Methods and tools for detecting the effects of compounds on protein, RNA, or DNA interactions

Abstract

There is a need for a reliable and simple cell-based assay that can be used to screen small molecule modulators of protein-protein interactions. [Solution] The present invention deals with an in vitro method for evaluating the ability of compounds to disrupt or stabilize interactions between one or more baits and one or more candidate plays in eukaryotic cells. In addition, the present invention deals with compounds identified by the in vitro method according to any one of claims 1 to 13 for use as pharmaceuticals.

Description

[Technical Field]

[0001] The present invention deals with an in vitro method for evaluating the ability of compounds to disrupt or stabilize interactions between one or more baits and one or more candidate plays in eukaryotic cells. In addition, the present invention deals with compounds identified by the in vitro method described in any one of claims 1 to 13 for use as pharmaceuticals.

[0002] (Field of invention) The present invention preferably relates to methods, uses, and tools for studying ligand interactions in eukaryotic cells and the effects of compounds on these interactions in order to identify novel drug candidates. [Background technology]

[0003] Protein-protein interactions (PPIs) are crucial for cellular function and exist in all types of organisms, with over 500,000 interactions estimated in humans (Li et al., 2017; Rolland et al., 2014). PPIs are involved in regulating biological pathways consistent with their involvement in the development of numerous disorders. Therefore, they represent attractive targets for the development of new therapeutics to treat various human diseases (Lu et al., 2020; Scott et al., 2016). Most PPI interactions were initially considered unsuitable for drug discovery due to their large interaction surface, before emerging as promising potential targets for identifying some efficient inhibitors of PPIs related to human diseases (Arkin and Wells, 2004; Arkin et al., 2014).

[0004] Targeting RNA-protein interactions (RPIs), which are critically involved in pathological mechanisms, is also a promising strategy for discovering a novel class of drug candidates, much of which remains undeveloped (Einstein, J. Met., 2021). Intracellular RPIs are highly diverse, including interactions with messenger RNA (mRNA) (Baltz, A. Get., 2012), ribosomal RNA (rRNA) (Simsek, D. et al., 2017), and non-coding RNA (ncRNA) (Lu, S. et al., 2019), which are crucial for fine-tuning spatiotemporal gene expression. As revealed by genomic approaches (Van Nostrand, E. et al. 2020 and Castello, A. et al. 2012), the human genome contains over 1000 transcripts encoding RNA-binding proteins (RBPs), and therefore offers a wide variety of interactions with coding and non-coding RNAs. However, while the diversity of RNA:protein interactions could enable the development of RPI alienation molecules (Wu, P. 2020), very little research has been done so far, and it has been limited to a few complexes such as Lin28-let7 interactions (Roos, M. et al. 2018 and Wang, L. et al. 2018), MSI-RNA (Minuesa, G. et al. 2019), and hnRNP A18-RNA (Solano-Gonzalez, E. et al. 2021).

[0005] As mentioned above, targeting PPIs and RPIs for drug discovery is an attractive strategy for the pharmaceutical industry. However, despite years of extensive research into developing new PPI or RPI modulators, only a few small molecules have made it to market. The large surface area of ​​PPIs and RPIs may be a contributing factor to drug development, but there is also a lack of availability of methods to accurately assess whether candidate modulators are effective in the cellular environment using high-content screening approaches.

[0006] Regarding PPIs in particular, the need for methods to conduct large-scale screening of putative PPI inhibitors is especially critical due to the aforementioned obstacles in the discovery of PPI inhibitors. For this purpose, several in vitro and cell-based methods have already been developed (Arkin et al., Titeca et al., 2019, Wade and Arkin). One of the most common assays for screening small molecules targeting specific protein-protein interactions is homogeneous time-resolved fluorescence (HTRF) (Degorce, 2009). This highly sensitive technique, based on energy transfer between donor and acceptor molecules, has been adapted for automation of HTS. However, like other in vitro methods, the main drawback of HTRF is the need to purify the protein of interest, which in most cases limits the use of specific domains and lacks information on the fate of the small molecules and their efficiency in inhibiting PPIs in the cellular environment. While HTRF methods can still monitor protein-protein interactions in cells, the adaptation of HTRF to cell-based assays required subsequent steps of antibody or chemical protein labeling, reducing the interest in using this approach in cell assays (Degorce, 2009). FRET technology was primarily used to assess interactions between two proteins in cells, but its lack of sensitivity and low signal-to-noise ratio made it unsuitable for accurately measuring disruption of interactions in the cellular environment (Piston and Kremers, 2007). Meanwhile, BRET, still based on energy transfer but using bioluminescence instead of fluorescence, has become a new area of ​​interest with the recent development of nanoluciferases. This small, bright, and stable donor has greatly expanded the potential applications of BRET assays through its use in NanoBRET assays (Dale et al., 2019, Machleidt et al., 2015).NanoBRET Technology is based on the expression of Nanoluc, a bright luciferase acting as an energy donor, and a full-length protein tagged with a fluorescent dye called HaloTag as an acceptor. The readout is the average luminescence signal per well recorded by a microplate reader. This technique enables the detection of PPIs in the natural cellular environment only if the donor and acceptor probes are close enough to each other to allow energy transfer. In addition, while NanoBRET assays have good sensitivity, their reproducibility depends on donor expression, as the BRET signal depends on the donor / acceptor ratio, as described by donor saturation assays (Couturier and Deprez, 2012). To overcome this limitation and ensure the stability and reproducibility of the BRET signal, stable cell lines expressing the donor are preferable for developing BRET-based screening assays, increasing the complexity of setting up such assays (Couturier and Deprez, 2012). Therefore, there is a need for a reliable and simple cell-based assay that can be used to screen small molecule modulators of protein-protein interactions.

[0007] Regarding RPi, previous approaches to finding potent inhibitors of the RNA:protein interface have involved testing the efficacy or selectivity of several hits using in vitro assays such as pull-down experiments with cell lysates or fluorescence polarization assays complemented by RNA ELISA (Roos, M. et al., 2016 and Minuesa, G. et al.). While in vitro approaches are important for defining putative hits and leading to validation of effective compounds, deciphering whether selected molecules are effective in cellular contexts generally relies on indirect measurements using techniques such as Cellular Engagement Thermal Shift Assay (CETSA) or functional assays, where the putative results of RPI disruption to cellular function have considerable uncertainty. In fact, multiple functions are associated with RPI, which complicates the interpretation of functional assay results. In addition, the low micromolar range of K is particularly common in the case of RPI inhibitors. D When using small molecules with certain properties, toxicity and off-target effects are always presumptive biases that are difficult to eliminate (Jung, J. et al. 2013 and Cambridge, L. et al. 2017). To bridge the gap between in vitro assays and functional assays, cell-based approaches such as FRET (fluorescence resonance energy transfer) or PLA (proximity ligation assay) have been adapted to detect RPIs in cells (Jung, J. et al. 2013 and Cambridge, L. et al. 2017), but several technical issues, such as the need for RNA adapters in FRET and PLA, the proximity of donor and acceptor proteins in FRET, and the use of antibodies in PLA, have hindered their application.

[0008] Beyond assays that examine RPIs in cells, other challenges arise from the drug discovery process, including the quality of computational models, the strategies used in in silico screening, and the lack of experimental feedback and validation of computationally predicted inhibitors, which are essential for directing rational drug design procedures to the most relevant molecules.

[0009] The patent family International Publication No. 2016012451, partially filed by the same inventors as the present invention, relates to a method for detecting interactions between one or more protein baits and one or more candidate plays in eukaryotic cells by using polymerized tubulin-binding moieties, such as microtubule-binding domains (MBDs), fused to one or more baits, as a tool for determining the occurrence of interactions between one or more baits and one or more plays in eukaryotic cells.

[0010] Starting from the technologies described in this patent family, the applicant has developed an innovative new technique that enables the evaluation of the ability of compounds to disrupt or stabilize interactions between DNA, RNA, and / or proteins, and in particular, the evaluation of the ability of compounds to disrupt or stabilize PPIs and RPIs by single-cell fluorescence microscopy on a high-content screening (HCS) imager. This technique enables the screening of molecular libraries to identify drug candidates.

[0011] This technology can address the above challenges by introducing a drug screening approach that integrates chemical, structural, and cellular data from experimental assays suitable for HCS to score interactions in cells, particularly RPIs or PPIs, as well as experimental techniques for identifying and developing molecules, especially small molecules, that target these interactions, particularly RPIs or PPIs. [Overview of the project]

[0012] The present invention relates to a method for evaluating the ability of compounds to disrupt or stabilize interactions between one or more baits, preferably proteins, and one or more candidate peys in eukaryotic cells, more specifically in living eukaryotic cell conditions.

[0013] The system consists of (i) a "bait," a known protein, ribonucleic acid, or deoxyribonucleic acid, transported to the microtubule as a result of microtubule fusion to the microtubule-binding domain, and (ii) a detection system based on the expression of the bait in cells, which can be, for example, a protein, nucleic acid, or deoxyribonucleic acid.

[0014] When the play interacts with the bait, it also localizes along microtubules, which therefore makes it easily detectable. As developed in this invention, it makes it possible to determine the difference in interaction before and after contact with the test molecule, as well as to identify molecules of interest in various pathological conditions in which PPIs and RPIs are involved, such as cancer, viral or bacterial infections, epigenetic diseases, or proteinopathy.

[0015] Here, the system described in International Publication No. 2016012451 (MT BENCH) was improved to bridge the gap between in vitro assays and functional assays by investigating whether the interaction between selected bait and selected play is affected by the compound in the cellular context.

[0016] The present invention further relates to baits, vectors, cell lines, kits suitable for the method, use, and compounds specified according to the method of the present invention.

[0017] Therefore, a first aspect of the present invention is an in vitro method for evaluating the ability of a compound to disrupt or stabilize the interaction between one or more baits and one or more candidate plays in eukaryotic cells, a. (i) one or more baits comprising a bait portion and a polymerization tubulin binding portion, and (ii) eukaryotic cells expressing one or more candidate baits, b. In eukaryotic cells, the steps include determining the occurrence of an interaction between one or more baits, which are bound to polymerized tubulin in eukaryotic cells, and one or more candidate plays, thereby localizing one or more candidate plays along polymerized tubulin, and thereby detecting the interaction. c. A step of bringing eukaryotic cells into contact with the compound, d. A step in which the occurrence of interactions between one or more baits and one or more candidate plays in eukaryotic cells is determined, as in step b, and this interaction is quantified. e. The comparison step of the occurrence of interactions performed in step b and step d, The objective is to provide a method comprising the step of concluding whether the effect of the compound is to disrupt or stabilize the interaction between one or more baits and one or more candidate plays.

[0018] In a preferred embodiment, one or more baits are proteins.

[0019] In a preferred embodiment, one or more candidate plays are proteins or ribonucleic acids.

[0020] A second aspect of the present invention relates to the use of a polymerization tubulin binding moiety as a tool for evaluating the ability of a compound to disrupt or stabilize interactions between one or more baits and one or more plays in a eukaryotic cell, wherein the baits are bound to polymerization tubulin in the eukaryotic cell, thereby localizing one or more candidate plays along the polymerization tubulin, and the one or more baits and one or more candidate plays are brought into contact with a compound, thereby enabling evaluation of the disruption or stabilization of interactions.

[0021] The third aspect of the present invention relates to a protein bait comprising a polymerized tubulin-binding portion containing one or more microtubule-binding domains and a bait portion.

[0022] The fourth aspect of the present invention relates to a ribonucleic acid bait comprising a polymerized tubulin-binding portion containing one or more microtubule-binding domains and a bait portion.

[0023] The fifth aspect of the present invention relates to a deoxyribonucleic acid bait comprising a polymerized tubulin-binding portion containing one or more microtubule-binding domains and a bait portion.

[0024] The sixth aspect of the present invention relates to a vector containing an expression cassette suitable for expressing the protein bait of the present invention.

[0025] The seventh aspect of the present invention relates to a vector containing an expression cassette suitable for expressing the ribonucleic acid bait of the present invention.

[0026] The eighth aspect of the present invention relates to a vector containing an expression cassette suitable for expressing the deoxyribonucleic acid bait of the present invention.

[0027] The ninth aspect of the present invention relates to a vector containing an expression cassette suitable for expressing the protein prey of the present invention.

[0028] The tenth aspect of the present invention relates to a vector containing an expression cassette suitable for expressing the ribonucleic acid prey of the present invention.

[0029] The eleventh aspect of the present invention relates to a vector containing an expression cassette suitable for expressing the deoxyribonucleic acid prey of the present invention.

[0030] The twelfth aspect of the present invention relates to a cell line transfected with the vector of the present invention and / or stably expressing a bait suitable for the method and use of the present invention.

[0031] A thirteenth aspect of the present invention relates to a cell line that is transfected with the vector of the present invention and / or stably expresses a prey suitable for the methods and uses of the present invention.

[0032] A fourteenth aspect of the present invention relates to a kit for evaluating the ability of compounds to disrupt or stabilize interactions between one or more baits and one or more plays in cells, comprising the method and baits, vectors, and / or cell lines suitable for use of the present invention.

