The polysaccharide glycogen is an evolutionarily conserved storage form of glucose. However, the physiological significance of glycogen metabolism on homeostatic control throughout the animal life cycle remains incomplete. Here, we describe Drosophila mutants that have defective glycogen metabolism. Null mutants of glycogen synthase (GlyS) and glycogen phosphorylase (GlyP) displayed growth defects and larval lethality, indicating that glycogen plays a crucial role in larval development. Unexpectedly, however, a certain population of larvae developed into adults with normal morphology. Semi-lethality in glycogen mutants during the larval period can be attributed to the presence of circulating sugar trehalose. Homozygous glycogen mutants produced offspring, indicating that glycogen stored in oocytes is dispensable for embryogenesis. GlyS and GlyP mutants showed distinct metabolic defects in the levels of circulating sugars and triglycerides in a life stage-specific manner. In adults, glycogen as an energy reserve is not crucial for physical fitness and lifespan under nourished conditions, but glycogen becomes important under energy stress conditions. This study provides a fundamental understanding of the stage-specific requirements for glycogen metabolism in the fruit fly.

Glucose serves as a major energy source and also donates its carbon to most other synthesized molecules, such as the amino acids, nucleotides and fatty acids. Surplus glucose derived from dietary carbohydrates is stored as branched polysaccharide glycogen or triglycerides (TAGs) in the body for future energy needs (Saltiel and Kahn, 2001; Chng et al., 2017; Mattila and Hietakangas, 2017). Proper regulation of anabolism and catabolism of stored energy reserves is crucial to sustain metabolic homeostasis throughout the animal life cycle.

The fruit fly Drosophila has two-programmed starvation periods, embryo and pupa, during its life cycle. Glycogen is stored at the late stage of oogenesis by metabolic remodeling in mitochondria (Sieber et al., 2016). Metabolic analysis has revealed that glycogen stored in oocytes is consumed during embryogenesis (Tennessen et al., 2014; Matsuda et al., 2015). Consistently, the onset of aerobic glycolysis occurs in late-stage embryos through transcriptional induction (Tennessen et al., 2011). Likewise, glycogen stored in feeding larvae gradually decreases during metamorphosis (Gáliková et al., 2015; Matsuda et al., 2015). These observations suggest that developing embryos and pupae are using glycogen to produce ATP energy as well as to generate biomolecules needed for cellular proliferation and differentiation to sustain embryonic development and metamorphosis. However, the vital role of glycogen during these developmental periods has not been directly tested.

Glycogen metabolism is governed by evolutionarily conserved glycogen synthase and phosphorylase by the concerted action of branching and de-branching enzymes (Roach et al., 2012). In mammals, glycogen is primarily stored in the cells of the liver and muscles. Liver glycogen plays an important role in the glucose cycle to maintain circulating sugar levels, whereas muscle glycogen is directly utilized through glycolysis to maintain muscle activities (Roach et al., 2012; Petersen et al., 2017). In Drosophila larvae, we have shown previously that the tissue-specific regulation of glycogen metabolism in the fat body, an organ equivalent to the mammalian liver, plays a crucial role in the maintenance of circulating sugars under fasting conditions (Yamada et al., 2018). In adults, glycogen is stored in flight muscles and consumed during flight (Wigglesworth, 1949). Interestingly, emerging evidence suggest that genetic manipulation of glycogen metabolism extends lifespan in several species, including Caenorhabditis elegans and Drosophila (Bai et al., 2013; Gusarov et al., 2017; Post et al., 2018). On the other hand, the progressive accumulation of glycogen in neurons leads to neuronal cell death, locomotion deficits and reduced lifespan, thereby contributing to physiological aging (Duran et al., 2012). Although the requirements for glycogen metabolism likely change over developmental time in a tissue-specific manner, it remains unclear how animals respond to loss of a major energy reserve and maintain body homeostasis through compensatory metabolic mechanisms.

Here, we generated and characterized Drosophila defective for glycogen metabolism. We observed that more than half of the mutants are lethal in larvae. However, to our surprise, a certain population of larvae undergoes pupariation and enters adulthood. Importantly, we did not observe an essential contribution of glycogen during embryogenesis and metamorphosis. Glycogen synthase (GlyS) mutant adults showed reduced physical fitness, but this could conceivably be an indirect consequence of metabolic defects, including a reduction in trehalose. Glycogen phosphorylase (GlyP) mutants displayed normal flight performance, climbing ability and adult lifespan, comparable to those of control flies, suggesting that glycogen as a fuel source is largely dispensable for adult fitness. In contrast, the importance of glycogen metabolism becomes apparent under fasting conditions. This study provides the first direct evidence that glycogen metabolism plays an important role in larval development and adult fitness but not in embryogenesis and metamorphosis in Drosophila.

Generation and validation of GlyS and GlyP null alleles

In Drosophila, a set of single orthologous genes are involved in glycogen metabolism (Fig. 1A); namely, GlyS, 1,4-α-glucan branching enzyme (AGBE), GlyP and the de-branching enzyme amylo-α-1,6-glucosidase, 4-α-glucanotransferase (AGL; FlyBase annotation symbol CG9485). To examine the requirements of glycogen metabolism over the life cycle, we generated null mutants of GlyS and GlyP (Fig. 1B,C). GlyS8 mutants, carrying a small deletion at the N-terminal coding region, were created using the CRISPR/Cas9 system. GlyP3-13 mutants, in which mutants lack the entire first exon, were made by imprecise excision of a P-element. Unexpectedly, we found that homozygous mutants of GlyS8 and GlyP3-13 were viable at a significantly lower ratio than heterozygotes (see below). Homozygous adults of GlyS8 and GlyP3-13 displayed normal morphology (Fig. 1D).

Fig. 1.

Generation and validation of GlyS and GlyP null alleles. (A) Overview of glucose and storage sugar metabolism. Genes that function in glycogen and trehalose metabolism are shown in red. (B) Schematic of the GlyS and GlyP loci and molecular nature of the mutants. Protein-coding regions and untranslated regions are represented by black boxes and white boxes, respectively. The P-element insertion sites are marked with inverted triangles. (C) Sequences of the sgRNA target site and the GlyS8 deletion mutant. The 20-bp target sequence is indicated in blue, and the cleavage site of Cas9 in indicated by the inverted triangle. (D) GlyS and GlyP null mutants were viable without obvious morphological defects. (E) Total amounts of glycogen were analyzed in the indicated genotypes and are plotted relative to that of control. n=9-12 (late third instar: wandering larvae) or 9 (adult) batches from 3-4 independent experiments. (F) Expression levels of GlyS and GlyP were analyzed by qRT-PCR. n=4 batches. (G) GlyS and GlyP enzyme activities were assessed in tissue homogenates at early third instar. The average values from two independent experiments are shown. A.U., arbitrary unit; Glucose-P, glucose phosphate. (H) Expression of GlyS and GlyP in each mutant background restored glycogenesis and glycogenolysis, respectively, as visualized by PAS staining. Tub-Gal4 was used for ubiquitous expression. Scale bars: 100 μm. (I) Expression of GlyS and GlyP in each mutant background fully restored glycogen levels in late third instar larvae. n=5-9 batches from two independent experiments. One-way ANOVA with Dunnett's post-hoc test (E,F,I); ***P<0.001.

Fig. 1.

Generation and validation of GlyS and GlyP null alleles. (A) Overview of glucose and storage sugar metabolism. Genes that function in glycogen and trehalose metabolism are shown in red. (B) Schematic of the GlyS and GlyP loci and molecular nature of the mutants. Protein-coding regions and untranslated regions are represented by black boxes and white boxes, respectively. The P-element insertion sites are marked with inverted triangles. (C) Sequences of the sgRNA target site and the GlyS8 deletion mutant. The 20-bp target sequence is indicated in blue, and the cleavage site of Cas9 in indicated by the inverted triangle. (D) GlyS and GlyP null mutants were viable without obvious morphological defects. (E) Total amounts of glycogen were analyzed in the indicated genotypes and are plotted relative to that of control. n=9-12 (late third instar: wandering larvae) or 9 (adult) batches from 3-4 independent experiments. (F) Expression levels of GlyS and GlyP were analyzed by qRT-PCR. n=4 batches. (G) GlyS and GlyP enzyme activities were assessed in tissue homogenates at early third instar. The average values from two independent experiments are shown. A.U., arbitrary unit; Glucose-P, glucose phosphate. (H) Expression of GlyS and GlyP in each mutant background restored glycogenesis and glycogenolysis, respectively, as visualized by PAS staining. Tub-Gal4 was used for ubiquitous expression. Scale bars: 100 μm. (I) Expression of GlyS and GlyP in each mutant background fully restored glycogen levels in late third instar larvae. n=5-9 batches from two independent experiments. One-way ANOVA with Dunnett's post-hoc test (E,F,I); ***P<0.001.

To confirm whether newly generated GlyS8 and GlyP3-13 mutants were functional null alleles, we examined glycogen levels during the wandering stage (late third-instar larvae) and in 1-week-old adult males and females. As expected, the amount of glycogen significantly decreased in GlyS mutants, but increased in GlyP mutants (Fig. 1E). GlyS mRNA levels were drastically decreased in adults (Fig. 1F), most likely due to nonsense-mediated mRNA decay that eliminates mRNAs containing a premature termination codon (Karousis et al., 2016). GlyP mRNA was almost undetectable owing to the lack of a transcriptional start site.

To demonstrate directly that GlyS8 and GlyP3-13 mutants completely abolish glycogenesis and glycogenolysis, we developed enzyme activity assays using isotope-labeled glucose, followed by mass spectrometric analyses. Glycogenesis by GlyS was assessed by measuring the incorporation rate of U-13C-glucose into glycogen from uridine diphosphate (UDP)-U-13C-glucose. As expected, no incorporation was observed in tissue homogenates from GlyS mutant larvae even though glycogen was exogenously added to the reaction mixture (Fig. 1G). Similarly, glycogenolysis by GlyP was assessed by measuring the release rate of U-13C-glucose-phosphate from 13C-labeled glycogen that was purified from adult flies fed a U-13C-glucose-containing diet (Fig. S1A,B). No release of 13C-glucose-phosphate was detected in GlyP mutant homogenates. Together, these results unambiguously indicate that GlyS8 and GlyP3-13 mutants are functional null alleles that completely eliminate the corresponding enzymatic steps in glycogen metabolism.

