FIELD OF THE INVENTION The present invention relates to the field of chemical compound assay technology, more specifically to a novel, potentially faster screening assay for chemical compounds that may be neuroprotective in Alzheimer's, Parkinson's or other human neurodegenerative diseases, and in particular to a novel, optionally fast screening assay for chemical compounds that may be neuroprotective against the toxic effects of human alpha synuclein (UniProt: P37840, Ensembl.ENSGOOOOOI 45335), and/or Abeta42 peptide (for the latter, whether delivered directly or derived from human APP, UniProt P05067, Ensembl.ENSGOOOOOI 42192).

BACKGROUND TO THE INVENTION

The field of drug development against neurodegenerative diseases is making slow progress due to the biological complexity of the disease processes involved, which means that only in vivo tests are really valid, and due to the fact that these are primarily diseases of aging so that currently available in vivo assays are very slow and thus not suited to drug screening. Adult Drosophila (imagos) have been used for low to medium throughput assays for human neurodegenerative diseases, as they show remarkable similarity to the human neurodegeneration process for Alzheimer's, Parkinson's, Huntington's and other neurodegenerative diseases, and data can be obtained in 20 to 35 days, depending on the severity of the effect in each model. Nevertheless, any method that allowed faster in vivo drug screening would be very important for the field of neuroprotective drug development. A method for screening small molecules in genetically modified Drosophila has been described: U.S. Pub. No: US 2002/0026648 A1 . In this method, compounds are injected into the hemolymph of Drosophila adults. The injection procedure is technically difficult and is not suited to medium to high throughput chemical screening. Adult Drosophila may also be provided with chemicals in their food, but adult Drosophila expressing human genes implicated in neurodegenerative disease must be aged for up to 40 days before a reliably measurable phenotype is observed.

Accordingly, a need remains for a practical method to allow faster drug screening in Drosophila. Specifically, there is a need for method that allows medium to high throughput chemical screening of chemicals for neuroprotective effects against APP derived Abeta 42 (for Alzheimer's disease) and/or alpha synuclein and its mutant variants (for Parkinson's disease). The invention described here relates to the use of Drosophila larvae to allow a significantly faster screen of chemicals than any other in vivo screening method.

BRIEF SUMMARY OF THE INVENTION

The invention is based on the unexpected discovery that Drosophila larvae expressing human Abeta42, mutant forms of Abeta42, and/or human alpha synuclein, or its mutant forms, in the nervous system show significant changes in locomotion and/or learning despite the very short time allowed for symptoms of nerve damage to develop. This has allowed the establishment of an assay procedure in which Drosophila larvae expressing human Abeta42, mutant forms of Abeta42, and/or human alpha synuclein, or its mutant forms, in the nervous system are provided with medium containing a certain concentration of a chemical compound, following which their movement/response is assessed under standard conditions. Other larvae are assessed without the chemical compound added to their food, and yet another control is provided by larvae from wild type Drosophila, including a Drosophila strain identical to the test strain except that it lacks the human Abeta42 or alpha synuclein. Data can be obtained in five days that requires 20 to 35 days using adult Drosophila and many months in rodents, many more tests can be run with the same manpower, and the method is amenable to automation, producing even higher throughput: this allows a real high throughput in vivo screen of chemicals leading to neuroprotection against either the effects of alpha synuclein or Abeta42.

