PPARs are ligand-activated nuclear transcription factors that regulate β-oxidation of fatty acids in the cardiovascular system and PPAR α isoform is a putative target for regulation of cardiovascular function. High salt diet is an injurious stimulus to cardiovascular function but its effect on PPAR α and PPAR α–associated profile of proteins is unknown. Quantitative proteomics involving a two-dimensional electrophoresis (2D-DIGE) followed by LC-MS/MS technology was used to characterize the changes in protein expression profile in the kidney, heart, and blood vessels from PPAR α null (KO) and wild type (WT) mice placed on normal (0.3%, NS) or high salt (4% NaCl, HS) diet. Initial biological variation analysis using DeCyder software (v. 6.0) revealed the presence of 20 upregulated proteins and 9 proteins that are downregulated in the kidney, aorta, and heart tissues from KO and WT mice. A multimodality comparison of the differentially expressed proteins showing ≥ 1.5-fold change, ≥20% appearance at P ≤ 0.05 between strains (WT vs KO) and treatment (NS vs HS) revealed that HS diet affected 20 proteins in WT mice and 17 proteins in KO mice. However, 9 proteins were altered between WT and KO placed on NS and 7 proteins were altered by HS between WT and KO mice. The identified proteins include but not limited to those involved in fatty acid oxidation (FAO), mitochondrial electron transport chain, amino acid metabolism, stress response, DNA synthesis, and programmed cell death. HS diet led to upregulation of FAO enzymes viz: acyl-coenzyme A dehydrogenase, transketolase, and electron-transferring-flavoprotein dehydrogenase to different extents in WT and KO mice. These data showed differential and protein-specific responses to HS diet in PPAR α WT and KO mice that probably reflect the functional capacities of PPAR α as a means to limiting any salt-induced injury to the heart, kidney, and blood vessels.
Peroxisome proliferator-activated receptors (PPARs) are ligand-dependent nuclear transcription factors that form a subfamily of the nuclear receptor belonging to the steroid-thyroid hormone superfamily. PPARs comprise 3 isoforms, viz., α, β/δ, and γ that exhibit tissue-specific distribution and ligand-specific effects [
In animal cells, mitochondria and peroxisomes oxidize fatty acids via β-oxidation, with long chain fatty acids (LCFAs) and very long chain fatty acids (VLCFAs) being preferentially oxidized by peroxisomes [
The above information is consistent with emerging data that demonstrated physiologic and pathological roles for PPARα in tissues that depend on oxidation of fatty acids for energy production (see [
This study was approved by the Animal Care Committee of the Texas Southern University and conforms to the institutional guidelines on animal care and use. Age- and weight-matched adult (18 - 22 gm) male PPARα knockout (−/−; KO) mice (Jackson Laboratory, Bar Harbor, ME) or their wild type (129S1/Sv; +/+; WT) littermates were used for the study. The animals were maintained in a 12:12-hr light:dark cycle at 23˚C - 25˚C. They were randomly divided into groups (n = 5 - 7 per group) and assigned to normal rat chow (0.3% NaCl diet [normal salt (NS) diet], Harlan Tekland, Houston, USA) or high salt (HS) (4.0% NaCl) diet. These resulted in the following experimental groups:
Group 1 PPARα KO mice (n = 5) placed on normal salt (0.3% NaCl) diet
Group 2 PPARα KO mice (n = 6) placed on high salt (4% NaCl) diet
Group 3 PPARα WT mice (n = 6) placed on normal salt (0.3% NaCl) diet
Group 4 PPARα WT mice (n = 5) placed on high salt (4% NaCl) diet
Tap water was provided ad libitum and the mice were placed on the respective diets for 3 weeks after which they were anesthetized (sodium pentobarbital, 60 mg/kg i.p.). Tissues including the kidney, the heart and the blood vessels were collected from each animal and stored at −80˚C until subjected to proteomic analysis. Unless specified otherwise in the text, all chemicals used in this study were obtained from Sigma-Aldrich (St Louis, MO, USA) and are of the highest analytical grade.
