Systematic review with meta-analysis: HIF-1α attenuates liver ischemia–reperfusion injury

Transplantation Reviews xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

Transplantation Reviews journal homepage: www.elsevier.com/locate/trre

Systematic review with meta-analysis: HIF-1α attenuates liver ischemia–reperfusion injury Yingjia Guo a, Li Feng a, Yanni Zhou a, Jiantong Sheng a, Dan Long a, Shengfu Li a, Youping Li a,b,⁎ a Key Laboratory of Transplant Engineering and Immunology of National Health and Family Planning Commission of the People's Republic of China, West China Hospital, Sichuan University, Chengdu, Sichuan, PR China b Chinese Cochrane Centre, Chinese Evidence-Based Medicine Centre, West China Hospital, Sichuan University, Chengdu, Sichuan, PR China

a b s t r a c t Ischemia–reperfusion injury (IRI) induces inevitable complications in liver transplantation. Many studies have demonstrated that hypoxia-inducible factor 1α (HIF-1α) plays an important role in IRI. However, the mechanism of its pleiotropic effect remains unclear. This systematic review provides a comprehensive evaluation of all available evidence concerning the function of HIF-1α in transplant-induced hepatic IRI. Data were obtained through a search of Medline (PubMed), Embase, and the Cochrane Library literature review on the effect of HIF-1α in IRI (from inception to 12/2014). RevMan was used to calculate standardized mean difference (SMD) and 95% confidence intervals (CIs). Forty articles met inclusion criteria with 2 clinical and 38 basic studies. Two clinical trials (n = 68) revealed ischemic preconditioning (IPC) aroused protection after hepatic IRI based on the higher level of HIF-1α in IPC group compared with control group. In vitro studies confirmed the salutary effect of IPC disappearance in the inhabitation of stabilized HIF-1α. In vivo animal studies showed different HIF-1α expression and distribution patterns in the ischemia and reperfusion stage due to distinctive partial oxygen pressure gradient intra-liver, and 5 animal studies (n = 66) showed that stabilized HIF-1α treatment was associated with lower alanine aminotransferase (ALT) (SMD = −1.58; 95% CI = −2.56, −0.52) when compared with unstabilized HIF-1α group. Not only decreased liver IR injury, stabilized HIF-1α during the acute phase of IR could also promote graft regeneration capacity leading to better initial function and survival rate. More rigorous studies are needed to gauge the effectiveness due to insufficient sample size and possible publication bias. © 2015 Elsevier Inc. All rights reserved.

1. Introduction The vulnerability of liver grafts to ischemia-reperfusion injury (IRI) is the main cause of liver grafts primary nonfunction and delayed grafts function [1], thus limiting long-term survival [2]. Mounting evidence suggests that stabilization of hypoxia-inducible factor 1α (HIF-1α) protects liver grafts from IRI, but none of therapeutic approaches have found the way into routine clinical practice to dampen liver IRI. In 1960, Jennings et al. first described IRI in canine coronary ligation animal models, where IRI was defined as tissue damage that occurred when blood supply returned to tissue after a period of ischemia [3]. In 1986, Murry et al. first reported that ischemic preconditioning (IPC) protects the heart from IRI [4]. Subsequent studies confirmed this phenomenon in other organs. In 2000, Clavien et al. noted that 10 minutes of IPC protected patients from liver resection [5]. In 2004, it was demonstrated that IPC functioned through activation of HIF-1α in heart [6], followed by a study in 2007 that a randomized controlled trial (RCT) of 10 minutes ⁎ Corresponding author at: Chinese Cochrane Centre, Chinese Evidence-Based Medicine Centre, West China Hospital, Sichuan University, No. 37 Guoxue Xiang, Chengdu 610041, Sichuan, P.R. China. Tel.: +86 28 85164032; fax: +86 28 85164034. E-mail addresses: [email protected], [email protected] (Y. Li).

IPC in liver grafts increased HIF-1α protein levels [7]. In vitro research using ECV 304 endothelial cells also demonstrated that inhibition of HIF-1α eliminates the protective effect of IPC [8]. In 1992, Semenza et al. first proved that HIF-1 (heterodimers a composed of an oxygen sensitive α-subunit and a constitutively expressed β-subunit) increased the expression of erythropoietin (EPO) by binding to the hypoxic response elements (HRE) in its promoter under hypoxia [9]. The central role of HIF-1 in maintaining oxygen homeostasis may largely involve changes in its target genes production; of which hundreds have been identified in 2008 [10]. In 2011, epidemiology research confirmed that HIF-1 not only regulates reperfusion and oxygen transport, but also promotes survival of organs in ischemic diseases by regulating target genes and energy metabolism [11]. However the biogenesis and activation of HIF-1α regulation with respect to the physiology of the liver have yet to be clarified. It has been reported that in HepG2 cells, transcriptional activity of HIF-1α does not necessarily increased, even though its expression and DNA binding ability were increased under hyperbaric oxygen [12]. Although several studies describing the evidence concerning the function of HIF-1α in liver IRI have been published, they have not been systematically reviewed. The current systematic review represents a comprehensive and critical review of all relevant clinical and basic

http://dx.doi.org/10.1016/j.trre.2015.05.001 0955-470X/© 2015 Elsevier Inc. All rights reserved.

Please cite this article as: Guo Y, et al, Systematic review with meta-analysis: HIF-1α attenuates liver ischemia–reperfusion injury, Transplant Rev (2015), http://dx.doi.org/10.1016/j.trre.2015.05.001

