In this study, we experimentally dissected p53 repression by ΔNp63α during the radiation response. TP63 is a member of the TP53 gene family and is homologous to TP53 in its TA, DNA-binding domain, and tetramerization regions. ΔNp63 is one of the proteins generated through selective alternative splicing after TP63 transcription. Therefore, since its discovery, ΔNp63, especially ΔNp63α, has been thought to work competitively with p53 against target genes, thereby inhibiting the typical function of p53. On the other hand, radiation is highly cell permeant and induces DNA oxidative damage, including DSBs, which causes gene mutations and chromosomal aberrations through the generation of ROS, such as hydroxyl radicals and singlet oxygen, in the vicinity of DNA in the nucleus. This in turn leads to the activation of kinases such as ATM to induce the p53-derived DDR, including cell cycle arrest and apoptosis. Therefore, radiation biology provides the environment necessary to determine how ΔNp63α affects DDR in these signalling transductions and whether it is a competitive inhibitor of p53.
We first performed siRNA knockdown experiments using HMECs, which exhibit mammary basal cell characteristics, and observed changes in the expression of DDR-related genes downstream of p53 by RT–qPCR. After the X-irradiation to HMECs, genes related to cell cycle arrest and apoptosis showed little response in the scr-treated group but exhibited a marked increase in expression after ΔNp63α knockdown. This was further confirmed by Western blotting and RNA-seq. Although these DDR-related genes are upregulated through p53 binding to the promoter regions after irradiation, the differential expression of these genes even in the absence of irradiation suggests that ΔNp63α suppresses the expression of these genes at all times.
To investigate whether the transcriptional repression of DDR-related genes is independent of the amount of DNA damage, we directly quantified the DSBs generated in HMECs after irradiation. We found that the amounts of DSBs and ROS generated after irradiation were increased in the sip63-treated group compared to the scr-treated group. However, similar increases were observed in the nonirradiated group, suggesting that the difference between groups was due to the upregulation of antioxidant genes, such as GPX2 and CYBG, by ΔNp63α [36, 37], resulting in a decrease in long-lived ROS, such as H2O2. Hence, ΔNp63α-expressing cells are resistant to less reactive long-lived ROS but not to more reactive short-lived ROS, such as hydroxyl radicals, generated in the vicinity of DNA by radiation.
The RNA-seq analysis of HMECs revealed that ΔNp63α upregulates genes related to the cell cycle and cell division while downregulating genes related to apoptosis and cell death. These findings are consistent with those of previous studies [13, 14] and suggest that ΔNp63α acts as a transcription factor that maintains the stemness/inhibits the differentiation of mammary stem cells, keeps them alive by preventing cell death, and turns on genes that promote proliferation. The properties of ΔNp63α are opposite those of cancer suppressor genes such as TP53 and PTEN, and it is thus thought that ΔNp63α initially inhibits the typical radiation responses triggered upon radiation-induced DNA damage. In addition, RNA-seq analysis suggests that ΔNp63α downregulates TP53 and PTEN. These findings suggest that ΔNp63α suppresses the expression of TP53 and other tumour suppressor genes and counteracts their effects. On the other hand, the protein analysis of HMECs showed that the p53 protein was upregulated by ΔNp63α expression, which was not observed in iPS-DNs. This suggest that ΔNp63α may increase the lifetime of the p53 protein in HMECs. Indeed, since the SAM domain in the C-terminus of ΔNp63α interacts with p300/CBP [9], it is possible that p53 undergoes acetylation, resulting in low expression but a longer lifetime. Taken together, the results indicate that the relationship between ΔNp63α and tumour suppressor genes requires further investigation..
In the three cell types used in this study, HMECs, iPS-DNs, and iPS-KCs, RT–qPCR analysis results showed that ΔNp63α significantly inhibited the expression of BAX. Consistent with this, the detection of apoptotic cells by FCM showed that ΔNp63α suppressed radiation-induced apoptosis, since the proportion of apoptotic cells decreased in ΔNp63α-expressing cells. Furthermore, ChIP–qPCR assays of iPS-DNs confirmed that the binding of p53 to its target gene promoter region was reduced by ΔNp63α expression. These results all suggest that ΔNp63α inhibits radiation-induced apoptosis by suppressing the expression of apoptosis-related genes such as BAX by p53. On the other hand, although ΔNp63α transcriptionally repressed CDKN1A expression and reduced p21 at the protein level, cell cycle analysis by FCM showed that cell cycle arrest occurred after irradiation regardless of ΔNp63α expression. This may be because cell cycle arrest-related genes can be regulated by genes other than p53. Consistent with this model, the analysis of mammary organoids also showed that p21 is expressed at the same level in ΔNp63α-positive cells as in ΔNp63α-negative cells after irradiation. With regard to cell cycle arrest, this study supports the results of Westfall et al.: ΔNp63α transcriptionally represses CDKN1A, but p21 is still expressed in certain amounts after transcriptional repression; moreover, the results of the cell cycle analysis by FCM suggest that the transcriptional repressive effect is limited during cell cycle arrest. At the same time, GADD45A in X-irradiated iPS-DNs was not transcriptionally inhibited at all, despite being a p53 downstream gene. Thus, these results indicate that the transcriptional repression of ΔNp63α preferably affects proapoptotic p53 target genes. Understanding whether this transcriptionally repressive effect is due to competitive inhibition against p53 RE, as conventionally suggested, or to gene expression regulated by ΔNp63α will require further discussion addressing a wider range of p53-related genes.
It is problematic that stem cells contained in basal cells may all have this DDR vulnerability. If epithelial stem cells expressing ΔNp63α are vulnerable to DDR, especially apoptosis, it can be inferred that not only the mammary gland but also the prostate, lung, skin, and other epithelial tissues are at risk. Since caspase-3, which is activated by the p53 pathway, is also inhibited in ΔNp63α-expressing cells, it is conceivable that stem cells, which are long-lived and capable of differentiating into other cells [21, 22], may survive the failure to repair radiation-induced DNA damage, leaving DNA damage and mutations behind, and then differentiate into other cells, including cancerous cells (Fig. 6). BRCA1/2 genes are involved in DNA repair, especially homologous recombination repair, which precisely repairs DSBs. Women with BRCA1 or 2 gene deficiency are more likely to develop breast cancer. Thus, this may suggest an additive or synergistic effect of BRCA gene deficiency and p53 repression by ΔNp63α. Indeed, BRCA1-deficient breast cancers are positive for CK14, which seems to be associated with the development of triple-negative breast cancer [45].
Epithelial stem cells such as mammary stem cells still lack definitive markers, and therefore, single-cell analysis, including single-cell RNA-seq and whole-genome sequencing, will be needed to elucidate DDR inhibition by ΔNp63α in epithelial stem cells and then characterize the mutational signatures and chromosomal aberrations induced by irradiation. The process of long-term mutation accumulation in cancer-initiating cells should be investigated in detail by further analysing the characteristics of DDR inhibition by ΔNp63α in epithelial stem cells.