Wide gene expression profiling of ischemia-reperfusion injury in human liver transplantation

Abstract

Ischemia-reperfusion injury (IRI) causes up to 10% of early liver failures in humans and can lead to a higher incidence of acute and chronic rejection. So far, very few studies have investigated wide gene expression profiles associated with the IRI process. The discovery of novel genes activated by IRI might lead to the identification of potential target genes for the prevention or treatment of the injury. In our study, we compared gene expression levels in reperfused livers (RL group) vs. the basal values before retrieval from the donor (basal liver [BL] group) using oligonucleotide array technology. We examined 10 biopsies from 5 livers, analyzing approximately 33,000 genes represented on the Affymetrix HG-U133APlus 2.0 oligonucleotide arrays (Affymetrix, Santa Clara, CA). About 13,000 individual genes were considered expressed in at least 1 condition. A total of 795 genes whose expression is significantly modified by ischemia-reperfusion in human liver transplantation were identified in this study. Some of them are likely to be completely activated by IRI, as they are not expressed in basal livers. The supervised gene expression analysis revealed that at least 12% of the genes involved in the apoptotic process, 12.5% of the genes involved in inflammatory processes, and 22.5% of the genes encoding for heat shock proteins are differentially expressed in RL samples vs. BL samples. Furthermore, IRI induces the upregulation of some genes' coding for adhesion molecules and integrins. In conclusion, we have identified a relevant amount of early genes regulated in the human liver after 7–9 hours of cold ischemia and 2 hours from reperfusion, many of them not having been described before in this process. Their analyses may help us to better understand the pathophysiology of IRI and to characterize potential target genes for the prevention or treatment of the liver injury in order to increase the number of patients that successfully undergo transplantation. Liver Transpl 13:99–113, 2007. © 2006 AASLD.

Orthotopic liver transplantation has become an effective therapeutic approach for end-stage liver diseases. Advances in surgical procedures and immunosuppression protocols have considerably improved patient survival after liver transplantation. However, ischemia-reperfusion injury (IRI), an antigen-independent component of the insult to the liver, represents a major problem affecting the outcome of liver transplantation. IRI causes up to 10% of early liver failures and can lead to a higher incidence of acute and chronic rejection.1 Moreover, with the increasing donor shortage, more functionally “suboptimal or “marginal” livers are being used. Such livers are more susceptible to the damage caused by IRI compared with normal livers.2

Liver IRI is mediated by several processes that lead to hepatocellular damage, which is triggered when the liver is transiently deprived of oxygen during the organ procurement for transplantation and later reoxygenated during reperfusion. These destructive effects arise mostly from acute generation of reactive oxygen species subsequent to reoxygenation that inflict direct tissue injury even though different mechanisms mediate the early, intermediate, and late phase of hepatic IRI. Ischemia activates Kupffer cells, which are the main sources of vascular reactive oxygen species formation during the initial reperfusion period.3–5 The consequent production of free radicals, cytokines, and chemokines by Kupffer cells principally mediates the middle phase of reperfusion injury for up to 6 hours after restoration of the blood flow.6 In addition to the molecular mechanism of direct vascular and parenchymal injury (inactivation of antiproteases and direct cytotoxic effects), reactive oxygen species can promote reperfusion injury through stimulation of the transcription factors nuclear factor [NF]-kB and activator protein-1 (AP-1).7, 8 The postischemic oxidant stress can enhance the expression of genes such as tumor necrosis factor α, inducible nitric oxide synthase, heme oxygenase-1 (HO-1), CXC chemokines, and adhesion molecules.9

The whole process involved in IRI is also regulated by a great number of transcription factors, through many signaling pathways. Gene expression variations following the IRI process have been investigated mainly in rat liver. To our knowledge, the wide gene expression profile associated with this process in human cadaveric liver transplantation has never been investigated in depth. Only 1 paper describing the gene expression profiling of acute liver stress during living donor liver transplantation has been published.10

The aim of the present study was to identify, in the human transplanted liver, genes that modulate their expression level in response to IRI early, using a larger-scale technology. By oligonucleotide microarray analysis of 33,000 known genes, we have identified more than 700 genes whose expression is significantly regulated by ischemia-reperfusion during transplantation of human cadaveric livers. Similarities and differences with previous studies are discussed.

Expression patterns of genes involved in IRI and comparison to the expression profile in basal conditions can provide insight into the changes in gene expression associated with cellular dysfunction and concomitant regulatory pathways underlying ischemia and reperfusion, suggesting better therapeutic protocols for both prevention and treatment of the injury.

IRI, ischemia-reperfusion injury; HO-1, heme oxygenase-1; BL, basal liver; RL, reperfused liver; GO, Gene Ontology; PCR, polymerase chain reaction; IL, interleukin; CCL, cystein cistein ligand; HSP, heat shock protein.

MATERIALS AND METHODS

Experimental Design

In the first 4 months of 2005, 16 OLTs were performed at the Transplantation Center of the Cardarelli Hospital in Naples, Italy. We examined 10 biopsies from 5 livers from cadaveric donors. Two biopsies were collected from each liver: 1 biopsy before explantation from the donor (basal liver [BL] group), as the first step, immediately after opening, before ice was applied, and 1 biopsy about 2–3 hours after liver reperfusion in the recipient organism (reperfused livers [RL] group).

Livers were retrieved from donors classified as “standard” and “not standard,” according to the criteria of the “Italian National Transplantation Center.” Of 5 donors, 4 were “standard”; only LT 10 was “not standard” because of age (83 years). In particular, the donors' ages ranged from 38 to 83 years. There was no hypotension, and steatosis was always less than 15%. Mean cold ischemia time was 7 hours, mean warm ischemia time was 45 minutes and mean hospitalization in the intensive care unit was 5 days. Age, recipient gender, and liver ischemia times are shown in Table 1 together with donor and graft information.

Table 1. Data of Liver Donors and Recipients and Liver Ischemia Time

Donor and Graft Characteristics				Recipient Characteristics		Liver Ischemia Time
Case Code	Age/Gender	Steatosis	ALT/ASTUI/L	Age/Gender	Pathology	Cold Ischemia Time	Warm Ischemia Time
Abbreviations: LT, liver transplantation; ALT, alanine aminotransferase; AST, aspartate aminotransferase; F, female; M, male; HBV, hepatitis B virus; HCV, hepatitis C virus.
LT 8	48/F	<10%	47/31	55/M	Cirrhosis HBV	8 h 50 min	60 min
LT 10	83/F	<10%	107/33	34/M	Cirrhosis HBV	8 h 15 min	40 min
LT 15	40/M	15%	89/75	53/M	Cirrhosis HCV	6 h 50 min	40 min
LT 17	38/M	<10%	41/78	48/F	Cirrhosis HCV	7 h 30 min	50 min
LT 18	58/F	15%	116/107	47/F	Cirrhosis HBV	6 h 50 min	45 min

BL and RL groups were compared by evaluating gene expression before and after liver ischemia-reperfusion. Ten microarray hybridizations were performed using 5 biological replicates per condition.

Sample Collection and RNA Extraction

Liver biopsies for BL samples were collected in different hospitals, whereas biopsies for RL samples were collected at the Transplantation Center of the Cardarelli Hospital. All biopsies were obtained with informed consent given according to protocols approved by the Institutional Ethics Committee. Biopsies were instantly submerged in the RNAlater solution (Quiagen, Courtaboeuf, France) and then frozen in liquid nitrogen and stored at −135°C until RNA extraction.

Liver Transplantation Procedure

For liver retrieval, the inferior vena cava (IVC), suprahepatic vena cava, portal vein, and bile duct were isolated in the donor. The aorta was dissected at the origin of the celiac axis, and the celiac artery dissected from its origin to junction with the hepatic artery. After intravenous administration of 200 IU of heparin, the liver was flushed via a cannula placed in the aorta with cold Celsior preservation solution (Imtix Sangstat, Lyon, France),11 and some ice was placed in the donor abdomen for local cooling. The liver was excised and separated from the diaphragm, the right adrenal gland was placed in a basin with cold Celsior preservation solution for preparation of IVC and portal vein cuffs and then stored in a sealed container with Celsior preservation solution at 4°C for about 7–9 hours before implantation.

