Abstract
Relatively little is known about the first step of insulin synthesis: the translation of the mRNA. Here we show that the translation of D. melanogaster insulin 2 mRNA (dilp2) is controlled by methylation of N6-adenosine (m6A) in the 3’ UTR. Mutations in the m6A writer Mettl3 and methylated-residues in the dilp2 3’UTR decreased the levels of dilp2 mRNA associated with the polysomes, and the total amount of dilp2 protein produced. This resulted in aberrant energy homeostasis and diabetic-like phenotypes, consistent with the specific function of dilp2 in adult metabolism. Conserved m6A signatures were also identified in the 3’ UTRs of vertebrate insulin mRNAs. These data identify m6A as a key regulator of insulin protein synthesis and energy homeostasis in metazoans and demonstrate an essential role for m6A in translation, with important implications for diabetes and metabolic disease.
One-Sentence Summary The most abundant modification in eukaryotic mRNAs controls the synthesis of insulin protein in D. melanogaster.
m6A is an abundant internal modification of eukaryotic mRNAs involved in regulating mRNA stability, turnover, and translation (1). As such, it plays an important role in many biological processes, including differentiation, development, and cancer (2). Recent studies have also implicated the m6A pathway in energy homeostasis, particularly in the pathogenesis of Type 2 Diabetes (T2D), a chronic disease characterized by the inability to regulate blood glucose via the insulin hormone. Genetic depletion of m6A levels in mammalian insulin β-cells resulted in higher circulating glucose and diabetic phenotypes (3, 4) as well as impaired insulin-cell mass and maturation in vitro (4). Reduced m6A levels were also observed in the pancreatic β-cells of humans with T2D (5). These observations suggest that m6A plays a role in energy homeostasis, especially glucose regulation and T2D. However, the specific mechanisms through which this pathway controls energy balance remain unresolved. Further, translational control is thought to be a critical step in the production of insulin, but the mechanism remains uncharacterized (6, 7). Here we took advantage of the ancestral conservation of the m6A (8) and insulin systems in invertebrates (9) to identify the molecular mechanisms through which m6A regulates glucose homeostasis and its contributions to diabetes and metabolic disease.
In D. melanogaster, 14 neuroendocrine cells located in the pars intercelebralis (Fig. 1A) regulate hemolymph glycemia and energy homeostasis by producing three insulin-like hormones (10). While these insulin-like hormones have redundant roles in development, the Drosophila insulin-like 2 (dilp2) peptide is necessary and sufficient to regulate hemolymph glycemia and fat levels in adult flies; dilp2 also has the highest homology to human insulin, while other dilps are related to the Relaxin family of hormones (9, 11–13). Mutations in dilp2 and ablation of the insulin-producing cells result in higher levels of circulating sugar and reduced metabolic function, phenotypes that can be rescued by injection of human insulin (9–12). Networks of transcripts involved in fly metabolism are methylated (14, 15). The accessible location of these cells, together with the genetic and metabolic analysis tools available, makes the fly a uniquely suited model to investigate the in vivo function of the m6A pathway in glucose and energy homeostasis.
To tackle this question, we first measured the levels of circulating glucose and of body triglycerides in flies mutant for the conserved methyltransferase writer enzyme Mettl3 (1). Compared to control flies, Mettl3−/− mutants showed higher fasted circulating sugar levels (Fig. 1B), as well as increased triglycerides, as previously reported for dilp2 mutants (12) (Fig. 1C). Cell-specific knockdown of Mettl3 in the insulin-producing cells using the Dilp2-GAL4 transgene (50% efficiency, Fig S1A) phenocopied the effects of Mettl3−/− mutants and resulted in higher circulating glucose and triglycerides (Fig. 1D, E). These phenotypes did not arise from developmental alterations in the dilp2+ cells because knocking down Mettl3 only in post-eclosion, adult dilp2>Mettl3RNAi; tubulin-GAL80ts flies showed the same energy homeostasis phenotypes as dilp2>Mettl3RNAi animals (Fig. S1B). Importantly, expression of wild-type Mettl3 only in the insulin cells in an otherwise Mettl3−/− mutant background rescued the energy balance defects, indicating that the phenotype arises from these cells (Fig. 1F). Thus, Mettl3 acts in the insulin cells to regulate energy balance; this is similar to the phenotypes of m6A writers mutants observed in mice (3).
