On the role of p53 in the cellular response to aneuploidy.

Most solid tumors are aneuploid, and p53 has been implicated as the guardian of the euploid genome. Previous experiments using human cell lines showed that aneuploidy induction leads to p53 accumulation and p21-mediated G1 cell cycle arrest. We find that adherent 2-dimensional (2D) cultures of human immortalized or cancer cell lines activate p53 upon aneuploidy induction, whereas suspension cultures of a human lymphoid cell line undergo a p53-independent cell cycle arrest. Surprisingly, 3D human and mouse organotypic cultures from neural, intestinal, or mammary epithelial tissues do not activate p53 or arrest in G1 following aneuploidy induction. p53-deficient colon organoids have increased aneuploidy and frequent lagging chromosomes and multipolar spindles during mitosis. These data suggest that p53 may not act as a universal surveillance factor restricting the proliferation of aneuploid cells but instead helps directly or indirectly ensure faithful chromosome transmission likely by preventing polyploidization and influencing spindle mechanics.

INTRODUCTION

Aneuploidy refers to the state of unequal chromosome copy numbers and is one of the most prominent genomic aberrations in solid tumors (Beroukhim et al., 2010; Lengauer et al., 1998; Taylor et al., 2018; Weaver and Cleveland, 2006; Zack et al., 2013). In unicellular eukaryotes, it was shown that aneuploidy, by altering the stoichiometry of a large number of genes, can result in dramatic changes in cellular phenotypes and physiology and confer evolutionary adaptation under selective pressure (Dephoure et al., 2014; Kaya et al., 2015; Pavelka et al., 2010; Selmecki et al., 2006; Sterkers et al., 2012; Sunshine et al., 2015; Torres et al., 2007; Yona et al., 2012). Such basic insight about aneuploidy helps explain recent findings that karyotype alterations are associated with cancer initiation as well as the emergence of drug resistance (Cai et al., 2016; Davoli et al., 2013; Graham et al., 2017; Lane et al., 2014; Lee et al., 2011; Navin et al., 2011; Sack et al., 2018; Stichel et al., 2018; Yang et al., 2019). Indeed, cancer may be viewed as a disease of cellular evolution in a multicellular setting, whereby cells of metazoans turn to resembling unicellular organisms that are free to undergo evolutionary adaptation for better survival and proliferation through gross genomic alterations (Chen et al., 2015; Duesberg et al., 2001; Gerstung et al., 2020; Nowell, 1976).

It is thought that a key difference between mammals and freely adapting unicellular eukaryotes is the presence of p53 that guards genome stability by regulating the DNA damage response, senescence, and apoptosis. (Aylon and Oren, 2011; Hafner et al., 2019; Kastenhuber and Lowe, 2017; Mello and Attardi, 2018; Mijit et al., 2020; Reinhardt and Schumacher, 2014). The loss of functional p53 has been associated with the onset of many metastatic cancers with heightened chromosomal instability; in contrast, an increased p53 gene copy number is thought to be chemopreventive (Bykov et al., 2018; Donehower et al., 2019; Sulak et al., 2016; Wasylishen and Lozano, 2016). Studies in recent years have further suggested roles for p53 in limiting the proliferation of aneuploid cells. However, these studies were limited to established human cell lines that were chromosomally stable and near diploid, such as RPE1, an hTERT-immortalized retinal pigmented epithelial cell line; HCT116, a colon carcinoma cell line; and a few other cell lines (Cianchi et al., 1999; Giam et al., 2019; Hinchcliffe et al., 2016; Janssen et al., 2011; Kurinna et al., 2013; Li et al., 2010; Potapova et al., 2016; Santaguida et al., 2017; Soto et al., 2017; Thompson and Compton, 2010). Recent studies also revealed complex interplay between p53 and several other genome-protective proteins, such as p38, H3.3, and BCL9L (Hinchcliffe et al., 2016; López-García et al., 2017; Simões-Sousa et al., 2018). However, it has been unclear whether a universal signal elicited by abnormal karyotypes may be sensed by the p53 pathway or whether karyotype-specific stress states are sensed through diverse mechanisms and converge upon p53 activation. It was also unknown whether cell type or growth environment could contribute to the p53-mediated response to aneuploidy.

Here, we investigated p53 regulation and downstream cell fate after aneuploidy induction in diverse cell culture models. We set out to answer the following three questions. (1) Is the relationship between aneuploidy induction and p53 activation universal? (2) What are the downstream cell fate consequences after acute aneuploidy induction in different types of cell culture models? (3) Does p53 play a direct or indirect role in faithful mitosis rather than sensing aneuploid cells after erroneous mitosis? Although we confirm that upon acute aneuploidy induction by treating cells with an inhibitor of the spindle assembly checkpoint (SAC) kinase MPS1 (MPS1i), p53 and p21 were upregulated and cause growth arrest in RPE1 and HCT116 cell lines, this response was not conserved in three-dimensional (3D) organotypic cultures of primary cells from mouse and human tissues. Live imaging in colon organoids further supports a role for p53 in mitotic fidelity, as opposed to limiting the proliferation of aneuploid cells.

