GR20190100453A - Cyclic peptide inhibitors of amyloid-b aggregation - Google Patents

Abstract

Aspects of the present invention disclose compounds that modulate the aggregation of amyloidogenic proteins or peptides. In some aspects, disclosed compounds modulate the aggregation of disease-associated proteins and natural β-amyloid peptides. In a preferred embodiment, the compounds can inhibit natural amyloid aggregation. Pharmaceutical compositions comprising the compounds of the embodiments, and diagnostic and treatment methods for diseases (e.g., amyloidogenic diseases) using the compounds, are also disclosed.

Description

CYCLIC PEPTIDE INHIBITORS OF AMYLOID-B PEPTIDE AGGREGATION

Field of invention

The present invention provides compounds that modulate the aggregation of amyloid-β peptides. In a preferred embodiment, the compounds can inhibit the natural accumulation of amyloid. In another preferred embodiment, the compounds are cyclic heptapeptides, where cyclization has taken place between the two terminal groups of the peptide (head-to-tail) or variants thereof bearing specific modifications, such that the compound alters aggregation or inhibits the neurotoxicity of natural β-amyloid peptides when in contact with the peptides. Pharmaceutical compositions comprising the compounds of the invention and diagnostic and therapeutic methods for amyloidogenic diseases using the compounds of the invention are also described.

Background of the invention

Protein misfolding is currently associated with more than 50 diseases such as Alzheimer's disease (AD), Parkinson's disease, Huntington's disease, type 2 diabetes, cystic fibrosis, amyotrophic lateral sclerosis, Gaucher disease , nephrogenic diabetes insipidus and Creutzfeldt-Jakob disease. These disorders are collectively called Protein Misfolding Diseases (PMDs). There are two ways in which misfolded proteins (MisPs) lead to disease: in one case the proteins lose their ability to perform their normal function (loss-of-function) and in the other case the proteins acquire a new harmful property (gain-of-function). Cellular or environmental factors such as changes in pH, oxidative stress, exposure to high concentrations of metal ions or other chemicals, as well as the presence of a mutation or multiple mutations in the amino acid sequence of certain proteins, can play a critical role in problematic protein folding. PMDs become more common with increasing life expectancy, as many of them are associated with increased age.

PMDs include very serious disorders with high incidence rates and a serious impact on the well-being of the human population, resulting in therapeutic compounds against these diseases being in huge demand. One of the most promising approaches for identifying new compounds with potential therapeutic properties against these diseases is the discovery of chemical molecules with the ability to repair problematic protein folding. Such molecules have already been identified for some MisPs. For example, linear peptides with homology to certain regions of β-amyloid peptide (Aβ) and small molecules, such as scyl-inositol, homotaurine, methylene blue, and bexarotene, have been found to modulate Aβ aggregation and inhibit associated neurotoxicity in vitro and in vivo, and some of these have subsequently progressed to clinical studies. Similarly, in the case of the protein α-synuclein (asyn) associated with Parkinson's disease, peptides with homology to the central hydrophobic region of the protein have been found as well as some natural products such as baicalein and epigallocatechin gallate (EGCG), which show promising anti-aggregation effects of asyn. Indeed, the small molecule tafamidis, with its ability to repair the misfolding of the transthyretin protein, has recently been approved for the treatment of familial amyloid polyneuropathy (FAP) in Europe and Japan and is marketed with name Vyndaqel® (Pfizer). This compound and its use for the treatment of FAP are described in European patent EP1587821.

Macrocycles have been characterized as a particularly promising class of potentially therapeutic compounds that have been largely unexplored (Driggers, E.M., Hale, S.P., Lee, J. & Terrett, N.K. The exploration of macrocycles for drug discovery — an underexploited structural class. Nat. Rev. Drug Discov. 7, 608-624 (2008)). These compounds occupy a space between small molecules and biological macromolecules and often exhibit the advantages of both classes, namely the high bioavailability of small molecules combined with the high specificity and fewer side effects of biological macromolecules. In addition, their typically larger size and more complex structure make these compounds particularly suitable for targeting difficult targets, such as those involved in protein interactions. Since many PMDs are characterized by protein aggregation, a process dependent on protein interactions, macrocyclic compounds are expected to be particularly active against this class of disorders. Their therapeutic potential is gradually beginning to increase in a wide range of diseases and is shown in patent US9308236, where macrocyclic agents are shown to inhibit the interactions of PD-1/PD-L1 (Programmed Death 1) and CD80/PD-L1 proteins and they are therefore useful in alleviating various diseases, including cancer and infectious diseases.

Alzheimer's disease (AD) is the most common progressive neurodegenerative disease causing dementia in older people. This condition affects more than 7 million people in Europe and 35 million people worldwide. In economic terms, these numbers translate into an annual cost of treating the disease of $818 billion for the United States alone (2015 World Alzheimer Report). Due to the aging of the human population, the incidence of AD has been steadily increasing in recent years and is expected to increase by 200% within the next 20 years.

AD is neuropathologically characterized by intracellular neurofibrillary tangles composed of hyperphosphorylated tau protein, extracellular accumulation of amyloid-β peptide (Aβ) in senile plaques, and the accumulation of soluble Aβ oligomers in the brain. The prevailing theory regarding the molecular sources behind the disease pathology is the amyloid cascade hypothesis, according to which the formation of Aβ oligomers and/or larger aggregates in the brain is neurotoxic and the accumulation of these protein forms results in neuronal degeneration, neuronal death and finally the onset of AD.

Aβ is produced by the cleavage of the precursor protein APP as a result of the action of the proteases β- and γ-secretase. Because of the broad amino acid recognition specificity of γ-secretase, its proteolytic action on APP leads to the formation of different forms of the Aβ peptide. The major APP cleavage product consists of 40 amino acids (Aβ40), while one of the secondary products is 42 amino acids long (Aβ42). Aβ is an amphiphilic peptide comprising a hydrophilic (amino acids 1-16) and a hydrophobic region (amino acids 17-40/42). Aβ42 contains two additional amino acids at its C-terminus and exhibits higher hydrophobicity and higher aggregation propensity than Aβ40. Due to the distribution of hydrophobic amino acids in its sequence, Aβ is unable to adopt a well-defined conformation. As a result, some hydrophobic regions of the Aβ sequence remain exposed to the aqueous environment of the cell and the protein tends to form oligomers and higher order aggregates. As mentioned above, these soluble oligomers and/or larger aggregates are thought to be the neurotoxic species of Aβ that cause AD. AD can therefore be considered a proteinopathy, which is initiated due to the peculiar biochemical/biophysical properties of Aβ and its associated folding problem, thus categorizing AD into the large group of disorders called PMDs, as described above.

Despite the enormous socio-economic impact of AD and the intense research efforts of the last decades, there are still no preventive or therapeutic treatments against this specific disease. Such therapies could be developed after the identification of inhibitors of any step in the molecular pathway that starts with Aβ biosynthesis and leads to neuronal degeneration and the development of dementia. For example, molecules with the ability to reduce the concentration of circulating Aβ could act as regulatory agents delaying the formation of neurotoxic Aβ forms and disease development. This can be achieved, for example, by using anti-Aβ antibodies that lead to isolation, degradation and/or purification of the peptide, or by using β- and γ-secretase inhibitors that inhibit APP cleavage and reduce the rate of its biosynthesis Av. Such compounds have already been discovered but have not yet demonstrated the desired therapeutic profiles, mainly due to unwanted side effects. For example, γ-secretase inhibitors have shown adverse effects on other normal molecular pathways, such as Notch signaling, causing severe side effects in mice (gastrointestinal symptoms, etc.).

Compounds with the ability to bind to Aβ, correct its misfolding and prevent the formation of Aβ oligomers/aggregates have the potential to act as inhibitors of Aβ-induced neurotoxicity and become effective anti-AD drugs. Small molecules, such as scyllo-inositol, homotaurine, methylene blue, and bexarotene, have been found to modulate Aβ aggregation and inhibit its neurotoxicity in vitro and in vivo. Additionally, linear peptides with homology to certain regions of Aβ have shown similar properties. These and other research efforts have led to the identification of several promising compounds that are being tested or have already been tested in clinical trials. Until now, however, no compound has demonstrated the desired preventive or therapeutic properties against AD. Regardless of whether it is the oligomeric or the larger aggregated forms of Aβ that are primarily responsible for neurotoxicity and other pathogenic effects, the discovery of chemical regulators of the natural process of Aβ oligomerization/aggregation is considered a very promising approach, since this process it is considered purely pathogenic, not involved in other normal cellular functions. Therefore, such molecules are expected to exhibit a safer pharmacological profile.

Summary of the invention

In one embodiment, the present invention provides small cyclic drug-like peptides that have the ability to rescue the misfolding and affect the natural aggregation process of β-amyloid peptide, which is involved in the pathogenesis of Alzheimer's disease. A variety of cyclic heptapeptides, where the cyclization has taken place between the two end groups of the peptide (head-to-tail), and which exhibit these properties are described in the present invention. In particular, the inventors have identified the general formula cyclo-NuX1X2X3X4, where Nu = cysteine (C), serine (S) or threonine (T) and X is any of the twenty naturally occurring amino acids, as a very rich source of chemical rescuers of Aβ and regulators of its aggregation.

In another embodiment, the present invention provides macrocyclic heptapeptides of the general formula cyclo-(C,T)XVWXXX, wherein the first amino acid is cysteine or threonine, the third amino acid is valine, the fourth amino acid is tryptophan, and X is any of the 20 natural amino acids, as effective and preferred regulators of the natural aggregation process of Aβ. More preferred modulators of Aβ aggregation are cyclic oligopeptide sequences exhibiting the motif cyclo-(C,T)(R, K)VW(P,A,M)X(Ψ,P) where n = Q, C, S or T , Ψ = L, V or I and X = any of the 20 naturally occurring amino acids. A small group of three cyclic heptapeptides with this general formula cyclo-(C,T)XVWXXX are analyzed further.

In yet another embodiment, the present invention provides macrocyclic heptapeptides of the general formula cyclo-(C,T)XXVP(S,A)X, wherein the first amino acid is cysteine or threonine, the fourth amino acid is valine, the fifth amino acid is proline, the sixth amino acid is serine or alanine, and X is any of the 20 naturally occurring amino acids, as effective and preferred regulators of the natural Aβ aggregation pathway. More preferred modulators of Aβ aggregation are cyclic oligopeptide sequences exhibiting the motif cyclo-(C,T)A(I,V)VP(S,A)Ψ, where Δ = R, I, V or Q and Ψ = L, V or I. A small group of two cyclic heptapeptides with this general formula cyclo-(C,T)XXVP(S,A)X are further analyzed.

In yet another embodiment of the present invention, isolated cyclic oligopeptides are provided which comprise the amino acid sequences of ALA. NO.

TAYT. 1, ALA. AP. TAYT. 2, AAA. AP. TAYT. 3, AAA. AP. TAYT. 4, ..., to AAA. AP. TAYT. 132, as effective and preferred regulators of the natural course of aggregation of Aβ. Also provided are the DNA sequences encoding the peptides of the invention, as well as plasmids containing such nucleic acids, and cells containing such plasmids.

In a further embodiment of the present invention sufficient evidence is provided that the selected macrocyclic peptides are capable of modulating Aβ aggregation. Indeed, the effect of five selected macrocyclic oligopeptides on the regulation of Aβ aggregation using appropriate biochemical and/or biophysical methods is provided.

In yet another embodiment of the present invention, the successful in vivo rescue of β-amyloid peptide misfolding and aggregation by the selected macrocyclic peptides is demonstrated. The protective effect of the two selected cyclic peptides against Aβ aggregation and in vivo toxicity is demonstrated in established models of AD in the nematode Caenorhabditis elegans (C. elegans).

In yet a further embodiment of the present invention, hybrid polypeptides are provided which comprise a peptide motif that specifically interacts with the target polypeptide, and which is then inserted into a suitable protein scaffold. The polypeptide motif is introduced into the scaffold in such a way as to retain any desired activity of the scaffold while at the same time the peptide retains its ability to specifically bind to the target and/or modulate the natural aggregation pathway of the target polypeptide. The scaffold may include, for example, neuroprotective agents that make amyloid plaques less toxic, molecules that destroy amyloid to eliminate plaques, reagents that prevent plaque formation, or reagents that are useful for specifically imaging plaques in the brain tissue.

Other embodiments and advantages of the present invention will become apparent to those skilled in the art in view of the following detailed description.

