Differential Proteomic Analysis of STAT6 Knockout Mice Reveals New Regulatory Function in Liver Lipid Homeostasis Joe¨l Iff,†,‡ Wei Wang,‡ Tatjana Sajic,‡ Nathalie Oudry,§ Estelle Gueneau,| Ge´rard Hopfgartner,§,| Emmanuel Varesio,*,†,§,| and Ildiko Szanto‡,⊥ Department of Cellular Physiology and Metabolism, University of Geneva, Switzerland, SVS-MS Mass Spectrometry Core Facility, University of Geneva, Switzerland, Life Sciences Mass Spectrometry Laboratory, School of Pharmaceutical Sciences, University of Geneva, University of Lausanne, Switzerland, and Department of Rehabilitation and Geriatrics, University of Geneva, Switzerland Received April 9, 2009
Increased inflammatory signaling is a key feature of metabolic disorders. In this context, the role of increased pro-inflammatory signals has been extensively studied. By contrast, no efforts have been dedicated to study the contrasting scenario: the attenuation of anti-inflammatory signals and their role in metabolic homeostasis. IL-4 and IL-13 are anti-inflammatory cytokines signaling through the Signal Transducer and Activator of Transcription 6 (STAT6). Our study was aimed at evaluating the lack of STAT6 signaling on liver homeostasis. To this end we analyzed the liver proteome of wild type and STAT6 knock-out mice using 2D nanoscale LC-MS/MS with iTRAQ labeling technique. The coordinated changes in proteins identified by this quantitative proteome analysis indicated disturbed lipid homeostasis and a state of hepatocellular stress. Most significantly, the expression of the liver fatty acid binding protein (FABP1) was increased in the knock-out mice. In line with the elevated FABP1 expression we found latent liver lipid accumulation in the STAT6-deficient mice which was further aggravated when mice were challenged by a high fat diet. In conclusion, our study revealed a so far uncharacterized role for STAT6 in regulating liver lipid homeostasis and demonstrates the importance of anti-inflammatory signaling in the defense against the development of liver steatosis. Keywords: iTRAQ • lipid homeostasis • liver • mass spectrometry • STAT6
Introduction Signal transducer and activator of transcription (STAT) proteins mediate different cytokine-induced gene transcription. STAT6 is a ubiquitously expressed member of this family transmitting interleukin 4 and 13 (IL-4 and IL-13) signaling. Upon IL-4/IL-13 receptor binding, STAT6 becomes tyrosine phosphorylated, forms homodimers and translocates into the nucleus. Once in the nucleus, STAT6 modulates gene transcription by binding to a specific palindromic DNA sequence present in the promoter regions of most of the STAT6dependent genes. In addition to its direct gene-regulatory function, STAT6 interacts with a wide variety of other transcription factors and serves as a recruitment platform for the different members of the transcriptional machinery (reviewed * To whom correspondence should be addressed. Dr. Emmanuel Varesio, Life Sciences Mass Spectrometry, School of Pharmaceutical Sciences, University of Geneva, 20, bd d’Yvoy, CH-1211 Geneva 4, Switzerland. Phone: +41 (0)22 379 6757. Fax: +41 (0)22 379 6808. E-mail: emmanuel.varesio@ unige.ch. † These authors contributed equally to this work. ‡ Department of Cellular Physiology and Metabolism. § SVS-MS Mass Spectrometry Core Facility. | School of Pharmaceutical Sciences. ⊥ Department of Rehabilitation and Geriatrics. 10.1021/pr9003272 CCC: $40.75
2009 American Chemical Society
in ref 1). Therefore, STAT6-deficient mice display complex STAT6-dependent and -independent transcriptional alterations.2 Most of the research effort to elucidate the molecular basis for the different effects of STAT6 has been attributed to its essential function in the immune system to predispose T lymphocytes toward a specific T helper type 2 (Th2) differentiation.3 Recently, however, several novel studies indicated a function for STAT6 in a variety of other cell types. Notably, STAT6 has been demonstrated to be involved in adipocyte differentiation, kidney epithelial cell mechanosensation, in apoptosis regulation in human hepatoma cells and in the inflammatory response in lung epithelial cells.4-6 In the liver in vivo, STAT6 has a protective role against ischemia/reperfusion (I/R)-induced injury.7 Liver I/R leads to abrupt changes in oxygen supply (hypoxia followed by reoxigenation) and substrate availability thus altering cellular metabolism. The close relationship between cellular oxygen supply, metabolism and liver injury was recently highlighted by studies demonstrating that intermittent hypoxia induces alterations in lipid homeostasis and also predisposes to liver injury.8-10 The adaptive metabolic response to hypoxia-induced liver injury is controlled to a large extent by different inflammatory and anti-inflammatory cytokines.11 In particular, IL-4 and IL-13 display a protective effect against I/R injury, however the Journal of Proteome Research 2009, 8, 4511–4524 4511 Published on Web 08/11/2009
research articles metabolic gene expression pattern regulated by their common effector molecule, STAT6, has not yet been investigated. Therefore, the present study was aimed at exploring the effect of ablation of IL-4 and IL-13 signaling on liver function by analyzing the proteomes of wild type and STAT6 knock-out mice using a differential proteomic method and validating the physiological relevance of the identified proteins.
Experimental Methods Animals. Balb/cJ wild type and STAT6 knock-out male mice were obtained from Charles River Laboratories (L’Arbresle, France) and were kept under regular animal housing conditions.. Mice had ad libitum access to water and standard chow (SDS Dietex, Saint Gratien, France) and were sacrified at the age of 20 weeks. In a second set of experiments wild type and STAT6 knock-out mice were fed a high fat containing diet supplying 60% of energy as fat (cat: D12108, Provimi Kliba) for a periods of ten weeks. Ob/Ob mice, Fa/Fa rats and their control littermates were purchased from Elevage Janvier (CERJ, Le Genest St Isle, France). The experimental protocols were accepted by the Ethical Committee of the University of Geneva and the Veterinary Office of the Canton of Geneva. All experiments were carried out in accordance with the regulatory guidelines of the Veterinary Office of the Canton of Geneva on the care and welfare of laboratory animals. Preparation of Liver Samples for iTRAQ Analysis. Liver samples were lysed by freeze-cracking in presence of liquid nitrogen. Homogenization with a mechanical douncer was performed on ice by adding a Tris buffer (pH 7.4) containing 250 mM sucrose and a tablet of Complete protease inhibitor cocktail per 50 mL of buffer (Roche Diagnostics, Mannheim, Germany). The buffer volume was adjusted in order to obtain liver amounts of 50 mg per 100 µL. Organelle enrichment fractionation was performed by differential centrifugation according to Arnold et al.12 and the different fractions were stored at -80 °C. Only cytosolic fractions were used for subsequent differential analyses. iTRAQ Labeling and Two-Dimensional Nanoscale LC Tandem Mass Spectrometry. Protein concentration of wild type or STAT6 knock-out cytosolic enriched fractions was determined by the Coomassie Plus Bradford assay (Pierce, Rockford, IL). Differential labeling was performed using iTRAQ reagents (Applied Biosystems) according to the method described by Ross et al.13 In brief, ca. 20 µL of 500 mM triethylammonium bicarbonate (TEAB) buffer (pH 8.5) and 1 µL of 2% SDS were added to each sample in order to obtain a protein concentration of 5 µg/µL. Proteins were then reduced by the addition of 2 µL of 50 mM Tris(2-carboxyethyl)phosphine (TCEP) followed by an incubation at 60 °C for one hour. At room temperature, proteins were alkylated with 1 µL of 200 mM methyl methane-thiosulfate (MMTS) reagent. Overnight digestion (37 °C) was performed by adding 10 µL of a 1 mg/mL trypsin aqueous solution. Finally iTRAQ differential labeling was carried out by adding separately 70 µL of reagent to the respective tubes containing the pooled wild type or pooled STAT6-KO mice liver cytosolic fractions. After an incubation of one hour at room temperature, labeled samples were combined and evaporated to a final volume of ca. 200 µL. Sample pH was adjusted to 3.0 with a 10% (v/v) aqueous formic acid solution before injection in the two-dimensional nanoscale LC-MS/MS (2D-nLC-MS/MS) system. Analyses were performed according to the method described by Varesio et al.14 and adapted as follows: 5 µL of sample was injected 4512
Journal of Proteome Research • Vol. 8, No. 10, 2009
Iff et al. and cation exchange chromatography was carried out on a 300 µm ID × 15 cm Poros 10S SCX column (packed by LC Packings-Dionex) at a flow rate of 6 µL/min. A step gradient of 20 KCl fractions ranging from 0 to 300 mM in 10 mM sodium phosphate buffer (pH 3.0)/acetonitrile (95:5, v/v) was performed over 24 h. Each salt fraction containing peptides eluted from the SCX column was trapped onto a PepMap C18 300 µm ID × 5 mm cartridge and washed at 20 µL/min for 10 min with a 0.1% aqueous TFA solution to remove salts from the SCX mobile phase. Then, the cartridge was backflushed in line with a C18 nanocolumn (PepMap, 75 µm ID × 15 cm, LC Packings-Dionex) at a flow rate of 300 nL/min and peptides were separated over a 30 min generic reverse-phase LC gradient (i.e., from 0-75% of 0.1% formic acid in acetonitrile, 70 min total runtime). In the meantime the following SCX step was performed and peptides were retained on the second C18 cartridge mounted in parallel on the 10-ports switching valve (Switchos II, LC Packings-Dionex). Mass spectrometry was performed on a QSTAR XL (AB/MDS Sciex) operating in information-dependent acquisition mode with two product ions scans per MS survey scan. The sample was run in triplicate with an exclusion list built from the peptides identified from the previous runs.15 Peak list generation and protein identification were carried out by searching the Uniref 100 database (release 7.4 - 3 334 551 sequences) using the ProteinPilot software (v.1.0 - Applied Biosystems). The following parameters were applied: no species restriction, trypsin digestion agent, cysteine residues alkylated by MMTS as fixed modification, search effort was set as thorough ID for the Paragon search algorithm.16 Results were grouped by the ProGroup algorithm (Applied Biosystems) within the ProteinPilot software17,18 to group proteins sharing the same set of identified peptides. Protein reporting was based on at least two peptides identified with a confidence level higher than 95% leading to a detected protein threshold (i.e., Unused ProtScore) greater than 1.3 which corresponds to a confidence level of 95% for the protein identification. Only proteins related to rodent species were reported. Proteins isoforms or members of a protein family sharing the same set of peptides were all reported as being equal hits. The estimation of false-positive identification rate was performed by searching the same data set against a decoy database made of random sequences generated by the “decoy.pl” script available on the Matrix Science Web site (www.matrixscience.com). A false positive rate of 0.55% was calculated for rodent proteins. Quantification was performed by taking the peak areas ratio with a correction for the overlapping isotopic contributions from the different reporter ions according to manufacturer’s certificate. Experimental labeling bias was also corrected by the software. Peptides used for quantification were selected with the following criteria: the sum of reporter ion areas should be greater than 40 counts, peptides with a ratio of 0 or 9999 or peptides shared by several proteins as well as overlapping precursors were excluded from quantification, only peptides with an identification confidence higher than 95% were selected for quantification. No outlier data points were removed. Results for proteins reported as upor down-regulated were manually validated. For the iTRAQ differential analysis, liver samples from wild type (n ) 3) or STAT6-KO (n ) 3) mice were pooled to obtain two average samples. These samples were then analyzed in triplicate (2D nLC-MS/MS) to assess protein expression differences in wild type and STAT6-KO mice liver samples. Biological variation was assessed at the mRNA and physiologi-
