Metabolic Crosstalk in Triple-Negative Breast Cancer Lung Metastasis: Differential Effects of Vitamin D and E in a Co-Culture System

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This manuscript investigates metabolic interactions between triple-negative breast cancer cells and lung fibroblasts using an indirect co-culture model and examines the differential effects of Vitamin D and Vitamin E on these interactions. The topic is timely and relevant, as metabolic crosstalk within the metastatic tumor microenvironment remains an area of active investigation. The study presents a coherent dataset and provides descriptive insight into cell-type–specific changes in metabolic enzyme expression. However, several conclusions rely heavily on inferred metabolic activity without direct functional validation, and aspects of the experimental design, statistical analysis, and data interpretation require clarification and greater restraint. Addressing these points will substantially improve the rigor and impact of the manuscript.

Lines 22–24
Statements describing “enhanced oxidative phosphorylation” are not directly supported by functional metabolic assays and should be rephrased to indicate increased expression of oxidative metabolism–associated enzymes.

Lines 26–29
The claim that Vitamin E reinforces metabolic interplay implies causality and tumor-promoting effects that are not directly demonstrated and should be framed more cautiously as an association.

Lines 48–52
The rationale for focusing specifically on lung fibroblasts is underdeveloped and would benefit from clearer justification linking lung-specific stroma to TNBC metastatic metabolism.

Lines 74–86
The discussion of Vitamin E does not clearly distinguish between α-tocopherol and other vitamin E isoforms used in prior studies, which may affect interpretation of the results.

Lines 119–123
The co-culture setup lacks sufficient detail regarding media sharing and nutrient exchange between compartments, which is critical for interpreting metabolic interactions.

Lines 124–125
The concentrations of Vitamin D and Vitamin E are not adequately justified in terms of physiological relevance or prior experimental precedent.

Lines 141–146
The use of a paired Student’s t-test and lack of correction for multiple comparisons are insufficiently justified given the number of proteins analyzed.

Lines 156–158
Fibroblast activation is inferred solely from α-SMA and FAP expression, which should be clearly described as marker-based inference rather than functional CAF validation.

Lines 160–164
The reduction of HIF-1α in cancer cells is not reconciled with later interpretations of increased metabolic activity, creating conceptual inconsistency.

Lines 187–195
Reduced GLUL expression is interpreted as reduced glutamine synthesis without direct metabolic flux data, making this conclusion speculative.

Lines 191–192
Increased ASCT2 expression is equated with increased glutamine uptake despite the absence of uptake measurements.

Lines 196–205
Conclusions regarding enhanced TCA cycle activity rely solely on enzyme expression and should avoid implying functional flux changes.

Lines 199–201
The opposing regulation of fumarase in cancer cells versus fibroblasts is biologically intriguing but insufficiently emphasized or interpreted.

Lines 214–222
The suggestion that pyruvate accumulates and is diverted to anabolic pathways is speculative without metabolite quantification.

Lines 224–228
Lactate export by fibroblasts is inferred from transporter expression without direct extracellular lactate measurements.

Lines 238–244
Vitamin D is described as suppressing TCA cycle activity despite mixed regulation of individual enzymes, which should be acknowledged more explicitly.

Lines 258–260
The increase in HIF-1α in cancer cells following Vitamin D treatment contradicts its proposed suppressive role and requires more cautious interpretation.

Lines 264–269
Upregulation of GLUT1 and MCT4 in fibroblasts conflicts with the conclusion of broadly suppressed fibroblast metabolism and should be framed as pathway-specific effects.

Lines 276–283
Vitamin E–induced enhancement of oxidative phosphorylation is inferred without functional mitochondrial measurements and should be described more conservatively.

Lines 299–301
The increase in HIF-1α under Vitamin E treatment is not mechanistically reconciled with increased oxidative metabolism.

Lines 330–336
Lactate-driven metabolic coupling is discussed as established despite not being directly measured in this study.

Lines 342–348
The link between reduced fumarase expression and EMT is speculative, as EMT markers or functional assays were not assessed.

Lines 389–396
The opposing effects of Vitamin D on HIF-1α in cancer cells versus fibroblasts are intriguing but insufficiently explored mechanistically.

Lines 427–431
The conclusion that Vitamin E sustains a tumor-promoting microenvironment is not directly supported by functional outcome data and should be toned down.

 

Author Response

We thank the reviewer for providing a comprehensive analysis of our manuscript. We have addressed all the issues raised and believe that this has significantly improved the manuscript. Please see below for our point-by-point response to each comment.

 

Comment 1

Lines 22–24

Statements describing “enhanced oxidative phosphorylation” are not directly supported by functional metabolic assays and should be rephrased to indicate increased expression of oxidative metabolism–associated enzymes.

Answer:

We thank the reviewer for this insightful comment. We agree that as functional metabolic assays were not performed in this study, we cannot directly infer metabolic activity from enzyme expression levels alone. Accordingly, we have revised the text to avoid claims of enhanced oxidative phosphorylation activity and now describe our findings in terms of increased expression of oxidative metabolism–associated enzymes (e.g. Page 6; lines 225-229):

“Overall, these results suggest a remodeling of TCA cycle-associated metabolic enzyme expression in MDA-MB-231 cancer cells following co-culture with fibroblasts. This is consistent with alterations in oxidative and glutamine-associated metabolic pathways, as reflected by changes in enzyme expression patterns.”

 

Comment 2

Lines 26–29

The claim that Vitamin E reinforces metabolic interplay implies causality and tumor-promoting effects that are not directly demonstrated and should be framed more cautiously as an association.

