Proteomics of the rat myocardium during development of type 2

Subscriber access provided by University of Winnipeg Library

Article

Proteomics of the rat myocardium during development of type 2 diabetes mellitus reveals progressive alterations in major metabolic pathways Anders Valdemar Edhager, Jonas Agerlund Povlsen, Bo Løfgren, Hans Erik Bøtker, and Johan Palmfeldt J. Proteome Res., Just Accepted Manuscript • DOI: 10.1021/acs.jproteome.8b00276 • Publication Date (Web): 30 May 2018 Downloaded from http://pubs.acs.org on May 31, 2018

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 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 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.

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 31 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

Proteomics of the rat myocardium during development of type 2 diabetes mellitus reveals progressive alterations in major metabolic pathways

Author list: Anders Valdemar Edhager*1,a, Jonas Agerlund Povlsen2,a, Bo Løfgren2, 3, Hans Erik Bøtker2, Johan Palmfeldt1 1

Research Unit for Molecular Medicine, Department of Clinical Medicine, Aarhus University and Aarhus

University Hospital, Aarhus, Denmark, 2Department of Cardiology, Aarhus University Hospital, Aarhus, Denmark, 3Institute for Experimental Clinical Research, Aarhus University, Aarhus, Denmark a

Equal contribution

* Corresponding author: Anders Valdemar Edhager, MSc. email: [email protected], Tel. +4578455410

1 ACS Paragon Plus Environment

Journal of Proteome Research 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

Page 2 of 31

ABSTRACT Congestive heart failure and poor clinical outcome after myocardial infarction are known complications in patients with type-2 diabetes mellitus (T2DM). Protein alterations may be involved in the mechanisms underlying these disarrays in the diabetic heart. Here we map proteins involved in intracellular metabolic pathways in the Zucker diabetic fatty rat heart as T2DM develops using MS based proteomics. The pre-diabetic state only induced minor pathway changes, whereas onset and late T2DM caused pronounced perturbations. Two actin-associated proteins, ARPC2 and TPM3, were up-regulated at the prediabetic state indicating increased actin dynamics. All differentially regulated proteins involved in fatty acid metabolism, both peroxisomal and mitochondrial, were up-regulated at late T2DM, whereas enzymes of branched chain amino acid degradation were all down-regulated. At both onset and late T2DM, two members of the serine protease inhibitor superfamily, SERPINA3K and SERPINA3L were down-regulated, Furthermore, we found alterations in proteins involved in clearance of advanced glycation end-products and lipotoxicity, DCXR and CBR1, at both onset and late T2DM. These proteins deserve elucidation with regard to their role in T2DM pathogenesis and their respective role in the deterioration of the diabetic heart. Data are available via ProteomeXchange with identifiers PXD009538, PXD009554, and PXD009555. Keywords: LC-MS/MS, Proteomics, Type 2 diabetes mellitus, Rat, Heart, Mitochondrion, Peroxisomes, BCAA metabolism


Page 3 of 31 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


1 INTRODUCTION

The number of patients suffering from diabetes, predominantly type 2 diabetes mellitus (T2DM), is increasing worldwide and was estimated to comprise 422 million people in 2014 (1). T2DM and obesity are important risk factors for developing congestive heart failure (2) and death after myocardial infarction (MI) (3). Clinical studies have shown that T2DM is associated with increased mortality rate after acute myocardial infarction due to more extensive atherosclerotic lesions and a hypertrophied and dysfunctional left ventricle (4-6). Large-scale clinical trials on the effect of glycemic control in T2DM patients suffering myocardial infarction did not show a significant improvement on mortality following myocardial infarction (7-10). Furthermore, the rates of cardiovascular death and heart failure increased in T2DM patients following acute myocardial infarction despite of comparable left-ventricular systolic function and left-ventricular remodeling in the two groups (11). These results indicate that other factors than elevated blood glucose and systolic ventricular dysfunction may be responsible for the impaired prognosis of myocardial infarction and other cardiovascular diseases (CVD) in patients with T2DM. Additionally, studies in experimental models have shown that mitochondrial dysfunction, more specifically reduced oxidative capacity and cardiac efficiency, is an inherent part of the pathophysiology of T2DM (12, 13). However, only few studies have examined what underlying changes in the heart proteome might be accountable for these findings (14-18) and no study has to our knowledge, investigated the effect of disease progression. The present study aimed to analyze the rat myocardial proteome at different stages of T2DM to identify potential alterations in metabolic pathways, which may add to the current knowledge about detrimental effects in the heart accompanying the development of T2DM.



