Flybow genetic multicolor cell labeling for neural circuit analysis in Drosophila melanogaster

Transcription

1 Nature Methods Flybow genetic multicolor cell labeling for neural circuit analysis in Drosophila melanogaster Dafni Hadjieconomou, Shay Rotkopf, Cyrille Alexandre, Donald M Bell, Barry J Dickson & Iris Salecker Supplementary Figure 1 Supplementary Figure 2 Supplementary Figure 3 Supplementary Figure 4 Supplementary Figure 5 Supplementary Table 1 Recombination specificity and efficiency of the mflp5/mfrt71 system. Activity of FB1.1 in the visual system. Quantification of recombination events for FB1.1 and FB2.0. Expression pattern of MzVum-Gal4 in the adult optic lobe. Schematic representation of the cloning strategy used to generate Flybow constructs. Quantification of recombination efficiency of Flp recombinases.

2 Supplementary Figure 1 Recombination specificity and efficiency of the mflp5/mfrt71 system. Flies expressing Flp, mflp4, mflp5, mflp6 or mflp7 under the control of the eyeless (ey) enhancer were crossed to act5c> -tub 3 UTR >nlacz lines, in which the stop cassettes were flanked either by pairs of wild-type or one of three modified FRT sites (mfrt11, mfrt71 and mfrt11-71). Inverted repeats are shown in upper case and asymmetric spacer regions in lower case letters. Single base pair changes are highlighted in green. The orientation of symbols > and indicate the polarity of wild-type FRT and mfrt71 sites, respectively. Third instar larval eye imaginal discs were labeled with anti- - Galactosidase ( -Gal, red). Asterisks indicate the eye field. Flp efficiently excises the stop cassette flanked by wild-type FRT sites (a), but does not recognize mfrt71 sites (b). mflp5 mediates very low levels of recombination between wild-type FRT (arrowheads) (c), but is highly efficient for mfrt71 sites (d). See also Supplementary Table 1. Scale bars, 50 μm.

3 Supplementary Figure 2 Activity of FB1.1 in the visual system. elav-gal4 c155 in conjunction with FB1.1 drives expression of four FPs in neurons of the third instar larval (3L) optic lobe. Panel h of Fig. 2 shown as merged four-color image (a) and as separate images of the four channels in grey scale (b-e). Color-coded asterisks and arrowheads highlight lineagerelated and single neurons, respectively to facilitate comparison across panels. la, lamina; ln, lamina neurons; me, medulla; mn, medulla neurons. Scale bar, 50 μm.

4 Supplementary Figure 3 Quantification of recombination events for FB1.1 and FB2.0. (a) Recombination events induced by mflp5 in the optic lobe of animals expressing FB1.1 under the control of the pan-neuronal driver elav-gal4 c155 after exposure to three 30 min heat shocks at 48, 72 and 96 hours AEL. Numbers of neurons expressing the four FPs were obtained from three z sections of 10 optic lobes, corresponding to 3,367 ± cells (mean and 95% confidence interval) per sample. Quantification of percentages indicated that an average of 48.2% of neurons expressed EGFP, 16.3% mcitrine, 17.2% mcherry and 18.3% Cerulean-V5. While EGFP is most abundantly expressed (P < , unpaired two tailed t-test), the differences in percentages of mcitrine, mcherry and Cerulean-V5 expressing cells are not statistically significant (P > 0.58), indicating that these are expressed with similar probability. (b) Recombination events in the optic lobe of animals expressing FB2.0 under the control of elav-gal4 c155 after exposure to two 90 min heat shocks at 48 and 72 hours AEL. Numbers of neurons expressing the four FPs were obtained from 10 optic lobes (three z sections, n = 9; two z sections n = 1), corresponding to ± labeled cells (mean and 95% confidence interval) per sample. This confirms that FB2.0 in conjunction with Flp and mflp5 leads to sparse labeling. An average of 73.1% of neurons expressed EGFP, 4.3% mcitrine, 17.5% mcherry and 5.1% Cerulean-V5. EGFP is most abundantly expressed (P < , t-test). Frequencies of mcitrine, mcherry and Cerulean-V5 expressing cells are highly variable and differences are not statistically significant (P > 0.06, t-test). The histograms and error bars (a,b) show mean percentages and 95% confidence intervals. Statistical analysis was performed using Excel.

5 Supplementary Figure 4 Expression pattern of MzVum-Gal4 in the adult optic lobe. (a,b) In the absence of hs-mflp5 activation, MzVum-Gal4 solely leads to expression of EGFP in medulla neurons. The positions of R8/R7 terminals, visualized with mab24b10 (magenta) in layers M3 and M6 relative to the branches of MzVum-Gal4-positive neurons provide landmarks to identify layers in the distal medulla (me). MzVum-Gal4-positive neurons extensively innervate the lower part of layer M2 and layers M4, as well as M6-M10, but form few branches in layers M1, upper M2, M3 and M5 (arrowheads). In this background, individual neuron subtypes cannot be discerned. la, lamina; lo, lobula; lop, lobula plate. Scale bar, 50 μm.

6 Supplementary Figure 5 Schematic representation of cloning strategy used to generate Flybow constructs. (a) Cloning of constructs relied on building four basic modules. Each module contained a mfrt71 sequence, a membrane-anchor, one of four FP-encoding sequences and one of two polyadenylation signals. Cerulean was also tagged with a V5-encoding sequence. (b) The four modules were subsequently subcloned into the UAS-vector pkc26 to assemble the final constructs in the order indicated. A detailed description is provided in the Online Methods.

7 Supplementary Table 1 Quantification of recombination efficiency of Flp recombinases. % recombination efficiency a Flp mflp4 b K82Y, V226A mflp5 b R281Y mflp6 b A36T, T50A, K82Y, G109N, E150G, R281V mflp7 b A36T, T50A, K82Y, G109N, T176A, R281V FRT ± ± ± mfrt mfrt ± ± 6.7 mfrt ± ± ± 9.2 a Recombination efficiency of Flp recombinases towards the different FRT sites is indicated as the percentage of eye imaginal disc areas positive for -Galactosidase (see Supplementary Fig. 1). Values are mean percentages ± standard deviation. n = 5-8 eye discs/genotype. b Amino acid changes in modified Flp recombinases are as indicated.