[0033] A fifteenth aspect of the present invention relates to compounds identified by the in vitro method according to the present invention for use as pharmaceuticals. [Brief explanation of the drawing]

[0034] [Figure 1] Principle of steps a and b of the method: High-throughput analysis and detection system. Application to endogenous mRNA interaction with the mRNA-binding protein YB-1. (A) Cells cultured on a microplate are transfected by an automated system. (B) After image acquisition, identification of a "microtubule-like pattern" in the fluorescent channel corresponding to prey (messenger RNA) demonstrates the interaction between bait (YB-1-GFP-tau) and prey (mRNA). (C) (Top figure, from left to right) Bait: Detection of YB-1-GFP-tau, prey: Detection of mRNA, Example of detection of protein-RNA interaction showing the presence of a microtubule-like pattern in the Cy3 channel (corresponding to a poly-T RNA Cy3 labeled probe for mRNA detection). (Bottom figure) YB-1-GFP is used as a control and is uniformly distributed in the cytoplasm. [Figure 2]Principles of the method for fixed cells and endogenous prey in steps a and b. Detection of endogenous prey transported by microtubules in two steps. The first step (step 1) comprises transfection of an expression vector encoding a protein bait containing a microtubule-binding domain and optionally a fluorescent label. The second step (step 2) comprises detection of endogenous prey using a reagent such as an antibody against native prey, or hybridization with nucleic acid. The prey is then detected on microtubules after its interaction with the bait. [Figure 3] Principles of the method for live cells and exogenous prey for steps a and b. The first step (step 1) comprises cotransfection of two expression vectors, the first plasmid expressing a protein bait containing a microtubule-binding domain and optionally a fluorescent label. The second plasmid expresses exogenous prey optionally labeled with a fluorescent label. The second step (step 2) comprises detection of exogenous prey along dynamic microtubules or in the cytoplasm using prey-specific fluorescent labels and / or antibodies. [Figure 4]Development of high-content screening conditions for case studies of PPIs (p53 / MDM2) in cellular conditions using a microtubule bench (MT bench). (A) Schematic diagram of the MT bench technique for detecting PPI regulation using small molecules. Bait (P53) is covalently bound to RFP (red fluorescent protein) and MBD (microtubule-binding domain). Therefore, the RFP signal follows the microtubule network. Prey (MDM2) is bound to GFP (green fluorescent protein) but is free in the cytoplasm. Both signals, showing the same microtubule network, mean that bait and prey interact and can be regulated by inhibitors (e.g., RG7112). This interaction can be monitored by HCS (Opera Phenix by Perkin Elmer) and results in numerous single-cell events per well (here, 10% of the surface of 96 wells). (B) Cell images of fixed U-2 OS cells overexpressing p53-FL_C_RFP-MBD (used as bait protein) for positive controls (control+), or the triple mutant p53-F19AW23AL26A_RFP-MBD (used as bait protein) and MDM2-FL_C_GFP (used as prey protein) for negative controls (control-). The control+ image shows bait / prey interaction (both fluorescence signals indicate the same microtubule network), while the control- image does not show interaction (both fluorescence signals do not indicate the same cell compartment), in which case MDM2 is located in the nucleus. Inhibition of the interaction between p53 and MDM2 was induced by treatment with increasing concentrations of the well-known MDM2 inhibitor RG7112 (Vu et al., 2013) ([0.05 μM], [0.5 μM], [5 μM], and [50 μM]), demonstrating a dose-dependent disruption of the p53 / MDM2 interaction and relocalization of MDM2 in the nucleus. [Figure 5]The effect of point mutations on p53 residues at the MDM2 binding site using MT bench technology. Comparison of PCC scores between the wild-type construct of p53 as bait and different mutant constructs with single mutations (p53-F19A_C_RFP-MBD, p53-W23A_C_RFP-MBD), double mutations (p53-F19AW23A_C_RFP-MBD, p53-W23AL26A_C_RFP-MBD), and triple mutations (p53-F19AW23AL26A_C_RFP-MBD) was normalized to enable comparison between experimental plates (Kevorkov and Makarenkov, 2005). PCC values ​​were normalized relative to the control by using the mean of negative control wells and positive control wells corresponding to the p53-FL / MDM2 interaction, as well as negative control wells with the triple mutant. [Figure 6] Analysis pipeline workflow. The analyses used in this study were developed using Harmony® and R software. The different stages of the pipeline are: 1) cell biology, 2) image acquisition, 3) image segmentation, 4) image processing, 5) colocalization, and 6) data analysis. [Figure 7]Development of screening conditions. (A) Z' factor graph representation demonstrating the robustness of the system developed here in a 384-well plate format. Z' factor (p53-FL_C_RFP-MBD / MDM2-FL_C_GFP) containing p53 and MDM2 proteins is used as a positive control. (B) Fixed U-2 OS cells overexpressing p53-FL_C_RFP-MBD (used as bait protein, red) and MDM2-FL_C_GFP (used as prey protein) for the positive control (control+), or G3BP1_GFP (used as prey protein) for the negative control (control-). In the positive control, p53 / MDM2 interact and are shown in the merged channel (both fluorescence signals indicate the same microtubule network). In contrast, in the negative control, p53 / G3BP1 do not interact. The G3BP1 protein diffuses into the cytoplasm and is not rearranged on the microtubule network (both fluorescence signals do not indicate the same microtubule network). Cells were acquired using an HCS microscope system (Opera Phenix, PerkinElmer) equipped with a 40× water-immersion (1.1NA) objective lens and confocal mode. (C) Z' factor graph representation using the non-P53-related protein G3BP1 as a negative control. Graphs and calculations were performed using R software. [Figure 8] The conditions were tested using an MT bench assay. U-2 OS cells were cotransfected with MDM2-FL-GFP and P53 mutants, and treated with RG7112 small molecules under several conditions. [Figure 9] Relative enrichment of MDM2 on microtubules by different p53 mutants. Asterisks indicate that MDM2 does not interact with p53 on microtubules under these conditions, thus hindering the measurement of relative enrichment. To allow comparison between conditions, enrichment was normalized using minimum-maximum normalization (using positive and negative controls as maximum and minimum). [Figure 10] Cell images from MTbench technology. Both YB1-FL constructs fused to GFP were used as bait, and mRNA (Cy3 labeled) was used as play. In both cases, the mRNA was localized to microtubules. [Figure 11]Relative enrichment of mRNA against microtubules in cells transfected with both YB1-FL constructs. Enrichment is calculated as defined in Materials and Methods. *** represents a p-value of 1.44e-5. Error bars represent the standard error calculated from 16 replicates. The negative control corresponds to a condition in which cells were transfected with GFP fused to MBD. No RNA binding occurs under this condition. [Figure 12] Relative enrichment of mRNA relative to microtubules in cells expressing different DNA constructs. Enrichment values ​​were obtained in 384-well plates. The negative control corresponds to a condition in which cells were transfected with GFP fused to MBD. Error bars represent the standard error calculated from 16 replicates. The p-values ​​are as follows: -p-value(MBD-GFP-YB1-FL / MBD-GFP-YB1-deltaCSD)=8.59e-7 -p-value(YB1-FL-GFP-MBD / YB1-deltaCSD-GFP-MBD)=9.13e-20 [Figure 13] This is a dose-response curve showing mRNA inhibition. The YB1 interaction, which involves increasing the concentration of C8, allows for the determination of its activity (IC50) in the cellular environment.

[0035] definition According to the present invention, "evaluating the ability of a compound to disrupt or stabilize an interaction" means enabling a method to determine whether a compound can have an effect on the interaction between one or more baits and one or more candidate plays. In particular, the method enables determining whether a compound disrupts or stabilizes this interaction.

[0036] According to the present invention, the "bait" includes at least one "bait portion" and at least one "polymerized tubulin-binding portion".

[0037] According to the present invention, the "bait portion" is a portion of a cell that tends to interact with one or more candidate pylori.

[0038] According to the present invention, the expressions "comprising" or "comprises" also include "consisting" or "consisting of".

[0039] The "polymerized tubulin binding moiety" comprises a peptide, protein, or nucleoprotein having the ability to specifically bind to polymerized tubulin. Preferably, the "polymerized tubulin binding moiety" binds with higher specificity to its polymerized form of tubulin than to non-polymerized forms of tubulin, including monomeric and / or heterodimeric forms (corresponding to alpha / beta tubulin heterodimers). Ideally, the polymerized tubulin binding moiety binds almost exclusively to polymerized tubulin containing microtubules. The "polymerized tubulin binding moiety" can ideally be selected from known antibodies, microtubule-binding proteins, or fragments thereof, which include (in a non-exclusive manner) microtubule stabilizers and destabilizers, molecular motors, microtubule-cleaving proteins, and end-tracking proteins such as plus-end tracking proteins.

[0040] Microtubules are components of the cytoskeleton found throughout the cytoplasm. They are part of the structural network ("cytoskeleton") within the cell's cytoplasm. While the primary role of the microtubule cytoskeleton is mechanical support, microtubules are also involved in many other processes. Therefore, microtubules are only a part of the so-called "microtubule cytoskeleton," as the latter further includes associated proteins such as microtubule-associated proteins (MAPs), along with other organizing structures such as centrosomes.

[0041] Polymerized tubulin (or Polymerized-tubulin) exclusively refers to aggregates of monomeric tubulin, or alternatively, aggregates of heterodimers of tubulin, that are arranged in a regular pattern and possess different polarities. Tubular polymers of tubulin are highly dynamic, growing to lengths of up to 50 micrometers with an average length of 25 μm. Microtubules generally have an outer diameter of approximately 24–25 nm, while their inner diameter is approximately 12 nm. They are found in eukaryotic cells and are formed by the polymerization of dimers of two globular proteins, α-tubulin and β-tubulin. Therefore, the expression "polymerized tubulin" encompasses microtubules.

[0042] Therefore, “microtubules” represent a specific rearrangement of “polymerized tubulin” that physiologically arises in eukaryotic cells and, together with additional partners, forms the “microtubule cytoskeleton.” The physiological assembly of microtubules is generally described as involving a first step of regulated assembly of alpha-tubulin and beta-tubulin heterodimers, which together form polarized protofilaments. The protofilaments are then thought to assemble into so-called microtubules as cylinders. Thus, microtubules are generally described as polymers of α- and β-tubulin dimers, consisting of 13 protofilaments assembled around a hollow core. However, it should be noted that so-called microtubules with different numbers of protofilaments, such as microtubules with 14 or 15 protofilaments, have also been described in the art. However, the physiological significance of such changes, i.e., “protofilament transitions,” remains unclear.

[0043] Tubulins refer to one of several members of a small family of globular proteins. The tubulin superfamily includes five distinct families: alpha-, beta-, gamma-, delta-, and epsilon-tubulins, as well as a sixth family (zeta-tubulin) found only in kinetoplastid protozoa. The most common members of the tubulin family are alpha-tubulin (α-tubulin) and beta-tubulin (β-tubulin), which are proteins that make up microtubules. The end of a microtubule corresponding to beta-tubulin is called the plus end. The end of a microtubule corresponding to alpha-tubulin is called the minus end.

[0044] Therefore, the "polymerized tubulin-binding portion" of the present invention may include alpha-tubulin and / or beta-tubulin-binding portions and / or combinations thereof, preferably binding to tubulin in polymer form rather than in monomer form or heterodimer form.

[0045] As is known in the art, the "heterodimerized form" of tubulin corresponds to the alpha / beta tubulin heterodimer. Therefore, the polymeric form of tubulin also corresponds to the heterodimer polymer, which in turn corresponds to more than one heterodimer of alpha / beta tubulin.

[0046] For reference, human alpha-tubulin is represented by the sequence of Sequence ID No. 1.

[0047] For reference, human beta-tubulin is represented by the sequence of Sequence ID No. 2.

[0048] Firstly, microtubules, due to their dimensions (25 nm in diameter and 1 / 10 micrometer in length), provide a large surface area within the cell. Considering a 110 μm long microtubule in a typical mammalian cell like a HeLa cell, the microtubule surface area is over 30 μm², and can be even larger in cells like neurons or muscle cells. Thus, such a large surface area is available for the binding of a vast number of baits without saturation (if bait requires an interaction surface area of ​​as large as 10 nm² on a microtubule, then virtually more than 3,000,000 bait copies per cell can theoretically be immobilized on microtubules). This is significantly higher than the typical number of proteins overexpressed in transfected mammalian cells (around 100,000 copies is already a large overexpression for most proteins).

[0049] A second advantage of using the microtubule surface lies in its dynamic behavior. Microtubules are inherently very dynamic, constantly alternating between shortening and growing phases. This behavior allows bait to detach from microtubules during the depolymerization phase and bind to another microtubule, or to bind after the microtubule repolymerization. During that time interval, the bait can detach from the microtubule and then capture prey located in the bulk cytoplasm rather than in the vicinity of the microtubule.

[0050] The detection of filamentous structures such as "microtubule-like patterns" has long been known to be highly sensitive and to consist of extrapolating straight lines in a fluorescence image so that the signal can be detected from uniformly distributed noise. This general principle is routinely used to image microtubules with fluorescence speckle microscopy (see Salmon & Waterman; How we discovered fluorescent speckle microscopy; Mol Biol Cell, 3940-2; 2011 for an overview). However, this straightforward detection method is not possible for more spherical structures or other poorly defined substrates. The method of the present invention is also compatible with real-time data analysis of live cell fluorescence images and can therefore be implemented to better detect bait bound to microtubules in live cells. Importantly, microtubules move with time due to their highly dynamic structure, which further improves the detection of microtubules from a large fluorescence background for detection in live cells.

[0051] The term "microtubule-binding domain" refers to one or more fragments of microtubule-binding proteins involved in the binding of polymerized tubulin, particularly to microtubules. A non-restrictive list of microtubule-binding proteins is provided in Table 1 below.

[0052] [Table 1]

[0053] Therefore, the "polymerized tubulin binding portion" can be selected from full-length microtubule-binding proteins, fragments of microtubule-binding proteins, or even isolated microtubule-binding domains, all of which are known to specifically bind to polymerized tubulin such as microtubules.

[0054] A "microtubule-binding domain" (MBD) refers to one or more fragments of microtubule-binding proteins involved in the binding of polymerized tubulin, particularly to microtubules.

[0055] According to the present invention, "determining the occurrence of an interaction" refers to identifying the contact between one or more protein baits and one or more candidate plays in the microtubule-binding system according to the present invention. The protein baits bind to polymerized tubulin in eukaryotic cells, thereby localizing one or more candidate plays along the polymerized tubulin, and thereby detecting the interaction when it occurs.

[0056] According to the present invention, "comparison of interaction occurrences" refers to the identification of potential changes (destruction or stabilization) in the interaction between step b and step d, making it possible to conclude whether the effect of the compound is destruction or stabilization of the interaction. In addition to destruction or stabilization, the method can also demonstrate that there is no change in the interaction.

[0057] According to the present invention, “disruption of interaction” means that one or more baits and one or more candidate plays are no longer interacting, or that the interaction is weaker than it was before the step in contact with the compound being tested.

[0058] According to the present invention, "stabilization of interaction" means that one or more baits and one or more candidate plays are interacting, or that the interaction is stronger than before the step in which they come into contact with the compound being tested.

[0059] According to the present invention, “quantification of interaction” refers to the evaluation of the intensity of the interaction that occurs between one or more protein baits and one or more candidate plays. This quantification allows for the concatenation of the data obtained in steps b and d from the intensity of the interaction between one or more baits and one or more candidate plays in order to conclude whether the effect of the compound is to disrupt or stabilize the interaction, and to evaluate the intensity of this disruption or stabilization.

[0060] According to the present invention, "bringing eukaryotic cells into contact with a compound" refers to introducing the compound into the culture medium in which the eukaryotic cells are being tested.

[0061] IC50, or half-inhibitory concentration, is the most widely used and useful measure of drug efficacy. It indicates the amount of drug required to half-inhibit a biological process and thus provides a measure of the potency of antagonist drugs in pharmacological studies. In this invention, IC50 measures the amount of compound being tested required to disrupt or stabilize the interaction between one or more protein baits and one or more candidate baits.

[0062] According to the present invention, “physiological” conditions may consist of an experimental setup in which eukaryotic cells are not under stress or apoptosis, and this includes an experimental setup in which eukaryotic cells are not in contact with a culture medium having non-physiological pH or salt conditions.

