Role of ischaemic preconditioning in liver regeneration following major liver resection and transplantation

Abstract

Liver ischaemic preconditioning (IPC) is known to protect the liver from the detrimental effects of ischaemic-reperfusion injury (IRI), which contributes significantly to the morbidity and mortality following major liver surgery. Recent studies have focused on the role of IPC in liver regeneration, the precise mechanism of which are not completely understood. This review discusses the current understanding of the mechanism of liver regeneration and the role of IPC in this setting. Relevant articles were reviewed from the published literature using the Medline database. The search was performed using the keywords “liver”, “ischaemic reperfusion”, “ischaemic preconditioning”, “regeneration”, “hepatectomy” and “transplantation”. The underlying mechanism of liver regeneration is a complex process involving the interaction of cytokines, growth factors and the metabolic demand of the liver. IPC, through various mediators, promotes liver regeneration by up-regulating growth-promoting factors and suppresses growth-inhibiting factors as well as damaging stresses. The increased understanding of the cellular mechanisms involved in IPC will enable the development of alternative treatment modalities aimed at promoting liver regeneration following major liver resection and transplantation.

Key Words: Liver regeneration, Ischaemic reperfusion, Ischaemic preconditioning, Hepatectomy, Transplantation

INTRODUCTION

Ischaemic-reperfusion injury (IRI) is an inevitable phenomenon that results following major liver surgery, including partial hepatectomy and liver transplantation. As a consequence, parenchymal cell injury and liver dysfunction[1,2] of varied severity leads to significant morbidity and mortality post-surgery[3-5], in particular, in patients with liver cirrhosis and steatosis[6-9]. In addition, IRI significantly impairs liver regeneration following hepatectomy[10,11].

Due to the inevitability of ischaemia and reperfusion in liver surgery, various investigators have attempted to elucidate methods to limit the detrimental effects of IRI and improve liver function and regeneration of the remnant liver[12,13]. These include hypothermic perfusion of the liver[14], intermittent liver inflow occlusion[15,16] and ischaemic preconditioning (IPC)[17]. Liver IPC is an endogenous mechanism consisting of a short period of vascular occlusion followed by reperfusion that renders the liver more tolerant to subsequent prolonged episodes of ischaemia. Besides having protective effects on IRI following major liver resection and transplantation[17-19], it has been suggested that IPC is also beneficial in liver regeneration[20]. This article describes the current understanding of the liver regeneration cascade as published and gives a balanced review on the mechanisms by which IPC influences liver regeneration.

MECHANISMS OF LIVER REGENERATION

Liver regeneration is a complex and multi-factorial process that is mediated by interactions between regenerative cytokines, growth factors and metabolic demand of the liver following surgery and IRI.

The regenerative cytokine network and the priming pathway

During the first few hours following IRI, regenerative cytokines are produced that render the resting hepatocytes responsive to growth factors required for cellular division and proliferation[21,22]. This period is known as the “priming phase”. Various mediators have been implicated as possible triggers of the regenerative cytokine network including gastrointestinal lipopolysaccharide (LPS)[23,24], Toll-like receptor-myeloid differentiation factor 88 (Myd88) signaling pathways[25-27], components from the complement cascade[28], nitric oxide (NO)[29-31] and prostaglandins[32].

Studies have identified tumour necrosis factor-alpha (TNF-α) and interleukin-6 (IL-6) as important regenerative cytokines[21,33,34]. Akerman and co-workers showed that anti-TNF antibodies led to delayed DNA synthesis in the regenerating rat liver and inhibited the increase in IL-6 levels following partial hepatectomy[35]. The initiation of liver growth by TNF-α was shown to be dependent on its binding to TNF-receptor type 1 (TNF-R1)[36]. Mice deficient of TNF-R1 exhibited a delay in liver regeneration and increased mortality following liver resection, which was subsequently reversed by recombinant IL-6 injection[37]. However, IL-6 lacking mice demonstrated impaired liver regeneration despite the presence of TNF-α, suggesting that TNF-R1 signaling results in the release of IL-6[38]. IL-6 deficient mice not only showed impaired ability to regenerate, but also had increased IRI following liver resection[39]. However, Wuestefeld et al[40] reported that mice deficient in IL-6 and its common signal transducer, glycoprotein 130 (gp130) had no defects in DNA replication following partial hepatectomy. The groups of Zimmers and Blindenbacher have suggested that the levels of serum IL-6 present following liver resection in mice are critical in modulating its regenerative effects[41,42]. This may account for the difference in results observed in studies attempting to determine the effect of IL-6 in liver regeneration. It has been suggested that other mediators such as stem cell factor and oncostatin M may play a role in enhancing the effects of IL-6 on hepatocyte regeneration[43-45].

Non-parenchymal liver cells [Kupffer cells and sinusoidal endothelial cells (SECs)] are involved in the priming phase (Figure 1)[46]. Following stimulation from cytokine triggering mechanisms, TNF-α binds to TNF-R1 on non-parenchymal liver cells and stimulates the production of IL-6[47], via the activation of the transcription factor nuclear factor-kappa B (NF-κB)[48-50]. Ping and colleagues demonstrated that the secretion of regenerative cytokines, such as IL-6 by rat liver SECs was mediated by the phosphatidylinositol 3-kinase (PI 3-kinase)/Akt signaling pathway[51], via NF-κB activation[50]. The transcription factor NF-κB is also known to be an important component of pro-survival cellular signaling responses, and hence its activation will not only stimulate IL-6 production but also activate survival genes[52,53].

