
==== Front
Proc Natl Acad Sci U S A
Proc Natl Acad Sci U S A
PNAS
Proceedings of the National Academy of Sciences of the United States of America
0027-8424
1091-6490
National Academy of Sciences

38252824
202319475
10.1073/pnas.2319475121
research-articleResearch ArticlephysioPhysiology427
Biological Sciences
Physiology
miR-137 regulates PTP61F, affecting insulin signaling, metabolic homeostasis, and starvation resistance in Drosophila
Saedi Hana a https://orcid.org/0009-0000-8221-7593

Waro Girma a
Giacchetta Lea a
Tsunoda Susan susan.tsunoda@colostate.edu
a 1 https://orcid.org/0000-0002-5742-8644

aDepartment of Biomedical Sciences, Colorado State University, Fort Collins, CO 80523
1To whom correspondence may be addressed. Email: susan.tsunoda@colostate.edu.
Edited by Nancy Bonini, University of Pennsylvania, Philadelphia, PA; received November 7, 2023; accepted December 13, 2023

22 1 2024
30 1 2024
22 7 2024
121 5 e231947512107 11 2023
13 12 2023
Copyright © 2024 the Author(s). Published by PNAS.
2024
https://creativecommons.org/licenses/by-nc-nd/4.0/ This article is distributed under Creative Commons Attribution-NonCommercial-NoDerivatives License 4.0 (CC BY-NC-ND).

Significance

Energy homeostasis involves multiple hormone-induced pathways, including the insulin signaling pathway, that coordinate nutrient availability with metabolism. Using Drosophila, we show that the brain-enriched microRNA, miR-137, is a strong regulator of insulin signaling, affecting the organism’s overall physiology and survival under conditions of nutrient deprivation. Flies lacking miR-137 display drastically reduced activation of the insulin receptor, increased body weight, enhanced levels of triglycerides, decreased activity, and prolonged survival under nutrient stress. We identify the phosphatase, PTP61F, as a likely target of miR-137, and present evidence that the misregulation of PTP61F in miR-137 mutants underlies the altered metabolic homeostasis and increased starvation resistance of miR-137 mutants. Our studies identify miR-137 as an important regulator of insulin signaling and energy homeostasis.

miR-137 is a highly conserved brain-enriched microRNA (miRNA) that has been associated with neuronal function and proliferation. Here, we show that Drosophila miR-137 null mutants display increased body weight with enhanced triglyceride content and decreased locomotor activity. In addition, when challenged by nutrient deprivation, miR-137 mutants exhibit reduced motivation to feed and prolonged survival. We show through genetic epistasis and rescue experiments that this starvation resistance is due to a disruption in insulin signaling. Our studies further show that miR-137 null mutants exhibit a drastic reduction in levels of the phosphorylated/activated insulin receptor, InR (InR-P). We investigated if this is due to the predicted miR-137 target, Protein Tyrosine Phosphatase 61F (PTP61F), ortholog of mammalian TC-PTP/PTP1B, which are known to dephosphorylate InR-P. Indeed, levels of an endogenously tagged GFP-PTP61F are significantly elevated in miR-137 null mutants, and we show that overexpression of PTP61F alone is sufficient to mimic many of the metabolic phenotypes of miR-137 mutants. Finally, we knocked-down elevated levels of PTP61F in the miR-137 null mutant background and show that this rescues levels of InR-P, restores normal body weight and triglyceride content, starvation sensitivity, as well as attenuates locomotor and starvation-induced feeding defects. Our study supports a model in which miR-137 is critical for dampening levels of PTP61F, thereby maintaining normal insulin signaling and energy homeostasis.

Drosophila
metabolism
miR-137
Colorado State University (CSU) 100007235 N/A Hana SaediGirma WaroLea GiacchettaSusan Tsunoda
==== Body
pmcMicroRNAs (miRNAs) are small (~20 to 24 nucleotides), endogenous, noncoding RNAs that typically pair with specific sites in the 3′ untranslated region of mRNA targets, typically resulting in mRNA destabilization and/or translational repression (1–3). A single miRNA may regulate a multitude of targets in different cells and/or under different conditions. The misregulation or loss of a single miRNA may therefore have a variety of consequences. In this study, we focus on the highly conserved and brain-enriched miR-137 (4–10), which has procured a lot of attention for its association with schizophrenia (SCZ) and other neuropsychiatric disorders. For example, miR-137 has been genetically linked to multiple brain disorders, including SCZ (11–14), autism spectrum disorder (11, 15), bipolar disorder (16), and intellectual disability (8, 9). Likely related to these associations, studies have shown that altered levels of miR-137 result in defects in neuronal differentiation (6, 10, 17–19), proliferation (6, 17), maturation and development (5, 9, 20), as well as synapse formation (21–25), synaptic transmission (23, 26), and learning and memory function (24–26). Studies have also explored the role and potential therapeutic function of miR-137 as a tumor suppressor in multiple forms of cancer (27–32). For example, in glioblastoma multiforme, miR-137 was found to be down-regulated (10), while overexpression of miR-137 in glioma cell lines has been shown to inhibit cell proliferation (33, 34). miR-137 has also been shown to induce apoptosis through targets involved in multiple pathways of melanoma development and progression (35, 36).

In this study, we reveal a role for miR-137 in metabolic homeostasis in Drosophila. Drosophila has proven to be a valuable model for studying obesity, energy metabolism, and diabetes because organs, signaling pathways, and proteins involved are analogous to those in humans, and most disease-associated genes are conserved between Drosophila and humans (reviewed in refs. 37–39). Insulin signaling in Drosophila, like in mammalian systems, regulates energy availability and storage. And misregulation of insulin/PI3K/AKT/FOXO signaling has been shown to affect body weight (40–43), circulating sugar levels and triglyceride storage (44–47), as well as survival under conditions of nutrient deprivation (48–59). Here, we show that miR-137 null mutants exhibit increased body weight, with abnormally high levels of triglycerides, and extended survival under starvation conditions. We show that miR-137 regulates metabolic homeostasis and consequent starvation resistance (SR) via its interaction with the insulin signaling pathway in the nervous system. Strikingly, we show that levels of the activated/phosphorylated insulin receptor (InR-P) are drastically reduced in miR-137 mutants. We provide evidence that the predicted target, Protein Tyrosine Phosphatase-61F (PTP61F), which is known to dephosphorylate InR (60–62), is regulated by miR-137 in vivo, thereby affecting activation of InR, metabolic homeostasis, and starvation sensitivity.

Results

miR-137 Null Mutants Exhibit Prolonged Survival Under Starvation Conditions.

In wild-type (WT) flies, we found miR-137 to be enriched in brains and stably expressed across gender, age, and with different kinds of stress (SI Appendix, Fig. S1). To gain insight into the physiological role of miR-137, we generated a null mutant, miR-137CR, using a CRISPR/Cas9 approach. We also obtained a second null mutant line, miR-137KO, generated by homologous recombination (63). The use of two independently generated null mutants and the ability to also examine transheterozygotes ensures that phenotype(s) identified are not a consequence of genetic background. Homozygous miR-137CR and miR-137KO mutants were both validated to exhibit undetectable levels of miR-137 by qRT-PCR (Fig. 1A) and found to be homozygous viable. To test for survival under nutrient stress, populations of newly eclosed WT and miR-137CRand miR-137KO mutants were collected, aged 7 d in a mixed population to avoid any SR due to the reported persistence of the larval fat body in the young adult (64), and then subjected to protein, sugar, or total starvation. In all cases, miR-137CR and miR-137KO males and females exhibited extended survival when compared to the WT background control line in which miR-137CR was generated (Fig. 1 and SI Appendix, Fig. S2 A–C); this is referred to as SR. Since total starvation resulted in the most robust SR phenotype, with miR-137 mutants exhibiting a median survival twice that of WT flies (median survival of miR-137CR and miR-137KO, 84 h and 72 h, respectively; median survival of wild type, 36 h; Fig. 1B), we continued all following studies under conditions of total starvation, using both miR-137CR and miR-137KO homozygotes, and/or miR-137CR/KO transheterozygotes, which similarly exhibited SR (median survival of 96 h; Fig. 1B). Interestingly, miR-137CR/+ and miR-137KO/+ heterozygotes exhibited a semidominant phenotype with median survival times of 48 h under starvation conditions (Fig. 1B). The prolonged survival of miR-137 mutants was specific to starvation conditions since longevity under fed conditions was not enhanced (SI Appendix, Fig. S2D).

