WO2017182437A1

WO2017182437A1 - Quantification of an interaction between a ras protein and a raf protein

Info

Publication number: WO2017182437A1
Application number: PCT/EP2017/059123
Authority: WO
Inventors: Eduard STEFAN; Verena BACHMANN; Philipp RAFFEINER; Ruth RÖCK
Original assignee: Universität Innsbruck
Priority date: 2016-04-18
Filing date: 2017-04-18
Publication date: 2017-10-26

Abstract

The present invention provides a method for quantification of an interaction between a Ras protein and a Raf protein, comprising the steps of a) providing a cell expressing two constructs for a protein-fragment complementation assay, wherein the first construct comprises the Ras protein and a first luciferase protein fragment, wherein the first luciferase protein fragment is located C-terminal from the Ras protein and wherein the second construct comprises the Raf protein and a second luciferase protein fragment, wherein the second luciferase protein fragment is located C-terminal from the Raf protein, wherein the second construct comprises a full-length Raf sequence section, and wherein the first and the second luciferase protein fragments are derived from different sections of the same luciferase protein b) allowing the first and second constructs to contact each other, c) providing a bioluminescence substrate for an intact luciferase protein and d) measuring for bioluminescence, wherein said bioluminescence indicates interaction between the two constructs upon reassembling of the first and second luciferase protein fragments.

Description

Quantification of an interaction between a Ras protein and a Raf protein

The invention relates to a method for quantification of an interaction between a Ras protein and a Raf protein, a method for measuring an effect of a candidate compound on the interaction between Ras and Raf as well as nucleotides and cell lines suitable for such methods.

Background of the invention

Raf kinases are a kinase family involved in the cellular signalling pathway known as MAPK (Mitogen-Activated Protein Kinase) signalling cascade. As MAP-3K kinases (MAP kinase kinase kinases) Raf kinases catalyse the phosphorylation of MEK1 and MEK2 (Mitogen/Extracellular signal-regulated Kinases) which subsequently are regulating MAP kinases. The nomenclature of Raf kinases originates from "rapidly accelerated fibrosarcoma" and highlights the important association of the protein family to tumor geneses and progression. Three human Raf kinases are known as ARaf, BRaf and CRaf (or RAF-1). A variety of activating BRaf mutations have been described and occur frequently in e.g. thyroid and colorectal cancers and in around 60% of melanomas. It is the BRaf-V600E mutation which is the most recurrent oncogenic disease mutation of the Raf isoforms (Lito, P., Rosen, N. & Solit, D.B. Tumor adaptation and resistance to RAF inhibitors. Nat Med 19, 1401-9 (2013)). It is believed that this amino acid exchange serves as phospho-mimetic substitution in the BRaf kinase domain (Wan, P.T. et al. Mechanism of activation of the RAF-ERK signaling pathway by oncogenic mutations of B-RAF. Cell 116, 855-67 (2004)), which creates catalytically active BRaf. Consequently, substantial endeavours have been initiated to generate a collection of clinical effective ATP-competitive Raf-inhibitors (Lito, P., Rosen, N. & Solit, D.B. Tumor adaptation and resistance to RAF inhibitors. Nat Med 19, 1401-9 (2013); Girotti, M.R., Saturno, G., Lorigan, P. & Marais, R. No longer an unbeatable disease: how targeted and immunotherapies have changed the management of melanoma patients. Mol Oncol 8, 1140-58 (2014)).

Besides the kinase domain Raf kinases have N-terminal regulatory domains. The so called Ras binding domain (RBD) is supposed to interact with GTP-bound Ras protein to relief auto- inhibition by the regulatory domains (Fig. 1 A). Ras proteins are membrane associated small G-proteins. Three family members are very well characterized in human cells, i.e. HRas, NRas, KRas, but other family members receive increasing attention (Karnoub A.E., Weinberg R.A. Ras oncogenes: split personalities. Nat Rev Mol Cell Biol 9, 517-531 (2008)). These small GTPase Ras proteins are the most frequently mutated oncogene in human cancers. A Ras GTPase is a molecular switch which is active in the GTP-bound form. Functional GTPases contain intrinsic GTPase-activity and switch to inactive GDP-bound forms through GTP hydrolysis. Alterations in the Ras-gene affect Ras-protein function through inhibition of the GTP-release. Therefore different Ras mutations cause that the Ras-Raf-Erk pathway is stuck in the ON-state leading to aberrant cell growth and proliferation. Consequently Ras- GTP protein overexpression and amplification can lead to continuous cell proliferation, which is a major step in the development of cancer. The conversion of Ras from a proto-oncogene into an oncogene usually occurs through different point mutations. Mutant Ras has been identified in cancers of many different origins, including: pancreas (90%), colon (50%), lung (30%), thyroid (50%), bladder (6%), ovarian (15%), breast, skin, liver, kidney, and some leukemias.