We previously reported that periodic acid-Schiff (PAS) staining reliably visualizes stored glycogen and also detects mobilization upon brief starvation in the fat body (Yamada et al., 2018). Consistent with our previous research (Yamada et al., 2018), GlyS mutants showed strong reductions in the PAS signal in tissues such as the CNS, the fat body and body wall muscles. This reduction was fully rescued by the ubiquitous expression of C-terminal Flag-tagged GlyS by Tub-Gal4 (Fig. 1H). GlyP mutants failed to mobilize fat body glycogen upon starvation, which was completely rescued by the expression of C-terminal Flag-tagged GlyP. The overexpression of GlyP had no effect on PAS signals under fed conditions, as described below. Moreover, these transgenes restored normal glycogen levels in whole animals (Fig. 1I). Thus, the UAS constructs we generated produced a functional protein.

GlyS and GlyP mutants display semi-lethality during the larval period

Because glycogen mutants showed a significant level of mortality prior to the adult emergence, we next examined the lethal stage and found that more than half of GlyS8 and GlyP3-13 mutants exhibited larval lethality (Fig. 2A). Consistent with results from homozygotes in each mutant allele, the transheterozygotes over a deficiency line showed similar lethality during the larval period. Importantly, the mutant lethality was fully rescued by ubiquitous expression of GlyS or GlyP (Fig. 2B), indicating that the observed larval lethality is specific to the loss of function in GlyS and GlyP genes. By contrast, after pupariation of surviving larvae, GlyS and GlyP mutants lethality was not observed during metamorphosis (Fig. 2C). We further examined the phenotypes of available transposon insertion mutants (Fig. 1B). Transheterozygotes of GlySMI01490 and GlyPMI00957 over a deficiency line exhibited semi-lethality in larvae, but not in pupae, resulting in the emergence of escaping adults (Fig. 2A,C). Based on the degrees of larval lethality, their insertion sites and the mRNA level (Fig. 1F), we can conclude that GlyPMI00957 is a null allele that is comparable to GlyP3-13, whereas GlySMI01490 is a strong hypomorphic allele, as previously reported (Yamada et al., 2018). The backcrossed GlyPk07918 (renamed GlyPlacW) is a weak hypomorphic allele. Taken together, these results indicate that GlyS and GlyP are required for normal development in larvae, but not in pupae.

Fig. 2.

GlyS and GlyP mutants display semi-lethality during the larval period. (A) GlyS and GlyP mutants exhibited lethality during the larval period. (B) Expression of GlyS or GlyP rescued larval lethality in the respective GlyS and GlyP mutants. Percentages of surviving larvae were determined by the ratio to heterozygotes in each vial. (C) Glycogen mutants did not exhibit lethality during the pupal period. (D) Glycogen mutants died at increasing rates during the third instar indicated by the gray region. (E) Larval volume in glycogen mutants was analyzed at the indicated time points. (F) Pupal volume in males and females. (G) Glycogen mutants did not exhibit developmental delay in the timing of puparium formation. (H) Glycogen mutants exhibited normal feeding behavior. Food intake levels were evaluated by the rate of blue food ingestion by early third instar larvae. (I) Adult body weight was analyzed in glycogen mutant males and females. (J) GlyP and Tps1 double mutants showed growth defects at 48 h ALH. The numbers of vials (A-D,G), animals (E,F,J) and batches (H,I) from at least two independent experiments are indicated. One-way ANOVA (A-C,E-J) or two-way ANOVA (D) with Dunnett's post-hoc test; *P<0.05, ***P<0.001; n.s., not significant.

Fig. 2.

GlyS and GlyP mutants display semi-lethality during the larval period. (A) GlyS and GlyP mutants exhibited lethality during the larval period. (B) Expression of GlyS or GlyP rescued larval lethality in the respective GlyS and GlyP mutants. Percentages of surviving larvae were determined by the ratio to heterozygotes in each vial. (C) Glycogen mutants did not exhibit lethality during the pupal period. (D) Glycogen mutants died at increasing rates during the third instar indicated by the gray region. (E) Larval volume in glycogen mutants was analyzed at the indicated time points. (F) Pupal volume in males and females. (G) Glycogen mutants did not exhibit developmental delay in the timing of puparium formation. (H) Glycogen mutants exhibited normal feeding behavior. Food intake levels were evaluated by the rate of blue food ingestion by early third instar larvae. (I) Adult body weight was analyzed in glycogen mutant males and females. (J) GlyP and Tps1 double mutants showed growth defects at 48 h ALH. The numbers of vials (A-D,G), animals (E,F,J) and batches (H,I) from at least two independent experiments are indicated. One-way ANOVA (A-C,E-J) or two-way ANOVA (D) with Dunnett's post-hoc test; *P<0.05, ***P<0.001; n.s., not significant.

To identify the timing of lethality, we counted the number of surviving larvae and found that the survival rate of both GlyS8 and GlyP3-13 mutants decreased progressively during the third instar (Fig. 2D). Molting was not related to the lethality of these mutants as evidenced by the completion of molting from the late second to third instar [100% of the GlyS8 mutants (n=63) and 100% of the GlyP3-13 mutants (n=71) from more than five independent experiments]. Body size of these mutants was normal at 24 h after larval hatching (ALH) (i.e. late first instar) and 48 h ALH (i.e. late second instar) (Fig. 2E). However, at 72 h ALH (i.e. mid-third instar), both GlyS and GlyP mutants were ∼20-34% smaller on average than a genetically matched control (w), indicating that glycogen metabolism is required for normal body growth during the third instar. However, GlyS mutants showed normal body size at the time of pupariation in both males and females. In contrast, GlyP mutants still showed smaller pupae, but the size differences averaged only ∼5-7% (Fig. 2F). Developmental timing until pupariation was not affected in GlyS and GlyP mutants that survived (Fig. 2G); thus, developmental delay cannot account for the restoration of body size in these mutants. It is of note that both GlyS and GlyP mutants consumed similar amounts of food compared with control at the early third instar stage (Fig. 2H), suggesting that feeding behavior is not affected in these mutants. Adult body weight in GlyS mutants was almost indistinguishable from that of control flies (Fig. 2I). In contrast, GlyP mutant males and females were smaller or leaner by ∼7-10% on average. The GlyP mutant-specific decreases in body weight might be partly attributed to the reduction in stored lipids, as described below.

We hypothesized that the observed semi-lethality of glycogen mutants in the larval period might be explained by the presence of circulating sugar trehalose (Fig. 1A), which also plays an important role in maintaining glucose homeostasis (Elbein et al., 2003; Chng et al., 2017; Mattila and Hietakangas, 2017). In contrast to GlyS and GlyP mutants, null alleles of the trehalose synthesis enzyme Tps1 show complete lethality during metamorphosis, but not during the larval period when dietary sugar is available (Matsuda et al., 2015; Yasugi et al., 2017). To test phenotypic interaction between glycogen and trehalose metabolism, we analyzed the lethality in larvae lacking both glycogen and trehalose metabolism. GlyP,Tps1 double mutants showed severe lethality during the second and third instar, which resulted in almost complete lethality before pupariation (Fig. 2D). Moreover, GlyP,Tps1 double mutants displayed growth defects at 48 h ALH, but each single mutant had no effect (Fig. 2E,J). These genetic interactions suggest that disrupting trehalose metabolism has no effect on larval survival, disrupting glycogen metabolism induces a semi-lethal phenotype, whereas disrupting both metabolisms likely results in a dramatic homeostatic defect.

GlyS and GlyP mutants exhibit distinct metabolic defects in a life stage-dependent manner

Glycogen metabolism is thought to play an important role in maintaining postprandial glucose homeostasis. Thus, we next analyzed the amount of glucose, trehalose and cellular lipid storage TAGs in glycogen mutants. GlyS and GlyP mutants showed distinct metabolic defects in a stage-dependent manner (Fig. 3A). In the wandering larvae, trehalose was significantly decreased in GlyP mutants but not in GlyS mutants. By contrast, no defects were observed in glucose and TAGs. Unexpectedly, at the adult stage, GlyS mutants showed significant reductions in steady-state levels of trehalose and glucose, whereas GlyP mutants showed mild, if any, reductions. By contrast, male and female GlyP mutants showed a drastic reduction (∼50%) in TAGs. Collectively, defects in glycogen metabolism in whole animals do not increase these metabolites; rather, the defects in glycogen metabolism result in a drop in either circulating sugar or lipid storage.

Fig. 3.

GlyS and GlyP mutants show distinct metabolic defects in a life stage-dependent manner. (A) Relative amounts of trehalose, glucose and TAG in glycogen mutants. n=9-12 (late third instar) or 9 (adult) batches from 3-4 independent experiments. (B) PC analysis of water-soluble metabolites in the control and glycogen mutants (late third instar). Ellipses of clusters show the 95% confidence regions for each sample group. (C) Numbers of significantly different metabolites (q<0.01 with respect to control) between GlyS and GlyP mutants. (D) Volcano plots showing q-values versus fold change (FC) from the metabolomics data. (E) Relative amounts of metabolites that significantly changed in glycogen mutants. Control (A-E): a genetically matched strain, w1118. n=6 batches (B-E); one-way ANOVA with Dunnett's post-hoc test (A,E); *P<0.05, **P<0.01, ***P<0.001; n.s., not significant.

Fig. 3.

GlyS and GlyP mutants show distinct metabolic defects in a life stage-dependent manner. (A) Relative amounts of trehalose, glucose and TAG in glycogen mutants. n=9-12 (late third instar) or 9 (adult) batches from 3-4 independent experiments. (B) PC analysis of water-soluble metabolites in the control and glycogen mutants (late third instar). Ellipses of clusters show the 95% confidence regions for each sample group. (C) Numbers of significantly different metabolites (q<0.01 with respect to control) between GlyS and GlyP mutants. (D) Volcano plots showing q-values versus fold change (FC) from the metabolomics data. (E) Relative amounts of metabolites that significantly changed in glycogen mutants. Control (A-E): a genetically matched strain, w1118. n=6 batches (B-E); one-way ANOVA with Dunnett's post-hoc test (A,E); *P<0.05, **P<0.01, ***P<0.001; n.s., not significant.