Description of the designs

Fig. 1A-1B: (A) (Left) Schematic representation of the pSICLOPPS-NuX1X2X X4X5X6 plasmid library encoding the combinatorial oligopeptide library cyclo-NUX1X2X3X4X5X6; Nu: Cys (C), Ser (S) or Thr (T). X: any of the 20 naturally occurring amino acids. NNS: random codons, where N is any of the nucleotide bases A, T, C or G and S = G or C. 1C: C-terminus of Ssp DnaE split intein. IN: N-terminal end of Ssp DnaE split intein; CBD: chitin binding segment. (Right) Cyclization of a peptide using intein and SICLOPPS technology. After interaction of the two ends of the intein, the active intein is cleaved and a peptide bond is created between the two ends of the peptide leading to its cyclization and ultimately the production of the cyclo-NuX1X2X3X4X5X6 library; (B) Theoretical and actual diversity of the constructed combinatorial oligopeptide library cyclo- NUX1X2X3X4X5X6;

Fig. 2A-2T: (A) Schematic representation of the bacterial platform used for the discovery of inhibitors of protein aggregation and for the high-throughput analysis of selected compounds. pMisP-GFP: plasmid encoding a MisP-GFP fusion protein, PSICLOPPS-NUX1X2X3X4X5X6: plasmid library encoding the combinatorial oligopeptide library cyclo-NuX1X2X3X4X5X6; Nu: Cys (C), Ser (S) or Thr (T). X: any of the twenty naturally occurring amino acids. NNS: random codons, where N = A, T, C or G and S = G or C. FSC-H: forward scatter. SSC-H: side scatter. R: sorting area. (B) Scan of Tuner E. coli (DE3) cells overexpressing the Aβ42-GFR chimera together with the pooled library of cyclo -NuX1X2X3X4X5X6 heptapeptides using FACS flow cytometry. M: mean GFP fluorescence in arbitrary units. FITC-A: filter for fluorescein isothiocyanate. (C) Relative fluorescence of Tuner E. coli (DE3) cells co-expressing the Aβ42-GFR chimera and ten individual circular peptide sequences isolated after the seventh round of FACS screening shown in (B) using wild-type Ssp DnaE intein or of a mutant Ssp DnaE intein, incapable of splicing Ssp DnaE (H24L/F26A). Two randomly selected cyclic peptide sequences (random 1 and 2) previously shown not to affect fluorescence and aggregation of Aβ42-GFR were used as negative controls. The fluorescence of the bacterial population producing random cyclic peptide 1 was arbitrarily set to 100. Mean values ± s.e.m. are given. (n = 3 independent experiments, each in triplicate). (D) Western blot of the ten individual selected clones probed in (C) using anti-CBD antibody. The ~25kDa protein band corresponds to the IC precursor -peptide-IN-CBD, while the protein band of ~20 kDa corresponds to the IN-CBD product, the appearance of which is an indication of successful splicing and formation of the circular peptide. CBD: chitin-binding moiety. (E) Tuner cell fluorescence E. coli (DE3) co-expressing the p53(Y220C)-GFP chimeric protein from the pETp53(Y220C)-GFP plasmid, together with the cyclic peptides encoded by the selected clones 1–10 investigated in (C). The fluorescence of the bacterial population producing the random cyclic peptide was arbitrarily set to 100. Experiments were performed in triplicate (n = 1 independent experiment), and the values reported are the mean ± s.e.m.(F) Immunostaining and western blots of total (left) and soluble (right) cell lysates of Tuner E. coli (DE3) cells overexpressing Aβ42-GFP and the 10 individual circular peptide sequences examined in (C), using the anti-Aβ antibody 6E10 . The predicted molecular mass of the Aβ42-GFP chimera is ~32 kDa. (G) A western blot using the anti-Aβ antibody 6E10 (left) and in-gel fluorescence (right) of the total cell lysate after native-PAGE of Tuner E cells .coli (DE3) overexpressing Aβ42-GFP and the 10 individual cyclic peptide sequences examined in (C). (H) Emission spectra of Tuner E. coli (DE3) cells overexpressing Aβ42 together with four of the selected cyclic heptapeptide sequences tested in (C) and after incubation with ThS. The maximum fluorescence of the bacterial population producing random cyclic peptide 1 was arbitrarily set to 100. Mean values ± s.e.m are shown (n = 1 experiment performed in triplicate). (I) DNA sequences of the peptide coding regions of the pSICLOPPS plasmids contained in the selected clones studied in (C) together with the predicted amino acid sequences of the cyclic heptapeptides they encode.

Fig. 3A-3H: (A) Relative fluorescence of Tuner E. coli (DE3) cells co-expressing the Aβ42-GFR chimera and three individual cyclic heptapeptide sequences occurring at low frequencies in the selected cell population. The fluorescence of the cell population producing a random cyclic peptide was arbitrarily set to 100. Experiments were performed in triplicate (n = 1 independent experiment) and reported values correspond to the mean ± s.e.m. (B) Frequency of occurrence of the 20 native amino acids at each amino acid sequence position of the heptapeptides selected after the seventh round of screening (Figure 2B). (C) Enrichment (blue) and depletion (red) of 20 amino acids in each of the amino acid sequence of the heptapeptides selected after the seventh round of screening (Figure 2B). Values represent the log2-change of peptide amino acid frequency of the selected population compared to the original library. (D) Graphical representation of the enrichment and depletion of the 20 native amino acids at each amino acid sequence position of the heptapeptides as in (C). (E) Schematic representation of the linear representatives of a cyclic heptapeptide. (F) Representation of the basic clusters formed according to the sequence similarities of the selected cyclic heptapeptides. Nodes represent different cyclic peptide sequences, and solid lines connect pairs of peptides with at least 70% sequence homology. The sequences of the members of the two dominant clusters (clusters I and II) are presented in the respective dendrograms. (G) Representation of all clusters identified using the Girvan-Newman algorithm of Gephi software as in (F). Gray nodes represent cyclic heptapeptides that do not show at least 70% homology to any other cyclic peptide from the selected group. (H) Distribution of heptapeptide sequences in the different groups identified.

Fig. 4A-4E: (A) Chemical structures of selected cyclic heptapeptides AβC7-1 and AβC7-14. (B) Kinetic study of 2 μM Aβ42 aggregation in the absence and presence of AβC7-1 at different molar ratios (left) and the normalized values of t1/2, tlag, and the corresponding aggregation trajectories (right). (C) As (B) for AβC7-14. In (B) and (C), mean values ± s.e.m. (n = 1 experiment performed in triplicate). (D) Study of the aggregation course of 2 μM Aβ42 in the presence and absence of 0.5 μM AβC7-14 using blot immunoblotting and the OC Aβ42-specific antibody. Mean values ± s.e.m. (n = 1 experiment performed in triplicate). (E) Transmission electron microscopy TEM images of samples of 2 μM Aβ42 fibrils taken at the endpoint of the aggregation reaction as measured by thioflavin T and in the absence (left) or presence (right) of either 10 μM AβC7-1 (top) or 4 μM AβC7- 14 (below).

Fig. 5A-5E: (A) Normalized motility (left) and normalized movement speed (right) of Aβ42 and wild-type worms in the absence and presence of 40 μM AβC7-1 and 5 μM AβC7-14 during days 5 to 10 of adulthood. (B) Motility (left) and velocity (right) of individual Aβ worms and wild-type worms in the absence and presence of AβC7-1 and AβC7-14 at day 7 of adulthood. (C) Gross fitness of worms as in (B). (D) Relative fluorescence of Aβ42 and wild-type worms at day 7 of adulthood showing a 50–60% reduction in aggregate formation in the presence of AβC7-1 and AβC7-14. (E) Representative images from (D). In (A) to (C), approximately 200 worms were analyzed on average, while in (D), 25 worms were analyzed in total. In all cases, mean values ± s.e.m. are presented. (n = number of worms tested in an experiment). Statistical significance is indicated by * P < 0.05 and * P < 0.0001, for differences relative to the worm Aβ42 sample without peptide.

Fig. 6A-6H: (A) Relative fluorescence of Tuner E. coli (DE3) cells co-expressing the Aβ42-GFR chimera and AβC7-1 (left) or AβC7-14 (right) as well as their indicated variants, such as was measured by flow cytometry. The fluorescence of the bacterial population producing the random cyclic peptide was arbitrarily set to 100. Experiments were performed in triplicate (n = 1 experiment) and reported values correspond to the mean ± s.e.m. (B) A western blot using the anti-Aβ antibody 6E10 (top) and in-gel fluorescence (bottom) of the total cell lysate after native-PAGE electrophoresis of Tuner E cells. coli (DE3) overexpressing Aβ42-GFP and AβC7-1 (left) or AβC7-14 (right) along with their indicated variants. (C) Heat map of the distribution of the 20 native amino acids at each amino acid sequence position of the heptapeptides corresponding to cluster I (Figure 3F), as demonstrated after next-generation sequencing. Total (left) or unique (right) heptapeptide sequences are included in the analysis. (D) Frequency of occurrence of the twenty native amino acids in positions 2 to 7 of the peptide coding region of plasmids pSICLOPPS-NUX1X2X3X4X5X6 contained in the bacterial clones corresponding to cluster I (303,245 reads corresponding to cyclic heptapeptides of cluster I out of 404,280 total reads occurred more than 20 times in the selected cell population). (E) Frequency of occurrence of the twenty natural amino acids in positions 2 to 7 of the unique cyclic heptapeptides encoded by the pSICLOPPS-NUX1X2X3X4X5X6 vectors contained in the bacterial clones corresponding to cluster I (107 unique peptide sequences corresponding to cluster I cyclic heptapeptides from a total 416 unique selected peptide sequences that appeared more than 20 times in the selected cell population). (F) Heat map of the distribution of the 20 native amino acids at each amino acid sequence position of the heptapeptides corresponding to cluster II (Figure 3F), as demonstrated after next-generation sequencing. Total (left) or unique (right) heptapeptide sequences are included in the analysis. (G) Frequency of occurrence of the twenty native amino acids in positions 2 to 7 of the peptide coding region of plasmids pSICLOPPS-NUX1X2X3X4X5X6 contained in the bacterial clones corresponding to cluster II (19,690 reads corresponding to cyclic heptapeptides of cluster II out of 404,280 total reads occurred more than 20 times in the selected cell population). (H) Frequency of occurrence of the twenty natural amino acids in positions 2 to 7 of the unique cyclic heptapeptides encoded by the vectors pSICLOPPS-NuX1X2X3X4X5X6 contained in the bacterial clones corresponding to cluster II (25 unique peptide sequences corresponding to cyclic heptapeptides of cluster II from a total 416 unique selected peptide sequences that appeared more than 20 times in the selected cell population).

Detailed description of the invention

I. General

According to the prevailing amyloid cascade hypothesis, the high propensity of Aβ to misfold and aggregate results in the formation of neurotoxic Aβ oligomers/aggregates, the accumulation of which ultimately leads to neuronal degeneration and disease development. The present invention relates to compounds and pharmaceutical compositions thereof, which can regulate the aggregation of starch-specific proteins and peptides, and specifically compounds that can regulate the aggregation of β-amyloid peptide (Aβ) and inhibit its neurotoxicity. A compound of the invention that modulates Aβ aggregation, referred to herein alternatively as an Aβ-modulating compound, an Aβ modulator, or simply a modulator, results in alteration of the aggregation of native Aβ when the modulator contacts native Av. Thus, a compound of the invention acts as a metabolite of the natural aggregation process or aggregation rate of Aβ, thus interrupting the normal course of this process. A modulator that inhibits Aβ aggregation (an "inhibitor") can be used to prevent or delay the onset of β-amyloid deposition. In addition, inhibitor compounds of the invention inhibit the formation and/or activity of neurotoxic aggregates of native Aβ peptide (ie, the inhibitor compounds can be used to inhibit Aβ neurotoxicity).

Alternatively, in another embodiment, a modulatory compound of the invention promotes the aggregation of native Aβ peptides. The various forms of the term "promote" refer to an increase in the amount and/or rate of aggregation of Aβ in the presence of the modulator, compared to the amount and/or rate of aggregation of Aβ in the absence of the modulator. Such a compound which promotes the aggregation of Aβ is referred to as a stimulatory compound. Stimulatory compounds may be useful, for example, in reducing the amounts of neurotoxic Aβ oligomers by driving the natural aggregation process of Aβ toward its (generally) less neurotoxic and larger aggregates.

The compounds of the present invention may inhibit Aβ aggregation and/or oligomerization. In particular, preferred regulatory compounds of the invention include cyclic oligopeptides of the general formula cyclo-NuX1X2X3X4X5X6, where X is any of the twenty natural amino acids and Ny = cysteine (Cys or C), serine (Ser or S) or threonine (Thr or T) , which are capable of altering (and preferably inhibiting) the natural aggregation process or aggregation rate of Aβ. This Aβ regulator may comprise seven or fewer amino acid residues (or derivatives, analogs or mimetics thereof).

II. Discovery of peptide regulators of Aβ

The present invention describes the use of a recently devised generalized bacterial platform for the discovery of macrocyclic peptides with the ability to rescue misfolded-prone proteins (MisPs) and which have been associated with human illnesses. This platform is described in the international patent PCT/IB2018/000622 and in the scientific article Matis, I. et al. An integrated bacterial system for the discovery of chemical rescuers of disease-associated protein misfolding. Nat. Biomed. Eng. 2, 48 (2018).