Whereas the small G-protein Ras was not yet successfully explored for drug development, mutated BRaf is an established drug target. Kinase inhibitors that specifically address the BRaf-V600E mutant are for example the drugs vemurafenib, dabrafenib or encorafenib which are approved or in development for treatment of cancers such as melanoma. However, the duration of the response is variable and the efficacy of the inhibitor is limited through the onset of drug resistance. Diverse mechanisms of drug resistance have been proposed involving enhanced receptor tyrosine kinase activities or other means of reactivating the MAPK pathway. Moreover, it has been shown that kinase inhibition depends on the cellular context. Dependent on the genetic background (oncogene mutations), approved and structurally diverse ATP-competitive BRaf inhibitors have the potential to reactivate Ras- dependent MAPK pathways which leads to the progression of cancer cell proliferation (Lito, P., Rosen, N. & Solit, D.B. Tumor adaptation and resistance to RAF inhibitors. Nat Med 19, 1401-9 (2013); Nazarian, R. et al. Melanomas acquire resistance to B-RAF(V600E) inhibition by RTK or N-RAS upregulation. Nature 468, 973-7 (2010); Heidorn, S.J. et al. Kinase-dead BRAF and oncogenic RAS cooperate to drive tumor progression through CRAF. Cell 140, 209-21 (2010); Poulikakos, P.I. et al. RAF inhibitor resistance is mediated by dimerization of aberrantly spliced BRAF(V600E). Nature 480, 387-90 (2011); Hatzivassiliou, G. et al. RAF inhibitors prime wild-type RAF to activate the MAPK pathway and enhance growth. Nature 464, 431-5 (2010)).

These observations underline that there is a need for complementary approaches to analyze lead molecules targeting the GTP- and drug-dependent complex formation of the different full length Ras:Raf proteins of the Ras and Raf protein families. The status quo for molecular analyzes of Ras:Raf interactions and their direct GTPase and kinase activities are biochemical assays which involves disruption of cells or tissues. Biochemical kinase assays are used to analyze ATP-competitive inhibitors, the drug of choice to hamper kinase activities. Biochemical approaches are used to track GTP-loading of Ras. Also the concept of protein-fragment complementation has been suggested to study the Ras:Raf interaction. With a protein-fragment complementation assay (PCA) a protein-protein interface, i.e. a molecular interaction, can be investigated by fusing the potentially interacting partners to different fragments of a reporter protein, wherein a reporting signal results from assembly of the fragments in case the protein-protein interaction of the interacting partners is established. US 2015/0010932 describes the general concept of PCA and discloses a method, wherein the report protein is obtained by introducing a nucleotide sequence into a host cell by cre-recombinase mediated cassette exchange (RCME). The Ras:Raf interface is mentioned as a potential application field without disclosing a respective assay for this interface. The abstract "The small molecule compound, MCP1, inhibits Ras-Raf interaction in mammalian cells and induces apoptosis in various hematopoietic cancer cells" by Juran Kato-Stankewicz et al. was published in: Proceedings of the 95th Annual Meeting of the American Association for Cancer Research; 2004 March 27-31, 2004; Orlando FL. Abstract Nr 5627. Therein, the authors refer to a split Renilla luciferase fusion protein-based bioluminescence assay to study the interaction of Ras with the Ras binding domain (RBD) of Raf. The N-terminal luciferase split protein was fused to Ras (NLuc-Ras) and the Ras-binding domain (RBD, 51-131 a.a.) of Raf fused to a C-terminal luciferase split protein (RafRBD-CLuc). The constructs NLuc- Ras(L61) and RafRBD-CLuc, coexpressed in COS7 cells, showed luciferase activity. However, an investigation of allosteric drug effects on cellularly expressed GTPase:kinase complexes is not possible with the known PCA assay and immunoprecipitation experiments are still the method of choice. Thus, it is of high interest to provide an assay that allows for analysing the protein-protein interaction (PPI) of both wild type and mutated Ras GTPases and Raf kinases within the appropriate cellular context.

Description of the invention

The present invention provides a method for quantification of an interaction between a Ras protein and a Raf protein, comprising the steps of

a) providing a cell expressing two constructs for a protein-fragment complementation assay,

wherein the first construct comprises the Ras protein and a first luciferase protein fragment, wherein the first luciferase protein fragment is located C-terminal from the Ras protein, and

wherein the second construct comprises the Raf protein and a second luciferase protein fragment, wherein the second luciferase protein fragment is located C-terminal from the Raf protein, wherein the second construct comprises a full-length Raf sequence section and wherein the first and the second luciferase protein fragments are derived from different sections of the same luciferase protein;

b) allowing the first and second constructs to contact each other;

c) providing a bioluminescence substrate for an intact luciferase protein; and

d) measuring for bioluminescence, wherein said bioluminescence indicates interaction between the two constructs upon reassembling of the first and second luciferase protein fragments.

The method is based on protein-fragment complementation assay (PCA), for which the general concept is for example described in EP 0 966 685 Bl. In the method according to the present invention the reporter protein is a luciferase protein, for example a Renilla luciferase (Rluc) protein. Each of the two construct comprises another fragment of the luciferase protein, and if the interaction between the two constructs establishes a functional unit between the luciferase fragments assembles (Fig. 1 B). The functional luciferase enzyme produces a bioluminescence signal upon oxidation of the substrate.

The inventors have found that Raf coupled to a first fragment of a Renilla luciferase protein interacts preferentially with Ras variants including oncogenic mutations coupled to a second fragment of the Renilla luciferase protein thereby increasing the PCA originating bioluminescence signal substantially. Similar results were obtained for constructs with the RBD domain of Raf and full-length Raf. Without wishing to be bound by theory, it is believed that the GTP-dependent character enables the delegability via a PCA assay. The G12V mutation in HRas, KRas and NRas increases the cellular amount of activated GTP-bound Ras proteins (Fig. 2 A). Thus, the findings allow for the provision of a specific PCA assay which enables to evaluate and quantify the GTP-dependent Ras:Raf interaction in any cellular system via measuring bioluminescence signals. This is supported as interaction studied with wild-type Ras did not provide a significant interaction signal using the PCA assay as read out. The Ras:Raf interaction seems to be less established with the wild type, whereas Ras variants with oncogenic mutations resulted in elevated relative luminescence units (RLU) indicating increased interaction. This observation is in agreement with the hypothesis that these constitutively active Ras mutants upregulate MAPK activities and signalling through increased interaction with Raf kinases.