Since the physiological effects of Mettl3 depletion are reminiscent of those caused by mutations in the dilp2 gene –but not other insulin-like peptides (12)– and murine data implicated potential changes in insulin levels (3, 5), we asked if Mettl3 affected the amount of this hormone. We found no changes in the abundance of dilp2 mRNA between Mettl3−/− and control flies (Fig. 2A); the amount of other insulin-like mRNAs produced in these cells, dilp3 and dilp5, were unchanged (Fig. S2A, B). In contrast, we observed a marked reduction in dilp2 protein in the insulin-producing cells of Mettl3−/− mutant flies (Fig. 2B, C); no changes in the number and morphology of the insulin cells, or the amounts of the closely related dilp3 hormone were observed (Fig. S2C, D, E). m6A regulates both RNA stability/turnover and translation in a context-dependent way (2). The observed decrease in dilp2 protein levels without accompanying changes in mRNA abundance suggests that m6A may contribute to the translation of the dilp2 transcript. This may also explain why we do not observe a concomitant transcriptional response from the other dilps (Fig. S2A, B) (16). To test the hypothesis that the m6A pathway affects the translational status of the dilp2 mRNA, we fractionated polysomes from control and Mettl3−/− mutants and measured the amount of dilp2 transcript associated with each fraction (Fig. 2D, E). In control flies, 89% of the dilp2 mRNAs cosedimented with polysome fractions, suggesting active and efficient translation (Fig. 2F, gray). Strikingly, this pattern was reversed in Mettl3−/− flies, where only 19% of dilp2 mRNA was found in the heavier fractions; instead, 80% of this mRNA was associated with early fractions, representing ribosomal individual subunits and monosomes (Fig. 2F, green). This suggests that m6A is required for loading the dilp2 mRNA onto polysomes and therefore proper translation of the dilp2 mRNA into protein.
To characterize the molecular causes of this phenotype, we used fly heads to perform m6A ultraviolet light-induced Cross-linking and Immunoprecipitation (miCLIP (17)), a technique that identifies transcripts marked by m6A. The three biological replicates showed good reproducibility with a mean correlation of 0.95 (Fig. S3A) and identified 4,506 m6A peaks corresponding to 1,828 genes involved in brain processes including development and plasticity (Data S1 and Fig. S3B). Forty-four percent (2009) of our peaks overlap with previous CLIP data from fly heads including Atpα (18) (Fig. S3C). Metagene analysis revealed that most m6A peaks mark the 5’ untranslated region (UTR) of transcripts, while a much smaller portion is present in the 3’ UTR, particularly near the stop codon (Fig. 3A); this is consistent with previous miCLIP studies in flies (18, 19). We also observed an enriched fly RRAC (R=purine) sequence motif at C-to-T crosslinking-induced mutation sites (CIMS) which represent m6A sites from CLIP/RIP data (14, 17–19) (Fig. 3B, Data S1).
