alexa In Vitro, In Vivo Comparison of Cyclosporin A Induced Hepatic Protein Expression Profiles | OMICS International
ISSN: 2161-0495
Journal of Clinical Toxicology
Like us on:
Make the best use of Scientific Research and information from our 700+ peer reviewed, Open Access Journals that operates with the help of 50,000+ Editorial Board Members and esteemed reviewers and 1000+ Scientific associations in Medical, Clinical, Pharmaceutical, Engineering, Technology and Management Fields.
Meet Inspiring Speakers and Experts at our 3000+ Global Conferenceseries Events with over 600+ Conferences, 1200+ Symposiums and 1200+ Workshops on
Medical, Pharma, Engineering, Science, Technology and Business

In Vitro, In Vivo Comparison of Cyclosporin A Induced Hepatic Protein Expression Profiles

Freek G Bouwman1#, Anke Van Summeren1,2#, Anne Kienhuis3, Leo van der Ven3, Ewoud N Speksnijder4, Jean-Paul Noben5, Johan Renes1, Jos C S Kleinjans2 and Edwin C M Mariman1*

1Department of Human Biology, Maastricht University, P.O. box 616, 6200 MD Maastricht, The Netherlands

2Department of Toxicogenomics, Maastricht University, P.O. box 616, 6200 MD Maastricht, The Netherlands

3Laboratory for Health Protection Research, National Institute of Public Health and the Environment (RIVM), Bilthoven, The Netherlands

4Department of Toxicogenetics, Leiden University Medical Center, 2300 RC, Leiden, the Netherlands

5Hasselt University, Biomedical Research Institute and Transnational University Limburg, School of Life Sciences, Diepenbeek, Belgium

#Both authors contributed equally to this manuscript

*Corresponding Author:
Dr. Edwin C. M. Mariman
Department of Human Biology, Maastricht University
P.O. box 616, 6200 MD Maastricht, The Netherlands
Tel: +31 (0) 43 38
Fax: +31 (0) 43 3670976
E-mail: [email protected]

Received Date April 07, 2016; Accepted Date May 12, 2016; Published Date May 19, 2016

Citation: Bouwman FG, Summeren AV, Kienhuis A, Ven LVD, Speksnijder EN et al. (2016) In Vitro, In Vivo Comparison of Cyclosporin A Induced Hepatic Protein Expression Profiles. J Clin Toxicol 6:299. doi: 10.4172/2161-0495.1000299

Copyright: © 2016 Bouwman FG, et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Visit for more related articles at Journal of Clinical Toxicology


To reduce the amount of laboratory animals which are used to analyze hepatotoxic properties of chemicals and drugs, the development of alternative in vitro models is necessary. Ideally these in vitro models reflect the in vivo toxicological response and cholestasis. In this study the protein expression in livers from C57BL/6 mice after cyclosporin A-induced cholestasis was analyzed. After 25 days of a daily cyclosporine A treatment the cholestatic phenotype was established. An in vitro to this in vivo study comparison was made by using the results of our previous studies with HepG2 and primary mouse hepatocytes. The in vivo proteomics data show cyclosporin Ainduced oxidative stress and mitochondrial dysfunction was actually induced, leading to a decreased mitochondrial ATP production and an altered urea cycle. These processes were also altered by cyclosporin A in the in vitro models HepG2 and primary mouse hepatocytes. In addition, detoxification enzymes like methyl- and glutathione-Stransferases were differentially expressed after cyclosporin A treatment. Changes in these detoxification enzymes were mainly detected in vivo, though primary mouse hepatocytes show a differential expression of some of these enzymes. By means of a functional classification of differentially expressed proteins we demonstrated similarities and differences between in vitro and in vivo models in the proteome response of cyclosporin A-induced hepatotoxicity.


Proteomics; Hepatotoxicity; Cyclosporin A; Liver; In vivo


2DE: 2-Dimensional Gel Electrophoresis; ABC- Transporters: ATP Binding Cassette Transporters; CHOL: Cholesterol; CsA: Cyclosporin A; DIGE: Difference Gel Electrophoresis; GSTs: Glutathione STransferases; MALDI-TOF/TOF-MS: Matrix Assisted Laser Desorption Ionization Time-of-Flight Tandem Mass Spectrometry; LC-MSMS: Nano Liquid Chromatography Tandem Mass Spectrometry; TBIL: Total Bilirubin; TBA: Total Bile Acids


Novel drugs should be recognized as safe for human exposure. With respect to drug-induced toxicity, hepatotoxicity is prominent, because most drugs are metabolized to be eliminated by the liver. The hepatotoxic properties of chemicals and drugs are usually analyzed in in vivo repeated-dose toxicity tests, which involve a high number of laboratory animals. To reduce the amount of laboratory animals, alternative in vitro models are currently developed and their screening properties evaluated [1-3]. Ideally these in vitro models reflect the in vivo toxicological response. Accordingly, in vitro to in vivo comparisons are necessary. By applying Omics technologies it is possible to measure similar endpoints of drug-induced changes between in vitro and in vivo which enable a global comparison of both models [4].

Conventional hepatotoxicity assays rely on the analysis of clinical, hematological, and histopathological parameters. While the conventional assays generally measure only a limited set of biological endpoints, Omics technologies offers the possibility to measure multiple endpoints simultaneously in a single experiment. Currently, transcriptomics studies, where thousands of genes are measured simultaneously, have shown to be successful for this purpose [4-6]. However, transcriptomics investigates the relative mRNA levels of genes which often only moderately correlate with the relative abundance of their protein product. This moderate correlation is due to turnover differences of proteins and mRNA [7].

In addition, post-translational modifications and protein interactions are not detected by transcriptomics, which emphasizes the relevance of proteomics. For example, by applying difference gel electrophoresis (DIGE), proteins are separated based on their pI and molecular weight, so different protein isoforms can be visualized [8].

Previously we investigated the proteome of HepG2 cells and primary mouse hepatocytes after exposure to three well-defined hepatotoxicants [9,10]. These were acetaminophen, amiodarone and cyclosporin A (CsA), of which CsA generated the most prominent response. CsA is an immunosuppressive drug; however, as an adverse side effect it induces cholestasis caused by the inhibition of the bile salt transporters in hepatocytes [11].

