Introduction

Somatic alterations in the TP53 gene, encoding the tumor suppressor p53, are the most common events in human tumors (Hollstein et al., 1991). The p53 protein is a stress sensor stabilized and activated in response to potentially oncogenic insults. In its tetrameric active form, wildtype (WT) p53 can trigger a transcriptional program to induce various responses including cell cycle arrest, senescence, apoptosis or metabolic changes (Beckerman and Prives, 2010). In human cancers, about 75% of all TP53 alterations are missense mutations, and the eight most frequent (“hotspot”) missense mutations (R175H, Y220C, G245S, R248Q, R248W, R273H, R273C and R282W) affect six residues localized in the p53 DNA-binding domain (DBD) and account for 20% of all TP53 mutations (Hainaut and Pfeifer, 2016). The fact that most TP53 mutations are missense substitutions suggested that cells expressing mutant p53 might have a selective advantage over cells lacking p53, and evidence for this was first gained by expressing various p53 mutants in p53-null cells, and observing enhanced tumorigenic potential in nude mice or increased plating efficiency in agar cell culture (Dittmer et al., 1993). Mouse models expressing hotspot p53 mutants next helped to define potential mechanisms accounting for accelerated tumorigenesis. First, a dominant-negative effect (DNE) may be observed in heterozygotes, if the mutant p53 inhibits the WT p53 protein in hetero-tetramers (Gencel- Augusto and Lozano, 2020). Evidence that this mechanism accounts for accelerated tumorigenesis was notably reported in human leukemia (Boettcher et al., 2019). A second mechanism is a gain-of-function (GOF), i.e. the acquisition by mutant p53 of new oncogenic properties (Amelio and Melino, 2020; Stein et al., 2020). Although the DNA sequence specificity of p53 mutants is impaired, leading to a loss of WT functions (LOF), many p53 mutant proteins are stabilized in the cell and might engage in aberrant interactions with other transcription factors or chromatin-modifying complexes, leading to the acquisition of GOF phenotypes (Kim and Deppert, 2007; Pfister and Prives, 2017). However, a few recent studies challenged the pathological importance of mutant p53 GOF, or at least its relevance in anti-cancer therapeutic strategies (Aubrey et al., 2018; Boettcher et al., 2019; Wang et al., 2023). An alternative hypothesis emerged, postulating that the tumorigenic properties of a p53 mutant might result from a separation-of-function (SOF), if the mutant retains pro-proliferative or pro-survival functions of WT p53 while losing its tumor suppressive activities (Kennedy and Lowe, 2022). The concept of SOF was first proposed for TP53 exon-6 truncating mutations, which mimic a naturally occurring alternative p53 splice variant (Shirole et al., 2016), but may apply to some p53 missense mutations, for example p53^R248W (Humpton et al., 2018). Importantly, the analysis of various mutant p53 alleles in vivo appears crucial for a better understanding of their contribution to tumorigenesis, because GOF/SOF phenotypes may vary depending on the mutated residue, its specific mutation, the cellular context or genetic background (Dibra et al., 2024; Hanel et al., 2013; Kadosh et al., 2020; Kim and Lozano, 2018; Kotler et al., 2018; McCann et al., 2022; Xiong et al., 2022).

In human cancers, the TP53^Y220C mutation is the most frequent missense mutation outside of the DNA-binding surface of p53 (Hainaut and Pfeifer, 2016). The somatic TP53^Y220C mutation accounts for over 100,000 new cancer cases per year worldwide, and a germline TP53^Y220Cmutation was reported in 15 families with the Li-Fraumeni syndrome (LFS) of cancer predisposition (Bouaoun et al., 2016). This mutation causes a structural change in the DBD localized at the periphery of the β−sandwich region of the protein, far from the surface contact of DNA. The mutation from a tyrosine to a cysteine markedly lowers the thermodynamic stability of the DBD, leading to a largely unfolded and inactive protein at body temperature (Joerger et al., 2006), and a molecule designed to bind p53^Y220C and restore WT protein conformation appears as a promising anti-cancer drug (Dumbrava et al., 2022). Analyses of the impact of p53^Y220C in cancer cell lines led to conflicting results. On one hand, overexpression of p53^Y220C in a p53-null cell line increased its capacity for migration or invasion (Zhou et al., 2022), and a decreased expression of p53^Y220C, caused by RNA interference or various chemical compounds, promoted cell death or decreased the migratory or invasive capacities of cells (Tseng et al., 2022; Vikhanskaya et al., 2007; Zhou et al., 2022). On the other hand, the removal by CRISPR/Cas9 of p53 mutants with reported GOF - including p53^Y220C - in diverse cancer cell lines had no significant impact on cell survival or proliferation, and the cell death caused by RNA interference against p53^Y220C might result from non-specific toxic effects (Wang et al., 2023). Here, to gain a better understanding of the impact of the p53^Y220C mutation in vivo, we generated a mouse model with a targeted Trp53^Y217C mutation - the mouse equivalent to human TP53^Y220C (Figure S1) - and analyzed animals and cells carrying one or two copies of the mutant allele.

