Evidence for a Tumor-Suppressive Role of SHP-1 in EMT Regulation in Bladder Cancer Cells

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

1. The manuscript presents that PTPN6/SHP1 is a tumor suppressor that attenuates AKT signaling and EMT in bladder cancer, and the abstract extends this further to a prognostic biomarker and potential therapeutic target. However, the conclusions make clear that most of the work is in vitro, that the tissue analysis is exploratory and limited in sample size. The cohort is also only male subjects. The claims should therefore be softened. 
2. Based on comment 1 above, the title is too strong for the current dataset and should be toned down or revised to better reflect that this is largely in vitro work with limited exploratory tissue analysis work.
3. Rationale - It is not fully clear to me why SHP1 was selected here, given that it has already been shown to act as a tumor suppressor and to inhibit EMT in several other cancers. What is the specific rationale for studying SHP1 in bladder cancer beyond contextual extension into another tumor type? The introduction mentions prior TCGA-based associations and one prior in vitro study in bladder cancer, but the abstract and opening framing do not clearly articulate what the real novelty is here. This weakens the argument for novelty and makes the study feel more of a confirmation of what is already known.
4. Mechanism issues - the mechanistic support presented here consists mainly of correlation with EMT markers, pathway level RNA-seq/GSEA, and pakt/akt measurements in a limited subset of lines. That is suggestive, but it is still largely correlative. On that basis, the manuscript should avoid strong causal language implying that SHP1 definitively suppresses bladder cancer progression through AKT attenuation and EMT regulation. The current framing should be revised to make clear that these are associations consistent with the proposed model rather than proof of mechanism.
5. The tissue arm is potentially useful, but it is still limited. 40 patients, yet the reported IHC is in 26 frozen bladder samples. These data is supportive, but they should not be used to make strong claims about clinical aggressiveness or biomarker potential at this stage.
6. In the abstract, SHP1 is described as functioning as an oncogene in glioblastoma and colorectal cancer, whereas in the introduction colorectal cancer is listed among malignancies in which SHP1 serves as a tumor suppressor. This inconsistency should be addressed. If the role of SHP1 is context dependent across tumor types, then the authors should explicitly say so and briefly speculate why its function may differ depending on context.
7. RNA-seq data should be in a repository rather than being available only on request.
8. All figures and results should report the exact n values and p values if possible, and the number of biological replicate. 
9. Formatting issues and concerns -Several figures could be reorganized better. For example, Figure 4 mainly validates SHP1 knockdown/overexpression and could likely be merged with Figure 5 as a multi-panel figure. Similarly, Figures 1 and 2 could probably be combined into a single figure with a unified theme around SHP1 expression in cancer lines & tissues. Each figure should ideally have a clear message or theme rather than separated unnecessarily
10. All figure fonts should be large enough to read and standardized across panels, ideally at a minimum readable font size throughout. Please check your graphical abstract and Figure 8. Those need some formatting.

Author Response

Please see attatchment

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

This manuscript addresses a relevant biological question and presents a coherent mechanistic narrative linking reduced SHP-1 expression with EMT, invasive behavior, and AKT pathway activation in bladder cancer. The combination of cell line screening, functional perturbation experiments, tissue-level validation, and RNA-seq is a strength. However, several methodological and reporting issues need to be resolved before the conclusions can be considered robust, especially regarding biological sample definition, cohort structure, ethics, and statistical reporting.

1. The manuscript uses multiple types of biological material, but the distinction between them is not explained sharply enough. The study includes 19 bladder cancer cell lines for expression screening, four cell lines for SHP-1 knockdown/rescue experiments, flash-frozen human bladder tissues for IHC, and corresponding FFPE resection material for grade/stage assignment. At present, the reader has to infer how these different sample types relate to one another. The Methods should explicitly state that SHP-1 protein expression was assessed on frozen tissue sections, whereas tumor grade and stage were derived from the matched FFPE specimen pathology reports. That distinction is methodologically important and should not be left implicit.
2. The tissue section of the Methods states that tissues were collected from 40 male patients undergoing cystectomy, but the Results only report IHC findings in 26 frozen bladder tissue samples that met inclusion and exclusion criteria. The manuscript does not explain what happened to the remaining 14 specimens. This needs to be clarified explicitly.
3. The manuscript refers to “urothelial tissue with no residual disease” as NRD and uses it as a comparator group. However, this is not equivalent to normal bladder tissue from cancer-free individuals. These samples come from patients undergoing cystectomy and may still reflect disease-associated background, field effects, prior treatment, or smoking-related biology. It could be mentioned as limitations of the study.
4. The tissue analysis is based on three relatively small groups: NRD n=10, NMI n=8, and MI n=8. This is acceptable for an exploratory study, but the sample size limits the strength of inference and should be reflected more carefully in the wording of the conclusions. This limitation should be integrated more clearly into the interpretation of the tissue findings earlier in the manuscript, not only at the end.
5. The manuscript states that the study was conducted under an IRB-approved protocol and later specifies Lahey Clinic protocol 2002-080, approved on 11/22/2002, with patient consent waived under Exempt Research Category 4. That is an important strength, because ethical oversight is clearly stated. However, the Methods also say that tissues were “collected at the time of surgery,” which sounds like prospective perioperative collection rather than analysis of pre-existing exempt archival material. The authors should clarify whether these were prospectively collected research specimens under a long-standing IRB protocol, or whether they were de-identified banked samples analyzed retrospectively. As written, the ethics description is formally present but conceptually ambiguous.
6. The statistical methods section is too general for the range of analyses performed. The authors state that Welch’s t-test or Kruskal–Wallis with Bonferroni correction were used for continuous variables, and Fisher’s exact test for categorical variables, but they do not specify which test was applied to which figure or dataset. There is also no explanation of how normality was assessed, how variance assumptions were handled, or whether multiple-testing correction was applied beyond the Kruskal–Wallis setting. For a study with several small groups and multiple experimental systems, this level of reporting is insufficient. The statistical plan should be expanded substantially.
7. The manuscript states that proliferation assays were performed in duplicate, that migration and invasion assays counted three image fields per well and three wells per experiment, and that each set of conditions was tested with a minimum of three independent trials. Western blot samples were also tested in duplicate. However, it is unclear throughout the Results whether the reported n values and p values are based on biological replicates, technical replicates, or pooled measurements. This needs to be stated explicitly for every figure. Otherwise, the statistical interpretation of the functional assays is difficult to evaluate.
8. The RNA-seq section provides technical sequencing details and reports that 16 samples were analyzed, representing the eight transduced cell line phenotypes tested in duplicate. GSEA was then performed using hallmark gene sets, and significant pathways were reported using FDR < 0.1. However, the manuscript does not describe the differential expression framework in enough detail, nor does it explain whether the duplicates were biological or technical replicates. The absence of a clearly described statistical model for transcriptomic comparison weakens confidence in the pathway-level interpretation. At minimum, the authors should clarify replicate structure, preprocessing choices, and the exact analytical contrast used for GSEA.
9. For the functional and correlation figures, the figure legends should specify the number of biological replicates and define what the error bars represent in each panel. Some legends report SEM, but the underlying sample structure remains unclear.

