∆133p53 isoform promotes tumour invasion and metastasis via interleukin-6 activation of JAK-STAT and RhoA-ROCK signalling

Abstract

∆122p53 mice (a model of ∆133p53 isoform) are tumour-prone, have extensive inflammation and elevated serum IL-6. To investigate the role of IL-6 we crossed ∆122p53 mice with IL-6 null mice. Here we show that loss of IL-6 reduced JAK-STAT signalling, tumour incidence and metastasis. We also show that ∆122p53 activates RhoA-ROCK signalling leading to tumour cell invasion, which is IL-6-dependent and can be reduced by inhibition of JAK-STAT and RhoA-ROCK pathways. Similarly, we show that Δ133p53 activates these pathways, resulting in invasive and migratory phenotypes in colorectal cancer cells. Gene expression analysis of colorectal tumours showed enrichment of GPCR signalling associated with ∆133TP53 mRNA. Patients with elevated ∆133TP53 mRNA levels had a shorter disease-free survival. Our results suggest that ∆133p53 promotes tumour invasion by activation of the JAK-STAT and RhoA-ROCK pathways, and that patients whose tumours have high ∆133TP53 may benefit from therapies targeting these pathways.

The Emerging Landscape of p53 Isoforms in Physiology, Cancer and Degenerative Diseases

“…Increased ∆133p53α isoform expression, in the context of WT TP53, has been associated with poorer disease-free survival in patients with colorectal cancer. A possible mechanism explaining this association is ∆133p53α isoform mediated tumour cell invasion via increased IL-6 expression activating JAK-STAT3 and RhoA-ROCK pathways [77]. Interestingly, in advanced serous ovarian cancer tissue with mutant TP53 gene, ∆133p53α expression was associated with improved disease-free survival and overall survival, providing preliminary evidence that the mutational status of TP53 can influence the association between p53 isoform expression and patients' clinical outcome.…”

Accessing a New Dimension in TP53 Biology: Multiplex Long Amplicon Digital PCR to Specifically Detect and Quantitate Individual TP53 Transcripts

“…Analysis of IIICF/c cells using our long amplicon ddPCR assays clearly detected two t1 transcripts, which was then confirmed by performing splice-aware mapping of the RNA sequencing data. Furthermore, our assays showed that this TP53 splicing mutation led to expression of very high levels of transcripts encoding all ∆133/∆160p53 isoforms, and given the association of some of the ∆133p53 isoforms with invasion and metastasis [12,33], may explain why this mutation is pathogenic.…”

“…They share five central exons in common, which span 618 bp, therefore to determine which 5' end and 3' end are part of the same transcript requires amplification of >618 bp. Most PCR-based assays of TP53 transcript abundance, including our own work, have focused on quantifying the individual transcript ends [8][9][10][11][12][13]. These studies have shown that the levels of certain 5' or 3' ends are significantly associated with patient prognosis in a number of cancer types, presumably reflecting the biological functions of the p53 protein sequences encoded by these alternatively-spliced RNA ends.…”

“…These studies have shown that the levels of certain 5' or 3' ends are significantly associated with patient prognosis in a number of cancer types, presumably reflecting the biological functions of the p53 protein sequences encoded by these alternatively-spliced RNA ends. For example, glioblastoma, colorectal or prostate cancer patients with the highest levels of the ∆133TP53 5' end or the TP53α 3' end have relatively poor prognosis [11][12][13]. In contrast, low levels of TP53β and TP53γ 3' ends have been associated with poor prognosis in breast cancer [9,10].…”

“…However, until now, investigators have been uncertain which TP53 5' end might be associated with which 3' end in any one tumour or cell line, and so are unable to identify specifically which TP53 transcripts have contributed to these clinical associations. Recently, correlations between TP53 transcript ends have been used to infer (but without certainty) the abundance of specific TP53 transcripts [11,12,14,15]. However, this method only provides estimates at best, and if there is low correlation, inferring which end is associated with which becomes impossible.…”