target protein.
The cell used to select and test the test compounds (steps C) and D) of the method of the invention) is co-expressing:
The signal producing system of the method of the invention is selected among signal producing system endogenous to the cell and signal producing system exogenous to the cell. Preferably, said signal producing system is endogenous to the cell. The signal producing system comprises a protein-bound component which is fused to said target protein or moiety so as to provide a chimeric protein and a peptide-bound component which is fused to said peptide so as to provide a chimeric peptide, whereby a signal is produced when the peptide-bound and protein-bound components are brought into physical proximity as a result of the binding of the peptide to the target protein. Said components is a DNA-binding domain (DBD) and another of said components is a complementary transactivation domain (AD), and the signal producing system further comprises at least one reporter gene operably linked to an operator bound by said DNA-binding domain, the binding of the peptide to the target protein resulting in the constitution of a functional transactivation activator protein which activates expression of said reporter gene.
Said signal producing system of the invention comprises:
A signal is “produced” when said signal is detectable by the means adapted and normally used by the man skilled in the art to perform the detection (i.e. human eye, spectrophotometer, etc.). The indication that an absence of binding had occurred is when the signal is no longer detectable by the means routinely used to detect the reporter gene expression by the man skilled in the art. The indication that a decrease in the binding had occurred is when the decrease of the signal generated by the reporter gene expression is at least 50%, preferably at least 75%, and more preferably at least 90% of the signal generated when the specific binding occurred.
When the signal is the death or survival of the cell in question, the assay is said to be a selection. When the signal merely results in detectable phenotype by which the signalling cell may be differentiated from the same cell in a non-signalling state, the assay is a screen. However the term “screening” in the present invention may be used in a broader sense to include a selection. According to the present invention, “reporter gene” means a gene, the expression of which in one cell, allows to detect said cell in a large population of cells; that is to say, that it allows to distinguish between the cells that are expressing or not said reporter gene. The reporter gene of the invention includes detectable phenotype reporter genes and resistance selection genes. Non-limiting examples of resistance selection genes include yeast genes: 3-isopropylmalate dehydrogenase (LEU2), phosphoribosylanthranilate isomerase (TRP1), imidazole-glycerol-phosphate dehydratase (HIS3), Orotidine-5′-phosphate (OMP) decarboxylase (URA3) and antibiotic resistance genes. Among the antibiotics, one can recite, neomycin, tetracycline, ampicilline, kanamycine, phleomycine, bleomycine, hygromycine, chloramphenicol, carbenicilline, geneticine, puromycine, blasticidine. The antibiotics resistance genes are well known to the man skilled in the art. For example, neomycine gene provides the cells with a resistance to G418 antibiotic added to the cell culture medium. Alternatively, non-limiting examples of detectable phenotype reporter genes include the following genes: DHFR, luciferase, chloramphenicol acetyltransferase, beta-lactamase, adenylate cyclase, alkaline phosphatase, and beta-galactosidase, and auto-fluorescent proteins (such as green fluorescent protein, red fluorescent protein, blue fluorescent protein, yellow fluorescent protein, and all the variants and derived fluorescent proteins). The signal producing system may include, besides reporter gene(s), additional genetic or biochemical elements which cooperate in the production of the signal. Such an element could be, for example, the substrate of a reporter gene which is an enzyme.
There may be more than one signal producing system, and the system may include more than one reporter gene.
According to a preferred embodiment, the method of selecting binding peptides (steps A) and B) of the method of the invention)) is performed in yeast cells, and the signal producing system is comprising at least two reporter genes in tandem, one resistance selection gene selected among HIS3, LEU2, TRP1, URA3 and one reporter gene selected among luciferase, auto-fluorescent proteins and beta-galactosidase. In another preferred embodiment, the method of selecting test compounds of the invention (steps C) and D) of the method of the invention)) is performed in mammalian cells, preferably a human cell, and the signal producing system is comprising at least one reporter gene selected among luciferase, auto-fluorescent proteins and beta-galactosidase. The test compound is endogenously or, preferably, exogenously added to the cell of steps C) and D).
The present invention is useful to predict whether a candidate drug will be active or not against a target protein and the variants of such target protein and to select the drug candidate with the broadest scope of action.
The invention also relates to the test compounds selected by the method of the invention. The selected test compounds of the invention may be used as an assay hit for the identification of hits which constitute new candidate molecule(s) for drug development. The invention also relates to the use of selected test compounds of the invention as a drug with a broad spectrum of action against target protein wild type, target protein polymorphisms, target protein resistance mutants, especially for the prophylactic and therapeutic treatment of diseases, such as, without limitation and as examples, human and animal viral diseases and human and animal cancers.
The invention is also directed to an isolated peptide selected from the group consisting of:
or recombinant vector or host cell comprising said peptide.
The invention also relates to the method for determining the ability of a test compound to modulate the biological activity of a variant of a hepatitis C virus protein according to the present invention characterized in that said methods comprises a step wherein the test compound is tested for its ability to decrease the binding of the peptide:
For the remainder of the description, reference will be made to the legend to the figures below.