RESULTS

Induction of aneuploidy with MPS1i in mammalian cell lines and 3D organotypic cultures

To investigate the response to aneuploidy in a broadly representative panel of cell models, we included not only the previously used RPE1 and HCT116 cell lines but also Nalm6, a chromosomally stable pre-B cell lymphoma line that grows in suspension (Hurwitz et al., 1979), and 3D organotypic cultures of primary cells. Potentially influential parameters also included different stages of development, for which we included embryonic and adult mouse neural progenitor cells (eNPCs and aNPCs, respectively). For 3D cultures of primary cells that better mimic in vivo tissues, we also included human mammary organoids (hMOs) (Cheung et al., 2013) and mouse colon organoids (mCOs) (Sato and Clevers, 2013). To induce aneuploidy, each type of culture was treated with the MPS1 inhibitor NMS-P715 (Colombo et al., 2010) for 24 h followed by drug washout and recovery growth for 24 or 48 h in drug-free media to alleviate any direct drug effects before various analyses (Figure 1A). We first quantified the percent numerical aneuploidy by chromosome counting of metaphase spreads 24 h after the MPS1i treatment. For RPE1, HCT116, Nalm6, hMOs, and mCOs, the basal whole-chromosome aneuploidy levels were low in untreated cultures, and a significant increase in aneuploidy percentages, to 45%–55%, was observed in each MPS1i-treated culture (Figure 1B). NPCs also showed increased aneuploidy after MPS1i treatment, but the basal aneuploidy percentages were higher, consistent with earlier reports (Rehen et al., 2005; Yang et al., 2003). The chromosome number distributions showed both gains and losses of chromosomes across most cell models, consistent with the production of aneuploidy as a result of random mitotic errors (Figures 1C and 1D). We also measured the doubling time for RPE1, HCT116, and mCOs and found it to be in the range of 16–22 h (Figure S1A). For metaphase spreads, Colcemid was used to arrest cells in metaphase. However, this limited our analysis to only cells that were able to enter mitosis, which could result in an underestimation of the true aneuploidy frequency. To also analyze interphase cells, we performed chromosome counting in RPE1, HCT116, and mCOs by using Calyculin A, which causes premature chromosome condensation and does not require mitotic entry. We observed close to 60% aneuploidy in RPE1, HCT116, and mCOs after MPS1i treatment using this method (Figure 1E). We also performed live-cell imaging of H2B-mNeon RPE1, HCT116, and mCOs and observed comparable increases in lagging chromosome frequency during MPS1i treatment. (Figure 1F). Together, these analyses demonstrate that we induced similar levels of aneuploidy in each cellular system. Importantly, our method of inducing aneuploidy did not involve cell synchronization, as some agents altering cell cycle progression may cause DNA damage (Darzynkiewicz et al., 2011; Halicka et al., 2016), which could activate p53. As expected, pH2AX immunoblots and immunofluorescence staining showed that the MPS1i treatments used in our experiments did not induce DNA damage (Figures S1B–S1D). We also quantified the frequency of micronuclei—a possible source of DNA damage following MPS1i-induced chromosome missegregation (Crasta et al., 2012; Terradas et al., 2009; Zhang et al., 2015). Micronuclei formed infrequently and at a similar frequency in MPS1i-treated RPE1, HCT116, and mCOs (Figure S1E), suggesting that micronuclei formation was not a source of DNA damage in our study.

Figure 1.

Induction of aneuploidy with MPS1i in mammalian cell lines and 3D organotypic cultures

(A) Schematic of the workflow to induce aneuploidy and detect p53 regulation in RPE1, HCT116, Nalm6, neural progenitor cells (NPCs), and organoids and downstream cell fate analysis.

(B) Representative images of metaphase spreads from untreated and MPS1i-treated RPE1 cells (top). Quantification of the percent of metaphase spreads with an aneuploid chromosome number in each cellular model (bottom). Cell lines RPE1, HCT116, Nalm6, human mammary organoid (hMO), and mouse colon organoid (mCO) were treated with 1 μM MPS1i and 400 nM for mouse NPCs for 24 h, followed by drug washout and recovery growth in drug-free media for 24 h before sample collection. n ≥ 3 biological replicates; n ≥ 30 mitoses counted for each condition per replicate. Scale bar, 10 μm.

(C and D) Histograms represent the number of spreads with the indicated number of chromosomes by metaphase spreads of cell lines and organoid cultures as described in (B).

(E) Percent aneuploidy for calyculin-A-treated RPE1, HCT116, and mCOs. n ≥ 30 mitotic spreads counted for each condition.

(F) Lagging chromosome frequency quantified by live-cell imaging of H2B-mNeon-labeled cells. MPS1i was added 1 h prior to the start of imaging. n = 3 biological replicates, n > 120 mitoses scored per condition for mCO and HCT116, and n > 40 mitoses scored per condition for RPE1.

Error bars represent mean ± SEM. *p ≤ % 0.05; **p ≤ % 0.01; ***p ≤ % 0.001; one-tailed Fisher’s exact test (B, D, and E) and one-way ANOVA (F); n.s., not significant.

Adherent RPE1 and HCT116 but not suspension Nalm6 cells depend on p53 for growth arrest after aneuploidy induction

We next determined the effect of acute aneuploidy induction on p53 and the CDK inhibitor CDKN1A (p21), a well-studied direct transcriptional target of p53 that controls cell cycle progression (Dulić et al., 1994; el-Deiry et al., 1993). p53 is known to be regulated predominantly by posttranslational modifications that affect its stability or nuclear accumulation (Haupt et al., 1997; Kruse and Gu, 2009; Oren, 1999). Acute induction of aneuploidy using drugs that perturb mitosis or SAC led to increased p53 stabilization and a consequent G1 arrest upon transcriptional activation of p21 (Kollu et al., 2015; Li et al., 2010). We observed significant increases in p53 levels in the adherent RPE1 and HCT116 cells and suspension Nalm6 cells, but not in the NPCs or 3D organotypic cultures (presented further below), after aneuploidy induction using MPS1i (Figures 2A and 2B). Additionally, nuclear localization of p53 was confirmed in HCT116 and RPE1 cells by immunofluorescence staining (Figures S1F and S1G). All cell lines showed increased p53 protein abundance upon treatment with nutlin-3 (Shen and Maki, 2011), which stabilizes p53 by disrupting the p53-MDM2 complex, and DNA damaging agents such as doxorubicin and bleomycin (Figures 2A, 3A, 3B, S1F, and S1G).