The aim of the present study is to identify cholestatic-specific mechanisms in vivo, with use of proteomics. Furthermore, we want to compare these results with our previous in vitro studies with HepG2 and primary mouse hepatocytes [9,10].

For this purpose the hepatic protein expression from C57BL/6 mice after CsA-induced cholestasis was examined. The development of cholestasis at the proteome level was analyzed after 4, 11 and 25 days of a daily dose of CsA. The cholestatic phenotype was established after 25 days and was confirmed by serum biochemistry and histopathology.

DIGE was used to analyze the differentially expressed proteins induced by CsA. The results from our previous studies with HepG2 [9] and primary mouse hepatocytes [10] were used to establish an in vitro - in vivo comparison of CsA-induced protein expression profiles.

Materials and Methods


CsA, CAS-no 59865-13-3, purity minimum 98%, was kindly provided by Novartis, Basel, Switzerland. N,N-dimethylformamide (anhydrous, 99.8%) was purchased from Sigma-Aldrich (Zwijndrecht, The Netherlands), the Protein Assay Kit was from Bio-Rad (Veenendaal, The Netherlands). All chemicals used for DIGE were purchased from GE Healthcare (Diegem, Belgium).


Male C57BL/6 mice, aged 10 weeks at the start of the treatment period (21-27 g), were obtained from Charles River GmbH, Sulzfeld, Germany. Animals were kept under controlled specific pathogen-free conditions (23°C, 40%-50% humidity) under a 12hour light-dark cycle, and housed in groups of five.

Food and tap water were available ad libitum during the whole experiment. Experiments were conducted at the animal facility of the Leiden University Medical Center, under ethical review in accordance with the Dutch law (DEC 09157).

Animal treatment

For the 25 day repeated dose study, forty animals were assigned to eight groups of five mice per group. For the 4 and 11 day repeated-dose study, twenty-four animals were assigned to six groups of four mice per group. After an acclimatization period, treatment protocols were used in which mice were dosed with CsA in olive oil or with the vehicle only, by oral gavage in a volume of 4 ml/kg body weight, for five times per week (working days) between 2:00 and 4:00 pm. In the 25 day study, mice were treated with CsA up to 80 mg/kg body weight.

This study was used for dose range finding, selecting the dose of 26.6 mg/kg body weight which was determined to be the critical effect dose that induced cholestatic clinical chemistry parameters at 25 days of exposure. The animals were sacrificed by inhalation of CO2 and heart puncture at four, eleven and twenty-five days post CsA administration. Blood was collected in 0.8 ml Minicollect serum collection tubes (Greiner Bio-One, Alphen aan de Rijn, The Netherlands) for serum chemistry analyses.

The liver samples were frozen in liquid nitrogen and stored at -80°C until the protein isolation. Liver samples of animals treated with 26.6 mg/kg body weight in the 25 day dose range finding study and liver samples of the 4 and 11 day study were used for proteome analysis.

Serum biochemistry

Alanine transaminase (ALT), Aspartate transaminase (AST), Cholesterol (CHOL), total bilirubin (TBIL) and total bile acids (TBA) were analyzed on a Beckman Coulter LX20 Clinical Chemistry Analyzer using Beckman reagent kits (Beckman Coulter B.V., Woerden, The Netherlands) for TBIL, and CHOL, and a Dialab reagent kit (DIALAB GmbH, Neudorf, Austria) for TBA. A student’s-T test was performed on the results to determine significant differences.


After 24 hours fixation in 4% neutral buffered formalin, liver samples were stored in 70% ethanol until further processing, which included automated dehydration, embedding in paraffin, sectioning at 5 μm, and staining with hematoxylin and eosin.

Sample preparation

Liver samples were ground into fine powder in liquid nitrogen and homogenized in DIGE labeling buffer containing 7 M urea, 2 M thiourea, 4% (w/v) CHAPS and 30 mM TrisHCl. This mixture was mixed thoroughly and subjected to three cycles of freeze thawing with liquid nitrogen to lyse the cells. The homogenate was vortexed for 1 minute and centrifuged at 20 000g for 30 min at 10°C. The supernatant was collected and stored at -80°C until further analysis. Protein concentrations were determined with the Protein Assay Kit from Bio- Rad (Veenendaal, The Netherlands).


The protein labeling and the DIGE were performed as described before [9]. A one-way ANOVA test (P ≤ 0.05) was used to select the significant differential spots between the experimental groups. In addition, two way ANOVA-treatment, two-way ANOVA-time, and two-way ANOVA-interaction were computed to assign statistically significant changes in spot intensity due to the treatment alone, time alone and due to both treatment and time.

The differentially expressed proteins were excised and identified by matrix assisted laser desorption ionization time of flight tandem mass spectrometry (MALDI-TOF/TOF-MS) according to Bouwman et al. [12]. Protein spots that could not be identified by MALDI-TOF/TOFMS were further analyzed by nano liquid chromatography tandem mass spectrometry (LC-MSMS) on an LCQ Classic (Thermo Finnigan) as described before [13].

Functional Classification

The proteins differentially expressed after exposure to CsA in HepG2 and primary mouse hepatocytes, were retrieved from previous studies [9,10]. The Panther classification system ( was used to compare the effect of CsA upon the protein expression in HepG2, primary mouse hepatocytes and in vivo mouse liver. From each experiment the differentially expressed proteins were uploaded onto the Panther classification system.

Furthermore, the Functional Classification Tool of the DAVID Bioinformatics resource 6.7 ( was used to cluster functionally related proteins. For this purpose the differentially expressed proteins from the three experiments were uploaded and classified with the lowest stringency. Afterwards it was retrieved in which experiment these proteins were differentially expressed.


Traditional toxicology parameters

To induce cholestasis C57BL/6 mice were treated with 26.6 mg/kg CsA. Histopathology showed submembraneous vacuolization suggesting cholestasis at 25 days (Figure 1). The plotted serum values per animal for CHOL, TBIL, TBA, ALT and AST are presented in Figure 2.