(a) Isolation of peptides interacting with the NS5BΔ21 in the yeast two-hybrid-system. Yeast cells carrying the bait (NS5BΔ21-LexA) were transformed with the prey represented by a library of random peptides (in frame with the viral VP16 activation domain). Colonies growing on selective medium and staining blue allowed the isolation of individual peptides that would specifically interact with the polymerase. (b) Isolation of conformation sensitive peptides. The interacting peptides were then tested in presence of ligands known to induce conformational changes in the bait NS5B. The “3D-sensors” were thus identified as the ones whose binding to the target polymerase was prevented by known anti-NS5B. The corresponding colonies didn't show any expression of the selection or reporter gene. (c) Characterization of NS5BΔ21 “3D-sensors” in a spot test. Here were represented 2 selected yeast colonies: one carrying the ER and its specific peptide called No 30-2 isolated previously and used as a control, the other carrying the NS5BΔ21 and its specific peptide called N21C272. Both clones were spotted on selective medium lacking histidine, leucine and tryptophane, supplemented or not with polymerase ligands and overlayed with an X-Gal staining The NS5BΔ21+N21C272 clone grew on selective medium in the absence of any NS5B ligand and stained positive for LacZ. Colonies issued from the same culture didn't show any growth on selective medium containing varying amounts of specific NS5B inhibitor (Benzothiadiazin CVVS023477). Non specific viral polymerase inhibitors (Foscarnet, PAA) or unrelated ligands (Estradiol) didn't show any effect on growth or LacZ staining Our control clone (ER positif specific 3D sensor No 30.2) was sensitive to its corresponding specific ligand (Estradiol) but not to NS5B or other viral polymerase inhibitors.
Hela cells were co-transfected with plasmids expressing: (i) the 3D-sensor isolated in yeast fused to the trans-activation domain (VP16-AD); (ii) the target polymerase (NS5BΔ21 fused to the yeast Gal4-DBD) and (iii) the luciferase reporter gene whose expression is inducible by the complex [NS5BΔ21/3D-sensor-VP16].
Hela cells were transiently transfected with the luciferase reporter, the 3D-sensor and vectors expressing either native NS5BΔ21 or different mutants. Measurements of luciferase expression in absence of any ligand were represented.
Hela cells were transiently transfected with the luciferase reporter, the 3D-sensor N21C272 and the fusion protein NS5BΔ21-Gal4 then incubated for 24 hours with either specific or non specific ligands. Percentages of inhibition of resulting luciferase's expression were calculated versus an equivalent amount of solvent (DMSO). NS5B specific inhibitor cVVS-023477 (Benzothiadiazin (Pratt J K 2005)) induced a strong inhibition of the luciferase signal at 10 μM final concentration (83.2%+/−7.5; n=15) while another specific NS5B inhibitor cVVS-023476 (Indol-N-Acetamid (Harper S 2005)) showed a noticeable effect starting at 1 μM final concentration (67%+/−8.3, n=6). Non specific ligands such as Foscarnet or Phosphonoacetic acid didn't yield any significant signal inhibition at 10 μM concentration (respectively 11.7%+/−4 and −2.2%+/−33.8; n=3). Another important negative control was Ribavirin, whose presumed mechanism of action seems not to be related to a direct NS5B inhibition (Lohmann V, 2000; Maag D, 2001), didn't show any noticeable effect on our system at 75 μM final concentration (inhibition 8.7%+/−16.3; n=6).
Percentages of luciferase's inhibition were represented for cells transfected with plasmids encoding the reporter luciferase, the NS5BΔ21-Gal4 fusion protein and the 3D-sensor N21C272-VP16 after incubation in presence of increasing concentrations of either cVVS023477 or cVVS023476 (between 0 and 100 μM final concentration) for 24 hours. The obtained curve showed a hyperbolic relationship in agreement with the Michaelis-Menten model.
Hela cells were transiently transfected with the luciferase reporter, the 3D-sensor N21C272 and vectors expressing either native NS5BΔ21 or different mutants fused in frame with the Gal4 DNA-binding domain. Percentages of reporter's inhibition were represented for cells incubated with a specific NS5B inhibitors at increasing concentrations for 24 hours.
One can notice that the 3 mutants resistant to Benzothidiazins challenged on
Measurements of luciferase expression, in presence or absence of specific ligands, were represented. The background signal emitted by cells transfected with the native NS5BΔ21 and a backbone vector without the 3D-sensor-VP16 7182 RLU+/−6643 (n=17) while an equivalent one was obtained with a denaturated NS5B (L30S and L30A conformational destabilizing mutants described by Labonté et al., 2002) in presence of the 3D-sensor (respectively 7636RLU+/−5932; n=5 and 3100RLU+/−1610; n=3). Another background signal could be considered as the residual luciferase expression obtained after maximal inhibition with high (but not toxic) concentrations (100 μM) of the anti-NS5B cVVS-023477 78947RLU+/−3071; n=7 which is far below the uninhibited signal (115,271 RLU+/−60,800; n=22). The different background signals were rather equivalent and the calculated Signal/Background reached 16 for minimum of 3 required.
Hela cells were transfected with the three 3D-Screen vectors then plated in twenty 96-well plates to mimic HTS runs. The cells were incubated for 24 hours with or without reference drugs.
Hela cells were co-transfected with plasmids expressing: (i) the 3D-sensor isolated in yeast fused to the trans-activation domain (VP16-AD); (ii) the target polymerase (NS4A-NS3 protease fused to the yeast Gal4-DBD) and (iii) the luciferase reporter gene whose expression is inducible by the complex [NS4A-NS3 protease/3D-sensor-VP16].
Hela cells were co-transfected with plasmids expressing: (i) the 3D-sensor isolated in yeast fused to the trans-activation domain (VP16-AD); (ii) the target polymerase (NS4A-NS3 full length fused to the yeast Gal4-DBD) and (iii) the luciferase reporter gene whose expression is inducible by the complex [NS4A-NS3 full length/3D-sensor-VP16].
Hela cells were transiently transfected with the luciferase reporter and each of the NS5B or NS3 targets together with their specific 3D-Sensor. Addition of increasing concentrations of the specific NS5B inhibitor cVVS-023476 (Indol-N-Acetamide) induced a strong inhibition only on the NS5B platform. Also, the specific NS3 inhibitor cVVS-023518 inhibited the NS3 protease and full length platforms without significant off target effects.