Total ischemia time was considered as the window time from devascularization in the donor and portal reperfusion in the recipient. It includes a long period of cold ischemia at 4°C (cold ischemia time), applied intentionally to reduce metabolic activities, and a short period of 40-60 minutes (warm ischemia time) at increasing variable temperature, while performing the vascular reconstruction. Portal circulation was restored by connecting the retrohepatic venae cavae with a continuous suture and performing a termino-terminal anastomosis with the 2 portal veins. The arterial flow and the bile duct were reconstituted with an end-to-end anastomosis.12

After surgery, alanine aminotransferase and aspartate aminotransferase enzymes and bilirubin were evaluated every day. The aspartate aminotransferase level ranged from 243 to 648 on the first day and from 23 to 53 on the fifth day. The alanine aminotransferase ranged from 239 to 424 on the first day and from 44 to 153 on the fifth day. Overall, maximum raise of bilirubin was 5.3 mg/dL (second patient on the third day post-OLT). No rejection episode was observed.

One-year follow-up showed good liver function in all recipients, and no major adverse event was noticed in the clinical course.

Array Processing

All the experiments were performed with Affymetrix HG-U133A Plus 2.0 oligonucleotide arrays (Affymetrix, Santa Clara, CA). Total RNA from each sample was extracted using Trizol reagent (Gibco/BRL Life Technologies, Inc., Gaithersburg, MD) and used to prepare biotinylated target complementary RNA, according to the Affymetrix protocols.13 Quality and amount of starting RNA was confirmed using the Agilent 2100 bioanalyzer (Agilent Technologies, Palo Alto, CA).

Ten micrograms of high-quality total RNA were used to generate first-strand complementary DNA using Affymetrix GeneChip T7- oligo(dT) promoter primer; the second-strand synthesis was performed using the Superscript Choice Kit (Invitrogen Life Technologies, Breda, The Netherlands). In vitro transcription and labeling of complementary RNA was performed with biotinylated UTP and CTP using the Affymetrix IVT labeling Kit. The target complementary RNA generated from each sample was fragmented, washed, and stained according to the instructions provided by Affymetrix. Briefly, spike controls were added to 10-μg fragmented complementary RNA before overnight hybridization. Arrays were then washed and stained with streptavidin-phycoerythrin, before being scanned on an Affymetrix GeneChip scanner 3000.14

Microarray Data Acquisition and Analysis

The Affymetrix Software Microarray Suite (version 5.0) was used to assign to each probe set an “average difference” value corresponding to the expression level of the gene it represented. To make comparisons across different chips, data sets on each chip were scaled to a targeted total fluorescence of 500.

GeneSpring software, version 7.3 (Silicon Genetics, Redwood City, CA), was used for data mining. Raw expression data per gene were normalized to median. Normalized data were log-transformed.

Expression data were prefiltered to reduce noise and so discard “unreliable” genes by using the Cross-Gene error model.15 Furthermore, genes were considered reliable when called present in at least 2 out of 10 samples and with a raw signal higher than 10.0. Data were excluded with standard deviation >0.3 within the control group.

Differentially expressed genes were grouped according to Gene Ontology (GO) consortium classification for biological processes and molecular functions.16 Using the GOTree Machine,17 GO categories have been identified with significantly enriched gene numbers in the gene lists when compared to the Affymetrix chip gene set, used as a reference list.18, 19

Real-Time Quantitative Polymerase Chain Reaction

Expression values of 14 genes from the 5 liver samples were checked by real-time quantitative polymerase chain reaction (PCR). The same batch of total RNA was used both for primary gene expression and validation experiments.

Complementary DNA was synthesized with random hexamer primers starting from 1.5 μg of total RNA using the reverse transcription protocol (Taqman Reverse Transcription, Applied Biosystems, Applera, Monza, Italy). Real-time PCR was performed using SYBR Green I Master Mix (Applied Biosystems, Applera) on the DNA Engine Opticon 2 System (MJ Research, Boston, MA) according to the manufacturer's protocols. Reactions were performed in 20 μL total volume with 0.2 μmol/L primers; nucleotides, Taq DNA polymerase, and buffer were included in the SYBR Green I Master mix. PCR reactions were performed in triplicate. The primers (Primm Biotech Products and Services, Milan, Italy) used for amplification are listed in Table 2. Primer pairs were designed using the Primer 3 software20 to obtain amplicons ranging from 100 to 150 base pair, and specifically designed to span introns or cross intron/exon boundaries. Data normalization was performed using hydroxymethyl-bilane synthase as housekeeping gene.21

Table 2. Primer Pairs Used for Real-Time Quantitative PCR

Probe Set	Gene Name	Acc. number ENSEMBL	Left Primer	Right Primer
202076_at	BIRC2	ENSG00000110330	GGAGAAGGAAAAACAAGCTGAA	TGTTGAAAGAGAGCCATTCTGTT
230499_at	BIRC3	ENSG00000023445	CCAGTTACCCTCATCTACTTGAACA	TGAATGGTCTTCTCCAGGTTC
201236_s_at	BTG2	ENSG00000159388	GTGAGCGAGCAGAGGCTTA	ATGCGAATGCAGCGGTAG
227558_at	CBX4	ENSG00000141582	ACCGGAGGAGAACATCCTG	CACTAGCGGTTTGGGCTTC
206432_at	HAS2	ENSG00000170961	CAAAAATGGGGTGGAAAAAG	ACAGATGAGGCTGGGTCAAG
205601_s_at	HOXB5	ENSG00000120075	TATTCCCCTGGATGAGGAAG	AGGGTCTGGTAGCGGGTAT
215485_s_at	ICAM1	ENSG00000090339	TGATGGGCAGTCAACAGCTA	GCAGCGTAGGGTAAGGTTCTT
228287_at	ING5	ENSG00000168395	GCTGATGCGAGAGCTGGA	AGGCGTTCTGGATCTTCTGC
202068_s_at	LDLR	ENSG00000130164	GACCTGTTCCCGACACCTC	CACGCTACTGGGCTTCTTCT
205192_at	MAP3K14	ENSG00000006062	ACAGGATGGAGGACAAGCAG	AAAGGGACAATTCTGGGTGA
209239_at	NFKB1	ENSG00000109320	CCTGGTGCCTCTAGTGAAAAG	TGAACAATAACCTTTGCTGGTC
202464_s_at	PFKFB3	ENSG00000170525	GTGTGCGACGACCCTACAGT	CTTCTGCCGAGTTGCAGTCT
200688_at	SF3B3	ENSG00000189091	GGAACACGGCAACTTCCTTA	CAGGGTCATCCAGGTCATTC
225721_at	SYNPO2	ENSG00000172403	AGCAGAAGCAGCCCTTACAA	GCTTTCCATGAGCTTGATGAC

Experiments were performed twice. The amplification protocol was as follows: 1 cycle of 10 minutes at 95°C, 43 cycles of 95°C for 15 seconds, 56°C for 20 seconds, 72°C for 20 seconds, plus an extension at 72°C for 3 minutes. The relative expression value was calculated with the formula 2^−ΔΔct.

Statistics

Data from different groups were compared using Student's t test, with Benjamini and Hochberg false discovery rate as multiple testing correction. Statistical significance was established at P < 0.01 during the unsupervised analysis and P < 0.05 during the supervised analysis. Genes were considered differentially expressed with a fold change >1.5 between the 2 conditions. Gene sets were considered enriched with a P < 0.01 when compared to the reference gene list.

RESULTS

Approximately 54,000 probe sets represented on the Affymetrix chip corresponding to about 33,000 human genes were analyzed. The Affymetrix Microarray Suite (version 5.0) software was used to identify presence calls and to quantify gene expression (see “Materials and Methods”). Affymetrix software assigns to each probe set an absolute call (present, absent, or marginal) that represents a qualitative indication of gene transcription within each sample. According to this flag, the total number of genes expressed in adult liver was determined first. About 14,000 probe sets, corresponding to about 9,500 genes, were never called present in our liver samples; 18,000 probe sets, corresponding to about 13,000 individual genes, were called present in 1 of the 2 conditions: basal and reperfused liver. Among them, about 1,300 genes were not expressed in the BL samples; they are likely to be induced by the ischemia-reperfusion injury. On the other hand, approximately 200 genes expressed in the basal samples were no more expressed in all the RL samples.

