Quantitative Phosphoproteomics Reveals Signaling Mechanisms

Subscriber access provided by UNIV OF NEWCASTLE

Article

Quantitative phosphoproteomics reveals signaling mechanisms associated with rapid cold hardening in a chill-tolerant fly Nicholas Mario Teets, and David L. Denlinger J. Proteome Res., Just Accepted Manuscript • DOI: 10.1021/acs.jproteome.6b00427 • Publication Date (Web): 30 Jun 2016 Downloaded from http://pubs.acs.org on July 1, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Journal of Proteome Research is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Proteome Research

Quantitative phosphoproteomics reveals signaling mechanisms associated with rapid cold hardening in a chill-tolerant fly Nicholas M. Teets*,†,‡ and David L. Denlinger†,§ †

Department of Entomology and §Department of Evolution, Ecology, and Organismal Biology,

Ohio State University, Columbus, OH 43210 USA KEYWORDS Cell signaling, cold-sensing, phosphoproteomics, phosphorylation, rapid cold hardening.

1 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ABSTRACT Rapid cold hardening (RCH) is a physiological adaptation in which brief chilling (minutes to hours) significantly enhances the cold tolerance of insects. RCH allows insects to cope with sudden cold snaps and diurnal variation in temperature, but the mechanistic basis of this rapid stress response is poorly understood. Here, we used phosphoproteomics to identify phosphorylation-mediated signaling events that are regulated by chilling that induces RCH. Phosphoproteomic changes were measured in both brain and fat body, two tissues that are essential for sensing cold and coordinating RCH at the organismal level. Tissues were chilled ex vivo, and changes in phosphoprotein abundance were measured using 2D electrophoresis coupled with Pro-Q diamond labeling of phosphoproteins, followed by protein identification via LC/MS/MS. In both tissues, we observed an abundance of protein phosphorylation events in response to chilling. Some of the proteins regulated by RCH-inducing chilling include proteins involved in cytoskeletal reorganization, heat shock proteins, and proteins involved in the degradation of damaged cellular components via the proteasome and autophagosome. Our results suggest that phosphorylation-mediated signaling cascades are major drivers of RCH and enhance our mechanistic understanding of this complex phenotype. INTRODUCTION Insects are the predominant animals in terrestrial ecosystems, and as small-bodied ectotherms their biology is closely tied to temperature.1 Temperature is one of the primary determinants of an insect’s range,2 and as such, insect populations are at particular risk from global climate change.3 While higher surface temperatures are the most conspicuous feature of climate change, the earth is also experiencing an increase in diel variation,4 and increased thermal variability may pose a greater risk to insects than climate warming alone.5,6 Furthermore, reduced snow cover in temperate areas may enhance temperature variation in winter hibernacula,


Page 2 of 29

Page 3 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


resulting in more freeze-thaw events.7 Thus, adaptations to cope with thermal variability, both at high and low temperatures, are critical for the current and future success of insect populations. While seasonal adaptations of insects to low temperature are well-characterized,8,9 insects are also capable of responding to low temperature on much shorter time scales. In a process called rapid cold hardening (RCH), brief exposure (i.e., minutes to hours) to nonlethal chilling significantly enhances cold shock tolerance.10,11 While RCH is readily observed at the organismal level, recent evidence indicates that insect cells and tissues are capable of detecting and responding to chilling ex vivo, with no nervous or hormonal input.12-14 Thus, RCH is at least partly driven at the cellular level, akin to the well-characterized heat shock response.15 Unlike the heat shock response, the molecular machinery of RCH is poorly understood.16 In contrast to many stress responses, there is no transcriptional component to RCH;17,18 the speed (minutes) and temperatures at which RCH occurs do not support transcription of new gene products. While a few proteins are produced in response to RCH conditions,19,20 RCH still occurs when protein synthesis is blocked,21 calling into question the functional role of these protein changes that have been observed. Thus, transcriptomic and proteomic evidence indicates RCH is primarily mediated by post-translational signaling mechanisms. Insect cells detect chilling via a gradual influx of calcium, and blocking this calcium entry prevents RCH.13,14 p38 MAP kinase is also rapidly phosphorylated by chilling,22,23 but the functional significance of this activation remains unexplored. At the biochemical level RCH is accompanied by modest cryoprotectant synthesis,24,25 although RCH has also been observed in the absence of cryoprotectant synthesis.17,26 Ultimately, RCH prevents mortality by reducing cold-induced apoptotic cell death,27,28 and a recent genome wide association study suggests autophagy may be involved in



