Excessive MYC-topoisome activity triggers acute DNA-damage, MYC-degradation and replacement by a p53-topoisome

Summary

Hyperproliferation driven by the protooncogene MYC may lead to tumor suppressor p53 activating DNA-damage that has been presumed to derive from hypertranscription and over-replication. Here, we report that excessive MYC-topoisome (MYC/Topoisomerase 1/Topoisomerase 2) activity acutely damages DNA activating pATM and p53. In turn MYC is shutoff and degraded releasing TOP1 and TOP2A from MYC-topoisomes in vitro and in vivo. To manage the topological and torsional stress generated at its target genes, p53 assembles a separate topoisome. Because topoisomerase activity is intrinsically DNA-damaging, p53-topoisomes provoke an initial burst of DNA-damage. Because p53 unlike MYC, upregulates the DNA-damage-response (DDR) and activates TDP1 and TDP2, it suppresses further topoisome-mediated damage. The physical coupling and activation of TOP1 and TOP2 by p53 creates a tool that supports p53-target expression while braking MYC-driven proliferation in mammalian cells.

Graphical Abstract

eTOC

Das et al. use biochemical and genomic methods to show that MYC-topoisome overactivity directly and acutely damages DNA triggering MYC degradation, p53 induction, and p53-TOP1-TOP2 topoisome formation, that in turn increases p53-target gene expression including TDP1 and TDP2 that help repair topoisomerase-mediated DNA damage.

Introduction

The clash between the MYC oncogene and the p53 tumor suppressor pervades oncogenesis^1–4. Physiologically, MYC amplifies transcription^5–8, activates DNA replication^9,10, and drives most cellular processes^11–14; sustained MYC overexpression supports the pathogenesis of most human cancers¹⁵. p53 activates DNA repair¹⁶, cell cycle arrest, senescence, and apoptosis while suppressing genes supporting tumorigenesis and tumor progression¹⁷, acting through transcriptional and post-transcriptional mechanisms. MYC and p53 target many of the same pathways promoting or opposing proliferation, cell growth, and anabolic metabolism^18,19. Enforced high MYC levels provoke genome instability and consequent p53-dependent apoptosis²⁰, that protects against oncogenic transformation^2,21. How the p53-system recognizes MYC-overexpression has not been fully resolved. One possibility is that MYC, a potent mitogen, triggers damage at mechanically and enzymatically stressed sites of DNA-synthesis⁹. A second notion posits that MYC directly overdrives transcription via direct effects on the transcription apparatus and chromatin^22,23 as well as indirectly via the products of MYC- upregulated mRNAs. This transcriptional overdrive creates transcriptional stress resulting in R-loops and other DNA chromatin states susceptible to mechanically or enzymatically induced DNA damage^24,25. Collisions between elongating RNA polymerases and replication forks may also generate mechanical and enzymatic DNA-damage^26,27. For example, replication through transcription-triggered single-strand breaks converts these nicks into double-strand breaks. In all these instances, MYC is an indirect instigator of DNA-damage²⁸. When sufficient damage accumulates, p53 is triggered and cell-cycle arrest, DNA repair, senescence and apoptosis ensue²⁸. A direct coupling between p53 and MYC has not been found.

MYC amplifies transcription^5,11 and stimulates replication⁹ via a large panel of factors and complexes. These processes generate torsional stress upstream (negative supercoils) and downstream (positive supercoils) of translocating transcription or replication machineries^29–32. MYC-guided chromatin transactions pump high levels of negative torsional stress into DNA fibers by releasing nucleosomes^31,33,34, or in the reverse reaction, spooling negative supercoils upon nucleosome³¹. For MYC-driven processes to proceed on-schedule and unimpeded, supercoiling must be properly managed. DNA topoisomerases (TOP1, TOP2A and TOP2B, TOP3A and TOP3B) remove dangerous levels of torsional stress, regulate conformational dynamics of DNA and chromatin, and maintain genome integrity^31,35. MYC assembles a “topoisome”³⁶ that supercharges the catalytic activities of TOP1 and TOP2 to better handle supercoils generated by genomic transactions. Two MYC-topoisomes have been found, topoisome A (TOP1, TOP2A and MYC) and topoisome B (TOP1, TOP2B and MYCN)¹⁵.

Single or double-strand DNA breaks are obligatory for topoisomerase catalysis. Failure to reseal even a miniscule fraction of these transient breaks by chance or through chemical or mechanical interference, could yield potentially lethal DNA-damage^37,38,39. As TOP1 and TOP2 reaction rates are increased ~5–10 times in topoisomes, their propensity for collateral genome-damage would be predicted to be much greater than free TOP1 or TOP2⁴⁰. MYC also intensifies the toxicity of DNA damaging agents such as UV or IR radiation due to transcription and replication defects^21,41,42. Whether MYC itself might inflict acute topoisome-mediated genotoxicity on cells, has not been explored.

In response to acute genomic stress¹⁷, p53 is stabilized and promotes cell-cycle arrest⁴³, DNA repair⁴⁴, apoptosis, and senescence⁴⁵. It attenuates global gene expression by downregulating MYC while upregulating several hundred p53-targets that execute its cellular programs^46,47. With MYC downregulated, we wondered how torsional and topological issues are managed at p53 targets. Previous immunoprecipitation experiments and topoisomerase assays suggested interaction of p53 with TOP1⁴⁸ or TOP2⁴⁹ but did not elaborate their mechanism or relevance.

We find that with supraphysiological MYC, topoisome activity becomes acutely genotoxic and p53-activating with p21 arresting MYC-driven proliferation^50–52. A“p53-topoisome” composed of p53 ,TOP1 and TOP2A, in which both enzymes are stimulated, forms to support p53-target gene expression while MYC-topoisomes decline. The conversion from MYC- to p53-topoisome predominance is mass-action driven. p53-topoisomes are less genotoxic than MYC’s because p53 sponsors DDR that repairs basal and topoisome-mediated damage using increased levels of TDP1 and TDP2. RNA-seq analysis suggests that p53-topoisomes enhance p53-target expression. p53-topoisomes are enriched at transcriptionally upregulated targets during DDR. By directly monitoring MYC-topoisome mediated DNA-damage moment-to-moment, and if excessive, replacing MYC-topoisomes with p53-topoisomes, p53 recognizes and antagonizes MYC-mediated genome instability. In the absence of p53, high MYC-topoisome levels may increase MYC’s oncogenicity early in carcinogenesis.

Results

MYC directly induces DNA-damage and apoptosis

Pathologically high transcription in cancer is associated with DNA-damage⁵³ yet the mechanisms provoking this damage remain incompletely and imprecisely described. When overexpressed, MYC enforces high global transcription and flux through all cellular components^54,55. Supra-physiological MYC drives replication and transcription to levels generating intrinsically DNA-damaging mechanical and enzymatic stress^25,56. Such damage is downstream and delayed from the immediate actions of MYC. Upon expression, MYC rapidly forms a TOP1-TOP2-MYC topoisomerase superactivating complex (topoisome)³⁶. Because all topoisomerases transiently break one or both strands of DNA⁵⁷, and because no biochemical reaction is 100% efficient, we wondered whether MYC-topoisomes might intrinsically damage DNA.

To assemble MYC-topoisomes and monitor DNA-damage before MYC-driven macromolecular synthesis ramped up, we examined HO15.19-MYCER12 cells⁵⁸ in the absence or presence of tamoxifen (tam), scoring the number and intensity of γH2AX foci, a hallmark of DNA-damage⁵⁹. Lacking endogenous MYC, MYCER12 constituively expresses a retrovirally encoded MYC-ER chimera, that upon tam treatment immediately translocates from the cytoplasm into the nucleus⁵⁸. Cells were briefly pulsed with tam in the presence or absence of TOP2 poison etoposide (Eto) or TOP1 poison camptothecin (CPT), and checked for γH2AX using confocal microscopy (Figure 1A and Figure S1B) or western blotting (Figure S1A and S1D). Strikingly, within 100 minutes, tam alone, elevated γH2AX ~30-fold over background to levels comparable with Eto or CPT treatments that dramatically increased double or single-strand breaks in untreated (vehicle only) cells by inhibiting TOP2 or TOP1 (Figure 1A and Figure S1A–D, Figure 1B and Figure S1C). Seen in all cells, this damage was cell-cycle independent.. Notably, the tam-induced DNA-damage required MYC-ER (Figure S1F–G). Additionally, γH2AX levels were ~3-fold further elevated upon treatment with tam+Eto (Figure 1A, Figure 1B). Tam-activated MYC-ER also exaggerated the DNA-damage due to the TOP1 CPT (Figure S1A–C). The rapidity (within 1hr, Figure S1H–I) of the tam-induced damage, and its escalation by TOP1 or TOP2 poisons argued that MYC - mediated genotoxicity precedes upregulation of transcription or replication. Indeed, tam-triggered DNA-damage was not blunted by pretreatment with triptolide, a TFIIH inhibitor, and therefore was transcription independent (Figure S1H–I). Triptolide pretreatment slightly augmented baseline DNA-damage most likely due to nucleotide excision repair (NER) inhibition by triptolide as the TFIIH p89/XPB helicase is essential for both transcription and NER⁶⁰.

Figure 1. The MYC-topoisome is directly DNA-damaging and triggers apoptosis.

(A) Immunofluorescence microscopy of tamoxifen (tam)-treated or untreated HO15.19-MYC-ER cells treated with Etoposide (25 μM; 20 min), and with or without pretreatent with ICRF-187 (10 μM; 1hr). γH2AX foci are in red. Nuclei were visualized with DAPI (scale bars, 25 μm). (B) Quantification of γH2AX intensity per nucleus were calculated for 100 cells (calculated value ± S.E.M.). Asterisks denote significant difference (****P<0.0001; t-test). (C) Quantification of percentage of DNA in the COMET tail for drug treatment as in (A) for 100 cells (calculated value ± S.E.M.). (**P<0.01; t-test). (D) Quantification of percentage of the total apoptotic+dead cells for asynchronous and synchronous HO15.19-MYC-ER cells upon tam (200 nM) treatment for the specified time with and without serum. (E) Quantification of cell-cycle analysis of synchronized HO15.19-MYC-ER cells treated or untreated with tam (200 nM) for indicated hours without serum and analyzed by flow cytometry. (F) Scheme of tam treament (200 nM; 1–8 hrs) for HO15.19-MYC-ER cells and the indicated proteins were checked by immunoblot. Actin served as a loading control. Protein molecular weight (kDa) are indicated at right. See also Figure S1.

In “topoisomes”, MYC or MYC-MAX stimulate TOP1 and TOP2, in vitro and in vivo³⁶. To investigate whether MYC-aggravated DNA-damage was topoisomerase dependent, we utilized ICRF-187 which binds and inhibits TOP2 prior to covalent enzyme-DNA adduct (TOPcc) formation⁶¹ thereby protecting DNA from TOP2-mediated damage. Cells were pretreated with ICRF-187, then with or without Eto or tam, and checked for γH2AX (Figure 1A). ICRF-187 decreased tam-induced H2AX-foci ~2 fold relative to tam-alone (Figure 1A, Figure 1B). Similarly, the intensity of tam + Eto generated γ foci decreased ~3 fold when cells were preincubated with ICRF-187 prior to Eto addition (Figure 1A, Figure 1B, These results suggest that intrinsic MYC-topoisome activity triggers DNA-damage.

To confirm that MYC acutely breaks DNA strands, alkaline comet assays⁶² (Figure 1C) were performed on MYC-ER cells treated with tam, Eto and ICRF-187 as in Figure 1A, then evaluated for broken DNA tail formation. . Tam-treated cells exhibit ~15-fold more DNA breaks than untreated cells; this breakage was exacerbated with Eto (Figure 1C). Remarkably, ICRF-187 preincubation protected cells from tam-induced DNA breaks, with or without Eto indicating that MYC-topoisomes damage DNA.

To observe cell death due to supraphysiological MYC, MYC-ER cells were treated with tam for 4–24 hours, stained for Annexin V(+) and with propidium iodide (PI) and examined by flow cytometry to reveal apoptotic and dead cells^63,64. With tam, the fraction of apoptotic+dead cells tripled after 12 hours (Figure 1D, Asynchronous panel). Therefore, MYC-directed genomic instability acutely provoked apoptosis. To observe the consequence of MYC-induced DNA-damage throughout the cell-cycle, cells were arrested in G0 by serum starvation, then reactivated with tam in the presence or absence of serum, and monitored for apoptosis/cell death (Figure 1D) and for cell-cycle phase (Figure 1E) by flow cytometery. The population of apoptotic+dead cells tripled with 4 hrs of tam treatment, well before S-phase entry, but continued to increase over the next 2 days as more cells progressed into the cell-cycle (Figure 1D, Figure 1E). Cell-death following acute MYC-topoisome activation was of the same magnitude as that attributed to S-phase replication stress. Tamoxifen induction of apoptotic and dead cells prior to S-phase suggests an earlier insult than replication-stress. Besides cell-death and apoptosis, tam-activated MYC-ERER augmented senescence (Figure S1J).

