TP53 Splice Mutations Have Tumour-Independent Effects on Genomic Stability and Prognosis: An In Silico Study

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Thank you for the opportunity to review the manuscript titled "TP53 Splice Mutations

Have Tumour-Independent Effects on Genomic Stability and Prognosis: An In-Silico

Study". After careful evaluation, I believe that there is a fundamental flaw in the

manuscript. The core proposition of the study lacks necessary functional experimental

verification, resulting in a lack of substantial support for its biological conclusions.

This renders the conclusions invalid. Therefore, my opinion is to reject the manuscript.

Author Response

Comment 1: Thank you for the opportunity to review the manuscript titled "TP53 Splice Mutations Have Tumour-Independent Effects on Genomic Stability and Prognosis: An In-Silico Study". After careful evaluation, I believe that there is a fundamental flaw in the manuscript. The core proposition of the study lacks necessary functional experimental verification, resulting in a lack of substantial support for its biological conclusions. This renders the conclusions invalid. Therefore, my opinion is to reject the manuscript.

Response 1: We thank the reviewer for their critical assessment. While we acknowledge the value of functional validation, we respectfully disagree that our conclusions are invalid. Our study presents a large-scale computational integration of >25,000 germline and somatic TP53 splice mutations, incorporating transcriptomic, proteomic, and genomic instability metrics and correlating these with clinical outcomes. This anchors our findings in biologically and clinically relevant human specimens rather than purely computational predictions.

Although functional experiments would strengthen causal inference, our primary aim was to systematically characterize the molecular and clinical associations of TP53 splice-site mutations. Taken together, these findings demonstrate that distinct TP53 splice mutations drive unique molecular programs that influence genomic instability and differentially shape patient outcomes, underscoring the biological and clinical heterogeneity conferred by TP53 splicing events.

We also acknowledge in the Discussion the limitations of short-read sequencing and the need for future work using long-read RNA sequencing, isoform-specific quantification, and isogenic models to establish causality. Nonetheless, our study provides meaningful, hypothesis-generating insights from human-derived germline and tumour specimens, identifies clinically relevant TP53 splice variants, and lays the groundwork for subsequent mechanistic and therapeutic investigations.

Reviewer 2 Report

Comments and Suggestions for Authors

The article presents a thorough examination of TP53 splice mutations, elucidating distinct molecular pathways and unforeseen clinical implications. To improve the manuscript, the author should explore the mechanisms underlying significant discrepancies and refine the proposed biological pathways.

1. The study observed a "striking discordance" between mRNA and protein expression levels at splice sites X32 and X33. Notably, X33 was linked to copy number deletion and increased mRNA levels, while protein expression was slightly diminished. What specific post-transcriptional regulatory mechanisms (such as altered protein turnover, translation efficiency, or targeted degradation) could account for this site-specific variability in protein abundance, despite the observed changes in copy number and transcript levels?
2. The X126 mutation demonstrates a "gain-of-dosage effect" (characterized by increased mRNA and protein levels) and exhibits lower genomic instability (as indicated by tumor mutational burden and fractional genome alteration) compared to other variants. However, it is paradoxically associated with a significantly worse clinical outcome. The authors propose that this association is driven by "isoform modulation" rather than genomic instability. What specific evidence or detailed hypotheses can be provided to connect the increased abundance of X126-derived transcripts/proteins to a particular non-genomic pathway (such as selective suppression of apoptosis or senescence) that contributes to the poor prognosis?
3. The study indicates that donor (e.g., X125, X331) and acceptor (e.g., X126, X332) mutations within the same junction yield distinct molecular and clinical outcomes. Given that X125 and X331 are associated with "loss-of-function" and high genomic instability, while their counterparts X126 and X332 exhibit lower instability, can the revision provide a more comprehensive explanation of how the positional differences (donor versus acceptor) consistently influence the final isoform composition, resulting in such contrasting phenotypes regarding genomic stability?
4. Transcriptomic profiling categorized tumors into two clusters, with Cluster 2 exhibiting a heterogeneous pattern of mixed activation and repression of p53 target genes, alongside TP53 mRNA levels comparable to or exceeding those of wild-type tumors. Which specific p53 isoforms, known to function as dominant-negative inhibitors (e.g., Delta 40p53) or to suppress canonical functions (e.g., Delta 133p53), are potentially produced or dysregulated by the splice mutations prevalent in Cluster 2, and how does this isoform balance elucidate the observed mixed transcriptional phenotype?
5. Mutations X224 and X225 displayed intermediate fractional genome alteration values and significant pathway dysregulation (such as downregulation of steroid metabolic and cholesterol homeostasis pathways). Despite these molecular alterations, they were not significantly associated with recurrence, progression, or mortality. The authors suggest that "additional contextual factors, including tumor type or microenvironment," may influence these effects. What evidence or analyses can substantiate the impact of these contextual factors specifically for X224 and X225, or how might the authors validate this hypothesis in their revision?
6. The study identifies a notable difference in nucleotide substitution patterns at donor sites: transversions are more prevalent than transitions in somatic data, whereas transitions are more common than transversions in germline data (p < 0.001). What biological or environmental mechanisms (such as differential repair pathway activity or specific known mutational signatures in germline versus somatic contexts) are proposed to account for this significant positional and nucleotide-specific bias?
7. The article emphasizes that reliance on standard exon-focused sequencing and immunohistochemistry may "underestimate the prevalence and impact of splice mutations." Based on the analysis of a large cohort exceeding 25,000 TP53 variants, can the authors provide a quantifiable estimate of the percentage of clinically relevant TP53 splice mutations (defined as those associated with poor prognosis, such as X126, X125, and X331) that would be overlooked if sequencing were limited to exonic regions?
8. In light of the discussion's recommendation for future studies to utilize long-read RNA sequencing and functional assays, what specific functional experiments or molecular assays are necessary to mechanistically delineate the isoform-specific effects of key sites, such as X126, which is implicated in poor prognosis through isoform modulation rather than genomic instability? For instance, should the functional assays specifically assess the ratio of full-length p53 to Delta 133p53 or Delta 40p53, and evaluate the dominant-negative capacity of the splice variant in vitro?