[0063] According to the present invention, a "projection domain" refers to a domain found in MAPs such as MAP2 or tau, which is involved in microtubule bundling and determining the spacing between microtubules. Projection domains can also interact with other cytoskeletal structures. An example of a projection domain is the projection domain of the tau protein having the sequence of Sequence ID No. 3.

[0064] According to the present invention, "proteinopathy" refers to any disease or condition resulting from abnormal synthesis, folding, post-translational modification, or deposition of proteins in cells or tissues. Examples of proteinopathy include, but are not limited to, Alzheimer's disease, Parkinson's disease, Huntington's disease, and TDP-43 proteinopathy such as ALS and FTLD.

[0065] According to the present invention, "TDP-43 proteinosis" refers to a disease characterized by the presence of abnormally phosphorylated, ubiquitinated, and cleaved DNA-binding protein TDP-43 in the affected brain and spinal cord. Examples of TDP-43 proteinosis include, without limitation, amyotrophic lateral sclerosis (ALS) and frontotemporal lobar degeneration (FTLD).

[0066] According to the present invention, “linker” refers to an unstructured domain, in particular an unstructured domain that enables a nanometer-length gap between the polymerized tubulin surface and the protein of interest. Examples of linkers are known in the art. In one embodiment, the linker is a protruding domain of a microtubule-associated protein (MAP).

[0067] According to the present invention, "high-content screening (HCS)" is an image-based method used in drug discovery to identify phenotypic-modifying molecules (biogens or small molecules) by simultaneously using several cellular parameters as readouts (Fraietta and Gasparri, 2016; Mattiazzi Usaj et al., 2016).

[0068] According to the present invention, “compound” refers to any natural or chemical substance that can act on the interaction between one or more baits and one or more candidate plays. The compound may be a known substance or may be derived from a library of substances based on in vivo, in vitro, or in silico data. According to embodiments of the present invention, the compound may be a pharmaceutical candidate if one or more baits and one or more candidate plays are characteristic of the interaction that occurs in a particular medical condition. In particular, the compound may be a small molecule, peptide, protein, RNA, or DNA.

[0069] According to the present invention, "endogenous" refers to bait and / or play that originates from within the organism of an individual.

[0070] According to the present invention, the term "mutated" includes deletions, cleavages, additions, or substitutions of domains, amino acids, or glycosylations. [Modes for carrying out the invention]

[0071] The present invention relates to an in vitro method for evaluating the ability of compounds to disrupt or stabilize interactions between one or more baits and one or more candidate plays in eukaryotic cells.

[0072] In particular, in vitro methods for evaluating the ability of compounds to disrupt or stabilize interactions between one or more baits and one or more candidate plays in eukaryotic cells are: a. (i) one or more baits comprising a bait portion and a polymerization tubulin binding portion, and (ii) eukaryotic cells expressing one or more candidate baits, b. In eukaryotic cells, the steps include determining the occurrence of an interaction between one or more baits, which are bound to polymerized tubulin in eukaryotic cells, and one or more candidate plays, thereby localizing one or more candidate plays along polymerized tubulin, and thereby detecting the interaction. c. A step of bringing eukaryotic cells into contact with the compound, d. A step in which the occurrence of interactions between one or more baits and one or more candidate plays in eukaryotic cells is determined, as in step b, and this interaction is quantified. e. The comparison step of the occurrence of interactions performed in step b and step d, f. The step of concluding whether the effect of the compound is to disrupt or stabilize the interaction between one or more baits and one or more candidate plays.

[0073] A method and use for detecting the interaction between bait and candidate play for steps a, b, and d. According to one embodiment, the “bait” comprises at least one “bait portion” and at least one “polymerized tubulin-binding portion” that do not originate from the same naturally occurring entity.

[0074] The bait can be delivered to microtubules using a fusion polymerization tubulin-binding moiety, such as a microtubule-associated protein (MAP) or microtubule-associated protein (MBD) containing tau (NCBI reference sequence: NP_005901.2 or SEQ ID NO: 14).

[0075] According to one embodiment, bait fixation is due to the repetition of microtubule-binding domains commonly present in MAP. These repetitions have a higher affinity for microtubules compared to free tubulin, which is a fundamental component of microtubules, and using this method, only the weaker portions are released in the cytoplasm.

[0076] When prey interacts with bait transported by microtubules, its rearrangement on the microtubule surface allows for its detection through the appearance of microtubule structures in fluorescence images of the prey.

[0077] The detection of filamentous structures such as "microtubule-like patterns" is highly sensitive and has long been known to involve extrapolating straight lines in a fluorescence image to enable signal detection from uniformly distributed noise. This general principle is routinely used to image microtubules with fluorescence speckle microscopy (for an overview, see Salmon & Waterman; How we discovered fluorescence speckle microscopy; Mol Biol Cell, 3940-2; 2011).

[0078] However, this easy detection method is not possible with more spherical structures or other poorly defined substrates. The method of the present invention is also compatible with real-time data analysis of live cell fluorescence images and can therefore be implemented to better detect bait bound to microtubules in live cells. Importantly, microtubules move with time due to their highly dynamic structure, which further improves the detection of microtubules from a large fluorescence background for detection in live cells.

[0079] For all these reasons, the selection of microtubules is highly relevant, even when compared to other types of filaments. The interaction between bait and play can be observed in real time, allowing for tracking of the effects of the compound being tested.

[0080] The methods and uses of the present invention have been further validated, as illustrated by the examples. Briefly, the plasmid is designed to instruct the expression of a known protein (bait) in mammalian cells, which is a known protein (bait) fused to a tau protein via its N-terminal or C-terminal domain (the end of the tau projection domain).

[0081] After fusing with tau, the two baits were successfully transported to microtubules: the transcription factor P53 (tumor protein 53, accession number: BAC16799) and the mRNA-binding protein YB-1 (nuclease-sensitive element-binding protein 1 or Y-box-binding protein, accession number: NP_004550.2), both of which were readily detected along microtubules by light microscopy in HeLa cells (human cancer cells, respectively).

[0082] After fusing with tau, the two baits were successfully transported to microtubules: the transcription factor P53 (tumor protein 53, accession number: BAC16799) and the mRNA-binding protein YB-1 (nuclease-sensitive element-binding protein 1 or Y-box-binding protein, accession number: NP_004550.2), both of which were readily detected along microtubules by light microscopy in U2-OS (a cell line with an epithelial morphology derived from moderately differentiated sarcoma).

[0083] The "bait portion" and candidate play may be the same or different, and are preferably selected from the group that includes (or is composed of) proteins or nucleic acids containing nucleoproteins and deoxyribonucleic acids.

[0084] In a non-restrictive form, “bait portion” and / or “candidate play” include any molecule of interest, in particular biologically relevant, including antibodies, nucleic acid-binding proteins, and various other bait receptor proteins or peptides, especially those of diagnostic and pharmacological relevance. It also includes nucleic acids such as messenger RNA, coding and non-coding RNA, transfer RNA, ribosomal RNA, interfering RNA, or silencing RNA. It also includes any type of deoxyribonucleic acid, such as antisense DNA, circular DNA, complementary DNA, spacer DNA, heterologous DNA, hybrid DNA, homozygous double-stranded DNA, ligated DNA, mobile DNA (transposons), ribosomal DNA, satellite DNA, single-stranded DNA, double-stranded DNA, superhelical DNA, or triple-stranded DNA.

[0085] In the method according to the present invention, RNA, protein, or DNA is endogenous.

[0086] The interaction between bait and play is evaluated based on the interaction of bait with polymerized tubulin by the polymerized tubulin binding moiety. Therefore, it is preferable that the bait moiety does not interact with polymerized tubulin at all, does not colocalize with it, or interacts with or colocalizes it in a limited manner, compared to the same moiety conjugated to the polymerized tubulin binding moiety under reference and / or physiological conditions.

[0087] The bait portion of particular consideration is the bait portion whose distribution is diffused into the cytoplasm of eukaryotic cells under reference and / or physiological conditions.

[0088] According to some embodiments, the “bait portion” is a portion that does not interact with or co-localize with polymerized tubulin (or microtubules) in eukaryotic cells under reference or physiological conditions.

[0089] According to one exemplary embodiment, the "bait portion" is a nucleic acid-binding portion such as a nucleic acid-binding protein, and the "candidate play" is a nucleic acid such as messenger RNA.

[0090] In another embodiment, the "bait portion" is an antibody or a fragment thereof, and the "candidate play" is a protein or peptide that readily binds to the antibody.

[0091] In another embodiment, the "bait portion" is a nucleic acid such as messenger RNA, and the "candidate play" is a nucleic acid binding portion such as a nucleic acid binding protein.

[0092] "Candidate plays" can be natural or modified, of the same or different types.

[0093] A "microtubule-associated protein" (MAP) can also be a polymerized tubulin-binding protein and / or a microtubule-binding protein, but the term "microtubule-binding domain" (MBD) refers to a domain that can specifically and directly bind to microtubules. For the same reason, the "polymerized tubulin-binding portion" refers to a portion that specifically and directly binds to polymerized tubulin.

[0094] Therefore, MBD may contain all possible amino acid sequences that lead to the binding of microtubule-binding proteins to microtubules.

[0095] For reference, the microtubule-binding domain of the present invention may be derived from tau proteins such as tau isoform 2 (accession number: NP_005901.2) which contains the sequence of sequence number 4, including the sequences of sequence numbers 5-8.

[0096] The microtubule-binding domain of the present invention may also be derived from the MAP1A protein (accession number: NP_002364) containing the sequence of Sequence ID No. 9.

[0097] The microtubule-binding domain of the present invention may also be derived from the MAP2 protein (accession number: NP_002365) containing the sequence of Sequence ID No. 10.

[0098] The microtubule-binding domain of the present invention may also be derived from the MAP4 protein (accession number: AAA67361) containing the sequence of SEQ ID NO: 11.

[0099] The microtubule-binding domain of the present invention may also be derived from the MAP6 protein (accession number: NP_149052) containing the sequence of Sequence ID No. 12.

[0100] The microtubule-binding domain of the present invention may also be derived from the EB1 protein (accession number: NP_036457) containing the sequence of Sequence ID No. 13.

[0101] Accordingly, the microtubule-binding domain of the present invention may be selected from the group comprising or consisting of any microtubule-binding domain derived from the sequence of SEQ ID NOs. 4 and SEQ ID NOs. 5-8, MAP1A of the sequence of SEQ ID NO. 9, MAP2 of the sequence of SEQ ID NO. 10, MAP4 of the sequence of SEQ ID NO. 11, MAP6 of the sequence of SEQ ID NO. 12, EB-1 of the sequence of SEQ ID NO. 13, and / or fragments thereof, and combinations thereof.

[0102] Methods for identifying polymerization tubulin-binding moieties and / or microtubule-binding domains in proteins have already been reported in the art. For reference, see Cravchik et al.; Identification of a novel microtubule-binding domain in microtubule-associated protein 1A (MAP1A). J Cell Sci, 107 (Pt 3), 661-72, 1994.

[0103] According to one embodiment, the bait is - A polymerized tubulin-binding moiety containing one or more microtubule-binding domains (MBDs), -Includes the bait portion.

[0104] According to one embodiment, the bait includes a linker (L) region located between the polymerized tubulin-binding portion and the bait portion.

[0105] According to one embodiment, the microtubule-binding domain is selected from the group consisting of tau sequences of SEQ ID NOs. 4 and 5-8, MAP1A sequence of SEQ ID NO. 9, MAP2 sequence of SEQ ID NO. 10, MAP4 sequence of SEQ ID NO. 11, MAP6 sequence of SEQ ID NO. 12, EB-1 sequence of SEQ ID NO. 13, and / or any microtubule-binding domain derived from microtubule-associated proteins, as well as fragments thereof and combinations thereof.

[0106] According to one embodiment, the candidate protein includes a fluorescent protein.

[0107] According to one embodiment, the method according to the present invention includes a step c' between steps c and d, which controls the entry of the compound into the cell. This optional step allows for confirmation that the compound under test can act on the interaction between bait and play. Cell treatment with various compounds involves the use of these molecules diluted in a solvent such as DMSO, and the treatment period can range from a few minutes to several hours, or even several days. This flexibility in the treatment period allows researchers to evaluate the effect of the compound on the cellular process.

[0108] According to one embodiment, steps b and d also include quantifying the interaction between one or more baits and one or more candidate baits, allowing for a comparison between pre- and post-contact interactions with the compound under test, and enabling the determination of the intensity of disruption or stabilization of the compound.

[0109] According to one embodiment, the cells in step b are fixed cells or living cells.

[0110] The methods provided in steps a, b, and d enable the rapid identification of changes in interactions in living, fixed eukaryotic cells, such as mammalian cells.

[0111] The nucleic acid sequence coding the bait is inserted into a plasmid that is fused to a given microtubule-binding domain. Nucleic acid interaction partners can then be detected by screening using antibodies against potential endogenous partners in fixed cells, or by using complementary oligonucleotides.

[0112] In both live and fixed cells, a given bait can be screened for potential partners using a classical GFP-tagged plasmid encoding a given set of plies related to the interaction being investigated, and compared to a control (e.g., tau-GFP protein only).

[0113] High-throughput analysis of protein and nucleic acid interactomes in 96, 384, and 1536-well plates, whether fixed or live cells, is also possible.

[0114] Steps a, b, and d can generate a "microtubule-like pattern" or a "polymerized tubulin-like pattern" if (i) the candidate play can interact with the bait, and (ii) the bait can simultaneously bind to microtubules or polymerized tubulin, respectively, in eukaryotic cells.

[0115] Therefore, the detection of a "microtubule-like pattern" or a "polymerized tubulin-like pattern" indicates an interaction between one or more baits and one or more candidate plays in eukaryotic cells.

[0116] In particular, the detection of candidate plays in eukaryotic cells can generate a "microtubule-like pattern" or a "polymerized tubulin-like pattern" if (i) the candidate play can interact with the bait portion, and (ii) the polymerized tubulin binding portion simultaneously binds to microtubules or polymerized tubulin, respectively, in eukaryotic cells.

[0117] Therefore, according to the embodiment, the detection of a "microtubule-like pattern" or a "polymerized tubulin-like pattern" also indicates an interaction between one or more bait regions and one or more candidate plays in eukaryotic cells.

[0118] According to one embodiment, one or more candidate plays and / or one or more baits are ribonucleic acid.

[0119] According to one embodiment, one or more candidate plays and / or one or more baits are deoxyribonucleic acid.

[0120] According to one embodiment, one or more candidate plays and / or one or more baits are proteins.

