Look both ways

The Krüppel-like factor 4 (KLF4) transcription factor can function as a tumour suppressor in various cancer types, and as an oncogene in others, including breast cancer. In their Nature Cell Biology paper, Daniel Peeper and colleagues have identified some of the molecular mechanisms that explain the Janus-like behaviour of KLF4 in tumorigenesis.

When the authors identified KLF4 as a protein that can bypass cellular senescence induced by oncogenic RAS (RAS^V12), they decided to investigate the molecular mechanisms that are at work. They established that KLF4 represses expression of the tumour suppressor p53, but induces expression of the cyclin-dependent-kinase inhibitor p21 (CDKN1A). p53 is a crucial mediator of RAS^V12-induced proliferative arrest, so the authors hypothesized that suppression of p53 by KLF4 is probably the means through which KLF4 bypasses senescence. This was confirmed when short hairpin RNAs were used to suppress p53 to levels similar to those achieved by expression of KLF4, and this prohibited RAS^V12-induced senescence.

How does KLF4 suppress the expression of p53? Northern blots demonstrated that KLF4 downregulates TP53 mRNA levels, and chromatin immunoprecipitation assays indicated that KLF4 achieves this by directly binding to the TP53 promoter.

Although KLF4-mediated repression of p53 explains how KLF4 prevents RAS^V12-driven arrest, it does not explain the reverse: how RAS^V12 bypasses KLF4-induced arrest. Indeed, KLF4, similar to most other tumour suppressors, normally triggers cell-cycle arrest. One of the known mitogenic targets of RAS is cyclin D1, and Peeper and colleagues found that RAS^V12 could not collaborate with KLF4 in stimulating cell proliferation in the absence of cyclin D1 expression, indicating a crucial role for cyclin D1 in this setting.

How does this tie in with the expression of p21, which KLF4 induces? Cyclin D1 neutralizes the p21-induced cell-cycle arrest, so the cells are able to proliferate. In agreement with this hypothesis, p21-null mouse embryonic fibroblasts were unable to arrest in the presence of KLF4. Together, these data indicate that KLF4 is a tumour suppressor that induces cell-cycle arrest by increasing the expression level of p21.

The authors also examined whether the KLF4-mediated repression of p53 could explain the oncogenic behaviour of KLF4 in breast cancer cells. They found that suppressing KLF4 in human breast cancer cell lines restored the expression of p53, which resulted in apoptosis. So, suppression of p53 by KLF4 seems to be important for the survival of breast cancer cells, and this probably explains the oncogenic effect of KLF4 in these cells.

The authors conclude that the systemic targeting of KLF4 would not be desirable, as it would inhibit both the oncogenic and the tumour-suppressor functions of KLF4.