Page 4 of 31

2 MATERIALS AND METHODS 2.1 ETHICAL STATEMENT Animals were handled in accordance with national and institutional guidelines for animal research. The experimental work was approved by the Danish Animal Experiments Inspectorate (license no. 2011/5612010-C2).

2.2 ANIMALS AND COLLECTION OF RAT MYOCARDIAL TISSUE Male Zucker diabetic fatty (ZDF) rats (homozygous (ZDFfa/fa)) and their age-matched lean controls (heterozygous (ZDFfa/+)) (Charles River Laboratories, Kisslegg, Germany) were studied at ages 6, 12 and 24 weeks corresponding to a prediabetic state, onset of, and late T2DM. Rats were fed a Purina 5008 diet as recommended by the supplier (Charles River) and housed under controlled conditions with 12:12h light-dark cycles. Animals were subjected to 12-16 hours of fasting before experimental procedures. Rats were anesthesized with Dormicum® (midazolam, 0.5 mg/kg bodyweight (bwt), Matrix Pharmaceuticals, Herlev, DK) and Hypnorm® (fentanylcitrate, 0.158 mg/kg bwt and fluanisone, 0.5 mg/kg bwt, Vetapharma Ltd., Leeds, UK) by subcutaneous injection. Intubation was carried out through a tracheal incision and the tracheal tube was connected to a ventilator (Ugo Basile, Varese, Italy). A midline thoracotomy was performed followed by excision of a left ventricular biopsy with immediate snap-freezing in liquid nitrogen. The tissue biopsy was stored at -80 °C. Biochemical parameters were analyzed which included fasting Blood (B)-glucose, Serum (S)-total cholesterol, S-triglyceride and S-free fatty acid as previously described (19, 20). They confirmed the prediabetic -, onset - and late diabetic phenotype, respectively.


Page 5 of 31 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


2.3 STUDY 1 (LABEL-FREE) 2.3.1

Preparation of samples for mass spectrometry analysis

Thirty-six excised heart tissues (60-75 mg each) were homogenized for 30 s at 5000 rpm (Ultra-Turrax T8 homogenizer; IKA Labortechnik, Staufen, Germany) in 2 mL lysis buffer (20 mmol/L Tris-HCl, 100 mmol/L NaCl, 10 mmol/L EDTA, 0.5 mmol/L dithiotreitol (DTT; reducing agent), 1.5% Nonidet P-40 (non-ionic surfactant), and protease inhibitor cocktail (Complete mini, Roche diagnostics, pH 7.4). Homogenized samples were centrifuged at 1000 g for 15 minutes at 4˚C and the supernatant transferred to new Eppendorf tubes and kept at -20°C until further analysis. Protein concentrations were determined using the Bradford assay using an iMARK microplate absorbance reader (Bio-Rad). After protein concentrations were determined the 36 samples were pooled two and two to 18 samples (3 ZDFfa/fa and 3 ZDFfa/+ heart tissues at each stage in the development of T2DM). Label free proteomics analyses were carried out essentially as previously described (21). In short, proteins from homogenized heart tissue were subjected to separation by SDS-PAGE (Criterion™ TGX™ Any-kD gel, 220 V, 5 °C, Bio-Rad Laboratories). Gel lanes were cut into 10 equally sized pieces, disulfide bonds reduced with tris (2-carboxyethyl)phosphine (TCEP), and thiols alkylated with iodoacetamide (IAA) before in-gel trypsin digestion of proteins (Trypsin Gold, mass spectrometry grade, Promega). Peptides were purified by Mixed Cation Exchange (MCX, Oasis, Waters) and analyzed by liquid chromatography – tandem mass spectrometry (nanoLC-MS/MS) (LTQ Orbitrap, Thermo Scientific). 2.3.2

Database searches and statistical analysis

Proteome Discoverer 1.4 (Thermo Scientific) equipped with the Mascot algorithm (www.matrixscience.com) was used for quality control with regards to number of proteins detected and mass accuracy. Furthermore, retention time coherence between runs was examined using Qual-browser (Thermo Scientific). Identification and quantification of proteins was then carried out using MaxQuant’s Andromeda algorithm version 1.4.0.2 (22) with common contaminants included in the database search. Mass spectra from ZDFfa/fa and ZDFfa/+ heart tissue were searched against a Rattus Norvegicus sequence database (.fasta file obtained