Results

Targeting of a p53^Y217C mutation in the mouse

We used homologous recombination in 129/SvJ ES cells to target the p53^Y217C mutation at the mouse Trp53 locus. Mice expressing p53^Y217C conditionally were generated by using a targeting vector containing transcriptional stops flanked by LoxP sites (LSL) upstream of coding sequences, and the p53^Y217C mutation in exon 6 (Figure 1A-D). p53^+/LSL-Y217C mice were then crossed with mice carrying the PGK-Cre transgene (Lallemand et al., 1998) to excise the LSL cassette and obtain p53^+/Y217C mice, expressing the mutant protein constitutively. After two backcrosses over a C57BL/6J background, mouse embryonic fibroblasts (MEFs) were prepared from intercrosses of p53^+/Y217C mice (Figure 1E). We extracted RNAs from p53^Y217C/Y217C MEFs then sequenced p53 mRNAs to verify that the p53^Y217C coding sequence was identical to the wildtype p53 (p53^WT) sequence, except for the desired missense mutation and a silent mutation introduced to facilitate mouse genotyping (Figure 1F). The quantification of p53 mRNA levels from WT and p53^Y217C/Y217C MEFs next revealed similar transcription from both alleles (Figure 1G).

Figure 1.

Deficient p53-dependent stress responses in p53^Y217C/Y217C cells

We used western blots to analyze protein extracts from p53^+/+, p53^+/-, p53^+/Y217C, p53^Y217C/Y217C and p53^-/- MEFs, unstressed or treated with Nutlin, an antagonist of Mdm2, the E3 ubiquitin ligase for p53 (Vassilev et al., 2004). Results indicated a high increase in p53 abundance in untreated and Nutlin-treated p53^Y217C/Y217C MEFs and a moderate increase in untreated and Nutlin-treated p53^+/Y217C MEFs, compared to WT cells (Figures 2A and S2). Protein levels for p21 and Mdm2, the products of two classical p53 target genes, appeared similar in p53^+/+ and p53^+/Y217C MEFs, and were undetectable or markedly decreased in p53^Y217C/Y217C and p53^-/- MEFs (Figures 2A and S2). Accordingly, p53^Y217C appeared unable to transactivate the Cdkn1a/p21 and Mdm2 genes or to bind their promoters (Figure 2B). The fractionation of cells before or after treatment with doxorubicin, a clastogenic drug, indicated that p53^WT accumulated in the nucleoplasm and chromatin fractions in response to DNA damage, whereas p53^Y217C appeared mostly cytoplasmic in both untreated and doxorubicin-treated cells, and undetectable or barely detectable in chromatin fractions (Figures 2C and S3). Furthermore, when Nutlin-treated MEFs were analyzed by immunofluorescence, p53^WT was only detected in nuclei, whereas p53^Y217C could be observed in nuclear and cytoplasmic compartments (Figure 2D). We next analyzed two well-known p53-mediated responses to DNA damage, i.e. cell cycle arrest in MEFs and apoptosis in thymocytes (Kastan et al., 1992; Lowe et al., 1993). WT MEFs exposed to increasing doses of ɣ-irradiation (3 Gy or 12 Gy) exhibited significant increases in G1/S ratios, whereas G1/S ratios were similar before or after irradiation in p53^Y217C/Y217C and p53^-/- MEFs (Figures 2E and S4). Furthermore, thymocytes recovered from irradiated WT mice underwent a massive apoptotic response, with an almost three-fold increase in apoptotic cells after a 10 Gy irradiation. By contrast, no increase in apoptotic cells was observed upon irradiation in the thymi from p53^Y217C/Y217C mice, and apoptotic thymocytes were equally rare in irradiated p53^Y217C/Y217C and p53^-/- mice (Figures 2F and S5). Together, these results indicated that the p53^Y217C mutation altered the abundance and intracellular distribution of the p53 protein, associated with a decrease in DNA-bound protein, and that p53^Y217C had completely lost the ability to induce cell cycle arrest and apoptosis upon DNA damage.

Figure 2.