Raw expression data were normalized and prefiltered to eliminate unreliable data (see “Materials and Methods”) before performing gene expression differential analysis. Then 28,000 probe sets, corresponding to about 16,000 genes expressed in at least 1 sample, were selected.

Power analysis performed using PowerAtlas software (University of Alabama at Birmingham, Birmingham, AL)22 demonstrated that at the threshold 0.01 at which we had chosen to discriminate differentially expressed genes, a sample size of 5 biological replicates per condition corresponded to a true positive probability >92% (Fig. 1).

Figure 1. Probabilities of true positive (PTP) differential expressed genes calculated using PowerAtlas Software. The PTP is the expected proportion of genes declared significantly differentially expressed between 2 conditions that are truly differentially expressed between the 2 populations. The graphic shows PTP in the comparison between RL and BL samples across a variety of sample sizes and significance levels. At the significance level cutoff 0.01 (blue line) that we have chosen to discriminate differentially expressed genes, a sample size of 5 biological replicates per condition corresponds to a true positive probability >92%. Red, green, black and yellow lines indicate, respectively, the significance level cut-off of 0.1, 0.05, 0.001 and 1.0E-4 for Student's t test with Benjamini and Hochberg false discovery rate as multiple testing correction.

Unsupervised Gene Expression Analysis

The early gene expression profile of reperfused livers (RL group) was compared to the basal values (BL group). Based on Student's t test with Benjamini and Hochberg false discovery rate multiple correction, 420 genes were differentially expressed between the 2 groups with a fold change >2 and P < 0.01. Of them, 12 genes were downregulated and 408 genes were upregulated in RL samples. About 375 more genes were differentially expressed; 65 genes were downregulated in RL and 310 were upregulated with a fold change ranging between 1.5 and 2; therefore the total number of differentially expressed genes, with very high statistical significance, was 795 (Fig. 2). Table 3 reports the 50 most upregulated genes in RL group. The downregulated genes are listed in Table 4. The number of downregulated genes may be underestimated because Affymetrix Microarray Suite 5.0 software may assign the flag absent to a gene either when it is really not expressed or when its expression rate is not reliable. Genes with weak expression often give unreliable results, so they may be discarded from the analysis.

Figure 2. Expression profile of 1,063 probe sets differentially expressed in RL vs. BL samples with P < 0.01. Expression values (normalized log transformed data) of 1,063 probe sets, corresponding to 718 genes upregulated and 77 genes downregulated in reperfused livers, are plotted both per patient and per condition. Basal values are plotted on the left side and values after reperfusion are plotted on the right side. Red lines indicate that genes are upregulated in the RL group vs. the control group, and white lines indicate the downregulated genes. Genes were considered differentially expressed by comparing the mean expression values of RL samples with that of BL samples using Student's t test with Benjamini and Hochberg false discovery rate, with a P value <0.01, after filtering expression values for noise and unreliable data. TF, liver transplantation; RL, reperfused liver; BL, basal liver.

Table 3. List of the 50 Most Upregulated Genes in the RL Group vs. the BL Group, With a P < 0.01 according to the Benjamini and Hochberg False Discovery Rate Test

Probe Set	Fold Change	P Value	Common name	Genbank	Description
NOTE: Probe sets are sorted according to descending fold change. Standard deviation of fold changes per gene was always <2.0. Abbreviations: cDNA, complementary DNA; mRNA, messenger RNA.
202768_at	266.3	4.56E-05	FOSB	NM_006732	FBJ murine osteosarcoma viral oncogene homolog B
205476_at	77.2	0.00113	CCL20	NM_004591	Chemokine (C-C motif) ligand 20
215078_at	39.54	0.00125	SOD2	AL050388	Superoxide dismutase 2, mitochondrial
36711_at	38.04	0.000133	MAFF	AL021977	Human DNA sequence from clone CTA-447C4
241716_at	37.33	0.00137	HSPD1	BF965447	Heat shock 60kDa protein 1 (chaperonin)
209189_at	33.1	0.00428	FOS	BC004490	v-fos FBJ murine osteosarcoma viral oncogene homolog
204575_s_at	26.88	0.000381	MMP19	U38321	Matrix metalloproteinase 19
230494_at	24.04	0.000393	SLC20A1	AI671885	Solute carrier family 20 (phosphate transporter), member 1
202581_at	22.8	0.000278	HSPA1B	NM_005346	Heat shock 70kDa protein 1B
202859_x_at	22.72	0.00126	IL8	NM_000584	Interleukin 8
202672_s_at	21.72	0.000285	ATF3	NM_001674	Activating transcription factor 3
202627_s_at	21.22	0.00171	SERPINE1	AL574210	Serine (or cysteine) proteinase inhibitor, clade E member 1
200800_s_at	19.47	0.000304	HSPA1A	NM_005345	Heat shock 70kDa protein 1A
202014_at	18.97	0.000399	PPP1R15A	NM_014330	Protein phosphatase 1, regulatory (inhibitor) subunit 15A
218723_s_at	18.69	0.00278	RGC32	NM_014059	Response gene to complement 32
202643_s_at	17.91	0.000465	TNFAIP3	AI738896	Tumor necrosis factor, alpha-induced protein 3
239818_x_at	17.3	0.00277	TRIB1	AA576947	Phosphoprotein regulated by mitogenic pathways
1554333_at	16.73	0.00171	DNAJA4	BC031044	DnaJ (Hsp40) homolog, subfamily A, member 4
244753_at	14.66	0.00151	ACTN4	BF000430	Actinin, alpha 4
204621_s_at	14.33	0.00163	NR4A2	AI935096	Nuclear receptor subfamily 4, group A, member 2
208744_x_at	13.56	0.000521	HSPH1	BG403660	Heat shock 105kDa/110kDa protein 1
201324_at	13.48	0.000287	EMP1	NM_001423	Epithelial membrane protein 1
203821_at	12.56	0.00361	DTR	NM_001945	Diphtheria toxin receptor
201464_x_at	12.16	0.000245	JUN	BG491844	v-jun sarcoma virus 17 oncogene homolog (avian)
212533_at	11.92	0.00156	WEE1	X62048	WEE1 homolog (S. pombe)
221841_s_at	11.79	0.000372	KLF4	BF514079	Kruppel-like factor 4 (gut)
202340_x_at	11.58	0.00118	NR4A1	NM_002135	Nuclear receptor subfamily 4, group A, member 1
202497_x_at	11.3	0.00185	SLC2A3	AI631159	Solute carrier family 2 member 3
206115_at	10.84	0.00387	EGR3	NM_004430	Bridging integrator 3
208891_at	10.62	0.00064	DUSP6	BC003143	Dual specificity phosphatase 6
204103_at	10.51	0.000304	CCL4	NM_002984	Chemokine (C-C motif) ligand 4
242726_at	10.36	0.00403	ACSL3	BF221850	Acyl-CoA synthetase long-chain family member 3
226553_at	10.09	0.00215	TMPRSS2	AI660243	Transmembrane protease, serine 2
217911_s_at	9.64	0.000225	BAG3	NM_004281	BCL2-associated athanogene 3
205214_at	9.562	0.00373	STK17B	NM_004226	Serine/threonine kinase 17b (apoptosis-inducing)
242963_at	9.5	0.00148	MGC26963	AI160370	CDNA FLJ41298 fis, clone BRAMY2040478
200664_s_at	9.464	0.000225	DNAJB1	BG537255	DNAJB1
242736_at	9.445	0.00207	SORBS1	AI377221	Sorbin and SH3 domain containing 1
232017_at	9.332	0.000443	TJP2	AK025185	Homo sapiens cDNA: FLJ21532 fis, clone COL06049.
242727_at	9.294	0.00064	ARL8	BG032269	ADP-ribosylation factor-like 8
221031_s_at	9.257	0.000234	DKFZP434F0318	NM_030817	Hypothetical protein DKFZp434F0318
209803_s_at	9.18	0.0012	PHLDA2	AF001294	Pleckstrin homology-like domain, family A, member 2
208960_s_at	9.088	0.00389	COPEB	BE675435	Core promoter element binding protein
202464_s_at	9.052	0.000268	PFKFB3	NM_004566	6-phosphofructo-2-kinase/fructose-2,6-biphosphatase 3
206432_at	8.798	0.00217	HAS2	NM_005328	Hyaluronan synthase 2
205114_s_at	8.405	0.00223	CCL3	NM_002983	Chemokine (C-C motif) ligand 3
203068_at	8.262	0.000398	KIAA0469	NM_014851	KIAA0469
205681_at	8.186	0.00284	BCL2A1	NM_004049	BCL2-related protein A1
203751_x_at	8.062	0.000465	JUND	AI762296	Jun D proto-oncogene
204094_s_at	7.873	0.000972	KIAA0669	NM_014779	Homo sapiens KIAA0669 gene product (KIAA0669), mRNA

