There are several different ways to stop cancer-cell proliferation. The most obvious approach is to induce apoptosis, and many chemotherapeutic agents are known to do just that. Less is known, however, about how to induce senescence — a state of terminal cell-cycle arrest — in cancer cells.

In normal cells, replicative senescence results from loss of telomeric repeats after several rounds of cell division. This generates a DNA-damage signal that activates p53. Some anticancer drugs induce cell senescence by inducing this DNA-damage response in both normal and malignant cells, but these drugs can lead to unwanted side effects.

Dimitri Lodygin et al. therefore looked for signalling pathways that were specifically downregulated during replicative senescence. They reasoned that pharmacological inhibition of these pathways should induce senescence in tumour cells. Using microarray analysis, they found that several components of the cGMP signalling pathway were downregulated during replicative senescence of primary human diploid fibroblasts. So could a compound such as 6-anilino-5,8-quinolinequinone (Ly83583, or LY), which is known to inhibit cGMP production, induce senescence?

Lodygin et al. showed that treatment of fibroblasts with LY completely and irreversibly inhibited proliferation of the cells by blocking S-phase entry. Microarray analysis of LY-treated cells revealed a significant overlap between the genetic profiles of cells during replicative senescence and after LY treatment. One gene that was induced under both conditions was CDKN1A , which encodes an inhibitor of cyclin-dependent kinases — called WAF1 (also known as p21). LY also induced WAF1 expression and prevented proliferation in colorectal cancer, breast cancer and melanoma cell lines, indicating its therapeutic potential.

Experiments in cGMP-null cell lines revealed that LY's effects depended on induction of p21. LY did not, however, activate the DNA-damage/p53 pathway, as LY treatment of p53-deficient cells still resulted in p21 upregulation and senescence. It is not clear whether LY upregulation of p21 occurs through its effects on the cGMP signalling pathway, or through some other mechanism.

As LY induces cell-cycle arrest in a p53-independent manner, it might be useful in treating tumours that have p53 mutations. Inactivation of p21, on the other hand, has not been reported in cancer cells. In cells with disruptions in the RB pathway, LY induced apoptosis, rather than cell-cycle arrest, so LY could also be used to treat tumours that have RB defects. The ability of LY to induce senescence without inducing DNA damage indicates that it might also have fewer side effects than current chemotherapeutic agents.