Page 6 of 31

from UniProt, 7964 entries, September 2016). Carbamidomethylation (CAM) of cysteines was set as static modification and oxidation of methionine as variable modification. Peptide and protein false discovery rate (FDR) of identification were both set at 0.01. Common contaminants were excluded from further analysis. For each analyzed proteome sample, protein intensities were normalized to the mean protein intensity of the respective dataset to adjust for possible variation between analyses. Only proteins with at least two unique peptides were included in further analyses. Missing values were imputed by random insertion of a value between zero and the minimum intensity measured for a protein in each sample. Additionally, proteins were excluded if the fold change was less than two times the global standard error of the mean (GSEM), calculated from the variance of the entire data set, before applying an unpaired, two-tailed, student´s t-test and correction for multiple testing (Benjamini-Hochberg, 5% FDR (BH-FDR)). The mass spectrometry proteomics data from Study 1 have been deposited to the ProteomeXchange Consortium via the PRIDE (23) partner repository with the dataset identifier PXD009555.

2.4 STUDY 2 (TMT-10PLEX LABELING, ONSET AND LATE T2DM (12 AND 24 WEEKS) Excised heart tissue samples from twenty of the same ZDF rats as in study 1 (5 ZDFfa/fa and 5 ZDFfa/+ at each stage) were analyzed separately by TMT-based proteomics. Tissue samples were homogenized by mixer milling (Precellys Evolution, Bertin Instruments) in lysis buffer as described in previous section with a minor modification (HEPES instead of Tris). Samples of 60 µg protein, measured by BCA-assay (Thermo Scientific), were in-solution trypsin digested and TMT-labeled according to manufacturer’s protocol (TMT10plex, Thermo Scientific). Labeled samples were pooled and subjected to strong cation exchange (SCX) purification (Strata SCX, 55µM, 70A, 100mg/mL, Phenomenex) followed by vacuum centrifugation and isoelectric focusing (IEF) in a Immobiline DryStrip gel strip (GE Healthcare, Immobiline DryStrip pH 3-10, 18 cm) as previously described (24). The strip was then cut in 10 equally sized pieces and peptides were extracted for further C18-spin column (8 mg C18 resin, Pierce) purification according to manufacturer’s instructions.


Page 7 of 31 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


2.4.1

NanoLC-MS/MS

Each of the 10 purified IEF fractions were injected twice on NanoLC (Easy-nLC 1000, Thermo Scientific)MS/MS (Q-Exactive Plus, Thermo Scientific, Bremen). Pre-column (Acclaim PepMap 100, 75µm x 2 cm, Nanoviper, Thermo Scientific) and analytical column (EASY-Spray column, PepMap RSLC C18, 2µm, 100 Å, 75 µM x 25 cm) was used to trap and separate peptides using a 180 minute gradient (4-40 % Acetonitrile, 0.1 % Formic acid) followed by wash and equilibration. All spectra were collected in positive mode using higher energy collisional dissociation (HCD) with a stepped normalized collision energy (NCE) ranging from 31-34%. In full scan (MS1) resolution was set at 70,000 (at 200 m/z) and AGC target set at 1x106 with scan range between 391-1500 m/z. Data dependent analysis (DDA) was applied to fragment up to 10 of the most intense peaks in MS1. Resolution for fragment scans (MS2) was set at 35,000 with first fixed mass at 121 m/z and AGC target at 2x105. Dynamic exclusion was set at 18 seconds and unassigned and +1 charge states were excluded. Furthermore, peptides with more than 8 peptide spectrum matches (PSMs) in the first analysis were excluded from the second analysis. Qual-browser and Proteome Discoverer 2.1 (Mascot) was used for quality control of all raw files for retention time coherence and mass accuracy before further data processing.