Hela cells were transiently transfected with the luciferase reporter, the selected 3D-sensors and vectors expressing either native NS3 or a resistance mutant fused in frame with the Gal4 DNA-binding domain. Percentages of reporter's inhibition were represented for cells incubated with a specific NS3 inhibitors at increasing concentrations for 24 hours. The peptidomimetic compound yielded a strong reporter inhibition for the native NS4A-NS3 protease and full length while the same compound didn't show any noticeable reporter inhibition for cells transfected with the resistant mutant A 156V.
The interaction between the 3D-sensor VF-9A11 and either the native or mutated NS4A-NS3 full length protein led to a strong activation of the luciferase reporter. The signal is over background (far above three) for all tested variants except for the R155K mutated protein.
The native 3D-Screen platform is sensitive to the compound's inhibition as well as the D168V and R109K described as so in the literature. The mutants A156V and A156T described as highly resistant to cVVS-023518 show the expected profile on the 3D-Screen platform. The A156S, T54A and V36M, known to be moderately resistant to the peptidomimetic cVVS-023518, also display moderate fold increase in EC50 values by comparison with the wild type NS3 on the 3D-Screen platform.
The 3D-Screen mutant profiling platform displays results in the same range as the ones reported in the literature with enzymatic or replicon based assays for Telaprevir®.
This compound is active on the 3D-Screen platform in the same range as the one reported in literature as well as on the A156S and V36M variants described as sensitive. The highly resistant mutants A156V and D168V showed a high resistance to the compounds inhibitory activity as expected.
The 3D-Screen mutant profiling platform displays results in the same range as the ones reported in the literature with enzymatic or replicon based assays for Ciluprevir.
Percentages of luciferase inhibition in the presence of the Indol Acetamid NS5B specific ligand cVVS-023476 were represented for N21 I4 peptide targeting NS5B from genotype 1a and 1b.
Estradiol, Ribavirin, Phosphonoacetic acid and Foscarnet were purchased from Sigma-Aldrich (Lyon, France). The compound cVVS-023477 (N-1-cyclobutyl-4-hydroxyquinolon-3-yl-benzothiadiazin) was synthesized as described respectively in Pratt J K (2005). The compound cVVS-023476 (1-{[6-Carboxy-2-(4-chlorophenyl)-3-cyclohexyl-1H-indol-1-yl-]acetyl}-N,N-diethylpiperidin-4-aminium Chloride) was synthesized as described respectively in Harper S (2005). The compound cVVS-023518 (2-(2-{2-Cyclohexyl-2-[(pyrazine-2-carbonyl)-amino]-acetylamino}-3,3-dimethyl-butyryl)-octahydro-cyclopenta[c]pyrrole-1-carboxylic acid (1-cyclopropylaminooxalyl-butyl)-amide), also named Telaprevir® or VX-950 (Vertex Pharmaceuticals Inc.) was synthesized as described respectively in WO 02/183A2 (Babine R, 2002). X-Gal (5-bromo-4-chloro-3-indolyl-β-D-galactopyranoside) was purchased from Sigma-Aldrich (Lyon, France).
Lysis buffer: Tris Phosphate pH 7.8 (25 mM)+EDTA (2 mM)+DTT (1 mM)+Glycerol (10%)+Triton (1%).
Luciferin buffer: Tris Phosphate pH 7.8 (20 mM)+DTT (500 mM)+MgCl2 (1.07 mM)+MgSO4 (2.7 mM)+EDTA (0.1 mM)+ATP (0.53 mM)+Acetyl coEnzyme A (0.27 mM)+Luciferin (0.47 mM).
All components were purchased from Sigma-Aldrich (Lyon, France) except Luciferin that was purchased from Promega (Charbonnières, France).
X-Gal solution: 10 ml of 0.5 M Na2HPO4+NaH2PO4+50 μl of 20% SDS+100 μl of 2% X-Gal (in DMF).
1.3—Oligonucleotides for PCR were Obtained from Invitrogen (Cergy Pontoise, France).
Restriction and ligation enzymes were purchased from Promega (Charbonnieres, France) and used according to the manufacturer's recommendations. PCRs were performed with Uptitherm polymerase according the instructions of the manufacturer (Interchim, Montlucon, France).
The Saccharomyces cerevisiae yeast strain L40 (MATa; his3-Δ200; leu2-3,112; trp1-901; [plexAop]-4::HIS3; [plexAop]8::lacZ)) (ATCC, Molsheim, France) was cultured in either liquid (with vigorous shaking) or solid YPD or SD media supplemented with amino acids at 30° C. YPD media was composed as follows: Bacto-yeast extract 1%, Bacto-peptone 2%, Glucose 2% with or without Bacto-agar 2% (purchased from Difco). SD media was composed as follows: Yeast nitrogen base without amino acids 0.67%, Glucose 2% with or without Bacto-agar 2% (purchased from Difco, Paris, France) Amino acids were purchased from Sigma-Aldrich (Lyon, France) and added as follows: Adenine sulphate (20 mg/l SD), L-Tryptophan (50 mg/l SD), L-Histidine (50 mg/l SD, L-Leucin, L-Isoleucin, L-Valin (60 mg/l SD for each).
pVVS468 was obtained by insertion of the genotype 1b HCV NS5BΔ21°s sequence (GenBank D89872) between XhoI and PstI restriction sites of the polylinker of pBTM116 (Bartel & Fields, 1995) in order to be in frame with LexA (Genbank U89960). Expression of the fusion protein is under the control of the ADH1 promoter. Genotype 1b HCV NS5BD21's sequence was obtained by PCR amplification of pIV294 plasmid (kind gift Dr Catherine Schuster, Strasbourg France) carrying HCV's NS5B full-length of genotype 1b. The following oligonucleotide primers were used:
PCR amplification was carried out for 3 min at 96° C. followed by 35 cycles (30s at 94° C., 30s at 55° C. and 2 min at 72° C.) using a T1 thermocycler (Biometra).