[0121] In summary, the combination of one or more candidate plays and one or more baits can be as follows: i. One or more baits: at least protein / One or more candidate plays: at least protein ii. One or more baits: at least protein / One or more candidate plays: at least ribonucleic acid iii. One or more baits: at least protein / One or more candidate plays: at least deoxyribonucleic acid iv. One or more baits: at least ribonucleic acid / One or more candidate plays: at least protein v. One or more baits: at least ribonucleic acid / One or more candidate plays: at least ribonucleic acid vi. One or more baits: at least ribonucleic acid / One or more candidate plays: at least deoxyribonucleic acid vii. One or more baits: at least deoxyribonucleic acid / One or more candidate plays: at least protein viii. One or more baits: at least deoxyribonucleic acid / One or more candidate plays: at least ribonucleic acid ix. One or more baits: at least deoxyribonucleic acid / one or more candidate plays: at least deoxyribonucleic acid According to a preferred embodiment, one or more baits are proteins.

[0122] According to a preferred embodiment, one or more candidate plays are proteins or ribonucleic acids.

[0123] According to one embodiment of the present invention, if play is a protein, one or more candidate play may include a wild-type protein and a mutant version of the same protein.

[0124] According to one embodiment, the wild-type and mutant forms of the protein of interest include the sequence, fluorescent tag, linker, and one or more microtubule-binding domains of the wild-type or mutant form of the protein of interest. Preferably, the one or more microtubule-binding domains include or are selected from the group consisting of tau of the sequences of SEQ ID NOs. 4 and 5-8, MAP1A of the sequence of SEQ ID NO. 9, MAP2 of the sequence of SEQ ID NO. 10, MAP4 of the sequence of SEQ ID NO. 11, MAP6 of the sequence of SEQ ID NO. 12, EB-1 of the sequence of SEQ ID NO. 13, and / or any microtubule-binding domains derived from microtubule-associated proteins, as well as fragments thereof and combinations thereof.

[0125] The methods and uses of the present invention are suitable for evaluating the ability of compounds to disrupt or stabilize interactions between one or more baits and one or more candidate plays in eukaryotic cells, including one bait and one candidate play, one bait and two or more candidate plays, and two or more baits and one candidate play.

[0126] The method of the present invention is also suitable for evaluating the ability of compounds to disrupt or stabilize interactions between multiple baits and multiple candidate plays.

[0127] According to another embodiment, the method is for evaluating the ability of a compound to disrupt or stabilize the interaction between one bait and one candidate play in a eukaryotic cell.

[0128] According to the present invention, "eukaryotic cells" include any eukaryotic cells that contain and / or readily express polymerized tubulin such as microtubules.

[0129] According to certain embodiments, eukaryotic cells are selected from the group including primary cells, stable cell lines, stem cells, and induced pluripotent stem cells (IPS). For example, eukaryotic cells may include or be selected from a list comprising mammalian cells and insect cells.

[0130] In particular, eukaryotic cells may include or be selected from a list comprising HEK cells, NR cells, U2-OS cells, and HeLa cells, especially HeLa cells.

[0131] The cells in step b can be fixed cells or living cells.

[0132] If the cells in step b are living cells, the bait and / or candidate play preferably include a detectable tag such as a fluorescent tag.

[0133] The use of live cells in step b allows for the detection of candidate plays along dynamic polymers of tubulin, such as microtubules, when interacting with the bait.

[0134] If the cells in step b are fixed cells, the method requires an additional fixation step prior to step b. Protocols for obtaining fixed cells are well known in the art.

[0135] In either case, microtubules appear as bright lines on the fluorescence image, which makes it easy to detect the binding of prey on them by, for example, examining the appearance of long bright lines using the prey's fluorescence signal, leading to a characteristic "microtubule-like pattern."

[0136] Candidate plays can be modified or unmodified. Candidate plays can also be homogeneous or heterogeneous.

[0137] The bait and / or candidate play may also contain detectable portions. In particular, the detectable portions may be fluorescent proteins.

[0138] The detectable portion may be selected from the group including GFP, YFP, XFP, RFP, CFP, DsRED, mCherry, luciferase, cyanine dyes such as Cy2, Cy3, or Cy5, fluorescein, rhodamine, and Alexa fluor dye.

[0139] In particular, fluorescent proteins can be selected from the group including GFP, YFP, XFP, RFP, CFP, DsRED, and mCherry.

[0140] Fluorescent proteins are widely known in the field and can be found, for example, in Shaner, Steinbach, and Tsien (A guide to choosing fluorescent proteins; Nat Methods; 2(12):905-9; 2005).

[0141] The bait binds to polymerized tubulin in eukaryotic cells, thereby localizing one or more candidate preplets along the polymerized tubulin, and thus detecting interactions.

[0142] According to another embodiment, the present invention relates to the method defined above, wherein in steps b and d, the determination of the occurrence of prey in eukaryotic cells is made using a detection method selected from the group including binding to an antibody, hybridization with nucleic acids, and / or fluorescence measurement.

[0143] For example, steps b and d, which determine the occurrence of an interaction, can be achieved using a video microscope.

[0144] Steps b and d, which determine the occurrence of interactions, can also be achieved using computer-aided recognition methods, such as the computer-aided recognition method taught by Altinok et al. (Activity analysis in microtubule videos by mixture of hidden Markov models; Computer Vision and Pattern Recognition, IEEE Computer Society Conference, 2, 1662-1669; 2006).

[0145] In addition, the present invention relates to the method defined above, comprising at least one step prior to steps b and d of depolymerizing cellular tubulin using a microtubule depolymerizing agent or cold exposure.

[0146] Microtubule-destroying agents, microtubule-depolymerizing agents, or microtubule-degrading agents, such as nocodazole, vinblastine, vincristine, colchicine, colsemid, podophyllotoxin, rizhoxin, or vinorelbine, can also be used for this purpose.

[0147] Cold exposure is known in the art and generally relates to the step of depolymerizing microtubules by exposing them to low temperatures. Protocols for cold exposure are known in the art and are taught, for example, in Ochoa et al., Cold exposure reveals two populations of microtubules in pulmonary endothelia; Am.J.physiol.Lung Cell.Mol.Physiol;300:L132-L138;2011).

[0148] The polymerization and depolymerization steps defined above may, advantageously, be repeated over time, resulting in a series of alternating phases of bait association and dissociation over time.

[0149] Therefore, regarding the in vitro method according to the present invention, the steps are: a. (i) one or more baits comprising a bait portion and a polymerized tubulin binding portion, and (ii) eukaryotic cells expressing one or more candidate baits, Optionally, a step of depolymerizing cell tubulin using a microtubule depolymerization agent or cold exposure, b. Determine the occurrence of a first interaction between one or more baits, which are bound to polymerized tubulin in eukaryotic cells, and one or more candidate baits in eukaryotic cells. Optionally, a step of depolymerizing cell tubulin using a microtubule depolymerization agent or cold exposure, b'. A step of determining the occurrence of a second interaction between one or more baits, which are bound to polymerized tubulin in eukaryotic cells, and one or more candidate baits in eukaryotic cells. c. Bring eukaryotic cells into contact with the compound, Optionally, a step of depolymerizing cell tubulin using a microtubule depolymerization agent or cold exposure, d. Determine the occurrence of a first interaction between one or more baits and one or more candidate plays in eukaryotic cells, as in step b, and quantify this interaction. Optionally, the step of depolymerizing cell tubulin using a microtubule depolymerization agent or cold exposure, and d'. A step in which a second interaction occurs between one or more baits and one or more candidate plays in a eukaryotic cell, as in step b, and this interaction is quantified, may be arranged as follows:

[0150] In step b, the bait binds to polymerized tubulin in eukaryotic cells, thereby localizing one or more candidate plays along the polymerized tubulin, and thereby detecting the first interaction.

[0151] In step b', the bait binds to polymerized tubulin in eukaryotic cells, thereby localizing one or more candidate plies along polymerized tubulin, and thereby detecting a second interaction.

[0152] In step d, the bait binds to polymerized tubulin in eukaryotic cells, thereby localizing one or more candidate plies along polymerized tubulin, and thereby detecting the first interaction.

[0153] In step d', the bait binds to polymerized tubulin in eukaryotic cells, thereby localizing one or more candidate plays along polymerized tubulin, and thereby detecting a second interaction.

[0154] Naturally, the depolymerization step can be repeated over time.

[0155] The method allows for the detection of not only the interaction between bait and play, but also the dynamics of polymerized tubulin in eukaryotic cells.

[0156] Since bait and play can interact in their living environment, it is also an object of the present invention to not only enable the influence of compounds on the interaction between bait and play, but also to detect post-translational modifications that may occur in eukaryotic cells and affect the potency of the compounds. Accordingly, the method of the present invention may further include, for example, the step of detecting the occurrence of post-translational modifications of the bait moiety and / or candidate play. Post-translational modifications are well known in the art and include, for example, modifications such as phosphorylation events.

[0157] Therefore, this method is also efficient for identifying post-translational events that modify bait-prey interactions, such as phosphorylation, in real time, for example, using video microscopy, as detailed in the example of eIF2B-eIF2A interactions.

[0158] The present invention also relates to the use of polymerized tubulin-binding moieties as a tool for evaluating the ability of compounds to disrupt or stabilize interactions between one or more protein baits and one or more candidate plays in eukaryotic cells.

[0159] The bait binds to polymerized tubulin in eukaryotic cells, thereby localizing one or more candidate plays along the polymerized tubulin, and the ability of a compound to disrupt or stabilize the interaction between one or more protein baits and one or more candidate plays is detected.

[0160] In particular, the present invention relates to the use of polymerized tubulin-binding moieties fused to one or more baits as a tool for evaluating the ability of compounds to disrupt or stabilize interactions between one or more protein baits and one or more candidate plays in eukaryotic cells. The baits are bound to polymerized tubulin in eukaryotic cells, thereby localizing one or more candidate plays along the polymerized tubulin, and thereby the ability of compounds to disrupt or stabilize interactions between one or more protein baits and one or more candidate plays is evaluated.

[0161] More specifically, the present invention relates to the use of at least one microtubule-binding domain fused to one or more baits as a tool for evaluating the ability of a compound to disrupt or stabilize the interaction between one or more protein baits and one or more candidate plays in eukaryotic cells. The baits bind to polymerized tubulin in eukaryotic cells, thereby localizing one or more candidate plays along polymerized tubulin, and thereby evaluating the ability of a compound to disrupt or stabilize the interaction between one or more protein baits and one or more candidate plays.

[0162] According to embodiments, the present invention also relates to the use of multiple microtubule-binding domains, such as a tandem of microtubule-binding domains fused to one or more baits, as a tool for evaluating the ability of a compound to disrupt or stabilize interactions between one or more baits and one or more candidate plays in eukaryotic cells.

[0163] Preferably, the polymerized tubulin-binding moiety and / or microtubule-binding domain may be used in the form of a bait, as further disclosed below.

[0164] Bait This specification also describes the methods and baits suitable for use as defined above.

[0165] In particular, this specification describes (i) polymerized tubulin-binding moieties containing one or more microtubule-binding domains (MBDs), and (ii) baits containing bait moieties.

[0166] According to some embodiments, the "bait portion" does not interact with or co-localize with polymerized tubulin or microtubules in eukaryotic cells.

[0167] According to some embodiments, the bait portion may consist of an antibody or a fragment thereof, or a nucleic acid-binding protein. Therefore, according to some embodiments, the protein bait is - A polymerized tubulin-binding moiety containing one or more microtubule-binding domains, -Includes a bait portion that does not interact with polymerized tubulin in eukaryotic cells.

[0168] According to some embodiments, the bait portion may consist of an antibody or a fragment thereof, or a nucleic acid-binding protein. Therefore, according to some embodiments, the protein bait is - A polymerized tubulin-binding moiety containing one or more microtubule-binding domains, -Includes a bait portion composed of an antibody or a fragment thereof, or a nucleic acid-binding protein.

[0169] A potential barrier to interaction is the reduced accessibility of the bait when fused to the polymerization tubulin binding portion, due to its proximity to the microtubule surface. To maximize the accessibility of the bait to the play, the bait can preferably bind to protruding domains, which are unstructured tails that allow for a nanometer-length gap between the microtubule or polymerization tubulin surface and the bait.

[0170] Accordingly, according to a particular embodiment, the method and bait suitable for use of the present invention comprises a polymerized tubulin-binding moiety containing one or more microtubule-binding domains (MBDs), and a bait moiety.

[0171] Advantageously, the microtubule-binding domain (MBD) is selected from the group including tau of the sequence of SEQ ID NO: 4, MAP1A of the sequence of SEQ ID NO: 9, MAP2 of the sequence of SEQ ID NO: 10, MAP4 of the sequence of SEQ ID NO: 11, MAP6 of the sequence of SEQ ID NO: 12, EB-1 of the sequence of SEQ ID NO: 13, or any other microtubule-binding domain derived from microtubule-associated proteins, as well as combinations thereof.

[0172] According to a preferred embodiment, one or more baits are proteins.

[0173] According to some embodiments, the protein bait comprises a polymerized tubulin-binding moiety containing one or more microtubule-binding domains selected from the group consisting of tau sequences of SEQ ID NOs. 4 and 5-8, MAP1A sequence of SEQ ID NO. 9, MAP2 sequence of SEQ ID NO. 10, MAP4 sequence of SEQ ID NO. 11, EB-1 sequence of SEQ ID NO. 13, and / or fragments thereof, and combinations thereof, and a bait moiety.

[0174] According to some embodiments, the protein bait comprises a polymerized tubulin-binding moiety containing one or more microtubule-binding domains selected from the group consisting of tau sequences of SEQ ID NOs. 4 and 5-8, MAP1A sequence of SEQ ID NO. 9, MAP2 sequence of SEQ ID NO. 10, MAP4 sequence of SEQ ID NO. 11, EB-1 sequence of SEQ ID NO. 13, and / or fragments thereof, and combinations thereof; and a bait moiety that does not interact with polymerized tubulin in eukaryotic cells.

[0175] According to some embodiments, the protein bait comprises a polymerized tubulin-binding moiety containing one or more microtubule-binding domains selected from the group consisting of tau sequences of SEQ ID NOs. 4 and 5-8, MAP1A sequence of SEQ ID NO. 9, MAP2 sequence of SEQ ID NO. 10, MAP4 sequence of SEQ ID NO. 11, EB-1 sequence of SEQ ID NO. 13, and / or fragments thereof, and combinations thereof; and a bait moiety composed of an antibody or fragment thereof, or a nucleic acid-binding protein.

[0176] Therefore, the bait may further include one linker (L) region located between the polymerization tubulin-binding portion and the bait portion.

[0177] Linker (L) regions are generally unstructured domains that enable nanometer-length gaps between the polymerized tubulin surface and the bait, which are important for increasing the accessibility of the bait to the play. Examples of linker regions are known in the art.

[0178] Projection domains are found in microtubule-associated proteins (MAPs) such as MAP2 or tau, and are involved in microtubule bundling and determining the spacing between microtubules. They can also interact with other cytoskeletal structures.