With the present cell-based method, the invention allows to investigate cellular complex formation of Ras:Raf in the context of full-length Raf proteins (Fig. 2 B). Primarily, either the Raf-RBD or the kinase domains are studied using distinct read outs for kinase activities and interactions and previous Raf:Ras PCA-experiments were limited to constructs of the Raf- RBD. With the present invention also allosteric effects of diverse but interacting (macro)molecules on the Ras:Raf interaction can be observed. The cell-based Ras-Raf PPI reporter allows the direct analyses of full length kinase:GTPase complexes in vivo, i.e. in the cellular context.

The invention provides a method for measuring an effect of a candidate compound on the interaction between Ras and Raf, wherein the method according to the invention is conducted in presence of the candidate compound and the effect of the candidate compound on the interaction is determined by comparing the bioluminescence as measured in step d) in presence of the candidate compound versus the bioluminescence in absence of the candidate compound.

The investigation of BRaf inhibitors shows how the defined cell-based assay can be applied to track efficacies and OFF-target effects of lead molecules targeting the Ras-Raf-Erk cascade. Moreover, the PPI reporter can be subjected to draw drug profiles in the appropriate cell setting or model organism. First of all, this unique PPI reporter will ease the selection process of lead molecules through analyses of allosteric consequences on selective Ras:Raf interactions at an early stage in the drug discovery pipeline. Second, this modular reporter toolbox will help to determine efficacies and OFF-target effects of drugs acting on oncogenic components of the Ras-Raf-Erk cascade. Third, besides drug screenings the impact of patient mutations and post-translational modification can be studied using this versatile PPI reporter platform. Given that many kinase inhibitors target structurally-related catalytic clefts of kinases the Ras-Raf reporter serves as unique cellular PPI reporter to determine kinase drug specificity and efficacy.

In another aspect, the invention provides nucleotides that are suitable for generation of the protein constructs and thus, for the respective methods and for establishing cell lines that can be used according to the invention.

The term "construct" as used according to the invention refers to a protein comprising different sequence sections from different origins. These sequence sections are fused together, thus they form a single amino acid chain within the hybrid protein. The fusion is achieved by providing a respective engineered fusion nucleotide which may be translated to the construct. The term "construct" may be used synonym to the term fusion protein, hybrid protein, or chimeric protein. The wording does not imply that the fused sequence sections are directly coupled to each other and it does not imply a specific ordering of the sequence sections to each other. Moreover, there is no need that the individual sequences comprise in the construct are full-length protein sequences, it may also refer to partial sequences such as e.g. domains.

In the following preferred embodiments of the constructs are described, which are useful in a method of quantification of an interaction between a Ras protein and a Raf protein as well as a method for measuring an effect of a candidate compound on this interaction. The specific features refer to the protein construct; however, it is clear for a person skilled in the art that these features may be also preferable for nucleotide sequences encoding for the respective protein as such polynucleotides provide useful starting material for the constructs as used in a method according to the invention.

The term "Ras protein" as used according to the invention refers to a protein selected from the family of Ras proteins. The term comprises small GTP binding proteins like the well investigated HRas, NRas and KRas (including the two isoforms of KRas) as well as the proteins Rapl, Rap2, Ritl, Rit2, RalA, RalB, ERas, RRas, MRas und TC21. Ras proteins are known to a the person skilled in the art (Ras oncogenes: split personalities. Karnoub AE1, Weinberg RA. Nat Rev Mol Cell Biol. 2008 Jul;9(7):517-31. doi: 10.1038/nrm2438). Especially, the two KRas isoforms (SEQ ID No: 1 and SEQ ID No: 2), NRas (SEQ ID No: 3), HRas (SEQ ID No: 4), and the two KRas isoforms are known for variants having oncogenic mutations of the residues G12, G13 and Q61 mutants. G12 and G13 may be substituted with A, C, D, R, S or V and Q61 with E, K, L, P or R. In a method according to the invention increased bioluminescence signals were observed for constructs with Ras proteins with oncogenic mutants, which are known to increase the fraction ON-state of Ras proteins i.e. GTP bound Ras. Such mutations are for example G12V/C or Q61L mutants of HRas, KRas or NRas. Therefore, in a preferred embodiment the Ras protein may have a sequence of a mutated Ras protein (e.g. SEQ ID No: 5 to 24), especially KRas-G12V, NRas-G12V or HRas- G12V (SEQ ID No: 5, 10, 15, 20).

The term "Raf protein" as used according to the invention refers to proteins from the family of Raf kinases including ARaf (or A-Raf, SEQ ID No: 25), BRaf (or B-Raf, SEQ ID No: 26), and CRaf (C-Raf, Raf-1, v-Raf, SEQ ID No: 27) with their isoforms and mutations. This includes the constantly growing collection of Raf mutations which deregulate Raf signalling such as the amino acid substitutions in BRaf-V600K (SEQ ID No: 32), BRaf-K601E (SEQ ID No: 33) and generally modifications of distinct amino acids in the region of BRaf 590-601. Other examples for activating mutations are CRaf-S257L (SEQ ID No: 34), CRaf-S259A (SEQ ID No: 35), and ARaf-S214C (SEQ ID No: 31). Herein disclosed are also Raf constructs, wherein only a domain of the full-length protein e.g. the RBD domains (SEQ ID No: 28 to 30) is comprised. However, in these later constructs the Raf protein within the second construct does not comprise a full-length Raf sequence section. Examples with such Raf-RBD constructs are included for comparison.