Answer:

We thank the reviewer for this important comment. We agree that our data do not directly demonstrate causality or tumor-promoting effects. The text has therefore been revised to frame Vitamin E–associated metabolic changes as an association rather than a causal reinforcement of metabolic interplay or tumor promotion (Page 1; lines 39-45):

“Conversely, Vitamin E treatment was associated with increased expression of TCA cycle and oxidative metabolism-related markers in BrCa cells without significantly affecting fi-broblast glycolysis. Such differential metabolic responses may contribute to metabolic heterogeneity within the tumour microenvironment. These results provide valuable insights into the metabolic dynamics of TNBC metastases in the lung TME and demonstrate that Vitamins D and E exert distinct effects on metabolic cross-talk between cancer cells and fibroblasts.”

 

(Page 11; lines 335-339):

“Overall, our findings suggest that VE enhances glutamine metabolism capacity alongside TCA cycle enzymes in co-cultured cancer cells, while also increasing glycolysis enzymes in co-cultured fibroblasts. This suggests that unlike VD, VE may potentially sustain metabolic cross-talk between BrCa cancer cells and fibroblasts under the co-culture conditions.”

 

(Page 15, lines 480-483):

“However, VE did not alter this interplay in the same way and was associated with coordinated changes in metabolic enzyme expression in both cell types, suggesting a continued metabolic interplay within the co-culture system.”

 

Comment 3

Lines 48–52

The rationale for focusing specifically on lung fibroblasts is underdeveloped and would benefit from clearer justification linking lung-specific stroma to TNBC metastatic metabolism.

Answer:

We thank the reviewer for this comment. We have expanded the rationale to explicitly justify our focus on lung fibroblasts by highlighting the lung tropism of TNBC and the metabolically distinct role of lung-resident fibroblasts in supporting metastatic colonization and outgrowth (Page 2; lines 63-71):

“TNBC displays strong propensity for lung metastasis, where lung associated fibroblast act as a major stromal component that actively contributes to the metastatic microenvironment [10]. The lung microenvironment is metabolically different because of high oxygen availability and active nutrient exchange, which may particularly benefit oxidative and glutamine mediated metabolism in metastatic TNBC cells [11]. Therefore, the complex metabolic cross-talk between TNBC cells and lung fibroblasts remains poorly characterized, yet is crucial to understand in order to develop more effective therapeutics against metastatic TNBC [12].”

 

Comment 4:

Lines 74–86

The discussion of Vitamin E does not clearly distinguish between α-tocopherol and other vitamin E isoforms used in prior studies, which may affect interpretation of the results.

Answer:

We thank the reviewer for this comment. Our study employed Vitamin E specifically in the form of α-tocopherol. To improve clarity and avoid ambiguity, we have revised the text to explicitly state the Vitamin E isoform where discussed.

 

Comment 5:

Lines 119–123

The co-culture setup lacks sufficient detail regarding media sharing and nutrient exchange between compartments, which is critical for interpreting metabolic interactions.

Answer:

We thank the reviewer for this comment. The Methods section has been clarified to indicate that the 0.4 μm porous membrane allows diffusion of soluble factors, metabolites, and cytokines, enabling metabolic and paracrine interactions between MDA-MB-231 and MRC-5 cells (Page 4; lines 141-146):

“Inserts containing one cell type were then placed into wells containing the other cell type, allowing the diffusion of soluble factors, metabolites and cytokines through the porous membrane enabling metabolic and paracrine interactions between cells. Cocultures were incubated for an additional 72 h. To assess the effects of VD or VE, 20 nM VD [32] or 20 µM VE [31] was added to the co-culture media for the entire 72 h incubation.”

 

Comment 6:

Lines 124–125

The concentrations of Vitamin D and Vitamin E are not adequately justified in terms of physiological relevance or prior experimental precedent.

Answer:

We thank the reviewer for this helpful comment. The rationale for the selected Vitamin D and Vitamin E concentrations, including their physiological relevance and support from prior experimental studies, has now been clarified and added to the Discussion section (Page 14; lines 423-426):

“In the present study, VD was evaluated across a range of physiologically relevant and supraphysiological (pharmacological) concentrations (10–100 nM) [32, 53], identifying 20 nM as an effective dose that was used for subsequent experiments.”

 

Page 14; lines 449-453:

“VE is an antioxidant nutraceutical whose effective concentrations vary across experi-mental settings. In vitro studies commonly employ VE within a broad micromolar range (approximately 5–80 µM) to evaluate its biological effects in different disease contexts. [31]. Following pilot studies examining VE at doses ranging from of 5 to 80 µM, we identified 20 µM as an effective dose that was used in the present study.”

 

Comment 7:

Lines 141–146

The use of a paired Student’s t-test and lack of correction for multiple comparisons are insufficiently justified given the number of proteins analyzed.

Answer:

We thank the reviewer for this comment. Statistical analyses were performed using paired Student’s t-tests for predefined comparisons between treatment and control conditions for individual proteins. As these analyses were hypothesis-driven and conducted independently for each target, no correction for multiple comparisons was applied. This clarification has now been added to the Methods section (Page 4; lines 165-167):

“Statistical significance was assessed using a paired Student’s t-test for predefined comparisons between treatment and control groups for each protein, with p ≤ 0.05 considered statistically significant.”

 

Comment 8:

Lines 156–158

Fibroblast activation is inferred solely from α-SMA and FAP expression, which should be clearly described as marker-based inference rather than functional CAF validation.

Answer:

We thank the reviewer for this comment. We have revised the text to clarify that fibroblast activation is inferred based on increased expression of established CAF markers (α-SMA and FAP), rather than representing functional validation of CAF conversion (Page 4; lines 178-181):

“Co-culture significantly increased the expression of both activation markers in MRC-5 fibroblasts, providing marker-based evidence of fibroblast activation towards a CAF-like phenotype (Figure 1A).”

 

Comment 9:

Lines 160–164

The reduction of HIF-1α in cancer cells is not reconciled with later interpretations of increased metabolic activity, creating conceptual inconsistency.

