Puberty is the period during which sexual maturation occurs and is controlled by the hypothalamic-pituitary-gonadal axis. In mammals, puberty is initiated by a sustained increase in the pulsatile release of gonadotropin releasing hormone from the hypothalamus and is set by complex interplays between genetic, metabolic, and environmental factors (Gajdos et al., 2009; Palmert and Hirschhorn, 2003; Tena-Sempere, 2013). In humans, puberty usually begins between the ages of 8 and 12 years in girls and 9 and 14 years in boys. Initiation of puberty at younger or older ages outside of these limits is regarded as precocious or delayed puberty, respectively. Precocious puberty has physical and cognitive outcomes including short stature and increased risks of obesity, type 2 diabetes, breast cancer, and cardiovascular disease (Carel et al., 2004; Golub et al., 2008), but the precise mechanisms triggering the puberty initiation remain uncertain.

Recently, mkrn3, the gene encoding makorin RING finger protein 3, was reported associated with central precocious puberty in several ethnic populations (Abreu et al., 2013; Christoforidis et al., 2017; Lee et al., 2016b; Nishioka et al., 2017; Schreiner et al., 2014). Makorin gene family members encode zinc finger proteins with a RING domain responsible for ubiquitin ligase activity and are well conserved across invertebrate and vertebrate species (Gray et al., 2000). mkrn1, 2, and 3 have been identified in vertebrates (Gray et al., 2000; 2001; Jong et al., 1999). Although the molecular functions of MKRN1 as an ubiquitin ligase of several target proteins e.g. hTERT, p53, phosphatase and tensin homolog (PTEN), anaphase-promoting complex (APC), and AMP-activated protein kinase (AMPK) have been well-studied in mammalian cells, but sparse for MKRN3 (Kim et al., 2005; Lee et al., 2009; 2015; 2018a; 2018b). Given that mkrn3 is expressed in the hypothalamus and begins to decline at the onset of puberty in mice, the inhibitory role of mkrn3 in puberty initiation has been suggested (Abreu et al., 2013). However, a causal relationship of mkrn3 in puberty regulation has not yet been elucidated.

Genetic studies of the fruit fly Drosophila melanogaster suggest that central mechanisms coordinating growth and sexual maturation are well conserved across invertebrates and vertebrates. The Drosophila life cycle, similar to that of other animals, consists of a juvenile growth phase (three larval instars; L1, L2, and L3), a sexual maturation phase called metamorphosis (pupae), and a reproductive adult stage. Notably, the transition from larval-to-pupal stages in Drosophila is primarily regulated by neuroendocrine mechanisms, similar to hypothalamic-pituitary-gonadal axis activation in mammals. Progression through each stage is controlled by surges of the steroid hormone 20-hydroxyecdysone. A single pulse of ecdysone triggers transition through L1 and L2, and three low pulses of ecdysone followed by a high level of ecdysone terminate L3 and larval growth, thus initiating metamorphosis (Rewitz et al., 2013).

Developmental and nutritional signals coordinate to fine-tune the timing and duration of ecdysone pulses. Most importantly, when larvae attain critical weight at L3, prothoracicotropic hormone (PTTH) released from the brain reaches the prothoracic gland (PG), which is a part of composite endocrine tissue called the ring gland and induces ecdysone production. The duration between critical weight attainment and pupariation is called the terminal growth period and is when most larval growth takes place, with the amount of growth during this period determining the final body size of adults. Loss of PTTH signaling prolongs larval development, resulting in larger body size (McBrayer et al., 2007). In addition, insulin/Tor signaling controls ecdysone synthesis by incorporating nutrient status (Caldwell et al., 2005; Colombani et al., 2005; Layalle et al., 2008; Mirth et al., 2005; Walkiewicz and Stern, 2009). Reduced insulin/Tor signaling specifically in the PG downregulates the ecdysone production, thereby delays the pupariation and increases animal size (Colombani et al., 2005; Layalle et al., 2008). On the other hand, increasing insulin signaling in the PG accelerates the ecdysone release and advances the metamorphosis (Caldwell et al., 2005). Ecdysone synthesis in the PG is catalyzed by a sequence of reactions mediated by enzymes encoded by the Halloween family of genes including phantom, disembodied, and shadow (Rewitz et al., 2006).