[0179] Preferably, in order to maximize the bait's accessibility to the play, the bait is coupled to a linker region which is a protruding domain from a MAP or fragment thereof, such as tau.

[0180] Therefore, according to a preferred embodiment, the protein bait comprises (i) a polymerized tubulin-binding moiety containing one or more microtubule-binding domains (MBDs), (ii) a projection domain, and (ii) a bait moiety.

[0181] The method and the "protruding domain" suitable for use of the present invention may include or be composed of the N-terminal fragment of tau.

[0182] According to a particular embodiment, the “projection domain” is a tau projection domain or a fragment thereof of sequence number 3.

[0183] The linker region, or projection domain, can be of various lengths and includes any region or domain defined above, ranging from 1 to 150 amino acids in length, such as 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 2 9, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76 ,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117, It includes amino acid lengths of 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, and 150.

[0184] If the bait includes (i) a polymerized tubulin-binding moiety containing one or more microtubule-binding domains, and (ii) a protruding domain, it is preferable that both sites are part of the same protein, particularly the same microtubule-associated protein such as tau.

[0185] The bait portion may or may not contain detectable portions such as fluorescent proteins.

[0186] Advantageously, the bait may further include any detectable portion as defined above, such as a fluorescent label detectable using a fluorescence microscope, at its N-terminus or C-terminus.

[0187] In a non-restrictive manner, the detectable portion may be a fluorescent protein, such as a protein selected from the group including GFP, YFP, XFP, RFP, CFP, DsRED, and mCherry.

[0188] Vectors and cell lines This specification also describes vectors containing expression cassettes suitable for expressing bait according to the use and methods of the present invention.

[0189] In particular, the present invention relates to a vector containing an expression cassette suitable for expressing the protein bait of the present invention.

[0190] In particular, the present invention relates to a vector containing an expression cassette suitable for expressing the ribonucleic acid bait of the present invention.

[0191] In particular, the present invention relates to a vector containing an expression cassette suitable for expressing the deoxyribonucleic acid bait of the present invention.

[0192] In particular, the present invention relates to a vector containing an expression cassette suitable for expressing the protein prey of the present invention.

[0193] In particular, the present invention relates to a vector containing an expression cassette suitable for expressing the ribonucleic acid prey of the present invention.

[0194] In particular, the present invention relates to a vector containing an expression cassette suitable for expressing the deoxyribonucleic acid prey of the present invention.

[0195] Therefore, a vector suitable for use and method of the present invention is suitable for expressing bait, and this is a vector, - An expression cassette encoding a bait moiety fused to a polymerized tubulin-binding moiety containing one or more microtubule-binding domains (MBDs), or alternatively, - A vector comprising an expression cassette containing one or more microtubule-binding domains (MBDs) and encoding a polymerized tubulin-binding moiety that is directly suitable for cloning at least one bait moiety as a fusion bait.

[0196] Expression cassettes are directly suitable for cloning at least one bait moiety as a fusion protein bait according to standard cloning protocols, provided they are suitable for introducing an insert that encodes a polymerized tubulin-binding moiety and a bait moiety in frame. Standard cloning protocols include those that include a restriction enzyme digestion step and / or site-directed recombination.

[0197] The vector may contain one or more selectable markers. The selectable markers are marker genes that confer traits suitable for artificial selection, which are introduced into cells, particularly bacteria or cells in culture.

[0198] According to a preferred embodiment, the expression cassette encodes a polymerized tubulin-binding moiety containing one or more microtubule-binding domains (MBDs) derived from the microtubule-associated protein of the present invention.

[0199] According to the most preferred embodiment, the expression cassette encodes a polymerized tubulin-binding moiety containing one or more microtubule-binding domains (MBDs) derived from protein tau of sequence number 14. A vector suitable for use and method of the present invention is: - An expression cassette encoding a protein library for a bait moiety fused to a polymerized tubulin-binding moiety containing one or more microtubule-binding domains (MBDs), or alternatively, -The fusion protein bait may include one or more microtubule-binding domains (MBDs) and an expression cassette encoding a polymerized tubulin-binding moiety that is directly suitable for cloning nucleic acids encoding a protein library for the bait portion.

[0200] An example of a vector suitable for expression in eukaryotic cells is the Gateway (copyright) pEF-Dest51 plasmid.

[0201] This specification also describes cell lines that are transfected with the vectors described above and / or that stably express baits suitable for the methods and uses of the present invention.

[0202] This specification further describes a kit for evaluating the ability of compounds to disrupt or stabilize interactions between one or more baits and one or more plays in cells, including the method of the present invention and baits, vectors, and / or cell lines suitable for use.

[0203] This specification also describes kits for evaluating the ability of compounds to disrupt or stabilize interactions between one or more baits and one or more plays in cells, including baits, vectors, and / or cell lines suitable for use in the present invention, as well as combinations thereof.

[0204] This specification also describes kits as defined above, which include nucleic acid libraries encoding bait or prey, and / or protein libraries containing bait or prey, and / or deoxyribonucleic acid libraries encoding bait and / or reagents suitable for detecting the development of bait or prey in cells, and combinations thereof.

[0205] This specification provides a kit for evaluating the ability of compounds to disrupt or stabilize interactions between one or more baits and one or more prey in eukaryotic cells, -The vector of the present invention, and / or the cell line of the present invention, -Optionally, a protein library of one or more bait regions. -Optionally, kits containing reagents suitable for detecting the development of protein baits and / or candidate plays in eukaryotic cells are also described.

[0206] This specification provides a kit for evaluating the ability of compounds to disrupt or stabilize interactions between one or more baits and one or more prey in eukaryotic cells, -The vector of the present invention, and / or the cell line of the present invention, - A nucleic acid library or protein library encoding one or more bait moieties, -Optionally, kits containing reagents suitable for detecting the development of protein baits and / or candidate plays in eukaryotic cells are also described.

[0207] According to one embodiment of the present invention, steps a and b are repeated before step c using different vector constructs of one or more baits and one or more candidate plays to determine the occurrence of interactions between one or more baits and one or more candidate plays in eukaryotic cells and to select the correct combination which will be evaluated in subsequent steps to be as close as possible to what occurs in vivo.

[0208] In a preferred embodiment, the vector library (pDEST plasmid) is combinatorially generated in a manner specific to the method according to the present invention and comprehensively to characterize the mode of interaction between PPIs or RPIs, in whatever system.

[0209] In the combinatorial approach, a pDEST vector is generated to consider all possible fusion combinations with respect to a given fluorescent dye.

[0210] For PPI interactions between protein A (bait) and protein B (prey), the following pDEST vectors can be used, and pDON can be protein A (A_pDON) and protein B (B_pDON).

[0211] Four pDEST vectors in Nter: a. Bait: A_pDON-GFP-MBD (SEQ ID NO: 14) b. Bait: A_pDON-RFP-MBD (Sequence ID 14) c. Play: B_pDON-GFP d. Play: B_pDON-RFP

[0212] Four pDEST vectors in Cter: a. Bait: MBD (SEQ ID NO: 14)-GFP-A_pDON b. Bait: MBD (Sequence ID 14)-RFP-A_pDON c. Play:GFP-B_pDON d. Play: RFP-B_pDON

[0213] The following eight plasmids can be obtained from this system.

[0214] About bait: aA-GFP-MBD (SEQ ID NO: 14) bA-RFP-MBD (Sequence ID 14) c. MBD (SEQ ID NO: 14)-GFP-A d. MBD (Sequence ID 14)-RFP-A

[0215] About playing: aB-GFP bB-RFP c.GFP-B d.RFP-B

[0216] The PPI A&B system allows for the comprehensive generation of 16 plasmids to characterize the modes of protein-protein pair interactions for two given fluorescent dyes: GFP and RFP.

[0217] In this case, the combination is linked to the following factors. a.2 Proteins (A / B) b. A combination of two types (bait / play) c. Fusion of two types (Nter / Cter) d. Two fluorescent dyes (GFP / RFP)

[0218] The advantages of repeating steps a and b before step c in order to determine the interaction between one or more baits and one or more candidate plays in eukaryotic cells and to select the correct combination which will be evaluated in the following steps to be closer to what happens in vivo are as follows, in the non-limiting case of PPI interactions: a. Improved cell microscopy results to identify and characterize the interaction binding mode (PPI (bait / play) or RPI: bait only), b. Selection of the best system for conducting molecular screening to identify regulatory mechanisms (stabilization or disruption of bait / prey interactions). c. The possibility of generating more vectors for each of the fluorescent dyes considered. d. Reduction in the number of validation steps (e.g., restriction enzymes) in classical molecular biology, and therefore, increased robustness and acceleration in obtaining the plasmid of interest. e. Application of this approach to identifying protein partners in the context of target identification.

[0219] This replication step relates to any combination of bait and play (DNA, RNA, or protein) according to the present invention.

[0220] General Protocols for High-Throughput Analysis and Evaluation Systems In the case of protein baits, a cDNA sequence encoding a known protein or polypeptide chain called "bait" is inserted into a plasmid that directs the synthesis of this bait within the cell, fused to a microtubule-binding protein (tau, MAP2) or a domain linked to a protrusion domain to facilitate the bait's access to the play. The bait can also be further fused with a fluorescent label such as a fluorescent protein (GFP, RFP, ...).

[0221] A cDNA sequence encoding a polypeptide or protein, known as "prey," is inserted into a plasmid to direct its expression within the cell. Prey may be fused to a fluorescent label, such as a fluorescent protein (i.e., GFP or XFP), for detection in living cells via fluorescence microscopy.

[0222] Arrays of cell samples in 6- to 96-well plates are co-transfected with different plays (see Figure 1) to identify various interaction partners for a given bait in high throughput. The automated detection system enables recognition of plays bound to microtubules to obtain a Boolean output (interacts or does not interact).

[0223] 4) Detection: -i) For detection in live or fixed cells (Figures 2 and 3), co-expression of fluorescent bait and play allows for fluorescence video microscopy detection of play along dynamic microtubules when interacting with bait. Time-lapse images taken less than every minute (15 images per minute) are sufficient to detect microtubule structures with high sensitivity. Sensitivity in live cells is enhanced compared to fixed cells because highly dynamic microtubules can enable detection of play bound to bait via differential fluorescence imaging of play (image at time t0 - image at time t0+CEit, (Desforges et al.; An intercellular polyamine transfer via gap junctions regulates proliferation and response to stress in epithelial cells. Mol Biol Cell 24, 1529-1543; 2013)), a classic technique to highlight microtubule motion, which is useful here to remove fluorescence background from play that is not interacting. To better improve the accessibility of bait to play and the detection sensitivity, microtubules can be depolymerized in living cells using a reversible drug such as nocodazole, which allows the bait to freely explore the cytoplasm and interact with the bait away from the microtubule surface. Once the microtubule depolymerizing drug is washed away, the microtubules rapidly regrow from the centrosome (<2 min), and such a pattern is captured by video microscopy using the fluorescence of play whenever interaction between bait and play occurs. Sensitivity can be enhanced by reducing the fluorescence signal-to-noise ratio using confocal fluorescence microscopy or total internal fluorescence microscopy (TIRF). -ii) For detection in fixed cells (Figure 3), the method allows for fluorescence detection of native prey (endogenous proteins, endogenous RNA, endogenous DNA, or other biomolecules) transported to microtubules thanks to the interaction of native prey with bait. Detection can be performed using a classical optical fluorescence microscope after labeling the prey with a specific antibody (a protein selected to potentially interact with bait and to use a negative control), or after in situ hybridization if the goal is to detect the interaction of nucleic acids (especially RNA including miRNA and mRNA) with bait. This option is also relevant if fluorescent labeling of prey (GFP, RFP, or other labels) may interfere with the prey-bait interaction, or if overexpression of prey may bias the results.

[0224] For detection in fixed cells co-expressing bait and fluorescent prey, fluorescence detection of prey transported to microtubules due to interaction with bait can be directly performed using a classical optical microscope due to its fluorescent labeling.

[0225] In either case, microtubules appear as bright lines on the fluorescence image, which makes it easy to detect the binding of Prey to them by examining the appearance of bright lines of μm length using Prey's fluorescence signal.

[0226] To improve the efficiency of co-expression and to facilitate the study of interactions between given protein baits, stable cell lines expressing the bait can also be used. If bait expression is toxic, an inducible expression system in a stably transfected cell line can be proposed to obtain significant expression within a given time.

[0227] After step c, in which eukaryotic cells are brought into contact with the compound, step d is performed as discussed in steps a and b in this section, the results are compared in step e, and in step f, it is concluded whether the compound disrupts or stabilizes the interaction.

[0228] Compounds to be tested (step c) and potential candidates for treating diseases The in vitro method according to the present invention comprises step c, which involves bringing eukaryotic cells into contact with the compound, which ultimately leads to a modification of the interaction between the bait and the play, allowing for the conclusion that the effect of the compound is either disruption or stabilization of the interaction between one or more baits and one or more candidate plays (step f).

[0229] Contact between eukaryotic cells and compounds is preferably performed by an automated robot using 96, 384, or 1536-well plates. Eukaryotic cells are treated with various compounds diluted in a solvent such as DMSO. Treatment duration can range from several minutes to several hours, or even several days.

[0230] Contact of the compound with eukaryotic cells is preferably carried out by diluting the compound in a solvent, more preferably DMSO solvent. The compound is diluted to different concentrations depending on the compound being tested.

[0231] Next, the compound to be tested is brought into contact with eukaryotic cells.

[0232] The contact period between the compound being tested and eukaryotic cells is typically several minutes to several days.

[0233] According to one embodiment of the present invention, the contact period between the compound to be tested and eukaryotic cells is selected from among 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, or 59 minutes.

[0234] According to one embodiment of the present invention, the contact period between the compound to be tested and eukaryotic cells is selected from among 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, or 24 hours.

[0235] According to one embodiment of the present invention, the contact period between the compound to be tested and eukaryotic cells is selected from among 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, or 31 days.

[0236] The compound may be a natural substance or a chemical substance. In one embodiment, the compound is a small molecule.

[0237] This compound may be a known substance, or it may be derived from a library of substances based on in vivo, in vitro, or in silico data.

[0238] In certain embodiments of the present invention, the method is repeated with different concentrations of compounds to determine the IC50 of a compound for the disruption or stabilization of a tested combination of one or more baits and one or more candidate plays.

[0239] In a particular embodiment of the present invention, the method according to the present invention enables screening a library of molecules to identify potential candidates and performing IC50 on these relevant candidates.

[0240] According to embodiments of the present invention, if one or more baits and one or more candidate plays are characteristic interactions that occur in a particular medical condition, the compound may be a pharmaceutical candidate.

[0241] This invention makes it possible to evaluate the ability of this compound to disrupt or stabilize interactions between one or more baits and one or more candidate prey in eukaryotic cells. Since the combination of prey and bait may be characteristic of the interactions (or lack thereof) that occur in certain disease conditions, this method makes it possible to identify drug candidates that may act on the interactions (or lack thereof) and improve the treatment of disease conditions.

