Review Reports
- Haiping Jiang 1,2,
- Feng Chen 1 and
- Jianqi Xiao 1,*
- et al.
Reviewer 1: Anonymous Reviewer 2: Anonymous
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsThis manuscript investigates the molecular mechanisms underlying the anticancer effects of pectolinarigenin (PECT) in glioblastoma (GBM), focusing on ROS-mediated suppression of E2F1-dependent RRM2 transcription. The authors employ bioinformatic analyses, gene silencing, overexpression experiments, promoter mutagenesis, and dual-luciferase reporter assays to investigate the proposed signaling pathway. Although the findings provide additional mechanistic insights into the anti-GBM activity of PECT, several important concerns regarding novelty, mechanistic validation, and biological relevance should be addressed before the manuscript can be considered for publication.
Major comments
- The authors have previously demonstrated that PECT suppresses GBM progression by inhibiting RRM2 transcription and promoting its autophagy–lysosome-mediated degradation. Furthermore, E2F1 has already been identified as a transcriptional regulator of RRM2, including in GBM. Therefore, the present study primarily extends these established observations by proposing ROS-dependent suppression of E2F1 as an upstream mechanism. Although this represents an incremental mechanistic advancement, the authors should more clearly distinguish the current findings from their previous publication and explain the specific conceptual novelty of the proposed ROS–E2F1–RRM2 regulatory axis.
- The authors use bioinformatic prediction, site-directed mutagenesis, and dual-luciferase reporter assays to demonstrate the functional importance of a GC-rich element within the RRM2 promoter. However, these experiments do not directly establish that endogenous E2F1 occupies this specific promoter region in GBM cells. Chromatin immunoprecipitation followed by quantitative PCR (ChIP-qPCR) would provide more direct evidence of E2F1 binding to the endogenous RRM2 promoter.
- Although E2F1 overexpression partially restores cell viability following PECT treatment, E2F1 regulates numerous genes involved in cell-cycle progression, DNA replication, and apoptosis. Therefore, the observed rescue cannot be attributed specifically to RRM2. The authors should determine whether E2F1 overexpression restores RRM2 expression under PECT treatment. Direct RRM2 overexpression rescue experiments would help establish whether RRM2 downregulation is functionally responsible for the anticancer effects of PECT.
- The manuscript contains several formatting and organizational issues. For example, the figure section is incorrectly numbered as Section 3.2 after Section 3.4. The authors should also check the consistency of figure references, experimental descriptions, and statistical annotations throughout the manuscript.
Author Response
Reviewer 1
General assessment: The reviewer rated the English language as "fine and does not require any improvement" and all quality dimensions as "can be improved", noting: "Although the findings provide additional mechanistic insights into the anti-GBM activity of PECT, several important concerns regarding novelty, mechanistic validation, and biological relevance should be addressed before the manuscript can be considered for publication."
Response: We are grateful for this balanced and constructive assessment and address each of the four major comments below.
Comment 1:
"The authors have previously demonstrated that PECT suppresses GBM progression by inhibiting RRM2 transcription and promoting its autophagy–lysosome-mediated degradation. Furthermore, E2F1 has already been identified as a transcriptional regulator of RRM2, including in GBM. Therefore, the present study primarily extends these established observations by proposing ROS-dependent suppression of E2F1 as an upstream mechanism. Although this represents an incremental mechanistic advancement, the authors should more clearly distinguish the current findings from their previous publication and explain the specific conceptual novelty of the proposed ROS–E2F1–RRM2 regulatory axis."
Response: We agree that the distinction must be made explicit. Our previous publication (ref. 18) established that PECT suppresses GBM—an efficacy validated in vitro and in vivo—and identified RRM2 as its downstream target (transcriptional repression plus autophagy–lysosome-mediated degradation), but it left the initiating signal and the transcriptional executor unknown. The present study answers precisely these two questions, converting a descriptive drug–target relationship into a defined, druggable signaling axis: (i) ROS accumulation is identified as the proximal initiating event (NAC reversal); (ii) E2F1 is identified as the executor, declines before RRM2, and is linked for the first time to GBM prognosis (HR = 1.5, log-rank p = 0.028); and (iii) PECT sensitivity is mapped, at nucleotide resolution, to a single GC-rich E2F1-response element. We have added two clarifying statements to the Introduction (Page 3, Lines 102–105 and 116–119):
"Addressing this question is conceptually distinct from re-establishing the anti-GBM efficacy of PECT, which we have previously validated in vitro and in vivo; the present study instead asks how the PECT signal is initiated and relayed to the RRM2 promoter." (Page 3, Lines 102–105)
"To our knowledge, this is the first definition of a ROS-initiated, E2F1-executed transcriptional checkpoint governing RRM2 expression in GBM, and the first mapping of PECT sensitivity to a single promoter element at nucleotide resolution." (Page 3, Lines 116–119)
The Discussion (Page 11-12, Lines 363–384) contains a dedicated three-point comparison with our previous report, which now reads together with these new Introduction statements.
Comment 2:
"The authors use bioinformatic prediction, site-directed mutagenesis, and dual-luciferase reporter assays to demonstrate the functional importance of a GC-rich element within the RRM2 promoter. However, these experiments do not directly establish that endogenous E2F1 occupies this specific promoter region in GBM cells. Chromatin immunoprecipitation followed by quantitative PCR (ChIP-qPCR) would provide more direct evidence of E2F1 binding to the endogenous RRM2 promoter."