This approach offers several important advantages. First, it allows the construction and screening of extensive molecular libraries with a variety of up to 10<10> different molecules. A diverse library of >200 million different cyclic peptides was investigated in the present invention. It is important to emphasize that E. coli can support the in vivo biosynthesis not only of cyclic peptides where cyclization has taken place between the two end groups of the peptide (head-to-tail), such as those examined here, but also of other macrocyclic structures. , such as cyclic peptides where the cyclization has taken place between a side chain and a terminal group of the peptide (side chain-to-tail), bicyclic peptides, cyclotides, macrolides and other macrocyclic structures, which can contain not only natural amino acids but also wide variety of non-physical.

In addition, the analysis of these large libraries is performed using a very high-throughput genetic method, which allows the identification of bioactive molecules simply by isolating compounds that enhance the fluorescence of E. coli cells expressing the protein chimera MisP-GFP , such as the Aβ-GFP chimera, by medium flow cell sorting (FACS). Compared to approaches that rely on screening DNA-encoded chemical libraries based on their affinity to an immobilized protein target, such as phage-display and mRNA-display technologies, the approach described here does not detect simply conjugation to target MisP, such as Aβ, but directly selects bioactive compounds with the ability to repair misfolding of MisP, such as Aβ, without requiring the availability of an amount of purified MisP, such as Aβ.

Furthermore, the synthesis of test compounds and their screening for bioactivity are performed in vivo as part of a one-step process, without the need for organic synthesis and product isolation, which can be very difficult and time-consuming. . Importantly, screening for bioactivity is performed in a completely unbiased manner without requiring a priori knowledge of the structure of the MisP target protein, such as monomer, oligomer or larger aggregates of Aβ, and without requiring specific assumptions on potential binding sites or prior purification of aggregated forms of MisP, such as Ab.

More specifically, combinatorial libraries of random cyclic heptapeptides have been prepared and the most important targets have been selected to be studied as potential correctors of Aβ misfolding. The technology described in the present invention uses a technique called split internal circular ligation of peptides and proteins (SICLOPPS) for the production of peptide libraries in E. coli. The SICLOPPS technology uses split intenins, which are self-splicing protein elements, to cyclize between the N-terminus and C-terminus of a peptide intermediate and biosynthesize cyclic products with a minimum length of four amino acids. The only requirement for protein splicing and peptide cyclization to occur is the presence of a nucleophilic amino acid, i.e., cysteine (C), serine (S), or threonine (T), as the first amino acid of the protein intermediate following the C- terminal end of the intein.

According to the present invention, peptides belonging to the general formula NuX1X2X3X4X5X6, can be used to repair problematic protein folding and regulate their aggregation, where X is any of the twenty naturally occurring amino acids and Nu is cysteine (Cys or C ), serine (Ser or S) or threonine (Thr or T). According to the preferred embodiment of the present invention the peptide is a cyclic peptide. According to a preferred embodiment of the present invention Nu is C.

In a preferred embodiment of the present invention the peptide of the general formula cyclo-NuX1X2X3X4X5X6, for use in repair of protein folding and aggregation, has the following specifications: where Nu is C or T, where Xi is any amino acid and more preferably is R or K, where X2 is preferably V, where X3 is preferably W, where X4 is any amino acid and more preferably is Q, C, S, T, A or M, where X5 is any amino acid, where X6 is any amino acid and is preferably selected from L, V, I or P. According to the above specifications, the preferred cyclic heptapeptide is selected from the amino acid sequences set forth in the ALAs. ID NO. 1, ALA. ID NO. 2, ALA. ID NO. 3, ALA. ID NO. 4, ..., to ALA. ID NO. 107. According to the above specification, the most preferred heptapeptide is selected from cyclo-CKVWQLL (ALA. NO. 1), cyclo-CKVWMPL (ALA. NO. 2), and cyclo-CRVWTEL (ALA. NO. ID 3).

In another preferred embodiment of the present invention the peptide of the general formula cyclo-NuX1X2X3X4X5X6, for use in repair of protein folding and aggregation, has the following specifications: where Nu is C or T, where Xi is any amino acid and more preferably is R, I, V or Q, where X2 is any amino acid and more preferably I or V, where X3 is preferably V, where X4 is preferably P, where X5 is preferably S or A, where X6 is any amino acid and is preferably selected from L, V or I. According to the above specifications, the preferred cyclic heptapeptide is selected from the amino acid sequences set forth in the ALAs. ID NO. 108, ALA. ID NO. 109, ALA. ID NO. 110, ALA. ID NO. 111, ..., to ALA. ID NO. 132. According to the above specification, the most preferred heptapeptide is selected from cyclo-CRIVPSL (A.D. NO. 108), and cyclo-CIVVPSI (A.D. NO. 109).

The maximum theoretical diversity of the combined cyclo-NuX1X2X3X4X5X6 library investigated in the present invention was ~200 million different sequences. Libraries of genes encoding this combinatorial library of random cyclic heptapeptides were constructed using degenerate codons. The inventors created the highly diverse library of PSICLOPPS-NuX1X2X3X4X5X6 plasmids which is expected to encode the large majority of theoretically possible designed cyclo-NuX1X2X3X4X5X6 cyclic heptapeptide sequences using molecular biology techniques already known and used in the field.

To identify cyclic oligopeptide sequences with the ability to interfere with Aβ misfolding and regulate its oligomerization/aggregation, a high-throughput bacterial genetic system was used. This system monitors Aβ folding and aggregation by measuring the fluorescence of E. coli cells overexpressing a chimeric fusion of human Aβ42 with GFP. It has previously been shown that due to the high aggregation propensity of Aβ, E. coli cells overexpressing Aβ-GFP fusions produce misfolded protein fusion that accumulates in insoluble aggregates that lack fluorescence, despite expressing these fusions at high levels. Mutations in the Aβ sequence, or addition of compounds that inhibit its aggregation, result in the formation of soluble and fluorescent Aβ-GFP, and bacterial cells expressing the chimeric Aβ-GFP protein under these conditions acquire a fluorescent phenotype. The inventors of the present invention have used an adapted version of this system to screen macrocyclic aggregation inhibitors in a very high-throughput manner using a fluorescence-activated cell sorting (FACS)-based flow cytometry method that allows the isolation of bacterial clones that produce bioactive cyclic heptapeptides and as a result show increased levels of Aβ42-GFP fluorescence, as described in international patent PCT/IB20 18/000622 and in the scientific article Matis, I. et al. An integrated bacterial system for the discovery of chemical rescuers of disease-associated protein misfolding. Nat. Biomed. Eng. 2, 48 (2018).

Specifically, Tuner E. coli (DE3) cells producing the combined cyclo-NuX1X2X3X4X5X6 library, while simultaneously overexpressing the Aβ42-GFP chimeric protein, were screened by FACS in order to isolate clones showing increased Aβ42-GFP fluorescence. The increase in partial fluorescence was then measured and ten random clones were selected from the bacterial population after the screening process. In each case, the Aβ42-GFR fluorescence of the selected clones was found to be dramatically increased compared to cells expressing the same Aβ42-GFR chimera in the presence of random cyclic peptides selected from the initial (before the screening process) cyclo-NuX1X2X3X4X5X6 library; In addition, the observed phenotypic effects were found to depend on the ability of the Ssp DnaE intein (used for peptide cyclization as part of SICLOPPS technology) to perform protein splicing, as mutation of the C-terminus of the intein at positions H24L/F26A, which is known to prevent protein splicing and peptide cyclization, was found to reduce Aβ42-GFP fluorescence back to baseline fluorescence levels. Finally, the observed increase in fluorescence was found to be specific for Aβ, as isolated pSICLOPPS-NuX1X2X3X4X5X6 plasmids from these selected clones were unable to enhance the fluorescence levels of cells in which the Aβ sequence has been replaced with that of the DNA binding domain ( DNA-binding domain) of the human p53 protein containing the Tyr220Cys mutation (p53C(Y220C)).

Analysis of expressed Aβ42-GFP fusions by polyacrylamide gel electrophoresis (SDS-PAGE) and western blotting revealed that bacterial clones expressing the selected cyclic peptides significantly increased soluble levels of Aβ42-GFP compared to random cyclic peptide sequences. Furthermore, when the same cell lysates were analyzed under non-denaturing conditions (native-PAGE), it was observed that co-expression of the selected cyclic peptides reduced the accumulation of large aggregates of Aβ42-GFR, which cannot enter the gel, and increased the abundance of species exhibiting increased electrophoretic mobility.

The inventors further analyzed the selected peptides by DNA sequencing of the coding region of the ten isolated clones. This method revealed five distinct cyclic peptide sequences with potential inhibitory activity against Aβ aggregation: CKVWQLL (present six times between sequences), CKVWTEL, CKVVMPL, CIVVPSI and CRIVPSL.

As illustrated in Example 2, the inventors performed next-generation sequencing of the PSICLOPPS-NuX1X2X3X4X5X6 plasmids contained in the selected bacterial population after the FACS screening process, in order to determine all of the potentially bioactive cyclic heptapeptides present in the enriched library. Sequence analysis of the selected cyclic heptapeptides revealed that the Cys residue was present at position 1 in the great majority of the selected cyclic heptapeptides and that the remaining positions were enriched by a small number of specific amino acids: Arg and Lys at position 2, Val at position 3 , Trp and Thr at position 4, lle, Gin, Cys, Met, Ser, Thr and Pro at position 5, Ala, Leu, Val, Glu, Lys and Pro at position 6, and lie, Leu and Pro at position 7. In contrast, the majority of amino acids, including those present in higher abundance in the original library, were strongly de-enriched, thus indicating a highly efficient selection process.

Sequence similarity analysis using all linear versions of the selected cyclic peptides and hierarchical cluster analysis revealed the formation of 20 distinct clusters with similar sequence features. The majority of peptides from the two most dominant clusters, clusters I and II, appear to belong to a cyclo-CxVWxxx and a cyclo-CxxVPSx sequence motif, respectively, in agreement with the DNA sequencing results presented above.

As illustrated in Examples 3-6, the inventors selected two cyclic heptapeptide sequences for further analysis. These were cyclo-CKVWQLL and cyclo-CRIVPSL, hereafter referred to as AβC7-1 and AβC7-14 (Aβ-targeting cyclic 7-peptide number 1 and 14), and were produced via solid-phase synthesis in mg quantities. The inventors decided to further study these cyclic peptides because they were both encoded by the selected bacterial clones initially examined, and, more importantly, they were the most frequently encountered members between the two dominant clusters (clusters I and II).

The effect of the selected heptapeptides on the aggregation process of Aβ42 was first assessed by monitoring the kinetic course of Aβ42 aggregation in the absence and presence of the two cyclic peptides after thioflavin T (ThT) staining. Both heptapeptides were found to inhibit Aβ42 aggregation very efficiently at sub-metric ratios, increasing both the tlag time (time required for ThT fluorescence to reach 10% of total amplitude) and the tgrowth time (time transition from 10% to 90% of the total fluorescence range of ThT) of the Aβ42 aggregation reaction, although to a different extent. Similar results were observed in the absence of ThT, when the course of aggregation was monitored by extracting protein samples at different time points and detecting Aβ fibril formation by blot immunoblotting and using the specific OC antibody. Finally, transmission electron microscopy (TEM) studies revealed that Aβ42 fibrils formed after completion of the aggregation reaction in the absence and presence of both cyclic peptides were similar in size and morphology, indicating that these selected cyclic peptides do not bind irreversible with Aβ42, and do not reorient the aggregation process outside the normal molecular pathway.

The protective effect of AβC7-1 and AβC7-14 against Aβ aggregation and its associated toxicity was evaluated in vivo, using an established AD disease model in the nematode Caenorhabditis elegans (C. elegans). The conservation of genetic and metabolic pathways between mammals and C. elegans, combined with its fully characterized nervous and muscular system, easy imaging and simple handling, has defined C. elegans as an excellent model of neurodegenerative diseases including AD disease, while the use of C. elegans in the process of drug discovery in AD is increasing. The motility and overall fitness of the C. elegans GMC101 strain, where human Aβ42 is expressed in animal muscle wall cells under the control of a heat-inducible promoter and where Aβ aggregate formation is accompanied by the appearance of a paralysis phenotype, was monitored in the absence and presence of AβC7-1 and AβC7-14 and compared to C. elegans wild-type worms, which do not express any Aβ42. Both peptides were found to increase the motility and speed of Aβ-expressing worms throughout their lifespan, eventually restoring their overall fitness to approximately the levels of wild-type animals. Furthermore, imaging of the worms revealed a significant reduction in Aβ deposits in the presence of either of the two selected cyclic peptides.