Author Response

We thank the reviewer for their thoughtful evaluation and constructive comments. A point-by-point response addressing each question is provided below.

Comment 1: The study observed a "striking discordance" between mRNA and protein expression levels at splice sites X32 and X33… What specific post-transcriptional regulatory mechanisms… could account for this site-specific variability in protein abundance, despite the observed changes in copy number and transcript levels?

Response 1: In the revised manuscript, we have now explicitly addressed this point by outlining plausible post-transcriptional mechanisms that could explain this site-specific variability (Section 2.2, Line 180 -186).

 

Comment 2: The X126 mutation demonstrates a "gain-of-dosage effect" (characterized by increased mRNA and protein levels) and exhibits lower genomic instability (as indicated by tumour mutational burden and fractional genome alteration) compared to other variants. However, it is paradoxically associated with a significantly worse clinical outcome. The authors propose that this association is driven by "isoform modulation" rather than genomic instability. What specific evidence or detailed hypotheses can be provided to connect the increased abundance of X126-derived transcripts/proteins to a particular non-genomic pathway (such as selective suppression of apoptosis or senescence) that contributes to the poor prognosis?

Response 2: We thank the reviewer for this insightful comment. To investigate the molecular basis of the paradoxical clinical behaviour of X126 tumours, we performed Spearman’s correlation analysis between TP53 mRNA levels and genome-wide gene expression using our RNA-sequencing data. This analysis revealed that X126 TP53 mRNA associates with genes involved in RNA processing, transcriptional regulation, chromatin organization, cell-cycle and mitotic processes, and genome maintenance, but not with canonical apoptotic or cell-cycle pathways. These findings suggest that the poor prognosis of X126 tumours reflects altered TP53 isoform–driven regulatory programs rather than genomic instability. We have incorporated this data into Results Section 2.4 for all donor–acceptor pairs, revised the section (Lines 235-284) to include this additional information, provided Figure 5 illustrating the overlap between donor–acceptor pairs and canonical TP53 pathways, and supplied the pathway details as an Excel file. The Discussion has also been updated to reflect these changes (Lines 332-340).

 

Comment 3: The study indicates that donor (e.g., X125, X331) and acceptor (e.g., X126, X332) mutations within the same junction yield distinct molecular and clinical outcomes. Given that X125 and X331 are associated with "loss-of-function" and high genomic instability, while their counterparts X126 and X332 exhibit lower instability, can the revision provide a more comprehensive explanation of how the positional differences (donor versus acceptor) consistently influence the final isoform composition, resulting in such contrasting phenotypes regarding genomic stability?

Response 3: We have expanded the Discussion (Lines 313-318) to provide a more detailed explanation of how positional differences at donor versus acceptor splice sites influence isoform composition.

 

Comment 4: Transcriptomic profiling categorized tumours into two clusters, with Cluster 2 exhibiting heterogeneous pattern of mixed activation and repression of p53 target genes, alongside TP53 mRNA levels comparable to or exceeding those of wild-type tumours. Which specific p53 isoforms, known to function as dominant-negative inhibitors (e.g., Delta 40p53) or to suppress canonical functions (e.g., Delta 133p53), are potentially produced or dysregulated by the splice mutations prevalent in Cluster 2, and how does this isoform balance elucidate the observed mixed transcriptional phenotype?