A comparative proteomics approach using the two-dimensional difference in gel electrophoresis (2D-DIGE) technology was adopted to identify proteins that showed differential expression as a function of the strains of the mice and salt content of the diet. Briefly, total protein was isolated from each tissue using the protocol described in ToPI-DIGE kit protein isolation kit (ITSI Biosciences, Johnstown, PA). The ToPI-DIGE lysis buffer contained 7 M urea, 2 M thiourea, 30 mM TRIS, 4% CHAPS, 5 mM magnesium acetate tetrahydrate and 1% Nonylphenylpolyethylene Glycol (NP40). Proteins were quantified with the Total Protein Assay kit (ToPA) a Bradford-based protein assay reagent (ToPA, ITSI Biosciences, Johnstown, PA) using bovine serum albumin (BSA) as standard. 50 µg of each sample was labeled with 200 pmol of either Cy3 or Cy5 fluorescent dye and an equal amount of each tissue sample was combined in a tube and labeled with Cy2 dye at the same concentration as above. The Cy2 labeled sample served as the universal internal standard to allow normalization and multiple gel comparisons. Equal amounts of the Cy3, Cy5 and Cy2 labeled samples were mixed and subjected to 2D-DIGE as previously described (Somiari et al., 2003). For iso electric focusing (IEF), the labeled samples were loaded on a 24 cm, pH 3 - 11 NL immobilized pH gradient (IPG) strips. The strips were re-hydrated for 12 hours with the sample buffer consisting of 7 M urea, 2 M thiourea, 30 mM TRIS, 4% CHAPS, 5 mM magnesium acetate tetrahydrate and 1% NP40 and focused for 65,000 vhrs. Following IEF, the strips were equilibrated in SDS equilibration buffer (6 M Urea, 30% glycerol, 15 mM TRIS and 2%) and placed on a 24 cm × 20 cm SDS- PAGE gel (12.5%) for second dimension separation. The second dimension separation was carried out at 15 watts/gel for approximately 4.5 hours. The 2D gels were scanned with the Typhoon Trio Variable Mode Imager (GE Healthcare) at there wavelengths to generate Cy2, Cy3 and Cy5 signals. Images were imported into DeCyder version 6.0 (GE Healthcare) and analyzed with the Difference In-gel Analysis (DIA) and Biological Variation Analysis (BVA) modules. Candidate protein spots on the gel that showed statistically significant difference in abundance between the “test” e.g. Cy3 labeled sample and “reference” e.g. Cy5 labeled sample were automatically identified and selected for picking and identification by tandem mass spectrometry.
The candidate protein spots of interest were picked from preparatory gels with a spot-picking robot (Ettan Spot Picker, GE Healthcare) and in-gel digested using the Ettan Digester (GE Healthcare) proteomics grade trypsin (Promega, Madison, WI) as previously described [
Multimodality analysis was used to identify proteins that were expressed in the kidney, heart, or blood vessels of WT and KO mice. Criteria for differential expression were a P < 0.05, a difference > 1.5-fold, and ≥20% appearance of the spots on the gel.