Answer:

We thank the reviewer for this comment. We have clarified in the Discussion that although HIF‑1α expression decreased in TNBC cells during co-culture, the observed increase in TCA cycle activity likely reflects a HIF‑1α-independent metabolic shift toward glutamine-dependent metabolism, driven by stromal interactions. Relevant literature supporting HIF‑1α-independent metabolic reprogramming in cancer-stroma crosstalk has also been cited (Page 14; lines 402-413):

“In our study, HIF-1α was downregulated in cancer cells but upregulated in fibroblasts in the co-culture system. This differential expression reflects the distinct metabolic adaptations of cancer cells and fibroblasts in the TME. Notably, a study demonstrated that elevated levels of reactive oxygen species (ROS) originating in cancer cells can induce oxidative stress in neighboring CAFs, leading to the activation and stabilization of HIF-1α [22]. Although HIF‑1α expression was reduced in TNBC cells during co-culture, the observed increase in TCA cycle enzyme expression, may reflect a shift toward glutamine-dependent metabolism, supported by increased glutamine uptake via ASCT2. This metabolic adaptation is likely driven by stromal interactions in the co-culture environment and occurs in-dependently of HIF‑1α. HIF‑1α-independent metabolic reprogramming in the context of cancer-stroma crosstalk has been previously reported, supporting the role of the TME in shaping cancer cell metabolism [47].”

 

Comment 10:

Lines 187–195

Reduced GLUL expression is interpreted as reduced glutamine synthesis without direct metabolic flux data, making this conclusion speculative.

Answer:

We thank the reviewer for this comment. The text has been revised to indicate that changes in GLUL, GDH, and ASCT2 expression suggest a shift in glutamine metabolism based on marker analysis, without implying direct measurement of metabolic flux (Page 6; lines 208-217):

“Examining glutamine metabolic enzymes, GLUL was significantly reduced in MDA-MB-231 cells when co-cultured with fibroblasts (Figure 1D). The expression of GDH, which enables glutamine entry into the TCA cycle [37], was significantly upregulated in cancer cells, while remaining unaffected in fibroblasts following coculture (Figure 1D). Interestingly, the glutamine transporter ASCT2 was significantly elevated in cancer cells upon co-culture with fibroblasts, providing marker-based evidence consistent with enhanced glutamine uptake (Figure 1D). These findings suggests that glutamine metabolism in TNBC cells may be affected by co-culture with fibroblasts, as characterized by marker-based evidence of decreased glutamine synthesis and increased glutamine uptake toward the TCA cycle.”

 

Comment 11:

Lines 191–192

Increased ASCT2 expression is equated with increased glutamine uptake despite the absence of uptake measurements.

Answer:

We thank the reviewer for this comment. We have revised the text to clarify that the observed increase in ASCT2 expression provides marker-based evidence suggestive of enhanced glutamine uptake, as addressed for the above comment (Page 6; lines 208-217).

 

Comment 12:

Lines 196–205

Conclusions regarding enhanced TCA cycle activity rely solely on enzyme expression and should avoid implying functional flux changes.

Answer:

We thank the reviewer for this comment. We have revised the text to clarify that our conclusions about the TCA cycle are based on changes in enzyme expression and do not imply direct measurements of metabolic activity or flux (Page 6; lines 225-229):

“Overall, these results suggest a remodeling of TCA cycle-associated metabolic enzyme expression in MDA-MB-231 cancer cells following co-culture with fibroblasts. This is consistent with alterations in oxidative and glutamine-associated metabolic pathways, as reflected by changes in enzyme expression patterns.”

 

Comment 13:

Lines 199–201

The opposing regulation of fumarase in cancer cells versus fibroblasts is biologically intriguing but insufficiently emphasized or interpreted.

Answer:

We thank the reviewer for highlighting this point. We have expanded the Discussion to better interpret the opposing regulation of fumarase by incorporating evidence that fumarase is subject to transcriptional and epigenetic regulation and also possesses non-metabolic roles, including involvement in DNA damage response pathways. This added context supports the biological plausibility of its differential regulation within the co-culture microenvironment (Page 13; lines 371-386):

“Notably, while most TCA cycle enzymes were upregulated in co-cultured MDA-MB-231 cells, fumarase expression was reduced. Further, its expression was in-creased in co-cultured fibroblasts. This opposing regulation of fumarase may reflect active cell-type specific metabolic reprogramming. Previous studies have reported that fumarase expression can be transcriptionally and epigenetically regulated by chromatin-remodeling factors such as lymphocyte-specific helicase (LSH), leading to its suppression in certain cancers [44]. Furthermore, fumarase has been reported to possess non-metabolic functions, including roles in DNA damage response pathways, which may influence its expression and subcellular localization in response to microenvironment-associated stress [45]. It has also been reported that a reduction in fumarase activity leads to accumulation of fumarate, which acts as an oncometabolite by driving epigenetic changes that promote the epithelial-to-mesenchymal transition (EMT), ultimately promoting tumour invasion and metastasis [46]. In line with these findings, the decreased levels of fumarase observed in our study indicates a metabolic adaptation in TNBC cells, which may contribute to tumour progression within the coculture context, although whether this is linked to EMT requires further examination.” 

 

Comment 14:

Lines 214–222

The suggestion that pyruvate accumulates and is diverted to anabolic pathways is speculative without metabolite quantification.

Answer:

We thank the reviewer for this comment. We have revised the text to focus on the observed changes in enzyme expression, framing them as indicative of altered regulation of pyruvate-utilizing pathways in cancer cells under co-culture conditions. Speculative statements regarding pyruvate accumulation or diversion to other metabolic pathways have been removed to ensure conclusions are fully supported by the data (Page 7; lines 236-245):

“However, the mitochondrial enzyme PDH, which converts pyruvate into Acetyl-CoA for entry into the TCA cycle, was significantly down-regulated in cancer cells following co-culture (Figure 1F), indicating reduced enzymatic capacity for pyruvate entry into TCA cycle in coculture conditions and potentially reflecting a shift in pyruvate use.