Fig. 1. Loss of miR-137 results in increased resistance to starvation. (A) Quantification of the relative expression of the mature miR-137 by RT-qPCR in WT and miR-137CR and miR-137KO mutant heads. miR-137 levels are normalized to levels of S2 rRNA from the same samples (N = 5 to 10), and 2−ΔΔct values are shown as representations of relative fold-changes compared to WT. Pairwise comparisons were made to WT using the Student’s t test. *P < 0.05, error bars indicate SEM. (B) Representative survival curves are shown for male WT (black), miR-137KO (blue) homozygotes, miR-137CR(red) homozygotes, and miR-137CR/KO transheterozygotes (dark red), and miR-137CR/+ and miR-137KO/+ heterozygotes (green and blue lines, as indicated) and under total starvation conditions. Transheterozygote miR-137CR/KO, similar to miR-137KO and miR-137CR homozygotes, showed significantly prolonged survival compared to WT (N = 100 for each genotype; log-rank test: P < 0.0001 for each homozygote and heterozygote line compared with WT). (C and D) Male flies in which elav-Gal4 transgene was used to drive UAS-miR-137 expression in miR-137CR(C) or miR-137KO (D) mutant backgrounds (miR-137CR+elav>miR-137 and miR-137KO+elav>miR-137 lines, as indicated) were compared with elav-Gal4 and miR-137 mutant parental lines under conditions of total starvation. miR-137CR+elav>miR-137 and miR-137KO+elav>miR-137 flies exhibited a significant leftward shift in their survival sensitivity compared with parental miR-137 mutant background lines (log-rank test: P < 0.0001 for each miR-137 mutant line compared to corresponding miR-137CR+elav>miR-137 or miR-137KO+elav>miR-137 line; N = 100 for each genotype), indicating a rescue of SR with the reintroduction of miR-137 expression driven by elav-Gal4. (E and F) Conditional overexpression of miR-137 was achieved using elav-Gal4 driving UAS-miR-137 in combination with tub-Gal80ts (elav-Gal4-Gal80ts>miR-137); these flies were compared with parental lines, UAS-miR-137 and elav-Gal4; tub-Gal80ts (elav-Gal4-Gal80). Flies were raised at 18 °C for normal development and then shifted to 30 °C for 7 d to overexpress miR-137. qRT-PCR confirming overexpression of miR-137 in the elav-Gal4-Gal80ts>miR-137 line (Left; N = 5 independent RNA samples), and survival curves (Right) of males under total starvation conditions are shown. The conditional overexpression of miR-137 in the CNS after development significantly increased the sensitivity to starvation compared to the parental controls (N = 100 flies per genotype; log-rank test: P < 0.0001 for elav-Gal4-Gal80ts>miR-137 compared to elav-Gal4; tub-Gal80ts or UAS-miR-137).

To further confirm that the SR phenotype of miR-137 mutants is due to the loss of miR-137, we tested whether reintroduction of miR-137 expression would rescue the SR phenotype. Since miR-137 is enriched in Drosophila heads (SI Appendix, Fig. S1A) and mammalian studies have also found miR-137 to exhibit widespread expression in the brain (4–10), we used the pan-neuronal elav-Gal4 transgene to drive expression of miR-137 under the control of the relevant yeast upstream activating sequence (UAS). Indeed, pan-neuronal expression of UAS-miR-137 in miR-137CR or miR-137KO backgrounds was sufficient to restore normal survivorship under starvation conditions when compared with the elav-Gal4 background control line (Fig. 1 C and D). It is unlikely that the increase in starvation sensitivity is due to mis/overexpression of miR-137 that causes general sickness since, as discussed below, elav-Gal4>UAS-miR-137 expression in miR-137 mutants also resulted in an improvement in locomotor and feeding behaviors. Thus, our results suggest that it is the loss of miR-137 expression in the nervous system that underlies the strong SR observed in these mutants.

Overexpression Of Mir-137 Results In Increased Starvation Sensitivity.

We also tested whether overexpression of miR-137 would have the opposite effect to the loss of miR-137. Since gain-of-function miR-137 mutants in mammalian systems have been shown to display developmental defects (9), we set out to conditionally overexpress miR-137 postdevelopment. We expressed miR-137 in the CNS using the elav-Gal4>UAS-miR-137 combination with ubiquitous expression of the temperature-sensitive Gal80ts protein which inhibits Gal4 activity at 18 °C and permits Gal4 activity at 30 °C. We raised elav-Gal4, tub-Gal80ts>UAS-miR-137 flies at 18 °C to allow for normal development, then shifted newly eclosed flies to 30 °C to overexpress miR-137 postdevelopment; overexpression of miR-137 was confirmed by qRT-PCR (Fig. 1E). We found that postdevelopmental overexpression of miR-137 significantly increased starvation sensitivity, with a median survival of 24 h, compared to genetic background controls with median survival times of 36 and 48 h at 30 °C (Fig. 1F). Together our results suggest that levels of miR-137 modulate SR.

miR-137 Mutants Exhibit Altered Energy Homeostasis.

Energy homeostasis involves the complex balancing of energy storage, energy expenditure, with nutrient consumption. The SR of miR-137 mutants suggests an alteration in energy homeostasis, and some studies have proposed that SR is enabled by enhanced energy stores (65). Interestingly, we found that miR-137 mutant flies are larger than age-matched WT flies, with a 43% enhancement in body weight (Fig. 2A). We tested levels of glucose and triglycerides, and found that miR-137 mutants displayed no change in total glucose levels, but significantly elevated levels of triglycerides levels (Fig. 2 B and C). While expression of UAS-miR-137 driven by elav-Gal4 in miR-137 mutants did not significantly reduce body weight, it did rescue the enhancement in triglyceride levels back to WT levels (SI Appendix, Fig. S3 A and B). Interestingly, body weight and triglyceride levels of newly eclosed flies were also enhanced in miR-137 mutants (SI Appendix, Fig. S4), suggesting that increased energy storage begins earlier in development; future studies will need to test whether the roles of miR-137 during development and postdevelopment are similar.

Fig. 2. miR-137 mutants exhibit altered energy homeostasis. (A–C) WT and miR-137CR/KO male flies were aged for 7 d in a mixed population on a standard fly food diet and then assayed for (A) average body weight (N = 10 assays of 15 flies/assay); (B) relative triglyceride levels, normalized to protein content (N = 7 to 10 samples of 5 flies/sample; each sample run in triplicate); and (C) relative glucose levels (N = 7 to 10 samples of 5 flies/sample; each sample run in triplicate), normalized to protein content. (D) Locomotor performance of 7-d-old WT and miR-137CR and miR-137KO males scored in countercurrent climbing assays, following 18-h exposure to standard food (fed) or total starvation conditions (starved). Ten assays of 30 to 35 flies/assay were performed per genotype. (E) Seven-day-old WT and miR-137CR/KO flies were fed or starved for 18 h prior to exposure to food containing a blue dye for 15 min. Shown are representative flies with blue coloring seen in their abdomens (Left) and quantification (Right; absorbance read at 625 nm) of fly homogenates (N = 8 to 10 samples of 15 flies/sample). *Indicates significant differences (P < 0.05) between the miR-137CR/KO and WT, or indicated conditions; data analyzed Student’s t test. Error bars indicate SEM.

To examine whether there was any difference in the general activity of miR-137 null mutants, we compared miR-137CR and miR-137KO null mutants in standard locomotor activity assays with the WT genetic background control line in which miR-137CR was generated. Newly eclosed flies were aged for 7 d, then subjected to countercurrent climbing assays in groups of 30 to 35; flies were scored for their ability to climb against gravity through 10 successive tubes. We found that both miR-137CR and miR-137KO mutants exhibited significantly reduced locomotor performance whether they were fed or starved for 18 h prior to testing (Fig. 2D), and expression of elav-Gal4>UAS-miR-137 mutants rescued locomotor activity (SI Appendix, Fig. S3D). These results suggest that miR-137 mutants are generally less active and may have reduced energy expenditure.

We also examined whether miR-137 mutants exhibit any differences in acute food consumption or motivated feeding behavior. We aged miR-137CR/KO transheterozygote and WT flies on standard fly food for 7 d, then either kept them fed on standard food, or starved them for 18 h, before transferring them to 1% agar media containing 5% sucrose and 2.5% erioglaucine, an acid aminotriphenylmethane dye with a blue color that has been used to track the feeding of flies (65). After 15 min, flies were homogenized and tested for the intensity of blue dye content (absorbance at 625 nm) as a measure of food consumption. We found that WT flies exhibited a dramatic increase in feeding following starvation (Fig. 2E), as expected from previous reports (65). While miR-137CR/KO mutants and WT flies displayed feeding that was relatively similar in the fed state, miR-137CR/KO mutants exhibited a lack of starvation-induced feeding (Fig. 2E). This phenotype was rescued with expression of elav-Gal4>UAS-miR-137 in the nervous system (SI Appendix, Fig. S3E). Together, our results suggest that the loss of miR-137 results in increased lipid stores, generally decreased energy expenditure, and reduced motivation to feed after acute starvation. These findings suggest that the loss of miR-137 affects overall metabolic homeostasis.

Genetic Epistasis Suggests that miR-137 Interacts with the Insulin Signaling Pathway to Affect SR.

Given the observed SR and altered levels of triglycerides in miR-137 null mutants, we hypothesized that miR-137 was likely to interact with either the glucagon-like adipokinetic hormone (Akh) and/or insulin signaling pathway(s). Both of these pathways are known to play important roles in coordinating nutrient availability with metabolism, and disruption of either pathway has been shown to result in SR similar to miR-137 null mutants. To test for the involvement of miR-137 in either of these signaling pathways, we employed a genetic epistasis approach using SR as our readout assay. We first used the null mutant Akh1 to disrupt the Akh signaling pathway (66). Akh1 mutants exhibited SR, as expected (Fig. 3A). We then generated double-mutants in which miR-137CR or miR-137  KO mutations were combined with the Akh1 mutation. If double-mutants exhibit an “additive” phenotype with SR increased beyond either single mutant, this would suggest that miR-137 and Akh act in different pathways, however, if double-mutants exhibit SR similar to the single miR-137 mutants, this would suggest that miR-137 and Akh act in the same pathway. We found that the survival of lines containing either the miR-137CR or miR-137KO mutation with the Akh1 mutation exhibited SR that was significantly enhanced beyond that of either single mutant (Fig. 3A), suggesting that the function of miR-137 in energy homeostasis is likely to be independent of the Akh pathway.