Impact of p53^Y217C on mouse development and parturition

We next determined the impact of the p53^Y217C mutation in vivo, by comparing mouse cohorts generated from intercrosses of heterozygous p53^+/- or p53^+/Y217C mice resulting from ≥ 5 backcrosses to the C57BL6/J background. Intercrosses of p53^+/- mice are known to yield a reduced proportion of p53^-/- mice, that is mainly due to defects in neural tube closure causing exencephaly in a fraction of p53^-/- female embryos (Armstrong et al., 1995; Sah et al., 1995) and, to a lesser extent, to lung dysfunction affecting a fraction of p53^-/- neonates (Tateossian et al., 2015). At weaning (on the 21^st day postpartum or P21), we observed one p53^-/- female mouse for 3.5 p53^-/- males from p53^+/- intercrosses, an underrepresentation of females consistent with frequencies reported in earlier studies (Armstrong et al., 1995; Sah et al., 1995). Strikingly, the underrepresentation of female mice at weaning age was even more acute for p53^Y217C/Y217C animals, with only one p53^Y217C/Y217C female for 19 p53^Y217C/Y217C males from p53^+/Y217C intercrosses (Figure 3A). We next analyzed p53^Y217C/Y217C embryos, generated from heterozygous intercrosses, or from heterozygous-homozygous (p53^+/Y217C x p53^Y217C/Y217C) crosses, at embryonic days E12.5-E16.5. An underrepresentation of p53^Y217C/Y217C female embryos was not observed, but 11/26 (42%) female embryos exhibited developmental abnormalities, including 10 with exencephaly, whereas all the male embryos appeared normal (Figure 3B). Importantly, the frequency of p53^Y217C/Y217C female embryos with exencephaly (38.5%) was much higher than the reported frequency (0-8%) of p53^-/- female embryos with exencephaly of C57BL/6 x 129/Sv genetic background (Donehower et al., 1992; Sah et al., 1995), suggesting a stronger effect of the p53^Y217C mutant on female embryonic development.

Figure 3.

In p53^-/- female mice, exencephaly was previously correlated with stochastic aberrant X chromosome inactivation, with a decreased expression of Xist and an increase in bi-allelic expression of X-linked genes including Maob, Pls3 and Usp9x (Delbridge et al., 2019). We prepared neurospheres from p53^+/+, p53^-/- and p53^Y217C/Y217C female embryos and quantified mRNAs for these genes in neurospheres. Compared to neurospheres from WT animals, p53^Y217C/Y217C and p53^-/- neurospheres exhibited an apparent decrease in Xist expression and significantly higher levels of Maob, Pls3 and Usp9x (Figure S6). Thus, as for female p53^-/- mice, aberrant chromosome X inactivation may contribute to the underrepresentation of p53^Y217C/Y217C female mice. Of note however, defects in chromosome X inactivation did not appear to be more pronounced in p53^Y217C/Y217C neurospheres compared to p53^-/- neurospheres.

To further analyze the impact of the p53^Y217C mutation during development, we also analyzed its potential effect in embryos lacking Mdm2 or Mdm4, two major p53 regulators. The embryonic lethality resulting from Mdm2 or Mdm4 loss is rescued by a concomitant p53 deficiency (Bardot et al., 2015; Finch et al., 2002; Jones et al., 1995; Migliorini et al., 2002; Montes de Oca Luna et al., 1995; Parant et al., 2001), which provides a powerful assay for analyzing the functionality of p53 mutant alleles (Bardot et al., 2015; Iwakuma et al., 2004; Marine et al., 2006; Toledo et al., 2006). We identified Mdm2^-/- p53^Y217C/Y217C and Mdm4^-/- p53^Y217C/Y217C viable mice of both sexes, consistent with a major loss of canonical WT p53 activities in the p53^Y217C mutant (Table S1). Of note, a deficit in homozygous p53^Y217C/Y217C females was also observed at weaning among animals lacking Mdm2 or Mdm4. Interestingly however, the female to male ratio for p53^Y217C/Y217C animals appeared markedly increased in genetic backgrounds with Mdm4 loss (1/6) or haploinsufficiency (4/15), but only marginally increased (if at all) in genetic backgrounds with Mdm2 loss (1/11) or haploinsufficiency (1/11). This suggested that altering the levels of p53 regulators, particularly Mdm4, might improve the survival of p53^Y217C/Y217C females (Table S1).

Importantly, the high frequency (42%) of abnormal p53^Y217C/Y217C female embryos at E12.5-E16.5 could only partially account for the acute deficit in p53^Y217C/Y217C females observed at weaning, suggesting either embryonic abnormalities undetectable macroscopically, or that an important fraction of p53^Y217C/Y217C females died later in development or postpartum. We performed additional crosses to analyze female to male ratios at postpartum day 0 or 1 (P0-P1) and observed 20 females for 18 males for p53^+/+ animals, but only 5 females for 22 males for p53^Y217C/Y217C animals (Figure 3C). Furthermore, only one of these p53^Y217C/Y217C females reached weaning age, and the 4 other neonates died before P2. Altogether, these data led to conclude that the p53^Y217C mutation caused female-specific developmental abnormalities at a higher penetrance than a null allele, leading to a perinatal (late embryonic or early postpartum) lethality for most homozygous mutant females.