research articles
Differential Proteomic Analysis of STAT6 Knockout Mice cal levels by analyzing several mice separately (i.e., Western blots, RT-PCR and physiological measurements). The exact number of mice used in each experiment is indicated in the legends of each figure. RNA Preparation and Real-Time PCR. Total RNA was prepared by homogenizing approximately 100-200 mg liver tissue in TRIZOL Reagent (Invitrogen, Basel, Switzerland) and was purified by using RNase free DNase in combination with the RNeasy Mini Kit (Qiagen, Hombrechtikon, Switzerland). cDNA was synthesized from 2 µg of DNA-free RNA by Superscript II Reverse Transcriptase (Invitrogen). Primers and probes were designed by Primer Express software (Applied Biosystems) and are listed online as Supporting Information in Table S1. The results were quantified by the ∆∆Ct method using cyclophillin A as the standard internal nonvariable gene to compensate for differences in RNA input and efficiency of cDNA synthesis. Results were expressed as arbitrary units compared to the average expression levels in wild type mice. Western Blot Analysis. Liver tissues were snap frozen in liquid nitrogen immediately upon removal and were stored at -80 °C until processing. Tissues were homogenized in lysis buffer (25 mM HEPES, 0.5% Triton X100, 65 mM NaCl, 2.5 mM EDTA, 25 mM sodium pyrophosphate, 50 mM NaF, 2 mM PMSF, 9 mM sodium orthovanadate, one tablet of Complete Inhibitor Cocktail per 20 mL buffer (Roche Diagnostics, Rotkreuz, Switzerland), pH 7.5. Protein concentration was measured by bicinchoninic acid (BCA) method (Pierce). Lysates were dissolved in Laemmli buffer (10 mM sodium phosphate, pH 7.0, 0.1% glycerol, 2% SDS, 100 mM DTT and a trace of BPB) and were resolved on a 5-20% gradient polyacrylamide gel. Gels were transferred onto nitrocellulose membranes (GE Healthcare). Nonspecific binding of the antibody was prevented by blocking the membranes with 0.05% polyvinyl alcohol (PVA) followed by incubation with the respective primary antibodies at 4 °C overnight. Antibodies were as follows: Ezrin, SOCS3, STAT3 and STAT6 (Santa Cruz Biotechnology, Santa Cruz, CA), SOD1, GPX1, FABP1, FDPS, ACAT2, CSAD, UGDH, MTTP (Abcam, Cambridge, UK), GLO1 (BioMac, Leipzig, Germany), RGN (Cosmo Bio, Tokyo, Japan). Blots were then washed 3 times for 5 min at room temperature with TBS supplied with 0.1% Tween20, and subsequently incubated with the applicable secondary horseradish peroxidase (HRP)-conjugated antibody: goat antirabbit IgG (Bio-Rad, Reinach, Switzerland) or rabbit antigoat IgG (Sigma-Aldrich, Buchs, Switzerland). Signals were revealed by enhanced chemiluminescence (ECL Advanced Western Blot Detection Kit, GE Healthcare) and were recorded in Chemidoc XRS system (Bio-Rad). Quantification of the detected bands was performed by using the Quantity One program (Bio-Rad). Protein expression was related to the amount of ezrin as a nonvariable reference protein and was expressed as arbitrary units compared to the average expression in wild type mice. Determination of Liver Lipid Content. Liver lipids were extracted according to a modified Bligh and Dyer method.19 Briefly, liver pieces were pulverized in a mortar using liquid nitrogen then left overnight in a chloroform/methanol (2:1, v/v) extraction solution. After filtration, lipids were washed once with water, then three times using 2 mM calcium chloride in water/methanol/chloroform (48:49:3, v/v/v) with the supernatant discarded each time. Finally, lipids were air-dried and weighted to quantify total lipid amount. Histological Analysis. Liver samples were fixed in Bouin’s solution and processed for hematoxylin-eosin staining or were
frozen and processed for Oil-Red O staining using standard histological methods. Serum Ketone Bodies. Acetoacetate and BHB levels were determined in wild type and STAT6 KO mice in fed state or after overnight starving. Blood was mixed with 0.6 N perchloric acid with vigorous agitation for 30 min and levels of the ketone bodies were determined after centrifugation using the Autokit Total Ketone Bodies (Wako Chemicals, Richmond, VA). Thiobarbituric Acid Reactive Substances (TBARS) Analysis. The degree of oxidative stress was assessed by measuring TBARS in liver homogenates using a commercially available kit (ZeptoMetrix Co, Buffalo, NY). In Silico Promoter Analysis. In silico promoter analysis was performed using consensus DNA binding sequences as described by Pastorelli et al.20 Briefly, genomic sequences were downloaded from the University of California Santa Cruz (UCSC) genome browser database for the most recent (mm8) assembly. The -5000 and +1000 regions relative to the RefSeq transcriptional start sites were extracted and searched for the presence of different transcription factor binding motifs (Table 3) using custom Bioperl based scripts generously provided by P.C. Boutros (University of Toronto, Toronto, Canada). Statistical Analysis. Results were analyzed by Student’s unpaired t test using the SigmaStat software (version 2.0, SPSS, Chicago, IL). n ) 8 WT and 9 KO mice for chow fed and n ) 5 WT and 6 KO mice for high fat diet fed conditions. Results with a p value less or equal than 0.05 were considered significant.
Results Analysis of Differentially Expressed Proteins Detected by iTRAQ 2D nLC-MS/MS Analysis. Mouse liver cytosolic fractions were differentially analyzed by 2D nLC-MS/MS after iTRAQ labeling. From the 155 validated proteins with a total of 861 unique peptides identified (Table S2 in the Supporting Information), 16 down-regulated and 18 up-regulated proteins were found in STAT6 knock-out mice. The down- and upregulated proteins identified are listed in Tables 1 and 2, respectively. The differentially expressed proteins suggested disturbed cellular lipid homeostasis and the presence of hepatocellular oxidative stress in the knock-out mice. Increased Lipid Deposition in the Livers of STAT6 Knock-Out Mice. Hepatocytes fulfill a complex regulatory role in lipid metabolism. The proteome of the livers of the STAT6 knock-out mice revealed differences in the expression levels of several enzymes involved in this process. In order to gain insight into the physiological significance of the identified proteins we explored the changes in liver lipid homeostasis. One of the identified lipid homeostasis proteins was FABP1, a member of the family of the intracellular fatty acid binding proteins (FABPs) (Table 2, #2). FABPs are involved in the uptake, intracellular transport and esterification of fatty acids and their cellular content is regulated at the transcriptional level.21,22 Indeed, the results of the iTRAQ analysis were confirmed by Western blot and real-time PCR showing an upregulation of both FABP1 protein and mRNA expression in the knock-out mice (Figure 1A). The physiological relevance of the elevated expression of this fatty acid transport protein was established by direct quantification of liver lipid content which was significantly increased in the STAT6-null mice (Figure 1B, chow diet). Liver lipid accumulation is linked to obesity and prompted by high levels of circulating serum lipids which can be provoked by high fat (HF) containing nutrition. The role of Journal of Proteome Research • Vol. 8, No. 10, 2009 4513
research articles
Iff et al.
Table 1. Down-Regulated Proteins in the Livers of STAT6 Knock-Out Mice iTRAQ 2D nLC-MS/MS protein number
gene symbol
1 2 3 4
ASS1 ADH1 FBP1 ETFA
5 6
PYGL UGDH
7 8 9 10 11
FDPS TUBA6 GSTM2 ACAT3 ACAT2 GPI1
12 13 14 15 16
ACLY GOT1 CYCS SELENBP1 CSAD
protein name (accession number)
Argininosuccinate synthase (P16460/Q3UEJ7) Alcohol dehydrogenase 1 (Q3UKA4) Fructose bisphosphatase 1 (Q3UEH1) Electron transferring flavoprotein, alpha polypeptide (Q8BMU7/Q5M7W0/Q4 V9 × 5/Q3THD7) Liver glycogen phosphorylase (Q91WP9/Q3UKJ0) UDP-glucose 6-dehydrogenase (O70475/Q3TJ71/Q3TJE8/ O70199) Farnesyl pyrophosphate synthetase (Q920E5/Q3TMB3) Tubulin alpha-6 chain (P68373/Q3TIZ0/Q9JJZ2)) Glutathione S-transferase Mu 2 (P15626) Acetyl CoA transferase-like protein (Q8R4 V3/Q80 × 81) Acetyl CoA acetyltransferase, cytosolic (Q8CAY6) Glucose phosphate isomerase 1 (Q5RJI3/Q3UZJ1/Q3UUX1/ Q3TW50/Q3TEE7) ATP-citrate synthase (Q91 V92/Q3 V117/Q3TED3) Glutamate oxaloacetate transaminase 1 (P05201/Q3UJH8) Cytochrome C protein, somatic (Q56A15) Selenium-binding protein 1 (Q91 × 87/P17563) Cysteine sulfinic acid decarboxylase (Q9DBE0/Q8K566)
KO/WT ratioa mean [95% E.I] (n)
sequence coverage
unique peptides
68% 37% 48% 33%
15 9 15 6
0.87 0.73 0.81 0.87
18% 26%
10 6
0.60 [0.49-0.74] (15) 0.52 [0.41-0.66] (13)
16% 24% 73% 15%
3 3 5 2
0.76 0.67 0.64 0.67
13%
3
0.63 [0.44-0.90] (7)
5% 10% 37% 75% 17%
3 2 2 18 2
0.51 0.59 0.60 0.73 0.67
[0.78-0.97] [0.67-0.80] [0.67-0.97] [0.78-0.97]
[0.58-0.99] [0.60-0.74] [0.42-0.97] [0.55-0.81]
[0.41-0.63] [0.34-0.99] [0.44-0.81] [0.65-0.82] [0.63-0.70]
(98) (55) (48) (32)
(12) (8) (7) (7)
(6) (4) (3) (2) (2)
a 95% E.I. ) 95% confidence Error Interval ) [(Mean/Error Factor) - (Mean * Error Factor)], which means that the true protein ratio is found in this interval with a 95% confidence level.68 n ) number of peptides used for quantification.