Fig. 3. Genetic epistasis analyses indicate that SR of miR-137 involves the insulin signaling pathway. Survival curves of single mutations, or transgenes for specific gene DN or RNAi knock-down (RNAi), are shown for components of the Akh (A) and insulin (B–E) pathways under conditions of total starvation. Survival curves of these mutations/driven DN/RNAi lines combined with either the miR-137CR (Left column) or miR-137KO (Right column) mutations are shown to evaluate genetic epistasis; miR-137CR and miR-137KO backgrounds are reshown here in each graph for comparison. (A) Akh1 mutants (green) exhibit a significantly extended lifespan, similar to miR-137CR and miR-137KO (blue), when compared to wild type (WT) (log-rank test: P < 0.0001 for pairwise comparisons between WT and each of Akh1, miR-137CR, and miR-137KO survival curves; N = 100 flies per genotype). miR-137CR; Akh1 (miR-137CR+Akh1) and miR-137KO; Akh1 (miR-137KO+Akh1) double mutants (red lines) show a lifespan extended significantly beyond miR-137CR, miR-137KO, and Akh1 mutants (N = 100 per genotype; log-rank tests: P < 0.0001 for each pairwise comparison between each double mutant to Akh1 or corresponding miR-137 mutant background). For (B–E) the parental elav-Gal4 line is used as a control, and lines in which elav-Gal4 is used to drive expression of (B) UAS-DN InRDN subunit (elav>InRDN), (C) UAS-InR-RNAi (elav>InR-RNAi), (D) UAS-DN PI3KDN subunit (elav>PI3KDN), and (E) UAS-AKT-RNAi (elav>AKT-RNAi) are shown for survival under total starvation conditions; miR-137CR and miR-137KO lines are shown again for reference. In all cases (B–E), the lines containing knock-down of components of the insulin signaling pathway in the nervous system all showed significant extension of lifespan (SR) when compared to the elav-Gal4 parental line (N = 100 flies/genotype; log-rank test, P < 0.0001 for pairwise comparisons between elav-Gal4 and each of: elav>InRDN, elav>InR-RNAi, elav>PI3KDN, elav>AKT-RNAi). However, when these knock-downs were combined with miR-137CR or miR-137KO, survival curves were shifted leftward and not “additive”; there was no further extension of survivorship when miR-137CR or miR-137KO mutations were combined with elav>InRDN, elav>InR-RNAi, elav>PI3KDN, or elav>AKT-RNAi when compared to miR-137CR or miR-137KO lines, or knock-downs alone (N = 97 to 112 flies/genotype; log-rank tests: P > 0.05 for all comparisons, and median survivorship for miR-137CR or miR-137KO combined with elav>InRDN, elav>InR-RNAi, elav>PI3KDN, or elav>AKT-RNAi were not greater than corresponding miR-137CR or miR-137KO, or elav>InRDN, elav>InR-RNAi, elav>PI3KDN, elav>AKT-RNAi).

To test whether miR-137 might act via the insulin signaling pathway in similar, formally defined, epistasis experiments, we needed an InR null mutant. Since Drosophila InR null mutants are lethal (42), we opted to use lines in which InR was knocked-down; although not perfectly ideal, we similarly did this for downstream signaling components. We began with genetic epistasis tests using the pan-neuronally expressed dominant-negative (DN) and RNAi transgenes that target the insulin receptor InR. We used the elav-Gal4 transgene to drive expression of UAS-InRDN or UAS-InR-RNAi; compared to genetic background control lines, neuronal expression of InRDN or InR-RNAi was sufficient to result in SR, similar to miR-137 null mutants (Fig. 3 B and C). We then combined elav-Gal4>UAS-InRDN and elav-Gal4>UAS-InR-RNAi lines with either the miR-137CR or miR-137KO mutation and tested for SR. If SR is extended beyond either single mutant/transgenic, indicating an “additive phenotype,” this would suggest that miR-137 acts in a pathway different from the insulin pathway; conversely, if the SR is not extended beyond either single mutant/transgenic, this would suggest that miR-137 acts in the same pathway as the insulin pathway. We found that the SR of these combined lines was not significantly enhanced beyond that of parental lines with disrupted insulin signaling alone (Fig. 3 B and C), suggesting a genetic interaction between miR-137 and the insulin signaling pathway. We also tested DN and RNAi lines for the downstream effector phosphoinositide-3-kinase PI3K, and the downstream protein kinase B, AKT. We similarly found that neuronal expression of PI3KDN or AKT-RNAi alone resulted in SR that was not enhanced with either miR-137 null mutant (Fig. 3 D and E). Together, our epistasis results suggest that the SR of miR-137 mutants is due to its interaction with the insulin signaling pathway in the nervous system.

SR of miR-137 Mutants Is Rescued by Activation of the Insulin Signaling Pathway.

We next reasoned that if the SR phenotype of miR-137 null mutants is due to a disruption of insulin signaling, then we should be able to rescue SR by activating the insulin signaling pathway in miR-137 mutants. To test this, we used genetic tools to activate insulin signaling in the nervous systems of miR-137CR and miR-137KO null mutants. We reasoned that overexpression of the Drosophila insulin-like peptide ligands Dilp2 or Dilp5, which are both produced in the brain, would potentially increase activation of InR. We first used elav-Gal4 to induce expression of UAS driven transgenes of Dilp2 and Dilp5. In miR-137CR and miR-137KO mutant backgrounds, expression of either Dilp2 or Dilp5 was sufficient to generate a significant leftward shift in survival curves, increasing starvation sensitivity closer to those of the elav-Gal4 background control line (Fig. 4 A and B). Since Dilp2/5 are normally secreted from insulin-producing cells (IPCs) in the brain, we also drove UAS-Dilp2 with a Dilp2-Gal4 transgene. This was genetically more complicated since the Dilp2-Gal4 transgene is on the second chromosome, like miR-137. To do this, we used the miR-137/+ heterozygous background, which exhibits a significant, albeit more modest increase in SR (Fig. 1B). The Dilp2-Gal4/+ heterozygote background alone exhibited an increased SR compared to UAS-Dilp2/+ flies, nevertheless, when Dilp2-Gal4/+ was used to drive UAS-Dilp2/+ expression, in both miR-137CR and miR-137KO backgrounds, we observed a significant shift in survival curves to the left, restoring more normal starvation sensitivity to flies (SI Appendix, Fig. S5A).

Fig. 4. SR of miR-137 is reversed by activation of the insulin signaling pathway. Representative survival curves for genotypes overexpressing components of the insulin signaling pathway in the miR-137CR (Left column) or miR-137KO (Right column) backgrounds under conditions of total starvation. Survival curves for the parental elav-Gal4 (black) and miR-137CR or miR-137KO (red) lines are shown on each graph as background controls. elav-Gal4 was used to drive neuronal expression of insulin signaling components in a WT background (green) or in miR-137CR or miR-137KO backgrounds (blue). Insulin signaling components expressed from transgenes include UAS-Dilp2 (A); UAS-Dilp5 (B); and a constitutively active InRCA subunit (C). (A) When elav-Gal4>UAS-Dilp2 was expressed in either a miR-137CR or miR-137KO backgrounds, survival curves (miR-137CR+elav>Dilp2 or miR-137KO+elav>Dilp2) were significantly shifted leftward from miR-137CR and miR-137KO backgrounds alone (N = 90 to 100 flies per genotype; log-rank test: pairwise comparisons miR-137CR+elav>Dilp2 or miR-137KO+elav>Dilp2, with miR-137CR or miR-137KO, respectively, yielded significant differences, P < 0.0001). (B) elav-Gal4>UAS-Dilp5 flies exhibited significantly increased SR compared to elav-Gal4 flies (N = 100 flies per genotype; log-rank test, P < 0.001). When elav-Gal4>UAS-Dilp5 was combined with miR-137CR or miR-137KO mutations, survival curves (miR-137CR+elav>Dilp5 or miR-137KO+elav>Dilp5) were significantly shifted leftward from miR-137CR and miR-137KO backgrounds alone (N = 100 flies per genotype; log-rank test: pairwise comparisons miR-137CR+elav>Dilp5 or miR-137KO+elav>Dilp5, with miR-137CR or miR-137KO, respectively, yielded significant differences, P < 0.0001). (C) elav-Gal4>UAS-InRCA flies exhibited significantly increased starvation sensitivity compared to elav-Gal4 flies (N = 100 flies per genotype; log-rank test, P < 0.0001). When elav-Gal4>UAS-InRCA was combined with miR-137CR or miR-137KO mutations, survival curves (miR-137CR+elav> InRCA or miR-137KO+elav> InRCA) were significantly shifted leftward from miR-137CR and miR-137KO backgrounds alone (N = 100 flies per genotype; log-rank test: pairwise comparisons miR-137CR+elav> InRCA or miR-137KO+elav> InRCA, with miR-137CR or miR-137KO, respectively, yielded significant differences, P < 0.0001), as well as significantly shifted leftward from elav-Gal4>UAS-InRCA lines (N = 100 flies per genotype; log-rank test: pairwise comparison of elav>UAS-InRCA with each of miR-137CR+elav> InRCA or miR-137KO+elav> InRCA were significant with P = 0.0004).

We similarly expressed a constitutively active form of InR (InRCA) to more directly increase activation of InR in miR-137 mutant backgrounds. Expression of InRCA in the nervous system of miR-137 mutants had an especially strong effect, increasing starvation sensitivity even beyond that of control lines (Fig. 4C). We also used an InR-Gal4 transgene to drive expression of InRCA. We show that survival curves were significantly shifted leftward toward wild type in the miR-137KO background (SI Appendix, Fig. S5 B, Right); it is unclear, however, why rescue was not permissive in the miR-137CR background (SI Appendix, Fig. S5 B, Left). To further test the involvement of the insulin pathway, we also tested downstream effectors, PI3K and AKT. We found that overexpression of downstream PI3K or AKT was also able to restore starvation sensitivity to miR-137 null mutants (SI Appendix, Fig. S5 C and D). These results further support the idea that miR-137 interacts with the insulin signaling pathway to affect starvation sensitivity.