MYC-topoisome mediated DNA-damage activates the p53 response

MYC-induced, transcription-independent DNA-damage was surprisingly rapid. Because DNA-damage induces p53, which then represses MYC transcription¹⁹, we expected this damage to decline over several hours. To attenuate topoisomerase mediated-damage, more rapidly, MYC-toposome activity or integrity might be downregulated. To monitor the response to MYC-driven damage, MYC-ER cells were treated with tam for 1–8 hours and the levels of MYC-ER and several DDR factors (pATM, p53, p21, p15/p16) were tracked. First, though tam allosterically activates MYC-ER post-translationally, it initially also increased MYC-ER levels, most likely reflecting MYC-ER amplification of its own constitutive transcription from a retroviral LTR-promoter⁵⁸. pATM, a very early player in the DDR, increased upon tam treatment, suggesting ongoing DNA-damage (Figure 1F, Top lane). MYC, as expected⁵⁰, initially blunted p21-induction by p53, but as DNA-damage increased, p53 accrued sufficiently to activate p21 (Figure 1F). Downregulation of MYC-ER ensued with longer tam exposure, coincident with increasing p53. The molecular antagonism between MYC and p53 parallels their respective oncogenic and tumor suppressive roles.

DNA-damage dissociates the MYC-topoisome complex

The loss of MYC-ER would release TOP1 and TOP2A from the MYC-topoisome (Figure 2). To evaluate the integrity of MYC-topoisomes under conditions of severe genomic stress, HCT116 cells were treated with Neocarzinostatin (NCS, a DNA-damaging radiomimetic anticancer drug s) in the presence or absence of ATM inhibitor KU-55933⁶⁵ (Figure S2B). HCT116 and HCT116 p53^−/−cells are a well-characterized model to study p53 where native MYC is expressed from its endogenous locus¹⁹. Cellular extracts immunoprecipitated using antibodies against MYC, TOP1, or TOP2A. Upon NCS treatment, interactions of MYC with TOP1 or TOP2A noticeably diminished, as evinced by reduced pull-down of both topoisomerases with anti-MYC (Figure 2A, left panel). Reciprocally, co-immunoprecipitation of MYC by anti-TOP1 or anti-TOP2A also decreased (Figure 2A, middle and right panels). NCS treatment rapidly and dramatically depressed MYC levels (Figure 2A, middle and right Input MYC panels and Figure S2B). The dissociation of MYC from topoisomes following DNA-damage occurred both in wild-type and p53^−/− cells (Figure 2B, Left panel; lanes 2 and 4). Treatment with ATM inhibitor preserved the association of MYC with TOP1 and TOP2A and preserved MYC levels despite NCS treatment for many hours (Figure 2A and Figure S2B). Thus, NCS-triggered pATM apparently participates in the precipitous decline of cellular MYC. Curiously, despite MYC loss and dissociation from topoisomerases, following DNA-damage TOP1 and TOP2 remained associated with each other in pulldowns. Either TOP1 and TOP2 had become directly associated or else they were held together by a new partner, perhaps p53.

Figure 2. DNA-damage disrupts MYC-topoisome complex and promotes formation of p53-topoisome.

(A) Scheme of treatment at top. Nuclear extracts of HCT116 Cells with indicated treatments were immunoprecipitated (IP) as mentioned and probed for indicated proteins. (B) Wild-type (WT) and p53-knockout (p53^−/−) HCT116 cells were treated with Neocarzinostatin (NCS) (400 ng/ml; 3hr) and nuclear extreacts were IP as depicted and anlyzed for indicated proteins. Recombinant p53 protein (Rec.p53) was utlized to supplement p53^−/− cellular extract (NCS treated) to mimic wild-type. (C) Same nuclear extracts as in (B) were IP as shown and examined for mentioned proteins. See also Figure S2.

DNA-damage switches topoisomes from MYC to p53

p53 induction by MYC-topoisome instigated DNA-damage would immediately couple the oncogene’s action with the tumor suppressor’s response. p53 safeguards the genome by activating repair genes, enforcing cell-cycle arrest and apoptosis, and repressing oncogenes⁶⁶. The upregulated transcription of p53-target genes must inevitably generate supercoils that oppose high expression^22,30,31,35. The dissolution of the MYC-topoisome upon DNA-damage eliminates an effective tool to remediate topological stress. Lacking MYC-topoisomes, p53-targets must either tolerate suboptimal transcription mechanics or control torsion differently, perhaps with a p53-topoisome. Though previous studies have reported interactions of p53 with TOP1 or with TOP2 in vitro, the biological and mechanistic implications of these observations remain unexplored. To provoke the hypothesized replacement of MYC-topoisomes by p53-topoisomes, wild-type and p53-knockout (p53^−/−) HCT116 cells were NCS-treated and cellular extracts were immunoprecipitated using anti-MYC or anti-p53. Robust association of p53 with either TOP1 or TOP2 rapidly followed NCS treatment (Figure 2B, right part, lane 2), coincident with MYC-topoisome decay (Figure 2B, left part, lane 2 and 4). Moreover, supplementing p53^−/− cellular extract (NCS treated) with wild-type levels of recombinant p53 promoted association of p53 with topoisomerases (Figure 2B, right part, lane 5). Hence, association of cellular p53 with topoisomerases required no post-translational modifications or upregulation of other p53-targets. Thus, p53 and MYC exhibit contrasting behaviors in response to DNA-damage, MYC-topoisomerase associations waned, when p53-topoisomerase associations waxed. To investigate p53 bridging of TOP1 and TOP2A in cells, WT and p53^−/− HCT116 cells were treated with NCS, cellular lysates were immunoprecipitated using anti-TOP2A (Figure 2C, left panel) or anti-TOP1 (Figure 2C, right panel) and examined by western blot. TOP1 co-precipitated with TOP2A similarly from extracts of wild-type cells in the presence or absence of NCS. However, pull down of TOP1 by anti-TOP2A was highly attenuated using extracts from NCS-treated p53^−/− cells (Figure 2C, left panel lane 4) or vice-versa for pulldown of TOP2A by anti-TOP1 (Figure 2C, right panel lane 4). The association of TOP1 with TOP2A was rescued when p53^−/− NCS-treated cell extracts were supplemented with recombinant p53 to wild-type levels (Figure 2C, right and left panel lane 5). To confirm p53-topoisome formation in another non-transformed cell-line, WT (untransfected), control (Ctr siRNA transfected) and p53-knockdown (p53 siRNA transfected) MCF10A cells were treated with NCS, and cellular lysates were immunoprecipitated using anti-TOP2A (Figure S2C) or anti-TOP1 (Figure S2D), and examined by western blot. TOP1 association with TOP2A dramatically declined after p53-knockdown cells were treated with NCS (Figure S2C, lane 6, TOP1 panel and Figure S2D, lane 6, TOP2A panel) suggesting that p53-topoisome formation is a general phenomenon. Hence, upon DNA-damage, MYC levels declined releasing TOP1 and TOP2 to form a complex with p53.

p53 assembles a “topoisome” complex that boosts topoisomerase activity

The experiments above show that p53 forms its own “topoisome” in cells with TOP1 and TOP2A.s. To test whether p53, TOP1 and TOP2A were sufficient to form a p53-topoisome, recombinant TOP1 and TOP2A were mixed together in the presence or absence of recombinant p53 versus MYC and then immunoprecipitated using anti-TOP2A or anti-TOP1. Immunprecipitates were examined by western blot for each protein. TOP1 and TOP2A required p53 or MYC to co-precipitate (Figure 3A–B), suggesting p53 forms a MYC-like topoisome. To size the reconstituted p53-topoisome and assess its components stoichiometry, recombinant TOP1, TOP2A and p53 were combined and fractionated by size exclusion chromatography (SEC, Superose 6). Fractions were analyzed via western blotting for coelution of TOP2A, TOP1, and p53. TOP1, TOP2A, and p53 were identified across fractions 13 to 24, with TOP1 and TOP2A peaking in fractions 14 to 18 consistent with a complex of ~450 to 750 kDa, according to size markers (Figure S3A), confirming the existence of a high molecular p53-topoisome. Although SEC resolution was inadequate to define the exact topoisome stoichiometry, if native tetrameric p53, TOP1, and TOP2 combined in a 1–1-1 ratio, the complex would be ~650 kDa. We propose that the most complex most likelyto includes one TOP2 dimer (~350 kDa), one TOP1 (~90 kDa), and a p53 tetramer (~210 kDa), although the broadness of the peak allows for heterogeneity.

Figure 3. p53 assembles a “topoisome” complex and stimulates topoisomerase activity.

(A) Recombinant proteins were mixed as indicated and immunoprecipitated (IP) with either anti-TOP1 or (B) anti-TOP2A and probed for TOP1 and TOP2A. (C) Supercoiled plasmid was added to TOP1 alone or pre-incubated with increasing p53 and assessed by gel electrophoresis. Chl=chloroquine. (D) Before addition of kDNA, TOP2A or (E) TOP2B was pre-incubated with or without increasing p53 and decatenation activity was examined by gel electrophoresis. (F) TOP1 relaxation assay using lysates of indicated cells with or without NCS (400 ng/ml; 3hr) treatment. (G) Quantification of percentage of DNA relaxation by TOP1 in 3F. (H) TOP2 decatenation assay using same cellular lysates as in (F). (I) decatenation by TOP2 in Figure 3H. Error bars represent the mean ± S.E.M. See also Figure S3.

To explore p53’s influence on topoisomerase activity, TOP1 activity was monitored using changes in the electrophoretic migration of a supercoiled DNA plasmid⁶⁷ While TOP2A assays followed the unlinking (decatenation) of double-stranded DNA minicircles from a meshwork of interlinked circles (kinetoplast DNA from Crithidia fasciculata)⁴⁹. p53 greatly enhanced plasmid DNA relaxation by TOP1 (see increase in the intensity of supercoiled DNA band, Figure 3C, Figure S3B) and DNA decatenation by TOP2A (note the nicked or intact minicircle in the middle and/or bottom of the gel, Figure 3D, Figure S3C). Decatenation assays demonstrated that p53 also stimulated TOP2B.(Figure 3E, Figure S3D). It is noteworthy that whereas MYC exhibits specificity for TOP2A, and MYCN prefers TOP2B, because p53 strongly activated both TOP2A and TOP2B, p53-topoisome formation would be enabled by DNA-damage irrespective of the prevailing MYC or TOP2 in any particular cell.

p53 specifically stimulates topoisomerase activity ex vivo

Demonstrating that p53 hyperactivates TOP1 and TOP2 activity in vitro, we next asked if p53 enhanced topoisomerase activity in a more complex environment; TOP1 relaxation and TOP2 decatenation assays were conducted using extracts of wild-type and p53^−/− HCT116 cells treated or not with NCS. TOP1 and TOP2 activites were increased by NCS treatment of wild-type HCT116 cells (Figure 3F & 3H, Figure 3G & 3I, respectively), whereas in p53^−/− cells, both TOP1 and TOP2 activity was diminshed (Figure 3F & 3H, Figure 3G & 3I, respectively). So, p53 builds and hyperactivates a topoisome in response to DNA-damage in cells.

The Switch from MYC- to p53-topoisome is mass action driven

MYC-topoisome induced DNA-damage, activated pATM triggering the p53 response and p53-topoisome assembly. So, the topoisome switch could be via a direct handoff of topoisomerases from MYC to p53 triggered by DNA-damage, or mass action driven by altered component concentrations. To explore the switchover, recombinant MYC and p53 were combined in the presence or absence of recombinant TOP1 and TOP2A and then immunoprecipitated with either anti-p53 or anti-MYC. The immunoprecipitates were examined by western blot for each protein. MYC could not pull down p53 nor vice-versa (Figure 4A–B), so MYC and p53 did not interact arguing against direct handoff. To explore the second possibility, fixed amounts of recombinant TOP1 and TOP2A were mixed either with a constant amount of MYC and increasing p53 (Figure 4A–B, lanes 1–4) or with a fixed amount of p53 and increasing MYC (Figure 4A–B, lanes 6–9). These mixtures were immunoprecipitatied using anti-MYC (Figure 4A) or anti-p53 (Figure 4B), and tested by western blot. Strikingly, the binding of p53 with either TOP1 or TOP2A increased with elevated p53 (Figure 4B, left part, lanes 1–4), but decreased with escalating MYC (Figure 4B, right section, lanes 6–9). Likewise, the immunoprecipitation of TOP1 or TOP2A rose as MYC was elevated (Figure 4A, right section, lane 6–9) but declined as increasing p53 competed with MYC for topoisomerases (Figure 4A, left part lane 1–4). p53 exhibited slightly greater affinity (or avidity) for topoisome formation than MYC (for formation, compare Figure 4A, lane 6–9 with Figure 4B, lane 1–4; for disappearence compare Figure 4A, lane 1–4 with Figure 4B, lane 6–9), perhaps reflecting p53 tetramerspossessing multiple interaction sites.