Interestingly, we observed that p53^Y217C/Y217C males were fertile, and included them in some of our crosses to generate homozygous mutants at higher frequencies (Figure 3B-C and Table S1). By contrast, we were able to mate two p53^Y217C/Y217C adult females with a p53^+/Y217C male, and both got pregnant but had to be sacrificed due to complications while giving birth. In both cases, extended labor (> 24 hr) was the main sign of dystocia (Table S2). These observations suggest that, even for the rare p53^Y217C/Y217C females able to reach adulthood, the p53^Y217C mutation caused pathological traits not observed in p53^-/- females, because the loss of p53 was not previously reported to cause dystocia (Embree-Ku and Boekelheide, 2002; Guimond et al., 1996; Hu et al., 2007). Together, our data indicated that the p53^Y217C mutation not only abolished canonical p53 activities, but also conferred additional effects compromising female perinatal viability or parturition.

Impact of p53^Y217C on tumorigenesis

We next analyzed the impact of the p53^Y217C mutation on the onset and spectrum of spontaneous tumors in mice. We first compared p53^+/Y217C, p53^+/- and p53^+/+ mouse cohorts for 2 years. p53^+/Y217C and p53^+/- cohorts exhibited similar spontaneous tumor onset and spectrum, with most mice developing sarcomas during their second year of life (Figure S7). Thus, at least for tumors arising spontaneously, the p53^Y217C mutant protein did not appear to exert a dominant negative effect over the wildtype p53 protein.

We next compared p53^Y217C/Y217C and p53^-/- mouse cohorts. Because of the difficulty to generate p53^Y217C/Y217C females, we restricted our comparison to males, to avoid potential biases that might result from different sex ratios. All p53^-/- mice are known to die in less than 1 year, from thymic lymphomas in most cases, or more rarely from sarcomas (Donehower et al., 1992; Jacks et al., 1994). p53^Y217C/Y217C males died faster than their p53^-/- counterparts: all the p53^Y217C/Y217C males were dead by the age of 7 months, whereas more than 20% of the p53^-/- males were still alive at that age (Figure 4A). Furthermore, most p53^-/- and p53^Y217C/Y217C mice died from thymic lymphomas, but histological analysis of tumor organs revealed that the lymphomas in p53^Y217C/Y217C males were more aggressive and invasive, with sites of metastases notably including the lungs and spleen (Figure 4B-C). Thus, the p53^Y217C mutation not only abolished canonical tumor suppressive activities but also apparently conferred oncogenic properties to the encoded protein. Consistent with this, when we performed an RNA-seq analysis of thymi from 8 weeks-old p53^+/+, p53^-/- and p53^Y217C/Y217C males, 81.5 % of the 717 differentially expressed genes indicated a loss of function (LOF) in the p53^Y217C mutant, but genes suggesting a gain of function (GOF) or a separation of function (SOF) were also observed (Figure S8A and Table S3). Of note, among the differentially expressed genes corresponding to a LOF in the p53^Y217C mutant were Bbc3/Puma, Cdkn1a/p21 and Zmat3 (Figure S8B), three p53-target genes known to play a major role in p53-mediated tumor suppression (Brennan et al., 2023).

Figure 4.

We next performed a comparative analysis focusing on p53^Y217C/Y217C and p53^-/- thymi, which revealed 192 differentially expressed genes (Figure 5A-B, Table S4). An analysis of these data with GOrilla, the Gene Ontology enRIchment anaLysis and visuaLizAtion tool (Eden et al., 2009), indicated that 141 of these genes were associated with at least one gene ontology (GO) term, and revealed a significant enrichment for genes associated with white blood cell chemotaxis/migration and inflammation (Figure 5C and Figure S9). Among these genes were notably Ccl17, Ccl9, Ccr3, Cxcl10, S100a8 and S100a9, six genes associated each with 10-15 GO terms related to white blood cell behavior or inflammation (Table S5). We next used RT- qPCR to quantify the expression of these genes in thymi from 8 weeks-old p53^+/+, p53^Y217C/Y217C and p53^-/- male mice, and found their expression to be significantly increased in p53^Y217C/Y217C thymi compared to p53^-/-, or to both p53^+/+ and p53^-/- mice, consistent with gain or separation of function (GOF/SOF) effects in the p53^Y217C mutant (Figure 5D). Furthermore, when we compared the transcriptomes of p53^Y217C/Y217C and p53^-/- thymocytes by gene set enrichment analysis (Subramanian et al., 2005), we found 13 gene sets significantly enriched in p53^Y217C/Y217C cells (with normalized enrichment scores > 2), among which 3 (“antimicrobial peptides”, “chemokine receptors bind chemokines” and “defensins”) were related to an immune response (Figures 5E and S10A), consistent with an inflammatory response associated with the p53^Y217C mutant. Other gene sets significantly enriched in p53^Y217C/Y217C cells notably included 5 sets (“electron transport chain”, “respiratory electron transport ATP synthesis by chemiosmotic coupling and heat production by uncoupling proteins”, “mitochondrial translation”, “respiratory electron transport” and “oxidative phosphorylation”) related to mitochondria function, and 3 sets (“deposition of new CENPA-containing nucleosomes at the centromere”, “arginine methyltransferases methylate histone arginines” and “PRC2 methylates histones and DNA”) related to chromatin plasticity (Figure 5E and S10B-C). These gene sets, differentially expressed between p53^Y217C/Y217C and p53^-/- cells, might contribute to accelerated tumorigenesis in p53^Y217C/Y217C mice, given the reported impact of inflammation, metabolism changes and epigenetic plasticity on cancer evolution (Feinberg et al., 2006; Greten and Grivennikov, 2019; Kroemer and Pouyssegur, 2008).