Levels of Activated/Phosphorylated InR Are Reduced in miR-137 Null Mutants.

Since activation of insulin signaling in miR-137 null mutants restored starvation sensitivity to miR-137 mutants (Fig. 4), we hypothesized that insulin signaling was likely to be defective in miR-137 mutants. Because epistasis tests indicated that the miR-137 mutants genetically interacted with the insulin signaling pathway as far up as the InR, we began by testing whether miR-137 mutants displayed any difference in levels of the activated/phosphorylated form of InR [in Drosophila Tyr1485; analogous to human Tyr1146 (67) within the kinase activation loop], referred to as InR-P. Using an antibody specific for InR-P on immunoblots of whole head homogenates, we found that levels of InR-P were near undetectable in miR-137CR and miR-137KO homozygotes, while an antibody against total InR recognized a similar size band that was no different between WT and miR-137 mutants (Fig. 5 A and B). These results suggest that very little of the InR protein in miR-137 mutants is phosphorylated and that InR activation is severely impaired in these mutant lines. When UAS-miR-137 is expressed, using the elav-Gal4 driver, we show that relative InR-P levels are increased to a small but significant degree (SI Appendix, Fig. S3C). Consistent with the semidominant SR phenotype we observed for miR-137CR/+ and miR-137KO/+ heterozygotes (Fig. 1B), we observed a semidominant reduction in InR-P levels in miR-137CR/+ and miR-137KO/+ heterozygotes that was about halfway between WT and miR-137 null mutants (Fig. 5B). To further test whether insulin receptor signaling is impaired in miR-137 mutants, we also tested for levels of the activated/phosphorylated form of AKT (in Drosophila, Ser505; analogous to human Ser473), referred to as AKT-P. We indeed observed a significant reduction in Akt-P in miR-137CR/KO mutants with no difference in total Akt (SI Appendix, Fig. S6 A and B); since AKT can be activated by multiple signaling pathways, it was not surprising that the reduction in AKT-P was not as severe as we observed for InR-P. We next explored how the absence of miR-137 results in a loss of phosphorylated InR.

Fig. 5. miR-137 null mutants exhibit reduced phosphorylation of InR likely due to increased levels of PTP61F. (A and B) Quantification and representative immunoblots of relative levels of total InR (A) and InR-P (B) from heads of 7-d-old WT, miR-137CR (CR), and miR-137KO (KO) homozygotes, as indicated. Anti-InR and anti-InR-P levels were normalized to anti-actin, which was used as a loading control (N = 6 to 30 samples per genotype; 5 heads/sample); relative normalized levels are shown. (C) Representative immunoblots and quantification of an endogenously GFP-tagged PTP61F expressed in WT (Ctrl) and miR-137CR mutant backgrounds. Two independent GFP-PTP61F lines were used and shown here and in SI Appendix, Fig. S6; the line used here is GFSTF-PTP61FMI03132. Anti-GFP signals were normalized to anti-actin, which was used as a loading control (N = 6 to 30 samples per genotype; 5 heads/sample); relative normalized levels are shown. (D–I) Overexpression of UAS-PTP61F was driven pan-neuronally with the elav-Gal4 transgene (elav>PTP61F) in a WT background. (D) Quantification and representative immunoblots of relative levels of total InR-P from heads of 7-d-old WT and elav>PTP61F. Anti-InR-P levels are normalized to anti-actin, which was used as a loading control (N = 6 to 30 samples per genotype; 5 heads/sample); relative normalized levels are shown. (E) Survival curves from elav-Gal4, UAS-PTP61F, and elav>PTP61F lines shown. Survivorship of elav>PTP61F flies was significantly extended under total starvation conditions compared to parental controls (N = 100 to 110 flies per genotype; log-rank test: elav>PTP61F was pairwise compared to elav-Gal4 and UAS-PTP61F, each test resulted in P < 0.0001). (F) Body weight was compared between WT and elav>PTP61F (N = 10 samples, 15 flies/sample). elav>PTP61F flies displayed a significantly increased body weight compared to WT. (G) Relative triglyceride levels, normalized to body weight per sample (N = 15 to 16 samples of 5 flies/sample; each sample run in triplicate); elav>PTP61F flies had significantly enhanced triglyceride content compared to WT. (H) Locomotor performance of 7-d-old WT and elav>PTP61F males scored in countercurrent climbing assays. Surprisingly, elav>PTP61F flies displayed enhanced locomotion compared with WT; 10 assays of 30 to 35 flies/assay were performed per genotype. (I) Seven-day-old WT and elav>PTP61F flies were fed or starved for 18 h prior to exposure to food containing a blue dye for 15 min. Quantification shown of normalized absorbance read at 625 nm from fly homogenates (N = 9 to 10 samples of 15 flies/sample). For all, *indicates significance by P < 0.05, Student’s t test. Error bars indicate SEM.

PTP61F Is Negatively Regulated by miR-137, and Overexpression of PTP61F Alone Results in Reduced Levels of InR-P, SR, Enhanced Body Weight, and Triglyceride Content.

We hypothesized that there must be a target of miR-137 that affects the phosphorylation of InR. We searched for predicted targets of miR-137 with TargetScan (http://targetscan.org), a well-established prediction algorithm that uses stringent seed pairing as a criterion across multiple genomes (3, 68, 69). Interestingly, one of the predicted targets of miR-137 is the non-receptor PTP61F, the Drosophila homolog of mammalian TC-PTP and PTP-1B, which are known across species to negatively regulate insulin signaling by dephosphorylating InR (60–62, 70–72). Since most miRNAs serve to dampen expression of target mRNAs, we reasoned that if PTP61F is a canonical target of miR-137, then miR-137 null mutants would exhibit elevated levels of PTP61F, which may in turn be responsible for the hyper-dephosphorylation of InR observed in miR-137 mutants (Fig. 5B). To test whether levels of PTP61F are indeed elevated in miR-137 null mutants, we used two independently generated fly lines in which an artificial exon, containing the coding sequence for GFP between splice acceptor and splice donor sequences (-SA-GFP-SD-) was inserted into an intron within the coding sequence of PTP61F such that GFP would be in-frame and fused to PTP61F (73). The resultant protein expressed in these GFP-PTP61F flies was ~68 kD, consistent with the expression of a GFP-PTP61F fusion protein. We then examined whether levels of GFP-PTP61F were elevated in miR-137 mutant backgrounds. We found that levels of GFP-PTP61F, in each of the two GFP-PTP61F insertion lines, were indeed significantly elevated in the absence of miR-137 (Fig. 5C and SI Appendix, Fig. S6 C and D). While these data do not confirm a direct regulatory interaction, they do suggest that PTP61F is negatively regulated by miR-137 in vivo.

Since the loss of miR-137 results in increased levels of PTP61F, we next wanted to investigate whether these elevated PTP61F levels are likely to be responsible for the near lack of phosphorylated InR and SR seen in miR-137 mutants. We used elav-Gal4 to overexpress UAS-PTP61F in a WT background and compared these flies to genetic background control lines. We found that elav-Gal4>UAS-PTP61F flies exhibited a drastic reduction in InR-P (Fig. 5D), enhanced SR (Fig. 5E), and increased body weight (Fig. 5F) and triglyceride levels (Fig. 5G), similar to miR-137 null mutants. Interestingly, the overexpression of PTP61F alone did not negatively affect locomotor or feeding behaviors (Fig. 5 H and I). This may indicate that miR-137 regulation of other targets also contributes to these more complex phenotypes. Together, our results suggest that PTP61F is likely regulated by miR-137 and that misregulation of PTP61F in miR-137 mutants leads to effects on InR activation, survivorship under nutrient deprivation, body weight and composition.

Knock-Down of PTP61F in miR-137 Mutants Ameliorates miR-137 Effects on Metabolic Homeostasis and Locomotor and Feeding Behaviors.

Our next goal was to mitigate the upregulation of PTP61F in miR-137 mutants by RNAi knock-down of PTP61F and test whether this would then restore a near-normal phenotype to miR-137 mutants. We used Gal4 transgenes to drive expression of UAS-PTP61F-RNAi. We first surveyed whether ubiquitous expression using armadillo-Gal4 (arm-Gal4), fat body expression using r4-Gal4, and/or neuronal expression using elav-Gal4 would rescue SR and body weight enhancement in miR-137 mutants. We found that arm-Gal4 and r4-Gal4 transgenes driving expression of UAS-PTP-RNAi resulted in the restoration of starvation sensitivity to miR-137 mutants (SI Appendix, Fig. S7), while enhanced body weight of miR-137 mutants was significantly reduced with arm-Gal4, but not with r4-Gal4 (SI Appendix, Fig. S7). Since miR-137 is brain enriched, and our other results indicated a role for miR-137 regulation of insulin signaling and metabolic homeostasis in the nervous system, we were most interested in whether knock-down of PTP61F in neurons would be sufficient to attenuate miR-137 mutant phenotypes. We used elav-Gal4 to drive expression of a UAS-PTP61F-RNAi transgene in the miR-137CR mutants, then tested for levels of InR-P, body weight, relative triglyceride content, survival under starvation conditions, as well as locomotor and feeding behaviors. We found that expression of PTP61F-RNAi was indeed able to significantly increase levels of InR-P in miR-137CR flies (Fig. 6A), suggesting that the loss of activated InR-P in miR-137 mutants is in large part due to the upregulation of PTP61F expression that results from the loss of miR-137. As a consequence, PTP61F-RNAi expression also significantly restored normal body weight and triglyceride content to miR-137CR flies (Fig. 6 B and C), and significantly increased their starvation sensitivity (Fig. 6D). Similarly, we found that knocking-down PTP61F also significantly increased locomotor and starvation-induced feeding behaviors in miR-137CR flies (Fig. 6 E and F), suggesting that elevated PTP61F levels are a contributing factor to these behavioral defects in miR-137CR mutants. We observed similar effects of elav-Gal4>UAS-PTP61F-RNAi expression in the miR-137KO and miR-137CR/KO backgrounds (SI Appendix, Figs. S8 and S9). Altogether, our data support a model in which misregulation of PTP61F expression in the nervous system significantly contributes to the altered metabolic homeostasis, increased SR, and defective locomotor and feeding behaviors observed in miR-137 mutants.