To identify the functionally important residues within the isolated peptides, the inventors performed substitution of the position 1 residue with the other two nucleophilic amino acids present in the original libraries, as well as alanine mutagenesis of positions 2-7 of the AβC7-1 and AβC7 heptapeptides -14. As judged by the ability of the resulting variants to increase the fluorescence of E. coli cells overexpressing Aβ42-GFR, in the case of both AβC7-1 and AβC7-14, substitution of Cys at position 1 with Ser had as resulted in a 50% decrease in fluorescence, while substitution with Thr resulted in similar levels of fluorescence and aggregation of Aβ42-GFR to those corresponding to the respective starting sequence (AβC7-1 or AβC7-14). Furthermore, for both peptides, mutagenesis to alanine at the majority of positions 2-7 resulted in a significant decrease in Aβ42-GFR fluorescence and a concomitant increase in aggregation. These observations indicate that several residues in both selected cyclic heptapeptides are necessary for optimal inhibitory activity against Aβ aggregation. Indeed, sequence analysis of all selected sequences belonging to either cluster I or cluster II revealed that the peptides most frequently occurring in each cluster have strong preferences for specific amino acids at each position.

In particular, for cluster I, Arg and Lys residues at position 2 appeared in >90% of the selected peptides, while Val at position 3, Tip at position 4, Gin, Cys, Ser, Met and Thr at position 5, and Ile , Val and Leu at position 7 occurred in >99% of selected clones. Similarly, for cluster II, the frequency of occurrence of Arg, Ile, Val, and Gin at position 2 was ~93%, while for lie and Val at position 3, Val at position 4, Pro at position 5, Ser and Ala at position 6 and lie, Leu, and Val at position 7 were >97%. Taken together, these results indicate that the most bioactive motifs against problematic Aβ folding and aggregation in the examined library are cyclo-(C,T)(R,K)VW(n,A,M) and cyclo-(C ,T)A(I,V)VP(S,A) Ψ, for complexes I and II respectively, where X is any of the 20 natural amino acids, Π: any of the polar amino acids Q, C, S and T, D: R, I, V or Q, and Ψ: any of the aliphatic amino acids L, V and I.

III. Peptide modifications

Peptides and polypeptides of the invention include those corresponding to linear versions of the described cyclic heptapeptides with the ability to regulate Aβ aggregation, i.e. to sequences where the main amino acid chain of a described cyclic heptapeptide has been cleaved and which subsequently contains a free NH2- terminal amino group and a free COOH-terminal carboxyl group. For example, for the cyclic heptapeptide AβC7-1 with the amino acid sequence cyclo-CKVWQLL (AL. NO. SEQ. 1), a preferred Aβ peptide modulator of the present invention is also a linear version of AβC7-1, specifically the oligopeptide NH2 - CKVWQLL-COOH. Furthermore, due to their cyclic nature, the Aβ regulators described in the present invention do not have a "start point" (eg, N-terminus) or "termination point" (eg, C-terminus). Thus, all linear variants of the described cyclic oligopeptide Aβ modulators resulting from cleavage of an existing peptide bond to introduce new ends elsewhere in the peptide sequence are also preferred oligopeptide Aβ modulators of the present invention. For example, for the cyclic heptapeptide AβC7-1 with amino acid sequence cyclo-CKVWQLL (ALA. NO. SEQ ID NO: 1), preferred Aβ peptide modulators of the present invention are also all equivalent cyclic versions of AβC7-1, namely cyclo oligopeptides -KVWQLLC, cyclo-VWQLLCK, cyclo-WQLLCKV, cyclo-QLLCKVW, cyclo-LLCKVWQ and cyclo-LCKVWQL. Likewise, preferred Aβ peptide modulators of the present invention are also the linear versions of all described cyclic oligopeptides and all equivalent cyclic variants thereof. For example, for AβC7-1, in addition to the modification mentioned above, preferred Aβ peptide modulators of the present invention are linear versions of all cyclic AβC7-1 equivalent variants, namely the oligopeptides NH 2 -KVWQLLC-COOH, NH 2 -VWQLLCK-COOH, NH2-WQLLCKV-COOH, NH2-QLLCKVW-COOH, NH2-LLCKVWQ-COOH and NH2-LCKVWQL-COOH.

Peptides and polypeptides of the invention include those containing conservative amino acid substitutions. Such peptides and polypeptides are encompassed by the invention provided that the peptide or polypeptide can bind to Aβ. As used herein, suitable conservative amino acid substitutions are known to those skilled in the art and can generally be made without altering the biological activity of the resulting molecule. Those skilled in the art recognize that, in general, single amino acid substitutions in nonessential regions of a polypeptide do not substantially alter its biological activity (see, e.g., Watson et al. Molecular Biology of the Gene, 4th Edition, 1987, The Benjamin /Cummings Pub.co., p. 224). Such substitutions may be made according to those set forth as follows: Ala (A) Gly; SerArg(R)LysAsn(N)Gin; His Cys (C) Ser Gin (Q) Asn Glu (E) Asp Gly (G) Ala; ProHis(H)Asn; Gin lie (I) Leu; Val Leu (L.) lie; Val Lys (K) Arg; Gin? GluMet(N)Leu; Tyr? Ile Phe (L) Met; Leu? Tyr Ser (S) Thr Thr (T) Ser Trp (W) Tyr Tyr (Y) Trp; Phe Val (V) lie; Leo. Other substitutions are also permissible and can be determined empirically or according to known conservative substitutions.

The peptide compounds of the present invention can inhibit the misfolding of proteins and their aggregation, where said peptide compounds can be cyclic peptides where the cyclization has been carried out between the two end groups of the peptide (head-to-tail), cyclic peptides where cyclization has taken place between a side group and a terminal end of the peptide (side chain-to-tail), bicyclic peptide, lanthipeptide, linaridin, proteusin, cyanobactin, thiopeptide ), bottromycin, microcin, lasso peptide, microviridin, amatoxin, phallotoxin, θ-defensin, orbitide or cyclotide ( cyclotide).

The peptide compounds of the present invention also serve as structural models for non-peptidic molecules or "mimetics" with similar biological activity. Peptides can also be readily modified by phosphorylation and other methods to prepare peptide derivatives of the compounds of the present invention as described by Hruby, et al. Biochem. J. 268(2): 249-262, 1990, incorporated herein by reference. Thus, the peptide compounds of the invention also serve as structural models for non-peptide compounds with similar biological activity. Those skilled in the art recognize that various techniques are available for making compounds with the same or similar desired biological activity as the lead peptide compound but with more favorable activity than the lead compound in terms of solubility, stability, and susceptibility to hydrolysis and proteolysis. . See Morgan and Gainor, Ann. Rep. Med. Chem. 24:243-252, 1989, incorporated herein by reference. These techniques involve replacing the peptide backbone with a backbone composed of phosphonates, amides, carbamates, sulfonamides, and secondary amines.

The term mimetic, and in particular peptidomimetic, is intended to include isostere. The term "isostere", as used herein, is intended to include a chemical structure which can substitute for a second chemical structure because the steric configuration of the first structure is adapted to a binding site specific to the second structure. The term specifically includes peptide backbone modifications (ie, amide bond mimetics), which are well known to those skilled in the art. These include modifications of the amide nitrogen, α-carbon, amide carbonyl, complete replacement of the amide bond, extensions, deletions or backbone crosslinks. Several modifications of the main peptide chain are known, including ψ[CH2S], ψ[CH2NH], ψ[CSNH2], ψ[NHCO], ψ[COCH2], and ψ[(E) or (Z) CH=CH], In the nomenclature used above, ψ indicates the absence of an amine bond. The structure replacing the amide group is specified in parentheses. Other examples of isosteres include peptides substituted with one or more benzodiazepine molecules (see, e.g., James, G. L. et al. (1993) Science 260: 1937-1942).

Other possible modifications include the formation of cross-links via N-alkyl (or aryl) substitution (ψ[CONR|) in order to make lactams and other cyclic structures, the substitution of all L-amino acids in the sequence to D-amino acids (inverse or inverso compounds) or inversion of the amino acid sequence and replacement of L-amino acids with D -amino acids (retro-inverso or enantio-retro) (ψ[NHCO|). For example, if the original peptide is Thr-Ala-Tyr, the inverted modified form is Tyr-Ala-Thr, the inverse form is thr-ala-tyr, and the retro-inverso form is tyr-ala-thr (the lowercase refer to D -amino acids). Compared to the parent peptide, a retro-inverso peptide has an inverted main chain essentially retaining the original steric configuration of its side chains, resulting in a retro-inverso isomer with a topology that closely resembles the parent peptide. See Goodman et al. "Perspectives in Peptide Chemistry" pp. 283-294 (1981). See also US. Pat. No. 4,522,752 to Sisto for further description of “retro-inverso” peptides.

Preferably, the modulatory compound inhibits aggregation of native β-amyloid peptides when contacted with native β-amyloid peptides and/or inhibits Aβ neurotoxicity. Alternatively, the modulator compound may promote aggregation of native β-amyloid peptides when contacted with native β-amyloid peptides. The type and number of modifying groups conjugated to the modulator are selected such that the compound alters (and preferably inhibits) the aggregation of native β-amyloid peptides when contacted with native β-amyloid peptides. A single modifying group can be conjugated to the regulator or, alternatively, multiple modifying groups can be conjugated to the regulator.

Within a compound-modulator of the invention, a peptide structure (such as a cyclic oligopeptide regulator of Aβ or an amino acid sequence corresponding to a rearranged or modified cyclic oligopeptide regulator of Aβ) is directly or indirectly conjugated to at least one modifying group. The term "modifying group" is intended to include structures that are directly attached to the peptide structure (e.g., by covalent conjugation), as well as those that are indirectly attached to the peptide structure (e.g., by fixed non-covalent linkage or by covalent conjugation with additional amino acid residue or mimetics, analogues or derivatives thereof, which may border the cyclic oligopeptide regulator of Aβ). For example, the modifying group can be conjugated to a side chain of at least one amino acid residue of a cyclic oligopeptide regulator of Aβ or to a peptide or peptide mimetic region flanking the cyclic oligopeptide regulator of Aβ (e.g., via the ε-amino group one or more lysine residues, through the carboxyl group of one or more aspartic acid or glutamic acid residues, through the hydroxyl group of one or more tyrosine, serine or threonine residues or through other suitable functional groups from amino acid side chains). Modifying groups covalently coupled to the peptide structure can be linked by means and using methods well known in the art for linking chemical structures, including, for example, amide, alkylamine, carbamate or urate linkages.

The modifying group(s) is (are) selected such that the buffering compound alters and preferably inhibits the aggregation of the bamyloid peptides upon contact with the β-amyloid peptides or inhibits the neurotoxicity of β-amyloid peptides when in contact with them. Although not intended to be limited by mechanism, the modifier group(s) of the modulators of the invention are believed to function as a key pharmacophore which is important to confer on the modulator the ability to alter the aggregation of Av.

In one embodiment, the modifying group is a "biotinyl structure", which includes biotinyl groups and analogs and derivatives thereof (such as a 2-iminobiotinyl group). In another embodiment, the modifying group may comprise a "fluorescein-containing group", such as a group derived from the reaction of an Aβ-derived peptide structure with 5-(and 6-) -carboxyfluorescein (5-(and 6-)-carboxyfluorescein), succinimidyl ester or fluorescein isothiocyanate. In various other embodiments, the modifying group may include an N-acetylneuraminyl group, a trans-4-cotininecarboxyl group, a 2-imino-1-imidazolidineacetyl group ( 2-imino-1-imidazolidineacetyl), one (S)-(-)-indoline-2-carboxyl group ((S)-(-)-indoline-2-carboxyl), one (-)-methoxyacetyl group ((-)- menthoxyacetyl), a 2-norbomaneacetyl group (2-norbomaneacetyl), a goxo-5-acenaphthenebutyryl group (γ-ocho-5-acenaphthenebutyryl), a (-)-2-oxo-4-thiazolidinecarboxyl group ((-)-2 -oxo-4-thiazolidinecarboxyl), a tetrahydro-3-furoyl group, a 2-iminobiotinyl group, a diethylenetriaminepentaacetyl group, a 4-morpholinecarbonyl group ), a 2-thiopheneacetyl group or a 2-thiophenesulfonyl group.

Preferred modifying groups include groups comprising cholyl structures, biotinyl structures, fluorescein-containing groups, a diethylene-triaminopentaacetyl group, a (-)-methoxyacetyl group, and an N-acetyl-neuraminyl group. More preferred modifying groups are those comprising a bile structure or an iminobiotinyl group. Yet another type of modifying group is a compound containing an unnatural amino acid.

The β-amyloid modulator compounds of the invention can be further modified to alter specific properties while retaining their ability to alter Aβ aggregation and inhibit its neurotoxicity. For example, in one embodiment, a compound is further modified to alter a pharmacokinetic property thereof, such as in vivo stability or half-life. In another embodiment, the compound is further modified to be labeled with a detectable substance. In yet another embodiment, the compound is further modified to bind to another therapeutic molecule. To further chemically modify a compound, such as to alter its pharmacokinetic properties, functional groups can be derivatized.