Figure 5.

Discussion

In this report, we generated a mouse model with a Trp53^Y217C allele, the murine equivalent of the human hotspot mutant TP53^Y220C. The analysis of this mouse model indicated that the p53^Y217C mutation leads to the loss of many canonical wildtype p53 activities, notably the capacity to transactivate target genes important for tumor suppression (e.g. Bbc3, Cdkn1a and Zmat3). The fact that p53^Y217C can rescue the embryonic lethality caused by a deficiency in Mdm2 or Mdm4 is also consistent with a major loss of canonical wildtype p53 functions in this mutant.

In addition, our analyses provide evidence that the p53^Y217C mutation exhibits oncogenic effects. The possibility that a mutant p53 might acquire oncogenic functions was first suggested over 30 years ago (Dittmer et al., 1993), but this notion became controversial in recent years. Indeed, studies of a few p53 mutants in human acute myeloid leukemia and a mouse model of B cell lymphoma indicated that their increased tumorigenicity would result from dominant- negative effects (DNE) rather than gain-of-function (GOF), and a recent study further suggested that the putative GOF of many p53 mutants would not be required to sustain the expansion of tumors (Aubrey et al., 2018; Boettcher et al., 2019; Wang et al., 2023). Here, however, we observed accelerated tumorigenesis in p53^Y217C/Y217C mice, but did not observe any evidence of DNE in p53^+/Y217C animals. Importantly, the fact that p53^Y217C did not exhibit a DNE over the WT protein is consistent with the report that Li-Fraumeni patients carrying an heterozygous p53^Y220C mutation display a similar age of cancer onset than Li-Fraumeni patients with an heterozygous non-sense p53 mutation (Xu et al., 2014). Furthermore, our evidence of an oncogenic GOF leading to aggressive metastatic tumors in p53^Y217C/Y217C male mice is consistent with the oncogenic GOF recently attributed to the p53^Y220C mutant in male patients with glioblastoma (Rockwell et al., 2021), or with experiments suggesting that the expression of p53^Y220C in p53-null cells, or its overexpression in MCF-10A cells, may increase their capacity for migration or invasion (Pal et al., 2023; Zhou et al., 2022). Our transcriptomic analyses suggested that the p53^Y217C mutant might accelerate tumorigenesis by promoting inflammation, a hallmark of cancer (Hanahan, 2022) previously associated with a few other hotspot p53 mutants with oncogenic GOF (Agupitan et al., 2020; Behring et al., 2019; Ham et al., 2019; Zhao et al., 2024). Interestingly, our GSEA comparison of transcriptomes from p53^Y217C/Y217C and p53^-/- male thymic cells indicated differences related to inflammation and chemokine signaling, but also to mitochondria function and chromatin plasticity, and a link between inflammation, mitochondrial copper and epigenetic plasticity was recently demonstrated (Solier et al., 2023).

A striking and unexpected effect of the p53^Y217C mutation was its severe impact on the perinatal viability of female mice, which led to observe only 1 p53^Y217C/Y217C female for 19 p53^Y217C/Y217C males at weaning from heterozygous intercrosses. The perinatal lethality of p53^Y217C/Y217C females correlated with a high frequency (38.5%) of exencephalic females at E12.5-E16.5 embryonic ages. By comparison, the female to male ratio at weaning was 1/3.5 for p53^-/- mice, and only 0-8% of exencephalic female p53^-/- embryos were reported in similar genetic backgrounds (Armstrong et al., 1995; Donehower et al., 1992; Sah et al., 1995). These observations provide evidence of teratogenic effects for the p53^Y217C mutant protein. Three other mouse p53 models were previously found to cause exencephaly at a higher frequency than a p53 null allele : p53^NLS1, with 3 mutations at residues 316-318 affecting a nuclear localization signal and leading to a predominantly cytoplasmic localization of p53^NLS1 in most cells (Regeling et al., 2011); p53^N236S, a mouse model of p53^N239S, a recurrent but uncommon mutant in human cancers (Zhao et al., 2019); and Bim^+/- p53^-/- mice, combining p53 loss with an haploinsufficiency in the cytoplasmic proapoptotic regulator Bim/Bcl2l11 (Delbridge et al., 2019). Interestingly, the female-specific lethality of p53^N236S mice was proposed to result from increased Xist expression, whereas a decrease in Xist expression was observed in Bim^+/- p53^-/- mice, suggesting distinct causes for neural tube defects (Delbridge et al., 2019; Zhao et al., 2019). We observed decreased Xist expression and an increased expression of X-linked genes in p53^Y217C/Y217C females, which is not consistent with the mechanism proposed for exencephaly in the p53^N236S mouse model, but might rather reflect a loss of WT p53 function (Delbridge et al., 2019). In addition, the analysis of the p53^NLS1 and Bim^+/- p53^-/- mouse models make it tempting to speculate that, in female p53^Y217C/Y217C embryos, the abundance of p53^Y217C in the cytoplasm might perturb mitochondria function (Blandino et al., 2020) to further promote neural tube closure defects. Consistent with this possibility, p53 mutant proteins accumulating in the cytoplasm were reported to inhibit autophagy (Morselli et al., 2008), a key determinant for mitochondrial integrity (Rambold and Lippincott-Schwartz, 2011), and autophagy impairment may promote neural tube defects (Fimia et al., 2007; Ye et al., 2020).