Proteomic profiles were compared in the kidney, heart, and aorta of PPARα WT and KO mice fed 0.3% NaCl (NS) or 4% NaCl (HS) diet for 3 weeks. A summary of the throughput of the proteomic analysis is shown in
Spot # | Description | Mass | PI | Accession # | GenBank # |
---|---|---|---|---|---|
646 | Transketolase | 71,142 | 8.33 | 12018252 | NM_022592 |
Acyl-Coenzyme A dehydrogenase, very long chain | 70,706 | 10.1 | 6978435 | NM_012891 | |
648 | Acyl-Coenzyme A dehydrogenase, very long chain | 70,706 | 10.1 | 6978435 | NM_012891 |
Programmed cell death 8 | 66,682 | 10.1 | 25742626 | NM_031356 | |
672 | Stress-induced-phosphoprotein 1 (Hsp70/Hsp90-organizing protein) | 62,531 | 6.0 | 20302113 | NM_138911 |
Electron-transferring-flavoprotein dehydrogenase | 68,122 | 7.17 | 52138635 | NM_198742 | |
5-aminoimidazole-4-carboxamide ribonucleotide formyltransferase | 64,169 | 6.0 | 48675845 | NM_031014 | |
Chaperonin containing TCP1, subunit 3 (gamma) | 60,609 | 6.0 | 40018616 | NM_199091 | |
678 | 5-aminoimidazole-4-carboxamide ribonucleotide formyltransferase | 64,169 | 6.0 | 48675845 | NM_031014 |
Electron-transferring-flavoprotein dehydrogenase | 68,122 | 7.17 | 52138635 | NM_198742 | |
Malic Enzyme | 65,524 | 6.0 | 7106353 | NM_012600 | |
Predicted: similar to keratin 6 alpha | 63,674 | 6.0 | 62653090 | XM_576346 | |
684 | Acyl-Coenzyme A dehydrogenase, very long chain | 70,706 | 10.1 | 6978435 | NM_012891 |
740 | Alpha isoform of regulatory subunit A, protein phosphatase 2 | 65,282 | 4.25 | 55926139 | NM_057140 |
Predicted: similar to Actin, cytoplasmic (gamma-actin) | 41,767 | 4.25 | 62645364 | XP_213540 | |
Serum deprivation response protein | 46,359 | 4.25 | 56090257 | NM_001007712 | |
Similar to RIKEN cDNA 4732495G21 gene | 41,937 | 4.25 | 27687455 | XM_226755 | |
828 | Prolyl 4-hydroxylase, beta polypeptide | 56,830 | 4.25 | 6981324 | NM_012998 |
883 | Predicted: similar to sarcalumenin | 54,400 | 6.0 | 109487785 | XM_220171 |
Selenium binding protein | 52,500 | 6.0 | 18266692 | NM_080892 | |
Peptidase (mitochondrial processing) alpha | 58,119 | 6.0 | 54234052 | NM_001003673 | |
1027 | Fibrinogen, gamma polypeptide | 49,621 | 5.13 | 61098186 | NM_012559 |
Predicted: similar to Actin, cytoplasmic (gamma-actin) | 41,767 | 4.25 | 62645364 | XM_213540 | |
1591 | Predicted: similar to FATZ related protein 2 | 29,772 | 7.17 | 27695760 | XM_215692 |
1746 | Predicted: similar to NADH-ubiquinone oxidoreductase 30 kD subunit | 30,209 | 6.0 | 27702072 | XM_215776 |
1764 | Enoyl Coenzyme A hydratase, short chain, 1, mitochondrial | 31,497 | 8.33 | 17530977 | NM_078623 |
Quinoid dihydropteridine reductase | 25,537 | 8.33 | 11693160 | NM_022390 | |
Predicted: similar to fumarylacetoacetate hydrolase domain | 24,466 | 8.33 | 68163417 | NM_001024991 | |
1840 | NADH dehydrogenase (ubiquinone) flavoprotein 2 | 27,362 | 6.0 | 51092268 | NM_031064 |
1855 | Predicted: similar to tumor protein, translationally-controlled | 19,526 | 4.25 | 62653000 | XM_576332 |
The table depicts the particulars of proteins identified by using MS/MS spectra searched against NCBI proteins sequence data base based on the SEQUEST computer algorithm. The master spot numbers that showed significance on the gels, the name of the protein, the mass, isoelectric point (PI), are accession # and GenBank # are also shown.