Response: We fully agree that ChIP-qPCR would provide the most direct evidence, and we address this comment without new wet-lab data for the following reasons. First, direct occupancy of the RRM2 promoter by endogenous E2F1 is already established in the literature, including by chromatin immunoprecipitation in GBM cells (ref. 4, Rasmussen et al.; ref. 8, DeGregori et al.); the question our study addresses is not whether E2F1 binds the promoter, but which element functionally transmits PECT’s repressive signal—and this causal question is answered at nucleotide resolution by our mutagenesis data: mutation of the GC-rich core (GCGC→ATAT) abolished both E2F1-driven transactivation and PECT-induced repression while leaving basal promoter activity intact (Fig. 4B, C). We now state this connection explicitly in Section 3.4 (Page 6, Lines 280–282):
"This functional assignment is consistent with prior chromatin immunoprecipitation evidence that E2F1 directly occupies the RRM2 promoter in GBM cells[4]." (Page 6, Lines 280–282)
Second, a ChIP-qPCR assay of adequate quality (antibody validation, optimization, and biological replicates) cannot be completed within the 10-day revision period without compromising rigor, and we prefer not to present rushed data. We have therefore acknowledged this transparently in the Limitations and committed to the experiment in our ongoing follow-up study:
"; ChIP-qPCR verification of endogenous E2F1 occupancy at this element, together with RRM2 re-expression rescue assays, is planned in our ongoing follow-up study." (Page 15, Lines 431–433)
Comment 3:
"Although E2F1 overexpression partially restores cell viability following PECT treatment, E2F1 regulates numerous genes involved in cell-cycle progression, DNA replication, and apoptosis. Therefore, the observed rescue cannot be attributed specifically to RRM2. The authors should determine whether E2F1 overexpression restores RRM2 expression under PECT treatment. Direct RRM2 overexpression rescue experiments would help establish whether RRM2 downregulation is functionally responsible for the anticancer effects of PECT."
Response: We agree with the premise—E2F1 is pleiotropic, and the partial viability rescue cannot be attributed to RRM2 alone—and we have revised the Discussion to state this boundary explicitly:
"Because E2F1 governs a broad transcriptional program spanning cell-cycle progression, DNA replication, and apoptosis, this rescue cannot be ascribed to RRM2 restoration alone; direct RRM2 re-expression under PECT treatment will be required to formally quantify the contribution of RRM2 downregulation to PECT cytotoxicity." (Page 14, Lines 385–389)
At the same time, three independent lines of existing evidence support RRM2 as a major conduit downstream of E2F1: (i) the temporal order—E2F1 mRNA declines as early as 12 h after PECT exposure, preceding significant RRM2 reduction at 24 h (Fig. S2C, D); (ii) the reporter data—PECT represses the wild-type but not the E2F1-site-mutant RRM2 promoter, and E2F1-driven transactivation through this element is attenuated by PECT (Fig. 4B, C); and (iii) our previous in vivo study (ref. 18), in which RRM2 suppression mediated PECT’s anti-GBM efficacy. A formal RRM2 re-expression rescue (construct validation and repetition of the full dosing matrix in two cell lines) cannot be completed to a rigorous standard within the 10-day revision window; it is now listed, together with ChIP-qPCR, in the Limitations as committed follow-up work (Page 15, Lines 431–433).
Comment 4:
"The manuscript contains several formatting and organizational issues. For example, the figure section is incorrectly numbered as Section 3.2 after Section 3.4. The authors should also check the consistency of figure references, experimental descriptions, and statistical annotations throughout the manuscript."
Response: We apologize for these oversights and have performed a complete consistency sweep. Corrections include: (i) the figure section heading renumbered as "3.5 Figures, Tables and Schemes" (Page 6, Line 288); (ii) an erroneous citation in Section 3.2 (our previous work was miscited as ref. 13 in the original submission and is now correctly cited as ref. 18; Page 5, Line 236); (iii) a missing space after "(Fig. 2D)" (Page 6, Line 245); (iv) punctuation in the author line (Page 1, Line 4); (v) typographical issues in the Abstract ("PECT’ s", missing spaces before the Methods/Results/Conclusions headings, and spacing around the response-element sequence; Page 1, Lines 28–42); (vi) a capitalization error in Section 2.3 (Page 3, Lines 139); and (vii) spacing inconsistencies in the Methods ("37 ℃ for 30 min", Page 4, Lines 161; "95 ℃ for 10 min", Page 5, Line 196; "48 h post-transfection", Page 4, Line 183) and in the Figure 3 legend (Page 11, Line 314). All figure callouts, experimental descriptions, and statistical annotations were cross-checked and are consistent. (viii) following the addition of the new references, the reference list has been renumbered in order of first appearance, and all in-text citations have been updated accordingly. All corrections are visible as tracked changes.). All figure callouts, experimental descriptions, and statistical annotations were cross-checked and are consistent. All corrections are visible as tracked changes.
Reviewer 2 Report
Comments and Suggestions for AuthorsDear Authors;
The manuscript investigates the effects of pectolinarigenin (PECT) on the ROS–E2F1–RRM2 signaling pathway in glioblastoma cells. The study is interesting and provides useful findings. However, several issues regarding the experimental design, methodological details, and interpretation of the results need to be addressed. Additional validation of the proposed mechanism would strengthen the study. My specific comments are provided below.