To further clarify potential metabolic alterations during the larval stage, a widely targeted metabolomics analysis was conducted using whole animals in the wandering larvae by liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS) (Table S1). The overall results of changes in 158 water-soluble metabolites were visualized using a principal component (PC) analysis. The score plot, in which each point represents an individual sample, revealed a clear separation between three genotypes (control w, GlyS mutants and GlyP mutants) (Fig. 3B). Venn diagram comparisons further revealed specific and common metabolic defects in GlyS and GlyP mutants (Fig. 3C). Interestingly, three of five common defects between GlyS and GlyP mutants were in the amino acids threonine, histidine and glutamate. A volcano plot showed significant reduction in threonine and an increase in histidine and glutamate (Fig. 3D,E). Because these three amino acids are glucogenic and used for energy production through pyruvate or 2-oxoglutarate, amino acid metabolism is likely to be altered in mutant larvae. Thiamine pyrophosphate (TPP) is an essential co-factor for several enzymes in carbohydrate metabolism, including pyruvate dehydrogenase and 2-oxoglutarate dehydrogenase, which is a rate-limiting step in the tricarboxylic acid (TCA) cycle. Reduction in TPP in both GlyS and GlyP mutants further implies that defects in glycogen metabolism potentially cause abnormalities in primary metabolic pathways.

In addition to changes in amino acids, both GlyS and GlyP mutants decreased acetyl-CoA and acetoacetyl-CoA (Fig. 3E). These decreases coincided with increases in acetylcarnitine and acylcarnitines. Although the total amount of TAGs was not changed (Fig. 3A), these results suggest active lipolysis or impaired lipid metabolism at the late larval stage. Together, these results and additional support by enrichment analysis of metabolomics data (Table S1) indicate that defects in glycogen metabolism impact amino acid and lipid metabolism.

Larval lethality in glycogen mutants is partly rescued by enriched foods

To elucidate the cause of larval lethality, we then investigated whether glycogen mutants exhibit sensitivity to malnutrition. We previously reported that Tps1 mutants are sensitive to sugar deprivation and poor dietary conditions (Matsuda et al., 2015; Yasugi et al., 2017). However, larval lethality in GlyS and GlyP mutants was not changed under a yeast-only (Y-only) diet that largely lacks carbohydrates compared with that of mutants on a normal diet (ND) (Fig. 4A). Similarly, glycogen mutants were able to reach adulthood on a poor diet (1/5Y) without apparent defects in body growth (Fig. 4A,B). Thus, unlike Tps1 mutants, undernutrition does not appear to be the major cause of larval lethality in GlyS and GlyP mutants.

Fig. 4.

Larval lethality in glycogen mutants is partly rescued by enriched foods. (A) The larval survival rate and eclosion rate of glycogen mutants under various dietary conditions. ND, normal diet; Y-only, yeast-only low-carbohydrate diet; 1/5Y, low-protein diet; 2Y, 2× yeast diet; 2G, 2× glucose diet. (B) Pupal volume of glycogen mutants grown under various dietary conditions. (C) The larval survival rate was analyzed in tissue-specific knockdown under ND. The total numbers of vials (A,C) and pupae of mixed gender (B) from at least two independent experiments are indicated in each graph. One-way ANOVA with Dunnett's post-hoc test with respect to ND in each genotype (A) or control in each diet (B) or Tukey's post-hoc test (C); *P<0.05, **P<0.01, ***P<0.001; n.s., not significant.

Fig. 4.

Larval lethality in glycogen mutants is partly rescued by enriched foods. (A) The larval survival rate and eclosion rate of glycogen mutants under various dietary conditions. ND, normal diet; Y-only, yeast-only low-carbohydrate diet; 1/5Y, low-protein diet; 2Y, 2× yeast diet; 2G, 2× glucose diet. (B) Pupal volume of glycogen mutants grown under various dietary conditions. (C) The larval survival rate was analyzed in tissue-specific knockdown under ND. The total numbers of vials (A,C) and pupae of mixed gender (B) from at least two independent experiments are indicated in each graph. One-way ANOVA with Dunnett's post-hoc test with respect to ND in each genotype (A) or control in each diet (B) or Tukey's post-hoc test (C); *P<0.05, **P<0.01, ***P<0.001; n.s., not significant.

We next tested whether enriched foods could rescue the mutant lethality. We found that larval lethality in GlyS, but not GlyP, mutants was partially rescued under high-yeast (2Y) and high-glucose (2G) diet conditions (Fig. 4A). Although GlyS mutants grew normally under these dietary conditions, mutant larvae that survived showed strong lethality during metamorphosis (Fig. 4A,B). These results suggest that enriched foods are able to overcome larval lethality in GlyS mutants, but the diet induces deleterious effect(s) during metamorphosis.

To determine which organ is crucial for the observed larval lethality in glycogen mutants, we conducted tissue-specific knockdown experiments. Similar to the mutant phenotype, ubiquitous knockdown of GlyS and GlyP by Tub-Gal4 led to semi-lethality in larvae, with a certain population becoming pupae and adults (Fig. 4C). Glycogen is mainly stored in the muscles, fat body and CNS in larvae (Yamada et al., 2018). Muscle-specific knockdown by Mef2-Gal4 induced strong lethality that varied considerably between vials and experiments (Fig. 4C). By contrast, pan-neuronal knockdown by Elav-Gal4 produced a slight reduction in the survival rate of larvae, and pan-glial knockdown by Repo-Gal4 and fat body-specific knockdown by Cg-Gal4 had modest impacts, if any, on viability (Fig. 4C). The combination of muscle- and neuron-specific knockdown fully recapitulated the lethality observed when the RNAi was induced ubiquitously (Fig. 4C). Thus, larval lethality in GlyS and GlyP mutants is caused by synergistic defects in multiple organs, including muscles and neurons.

Differences in glycogen metabolism and trehalose metabolism

We next examined the consequences of GlyS and GlyP overexpression. The GlyS and GlyP proteins overexpressed by Tub-Gal4 were distributed uniformly in the cytoplasm in the larval fat body (Fig. 5A). However, despite high-level overexpression of GlyP, there was no visible impact on the level of stored glycogen in several tissues (Fig. 5B, Fig. S2A). These results were confirmed by biochemical quantification in whole animals at the early third instar and in the wandering larvae (Fig. 5C, Fig. S2B). Furthermore, overexpression of GlyS slightly increased the amount of glycogen. Consistent with these observations, flies overexpressing either GlyS or GlyP by Tub-Gal4 were viable.

Fig. 5.

Differences in glycogen and trehalose metabolism. (A) Localization of overexpressed GlyS, GlyP and Treh in the fat body. The punctate localization of sTreh-Flag suggests secretory vesicles. Scale bar: 50 μm. (B) The effects of GlyS and GlyP overexpression on the levels of stored glycogen as visualized by PAS staining at the wandering stage. Scale bar: 100 μm. (C) The effects of GlyS, GlyP and Treh overexpression on glycogen and trehalose levels at the wandering stage. The total numbers of batches in each genotype are indicated. (D) The effects of transient induction of genes involved in glycogen and trehalose metabolism using a heat shock-Gal4 (hs-Gal4). Early third instar larvae (0 h) on a normal diet were heat-treated for 1 h and analyzed after 6 h or 18 h. Heat shock itself induced the mobilization of glycogen. Values shown are mean±s.e.m. n=3 batches. One-way ANOVA with Dunnett's post-hoc test (C), two-way ANOVA with Dunnett's post-hoc test with respect to values in control (D); **P<0.01, ***P<0.001.

Fig. 5.

Differences in glycogen and trehalose metabolism. (A) Localization of overexpressed GlyS, GlyP and Treh in the fat body. The punctate localization of sTreh-Flag suggests secretory vesicles. Scale bar: 50 μm. (B) The effects of GlyS and GlyP overexpression on the levels of stored glycogen as visualized by PAS staining at the wandering stage. Scale bar: 100 μm. (C) The effects of GlyS, GlyP and Treh overexpression on glycogen and trehalose levels at the wandering stage. The total numbers of batches in each genotype are indicated. (D) The effects of transient induction of genes involved in glycogen and trehalose metabolism using a heat shock-Gal4 (hs-Gal4). Early third instar larvae (0 h) on a normal diet were heat-treated for 1 h and analyzed after 6 h or 18 h. Heat shock itself induced the mobilization of glycogen. Values shown are mean±s.e.m. n=3 batches. One-way ANOVA with Dunnett's post-hoc test (C), two-way ANOVA with Dunnett's post-hoc test with respect to values in control (D); **P<0.01, ***P<0.001.

These glycogen results were in striking contrast to those regarding trehalose metabolism. The overexpression of Tps1 drastically increased trehalose (Fig. 5D) (Matsuda et al., 2015). Likewise, overexpression of the trehalose-hydrolysis enzyme Treh, especially secreted Treh (sTreh), strongly depleted trehalose (Fig. 5C,D). Chronic overexpression of cytoplasmic Treh (cTreh) had a marginal effect. However, the transient induction of cTreh using heat-shock Gal4 (hsGal4) transiently decreased trehalose (Fig. 5D). Thus, the overexpression of Tps1 and Treh was sufficient to promote trehalose synthesis and breakdown, respectively. These results suggest that the regulatory mechanisms for glycogen metabolism and trehalose metabolism are different.

Stored glycogen in oocytes is required for trehalose synthesis but largely dispensable for embryogenesis

Our genetic analyses demonstrated that homozygous glycogen mutants are viable. Consistent with a previous report (Sieber et al., 2016), glycogen accumulated in matured oocytes during oogenesis (Fig. S3A), suggesting that stored glycogen is a crucial energy reserve in embryogenesis. However, contrary to expectations, we found that embryos laid by either GlyS and GlyP homozygous females exhibited only a slight reduction in hatching rates (Fig. 6A). Moreover, GlyS and GlyP homozygous mutants could be maintained for more than a year. We noticed that the timing of hatching was significantly delayed by 2 h in both GlyS and GlyP mutants (Fig. 6B), suggesting that glycogen metabolism plays a role in timely progression of embryogenesis.

Fig. 6.

Stored glycogen in oocyte is required for trehalose synthesis but is largely dispensable for embryogenesis. (A) The hatching rate of embryos laid by homozygous GlyS and GlyP mutant females. The numbers of batches and the total number of embryos (in parentheses) are indicated. (B) Glycogen mutants exhibited 2-h delays in timing of larval hatching. (C) PC analysis of water-soluble metabolites at early (0-2 h) and late (22-24 h) embryonic stages. Ellipses of clusters show the 95% confidence regions for each sample group. (D) Numbers of significantly different metabolites (q<0.001 with respect to control) between GlyS and GlyP mutants at each stage. (E) Relative changes in the amounts of metabolites during embryogenesis. Metabolite levels were directly analyzed by LC-MS/MS or after enzymatic treatment (for glycogen). TAG levels indicate the sum of detected TAG lipid classes. Control (A-E): a genetically matched strain, w1118. n=8 batches (C-E); one-way ANOVA with Dunnett's post-hoc test (A) or Tukey's post-hoc test (E), Kolmogorov–Smirnov test (B); **P<0.01, ***P<0.001.