Response 4: We thank the reviewer for this comment. We are currently unable to determine which specific TP53 isoforms are produced or their relative ratios due to the limitations of short-read sequencing. We have added this as a limitation in the Discussion (Lines 359-361): “A central limitation of our study is the inability of short-read sequencing to determine which TP53 isoforms are produced, or in what ratios, given the complexity of the locus (ref); thus, the precise isoform signatures associated with individual splice-site mutations remain unresolved.”

 

Comment 5: Mutations X224 and X225 displayed intermediate fractional genome alteration values and significant pathway dysregulation (such as downregulation of steroid metabolic and cholesterol homeostasis pathways). Despite these molecular alterations, they were not significantly associated with recurrence, progression, or mortality. The authors suggest that "additional contextual factors, including tumour type or microenvironment," may influence these effects. What evidence or analyses can substantiate the impact of these contextual factors specifically for X224 and X225, or how might the authors validate this hypothesis in their revision?

Response 5: We thank the reviewer for this comment. We have removed the sentences referring to “additional contextual factors, including tumour type or microenvironment” because we cannot accurately assess these influences with the available data.

 

Comment 6: The study identifies a notable difference in nucleotide substitution patterns at donor sites: transversions are more prevalent than transitions in somatic data, whereas transitions are more common than transversions in germline data (p < 0.001). What biological or environmental mechanisms (such as differential repair pathway activity or specific known mutational signatures in germline versus somatic contexts) are proposed to account for this significant positional and nucleotide-specific bias?

Response 6: We thank the reviewer for this insightful question. While the observed difference in nucleotide substitution patterns at donor sites between somatic and germline contexts is intriguing, we are unable to rigorously explore the underlying biological or environmental mechanisms. Specifically, we do not have access to environmental exposure data, which likely differ between germline and somatic contexts and can influence mutational spectra. Additionally, somatic mutations arise under distinct selective pressures within tumorigenic environments compared to germline variants, affecting which substitutions are tolerated. Finally, while differential DNA repair activity (e.g., mismatch repair or nucleotide excision repair) could contribute to the observed bias, we lack functional or epigenomic data in these samples to validate such mechanisms. Collectively, these limitations prevent us from attributing the positional and nucleotide-specific bias to specific mutational processes. We have added these limitations to the Discussion of our Manuscript (Lines 319-323).

 

Comment 7: The article emphasizes that reliance on standard exon-focused sequencing and immunohistochemistry may "underestimate the prevalence and impact of splice mutations." Based on the analysis of a large cohort exceeding 25,000 TP53 variants, can the authors provide a quantifiable estimate of the percentage of clinically relevant TP53 splice mutations (defined as those associated with poor prognosis, such as X126, X125, and X331) that would be overlooked if sequencing were limited to exonic regions?

Response 7: We thank the reviewer for this comment. Based on our analysis of over 25,000 TP53 variants, we estimate that approximately 84% of splice mutations occur beyond the high-confidence 1 bp exon–intron window and are therefore likely to be missed by standard exon-focused sequencing. This includes clinically relevant mutations such as X125, X126, and X331, which are associated with poor prognosis. We have added this to our Discussion section (Lines 347-354).

 

Comment 8: In light of the discussion's recommendation for future studies to utilize long-read RNA sequencing and functional assays, what specific functional experiments or molecular assays are necessary to mechanistically delineate the isoform-specific effects of key sites, such as X126, which is implicated in poor prognosis through isoform modulation rather than genomic instability? For instance, should the functional assays specifically assess the ratio of full-length p53 to Delta 133p53 or Delta 40p53, and evaluate the dominant-negative capacity of the splice variant in vitro?

Response 8: As described in our revised discussion (Lines 361-374), mechanistic evaluation of splice-site mutations will require long-read RNA sequencing and isoform-specific RNA and protein quantification to define the full-length and truncated TP53 isoform repertoire. Isogenic cellular models, combined with functional assays of p53 transcriptional activity, DNA-damage response, cell fate, and genome stability, will help determine whether these effects arise from altered p53 dosage rather than classical loss-of-function. Complementary analyses of chromatin occupancy, protein stability, and post-translational regulation, alongside in vivo validation in patient-derived organoids or xenografts, will be essential to establish the clinical significance of splice-dependent changes in isoform architecture.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Thank you to the author for responding to the previous review comments.

After careful review and revision of the manuscript, I believe that there are still significant shortcomings in the innovation aspect of this study, and therefore maintain the rejection recommendation.

Author Response

Comment 1:  Thank you to the author for responding to the previous review comments. After careful review and revision of the manuscript, I believe that there are still significant shortcomings in the innovation aspect of this study, and therefore maintain the rejection recommendation.

Response 1: We thank the reviewer for the additional assessment of our revised manuscript. We respectfully acknowledge the reviewer’s position and appreciate the time taken to re-evaluate our work. We have no further clarifications to add beyond those already provided in our responses and revisions.