If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Division of Gastroenterology and Hepatology, Department of Internal Medicine III, Medical University of Vienna, 1090 Vienna, AustriaLudwig Boltzmann Institute for Rare and Undiagnosed Diseases (LBI-RUD), 1090 Vienna, AustriaCeMM Research Center for Molecular Medicine of the Austrian Academy of Sciences, 1090 Vienna, Austria
Corresponding author. Address: Institute for Developmental Immunology, Biocenter, Medical University of Innsbruck, Innrain 80, A-6020, Innsbruck, Austria; Tel.: +43-512-9003-70380, fax: +43-512-9003-73960.
Institute for Developmental Immunology, Biocenter, Medical University of Innsbruck, Innsbruck, AustriaLudwig Boltzmann Institute for Rare and Undiagnosed Diseases (LBI-RUD), 1090 Vienna, AustriaCeMM Research Center for Molecular Medicine of the Austrian Academy of Sciences, 1090 Vienna, Austria
A balanced increase in DNA content (ploidy) is observed in some human cell types, including bone-resorbing osteoclasts, platelet-producing megakaryocytes, cardiomyocytes or hepatocytes. The impact of increased hepatocyte ploidy on normal physiology and diverse liver pathologies is still poorly understood. Recent findings suggest swift genetic adaptation to hepatotoxic stress and the protection from malignant transformation as beneficial effects. Herein, we discuss the molecular mechanisms regulating hepatocyte polyploidisation and its implication for different liver diseases and hepatocellular carcinoma. We report on centrosomes’ role in limiting polyploidy by activating the p53 signalling network (via the PIDDosome multiprotein complex) and we discuss the role of this pathway in liver disease. Increased hepatocyte ploidy is a hallmark of hepatic inflammation and may play a protective role against liver cancer. Our evolving understanding of hepatocyte ploidy is discussed from the perspective of its potential clinical application for risk stratification, prognosis, and novel therapeutic strategies in liver disease and hepatocellular carcinoma.
Polyploidisation describes the process of whole genome duplication. Polyploid cells harbour multiple copies of the entire genome, which, in contrast to aneuploidy, allows them to maintain balanced gene expression.
For many of these cell types the physiological relevance of polyploidisation is still unclear (Fig. 1A).
Polyploidisation can be achieved through different mechanisms (Fig. 1B). Endocycles and endomitosis are abortive cell cycle modes in which cells, like megakaryocytes, skip nuclear (karyokinesis) and cellular division (cytokinesis).
Polyploidisation of hepatocytes and cardiomyocytes is mostly the result of incomplete cytokinesis, generating binucleated cells. Both of these cell types can also undergo endocycles, which leads to an increase in nuclear ploidy.
Similar to polyploid cardiomyocytes or epithelial cells found in the lactating mammary gland, the increase in cell mass and ploidy is thought to be necessary for hepatocytes to carry out their excessive, physical, secretory and metabolic workload (Fig. 1A).
The spatial distribution of polyploidy in the liver is mostly independent of metabolic liver zonation but a tendency toward lower ploidy in the periportal region (zone 1) has been reported. Considering temporal frequency, polyploid hepatocytes usually first appear in midlobular zone 2.
Hepatocyte polyploidisation is a developmental process that is regulated by transcriptional and post-transcriptional mechanisms.
Drug- and diet-induced liver injury primarily affect the periportal (zone 1) and pericentral (zone 3) hepatocyte populations (Fig. 2). Two recent studies show that zone 2 hepatocytes can regenerate the liver upon such types of zonal injury as this hepatocyte population is shielded from the insult.
Upon partial hepatectomy, a type of injury that induces regeneration without a zonal bias, a wave-like propagation of proliferation from the periportal vein (zone 1) to the central vein (zone 3) can be detected.
The polyploidisation process has mostly been studied in mice and rats. It was shown to be triggered by weaning, giving rise to binucleated tetraploid daughter cells, but it is limited to a short window of time during this period of development.
Further proliferation of such a polyploid cell either generates 2 mononucleated tetraploids or a binucleated octaploid cell, if cytokinesis fails again. After this initial wave, hepatocyte polyploidy increases only modestly over the lifetime of the organism. Consistent with this, a recent study shows that homeostatic proliferation is very low (an estimated cell fraction of only 0.042% in zone 2) but coincides with successful cytokinesis in >90% of proliferation events.