Figure 2.

Adherent RPE1 and HCT116 but not suspension Nalm6 cells depend on p53 for growth arrest after aneuploidy induction

(A) Levels of p53 and p21 as determined by immunoblot analysis. Unsynchronized populations of mammalian cell lines RPE1, HCT116, and Nalm6 and NPCs either untreated or MPS1i treated for 24 h followed by drug washout and recovery growth in drug-free media for 24 h or 48 h. Bleomycin, doxorubicin, and nutlin were used as positive controls for p53 and p21 induction. GAPDH served as a loading control.

(B) Quantification of p53 fold change by densitometry analysis; n ≥ 3 biological replicates of the data in (A).

(C) Representative fluorescence-activated cell sorting (FACS) profiles of EdU incorporation in untreated, MPS1i-treated with drug-free recovery, and nutlintreated TP53+/+ or TP53 RPE1 cells.

(D) Quantification of the percentages of EdU+ cells from the FACS data; n ≥ 3 biological replicates in the same experiment as in (C). Graphs show mean ± SEM. *p ≤ % 0.05, **p ≤ % 0.01, ***p ≤ % 0.001; multiple t test (B) and one-tailed Fisher’s exact test (D); n.s., not significant.

Figure 3.

3D organotypic cultures do not activate p53 or undergo growth arrest in response to aneuploidy

(A) Representative immunoblots from mCOs and hMOs showing p53 and p21 protein abundance after aneuploidy induction with MPS1i and positive-control cells treated with bleomycin, nutlin, or doxorubicin.

(B) Quantification of p53 and p21 fold change by densitometry analysis; n ≥ 3 biological repeats of the data in (A).

(C) Quantification of the percentages of EdU+ cells from the FACS data; n ≥ 3 biological replicates in Trp53+/+ and Trp53 mCOs either untreated or treated with MPS1i, followed by EdU incorporation for 16 h.

(D) 3D reconstruction of a hMOs showing EdU+ cells along with quantification; n ≥ 3 biological replicates. Scale bar, 20 μm.

Graphs show mean ± SEM. *p ≤ % 0.05, **p ≤ % 0.01, ***p ≤ % 0.001; multiple t test (B) and one-tailed Fisher’s exact test (D); n.s., not significant.

p53 plays a critical role in controlling cell proliferation and death in response to stress (Kastenhuber and Lowe, 2017; McKinley and Cheeseman, 2017). We examined whether p53 induction was associated with either cell cycle arrest or apoptosis by assaying DNA synthesis using the 5-ethynyl-2′-deoxyuridine (EdU) incorporation assay (Salic and Mitchison, 2008) and apoptosis using Annexin V and propidium iodide. In MPS1i-treated RPE1, HCT116, and Nalm6 cells, there was a significant reduction in EdU-positive cells (Figures 2C and 2D), consistent with a suppression of proliferation, but these populations did not display a significant increase in apoptotic cells (Figure S2E). Cell cycle analysis for RPE1, HCT116, and Nalm6 cells showed significant G1 arrest after aneuploidy induction (Figures S3A and S3B). To test the requirement for p53 in mediating this arrest, we used TP53 HCT116 (Bunz et al., 2002), RPE1 TP53 (Lambrus et al., 2016), and TP53 Nalm6 cells generated using CRISPR-Cas9. p53 knockout was confirmed in RPE1, HCT116, and Nalm6 cell lines by immunoblotting (Figure S2A). These lines lacking p53 did not show p21 upregulation after aneuploidy induction. Additional genotyping validation was performed for Nalm6 (Figure S2B). Deviation from the diploid chromosome number in these cells lacking p53 was observed even without MPS1i treatment (Figure S2C). p53 knockout rescued cell proliferation, as indicated by the increased EdU-positive cells, in MPS1i-treated HCT116 and RPE1 populations (Figures 2C and 2D). However, this rescue was not observed in Nalm6 TP53 suspension cells (Figure 2D), suggesting that, although p53 was induced after aneuploidy, it was not required for the reduced proliferation in Nalm6 cells. TP53 Nalm6, but not HCT116 or RPE1, cells also showed an increase in apoptosis after aneuploidy induction (Figure S2E).

3D organotypic cultures do not activate p53 or undergo growth arrest in response to aneuploidy

The three established cell lines (RPE1, HCT116, and Nalm6) responded to aneuploidy as expected despite a lack of p53 dependence for the growth arrest in Nalm6 cells, whereas we did not observe an increase in p53 or p21 protein abundance after aneuploidy induction in 3D cultures of NPCs, mCOs, or hMOs (Figures 2B, 3A, and 3B). As a positive control, treatment with nutlin, doxorubicin, or bleomycin increased p53 and p21 protein abundance, suggesting that p53 is generally functional in these cell models. Consistent with a lack of p53 activation in 3D organotypic cultures after aneuploidy induction, an EdU incorporation assay showed unimpeded cell proliferation, in contrast to reduced proliferation of the same cultures treated with nutlin or DNA-damaging agents (Figures 3C, 3D, and S2D). Also, cell cycle analysis for mCOs did not show significant G1 arrest after aneuploidy (Figure S3B). A recent study demonstrated that cellular architecture has profound impacts on chromosome segregation fidelity (Knouse et al., 2018). To rule out the possibility that the differential responses to aneuploidy observed between the adherent epithelial cell lines and organotypic cultures was due to a 2D versus 3D culture environment, we cultured HCT116 cells as 3D spheres in Matrigel (Figure S4A). These cells grew into spherical structures with high-level EdU incorporation, indicating active proliferation (Figure S4C). Upon aneuploidy induction, HCT116 cells cultured in 3D exhibited similar levels of p53 and p21 increase and growth arrest, as observed for those cultured in 2D (Figures S4B, S4C, and S4D). This result suggests that the difference in aneuploidy responses observed among the different cell types tested was unlikely attributed to 2D versus 3D culture environment.