A moderator compound can be further modified to be labeled with a detectable substance. Suitable detectable substances include various enzymes, prosthetic groups, fluorescent materials, luminescent materials and radioactive materials. Examples of suitable enzymes include horseradish peroxidase, alkaline phosphatase, β-galactosidase or acetylcholinesterase. Examples of suitable prosthetic group complexes include streptavidin/biotin and avidin/biotin. Examples of suitable fluorescent materials include umbelliferone, fluorescein, fluorescein isothiocyanate, rhodamine, dichlorotriazinylamine fluorescein, dansyl chloride or phycoerythrin. ). An example of a luminescent material includes luminol. Examples of suitable radioactive material include <14>C, <123>I, <124>I, <125>I, <131>I, <99m>Tc, <35>S or <3>H. In a preferred embodiment, a moderator compound is radioactively labeled with <14>C, either by incorporating <14>C into the modifying group or into one or more amino acid structures in the moderator compound. Labeled compounds can be used to assess the in vivo pharmacokinetics of compounds, as well as to detect Aβ aggregation, for example for diagnostic purposes. Aggregation of Aβ can be detected using a labeled regulator compound either in vivo or in an in vitro sample derived from a patient.

Preferably, for use as an in vivo diagnostic agent, a compound regulator of the invention is labeled with radioactive technetium or iodine. Accordingly, in one embodiment, the invention provides a moderator compound labeled with technetium, preferably<99m>Tc. Methods for labeling peptide compounds with technetium are well known in the art (see, e.g., U.S. Pat. Nos. 5,443,815, 5,225,180 and 5,405,597, all by Dean et al., Stepniak-Biniakiewicz, D., et al. (1992) J. Med. Chem. 35:274-279, Fritzberg, A. R., et al. (1988) Proc. Natl. Acad. Sci. USA 85:4025-4029, Baidoo, K. E., et al. (1990) Cancer Res. Suppl. 50:799s-803s, and Regan, L. and Smith, C.K. (1995) Science 270:980-982).

Further, an additional modification of a modulator compound of the invention may serve to impart an additional therapeutic property to the compound. That is, the additional chemical modification may include an additional functional moiety. For example, a functional moiety with the ability to break down or dissolve amyloid plaques can be conjugated to the modulator compound. In this form, the modified-module targets Aβ peptides and disrupts their polymerization, while the additional functional unit disrupts or dissolves amyloid plaques after targeting the compound to these sites.

In an alternative chemical modification, a β-amyloid compound of the invention is prepared in a "prodrug" form, where the compound itself does not modulate aggregation, but rather is capable of being transformed, during in vivo metabolism, into a modulator of Aβ such as defined here. For example, in this type of compound, the modulating group may be present in a prodrug form that is capable of being converted upon metabolism to an active modulating group. This prodrug form of a modifying group is referred to herein as a "secondary modifying group". Numerous strategies for the preparation of peptide prodrugs, which limit metabolism in order to optimize the release of the active form of the peptide drug, are well known in the art (see, e.g., Moss, J. (1995) in Peptide-Based Drug Design : Controlling Transport and Metabolism, Taylor, M. D. and Amidon, G. L. (eds), Chapter 18). Additional strategies have been specifically designed to deliver a drug to the central nervous system (CNS) relying on "sequential metabolism" (see, e.g., Bodor, N., et al. (1992) Science 257:1698-1700; Prokai, L., et al. (1994) J. Am. Chem. Soc. 116:2643-2644, Bodor, N. and Prokai, L. (1995) in Peptide-Based Drug Design: Controlling Transport and Metabolism, Taylor, M. D. and Amidon, G.L. (eds), Chapter 14). In one embodiment of a prodrug form of a modulator of the invention, the modifying group comprises an alkyl ester to facilitate permeability of the blood brain barrier.

The modulator compounds of the invention can be prepared by chemical synthesis using techniques well known in the art. The peptide component of a regulator consisting, at least in part, of a peptide can be synthesized using standard techniques such as those described in Bodansky, M. Principles of Peptide Synthesis, Springer Verlag, Berlin (1993) and Grant, GA (ed. ). Synthetic Peptides: A User's Guide, W. H. Freeman and Company, New York (1992). Automated peptide synthesis systems are also commercially available (eg, Advanced ChemTech Model 396, Milligen/Biosearch 9600). In addition, one or more regulatory groups can be attached to a regulator of Aβ (e.g., a nuclear aggregation domain of Aβ) by standard methods, for example using reaction methods via an amino group, a carboxyl group, a hydroxyl group (e.g. tyrosine, serine or threonine residues) or through other suitable functional groups from amino acid side groups (see e.g. Greene, T. W. and Wuts, P. G. M. Protective Groups in Organic Synthesis, John Wiley and Sons, Inc ., New York (1991)).

Alternatively, modulator compounds of the invention can be prepared biosynthetically and isolated in pure or enriched form from a host that allows recombinant protein production, such as bacteria, yeast, plant cells, or mammalian cells (see, e.g., Scott CP, Abel-Santos E, Jones AD, Benkovic SJ, Structural requirements for the biosynthesis of backbone cyclic peptide libraries. Chem Biol. 2001 Aug;8(8):801-15, as an example of recombinant production of cyclic oligopeptides in bacterial cells).

Alternatively, the modulator compounds of the invention can be biosynthetically prepared from a host that allows recombinant protein production, such as bacteria, yeast, plant cells or mammalian cells, but cannot be isolated in pure or enriched form and instead they can therefore be provided to the patient in the form of a probiotic, as part of the recombinant host, such as a bacterium, a yeast, a plant cell or a mammalian cell, that produces the specific regulatory compound. By "probiotic" we mean live microorganisms or other cultured cells that can provide health benefits when administered and consumed in adequate amounts (see e.g. O'Toole PW, Marchesi JR, Hill C, Next-generation probiotics: the spectrum from probiotics to live biotherapeutics, Nat Microbiol. 2017 Apr 25;2:17057).

IV. Hybrid regulators

A hybrid molecule of the invention comprises a peptide or polypeptide that binds to the amyloid form of Aβ as well as a scaffold molecule. The scaffold molecule may comprise a diagnostic or therapeutic reagent. The therapeutic or diagnostic reagent may be a polypeptide, small molecule or compound.

Specifically, provided herein are hybrid molecules, such as hybrid polypeptides, comprising a peptide or polypeptide provided in the present invention, and additional amino acid residues (typically 5, 10, 15, 20, 30, 40, 50, 100 or more) such that the resulting hybrid molecule to specifically interact with Aβ. The motif can be modified, such as by replacing certain amino acids or by directed and random evolution methods, to produce motifs with higher affinity. As used herein, a hybrid polypeptide refers to a polypeptide comprising domains from at least two sources, such as from an antibody or enzyme or other scaffold that may be an acceptor, and from a binding motif, such as a polypeptide or a peptide that binds to a amyloid form of the Aβ peptide.

Thus, among the hybrid molecules provided herein are hybrid molecules, namely hybrid polypeptides produced by grafting a binding motif (e.g., peptide) from a molecule to a scaffold, such as an antibody or fragment thereof or an enzyme or other molecule reference. The hybrid polypeptides provided herein, even the hybrid immunoglobulins, are not antibodies per se, but are polypeptides that are hybrid molecules that contain a selected motif (eg, a peptide that binds to the amyloid form of the Aβ peptide) the which is inserted into another polypeptide so that the motif retains or acquires the ability to bind to a protein associated with a disease caused by protein aggregation. Hybrid polypeptides may include portions of antibodies or other scaffolds, but may also include a portion not derived from an antibody or other scaffold, which is grafted thereto. The non-immunoglobulin moiety is recognized by its ability to bind specifically to a target polypeptide isomeric form. The hybrid polypeptide can bind specifically to the target infectious or disease-associated or a selected isoform of a polypeptide as a monomer with sufficient affinity to detect the resulting complex or to precipitate the target polypeptide.

The scaffold is chosen such that insertion of this motif does not substantially alter (ie, preserve) the desired specific binding of the motif. The scaffold can additionally be selected for its properties, such as its ability to act as a reference molecule.

Provided in the present invention are methods for producing hybrid molecules that specifically interact with a form of a protein that is associated with a disease associated with protein folding or protein aggregation. In these methods a polypeptide motif from the protein is inserted into a scaffold so that the resulting molecule exhibits specific binding to one protein form compared to others. In particular, the hybrid molecule may exhibit specific binding to the amyloid form of the Aβ peptide. The peptides of the invention have been shown to bind to Aβ in vitro and in vivo. Peptides can be incorporated into a scaffold which includes additional amino acid sequences and/or compounds. The hybrid molecule can then be used to label or treat plaques associated with Aβ amyloid. The polypeptides, nucleic acids encoding the polypeptides, and methods of using the polypeptides or nucleic acids can be used to detect, diagnose, and/or treat disorders associated with the formation of plaques in brain tissue.

Any molecule, such as a polypeptide, into which the selected polypeptide motif is inserted (or linked) so that the resulting hybrid polypeptide has the desired binding specificity is contemplated as part of the hybrid molecules of the present invention. The polypeptides can be inserted into any amino acid sequence that contains at least a sufficient number (10, 20, 30, 50, 100 or more amino acids) to present the correct motif for binding to the target amyloid plaque. The purpose of the scaffold is to present the motif to the target polypeptide in a form that binds to it. The scaffold can be designed or selected to have additional properties, such as serving as a detectable marker or label, or having additional binding specificity to allow or aid its use in assays to detect specific isoforms of a target protein (e.g., e.g., the amyloid form of the Aβ peptide) or for screening therapeutic molecules or other assays and methods.

Scaffolds include reporter molecules, such as fluorescent proteins and enzymes or fragments thereof, and binding molecules, such as antibodies or fragments thereof. Selected scaffolds include whole or parts of antibodies, enzymes such as luciferases, alkaline phosphatases, β-galactosidases and other signal producing enzymes, chemiluminescence generators such as horseradish peroxidase, fluorescent proteins such as red, green and blue fluorescent proteins, which are well known, and chromogenic proteins.

The peptide motif is introduced into the scaffold in a region that does not interfere with any desired activity. Scaffolds can also include other functional domains, such as an additional binding site, such as one specific for a second detection group.

V. Nucleic acid molecules

Nucleic acid molecules encoding any of the peptides, polypeptides or hybrid polypeptides provided herein are provided in the general experimental procedures. Such molecules can be introduced into plasmids and vectors for expression in suitable host cells. As used herein, the term "nucleic acid" refers to single-stranded and/or double-stranded polynucleotides such as deoxyribonucleic acid (DNA) and ribonucleic acid (RNA) as well as analogs or derivatives of RNA or DNA. Also included in the term "nucleic acid" are analogues of nucleic acids such as peptide nucleic acid (PNA), phosphorothioate DNA and other such analogues and derivatives or combinations thereof. The term should be understood to include, as equivalents, derivatives, variants, and analogs of either RNA or DNA produced from nucleotide analogs, single-stranded (of the coding or template strand), or double-stranded polynucleotides. Deoxyribonucleotides include deoxyadenosine, deoxycytidine, deoxyguanosine and deoxythymidine. For RNA, the uracil base is uridine.

Also provided in the general experimental procedures are plasmids and vectors containing the nucleic acid molecules, as well as cells containing these vectors, including cells expressing the encoded proteins. The cell may be a bacterial cell, a yeast cell, a fungal cell, a plant cell, an insect cell or an animal cell. Also provided in the present invention are the methods for producing a cyclic oligopeptide or a hybrid polypeptide, such as, for example, growing the cell under conditions where the encoded polypeptide is expressed by the cell and recovering the expressed protein. Cells are used to express the cyclic oligopeptide or protein, which may be secreted or expressed in the cytoplasm. Hybrid polypeptides can also be chemically synthesized using standard protein synthesis methods well known in the art.

VI. Pharmaceutical compositions

It is envisioned that one would use the modulators of the present invention as a treatment for Alzheimer's disease. If the regulator is peptide in nature, a gene therapy technique can be used to deliver the DNA molecule encoding the regulator to the affected sites. As for pharmaceutical compositions, they must reach and be effective in the affected areas. These modulators are more likely to be mixtures of carbohydrates and peptides, and particular mixtures capable of crossing the blood-brain barrier. For other examples, see Tamai et al., Adv. Drug Delivery Review 19: 401-424, 1996, incorporated herein by reference. In these cases, the active component of the regulators would also facilitate transport across the blood-brain barrier.

Thus, the present invention provides methods for the treatment of Alzheimer's disease, comprising the administration of a compound of the invention in amounts sufficient to modulate the natural course of Aβ aggregation. Accordingly, the present invention provides pharmaceutical compositions comprising, as an active ingredient, at least one of the peptides or other compounds of the invention in combination with a pharmaceutical carrier or excipient. The compounds of the invention can be administered by the oral, parenteral (intramuscular, intraperitoneal, intravenous (IV) or subcutaneous injection), transdermal, nasal, vaginal, rectal or sublingual routes.

