The present invention relates to a method for the quantification of seeding activity (At ₅₀ ) of an amyloidogenic aggregate, methods for assessing the risk for development, predicting the onset or assessing the progression of a polyQ disease, and a method for identifying compounds that inhibit mHTT seeding activity (HSA) in vitro. Further, uses of fluorophore-bearing polyQ proteins, particularly mutant N-terminal huntingtin fragments comprising exon-1 and related soluble protein constructs are provided.

BACKGROUND OF THE INVENTION

Self-propagating protein aggregates are a pathological hallmark of a large number of neurodegenerative diseases (NDs) including Huntington’s disease (HD) (Chiti and Dobson, 2017; Jucker and Walker, 2013). Recent studies indicate that aggregate pathology and associated tissue atrophy do not appear randomly throughout the brain but instead progress along distinct neuronal networks (Brundin et al., 2010). Evidence was provided that amyloidogenic protein assemblies spread from cell to cell, converting free molecules of the same protein into aggregated species. This transcellular propagation may drive pathogenesis in NDs (Guo and Lee, 2014; Pecho-Vrieseling et al., 2014). To understand the mechanisms of disease development and progression, it is of critical importance to specifically monitor the activity of self-propagating protein aggregates in complex biosamples.

A number of assays have been established that allow the quantification of seeding activity of amyloidogenic aggregates in crude protein homogenates (Atarashi et al., 2007; Holmes et al., 2014; Tan et al., 2015). These methods take advantage of the phenomenon that ordered protein aggregates are formed from monomers by a nucleation-dependent process (Jarrett and Lansbury, 1993; Scherzinger et al., 1999), a relatively slow process in vitro. However, spontaneous amyloid formation can be accelerated by addition of preformed aggregates that function as seeds for the conversion of monomers from a soluble into an aggregated state (Cohen et al., 2012; Jarrett and Lansbury, 1993). Biosamples that contain seeding-competent protein aggregates might therefore stimulate the polymerization of soluble amyloidogenic proteins with related amino acid sequences in cell-free or cell-based seeding assays.

Based on this premise, the protein misfolding cyclic amplification (PMCA) technology and related methods (Atarashi et al., 2007; Atarashi et al., 2011 ; Saborio et al., 2001 ) have been developed, which allow the detection of minute quantities of seeding-competent PrP ^Sc aggregates in various biomaterials prepared from patients or rodent models with prion disease (Castilla et al., 2005). Variants of the PMCA technology have also been applied for the amplification of amyloid-b and a-synuclein aggregates from biosamples (Du et al., 2011 ; Herva et al., 2014). A key feature of PMCA methods is that seed- mediated amyloid polymerization is indirectly monitored through the reporter dye Thioflavin T (ThT), which changes its fluorescence emission upon binding to ordered amyloid fibrils (Biancalana and Koide, 2010). Also cell-based amyloid polymerization assays have been developed (Holmes et al., 2014; Tan et al., 2015). In these assays, ectopically expressed aggregation -prone reporter proteins with fluorescent tags are utilized as biosensors for detecting amyloidogenic aggregates.

Recent studies with brain slices, fly and mouse models provide evidence that mHTT aggregates with pathogenic polyQ tracts indeed possess seeding activity and spread from cell to cell (Pecho-Vrieseling et al., 2014; Babcock and Ganetzky, 2015; Pearce et al., 2015), suggesting that proteopathic mHTT seeding in HD patient brains or mouse models drives pathogenesis (Brundin et al., 2010; Jeon et al., 2016). Several mHTT aggregate species, i.e. small oligomers and fibrils, have been described as potentially pathogenic (Nucifora et al., 2012; Pieri et al., 2012; Scherzinger et al., 1997). To be regarded as disease relevant, the inventors propose that seeding- competent aggregates, e.g. mHTT aggregates, need to be detectable in affected brain regions in patients and transgenic HD models, e.g. mouse models. To promote disease development, such structures should be present in model systems prior to the appearance of a disease phenotype. Also, their abundance in affected tissues should increase with the severity of disease symptoms. Finally, perturbation of seeding activity through genetic manipulation should influence the disease phenotype in model systems. In summary, to elucidate the potential importance of seeding in disease, it is crucial to detect seeding-competent structures in relevant biosamples, and to investigate their potential impact on biological functions and phenotypes.

BRIEF SUMMARY OF THE INVENTION

The present inventors surprisingly found that quantification of seeding activity of amyloidogenic proteins like mHTT is possible with high sensitivity and specificity even in complex biosamples with FRET-based assay.

Accordingly, the present invention provides methods involving FRET-based determination of seeding activity (At ₅₀ ) in a variety of biological samples.

In a first aspect, a method for the quantification of seeding activity (At ₅₀ ) of an amyloidogenic aggregate is provided, which comprises the steps of:

(i) providing, in a solution, a mixture of an amyloidogenic protein A which is N- terminally or C-terminally fused to a donor fluorophore molecule, e.g. a cyan fluorescent protein (CFP), and an amyloidogenic protein B which is N- terminally or C-terminally fused to an acceptor fluorophore molecule, e.g. a yellow fluorescent protein (YFP), wherein the amyloidogenic proteins A and B are preferably identical and wherein the donor fluorophore molecule and the acceptor fluorophore molecule are capable of Forster Resonance Energy Transfer (FRET) if they are in close proximity to each other; (ii) adding a sample containing an amyloidogenic protein aggregate C to the mixture of step (i), wherein the amyloidogenic protein aggregate C preferably comprises or consists of amyloidogenic proteins A and/or B;

(iii) shaking the mixture of step (ii);

(iv) measuring fluorescence signals in the donor, e.g. cyan, channel, the acceptor, e.g. yellow, channel and the Forster Resonance Energy Transfer (FRET) channel at predetermined intervals after completion of step (iii);

(v) calculating FRET efficiency (E) from the signals obtained in step (iv); and

(vi) quantifying seeding activity (At ₅₀ ) by subtracting the time at half-maximal FRET efficiency of a sample (t ₅₀ (S))from the time at half-maximal FRET efficiency of a negative control (t ₅ o(0)).

In a further aspect, a method for assessing the risk for development of a polyglutamine (polyQ) disease in a subject is provided, which comprises the steps of:

(i) quantification of seeding activity (At ₅₀ ) of a polyQ-containing protein in a sample;

(ii) correlating that the subject is at risk for development of the polyQ disease when the seeding activity in the sample is increased as compared to a reference sample.

In a further aspect, a method for predicting the onset of a polyglutamine (polyQ) disease in a subject is provided, which comprises the steps of:

(i) quantification of seeding activity (At ₅ o) of a polyQ-containing protein in a sample;

(ii) correlating that the onset of the polyQ disease has occurred or will occur soon when the seeding activity in the sample is increased as compared to a reference sample.

In a further aspect, a method for assessing the progression of a polyglutamine (polyQ) disease in a subject is provided, which comprises the steps of:

(i) quantification of seeding activity (At ₅ o) of a polyQ-containing protein in a sample collected at a timepoint t-i; (ii) quantification of seeding activity (At ₅ o) of a polyQ-containing protein in a sample collected at a timepoint t ₂ , wherein t ₂ is later than t-i, comprising the steps (a) to (f) as defined in (i); and

(iii) correlating that the polyQ disease has progressed when the seeding activity in the sample taken at t ₂ is increased as compared to the sample taken at ti.

In yet a further aspect, a method for identifying compounds that inhibit mHTT seeding activity (HSA) in vitro is provided, which comprises the steps of:

(i) determining HSA in a test sample;

(ii) determining HSA in a control sample without the compound or mixture of compounds to be tested for inhibiting HSA in vitro ; and

(iii) selecting compounds or mixtures of compounds which show decreased seeding activity as compared to control samples.

In yet a further aspect, the use of soluble glutathione S-transferase HTT exon-1 fusion proteins in an aggregation assay is provided, wherein a first fusion protein comprises from about 35 to about 75, particularly about 40 to about 55, e.g. 40 or 48 or 49 glutamine residues which is C-terminally fused to a donor fluorophore molecule, e.g. a CFP such as CyPet, and a second fusion protein comprises from about 35 to about 75, particularly about 40 to about 55, e.g. 40 or 48 or 49 glutamine residues which is C-terminally fused to an acceptor fluorophore molecule, e.g. a YFP such as YPet.

Likewise, the use of soluble HTT exon-1 fusion proteins in an aggregation assay is provided, wherein a first fusion protein comprises from about 35 to about 75, particularly about 40 to about 55, e.g. 40 or 48 or 49 glutamine residues which is C-terminally fused to a donor fluorophore molecule, e.g. a CFP such as CyPet, and a second fusion protein comprises from about 35 to about 75, particularly about 40 to about 55, e.g. 40 or 48 or 49 glutamine residues which is C-terminally fused to an acceptor fluorophore molecule, e.g. a YFP such as YPet. In a further aspect, a soluble protein is provided, which comprises, from N- to C- terminus: (i) optionally glutathione S-transferase (GST), particularly of SEQ ID NO.: 6; (ii) exon 1 of huntingtin with 48 glutamine residues (HTTEx1 Q48), particularly of SEQ ID NO.: 2; and (iii) CyPet, particularly of SEQ ID NO.: 23, or YPet, particularly of SEQ ID NO.: 21.

DETAILED DESCRIPTION OF THE INVENTION

There is increasing experimental evidence that self-propagating aggregates, or seeds, of proteins like mHTT, play an important role in model organisms, e.g. HD model organisms (Babcock and Ganetzky, 2015; Pecho-Vrieseling et al ., 2014.

The fluorescent dye Thioflavin T (ThT) is currently utilized in a large number of cell-free assays as a reporter molecule to monitor the seeding activity of amyloidogenic protein aggregates (Gupta et al., 2012). ThT exhibits enhanced fluorescence when it is bound to b-sheet-rich amyloid structures (LeVine, 1993). However, its binding to such structures is significantly decreased, when competing proteins are present in complex amyloid polymerization reactions (Biancalana and Koide, 2010). Therefore, previously established ThT-based seeding assays are relatively insensitive when complex biosamples such as brain homogenates are analyzed (Gupta et al., 2012).

To overcome these limitations, the present inventors have developed a sensitive FRET-based biosensor assay, which in the context of HD has been termed FRET-based mHTT aggregate seeding (FRASE) assay, which does not require ThT reporter molecules for the quantification of HSA in biosamples. Two fluorescently tagged aggregation-prone fusion proteins are used as reporter molecules to monitor seeding activity. This assay is highly robust and affected by contaminating proteins in complex biosamples only to a very small extent. Therefore, the assay can be employed without the need for upstream purification of seeds, which would complicate the protocol and decrease accuracy of quantification. For example, the present inventors surprisingly found that the FRASE assay enables the quantification of mHTT seeding activity (HSA) in complex biosamples.

Accordingly, the present invention provides a method for the quantification of seeding activity (At ₅ o) of an amyloidogenic aggregate, comprising the steps of:

(i) providing, in a solution, a mixture of an amyloidogenic protein A which is N- terminally or C-terminally fused to a donor fluorophore molecule, e.g. a cyan fluorescent protein (CFP), and an amyloidogenic protein B which is N- terminally or C-terminally fused to an acceptor fluorophore molecule, e.g. a yellow fluorescent protein (YFP), wherein the amyloidogenic proteins A and B are preferably identical and wherein the donor fluorophore molecule and the acceptor fluorophore molecule are capable of Forster Resonance Energy Transfer (FRET) if they are in close proximity to each other;

(ii) adding a sample containing an amyloidogenic protein aggregate C to the mixture of step (i), wherein the amyloidogenic protein aggregate C preferably comprises or consists of amyloidogenic proteins A and/or B;

(iii) shaking the mixture of step (ii);

(iv) measuring fluorescence signals in the donor, e.g. cyan, channel, the acceptor, e.g. yellow, channel and the Forster Resonance Energy Transfer (FRET) channel at predetermined intervals after completion of step (iii);

(v) calculating FRET efficiency (E) from the signals obtained in step (iv); and

(vi) quantifying seeding activity (At ₅₀ ) by subtracting the time at half-maximal FRET efficiency of a sample (t ₅₀ (S))from the time at half-maximal FRET efficiency of a negative control (t ₅₀ (0)).

Amyloidogenic proteins are proteins that undergo non-native cross- -assembly to form linear polymers (fibrils), which is a central feature of diseases of toxic misfolding. In the context of the present invention, an amyloidogenic protein is a protein which can, in vivo or in vitro, self-assemble into amyloid fibrils. A particular group of amyloidogenic proteins, from which proteins A and B may be selected from, are wild-type or mutant forms of the group consisting of proteins comprising repeats of at least 15 consecutive glutamine residues in their amino acid sequence, and amyloidogenic fragments thereof. These proteins are also referred to as polyQ proteins.

Specific examples of polyQ proteins according to the invention are wild -type or mutant forms of the group consisting of huntingtin (HTT), androgen receptor (AR), atrophin 1 (ATN1 ), ataxin 1 (ATXN1 ), ataxin 2 (ATXN1 ), ataxin 3 (ATXN1 ), ataxin 7 (ATXN1 ), TATA-box binding protein (TBP), a-i _A -voltage dependent calcium channel subunit (CACNA1A), and polyglutamine repeat containing fragments thereof. An exemplary polyQ containing fragment of HTT is exon 1 .

A“mutant form” in the context of the invention is particularly a protein that contains at least one, e.g. from 1 to 100 such as from 1 to 20, insertions, deletions (such as a truncation) or substitutions of amino acids in its primary sequence that increase its amyloidogenic properties, e.g. leads to facilitated amyloid formation. In particular, mutant forms of amyloidogenic proteins, such as polyQ proteins, are characterized by an increased number of glutamine residues as compared to the corresponding wild-type form, i.e. additional glutamine residues are inserted into the wild-type sequence. The number of glutamine residues may be increased by e.g. 5-75 glutamines.

In another embodiment, mutant forms of amyloidogenic proteins, such as polyQ proteins, are characterized by an increased number of glutamine residues and proline residues as compared to the corresponding wild-type form, i.e. additional glutamine and proline residues are inserted into the wild-type sequence. The number of glutamine residues may be increased by e.g. 5-75 glutamines. The number of proline residues may be increased by e.g. 1 -20 prolines. Additionally, these mutant forms may optionally have 1 -75 deletions (such as a truncation) of amino acids in their primary sequence that increase the amyloidogenic properties, e.g. lead to facilitated amyloid formation. Such truncation may e.g. concern the N17 region and/or the proline-rich domain (PRD) region of a polyQ protein, such as a HTT protein. In a preferred embodiment, the truncation concerns the N17 region of a polyQ protein, such as a HTT protein. In another preferred embodiment, the truncation concerns the N17 region and the PRD region of a polyQ protein, such as a HTT protein.

Other exemplary amyloidogenic proteins according to the invention are wild- type or mutant forms of the group consisting of Tau protein, TAR DNA-binding protein 43 (TDP-43), Fused in Sarcoma (FUS) protein, Suppressor 35 (SUP35) protein, alpha-synuclein (a-Syn), and amyloidogenic fragments thereof.

Huntingtin is a large cytoplasmic protein (348 kDa in humans) that is widely expressed in the body. The first exon (exon 1 ) of the HTT gene contains a naturally polymorphic CAG repeat, leading to variable numbers of glutamine residues in the huntingtin protein. The amino acid sequence of wild-type human huntingtin (with 21 consecutive glutamine residues in exon 1 ), accessible via UniProtKB (accession no. P42858) is shown in SEQ ID NO.:1. A number of up to 35 glutamine residues (within exon 1 , which is always referred to herein with regard to polyQ repeats unless indicated otherwise) is considered wild -type. Conversely, HTT with 36 glutamine residues or more in exon 1 is classified as mutant huntingtin (mHTT) herein. In certain preferred embodiments of the invention, the amyloidogenic proteins A and B are mutant huntingtin (mHTT), particularly N-terminal fragments of mHTT, more particularly N-terminal fragments of mHTT comprising mHTT exon 1 , more particularly N-terminal fragments of mHTT consisting of mHTT exon 1 . In the context of the invention, mHTT exon 1 is also referred to as mHTTexl . Exemplary nucleotide and amino acid sequences of human wild -type exon 1 with 23 consecutive glutamine residues (Ex1 Q23) and mutant huntingtin exon 1 with 48 or 49 consecutive glutamine residues (Ex1 Q48, Ex1Q49) are shown in SEQ ID NOs.: 2-5. Further mutant huntingtin exon 1 sequences according to the invention with 48 or 40 consecutive glutamine residues and an adjacent modified proline rich domain (PRD) comprising 6 or more consecutive proline residues are e.g. K2Q48P6 (SEQ ID NO: 71 ), AN17Q48+6PRD (SEQ ID NO: 72) and AN17Q40+6PRD (SEQ ID NO: 73). Of particular interest in the present invention are amyloidogenic proteins of human origin and of genetic model organisms, e.g. Caenorhabditis elegans, fruit fly, zebrafish and Xenopus laevis. According to some embodiments, the amyloidogenic proteins A and B are Homo sapiens proteins or Drosophila melanogaster proteins.

The amyloidogenic proteins A and B may be different proteins or may be isoforms of the same protein, but preferably share a high degree of amino acid sequence identity, e.g. at least 80%, at least 90%, at least 95%, or at least 98%. Particularly, amyloidogenic proteins A and B have an amino acid sequence identity of at least 95%, more particularly at least 98%. In some embodiments, the amino acid sequence identity of proteins A and B is 100%, i.e. amyloidogenic proteins A and B are identical.

The percent sequence identity may be determined according to the following formula:

I = n : L

wherein I is the identity in percent, n is the number of identical amino acids between a given sequence and a comparative sequence, and L is the length of the comparative sequence. Importantly, when calculating the percent sequence identity according to this formula, an alignment of the two sequences shall be carried out without gaps between complementary portions and over the whole length of the comparative sequence.

Each of the amyloidogenic proteins A and B fused to the fluorophore may be prepared synthetically, e.g. by solid phase peptide synthesis, or it may be prepared recombinantly, e.g. from expression of suitable vectors in bacteria, yeast, or expression cell lines known to the skilled person. Accordingly, in preferred embodiments, each of the amyloidogen ic proteins A and B is independently a synthetically produced protein or a recombinantly produced protein.

The amyloidogenic protein aggregate C by definition comprises at least one amyloidogenic protein, which has or which have self-aggregated. For example, protein aggregate C may comprise mutant huntingtin. In particular embodiments, aggregate C comprises amyloidogenic proteins A and/or B. In other embodiments, aggregate C consists of amyloidogenic proteins A and/or B, i.e. it is constituted exclusively by protein A or by protein B or by a mixture of both protein A and B.

The aggregate may be formed in the presence of one or more other compounds, e.g. chaperones, which may still be present in the aggregate C used in the inventive method. Also, it is possible that preformed aggregate C has subsequently (i.e. after formation) been brought into contact with a further compound, which may for instance positively or negatively influence its seeding properties.

The protein aggregate C may be synthetically produced or recombinantly produced, e.g. from synthetic or recombinantly expressed proteins A and/or B. In some embodiments, the synthetically or recombinantly produced aggregate C may optionally be sonicated before being added to the mixture of step (i) as defined above.

Alternatively, it is possible that protein aggregate C is present in a sample, e.g. a biological sample. Such samples can be tissue samples or body fluid samples or culture samples, e.g. cell culture samples. Any of these samples may optionally be pretreated. Pretreatment can be effected by at least one of centrifugation, cell lysis, protein extraction, dialysis, sonication and similar procedures used in protein purification known to the skilled person. Accordingly, the sample containing aggregate C may be selected from the group consisting of an optionally pretreated tissue sample, an optionally pretreated body fluid sample and an optionally pretreated cell culture sample.

In exemplary embodiments, the sample is pretreated. Such a pretreated tissue, body fluid or cell culture sample may be selected from the group consisting of a homogenate, an extract, a pellet and a lysate. As indicated above, the pretreated sample may, in some embodiments and in addition to previous pretreatment steps, be sonicated. A “homogenate” particularly is a cell suspension or animal tissue that has been ground in an all-glass homogenizer (douncer) to disrupt cells. An“extract” is particularly a cell suspension or animal tissue, wherein cells have been chemically or physically lysed, and the cell debris has subsequently been removed, e.g. by centrifugation (e.g. at 30,000 g or 100,000 g). A“pellet” may be obtained, e.g., by centrifugation of a cell suspension etc., which is then washed with suitable buffers or further purified, as is known to the skilled person. A“lysate” is the material that remains when cells are lysed by enzymes, inorganic chemicals, or physical means.