Trp53 mCOs exhibit frequent mitotic aberrations

Although mCOs did not exhibit elevated p53 or G1 arrest after aneuploidy induction, mCOs generated from Trp53 mice showed around 40% aneuploidy (Figures S5A and S5B) compared to (Figure 1B) 15% aneuploidy in Trp53+/+ mCOs, suggesting that p53 may have a role in preventing aneuploidy occurrence instead of inhibiting the proliferation of aneuploid cells. To test this, Trp53+/+ and Trp53 mCOs were transduced with H2B-mNeon, and mitotic cell divisions were observed using live 3D confocal imaging (Videos S1 and S2). Mitotic aberrations including multipolar divisions and lagging chromosomes were quantified from the live movies. The fraction of cells exhibiting lagging chromosomes was increased in Trp53 when compared to Trp53+/+ mCOs, and there was also an increase in multipolar divisions (Figures 4A and 4B; Videos S3 and S4). In the presence of low-dose nocodazole, the time in mitosis (from nuclear envelope breakdown [NEBD] to anaphase onset) was lengthened significantly in both Trp53+/+ and Trp53 mCOs, suggesting that p53 loss did not disrupt canonical kinetochore-based SAC signaling (Figure 4C). Interestingly, induction of SAC by low-dose nocodazole reduced the frequency of lagging chromosomes in Trp53 mCOs, suggesting that lengthening mitotic duration facilitates error correction in the p53-deficient background. The lagging chromosomes could occur secondary to polyploidization or centrosome amplification (Fukasawa et al., 1996; Galipeau et al., 1996; Lopes et al., 2018). Analysis of nuclear volume immediately before (G2) and after (G1) mitosis from the videos confirmed nuclear volume scales with DNA content (Figure 4D). The G2 nuclear volume of cells with lagging chromosomes in bipolar spindles was not significantly different from cells undergoing normal mitoses (Figure 4D). In contrast, the average G2 nuclear volume of cells that underwent multipolar divisions was approximately double that of cells with normal divisions (Figure 4D). This finding suggests that tetraploidization occurred prior to multipolar divisions but was not the cause of lagging chromosomes in Trp53 cells with bipolar mitoses. Centrosome amplification, independent of genome doubling, gives rise to lagging chromosomes (Ganem et al., 2009). In Trp53 mCOs, increased centrosome number determined by pericentrin immunofluorescence was exclusively associated with increased mitotic chromatin volume (Figure 4E). Total pericentrin fluorescence intensity per centrosome was similar in cells with two centrosomes and those with more than two centrosomes in Trp53+/+ and Trp53 mCOs (Figure S5C), indicating centrosome clustering was not impeding our ability to accurately count centrosome number. This result suggests that extra centrosomes are associated with tetraploidization in Trp53 mCOs. Therefore, centrosome amplification is unlikely to be the cause of lagging chromosomes in diploid Trp53 cells in mCOs.

Figure 4.

Trp53 mCOs exhibit frequent mitotic aberrations

(A) Representative frames from time-lapse videos of Trp53+/+ and Trp53 mCOs with chromosomes labeled with H2B-mNeon undergoing normal mitosis (top row), mitosis with lagging chromosomes (middle row), and multipolar mitosis (bottom row). Arrowheads indicate relevant mitotic aberrations. Scale bars, 10 μm.

(B) Quantification of the percentage of mitoses with a mitotic error in untreated and 80 nM nocodazole-treated Trp53+/+ and Trp53 mCOs. Each data point represents one organoid. Percentage was obtained by dividing the number of mitoses with an error by the total number of mitoses observed. n ≥ 6 organoids observed per condition, n > 150 divisions scored for untreated Trp53+/+ and Trp53, and n > 80 divisions scored for nocodazole-treated Trp53+/+ and Trp53.

(C) Quantification of time from nuclear envelope breakdown (NEBD) to anaphase onset. Points with black outlines represent the mean of one organoid. Light-gray points represent one mitosis. n > 150 divisions scored for untreated Trp53+/+ and Trp53, and n > 80 divisions scored for nocodazole-treated Trp53+/+ and Trp53.

(D) Quantification of nuclear volume immediately prior to (G2) or after (G1) mitosis in H2B-mNeon Trp53 mCO.s Each point represents one nucleus. n ≥ 6 organoids observed per replicate.

(E) Quantification of mitotic chromatin volume and centrosome number of dividing cells in Trp53+/+ and Trp53 mCOs. Chromatin was labeled using DAPI, and centrosome number was quantified by pericentrin immunofluorescence (IF) staining. n ≥ 8 organoids per replicate.

Bars in (B) and (C) represent mean ± SEM of n ≥ 6 total organoids for n = 3 independent experiments. Lines in (D) and (E) represent mean of all data points. *p ≤ % 0.05, **p ≤ % 0.01, ***p ≤ % 0.001; Wilcoxon rank-sum test (B and C) or unpaired Welch’s unequal variances t test (D and E). See also Videos S1, S2, S3, and S4.

DISCUSSION

We have examined the generality of p53 upregulation and cellular responses after acute induction of aneuploidy in a variety of cell models linked to different tissue origins and growth environments. Based on earlier findings in immortalized and cancer cell lines such as RPE1 and HCT116, our original interest was to identify aneuploidy-associated signals that induce p53 activation in these cells. However, our inability to generate a similar response in NPCs and the observation of a p53-independent response in Nalm6 pre-B lymphocytes suggested to us that p53 activation after aneuploidy induction is not necessarily a universal response but may be more cell line specific. This led us to conduct a more comprehensive investigation of cellular responses in 3D organoid cultures. We found that induction of aneuploidy does not result in p53 accumulation or cell proliferation arrest in organotypic cultures from epithelial tissues of neural, intestinal, and mammary origins from mouse or human. The p53 response and cell cycle fate after aneuploidy induction in each of the cellular systems used in our study are summarized in Table S1. Testing organoids derived from a broader range of tissues will allow further assessment of the generality of our findings.