Table 4. List of All the Downregulated Genes in the RL Samples vs. BL Samples With a Fold Change >1.5 and P < 0.01 When Compared With the Benjamini and Hochberg False Discovery Rate

Probe Set	Fold Change	P Value	Common name	Genbank	Description
NOTE: Standard deviation of fold changes per gene was always <0.2. Abbreviations: mRNA, messenger RNA; cDNA, complementary DNA.
222518_at	0.661	0.00776	ARFGEF2	BF525399	ADP-ribosylation factor guanine nucleotide-exchange factor 2
200614_at	0.658	0.00373	CLTC	NM_004859	Clathrin, heavy polypeptide (Hc)
1555815_a_at	0.657	0.00576	L3MBTL2	AL136564	l(3)mbt-like 2 (Drosophila)
206313_at	0.651	0.0033	HLA-DOA	NM_002119	Major histocompatibility complex, class II, DO alpha
217914_at	0.646	0.00925	TPCN1	NM_017901	Two pore segment channel 1
235471_at	0.645	0.00811	C10orf72	BE858464	7g29d12.x1 NCI_CGAP_Brn23
225910_at	0.644	0.00418	LOC284019	BF514723	Helicase with zinc finger domain
226093_at	0.643	0.00723	DCP1B	AW204088	Decapping enzyme hDcp1b
204594_s_at	0.643	0.00246	FLJ20232	NM_013298	FLJ20232
226510_at	0.642	0.00703	C14orf125	BF435286	Chromosome 14 open reading frame 125
235303_at	0.641	0.00555	RG9MTD3	AV728846	RNA (guanine-9-) methyltransferase domain containing 3
1566956_at	0.64	0.00985	OR7E104P	AL137719	Olfactory receptor, family 7, subfamily E, member 104 pseudogene
242895_x_at	0.636	0.00317	RNF39	R63837	Ring finger protein 39
205752_s_at	0.635	0.00519	GSTM5	NM_000851	Glutathione S-transferase M5
209316_s_at	0.635	0.00286	HBS1L	BC001465	HBS1-like (S. cerevisiae)
221696_s_at	0.635	0.00894	STYK1	AF251059	Protein kinase STYK1
227502_at	0.634	0.00304	FLJ10842	AV649579	LCHN protein
216885_s_at	0.634	0.00819	H326	AK026481	Homo sapiens cDNA: FLJ22828 fis, clone KAIA4051
221063_x_at	0.632	0.00568	RNF123	NM_022064	Ring finger protein 123
223483_at	0.627	0.00737	DELGEF	BC000707	Deafness locus associated putative guanine nucleotide exchange
230214_at	0.62	0.00334	MRVI1	AL044056	Murine retrovirus integration site 1 homolog
1557189_at	0.616	0.00691	DNASE1	AW468509	Deoxyribonuclease I
1557953_at	0.615	0.00712	ZNF36	BG761185	602718526F1 NIH_MGC_49
225499_at	0.614	0.0075	KIAA1272	AW296194	Chromosome 20 open reading frame 74
1556464_a_at	0.614	0.0039	LOC257407	AF086098	Hypothetical protein LOC257407
202607_at	0.614	0.00983	NDST1	AL526632	AL526632 Homo sapiens NEUROBLASTOMA COT 25-NORMAL
224686_x_at	0.611	0.0093	LOC474170	AA045233	FLJ34306 protein/// Hypothetical LOC388397
221871_s_at	0.61	0.00644	TFG	BF057492	TRK-fused gene
202288_at	0.609	0.00801	FRAP1	U88966	FK506 binding protein 12-rapamycin associated protein 1
201620_at	0.609	0.00738	MBTPS1	NM_003791	Membrane-bound transcription factor protease, site 1
217902_s_at	0.607	0.00571	HERC2	NM_004667	Hect domain and RLD 2
219320_at	0.604	0.00532	MYOHD1	NM_025109	Myosin head domain containing 1
223760_s_at	0.604	0.00311	RAB7	AF119891	Predicted protein of HQ2706; Homo sapiens PRO2706 mRNA
49452_at	0.602	0.00243	LOC283445	AI057637	Acetyl-coenzyme A carboxylase beta
212880_at	0.6	0.00894	WDR7	AB011113	WD repeat domain 7
202204_s_at	0.596	0.00489	AMFR	AF124145	Autocrine motility factor receptor
1552279_a_at	0.593	0.000972	MGC9564	AK074161	Sterile alpha and TIR motif containing 1
204691_x_at	0.59	0.00323	PLA2G6	NM_003560	Phospholipase A2, group VI (cytosolic, calcium-independent)
219245_s_at	0.589	0.00311	FLJ13491	AI309636	Hypothetical protein FLJ13491
223351_at	0.589	0.00137	HLC-8	N21028	yx46f01.s1 Soares melanocyte 2NbHM
1565641_at	0.578	0.00321	C16orf45	BE503823	Chromosome 16 open reading frame 45
205601_s_at	0.577	0.00492	HOXB5	NM_002147	Homeo box B5
229223_at	0.576	0.00784	NFATC3	AI038402	Nuclear factor of activated T-cells, cytoplasmic, calcineurin-dep
226450_at	0.574	0.00874	INSR	AV703054	Insulin receptor
242759_at	0.573	0.00762	ZFP64	AI821726	Zinc finger protein 64 homolog (mouse)
228456_s_at	0.572	0.00283	LOC149832	AU151357	Clone IMAGE:5729395, mRNA
225509_at	0.57	0.00516	LOC56757	AI862477	Hypothetical protein FLJ11526
218873_at	0.568	0.00514	FLJ20203	NM_017710	Synonyms: FLJ12923, FLJ23040, KIAA1606, DKFZp761I241
217225_x_at	0.566	0.00284	LOC283820	AL512687	Hypothetical protein LOC283820
200637_s_at	0.566	0.00285	PTPRF	AI762627	Protein tyrosine phosphatase, receptor type, F
205783_at	0.564	0.00995	KLK13	NM_015596	Kallikrein 13
1566646_at	0.563	0.00982	LOC149086	AK057562	Hypothetical protein LOC149086
1560198_at	0.554	0.00278	C14orf70	AV701600	AV701600 ADB Homo sapiens cDNA clone ADBCWF03 5′
225065_x_at	0.548	0.00206	MGC40157	AI826279	wk33e07.x1 NCI_CGAP_Pr22 3′
211973_at	0.548	0.00137	NUDT3	AW341200	Nudix (nucleoside diphosphate linked moiety X)-type motif 3
214955_at	0.543	0.00243	TMPRSS6	AI912086	Transmembrane serine protease 6
225115_at	0.541	0.00679	HIPK2	BF529628	Homeodomain interacting protein kinase 2
228688_at	0.538	0.00838	PSMD7	AA843726	Transcribed sequence with strong similarity prf:2111281A
239586_at	0.534	0.00568	BJ-TSA-9	AA085776	Hypothetical protein MGC14128
225504_at	0.534	0.00698	FLJ21616	AW294031	Hypothetical protein FLJ21616
203279_at	0.525	0.00613	EDEM1	NM_014674	EDEM1
230888_at	0.519	0.00264	HSPC049	AW300278	HSPC049 protein
236153_at	0.517	0.00223	SUHW4	BF447323	Suppressor of hairy wing homolog 4 (Drosophila)
227319_at	0.51	0.00308	C16orf44	AI693862	Chromosome 16 open reading frame 44
160020_at	0.503	0.00506	MMP14	Z48481	Matrix metalloproteinase 14 (membrane-inserted)
203656_at	0.484	0.00238	KIAA0274	NM_014845	KIAA0274
228287_at	0.477	0.00172	ING5	BG054893	Inhibitor of growth family, member 5
231304_at	0.471	0.00307	PPP3R2	AI936596	Glutamate receptor, ionotropic, N-methyl-D-aspartate 3A
200688_at	0.463	0.00342	SF3B3	D13642	Splicing factor 3b, subunit 3, 130kDa
218019_s_at	0.459	0.00304	C21orf97	NM_021941	Chromosome 21 open reading frame 97
1552568_at	0.45	0.00114	TMEM7	NM_031440	Transmembrane protein 7
224653_at	0.444	0.000245	EIF4EBP2	U88989	Eukaryotic translation initiation factor 4E binding protein 2
200617_at	0.392	0.003	KIAA0152	NM_014730	Go_component: integral to membrane [goid 0016021]
226495_at	0.379	0.00349	KIAA1271	BE727883	601564286F1 NIH_MGC_20.
200784_s_at	0.341	0.000225	LRP1	BF304759	Low density lipoprotein-related protein 1
201729_s_at	0.297	0.00258	KIAA0100	NM_014680	KIAA0100

