WO2010040842A1 - Linking protein aggregation and yeast survival - Google Patents

Abstract

The invention relates to methods for the identification of compounds capable of preventing the aggregation of aggregation-prone polypeptides based on the use of fusion proteins which comprise an aggregation-prone polypeptide and an enzyme which is able to metabolize a compound showing toxicity of the cell whereby if the compound prevents aggregation of the aggregation-prone polypeptide, the enzyme will remain soluble and in active form thus preventing the toxic effect of the compound on the cell and increasing cell viability. The invention also provides polypeptides comprising an aggregation-prone polypeptide and an enzyme and uses thereof.

Description

LINKING PROTEIN AGGREGATION AND YEAST SURVIVAL

BACKGROUND OF THE INVENTION

In the last few years, protein aggregation has emerged from a neglected area of protein chemistry as a transcendental issue in biological and medical sciences. In this regard, an increasing body of evidence points out at the anomalous misassembly of proteins into insoluble amyloid deposits as the fundamental cause behind some debilitating human disorders of growing incidence such as Alzheimer's disease (AD), Parkinson' s disease (PD), type II diabetes, the transmissible spongiform encephalopathies and many others (Dobson, CM. Protein-misfolding diseases: Getting out of shape. Nature 418, 729 - 730 (2002)). A common trait of these disorders is that the aggregated protein deposits in internal organs and interferes with normal cellular function, sometimes lethally. On the other hand, protein aggregation in cell factories represents a major bottleneck in recombinant protein production, narrowing the spectrum of polypeptides obtained by recombinant techniques and hampering the development of priority research areas such as structural genomics and proteomics. Therefore, there is an increasing interest in the development of protein solubility screening methods that allow foreseeing genes, chemical compounds or culture conditions that would modulate protein aggregation.

To date, several protein solubility assays have been developed in order to study and detect aggregation in prokaryotic cells, specifically in Escherichia coli (Maxwell, K.L., Mittermaier, A. K., Forman-Kay, J.D. & Davidson, A.R. A simple in vivo assay for increased protein solubility, Protein Sci 8, 1908-1911 (1999)), (Waldo, G. S., Standish, B. M., Berendzen, J. & Terwilliger, T. C. Rapid protein- folding assay using green fluorescent protein. Nat Biotechnol 17, 691-695 (1999)), (Wigley, W.C., Stidham, R.D., Smith, N. M., Hunt, J. F. & Thomas, P.J. Protein solubility and folding monitored in vivo by structural complementation of a genetic marker protein. Nat Biotechnol 19, 131-136 (2001)). The majority of these systems rely on the ability of aggregating proteins to co-trans lationally induce improper folding of a fused marker protein. Thus, different reporter proteins have been used allowing different readouts as fluorescence, chloramphenicol resistance or β-galactosidase activity. Another approach has been the use of fluorescence based methods. One of the first developed methods used the GFP as a reporter of concomitant target protein aggregation (Waldo, G.S., Standish, B. M., Berendzen, J. & Terwilliger, T. C. Rapid protein- folding assay using green fluorescent protein. Nat Biotechnol 17, 691-695 (1999)). It is based on the finding that the fluorescence in GFP fusions correlates with the tendency of the N-terminal fusion partner to form insoluble aggregates. Fusions with non-aggregating partners allow proper folding of the GFP reporter. Therefore, colonies expressing such fusions exhibit green fluorescence. On the contrary, GFP fused to an aggregating-prone protein do not fold properly and the fluorescence emission decreases. As a test case, amyloid peptide Aβ42 was fused to GFP and expressed in bacteria. After performing random mutagenesis over this peptide, mutations that reduce Aβ42 tendency to aggregate resulted in increased fluorescence of the bacteria expressing such fusions.

Fluorescence resonance energy transfer (FRET) has also been used to establish a novel in vivo screening system that allows rapid detection of protein folding and protein variants with increased thermodynamic stability in the E.coli cytoplasm (Philipps, B., Hennecke, J. & Glockshuber, R. FRET-based in vivo screening for protein folding and increased protein stability. JMo/ Biol 327, 239-249 (2003)).

The system is based on the simultaneous fusion of the green fluorescent protein (GFP) to the C terminus of a protein X of interest, and of blue fluorescent protein (BFP) to the N terminus. Efficient FRET from BFP to GFP in the ternary fusion protein is observed in vivo only when protein X is folded and brings BFP and GFP into close proximity; whereas FRET is lost when BFP and GFP are far apart due to unfolding or intracellular degradation of protein X.

Other approaches have exploited the intracellular activity of β-galactosidase to report on protein solubility or aggregation. One method is an adaptation of the classical β-galactosidase protein complementation assays (Wigley, W.C., Stidham, R.D., Smith,

N. M., Hunt, J.F. & Thomas, P.J. Protein solubility and folding monitored in vivo by structural complementation of a genetic marker protein. Nat Biotechnol 19, 131-136 (2001)). Each monomer of the homotetrameric enzyme can be divided into two fragments, the small α-fragment and the larger co-fragment. In the presence of α- fragment, dimers of co-fragments achieve a dynamic equilibrium to form a tetramer with enzymatic activity. If the target protein is fused to the α-fragment, redistribution of the α-fragment into the insoluble cellular fraction will lead to a reduction in the level of β- gal activity, reporting on the solubility of the target.

Another assay has been developed as a result of the identification of specific genes responding to protein misfolding. In this method, one of these genes promoters (IbpAB) is fused to β-galactosidase in order to quantify the response of the promoter to intracellular misfolding and aggregation (Lesley, S. A., Graziano, J., Cho, CY. , Knuth, M. W. & Klock, H. E. Gene expression response to misfolded protein as a screen for soluble recombinant protein. Protein Eng 15, 153-160 (2002)). In this way, β-gal expression (and activity) becomes linked to protein aggregation inside the cell. Recently, it has been demonstrated the utility of this approach to evaluate different factors that modulate solubility during recombinant expression in E.coli (Schultz, T., Martinez, L. & de Marco, A. The evaluation of the factors that cause aggregation during recombinant expression in E. coli is simplified by the employment of an aggregation- sensitive reporter. Microb Cell Fact 5, 28 (2006)).

Chloramphenicol resistance has also been used as a readout to detect soluble mutants of an aggregating-prone protein in vivo in E. coli (Maxwell, K.L., Mittermaier, A.K., Forman-Kay, J.D. & Davidson, A.R. A simple in vivo assay for increased protein solubility. Protein Sci 8, 1908-1911 (1999)). In this case, the reporter protein is chloramphenicol acetyltransferase (CAT). Resistance to high levels of chloramphenicol will be equivalent to the expression of soluble mutant fusions of the target protein. The selection can be carried out growing the cells in plates with high concentration of the antibiotic.

Methods based on the twin-arginine translocation pathway have also been used for detection of protein misfolding. The basis for these assays is the protein dependence on correct folding in order to be transported through the bacterial twin-arginine translocation (Tat) pathway (Fisher, A.C., Kim, W. & DeLisa, M. P. Genetic selection for protein solubility enabled by the folding quality control feature of the twin-arginine translocation pathway. Protein Sci 15, 449-458 (2006)). In this system, a target protein is expressed as a tripartite fusion between a N-terminal Tat signal peptide and a C- terminal TEMl β-lactamase reporter protein (BIa). If the protein folds correctly, it will be translocated through the TAT pathway to the periplasm. Due to the fact that the target protein is also fused to β-lactamase, it will confer ampicillin resistance to the bacteria. Then, survival of E. coli cells expressing a Tat-targeted test protein/β- lactamase fusion on selective medium correlates with the solubility of the protein of interest. Using this assay, variants of the Alzheimer's Aβ42 peptide with an enhanced solubility could be detected and isolated from a large combinatorial library.

All these methods are based on bacterial cells, therefore presenting the drawbacks associated to heterologous protein expression in bacteriae.

WO2007/103788 describes a method for determining protein aggregation in yeast. It is based on the capacity of the translational termination factor Sup35p to form self- propagating infectious amyloid aggregates. This factor manifests a prion phenotype referred to as [PSI+] and it is composed of three domains. The N-terminal domain (N) is dispensable for viability, and it is required and sufficient for the prion properties of Sup35p. While the function of the highly charged middle (M) domain remains unclear, the C-terminal RF (release factor) domain performs termination of protein translation and is essential for viability

In this assay, the activity of the termination factor Sup35p (NMRF) is conveniently assayed in vivo by examining the efficiency with which protein synthesis terminates at a premature stop codon (a nonsense-suppression assay). The assay uses the adel-14 nonsense allele. Strains carrying this mutation and bearing fully active NMRF produce only a truncated (inactive) version of Adelp, and as a result cannot grow on synthetic medium lacking adenine (-Ade), while they grow normally on synthetic medium supplemented with adenine (+Ade). In addition, these cells accumulate a red intermediate of the adenine synthesis pathway when grown on complex medium. However, if the efficiency of translational termination at the premature stop codon of the adel-14 allele is compromised, the cells gain the ability to grow on -Ade (i.e. they become Ade+) and do not accumulate red pigment. For example, cells expressing the complete Sup35p containing all three domains are white and Ade+ when NMRF is in the aggregated [PSI+] prion form. Cells expressing an aggregation-deficient and therefore fully functional form of Sup35p lacking the non- essential N-terminal domain (MRF) are red and Ade-. Thus, this well established system reliably distinguishes between fully active monomer, and malfunctioning aggregated forms of NMRF. It can be applied to study the aggregation of a target protein like Aβ₄₂. If this aggregation-prone peptide is fused to MRF (AβMRF), the yeast cells become white and grow on -Ade (they will be Ade+), whereas mutations that promote increased solubility of Aβ₄₂ render dark pink, Ade- cells.

This method, however, provides a negative assay for determining protein folding, since the yeast which do not aggregate the protein of interest (due, for example, to the presence of an inhibitor of aggregation) are not viable.