We recognized that two previous studies found that not all aneuploid RPE1 cells resulting from erroneous mitoses activate p53 (Santaguida et al., 2017; Soto et al., 2017). Those studies concluded that the p53-mediated G1 arrest may be more associated with complex aneuploidy or structural aneuploidy in RPE1 cells. Our data in RPE1 and HCT116 cells do not contradict these conclusions but differ in the degree of p53 induction. Also, our data do not address the nature of the signal that activates p53. In our experiments, the 24-h MPS1i treatment did not induce DNA damage, although we cannot rule out that continued propagation of aneuploid cells in organotypic cultures could lead to DNA damage and p53 activation at a later time. Responses to aneuploidy induction in the Drosophila brain varied in different cell types and at different times (Mirkovic et al., 2019). The different p53 response observed in our study was also unlikely to be due strictly to a 2D versus 3D culturing system because 3D cultures of HCT116 cells showed a similar p53 response to those cultured in 2D.

Given that 3D organotypic cultures of primary cells are better mimics of native tissues than immortalized or cancer cell lines (Schutgens and Clevers, 2020), our findings do not support the notion that p53 acts in a surveillance mechanism against aneuploidy in vivo. It is possible that stress associated with artificially immortalized or cancer cell lines passaged in the conventional cell culture environment combined with the stress caused by aneuploidy can trigger p53 accumulation and cell cycle arrest. Interestingly, the Nalm6 pre-B lymphocyte cell line showed a p53-independent cell cycle arrest following aneuploidy induction, suggesting that another yet unidentified mechanism exists to limit the proliferation of aneuploid blood cells. This may be in line with the observation that certain hematological cancers show lower TP53 mutational rates (Olivier et al., 2010; Prokocimer et al., 2017; Robles et al., 2016). Further studies are warranted to understand the specific molecular mechanisms driving the differences in p53 response to aneuploidy induction.

Although our data do not support a role for p53 in aneuploidy surveillance in organotypic cultures, there is strong evidence indicating that the loss of p53 is associated with chromosome instability and poor prognosis in the development of several cancers (Donehower et al., 2019; Foijer et al., 2014; Fujiwara et al., 2005; Watson and Elledge, 2017), and a majority of tumors show varied TP53 mutations (Clausen et al., 1998; Muller and Vousden, 2013). Live-cell imaging of Trp53+/+ and Trp53 mCOs showed that mitotic errors, including lagging chromosomes and multipolar mitoses, occurred frequently in cells lacking p53, consistent with several published studies (Artegiani et al., 2020; Drost et al., 2015). Our analysis also indicates that tetraploidization is likely to be a precursor to multipolar divisions in Trp53 mCOs. There has been significant debate surrounding the role of p53 in suppressing the division of tetraploid cells (Andreassen et al., 2001; Ganem and Pellman, 2007; Horii et al., 2015; Uetake and Sluder, 2004; Wong and Stearns, 2005). Recent work showed that p53 transcriptionally regulates multiple genes important for mitotic processes, including Aurka, Foxm1, and Plk4, and that a loss of p53 leads to polyploidization during liver regeneration in vivo (Kurinna et al., 2013). Aberrant expression of these mitotic genes in the absence of p53 may drive tetraploidization and lead to multipolar divisions in Trp53 mCOs. Importantly, tetraploidization was found to be sufficient to drive tumorigenesis in cells lacking p53 (Fujiwara et al., 2005).

The underlying cause of the observed lagging chromosomes in Trp53 mCOs remains unclear. An increase in lagging chromosomes was also reported for p53-deficient human liver and colon organoids (Artegiani et al., 2020; Drost et al., 2015). Our results suggest that lagging chromosomes in bipolar divisions do not occur secondarily to polyploidization or centrosome amplification. p53 loss does not appear to disrupt SAC activation induced by nocodazole treatment, and in fact, SAC activation reduced the occurrence of lagging chromosomes. It is thus possible that the loss of p53 directly or indirectly leads to the type of mitotic problems, such as faulty kinetochore-microtubule attachments, that can be corrected when mitosis is lengthened by SAC. Interestingly, lagging chromosomes can impede cytokinesis and result in the generation of tetraploid cells (Lens and Medema, 2019; Shi and King, 2005; Steigemann et al., 2009). Thus, it is possible that lagging chromosomes in Trp53 mCOs contribute to the generation of tetraploid cells that then undergo multipolar divisions. A better understanding of the precise role for p53 in suppressing aneuploidy in physiological settings may allow for the development of therapeutics for the prevention of mitotic abnormalities and, ultimately, cancers associated with p53 loss.

STAR★METHODS

RESOURCE AVAILABILITY

Lead contact

Further information and requests for resources and reagents should be directed to and will be fulfilled by the Lead Contact, Rong Li (rong@jhu.edu).

Materials availability

This study did not generate new unique reagents.

Data and code availability

The published article includes all datasets and code generated or analyzed during this study.

EXPERIMENTAL MODEL AND SUBJECT DETAILS

Animals

Wild-type C57BL/6J were purchased from the Jackson Laboratory and Trp53 mice (The Jackson Laboratory) were a gift from Andrew Holland’s lab. Mice used for NPC and colon organoid isolation were between 6 and 12 weeks old. Male and female wild-type mice were used for NPC and colon organoid isolation and male Trp53 mice were used for colonoid isolation. All mouse care and use were approved by the Institutional Animal Care and Use Committee at the Johns Hopkins University.