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

inhibiting cold-induced cell death.29 However, despite recent progress, RCH is still poorly understood at the cellular level. Here, we test the hypothesis that cold-sensing and RCH are governed in part by rapid changes in protein phosphorylation. Using quantitative phosphoproteomics, we identify numerous changes in protein phosphorylation induced by chilling in brain and fat body of the flesh fly, Sarcophaga bullata. The brain coordinates behavioral and hormonal responses to cold stress30,31 and has been shown to augment the RCH response when present32. The fat body is the metabolic hub of an insect (akin to the liver) and executes many of the metabolic responses to low temperature33. S. bullata is a well-studied model of insect dormancy, and it also has a robust RCH response. Furthermore, these flies are roughly 100x larger than Drosophila, making them good subjects to investigate tissue-specific stress responses in experiments where large amounts of cellular material are needed. We report an abundance of chilling-induced phosphorylation events that expand our knowledge of potential signaling targets used by insect cells to detect and respond to chilling. EXPERIMENTAL SECTION Insects Flesh flies (Sarcophaga bullata) were reared at 25°C, 16:8 L:D. Adults were fed sugar, water, and beef liver ad lib, and a fresh piece of liver was provided for larviposition 11 d after eclosion. Larvae were fed beef liver and allowed to wander and pupariate on aspen chips.Puparia were transferred to a new cage shortly before eclostion. Adult males 5-8 d post-eclosion were used for experiments. To confirm the phenotypic effects of RCH in adult flies, we exposed 5-8 d old males to either direct cold shock (direct transfer from 25 to -8°C for 2 h) or RCH (transfer from 25 to 0°C


Page 4 of 29

Page 5 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


for 2 h, then immediately transferred to -8°C for 2 h). Each treatment was replicated four times, with each replicate consisting of 10 flies. Survival was assessed 24 h later by the ability of flies to right themselves after being turned on their back.

Dissections and Treatments Phosphoproteomic changes in response to chilling were assessed in the brain and fat body. Brains and abdominal fat body (sampled in separate flies) were dissected from male flies in room temperature insect saline (in mM: 187 NaCl, 21 KCl, 5.6 CaCl2-2H2O, 4.1 MgCl26H2O) and immediately transferred to Petri dishes containing room temperature Coast’s solution (in mM: 100 NaCl, 8.6 KCl, 4.0 NaHCO3, 4.0 NaH2PO4-H2O, 1.5 CaCl2-2H2O, 8.5 MgCl26H2O, 24 glucose, 25 Hepes, and 56 sucrose).34 Tissues were pooled to obtain enough protein; each brain sample consisted of 20 brains, while each fat body sample consisted of the abdominal fat body from 10 flies. Following dissection, control samples were maintained at 25 for 2 h while the chilled samples were kept at 0°C for 2 h in an ice water slurry. We elected to expose tissues to cold ex vivo to capture the cell-specific signaling events that occur in response to chilling, without input from the sensory nervous system or hormones. After treatment, the medium was aspirated and samples were snap-frozen in liquid nitrogen and stored at -80°C. For each tissue and treatment, we sampled four biological replicates. Experiments were run in blocks, so that only one control and one chilled sample were treated concurrently, to minimize the time between dissection and sampling. A power analysis using JMP 12 (SAS Institute, Cary, NC) indicated that four replicates would allow us to detect phosphoprotein differences of 50% at a power level of 0.8, assuming 20% standard deviation. Protein Extraction 5 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Tissue samples were removed individually from the freezer and proteins were extracted in ice cold RIPA buffer (25 mM Tris-HCl pH 7.6, 150 mM NaCl, 1% NP-40, 1% sodium deoxycholate, 0.1% SDS) containing Halt Protease and Phosphatase Inhibitor Cocktail (Thermo Fisher, Waltham, MA). The protein concentration of each sample was measured using the Pierce BCA Protein Assay Kit (Thermo Fisher). Homegenates were stored at -80°C and sent to the Ohio State Mass Spectrometry and Proteomics Facility for phosphoproteome analysis. 2D Gel Electrophoresis We used 2D gel electrophoresis coupled with the phosphoprotein-specific stain Pro-Q Diamond (Thermo Fisher) to measure changes in phosphoprotein abundance. Proteins were purified using methanol/chloroform precipitation and resuspended in rehydration buffer (7M Urea, 2M thiourea, 2% CHAPS, 1% pH 3-10 IPG buffer [GE Healthcare, Pittsburgh, PA], 50mM DTT, 1% saturated bromophenol blue solution) and centrifuged. Protein concentrations were normalized, and protein samples (600 µg for fat body, 300 µg for brains) were applied to pH 4-7 IEF strips (GE Healthcare). The IEF strips were focused at 500 V for 1 h, a gradient to 1000 V over 1 h, a gradient to 10,000 V over 3 hours, and held at 10,000 V for 3.25 hours. After IEF, strips were equilibrated at room temperature in 5 ml equilibration buffer A (50 mM Tris pH 8.9, 6M urea, 30% glycerol, 2% SDS, 0.5% DTT) for 15 minutes, followed by 5 ml of equilibration buffer B (50 mM Tris pH 8.8, 6M urea, 30% glycerol, 2% SDS, 4.5% iodoacetamide). Strips were rinsed in 1x SDS-PAGE running buffer (50 mM Tris, 384 mM glycine, 0.2% SDS) and placed in 20x24 cm 12% SDS-PAGE gels. Gels were run at 2 watts for 45 minutes, followed by 15 watts until the dye front reached the bottom of the gel. After electrophoresis gels were stained with Pro-Q Diamond according to the manufacturer’s