Author Response

Please see attatchment

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

1. SHP-1 is a protein tyrosine phosphatase. AKT (AKT1) is a serine/threonine kinase. The manuscript does not explain and discuss (in the Intro or Discussion) how SHP-1 directly regulates AKT phosphorylation (occurs at Ser473 and Thr308).
2. In Fig 8, the pAKT/AKT rescue experiment is presented for RT-112 (knockdown) and TCCSUP (rescue) only, but the knockdown and rescue were conducted in four lines total. Clarify.
3. Figure 4 shows approximately 50% residual SHP-1 expression in CUBIII shSHP-1 and RT-112 shSHP-1 relative to shNC. This is a partial knockdown. Did the authors attempt a complete knockdown? Authors need to acknowledge that incomplete silencing may be attenuating effect sizes, etc.
4. Most studies in this area use FFPE tissue, which is better suited for antibody-based detection and better preserved. The authors used frozen sections, however, and they applied a single DAB optical density threshold (0.1) to define positivity (line 206). Clarify.
5. The RNA-seq analysis covers 16 samples representing only 4 transduced cell lines tested in duplicate. Running GSEA on this limited transcriptomic dataset introduces substantial variability, and an FDR cutoff of q<0.1 is lenient. The authors should either justify the FDR threshold or consider reporting only those gene sets meeting q<0.05. 
6. Why is there no mention of the STR profiling or mycoplasma testing for any of the 19 bladder cancer cell lines used? These are now standard requirements for most high-impact journals. 
7. The first citation "World Cancer Day 2026" should be replaced as it appears to be a press or policy release. Authors can refer to Cancer Statistics 2026 by Siegel et al., and extract data from this most updated reference.
8. The authors described lentiviral overexpression of SHP-1 as a "knock-in" at line 289. In molecular biology, "knock-in" refers to targeted allele replacement, typically through homologous recombination. The correct term here is "stable overexpression" or "SHP-1 rescue by lentiviral transduction."
9. Gene and protein nomenclature: The gene symbol PTPN6 should be italicized throughout per HGNC convention. The manuscript inconsistently uses "AKT" and "Akt"; for example, "PI3K/Akt" at line 327 vs. "pAKT/AKT" at line 38 and line 327-329.
10. Line 35 "effected proliferation, invasion, and migration" should be "affected."
11. Line 211: "Bonferonni" should be "Bonferroni."
12. Line 162: "40ug/ml" should be written as "40 μg/mL" with the correct SI unit and spacing.
13. Mean vs. median notation error (lines 257–261): The symbol x̄ denotes the arithmetic mean, not the median. The text uses x̄=0.81, x̄=0.72, and x̄=0 in the context of median percentages. This should be corrected to "median" with the appropriate symbol (Md or the actual median values stated explicitly).
14. "5% CO2" should use subscript notation: CO₂.
15. Figure 2B: The H&E images for NRD and Ta appear to show predominantly adipose or loose connective tissue rather than representative urothelium or papillary tumor.
16. Figures 6B and 6D: Labeling of the scale bars are very small and difficult to read.

Author Response

Please see the attatchment

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

I appreciate the effort the authors have made to improve the manuscript, particularly the revision of the title and the clearer articulation of the contextual novelty of the study. The revised framing better distinguishes this work from prior studies by emphasizing the possible association of SHP1 loss with progression to muscle-invasive disease/bladder cancer and with Akt signaling. The remaining issues, including figure size and overall presentation, are primarily editorial in nature.

Reviewer 3 Report

Comments and Suggestions for Authors

Authors addressed all my comments. Thank you.