[0242] In examples of the present invention, not limited to the range of combinations addressed by this application, the applicant has been able to identify compounds that have an effect on RPI interactions (protein bait and RNA prey) and PPI interactions (protein bait and protein prey), demonstrating the effectiveness and potential of the methods of the present invention.

[0243] According to one embodiment, the present invention relates to a compound identified by the in vitro method according to the present invention for use as a pharmaceutical.

[0244] In certain embodiments of the present invention, one or more baits and one or more candidate plays are cancer-specific.

[0245] In preferred embodiments of the present invention, cancer is selected from the group including or comprising colorectal cancer, malignant brain tumor, lung cancer, drug-resistant cancer, and pancreatic cancer.

[0246] In this embodiment, the method and use according to the present invention can be used to conclude whether the effect of the compound under test is disruption or stabilization of interactions, and thus whether one or more compounds under test may be of interest as potential drug candidates in a combination of bait and play cancer features.

[0247] According to one embodiment, the compound is intended for use in the treatment of cancer in a subject.

[0248] In certain embodiments of the present invention, one or more baits, preferably protein baits, and one or more candidate baits are characteristic of proteinosis.

[0249] According to one embodiment, the compound is intended for use in treating protein disorders in a subject.

[0250] In preferred embodiments of the present invention, the proteinosis is selected from or includes Alzheimer's disease, Parkinson's disease, Huntington's disease, amyotrophic lateral sclerosis, frontotemporal dementia, TDP-43, or FUS proteinosis.

[0251] In a more preferred embodiment of the present invention, the proteinopathy is a TDP-43 proteinopathy, preferably amyotrophic lateral sclerosis (ALS) or frontotemporal lobar degeneration (FTLD), and the protein of interest is TDP-43.

[0252] According to one embodiment, the mutant forms of TDP-43 are selected from the group consisting of or including A90V, D169G, N267S, G287S, G290A, S292N, G294A / V, G295C / R / S, G298S, M311V, A315T / E, A321G / V, Q331K, S332N, G335D, M337V, Q343R, N345K, G348C / R / V, N352S / T, G357R / S, R361S / T, P363A, G368S, Y347X, G376D, N378D / S, S379C / P, A382P / T, I383V, G384R, W385G, N390S / D, and S393L.

[0253] The method according to the present invention was able to identify compounds associated with TDP-43 proteinopathy.

[0254] Some mutations in the C-terminal domain of TDP-43 promote the transition of the protein from the liquid state of TDP-43 where the protein is present in reversible liquid intracellular compartments to a solid aggregate state where the protein accumulates and forms irreversible solid aggregates within the cell.

[0255] The process according to International Publication No. 2016012451 patent family has been used to investigate mutations involved in the transition of TDP-43 to the solid aggregate state, which is characteristic of ALS pathology.

[0256] Since the mutant forms of TDP-43 showed various defects in cell tests, the approach was for finding compounds that could recover these defects.

[0257] 134 compounds from a library of kinase or phosphatase inhibitors were tested by the method according to the present invention using the system TDP43-GFP-MBD (SEQ ID NO: 14)[wt] / SE3-RFP-MBD (SEQ ID NO: 14).

[0258] Following the initial screening, 13 compounds were selected that either increased protein segregation (i.e., exhibited a solid aggregate state of the TDP-43 variant) or, conversely, increased protein mixing (i.e., exhibited a liquid state of the TDP-43 variant), and no toxicity issues were found.

[0259] Next, 13 selected compounds were tested in three cell studies: enrichment of wt TDP-43 in stress granules and their intracellular localization in the presence of the compound under test, and the effect of the compound under test on alternative splicing of the cystic fibrosis transmembrane conductance regulator (CFTR) gene by wt TDP-43. Finally, the effect of the 13 compounds on the aggregation of the G146A variant of TDP-43 was tested.

[0260] Kinase inhibitors increased protein segregation and aggregation of mutant G146A, while phosphatase inhibitors increased protein mixing and decreased aggregation of mutant G146A. Of these compounds, four showed efficacy in at least two of the four tests: three kinase inhibitors (Ro-mesylate, PKRi, 5-iodotubercidin) and one phosphatase inhibitor (9,10-phenanthrenequinone). These results therefore indicate that phosphatase inhibitors identified by this method can restore defects observed in ALS, such as the accumulation of mutant forms of TDP-43.

[0261] These results demonstrate that the methods according to the present invention can be used to test mutational forms of proteins of interest, particularly those accumulating in subjects suffering from proteinopathy, and to identify compounds that can be used to treat proteinopathy. [Examples]

[0262] Example 1: Targeted Proteins: Protein Interactions for Identifying Modulators in the Cellular Environment The widely studied system p53 / MDM2 (Moll and Petrenko, 2003; Nayak et al., 2018; Zhao et al., 2015) was selected as a proof-of-concept to investigate the performance of MT bench technology for screening PPI inhibitors using a high-content screening system. High-content screening (HCS) is an image-based method used in drug discovery to identify phenotypic-modifying molecules (biogens or small molecules) using several cellular parameters simultaneously as readout (Fraietta and Gasparri, 2016; Mattiazzi Usaj et al., 2016). An analytical pipeline was developed to measure colocalization on individual cells from HCS images by extracting correlation scores. Correlation scores were calculated using the Pearson Correlation Coefficient (PCC) between enhanced texture intensities from play and bait. The inventors measured a good Z' factor of 0.69 (0.5–1, indicating a superior assay) (Zhang, 1999; Zhang et al., 2000), reflecting the robustness of the technique for screening small inhibitors of p53 / MDM2 with high content capability. The inventors then observed that the interaction detected on microtubules depended on specific residues of the MDM2 binding site with p53. Having demonstrated the effectiveness of the MT bench for quantifying p53 / MDM2 interaction, the inventors initiated screening a small set of 10 molecules with known activity against MDM2, as previously characterized by the HTRF assay (Gicquel et al., 2018) for hit identification. Both false-negative and false-positive molecules identified from the HTRF assay were successfully detected and independently confirmed by measuring the accumulation of p53 and p53-related proteins. The MT bench assay was also used to calculate IC50 from HCS images. The inventors obtained cellular IC50 values ​​of 74 nM and 426 nM for the known p53 / MDM2 inhibitors, RG7112, and Nutrin-3a (Vu et al., 2013).These values ​​are consistent with those found in the literature from in vitro HTRF assays (Liu et al., 2019, Neochoritis et al., 2019a).

[0263] (Vu et al., 2013), and also obtained from cellular nanoBRET assays (Itatani and Olsen). The inventors also used the results to characterize the inhibition constants for all tested molecules that enable structural activity relationships (SARs). In summary, the results presented in this study acknowledge the MT bench technique in an HCS setting as a robust method for screening and characterizing PPI modulators in cells.

[0264] Methods and materials: Genetic construction Cloning of the gene of interest was performed using Invitrogen's gateway system. Sets of pDEST plasmids derived from pDEST-CMV-N-EGFP and pDEST-CMV-C-EGFP were designed using synthetic genes purchased from Twist Bioscience to generate constructs with N-terminal and C-terminal fusion proteins for play (the protein of interest fused with the eGFP protein) and bait (the protein of interest fused with the RFP fluorescent dye and the microtubule-binding domain (MBD (SEQ ID NO: 14)) corresponding to the longest isoform of human tau protein as previously described (Boca et al., 2015)). Human p53 and MDM2 full-length cDNA were amplified by PCR using primers that either contained the kozak sequence and did not contain a stop codon to generate a "fusion" entry clone in pDONR221 for recombination into the pDEST-CMV-C vector (for the C-terminal fusion protein), or did not contain ATG but contained a stop codon to generate a "proximity" entry clone for recombination into the pDEST-CMV-N vector (for the N-terminal fusion protein).

[0265] For the MT bench, we selected a p53 construct (SEQ ID NO: 15) (p53-FL_C_RFP-MBD) as a bait with a C-terminal fusion of RFP-MBD, and an MDM2 construct (SEQ ID NO: 21) (MDM2-FL_C_GFP) as a play with a C-terminal fusion of GFP. We obtained p53 mutants by subcloning synthetic genes purchased from Twist Bioscience to generate mutant constructs: p53F19A(SEQ ID NO: 17)_C_RFP-MBD and p53W23A(SEQ ID NO: 16)_C_RFP-MBD with single mutations, p53F19AW23A(SEQ ID NO: 18)_C_RFP-MBD and p53W23AL26A(SEQ ID NO: 19)_C_RFP-MBD with double mutations, and p53F19AW23AL26A(SEQ ID NO: 20)_C_RFP-MBD with triple mutations. All generated plasmids were purified using the MACHEREY-NAGEL NucleoSpin Plasmid DNA Purification Kit, and the integrity of different DNA sequences was confirmed by sequencing using Eurofins Genomics.

[0266] Cell plate generation The MT bench was performed using the human osteosarcoma (U-2 OS) cell line, which is the standard for cell imaging. U-2 OS cells were cultured at 37°C in a humidified atmosphere under 5% CO2 in Dulbecco's modified Eagle's medium (DMEM, Life Technologies) supplemented with 10% FBS (fetal bovine serum, Life Technologies) and 1% penicillin / streptomycin.

[0267] Cell plate preparation (seeding, transfection, and processing) was automated using an Agilent BRAVO liquid handler equipped with a 96-LT (Large Tips) head. U-2 OS cells were seeded at a density of 18,000 cells / well on a black 96-well plate cell carrier ultra (PerkinElmer) and incubated in a humidified incubator at 37°C and 5% CO2 for 24 hours. Then, the positive control (untreated condition) was co-transfected for 18 hours with 1.2 μg of human p53-FL (SEQ ID NO: 15)_C_RFP-MBD and 0.3 μg of human MDM2-FL (SEQ ID NO: 21)_C_GFP plasmid, or the negative control (non-interaction condition / treated condition) was co-transfected with 1.2 μg of human p53-F19AW23AL26A (SEQ ID NO: 20)_C_RFP-MBD and 0.3 μg of human MDM2-FL (SEQ ID NO: 21)_C_GFP plasmid using 0.5 μg of Lipofectamine 2000 (Invitrogen) in optiMEM.

[0268] To identify hits, cells were treated at 37°C for 2 hours with two concentrations (1 μM and 10 μM) of RG7112, Nutrin-3a, and eight other compounds (Rac-7b'', Rac-4a'', Rac-4c', Rac-4d', Rac-8', (-)-8', (+)-8', Rac-MG-1133) diluted from 50 mM or 10 mM stock solutions in culture medium, using 0.1% DMSO for the control wells. IC 50 For the control wells, cell treatment was carried out by adding 0.1% DMSO to the culture medium at 37°C for 2 hours, along with 10 different concentrations of the same compound ranging from 0.001 μM to 25 μM or 0.01 μM to 100 μM.

[0269] Eighteen hours after transfection, cells were fixed with 100% ice-cold methanol at -20°C for 10 minutes, washed with PBS, and then further fixed at room temperature for 10 minutes with 4% paraformaldehyde (PFA) in freshly prepared PBS. This double fixation (methanol / PFA) is used as the best method for elucidating microtubules and co-localization events. After washing with PBS, cell nuclei were stained with DAPI (0.1 μg / mL) at room temperature for 5 minutes, and the samples were retained in PBS for fluorescence microscopy imaging. Cellular fluorescence signals were acquired using a PerkinElmer Opera Phenix HCS imaging system with a 40× water-immersion 1.1 NA objective lens and Harmony® software.

[0270] Image and statistical analysis To quantify the level of co-localization between MBD-fused protein bait and putative protein play, we developed an analytical pipeline adapted from a previously described method (French et al., 2008). After image acquisition, sequential fundamental elements from Harmony® software were used for image segmentation (finding nuclei, finding cytoplasm, and selecting populations by threshold against cell area, along with maximum and average intensities of GFP and RFP). Both images were then filtered using a SER ridge filter to highlight microtubule structures in the cytoplasm.

[0271] To quantify the degree of co-localization between fluorescent dyes, the inventors used the Pearson correlation coefficient (PCC) (Benesty et al., 2009). The inventors also used Mander's overlap coefficient (MOC) (Manders et al., 1993) (data not shown). These two coefficients are mathematically similar, but the PCC seems to be more precise in quantifying co-localization by using the deviation from the mean (Adler and Parmryd, 2010). The PCC is a well-established correlation score for measuring co-localization and is calculated as follows

[0272] [Number] In the formula, R i and G i are pixel intensity values, and R av and G av are the means of the red and green channels, respectively. The value ranges from +1 (perfect correlation) to -1 (perfect inverse correlation), and 0 indicates no co-localization. Therefore, the PCC value reflects the co-localization in the cytoplasm between the bait protein and the prey protein and translates the interaction on the microtubule network.

[0273] In the case of hit identification, raw data (PCC scores) were normalized (Kevorkov and Makarenkov, 2005) to enable comparison between experimental plates. This normalization was performed by extrapolating the PCC value relative to the control by using the means of the negative control wells and the positive control wells.

[0274] The Z' factor is related to the separation band between the distributions of the positive and negative control values. It is calculated by the following formula:

[0275] [Number] In the formula, σC+ and σC− are the standard deviations of the positive and negative controls, respectively, and μC+ and μC− are their means (Zhang, 1999; Zhang et al., 2000). The Z factor is the equivalent of the Z’ factor when the sample is used instead of the negative control. The quality of the HTS assay is classified according to the values of the obtained Z’ and Z factors. A value of 0.5 is the generally accepted threshold required for a good-quality screening.

[0276] The mean ± SD of the independent experiments in the quadruplicate experiments was represented as a histogram. R statistical software and GraphPad Prism® software were used for data calculation, statistical analysis, and plotting. Statistical comparisons were performed using the Student's t-test in GraphPad Prism®. The following symbols were used: * p < 0.1, ** p < 0.01, *** p < 0.001, **** p < 0.0001.