Fig. 6.

Stored glycogen in oocyte is required for trehalose synthesis but is largely dispensable for embryogenesis. (A) The hatching rate of embryos laid by homozygous GlyS and GlyP mutant females. The numbers of batches and the total number of embryos (in parentheses) are indicated. (B) Glycogen mutants exhibited 2-h delays in timing of larval hatching. (C) PC analysis of water-soluble metabolites at early (0-2 h) and late (22-24 h) embryonic stages. Ellipses of clusters show the 95% confidence regions for each sample group. (D) Numbers of significantly different metabolites (q<0.001 with respect to control) between GlyS and GlyP mutants at each stage. (E) Relative changes in the amounts of metabolites during embryogenesis. Metabolite levels were directly analyzed by LC-MS/MS or after enzymatic treatment (for glycogen). TAG levels indicate the sum of detected TAG lipid classes. Control (A-E): a genetically matched strain, w1118. n=8 batches (C-E); one-way ANOVA with Dunnett's post-hoc test (A) or Tukey's post-hoc test (E), Kolmogorov–Smirnov test (B); **P<0.01, ***P<0.001.

One possibility is that GlyS and GlyP mutants compensate for energy production by oxidizing other sources, such as stored lipids and amino acids. To understand the global change in metabolism caused by defective glycogen metabolism during embryogenesis, we conducted a comparative metabolomics analysis of water-soluble metabolites and TAGs (Table S2, Fig. S3B). PC analysis using 161 water-soluble metabolites revealed a separation between early embryo [0-2 h after egg laying (AEL)] and late embryo (22-24 h AEL) in the direction of PC1, whereas PC2 and PC3 separated three genotypes (i.e. control w, GlyS and GlyP mutants) (Fig. 6C). Venn diagram comparisons using the metabolomics data including water-soluble metabolites and TAGs revealed that GlyS and GlyP mutants showed common metabolic changes, especially at the late embryo stage (Fig. 6D, Fig. S3C), suggesting that glycogen mutants induce metabolic stress by the progression of embryogenesis.

We confirmed that embryos laid by GlyS mutant females did not retain glycogen (Fig. 6E). Embryos laid by GlyP mutant females had more glycogen than control embryos. Interestingly, a significant reduction of stored glycogen was observed in GlyP mutants during embryogenesis, suggesting that GlyP-independent degradation of glycogen occurs. We further found that consumption of TAGs during embryogenesis did not vary between control and glycogen mutants, although significant changes in the intermediates of fatty acid oxidation, such as even-chain acylcarnitines and acetylcarnitine, were detected (Fig. 6E). The increase of uric acid, a nitrogen-containing waste product of amino acid and purine catabolism, occurred normally in GlyS and GlyP mutants. Furthermore, reduction of ATP was not observed in glycogen mutants (Fig. 6E). These results suggest that loss of glycogen metabolism does not force a detectable switch to fatty acid or amino acid oxidation.

Interestingly, the increase of trehalose and glucose at late embryo was strongly impaired in GlyS and GlyP mutants (Fig. 6E), indicating that stored glycogen is utilized as a carbon source to produce trehalose and glucose during embryogenesis. Taken together, these results support the conclusion that glycogen metabolism, per se, is largely dispensable for embryogenesis; however, glycogen mutants do exhibit metabolic alterations.

Glycogen metabolism is crucial to maintain adult fitness under starvation conditions

To further understand the physiological significance of glycogen in adult fitness, we next analyzed physical ability. Adult flies attempt to climb to the top of a vial, in opposition to gravity, as an innate behavior (Rhodenizer et al., 2008). By recording their locomotion, we measured the average speed at which individual flies climbed. We found that GlyS mutant males had significantly decreased climbing ability, which was fully rescued by ubiquitous expression of GlyS (Fig. 7A). By contrast, GlyP mutants did not show remarkable deficits in climbing speed. Aging slightly reduced physical activity in GlyP mutants (Fig. 7B). We next conducted a flight performance test in 5-day-old males. Approximately 30% of GlyS mutants showed compromised flight ability, whereas GlyP mutants were largely normal (Fig. 7C). We further examined adult lifespan and found that GlyS mutant males had a slightly shortened lifespan (6-day difference in the median survival), whereas GlyP mutants had a normal lifespan (Fig. 7D). Taken together, these results suggest that GlyS is required for normal physical performance and adult fitness, whereas GlyP is largely dispensable under normal dietary conditions.

Fig. 7.

Glycogen metabolism is crucial for adult fitness under starvation conditions. (A) GlyS, but not GlyP, mutant males showed impaired physical fitness. Adult locomotion ability was assessed by a climbing assay. The average speeds of 5-day-old flies are shown. (B) Aging slightly enhanced the physical defect in GlyP mutants. The percentage of flies that climbed up to 10 cm was plotted. (C) A certain population of GlyS, but not GlyP, mutant males displayed compromised flight ability. The x-axis represents the vertical positions where the flies landed in the cylinder, and the y-axis represents the percentage of flies. The gray region indicates the bottom of the cylinder. (D) Adult lifespan in GlyS and GlyP mutant males. (E) Tps1MIC mutant males showed impaired physical activity. The average climbing speeds of 5-day-old flies are shown. (F) Both GlyS and GlyP mutants showed reduced survival rates after starvation. Ten-day-old flies were used for the starvation assay. (G) One-day starvation decreased the climbing activity in GlyP mutant males. The average speeds of 5-day-old flies are shown. Control (A-G): a genetically matched strain, w1118. The total numbers of flies from at least two independent experiments are indicated. One-way ANOVA with Dunnett's post-hoc test (A,E) or Tukey's post-hoc test (G), Kruskal–Wallis test with Dunn's correction (B,C), log-rank test (D,F); *P<0.05, ***P<0.001; n.s., not significant.

Fig. 7.

Glycogen metabolism is crucial for adult fitness under starvation conditions. (A) GlyS, but not GlyP, mutant males showed impaired physical fitness. Adult locomotion ability was assessed by a climbing assay. The average speeds of 5-day-old flies are shown. (B) Aging slightly enhanced the physical defect in GlyP mutants. The percentage of flies that climbed up to 10 cm was plotted. (C) A certain population of GlyS, but not GlyP, mutant males displayed compromised flight ability. The x-axis represents the vertical positions where the flies landed in the cylinder, and the y-axis represents the percentage of flies. The gray region indicates the bottom of the cylinder. (D) Adult lifespan in GlyS and GlyP mutant males. (E) Tps1MIC mutant males showed impaired physical activity. The average climbing speeds of 5-day-old flies are shown. (F) Both GlyS and GlyP mutants showed reduced survival rates after starvation. Ten-day-old flies were used for the starvation assay. (G) One-day starvation decreased the climbing activity in GlyP mutant males. The average speeds of 5-day-old flies are shown. Control (A-G): a genetically matched strain, w1118. The total numbers of flies from at least two independent experiments are indicated. One-way ANOVA with Dunnett's post-hoc test (A,E) or Tukey's post-hoc test (G), Kruskal–Wallis test with Dunn's correction (B,C), log-rank test (D,F); *P<0.05, ***P<0.001; n.s., not significant.

These results raise the question of why only GlyS mutants showed an apparent deficit in physical activity under fed conditions. GlyS mutants exhibited a specific decrease in circulating trehalose and glucose (Fig. 3A), so perhaps a reduction in circulating sugars explains the lower physical activity. To test this hypothesis, we examined a hypomorphic allele in Tps1 (Matsuda et al., 2015). As expected, Tps1MIC homozygous adults showed a significant decrease in climbing ability (Fig. 7E). These results support the possibility that the lower physical fitness in GlyS mutants is an indirect consequence of metabolic defects caused by impaired glycogenesis rather than the lack of glycogen metabolism as a fuel source.

If glycogen metabolism plays a role in maintaining energy homeostasis in times of metabolic need, GlyP mutants may decrease their physical activity after starvation. Indeed, we found that both GlyS and GlyP mutants had a lower survival rate after starvation (Fig. 7F). Control flies maintained their climbing activity 1 day after starvation compared with those under fed conditions (Fig. 7G, Fig. S4). However, both GlyS and GlyP mutants had decreased climbing speed when climbing performance was tested 1 day after starvation. Taken together, these results demonstrate that glycogen as a carbon source contributes to the maintenance of physical performance under fluctuations of dietary conditions.

In Drosophila, the importance of glycogen in homeostatic control, such as starvation tolerance, hypoxia tolerance, muscle function, protection from glucose toxicity, and aging, has been analyzed in tissue-specific overexpression and knockdown experiments (Duran et al., 2012; Paik et al., 2012; Zirin et al., 2013; Saez et al., 2014; Sinadinos et al., 2014; Garrido et al., 2015; Yamada et al., 2018). However, genetic null mutants of glycogen metabolism enzymes have not been reported.

In this study, we demonstrate that glycogen metabolism plays an important role in larval survival in collaboration with the circulating sugar trehalose. We also determine that muscle is an important organ for the observed lethality in GlyS and GlyP mutants. Interestingly, 90% of mice lacking muscle glycogen synthase (GYS1) die shortly after birth owing to abnormal cardiac development, but the surviving null mice grow up without obvious abnormalities, other than being smaller than wild-type mice (Pederson et al., 2004, 2005). Muscles are primarily defined as the organs for movement. However, muscle-specific knockdown of GlyS has no effect on larval locomotion under fed conditions (Zirin et al., 2013). Consistent with this, we did not detect a reduction in food intake, which depends on muscles for mouth hook movement and gut peristalsis (Min et al., 2017). Thus, the larval lethal effect appears not to rely on the movement function of muscles. Because of its mass and high metabolic rate during exercise, muscle has a profound influence on body metabolism. Recently, muscle has been shown to orchestrate many aspects of animal physiology, including body growth and lifespan, by regulating organismal energy homeostasis and governing systemic signaling networks (Droujinine and Perrimon, 2016; Rai and Demontis, 2016). For example, muscle-derived growth factors and cytokines, known as myokines, mediate the endocrine functions of muscles on other tissues. Thus, glycogen metabolism in muscles is important for larval development, likely through a systemic effect on body homeostasis.