Here, we show that loss of mkrn1, an ortholog of vertebrate mkrn genes, lengthened the duration from 3^rd instar-to-pupariation in Drosophila and produced bigger size pupae. MKRN1 protein is strongly expressed in the Drosophila endocrine tissue ring gland. Moreover, phm and E74 mRNA levels were reduced in mkrn1 null larvae, indicating downregulation of ecdysone-mediated signaling. Taken together, our results demonstrate that MKRN1 controls larval developmental timing and body size by regulating steroid hormone ecdysone production. Furthermore, our study supports the notion that malfunction of makorin gene family member, mkrn3 leads to puberty timing dysregulation in mammals.

Generation of mkrn1 mutants and fly strains

To create an mkrn1-null allele, we exploited P element excision mutagenesis. A fly line harboring a P element near the mkrn1 gene, P{EPgy}mkrn1^EY14602, was mated to Dr/TM3,p{Δ2-3} flies. F1 males were then mated to MKRS/TM6B balancer females, and F2 progeny with TM6B were screened for white eye color. Approximately 100 white-eyed F2 flies were analyzed for deletion via polymerase chain reaction (PCR) analysis using genomic DNA as template. Finally, one fly line harboring a precisely excised chromosome (CTRL) and two fly lines harboring small and large deletions, named mkrn1^exS and mkrn1^exL, respectively, were mated to MKRS/TM6B balancer flies to establish a stock. The primer sequences used for genomic DNA PCR analysis were as follows: P1, forward, 5′-CCCGCCTTTTCCA TAATCGTGTCA-3′; P2, reverse, 5′-TCATTGCCGCGTCATTATT AGGAG-3′; P3, reverse, 5′-TTCATGCGCGGCTGGTCTATCG-3′; P4, forward, 5′-TACTGAGCCAAATATCGAATGCG-3′. The precision of deleted regions was confirmed by sequencing.

Other fly strains used included w¹¹¹⁸ (BL5905), UASdicer2; actin-Gal4/CyO (BL25708), and P{UAS-GFP.nls} (BL4776) obtained from the Bloomington Drosophila stock center and UAS-ptth (McBrayer et al., 2007), Tubulin-Gal4/TM6B, and phm22-Gal4 kindly provided by Michael B. O’Connor (University of Minnesota, USA), Jongkyeong Chung (Seoul University, Republic of Korea), and Seogang Hyun (Chung-Ang University, Republic of Korea), respectively. RNAi lines obtained from the Vienna Drosophila Resource Center (VDRC) were CG12477 (VDRC 102882, VDRC 31944), CG5334 (VDRC 102765), and CG5347 (VDRC 22089 and 110427). Identification of trans-heterozygotes harboring both actin-Gal4 and UAS-RNAi was performed by selecting larvae without visible phenotypic bc marker after mating UASdicer2; actin-Gal4/bc to each line of UAS-RNAi flies.

Measurement of developmental timing

Virgin and male flies were crossed and transferred every 2 h at 25°C. 2^nd instar larvae were collected 66 h after egg laying (AEL). Approximately 8–10 h later, 3^rd instar larvae were collected and transferred to vials with 20 larvae per vial. Larval stage was determined by floating the larvae in chilled 25% sucrose solution and observing spiracle and mouth hook morphology. At this time, phenotypic marker selection (either bc- or TM6B-harboring depending on the cross) was performed. Pupariation was scored at 2-h intervals in an incubator maintained at 25°C. Pupa length was measured 1 day after pupariation from images taken by a CCD camera using Optinity OptiView 3.7 software (Korea Lab Tech).

RT-PCR analysis

Total RNA was isolated from frozen larvae using QIAzol lysis reagent (QIAGEN). Total RNA (1 μg) was reverse-transcribed with oligo-dT primer using Prime Script reverse transcriptase (TAKARA), and PCR was performed using rTaq DNA polymerase (TAKARA). The following primers were used: mkrn1-forward, 5′-GCGTTCTGGATGGAGACCAA-3′; mkrn1-reverse, 5′-GTAACTTTCGGGTGCGCTTC-3′; CG32442-forward, 5′-TG CTGCTAATAATCACCGCCA-3′; CG32442-reverse, 5′-CGGTC GTGTTATTTCGCCAC-3′; ppk5-forward, 5′-CGGGAGTTGGA GTTGGTACC-3′; ppk5-reverse, 5′-AAAACTCGTCTGGTGCCC AA-3′; gapdh-forward, 5′-ACCGACTTCTTCAGCGACAC-3′; gapdh-reverse, 5′-GAGTTCGGTTACTCCAACCG-3′. ptth-forward, 5′-AAGGTTTGGACGAGATGGTG-3′; ptth-reverse, 5′-GCTGGGACTGAATCCGAATA-3′; RpL32-forward, 5′-TAA GCTGTCGCACAAATGG-3′; and RpL32-reverse, 5′-GTTCGAT CCGTAACCGATGT-3′.