Apart from entry into the TCA cycle, the other fate of pyruvate is conversion into lactate by the enzyme LDHA, followed by export from cells via the MCT4 transporter [38]. Notably, both LDHA and MCT4 expression were also reduced in cancer cells following co-culture (Figure 1F), reflecting a coordinated downregulation of the pathways for lactate metabolism. Together, these changes in enzyme expression suggest an altered regulation of pyruvate-utilizing pathways in cancer cells under co-culture conditions.”

 

Comment 15:

Lines 224–228

Lactate export by fibroblasts is inferred from transporter expression without direct extracellular lactate measurements.

Answer:

We thank the reviewer for the comment. We have toned down the text to describe these changes as indicative of increased glycolytic capacity in fibroblasts, without implying direct lactate export (Page 7; lines 245-251):

“Analysis of LDHA and MCT4 in fibroblasts revealed significant upregulation of both enzymes after co-culture (Figure 1F). Further, the lactate transporter MCT1, which was only detected in fibroblasts, was also up-regulated following co-culture with cancer cells (Figure 1F). These changes, together with the elevated GLUT1 and HK2 are indicative of increased glycolytic activity in cocultured fibroblasts and may reflect enhanced lactate production and its export into the TME.”

 

Comment 16:

Lines 238–244

Vitamin D is described as suppressing TCA cycle activity despite mixed regulation of individual enzymes, which should be acknowledged more explicitly.

Answer:

We thank the reviewer for this comment. We have revised the text to explicitly acknowledge the mixed regulation of TCA cycle enzymes by VD, emphasizing that these changes reflect selective, enzyme-specific effects rather than uniform suppression of TCA cycle activity (Page 7; lines 262-267).

“Examining the TCA cycle enzymes, VD treatment significantly down-regulated key enzymes such as SDHA, CS, and fumarase in cancer cells (Figure 2B). However, it did not alter the expression of aconitase or MPC2, while notably upregulating IDH2 levels (Figure 2B). Overall, these findings suggest that VD modulates TCA cycle–associated enzymes in a selective manner, reflecting complex and enzyme-specific effects on TCA cycle activity.”

 

Comment 17:

Lines 258–260

The increase in HIF-1α in cancer cells following Vitamin D treatment contradicts its proposed suppressive role and requires more cautious interpretation.

 Answer:

We thank the reviewer for this insightful comment. The conclusion has been revised to adopt a more cautious interpretation, emphasizing selective and context-dependent metabolic modulation by VD (Page 9; lines 278-284):

“Notably, expression of the vitamin D receptor (VDR) and HIF-1α were significantly elevated in the co-cultured cancer cells (Figure 2C). Collectively, these observations reflect that VD induces selective and context-dependent metabolic changes in cancer cells under coculture conditions. The concurrent downregulation of specific metabolic enzymes alongside enhanced HIF-1α and ASCT2 expression highlights a complex and nuanced metabolic response to VD within the coculture microenvironment.”

(Page 14; lines 438-445):

“We observed that in TNBC cancer cells, both VDR and HIF-1α are enhanced following VD treatment, whereas in fibroblasts, VD treatment up-regulated VDR but suppressed HIF-1α expression. This was accompanied by reduced glycolysis in fibroblasts, an effect that is likely mediated by HIF-1α down-regulation [58]. The opposing effects of VD on HIF-1α in cancer cells and fibroblasts likely reflect cell-type specific regulatory mechanisms and microenvironmental influences, highlighting complex metabolic crosstalk within the TME, necessitating more exploration of underlying signalling mechanisms of these interactions.”

 

Comment 18:

Lines 264–269

Upregulation of GLUT1 and MCT4 in fibroblasts conflicts with the conclusion of broadly suppressed fibroblast metabolism and should be framed as pathway-specific effects.

Answer:

We thank the reviewer for this comment. We have revised the text to avoid suggesting a global suppression of fibroblast metabolism. The results and conclusions now emphasize that VD induces selective, pathway-specific metabolic modulation in fibroblasts (Page 9; lines 291-296):

“Notably, the glucose importer GLUT1 and lactate exporter MCT4 were up-regulated in these cells, highlighting a selective and pathway specific metabolic response rather than a uniform metabolic suppression.

Overall, our findings suggest that VD treatment partially counteracts cancer-driven metabolic re-programming of fibroblasts in co-culture by selectively modulating glycolytic and lactate processing pathways.”

 

Comment 19:

Lines 276–283

Vitamin E–induced enhancement of oxidative phosphorylation is inferred without functional mitochondrial measurements and should be described more conservatively.

Answer:

We thank the reviewer for this suggestion. The Results text has been revised to indicate that VE treatment is associated with increased expression of glutamine- and mitochondria-related metabolic enzymes in cancer cells, without implying direct enhancement of metabolic flux or oxidative phosphorylation (Page 9; lines 305-307):

“Collectively, these observations indicate that VE treatment is associated with increased expression of glutamine and mitochondria associated metabolic enzymes in cancer cells under coculture conditions.”

 

Comment 20:

Lines 299–301

The increase in HIF-1α under Vitamin E treatment is not mechanistically reconciled with increased oxidative metabolism.

Answer:

We thank the reviewer for this comment. The text has been revised to clarify that the concurrent upregulation of HIF-1α and TCA cycle–associated enzymes under VE treatment may reflect an adaptive metabolic response to the co-culture environment, emphasizing the complex regulation of cancer cell metabolism without implying a direct enhancement of oxidative metabolism (Page 11; lines 324-328).

“Finally, HIF-1α was increased in the cancer cells following VE treatment, an effect that was similar to that observed with VD treatment. The simultaneous increase of HIF-1α and TCA cycle associated enzymes may indicate an adaptive metabolic response to the coculture microenvironment, highlighting the complexity of metabolic regulation in response to VE treatment.”

 

Comment 21:

Lines 330–336

Lactate-driven metabolic coupling is discussed as established despite not being directly measured in this study.