2

Y. Guo et al. / Transplantation Reviews xxx (2015) xxx–xxx

Table 1 Methodological quality criteria for studies. Grade Clinical trial A B C

Basic studies in vitro

Animal in vivo studies

RCT With comparable baseline ≤6 Controlled study Baseline unknown ≤3 and b6 Non-controlled study No comparable baseline b3

publications. An in-depth analysis of phenomenon including possible mechanisms is presented along with remaining questions requiring further research and study. 2. Methods 2.1. Publication search and inclusion criteria Medline (PubMed), Embase (Ovid) and the Cochrane Library (all from inception to December 2014) were searched for identification of relevant studies, using strategy combining the following MeSH headings or text words: “liver”, “hepatocytes”, and “ischemia”, “anoxia”, “reperfusion injury”, and “hypoxia-inducible factor 1, alpha subunit”. Studies, whether clinical or basic researches, focusing on the impact of HIF-1α on IRI of liver transplantation were included. Review articles and abstracts were excluded. 2.2. Quality assessment We rated quality criteria for animal in vivo studies as follows [13]: (1) peer-reviewed publication; (2) control of temperature; (3) randomization to treatment or control; (4) blinded assessment of

outcome; (5) animal model (old animals, fatty liver, alcoholic hepatitis or HIF-1α hepatocyte-specific knockout); (6) statement of compliance with animal welfare requirements; and (7) statement of possible conflict of interest. If a study was conducted using inbred animal models, we considered it equivalent to a random allocation in the absence of individual heterogeneity. Since there is no accepted universal evaluation standard for both clinical and basic studies, so we set criteria referring to quality assessment criteria of Cochrane Reviewer's Handbook (version 5.1.0) and CAMARADES [13]. Study quality was stratified into three ranks for (1) clinical trial or (2) basic studies in vitro or (3) animal in vivo studies as follows: A, (1) RCT or (2) with comparable baseline or (3) scores ≥6; B, (1) controlled study or (2) baseline unknown or (3) 6 N scores ≥3; C, (1) non-controlled study or (2) no comparable baseline or (3) scores b 3. Y.G. and L.F. assessed literature independently according to the selection criteria (Table 1). Discrepancies were resolved by Y.L.

2.3. Data extraction Y.G. and L.F. extracted data independently and recorded the following information according to PICOS principle: participants (included subjects, i.e., patients, cell, mouse, rat, etc.), interventions (ischemia/hypoxia, refusion/reoxygenation), comparisons (normoxic), outcomes (O1: liver function (tests include: ALT, apoptosis and angiogenesis) after ischemia/ refusion (IR) or hypoxia/reoxygenation (HR) or anoxia/reoxygenation (AR); O2: HIF-1α mRNA expression; O3: HIF-1α protein expression; O4: HIF-1α/β dimerization; O5: HIF-1 DNA-binding activated; O6: HIF-1 regulated gene expression), and study design (observational or interventional study, an observational study: under ischemia/hypoxia,

Fig. 1. The literature screening process. Note: Data for 2014 are until December.

Please cite this article as: Guo Y, et al, Systematic review with meta-analysis: HIF-1α attenuates liver ischemia–reperfusion injury, Transplant Rev (2015), http://dx.doi.org/10.1016/j.trre.2015.05.001

Y. Guo et al. / Transplantation Reviews xxx (2015) xxx–xxx

reperfusion/reoxygenation, the expression of HIF-1α in liver; an interventional study: under different experiments condition, the effect of stabilization/unstabilization of HIF-1α in liver IRI). In addition to these parameters, we also summarized the main findings of each study and grade. Important unpublished data were obtained by contacting corresponding authors whenever possible. Discrepancies were resolved by a third reviewer (Y.L.). 2.4. Data analysis The ALT after liver IR/HR/AR was selected as the endpoint outcome. Data were analyzed by RevMan 5.3.5 software (Cochrane Collaboration). Continuous data were expressed as standard mean difference (SMD) with 95% confidence intervals (CIs). The I2 test was conducted to evaluate heterogeneity between studies. Heterogeneity was considered high when I 2 N 50% and very high when I 2 N 75%. Meta-analysis with random-effects models was applied to pool SMD across studies if heterogeneity was present (P value b 0.10 or I 2 ≥ 25%). In other cases, fixedeffects models were used. The statistical significance of SMD was analyzed by the Z test. Significance level was set at p b 0.05. Most of the literature of the remaining could not be performed meta-analysis, so we carried out a narrative synthesis about oxygen concentration, various experimental models and exposure duration of liver IR. 3. Literature review 3.1. Literature search and selection Two clinical and 38 basic researches (in vivo and in vitro) were finally included (Fig. 1). The relatively small number of studies suggests that research on this theme is still in an early stage. Basic research was initiated in 1996 [14], which is 7 years earlier than clinical research (Fig. 2). To our knowledge, there has not been a systematic review of the literature using similar criteria. Five [15–19] of these studies were suitable for meta-analysis. 3.2. Characteristics of included studies Twenty-six of included 40 articles are observational studies [14,19–43], while the remaining are interventional studies [7,8,15–18,43–51]. Nine in vitro cell culture studies were reported using various cell lines. Oxygen concentrations for in vitro hypoxia were set as 1% in 7 studies [14,21–24,40,41], 0% in one study [8], and 1% or 0.1% in one study [25]. In animal hypoxia models, oxygen concentration was varied from 0% to 21% (asphyxia, hypoxia, and normoxia) with time spans from acute hypoxia to tolerant hypoxia. In 21 IR animal models, 11 adopted 70% liver (left and middle lobe) IR (I: 1–6 h, R:

3

Table 2 Methodological quality of included animal in vivo studies. Study

(1)

(2) (3) (4) (5) (6) (7) Score

Grade

The expression of HIF-1α in liver (hepatocyte) under ischemia (hypoxia) Yeo, E. J. [26] + − − − − + − 2 C Minchenko, O. [27] + + + − − + − 4 B Frenkel-Denkberg, + − + − + + − 4 B G. [28] Zhang, B. L. [29] + − + − − − − 2 C Bianciardi, P. [30] + + + + − + − 5 B Abdulmalek, K.[31] + + + − − + − 4 B Stroka, D. M. [32] + − + − − + − 3 B Baze, M. M. [33] + + + − − + − 4 B Kai, S. [34] + + + − − + + 5 B Abe, Y. [35] + + + + − + − 5 B Lau, T. Y. [42] + − − − − + − 2 C 3.64 ± B(8B3C) 0.36 The expression of HIF-1α in liver (hepatocyte) under reperfusion (reoxygenation) Tacchini, L. [36] + + + − − + − 4 B Cursio, R . [37] + + + + − + − 5 B + − − − + + − 3 B Li, J. [38] Tacchini, L. [39] + − − − − + − 2 C Namas, R. A. [23] + − + − − + − 3 B Lehwald, N. [24] + − + − − + + 4 B + + + − + − 5 B Zimmerman, M. A. + [43] Ke, B. [19] + + + + − + + 6 A 4 ± 0.46 B(1A6B1C) The effect of stabilization/unstabilization of HIF-1α in liver IRI under different experimental conditions Schmeding, M. [18] + + + + + + − 6 A Zaouali, M. A. [44] + + + − + + − 5 B Schneider, M. [45] + + + + − + + 6 A Zhong, Z. [46] + + + + − + − 5 B Shi, J. [47] + + + − − + − 4 B Rong, L. [17] + − + + − + + 5 B Guo, Y. [15] + + + − − + + 5 B Knudsen, A.R. [48] + + + − − + + 5 B B Ren, P. [49] + + + + − + − 5 Guo, J. Y. [16] + + + − − + + 5 B Ben Mosbah, I. [50] + + + + − + + 6 A Yang, Y.Y. [51] + − − − + + − 3 B 5 ± 0.25 B(3A9B) Total 31 20 26 11 5 30 9 “−” Articles did not report relevant information. “+” Articles reported information. (1) Peer-reviewed publication; (2) control of temperature; (3) randomization to treatment or control; (4) blinded assessment of outcome; (5) animal model (old animals, fatty liver, alcoholic hepatitis or HIF-1α hepatocyte-specific knockout); (6) statement of compliance with animal welfare requirements; and (7) statement of possible conflict of interest.