Considering the list of 408 genes upregulated in RL group with a fold change >2 and P < 0.01, 74 GO categories for molecular process were relatively enriched when compared with the complete Affymetrix chip gene set, with a ratio >2 between observed and expected values. Seven categories were relatively enriched in the list of 77 downregulated genes. Table 5 reports the observed and the expected gene numbers in these lists.

Table 5. Gene Ontology Categories for Biological Process Relatively Highly Enriched When Compared With the Complete Affymetrix Chip Gene Set, in the List of 408 Upregulated Genes and 77 Downregulated Genes (Fold Change >2 and P < 0.01)

Upregulated Genes
GO Categories	Observed	Expected	Ratio	Significance
NOTE: The observed and the expected gene numbers in these lists are reported. Data are ordered per observed/expected ratio. Abbreviations: MAPK, mitogen activated protein kinase; JAK-STAT, janus kinase-signal transducers and activators of the transcriptors; MAPKKK, mitogen activated protein kinase cascade.
Circadian rhythm	4	0.40	10.00	0.000524
Response to protein stimulus	10	1.19	8.40	0.000000
Response to unfolded protein	10	1.19	8.40	0.000000
Positive regulation of cell differentiation	4	0.50	8.00	0.001368
Actin filament organization	4	0.58	6.90	0.002425
Negative regulation of transcription from RNA polymerase II promoter	11	1.75	6.29	0.000001
Rhythmic process	5	0.80	6.25	0.001045
Positive regulation of development	4	0.72	5.56	0.005244
Angiogenesis	9	1.62	5.56	0.000030
Blood vessel morphogenesis	9	1.70	5.29	0.000044
Vasculature development	9	1.70	5.29	0.000044
Blood vessel development	9	1.70	5.29	0.000044
Regulation of cell adhesion	5	0.98	5.10	0.002749
Regulation of MAPK activity	7	1.38	5.07	0.000412
JAK-STAT cascade	4	0.80	5.00	0.007696
Anti-apoptosis	15	3.07	4.89	0.000000
Negative regulation of apoptosis	17	3.55	4.79	0.000000
Negative regulation of programmed cell death	17	3.58	4.75	0.000000
Cell cycle arrest	8	1.70	4.71	0.000270
Negative regulation of transcription\. DNA-dependent	13	2.81	4.63	0.000004
Regulation of kinase activity	14	3.31	4.23	0.000005
Regulation of protein kinase activity	14	3.31	4.23	0.000005
Regulation of transferase activity	14	3.34	4.19	0.000006
Positive regulation of protein kinase activity	6	1.46	4.11	0.003209
Positive regulation of transferase activity	6	1.48	4.05	0.003516
Negative regulation of signal transduction	6	1.51	3.97	0.003844
Negative regulation of transcription	15	4.16	3.61	0.000018
Taxis	11	3.13	3.51	0.000290
Chemotaxis	11	3.13	3.51	0.000290
Regulation of transcription from RNA polymerase II promoter	24	6.92	3.47	0.000000
Regulation of apoptosis	30	8.72	3.44	0.000000
Regulation of programmed cell death	30	8.77	3.42	0.000000
Negative regulation of nucleobase\. nucleoside\. nucleotide and nucleic acid metabolism	15	4.40	3.41	0.000034
Negative regulation of cellular metabolism	17	4.98	3.41	0.000010
Locomotory behavior	11	3.26	3.37	0.000416
Negative regulation of cellular physiological process	53	16.04	3.30	0.000000
Apoptosis	43	13.28	3.24	0.000000
Programmed cell death	43	13.33	3.23	0.000000
Cell death	45	14.02	3.21	0.000000
Regulation of cell growth	9	2.81	3.20	0.001981
Negative regulation of progression through cell cycle	13	4.06	3.20	0.000212
Death	45	14.07	3.20	0.000000
Negative regulation of physiological process	53	16.67	3.18	0.000000
Negative regulation of cellular process	56	17.78	3.15	0.000000
Transcription from RNA polymerase II promoter	38	12.14	3.13	0.000000
Negative regulation of cell proliferation	12	3.90	3.08	0.000531
Cell growth	11	3.60	3.06	0.000971
Regulation of cell size	11	3.60	3.06	0.000971
Negative regulation of biological process	58	19.22	3.02	0.000000
MAPKKK cascade	7	2.33	3.00	0.008717
Induction of apoptosis	11	3.74	2.94	0.001305
Induction of programmed cell death	11	3.74	2.94	0.001305
Regulation of growth	9	3.07	2.93	0.003657
Response to chemical stimulus	25	8.61	2.90	0.000002
Negative regulation of metabolism	17	5.88	2.89	0.000086
Regulation of enzyme activity	17	5.99	2.84	0.000107
Inflammatory response	15	5.33	2.81	0.000296
Growth	12	4.45	2.70	0.001711
Regulation of cell proliferation	20	7.45	2.68	0.000058
Positive regulation of cell proliferation	9	3.37	2.67	0.006613
Positive regulation of apoptosis	11	4.16	2.64	0.003060
Positive regulation of programmed cell death	11	4.19	2.63	0.003213
Response to abiotic stimulus	26	10.10	2.57	0.000009
Cell motility	15	5.88	2.55	0.000840
Localization of cell	15	5.88	2.55	0.000840
Locomotion	15	5.88	2.55	0.000840
Protein kinase cascade	18	7.08	2.54	0.000268
Regulation of signal transduction	13	5.41	2.40	0.003071
Behavior	12	5.04	2.38	0.004700
Response to wounding	23	9.67	2.38	0.000107
Regulation of progression through cell cycle	26	11.42	2.28	0.000078
Regulation of cell cycle	26	11.45	2.27	0.000081
Cell proliferation	29	13.36	2.17	0.000070
Cellular morphogenesis	14	6.57	2.13	0.006355
Downregulated Genes
GO categories	Observed	Expected	Ratio	Significance
Catabolism	9	1.96	4.59	0.00011129
Cellular catabolism	7	1.67	4.19	0.00120548
Cellular protein metabolism	17	9.15	1.86	0.00492348
Cellular macromolecule metabolism	17	9.32	1.82	0.00598653
Biopolymer metabolism	16	8.84	1.81	0.00874002
Protein metabolism	18	10.01	1.80	0.00517154
Macromolecule metabolism	23	13.84	1.66	0.00304722
Primary metabolism	32	22.77	1.41	0.00318918

More than 700 genes expressed in adult liver are likely to not be affected by the ischemia-reperfusion process, because their expression is significantly homogeneous (fold change <1.1) in the 2 conditions.

Supervised Gene Expression Analysis

Next, the analysis addressed genes involved in pathways already investigated in the physiopathology of IRI. The genes involved in the apoptotic process were investigated in order to identify genes whose expression is modulated during ischemia-reperfusion. The HG-U133A Plus 2.0 chip contains about 1,000 probe sets, corresponding to about 600 genes, related to apoptosis. Seventy-two apoptosis related genes may be considered early genes in the IRI process, as they are regulated after 2 hours of reperfusion with a fold change >1.5 and P < 0.05. Three of them are downregulated. The complete list of these genes is reported in Table 6.

