Complex interplay between p53 and chromosome stability

ABSTRACT

TP53-dependent cell cycle arrest has been proposed to limit the proliferation of aneuploid cells. We investigated the cellular response to aneuploidy in cell lines and organoid cultures and found that TP53 (also known as p53) is not activated following aneuploidy induction in organoids. However, we confirmed that p53 is required for high mitotic fidelity. Our findings provide a revised view on how p53 safeguards against aneuploidy.

Introduction

Aneuploid cells are far more prevalent in tumors than normal tissues. The tumor suppressor protein p53 (TP53 in humans and Trp53 in mice, best known as p53) is an important barrier to chromosomal instability as p53 mutation is a near universal feature of highly aneuploid tumors including ovarian and colitis-associated cancers.¹ p53 could limit aneuploidy either by inhibiting the survival or growth of aneuploid cells or by preventing mitotic errors that result in chromosome gains and losses. p53 has been proposed to play a surveillance role against aneuploid cells. It was shown that aneuploidy induction activated p53 leading to cyclin-dependent kinase inhibitor 1 (p21)-mediated G1 cell cycle arrest in human cell lines.² However, it was unclear if there was a universal aneuploidy-induced signal that activated p53, or if this response was context dependent. In our recently published work,³ we investigated p53 activation and cell cycle progression after aneuploidy induction in various cell lines and organoid model systems.

Adherent cells, but not suspension cells or organoids, undergo p53-dependent cell cycle arrest after aneuploidy induction

To broadly investigate the cellular response to aneuploidy, we examined 2D adherent cell lines (RPE1, HCT116), suspension cells (Nalm6), and 3D organotypic cultures (human mammary organoids [hMO], mouse colon organoids [mCO], neural progenitor cells [NPC]). By disrupting the spindle assembly checkpoint (SAC) using NMS-P715, a monopolar spindle 1 (MPS1) kinase inhibitor, we induced aneuploidy in 45–60% of cells in each cellular system. p53 and p21 protein abundance increased after aneuploidy induction in RPE1, HCT116, and Nalm6 cells. However, activation of the p53 pathway was not seen in hMO, mCO, or NPC. Accordingly, 5-ethynyl-2ʹ-deoxyuridine (EdU) incorporation and cell cycle analysis supported p53-dependent cell cycle arrest in RPE1 and HCT116 cells but not organotypic cultures. Nalm6 suspension cells arrested in G1 phase after aneuploidy induction, but this arrest was not p53 dependent suggesting that alternative pathway(s) exist to limit aneuploidy in these cells. We are currently pursuing a genome wide screen to uncover p53-independent aneuploidy surveillance mechanisms in blood cells. To explore if cellular architecture may underlie the differential responses to aneuploidy, we grew HCT116 cells as 3D spheres. In 3D, they exhibited increased p53 and p21 protein abundance and cell cycle arrest after aneuploidy induction suggesting that cellular architecture was not solely responsible for the difference in p53 activation between cell lines and organoids. We also attempted to culture mouse colon organoids in 2D, but constitutive stress associated with this condition activated p53 even without aneuploidy induction. Our results demonstrate the cellular response to aneuploidy is variable and p53 is not activated in all contexts (Figure 1).

Figure 1.

Response to aneuploidy in cultured cell lines and organoids. Aneuploidy was induced by treating cells with a monopolar spindle kinase 1 inhibitor (MPS1i) for 24 hours and cellular responses were assayed 24 hours after drug washout. Aneuploidy induction led to TP53 (best known as p53) activation and p53-dependent cell cycle arrest in 2D adherent cell lines. In suspension cells, aneuploidy induction activated p53, but the cell cycle arrest was p53-independent. Aneuploidy induction did not result in p53 activation in organoids and no cell cycle arrest was observed

p53 loss causes frequent mitotic aberrations

Around 40% of cells in mCO generated from Trp53^-/- mice, compared to less than 20% in Trp53^+/+ mCO, were aneuploid. This indicated that p53 plays a role in regulating the frequency of aneuploidy and may help prevent the emergence of aneuploid cells by ensuring mitotic fidelity. To test this hypothesis, we labeled mCO with histone H2B-mNeon and performed live cell imaging of mitosis. Trp53^-/- cells had increased lagging chromosomes and multipolar divisions, two types of mitotic errors known to lead to aneuploidy. Activation of the SAC by low-dose nocodazole reduced the frequency of lagging chromosomes in Trp53^-/- mCO, suggesting that lengthening mitotic duration facilitates error correction in the absence of p53. We determined that tetraploidization occurred prior to multipolar divisions in Trp53^-/- mCO. Multipolar division was not an underlying cause of lagging chromosomes, as they were observed in diploid cells undergoing bipolar division. We did observe centrosome amplification in Trp53^-/- mCO; however, extra centrosomes were associated with tetraploidization and are thus unlikely to be the cause of lagging chromosomes in diploid cells.

Discussion and perspective

In contrast to cell lines, aneuploidy induction did not activate p53 or lead to cell cycle arrest in organoids. Organoids do not require artificial immortalization and recapitulate many important features of in vivo epithelium including 3D cellular organization and extracellular matrix interactions.⁴ It is therefore possible that the aneuploidy response observed in organoids is more representative of the response in tissues and an aneuploidy-associated stress⁵ synergizes with another stress present in cell lines to activate p53. For example, hypo-osmotic stress was recently shown to be associated with aneuploidy due to proteome imbalance.⁶ The transmission of such mechanical stress is likely to be very different depending on the tissue architecture.

Although our data does not support a role for p53 in aneuploidy surveillance in organotypic cultures, p53 mutations are associated with chromosomal instability (CIN) and poor prognosis in cancer.⁷ We showed that mitotic errors, such as lagging chromosomes and multipolar mitoses, occurred frequently in cells lacking p53. Our work provided insights into how CIN could be reduced when p53 is lost. When Trp53^-/- mCO were treated with nocodazole, an inhibitor of microtubule polymerization that results in SAC activation, fewer lagging chromosomes were seen. This suggests that p53 is unlikely to function through SAC and that mitotic errors that occur in the absence of p53 can, at least in part, be prevented by activating the intrinsic checkpoint mechanism.

The mechanism(s) underlying CIN in the absence of p53 remains unclear. p53 has been reported to localize to centrosomes and TP53-mutant liver organoids had frequent mitotic spindle abnormalities supporting a role for p53 in mitotic spindle function.^8,9 p53 may also be involved in aspects of cellular organization, such as mitotic spindle positioning, establishment of cell polarity, or cell-cell interactions, which can affect mitotic fidelity. Additionally, p53 interacts with many cell cycle and mitotic regulators including cyclin-dependent, Aurora, Polo-like, and SAC kinases.¹⁰ Together, these findings illustrate that p53 is a key player in cell cycle and mitotic signaling networks and underscore the notion that the cause of CIN in the absence of p53 is likely complex and multifaceted. As our work highlights, p53 function is context dependent, adding an additional layer of complexity to this critical tumor suppressor. Ultimately, improving our understanding of how p53 loss leads to CIN could pave the way for therapies that promote chromosomal stability and may be useful for preventing cancer initiation or progression.

Funding Statement

This work was supported by the grant R35 GM118172 from the National Institutes of Health and the ASPIRE Award from the Mark Foundation for Cancer Research to R.L. B.A.J. received support from the NIH Medical Scientist Training Program T32GM136577.

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.