Table 2. Up-Regulated Proteins in the Livers of STAT6 Knock-Out Mice iTRAQ 2D nLC-MS/MS protein number
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
unique peptides
KO/WT ratioa mean [95% E.I.] (n)
1.16 [1.02-1.31] (130) 1.45 [1.26-1.66] (101) 1.34 [1.21-1.48] (59) 1.66 [1.38-2.01] (50) 1.16 [1.04-1.29] (44) 1.13 [1.02-1.25] (42) 1.31 [1.13-1.52] (37) 1.37 [1.13-1.65] (16) 2.54 [2.14-3.02] (11) 1.70 [1.48-1.97] (9) 1.18 [1.07-1.31] (9) 1.14 [1.01-1.29] (8) 25.54 [n/a] (1) 15.73 [5.29-46.77] (7)b 15.73 [5.29-46.77] (7)b 15.73 [5.29-46.77] (7)b
gene symbol
protein name (accession number)
sequence coverage
CA3 FABP1 HBA-A1 GSTP1 SOD1 ACAA2 SCP2 RGN GLO1 SELENBP2 NME2 AKR1C6 MUP1a MUP1b MUP6 MUP8, MUP11 TST GPX1
Carbonic anhydrase 3 (P16015) Fatty acid-binding protein|L-FABP (P12710) Hemoglobin alpha chain (P01942/Q91 VB8/Q8BPF4/Q9CY10) Glutathione S-transferase P1 (P19157) Superoxide dismutase [Cu-Zn] (P08228) 3-ketoacyl-CoA thiolase, mitochondrial (Q8BWT1/Q3TIT9) Nonspecific lipid-transfer protein (P32020) Regucalcin|SMP 30 (Q64374) Glyoxalase 1|LGUL (Q9CPU0) Selenium binding protein 2 (Q8R1T6/Q63836) Nucleoside diphosphate kinase B (Q01768) Estradiol 17 beta-dehydrogenase 5 (P70694) Major urinary protein 1|MUP (Q4FZE8) Major urinary protein 1|MUP (Q58EV3/P11588/Q9CXU6) Major urinary protein 6|MUP (P02762) Major urinary protein 8 and 11|MUP (P04938)
74% 84% 89% 90% 79% 64% 29% 52% 47% 79% 79% 65% 38% 31% 38% 45%
16 9 6 10 6 11 8 8 3 21 5 4 3 3 3 3
Thiosulfate sulfurtransferase (Q545S0) Glutathione peroxidase 1 (P11352/Q5RJH8)
14% 45%
2 3
1.23 [1.09-1.39] (7) 1.33 [1.07-1.65] (4)
a 95% E.I. ) 95% confidence Error Interval ) [(Mean/Error Factor) - (Mean * Error Factor)], which means that the true protein ratio is found in this interval with a 95% confidence level.68 n ) number of peptides used for quantification. b This ratio is the same for the three MUP proteins and was entered as a single entry in Table S2, Supporting Information (MUP 1/11&8/6 - entry #152).
FABP1 in this process is emphasized by the findings that FABP1 knock-out mice are protected from high fat diet-induced hepatic lipid accumulation (steatosis).23 Therefore, in order to highlight the physiological relevance of elevated FABP1 expression in the STAT6-deficient mice, we challenged wild type and knock-out mice by a high fat diet for a period of ten weeks. Indeed, as expected, HF diet feeding resulted in more pronounced liver lipid accumulation in the knock-out mice compared to their wild-type littermates, in line with the increased expression of FABP1 identified in the proteomic analysis (Figure 1B, HF diet). Previous studies conducted in regular diet fed STAT6-deficient mice failed to describe mor4514
Journal of Proteome Research • Vol. 8, No. 10, 2009
phological signs of hepatic lipid accumulation when examined by hematoxylin-eosin staining.24 In accordance with these studies we found no gross structural alterations in similarly stained liver sections (Figure 1C, upper panels). However, when applying the neutral lipid dye Oil-red-O, the generalized lipid deposition in the livers of the knock-out mice became evident (Figure 1C, lower panels). Besides increased external fatty acid uptake, liver lipid accumulation can also be inflicted by increased intrinsic triglyceride synthesis or by a decrease in fatty acid oxidation capacity. To further prove the importance of FABP1 in the observed liver steatosis in the STAT6-null mice we examined
Differential Proteomic Analysis of STAT6 Knockout Mice
research articles
Figure 1. Increased liver lipid accumulation in STAT6 knock-out mice. A. mRNA and protein expression of the Fatty Acid Binding Protein 1 (FABP1). Each bar represents the average mRNA or protein level expressed as arbitrary units normalized to the mean of the wild type controls ( S.E.M.; * ) p e 0.05 (n ) 5). B. Total liver chloroform/methanol-extractable lipid content; * ) p e 0.05 wild type vs knock-out mice, ## ) p e 0.01 high fat diet vs chow diet (n ) 9 for chow diet and n ) 6 for high fat diet). C. Hematoxylin-Eosin (H.E.) (magnification: 20x) and Oil Red O (magnification: 10x) staining of liver sections of chow diet fed mice. D. mRNA expression of enzymes involved in intrinsic lipid metabolism. Each bar represents the average mRNA level expressed as arbitrary units normalized to the mean of the wild type controls ( S.E.M; * ) p e 0.01 (n ) 5). ACC: Acetyl CoA Carboxylase, FAS: Fatty Acid Synthase, CPT1: Carnitin Palmoyltransferase 1. E. Ketone body concentrations. Each bar represents the mean values ( S.E.M.; ### ) p e 0.001 (n ) 9) fed vs starved state in mice of the same genotype. HO-butyrate: hydroxybutyrate. F. Microsomal Triglyceride Transfer Protein (MTTP) mRNA and protein expression in wild type and knock-out mice. Each bar represents the average mRNA and protein levels expressed as arbitrary units normalized to the mean of the wild type controls ( S.E.M; (n ) 5-6 mice/group). A representative Western blot is shown for MTTP and Ezrin.
the expression levels of several intrinsic lipid metabolic enzymes whose function is known to be regulated at the mRNA level. In line with our expectations, no changes were observed in the expression of two key regulatory enzymes of the de novo fatty acid synthesis, acetyl-CoA carboxylase (ACC) and fatty acid synthase (FAS) (Figure 1D). This finding was also confirmed at the protein level for FAS (Table S2 #39, Supporting Information) while ACC was not identified in this analysis. The rate limiting step of mitochondrial FA degradation (oxidation) is the acyl-CoA transport across the mitochondrial membrane by the carnitin-palmytoil CoA transferase 1 (CPT1). CPT1 mRNA levels were not different between wild type and knock-out mice suggesting that FA oxidation rates were within the limits of physiological capacity of this transporter (Figure 1D). Fatty acid oxidation
is enhanced during starvation resulting in the formation of beta-hydroxybutyrate (BHB) and acetoacetate, commonly referred to as “ketone bodies”. In line with the similar CPT1 mRNA expression we found identical BHB and acetoacetate concentrations in wild type and knock-out mice regardless if mice were fed or were challenged by overnight starving (Figure 1E). Microsomal triglyceride transport protein (MTTP) plays a crucial role in the export of very low density lipoprotein (VLDL) from the liver. Deletion of MTTP led to the development of liver steatosis.25 In STAT6 mice MTTP mRNA and protein levels showed a tendency toward increased expression though the differences did not reach statistical significance (Figure 1F). Taken together, these data indicate that the observed lipid accumulation is due to changes in FABP1-mediated fatty acid uptake and storage Journal of Proteome Research • Vol. 8, No. 10, 2009 4515
research articles
Iff et al.
Figure 2. Verification of expression levels of proteins identified by the differential proteomics. (A and B) Protein and mRNA expression of Farnesyl diphosphate synthase (FDPS), acetyl CoA acetyltransferase (ACAT2) and cysteine sulfinic acid decarboxylase (CSAD) as determined by Western blot (A) and real-time PCR (B) analysis. Bars represent the average mRNA and protein levels expressed as arbitrary units normalized to the mean of the wild type controls ( S.E.M; (n ) 5-6 mice/group). A representative Western blot is shown for each protein along with the loading control Ezrin. Cytochome P7A1 mRNA expression, ** ) p e 0.01 wild type vs knock-out mice. (C) UDP-glucose 6-dehydrogenase. (UGDH) mRNA and protein expression, n ) 4-6mice/group; ** ) p e 0.01 wild type vs knock-out mice. A representative UGDH and Ezrin Western blot image is shown. (D) Ratio of conjugated plasma bilirubin. Data are expressed as the mean ( S.E.M.; * ) p e 0.05 (n ) 9).
and not to an increase in endogenous fatty acid synthesis or a decrease in fatty acid oxidation or VLDL export. Liver plays a crucial role in the metabolism of another important aspect of lipid homeostasis: the cholesterol/bile acid synthesis pathway. Indeed, iTRAQ analysis revealed decreased expression of two enzymes involved in this process, e.g. acetyl CoA acetyltransferase (ACAT2), farnesyl pyrophosphate synthetase (FPP Synthase, FPDS) (Table 1, #10 and #7, Figure 3). These results were confirmed by Western blot and real-time PCR analysis which showed corresponding expression levels, 4516
Journal of Proteome Research • Vol. 8, No. 10, 2009
though the differences did not always reach statistical significance (Figure 2A and B). In line with the down-regulation of these enzymes STAT6 knock-out mice showed decreased expression of cysteine sulfinic acid decarboxylase (CSAD) (Table 1, #16, Figure 2A and B). CSAD converts cystein into taurin, a molecule used for the conjugation of bile acids prior to their secretion. In accordance with the decreased expression of these proteins we confirmed the diminished mRNA expression of CYP7A1, the rate-limiting enzyme of bile acid synthesis in rodents (Figure 2B).
