Liver regeneration in the clinical setting

With its unique regenerative capacity and the potential of simultaneously maintaining vital organ specific functions, the liver is one of the most intriguing organs of the human body. Liver resection is considered as the only curative treatment option for several neoplastic entities of the liver [1, 2]. Accordingly, partial hepatectomies are performed for eligible patients based on the resection of all radiologically as well as macroscopically detectable tumor while preserving at least 20–25 % of healthy total liver volume [3, 4]. Despite substantial improvements in surgical techniques and perioperative care, postoperative morbidity and mortality remain an important concern after liver resection [5, 6]. The most significant factor determining morbidity and mortality following partial hepatectomy is the ability of the remnant liver to regenerate. To date, no efficient treatment option for patients suffering from postoperative liver failure exists. Accordingly, understanding the process of liver regeneration, to identify potential therapeutic targets, is of utmost importance. Although basic research has put forward several highly promising target molecules, only a few have been validated in the clinical setting [7–10].

A tightly regulated interplay of different stimuli

Although the process of liver regeneration requires a complex interplay of various factors, it follows a controlled pattern and can be divided into a total of three different phases: the initiation/priming phase, followed by the proliferation, and ultimately the termination phase [11, 12]. Diverse priming factors, growth factors, comitogens, and their suppressors are associated with these stages and are overlapping in their mode of action, allowing a continuous progression of liver regeneration. The process of regeneration starts immediately after liver resection and lasts for about 8–15 days and initial changes can be observed already 5 min after surgery [12]. During this tightly regulated process, hepatocytes enter one first round of replication (leading to 60 % of hepatocytes). Another second round of DNA synthesis is entered by a smaller proportion of hepatocytes ultimately establishing the original liver mass. Apoptosis of hepatocytes marks the termination of liver regeneration, presumably to an overshooting of the regenerative response [12, 13]. Hepatocytes are the first cell population to enter the process of regeneration, followed by nonparenchymal cells of the liver and cells of the biliary tract [14]. While liver mass is be restored quickly, lobular reorganization seems to take place for several weeks after surgery [15].

Several different molecules have now been identified to affect liver regeneration and studies showed that blockage of a single molecule generally only delays liver regeneration [16]. Ultimately, liver regeneration is completed even if apparently central mediators are blocked [16]. Indeed, downstream effects are highly redundant between distinct effector molecules, so that blocking of a single component can be compensated by another. However, this delay in regeneration is not without clinical relevance. Indeed, partial postoperative liver dysfunction, caused by delayed liver regeneration, is associated with an increased incidence of postoperative morbidity as we have also observed in our patients (Fig. 1). However, only a few of these patients go into complete liver failure, which is fatal and currently untreatable. Thus, an in-depth understanding of the molecular processes involved in liver regeneration of patients undergoing liver resection is essential for an improvement of current therapeutical interventions. Animal models give important hints for a better understanding of liver pathophysiology but may not completely reflect the human situation. Rodent knock-out models often represent with altered liver histology and long-term adaptation to specific genetic alterations that might substantially affect liver regeneration [16]. With respect to the inevitable differences between rodent models and humans, we placed our focus on prospective translational clinical trials to evaluate the relevance of basic research findings in the human setting.

Fig. 1
figure 1

Postoperative delayed liver regeneration is associated with increased postoperative morbidity. A total of 134 patients were evaluated for postoperative liver dysfunction using the ISGLS criteria to reflect delayed postoperative liver regeneration [65]. These groups were then compared for the incidence of postoperative morbidity a and severe morbidity b according to Dindo et al. [66] (severe morbidity = grade III–V)

Full size image

Platelets and liver regeneration

Several factors have been identified to affect liver regeneration [12, 17]. Among those, hepatocyte growth factor, vascular endothelial growth factor (VEGF), epidermal growth factor (EGF), thrombospondin-1 (TSP-1), serotonin (5-HT) and transforming growth factor β (TGF-β) have been described to regulate the complex process of liver function recovery [17]. More recently, animal studies have shown that platelets are able to modulate liver regeneration [18–21]. Platelet inhibition by the P2Y12 inhibitor, clopidogrel, has previously been shown to prevent hepatocyte proliferation after 70 % partial hepatectomy in mice, emphasizing a critical role of platelets in liver regeneration [21]. Accordingly, in platelet-depleted mice liver regeneration is severely suppressed [19–22]. In contrast, thrombopoietin (TPO) induced thrombocytosis was found to promote liver regeneration after partial hepatectomy [18, 20]. Even direct application of platelet-rich plasma via the portal vein seems to accelerate postoperative liver regeneration [23]. All these experimental data document the beneficial effect of platelets in postoperative liver regeneration. Of note, also clinical studies suggest that platelets affect postoperative liver regeneration. In particular, preoperative platelet counts have been identified as outcome predictors after liver resection [24, 25]. Moreover, Alkozai et al. [26] demonstrated that immediate postoperative low platelets, at the onset of liver regeneration, are independently associated with delayed liver function recovery. In line with the experimental data on TPO, we were able to document that patients who fail to increase circulating TPO levels after liver surgery suffer from a significantly increased risk of postoperative liver dysfunction [27]. Accordingly, several lines of evidence clearly suggest a central role of platelets in liver regeneration.