Fig. 6. Knock-down of PTP61F in miR-137CR mutants ameliorates the loss of InR-P, increased body weight and triglyceride content, and restores starvation sensitivity and more normal locomotor and feeding behaviors. We compared WT, miR-137CR, and miR-137CR lines combined with elav-Gal4 driving the expression of UAS-PTP61F-RNAi (miR-137CR+elav>PTP61F-RNAi) for all the phenotypes associated with miR-137 mutants (Figs. 1 and 2). (A) Representative immunoblots and quantification for InR-P levels in WT, miR-137CR, and miR-137CR+elav>PTP61F-RNAi samples. Anti-InR-P signals were normalized to anti-actin, which was used as a loading control (N = 9 samples per genotype; 5 heads/sample); relative normalized levels are shown. (B) Body weight was compared between WT, miR-137CR, and miR-137CR+elav>PTP61F-RNAi flies (N = 5 to 8 samples, 15 flies/sample). miR-137CR flies exhibit significantly enhanced body weight compared to WT, as expected. miR-137CR+elav>PTP61F-RNAi flies displayed a significant decrease body weight compared to miR-137CR, and body weight not significantly different (ns) from WT. (C) Relative triglyceride levels, normalized to body weight per sample (N = 7 to 11 samples of 5 flies/sample; each sample run in triplicate). miR-137CR flies exhibited significantly enhanced triglyceride content compared to WT, as expected. miR-137CR+elav>PTP61F-RNAi flies displayed a significantly decreased triglyceride content compared to miR-137CR. (D) Survival curves from elav-Gal4, miR-137CR, UAS-PTP61F-RNAi, and miR-137CR+elav>PTP61F-RNAi lines are shown. Survivorship of miR-137CR+elav>PTP61F-RNAi flies was significantly reduced compared to miR-137CR background (N = 98 to 100 flies per genotype; log-rank test: miR-137CR+elav>PTP61F-RNAi was compared to miR-137CR, P < 0.0001), and more similar to elav-Gal4 and UAS-PTP61F-RNAi parental controls. (E) Locomotor performance of 7-d-old WT, miR-137CR, and miR-137CR+elav>PTP61F-RNAi males scored in countercurrent climbing assays. Ten assays of 30 to 35 flies/assay were performed per genotype. Locomotor performance of miR-137CR+elav>PTP61F-RNAi was significantly improved compared to the miR-137CR background and not significantly different from WT. (F) Seven-day-old WT, miR-137CR, and miR-137CR+elav>PTP61F-RNAi flies were fed or starved for 18 h prior to exposure to food containing a blue dye for 15 min. Quantification shown of normalized absorbance read at 625 nm from fly homogenates (N = 7 to 17 samples of 15 flies/sample). Starvation-induced feeding of miR-137CR+elav>PTP61F-RNAi flies was significantly enhanced compared to the miR-137CR background. For all, *indicates significance by P < 0.05, Student’s t test. Error bars indicate SEM.

Discussion

Fitness and survival depend on an organism’s balance of energy storage, metabolism, adaptation to changes in nutrient availability, as well as energy expenditure. Multiple nutrient-sensing, metabolic and growth signaling pathways contribute to these homeostatic processes, including the insulin signaling pathway. In this study, several lines of evidence suggest that miR-137 regulates insulin signaling: genetic epistasis experiments with mutants in the insulin signaling pathway that yield a similar SR phenotype, experiments in which starvation sensitivity of miR-137 mutants is restored by activating insulin signaling, and finally, near undetectable levels of activated/phosphorylated InR. Disruption in insulin signaling in miR-137 mutants is likely to be a major contributor to the increased body weight, enhanced levels of triglycerides, and SR observed in these mutants. Genetic disruption of insulin signaling in both Drosophila and mammals has indeed been shown to result in changes in body weight (40–43) and elevated levels of sugars and triglycerides (44–47). miR-137 null mutants, however, did not mimic insulin signaling mutants in all ways. Body size, for example, has been shown to be reduced in chico mutants (40), while miR-137 mutants are obese. Chico mutants, and lines in which the IPCs were ablated, thereby eliminating the secretion of Dilps in the brain, however, did exhibit increased lipid content (40, 44) similar to miR-137 mutants. IPC-ablation also resulted in elevated fasting glucose levels in the hemolymph (44). Although we did not observe any difference in glucose levels, we did not test fasting levels nor restrict our collection to the hemolymph. Interestingly, mice and humans that lack insulin receptor(s) also exhibit paradoxical differences in glycemic effects (74, 75). Results may also be more complex in miR-137 mutants since miR-137 is a microRNA that has multiple targets, including some in other intersecting pathways. For example, miR-137 has been associated with growth and target of rapamycin signaling (76–79), and mammalian AKT2 has been identified as a miR-137 target (76, 80). Future studies will need to address how the effect of miR-137 is integrated across these different signaling pathways.

The extended survival of miR-137 mutants under starvation conditions is likely to be a consequence of the lowered levels of InR-P seen in these mutants. Mutations that decrease insulin/PI3K/AKT/FOXO signaling exhibit increased SR across species (48–59). Ablation of IPCs that completely disrupts insulin signaling in the brain similarly results in SR (44). SR in miR-137 mutants is likely to be bolstered by their enhanced energy stores since accumulation of lipid stores has been shown to correlate with resistance to starvation (81–83). In this study, we present evidence that altered insulin signaling in the nervous system is a significant factor contributing to SR of miR-137 mutants. It is, however, possible that miR-137 also affects insulin signaling in the fat body, which is known to regulate lipid storage. Interestingly, when we used the fat body r4-Gal4 transgene to drive expression of PTP61F-RNAi in miR-137 mutants, we found that starvation sensitivity was also restored (SI Appendix, Fig. S7), suggesting a role for insulin signaling in the fat body. Future studies will need to dissect whether miR-137 functions in particular neurons and/or in other cell types such as glial and fat body cells. We also show that miR-137 mutants exhibit decreased energy expenditure as seen in their decreased locomotor and feeding behavior, and this may also contribute to prolonged survival during starvation.

What are the mechanism(s) underlying the reduced levels of InR-P in miR-137 mutants? Using the algorithm TargetScanFly7.2 (https://www.targetscan.org/fly_72/) to gain insight into predicted targets of miR-137 in Drosophila, we identified the candidate PTP61F, which has been shown to dephosphorylate InR in Drosophila (60–62). Since miRNA regulation most commonly results in target mRNA destabilization and/or translational repression (84–86), we expected that the loss of miR-137 could lead to enhanced expression of PTP61F, which would then explain the near absence of InR-P. We present evidence here that the loss of miR-137 does indeed result in increased levels of PTP61F, that overexpression of PTP61F is sufficient to mimic the metabolic phenotypes associated with the loss of miR-137, and that knocking-down enhanced levels of PTP61F expression in miR-137 mutants partially or fully rescues all phenotypes, metabolic and behavioral. Interestingly, when we used TargetScanMouse8.0 (https://www.targetscan.org/mmu_80/) and TargetScanHuman8.0 (https://www.targetscan.org/vert_80/), we found that both mouse and human miR-137 are similarly predicted to target a mammalian ortholog of PTP61F, TC-PTP, which is encoded by the PTPN2 gene; this phosphatase has also been shown to dephosphorylate the mammalian InR (70–72), suggesting that regulation of insulin signaling by miR-137 is likely to be conserved across species. PTP61F is associated with multiple other signaling pathways (87–91) and exhibits roles in tumor suppression and regulating organ size (87). It will be interesting to investigate whether the regulation of PTP61F by miR-137 also affects these pathways and roles.

Does misregulation of miR-137 in neuropsychiatric disorders contribute to metabolic comorbidities associated with these disorders? Multiple studies have found miR-137 to be strongly linked to SCZ with genome-wide significance (11–14). In addition to neuropsychiatric symptoms, individuals with SCZ have a ~10 to 15 y reduction in life expectancy (92–94), and this high mortality rate is largely due to the high prevalence of obesity, T2D, elevated cholesterol and triglycerides, high blood pressure, heart disease, high blood glucose, and irregularities in insulin signaling (94–101). Although there are many studies that have examined metabolic syndrome associated with SCZ patients under antipsychotic treatment, there is recent and substantial evidence that drug-naive patients also exhibit metabolic disturbances (102–106) and that antipsychotics only worsen these metabolic symptoms (107–109). Interestingly, impairment in insulin signaling in the brain has been suggested to underlie the metabolic dysfunction in drug-naive SCZ patients (110–113). miR-137 is also associated with bipolar disorder (16), and individuals with bipolar disorder have the highest incidence of obesity and metabolic disorders [~45%; (114)], resulting in a reduced life expectancy (92, 114, 115), similar to SCZ patients. Thus, understanding how miR-137 regulates insulin signaling may also help explain the metabolic comorbidities that arise with SCZ and bipolar disorder, and perhaps reveal avenues of treatment.