In some preferred embodiments, the tissue used in the inventive method is muscle tissue. Exemplary pretreated tissue samples are selected from the group consisting of brain homogenates, brain extracts, and protein extracts from post-mortem tissue such as cerebral cortex, caudate nucleus and cerebellum.

Body fluids suitable for use in the inventive method are, inter alia, blood, preferably full blood, or cerebrospinal fluid. An exemplary pretreated body fluid sample is blood plasma.

Suitable cell culture samples according to the invention include cell line samples, e.g. from a fibroblast cell line or iPSC-derived neurons, stem cell samples, e.g. induced pluripotent stem cells (iPSC), and primary cell culture samples. The sources of primary cultures comprise excised animal tissue that is cultured either as an explant culture, suspension or as monolayer and maintained in vitro. The excised tissue is subjected to enzyme treatment and the dissociated cells are cultured under the appropriate conditions in culture medium until they reach adequate numbers.

The solution in which the mixture of amyloidogenic proteins A and B is provided preferably is a buffer, e.g. an aggregation buffer. For example, an aggregation buffer may be made up of a buffering agent (e.g. Tris-HCI, HEPES-KOH, etc.), particularly at a pH of from 7 to 8 and concentration of 25-100 mM, at least one salt (e.g. NaCI, (NH ₄ ) ₂ S0 ₄ , MgCI ₂ , KCI), preferably at a total concentration of 50-200 mM, a chelator (e.g. EDTA) and a reducing agent (e.g. b- mercaptoethanol or dithiothreitol (DTT)). A specific example of a buffer for use according to the invention consists of an aqueous solution of 50 mM Tris-HCI pH 7.4, 150 mM NaCI, 1 mM EDTA and 1 mM DTT.

Within the context of the above-described method for the quantification of seeding activity (Dΐ ₅ o) of an amyloidogenic aggregate, it is also encompassed by the invention that the preformed aggregate may be used to screen for compounds that influence seeding competence of the aggregate. For example, a given amyloidogenic aggregate C may be pretreated with a compound of interest (D) before being used in the described screening method. Then, seeding activity is determined once with a sample containing aggregate C without pretreatment, and once with a sample containing aggregate C pretreated with the compound D. Compound D may be e.g. a protein, peptide or small molecule as defined hereinbelow.

Alternatively, a given amyloidogenic aggregate C may be formed in the presence of a compound D. Then, seeding activity is determined once with a sample containing aggregate C formed in the absence of compound D, and once with a sample containing aggregate C formed in the presence of compound D. Compound D may be e.g. a protein, peptide or small molecule as defined hereinbelow.

Accordingly, the invention provides a method for identifying compounds that inhibit seeding activity of amyloidogenic aggregates, comprising the steps of:

(i) quantifying seeding activity (At ₅ o) in a first test sample, comprising the steps of:

(a) providing, in a solution, a mixture of an amyloidogenic protein A which is N-terminally or C-terminally fused to a donor fluorophore molecule, e.g. a CFP, and an amyloidogenic protein B which is N- terminally or C-terminally fused to an acceptor fluorophore molecule, e.g. a YFP, wherein the amyloidogenic proteins A and B are preferably identical and wherein the donor fluorophore molecule and the acceptor fluorophore molecule are capable of FRET if they are in close proximity to each other; (b) adding a sample containing an amyloidogenic protein aggregate C to the mixture of step (a), wherein the amyloidogenic protein aggregate C preferably comprises or consists of amyloidogenic proteins A and/or B;

(c) shaking the mixture of step (b);

(d) measuring fluorescence signals in the donor, e.g. cyan, channel, the acceptor, e.g. yellow, channel and the FRET channel at predetermined intervals after completion of step (c);

(e) calculating FRET efficiency (E) from the signals obtained in step (d); and

(f) quantifying seeding activity (At ₅₀ ) by subtracting the time at half- maximal FRET efficiency of a sample (t ₅₀ (S))from the time at half- maximal FRET efficiency of a negative control (t ₅₀ (0));

(ii) quantifying seeding activity (At ₅ o) in a second test sample, comprising the steps (a) to (f) as described in (i), wherein the aggregate C further has been pretreated with a compound D or has been formed in the presence of a compound D;

(iii) correlating that the compound of interest has an inhibitory effect on the seeding activity of aggregate C, if the seeding activity obtained in step (ii) is lower than the seeding activity obtained in step (i).

Preferably, the amyloidogenic proteins A and B are mutant huntingtin (mFITT), particularly N-terminal fragments of mFITT, more particularly N-terminal fragments of mFITT comprising mFITT exon 1 , more particularly N-terminal fragments of mFITT consisting of mFITT exon 1. In the context of the invention, mFITT exon 1 is also referred to as mFITTexl . Exemplary nucleotide and amino acid sequences of human wild-type exon 1 with 23 consecutive glutamine residues (Ex1Q23) and mutant huntingtin exon 1 with 48 or 49 consecutive glutamine residues (Ex1 Q48, Ex1Q49) are shown in SEQ ID NOs.: 2-5. Further mutant huntingtin exon 1 sequences according to the invention with 48 or 40 consecutive glutamine residues and an adjacent modified proline rich domain (PRD) comprising 6 or more consecutive proline residues are e.g. K2Q48P6 (SEQ ID NO: 71 ), AN17Q48+6PRD (SEQ ID NO: 72) and AN17Q40+6PRD (SEQ ID NO: 73). The present invention further provides a method for assessing the risk for development of a polyglutamine (polyQ) disease in a subject, comprising

(i) quantification of seeding activity (Dΐ ₅₀ ) of a polyQ-containing protein in a sample, comprising the steps of:

(a) providing, in a solution, a mixture of a first, preferably recombinant, polyQ-containing protein which is C-terminally fused to a donor fluorophore molecule, e.g. a CFP, and a second, preferably recombinant, polyQ-containing protein which is C-terminally fused to an acceptor fluorophore molecule, e.g. a YFP, wherein the first and second proteins are the same polyQ-containing protein and wherein the donor fluorophore molecule and the acceptor fluorophore molecule are capable of FRET if they are in close proximity to each other;

(b) adding a sample collected from the subject to the mixture of step (a);

(c) shaking the mixture of step (b);

(d) measuring fluorescence signals in the donor, e.g. cyan, channel, the acceptor, e.g. yellow, channel and the FRET channel at predetermined intervals after completion of step (c);

(e) calculating FRET efficiency (E) from the signals obtained in step (d); and

(f) quantifying seeding activity (Dί ₅₀ ) by subtracting the time at half- maximal FRET efficiency of a sample (t ₅₀ (S)) from the time at half- maximal FRET efficiency of a negative control (t ₅₀ (0)); and

(ii) correlating that the subject is at risk for development of the polyQ disease when the seeding activity in the sample is increased as compared to a reference sample.

In some embodiments, a Dί ₅₀ of the collected sample which is significantly higher than Dΐ ₅₀ of a reference sample indicates that the subject is at risk for developing a polyQ disease.

Typically, At ₅₀ of the reference sample will be about 0 with standard error of the mean (SEM). Dί ₅₀ of the collected sample to be analyzed may be of about the same value (i.e. about 0±SEM) if the subject is not at risk for developing the polyQ disease of interest. However, if the subject is at risk for developi ng the polyQ disease of interest, Dΐ ₅₀ of the collected sample to be analyzed may be significantly higher.

Of course, it should be verified that the reference sample is obtained from a source which is not at risk of developing the polyQ disease of interest (i.e., a healthy subject or cell culture).

The present invention further provides a method for predicting the onset of a polyglutamine (polyQ) disease in a subject, comprising

(i) quantification of seeding activity (Dΐ ₅₀ ) of a polyQ-containing protein in a sample, comprising the steps of:

(a) providing, in a solution, a mixture of a first, preferably recombinant, polyQ-containing protein which is C-terminally fused to a donor fluorophore molecule, e.g. a CFP, and a second , preferably recombinant, polyQ-containing protein which is C-terminally fused to an acceptor fluorophore molecule, e.g. a YFP, wherein the first and second proteins are the same polyQ-containing protein and wherein the donor fluorophore molecule and the acceptor fluorophore molecule are capable of FRET if they are in close proximity to each other;

(b) adding a sample collected from the subject to the mixture of step (a);

(c) shaking the mixture of step (b);

(d) measuring fluorescence signals in the donor, e.g. cyan, channel, the acceptor, e.g. yellow, channel and the FRET channel at predetermined intervals after completion of step (c);

(e) calculating FRET efficiency (E) from the signals obtained in step (d); and

(f) quantifying seeding activity (Dί ₅₀ ) by subtracting the time at half- maximal FRET efficiency of a sample (t ₅₀ (S)) from the time at half- maximal FRET efficiency of a negative control (t ₅₀ (0)); and (ii) correlating that the onset of the polyQ disease has occurred or will occur soon when the seeding activity in the sample is increased as compared to a reference sample.

In some embodiments, a At ₅₀ of the collected sample which is significantly higher than At ₅₀ of a reference sample indicates that onset of the analyzed polyQ disease will occur soon (i.e. is imminent). In other embodiments, a Dί ₅₀ of the collected sample which is significantly higher than Dΐ ₅₀ of a reference sample indicates that onset of the analyzed polyQ disease has already occurred.

Typically, Dί ₅₀ of the reference sample will be from about 0 to about 1. Dΐ ₅₀ of the collected sample to be analyzed may be of about the same value (i.e. about 0±SEM to about 1 ±SEM) if there is no imminent onset of the polyQ disease of interest. If onset of the polyQ disease of interest will occur soon (e.g. within a time interval of between 3 and 9 months), Dΐ ₅₀ of the collected sample to be analyzed may be at least about 2, e.g. from about 2 to about 3, particularly 2.5 to 3.5. If onset of the polyQ disease of interest has occurred already, Dί ₅ o of the collected sample to be analyzed may be at least about 4, e.g. from about 4 to about 6, particularly 3.6 to 6.0. In the context of the invention, the“onset” of a disease is the timepoint, from which onwards established clinical signs or symptoms can be detected.

Of course, it should be verified that the reference sample is obtained from a source where there is at least no imminent onset of the polyQ disease of interest, or even better, which is not at all developing the polyQ disease of interest (i.e., a healthy subject or cell culture).

In general, the reference sample in the above-described method for assessing the risk for development of a polyQ disease and method for predicting the onset of a polyglutamine (polyQ) disease should be a sample that is closely related to the sample collected from the subject to be analyzed; typically, it may be derived from the same organism and from the same tissue or a cell culture of the respective tissue or the same body fluid as the sample of step (i)(b). If the sample of step (i)(b) is pretreated, the reference sample should be subjected to the same pretreatment. Seeding activity (Dΐ ₅₀ ) is quantified in the reference sample according to steps (i)(a)-(f), using the reference sample instead of the sample collected from the subject to be analyzed. The obtained value can be compared with the seeding activity (Dΐ ₅₀ ) value of the sample collected from the subject to be analyzed.

Alternatively, it is possible to rely on known values of reference samples (of corresponding origin) obtained according to a standardized procedure analogous to the method steps (i)(a)-(f) described above.

The present invention further provides a method for assessing the progression of a polyglutamine (polyQ) disease in a subject, comprising

(i) quantification of seeding activity (Dί ₅ o) of a polyQ-containing protein in a sample collected at a timepoint t-i, comprising the steps of:

(a) providing, in a solution, a mixture of a first, preferably recombinant, polyQ-containing protein which is C-terminally fused to a donor fluorophore molecule, e.g. a CFP, and a second , preferably recombinant, polyQ-containing protein which is C-terminally fused to an acceptor fluorophore molecule, e.g. a YFP, wherein the first and second proteins are the same polyQ-containing protein and wherein the donor fluorophore molecule and the acceptor fluorophore molecule are capable of FRET if they are in close proximity to each other;

(b) adding a sample collected from the subject to the mixture of step (a);

(c) shaking the mixture of step (b);

(d) measuring fluorescence signals in the donor, e.g. cyan, channel, the acceptor, e.g. yellow, channel and the FRET channel at predetermined intervals after completion of step (c);

(e) calculating FRET efficiency (E) from the signals obtained in step (d);

(f) quantifying seeding activity (Dί ₅₀ ) by subtracting the time at half- maximal FRET efficiency of a sample (t ₅₀ (S)) from the time at half- maximal FRET efficiency of a negative control (t ₅₀ (0)); (ii) quantification of seeding activity (At ₅ o) of a polyQ-containing protein in a sample collected at a timepoint t ₂ , wherein t ₂ is later than t-i, comprising the steps (a) to (f) as defined in (i); and

(iii) correlating that the polyQ disease has progressed when the seeding activity in the sample taken at t ₂ is increased as compared to the sample taken at t-i.

PolyQ diseases according to the invention are particularly selected from the group consisting of Huntington’s disease (HD), Machado-Joseph disease (MJD/SCA3), dentatorubral pallidoluysian atrophy (DRPLA), spinocerebellar ataxia (SCA) type 1 , SCA type 2, SCA type 6, SCA type 7, SCA type 17 and spinal and bulbar muscular atrophy, X-linked 1 (SMAX1/SBMA).

A polyQ disease of particular interest in preferred embodiments of the invention is Huntington’s disease (HD).

In the above-described methods for assessing the risk for development, predicting the onset or assessing the progression of a polyQ disease, the polyQ protein of the respective step (i)(a) may be a synthetically produced or recombinant protein. Preferably, it is recombinantly produced. The polyQ protein is particularly selected from wild-type or mutant forms of the group consisting of huntingtin (HTT), androgen receptor (AR), atrophin 1 (ATN1 ), ataxin 1 (ATXN1 ), ataxin 2 (ATXN1 ), ataxin 3 (ATXN1 ), ataxin 7 (ATXN1 ), TATA-box binding protein (TBP), a-i _A -voltage dependent calcium channel subunit (CACNA1A), and polyglutamine repeat containing fragments thereof. In certain preferred embodiments, the polyQ protein is mutant huntingtin (mHTT) or an N-terminal fragment thereof, characterized by an increased number of glutamine residues as compared to the corresponding wild-type form. More preferably, the polyQ protein is selected from N-terminal fragments of mHTT comprising mHTT exon 1 , particularly N-terminal fragments of mHTT consisting of mHTT exon 1 (mHTTexl ). Specific examples of mHTTexl according to the invention include mHTT Ex1 Q48 (SEQ ID NO.: 2), Ex1 Q49 (SEQ ID NO.: 4), K2Q48P6 (SEQ ID NO: 71 ), AN17Q48+6PRD (SEQ ID NO: 72) and AN17Q40+6PRD (SEQ ID NO: 73). In the above-described methods for assessing the risk for development, predicting the onset or assessing the progression of a polyQ disease, the subject may be a mammalian subject, particularly a subject selected from the group consisting of mouse, rat, monkey, sheep, pig and human. In preferred embodiments, the subject is a human. The subject may also be a non- mammalian model organism for studying a polyQ disease, e.g. a model organism used to study HD. In particular, the model organism may be selected from the group consisting of yeast, Caenorhabditis elegans, fruit fly, zebrafish and Xenopus laevis. Alternatively, the subject is a population of cells from cell culture. The cell population may, for example, be stem cells, including pluripotent stem cells, such as induced pluripotent stem cells (iPSC), fibroblasts, or iPSC-derived neurons. Also, cells from established cell lines or primary cell culture may be suitable.

The sample used in step (i)(b) of the above-described methods for assessing the risk for development, predicting the onset or assessing the progression of a polyQ disease may be obtained from a variety of tissues, body fluids, whole organisms and cultured cells from a number of different organisms. The sample may be collected from a living organism, or it may be collected post mortem.

In some embodiments, the sample is a tissue sample, which is optionally pretreated as described above. For example, the sample may be selected from the group consisting of muscle tissue, protein extracts from muscle tissue, homogenated muscle tissue, nasal brushing and nasal epithelial tissue. Among samples that can only be obtained post mortem are mammalian, e.g. human, brain samples. In some embodiments, the sample is a human brain sample.

In other embodiments, the sample is a body fluid sample. Suitable body fluids include blood, sputum and cerebrospinal fluid. For example, the sample may be selected from the group consisting of full blood, homogenated blood, blood plasma and cerebrospinal fluid. The solution in which the mixture of the first and second polyQ proteins is provided preferably is a buffer, e.g. an aggregation buffer. For example, an aggregation buffer may be made up of a buffering agent (e.g. Tris-HCI, HEPES- KOH, etc.), particularly at a pH of from 7 to 8 and concentration of 25-100 mM, at least one salt (e.g. NaCI, (NH ₄ ) ₂ S0 ₄ , MgCI ₂ , KCI), preferably at a total concentration of 50-200 mM, a chelator (e.g. EDTA) and a reducing agent (e.g. b-mercaptoethanol or dithiothreitol (DTT)). A specific example of a buffer for use according to the invention consists of an aqueous solution of 50 mM Tris-HCI pH 7.4, 150 mM NaCI, 1 mM EDTA and 1 mM DTT.

The present invention further provides a method for identifying compounds that inhibit mHTT seeding activity (HSA) in vitro, comprising the steps of:

(i) determining HSA in a test sample, comprising the steps of:

(a) providing, in a solution, a mixture of a purified, preferably recombinant, mutant form of an N-terminal huntingtin fragment comprising exon 1 (mHTTexl ) characterized by an increased number of glutamine residues as compared to the corresponding wild-type form, which is (1 ) N-terminally fused to a globular peptide comprising a protease recognition sequence at its C-terminus, and (2) C-terminally fused to a donor fluorophore molecule, e.g. a CFP, and a purified, preferably recombinant, mutant form of an N-terminal huntingtin fragment comprising exon 1 (mHTTexl ) characterized by an increased number of glutamine residues as compared to the corresponding wild-type form, which is (1 ) N-terminally fused to a globular peptide comprising a protease recognition sequence at its C-terminus, and (2) C-terminally fused to an acceptor fluorophore molecule, e.g. a YFP;

(b) adding a compound or a mixture of compounds to be tested for inhibiting HSA in vitro to the mixture of step (a);

(c) adding a protease specifically recognizing the protease recognition sequence within the globular peptide to the mixture of step (b);

(d) optionally adding preformed aggregates of mHTT or N-terminal fragments thereof to the mixture of step (c);

(e) shaking the mixture of step (c) or step (d); (f) measuring fluorescence signals in the donor, e.g. cyan, channel, the acceptor, e.g. yellow, channel and the FRET channel at predetermined intervals after completion of step (e);

(g) calculating FRET efficiency (E) from the signals obtained in step (f); and

(h) quantifying seeding activity (Dΐ ₅₀ ) by subtracting the time at half- maximal FRET efficiency of a sample (t ₅₀ (S)) from the time at half- maximal FRET efficiency of a negative control (tso(0));

(ii) determining FISA in a control sample without the compound or mixture of compounds to be tested for inhibiting FISA in vitro, comprising steps (a), (c), optionally (d), (e), (f), (g) and (h) as defined in (i); and

(iii) selecting compounds or mixtures of compounds which show decreased seeding activity as compared to control samples.

With the above-described method, it is possible to screen for compounds directly targeting mFITT seeding in vitro, which may have therapeutic potential. A“compound” in the context of this screening method includes small molecules (i.e. naturally occurring, modified or synthetic organic molecules with a molecular weight of about 900 Da or less) as well as peptides (up to 100 amino acids) and proteins (more than 100 amino acids), which may be further modified, e.g. by attaching moieties like polyethylene glycol (PEG) to increase half-life. It is also possible to test a mixture of compounds, e.g. of two or three compounds, which only in combination inhibit mFITT seeding activity in vitro.

The fusion protein comprising the donor fluorophore molecule and the fusion protein comprising the acceptor fluorophore molecule comprise, each independently, a purified mutant form of an N-terminal huntingtin fragment comprising exon 1 (mFITTexl ). This mutant form is characterized by an increased number of glutamine residues as compared to the corresponding wild-type form. In certain embodiments, the N-terminal huntingtin fragment consists of exon 1 (mFITTexl ). The N-terminal huntingtin fragment may have, e.g., from 35 to 75 glutamines in exon 1 (mFITT Ex1 Q35 - mFITT Ex1 Q75). Specific examples of mFITTexl for use in this screening method include mFITT Ex1 Q48 (SEQ ID NO.: 2), Ex1 Q49 (SEQ ID NO.: 4), K2Q48P6 (ID NO: 71 ), AN17Q48+6PRD (SEQ ID NO: 72) and AN17Q40+6PRD (SEQ ID NO: 73). The N-terminal huntingtin fragments of step (i)(a) of the above-described screening method are N-temninally fused to a globular peptide. The globular peptide comprises a protease recognition sequence at its C-terminus. Suitable globular peptide tags include glutathione-S-transferase (GST) and maltose binding protein (MBP). In preferred embodiments, the globular peptide is GST, particularly GST as shown in SEQ ID NO.: 6. Suitable protease recognition sequences are known to the skilled person. An exemplary protease recognition sequence, which is recognized by the commercially available PreScission ^® Protease (or PSP; GE Healthcare), is LEVLFQGP (SEQ ID NO.: 8).