Title
- The title states that PECT inhibits glioblastoma growth; however, the study only evaluates cell viability using a 48-hour MTT assay. No direct proliferation, colony formation, or in vivo tumor growth assays were performed. Therefore, the authors should revise the title to more accurately reflect the experimental findings. As a suggestion, progression rather than growth would be more appropriate.
Introduction
- The regulatory relationship between E2F1 and RRM2 has been established in prior studies (Refs. 4 and 8). The authors should more clearly define the novelty of the present study, particularly how PECT affects this established regulatory axis.
- The claim that E2F1 activity is upregulated in high-grade gliomas and associated with adverse clinical features requires appropriate references.
- The Introduction provides limited background on the relationship between ROS and E2F1. Because oxidative stress and DNA damage can also promote E2F1 stabilization or activation under certain conditions, the authors should discuss these context-dependent effects and provide relevant references to better support their proposed mechanism.
Materials and methods
- Only two PECT concentrations (10 and 20 µM) were tested. The authors should provide a rationale for selecting these concentrations and clarify whether they were based on prior dose-response or IC50
- In Supplementary Table S1, siE2F1#1 and siE2F1#2 share substantial sequence overlap. The authors should clarify how they selected these siRNAs and present their sequences in a consistent format.
- Molecular weight markers are not shown in the Western blot figures. The authors should provide the uncropped, original Western blot images, including molecular weight information, as supplementary material. In addition, some Western bands are not labeled in the current uncropped supplementary material file. For example, which proteins do the bands below the E2F1 bands in Fig.2E correspond to? Additionally, apart from the b-actin bands presented in the publication, each membrane's own β-actin should also be included in the supp file.
- Some figures show approximately five individual data points per group, whereas the legends report n = 3. The authors should clarify the distinction between biological and technical replicates and specify the actual sample size used for statistical analysis.
- The statistical analysis requires further clarification. Experiments involving multiple factors, such as promoter type (WT/Mut) and treatment, would benefit from an appropriate factorial analysis, such as two-way ANOVA, including relevant interaction effects.
Results
- ROS levels were measured only after 48 h of PECT treatment, when cell viability was already reduced. The authors should assess ROS levels at earlier time points and in the PECT + NAC group to confirm the proposed ROS-dependent mechanism.
- In Figure 1C–D, approximately five data points are shown per group, whereas the figure legend reports n = 3. The authors should clarify the number of biological and technical replicates. Additionally, Fig. 1B does not show data points. Data points should be added to all graphs throughout the article.
- The scale bars are not visible in Figure 1A. Additionally, the U87 20 µM panel appears to differ in cell density and image appearance. The authors should confirm that consistent imaging conditions and magnification were used.
- In Figure 3E–H, NAC treatment restores E2F1 expression, but its effect on RRM2 remains unclear. The authors should evaluate RRM2 expression in the Control, PECT, NAC, and PECT + NAC groups to further support the proposed ROS–E2F1–RRM2 signaling pathway. If it is not included, explain why and, if necessary, add it to the limitations.
Discussion
- The Discussion section largely repeats the results and should be shortened to avoid redundancy and to focus more on interpreting and discussing the significance of the findings.
- The therapeutic implications are somewhat overstated, given that the study is limited to two GBM cell lines without normal-cell controls or in vivo validation. The authors should moderate these claims in the Discussion and Conclusions to better reflect the available experimental evidence.
Author Response
R
Reviewer 2
General assessment: The reviewer rated the English language as "fine and does not require any improvement", rated the introduction and conclusions as satisfactory ("yes"), and the remaining dimensions as "can be improved", noting: "The study is interesting and provides useful findings. However, several issues regarding the experimental design, methodological details, and interpretation of the results need to be addressed. Additional validation of the proposed mechanism would strengthen the study."
Response: We thank the reviewer for this constructive evaluation and address each comment below, grouped by section as in the report.
Comment 1 (Title):
"The title states that PECT inhibits glioblastoma growth; however, the study only evaluates cell viability using a 48-hour MTT assay. No direct proliferation, colony formation, or in vivo tumor growth assays were performed. Therefore, the authors should revise the title to more accurately reflect the experimental findings. As a suggestion, progression rather than growth would be more appropriate."
Response: We thank the reviewer for this apt suggestion and have adopted it verbatim. The title now reads:
"Pectolinarigenin suppresses E2F1-dependent RRM2 transcription through ROS accumulation to inhibit glioblastoma progression" (Page 1, Lines 2–3)
We note for completeness that in vivo anti-GBM efficacy of PECT was demonstrated in our previous study (ref. 18), but we agree the title of the present paper should reflect its own experimental scope.
Comment 2 (Introduction):
"The regulatory relationship between E2F1 and RRM2 has been established in prior studies (Refs. 4 and 8). The authors should more clearly define the novelty of the present study, particularly how PECT affects this established regulatory axis."
Response: We fully agree—please see our response to Reviewer 1, Comment 1. Briefly, the novelty is not the E2F1–RRM2 relationship itself but (i) ROS as the initiating signal, (ii) E2F1 as the pharmacologically suppressible executor of the axis, and (iii) the mapping of PECT sensitivity to a single promoter element. Two sentences clarifying this have been added to the Introduction (Page 3, Lines 102–105 and 116–119; quoted under Reviewer 1, Comment 1).