0.5–12 h) [15,16,23,24,36,37,39,43,46,47,51], 4 liver transplantation induced IR (I: 6–24 h, R: 2 h–14 d) [18,19,44,50], 3 whole liver IR (I: 2–30 min, R: 4–30 min) [35,38,48], 1 liver artery IR (I: 48 h, R: 8 h)

Fig. 2. Year of publication of included sources (2 literatures have both animal and in vitro part).

Please cite this article as: Guo Y, et al, Systematic review with meta-analysis: HIF-1α attenuates liver ischemia–reperfusion injury, Transplant Rev (2015), http://dx.doi.org/10.1016/j.trre.2015.05.001

4

Y. Guo et al. / Transplantation Reviews xxx (2015) xxx–xxx

Table 3 Characteristics of included studies. Study

Object of study (N)

Group of study

Outcome of study

Type of study

Conclusion of study

G

N

I

R

O1

O2

O3

O4

O5

O6

O

I

H (60) H (8)

+ +

— —

+ +

+ —

— —

+ —

— —

— +

— +

— +

+ —

IPC↑⁎ biochemical markers of liver cell function, and↑⁎ HIF-1α concentrations HO-1 contributed to the early protective effects of IPC and HIF-1 is a hypoxia responsive transcription factor with a binding sequence in the HO-1 promoter

A B

Basic studies in vitro Shi, L.B. [8]

H

+

—

+

+

+

+

—

—

—

—

+

A

Makita, T. [21]

H/M

+

+

—

—

—

+

+

+

+

+

—

Forsythe, J.A. [14]

H

—

+

—

—

+

+

+

+

+

+

—

Wang, Y. [22]

R

+

+

—

—

+

+

—

—

—

+

—

Khan, Z. [40]

R

+

+

+

—

+

+

—

—

+

+

—

Namas, R. A.[23]

M

+

+

—

+

+

+

—

—

+

+

—

IPC can ↑⁎ cell viability, the HIF-1α mRNA and protein expression, meanwhile,↓⁎ LDH and ICAM-1 in EVC-304, but these protection reversed by HIF-1α inhibitor Epo is essential factor in the fetal liver and HIF-1 response element in the Epo enhancer, whereas the Epo promoter is active in Hep3B (cancer cells) but is markedly less in E12 hepatocytes (primary cells) Demonstrated for the first time the involvement of HIF-1α in the activation of VEGF transcription in hypoxic Hep3B cells Hypoxia-induced expression of HIF-1α was slowly decreased after reaching its peak level at 3 h in T6-HSCs. A similar expression pattern was also observed in hypoxic macrophage In hepatocytes, hypoxia targeted HIF-1α to the peroxisome, rather than the nucleus, and the peroxisomal pool exists in normoxia, but increases with hypoxia-reoxygenation Mouse primary hepatocytes were exposed to 1% O2 for 6 h, and the mRNA and protein had no significant change, but protein of BNIP3↑⁎

Lehwald, N. [24]

M

+

+

+

+

—

—

—

+

—

+

—

Laemmle, A. [25]

H

+

+

—

—

+

+

—

—

+

+

—

Sun, G. [41]

H

+

+

—

—

+

+

—

—

+

+

—

Animal in vivo studies The expression of HIF-1α in liver (hepatocyte) under ischemia (hypoxia) Yeo, E.J. [26] M (51) + + — — — — — — +

+

—

Clinical studies Amador, A. [7] Patel, A. [20]

Minchenko, O. [27]

M (3)

+

+

—

—

—

—

—

—

+

+

—

Frenkel-Denkberg, G. [28]

M (6)

—

+

—

—

—

+

—

+

—

+

—

Zhang, B.L. [29] Bianciardi, P. [30]

R (54) R (14)

— +

+ +

— —

+ +

— —

+ +

— —

— —

— —

+ +

— —

Abdulmalek, K. [31]

R (18)

+

+

—

—

+

—

—

—

+

+

—

Stroka, D.M. [32]

M (79)

—

+

+

—

—

+

—

—

+

+

—

Baze, M.M. [33]

M (36)

—

+

—

+

+

—

—

—

—

+

—

Kai, S. [34]

M (12)

+

+

—

—

+

—

—

—

+

+

—

Abe, Y. [35]

R (36)

+

+

—

+

+

+

—

—

+

+

—

Lau, T.Y. [42]

R (48)

+

+

—

+

—

—

—

+

+

+

—

The expression of HIF-1α in liver (hepatocyte) under reperfusion (reoxygenation) Tacchini, L. [36] R (24) + + + + — — — + + + — Cursio, R. [37]

R (20)

+

—

+

+

—

+

—

—

—

+

—

Li, J. [38]

R (41)

+

—

+

+

+

+

—

—

+

+

—

Tacchini, L. [39]

R (18)

+

—

+

—

—

—

—

+

+

+

—

β-Catenin binds to HIF-1α under hypoxia conditions to support hepatocyte survival HIF-1α protein accumulation and activation of target genes involved must have the SIRT1 under hypoxic conditions Overexpression of miR-494 upregulated HIF-1α against apoptosis by activating PI3K/Akt pathway under both normoxia and hypoxia