Solid forms for oral administration include capsules, tablets, pills, powders and granules. In such solid dosage forms, the active compound is mixed with at least one pharmaceutically inert acceptable carrier such as sucrose, lactose or starch. Such dosage forms may also include, as is common practice, additional substances other than inert molecules, such as e.g. lubricants such as magnesium stearate. In the case of capsules, tablets and pills, the dosage forms may also include buffering agents. Tablets and pills can additionally be prepared with enteric coatings.

Liquid dosage forms for oral administration include pharmaceutically acceptable emulsions, solutions, suspensions, syrups, the formulations containing inert agents commonly used in the field, such as water. In addition to such inert molecules, the compositions may also include adjuvants such as wetting agents, emulsifying and suspending agents, and sweetening, flavoring and flavoring agents.

According to the present invention, preparations for parenteral administration include sterile aqueous or non-aqueous solutions, suspensions or emulsions. Examples of non-aqueous solvents or carriers are propylene glycol, polyethylene glycol, vegetable oils such as olive oil and corn oil, gelatin, and injectable organic esters such as ethyl oleate. Such dosage forms may also contain adjuvants such as preserving, wetting, emulsifying and dispersing agents. They can also be sterilized, for example, by filtration through bacteria-retaining filters, by incorporating sterilizing agents into their compositions, by irradiating the compositions, or by heating the compositions. In addition, they can be made up immediately before use, using sterile water or some other sterile injectable medium.

Compositions for rectal or vaginal administration are preferably suppositories which may contain, in addition to the active substance, excipients such as cocoa butter or wax. Formulations for nasal or sublingual administration are also prepared with standard excipients well known in the art.

The dosage of the active ingredient in the compositions of this invention may vary. However, it is necessary that the amount of the active ingredient is such that a suitable dosage form is achieved. The selected dosage depends on the desired therapeutic effect, on the route of administration and on the duration of the desired treatment.

The following examples illustrate aspects of the present invention that include the construction and screening of a library of macrocyclic peptides, the identification of macrocyclic peptides with the ability to repair misfolding and aggregation of the prominent protein target Aβ associated with AD disease, and ultimately the use in rescuing toy Aβ aggregation and toxicity in vitro and in vivo. The examples do not in any way limit the scope of the invention.

Examples

Example 1

Combinatorial libraries of random cyclic heptapeptides have been selected to be studied as potential rescuers of Aβ misfolding and its pathological aggregation. A technique called split internal circular ligation of peptides and proteins (SICLOPPS) (US 7354756 B1 “Intein-mediated cyclization of peptides”) is used to produce the peptide libraries in E. coli bacteria. The SICLOPPS technology uses split inteins, which are self-splicing protein elements, to cyclize between the N-terminus and C-terminus of a peptide intermediate and biosynthesize cyclic products with a minimum length of four amino acids. The only requirement for protein splicing and peptide cyclization to occur is the presence of a nucleophilic amino acid, i.e., cysteine (C), serine (S), or threonine (T), as the first amino acid of the protein intermediate following the C- terminal end of the intein.

In order to maximize the diversity of the libraries, the inventors chose to study peptides of the general formula cyclo-NuX1X2X3X6, where Nu = C, S or T and X is any of the twenty natural amino acids (Fig. 1A). The maximum theoretical polymorphism of the combined cyclo-NuX library, X2X3X4X5X6, is 192 million different sequences (Fig. 1B). Libraries of genes encoding this combinatorial library of random cyclic heptapeptides were constructed using degenerate codons. The inventors created the highly diverse plasmid library PSICLOPPS-NUX1X2X3X4X5X6 which is expected to encode all theoretically possible NUX1X2X3X4X4X6 cyclic heptapeptide designed sequences using molecular biology techniques already known and used in the field.

To identify cyclic heptapeptide sequences with the ability to interfere with the misfolding of Aβ and inhibit its oligomerization/aggregation, the inventors used a high-throughput bacterial genetic system. This system monitors Aβ folding and aggregation by measuring the fluorescence of E. coli cells overexpressing a chimeric fusion of human Aβ42 with GFP (US20070077552A1 “High throughput screen for inhibitors of polypeptide aggregation”). It has previously been shown that due to the high aggregation propensity of Aβ, E. coli cells overexpressing Aβ-GFP fusions produce misfolded protein fusion that accumulates in insoluble aggregates that lack fluorescence, despite expressing these fusions at high levels. Mutations in the Aβ sequence, or addition of compounds that inhibit its aggregation, result in the formation of soluble and fluorescent Aβ-GFP, and bacterial cells expressing the chimeric Aβ-GFP protein under these conditions acquire a fluorescent phenotype. The inventors of the present invention have used an adapted version of this system to screen macrocyclic aggregation inhibitors in a very high-throughput manner using a fluorescence-activated cell sorting (FACS)-based flow cytometry method that allows the isolation of bacterial clones that produce bioactive cyclic heptapeptides and as a result exhibit increased levels of Aβ42-GFR fluorescence, as shown in Fig. 2A and described in international patent PCT/IB20 18/000622 and in the scientific article Matis, I. et al. An integrated bacterial system for the discovery of chemical rescuers of disease-associated protein misfolding. Nat. Biomed. Eng. 2, 48 (2018).

Tuner E. coli (DE3) cells producing the combined cyclo-NuX1X2X3X4X5X6 library, while simultaneously overexpressing the Aβ42-GFP chimeric protein, were screened by FACS to isolate clones exhibiting increased Aβ42-GFP fluorescence. After seven rounds of screening, the mean fluorescence of the bacterial population increased almost sixfold compared to that of the original library (Fig. 2B). Ten individual clones were randomly selected from the enriched population after the screening process and cellular Aβ42-GFR fluorescence was measured using a plate reader. Aβ42-GFP fluorescence of the selected clones was found to increase dramatically compared to cells expressing the same Aβ42-GFP fusion in the presence of random cyclic peptide sequences previously shown not to affect Aβ42-GFP fluorescence and aggregation (Fig. 2C ). In addition, the observed phenotypic effects were found to depend on the ability of the Ssp DnaE intein to perform protein splicing, as mutation of the C-terminus of the intein at positions H24L/F26A, which is known to block protein splicing and peptide cyclization , was found to reduce Aβ42-GFP fluorescence back to baseline fluorescence levels (Fig. 2C–2D). Finally, the observed increase in fluorescence was found to be specific for Aβ, as isolated pSICLOPPS-NuX1X2X3X4X5X6 plasmids from these selected clones were unable to enhance the fluorescence levels of cells in which the Aβ sequence has been replaced with that of the DNA binding domain ( DNA-binding domain) of the human p53 protein containing the Tyr220Cys mutation (p53C(Y220C)) (Fig. 2E).

Analysis of expressed Aβ42-GFP fusions by polyacrylamide gel electrophoresis (SDS-PAGE) and western blotting revealed that bacterial clones expressing the selected cyclic peptides significantly increased soluble levels of Aβ42-GFP compared to random cyclic peptide sequences, despite the fact that the total amount of Aβ42-GFP chimera remained at the same levels in each case (Fig. 2F). Furthermore, when the same cell lysates were analyzed under non-degradative conditions (native-PAGE) and western blotting, it was observed that co-expression of the selected cyclic peptides reduced the accumulation of large aggregates of Aβ42-GFR, which cannot enter the gel and increased the abundance of species exhibiting increased electrophoretic mobility (Fig. 2G, left). As revealed by in-gel fluorescence analysis, these higher electrophoretic mobility species correspond to the fraction of the total Aβ42-GFP chimera that exhibits fluorescence (Fig. 2G, right). Since the solubility and fluorescence of bacterially expressed Aβ-GFP have been found to be inversely proportional to the propensity for Aβ aggregation, these results suggest that Aβ aggregation is significantly reduced in the presence of the selected cyclic peptides.

Similar results were obtained when Aβ42 was produced in a free form, i.e. not fused to GFP. The inventors studied the effect of selected cyclic heptapeptides on Aβ42 aggregation using an in vivo method employing thioflavin S (ThS) staining of whole cells and intracellularly formed Aβ42 aggregates (Espargaro A, Sabate R, Ventura S. Mol BioSyst ., 2012, 8: 2839-2844) and observed that co-production of the selected peptides resulted in reduced levels of ThS fluorescence, further indicating reduced aggregate formation (Fig.

2H).

DNA sequencing of the region of the ten isolated clones encoding the selected peptides revealed five distinct cyclic peptide sequences with potential inhibitory activity against Aβ aggregation: CKVWQLL (present six times between sequences), CKVWTEL, CKVVMPL, CIVVPSI, and the CRIVPSL (Fig. 2Θ).

Example 2

In order to determine the set of potentially bioactive cyclic heptapeptides present in the enriched library, the inventors performed next-generation sequencing on > 0.4 million PSICLOPPS-NuX1X2X3X4X5X6 plasmids from the selected bacterial population after the FACS screening process (Fig.

2B). This analysis revealed the existence of 416 distinct cyclic heptapeptide sequences that occur at least 20 times within the selected population, thus suggesting that their presence in the enriched population is not random. Cloning of three randomly selected cyclic heptapeptide sequences occurring at low frequencies in the selected cell population revealed that they are also effective in increasing the fluorescence of the bacterially expressed Aβ42-GFP chimera (Fig. 3A).

Sequence analysis of the selected cyclic heptapeptides revealed that the Cys residue was present at position 1 in the large majority of the selected cyclic heptapeptides (99.6% of the selected sequences) (Fig. 3B). Also, the inventors observed that the remaining positions were enriched by a small number of specific amino acids: Arg and Lys at position 2, Val at position 3, Tip and Thr at position 4, Ile, Gin, Cys, Met, Ser, Thr and Pro at position 5, Ala, Leu, Val, Glu, Lys, and Pro at position 6, and lie, Leu, and Pro at position 7 (Figs. 3C-3D). In contrast, the majority of amino acids, including those present in higher abundance in the original library, were strongly de-enriched (Figs. 3C-3D), thus indicating a highly efficient selection process.

In order to identify possible relationships between the selected cyclic heptapeptides, the inventors performed sequence similarity analysis and hierarchical cluster analysis. As the similarity analysis is performed using only linear sequences, the inventors considered all linear versions of the selected cyclic peptides (Fig. 3E). Of the 416 cyclic heptapeptides selected, 323 of them formed 1467 unique pairs with at least 70% sequence homology and formed 20 distinct clusters with similar sequence features (Fig. 3F-3H). Clusters I and II were the most dominant, containing 75.0% and 4.6% of the selected bacterial clones, respectively, as well as 25.7% and 6% of the selected cyclic heptapeptide sequences (Fig. 3F-3H). . The majority of peptides from clusters I and II appear to belong to a cyclo-CxVWxxx sequence motif and a cyclo-CxxVPSx sequence motif, respectively, in agreement with previous observations (Fig. 2T).

Example 3

Two of the selected heptapeptides, cyclo-CKVWQLL and cyclo-CRIVPSL, named AβC7-1 and AβC7-14 (Aβ-targeting cyclic 7-peptide number 1 and 14), respectively (Fig. 4A) were selected for further analysis and which were chemically synthesized in mg quantities. These cyclic peptides were chosen because they were both encoded by the selected bacterial clones initially screened (Fig. 2T), and, more importantly, they were the most frequently encountered members between the two dominant clusters (clusters I and P) (Tables 1 -2).

The effect of the selected heptapeptides on the aggregation process of Aβ42 was first assessed by monitoring the kinetic course of Aβ42 aggregation in the absence and presence of the two cyclic peptides after thioflavin T (ThT) staining. Both heptapeptides were found to inhibit Aβ42 aggregation very efficiently at sub-stoichiometric ratios, starting at 0.5 molar equivalents for AβC7-1 and 0.1 molar equivalents for AβC7-14 (Fig. 4B-4C). Specifically, both cyclic peptides were found to increase both the tlag time (time required for ThT fluorescence to reach 10% of the total range) and the tgrowth time (transition time from 10% to 90% of the total ThT fluorescence range) of the Aβ42 aggregation reaction, although to a different extent (Figs. 4B-4C). Similar results were observed in the absence of ThT, when the course of aggregation was monitored by extracting protein samples at different time points and detecting Aβ fibril formation by blot immunoblotting using the specific OC antibody (Fig. 4D). Finally, transmission electron microscopy (TEM) studies revealed that Aβ42 fibrils formed after completion of the aggregation reaction in the absence and presence of both cyclic peptides were similar in size and morphology, indicating that these selected cyclic peptides do not bind irreversible with Aβ42, and do not reorient the aggregation process outside the normal molecular pathway (Fig. 4E).