"Addressing this question is conceptually distinct from re-establishing the anti-GBM efficacy of PECT, which we have previously validated in vitro and in vivo; the present study instead asks how the PECT signal is initiated and relayed to the RRM2 promoter." (Page 3, Lines 102–105)
"To our knowledge, this is the first definition of a ROS-initiated, E2F1-executed transcriptional checkpoint governing RRM2 expression in GBM, and the first mapping of PECT sensitivity to a single promoter element at nucleotide resolution." (Page 3, Lines 116–119)
Comment 3 (Introduction):
"The claim that E2F1 activity is upregulated in high-grade gliomas and associated with adverse clinical features requires appropriate references."
Response: We agree. To eliminate redundancy with the preceding statement that E2F1 cooperates with BRCA1 to sustain RRM2 transcription in GBM (ref. 4), the two sentences have been consolidated into a single statement on the clinical relevance of E2F1, now supported by a newly added reference systematically analyzing E2F expression and prognosis in high-grade glioma (ref. 10, Yu et al., CNS Neurosci Ther 2020):We agree and have revised the sentence, which is now supported by ref. 4 and by a newly added reference systematically analyzing E2F expression and prognosis in high-grade glioma (ref. 10, Yu et al., CNS Neurosci Ther 2020):
"Moreover, E2F1 is upregulated in high-grade gliomas and has been linked to adverse clinical features and poor survival in glioma patients[10]." (Page 2, Lines 79–81; new ref. 10 at Page 17, Lines 514–515)
Comment 4 (Introduction):
"The Introduction provides limited background on the relationship between ROS and E2F1. Because oxidative stress and DNA damage can also promote E2F1 stabilization or activation under certain conditions, the authors should discuss these context-dependent effects and provide relevant references to better support their proposed mechanism."
Response: This is an excellent point. The ROS–E2F1 relationship is indeed bidirectional and context-dependent—acute genotoxic stress stabilizes E2F1 via ATM/ATR-dependent phosphorylation (Lin et al., Genes Dev 2001, new ref. 13), whereas sustained oxidative stress can dismantle the E2F1-driven S-phase program. We have added this background, with references (new refs. 13 and 14), to the Introduction:
"Notably, the ROS–E2F1 relationship is context-dependent: acute genotoxic stress can stabilize and activate E2F1 through ATM/ATR-dependent phosphorylation[13], and redox signaling intersects with cell-cycle control in a phase-specific manner[14]; whether oxidative stress stabilizes or suppresses E2F1 therefore depends on the intensity and duration of the insult." (Pages 2–3, Lines 88–93; new refs. 13–14 at Page 17, Lines 520–523)
In our system, the direction of the relationship was defined experimentally: NAC-mediated ROS scavenging restored E2F1 (Fig. 3E–H), indicating that the sustained oxidative stress induced by PECT suppresses, rather than stabilizes, E2F1 in GBM cells.
Comment 5 (Materials and methods):
"Only two PECT concentrations (10 and 20 µM) were tested. The authors should provide a rationale for selecting these concentrations and clarify whether they were based on prior dose-response or IC50."
Response: We have added the dose rationale to Section 2.4 (Page 4, Lines 153–157):
"PECT concentrations (10 and 20 μM) were selected based on the dose range established in our previous study[18], in which PECT inhibited GBM cell viability in the low-micromolar range; 20 μM approximates the half-maximal inhibitory concentration under the 48-h conditions used here, and 10 μM serves as a sub-maximal dose to assess dose dependence." (Page 4, Lines 153–157)
Comment 6 (Materials and methods):
"In Supplementary Table S1, siE2F1#1 and siE2F1#2 share substantial sequence overlap. The authors should clarify how they selected these siRNAs and present their sequences in a consistent format."
Response: We thank the reviewer for noting this. The three siRNAs were not designed de novo but were selected from previously published, experimentally validated E2F1-targeting siRNAs: siE2F1#1 was reported by Subtil-Rodríguez et al. (Nucleic Acids Res 2014) [19], siE2F1#2 by Moniz et al. (J Cell Sci 2015) [20] and Sang et al. (Oncotarget 2017) [21], and siE2F1#3 by Jesenko et al. (Bioelectrochemistry 2025) [22]. As the reviewer correctly observes, siE2F1#1 and siE2F1#2 share substantial overlap (15 nucleotides); this is because both established siRNAs were independently validated against the same high-efficacy region of the E2F1 transcript (NM_005225.3, nt 494–517), whereas siE2F1#3 targets a distinct, non-overlapping region (nt 1310–1328). All three siRNAs achieved efficient knockdown (Fig. 2E) and produced concordant phenotypes; because siE2F1#3 is fully independent, these concordant results minimize concerns that the observed phenotype arises from an off-target effect of any single reagent. A clarifying sentence has been added to Section 2.6 (Page 4, Lines 171–176):
"Three siRNAs targeting E2F1 (NM_005225.3) were selected from previously validated sequences [19-22]; siE2F1#1 and siE2F1#2 target an overlapping high-efficacy region, whereas siE2F1#3 targets a non-overlapping region. All three achieved efficient knockdown (Fig. 2E), and the concordant phenotypes across siRNAs minimize concerns of off-target effects." (Page 4, Lines 171–176)
Supplementary Table S1 has been reformatted in a uniform 5′→3′ notation, with 3′-TT overhangs indicated, target coordinates on NM_005225.3 annotated, and the published source of each siRNA cited.
Comment 7 (Materials and methods):
"Molecular weight markers are not shown in the Western blot figures. The authors should provide the uncropped, original Western blot images, including molecular weight information, as supplementary material. In addition, some Western bands are not labeled in the current uncropped supplementary material file. For example, which proteins do the bands below the E2F1 bands in Fig.2E correspond to? Additionally, apart from the b-actin bands presented in the publication, each membrane's own β-actin should also be included in the supp file."