Exposed to mild whole body hypoxia (8% or 12% O2 for 1–8 h), HIF-1α and HIF-2α were both induced in all tissues examined Cells shift primarily to a glycolytic mode for generation of energy when oxygen supply is limited, and PFKFB (key regulators of glycolytic) was induced by HIF-1, but the intensity of the hypoxic induction appears to vary in an organ-, tissue- or cell-specific manner. The liver has higher basal levels of PFKFB-1 and much lower PFKFB-2 and PFKFB-3 mRNA Senescent organisms respond poorly to hypoxic stress not due to less HIF protein but rather due to loss of the ability of HIF-1 to bind to the HRE The ability of HIF-1 to bind to the HRE under hypoxic is exponentially changing Two weeks 10% O2, arterial blood pO2 (mmHg)↓⁎(N: 58 ± 2.8 vs H: 35.3 ± 1.1), hepatic tissue did not exhibit neither HIF-1α accumulation nor apoptosis HIF-1α mRNA was abundant in normoxia, fraction of inspired O2 9–10% for 12 or 24 h only in the kidney and diaphragm, HIF-1α mRNA↑⁎, Tie-2 mRNA and Pro↑⁎ Hypoxia leads to a timely and spatially distinct pattern of HIF-1 protein expression and it may have an important role in tissue homeostasis. In liver, HIF-1α protein reaches maximal levels after 1 h and gradually decreases to baseline levels after 4 h of continuous hypoxia, but HIF-1α mRNA levels had no significant changes Hypoxic conditions were effective in eliciting significant physiological responses. However, the mRNA expression in the liver after 32 days of hypoxia was different from expression patterns reported in studies of acute hypoxia studies, these findings indicate HIF-1 is an important transcription factor in regulating the initial responses to acute hypoxia, it is not as important in the maintenance of acclimatization to chronic hypoxia stress in the liver First demonstration that H2S inhibits HIF-1 activation and its downstream gene expression dependent on mitochondrial and VHL under hypoxic conditions Through HIF-1 target genes, liver epithelial cell proliferation under hypoxic conditions and protects against liver IRI The activity of HIF-1α is progressively increased from day 7 to day 28 in the chronic hypoxia (10% O2) Liver ischemia activates the HIF-1 DNA binding capacity and its target gene mRNA levels. Upon reperfusion, the binding capacity decrease occurred rapidly A significant increase ↑⁎ in liver HIF-1α protein levels after reperfusion 1, 3 h, and apoptosis ↑⁎ after reperfusion 3, 6 h. Author presumed HIF-1α may trigger apoptosis through upregulation of VEGF expression in the ischemic liver Results show in vivo that the pO2 levels in the liver are not the reason of IRI, but during recovery from hypoxia, reoxygenation increases the gene expression of proline hydroxylase, which stops the hypoxic response by destabilizing HIF-1α which will lead to lower expression Because of the impairment of the liver RNA synthesizing machinery during ischemia, HIF-1α mediated activation of TfR gene transcription and IRP-mediated increase of TfR mRNA stability ensure a steady induction of TfR, and hence higher

A

A A

A

A A A A

B C B

B C B B B

B

B B C B B B

B

C

Please cite this article as: Guo Y, et al, Systematic review with meta-analysis: HIF-1α attenuates liver ischemia–reperfusion injury, Transplant Rev (2015), http://dx.doi.org/10.1016/j.trre.2015.05.001

Y. Guo et al. / Transplantation Reviews xxx (2015) xxx–xxx

5

Table 3 (continued) Study

Object of study (N)

Group of study N

I

Outcome of study R

O1

O2

O3

O4

Type of study O5

O6

O

Conclusion of study

G

I

iron uptake in reperfused rat liver. TfR-mediated entry of the metal into liver cells may represent a source of catalytically active iron, which may increase continuously of liver reperfusion damage Namas, R.A. [23] M (40) + + + + — — — — + + — In addition to hypoxic signals, the HIF-1 pathway is also activated by a wide variety of oxygen-independent signals, including growth factors and cell density Lehwald, N. [24] M (20) + + + — — + — — + + — HIF-1α can compete with TCF for binding β-catenin in response to hypoxia. Zimmerman, M. A. [43] M (12) + — + + — + — — + + — ENT1 and Adora2b are transcriptionally regulated by HIF-1α during liver IRI Ke, B. [19] M (50) + — + + + + — — + + — Hepatocyte Keap1 deficiency facilitated activated Trx1 which promoted PI3K/Akt, crucial for HIF-1α activity The effect of stabilization/unstabilization of HIF-1α in liver IRI under different experimental conditions Schmeding, M. [18] R (210) — — + + + — — — — — + IRI ↓⁎, graft survival ↑⁎, the levels of rhuEpo, HIF-1α mRNA expression ↑ in time points (2, 4.5, 24, 48 h, and 7 days postoperatively) among Epo-treated donor fatty liver. Akt activation induces HIF-gene transcription Zaouali, M.A. [44] R (208) + — + + — + — — + — + Fatty liver preserved in IGL-1 by the addition of TMZ (which induces NO) for 24 h, 4 °C, and after reperfusion 2 h, HIF-1α ↑⁎, function parameters of liver injury ↓⁎ than UW, IGL-1 solutions, but they were abolished by the addition of L-NAME (which inhibits NO), which demonstrated vital function of HIF-1α/NO system in fatty liver preservation Schneider, M. [45] R (19) + — + + — + — — + — + PDK1 (a direct target for HIF-1α)↑⁎, anaerobic glucose catabolism↑⁎, oxygen consumption↓⁎, ROS↓⁎, hypoxic hepatocyte damage↓⁎ in PHD−/− livers after ischemia, short-term inhibition of PHD1 just reduces liver apoptosis⁎ Zhong, Z. [46] M (12) — — + + — — — + + — + PHD inhibitors can ↓⁎IRI, and ↑⁎HIF-1α and HO-1, prevents onset of the MPT in liver, but this protection reversed by HO-1 inhibitor Shi, J. [47] M (90) + — + + — + — — — — + ↑Receptor activator for NF-kappa B can protect against hepatic IRI at least in part via the inhibition of the proinflammatory NF-kappa B pathway as well as proapoptotic JNK and HIF-1α pathway activation Rong, L. [17] R (66) + — + + — + — — — — + High-altitude hypoxia may result in liver injury and the key factor of hepatic steatosis. The experimental salidroside group may upregulate HIF-1α protein levels⁎ and inhibit liver cell apoptosis and improve liver function parameters Guo, Y. [15] M (42) + — + + — — — + + — + Rb1 can ↑⁎NO, iNOS, SOD, p-Akt, HIF-1α, meanwhile, ↓⁎ROS, TNF-α, ICAM-1. But they were abolished by the addition of L-NAME 5 min before reperfusion. Thus, the protection conferred by Rb1 appears to be mediated by ROS-NO-HIF pathway Knudsen, A.R. [48] R (28) + — + + + — — — + — + Ischemia (30 min) and reperfusion (30 min) might be too short, and might explain why IPO, IPC, IPO + IPC were not protective effects as judged by liver parameters Ren, P. [49] R (56) + — + + + + — + + — + HBO-PC and 70% partial hepatectomy, respectively, can ↑⁎HIF-1α, ↑⁎ VEGF, ↑⁎ the protein and the DNA-binding activity of nuclear HIF-1α subunit, while the mRNA level kept unchanged. However, HBO-PC and 70% partial hepatectomy together failed to produce additional effect Guo, J. Y. [16] M (42) + — + + + — — — + — + Postconditioning (10 s × 3 cycles) protected the liver IRI. The protection was abrogated in the presence of the NO inhibitor L-NAME Ben Mosbah, I. [50] R (30) + + + + — + — — + — + MnDPDP pretreatment improves liver graft tolerance to IRI by activation of Nfr2 and HIF-1α pathways Yang, Y. Y. [51] R (36) + — + + + + — — — — + Activation of the Wnt3–HIF axis is involved in the protection of losartan on fatty liver graft with IRI