Example 4

In order to evaluate in vivo the protective effects of AβC7-1 and AβC7-14 against Aβ aggregation and its associated toxicity, the inventors used an established AD disease model in the nematode Caenorhabditis elegans (C. elegans). The conservation of genetic and metabolic pathways between mammals and C. elegans, combined with its fully characterized nervous and muscular system, easy imaging and simple handling, has defined C. elegans as an excellent model of neurodegenerative diseases including AD disease, while the use of C. elegans in the process of drug discovery in AD is increasing. The motility and overall fitness of the C. elegans GMC101 strain, where human Aβ42 is expressed in animal muscle wall cells under the control of a heat-inducible promoter and where Aβ aggregate formation is accompanied by the appearance of a paralysis phenotype, was monitored in the absence and presence of AβC7-1 and AβC7-14 and compared to C. elegans wild-type worms, which do not express any Aβ42. Both peptides were found to increase the motility and speed of Aβ-expressing worms throughout their lifespan and eventually restoring their overall fitness to approximately the levels of wild-type animals (Fig. 5A–5C). Furthermore, imaging of the worms revealed a significant reduction in Aβ deposits in the presence of either of the two selected cyclic peptides (Figs. 5D-5E). Taken together, these results demonstrate the protective effect of the two cyclic peptides in the context of an animal, as shown by the reduction of Aβ42 deposits, their increased mobility and the recovery of their native state.

Example 5

In order to identify the functionally important residues of the isolated peptides, the inventors performed substitution of the position 1 residue with the other two nucleophilic amino acids present in the original libraries, as well as alanine mutagenesis of positions 2-7 of the heptapeptides AβC7-1 and AβC7 -14. As judged by the ability of the generated variants to increase the fluorescence of E. coli cells overexpressing Aβ42-GFR, in the case of both AβC7-1 and AβC7-14, substitution of Cys at position 1 with Ser resulted in 50% decrease in fluorescence, whereas substitution with Thr resulted in similar levels of fluorescence and aggregation of Aβ42-GFR to those corresponding to the respective parent sequence (AβC7-1 or AβC7-14) (Fig. 6A-6B). Furthermore, for both peptides, mutagenesis to alanine at the majority of positions 2-7 resulted in a significant decrease in Aβ42-PPF fluorescence and a concomitant increase in aggregation (Fig. 6A-6B). These observations indicate that several residues in both selected cyclic heptapeptides are necessary for optimal inhibitory activity against Aβ aggregation. Indeed, sequence analysis of all selected sequences belonging to either cluster I or cluster II revealed that the peptides most frequently occurring in each cluster have strong preferences for specific amino acids at each position (Tables 1–2). In particular, for cluster I, Arg and Lys residues at position 2 appeared in >90% of the selected peptides, while Val at position 3, Trp at position 4, Gin, Cys, Ser, Met and Thr at position 5, and Ile , Val and Leu at position 7 occurred in >99% of selected clones (Figs. 6C-6E). Similarly, for cluster II, the frequency of occurrence of Arg, Il, Val, and Gin at position 2 was ~93%, while for lie and Val at position 3, Val at position 4, Pro at position 5, Ser and Ala at position 6 and lie, Leu, and Val at position 7 were >97% (Fig. 6F-6H). Taken together, these results indicate that the most bioactive motifs against problematic Aβ folding and aggregation in the library examined are cyclo-(C,T)(R,K)VW(n,A,M) and cyclo-(C ,T)A(I,V)VP(S,A) Ψ, for complexes I and II respectively, where X is any of the 20 natural amino acids, Π: any of the polar amino acids Q, C, S and T, D: R, I, V or Q, and Ψ: any of the aliphatic amino acids L, V and I.

Example 6

Materials

All enzymes used for DNA cloning were purchased from New England Biolabs with the sole exception of FastAP alkaline phosphatase which was purchased from ThermoFisher Scientific. Recombinant plasmids were isolated using Macherey-Nagel NucleoSpin Plasmid and Qiagen Plasmid Mini and Midi isolation kits. PCR products and DNA extracted from agarose gels were purified using the Macherey-Nagel Nucleospin Gel and PCR Clean-up isolation kit. Synthetically prepared AβC7-1 and AβC7-14 were purchased from Proteogenix (France). Before each experiment, AβC7-1 was dissolved in 50% acetonitrile/0.1% Tween-20 and AβC7-14 was dissolved in 30% acetonitrile/0.025% Tween-20 to make 10 mM solutions. Under these conditions, both cyclic heptapeptides were stable in monomeric form for the duration of the in vitro experiments, as determined by dynamic light scattering (DLS).

Construction of the Combinatorial Library of Cyclic Heptapeptides

For the construction of the plasmids of the pSICLOPPS-Cys X1X2X3X4X5X6, pSICLOPPS-Ser X1X2X3X4X5X6 and pSICLOPPS-Thr X1X2X3X4X5X6 sub-libraries, degenerate forward primers were used in which the amino acids Cys, Ser and Thr were encoded by the codons TGC, AGC and ACC, respectively, while the random amino acids (X) were coded using random NNS codons, where N = A, T, G or C and S = G or C. A second PCR reaction was performed in each case to eliminate mismatches. The resulting PCR products were then digested with BglII and HindIII for 5 h and inserted into a similarly cut and dephosphorylated plasmid pSICLOPPSKanR (see below). Ligase reactions were optimized with a 12:1 insert:vehicle ratio and samples were incubated at 16 °C for 4 h. Approximately 10 µg of the pSICLOPPSKanR vector was used for each sub-library. The recombinant DNA was then purified using centrifugation columns, and used to transform electroreceptive MC1061 cells, which were plated on Luria-Bertani (LB) agar plates containing 25 pg/ml chloramphenicol and incubated at 37 °C for 14–16 h . This procedure resulted in the growth of approximately 1.2 billion independent colonies as judged after serial dilutions on plates.

A PCR reaction on 150 randomly selected clones using intein-specific primers revealed that 82 of them (~55%) contained the correct insert. Overexpressing the IC-peptide-IN-CBD tetrameric fusion protein with 0.002% arabinose in these randomly selected clones and monitoring the production of this fusion protein by western blotting using the rat anti-CBD primary antibody (New England Biolabs, USA, 1:100,000 dilution) and horseradish peroxidase (HRP)-conjugated goat antimouse secondary antibody (Bio-Rad, USA, 1:4000 dilution), showed that 74 of them (~49%) produced high amounts of the tetrameric protein fusion. Among these 74 clones that produced the fusion protein precursor (with a molecular mass of ~25kDa), 67 clones (~45% of the total clones screened) also gave a lower molecular weight band (molecular mass ~20kDa), which corresponds to a from the protein splicing products, IN-CBD, which is released after peptide cyclization. Therefore, according to these results, the generated bacterial libraries encoding the cyclic heptapeptides contain approximately 5,400,000,000 clones, which express the tetramer protein fusion at high levels and which are capable of undergoing protein splicing and potentially producing a cyclic peptide. This diversity fully covers the theoretical diversity of the combined cyclo-Nu libraries X1X2X3X4X5X6(3 x 20<6>= 192,000,000) by nearly threefold (Fig. 1B).

Construction of expression vectors

To construct the helper vector pSICLOPPSKanR, a PCR reaction was performed to amplify the gene encoding aminoglycoside 3'-phosphotransferase (KanR - the enzyme that confers resistance to the antibiotic kanamycin) from pET28a(+) (Novagen , USA) followed by digestion with Bgll and Hindlll and insertion into similarly cut plasmid pSICLOPPS. To construct the pSICLOPPS(H24L/F26A)KanR helper vector, the gene encoding the C-terminal region of the split Ssp DnaE intein (1C) was mutated using mutagenic primers and the pSICLOPPS plasmid as a template. The IC(H24L/F26A) PCR product was then digested with NcoI and Bg11, the KanR gene was digested from pSICLOPPSKanR with Bg11 and HindIII, and the pSICLOPPS vector was digested with NcoI and HindIII. The aforementioned digests (IC(H24L/F26A), KanR and pSICLOPPS) were ligated by a ligase reaction, leading to the production of the desired vector pSICLOPPS(H24L/F26A)KanR.

For the construction of pSICLOPPS (H24L/F26A)-AβC7-1, pSICLOPPS (H24L/F26A)-AβC7-2, pSICLOPPS (H24L/F26A)-AβC7-3, pSICLOPPS (H24L/F26A)-AβC7-7 and pSICLOPPS H24L /F26A)-AβC7-14, plasmids pSICLOPPS-AβC7-1, pSICLOPPS-AβC7-2 pSICLOPPS-AβC7-3, pSICLOPPS-AβC7-7 and pSICLOPPS-AβC7-14 were digested with Bgll and HindIII and the resulting inserts were ligated by reaction ligase with the similarly digested pSICLOPPS (H24L/F26A)KanR vector.

For the construction of pSICLOPPS-AβCλ-1 (C1 S), pSICLOPPS-AβCλ-1 (C1T), pSICLOPPS-AβC7-1 (K2 A), pSICLOPPS-AβC7-1(V3A), pSICLOPPS-AβC7-1(W4A) . , pSICLOPPS-AβC7-14(R2A), pSICLOPPS-AβC7-14(13 A), pSICLOPPS-AβC7-14(V4A), pSICLOPPS-AβC7-14(P5A), pSICLOPPS-AβC7-14(S6A) and pSICLOPPS-AβC7 -14(L7A), a mutagenesis PCR reaction was performed starting from pSICLOPPS-AβC7-1 or pSICLOPPS-AβC7-14 respectively, followed by digestion with BglII and HindIII and ligation into a similarly digested pSICLOPPSKanR vector. pSICLOPPS-AβC7-371, pSICLOPPS-AβC7-405 and pSICLOPPS-AβC7-416 were similarly prepared.

To construct pET-Aβ42, the Aβ42 gene was amplified by PCR from pETAβ42-GFR. The resulting PCR product was digested with NcoI and XhoI and inserted into a similarly digested pET28a(+) plasmid.

Screening of the cyclic heptapeptide library

E. coli Tuner (DE3) electroreceptive cells (Novagen, USA) carrying the pETAβ42-GFR expression vector, which produces Aβ42-GFR under the control of the T7 promoter, were co-transformed with the pooled library of plasmids pSICLOPPS-NuX1X2X3X4X5X6 - Approximately 10<9> colonies carrying both vectors were collected, pooled together and diluted until the optical density at 600 nm (OD600) was approximately equal to 0.1 in a liquid LB culture containing 0.005% L( +)-arabinose to induce the production of cyclic peptides. The culture was incubated at 37 °C with shaking until the OD600 was 0.4-0. 5, at which point 0.1 mM isopropyl-β-D-thiogalactoside (IPTG) was added to the sample to induce Aβ42-GFR overexpression. After 2 h induction at 37 °C the fluorescence of 50,000 cells was recorded using the BD FACSAria II system (BD Biosciences, USA) which included a laser for GFP excitation at 488 nm and a 530/30 band pass filter for detection of the signal. Next, ∼3xl 0<9> cells were gated on a scatter plot of side scatter (SSC-H) versus forward scatter (FSC-H) to eliminate non-cellular events and FACS sorted to isolate bacterial population that exhibited the highest 2% of fluorescence. The isolated cells were recultured and screened for six additional rounds in the same manner, at which point DNA was isolated from the enriched population using the Qiagen Plasmid Mini Kit isolation kit.

Protein / cyclic peptide production in liquid cultures

Tuner E. coli (DE3) cells, freshly transformed with the appropriate expression vectors, were used for all protein production experiments. Single bacterial colonies were used to inoculate liquid LB cultures, which contained 40 pg/mL chloramphenicol and 50 pg/mL kanamycin. These cultures were used at a 1:100 dilution to inoculate fresh LB cultures containing 0.02% y(+)-arabinose and incubated at 37 °C with agitation until their optical density at 600 nm was approximately equal to 0 .4 (OD600 ~ 0.4). At this point, protein production was challenged for 2 h, after addition of 0.1 mM IPTG.

Bacterial cell fluorescence

After protein overexpression, bacterial cells corresponding to 1 mL of culture with OD600 = 1 were collected by centrifugation, resuspended in 100 µL of phosphate-buffered saline (PBS), transferred to a FLUOTRAC 200 96-well microplate (Greiner Bio One International, Austria ) and after excitation at 488 nm, their fluorescence was measured at 510 nm using the TECAN Safire II-Basic microplate fluorometer (Tecan, Austria). For ThS fluorescence measurements, a similar procedure was followed with the exception that cells were resuspended in 250 μM ThS in PBS, equilibrated for 15 min, and then their emission spectra were recorded in the 460–600 nm range. For medium flow cell measurements, after protein overexpression, the fluorescence of 50,000 cells resuspended in PBS was recorded at 530/30 nm after GFP excitation at 488 nm, using the BD FACSAria II system (BD Biosciences, USA).