Response: We sincerely thank the reviewer for this constructive comment, which substantially improves the transparency and rigor of our Western blot data presentation. We have addressed all three points as follows:
(1) Molecular weight markers: In the revised Supplementary File (Original Western Blot Images), we now provide all available uncropped, unprocessed exposure images for every blot presented in the manuscript. The positions of the pre-stained molecular weight markers were recorded on the same gel/membrane for each experiment and are now indicated on every image, with the molecular weight of each target protein labeled alongside the corresponding band.
(2) Band labeling: All bands in the uncropped images have now been annotated with the protein identity and the corresponding figure panel. Specifically, regarding Fig. 2E: the bands below the E2F1 band correspond, from top to bottom, to β-actin (~42 kDa, loading control) and LC3-I/LC3-II (~16–18 kDa). These lysates from the siNC/siE2F1-transfected U251 cells were electrophoresed in the same gel run, and the membrane was additionally probed for LC3 as part of a parallel assessment of autophagy-related markers in this batch of samples; the LC3 data were not part of Fig. 2E and are labeled in the supplementary file solely for transparency.
(3) β-actin for each membrane: The β-actin blot corresponding to each individual membrane has now been included in the supplementary file, matched one-to-one with the figure panels in which the respective target-protein bands appear.
We must also candidly explain a limitation. The Western blot experiments of this study were performed using a membrane-cutting strategy, in which membranes were sectioned according to the molecular weights of the target proteins prior to antibody incubation. This approach was adopted to maximize the number of proteins that could be analyzed from limited GBM cell lysates and to avoid signal loss associated with repeated stripping and reprobing; consequently, intact full-length membrane scans were not systematically archived at the time. Moreover, the protein samples from these experiments have been entirely consumed by the assays reported in this study, and no residual material is available to repeat the electrophoresis for full-membrane imaging. We acknowledge that retaining complete full-membrane records should have been part of our standard practice.
In response to this comment, we have now adopted a strict full-membrane policy: all ongoing and future Western blot experiments in our laboratory retain complete uncut membrane images with molecular weight markers, and such images will be provided as supplementary material in any subsequent submission. We hope the reviewer will find the annotated uncropped images, molecular weight indications, and per-membrane β-actin controls provided in the revised supplementary file sufficient to verify the integrity of the data, and we are fully willing to repeat any key blot with full-membrane documentation should the reviewer or editor deem it necessary.
Comment 8 (Materials and methods):
"Some figures show approximately five individual data points per group, whereas the legends report n = 3. The authors should clarify the distinction between biological and technical replicates and specify the actual sample size used for statistical analysis."
Response: We apologize for the ambiguity. Throughout this study, n denotes the number of independent biological experiments; technical replicates within each biological replicate (e.g., replicate wells) were averaged to a single value before statistical analysis. The sample size differs among experiments: Fig. 1A and 1B were analyzed with n = 3 independent biological replicates, whereas Fig. 1C and 1D were analyzed with n = 5; the exact n for each experiment is now stated in the corresponding figure legend. We have added the following clarification to Section 2.10 (Page 5, Lines 217–220):
"Unless otherwise indicated, n denotes the number of independent biological experiments, and each plotted data point represents one independent biological replicate, measured as the mean of technical replicates." (Page 5, Lines 217–220) In addition, the exact n for each experiment is now stated in the corresponding figure legend.
All figure legends and graphs have been revised so that the number of plotted data points matches the stated n for every panel (Fig. 1A, B, n = 3; Fig. 1C, D, n = 5). The same correction has been applied to the Supplementary Materials: the viability assays in Fig. S3D, E are now explicitly stated as n = 5 independent experiments (individual data points plotted), whereas the qPCR and immunoblot-quantification panels (Fig. S1; Fig. S2; Fig. S3A, C) remain n = 3 independent experiments, consistent with the replicate definition in Section 2.10.
Comment 9 (Materials and methods):
"The statistical analysis requires further clarification. Experiments involving multiple factors, such as promoter type (WT/Mut) and treatment, would benefit from an appropriate factorial analysis, such as two-way ANOVA, including relevant interaction effects."
Response: We agree and have re-analyzed the two-factor experiments (Fig. 4B, C; promoter construct × treatment) by two-way ANOVA including the interaction term, followed by Šídák’s multiple comparisons test. Section 2.10 (Page 5, Lines 212–214) and the Figure 4 legend (Page 10, Lines 327–329) have been updated accordingly:
"…experiments involving two factors (e.g., promoter construct × treatment) were analyzed by two-way ANOVA including the interaction term, followed by Šídák’s multiple comparisons test." (Page 5, Lines 212–214)
"For panels B and C, data were analyzed by two-way ANOVA (promoter construct × treatment, including the interaction term) followed by Šídák’s multiple comparisons test; individual data points represent independent biological replicates." (Page 13, Lines 331–333)
Comment 10 (Results):
"ROS levels were measured only after 48 h of PECT treatment, when cell viability was already reduced. The authors should assess ROS levels at earlier time points and in the PECT + NAC group to confirm the proposed ROS-dependent mechanism."