pVVS601 was obtained by insertion of the genotype 1b HCV NS3's sequence (GenBank D89872) between XhoI and BamHI restriction sites of the polylinker of pBTM116 (Bartel & Fields, 1995) in order to be in frame with LexA (Genbank U89960). Expression of the fusion protein is under the control of the ADH1 promoter. Genotype 1b HCV NS3's sequence was obtained by PCR amplification of pIV294 plasmid (kind gift Dr Catherine Schuster, Strasbourg France) carrying HCV's NS3 protease of genotype 1b. In fact, the NS3 protease has been shown to be complexed to NS4A (after cis cleavage between the two proteins) for optimal conformation and activity (Lin et al., 1995). A single chain recombinant NS4A-NS3 protease domain has shown full proteolytic activity (Taremi et al., 1998). The following oligonucleotide primers were used in order to obtain a single chain construct NS4A (residues 21-32), the GSGS linker and the NS3 protease (residues 3-181):
PCR amplification was carried out for 3 min at 96° C. followed by 35 cycles (30s at 94° C., 30s at 55° and 2 min at 72° C.) using a T1 thermocycler (Biometra).
pVVS 775 was obtained by insertion of the genotype 1b HCV NS3 full length's sequence (GenBank D89872) between XhoI and BglII restriction sites of the polylinker of pBTM116 (Bartel, 1995) in order to be in frame with LexA (Genbank U89960). Expression of the fusion protein is under the control of the ADH1 promoter. Genotype 1b HCV NS3's sequence was obtained by PCR amplification of pIV294 plasmid (kind gift of Dr Catherine Schuster, Strasbourg, France) carrying HCV's NS3 full-length of genotype 1b. Since the NS3 protease has been shown to be complexed to NS4A for optimal conformation and activity (Lin et al., 1995), a single chain recombinant NS4A-NS3 full length protein has shown full proteolytic and helicase activity (Howe et al., 1999). Moreover, it has been shown that the crystal structure of such recombinant single chain protein was superposed to the native complex NS3-NS4A (Yao et al., 1999). The following oligonucleotide primers were used in order to obtain a single chain construct NS4A (residues 21-32), the GSGS linker and the NS3 full length (residues 3-631):
PCR amplification was carried out for 3 min at 96° C. followed by 35 cycles (30s at 94° C., 30s at 55° and 2 min at 72° C.) using a T1 thermocycler (Biometra).
pVVS625, was derived from pVP16 (Hollenberg et al., 1995) by digestion with SfiI and insertion of a random 15 amino acids peptide library generated by PCR.
pVVS578 was derived from plasmid pUASLuc (Thiel et al., 2000) and includes a reporter construct composed of 5 Gal4-binding sites and the minimal promoter of the human{tilde over (-)}βGlobin gene followed by the firefly luciferase coding sequence.
pVVS478 was derived from pG4 mpolyII (Webster N J G, 1989) by digestion of the polylinker with Xho and BamH1 and insertion of the genotype 1b HCV's NS5B-Delta 21 sequence in order to be in frame with the carboxy-terminus of the yeast Gal4 DBD (GenBank AY647987). Expression of the fusion protein is under the control of the constitutive SV40 early promoter. NS5B-delta 21 insert was obtained by PCR amplification of pIV294 plasmid using the following oligonucleotide primers:
PCR amplification was carried out for 3 min at 96° C. followed by 35 cycles (30s at 94° C., 30s at 55° C. and 2 min at 72° C.) using a T1 thermocycler (Biometra).
Single mutations in the NS5B-delta21 sequence were introduced using QuickChange II XL Site-Directed mutagenesis kit according to the manufacturer's instructions (Stratagene, Amsterdam, The Netherlands).
The following primers were used to generate the plasmid pVVS474:
The following primers were used to generate the plasmid pVVS475:
The following primers were used to generate the plasmid pVVS517:
The following primers were used to generate the plasmid pVVS516:
The following primers were used to generate the plasmid pVVS789:
The following primers were used to generate the plasmid pVVS795:
The following primers were used to generate the plasmid pWS788:
The following primers were used to generate the plasmid pVVS790:
The following primers were used to generate the plasmid pVVS797:
The following primers were used to generate the plasmid pVVS791:
The following primers were used to generate the plasmid pVVS799:
The following primers were used to generate the plasmid pVVS794:
The following primers were used to generate the plasmid pVVS787:
The following primers were used to generate the plasmid pVVS796:
pVVS602 was derived from pG4 mpolyII (Webster N J G, 1989) by digestion of the polylinker with XhoI and BamH1 and insertion of the genotype 1b HCV NS3's sequence in order to be in frame with the carboxy-terminus of the yeast Gal4 DBD (GenBank AY647987). Expression of the fusion protein is under the control of the constitutive SV40 early promoter. A single chain constructed NS4A (residues 21-32), the GSGS linker and the NS3 protease (residues 3-181) (Taremi et al., 1998) was obtained by PCR amplification of pIV294 plasmid using the following oligonucleotide primers:
PCR amplification was carried out for 3 min at 96° C. followed by 35 cycles (30s at 94° C., 30s at 55° C. and 2 min at 72° C.) using a T1 thermocycler (Biometra).
Single mutations in the NS4A-NS3 protease sequence were introduced using QuickChange II XL Site-Directed mutagenesis kit according to the manufacturer's instructions (Stratagene).
The following primers were used to generate the plasmid pVVS781:
pVVS772 was derived from pG4 mpolyII (Webster N J G, 1989) by digestion of the polylinker with Xho and BamH1 and insertion of the genotype 1b HCV's NS3 full length sequence (GenBank D89872) in order to be in frame with the carboxy-terminus of the yeast Gal4 DBD (GenBank AY647987). Expression of the fusion protein is under the control of the constitutive SV40 early promoter. A single chain constructs NS4A (residues 21-32), the GSGS linker and the NS3 protease (residues 3-631) (Howe et al., 1999) was obtained by PCR amplification of pIV294 plasmid using the following oligonucleotide primers:
PCR amplification was carried out for 3 min at 96° C. followed by 35 cycles (30s at 94° C., 30s at 55° C. and 2 min at 72° C.) using a T1 thermocycler (Biometra).