Indeed, platelets store most of the relevant factors for liver regeneration in their granules [28, 29]. Platelets contain three types of granula: α-, dense, and lysosomal granula. Accumulating evidence suggests that dense and α-granules exhibit distinct functions in liver regeneration and will therefore be discussed separately.

Dense granules

Several researchers aimed to identify the mechanism how platelets modify postoperative liver regeneration. Lesurtel et al. reported on the role of platelet-derived 5-HT in a mouse model. The majority of the body’s 5-HT pool is generated in the enterochromaffin cells of the gastrointestinal tract from L–tryptophan through by tryptophan hydroxylase activity. Apart from its relatively short half-life in the extracellular matrix, 5-HT is rapidly taken up and stored by platelets. Lesurtel et al. [21] used knockout mice, lacking the rate-limiting enzyme of 5-HT biosynthesis, to demonstrate that these mice fail to show adequate liver regeneration after partial hepatectomy. As a result, platelet-derived 5-HT has been postulated to be a relevant effector of platelets on hepatocyte growth and liver regeneration. Many experimental models have now verified that liver regeneration is affected by platelet-derived 5-HT and the interaction with its receptors [21, 30–35]. Of note, we were able to provide the first clinical evidence that intra platelet (IP) 5-HT may promote liver regeneration in humans [7]. Accordingly, patients with a reduced preoperative IP 5-HT pool, that may be released upon activation at the site of liver regeneration, were found to have an increased incidence of postoperative complications and more frequently suffered from postoperative liver dysfunction and delayed liver function recovery.

Alpha granules

Besides dense granules, which contain small molecules including 5-HT, α-granules, which contain larger molecules, are of major interest for liver regeneration, as they contain several important regulatory factors such as VEGF or TSP-1. While dense granules are the first to be discharged upon activation, the release of α-granules has been discussed controversially within the past few years. In 2008, Italiano et al. [36] reported that angiogenic factors are stored in different α-granule populations and can be specifically released in response to distinct stimuli. In particular, protease activating receptor 1 (PAR-1) activation caused a release of pro-angiogenic factors (such as VEGF), while PAR-4 stimulation induced anti-angiogenic factor secretion (such as TSP-1). While some studies failed to confirm this particular mechanism [37, 38], other studies support the possibility that α-granules differ in their composition as they observed agonist-specific release of distinct granules [39, 40]. The specific release of α-granule content is of major importance during the process of liver regeneration as α-granules contain regulatory factors with opposing function. In particular, while VEGF is a well-established promotor of liver regeneration [41–44], TSP-1 has recently been shown to represent a negative regulator of liver regeneration after partial hepatectomy [45]. TSP-1 is a homotrimeric glycoprotein with well-known functions in hemostasis and angiogenesis [46–48]. In adult tissue, TSP-1 is primarily released at sites of tissue remodeling where it acts in the pericellular space to regulate cell proliferation and extracellular matrix remodeling [49]. Hayashi et al. [45] reported that TSP-1 suppresses liver regeneration by the activation of latent TGF-β1. Based on a mouse model of partial hepatectomy, they were able to demonstrate that TSP-1 was induced as an immediate early gene in the initial response to liver resection. Moreover, they were recently able to demonstrate that a leucine-serine-lysine-leucine peptide, that inhibits TSP-1 mediated TGF-β activation, promoted liver regeneration in mice after partial hepatectomy, suggesting potential clinical applicability [50]. In line with these results, we were recently able to provide the first clinical evidence for TSP-1 as a relevant suppressor of liver regeneration after partial hepatectomy [10]. Patients with elevated TSP-1 levels, on the first postoperative day (the onset of liver regeneration), were found to suffer from an increased incidence of postoperative liver dysfunction and poor postoperative clinical outcome, as a reflection of reduced postoperative liver regeneration. Importantly, we were able to confirm these observations and the predictive marker potential of TSP-1 in an independent prospective validation cohort [9].

Considering the opposing roles of VEGF and TSP-1 in liver regeneration, a differential release of granule-stored mediators via specific activation of platelets would enable a distinct regulation of liver regeneration by platelets. However, proof that the selective α-granule release does indeed occur in vivo is still missing. Moreover, potential mechanisms on how differential α-granule release could be mediated have to be further defined.