Further, the N-terminal huntingtin fragments of step (i)(a) of the above- described screening method is C-terminally fused to a donor fluorophore molecule and an acceptor fluorophore molecule, respectively. The fluorophore molecules are selected from suitable FRET pairs, for instance those described below.

The protease used for cleaving off the globular protein at the N-terminus of the mHTT fragments (step (i)(c) of the above-described screening method), starts the aggregation reaction. It will be selected according to the used protease recognition sequence. A number of recognition sequence/protease combinations are known to the skilled person. For example, when the protease recognition sequence LEVLFQGP (SEQ ID NO.: 8) is used, a suitable protease is PreScission ^® Protease (or PSP; GE Healthcare).

The preformed aggregate of step (i)(d) of the above -described screening method may optionally be added to the reaction as an aggregation seed. Such a seed can increase aggregation rate of the mHTT fragments fused to the fluorophore molecules. The preformed aggregate may be obtained from various sources. According to some embodiments the preformed aggregate may be obtained ex vivo, e.g. from a subject suffering from HD or from a HD model. According to other embodiments, it may be obtained in vitro, e.g. from recombinantly expressed or chemically synthesized and subsequently aggregated mHTT or N-terminal fragment thereof. When obtained in vitro, the aggregated mHTT or N -terminal fragment thereof is characterized by an increased number of glutamine residues as compared to the corresponding wild-type form. The aggregate of N-terminal huntingtin fragment may consist of exon 1 (mHTTexl ). In certain preferred embodiments, the preformed aggregates of mHTT or N-terminal fragment thereof consist of mHTT exon 1 having from 35 to 75 glutamines in exon 1 (Ex1 Q35 - Ex1 Q75). Specific examples of mHTTexl aggregates for use in this screening method include aggregates consisting of Ex1 Q48 (SEQ ID NO.: 2), Ex1 Q49 (SEQ ID NO.: 4), K2Q48P6 (SEQ ID NO.: 71 ), AN17Q48+6PRD (SEQ ID NO.: 72) or AN17Q40+6PRD (SEQ ID NO.: 73). It is possible to use a similar recombinant construct, but without fusion to fluorophore molecules, for obtaining the preformed aggregates. For example, the preformed aggregates of mHTT or an N-terminal fragment thereof may be obtained by adding a suitable protease to purified mHTT GST-Ex1 Q48 (SEQ ID NO.: 9), purified mHTT GST-Ex1 Q49 (SEQ ID NO.: 11 ), purified GST-K2Q48P6 (SEQ ID NO: 60), purified GST- AN17Q48+6PRD (SEQ ID NO: 62) or purified GST-AN17Q40+6PRD (SEQ ID NO: 64). As mentioned above, selection of a suitable protease depends on the protease recognition sequence in the fusion protein. A number of recognition sequence/protease combinations are known to the skilled person. For example, when the protease recognition sequence LEVLFQGP (SEQ ID NO.: 8) is used, a suitable protease is PreScission ^® Protease (or PSP; GE Healthcare).

The solution in which the mixture of the purified mutant N-terminal huntingtin fragments comprising exon 1 (mHTTexl ) fused to the FRET pair fluorophores is provided preferably is a buffer, e.g. an aggregation buffer. For example, an aggregation buffer may be made up of a buffering agent (e.g. Tris-HCI, HEPES- KOH, etc.), particularly at a pH of from 7 to 8 and concentration of 25-100 mM, at least one salt (e.g. NaCI, (NH ₄ ) ₂ S0 ₄ , MgCI ₂ , KCI), preferably at a total concentration of 50-200 mM, a chelator (e.g. EDTA) and a reducing agent (e.g. b-mercaptoethanol or dithiothreitol (DTT)). A specific example of a buffer for use according to the invention consists of an aqueous solution of 50 mM Tris-HCI pH 7.4, 150 mM NaCI, 1 mM EDTA and 1 mM DTT. In all methods described herein, advantage is taken of the capability of certain fluorophore molecules to effect an energy transfer. This phenomenon is termed Forster or fluorescence resonance energy transfer (FRET) and is characterized in that a donor fluorophore molecule is excited, and in its excited state transfers (non-radiatively) excitation energy to a fluorophore molecule in close proximity. Close proximity is usually given when the distance between donor and acceptor fluorophore is in the range of 1 -10 nm. A number of optical methods is known to measure FRET and to calculate FRET efficiency therefrom.

The donor and acceptor molecules capable of FRET are also termed FRET pairs. Nowadays, a variety of FRET pairs is available. Within the scope of the invention, it is encompassed to use, e.g., blue/yellow FRET pairs, green/red FRET pairs, and far-red/infra-red FRET pairs. Specific examples of FRET pairs that may in principle be used are ECFP/EYFP, mTurquoise2/sEYFP, SCFP/SYFP, CyPet/YPet, ECFP/mVenus, mCerulean/mCitrine, EGFP/mCherry, mNeonGreen/mRuby3, EGFP/mRFP1 , mCerulean/mVenus and mTurquoise/mVenus.

Particularly, the acceptor fluorophore molecule according to the present invention is a yellow fluorescent protein (YFP). A number of YFPs are presently known, and it is encompassed by the invention that the YFP may, e.g., be selected from the group consisting of EYFP, Venus (e.g. mVenus), Citrine (e.g. mCitrine) and YPet. A preferred YFP according to the invention is YPet. A specific example of YPet used according to the invention is shown in SEQ ID NO.: 21.

The donor fluorophore molecule according to the invention is particularly a cyan fluorescent protein (CFP), which may, e.g., be selected from the group consisting of ECFP, SCFP, Cerulean (e.g. Cerulean, mCerulean2, mCerulean3), Turquoise (e.g. mTurquoise, mTurquoise2) and CyPet. A preferred CFP according to the invention is CyPet. A specific example of CyPet used according to the invention is shown in SEQ ID NO.: 23. According to some embodiments of the invention, the donor fluorophore molecule is a CFP, which is preferably selected from the group consisting of ECFP, SCFP and CyPet, more preferably CyPet, and the acceptor fluorophore molecule is a YFP, which is preferably selected from the group consisting of EYFP, Venus, Citrine and YPet, more preferably YPet. In certain preferred embodiments of the invention, CyPet/YPet is used as the FRET pair.

Thus, in some preferred embodiments of the method for the quantification of seeding activity (At ₅ o) of an amyloidogenic aggregate described herein, the amyloidogenic proteins A and B are mFITT Ex1 Q48-YPet (SEQ ID NO.: 13) and mHTT Ex1 Q48-CyPet (SEQ ID NO.: 15).

In some preferred embodiments of the method for the quantification of seeding activity (At ₅ o) of an amyloidogenic aggregate described herein, the amyloidogenic proteins A and B are K2Q48P6-YPet (SEQ ID NO.: 66) and K2Q48P6-CyPet (SEQ ID NO.: 65).

In some preferred embodiments of the method for the quantification of seeding activity (At ₅ o) of an amyloidogenic aggregate described herein, the amyloidogenic proteins A and B are AN17Q48+6PRD-YPet (SEQ ID NO.: 68) and DN17Q48+6PRD-CyPet (SEQ ID NO.: 67).

In some preferred embodiments of the method for the quantification of seeding activity (At ₅ o) of an amyloidogenic aggregate described herein, the amyloidogenic proteins A and B are AN17Q40+6PRD-YPet (SEQ ID NO.: 70) and DN17Q40+6PRD-CyPet (SEQ ID NO.: 69).

Likewise, in some preferred embodiments of the methods for assessing the risk for development, predicting the onset or assessing the progression of a polyQ disease, the first and second recombinant polyQ-containing proteins of step (a) are mHTT Ex1 Q48-YPet (SEQ ID NO.: 13) and mHTT Ex1 Q48-CyPet (SEQ ID NO.: 15). In some preferred embodiments of the methods for assessing the risk for development, predicting the onset or assessing the progression of a polyQ disease, the first and second recombinant polyQ-containing proteins of step (a) are K2Q48P6-YPet (SEQ ID NO.: 66) and K2Q48P6-CyPet (SEQ ID NO.: 65).

In some preferred embodiments of the methods for assessing the risk for development, predicting the onset or assessing the progression of a polyQ disease, the first and second recombinant polyQ-containing proteins of step (a) are AN17Q48+6PRD-YPet (SEQ ID NO.: 68) and AN17Q48+6PRD-CyPet (SEQ ID NO.: 67).

In some preferred embodiments of the methods for assessing the risk for development, predicting the onset or assessing the progression of a polyQ disease, the first and second recombinant polyQ-containing proteins of step (a) are AN17Q40+6PRD-YPet (SEQ ID NO.: 70) and AN17Q40+6PRD-CyPet (SEQ ID NO.: 69).

In some preferred embodiments of the method for identifying compounds that inhibit mHTT seeding activity (HSA) in vitro, the purified recombinant mutant forms of huntingtin (mHTT) in step (i)(a) and step (ii)(a) are mHTT GST- Ex1 Q48-YPet (SEQ ID NO.: 17) and mHTT GST-Ex1 Q48-CyPet (SEQ ID NO.: 19).

In some preferred embodiments of the method for identifying compounds that inhibit mHTT seeding activity (HSA) in vitro, the purified recombinant mutant forms of huntingtin (mHTT) in step (i)(a) and step (ii)(a) are mHTT GST- K2Q48P6-CyPet (SEQ ID NO.: 48) and mHTT GST-K2Q48P6-YPet (SEQ ID NO.: 50).

In some preferred embodiments of the method for identifying compounds that inhibit mHTT seeding activity (HSA) in vitro, the purified recombinant mutant forms of huntingtin (mHTT) in step (i)(a) and step (ii)(a) are mHTT GST- DN17Q48+6PRD-CyPet (SEQ ID NO.: 52) and mHTT GST-AN17Q48+6PRD- YPet (SEQ ID NO.: 54).

In some preferred embodiments of the method for identifying compounds that inhibit mHTT seeding activity (HSA) in vitro, the purified recombinant mutant forms of huntingtin (mHTT) in step (i)(a) and step (ii)(a) are mHTT GST- DN17Q40+6PRD-CyPet (SEQ ID NO.: 56) and mHTT GST-AN17Q40+6PRD- YPet (SEQ ID NO.: 58).

The ratio in which the protein fused to the donor fluorophore molecule and the protein fused to the acceptor fluorophore molecule (both of which are also referred to herein as“the fluorophore fusion proteins”) are provided will typically be selected in order to provide the best FRET results. In the context of the present invention, it is preferred that the ratio of the protein fused to the donor fluorophore molecule to the protein fused to the acceptor fluorophore molecule ranges from 2:3 to 3:2, e.g. 2:3, 4:5, 7:8, 1 :1 , 8:7, 5:4, 3:2. In certain preferred embodiments, the ratio is about 1 :1 or exactly 1 :1 , i.e. the fluorophore fusion proteins are provided as an equimolar mixture. Suitable concentrations of the fluorophore fusion proteins in the reaction mixture are typically in the low micromolar range. For example, the fluorophore fusion proteins can be provided in a concentration of from about 0.1 mM to about 2.0 pM each (i.e. when using an equimolar mixture, the final concentration of both fluorophore fusion proteins will be from about 0.2 pM to about 4.0 pM). Specific examples of fluorophore fusion protein concentrations, e.g. of GST-Ex1 Q48-YPet and GST-Ex1 Q48- CyPet, are 0.5 pM each, 0.6 pM each, 0.7 pM each, 0.75 pM each or 0.8 pM each, particularly in an equimolar mixture (i.e. final concentrations 1.0 pM, 1.2 pM, 1.4 pM, 1.5 pM or 1.6 pM).

In this context it is noted that the sample added to the mixture of fluorophore fusion proteins may be in the range of from about 1 % (v/v) to about 30% (v/v), particularly 1 % (v/v) to 10% (v/v).

Measurement of fluorescence signals is performed as known in the field of FRET, for instance in a fluorescence plate reader. Fluorescence signals in the donor channel are usually measured at the peak excitation (Ex) wavelength and the peak emission (Em) wavelength. For cyan fluorescent proteins, signals may e.g. be recorded at the following wavelengths:

In certain preferred embodiments of the invention, fluorescence signals in the donor channel are measured at 435 nm (excitation) and 475 nm (emission).

Likewise, fluorescence signals in the acceptor channel are usually measured at the peak excitation (Ex) wavelength and the peak emission (Em) wavelength. For yellow fluorescent proteins, signals may e.g. be recorded at the following wavelengths:

In certain preferred embodiments of the invention, fluorescence signals in the acceptor channel are measured at 500 nm (excitation) and 530 nm (emission).

Fluorescence signals in the FRET channel (also referred to as DA) are typically recorded at the excitation wavelength of the donor fluorophore and the acceptor wavelength of the acceptor fluorophore. In certain preferred embodiments of the invention, fluorescence signals in the FRET channel are measured at 435 nm (excitation) and 530 nm (emission). The raw signals obtained are then processed to calculate FRET efficiency. To this end, raw signals were processed by subtracting the background fluorescence of unlabeled Ex1 Q48, or unlabeled K2Q48P6, or unlabeled AN17Q48+6PRD or unlabeled AN17Q40+6PRD, respectively, in all channels. Signals in the FRET channel were corrected for donor bleed -through (CD) and acceptor cross excitation (c _A ) using donor- and acceptor-only samples to obtain sensitized emission. Finally, sensitized emission was normalized to the acceptor signals (Jiang and Sorkin, 2002). In preferred embodiments of the invention, FRET efficiency (E) is calculated according to formula (1 )

E = (DA - c _D DD - c _A AA)/AA (1 ),

wherein DA is the FRET channel signal

C _D is the donor bleed-through

DD is the donor channel signal

c _A is the acceptor cross-excitation

AA is the acceptor channel signal.

Measurement of fluorescence signals can be performed at a predetermined timepoint or, preferably, it can be measured at certain intervals after the assay where FRET signals are expected has been started. This intervals will typically be in the range of minutes, e.g. from 1 minute to 50 minutes, particularly 10 to 30 minutes. A specific example according to the invention is an interval of 20 minutes. Measurement is usually stopped when FRET efficiency has reached saturation. This can be after, e.g., 12 hours, 24 hours, 48 hours or even longer, depending on the experimental conditions. A specific endpoint for fluorescence measurement in the methods according to the invention is 48 hours.

Accordingly, in some embodiments, the fluorescence signals are measured at intervals of 20 minutes for up to 48 h.

The FRET efficiency results can be further used to quantify seeding activity (At ₅₀ in hours), which in the case of mutant huntingtin, is referred to as mFITT seeding activity (FISA). To this end, the t ₅₀ values (time at half-maximal FRET efficiency) of the respective sample are subtracted from the negative control. To obtain the t ₅₀ values, the aggregation kinetics are preferably curve fitted by Richard’s five-parameter dose-response curve (Formula (2) below), e.g. using the software GraphPad Prism (GraphPad Software, La Jolla, USA). y y¥-Vo

= Vo + 0 [l+lO lLogxb-x)xHUISlope]' ύ (2) wherein

y ₀ = minimum asymptote

y-= maximum asymptote

Logxb = LogEC50 + (1/HillSlope) ^* Log((2 ^A (1/S))-1 )

LogEC50 = concentrations that give half-maximal effects, in the same units as X

x = time

HillSIope = unitless slope factor or Hill slope

S = asymmetry factor

All methods involving quantification of seeding activity described herein comprise a step of shaking, in order to properly mix the fluorophore containing molecules with the added sample to be analyzed. Shaking can be performed, e.g., manually or automatically, e.g . in vertical direction, in a fluorescence reader, for instance from about 3 seconds to about 10 seconds. In certain preferred embodiments, shaking the mixture is performed by 5 seconds vertical shaking. The present invention further provides the use of certain constructs in an aggregation assay. For example, these constructs may be used in the methods involving quantification of seeding activity described herein. In some embodiments, the invention provides the use of a soluble glutathione S- transferase HTT exon-1 fusion protein comprising from about 35 to about 75, particularly about 40 to about 55, e.g. 40 or 48 or 49 glutamine residues which is C-terminally fused to a donor fluorophore molecule, e.g. a CFP such as CyPet, and of a soluble glutathione S-transferase HTT exon-1 fusion protein comprising from about 35 to about 75, particularly about 40 to about 55, e.g. 40 or 48 or 49 glutamine residues which is C-terminally fused to an acceptor fluorophore molecule, e.g. a YFP such as YPet in an aggregation assay.

In preferred embodiments of the described use, the donor fluorophore fusion protein comprises 48 glutamine residues and/or the donor fluorophore is CyPet. In other preferred embodiments of the described use, the donor fluorophore fusion protein comprises 40 glutamine residues and/or the donor fluorophore is CyPet. Between GST and the HTT sequence, a protease recognition sequence as described herein is typically present. Preferably, the donor fluorophore fusion protein consists of mHTT GST-Ex1 Q48-CyPet. A specific example of such a fusion protein has the amino acid sequence as shown in SEQ ID NO.: 19. In other preferred embodiment, the donor fluorophore fusion protein consists of mHTT GST-K2Q48P6-CyPet. A specific example of such a fusion protein has the amino acid sequence as shown in SEQ ID NO.: 48. In another preferred embodiment, the donor fluorophore fusion protein consists of mHTT GST- AN17Q48+6PRD-CyPet. A specific example of such a fusion protein has the amino acid sequence as shown in SEQ ID NO.: 52. In another preferred embodiment, the donor fluorophore fusion protein consists of mHTT GST- AN17Q40+6PRD-CyPet. A specific example of such a fusion protein has the amino acid sequence as shown in SEQ ID NO.: 56.

Also, in preferred embodiments of the described use, the acceptor fluorophore fusion protein comprises 48 glutamine residues and/or the acceptor fluorophore is YPet. In other preferred embodiments of the described use, the acceptor fluorophore fusion protein comprises 40 glutamine residues and/or the acceptor fluorophore is YPet. Between GST and the HTT sequence, a protease recognition sequence as described herein is typically present. Preferably, the acceptor fluorophore fusion protein consists of mHTT GST-Ex1 Q48-YPet. A specific example of such a fusion protein has the amino acid sequence as shown in SEQ ID NO.: 17. In another preferred embodiment, the acceptor fluorophore fusion protein consists of mHTT GST-K2Q48P6-YPet. A specific example of such a fusion protein has the amino acid sequence as shown in SEQ ID NO.: 50. In another preferred embodiment, the acceptor fluorophore fusion protein consists of mHTT GST-AN17Q48+6PRD-YPet. A specific example of such a fusion protein has the amino acid sequence as shown in SEQ ID NO.: 54. In another preferred embodiment, the acceptor fluorophore fusion protein consists of mHTT GST-AN17Q40+6PRD-YPet. A specific example of such a fusion protein has the amino acid sequence as shown in SEQ ID NO.: 58.

In yet further embodiments of the invention, the use of a soluble HTT exon-1 fusion protein comprising from about 35 to about 75, particularly about 40 to about 55, e.g. 40 or 48 or 49 glutamine residues which is C-terminally fused to a donor fluorophore molecule, e.g. a CFP such as CyPet, and of a soluble HTT exon-1 fusion protein comprising from about 35 to about 75, particularly about 40 to about 55, e.g. 40 or 48 or 49 glutamine residues which is C-terminally fused to an acceptor fluorophore molecule, e.g. a YFP such as YPet in an aggregation assay is provided.

In preferred embodiments of the described use, the donor fluorophore fusion protein comprises 48 glutamine residues and/or the donor fluorophore is CyPet. Preferably, the donor fluorophore fusion protein consists of mHTT Ex1 Q48- CyPet. A specific example of such a fusion protein has the amino acid sequence as shown in SEQ ID NO.: 15. In another preferred embodiment, the donor fluorophore fusion protein consists of K2Q48P6 -CyPet. A specific example of such a fusion protein has the amino acid sequence as shown in SEQ ID NO.: 65. In another preferred embodiment, the donor fluorophore fusion protein consists of AN17Q48+6PRD-CyPet. A specific example of such a fusion protein has the amino acid sequence as shown in SEQ ID NO.: 67. In another preferred embodiment, the donor fluorophore fusion protein consists of AN17Q40+6PRD-CyPet. A specific example of such a fusion protein has the amino acid sequence as shown in SEQ ID NO.: 69.