Our results indicate that disrupting two enzymes acting in opposite directions provokes at least in part similar phenotypic traits. Potentially, common phenotypic traits in both GlyS and GlyP mutants reveal problems in carbohydrate homeostasis, whereas specific phenotypes in either GlyS or GlyP mutants reveal the consequence of lack or excess of glycogen granules. Interestingly, GlyS and GlyP mutants showed distinct metabolic defects in a life stage-specific manner. One possibility is that the lack or over-accumulation of glycogen influences intracellular signaling and transcription, resulting in changes in sugar and lipid metabolism. This idea is supported by the fact that glycogen inhibits the kinase activity of AMPK by binding to the glycogen-binding domain of the β-subunit (McBride et al., 2009). Moreover, glycogen granules act as a signaling scaffold via physical interaction with a variety of proteins (Graham, 2009; Philp et al., 2012; Stapleton et al., 2013). Thus, the flux of either glycolysis or fat oxidation may be promoted through a metabolic shift to compensate for the energy deficit in a mutant-specific manner. Considering that metabolism is intricately connected to diverse pathways (Kamleh et al., 2008; St Clair et al., 2017), impaired glycogenesis and glycogenolysis may also lead to distinct metabolic outcomes. Although the detailed mechanisms remain largely unknown, accumulating evidence has revealed a profound difference in carbohydrate metabolism between larvae and adults (Mattila and Hietakangas, 2017). Further analysis will be required to clarify the mutant- and life stage-specific metabolic changes.

We found that the majority of glycogen mutant embryos successfully hatch, but with a 2-h delay. In contrast to glycogen, the amount of trehalose increases during embryogenesis (An et al., 2014; Tennessen et al., 2014; Matsuda et al., 2015). Our results indicate that the carbon source of trehalose synthesis during embryogenesis is highly dependent on stored glycogen in oocytes. Nevertheless, trehalose synthesized in the embryo is dispensable for embryogenesis. This result is consistent with our previous report showing that maternal and zygotic mutation of Tps1 does not result in embryonic lethality (Matsuda et al., 2015). Because Tps1 and Treh mutant larvae are extremely sensitive to dietary deprivation (Matsuda et al., 2015; Yoshida et al., 2016), the physiological significance of trehalose synthesis during embryogenesis is the most likely explanation for survival immediately after hatching. These results are consistent with previous reports showing that global expression of genes involved in aerobic glycolysis occurs at the end of the embryonic stage for metabolic adaptation in larvae (Tennessen et al., 2011, 2014). In this regard, a major purpose of glycogen deposition in matured oocytes is to distribute a source of glucose for larval survival but not a fuel source for embryogenesis. The amount of trehalose should be spatiotemporally regulated in embryogenesis because ectopic trehalose synthesis potentially induces uncontrolled osmotic swelling that results in lethality (Matsuda et al., 2015; Yoshida et al., 2016) and that precludes direct deposition of trehalose in oocytes. These observations further imply that stored TAGs are likely an indispensable fuel source to complete embryogenesis. Interestingly, GlyP mutants deposit normal levels of TAGs in embryos, despite a significant reduction in the whole body levels of TAGs in adults, suggesting that adult females are able to control the correct deposition of stored lipids in oocytes, presumably through a mechanism involved in local steroid signaling (Sieber and Spradling, 2015).

Our results also indicate that stored glycogen is partly consumed during embryogenesis in GlyP mutants. It appears that GlyP-independent glycogenolysis occur at this stage, most likely via lysosome/autophagy-dependent degradation of glycogen (Roach et al., 2012; Zirin et al., 2013). The partial mobilization of stored glycogen in GlyP mutants is strongly supported by the fact that GlyP mutants synthesize trehalose to some degree during embryogenesis. In addition, relatively weak embryonic lethality in GlyP mutants compared with GlyS mutants could also be attributed to the partial glycogenolysis.

We found that the ectopic induction of GlyS has a modest impact on glycogen levels, whereas that of GlyP is not sufficient to promote glycogen mobilization in larvae. Because glycogen metabolism is tightly regulated by allosteric and post-translational modifications (Bouskila et al., 2010; Roach et al., 2012), it is reasonable to assume that increased levels of GlyS and GlyP fail to facilitate either glycogenesis or glycogenolysis. Levels of glycogenin, a core protein for glycogen synthesis, may also restrict the amount of glycogen (Lomako et al., 2004; Roach et al., 2012). It should be noted, however, that GlyP overexpression reduces glycogen in adults (Post et al., 2018) and the expression of human GYS1 increases the amount of glycogen in Drosophila adult neurons (Sinadinos et al., 2014). Furthermore, GlyS and GlyP expression is elevated in whole larvae on a high-sugar diet (Musselman et al., 2011; Garrido et al., 2015; Mattila et al., 2015). Thus, transcriptional regulation of GlyS and GlyP may facilitate glycogen metabolism depending on life stages or in response to dietary conditions. By contrast, the overexpression of Treh is sufficient to hydrolyze circulating trehalose. Considering that Tps1 overexpression increases trehalose and leads to pupal lethality (Matsuda et al., 2015), transcriptional regulation is likely sufficient to facilitate trehalose metabolism. It will be interesting to explore further how the distinct regulatory systems in glycogen and trehalose metabolism mirror the functional relevance of these storage sugars throughout the life cycle.

A previous report indicated a significant drop in flight performance in GlyP mutants (Eanes et al., 2006), but we did not observe any such abnormality in our GlyP mutants. This discrepancy might arise from the difference in experimental conditions; namely, a tethered flight performance test focusing on wing beat frequency (in the previous work) versus a release-based flight test without tethering (in the present study). Alternatively, dietary conditions may affect the flight performance because starvation significantly decreases climbing activity in GlyP mutants. Interestingly, loss of neuronal glycogen improves locomotion ability during aging in Drosophila (Sinadinos et al., 2014). These observations raise the possibility that local defects in glycogen metabolism result in distinct outcomes compared with inherited mutations. In humans, several inherited metabolic disorders, collectively referred to as glycogen storage diseases, are caused by deficiencies of enzymes involved in glycogen synthesis or breakdown (Özen, 2007; Adeva-Andany et al., 2016). An explicit comparison of GlyS and GlyP mutants will advance our understanding of life stage-specific requirements for glycogen metabolism with respect to carbohydrate metabolism and body homeostasis.

Drosophila strains

The following Drosophila melanogaster strains were used: w1118 (used as a control), UAS-Tps1, Tps1d2, Tps1MI03087 (Matsuda et al., 2015). y2 cho2 v1, P{nos-Cas9, y+, v+}/FM7c, KrGal4, UAS-GFP (CAS-0002) was obtained from the National Institute of Genetics (NIG) Drosophila Stock Center. Df(3R)ED10561 (a deficiency strain of the GlyS locus, no. 150359) and Df(2L)ED119 (a deficiency strain of the GlyP locus, no. 150067) were obtained from the Kyoto Drosophila Stock Center. The following stocks were obtained from the Bloomington Drosophila Stock Center (BDSC): GlyS-RNAi (34930), GlyP-RNAi (33634), hs-Gal4 (1799), tub-Gal4 (5138), Mef2-Gal4 (27390), elav-Gal4 (8760, 8765), Repo-Gal4 (7415), Cg-Gal4 (7011), P{lacW}GlyPk07918 (10692), Mi{MIC}GlyPMI00957 (34131) and Mi{MIC}GlySMI01490 (34440).

Of note, the original GlySMI01490 and GlyPMI00957 lines from the BDSC showed complete lethality in the early larvae or embryo; however, transheterozygotes of each line over a deficiency line exhibited semi-lethality in larvae that resulted in the emergence of escaping adult flies, indicating that these Minos-insertion lines had second-site mutation(s).

Generation of the GlyS mutants

Generation of the GlyS mutant allele was carried out using the CRISPR/Cas9 system. Sense and antisense oligonucleotides corresponding to sgRNA target sequences were annealed and inserted into BbsI-digested pBFv-U6.2 vector (obtained from NIG). The GlyS sgRNA vector was injected into embryos carrying attP40 and nos-phiC31, and transgenic strains were generated (BestGene). The nos-Cas9-based gene targeting was carried out as previously described (Kondo and Ueda, 2013). Independent isogenized strains for each sgRNA construct were established. Indel mutations were analyzed via genome DNA extraction and PCR amplification of the DNA fragment including the target site, followed by sequence analysis. We isolated several frameshift mutations for GlyS using two different sgRNAs. All isolated GlyS mutants were semi-viable; that is, approximately 10-30% survived to adulthood. We chose one strain, GlyS8, for further analyses after backcrossing four times to the w control strain. The backcrossed GlyS8 homozygous mutants in w background were used in all experiments unless otherwise noted.

Characterization and generation of GlyP mutants

A transposable P-element insertion P{lacW}k07918 (BDSC, 10692) was backcrossed to the w control strain to remove the lethal mutation in the ft locus as described previously (Tick et al., 1999). Consistent with this finding, transheterozygotes of the original P{lacW}k07918 line with a deficiency allele lacking the GlyP locus, but not the ft locus, displayed homozygous viability. The backcrossed P-element line (w+) was used to generate GlyP mutants by imprecise excision. The progeny were first screened for the loss of an eye color marker (w+), and the extent of the deletion in each mutant was determined by PCR and subsequent DNA sequencing. GlyP3-11 and GlyP3-13 carried deletions of 1463 base pairs and 1710 base pairs, respectively, both of which resulted in the lack of the entire first exon, including the transcriptional and translational start sites. GlyP3-13 was backcrossed three times to the w control strain. The backcrossed GlyP3-13 homozygous mutants in w background were used in all experiments unless otherwise noted.

Fly food

Drosophila were reared on fly food (normal diet, ND) containing 8 g agar, 100 g glucose, 45 g dry yeast, 40 g corn flour, 4 ml propionic acid and 0.45 g butylparaben (in ethanol) per liter (1× recipe). For the analysis of restricted or enriched dietary conditions, fly food was prepared as described previously (Matsuda et al., 2015; Yasugi et al., 2017). In brief, fly food containing a reduced or increased amount of yeast or glucose was used (0.2× or 2× the amount used for the 1× recipe). The yeast-only diet was prepared according to the 1× recipe. No yeast paste was added to the fly tubes in any of the experiments. All the experiments were conducted under non-crowded conditions at 25°C. The food intake assay was carried out as described previously (Yasugi et al., 2017).

qRT-PCR experiments

qRT-PCR analysis was performed as described previously (Okamoto et al., 2012; Okamoto and Nishimura, 2015). The primers used in this study have also been described before (Yamada et al., 2018). The primers for GlyS and GlyP genes were located in the C terminus of the coding region.