Table 6. Apoptosis-Related Genes Differentially Expressed in RL Samples vs. BL Samples With a Fold Change >1.5 and P < 0.05

Probe Set	P Value	Fold Change	Common Name	Description
Abbreviations: cDNA, complementary DNA; mRNA, messenger RNA.
201466_s_at	0.00748	21.83	JUN	v-jun sarcoma virus 17 oncogene homolog (avian)
202014_at	0.00063	18.97	PPP1R15A	Protein phosphatase 1, regulatory (inhibitor) subunit 15A
202643_s_at	0.000783	17.91	TNFAIP3	Tumor necrosis factor, alpha-induced protein 3
208891_at	0.00103	10.62	DUSP6	Dual specificity phosphatase 6
217911_s_at	0.000171	9.64	BAG3	BCL2-associated athanogene 3
205214_at	0.00827	9.562	STK17B	Serine/threonine kinase 17b (apoptosis-inducing)
205681_at	0.00649	8.186	BCL2A1	BCL2-related protein A1
217997_at	0.00191	7.719	PHLDA1	Pleckstrin homology-like domain, family A, member 1
202431_s_at	0.0028	7.607	MYC	v-myc myelocytomatosis viral oncogene homolog (avian)
39402_at	0.0152	7.392	IL1B	Interleukin 1, beta
200798_x_at	0.000171	6.484	MCL1	Myeloid cell leukemia sequence 1 (BCL2-related)
210655_s_at	0.0256	5.604	FOXO3A	Forkhead box O3A
201502_s_at	0.000171	5.571	NFKBIA	Nuclear factor of kappa light polypeptide gene enhancer in B-cells inhibitor, alpha
218368_s_at	0.00514	5.342	TNFRSF12A	Tumor necrosis factor receptor superfamily, member 12A
207574_s_at	0.000171	5.143	GADD45B	Growth arrest and DNA-damage-inducible, beta
201631_s_at	0.00595	4.78	IER3	Immediate early response 3
203725_at	0.0011	4.63	GADD45A	Growth arrest and DNA-damage-inducible, alpha
203120_at	0.00335	4.59	TP53BP2	Tumor protein p53 binding protein, 2
221009_s_at	0.0028	4.434	ANGPTL4	Angiopoietin-like 4
202284_s_at	0.00133	4.383	CDKN1A	Cyclin-dependent kinase inhibitor 1A (p21, Cip1)
236402_at	0.0424	3.776	BRAF	v-raf murine sarcoma viral oncogene homolog B1
209308_s_at	0.000852	3.523	BNIP2	BCL2/adenovirus E1B 19kDa interacting protein 2
227345_at	0.00149	3.514	TNFRSF10D	Tumor necrosis factor receptor superfamily, member 10d, decoy with truncated death domain
204121_at	0.0161	3.323	GADD45G	Growth arrest and DNA-damage-inducible, gamma
226525_at	0.0409	3.283	STK17B	yz03f02.s1 Soares_multiple_sclerosis_2NbHMSP Homo sapiens
238509_at	0.0156	3.271	CUL1	cullin 1
202724_s_at	0.00514	3.053	FOXO1A	Forkhead box O1A (rhabdomyosarcoma)
243664_at	0.0142	2.942	TXNL	Thioredoxin-like, 32kDa
208309_s_at	0.0297	2.778	MALT1	Mucosa associated lymphoid tissue lymphoma translocation gene 1
209545_s_at	0.00178	2.771	RIPK2	Receptor-interacting serine-threonine kinase 2
239629_at	0.0042	2.735	CFLAR	CASP8 and FADD-like apoptosis regulator
231775_at	0.0254	2.7	TNFRSF10A	zd33e01.r1 Soares_fetal_heart_NbHH19W Homo sapiens cDNA clone IMAGE:342456 5′ similar to contains Alu repetitive element;, mRNA sequence.
227558_at	0.0376	2.668	CBX4	Chromobox homolog 4 (Pc class homolog, Drosophila)
206665_s_at	0.0409	2.63	BCL2L1	BCL2-like 1
213596_at	0.00514	2.608	CASP4	Caspase 4, apoptosis-related cysteine protease
230499_at	0.00178	2.544	BIRC3	Baculoviral IAP repeat-containing 3
205263_at	0.0191	2.497	BCL10	B-cell CLL/lymphoma 10
212312_at	0.0177	2.492	COX4I2
211919_s_at	0.00824	2.485	CXCR4	Chemokine (C-X-C motif) receptor 4
210405_x_at	0.00576	2.411	TNFRSF10B	Tumor necrosis factor receptor superfamily, member 10b
209239_at	0.00119	2.37	NFKB1	Nuclear factor of kappa light polypeptide gene enhancer in B-cells 1 (p105)
202693_s_at	0.0282	2.333	STK17A	Serine/threonine kinase 17a (apoptosis-inducing)
222728_s_at	0.0297	2.332	MGC5306	NB4 apoptosis/ differentiation related protein; Homo sapiens PNAS-104
205192_at	0.0222	2.33	MAP3K14	Mitogen-activated protein kinase kinase kinase 14
229519_at	0.00514	2.264	FXR1	Fragile X mental retardation, autosomal homolog 1
201739_at	0.0161	2.258	SGK	Serum/glucocorticoid regulated kinase
202076_at	0.0131	2.177	BIRC2	Baculoviral IAP repeat-containing 2
218088_s_at	0.00514	2.167	RRAGC	Ras-related GTP binding C
204005_s_at	0.0382	2.028	PAWR	PRKC, apoptosis, WT1, regulator
201375_s_at	0.0234	2.014	PPP2CB	Protein phosphatase 2 (formerly 2A), catalytic subunit, beta isoform
208652_at	0.00756	2.012	PPP2CA	Protein phosphatase 2 (formerly 2A), catalytic subunit, alpha isoform
225434_at	0.0131	2.003	DEDD2	Death effector domain containing 2
214499_s_at	0.00654	1.925	BCLAF1	BCL2-associated transcription factor 1
200608_s_at	0.041	1.902	RAD21	RAD21 homolog (S. pombe)
222985_at	0.0134	1.879	YWHAG	Tyrosine 3-monooxygenase/tryptophan 5-monooxygenase activation protein
231809_x_at	0.0191	1.867	PDCD7	EST365840 MAGE resequences, MAGC Homo sapiens cDNA, mRNA sequence.
209941_at	0.025	1.821	RIPK1	Receptor (TNFRSF)-interacting serine-threonine kinase 1
210792_x_at	0.0299	1.816	SIVA	CD27-binding (Siva) protein
200071_at	0.025	1.801	SMNDC1	Survival motor neuron domain containing 1
208905_at	0.00819	1.69	CYCS	Cytochrome c, somatic
202886_s_at	0.0146	1.596	PPP2R1B	Protein phosphatase 2 (formerly 2A), regulatory subunit A (PR 65), beta isoform
222158_s_at	0.0278	1.585	PNAS-4	CGI-146 protein
212373_at	0.025	1.582	FEM1B	fem-1 homolog b (C. elegans)
213026_at	0.0225	1.561	APG12L	APG12 autophagy 12-like (S. cerevisiae)
221478_at	0.0297	1.558	BNIP3L	BCL2/adenovirus E1B 19kDa interacting protein 3-like
204780_s_at	0.0409	1.552	TNFRSF6	Tumor necrosis factor receptor superfamily, member 6
201149_s_at	0.0142	1.528	TIMP3	Tissue inhibitor of metalloproteinase 3 (Sorsby fundus dystrophy, pseudoinflammatory)
210101_x_at	0.0316	1.507	SH3GLB1	SH3-domain GRB2-like endophilin B1
212401_s_at	0.00748	1.505	CDC2L1	Cell division cycle 2-like 2
1553178_a_at	0.0297	0.639	SSTR3	Somatostatin receptor 3
205488_at	0.024	0.632	GZMA	Granzyme A (granzyme 1, cytotoxic T-lymphocyte-associated serine esterase 3)
210685_s_at	0.00649	0.564	UBE4B	Ubiquitination factor E4B (UFD2 homolog, yeast)

Genes related to the inflammatory response and heat shock genes were also analyzed. Spotted on the HG133 Plus 2.0 chip are 195 probe sets related to inflammatory response. They correspond to 132 genes, 17 of which are differentially expressed, upregulated or downregulated, and mostly chemokines CC and CXC and interleukins 1 and 8 (Table 7). Other interleukins are not significantly regulated after IRI.

