Ohgata, the Single Drosophila Ortholog of Human Cereblon, Regulates Insulin Signaling-dependent Organismic Growth

Abstract

Cereblon (CRBN) is a substrate receptor of the E3 ubiquitin ligase complex that is highly conserved in animals and plants. CRBN proteins have been implicated in various biological processes such as development, metabolism, learning, and memory formation, and their impairment has been linked to autosomal recessive non-syndromic intellectual disability and cancer. Furthermore, human CRBN was identified as the primary target of thalidomide teratogenicity. Data on functional analysis of CRBN family members in vivo, however, are still scarce. Here we identify Ohgata (OHGT), the Drosophila ortholog of CRBN, as a regulator of insulin signaling-mediated growth. Using ohgt mutants that we generated by targeted mutagenesis, we show that its loss results in increased body weight and organ size without changes of the body proportions. We demonstrate that ohgt knockdown in the fat body, an organ analogous to mammalian liver and adipose tissue, phenocopies the growth phenotypes. We further show that overgrowth is due to an elevation of insulin signaling in ohgt mutants and to the down-regulation of inhibitory cofactors of circulating Drosophila insulin-like peptides (DILPs), named acid-labile subunit and imaginal morphogenesis protein-late 2. The two inhibitory proteins were previously shown to be components of a heterotrimeric complex with growth-promoting DILP2 and DILP5. Our study reveals OHGT as a novel regulator of insulin-dependent organismic growth in Drosophila.

Keywords: CRISPR/Cas; Cereblon; Drosophila genetics; E3 ubiquitin ligase; development; insulin.

FIGURE 1.

CRBN is conserved across the animal kingdom. A, phylogenetic tree of CRBN homologs. Depicted values indicate posterior probabilities. The scale bar references branch lengths. B, schemes representing the domain structure of human CRBN and its murine and Drosophila homologs (the structure of human CRBN is based on a previous study (64) and structures of murine and Drosophila homologs are based on those described in the Universal Protein Resource database (entries Q8C7D2 and Q9VH36, respectively). The two evolutionary conserved domains, LON domain and TBD, are indicated as filled boxes (blue and gray, respectively). Percentages of identity at the amino acid level are noted. Two immunogens used to generate OHGT-specific antibodies (rabbit anti-OHGT(1–187) and guinea pig anti-OHGT(15–33)) are represented. KLH, keyhole limpet hemocyanin.

FIGURE 2.

Generating an ohgt mutant. A, a scheme representing the ohgt locus and the target site. A 20-bp-long sequence within exon 2 was chosen for targeting by the CRISPR-Cas9 system. The predicted cleavage site is indicated by a vertical bar. Two nucleotides deleted by Cas9 activity are indicated by horizontal bars (-). The protospacer adjacent motif (PAM) sequence is shown in green. The putative stop codon after the frameshift is shown in red. (B) A scheme to recover flies with ohgt^Ex2 allele. Microinjection survivors were crossed with TM3-GFP balancer flies and subsequently sacrificed to perform the HRMA-based mutagenesis screening. The F₁ progenies from the cross were again crossed to the same strain, and further HRMA was performed to confirm germ line transmission of the mutation. The amplicon was inserted into pCRII-TOPO vector and subjected to molecular characterization. C, HRMA of an F₁ mutant candidate. Melt curves of ohgt^Ex2/TM3-GFP (orange) and TM3-GFP balancer (gray) can be easily distinguished due to heteroduplex formation in candidate-derived sample. D, change in relative fluorescence units (RFU) relative to TM3-GFP balancer emphasizes the change of melt curve shown in C. E, total lysates were prepared from control (w*), homozygous ohgt^Ex2 (Ex2/Ex2), and transheterozygous ohgt^Ex2/ohgt^Df (Ex2/Df) third instar larvae. Proteins from each lysate were subjected to Western blotting with antibodies to the indicated proteins.

FIGURE 3.

The ohgt^Ex2 mutants show overgrowth phenotype. A, percentage of w* and ohgt^Ex2 animals pupated at each time point after egg deposition. B, w*, ohgt^Ex2/Ex2, and ohgt^Ex2/ohgt^Df (ohgt^Ex2/Df) adult male flies. The scale bar represents 1 mm. C, the ohgt^Ex2 larvae showed food intake similar to that of w* larvae (n = 4). D, average wet weight of w*, ohgt^Ex2/Ex2, and ohgt^Ex2/Df adult males. Animals were weighed in batches of 10, and average weight per animal was calculated (n = 10). E, average wing area of w*, ohgt^Ex2/Ex2, and ohgt^Ex2/Df adult males (n = 30). F, average number of cells in the posterior compartment of the wing of w*, ohgt^Ex2/Ex2, and ohgt^Ex2/Df adult males (n = 30). G, w*, ohgt^Ex2/Ex2, and ohgt^Ex2/Df pupae. An animal whose body length is closest to the average was chosen from each genotype. The scale bar represents 1 mm. H, average body length of w* (n = 66), ohgt^Ex2/Ex2 (n = 80), and ohgt^Ex2/Df (n = 73) pupae. Error bars indicate S.E. * indicates p < 0.05, and ** indicates p < 0.01. p values were calculated by Student's t test. n.s. indicates no significance.