Figure 4. The switch from the MYC- to the p53-topoisome is mass action driven.

(A and B) Recombinant proteins were mixed as indicated and immunoprecipitated (IP) with either (A) anti-MYC or (B) anti-p53 and probed for indicated proteins. (C) Supercoiled plasmid was added to TOP1 alone or pre-incubated with either increasing p53 or increasing MYC as indicated and assessed by gel electrophoresis. Chl=chloroquine. (D) Quantification for comparative DNA relaxation by TOP1 shown in 4C. (E) Scheme of nutlin treatments at top. Following treaments, nuclear extracts of mentioned cells were IP with anti-MYC or anti-p53 and checked for mentioned proteins. (F) TOP1 relaxation assay using untreated, tam (200 nM; 20 min) or nutlin (10 μM; 3hr) treated HO15.19-MYC-ER lysates. See also Figure S4.

To evaluate the impact of MYC and p53 on TOP1 enzymatic activity, TOP1 relaxation assays were performed with increasing concentrations of MYC or p53, and a constant amount of TOP1. p53 stimulated TOP1 activity 2–3x more than MYC, adjusted for equal monomer concentrations (Figure 4C, lane 4–9 with lane 11–16 and Figure 4D); as a tetramer p53 is a more potent stimulator of topoisomerase activity than MYC.

The p53-topoisome is even more potent than the MYC-topoisome

To determine whether p53-topoisome formation requires DNA-damage-triggered MYC degradation to release TOP1 and TOP2, we inhibited MDM2, the E3-ligase that targets p53 for degradation, with nutlin-3a⁶⁸; p53 accummulates in nutlin-treated cells. Cellular extracts from nutlin-treated wild-type and p53^−/− HCT116 cells were immunoprecipitated using anti-MYC and anti-p53. Nutlin diminished MYC-topoisomes, but enhanced p53-topoisome formation in wild-type cells (Figure 4E, compare lane 2 and lane 6), while, in nutlin-treated p53^−/− HCT116 cells, MYC-topoisomes persisted. Because p53 is able to form topoisomerases in the absence of DNA-damage or activated ATM, we infer that it has greater affinity for these enzymes than MYC. The perdurance of the MYC-topoisome in p53-knockout cells may sustain DNA-damage.

To evaluate the impact of elevated MYC versus p53 on cellular TOP1 activity, MYC-ER cells were treated with nutlin or tam and cellular extracts were then assayed for TOP1 using chloroquine gel electrophoresis. Nutlin stimulated cellular TOP1 activity more than tam in a dose dependent manner (Figure 4F, compare lane 6–7 with lane 9–10, Figure S4A), revealing p53 to be a more potent activator than MYC. Enhanced Topoisomerase activity increased transient-TOPccs whether in vitro or in cells. To compare the p53- and MYC-induced TOPccs in cells, we performed covalent adduct detection (CAD) assays. MYC-ER cells were treated with the Eto to trap TOPcc in the presence or absence of either nutlin or tam. Strikingly, nutlin treatment yielded more TOP2Accs than tam (Figure S4B, Eto panel, Figure S4C) either in the presence or absence of Eto. When TOP1 and TOP2A are joined in topoisomes, each increases the likelihood that the other will be tethered to nearby DNA.. Therefore, cross-CAD assays³⁶ were performed to compare TOPccs generated by p53 versus MYC. MYC-ER cells were treated with TOP1 poison CPT in the presence and absence of nutlin or tam. Immunoblotting with anti-TOP2A revealed the predicted cross-CAD enhancement of TOP2Acc due to CPT-mediated TOP1 topoisome tethering (Figure S4B, CPT panel, Figure S4C). Nutlin augmented the cross-CAD TOP2Acc more than tam did in the presence of CPT. In summary, p53 establishes a more potent topoisome than MYC.

p53-topoisome is less genotoxic than the MYC-topoiosme

Supra-physiological MYC levels promoted topoisome assembly in turn increasing TOP1 and TOP2 activity that, increased DNA-damage. The potent p53-topoisome presumably resolves topological issues caused by elevated transcription of p53-target genes. Wouldn’t the p53-topoisome also be DNA-damaging in? MYC-ER cells were treated or not, with tam and nutlin singly or in combination for 3 and 5 hours and checked for the elevation of pATM and formation of γH2AX by western blotting and confocal microscopy. Nutlin alone did not elevate pATM, but increased γH2AX a bit less than tam (Figure S5A, lane 1–4). At 3 hours, tam and nutlin treatments were additive, intesifying γH2AX by ~2 fold over either alone (Figure 5A, Top and middle panel, Figure 5B). 5 hours treatment with nutlin markedly attenuated γH2AX intensity compared to 3 hours, regardless of tam treatment (Figure 5A, compare middle and lower panel, Figure 5B), suggesting that an initial burst of nutlin-driven DNA-damage is reduced upon prolonged treatment. To monitor p53-triggered DNA-damage, HCT116 cells were examined for γH2AX by confocal microscopy and western blotting following nutlin treatment. As p53 increased, γH2AX spiked with 2–3 hours of nutlin declinng thereafter, despite p53 increasing further (Figure 5C left panels, Figure 5D and Figure S5B). In p53^−/− HCT116 cells, γH2AX levels continued to rise as nutlin treatment was prolonged (Figure S5B and Figure 5C, Figure 5D), revealing a nutlin-induced, p53-independent DNA-damag^68,69. To document that p53 induced acute DNA-damage, γH2AX foci were documented in non-transformed p53 wild-type MCF10A cells transfected with control (ctr) or p53 siRNA and then treated or not with nutlin for 3 to 5 hrs. Whereas, in control siRNA-transfected cells, γH2AX foci intensfied for 3 hours but declined by 5 hours (Figure 5E, Figure 5F), with p53-knockdown, γH2AX levels continued to climb suggesting that p53 attenuates both MYC- and p53-topoisome-mediated damage. Notably, p53^−/− HCT116 cells are impaired for repair of etoposide/TOP2 induced damage⁷⁰.

Figure 5. The p53-topoisome is less genotoxic than the MYC-topoisome.

(A, C and E) Immunofluorescence microscopy of (A) tam-treated or untreated HO15.19-MYC-ER cells, (C) wild-type (WT) and p53-knockout (p53^−/−) HCT116 cells, and (E) wild-type (ctr siRNA) and p53-knockdown (p53 siRNA) MCF10A cells with or without nutlin (10 μM; 3/5hr) treatment. γH2AX foci are in red. Nuclei were stained with DAPI (scale bars, 25 μm). (B, D and F) Quantification of γH2AX intensity per nucleus for 100 cells (B) as presented in 5A, (D) in 5C, and (F) in 5E (calculated value ± S.E.M.) and plotted for each scenario. Asterisks denote significant difference (****P<0.0001; t-test). (G) Quantification of DNA strand breaks for untreated or nutlin treated (for indicated hours) WT and p53^−/− HCT116 cells for 100 cells (calculated value ± S.E.M.) and plotted for each scenario. (**P<0.01; t-test). (H) Quantification of percentage of total apoptotic and dead cells for HO15.19-MYC-ER cells with or without nutlin treatment (10 μM) for indicated time⁶⁴. See also Figure S5.

Nutlin stabilizes p63 and p73 as well as p53⁷¹. p53/p63/p73 may form a ternary complex and complement each other’s functions^72,73. If p63/p73 normally rely on the p53-topoisome to manage DNA topology, then p53^−/− cells with elevated p73/p63 may suffer torsional stress inflicted DNA-damage. Strikingly, nutlin elevated p63 and p73 expression more rapidly in p53^−/− HCT116 cells than in wild-type (Figure S5B). To investigate whether p63/p73 associate with TOP1 and TOP2A in cells⁷⁴, WT and p53^−/− HCT116 cells were treated with nutlin for 3–5 hours, cellular lysates were immunoprecipitated using anti-p73 (Figure S5C) or anti-p63 (Figure S5D) and examined by immunolot for TOP1 and TOP2A. Anti-p63/p73 efficiently immunoprecipitated TOP1 but not TOP2A in nutlin-treated p53^−/− cell extracts (Figure S5C–D, lane 4–6), suggesting that they did not form topoisomes. Persistence of the MYC-topoisome in p53^−/− cells and absence of the p53-topoisome may impair management of torsion, topology and chromatin at p63/p73 targets and exacerbate DNA-damage.

p53-topoisome induced DNA-damage is blunted by the p53-response

Nutlin-induced DNA-damage was confirmed using alkaline comet assays (Figure 5G) of wild-type or p53^−/− HCT116 cells. Whereas the amount of DNA in the comet tails increased ~4-fold with 3 hours of treatment thereafter stabilizing in the wild-type cells (Figure 5G, WT panel), damage continued to accrue in p53^−/− cells (Figure 5G, p53^−/− panel). Despite the p53-topoisome acutely damaging DNA, the p53-response attenuated its own genotoxicity.

To monitor p53-topoisome induced apoptosis and cellular death, MYC-ER cells were treated with tam and/or nutlin for 3–12 hours, annexin V(+) and propidium iodide (PI) stained, and studied by flow cytometry. 3 hours with nutlin increased the apoptotic+dead cell populations by ~3 fold compared to untreated, increasing no further (Figure 5H). p53-topoisome formation initially promoted genomic instabilty but then activated DDR to limit topoisome damage and protect cells.

TDP1 and TDP2 are the target of p53

p53-topoisome formation following nutlin treatment directed an immediate burst of DNA-damage, apoptosis and cell-death. As the guardian of the genome, p53 would be expected to quell this burst⁴⁴. To handle topoisomerase-mediated DNA-damage, p53 must direct hydrolysis or excision of tyrosyl-DNA-adducts at topoisomerase generated DNA-breaks^40,75. Tyrosyl-DNA-phosphodiesterases 1 and 2 (TDP1 & 2) yield clean, ligatable 3’ and 5’ ends^76,77. The Expression of TDP1 and TDP2 proteins were monitored by immunoblot of wild-type or p53^−/− HCT116 cells, treated or untreated with nutlin for 3 and 5 hours. In wild-type cells, nutlin treatment augmented TDP1 and TDP2 protein expression by ~2 fold compared to untreated (Figure 6A, Figure 6B–C, WT panel), partly reflecting a modest increase in their mRNAs (Figure S6A–B). Surprisingly, in p53^−/− HCT116, the basal TDP1 and TDP2 were markedly lower than in wild-type, and only marginally increased by nutlin (Figure 6A, Figure 6B–C, p53^−/− panel). These data indicate that basal p53 sustains a low level of TDP1 and TDP2 expression in untreated wild-type cells, perhaps reflecting ongoing topoisomerase-mediated DNA-damage at steady state. As p53 increased, TDP1 and TDP2 rose to levels sufficient to offset ongoing p53-topoisome mediated damage.

Figure 6. p53-topoisome is essential for upregulating p53-target gene expression to mitigate DNA-damage.

(A) WT and p53^−/− HCT116 cells were treated as shown in the above scheme and checked for the indicated proteins. Actin served as loading control. (B) Quantification shows the fold change in expression of TDP1 or (C) TDP2 in 6A. Quantification of the fold change in the binding of p53 with either (D) TDP1 or (E) TDP2 promoter for WT HCT116 cells for ChIP-PCR with indicated treatment and assayed by RT-qPCR using IgG as control. (Error bars represent the mean ± S.E.M.). (***P<0.001; t-test). Quantification of the fold change in expression of (F) TDP1 and (G) TDP2 for Wild-type (si control) and TOP1+TOP2A-knockdown (si TOP1+ si TOP2A) HCT116 cells untreated or treated with Nutlin (10 μM, 3/5hrs). (H) Heatmap of gene expression of p53-targets arranged in descending order of the expression for indicated cells under mentioned condition. (I) Scatter plot of the difference in fold changes of p53-target transcripts between Wild-type (si control) and TOP1+TOP2A-knockdown (si TOP1+ si TOP2A) HCT116 cells under Nutlin as a function of the fold change in Wild-type (si control) cells. Black line indicates the best fit linear trend and the legend indicates the corresponding Spearman correlation coefficient and p-value. See also Figure S6.