Strikingly, in response to injury, quiescent polyploid hepatocytes re-enter the cell cycle to regenerate the liver. Although the overall liver ploidy is maintained in the regenerated liver, the ratio between mononucleated and binucleated hepatocytes changes. Dependent on the degree of injury, hepatocytes can undergo endoreduplication which increases the fraction of mononuclear hepatocytes.
The main mechanisms regulating polyploidisation are linked to insulin signalling in response to nutritional changes and transcriptional control of cytokinesis effector genes, but post-transcriptional effects involving microRNA (miR) action or direct interference with the process of cytokinesis have also been reported (see below).
Transcriptional control of liver ploidy
E2F transcription factors play a crucial role in the regulation of hepatocyte polyploidy.
Upon cell cycle entry, activators E2F1-3 induce transcription of their target genes, including the S-phase cyclins E and A, and several cytokinesis effectors. Additionally, E2F1-3 activate the expression of their own antagonists, E2F7/8, which immediately repress the same set of target genes.
This could be ascribed to the fact that some of the top transcriptional targets of E2F7/8 are essential cytokinesis factors: the guanine nucleotide exchange factor ECT2 (epithelial cell transforming 2), Aurora B kinase, RACGAP1 (Rac GTPase activating protein 1), MKLP1 (or KIF23), and other genes involved in DNA replication, mitosis, and p53-dependent cell cycle arrest.
The tumour suppressor p53 and its most prominent transcriptional target, the cyclin-dependent kinase (CDK) inhibitor p21, have also been shown to regulate hepatic ploidy. Consistently, loss of either p53 or p21 promote hepatocyte polyploidisation during both normal development and regeneration.
A mouse model deficient in all Period genes (Per1-3) displayed transcriptional increases in Mkp1 (mitogen-activated protein kinase phosphatase 1). This phosphatase, in turn, inactivated ERK1/2 during cytokinesis, preventing successful cell division (Fig. 3A).
Weaning, the transition from suckling to solid food, causes an increase in insulin levels, which triggers the failure of cytokinesis in rodent hepatocytes. Insulin activates the PI3K/AKT pathway, which is suggested to impair Ras homolog family member A (RhoA) activity.
Desdouets and colleagues found that in rat weanlings RhoA and other essential cytokinesis effectors are delocalised upon increased insulin signalling, thereby preventing cleavage furrow formation and ingression.
The centrosome-PIDDosome-p53 signalling axis is part of the E2F transcriptional regulatory circuit and functions as a break on hyperpolyploidisation.
Another way to repress cytokinesis factors and facilitate hepatocyte polyploidisation involves miR-122, which is mainly expressed in the liver. The mRNA targets that are repressed or degraded by miR-122 include those encoding cytokinesis factors such as RhoA and CUX1 (cut like homeobox 1), a transcriptional activator of centralspindlin components. Consistently, loss of this miR significantly reduces hepatic ploidy.
Several other pathways have been found to negatively regulate polyploidisation, thereby counteracting the effects of insulin, miR-122 and E2Fs. Loss of LKB1 (liver kinase B1), cyclin E1, or CDK1 leads to hyperpolyploidy.
Intriguingly, the Hippo pathway also acts through SKP2 to regulate hepatocyte polyploidy. Yes-associated protein (YAP) activation causes AKT phosphorylation, leading to cytoplasmic retention of SKP2 and, therefore, p27 accumulation in the nucleus (Fig. 3B). In addition, degradation of the transcription factor FOXO1/3 (forkhead box O1/3) is deregulated, which together results in hepatic hyperploidy.
The PIDDosome comprises the death domain (DD) containing proteins p53-induced DD protein 1 (PIDD1) and CRADD (also known as RAIDD). CRADD/RAIDD also contains a caspase-recruitment domain, which allows it to interact with and activate an aspartate-specific protease (CASP2). This interaction was initially reported to promote cell death upon DNA damage.
Over the last few years, however, we have gained deeper insights into how CASP2 activation is accomplished in normal physiology. First reports by Fava et al. suggested that the PIDDosome becomes activated in transformed or immortalised cells that fail cytokinesis.
The trigger for PIDDosome activation appeared to be the extra centrosome that accumulates in such cells. Centrosomes consist of a pair of centrioles and the surrounding pericentriolar material; they organise the mitotic spindle apparatus and template cilia formation in multiple differentiated cell types. As with DNA, the centrosome duplicates exactly once per cell cycle to enable the formation of 2 spindle poles for chromosome segregation.