Answer:

We thank the reviewer for this comment. The text has been revised to clarify that the observed enzyme and transporter expression changes indicate possible metabolic adaptations and crosstalk between cancer cells and fibroblasts, without implying direct measurements of flux or metabolite transfer (Page 13; lines 362-366):

“In parallel, glycolytic enzymes in co-cultured fibroblasts were upregulated, which may suggest increased lactate production and export. This lactate could possibly contribute to metabolic crosstalk with cancer cells, to support mitochondrial metabolism and anabolic processes (Figure 4A) [42], although further direct assessment of lactate transfer between these cells is required.”

 

Comment 22:

Lines 342–348

The link between reduced fumarase expression and EMT is speculative, as EMT markers or functional assays were not assessed.

Answer:

We thank the reviewer for this comment. The text has been revised to remove the implication that reduced fumarase leads to EMT in the current study, and to acknowledge that this would need to be directly assessed (Page 13; lines 379-386).

“It has also been reported that a reduction in fumarase activity leads to accumulation of fumarate, which acts as an oncometabolite by driving epigenetic changes that promote the epithelial-to-mesenchymal transition (EMT), ultimately promoting tumour invasion and metastasis [46]. In line with these findings, the decreased levels of fumarase observed in our study indicates a metabolic adaptation in TNBC cells, which may contribute to tumour progression within the coculture context, although whether this is linked to EMT requires further examination.”  

 

Comment 23:

Lines 389–396

The opposing effects of Vitamin D on HIF-1α in cancer cells versus fibroblasts are intriguing but insufficiently explored mechanistically.

Answer:

We thank the reviewer for this insightful comment. The discussion has been revised and appropriately tempered to avoid mechanistic over-interpretation (Page 14; lines 442-447):

“The opposing effects of VD on HIF-1α in cancer cells and fibroblasts likely reflect cell-type specific regulatory mechanisms and microenvironmental influences, highlighting complex metabolic crosstalk within the TME, necessitating more exploration of underlying signalling mechanisms of these interactions. Nonetheless, our study is the first to report that VD potentially suppresses glycolysis in co-cultured CAFs, highlighting a novel mechanism by which VD potentially modulates the TME (Figure 4B).”

 

 Comment 24:

Lines 427–431

The conclusion that Vitamin E sustains a tumor-promoting microenvironment is not directly supported by functional outcome data and should be toned down.

Answer:

We thank the reviewer for this comment. The conclusion has been revised to avoid causal claims and to describe the effects of Vitamin E more cautiously as associative metabolic changes rather than tumour-promoting outcomes (Page 15; lines 486-489):

“However, VE did not alter this interplay in the same way and was associated with coordinated changes in metabolic enzyme expression in both cell types, suggesting a continued metabolic interplay within the co-culture system.”

Reviewer 2 Report

Comments and Suggestions for Authors

 

The authors investigated a clinically important subject, metabolic interactions within the tumor microenvironment of triple-negative breast cancer (TNBC) lung metastasis, by employing a Transwell co-culture model of MDA-MB-231 cells and MRC-5 lung fibroblasts. They propose that this co-culture leads to a "Reverse Warburg" metabolic phenotype, which is variably affected by Vitamins D and E. While the study addresses a crucial question and presents intriguing, potentially contradictory findings concerning Vitamin E, the manuscript is currently in a preliminary stage. The primary conclusions were predominantly drawn from protein expression data without validation through functional metabolic or phenotypic assays. However, substantial additional evidence is required to support these claims.

The primary limitation of this study is the reliance on Western blot data alone to infer alterations in metabolic flux, such as glycolysis and TCA cycle activity. Although the upregulation or downregulation of key enzymes, including GLUT1, HKII, and CS, may suggest changes, it does not provide direct evidence of the modified pathway activity.

Functional assays are crucial for validating these metabolic claims. The authors should incorporate Seahorse Analyzer assays to assess the extracellular acidification rate (ECAR) and oxygen consumption rate (OCR) under mono- and co-culture conditions, with and without vitamin treatment. Additionally, direct metabolite measurements, such as lactate secretion (glycolytic flux) and glucose/glutamine consumption from conditioned media, should be included. Enzymatic activity assays for key enzymes, such as hexokinase and citrate synthase, are necessary to confirm that changes in protein levels correspond to the functional activity.

The study lacks a connection to functional cancer cell phenotypes, as it concludes with the measurement of protein markers without demonstrating the functional implications of the observed metabolic shifts on the behavior of cancer cells. To substantiate claims regarding a "tumor-promoting microenvironment," it is essential to provide data linking these metabolic changes to the aggressiveness of cancer cells. Assays for proliferation (e.g., MTT and colony formation), migration, and invasion in the co-culture setting, with and without vitamin treatment, are required.

The validation of the "Reverse Warburg" or Lactate Shuttle Hypothesis remains incomplete. This model suggests that fibroblasts export lactate via MCT4 to support oxidative phosphorylation (OXPHOS) in cancer cells through MCT1. However, Figure 1F indicates that MCT1 expression was assessed solely in fibroblasts and not in cancer cells. The authors are required to present western blot data to confirm MCT1 expression in MDA-MB-231 cells under the specified experimental conditions.

The phenomenon referred to as the "Mechanistic Gap in the Vitamin E Paradox," wherein the antioxidant Vitamin E appears to foster a pro-tumorigenic metabolic profile characterized by increased glycolysis and HIF-1α expression, remains intriguing yet mechanistically unresolved. Although the authors cited conflicting literature, they did not explore the underlying cause within their experimental framework. The authors should examine whether Vitamin E, at the 20 µM concentration employed, functions as a pro-oxidant in this specific co-culture context. Therefore, assessing intracellular ROS levels in both cell types after treatment is essential. Additionally, exploring the impact of Vitamin E on other regulators of HIF-1α, such as PHD activity and α-KG levels, could yield valuable mechanistic insights.