Differential Proteomic Analysis of STAT6 Knockout Mice A change in liver function in the STAT6 knock-out mice was also suggested by the decreased expression of another enzyme involved in a different conjugation process, the UDP-glucose 6-dehydrogenase (UGDH) (Table 1 #6, Figure 2C). UGDH catalizes the synthesis of UDP-glucoronic acid, used for the conjugation of bilirubin, the degradation product of hemoglobin. The physiological importance of the observed decrease in UGDH expression was reflected by a significant decrease in the ratio of conjugated bilirubin in the STAT6 knock-out mice (Figure 2D). The functions of the different lipid metabolic enzymes indentified by differential proteomics are summarized schematically in Figure 3. Expressions of Stress-Related Proteins in STAT6 KnockOut Mice. Eight proteins involved in cellular defensive strategies against oxidative, metabolic and heat shock stress showed coordinated changes in their expression levels: glutathione S-transferase Mu 2 (GSTM2), selenium binding protein 1 (SBP1), superoxide dismutase 1 (SOD1), regucalcin (RGN), lactoylglutathione lyase (GLO1), selenium binding protein 2 (SBP2), glutathione peroxidase 1 (GPX1) and glutathione Stransferase P1 (GSTP1) (Table 1, # 9, 15 and Table 2, # 5, 8, 9, 10, 18, 4). Taken together, these proteins indicated disturbed calcium and selenium homeostasis (regucalcin and SBP1, SBP2, respectively) and an increased need to eliminate toxic metabolic and oxidative byproducts (GLO1 and SOD1, GPX1, respectively). The highest levels of induction were detected in RGN and GLO1 expression, and these changes were also confirmed by Western blotting. Moreover, the increase in protein amount was accompanied by an elevation in the corresponding mRNA levels for both proteins (Figure 4A and B). Western blot analysis also validated the increased protein expression of GPX1 and SOD. However, in contrast to RGN and GLO1, where a clear increase in mRNA expression was observed, the changes were less evident in the case of GPX1 and SOD1 (Figure 4A and B). This might be due to the different sensitivity of the two methods applied or might reflect a post-transcriptional mechanism as suggested in another study.26 Glutathione is one of the major cellular protective factors against oxidative stress as it can act as a “buffer” against different reactive oxygen species by a reversible change between a reduced (GSH) monomer and an oxidized (GSSG) dimer form.27 The proteomic analysis revealed several proteins implicated in glutathione metabolism that were differentially expressed in the knock-out mice. The summary of the functions of the differentially expressed stress proteins is depicted in Figure 4C. GPX1 and GLO1 are two important detoxifying enzymes using the glutathione system. GPX1 catalyzes the reduction of peroxides while GLO1 is involved in the elimination of toxic glucose and lipid metabolic byproduct, mainly methylglyoxal.28 Increased amount of GPX1 and GLO1 are likely to reflect an increase in toxic metabolite production and thus can be regarded as a sign of developing oxidative stress. Other stress-related proteins regulated in the livers of STAT6 KO mice were the selenium binding proteins 1 and 2 (SBP1, SBP2). In line with other proteomic analyses we found that SBP1 and SBP2 were regulated in an opposite manner (Table 1, #15 and Table 2, #10).20,29,30 To validate the physiological significance of the identified stress proteins and to confirm the presence of cellular oxidative stress, we compared the amount of liver thiobarbituric acid reactive substances (TBARS) between wild type and STAT6 knock-out mice. TBARS reflect the degree of lipid peroxidation, a major indicator of oxidative stress.31-33
research articles
Figure 3. Overview of the functions of lipid metabolic enzymes identified by differential proteomics grouped by their metabolic function as proteins involved in (i) fatty acid uptake and storage, (ii) cholesterol and bile acid synthesis and (iii) bilirubin conjugation. Up- or down-regulated proteins in the STAT6 knock-out mice are marked in red and green, respectively. Proteins marked in black are enzymes that were not identified in the proteomic study but whose expression was established by real-time PCR. Box around proteins refers to rate-regulating enzymes. Arrows represent the direction of change, ) represents no change. Upregulated protein: FABP1, Fatty acid-binding protein. Downregulated proteins: ACAT2, acetyl CoA acetyltransferase; CSAD, cysteine sulfinic acid decarboxylase; FDPS, Farnesyl diphosphate synthase; CYP7A1, Cytochrome P450 7A1; UGDH, UDP-glucose 6-dehydrogenase. Other abbreviations: ACC, Acetyl-Coenzyme A carboxylase; CoA, Coenzyme A; CPT1, Carnitin Palmoyltransferase; FAS, Fatty acid synthase.
Indeed, in line with the conclusion drawn from the proteomic analysis, we found increased levels of oxidative stress in the knock-out mice (Figure 4D). Unaltered Expression of STAT3 and SOCS3. Decreased expression of another member of the STAT family, STAT3, and increased expression of the Suppressor of Cytokine Signaling (SOCS) proteins, negative regulators of STAT signaling were both shown to be related to the development of liver steatosis.34,35 To exclude changes in the expression of STAT3 and/or SOCS3 due to the elimination of STAT6 signaling, we evaluated STAT3 and SOCS3 mRNA and protein levels by realJournal of Proteome Research • Vol. 8, No. 10, 2009 4517
research articles
Iff et al. unchanged in the STAT6 knock-out mice, reinforcing a direct role for the lack of STAT6 signaling in the development of the observed liver lipid accumulation (Figure 5A and B, respectively). Increased Expression of Major Urinary Proteins. The most prominent change in protein expression was observed in case of the different isoforms of the major urinary proteins (MUP1, MUP6, MUP8 and MUP11). MUPs belong to the family of extracellular lipid-binding proteins, also referred to as “lipocalins”.36 The different MUP isoforms showed a 15-25-fold increase in the livers of the STAT6-null mice as determined by iTRAQ analysis (Table 2, #13, #14, #15 and #16) and this enhanced protein expression of was mirrored in their mRNA levels revealing a more than 500-fold increase in the knockout mice (Figure 6A, males). MUPs are synthesized in the liver and secreted in the male urine to serve as territorial marks, therefore MUP expression is generally higher in males than in females.37 In line with its sex-specific expression, female knockout mice showed a much less robust increase in their MUP expression when compared to males (Figure 6A, females). To assess the possible association between increased MUP expression and liver lipid accumulation we assessed their expression by real-time PCR in two different genetically obese rodent models characterized by liver steatosis: Ob/Ob mice and Fa/ Fa rats. In these two models we found a decrease in MUP expression suggesting that the observed MUP up-regulation in the livers of STAT6 KO mice is not directly linked to the development of liver lipid deposition (Figure 6B). In silico Promoter Analysis. STAT6 is a transcription factor; therefore changes in protein amount observed in the STAT6deficient livers could be attributed to alterations in mRNA transcription of STAT6-dependent genes. Indeed, several proteins, for example, FABP1, GLO1, RGN and MUPs showed tight correlations between their mRNA and protein levels. To gain a better insight to the mechanism underlying the changes observed in the STAT6-deficient mice, we performed in silico promoter analysis. The binding elements used for the analysis and the number of their occurrence in the promoter regions of the proteins identified in the study are listed in Table 3.
Figure 4. Increased cellular stress in the livers of STAT6 knock-out mice. Verification of up-regulation of identified proteins by Western blot (A) and real-time PCR (B). Each bar represents the average protein or mRNA level expressed as arbitrary units normalized to the mean of wild type controls ( SEM. RGN, regucalcin; GLO1, lactoylglutathione lyase; GPX1, glutathione peroxidase; SOD1, superoxide dismutase 1; * ) p e 0.05, ** ) p e 0.01, *** ) p e 0.001 (n ) 10 for Western blot, n ) 5 for real-time PCR). (C) Changes in the glutathione biochemical cycle in STAT6 knock-out mice. Proteins identified as up-regulated in the STAT6 knock-out mice are represented in red. Other abbreviations: G6PD, glucose-6-phosphate dehydrogenase; GR, glutathione reductase; GSH, glutathione (reduced form); GSSG, glutathione (oxidized form); NADPH, nicotinamide-adenine-dinucleotide phosphate. (D) Thiobarbituric acid reactive substances (TBARS) analysis (n ) 6-8, ** ) p e 0.01).
time PCR and Western blot analysis. As judged by these two methods the expression levels of the two proteins remained 4518
Journal of Proteome Research • Vol. 8, No. 10, 2009
First we analyzed the presence of STAT6 binding sequence in the promoter regions. Thirteen of the 17 down-regulated and 15 of the 17 up-regulated proteins contained one or more consensus STAT6 binding elements in their promoter regions, indicating that the presence of the binding element is not directly linked to the observed changes in expression levels in the knock-out mice and that probably some of the regulated genes are the results of an indirect effect of STAT6 deletion (Tables 4 and 5). Indeed, the homologue selenium-binding proteins 1 and 2 were down-and up-regulated, respectively, in spite of the presence of several STAT6 consensus elements in the promoters of both proteins. These results confirm a more general, two-way (inhibitory and stimulatory) direct and indirect transcription regulatory role for STAT6; a situation similar to its role in lymphocyte Th2 differentiation.38 Another family of transcription factors, the CCAAT/Enhancer Binding Proteins (C/EBPs), is a well-known modifier of STAT6 regulated transcriptional activation.1 In addition, liver specific C/EBPR knockout mice display age-dependent hepatosteatosis, a similar feature to the phenotype uncovered in the STAT6 knock-out mice by this study.39 In our analysis none of the promoter regions of the identified proteins contained the classical binding element for C/EBP rendering it unlikely that the
Differential Proteomic Analysis of STAT6 Knockout Mice
research articles
Figure 5. Unaltered STAT3 and SOCS3 expression in STAT6-null mice. (A) mRNA expression of STAT3 and SOCS3. Bars represent average mRNA levels expressed as arbitrary units normalized to the mean of wild type mice ( SEM (n ) 6). (B) Western blot analysis of protein expression of STAT6, STAT3 and SOCS3. Ezrin is a nonvariable protein showing equal protein loading. Each blot shows three representative samples from a blot containing six individual samples.