It is expected that the method allows a sequence variability regarding the Ras or Raf protein sequences comprised in the constructs. Thus, the up to 5% of the sequence of the Ras or Raf protein may differ from the native human sequences. This may allow investigating also the proteins of other species. Functional motives, especially those involved in the protein interface of Ras and Raf may be identified with the method of the invention and should be preferably conserved to study other effects of sequence variability. The 5% tolerance regarding the sequence diversity should also be understood as covering sequence omissions such as, e.g. gaps within the sequence and/or truncations at the ends, preferably terminal truncations.

Thus, in one embodiment, the Ras protein has a sequence identity of at least 95 % to a sequence selected out of the group consisting of SEQ ID No: 1 to SEQ ID No: 4.

In another embodiment, the Raf protein has a sequence identity of at least 95 % to a sequence selected out of the group consisting of SEQ ID No: 25, 26, 27, 31, 32, 33, 34, 35, 36.

Moreover, it may be preferred that the Ras protein has a sequence identity of at least 98 % to a sequence selected out of the group comprising SEQ ID No: 1 to SEQ ID No: 24. Preferably the Ras protein has a sequence selected out of the group comprising SEQ ID No: 1 to SEQ ID No: 24, preferably SEQ ID No: 1, 2, 3, 4, 5, 10, 15 and 20, in particular preferred SEQ ID No: 5, 15, or 20. Similarly, it may be preferred that the Raf protein has a sequence identity of at least 98 % to a sequence selected out of the group comprising SEQ ID No: 25 to SEQ ID No 27. Preferably, the Raf protein has a sequence selected out of the group comprising SEQ ID No: 25 to SEQ ID No: 27 and SEQ ID No: 31 to SEQ ID No: 35, preferably SEQ ID No: 26 and SEQ ID No: 32.

It was shown that the PCA assay was functional with the Raf containing constructs comprised a Ras binding domain (RBD, SEQ ID No: 28 to SEQ ID No: 30) as well as constructs comprising a full-length Raf sequence section (SEQ ID No: 25 to 27 and SEQ ID No: 31 to 35). The second construct according to the invention comprises a full-length Raf sequence section which allows to study allosteric effects. Especially, the BRaf sequence sections and therein the ones with oncogenic mutation V600E are preferred due to their pathological importance. Thus, in a preferred embodiment the Ras protein comprises the sequence of full- length BRaf (SEQ ID No: 26) or its mutant V600E (SEQ ID No: 32), preferably BRAF V600E.

The constructs according to the invention comprise an N-terminal sequence section comprising a Ras or Raf sequence as described above and C-terminal section being derived from a lucif erase protein. Additionally, it is consider suitable that the construct comprises a linker intervening between the N-terminal sequence section of the Ras or the Raf protein and the C-terminal luciferase protein fragment. Various linkers are known and could be applied equivalently. Preferably, the sequence of the first construct is composed of a Ras protein sequence, a linker sequence and a luciferase protein fragment consecutively arranged in this order from the N-terminus to the C-terminus of the construct. Preferably, the sequence of the second construct is composed of a Raf protein sequence, a linker sequence and a luciferase protein fragment consecutively arranged in this order from the N-terminus to the C-terminus of the construct. Preferably linker sequences comprise small and flexible residues such as glycine and/or serine residues. Linker sequences may comprise for example 5 to 50, 5 to 20, 8 to 16 or 10 to 12 residues. A linker sequence may for example be GGGGSGGGGS or similar flexible regions with amino acids with no or small side chains.

In a preferred embodiment of the invention achieved via a glycine rich linker, preferably the linker has a sequence according to SEQ ID No: 36. The term "luciferase fragment protein" refers to a protein fragment, wherein the sequence of this protein is derived from a section of full-length reporter protein. The Renilla luciferase fragments, i.e. derived from a Renilla luciferase (Rluc) sequence, turned out to be valuable reporter protein fragments. In addition to Rluc other lucif erases may be used to generate a functional Ras:Raf reporter platform according to the invention. Any luciferase based PCA reporter enzyme might be applicable to generate a functional Ras:Raf interaction reporter according to the invention. Preferably the luciferase protein fragment is derived from a luciferase selected out of the group consisting of Renilla luciferase, Gaussia luciferase, firefly luciferase, and Nanoluc. The optimization of Rluc fragmentation point for another PCA assay has been described before. It is preferred that the fragmentation of the Renilla luciferase signal applies after residue 110 (Stefan, E. et al., Proc Natl Acad Sci U S A. 2007 Oct 23; 104(43): 16916-16921). The fragments are derived from the N-terminal residues 3 to 110 or the C-terminal residues 111 to 311 of the sequence of native luciferase from Renilla reniformis, respectively. SEQ ID No: 37 comprises a 109 residue beginning with an additional alanine residue located N-terminally and SEQ ID No: 38 comprises a 201 residues sequence from the C-terminal part.

In a preferred embodiment the method is characterized in that the luciferase protein fragments are derived from Renilla luciferase. More preferably, the first and the luciferase protein fragments have a sequence identity of at least 95 % to one of the sequence out of SEQ ID No: 37 or SEQ ID No: 38.