Subsequent stabilisation of p53 and the expression of its target genes, most notably p21, results in cell cycle arrest and therefore ensures that polyploid cells with extra centrosomes no longer proliferate.
Nevertheless, 5 days after weaning most hepatocytes fail cytokinesis and become tetraploid, which was paralleled by PIDDosome-dependent p53 accumulation. PIDDosome-deficient mice present considerably increased hepatocyte ploidy levels, mimicking the well-known p53 and p21 loss-of-function phenotypes.
Together, this confirms the physiological relevance of the PIDDosome-p53 axis during liver development. Strikingly, chromatin immunoprecipitation sequencing analysis revealed Casp2 and Pidd1 as direct targets of E2F1 and E2F8, identifying them as essential effectors of the E2F ploidy control circuit. Consistently, hepatocyte-specific loss of E2F7/8 in mice leads to increased expression of Pidd1 and Casp2, which is no longer detectable upon additional deletion of E2F1 (Fig. 3B).
Inflammatory liver pathologies can directly or indirectly promote hepatocyte polyploidisation.
PIDD1 and CASP2 are barely detectable in adult hepatocytes, raising the possibility that the reported role for p53 in limiting ploidy during liver regeneration may involve alternative modes of p53 activation. Yet, during regeneration following two-thirds partial hepatectomy, hepatocytes re-express PIDD1 and CASP2 and the lack of a functional PIDDosome-p53-p21 axis resulted in faster regeneration of liver mass.
The PIDDosome component CRADD/RAIDD is constantly expressed in mouse livers and recent work suggests that it may be preferentially found in zone 2 hepatocytes. More strikingly, CRADD/RAIDD appears to suppress liver regeneration emanating from zone 2 hepatocytes. This is suggested by the results from an elegant targeted transposon-based CRISPR loss-of-function and a trans-activation screen, using selected single guide RNAs for genes preferentially expressed in zone 2 hepatocytes.
Together, these observations open up the possibility that breaking the ploidy barrier by chemical inhibition of the PIDDosome may promote liver regeneration in transplantation or other regenerative settings. In support of conservation of mechanism, we confirmed similar dynamics of E2F1, CASP2 and p21 activation, along with increased ploidy, in liver biopsies sequentially taken from patients undergoing the ALPPS procedure (associating liver partition and portal vein ligation for staged hepatectomy).
Aside from beneficial polyploidy in these specialised cell types, several pathogens can induce polyploidisation. Macrophages can become polyploid as macrophage giant cells in response to bacterial infections.
Infections with oncogenic viruses, such as human papilloma virus, contribute to the polyploidy seen in cancer. However, the relative contribution of polyploidy to the pathogenesis of virus-driven cancers remains to be clarified.
Polyploidy is a common feature of tumorigenesis, found in more than one-third of human cancers.
However, in the liver, polyploidy has been suggested to promote the generation and expansion of beneficial karyotypes to cope with stress. This has been investigated using a mouse model for hereditary chronic tyrosinemia caused by deficiency of fumarylacetoacetate hydrolase (Fah-/-). This results in the accumulation of a toxic intermediate that causes cell death. Inhibition or loss of the upstream enzyme, homogentisate 1,2 dioxygenase (Hgd), that generates the toxic intermediate can reverse this effect. Livers of Fah-/-Hgd+/- animals formed regenerative nodules, mainly consisting of a subpopulation of cells with a specific karyotype that had lost the chromosome with the remaining Hgd allele.
Wilkinson et al. utilised liver-specific E2F7/8 knock-out to achieve lower ploidy in the Fah-/-Hgd+/- model for chronic liver injury. Lower polyploidy reduced the degree of aneuploidy and impaired successful regeneration, as monitored by animal survival.
While the immediate functional advantage of being polyploid is clear for many of the aformentioned cell types, the benefit for liver function and the organism as a whole is less well understood. Recent studies suggest differences between diploid and polyploid hepatocytes in their metabolic capacity and specialisation. Kreutz et al. performed transcriptome analysis on diploid vs. polyploid hepatocytes, which were sorted based on their insulin affinity. Most differentially regulated genes were associated with RNA, protein, and fatty acid metabolism, as well as transport of fatty acids and cholesterol.