Figure 1 Current proposed mechanisms of the priming pathway of liver regeneration. IRI: Ischaemic reperfusion injury; LPS: Lipo-polysaccharide; TLR: Toll-like receptor; MyD88: Myeloid differentiation factor 88; NO: Nitric oxide; PG: Prostaglandins; NPCs: Non-parenchymal cells; TNF: Tumour necrosis factor; TNFR1: TNF receptor type I; NF-κB: Nuclear factor-kappaB; IL-6: Interleukin-6; IL-6 RC: IL-6 receptor complex; ERK-1/2: Extracellular regulated kinases 1/2; STAT-3: Signal transducer and activator of transcription-3.

IL-6 acts directly on hepatocytes by binding to the IL-6 receptor complex and induces the translocation of signal transducer and activator of transcription-3 (STAT-3) to the nucleus. This initiates a cascade of events that leads to progression of the cell cycle, culminating in the synthesis of DNA and subsequent cellular mitosis[54,55].

Growth factors and growth-factor signaling systems in liver regeneration

Following this priming phase, cell cycle progression is then dependent on growth factors, such as hepatocyte growth factor (HGF), transforming growth factor-alpha (TGF-α) and epidermal growth factor (EGF)[56,57]. The two main growth-promoting signaling systems involved in liver regeneration are the HGF and its receptor (Met) and the epidermal growth factor receptor (EGFR) and its relatively large family of ligands. The effect of growth factors and their corresponding signaling systems may be dependent on the metabolic state of the hepatocytes and the presence or absence of other effectors[58].

HGF is produced by non-parenchymal cells of the rat liver following liver injury[59-61] and partial hepatectomy[62], and acts on its receptor on hepatocytes. Several authors have demonstrated that HGF is crucial in promoting liver regeneration following partial hepatectomy and transplantation in animal models[63-67]. HGF administration to recipients of reduced-size liver grafts in rats illustrated early regeneration and provided hepatoprotection against rejection-related injuries[64,68,69]. These essential signals are regulated by HGF, via c-met, the gene on the HGF-receptor.

EGF is mainly produced in the salivary glands in rodents and plays an important role in hepatocyte proliferation by binding to the EGFR on hepatocytes[57]. Sialoadenectomy-induced decrease in circulating EGF in mice and rat models resulted in impaired liver regeneration following partial hepatectomy, which was reversed by the administration of EGF[70,71]. In addition, combined administration of EGF and insulin increases the DNA synthesis following liver resection in cirrhotic rats[72]. Results from these studies suggest that EGF has a direct effect on hepatocyte proliferation.

TGF-α is produced in non-parenchymal cells, mainly the Kupffer cells. Mead and Fausto demonstrated that TGF-α may function as a physiological inducer of hepato-cyte DNA synthesis during liver regeneration by an autocrine mechanism by binding to EGFR in both rat and culture models[73]. Besides TGF-α, there are many ligands for EGFR, including EGF, amphiregulin, heparin-binding EGF-like growth factor (HB-EGF) and epiregulin[57]. Although TGF-α expression increases following partial hepatectomy in mice, TGF-α lacking mice do not display diminished liver regeneration[74,75]. This may be due to the fact that ligands such as EGF can also stimulate EGFR and activate common intracellular growth signaling pathways. Studies have shown that other growth factors, such as HB-EGF[50,76,77], amphiregulin[78], insulin and glucagon[79,80] may also play a role in liver regeneration.

Although these two growth-promoting signaling systems are largely independent, some integration may exist. Scheving et al[81] demonstrated that EGFR kinase inhibition by PKI166 (selective, potent inhibitor of EGFR kinase) blocked the mitogenic effects of HGF in cultured rat hepatocytes, suggesting EGFR may regulate HGF-mediated hepatocyte proliferation. Nevertheless, various regenerative pathways are initiated by HGF/c-met and the EGFR signaling mechanisms (Figure 2), which include the activation of mitogen-activated protein kinases (MAPKs), such as extracellular signal-regulated kinases 1 and 2 (ERK-1/2; aka p42/44 MAPKs), c-jun-NH2-terminal kinases 1 and 2 (JNK-1/2; aka p46/p54 SAPK) and p38 MAPKs. Collectively, these MAPKs have been shown to play essential roles in cell growth, transformation differentiation and apoptosis[82-84].

Figure 2 Potentiation of growth factor signaling pathways involved in cell cycle progression in liver regeneration. NPCs: Non-parenchymal cells; HGF: Hepatocyte growth factor; EGF: Epidermal growth factor; TGF-α: Transforming growth factor-α; EGFR: Epidermal growth factor receptor; ERK-1/2: Extracellular regulated kinases 1/2; JNK: c-jun-NH2-terminal kinase; PI 3K/Akt: Phosphatidylinositol 3-kinase/Akt-signal pathway.