Developmental staging and survival rate

Developmental staging was performed essentially as previously described (Okamoto et al., 2012; Okamoto and Nishimura, 2015). Early third instar larvae were defined as 0-6 h after second ecdysis. A defined number of newly hatched larvae were placed on each vial, and the numbers of pupae and adults were counted. Embryos laid over a time period of 4 h were used to determine the timing of pupariation. Pupariation was visually assessed two times per day, and the average time was calculated for each vial. Experiments were repeated at least twice, and all data were summed.

Quantification of weight and body size

Quantification of adult weight was performed as previously described (Okamoto et al., 2013). Adult flies, pupae and larvae were photographed under a Zeiss Stemi 2000-C stereomicroscope equipped with a Canon PowerShot G15 digital camera (Canon). The volumes of larvae and pupae were determined using the formula 4/3π(L/2)(l/2)2, where L is the length and l is the diameter.

Plasmid construction

The cDNAs encoding sTreh, cTreh, GlyS and GlyP were cloned by RT--PCR using sequenced strains (BDSC, 2057) and then subcloned into the pCR-BluntII-TOPO vector (Invitrogen). The cDNAs were subcloned into a modified pUAST vector that contained a C-terminal 1xFlag tag. Transformants were obtained using a standard injection method (BestGene).

Measurements of protein, TAG and sugar levels

Protein, TAG, trehalose, glycogen and glucose levels were measured as described previously, with minor modifications (Matsuda et al., 2015).

Frozen samples in tubes were homogenized using a pellet pestle in 100 µl of cold PBS containing 0.1% Triton X-100, immediately heat-inactivated at 80°C for 10 min, and then cooled to room temperature (RT). Samples were further crushed to obtain uniform homogenates with 1× φ3-mm zirconia beads using an automill (Tokken) at 41.6 Hz for 2 min. A portion of homogenate was mixed with triglyceride reagent (Sigma-Aldrich,), incubated at 37°C for at least 30 min, and then cleared by centrifugation at 20,000 g for 10 min. The supernatant was used for measurement of TAG using free glycerol reagent (Sigma-Aldrich). A triolein equivalent glycerol (Sigma-Aldrich) was used as the standard. The amount of TAG was normalized to the total protein level.

Homogenate samples were further used to determine the glycogen levels, and the cleared samples after centrifugation at 10,000 rpm (9000 g) for 10 min at RT were used to determine trehalose and glucose levels. A portion of homogenate was incubated with PBS containing bacterially produced recombinant His-tagged Drosophila cTreh (Yoshida et al., 2016) or amyloglucosidase (Sigma-Aldrich) at 37°C overnight. A portion of sample was incubated with PBS without enzymes in parallel for the determination of glucose levels. The amounts of samples were adjusted based on prior experience to obtain linearity within the range of standards. The reaction was carried out in a 15-µl assay mixture. Glucose levels were determined using a glucose assay kit (Sigma-Aldrich). A serial dilution of glucose was used as a standard. The trehalose and glycogen concentrations for each sample were determined by subtracting the values of free glucose in the untreated samples. Trehalose, glycogen and glucose levels were normalized to the protein level, which was determined by a BCA protein assay kit (Thermo Scientific).

Metabolite extraction and a widely targeted metabolomics profile

Frozen samples in 1.5 ml plastic tubes were homogenized in 300 µl of cold methanol with 1× φ3-mm zirconia beads using an automill (Tokken) at 41.6 Hz for 2 min. The homogenates were mixed with 200 µl of methanol, 200 µl of H2O and 200 µl of CHCl3 and then vortexed for 20 min at RT. The samples were centrifuged at 15,000 rpm (20,000 g) for 15 min at 4°C. The supernatant was mixed with 350 µl of H2O and vortexed for 10 min at RT. The aqueous phase was collected after centrifugation and dried in a vacuum concentrator. The samples were re-dissolved in 2 mM ammonium bicarbonate (pH 8.0) and analyzed by liquid chromatography with tandem mass spectroscopy (LC-MS/MS). The insoluble pellets were washed with 90% ethanol, re-dissolved with PBS containing 0.1% Triton X-100, and used for quantification of glycogen after treatment with amyloglucosidase. The samples were further mixed with two volumes of 0.2 N NaOH, heat-denatured, and used to quantify total protein using a BCA protein assay kit (Thermo Scientific).

An in-house platform for a widely targeted analysis was established by making scheduled multiple reaction monitoring (MRM) methods using individual authentic compounds and biological samples. Chromatographic separations in an Acquity UPLC H-Class System (Waters) were carried out under reverse-phase conditions using an Acquity UPLC HSS T3 column and under HILIC conditions using an Acquity UPLC BEH Amide column. The ionized compounds were detected using a Xevo TQD triple quadrupole mass spectrometer coupled with an electro-spray ionization source (Waters). The peak area of a target metabolite was analyzed using MassLynx 4.1 software (Waters). Metabolite signals were then normalized to the total protein level of the corresponding sample after subtracting the values from the blank sample. P-values were calculated by two-tailed Welch's t-test using Microsoft Excel. An estimate of the false discovery rate (FDR), which is given by the q-value, was calculated to take into account the multiple comparisons. Further statistical analyses were performed using MetaboAnalyst 4.0 (Xia et al., 2009). For multivariate analysis, PC analysis was performed to identify biologically relevant classifications (Worley and Powers, 2013). Data were normalized to the median per sample. Heat map and hierarchical clustering were generated using Pearson correlations and Ward's method. Enrichment analysis was performed using MetaCore software ver. 6.37 (Clarivate Analytics). Networks with In Data>2 are shown in Table S1.

For the TAG analysis, the organic phase was collected after the addition of hexane. Dried samples were re-dissolved in isopropanol and analyzed by LC-MS/MS using an Acquity BEH C18 column (Waters).

Preparation of 13C-labeled glycogen

Control pupae raised on a normal diet were transferred to a new vial containing a diet composed of 5% yeast and either 5% D-glucose (Wako Chemical) or 5% D-[U-13C6] glucose (Cambridge Isotope Laboratories). Adult flies were collected 4 days after eclosion and transferred to 1.5 ml tubes (100 flies per tube). Samples were homogenized in 1 ml of 10% trichloroacetic acid and cleared by centrifugation at 15,000 rpm (20,000 g) for 10 min at RT. The supernatant was diluted with 10 times volume of ethanol in a 50 ml tube. The pellet containing glycogen was recovered by centrifugation at 15,000 rpm (20,000 g) for 10 min at 4°C and washed once with 90% ethanol. Glycogen was dissolved in 1× PBS, filtered through an Amicon filter (100 kDa MWCO), and washed five times with 1× PBS to remove low-molecular weight contaminants.

The purity of glycogen was confirmed by a high-performance liquid chromatography (Alliance 2695, Waters) coupled with a refractive index detector (Waters 2414). Size-exclusion chromatography was performed at 40°C using a Shodex OHpak SB-806M HQ column (8.0×300 mm, Shodex). Elution was carried out with 1× PBS at a flow rate of 0.5 ml/min isocratic. Pullulan standards were obtained from Shodex. The proportion of glucose isotopologues in purified glycogen was analyzed by LC-MS/MS after treatment with amyloglucosidase (Sigma).

GlyS enzyme assay

Ten early third instar larvae were dissected in PBS. Dissected tissues were homogenized in cold 100 µl 2× homogenize buffer [2× PBS, 4 mM MgCl2, 0.2% Triton X-100, 2× complete proteinase inhibitor cocktail (Roche) and 1 mM PMSF] with motorized pestles. Samples (10 µl) were mixed on ice with 10 µl of 2× GlyS assay mixture containing 3.4 mM UDP-[U-13C6] glucose (Omicron Biochemicals), 76 mM glucose-6-phosphate (Wako Chemical) and 200 µg rabbit liver glycogen (Wako Chemical) in H2O. Reactions were started by placing samples on a 30°C block incubator for the indicated times and stopped by the addition of 200 µl 10% trichloroacetic acid. The samples were cleared by centrifugation at 15,000 rpm (20,000 g) for 10 min at RT. The supernatant was diluted with 1 ml ethanol. The pellet containing glycogen was recovered by centrifugation at 15,000 rpm (20,000 g) for 10 min at 4°C and washed once with 90% ethanol. The pellet was dissolved in 20 µl 1× PBS containing amyloglucosidase (Sigma). After incubation overnight at 37°C, samples were mixed with 500 µl acetonitril containing 50 ng 13C1-mannitol (Sigma) as an internal standard and cleared by centrifugation at 15,000 rpm (20,000 g) for 10 min at 4°C. The supernatant was loaded onto a MonoSpin Amide column (GL Science). The bound metabolites were eluted with 300 µl H2O and then vacuum dried. Samples were re-dissolved in 2 mM ammonium bicarbonate and analyzed by UPLC-MS/MS using an Acquity BEH Amide column (Waters). The MRM transitions (negative ion mode) were as follows: 13C1-Mannitol m/z 182>89, 13C6-glucose m/z 185>92. The incorporation rate of 13C6-glucose into glycogen was determined after normalization with values of 13C1-mannitol.

GlyP enzyme assay

Tissue homogenates were prepared as described above. Samples (10 µl) were mixed on ice with 10 µl 2× GlyP assay mixture containing 200 µg 13C6-labeled glycogen in H2O. Reactions were started by placing samples on a 30°C block incubator for the indicated times and stopped by the addition of 500 µl acetonitrile containing 50 ng 13C1-mannitol (Sigma). The samples were cleared by centrifugation at 15,000 rpm (20,000 g) for 10 min at 4°C. The supernatant was loaded onto a MonoSpin Amide column (GL Science). The bound metabolites were eluted with 300 µl H2O and vacuum dried. Samples were re-dissolved in 2 mM ammonium bicarbonate and analyzed by LC-MS/MS using an Acquity HSS T3 column (Waters). MRM transitions (positive ion mode) were as follows: 13C6-glucose-phosphate (mixture of glucose-1-phosphate and glucose-6-phosphate) m/z 267>99. The release rate of 13C6-glucose-phosphate from 13C6-labeled glycogen was determined after normalization with values of 13C1-mannitol.