In analogy to other polyploid cell types, hepatocyte polyploidisation could be a way to save energy in periods of stress and/or extensive liver growth. Consistent with this idea, the first wave of cytokinesis failure occurs during weaning, when liver growth coincides with a high metabolic load due to the severe nutritional change.
Similarly, several mouse models, but also studies with human patient material, suggest that hepatocyte ploidy increases in chronic diseases where proliferation is induced to compensate for ongoing loss of liver tissue.
In contrast to most other tissues, in the liver, hepatocyte polyploidy protects from tumorigenesis by buffering tumour suppressor gene loss.
Ploidy in liver injury and inflammation
Chronic or repetitive liver injury can cause apoptosis and/or necrosis of hepatocytes, creating a proinflammatory microenvironment that can lead to fibrosis and cirrhosis. All of these conditions are accompanied by compensatory hepatocyte proliferation which may ultimately culminate in hepatocellular carcinoma (HCC).
Several types of chronic liver diseases have been associated with increased hepatocyte polyploidy, consistent with the notion that proliferation under stress conditions promotes alternative cell cycle modes, including failed cytokinesis and endoreduplication. Unlike weaning-induced polyploidisation, during regeneration following partial hepatectomy, hepatocytes tend to switch to endoreduplication rather than cytokinesis failure in mice.
Whether the higher frequency of mononucleated polyploids has any direct consequences for the liver is unclear. It seems more likely that the presence of these cells is indicative of compensatory regeneration and proliferation phases in the disease context. Thus, the frequency of these cells could potentially correlate with the severity of the chronic liver disease.
Polyploidisation in response to injury is also found in other tissues and organisms, such as wound healing in drosophila, or heart regeneration in zebra fish.
Together, these data support the hypothesis that polyploidisation might be a way to achieve rapid regeneration by saving the time and the energy required for actual cell division in periods of stressful growth, triggered by acute or chronic injury. Interestingly, however, recent data point towards a mutual relationship of polyploidy and fibrosis, suggesting that the situation is actually more complex.
Toyoda et al. reported a clear correlation between the number of mononuclear polyploids and the severity of HBV- or HCV-driven chronic hepatitis, which was inferred from the degree of fibrosis and the severity of hepatitis. Interestingly, the fraction of binucleated polyploids was higher in patients with HBV than in those with HCV, independent of the severity of the disease.
As HCV and HBV are unrelated viruses, they might act through distinct routes to promote polyploidisation. Overexpression of the HCV core protein in different cell lines in vitro has been suggested to increase polyploidy by interfering with the retinoblastoma pathway, but further work is needed to dissect this observation.
While the interrelation of HCV and polyploidy is poorly understood, the HBV protein HBx has been directly implicated in modifying the cell cycle. HBx was found to induce premature DNA replication in G2-phase by deregulation of replication licensing factors and Polo like kinase 1 (PLK1) in an immortalised hepatocyte cell line. This unscheduled DNA replication was accompanied by a high degree of DNA damage and increased polyploidy, the latter was further augmented in the absence of p53.
Moreover, livers of these mice showed persistent activation of p38/MAPK and increased expression of PLK1. p38 signalling clearly increased the fraction of mononucleated tetraploids, yet the consequences and the detailed mechanism of how HBx modifies ploidy remained elusive.
While alterations in cell cycle progression in the presence of viral proteins can directly increase ploidy, it remains to be explored to what extent these hepatotropic viruses may increase ploidy by promoting hepatocyte fusion during viral entry. Polyploidisation via cell-to-cell fusion has been observed upon infection with human papilloma virus or human cytomegalovirus, with the latter also being a hepatotropic virus.
The NAFLD spectrum comprises liver steatosis resulting from hepatocyte triglyceride accumulation that can progress to steatohepatitis (NASH). Importantly, NAFLD/NASH is most often caused by caloric excess and sedentary lifestyle.
Gentric et al. investigated the ploidy profiles in different NASH models and found increased hepatocyte ploidy in obese mice as well as mice on NASH-inducing diets, such as high-fat or methionine- and choline-deficient diets.
Interestingly, hepatocytes in these NAFLD/NASH mouse models were predominantly mononucleated polyploids, suggesting that the cell cycle mode is different from the cytokinesis failure that generates binucleated cells, as seen after weaning.
Further experiments with hepatocytes from NAFLD/NASH mice showed that oxidative stress, one of the major drivers of NASH, causes endoreduplication cycles by attenuation of CDK1 via ATR kinase signalling.