ERK-1/2 activation has been shown to correlate with hepatocyte proliferation in animal studies and in vitro models[85-87]. Besides being responsive to growth factor signals, studies have demonstrated that ERK-1/2 activity can also be induced by cytokines such as TNF-α[88]. Hence, the ERK-1/2 may be a signaling pathway that integrates both growth factor and cytokine signaling. Serandour et al[89] demonstrated that a combination of EGF and TNF-α induced hepatocyte proliferation by 30%, compared to EGF alone in a hepatocyte-liver epithelial cell co-culture model. This study suggested that TNF-α mediated extracellular matrix remodeling was required for continued hepatocyte replication and proliferation. Hence, this study re-iterates the importance of the interaction of cytokines and growth factors in the liver regeneration cascade. Although ERK-1/2 has been identified to have a key role in hepatocyte growth, inhibition of ERK-1/2 does not significantly alter proliferation of regenerating rat hepatocytes. However, inhibition of p38 MAPKs results in decreased DNA synthesis, suggesting that p38 MAPKs activation is prerequisite for hepatocyte proliferation[90]. JNK-1 and p38 MAPKs are involved in the regulatory control of the induction of nuclear proteins, such as cyclin D1. The activation of cyclin D1 is one of the earliest steps in the pathway of resting cells to enter the pre-replicative phase of the cell cycle.

Metabolic demand of the liver

The increased metabolic demand imposed on the remnant liver following partial hepatectomy are likely to be inter-connected with the activation of the mechanisms involved in DNA replication to sustain liver function. This is likely to be dependent on energy levels and nutrient availability.

Mitochondria are the predominant source of the high energy phosphates that are essential for energy-dependent processes in cells. Mitochondrial activity has been shown to be correlated with the recovery of liver function and subsequent regeneration[91,92]. Maruyama et al[93] showed the recovery of liver weight following hepatectomy was proportional to the energy (ATP) levels of the remnant liver in rats.

Amino acid deprivation has been shown to inhibit the regeneration process in rat livers following liver resection[94]. Nelsen et al[95] showed that selective amino acid deprivation in culture and protein deprivation in mice impaired hepatocyte cyclin D1 expression and that transfection of cyclin D1 promoted cell cycle progression under these conditions. Hence, amino acids regulate hepatocyte proliferation via cyclin D1.

The absence of bile acids in the intestine has been shown to delay liver regeneration following partial hepatectomy in rats[96]. Following liver surgery or injury, bile flow is stimulated[97] which results in the release of bile from the gallbladder and its return through the entero-hepatic circulation exposes the remnant hepatocytes to an increase in relative bile acid flux. This early phase of bile acid overload and subsequent bile acid signaling is necessary for normal liver regeneration. However, liver-specific functions such as synthesis of clotting factors and albumin and the continuous formation of bile are impaired following partial hepatectomy and results in transient cholestasis[98]. The formation of bile is dependent on the active secretion of bile salts and other biliary constituents into the bile canaliculus by specific bile acid and organic anion transporters[99]. Gerloff et al[100] demonstrated that the expression of two ATP-dependent transporters [bile salt acid pump (Bsep) and multi-organic anion transporter (Mrp2)] was unchanged or slightly increased following partial hepatectomy in rats and this provided a potential mechanism by which the regenerating liver cells maintained or increased bile secretion. The authors in this study also suggested that the down-regulation of certain transporters [sodium-taurocholate cotransporter (Ntcp) and organic anion transporting polypeptides (Oatp1 and Oatp2)] could be a protective mechanism against the potentially hepatotoxic bile salts[100]. The differential regulation of hepatobiliary transporters during the regeneration process are likely to be mediated by cytokines such as TNF-α[101]. Recently, Huang et al[102] showed that bile acid activation of nuclear receptor-dependent signaling pathways regulated the regeneration process by sensing the liver’s functional capacity following partial hepatectomy in mice. When inadequate function causes bile acids to build up, the resultant nuclear receptors activation not only induces negative feedback pathways that protect hepatocytes from bile acid toxicity but also increases the capacity of the liver to manage the overload by promoting liver growth[102].

Results from recent studies have implicated the mammalian target of rapamycin (mTOR) complex as a sensor of nutrient-energy levels and its downstream mediators, such as p70 S6 kinase, are thought to regulate protein translation and cell growth[103,104]. Inhibition of the mTOR complex leads to diminish DNA replication following partial hepatectomy in mice and rat models[105,106]. The activity of p70 S6 kinase has been shown to increase following partial hepatectomy[106] and mice lacking the S6 kinase demonstrated diminished hepatocyte proliferation[107]. Hence, energy is required for energy-dependent signaling pathways and processing the available nutrients for the regeneration process to proceed.

ISCHAEMIC PRECONDITIONING IN LIVER REGENERATION

Since IPC was first described by Murry et al[108], this strategy has been developed more widely and is currently practiced in major liver surgery in several centers. The mechanism of IPC is thought to be divided into two phases; early (classical/acute) pre-conditioning and delayed pre-conditioning. The hepatoprotective effects of early preconditioning occur within minutes after reperfusion and are maintained for 1 to 2 h[109]. This phase is thought to be mediated by pre-existing substances. The “second window of protection”, delayed pre-conditioning, re-appears 24 to 72 h following IPC. The underlying mechanism is thought to rely on the modification of gene expression resulting in protein production as its effectors[110]. Various substances have been implicated as key effectors in liver IPC including adenosine[111-114], protein kinase C[115-117], NO[118-121], heat-shock proteins[122,123], tyrosine kinases[124], MAPKs[117], oxidative stress[125,126] and NF-κB[127,128].