Quantitative real-time PCR was performed as previously described (Lee et al., 2016a). Real-time PCR was performed using a Rotor Gene 6000 (Qiagen) with SYBR® Premix Ex Taq™ (Tli RNaseH Plus; Takara). The following quantitative reverse-transcription (qRT)-PCR primers were used: mkrn1-forward, 5′-GACGTGCGGCATCTGCTTTG-3′; mkrn1-reverse, 5′-TGTTTGGCCTGACGCCATGT-3′; CG12477-forward, 5′-TTCGTCCGTCAGGTGTTGCC-3′, CG12477-reverse, 5′-TGCC GCACTTCTTGTCCTGG-3′; CG5334-forward 5′-CAATGGGC GCCAAGGATTGC-3′; CG5334-reverse 5′-ATGCGGTTATCAC CCGCCAG-3′; CG5347-forward 5′-GCATTCGCACATGGCGT CAG-3′; CG5347-reverse, 5′-GCCCAAGCCCCCATTGAAGT-3′; ptth-forward, 5′-AAGGTTTGGACGAGATGGTG-3′; ptth-reverse, 5′-GCTGGGACTGAATCCGAATA-3′; Phm-forward, 5′-GCTTGCATTTCCGAGACGAT-3′; Phm-reverse, 5′-ACGATC ATCGAACCACCCTT-3′; E74-forward, 5′-CAAACCGAAGCT GGAGATGG-3′; and E74-reverse, 5′-TCGTCCACTTGATGAA ACGC-3′. mRNA encoding cpb20 was used as a reference gene to normalize gene expression levels using the following primer sequence: cbp20-forward, 5′-GTATAAGAAGACGCC CTGC-3′; cbp20-reverse, 5′-TTCACAAATCTCATGGCCG-3′. Data were analyzed using Rotor Gene 6000 software, and relative mRNA levels were quantified using the 2^−ΔΔCt method.

Western blot analysis

Protein extracts from 3^rd instar larvae were prepared using lysis buffer (10 mM HEPES, pH 7.5; 50 mM KCl; 10% glycerol; 5 mM Tris-HCl, pH 7.5) with freshly added 5 mM EDTA, 1 mM DTT, 0.1% Triton X-100, protease inhibitor (Sigma), 1 mM Na₃VO₄, and 0.25 mM NaF (final concentration). Protein extracts were resolved by 10% SDS-polyacrylamide gel electrophoresis (PAGE), and blots were probed by the primary antibodies anti-MKRN1 (Rb-1, 1:3,000) or anti-ERK (Cell Signaling Technology, 1:2000).

Immunostaining

At the 3^rd instar stage, brains attached with ring gland were dissected, fixed in 3.7% formaldehyde, and washed with PAXD buffer (1× PBS, 5% BSA, 0.03% sodium deoxycholate, 0.03% Triton X-100) (Gunawardhana and Hardin, 2017). Fixed larval brains were blocked in 10% horse serum in PAXD buffer (blocking solution) for 1 h. Anti-MKRN1 antibody (Rb-1, 1:1,000) was directly added and incubated overnight at 4°C. The next day, brains were washed and incubated with secondary antibody (Alexa Flour® 555 goat anti-rabbit IgG, 1:200) for 24 h at 4°C. Stained brain samples were washed with PAXD buffer, further stained with Hoechst 33342 (Sigma, 1:1,000), and mounted. Confocal images were obtained with an LSM 800 confocal microscope (Carl Zeiss) and processed with Zen software (Carl Zeiss). Representative images were selected from > 10 larval brains per genotype.