Since we observed lower translation of the dilp2 mRNA in Mettl3−/− mutants, we asked if this transcript was methylated. Indeed, an m6A peak was present in the 3’ UTR of dilp2, shortly after the stop codon (Fig. 3C); none of the mRNAs encoding for other insulin-like peptides expressed in the insulin-producing cells, such as dilp3 or dilp5, showed any signatures of methylation (Fig S3D, E). Thus, the dilp2 mRNA is methylated in vivo. To better characterize the location of the modified A we turned to direct-RNA sequencing (Oxford Nanopore) where modified bases can be detected as changes in normalized current through the nanopore (20–22). We first in vitro transcribed RNA oligomers containing only one methylated or unmethylated A, sequenced them on the Nanopore, and then used the EpiNano algorithm to assess deviations in base-calling between these two identical oligomers (Fig. S4A, Table S1) (21). Direct-RNA sequencing of these RNAs revealed a shift in the raw current near the methylated bases (Fig. S4B) which results in a significant difference in base-call “errors” (Fig. S4C). Thus, we can detect N6-methylation as a deviation in base calling at or near to the methylated A via direct-RNA sequencing. To define if the native dilp2 mRNA was methylated in vivo, we enriched for insulin-cell-specific mRNAs by expressing the FLAG-tagged Ribosomal Protein 3 UAS-RPL3::FLAG transgene with Dilp2-GAL4. We then direct-RNA sequenced the purified poly(A)+ RNA and defined deviations between base calls of in vitro transcribed and native RNA (21). This experiment revealed a significant difference in the 3’ UTR between the in vitro transcribed and native dilp2 RNAs (Fig. 3D, E). In particular, three bases in the 3’UTR (Positions G519, C614, C664) had significantly higher base-call deviations in the native RNA compared to the in vitro transcribed dilp2 RNA. Position G519 corresponds to the CLIP peak and positions C614 and C664 are part of AC dinucleotides (Table S2). Together with the results from miCLIP, these data show that at least three specific ACs in the 3’UTR of the dilp2 mRNA are methylated in vivo in the D. melanogaster insulin-producing cells.
To investigate whether N-6 RNA methylation of the dilp2 3’ UTR directly affects its translation, we generated flies that lacked “methylateable” A’s in the 3’ UTR (dilp2 m6A−/−). Using the miCLIP and Nanopore analyses, we selected 11 “AC” nucleotides and mutated these into “UC” using CRISPR (Fig. S5A). These included the AC near G519 and C614 with the strongest methylation signals (Fig. 3C, E); A663 could not be removed for technical reasons. Clone 364-4 was selected for further study after confirming all the A>U mutations by Sanger sequencing (Fig. S5B). Sucrose density polysome gradients showed that compared to control flies, where 92% of the dilp2 mRNAs were in heavier fractions, representing the polysomes, 69% of dilp2 mRNAs from dilp2 m6A−/− flies were associated with early fractions (Fig 4A, S6A, B). Consistent with this, quantification of the total levels of dilp2 protein with anti-dilp2 specific antibodies in the insulin cells revealed a 20 % decrease in dilp2 m6A−/− flies compared to controls (Fig. 4B, C). Strikingly, we found that dilp2 m6A−/− mutants recapitulated the deficits in glucose and energy homeostasis observed in Mettl3 mutant flies, with an increase in fasting glucose levels and in triglycerides (Fig 4 D, E). Together, these findings suggest that, in flies, m6A modification of the dilp2 mRNA controls the effective translation, and thus synthesis, of the insulin protein.
Elements in the 3’ UTR of mammalian insulin are thought to play an important role in its translation (6, 7). Given the conservation of the insulin and m6A pathways in metazoa, we asked if signatures of m6A were also present in vertebrate insulin mRNAs. To do this, we obtained polyA-selected mRNAs from Atlantic salmon pancreatic tissue (Salmo salar) and mouse islet cells and analyzed them by direct-RNA sequencing. We analyzed 5,000 and 10,000 reads that mapped to the INS and INS2 insulin genes from salmon and mouse, respectively. This analysis revealed one putative m6A site in the 3’UTR of salmon insulin (Fig. 4F, Table S2, C464) and two putative m6A sites in INS2 3’ UTR (Fig. 4G, Table S2, A531, and C581). This suggests that regulation of insulin translation by m6A may be conserved.