Also, in preferred embodiments of the described use, the acceptor fluorophore fusion protein comprises 48 glutamine residues and/or the acceptor fluorophore is YPet. Preferably, the acceptor fluorophore fusion protein consists of mHTT Ex1 Q48-YPet. A specific example of such a fusion protein has the amino acid sequence as shown in SEQ ID NO.: 13. In another preferred embodiment, the acceptor fluorophore fusion protein consists of K2Q48P6-YPet. A specific example of such a fusion protein has the amino acid sequence as shown in SEQ ID NO.: 66. In another preferred embodiment, the acceptor fluorophore fusion protein consists of AN17Q48+6PRD-YPet. A specific example of such a fusion protein has the amino acid sequence as shown in SEQ ID NO.: 68. In another preferred embodiment, the acceptor fluorophore fusion protein consists of AN17Q40+6PRD-YPet. A specific example of such a fusion protein has the amino acid sequence as shown in SEQ ID NO.: 70.

The aggregation assays mentioned above are preferably cell -free aggregation assays, i.e. no intact living cells are present in the assay.

Within the above-described uses, the inventors have surprisingly found that it is possible to obtain good results even when the sample to be assayed is a complex biosample. Thus, according to preferred embodiments, the above- described fusion proteins are used in an aggregation assay, wherein the aggregation assay involves monitoring mHTT seeding activity (HSA) in a complex biosample.

A “complex biosample” according to the invention is a sample that may comprise a variety of proteins and other factors. In particular, a complex biosample within the context of the invention is selected from the group consisting of optionally pretreated tissue samples, optionally pretreated body fluid samples and optionally pretreated cell culture samples. As described above, pretreatment can be effected by at least one of centrifugation, cell lysis, protein extraction, dialysis, sonication and similar procedures used in protein purification known to the skilled person. Pretreated tissue, body fluid or cell culture samples may, e.g., be selected from the group consisting of a homogenate, an extract, a pellet and a lysate. As indicated above, the pretreated sample may, in some embodiments and in addition to previous pretreatment steps, be sonicated. More particularly, the complex biosample is selected from muscle tissue, homogenated muscle tissue, protein extracts from muscle tissue, nasal brushing, nasal epithelial tissue, full blood, homogenated blood, blood plasma, cerebrospinal fluid, cells, cell extracts, cell pellets, cell lysates, e.g. from a cell line, from stem cells, or from primary cell culture, brain homogenates, brain extracts that are optionally sonicated, and protein extracts from post-mortem tissue such as cerebral cortex, caudate nucleus and cerebellum.

The present invention further provides soluble protein constructs, which find application, e.g., in the methods and uses described herein. These protein constructs mandatorily comprise exon 1 of huntingtin with 48 glutamine residues (HTTEx1 Q48), and, fused to the C-terminus of the HTTEx1 Q48 sequence, a fluorophore, i.e. either CyPet or YPet. Optionally, glutathione S- transferase (GST) may be fused to the N-terminus of the HTTEx1 Q48 sequence. HTTEx1 Q48 particularly has the sequence shown in SEQ ID NO.: 2. CyPet particularly has the sequence as shown in SEQ ID NO.: 23. YPet particularly has the sequence as shown in SEQ ID NO.: 21. GST, if present, particularly has the sequence as shown in SEQ ID NO.: 6. Specific examples of these soluble protein constructs are mHTT Ex1 Q48-YPet (SEQ ID NO.: 13), mHTT Ex1Q48-CyPet (SEQ ID NO.: 15), mHTT GST-Ex1 Q48-YPet (SEQ ID NO.: 17) and mHTT GST-Ex1 Q48-CyPet (SEQ ID NO.: 19).

The present invention further provides soluble protein constructs, which find application, e.g., in the methods and uses described herein, comprising mutant exon 1 of huntingtin with 40 or 48 glutamine residues and an adjacent modified PRD comprising 6 or more, preferably 6-20, consecutive proline residues (Q40Pn and Q48Pn proteins, respectively, with n representing the number of consecutive proline residues), and, fused to the C-terminus of the mutant exon 1 sequence, a fluorophore, i.e. either CyPet or YPet. Optionally, glutathione S- transferase (GST) may be fused to the N-terminus of the mutant exon 1 sequence. A Q40P17 protein particularly has the sequence shown in SEQ ID NO.: 73 (AN17Q40+6PRD); a Q48P6 protein particularly has the sequence shown in SEQ ID NO.: 72 (AN17Q48+6PRD); a Q48P17 protein particularly has the sequence shown in SEQ ID NO: 71 (K2Q48P6). CyPet particularly has the sequence as shown in SEQ ID NO.: 23. YPet particularly has the sequence as shown in SEQ ID NO.: 21. GST, if present, particularly has the sequence as shown in SEQ ID NO.: 6. Specific examples of these soluble protein constructs are mHTT K2Q48P6-YPet (SEQ ID NO.: 66), mHTT K2Q48P6-CyPet (SEQ ID NO.: 65), mHTT GST-K2Q48P6-YPet (SEQ ID NO.: 50), mHTT GST-K2Q48P6- CyPet (SEQ ID NO.: 48), mHTT AN17Q48+6PRD-YPet (SEQ ID NO.: 68), mHTT DN17Q48+6PRD-CyPet (SEQ ID NO.: 67), mHTT GST- DN17Q48+6PRD-YPet (SEQ ID NO.: 54), mHTT GST-AN17Q48+6PRD-CyPet (SEQ ID NO.: 52), mHTT AN17Q40+6PRD-YPet (SEQ ID NO.: 70), mHTT DN17Q40+6PRD-CyPet (SEQ ID NO.: 69), mHTT GST-AN17Q40+6PRD-YPet (SEQ ID NO.: 58) and mHTT GST-AN17Q40+6PRD-CyPet (SEQ ID NO.: 56).

FIGURE LEGENDS

Figure 1. Establishment of a FRET-based mutant HTT aggregate seeding assay

(A) Time-dependent aggregation of Ex1Q48-CyPet and -YPet fusion proteins (3 mM) monitored by FRA (500 ng protein per dot). Immunblot, anti-GFP antibody.

(B) Analysis of spontaneously formed Ex1 Q48-CyPet, Ex1Q48-YPet and Ex1 Q48 aggregates by AFM (3 pM) after 24 h. Scale bars: 1 pm; color gradient represents 0-20 nm height.

(C) Schematic model of FRET-inducing co-aggregating Ex1 Q48-CyPet and - YPet upon cleavage of GST fusion proteins with PSP.

(D) Spontaneous time-dependent co-aggregation of Ex1 Q48-CyPet and -YPet sensor proteins (1 :1 mixture) upon incubation of GST fusion proteins with PSP at 25 °C. FRET efficiency is displayed as mean ± SD of technical triplicates.

(E) Fibrillar Ex1 Q48 aggregates (seeds) induce a concentration-dependent shortening of the lag phase in Ex1 Q48-CyPet and -YPet (1 :1 mixture, total cone. 1.2 pM) co-polymerization reactions. Seed concentrations are equivalent to monomer concentrations. FRET efficiency is displayed as mean ± SD of technical triplicates. (F) Quantification of mHTT seeding activities (HSAs, At ₅₀ values) from aggregation profiles in E. At ₅₀ is displayed as individual values (·) and mean ± SD of technical triplicates.

Figure 2. FRASE assays facilitate detection of HSA with high specificity and sensitivity

(A) Analysis of sonicated (1 min) and non-sonicated fibrillar Ex1 Q48 aggregates by blue native (BN) PAGE and immunoblotting using HD1 antibody.

(B) Effects of small, preformed Ex1 Q48 seeds (1250 kDa) on Ex1 Q48-CyPet and -YPet (1 :1 mixture, total cone. 1.2 mM) co-aggregation. Data are mean ± SEM (n = 5).

(C) Calculation of HSAs (At ₅₀ values) from aggregation profiles in B. Data are mean ± SEM (n = 5).

(D) Effects of non-polyQ fibrils on Ex1 Q48-CyPet and -YPet (1 :1 mixture, total cone. 1.2 pM) co-aggregation. Data are mean ± SD of triplicates.

(E) Analysis of a-synuclein (a-Syn), amyloid^42 (Ab), islet amyloid polypeptide (IAPP) and Tau (Tau40) fibrils by AFM. Scale bars: 1 pm; Height of color gradients: 0-10 nm (a-Syn), 0-5 nm (Ab), 0-30 nm (IAPP) and 0-10 nm (Tau40).

Figure 3. Quantification of HSA in brain extracts of patients and HD mice

(A) Schematic workflow for quantifying HSA in brain tissue homogenates.

(B) Effects of mouse brain homogenates on Ex1Q48-CyPet and -YPet (1 :1 mixture, total cone. 3 pM) co-aggregation. Data are mean ± SD of technical triplicates.

(C) HSA (At ₅₀ values) of mouse brain extracts investigated in B. Statistical analysis: two-way ANOVA followed by Bonferroni’s multiple comparison post hoc test against the respective WT controls. At ₅₀ is displayed as individual values (·) and as mean ± SD of technical triplicates.

(D) Quantification of HSA in brain homogenates prepared from HD patients and controls. For clarity, the average At ₅₀ values obtained from 3 healthy control samples are depicted (Average Ctrl). Individual values of At ₅₀ (·) and mean ± SD of triplicates are displayed. (E) Quantification of HSA in brain extracts of presymptomatic R6/2Q212 and wild-type (WT) control mice (3 mice per age). Results from WT mice are shown as an average At ₅ o value.

(F) Quantification of HSA in brain extracts of presymptomatic R6/2Q212 and wild-type (WT) control mice after sonication. Results from WT mice are shown as an average At ₅₀ value.

(G) Quantification of HSA in brain tissue extracts of HdhQ'\50 heterozygous knock-in and WT mice. Data are mean ± SEM (n = 3). One-Way ANOVA followed by Dunnett's multiple comparisons test.

Figure 4. Detection of small HTTexl fibrils in soluble brain fractions with high HSA

(A) Schematic workflow for preparing soluble and insoluble protein fractions from crude tissue homogenates by centrifugation.

(B) Quantification of HSA in soluble and insoluble fractions prepared from brains of R6/2Q212 transgenic mice. In all cases, data were normalized to average At ₅₀ values of age-matched WT control mice. Bars are mean ± SEM (n = 2). HSAs measured for individual mice are displayed as black dots (·).

(C) Detection of large, SDS-stable fibrillar mHTTexl aggregates in prepared protein fractions by FRA.

(D) Detection of small, fibrillar HTTexl aggregates in P2 fractions of R6/2Q212 mice by immunoelectron microscopy using the anti-HTT antibody Agg53. Scale bar: 100 nm.

Figure 5. The short-term HTTexl Q97 expression -induced decrease in Drosophila lifespan correlates with HSA but not with aggregate load in general

(A) Scheme for short- and long-time RU486 treatment of adult HD transgenic flies.

(B) Survival of RU486 treated and untreated elavGS;HTTex1 Q97 and elavGS;HTTex1 Q17 flies (n = -100 flies/group). Survival was plotted as the percentage of surviving flies of 3 biological replicates.

(C) Median life span calculated from survival curves in B. Average survival of each experiment (n = -100 flies/group) is presented as black dots (·). Bars are mean ± SEM from 3 independent experiments; One-way ANOVA Dunnett’s post-hoc test; data were compared to elavGS;HTTex1 Q97 ^OFF transgenic flies [statistically significant differences are indicated by asterisks ( ^* )] or to elavGS;HTTex1 Q97 ^ON flies [statistically significant differences are indicated by hashtags (#)].

(D) Analysis of motor performance of RU486 treated and untreated elavGS;HTTex1 Q97 flies (n = -100 flies/group; three independent experiments).

(E) Quantification of large, SDS-stable fibrillar HTTexl aggregates in heads of RU486 treated and untreated elavGS;HTTex1 Q97 flies by FRA using the MW8 antibody. Representative images for each condition are shown. Data are mean ± SEM of individual measurements (·); One-way ANOVA Dunnett’s post hoc test.

(F) FRASE analysis of head lysates analyzed in £. Values are means ± SEM of three biological replicates each performed in triplicates.

(G) Quantification of FISA (At ₅₀ values) from FRET-based aggregation profiles depicted in F. Results are displayed as mean ± SEM; individual measurements (·); One-way ANOVA Dunnett’s post hoc test compared to elavGS;HTTex1 Q97 ^OFF flies.

(FI) Pearson correlation analysis shows a significant linear relationship between the survival of RU486 treated and untreated elavGS;FITTex1 Q97 flies and the FISA measured by FRASE assays (p = 0.020). No such correlation was observed between fly survival and the abundance of large, fibrillar FITTex1 Q97 aggregates detected by FRAs [p = 0.368 (FRA, MAB5492), p = 0.411 (FRA, MW8)]. Data are presented as mean ± SEM of the three independent experiments.

Figure 6. Co-production of Flsp70 extends survival of FID flies

(A) Scheme depicting the workflow for the temporary treatment of FID transgenic flies with RU486.

(B) Life span analysis of short-time RU486 treated elavGS;FISPA1 L;FITTex1 Q97 flies (n = -90 flies/group). In control experiments, hormone-treated and untreated elavGS;FITTex1 Q97 (n = -40 flies/group) were analyzed. The percentage of surviving flies of three biological replicates are shown.

(C) Median life span calculated from survival curves in B. Bars are mean ± SEM from three independent replicates; Unpaired t test.

(D) Quantification of large, SDS-stable HTTex1Q97 aggregates in fly heads by FRAs using the MW8 antibody. Representative images for each condition are shown. Data are mean ± SEM; Individual measurements are presented as dots (•);Unpaired t-test.

(E) FRASE analysis of fly head lysates. Values are plotted as mean ± SEM of 3 biological replicates each performed in triplicates.

(F) Quantification of FISA from FRET-based aggregation profiles in E. Data are mean ± SEM; Individual measurements (·); Unpaired t test.

(G) Representative confocal images of the right central brain region of elavGS;FISPA1 L;FITTex1 Q97 flies treated for 6 days with RU486 and immunostained for FITT (Mab5492, green) and Flsp70 (anti-FISP70/FISP72, red). White arrows indicate co-localization of Flsp70 with FITTexl aggregates; grey arrow indicates areas with no co-localization. Scale bars: 20 pm.

Figure 7. Characterization of FITTexl sensor proteins

(A) Schematic representation of the applied GST -tagged FITTexl fusion proteins with pathogenic and non-pathogenic polyQ tracts. P, proline-rich regions. (B) The recombinant GST-Ex1 Q48-CyPet and -YPet fusion proteins were affinity purified using glutathione-coated sepharose beads. Purity was assessed by SDS-PAGE and subsequent Coomassie blue staining. (C) Investigation of PreScission protease (PSP)-mediated cleavage of GST- Ex1 Q48-CyPet and -YPet fusion proteins. 3 pM of GST fusion proteins were incubated in the presence or absence of PSP. Aliquots were taken at the indicated time points; cleavage was confirmed by SDS-PAGE and immunoblotting with a polyclonal anti-GFP antibody. (D) AFM analysis of co- aggregated sensor proteins Ex1 Q48-CyPet/-YPet (3 pM). Scale bar: 1 pm; color gradient represents 0-20 nm height. (E) Preformed, fibrillar Ex1 Q48 aggregates (seeds) shorten the lag phase of Ex1 Q48-CyPet/-YPet polymerization; no seeding effect was observed with the sensor proteins Ex1 Q23-CyPet/-YPet. Seeds were produced by incubating GST-Ex1 Q48 fusion protein with PSP for 24 h at 25 °C. Indicated seed concentrations are equivalent to monomer concentrations. Co-aggregation of the sensor proteins (1 :1 mixture; 1 .2 mM) was monitored by quantification of FRET; the resulting aggregation kinetics were curve fitted by non-linear regression. FRET efficiency is plotted as means ± SD of 3 technical replicates. (F) Ex1 Q48-CyPet/-YPet (1 :1 mixture; 3 mM) sensor protein co-aggregation was accelerated by addition of preformed fibrillary Ex1 Q48 seeds. Ex1 Q23 protein was prepared under identical conditions (no fibrillar aggregates observed; data not shown). Co-aggregation of sensor proteins was not influenced by the addition of uncleaved GST-Ex1 Q48 or GST- Ex1 Q23 fusion proteins, respectively. Co-aggregation of the fluorescence sensor proteins was monitored by quantification of FRET; the resulting aggregation kinetics were curve fitted by non-linear regression. Indicated seed concentrations are equivalent to monomer concentrations. Data is shown as means ± SD of 3 technical replicates.

Figure 8. Both small and large fibrillar Ex1Q48 aggregates exhibit seeding activity in FRASE assays

(A) Sonication of preformed, fibrillar Ex1 Q48 aggregates reveals protein fractions with high FISA in FRASE assays. Fibrillar Ex1 Q48 aggregates were produced by incubating GST-Ex1 Q48 fusion protein (2 pM) for 24 h at 25 °C. 1 nM preformed Ex1 Q48 aggregates (seeds) were added to Ex1 Q48-CyPet/- YPet (1 :1 mixture; 1 .2 pM) co-aggregation reactions. The added seed concentration is equivalent to the monomer concentration. Data is shown as means ± SD of 3 technical replicates. (B) Quantification of FISA. Calculated At ₅₀ values from Ex1 Q48-CyPet/-YPet aggregation profiles in A. At ₅₀ is displayed as individual values (·) and mean ± SD of technical triplicates. (C) Analysis of sonicated and non-sonicated Ex1 Q48 seeds by denaturing filter retardation (FRA, left panel) and dot blot (DB, right panel) assays. Fragmentation of large fibrillar Ex1 Q48 aggregates by sonication prevents their detection by FRAs. (D) Preformed Ex1 Q48 fibrils were sonicated for the indicated times and visualized by AFM. Sonication reduces the size of preformed Ex1 Q48 fibrils. Scale bar: 1 pm; color gradient represents 0-20 nm height. Figure 9. Detection of mHTT seeding activity in brain extracts of various HD mouse models

(A) Brain homogenates (7.5 pg) prepared from R6/2Q212 and WT mice were added to Ex1 Q48-CyPet/-YPet or Ex1 Q23-CyPet/-YPet sensor proteins. Brain homogenates prepared from R6/2Q212 mice accelerated Ex1 Q48-CyPet/-YPet polymerization but do not induce co-aggregation of Ex1 Q23-CyPet/-YPet sensor proteins. Co-aggregation of sensor proteins (1 :1 mixture; 1.2 mM) was monitored by quantification of FRET and displayed as mean ± S D of technical triplicates; the resulting aggregation kinetics were curve fitted by non-linear regression. (B) Quantification of HSA in brain extracts prepared from R6/2Q51 transgenic mice and controls using FRASE assays (1.2 pM Ex1Q48-CyPet/- YPet). HSA measured for each mouse is displayed as dots (·). Bars are mean ± SEM. Statistical significance was assessed by One-Way ANOVA followed by Dunnett's multiple comparisons test (n = 2). (C) Analysis of body weight of FVB/N mice expressing the proteins HTT853Q18 or HTT853Q79 for 8 weeks. Body weight of individual mice is displayed as dots (·). Bars are mean ± SEM. Statistical significance was assessed by One-Way ANOVA followed by Dunnett's multiple comparisons test (n = 3). (D) Quantification of HSA in hypothalamic brain homogenates of FVB/N mice expressing the proteins HTT853Q18 or HTT853Q79. The total concentration of sensor proteins was 3 pM. Data are mean ± SEM (n = 3). Individual measurements are displayed as dots (·). Statistical significance was assessed by One-Way ANOVA followed by Dunnett's multiple comparisons test. (E) Brain homogenates prepared from tissues of HD (caudate nucleus) and AD (temporal cortex) patients were analyzed by FRASE assays and compared to corresponding brain tissue of control individuals. HSA values are plotted individually as dots (·) and as mean ± SEM (n = 3); caudate tissue from HD patients (Grade 4, CAG repeat length: 52.3 ± 1.2, Age 42.3 ± 2.1 ), caudate tissue from controls (Age 60.3 ± 1.2), cortical tissue from AD patients (Braak 6, Age 73.3 ± 4.6), cortical tissue from controls (Braak 0, Age 66.3 ± 7). Statistical significance was assessed by unpaired t test. (F) Immunodepletion of mHTTexl aggregates from R6/2Q212 mouse brain homogenates decreases their seeding activity in FRASE assays. Brain homogenates prepared from transgenic mice and littermate controls (12 weeks) were incubated with MW8 antibody-coated protein G beads; supernatant (post-IP) and input samples were applied to FRASE analysis using 3 mM of sensor proteins. FRET efficiency is plotted as mean ± SD of technical triplicates. (G) Same procedure as in F but with an IgG isotype control antibody. (FI and I) Immunoblots of samples analyzed in F and G. mFITTexl aggregates appear as a smear at the upper edge of the blot (red rectangles). Input, brain extract before immunodepletion; Sup, supernatant after immunodepletion; Beads, antibody-coated protein G beads after immunodepletion. mFITTexl aggregates are depleted from mouse brain homogenates with the anti-FITT antibody MW8 but not with an IgG isotype control antibody.