Figure 1: (A) Mouse liver treated with vehicle for 25 days. (B) Submembraneous vacuolization suggestive of cholestasis in mouse liver upon treatment with 26.6 mg/kg body weight for 25 days. Magnification: 60X.


Figure 2: Plotted serum values per animal for (A) Cholesterol, (B) Total Bilirubine, (C) Total Bile Acids, (D) ALT and (E) AST at 4, 11, and 25 days of exposure to 26.6 mg/kg body weight CsA.

A change (P<0.1) was observed as early as 4 days treatment up till 25 days for the cholestatic parameters CHOL, TBIL, TBA. The general hepatotoxicity markers ALT and AST did not show a significant increase, indicating that a severe stage with liver damage was not yet reached.

DIGE analysis

To analyze the in vivo hepatotoxic effects of CsA, C57BL/6 mice were exposed to CsA for 4, 11 and 25 days with olive oil as a vehicle control. On these time-points the animals were sacrificed and the liver was isolated. The proteins were extracted from these liver samples and the differentially expressed proteins were determined using DIGE. In total 3235 spots could be matched within all images.

With a one-way ANOVA 60 spots were found significantly different (P ≤ 0.05) between all groups. With a two-way ANOVA analysis 92 spots were significantly different (P ≤ 0.05) in response to treatment, 12 spots were significantly different (P ≤ 0.05) in time and 8 spots were differentially expressed for the interaction of the treatment and time (P ≤ 0.05).

Protein identification

The differential spots were included in a pick list. For spot picking and identification a preparative gel was loaded with 150 μg of the internal standard labeled with 300 pmol Cy2 and run in the same way as the analytical gels. Afterwards, with use of the DeCyder™ 7.0 software (GE Healthcare) the preparative gel was matched with the analytical gels.

Protein identification was performed by in-gel digestion followed by MALDI-TOF/TOF-MS and/or LC-MS/MS analysis. The 96 selected protein spots were identified belonging to 86 different proteins. A total of 19 protein spots were isoforms from 9 proteins due to posttranslational modifications or processing of the protein. Figure 3 shows the 2-DE map made from the master gel with the identified differential spots indicated with a number which corresponds to the numbers presented in Table 1.

