Dr. David Tosh, University of Bath, on 22 November 2002

Dr. Tosh described some recent work on cellular plasticity, or how cells differentiate and turn into one or another organ. For example, the bone marrow cells can be made to convert themselves into many other cell types, including parts of the brain, liver, skeletal muscle and heart muscle. They circulate freely in bloodstream and seem to recognise a damaged organ and to some extent can repair the damage by becoming a cell suitable for the repair, a process known as trans-differentiation. The question is whether other cell types have a similar capability, and Dr. Tosh's specific interest is in pancreas and liver cells, which are known to develop next to each other from the endoderm cells of the embryo, and so may share many characteristics and gene expressions.

The pancreas has a number of functions, especially the production of various digestive enzymes (the exocrine function) and some proteins, for example insulin (the endocrine function). Part of it consists of the Islets of Langerhans, named in the late 1800s and which are identified as the site where insulin is produced, and become disabled in diabetes mellitus. Diabetes was first clinically described by Aretus of Cappadocia around 200-300AD, and is characterised by polyphagia (over eating), polydypsia (excessive thirst) and polyuria (excessive excretion of urine). The normal metabolism uses sugars, like glucose, to obtain energy, but must first get the hydrophilic sugar through the fatty wall of a cell. Insulin enables this transport to take place; and in its absence, glucose levels remain high in the blood, which triggers the kidneys to excrete sugar in the urine. If any other cell types in the body could be induced to manufacture insulin, then this could help in treating the disease.

The liver is an organ that can regenerate. Curiously, the myth of Prometheus was that Zeus punished him for passing fire on to humans by chaining him where Zeus visited him in the guise of an eagle to tear at his liver during the day, and it grew back again each night.

The liver produces many sorts of proteins, among them clotting agents for the blood. It can also respond to insulin by storing it for later use. but one of its primary functions is general detoxification of the blood stream of various by-products of normal metabolism, like bilirubin, which must be removed before it can collect as yellow deposits in different cells. Bilirubin, the result of recycling the iron in haemoglobin, is responsible for the yellow colour of jaundice and is normally excreted in the bile by the liver.

Why try to convert pancreas cells to liver cells? There are three principal reasons.

First, as the two organs are derived from the embryo's endoderm at adjacent sites, they are probably derived from the same cells, and so determining their difference may shed light on the normal course of development. If that is understood, then it may be possible to convert cells in the adult too.

Secondly, it will help us to understand some diseases, like Barrett's oesophagus in which some cells in the oesophagus have converted to an intestinal type.

Thirdly, we may then be able to treat several diseases, like neuro-degenerative disorders, diabetes, and liver disorders, and overcome organ rejection by using the patient's own cells, converting the cell type, and re-introducing them into a suitable place, or even converting them in situ.
When working with these cells, it is important to know what type they are, and this is done by creating (or buying ready made) antibodies to the protein products of a cell, and then further antibodies to those antibodies. Then by looking for cells containing these antibodies, we can deduce the presence of the proteins, and hence that the cells of probably of a particular type.

Recent work was done on a cell strain known as B13, a cancerous and apparently immortal strain from a pancreas and which is now 20 years old. It was known that treating these with a hormone-like substance, dexamethasone (dex) made them change shape. On further analysis, it was discovered that if the B13 cells were cultured for 2 weeks, then a significant number were shown no longer to contain amylase, a typical pancreas constituent, and instead were producing albumin, a typical liver protein, indicating that some conversion had taken place.

Further work has identified one gene, called in the talk `gene A' that is involved in distinguishing pancreas and liver. Gene A is not active in the pancreatic cells but is active in the liver cells. Somehow the dex has turned on Gene A in the pancreatic cells, but the mechanism is not yet understood.

Since that work was done on a culture of cells, a test was conducted on normal pancreatic cells; and indeed, adding dex produced liver cells. In addition, more liver proteins were tested for, and all 6 types of liver cells could be found in the resulting mix of cells, although they were not organised, as a real liver would be.

Can we convert from liver to pancreas? Dr. Tosh and colleagues have found a way to do this and will report their findings in a paper to be published in February.

What of the future? Some of the questions that researchers would like to address are:

- Can we take any type of cell and convert it to any other type?

- Can we reconstitute a liver?

- Is it possible to show the same effect in humans?

- Can we cure diseases?


During the lively questions after the talk, some further details were given. It seems that it is not necessary for cells to reproduce before they convert their type, which was a great surprise. And after the dex is withdrawn, the liver cells appear to stay as liver cells and do not revert, even though they are in contact with normal pancreatic cells.

If we do manage to create insulin-producing cells in a liver there are other problems to overcome before we could say it was useful. For example, how would we ensure that the insulin was generated where it could be sent into the body as efficiently as it is from the pancreas? We might find that `'the plumbing is wrong''.

Andy Pepperdine