Thus, there is a need for a system where one can identify protein folding and therapeutic agents for diseases associated with protein oligomerization which may have their therapeutic effect due to being either regulators of protein folding, and/or inhibitors of protein aggregation in a positive assay, wherein the viable cells are those which do not have protein aggregation and therefore are easier to analyze.

SUMMARY OF THE INVENTION

The invention is related to a method that could couple an easily measurable phenotype like cell survival to protein aggregation using yeast as a model of eukaryotic organism. It is based in the fusion of the target protein to the human dihydrofolate reductase (h-DHFR). DHFR is a key enzyme in thymidine synthesis that catalyses the reduction of 7,8-dihydrofolate to 5,6,7,8-tetrahydrofolate with NADPH as a coenzyme and in the three-hybrid method.

Prokaryotic and eukaryotic DHFRs are central to cellular one-carbon metabolism and are absolutely required for cell survival. And its activity can be specifically inhibited by the drug methotrexate (MTX). In the present work, we sought to exploit our observation that yeast cells can become insensitive to MTX if they express high levels of h-DHFR. Other specific inhibitors of the DHFR could be used (such as, for example, trimethoprim in case DHFR of bacterial origin was used).

This enzyme is a very soluble protein and, in our approach, it is expressed at concentrations that allow cell survival in MTX concentrations that otherwise would be lethal. Therefore, the invention is based on the surprising effect that the fusion of h- DHFR to aggregation-prone polypeptides might inactivate the enzyme and render the cells expressing these kinds of fusions MTX susceptible.

In principle, this might allow for the design of an easy high-throughput method to monitor protein folding inside yeast based on the reversal of MTX growth inhibition. And it might permit to monitor the effect of different factors (mutations, genetic backgrounds, chemical compounds or growth conditions) over protein aggregation. To demonstrate the applicability of the assay, different aggregation prone proteins (Alzheimer's amyloid β (Aβ) peptide, polyglutamine expansions in the huntingtin protein (poliQ) and alpha-synuclein (α-Syn) will be used as models.

The approach discussed here aims at the easy and reliable evaluation of the effects of intrinsic and extrinsic factors on protein aggregation. And it is based on the correspondence between the intracellular activity and solubility of recombinant h- DHFR and cell growth in the presence of lethal concentrations of MTX. Furthermore, the use of fMTX (a MTX labeled with a fluorescent compound) enables to monitor simultaneously the cell viability and the localization of the aggregates inside the cell.

Overall, the method is able to anticipate the intracellular aggregation propensity of genetic variants of three unrelated polypeptides linked to important human disorders.

The system could become also a convenient platform for chemical screening of agents that interfere with protein aggregation in order to assist in the development of new therapeutic leader compounds targeting protein aggregation and toxicity. The use of S. cerevisiae is compatible with these applications due to the availability of drug- permeable strains (i.e. ergόA), although any other yeast cell capable of expressing the gens of interest could be used.

Thus, in a first aspect, the invention relates to a method for the identification of compounds that are capable of decreasing aggregation of an aggregation-prone polypeptide comprising:

(i) contacting one or more yeast cells with a candidate compound, wherein the yeast cells express a fusion protein which comprises an aggregation-prone polypeptide and an enzyme, wherein the enzyme is capable of modifying a compound which adversely affects yeast cell viability into a metabolite with a reduced adverse effect on said yeast cell viability,

(ii) adding the toxic compound to the yeast cell of step (i) in an amount which, without the presence of the activity of the enzyme forming part of the fusion protein used in step (i), would adversely affect the yeast cell viability and

(iii) determining the viability of the yeast cells wherein an increased viability of the cells with respect of the cells which have not been exposed to the candidate compound is indicative that the compound is capable of decreasing aggregation of the aggregation-prone polypeptide.

In another aspect, the invention relates to a polypeptide comprising an aggregation- prone polypeptide and a polypeptide having enzymatic activity wherein said polypeptide having enzymatic activity is capable of modifying a compound which adversely affects yeast cell viability into a metabolite of said compound with a reduced adverse effect on said yeast cell viability.

In further aspects, the invention relates to a polynucleotide encoding a polypeptide of the invention, a vector comprising a polynucleotide of the invention, and a host cell comprising a polypeptide, a polynucleotide or a vector of the invention. In another aspect, the invention relates to the use of a host cell of the invention for the identification of compounds which are capable of inhibiting aggregation of an aggregation-prone polypeptide.

In one aspect, the invention is related to a method of screening for a compound that decreases aggregation of aggregation-prone polypetides wherein the method comprises (a) contacting one or more yeast cells with a candidate compound, wherein the yeast cells express a fusion protein comprising an aggregation-prone polypeptide, such as a amyloidogenic protein, and an enzyme which inhibits a toxic compound which affects yeast cell viability, or which prevents the toxic compound from acting and affecting the cell viability (b) adding the toxic compound to the yeast cell in an amount which, without the presence of the enzyme, would affect cell viability.

Using this method, if a compound which decreases aggregation of amyloidogenic peptides is added to the yeast cell, the enzyme will have activity and will inhibit the toxic compound, therefore the cell will survive.

Different yeast cells could be used, for example Saccharomyces uvae, Saccharomyces kluyveri, Schizosaccharomyces pombe, Saccharomyces uvarum, Kluyveromyces lactis, Hansenula polymorpha, Pichia pastor is, Pichia methanolica, Pichia kluyveri, Yarrowia lipolytica, Candida sp., Candida utilis, Candida cacaoi, Geotrichum sp., Geotrichum fermentans, and Saccharomyces cerevisiae. Preferred is Saccharomyces cerevisiae. More preferred is Saccharomyces cerevisiae FY384, since the lethality of Methotrexate (MTX) on S . cerevisiae FY384 cells above certain concentrations (1 mM of sulfonamide, which is used to promote MTX cell intake) can be overcome by heterologous expression of human DHFR under the control of Gall 10 promoter.

Different degrees of sensitivity to MTX may thus be correlated with the intracellular activity of the enzyme. In case the human DHFR is expressed as a fusion with aggregation prone polypeptides, the fused protein would promote, at least partially, its deposition lowering the intracellular activity and increasing sensitivity to MTX. Therefore, aggregation state of the fused protein is linked directly to yeast cell survival in the presence of methotrexate. The invention is based in the surprising effect that the fusion of an h-DHFR to an amyloidogenic protein, inactivate the enzyme and render these cells expressing these kinds of fusion susceptible to MTX.

Therefore, the fusion of an enzyme to an aggregation-prone polypeptide, inactivate the enzyme and render these cells expressing these kinds of fusion susceptible to a toxic compound which would be inhibited or otherwise inactivated by the enzyme, as it is MTX with h-DHFR.

Conveniently, the aggregation-prone peptides are amyloidogenic peptides, such as Aβ42, PoIyQ expansions, or α-synuclein variants.

The method of the invention can be used, in a more preferred way, to test compounds related to the prevention of treatment of a disease selected from the group consisting of Alzheimer's disease, Parkinson's disease, Familial Amyloid Polyneuropathy, a Tauopathy, Trinucliotide disease, transmissible spongiform encephalopathies (TSEs), Alzheimer's (AD), and Huntingdon's Disease (HD), which are known to be caused by the aggregation of amyloidogenic peptides.

In the case a fluorescent reported is used, apart from cell viability the method could be measured also by fluorescence, allowing a second control regarding inhibiting properties of the tested compound. The person skilled in the art would recognize several different fluorescent markers suitable for the invention. Preferred is AlexA.

In another aspect of the invention, the invention is related to a kit useful for screening compounds that inhibit protein aggregation which uses the method of the invention.

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 shows a scheme of the plasmid used in this assay. The target protein fused to DHFR was cloned between the restriction sites CIaI i BgIII in the plasmid pESC. Figure 2 shows A) Visualization of intracellular wild-type and Aβ42(F19D) distribution with GFP expressed in S. cerevisiae B) Co-staining of the cell nucleus with Hoechst (blue). The aggregated Aβ42-GFP has a juxtanuclear position.

Figure 3 shows cell viability (spotting) assays for yeast expressing DHFR, peptide Aβ42-DHFR or peptide Aβ42 F19D-DHFR at different temperatures and MTX concentrations. Four- fold serial dilutions starting with equal number of cells are shown.

Figure 4 shows growth assays of FY834 yeast cells expressing DHFR (empty circles), peptide Aβ42-DHFR (empty squares) or peptide Aβ42(Fl 9D)-DHFR (solid circles) in the presence of 0 μM (left) and 20 μM of MTX (right).

Figure 5 shows filter trap assay of cells expressing DHFR, Aβ42(Fl 9D)-DHFR or

Aβ42-DHFR. Protein aggregates were detected by immunoblot analysis using specific antibody against DHFR.

Figure 6 shows the addition of a fluorescent inhibitor of DHFR (fMTX) enables the visualization of the intracellular distribution of wild-type and F19D mutant Aβ42 fused to DHFR.

Figure 7 shows A) Fluorescence microscopy of yeast cells expressing different polyQ expansions (Q25, Q72 or Q103) fused to GFP B) Growth assays of yeast cells expressing the different polyQ expansions fused to DHFR in the presence of 20 μM MTX.

Figure 8 shows the addition of a fluorescent inhibitor of DHFR (fMTX) enables the visualization of the intracellular distribution of the different poliQ expansions fused to DHFR.

Figure 9 shows A) Fluorescence microscopy of yeast cells expressing different α- synuclein variants fused to GFP. B) Growth assays of yeast cells expressing α-synuclein variants fused to DHFR in the presence of 100 μM MTX. Figure 10 shows the addition of a fluorescent inhibitor of DHFR (fMTX) and how it enables the visualization of the intracellular distribution of the different α-synuclein variants fused to DHFR.

Figure 11 shows A) Growth restoration of ergo A yeast cells expressing in Aβ42-DHFR in the presence of 20 μM MTX, ImM sulfanilamide and selected concentrations of quercetin (30 μM and 100 μM) and CR (10 μM). Growth is normalized to 0 μM compound. Significant differences are marked with an asterisk. B) Fluorescence microscopic assessment of Aβ42-GFP aggregation in control or compound treated ergo A cells.