DETAILED DESCRIPTION OF THE INVENTION

Drosophila stock maintenance, rearing and genetic crossing are carried out on standard cornmeal medium. In all studies described here, Drosophila were kept at 25°C with a 12 hour dark, 12 hour light cycle. Drosophila eggs from appropriate Drosophila strains are obtained by placing a suitable number of males and females (for example, 10 and 20 respectively) in a population cage with a "red wine plate" containing an agar-grape juice mixture, which is changed at regular intervals to collect eggs. Drosophila larval rearing procedures are described, for example, in Cold Spring Harbor Protocol 2012, doi:10-1 101/pdb.prot069302. Eggs are collected, allowed to hatch and larvae grown on Jazzmix medium (Applied Scientific Products AS153) or any other suitable Drosophila food medium containing either 1 % DMSO as a control to allow for solvent effects, or 1 % DMSO together with a suitable concentration of the chemical to be tested. At a suitable stage, for example third instar, larvae are transferred to black agar plates (for example 9 cm plates). The agar recipe is exemplified as 20 grams of Agar (for example, Sigma Agar A-7002), 25 grams of sucrose, 1 .5 grams of Nipagin A (for example, from Clariant UK), 125 millilitres of red wine concentrate, 6 grams of activated charcoal, and 875 millilitres of water. Each black agar plate is prepared with a ring of high salt agar around the edge (using the same recipe with 5M sodium chloride) to inhibit the larvae from leaving the field of view. Larvae are allowed to adapt for 45 seconds, then repositioned to the central area of the plate and video recording initiated, where the video recorder is linked to a computer running suitable tracking software. A tracking system, for example, Actual Analytics Actual Track version 3.0, is used to follow and record larval tracks, and the distance moved in 6 minutes, or other suitable period, is used to calculate mean larval velocity. The number of larvae per plate is adjusted to plate size and tracker capability: in the examples typically 6 larvae were placed on a 9 cm plate. One-way ANOVA with Bonferroni connection (95% CI) statistical analysis is carried out using GraphPad (Prism6) or other suitable analysis software. Statistically significant differences between chemical + DMSO treated neurodegeneration test larvae and DMSO-only treated neurodegeneration test larvae, with appropriate control data with controls showing no significant differences between chemical + DMSO treated neurodegeneration control larvae and DMSO-only treated control larvae, indicate a potentially neuroprotective or otherwise beneficial effect specific to the human protein or peptide that is expressed in the neurodegeneration test larvae.

Drosophila strains are prepared that express a human protein or proteins in all or some neurons in the Drosophila larval nervous system, using methods well known to those skilled in the art, for example as in Feany and Bender 2000, Nature 404, 394-8. For example, the test strain may be one expressing human wild-type alpha synuclein, or the human alpha synuclein mutants A30P, A53T, E46K, G51 D, H50Q, artificial alpha synuclein mutants such as TP S (A30P, A56P and A76P), or any combination of these, in the larval nervous system, or a strain expressing human wild-type Abeta42 Arctic (E693G) or other forms of the Abeta peptide in the larval nervous system. For example, the genetic construct to be expressed may be linked to the yeast UAS sequence, expression from which is driven by the yeast Gal4 transcription factor, as in Brand A.H. and Perrimon N., 1993: the temporal and cell/tissue expression pattern of the Gal4 protein is determined by the gene regulatory sequences in its vicinity whether naturally occurring or artificial, and in turn determines the temporal and cell/tissue expression pattern of the human protein or peptide. For example, the elav-Gal4 driver construct is used to obtain wide expression in most neurons; Yao K.M. and White K. 1994. Other approaches to gene expression control known to those skilled in the art may also be used. In one version of such strains, the human protein or peptide is inserted into a known cpc31 insertion (att) site, Markstein M. et al., 2008: the att strain without the inserted expression construct is then used as an otherwise genetically identical control strain.

Automated Screening. Preparation of the microtiter or other plates with the growth medium, addition of candidate compounds with or without DMSO or other solvent, addition of Drosophila eggs and larval handling can be performed manually or using a robotic system or systems, such as COPAS systems, for example as described in http://www.unionbio.com/documents/DrosophilaANS01 .pdf. For example, plating of the growth medium and of candidate compounds in solution on the microtiter or other plates can be readily adapted to known robotic systems that can be configured to repeatedly inject a predetermined volume of the growth medium and of the test solutions into each well of the microtiter plate. Similarly, the operation of the video tracking system assay can be done manually, or can be automated.

EXAMPLE 1 Example 1 relates to a method of testing chemicals for neuroprotective or otherwise useful activity as treatments for Alzheimer's disease. In this example, a Drosophila strain expressing human Abeta42 Arctic mutant peptide under control of the pan- neuronal Elav-Gal4-C155 Gal4 driver construct was tested. Figure 1 shows a significant reduction in crawling speed of the Abeta42 Arctic larvae with reference to the v525 strain, which is genetically identical to the Abeta42 expressing strain except that it lacks the human Abeta42 Arctic mutant expression construct. A Drosophila strain expressing a carboxy-terminal fragment of APP (from the beta secretase cleavage site to the carboxy terminus, including the Abeta42 sequence) together with human tau, under control of the pan-neuronal Elav-Gal4-C155 Gal4 driver construct, also shows a significant reduction in crawling speed with reference to the v525 strain.

