The gene FBW7 is a tumour suppressor, that is, it protects cells from carcinogenic changes. It is known to be inactivated in several tumour types, particularly in T-cell acute lymphoblastic leukaemia (T-ALL) and in breast and colon cancer. The fact that selective deletion of the equivalent gene Fbw7 in mouse T-cells gives rise to T-ALL provides further evidence for the tumour suppressor role of this gene. The protein product of this gene is known to form part of a multi-component enzyme, ubiquitin ligase, which tags proteins that are targeted for degradation with a small protein, ubiquitin, after which those proteins are broken down into small peptides by proteasomes. Although the proteins targeted by this ligase include some that are involved in the cell cycle and linked to cancer, the precise mechanism through which FBW7 protects cells from carcinogenic transformation is so far mostly unknown.  Significant light has now been shed on this mechanism by two studies published back-to-back in the 3 March 2011 issue of the journal Nature.

The first of these was led by Wenyi Wei from Harvard Medical School, Boston, Massachusetts, USA with collaborators from the Dana-Farber Cancer Institute, also in Boston, and from Kyushu University, Fukoka, Japan and the Jackson Laboratory, Bar Harbor, Maine, USA [1]. Wei and co-workers recognised that the ubiquitin ligase complex that includes FBW7, SCF^FBW7, is likely to be involved in the regulation of a protein named MCL1, a member of the BCL2 family of proteins that regulate cell survival and apoptosis, and that is particularly unstable. The researchers tested the theory that FBW7 tags MCL1 for degradation by the proteasome in response to the latter protein’s phosphorylation by glycogen synthase kinase 3 (GSK3). They found that depletion of FBW7 and of certain other components of the ubiquitin ligase complex, but not all, resulted in a significant increase in cellular concentrations of MCL1. There was, however, no increase in MCL1 mRNA levels, which is consistent with the theory that this regulation takes place after protein translation. 

The researchers then used mass spectroscopy to investigate the phosphorylation steps that are needed for FBW7 to recognise MCL1 and tag it for proteasomal destruction. This protein was found to be phosphorylated at many sites in vivo. Inactivation of four of these phosphorylation sites, either by mutating the amino acids – three serines and one threonine – into ones that cannot be phosphorylated, or by inhibiting the kinase GSK3, caused increases in MCL1 concentrations. Furthermore, co-expression of GSK3 and FBW7 with wild type MCL1 caused a reduction in the half-life of MCL1 that was not observed with mutant protein. Taken together, these results suggest that the hypothesis that FBW7 is a necessary component for the tagging and degradation of MCL1 that has been phosphorylated by GSK3 is correct.

Some, but not all, lines of human T-ALL cells contain FBW7 protein that has been inactivated by mutations and are therefore unable to degrade MCL1. Cells with defective FBW7 were found to be particularly sensitive to the multi-kinase inhibitor sorafenib, which represses MCL1 levels, but resistant to the BCL2 antagonist ABT-737, which does not. Both restoration of wild type FBW7 and depletion of MCL1 at the genetic level were found to restore sensitivity to ABT-737 in these cells. Wei and co-workers concluded that drugs that reduce the expression or activity of MCL1 would be likely to be effective in treating the significant proportion of T-ALL cases in which the protein FBW7 is defective in the leukaemia cells.

The second study, by Irene Wertz and colleagues at Genenetch, San Francisco, USA with collaborators at Abbott Laboratories, Illinois, USA, and at the Walter and Eliza Hall Institute of Medical Research, Parkville, Victoria, Australia, investigated the role of MCL1 and FBW7 in regulating the sensitivity of tumour cells to anti-cancer drugs such as paclitaxel (Taxol^®). These anti-tubulin drugs bind to, and stabilise, the microtubules that are formed during mitosis, which stops mitosis and triggers apoptosis. Aberrations in the expression pattern of proteins in the BCL2 family, including MCL1, can alter this response in many cell types.

Wertz and her co-workers observed that mouse embryonic fibroblasts (MEFs) with deletions in the pro-apoptotic BCL2 genes Bax and Bak were resistant to apoptosis induced by paclitaxel and other anti-tubulin drugs. Cells with MCL1 deletions, however, were more sensitive to these drugs than wild type cells, thus confirming the critical importance of the ratio of pro- to anti-apoptotic BCL2 proteins in determining cells’ response to apoptotic signals. The researchers then monitored the levels of these proteins and the associated mRNAs during drug-induced mitotic arrest and discovered that the concentration of MCL1 protein but not its mRNA decreased significantly. This suggested that MCL1 concentration is regulated post-translationally through the ubiquitin-proteasome system, a hypothesis that was strengthened by the observations that MCL1 was ubiquitinylated and that its concentration was increased in the presence of proteasome inhibitors. MCL1 degradation was also decreased in cells in which levels of the protein FBW7 had been knocked down using RNA interference (RNAi).

The researchers then used mass spectroscopy to evaluate the phosphorylation status of MCL1 in the same cells during mitotic arrest, and identified the same four phosphorylated serine and threonine residues as in the study by Wei and co-workers. MCL1 mutants in which several of these residues had been replaced by alanine were degraded less efficiently than wild type MCL1. Several kinases other than the previously identified GSK3 were found to be able to phosphorylate MCL1 at these sites and therefore regulate its binding to FBW7 and subsequent degradation.

The researchers then showed that human colon and ovarian tumour cell lines with mutated FBW7 contained higher concentrations of MCL1 in mitotic arrest than similar cells with wild type FBW7, and that they, like the mouse fibroblasts, were resistant to anti-tubulin drugs. These cells evade chemotherapy-induced apoptosis and may duplicate their DNA without dividing, becoming polyploid. This evasion of apoptosis is likely to prove as important a mechanism of resistance to antitubulin drugs as cell efflux or mutations in tubulin itself, and monitoring FBW7 activity in tumour cells could prove useful for predicting response to drugs like paclitaxel.

Both these studies, therefore, provide important insights into the mechanism of action of the tumour suppressor FBW7 as a promoter of apoptosis via degradation of MCL1. This protein may become an important biomarker for both targeting treatment and predicting drug response in a number of tumour types.

References

1: Inuzuka, H., Shaik, S., Onoyama, I. and 17 others (2011). SCF^FBW7 regulates cellular apoptosis by targeting MCL1 for ubiquitylation and destruction. Nature 471, 104-109.  doi:10.1038/nature09732

2: Wertz, I.E., Kusam, S., Lam, C. and 24 others (2011). Sensitivity to antitubulin chemotherapeutics is regulated by MCL1 and FBW7. Nature 471, 110-114. doi:10.1038/nature09779