Mutations in the native sequence of Rluc may be tolerated or even beneficial for bioluminescence activity of the reassembling fragments. Thus, 95% sequences identity with the native Rluc sequence may be sufficient for the fragments. Generally, it did not turn out to be critical to combine a specific construct with either the first or the second Rluc fragment (data not shown). In one embodiment, it may a be preferred embodiment that the Ras protein is coupled to SEQ ID No: 37 and the Raf protein is coupled to SEQ ID No: 38. Thus, a Ras protein construct used as first construct according to the invention may exemplarily have a sequence as SEQ ID No 38: to SEQ ID No: 44. A Raf protein construct as used as second construct according to the invention may exemplarily have a sequence as SEQ ID No: 45 or SEQ ID No: 46. The term "luciferase substrate" refers to so called luciferins, which are compounds that are oxidized by an active luciferase to form a light emitting molecule. The luciferase substrate provided in step c) may be for example a selected out of the chemical group Coelenterazine compounds, which are also referred to as CTZ or CLZN. Suitable examples may be benzylcoelenterazine (also known as coelenterazine h, 2,8-dibenzyl-6-(4- hydroxyphenyl)imidazo[l,2-a]pyrazin-3(7H)-one, CAS: 50909-86-9). This substrate of Renilla luciferase may be preferred in combination with constructs comprising SEQ ID RlucFl and RlucF2. Alternatively, native coolenterazine may be used (6-(4-hydroxyphenyl)- 2-[(4-hydroxyphenyl)methyl]-8-(phenylmethyl)-7H-imidazo[3,2-a] pyrazin-3-one, CAS: 55779-48-1), as well as other luciferins from the Colelenterzine class including the non-native derivates, e.g. Coelenterazine 400a (Bisdeoxycoelenterazine, 2,8-dibenzyl-6-phenyl- imidazo[l,2A]pyrazin-3-(7H)-l, CAS 70217-82-2), e-Coelenterazine (Coelenterazine-E, Benz[f]imidazol[l,2-a]quinoxalin-3(6H)-one,5,l l-dihydro-8-hydroxy-2-[(4-hydroxyphenyl- methyl]-12-(phenylmethyl), CAS: 114496-02-5), Coelenterazine-Fluoride (Coelenterazine F, 8-benzyl-2-(4-fluorobenzyl)-6-(4-hydroxyphenyl)imidazo[l,2- a]pyrazin-3(7H)-one, CAS: 123437-16-1), e-Coelenterazine-F (Benz[f]imidazol[l,2-a]quinoxalin-3(6H)-one,5,l l- dihydro-8-hydroxy-2-[(4-fluorophenyl-methyl]-12-(phenylmethyl)), v-Coelenterazine (Coelenterazine-v, 16-benzyl-5-hydroxy-13-[(4-hydroxyphenyl)methyl]-l l,14,17- triazatetracyclo[8.7.0.0^A{2,7}.0^A{ l l,15}]heptadeca-l(10),2(7),3,5,8,13,15-heptaen-12-one), Coelenterazine hep (2-benzyl-8-(cyclopentylmethyl)-6-(4-hydroxyphenyl)imidazo[ 1 ,2- a]pyrazin-3(7H)-one CAS: 123437-32-1), Coelenterazine cp (8-(cyclopentylmethyl)-2-(4- hydroxybenzyl)-6-(4-hydroxyphenyl)imidazo[l,2-a]pyrazin-3(7H)-one, CAS: 123437-25-2), Coelenterazine fcp (8-(cyclopentylmethyl)-2-(4-fluorobenzyl)-6-(4-hydroxyphenyl)imidazo- [l,2-a]pyrazin-3(7H)-one CAS: 123437-33-2), Coelenterazine ip (8-(isopropylmethyl)-2-(4- hydroxybenzyl)-6-(4-hydroxyphenyl)imidazo[l,2-a]pyrazin-3(7H)-one). These compounds may be used in step c) of a method according to the invention as long as they are substrate of the intact luciferase formed by the fragments.

A "candidate compound" could be any (macro)molecule for which it is of interest to study the influence on the Ras:Raf interaction. Thus, the method may be applied to screen for compounds influencing the Ras:Raf interaction. In such situation the candidate compound may be any compound, preferably a small organic molecule. Alternatively, the compound may be a compound which is known to interact with either Ras or Raf in order to study allosteric effects on the Ras:Raf interaction.

In a preferred embodiment or the invention the candidate compound interacts with the Raf protein, preferably the candidate compound is a known inhibitor of the Raf kinase activity.

Regarding the polynucleotides provided according to the invention a first polynucleotide encoding for a fused protein construct is characterized in that the construct comprises the Ras protein and a first luciferase protein fragment, wherein the first luciferase protein fragment is located C-terminal from the Ras protein, wherein preferably the Ras protein has a sequence identity of at least 95 % to a sequence selected out of the group comprising SEQ ID No: 1 to SEQ ID No: 24 and wherein preferably the luciferase protein fragment has a sequence identity of at least 95 % to one of the sequence out of SEQ ID No: 37 or SEQ ID No: 38.

Another polynucleotide encoding for a fused protein construct is characterized in that the construct comprises the Raf protein and a second luciferase protein fragment, wherein the second luciferase protein fragment is located C-terminal from the Raf protein and wherein the Raf protein has a sequence identity of at least 95 % to a sequence selected out of the group comprising SEQ ID No: 25 to SEQ ID No: 27 and wherein preferably the luciferase protein fragment has a sequence identity of at least 95 % to the sequence out of SEQ ID No: 37 or SEQ ID No: 38. Preferably, the Raf protein sequence encoded by the polynucleotide is a sequence selected out of the group comprising SEQ ID No: 25, 26, 27, 31, 32, 33, 34, 35, preferably SEQ ID No: 26 or SEQ ID No: 32.