2.4.2

Database search and statistical analysis

Final database search was conducted in Proteome Discover 2.1 (Thermo Scientific) with Mascot (Matrix Science) on all raw files merged for each stage. Swiss-Prot was used as database with maximum two missed cleavages using trypsin as enzyme and taxonomy was set for Rattus Norvegicus (September 2016). Precursor and fragment mass tolerance was set at 10 ppm and 20 mmu, respectively. Oxidation of methionine was set as dynamic modification and static modifications were CAM of cysteines and TMT-labels on lysine and Nterminus. In the quantification node unique and razor peptides were included. Furthermore, correction factors from each TMT kit was added and co-isolation threshold set at 45 %. Within the percolator node, strict FDR of identification was set to 0.01. Output data files (.csv) from the Mascot database search was imported into R version 3.4.2 (25) with R-studio interface. Proteins with a minimum of two unique peptides and four



Page 8 of 31

quantifiable PSMs were included in further statistical analysis. TMT ratios were normalized to the median ratio of each TMT channel. Proteins with a fold change > 2 x GSEM were log-transformed before applying of an unpaired student’s t-test between ZDFfa/fa and ZDFfa/+ mean protein intensities and correction for multiple testing (BH-FDR). The mass spectrometry proteomics data from Study 2 have been deposited to the ProteomeXchange Consortium via the PRIDE (23) partner repository with the dataset identifiers PXD009538 and PXD009554.

2.5 PATHWAY ANALYSIS The database for annotation, visualization and integrated discovery (DAVID, version 6.8) was used for functional annotation cluster analysis (26, 27). Proteins, from the pre-diabetic samples, with a p-value 2x GSEM were considered nominally differentially abundant proteins (nDAPs). At onset and late stages of T2DM, proteins were considered nDAPs when they passed fold change criterion of > 2x GSEM together with a p-value < 0.05 in either study 1 or 2, and p-value < 0.1 in the opposite study. nDAPs were then subjected to functional annotation cluster analysis. The total list of identified proteins at each stage was used as background proteome. In DAVID, default settings were applied. All proteins in the text will be referred to by its gene name in capital letters.

3 RESULTS AND DISCUSSION 3.1 ANIMAL CHARACTERISTICS AND BIOCHEMICAL PROFILE We measured animal characteristics and biochemical profiles with respect to variables relevant for T2DM pathogenesis (Table 1). ZDFfa/fa-rats had an increased body weight at 12 weeks, compared with controls, but not at 6 or 24 weeks of age. Blood-glucose, plasma-cholesterol, and plasma-triglyceride concentrations were significantly higher in ZDFfa/fa-rats than in their age matched ZDFfa/+ controls and the amount increased over time for all three variables. Serum-free fatty acid levels were significantly increased in ZDFfa/fa-rats at 6 and


Page 9 of 31 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


12 weeks however not at 24 weeks. Furthermore, published data of the same ZDF rat model describe significantly elevated insulin levels at 6 and 12 weeks, but not at 24 weeks, for ZDFfa/fa rats compared to their age matched lean controls (19).

Table 1. Animal characteristics and biochemical profile measured in fasting blood (B) and serum (S). The values display mean ± standard deviation (SD). Level of significance is * = p < 0.01, comparing the difference between ZDFfa/fa and ZDFfa/+ with an unpaired student´s t-test within each stage.

Pre-diabetic (6 weeks)

Onset T2DM (12 weeks)

Late T2DM (24 weeks)

fa/fa

ZDF (n=6)

ZDFfa/+ (n=6)

ZDFfa/fa (n=6)

ZDFfa/+ (n=6)

ZDFfa/fa (n=5)

ZDFfa/+ (n=6)

Body weight (g)

193.0 ± 10.7

193.5 ± 6.8

397.5 ± 17.1*

320.2 ± 8.5

405.8 ± 21.3

432.5 ± 24.4

B-Glucose (mmol/L)

12.11 ± 3.90* 6.60 ± 0.63

19.77 ± 2.67*

10.39 ± 1.71

31.21 ± 6.91*

12.37 ± 1.81

S-total Cholesterol (mmol/L)

2.45 ± 0.04*

1.99 ± 0.16

3.58 ± 0.18*

1.79 ± 0.10

8.92 ±1.63*

2.33 ± 0.10

S-Triglycerides (mmol/L)

2.80 ± 0.74*

0.54 ± 0.07

7.38 ± 1.18*

0.59 ± 0.13

13.95 ± 3.90*

0.84 ± 0.17

S-Free fatty acids (mmol/L)

5.58 ± 3.16*

1.66 ± 0.61

6.26 ± 3.34*

1.13 ± 0.48

8.43 ± 8.52

2.68 ± 0.69

3.2 PROTEOMICS To assess proteome alterations in myocardial tissue from ZDF-rats during development of T2DM, we employed quantitative proteomics using label-free (Study 1) and TMT labeling (Study 2). Male ZDF rats with homozygous mutation in the leptin receptor gene (ZDFfa/fa) gradually develops insulin resistance, hyperglycemia, and diabetes (28). To identify molecular metabolic pathways affected by the progression of T2DM, we identified nominally differentially abundant proteins (nDAPs) (Figure 1A and B) and performed functional annotation cluster analyses (Figure 1 C).



