RPE-1 cells were obtained from Arshad Desai Lab (UCSD). DLD1 cells were obtained from ATCC. MDA-MB-231 and HEK293T cells were obtained from NCCS, Pune. RPE-1 cells were grown in DMEM-F12 (Gibco-11330–032), DLD1 cells in RPMI (12633–012), MDA-MB-231 and HEK293T in DMEM (Himedia-AL007S) supplemented with 10% FCS (Thermo Fisher Scientific, A3160401), 0.1 mg/mL Penicillin Streptomycin (Gibco, 15140–122), 0.25 µg/mL Amphotericin B (Sigma, A2942), with (2 mM) L-Glutamine (Gibco, 25030–08). The inhibitors Cytochalasin-B (Abcam#ab143482), Aurora Kinase Inhibitor (Sigma#189410), and Riversine (Sigma#R3904) were dissolved in DMSO and used at the concentrations specified in the result section.

All Antisense Oligonucleotides (ASO) transfections were performed for 72 h using Lipofectamine RNAi Max reagent (Thermo Fisher Scientific, 13778075) as per manufacturer’s instructions. For RNAi, ASO were diluted in Opti-MEM (Thermo Fisher Scientific, 31985088). ASO against PURPL (5′-/52MOErG/*/i2MOErA/*/i2MOErA/*/i2MOErG/*/) from IDT and control ASO (5′-/52MOErT/*/i2MOErT/*/i2MOErC/*/i2MOErT/*/) from IDT were used at a final concentration of 100 nM.

Cell lysates were prepared using RIPA lysis buffer. Primary antibodies used were p21 (1:1,000; Cell Signaling Technology, 12D1), GAPDH (1:5,000; Invitrogen, MA5-15738) and p53 (1:1,000; Santa Cruz Biotechnology, SC-126Secondary antibodies (1:5,000 dilutions) were HRP-conjugated against rabbit (Cyvita, NA934V) and mouse (Cyvita, NA931V).

For immunostaining, cells were fixed with ice-cold methanol for 5 min. Fixation was followed by blocking with 1% BSA in PBS/0.1% Tween (PBST) for 30 min at room temperature. Coverslips with cells were incubated in DAPI for 1 h and then washed 3 times in PBST. Cells were then mounted on slides using Prolong gold antifade mountant (Thermofisher scientific-P36935).

Immunostained cells were imaged on Delta Vision Core system (Applied Precision/GE Healthcare, Issaquah, WA) consisting of Olympus IX70 inverted microscope (Olympus America, Inc., Melville, NY) with 100× NA oil immersion objective and a CoolSnap HQ 12-bit camera (Photometrics, Tucson, AZ) controlled by SoftWoRX software. Filters used for imaging were DAPI (Ex360/40; Em 457/50) of the 86000 Sedat Quadruple Filter Set (Chroma Technology Corp, Bellows Falls, VT). Z-stacks of at least 10 focal planes were acquired with an exposure of 0.1–0.5 s, depending on the filter. To prepare the figures, images were deconvolved with Softworx and scaled manually to 8-bit using Fiji and the same range of scaling for all the images.

Total RNA was extracted with TRIZOL reagent (Ambion, 15596018) and 1–2 µg of total RNA was reverse transcribed using Superscript III Reverse Transcriptase as per manufacturer instructions (Invitrogen, 18080093). PCR was performed on 1/10th volume (2 µL) of the cDNA using HotStar Taq Master Mix following manufacturer instructions (Qiagen, 203443).

He et al. (2018) found that senescence happened when there were more cells in a population with abnormal chromosome numbers, specifically aneuploid and near-polyploid cells. Furthermore, chromosome missegregation causes the formation of senescent cells (Santaguida et al., 2017). We were curious to investigate whether PURPL, a gene that has been found to be increased in colorectal cancer, which is characterized by a high number of abnormal chromosomes, is associated with aneuploidy. Casella et al. (2019) demonstrated that PURPL may serve as a more resilient indicator of senescence compared to p16 mRNA or p21 mRNA. This prompted us to examine the function of PURPL in chromosomal instability (CIN) or aneuploidy.