Collectively, hepatocytes increase their ploidy during the proliferative stress that underlies various disease conditions. Even though the mechanisms altering the cell cycle mode towards endoreduplication or failed cytokinesis vary between these aetiologies, ploidy increases are observed in liver disorders that cause compensatory liver growth. The role of polyploidy in the interrelation between inflammation, fibrosis, and hepatocarcinogenesis requires further investigation. Intriguingly, a recent study using liver-specific Cdk1 or Ccna2 deficient mice to model high mononuclear hepatic ploidy found that this condition was sufficient to cause DNA damage, liver inflammation and fibrosis, independent of any other trigger. Elevated ALT (alanine aminotransferase) levels, as well as increased cytokine production and the presence of Kupffer cells were observed in animals as young as 2 weeks old.
These reports suggest that mononuclear polyploidy, a state often referred to as “pathological polyploidy”, could by itself produce an inflammatory liver disease. Of note, the latter reports used a mouse model deficient in a major cell cycle regulator. Thus, future studies will have to carefully dissect whether mononucleation is a cause or consequence of the described pathologies.
In general, infection and/or inflammation-triggered compensatory proliferation may eventually be limited by the induction of a cellular senescence programme. Induction of senescence also depends on the p53/p21 axis and can eventually rewire hepatocyte metabolism, leading to steatosis. Accordingly, using senolytic drugs to remove senescent hepatocytes can revert NAFLD-related steatosis,
Mouse models show that elevated hepatocyte polyploidy is protective, while poorly polyploid livers develop spontaneous HCC.
Ploidy and liver cancer
Chronic liver disease is associated with ongoing hepatocyte injury, inflammation and fibrosis that may ultimately lead to cirrhosis (Kulik L, Gastroenterology 2019). Cirrhosis is an important pre-cancerogenic condition as most cases of HCC arise in cirrhotic livers, while non-cirrhotic HCC is rare – except in patients with HBV infection or with NASH.
The “proliferation HCC” subclass is characterised by p53 mutations and CIN, with patients often showing high AFP (alpha-fetoprotein) levels, higher tumour grades and worse clinical outcomes. In contrast, the “non-proliferation HCC” subclass often show mutations in CTNNB1 (beta-catenin) that also promote immune escape (i.e. a low level of immune cell infiltration), and are characterised by a lower tumour grade and better clinical outcomes.
Even though polyploidy is generally associated with cancer and shown to increase in a variety of diseases that promote tumorigenesis in the liver, several reports demonstrated that the interrelation of hepatocyte polyploidy and hepatocarcinogenesis is quite the opposite. Studies by the Zhu laboratory support the idea that higher polyploidy can protect from HCC in different carcinogen-driven models, which was explained by increasing the copy number of tumour suppressor genes.
The major role of the liver in detoxification constantly exposes hepatocytes to genotoxic compounds and metabolites. Hence, hepatocytes must have adapted to this condition since the frequency of spontaneous HCC without underlying cirrhosis is rather low.
There have been recent advances in our understanding of the role of hepatocyte ploidy in tumorigenesis. Studies have utilised mouse models in which the ploidy of hepatocytes was either reduced by loss of E2F7/8, or increased by premature weaning or the knock-down of the cytokinesis effector anillin.
It has been demonstrated that liver tumours arise mostly from poorly polyploid, presumably diploid hepatocytes (Fig. 5). Based on targeted sequencing analyses, the authors attributed these effects to the increased tumour suppressor gene copy numbers in polyploid cells, while in turn the low ploidy hepatocytes were prone to lose tumour suppressor gene copies. While tumour promoting driver mutations were similar in low- and high-ploidy lesions, high polyploidy was clearly protective against loss of heterozygosity of tumour suppressor genes such as p53.
In addition to low polyploidy, ploidy-reducing divisions of highly polyploid hepatocytes have also been proposed to drive tumorigenesis in the liver. This concept has been described in the “ploidy conveyor” model. Polyploid hepatocytes can form multipolar mitotic spindles, which were found to generate daughter cells of reduced ploidy in vitro.
On the other hand, loss of supernumerary centrosomes will inactivate the PIDDosome-p53 axis that would otherwise reduce the proliferative potential of polyploid hepatocytes, thus alleviating the competitive disadvantage of polyploids.