Figure 12 shows A) Growth of yeast FY834 strains overexpressing a chaperone and co- expressing peptide Aβ42-DHFR in the presence of MTX in liquid media. Growth is normalized to the same strain expressing the corresponding chaperone and DHFR. Significant differences are marked with an asterisk. B) Cell viability (spotting) assays for the different strains overexpressing a chaperone and DHFR or Aβ42-DHFR. In each case, four- fold serial dilutions starting with equal number of cells are shown.

Figure 13 shows A) Fluorescence microscopy of different strains overexpressing a chaperone and peptide Aβ42 fused to GFP. B) Western Blot analysis of cells co- expressing different chaperones and Aβ42-DHFR. The concentration of Aβ42-DHFR did not differ due to the expression of one chaperone. The different bands were correlated to the different oligomerization degrees of Aβ42-DHFR.

Figure 14 shows Growth of yeast FY834 strains with a deletion in one chaperone and expressing peptide Aβ42-DHFR in the presence of MTX in liquid media. Growth is normalized to the same strain expressing the corresponding chaperone and DHFR. Significant differences are marked by an asterisk.

Figure 15 shows A) Fluorescence microscopy of different strains with chaperone knockouts and expressing peptide Aβ42 fused to GFP. B) Western Blot analysis of different yeast strains with a deletion in a chaperone and cells expressing Aβ42-DHFR. The detected bands correlated with the different oligomerization degrees of Aβ42- DHFR.

DETAILED DESCRIPTION OF THE INVENTION

An objective of the invention is to develop a method to detect protein aggregation using simultaneously yeast survival and fluorescence emission as reporter signals. As a test models different aggregation-prone proteins involved in neurodegenerative diseases will be used: Alzheimer's amyloid β (Aβ) peptide, polyglutamine expansions in the huntingtin protein (poliQ) and alpha-synuclein (α-Syn).

A second objective is to study the ability of the method to detect the effect of different factors that modulate protein aggregation in vivo, such as chemical compounds, overexpression of chaperones, deletion of chaperones and growth conditions (i.e. temperature).

In a first aspect, the invention relates to a method for the identification of compounds that are capable of decreasing aggregation of an aggregation-prone polypeptide comprising:

(i) contacting one or more yeast cells with a candidate compound, wherein the yeast cells express a fusion protein which comprises an aggregation-prone polypeptide and an enzyme, wherein the enzyme is capable of modifying a compound which adversely affects yeast cell viability into a metabolite with a reduced adverse effect on said yeast cell viability,

(ii) adding the toxic compound to the yeast cell of step (i) in an amount which, without the presence of the activity of the enzyme forming part of the fusion protein used in step (i), would adversely affect the yeast cell viability and (iii) determining the viability of the yeast cells wherein an increased viability of the cells with respect of the cells which have not been exposed to the candidate compound is indicative that the compound is capable of decreasing aggregation of the aggregation-prone polypeptide.

The term "aggregation-prone polypeptide" refers to a polypeptide which is able to adopt a beta-pleated sheet conformation and/or to form oligomers, fibrils and plaques. For example, peptides having a potential for self- assembling and fϊbrillogenesis are fibrillaric proteins derived from at least one of the following precursor proteins: Tau, alpha-synuclein, huntingtin, ataxin, superoxide dismutase, TDP-43, SAA (Serum- Amyloid-Protein A), AL (k or Might chains of Immunoglobulins), AH (gl Ig-heavy chains), ATTR (Transthyretin, Serum-Prealbumin), AApo-A-1 (Apolipoprotein Al), AApoA2 (Apolipoprotein A2), AGeI (Gelsolin), ACys (Cystatin C), ALys (Lysozyme), AFib (Fibrinogen), Beta-amyloid (Amyloid precursor protein), Beta-amyloid2M (beta2 -microglobulin), APrP (Prion protein), ACaI (Pro calcitonin), AIAPP (islet amyloid polypeptide); APro (Prolactin), AIns (Insulin); AMed (Lactadherin); Aker (Kerato- epithelin); ALac (Lactoferrin), Abri (AbriPP), ADan (ADanPP); or AANP (Atrial natriuretical peptide); for review see, e.g., Skovronsky at al, Annu. Rev. Pathol. Mech. Dis. 1 (2006), 151-70 and Buxbaum, Curr. Opin. Rheumatol. 16 (2003), 67-75). In a preferred embodiment, the aggregation-prone polypeotide is selected from the group of mutant-huntingtin, beta-amyloid, tau, alpha-synuclein, mutant androgen receptor, mutant SODI, mutant ataxin and the like. In a still more preferred embodiment, the aggregation-prone polypeptide is an amyloidogenic peptide. In a still more preferred embodiment, the aggregation-prone polypeptide is a polypeptide selected from the group of Aβ42, a peptide comprising a Poly-glutamine expansions and α-synuclein or a variant thereof.

The term "yeast cells" , as used herein, includes ascosporogenous yeast (Endomycetales), basidiosporogenous yeast, and yeast belonging to the Fungi lmperfecti (Blastomycetes). Since the classification of yeast may change in the future, for the purposes of this invention, yeast shall be defined as described in Biology and Activities of Yeast (Skinner, F. A., Passmore, S. M., and Davenport, R. R., eds, Soc. App. Bacteriol. Symposium Series No. 9, 1980). In a preferred aspect, the yeast host cell is a Candida, Hansenula, Kluyveromyces, Pichia, Saccharomyces, Schizosaccharomyces, or Yarrowia cell. In a most preferred aspect, the yeast host cell is selected from the group of Saccharomyces cererevisiae, Saccharomyces uvae, Saccharomyces kluyveri, Schizosaccharomyces pombe, Saccharomyces uvarum, Kluyveromyces lactis, Hansenula polymorpha, Pichia pastoris, Pichia methanolica, Pichia kluyveri, Yarrowia lipolytica, Candida sp., Candida utilis, Candida cacaoi, Geotrichum sp. and Geotrichum fermentans

The term "contacting a cell with a candidate compound", as used herein, includes any possible way of taking the candidate compound inside the cell expressing the fusion protein. Thus, in the event that the candidate compound is a molecule with low molecular weight, it is enough to add said molecule to the culture medium. In the event that the candidate compound is a molecule with a high molecular weight (for example, biological polymers such as a nucleic acid or a protein), it is necessary to provide the means so that this molecule can access the cell interior. In the event that the candidate molecule is a nucleic acid, conventional trans fection means can be used, as described previously for the introduction of the the polynucleotide. In the event that the candidate compound is a protein, the cell can be put in contact with the protein directly or with the nucleic acid encoding it coupled to elements allowing its transcription / translation once they are in the cell interior. To that end, any of the aforementioned methods can be used to allow its entrance in the cell interior. Alternatively, it is possible to put the cell in contact with a variant of the protein to be studied which has been modified with a peptide which can promote the translocation of the protein to the cell interior, such as the Tat peptide derived from the HIV-I TAT protein, the third helix of the Antennapedia homeodomain protein from D.melanogaster, the VP22 protein of the herpes simplex virus and arginine oligomers (Lindgren, A. et al, 2000, Trends Pharmacol. Sci, 21 :99-103, Schwarze, S.R. et al., 2000, Trends Pharmacol. Sd., 21 :45- 48, Lundberg, M et al., 2003, MoI. Therapy 8:143-150 and Snyder, EX. and Dowdy, S.F., 2004, Pharm. Res. 21 :389-393).

The compound to be assayed is preferably not isolated but forms part of a more or less complex mixture derived from a natural source or forming part of a library of compounds. Examples of libraries of compounds which can be assayed according to the method of the present invention include, but are not limited to, libraries of peptides including both peptides and peptide analogs comprising D-amino acids or peptides comprising non-peptide bonds, libraries of nucleic acids including nucleic acids with phosphothioate type non-phosphodiester bonds or peptide nucleic acids, libraries of antibodies, of carbohydrates, of compounds with a low molecular weight, preferably organic molecules, of peptide mimetics and the like. In the event that a library of organic compounds with a low molecular weight is used, the library can have been preselected so that it contains compounds which can access the cell interior more easily. The compounds can thus be selected based on certain parameters such as size, lipophilicity, hydrophilicity, capacity to form hydrogen bonds.

The compounds to be assayed can alternatively form part of an extract obtained from a natural source. The natural source can be an animal, plant source obtained from any environment, including but not limited to extracts of land, air, marine organisms and the like.

The term "fusion protein" or "chimeric protein", as used herein, comprises a polypeptide of the invention operatively linked to another polypeptide. Within the fusion protein, the term "operatively linked" is intended to indicate that the polypeptide(s) according to the invention and the other polypeptide(s) are fused in- frame to each other.

The term "yeast cell viability", as used herein, refers to the ability of cells in culture to survive under a given set of culture conditions or experimental variations.

As used herein, an compound which adversely affects yeast cell viability or toxic compound is any compound whose presence in the host cell will prevent the host cell in culture from achieving the normal logarithmic growth it would have achieved but for the expression of the compound. IN a preferred embodiment, the enzyme which adversely affects yeast cell viability is human dehydrofolate reductase (h-DHFR) and the toxic compound is methotrexate (MTX).

In a preferred embodiment, the method according to the invention is carried out using a toxic compound whcih is fluorescently labelled. Suitable fluoresenct compounds that can be used to label teh toxic compound include, without limitation, FAM™, TET™, JOE™, VIC™, SYBR(R) Green; 6 FAM, HEX, TET, TAMRA, JOE, ROX, Fluorescein, Cy3, Cy5, Cy55, Texas Red, Rhodamine, Rhodamine Green, Rhodamine Red, 6-CarboxyRhodamine 6G, Oregon Green 488, Alexa Flour, Oregon Green 500 or Oregon Green 514. In a preferred embodiment, the fluorescently labelled toxic compound is fluoresencently labelled MTX and, in particular, fMTX.