Taken together, our observations support a general model where methylation of adenosines in the 3’UTR of the dilp2 mRNA enhances translation by promoting the association of this mRNA with polysomes which is consistent with a possible defect in translation initiation (23). In contrast, m6A in the 5’ UTR has been linked to non-canonical translational control (24). Although we cannot exclude that this mechanism also controls other aspects of dilp2 mRNA, our polysome profiling data links m6A marks in the dilp2 mRNA with translational initiation. In the absence of this epitranscriptomic signal, less insulin is produced, compromising the animal's ability to regulate sugar and energy homeostasis. Interestingly, this mark only controls the translation of the dilp2 mRNA and not the closely related dilp3 and dilp5 genes, consistent with the fact that the deposition of the signal is specific (25). Further, it suggests that previously observed transcriptional compensations among different insulin-like peptides in D. melanogaster are tied to mRNA abundance, not protein levels (16).
Based on our discovery that adenosines in the 3’UTR of mouse and salmon insulin mRNAs are marked with m6A, and on previous data showing that Mettl3 and Mettl14 are involved in glucose homeostasis, we anticipate that regulation of insulin translation and energy homeostasis by this RNA modification will be a conserved mechanism in vertebrates. Since m6A has been linked to cellular metabolism and signaling, this mode of regulation could provide a mechanism to time insulin translation with the presence of specific physiological signals, as well as to prepare for its fast production. This is consistent with findings that elements in the 3’UTR of mammalian insulin are important for its regulated production (6, 7) and that altered levels of m6A are found in the islets of people with T2D (5). However, it is still unclear if the m6A mark is dynamic and could really be used to switch translation. However, rather than toggling translation on and off, m6A could promote protein synthesis by direct cellular compartmentalization via phase separation (1). This work also provides direct evidence for the hypothesis that m6A in the 3’ UTR plays an essential and robust role in translation, although the mechanisms remain to be defined (1). Besides its relevance to the study of insulin and physiology, the methylation of dilp2 mRNA could be a viable model to address these and future questions about the biology of m6A. In summary, our work has uncovered a new fundamental step in the biosynthesis of insulin and the biology of m6A, with critical implications for understanding RNA control (or translational control) mechanisms in the regulation of energy homeostasis and the etiology of metabolic disease.
Funding
NIH R00 DK-97141 (MD)
NIH 1DP2DK-113750 (MD)
Rita Allen Foundation (MD)
NSF CAREER 1941822 (MD)
NIH T32 DA007268 (DW)
NIH P30 DK089503 (MD and DW)
Author contributions
Conceptualization: DW and MD
Methodology: DW
Investigation: DW and MD
Visualization: DW and MD
Funding acquisition: MD
Project administration: MD
Supervision: MD
Writing – DW and MD
Writing – review & editing: DW and MD
Competing interests
Authors declare that they have no competing interests.
Data and materials availability
All data are available in the main text or supplementary materials. RNA-sequencing reads were deposited in Gene Expression Omnibus accession number GSE207547. (The following secure token has been created to allow review of record GSE207547 while it remains in private status: cfkreeumnfajzkn) dIlp2 3’UTR mutants are available upon request.
Supplementary Materials
Materials and Methods
Fly lines and husbandry
All flies were maintained at 25 °C in a humidity-controlled incubator with a 12:12 hr light/dark cycle. Animals were fed Bloomington Food B (cornmeal-glucose) ad-lib and provided fresh food every other day. For all experiments, flies were collected under CO2 anesthesia, 2–4 days following eclosion, and housed in groups of 20–30 to age until testing (6-10 days old). The stocks used are listed in Table S3. Depending on the genetic background of the mutations, we used w1118 (Rainbow Transgenic Flies, Camarillo, CA) or w1118 Canton-S flies (Benzer lab, Caltech) as controls.