Figure 6. MUP mRNA expression. (A) Quantification of MUP mRNA expression in male and female mice. Each bar represents the average expression level expressed as arbitrary units normalized to the mean of the wild type males ( SEM (males: n ) 5; females: n ) 11). Y axis is in logarithmic scale. * ) p e 0.05, ** ) p e 0.01 wild type vs knock-out mice; ### ) p < 0.001 males vs females of the same genotype. (B) MUP mRNA levels determined by real-time PCR in obese murine models of hepatic steatosis. ** ) p e 0.01, n ) 6. Table 3. Oligo Binding Motives Used for in silico Promoter Analysis motive
sequence
down-regulated proteins
up-regulated proteins
STAT6 C/EBP
TTCNNNNGAA (A/G)TTGCG(C/T)AA(C/T)
34 0
27 0
observed liver lipid deposition is indirectly due to the lack of signaling through C/EBP.
Discussion The link between increased inflammatory signaling and the development of metabolic disorders is a current and highly investigated topic.40 Indeed, the role of increased pro-inflammatory signals from the adaptive immune system (e.g., Tumor Necrosis Factor R, TNF-R), and from the innate immune system (e.g., Tol-like Receptor 4, TLR4) has been extensively studied.41-43 By contrast, no major research efforts have been dedicated to study the contrasting scenario: the attenuation of counterbalancing anti-inflammatory signals. IL-4 and IL-13 are two major anti-inflammatory cytokines signaling through the transcription factor STAT6. Therefore, the aim of our investigation was to assess the impact of loss of signaling by STAT6 in liver, an organ playing a major role in the regulation of metabolic homeostasis. In the first part of our study we analyzed the liver proteomes of wild type and STAT6 knock-out mice using 2D nanoscale LC tandem mass spectrometry, also known as multidimensional protein identification technology (MudPIT).44 We identified 16 down-regulated and 18 up-regulated proteins in the
STAT6 knock-out mice. Taken together, these proteins suggested alterations in lipid homeostasis and the presence of hepatocellular stress. The changes in the expression levels of enzymes controlling lipid metabolism suggested a balance biased toward liver lipid accumulation due to excess exogenous fatty acid loading related to the up-regulation of the liver fatty acid binding protein, FABP1. Indeed, in line with the elevated FABP1 expression, liver lipid content in the knock-out mice was significantly increased and this augmentation was further aggravated when mice were challenged with a high fat containing diet. These findings are in perfect correlation with data demonstrating the protective effect of FABP1 deficiency against the onset of diet-induced hepatic steatosis in the FABP1 knockout mice.23 Interestingly, another member of the family of fatty acid binding proteins, the adipocyte/macrophage specific aP2 has been shown to be inducible by IL-4 in human bronchial cells in a STAT6 dependent manner in vitro, and inhibition of aP2 ameliorated the fatty liver phenotype in vivo in obese mice.6,45 Thus, our data provides a novel aspect to these findings demonstrating that the expression of the liver specific fatty acid binding protein isoform, FABP1, is regulated by STAT6 and is linked to the onset of liver lipid accumulation. In this context it is important to point out two major aspects of our finding concerning the up-regulation of FABP1 expression: the presence of latent liver steatosis that can be aggravated by metabolic challenges and the fact that this steatosis develops independent of general obesity. Indeed, the presence of fatty liver of unknown origin has been documented in 10-15% of normal individuals and is speculated to give the “first hit” insult toward the development of more severe liver pathologies, for Journal of Proteome Research • Vol. 8, No. 10, 2009 4519
research articles
Iff et al.
Table 4. STAT6 Binding Motives Identified in the +5000/-1000 Basepair Regions of Down-Regulated Proteins number
gene symbol
protein name
RefSeq
chromosome
strand
STAT6
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Ass1 Adh1 Fbp1 Etfa Pygl Ugdh Fdps Tuba6 Gstm2 Acat3 Acat2 Gpi1 Acly Got1 Cycs Selenbp1 Csad
Argininosuccinate synthase Alcohol dehydrogenase 1 Fructose bisphosphatase 1 Electron transferring flavoprotein Liver glycogen phosphorylase UDP-glucose 6-dehydrogenase Farnesyl pyrophosphate synthetase Tubulin alpha-6 chain Glutathione S-transferase Mu 2 Acetyl CoA transferase-like protein Acetyl CoA acetyltransferase, cytosolic Glucose phosphate isomerase 1 ATP-citrate synthase Glutamate oxaloacetate transaminase 1 Cytochrome C protein, somatic Selenium-binding protein 1 Cysteine sulfinic acid decarboxylase
NM_007494 NM_007409 NM_019395 NM_145615 NM_133198 NM_009466 NM_134469 NM_009448 NM_010359 NM_153151 NM_009338 NM_008155 NM_134037 NM_010324 NM_007808 NM_009150 NM_144942
chr2 chr3 chr13 chr9 chr12 chr5 chr3 chr15 chr3 chr17 chr17 chr7 chr11 chr19 chr6 chr3 chr15
+ + + + -
5 0 0 2 1 0 1 1 2 4 4 3 3 2 0 2 4
Table 5. STAT6 Binding Motives Identified in the +5000/-1000 Basepair Regions of Up-Regulated Proteins number
gene symbol
protein name
RefSeq
chromosome
strand
STAT6
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Ca3 Fabp1 Hba-a1 Gstp1 Sod1 Acaa2 Scp2 Rgn Glo1 Selenbp2 Nme2 Akr1c6 Mup1a Mup6 Mup8, Mup11 Tst Gpx1
Carbonic anhydrase 3 Fatty acid-binding protein | FABPL Hemoglobin alpha chain Glutathione S-transferase P1 Superoxide dismutase [Cu-Zn] 3-ketoacyl-CoA thiolase, mitochondrial Nonspecific lipid-transfer protein Regulcalcin | SMP 30 Lactoylglutathione lyase LGUL Selenium binding protein 2 Nucleoside diphosphate kinase B Estradiol 17 beta-dehydrogenase 5 Major urinary protein 1 MUP Major urinary protein 6 MUP Major urinary protein 8 and 11 MUP Thiosulfate sulfurtransferase Glutathione peroxidase 1
NM_007606 NM_017399 NM_008218 NM_013541 NM_011434 NM_177470 NM_011327 NM_009060 NM_025374 NM_019414 NM_008705 NM_030611 NM_031188 NM_008648 NM_008649 NM_009437 NM_008160
chr3 chr6 chr11 chr19 chr16 chr18 chr4 chrX chr17 chr3 chr11 chr13 chr4 chr4 chr4 chr15 chr9
+ + + + + + + + +
1 1 1 0 0 1 1 1 1 3 4 1 3 2 3 3 1
example, fibrosis and cirrhosis.46 Thus, identifying FABP1 and ultimately STAT6 as crucial factors predisposing toward a silent liver lipid accumulation is of clinical significance. Along the same line, increased aP2 expression has been linked to higher cardiovascular risk in humans and selective inhibition of aP2 has been shown to be of potential value for the treatment against vascular wall lipid deposition and atherosclerosis in a mouse model.45 Based upon the data of our study, inhibiting the hepatocyte specific FABP1 may provide a novel mean to ameliorate or to prevent the onset of liver steatosis and the following fibrosis/cirrhosis. Moreover, recognizing mutations leading to decreased STAT6 expression or function might be useful for identifying subjects with increased risk to develop liver steatosis and hepatocyte specific targeting of STAT6 might be considered as a therapeutic mean.47 As mentioned earlier, the most physiologically important change observed in the STAT6 knock-out mice was the upregulation of the fatty acid binding protein, FABP1. Beside their role in regulation cellular lipid homeostasis, fatty acid binding proteins have been proposed as molecular links between metabolic and inflammatory pathways in the context of adipose tissue.48 Similarly, in the liver, the increased FABP1 expression was accompanied by the coordinated regulation of several members of diverse but interconnected stress-related cellular networks, notably proteins involved the regulation of cytoplas4520
Journal of Proteome Research • Vol. 8, No. 10, 2009
mic calcium, selenium and glutathione concentrations. Altogether, the changes in these proteins indicated the presence of cellular stress and a decreased defense capacity against further metabolic or oxidative insults. While our results showed a concomitant increase in FABP1 expression along with elevated TBARs we cannot rule out an independent STAT6related effect on oxidative stress genes. To prove a causative effect of FABP1 in the development of oxidative stress further experiments will be required. Among the identified stress-related proteins the first group is linked to the regulation intracellular calcium homeostasis and -as one of the most important calcium store- the function of the endoplasmatic reticulum (ER). ER also plays a crucial role in the protein refolding process of proteins damaged by intracellular metabolic/oxidative stress. Our proteomic analysis showed an up-regulation of two proteins related to this function: regucalcin (RGN) and glyoxalase 1 (GLO1). Indeed, a protective role for RGN against liver steatosis and disturbed function was suggested by studies showing RGN down-regulation in livers of C57Bl/6 mice, a strain susceptible to atherogenic diet, compared to the resistant C3H/HeJ mice; and in the livers of senescence accelerated mice (SAM).29,49 GLO1 plays a central role in the elimination of toxic byproducts derived from lipid peroxidation and the formation of advanced glycation end-products (AGEs).50,51 Therefore, increased RGN