Histochemistry

Polysaccharide staining, including glycogen, was performed as described previously (Yamada et al., 2018). Larval tissues were dissected in PBS containing 1% bovine serum albumin (PBSB), fixed using 3.7% formaldehyde in PBS for 20 min, washed twice in PBSB, incubated with periodic acid solution (Merck) for 5 min, and washed twice in PBSB. Samples were stained with Schiff's reagent (Merck) for 15 min, washed twice in PBSB, and mounted in 50% glycerol in PBS. Images were acquired with a Zeiss Primo Star microscope equipped with an AxioCam ERc (Zeiss). The experiments were conducted independently at least twice, and representative images obtained from several larvae are presented.

Immunohistochemistry

Larval tissues were dissected in PBS, fixed for 10 min in 3.7% formaldehyde in PBS containing 0.2% Triton X-100, and processed as previously described (Wirtz-Peitz et al., 2008). Mouse anti-Flag M2 antibody (Sigma, F1804, 1:500) was used as the primary antibody. Alexa-conjugated secondary antibodies (A11029, 1:500) and phalloidin (Thermo Fisher Scientific) were used. Images were acquired with a Zeiss LSM700 confocal microscope and processed in Photoshop (Adobe Systems).

Climbing assay

Groups of 15-20 male flies, reared under conditions of controlled population density, were collected within 24 h of eclosion. The flies were transferred to fresh vials every second or third day until the indicated ages, without further exposure to CO2. Before testing, the flies were transferred to a 100 ml plastic graduated cylinder and allowed to rest for 10 min. The flies were tapped down to the bottom of the cylinder, and their locomotion was recorded using a Canon PowerShot G1X digital camera (Canon). The first five trials were conducted for habituation to the environment, and the following three trials were used for the analysis. The average climbing speed (cm/s) was determined in ImageJ software by tracing the position of flies every second until they reached the top. The experiment was repeated with at least two independently reared replicates, and all data were summed.

Flight test

A φ20×100 cm acrylic cylinder was used for the flight test. Groups of ten male flies (2-5 days after eclosion) were placed in the top of the cylinder through a funnel to initiate flight. The height at which flies landed in the cylinder was recorded by a Canon PowerShot G15 digital camera (Canon) and judged as their flight ability. The experiment was repeated with at least two independently reared populations, and all data were summed.

Lifespan and transient starvation assay

Flies were reared under conditions of controlled population density and temperature. Groups of 20 flies (males for lifespan assay, males and females separately for starvation assay) were collected within 24 h of eclosion. The flies were transferred to fresh vials twice per week, and the number of dead flies was recorded. For the transient starvation assay, 10-day-old flies were transferred to a vial that contained 0.8% agar, and the number of dead flies was recorded. For each genotype, at least four independent cohorts of flies were analyzed. Kaplan–Meier lifespan analysis was performed, and P-values were calculated using log-rank statistical analysis.

Statistical analysis and data presentation

Statistical tests were performed using Microsoft Excel or GraphPad Prism 6 software. Box and violin plots were drawn online using the BoxPlotR application (http://boxplot.tyerslab.com/). For box plots, centerlines show the medians, box limits indicate the 25th and 75th percentiles, whiskers extend 1.5 times the interquartile range from the 25th and 75th percentiles, and outliers are represented by dots. For violin plots, white circles show the medians, box limits indicate the 25th and 75th percentiles, whiskers extend 1.5 times the interquartile range from the 25th and 75th percentiles, and polygons represent density estimates of data and extend to extreme values.

We thank the Bloomington Drosophila Stock Center, the Kyoto Stock Center, and the National Institute of Genetics Drosophila Stock Center for the fly stocks. We also thank members of fly laboratories in RIKEN BDR for their valuable support and discussion; the genome resource and analysis unit in RIKEN BDR for their technical support; and K. Hironaka, R. Niwa and S. K. Yoo for discussions and comments on the manuscript.

Author contributions

Conceptualization: T.Y., T.N.; Validation: T.Y., O.H., T.N.; Formal analysis: T.Y., O.H., Y.Y., R.M., H.K., Y.N., T.N.; Investigation: T.Y., O.H., Y.Y., R.M., H.K., T.N.; Writing - original draft: T.N.; Writing - review & editing: T.Y., Y.Y., Y.N.; Visualization: T.Y., T.N.; Supervision: T.N.; Project administration: T.N.; Funding acquisition: T.N.

Funding

This work was supported, in part, by the Japan Society for the Promotion of Science (JSPS) [KAKENHI grants JP17K19433 and JP17H03658 to T.N.].