Hepatocyte and HCC tumour cell ploidy may be used for risk stratification, prognosis, and therapeutic strategies in clinics.
How multipolar divisions of hepatocytes can generate near-euploid, viable daughter cells is not clear. This could either be the result of an unidentified counting mechanism for the individual chromosomes or, more likely, multipolar divisions lead to random chromosome segregation within multiple daughter cells and only those with viable karyotypes survive and repopulate the liver. Another way could be the formation of 2 independent spindles in binucleated hepatocytes for faithful chromosome segregation, however, this is a rarely observed event.
Answering this question in detail will be essential to understand the role of hepatocyte ploidy reduction in hepatocarcinogenesis.
Whether mononucleated polyploids that are frequently observed in various liver diseases play a role in hepatocarcinogenesis remains unknown. Interestingly, serial transplantations in the Fah-/- mouse model selected for mononuclear polyploids that were less prone to ploidy-reducing divisions.
This is reminiscent of chronic liver diseases where mononucleated polyploid hepatocytes occur at a higher frequency, presumably also after compensatory regeneration that requires the cells to cycle several rounds.
Aside from the “model organism”, one possible explanation for this discrepancy is the proinflammatory disease environment in humans. Nonetheless, the impact of nuclear number on the progression of the disease and eventually HCC development requires further clarification.
The PIDDosome in liver cancer
Beside its role in physiological development and regeneration, we also explored the role of the PIDDosome in HCC, as the PIDDosome acts upstream of p53, which is mutated in about 30% of patients with HCC.
To investigate the impact of the PIDDosome pathway on liver cancer, we used the diethylnitrosamine (DEN) model of HCC in mice. Interestingly, we observed that PIDDosome-deficient mice developed significantly less tumour nodules and reduced overall tumour burden than wild-type mice, contrasting with our expectation that the loss of p53 activation after polyploidisation would favour tumourigenesis.
As there was no difference in tumour ploidy among the genotypes, it seems likely that the tumours develop from low ploidy cells and therefore higher hepatocyte ploidy hampers tumour development owing to a lower abundance of diploid/tetraploid cells.
Intriguingly, our data show that high polyploidy is protective even if the tumourigenic event occurs in a mostly diploid stage of the liver. This indicates that increased polyploidy can buffer secondary hits that are acquired later in life. The recent study by the Grompe lab mentioned earlier actually confirms this notion and provides evidence that cell divisions of polyploid hepatocytes leading to reduced ploidy are more error prone and create a pool of low ploidy cells that are more vulnerable to transformation.
As such, while high ploidy protects, division of polyploid hepatocytes that cause genome size reduction in regenerative settings or chronic inflammatory liver disease may promote malignant disease. This hypothesis is supported by observations made by Lin et al., who report DEN-driven hyperpolyploidisation of centrilobular hepatocytes that foster the formation of premalignant lesions. However, within these preneoplastic lesions, cells were found to be substantially smaller, suggesting that they undergo nuclear- and genome-content reduction during transformation. Blocking DEN-induced hyperpolyploidisation by inhibiting Aurora B kinase reduced tumour number and size, establishing causality.
It will be interesting to see if higher basal liver ploidy, as seen in PIDDosome-mutant mice may alter HCC risk in adequate model systems or correlate with HCC penetrance in humans.
Analysing the “Liver Hepatocellular Carcinoma” TCGA data set containing information on 372 HCC patient samples, we also noted that higher expression of CASP2 and PIDD1 are associated with more severe disease stages and with reduced recurrence-free survival.
The expression of these PIDDosome components was even higher in patients with mutated p53 and their increased expression was confirmed in primary HCC samples compared to neighbouring non-malignant liver tissue, albeit the p53 status of these patients remained unknown.
While this indicates that CASP2 or PIDD1 expression may be used as a prognostic marker of disease progression in HCC, this correlation may simply be a consequence of increased cellular proliferation rates that depend on E2F transcription factors that presumably also control the expression of Casp2 and PIDD1 in malignant cells.
Ploidy as a prognostic marker in patients with HCC
Recently, the ploidy state in human HCC has indeed been linked to patient survival. However, in contrast to studies in mouse models, patients with highly polyploid tumours and mutated p53 were found to have significantly decreased progression-free survival.
We investigated a large set of HCC specimens from patients with regards to cellular size (specifically cell density), as a measure for tumour cell ploidy, and found that high tumour cell ploidy was a surrogate of better prognosis.