Tissues

Normal mammary reduction samples were acquired from the Cooperative Human Tissue Network in compliance with a study protocol (NA_00077976) that was categorized as exempt/not human subject research by the Johns Hopkins School of Medicine Institutional Review Board. Each sample was deidentified prior to receipt and shipped overnight in basal DMEM/RPMI medium.

Cell culture

hTERT-RPE1 were grown at 37°C in 5% CO2 in DMEM supplemented with 10% fetal bovine serum (FBS). RPE-1 TP53 cells were a gift from Andrew Holland’s lab. HCT116 cells were grown at 37°C in 5% CO2 in McCoy’s media supplemented with 10% fetal bovine serum (FBS). HCT116 TP53 cells were a gift from Bert Vogelstein lab. Nalm6 cells were a gift from Mike Tyers’ lab and were grown at 37°C in 5% CO2 in RPMI supplemented with 10% fetal bovine serum (FBS). Nalm6 TP53 cells were generated in-house using lentiCRISPRV2. TP53 knockout was confirmed by sequencing and western blot validation. Neuronal Progenitor cells isolated from embryonic mouse brain (frontal) and adult 10-week-old mice SVZ and DG areas and cultured as neurospheres on ultra-low attachment plates using DMEM and Ham’s F-12 medium in a 1:1 ratio (DMEM/F-12; Omega Scientific, cat. no. DM-25) B27 serum-free supplement (B27; Invitrogen/GIBCO, cat. no. 17504-044) Basic fibroblast growth factor-2 (FGF-2; PeproTech, cat. no. 100-18B-B) Epidermal growth factor (EGF; PeproTech, cat. no. 100-15) l-Glutamine (Invitrogen/GIBCO, cat. no. 2503-081) Antibiotic-antimycotic (Anti-Anti; Invitrogen/GIBCO, cat. no. 15240-062) as previously described in Guo et al. (2012).

Colonoid isolation and culture

Mouse colon organoids were generated from wild-type C57BL/6J and Trp53 mice. All mouse care and use were approved by the Institutional Animal Care and Use Committee at the Johns Hopkins University. Trp53 genotype was determined using the following primers: 5′-CCCGAGTATCTGGAAGACAG-3′ and 5′-ATAGGTCGGCGGTTCAT-3′. Murine colon organoids were generated as previously described (Sato et al., 2011). The distal third of the colon was removed and rinsed with PBS. The distal colon was placed in complete chelating solution and minced. Minced tissue was then incubated with 150 rpm shaking in 31.25 mM EDTA in chelating solution for 30 minutes-1 hours at 4°C to release crypts. Tissue fragments were allowed to settle and supernatant containing crypts was collected. Crypts were centrifuged at 200 × g for 5 minutes and embedded in Matrigel and seeded on 24 well plates. After solidification of Matrigel, 500 μl WENR culture medium was added (Sato et al., 2011). Media was changed every 2 days and mCO were passaged using enzymatic dissociation with TrypLE Express every 7 days.

Generation of primary human mammary organoids

Normal mammary reduction samples were acquired from the Cooperative Human Tissue Network in compliance with a study protocol (NA_00077976) that was categorized as exempt/not human subject research by the Johns Hopkins School of Medicine Institutional Review Board. Each sample was deidentified prior to receipt and shipped overnight in basal DMEM/RPMI medium. Upon receipt, samples were washed with an anti-fungal solution, minced, and digested in a collagenase (Sigma C2139) solution. Normal mammary epithelial organoids were then enriched by performing a series of differential centrifugation steps, as previously described (Nguyen-Ngoc et al., 2012). Mammary organoids were then embedded in growth factor reduced Matrigel and cultured in the presence of medium containing EGF, hydrocortisone, insulin, and cholera toxin (Cheung et al., 2013).

METHOD DETAILS

Lentivirus production

Plasmids used: pMD2.G, psPAX2, pLV-H2B-Neon-ires-Puro. HEK293FT cells were co-transfected with the lentiviral transfer plasmid, packaging plasmid, and envelope plasmid. Media containing lentivirus was collected 24 and 48 hours after transfection. Lentivirus was concentrated using a centrifugal filter (Amicon Ultra-15, 100,000 NMWL). Lentiviral titer was determined by qPCR (abm qPCR Lentivirus Titration Kit, cat. # LV900).

Live imaging

mCO, HCT116 and RPE1 were transduced with H2B-Neon lentivirus as described previously (Bolhaqueiro et al., 2018, 2019). Following transduction, mCO were cultured in WENR culture media for 5–7 days. mCO were then dissociated to single cells and allowed to grow for 2 days in WENR culture media. Then, cells stably expressing the construct were selected with 1 μg/ml puromycin for 2–4 days. Transduced HCT116 and RPE1 cells were selected beginning 48 hours after addition of lentivirus using 1 μg/ml and 3 μg/ml puromycin, respectively. For live imaging, mCO, HCT116 and RPE1 were seeded on 8-well coverglass bottom chamber slides. 3–5 days after seeding for mCO and 2 days after seeding HCT116 and RPE1 cells were imaged using a laser-scanning confocal microscope (Zeiss 780). The environmental chamber was maintained at 37°C with 5% CO2. Cells were imaged with a 40x water-immersion objective. Cells were imaged every 3–3.5 minutes with 15–20 2.0 μM z slices for mCO and 15–20 1.0 μM z slices for HCT116 and RPE1. Images were processed using a custom ImageJ macro modified from the ImageJ plugin “Temporal-Color Code” (Miura, 2010). Mitotic errors were scored manually.