Consistent with TDP1 and TDP2 being p53-targets, the Encode database revealed clear TP53 binding peaks at their promoters⁷⁸ (https://www.encodeproject.org/search/?searchTerm=tp53+chip-seq+myers). Further confirming p53 acting at the TDP1 and TDP2 promoters to mitigate p53-topoisome mediated DNA-damage, chromatin immunoprecipitation (ChIP) using anti-p53 and subsequent RT qPCR was performed on wild-type HCT116 cells treated or untreated with nutlin for 3 and 5 hours. In wild-type cells, TDP1 and TDP2 expression wincreased ~2 fold compared to untreated following 3 hour treatment (Figure 6D for TDP1 and Figure 6E for TDP2). Expression was further elevated after 5hrs of treatment. TDP1 and TDP2 are p53-targets for p53.

The topoisome is necessary for the p53-transcriptional response

p53 orchestrates the upregulation of diverse target genes associated with DNA repair, cell-cycle arrest, and apoptosis. Without the globally acting MYC-topoisome, p53-targets would depend on the p53-topoisome to mitigate torsional stress. Are the combined actions of TOP1 and TOP2 required to express TDP1, TDP2 (Figure 6A–E) and other p53-targets? Wild-type (ctr siRNA transfected) and TOP1+TOP2A knockdown (TOP1 siRNA+ TOP2A siRNA transfected) HCT116 cells were untreated or treated with nutlin for 3 to 5 hrs (Figure 6H–I) to acutely elevate p53 and RNA-seq was performed. Although RT qPCR confirmed knockdown of TOP1 and TOP2A, this was not obvious in RNA-seq data. Apparently, knockdown of TOP1 and TOP2 resulted reduced global RNA levels. This notion was validated by RNA-seq data was normalized to total cellular RNA, and RNA estimation by RT qPCR (as detailed in Methods). Besides global RNA reduction following TOP1+TOP2A knockdown, p53-target upregulation was impaired (Figure 6H and Figure S6D). Basal expression of TDP1 and TDP2 was depressed in topoisome-knockdown cells in the absence of nutlin (Figure 6F–G) while nutlin treatment impaired p53-driven upregulation to less than 40% of the control. TOP1 and TOP2A silencing reduced nutlin induction (fold-change) at all p53-targets (Figure 6I). TP53-targets with most induced by nutlin with control siRNA transfection cells, were more sensitive to topoisome-knockdown both with and without nutlin (Figure 6H–I, Figure S6C). That highly induced p53-targets were generally more affected by TOP1 and TOP2A silencing than less induced genes, confirms that topoisomerase activation by p53 intensifies p53 activity.

The p53-topoisome enhances transcription at p53-targets

After MYC-topoisome provoked DNA-damage, MYC-topoisomes are lost and replaced by p53-topoisomes. Concurrently, the cell’s output shifts from MYC-amplification transcription during growth to the temporally complex but smaller set of genes upregulated by p53^6,19. To enumerate sites enriched for p53-topoisomes, we employed a technique to enrich MYC-topoisome-bound DNA that distinguishes topoisomes from solitary topoisomerases³⁶. Usually, the TOP2 poison etoposide traps covalent TOP2cc’s but not TOP1-DNA complexes. However, when the MYC-topoisome is locked onto the DNA by etoposide, the indirectly tethered TOP1 more frequently engages nearby DNA. To confirm that TOP1 and TOP2 are active within the same p53-topoisome, we challenged MYC-ER cells with nutlin, but not tam to elevate p53-topoisome formation. As reported³⁶, the “cross-CAD” technique uses TOP2 inhibitors to trap TOP1 on DNA (and vice versa) when topoisomes are present. (Figure 7A–B). MYC-ER cells in the presence or absence of nutlin, were treated with Eto, to lock TOP2 covalently onto DNA (Figure 7B). Strikingly, in the presence of Nutlin+Eto, cross-CAD confirmed the predicted enhancement of TOP1cc by Eto-mediated topoisome tethering through TOP2 (Figure 7C, right panel). TOP1cc recovery was notably higher following p53 stabilization, as topoisomes were DNA-linked via Eto-TOP2cc-DNA adducts.

Figure 7. p53-topoisome supports the elevated transcriptional activity of p53-target genes.

(A-B) Schematic representation of “Cross-CAD” assay. CPT poisoning of TOP1 also traps TOP2A in the presence of p53, while Eto poisoning of TOP2A also traps as well as TOP1. Red dots with Y indicate the active site of TOP1 (Y723) or TOP2 (Y805). (C) Detection by anti-TOP1 of p53-tethered TOP1 along with Eto-trapped TOP2Acc by cross-CAD assays in HO15.19-MYC-ER cells with or without nutlin. (D) Average TOP1 Cross-CAD-seq (RPM) at all RNAPII genes (between −3 kb, TSS and +3kb) or (E) at p53-target genes for Rat and (F) at Top (90–100%), middle (61–90%) or low (<60%) expressed genes. (G) The model describing how a p53-topoisome replaces the acutely genotoxic MYC-topoisome to empower DNA-damage repair in cells.

Next to identify and map catalytically engaged p53-topoisome, “cross-CAD-seq” was performed, as shown in Figure 7A–B. MYC-ER cells were treated with Eto in presence/absence of nutlin to capture TOP2A and TOP2B-DNA adducts. Immunoprecipitating lysates using anti-TOP1 instead of anti-TOP2A recovered the bound topoisome population following “Cross-CAD-seq”⁷⁹. Lacking formaldehyde or any other cross-linker, this method depends on TOPccs. The DNA recovered without any treatment must indicate either TOP1 catalytically engaged during lysis or nonspecific background (Input). In untreated cells, TOP1 engagement was at a minimum at transcription start sites (TSS) of protein coding genes, supporting previous findings showing thatTOP1cc formation at active genes is suppressed until pause release, when the TOP1 is stimulated by BRD4-dependent RNAPII-CTD phosphorylation⁸⁰ (Figure 7D). Following nutlin+Eto treatment this minimum was dramatically elevated supporting the notion that p53 sponsored covalent engagement of p53-topoisome with TSS increased during transcription.

In response to DNA-damage, p53 upregulates a variety of target genes involved in DNA repair, cell-cycle arrest, and apoptosis. Elevated transcription of genes provokes opposing torsional stress³⁰. Without the MYC-topoisome, highly expressed p53-targets would depend on the p53-topoisome⁸¹ (Table S1) were analyzed for the recruitment of the p53-topoisome upon p53 activation. In untreated cells, there was a low baseline of topoisomes at the TSS of p53-targets that was transformed into a peak upon nutlin+Eto treatment (Figure 7E). This data confirms a role for p53-topoisomes in p53-target gene expression.

Previous research has indicated a relationship between the accumulation of topoisomerase and the expression of mRNAs⁸⁰. If the mRNA levels reflect the transcription intensity, then p53-topoisome recruitment should parallel mRNA output. Protein coding genes were classified into three categories based on their expression levels: Top 10%, mid (61–90%), and low (<60%). Indeed, topoisome complex recruitment was highest for highly expressed genes and decreased with lower gene expression (Figure 7F). Overall, these results suggest that p53-topoisomes are effective tools to promote p53-target gene expression while opposing MYC-driven cell growth. The reliance of oncogenic transcription and replication on MYC-topoisomes may lead to p53-activating DNA-damage that would be compounded by other genomic stressors.

Discussion

MYC-topoisome is detrimental for cells

MYC hyperactivated transcription (Figure 7G, arrow 1) and replication (Figure 7G, arrow 2 & 3) pumps positive supercoils upstream and negative supercoils downstream of sites of active transcription and/or replication (Figure 7G, arrows 4 & 5)^{30,31,35,83,84}. Unless mitigated this stress could inhibit RNA and DNA synthesis (Figure 7G, arrows 6 & 7). To manage topological stress, (Figure 7G, arrow 8) MYC and MYCN bind and stimulate TOP1 and TOP2 activity in topoisome A in proliferating cells and topoisome B in differentiated tissues (Figure 7G, arrow 8 & 9)³⁶. But topoisomerase overactivity is genotoxic (Figure 7G, arrow 10)^37,40. Topoisomerases obligatorily break one or both DNA strand(s) allowing passage of the complementary strand or a double helix through the break before resealing changes twist or writhe³⁹. As no biological process is 100% efficient, rare failures of break closure elicit DDR⁴⁰. Failures might occur stochastically or accrue during transcription, replication, chromatin remodeling or other processes that develop topological and mechanical impediments to topoisomerase activity⁸⁴. Here we find a pulse of DNA-damage concurrent with MYC activation due to activated topoisomes (Figure 1); consistent with this, preincubation with TOP2A catalytic inhibitor ICRF-187, reduced MYC-induced DNA breakage in tam-treated cells in the presence and absence of etoposide (Figure 1A–C). DNA-damage coincident with MYC-topoisome assembly, increased apoptosis and cell-death only 4hrs after tam treatment (Figure 1D), well before S-phase entry (Figure 1E). Progressing further into the cell-cycle, doubled apotosis/cell-death as topoisome-mediated damage and was compounded with equivalent replication/transcription-stress. Despite ongoing DNA-damage, cell-cycle arrest is delayed as MYC together with MIZ1 block the expression of G1-arresters CDKN1A, CDKN4A, and CDKN4B ( p21 and p15/p16, respectively)^50,52. Physiologically, high MYC levels are transient, and so MYC-topoisome DNA-damage is quickly repaired. When high MYC levels are pathologically sustained, MYC-topoisomes damage enough DNA to activate pATM, in turn, triggering MYC degradation (Figure 7G, arrow 11) while activating p53 and p21 (Figure 7G, arrow 12) that further repress MYC (Figure 1F). These conditions promote assembly of the p53-topoisomes and loss of MYC-topoisomes (Figure 7G, arrows 13 & 14) enabling p53-driven DDR to override MYC. Thus, MYC-induced DNA-damage immediately and directly trigger p53, providing a strong impetus for tumors to inactivate p53 or establish other survival pathways²⁰.

Why assemble p53-topoisomes?

Following acute genotoxic stress, p53 upregulates multiple target genes while globally downregulating transcription by reducing MYC (Figure 7G, arrows 15 & 16)^18,46. The hundreds of upregulated p53-targets enforce cell-cycle arrest, apoptosis and DDR (Figure 7G, arrows 18 &19). Robust transcription of p53-targets must generate topological and mechanical stresses that would compromise their expression if not resolved; the p53-topoisome provides this resolution (Figure 4A) (Figure 7G, arrow 14 &17).

The p53-topoisome is more robust and potent than MYC(N)-topoisomes. As with MYC, p53-topoisome formation transcription-independent (Figure 2C, Figure 3A–B). p53-target expression is facilitated by p53-topoisomes (Figure 6). Absent DNA-damage, MYC-topoisomes prevail, but p53-topoisomes predominate after DNA-damage. A universal amplifier, MYC upregulates global gene expression globally, whereas, upregulation by p53 is more restricted. These preferences are shared by MYC- and p53-topoisomes. Operating preferentially on highly expressed genes, MYC doesn’t amplify DNA repair genes, and cannot repair MYC-topoisome mediated damage. By upregulating repair genes such as TDP1 & 2, p53 supports DDR including damage from p53-topoisomes. Upon induction by nutlin, p53-topoisomes assemble in the presence of MYC and absence of DNA-damage (Figure 4E). p53-topoisomes relax supercoils and decatenate kDNA more efficiently than MYC-topoisomes. The efficiency of p53 topoisomes may partly derive from p53’s tetrameric nature¹⁶ that provides a 4-fold increase in the determinants of topoisomerase binding. p53, unlike MYC, enhances the activity of both TOP2A and TOP2B, making p53-topoisomes useful in both proliferating cells and differentiated tissues.