RNA extraction
Standard methods to isolate total RNA were used for quantitative polymerase chain reaction (qPCR), crosslinking and immunoprecipitation (CLIP), and direct-RNA sequencing. Briefly, heads from 10 to 20 flies were dissected and immediately frozen on dry ice for qPCR. 800 fly heads were isolated from bodies by sieving for CLIP and direct RNA-sequencing. All samples were stored at −80 °C until extraction. Phenol-chloroform (Invitrogen, 15596018) and RIPA buffer (150 mm NaCl, 1% Nonidet P-40, 0.5% Sodium deoxycholate, 0.1% SDS, 50mM Tris pH 7.5) were added to frozen samples and homogenized by bead bashing (Bead Ruptor, 19-040E). Phenol-chloroform extracted RNA was precipitated by isopropanol with GlycoBlue coprecipitant (Invitrogen AM9515). RNA was stored at −80 °C until further processing.
CLIP
Library preparation
We adapted miCLIP (17) to use infrared-dye-conjugated irCLIP adaptors (26). Briefly, 20 ug of poly(A) selected (Invitrogen, 61012) RNA for each biological replicate from 800 heads of w1118CS flies was used as input to the CLIP, no antibody control, and input only (no CLIP) reactions. Fragmented RNA (Ambion, E6150S) was incubated for 2 hr at 4 °C with antibody against m6A (Abcam, ab151230), UV-crosslinked, then immunoprecipitated with magnetic beads (Invitrogen, 10004D) as in (17). Pre-adenylated dye-conjugated linkers were ligated (NEB, M0351S) to dephosphorylated 3’ ends of RNA fragments overnight at 16 °C (Table S1). RNA was extracted with NuPAGE SDS buffer and separated by NuPAGE gel (Invitrogen, NP0321), transferred to nitrocellulose membrane and visualized for excision at 800 nm. RNA-antibody complexes were released from the membrane by proteinase K digestion and RNA was purified with phenol-chloroform extraction. RNA was reverse transcribed (Invitrogen, 18080085), circularized (Lucigen, CL4111K), and PCR-amplified (Thermo Scientific, F530S) following the published protocol (17). Libraries were subjected to 151 bp paired-end sequencing according to the manufacturer’s protocol on an Illumina NovaSeq 6000 at the University of Michigan Genomics core.
Bioinformatics
Sequencing reads were de-multiplexed using the Bcl2fastq2 Conversion Software (Illumina). Five prime end unique molecular identifiers (UMI, 9 nt random sequence) were used to remove PCR duplicates with a custom script. Then UMI and sequences and sequencing adapters were removed (fastx_clipper, 0.0.14). Reads were mapped to the Ensembl D. melanogaster genome (BDGP6) genome using Star (v 2.7.5a, (27)) with default settings (Table S4). Aligned reads were peak-called using Piranha (v 1.2.1) (28) (Data S1). Metagene analysis was performed using MetaPlotR (29). Crosslink induced mutation site analysis we performed as in (17). Sequence logo was generated using weblogo (v 2.8.2, (30)). Gene ontology (GO) analysis was performed using Gene Set Enrichment Analysis (GESA) implemented in the ClusterProfiler R package version 4.2.2 (31). We used Benjamini–Hochberg to correct for multiple hypothesis testing and “biological process” from the Bioconductor package “Genome Wide Annotations for Fly” version 3.8.2 for GESA (32).
Polysome fractionation
Polysome profiles were performed as previously described (33). Briefly, 300 fly heads were homogenized using a bead beater (Bead Ruptor, 19-040E) in 800 ul polysome extraction buffer (300 mM NaCl, 50 mM Tris-HCL (pH 7.5), 10 mM MgCl2, 200 mg heparin/ml, 400 U RNasin/ml, 1.0 mM, phenylmethylsulfonyl fluoride, 0.2 mg/ml cycloheximide, 1% Triton X-100, 0.1% sodium deoxycholate) then incubated on ice for 10 minutes. Lysate was cleared by centrifugation at 10,000 x g for 10 min at 4 °C. Equal A260 units were layered onto a 10–50% sucrose gradient in resolving buffer (20mM Tris-HCl (pH7.5), 150mM NaCl, 15mM MgCl2, 1mM DTT, 100ug/mL cycloheximide) and separated using a Beckman SW41Ti rotor (30,000 rpm for 3 hr at 4 °C). The absorbance (254 nm) was monitored and 750 ul fractions were collected using a Brandel pump set to a flow rate of 1.5mL/min. Equal molar concentrations of Saccharomyces cerevisiae enolase-2 (Eno2) transcript was added to all fractions before RNA isolation. Nucleic acid was precipitated from each fraction then pellets were resuspended in water and phenol chloroform extracted following the manufacturer's protocol (Invitrogen, 15596026). The RNA was precipitated with isopropanol and GlycoBlue. RNA was resuspended in 10 ul water and equal volumes from each fraction were reverse transcribed following the manufacture recommendations (Invitrogen, 18080085). Fractions 5-12 were probed for dilp2 and Eno2 by qPCR (Table S1).