The beneficial effects of IPC on the liver following IRI include decrease in severity of liver necrosis[129], anti-apoptotic effects[130], preservation of liver microcirculati-on[131,132] and improvement in survival rate[119], and more recently, its role in liver regeneration is currently being evaluated. For IPC to influence liver regeneration, its key effectors must be involved in either promoting or up-regulating mediators that are involved in the regeneration cascade or exert an inhibitory effect on growth-inhibitory factors of liver regeneration or vice versa (Figure 3).

Figure 3 The effect of ischaemic preconditioning and its effectors on the signaling pathways of liver regeneration. TK: Tyrosine kinase; PKC: Protein kinase C; NF-κB: Nuclear factor-kappaB; IκBα: Inhibitory binding protein for NF-κB; ERK-1/2: Extracellular regulated kinases 1/2; HSPs: Heat shock proteins; NO: Nitric oxide; IL1-RA: Interleukin-1 receptor antagonist; HGF: Hepatocyte growth factor; ROS: Reactive oxygen species; PI 3K/Akt: Phosphatidylinositol 3-kinase/Akt-signal pathway; JNK: c-jun-NH2-terminal kinase; C-D1: Cyclin D1; IL-1: Interleukin-1; TGF-β: Transforming growth factor-β.

Up-regulation of the regenerative cytokine network

Tumour necrosis factor-alpha and interleukin-6: The initiation of the regenerative response is dependent on the early activation of TNF-α and IL-6 responsive transcription factors[38,54]. Studies evaluating the effect of IPC in stimulating the release of TNF-α and IL-6 have produced conflicting findings. Tsuyama et al[133] reported that IPC prolonged survival, suppressed liver necrosis induced by IRI and inhibited the release of TNF-α at 1 h, and IL-6 at 2 and 5 h of the reperfusion phase in mice. IPC has also been shown to reduce TNF-α production in normothermic and cold ischaemic conditions[125,134,135].

In contrast, Bedirli and others demonstrated that IPC inhibited the production of TNF-α, but not IL-6 during the late (24 and 48 h) phases of reperfusion in rats following partial hepatectomy[136]. This study suggested that IL-6 is an important mediator in IPC-treated rats in promoting hepatocyte proliferation following liver resection[136]. Using a TNF gene-deleted mice model, Teoh et al[137] demonstrated that pre-treatment with low dose IL-6 prior to hepatic ischaemia conferred equivalent hepatoprotection and earlier cell cycle entry as IPC compared to non-IPC-treated mice at 2 h of reperfusion. Taken together, these studies emphasize the importance of IL-6 as a hepatoprotective and pro-proliferative mediator during the early and late phases of reperfusion following IRI. Although IPC attenuates the late onset prolonged release of TNF-α (up to 44 h) that mediates liver IRI in rats[138,139] and mice[140], IPC itself is associated with the early increase (10 min) in liver and serum TNF-α following hepatic ischaemia[140]. Injection of low dose TNF-α 30 min prior to liver ischaemia conferred similar hepatoprotection as IPC[140]. Both IPC and pre-treatment with low dose TNF-α injection were shown to stimulate earlier and more vigorous cell cycle entry following liver IRI compared to naïve mice during the first 24 h of the reperfusion period[140].

The differences reported among studies investigating the influence of IPC on the regenerative cytokine network may be related to the differences in experimental models and protocols employed. In addition, both TNF-α and IL-6 are known to be important components in other cellular signaling responses. For example, high levels of serum TNF-α following liver ischaemia in rats is thought to mediate liver IRI[138,139]. Investigators have suggested that the mechanism of action of IL-6 in promoting liver regeneration appear to be separate from those involved in the modulation of IRI[39,141]. The hepatoprotective mechanisms against IRI may involve the anti-inflammatory properties of IL-6 and the down regulation of TNF-α production[39,142,143]. This could potentially explain the differences in results obtained.

Nevertheless, the influence of IPC in promoting the early release of TNF-α and the up regulation of IL-6 during the reperfusion phase, and subsequent modulation of the hepatocellular proliferation process should not be discounted. In addition, these conflicting results may indicate that IPC can potentiate hepatocyte proliferation via at least two different pathways; liver regeneration via a TNF-α/IL-6-dependent pathway and a mitogen-induced proliferative pathway that does not require TNF-α or IL-6[144,145].

Down-regulation of inhibitory mediators of liver regeneration

Hepatocyte growth is controlled by a balance of both growth-promoting and growth-inhibiting factors[65,146]. While those cytokines and pathways previously described promote liver regeneration following liver resection, there are various mediators that are involved directly or indirectly in inhibiting the liver regeneration process.