Metaphase chromosome spreads

Chromosome spreads were obtained by arresting cells in mitosis in Karyomax Colcemid solution (1:100; Life Technologies) for 4–8 h, followed by harvesting using trypsin and/or accutase. Trypsinized cells were collected by centrifugation for 5 min at 300 × g and gently resuspended in a small amount of medium (~1 ml). Resuspended cells were allowed to swell for 7–10 min in 0.4% KCl solution at room temperature and prefixed by addition of freshly prepared methanol: acetic acid (3:1) fixative solution (~100 μL per 10 mL of total volume). Prefixed cells were collected by centrifugation and fixed in methanol:acetic acid (3:1) fixative solution. Spreads were dropped on a glass slide and incubated at 65°C for at least 1 h. For counting, spreads were mounted in Vectashield containing 4′,6-diamidino-2-phenylindole (DAPI; Vector Laboratories, Burlingame, CA), imaged on a Nikon TiE-Eclipse epifluorescence microscope (60 × oil immersion objective), and counted using the Cell Counter plug-in in ImageJ (Schneider et al., 2012)

Immunoblotting

Mammary organoids were isolated from Matrigel by incubation with dispase (Corning) at 37°C for 1 hr. Digestion was terminated by addition of 5 mM EDTA and organoids resuspended by trituration. Colon organoids were isolated from Matrigel by shaking at 150 rpm at 4°C in Cell Recovery Solution (Corning). Organoids and cells were pelleted at 200 g for 5 min and resuspended in RIPA buffer (50 mM Tris pH 8.0, 150 mM sodium chloride, 1% NP-40, 0.5% sodium deoxycholate, 0.1% SDS) containing Complete Mini Protease Inhibitor Cocktail (Roche). Lysates were homogenized by vortexing, incubated on ice for 20 min, and centrifuged at 13,000 rpm at 4°C for 20 min. Protein concentration of supernatants was measured using Bradford dye (Bio-Rad) on a spectrophotometer. Lysates were diluted in 5X sample buffer (250 mM Tris pH 6.8, 50% glycerol, 5% β-mercaptoethanol 0.025% bromophenol blue, 5% SDS). Samples were separated on homemade polyacrylamide gels or Bolt 4–12% Bis-Tris Plus gels (ThermoFisher) and transferred to Immobilon-P membranes (Millipore) via wet transfer or PVDF membranes via iBlot dry transfer (ThermoFisher). Membranes were blocked with 5% milk in TBST (50 mM Tris pH 8.0, 150 mM NaCl, 0.1% Tween-20) for 1 hr at room temperature. Membranes were incubated in primary antibodies diluted in 5% milk in TBST at 4°C with rocking for 16 hr and washed with TBST for 10 min thrice. Membranes were incubated in secondary antibodies diluted in TBST at room temperature with rocking for 1 hr and washed with TBST for 10 min thrice. Membranes were incubated in ECL Prime Western Blotting Detection Reagent (GE Healthcare) for 5 min and imaged on an ImageQuant LAS 4000 luminescent image analyzer (GE Healthcare).

Immunofluorescence

For p53 and pH2AX immunofluorescence, cells were fixed in 4% PFA/PBS for 15 min and permeabilized with 0.5% Triton X-100. Blocking was done with 5% boiled normal goat serum in PBS/0.5% Triton X-100. Primary and secondary antibodies were diluted in 2.5% (wt/vol) bovine serum albumin (BSA)/PBS/0.5% Triton X-100. Specimens were incubated with primary antibodies overnight, washed three times for 5–10 min, and incubated with fluorescently conjugated secondary antibodies for 2–4 h. All washes were performed with PBS/0.5% Triton X-100. DNA was counterstained with DAPI or Hoechst 33342 (Thermo Fisher Scientific). Vectashield (Vector Laboratories) was used for mounting. p53 and pH2AX images were acquired with the Nikon TiE-Eclipse fluorescence microscope (20 × air and 60 × oil immersion objectives). For p53 immunofluorescence, three or four tiled images containing hundreds of cells were analyzed to determine the mean number of positive cells and the SD. At least 30 cells were imaged for each condition. Data were visualized and analyzed using ImageJ. After background subtraction, images were thresholded and the area of each object quantified using MetaMorph software. For pericentrin IF in mCO, cells were grown in 8 well glass-bottom coverslides and fixed with 4% paraformaldehyde for 10 min. Cells were permeabilized with for 1 hour at room temperature with 0.5% Triton X-100 in PBS and blocked for 3 hours at room temperature with 10% FBS + 0.1% Triton X-100 in PBS. Primary and secondary antibodies were diluted in 2% FBS + 0.1% Triton X-100. Specimens were incubated with primary antibodies overnight, washed three times for 10 minutes, and incubated with fluorescently conjugated secondary antibodies for 2 hours. All washes were performed with PBS. DNA was counterstained with DAPI and F-actin was labeled with Alexa Fluor 555 Phalloidin. Images were acquired using an LSM780 confocal microscope (40x water immersion objective). At least 8 organoids were imaged per condition.

EdU labeling

For EdU incorporation assays, cells were typically seeded in multiwell black, optically clear bottom, tissue culture–treated plates (PerkinElmer) for imaging and treated with 10 μM EdU (Thermo Fisher Scientific) for the indicated times. Cells were fixed in 4% paraformaldehyde (PFA)/phosphate-buffered saline (PBS) for 15 min and then permeabilized with 0.5% Triton X-100. Fixed cells were washed with PBS and stained with 1 μM Alexa Fluor 488 or Alexa Fluor 555–conjugated azide diluted in PBS containing 2 mM CuSO4 and 50 mM ascorbic acid. To counterstain the DNA, Hoechst 33342 was added to 2 μg/ml. Cells were incubated for several hours or overnight at room temperature protected from light and evaporation and then washed three times with PBS. Later they were subjected to imaging using Nikon TiE-Eclipse epifluorescence microscope (60 × oil immersion objective) and/or Attune flow analyzer.