qPCR
Reverse transcription was performed using SuperScript III (Invitrogen, 18080085) with 1 ug of total RNA as input and primed with oligo(dT) (Invitrogen, 18418012) according to the manufacturer's protocol for transcript abundance analysis. Quantitative-PCR was performed following manufacturer's directions (Applied Biosystems, 4367659) for all experiments. Primers were added at a 2.5 μM concentration in 20 ul reactions. Reactions were run on the StepOnePlus Real-Time PCR System (Applied Biosystems), and quantifications were normalized relative to the reference gene ribosomal protein 49 (Rp49) for transcript abundance or spike-in, Eno2, for polysome fractionation (delta Ct method).
Fat and lean mass analysis
Colormetric measurements of triglycerides (Stanbio, SB-2100-430) and protein (Pierce, PI23225) were done as previously described (34), where 1 biological replicate n=2 male flies. Flies were collected 3-5 days post eclosion and homogenized with a Bead Ruptor. Standard curves were generated for each to normalize the concentration of samples. Samples were quantified using a Tecan Spark plate reader at 562 nm for protein or 500 nm for triglycerides.
Immunofluorescence
Immunofluorescence was done essentially as (35). Male flies 3-5 days post eclosion were sorted then aged for 2-5 days on standard food. Flies were then fasted in vials with a wetted Kimwipe for 16-18 hours before dissection. Brains were dissected in PBS, fixed (4% paraformaldehyde aqueous solution in 1X PBS), blocked (10% normal goat serum, 2% Triton X-100, 1X PBS), and then incubated overnight in primary antibody (rat anti-dilp2 (1:500) (36), rabbit anti-dilp3 (1:500) (37)). After washing (3% NaCl, 1% Triton X-100, 1X PBS), brains were then incubated at room temp overnight in secondary antibody: either goat anti-rat Alexa Fluor 647 (Invitrogen, A-21247) or goat anti-rabbit Alexa Fluor 488 (Invitrogen, A-11008). Brains were mounted in FocusClear (CelExplorer, FC-101) on coverslips, and the cell bodies were imaged using an FV1200 Olympus confocal with a 20x objective. Median intensity of individual insulin-producing cells was quantified using Fiji (38).
in vitro transcription for direct-RNA sequencing
DNA insulin templates were synthesized by IDT based on transcripts FBtr0076329 (fly Dilp2), NM_001185083.2 (mouse Ins2), and two randomized sequences containing only one adenine (Supplemental Table 2). To generate the salmon Ins template, we PCR amplified (Thermo Scientific, F530S) DNA from cDNA and Sanger sequence-verified the product (XM_014198195.2). Each template DNA sequence was used for in vitro transcription (Invitrogen, AM1334). To generate the randomized DNA template “rand-A” the reaction mixture contained bases adenosine, uracil, cytosine, and guanosine while the randomized DNA template “rand-m6A” used N6-methyladenosine in the palace of adenosine. All the other in vitro transcription reactions used only the standard bases. The reactions were performed following the manufacturer's protocol overnight.