Loss of MKRN1 delayed development

In Drosophila, four members of the mkrn gene family have been annotated in the genome: mkrn1, CG12477, CG5347, and CG5334. Of these, CG12477, CG5347, and CG5334 are intronless and correspond to retrocopies of mkrn1 (Bohne et al., 2010; Gray et al., 2000). Thus, we reasoned that mkrn1 is a bona fide ortholog of vertebrate mkrn genes. To investigate the role of mkrn1 in developmental timing and maturation, an mkrn1 loss-of-function mutation was generated in flies by imprecise excision of the P element inserted in the line P{EY14602}. Two lines of deletion mutants were obtained and named mkrn1^exS and mkrn1^exL. Genomic DNA PCR and sequencing revealed the deleted regions of chromosome in two alleles (Figs. 1A and 1B). Of the two alleles, mkrn1^exS flies showed a specific disruption of mkrn1 expression without affecting nearby CG32442 and ppk5 gene expression (Fig. 1C). Thus, we used mkrn1^exS flies for further analysis, with precisely excised flies serving as control flies.

To determine whether MKRN1 regulates developmental timing, we compared pupariation timing between control and mkrn1^exS flies. We first validated our experimental setup by measuring the pupariation timing of ptth-overexpressing larvae. ptth expression was driven by tubulin-Gal4 throughout the body, with a significant increase in ptth expression confirmed by RT-PCR analysis (Fig. 2A). Given that PTTH released from the brain induces ecdysone synthesis in the PG, we expected advanced pupariation in ptth-overexpressing larvae. Consistent with a previous report (McBrayer et al., 2007) and our expectation, pupariation occurred 10–12 h earlier in ptth-overexpressing larvae than in two parental lines when measured from the 3^rd instar transition (Fig. 2B). This premature pupariation resulted in shorter pupa length compared with control larvae (Fig. 2C).

Next, we performed developmental timing analysis of control and mkrn1^exS flies by measuring the duration of time AEL until puparium formation (Fig. 2D). The average total time AEL to pupariation increased from 121 h in controls to 125 h in mkrn1^exS larvae. To understand which developmental transition phase was affected by loss of MKRN1, we measured the duration of time from 3^rd instar-to-pupariation (Fig. 2E). During this stage, widespread changes throughout larvae occur to signal the onset of maturation (Thummel, 2001). mkrn1^exS flies exhibited delayed pupariation by ~7 h compared with control flies, similar to the delay in timing AEL. This result indicates that MKRN1 mainly affects developmental timing from 3^rd instar-to-pupariation. In Drosophila, the growth of larvae defines body size (Mirth and Shingleton, 2012). Thus, we measured pupa length and observed that both male and female mkrn1^exS pupae were slightly longer than control pupae (Fig. 2F). Taken together, our findings suggest that deletion of mkrn1 lengthens the 3^rd instar-to-pupariation duration, resulting in a slight increase in body size.

Knockdown of mkrn1 paralogs did not delay development

There are three mkrn1 paralogs in the Drosophila genome: CG12477, CG5334, and CG5347. Although mkrn1 seems to be the founder gene of makorin genes (Bohne et al., 2010), we tested whether these three genes affect developmental timing in Drosophila. We measured the pupariation timing of the 3^rd instar, wherein the expression of genes was downregulated by actin-Gal4 driving UAS-RNAi for each gene. First, the extent of knockdown was verified by real-time qRT-PCR analysis. For CG12477, two RNAi lines (V102882 and V31944) showed a 60–80% reduction compared with parental control larvae (Figs. 3A and 3B). For CG5334, one RNAi line (V102765) showed a 60% reduction compared with parental control larvae (Fig. 3C). For CG5347, two RNAi lines (V110427 and V22089) showed a 40–50% reduction (Figs. 3D and 3E). Thus, we examined the pupariation timing of CG12477 and CG5334 knockdown larvae as these lines showed the significant reduction in mRNA levels. One line of CG12477 knockdown larvae (V102882) formed pupae in a slightly delayed manner compared with two parental lines considering 50% pupariation, but all three genotypes of larvae eventually formed pupae about the same time (Fig. 3F). Two other larvae expressing CG12477 and CG5334 RNAi did not show developmental delays compared with two parental control lines (Figs. 3G and 3H). Thus, we conclude that knockdown of the three mkrn1 paralogs does not exert strong effects on developmental timing. Nonetheless, the possibility that the knockdown was not sufficient to produce significant effects on developmental timing cannot be ruled out.