Single mutations in the NS4A-NS3 full length sequence were introduced using QuickChange II XL Site-Directed mutagenesis kit according to the manufacturer's instructions (Stratagene, Amsterdam, The Netherlands).
The following primers were used to generate the plasmid pVVS783:
pVVS527 was derived from pCINéo (Promega) in order to be the recipient of the selected peptide sequences in frame with frame with the carboxy-terminus of the viral VP16 activation domain (GenBank AF294982).
pVVS550 was derived from the pVVS527 by Sfi I digestion and insertion of the selected peptide N21C272 coding sequence. Expression of the fusion protein is under the control of the constitutive CMV promoter.
The insert was obtained by PCR amplification using the following oligonucleotide primers:
PCR amplification was carried out for 3 min at 96° C. followed by 35 cycles (1 min at 94° C., 1 min at 60° C. and 1 min at 72° C.) using a T1 thermocycler (Biometra).
The naïve yeast was transformed with the pVVS468 (NS5BΔ21 fused to LexA) according to the Lithium Acetate transformation protocol (Schiestl et al, 1989). Transformants were grown in liquid culture under selective pressure (SD medium without Tryptophan), before being transformed with pVVS625 (library of random peptides in frame with the viral VP16 activation domain). Resulting double transformants were spread on SD Agar medium lacking Tryptophan and Leucine, and incubated for 48 hours at 30° C. Colonies were then scraped and mixed. 1/500th of the suspension obtained was plated on SD Agar medium lacking Tryptophan, Leucine and Histidine to identify HIS3 reporter's activation. After 3 to 5 days of growth, an X-Gal overlay was performed to identify LacZ activation. An X-Gal solution was mixed with an equal volume of 1% agarose cooled to 50° C. then poured slowly over yeast colonies on the surface of agar plates. The individual clones that stained blue were picked up to be grown overnight in deep wells in order to be further analyzed.
Yeast colonies containing a peptide interacting with the bait were grown overnight in 96-well deepwells in trp-leu-liquid medium. They were then spotted using a plugger either on trp-leu-his-or trp-leu-solid medium supplemented with different concentrations of specific or non specific drugs or an equivalent amount of solvent (DMSO). Yeast colonies loosing expression of reporter genes were selected. Result was confirmed at least 3 times.
Another method used was a semi-quantitative β-galactosidase liquid assay. Yeast colonies were grown as described above in liquid selective medium with or without the reference drug at different concentrations. The cell pellet was resuspended in 100 ul of Y-PER protein extraction reagent (Pierce, Brebières, France). After 30 nm of shaking at 300 rpm, the supernatant was clarified and the total protein concentration determined using the Micro BCA™ Protein Assay Reagent Kit from Pierce according to the manufacturer's instructions. 5 ul of the protein extract were used to determine Lac Z activity by adding 180 ul of ONPG (O-nitrophenyl-β-D-galactopyranoside) solution at 0.66 mg/ml (Sigma-Aldrich, Lyon, France)). The reaction was stopped as soon as a yellow coloration has developed by adding 75 ul of 1 molar solution of Na2CO3. OD was measured at 420 nm on a spectrophotometer and the specific activity was calculated using the following formula:
Specific activity (nmol/mg/min)=OD420X Vt/(0.0045×Protein×Ve×Time) where Vt is the total volume reaction (160 ul), Protein is the protein concentration of the yeast extract in mg/ml, Ve is the extract volume assayed in ml, and Time is the time of reaction in minutes.
For plasmid extraction, yeast clones whose reporter expression was inhibited by specific drugs were grown in trp-leu-liquid medium. 2 ml of overnight saturated culture were spinned down for 5 min in a microfuge and the pellet was resuspended in 80 ul of lysis buffer supplemented with 20 μl of 2500 Units/ml of Lyticase (Sigma-Aldrich, Lyon, France). The suspension was incubated 1 h at 37° C. with shaking at 220 rpm. 20 μl of 20% SDS were added and the mix was vortexed for 1 min. The mixture was then purified using the Nucleospin Plasmid Quickpure according to the manufacturer's instructions (Macherey Nagel, Hoerdt, France). PCR was performed on the purified eluate using the following primers:
PCR products were sequenced and cloned in the pVVS527 vector as described above. Plasmids were purified and verified by enzymatic restriction and DNA sequencing.
Hela cells (ATCC, Molsheim, France) were maintained in EMEM (Minimum Essential Medium Eagle with Earle's Balanced Salt solution without L-Glutamine; Biowhittaker/Cambrex, East Rutherford, N.J., USA) supplemented with 10% FBS (JRH), 2 mM Glutamine (Biowhittaker/Cambrex), 1% Non-essential Amino acids (Biowhittaker/Cambrex), 100 units/ml Penicillin and 100 ug/ml Streptomycin. Cells were kept in a humidified 5% CO2 atmosphere at 37° C. and dissociated using Trypsine-EDTA (1×) when reaching 80-90% confluence. Huh-7 hepatoma derived cells (Nakabayashi H et al, 1982) were maintained in half volume of DMEM (Dulbecco's Minimum Essential Medium Eagle Medium without L-Glutamine; Lonza, Basel, Switzerland) and half volume of Ham's F12 medium (Lonza, Basel, Switzerland) supplemented with 10% FBS (SAFC, USA), 1 mM Glutamine (Lonza, Basel, Switzerland), 0.5 mM Sodium pyruvate (Lonza, Basel, Switzerland), 100 units/ml Penicillin and 100 ug/ml Streptomycin. Cells were kept in a humidified 5% CO2 atmosphere at 37° C. and dissociated using Trypsine-EDTA (1×) when reaching 80-90% confluence.