Based on the criteria of ≥1.5-fold change in expression, ≥20% appearance on the gel, and a P value ≤ 0.05, at least 24 proteins were positively identified from 14 master spots (
When the expression of the different proteins was compared in the kidney, heart, and aorta of PPARα WT mice placed on high or normal salt diet, there was a significant difference in the expression in 9 of the 14 master spots,
When the expression of the different proteins was compared in the kidney, heart, and aorta of PPARα KO mice placed on high versus normal salt diet,
Comparison of the effect of high salt diet on proteomic profile in PPARα KO and WT mice
When the expression profile of proteins was compared between tissues collected from PPARα KO and WT mice placed on high salt diet,
Spot No | Protein(s) | Ratio (KO:WT) |
---|---|---|
646 | Acyl-Coenzyme A dehydrogenase, very long chain# Transketolase# | −2.41 |
648 | Acyl-Coenzyme A dehydrogenase, very long chain# Programmed cell death 8 | −1.53 |
672 | Stress-induced-phosphoprotein1 (HSP70/90) Electron transferring-flavoprotein dehydrogenase# 5-aminoimidazole-4-carboxamide ribonucleotideformyltransferase# Chaperonin containing TCP-1, subunit 3 (gamma) | 1.68 |
678 | 5-aminoimidazole-4-carboxamide ribonucleotideformyltransferase# Electron-transferring-flavoprotein dehydrogenase# Malic enzyme (Malate dehydrogenase)# Predicted: similar to keratin 6 alpha | 1.95 |
Spots that showed significant differences in protein levels in the gel samples in tissues harvested from WT or KO mice placed on 0.3% NaCl (normal salt) diet. The average ratio of the levels of the proteins in KO relative to WT mice is presented: + indicates increase; − indicates decrease; # indicates proteins that are directly involved in fatty acid oxidation.
Spot No | Protein(s) | Ratio (HS:NS) |
---|---|---|
672 | Stress-induced-phosphoprotein1 (HSP70/90) Electron transferring-flavoprotein dehydrogenase# 5-aminoimidazole-4-carboxamide ribonucleotide formyltransferase# Chaperonin containing TCP-1, subunit 3 (gamma) | 1.55 |
740 | α isoform of regulatory subunit A, protein phosphatase 2 Predicted: Similar to actin, cytoplasmic (γ-actin) Serum deprivation response protein Similar to RIKEN cDNA 4732495G21 | −1.62 |
828 | Prolyl 4-hydroxylase, β polypeptide | 3.07 |
883 | Predicted: Similar to sarcalumenin Selenium binding protein Peptidase (mitochondrial processing) α | 5.21 |
1027 | Fibrinogen, γ polypeptide Predicted: Similar to actin, cytoplasmic (γ-actin) | 2.63 |
1746 | Predicted: Similar to NADPH-ubiquinone oxidoreductase 30 kD subunit# | 2.03 |
1764 | Enoyl Coenzyme A hydratase, short chain, mitochondrial# Quinoid dihydropteridine reductase Predicted: Similar to fumarylacetoacetate hydrolase domain | 3.6 |
1840 | NADH dehydrogenase (ubiquinone) flavoprotein 2# | 2.86 |
1855 | Predicted: Similar to tumor protein, translationally controlled | 1.8 |
The spots that showed significant differences in protein levels in tissues harvested from WT mice were placed on 0.3 % (normal, NS) or 4% NaCl (high salt, HS) diet. The average ratio of the levels of the protein in mice placed on HS diet is presented: + indicates increase; − indicates decrease; # indicates proteins that are directly involved in fatty acid oxidation.
Spot No | Protein(s) | Average Ratio (HS:NS) |
---|---|---|
646 | Acyl-Coenzyme A dehydrogenase, very long chain# Transketolase# | 2.35 |
648 | Acyl-Coenzyme A dehydrogenase, very long chain# Programmed cell death 8 | 2.07 |
678 | 5-aminoimidazole-4-carboxamide ribonucleotide formyltransferase# Electron-transferring-flavoprotein dehydrogenase# Malic enzyme (Malate dehydrogenase)# Predicted: similar to keratin 6 alpha | −1.86 |
684 | Acyl-Coenzyme A dehydrogenase, very long chain# | 2.43 |
740 | Alpha isoform of regulatory subunit A, protein phosphatase 2 Predicted: Similar to actin, cytoplasmic (γ-actin) Serum deprivation response protein Similar to RIKEN cDNA 4732495G21 | 2.34 |
828 | Prolyl-4-hydroxylase, β polypeptide | 4.32 |
1027 | Fibrinogen γ polypeptide | 2.87 |
1591 | Predicted: Similar to FATZ related protein 2 | −2.5 |
1746 | Predicted: similar to NADH-ubiquinone oxidoreductase 30 kD subunit# | 2.99 |
1840 | NADH dehydrogenase (ubiquinone) flavoprotein 2# | 1.91 |
1855 | Predicted: similar to tumor protein | 1.41 |
The spots that showed significant differences in tissues from KO mice placed on 0.3% (normal, NS) or 4% NaCl (high salt, HS) diet. The average ratio of the levels of the proteins in mice placed on HS diet is presented: + indicates increase; − indicates decrease; # indicates proteins that are directly involved in fatty acid oxidation.