Figure 1 shows Larval crawling speed with exact genetic control for the human Abeta42 Arctic expressing strain.

The same test strains (human Abeta42 Arctic and human APP C-terminus/tau) were tested for larval crawling speed in the presence of DMSO. 1% DMSO was added to the medium to simulate the conditions of adding a test chemical in DMSO as a solvent.

Figure 2 shows that both strains showed significant reductions in larval crawling speed compared to a wild type Oregon R strain. Data here show Larval crawling speed with 1 % DMSO added (a vehicle for drug administration)

Figure 3 shows a similar experiment to that described and shown for Figure 1 , with no DMSO and Oregon R as a wild type control. Both strains showed significant reductions in larval crawling speed compared to a wild type Oregon R strain. Figure 4 shows a similar experiment to that described and shown for Figure 2, with 1 % DMSO and OregonR as a wild type control. Both strains showed significant reductions in larval crawling speed compared to a wild type Oregon R strain. This experiment shows that adding the Elav-Gal4-C155 driver to the wild type control (OreRxC155-D) does not change the significance of the difference in crawling speed.

Figure 5 demonstrates an example of chemical partial rescue of the reduced crawling speed phenotype for the human human Abeta42 Arctic strain. Drosophila larvae in the light blue column of Figure 5 were grown on medium with 1 % DMSO alone, whereas Drosophila larvae in the dark blue column of Figure 5 were grown on medium with 1 % DMSO and 100 micromolar tacrine.

Figure 6 shows a similar partial chemical rescue for the human APP C-terminus/tau expressing strain It should be noted that, in the studies that yielded the data of Figures 3 to 6, the Drosophila larvae were grown on a slightly different version of Drosophila growth medium, which used standard Drosophila food mix known to all those skilled in the art, instead of Jazzmix. EXAMPLE 2

Example 2 relates to a method of testing chemicals for neuroprotective or otherwise useful activity as treatments for Parkinson's disease. In this example, Drosophila strains expressing various mutant forms of human alpha synuclein peptide under control of the pan-neuronal Elav-Gal4-C155 Gal4 driver construct were tested: human alpha synuclein H50Q, human alpha synuclein E46K, and human alpha synuclein TP S triple mutant (A30P/A56P/A76P). Each UAS-alpha synuclein construct was inserted into the AttP40 site, so that the AttP40 strain provides an exact genetic wild type control. 1 % DMSO was added to the medium to simulate the conditions of adding a test chemical in DMSO as a solvent. Figure 7 shows a significant reduction in crawling speed of the human alpha synuclein H50Q, human alpha synuclein E46K, and human alpha synuclein TPaS triple mutant (A30P/A56P/A76P) relative to the attP40 exact genetic wild type control larvae.

Figure 8 shows an otherwise identical study to that shown in Figure 7, with the human alpha synuclein expressing strains compared to the wild type Oregon R strain for larval crawling speed on 1 % DMSO. This study also shows a significant reduction in crawling speed of the human alpha synuclein H50Q, human alpha synuclein E46K, and human alpha synuclein TPaS triple mutant (A30P/A56P/A76P) relative to the attP40 exact genetic wild type control larvae.

Example 3 We performed an experiment where we expressed the human Abeta42 Arctic peptide into a subset of larval neurons, specifically Kenyon cells in the mushroom bodies using PGAL4 strain MB247. These neurons are widely known to be important in olfactory associative learning in insects. When we do this, we see a defect in the ability of the larvae to learn.

To test this we used the protocol for odour-taste learning assay in Drosophila larvae, described by Gerber et al (2013). For these specific experiments we used the reciprocal principle described by Gerber et al (2013): two groups of animals were trained at the same time, reciprocally with two different odours (in our case 1 -Hexanol and mineral oil, Selcho et al 2009) rewarding respectively one of the two odours presented per group with the presence of a Fructose reward in the agar plates. The genotype tested are w; MB247Gal4/iso2; UAS-Ap42[Arc]/+ (as disease model) and w; MB247Gal4/iso2; iso3/+ (as its respective genetic control). The disease model larvae show a subtle, but significant defect in larval learning when compared to controls (figure 9).