Transfections were performed for 2 hours in 100 mm dishes in 2 to 4 million cells with 5 to 20 ug of total DNA complexed to jetPEI (PolyPLus transfection, Illkirch, France) with a N/P ratio of 8 The optimal ratio between the plasmids encoding for the luciferase reporter, the target and the selected 3Dsensor was: 3 ug/1 ug/0.5 ug. Cells were dissociated and 25000 cells per well were seeded in 96 well culture treated plates (BD Biosciences, Erembodegem, Belgium).
Compound library's plates were thawed at room temperature. Drugs were handled by a Genesis workstation (Tecan, Lyon, France) and added to the cells at a 5 uM (in 0.5% DMSO) final concentration. Screening was performed by runs of twenty plates. Hits were identified by measurement of the luciferase activity.
Cell culture supernatant was discarded and 100 ul of Lysis buffer were added to each well. Following 10 nm of vigorous shaking, the lysate was transferred to white opaque 96 well-plates and 100 ul of Luciferin buffer were added. Chemi-luminescence measurement was performed immediately using a Microlumat LB96 luminometer (Berthold, Thoiry, France) with 1 or 10 sec integration time.
Viability measurements were performed on cells transfected and seeded in 96 well plates as described above and incubated with drugs for 24 hours. 0.15 mg/ml of XTT solution (Roche Diagnostics, Meylan, France) were added to each well and the plates incubated for 1 hour at 37° C. The optical density of each well's cell culture supernatant was analysed using a spectrophotometer (at 450 nm with deduction of background signal at 690 nm).
Selected hits were synthesized and tested at concentrations varying between 0.01 uM and 100 uM final concentration so as to determine their IC50 on 3 independent experiments.
4.1—Identification of Conformation Sensitive Peptides Interacting with NS5B-Delta21 in Yeast
Yeast two-hybrid experiments were carried in the L40 strain (ATCC, Molsheim, France) transformed with a lexA/truncated HCV polymerase lacking the C-terminal 21 amino-acids fusion protein as a bait and the VVS001 peptide library as a prey. The C-terminal 21 amino-acid domain anchors the protein to the cell membrane which compromises the feasibility of a two-hybrid approach (Yamashita et al., 1998; Dimitriva et al., 2003). Removal of the 21 amino-acids domain does not compromise the NS5B activity and introduces minor changes in the overall protein conformation (Yamashita et al., 1998; Toméi et al., 2000). A total of 576 colonies were selected for their ability to grow under selective conditions (
Plasmids encoding for the selected peptides were isolated and used to confirm in yeast the ability of each 3D-sensor to bind to the target as well as to study specificities of interaction.
None of the peptides' sequences exhibited any significant homology with known proteins (A BLAST search of the National Center for Biotechnology Information database). All sequences were inserted in frame with VP16 AD in expression vectors to perform the screening of small molecules in mammalian cells.
3D-sensors confirmed in yeast were further investigated in human cells. Among the reasons of choosing Hela cells figured their wide use, easiness of manipulation and capacity of replicating subgenomic HCV replicons (Zhu et al., 2003; Kato et al., 2005). The cells were co-transfected with plasmids expressing: (i) the 3D sensor peptide candidate fused to a trans-activation domain (VP16-AD); (ii) the target polymerase (NS5BΔ21 fused to the yeast Gal4-DNA Binding Domain) and (iii) the luciferase reporter gene whose expression is inducible by the complex [NS5BΔ21/3D-sensor-VP16]. 6 of the selected 3D-sensors supported activation of the luciferase reporter gene in the absence of any ligand (
To further characterize the 3D-Screen platform, the selectivity of binding of peptide N21C272 was investigated. Hela cells were transiently transfected with the luciferase reporter, the 3D-sensor N21C272 and vectors expressing either the native NS5BΔ21 or different mutants (
On the other hand, as shown in
The inventors next confirmed that dissociation of the 3D-Sensor from the polymerase resulted from binding of relevant drugs. Hela cells were transiently transfected with the luciferase reporter, the 3D-sensor N21C272/VP16 and the fusion protein NS5BΔ21-Gal4. Addition of the specific NS5B inhibitors cVVS-023477 and cVVS-023476 induced a strong inhibition of the luciferase signal at 10 uM final concentration (respectively 83.2%+/−7.5; n=15 and 82.6%+/−8.3; n=6) as shown on
To validate whether drug's action on our system was stokiometric or not, Hela cells were transiently transfected with three expression plasmids encoding the reporter gene luciferase, the target protein and the conformation sensitive peptide N21C272. After incubation with increasing amounts of reference drugs cVVS-023476 or cVVS-023477, luciferase expression was measured.
Since the specificity of drugs' action on the 3D-Screen platform could be useful for screening of molecules active on resistant mutants, the inventors transfected Hela cells with the luciferase reporter, the 3D-sensor N21C272 and vectors expressing either native NS5B-Delta21 or different mutants described in the literature. After incubation for 24 hours with increasing concentrations of cVVS023477 (a compound belonging to the Benzothiadiazin family) or cVVS-023476 (a compound belonging to the Indol N Acetamid family), luciferase expression was measured.
These results show that one could generate a cellular platform that is selective to the target's conformation as far as it is not disturbed by destabilising mutations. Furthermore, the resistance mutants platform of the invention allows to discriminate between active and inactive drugs according to the target's resistance profile.
Several other peptides identified with the same technology among which one was named N21 I4 (corresponding to the sequence SEQ ID No 62) was as sensitive as the 3D-Sensor N21C272 to the compounds cVVS-023476 and cVVS-0234777. This 3D-Sensor was used to generate a 3D-Screen-NS5B mutant profiling platform in Hela and Huh7 hepatoma derived cells.
4.3.5) Compatibility of the NS5BΔ21/N21C272-3D-Screen Platform with High Throughput Screening
The assay was further optimized to meet high throughput requirements starting with the signal window.