Interleukin-1 and transforming growth factor-beta as inhibitors of liver regeneration: IL-1 is a mediator of acute inflammation[147] and is a significant down-regulator of hepatocyte proliferation[148]. IL-1β delays and inhibits hepatocyte proliferation in both culture models[149] and following partial hepatectomy in rats[148]. Other investigators have implicated IL-1α as a mediator of IRI and an inhibitor of liver regeneration[150]. Besides antagonizing the stimulatory effects of growth factors such as HGF[150], IL-1 strongly inhibits hepatocyte DNA synthesis leading to impaired liver regeneration in primary culture models[149]. TGF-β is another inhibitor of hepatocyte DNA synthesis[151-153] and antagonizes the stimulatory effects of HGF during liver regeneration in both in vitro and experimental hepatectomy models[152,154]. This anti-regenerative effect of TGF-β is thought to be modulated by the induction of oxidative stress in hepatocytes[155,156].

Results from several studies have suggested that IPC antagonizes the effects of these inhibitory cytokines. Previous studies have shown that the release of IL-1 is potentially influenced by NO during IRI in a variety of cell types[157-159]. In IPC-treated rats that underwent reduced size liver transplantation, an increase in IL-1α levels were noted following NO synthesis inhibition[150]. This practice abolished the benefits of IPC on hepatic IRI, oxidative stress and liver regeneration[3,150]. Furthermore, the detrimental effects of NO inhibition were not observed when rats subjected to this treatment were subsequently treated with an IL-1 receptor antagonist (IL-1-RA)[150]. IL-1α has also been shown to be involved in pulmonary injury following liver IRI. IPC mediated by NO, reduced IL-1α release and protected against pulmonary damage[160]. Data from these studies suggests that IPC, through increased NO availability, inhibits the release of IL-1, thereby protecting the liver graft and the lungs against liver IRI and preventing inhibition of liver regeneration. Hence, one proposed mechanism by which IPC promotes liver regeneration is by inhibiting the release of growth-inhibitory cytokines from Kupffer cells, such as IL-1, which is dependent on up-regulation of the NO pathway.

The mechanism by which IPC inhibits IL-1 production may also be related to the induction of intracellular stress proteins[161,162], such as heat shock protein 70 (HSP 70). IPC induces over-expression of HSP 70 in isolated hepatocytes[163], and reduces liver IL-1 synthesis under normothermic conditions[164]. Besides expressing cytoprotective effects[165,166], HSP induction leads to the down-regulation of IL-1 synthesis in pancreatic[161,167] and lung[162,168] cell studies. The induction of HSP mediated by IPC may be independent of the NO pathway as the increase in HSP by IPC was not modified when NO synthesis was inhibited[150]. These results suggest that IPC potentiates the overexpression of HSP, via an NO independent pathway, leading to hepato-protection against IRI and ameliorates liver regeneration due to decrease inhibitor production.

Another proposed mechanism by which IPC promotes liver regeneration could be due to the stimulation of IL-1-RA. IL-1-RA is an acute phase protein that has been shown to inhibit the effects of IL-1α and IL-1β by competing for typeIand type II IL-1 receptors[169]. This leads to a decrease in the inflammatory response[169] and abrogates liver IRI in vivo[170]. A recent study on gene expression profiling on patients undergoing partial hepatectomy revealed that IPC stimulated the expression of the IL-1-RA gene[171]. Hence, liver over-expression of IL-1-RA following IPC directly inhibits the effect of high IL-1 concentrations induced by IRI and results in reduced liver injury and necrosis[171]. Although no study has formally assessed the effect of IL-1-RA on markers of liver regeneration, IL-1-RA could be indirectly involved in the regeneration process by antagonizing the inhibitory effects of IL-1 on hepatocyte proliferation.

Inhibitory binding protein for NF-κB (IκB-α) and NF-κB activity: The transcriptional activities of NF-κB are tightly controlled by its inhibitory proteins, especially IκB-α[172]. The phosphophorylation and subsequent degradation of IκB-α leads to the liberation of NF-κB proteins allowing binding to a variety of promoters and triggers gene expression. Data from studies examining the role of NF-κB as an effector of IPC have demonstrated conflicting results. IPC facilitated the activation of transcription factor NF-κB in an in vivo murine model, and this was parallel to the degradation of its inhibitory protein, IκB-α[173]. Ricciardi and co-workers found that IPC increased IκB-α phosphorylation and NF-κB concentration prior to cold ischemia in pig liver grafts. Data from this study suggested that the underlying mechanism involved was related to the activation of second messengers of tyrosine kinase[127]. Another proposed mechanism for NF-κB activation and degradation of IκB-α could be mediated by TNF-α. The release of TNF-α by IPC and pre-treatment with low-dose TNF-α was shown to increase IκB-α degradation and increase NF-κB DNA binding in a mouse model[140]. However, Li et al[174] observed that IPC inhibited the activity of the transcription factor NF-κB during the early reperfusion phase (1 and 2 h), and this was accompanied by diminished TNF-α expression and reduced IRI in liver transplantation rat model.

One explanation for this contradictory evidence might be the different experimental models and methodology used. Nevertheless, these results indicate that IPC does attenuate the nuclear levels of the transcription factor NF-κB. It is possible that other mediators may have an effect in determining the increase or inhibition of NF-κB activity and its corresponding cellular signaling responses.