The p53-topoisome is less DNA-damaging than the MYC-topoisome

MYC-topoisome hyperactivity is genotoxic; experimentally, this DNA-damage might mask damage generated by newly formed p53-topoisomes. Similarly, genotoxic maneuvers that commonly trigger the p53-response might camouflage damage due to surging p53-topoisome levels. Forming after mdm2 inhibition by nutlin, p53-topoisomes trigger DNA-damage in the absence of extrinsic genotoxic stress (Figure 7G, arrow 20) (Figure 4E, and Figure S5A). This pulse of DNA-damage rapidly waned as the p53-response progressed (Figure 5). Apparently, p53-topoisome genotoxicity is countered by p53-programmed DDR. The repair of TOP1-DNA-damage involves multiple pathways, such as proteolysis and TDP1-pathway (involving TDP1-PNKP-XRCC1-PARP1-LigIII-polβ complex), the endonuclease pathway (involving Rad-Mus-Mre complex), and 5’ end religation mediated by RecQ helicases⁴⁰. The processing of TOP2-damage includes ubiquitination (involving MRE11-CtIP-BRCA1), TDP2-pathways, and sumoylation-mediated proteolysis (involving SPRTN-ZATT-SUMO)⁷⁶. Tyrosyl-DNA-phosphodiesterases (TDP) unlink topoisomerase active-site tyrosines from dead or stalled complexes to enable resealing of broken DNA. Topoisome activity augments both TOP1cc and TOP2cc formation.

HCT116 p53^−/− cells suffer twice as much basal DNA-damage while expressing half the TDP1 and TDP2 as their wild-type counterparts. With nutlin treatment, this damage further accumulates deriving from at least two sources: (1) peristence of the MYC-topoisome in the absence of pATM activation, and (2) the mismanagement of DNA topology, torsional stress and chromatin by p63- and p73 due to their inablility to assemble a topoisome. This deficiency may create problems for upregulating DDR and unmitigated torsional stress that may sponsor alternative DNA structures, R-loops, back-tracked RNA polymerase or other DNA-damage-sensitizing situations. Lacking p53, TDP1, TDP2 and perhaps other DDR components, repair pathways are less effective, and so damage accrues (Figure 5G–H). Despite being a universal amplifier of transcription, MYC fails to upregulate weakly expressed TDP1 or TDP2. MYC’s inability to elicit DDR without p53 may help to limit its oncogenic potential. Pulses of upregulated MYC are provoked by a variety of signals; this upregulated MYC amplifies and accelerates stress responses and other cellular programs. Sustained high MYC is usually pathologic, reinforcing proliferation and promoting neoplastic transformation; high MYC-states foster high levels of p53-activating MYC-topoisome-mediated DNA-damage. The p53 response quells MYC action delimiting MYC-mediated mitogenesis^47,85. If MYC were to sponsor DDR directly, and so rescue topoisome-mediated damage, its oncogenic potential would be unopposed. When very high MYC levels are required during a physiological response (i.e. B-cell activation), we speculate that p53 variants (such as Δ133) that support DDR but are impaired for induction of apoptosis or senescence^86–88, might mitigate MYC-topoisome genotoxicity. Pathologically, such variants or similarly acting mutant p53 proteins would allow MYC to persist at oncogenic levels. The dependence of MYC-driven oncogenic processes on topoisome activity creates a vulnerability that is exploited by p53 during periods of genomic stress and may inform strategies to intercept neoplastic growth.

Limitations of the Study

Rapid MYC loss upon pATM activation was unexpected. We do not know if pATM directly or indirectly triggered MYC degradation. Presumably MYC loss is through ubiquitin-directed proteasomal degradation. The responsible E3-ligase(s) are not known. In MYCER12 cells, MYC-topoisomes cause DNA-damage without exogenous DNA-damaging agents, We have yet not determined whether ATR-mediated pathways similarly lead to replacement of MYC-topoisomes by p53-topoisomes. The switch from MYC- to p53-topoisomes has been characterized mainly in MYCER12. We would not be surprised if MYC overactivity provokes similar switching in other cells. The fate of the MYCN-topoisome (MYCN-TOP2B-TOP1) in response to DNA-damage has not been examined. Because TOP2B may predominate in post-mitotic, differentiated cells, its relationship to p53, DNA-damage and the cell-cycle need not parallel MYC. We note that these experiments were performed in tissue culture, and so organismal parameters that modify topoisome biology have not been examined.

Resource availability

Lead contact

Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact David Levens (levensd@mail.nih.gov).

Materials availability

The material generated in this study is available upon request.

Data and code availability

• The ChIP-seq data presented in this paper have been submitted to the GEO database under the accession number GSE243792. The data are publicly available as the date of publication.

• This paper does not include original code. Detailed method of normalization for RNA-seq data was discussed in the method section. Published software, code, and the accession numbers for all previously mentioned datasets used in this study are listed in the Key-Resource-Table.

• Any Additional information necessary to reanalyze the data reported in this paper is available from the lead contacts upon request.

Key resources table

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Antibodies
Rabbit monoclonal Anti-Topoisomerase I antibody	Abcam	Cat# ab109374; RRID:AB_10861978
Mouse monoclonal Anti-Topo I Antibody (C-21)	Santa cruz Biotechnology	Cat# sc-32736; RRID:AB_628382
Rabbit monoclonal Anti-Topoisomerase II alpha antibody	Abcam	Cat# ab52934; RRID:AB_883143
Rabbit monoclonal Recombinant Anti-c-Myc antibody [Y69]	Abcam	Cat# ab32072; RRID:AB_731658
anti-p53 (DO-1)(Sc-126)	Santa Cruz Biotechnology	Cat# sc-126, RRID:AB_628082
anti-p15/p16 (C-7)	Santa Cruz Biotechnology	Cat# sc-377412, RRID:AB_2936231
anti-p21 (F-5)	Santa Cruz Biotechnology	Cat# sc-6246, RRID:AB_628073
anti-pATM	Abcam	Cat# ab81292, RRID:AB_1640207
normal mouse IgG antibody	Santa Cruz Biotechnology	Cat# sc-2025; RRID:AB_737182
normal rabbit IgG antibody	Santa Cruz Biotechnology	Cat# sc-2027; RRID:AB_737197
Goat Anti-Rabbit IgG H&L (HRP) antibody	Abcam	Cat# ab205718; RRID:AB_2819160
Goat Anti-Mouse IgG H&L (HRP) antibody	Abcam	Cat# ab205719; RRID:AB_2755049
Rabbit polyclonal anti-TDP2	Thermo Fisher Scientific	Cat# A302-737A, RRID:AB_10631698
Rabbit monoclonal anti-TDP1 antibody	Cell Signaling Technology	Cat# 59710, RRID:AB_2799572
mouse monoclonal anti-actin	Cell Signaling Technology	Cat# 3700, RRID:AB_2242334
Mouse monoclonal anti-YH2ax (JBW301)	Millipore-Sigma	Cat# 05-636, RRID:AB_309864
Rabbit monoclonal anti-p73 antibody	Abcam	Cat# ab40658, RRID:AB_776999
Rabbit monoclonal anti-p63 antibody	Abcam	Cat# ab124762, RRID:AB_10971840
Goat anti-Mouse IgG (H+L) Cross-Adsorbed Secondary Antibody, Alexa Fluor™ 488	Thermo Fisher Scientific	Cat# A-11001, RRID:AB_2534069
Chemicals, peptides, and recombinant proteins
DMEM, high glucose	Thermo Fisher Scientific	Cat# 11965092
Opti-MEM™ I Reduced Serum Medium	Thermo Fisher Scientific	Cat# 31985062
Corning® 100 mL MEM Nonessential Amino Acids, 100x	Corning	Cat# 25-025-CI
Penicillin-Streptomycin (10,000 U/mL)	Gibco	Cat# 15140122
Bovine Serum, heat inactivated, New Zealand origin	Thermo Fisher Scientific	Cat# 26170043
DMEM/F-12	Invitrogen	Cat# 11330032
Horse Serum, New Zealand origin	Invitrogen	Cat# 16050122
EGF	Peprotech	Cat# AF-100-15
Hydrocortisone	Milipore-Sigma	Cat# H0888
Cholera Toxin	Milipore-Sigma	Cat# C8052
Insulin	Milipore-Sigma	Cat# I1882
Lipofectamine™ 2000 Transfection Reagent	Thermo Fisher Scientific	Cat# 11668019
ON-TARGETplus Smartpool Human TOP1 siRNA	Horizon	Cat# L-005278-00-0005
ON-TARGETplus Smartpool Human TOP2A siRNA	Horizon	Cat# L-004239-00-0005
ON-TARGETplus Smartpool Human p53 siRNA	Horizon	Cat# L-003329-00-0005
cOmplete™, Mini, EDTA-free Protease Inhibitor Cocktail	Milipore-Sigma	Cat# 11836170001
Puromycin Dihydrochloride	Thermo Fisher Scientific	Cat# A1113803
Tamoxifen	Milipore-Sigma	Cat# T5648-1G
ICRF-187	Selleck Chemicals LLC	Cat# S1222
Etoposide	Milipore-Sigma	Cat# E1383-250MG
ATM Kinase inhibitor (KU-55933)	Selleck Chemicals LLC	Cat# S1092
Neocarzinostatin	Milipore-Sigma	Cat# N9162
Camptothecin	Milipore-Sigma	Cat# C9911-250MG
Triptolide from Tripterygium wilfordii, ≥98% (HPLC), solid	Milipore-Sigma	Cat# T3652-5MG
Nutlin-3	Milipore-Sigma	Cat# N6287
(R)-MG132	Cayman chemicals	Cat# 13697
Benzonase® Nuclease, ultrapure ≥250 units/μL, ≥99% (SDS-PAGE)	Milipore-Sigma	Cat# E8263-25KU
UltraPure™ Agarose	Thermo Fisher Scientific	Cat# 16500500
Chloroquine diphosphate salt	Milipore-Sigma	Cat# C6628-25G
VECTASHIELD® Antifade Mounting Medium with DAPI	Vector laboratories	Cat# H-1200-10
Pierce™ Protein A/G Magnetic Beads	Thermo Fisher Scientific	Cat# 88803
Diamond™ Nucleic Acid Dye	Promega	Cat# H1181
RNase A, DNase and protease-free (10 mg/mL)	Thermo Fisher Scientific	Cat# EN0531
Proteinase K, Molecular Biology Grade	New England Biolabs Inc.	Cat# P8107S
Phenol:Chloroform:Isoa myl Alcohol 25:24:1 Saturated with 10 mM Tris, pH 8.0, 1 mM EDTA	Milipore-Sigma	Cat# P2069-400ML
Sodium dodecyl sulfate	Milipore-Sigma	Cat# L3771-100G
Glycerol for molecular biology, ≥99.0%	Milipore-Sigma	Cat# G5516-1L
Lithium chloride anhydrous, free-flowing, Redi-Dri™, ReagentPlus®, 99%	Milipore-Sigma	Cat# 793620-100G
TWEEN® 20 for molecular biology, viscous liquid	Milipore-Sigma	Cat# P9416-50ML
Triton™ X-100 for molecular biology	Milipore-Sigma	Cat# T8787-50ML
N-Lauroylsarcosine sodium salt BioXtra, ≥97% (TLC)	Milipore-Sigma	Cat# L5777-50G
Sodium deoxycholate monohydrate BioXtra, ≥99.0% (titration)	Milipore-Sigma	Cat# D5670-25G
Urea ACS reagent, 99.0-100.5%	Milipore-Sigma	Cat# U5128-5KG
DTT 1,4-Dithiothreitol	Milipore-Sigma	Cat# 11583786001
Nonidet™ P 40 Substitute	Milipore-Sigma	Cat# 74385
Ethylenediaminetetraac etic acid disodium salt dihydrate	Milipore-Sigma	Cat# 03685-500G
Potassium chloride	Milipore-Sigma	Cat# P9541-500G
UltraPure™ Ethidium Bromide, 10 mg/mL	Thermo Fisher Scientific	Cat# 15585011
Exonuclease VII	New England Biolabs Inc.	Cat# M0379S
Recombinant Human Topoisomerase IIα	NCI core facility	N/A
Recombinat Human TOPIIβ protein	This paper	N/A
Recombinat Human TOP1 protein	NCI core facility	N/A
Recombinat Human cMYC protein	NCI core facility	N/A
Recombinat Human p53 protein	Active Motif	Cat# 81091
NuPAGE™ LDS Sample Buffer (4X)	Thermo Fisher Scientific	Cat# NP0007
AMPure XP	Beckman	Cat# A63880
Ibidi 15 μ-slide angiogenesis slide	Ibidi-treat	Cat# 81506
Pierce™ 16% Formaldehyde (w/v), Methanol-free	Thermo Fisher Scientific	Cat# 28906
4-12% Bis-Tris NuPAGE protein gel	Thermo Fisher Scientific	Cat# NP0323
SYBR™ Green I Nucleic Acid Gel Stain	Thermo Fisher Scientific	Cat# S7563
SuperSignal™ West Femto Maximum Sensitivity Substrate	Thermo Fisher Scientific	Cat# 34095
Nitrocellulose Membranes, 0.45 μm	Thermo Fisher Scientific	Cat# 77010
Fetal Bovine Serum, GIBCO	Thermo Fisher Scientific	Cat# A5256801
Sodium Chloride	Milipore-Sigma	Cat# S9888
Bovine Serum Albumin	Milipore-Sigma	Cat# A7906
Ethyl Alcohol	Milipore-Sigma	Cat# E7023
Adenosine 5’-triphosphate disodium salt hydrate	Milipore-Sigma	Cat# A2383
Magnesium chloride	Milipore-Sigma	Cat# M8266
Critical commercial assays
Comet Assay Kit	Abcam	Cat# ab238544
Dead Cell Apoptosis Kits with Annexin V for Flow Cytometry	Thermo Fisher Scientific	Cat# V13242
FxCycle™ PI/RNase cell cycyle kit	Thermo Fisher Scientific	Cat# F10797
Senescence Cell Detection kit	Dojindo	Cat# SG04
Gel Filtration Calibration	Cytiva	Cat# 28403842
Kits for HMW
iTaq™ Universal SYBR® Green Supermix	Bio Rad	Cat# #1725121
Direct-Zol TM RNA Miniprep kit	Zymo Research	Cat# R2051-A
Reverse transcription kit	Invitrogen	Cat# 18080051
QIAquick PCR purification kit	QIAGEN	Cat# 28006
ThruPLEX DNA-seq kit	Takara	Cat# R400676
Agilent High Sensitivity DNA Kit	Agilent	Cat# 5067-4626
Qubit™ 1X dsDNA HS Assay Kit	Thermo Fisher	Cat# Q33230
Minielute PCR purification kit	Qiagen	Cat# 28004
NextSeq 2000 P3 Reagents (100 Cycles) (for ChIP-seq)	Illumina	Cat# 20040559
RNA seq Library prep kit	Illumina	Cat# E7760L
NextSeq 2000 P2 Reagents (200 Cycles) (for RNA-seq)	Illumina	Cat# 20046812
Deposited data
Raw and processed sequencing data	This Paper	GEO: GSE243792
Previously published Topoisome ChIP-seq data	Das et al., 202236	GEO: GSE181450
Previously published Encode data sets	Partridge, E.C. et al, 202078	GEO: GSE104247
Experimental models: Cell lines
HCT116	Developmental Therapeutics Programme	N/A
p53−/− HCT116	Developmental Therapeutics Programme	N/A
Rat HO15.19-MYC-ER	Das et al., 202236	N/A
MCF10A	Developmental Therapeutics Programme	N/A
Oligonucleotides
For Real time PCR using the Forward primer TDP1:5’-ATTGACTGTCCTCAC AGCGG-3’	IDT oligo	N/A
For Real time PCR using the Reverse primer TDP1:5’-CAGGGAGAGAGGAG AAGCCT-3’	IDT oligo	N/A
For Real time PCR using the Forward primer TDP2:5’-CCGAGAACGACTGGG AGATG-3’	IDT oligo	N/A
For Real time PCR using the Reverse primer TDP2:5’-AGAGATGGTTTCAGG TCGGC-3’	IDT oligo	N/A
For Real time PCR using the Forward primer TOP1:5’-GAACAAGCAGCCCGA GGATGAT-3’	IDT oligo	N/A
For Real time PCR using the Reverse primer TOP1:5’-TGCTGTAGCGTGATG GAGGCAT-3’	IDT oligo	N/A
For Real time PCR using the Forward primer TOP2A:5’-GTGGCAAGGATTCTG CTAGTCC-3’	IDT oligo	N/A
For Real time PCR using the Reverse primer TOP2A:5’-ACCATTCAGGCTCAA CACGCTG-3’	IDT oligo	N/A
For Real time PCR using the Forward primer μ-actin:5’-CACCATTGGCAATGA GCGGTTC-3’	IDT oligo	N/A
For Real time PCR using the Reverse primer μ-actin:5’-AGGTCTTTGCGGATG TCCACGT-3’	IDT oligo	N/A
Recombinant DNA
Kinetoplast DNA (kDNA)	Topogen	Cat# TG2013-3
Software and algorithms
GraphPad Prism 9.1.0 (software for graph and statistics analysis)	GraphPad	https://www.graphpad.com/scientific-software/prism/; RRID: SCR_002798
Adobe acrobat (illustrator for preparation of manuscript)	Adobe	https://adobe.com/products/illustrator; RRID:SCR_010279
Fiji	Schindelin et al., 2012 93	https://fiji.sc/; RRID:SCR_002285
Bowtie2 2.3.5.1	Langmead and Salzberg, 2012 94	http://bowtie-bio.sourceforge.net/bowtie2/index.shtml; RRID: SCR 016368
Deeptools 2.5.0	Ramirez et al., 201495	https://deeptools.readthedocs.io/en/develop/; RRID:SCR_016366
FastQC 0.11.9	http://journal.embnet.org/index.php/embnetjournal/article/view/200	https://github.com/s-andrews/FastQC; RRID:SCR_014583
MultiQC 1.10	Ewels et al., 201696	http://multiqc.info/; RRID:SCR_014982
Picard tools 2.10.3	Broad Institute	https://broadinstitute.github.io/picard/; RRID:SCR_006525
Samtools 1.8	Li et al., 200997	http://www.htslib.org/; RRID:SCR_002105
R 4.1.0	R Core Team (2021). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria.	http://www.r-project.org/; RRID:SCR_001905
Bioconductor	Huber et al., 201598	https://www.bioconductor.org/; RRID:SCR_006442
Cutadapt 3.1	http://journal.embnet.org/index.php/embnetjournal/article/view/200	https://www.bioconductor.org/; RRID:SCR_006442 https://cutadapt.readthedocs.io/en/stable/; RRID:SCR_011841
UCSC Genome Browser	Kent et al., 200299	http://genome.ucsc.edu/; RRID: SCR_005780
ggplot2	https://onlinelibrary.wiley.com/doi/10.1002/wics.147	https://cran.r-project.org/web/packages/ggplot2/index.html; RRID:SCR_014601
ngsplot 2.61	Shen et al., 2014100	https://github.com/shenlab-sinai/ngsplot;RRID: SCR_011795
MACS3 3.0.0a6	Zhang et al., 2008101	https://github.com/macs3-project/MACS; RRID:SCR_013291
Bedtools 2.27.1	Quinlan et al., 2010102	https://bedtools.readthedocs.io/en/latest/index.html; RRID:SCR_006646
TOP1 CAD-seq	Vladislav Kuzin et al., 202279	N/A

