The TP53 gene is mutated or inactivated in approximately 50% of all human cancers, making it the most frequently altered tumor suppressor across cancer types (Olivier et al., 2010). Yet mutations alone do not fully explain why some tumors resist chemotherapy and radiotherapy while others respond. Understanding the role of p53 in chemotherapy resistance requires looking beyond the genetic code: a growing body of evidence points to the dynamic behavior of the p53 protein itself — how its concentration changes over time, how long it remains bound to the DNA, and how these patterns determine whether a cancer cell lives or dies.
A new research project at Vita-Salute San Raffaele University, funded by the Fondo Italiano per la Scienza (FIS) grant of the Italian Ministry of University and Research, is now investigating this question at the single-molecule level. Led by Davide Mazza, Associate Professor at Vita-Salute San Raffaele University and Deputy Director of the Experimental Imaging Center at San Raffaele Hospital, the project aims to decode how p53 dynamics influence gene expression and, ultimately, cancer cell fate in response to therapy.
«This question is particularly relevant if we want to understand and solve cancer resistance to conventional therapies, such as chemotherapy or radiotherapy. While it is known that gene mutations can partially explain such drug resistance of tumor cells, there are additional non-genetic factors that can be involved. p53 dynamics might be one of those factors», explains Professor Mazza.
The p53 protein, often called the guardian of the genome, is a transcription factor that controls the expression of genes promoting either cell survival or cell death in response to DNA damage. For decades, research has focused primarily on how TP53 mutations disable this protective function: somatic TP53 mutations occur at rates ranging from 38–50% in ovarian, colorectal, and lung cancers to about 5% in primary leukemia and melanoma (Olivier et al., 2010; Chen et al., 2022). But even in tumors where p53 retains its wild-type form, cancer cells can resist therapy, suggesting that the protein’s dynamic behavior ( François et al., 2026), not just its genetic sequence, plays a decisive role in chemotherapy resistance.
Professor Mazza’s laboratory at the Experimental Imaging Center has pioneered the use of custom-built microscopes that can track individual fluorescently labeled p53 molecules as they move inside living cells with nanometric resolution. This single-molecule approach allows researchers to observe not just how much p53 is present, but how it interacts with DNA and target genes in real time.
Earlier work from the group characterized, in quantitative biophysical terms, how p53 navigates cell nuclear architecture and how different protein domains, often mutated in cancer, contribute to target recognition and oncosuppressive function (Loffreda et al., 2017; Mazzocca et al., 2023).
The FIS-funded project builds upon preliminary experiments that revealed a surprising pattern in how p53 controls gene expression during chemotherapy-induced DNA damage.
In a breast cancer cell line treated with chemotherapy, Professor Mazza and his team observed that p53 levels increase gradually after DNA damage, yet target genes remain completely silent for hours. Then, seemingly abruptly, their transcription switches ON — and stays ON. The timing of this switch varies from cell to cell and does not correlate perfectly with the instantaneous concentration of p53.
«The idea behind this preliminary experiment was to investigate how p53 dynamics correlate with the expression of downstream genes regulating tumor cell survival. The findings were unexpected: p53 seemed to behave as a digital switch-that is, protein levels increased for hours before any downstream change in the expression of the CDKN1A transcript could be detected. However, once CDKN1A was on, it kept being on, independently of p53 levels, and this is correlated with a positive response of cancer cells to chemotherapy», explains Prof. Mazza.
A digital switch, in this context, means that p53 does not gradually increase the activity of its target genes in proportion to its own concentration. Instead, it acts more like a binary mechanism: gene expression remains off until a critical condition is met, then flips decisively to on. This behavior was specifically observed with CDKN1A — a gene encoding the p21 protein, which blocks cell division by inhibiting cyclin-dependent kinases. When CDKN1A is activated, the cell enters cell cycle arrest: it stops dividing, giving the DNA repair machinery time to fix the damage caused by chemotherapy. This mechanism is a key step in preventing damaged cancer cells from continuing to proliferate.
In different experimental conditions, p53 activation produced only a transient increase in target gene transcription: an ON/then OFF behavior that ultimately allowed cancer cells to repair therapy-induced DNA damage, promoting cell survival rather than death. This difference between sustained and transient gene activation may be one of the non-genetic mechanisms behind p53-driven chemotherapy resistance.
Fixed-cell image obtained using the smFISH (single-molecule Fluorescence In Situ Hybridization) technique, which allows counting how many molecules of a specific RNA/gene are expressed at that moment in each individual cell (each dot = 1 RNA molecule).
Two conceptual models could explain how p53 determines whether a target gene switches permanently or only transiently — a distinction that may be central to understanding cancer drug resistance mechanisms. Discriminating between these two scenarios is essential to clarify the role of p53 in chemotherapy resistance at the mechanistic level.
In a threshold model, a gene activates once p53 crosses a fixed concentration level. This is the simpler hypothesis: the cell sets a bar, and once p53 protein levels exceed it, transcription begins.
In a memory model, the gene integrates p53 exposure over time, effectively remembering how much signal it has accumulated. In other words, the target gene has a form of molecular memory: it tracks how much p53 it has “seen” early on, and this cumulative history, rather than the instantaneous concentration, determines whether transcription is triggered.
«We aim to test both models, dissect the molecular mechanisms underlying them. and eventually correlate p53 dynamics and target gene transcription with pro-survival or pro-death cell fate, first in cancer cell lines and second in 3D organoids of metastatic colorectal cancer derived from patients. The latter will be developed thanks to our collaboration with the group led by Professor Giovanni Tonon, director of the Center for Omics Sciences and group leader of the Functional Genomics of Cancer lab at San Raffaele», says Prof. Mazza.
Imaging three-dimensional structures like patient-derived cancer organoids — miniature, lab-grown replicas of a patient’s tumor that preserve the original tissue architecture and genetic features — requires new instrumentation that will be developed within the FIS-funded project.
«We are working to build an advanced oblique plane microscope capable of fast, low-phototoxicity volumetric imaging. In the end, we will use this to image live dynamics of p53 in patients derived cancer organoids».
The live imaging data will then be integrated with high-multiplex single-molecule FISH (fluorescence in situ hybridization); a technique that uses fluorescent probes to light up specific RNA molecules, allowing researchers to visualize the expression of dozens of target genes simultaneously in fixed tissue samples such as cancer biopsies.
This integration of live-cell microscopy with gene expression snapshots will also serve to train a machine learning model. Once trained, the artificial intelligence system should be capable of predicting how the transcription of several target genes, as captured with FISH, reflects underlying p53 dynamics and how this translates into pro-survival or pro-death cancer cell fate upon chemotherapy or radiotherapy. In practice, this means that patterns of p53 gene expression regulation observed in the lab could become predictive biomarkers of chemotherapy resistance in clinical settings.
«If successful, this approach could help predict how a tumor will respond to chemotherapy or radiotherapy using only fixed biopsy material. This strategy, which results from the powerful application of advanced microscopy to basic research in biology, hopefully will inform future therapeutic choices in the clinical setting», concludes Professor Mazza.
The ability to predict therapy response from a standard biopsy, without the need for live-cell imaging in a clinical setting, could represent a significant step forward. By revealing how p53 dynamics shape chemotherapy resistance at the molecular level, this research opens the door to more precise and personalized cancer treatment strategies.