Response: We agree that early-kinetic ROS data would strengthen the causal argument, and we explain our reasoning as follows. The 48-h endpoint was chosen to match the viability endpoint, and the ROS dependence of PECT’s effect is established functionally rather than solely by endpoint measurement: (i) NAC co-treatment substantially rescued viability in both cell lines (Fig. 1C, D), demonstrating that ROS accumulation is required for the cytotoxicity; (ii) NAC restored E2F1 protein (Fig. 3E–H), placing ROS upstream of E2F1; and (iii) the time-course analysis shows E2F1 mRNA declining at 12 h, before any substantial loss of viability, arguing against the decline being a secondary consequence of cell death. Additional early-time-point ROS measurements and ROS quantification in NAC rescued cells would require a new imaging series that cannot be completed and validated within the 10-day revision window; we have therefore added this explicitly to the Limitations:
"Third, ROS was measured at a single 48 h endpoint matched to the viability assay; early-kinetic ROS measurements, direct ROS quantification in NAC rescued cells, and RRM2 immunoblotting of the NAC rescue experiment would further strengthen the causal chain and are included in our ongoing follow-up study." (Page 15, Lines 438–441)
Comment 11 (Results):
"In Figure 1C–D, approximately five data points are shown per group, whereas the figure legend reports n = 3. The authors should clarify the number of biological and technical replicates. Additionally, Fig. 1B does not show data points. Data points should be added to all graphs throughout the article."
Response: Please see our response to Comment 8 regarding the replicate definition (now clarified in Section 2.10, Page 5, Lines 217–220). Following the reviewer’s suggestion, individual data points have been added to the graphs in the main text, including Fig. 1B, and the plotted points now correspond exactly to the stated n in every main-text figure. In addition, the legend of Figure S3 in the Supplementary Materials has been corrected: the previous blanket statement "n = 3 independent experiments" did not match the viability data shown in Fig. S3D, E, which are plotted with five individual data points per group; the legend now specifies n = 5 independent experiments for Fig. S3D, E and n = 3 independent experiments for Fig. S3A, C (and, unchanged, Fig. S1 and Fig. S2), so that the stated n matches the plotted points throughout.
Comment 12 (Results):
"The scale bars are not visible in Figure 1A. Additionally, the U87 20 µM panel appears to differ in cell density and image appearance. The authors should confirm that consistent imaging conditions and magnification were used."
Response: We apologize for the missing scale bars; they have been added to the revised Figure 1A (100 µm, as stated in the legend; the scale bar applies to all panels). Regarding the U87 20 µM panel, we thank the reviewer for prompting a closer look. We confirm that all panels were acquired under identical imaging conditions and magnification. The apparently different appearance of the U87 20 µM panel reflects the PECT treatment effect rather than inconsistent imaging: at the highest dose, the field is dominated by numerous rounded, brightly DCFH-DA–positive cells that are condensing and detaching as they undergo death, whereas the control field contains sparser, dimly stained, spread-out viable cells. This morphological interpretation is consistent with the corresponding viability quantification (Fig. 1D).
Comment 13 (Results):
"In Figure 3E–H, NAC treatment restores E2F1 expression, but its effect on RRM2 remains unclear. The authors should evaluate RRM2 expression in the Control, PECT, NAC, and PECT + NAC groups to further support the proposed ROS–E2F1–RRM2 signaling pathway. If it is not included, explain why and, if necessary, add it to the limitations."
Response: We thank the reviewer for explicitly offering the limitation route. We agree that RRM2 immunoblotting of the NAC rescue would complete the chain, but the samples from that experiment were not retained for re-blotting, and repeating the full four-group, two-cell-line experiment cannot be completed within the 10-day revision window. We respectfully note that the existing data already close most of this gap: NAC restores E2F1 (Fig. 3E–H); E2F1 knockdown reduces RRM2 (Fig. 2E–G); and E2F1 decline precedes RRM2 decline (Fig. S2C, D)—together supporting that ROS-dependent E2F1 suppression drives RRM2 downregulation. Following the reviewer’s suggestion, the missing validation is now explicitly acknowledged in the Limitations (Page 15, Lines 438–441; quoted under Comment 10), and the experiment is included in our ongoing follow-up study.
Comment 14 (Discussion):
"The Discussion section largely repeats the results and should be shortened to avoid redundancy and to focus more on interpreting and discussing the significance of the findings."
Response: We have condensed the Discussion. The principal-findings summary in the opening paragraph was shortened by approximately 40% (Page 13, Lines 347–352) and now reads:
"Our principal findings can be summarized as follows: PECT induces ROS accumulation that is required for its cytotoxicity; ROS-dependent suppression of E2F1 precedes and drives RRM2 downregulation; and a single GC-rich E2F1-response element in the RRM2 promoter is required for both E2F1-mediated transactivation and PECT-induced repression." (Page 13, Lines 347–352)
The remaining paragraphs were reviewed to retain interpretation and significance (comparison with prior work, the ROS–E2F1 context, limitations, and moderated translational implications) rather than result restatement.
Comment 15 (Discussion):
"The therapeutic implications are somewhat overstated, given that the study is limited to two GBM cell lines without normal-cell controls or in vivo validation. The authors should moderate these claims in the Discussion and Conclusions to better reflect the available experimental evidence."