The generalizability of this study is limited by the use of a single TNBC cell line and a single lung fibroblast line. Given the heterogeneity of TNBC and variability in the stromal compartment, it is recommended that key experiments be replicated using at least one additional TNBC cell line (e.g., BT-549 or HCC1806) and/or a different lung fibroblast line to enhance the broader applicability of the findings.

Statistical Analysis: The employment of n=2 replicates, as noted in certain figure legends, is inadequate for conducting a rigorous statistical analysis. All quantitative data were obtained from a minimum of three independent biological replicates. Western Blot Quantification: The loading controls (β-actin) in several figures (e.g., 2C, 3C) exhibit inconsistent loading or saturation issues. The authors must ensure that the blots fall within the linear dynamic range for densitometry and provide complete, uncropped blot images in the supplementary materials.

Physiological Relevance of Doses: A concise justification for the selected concentrations of Vitamin D (20 nM) and Vitamin E (20 µM), referencing achievable physiological or pharmacologically relevant levels in serum or tissue, would enhance the translational relevance of the study. Schematic Model (Figure 4): The model incorporates elements (e.g., “ROS” transfer) that were not directly measured in this study. The figure legend should explicitly indicate that it is a “proposed model” based on the current data and literature, clearly distinguishing between the observed and hypothesized interactions.

This manuscript presents a promising pilot study with novel observations on vitamin-mediated modulation of stromal-tumor metabolic crosstalk. However, the conclusions appear to be overstated in relation to the evidence. The primary concern is the discrepancy between protein expression data and claims of functional metabolic reprogramming. I recommend major revisions. Consideration for publication should occur only after the authors address the essential requirements for functional metabolic assays (Seahorse/metabolomics) and establish a connection between their findings and relevant cancer cell phenotypes. Providing mechanistic data on the effect of Vitamin E and validating key findings in an additional cell line model would substantially enhance the manuscript.

Author Response

We sincerely thank the reviewer for their thorough, constructive, and insightful evaluation of our manuscript. We have carefully considered all comments and agree that many of the suggested functional, metabolic, and mechanistic studies would substantially strengthen the work. While these experiments are beyond the scope of the present study, which was designed as a pilot investigation, we have made extensive revisions to ensure that the data are not over-interpreted and that the conclusions are appropriately aligned with the experimental evidence presented. We have also explicitly acknowledged the study’s limitations and incorporated the reviewer’s suggestions as directions for future research. We hope that these revisions adequately address the reviewer’s concerns and clarify the intent and contribution of this preliminary study.

 

Comment 1:

While the study addresses a crucial question and presents intriguing, potentially contradictory findings concerning Vitamin E, the manuscript is currently in a preliminary stage. The primary conclusions were predominantly drawn from protein expression data without validation through functional metabolic or phenotypic assays. However, substantial additional evidence is required to support these claims.

Answer:

We thank the reviewer for this important point and agree that further functional validation and metabolic assays are required to validate the primary conclusions. However, this was a pilot study to investigate the potential impact of Vitamins D and E on the metabolic interactions between TNBC cells and lung fibroblasts, with the preliminary data presented now justifying further in-depth analysis. To clarify this, we have extensively revised the manuscript to ensure the data presented is not over-interpreted (as also suggested by Revier 1). For a detailed description of the changes that were made to address the issue of conclusions being drawn from protein expression data, we kindly ask the reviewer to refer to our responses to Revier 1.

 

Comment 2:

The primary limitation of this study is the reliance on Western blot data alone to infer alterations in metabolic flux, such as glycolysis and TCA cycle activity. Although the upregulation or downregulation of key enzymes, including GLUT1, HKII, and CS, may suggest changes, it does not provide direct evidence of the modified pathway activity. Functional assays are crucial for validating these metabolic claims. The authors should incorporate Seahorse Analyzer assays to assess the extracellular acidification rate (ECAR) and oxygen consumption rate (OCR) under mono- and co-culture conditions, with and without vitamin treatment. Additionally, direct metabolite measurements, such as lactate secretion (glycolytic flux) and glucose/glutamine consumption from conditioned media, should be included. Enzymatic activity assays for key enzymes, such as hexokinase and citrate synthase, are necessary to confirm that changes in protein levels correspond to the functional activity.

 

Answer:

We agree with this important point and have now revised the manuscript to ensure that our interpretation and conclusions drawn are based on the expression of metabolic enzymes. We note that further functional and metabolic assays are beyond the aim and scope of the current study, which was designed as a pilot study to determine whether further investigation into the effects of Vitamins D and E on the metabolic cross-talk within the TNBC tumour microenvironment is justified. These limitations and suggested further experimental validation has also now been added to the manuscript.

 

Comment 3:

The study lacks a connection to functional cancer cell phenotypes, as it concludes with the measurement of protein markers without demonstrating the functional implications of the observed metabolic shifts on the behavior of cancer cells. To substantiate claims regarding a "tumor-promoting microenvironment," it is essential to provide data linking these metabolic changes to the aggressiveness of cancer cells. Assays for proliferation (e.g., MTT and colony formation), migration, and invasion in the co-culture setting, with and without vitamin treatment, are required.

Answer:

We agree with this important point and have now revised the manuscript to ensure that our interpretation and conclusions drawn are based on the expression of the metabolic enzymes assessed, with the acknowledgment that further functional and metabolic assays are required to validate this. We note that further functional and metabolic assays are beyond the aim and scope of the current study, which was designed as a pilot study to determine whether further investigation into the effects of Vitamins D and E on the metabolic cross-talk within the TNBC tumour microenvironment is justified.

 

Comment 4:

The validation of the "Reverse Warburg" or Lactate Shuttle Hypothesis remains incomplete. This model suggests that fibroblasts export lactate via MCT4 to support oxidative phosphorylation (OXPHOS) in cancer cells through MCT1. However, Figure 1F indicates that MCT1 expression was assessed solely in fibroblasts and not in cancer cells. The authors are required to present western blot data to confirm MCT1 expression in MDA-MB-231 cells under the specified experimental conditions.