Materials and Methods

Fly Husbandry and Strains.

Flies were reared on a standard Drosophila cornmeal-yeast-agar-based food in a controlled environment with continuous 12/12 h light/dark exposure and 65% humidity; experiments were conducted at 25 °C unless otherwise stated. Fly strains used and created are described in SI Appendix.

Quantitative RT-PCR.

For microRNA quantification, total RNA was extracted from five heads or five bodies per sample using the miRNeasy Mini Kit (Qiagen). We used TaqMan™ microRNA assays (Applied Biosystems) customized for the reverse transcription and qRT-PCR of miR-137 and small RNA controls, 2S rRNA (30 nt) and U14 (51 nt), which were used as reference genes for real-time PCR amplification. Forward and reverse primers to amplify target genes from these cDNA products were customized and included with specific probes for quantification (TaqMan™ microRNA assays; Applied Biosystems) of the mature miR-137 sequence. Procedures, primer sequences, and statistical analyses are detailed in SI Appendix.

Survival Analyses Under Starvation Conditions.

Newly eclosed flies (<24 h) were collected and raised in a mixed population of males and females on standard fly food for 6 to 7 d under uncrowded (<50 flies/large bottle) and constant conditions of 25 °C temperature and 65% humidity. Flies were transferred to fresh food every 3 to 4 d to ensure constant conditions; we further confirmed that results were similar even when living conditions during development included sparsely populated (40 larvae/tube) larvae (SI Appendix, Fig. S4). Flies were sorted into groups of 10 per vial >24 h before transfer to starvation media. Under conditions of total starvation, survival was scored every 12 h until all flies were dead. Detailed procedures and statistical analysis are given in SI Appendix.

Body Weight and Metabolite Assays.

Newly eclosed flies (<24 h) were collected from standard cultures and raised in a mixed population of males and females on standard fly food for 6 to 7 d under uncrowded (<50 flies/large bottle) and constant conditions of 25 °C temperature and 65% humidity. Flies were transferred to fresh food every 3 to 4 d to ensure constant conditions; we further confirmed that results were similar even when living conditions during development included sparsely populated (40 larvae/tube) larvae (SI Appendix, Fig. S4). Dry body weight was determined by weighing 10 groups of 15 flies for each genotype (SI Appendix).

Triglyceride quantification.

Triglyceride content was determined in a modified protocol from previously described assays (116) (SI Appendix). Each experiment was performed across at least two plates with internal controls and standard curves; quantified triglyceride content for each sample was normalized to protein or body weight of flies in each sample.

Glucose quantification.

The Glucose (GO) Assay Kit GAGO20-1KT Sigma was used to quantify the total glucose content in flies, as previously described (116). Quantification of glucose concentrations was processed and compared, as described for triglyceride quantification.

Protein quantification.

Protein content was quantified by using the Pierce™ BCA Protein Assay Kit from five adult flies per sample, as detailed in SI Appendix. Quantification of protein concentrations was processed for each sample and used for triglyceride and glucose content normalization.

Countercurrent Climbing Assays.

30 to 35 adult males, aged at 25 °C for 7 d were compared in countercurrent climbing assays as we and others have previously described (117). From resulting countercurrent distribution of flies (118), each fly was given a score of 0.5 for each tube that it climbed out of. Ten assays, each with ~30 to 35 naive flies, were performed for each genotype/condition tested.

Feeding Assays.

Fifteen adult flies (7 d old) per vial were kept free from CO2 for 24 h to avoid the negative impact of anesthesia exposure on feeding behavior. Flies were either kept on standard fly food (fed) or starved for 18 h on 1% agar (starved) before they were transferred to vials containing 1% agar, 5% sucrose, and 2.5% blue food dye (Erioglaucine Disodium Salt, Sigma, cat. # 861146) and allowed to feed for 15 min at 25 °C, as previously described (65). The absorbance of fly homogenates was read at 625 nm. Sample absorbances were normalized to the mean of the absorbance of samples from a WT starved condition performed in the same experiment.

Immunoblot Analyses.

For each sample, homogenates from 5 adult fly heads were separated on a 10% SDS polyacrylamide gel, then transferred to a 0.45 µm nitrocellulose membrane, which was incubated with primary then secondary antibodies, as detailed in SI Appendix. Recognized protein bands were visualized by chemiluminescence and quantified using ImageJ software. Quantified bands of interest were normalized to anti-actin bands from the same lanes. Secondary normalization was made to a control condition before values were combined and compared across the blots.

Supplementary Material

Appendix 01 (PDF)

Click here for additional data file.

We acknowledge the Bloomington Drosophila Stock Center for many of the Drosophila stocks used in this study (Materials and Methods), as well as FlyORF for the UAS-AKT line, and the Vienna Drosophila Resource Center for the UAS-AKT-RNAi and UAS-InR-RNAi lines. We also thank colleagues in the field for generously sharing Drosophila stocks with us. We thank Dr. Michal Zurovec and Dr. Dalibor Kodrik for the Akh1 mutant line.

Author contributions

H.S. and S.T. designed research; H.S., L.G., and S.T. performed research; H.S. and G.W. contributed new reagents/analytic tools; H.S., L.G., and S.T. analyzed data; and S.T. wrote the paper.

Competing interests

The authors declare no competing interest.

Data, Materials, and Software Availability

All study data are included in the article and/or SI Appendix.