FIGURE 4.

OHGT is expressed in the nucleus of fat body cells. A and B, co-immunostaining of OHGT and GFP in the IPCs of dilp2>GFP third instar larva. C and D, co-immunostaining of OHGT and plasma membrane marker α-spectrin in the fat body of w* early third instar larva (72 h AED). E and F, co-immunostaining of OHGT and α-spectrin in the fat body of ohgt^Ex2/Df early third instar larva (72h AED). G and H, co-immunostaining of OHGT and GFP after clonal ohgt RNAi induction in fat body cells. Genotype, y,hs-flp/+; Dcr-2/+; act-FRT-CD2-FRT-Gal4, UAS-GFP/UAS-ohgt^RNAiv40486. The scale bars in images represent 50 μm. I–L, normalized staining intensities in whole cells, the cytoplasm, and the nucleus as well as the nuclear to cytoplasm ratio of wild type clones (n = 34) and ohgt RNAi induced clones (n = 17). Error bars indicate S.E. * indicates p < 0.05, and ** indicates p < 0.01. p values were calculated by Mann-Whitney U test.

FIGURE 5.

Fat body-specific ohgt knockdown phenocopies the ohgt mutant phenotype. A, transcript levels of ohgt in the fat body and carcass of Cg>w* and Cg>ohgt^RNAiv40486 third instar larvae were quantified using qRT-PCR (n = 4). B, Cg>w* and Cg>ohgt^RNAiv40486 pupae. An animal whose length is closest to the average was chosen from each genotype. The scale bar represents 1 mm. C, average body length of Cg>w* (n = 48) versus Cg>ohgt^RNAiv40486 (n = 66) pupae, Cg>w* (n = 40) versus Cg>ohgt^RNAiNIG3 (n = 48) pupae, and FB>w* (n = 48) versus FB>ohgt^RNAiv40486 (n = 52) pupae. D, Cg>w* and Cg>ohgt^RNAiv40486 adult male flies. The scale bar represents 1 mm. E, average wet weight of Cg>w* and Cg>ohgt^RNAiv40486 adult male flies. Animals were weighed in batches of 10, and average weight per animal was calculated (n = 8). F, average wing area of Cg>w* and Cg>ohgt^RNAiv40486 adult males (n = 20). Error bars indicate S.E. * indicates p < 0.05, and ** indicates p < 0.01. Each p value was calculated by Student's t test. n.s. indicates no significance.

FIGURE 6.

The ohgt mutation and ohgt knockdown in the fat body are associated with insulin signaling up-regulation and altered expression of fat body-derived cofactors of DILPs. A, the expression level of phospho-dAkt (p-dAkt) was increased in ohgt^Ex2/Ex2 compared with w*. The experiment was performed in triplicate. B, transcription of the genes d4EBP, dInR, and Lip3 (included as a starvation marker) in ohgt^Ex2/Ex2 and w* (n = 4). C, the expression level of phospho-dAkt was increased in Cg>ohgt^RNAiv40486 compared with Cg>w* early third instar larval carcasses. The experiment was performed in triplicate. D, transcription of d4EBP, dInR, and Lip3 in Cg>ohgt^RNAiv40486 and Cg>w* in early third instar larval carcasses (n = 4). E, transcription of the genes dALS, Imp-L2, and upd2 in the total ohgt^Ex2/Ex2 and w* early third instar larvae was quantified (n = 4). F, transcription of dALS, Imp-L2, and upd2 in the fat bodies of Cg>ohgt^RNAiv40486 and Cg>w* early third instar larvae was quantified (n = 4). Error bars indicate S.E.

FIGURE 7.

OHGT interacts with PIC, the Drosophila homolog of human DDB1. A, 3D model for OHGT-PIC complex was created using the SWISS-MODEL service and then aligned to respective positions in Protein Data Bank code 4TZ4. Amino acid residues and domains depicted in the structures are shown. B, the crystal structure of the human CRBN-DDB1 complex (Protein Data Bank code 4TZ4). C and D, a close-up view of the OHGT-PIC/CRBN-DDB1 interaction interfaces. C′ and D′, amino acid side chains of OHGT/CRBN that point toward the PIC/DDB1 interface are shown in red. E, S2 cells were transfected with HA-OHGT expression vector, and protein complexes containing HA-OHGT were immunoprecipitated using HA tag-specific antibody. Total lysates (Input) and immunoprecipitates (IP) were subjected to Western blotting with antibodies to the indicated proteins (for detecting OHGT, guinea pig OHGT-specific antibody was used). F, protein complexes containing endogenous OHGT were immunoprecipitated using rabbit OHGT-specific antibody. Total lysates (Input) and immunoprecipitates (IP) were subjected to Western blotting with antibodies to the indicated proteins (for detecting OHGT, guinea pig OHGT-specific antibody was used). G, S2 cells were treated with MG132, and endogenous OHGT was immunoprecipitated using rabbit OHGT-specific antibody. Total lysates (Input) and immunoprecipitates (IP) were subjected to Western blotting with antibodies to the indicated proteins (for detecting OHGT, guinea pig OHGT-specific antibody was used).