In those cases wherein the method is carried out using a fluorescently labelled toxic compound, the method further comprises detecting the fluorescence in the yeast cell wherein an increased intracellular fluorescence is indicative that the compound is capable of decreasing aggregation of the aggregation-prone polypeptide. Suitable method for detecting the fluorescence in the yeast cell includes, without limitation, FACS, immunofluorescence, immunohistochemistry and the like.

In another preferred embodiment, the method is carried out using yeast strains showing an increased membrane permeability. Strains having an increased cell permeability are widely known to the skilled person and can be identified using standard technology. In a preferred embodiment, the yeast strain carries an inactivating mutation in the ergo gene.

In a preferred embodiment, the method of the invention is carried out in a cell which carries an inactivating mutation in one or more molecular chaperones. Molecular chaperones, as used herein, refers to any of a group of proteins that are involved in the correct intracellular folding and assembly of polypeptides without being components of the final structure. Herein, "molecular chaperones" and "chaperones" are used interchangeably. In a preferred embodiment, the molecular chaperone is selected from the group of a member of the HsplOO protein family, a member of the Hsp90 protein family, a member of the Hsp70 protein family, a member of the Hsp40 protein family or a small heat shock protein. In a still more preferred embodiment, the member of the HsplOO protein family is HsplO4, the member of the Hsp90 protein family is selected from the group of Hsc82 and Hsp82, the member of the Hsp70 protein family is selected from the group of Ssal , Ssa2, Ssa3 and Ssa4, the member of the Hsp40 protein family is selected from the group of Ydjl and Sisl and/or the small heat shock protein is selected from the group of Hsp26 and Hsp42.

The invention also relates to polypeptides comprising an aggregation-prone polypeptide and a polypeptide having enzymatic activity wherein said polypeptide having enzymatic activity is capable of modifying a compound which adversely affects yeast cell viability into a metabolite of said compound with a reduced adverse effect on said yeast cell viability.

Suitable aggregation-prone polypeptides, polypeptide having enzymatic activity have been described above in detail in relation to the method of the invention. In a preferred embodiment, the polypeptide of the invention further comprises a reporter polypeptide. As used herein "reporter polypeptide" refers to a polypeptide gene product, which, can be quantitated either directly or indirectly. Suitable reporter genes include, without limitation, a beta-galactosidase (lacZ), beta-glucuronidase (GUS), luciferase, alkaline phosphatase, nopaline synthase (NOS), chloramphenicol acetyltransferase (CAT), horseradish peroxidase (HRP). In a preferred embodiment, the reporter polypeptide is a fluorescent protein. As used herein, the term "fluorescent protein" as used herein is a protein that has intrinsic fluorescence when excited with electromagnetic radiation at the appropriate wave length. Representative fluorescent proteins can include, but are not limited to, sgGFP, sgBFP, BFP blue-shifted GFP (Y66H), Blue Fluorescent Protein, CFP~Cyan Fluorescent Protein, Cyan GFP, DsRed, monomeric RFP, EBFP, ECFP, EGFP, GFP (S65T), GFP red shifted (rsGFP), GFP wild type, non-UV excitation (wtGFP), GFP wild type, UV excitation (wtGFP), GFPuv, HcRed, rsGFP, Sapphire GFP, sgBFP.TM., sgBFP.TM. (super glow BFP), sgGFP.TM., sgGFP.TM. (super glow GFP), wt GFP, Yellow GFP and YFP. In a preferred embodiment, the fluorescent protein is GFP.

In yet another aspect, the invention relates to a polynucleotide encoding a polypeptide of the invention.

The term "polynucleotide(s)," as used herein, means a single or doublestranded polymer of deoxyribonucleotide or ribonucleotide bases and includes DNA and corresponding RNA molecules, including HnRNA and mRNA molecules, both sense and anti-sense strands, and comprehends cDNA, genomic DNA and recombinant DNA, as well as wholly or partially synthesized polynucleotides

In another aspect, the invention relates to vector comprising the polynucleotide according to the invention. A "vector", as used herein, refers to a nucleic acid molecule capable of transporting another nucleic acid to which it has been linked. The term "yeast expression vector" as used herein refers to DNA expression constructs, e.g., nucleic acid segments, plasmids, cosmids, phages, viruses or virus particles capable of synthesizing the subject proteins encoded by their respective recombinant genes (carried by the vector in a yeast. Alternatively, nucleic acid segments may also be used to create transgenic yeast cells, using non-directional or homologous recombination, in which the gene or genes of interest are stably integrated into the yeast genome.

The polynucleotides of the invention or the gene constructs forming them can form part of a vector. Thus, in another aspect, the invention relates to a vector comprising a polynucleotide or a gene construct of the invention. A person skilled in the art will understand that there is no limitation as regards the type of vector which can be used because said vector can be a cloning vector suitable for propagation and for obtaining the polynucleotides or suitable gene constructs or expression vectors in different heterologous organisms suitable for purifying the conjugates. Thus, suitable vectors according to the present invention include expression vectors in prokaryotes such as pUC18, pUC19, Bluescript and their derivatives, mpl8, mpl9, pBR322, pMB9, CoIEl, pCRl, RP4, phages and shuttle vectors such as pSA3 and pAT28, expression vectors in yeasts such as vectors of the type of 2 micron plasmids, integration plasmids, YEP vectors, centromeric plasmids and the like, expression vectors in insect cells such as the pAC series and pVL series vectors, expression vectors in plants such as vectors of expression in plants such as pIBI, pEarleyGate, pAVA, pCAMBIA, pGSA, pGWB, pMDC, pMY, pORE series vectors and the like and expression vectors in superior eukaryotic cells based on viral vectors (adenoviruses, viruses associated to adenoviruses as well as retroviruses and lentiviruses) as well as non-viral vectors such as pSilencer 4.1-CMV (Ambion), pcDNA3, pcDNA3.1/hyg pHCMV/Zeo, pCR3.1 , pEFl/His, pIND/GS, pRc/HCMV2, pSV40/Zeo2, pTRACER-HCMV, pUB6/V5-His, pVAXl, pZeoSV2, pCI, pSVL and pKSV-10, pBPV-1, pML2d and pTDTl.

Vectors for use with the invention are, for example, vectors capable of autonomous replication and/or expression of nucleic acids to which they are linked in yeast cells. In the present specification, the terms "plasmid" and "vector" are used interchangeably as the plasmid is the most commonly used form of a vector. Moreover, the invention is intended to include such other forms of expression vectors that serve equivalent functions and which become known in the art subsequently hereto. Said yeast expression vector may be a yeast episomal expression vector or a yeast integrative expression vector, and they can be obtained by conventional techniques known for the skilled person in the art.

Thus, in an embodiment, said yeast expression vector is a yeast episomal expression vector. The term "yeast episomal expression vector" as used herein refers to an expression vector that is maintained as an extra-chromosomal DNA molecule in the yeast cytoplasm. In a particular embodiment, said yeast episomal expression vector, in addition to the nucleotide sequence coding for TF protein or a fragment thereof having pro-coagulant activity operatively linked to a yeast-functional promoter, further comprises: (i) a yeast selection gene; (ii) a yeast replication origin; (iii) a bacterial selection gene; and (iv) a yeast transcription termination signal. Advantageously, said yeast episomal expression vector further comprises a unique restriction t greater than Ue for cloning the selected gene (l"b protein or a fragment thereof having pro-coagulant activity) under the control of the yeast-functional promoter and followed by the yeast transcription termination signal. Practically any yeast-functional promoter, yeast selection gene, yeast replication origin, bacterial selection gene, yeast transcription termination signal, and restriction site for cloning, can be used in the manufacture of said yeast episomal expression vector; nevertheless, in a particular embodiment, the glyceraldehyde-3-phosrholiatc dehydrogenase promoter (pGPDj is used as the yeast-functional promoter; in another particular embodiment, the I R A3 gene (UR A3) a , used as yeast selection gene; in another particular embodiment, the yeast 2 microns (2mu) replication origin is used as the yeast replication origin; in another particular embodiment, the am pi oil Hn resistance gene S Amp) is used as the bacterial selection gene; and in another particular embodiment, the transcription termination signal of the phosphogSyceraie kinase (PGKt) is used art the specific yeast transcription termination signal. Thus, in a specific embodiment (Examples 1-3), the yeast episornal expression vector comprises (i) the IJRA 3 gerse; (U) the Amp gene for selecting and propagating the vector in E. roll; fiiij the yeat greater than t 2mu replication origin; (iv) the pGPIX (v) the specific yeat greater than t transcription termination signal of PGKt: and (vi) a unique BawHl restriction site that allows cloning of selected genes under the control of the pGPD, and followed by the PGKt sequence.

In other embodiment, said yeast expression vector is a yeast integrative expression vector. The term "yeast integrative expression vector" as used herein refers to a vector which is capable of integrating into the yeast genome. In a particular embodiment, said yeast integrative expression vector comprises: (i) a bacterial selection gene; and (ii) an expression cassette inserted in a yeast selection gene, said expression cassette further comprising a yeast-functional promoter, a yeast transcription termination signal and a unique restriction site for cloning the selected gene (TF protein or a fragment thereof having pro-coagulant activity).

Practically any bacterial selection gene, expression cassette inserted in a yeast selection gene, yeast-functional promoter, yeast transcription termination signal, and unique restriction site for cloning the selected gene, can be used in the manufacture of said yeast integrative expression vector; nevertheless, in a particular embodiment, the ampicillin resistance gene (Amp) is used as the bacterial selection gene.

The vector of the invention can be used to transform, transfect or infect cells which can be transformed, transfected or infected by said vector. Said cells can be prokaryotic or eukaryotic. By way of example, the vector wherein said DNA sequence is introduced can be a plasmid or a vector which, when it is introduced in a host cell, is integrated in the genome of said cell and replicates together with the chromosome (or chromosomes) in which it has been integrated. Said vector can be obtained by conventional methods known by the persons skilled in the art (Sambrok et al., 2001, mentioned above). Thus, in another aspect, the invention relates to a cell comprising a polynucleotide, a gene construct or a vector of the invention, for which said cell has been able to be transformed, transfected or infected with a construct or vector provided by this invention. The transformed, transfected or infected cells can be obtained by conventional methods known by persons skilled in the art (Sambrok et al., 2001 , mentioned above). In a particular embodiment, said host cell is an animal cell transfected or infected with a suitable vector.