Response: We agree and have moderated the claims throughout. Specifically:
"PECT may therefore represent a candidate modulator of this axis; however, our current evidence is limited to two GBM cell lines without normal-cell controls, and any therapeutic extrapolation would be premature." (Page 14, Lines 415–419)
"or RRM2-directed agents like triapine. Pharmacokinetic profiling, blood–brain barrier penetration, normal-glial toxicity, and orthotopic efficacy studies will be essential before PECT—alone or in rational combinations—can be evaluated in these settings." (Page 14, Lines 423–426)
In addition: (i) the Simple Summary now states that the findings "support further preclinical investigation of PECT-based strategies" (Page 1, Lines 21–23); (ii) the Abstract Conclusions now read "identifying this axis as a candidate for therapeutic targeting" (Page 2, Lines 44–46); (iii) the Conclusions now call for "merits further preclinical exploration, including validation in normal glial cells and in vivo models" (Page 15, Lines 453–455). For transparency, we note that in vivo anti-GBM efficacy of PECT was previously demonstrated in our earlier study (ref. 18), but we agree that the present manuscript should stand on its own evidence, and the text has been moderated accordingly.
Author Response File:
Author Response.pdf
Round 2
Reviewer 1 Report
Comments and Suggestions for AuthorsThe authors have provided detailed responses to the previous comments, however, some important mechanistic concerns remain insufficiently addressed. Overall, the revised manuscript has improved, but the central mechanistic conclusion remains incompletely supported, particularly regarding the functional involvement of RRM2. Additional experimental evidence or substantial moderation of the mechanistic claims is necessary before the manuscript can be considered for publication.
- The functional contribution of RRM2 to PECT-induced cytotoxicity remains insufficiently established. Although the authors have clarified the temporal relationship between E2F1 and RRM2 expression and provided additional discussion, these observations do not replace functional rescue experiments. In particular, it remains unclear whether E2F1 overexpression restores RRM2 protein expression following PECT treatment. Furthermore, direct RRM2 overexpression rescue experiments would provide important evidence that RRM2 downregulation contributes to the anticancer effects of PECT. This issue is central to the proposed molecular mechanism and should be addressed more convincingly.
Author Response
Dear Editor and Reviewer,
We sincerely thank the reviewer for the careful and constructive re-evaluation of our revised manuscript (review dated 07 November 2026). We are grateful for the recognition that the manuscript has improved, and for the precise identification of the remaining central concern — the functional involvement of RRM2 in PECT-induced cytotoxicity — together with the quality-of-reporting checklist. In this revision we have addressed these points along the two routes proposed by the reviewer, namely additional experimental evidence and moderation of the mechanistic claims. A point-by-point response is provided below. All changes are marked in blue in the tracked-changes version, in which only the changes of the present round are highlighted (the tracked changes of the previous round have been accepted for clarity) and continuous line numbering has been enabled; a clean version is also supplied. Page and line numbers below refer to the marked revised manuscript with continuous line numbering enabled, viewed in Final/Simple Markup mode.
Reviewer Comment 1 (general assessment):
"The authors have provided detailed responses to the previous comments, however, some important mechanistic concerns remain insufficiently addressed. Overall, the revised manuscript has improved, but the central mechanistic conclusion remains incompletely supported, particularly regarding the functional involvement of RRM2. Additional experimental evidence or substantial moderation of the mechanistic claims is necessary before the manuscript can be considered for publication."
Response: We agree with this balanced framing, and we have responded along exactly the two routes the reviewer proposes — additional experimental evidence and moderation of the mechanistic claims — as detailed point by point below.
Reviewer Comment 1 (specific):
"The functional contribution of RRM2 to PECT-induced cytotoxicity remains insufficiently established. Although the authors have clarified the temporal relationship between E2F1 and RRM2 expression and provided additional discussion, these observations do not replace functional rescue experiments. In particular, it remains unclear whether E2F1 overexpression restores RRM2 protein expression following PECT treatment. Furthermore, direct RRM2 overexpression rescue experiments would provide important evidence that RRM2 downregulation contributes to the anticancer effects of PECT. This issue is central to the proposed molecular mechanism and should be addressed more convincingly."
Response 1: We thank the reviewer for this precise and fair assessment. We have responded on three levels:
(a) NEW EXPERIMENTAL DATA. We have extended the NAC-rescue experiment to include RRM2 immunoblotting. In the UPDATED Figure 3 (panels E–H), NAC co-treatment restored not only E2F1 but also RRM2 protein levels in both U251 and U87 cells. The corresponding sentence in the revised Results (Section 3.3) now reads: "...whereas NAC alone had no detectable effect; importantly, NAC co-treatment also restored RRM2 protein levels in both cell lines (Fig. 3E–H), demonstrating that PECT suppresses the E2F1–RRM2 axis in a ROS-dependent manner." (Page 6, Lines 276–279) The Figure 3 legend has been updated accordingly (panels E–H now describe RRM2 and E2F1 blots with quantification; Figure 3 and legend, Page 13, Lines 335 and 337), and Limitations item 3 has been rewritten to state that "RRM2 immunoblotting has now been incorporated into the NAC rescue experiment (Fig. 3E–H)" (Page 17, Lines 472–476). These data close the gap previously acknowledged in the Limitations and directly demonstrate that RRM2, like E2F1, is restored when PECT-induced ROS accumulation is prevented.