B B B A B A

B

A

B B

B

B

B

B

B A B

“—” Articles did not report relevant information. “+” Articles reported information. Formation and activation of the biological regulation mechanism of HIF-1α in the liver have not completely clear, our study has focused on the following six outcomes of included studies, O1: the study outcome 1, liver function indices; O2: the levels of the HIF-1α mRNA; O3: the levels of the HIF-1α protein; O4: the levels of the HIF-1α dimerization; O5: the levels of the activated DNA binding of HIF-1; O6: the levels of the transcription of hypoxia-responsive genes. N: normoxic groups, I: ischemia group (hypoxia group), R: refusion group (reoxygenation group), IRI: ischemia reperfusion injury, O: an observational study, I: an intervention study, G: grade of study. IPC: ischemic preconditioning, IPO: ischemic postconditioning, HRE: hypoxic response element, H: human, M: mouse, R: rat, ↑: up-regulation, ↓: down-regulation, ECV-304: sinusoidal endothelial cell lines, Tie: tyrosine kinases that contain immunoglobulin-like loops and epidermal growth factor-similar domains, MPT: mitochondrial permeability transition, HBO-PC: hyperbaric oxygen preconditioning, T6-HSCs: rat hepatic stellate cell line, PHD: prolyl hydroxylase domain, SIRT: Sirtuin. ⁎p b 0.05 compared to control group.

[17], 1 liver left lobe IR (I: 1.5 h, R: 8 h) [45], and 1 resection of the 70% liver [49].

B [15–17,23,24,27,28,30–38,43,44,46–49,51], 4 (10%) level C [26,29,39,42]. Generally, the quality of included studies was B in these criteria.

3.3. Quality assessment of included studies 3.4. Outcomes The 31 included animal studies score ranging from 2 to 6 (see Table 2), and contained 4 studies ranked A [18,19,45,50], 23 ranked B [15–17,23,24,27,28,30–38,43,44,46–49,51], and 4 ranked C [26,29,39,42]. In subject classification, the study of intervention gets the highest marks, owing to better control of temperature and blinded assessment of outcome. The marks of the rest of the groups are equal. Of all 40 included studies, two articles [23,24] have both animal and in vitro parts resulting in total of 42 studies (see Table 3). Of these, 14 (33%) studies rank level A [7,8,14,18,19,21–25,40,41,45,50], 24 (57%) level

Main findings of each study are summarized in Table 3. Function parameters detected in the included literatures are ordered from most to least as follows: (1) metabolism and inflammatory response(ALT, lactic acid dehydrogenase), (2) apoptosis and dropsy, (3) regeneration (weight of liver and number of regenerative cells). Of all included studies, 29 monitored HIF-1 target gene expression, 26 measured protein level of HIF-1α, 18 followed mRNA level of HIF-1α, 11 detected DNA combining ability of HIF-1, and 2 monitored dimerization of HIF-1.

Please cite this article as: Guo Y, et al, Systematic review with meta-analysis: HIF-1α attenuates liver ischemia–reperfusion injury, Transplant Rev (2015), http://dx.doi.org/10.1016/j.trre.2015.05.001

6

Y. Guo et al. / Transplantation Reviews xxx (2015) xxx–xxx

Fig. 3. Meta-analysis for stabilized HIF-1α on ALT (IU/L) (random-effects models).

In the included studies, the HIF-1 downstream factors detected from most to least are: angiectasis to increase local oxygen availability, facilitating oxygen transport, improving glucolysis, and promoting cell proliferation.

4.1. Stabilization of HIF-1α protects liver from IRI under different experimental conditions

3.6. Publication bias

In sinusoidal endothelial cell lines ECV-304 of IPC diminishing IRI through HIF-1 [8] shows consistency with clinical trial [7]. Stabilizing HIF-1α by inhibition of prolyl hydroxylase domain (PHD) [45,46], drug treatment [15,17], hepatectomy and hyperbaric oxygen preconditioning [49] attenuates liver IRI (ALT significantly reduced) in animal models. The angiogenesis-related genes such as VEGF, HO-1, EPO and iNOS, are among the most studied HIF-1 target genes, suggesting that revascularization was the focus of many studies, and revascularization plays a key role in vascular endothelium IRI. Besides, sinusoidal endothelial cells and biliary epithelial cells are most vulnerable to liver IRI, while the integrity and dysfunction of endothelia cells are important to grafts' immunogenicity [53]. The effect of stabilized HIF-1α could also promote graft regeneration capacity [49] leading to better initial function and survival rate. But the relationship between the HIF-1α stabilization time course and IR time length is still unclear in the liver. More evidence is needed to determine the best time to stabilize HIF-1α.

Because the number of included studies was b 10, we did not assess the publication bias [52].