In order to evaluate the assay's variability, twenty 96-well plates containing cells transfected with the plasmids encoding for the luciferase reporter, NS5BΔ21 and the 3D sensor N21C272 were plated in the presence of 0.5% DMSO for 24 hours in 3 independent experiments. Calculated S/B reached 14+/−0.9 (
The 23510 original compounds issued from VIVALIS proprietary combinatorial library (i.e the Vivatheque) were screened in HeLa cells. The three expression plasmids: luciferase reporter, NS5B-Delta21-Gal4 and 3D-sensor/VP16AD (N21C272) were co-transfected in 4 million Hela cells in 10 cm dishes. After 2 hours, cells were dissociated and 25,000 cells/well were seeded in 96-well plates. Chemical compounds were added after cell adhesion. Luciferase activity was measured 24 H later. 1437 hits yielding 40% of luciferase inhibition were identified during the primary screening (hit rate 6.1%).
342 hits (1.45% of total screened compounds) were confirmed on a second and third round of transfections which represents a confirmation rate of 23.8%. Among those, 215 compounds showed a cytotoxicity less than 40%.
102 of the selected compounds were purchased as pure powders so as to determine their 50% inhibitory concentration. 29 showed interesting dose response profiles independently from cytotoxicity. For 18 molecules, the IC50 value was below 10 uM with a toxicity less than 20% at that concentration. The viral inhibitory potential of 8 of them was confirmed using both the reporter replicon system in Huh7(5,2) and the RT-PCR quantitative replicon system in Huh7(9,13). Three of the molecules identified by our platform were previously patented for their anti-HCV properties. The first one, cVVS 013005 (aminothiazol) was reported in the patent WO 2006/122011, the second one cVVS019534 (thiosemicarbazone) was described in the patent WO 2004/060308 and the third one a thienopyridine was described in the patent WO 2006/019832.
5.1.1) Identification of Conformation Sensitive Peptides Interacting with NS4A-NS3 Protease in Yeast
Yeast two-hybrid experiments were carried in the L40/yDG1 strain transformed with a lexA/HCV NS4A-NS3 protease domain (amino-acids 1-181) fusion protein as a bait and the VVS001 peptide library as a prey. The inventors screened 864 interacting peptides in presence or not of a reference drug cVVS023518 (a peptidomimetic inhibitor in phase 2 of clinical trial (at 50 uM concentration) (Perni et al., 2006). 23 of them were selected for their signal extinction in presence of the reference drug using a quantitative Beta Galactosidase liquid assay.
Plasmids encoding for the selected peptides were isolated and used to confirm in yeast the ability of each 3D-sensor to bind to the target as well as to study specificities of interaction. Amino acid sequences of the 3D-sensor peptides identified with the cVVS023518 were determined None of the peptides' sequences exhibited any significant homology with known proteins (BLAST search of the National Center for Biotechnology Information database). All sequences were inserted in frame with VP16 AD in expression vectors to perform the screening of small molecules in mammalian cells.
5.1.2) Identification of Conformation Sensitive Peptides Interacting with NS4A-NS3 Full Length Protein in Yeast
Yeast two-hybrid experiments were carried in the L40 strain transformed with a lexA/HCV NS4A-NS3 full length (amino-acids 1-632) fusion protein as a bait and the VVS001 peptide library as a prey. We screened 768 interacting peptides in presence or not of a reference drug cVVS023518 (a peptidomimetic inhibitor in phase 2 of clinical trial (at 50 uM concentration) (Perni et al., 2006). 15 of them were selected for their reactivity to the reference drug using a quantitative Beta Galactosidase liquid assay. Plasmids encoding for the selected peptides were isolated and used to confirm in yeast the ability of each 3D-sensor to bind to the target as well as to study specificities of interaction.
Amino acid sequences of the 3D-sensor peptides identified with the cVVS023518 were determined None of the peptides' sequences exhibited any significant homology with known proteins (BLAST search of the National Center for Biotechnology Information database. All sequences were inserted in frame with VP16 AD in expression vectors to perform the screening of small molecules in mammalian cells.
5.2.1) Validation of the 3D-Sensors Interacting with NS4A-NS3Protease
3D-sensors confirmed in yeast were further investigated in Hela cells. The cells were co-transfected with plasmids expressing: (i) the 3D sensor peptide candidate fused to a trans-activation domain (VP16-AD); (ii) the target NS4A-NS3 protease fused to the yeast Gal4-DNA Binding Domain; and (iii) the luciferase reporter gene whose expression is inducible by the complex [NS4A-NS3 protease/3D-sensor-VP16].
4 of the 23 tested 3D-sensors supported activation of the luciferase reporter gene in the absence of any ligand (
The 3D-sensor V7-62 was selected to generate the 3D-Screen platform on the basis of (i) signal intensity, (ii) inhibition of the signal in the presence of cVVS-023518 (90% luciferase inhibition at 2 uM final concentration).
5.2.2) Validation of the 3D-Sensors Interacting with NS4A-NS3Full Length Protein
3D-sensors confirmed in yeast were further investigated in Hela cells. The cells were co-transfected with plasmids expressing: (i) the 3D sensor peptide candidate fused to a trans-activation domain (VP16-AD); (ii) the target NS4A-NS3 full length protein fused to the yeast Gal4-DNA Binding Domain; and (iii) the luciferase reporter gene whose expression is inducible by the complex [NS4A-NS3 full length protein/3D-sensor-VP16].