Increased production of growth factors

Hepatocyte growth factor (HGF): Franco-Gou et al[175] demonstrated an increase in both liver and plasma HGF levels following IPC in reduced size liver transplantation rat model, and this was associated with an increase in hepatocyte proliferation. The modulation of HGF levels by IPC could be mediated by the generation of NO and its effect on TGF-β and HGF concentrations. IPC reduced the levels of TGF-β with an associated increase in HGF levels[150]. Similar results were seen with NO preconditioning treatment[150]. This suggests that IPC reduced TGF-β levels with a parallel increase in HGF and subsequent hepatocyte proliferation, possibly mediated by NO.

HGF may also exhibit the ability to promote hepatocyte survival. In a liver IRI rat model, pre-treatment with HGF inhibited the production of reactive oxygen species and its damaging effects[176]. Similar results were observed in a hypoxia-reoxygenation-induced oxidative stress model in hepatocytes[177]. These results suggest that the anti-apoptotic effect of HGF could pave the way for hepatocyte proliferation following IRI.

Activation of downstream mitogen-activated protein kinases (MAPKs)

Extracellular signal-regulated kinases 1 and 2 (ERK-1/2): ERK-1/2 is predominantly activated by growth-promoting factors. Studies have shown that protein kinase C plays a pivotal role in the activation of ERK-1/2 signaling pathway[163,178] that participates in the preservation of hepatocytes[163]. A number of reports have indicated that protein kinase C was critical for the development of IPC in rat, rabbit and human myocardiocytes[179,180]. Gao and associates showed that the activation of protein kinase C and its downstream ERK-1/2 mediators were increased in IPC-treated in vivo and in vitro models[163]. Data from this study also suggested that the expression of HSP70 was reduced and the protective effect of IPC was diminished when ERK-1/2 activity was reduced by a MAPK inhibitor (PD-98059). HSP70 expression and the cytoprotective effect of IPC were also reduced by a protein kinase C inhibitor (chelerythrine) in both in vitro and in vivo settings. Hence, this suggests that IPC increased the activation of ERK-1/2, via protein kinase C. The increased activation of protein kinase C-dependent ERK-1/2 by IPC may also increase the expression of HSP70. The up-regulation of ERK-1/2 by IPC may help in promoting liver regeneration in both the priming phase and growth factor signaling pathways.

p38 mitogen-activated protein kinases and c-jun-NH2-terminal kinase 1: Both p38 MAPKs and JNK-1 are known to modulate proliferative or apoptotic signaling pathways[181]. The activation of MAPKs and its corresponding downstream signaling pathway is regulated by specific stimuli and is also dependent on cell type. The co-activation of NF-κB protects the hepatocyte from apoptosis and is involved in the priming of hepatocytes to enter the cell cycle[22,182].

Carini and co-investigators demonstrated that hypoxic preconditioning activated the p38 signaling pathway in rat hepatocytes subjected to hypoxia-re-oxygenation injury in vitro[117]. Teoh et al[173] demonstrated that proliferating hepatocytes were identified earlier in IPC-treated livers in a murine model of partial liver ischaemia, and this corresponded with the earlier activation and sustained maintenance of p38 and JNK-1. This suggests that IPC stimulus could prime quiescent hepatocytes to enter the cell cycle early, hence, setting up a regenerative response to compensate for hepatocyte injury by IRI.

Activation of survival and proliferative pathways

Phosphatidylinositol 3-kinase (PI 3-kinase)/Akt signaling pathway: The activation of the PI 3-kinase/Akt cascade has been shown to have a positive impact on cell survival and proliferation[183-186], inhibition of apoptosis and encourage the uptake of glucose and amino acids following stimulation by various growth-promoting factors in certain cells[187-189]. Data from the Izuishi group showed significant Akt activation following IPC in an in vivo model and suggested that this might contribute to the up regulation of glucose and amino acid transport after IRI required for liver regeneration[190].

Activation of nuclear proteins

Cyclin D1: Cyclin and cyclin-dependent kinases are involved in cell cycle regulation[191], in particular, the D group cyclins[192]. Cyclin D1 is a nuclear protein required for cell cycle progression in the G₁ phase[192-194], and controls hepatocyte proliferation[195]. IPC-treated livers showed earlier expression of cyclin D1 protein that corresponded with enhanced entry of hepatocytes into the cell cycle[173]. Cai and co-workers demonstrated that IPC stimulated cyclin D1 mRNA and protein expression during the early reperfusion phase in IPC-treated rat livers[196]. The early production of cyclin D1 could be mediated by the activation of the p38 MAPK pathway[182]. The cyclin D1 promoter region includes binding sites for NF-κB[197] and NF-κB activation is evident in hepatocytes during the early phase of regeneration following partial hepatectomy[48,198]. There is a close link between the activation of NF-κB by degradation of IκB-α cyclin D1 activation and cell cycle progression. IPC can modulate these transcription factors and increase cell proliferation, which is possibly one of the protective mechanisms against IRI.