No Uniprot Gene name Protein description P value Fold change5
one-way anova1 two-way anova, treatment2 two-way anova, time3 interaction treatment/time4 CsA4/C4 CsA11/C11 CsA25/C25
Tricarboxylic acid cycle
39 Q99KI0 Aco2 Aconitatehydratase. mitochondrial precursor 0.018 0.0097 0.11 0.11 -1.17** -1.01 -1.07
59 Q99KI0 Aco2 Aconitatehydratase. mitochondrial precursor 0.13 0.02 0.93 0.23 -1.13 -1 -1.11
72 Q99KI0 Aco2 Aconitatehydratase. mitochondrial precursor 0.021 0.03 0.036 0.17 -1.14 1 -1.07
32 O88844 Idh1 Isocitrate dehydrogenase [NADP] cytoplasmic 0.038 0.0068 0.4 0.18 1.01 1.11 1.14
56 Q9Z2I9 Sucla2 Succinyl-CoA ligase [ADP-forming] subunit beta. mitochondrial 0.068 0.015 0.11 0.89 -1.09 -1.06 -1.09
69 P16332 Mut Methylmalonyl-Coenzyme A mutase 0.01 0.028 0.2 0.29 -1.04 -1.14 -1.33
Carbohydrate metabolism
49v P13707 Gpd1 Glycerol-3-phosphate dehydrogenase [NAD+]. cytoplasmic 0.062 0.014 0.093 0.9 1.1 1.09 1.13
82 Q91Y97 Aldob Fructose-bisphosphatealdolase B 0.087 0.042 0.6 0.083 -1.05 1.16 1.13
4 P17182 Eno1 Alpha enolase 0.0024 0.00028 0.47 0.052 1.02 1.11** 1.05
25 Q9QXD6 Fbp1 Fructose-1.6-bisphosphatase 0.001 0.0047 0.0009 0.4 1.04 1.1* 1.04
58 Q9DBJ1. Pgam1 Phosphoglyceratemutase 1 0.0075 0.018 0.0047 0.54 1.07 1.02 1.06
30 P53657 Pklr Pyruvate kinase. isozymes R/L 0.036 0.0058 0.046 0.17 1.07 1.24** 1.07
85 Q93092 Taldo1 Transaldolase 0.038 0.045 0.18 0.061 -1.04 1.1 1.13
50 P97328 Khk Ketohexokinase 0.073 0.014 0.75 0.15 -1.01 1.15 1.14
93 Q9DBB8 Dhdh Trans-1.2-dihydrobenzene-1.2-diol dehydrogenase 0.037 0.22 0.038 0.085 -1.05 1.09 1.07
88 Q9JLJ2 Aldh9a1 4-trimethylaminobutyraldehyde dehydrogenase 0.19 0.05 0.87 0.18 -1.02 1.12 1.09
Urea cycle
17 Q61176 Arg1 Arginase-1 0.0069 0.0025 0.04 0.24 1.02 1.12* 1.11
78 Q61176 Arg1 Arginase-1 0.25 0.038 0.52 0.61 -1.12 -1.1 -1.03
42 P16460 Ass1 Argininosuccinate synthase 0.15 0.011 0.62 0.8 -1.18 -1.22 -1.29
86 P16460 Ass1 Argininosuccinate synthase 0.3 0.049 0.65 0.55 -1.12 -1.16 -1.35
3 Q8C196 Cps1 Carbamoyl-phosphate synthase I 0.0049 0.00021 0.9 0.36 -1.31** -1.15 -1.18
9 Q8C196 Cps1 Carbamoyl-phosphate synthase I 0.014 0.00081 0.52 0.37 -1.29** -1.13 -1.14
11 P26443 Glud1 Glutamate dehydrogenase 1. mitochondrial precursor 0.026 0.0011 0.54 0.87 -1.13 -1.13 -1.18
Cholesterol and lipid metabolic processes
12 Q920E5 Fdps Farnesyl pyrophosphate synthetase 0.0032 0.0012 0.31 0.019 -1.05 1.37* 1.65**
66 Q920E5 Fdps Farnesyl pyrophosphate synthetase 0.064 0.026 0.091 0.51 1.07 1.13 1.32
19 P52430 Pon1 Serum paraoxonase/arylesterase 1 0.048 0.0033 0.72 0.4 -1.15 -1.36* -1.12
55 Q9QXE0 Hacl1 2-hydroxyphytanoyl-CoA lyase 0.097 0.015 0.38 0.41 1.05 1.15 1.07
45 P50544 Acadvl Acyl-CoA dehydrogenase. very-long-chain specific. mitochondrial precursor 0.02 0.012 0.019 0.87 -1.08 -1.13 -1.11
81 Q8VCW8 Acsf2 Acyl-CoA synthetase family member 2. mitochondrial 0.043 0.042 0.19 0.1 -1.01 1.19 1.06*
26 Q8BWT1 Acaa2 3-ketoacyl-CoA thiolase. mitochondrial 0.041 0.0048 0.31 0.42 -1.1 -1.04 -1.13
68 Q8VCC1 Hpgd 5-hydroxyprostaglandin dehydrogenase [NAD+] 0.16 0.027 0.91 0.29 1.1 1.43 1.14
2 P06801 Me1 NADP-dependent malic enzyme 0.0029 0.00019 0.13 0.53 1.16 1.26** 1.14
37 Q91V92 Acly ATP citrate lyase 0.036 0.0088 0.38 0.083 -1.01 1.22* 1.17
35 P56480 Atp5b ATP synthase beta chain. mitochondrial precursor 0.033 0.0081 0.067 0.8 -1.09 -1.05 -1.1
Protein metabolic processes
80 Q8BWY3 Etf1 Eukaryotic peptide chain release factor subunit 1 0.05 0.041 0.64 0.035 1.05 -1.42** -1.1
43 P49722 Psma2 Proteasome subunit alpha type-2 0.12 0.011 0.67 0.48 1.04 1.13 1.12
64 O88685 Psmc3 26S protease regulatory subunit 6A 0.12 0.023 0.88 0.18 1.01 1.13 1.36*
77 P62334 Psmc6 26S protease regulatory subunit S10B 0.0032 0.038 6E-05 0.52 -1.03 -1.09 -1.03
46 P97371 Psme1 Proteasome activator complex subunit 1 0.15 0.013 0.58 0.77 1.07 1.11 1.14
75 Q9D0R2 Tars Threonyl-tRNAsynthetase. cytoplasmic 0.047 0.033 0.034 0.75 1.06 1.03 1.08
Other metabolic processes
92 Q80X81 Acat3 acetyl-Coenzyme A acetyltransferase 3 0.03 0.12 0.29 0.015 -1.07 1.12 1.08
62 P97355 Srm Spermidine synthase 0.11 0.021 0.28 0.36 1.25 1.06 1.34
14 Q99MR8 Mccc1 Methylcrotonoyl-CoA carboxylase alpha chain. mitochondrial precursor 0.0094 0.0014 0.0049 0.08 -1.15** -1.09 -1.01
6 P40142 Tkt Transketolase 0.00088 0.00044 0.0091 0.16 1.03 1.14** 1.13
23 P40142 Tkt Transketolase 0.014 0.004 0.11 0.29 1.14 1.5* 1.25
22 Q99LB7 Sardh Sarcosine dehydrogenase. mitochondrial precursor 0.026 0.0038 0.11 0.72 -1.07 -1.05 -1.1
54 Q9DBT9 ME2GLYDH Dimethylglycine dehydrogenase. mitochondrial precursor 0.17 0.015 0.58 0.73 -1.14 -1.06 -1.09