Host cells suitable for the expression of the conjugates of the invention include, without being limited to, mammal, plant, insect, fungal and bacterial cells. Bacterial cells include, without being limited to, Gram-positive bacterial cells such as species of the

Bacillus, Streptomyces and Staphylococcus genus and Gram-negative bacterial cells such as cells of the Escherichia and Pseudomonas genus. Fungal cells preferably include cells of yeasts such as Saccharomyces, Pichia pastoris and Hansenula polymorpha. Insect cells include, without being limited to, Drosophila cells and Sf9 cells. Plant cells include, among others, cells of crop plants such as cereals, medicinal, ornamental or bulbous plants. Suitable mammal cells in the present invention include epithelial cell lines (porcine, etc.), osteosarcoma cell lines (human, etc.), neuroblastoma cell lines (human, etc.), epithelial carcinomas (human, etc.), glial cells (murine, etc.), hepatic cell lines (from monkey, etc.), CHO (Chinese Hamster Ovary) cells, COS cells,

BHK cells, HeLa cells, 911, AT1080, A549, 293 or PER.C6, NTERA-2 human ECC cells, D3 cells of the mESC line, human embryonic stem cells such as HS293 and BGVOl, SHEFl, SHEF2 and HS181, NIH3T3 cells, 293T, REH and MCF-7 and hMSC cells.

Suitable host cells includes those showing enhanced membrane permeability as well as having inactivating mutation in one or more molecular chaperones and have been described in detail above in respect to the method of the invention.

In another aspect, the invention relates to the use of a host cell according to the invention for the identification of compounds which are capable of inhibiting aggregation of an aggregation-prone polypeptide.

In further embodiments, the invention relates to:

[1] Method of screening for a compound that decreases aggregation of aggregation-prone polypeptides, wherein the method comprises (a) contacting one or more yeast cells with a candidate compound, wherein the yeast cells express a fusion protein comprising an aggregation-prone polypeptides and an enzyme which inhibits a toxic compound which affects yeast cell viability (b) adding the toxic compound to the yeast cell in an amount which, without the presence of the enzyme, would affect cell viability.

[2] Method, according to [1] wherein the aggregation prone polypeptide is an amyloidogenic peptide.

[3] Method, according to [2] wherein the amyloidogenic peptide is Aβ42.

[4] Method, according to [2] wherein amyloidogenic peptide is PoIyQ expansions.

[5] Method according to [2] wherein amyloidogenic peptide is α-synuclein variant.

[6] Method, according to [1] to [5] any of the previous claims wherein the enzyme is human dehydrofolate reductase (h-DHFR) and the toxic compound is methotrexate (MTX).

[7] Method, according to any of [1] to [6], wherein the yeast cell is selected from the group consisting of Saccharomyces uvae, Saccharomyces kluyveri, Schizosaccharomyces pombe, Saccharomyces uvarum, Kluyveromyces lactis, Hansenula polymorpha, Pichia pastoris, Pichia methanolica, Pichia kluyveri, Yarrowia lipolytica, Candida sp., Candida utilis, Candida cacaoi, Geotrichum sp. and Geotrichum fermentans.

[8] Method, according to any of [1] to [6] wherein the yeast cell is Saccharomyces cererevisiae.

[9] Method according to any of [1] to [8] wherein the aggregation of the amyloidogenic peptide results in a disease selected from the group consisting of Alzheimer's disease, Parkinson's disease, Familial Amyloid Polyneuropathy, a Tauopathy, Trinucleotide disease, transmissible spongiform encephalopathies

(TSEs), Alzheimer's (AD), and Huntingdon's Disease (HD).

[10] Kit useful for screening compounds that inhibit protein aggregation, characterized in that it uses the method according to [I].

EXAMPLES

In this invention we used different yeast strains depending on the assay to be performed. The strain FY834 was used in the preliminary assays and in the study related to overexpression of chaperones whereas drug testing was done in the drug permeable strain ergo A in the BY4741 parental background. Strains with a deletion of specific chaperone were provided by Euroscarf and were also based in the BY4741 strain. Methotrexate (MTX), sulfanilamide, quercetin and Congo Red were purchased from Sigma Aldrich.

All the polypeptides (Aβ42, polyQ expansions and α-synuclein as well as their variants) were fused to DHFR by a linker with the sequence GSAGSAAGSG (SEQ ID NO;1). The table 4.1 summarizes the protein fusions designed for the experiments.. DHFR was also cloned in the same plasmids (Figure 1). Table 4.1 Design of the fusion proteins

Target proteins Fusion proteins Plasmid Restriction sites Selectable markers

DHFR pESC CM, BgRl Ura/Trp

Aβ42 GFP pESC CM, BgRl Trp

DHFR pESC CM, BgRl Ura/Trp

Aβ42 F19D GFP pESC CM, BgRl Trp

DHFR pESC CM, BgRl Ura/Trp

Q25 GFP p416 SaR, BamHl Ura

DHFR pESC CM, BgRl UraTrp

Q72 GFP p416 SaR, BamHl Ura

DHFR pESC CM, BgRl UraTrp

Q103 GFP p416 SaR, BamHl Ura

DHFR pESC CM, BgRl UraTrp α-syn wt GFP p426 Spel,Xhol Ura

DHFR pESC CM, BgRl UraTrp α-syn A3 OP GFP p426 Spel,Xhol Ura

DHFR pESC CM, BgRl UraTrp α-syn A53T GFP p426 Spel,Xhol Ura

Plasmids encoding for the different yeast chaperones are listed in the following table.

Table 4.2 Plasmids encoding for the chaperones used in this study

Selectable

Chaperone Plasmid Restriction sites markers

Ssal pRS425 BamHl, Hindlll Leu

HsplO4 p2HG BamHl His

Hsp82 pTGpd BamHl Trp

Sisl pTV3 BamHl Trp

Cells were grown overnight at 30⁰C in a selective synthetic complete (SC) media containing raffinose. They were inoculated at OD₆Oo of 0.02 in minimal galactose medium containing the appropriate MTX concentration (20-100 μM) and 1 mM of sulfanilamide. Growth was followed measuring OD₆Oo using Cary 400Bio spectrophotometer. In drug screenings, the protocol was the same being the cells inoculated in a minimal galactose medium containing 20 μM MTX, ImM sulfanilamide and the tested compound: quercetin (30 μM) or CR (10 μM). In this case, a previous incubation before the addition of galactose was performed to assure the presence of the compound in the cell when the fusion protein started to be expressed. This way, before the induction with galactose, quercetin or Congo Red were added at the same concentration as the final assays during 90 minutes. After this period of time, the cells were inoculated in a media containing galactose, MTX and sulfanilamide.

It has to be taken into account that some compounds were dissolved in DMSO.

To control the effects of DMSO on the cells, equal amounts of this solvent were added in the negative control. After 30 hours from the induction, the cultures were diluted 1/100 and its OD₆₀₀ was measured. In all the experiments OD₆₀₀ measurement was the average of triplicate measurements from several independent transformants.

Yeast cells were grown overnight in minimal media at 30⁰C containing raffinose. Cell density was determined by measuring the OD₆₀₀ and cells were diluted to a final OD₆₀₀ of 0.18 using PBS. Afterwards, 10 μl of each dilution (1/10, 1/10 and 1/100) was spotted in plates of selective synthetic complete media with galactose, MTX (20 μM) and sulfanilamide (1 mM). The Petri dishes were incubated for 48 hours at 30⁰C. Images of the plates were taken using the molecular imager Gel Doc XR system from Bio-Rad.

Yeast cells transformed with plasmids encoding the target proteins fused to DHFR were grown overnight in minimal media at 30⁰C containing raffinose. They were inoculated at OD₆₀₀ of 0.02 in minimal galactose medium. Yeast cells were grown during 24 hours and then, they were incubated with sulfanilamide (1 mM) and 10 μM of

MTX labeled with the fluorescent molecule Alexa (Invitrogen) for another period of 24 hours. Afterwards, the medium was removed and the cells were washed with PBS and reincubated for 30 min in the selective medium to allow for efflux of unbound flvlTX.

Then, the cells were centrifuged (1300xg for 5 min) and washed with 50 mM Tris-HCl pH=7 five times. Cells were visualized by microscopy. The main component of AD lesions is the hydrophobic polypeptide Aβ(l-42)²⁰. In previous works, our group has shown that mutation of Phenylalanine 19 to Aspartate (F 19D) abolishes the amyloidogenicity of Aβ42 in vitro²¹ and also reduces Aβ42 aggregation propensity inside E.coli²². These two extreme behaviours make of peptide Aβ42 and its F19D point mutant a promising pair of targets to explore factors influencing protein aggregation in eukaryotic backgrounds, particularly in yeast.

In order to study the aggregation behaviour of the proteins in an eukaryotic background, both variants were expressed in S. cerevisiae as fusion proteins with GFP. While the GFP-fusion with mutant Aβ42(F19D) did not aggregate intracellularly and its fluorescence was distributed diffusely throughout the cell, the Aβ42-GFP protein fluorescence was concentrated in a single large aggregate in a juxtanuclear position, as shown by co-staining with Hoechst (Figure 2). Immunob lotting of total cellular protein indicated that both proteins were expressed at similar levels, demonstrating that the degree of coalescence exhibited by Aβ42 forms is more dependent on their sequence than on the level of protein expressed.

The lethality of MTX on S. cerevisiae FY384 cells above certain concentrations (such as 25 μM with 1 mM of sulfanilamide) can be overcome by transformation with a plasmid encoding human DHFR (h-DHFR) under the control of GaIl O promoter. Different degrees of sensitivity to MTX may thus be correlated with the intracellular activity of the heterologously expressed enzyme, which is likely to vary depending on its expression alone, as a fusion with soluble molecules or as a fusion with an aggregation-prone polypeptide. In the latter case, the fused protein would promote, at least in part, its deposition lowering the intracellular DHFR activity and causing a higher sensitivity to MTX. Since the intrinsic aggregation propensities of Aβ42 and Aβ42(F19D) peptides determined the fate of the fused GFP within yeast, both peptides were fused to h-DHFR. The differential growth abilities of cells expressing these fusions were compared with that of cells expressing h-DHFR alone.