Various methods have been used to induce chromosome mis-segregation in cell culture. For example, compounds that interfere with microtubule dynamics or microtubule-kinetochore attachment cause a SAC-dependent delays in mitosis and induces chromosome mis-segregation (Foley and Kapoor, 2012). When cells are exposed to inhibitors of SAC function, proper alignment of the chromosomes to the spindle is hindered resulting in generation of aneuploid progeny (Santaguida et al., 2017). We examined hTERT immortalized RPE-1 cells grown in the presence or absence of reversine at 24 and 48 h. Reversine inhibits the SAC kinase Mps1 (Zhang et al., 2016). We investigated whether Reversine treatment promotes chromosomal segregation defects in RPE-1 cells. We examined the consequences of Reversine treatment on mitosis by analyzing the incidence of micronuclei. Chromosomal Instability manifests with a higher incidence of micronuclei that often results from DNA bridges due to defective chromosome segregation. Fixed and immunostained RPE-1 ^control and RPE-1 ^Reversine cells were assayed for micronuclei due to Reversine treatment and observed that RPE-1 ^Reversine cells showed a higher proportion of cells with micronuclei (Figure 1A) when compared to RPE-1 ^Control. This result indicated that Reversine treatment induces chromosomal instability. To measure the consequences of reversine on RPE-1, we did qRT-PCR to measure the gene changes associated with chromosome segregation defects. Further, we did RT -PCR and qRT-PCR to measure PURPL expression. Analysis of qRT-PCR data showed upregulation of p53 targets namely p21 and MDM2 at both 24 and 48 Hrs (Figure 1B; Supplementary Figures S1A, S1B). Once the p53 targets are validated, we measured the expression of PURPL through both RT-PCR (Figure 1C) and qRT-PCR (Figure 1B; Supplementary Figure S1C), both measurements showed increased expression of PURPL at both measured time points (Figure 1B; Supplementary Figure S1C). qRT-PCR analysis showed increased expression of PURPL along with p21 and MDM2 indicating the involvement of PURPL and p53 targets (p21 and MDM2) in chromosome segregation defects or CIN.

Cytochalasin B is a chemical agent which inhibits cytokinesis without preventing nuclear division (Minissi et al., 1999). Cytochalasin B was shown to induce polyploidy in several studies (Defendi and Stoker, 1973; Stanley et al., 1981). Cytokinesis failure induced in embryos through cytochalasin-B displayed segregation defects (Paim and FitzHarris, 2019). Tetraploidy perturbs metaphase chromosome alignment, kinetochore-microtubule establishment leading to CIN (Paim and FitzHarris, 2019). The other described mechanism by which tetraploidy leads to CIN in mammalian cells is that the acquisition of supernumerary centrioles/centrosomes leads to the formation of hazardous multipolar spindles resulting in segregation errors (Storchova and Kuffer, 2008). We examined hTERT immortalized RPE-1 cells grown in the presence or absence of Cytochalasin-B at 24 and 48 h. We investigated whether Cytochalasin B treatment promotes chromosomal instability defects in RPE-1 cells. We examined the consequences of Cytochalasin B treatment by analyzing the incidence of micronuclei. We scored for cells with the incidence of micronuclei. Fixed and immunostained RPE-1 ^control and RPE-1 ^{Cytochalasin−B} cells were assayed for incidence of micronuclei status due to cytochalasin-B treatment and observed that RPE-1 ^{Cytochalasin B} cells showed a higher proportion of cells with micronuclei (Figures 2A, B; Supplementary Figures S2A, S2B), further supporting the conclusion that Cytochalasin-B treatment contribute to CIN. Cell multinucleation is closely associated with chromosomal instability (Pantelias et al., 2019). Polyploidy cells were shown to have defective chromosome segregation resulting in aneuploidy (Potapova et al., 2013). Furthermore, polyploidy is also associated with high levels of chromosome instability (Mayer and Aguilera, 1990). Fixed and immunostained RPE-1 ^control and RPE-1 ^{Cytochalasin−B} cells were assayed for incidence of multinucleation due to cytochalasin-B treatment and observed that RPE-1 ^{Cytochalasin B} cells showed a higher proportion of multinucleated cells (Figures 2A, B; Supplementary Figures S2A, S2B), further supporting the conclusion that Cytochalasin-B treatment contributes to CIN. Treatment with Cytochalasin-B led to micronuclei formation and polyploidy generation in a time-dependent manner (Figure 2A; Supplementary Figure S2A). To measure the consequences of Cytochalasin-B on PURPL expression, we did qRT-PCR to measure PURPL, p21, and MDM2 expression. The qRT-PCR measurements showed increased expression of PURPL at both measured time points (Figure 2C; Supplementary Figure S2C). In addition, qRT-PCR showed increased expression of p21 and MDM2 at 48 h, but at 24 h, only MDM2 showed increased expression, but p21 expression did not change (Supplementary Figure S2D). These results indicated the involvement of PURPL and p53 targets (p21 and MDM2) in chromosome segregation defects or CIN.