Figure 10. Detection of mFITTexl aggregates in Drosophila FID models

(A) Illustration of hormone treatment utilized to assess the dynamics of FITTex1 Q97 transcriptional repression upon hormone removal. Grey lines indicate time periods without RU486 treatment; black lines indicate time periods of exposure to RU486. (B) Relative mRNA levels assessed by qPCR show transcriptional repression of FITTex1 Q97 upon removal of RU486. Values are depicted as mean ± SEM of 3 biological replicates; individual measurements (·); significance assessment with One-way ANOVA Dunnett’s post-hoc test. (C) Treatment scheme for the preparation of Drosophila samples for FRA and FRASE analyses. (D) Analysis of mFITTexl aggregate load in fly head lysates by FRAs (75 pg protein) using the MAB5492 antibody. Data is displayed as mean ± SEM of 3 biological replicates; individual measurements (·); statistical significance was assessed by One-way ANOVA Dunnett’s post-hoc test. (E) Representative confocal images of elavGS;FITTex1 Q97 whole fly brains (hormone treatment as in C. The RBP staining is shown in magenta and the MAB5492 staining in green (Scale bars: 200 pm). Magnifications are shown below (Scale bars: 20 pm).

Figure 11. Effects of chaperone levels on FISA in transgenic flies and worms

(A) Relative mRNA levels assessed by qPCR show similar expression of HTTex1 Q97 in elavGS;HTTex1 Q97 and elavGS;HSPA1 L;HTTex1 Q97 fl ies. Bars are mean ± SEM of three biological replicates and individual measurements are displayed as dots (·); significance assessment with unpaired t-test. (B) Comparison of Hsp70 protein levels in elavGS;HSPA1 L;HTTex1 Q97 and elavGS;HSPA1 L flies untreated and treatment with RU486. Protein extracts prepared from fly heads were analyzed by SDS-PAGE and immunoblotting (20 pg of total protein). (C) Motility phenotype (% motility) of RNAi-treated and untreated Q35-YFP expressing transgenic worms at day 5. In all cases, data were normalized to age-matched control worms. Data are mean ± SEM (n = 20). Significance assessment with unpaired t test. (D) FRASE analysis of Q35-YFP seeding activity in RNAi- treated and untreated worms after 5 days. FRET efficiency is displayed as mean ± SD. (E) Quantification of results shown in D. FISA values are plotted individually as dots (·) and as mean ± SD.

Figure 12. FISA detected in the putamen of FID patients increases with the advancement of neuropathological changes

Assessment of FISA in homogenates prepared from putamen of control individuals and FID patients at different disease stages. At ₅₀ values for each individual are plotted as circles; Boxes show first and third quartiles, the central band shows the median, and the whiskers show data within 1.5 IQR of the median; putamen tissue from FID patients (Grade 2 (n = 4), CAG repeat length 45.8 ± 0.96, Age 60.25 ± 12.1 ; Grade 3 (n = 4), CAG repeat length 47.5 ± 1.7, Age 54.5 ± 6.6; Grade 4 (n=5), CAG repeat length 52.0 ± 1.0, Age 44.6 ± 4.9), caudate tissue from controls (n = 10, Age 60.6 ± 9.1 ); statistical significance was assessed by One Way ANOVA followed by Dunnett's multiple comparisons test.

Figure 13. Modulation of mFITT seeding activity by small molecule compounds

Flttex1 Q48 seeds were pre-incubated with 04 in 200-fold molar excess at 25 °C for 20 h and subsequently analyzed for their seeding activity using FRASE assays. (A) Aggregation profiles of sensor proteins in the presence or absence of 04 treated seeds and respective controls. Data are mean ± SD of technical triplicates. (B) Calculation of At ₅ o values from aggregation profiles in A. At ₅ o is displayed as individual values (black ·) and as mean ± SD of technical triplicates. METHODS

EXPERIMENTAL MODEL AND SUBJECT DETAILS HD mouse models

Hemizygous R6/2Q212 mice (Mangiarini et al., 1996) were bred by backcrossing R6/2Q212 males to (CBA/Ca x C57BI/6J) F1 females (B6CBAF1/OlaHsd, Harlan Olac, Bicester, UK). HdhQ'\ 50 heterozygous knock- in mice (Lin et al., 2001 ; Woodman et al., 2007) on a (CBA/Ca x C57BI/6J) F1 background were generated by intercrossing HdhQ'\ 50 heterozygous CBA/Ca and C57BL/6J congenic lines (inbred lines from Harlan Olac, Bicester, UK). All animals were subject to a 12 h light/dark cycle with unlimited access to drinking water and breeding chow (Special Diet Services, Witham, UK). Housing conditions and environmental enrichment were described previously (Hockly et al., 2003). R6/2 mice were always housed with wild-type mice. The CAG repeat size of the R6/2Q212 mice used in this study was 212 ± 5.27 (s.d.) and that of the HdhQ'\50 heterozygotes was 160 ± 2.86 (s.d.). Hemizygous R6/2Q51 mice were derived from R6/2 parent lines by selective breeding (Larson et al., 2015) and bred by backcrossing R6/2Q51 males to (CBA x C57BI/6) F1 females (Charles Rivers, UK). R6/2Q51 mice were maintained and bred as described previously (Larson et al., 2015). Female mice from the FVB/N strain (Charles River Laboratories, Germany) were injected at eight to ten weeks of age with recombinant adeno-associated viral (AAV) vectors of serotype 5 encoding the first 853 amino acids of either the WT form of HTT with 18Q (HTT853- Q18) or the mutant form of the protein with 79Q (HTT853-Q97) (Baldo et al., 2013). All mice were housed in groups at a 12 h light/dark cycle. At 8 weeks post- injection, FVB/N mice were sacrificed. Overall, mice were sacrificed at different ages from 1 day up to 2 years. Tissues were strictly stored at -80 °C until use.

All animal work with R6/2Q212 and HdhQ 150m ice was approved by the King’s College London Ethical Review panel and performed under a Home Office project license in accordance with the United Kingdom 1986 Animals (Scientific procedures) Act. All animal work with R6/2Q51 mice was approved by the University of Cambridge Ethical Review panel and performed under a Home Office project license in accordance with the United Kingdom 1986 Animals (Scientific procedures) Act. All experimental procedures with FVB/N mice were approved by the Regional Ethical Committee in Lund, Sweden.

Human brain tissue

Post mortem brain tissues from human HD and AD patients and unaffected control individuals (both male and female) were obtained from the Newcastle Brain Tissue Resource (NBTR, Newcastle University, UK). Experiments were performed in accordance with the approval of the joint Ethics Committee of Newcastle and North Tyneside Health Authority and following NBTR brain banking procedures. Tissues were collected at 34.5 ± 21.0 h post-mortem from HD patients and controls with an average age of 57.8 ± 10.7 years.

Generation and maintenance of Drosophila strains

ElavGS-GAL4, Elav-GAL4 and HSPA1 L lines were obtained from the Bloomington Drosophila Stock Center. Transgenic flies were generated through cloning of cDNAs encoding HTTex1 Q17 and HTTex1 Q97 into pUAST-attB-rfA (provided by Prof. S. Sigrist, Freie Universitat, Berlin) and subsequent site- directed insertion on the third chromosome (68E) using the PhiC31 integrase [Rainbow Transgenic Flies Inc. (Camarillo, CA, USA)]. All Drosophila strains were cultured on standard medium at 25°C and 65% humidity with a 12 h light- dark cycle. Expression of transgenes was induced by culturing flies on standard medium containing 400 mM RU486.

C. elegans strains and maintenance

C. elegans Q35 AM140 (rmls132 (unc-54p::Q35::YFP)) were grown on NGM plates seeded with the E. coli OP50 strain at 20 °C. Nematodes were transferred to fresh wells or plates every day in the course of the experiment to separate them from their progeny.

METHOD DETAILS

Cloning of expression vectors

For the construction of plasmids encoding CyPet- and YPet-tagged HTTEx1 Q48 fusion proteins, the coding sequence of HTTEx1 Q48 was PCR- amplified from pGEX-6P1 -HTTEx1Q48 using the primers 5’- qacqacqaattcatqqcqaccctq-3’ (SEQ ID NO.: 37) and 5’- qacqacctcqaq tggtcggtgcagcgg-3’ (SEQ ID NO.: 38). The resulting PCR product was digested with the restriction enzymes EcoRI and Notl. Additionally, CyPet cDNA was PCR amplified from pBAD33-CyPet-His (Addgene plasmid #14030) (Nguyen and Daugherty, 2005) with the primers 5’- acqacctcqaqqqtqqcqqtqqcqqtatqtctaaaqqtqaaqaattattcqq-3’ (SEQ ID NO.: 39) and 5’-qacqacqcqqccqcttatttqtacaattcatccataccatq-3’ (SEQ ID NO.: 40). YPet cDNA was amplified from pBAD33-YPet-His (Addgene plasmid #14031 ) (Nguyen and Daugherty, 2005) with the primers 5’- qacqacctcqaqqqtqqcqqtqqcqqtatqtctaaaqqtqaaqaattattcactqq-3� � (SEQ ID NO.: 41 ) and 5’-qacqacqcqqccqcttatttqtacaattcattcataccctcq-3’ (SEQ ID NO.: 42). The resulting PCR fragments were cloned into the plasmids pGEX-6P1 using the EcoRI/Xhol/Notl restriction sites to obtain plasmids pGEX-6P1 -HTTExl Q48- CyPet and -YPet, respectively. To generate the plasmids encoding GST- Ex1 Q23-CyPet and -YPet or GST-Ex1 Q23 the coding sequence of HTTEx1 Q23 was PCR-amplified from pDONR221 -HTTEx1Q23 using the primers 5’- gacgacgaattcatggcgaccctg-3’ (SEQ ID NO.: 43) and 5’-gacgacgcggccgcct cgagtggtcggtgcagcgg-3’ (SEQ ID NO.: 44) (GST-Ex1 Q23-CyPet and -YPet) or 5’-gacgacgaattcatggcgaccctg-3’ (SEQ ID NO.: 45) and 5’-gacgacgcggccgcct cgagttatggtcggtgcagcgg-3’ (SEQ ID NO.: 46). The resulting PCR products were digested using EcoRI and Xhol endonucleases and cloned into the plasmids pGEX-6P1 -HTTExl Q48-CyPet, -YPet or pGEX-6P1 -HTTEx1 Q48 after excision of HTTExl Q48 fragments by EcoRI/Xhol endonucleases.

To generate the plasmids encoding GST-K2Q48P6-CyPet and -YPet the cDNA 5’-GGgatccAAgAAAcagcagcagcagcagcagcagcagcagcagcagcagcagcag cagcagc agcagcagcagcagcagcagcagcagcagcagcagcagcagcagcagcagcagcagcagc agcag cagcagcagcagcagcagcagcagcaacagccgccaccgccgccgccgctcgag-3’ was generated by gene synthesis and subcloned into the plasmids pGEX-6P1 - HTTExl Q48-CyPet, -YPet using endonucleases BamH1 and Xho1.

To generate the plasmids encoding GST-AN17Q48+6PRD-CyPet and -YPet the cDNA 5’-GGgatcccagcagcagcagcagcagcagcagcagcagcagcagcagcagcagcag cagcagcagcagcagcagcagcagcagcagcagcagcagcagcagcagcagcagcagcag cagca gcagcagcagcagcagcagcagcagcaacagccgccaccgccgccgccgccgccgccgcc tcctccac cgccgccgccgccacagcttcctcagccgccgccgcaggcacagccgctgctgcctcagc tgcagccgcc cccgccgccgcccccgccgccacccggcccggccgcggctgaggagccgctgcaccgacc actcgag- 3’ was generated by gene synthesis and subcloned into the plasmids pGEX- 6P1 -HTTEx1 Q48-CyPet, -YPet using endonucleases BamH1 and Xho1.

To generate the plasmids encoding GST-AN17Q40+6PRD-CyPet and -YPet the cDNA 5’-GGgatcccagcagcagcagcagcagcagcagcagcagcagcagcagcagcagcag cagcagcagcagcagcagcagcagcagcagcagcagcagcagcagcagcagcagcagcag cagca gcaacagccgccaccgccgccgccgccgccgccgcctcctccaccgccgccgccgccaca gcttcctcag ccgccgccgcaggcacagccgctgctgcctcagctgcagccgcccccgccgccgcccccg ccgccaccc ggcccggccgcggctgaggagccgctgcaccgaccactcgag-3’ was generated by gene synthesis and subcloned into the plasmids pGEX-6P1 -HTTEx1 Q48-CyPet, - YPet using endonucleases BamH1 and Xho1.

Recombinant protein expression

The proteins GST-Ex1 Q23, -Ex1 Q48, -Ex1 Q23-CyPet, -Ex1 Q23-YPet, - Ex1 Q48-CyPet, -Ex1 Q48-YPet, -K2Q48P6, -K2Q48P6-CyPet, -K2Q48P6-YPet, -DN17Q48+6PRD, -AN17Q48+6PRD-CyPet, -AN17Q48+6PRD-YPet, - AN17Q40+6PRD, -AN17Q40+6PRD-CyPet, and -AN17Q40+6PRD-YPet were produced in E. coli BL21 -CodonPlus-RP and affinity-purified on glutathione- sepharose beads. Purified proteins were dialyzed over night at 4 °C against 50 mM Tris-HCI pH 7.4, 150 mM NaCI, 1 mM EDTA and 5% glycerol, snap-frozen in liquid N ₂ and stored at -80 °C. Protein concentrations were determined with a NanoDrop spectrophotometer. Prior to use, protein solutions were ultra- centrifuged at 190,000 x g for 40 min to remove aggregated material a- Synuclein (a-Syn) was produced in E. coli BL21 (DE3) and monomeric a-Syn was purified as described elsewhere (Theillet et al., 2016). Expression of Tau40 protein was performed in E. coli BL21 using a 50 I bioreactor. After cell disruption using a French press, Tau40 protein was purified via cation exchange chromatography and gel filtration. Expression and purification of Tau were performed by InVivo BioTech Services (Hennigsdorf, Germany) using proprietary company protocols. Preparation of in vitro seeds

Spontaneous Ex1 Q48 aggregation was initiated by addition of 14 U PreScission protease (GE Healthcare) per nmol purified GST-Ex1Q48 fusion protein (2 mM). The aggregation reaction was performed in 50 mM Tris-HCI pH 7.4, 150 mM NaCI, 1 mM EDTA and 1 mM DTT at 25 °C and constant agitation (450 rpm) for 24 h. Ex1 Q23 protein for seeding experiments was prepared from GST-Ex1 Q23 fusion protein using the same protocol. K2Q48P6 protein for seeding experiments was prepared from GST-K2Q48P6 fusion protein using the same protocol. AN17Q48+6PRD protein for seeding experiments was prepared from GST-AN17Q48+6PRD fusion protein using the same protocol. AN17Q40+6PRD protein for seeding experiments was prepared from GST-AN17Q40+6PRD fusion protein using the same protocol. Synthetic human IAPP was aggregated as described previously (Gao et al 2015). Lyophilized a-Syn was dissolved in PBS at 500 mM and centrifuged (4 °C, 265.000 x g) after a 5 min sonication step to remove aggregated material. The supernatant was incubated for 7 d at 37 °C under constant shaking in the presence of a single glass bead. Tau40 was aggregated for 6 d at 37 °C under constant shaking in 100 mM sodium acetate, pH 7.4, and 1 mM DTT in the presence of heparin. Synthetic human Abi _-42 was dissolved in 100 mM NaOH and diluted to 200 pM in low salt buffer (10 mM K ₃ P0 ₄ , pH 7.4, 10 mM NaCI). Aggregation was performed for 6 h at 37 °C under constant agitation.

Atomic force microscopy

Aliquots of 15 pi aggregation reactions (24 h) were spotted onto freshly cleaved mica glued to a microscope slide. After incubation for 30 min to allow adsorption, samples were rinsed 4 times with 40 pi distilled water and dried over night at RT. Samples were imaged with a digital multimode Nanowizard II (JPK, Germany) atomic force microscope operating in intermittent-contact mode.

Filter retardation assays

FRAs were essentially performed as described previously (Wanker et al., 1999). Briefly, equal volumes of 500 ng of Ex1 Q48 aggregates and 4 % SDS solution with 100 mM DTT were mixed and boiled at 95 °C for 5 min. By applying vacuum, samples were filtered through a cellulose acetate membrane with 0.2 pm pores (Schleicher and Schuell, Germany) and washed twice with 100 pi 0.1 % SDS. For analysis of tissue homogenates, 60 pg of total protein for mouse brain and 75 pg of total protein for Drosophila heads were filtered per dot. Membranes were blocked with 5% skim milk in PBS/0.05 % Tween20 (PBS-T) for at least 30 min. Aggregates retained on the membrane were detected using GFP, N18, MW8, Mab5492 or HD1 antibody followed by an appropriate peroxidase-coupled secondary antibody. Signals were quantified using the AIDA image analysis software (Raytest, Germany).

Dot blot assays

To estimate total FITT protein, native dot blot (DB) assays were performed as described previously (Kayed et al., 2003). Briefly, 250 ng of Ex1Q48 protein were filtered onto a nitrocellulose membrane and blocked with 5% skim milk in PBS-T. For detection, the membrane was incubated with HD1 antibody followed by an appropriate peroxidase-coupled secondary antibody. Signals were quantified using the AIDA image analysis software (Raytest, Germany).

Native gels

Protein solutions were mixed with sample buffer and loaded onto a Novex NativePAGE 3-12% Bis-Tris gradient gel (Life Technologies). NativePAGE and immunoblotting were performed according to manufacturer recommendations. Ex1 Q48 aggregates were visualized as for SDS-PAGE.

SDS-PAGE and Western blotting

Samples of aggregation reactions were mixed with loading buffer (50 mM Tris- HCI pH 6.8, 2% SDS, 10% glycerol and 0.1 % bromophenol blue) and boiled at 95 °C for 5 min. Samples were loaded onto Novex NuPAGE 4-12% Bis-Tris gradient gels (Life Technologies). SDS-PAGE and immunoblotting were performed according to manufacturer recommendations. Ex1 Q48 distribution was visualized by N18 antibody (Santa Cruz) or Ex1 Q48-CyPet/-YPet with a GFP antibody (Abeam) followed by appropriate peroxidase labeled secondary antibodies. Genotyping of Drosophila strains

Total genomic DNA from transgenic flies was extracted using the DNeasy® Blood & Tissue Kit (Qiagen). cDNAs encoding HTTex1 Q17 and HTTex1 Q97 were PCR amplified using the Pwo DNA polymerase (Roche) and the primers 5’-aaccccgtaaatcaactgc- 3’ and 5’-atctctgtaggtagtttgtc-3’). The sizes of the resulting PCR products were analyzed by agarose gel electrophoresis.

Quantitative polymerase chain reaction (qPCR)

RNA was extracted from Drosophila heads using TRIzol™ Reagent (Invitrogen). cDNA was synthesized using M-MLV Reverse Transcriptase (Thermo Scientific) and qPCR was performed using the SYBR Green PCR Master Mix (Thermo Scientific). Primer pairs for HTT (sense, 5'-gacctggaaaagctgatga-3’ and antisense 5'-tcatggtcggtgcagcggct-3'), and control primers for rp49 (sense 5'- tacaggcccaagatcgtgaa-3', and antisense 5'-acgttgtgcaccaggaactt-3') were utilized. SYBR Green analysis was performed using the ViiA7 Real -time PCR system (Thermo Scientific). The amount of mRNA detected was normalized to control rp49 mRNA values. Viability analysis of adult Drosophila melanogaster

Viability assays were performed with elavGS;HTTex1 Q17, elavGS;HTTex1 Q97 and elavGS;HSPA1 L;HTTex1 Q97 transgenic flies by quantification of lethality of at least 100 females of each genotype and expression condition in three independent biological replicates. Flies were aged at 25°C, with 10 flies per vial and were transferred every 3-4 days. Median lifespan (age at which half of the tested population has died) was calculated by fitting survival curves to the log(inhibitor) vs. normalized response (variable slope) equation using GraphPad Prism. Statistical significance was assessed by one-way ANOVA followed by Dunnett’s multiple comparison post hoc test. ^* , p<0.05; ^** , p<0.01 ; ^*** , p<0.001 .

