Review Article

Modelling Tauopathies in Drosophila: Insights from the Fruit Fly

Table 1

Drosophila genetic techniques used in tauopathy models.

MethodPurposeHow it worksExamples

Gal4/UAS [19, 28]Allows tissue-specific expression of the gene of interest, in a modular fashion.One line of flies with Gal4 protein under a promoter expressed in the tissue of interest is crossed with another line harbouring the gene of interest downstream of GAL4-binding UAS DNA sequences and a minimal promotor. In F1 flies with copies of both transgenes, Gal4 will bind to UAS and drive expression, and thus the gene of interest will be expressed only in the tissue of interest.The majority of Drosophila models of tauopathy have used this technique. Tau constructs of different species and isoform, and with a variety of mutations, have been expressed in retina with GMR-, Ey- or Sev-Gal4 drivers [15, 27, 2946]; motor neurons with D42- or OK6-GAL4 [35, 45, 4749]; sensory neurons with C161- or el6E2-GAL4 [50]: mushroom body neurons with c772- or C492-GAL4 [45, 51]; dopaminergic neurons with Ppk- or ddc-GAL4 [38, 52]; cholinergic neurons with Cha-GAL4 [43, 53]; notum with Eq-GAL4 [38]; glia with repo-GAL4 [25] or panneuronally with Elav- or Appl-GAL4 [30, 40, 42, 43, 45, 49, 5263].

Gal80TS (TARGET) [21, 64]Provides temporal control of the Gal4/UAS system.Adding another layer to the Gal4/UAS system to drive a gene of interest at the time and place of the experimenter’s choosing. Gal80TS is a temperature-dependent repressor of Gal4. Under permissive conditions (18°C), Gal80TS functions normally and blocks Gal4-mediated transactivation of UAS-transgenes. At 29°C, Gal80TS is dysfunctional and therefore allows GAL4 to function normally.Colodner and Feany (2010) used this technique in order to cease expression of human tau in glia after NFTs had already formed (see text) [25]. Papanikolopoulou et al., (2010) used TARGET to study temporal effects of tau phosphorylation in mushroom body neurons [65].

EP lines [66]A large set of lines of flies with gain or loss of function in numerous genes can be screened for interaction.Enhancer-promotor-transposable (EP) elements are inserted into the genome at random. They contain GAL4-binding UAS sequences and a promotor. When they land near a gene in the same direction, their activation by Gal4 may promote that gene’s transcription in the tissue of interest.These lines have been screened for modifiers of rough-eye phenotype in flies expressing human tau in the retina [32, 35]. Others have screened the set of lines for modifiers of A-beta [67] or Ataxin-3 [44] toxicity, then coexpressed the hits with tau to identify commonalities.

UAS-dsRNAi lines [68, 69]A comprehensive set of fly lines to knockdown the expression of Drosophila genes under GAL4/UAS control.Interfering double-stranded RNA for a Drosophila gene is expressed under UAS. Thus, when crossed with a Gal4 driver, the relevant protein will be knocked down in the tissue of interest. This can also be combined with Gal80TS for temporal control, or the lines used for an enhancer/suppressor screen.Individual UAS-RNAi lines have been used either to knock down tau, or to demonstrate a genetic interaction with tau [46, 52, 57, 70, 71]. A set of 19 lines were used in a secondary screen for modifiers of the tau-induced rough-eye phenotype [27].

MARCM (Mosaic analysis with a repressible cell marker) [72]Generates a mosaic animal with GFP-marked clonal cells homozygous for an allele of interest, for direct comparison with non-GFP heterozygous or wildtype cells in the same animal.MARCM relies upon Flp/FRT-mediated homologous recombination in mitotic cells to generate clonal subsets of daughter cells that are either (i) marked with GFP and homozygous for an allele of interest or (ii) heterozygous or wildtype for the allele and unmarked by GFP. Parental cells contain the allele of interest and Gal80 on homologous chromosomes distal to FRT sites as well as hsFLP, UAS-GFP, and a cell/tissue-specific Gal4 driver. After recombination, daughter cells which are homozygous for the allele of interest have no Gal80 and will thus express GFP, while nonmarked clones will be wild type or heterozygous.Nishimura et al., (2004) used this technique to show that par-1 is required for tau toxicity. Using elav-Gal4, they made clones of neurons that were mutant or heterozygous for par-1 in the presence or absence of human tau overexpression in each combination [43]. This would otherwise have been difficult to test, as par-1 null flies are not viable.