In order to verify whether the increased expression of PURPL upon cytochalasin-B treatment is cell line-specific, we treated 293T cells with cytochalasin-B and measured PURPL expression. Chromosomal instability and aneuploidy are associated with cell cycle defects, resulting in an asynchronous cell cycle. We measured the cell cycle profile of 293T cells treated with or without cytochalasin-B through flow cytometry. 293T cells treated with cytochalasin-B resulted in an asynchronous cell cycle (Figure 2D lower panel) as compared to control cells (Figure 2D upper panel), further supporting the conclusion that cytochalasin-B treatment contributes to CIN, or aneuploidy. To measure the consequences of Cytochalasin-B on PURPL expression in 293T cells, we did qRT-PCR and RT-PCR to measure PURPL expression. The qRT-PCR measurements showed increased expression of PURPL at 72 h (Figure 2E). This result was further validated through increased expression of PURPL through RT-PCR (Figure 2F). These results indicated the involvement of PURPL in CIN or aneuploidy, and this phenomenon is not cell line-specific.

ZM447439 (“ZM”), the first Aurora family kinase inhibitor to be developed and characterized was previously found to interfere with the mitotic spindle integrity checkpoint and chromosome segregation (Gadea and Ruderman, 2005). When ZM was added to mammalian somatic tissue culture cells, the cells mitotic spindle was disorganized, chromosomes did not align properly, and cytokinesis was blocked (Gadea and Ruderman, 2005).

We examined the consequences of ZM447439 treatment by analyzing the incidence of micronuclei. We scored for cells with the incidence of micronuclei. Fixed and immunostained, RPE-1 ^control, and RPE-1 ^ZM447439 cells were assayed for incidence of micronuclei status due to ZM447439 treatment and observed that RPE-1 ^ZM447439 cells showed a higher proportion of cells with micronuclei (Figure 3A; Supplementary Figure S3A), further supporting the conclusion that ZM447439 treatment contributes to CIN. Abnormal nuclear shape and chromosomal instability are closely related (Gisselsson et al., 2001). A linear and positive correlation was observed between the abnormal nuclear shape on one hand and cells with chromosomal instability feature like unstable chromosomes and anaphase bridges on the other hand (Gisselsson et al., 2001). Fixed and immunostained RPE-1 ^control, and RPE-1 ^ZM447439 cells were assayed for incidence of abnormal nuclear shape due to ZM447439 treatment and observed that RPE-1 ^ZM447439 cells showed a higher proportion of abnormal nuclear shape (Figures 3A, B; Supplementary Figures S3A, S3B), further supporting the conclusion that ZM447439 treatment contributes to CIN. Treatment with ZM447439 showed chromosome instability features in a time-dependent manner (Figure 3A; Supplementary Figure S3A). To measure the consequences of ZM447439 on RPE-1, we did qRT-PCR to measure PURPL, p21 and MDM2 expression, indeed qRT-PCR showed increased expression of p21 and MDM2 at both 24 and 48 h (Supplementary Figures S3B, S3C). Once the p53 targets are validated, we measured the expression of PURPL through qRT-PCR at both measured time points. qRT-PCR analysis showed increased expression of PURPL (Figure 3C; Supplementary Figure S3B) indicating the involvement of PURPL and p53 targets (p21 and MDM2) in CIN or aneuploidy.