Figure 9: The disease model flies (PI AB42) show subtle but significant decrease in their performance in the learning task when compared to controls (PI Gen. Ctrl)

In addition to the data described above, Ping et al 2015, published a set of results using very similar strains and protocols. Their results also show a learning defect in larvae expressing human neurodegenerative proteins (Ping et al. , figure 6A).

Example 4 We performed an experiment where we expressed the human Abeta42 Arctic peptide into a subset of larval neurons, specifically Kenyon cells in the mushroom bodies and then examined the structure of the neurons and their ability to form processes that were outwardly normal. Briefly, brains from the larvae were dissected and stained for human Abeta42 and compared to control brains where there was no human Abeta42 Arctic peptide. We observed that the human Abeta42 Arctic peptide was present in the larval brain. We saw no obvious evidence for damage at the level of the brain structure in these animals. In particular we looked at key brain structures including the mushroom bodies where we described the learning defect above but they appear to be normally formed. This suggests that the behavioural changes we see in the larvae are more likely to be acute toxic effects of human Abeta42 rather than the pathological effects of downstream neurodegeneration (at this age of animal).

References cited. All references cited above are incorporated herein by reference in their entirety and for all purposes to the same extent as if each individual publication, patent or patent application was specifically and individually indicated to be incorporated by reference in its entirely for ail purposes. Citation of a reference herein shall not be construed as an admission that such is prior art to the present invention. Alpha synuclein. http://www. u n iprot . org/u n prot/ P37840 SNCA:Ensembl.ENSG00000145335

http://www.ensembl.org/Homo_sapiens/Gene/Summary?g=ENSG00 000145335;r=4:89 724099-89838315

Amyloid beta A4 protein (APP).,hi¾3^ APP gene Ensembl. ENSG00000142192. http://www.ensembl.org/Homo_sapiens/Gene/Sequence?g=ENSG0000 0142192;r=21 : 25880550-26171 128

Preparing Drosophila larvae for feeding assays. Cold Spring Harbor Protocol 2012, doi:10-1101/pdb.prot069302

The "Arctic" APP mutation (E693G) causes Alzheimer's disease by enhanced Abeta protofibril formation. Nilsberth, C, Westlind-Danielsson A., Eckman C.B., Condron M.M., Axelman K., Forsell C, Stenh C, Luthman J., Teplow D.B., Younkin S.G., Naslund J., Lannfelt L. 2001 . Nature Neuroscience 4, 887-893.

A Drosophila model of Parkinson's disease. Feany M.B. and Bender W.W. 2000, Nature 404, 394-8.

Targeted gene expression as a means of altering cell fates and generating dominant phenotypes. Brand, A.H. and Perrimon, N. 1993. Development 1 18, 401 -415.

Neuronal specificity of elav expression: defining a Drosophila promoter for directing expression to the nervous system. Yao, K.M. and White, K. 1994. J. Neurochemistry 63, 41 -51 .

Exploiting position effects and the gypsy retrovirus insulator to engineer precisely expressed transgenes. Markstein M., Pitsouli C, Villalta C, Celniker S.E. and Perrimon N., 2008. Nature Genetics 40, 476-83. COPAS http:/ mvw.unionbjo.com/documents/DrosophiiaANS01 .pdf. COPAS Application Note S-01 , Rapid Drosophila Embryo Sorting.

Ping Y, Hahm E-T, Waro G, Song Q, Vo-Ba, D-A, Licursi A, et al. (2015) Linking Αβ42- Induced Hyperexcitability to Neurodegeneration, Learning and Motor Deficits, and a Shorter Lifespan in an Alzheimer's Model. PLoS Genet 1 1 (3): e1005025. doi:10.1371/journal.pgen.1005025

Gerber, B., Biernacki, R., & Thum, J. (2013). Odor - Taste Learning Assays in Drosophila Larvae Odor - Taste Learning Assays in Drosophila Larvae. Cold Spring Harbor Protocols doi:10.1 101/pdb.prot071639