27 Q8VC30 Dak Bifunctional ATP-dependent dihydroxyacetone kinase/FAD-AMP lyase 0.023 0.0048 0.42 0.1 1 1.28* 1.26
70 Q8VC30 Dak Bifunctional ATP-dependent dihydroxyacetone kinase/FAD-AMP lyase 0.077 0.028 0.22 0.24 1 1.14 1.08
63 Q9CWS0 Ddah1 NG.NG-dimethylargininedimethylaminohydrolase 1 0.12 0.021 0.14 0.79 1.05 1.1 1.07
24 P52196 Tst Thiosulfate sulfurtransferase 0.0097 0.0041 0.025 0.5 1.19 1.38 1.07
94 P00920 Ca2 Carbonic anhydrase 2 0.000071 0.32 5E-06 0.98 1.06 1.07 1.04
10 Q78JT3 Haao 3-hydroxyanthranilate 3.4-dioxygenase 0.012 0.001 0.46 0.22 1.03 1.13 1.14*
76 Q922D8 Mthfd1 C-1-tetrahydrofolate synthase. cytoplasmic 0.076 0.034 0.2 0.21 -1.02 1.21 1.29
71 P38647 GRP 75 Stress-70 protein. mitochondrial precursor 0.0013 0.03 0.0003 0.58 -1.11 -1.03 -1.11
36 P63038 Hspd1 60 kDa heat shock protein. mitochondrial precursor 0.013 0.0085 0.014 0.78 -1.09 -1.11 -1.06
67 P63038 Hspd1 60 kDa heat shock protein. mitochondrial precursor 0.23 0.027 0.44 0.89 -1.16 -1.12 -1.09
57 Q8CGK3 Lonp1 Lon protease homolog 0.16 0.017 0.87 0.41 -1.08 -1.04 -1.16
79 Q8CGK3 Lonp1 Lon protease homolog 0.064 0.038 0.083 0.38 -1.01 -1.04 -1.08
60 P17742 Ppia Peptidyl-prolylcis-trans isomerase 0.08 0.021 0.23 0.33 1.04 1.19 1.07
95 P24369 Ppib Peptidyl-prolylcis-trans isomerase B 0.021 0.36 0.079 0.014 -1.03 -1.23 1.14
8 P09103 P4hb Protein disulfide-isomerase 0.005 0.00081 0.059 0.38 1.17 1.1 1.27**
89 P07724 Alb Serum albumin precursor 0.00019 0.061 5E-05 0.12 -1.1 1.03 -1.18
44 Q921I1 Tf Serotransferrin precursor 0.01 0.011 0.012 0.45 -1.06 -1.26 -1.19
83 Q921I1 Tf Serotransferrin precursor 0.036 0.043 0.35 0.033 1.13 -1.29 -1.4*
87 Q921I1 Tf Serotransferrin precursor 0.15 0.049 0.94 0.12 1.05 -1.21 -1.31
90 P04938 Mup8 and 10 Major urinary proteins 11 and 8 0.043 0.076 0.082 0.19 -1.07 -1.7 -1.05
29 P40124 Cap1 Adenylyl cyclase-associated protein 1 0.042 0.0054 0.27 0.43 -1.39 -1.3 -1.1
20 P68134 Acta1 Actin. alpha skeletal muscle 0.048 0.0033 0.72 0.4 -1.15 -1.36* -1.12
Xenobiotic metabolism
84 P24472 Gsta4 Glutathione S-transferase 5.7 0.13 0.044 0.19 0.48 -1.04 -1.1 -1.22
1 P15626 Gstm2 Glutathione S-transferase Mu 2 0.00004 5.10E-06 0.26 0.005 1.04 1.56 1.66**
52 O35660 Gstm6 Glutathione S-transferase Mu 6 0.11 0.015 0.78 0.25 1.03 1.17* 1.08
61 P19157 Gstp1 Glutathione S-transferase P 1 0.078 0.021 0.34 0.35 -1.04 -1.32 -1.24
28 Q9WVL0 Gstz1 Maleylacetoacetateisomerase 0.032 0.0052 0.14 0.54 1.05 1.12 1.08
21 P08228 Sod1 Superoxide dismutase [Cu-Zn] 0.042 0.0034 0.48 0.45 1.08 1.14 1.06
18 P09671 Sod2 Superoxide dismutase 0.033 0.0031 0.56 0.28 -1.02 -1.08** -1.04
51 Q9QXF8 Gnmt Glycine N-methyltransferase 0.047 0.014 0.22 0.35 1.02 1.16 1.14
96 Q9QXF8 Gnmt Glycine N-methyltransferase 0.023 0.9 0.0045 0.33 -1.09 1.09 -1.03
74 P40936 Inmt Indolethylamine N-methyltransferase 0.094 0.031 0.81 0.098 1.07 1.4** 1.01
16 O55060 Tpmt Thiopurine S-methyltransferase 0.02 0.0018 0.68 0.19 1.09 1.05 1.2**
15 Q91VF2 Hnmt Histamine N-methyltransferase 0.038 0.0015 0.7 0.82 1.16 1.22 1.25
91 P50247 Ahcy Adenosylhomocysteinase 0.0044 0.092 0.0003 0.02 1.2 -1.12 1.18
34 Q60967 Papss1 Bifunctional 3'-phosphoadenosine 5'-phosphosulfate synthase 2 0.038 0.0075 0.23 0.27 1.07 1.27* 1.1
33 P26443 Glud1 Glutamate dehydrogenase 1. mitochondrial precursor 0.0043 0.0068 0.035 0.72 -1.07 -1.1 -1.11*
65 Q9Z0X1 Aifm1 Programmed cell death protein 8. mitochondrial precursor 0.074 0.024 0.012 0.51 -1.11 -1.03 -1.05
47 Q91VD9 Ndufs1 NADH-ubiquinone oxidoreductase 75 kDa subunit. mitochondrial 0.0063 0.013 0.012 0.13 -1.09 -1.01 -1.17*
53 P05784 Krt18 Keratin. type I cytoskeletal 18 0.063 0.015 0.23 0.31 1.03 1.18 1.27
Not listed
40 Q00896 Serpina1c Alpha-1-antitrypsin 1-3 0.11 0.0098 0.62 0.56 -1.1 -1.34 -1.29
31 Q00897 Serpina1d Alpha-1-antitrypsin 1-4 precursor 0.004 0.0065 0.0033 0.79 -1.13 -1.2 -1.14
73 P00920 Ca2 Carbonic anhydrase 2 0.002 0.031 3E-05 0.62 1.1 1.21 1.07
5 Q01768 Nme2 Nucleoside diphosphate kinase B 0.0053 0.00028 0.2 0.69 1.11 1.08 1.13*
7 Q63836 Selenbp2 Selenium-binding protein 2 0.001 0.00079 0.29 0.004 -1.03 1.16** 1.16**
48 Q9QYG0 Ndrg2 Isoform 1 of Protein NDRG2 0.14 0.013 0.8 0.42 -1.09 1.06* -1.3
13 Q8R086 Suox Sulfite oxidase. mitochondrial precursor 0.00061 0.0013 0.0013 0.22 -1.09 -1.08 -1.22**
41 Q9CZ13 Uqcrc1 Ubiquinol-cytochrome-c reductase complex core protein I. mitochondrial precursor 0.15 0.01 1 0.57 1.08 -1.03 -1.06
38 P70296 Pebp1 Phosphatidylethanolamine-binding protein 1 0.079 0.0093 0.22 0.97 1.13 1.15 1.14