8 of the 16 identified 3D-sensors supported activation of the luciferase reporter gene in the absence of any ligand (
The inventors confirmed that dissociation of the 3D-Sensor from the NS3 target resulted from binding of relevant drugs and not due to specific events. Hela cells were transiently transfected with the luciferase reporter and each of the NS5B and NS3 targets together with their specific 3D-Sensor. Addition of increasing concentrations of the specific NS5B inhibitor cVVS-023476 (Indol-N-Acetamide) induced a strong inhibition of only the NS5B platform. Also, the specific NS3 inhibitor cVVS-023518 inhibited the NS4A-NS3 protease and full length platforms without significant off target effects (
Since the specificity of drugs' action on the 3D-Screen platform could be useful for screening of molecules active on resistant mutants, the inventors transfected Hela cells with the luciferase reporter, the 3D-sensor V7-62 and vectors expressing either native NS4A-NS3 protease or one of the most frequently described mutants resistant to VX950 described in the literature (Lin et al., 2005). The interaction between the 3D-sensor V7-62 and either the native or A156V mutated NS4A-NS3 protease led to a strong activation of the luciferase reporter even if it is lower for the mutated form (33 times the background luciferase activity without the 3D sensor-VP16AD versus 80 times for the native one) (
Several other peptides identified with the same technology among which one was named T1E7 (corresponding to the sequence SEQ ID No 63) were as sensitive as the 3D-sensor V7-62 to Telaprevir and Ciluprevir. This 3D-sensor was used to generate 3D-Screen-NS4A-NS3 protease mutant profiling platform.
The inventors transfected Hela cells with the luciferase reporter, the 3D-sensor H5-34 and vectors expressing either native NS4A-NS3 full length protein or one of the most frequently described mutants resistant to the peptidomimetic described in the literature (Lin et al., 2005). The interaction between the 3D-sensor H5-34 and either the native or A156V mutated NS4A-NS3 full length protein led to a strong activation of the luciferase reporter (42 times the background luciferase activity without the 3D sensor-VP16AD for the native protein versus 56 times for the mutated one) (
These results show that one could generate a cellular platform that is selective of the target's conformation as far as it is not disturbed by destabilising mutations. Furthermore, the resistance mutants platform of the invention could discriminate active and inactive drugs according to the target's resistance profile.
The inventors transfected Huh7 hepatoma cells with the luciferase reporter, the 3D-sensor VF-9A11 (SEQ ID No 64) isolated in yeast and vectors expressing either native NS4A-NS3 full length protein or one of the most frequently described mutants resistant to the peptidomimetic described in the literature (Lin et al, 2005). The interaction between the 3D-sensor VF-9A11 and either the native or mutated NS4A-NS3 full length protein led to a strong activation of the luciferase reporter (signal over background far above three except for the R155K mutated protein) as shown in
The interaction of the VF-9A11 peptide with the target variants enabled to generate as many 3D-Screen platforms on which the peptidomimetic compound cVVS-023518 was tested. As shown in
The A156S, T54A and V36M, known to be moderately resistant to the peptidomimetic cVVS-023518, also display moderate fold increase in EC50 values by comparison with the wild type NS3 on the 3D-Screen platform. As depicted in
These results support the notion that the 3D-Screen technology could be exploited to develop a profiling platform to predict the spectrum of activity of a given drug against clinically described resistant mutants. Such a profiling platform would allow the ranking of the isolated hits according to their range of anti-viral activities, enhancing at an early stage the chance of selecting the most promising molecules.
The 3D-Screen mutant profiling platform relevance was further challenged with a second peptidomimetic inhibitor cVVS-23591 described to be active on the protease in the low nanomolar range and validated in phase 1 of clinical trials (Lamarre et al, 2003). As shown on
The highly resistant mutants A156V and D168V (Lin et al, 2004 & 2005) showed a resistance in the same range as what was previously reported on enzymatic or replicon assays as depicted in
Several other peptides identified with the same technology and named VFII-N4, VFII-N13, T4C11 and T3G5 (respectively corresponding to the sequences SEQ ID No 65, 66, 67 and 68) were as sensitive as the 3D-Sensor VF-9A11 to Telaprevir and Ciluprevir. All these 3D-Sensors were used to generate 3D-Screen-NS4A-NS3 full length mutant profiling platforms.
Thus, the present results demonstrate the generation of a robust and reliable profiling platform to predict the spectrum of activity of a given drug against clinically described resistant mutants. The 3D-Screen mutant profiling platform has been developed and validated for the NS5B polymerase with two different inhibitors. It has been also developed for the NS3 protein and validated with two reference protease inhibitors. Its resistance phenotyping was accurate, predictive and correlated with results reported in the literature with enzymatic or replicon based assays. In addition, the cell-based 3D-Screen platform has the advantages of rapidity and simplicity of implementation as well as high throughput.
HCV variants are classified into six genotypes (from 1 to 6) associated with a lower case letter to indicate the subtype (Le Guillou-Guillemette et al., 2007). Since sensitivity to drugs could vary from one HCV genotype to the other (Manns et al., 2007), the investors tested whether the 3D-Screen platform could be able to discriminate such differences which is highly important for preclinical profiling of drug candidates. The inventors thus verified that the 3D-Sensor N21I4 did interact with the polymerase NS5B from genotypes 1a and 1b in Hela and Huh7 cells in an equivalent way.
Huh-7 cells were transiently transfected with three expression plasmids encoding the reporter gene luciferase, the target NS5B from genotype 1a or 1b and the conformation sensitive peptide N21I4. After incubation with increasing amounts of cVVS-023476, luciferase expression was measured.
These results show that it is possible to generate a cellular platform that is selective of the target's conformation even if issued from different genotypes. Furthermore, the screening platform of the invention could be used to discriminate between active and less active drugs according to the target's genotype background. The same principle was applied to the targets NS4A-NS3 protease and NS4A-NS3 full length proteins.
| Number | Date | Country | Kind |
|---|---|---|---|
| 08300146.1 | Mar 2008 | EP | regional |
| Filing Document | Filing Date | Country | Kind | 371c Date |
|---|---|---|---|---|
| PCT/EP2009/053101 | 3/16/2009 | WO | 00 | 10/26/2010 |
