Fungal–Bacterial Crosstalk Modulates Glucocorticoid-Primed TLR2 Signaling in the Human Skin

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Major comments

1. The premise that the yeast M. restricta suppresses activation by C. acnes occurs through changes in TLR2 expression is interesting. However, both C. acnes and M. restricta could be activating the cells directly through TLR2 and all of these effects could simply be explained as receptor co-ligation. This could be answered with an "M. restricta-only" control in all of these experiments. It must be included if any meaningful conclusions are to be reached. The "Dex alone" control must also be presented throughout the manuscript.
2. It is very difficult to follow the rationale as presented in the introduction. If Dex causes acne erruptions in patients, then it makes sense that it may be increasing TLR2 expression -- thereby making cells more susceptible to activation by C. acnes. But how does decreasing TLR2 expression by M. restricta make this situation worse? It was difficult to follow the logic and the reasoning behind the analysis of the effect of M. restricta.
3. The role of MKP-1 is important in the processes that are examined in this manuscript. It should be added to the schematic in Fig. 1. 
4. Sentences regarding attribution of schematic in lines 62-63 can be moved to the figure legend.
5. In Fig 1, the authors show the changes in TLR2 expression after exposure to the fungus and bacteria. On line 158 they state "However, co-culture with M. restricta significantly supressed this Dex-enchanced induction, with TLR2 expression remaining approximately 50-fold relative to the control, which was significantly lower than that of C. acnes stimulation alone (p <0.01)". However, according to panel A, the Dex+M.restricta response is not lower than the C.acnes response alone, so this sentences is either incorrect or requires re-wording.
6. The use of ELISA to measure TLR2 is unusual for this type of study. This is mainly because the membrane-associated, extracellular TLR2 is necessary for signaling and response to the bacterium. As such, membrane associated TLR2 expressed on the outer surface of the cells would be a better measure of their capability of responding to C. acnes. Flow cytometry or even membrane-fraction western blot would be a more appropriate approach.
7. The immunofluoresence images in Fig. 3 are difficult to interpret. There should be a control label for the nucleus (ex: DAPI) to differentiate cells. Also, the scale bars are too small to read. Higher magnification is necessary to fully visualize the distinct cells. Presumably, these keratinocytes express some constitutive TLR2 protein -- why is it not visible in panel A? This is missing the Dex alone and M. restricta alone control -- and the isotype control images should be presented as a supplementary figure.
8. In Fig. 4, the data suggests that M. restricta promotes the phosphorylation of p38, without affecting expression, which would presumably inhibit TLR2 expression. At the same time, M. restricta is inhibiting Dex+C.acnes-induced NF-kB translocation to the nucleus. Does this occur when the cells are only stimulated with C. acnes? In other words, there should be a C. acnes+M. restricta control to determine whether the presence of Dex is necessary for this interaction.
9. Figure 5 is again missing key controls (as mentioned above) but also the effect of the drugs alone should be tested. The concentrations of the drugs should be stated in the figure legends and the rationale for choosing these concentrations should be explained in the Materials and Methods or a supplementary data showing some concentration curves.
10. All of the experiments in this manuscrip were performed with 100 nM of Dex, 50 MOI of C. acnes and 10 MOI of M. restricta. Why were these concentrations and doses chosen? Were these conditions optimized and how?
11. Dexamethasone was added to the cultures 20 min prior to infection with the bacteria and fungi. Why was this time point chosen? If the hypothesis is that Dex increases TLR2 expression on its own, this is insufficient time to allow for changes in TLR2 protein expression. 
12. M. restricta on its own has been shown to upregulate TLR2 expression in previous studies. How do the authors reconcile their observations with these previous published reports? The discussion does not adequately adress this misalignment. Why would the addition of Dex, which also enhances TLR2 expression, cause this drop in TLR2 expression that they observed in Fig. 2 and Fig. 3?
13. The Author contributions section indicates that funding was obtained but the funding sections states that "This research recieved no external funding."

Author Response

To Reviewer 1

Thank you very much for your valuable suggestions and comments on our manuscript. We have carefully addressed each comment point by point, and all revisions have been clearly highlighted in the revised manuscript.