Answer:

We thank the reviewer for this important point. We note that the expression of MCT1 was assessed in both fibroblasts and MDA-MB-231 cells, however MCT1 was not detected in the cancer cells under the conditions used. This was despite multiple independent repeats and use of longer exposure times.

 

Comment 5:

The phenomenon referred to as the "Mechanistic Gap in the Vitamin E Paradox," wherein the antioxidant Vitamin E appears to foster a pro-tumorigenic metabolic profile characterized by increased glycolysis and HIF-1α expression, remains intriguing yet mechanistically unresolved. Although the authors cited conflicting literature, they did not explore the underlying cause within their experimental framework. The authors should examine whether Vitamin E, at the 20 µM concentration employed, functions as a pro-oxidant in this specific co-culture context. Therefore, assessing intracellular ROS levels in both cell types after treatment is essential. Additionally, exploring the impact of Vitamin E on other regulators of HIF-1α, such as PHD activity and α-KG levels, could yield valuable mechanistic insights.

Answer:

We thank the reviewer for this important and insightful comment. Yes, it would certainly be important to assess whether Vitamin E acts as an anti- or pro-oxidant in the co-culture context. While this was beyond the scope of our study, we have now added this important point into the discussion section (Page 15; lines 452-456)

“Following pilot studies examining VE at doses ranging from of 5 to 80 µM, we identified 20 µM as an effective dose that  was used in the present study. Whether this dose influ-ences the generation of reactive oxygen species (ROS) within the co-culture context was not assessed in this study and remains to be established. However, this dose of VE did influence the expression of numerous metabolic enzymes in both cancer cells and fibro-blasts.”

 

(Page 15; lines 467-470)

“Overall, our study reveals that VE may potentially enhance the metabolic interplay be-tween MDA-MB-231 cells and fibroblasts (Figure 4C), an effect that requires further investigation to understand its biological significance and potential redox effects in TNBC lung metastases.”

 

Comment 6:

The generalizability of this study is limited by the use of a single TNBC cell line and a single lung fibroblast line. Given the heterogeneity of TNBC and variability in the stromal compartment, it is recommended that key experiments be replicated using at least one additional TNBC cell line (e.g., BT-549 or HCC1806) and/or a different lung fibroblast line to enhance the broader applicability of the findings.

Answer:

We agree with this important point. While we completely agree that further cell lines are required to validate these effects and ensure they are not cell-specific, the pilot nature of this study meant that this was not feasible due to time and funding constraints. We have now made the need for further experimental validation in additional cell lines and models clear in our “Limitations and directions for future research” section (Page 15; lines 472-481):

“While these findings provide valuable insights into metabolic interactions in meta-static BrCa, our study had limitations. First, the in vitro model does not fully recapitulate the complexity of the TME, including the effect of immune cells, extracellular matrix, and systemic factors. Moreover, the study examined a limited number of metabolic enzymes in a single TNBC cell line. To further validate these results, future studies should assess mul-tiple TNBC and fibroblast cell lines and perform unbiassed metabolic analysis that in-cludes untargeted metabolomics to more accurately dissect the metabolic alterations at play. More physiological conditions such as 3D co-culture systems, patient-derived or-ganoids, or in vivo metastatic models, are needed to validate the intricate metabolic dynamics within the TME and how VD and VE influence this.”

 

Comment 7:

Statistical Analysis: The employment of n=2 replicates, as noted in certain figure legends, is inadequate for conducting a rigorous statistical analysis. All quantitative data were obtained from a minimum of three independent biological replicates. Western Blot Quantification: The loading controls (β-actin) in several figures (e.g., 2C, 3C) exhibit inconsistent loading or saturation issues. The authors must ensure that the blots fall within the linear dynamic range for densitometry and provide complete, uncropped blot images in the supplementary materials.

Answer:

We would like to note that for all experiments presented, at least 3 independent replicates were performed, as indicated in both the Materials and Methods section (section 2.6) and in the figure legends for Figures 1, 2 and 3, thus enabling statistical analysis.

 

Comment 8:

Physiological Relevance of Doses: A concise justification for the selected concentrations of Vitamin D (20 nM) and Vitamin E (20 µM), referencing achievable physiological or pharmacologically relevant levels in serum or tissue, would enhance the translational relevance of the study.

Answer:

We thank the reviewer for this important point and have now added further justification for the selected concentrations of Vitamin D and E (Page 14; lines 423-426):

“In the present study, VD was evaluated across a range of physiologically relevant and supraphysiological (pharmacological) concentrations (10–100 nM) [32, 53], identifying 20 nM as an effective dose that was used for subsequent experiments.”

 

Page 14; lines 449-453:

“VE is an antioxidant nutraceutical whose effective concentrations vary across experi-mental settings. In vitro studies commonly employ VE within a broad micromolar range (approximately 5–80 µM) to evaluate its biological effects in different disease contexts. [31]. Following pilot studies examining VE at doses ranging from of 5 to 80 µM, we identified 20 µM as an effective dose that was used in the present study.”

 

Comment 9:

Schematic Model (Figure 4): The model incorporates elements (e.g., “ROS” transfer) that were not directly measured in this study. The figure legend should explicitly indicate that it is a “proposed model” based on the current data and literature, clearly distinguishing between the observed and hypothesized interactions.

Answer:

This is an important point, and we have now changed the wording of Figure 4 legend to ensure that it is clear that this is a proposed model:

“Figure 4. Proposed model of metabolic crosstalk between BrCa cells and lung fibroblasts. This schematic represents the proposed metabolic interactions between MDA-MB-231 cells and MRC-5 lung fibroblasts in co-culture. (A) TNBC cells release signalling molecules that activate fibroblasts into CAFs, promoting metabolic crosstalk that supports cancer progression. (B) Vitamin D potentially enhances glutamine uptake enzymes while reducing TCA cycle enzymes in cancer cells and suppresses lactate metabolism and HIF-1α expression in fibroblasts. (C) Vitamin E potentially increases glutamine metabolism and OXPHOS dependence in cancer cells while potentially promoting glucose uptake and lactate export in fibroblasts within the co-culture system.”