Analysis of motor performance (climbing assay)

Ten female flies were placed in a closed empty vial and gently tapped to the bottom of the vial. The percentage of flies that climbed 8 cm within 15 s was recorded. Flies were aged at 25°C (10 flies per vial) and were monitored and transferred twice a week. Motor performance was assessed for elavGS;HTTex1 Q17 and elavGS;HTTex1 Q97 flies expressing the HTT transgenes for the indicated times. 100 females of each genotype and expression condition in each of the three biological replicates were investigated.

Preparation of Drosophila head lysates for FRAs

Drosophila head lysates were produced by homogenizing fly heads in 2 % SDS and complete protease inhibitor cocktail using a micro pestle. Lysates were centrifuged for 10 min at 8,000 rpm (4°C). The supernatant was transferred to a new tube and total protein concentration was determined with a Pierce™ BCA assay using BSA as a standard.

Dissection and immunostaining of Drosophila adult brain

The whole brains of adult flies were dissected in ice-cold haemolymph-like saline (HL3) solution (Stewart et al., 1994), fixed for 20 min in 4 % paraformaldehyde (PFA) in PBS and permeabilized in PBS-T (1 % Triton™ X- 100) for 20 min at RT. Samples were blocked in 10 % normal goat serum (NGS) in PBS-T (0.3 % Triton™ X-100) for at least two hours. Brains were incubated with the indicated primary antibody (1 :500) in brain staining buffer (5 % NGS, 0.1 % NaN ₃ in PBS-T (0.3 % Triton™ X-100)) for 48 hours at 4 °C. Subsequently, brains were washed in PBS-T (0.3 % Triton™ X-100) for 24 hours at 4°C with multiple buffer exchanges. Next, samples were incubated with appropriate secondary antibody in brain staining buffer for 24 hours at 4 °C, washed six times for 30 min in PBS-T (0.3 % Triton™ X-100) at RT and stored in VectaShield H-100 (Vector Laboratories) at least for one day at -20 °C. Brains were mounted and imaged using the Leica TCS SP8 Confocal Microscope. Images were analyzed using Fiji.

RNA interference

For synchronization, gravid adults from one 10 cm NGM plate were collected in a canonical tube and treated with 20% alkaline hypochlorite solution under vigorous agitation for 4 min. The eggs were then washed three times with cold 0.1 M NaCI solution. The eggs were allowed to hatch in M9 medium at 20 °C for 22 h. Animals were then placed as L1 larvae onto RNAi plates that were seeded with E. coli expressing dsRNAi against hsp-1 or the empty vector L4440 (control).

Fluorescence microscopy

The aggregation propensities of Q35-YFP were analyzed throughout adulthood. Animals were subjected to RNAi treatment from the first larval stage on and maintained on RNAi plates throughout the experiment. For imaging, nematodes were mounted onto 2 % agarose (Sigma) pads on glass slides and immobilized with 2 mM Levamisole (Sigma). Images were taken on a Zeiss LSM780 confocal microscope at 20x magnification. The Q35-YFP expressing nematodes were analyzed as whole nematode for quantification of the aggregates and an image was taken of the head region of every animal. 20 animals were analyzed for each condition.

Motility assay

Nematodes were transferred from liquid culture onto a blank (unseeded) NGM plate and allowed to acclimate for 15 min. The movement of the animals was digitally recorded at 20 °C using a Leica M165FC microscope with a DFC3000G digital camera and the Leica LASX Software. Movies of 10 s were captured at 10 frames/s. Animals that crossed each other or those that escaped from the field of view were excluded from analysis. 20 animals were analyzed for each condition. Captured frames were merged into ^* .avi format, imported into Fiji (Schindelin et al., 2012) and analyzed using the wrMTrck plugin (http://www.phage.dk/plugins). The average speed of each animal was calculated by dividing its body length by the duration of each track (body length per second).

Tissue homogenization

Frozen brain tissue was cut on dry ice, weighed and homogenized in a 10-fold excess (w/v) of ice-cold 10 mM Tris-HCI pH 7.4, 0.8 M NaCI, 1 mM EDTA, 10% sucrose, 0.25 U/mI benzonase and complete protease inhibitor cocktail with a dounce homogenizer. The homogenate was incubated for 1 h at 4 °C on a rotating wheel and centrifuged for 20 min at 2,700 x g (4 °C) to remove cell debris. Drosophila heads were processed comparably using 10 m I of ice-cold 10 mM Tris-HCI pH 7.4, 0.8 M NaCI, 1 mM EDTA, 10% sucrose and a complete protease inhibitor cocktail per fly head. Homogenates were centrifuged for 10 min at 8,000 rpm (4 °C). After centrifugation, supernatants were transferred to a new tube and total protein concentration was determined with a Pierce™ BCA assay using BSA as a standard. For FRASE analysis, 0.8-5 pg total protein per replicate were applied.

Electron microscopy

Total brain homogenate was centrifuged at 18,000 x g at 4 °C for 20 min; the resulting supernatant was pelleted by ultra-centrifugation at 190,000 x g for 40 min and resuspended in 10 mM Tris-HCI (pH 7.4). Immunolabeling was performed with minor modifications as described (Laue, 2010). Briefly, samples were incubated on formvar-coated copper grids (Plano) for 10 min before immunolabeling. Grids were blocked and washed in PBS supplemented with 1 % BSA and 0.1 % glycine. Labeling was performed with the anti-HTT aggregate antibody AGG and an appropriate 12 nm colloidal gold -labeled secondary antibody (Jackson ImmunoResearch). Samples were stained with 2% uranyl acetate and imaged with a Zeiss EM 910 transmission electron microscope at 80 kV. Acquisition was performed with a CDD camera (Quemesa, Olympus Viewing System).

Immunodepletion of HTTexl aggregates from mouse brain extracts

Protein G-coupled magnetic beads (Life Technologies) were incubated with 4 pg MW8 (Developmental Studies Hybridoma Bank, DSHB) or IgG isotype control (Invitrogen) antibody, respectively, for 10 min at RT to allow antibody binding. Free binding sites were saturated with Pierce protein-free blocking solution according to manufacturer recommendations. 500 pg brain homogenate in brain lysis buffer were incubated with antibody coupled beads for 3 h at 4 °C under constant overhead rotation. Subsequently, aliquots from the supernatants were taken and analyzed with the FRASE assay.

FRASE assay

Purified GST-Ex1 Q48-CyPet and GST-Ex1 Q48-YPet were diluted in aggregation buffer at an equimolar ratio to a final concentration of 1.2 pM (0.6 mM each) with 14 U PSP per nmol sensor proteins if not stated otherwise. The solution was then mixed with preformed aggregates of Ex1 Q48 (seeds) at varying concentrations with or without prior sonication and transferred to a black 384-well plate (with a final reaction volume of 30 pi per well and a sensor protein concentration of 1.2 mM). For quantification of seeding -competent HTT species in tissue samples, the sensor-protein mixture was supplemented with up to 10% (v/v) tissue homogenate. Fluorescence signals were measured every 20 min following a 5 s pulse of vertical shaking with a Tecan M200 fluorescence plate reader at 25 °C for up to 48 h. CyPet donor fluorescence was measured at excitation (Ex): 435 nm/emission (Em): 475 nm; YPet acceptor fluorescence at Ex: 500 nm/Em: 530 nm; the FRET channel (DA) was recorded at Ex: 435 nm/Em: 530 nm. Raw signals were processed by subtracting the background fluorescence of unlabeled Ex1 Q48 in all channels. Signals in the FRET channel were corrected for donor bleed-through (c _D ) and acceptor cross excitation (c _A ) using donor- and acceptor-only samples to obtain sensitized emission. Finally, sensitized emission was normalized to the acceptor signals (Jiang and Sorkin, 2002). In brief, the FRET efficiency E (in %) was calculated as follows: E = (DA- C _D XDD-C _A XAA)/AA with DD = donor channel signal and AA = acceptor channel signal.

Quantification of mutant HTT seeding activity (HSA)

Seeding effects (At ₅₀ [h]) were quantified by subtracting the t ₅₀ values (time at half-maximal FRET efficiency) of the respective sample from the negative control. To obtain the t ₅₀ values, the aggregation kinetics were curve fitted by Richard’s five-parameter dose-response curve using GraphPad Prism.

QUANTIFICATION AND STATISTICAL ANALYSIS

Statistical parameters including the exact value of n, the definition of center, dispersion and precision measures (mean ± SEM or mean ± SD) as well as the statistical analysis chosen and statistical significance are reported in the figures and figure legends. Data is judged to be statistically significant when p < 0.05 by the indicated statistical test. In figures, asterisks denote statistical significance as calculated by Student’s t test ( ^* , p < 0.05; ^** , p < 0.01 ; ^*** , p < 0.001 ). Statistical analysis was performed in GraphPad PRISM 7.

The invention is further illustrated by the following examples.

EXAMPLES

Self-propagating, amyloidogenic mutant huntingtin (mHTT) aggregates may drive progression of Huntington’s disease (HD). Here, we report the development of a FRET-based mHTT aggregate seeding (FRASE) biosensor assay that enables the quantification of mHTT seeding activity (HSA) in complex biosamples from HD patients and disease models with high sensitivity and specificity. Application of the FRASE assay revealed HSA in disease- affected brain tissues of HD patients and mouse models, e.g. brain homogenates of presymptomatic HD transgenic and knock-in mice, and its progressive increase with phenotypic changes, suggesting that HSA quantitatively tracks disease progression. Biochemical investigations of mouse brain homogenates demonstrated that small rather than large mHTT structures are responsible for the HSA measured in FRASE assays. Finally, we assessed the neurotoxicity of mHTT seeds in an inducible Drosophila model transgenic for HTTexl. We found a strong correlation between HSA measured in adult neurons and the increased mortality of transgenic HD flies, indicating that FRASE assays detect disease-relevant, neurotoxic, mHTT structures with severe phenotypic consequences in vivo.

Example 1 : Establishment of a FRET-based mHTT aggregate seeding assay

To monitor mHTT seeding activity, we first developed a cell-free aggregation assay with recombinant fluorescent reporter proteins (Figure 7A). Two soluble glutathione S-transferase HTT exon-1 (HTTexl ) fusion proteins with 48 glutamines C-terminally fused to CyPet or YPet (GST-Ex1 Q48-CyPet or -YPet) were produced in E. coli and purified to -90% homogeneity using glutathione sepharose chromatography (Figure 7B). Recombinant proteins were cleaved with PreScission protease (PSP) to release GST and to initiate the spontaneous aggregation of the fusion proteins Ex1 Q48- CyPet and -YPet. The assembly of the tagged Ex1 Q48 proteins into insoluble aggregates over time was monitored using an established filter retardation assay (FRA), which specifically detects large SDS-stable mHTT aggregates (Wanker et al., 1999). We found that the proteins Ex1 Q48-CyPet and -YPet rapidly self-assemble into SDS-stable aggregates in vitro (Figure 1A), confirming recently reported results (Wagner et al., 2018). To investigate the morphology of spontaneously formed Ex1 Q48-CyPet and -YPet aggregates, we analyzed the aggregation reactions with atomic force microscopy (AFM). We observed that the tagged Ex1Q48 fusion proteins, similar to the untagged Ex1 Q48 protein (Wagner et al., 2018), form large fibrillar protein aggregates (Figure 1 B).

We hypothesized that co-aggregation of CyPet- and YPet-tagged FITTexl fragments should lead to a time-dependent increase of FRET as the fluorescent tags are brought in close proximity when fibrillar aggregates are formed (Figure 1C). We treated mixtures of fusion proteins (1 :1 molar ratio; 1 - 3 mM concentrations) with PSP and quantified the spontaneous formation of Ex1 Q48- CyPet/-YPet co-aggregates by repeated FRET measurements. We observed a time- and concentration-dependent increase of FRET efficiency (Figure 1 D), indicating that FRET measurements are suitable to quantify HTTexl co- aggregation. In contrast, no time-dependent increase of FRET efficiency was observed in samples that were not treated with PSP , underlining that the removal of the GST tag from CyPet- and YPet-tagged Ex1Q48 fragments is critical for the self-assembly of co-aggregates. Proteolytic cleavage of the GST fusion proteins with PSP was confirmed by SDS-PAGE and immunoblotting (Figure 7C). Finally, AFM analysis confirmed that the samples indeed contain typical fibrillar HTTexl co-aggregates (Figure 7D).

In order to assess whether preformed Ex1 Q48 fibrils can seed the co- aggregation of Ex1 Q48-CyPet/-YPet, we incubated a 1 :1 mixture of the GST fusion proteins with PSP and different amounts of preformed Ex1 Q48 fibrils as seeds. We observed that addition of fibrils shortens the lag phase of Ex1 Q48- CyPet/-YPet polymerization in a concentration -dependent manner (Figure 1 E and 1F), indicating that they possess seeding activity. We termed the established method, which permits the quantification of in samples of interest, (FRASE) assay.

In independent control experiments, we also investigated whether a mixture of fusion proteins with non-pathogenic polyQ tracts, GST-Ex1 Q23-CyPet/-YPet (Figure 7A), can be applied as reporter molecules to monitor HSA. We found that preformed, fibrillar Ex1 Q48 seeds do not induce FRET when they are added to PSP treated GST-Ex1 Q23-CyPet/-YPet fusion proteins (Figure 7E). Finally, we confirmed that addition of proteolytically cleaved GST-Ex1 Q23 fusion protein as well as of uncleaved GST-Ex1 Q23 or GST-Ex1 Q48 fusion proteins do not shorten the lag phase of Ex1 Q48-CyPet/-YPet polymerization (Figure 7F).

Example 2: Both small and large Ex1Q48 fibrils exhibit HSA in FRASE assays Our initial experiments indicate that large bundles of Ex1 Q48 fibrils (~1 -2 miti in length; Figure 1 B) possess HSA (Figure 1E and 1 F). We next investigated whether such an activity can also be detected, when small fibrillar Ex1 Q48 seeds are added to FRASE assays. We sonicated large preformed Ex1 Q48 fibrils for different periods of time and subsequently determined the HSA. We found that seeding activity is high in sonicated Ex1 Q48 preparations (Figure 8A and 8B), indicating that besides large also small Ex1 Q48 fibrils possess HSA. To confirm that indeed small fibrils are produced, we analyzed the generated samples by FRA (Wanker et al., 1999). We detected large Ex1 Q48 aggregates in non-sonicated samples (Figure 8C), while they were not observed in sonicated samples (>30 sec). This suggests that sonication (>30 sec) leads to fibril breakage and the formation of small HTTexl structures that are no longer retained on filter membranes. Next, the samples were analyzed by dot blot (DB) assays, which allow the identification of protein assemblies on filter membranes independent of their size (Kayed et al., 2003). These experiments revealed Ex1 Q48 immunoreactivity in both sonicated and non-sonicated samples (Figure 8C), confirming the presence of FITT protein in all samples. Finally, we analyzed the generated samples with AFM, confirming that small fibrillar Ex1 Q48 structures are produced by sonication (Figure 8D).

Example 3: FRASE assays detect HSA with high sensitivity and specificity

To investigate the sensitivity and specificity of FRASE assays, we generated recombinant Ex1 Q48 seeds by sonication and analyzed them by blue native PAGE and immunoblotting. We found that sonication for 60 sec leads to the formation of Ex1 Q48 structures with an average molecular weight of -1 ,250 kDa (~90mers) (Figure 2A), while aggregates with a much larger in size were detected in non-sonicated samples.

Next, a large range of concentrations of sonicated Ex1Q48 seeds were analyzed for their activity in FRASE assays. As expected, we observed a dose- dependent shortening of the lag phase when Ex1 Q48 structures were added to polymerization reactions (Figure 2B and 2C). We determined a threshold of -60 fM for detecting Ex1 Q48 seeds. Furthermore, FRASE assays responded quantitatively to seeds over a dynamic range of 4 orders of magnitude (Figure 2C). At a concentration of -560 fM the Z’ factor (Zhang et al., 1999) was 0.67 (Figure 2C). Finally, we investigated the specificity of the FRASE assay for detecting FITTexl aggregates. We produced fibrillar a-synuclein, tau, amyloid-b and IAPP aggregates in vitro and subsequently analyzed them in FRASE assays. The unrelated fibrillar aggregates did not significantly influence Ex1 Q48-CyPet/-YPet polymerization (Figure 2D), indicating that the FRASE assay specifically detects amyloidogenic HTTexl aggregates. AFM analysis confirmed that fibrillar a-synuclein, tau, amyloid-b and IAPP aggregates were added to reactions (Figure 2E). Example 4: HSA is detectable in brains of HD mice and patients

To investigate whether FRASE assays detect HSA in complex biosamples (Figure 3A), we first assessed brain homogenates prepared from 12-week-old R6/2Q212 transgenic mice (carrying -212 CAGs) and age-matched controls. R6/2Q212 mice express low levels of the human HTTex1 Q212 protein (Sathasivam et al., 2013), show motor abnormalities from 8 weeks of age (Carter et al., 1999) and typical HTTexl inclusion bodies from 3-4 weeks onwards (Li et al., 1999). We detected high levels of HSA in brain homogenates of R6/2Q212 mice but not in those of age-matched littermate controls (Figures 3B and 3C), indicating the presence of seeding-competent HTTexl structures. Independent control experiments with the non -pathogenic reporter molecules Ex1 Q23-CyPet/-YPet did not reveal detectable HSA in R6/2Q212 brain homogenates (Figure 9A).

Next, we assessed whether HSA is detectable in brain extracts of 12-week-old R6/2Q51 (Larson et al., 2015) mice, which express a mutant HTTexl Q51 fragment. In comparison to R6/2Q212 mice, these mice do not yet have a disease phenotype at 12 weeks of age, suggesting that HSA should be lower. FRASE analysis revealed that brain homogenates of prodromal 12-week-old R6/2Q51 mice do not possess significant HSA (Figure 9B), while activity was detectable in extracts of very old R6/2Q51 mice (104-105 weeks), which show pathological signs of disease.

We also investigated whether HSA is detectable in the hypothalamus of mouse brains, in which the proteins HTT853-Q79 or HTT853-Q18 were overexpressed for 8 weeks using viral vectors. Previous studies have demonstrated that hypothalamic expression of HTT853-Q79 leads to a gain of body weight and the formation of insoluble HTT protein aggregates (Hult et al., 2011 ). We found that HTT853-Q79 mice were significantly heavier than control and HTT853-Q18 mice (Figure 9C). Furthermore, we also observed a significantly higher HSA in brain homogenates of HTT853-Q79 compared to HTT853-Q18 and controls mice (Figure 9D), indicating that seeding activity and alterations in body weight are associated. We next examined HSA in brain regions of HD patients. Protein extracts prepared from postmortem tissue (cerebral cortex, caudate nucleus and cerebellum) from HD patients and control individuals were systematically analyzed using the FRASE assay. HSA was invariably detected in HD but not in control samples (Figure 3D), indicating that the method is suitable to discriminate between patients and healthy individuals. Interestingly, HSA was detectable in the cerebral cortex and the caudate nucleus, which are severely affected in HD patients (Zuccato et al., 2010) while it was not observed in the cerebellum, which is less affected in disease (DiFiglia et al., 1997). Similarly, no HSA was detectable in postmortem brains of patients with Alzheimer’s disease (AD) that do not contain abnormal polyQ aggregates (Figure 9E).

Finally, we investigated whether HSA in biosamples indeed originates from mHTT seeds. We produced brain extracts from symptomatic 12-week-old R6/2Q212 mice and littermate controls and immunodepleted potential seeding - competent mHTTexl seeds using the monoclonal anti-HTT antibody MW8 (Ko et al., 2001 ). Then, samples were analyzed using FRASE assays. We observed a dramatic decrease of HSA in MW8-immunodepleted R6/2Q212 brain homogenates but not in homogenates treated with an isotype control antibody (Figures 9F and 9G), indicating that antibody treatment removes seeding- competent mHTTexl aggregates from mouse brain extracts. As expected, we did not detect HSA in crude brain extracts of age-matched wild-type control mice. SDS-PAGE and immunoblotting confirmed depletion of mHTTexl protein aggregates from brain homogenates by MW8 antibody treatment (Figures 9H and 91).

