Essential role for centromeric factors following p53 loss and oncogenic transformation

Abstract

In mammals, centromere definition involves the histone variant CENP-A (centromere protein A), deposited by its chaperone, HJURP (Holliday junction recognition protein). Alterations in this process impair chromosome segregation and genome stability, which are also compromised by p53 inactivation in cancer. Here we found that CENP-A and HJURP are transcriptionally up-regulated in p53-null human tumors. Using an established mouse embryonic fibroblast (MEF) model combining p53 inactivation with E1A or HRas-V12 oncogene expression, we reproduced a similar up-regulation of HJURP and CENP-A. We delineate functional CDE/CHR motifs within the Hjurp and Cenpa promoters and demonstrate their roles in p53-mediated repression. To assess the importance of HJURP up-regulation in transformed murine and human cells, we used a CRISPR/Cas9 approach. Remarkably, depletion of HJURP leads to distinct outcomes depending on their p53 status. Functional p53 elicits a cell cycle arrest response, whereas, in p53-null transformed cells, the absence of arrest enables the loss of HJURP to induce severe aneuploidy and, ultimately, apoptotic cell death. We thus tested the impact of HJURP depletion in pre-established allograft tumors in mice and revealed a major block of tumor progression in vivo. We discuss a model in which an "epigenetic addiction" to the HJURP chaperone represents an Achilles' heel in p53-deficient transformed cells.

Shaping Chromatin in the Nucleus: The Bricks and the Architects

“…To determine whether this aspect of p53-mediated up-regulation involves other transcription factors, such as FoxM1 (a p53-partner), it will be necessary to extend experiments to new mutant mouse models (Barsotti and Prives 2009;Bieging et al 2014). Importantly, although CENP-A and HJURP levels are high in p53-deficient transformed cells, this elevation cannot be merely attributed to hyperproliferation (Filipescu et al 2017). To determine if an increased dose of a histone variant or its chaperone could drive transformation, we overproduced the two factors individually in mouse cells.…”

“…This finding indicates that increased CENP-A and HJURP levels are linked to loss of the transcriptional function of p53 in cancer cells. Interestingly, CENP-A and HJURP are overexpressed not only in cancers with p53-inactivating genetic lesions, but also in a subset of tumors with gainof-function mutations in p53 (Filipescu et al 2017). To determine whether this aspect of p53-mediated up-regulation involves other transcription factors, such as FoxM1 (a p53-partner), it will be necessary to extend experiments to new mutant mouse models (Barsotti and Prives 2009;Bieging et al 2014).…”

“…3B). We have generated a model of oncogene-transformed MEFs (mouse embryonic fibroblasts) that recapitulates the increase in CENP-A and HJURP levels (Filipescu et al 2017). Notably, in MEFs, altering the dose of the brick or its architect does not necessarily lead to the same outcome.…”

“…4B). In a reporter system, p53 can associate at the promoters of CENP-A and HJURP, as part of the E2F-DREAM repressive complex (Filipescu et al 2017). This finding indicates that increased CENP-A and HJURP levels are linked to loss of the transcriptional function of p53 in cancer cells.…”

“…Here, we summarize recent work suggesting that therapies targeted against specific histone chaperones and their "client" variants represent a promising opportunity for improved cancer care. In addition, we highlight how dynamics of CENP-A in cancer (Lacoste et al 2014;Filipescu et al 2017) and H3.3 in development can open new avenues for fundamental and translational research in the future.…”