STAR methods

Experimental model and study participant details

Cell Cultures, Drug Treatments

Human colon carcinoma wild-type and p53^−/− HCT116 cells were grown in high glucose (4.5g/L) DMEM (Thermo Fisher, 11965092), MEM Nonessential Amino Acids (Corning, 25–025-Cl), 50 U/ml penicillin G and containing 10% FBS (Thermo Fisher, 26170043) in a 37°C incubator supplied with 5% CO₂. NCS and nutlin treatments were performed for 3–5 hrs on wild-type and p53^−/− HCT116 cells at 37°C with a final concentration of 400 ng/ml or 10μM, respectively. ATMi treatments was performed for 1 hr on wild-type and p53^−/− HCT116 cells at 37°C with a final concentration of 10μM. Human breast epithelial MCF10A cells were grown in high glucose (4.5g/L) DMEM/F-12 (Invitrogen, 11330032), 5% Horse serum (Invitrogen, 16050122), 20 ng/ml EGF, 0.5 mg/ml Hydrocortisone, 100 ng/ml Cholera Toxin, 10 μg/ml Insulin and 50 U/ml penicillin G (Invitrogen, 15070063) in a 37°C incubator supplied with 5% CO₂. NCS and nutlin treatments were performed for 3–5 hrs on MCF10A cells at 37°C with a final concentration of 400 ng/ml or 10 μM, respectively. Rat1A or HO15.19-MYC-ER cells were grown with DMEM and 10% of calf serum (Gibco, 26170–043) as previously described³⁶. 4-OH tam treatment (200 nM final concentration) was performed for 15 min on HO15.19-MYC-ER cells to express a retrovirally encoded MYC-ER chimeric protein that immediately translocate to the nucleus. TOP2 catalytic inhibitor (ICRF-187) (Selleck Chemicals LLC, TX, USA) treatment was performed for 1hr on HO15.19-MYC-ER with a final concentration of 10 μM. CPT (Sigma, C9911) (20 μM) and Eto (Sigma, E1383) (25 μM) were added for the last 20 min. Nutlin treatment was performed for 3hr on HO15.19-MYC-ER with a final concentration of 10 μM and Eto (Sigma, E1383) (25 μM) were added for the last 30 min.

siRNA transfection

All transfections were carried out as described previously⁸⁹. In brief, cells (1.5×10⁵) were transfected with control siRNA or 20 nM p53 siRNA (GE Dharmacon, SiRNA-SMARTpool), TOP1 siRNA and TOP2A siRNA (GE Dharmacon, SiRNA-SMARTpool) using Opti-MEM^™ I Reduced Serum medium (Thermo Fisher, 31985062) according to the manufacturer’s protocol. Different time course experiments revealed a maximum suppression of specific protein at day 3 after transfection, as analyzed by western blotting.

Method details

Drug and antibodies

Camptothecin (CPT), etoposide (Eto), 4-OH tamoxifen (tam), doxycycline (Dox), Triptolide, Neocarzinostatin (NCS), and Nutlin-3 were purchased from Sigma-Aldrich (St Louis, MO, USA). TOP2 catalytic inhibitor (ICRF-187) and ATM Kinase Inhibitor (KU-55933) were purchased from Selleck Chemicals LLC (Houston, TX, USA). MG132 was purchased from Cayman Chemical (Ann Arbor, MI, USA). Mouse monoclonal anti-TOP1 (C-21), anti-p53 (DO-1)(Sc-126), anti-p15/p16 (C-7), anti-p21 (F-5), and anti-IgG antibodies were from Santa Cruz Biotechnology (Dallas, TX, USA). Rabbit monoclonal anti-c-MYC antibody (N-ter) (Y69) (ab32072), anti-TOP1 (N-ter) (ab109374), anti-pATM (ab81292), anti-TOP2A (ab52934), anti-p73 (ab40658) and anti-p63 (ab124762) were from Abcam (Cambridge, United Kingdom). Rabbit polyclonal anti-TDP2 (A302–737A) was from Bethyl, Thermo Fisher (Waltham, MA, USA). Rabbit monoclonal anti-TDP1 antibody (D8D1B) and mouse monoclonal anti-actin (8H10D10) was from Cell-signalling (Danvers, MA, USA). Mouse monoclonal anti-γH2ax (JBW301) was from Milipore Sigma. Secondary antibodies and horseradish peroxidase conjugated anti-rabbit IgG or anti-mouse IgG were purchased from Abcam.

Expression and purification of recombinant protein

Recombinant human TOP1 and TOP2A were obtained from NCI protein expression core (NCI, Frederick, MD). Recombinant human TOP2A was purchased from Topogen (Buena Vista, CO, USA). Recombinant p53 was purchased from Active Motif (81091). Recombinant MYC was produced using using the protocol as discussed previously³⁶ with slight modifications. All of the protein’s purity and activity was confirmed by protein staining and immunoblotting.

Immunoprecipitation and western blotting

Cells were washed twice with PBS, scraped off and lysed in RIPA buffer (10 mM Tris-HCl pH 7.5, 150 mM NaCl, 1 mM EDTA, 0.2 % Triton X-100, 1 % NP40, 5 % glycerol supplemented with complete protease inhibitors). After vortexing and incubation at 4°C for 1hr, lysates were centrifuged at 14000 rpm for 10 min at 4°C. Supernatants were collected and stored at −80°C. Nuclear extract was prepared as discussed previuosly³⁶. Briefly, Cells were spun down, resuspended in HP buffer (50 mM HEPES pH 7.5, plus protease inhibitors), passed 10 times through a 20 G needle (to disrupt the cell membrane), then centrifuged for 10 min, 4000 rpm at 4°C. The pellet was resuspended in 300 μl HLSP buffer (50 mM Tris-HCl pH 7.5, 100 mM LiCl, 0.2% sarkosyl and protease inhibitors), treated with 1 μl benzonase (250 units/μl) for 1 hr at 4°C and further dissolved through sonication and passing through a 25G needle. The sample was centrifuged at 15000 rpm for 15 min and the supernatant was collected and quantified. For the immunoprecipitation of nuclear extract, 1 mg of protein was pre-cleared with 25 μl of Protein A/G for 1 hr. The precleared supernatant was incubated overnight with 2 μg of specific antibody and 30μl of Protein A/G magnetic beads (blocked by 3% BSA) at 4°C. The immunocomplexes were washed four times with HLSP buffer and 2 times with RIPA buffer with 300 mM NaCl and 0.2 % sarkosyl. The immunoprecipitated proteins were eluted by heating in 40μl of 1 x NuPAGE LDS loading dye (Thermo Fisher, NP0007) for 10 min at high temp. Protein samples were examined by western blotting. Briefly, samples were run on 4–12% Bis-Tris NuPAGE protein gel (Thermo Fisher, NP0323), transferred onto nitrocellulose blots, probed with antibodies against the protein(s) of interest, and imaged by chemiluminescence using ECL chemiluminescence reaction (Thermo Scientific) under the Odyssey infrared scanner (Li-Cor). All densitometric quantifications of immunoblots used Image J software.