Ecdysone-synthesizing enzyme expression was reduced in MKRN1-null larvae

Developmental timing in Drosophila is regulated by concerted actions of the neuroendocrine system (Yamanaka et al., 2013). The PG in the ring gland is the principal organ producing the steroid hormone ecdysone, which controls all developmental transitions (Huang et al., 2008; Rewitz et al., 2006). Thus, we examined whether MKRN1 is expressed in the ring gland. To accomplish this, we raised MKRN1 antibody and confirmed its specificity by western blot analysis (Fig. 4A). In control larvae, the PG was marked by green fluorescent protein (GFP) expressed via the phantom-GAL4 driver (Fig. 4B). MKRN1 expression was evident in the cytoplasm of cells in the ring gland, including PG cells. By contrast, specific MKRN1 staining was not observed in the ring gland of mkrn1^exS 3^rd instar larvae. These results indicate that MKRN1 is expressed in endocrine tissue in Drosophila, where it may control growth and maturation.

Given that surges of ecdysone at the 3^rd instar are required for proper larval-to-pupal transition, we examined whether ecdysone signaling is perturbed in mkrn1^exS larvae. To test this possibility, we first measured mRNA levels of phantom, a Halloween family gene required for synthesis of ecdysone in the PG (Rewitz et al., 2006; Warren et al., 2004). phantom mRNA levels were significantly reduced in mkrn1^exS larvae compared with control larvae at the early (~74 h AEL) and late (~94 h AEL) 3^rd instar (Fig. 5A). Next, we measured the transcription of E74, a downstream target gene whose transcription is induced by ecdysone pulses in various target tissues (Boyd et al., 1991). Control larvae showed an ~8-fold increase in E74 transcription from the early to the late 3^rd instar stage, corresponding to the steep increase of ecdysone toward the end of the larval period (Fig. 5B). Consistent with our finding that ecdysone synthesis was reduced in mkrn1^exS larvae, E74 mRNA levels were reduced at the early and late 3^rd instar stages, although the difference at the late 3^rd instar stage was not statistically significant (Fig. 5B). As PTTH is required for proper production of ecdysone at metamorphosis onset (McBrayer et al., 2007; Rybczynski and Gilbert, 2003), we tested the possibility that a reduction in PTTH would downregulate ecdysone synthesis. No significant difference in ptth expression was observed in control and mkrn1^exS 3^rd instar larvae at either early or late stages (Fig. 5C). Taken together, our findings suggest that the loss of MKRN1 reduces the expression of ecdysone-synthesizing enzyme and ecdysone downstream target gene expression in a PG cell-autonomous manner, thereby lengthening the duration from the 3^rd instar stage to pupariation.

mkrn3 mutations are associated with precocious puberty in humans (Abreu et al., 2013; Christoforidis et al., 2017; Lee et al., 2016b; Nishioka et al., 2017; Schreiner et al., 2014). Although prolonged mkrn3 expression before the onset of puberty in mice is consistent with an inhibitory role of mkrn3 in puberty initiation (Abreu et al., 2013), a causal relationship between makorin genes and sexual maturation has not been elucidated. Our present results using Drosophila as a model system demonstrate that the loss of MKRN1, an ortholog of mammalian makorin genes, lengthened the 3^rd instar larval growth period and delayed the onset of metamorphosis, resulting in larger body size. These results support the notion that makorin proteins play important roles in fine-tuning the timing of puberty in Drosophila and mammals.

Whereas deletion of mkrn1 in Drosophila delayed pupariation in our study, mkrn3 mutations in humans induce precocious sexual maturation (Abreu et al., 2013; Christoforidis et al., 2017; Lee et al., 2016b; Nishioka et al., 2017; Schreiner et al., 2014). The human mkrn3 gene was originally identified as a Prader-Willi syndrome (PWS)-associated gene. The mkrn3 gene resides in the maternally imprinted chromosome region 15q11.2-q13, which is deleted in a group of PWS patients (Jong et al., 1999; Butler, 2011). Clinical symptoms of PWS include hypogonadotropic hypogonadism with central adrenal insufficiency, general developmental delay, and obesity in children (Angulo et al., 2015). Several genes other than mkrn3 are included in the deleted chromosomal region in PWS, with each gene possibly involved in different clinical symptoms (Butler, 2011). Nonetheless, it is intriguing to note the presence of phenotypic similarities between PWS and mkrn1-null flies, which exhibit hypogonadotropic hypogonadism and a developmental delay. Although speculative at this stage, complete deletion of mkrn3 in PWS patients, similar to that in mkrn1^exS Drosophila, might produce a developmental delay, whereas site-substitutive mutations of mkrn3 in children with precocious puberty might induce premature development via an antimorphic effect (Abreu et al., 2013; Christoforidis et al., 2017; Lee et al., 2016b; Nishioka et al., 2017; Schreiner et al., 2014).

