by ecancer reporter Clare Sansom

Chemotherapy drugs and activated oncogenes can both induce a state of growth arrest in viable tumour cells.

This state, which is characterised by a specific histone methylation that blocks entry of the cells into the S phase of the cell cycle, is termed cellular senescence.

When this senescence is induced by drugs – a state known as therapy induced senescence (TIS) – tumour growth will be blocked, but these cells themselves have potentially toxic properties that make their removal a priority for further treatment.

A group of researchers led by Clemens Schmitt of Charité – Universitätsmedizin, Berlin, Germany has now used a mouse model of B-cell lymphoma to investigate differences in metabolism between senescent and non-senescent tumour cells, and thus to demonstrate how these cells can be selectively targeted.

In this lymphoma model, TIS is dependent on the presence of a H3K9 histone methyltransferase known as Suv39h1.

Lymphoma cells with and without the expression of functional Suv39h1 (Suv39h1 and Suv39h1- respectively) were injected into transgenic mice and treated with the DNA-damaging anti-cancer drug cyclophosphamide once the mouse lymph nodes had enlarged.

After five days of treatment, Suv39h1 but not Suv39h1- lymphoma cells became uniformly senescent.

Senescent lymphoma cells were stalled in the cell cycle at entry into S phase, and showed increased activity of senescence-associated beta-galactosidase (SA-β-gal) and reduced cell growth when visualised using [18F] fluorothymidine positron emission tomography (PET).

Interestingly, glucose levels, glucose uptake and ATP production increased in Suv39h1 (also known as senescence competent) lymphoma cells after chemotherapy; this effect was not observed in Suv39h1- (senescence incompetent) cells.

This indicates that it can be possible to detect the senescent phenotype non-invasively, using PET imaging.

The researchers next investigated the metabolic alterations associated with therapy induced senescence in these lymphoma cells.

Despite the fact that the cells were not dividing, they were found to metabolise glucose rapidly to pyruvate, lactate and citrate, and to activate a protein kinase that is associated with energy use.
These results are consistent with an enhancement of the so-called Warburg effect – an increase in non-oxidative glucose metabolism often observed in tumours – in senescent cells in particular.

This increased level of glycolysis in senescent cells was associated with increased levels of one isoform of pyruvate kinase, high oxygen consumption and high levels of intracellular ATP.

An increase in energy production is the characteristic feature of this phenotype, which can therefore be described as ‘hypermetabolic’.

Senescent cells were found to be particularly susceptible to genetic or pharmacological inhibition of glucose transport, showing that these cells depend on the hypermetabolic phenotype for survival.

Protein synthesis, and in particular the production of pro-inflammatory cytokines previously associated with senescence, was also shown to increase in senescent cells: an increase in the secretion of these has been termed the senescence-associated secretory phenotype (SASP).

Schmitt and his co-workers proposed that the production of these cytokines leads to an increase in proteotoxic stress, and that this in turn leads to an increase in energy use as the cells attempt to tag and degrade these unwanted proteins through autophagy.

The researchers exposed non-senescent and senescent lymphoma cells to drugs that block protein degradation, including lysosomal ATPase and protease inhibitors, and found that these inhibitors were far more toxic to the senescent cells.

These drugs induced apoptosis in the senescent cells through induction of caspase-12 and caspase-3.

Taken together, these results indicated that TIS cells are uniquely dependent on efficient protein degradation through lysosomes, and that drugs that target this pathway might be able to selectively destroy these cells.

Accordingly, Schmitt and his co-workers evaluated a number of human cancer cell lines for the TIS phenotype and found that TIS-capable cell lines were more sensitive to drugs such as bafilomycin A1, which inhibits lysosomal V-ATPases.

This enhanced vulnerability of therapy-induced senescent tumour cells to drugs that block lysosomal protein degradation or autophagy represents not only a way of targeting these harmful cells but a potential ‘Achilles’ heel’ through which a tumour can itself be targeted.    

Reference

Dörr, J.R., Yu, Y., Milanovic, M. and 21 others (2013). Synthetic lethal metabolic targeting of cellular senescence in cancer therapy. Nature, published online ahead of print 15 August 2013. doi:10.1038/nature12437