Adeva-Andany
,
M. M.
,
González-Lucán
,
M.
,
Donapetry-García
,
C.
,
Fernández-Fernández
,
C.
and
Ameneiros-Rodríguez
,
E.
(
2016
).
Glycogen metabolism in humans
.
Biochim. Biophys. Acta Clin.
5
,
85
-
100
.
An
,
P. N. T.
,
Yamaguchi
,
M.
,
Bamba
,
T.
and
Fukusaki
,
E.
(
2014
).
Metabolome analysis of Drosophila melanogaster during embryogenesis
.
PLoS ONE
9
,
e99519
.
Bai
,
H.
,
Kang
,
P.
,
Hernandez
,
A. M.
and
Tatar
,
M.
(
2013
).
Activin signaling targeted by insulin/dFOXO regulates aging and muscle proteostasis in Drosophila
.
PLoS Genet.
9
,
e1003941
.
Bouskila
,
M.
,
Hunter
,
R. W.
,
Ibrahim
,
A. F. M.
,
Delattre
,
L.
,
Peggie
,
M.
,
van Diepen
,
J. A.
,
Voshol
,
P. J.
,
Jensen
,
J.
and
Sakamoto
,
K.
(
2010
).
Allosteric regulation of glycogen synthase controls glycogen synthesis in muscle
.
Cell Metab.
12
,
456
-
466
.
Chng
,
W.-A.
,
Hietakangas
,
V.
and
Lemaitre
,
B.
(
2017
).
Physiological adaptations to sugar intake: new paradigms from Drosophila melanogaster
.
Trends Endocrinol. Metab.
28
,
131
-
142
.
Droujinine
,
I. A.
and
Perrimon
,
N.
(
2016
).
Interorgan communication pathways in physiology: focus on Drosophila
.
Annu. Rev. Genet.
50
,
539
-
570
.
Duran
,
J.
,
Tevy
,
M. F.
,
Garcia-Rocha
,
M.
,
Calbó
,
J.
,
Milán
,
M.
and
Guinovart
,
J. J.
(
2012
).
Deleterious effects of neuronal accumulation of glycogen in flies and mice
.
EMBO Mol. Med.
4
,
719
-
729
.
Eanes
,
W. F.
,
Merritt
,
T. J. S.
,
Flowers
,
J. M.
,
Kumagai
,
S.
,
Sezgin
,
E.
and
Zhu
,
C.-T.
(
2006
).
Flux control and excess capacity in the enzymes of glycolysis and their relationship to flight metabolism in Drosophila melanogaster
.
Proc. Natl. Acad. Sci. USA
103
,
19413
-
19418
.
Elbein
,
A. D.
,
Pan
,
Y. T.
,
Pastuszak
,
I.
and
Carroll
,
D.
(
2003
).
New insights on trehalose: a multifunctional molecule
.
Glycobiology
13
,
17R
-
27R
.
Gáliková
,
M.
,
Diesner
,
M.
,
Klepsatel
,
P.
,
Hehlert
,
P.
,
Xu
,
Y.
,
Bickmeyer
,
I.
,
Predel
,
R.
and
Kühnlein
,
R. P.
(
2015
).
Energy homeostasis control in Drosophila adipokinetic hormone mutants
.
Genetics
201
,
665
-
683
.
Garrido
,
D.
,
Rubin
,
T.
,
Poidevin
,
M.
,
Maroni
,
B.
,
Le Rouzic
,
A.
,
Parvy
,
J.-P.
and
Montagne
,
J.
(
2015
).
Fatty acid synthase cooperates with glyoxalase 1 to protect against sugar toxicity
.
PLoS Genet.
11
,
e1004995
.
Graham
,
T. E.
(
2009
).
Glycogen: an overview of possible regulatory roles of the proteins associated with the granule
.
Appl. Physiol. Nutr. Metab.
34
,
488
-
492
.
Gusarov
,
I.
,
Pani
,
B.
,
Gautier
,
L.
,
Smolentseva
,
O.
,
Eremina
,
S.
,
Shamovsky
,
I.
,
Katkova-Zhukotskaya
,
O.
,
Mironov
,
A.
and
Nudler
,
E.
(
2017
).
Glycogen controls Caenorhabditis elegans lifespan and resistance to oxidative stress
.
Nat. Commun.
8
,
15868
.
Kamleh
,
M. A.
,
Hobani
,
Y.
,
Dow
,
J. A. T.
and
Watson
,
D. G.
(
2008
).
Metabolomic profiling of Drosophila using liquid chromatography fourier transform mass spectrometry
.
FEBS Lett.
582
,
2916
-
2922
.
Karousis
,
E. D.
,
Nasif
,
S.
and
Mühlemann
,
O.
(
2016
).
Nonsense-mediated mRNA decay: novel mechanistic insights and biological impact
.
Wiley Interdiscip. Rev. RNA
7
,
661
-
682
.
Kondo
,
S.
and
Ueda
,
R.
(
2013
).
Highly improved gene targeting by germline-specific Cas9 expression in Drosophila
.
Genetics
195
,
715
-
721
.
Lomako
,
J.
,
Lomako
,
W. M.
and
Whelan
,
W. J.
(
2004
).
Glycogenin: the primer for mammalian and yeast glycogen synthesis
.
Biochim. Biophys. Acta
1673
,
45
-
55
.
Matsuda
,
H.
,
Yamada
,
T.
,
Yoshida
,
M.
and
Nishimura
,
T.
(
2015
).
Flies without trehalose
.
J. Biol. Chem.
290
,
1244
-
1255
.
Mattila
,
J.
and
Hietakangas
,
V.
(
2017
).
Regulation of carbohydrate energy metabolism in Drosophila melanogaster
.
Genetics
207
,
1231
-
1253
.
Mattila
,
J.
,
Havula
,
E.
,
Suominen
,
E.
,
Teesalu
,
M.
,
Surakka
,
I.
,
Hynynen
,
R.
,
Kilpinen
,
H.
,
Väänänen
,
J.
,
Hovatta
,
I.
,
Käkelä
,
R.
, et al.
(
2015
).
Mondo-Mlx mediates organismal sugar sensing through the Gli-similar transcription factor Sugarbabe
.
Cell Rep.
13
,
350
-
364
.
McBride
,
A.
,
Ghilagaber
,
S.
,
Nikolaev
,
A.
and
Hardie
,
D. G.
(
2009
).
The glycogen-binding domain on the AMPK beta subunit allows the kinase to act as a glycogen sensor
.
Cell Metab.
9
,
23
-
34
.
Min
,
S.
,
Yoon
,
W.
,
Cho
,
H.
and
Chung
,
J.
(
2017
).
Misato underlies visceral myopathy in Drosophila
.
Sci. Rep.
7
,
17700
.
Musselman
,
L. P.
,
Fink
,
J. L.
,
Narzinski
,
K.
,
Ramachandran
,
P. V.
,
Hathiramani
,
S. S.
,
Cagan
,
R. L.
and
Baranski
,
T. J.
(
2011
).
A high-sugar diet produces obesity and insulin resistance in wild-type Drosophila
.
Dis. Model Mech.
4
,
842
-
849
.
Okamoto
,
N.
and
Nishimura
,
T.
(
2015
).
Signaling from glia and cholinergic neurons controls nutrient-dependent production of an insulin-like peptide for Drosophila body growth
.
Dev. Cell
35
,
295
-
310
.
Okamoto
,
N.
,
Nishimori
,
Y.
and
Nishimura
,
T.
(
2012
).
Conserved role for the Dachshund protein with Drosophila Pax6 homolog Eyeless in insulin expression
.
Proc. Natl. Acad. Sci. USA
109
,
2406
-
2411
.
Okamoto
,
N.
,
Nakamori
,
R.
,
Murai
,
T.
,
Yamauchi
,
Y.
,
Masuda
,
A.
and
Nishimura
,
T.
(
2013
).
A secreted decoy of InR antagonizes insulin/IGF signaling to restrict body growth in Drosophila
.
Genes Dev.
27
,
87
-
97
.
Özen
,
H.
(
2007
).
Glycogen storage diseases: new perspectives
.
World J. Gastroenterol.
13
,
2541
-
2553
.
Paik
,
D.
,
Jang
,
Y. G.
,
Lee
,
Y. E.
,
Lee
,
Y. N.
,
Yamamoto
,
R.
,
Gee
,
H. Y.
,
Yoo
,
S.
,
Bae
,
E.
,
Min
,
K.-J.
,
Tatar
,
M.
, et al.
(
2012
).
Misexpression screen delineates novel genes controlling Drosophila lifespan
.
Mech. Ageing Dev.
133
,
234
-
245
.
Pederson
,
B. A.
,
Chen
,
H.
,
Schroeder
,
J. M.
,
Shou
,
W.
,
DePaoli-Roach
,
A. A.
and
Roach
,
P. J.
(
2004
).
Abnormal cardiac development in the absence of heart glycogen
.
Mol. Cell. Biol.
24
,
7179
-
7187
.
Pederson
,
B. A.
,
Schroeder
,
J. M.
,
Parker
,
G. E.
,
Smith
,
M. W.
,
DePaoli-Roach
,
A. A.
and
Roach
,
P. J.
(
2005
).
Glucose metabolism in mice lacking muscle glycogen synthase
.
Diabetes
54
,
3466
-
3473
.
Petersen
,
M. C.
,
Vatner
,
D. F.
and
Shulman
,
G. I.
(
2017
).
Regulation of hepatic glucose metabolism in health and disease
.
Nat. Rev. Endocrinol.
13
,
572
-
587
.
Philp
,
A.
,
Hargreaves
,
M.
and
Baar
,
K.
(
2012
).
More than a store: regulatory roles for glycogen in skeletal muscle adaptation to exercise
.
Am. J. Physiol. Endocrinol. Metab.
302
,
E1343
-
E1351
.
Post
,
S.
,
Karashchuk
,
G.
,
Wade
,
J. D.
,
Sajid
,
W.
,
De Meyts
,
P.
and
Tatar
,
M.
(
2018
).
Drosophila insulin-like peptides DILP2 and DILP5 differentially stimulate cell signaling and glycogen phosphorylase to regulate longevity
.
Front. Endocrinol.
9
,
245
.
Rai
,
M.
and
Demontis
,
F.
(
2016
).
Systemic nutrient and stress signaling via myokines and myometabolites
.
Annu. Rev. Physiol.
78
,
85
-
107
.
Rhodenizer
,
D.
,
Martin
,
I.
,
Bhandari
,
P.
,
Pletcher
,
S. D.
and
Grotewiel
,
M.
(
2008
).
Genetic and environmental factors impact age-related impairment of negative geotaxis in Drosophila by altering age-dependent climbing speed
.
Exp. Gerontol.
43
,
739
-
748
.
Roach
,
P. J.
,
Depaoli-Roach
,
A. A.
,
Hurley
,
T. D.
and
Tagliabracci
,
V. S.
(
2012
).
Glycogen and its metabolism: some new developments and old themes
.
Biochem. J.
441
,
763
-
787
.
Saez
,
I.
,
Duran
,
J.
,
Sinadinos
,
C.
,
Beltran
,
A.
,
Yanes
,
O.
,
Tevy
,
M. F.
,
Martínez-Pons
,
C.
,
Milán
,
M.
and
Guinovart
,
J. J.
(
2014
).
Neurons have an active glycogen metabolism that contributes to tolerance to hypoxia
.
J. Cereb. Blood Flow Metab.
34
,
945
-
955
.
Saltiel
,
A. R.
and
Kahn
,
C. R.
(
2001
).
Insulin signalling and the regulation of glucose and lipid metabolism
.
Nature
414
,
799
-
806
.
Sieber
,
M. H.
and
Spradling
,
A. C.
(
2015
).
Steroid signaling establishes a female metabolic state and regulates SREBP to control oocyte lipid accumulation
.
Curr. Biol.
25
,
993
-
1004
.
Sieber
,
M. H.
,
Thomsen
,
M. B.
and
Spradling
,
A. C.
(
2016
).
Electron transport chain remodeling by GSK3 during oogenesis connects nutrient state to reproduction
.
Cell
164
,
420
-
432
.
Sinadinos
,
C.
,
Valles-Ortega
,
J.
,
Boulan
,
L.
,
Solsona
,
E.
,
Tevy
,
M. F.
,
Marquez
,
M.
,
Duran
,
J.
,
Lopez-Iglesias
,
C.
,
Calbó
,
J.
,
Blasco
,
E.
, et al.
(
2014
).
Neuronal glycogen synthesis contributes to physiological aging
.
Aging Cell
13
,
935
-
945
.
St Clair
,
S. L.
,
Li
,
H.
,
Ashraf
,
U.
,
Karty
,
J. A.
and
Tennessen
,
J. M.
(
2017
).
Metabolomic analysis reveals that the Drosophila melanogaster gene lysine influences diverse aspects of metabolism
.
Genetics
207
,
1255
-
1261
.
Stapleton
,
D.
,
Nelson
,
C.
,
Parsawar
,
K.
,
Flores-Opazo
,
M.
,
McClain
,
D.
and
Parker
,
G.
(
2013
).
The 3T3-L1 adipocyte glycogen proteome
.
Proteome Sci.
11
,
11
.
Tennessen
,
J. M.
,
Baker
,
K. D.
,
Lam
,
G.
,
Evans
,
J.
and
Thummel
,
C. S.
(
2011
).
The Drosophila estrogen-related receptor directs a metabolic switch that supports developmental growth
.
Cell Metab.
13
,
139
-
148
.
Tennessen
,
J. M.
,
Bertagnolli
,
N. M.
,
Evans
,
J.
,
Sieber
,
M. H.
,
Cox
,
J.
and
Thummel
,
C. S.
(
2014
).
Coordinated metabolic transitions during Drosophila embryogenesis and the onset of aerobic glycolysis
.
G3 (Bethesda)
4
,
839
-
850
.
Tick
,
G.
,
Cserpán
,
I.
,
Dombrádi
,
V.
,
Mechler
,
B. M.
,
Török
,
I.
and
Kiss
,
I.
(
1999
).
Structural and functional characterization of the Drosophila glycogen phosphorylase gene
.
Biochem. Biophys. Res. Commun.
257
,
34
-
43
.
Wigglesworth
,
V. B.
(
1949
).
The utilization of reserve substances in Drosophila during flight
.
J. Exp. Biol.
26
,
150
-
163
.
Wirtz-Peitz
,
F.
,
Nishimura
,
T.
and
Knoblich
,
J. A.
(
2008
).
Linking cell cycle to asymmetric division: Aurora-A phosphorylates the Par complex to regulate Numb localization
.
Cell
135
,
161
-
173
.
Worley
,
B.
and
Powers
,
R.
(
2013
).
Multivariate Analysis in Metabolomics
.
Curr. Metabolomics.
1
,
92
-
107
.
Xia
,
J.
,
Psychogios
,
N.
,
Young
,
N.
and
Wishart
,
D. S.
(
2009
).
MetaboAnalyst: a web server for metabolomic data analysis and interpretation
.
Nucleic Acids Res.
37
,
W652
-
W660
.
Yamada
,
T.
,
Habara
,
O.
,
Kubo
,
H.
and
Nishimura
,
T.
(
2018
).
Fat body glycogen serves as a metabolic safeguard for the maintenance of sugar levels in Drosophila
.
Development
145
,
dev158865
.
Yasugi
,
T.
,
Yamada
,
T.
and
Nishimura
,
T.
(
2017
).
Adaptation to dietary conditions by trehalose metabolism in Drosophila
.
Sci. Rep.
7
,
1619
.
Yoshida
,
M.
,
Matsuda
,
H.
,
Kubo
,
H.
and
Nishimura
,
T.
(
2016
).
Molecular characterization of Tps1 and Treh genes in Drosophila and their role in body water homeostasis
.
Sci. Rep.
6
,
30582
.
Zirin
,
J.
,
Nieuwenhuis
,
J.
and
Perrimon
,
N.
(
2013
).
Role of autophagy in glycogen breakdown and its relevance to chloroquine myopathy
.
PLoS Biol.
11
,
e1001708
.

Competing interests

The authors declare no competing or financial interests.

Supplementary information