mkrn1^exS larvae exhibited delayed pupariation and prolonged developmental duration at the 3^rd instar stage. Moreover, pupa length was increased in mkrn1^exS larvae compared with control larvae, indicating that loss of MKRN1 did not change growth rate but prolonged larval development. What might have caused this pupariation delay in mkrn1^exS larvae? Similar to that juvenile growth concludes with a surge of steroid hormone production in vertebrates, 3^rd instar-to-pupa metamorphosis is triggered by a surge of the steroid hormone ecdysone in Drosophila. Thus, we reasoned that delayed pupariation might be caused by reduced ecdysone signaling. Indeed, the induction of E74, a direct target of ecdysone signaling, was reduced in mkrn1^exS larvae as compared with control larvae at the 3^rd instar, likely due to reduced ecdysone synthesis. As ecdysone synthesis in the PG is induced by PTTH released from small neuronal groups in the brain, we checked whether PTTH levels were reduced in mkrn1^exS larvae. There was no difference in PTTH mRNA levels between control and mkrn1^exS larvae, suggesting that loss of MKRN1 function does not affect PTTH release. During 3^rd instar-to-pupa transition, the other crucial factor affecting development is nutrient signaling. Indeed, PTTH is released when larvae reach critical weight, the attainment of which is governed by larval nutrition. Nutrient status is conveyed via insulin/Tor signaling, and mutations that downregulate insulin/Tor signaling throughout the organism delay larval development and produce small adult flies, indicating that both growth rate and developmental timing is affected (Bohni et al., 1999; Oldham et al., 2000; Shingleton et al., 2005). Intriguingly, downregulation of insulin/Tor signaling in a PG-specific manner before the critical weight checkpoint delays metamorphosis and increases body size due mainly to a lengthening of the 3^rd instar duration (Colombani et al., 2005; Layalle et al., 2008). Moreover, reduced Tor signaling in PG cells significantly delays transcriptional induction of ecdysone-synthesizing enzymes (Layalle et al., 2008). We are intrigued that these phenotypes are remarkably similar to those observed in mkrn1^exS larvae. Thus, we proposed the model that in the mkrn1^exS 3^rd instar stage, insulin/Tor signaling is compromised in PG tissue autonomously before the critical weight check-point, resulting in insufficient ecdysone synthesis and prolonged 3^rd instar duration (Fig. 5D, See below). Immunostaining with our newly generated MKRN1 antibody revealed expression of MKRN1 in the cytoplasm of ring gland composite cells including PG, corpus allatum, and corpus cardiacum. This result supports the notion that MKRN1 functions in a PG cell-autonomous manner to control ecdysone synthesis.

makorin gene family members encode zinc finger proteins with a RING domain that have ubiquitin ligase activity and are well conserved across invertebrate and vertebrate species (Gray et al., 2000). Three functional genes, mkrn1, 2, and 3, are identified in mammals, with the molecular functions of MKRN1 being extensively studied in mammals. Several substrates targeted by MKRN1 as an ubiquitin ligase have been identified including hTERT, p53, PTEN, APC, and AMPK. Downregulation of MKRN1 results in accumulation of p53, PTEN, APC, and AMPK in mammalian cancer cell lines or tissues and affects tumorigenesis (Kim et al., 2005; Lee et al., 2009; 2015; 2018a) and cellular energy metabolism (Lee et al., 2018b). Among substrates targeted by MKRN1, PTEN and AMPK are notable negative regulator of insulin/TOR signaling (Nakashima et al., 2000; Shaw et al., 2004). Although the direct target of MKRN1 is not known in Drosophila, it is possible that MKRN1 might function as ubiquitin ligase of PTEN and/or AMPK as in mammals, thereby loss of MKRN1 reduces insulin/Tor activity in mkrn1^exS larvae ultimately downregulating ecdysone synthesis (Fig. 5D). We cannot rule out the possibility of other substrates of MKRN1 that might function in a PG cell-specific manner. Indeed, a previous study reports that MKRN1 functions as a ribonucleoprotein in embryonic stem cells (Cassar et al., 2015), with MKRN1 being associated with mRNAs and RNA-binding proteins and controlling mRNA metabolism independently of its ubiquitin ligase activity. Further studies are needed to identify MKRN1 downstream targets in the PG. In addition, it would be interesting to test whether MKRN3 regulates developmental timing in mammals by affecting hypothalamic cell nutrient signaling.