Page 6 of 29

Page 7 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


instructions and immediately scanned in a Typhoon 9400 scaner (GE Healthcare) using a 532 nm laser and a 560 nm long-pass filter. Image Analysis Gel images from fat body samples were initially analyzed using DeCyder software (GE Healthcare). Pro-Q-positive spots from each gel were matched using the Biological Variation Analysis module, and spot volumes were exported. Spot volumes were normalized to the percentage of total ProQ staining in each gel. A t-test was performed on the normalized ProQ signal percentage to identify protein spots that significantly varied across treatments. For the DeCyder analysis, we set statistical significance at α = 0.1. After using DeCyder to create a spot-picking list for identifying the proteins (see below), we revisited our analysis with SameSpots software (Totallabs, Newcastle, UK), which is more appropriate for single stain analyses. SameSpots analysis is similar to DeCyder with the following exception: in addition to normalizing to total Pro-Q intensity, SameSpots selects a reference gel that is most similar to all other gels in the experiment. Then, the mean log volume ratio of each spot, relative to the reference gel, was calculated to obtain a scaling factor to further normalize the gels. For SameSpots analysis, significance was set at α = 0.05. SameSpots analysis was superior to DeCyder and resulted in lower technical variation between samples; thus, the brain samples were only analyzed with SameSpots. However, for the fat body we present both SameSpots and DeCyder analyses, to provide a comprehensive dataset of candidate proteins. Protein Identification Protein spots of interest were cored with an Ettan Spot Handling Workstation (GE Healthcare) and placed in a 96 well plate. Gel pieces were washed twice in 100 µl 50% methanol 7 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