A particularly intriguing question is whether common mechanisms might contribute to the teratogenic effects causing the perinatal lethality of p53^Y217C/Y217C females and the oncogenic effects accelerating tumorigenesis in p53^Y217C/Y217C males. Interestingly, our data indicated that inflammation might promote oncogenesis in p53^Y217C/Y217C males, and inflammation was proposed to promote neural tube defects in a few mouse models (Lian et al., 2004; Zhao et al., 2013). Furthermore CD44, which drives inflammation and cancer progression (Solier et al., 2023), was recently identified as a key gene for the diagnosis and early detection of open neural tube defects (Karthik et al., 2022). Independently, our mating attempts with p53^Y217C/Y217C mice indicated that p53^Y217C/Y217C pregnant females exhibited dystocia, and parturition is considered to be a finely regulated inflammatory process (Zhang and Wei, 2021). Of note, dystocia was not observed in p53^-/- mice, but was previously reported in pregnant females with a combined deficiency of p53 and FasL (Embree-Ku and Boekelheide, 2002). The fact that p53^Y217C/Y217C females shared phenotypic traits with both Bim^+/- p53^-/- mice (neural tube defects) and FasL^-/- p53^-/- mice (dystocia) may seem relevant, because Bim and FasL are both regulators of apoptosis and autophagy that also regulate immune responses (Sionov et al., 2015; Taylor and Ng, 2018). Altogether, these data raise the possibility that inflammation may cause accelerated tumorigenesis in p53^Y217C/Y217C male mice and might also contribute to promote neural tube defects in p53^Y217C/Y217C female embryos or dystocia in p53^Y217C/Y217C pregnant females.

In conclusion, we generated a mouse model expressing p53^Y217C, the murine equivalent of human p53^Y220C, and its analysis revealed that the gain-of-function (or separation-of-function) of a hotspot mutant p53 may not be limited to oncogenic effects. This is an important notion to consider given the ever-expanding functions ascribed to p53 (Hu, 2009; Rakotopare and Toledo, 2023; Voskarides and Giannopoulou, 2023; Vousden and Lane, 2007).

Materials and methods

LSL-Y217C construct

We used mouse genomic Trp53 DNA from previous constructs, containing a portion of intron 1 with a LoxP-Stop-LoxP (LSL) cassette (Ventura et al., 2007) in which the puromycin-resistance gene was replaced by a neomycin-resistance gene (Simeonova et al., 2013). The point mutation (A to G) encoding a Tyr to Cys substitution (TAT to TGT) at codon 217, together with a silent mutation nearby creating a Ban II restriction site, were introduced by PCR directed mutagenesis by using primers Y217C-F and Y217C-R (See Table S6 for primer sequences). The resulting p53^LSL-Y217C targeting vector was fully sequenced before use.

Targeting in ES Cells and Genotyping

129/SvJ embryonic stem (ES) cells were electroporated with the targeting construct linearized with Not I. Two recombinant clones, identified by long-range PCR and confirmed by internal PCR and Southern Blot, were injected into C57BL/6J blastocysts to generate chimeras, and germline transmission was verified by genotyping their offspring. In vivo excision of the LSL cassette was performed by breeding p53^+/LSL-Y217C male mice with females carrying the PGK-Cre transgene (Lallemand et al., 1998) and genotyping their offspring. The p53^Y217C (p53^YC) mutation was routinely genotyped by PCR using primers c and d (Figure 1, Table S6) followed by BanII digestion. Tumorigenesis studies were performed on mouse cohorts resulting from ≥ 5 generations of backcrosses with C57BL6/J mice. All experiments were performed according to IACUC regulations as supervised by the Ethics Committee of Institut Curie.

