The Drosophila ubiquitin-specific protease Puffyeye regulates dMyc-mediated growth

Abstract

The essential and highly conserved role of Myc in organismal growth and development is dependent on the control of Myc protein abundance. It is now well established that Myc levels are in part regulated by ubiquitin-dependent proteasomal degradation. Using a genetic screen for modifiers of Drosophila Myc (dMyc)-induced growth, we identified and characterized a ubiquitin-specific protease (USP), Puffyeye (Puf), as a novel regulator of dMyc levels and function in vivo. We show that puf genetically and physically interacts with dMyc and the ubiquitin ligase archipelago (ago) to modulate a dMyc-dependent cell growth phenotype, and that varying Puf levels in both the eye and wing phenocopies the effects of altered dMyc abundance. Puf containing point mutations within its USP enzymatic domain failed to alter dMyc levels and displayed no detectable phenotype, indicating the importance of deubiquitylating activity for Puf function. We find that dMyc induces Ago, indicating that dMyc triggers a negative-feedback pathway that is modulated by Puf. In addition to its effects on dMyc, Puf regulates both Ago and its cell cycle substrate Cyclin E. Therefore, Puf influences cell growth by controlling the stability of key regulatory proteins.

Keywords: Deubiquitinase; Drosophila; Growth; Myc; Protein degradation.

Fig. 1.

puffyeye (CG5794) is a novel regulator of the dMyc-dependent rough eye phenotype. (A) The puf locus. Two of the five predicted transcript isoforms (puf-RD and -RE) are indicated (FlyBase R5.48). Coding exons are indicated as black boxes. The puf^A459 insertion is situated downstream of puf-RE. The insertion sites of EP(3)3472, EY03971 and two RNAi targeting regions are indicated. (B) Schematic of Puf^WT, Puf^WT-S, Puf^CA/HA and USP34 proteins. Dark-grey box indicates the USP catalytic domain. Overall percentage of amino acid sequence identity and the USP domain between Puf and human USP34 are shown. Light-grey bars indicate regions used for antisera production. Black dots indicate the wild-type conserved cysteine and histidine boxes, and the white dots indicate the mutated catalytic sites. (C-N′) Scanning electron micrographs (SEMs) of the lateral view of adult compound eyes. (C-J′) SEM images show that Puf modifies GMM eye phenotype. (K-N′) SEM images show that Puf and Ago genetically interact with each other. ago¹ and ago⁴ represent different mutant alleles. Original SEM magnification: 160× for for C-N; 750× for C′-N′.

Fig. 2.

Puf promotes growth in the eye in a dMyc-dependent manner. (A-C′) Lateral view of scanning electron micrographs (SEMs) show eye phenotypes caused by dMyc or Puf activation. (D-G′) SEM shows dMyc and Puf genetically interact to modify eye phenotype. (H,H′,K-N′) SEM shows eye phenotype of UAS-puf transgenes. (I-J′) Transmission electron micrographs (TEM) show ommatidia of adult compound eyes. (O-Q′) SEM shows modification of GMM eye phenotype by UAS-puf transgenes. Original SEM magnification: 160× for A-H,K-Q; 750× for A′-H′,K′-Q′. Original TEM magnification: 600× for I,J; 1000× for I′,J′.

Fig. 3.

Puf is necessary for growth in the wing. (A-H) Wing phenotypes caused by BxΔK-Gal4-driven transgene expression in the dorsal compartment of the wing. (I-P) Wing phenotypes caused by en-Gal4-induced transgene expression in the posterior compartment of the wing. Dashed line in I indicates the anterior-posterior boundary. Arrows indicate vein phenotype. (Q) Quantification of the areas of anterior and posterior compartments of the adult wing (left panel) and the ratio of posterior:anterior compartments (right panel). Anterior (red) and posterior (blue) areas were measured using ImageJ software (10 wings per genotype). Error bars indicate s.e.m. ^*P<0.001, Student’s t-test.

Fig. 4.