[0277] Western blotting Human osteosarcoma (U-2 OS) was cultured in DMEM (Life Technologies) supplemented with 10% FBS (Life Technologies) and 1% penicillin / streptomycin, under a humidified atmosphere at 37 °C and 5% CO2, at 0.5 × 10 6Cells were cultured in a 6-well plate at a cell density of cells / well. After 24 hours of incubation, the compound was added at a final concentration of 10 μM. Six hours after treatment, the cells were lysed in RIPA buffer (Sigma) supplemented with a protease inhibitor cocktail (Sigma). After protein administration using a BCA protein assay kit (Pierce®), an equal volume of protein lysate (10 μg) was loaded into the wells of a 12% SDS-PAGE gel after denaturation. After protein separation by electrophoresis (120 V, 45 min), the proteins were transferred to a PVDF membrane (Amersham Hybond-P Membrane, GE Heathcare) by electroblotting (100 V, 1 hour 15 min) using a Trans-Blot® Turbo® Blotting System. The membrane was blocked at room temperature for 2 hours with 5% skim milk in 0.1% Tween-PBS (T-PBS). Subsequently, the membranes were examined with the following primary antibodies: mouse monoclonal anti-human p53 (DO-1, #sc-126, Santa Cruz Biotechnology Inc), rabbit anti-human p21 (C-19, #sc-397, Santa Cruz Biotechnology Inc), and mouse monoclonal anti-actin (#A5441, Sigma). As secondary antibodies, the inventors used horseradish peroxidase (HRP)-conjugated goat anti-mouse IgG (#PO447, Dako) and goat anti-rabbit IgG (#AP187P, Millipore). The blots were developed using the ECL Western Blotting Kit (Bio-Rad), visualized using the ChemiDoc MP imaging system (Bio-Rad), and quantified.

[0278] result: High-content screening of p53 / MDM2 interactions using microtubule bench cell-based assays To perform the MT bench technique on the HCS system (Figure 4A), the inventors used the well-known PPI, p53-MDM2 (Moll and Petrenko, 2003; Nayak et al., 2018; Zhao et al., 2015), as a model. To analyze the interaction between p53 and MDM2, the inventors transiently expressed both p53 and MDM2 fusion proteins in U-2OS cells. The inventors were able to transfer the assay from microscopy to the HCS system, enabling robust quantification of binding events.

[0279] The bait protein, full-length p53 (SEQ ID NO: 15), was fused to both RFP (red fluorescent protein) and MBD (microtubule-binding domain) (SEQ ID NO: 14) derived from MAP (microtubule-associated protein) to transport P53 onto microtubules, allowing observation of the bait along the microtubule network in the cell using the red channel. In contrast, the prey protein MDM2 was not transported onto microtubules and was fused alone to GFP (green fluorescent protein) to enable its detection by fluorescence microscopy using the green channel. When the red and green channels were merged, both fluorescent signals co-localized along the same microtubule, which corresponds to the interaction between bait p53 and prey MDM2 (Figure 4A). While the native partial localization of p53 and MDM2 is mostly in the nucleus, the ability of techniques to rearrange nuclear proteins in the cytoplasm allowed for the detection of p53 / MDM2 interactions along microtubules. As a negative control, a plasmid containing full-length p53 as bait (p53-FL(SEQ ID NO: 15)_C_RFP-MBD) was mutated with three hotspot residues (Phe19, Trp23, and Leu26) that cause loss of interaction with MDM2 (Bista et al., 2013). Subsequently, we observed loss of co-localization in microtubules accompanied by relocalization of MDM2 into the nucleus (Figure 4B). Interestingly, a single mutation in one of these three hotspot residues was sufficient to cause loss of more than 90% of the interactions detected on microtubules (Figure 5).

[0280] To provide an interaction score, the inventors developed an image analysis pipeline for measuring the co-localization of p53 and MDM2 on microtubules using Harmony® software from PerkinElmer, as described in the Materials and Methods section (Figure 6). For this purpose, the inventors measured the Pearson correlation coefficient (PCC) in the cytoplasm of each cotransfected cell (Adler and Parmryd, 2010; Benesty et al., 2009), and then calculated the average PCC per well to give a statistically reliable score (Figure 6). After detecting the interaction between p53 and MDM2 in the cellular context, the inventors investigated whether perturbations by small molecules could be quantified on the MT bench using a well-characterized MDM2 inhibitor, RG7112 (Liu et al., 2019; Vu et al., 2013). When cells were treated with gradually increasing concentrations of RG7112 ([0.05 μM], [0.5 μM], [5 μM], and [50 μM]), the inventors observed a dose-dependent decrease in p53 and MDM2 interaction scores by MT bench assay (Figure 4C). The dose-dependent response to RG7112 was characterized by weak inhibition of the interaction at low concentrations ([0.05 μM]), moderate inhibition at medium concentrations ([0.5 μM]), and strong inhibition of the interaction at high concentrations ([5 μM] and [50 μM]) (Figure 4C). In addition, the inventors noticed a gradual disappearance of MDM2 on the microtubule network in the green channel, reflecting a progressive loss of interaction between p53 and MDM2 induced by small molecules. Interestingly, in the MT bench assay, the presence of a p53 / MDM2 inhibitor not only induces loss of interaction but also correlates with the progressive relocalization of MDM2 into the nucleus.

[0281] Next, the inventors tested whether the MT bench assay could be used to screen PPI inhibitors using HCS. They then measured the Z' factor, a statistical parameter widely used in the 96-well plate format to validate the assay's suitability for high-throughput screening (Zhang, 1999) (see Methods). Using the p53 triple mutant (p53-F19AW23AL26A) as a negative control, the resulting Z' factor was 0.69, which is above 0.5, suggesting a good assay threshold (Figure 15A). The system was also performed on a 384-well plate format with Z'=0.65 (Figure 7A). It is noteworthy that the Z' factor calculated after using the G3BP1 protein as the negative control bait was 0.59. This value falls within the same range as the p53 triple mutant that loses PPI, meaning that the inventors were able to use another unrelated protein as a negative control without prior knowledge of the hotspot residues that prevent specific PPI (Figures 7B-7C). In addition, the inventors measured the Z factor (Zhang et al., 2000) when the negative control corresponded to a sample in which complete disruption of the interaction complex by the characterized inhibitor was obtained. Here, the inventors measured the minimum colocalization score with 5 μM RG7112 and measured a Z' factor of 0.67 when this condition was used as the negative control (Figure 15B). In summary, this dataset demonstrates that the system p53 / MDM2 setup using MT bench technique on HCS is robust, reproducible, and highly sensitive for hit identification.

[0282] Consideration: Direct visualization and quantification of full-length protein interactions in human cells offers specific advantages compared to other methods. In particular, we demonstrate that MT bench technology enables the identification of hotspot residues and small molecule modulators of PPIs that represent strategically important targets for drug discovery. Furthermore, we also measure dose-response effects and their cellular activity (IC). 50) can also be extracted. Such cell structure-activity relationship (SAR) information can pave the way for modulator optimization by using cheminformatics tools.

[0283] A key feature of the MT bench technique is its ability to reduce the number of false hits compared to HTRF, likely due to its high sensitivity and the fact that the interaction involves full-length proteins and occurs in human cells. Indeed, based on the screening reported herein, the molecule Rac-MG-1133 was not identified as a potential p53 / MDM2 disruptor in HTRF, but it significantly inhibited the p53 / MDM2 interaction in cells. Therefore, Rac-MG-1133 can be considered a false negative. The MT bench assay can distinguish the effects of enantiomers from racemic mixtures, as exemplified for the molecule Rac-8. The dextrorotatory enantiomer (+)-8' induced strong inhibition in both assays, while the levorotatory enantiomer (-)-8' was a false positive in the MT bench assay, most likely because this enantiomer is not efficient in cells but still exhibits inhibition in HTRF in vitro.

[0284] The robustness of the MT bench technique lies in the fact that it is based on single-cell analysis. Therefore, the output per well is an average of hundreds of events, which leads to a quantitatively accurate scoring of the interaction between two proteins, and thus its perturbation by the modulator. In fact, similar Z' factors using either an unbound protein as bait, a non-interacting mutant as a control, or a small molecule that results in the complete disruption of the interaction demonstrate the reproducibility of the MT bench assay (Figures 15A-15B and 7C).

[0285] In addition to the variety of negative controls available, the simplicity and flexibility of the technique of working with full-length proteins in cells allows for screening protein interactions without prior structural data or knowledge of the interaction interface. Another advantage of using full-length proteins is the fact that either orthosteric or allosteric modulators can be identified. The MT bench can also be used to confirm interactions between two partners and to perform hotspot identification in cells, providing valuable information for the discovery of PPI inhibitors using a virtual screening approach (Guo et al., 2014).

[0286] Here, we have shown that the MT bench is a reliable cell-based technology for the discovery of PPI inhibitors, providing a solution for the discovery of novel chemical scaffolds with improved ADMET properties (Neochoritis et al., 2019b), and can assist in the clinical development of MDM2 / MDMX inhibitors for cancer treatment.

[0287] In conclusion, the inventors demonstrated that MT Bench is a robust, quantitative, and highly sensitive cell-based screening assay. Combined with in silico expertise, this technology can significantly facilitate early drug discovery in a time-efficient manner for challenging targets such as PPIs. By reducing the risk of false positives and false negatives in the hit identification phase and by characterizing the mechanism of action of PPI modulators, MT Bench can facilitate the optimization of small molecule leads with properties suitable for preclinical studies.

[0288] Example 2: Targeted protein: Protein interaction to identify modulators and determine IC50 data in cells. Methods and materials: Genetic construction Cloning of the gene of interest was performed using Invitrogen's gateway system. To generate combinations of constructs with N-terminal and C-terminal fusion proteins for play (the protein of interest fused with the eGFP protein) and bait (the protein of interest fused with the RFP fluorescent dye and the microtubule-binding domain (MBD (SEQ ID NO: 14)) corresponding to the longest isoform of human tau protein as previously described (Boca et al., 2015)), we designed sets of pDEST plasmids derived from pDEST-CMV-N-EGFP and pDEST-CMV-C-EGFP using synthetic genes purchased from Twist Bioscience. Human p53 and MDM2 full-length (SEQ ID NO: 21) cDNA were amplified by PCR using primers that either contained the kozak sequence and did not contain a stop codon to generate a "fusion" entry clone in pDONR221 for recombination into the pDEST-CMV-C vector (for the C-terminal fusion protein), or did not contain ATG but contained a stop codon to generate a "proximity" entry clone for recombination into the pDEST-CMV-N vector (for the N-terminal fusion protein).

[0289] For the MT bench, we selected a p53 construct as a bait with a C-terminal fusion of RFP-MBD (p53-FL(SEQ ID NO: 15)-RFP-MBD(SEQ ID NO: 14)) and an MDM2 construct as a play with a C-terminal fusion of GFP (SEQ ID NO: 21) (MDM2-FL-GFP). We subcloned p53 mutants from synthetic genes purchased from Twist Bioscience to generate mutant constructs: p53F19A-RFP-MBD and p53W23A-RFP-MBD with single mutations, p53F19AW23A-RFP-MBD and p53W23AL26A-RFP-MBD with double mutations, and p53F19AW23AL26A-RFP-MBD with triple mutations. All generated plasmids were purified using the MACHEREY-NAGEL NucleoSpin Plasmid DNA Purification Kit, and the integrity of different DNA sequences was confirmed by sequencing using Eurofins Genomics.

[0290] Cell plate generation The MT bench was performed using the human osteosarcoma (U-2 OS) cell line. U-2 OS cells were cultured at 37°C in a humidified atmosphere at 5% CO2 in Dulbecco's modified Eagle medium (DMEM, Life Technologies) supplemented with 10% FBS (fetal bovine serum, Life Technologies) and 1% penicillin / streptomycin.

[0291] Cell plate preparation (seeding, transfection, and processing) was automated using an Agilent BRAVO liquid handler with a 96-LT (Large Tips) head. U-2 OS cells were seeded at a density of 18,000 cells / well into black 96-well cell carrier ultra (PerkinElmer) plates and incubated in a humidified incubator at 37°C and 5% CO2 for 24 hours. The cells were then co-transfected for 18 hours with the indicated amount of plasmid of interest using 0.5 μg of Lipofectamine 2000 (Invitrogen) in optiMEM.

[0292] For IC50, cell treatment was performed by adding 0.1% DMSO to the culture medium in the control well, along with RG7112 at 10 different concentrations ranging from 0.001 μM to 25 μM, at 37°C for 2 hours.

[0293] Eighteen hours after transfection, cells were fixed with 100% ice-cold methanol at -20°C for 10 minutes, washed with PBS, and then further fixed at room temperature for 10 minutes with 4% paraformaldehyde (PFA) in freshly prepared PBS. This double fixation (methanol / PFA) is used as the best method to elucidate microtubules and co-localization events. After washing with PBS, cell nuclei were stained with DAPI (0.1 μg / mL) at room temperature for 5 minutes, and the samples were retained in PBS for fluorescence microscopy imaging. Cellular fluorescence signals were acquired using a PerkinElmer Opera Phenix HCS imaging system with a 40× water-immersion 1.1 NA objective lens and Harmony® software.

[0294] Image and statistical analysis To quantify the enrichment level of protein play on protein bait fused with MBD, the inventors developed an analytical pipeline adapted from a previously described method (French et al., 2008). Image acquisition and statistical analysis of cell fluorescence signals were performed using a Perkin Elmer Opera Phenix Plus high-content imaging system on a 40× immersion objective lens with a numerical aperture of 1.1, enabling good resolution in confocal mode. Several fields of view were adopted for each well, resulting in thousands of cells to be analyzed per well in 96 or 384-well plate formats, for the strongest statistical significance. Data were extracted using SImA® software with an analytical pipeline that included sequential fundamental elements for image segmentation, selection of populations of interest, and calculation of signal enrichment on microtubules (MTs). Enrichment is calculated on identified spots using an RFP (red fluorescent protein) channel corresponding to the signal of the bait protein localized to MTs due to fusion of the bait protein with a microtubule-binding domain (MBD). Spots representing segments of MT were selected based on their shape and the intensity of the RFP channel signal (corresponding to the presence of bait on the MT). Other proteins were detected by GFP (green fluorescent protein) and could be transported onto the MT due to potential interactions with bait. Calculated RFP and GFP intensities on the spots and in the cytoplasm were extracted from SImA® software and then processed to measure the slope of bait protein enrichment on the MT (average spot intensity divided by average cytoplasmic intensity versus average bait spot intensity).

[0295] Spotfire® data analysis software was used for data calculation, statistical analysis, and plotting. Raw data was normalized using minimum-maximum normalization (using positive and negative controls as maximum and minimum) to enable comparisons between experimental plates (Kevorkov and Makarenkov, 2005).

[0296] Results and analysis: To implement MT bench technology on an HCS system, the inventors used the well-known PPI, p53-MDM2 (Moll and Petrenko, 2003; Nayak et al., 2018; Zhao et al., 2015), as a model. To analyze the interaction between p53 and MDM2, the inventors transiently expressed both p53 and MDM2 fusion proteins in U-2OS cells.

[0297] The bait protein, p53, is fused to both RFP (red fluorescent protein) and MBD (microtubule-binding domain) derived from MAP (microtubule-associated protein) to transport P53 onto microtubules, enabling observation of the bait along the microtubule network in the cell using the red channel. In contrast, the bait protein MDM2 is not transported onto microtubules and is fused alone to GFP (green fluorescent protein), enabling its detection by fluorescence microscopy using the green channel (Figure 4A).