 

Comment 10:

This manuscript presents a promising pilot study with novel observations on vitamin-mediated modulation of stromal-tumor metabolic crosstalk. However, the conclusions appear to be overstated in relation to the evidence. The primary concern is the discrepancy between protein expression data and claims of functional metabolic reprogramming. I recommend major revisions. Consideration for publication should occur only after the authors address the essential requirements for functional metabolic assays (Seahorse/metabolomics) and establish a connection between their findings and relevant cancer cell phenotypes. Providing mechanistic data on the effect of Vitamin E and validating key findings in an additional cell line model would substantially enhance the manuscript.

Answer:

We thank the reviewer for these important points and acknowledge the preliminary nature of this pilot study. While it is beyond the scope of this project to perform the further validation and functional studies suggested, we wholeheartedly agree that these are important experiments that do need to be performed in future studies to validate our preliminary findings. We agree that some of the conclusions and metabolic implications were overstated and have thus extensively revised the manuscript to ensure that our interpretations are based on the assessed metabolic enzyme expression levels. While preliminary, we do feel that this study is important to publish to inform the field and provide a justification for a further in-depth analysis into the metabolic influence of Vitamins D and E within the TNBC metastatic microenvironment, particularly as this is very much an underexplored area of research that might provide critical evidence for the potential use of these vitamins in patients.

Reviewer 3 Report

Comments and Suggestions for Authors

The Authors investigated the metabolic interaction between MDA-MB-231 TNBC cells and MRC-5 lung fibroblasts in a co-culture system to elucidate the interactions between these cells. They also examined the effects of vitamin D and vitamin E on these interactions. Using Western blot analysis, they were able to demonstrate that co-culture induced the transformation of fibroblasts into cancer-associated fibroblasts (CAFs), which was confirmed by increased levels of α-SMA and FAP proteins. Interesting observations in the presented study include the demonstration of increased oxidative phosphorylation (OXPHOS) and glutamine metabolism in TNBC cells in the co-culture, and the demonstration that fibroblasts relied on glycolysis and lactate metabolism. Vitamin D also demonstrated inhibition of lactate metabolism and decreased HIF-1α protein levels, but promoted TCA cycle activity in cancer cells. The Authors suggest that this may lead to disruption of the oncogenic metabolic interaction. In the case of vitamin E, a promotion of TCA cycle activity and oxidative metabolism was noticed in BrCa cells, but no significant effect on fibroblast glycolysis was observed. The Authors suggest that this may enhance metabolic interactions between these cells and create a microenvironment conducive to tumor development. The results of the presented study are interesting because they provide new information on the metabolic dynamics of TNBC metastases in the lung TME and demonstrate that vitamins D and E influence the metabolic interaction between tumor cells and fibroblasts. The results of the presented study may support possible new directions for vitamin D and E supplementation in patients with metastatic TNBC.

Furthermore, the Authors emphasize that this study is the first proving that vitamin D (VD) suppresses glycolysis in co-cultured CAFs, what might suggest a new mechanism of action of vitamin D in the process of modulating the TME.

 Vitamin D appears to inhibit the oncogenic metabolic interaction between BrCa and CAF cells, which may promote tumor suppression. Vitamin E, on the other hand, did not demonstrate such an interaction, suggesting an opposing effect.

In my opinion, the work is carefully prepared, provides a good introduction to the process of cellular metabolic reprogramming, and the methodological approach is engaging, beginning with cell co-culture. The results of the Western blot analysis are very broad and have been subjected to quantitative analysis, and the obtained results are presented in the form of compensated figures containing diagrams that clearly illustrate the phenomena described.

The results were carefully discussed and related to previously published data. The Authors also highlighted limitations of the presented work and directions for further research. References were carefully selected.

After reading the paper, some minor editorial comments come to mind:

Although the abbreviations used in the paper are carefully explained in the text, it seems that adding the collected abbreviations could improve the reading experience.

On page 13, publication number 42 is cited, followed by publication number 46 – publications numbered 43-45 are cited later (after reference 46).

On page 15, in the second paragraph, the Authors cite: "Interestingly, Anna et al., demonstrated a negative correlation between VDR ...." Anna is the author's first name; the citation should include her last name – Brożyna.

 

Author Response

We thank the reviewer for their positive comments regarding our manuscript. We have now addressed all the errors that were identified:

 Comment 1:

Although the abbreviations used in the paper are carefully explained in the text, it seems that adding the collected abbreviations could improve the reading experience.

Answer: We thank the reviewer for this helpful suggestion. Abbreviations used in the manuscript are defined at first mention in the text, and the journal’s formatting system automatically provides a consolidated list of abbreviations. We have ensured that all abbreviations are clearly and consistently defined to maintain readability.

 

Comment 2:

On page 13, publication number 42 is cited, followed by publication number 46 – publications numbered 43-45 are cited later (after reference 46).

Answer: We thank the reviewer for pointing this out. The reference numbering has been corrected to ensure that citations now appear in the proper sequential order throughout the manuscript.

 

Comment 2:

On page 15, in the second paragraph, the Authors cite: "Interestingly, Anna et al., demonstrated a negative correlation between VDR ...." Anna is the author's first name; the citation should include her last name – Brożyna.

Answer: We thank the reviewer for noting this error. The citation has been corrected to replace the author’s first name with the appropriate surname (Brożyna) in the revised manuscript (Page 14; Line 434).

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The manuscript is now significantly improved, and I believe it meets the standards for publication. I have no further concerns, and I recommend the paper for acceptance in its current form.

   

Reviewer 2 Report

Comments and Suggestions for Authors

The authors have worked on the said comments and I think there some potential missing but they clearify that it should be considered in future studies, so I would go with the recommendation.