4.2. Ischemia-induced HIF-1α shows weak protection during IRI under physiological conditions

3.5. Stabilized HIF-1α decreased ALT after liver IR Almost all the included animal studies mentioned ALT. However, the detailed number and/or SMD were not reported in all studies. Additionally, various experimental animal models were employed in the different studies. As a result, only the peak levels of ALT are reported. The data [15–19] of I 2 = 60%, and we used a random-effects models, and there was statistically significant difference between the 2 groups. As shown in Fig. 3, 5 studies (n = 66) showed that stabilized HIF-1α treatment was associated with lower ALT (SMD = − 1.58; 95% CI = − 2.56, −0.52) when compared with unstabilized HIF-1α group.

4. Discussion We hereby expect to provide more information about endogenous HIF-1α related to IRI in the liver transplant, with both clinical and experimental studies to discuss the following questions.

Ischemia by itself can trigger cell killing, and reperfusion accelerates the process. Liver transplantation goes through cold ischemia (graft stored at 0 °C–4 °C), warm ischemia (in period of anastomoses, graft up toward body temperature 0.5 °C/min), and reperfusion (37 °C). Injury occurs from non-parenchymal cells (20–40% of liver) to parenchymal cells (60–80%). The essence of injury caused by reperfusion is death of

Fig. 4. HIF-1α responses to different stress.

Please cite this article as: Guo Y, et al, Systematic review with meta-analysis: HIF-1α attenuates liver ischemia–reperfusion injury, Transplant Rev (2015), http://dx.doi.org/10.1016/j.trre.2015.05.001

Y. Guo et al. / Transplantation Reviews xxx (2015) xxx–xxx

endothelial cells and activation of Kupffer cells. HIF-1 plays critical roles in revascularization by regulating the expression of its target genes, which may further affect the immunogenicity of the graft [19]. It is well known that HIF-1α combined with HRE to regulate the transcription of downstream factors after ischemia or hypoxia. However, Tacchini et al. confirmed that RNA synthesis device in the liver is damaged during ischemia, such that the effective mRNA synthesis in the downstream regulated gene can only occur in the reconstruction of blood supply during the reperfusion period [39]. Moreover, Li et al. demonstrated that HIF-1α translocated to the peroxisome, which sharply increased after hypoxia/reperfusion, rather than nucleus under hypoxia in rat primary hepatic cell line [38]. Finally the reoxygenation process is also the process of rapid degradation of HIF-1α, thus it shows weak protection against IRI under physiological condition.

4.3. The possible mechanisms of differently HIF-1α responses to stress As shown in Fig. 4, the responses of HIF-1α under the normoxic, acute and chronic hypoxic stimulus were very different. In particular, HIF-1α protein expression from acute oxygen deprivation to hypoxiatolerance was significantly different. Baze et al. proposed that most mammals restore oxygen homeostasis after hypoxia tolerance, thus unnecessarily activating HIF-1 pathway [33]. There are negative-feedback loops in the HIF-1α pathway: in normoxia, the HIF-1α proteins are rapidly degraded via the ubiquitin protein family [54]; at the very beginning of hypoxia sirtuin 1 (SIRT1) inhibits activation of HIF-1α [25,55]; in the early period of hypoxia, PHD2 (HIF target gene) inhibits HIF-1α protein stability [54]; when hypoxia lasts for longer periods of time, miR-155 (HIF target gene) inhibited HIF-1α at its transcription level [56]. Both PHD and miRNAs appear to be regulated by hypoxia in a cell type-/tissue-specific manner. Overall, HIF-1α plays an important role in acute hypoxia, and a deficiency of HIF-1α dramatically inhibits long-term physiological responses to hypoxia [57]. However, in chronic hypoxia HIF-1α becomes a lethal factor, because high level of HIF-1 causes cell death [58].

4.4. Limitations of our review Herein our observations have significant clinical relevance because ischemia and reperfusion are common features of many liver diseases. In light of the clinical and basic studies presented here, stabilized HIF-1α can significantly reduce liver IRI or promote graft survival rate. Nevertheless, the liver is composed of hepatocytes, many immune cells and blood cells. It is difficult to identify from which cell types is the source of the gene expression. Moreover, the anti-hypoxia activity of different cells is exerted via different mechanisms. Responses to ischemia (hypoxia) include changes in a wide variety of signals that correlate with HIF-1α, such as nuclear factor kappa B (NF-κB) [47,42], Wnt–β-catenin signaling [24,51], etc. However, studies are still in the exploratory stage and clinical trials are just beginning. Furthermore, the quality and sample size of included studies are insufficient.

4.5. Conclusions In summary, our systematic review indicated that stabilized HIF-1α treatment can protect the liver in IRI, and through a variety of means to stabilize HIF-1α which only plays a role in the acute stage of liver ischemia or hypoxia. But unfortunately, physiological responses of ischemia-induced HIF-1α show weak protection during reperfusion in liver transplantation. Better understanding the molecular mechanisms underlying the effect of HIF-1α in liver IRI will help us to optimize the clinical prevention or treatment of graft IRI, and to better understand the mechanism of why liver is intolerant to ischemia or hypoxia.