Table 7. Inflammatory Response Genes Differentially Expressed in RL Samples vs. BL Samples With a Fold Change >1.5 and P < 0.05

Probe set	P Value	Fold Change	Common Name	Description
203140_at	0.00104	2.283	BCL6	B-cell CLL/lymphoma 6 (zinc finger protein 51)
207655_s_at	0.0218	1.811	BLNK	B-cell linker
205476_at	0.00211	77.2	CCL20	Chemokine (C-C motif) ligand 20
205114_s_at	0.00534	8.405	CCL3	Chemokine (C-C motif) ligand 3
204103_at	0.000866	10.51	CCL4	Chemokine (C-C motif) ligand 4
209774_x_at	0.00161	2.194	CXCL2	Chemokine (C-X-C motif) ligand 2
206336_at	0.0267	2.481	CXCL6	Chemokine (C-X-C motif) ligand 6 (granulocyte chemotactic protein 2)
211919_s_at	0.0114	2.485	CXCR4	Chemokine (C-X-C motif) receptor 4
221664_s_at	0.0425	1.782	F11R	F11 receptor
209189_at	0.0127	33.1	FOS	v-fos FBJ murine osteosarcoma viral oncogene homolog
205119_s_at	0.00295	3.477	FPR1	Formyl peptide receptor 1
39402_at	0.0181	7.392	IL1B	Interleukin 1, beta
207008_at	0.0127	3.103	IL8RB	Interleukin 8 receptor, beta
206157_at	0.04	4.588	PTX3	Pentaxin-related gene, rapidly induced by IL-1 beta
209545_s_at	0.00211	2.771	RIPK2	Receptor-interacting serine-threonine kinase 2
204924_at	0.0212	1.897	TLR2	Toll-like receptor 2
202859_x_at	0.0002	40.8	IL8	Interleukin 8

For heat shock genes, 261 probe sets that correspond to 180 genes have been tested. Twenty-six genes are differentially expressed between BL and RL samples with a fold change >1.5 and P < 0.05. They are all upregulated except for the heat shock protein HSPC049 gene, which that is 2-fold downregulated in RL samples. The list of these genes is reported in Table 8.

Table 8. Heat Shock Proteins Differentially Expressed in RL Samples vs. BL Samples With a Fold Change >1.5 and P < 0.05

Probe Set	P Value	Fold Change	Common Name	Description
202581_at	0.0004	22.8	HSPA1B	Heat shock 70kDa protein 1B
200800_s_at	0.0004	19.47	HSPA1A	Heat shock 70kDa protein 1A
208744_x_at	0.0006	13.56	HSPH1	Heat shock 105kDa/110kDa protein 1
207714_s_at	0.0183	10.67	SERPINH1	CBP1, CBP2; collagen-binding protein 1
1554334_a_at	0.0006	9.787	DNAJA4	DnaJ (Hsp40) homolog, subfamily A, member 4
200666_s_at	0.0004	7.331	DNAJB1	DnaJ (Hsp40) homolog, subfamily B, member 1
213418_at	0.0115	7.044	HSPA6	Heat shock 70kDa protein 6 (HSP70B')
203811_s_at	0.0042	4.073	DNAJB4	DnaJ (Hsp40) homolog, subfamily B, member 4
219343_at	0.0052	3.626	CDC37L1	Hsp90-associating relative of Cdc37
201841_s_at	0.0151	3.247	HSPB1	Heat shock 27kDa protein 1
210211_s_at	0.0151	2.589	HSPCA	Heat shock 90kDa protein 1, alpha
213330_s_at	0.0372	2.518	STIP1	Stress-induced-phosphoprotein 1 (Hsp70/Hsp90-organizprotein)
200881_s_at	0.0151	2.252	DNAJA1	DnaJ (Hsp40) homolog, subfamily A, member 1
218936_s_at	0.0325	2.209	HSPC128	HSPC128 protein
208811_s_at	0.0119	2.165	DNAJB6	DnaJ (Hsp40) homolog, subfamily B, member 6
214328_s_at	0.0184	2.133	HSPCA	Transcribed sequences
200064_at	0.0142	2.127	HSPCB	Heat shock 90kDa protein 1, beta
202843_at	0.0325	2.117	DNAJB9	DnaJ (Hsp40) homolog, subfamily B, member 9
223486_at	0.0372	2.007	HSPC135	HSPC135 protein
200806_s_at	0.0325	1.93	HSPD1	Transcribed sequence with strong similarity to protein pir:A32800
205133_s_at	0.0372	1.833	HSPE1	Heat shock 10kDa protein 1 (chaperonin 10)
223271_s_at	0.0303	1.772	HSPC129	Hypothetical protein HSPC129
208815_x_at	0.0134	1.629	HSPA4	Heat shock 70kDa protein 4
218728_s_at	0.0368	1.542	HSPC163	HSPC163 protein
230888_at	0.0055	0.519	HSPC049	HSPC049 protein
203665_at	0.0123	0.512	HMOX1	Heme oxygenase (decycling) 1, HO-1

Quantitative real-time PCR analysis on 14 genes was carried out to verify the results of microarray analysis. The genes were chosen as follows: 10 genes upregulated in RL group, 2 downregulated genes, and 2 normoregulated genes. The results confirm the microarray data for these genes (Table 9). Correlation rates between quantitative real-time PCR and microarray data were satisfactory for all the tested genes (r > 0.85) except for the homeobox B5 gene, which is downregulated in RL according to microarray data and normoregulated according to real-time PCR.

Table 9. Comparison Between Quantitative Real-time PCR Analysis and Microarray Data on 14 Genes

Probe Set	Gene Description	RL/BL Microarray	RL/BLqPCR
Abbreviation: qPCR, quantitative PCR.
202076_at	Baculoviral IAP repeat-containing 2 (BIRC2)	2.2	2.0
230499_at	Baculoviral IAP repeat-containing 3 (BIRC3)	2.5	11.8
201236_s_at	BTG family, member 2 (BTG2)	2.4	6.0
227558_at	Chromobox homolog 4 (Pc class homolog, Drosophila) (CBX4)	2.7	2.5
206432_at	Hyaluronan synthase 2 (HAS2)	8.8	34.2
205601_s_at	Homeobox B5 (HOXB5)	0.6	1.1
215485_s_at	Intercellular adhesion molecule 1 (CD54), human rhinovirus receptor (ICAM1)	2.0	11.0
228287_at	Inhibitor of growth family, member 5 (ING5)	0.5	0.6
202068_s_at	Low density lipoprotein receptor (familial hypercholesterolemia) (LDLR)	8.7	11.3
205192_at	Mitogen-activated protein kinase kinase kinase 14 (MAP3K14)	2.3	3.3
209239_at	Nuclear factor of kappa light polypeptide gene enhancer in B-cells 1 (p105) (NFKB1)	2.3	2.0
202464_s_at	6-phosphofructo-2-kinase/fructose-2,6-biphosphatase 3 (PFKFB3)	9.0	5.6
200688_at	Splicing factor 3b, subunit 3, 130kDa (SF3B3)	0.5	0.8
225721_at	Synaptopodin 2 (SYNPO2)	0.8	1.2

DISCUSSION

A total of 795 genes whose expression is significantly modified by ischemia-reperfusion in human liver have been identified during transplantation of livers from cadaveric donors. All of them are early-responding genes, as their expression is significantly altered within 3 hours after reperfusion. The analysis of the enriched GO categories for biological processes in the list of upregulated genes showed a very significant presence of genes involved in signal transduction, intracellular signaling cascade, Janus kinase-signal transducers and activators of the transcriptors (JAK-STAT) and mitogen activated protein kinase (MAPK) activity, apoptosis induction and regulation, cell death, and cell growth. Many genes included in the group of significantly upregulated genes are likely to be completely activated by IRI, as these genes are not expressed at all in the basal samples.

