Research > Liver Regeneration


The liver is almost unique amongst the tissues of the body in its capacity for regeneration. The ability of the liver to regenerate after resection has been known since the late 1800's and it is now recognised that hepatocytes have an apparently unlimited capacity for division. Although the hepatocytes of adult liver divide only rarely under normal conditions (approximately 1 hepatocyte in 20,000), the loss of liver functional mass (by surgical removal or following chemical or viral insult), leads to rapid proliferation and restoration of functional liver tissue. 24 hours after a 70% partial hepatectomy in the rat, greater than 90% of the hepatocytes in the remnant 30% are in the process of dividing and the liver mass if fully restored some 7 to 10 days later.

During regeneration the different liver cell types do not divide together. In rats the hepatocytes divide first at about 24 h post hepatectomy, followed by the biliary ductular cells, then the Kupffer cells and Stellate cells and finally the endothelial cells. The precise timing depends, at least in part, on the age of the animal (the process is longer and more protracted in older animals) and the magnitude of the response increases with the severity of the hepatectomy.

It is generally assumed that the regenerative process in man is the same as that in rodents. However, even rats and mice differ in their temporal responses and after hepatectomy the hepatocytes of young rats show a peak of DNA synthesis 12-16 h before those of young mice and undergo division some 24 h earlier. Thus, in even closely related species differences in the timing of events do occur. Interestingly, the timing of hepatocyte entry into DNA synthesis following partial hepatectomy appears to be cell autonomous.

Although the precise timing of events in man is currently much less clear, the available evidence suggests there is rapid regeneration during the first two weeks after major hepatectomy, followed by a more gradual increase in size which ceases some 6 to 12 month later. Whether regeneration is intrinsically slower in humans than in rats, or is a reflection of the relatively older population in the human studies, is not clear.

Following resection for the removal of liver metastases in patients admitted to this unit, the remnant liver undergoes regeneration. We are able to obtain samples of liver from consenting patients undergoing liver resections at the time of operation and are thus in a position to investigate some of the very early events associated with the regenerative process in man.

The Early Events in Liver Regeneration

Among the earliest documented events following hepatectomy in rats are increases in uPA (urokinase-like plasminogen activator) activity and uPAR (uPA receptor). Increased uPA activity is detected at 1 minute post hepatectomy, and continues to increase for at least 60 min. Increased uPAR (as detected on Western blots) is also seen at 1 min post hepatectomy, and more clearly at 1 hour. These observations suggest uPA is a key initiator of the metalloproteinase cascade leading to matrix degradation.

Further evidence that these changes in uPA and extracellular remodelling are of significance comes from experiments with rat hepatocytes cultured in serum-free medium, and from studies of uPA-deficient mice (uPA-/-) and from transgenic mice overexpressing uPA.

Collectively these changes are associated with, or initiate alterations in the extracellular matrix which permit the subsequent division firstly of the hepatocytes and then the other liver cell types. This is considered to be part of the priming phase of the regenerative response, necessary to allow subsequent increases in growth factors to initiate hepatocyte cell division but is not of itself sufficient to produce hepatocyte proliferation. The proliferation step requires a combination of cytokines and growth factors.

Early changes have also been recently described in the Wnt/beta-catenin pathway. Transient increases in hepatocyte levels of b-catenin and increased translocation to the nucleus occur during the first 5 min after hepatectomy in the rat. The increase in beta-catenin levels results from a decreased rate of beta-catenin degradation. Shortly afterwards there is active b-catenin breakdown to below normal levels which do not return to normal until 48 to 72 h. This early modulation of beta-catenin may be important in the regenerative process since within the nucleus it can act as a transcription factor affecting uPAR, cyclin D1 and c-myc expression, all of which have important potential. Determination of the precise role of beta-catenin in liver regeneration should be enlightening.

Activation of immediate-early genes also occurs shortly after hepatectomy, with transiently increased expression of c-fos, c-jun and c-myc within the first 3 h. allowing a subsequent increase in the AP-1 binding as c-Jun protein and c-Jun nuclear kinase ( JNK) increase.

Clearly this early stage in the regenerative process is crucial to orchestrating subsequent events, and the transient nature of many of the changes observed so far suggest a series of biological switches are thrown which initiate and commit the liver to regenerate. A more detailed review of the molecular physiology of liver regeneration has recently been published by our group.

Potential Clinical Applications and Relevance for the Surgeon

One of the interesting features of the work from rodents is that the hepatic regenerative signals are not confined to the liver. Both isolated hepatocytes and small liver explants introduced into sites well away from the liver show mitotic responses following partial hepatectomy. The effect of hepatectomy on any residual tumour left in the liver remains controversial.

Liver regeneration has a particular relevance to surgeons who undertake major hepatic resections. Recent introduction of split liver transplantation and live related liver donor transplantation has raised the problem of the small-for-size syndrome. These patients suffer from profound liver insufficiency because of the small volume of residual or transplanted liver tissue.

One strategy to increase the volume of residual liver (and thus overcome post-operative liver insufficiency) is selective right portal vein embolisation prior to major hepatic resection and may provide an excellent in-vivo model for studying hepatic growth factors. In a recent study, tumours within the liver underwent enhanced growth after portal vein embolisation again suggesting the mitogenecity of hepatic growth factors.

In addition, an increased understanding of the liver regeneration cascade in humans could lead to improved therapies for the treatment of acute or chronic liver pathologies, where the ability to specifically stimulate liver cells would be valuable. In fulminant liver failure for example, (which has a 80% mortality unless transplantation is performed), the ability to stimulate the remaining viable liver cells to divide would be potentially life saving. Similarly, in end-stage liver disease, increasing the functional liver mass by stimulating hepatocyte growth may improve the well being of these patients, possibly to the extent that transplant surgery could be avoided. To develop such therapies requires a better understanding of the regeneration cascade in humans.

Our Studies

Work from this Unit has focussed so far on changes in uPA activity in membrane fractions prepared from liver samples obtained at operation. Increased uPA activity in the remnant is seen in patients in whom the resection size is estimated to be a least 50%, but not with smaller resections. Unlike the rodent experiments however, we have not seen increases in uPA as early as 1 minute post resection and in our group of subjects the earliest time that we have seen increased uPA is 15 min post resection. Thus in both rodents and humans, large liver resection does appear to be related to early rises in remnant liver uPA activity.

Our future studies will be directed towards other aspects of the early changes in liver regeneration, particularly those involving the phosphorylation state of c-met ( the receptor for hepatocyte growth factor ); the association of c-met with uPA and the association of c-met with beta-catenin.