7

Acknowledgments This work was financially supported by National Basic Research Program of China No. 2009CB522401, by the Natural Science Foundation of China (NSFC) No. 81270552 and 81273255, and by 2013 program of Key Laboratory of National Health and Family Planning Commission. The funders had no role in study design, data collection and analysis, decision to publish or preparation of the manuscript. The authors have declared no conflicts of interest. References [1] Bilzer M, Gerbes AL. Preservation injury of the liver: mechanisms and novel therapeutic strategies. J Hepatol 2000;32:508–15. [2] Ohkohchi N. Mechanisms of preservation and ischemic/reperfusion injury in liver transplantation. Transplant Proc 2002;34:2670–3. [3] Jennings RB, Sommers HM, Smyth GA, Flack HA, Linn H. Myocardial necrosis induced by temporary occlusion of a coronary artery in the dog. Arch Pathol 1960;70:68–78. [4] Murry CE, Jennings RB, Reimer KA. Preconditioning with ischemia: a delay of lethal cell injury in ischemic myocardium. Circulation 1986;74:1124–36. [5] Clavien PA, Yadav S, Sindram D, Bentley RC. Protective effects of ischemic preconditioning for liver resection performed under inflow occlusion in humans. Ann Surg 2000;232:155–62. [6] Lee JW, Bae SH, Jeong JW, Kim SH, Kim KW. Hypoxia-inducible factor (HIF-1)alpha: its protein stability and biological functions. Exp Mol Med 2004;36:1–12. [7] Amador A, Grande L, Marti J, et al. Ischemic pre-conditioning in deceased donor liver transplantation: a prospective randomized clinical trial. Am J Transplant 2007;7: 2180–9. [8] Shi LB, Huang JH, Han BS. Hypoxia inducible factor-1alpha mediates protective effects of ischemic preconditioning on ECV-304 endothelial cells. World J Gastroenterol 2007;13:2369–73. [9] Semenza GL, Wang GL. A nuclear factor induced by hypoxia via de novo protein synthesis binds to the human erythropoietin gene enhancer at a site required for transcriptional activation. Mol Cell Biol 1992;12:5447–54. [10] Zhang H, Bosch-Marce M, Shimoda LA, et al. Mitochondrial autophagy is an HIF-1dependent adaptive metabolic response to hypoxia. J Biol Chem 2008;283: 10892–903. [11] Semenza GL. Oxygen sensing, homeostasis, and disease. N Engl J Med 2011;365: 537–47. [12] Salhanick SD, Belikoff B, Orlow D, Holt D, Reenstra W, Buras JA. Hyperbaric oxygen reduces acetaminophen toxicity and increases HIF-1alpha expression. Acad Emerg Med 2006;13:707–14. [13] Macleod MR, O'Collins T, Howells DW, Donnan GA. Pooling of animal experimental data reveals influence of study design and publication bias. Stroke 2004;35:1203–8. [14] Forsythe JA, Jiang BH, Iyer NV, et al. Activation of vascular endothelial growth factor gene transcription by hypoxia-inducible factor 1. Mol Cell Biol 1996;16:4604–13. [15] Guo Y, Yang T, Lu J, et al. Rb1 postconditioning attenuates liver warm ischemiareperfusion injury through ROS-NO-HIF pathway. Life Sci 2011;88:598–605. [16] Guo JY, Yang T, Sun XG, et al. Ischemic postconditioning attenuates liver warm ischemia-reperfusion injury through Akt-eNOS-NO-HIF pathway. J Biomed Sci 2011;18:79–91. [17] Rong L, Dai LL, ZENG WZ, Jiang MD, Wu XL, Zhang H. Protective effect of salidroside on liver injury of rats under altitude hypoxia simulation. J Clin Rehabil Tissue Eng Res 2010;14:5813–7. [18] Schmeding M, Rademacher S, Boas-Knoop S, et al. rHuEPo reduces ischemiareperfusion injury and improves survival after transplantation of fatty livers in rats. Transplantation 2010;89:161–8. [19] Ke B, Shen XD, Zhang Y, et al. KEAP1-NRF2 complex in ischemia-induced hepatocellular damage of mouse liver transplants. J Hepatol 2013;59:1200–7. [20] Patel A, van de Poll MC, Greve JW, et al. Early stress protein gene expression in a human model of ischemic preconditioning. Transplantation 2004;78:1479–87. [21] Makita T, Duncan SA, Sucov HM. Retinoic acid, hypoxia, and GATA factors cooperatively control the onset of fetal liver erythropoietin expression and erythropoietic differentiation. Dev Biol 2005;280:59–72. [22] Wang YQ, Luk JM, Ikeda K, et al. Regulatory role of vHL/HIF-1a in hypoxia-induced VEGF production in hepatic stellate cells. Biochem Biophys Res Commun 2004; 317:358–62. [23] Namas RA, Metukuri MR, Dhupar R, et al. Hypoxia-induced overexpression of BNIP3 is not dependent on hypoxia-inducible factor 1alpha in mouse hepatocytes. Shock 2011;36:196–202. [24] Lehwald N, Tao GZ, Jang KY, et al. Wnt-beta-catenin signaling protects against hepatic ischemia and reperfusion injury in mice. Gastroenterology 2011;141: 707–718.e1-5. [25] Laemmle A, Lechleiter A, Roh V, et al. Inhibition of SIRT1 impairs the accumulation and transcriptional activity of HIF-1alpha protein under hypoxic conditions. PLoS One 2012;7:e33433. http://dx.doi.org/10.1371/journal.pone.0033433. [26] Yeo EJ, Cho YS, Kim MS, Park JW. Contribution of HIF-1alpha or HIF-2alpha to erythropoietin expression: in vivo evidence based on chromatin immunoprecipitation. Ann Hematol 2008;87:11–7. [27] Minchenko O, Opentanova I, Caro J. Hypoxic regulation of the 6-phosphofructo-2kinase/fructose-2,6-bisphosphatase gene family (PFKFB-1-4) expression in vivo. FEBS Lett 2003;554:264–70.

Please cite this article as: Guo Y, et al, Systematic review with meta-analysis: HIF-1α attenuates liver ischemia–reperfusion injury, Transplant Rev (2015), http://dx.doi.org/10.1016/j.trre.2015.05.001