Energy metabolism

Results obtained from an experimental model of 70% hepatectomy indicated that liver regeneration was closely correlated to the ATP levels of the liver remnant[93]. Studies have shown that IPC in normothermic conditions preserved the adenine nucleotide pool[199,200] and this is thought to be a consequence of the down regulation of cellular metabolism[199,201]. In a rat model of hypothermic transplant preservation injury, hepatocytes exposed to IPC had higher ATP concentrations and increased protein synthesis[202]. This improvement in energy metabolism is thought to contribute to hepatocyte viability following IRI[202]. Besides improvement in energy metabolism, Yoshizumi et al[203] also demonstrated an increase in bile production in IPC-treated rats. However, Franco-Gou and colleagues demonstrated similar energy metabolism (ATP, adenine nucleotides, ATP/ADP ratio and energy charge) in the IPC- and non-IPC-treated rat livers[175]. The difference in results obtained could be related to the difference in experimental models used. However, the role of IPC in modulating liver energy metabolism, which involves the preservation of ATP should not be discounted.

CLINICAL IMPLICATIONS OF ISCHAEMIC PRECONDITIONING IN LIVER SURGERY

The use of IPC as a surgical strategy to limit the detrimental effects of IRI during liver surgery has been extensively researched. Encouraging findings in animal studies in both warm ischaemia[203] and transplantation models[119], led to the first human trial which demonstrated that IPC reduced the severity of post-operative liver injury as well as alleviating endothelial cell injury[17]. Further human liver resection studies have shown that IPC reduces post-operative serum aminotransferase levels in both steatotic[204] and cirrhotic[205] livers and improves post-reperfusion haemodynamic stability[206] (Table 1). Although IPC is protective against IRI[207,208], it did not influence the morbidity and mortality rates in human studies[204,209,210]. In the transplant setting, Jassem et al[211] reported lower serum aminotransferase levels and shorter intensive care stay in IPC-treated cadaveric donor allografts. Other studies have not shown IPC to be beneficial in terms of graft function. Azoulay et al[212] found that although IPC protected cadaveric liver grafts against IRI, this beneficial effect was counter-balanced by decreased early graft function. Cescon and colleagues demonstrated similar protective effects against IRI, but IPC showed no clinical benefit (primary graft function and survival rates) in liver transplantation from deceased donors[213]. Although gaining popularity, the incongruous evidence of the clinical effects of IPC has precluded its widespread adoption in liver transplantation units.

Table 1 Previous published human studies on the results of ischaemic preconditioning following liver resection and transplantation.

Study group	Sample1	Surgery	IPC2	Ischaemia and reperfusion time (min)3	Parameters assessed	Outcome of IPC
Clavien et al[17] (2000)	24 (12)	Liver resection	10I + 10R	IPC and control (TI: 30)	PT, Bilirubin, ALT, AST, Histology, Caspase-3 and 8 activity, SEC apoptosis, Blood loss, Transfusion, ITU stay, LOS	Protective against IRI Beneficial in patients with steatosis
Clavien et al[204] (2003)	100 (50)	Liver resection	10I + 10R	IPC (TI: 36 ± 5.9, Op: 225 ± 73), Control (TI: 35 ± 6.8, Op: 240 ± 92)	PT, Bilirubin, ALT, AST, Histology, Hepatic ATP, Blood loss, Transfusion, ITU stay, LOS	Protective against IRI Beneficial in younger patients, those with steatosis and longer periods of occlusion
Li et al[205] (2004)	29 (14)	Liver resection	5I + 5R	IPC (TI: 18 ± 3.6, Op: 191.3 ± 74.9), Control (TI: 17.4 ± 2.3, Op: 208.2 ± 45.3)	Bilirubin, ALT, AST, Histology, Caspase-3 activity, SEC apoptosis, LOS	Protective against IRI, mainly HCC patients with cirrhosis Shorter hospital stay
Nuzzo et al[209] (2004)	42 (21)	Liver resection	10I + 10R	IPC (TI: 54 ± 19, Op: 321 ± 92), Control (TI: 36 ± 14, Op: 339 ± 112)	PT, Bilirubin, ALT, AST, Transfusion, Morbidity, Mortality	Reduces operative bleeding Protective against IRI
Chouker et al[206] (2004)	68 (22)	Liver resection	10I + 10R	IPC (TI: 32 ± 6.3, Op: 251 ± 46), Control without PR (TI: NA, Op: 52 ± 30), Control with PR (TI: 35 ± 11, Op: 257 ± 83)	ALT, AST, Fluid loss, Transfusion, 4Cardiovascular status	Protective against IRI Improves haemodynamic stability
Chouker et al[207] (2005)	75 (25)	Liver resection	10I + 10R	IPC (TI: 35.5 ± 2.7, LR: 32.2 ± 2.0), Control without PR (TI: NA, LR: 39 ± 4.5), Control with PR (TI: 35.6 ± 2.6, LR: 33.2 ± 2.3)	IL-6, IL-8, Cytochrome c, Adhesion molecules [B2-integrins (CD18)], Histology (neutrophil infiltration)	Protective against IRI by attenuating neutrophil activation and IL-8 release
Chouker et al[208] (2005)	73 (25)	Liver resection	10I + 10R	IPC (TI: 35.12 ± 13.6, LR: 31.50 ± 9.1), Control without PR (TI: NA, LR: 34.77 ± 16.5), Control with PR (TI: 34.2 ± 10.9, LR: 32.13 ± 10)	PT, ALT, AST, α-GST	Protective against IRI Prevented early rise of α-GST
Koneru et al[210] (2005)	62 (34)	Transplant	5I + 5R	IPC (CI: 384 ± 92, WI: 41 ± 5.8), Control (CI: 415 ± 87, WI: 37 ± 5.6)	INR, Bilirubin, ALT, AST, Histology (apoptosis, hepatocyte swelling), LOS, Survival (6 mo)	No beneficial effect
Azoulay et al[212] (2005)	91 (46)	Transplant	10I + 10R	IPC (CI: 436 ± 116, Op: 441 ± 119), Control (CI: 461 ± 96, Op: 462 ± 98)	PT, Bilirubin, ALT, AST, Histology, Graft function, Morbidity, Mortality	Better ischaemic tolerance Decreased early graft function
Jassem et al[211] (2005)	23 (9)	Transplant	10I + 10R	IPC (CI: 620 ± 190, WI: 43.9 ± 13), Control (CI: 665 ± 280, WI: 40.4 ± 9)	AST, INR, Lactate, ITU stay, Histology (neutrophil infiltration, platelet deposition), Graft function	Protective against IRI Reduces inflammatory response Shorter ITU stay
Cescon et al[213] (2006)	47 (23)	Transplant	10I + 15R	5IPC [TI: 388 (259-830), Op: 440 (225-725)], Control [TI: 383 (279-695), Op: 465 (280-1015)]	PT, Bilirubin, ALT, AST, Histology (neutrophil, lymphocyte infiltration, iNOS, apoptosis), Graft function, Survival (1 yr)	Protective against IRI No clinical benefit