Cells and Cell Culture Reagents

Mouse embryonic fibroblasts (MEFs) isolated from 13.5-day embryos were cultured in a 5% CO2 and 3% O2 incubator, in DMEM GlutaMAX (Gibco), with 15% FBS (Biowest), 100 mM 2- mercaptoethanol (Millipore), 0.01 mM Non-Essential Amino-Acids, and penicillin/streptomycin (Gibco) for ≤ 6 passages. Cells were irradiated with a Cs γ-irradiator or treated for 24 hr with 10 mM Nutlin 3a (Sigma-Aldrich) or 10 mM Doxorubicin (Sigma- Aldrich). Neurospheres were prepared by isolating neural stem/progenitor cells (NPCs) from the lateral and median ganglionic eminences of E14.5 embryos. NPCs were cultured in a 5% CO2 incubator, in DMEM/F12 medium (with L-Glutamine, Invitrogen) supplemented with 1% B-27 (Invitrogen), 1% N2 (Invitrogen), 0.3% glucose (Invitrogen), 25 μg/ml insulin (Sigma-Aldrich), 20 ng/ml EGF (PeproTech), 10 ng/ml bFGF (PeproTech) and penicillin/streptomycin (Gibco). Neurospheres were dissociated once a week by Accutase treatment and mechanically, then seeded at 10⁶ cells per 6 cm diameter dish.

Quantitative RT-PCR

Total RNA, extracted using NucleoSpin RNA II (Macherey-Nagel), was reverse transcribed using SuperScript IV (Invitrogen). Real-time quantitative PCRs were performed as previously described (Simeonova et al., 2012), on an ABI PRISM 7500 using Power SYBR Green (Applied Biosystems). Primer sequences are reported in Table S6.

Western Blots

Cells were lysed in RIPA buffer (50 mM Tris-HCl pH 8, 150 mM NaCl, 5 mM EDTA, 0.5% deoxycholic acid, 0.1% SDS, 1% NP-40) supplemented with a Protease inhibitors cocktail (Roche) and 1 mM PMSF (Sigma). Whole-cell extracts were sonicated three times for 10 s and centrifuged at 13 000 rpm for 30 min to remove cell debris. Protein lysate concentration was determined by BCA assay (Thermo Scientific) and 30 μg of each lysate was fractionated by SDS–PAGE on a 4-12% polyacrylamide gel and transferred onto PDVF membranes (Amersham). Membranes were incubated with antibodies raised against Mdm2 (MDM2-4B2, Abcam, 1/500), p53 (CM-5, Novocastra, 1/2000), p21 (F5, Santa Cruz Biotechnology, 1/200) and actin (actin-HRP sc47778, Santa Cruz Biotechnology, 1/5000). Chemiluminescence revelation of western blots was achieved with the SuperSignal West Dura (Perbio).

ChIP Assay

ChIP analysis was performed as previously described (Simeonova et al., 2013). p53-DNA complexes were immunoprecipitated from total extracts by using 50 μg of a polyclonal antibody against p53 (FL-393, Santa Cruz Biotechnology) and 300 μg of sonicated chromatin. Rabbit IgG (Abcam) was used for control precipitation. Quantitative PCR was performed on ABI PRISM 7500. Primer sequences are reported in Table S6.

Cell Fractionation Assay

Cells were lysed in hypotonic lysis buffer (10 mM Tris pH 7.5, 10 mM NaCl, 3 mM Mgcl2, 0.3% NP-40, 10% glycerol) on ice for 10 min, then centrifuged at 4°C for 3 min at 1000g. The supernatant was recovered as the cytoplasmic fraction, and the pellet was incubated in a modified Wuarin-Schibler buffer (10 mM Tris-HCl pH 7.0, 4 mM EDTA, 0.3 M NaCl, 1 M urea, 1% NP-40) on ice for 5 min, vortexed, incubated on ice for 10 min, then centrifuged at 4°C for 3 min at 1000g. The supernatant was recovered as the nucleoplasmic fraction, and the pellet was washed, then incubated in a nuclear lysis buffer (20 mM Tris pH 7.5, 150 mM KCl, 3 mM MgCl2, 0.3% NP-40, 10% glycerol) on ice for 5 min, sonicated, then centrifuged at 13000 rpm at 4°C and the supernatant was recovered as the chromatin fraction. All buffers were supplemented with a Protease inhibitors cocktail (Roche). Cellular fractions were analyzed by western blots with antibodies against p53 (AF-1355, R&D systems, 1/600), Tubulin (ab15568, Abcam, 1/1000), Nup98 (ab50610, Abcam, 1/1000) and histone H3 (ab1791, Abcam, 1/1000). Chemiluminescence revelation was achieved with the SuperSignal West Pico or Femto (Thermoscientific) and analyzed with a ChemiDoc imaging system (Bio-Rad).

Immunofluorescence

MEFs were cultured on collagen-coated coverslips, treated with 10 μM Nutlin 3a (Sigma-Aldrich) for 24 hr and analyzed. Coverslips were stained with rabbit anti-p53 FL-393 (Santa Cruz Biotechnology), mouse anti-actin primary antibody and with Alexa Fluor 647 anti-Rabbit and Alexa Fluor 488 anti-Mouse secondary antibodies (Molecular Probes). DNA was counterstained with DAPI. Images were captured on an epifluorescence microscope using equal exposure times for all images for each fluor.