Immunoprecipitation of recombinant proteins were carried out by mixing equimolar concentrations of proteins in PDB buffer (10 mM Tris-HCl pH 7.5, 50 mM KCl, 0.5% NP40, 7.5 % Glycerol, 2 mM EDTA, 0.5 mM DTT and protease inhibitors) in total 20 μl on rotor at 4°C for 15 min and 500 ng of specific antibody was added, followed by addition of 130 μl PDB. The mixture was incubated on rotor at 4°C for 90 min. For each sample, 3 μl Protein A/G beads were blocked in PDB buffer with 5 % skim milk powder for 1 hr at 4°C, washed in PDB buffer, added to the protein solution and incubated at 4°C for 1 hr. The immunocomplexes were washed four times with wash buffer (10 mM Tris-HCl pH 8, 300 mM NaCl, 1mM EDTA, 1% Triton X-100, 0.1% Na-Deoxycholate, 0.1% SDS, and protease inhibitors). Elution was performed by heating samples in presence of protein 1 x NuPAGE LDS loading dye. Protein samples were subjected to standard western blotting protocol as discussed before.

Alkaline COMET assay

The extent of DNA-damage was assayed by alkaline COMET assay according to the manufacturer’s protocol (abcam, ab238544) as described previously⁶². Briefly, HO15.19-MYC-ER cells were treated with tamoxifen in presence or absence of Etoposide to induce DNA-damage, with or without pretreatment of ICRF-187. Approximately 1 ×10³ cells were collected and mixed with low melting agarose (provided in the kit) and plated in the chamber slide from kit. Slides were immersed in lysis solution at 4°C for 30 min and then transferred into alkaline buffer (Kit) for another 30 min followed by electrophoresis at a constant current of 300 mA and 20 V for 20 min at 4°C. After electrophoresis, slides were rinsed with pre-chilled double distilled water, dehydrated in ice-cold 70% ethanol for 5 min, and air-dried. Processed sildes were stained with Vista Green DNA dye and observed with epifluorescence microscopy EVOS M5000 Multi-Photon using a FITC filter. The distinct COMET tail was generated due to DNA-damage. The relative length and intensity of EtBr-stained DNA tails to heads, is proportional to the amount of DNA-damage present in the individual nucleus. Comet length and percentage of DNA in the COMET tail was measured and was scored for at least 100 cells by (Image J Fiji) software. Distributions of comet lengths were compared using the Student t-test.

Double Staining with Annexin V and PI

The outer membrane externalization of phosphatidylserine (PS) of HO15.19MYC-ER cells due to tam treatment ( 200 nM for 4,12 and 24 hrs) was measured by the Dead Cell Apoptosis Kit (Invitrogen V13242, USA) with the attachment of FITC-Annexin V FITC & Propidium Iodide for flow cytometry following manufacture’s protocol⁶⁴. For Synchronizing, HO15.19-MYC-ER cells were arrested in G0 by serum starvation for 16 hrs, and were treated or untreated with tam (200 nM) for 4, 12, 24 and 48 hrs in presence or absence of serum and then labelled with FITC Annexin V and PI. Briefly, cells were harvested and washed with cold PBS. Approximately, 1×10⁵ cells were resuspended in 100 μl annexin-binding buffer (1X) and labelled with 5 μl FITC Annexin V and 1 μl of the 100 μg/mL Propidium Iodide (PI) for 15 minutes at room temperature. After the incubation, 400 μl of annexin-binding buffer (1X) were added to the cells and analyzed for apoptotic and dead cells using CytoFLEX Flow Cytometer (Beckman Coulter, USA). The populations of live ((negative for both Annexin V and PI; green), apoptotic (Annexin V positive and PI negative; green) and dead (positive for both Annexin V and PI; red) cells were distinguished using a flow cytometer with the 488 nm line of an argon-ion laser for excitation. The gating was fixed so that the mean intensity of FITC-annexin V and the mean intensity of PI were depicted by FL-1 and FL-2 channels, respectively. The total apoptotic and dead population for each condition was calculated by addition of apoptotic dead cell populations. The data presented are the mean of three independent experiments.

Cell cycle analysis

The cell-cycle analysis was performed by Flow cytometry (FACS) according to the manufacturer’s protocol (Thermo, F10797). HO15.19-MYC-ER cells (1 ×10⁵ cells) were serum starved for 14–16 hrs (to synchronize) and treated with tam (200 nM) for 2–48 hrs and harvested. Fixation and staining of cells were performed with propidium iodide (PI) and were analyzed via flow cytometry as suggested⁶⁴. Detection of cells in individual sub G0, G1, S, and G2/M phases of the cell cycle were carried out using CytoFLEX Flow Cytometer (Beckman Coulter, USA).

Cellular Senescence Detection assay

The extent of Cellular Senescence was detected by senescence setection kit according to the manufacturer’s protocol (dojindo, SG04). Briefly, 1 ×10⁶ cells were plated on 6-well plate (Corning, 353046) for overnight and treated with tam for the indicated time followed by adding 1 ml of Bafilomycin A1 (supplied with kit) working solution and incubate at 37°C for 1 hour in a 5% CO₂ incubator. Next, Add 1 ml of SPiDER-βGal working solution (supplied with kit) was added to cells and incubate at 37°C for 30 minutes in a 5% CO₂ incubator. After incubation, cells were rinsed with HBSS twice and observed with epifluorescence microscopy EVOS M5000 Multi-Photon using a FITC filter. Detection of cells with senesence positive was carried out according to the manufacturer’s protocol (dojindo, SG04). The quantification was performed for at least 100 cells.

Immunocytochemistry and confocal microscopy

Immunofluorescence staining and confocal microscopy were performed as described previously^75,89,90. Briefly, 1 ×10⁴ cells were plated on Ibidi 15 μ-slide angiogenesis slide (ibidi-treat, 81506) for 1 hr and treated with the indicated concentration of drugs followed by fixation with 4% paraformaldehyde for 20 min at room temperature. Primary antibody against γH2AX (1: 200) was detected using anti-mouse IgG secondary antibodies (1: 1000) labeled with Alexa 488/568 (Invitrogen). Cells were mounted in anti-fade solution with 4 ,6-diamidino-2-phenylindole (DAPI) (Vector Laboratories, Burlingame, CA, USA) and observed with with Zeiss LSM880 Multi-Photon Confocal Microscope with a 63×/1.4 NA oil objective. Images were collected and processed using using Macro for ImageJ (Image J Fiji) software and sized in Adobe Photoshop 7.0. The γH2AX intensity per nucleus was determined with Adobe Photoshop 7.0 by measuring the fluorescence intensities normalized to cell count. Data was plotted and the statistical analysis was carried out using GraphPad Prism 9.1.0. The number of replicates for each experiment are reported in the figure legends. Statistical significance was calculated by two-way ANOVA using Dunnett’s correction for multiple comparisons. Asterisks indicate significant difference *: p < 0.05; **: p < 0.01; ***: p < 0.001; ****: p < 0.0001.

Gel filtration size exclusion chromatography

Recombinant TOP1, TOP2A and p53 was mixed in equimolar ratio in TNE buffer (10 mM Tris, pH.7.5, 300 mM NaCl, 1mM EDTA) and incubate at 4°C for 1hr. The protein mixture was then injected onto a Superose 6 Increase 5/150 column (Cytiva) and isocractically fractionated using TNE buffer using an AKTA Pure system. Calibration of Superose 6 Increase 5/150 column was performed with high molecular weight marker (calibration each fraction volume was 500 μl). Fractions were collected at 1 % of the total column volume (each fraction volume was 500 μl) and were analyzed by western blot for the co-elution of p53, TOP1, TOP2A.

Topoisomerase I relaxation assay

The activity assay of type I DNA topoisomerases was performed by analyzing the mobility of supercoiled plasmid DNA (FLIP-FUSE plasmid) on agarose gel as discussed previosly^36,62,67. Briefly, recombinant human TOP1 (12 nM) was incubated with increasing amount of either recombinant full length p53 (15–75 nM) and (5–30 nM) or full length cMYC (10–60 nM) for 5 min in ice in the TOP1 buffer (10 mM Tris-HCl pH 7.5, 150 mM KCl, 5 mM MgCl₂, 3% glycerol, 0.1 mM EDTA, 0.5 mM DTT, 15 μg/ml BSA). Supercoiled plasmid DNA (200 ng) was added to the mixture, incubated at 37°C for 15 min and reactions were terminated by adding TE-SDS 1% plus proteinase K (200 μg/ml). Reaction products were purified by phenol: chloroform and analyzed on 1.4% (w/v) agarose gel electrophoresis in 1XTAE buffer containing 15 μM chloroquine. To visualize the relaxed species, gels were stained with Diamond^™ Nucleic acid dye (Promega, H1181) and imaged under the Odyssey infrared scanner (Li-Cor). Each experiment was performed more than 4 times.

Ex-vivo relaxation assays were carried out with whole cell extracts from wild-type and p53^−/− HCT116 cells in presence and absence of NCS as described³⁶. Briefly, the indicated amount of whole cell extracts from untreated or NCS-treated (3 hrs, 400 ng/ml) wild-type and p53^−/− HCT116 cells were incubated with supercoiled DNA (200 ng) in TOP1 buffer at 37°C for 15 min, terminated with TE-SDS 1% plus proteinase K (200 μg/ml) and purified with phenol: chloroform. Purified products from relaxation assay were analyzed by agarose gel electrophoresis in presence of chloroquine as discussed before. The data presented are the mean of four seperate experiments. All densitometric quantifications of relaxation gels were carried out using Image J software.

Topoisomerase II decatenation assay

TOP2 activity was assayed by decatenation assay of kinetoplast DNA⁴⁹. Briefly, recombinant human TOP2A and TOP2B (10 nM) was incubated with increasing amount of recombinant full length full length p53 (10–40 nM) in ice for 5 min in the TOP2 reaction buffer (50 mM Tris-HCl pH 8.0, 10 mM MgCl2, 2 mM ATP, 0.5 mM DTT, 30 μg/ml BSA) before addition of 200 ng of kinetoplast DNA (Topogen) and was incubated at 37°C for 10 min. The reaction was stopped by adding TE-SDS 1% plus proteinase K (200 μg/ml) and reaction products were purified by phenol: chloroform and run on a 1% agarose gel containing 0.5 μg/ml SYBR Green. Image was detected under the ChemiDoc System (Bio-Rad).

Nuclear extracts were harvested for the TOP2 activity assay with a modified version of a previously described protocol³⁶. Briefly, for the ex vivo decatenation assay, the indicated amount of nuclear extracts from untreated or NCS-treated (3 hrs, 400 ng/ml) wild-type and p53^−/− HCT116 cells were incubated incubated with 200 ng of kinetoplast DNA in TOP2 reaction buffer (50 mM Tris pH 8.0, 10 mM MgCl2, 2 mM ATP, 0.5 mM DTT, 30 μg/ml BSA) at 37°C for 10 min. Reactions were stopped with the addition of adding TE-SDS 1% plus proteinase K (200 μg/ml) and purified with phenol:chloroform and checked on agarose gel containing 0.5 μg/ml SYBR Green and image was detected under the ChemiDoc System (Bio-Rad). All densitometric quantifications of relaxation gels were carried out using Image J software.

