Identification of Novel Molecular Candidates for Acute Liver Failure in Plasma of BALB/c Murine Model Sa Lv,† Lai Wei,*,† Jiang-hua Wang,† Jing-yan Wang,‡ and Feng Liu† Peking University People’s Hospital, Peking University Hepatology Institute, Beijing 100044, China, and Department of Infectious Disease, the Second Affiliated Hospital of China Medical University, Shenyang 110004, China Received March 28, 2007
In an effort to identify proteins involved in the disease process of acute liver failure (ALF), we investigated changes in the plasma proteome associated with D-galactosamine/lipopolysaccharide (GalN/LPS) treatment of BALB/c mice. The plasma samples from mice with ALF and control were screened for potential differences using two-dimensional electrophoresis followed by liquid chromatographyelectrospray ionization-tandem mass spectrometry or matrix associated laser desorption ionizationtime-of-flight mass spectrometry. The expression levels of candidate protein named phosphatidylethanolamine binding protein (PEBP) in plasma and liver, brain tissues were confirmed by western blot and RT-PCR analyses. Results were confirmed in plasma samples of human beings. Seven proteins existed in plasma of GalN/LPS-treatment animals only but not in controls. They included PEBP, regucalcin, Cu/Zn superoxide dismutase, glyoxalase 1, malate dehydrogenase, proteasome subunit alpha type 1, and HPMS haptoglobin precursor. Two proteins, proteasome subunit alpha type 5 and apolipoprotein A-I precursor, were up-regulated by GalN/LPS, and one protein, HPMS haptoglobin precursor, was down-regulated by this treatment. Western blot analysis confirmed the results that PEBP protein levels increased significantly in plasma and liver tissues only in ALF mice, but not in surviving mice treated with GalN/LPS. Further analysis revealed that GalN/LPS also induced up-regulation of PEBP mRNA levels in liver tissues. Importantly, plasma obtained from ALF patients, but not from healthy volunteers or from hepatitis patients, also contained detectable levels of PEBP. The present study show that PEBP may be a potential plasma biomarker for ALF diagnosis and participate in the pathphysiological process of ALF. Keywords: proteomics • phosphatidylethanolamine binding protein • D-galactosamine • lipopolysaccharide • twodimensional electrophoresis
Introduction Acute liver failure (ALF) is a rare condition in which rapid deterioration of liver function results in coagulopathy and altered mentation in previously normal individuals.1 This fascinating clinical syndrome may be induced by a number of noxious agents, especially viruses and drugs. Patients with ALF have a high mortality. Two recent studies yielded spontaneous survival rates of 252and 14%.3 The only definitive therapy is liver transplantation, which is associated with a short-term survival of 67%.4 However, because of the limited number of liver donors and the high expense of transplantation, developing early diagnosis biomarkers and effective non-surgery therapies remain key goals for improving overall survival rates. The high throughput analysis of proteomes is a powerful tool for the identification disease-associated proteins, some of * To whom correspondence should be addressed. Lai Wei, MD, Professor, No.11 Xizhimen South Street, Beijing 100044, China; Telephone, 861088325566; Fax, 8610-68322662; E-mail,
[email protected],
[email protected]. † Peking University People’s Hospital. ‡ the Second Affiliated Hospital of China Medical University.
2746
Journal of Proteome Research 2007, 6, 2746-2752
Published on Web 06/15/2007
which may be potential drug targets and diagnostic markers. Up to now, proteomics has led to many advances in the understanding of liver function and disease, especially with regard to hepatocellular carcinoma,5 liver cirrosis,6,7 and hepatitis.8,9 However, no proteomic analyses of liver failure have been reported. In fact, because of the rarity of ALF and the limited number clinical samples, it is difficult to study this disease in depth, and very few controlled therapy trials have been performed. Animal models provide an alternative strategy to investigate the pathophysiological mechanisms of ALF. Administration of D-galactosamine (GalN) together with lipopolysaccharide (LPS) is often used as an animal model of ALF.10,11 Plasma proteins, which comprise not only “true” plasma proteins but also “leakage” proteins from tissues, generally function to maintain physiological homeostasis. Tissue leakage products can be released into the plasma as the result of cell death or injury. These proteins may provide direct information about the disease process and may also serve as novel biomarkers for diagnosis and treatment. 10.1021/pr0701759 CCC: $37.00
2007 American Chemical Society
research articles
Identification of Molecular Candidates for ALF in Plasma
Here, we investigated changes in the plasma proteome associated with GalN/LPS-induced ALF. Many differentially altered plasma proteins were identified, including phosphatidylethanolamine binding protein (PEBP). Further experimentation revealed that PEBP is selectively present in animal liver tissues exposed to GalN/LPS as well as in the plasma of patients with liver failure, making it a potential plasma biomarker for ALF diagnosis. Thus, comparative proteomic analyses, such as employed here, will be a useful tool for identifying ALF biomarkers.