(b) DIRECT RRM2-OVEREXPRESSION RESCUE EVIDENCE (ref. 18). Direct RRM2-overexpression rescue experiments in PECT-treated U251 and U87 cells were reported in our previous study (ref. 18, Jiang et al., Front. Oncol. 2022): RRM2 overexpression partially rescued GBM cells from PECT-induced proliferation inhibition, G2/M cell-cycle arrest, and autophagic-flux alterations. This published evidence is now explicitly integrated at five locations: the Simple Summary (Page 1, Lines 20–27); the Abstract, Results (Page 2, Lines 46–49); Section 3.4 (Page 7, Lines 302–304); the Discussion, third paragraph (Page 16, Lines 407–416); and Limitations item 1 (Page 17, Lines 460–462). Taken together with the new NAC-rescue data in (a), these data directly support the functional contribution of RRM2 downregulation to the anticancer effects of PECT. We further note the triangulation of three independent lines of evidence — ectopic E2F1 partially restores viability under PECT (Fig. S3D, E); published ectopic RRM2 partially restores proliferation and cell-cycle phenotypes under PECT (ref. 18); and ROS scavenging simultaneously restores both E2F1 and RRM2 (Fig. 3E–H) — each consistent with, and mutually reinforcing of, the proposed ROS–E2F1–RRM2 cascade.
(c) MODERATION OF MECHANISTIC CLAIMS. In agreement with the reviewer view that our previous wording overstated the evidence, we have moderated the central claims throughout the manuscript: (i) the Abstract Conclusions now read "Taken together, our data support a model in which PECT suppresses RRM2 transcription through ROS-dependent downregulation of its upstream activator E2F1, identifying the ROS–E2F1–RRM2 axis as a candidate for therapeutic targeting in GBM" (Page 2, Lines 49–55) (previously a definitive statement); (ii) the Simple Summary now states that restoring E2F1 "only partially protected tumor cells from PECT" and that "restoring RRM2 likewise provides only partial protection, indicating that additional mechanisms contribute" (Page 1, Lines 20–27); (iii) the novelty claim in the Introduction was softened ("this study provides the first definition of a ROS-initiated, E2F1-executed transcriptional checkpoint governing RRM2 expression in GBM, and maps PECT sensitivity to a single promoter element at nucleotide resolution" (Page 3, Lines 125–131)); (iv) in Results 3.3, the causal clause "and consequent impairment of GBM cell survival" was deleted, and the summary sentence now ends at "leading to reduced RRM2 transcription" (Page 6, Line 282–283); and (v) the Abstract (Results) and Discussion no longer imply that the E2F1–RRM2 axis alone accounts for the effects of PECT. Finally, in the spirit of full transparency, Limitations item 1 now explicitly states that "nevertheless, whether ectopic E2F1 restores endogenous RRM2 protein expression under PECT treatment remains to be directly measured" (Pages 17, Lines 462–464), and the planned follow-up studies have been revised accordingly.
Reviewer Comment 2 (quality-of-reporting checklist):
"The reviewer confirmed that the English is fine and rated the introduction/background, research design, methods, results, conclusions, and figures/tables as Can be improved."
Response 2: We have addressed every Can be improved rating as follows: the dimension Are the conclusions supported by the results? is met by the new experimental data and the moderated claims described in Response 1; Are the methods adequately described?, Are the results clearly presented?, and Are all figures and tables clear and well-presented? are met by corrections 2–6, 7 and 9 below; and Does the introduction provide sufficient background and include all relevant references? is met by the tempered novelty framing and citation corrections (corrections 2 and 9; Response 1c). Specific corrections and formatting improvements:
- Title page: changed the period to a comma in the author byline ("Shiguang Zhao 2,*"); capitalized affiliation 1 ("Department of Neurosurgery"). (Page 1, Lines 4–5)
- Methods 2.4: corrected the citation supporting the PECT dose range from ref. [15] to our own prior PECT study, ref. [18]. (Page 4, Lines 166)
- Methods: inserted spaces between numerals and temperature units throughout (37 °C; 37 ℃; 95 ℃; 4 ℃). (Pages 4–5, Lines 139, 162, 208, 215)
- Figure 2 legend: "(N = 207)" corrected to "(n = 207)" for consistency with all other sample-size notations. (Page 10, Lines 322)
- Figure 3: replaced with the updated figure (RRM2 immunoblotting added to the NAC-rescue panels E–H); the legend was updated to match. (Figure 3 and legend, Page 13, Lines 335, 337)
- 6. Discussion: paragraph 3 rewritten to integrate the published RRM2-overexpression rescue evidence (ref. 18) and to attribute the partial nature of both rescue experiments to additional PECT effector mechanisms. (Page 16, Lines 407–418)
- 7. Limitations: items 1 and 3 revised as described in Response 1(a) and 1(c). (Pages 17, Lines 459–464and 469–476)
- 8. Acknowledgments: "www. editage.cn" corrected to "www.editage.cn". (Page 18, Line 518)
- 9. References [4] and [8]: removed duplicated DOI entries while retaining the complete original citation details (journal, year, volume, pages). (Page 19, Lines 534and 545)
We hope that the new RRM2 immunoblot data, the explicit integration of the published RRM2-overexpression rescue experiments, and the moderated mechanistic claims satisfactorily address the reviewer concerns. As now transparently stated in the Limitations (Page 17, Lines 457–464), we remain fully committed to completing (i) the direct E2F1→RRM2 restoration experiment under PECT treatment and (ii) ChIP-qPCR verification of endogenous E2F1 occupancy at the predicted promoter element in our ongoing follow-up study.
Yours sincerely,
Haiping Jiang, on behalf of all authors
Department of Neurosurgery, The Affiliated Hospital of Guangdong Medical University, Zhanjiang, Guangdong, China
E-mail: 17390601611@163.com / 1404266177@qq.com; Tel: +86-17390601611
Round 3
Reviewer 1 Report
Comments and Suggestions for AuthorsThe revised manuscript address the reviewers' concerns