Systematic discovery of Rab GTPases with synaptic functions in Drosophila

Abstract

Background: Neurons require highly specialized intracellular membrane trafficking, especially at synapses. Rab GTPases are considered master regulators of membrane trafficking in all cells, and only very few Rabs have known neuron-specific functions. Here, we present the first systematic characterization of neuronal expression, subcellular localization, and function of Rab GTPases in an organism with a brain.

Results: We report the surprising discovery that half of all Drosophila Rabs function specifically or predominantly in distinct subsets of neurons in the brain. Furthermore, functional profiling of the GTP/GDP-bound states reveals that these neuronal Rabs are almost exclusively active at synapses and the majority of these synaptic Rabs specifically mark synaptic recycling endosomal compartments. Our profiling strategy is based on Gal4 knockins in large genomic fragments that are additionally designed to generate mutants by ends-out homologous recombination. We generated 36 large genomic targeting vectors and transgenic rab-Gal4 fly strains for 25 rab genes. Proof-of-principle knockout of the synaptic rab27 reveals a sleep phenotype that matches its cell-specific expression.

Conclusions: Our findings suggest that up to half of all Drosophila Rabs exert specialized synaptic functions. The tools presented here allow systematic functional studies of these Rabs and provide a method that is applicable to any large gene family in Drosophila.

Figure 1. Design of P[acman]-KO: Combining BAC Recombineering, PhiC31 transgenesis, and ends-out homologous recombination

(A) Vector design. First, a [Frt, ISce1] cassette was inserted into the AscI and PacI sites of attB-P[acman]-Ap^R to add mobilization capability for endogenous targeting. Second, a Gateway^TM cassette for bacteriophage λ–mediated BP recombination was introduced with new AscI and PacI sites. Hence, homology arm cassettes can either be integrated using AscI, PacI conventional cloning or by including Gateway^TM attB sites into the primers used to create such cassettes. See Supplemental Materials for more details. All other features of the vector are described in Venken et al., (2006). (B) Design of the Gal4 knock-in cassette. This 6.7kb cassette is adapted for desired loci by including gene-specific 100bp homology arms in the primers used to amplify the cassette. Note that optimized PCR conditions are identical for any primer pair (see Suppl. Materials). The floxed 3×P3-RFP, Kan cassette serves for selection during bacterial recombineering cloning (Kan) and positive selection of the targeting cassette in a homologous recombination experiment in vivo in the fly (3×P3-RFP) and can be removed in transgenic or gene-targeted flies easily by crossing to available Cre strains. (C) Strategies for ORF and ATG knock-ins. The Gal4 knock-in cassette replaces the complete open reading frame in ORF knock-ins, whereas in ATG knock-ins only the start codon and the remaining part of the start-codon containing exon are replaced. Note that in both cases the ATG of Gal4 replaces the ATG of the rab gene. Brain: L3 larval brain; Eye: L3 eye disc; wing: L3 wing disc; leg: L3 leg disc; Gland: L3 salivary gland. Scale bar for each tissue: 100μm.

Figure 2. Systematic Expression Profiling in larval tissues reveals that half of all rab GTPases in Drosophila are neuron-specific or neuron-enriched

Shown are the five larval tissues brain, eye disc, wing disc, leg disc and salivary gland (from left to right) for a total of 23 rab-Gal4 lines crossed to UAS-CD8-GFP (green). Toto-3 labels cell bodies/nuclei (blue), and the 3×P3-RFP cassette from our knock-in cassette marks the termini of the larval photoreceptor organs in the brain (red). On the top left five control Gal4 lines are shown: act-Gal4 and tub-Gal4 (both showing ubiquitous expression), elav-Gal4 (showing expression in developing and functional neurons as well as low levels in some other cells), n-syb-Gal4 (showing panneurononal expression), and repo-Gal4 (showing expression in all glial cells). All rabs are sorted from the most neuron-specific in red (starting with rab3 and rabX4), via lines with somewhat specialized patterns in grey to the most ubiquitous Gal4 driver lines in green (rab5-Gal4 and rab11-Gal4).

