Comprehensive analysis of differential gene expression profiles on diclofenac-induced acute mouse liver injury and recovery

　

Heekyoung Chung,a Hyun-Jun Kim,a Ki-Seok Jang,a Mingoo Kim,b Jungeun Yang,c Ju Han Kim,b Yong-Sung Lee,c and Gu Kong a,*

Affiliation and Addresses: a Department of Pathology, and c Department of Biochemistry, College of Medicine, Hanyang University, Seoul 133-791, Republic of Korea;
 and b Seoul National University Biomedical Informatics (SNUBI), Seoul National University College of Medicine, Seoul 110-799, Republic of Korea 

*Corresponding Author. Department of Pathology, College of Medicine, Hanyang University, 17 Haengdang-dong, Seongdong-gu, Seoul, 133-791, Republic of Korea.
 Tel: +82-2-2290-8251, Fax: +82-2-2295-1091, E-mail address: gkong@hanyang.ac.kr (G. Kong).

 Key Words: diclofenac; mouse; liver; toxicogenomics; Applied Biosystems Mouse Genome Survey Microarray

　

　

Abstract

Microarray analysis of RNA from diclofenac-administered mouse livers was performed to establish a global gene expression profile during injury and recovery stages at two different doses. A single dose of diclofenac at 9.5 or 0.95 mg/kg body weight was given orally, and the liver samples were obtained after 6, 24, and 72 h. Histopathologic studies enabled the classification of the diclofenac effect into injury (6, 24 h) and recovery (72 h) stages. By using the Applied Biosystems Mouse Genome Survey Microarray, a total of 7,370 out of 33,012 (22.3%) genes were found to be statistically reliable at p < 0.05 by 2-way ANOVA, and 602 (1.8%) probes at false discovery rate < 5% by Significance Analysis of Microarray. Among the statistically reliable clones by both analytical methods, 49 genes were differentially expressed with more than a 1.625-fold difference (which equals 0.7 in log2 scale) at one or more treatment conditions. Forty genes and 2 genes were identified as injury- and recovery-specific genes, respectively, showing that most of the transcriptomic changes were seen during the injury stage. Furthermore, multiple genes involved in oxidative stress, eicosanoid synthesis, apoptosis, and ATP synthesis showed variable transcript levels upon acute diclofenac administration.

　

　

1. Experimental Design  

 

Array: Applied Biosystems AB1700 mouse chip (33315 probes) 

30 arrays in total with triple biological replicates

　

2. Data

2.1. Raw data

Name Files
DIC_Nontreat_NA_1  
DIC_Nontreat_NA_2  
DIC_Nontreat_NA_3  
DIC_Vehicle_06hr_1  
DIC_Vehicle_06hr_2  
DIC_Vehicle_06hr_3  
DIC_Vehicle_24hr_1  
DIC_Vehicle_24hr_2  
DIC_Vehicle_24hr_3  
DIC_Vehicle_72hr_1  
DIC_Vehicle_72hr_2  
DIC_Vehicle_72hr_3  
DIC_Low_06hr_1  
DIC_Low_06hr_2  
DIC_Low_06hr_3  
DIC_Low_24hr_1  
DIC_Low_24hr_2  
DIC_Low_24hr_3  
DIC_Low_72hr_1  
DIC_Low_72hr_2  
DIC_Low_72hr_3  
DIC_High_06hr_1  
DIC_High_06hr_2  
DIC_High_06hr_3  
DIC_High_24hr_1  
DIC_High_24hr_2  
DIC_High_24hr_3  
DIC_High_72hr_1  
DIC_High_72hr_2  
DIC_High_72hr_3  

　

2.2. Normalized data

    Download normalized data (33315 probes x 30 samples)

    Data are normalized in following procedure.

  1. Assay Normalized Signals are used.
  2. Only probes with Flag < 100 are used.
  3. Missing and filtered valued are imputed with KNN imputation algorithm.
  4. Values are VSN transformed.
  5. Arrays are scaled with quantile normalization.
  6. Normalized valued are rescaled with

　

3. Diagnostic Plots

3.1. Single-chip diagnostic plots

Name Analysis
DIC_Nontreat_NA_1
DIC_Nontreat_NA_2
DIC_Nontreat_NA_3
DIC_Vehicle_06hr_1
DIC_Vehicle_06hr_2
DIC_Vehicle_06hr_3
DIC_Vehicle_24hr_1
DIC_Vehicle_24hr_2
DIC_Vehicle_24hr_3
DIC_Vehicle_72hr_1
DIC_Vehicle_72hr_2
DIC_Vehicle_72hr_3
DIC_Low_06hr_1
DIC_Low_06hr_2
DIC_Low_06hr_3
DIC_Low_24hr_1
DIC_Low_24hr_2
DIC_Low_24hr_3
DIC_Low_72hr_1
DIC_Low_72hr_2
DIC_Low_72hr_3
DIC_High_06hr_1
DIC_High_06hr_2
DIC_High_06hr_3
DIC_High_24hr_1
DIC_High_24hr_2
DIC_High_24hr_3
DIC_High_72hr_1
DIC_High_72hr_2
DIC_High_72hr_3

　

3.2. Between-chip diagnostic plots

[View between-chip diagnostic plots]

　

4. Differentially Expressed Genes (DEGs)

        4.1.  Lists of DEGs

Supplementary Table 1. Microarray results of 7,370 probes found to be statistically reliable at p < 0.05 by 2-way ANOVA
Supplementary Table 2. Microarray results of 4,566 "differentially regulated" probes found to be statistically reliable at p < 0.05 by 2-way ANOVA
Supplementary Table 3. Microarray results of 602 probes found to be statistically reliable by SAM at FDR < 5% 
Supplementary Table 4. Microarray results of 58 "differentially regulated" probes found to be statistically reliable by SAM at FDR < 5%
Supplementary Table 5. Microarray results of 49 "differentially regulated" probes identified by SAM and 2-way ANOVA
Supplementary Table 6. Microarray results of 1,355 injury-specific genes and 392 recovery-specific gene identified by 2-way ANOVA
　

4.2. SOM clustering and Gene ontology and pathway functional enrichment study

                    SOM clustering was performed on each ANOVA group. Then Enrichment study using Gene ontology and pathway information was performed on each cluster.
                    Broken line graphs show general tendency of expression profile for each cluster. (Installation of SVG viewer under Internet Explore is requiered.)   

P-value < 0.05
MTC MDTI MDTx MDxI MDxx MxTI MxTx MxxI Mxxx
UN 176 C A 470 C A 117 C A 547 C A 746 C A 4803 C A 408 C A 103 C A
BF 0 C A 0 C A 0 C A 0 C A 0 C A 371 C A 0 C A 0 C A
BH 0 C A 0 C A 0 C A 0 C A 0 C A 2916 C A 0 C A 0 C A
BY 0 C A 0 C A 0 C A 0 C A 0 C A 1220 C A 0 C A 0 C A

D = Dose
T = TimeClick C for Cluster viewer, GO enrichement analysis
Click A for ArrayXPath, Pathway enrichment analysis
UN = Unadjusted
BF = Bonferoni multiple test correction
BH = Benjamini :: Hochberg multiple test correction
BY = Benjamini & Yekutieli multiple test correction

　

　

Last updated: 2006-05-25