Supporting Information

This article is a PNAS Direct Submission.
==== Refs
1 S. M. Hammond, An overview of microRNAs. Adv. Drug Deliv. Rev. 87 , 3–14 (2015).25979468
2 M. Ghildiyal, P. D. Zamore, Small silencing RNAs: An expanding universe. Nat. Rev. Genet. 10 , 94–108 (2009).19148191
3 D. P. Bartel, MicroRNAs: Target recognition and regulatory functions. Cell 136 , 215–233 (2009).19167326
4 I. Guella , Analysis of miR-137 expression and rs1625579 in dorsolateral prefrontal cortex. J. Psychiatr. Res. 47 , 1215–1221 (2013).23786914
5 J. J. Crowley , Disruption of the microRNA 137 primary transcript results in early embryonic lethality in mice. Biol. Psychiatry 77 , e5-7 (2015).25016316
6 G. Sun , miR-137 forms a regulatory loop with nuclear receptor TLX and LSD1 in neural stem cells. Nat. Commun. 2 , 529 (2011).22068596
7 K. Sakamoto, J. J. Crowley, A comprehensive review of the genetic and biological evidence supports a role for MicroRNA-137 in the etiology of schizophrenia. Am. J. Med. Genet. B Neuropsychiatr. Genet. 177 , 242–256 (2018).29442441
8 M. H. Willemsen , Chromosome 1p21.3 microdeletions comprising DPYD and MIR137 are associated with intellectual disability. J. Med. Genet. 48 , 810–818 (2011).22003227
9 R. D. Smrt , MicroRNA miR-137 regulates neuronal maturation by targeting ubiquitin ligase mind bomb-1. Stem Cells 28 , 1060–1070 (2010).20506192
10 J. Silber , miR-124 and miR-137 inhibit proliferation of glioblastoma multiforme cells and induce differentiation of brain tumor stem cells. BMC Med. 6 , 14 (2008).18577219
11 Schizophrenia Psychiatric Genome-Wide Association Study (GWAS) Consortium, Genome-wide association study identifies five new schizophrenia loci. Nat. Genet. 43 , 969–976 (2011).21926974
12 Schizophrenia Working Group of the Psychiatric Genomics Consortium, Biological insights from 108 schizophrenia-associated genetic loci. Nature 511 , 421–427 (2014).25056061
13 S. Ripke , Genome-wide association analysis identifies 13 new risk loci for schizophrenia. Nat. Genet. 45 , 1150–1159 (2013).23974872
14 Z. Li , Genome-wide association analysis identifies 30 new susceptibility loci for schizophrenia. Nat. Genet. 49 , 1576–1583 (2017).28991256
15 D. Pinto , Convergence of genes and cellular pathways dysregulated in autism spectrum disorders. Am. J. Hum. Genet. 94 , 677–694 (2014).24768552
16 J. Duan , A rare functional noncoding variant at the GWAS-implicated MIR137/MIR2682 locus might confer risk to schizophrenia and bipolar disorder. Am. J. Hum. Genet. 95 , 744–753 (2014).25434007
17 K. E. Szulwach , Cross talk between microRNA and epigenetic regulation in adult neurogenesis. J. Cell Biol. 189 , 127–141 (2010).20368621
18 K. Jiang, C. Ren, V. D. Nair, MicroRNA-137 represses Klf4 and Tbx3 during differentiation of mouse embryonic stem cells. Stem Cell Res. 11 , 1299–1313 (2013).24084696
19 C. Tarantino , miRNA 34a, 100, and 137 modulate differentiation of mouse embryonic stem cells. FASEB J. 24 , 3255–3263 (2010).20439489
20 S. L. Hollins , Ontogeny of small RNA in the regulation of mammalian brain development. BMC Genomics 15 , 777 (2014).25204312
21 C. Jeck, J. Pinto, P. Boyden, Transient outward currents in subendocardial Purkinje myocytes surviving in the infarcted heart. Circulation 92 , 465–473 (1995).7634462
22 M. Strazisar , MIR137 variants identified in psychiatric patients affect synaptogenesis and neuronal transmission gene sets. Mol. Psychiatry 20 , 472–481 (2015).24888363
23 E. He , MIR137 schizophrenia-associated locus controls synaptic function by regulating synaptogenesis, synapse maturation and synaptic transmission. Hum. Mol. Genet. 27 , 1879–1891 (2018).29635364
24 N. F. Olde Loohuis , MicroRNA-137 controls AMPA-receptor-mediated transmission and mGluR-dependent LTD. Cell Rep. 11 , 1876–1884 (2015).26095359
25 Y. Cheng , Partial loss of psychiatric risk gene Mir137 in mice causes repetitive behavior and impairs sociability and learning via increased Pde10a. Nat. Neurosci. 21 , 1689–1703 (2018).30397325
26 S. Siegert , The schizophrenia risk gene product miR-137 alters presynaptic plasticity. Nat. Neurosci. 18 , 1008–1016 (2015).26005852
27 Q. Chen , miR-137 is frequently down-regulated in gastric cancer and is a negative regulator of Cdc42. Dig. Dis. Sci. 56 , 2009–2016 (2011).21221794
28 S. M. Langevin , MicroRNA-137 promoter methylation in oral rinses from patients with squamous cell carcinoma of the head and neck is associated with gender and body mass index. Carcinogenesis 31 , 864–870 (2010).20197299
29 A. R. Smith , Tumor suppressive microRNA-137 negatively regulates Musashi-1 and colorectal cancer progression. Oncotarget 6 , 12558–12573 (2015).25940441
30 J. Xiao , microRNA-137 modulates pancreatic cancer cells tumor growth, invasion and sensitivity to chemotherapy. Int. J. Clin. Exp. Pathol. 7 , 7442–7450 (2014).25550779
31 Z.-M. Li, H.-Y. Zhang, Y.-X. Wang, W.-B. Wang, MicroRNA-137 is downregulated in human osteosarcoma and regulates cell proliferation and migration through targeting FXYD6. J. Drug Target. 24 , 102–110 (2016).26302771
32 J. Guo, B. Xia, F. Meng, G. Lou, miR-137 suppresses cell growth in ovarian cancer by targeting AEG-1. Biochem. Biophys. Res. Commun. 441 , 357–363 (2013).24144591
33 L. Chen , miR-137 is frequently down-regulated in glioblastoma and is a negative regulator of Cox-2. Eur. J. Cancer 48 , 3104–3111 (2012).22406049
34 K. K. Li , MIR-137 suppresses growth and invasion, is downregulated in oligodendroglial tumors and targets CSE1L. Brain Pathol. 23 , 426–439 (2013).23252729
35 S. Hao , miR-137 inhibits proliferation of melanoma cells by targeting PAK2. Exp. Dermatol. 24 , 947–952 (2015).26186482
36 C. Luo , miR-137 inhibits the invasion of melanoma cells through downregulation of multiple oncogenic target genes. J. Invest. Dermatol. 133 , 768–775 (2013).23151846
37 N. Chatterjee, N. Perrimon, What fuels the fly: Energy metabolism in Drosophila and its application to the study of obesity and diabetes. Sci. Adv. 7 , eabg4336 (2021).34108216
38 S. K. Kim, D. D. Tsao, G. S. B. Suh, I. Miguel-Aliaga, Discovering signaling mechanisms governing metabolism and metabolic diseases with Drosophila. Cell Metab. 33 , 1279–1292 (2021).34139200
39 D. R. Nassel, Y. Liu, J. Luo, Insulin/IGF signaling and its regulation in Drosophila. Gen. Comp. Endocrinol. 221 , 255–266 (2015).25616197
40 R. Bohni , Autonomous control of cell and organ size by CHICO, a Drosophila homolog of vertebrate IRS1-4. Cell 97 , 865–875 (1999).10399915
41 C. Werz, K. Kohler, E. Hafen, H. Stocker, The Drosophila SH2B family adaptor Lnk acts in parallel to chico in the insulin signaling pathway. PLoS Genet. 5 , e1000596 (2009).19680438
42 C. Chen, J. Jack, R. S. Garofalo, The Drosophila insulin receptor is required for normal growth. Endocrinology 137 , 846–856 (1996).8603594
43 D. J. Withers , Disruption of IRS-2 causes type 2 diabetes in mice. Nature 391 , 900–904 (1998).9495343
44 S. J. Broughton , Longer lifespan, altered metabolism, and stress resistance in Drosophila from ablation of cells making insulin-like ligands. Proc. Natl. Acad. Sci. U.S.A. 102 , 3105–3110 (2005).15708981
45 S. Gronke, D. F. Clarke, S. Broughton, T. D. Andrews, L. Partridge, Molecular evolution and functional characterization of Drosophila insulin-like peptides. PLoS Genet. 6 , e1000857 (2010).20195512
46 T. Ikeya, M. Galic, P. Belawat, K. Nairz, E. Hafen, Nutrient-dependent expression of insulin-like peptides from neuroendocrine cells in the CNS contributes to growth regulation in Drosophila. Curr. Biol. 12 , 1293–1300 (2002).12176357
47 E. J. Rulifson, S. K. Kim, R. Nusse, Ablation of insulin-producing neurons in flies: Growth and diabetic phenotypes. Science 296 , 1118–1120 (2002).12004130
48 D. J. Clancy , Extension of life-span by loss of CHICO, a Drosophila insulin receptor substrate protein. Science 292 , 104–106 (2001).11292874
49 S. J. Kwak , Drosophila adiponectin receptor in insulin producing cells regulates glucose and lipid metabolism by controlling insulin secretion. PLoS One 8 , e68641 (2013).23874700
50 K. D. Kimura, H. A. Tissenbaum, Y. Liu, G. Ruvkun, daf-2, an insulin receptor-like gene that regulates longevity and diapause in Caenorhabditis elegans. Science 277 , 942–946 (1997).9252323
51 C. J. Kenyon, The genetics of ageing. Nature 464 , 504–512 (2010).20336132
52 C. Kenyon, J. Chang, E. Gensch, A. Rudner, R. Tabtiang, A C. elegans mutant that lives twice as long as wild type. Nature 366 , 461–464 (1993).8247153
53 M. Tatar , A mutant Drosophila insulin receptor homolog that extends life-span and impairs neuroendocrine function. Science 292 , 107–110 (2001).11292875
54 M. Holzenberger , IGF-1 receptor regulates lifespan and resistance to oxidative stress in mice. Nature 421 , 182–187 (2003).12483226
55 M. Bluher, B. B. Kahn, C. R. Kahn, Extended longevity in mice lacking the insulin receptor in adipose tissue. Science 299 , 572–574 (2003).12543978
56 M. E. Giannakou , Long-lived Drosophila with overexpressed dFOXO in adult fat body. Science 305 , 361 (2004).15192154
57 D. S. Hwangbo, B. Gershman, M. P. Tu, M. Palmer, M. Tatar, Drosophila dFOXO controls lifespan and regulates insulin signalling in brain and fat body. Nature 429 , 562–566 (2004).15175753
58 J. M. Bao , Association between FOXO3A gene polymorphisms and human longevity: A meta-analysis. Asian J. Androl. 16 , 446–452 (2014).24589462
59 Y. Suh , Functionally significant insulin-like growth factor I receptor mutations in centenarians. Proc. Natl. Acad. Sci. U.S.A. 105 , 3438–3442 (2008).18316725
60 B. J. Buszard , The nucleus- and endoplasmic reticulum-targeted forms of protein tyrosine phosphatase 61F regulate Drosophila growth, life span, and fecundity. Mol. Cell Biol. 33 , 1345–1356 (2013).23339871
61 C. L. Wu , Dock/Nck facilitates PTP61F/PTP1B regulation of insulin signalling. Biochem. J. 439 , 151–159 (2011).21707536
62 L. F. Willoughby , Differential regulation of protein tyrosine kinase signalling by Dock and the PTP61F variants. FEBS J. 284 , 2231–2250 (2017).28544778