Differential Proteomic Analysis of STAT6 Knockout Mice and GLO1 expression in the STAT6 KO mice may be considered as part of the cellular defensive mechanism to stabilize cytoplasmic Ca2+ levels and to increase refolding of proteins damaged by oxidative stress and/or increased glycosylation due to the metabolic changes induced by fatty acid overload. The second group of stress-related proteins identified in the study is related to cellular selenium homeostasis, namely the selenium binding proteins (SBPs). There are three different proteins belonging to this class: the liver fatty acid-binding protein FABP1, mentioned already in context of its role in fatty acid uptake and two homologue proteins called selenium binding protein 1 and 2 (SBP1 and SBP2). While the role of FABP1 is characterized in a detailed fashion the functions of SBP1 and SBP2 remained relatively unexplored. The two SBPs share a high degree of identity both at the mRNA (98%) and at the protein (96%) levels with a difference of mere 18 amino acids between their sequences. In contrast to SBP1, the expression of SBP2 was significantly up-regulated in STAT6deficient mice. Indeed, in spite of their structural and apparent functional similarity SBP1 and SBP2 expression were found to be divergent in several previous proteomic studies.29,52,53 Most importantly, treatment of mice with ciprofibrate, a hypolipidemic agent led to the selective down-regulation of SBP2 expression coinciding with a decrease in liver lipid content.54 An even more specific link between liver lipid homeostasis and SBP2 expression was indicated by the study of Park et al. where feeding of a hypercholesterinemia-inducing diet led to the down-regulation of SBP2 expression in the steatosis susceptible C57BL6/J but not in the resistant C3H strain.49 In this regard the increased SBP2 expression in the STAT6 knock-out mice could be viewed as an adaptive mechanism against the gradually developing lipid deposition and the consequential ROS accumulation leading to oxidative stress. Indeed, in addition to their role in selenium homeostasis, a direct antioxidant activity has been postulated for SBPs in analogy to other selenocysteine-containing enzymes, for example, glutathione peroxidase 1 (GPX1).55 This point is substantiated by the fact that GPX1 has also been identified as up-regulated in STAT6 knock-out mice. The third group of identified stress proteins was enzymes involved in the cellular defense against the accumulation of toxic metabolic byproduct or reactive oxygen radicals utilizing reduced glutathione. The coordinated regulation of these enzymes suggests a decrease in glutathione antioxidative capacity due to its increased utilization by GLO1 and GPX1. This picture is in accordance with the sensitivity of STAT6 knock-out mice toward liver and kidney ischemia-reperfusion induced cellular oxidative injury and with the protective effect of IL-4 and IL-13, the two STAT6-dependent cytokines against these insults.56 The significance of the intact function of the glutathione system in humans was established by a recently study reporting a decrease in glutathione S-transferase (GST) expression in steatotic livers; a finding emphasizing the relevance of down-regulation of this enzyme in the STAT6 knockout mice.57 In summary, the proteome of the STAT6-deficient mice presented an overall picture with a up-regulation of enzymes involved in the elimination of different reactive oxygen species and damaging metabolites along with a depletion of cellular defensive reserve to withstand oxidative stress. The physiological relevance of the expression changes in these stress-related proteins was verified by direct measurement of the degree of lipid peroxidation, a major indicator of cellular oxidative stress.
research articles In accordance with our expectations, STAT6 knock-out mice showed elevated oxidative stress thus confirming our conclusions derived from the proteomic analysis. The highest increase in mRNA and protein expression was detected in case of the major urinary proteins (MUPs). MUPs received their name by their predominance in mouse urine constituting a “physiological proteinuria” (reviewed in ref 58). MUPs are filtrated by kidney glomeruli freely due to their low (approximately 18 kDa) molecular weights and their globular form. Major urinary proteins belong to the family of lipocalins, characterized by an overall structure of eight β-sheets defining a β-barrel configuration. The hydrophobic core of the molecule binds volatile polar pheromones in the male urine and releases them slowly once deposited as a territorial “scent mark”.59,60 MUPs are primarily produced in the liver in a sex dependent manner with the males having significantly higher expression levels than females.61,62 MUPs are the products of a multigene family of approximately 30 genes and pseudogenes localized on mouse chromosome 4 displaying high sequence homology both at the mRNA and the protein levels.63 Transcription of MUPs can be induced by testosterone, growth hormone, thyroxin, insulin and dexamethasone administration. None of these factors could be held accountable for the spectacular upregulation observed in the STAT6 KO mice indicating the involvement of currently unidentified factor(s). MUP upregulation, however, was not casually related to the development of liver steatosis as other murine models of hepatoteatosis showed decreased MUP expression. Indeed, while this manuscript was in revision a paper published by Hui et al. showed a role for MUP-1 in enhancing energy expenditure in muscle.64 In this respect, the spectacular up-regulation of MUP-1 expression in the STAT6 knock-out mice can be viewed as a counterregulatory mechanism, actually mitigating the observed liver steatotic phenotype. In summary, the data of our proteomic comparison suggested that STAT6 exerts a protective effect against the development of liver steatosis through a complex regulatory network concerning metabolic enzymes and proteins involved in cellular redox state regulation. The physiological relevance and potential therapeutic interest of STAT6 in liver pathology are supported by a variety of mouse and human studies. The occurrence of liver steatosis is in accordance with previous results demonstrating that STAT6 knock-out mice are prone to develop more serious atherosclerotic lesions with enhanced lipid deposition in the aortic wall when challenged by a high-fat diet.65 Furthermore, the relevance of these results to human pathology was recently highlighted by data identifying STAT6 as one of the three mRNAs showing the greatest up-regulation in aortic atherosclerotic plaques.66 In addition, a recent study suggested an association between atherosclerosis and nonalcoholic fatty liver disease implying the involvement of similar factors in their development.67 While proteomic studies conducted in human liver steatosis did not identify STAT6 as a differentially expressed protein it is not surprising given the very low abundance of this protein as demonstrated by the lack of detection by iTRAQ analysis in our study. Taken together, the above results confirm that the phenotype of STAT6 knockout mice bears resemblance to the human pathology and imply a yet unappreciated contribution of STAT6 to the development of liver steatosis. In conclusion, our study explored the effect of the suppression of IL-4 and IL-13-mediated anti-inflammatory signals on liver function by comparing the proteomes of wild type and Journal of Proteome Research • Vol. 8, No. 10, 2009 4521
research articles STAT6 knock-out mice. Based upon the identified proteins we revealed a so far unknown metabolic phenotype in the STAT6deficient mice by demonstrating the presence of latent liver steatosis and their sensitivity toward high-fat diet feeding. Taken together, these results validate a protective role for STAT6 against hepatic lipid deposition; a finding reminiscent of the protective role it plays against the development of atherosclerotic lipid accumulation. According to our results, a role for STAT6 in the context of nonalcoholic fatty liver disease in humans is a topic worth of further exploration. Abbreviations: 2D nLC-MS/MS, two-dimensional nanoscale LC tandem mass spectrometry; ACAA2, 3-ketoacyl-CoA thiolase; ACAT2, acetyl CoA acetyltransferase; ACC, acetyl-CoA carboxylase; acetyl-CoA, acetyl coenzyme A; ACLY, ATP-citrate synthase; AGEs, advanced glycation endproducts; BCA, bicinchoninic acid; BHB, beta-hydroxybutyrate; BPB, bromo phenol blue; CPT1, carnitin-palmytoil CoA transferase 1; CSAD, cysteine sulfinic acid decarboxylase; CYCS, Cytochrome C protein, somatic; CYP7A1, cytochrome family 7 subunit A isoform 1; FABP1, fatty acid-binding protein; FAS, fatty acid synthase; FPP synthase, farnesyl pyrophosphate synthase; GLO1, glyoxalase 1; GPX1, glutathione peroxidase 1; GSTM2, glutathione Stransferase 2; GST, glutathione S-transferase; GSTP1, glutathione S-transferase P1; HRP, horseradish peroxidase; KO, knock-out; MMTS, methyl methane-thiosulfate; MTTP, microsomal triglyceride transfer protein; MudPIT, multidimensional protein identification technology; MUP, major urinary protein; NaF, sodium fluoride; NAFLD, nonalcoholic fatty liver disease; nLC, nanoscale liquid chromatography; RGN, regucalcin; ROS, reactive oxygen species; SAM, senescence accelerated mice; SBP1, selenium binding protein 1; SBP2, selenium binding protein 2; SCP2, nonspecific lipid-transfer protein; S.E.M., standard error of the mean; SMP 30, 30 kDa senescence marker protein; SOCS, suppressor of cytokine signaling; SOD1, superoxide dismutase 1; STAT3, STAT6, signal transducer and activator of transcription 3 and 6; TBARS, thiobarbituric acid reactive substances; TCEP, tris(2-carboxyethyl)phosphine; TEAB, triethylammonium bicarbonate; TRACE, time-resolved amplified cryptated emission; UGDH, UDP-glucose 6-dehydrogenase; WT, wild type.
Acknowledgment. We are grateful to colleagues who generously provided advice and reagents. We thank N. Desmeules, C. Manzin and S. Mouche (Department of Cell Physiology and Metabolism, University of Geneva) for excellent technical help; Prof. D. Hochstrasser and Dr. O. Golaz (Central Clinical Chemistry Laboratory of the University Hospital of Geneva) for serum measurements; Dr. P. Descombes and C. Delucinge (Genomics Platform, NCCR “Frontiers in Genetics”, University of Geneva) for the promoter analysis; and P. Boutros (Department of Pharmacology, University of Toronto) for providing the Bioperl scripts and valuable scientific comments. This work was supported from a grant from COST Action B17, No.C02.0097. Supporting Information Available: List of primers used for real-time PCR and list of identified proteins in the livers of wild type and STAT6 knock-out mice by iTRAQ 2D nLC-MS/MS technique. This material is available free of charge via the Internet at http://pubs.acs.org. 4522
Journal of Proteome Research • Vol. 8, No. 10, 2009
Iff et al.