Figure 3. rab-Gal4 expression patterns in the pupal and adult brain

(A) Pupal brains (30-40% pupal development). Shown are maximum projections with CD8-GFP driven by the denoted rab-Gal4 lines (green) and nuclear labeling with Toto-3 (blue). The top left corner shows a schematic with a few prominently labeled landmark structures: the developing eyes (red), glia (yellow), and developing mushroom bodies and antennal lobes (green). (B) Adult brains. Shown are partial maximum projections of 20μm depth of the anterior adult brain on top and 20μm depth of the posterior brain on the bottom. CD8-GFP is driven by the denoted rab-Gal4 line (green), the 3×P3-RFP marker from the knock-in cassette labels the photoreceptor projections(red) and Toto-3 labels nuclei (blue). The top left corner shows an anterior brain schematic with the lamina (red) and the mushroom bodies and antennal lobes (green), and a posterior brain schematic showing the cell bodies of the Kenyon cells that form the mushroom body (green) and the optic neuropils medulla and lobula complex (red). Glia is marked in yellow. The different rab-Gal4 lines drive expression in these structures with highly varying intensity. Scale bar for all panels: 100μm.

Figure 4. Subcellular Localization Profiling of YFP-tagged Rab proteins expressed under control of their endogenous regulatory elements

All neuronal rabs encode synaptic proteins that colocalize with recycling endosome or synaptic vesicle markers. Double immunolabelings of the posterior larval brain ventral ganglion at high resolution are shown for selected YFP-Rabs driven by their respective rab-Gal4 lines (green), anti-Rab11 (red, recycling endosomes) and anti-Rab5 (blue, early endosomes) labeling in the first column and anti-CSP (red, synaptic vesicles) and anti-Rab7 (blue, late endosomes) in the second column. Cell bodies are peripherally and synaptic neuropils centrally located. Single channels of the colocalizing labels are depicted in the two columns on the right. Shown are only the seven most neuron-specific lines and rab5, rab7 and rab11 as controls; see Figure S5A for the complete dataset. Arrows point to colocalizing compartments. Scale bar for all panels: 20μm.

Figure 5. Subcellular Localization Profiling as a function of GTP- and GDP-bound states

Proximal ventral ganglion sections are shown sorted for all Rabs from most neuronal (top, red) to most ubiquitously expressed (bottom, green) similar to Figure 4. Corresponding Gal4-lines drive the expression of wild type YFP-tagged Rabs in the left column, constitutively active (GTP-bound) YFP-tagged Rabs in the middle column and dominant negative (GDP-bound) YFP-tagged Rabs in the right column. Toto-3 labels nuclei (blue). Note that most neuronal Rabs show synaptic localization and little or no cell body localization that is maintained and further enriched as constitutively active, but lost as dominant negative versions. Further note that only expression of the dominant negatives of Rab1, Rab5 and Rab11 cause embryonic or early larval lethality, while Rab35 dominant negative is semilethal with few adult escapers. A high-resolution version of this Figure is available online. Scale bar for all panels: 50 μm.

Figure 6. Generation of a rab27 knock-out by ends-out homologous recombination of the Gal4 knock-in cassette reveals a specific sleep phenotype

(A) PCR verification of 3×P3-RFP positive targeting cassette mobilizations and re-integrations in the genome indicate loss of the endogenous rab27 locus for three out of 12 potential knock-out lines. (B) Verification of two of the knock-outs by Southern with an ORF probe (comp. Suppl. Fig. 6). (C, D) rab27-Gal4 in the landing site and rab27^Gal4-KO in the endogenous site exhibit very similar expression patterns (CD8-GFP in green) specific to the mushroom bodies in the adult brain. (E, F) YFP-tagged Rab27 expression is identical between YFP-Rab27 in wild type using rab27-Gal4 and YFP-Rab27 rescuing expression in a homozygous null mutant with rab27^Gal4-KO knock-in. (G, H) Sleep phenotype in rab27^Gal4-KO flies. rab27^Gal4-KO flies show decreased daytime sleep bout length (median, quartiles, 90^th percentile) compared to controls (G), a phenotype rescued by introducing the UAS-Rab27-YFP transgene (H). (I, J) Representative single fly sleepograms showing decreased bout length. Each bar represents a sleep bout, with the height indicating the duration (Y-axis =120 min, X-axis = 24 hours, light/dark cycle indicated by white/black boxes). rab27^Gal4-KO flies have a decreased number of long, >60 min sleep bouts (indicated by asterisk), especially during the daytime, compared to rab27^Gal4-KO;UAS-rab27⁺-YFP rescue flies.