[0298] When the red and green channels were merged, both fluorescent signals co-localized along the same microtubules, which corresponds to the interaction between bait p53 and play MDM2. While the native partial localization of p53 and MDM2 is mostly in the nucleus, the ability of techniques to rearrange nuclear proteins in the cytoplasm allowed for the detection of p53 / MDM2 interactions along microtubules. A plasmid containing full-length p53 as bait (p53-FL-RFP-MBD) was mutated. When the red and green channels were merged, both fluorescent signals co-localized along the same microtubules, which corresponds to the interaction between bait p53 and play MDM2. While the native partial localization of p53 and MDM2 is mostly in the nucleus, the ability of techniques to rearrange nuclear proteins in the cytoplasm allowed for the detection of p53 / MDM2 interactions along microtubules. A plasmid containing full-length p53 (p53-FL-RFP-MBD) was mutated with three hotspot residues (Phe19, Trp23, and Leu26) that caused a loss of interaction with MDM2 (Bista et al., 2013). Subsequently, the inventors observed a loss of co-localization in microtubules accompanied by the re-localization of MDM2 into the nucleus (Figure 4B).

[0299] To quantitatively estimate protein-protein interactions, we developed an image analysis pipeline for measuring the enrichment of MDM2 on microtubules via interaction with p53, as described in the Materials and Methods section, and tested whether the MT bench assay could be used for screening PPI inhibitors using HCS.

[0300] The inventors first measured the enrichment of MDM2 into microtubules when different p53 mutants were overexpressed (Figure 8).

[0301] The inventors evaluated the enrichment levels under each condition (Figure 9). Expression of the p53 mutant led to less enrichment on microtubules, close to the negative control enrichment level. Under conditions in which the single mutant p53-L26A-RFP-MBD or the triple mutant p53-F19AW32AL26A-RFP-MBD was expressed, MDM2 was never found on microtubules, meaning these mutants completely abolish protein-protein interactions. These data demonstrate that MT bench technology enables the identification of hotspot residue modulators of PPIs.

[0302] After detecting the interaction between p53 and MDM2, the inventors investigated whether perturbations by a small molecule could be quantified using the MT bench assay with a well-characterized MDM2 inhibitor, RG7112 (Liu et al., 2019; Vu et al., 2013). The inventors determined the IC50 of this molecule in cells at 10 different concentrations using the MT bench assay. The calculated IC50 value for RG7112 was 0.074 ± 0.033 μM, which is consistent with the IC50 reported from the HTRF assay (IC50 = 0.018 μM) (Vu et al., 2013). Several cell data have also been reported using the MTT assay for RG7112 (IC50 = 0.4 μM) (Vu et al., 2013). (These results demonstrate that the MT bench technique has sufficient sensitivity to detect perturbations in cells even at the lowest inhibitor concentrations, enabling more accurate IC50 data in cells.)

[0303] Example 3: Targeted RNA: Protein interactions to identify modulators and determine IC50 data in cells. Methods and materials: Genetic construction Cloning of the genes of interest was performed using Invitrogen's gateway system. Sets of pDEST plasmids derived from pDEST-CMV-N-EGFP and pDEST-CMV-C-EGFP (pDEST-N-MBD-EGFP and pDEST-CMV-C-EGFP-MBD) were designed using synthetic genes purchased from Twist Bioscience to generate construct combinations in which the bait protein is fused to EGFP at either the N-terminus or C-terminus, and then fused to the microtubule-binding domain (MBD) corresponding to the longest isoform of the human tau protein, as previously described (Boca et al., 2015). Human YB1FL (full length) cDNA was amplified by PCR using primers that either contained the kozak sequence and did not contain a stop codon to generate a "fusion" entry clone in pDONR221 for recombination into the pDEST-C-GFP-MBD vector (for the C-terminal fusion protein), or did not contain ATG but contained a stop codon to generate a "proximity" entry clone for recombination into the pDEST-CMV-N vector (for the N-terminal fusion protein). YB1ΔCSD (SEQ ID NO: 23) cDNA (with amino acids 52-129, corresponding to the CSD domain of YB1, deleted) was synthesized by Eurofins Genomics, and the construct was then generated in the same manner as for YB1 FL (SEQ ID NO: 22).

[0304] All generated plasmids were purified using the MACHEREY-NAGEL NucleoSpin Plasmid DNA Purification Kit, and the integrity of different DNA sequences was confirmed by sequencing using Eurofins Genomics.

[0305] Cell plate generation The MT bench was performed using HeLa cells (ATCC CCL-2). HeLa cells were cultured at 37°C in a humidified atmosphere under 5% CO2 in Dulbecco's modified Eagle medium (DMEM, Life Technologies) supplemented with 10% FBS (fetal bovine serum, Life Technologies) and 1% penicillin / streptomycin.

[0306] Cell plate preparation (seeding, transfection, and processing) was automated using an Agilent BRAVO liquid handler with a 96-LT (Large Tips) head. HeLa cells were seeded at a density of 5000 cells / well into black 384-well cell carrier ultra (PerkinElmer) plates and incubated in a humidified incubator at 37°C and 5% CO2 for 24 hours. The cells were then transfected with 0.5 μg of the plasmid of interest for 18 hours using 0.1 μg of Lipofectamine 2000 (Invitrogen) in optiMEM.

[0307] Next, the cells were first fixed with 100% ice-cold methanol at -20°C for 10 minutes, washed with PBS, and then further fixed at room temperature for 10 minutes with 4% PFA in freshly prepared PBS. After fixation, the cells were incubated at 37°C for 2 hours for RNA visualization using oligo-dT-[Cy3] diluted in 2× SSC, 1 mg / ml-1 yeast tRNA, 0.005% BSA, 10% dextran sulfate, and 25% formamide. The washing step was performed using 4× and then 2× SSC buffer (0.88% sodium citrate, 1.75% NaCl, pH 7.0). Cell nuclei were stained with DAPI (0.1 μM) at room temperature for 5 minutes.

[0308] After washing with PBS, the cell nuclei were stained with DAPI (0.1 μg / mL) at room temperature for 5 minutes, and the samples were retained in PBS for fluorescence microscopy imaging. Cellular fluorescence signals were acquired using a PerkinElmer Opera Phenix Plus HCS imaging system with a 40× 1.1 NA objective lens and Harmony® software.

[0309] Image analysis To quantify the enrichment level of mRNA play on protein bait fused with MBDs, we developed an analytical pipeline adapted from a previously described method (French et al., 2008). Image acquisition and statistical analysis of cell fluorescence signals were performed using a Perkin Elmer Opera Phenix Plus high-content imaging system on a 40× immersion objective lens with a numerical aperture of 1.1, allowing for good resolution in confocal mode. Twenty fields of view were employed per well, resulting in thousands of cells per well to be analyzed for the strongest statistical significance. Data were extracted using SImA® software with an analytical pipeline that included sequential foundational elements for image segmentation, selection of populations of interest, and calculation of signal enrichment on microtubules (MTs). Enrichment was calculated on identified spots using a GFP (green fluorescent protein) channel corresponding to the signal of the bait protein localized to MTs due to fusion of the bait protein with microtubule-binding domains (MBDs). Spots representing segments of MTs were selected based on their shape. mRNA was detected using Cy3-labeled poly(dT) and could be transported onto MTs due to potential interactions with bait. Calculated GFP and Cy3 intensities in the spots and cytoplasm were extracted from SImA® software and then processed to measure the relative enrichment of mRNA on MTs (average spot intensity divided by average cytoplasmic intensity (GFP), and then further divided by average bait spot intensity divided by average cytoplasmic intensity (Cy3)).

[0310] Spotfire® data analysis software was used for data calculation, statistical analysis, and plotting. Raw data was normalized using minimum-maximum normalization (using positive and negative controls as maximum and minimum) to enable comparisons between experimental plates (Kevorkov and Makarenkov, 2005).

[0311] statistical analysis To determine whether there was a significant difference in relative enrichment between the conditions, the inventors performed Student's t-test. This statistical test allows for the comparison of the means of two groups and the determination of whether the difference between them is statistically significant. The following hypothesis was formulated for the t-test.

[0312] Null hypothesis (H0): There is no significant difference in relative enrichment between the conditions.

[0313] Alternative hypothesis (H1): There is a significant difference in relative enrichment between the conditions.

[0314] To ensure the reliability of the t-test, certain conditions must be met. First, observations within each group should be independent of one another. In our experiments, we used multiple copies for each condition, which helps ensure the independence of the observations.

[0315] Next, the data should be normally distributed within each group. We checked the normality of the data using histograms and normal probability plots and found that the data is approximately normally distributed.

[0316] The inventors also used the F-test to check for equal variances and found that the variances of the two groups were not significantly different. This is important because the t-test assumes equal variances, and failure to satisfy this assumption can affect the reliability of the test.

[0317] Finally, the inventors checked the sample sizes of the two groups and found that they were equal. This is important because the t-test assumes equal sample sizes, and failure to satisfy this assumption can affect the reliability of the test.

[0318] Overall, the inventors believe that the conditions for the t-test were met in their experiment, and that the test results are reliable for accurately comparing the relative enrichment of the two conditions.

[0319] Based on the results of the t-test, the inventors can conclude that the null hypothesis should be rejected. The p-value for the test is less than 0.001, which indicates that the difference in relative enrichment between the two experiments is statistically significant. This suggests that there is a true difference between the two groups, rather than mere random variation.

[0320] The inventors compare the relative enrichment of different constructs using the geometric mean with a negative control as a reference for normalization.

[0321] To calculate the IC50 curve and values, the raw data was normalized using minimum-maximum normalization (including positive and negative controls as maximum and minimum values, Kevorkov and Makarenkov, 2005).

[0322] Results and analysis: Nter and Cter fusion tags are useful tools for studying protein-RNA interactions, but their specific localization and interaction with RNA can influence the results of these studies. We first investigated the effect of GFP fusion protein localization on mRNA enrichment on YB1 (Figure 10).

[0323] Our results showed that YB1 and RNA can still interact regardless of the localization of the GFP tag (Nter or Cter) (Figure 11).

[0324] However, enrichment differs significantly, with Nter GFP-tagged constructs exhibiting lower enrichment. The mechanism of YB1: The RNA interactions explaining this difference still need to be investigated.

[0325] Next, the inventors investigated the effect of deleting the cold-shock domain (CSD) of YB1 (SEQ ID NO: 23), a conserved RNA-binding domain that has been found to be drug-potential (El Hage et al., under review) (Figure 12).

[0326] Our results show that the deletion of the RNA-binding domain results in a significant decrease in RNA binding and enrichment for both Nter and Cter GFP-tagged proteins. These results indicate that the presence of the RNA-binding domain is necessary for significant enrichment to occur.

[0327] Similar to the PPI system, the inventors attempted to investigate whether perturbations by small molecules could be quantified using the MT bench. Preliminary results in NMR and cells showed that the C8 compound reduced RNA:YB1 interaction. Using the MT bench assay, the inventors determined the IC50 of this molecule in HeLa cells at 19 different concentrations: 31.3 μM (Figure 13).

[0328] These combined results demonstrate that the MT bench technique is sensitive enough to detect variations in protein / RNA interactions. Combined with modifications / recombinations / mutations of proteins expressed in cells (via gateway vectors), this technique will enable the characterization and quantification of protein / RNA interactions within cells.

[0329] [Table 2-1]

[0330] Table 2-2

[0331] Table 2-3

Claims

1. An in vitro method for evaluating the ability of a compound to disrupt or stabilize the interaction between one or more baits and one or more candidate plays in eukaryotic cells, a. (i) providing one or more baits comprising a bait portion and a polymerized tubulin binding portion, and (ii) providing eukaryotic cells expressing one or more candidate baits, b. In the eukaryotic cell, the steps of determining the occurrence of an interaction between the one or more baits, which are bound to polymerized tubulin in the eukaryotic cell, and the one or more candidate plays, thereby localizing the one or more candidate plays along the polymerized tubulin, and thereby detecting the interaction, c. The step of bringing the eukaryotic cells into contact with the compound, d. A step of determining the occurrence of the interaction between the one or more baits and the one or more candidate plays in the eukaryotic cell as in step b, and quantifying this interaction, e. The comparison step of the occurrence of the interaction performed in step b and step d, f. An in vitro method comprising the step of concluding whether the effect of the compound is the disruption or stabilization of the interaction between the one or more baits and the one or more candidate plays.

2. The aforementioned bait, - A polymerized tubulin-binding moiety containing one or more microtubule-binding domains (MBDs), - The in vitro method according to claim 1, comprising a bait portion.

3. The in vitro method according to claim 1 or 2, wherein the bait includes a linker (L) region located between the polymerization tubulin-binding portion and the bait portion.

4. The in vitro method according to any one of claims 1 to 3, wherein the candidate play includes a fluorescent protein.

5. The in vitro method according to any one of claims 1 to 4, wherein determining the occurrence of the prey in the eukaryotic cells in step b and step d is a detection method selected from the group including binding to an antibody, hybridization with nucleic acids, and fluorescence measurement.

6. The in vitro method according to any one of claims 1 to 5, wherein the cells in step b are fixed cells or living cells.

7. The in vitro method according to any one of claims 1 to 6, wherein the method is repeated using different concentrations of the compound to determine the IC50 of the compound for tested combinations of one or more baits and one or more candidate plays.

8. The in vitro method according to any one of claims 1 to 7, wherein steps a and b are repeated before step c using different vector constructs of the one or more baits and the one or more candidate plays to determine the occurrence of the interaction between the one or more baits and the one or more candidate plays in the eukaryotic cells and to select the correct combination which will be evaluated in subsequent steps so as to be closer to what occurs in vivo.

9. The in vitro method according to any one of claims 1 to 8, wherein one or more candidate plays and / or one or more baits are selected from ribonucleic acid and / or deoxyribonucleic acid and / or proteins.

10. The in vitro method according to any one of claims 1 to 9, wherein the one or more baits are proteins.

11. The in vitro method according to any one of claims 1 to 10, wherein one or more candidate plays are ribonucleic acid.

12. The in vitro method according to any one of claims 1 to 10, wherein one or more candidate plays are proteins.

13. The in vitro method according to any one of claims 1 to 12, wherein the one or more baits and the one or more candidate plays are characteristic of proteinosis or cancer.

14. A compound identified by the in vitro method described in any one of claims 1 to 13, for use as a pharmaceutical.

15. Use of a polymer tubulin-binding moiety as a tool for evaluating the ability of a compound to disrupt or stabilize the interaction between one or more baits and one or more play in a eukaryotic cell, wherein the bait is bound to polymer tubulin in the eukaryotic cell, thereby localizing the one or more candidate play along the polymer tubulin, and the one or more baits and the one or more candidate play are brought into contact with the compound, thereby enabling the evaluation of the disruption or stabilization of the interaction.

Images