We compared our data with data previously obtained by Borozan et al.10 during living donor transplantation. The authors analyzed 19,000 complementary DNA clones, identifying a set of genes whose expression is impaired in acute but not in chronic liver stress. They demonstrated that 2 hours after reperfusion of transplanted liver, 125 genes were at least 2-fold upregulated and 106 were downregulated with P < 0.01 when compared with basal gene expression. Even though experimental conditions are different, as we considered livers from cadaveric donors instead of living donors and our ischemia times were far longer, our data showed 90% agreement (95/125) with the list of upregulated genes. In particular, when we considered the list of 25 most dysregulated genes that the authors confirmed by real-time PCR, the agreement was quite complete for the upregulated genes. The genes downregulated after reperfusion showed a similar fold change but with poor significance in our data (Table 10). We hypothesize that the expression of many downregulated genes was already impaired in basal samples at BL, as we began our analysis with cadaveric livers.

Table 10. Comparison Between Borozan Data and Our Microarray Data

Probe Set	Fold Change, Borozan et al.	Fold Change, Our Data	P Value, Our Data	Common Name	GenBank Acc.
Abbreviations: ID, identification; Acc., accession number.
Upregulated Genes
202284_s_at	6.85	4.383	0.000136	CDKN1A	NM_000389
224654_at	6.54	3.228	0.000226	DDX21	BG164358
209457_at	3.83	7.78	4.53E-05	DUSP5	U16996
232174_at	3.82	5.155	0.000977	EXT1	AA480392
202768_at	3.39	266	1.70E-07	FOSB	NM_006732
208785_s_at	3.84	2.476	1.00E-05	MAP1LC3B	BE893893
210524_x_at	3.51	1.597	0.202	MT1F	AF078844
236495_at	8.05	3.476	0.000549	PBEF1	AI681868
202071_at	3.4	2.911	0.001	SDC4	NM_002999
211573_x_at	4.73	1.463	0.213	TGM2	M98478
235086_at	5.77	8.078	0.00231	THBS1	AW956580
215033_at	15.3	13.29	0.00468	TM4SF1	AI189753
Downregulated Genes
229229_at	0.26	0.907	0.626	AGXT2	AJ292204
206913_at	0.14	1.913	0.113	BAAT	NM_001701
1554931_at	0.35	0.32	0.168	CYP4A11	BC022851
206930_at	0.3	0.807	0.661	GLYAT	NM_005838
1554303_at	0.34	0.599	0.201	HNMT	AF523356
1562089_at	0.23	0.73	0.569	MGC15937	BC013929
240370_at	0.33	0.722	0.54	PKD2	AI432451
1558027_s_at	0.34	0.615	0.0376	PRKAB2	AL552001
205790_at	0.32	0.742	0.0796	SCAP1	NM_003726
205152_at	0.3	1.259	0.531	SLC6A1	AI003579
203868_s_at	0.34	0.973	0.911	VCAM1	NM_001078

The supervised gene expression analysis revealed that 12% of the genes involved in the apoptotic process are differentially expressed in RL samples vs. BL samples. They are mostly upregulated except for the somatostatin receptor 3, the granzyme A, and the ubiquitination factor E4B, which are downregulated after reperfusion. The 72 upregulated genes are subdivided as follows: 41 proapoptotic genes, 31 antiapoptotic genes.

The 12.5% of the genes involved in the inflammatory process were induced by IRI. The most upregulated genes in this category encoded for chemotactic agents like C-C chemokines cystein cistein ligand (CCL) CCL4, and CCL20. Interleukin (IL) 1 and IL-8 were more than 5-fold upregulated. These results agree with the published data in mouse lung ischemia-reperfusion previously reported by Yamane.23 IL-1 is a strong inducer of hepatic chemokine synthesis.24 CXC chemokines recruit neutrophils into the reperfused liver due to their strong chemotactic activity. There is a strong activation of IL-6 gene in human liver. This result confirms the previous data obtained in lung experiments. The IL-6 receptor and the IL-6 signal transducer are deeply upregulated. IL-6 is an antiinflammatory citokine, as it can downregulate the tumor necrosis factorα. Its expression in liver is completely induced by IRI, as IL-6 is not expressed at all in the basal samples.

The activation of genes encoding for heat shock protein (HSP) is also in agreement with previously reported data.25 About 22.5% of the heat shock genes were differentially expressed in the 2 tested conditions. The most upregulated genes were HSP70, HSP105 and HSP90. HSPs are mostly molecular chaperons. They have already been used to evaluate IRI in rat liver.26

HSP induction can reduce the nuclear binding of proinflammatory transcription factors27 and increase the antioxidant power of cells.28 HSP and HO-1 are considered to be key factors in protection because of their preconditioning mechanisms such as ischemic and heat shock preconditioning.29 HO-1 was about 2-fold upregulated in samples after reperfusion. HO-1 degrades heme into carbon monoxide, iron, and biliverdin and is ubiquitously distributed in mammalian tissues. HO-1 is strongly and rapidly induced by a variety of stimuli and agents that cause oxidative stress under pathological conditions. Increased HO-1 expression has been proposed to reduce inducible nitric oxide synthase activity through accelerated degradation of heme thereby exerting a protective effect in animal models of inflammation.

The comparison between these data and data from literature confirms the activation of the adhesion molecule intercellular adhesion molecule (ICAM1). Other adhesion molecules induced by IRI in our liver samples were chemotactic factors such as interleukin-8 (IL-8) and CCL4 and angiogenic inducers such as the cystein-rich angiogenic inducer 61 (CYR61) and the tumor necrosis factor receptor 12A and integrins such as beta-1 integrins. It is known that interaction between integrins and other adhesion molecules is responsible for the process of extravasation of neutrophils, a crucial prerequisite for hepatocyte damage.30 Inhibiting adhesion molecules production should be helpful in reducing leukocyte adhesion to vasal endothelium and therefore cell injury consequent to this event.

Pathway analysis of all of the differential expressed genes, performed using the Pathway Express software (Department of Computer Science, Wayne State University, Detroit, MI), revealed a possible orderly sequence of gene activation in the dysregulated pathways. Cytokine-cytokine receptor interaction pathway was one of the most altered pathways, as 38 out of 229 genes represented on the chip are upregulated in RL samples, including IL-8 (22-fold upregulated) and chemokine (C-C motif) ligand 3 and 4 and IL-6 and -1 (more than 7-fold upregulated). This pathway regulates the focal adhesion pathway (dysregulated with 28/155 upregulated genes) that in turn may influence 3 other pathways: WNT signaling, apoptosis, and mitogen activated protein kinase (MAPK) signaling pathways (all dysregulated with 23/128, 21/92, and 52/232 upregulated genes in RL, respectively). These 3 pathways cooperate to induce cyclin D2 expression, regulating the cell cycle pathway and therefore promoting the liver regeneration process. Cyclin D2, indeed, is more than 2-fold upregulated in RL samples. In this manner, quiescent hepatocytes are induced to reenter the cell cycle, also stimulated by IL-6. Induction of the liver regeneration process is demonstrated the upregulation of MYC, JUND, JUNB, FOS, and FOSB genes, already identified during an early G1 phase of cell cycle in the regenerating liver.31 These genes are highly expressed in RL samples together with other genes upregulated in G1 such as DUSP6 and the cyclin dependent kinase inhibitor p21 (CDKN1A). MYC, FOS, JUND, and FOSB were not expressed in our basal samples. FOSB was very highly upregulated in RL samples. Other genes activated during late cell cycle phases, such as cyclin D1 and genes involved in DNA replication and mitotic checkpoint control were not yet upregulated in our samples 2–3 hours after reperfusion.

In conclusion we have identified a huge amount of early human genes regulated in surgical phases of liver transplantation. Some of these genes are involved in liver injury and cell death, while other genes play a pivotal role in potential anti-inflammatory, regenerating, and protective mechanisms. The present study focused on the molecular mechanisms of liver injury. Obviously it is possible to hypothesize that there are other mechanisms that influence protein levels, despite gene expression behavior. The discovery of novel genes activated by IRI might lead to identification of potential target genes for the prevention or treatment of the liver injury. In that case we would consider it very helpful to extend the analysis to a proteomic level. The expression rate of the 50 most upregulated genes (Table 3) could potentially be used to check liver conditions according to specific therapeutic protocols.

Acknowledgements

The authors thank Myriam Alcalay, Simone Minardi (FIRC Institute of Molecular Oncology Foundation, Milan), and Pasquale De Luca (Genetic Engineering Center, Naples) for bioinformatic assistance.