Table 1: Protein identification of differentially expressed proteins in from mouse liver after exposure to CsA after 4, 11 and 25 days. 1P-value from one way ANOVA statistical test between the six groups with each four biological replicates. 2P-value from two way ANOVA (treatment) statistical test between the six groups, which indicates the differences between the control and exposed groups. 3P-value from two way ANOVA (time) statistical test between the six groups, which indicates the differences between the day 4, 11 and 25. 4P-value from two way ANOVA (interaction) statistical test between the six groups, which indicates the interaction between time and treatment. 5The difference in the standardized abundance of the proteins is expressed as the fold change between the control (C) and the treated groups (T). The fold change is calculated by taking the means of standardized volume values for the protein spot in the corresponding groups (C=control, CsA=cyclosporin A, 4=day 4, 11=day 11, 25=day25), values are calculated as T/C and displayed in the range of +1 to + ∞ for increases in expression and calculated as-C/T and displayed in the range of -∞ to -1 for decreased expression. **Indicates significant fold changes (P ≤ 0.05) between the control and the treated group, calculated with a multiple comparison test. *Indicates significant fold changes (P ≤ 0.1) between the control and the treated group, calculated with a multiple comparison test.


Figure 3: Proteome map of the differentially expressed proteins. The identified spots are indicated with a number which corresponds to the numbers used in Table 1.

For the identified protein spots a Tukey’s multiple comparison test was performed, from which 4 spots (methylcrotonoyl-CoA carboxylase alpha chain, aconitate hydratase, and two isoforms of carbamoylphosphate synthase I) were significantly differential after 4 days treatment of CsA. After 11 days of treatment 8 spots were differentially expressed (alpha enolase, selenium-binding protein 2, NADPdependent malic enzyme, transketolase, eukaryotic peptide chain release factor subunit 1, pyruvate kinase, superoxide dismutase 2, and indolethylamine N-methyltransferase).

Twenty five days of treatment induced the differential expression of 6 spots (glutathione S-transferase Mu 2, farnesyl pyrophosphate synthetase, selenium-binding protein 2, sulfite oxidase, thiopurine Smethyltransferase and protein disulfide-isomerase).

In Table 1 the spots are listed with their protein identification and their fold changes between the control and compound. The significant changes are marked with **P ≤ 0.05 or *P ≤ 0.1 accordingly Tukey's multiple comparison test.

Functional classification

Data from HepG2 and primary mouse hepatocytes, were retrieved from previous studies [9,10]. The Venn diagram in Figure 4 illustrates the overlap of differentially expressed proteins induced by CsA in in vivo mouse liver, primary mouse hepatocytes and HepG2 cells.


Figure 4: Venn diagram of the significant differentially expressed proteins induced by CsA in in vivo mouse liver, HepG2 and primary mouse hepatocytes.

In order to compare the protein expression results in mouse and human cells, only mouse orthologues were used. The overlap of the differentially expressed proteins is the highest between the in vitro models PMH and HepG2.

However the in vivo-in vitro comparison of CsA-induced hepatotoxicity based on single protein expression shows only a small overlap in the differentially expressed proteins from the different models.

For that reason we made use of the Panther classification system to identify the functional properties of the identified proteins. The differentially expressed proteins in the liver from exposed mice were mostly involved in metabolic process, immune system process and generation of precursor metabolites and energy (Figure 5).


Figure 5: Classification of the differentially expressed proteins in primary mouse hepatocytes, HepG2 and liver after exposure to CsA with the Panther classification system (

For HepG2 cells and primary mouse hepatocytes the majority of the differential proteins are involved in transport, metabolic and cellular processes (Figure 5). Similar processes between the analyzed in vitro systems are cell cycle, cellular processes, developmental processes and cell adhesion (Figure 5).

In all three models CsA altered proteins which belong to transport and a response to stimulus (Figure 5). The Functional Classification Tool of the DAVID Bioinformatics resource 6.7, revealed 12 clusters which are presented in Table 2.

Cluster Function Enrichment Score in vivo PMH HepG2
1 ATP-binding 9.62041 40 5 16
2 Carbohydrate metabolism 9.293973 8 1 9
3 metal binding 7.999105 10 1 1
4 Methyltransferases 5.247008 4 1 0
5 metal binding 4.844126 8 2 5
6 NAD cofactor 4.807059 1 0 3
7 Detoxification glutathione S-transferase 4.368559 5 0 0
8 Chaperone activity 3.910156 1 7 9
9 Cytoskelet 3.231529 1 1 5
10 mRNA processing 2.748475 0 0 5
11 Protein transport 1.503542 1 1 1
12 response to organic substrate 0.489311 2 1 0

Table 2: Gene classification of the differentially expressed proteins after CsA treatment in vivo, primary mouse hepatocytes and HepG2.


The aim of this study was to identify cholestatic-specific mechanisms in vivo, with use of proteomics. In addition, an in vitro-in vivo comparison of CsA-induced protein expression profiles was established by comparing these results with the results from our previous in vitro studies with HepG2 and primary mouse hepatocytes [9,10] For this purpose, we analyzed the hepatic protein expression in C57BL/6 mice after CsA-induced cholestasis. The cholestatic phenotype was established after 25 days and confirmed by histopathology and serum parameters, which allowed us to search for cholestatic-specific mechanisms in vivo.

Differential protein expression in mice after CsA treatment

CsA inhibits the bile salt export pump (ABCB11), multidrug resistance protein 2 (ABCC2) and P-glycoprotein (ABCB1) in the canalicular membrane vesicles. These ATP Binding Cassette transporters (ABC transporters) are responsible for the bile secretion into the bile canaliculus [14]. Therefore, inhibition of these transport proteins causes the hepatic accumulation of bile salts resulting in cholestasis [15]. Previous studies suggest that accumulated bile acids induce oxidative stress and can cause mitochondrial dysfunction in the liver [16]. Moreover, CsA induced in vitro as well as in vivo oxidative stress, increases lipid peroxidation and depletes the hepatic pool of glutathione [17,18] Superoxide is one of the main reactive oxygen species in the cell which can be converted into oxygen and hydrogen peroxide by superoxide dismutase. In our study cytoplasmatic superoxide dismutase 1 was up-regulated after cyclopsorin A treatment, while mitochondrial superoxide dismutase 2 was downregulated, indicating that CsA induced oxidative stress and mitochondrial dysfunction. Indications for mitochondrial dysfunction are already visible after 4 days of CsA treatment, since all proteins with a significantly differential expression are mitochondrial.

Furthermore, the down-regulation of several enzymes contributing the TCA-cycle like aconitate hydratase, succinyl-CoA ligase (ADPforming) subunit beta and methylmalonyl-coenzyme A mutase suggest a deficient ATP production by the mitochondria. Previously, others have demonstrated an ATP reduction in hepatocytes after exposure to necrotic concentrations of toxic bile salts [19,20]. The reduced ATP was directly due to mitochondrial dysfunction as glycolytic ATP generation was intact [20]. In our study several proteins from the glycolysis pathway were found to be up-regulated. This suggests a compensative mechanism for mitochondrial dysfunction via the glycolytic pathway.