References (1) Hebenstreit, D.; Wirnsberger, G.; Horejs-Hoeck, J.; Duschl, A. Signaling mechanisms, interaction partners, and target genes of stat6. Cytokine Growth Factor Rev. 2006, 17 (3), 173–88. (2) Zimmermann, N.; Mishra, A.; King, N. E.; Fulkerson, P. C.; Doepker, M. P.; Nikolaidis, N. M.; Kindinger, L. E.; Moulton, E. A.; Aronow, B. J.; Rothenberg, M. E. Transcript signatures in experimental asthma: Identification of stat6-dependent and -independent pathways. J. Immunol. 2004, 172 (3), 1815–24. (3) Kaplan, M. H.; Schindler, U.; Smiley, S. T.; Grusby, M. J. Stat6 is required for mediating responses to il-4 and for development of th2 cells. Immunity 1996, 4 (3), 313–9. (4) Aoudjehane, L.; Podevin, P.; Scatton, O.; Jaffray, P.; Dusanter-Fourt, I.; Feldmann, G.; Massault, P. P.; Grira, L.; Bringuier, A.; Dousset, B.; Chouzenoux, S.; Soubrane, O.; Calmus, Y.; Conti, F. Interleukin-4 induces human hepatocyte apoptosis through a fasindependent pathway. FASEB J. 2007, 21 (7), 1433–44. (5) Low, S. H.; Vasanth, S.; Larson, C. H.; Mukherjee, S.; Sharma, N.; Kinter, M. T.; Kane, M. E.; Obara, T.; Weimbs, T. Polycystin-1, stat6, and p100 function in a pathway that transduces ciliary mechanosensation and is activated in polycystic kidney disease. Dev. Cell 2006, 10 (1), 57–69. (6) Shum, B. O.; Mackay, C. R.; Gorgun, C. Z.; Frost, M. J.; Kumar, R. K.; Hotamisligil, G. S.; Rolph, M. S. The adipocyte fatty acidbinding protein ap2 is required in allergic airway inflammation. J. Clin. Invest. 2006, 116 (8), 2183–2192. (7) Gao, B. Cytokines, stats and liver disease. Cell Mol. Immunol. 2005, 2 (2), 92–100. (8) Li, J.; Thorne, L. N.; Punjabi, N. M.; Sun, C. K.; Schwartz, A. R.; Smith, P. L.; Marino, R. L.; Rodriguez, A.; Hubbard, W. C.; O’Donnell, C. P.; Polotsky, V. Y. Intermittent hypoxia induces hyperlipidemia in lean mice. Circ. Res. 2005, 97 (7), 698–706. (9) Savransky, V.; Nanayakkara, A.; Vivero, A.; Li, J.; Bevans, S.; Smith, P. L.; Torbenson, M. S.; Polotsky, V. Y. Chronic intermittent hypoxia predisposes to liver injury. Hepatology 2007, 45 (4), 1007–13. (10) Li, J.; Grigoryev, D. N.; Ye, S. Q.; Thorne, L.; Schwartz, A. R.; Smith, P. L.; O’Donnell, C. P.; Polotsky, V. Y. Chronic intermittent hypoxia upregulates genes of lipid biosynthesis in obese mice. J. Appl. Physiol. 2005, 99 (5), 1643–8. (11) Husted, T. L.; Lentsch, A. B. The role of cytokines in pharmacological modulation of hepatic ischemia/reperfusion injury. Curr. Pharm. Des. 2006, 12 (23), 2867–73. (12) Arnold, R. J.; Hrncirova, P.; Annaiah, K.; Novotny, M. V. Fast proteolytic digestion coupled with organelle enrichment for proteomic analysis of rat liver. J. Proteome Res. 2004, 3 (3), 653–657. (13) Ross, P. L.; Huang, Y. N.; Marchese, J. N.; Williamson, B.; Parker, K.; Hattan, S.; Khainovski, N.; Pillai, S.; Dey, S.; Daniels, S.; Purkayastha, S.; Juhasz, P.; Martin, S.; Bartlet-Jones, M.; He, F.; Jacobson, A.; Pappin, D. J. Multiplexed protein quantitation in saccharomyces cerevisiae using amine-reactive isobaric tagging reagents. Mol. Cell. Proteomics 2004, 3 (12), 1154–1169. (14) Tscha¨ppa¨t, V.; Varesio, E.; Signor, L.; Hopfgartner, G. The application of 2-d dual nanoscale liquid chromatography and triple quadrupole-linear ion trap system for the identification of proteins. J. Sep. Sci. 2005, 28 (14), 1704–1711. (15) Hirsch, J.; Hansen, K. C.; Choi, S.; Noh, J.; Hirose, R.; Roberts, J. P.; Matthay, M. A.; Burlingame, A. L.; Maher, J. J.; Niemann, C. U. Warm ischemia-induced alterations in oxidative and inflammatory proteins in hepatic kupffer cells in rats. Mol. Cell. Proteomics 2006, 5 (6), 979–986. (16) Shilov, I. V.; Seymour, S. L.; Patel, A. A.; Loboda, A.; Tang, W. H.; Keating, S. P.; Hunter, C. L.; Nuwaysir, L. M.; Schaeffer, D. A. The paragon algorithm: A next generation search engine that uses sequence temperature values and feature probabilities to identify peptides from tandem mass spectra. Mol. Cell. Proteomics 2007, 6, 1638–1655. (17) Seymour, S. L. In MCP Workshop: Criteria for Publication of Proteomic Data; Pro group: Paris, 2005. (18) Guo, Y.; Singleton, P. A.; Rowshan, A.; Gucek, M.; Cole, R. N.; Graham, D. R. M.; Van Eyk, J. E.; Garcia, J. G. N. Quantitative proteomics analysis of human endothelial cell membrane rafts. Mol. Cell. Proteomics 2007, 6 (4), 689–696. (19) Bligh, E. G.; Dyer, W. J. A rapid method of total lipid extraction and purification. Can. J. Biochem. Physiol. 1959, 37 (8), 911–7. (20) Pastorelli, R.; Carpi, D.; Campagna, R.; Airoldi, L.; Pohjanvirta, R.; Viluksela, M.; Hakansson, H.; Boutros, P. C.; Moffat, I. D.; Okey, A. B.; Fanelli, R. Differential expression profiling of the hepatic proteome in a rat model of dioxin resistance: Correlation with genomic and transcriptomic analyses. Mol. Cell. Proteomics 2006, 5 (5), 882–94.
research articles
Differential Proteomic Analysis of STAT6 Knockout Mice (21) Atshaves, B. P.; McIntosh, A. M.; Lyuksyutova, O. I.; Zipfel, W.; Webb, W. W.; Schroeder, F. Liver fatty acid-binding protein gene ablation inhibits branched-chain fatty acid metabolism in cultured primary hepatocytes. J. Biol. Chem. 2004, 279 (30), 30954–65. (22) Glatz, J. F.; van der Vusse, G. J. Cellular fatty acid-binding proteins: Their function and physiological significance. Prog. Lipid Res. 1996, 35 (3), 243–82. (23) Newberry, E. P.; Xie, Y.; Kennedy, S. M.; Luo, J.; Davidson, N. O. Protection against western diet-induced obesity and hepatic steatosis in liver fatty acid-binding protein knockout mice. Hepatology 2006, 44 (5), 1191–205. (24) Shen, X. D.; Ke, B.; Zhai, Y.; Gao, F.; Anselmo, D.; Lassman, C. R.; Busuttil, R. W.; Kupiec-Weglinski, J. W. Stat4 and stat6 signaling in hepatic ischemia/reperfusion injury in mice: Ho-1 dependence of stat4 disruption-mediated cytoprotection. Hepatology 2003, 37 (2), 296–303. (25) Minehira, K.; Young, S. G.; Villanueva, C. J.; Yetukuri, L.; Oresic, M.; Hellerstein, M. K.; Farese, R. V., Jr.; Horton, J. D.; Preitner, F.; Thorens, B.; Tappy, L. Blocking vldl secretion causes hepatic steatosis but does not affect peripheral lipid stores or insulin sensitivity in mice. J. Lipid Res. 2008, 49 (9), 2038–2044. (26) Handy, D. E.; Zhang, Y.; Loscalzo, J. Homocysteine down-regulates cellular glutathione peroxidase (gpx1) by decreasing translation. J. Biol. Chem. 2005, 280 (16), 15518–25. (27) Rhee, S. G.; Chae, H. Z.; Kim, K. Peroxiredoxins: A historical overview and speculative preview of novel mechanisms and emerging concepts in cell signaling. Free Radic. Biol. Med. 2005, 38 (12), 1543–52. (28) Behne, D.; Kyriakopoulos, A. Mammalian selenium-containing proteins. Annu. Rev. Nutr. 2001, 21, 453–73. (29) Cho, Y. M.; Bae, S. H.; Choi, B. K.; Cho, S. Y.; Song, C. W.; Yoo, J. K.; Paik, Y. K. Differential expression of the liver proteome in senescence accelerated mice. Proteomics 2003, 3 (10), 1883–94. (30) Chu, R.; Lim, H.; Brumfield, L.; Liu, H.; Herring, C.; Ulintz, P.; Reddy, J. K.; Davison, M. Protein profiling of mouse livers with peroxisome proliferator-activated receptor alpha activation. Mol. Cell. Biol. 2004, 24 (14), 6288–97. (31) Armstrong, D.; Browne, R. The analysis of free radicals, lipid peroxides, antioxydant enzymes and compounds related to oxidative stress as applied to the clinical chemistry laboratory. Adv. Exp. Med. Biol. 1994, 366, 43–58. (32) Yagi, K. Simple procedure for specific assay of lipid hydroperoxides in serum or plasma. Methods Mol. Biol. 1998, 108, 107–10. (33) Lefe`vre, G.; Beljean-Leymarie, M.; Beyerle, F.; Bonnefont-Rousselot, D.; Cristol, J.; The´rond, P.; Torreilles, J. Evaluation of lipid peroxidation by measuring thiobarbituric acid reactive substances. Ann. Biol. Clin. 1998, 56 (3), 305–319. (34) Inoue, H.; Ogawa, W.; Asakawa, A.; Okamoto, Y.; Nishizawa, A.; Matsumoto, M.; Teshigawara, K.; Matsuki, Y.; Watanabe, E.; Hiramatsu, R.; Notohara, K.; Katayose, K.; Okamura, H.; Kahn, C. R.; Noda, T.; Takeda, K.; Akira, S.; Inui, A.; Kasuga, M. Role of hepatic stat3 in brain-insulin action on hepatic glucose production. Cell. Metab. 