Example 5: FRASE assay detects HSA in brains of presymptomatic HD mice

To address whether HSA is detectable in brains of presymptomatic HD mice, we first analyzed non-sonicated brain homogenates of young R6/2Q212 mice and age-matched controls using the FRASE assay. We detected significant HSA in brain extracts of 2-week-old R6/2Q212 mice (Figure 3E) that progressively increased over time. A similar result was also obtained with sonicated brain extracts (Figure 3F). With sonication, significant HSA was already detectable in brains of 1 -day-old R6/2Q212 transgenic mice, indicating that seeding-competent mHTTexl structures are present in brains of R6/2Q212 mice long before inclusion bodies or motor abnormities can be detected (Davies et al., 1997; Zuccato et al., 2010).

Next, we investigated whether HSA is detectable in presymptomatic HdhQ'\ 50 knock-in mice that express a full-length HTT protein with a pathogenic polyQ tract of 150 glutamines (Lin et al., 2001 ). These mice show onset of depressive- like symptoms by 12 months of age (Ciamei et al., 2015) and impairment of motor function at ~18 months of age. Widespread deposition of mHTT aggregates throughout the brain is observed by 8 months of age (Woodman et al., 2007). We systematically analyzed tissue homogenates prepared from cortex, striatum and hippocampus of 2-, 5- and 8-month-old heterozygous HdhQ'\ 50 mice and littermate controls using the FRASE assay. We observed progressively increasing HSA in protein extracts from all three brain regions of HdhQ'\ 50 but not from control mice (Figure 3G), confirming that mHTT seeds are detectable in HD mouse brains long before the appearance of neuronal inclusion bodies and motor abnormalities (Woodman et al., 2007).

Example 6: HSA is detectable in protein fractions after depletion of large mHTTexl aggregates by centrifugation

To investigate whether HSA in HD mouse brains originates predominantly from soluble or insoluble mHTTexl aggregates, non-sonicated brain homogenates prepared from symptomatic 12-week-old R6/2Q212 mice were centrifuged for 20 min at 2,700 x g (low speed) or 18,000 x g (medium speed), respectively, and the resulting supernatant and pellet fractions (S1 _L0 w, P1 LOW and S1 _Med , P1 _Med ! Figure 4A) were analyzed with FRASE assays. Interestingly, HSA was high in the parental crude lysate and in the S1 _LOW fraction, while it was relatively low in the P1 _Low fraction (Figure 4B), suggesting that it predominantly originates from soluble rather than insoluble mHTTexl aggregates. A similar result was obtained when the fractions S1 _M e _d and P1 _M e _d were analyzed (Figure 4B). However, after medium speed centrifugation HSA in the P1 Med fraction was higher than in the P1 _L0w fraction, indicating that mHTTexl seeds can be removed from supernatant fractions using a higher centrifugation speed. This trend was even more pronounced when the generated S1 _Med fraction was subjected to a high-speed centrifugation (190,000 x g), resulting in the supernatant and pellet fractions S2 and P2 (Figure 4A). FRASE analysis revealed a significantly higher HSA in the P2 than in the S2 fraction, indicating that small seeding-competent mHTTexl aggregates can be removed from the soluble S1 Med fraction by high-speed centrifugation (Figures 4A and B).

To obtain a first hint about the size of the seeding-competent mHTTexl aggregates in the brains of R6/2Q212 mice, the supernatant and pellet fractions were analyzed by FRA (Wanker et al., 1999). We found mHTTexl immunoreactivity predominantly in the P1 _LOW and Pl _Med fractions. In comparison, weak or no immunoreactivity was detected in the fractions S1 _L0w , S1 _Med , P2 and S2 (Figure 4C), suggesting that HSA in R6/2Q212 mouse brain extracts predominately originates from small rather than large mHTTexl protein assemblies.

Finally, we used transmission immunoelectron microscopy to assess the size and morphology of mHTTexl seeds present in P2 fractions. They exhibit high HSA in FRASE assays but does not contain large mHTTexl aggregates. We detected small, immunoreactive mHTTexl fibrils with diameters of 10.2 ± 3.6 nm and lengths of 157.8 ± 64.1 nm exclusively in P2 fractions of R6/2Q212 mice (Figure 4D), suggesting that HSA originates from such structures in P2 fractions.

Example 7: Short-time expression of HTTex1Q97 in adult neurons decreases lifespan and locomotor activity of HD flies We first confirmed that HTTexl transcripts decline in neurons when HTTexl Q97 (a protein with pathogenic polyQ tracts) expressing elavGS;HTTex1 Q97 flies, a newly established inducible HD Drosophila model, are placed back on food without the inducer (Figures 10A and 10B).

Next, we investigated whether both long- and short-time expression of HTTexl Q17 or HTTexl Q97 in adult neurons influences survival of HD flies. Starting at an age of 3 days, we treated elavGS;HTTex1 Q17 and elavGS;HTTex1 Q97 flies either continuously or only for a short time of 3 or 6 days with RU486 (Figure 5A); survival was measured by counting dead flies. We found that the lifespan of chronically RU486 treated elavGS;HTTex1 Q97 flies was significantly reduced in comparison to untreated flies (Figure 5B and 5C). In strong contrast, chronic treatment with RU486 did not shorten the lifespan of elavGS;HTTex1 Q17 flies. We calculated a median lifespan of ~30 and ~85 days for treated and untreated elavGS;HTTex1 Q97 flies, respectively. Strikingly, median lifespans of short-time (~38 and ~33 days) and chronically treated elavGS;HTTex1 Q97 flies was similar.

As a behavioral measure of neuronal dysfunction, locomotor activity of HD flies was assessed using a negative geotaxis (climbing) assay (Latouche et al., 2007). We observed that RU486-treated elavGS;HTTex1 Q97 flies show a significant decline in climbing behavior in comparison to untreated controls (Figure 5D), confirming that both short and long-time expression of HTTexl Q97 in adult neurons induces neurotoxicity in HD flies.

Example 8: Formation of small, seeding-competent HTTex1Q97 structures in adult neurons is associated with reduced survival

We first assessed the correlation between the formation of large, SDS -stable HTTexl aggregates in neurons and the survival of RU486-treated elavGS;HTTex1 Q97 flies (Figure 5B and 5C). Head lysates were prepared from continuously and short-time (3 and 6 days) RU486-treated and untreated flies (Figure 10C) and analyzed by FRA using the anti-HTT antibody MW8 (Ko et al., 2001 ). We found that the abundance of large, SDS-stable HTTex1 Q97 aggregates was high in heads of chronically RU486-treated elavGS;HTTex1 Q97 flies but relatively low in short-time treated flies (Figure 5E). A similar result was obtained when the formation of large HTTex1 Q97 aggregates in fly heads was quantified by FRAs using the anti-HTT antibody MAB5492 (Figure 10D). These results indicate that large HTTex1Q97 aggregates detected by FRAs in adult neurons cannot well predict the observed survival phenotypes, which are very similar for short- and long-time RU486- treated flies (Figure 5B and 5C).

We next investigated mFITTexl aggregate formation in brains of hormone- treated elavGS;FITTex1 Q97 flies using an immunohistochemical method . We dissected whole brains of short- and long-time treated elavGS;FITTex1 Q97 flies (Figure 10C) and incubated them with the antibody MAB5492. As a control, the brain sections were also immunoassayed with an anti-RBP (RIM-binding protein) antibody, which detects synapses in fly brains (Liu et al., 2011 ). As expected, we detected high amounts of HTTex1 Q97 aggregates (green puncta) in long-time and lower amounts in short-time (3 and 6 days) hormone-treated HD flies (Figure 10E), confirming the results obtained by FRAs (Figure 5E). Interestingly, these investigations also revealed very low amounts of HTTex1 Q97 aggregates in brains of non-induced elavGS;HTTex1 Q97 flies (Figure 10E), indicating that the elavGS expression system is leaky and very low levels of HTTex1Q97 protein are also produced in the absence of hormone treatment. However, such low expression of HTTex1 Q97 was not sufficient to significantly shorten the lifespan of HD flies (Figure 5C).

Finally, we used the FRASE assay to quantify HSA in head lysates of RU486- treated elavGS;HTTex1 Q97 flies. Strikingly, we measured high HSA in protein lysates of both short- and long-time hormone-treated flies (Figure 5F and 5G), demonstrating that FRASE assays provide information that is fundamentally different from that obtained by FRAs. As the abundance of large fibrillar aggregates is very low in protein extracts of short-time treated flies (Figure 5E and 10D), HSA in these fractions must predominantly result from smaller structures that are not retained by the filter membrane. In contrast to the FRA results (Figure 5E and 11 D), HSA levels measured with the FRASE assay (Figure 5F and 5G) correlate significantly better with the increased mortality of RU486-treated elavGS;HTTex1 Q97 flies (Figure 5H). As expected, HSA was undetectable in head lysates of 27-day-old elav;HTTex1 Q17 control flies, which constitutively express the protein HTTex1 Q17 in neurons (Figure 5F).

Example 9: Short-time expression of Hsp70 extends the lifespan of HD flies and decreases HSA in neurons We hypothesized that co-expression of the molecular chaperone Hsp70 (HSPA1L) (Chan et al., 2000) might influence HSA and neurotoxicity in HD transgenic flies. To address this question, we generated elavGS;HSPA1 L;HTTex1 Q97 flies, which upon hormone treatment co-produce both Hsp70 and HTTex1 Q97 in adult neurons. We first assessed whether in brains of RU486-treated elavGS;HSPA1 L;HTTex1 Q97 and elavGS;HTTex1 Q97 flies similar levels of HTTex1 Q97 transcripts are expressed. We treated 3-day- old flies for 6 days with RU486 (400 mM) and subsequently quantified mHTTexl transcript levels in fly heads by qPCR. Similar HTTex1 Q97 transcript levels were observed in both strains, indicating that co-expression of HSPA1L does not significantly influence mHTTexl expression in elavGS;HSPA1 L;HTTex1 Q97 flies (Figure 11 A). To confirm the expression of Hsp70 in elavGS;HSPA1 L;HTTex1 Q97 flies, we also analyzed protein extracts of hormone-treated animals by SDS-PAGE and immunoblotting. As expected, similar Hsp70 protein levels were detectable in head lysates of hormone -treated elavGS;HSPA1 L;HTTex1 Q97 and elavGS;HSPA1 L control flies (Figure 11 B), indicating that HTTex1 Q97 co-expression does not significantly influence Hsp70 production in neurons of elavGS;HSPA1 L;HTTex1 Q97 flies.

Next, we assessed whether short-time co-expression of Hsp70 (for 6 days) in adult neurons influences the survival of elavGS;HSPA1 L;HTTex1 Q97 HD flies. In control experiments, the survival of short-time RU486-treated elavGS;HTTex1 Q97 flies was analyzed. We determined median lifespans of ~39 and ~33 days for RU486 treated elavGS;HSPA1 L;HTTex1 Q97 and elavGS;HTTex1 Q97 flies, respectively, (Figure 6A-6C), indicating that short- time co-expression of Hsp70 in adult neurons improves the survival of elavGS;HSPA1 L;HTTex1 Q97 flies. As expected, a median lifespan of ~89 days was observed for non-treated elavGS;HTTex1 Q97 flies, confirming our initial results (Figure 5C).

To examine whether short-time co-expression (6 days) of Hsp70 in adult neurons influences mHTTexl aggregation, head lysates of 13-day-old RU486- treated elavGS;HSPA1 L;HTTex1 Q97 and elavGS;HTTex1 Q97 flies were analyzed by FRAs and FRASE assays. We found that the abundance of large HTTex1 Q97 aggregates and HSA both were significantly decreased in brains of hormone treated elavGS;HSPA1 L;HTTex1 Q97 flies in comparison to elavGS;HTTex1 Q97 flies (Figure 6D-6F), substantiating our hypotheses that quantification of HSA predicts survival of HD transgenic fl ies.

Finally, we addressed the question whether the molecular chaperone Hsp70 associates directly with mHTTexl aggregates in fly neurons.

Immunohistochemical investigations of brains prepared from 9-day-old elavGS;HSPA1 L;HTTex1 Q97 flies treated for 6 days with RU486 revealed partial co-localization of Hsp70 and HTTex1 Q97 aggregates (Figure 6G), supporting previous observations that Hsp70 directly targets aggregation -prone polyQ-containing HTTexl fragments (Warrick JM, 1999).

Example 10: Depletion of Hsc70 increases Q35-YFP seeding activity in a C.elegans model

Our studies in HD flies indicate that short-time overproduction of Hsp70 decreases HSA in neurons (Figure 6F), suggesting that a decrease of chaperone expression might have the opposite effects. To address this question, we performed RNAi knockdown experiments in transgenic worms that overproduce the aggregation-prone protein Q35-YFP in body wall muscle cells. Previous studies have demonstrated that Q35-YFP aggregation in these cells leads to motor impairment and that this phenotype increases in severity upon knock-down of hsp-1 (Hsc70) gene expression by RNAi (Brehme et al 2014). We treated Q35-YFP expressing worms with hsp-1 RNAi and assessed their motility at day five. We observed a significant reduction of motility in RNAi- treated in comparison to untreated worms (Figure 11C), confirming previously published results (Brehme et al., 2014). Furthermore, this phenotypic change was associated with a significant increase in Q35-YFP seeding activity measured by FRASE assays (Figure 11 D and 11E), supporting our hypothesis that FISA is a marker of dysfunction and toxicity in model systems.

Example 11 : Application of the FRASE assay for the detection of FISA in human brain tissue

Using the FRASE assay, FISA was detectable in brain tissues of various FID mice at symptomatic stage, regardless whether an N-terminal fragment of mutant FITT or the full-length protein was expressed. In addition, FISA was detectable in severely affected brain regions of FID patients but not in control individuals. This suggests that the presence of seeding-competent mFITT aggregates is a general phenomenon in FID models and patients and further implies a potential role of these structures in disease development or progression. Flowever, to be regarded as disease relevant and capable of promoting pathogenesis, such structures would need to be present early in disease development and their abundance in affected tissues should increase with the severity of disease symptoms. Initial experiments using R6/2Q212 and Hc//?Q150 mice demonstrated that mFITT seeds are detectable in FID mouse brains long before the appearance of inclusion bodies and motor abnormalities and increase in abundance with the development of disease pathology (Woodman et al., 2007). Subsequently, we asked whether similar results can be obtained with brain tissues prepared from FID patients. Based on the temporospatial pattern of degeneration in the striatum, Vonsattel et al. developed a grading system to classify the severity of neuropathological changes into five distinct grades (0 - 5) (Vonsattel et al., 1985). Longitudinal examination of FID patients prior to death showed significant correlation between clinical features and the neuropathological grade assigned postmortem (Rosenblatt et al., 2003).

Here, we analyzed brain homogenates prepared from the putamen of HD patients and control individuals. Neuropathological changes of HD patients were classified and ranged from grade 2 to grade 4. FRASE analysis demonstrates a significant elevation of HSA in brain tissue with mild neuropathological changes (Figure 12, Grade 2). With the advancement of neuropathological changes, a successive increase of HSA (Figure 12, Grade 3 and 4) was observed, indicating that HSA correlates with the severity of neuropathological changes.

Example 12: Application of the FRASE assay for the detection of HSA in human brain tissue

The FRASE assay could be further optimized regarding its sensitivity. Impeding primary nucleation and thereby delaying spontaneous self-assembly of the reporter proteins might improve the detection limit. Primary nucleation of mHTTexl depends on the concentration of the protein and the length of the polyQ tract (Li and Li, 1998; Scherzinger et al., 1999). In addition, primary nucleation is greatly influenced by the amino acid sequences flanking the polyQ domain (N17 and PRD domain). Whereas the N17 domain has been reported to facilitate aggregation, the PRD was shown to counteract this process (Bhattacharyya et al., 2006; Crick et al., 2013; Mishra et al., 2012). Therefore, the deletion of the N17 region on the one hand or the expansion of the PRD on the other hand, might be strategies to generate sensor proteins of even higher sensitivity. In addition, minimizing or removing both flanking sequences (N17 and PRD) might generate a generic seeding sensor that can detect seeding - competent aggregates associated to other polyQ diseases.

We generated three additional mHTTexl based reporter protein pairs. For the constructs GST-K2Q48P6-CyPet/GST-K2Q48P6-YPet the N17 region was replaced by two lysine residues and the proline-rich domain (PRD) was replaced by 6 proline residues. The construct contains, in consecutive order, 2 lysines, 48 glutamines and 6 prolines, and was fused N-terminally to GST and C-terminally to CyPet or YPet (GST-K2Q48P6-CyPet/GST-K2Q48P6-YPet; SEQ ID NOs: 48 and 50).

In order to generate the constructs GST-AN17Q48+6PRD-CyPet/GST- AN17Q48+6PRD-YPet the N17 region was deleted and the proline rich domain was extended by 6 proline residues. The construct contains, in consecutive order, 48 glutamines and an extended PRD, and was fused N-terminally to GST and C-terminally to CyPet or YPet (GST-AN17Q48+6PRD-CyPet/GST- DN17Q48+6PRD-YPet; SEQ ID NOs: 52 and 54).

In order to generate the constructs GST-AN17Q40+6PRD-CyPet/GST- AN17Q40+6PRD-YPet the N17 region was deleted and the proline rich domain was extended by 6 proline residues. The construct contains, in consecutive order, 40 glutamines and an extended PRD, and was fused N-terminally to GST and C-terminally to CyPet or YPet (GST-AN17Q40+6PRD-CyPet/GST- DN17Q40+6PRD-YPet; SEQ ID NOs: 56 and 58).

Using the above fusion proteins, the FRASE assay’s sensitivity was improved and thus, its applicability was broadened.

Example 13: Analyzing compound effects on seeded aggregation and seed modulation

Within the context of the above-described method for the quantification of seeding activity (Dί ₅ o) of an amyloidogenic aggregate, it is also encompassed by the invention that the preformed aggregate may be used to screen for compounds that influence seeding competence of the aggregate. For example, a given amyloidogenic aggregate C may be pretreated with a compound of interest (D) before being used in the described screening method. Then, seeding activity is determined once with a sample containing aggregate C without pretreatment, and once with a sample containing aggregate C pretreated with the compound D. Compound D may be e.g. a protein, peptide or small molecule as defined herein. The small molecule 04

has previously been shown to interfere with the aggregation of HTTex1 Q48 (Wagner et al., 2018), which makes it a promising candidate to modulate preformed aggregates and alter their seeding activity. In order to test this hypothesis we performed proof of principle experiments in which preformed, sonicated HTTex1 Q48 aggregates were incubated with 04 in a 200-fold molar excess at 25°C for 20 hours and subsequently analyzed for their seeding activity using the FRASE assay.

Pre-incubation of seeds with 04 strongly reduces HTTex1 Q48 seeding activity in comparison to HTTex1 Q48 seeds incubated with DMSO (Figure 13 A and B). In order to confirm that the reduction in FISA results from the modification of FITTex1 Q48 seed and not from the modulation of sensor protein aggregation by residual 04, sensor proteins were aggregated in the presence of 04 but in the absence of FITTex1 Q48 seeds. At this concentration, 04 has no significant effect on sensor protein aggregation, showing the reduction in FISA of pre- incubated seeds originates from the modulation of FITT aggregates.

DISCUSSION

With the described assay in hand, we assessed the potential correlation between FISA in affected tissues and the appearance of disease phenotypes in various FID models (Figures 3-6, Figure 12). We e.g., detected FISA in crude brain extracts of mice weeks before manifestation of disease (Figures 3E-3G). Furthermore, we observed an increase of mutant FISA in mouse brain extracts and homogenates prepared from putamen of FID patients concomitantly with the appearance of symptoms, suggesting that it quantitatively tracks disease progression. Finally, mechanistic studies with a newly established inducible Drosophila model of HD indicate a correlation between HSA in adult neurons and reduced survival of HD flies, supporting our hypothesis that mHTT seeding is a disease-relevant process. Taken together, these studies indicate that HSA is a valuable early disease marker that can predict severe downstream phenotypic changes in various HD models.