Glutamate dehydrogenase is responsible for the conversion of glutamate to α-ketoglutarate and ammonium, which will bled off to the urea-cycle. The first step of the urea cycle requires ATP for the conversion of NH4+ and HCO3 to carbamoyl phosphate, catalyzed by carbamoyl-phosphate synthase (Cps1). In a later step, ATP is necessary for the conversion of citrulline and aspartate in argininosuccinate, catalyzed by argininosuccinate synthase (Ass1). In our study glutamate dehydrogenase 1, Ass1 and Cps1, were all down-regulated together with other enzymes from the urea cycle.

Rare autosomal recessive disorders of the urea cycle like arginase deficiency and citrin deficiency are associated with neonatal intrahepatic cholestasis [21,22]. Neonatal intrahepatic cholestasis is defined as impaired bilirubin excretion, resulting in jaundice and conjugated hyperbilirubinemia, detected either in a newborn or an infant up to 4 months old [21-23]. The pathogenesis of cholestasis in these urea cycle disorders remains unclear, probably the combination of a primary mitochondrial defect and a delayed maturity of bile acid metabolism, may form a vicious circle in transient neonatal intrahepatic cholestasis [22]. A similar mechanism for cholestasis as observed in neonatal intrahepatic cholestasis may explain downregulation of enzymes from the urea cycle in our study.

Methylation enzymes indolethylamine N-methyltransferase (Inmt), thiopurine S-methyltransferase, glycine N-methyltransferase and histamine N-methyltransferase showed an increased expression after CsA treatment. Methyl conjugation is mainly used for the metabolism of small endogenous compounds such as epinephrine, norepinephrine, dopamine, and histamine but is also involved in the metabolism of macromolecules such as nucleic acids and in the biotransformation of certain drugs [24]. In contrast to other conjugative reactions, methylation leads to less polar compounds that may be less readily excreted from the body [24]. Inmt, thiopurine S-methyltransferase, glycine N-methyltransferase and histamine N-methyltransferase all use S-adenosylmethionine as methyldonor. Previously, it was shown that Sadenosylmethionine protects against CsA [25,26], chlorpromazine [27] and ethenylestradiol-induced cholestasis [28]. Cholestatic rat liver, induced by common bile duct ligation, exhibited increased enzymatic activities of Inmt and thiol methyltransferase [29].

The increased expression of these N-methyltransferases results in an increased conversion of S-adenosylmethionine to S-adenosylhomocysteine, which in turn is converted in homocysteine and adenosine by adenosylhomocysteinase, here differentially expressed. The increased conversion of S-adenosylmethionine and S-adenosylhomocystein activates the trans-sulfuration pathway, leading to the formation of glutathione [27]. Glutathione is responsible for the detoxification of various compounds to protect the cells from oxidative stress. Moreover it plays an important role in bile formation [30]. Previously it has been demonstrated that glutathione is depleted during cholestasis, therefore activation of the trans-sulfuration pathway is necessary to maintain the glutathione levels in the cholestasic liver [30,31]. In addition, sulfite oxidase and bifunctional 3'-phosphoadenosine 5'-phosphosulfate synthase 2, other enzymes of the trans-sulfuration pathway, were also differentially expressed.

Glutathione is a substrate of both conjugation and reduction reactions, catalyzed by glutathione S-transferase enzymes. Mice exposed to CsA, show a differential expression of several glutathione Stransferases (GSTs). GSTs are not only important phase II detoxification enzymes; they are able to bind bile acids and are thought to play a role in the intracellular trafficking of bile acids [32].

Furthermore, CsA induced the differential expression of proteins related to cholesterol biosynthesis and lipid metabolism. Bile acid synthesis is the main route for cholesterol metabolism and is initiated by cholesterol 7α-hydroxylase (CYP7A1). In our study Farnesyl diphosphate synthase (Fdps) showed an increased expression after 11 days of CsA exposure. Fdps is responsible for the formation of farnesyldiphosphate, a key intermediate in cholesterol synthesis and protein farnesylation. Previously drug-induced cholestasis was associated with an increased hepatic cholesterol synthesis, which is in line with the presently observed Fdps expression [33].

In addition, cholesterol synthesis requires the presence of cytosolic acetyl-coA. Previously, we mentioned a down-regulation of the TCAcycle, a mitochondrial source of acetyl-coA. The up-regulation of citrate-lyase and NADP-dependent malic enzyme in our study, suggests a drain of mitochondrial acetyl-coA to the cytosol for cholesterol synthesis.

Furthermore, an increased cholesterol synthesis is associated with a decreased CYP7A1 expression as a protective adaptive response to reduce cellular bile accumulation [33]. Probably these two mechanisms are the cause of high plasma cholesterol in cholestasis. In a previous study, in which transcriptomics analysis was performed on liver samples used for proteome analysis in this study, CYP7A1 downregulation was observed as one of the strongest effects upon progression of cholestasis in mice and was interpreted as an adaptive response [34].

Serum paraoxonase/arylesterase 1 (Pon1) is an antioxidant enzyme responsible for the detoxification of organophosphates and prevention of low-density lipoprotein oxidative modification. Our study shows a down-regulation of Pon1 after CsA treatment. Rats treated with CCl4 showed a reduced activity of hepatic Pon1 together with increased lipid peroxidation [35]. Furthermore, Pon1 was down-regulated in the protein extract from rat liver exposed to acetaminophen [36] However, Pon1 failed as candidate marker for drug-induced hepatotoxicity, because it was not consistently altered in response to several hepatotoxicants [37].