Yet, while this study could not determine the p53 status of the included patients, the on first sight ‘different’ observations by Bou-Nader et al. and by Sladky et al. may still be compatible. Loss of p53 enables polyploid cancer cells to proliferate in the presence of extra centrosomes, driving evolution of aggressive disease, as observed by Bou-Nader and colleagues.
In the presence of functional p53, which reflects still around 70% of all HCC cases analysed, the protective effect may dominate. In our large cohort, this may dilute out the negative effects on survival, imposed by p53 mutation. It will be of interest to investigate if discriminating p53 status on top of tumour cell size will constitute a novel and simple prognostic pair of genetic and morphological markers in HCC. Follow-up studies including larger patient cohorts with detailed genetic information combined with morphometric analyses are needed (i) to confirm the degree of hepatocyte polyploidy as a risk factor for subsequent HCC development, and (ii) to assess if HCC tumour cell density/ploidy is of prognostic value in patients with HCC. Ultimately, all future studies on hepatic ploidy should consider different molecular subtypes of HCC, assess genetic loss/mutations of tumour promoter/suppressor genes, as well as consider age and liver disease aetiologies as potential modulators of hepatocyte ploidy state.
Despite all the efforts made to understand the role of liver ploidy in health and disease, there are still considerable controversies regarding its clinical relevance. The protective role of hepatocyte ploidy against development of cancer was, however, consistently reported by various groups using different tools and approaches to investigate its impact on liver cancer. This finding may thus be of prognostic value as low ploidy seems to predict HCC progression or recurrence after liver transplantation, especially when combined with the assessment of distinct genetic signatures. In addition, inhibition of molecular mechanisms that restrict ploidy may improve liver regeneration without negatively affecting overall liver function. However, the impact of hepatocyte ploidy on disease progression in different liver pathologies remains to be further explored, especially with regard to ploidy-reducing cell divisions that promote CIN and thus potentially hepatocarcinogenesis. Yet, pharmacological modulation of ploidy regulators, such as the PIDDosome, that facilitate polyploidisation may reduce the risk of liver cancer in patients with viral or metabolic liver disease (i.e. chemoprevention). Such strategies may also help to promote liver regeneration in liver failure, to precondition livers for extensive curative resections, or to allow successful bridging to liver transplantation.
ALD, alcohol-related liver disease; ASH, alcohol-related steatohepatitis; CASP2, caspase-2; CDK, cyclin-dependent kinase; CIN, chromosomal instability; DD, death domain; DEN, diethylnitrosamine; FAH, fumarylacetoacetate hydrolase; HCC, hepatocellular carcinoma; HGD, homogentisate 1,2 dioxygenase; miR, microRNA; NAFLD, non-alcoholic fatty liver disease; NASH, non-alcoholic steatohepatitis; PIDD1, p53-induced death domain protein 1; PLK1, Polo like kinase 1; RhoA, Ras homolog family member A; SKP2, S-phase kinase associated protein 2; YAP, yes-associated protein.
AV: ERC-AdG “POLICE” (# 787171 ), Ludwig Boltzmann Society . TR: Christian-Doppler Society , Ludwig Boltzmann society . VCS: EMBO Postdoctoral fellowship program ( ALTF 1194-2020 ).
VCS, FE, TR, AV wrote and edited the manuscript; FE created the figures; AV coordinated the work.
Conflict of Interest
TR received grant support from Abbvie, Boehringer-Ingelheim, Gilead, MSD, Philips Healthcare, Gore; speaking honoraria from Abbvie, Gilead, Gore, Intercept, Roche, MSD; consulting/advisory board fee from Abbvie, Bayer, Boehringer-Ingelheim, Gilead, Intercept, MSD, Siemens; and travel support from Abbvie, Boehringer-Ingelheim, Gilead and Roche. All other authors declare no conflict of interest.
Please refer to the accompanying ICMJE disclosure forms for further details.
We want to thank all members of our teams for fruitful discussion and Lauren Evans for help with editing the manuscript. We apologize to all those whose valuable contribution was not cited due to space constrains. Work in our laboratories is supported by the ERC-AdG “POLICE”, and the Ludwig Boltzmann Society. TR is supported by the Christian-Doppler Society. VCS acknowledges support by the EMBO Postdoctoral fellowship program (ALTF 1194-2020).