63 Y. W. Chen , Systematic study of Drosophila microRNA functions using a collection of targeted knockout mutations. Dev. Cell 31 , 784–800 (2014).25535920
64 J. R. Aguila, J. Suszko, A. G. Gibbs, D. K. Hoshizaki, The role of larval fat cells in adult Drosophila melanogaster. J. Exp. Biol. 210 , 956–963 (2007).17337708
65 J. E. Park , Flightless-I controls fat storage in Drosophila. Mol. Cells 41 , 603–611 (2018).29890821
66 S. Sajwan , Targeted mutagenesis and functional analysis of adipokinetic hormone-encoding gene in Drosophila. Insect Biochem. Mol. Biol. 61 , 79–86 (2015).25641265
67 R. Fernandez, D. Tabarini, N. Azpiazu, M. Frasch, J. Schlessinger, The Drosophila insulin receptor homolog: A gene essential for embryonic development encodes two receptor isoforms with different signaling potential. EMBO J. 14 , 3373–3384 (1995).7628438
68 J. G. Ruby , Large-scale sequencing reveals 21U-RNAs and additional microRNAs and endogenous siRNAs in C. elegans. Cell 127 , 1193–1207 (2006).17174894
69 J. G. Ruby , Evolution, biogenesis, expression, and target predictions of a substantially expanded set of Drosophila microRNAs. Genome Res. 17 , 1850–1864 (2007).17989254
70 A. Salmeen, J. N. Andersen, M. P. Myers, N. K. Tonks, D. Barford, Molecular basis for the dephosphorylation of the activation segment of the insulin receptor by protein tyrosine phosphatase 1B. Mol. Cell 6 , 1401–1412 (2000).11163213
71 S. Galic , Coordinated regulation of insulin signaling by the protein tyrosine phosphatases PTP1B and TCPTP. Mol. Cell Biol. 25 , 819–829 (2005).15632081
72 F. G. Haj, J. M. Zabolotny, Y. B. Kim, B. B. Kahn, B. G. Neel, Liver-specific protein-tyrosine phosphatase 1B (PTP1B) re-expression alters glucose homeostasis of PTP1B-/-mice. J. Biol. Chem. 280 , 15038–15046 (2005).15699041
73 K. J. Venken , MiMIC: A highly versatile transposon insertion resource for engineering Drosophila melanogaster genes. Nat. Methods 8 , 737–743 (2011).21985007
74 T. Kitamura, C. R. Kahn, D. Accili, Insulin receptor knockout mice. Annu. Rev. Physiol. 65 , 313–332 (2003).12471165
75 A. E. Bunner, P. C. Chandrasekera, N. D. Barnard, Knockout mouse models of insulin signaling: Relevance past and future. World J. Diabetes 5 , 146–159 (2014).24748928
76 K. T. Thomas , Inhibition of the schizophrenia-associated microRNA miR-137 disrupts Nrg1alpha neurodevelopmental signal transduction. Cell Rep. 20 , 1–12 (2017).28683304
77 L. Li , MiR-137 promotes anoikis through modulating the AKT signaling pathways in pancreatic cancer. J. Cancer 11 , 6277–6285 (2020).33033511
78 J. Wang, Z. Wang, J. Yuan, Q. Wang, X. Shen, Upregulation of miR-137 expression suppresses tumor growth and progression via interacting with DNMT3a through inhibiting the PTEN/Akt signaling in HCC. Onco Targets Ther. 14 , 165–176 (2021).33447058
79 Y. Gui, L. Wang, Z. Huang, MiR-137 inhibits cervical cancer progression via down-modulating Notch1 and inhibiting the PI3K/AKT/mTOR signaling pathway. Transl. Cancer Res. 10 , 3748–3756 (2021).35116675
80 L. L. Liu , FoxD3-regulated microRNA-137 suppresses tumour growth and metastasis in human hepatocellular carcinoma by targeting AKT2. Oncotarget 5 , 5113–5124 (2014).24970808
81 A. A. Hoffmann, R. Hallas, A. R. Anderson, M. Telonis-Scott, Evidence for a robust sex-specific trade-off between cold resistance and starvation resistance in Drosophila melanogaster. J. Evol. Biol. 18 , 804–810 (2005).16033551
82 A. A. Hoffmann, L. G. Harshman, Desiccation and starvation resistance in Drosophila: Patterns of variation at the species, population and intrapopulation levels. Heredity (Edinb) 83 , 637–643 (1999).10651907
83 D. J. Borash, G. T. Ho, Patterns of selection: Stress resistance and energy storage in density-dependent populations of Drosophila melanogaster. J. Insect Physiol. 47 , 1349–1356 (2001).12770141
84 D. P. Bartel, Metazoan microRNAs. Cell 173 , 20–51 (2018).29570994
85 S. W. Eichhorn , mRNA destabilization is the dominant effect of mammalian microRNAs by the time substantial repression ensues. Mol. Cell 56 , 104–115 (2014).25263593
86 K. Saliminejad, H. R. Khorram Khorshid, S. Soleymani Fard, S. H. Ghaffari, An overview of microRNAs: Biology, functions, therapeutics, and analysis methods. J. Cell Physiol. 234 , 5451–5465 (2019).30471116
87 J. E. La Marca , PTP61F mediates cell competition and mitigates tumorigenesis. Int. J. Mol. Sci. 22 , 12732 (2021).34884538
88 G. H. Baeg, R. Zhou, N. Perrimon, Genome-wide RNAi analysis of JAK/STAT signaling components in Drosophila. Genes Dev. 19 , 1861–1870 (2005).16055650
89 C. H. Huang, T. Y. Lin, R. L. Pan, J. L. Juang, The involvement of Abl and PTP61F in the regulation of Abi protein localization and stability and lamella formation in Drosophila S2 cells. J. Biol. Chem. 282 , 32442–32452 (2007).17804420
90 H. Y. Ku, C. L. Wu, L. Rabinow, G. C. Chen, T. C. Meng, Organization of F-actin via concerted regulation of Kette by PTP61F and dAbl. Mol. Cell Biol. 29 , 3623–3632 (2009).19398577
91 S. Tchankouo-Nguetcheu , Negative regulation of MAP kinase signaling in Drosophila by Ptp61F/PTP1B. Mol. Genet. Genomics 289 , 795–806 (2014).24752400
92 N. Weye , Association of specific mental disorders with premature mortality in the danish population using alternative measurement methods. JAMA Netw. Open 3 , e206646 (2020).32492163
93 O. Plana-Ripoll , Changes over time in the differential mortality gap in individuals with mental disorders. JAMA Psychiatry 77 , 648–650 (2020).32267492
94 S. Saha, D. Chant, J. McGrath, A systematic review of mortality in schizophrenia: Is the differential mortality gap worsening over time? Arch. Gen. Psychiatry 64 , 1123–1131 (2007).17909124
95 N. C. Momen , Association between mental disorders and subsequent medical conditions. N. Engl. J. Med. 382 , 1721–1731 (2020).32348643
96 E. C. Harris, B. Barraclough, Excess mortality of mental disorder. Br J. Psychiatry 173 , 11–53 (1998).9850203
97 T. M. Laursen, T. Munk-Olsen, M. Vestergaard, Life expectancy and cardiovascular mortality in persons with schizophrenia. Curr. Opin. Psychiatry 25 , 83–88 (2012).22249081
98 O. Plana-Ripoll , A comprehensive analysis of mortality-related health metrics associated with mental disorders: A nationwide, register-based cohort study. Lancet 394 , 1827–1835 (2019).31668728
99 S. Podolsky, N. A. Leopold, D. S. Sax, Increased frequency of diabetes mellitus in patients with Huntington’s chorea. Lancet 1 , 1356–1358 (1972).4113563
100 Z. Li , Glucose and insulin-related traits, type 2 diabetes and risk of schizophrenia: A mendelian randomization study. EBioMedicine 34 , 182–188 (2018).30100396
101 S. M. Grundy , Diagnosis and management of the metabolic syndrome: An American Heart Association/National Heart, Lung, and Blood Institute scientific statement: Executive summary. Crit. Pathw. Cardiol. 4 , 198–203 (2005).18340209
102 X. Lang , The prevalence and clinical correlates of metabolic syndrome and cardiometabolic alterations in 430 drug-naive patients in their first episode of schizophrenia. Psychopharmacology (Berl) 238 , 3643–3652 (2021).34586464
103 K. Coello , Metabolic profile in patients with newly diagnosed bipolar disorder and their unaffected first-degree relatives. Int. J. Bipolar Disord. 7 , 8 (2019).30937579
104 T. Pillinger , Impaired glucose homeostasis in first-episode schizophrenia: A systematic review and meta-analysis. JAMA Psychiatry 74 , 261–269 (2017).28097367
105 B. I. Perry, G. McIntosh, S. Weich, S. Singh, K. Rees, The association between first-episode psychosis and abnormal glycaemic control: Systematic review and meta-analysis. Lancet Psychiatry 3 , 1049–1058 (2016).27720402
106 L. J. van Nimwegen , Hepatic insulin resistance in antipsychotic naive schizophrenic patients: Stable isotope studies of glucose metabolism. J. Clin. Endocrinol. Metab. 93 , 572–577 (2008).18029467
107 C. Deng, K. Weston-Green, X. F. Huang, The role of histaminergic H1 and H3 receptors in food intake: A mechanism for atypical antipsychotic-induced weight gain? Prog. Neuropsychopharmacol. Biol. Psychiatry 34 , 1–4 (2010).19922755
108 R. Singh, Y. Bansal, B. Medhi, A. Kuhad, Antipsychotics-induced metabolic alterations: Recounting the mechanistic insights, therapeutic targets and pharmacological alternatives. Eur. J. Pharmacol. 844 , 231–240 (2019).30529195
109 R. Lopes, B. D. Pereira, Delirium and psychotic symptoms associated with hyperglycemia in a patient with poorly controlled type 2 diabetes mellitus. Innov. Clin. Neurosci. 15 , 30–33 (2018).
110 S. Dash, C. Xiao, C. Morgantini, K. Koulajian, G. F. Lewis, Intranasal insulin suppresses endogenous glucose production in humans compared with placebo in the presence of similar venous insulin concentrations. Diabetes 64 , 766–774 (2015).25288674
111 M. A. Abraham , Hypothalamic glucagon signals through the KATP channels to regulate glucose production. Mol. Metab. 3 , 202–208 (2014).24634823
112 S. Obici, B. B. Zhang, G. Karkanias, L. Rossetti, Hypothalamic insulin signaling is required for inhibition of glucose production. Nat. Med. 8 , 1376–1382 (2002).12426561
113 A. Pocai , Hypothalamic K(ATP) channels control hepatic glucose production. Nature 434 , 1026–1031 (2005).15846348
114 S. Grover , Comparative study of prevalence of metabolic syndrome in bipolar disorder and schizophrenia from North India. Nord. J. Psychiatry 68 , 72–77 (2014).23293896
115 F. P. Moreira , Metabolic syndrome and psychiatric disorders: A population-based study. Braz. J. Psychiatry 41 , 38–43 (2019).30328961
116 J. M. Tennessen, W. E. Barry, J. Cox, C. S. Thummel, Methods for studying metabolism in Drosophila. Methods 68 , 105–115 (2014).24631891
117 Y. Ping , Linking abeta42-induced hyperexcitability to neurodegeneration, learning and motor deficits, and a shorter lifespan in an Alzheimer’s model. PLoS Genet. 11 , e1005025 (2015).25774758
118 S. Benzer, Behavioral mutants of Drosophila isolated by countercurrent distribution. Proc. Natl. Acad. Sci. U.S.A. 58 , 1112–1119 (1967).16578662