ChIP-PCR

To assess p53 binding to target sites of indicated protein, ChIP-PCR was performed on wild-type HCT116 cells treated (in biological duplicates) with or without nutlin (10 μM, 3 or 5hr) treatment using anti-p53. Cells were processed for ChIP-seq as discussed previously⁹¹. Briefly, cells were crosslinked by adding 1 ml of 16% formaldehyde (1% final concentration) and to neutralize add 2ml of 1M glycine to reach final concentration of 0.125M. samples were sonicated with a Bandelin probe sonicator to produce chromatin fragments ~200–400 bp on average. 2 μg of anti-p53 (DO-1) (Sc-126) antibody or pre-Immune normal rabbit IgG (Santa Cruz Biotechnology, catalog number: sc-2017) (as control) were mixed with 40 μl Protein A/G magnetic beads (Pierce, 88803) and incubated at 4°C for 4 hrs with controlled rotation. Eluted samples from 1×10⁷ cells were added to the Protein A/G-antibody complexes and incubated with rotation overnight at 4°C. Washing was performed as described previously⁹¹. The beads were then resuspended in 100 μl TE plus 0.5% SDS supplemented with proteinase K (500 μg/ml, Roche) and incubated for 4 hrs at 65°C. Samples were then purified by phenol-chloroform and precepitated by ethanol and sodium acetate.

Quantitative PCR was conducted using iTaq Universal SYBR Green Supermix (1725121, Bio-Rad) on the LightCycler 480 Instrument II (Roche). Each reaction mixture comprised 6 μl of 2× SYBR-Green PCR master mix and 1 μl of reverse transcriptase-generated cDNA diluted 10-fold in a final volume of 12 μl, with primers at 10 μM concentration. Thermocycling conditions consisted of initial denaturation (50°C for 2 min and 95°C for 5 min), followed by 45 cycles of amplification (95°C for 10 sec, 59°C for 1 min, and 72°C for 30 sec) and a melting step (95°C for 10 sec, 60°C for 1 min). The relative fold change in gene expression was determined as the ratio of the gene of interest expression level to that of IgG, with values normalized to 100% in wild-type cells. Primer sequences are provided in Table S2.

Quantitative Real Time PCR

The total RNA was isolated from indicated cells (5× 10⁶) using Direct-Zol TM RNA Miniprep kit (15596018, Zymo Research, R2051-A) according to the manufacturer’s protocols, which includes DNase I (supplied with the kit) treatment for each sample⁵. An aliquot of 1 μg of RNA was reverse-transcribed using a reverse transcription kit (18080–051, Invitrogen). Real-time PCR was performed with iTaq universal SYBR Green Supermix (1725121, Bio-Rad) on the LightCycler-^® 480 Instrument II (Roche). Reaction mixtures contained 6 μl of 2× SYBR-Green PCR master mix and l μl of reverse transcriptase-generated cDNA diluted 10-fold in a final volume of 12 μl containing primers at 10 μM. The thermocycling conditions include denaturation (50°C for 2 min and 95°C for 5 min), followed by amplification for 45 cycles (95°C for 10 sec, 59°C for 1 min, and 72°C for 30s) and melting (95°C for 10 sec, 60°C for 1 min). Relative fold change in gene expression was expressed as a ratio of the expression level of the gene of interest to that of β-actin RNA, with values in wild-type cells defined as 100%. The sequences of the primers are listed in Table S2.

Covalent adduct detection (CAD) in living cells

HO15.19-MYC-ER cells (1.5 ×10⁷) were treated as indicated in the scheme of top of each figure. Covalent adduct detection (CAD) assay was performed as discussed previously³⁶. Briefly, Cells were lysed by STE buffer (2% SDS, 0.5 M Tris-HCl pH 8.1, 10 mM EDTA) and disrupted using a Dounce homogenizer tight pestle shearing apparatus followed by phase seperation with phenol: chloroform. Protein/peptides covalently linked to DNA were precipitated with ethanol and sodium acetate and dissolved in elution buffer (10 mM Tris pH 8.0) following purification using QIAquick PCR purification kits (QIAGEN, 28006). After quantifying, each CAD sample (1–3 μg) was applied to a nitrocellulose membrane (Bio-Rad) using a vacuum slot-blot (Minifold II), blocked for 1 hr in 3% BSA and incubated with specific primary and secondary antibodies. The peptide linked DNA was detected using ECL chemiluminescence reaction under the Odyssey infrared scanner (Li-Cor).

TOP1 Cross Covalent adduct detection and ChIP for TOP1 (TOP1 Cross-CAD-seq)

We performed the TOP1 Cross-CAD-seq in the HO15.19-MYC-ER cells were treated (in biological duplicates) with nutlin (10 μM) with or without Eto (25 μM) as indicated in the figure scheme. Cells were processed for harvesting covalent DNA adducts as discussed previously^36,79. Briefly, CAD samples were sonicated with a Bandelin probe sonicator to produce chromatin fragments <1 kb on average. 2 μg of anti-TOP1 antibody (ab109374) were mixed with 40 μl Protein A/G magnetic beads (Pierce, 88803) and incubated at 4°C for 4 hrs with controlled rotation. Eluted CAD samples from 1×10⁷ cells were added to the Protein A/G-antibody complexes and incubated with rotation overnight at 4°C. Washing was performed as described⁷⁹ for the ChIP protocol but twice with every buffer and always in presence of 0.1% SDS. The beads were then resuspended in 100 μl TE plus 0.5% SDS supplemented with proteinase K (500 μg/ml, Roche) and incubated for 4 hrs at 65°C. Samples were then purified by MiniElute PCR purification kit (QIAGEN, 28006).

Library preparation and sequencing of CAD-seq and Cross-CAD-seq

To cleave off covalently bound tyrosyl residues from immunoprecipitated DNA samples, the samples were additionally treated with ExoVII (0.5–1U per 10 ng of DNA) and purified by PCR purification Kit (QIAGEN). CAD-seq libraries were prepared using the ThruPLEX DNA-seq kit (Takara, R400676) following manufacture’s protocol. Next, library size selection was performed using AMPure XP beads (Beckman, A63880) as previously described^36,79. Efficiency of size selection of the Cross-CAD libraries was assessed with a Bioanalyzer using an Agilent High Sensitivity DNA Kit (D1000) and sequenced on NextSeq 2000 (100 cycle) in Single End mode with 100 bp reads. NCI sequencing core, (Bethesda, Maryland, USA).

Data analysis

Data analysis followed the previously discussed protocol^36,79. All of the steps were performed using Biowulf (NCI facility). Briefly, the fastq files were quality controlled with FastQC and MultiQC, trimmed with cutadapt, aligned to Rn6 reference genome with bowtie2, deduplicated, sorted and indexed using samtools and Picard. BigWig files for visualization were generated using Deeptools from combined reads of two replicas. The short reads average distributions near TSSs and along normalized gene bodies were generated using Deeptools. Combined reads from two replicas were used for average profile generation unless stated otherwise. Only the protein coding genes from Ensembl 76 database were used to generate profiles. The table of p53-target genes for Rat (Table S1) was generated using the Human p53-target gene table and Ensemble (https://useast.ensembl.org/index.html). The corresponding Bed file for the p53-target genes⁸¹ of rat genome was built by UCSC genome browser table (https://genome.ucsc.edu/cgi) (selecting genome-Rat; group- Genes and Gene predictions and track-UCSC RefSeq (refgene)). This bed file was used for generating the metagene plot using Deeptools to check the distribution near TSSs and along normalized gene bodies of TOP1 for p53-target genes. The expression based selection of genes was carried out as previously described³⁶. Top (90–100%), middle (61–90%) or low (below 60%) expressed genes were determined using previously published RNA-seq⁹². The deep-sequencing data reported in this paper were deposited in the Gene Expression Omnibus (GEO) under accession number GSE243792 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE243792) (reviewer token number-wfsbgkgmdtcbpel). Rat RNA-seq data was published under accession number GSE167513⁹².

RNA-seq

Wild-type (si control) and TOP1 and TOP2A knockdown (si TOP1+ si TOP2A) HCT116 cells were untreated or treated (in biological duplicates for the sequencing assays and in quadruplicates for the qPCRs) with nutlin 10 μM for 3–5 hrs. The total RNA was extracted from the specified cells (1×10⁷) using the Direct-ZolTM RNA Miniprep kit (15596018, Zymo Research, R2051-A), following the manufacturer’s instructions, which involved DNase I treatment (provided with the kit) for each sample. Next, the RNA library was prepared by using the polyA-selected RNASeq. Efficiency of size selection of the RNA samples was assessed with a Bioanalyzer using an Agilent High Sensitivity DNA Kit (D1000) and sequenced on polyA selected RNASeq (100 cycle) in Single End mode with 100 bp reads (~20–30M read per sample). NCI sequencing core (Bethesda, Maryland, USA)ran the RNA-seq for us.

RNA sequencing data analysis

Transcript quantification from fastq files

Transcript quantification was performed using Salmon (v. 1.2.1). We first created a Salmon index corresponding to all human cDNA from the Ensembl v111 release using the ‘salmon index’ command. This index was then used to quantify the transcript expression levels from paired-end fastq files using the ‘salmon quant’ command. Samples were processed using snakemake v7.32.4 on HPC resources from the University of Minnesota. Gene annotation for downstream analyses were carried out using the Annotation Hub R package and EnsDb.Hsapiens.v111 obtained from Bioconductor in RStudio (R version 4.3.0).

Normalization to qRT-PCR data

TOP1 and TOP2A silencing leads to changes in the total mRNA per cell, invalidating the assumption of similar total RNA expression used in typical normalization methods. Using library size normalization or median of ratios normalization did not show the expected siRNA-mediated downregulation of TOP1 and TOP2A that was evident from qRT-PCR. Therefore, we globally rescaled the RNAseq TPM values for each sample such that the beta-actin (ACTB)-normalized fold changes in TOP1 and TOP2A under silencing were equal to the average ACTB-normalized fold change in these transcripts as measured by qRT-PCR.

We first normalized the TPM of TOP1 and TOP2A for each sample by the TPM of ACTB and then averaged this normalized value across the two replicates for each condition j.

$$TOP{1}_{\mathit{\text{norm}}}(j)=\frac{1}{2}\sum _{i=1}^2{(\frac{TOP1}{\mathit{\text{ACTB}}})}_i$$

$$TOP2A_{\mathit{\text{norm}}}(j)=\frac{1}{2}\sum _{i=1}^2{(\frac{TOP2A}{\mathit{\text{ACTB}}})}_i$$

We then calculated the fold-change in the expression of these transcripts for each sample.

$$TOP{1}_{FC,\mathit{\text{RNASeq}}}(j)=TOP{1}_{\mathit{\text{norm}}}(j)/TOP{1}_{\mathit{\text{norm}}}(\mathit{\text{siControl}})$$

$$TOP2A_{FC,\mathit{\text{RNASeq}}}(j)=TOP2A_{\mathit{\text{norm}}}(j)/TOP2A_{\mathit{\text{norm}}}(\mathit{\text{siControl}})$$

These quantities are now analogous to the fold changes in these transcripts as measured by qRT-PCR. Thus, we can calculate a normalization factor Z_j for each sample such that the RNASeq measurements best match the qRT-PCR results.

$$Z_j=\frac{Z_{j,TOP1}+Z_{j,TOP2A}}{2}$$

Where,

$$Z_{j,TOP1}=TOP{1}_{FC,\mathit{\text{qPCR}}}(J)/TOP{1}_{FC,\mathit{\text{RNASeq}}}(j)$$

$$Z_{j,TOP2A}=TOP2A_{FC,\mathit{\text{qPCR}}}(J)/TOP2A_{FC,\mathit{\text{RNASeq}}}(j)$$

J is the qRT-PCR sample corresponding to the same conditions as the RNASeq sample j.

We now multiply this normalization factor to the entire sample to get normalized transcript expression values. For a gene i in sample j, the normalized expression is given by –

$$\mathit{\text{normalizedExpression}}(i,j)=TPM(i,j)*Z_j$$

Analysis and visualization of normalized gene expression

Normalized gene expression matrices were exported to .csv files and were visualized using MATLAB 2023a (Mathworks). Density plots were generated using the in-built histogram2 function. Heatmaps were generated using the in-built heatmap function. For heatmap generation, the gene expression was also first log₂+1 normalized and then centered such that the mean normalized expression of the gene across all samples is zero.

Quantification and statistical analysis

Statistical testing

All statistical tests were performed using Graphpad Prism version 8 or 9 or statistical functions in R using the tests described for each experiment. *: p < 0.05; **: p < 0.01; ***: p < 0.001; ****: p < 0.0001. Detailed information about statistical tests is provided in figure legends for respective figures.

Supplementary Material

2

Table S1. List of p53-target genes for Rat, related to Figure 7. Panel A) Rat gene stable ID. B) Rat gene name.

Highlights.

Hyperactive MYC-topoisome directly and acutely damages DNA.

ATM-activating DNA damage triggers the p53 response and MYC degradation.

The p53-topoisome (a p53/TOP1/TOP2 complex) increases p53 target gene expression.

p53 and its topoisome, unlike MYC, induce TDP1 &2 to repair topoisomerase-damaged DNA.