Direct-RNA sequencing
500 ng of poly(A) selected (Invitrogen) RNA from heads of flies or a total 500 ng of pooled in vitro transcribed RNA was used for library preparation following the manufacturer's protocol (Oxford Nanopore, SQK-RNA002, Version DRS_9080_v2_revB_22Nov2018). Briefly, the RT adapter was ligated to the RNA, reverse transcription was performed, and the RNA-cDNA hybrids were purified. Next, the second adapter was ligated to the RNA and again purified. The libraries were loaded onto the MinION flowcell (R9.4.1). The Oxford Nanopore sequencer was run for 24-36 hours. Data were base called using Oxford Nanopore’s Guppy (v 3.1.5) and aligned to the reference sequences using MiniMap2 -ax splice -uf -k14 (v 2.17) (39). Only reads that passed filtering and that mapped to the reference were considered for further analysis. Next, aligned reads from biological samples (modified) and matched in vitro transcribed RNAs (unmodified) were used as input to EpiNano (EpiNano-Error, v 1.2) to determine the positions of modifications (21). We plotted the data from the longest transcript (NM_001185083.2) for EpiNano mouse data. The Tombo suite of tools was used to visualize reads (40).
Circulating glucose assay
Hemolymph was collected from starved (12-16 hours) male flies from 40-50 flies per replicate by centrifugation. Circulating glucose levels were measured as previously described (41). Briefly, 0.5 ul of hemolymph was added to 100 ul of HexoKinase (HK) reagent (Sigma, GAHK20), incubated at room temperature for 15 min, then absorbance at 340 nm was measured on a Tecan Spark plate reader.
Generation of dilp2 m6A mutant flies
CRISPR constructs were synthesized and micro-injected into w1118 flies by Rainbow Transgenic Flies (Camarillo, CA). The pScarless donor vector (dsRed marker) was introduced to remove the endogenous Dilp2 3’UTR and replace it with a mutant Dilp2 3’UTR that replaced 11 AC dinucleotides to TC (Supplement Fig. 4A). F1 progeny were screened for transformation with dsRed fluorescence. Positive transformants were Sanger sequenced (below) to verify the correct insertion.
Sanger Sequencing
Genomic DNA was extracted from two male flies from each fly line with positive dsRed expression by silica column purification (Invitrogen, K182002). The dilp2 locus was PCR amplified (Thermo Scientific, F-530XL), PCR products were purified and normalized to 5ng/ul with 10 pmol/ul of the with appropriate primer added. Samples were submitted to Eurofins Genomics for sequencing and traces were analyzed with Benchling software.
Mouse islet isolations
Islet cells were collected from 4 fasted male mice from Vil-Cre backcrossed to C57 9x by the University of Michigan Islet Isolation core. The pooled tissue was added directly to Trizol (Invitrogen, 15596018) and stored at −80 °C.
Salmon pancreatic tissue isolation
The Atlantic salmon used was approximately 2 years old and post-smolt. The pancreatic tissue from one individual was isolated from the surrounding pyloric caeca. Upon removal, the tissue was immediately added to Trizol and stored at −80 °C.
Table S4. Summary of sequencing reads.
Data S1. List of CLIP peaks.
Complete list of peaks from each biological replicate and the union of the data sets. CIMS output of C to T transitions.
Acknowledgments
General: We thank Pierre Léopold for the kind gift of Dilp2 antibody, Patrick Callaerts for the gift of Dilp3 antibody, Randy Seeley for mice and Corentin Cras-Méneur for isolating Mouse islets, Jean-Yves Roignant for Mettl3 mutant flies, and the Blooming Drosophila Stock center for other flies used in this study. Salmon tissue was a gift from Brian Peterson (USDA); we thank Cunmin Duan for advice. We are grateful to Peter Todd and Shannon Miller for training and for the use of their polysome fractionation equipment and training; Christopher Lapointe for thoughtful comments on the manuscript. We also thank Julia Kuhl for designing some of the graphics in this manuscript.