Page 10 of 31

Figure 1. The bioinformatics study design and results at the pre-diabetic state (6 weeks), onset T2DM (12 weeks), and late T2DM (24 weeks). A) at each stage a volcano plot was created to visualize the distribution of identified and quantified proteins. Fold change was calculated from the mean normalized protein intensity of ZDFfa/fa samples divided by the mean normalized protein intensity of ZDFfa/+ samples. The Volcano plots depict the log2 fold change against -log10 of the p-value at 6, 12, and 24 weeks. Horizontal dotted line indicates cut-off p-value at 0.05. Vertical dotted lines denote the fold change threshold (2 x global standard error of the mean). B) Light blue circles in the Venn diagram depict amount of quantitated proteins, from each study, out of which 99 proteins were regarded as nDAPs at 6 weeks, 44 nDAPs at 12 weeks, and 60 nDAPs at 24 weeks. C) Functional cluster analysis and the obtained enrichment scores and number of proteins for each cluster. ES = Enrichment score, # proteins = number of proteins annotated to respective cluster.

3.2.1

Proteome alterations in pre-diabetic hearts (6 weeks)

Label-free proteomics analysis (Study 1) identified 99 nDAPs in the pre-diabetic state (6 weeks), and 57 of these were up-regulated while 16 were down-regulated. The only enriched cluster during DAVID functional clustering analysis was Keratin (Figure 1C). Complete lists of quantified proteins are in Supplementary Table S1A-C and functional clusters, obtained through DAVID in Supplementary Table S2A-C. The effects of pre-diabetes on the cardiac proteome have previously only been sparsely addressed. Cruz-Topete et al.


Page 11 of 31 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


used C57BL/6J mice to assess protein changes at the pre-diabetic state and showed dysregulation in proteins of energy metabolism, Krebs cycle, fatty acid oxidation (FAO), and cardiac contractile function (15). In the present study, we could not detect enrichment in mitochondrial metabolic pathways at the pre-diabetic state, although a few altered mitochondrial proteins were differentially regulated. Seven proteins were statistically significant after correction for multiple testing (BH-FDR) (Table 2). Propionyl-CoA carboxylase beta chain (PCCB) was down-regulated. PCCB is, together with PCCA, a part of the mitochondrial propionyl-CoA carboxylase complex (PCC), which is responsible for the conversion of propionyl-CoA to methylmalonylCoA from various substrates including branched chain amino acids (BCAA) (29). The lower abundance of PCCB, and PCC, might lead to accumulation of circulating BCAA metabolites, which have been associated with prediction and increased risk of developing insulin resistance and T2DM (30). 3-ketoacyl CoA thiolase (ACAA2), which we found down-regulated, is involved in both BCAA metabolism, upstream of the PCC, and in the last step of mitochondrial FAO. Inhibition of ACAA2 have shown to shift the cardiac energy metabolism from FAO towards glucose oxidation in Sprague-Dawley rats (31). An increase in glucose oxidation might counteract possible glucotoxic effects from increased glucose levels as seen at the prediabetic state in this study (Table 1). Peroxisomal acyl-coenzyme A oxidase 1 (ACOX1) was downregulated. ACOX1 is an enzyme of the acyl-CoA oxidase family and is involved in initial step of peroxisomal FAO (32). Prostaglandin E synthase 3 (PTGES3), found down-regulated, is involved in conversion of prostaglandin endoperoxide H2 (PGH2) to prostaglandin E2 (PGE2), which is a part of arachidonic acid metabolism, and is furthermore co-regulated with/and co-chaperoning Heat shock protein 90 (HSP90) (33). Actin-related protein 2/3 complex subunit 2 (ARPC2), found up-regulated, is one of the main components of the Arp2/3 complex, essential for actin polymerization (34). We also found Tropomyosin alpha-3 chain (TPM3), which is important for muscle contraction (35), up-regulated. Tropomyosin can inhibit nucleation of filaments by competing for the same binding site as the Arp2/3-complex on actin (36). The up-regulation of ARPC2 and TPM3 may indicate perturbations in cardiac actin assembly and regulation.