Image analysis

Nuclear and mitotic chromatin volumes in mCO were calculated using the commercial software Imaris, version 9.5. Nuclei and mitotic chromatin were segmented by H2B-mNeon (live imaging) or DAPI (fixed cells) fluorescence using the “Surface” function of Imaris. Volume of the generated surfaces was quantified and reported. Centrosomes were segmented using the “Spots” function of Imaris. Centrosome number per mitotic cell was manually counted following segmentation.

Doubling time analysis

Cell population doubling time was quantified in HCT116, RPE1, and Trp53+/+ H2B-mNeon mCO. For the 2D cell lines (HCT116 and RPE1), cell number was counted using flow cytometry. Dead cells were identified and excluded from the cell count by staining with 1 μg/mL propidium iodide. To quantify mCO doubling time, Trp53+/+ H2B-mNeon mCO were dissociated to single cells using TrypLE Express Enzyme (GIBCO, catalog 12604013) and plated in a 96 well glass-bottom imaging plate at a density of 3,000 cells per well. Twenty randomly selected mCO were imaged using a using a laser-scanning confocal microscope (Zeiss 780). Nuclei number per organoid was quantified by segmenting individual nuclei using the “Surface” function of Imaris as described above. Doubling time was calculated by: . g = growth rate, N = average cell number, t = 24 hours. .

Drug Treatments

We used the following chemical inhibitors: MPS1 Inhibitor (NMS-P715) 0.4 μM for NPC and 1μM for all other cell lines and organoid cultures, Nutlin-3 at 1–5 μM, Bleomycin at 2.5 μg/ml, Doxorubicin at 0.5–5 μM, ZM 447439 at 1 μM, Nocodazole at 80nM and CalyculinA at 50 nM.

QUANTIFICATION AND STATISTICAL ANALYSIS

The number of biological replicates, the number of cells analyzed per biological replicate, and the statistical test performed are indicated in the figure legends. A biological replicate was defined as tissue or cells harvested from a single mouse or patient sample. Statistical tests performed and figures generated using Prism (GraphPad) and/or R-Studio. For all statistical tests, a cutoff of p < 0.05 was used to indicate significance.

KEY RESOURCES TABLE

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Antibodies
Anti-p53 [DO-1], mouse monoclonal	Abcam	Cat# ab1101; RRID:AB_297667
Anti-p53 (1C12), mouse monoclonal	Cell Signaling	Cat# 2524; RRID:AB_331743
Anti-p21 Waf1/Cip1 (12D1), rabbit monoclonal	Cell Signaling	Cat# 2947; RRID:AB_823586
Anti-p21 Waf1/Cip1 (F-5), mouse monoclonal	Santa Cruz	Cat# sc-6246; RRID:AB_628073
Anti-GAPDH (D16H11), rabbit monoclonal	Cell Signaling	Cat# 5174; RRID:AB_10622025
Anti-pericentrin, mouse monoclonal	BD Biosciences	Cat# 611814; RRID:AB_399294
Anti-phospho-histone H2A.X (Ser-139), rabbit monoclonal	Cell Signaling	Cat #9718; RRID:AB_2118009
Anti-rabbit IgG, HRP-linked Antibody	Cell Signaling	Cat# 7074; RRID:AB_2099233
Anti-mouse IgG, HRP-linked Antibody	Cell Signaling	Cat# 7076; RRID:AB_330924
Alexa Fluor 488 Polyclonal Antibody	ThermoFisher	Cat# A-11094; RRID:AB_221544
Biological samples
Normal human mammary reduction samples	Cooperative Human Tissue Network	https://www.chtn.org/
Chemicals, peptides, and recombinant proteins
NMS-P715	EMD/Millipore	Cat# 475949
Nutlin-3	Sigma Aldrich	Cat# N6287
Bleomycin	EMD/Millipore	Cat# 203410
Nocodazole	Sigma Aldrich	Cat# M1404
ZM 447439	Selleckchem	Cat# S1103
Calyculin A	Cayman Chemical	Cat# 19246
Doxorubicin	Sigma Aldrich	Cat# AMBH324A4B72
Critical commercial assays
Click-iT EdU Alexa Fluor 488	Invitrogen	Cat# C10420
Annexin V, Alexa Fluor 488 conjugate	Invitrogen	Cat# A13201
Experimental models: cell lines
RPE-1 hTERT	ATCC	CRL-4000
RPE-1 TP53−/−	Laboratory of Andrew Holland	N/A
HCT116	Laboratory of Bert Vogelstein	N/A
HCT116 TP53−/−	Laboratory of Bert Vogelstein	N/A
Nalm6	Laboratory of Mike Tyers	N/A
Nalm6 TP53−/−	This study	N/A
Experimental models: organisms/strains
Mouse: wild-type C57BL/6J	The Jackson Laboratory	RRID:IMSR_JAX:000664
Mouse: B6.129S2-Trp53tm1Tyj/J	The Jackson Laboratory	RRID:IMSR_JAX:002101
Recombinant DNA
pLV-H2B-Neon-ires-Puro	Bolhaqueiro et al., 2018, 2019	N/A
pMD2.G	Laboratory of Didier Trono	RRID:Addgene_12259
psPAX2	Laboratory of Didier Trono	RRID:Addgene_12260
Software and algorithms
ImageJ	Schneider et al., 2012	https://imagej.nih.gov/ij/
IMARIS	Oxford Instruments 2021	https://imaris.oxinst.com/
Prism (version 8)	Graphpad	N/A
R, version 3.6.1	R Core Team, 2019	https://www.R-project.org/
RStudio, version 1.1.463	RStudio Team, 2020	https://www.rstudio.com/
ggplot2, version 3.2.1	Wickham, 2016	https://ggplot2.tidyverse.org/