In order to verify whether the increased expression of PURPL upon ZM447439 treatment is cell line-specific, we treated MDA-MB-231 cells with ZM447439 and measured PURPL expression. Chromosomal instability and aneuploidy are associated with cell cycle defects, resulting in an asynchronous cell cycle. We measured the cell cycle profile of MDA-MB-231 cells treated with or without ZM447439 through flow cytometry. MDA-MB-231 cells treated with cytochalasin-B resulted in an asynchronous cell cycle (Figure 3D lower panel) as compared to control cells (Figure 3D upper panel), further supporting the conclusion that ZM447439 treatment contributes to CIN, or aneuploidy. To measure the consequences of ZM447439 on PURPL expression in MDA-MB-231 cells, we did qRT-PCR and RT-PCR to measure PURPL expression. The qRT-PCR measurements showed increased expression of PURPL at 72 h (Figure 3E). This result was further validated through increased expression of PURPL through RT-PCR (Figure 3F). These results indicated the involvement of PURPL in chromosome CIN or aneuploidy, and this phenomenon is not cell line-specific.

PURPL was shown to be upregulated upon DOXO treatment in a p53-dependent manner (Li et al., 2017). p53-dependent upregulation of PURPL was further strengthened by qRT-PCR upon genetic loss of p53 or upon p53 knockdown in HCT116 in the presence or absence of DOXO (Li et al., 2017). PURPL, an abundant lncRNA upregulated after DOXO treatment was abolished upon genetic loss of 53 or upon p53 knockdown (Li et al., 2017). In order to look at the role of p53 in the regulation of PURPL expression due to CIN, we looked at the DLD1 cells. DLD1 is a pseudodiploid human cell line with p53 mutation (Galofré et al., 2020). The p53 protein produced in DLD1 cells has a point mutation at position 241 (C -> T a mutation resulting in Ser -> Phe). We examined DLD1 cells grown in the presence or absence of reversine at 48 and 72 h. We did RT-PCR and qRT-PCR to measure PURPL expression, indeed RT-PCR (Figure 4A) and qRT-PCR (Figures 4B, C) showed increased expression of PURPL at both 48 and 72 h (Figures 4A–C). Once the PURPL exression was validated, we measured the expression of p53 targets through qRT-PCR. qRT-PCR (Figures 4D–G) measurement showed increased expression of p21 and MDM2 at both measured time points. The magnitude of PURPL, p21, and MDM2 expression was less than that observed in the case of RPE-1 cells treated with Reversine (Figures 4, 1B; Supplementary Figure S1). The decreased expression of PURPL along with p21 and MDM2 in DLD1 cells indicate the involvement of p53 in PURPL expression in CIN or aneuploidy.

To validate the p53-dependent upregulation of PURPL by reversine through qRT-PCR, RPE-1 cells were treated with reversine, in the presence or absence of Pifithrin-α. Pifithrin-α is a reversible inhibitor of p53 (Zhu et al., 2020). Reversine treatment resulted in increased expression of PURPL in absence of a p53 inhibitor (Figures 5A, B; Supplementary Figures S4A, S4B; Supplementary Figures S4C, S4D). Furthermore, in presence of Pifithrin-α (50 uM), reversine treatment resulted in a weakened response of increased expression of PURPL (Figures 5A, B; Supplementary Figures S4C, S4D).

The increased expression of PURPL did not decrease much in the presence of 10 μM and 20 µM of Pifithrin-α (Supplementary Figures S4A, S4B). Furthermore, the increased transcriptional activation of p21 upon reversine treatment did not show any change in the expression with and without Pifithrin-α (50 uM) (Figure 5C). In addition, qRT-PCR analysis showed weakened response of increased expression of MDM2 in presence of reversine along with Pifithrin-α (50 uM) (Figure 5D). These results indicated that the strength or magnitude of increased expression of PURPL upon reversine treatment is dependent upon p53 and its target MDM2.