Cell-Cycle Assay

Log phase cells were irradiated at room temperature (RT) with a Cs γ-irradiator at doses of 3 or 12 Gy, incubated for 24 hr, then pulse labeled for 1 hr with BrdU (10 μM), fixed in 70% ethanol, double stained with FITC anti-BrdU and propidium iodide, and analyzed by flow cytometry with a BD Biosciences FACSort and the FlowJo software.

Apoptosis Assay

Six to eight weeks-old mice were left untreated or submitted to 10 Gy whole-body ɣ-irradiation. Mice were sacrificed 4 hr later and thymi were extracted. Thymocytes were recovered by filtration through a 70 μm smash, stained with Annexin V-FITC Apoptosis detection kit (Abcam) and propidium iodide, then analyzed with a LSRII FACS machine. Data were analyzed using Flowjo software.

Histology

Organs were fixed in formol 4% for 24 hr, then ethanol 70%, and embedded in paraffin wax. Serial sections were stained as described (Simeonova et al., 2013), with hematoxylin and eosin using standard procedures.

RNA-Seq analysis

Total RNA was extracted from the thymi of 8 weeks-old asymptomatic male mice using nucleospin RNA II (Macherey-Nagel). The quality of RNA was checked with Bioanalyzer Agilent 2100 and RNAs with a RIN (RNA integrity number) ≥ 7 were retained for further analysis. RNA was depleted from ribosomal RNA, then converted into cDNA libraries using a TruSeq Stranded Total Library preparation kit (Illumina). Paired-end sequencing was performed on an Illumina MiSeq platform. Reads were mapped to the mouse genome version GRCm38 and counted on gene annotation gencode.vM18 with featureCounts (Liao et al., 2014). Differentially expressed genes with an adjusted p-value < 0.05 were identified using the DESeq2 R package (Love et al., 2014).

Gene Ontology analysis

Gene ontology analysis of differentially expressed genes was performed by using the GOrilla (Technion) software as previously described (Rakotopare et al., 2023). Enrichment analyses were carried out by comparing the list of 192 differentially expressed genes between p53^Y217C/Y217C and p53^-/- thymi to the full list of genes (background), with ontology searches for biological processes and default P-value settings (10⁻³).

Gene set enrichment analysis

Gene set enrichment analysis was performed by using the GSEA software with canonical pathway gene sets from the Mouse Molecular Signature Database (MSigDB) (Subramanian et al., 2005). Gene sets that displayed a normalized enrichment score (NES) > 2, a nominal (NOM) P-value <0.01, and a false discovery rate (FDR) q-value < 0.01 were regarded as significantly enriched.

Statistical analyses

The Student’s t test was used in all figures to analyze differences between two groups of values. For the proportions of mice at weaning, the observed mice count was compared to the expected count according to Mendel’s distribution and a Chi-square (χ²) test, and a Fischer’s test was applied to compare frequencies of mutant females. Survival curves were analyzed with log-rank Mantel-Cox tests. Analyses were performed by using GraphPad Prism, and values of P ≤ 0.05 were considered significant.

Data availability

RNA sequencing data have been deposited in the Gene Expression Omnibus (GEO) under the accession code GSE248936.

Acknowledgements

This project was supported by grants attributed to F.T., from the Ligue Nationale Contre le Cancer, the Fondation ARC pour la recherche sur le Cancer and the Gefluc. S.J., E.E. and J.R. were PhD fellows of the Ministère de la Recherche, and S.J. and E.E. received additional support from the Fondation ARC; M.G. was paid by European Research Council 875532-Prostator-ERC-2019-PoC attributed to A.M. We thank V. Volochtchouk and M. Licaj for technical help, A. Fajac for comments on the manuscript, and the following members of the Institut Curie platforms: I. Grandjean, H. Gautier, C. Daviaud, M. Garcia, M. Verlhac, A. Fosse and D. Andreau (animal facility); S. Baulande and S. Lameiras (NGS); M. Huerre, A. Nicolas and R. Leclere (histopathology); Z. Maciorowski, A. Viguier and S. Grondin (flow cytometry).

Additional information

Author Contributions

Conceptualization: F.T.; investigation: S.J., E.E., J.R., V.L., I.S, B.B., F.T.; formal analysis: M.G., B.B., F.T., A.M.; supervision, writing and visualization: F.T., B.B; project administration and funding acquisition: F.T.

Declaration of interests

The authors declare no competing financial interests.

Supplemental figures and tables

Figure S1.

Figure S2.

Figure S3.

Figure S4.

Figure S5.

Figure S6.

Figure S7.

Figure S8.

Figure S9.

Figure S10.

Table S1.

Table S2.

Table S3.

Table S4.

Table 55.

Table S6.