In the presence of MTX, the growth of yeast expressing Aβ42-DHFR became stationary after a short period of time; whereas cells expressing h-DHFR or peptide Aβ42(Fl 9D)-DHFR displayed clearly higher and similar growth rates (Figure 4). In the absence of MTX, no significant differences in growth rates could be observed indicating that the growth divergence could not be attributed to a differential toxicity of the expressed proteins.

In order to assess if the detected differences in viability were caused by the dissimilar solubility of h-DHFR due to its fusion to the Aβ42 variants, a filter trap assay was performed. As it could be inferred from the target protein fusions with GFP, protein aggregates were only detected in the case of yeast cells expressing the wild-type form fused to DHFR, whereas Aβ42(F19D)-DHFR or DHFR alone did not present any aggregate within the cell.

The system should allow also monitoring the influence of culture conditions on polypeptide aggregation. It is assumed that high temperatures promote in vivo and in vitro protein aggregation by reinforcing hydrophobic intermolecular interactions among polypeptides. Increasing the growth temperature from 30⁰C to 37°C resulted in a decrease in viability of cells expressing Aβ42-DHFR at all the MTX concentrations assayed, but also of cells expressing Aβ42(F 19D)-DHFR at high MTX concentrations (Figure 3). No such phenotypic effect was observed in cells expressing h-DHFR alone. This result suggested that temperature was specifically increasing, directly or indirectly, the aggregation propensity of the Aβ moiety and that the system was sensitive enough to detect such effect.

Overall, these results indicated that h-DHFR activity, as reflected in cell growth, could be used as a reporter to monitor the influence of both intrinsic and extrinsic factors on the aggregation of a given polypeptide.

The reliability of the indirect methods to study in vivo aggregation has to be always confirmed by measuring the solubility of the target protein using another approach. In our case, one strategy was the fusion of the same target to GFP. Then, it was expressed in the same eukaryotic background and the aggregation state of the target protein was imaged by monitoring the localization of the fluorescence. Alternatively, the solubility of the target protein fused to DHFR was also detected by filter trap assay. In this context, we sought to exploit the ability of h-DHFR to bind MTX with high affinity in a 1 : 1 complex in order to visualize the presence of the active enzyme within the cells. With this purpose, we used a version of the inhibitor labeled with a fluorescent compound (fMTX). It has been proved before that fMTX is retained into the cells through this binding to DHFR, whereas the unbound fMTX is actively and rapidly transported outside. In addition, the interaction of fMTX and DHFR results in a 4.5-fold increase in quantum yield. Thus, bound fMTX, and by inference the properly folded DHFR, could be monitored by fluorescence microscopy.

The data obtained by imaging the location of fMTX in cells expressing h-DHFR, Aβ42-DHFR and Aβ42(Fl 9D)-DHFR were absolutely coincident with those obtained using GFP as a tag.

While cells expressing h-DHFR and Aβ42(F 19D)-DHFR had no aggregates and their fluorescence was high and distributed diffusely throughout the cell, the fluorescence of the ones expressing Aβ42-DHFR was lower, indicating less amount of folded enzyme, and it was concentrated in a single large aggregate per cell. It could be deduced that the sequence of the target protein determined the aggregated state of the enzyme. Also, the solubility of the target protein could be easily and simultaneously assayed by measuring cellular viability in the presence of MTX as well as by imaging the location of the fluorescent inhibitor inside the cells.

To test the general applicability of the method, other aggregation-prone proteins should be fused to DHFR verifying the link between protein aggregation and yeast survival. With this purpose, polyglutamine expansions and α-synuclein protein were chosen due to their aggregation propensity and their implication in human pathologies:

Huntington's and Parkinson's diseases, respectively. In the case of polyglutamine expansions, it has been previously demonstrated that their aggregation propensity depends on their length. As it was performed in the previous case, the distribution of the different variants in living yeast cells was studied through their fusion to GFP. Whereas cells expressing extensions of 25 glutamines (Q25) displayed fluorescence diffusely distributed trough the cell, yeasts expressing 72 (Q72) or 103 glutamines (Q 103) exhibited various fluorescent foci. Also, it was noticeable that the number and size of foci increased with the length of polyQ expansions (figure 7).

In the absence of MTX, the fusion of po Iy-Q fragments to DHFR reduced cell viability. Nevertheless, all polyQ repeats displayed similar effect on viability independently of their length.

In the presence of MTX, the yeast growth was inversely proportional to polyQ expansions length. In other words, yeasts expressing Q25-DHFR exhibited the highest growth rate. Therefore, the survival of yeast cells expressing different polyQ fused to DHFR correlated with the observed solubility of the GFP fusions. This link indicated that, under the conditions of the assay, the presence of properly folded DHFR moiety determined the cell growth.

The use of fMTX provided results in excellent agreement with these data. In cells expressing Q25-DHFR, the enzyme was soluble because the fluorescence was high and homogenously distributed in the cell. On the contrary, the presence of Q72 or Q 103 expansions strongly reduced the fluorescence emission and the active enzyme became mainly located into aggregates (figure 8).

The protein α-synuclein (α-Syn) forms the fibrous portion of Lewy Bodies, cytoplasmatic inclusions present in Parkinson's disease (PD). Two rare early-onset forms of PD are linked with mutations in the α-Syn gene: A53T and A30P. Both variants have distinct physical properties: A53T is accumulated at the plasma membrane or in cytoplasmatic foci like wild-type α-Syn; whereas A30P is dispersed through the cell.

When all the α-Syn variants were fused to GFP, the above described phenotypes were reproduced (Figure 9). When the variants were fused to DHFR and expressed in yeast in the absence of MTX, cell viability was reduced in a similar extent indicating certain toxicity of the α-Syn gene when it was expressed in yeast. In the presence of MTX, the expression of α-SynA30P-DHFR allowed a cell growth rate very close to the one of cells expressing h-DHFR alone, whereas the α- SynA53T-DHFR or α-Syn-DHFR variants displayed a significantly reduced viability.

Overall, the survival of yeast expressing the α-Syn variants fused to DHFR correlated well with the observed solubility of the GFP-fusions. This fact demonstrated again that the presence of properly folded DHFR controlled cell growth under the conditions of the assay (figure 10).

Accordingly to all these data, the fMTX fluorescence of cells expressing α-

SynA3 OP-DHFR was higher and no aggregates were observed. On the other hand, punctuated nuclei, usually close to the cytoplasmic membrane, were observed for the wild-type and A53T mutant.

The identification of small chemical compounds that affect the size or the number of protein inclusions inside cells is one of the approaches towards therapeutic intervention against depositional diseases. We wanted to explore whether the previously outlined method could be useful to screen for such substances. As test case, ergόΔ yeast strain cells expressing h-DHFR, and Aβ42-DHFR were grown in the presence of selected concentrations of the compounds quercetin and Congo Red (CR), both previously shown to in vitro bind Aβ aggregates.

This way, the assay could detect the previously reported inhibitory aggregation activity of quercetin (CAS Number [117-39-5]). The presence of this flavonoid in the medium restored the growth of cells expressing wild-type Aβ42 and reduced the size of intracellular Aβ aggregates, suggesting that it targets in vivo Aβ aggregation. In contrast, under the conditions of the assay, Congo Red (hereinafter also referred as CR) (CAS Number [573-58-0]) had not positive effect on growth rate. Consistently, it has been recently shown that, in vitro, CR inhibits Aβ42 oligomerization but does not affect its fibrillization. Furthermore, it has been demonstrated that low concentrations of CR can promote fibril formation as shown for the Aβl 1-28 fragment, the prion protein or immunoglobulin light chains, questioning the therapeutic utility of this compound or its analogs to inhibit amyloidosis. The increased number of aggregation foci promoted by CR in our system might reflect a similar effect on Aβ aggregation inside yeast explaining the observed rather negative impact on cell survival.

The analysis of the effect of overexpression or deletion of chaperones demonstrated the applicability of the method in genetic screening. Overall, modification of chaperones levels had huge impact in the survival of cells and in the size and distribution of intracellular Aβ aggregates.

Among the entire chaperone set, overexpression of the Sisl gene (the yeast homologue of human HDJl) had the most dramatic effect on viability. This result is not surprising because the knockout of sisl gene is not viable, being thus essential for yeast. Its effect is clearly higher than this promoted by Ydjl (the yeast Hsp40 homologue of human HD J2). In agreement with our data, both in mammalian and in yeast cells, the Sisl/HDJl chaperones had a much stronger effect on modulating the aggregation of polyQ than Ydjl/HDJ2.

Overexpression of HsplO4 also increased cell viability. HsplO4 allows rescuing proteins from aggregated states regaining their function. It is a key protein in the chaperone network. Puzzlingly, Sisl or specially HsplO4 promoted an increase of fluorescence signal in the cytoplasm and in the aggregates. This coincides with their effect in the yeast model of HD, where it was shown that the protein in those aggregates was more loosely packed. Taking into account the relationship between packing of Aβ- GFP aggregates and the activity of the embedded protein that was established in bacteria, these chaperones might promote loosely packed and probably more active Aβ- DHFR aggregates.

In contrast, Hsp82 promoted viability seems to be dependent on direct reduction of Aβ42 aggregation, resulting in smaller intracellular foci. Interestingly, also human homolog interacts with amyloid precursors in AD and its upregulation protects neurons from Aβ toxicity. Surprisingly, the knockout of Hsp35, Hsp26 and Hsp42 chaperones promoted cell growth, implying lower Aβ aggregation. Hsp35 is a member of the glyceraldehyde- 3-phosphate dehydrogenase (GAPDH) family. And it is supposed to be a chaperone because of its heat inducibility and its high abundance in yeast. Intriguingly, in humans, polymorphic variation within GAPDH genes is associated with an elevated risk of developing AD and this effect depends specifically on the interaction with Aβ peptide. Also, the Hsp35 Caenorhabditis elegans homolog is down regulated in transgenic animals expressing Aβ42.