To see if the lower level of PURPL in cells treated with reversine after p53 knockdown was because of PFT-α′s effect on proliferation, we looked at how PFT-α affected proliferation. 10 μM of PFT-α did not show any significant effect on proliferation (Figure 5E). 20 µM and 50 µM of PFT-α showed some effect on proliferation (Figure 5E), but this could not be corroborated with the effect on PURPL expression upon reversine treatment. At both 20 µM and 50 µM of PFT-α treatment, there was a similar reduced effect on proliferation (Figure 5E), but the effect on PURPL expression was greater for 50 µM of PFT-α as compared to 20 µM (Supplementary Figures S4A, S4C). The results (Figure 5; Supplementary Figure S4) did not show a direct link between the lower levels of PURPL after reversine treatment for cells that had been treated with 50 µM of PFT-α and the lower cell growth that 50 µM of PFT-α caused.

To elucidate potential functions of PURPL, we knockdown PURPL with GAPMERS and looked for its role on chromosomal instability in RPE-1 cells. The efficiency of the PURPL knockdown was checked through qRT-PCR. Use of PURPL GAPMERS resulted in almost 80% knockdown of PURPL as compared to control (Figure 6A). We examined the consequences of PURPL knockdown by analyzing the incidence of micronuclei and deformed nuclei. We scored for cells with defective segregation as depicted by the presence of deformed nucleus and incidence of micronuclei. Fixed and immunostained RPE-1 GAPMERS ^Control, and RPE-1 ^{GAPMERS PURPL} cells were assayed for micronuclei due to PURPL knockdown and observed that RPE-1 ^{GAPMERS PURPL} cells displayed a significant increase in the proportion of cells with micronuclei as compared to RPE-1 GAPMERS ^Control (Figure 6B). Hence, we concluded that PURPL knockdown correlates with defects in chromosome segregation as shown by an increased incidence of micronuclei. In addition to the incidence of micronuclei, CIN also manifests abnormal nuclear shape (Gisselsson et al., 2001). A positive correlation was observed between the abnormal nuclear shape and chromosomal instability features like unstable chromosomes and anaphase bridges (Gisselsson et al., 2001). Fixed and immunostained RPE-1 GAPMERS ^Control and RPE-1 ^{GAPMERS PURPL} cells were assayed for nuclear shape due to downregulation of PURPL and observed that RPE-1 ^{GAPMERS PURPL} cells displayed a significant increase in the proportion of cells with a deformed nucleus (Figure 6C). Cells with PURPL knockdown, developed a higher proportion of misshapen, multilobed nuclei, and contained micronuclei. These morphological abnormalities were phenocopies of siRNA-mediated reduction in centromeric proteins itself that drives chromosome mis-segregation events underlying the interphase nuclear defects. The nuclei of normal cells are normally ellipsoid shapes with smooth outlines (Gisselsson et al., 2001). Deformed nuclei are characteristics of many cancer cells where they are easily identifiable by increased nuclear size, irregular nuclear contours, and disturbed chromatin distribution, making nuclear morphology one of the oldest and most commonly used cancer markers (Gisselsson et al., 2001; De las Heras et al., 2013; Denais and Lammerding, 2014). The irregular nuclear outline in cancer cells is mainly the result of grooving, convolutions, and invaginations of the nuclear envelope (Van Bortle and Corces, 2012; De las Heras et al., 2013; Denais and Lammerding, 2014). These data indicate that PURPL loss results in perturbing nuclear architecture ultimately leading to chromosomal instability with the potential to play an important role in cancer. Furthermore, these results indicated that proper levels of PURPL expression are critical for organized nuclear architecture. These studies indicate that expression levels of PURPL must be precisely regulated for maintaining genomic instability.

When we knocked down PURPL, we observed a change in the nucleus’ shape and an increase in micronuclei, prompting us to investigate its role in CIN/aneuploidy. We transiently knocked down the expression of PURPL in RPE-1 cells (Figures 7–9) and then compared how well they did when treated with CIN or aneuploidy inducers like reversine, cytochalasin-B, or ZM447439 (Figures 7–9).

To cause aneuploidy, cells with or without PURPL knockdown were given 1 µM Reversine, 2.5 µM Cytochalasin-B, and 10 µM ZM447439 for 24 h. PURPL knockdown cells showed morphological changes as compared to control cells. Cells with PURPL knockdown were more elongated and stressed as compared to control cells upon reversine treatment, showing that reversine is inducing more changes in PURPL knockdown cells (Figures 7A, B). In addition, the cells are more stressed, with clear morphological changes following cytochalasin-B treatment (Figures 8A, B). Treatment with ZM447439 also induces more morphological changes in PURPL knockdown cells (Figures 9A, B). PURPL knockdown cells appear more elongated and stressed (Figures 9A, B).