Page 12 of 31

Receptor of activated protein C kinase 1 (RACK1) was up-regulated. RACK1 is an important scaffolding protein for signaling pathways, which can increase kinase activity when bound to protein kinase c (PKC), and furthermore, central for transcriptional regulation (37). In addition, eight ribosomal subunit proteins and six proteasome subunit proteins, regarded as nDAPs, were all up-regulated, indicating a higher protein turnover at this stage.

Table 2. Fold change for statistical significant differentially abundant proteins after correction for multiple testing at each stage

(Bold) and non-statistically significant proteins (Italic). Hyphen (–) denotes not detected proteins. Fold change was calculated from the mean normalized protein intensity of ZDFfa/fa samples divided by the mean normalized protein intensity of ZDFfa/+ samples. UP Acc.: UniProt accession, Description, Gene name, Pre-diabetic state, Onset T2DM, Late T2DM, FC1: Fold change calculated from Study 1. FC 2: Fold change calculated from Study 2. Proteins were sorted into functional groups within each stage. AA metabolism: Amino acid metabolism, Carbo.: Carbohydrate metabolism, FA metabolism: Fatty acid metabolism. Statistical data can be found in Supplementary Table S1A-C, S3A-F and S5A-F.

UP Acc. Description P07633 Propionyl-CoA carboxylase beta chain, mitochondrial P85970 Actin-related protein 2/3 complex subunit 2 Q63610 Tropomyosin alpha-3 chain P13437 3-ketoacyl-CoA thiolase, mitochondrial P07872 Peroxisomal acyl-coenzyme A oxidase 1 P63245 Receptor of activated protein C kinase 1 P83868 Prostaglandin E synthase 3 P97852 Peroxisomal multifunctional enzyme type 2 O55171 Acyl-coenzyme A thioesterase 2, mitochondrial P05545 Serine protease inhibitor A3K P05544 Serine protease inhibitor A3L P12007 Isovaleryl-CoA dehydrogenase, mitochondrial P0C2X9 Delta-1-pyrroline-5carboxylate dehydrogenase, mitochondrial P70584 Short/branched chain specific acyl-CoA dehydrogenase,

6 weeks 12 weeks Gene names FC 1 FC 1 FC 2 0.85 1.00 Pccb 0.75

24 weeks FC 1 FC 2 Group 0.58 0.75 AA metabolism

Arpc2

1.85

0.95

1.15

1.61

0.96

Actin associated

Tpm3 Acaa2

1.76 0.72

0.71

0.99

1.42

0.98

1.16

1.06

1.73

1.69

Actin associated FA metabolism

Acox1

0.51

1.92

1.27

1.90

1.40

FA metabolism

Rack1

1.81

1.17

0.96

1.09

0.92

Miscellaneous

Ptges3 Hsd17b4

0.04

1.04

1.18

1.13

0.99

0.69

2.31

1.19

1.91

1.49

Miscellaneous FA metabolism

Acot2

0.82

1.46

1.30

3.70

2.60

FA metabolism

Serpina3k Serpina3l Ivd

0.50 0.52

0.30 0.25

0.55 0.54

-

0.88

1.07

0.07 0.05 0.38

0.39 0.28 0.53

Miscellaneous Miscellaneous AA metabolism

Aldh4a1

0.73

0.86

0.96

0.37

0.61

AA metabolism

Acadsb

-

0.82

0.92

0.35

0.59

AA metabolism


Page 13 of 31 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


mitochondrial Maleylacetoacetate isomerase Branched-chain-amino-acid aminotransferase, mitochondrial Q8CIN7 Inositol monophosphatase 2 P15429 Beta-enolase P27881 Hexokinase-2 P04639 Apolipoprotein AI;Proapolipoprotein A-I Q64591 2,4-dienoyl-CoA reductase, mitochondrial Q64428 Trifunctional enzyme subunit alpha, mitochondrial P23965 Enoyl-CoA delta isomerase 1, mitochondrial Q60587 Trifunctional enzyme subunit beta, mitochondrial P15650 Long-chain specific acyl-CoA dehydrogenase, mitochondrial B2GV06 Succinyl-CoA:3-ketoacid coenzyme A transferase 1, mitochondrial P29147 D-beta-hydroxybutyrate dehydrogenase, mitochondrial P07335 Creatine kinase B-type P57113 O35854