The term "polynucleotide" is to be understood synonym to oligonucleotide, and nucleic acid single- stranded and double- stranded polymers of nucleotide monomers, including 2'- deoxyribonucleotides (DNA) and ribonucleotides (RNA). The person skilled in the art may derive the respective RNA or DNA sequence of a polynucleotide easily from the protein construct for which is should encode. The general approach for generating a polynucleotide with a luciferase fragment for a PCA assay was described previously (Stefan, E. et al. Quantification of dynamic protein complexes using Renilla luciferase fragment complementation applied to protein kinase A activities in vivo. Proc Natl Acad Sci U S A 104, 16916-21 (2007)).

Preferably, a polynucleotide according to the invention for providing a Ras protein construct encodes for a sequence selected out of the group consisting of SEQ ID No 38: to SEQ ID No: 44. A polynucleotide according to the invention for providing a Raf protein construct encodes preferably for a sequence selected out of the group consisting of SEQ ID No: 45 or SEQ ID No: 46.

Moreover, the invention provides a cell comprising the nucleotides according to the invention.

Thus, a cell according to invention is capable of expressing both constructs useful for a method according to the invention. The polynucleotides may be transfected to any modified cancer cell line. In one embodiment the cell line may be a melanoma cell line. The procedure of obtaining a cell according to the invention is given in the detailed description. Exemplary, the cell according to the invention may be derived from an established cell line such as a cell line selected out of the group comprising HEK293, SW480 and U20S. Preferably, a cell according to the invention is derived from a HEK293 cell line.

Detailed description of the invention The invention will now be described in more detail by the following figures and non-limiting examples.

The figures show: Figure 1A shows the activation mechanism of Raf kinases via the membrane associated GTP binding Ras protein as schematic drawing. The Raf kinase protein is composed of a N- terminal Ras binding domain (RBD), the cysteine rich domain (CRD) and the C-terminal kinase domain (KD). Figure IB shows the principle of the specific protein-fragment complementation assay (PCA) for Raf and Ras, wherein Raf is fused with Renilla luciferase fragment 1 (F[l]) and Ras is fused with Renilla luciferase fragment 2 (F[2]). Following Ras:BRaf interaction (right side of the figure) folding and complementation of Renilla luciferase (Rluc) fragments is initiated leading to a functional Rluc. The quantification of the resulting PCA signal correlates with the amount of formed protein-protein interactions (PPIs). The signal can be measured as relative luminescence (or light) units resulting from turnover of the luciferase substrate benzylcoelenterazine.

Figures 2A, 2B, 2C show results from comparative PCA experiments with the expression constructs of the Ras binding domain (RBD) of BRaf in (Fig. 2A) and the full-length BRaf (Fig 2B and Fig. 2C) fused to F[l] versus experiments either with wild type expression constructs of Ras proteins K/N/H-Ras (wt) or the one with the oncogenic Ras-G12V mutations fused to F[2]. The upper parts of the figures schematically show investigated constructs and the lower parts show results from the assay given as relative luminescence units (RLU) or fold change of RLU in comparison to a reference value. Rluc PCA constructs which are highly expressed in cells, i.e. the constructs comprising Rllb and Max, were used as negative controls for non-interacting protein pairs in the PCA analyses.

Figures 3A, 3B, 3C show the effect of the ATP-competitive kinase inhibitor vemurafenib, which specifically inhibits kinase activity of BRaf V600E, on the Raf:Ras interface as measured by PCA assay (Fig. 3A and Fig. 3B) or immunoprecipitation (Fig. 3C). Controls include a Rllb dimerization assay and intact luciferase which are not influenced by the presence of vemurafenib. Similar, the PCA assays using only the RBD expression construct of BRaf are not affected.

Figures 4A, 4B shows concentration dependent effects of vemurafenib (Fig. 4A) and also the kinase inhibitors dabrafenib and encorafenib (Fig. 4B) on the Raf:Ras interaction measured by the PCA assay with the indicated constructs.

Examples

The inventors have found that the RBD domain of Raf coupled to a first fragment of a Renilla luciferase protein (SEQ ID No: 37) interacts preferentially with Ras variants including the oncogenic G12V mutation coupled to a second fragment of the Renilla luciferase protein (SEQ ID No: 38) thereby increasing the PCA originating bioluminescence signal substantially. Surprisingly, it was found that the C-terminal location of the luciferase protein fragment was critical to detect the formation of the Ras:Raf PPL The G12V mutation increases the cellular amount of activated GTP-bound Ras proteins (Fig. 2A). Thus, the findings allow for the provision of a specific PCA assay which enables to evaluate and quantify the GTP-dependent Ras:Raf interaction in any cellular system via measuring bioluminescence signals.

This is particularly interesting as interaction studies with wild-type Ras did not provide a significant interaction signal using the PCA assay as read out. The Ras:Raf interaction seems to be less established with the wild type, whereas Ras variants with oncogenic mutations resulted in elevated relative luminescence units (RLU) indicating increased interaction. This observation is in agreement with the hypothesis that these constitutively active Ras mutants upregulate MAPK activities and signalling through increased interaction with Raf kinases.