Portal hemodynamics in liver regeneration

Under physiological conditions, the liver is supplied with arterial as well as venous blood. After partial hepatectomy the highly regulated arterial blood flow remains unchanged, while changes in the venous blood flow occur and are highly dependent on the volume of the remaining liver tissue as well as the architecture of the liver as determined by fibrosis/cirrhosis or steatosis [16]. This increase in portal venous pressure is associated with the activation of a variety of signaling pathways within the liver that contribute to liver regeneration [16]. Accordingly, portal blood flow has been proposed to act as one of the major inducers of liver regeneration after hepatectomy. However, excessive increase in portal venous pressure, as observed after major liver resection or in severely diseased livers impairs liver regeneration and can result in postoperative liver dysfunction or even liver failure [51]. In this context, Asencio et al. [52] postulated that the development of postoperative liver failure is not per se determined by the mass of the liver remnant, but is rather determined by the hemodynamic parameters of the hepatic circulation. Summarizing growing experimental as well as clinical evidence they suggest that not only the absolute amount of remaining liver tissue, the so-called “small for size” aspect, but rather the postoperative portal venous flow, a “small for flow” syndrome, seems to be critical in liver regeneration [52]. Indeed, portal hypertension has been found to be associated with several morphological changes within hepatic ultrastructure with a pronounced effect on the sinusoidal endothelial cell lining [53, 54]. These observations are of major clinical importance as there are several options to interact with portal venous pressure, while the therapeutic options for postoperative liver failure are currently very limited. In particular, Fahrner et al. [55] were able to document that the application of terlipressin, a vasoconstrictor with primary effects on the splanchnic vasculature and the ability to decrease hepatic blood flow, was able to accelerate postoperative liver regeneration in a mouse model. In particular, one of the most relevant beneficial effects of terlipressin seemed to be a reduction in the disruption of hepatic sinusoids. Furthermore, the beneficial effects seemed to be even more pronounced in steatotic mouse livers. Importantly, these experimental results are currently evaluated in a therapeutic intervention trial in patients undergoing major liver resection (ClinicalTrials.gov Identifier: NCT01921985).

While there is extensive evidence for the role of portal venous pressure in liver regeneration, [56] the simple increase of portal flow itself is not sufficient to induce liver regeneration [57]. In particular, Mortensen et al. [57] demonstrated that an increase in portal flow by an aorto-portal shunt did not induce liver regeneration as seen after partial hepatectomy, suggesting a substantial contribution of other, potentially humoral factors to the regenerative process .

Potential links between platelets and portal venous pressure

As an isolated increase of portal pressure is not sufficient to induce liver regeneration [57], other circulating factors, affected by the changes in hemodynamics after liver resection, seem to be of crucial relevance. In this context, platelets, loaded with regulatory factors, could be central players that sense and respond to fluctuations of portal venous pressure thereby translating changes in blood flow dynamics to molecular signals for liver cells. Indeed, Schlegel et al. [58] recently proposed that circulating growth factors are more important in liver regeneration than the increase in portal venous pressure itself. Of note, the investigators were able to demonstrate that the simple injection of plasma derived from mice with accelerated liver regeneration was vital and sufficient to promote liver regeneration in mice undergoing singular portal vein embolization without liver resection. As platelets represent one of the major sources for hepatotrophic growth factors and are also sensitive for changes in blood flow dynamics and one of the first cells to arrive within the liver after resection, it is reasonable to speculate that they contribute to accelerating liver regeneration in response to elevated portal pressure. The sudden increase in venous pressure after liver resection exposes the liver to excessive hemodynamic forces, which greatly affects liver sinusoidal endothelial cells (LSEC). Disruption of the fenestrations of the LSEC layer shortly after hepatectomy seems to be of central relevance as it allows rapid accumulation of platelets within the liver sinusoids [20, 59–61]. Moreover, endothelial cell damage, caused by an increase in portal venous pressure, has been shown to promote platelet adhesion and activation [62, 63]. Platelets directly interact with LSEC thereby modifying hepatocyte proliferation [64]. Accordingly, several experimental reports support the hypothesis that postoperative changes in intrahepatic hemodynamics result in postoperative platelet activation thereby promoting liver regeneration, as summarized in Fig. 2.

Fig. 2
figure 2

Possible relation between portal venous pressure and platelet activation. a The intrahepatic microcirculation under normal conditions with an intact liver sinusoidal endothelial cells lining and no substantial platelet extravasation into the space of Disse. b Potential changes after partial hepatectomy with a disruption of the endothelial cell lining and platelet adhesion to liver cells resulting in the local release of platelet-stored growth factors, which in turn modulates liver regeneration

Full size image

Summary

Several experimental as well as clinical studies have indisputably demonstrated that platelets play a critical role in postoperative liver regeneration. However, the exact mechanism of specific postoperative platelet activation remains to be elucidated. Postoperative portal venous pressure seems to be a likely candidate to mediate postoperative platelet activation and thereby affect liver regeneration via the release of platelet-stored factors. By exploring the connections between postoperative platelet activation and postoperative portal venous pressure in the clinical setting, we might be able to identify therapeutic targets to improve clinical care of patients after liver resection.

Conflict of interest

No conflict of interest.