Studies have shown that cells with chromosome instability or aneuploidy have more S-phase (Look et al., 1982; Gemble et al., 2022) or sub-G0 phase cells (Santaguida et al., 2017) in the cell cycle compared to control cells. FACS is a quantitative way to measure cells in different phases of the cell cycle. The flow cytometry analysis of PURPL knockdown cells showed that they had more Sub G0 (8.9% vs. 6.1%) and S-Phase (25.1% vs. 17.4%) cells than control cells upon reversine treatment (Figure 7C). Cytochalasin-B treatment increased the proportion of S-phase cells to 21% from 10.2% in comparison to control cells (Figure 8C). Once ZM447439 was added to PURPL knockdown cells, there were more cells in Sub G0 (11% vs. 4.5%) and S-Phase (22.6% vs. 12.4%) (Figure 9C). These results clearly indicate that PURPL knockdown cells are more vulnerable to aneuploidy as compared to control cells.

We checked the level of p21 due to its association with chromosomal instability or aneuploidy (Kumari et al., 2014). Once we treated PURPL knockdown cells with cytochalasin-B (Figure 8D) and ZM447439 (Figure 9D), we saw that they expressed more p21 than control cells. These results clearly indicate that PURPL knockdown cells are more susceptible to chromosomal instability, or aneuploidy, as compared to control cells.

In order to find the molecular mechanism by which PURPL can regulate nuclear morphology or chromosomal stability, we focused on MDM2 for the following reasons: The MDM2 gene was subsequently found to be amplified in cancer (Oliner et al., 2016). MDM2 is a negative regulator of the tumor suppressor p53 (Momand et al., 2000). Elevated MDM2 expression was shown to induce chromosomal instability in aging mice (Lushnikova et al., 2011). In addition, it was also shown to induce chromosomal instability when overexpressed in B cells (Wang et al., 2007). The capability of overexpressed MDM2 to affect genomic instability can also happen independently of p53 (Bohlman and Manfredi, 2014). PURPL was shown to be regulated by p53 and, at the same time, regulate the expression of p53 (Li et al., 2017). Since both PURPL and MDM2 are shown to be regulated by p53 and, in turn, they regulate the expression of p53, we looked at the regulation of MDM2 by PURPL. We looked at the expression of MDM2 upon transient knockdown of PURPL using GAPMERS in RPE-1 cells (Supplementary Figure S5A). Knockdown of PURPL using GAPMERS resulted in increased cell death in RPE-1 cells (Supplementary Figure S5B). qRT-PCR analysis showed knockdown of PURPL resulted in increased expression of MDM2 (Supplementary Figure S5C). This result suggested that PURPL regulates the expression of MDM2. Further, we used a bioinformatics tool called catRAPID to predict MDM2 and PURPL interactions (Agostini et al., 2013). catRAPID predicted an interaction propensity of 0.94 between PURPL and MDM2, suggesting an average probability of interaction between PURPL and MDM2 (Supplementary Figure S5D). The interaction profile of the nucleotide position of long noncoding RNA PURPL with MDM2 showed that the nucleotide position from 400 to 1,000 nucleotides showed more probability of interaction with MDM2 as compared to the first 400 nucleotides (Supplementary Figure S5E). In addition, an interaction matrix involving long noncoding RNA PURPL nucleotide positions and individual protein residue fragments of MDM2 indicated that MDM2 fragments from 100 to 450 amino acid residues have a greater chance of interacting with 400–1,000 nucleotides of PURPL (Supplementary Figure S5F). These bioinformatics analyses indicate that PURPL may interact with MDM2 directly, although more extensive biochemical characterization is required to confirm this. Through probable interactions with MDM2, PURPL may regulate MDM2 expression. The knockdown of PURPL led to an increase in MDM2 expression, potentially explaining the observed increase in CIN or aneuploidy and subsequent cell death.