Table 1. Characteristics of the Acute Liver Failure (ALF) Patients patient no.
gender
age
ALF etiology
HEa
outcome
PEBPb expression
HBV-infection anti-tuberculosis drug anti-tuberculosis drug
ΙΙ ΙΙΙ
died recovery
(+) (+)
-
died
(+)
1 2
M F
66 23
3
F
51
Materials and Methods
a Hepatic encephalopathy (HE) was graded as Acharya et al. described.12 Plasma phosphatidylethanolamine binding protein (PEBP) levels as measured by western blot.
Animals and Experimental Protocols. Male BALB/c mice (Academy of Military Medical Sciences), aged six to eight weeks, were randomly divided into eight groups (n ) 8 per group). For induction of ALF, mice were given intraperitoneal injections of D-Galactosamine (600 mg/kg body weight; Sigma, USA) and lipopolysaccharide (LPS; 8 µg/kg body weight; Sigma). Control animals were intraperitoneally injected with GalN, LPS, or saline. Plasma samples, liver tissue samples, and brain tissue samples were obtained at 5 and 24 h after injection and were stored at -80 °C until analysis. All animal experimentation procedures were approved by the Ethics Committee of Peking University People’s Hospital, prior to study commencement. All animals were housed and cared for in rooms maintained at a constant temperature and humidity. Food and water were allowed ad libitum. Food was withdrawn the evening prior to experimentation. Blood Biochemistry and Histopathological Analysis of Liver Tissue. Plasma alanine transaminase (ALT) levels were determined using an automatic analyzer (HITACHI, Japan). The liver tissues were fixed in 10% phosphate-buffered formalin and embedded in paraffin. Serial sections of 5 µm thickness were obtained using a rotatory microtome (Microm GmbH, Germany) and were stained with hematoxylin-eosin for histopathological analysis. Two-Dimensional Electrophoresis (2-DE) of Plasma Samples. Plasma protein samples were pooled, and lipid was removed by chloroform extraction. Proteins were precipitated with cold acetone and dissolved in buffer containing 7 M urea, 2 M thiourea, 4% CHAPS, 1 mM EDTA, 50 mM DTT, and 1 mM PMSF. Total protein concentrations were determined with the Bradford method. Proteins (2 mg) were solubilized in rehydration buffer containing 6 M urea, 2% CHAPS, 20 mM DTT, 0.5% IPG buffer, and a trace amount of bromophenol blue. Isoelectric focusing (IEF) was performed with commercially available preformed immobilized pH gradients (nonlinear, pH 3-7, 24 cm; Amersham, Sweden) using an IPGphore Isoelectric Focusing System (Amersham). IPG strips were rehydrated with samples at 40 V for 10 h. Proteins were then focused at 8000 V for a total of 15 000 Vh at 20 °C. Following a threestep equilibration (two 15 min incubations in DTT buffer and then a 15 min incubation in iodoacetamide buffer), the IPG strips were positioned on 12% polyacrylamide gels. SDS-PAGE was run at constant current of 5 W/gel for 1 h, and the current was switched to 10 W/gel until the bromphenol blue dye front reached the bottom of the gel. The gels were then stained by Coomassie Brilliant Blue G-250 and scanned using Labscan (Amersham). 2-DE of pooled samples was repeated three times. Image Analysis. Digitized images of stained gels were analyzed using ImageMaster 2D ver5.1 software (Amersham). The match analysis was done in an automatic mode, and
further manual editing was performed to correct the mismatched and unmatched spots. The image with the highest number of spots was selected as the reference gel. The relative volume of each spot was considered to represent its expression level. Only spots that were present in experimental gels and were altered at least 2-fold compared to control gels were considered to be significant and were identified by mass spectrometry. Protein Digestion and Identification. Spots of interest were excised and digested in-gel with digestion buffer (50 mM NH4HCO3) containing 50 ng of trypsin (sequencing grade; Roche). The digestion buffer containing the digested peptides was vacuum-dried to final volume of about 10 µL and stored at -20 °C. For liquid chromatography-electrospray ionization-tandem mass spectrometry (LC-ESI-MS/MS), vacuum-dried peptide extracts were dissolved in 0.1% formic acid/ 2% acetonitrile to a final volume of 20 µL and eluted with a 10-30% gradient of organic phase (0.1% formic acid/ acetonitrile) for 30 min. Capillary voltage was 22 V at 200 °C, and tandem MS was performed on precursors with charge states of 2, 3, and 4 covering m/z from 400 to 1800. Spectra were recorded on an LCQ DECA XPPLUS (ThermoFinnigan, San Jose, CA). MS/MS spectra were searched against NCBI nr-databank by using the SEQUEST algorithm. In cases where a positive result could still not be obtained, protein spot samples were processed through peptide mass fingerprinting on matrix associated laser desorption ionizationtime-of-flight mass spectrometry (MALDI-TOF MS, Applied Biosystems, MA). The digestion buffer containing digested peptides was suspended in an equal volume of matrix solution (R-cyano-4-hydroxycinnamic acid saturated in 5% trifluoroacetic acid/50% acetonitrile). One microliter of the mixture was applied onto the sample plate of the mass spectrometer. Mass spectra were recorded in the positive ion reflection mode of a MALDI-TOF Voyager DEPRO mass spectrometer (Applied Biosystems). Spectra were externally calibrated using angiotensin I (1296.68 Da) and ACTH 18-39 fragment (2465.20 Da). Internal calibration was carried out with trypsin auto-digestion at 2163.05 Da. MS data were searched against the NCBI nrdatabank by using the Mascot search engine (www.matrixscience.com). Western Blot Analysis. Liver and brain tissues were homogenized in lysis buffer (50 mM Tris-Cl, pH 8.0; 1% NP-40; 150 mM NaCl; 0.1% SDS; 0.02% sodium azide, 100 µg/mL PMSF). Plasma or tissue proteins (50 µg) were electrophoresed on 12% SDS-PAGE and electrotransferred to a nitrocellulose membrane (40 V overnight). Membranes were blocked with nonfat dried milk in TBS containing 0.2% Tween-20 (TTBS) for 1 h at room temperature. Membranes were then incubated with rabbit
b
Journal of Proteome Research • Vol. 6, No. 7, 2007 2747
research articles
Lv et al.