Spot No | Protein(s) | Ratio (WT: KO) |
---|---|---|
648 | Acyl-Coenzyme A dehydrogenase, very long chain# Programmed cell death 8 | 1.61 |
740 | Alpha isoform of regulatory subunit A, protein phosphatase 2 Predicted: Similar to actin, cytoplasmic (γ-actin) Serum deprivation response protein Similar to RIKEN cDNA 4732495G21 | 2.78 |
1591 | Predicted: Similar to FATZ related protein 2 | −1.45 |
The spots that showed significant differences in protein levels in tissues from WT or KO mice placed on 4% NaCl in (high salt, HS) diet. The average ratio of the levels of the proteins in mice placed on HS diet is presented: + indicates increase; − indicates decrease; # indicates proteins that are directly involved in fatty acid oxidation.
and 740 (2.78-fold). The only protein of significance in relation to fatty acid metabolism is acyl-Coenzyme A dehydrogenase (spot # 648).
The results of this study demonstrate potentially important interactions between high salt diet and PPARα as they affect fatty acid oxidation enzymes and other proteins involved in important cellular functions. The differentially expressed proteins include but not limited to those primarily associated with fatty acid oxidation phenotype, reflecting the pleiotropic nature of PPARα gene and its activation. Thus, apart from fatty acid oxidation, the proteins identified include those involved in such cellular functions or structures such as signal transduction, energy metabolism, and cytoskeleton. These include heat shock protein e.g. HSP70/90, mitochondrial enzymes, transcriptional factors, structural proteins such as prolyl 4-hydroxylase and others.
PPARα is well characterized in cardiovascular tissues/organs including the heart, blood vessel, and kidney. A protective role has long been established for PPARα in the heart and blood vessel [
Fatty acid oxidation was long known to be important in organs with high metabolism such as the heart, liver, blood vessels [
The present study focused on proteins that are directly associated with fatty acid oxidation. Thus, comparison between tissues from WT and KO mice placed on NS diet revealed a surprising decrease in protein levels of acyl-Coenzyme A dehydrogenase and transketolase in WT mice, an observation at odds with our hypothesis and inconsistent with an intact PPARα gene in WT mice to effect metabolism of long chain fatty acids. By contrast, there were increases in the levels of malic enzyme, EFTDH, and AICAR in KO mice as expected. Malic enzyme is involved in the oxidative decarboxylation of malate to produce pyruvate and CO2 coupled with the reduction of NAD+ or NADP+. An increase in its level and that of ETFDH in KO mice could be explained on the necessity to compensate for depleted fatty acid oxidation proteins due to lack of PPARa gene. When placed on HS diet, tissues from WT mice showed increases in protein levels of ETFDH, NADPH ubiquinone-oxidoreductase, NADH flavoprotein dehydrogenase, and enoyl coenzyme A hydratase, the enzyme involved the second step of mitochondrial beta oxidation of short chain fatty acid metabolism. Consistent with our hypothesis and the known protective role of PPARα [
The increase in FAO enzyme in response to HS diet in this study is in agreement with the findings that HS diet affected a number of proteins in the Dahl salt-sensitive rat kidney including fumarate hydratase and other proteins involved in the metabolism of fatty acids [
PPARα appears to be a salt-responsive gene controlling multiple gene targets through changes in the expression of different proteins that mitigate any possible injury to the heart, kidney, and the blood vessel. These multiple proteins and/or targets involve fatty acid oxidation, lipid and amino acid metabolism, and other cellular pathways that serve critical protective roles in the body.