3.2.2

0.31 0.27

0.58 0.51

AA metabolism AA metabolism

1.47

6.45 0.56 0.07 5.55

1.47 0.68 0.58 4.30

Carbo. Carbo. Carbo. FA metabolism

1.42

1.42

5.25

3.32

FA metabolism

-

1.11

1.19

2.01

1.66

FA metabolism

Eci1

-

1.16

1.23

1.98

1.84

FA metabolism

Hadhb

-

1.22

1.17

1.82

1.51

FA metabolism

Acadl

-

1.21

1.14

1.35

1.27

FA metabolism

Oxct1

-

0.71

0.92

0.53

0.62

Ketone body degradation

Bdh1

-

0.94

0.77

0.25

0.51

1.41

0.95

1.01

0.59

0.55

Ketone body degradation Miscellaneous

Gstz1 Bcat2

-

1.04

0.90

-

0.79

0.96

Impa2 Eno3 Hk2 Apoa1

0.70

0.93

1.00

1.55

0.92

1.12

-

0.60

0.80

1.41

1.64

Decr1

-

Hadha

Ckb

Proteome alterations at onset of T2DM (12 weeks)

In Study 1 and Study 2, 815 and 1618 proteins, respectively, were identified and quantified, of which 786 proteins were common in both studies (Figure 1B, 12 weeks). After Benjamini-Hochberg correction for multiple testing (BH-FDR), four proteins were statistically significant; Peroxisomal multifunctional enzyme type 2 (HSD17B4) and Acyl-coenzyme A thioesterase 2 (ACOT2) were up-regulated, and Serine protease inhibitor A3K (SERPINA3K) and Serine protease inhibitor A3L (SERPINA3L) down-regulated. Fasting, streptozotocin-induced diabetes, and peroxisome proliferator-activated receptor (PPAR) alpha agonists have all been found to upregulate ACOT2 (38). Increased levels of ACOT2 have also been found in high fat diet induced Wistar rat hearts, implicating its importance in protection of the heart from mitochondrial stress during times of high β-oxidation flux due to prolonged high fat exposure (39). HSD17B4 is an NAD+ dependent enzyme involved in peroxisomal FAO (40). Peroxisomal dysregulation has been implicated in insulin resistance, T2DM, and other age related diseases (41).



Page 14 of 31

SERPINA3K and SERPINA3L, found down-regulated, are both members of the Serpin superfamily clade A, member 3; α-proteinase inhibitors (α1-antichymotrypsin) (42). Members of this superfamily have recently been shown to be transcriptional targets of leptin action in leptin deficient ob/ob mice (43). SERPINA3K (Kallikrein-binding protein), has been given focus because of its involvement in diabetic retinopathy. In streptozotocin diabetes induced Sprague-Dawley rats the protein was found down-regulated, implicating a contribution to the pathology of diabetic retinopathy (44). Furthermore, SERPINA3K rat liver mRNA levels are decreased in a state of acute inflammation (45) and decreased protein levels in spontaneous hypertensive rats (46). In a rat model inducing myocardial infarction (MI), SERPINA3N was characterized as downregulated a factor two, 14 days post MI insult (47). mRNA data of the human paralogue SERPINA3 has described that decreased levels could be a biomarker of right ventricular myocardial heart failure (48) and finally, gene expression of SERPINA3 has been found down-regulated in failing human hearts (49). The SERPINAs possible mechanistic contributions to the pathology of CVD and T2DM are still to be elucidated. 3.2.3

Functional cluster analysis

Forty-four nDAPs were used for functional cluster analysis, which revealed four statistical significant clusters. Starch and sucrose metabolism was the most enriched cluster (enrichment score = 1.95) at 12 weeks followed by Peroxisomes, NAD(P)-binding domains, and Oxidoreductase activity (Figure 1C, 12 weeks). Quantified proteins and DAVID functional annotation cluster analysis are found in Supplementary Table S3A-F and S4A-C, respectively. In the next sections, clusters and nDAPs with at least p

Proteomics of the rat myocardium during development of type 2

Recommend Documents