We observed high HSA in soluble fractions of HD mouse brain extracts (Figure 4B and 4C), suggesting that seeding activity in transgenic animals predominately originates from small rather than large mHTTexl aggregates. However, inclusions with insoluble fibrillar HTTexl aggregates (Bauerlein et al., 2017) may also possess seeding activity. Further studies will be necessary to purify fibrillar HTTexl structures of different sizes from mouse and fly brains and to compare their specific seeding activity (i.e., seeding activity per unit of protein). Our results are in agreement with previous investigations indicating that small, fibrillar polyQ-containing HTT assemblies are detectable in the cytoplasm of cells besides large inclusions with fibrillar mHTT aggregates (Sahl et al., 2012). They are also consistent with studies demonstrating that proteotoxicity in mammalian cells is associated with small, diffusible HTT oligomers rather than large inclusions (Arrasate et al., 2004; Leitman et al., 2013). However, our present study advances beyond the state-of-the-art. For the first time, we provide experimental evidence that the abundance of small seeding-competent polyQ structures correlates with dysfunction and toxicity in HD transgenic flies and worms (Figures 5, 6 and 11).

Together, our investigations suggest that HSA in HD mouse and fly brain extracts is a biological marker of disease long before its onset. Previous studies argue that the abundance of large inclusions with insoluble mHTT aggregates in brains of HD mice and patients is not predictive for the development of symptoms (Kuemmerle et al., 1999). However, neuronal inclusions are commonly detected with immunohistological methods, which fail to identify small, seeding-competent mHTT assemblies in disease brains. The application of the FRASE assay overcomes this important limitation associated with standard histology and is likely to yield new mechanistic insights into the progressive development of HD. We propose that in future drug trials with HD mice HSA could be utilized as an outcome marker to monitor the efficacy of therapeutic molecules in vivo, before and independent of changes in phenotype. As we detected robust HSA in the striatum of 2-month-old HdhQ'\ 50 knock-in mice (Figure 3G), drug treatment could start before that point in time and animals could be assessed for HSA at any later age. Furthermore, we found that the FRASE method can be applied as a drug screening assay to identify therapeutic molecules such as small molecule compounds that directly target mHTT seeding in vitro (Figure 13). As the assay can monitor mutant HSA in protein extracts from postmortem patient brain and transgenic animals (Figure 3D-3G, Figure 12), it seems now feasible to investigate aggregate-targeting therapeutic candidate molecules in assays which contain disease-relevant seeds.

Through the application of FRASE assays, we have demonstrated that HSA is a robust, early disease biomarker in HD transgenic mice and flies. We propose that it also might be of high value for monitoring disease onset and progression in HD patients if HSA could be quantified in biosamples whose collection is technically and ethically possible, like cerebrospinal fluid, blood or muscle tissue. Through the quantification of HSA in patient samples, the optimal time point for the initiation of clinical trials could be determined and the efficacy of therapeutic interventions could be monitored. In this way, our findings may help to develop novel disease-modifying therapeutic strategies for HD and other polyQ diseases.

REFERENCES

Arrasate, M., Mitra, S., Schweitzer, E.S., Segal, M.R., and Finkbeiner, S. (2004). Inclusion body formation reduces levels of mutant huntingtin and the risk of neuronal death. Nature 431, 805-810.

Atarashi, R., Moore, R.A., Sim, V.L., Hughson, A.G., Dorward, D.W., Onwubiko, H.A., Priola, S.A., and Caughey, B. (2007). Ultrasensitive detection of scrapie prion protein using seeded conversion of recombinant prion protein. Nature methods 4, 645-650.

Atarashi, R., Satoh, K., Sano, K., Fuse, T., Yamaguchi, N., Ishibashi, D., Matsubara, T., Nakagaki, T., Yamanaka, H., Shirabe, S., et al. (2011 ). Ultrasensitive human prion detection in cerebrospinal fluid by real-time quaking- induced conversion. Nature medicine 17, 175-178.

Babcock, D.T., and Ganetzky, B. (2015). Transcellular spreading of huntingtin aggregates in the Drosophila brain. Proceedings of the National Academy of Sciences of the United States of America 112, E5427-5433.

Baldo, B., Soylu, R., and Petersen, A. (2013). Maintenance of basal levels of autophagy in Huntington's disease mouse models displaying metabolic dysfunction. PloS one 8, e83050.

Bhattacharyya, A., Thakur, A.K., Chellgren, V.M., Thiagarajan, G., Williams, A.D., Chellgren, B.W., Creamer, T.P., and Wetzel, R. (2006). Oligoproline Effects on Polyglutamine Conformation and Aggregation. Journal of Molecular Biology 355, 524-535.

Biancalana, M., and Koide, S. (2010). Molecular mechanism of Thioflavin-T binding to amyloid fibrils. Biochimica et biophysica acta 1804, 1405-1412. Brehme, M., Voisine, C., Rolland, T., Wachi, S., Soper, J.H., Zhu, Y., Orton, K., Villella, A., Garza, D., Vidal, M., et al. (2014). A chaperome subnetwork safeguards proteostasis in aging and neurodegenerative disease. Cell reports 9, 1135-1150. Brundin, P., Melki, R., and Kopito, R. (2010). Prion-like transmission of protein aggregates in neurodegenerative diseases. Nature reviews. Molecular cell biology 11, 301 -307.

Carter, R.J., Lione, L.A., Humby, T., Mangiarini, L, Mahal, A., Bates, G.P., Dunnett, S.B., and Morton, A.J. (1999). Characterization of progressive motor deficits in mice transgenic for the human Huntington's disease mutation. The Journal of neuroscience : the official journal of the Society for Neuroscience 19, 3248-3257. Castilla, J., Saa, P., and Soto, C. (2005). Detection of prions in blood. Nature medicine 11, 982-985.

Chan, H.Y., Warrick, J.M., Gray-Board, G.L., Paulson, H.L., and Bonini, N .M. (2000). Mechanisms of chaperone suppression of polyglutamine disease: selectivity, synergy and modulation of protein solubility in Drosophila. Human molecular genetics 9, 2811 -2820.

Chiti, F., and Dobson, C.M. (2017). Protein Misfolding, Amyloid Formation, and Human Disease: A Summary of Progress Over the Last Decade. Annu Rev Biochem 86, 27-68.

Ciamei, A., Detloff, P.J., and Morton, A.J. (2015). Progression of behavioural despair in R6/2 and Hdh knock-in mouse models recapitulates depression in Huntington's disease. Behavioural brain research 291, 140-146. Cohen, S.I., Vendruscolo, M., Dobson, C.M., and Knowles, T.P. (2012). From macroscopic measurements to microscopic mechanisms of protein aggregation. Journal of molecular biology 421, 160-171.

Crick, S.L., Ruff, K.M., Garai, K., Frieden, C., and Pappu, R.V. (2013). Unmasking the roles of N- and C-terminal flanking sequences from exon 1 of huntingtin as modulators of polyglutamine aggregation. Proceedings of the National Academy of Sciences 110, 20075-20080.

Davies, S.W., Turmaine, M., Cozens, B.A., DiFiglia, M., Sharp, A.FI., Ross, C.A., Scherzinger, E., Wanker, E.E., Mangiarini, L, and Bates, G.P. (1997). Formation of neuronal intranuclear inclusions underlies the neurological dysfunction in mice transgenic for the FID mutation. Cell 90, 537-548.

DiFiglia, M., Sapp, E., Chase, K.O., Davies, S.W., Bates, G.P., Vonsattel, J.P., and Aronin, N. (1997). Aggregation of huntingtin in neuronal intranuclear inclusions and dystrophic neurites in brain. Science 277, 1990-1993.

Du, D., Murray, A.N., Cohen, E., Kim, H .E., Simkovsky, R., Dillin, A., and Kelly, J.W. (2011 ). A kinetic aggregation assay allowing selective and sensitive amyloid-beta quantification in cells and tissues. Biochemistry 50, 1607-1617.

Gao, M., Estel, K., Seeliger, J., Friedrich, R.P., Dogan, S., Wanker, E.E., Winter, R., and Ebbinghaus, S. (2015). Modulation of human IAPP fibrillation: cosolutes, crowders and chaperones. Physical chemistry chemical physics : PCCP 17, 8338-8348.

Guo, J.L., and Lee, V.M. (2014). Cell-to-cell transmission of pathogenic proteins in neurodegenerative diseases. Nature medicine 20, 130-138.

Gupta, S., Jie, S., and Colby, D.W. (2012). Protein misfolding detected early in pathogenesis of transgenic mouse model of Huntington disease using amyloid seeding assay. The Journal of biological chemistry 287, 9982-9989.

Herva, M.E., Zibaee, S., Fraser, G., Barker, R.A., Goedert, M., and Spillantini, M.G. (2014). Anti-amyloid compounds inhibit alpha-synuclein aggregation induced by protein misfolding cyclic amplification (PMCA). The Journal of biological chemistry 289, 11897-11905.

Hockly, E., Woodman, B., Mahal, A., Lewis, C.M., and Bates, G. (2003). Standardization and statistical approaches to therapeutic trials in the R6/2 mouse. Brain research bulletin 61, 469-479.

Holmes, B.B., Furman, J.L., Mahan, T.E., Yamasaki, T.R., Mirbaha, H., Eades, W.C., Belaygorod, L., Cairns, N.J., Holtzman, D.M., and Diamond, M.l. (2014). Proteopathic tau seeding predicts tauopathy in vivo. Proceedings of the National Academy of Sciences of the United States of America 111, E4376- 4385. Hult, S., Soylu, R., Bjorklund, T., Belgardt, B.F., Mauer, J., Bruning, J.C., Kirik, D., and Petersen, A. (2011 ). Mutant huntingtin causes metabolic imbalance by disruption of hypothalamic neurocircuits. Cell metabolism 13, 428-439.

Jarrett, J.T., and Lansbury, P.T., Jr. (1993). Seeding "one-dimensional crystallization" of amyloid: a pathogenic mechanism in Alzheimer's disease and scrapie? Cell 73, 1055-1058.

Jeon, I., Cicchetti, F., Cisbani, G., Lee, S., Li, E., Bae, J., Lee, N., Li, L., Im, W., Kim, M., et al. (2016). Human-to-mouse prion-like propagation of mutant huntingtin protein. Acta neuropathologica.

Jiang, X., and Sorkin, A. (2002). Coordinated traffic of Grb2 and Ras during epidermal growth factor receptor endocytosis visualized in living cells. Molecular biology of the cell 13, 1522-1535.

Jucker, M., and Walker, L.C. (2013). Self-propagation of pathogenic protein aggregates in neurodegenerative diseases. Nature 501, 45-51.

Kayed, R., Head, E., Thompson, J.L., Mclntire, T.M., Milton, S.C., Cotman, C.W., and Glabe, C.G. (2003). Common structure of soluble amyloid oligomers implies common mechanism of pathogenesis. Science 300, 486-489.

Ko, J., Ou, S., and Patterson, P.H. (2001 ). New anti-huntingtin monoclonal antibodies: implications for huntingtin conformation and its binding proteins. Brain research bulletin 56, 319-329.

Kuemmerle, S., Gutekunst, C.A., Klein, A.M., Li, X.J., Li, S.H., Beal, M.F., Hersch, S.M., and Ferrante, R.J. (1999). Huntington aggregates may not predict neuronal death in Huntington's disease. Annals of neurology 46, 842-849.

Larson, E., Fyfe, I., Morton, A.J., and Monckton, D.G. (2015). Age-, tissue- and length-dependent bidirectional somatic CAG*CTG repeat instability in an allelic series of R6/2 Huntington disease mice. Neurobiology of disease 76, 98-111.

Latouche, M., Lasbleiz, C., Martin, E., Monnier, V., Debeir, T., Mouatt-Prigent, A., Muriel, M.P., Morel, L., Ruberg, M., Brice, A., et al. (2007). A conditional pan-neuronal Drosophila model of spinocerebellar ataxia 7 with a reversible adult phenotype suitable for identifying modifier genes. The Journal of neuroscience : the official journal of the Society for Neuroscience 27, 2483- 2492.

Laue, M. (2010). Electron microscopy of viruses. Methods in cell biology 96, 1 - 20.

Leitman, J., Ulrich Hartl, F., and Lederkremer, G.Z. (2013). Soluble forms of polyQ-expanded huntingtin rather than large aggregates cause endoplasm ic reticulum stress. Nature communications 4, 2753. LeVine, H., 3rd (1993). Thioflavine T interaction with synthetic Alzheimer's disease beta-amyloid peptides: detection of amyloid aggregation in solution. Protein science : a publication of the Protein Society 2, 404-410.

Li, S.-H., and Li, X.-J. (1998). Aggregation of N-Terminal Huntingtin is Dependent on the Length of Its Glutamine Repeats. Human Molecular Genetics 7, 777-782.

Li, H., Li, S.H., Cheng, A.L., Mangiarini, L., Bates, G.P., and Li, X.J. (1999). Ultrastructural localization and progressive formation of neuropil aggregates in Huntington's disease transgenic mice. Human molecular genetics 8, 1227-1236.

Lin, C.H., Tallaksen-Greene, S., Chien, W.M., Cearley, J.A., Jackson, W.S., Crouse, A.B., Ren, S., Li, X.J., Albin, R.L., and Detloff, P.J. (2001 ). Neurological abnormalities in a knock-in mouse model of Huntington's disease. Human molecular genetics 10, 137-144.

Liu, K.S.Y., Siebert, M., Mertel, S., Knoche, E., Wegener, S., Wichmann, C., Matkovic, T., Muhammad, K., Depner, H., Mettke, C., et al. (2011 ). RIM-Binding Protein, a Central Part of the Active Zone, Is Essential for Neurotransmitter Release. Science 334, 1565-1569.

Mangiarini, L., Sathasivam, K., Seller, M., Cozens, B., Harper, A., Hetherington, C., Lawton, M., Trottier, Y., Lehrach, H., Davies, S.W., et al. (1996). Exon 1 of the HD gene with an expanded CAG repeat is sufficient to cause a progressive neurological phenotype in transgenic mice. Cell 87, 493-506.

Mishra, R., Jayaraman, M., Roland, B.P., Landrum, E., Fullam, T., Kodali, R., Thakur, A.K., Arduini, L, and Wetzel, R. (2012). Inhibiting the Nucleation of Amyloid Structure in a Huntingtin Fragment by Targeting a-Helix-Rich Oligomeric Intermediates. Journal of Molecular Biology 415, 900-917.

Nguyen, A.W., and Daugherty, P.S. (2005). Evolutionary optimization of fluorescent proteins for intracellular FRET. Nature biotechnology 23, 355-360.

Nucifora, L.G., Burke, K.A., Feng, X., Arbez, N., Zhu, S., Miller, J., Yang, G., Ratovitski, T., Delannoy, M., Muchowski, P.J., et al. (2012). Identification of novel potentially toxic oligomers formed in vitro from mammalian -derived expanded huntingtin exon-1 protein. The Journal of biological chemistry 287, 16017-16028.

Osterwalder, T., Yoon, K.S., White, B.H., and Keshishian, H. (2001 ). A conditional tissue-specific transgene expression system using inducible GAL4. Proceedings of the National Academy of Sciences of the United States of America 98, 12596-12601.

Pecho-Vrieseling, E., Rieker, C., Fuchs, S., Bleckmann, D., Esposito, M.S., Botta, P., Goldstein, C., Bernhard, M., Galimberti, L, Muller, M., et al. (2014). Transneuronal propagation of mutant huntingtin contributes to non -cell autonomous pathology in neurons. Nature neuroscience 17, 1064-1072. Pieri, L., Madiona, K., Bousset, L., and Melki, R. (2012). Fibrillar alpha- synuclein and huntingtin exon 1 assemblies are toxic to the cells. Biophysical journal 102, 2894-2905.

Rosenblatt, A., Abbott, M.H., Gourley, L.M., Troncoso, J.C., Margolis, R.L., Brandt, J., and Ross, C.A. (2003). Predictors of neuropathological severity in 100 patients with Huntington's disease. Ann Neurol 54, 488-493. Saborio, G.P., Permanne, B., and Soto, C. (2001 ). Sensitive detection of pathological prion protein by cyclic amplification of protein misfolding. Nature 411, 810-813.

Sahl, S.J., Weiss, L.E., Duim, W.C., Frydman, J., and Moerner, W.E. (2012). Cellular inclusion bodies of mutant huntingtin exon 1 obscure small fibrillar aggregate species. Scientific reports 2, 895.

Sathasivam, K., Neueder, A., Gipson, T.A., Landles, C., Benjamin, A.C., Bondulich, M.K., Smith, D.L., Faull, R.L., Roos, R.A., Howland, D., et al. (2013). Aberrant splicing of HTT generates the pathogenic exon 1 protein in Huntington disease. Proceedings of the National Academy of Sciences of the United States of America 110, 2366-2370.

Scherzinger, E., Lurz, R., Turmaine, M., Mangiarini, L., Hollenbach, B., Hasenbank, R., Bates, G.P., Davies, S.W., Lehrach, H., and Wanker, E.E. (1997). Huntingtin-encoded polyglutamine expansions form amyloid-like protein aggregates in vitro and in vivo. Cell 90, 549-558.

Scherzinger, E., Sittler, A., Schweiger, K., Heiser, V., Lurz, R., Hasenbank, R., Bates, G.P., Lehrach, H., and Wanker, E.E. (1999). Self-assembly of polyglutamine-containing huntingtin fragments into amyloid-like fibrils: implications for Huntington's disease pathology. Proceedings of the National Academy of Sciences of the United States of America 96, 4604-4609. Schindelin, J., Arganda-Carreras, L, Frise, E., Kaynig, V., Longair, M., Pietzsch, T., Preibisch, S., Rueden, C., Saalfeld, S., Schmid, B., et al. (2012). Fiji: an open-source platform for biological-image analysis. Nature methods 9, 676-682.

Stewart, B.A., Atwood, H.L., Renger, J.J., Wang, J., and Wu, C.F. (1994). Improved Stability of Drosophila Larval Neuromuscular Preparations in Hemolymph-Like Physiological Solutions. J Comp Physiol A 175, 179-191.

Tan, Z., Dai, W., van Erp, T.G., Overman, J., Demuro, A., Digman, M.A., Hatami, A., Albay, R., Sontag, E.M., Potkin, K.T., et al. (2015). Huntington's disease cerebrospinal fluid seeds aggregation of mutant huntingtin. Molecular psychiatry 20, 1286-1293.

Theillet, F.X., Binolfi, A., Bekei, B., Martorana, A., Rose, H.M., Stuiver, M., Verzini, S., Lorenz, D., van Rossum, M., Goldfarb, D., et al. (2016). Structural disorder of monomeric alpha-synuclein persists in mammalian cells. Nature 530, 45-50.

Vonsattel, J.P., Myers, R.H., Stevens, T.J., Ferrante, R.J., Bird, E.D., and Richardson, E.P., Jr. (1985). Neuropathological classification of Huntington's disease. Journal of neuropathology and experimental neurology 44, 559-577.

Wagner, A.S., Politi, A.Z., Ast, A., Bravo-Rodriguez, K., Baum, K., Buntru, A., Strempel, N.U., Brusendorf, L, Hanig, C., Boeddrich, A., et al. (2018). Self- assembly of Mutant Huntingtin Exon-1 Fragments into Large Complex Fibrillar Structures Involves Nucleated Branching. Journal of molecular biology 430, 1725-1744.

Wanker, E.E., Scherzinger, E., Heiser, V., Sittler, A., Eickhoff, H., and Lehrach, H. (1999). Membrane filter assay for detection of amyloid -like polyglutamine- containing protein aggregates. Methods in enzymology 309, 375-386.

Warrick JM, C.H., Gray-Board GL, Chai Y, Paulson HL, Bonini NM. (1999). Suppression of polyglutamine-mediated neurodegeneration in Drosophila by the molecular chaperone HSP70. Nat Genet.

Woodman, B., Butler, R., Landles, C., Lupton, M.K., Tse, J., Hockly, E., Moffitt, H., Sathasivam, K., and Bates, G.P. (2007). The Hdh(Q150/Q150) knock-in mouse model of HD and the R6/2 exon 1 model develop comparable and widespread molecular phenotypes. Brain research bulletin 72, 83-97.

Zhang, J.H., Chung, T.D., and Oldenburg, K.R. (1999). A Simple Statistical Parameter for Use in Evaluation and Validation of High Throughput Screening Assays. Journal of biomolecular screening 4, 67-73.

Zuccato, C., Valenza, M., and Cattaneo, E. (2010). Molecular mechanisms and potential therapeutical targets in Huntington's disease. Physiological reviews 90, 905-981.