¹Patients stated in brackets are the number of patients who had ischaemic preconditioning treatment;

²IPC was performed by portal triad clamping in all these studies;

³Ischaemia and operative times are presented as mean ± SD unless otherwise stated;

⁴Cardiovascular status refers to mean arterial pressure, central venous pressure, heart rate, stroke volume index, systemic vascular resistance index, fluid infusion and catecholamines requirements;

⁵Ischaemia and reperfusion times in this study were presented as median (range). I: Ischaemia; R: Reperfusion; IPC: Ischaemic preconditioning group; PR: Pringle maneuver; TI: Total ischaemia time; CI: Cold ischaemia time; WI: Warm ischaemia time; Op: Total operative time; LR: Liver resection time; ITU: Intensive therapy unit; LOS: Length of hospital stay; PT: Pro-thrombin time; INR: International normalized ratio of pro-thrombin time; ALT: Alanine aminotransferase; AST: Aspartate aminotransferase; α-GST: Alpha-Gluthathione S-Transferase; iNOS: Inducible form of nitric oxide synthase; ATP: Adenosine triphosphate; HCC: Hepatocellular carcinoma; IL: Interleukin; SEC: Sinusoidal endothelial cell; IRI: Ischaemia-reperfusion injury.

Following major liver resection and IRI, the ability of the liver to regenerate is crucial to maintain liver function. This also has implication in live donor orthotopic liver transplantation and transplantation of segmental liver grafts. Since Yamada and co-workers first demonstrated that IPC significantly increased the regenerative capacity of the remaining hepatocytes in a rat model of IRI[20], various investigators have attempted to elucidate the role of IPC on liver regeneration using both culture and animal model studies as described above. At present, although there is no clinical trial published on the effect of IPC on liver regeneration, there are studies evaluating methods of monitoring liver regeneration. Special radiological imaging techniques currently available not only show volume of regeneration, but also determine functional ability of the remnant and regenerated liver[214,215]. This is another step towards assessing the role of IPC in liver regeneration in the clinical scenario.

With the increasing laboratory evidence of protection against IRI and improved liver regeneration by IPC, several aspects of this strategy could be developed pharmacologically that may be more clinical applicable than IPC itself. This is especially in cases with a background of chemotherapy-induced steatohepatitis[216] and cirrhosis[217,218]. Pharmacologic agents targeting mediators of IPC that can be potentially developed include HGF, IL-6 and IL-1-RA. However, several issues such as the timing of administration of these agents, therapeutic doses and immunological response of the recipient need to be determined. Further understanding of the mechanistic pathways of IPC may pave the way for the development of these agents that are capable of conferring protection against IRI and promote liver regeneration.

CONCLUSION

Liver regeneration is of clinical significance in view of the increasing number of major liver resections and the increasing use of marginal donor liver and split-liver allografts for transplantation. Successful patient outcome often depends on liver regeneration, particularly in patients with cirrhotic and steatotic livers. Regeneration of the liver following IRI and major liver surgery is a complex process that involves the integration of a network of cytokines, growth factors, kinases, transcription factors and metabolic demands of the liver.

In comparison with the evidence available on the effect of IPC on IRI, its role in liver regeneration is still undetermined. However, current research has demonstrated that the beneficial effects of IPC on liver regeneration is mediated by up regulating growth-promoting factors, suppressing growth-inhibitory factors and preserving energy levels for regeneration. Nevertheless, more studies are still required to further delineate the underlying pathophysiology of IPC and impact on mediators of liver regeneration. It is also important to determine whether the beneficial effect of IPC in the laboratory setting is reproducible in clinical practice. By understanding the underlying mechanisms by which IPC influences liver regeneration, other strategies as alternatives to IPC, could be developed to modulate the regenerative pathways in the clinical setting and improve outcomes of patients following major liver resection and transplantation. The assessment of IPC on liver regeneration in human studies is clearly the next step.