8

Y. Guo et al. / Transplantation Reviews xxx (2015) xxx–xxx

[28] Frenkel-Denkberg G, Gershon D, Levy AP. The function of hypoxia-inducible factor 1 (HIF-1) is impaired in senescent mice. FEBS Lett 1999;462:341–4. [29] Zhang BL, Yang ZH, Ran P, et al. Induction of hypoxia-inducible factor-1alpha in two kinds of rats asphyxiation death models. Fa Yi Xue Za Zhi 2007;23:4–7. [30] Bianciardi P, Fantacci M, Caretti A, et al. Chronic in vivo hypoxia in various organs: hypoxia-inducible factor-1alpha and apoptosis. Biochem Biophys Res Commun 2006;342:875–80. [31] Abdulmalek K, Ashur F, Ezer N, Ye F, Magder S, Hussain SN. Differential expression of Tie-2 receptors and angiopoietins in response to in vivo hypoxia in rats. Am J Physiol Lung Cell Mol Physiol 2001;281:L582–90. [32] Stroka DM, Burkhardt T, Desbaillets I, et al. HIF-1 is expressed in normoxic tissue and displays an organ-specific regulation under systemic hypoxia. FASEB J 2001;15:2445–53. [33] Baze MM, Schlauch K, Hayes JP. Gene expression of the liver in response to chronic hypoxia. Physiol Genomics 2010;41:275–88. [34] Kai S, Tanaka T, Daijo H, et al. Hydrogen sulfide inhibits hypoxia- but not anoxiainduced hypoxia-inducible factor 1 activation in a von hippel-lindau- and mitochondria-dependent manner. Antioxid Redox Signal 2012;16:203–16. [35] Abe Y, Uchinami H, Kudoh K, et al. Liver epithelial cells proliferate under hypoxia and protect the liver from ischemic injury via expression of HIF-1 alpha target genes. Surgery 2012;152:869–78. [36] Tacchini L, Radice L, Bernelli-Zazzera A. Differential activation of some transcription factors during rat liver ischemia, reperfusion, and heat shock. J Cell Physiol 1999; 180:255–62. [37] Cursio R, Miele C, Filippa N, Van Obberghen E, Gugenheim J. Liver HIF-1 alpha induction precedes apoptosis following normothermic ischemia-reperfusion in rats. Transplant Proc 2008;40:2042–5. [38] Li J, French B, Wu Y, et al. Liver hypoxia and lack of recovery after reperfusion at high blood alcohol levels in the intragastric feeding model of alcohol liver disease. Exp Mol Pathol 2004;77:184–92. [39] Tacchini L, Fusar Poli D, Bernelli-Zazzera A, Cairo G. Transferrin receptor gene expression and transferrin-bound iron uptake are increased during postischemic rat liver reperfusion. Hepatology 2002;36:103–11. [40] Khan Z, Michalopoulos GK, Stolz DB. Peroxisomal localization of hypoxia-inducible factors and hypoxia-inducible factor regulatory hydroxylases in primary rat hepatocytes exposed to hypoxia-reoxygenation. Am J Pathol 2006;169:1251–69. [41] Sun G, Zhou Y, Li H, et al. Over-expression of microRNA-494 up-regulates hypoxiainducible factor-1 alpha expression via PI3K/Akt pathway and protects against hypoxia-induced apoptosis. J Biomed Sci 2013;20:100. http://dx.doi.org/10.1186/ 1423-0127-20-100. [42] Lau TY, Xiao J, Liong EC, et al. Hepatic response to chronic hypoxia in experimental rat model through HIF-1 alpha, activator protein-1 and NF-kappa B. Histol Histopathol 2013;28:463–71. [43] Zimmerman MA, Tak E, Ehrentraut SF, et al. Equilibrative nucleoside transporter (ENT)-1-dependent elevation of extracellular adenosine protects the liver during ischemia and reperfusion. Hepatology 2013;58:1766–78.

[44] Zaouali MA, Ben Mosbah I, Boncompagni E, et al. Hypoxia inducible factor-1alpha accumulation in steatotic liver preservation: role of nitric oxide. World J Gastroenterol 2010;16:3499–509. [45] Schneider M, Van Geyte K, Fraisl P, et al. Loss or silencing of the PHD1 prolyl hydroxylase protects livers of mice against ischemia/reperfusion injury. Gastroenterology 2010; 138:1143–1154.e1-2. [46] Zhong Z, Ramshesh VK, Rehman H, et al. Activation of the oxygen-sensing signal cascade prevents mitochondrial injury after mouse liver ischemia-reperfusion. Am J Physiol Gastrointest Liver Physiol 2008;295:G823–32. [47] Shi J, Shao W, Yang D, et al. Hydrodynamics-based transfection of plasmid encoding receptor activator for nuclear factor kappa B-Fc protects against hepatic ischemia/ reperfusion injury in mice. Liver Transpl 2010;16:611–20. [48] Knudsen AR, Kannerup AS, Gronbaek H, et al. Effects of ischemic pre- and postconditioning on HIF-1alpha, VEGF and TGF-beta expression after warm ischemia and reperfusion in the rat liver. Comp Hepatol 2011;10:3. http://dx.doi.org/10.1186/ 1476-5926-10-3. [49] Ren P, Kang Z, Gu G, et al. Hyperbaric oxygen preconditioning promotes angiogenesis in rat liver after partial hepatectomy. Life Sci 2008;83:236–41. [50] Ben Mosbah I, Mouchel Y, Pajaud J, et al. Pretreatment with mangafodipir improves liver graft tolerance to ischemia/reperfusion injury in rat. PLoS One 2012;7:e50235. http://dx.doi.org/10.1371/journal.pone.0050235. [51] Yang YY, Lee PC, Huang YT, et al. Involvement of the HIF-1alpha and Wnt/beta-catenin pathways in the protective effects of losartan on fatty liver graft with ischaemia/ reperfusion injury. Clin Sci (Lond) 2014;126:163–74. http://dx.doi.org/10.1371/ journal.pone.0050235. [52] Song F, Hooper Y Loke. Publication bias: what is it? How do we measure it? How do we avoid it? Open Access J Clin Trials 2013;71. [53] Yuan X, Theruvath AJ, Ge X, et al. Machine perfusion or cold storage in organ transplantation: indication, mechanisms, and future perspectives. Transpl Int 2010;23:561–70. [54] Hyvarinen J, Hassinen IE, Sormunen R, et al. Hearts of hypoxia-inducible factor prolyl 4-hydroxylase-2 hypomorphic mice show protection against acute ischemiareperfusion injury. J Biol Chem 2010;285:13646–57. [55] Lim JH, Lee YM, Chun YS, Chen J, Kim JE, Park JW. Sirtuin 1 modulates cellular responses to hypoxia by deacetylating hypoxia-inducible factor 1alpha. Mol Cell 2010;38:864–78. [56] Bruning U, Cerone L, Neufeld Z, et al. MicroRNA-155 promotes resolution of hypoxiainducible factor 1alpha activity during prolonged hypoxia. Mol Cell Biol 2011;31: 4087–96. [57] Yu AY, Shimoda LA, Iyer NV, et al. Impaired physiological responses to chronic hypoxia in mice partially deficient for hypoxia-inducible factor 1alpha. J Clin Invest 1999;103:691–6. [58] Kline DD, Peng YJ, Manalo DJ, Semenza GL, Prabhakar NR. Defective carotid body function and impaired ventilatory responses to chronic hypoxia in mice partially deficient for hypoxia-inducible factor 1 alpha. Proc Natl Acad Sci U S A 2002;99: 821–6.

Please cite this article as: Guo Y, et al, Systematic review with meta-analysis: HIF-1α attenuates liver ischemia–reperfusion injury, Transplant Rev (2015), http://dx.doi.org/10.1016/j.trre.2015.05.001