A genome-wide resource for the analysis of protein localisation inDrosophila

Abstract

30The Drosophila genome contains >13,000 protein coding genes, the majority of 31 which remain poorly investigated. Important reasons include the lack of 32 antibodies or reporter constructs to visualise these proteins. Here we present a 33 genome-wide fosmid library of ≈10,000 GFP-tagged clones, comprising tagged 34 genes and most of their regulatory information. For 880 tagged proteins we have 35 created transgenic lines and for a total of 207 lines we have assessed protein 36 expression and localisation in ovaries, embryos, pupae or adults by stainings and 37 live imaging approaches. Importantly, we can visualise many proteins at 38 endogenous expression levels and find a large fraction of them localising to 39 subcellular compartments. Using complementation tests we demonstrate that 40 two-thirds of the tagged proteins are fully functional. Moreover, our clones also 41 enable interaction proteomics from developing pupae and adult flies. Taken 42 together, this resource will enable systematic analysis of protein expression and 43 localisation in various cellular and developmental contexts. 44 Impact statement 45We provide a large-scale transgenic resource, which enables live imaging, subcellular 46 localisation and interaction proteomics of selected gene products at all stages of 47 Results 113Our goal was to generate a comprehensive resource that allows the investigation of 114 protein localisation and physical interactions for any fly protein of interest through a 115 robust, generic tagging pipeline in bacteria, which is followed by a large-scale 116 transgenesis approach (Figure 1). We based our strategy on a Drosophila 117 melanogaster FlyFos library of genomic fosmid clones, with an average size of 36 kb, 118 which covers most Drosophila genes (Ejsmont et al., 2009). Our two-step tagging 119 strategy first inserts a generic 'pre-tag' at the C-terminus of the protein, which is then 120 replaced by any tag of choice at the second tagging step, for example with a 121 superfolder-GFP (sGFP) tag to generate the sGFP TransgeneOme clone library. These 122 tagged clones are injected into fly embryos to generate transgenic fly-TransgeneOme 123 (fTRG) lines, which can be used for multiple in vivo applications. (Figure 1). 124 125 sGFP TransgeneOme -a genome-wide tagged FlyFos clone library 126We aimed to tag all protein coding genes at the C-terminus of the protein, because a 127 large number of regulatory elements reside within or overlap with the start of genes, 128 including alternative promoters, enhancer elements, nucleosome positioning 129 sequences, etc. These are more likely to be affected by a tag insertion directly after 130 the start codon. Signal sequences would also be compromised by an inserted tag after 131 the start codon. Additionally, the C-termini in the gene models are generally better 132 supported by experimental data than the N-termini due to an historical bias for 3' end 133 sequencing of ESTs. Thus, C-terminal tagging is more likely to result in a functional 134 tagged protein than N-termi...

Inter-adipocyte Adhesion and Signaling by Collagen IV Intercellular Concentrations in Drosophila

“…Insertion sites of these traps are: between coding exons 1 and 2 for if, exons 1 and 2 for Ilk, exons 10 and 11 for rhea, exons 2 and 3 for Sdc, and at intronic position X 2,499,173 in domain II for trol (many variables exons exist in this gene). In vivo protein-fusion fluorescent reporters LanB1 fTRG.681 .sGFP and Ndg fTRG.638 .sGFP are not protein traps generated from the endogenous loci, but from transgenic fosmids containing an extra copy of the gene to which a C-terminal sGFP exon has been added [50]. LanB1.sGFP and Ndg.sGFP are functional on account of their ability to rescue LanB1 [50] and Ndg mutants (not shown).…”

“…In vivo protein-fusion fluorescent reporters LanB1 fTRG.681 .sGFP and Ndg fTRG.638 .sGFP are not protein traps generated from the endogenous loci, but from transgenic fosmids containing an extra copy of the gene to which a C-terminal sGFP exon has been added [50]. LanB1.sGFP and Ndg.sGFP are functional on account of their ability to rescue LanB1 [50] and Ndg mutants (not shown). mys.sGFP fTRG.932 (Integrin b) fosmid reporter was unable to rescue lethality of mys mutants (mys M2 and mys G4 ) and therefore was not used in this study.…”