In vivo-in vitro comparison

Previously we analyzed CsA-induced cholestasis in HepG2 cells [9] and in primary mouse hepatocytes [10]. An in vivo-in vitro comparison of CsA-induced hepatotoxicity based on single protein expression is difficult, partly because there is only a small overlap in the differentially expressed proteins from the different models (Figure 4). Therefore the differentially expressed proteins from these models were classified based on the GO-terms with the Panther classification system. This classification revealed a similar outcome for HepG2 and primary mouse hepatocytes. The majority of the differentially expressed proteins are involved in metabolic processes, cellular processes and transport. Similar processes between the analyzed in vitro systems are cell cycle, cellular processes, developmental processes and cell adhesion (Figure 5). In all three models CsA altered proteins belonging to transport and a response to stimulus (Figure 5). For a more detailed overview the proteins were clustered according to their function with DAVID Bioinformatics Resources 6.7. Cluster 1, 2 and 5 contain differentially expressed proteins from the three models. Cluster 1 involves ATP-binding and mitochondrial proteins and cluster 2 contains proteins from carbohydrate metabolism. Cluster 5 refers to binding of metal ions, which is a characteristic of many metabolic enzymes like the glycolytic enzymes. These clusters show that CsA induces in vivo as well as in vitro mitochondrial dysfunction and changes in the energy metabolism, as also described before. Mitochondria are critical targets for drug toxicity; therefore mitochondrial dysfunction and oxidative stress are often seen in druginduced hepatotoxicity [38]. The differential expressions of proteins from energy metabolism are probably a first indication of a toxicological response. However, several proteins from energy metabolism are often detected in comparative proteomics and considered as proteins from a general stress response [39]. Therefore using those proteins as specific markers for cholestasis should be done with caution.

Methyltransferases which are described earlier as detoxifying enzymes constitute cluster 4. Almost all methyltransferases were detected only in vivo, however Inmt was also detected in primary mouse hepatocytes. Furthermore, in vivo CsA induced cholestasis was accompanied with the differential expression of glutathione-Stransferases (cluster 7). In primary mouse hepatocytes we have also detected some glutathione-S-transferases, although they were not significantly changed after CsA treatment and were therefore excluded from the cluster analysis. We believe that both methyltransferases and glutathione-S-transfereses are important detoxification proteins in CsA-induced cholestasis. These proteins were mainly detected in vivo but some also in primary mouse hepatocytes. However, they were not observed in HepG2, suggesting an under-representation of drugmetabolizing enzymes in HepG2 [40,41].

Characteristic for HepG2 cells, proteins responsible for mRNA processing were differentially expressed (cluster 10). These proteins have central roles in DNA repair, telomere elongation, cell signaling and in regulating gene expression at transcriptional and translational level [42]. Furthermore, heterogeneous nuclear ribonucleoproteins (hnRNPs) play a role in tumor development and are up-regulated in various cancers [42]. Previously it was shown that carcinoma cell lines including HepG2 cells have a higher expression of hnRNPs than human intestinal epithelium [43]. Therefore the expression of these proteins can be ascribed to the carcinoma character of the HepG2 cell line. Apparently CsA induced a down-regulation of these proteins and probably decreased cell proliferation.

In vitro, CsA mainly induced the differential expression of chaperone proteins (cluster 8), while this was less observed in vivo . Previously we hypothesized that CsA induces ER stress, with an altered chaperone activity together with a disturbed protein transport resulting in a decreased protein secretion [9]. However, this reaction of CsA seems characteristic for in vitro models. Possibly in vitro cell cultures are more sensitive because they are in direct contact with the toxicant and they function independent from other cells/organs.

Our analysis identified the similarities and differences in in vitro and in vivo models with respect to the response to CsA induced hepatotoxicity. Previous studies have shown the potential of the current in vitro models to detect drug-induced hepatotoxicity [2,9,44]. However the different toxicant-induced responses between in vivo and in vitro models explain the current difficulties to validate in vitro biomarkers against in vivo models.


This work was supported by the Netherlands Genomics Initiative/ Netherlands Organization for Scientific Research (NWO), [grant number: 050-060-510]. We thank Erik Royackers from the Biomedical Research Institute of Hasselt University for his technical support of the LC-MS/MS analysis. The authors thank Piet Beekhof and Dr. Eugene Janssen for clinical chemistry analyses and Joke Robinson for histopathology analyses.


Select your language of interest to view the total content in your interested language
Post your comment

Share This Article

Relevant Topics

Article Usage

  • Total views: 8196
  • [From(publication date):
    June-2016 - Mar 18, 2018]
  • Breakdown by view type
  • HTML page views : 8080
  • PDF downloads : 116

Post your comment

captcha   Reload  Can't read the image? click here to refresh

Peer Reviewed Journals
Make the best use of Scientific Research and information from our 700 + peer reviewed, Open Access Journals
International Conferences 2018-19
Meet Inspiring Speakers and Experts at our 3000+ Global Annual Meetings

Contact Us

Agri & Aquaculture Journals

Dr. Krish

[email protected]

1-702-714-7001Extn: 9040

Biochemistry Journals

Datta A

[email protected]

1-702-714-7001Extn: 9037

Business & Management Journals


[email protected]

1-702-714-7001Extn: 9042

Chemistry Journals

Gabriel Shaw

[email protected]

1-702-714-7001Extn: 9040

Clinical Journals

Datta A

[email protected]

1-702-714-7001Extn: 9037

Engineering Journals

James Franklin

[email protected]

1-702-714-7001Extn: 9042

Food & Nutrition Journals

Katie Wilson

[email protected]

1-702-714-7001Extn: 9042

General Science

Andrea Jason

[email protected]

1-702-714-7001Extn: 9043

Genetics & Molecular Biology Journals

Anna Melissa

[email protected]

1-702-714-7001Extn: 9006

Immunology & Microbiology Journals

David Gorantl

[email protected]

1-702-714-7001Extn: 9014

Materials Science Journals

Rachle Green

[email protected]

1-702-714-7001Extn: 9039

Nursing & Health Care Journals

Stephanie Skinner

[email protected]

1-702-714-7001Extn: 9039

Medical Journals

Nimmi Anna

[email protected]

1-702-714-7001Extn: 9038

Neuroscience & Psychology Journals

Nathan T

[email protected]

1-702-714-7001Extn: 9041

Pharmaceutical Sciences Journals

Ann Jose

[email protected]

1-702-714-7001Extn: 9007

Social & Political Science Journals

Steve Harry

[email protected]

1-702-714-7001Extn: 9042

© 2008- 2018 OMICS International - Open Access Publisher. Best viewed in Mozilla Firefox | Google Chrome | Above IE 7.0 version