Figure 1. Two-dimensional gels of murine plasma proteins. Results shown are representative of three independent experiments. The images were analyzed by ImageMaster 2D software and proteins (arrows) were identified by LC-ESI-MS/MS or MALDI-TOF MS. Gels are numbered according to experimental group as follows: 1, 5 h after saline injection; 2, 5 h after LPS injection; 3, 5 h after GalN injection; 4, 5 h after GalN /LPS injection; 5, 24 h after saline injection; 6, 24 h after LPS injection; 7, 24 h after GalN injection; 8, 24 h after GalN /LPS injection. The proteins were identified as: PEBP (1), SMP30 (2, 3), Cu/Zn superoxide dismutase (4), glyoxalase 1 (5), malate dehydrogenase (6), proteasome subunit alpha type 1 (7), HPMS haptoglobin precursor (8), apolipoprotein A-I precursor (9), proteasome subunit alpha type 5 (10), and HPMS haptoglobin precursor (11-13).
Figure 2. Spectrum from LC-ESI-MS/MS analysis of the trypsin digestion product of phosphatidylethanolamine binding protein (PEBP).
polyclonal anti-PEBP antibody (dilution 1:1500 for plasma and 1:10 000 for tissues, Upstate Biotechnology, Lake Placid, NY) or rabbit polyclonal antibody β-actin (dilution 1:300, Santa Cruz, CA) overnight at 4 °C. After washing in TTBS three times, membranes were reacted with HRP-labeled goat anti-rabbitIgG (dilution 1:3000, Santa Cruz, CA) for 1 h at room temperature. Immunodetection was performed with the ECL-Plus kit (Pierce Biotechnology) according to the manufacturer’s instructions. Reverse Transcriptase-PCR. Total RNA was isolated from liver tissues by TRIZOL Reagent (Invitrogen, U.S.A.). First-strand cDNA was generated from 500 ng of total RNA using an oligo (dT) primer and SuperscriptIII reverse transcriptase (Invitrogen). PCR (20 cycles) was performed using Taq polymerase (QIAGEN, Valencia) and oligonucleotide primers for mouse PEBP (forward: 5′-AGTATGAGAGTAGGTGTCCCGCC-3′, reverse: 5′-AAATGCTT GGGAACTGCCTGGGG-3′, amplicon size 350 bp) and GAPDH (forward: 5′-GTGAAGGTCGGTGTGAACGGAT-3′, reverse: 5′-GCATCCTGCTTCACCAC CTTCTT-3′, amplicon size 788 bp). PCR products were analyzed by fractionation on a 1.2% agarose gel and visualized by ethidium bromide staining. Images were captured using a gel documentation system (Quantity One, Bio-Rad). Patient Protocol. Plasma samples were obtained from three ALF patients. ALF was defined as liver failure with jaundice and 2748
Journal of Proteome Research • Vol. 6, No. 7, 2007
Figure 3. Western blot analysis of plasma PEBP levels in GalN / LPS-treated mice. Plasma PEBP levels were markedly elevated in mice with ALF. PEBP levels for each group were normalized to those of the Group 1 (5 h post saline injection). Similar results were obtained with 2-D gel electrophoresis, as shown in (B). The protein spot corresponding to PEBP is indicated with an arrow. Lane and gel 1: 5 h after saline injection, Lane and gel 2: 5 h after LPS injection, Lane and gel 3: 5 h after GalN injection, Lane and gel 4: 5 h after GalN /LPS injection, Lane and gel 5: 24 h after saline injection, Lane and gel 6: 24 h after LPS injection, Lane and gel 7: 24 h after GalN injection, Lane and gel 8: 24 h after GalN /LPS injection.
an international normalized ratio (INR) value >1.5 in patients without re-existing chronic liver disease and with an illness