Succession of new generation

Next-generation sequencing was performed using an Ion Torrent high-throughput sequencing platform. The combined PSICLOPPS-NuX1X2X3X4X5X6 plasmid library and the enriched peptide library after the seventh round of screening were digested with NcoI and BsrGI, and the resulting products of size -250 bp containing the coding region of each peptide were isolated and analyzed. Sequencing results were aligned to a reference sequence using the bowtie2 program (v2.2.8). The alignment information stored in the CIGAR string of the resulting SAM file was analyzed and mapped to match and mismatch sequences using the JVarkit utilities biostar59647 tool. From the resulting XML file, a custom awk script extracted the mismatch sequences, which were then clustered using the CD-HIT tool (v4.6.1) along with their read counts. From the obtained data, only sequences encoding 21 bp peptides with NNS codons were subjected to further analysis. For the enriched peptide library, all sequences that included stop codons were also discarded from subsequent analysis.

Peptide sequence similarity analysis and cluster analysis

Sequence similarity analysis was performed using the Immune Epitope Database (IEDB) clustering tool (http://tools.iedb.org/cluster2/) and the fully connected cliques method. This approach allows all peptides in a clique to share a minimal level of homology, while at the same time a peptide can be part of multiple cliques. As the sequence similarity analysis is performed using only linear sequences, all linear versions of the selected cyclic peptides that occur at least 20 times within the enriched population were identified and considered, resulting in a total of 2912 linear versions of the 416 cyclic heptapeptides. From this analysis, 5087 cliques sharing at least 70% sequence homology were detected, and after rejoining the different linear versions to the original sequence of each cyclic peptide, 617 unique cliques remained. Of the 416 cyclic heptapeptides, 323 participated in these cliques forming a total of 1467 unique pairs with at least 70% sequence homology. The remaining 93 cyclic peptides do not share the minimum homology level of 70% with any other of the selected peptides. Results were then presented in an undirected network graph using Gephi graphical visualization software, and cluster identification was performed using the Girvan-Newman algorithm.

In-gel fluorescence and western blot analyses

After protein overexpression, cells corresponding to 1 mL culture with OD600 = 1 were harvested by centrifugation and resuspended in 100 µL PBS. Samples were lysed with short cycles of sonication on ice and the resulting lysates (total fraction) were centrifuged at 13,000 rpm for 25 min (soluble fraction). The samples were analyzed by electrophoresis on a 10 or 15% polyacrylamide gel under denaturing conditions (SDS-PAGE) or under non-denaturing conditions (native-PAGE) respectively and without prior boiling (for in-gel fluorescence) or after boiling the samples for 10 minutes (for western immunoblotting). In-gel fluorescence was measured using the ChemiDoc-It<2> imaging system equipped with a CCD camera and a GFP filter (UVP, UK), after exposure for approximately 3 seconds. For western immunoblotting, proteins were transferred to PVDF membranes (Merck, Germany) for 1 h at 12 V using a semi-dry transfer device (Thermo Fisher, USA). Membranes were fixed after incubation for 1 h at room temperature in a 5% milk solution in Tris buffered saline containing 0.1% Tween-20 (TBST). After washing with TBST three times, the membranes were incubated for 1 h at room temperature with the appropriate antibody diluted in 0.5% milk solution in TBST. Antibodies used were rat anti-Aβ (6E10) primary antibody (Covance, USA) at a dilution of 1:2000, rat anti-CBD primary antibody (New England Biolabs, USA) at a dilution of 1:100,000 and secondary antibody from anti -goat anti-mouse serum conjugated with horseradish peroxidase (HRP) (Bio-Rad, USA) at a dilution of 1:4000. Proteins were then visualized using the ChemiDoc-It<2> imaging system (UVP, UK ).

Preparation of the monomeric peptide AB42

The recombinant Aβ(M1-42) peptide (MDAEFRHDSGYEVHHQKLVFFAED VGSNKGAIIGLMVGGVVIA), here named Aβ42, was expressed in E. coli and purified as previously described (Habchi J, Arosio P, Pemi M, Costa AR, Yagi-Utsumi M, Joshi P, Chia S, Cohen SI, Muller MB, Linse S, Nollen EA, Dobson CM, Knowles TPJ, Vendruscolo M. Sci Adv. 2016, 2: el501244). The lyophilized peptide was then dissolved in 6M GuHCl, further purified using a Superdex 75 10/300 GL column (GE Healthcare), and eluted in 20 mM sodium phosphate buffer, pH 8, containing 200 μM EDTA and 0.02% NaN3 . Finally, the center of the peak was collected and the concentration of Aβ42 was determined from the absorbance of the peak area using the molar absorbance ε280<=>1.490 M<-1>cm<-1>.

Experiments to study the kinetics of Aβ42 aggregation

Appropriate amounts of the synthetic cyclic peptides were added to 2 μM Aβ42 monomer to obtain the desired molar ratios of cyclic peptide:Aβ42, and samples were supplemented with 20 μM ThT, 1% v/v acetonitrile, and 0.025% or 0.1% v/v Tween. -20 for AβC7-1 and AβC7-14, respectively. Under these conditions, AβC7-1 and AβC7-14 remained stable in the monomeric state for the duration of the in vitro experiments, as determined by dynamic light scattering (DLS) measurements. All samples were prepared in low protein-binding Eppendorf tubes, on ice and with great care to avoid entrainment of air bubbles, and 80 pL of each sample was transferred to three wells of a 96-well, clear-bottom, polyethylene glycol-coated microplate ( PEG), which exhibits low protein binding and where each well is half the area of conventional 96-well microplates (Coming 3881). This microplate with the protein samples was incubated at 37 °C under resting conditions in a microplate fluorescence meter (Fluostar Omega, Fluostar Optima or Fluostar Galaxy, BMG Labtech) and after excitation at 440 nm, ThT fluorescence was measured at 480 nm, through the bottom of the plate.

Blot immunoprint

The samples for this assay were prepared as for the kinetic experiments above with the only difference being the absence of the ThT dye. 2 µl of each sample was removed at different time points and blotted onto a nitrocellulose membrane (0.2 µm, Whatman). Membranes were then dried, incubated with a 5% milk solution in PBST (0.1%), and then incubated with an Aβ42 fibril-specific primary antibody (OC, Millipore) and an Alexa-Fluor 488-conjugated secondary antibody (Life Technologies) according to the manufacturer's instructions. Fluorescence detection was performed using the Typhoon Trio Imager system (GE Healthcare), and fluorescence quantification was performed using ImageJ software (National Institutes of Health).

Transmission Electron Microscopy (TEM)

5 μl of each sample was removed at the end of the Aβ aggregation reaction and placed on a carbon support film on 400-mesh copper grid (EM Resolutions Ltd.). After adsorption, the grids were stained negatively with 2% w/v uranyl acetate, TEM images were captured using the FEI Tecnai G2 transmission electron microscope (Cambridge Advanced Imaging Center) and finally the images were analyzed using the SIS Megaview II Image Capture system (Olympus).

Study of the motility of the nematode C. elegans

Stelegi. The following strains were used for this experiment: (1) GMC101, herein referred to as Aβ worms, genotyped [unc-54p::A-beta-1-42::unc-54 3'- UTR mtl-2p: :GFP], mt1-2p::GFP constitutively expresses green fluorescent protein (GFP) in intestinal cells. unc-54p::A-beta-1-42 expresses Aβ42 in body wall muscle cells, resulting in Aβ42 aggregation and paralysis of the worm after the temperature shift from 20 to 25 °C. (2) N2, wild-type C. elegans var Bristol, referred to herein as wild-type worms.

Virus propagation processes. C. elegans worms were propagated using standard conditions. Briefly, worms were treated with sodium hypochlorite, eggs hatched overnight in M9 buffer (3 g/L KH2PO4, 6 g/L Na2HPO4, 5 g/L NaCl and 1 mM MgSO4) and then were distributed in plates of NGM nematode growth medium [1 mM CaCl2, 1 mM MgSO4, 5 mg/ml cholesterol, 250 mM KH2PO4(pH 6), 17 g/L agar, 3 g/L NaCl and 7.5 g/L casein] the which were infected with E. cob OP50 cells and incubated at 20 °C. After reaching the L4 developmental stage, ∼700 worms were placed in NGM plates containing the desired concentration of the cyclic peptides in a 1% v/v acetonitrile solution. The synthetic cyclic peptides were provided to the worms without additional steps to enhance their permeability. At that point, 75 µM 5-fluoro-2'-deoxyuridine (FUDR) was also added to the plates to inhibit the growth of the progeny worms. The plates were then transferred to 24 °C to promote Aβ42 expression and aggregation.

Mobility test. On days 5 to 10 of adulthood, worms were harvested using M9 buffer and plated on 9 cm NGM plates. The movements of the worms were recorded at 30 frames per second for 1 minute using a proprietary construction including a microscope setup, and the body bends of the worms were quantified using a custom tracking algorithm. In total, ~2300 worms were analyzed for each peptide with an average of ~200 worms per experiment. Overall fitness refers to the sum of worms' mobility and speed.

Quantification of aggregates. Live worms were stained after incubation with 1 μM NIAD-4 (0.1% v/v DMSO in M9 buffer) for 6 h at room temperature and then transferred to NGM plates to be destained for approximately 16 h. Stained worms were then anesthetized by adding 40 mM NaN 3 and mounted on 2% agarose slides on microscope slides. Images were collected using a Zeiss Axio Observer D1 fluorescence microscope (Carl Zeiss Microscopy GmbH) with a 20 x objective and the 49004 ET-CY3/TRITC filter (Chroma Technology Corp), and fluorescence intensity was calculated using ImageJ software (National Institutes of of Health). Only the head region of the worms was calculated in the analysis due to the high non-specific signal in the gut.

Statistical analyses

Statistical analyzes were performed using the Prism program (GraphPad Software Inc., La Jolla, CA, USA) and mean values were compared using the unpaired two-tailed t-test. For animal experiments, group sizes were selected based on previous experience and the literature, so that sufficient numbers remained at the end of the experiment. No samples, worms, or experimental data were excluded from the reported analyses.

While the invention has been described with respect to specific embodiments, it is apparent that modifications are possible without departing from the scope of the invention.

Claims (16)

CYCLIC PEPTIDE INHIBITORS OF AMYLOID-B PEPTIDE AGGREGATION Claims 1. A peptide consisting of the amino acid sequence NuX1X2X3X6, where: Nu is C, S or T; Is X1 any amino acid other than C, M, D, E, P, and G? Is X2 A, V, W or P? Is X3 A, L, F, W, Y, S, T, R or G? Is X4 any amino acid other than I, L, F, W, N, D, and K? Xs is any amino acid other than Y, N, R and K? X6 is any amino acid except Y, N, Q, S, D, E, and H. 2. The peptide of claim 1, wherein N y is C or T; Is X1 R or K? X2is V? X3is W? Is X4 Q, C, M, S or T? Is X5 A, L, V, T, E or P? X6 is L, V or P. 3. A peptide consisting of the amino acid sequence NyX1X2X3X4X5X6, where: Ny is C, S or T; Is Chi any amino acid other than A, F, N, C, S, H, P, and G? Is X2 I, L, V, F, W, Y, M or T? Is X3 I or V? Is X4 C, P or G? Is X5 A, S, T or P? X6 is I, L, or V. 4. The peptide of claim 3, wherein N y is C or T; Is X1 I, L, Q or R? Is X2 I or V? X3is V? X4 is P? Are Xs A or S? X6 is I or L. 5. The peptide of any one of claims 1 to 4, wherein the amino acid sequence is further defined by any of ALA. ID NO. 1-132. 6. The peptide of any one of claims 1 to 5, wherein the peptide consists of any of the amino acid sequences CKVWQLL, CKVWMPL, CRVWTEL, CRIVPSL or CIVVPSI. The peptide of any one of claims 1 to 6, wherein at least one residue has a D configuration. The peptide of any one of claims 1 to 7, wherein the peptide is cyclic. 9. A hybrid molecule consisting of (1) the peptide of any one of claims 1 to 8 and (2) a polypeptide molecular template. 10. The hybrid molecule of claim 9, wherein the molecular template consists of all or a sufficient part of a protein belonging to any of the classes of antibodies, enzymes, chromogenic proteins, fluorescent proteins and parts thereof. 11. The peptide of any one of claims 1 to 8 or the hybrid molecule of any one of claims 9 to 10 for use in the treatment, diagnosis or prevention of Alzheimer's disease. 12. A pharmaceutical composition consisting of the peptide of any one of claims 1 to 8 or the hybrid molecule of any one of claims 9 to 10 and a pharmaceutical excipient. 13. The pharmaceutical composition of claim 12 for use in the treatment, diagnosis or prevention of Alzheimer's disease. A plasmid vector containing the deoxynucleotide sequence encoding the peptide of any one of claims 1 to 8. The plasmid vector of claim 14, wherein the plasmid vector is an expression plasmid. A prokaryotic or eukaryotic host cell containing the plasmid vector of any one of claims 14 to 15.