Hsp26 and Hsp42 are small heat shock proteins (sHsp), which trap misfolded proteins into aggregates that are subsequently reactivated by the Hspl04/Hsp70/Hsp40 chaperone system. Nevertheless, larger substrate/sHsp relationships result in larger (and maybe tighter) complexes, which are poorly reactivated by other chaperones. The knockout of sHsp might prevent the incorporation of overexpressed Aβ42 into sHsp aggregates reducing indirectly its deposition.

The loss of function of Ydjl has been shown to reduce polyQ aggregation, an observation compatible with the increase in yeast survival promoted by its knockout in our model.

On the other hand, deletion of the members of the Hsp70 cytosolic family (Ssal, Ssa2 and specially Ssa4) reduced cell viability and increased the number of fluorescent foci within the cells. Importantly, in a yeast model of polyQ aggregation, mutation of the SSAl, SSA2 genes also inhibited the expansion of small aggregate foci into a large inclusion body. Finally, the huge impact on both viability and aggregation of SSa4 deletion suggest that it could play an important role in protein folding and/or deposition. Accordingly, under conformational stress the amounts of Ssa4 mRNA increase several fold.

Overall, and although the present study was not aimed to characterize in detail the effect of each specific chaperone on the intracellular aggregation of Aβ, the obtained data confirm that the aggregation of disease-related polypeptides in yeast share several hallmarks suggesting also new and differential targets to study Aβ aggregation.

The prevalence of the conformational diseases in our society presents a new challenge both for basic and applied research. In the last years the simple, yet powerful, genetics of S. cerevisiae has been exploited for the study of intracellular amyloid protein aggregation. From the present study, it appears that Alzheimer's disease may be yet another disorder whose modelling in yeast could contribute to decipher conserved and/or differential mechanisms of amyloid aggregation and help in the identification of potential therapeutic targets. The ability of the method discussed here to link protein aggregation to cell survival is expected to allow a fast, visual and easily automated screening of target mutations, genes or compounds that modulate protein aggregation of disease-related polypeptides inside eukaryotic cells. Besides, the method should have wide applicability in protein production and design as well as in folding studies.

The erg6 mutation inhibits ergosterol biosynthesis, which enhances membrane fluidity and permeability to various chemical compounds. This results in a four-fold higher sensitivity to MTX when compared to the FY384 strain and in a concomitant decrease in the viability of ergόA cells expressing Aβ42-DHFR relative to those expressing h-DHFR. Thus, to specifically monitor the effects of compounds on the viability of cells expressing the Aβ42 fusion, growth was always referenced to that of cells expressing h-DHFR under the same conditions. Quercetin is a flavonoid compound shown to inhibit in vitro Aβ fibril formation and to reduce the toxicity of Aβ fragments in neuroblastoma cells. In the presence of MTX, cells expressing peptide Aβ42-DHFR displayed a growth restoration dependent on quercetin concentration (Figure 11 A). Interestingly enough, this effect was also reported in a yeast-based model of α-synucleinopathy at similar quercetin concentrations.

On the other hand, despite the fact that CR function as amyloid ligand was widely reported, no significant effect could be observed at 10 μM (Figure 11 A). Strikingly, this CR concentration was shown to reduce significantly the aggregation of huntingtin with expanded polyglutamine (polyQ) inside mammalian cells. The different activity of both substances on Aβ42 aggregation in yeast could be rationalized by analyzing fluorescence microscopy images of cells expressing peptide Aβ42 fused to GFP in media containing selected concentrations of quercetin or CR (Figure 11 B). Cells grown in the presence of 30 μM quercetin displayed a unique aggregate per cell with a smaller average diameter (0.6 μm) than those formed in the absence of the compound (1.0 μm), suggesting that it effectively targets in vivo intracellular protein insolubility. Cells treated with 10 μM CR presented one big fluorescent focus with an average diameter of 1.1 μm as well as many smaller ones. Thus, although CR might interfere with the aggregation process of Aβ42 in yeast, it did not increase the effective soluble protein concentration and rather promoted the appearance of new aggregation foci.

It has been shown that changes in the levels of chaperones in the cell modulate protein aggregation in yeast models of both Huntington's and Parkinson's diseases. To evaluate whether our method could be also sensitive to differences in the composition and concentration of proteins involved in the cellular folding machinery, a set of chaperone knockouts and chaperone overexpressing strains producing h-DHFR and

Aβ42-DHFR were engineered. In the table 4.3, there is a classification off the all chaperones used in this section according to their family.

Table 4.3 List of the chaperones studied in this section Family Members

HsplOO HsplO4 Hsp90 Hsc82, Hsp82 Hsp70 Ssal, Ssa2, Ssa3, Ssa4 Hsp40 Ydjl, Sisl Small heat shock proteins Hsp26, Hsp42

Overexpression of all tested chaperones promoted an increase in the viability of cells expressing Aβ42-DHFR in liquid media containing MTX. The chaperones with greater effect were in decreasing order: Sisl, HsplO4, Hsp82 and Ssal with Ydjl having a rather moderate impact on yeast survival (Figure 12 A). However, in parallel spotting assays in the presence of MTX only Sisl and Hsp82 overexpression promoted clear increase in the cell survival. (Figure 12 B).

Western blot analysis against Aβ42 moiety demonstrated the formation of SDS- stable oligomers (Figure 12). These oligomers were reminiscent of the ones formed by the Aβ42 peptide in vitro, in mammalian cell culture, and in the human brain.

Interestingly, the expression in yeast of Aβ42 fused to the MRF domain of the Sup35 prion resulted in a very similar distribution of SDS-stable oligomers. Overall, the levels and distribution of Aβ42-DHFR were not affected by chaperone overexpression suggesting that they target larger aggregates causing the observed differences in viability.

Accordingly, fluorescence microscopy images of yeast expressing Aβ42-GFP showed that chaperones strongly affected the distribution of fluorescence (Figure 12). Cells overexpressing HsplO4 dramatically increased the number and intensity of fluorescent inclusions as well as of background fluorescence. Sisl overexpressing cells presented generally one big aggregate and several minor aggregation nuclei with high background fluorescence. On the other hand, overexpression of Hsp82 and Ssal resulted in a phenotype similar to the one reported in the presence of quercetin, with only a small fluorescent focus per cell. And the overexpression of Ydj caused the appearance of several small foci per cell (figure 13A)

To test if the depletion of chaperones also affected the intracellular aggregation of Aβ42, strains with a knockout in one specific chaperone gene were transformed with plasmids encoding h-DHRF or peptide Aβ42-DHFR and their growth was measured in the presence of MTX. Surprisingly, depletion of some chaperones resulted in increased cell growth relative to the wild-type strain (hsp35A, hsp42A and hsp26A) whereas others resulted in decreased cell density (ssal A, ssa2A, hsc82A or hsplO4A) and one strain did not grow (ssa4A) (figure 14). Western blot analysis indicated that no significant differences in global expression of Aβ42-DHFR existed between knockouts but some chaperones affected oligomer distribution, (figure 15B).

To test whether the deletion of chaperones influenced the intracellular distribution of peptide Aβ42 aggregates, some knockout strains were transformed with this peptide fused to GFP (Figure 15A). hsp35A cells exhibited a single fluorescent focus with a smaller diameter (0.6 μm) than cells with wild-type background. On the other hand, deletions of Ssal, Ssa2 or HsplO4 resulted in the appearance of more than one Aβ42-GFP aggregate per cell. It is worth to notice that the ssa4A strain, which has lost the ability to grow in the presence of MTX albeit overexpressing h-DHFR, displayed a large number of Aβ42-GFP aggregates indicating an anomalous increased aggregation of polypeptides in the absence of this gene.

The following strains of Saccharomyces cerevisiae were used:

Saccharomvces cerevisiae strain FY834

Genotype: MAT a ura3-52 leu2-\ trpl-63 his3-200 fys2-202

Saccharomvces cerevisiae strain BY4741

Genotype: MAT a his3Δl; leu2A0; met 15A0; urα3Δ0

It is provided by Euroscarf (European Saccharomyces cerevisiae Archive for Functional analysis)

Saccharomvces cerevisiae strain ergόA

Genotype: MAT a his3Al leu2A0 metlδAO ura3A0 ergόA: :kanMX4.

This strain (based in the BY4741 parental background) has a mutation that affects cell permeability, specifically, in the gene codifying C-24 sterol methyltransferase Erg6p.

The following Saccharomyces cerevisiae vectors were used: pESC vectors (Stratagene) The pESC vectors are a series of epitope-tagging vectors designed for expression and functional analysis of eukaryotic genes in the yeast S. cerevisiae. These vectors contain the GALl and GALlO yeast promoters and the yeast 2μ origin, which enables autonomous replication of the plasmids in S. cerevisiae. They have a selectable marker gene (HIS3, TRPl, LEU2, or URA3) to select and maintain the expression vector in yeast cells.

The following Saccharomyces cerevisiae media has been used:

YDP (Yeast Extract Peptone Dextrose) media

Dissolve the following compounds to 800 ml H₂O:

1O g of Bacto Yeast extract 20 g of BactoPeptone 20 g Dextrose Adjust volume to 1 liter with ddH₂O. Sterilize by autoclaving.

SC (Synthetic Complete) drop-out media Dissolve the following compounds in 1 liter ddH₂O 6.7 g Yeast nitrogen base without amino acids 100 ml of the appropriate sterile 10x Drop Out Solution Adjust the pH to 5.8 if necessary, and autoclave. Add the appropriate sterile carbon source, usually dextrose (glucose) to 2%.

Drop Out solution

To make one liter of 10x -Leu/-Trp/-URA/-His Drop Out solution, combine the components listed in the table below taking into account the amino acid that has to be omitted from the Drop Out. For example, if the Drop Out is -Leu, do not add this component to the final solution.

