Systematic analysis of the physiological importance of deubiquitinating enzymes

Abstract

Deubiquitinating enzymes (DUBs) are proteases that control the post-translational modification of proteins by ubiquitin and in turn regulate diverse cellular pathways. Despite a growing understanding of DUB biology at the structural and molecular level, little is known about the physiological importance of most DUBs. Here, we systematically identify DUBs encoded by the genome of Drosophila melanogaster and examine their physiological importance in vivo. Through domain analyses we uncovered 41 Drosophila DUBs, most of which have human orthologues. Systematic knockdown of the vast majority of DUBs throughout the fly or in specific cell types had dramatic consequences for Drosophila development, adult motility or longevity. Specific DUB subclasses proved to be particularly necessary during development, while others were important in adults. Several DUBs were indispensable in neurons or glial cells during developmental stages; knockdown of others perturbed the homeostasis of ubiquitinated proteins in adult flies, or had adverse effects on wing positioning as a result of neuronal requirements. We demonstrate the physiological significance of the DUB family of enzymes in intact animals, find that there is little functional redundancy among members of this family of proteases, and provide insight for future investigations to understand DUB biology at the molecular, cellular and organismal levels.

Figure 1. Human and Drosophila DUBs.

A) Shown are human DUBs categorized into five subclasses based on homology at the catalytic domain , . UCH: Ubiquitin C-terminal Hydrolases, USP: Ubiquitin-Specific Proteases, MJD: Machado-Joseph Disease Proteases, OTU: Otubain proteases, JAMM: JAB1/MPN/Mov34 Metalloenzymes. All DUB subclasses are cysteine proteases, except JAMM domain DUBs, which are Zn²⁺-dependent metalloproteases , . Underlined and bolded: human DUBs that aligned with Drosophila DUBs (also see File S2). Inset: histograms show the total numbers of human DUBs (black) and fly DUBs (gray) that we identified for each subclass. B) Listed are all the Drosophila DUBs that we identified. Arrowheads highlight orthologues based on coverage and domain analyses (complete analysis in File S2). Not all fly DUB genes have symbols designated to them, indicating that they have not been previously characterized. Symbols were obtained from www.flybase.org.

Figure 2. Gal4 drivers.

Flies carrying UAS-GFP were crossed to flies carrying the Gal4 driver specified in the figure. Newly-eclosed adults heterozygous for UAS-GFP and the Gal4 driver were homogenized in SDS lysis buffer and loaded into SDS/PAGE gels. Western blots show how strongly each Gal4 driver expresses UAS-GFP. Blots were probed with anti-GFP and anti-tubulin antibodies. Ubiquitous driver: sqh-Gal4. Pan-neuronal drivers: elav-Gal4.

Figure 3. List of phenotypes when individual DUBs are knocked down throughout the fly.

Listed are the strongest phenotypes associated with ubiquitous knockdown of individual fly DUBs. Diamonds highlight DUBs whose knockdown led to observable phenotypes. Experimental groups consisted of flies heterozygous for sqh-Gal4 and UAS-RNAi. Controls were heterozygous for sqh-Gal4 on the isogenic background of RNAi lines. “Dead soon after eclosion” category denotes flies that eclosed successfully from the pupal case but fell on food and died within a few hours. Late motility phenotype means it was first observed 20 or more days after eclosing from the pupal case.

Figure 4. Phenotypic distribution by developmental stage.

Histograms show the number of DUBs whose knockdown led to phenotype at each stage. A) sqh-Gal4 (ubiquitous) driver; details in Figure 3. B) strong elav-Gal4 (pan-neuronal) driver; details in Figure 5. In cases where two different RNAi lines targeting the same DUB led to phenotype in different stages, only the earlier stage was counted. Instances where knockdown of a specific DUB by one RNAi line led to defects in two stages were counted for each stage (e.g. knockdown of CG3416 led to death in both larval and pupal stages).

Figure 5. List of phenotypes when individual DUBs are knocked down pan-neuronally.

Listed are the strongest phenotypes associated with pan-neuronal knockdown of individual fly DUBs. Diamonds highlight DUBs whose knockdown led to phenotypes. Experimental groups consisted of flies heterozygous for the strong elav-Gal4 driver and UAS-RNAi. Controls were heterozygous for the strong elav-Gal4 driver on the isogenic background of RNAi lines. “Dead soon after eclosion” category denotes flies that eclosed successfully from the pupal case but fell on food and died within a few hours. Late motility phenotype means it was first observed 20 or more days after eclosing from the pupal case.

Figure 6. Phenotype when select DUBs are knocked down in glial cells.

List compares phenotypes associated with knockdown of select fly DUBs in all cells and tissues, pan-neuronally or only in glial cells.

Figure 7. Gene dosage effects with pan-neuronal knockdown.

List of phenotypes observed when individual fly DUBs were knocked down by pan-neuronal drivers with different expression strengths.

Figure 8. Differences in distribution and levels of ubiquitinated species in whole fly lysates.

Shown are western blots of whole, newly-eclosed adult flies homogenized in SDS lysis buffer and electrophoresed in SDS-PAGE gels. Experimental groups were heterozygous for sqh-Gal4 and UAS-RNAi. Controls were heterozygous for sqh-Gal4 on the isogenic background of RNAi lines. Boxes highlight some areas with visible differences in ubiquitinated species. Western blots are representative of at least three independent repeats with similar results. Underlined: ubiquitous knockdown led to phenotype (Figure 3).

Figure 9. Wing postural defects.

Shown are representative cases of flies with normal or drooping wings. Drooping wings was generally an age-dependent phenotype. Box lists DUBs whose knockdown throughout the fly or only in neurons led to this phenotype. Ubiquitous knockdown of CG1490 was lethal before adults eclosed from the pupal case.

Figure 10. Comprehensive list of fly DUBs and their physiological significance.

Listed are all the fly DUBs that we identified, highlighting previously reported functions and our current findings. Not tested: DUBs that we did not examine either because of lack of reagents, too many non-specific targets from existing RNAi lines, or because the function of these DUBs is well characterized in flies. Empty cells: as of this publication, no information had been previously reported for these DUBs.