The cell-based assay with full-length Raf protein allows to study allosteric effects (Fig. 2B). Exemplarily, the method was applied to study the effect of the oncogenic mutation V600E in the BRaf kinase domain (Fig. 2C). The introduction of V600E in BRaf leads to an increased RLU signal with GTP-loaded wild-type HRas in comparison to the RBD of Raf with wild- type HRas. Thus, it is confirmed that for the method according to the invention it may be appropriate to select also a Ras protein with a wild-type sequence e.g. depending on the interaction partner. By the PCA method including full-length BRaf constructs, it became possible to elucidate that several BRaf-V600E kinase inhibitors affect the Ras:Raf interaction via an allosteric mechanism. The collection of PPI reporters was subjected to time dependent treatments with the BRaf-V600E kinase inhibitor vemurafenib (Figure 3). Surprisingly, an elevation of the PPI was observed primarily in experiments analyzing BRaf complexes with the V600E mutation. Both the interaction BRaf-V600E with wild type and the interaction with the GTP- bound HRas was significantly elevated following the drug exposure (Fig. 3B). To exclude that the PPI reporter system is responsible for this observation an immunoprecipitation assays was performed. It confirmed that vemurafenib treatment indeed elevates complexes of overexpressed BRaf-V600E:HRas-G12V (Fig. 3C).

In the present case vemurafenib was a candidate compound. Besides vemurafenib also other known BRaf-V600E kinase inhibitors such as dabrafenib and encorafenib showed a similar effect. Other Raf kinase inhibitors which are known to interact with Raf are candidate compounds which may be tested using this cell based PPI reporter (Figure 4).

With the results obtained in presence of kinase inhibitors, valuable information on an allosteric effect of the V600E-specific BRaf inhibitors could be derived. In presence of e.g. vemurafenib a time and dose dependent increase of complex formation was observed for overexpressed BRaf-V600E with the HRas wild type and the oncogenic G12V mutations. The effect was also shown for the other G12V mutated Ras family members. Similar to vemurafenib, the other BRaf-V600E inhibitors increased the BRaf-V600E interaction with the mutated and constitutive active Ras variants. This allosteric effect of structurally diverse ATP-competitive kinase on the defined and regulatory PPI might be an additional explanation how BRaf inhibitors cause OFF-target effects and/or drug resistance via interference with an upstream located signalling interface. This phenomenon of drug-driven changes of PPIs needs to be considered in analyses of lead molecule efficacies and of approved drugs targeting specific kinases such as Raf.

Methodological aspects

Renilla luciferase constructs

The Rluc PCA based hybrid proteins RIIb-F[l] (a regulatory subunit of type II from PKA forming a homo dimeric PPI) has been designed as previously described (Stefan, E. et al. Quantification of dynamic protein complexes using Renilla luciferase fragment complementation applied to protein kinase A activities in vivo. Proc Natl Acad Sci U S A 104, 16916-21 (2007)). Rluc PCA fusions with Ras and Raf have been generated using an analogous cloning approach. Following PCR amplification of human Ras and Raf sequences they were fused C-terminally with either F[l] or F[2] of the Rluc PCA. Site directed mutagenesis have been performed to generate indicated Ras and Raf mutants.

Stable RasrRaf Rluc PCA cell lines

Stable transfection provides persistent expression of the introduced DNA constructs. After effective DNA delivery cells which have acquired the expression vector/s are selected by using eukaryotic antibiotics. Different cellular expression vectors for either Ras or Raf Rluc PCA constructs mediate resistance to Hygromycin B and to Zeocin. The aminoglycoside Hygromycin B produced from Streptomyces hygro-scopicus acts by inhibiting ribosomal translocation during protein synthesis. The glycopeptide antibiotic Zeocin (Streptomyces verticillus) intercalates into DNA and leads to DNA strand breaks. Following seeding and transient cell transfection of combinations of the Rluc PCA constructs into e.g. HEK293, SW480 or U20S cells, the antibiotics are added with varying concentrations (dependent on the cell line). Within 2-3 weeks stable Ras:Raf Rluc PCA reporter colonies which are resistant to the antibiotic(s) should be visible.

PPI measurements by PCA assay

HEK293 cells were grown in DMEM supplemented with 10% FBS. Indicated versions of the Rluc PCA based biosensor were transiently overexpressed in 24 well plate format. 24 or 48 hours post-transfection the growth medium was exchanged and cells resuspended PBS. The amount of PPI needs to be determined in Rluc PCA measurements of the detached cells. Cell suspensions were transferred to 96-well plates and subjected to luminescence analysis using the LMaxTM-II-384 luminometer (Molecular Devices). Rluc luminescence signals were integrated for 10 seconds following addition of the Rluc substrate benzyl-coelenterazine (5 μΜ; Nanolight). We observed elevation of cellular Ras:Raf complexes using GTPase deficient Ras mutants underlining the specificity of the PPI (Figure 2). Max and Rllb constructs are included as controls.

PPI measurements in presence of kinase inhibitors

Following treatments with indicated kinase inhibitors the medium was exchanged and detached cells subjected to Rluc PCA measurements (Figure 3). Dose-dependent effects of drug exposure on PPI-mediated luminescence signals were measured for indicated times in Figure 4. Following co-expression of indicated Rluc PCA constructs including the three Ras G12V mutants, the cells were treated with increasing vemurafenib concentrations (0.1 and 1 μΜ). Shown is the PPI analyses of n=3 independent experiments (HEK293, +SEM Fig. 4A). Following coexpression of the Rluc PCA tagged constructs BRaf-V600E:HRas-G12V, BRaf:HRas-G12V, and RBD:HRas-G12V, the effect of increasing concentrations of GSK2118436A (Dabrafenib), LGX818 (Encorafenib) and PLX4032 (Vemurafenib) was studied. Following bioluminescence measurements the obtained bioluminescence signals were normalized (i) on the interaction of RBD: HRas-G12V and (ii) on the signals obtained with 0.1 nM drug treatment. Shown is the PPI analyses of n=3 independent experiments (HEK293, +SEM; Fig. 4B).