Puf regulates dMyc levels in a deubiquitylating activity-dependent manner. (A,B) Western blots of protein lysates from 3rd instar larval wing discs. (A) w¹¹¹⁸ wild-type control (lanes 1, 2) and puf^A459 homozygous mutant (lanes 3, 4). Each lane used protein lysates from 20 wing discs. γ-Tubulin served as loading control. (B) Effect of Puf ^WT and catalytic mutants on endogenous dMyc levels. Each lane used protein lysates from 10 wing discs. (C-H′) Immunostaining of 3rd instar larval wing disc showing effect of Puf on dMyc levels. (I,J) Effect of Puf ^WT and catalytic mutants on overexpressed dMyc protein levels. Each lane used protein lysates from 10 wing discs after a 20-hour induction.

Fig. 5.

Post-translational regulation of dMyc abundance by Puf. (A) Endogenous dMyc levels at increasing times following treatment with cycloheximide (CHX chase) to determine the stability of dMyc in control wing disc cells (lanes 1-4) or in cells overexpressing Puf^WT (lanes 5-8) or Puf^CA/HA (lanes 9-11). Protein lysates were isolated from 12 wing discs at the indicated time points after CHX treatment. (B) The quantification of dMyc levels after the addition of CHX. The amount of dMyc remaining was obtained by normalizing signals generated by mouse monoclonal dMyc antibody to that of γ-tubulin using ImageJ software. Time zero was set at 100%. The data are plotted as percentages of dMyc remaining over time zero for each point. (C) Stability of overexpressed dMyc measured in CHX-chase experiments. Protein lysates from wing discs overexpressing dMyc (lanes 1-4), dMyc+ Puf^WT (lanes 5-8) or dMyc+Puf^CA/HA (lanes 9-11). (D) qRT-PCR determination of relative levels of dMyc mRNA following overexpression of Puf (tub-Gal4, UAS-puf^WT). Relative expression levels (ΔΔCT) were calculated using RpS16 as internal control. Data represent mean of three biological samples analyzed in duplicate. Error bars indicate s.e.m. Transgenes were induced for 20 hours using temperature-inducible Gal4 drivers.

Fig. 6.

Puf is a nuclear protein and interacts with dMyc and Ago. (A-A′) Immunostaining of the 3rd instar larval wing disc of w¹¹¹⁸ stained with anti-Puf (red, A) and anti-lamin C (A′, green) showing nuclear localization of endogenous Puf. (A′) Merge of A and A′. (B-B′) Immunostaining of the 3rd instar larval wing disc expressing UAS-dMyc, UAS-puf^WT stained with anti-dMyc (red, B′) and anti-Puf (B, green) showing colocalization of Puf and dMyc in the nucleus. (B′) Merge of B and B′. (C) Co-immunoprecipitation showing that both Puf^WT (left panel) and Puf^CA/HA (right panel) form a protein complex with dMyc. (D) Co-immunoprecipitation showing endogenous dMyc forms a complex with Puf^WT or Puf^CA/HA. (E) Co-immunoprecipitation showing Puf forms protein complexes with Ago or dMyc. All co-immunoprecipitations were carried out using protein lysates isolated from 3rd instar larval wing discs of indicated genotypes, immunoprecipitated with pre-immune or anti-Puf antisera, and analyzed by western blot using appropriate antisera. Transgenes were induced for 20 hours using temperature-inducible Gal4 drivers. (F) Western blot showing specificity of anti-Ago and effect of Puf on endogenous Ago (lanes 1-5; 30 discs/lane) and overexpressed Ago (lanes 6-8; 10 discs/lane).

Fig. 7.

Post-translational regulation of Ago and CycE levels by Puf. (A-G′) Immunostaining of the 3rd instar larval wing disc showing that Puf and dMyc regulate Ago levels. All transgene expression was marked by GFP, induction time is indicated. (H) Western blot using anti-CycE showing effect of Puf on CycE levels. (I) CHX chase experiments for overexpressed CycE alone (lanes 1-3), CycE+ Puf^WT (lanes 4-6) or CycE+ Puf^CA (lanes 7-9). Fifteen discs per time point. (J) The complex regulatory loop among Puf, Myc and Ago.