2006, 3 (4), 267–75. (35) Ueki, K.; Kadowaki, T.; Kahn, C. R. Role of suppressors of cytokine signaling socs-1 and socs-3 in hepatic steatosis and the metabolic syndrome. Hepatol. Res. 2005, 33 (2), 185–92. (36) de Wolf, F. A.; Brett, G. M. Ligand-binding proteins: Their potential for application in systems for controlled delivery and uptake of ligands. Pharmacol Rev. 2000, 52 (2), 207–36. (37) Knopf, J. L.; Gallagher, J. F.; Held, W. A. Differential, multihormonal regulation of the mouse major urinary protein gene family in the liver. Mol. Cell. Biol. 1983, 3 (12), 2232–40. (38) Filen, J.-J.; Filen, S.; Moulder, R.; Tuomela, S.; Ahlfors, H.; West, A.; Kouvonen, P.; Kantola, S.; Bjorkman, M.; Katajamaa, M.; Rasool, O.; Nyman, T. A.; Lahesmaa, R. Quantitative proteomics reveals gimap family proteins 1 and 4 to be differentially regulated during human t helper cell differentiation. Mol. Cell. Proteomics 2009, 8 (1), 32–44. (39) Inoue, Y.; Inoue, J.; Lambert, G.; Yim, S. H.; Gonzalez, F. J. Disruption of hepatic c/ebpalpha results in impaired glucose tolerance and age-dependent hepatosteatosis. J. Biol. Chem. 2004, 279 (43), 44740–8. (40) Hotamisligil, G. S. Inflammation and metabolic disorders. Nature 2006, 444 (7121), 860–7. (41) Feldstein, A. E.; Werneburg, N. W.; Canbay, A.; Guicciardi, M. E.; Bronk, S. F.; Rydzewski, R.; Burgart, L. J.; Gores, G. J. Free fatty acids promote hepatic lipotoxicity by stimulating tnf-alpha expression via a lysosomal pathway. Hepatology 2004, 40 (1), 185–94. (42) Poggi, M.; Bastelica, D.; Gual, P.; Iglesias, M. A.; Gremeaux, T.; Knauf, C.; Peiretti, F.; Verdier, M.; Juhan-Vague, I.; Tanti, J. F.; Burcelin, R.; Alessi, M. C. C3h/hej mice carrying a toll-like receptor
(43) (44) (45)
(46) (47)
(48) (49) (50) (51)
(52)
(53)
(54)
(55)
(56) (57)
(58) (59) (60)
(61) (62)
(63)
4 mutation are protected against the development of insulin resistance in white adipose tissue in response to a high-fat diet. Diabetologia 2007, 50 (6), 1267–76. Shi, H.; Kokoeva, M. V.; Inouye, K.; Tzameli, I.; Yin, H.; Flier, J. S. Tlr4 links innate immunity and fatty acid-induced insulin resistance. J. Clin. Invest. 2006, 116 (11), 3015–25. Washburn, M. P.; Wolters, D.; Yates, J. R., III. Large-scale analysis of the yeast proteome by multidimensional protein identification technology. Nat. Biotechnol. 2001, 19 (March), 242–247. Furuhashi, M.; Tuncman, G.; Gorgun, C. Z.; Makowski, L.; Atsumi, G.; Vaillancourt, E.; Kono, K.; Babaev, V. R.; Fazio, S.; Linton, M. F.; Sulsky, R.; Robl, J. A.; Parker, R. A.; Hotamisligil, G. S. Treatment of diabetes and atherosclerosis by inhibiting fatty-acid-binding protein ap2. Nature 2007, 447 (7147), 959–65. Duvnjak, M.; Lerotic, I.; Barsic, N.; Tomasic, V.; Virovic Jukic, L.; Velagic, V. Pathogenesis and management issues for non-alcoholic fatty liver disease. World J. Gastroenterol. 2007, 13 (34), 4539–50. Borger, P.; Matsumoto, H.; Boustany, S.; Gencay, M. M.; Burgess, J. K.; King, G. G.; Black, J. L.; Tamm, M.; Roth, M. Disease-specific expression and regulation of ccaat/enhancer-binding proteins in asthma and chronic obstructive pulmonary disease. J. Allergy Clin. Immunol. 2007, 119 (1), 98–105. Makowski, L.; Hotamisligil, G. S. Fatty acid binding proteins--the evolutionary crossroads of inflammatory and metabolic responses. J. Nutr. 2004, 134 (9), 2464S–2468S. Park, J. Y.; Seong, J. K.; Paik, Y. K. Proteomic analysis of dietinduced hypercholesterolemic mice. Proteomics 2004, 4 (2), 514– 23. Mlakar, A.; Batna, A.; Dudda, A.; Spiteller, G. Iron (ii) ions induced oxidation of ascorbic acid and glucose. Free Radic. Res. 1996, 25 (6), 525–39. Wells-Knecht, K. J.; Zyzak, D. V.; Litchfield, J. E.; Thorpe, S. R.; Baynes, J. W. Mechanism of autoxidative glycosylation: Identification of glyoxal and arabinose as intermediates in the autoxidative modification of proteins by glucose. Biochemistry 1995, 34 (11), 3702–9. Henkel, C.; Roderfeld, M.; Weiskirchen, R.; Berres, M. L.; Hillebrandt, S.; Lammert, F.; Meyer, H. E.; Stuhler, K.; Graf, J.; Roeb, E. Changes of the hepatic proteome in murine models for toxically induced fibrogenesis and sclerosing cholangitis. Proteomics 2006, 6 (24), 6538–48. Huang, K. C.; Park, D. C.; Ng, S. K.; Lee, J. Y.; Ni, X.; Ng, W. C.; Bandera, C. A.; Welch, W. R.; Berkowitz, R. S.; Mok, S. C.; Ng, S. W. Selenium binding protein 1 in ovarian cancer. Int. J. Cancer 2006, 118 (10), 2433–40. Giometti, C. S.; Liang, X.; Tollaksen, S. L.; Wall, D. B.; Lubman, D. M.; Subbarao, V.; Rao, M. S. Mouse liver selenium-binding protein decreased in abundance by peroxisome proliferators. Electrophoresis 2000, 21 (11), 2162–9. Mayr, M.; Chung, Y. L.; Mayr, U.; McGregor, E.; Troy, H.; Baier, G.; Leitges, M.; Dunn, M. J.; Griffiths, J. R.; Xu, Q. Loss of pkcdelta alters cardiac metabolism. Am. J. Physiol. Heart Circ. Physiol. 2004, 287 (2), H937–45. Kato, A.; Yoshidome, H.; Edwards, M. J.; Lentsch, A. B. Regulation of liver inflammatory injury by signal transducer and activator of transcription-6. Am. J. Pathol. 2000, 157 (1), 297–302. Younossi, Z. M.; Baranova, A.; Ziegler, K.; Del Giacco, L.; Schlauch, K.; Born, T. L.; Elariny, H.; Gorreta, F.; VanMeter, A.; Younoszai, A.; Ong, J. P.; Goodman, Z.; Chandhoke, V. A genomic and proteomic study of the spectrum of nonalcoholic fatty liver disease. Hepatology 2005, 42 (3), 665–74. Cavaggioni, A.; Mucignat-Caretta, C. Major urinary proteins, alpha(2u)-globulins and aphrodisin. Biochim. Biophys. Acta 2000, 1482 (1-2), 218–28. Bacchini, A.; Gaetani, E.; Cavaggioni, A. Pheromone binding proteins of the mouse, mus musculus. Experientia 1992, 48 (4), 419–21. Zidek, L.; Stone, M. J.; Lato, S. M.; Pagel, M. D.; Miao, Z.; Ellington, A. D.; Novotny, M. V. Nmr mapping of the recombinant mouse major urinary protein i binding site occupied by the pheromone 2-sec-butyl-4,5-dihydrothiazole. Biochemistry 1999, 38 (31), 9850– 61. Finlayson, J. S.; Baumann, C. A. Mouse proteinuria. Am. J. Physiol. 1958, 192 (1), 69–72. Finlayson, J. S.; Potter, M.; Runner, C. R. Electrophoretic variation and sex dimorphism of the major urinary protein complex in inbred mice: A new genetic marker. J. Natl. Cancer Inst. 1963, 31, 91–107. Clissold, P. M.; Bishop, J. O. Variation in mouse major urinary protein (mup) genes and the mup gene products within and between inbred lines. Gene 1982, 18 (3), 211–20.
Journal of Proteome Research • Vol. 8, No. 10, 2009 4523
research articles (64) Hui, X.; Zhu, W.; Wang, Y.; Lam, K. S. L.; Zhang, J.; Wu, D.; Kraegen, E. W.; Li, Y.; Xu, A. Major urinary protein-1 increases energy expenditure and improves glucose intolerance through enhancing mitochondrial function in skeletal muscle of diabetic mice. J. Biol. Chem. 2009, 284 (21), 14050–14057. (65) Huber, S. A.; Sakkinen, P.; David, C.; Newell, M. K.; Tracy, R. P. T helper-cell phenotype regulates atherosclerosis in mice under conditions of mild hypercholesterolemia. Circulation 2001, 103 (21), 2610–6. (66) Satterthwaite, G.; Francis, S. E.; Suvarna, K.; Blakemore, S.; Ward, C.; Wallace, D.; Braddock, M.; Crossman, D. Differential gene
4524
Journal of Proteome Research • Vol. 8, No. 10, 2009
Iff et al. expression in coronary arteries from patients presenting with ischemic heart disease: Further evidence for the inflammatory basis of atherosclerosis. Am. Heart J. 2005, 150 (3), 488–99. (67) Brea, A.; Mosquera, D.; Martin, E.; Arizti, A.; Cordero, J. L.; Ros, E. Nonalcoholic fatty liver disease is associated with carotid atherosclerosis: A case-control study. Arterioscler. Thromb. Vasc. Biol. 2005, 25 (5), 1045–50. (68) Applera Corporation and MDS Sciex. ProteinPilot software online help v.1.0.15, 2006.
PR9003272