Immunoprecipitation (comparative experiment)

Following transient overexpression of indicated Ras and Raf expression constructs the cells were treated with 10 μΜ vemurafenib for 16h. Subsequent to PBS washing steps, they were homogenized using a Potter S (B. Braun Biotech International) with 15 strikes (standard lysis buffer: 10 mM sodium phosphate pH 7.2, 150 mM NaCl, 0.5% Triton-XlOO supplemented with standard protease inhibitors and phosphatase inhibitors). The lysate (13 000 rpm, 15 min) was clarified and immunoprecipitations performed using ProteinA/G mixtures and 2 μg of control and anti-flag tag antibodies for three hours at 4°C. Resin associated proteins were washed four times with standard lysis buffer and eluted with Laemmli sample buffer. SDS- PAGE followed by immunoblotting was performed with flag tag antibodies and an antibody versus the Fl of Rluc (MAB4410, Millipore). Figure 3C shows the result of the co- immunoprecipitation experiments.

Claims

1. A method for quantification of an interaction between a Ras protein and a Raf protein, comprising the steps of

wherein the second construct comprises the Raf protein and a second luciferase protein fragment, wherein the second luciferase protein fragment is located C-terminal from the Raf protein, wherein the second construct comprises a full-length Raf sequence section and

wherein the first and the second luciferase protein fragments are derived from different sections of the same luciferase protein;

b) allowing the first and second constructs to contact each other;

c) providing a bioluminescence substrate for an intact luciferase protein; and

2. A method for measuring an effect of a candidate compound on the interaction between a Ras protein and a Raf protein, wherein the method according claim 1 is conducted in presence of the candidate compound and the effect of the candidate compound on the interaction is determined by comparing the bioluminescence as measured in step d) in presence of the candidate compound versus the bioluminescence in absence of the candidate compound.

3. A method according to claim 1 or claim 2, characterized in that the Ras protein has a sequence identity of at least 95 % to a sequence selected out of the group consisting of SEQ ID No: 1, 2, 3, and 4.

4. A method according to claim 3, characterized in that the Ras protein has a sequence selected out of the group consisting of SEQ ID No: 1 to SEQ ID No: 24, preferably SEQ ID No: 5, 15 and 20.

5. A method according to any one of claims 1 to 4, characterized in that the Raf protein has a sequence identity of at least 95 % to a sequence selected out of the group consisting of SEQ ID No: 25, 26, 27, 31, 32, 33, 34 and 35.

6. A method according to claim 4, characterized in that the Raf protein has a sequence selected out of the group comprising SEQ ID No: 25, 26, 27, 31, 32, 33, 34 and 35, preferably SEQ ID No: 26 or 32.

7. A method according to any one of claims 1 to 6, characterized in that at least one of the constructs comprises a linker intervening between the Ras protein or the Raf protein and the first or second luciferase protein fragment, wherein preferably the linker is a glycine rich linker, more preferably the linker has a sequence selected out of the group consisting of SEQ ID No: 36.

8. A method according to any of claims 1 to 7, characterized in that the luciferase protein fragments are derived from Renilla luciferase.

9. A method according to claim 8, characterized in that the first and the luciferase protein fragments have a sequence identity of at least 95 % to one of the sequence out of SEQ ID No: 37 or SEQ ID No: 38.

10. A method according to claim 11, characterized in that the Ras protein is coupled to SEQ ID No: 38 and the Raf protein is coupled to SEQ ID No: 39.

11. A method according to any of claims 1 to 10, characterized in that the bioluminescence substrate is selected out of the group consisting of Benzylcoelentrerazine, native Coelenterazine, Coelenterazine h, Coelenterazine 400a, e-Coelenterazine, Coelenterazine-Fluoride, e-Coelenterazine-F, v-Coelenterazine, Coelenterazine hep, Coelenterazine cp, Coelenterazine fcp, and Coelenterazine ip, preferably Benzylcoelentrerazine.

12. A method according to any of claims 2 to 11, characterized in that the candidate compound interacts with the Raf protein, preferably the candidate compound is a known inhibitor of the Raf kinase activity.

13. A polynucleotide encoding for a fused protein construct, characterized in that the construct comprises the Ras protein and a first luciferase protein fragment, wherein the first luciferase protein fragment is located C-terminal from the Ras protein,

wherein preferably the Ras protein has a sequence identity of at least 95 % to a sequence selected out of the group consisting of SEQ ID No: 1 to SEQ ID No: 24,

and wherein preferably the luciferase protein fragment has a sequence identity of at least 95 % to one of the sequence out of SEQ ID No: 37 or SEQ ID No: 38.

14. A polynucleotide encoding for a fused protein construct, characterized in that the construct comprises the Raf protein and a second luciferase protein fragment, wherein the second luciferase protein fragment is located C-terminal from the Raf protein, and

wherein the Raf protein has a sequence identity of at least 95 % to a sequence selected out of the group consisting of SEQ ID No: 25, 26, 27, 31, 32, 33, 34, 35, 36, and wherein preferably the luciferase protein fragment has a sequence identity of at least 95 % to the sequence out of SEQ ID No: 37 or SEQ ID No: 38.

15. A cell comprising a nucleotide according to claim 14 or a nucleotide according to claim 15.

16. A cell according to claim 15, wherein the cell line is established from a cell line selected out of HEK293, SW480, U20S, or any modified melanoma cell line, preferably the cell is a HEK293 cell line.