Table M.1 Components of the Drop Out solution

Constituent Final mg/ml Stock solution for 100 ml dH2O ml stock for 11 media

Adenine sulfate 20 200 mg* 10

Uracil 20 200 mg* 10

L-tryptophan 20 I g 2

L-histidine-HCL 20 I g 2

L-arginine-HCL 40 1 g 4

Some stocks should be stored at room temperature (indicated by an asterisk) to prevent precipitation, while the others should be refrigerated. Afterwards, the solution has to be autoclaved.

The preparation and of yeast competent cells will be known by the skilled person in the art, and can be, for example, those described in Gietz, R.D. & Schiestl, R.H. Frozen competent yeast cells that can be transformed with high efficiency using the LiAc/SS carrier DNA/PEG method. Nat Protoc 2, 1-4 (2007), which are the method used in the present invention.

The transformation of yeast competent cells will be known to a skilled person in the art. As an example of such methods, the process used in the present invention is the following:

1) Thaw cell samples in a 37°C water bath for 15-30 s.

2) Centrifuge at 13000xg for 2 min and remove the supernatant.

3) Make up frozen competent cell (FCC) transformation mix for the planned number of transformations plus one extra. Include an extra tube for a negative control tube for no plasmid DNA. Add this to the pellet and vortex mix vigorously to resuspend the cell pellet.

Table M.3 Components of the transformation mix solution for yeast.

Transformation mix components Volume (ml)

PEG 3350 (50% (w/v)) 260

LiAc 36

Single-stranded carrier DNA (2mg/ml) 50

Plasmid DNA plus sterile water 14 Total volume 360

4) Add 70 μl of DMSO. Mix well by gentle inversion or swirling.

5) Incubate at 30⁰C for 30 min with shaking (200 r.p.m).

6) Heat shock at 42°C in a water bath for 15 min. 7) Chill cells on ice for 1-2 min.

8) Centrifuge the tubes at 13000xg for 30 s and remove the supernatant.

9) Pipette 1.0 ml of sterile water into the transformation tube.

10) Plate and spread 200 μl of the cell suspension onto the appropriate SC selection medium. Cells should be plated less densely when possible because plating density negatively affects transformation efficiency.

Incubate the plates at 30⁰C for 3-4 days and recover the transformants.

Claims

1. A method for the identification of compounds that are capable of decreasing aggregation of an aggregation-prone polypeptide comprising: (i) contacting one or more yeast cells with a candidate compound, wherein the yeast cells express a fusion protein which comprises an aggregation-prone polypeptide and an enzyme, wherein the enzyme is capable of modifying a compound which adversely affects yeast cell viability into a metabolite with a reduced adverse effect on said yeast cell viability,

(ii) adding the toxic compound to the yeast cell of step (i) in an amount which, without the presence of the activity of the enzyme forming part of the fusion protein used in step (i), would adversely affect the yeast cell viability and (iii) determining the viability of the yeast cells wherein an increased viability of the cells with respect of the cells which have not been exposed to the candidate compound is indicative that the compound is capable of decreasing aggregation of the aggregation-prone polypeptide.

2. Method, according to claim 1 wherein the aggregation prone polypeptide is an amyloidogenic peptide.

3. Method, according to claim 2 wherein the amyloidogenic peptide is selected from the group of Aβ42, a peptide comprising a Poly-glutamine expansions and a α- synuclein variant.

4. Method, according to any of claims 1 to 3 wherein the enzyme which adversely affects yeast cell viability is human dehydrofolate reductase (h-DHFR) and the toxic compound is methotrexate (MTX).

5. Method according to any of claims 1 to 4 wherein the toxic compound is fluorescently labelled.

6. Method according to claim 5 wherein step (iii) further comprises detecting the fluorescence in the yeast cell wherein an increased intracellular fluorescence is indicative that the compound is capable of decreasing aggregation of the aggregation-prone polypeptide.

7. Method, according to any of claims 1 to 6, wherein the yeast cell is selected from the group consisting of Saccharomyces cererevisiae, Saccharomyces uvae, Saccharomyces kluyveri, Schizosaccharomyces pombe, Saccharomyces uvarum, Kluyveromyces lactis, Hansenula polymorpha, Pichia pastoris, Pichia methanolica, Pichia kluyveri, Yarrowia lipolytica, Candida sp., Candida utilis,

Candida cacaoi, Geotrichum sp. and Geotrichum fermentans

8. Method according to any of claims 1 to 7 wherein the yeast cell shows an increased membrane permeability.

9. Method according to claim 8 wherein the yeast strain carries an inactivating mutation in the ergo gene.

10. Method according to any of claims 1 to 9 wherein the yeast cell carries an inactivating mutation in one or more molecular chaperones selected from the group of a member of the HsplOO protein family, a member of the Hsp90 protein family, a member of the Hsp70 protein family, a member of the Hsp40 protein family or a small heat shock protein.

11. A method as defined in claim 10 wherein the member of the HsplOO protein family is HsplO4, the member of the Hsp90 protein family is selected from the group of Hsc82 and Hsp82, the member of the Hsp70 protein family is selected from the group of Ssal, Ssa2, Ssa3 and Ssa4, the member of the Hsp40 protein family is selected from the group of Ydj l and Sisl and/or the small heat shock protein is selected from the group of Hsp26 and Hsp42.

12. A polypeptide comprising an aggregation-prone polypeptide and a polypeptide having enzymatic activity wherein said polypeptide having enzymatic activity is capable of modifying a compound which adversely affects yeast cell viability into a metabolite of said compound with a reduced adverse effect on said yeast cell viability.

13. A polypeptide as defined in claim 12 wherein the aggregation-prone polypeptide is an amyloidogenic peptide.

14. A polypeptide as defined in clam 13 wherein the amyloidogenic peptide is selected from the group of Aβ42, a polypeptide comprising a poly-glutamine repeat and an α-synuclein or a variant thereof.

15. A polypeptide as defined in any of claims 12 to 14 wherein the polypeptide having enzymatic activity is DHFR.

16. A polypeptide as defined in claim 15 wherein DHFR is human DHFR.

17. A polypeptide as defined in any of claims 12 to 16 further comprising a reporter polypeptide.

18. A polypeptide as defined in claim 17 wherein the reporter polypeptide is a fluorescent protein.

19. A polypeptide as defined in claim 18 wherein the fluorescent protein is GFP.

20. A polynucleotide encoding a polypeptide as defined in any of claims 12 to 19.

21. A vector comprising a polynucleotide as defined in claim 20.

22. A host cell comprising a polypeptide as defined in any of claims 12 to 19, a polynucleotide as defined in claim 20 or a vector as defined in claim 21.

23. A host cell as defined in claim 22 wherein the host cell is a yeast cell.

24. A host cell as defined in claim 23 wherein the yeast cell is Saccharomyces cerevisiae.

25. A host cell as defined in any of claims 22 to 24 wherein the yeast cell shows enhanced membrane permeability.

26. A host cell as defined in claim 25 wherein the yeast cell carries an inactivating mutation in the ergβ gene.

27. A host cell as defined in any of claims 22 to 26 wherein the cell carries an inactivating mutation in one or more molecular chaperones selected from the group of a member of the HsplOO protein family, a member of the Hsp90 protein family, a member of the Hsp70 protein family, a member of the Hsp40 protein family or a small heat shock protein.

28. A host cell as defined in claim 27 wherein the member of the HsplOO protein family is HsplO4, the member of the Hsp90 protein family is selected from the group of Hsc82 and Hsp82, the member of the Hsp70 protein family is selected from the group of Ssal, Ssa2, Ssa3 and Ssa4, the member of the Hsp40 protein family is selected from the group of Ydj l and Sisl and/or the small heat shock protein is selected from the group of Hsp26 and Hsp42.

29. Use of a host cell as defined in any of claims 22 to 28 for the identification of compounds which are capable of inhibiting aggregation of an aggregation-prone polypeptide.

30. Method of screening for a compound that decreases aggregation of aggregation- prone polypeptides, wherein the method comprises (a) contacting one or more yeast cells with a candidate compound, wherein the yeast cells express a fusion protein comprising an aggregation-prone polypeptides and an enzyme which inhibits a toxic compound which affects yeast cell viability (b) adding the toxic compound to the yeast cell in an amount which, without the presence of the enzyme, would affect cell viability.

31. Method, according to claim 1 wherein the aggregation prone polypeptide is an amyloidogenic peptide.

32. Method, according to claim 2 wherein the amyloidogenic peptide is Aβ42.

33. Method, according to claim 2 wherein amyloidogenic peptide is PoIyQ expansions.

34. Method, according to claim 2 wherein amyloidogenic peptide is α-synuclein variant.

35. Method, according to any of the previous claims wherein the enzyme is human dehydrofolate reductase (h-DHFR) and the toxic compound is methotrexate (MTX).

36. Method according to any of claims 1 to 6 wherein the toxic compound is fluorescently labelled.

37. Method, according to any previous claims, wherein the yeast cell is selected from the group consisting of Saccharomyces uvae, Saccharomyces kluyveri,

Schizosaccharomyces pombe, Saccharomyces uvarum, Kluyveromyces lactis, Hansenula polymorpha, Pichia pastoris, Pichia methanolica, Pichia kluyveri, Yarrowia lipolytica, Candida sp., Candida utilis, Candida cacaoi, Geotrichum sp. and Geotrichum fermentans

38. Method, according to any of claims 1 to 6 wherein the yeast cell is

Saccharomyces cererevisiae.

39. Method according to any previous claims, wherein the aggregation of the amyloidogenic peptide results in a disease selected from the group consisting of

Alzheimer's disease, Parkinson's disease, Familial Amyloid Polyneuropathy, a Tauopathy, Trinucleotide disease, transmissible spongiform encephalopathies (TSEs), Alzheimer's (AD), and Huntingdon's Disease (HD).

40. Kit useful for screening compounds that inhibit protein aggregation, characterized in that it uses the method according to claim 1.