with 5% acetic acid, washed with acetonitrile for 5 min, dried, and resuspended in 50 mM NH4HCO3 three times. Rehydrated gel pieces were then digested with 5 µg/ml sequencing grade trypsin (Promege, Madison, WI) for 3 h at 37°C. Peptides were extracted three times with 50 µl 50% acetonitrile containing 5% formic acid, and extracts were concentrated. Peptides were analyzed using Capillary-LC/MS/MS on a Thermo Finnigan LTQ orbitrap mass spectrometer equipped with a microspray source (Michrom Bioresources, Auburn, CA) operated in positive ion mode. Samples were separated on a Magic C18AQ column (Bruker Daltonics, Billerica, MA) using an UltiMate 300 HPLC system (Thermo Fisher). Peptides were separated on a 2-90% acetonitrile gradient with 50 mM acetic acid. MS data were acquired using the data dependent TopTen method. Peak lists were searched against all Drosophila melanogaster proteins in the NCBI RefSeq database using Mascot Daemon v.2.2.1 (Matrix Science, Boston, MA) with the following parameters: mass accuracy of precursor ions set to 1.8 Da, fragment mass accuracy set to 0.8 Da, considered modifications were methionine oxidation, deamidation, carbamidomethyl cysteine, and phosphorylation, and two missed cleavages were permitted. Protein identifications were checked manually and proteins with a Mascot score >100 with at least two unique peptides were accepted. Functional Analyses Enrichment analyses and protein interaction analyses were conducted in STRING (stringdb.org).35 We tested for enrichment of Gene Ontology (GO) biological process terms and KEGG pathways, with the Drosophila melanogaster genome as the background. For enrichment analyses, we considered terms with FDR0.7. RESULTS AND DISCUSSION Phenotypic Effects of RCH To confirm the phenotypic effects of RCH in S. bullata, adult male flies were exposed to two treatments: a direct cold shock treatment, in which flies reared at 25°C were directly transferred to -8°C, and an RCH treatment in which flies reared at 25°C were first exposed to 0°C before being transferred to -8°C for 2 h. When survival was scored 24 h later, fewer than 10% of flies survived a direct cold shock, while nearly 100% of the flies exposed to RCH survived (Figure 1). Thus, our chilling conditions elicited a robust RCH response; these same conditions (2 h at 0°C) were used to quantify tissue-specific changes in phosphoprotein abundance. Brief Chilling Causes Many Protein Phosphorylation Events In response to chilling, there were numerous changes in phosphoprotein abundance in both the fat body and the brain (Figure 2). In the fat body, there were 64 phosphoproteins that were differentially regulated between control and chilled tissues, while the brain had 82 differentially regulated phosphoproteins (Supporting Information Table 1, 2). We observed that many proteins were represented by multiple differentially regulated spots; in both tissues 13 proteins contained 2 or more differentially abundant spots. Thus, while RCH elicits no or very few changes at the levels of transcription17 and translation,19 our phosphoproteomic results suggest that reversible protein phosphorylation is a significant contributor to the RCH phenotype. 9 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The top ten differentially regulated phosphoproteins for each tissue are shown in Figure 3. In the fat body, proteins in the top 10 included eukaryotic initiation factor 4a, three proteins involved in cytoskeletal dynamics and cell shape (α tubulin at 84B, actin 5c, myosin light chain 2), two spots belonging to ATP synthase β, two spots belonging to a lipid storage droplet protein, and two proteins with known function in stress responses (heat shock protein 83 and lethal (2) 37Cc). In the brain, the top 10 differentially phosphorylated proteins include Cdc42 (a Rho family GTPase involved in cytoskeletal dynamics and cell shape), ATP synthase β (three spots in top 10), CG7646 (a calcium binding protein), two proteins involved in protein synthesis (eukaryotic initiation factor 4a and ribosomal protein S3), transport and Golgi organization 7 (involved in Golgi dynamics and apoptosis signaling), and CG7461 (a predicted acyl-CoA dehydrogenase involved in beta oxidation). One of our motivations for profiling two tissues was to identify conserved protein phosphorylation events operating across both tissues. Nine common proteins were differentially phosphorylated in both tissues, although not always in the same direction (Figure 4). These proteins included proteins involved in cytoskeletal dynamics (four proteins), protein synthesis, and ATP synthesis. In addition, both tissues exhibited differential phosphorylation of lethal (2) 37 Cc, a gene of unknown molecular function that is involved in the cellular response to hypoxia.36 Cold and hypoxia share many characteristics at the cellular level (e.g., ATP depletion, reduced metabolic turnover, and a loss of ion homeostasis),37 and this results suggests mechanistic overlap. Both tissues also exhibited differential phosphorylation of Grasp65, a Golgi stacking protein that is involved in stress sensing and apoptosis regulation in mammals.38 Finally, 14-3-3ζ, a signaling protein that regulates Ras and MAPK signaling in flies,39 was


Page 10 of 29

Page 11 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


differentially phosphorylated in both tissues. The MAPK p38 is also rapidly phosphorylated by chilling in flies,22 suggesting 14-3-3ζ regulation may be occurring upstream of p38 activation. Without targeted functional validation of each protein, it is impossible to say whether a phosphorylation event is directly related to RCH or is a byproduct of chilling. However, if multiple proteins belonging to a common pathway or functional group are all differentially phosphorylated, that provides stronger evidence for a regulatory process. Thus, we conducted functional enrichment analyses to extract functional information from our dataset, and the majority of our discussion will focus on these results. Lists of differentially phosphorylated proteins from both tissues were combined and

evaluated for enrichment of GO terms and KEGG pathways (Supporting Information Table 3). In total, 21 GO biological processes and 7 KEGG pathways were enriched at FDR

Quantitative Phosphoproteomics Reveals Signaling Mechanisms

Recommend Documents