Q1 The premise that the yeast M. restricta suppresses activation by C. acnes occurs through changes in TLR2 expression is interesting. However, both C. acnes and M. restricta could be activating the cells directly through TLR2 and all of these effects could simply be explained as receptor co-ligation. This could be answered with an "M. restricta-only" control in all of these experiments. It must be included if any meaningful conclusions are to be reached. The "Dex alone" control must also be presented throughout the manuscript.

 

Response

We thank the Reviewer for this important comment regarding the possibility that the observed effects could be explained by simple receptor co-ligation through TLR2.

To directly address this concern, we have now incorporated additional controls performed in parallel with the original experiments but not included in the initial submission.

First, Dex alone and M. restricta alone conditions are now included as controls in Figures 2 and 4. These data demonstrate that M. restricta alone does not induce TLR2 expression while Dex alone does.

Second, we have added Figure S2, demonstrating how Dex alone affects TLR2 expression as well as the lack of TLR2 induction by M. restricta alone across the tested concentrations.

Third, to rule out the possibility that the observed suppression was due to simple receptor co-ligation, we performed C. acnes and M. restricta co-culture experiments in the absence of Dex. As presented in Figure S3, we observed no suppressive or synergistic effects on TLR2 expression under these conditions, indicating that the suppressive effect of M. restricta is Dex-dependent and cannot be explained by direct TLR2 co-ligation by microbial ligands.

In addition, to determine whether TLR2 expression modulation directly translates into altered inflammatory output, we have analyzed proinflammatory cytokine expressions under the same experimental conditions, revealing that dexamethasone markedly suppressed C. acnes–induced IL-6, IL-8, and TNF-α expression, and co-culture with M. restricta did not further enhance this suppressive effect (Figure S4). These results indicate that M. restricta-induced TLR2 expression changes under glucocorticoid exposure do not directly correlate with further suppression of inflammatory cytokine production, supporting the conclusion that M. restricta modulates glucocorticoid-primed signaling context rather than simply attenuating inflammatory output. Based on these additional data, we conclude that the suppressive effect of M. restricta on C. acnes-induced TLR2 expression could unlikely be explained by simple receptor co-ligation, and it reflects Dex-dependent host signaling pathway modulations instead.

 

We have added this description in Lines 207–2103.

 

Q2 It is very difficult to follow the rationale as presented in the introduction. If Dex causes acne erruptions in patients, then it makes sense that it may be increasing TLR2 expression -- thereby making cells more susceptible to activation by C. acnes. But how does decreasing TLR2 expression by M. restricta make this situation worse? It was difficult to follow the logic and the reasoning behind the analysis of the effect of M. restricta.

 

Response

We thank the Reviewer for pointing out the lack of clarity in the rationale presented in the Introduction. We agree that, in its original form, the logic linking glucocorticoid-induced TLR2 upregulation to the effect analysis of Malassezia restricta was not sufficiently explicit.

Therefore, we have revised the Introduction to more clearly articulate our conceptual framework. Specifically, we have now emphasized that glucocorticoids induce a TLR2-sensitized state in keratinocytes, which enhances C. acnes responsiveness and contributes to steroid-induced acne. However, keratinocytes are exposed to C. acnes in the context of a complex follicular microbiome that includes commensal fungi, and inflammatory signaling is shaped by microbial crosstalk rather than by bacterial stimuli alone.

Importantly, our study does not assume that reduced TLR2 expression necessarily leads to disease attenuation. Instead, we propose that M. restricta actively modulates glucocorticoid-dependent host signaling, thereby altering the balance between inflammatory priming and downstream signal execution. In this context, changes in TLR2 expression reflect a broader rewiring of the signaling pathways, as supported by our findings on p38 MAPK activation and the inhibition of NF-κB nuclear translocation. Importantly, our aim was not to equate reduced TLR2 expression with disease suppression, but to determine whether commensal fungi reshape glucocorticoid-dependent inflammatory signaling in keratinocytes exposed to a polymicrobial follicular environment.

To address this point, we have revised the Introduction to explicitly state why M. restricta was examined and how its suppressive effect on TLR2 expression is mechanistically and biologically relevant under glucocorticoid exposure. With these revisions, we aim to clarify the rationale of the study and align the Introduction with the experimental design and results.

 

We have added this description in Lines 58-65, 205-207, 294-302.

 

Q3 The role of MKP-1 is important in the processes that are examined in this manuscript. It should be added to the schematic in Fig. 1. 

 

Response

We agree with the Reviewer that MAP kinase phosphatase-1 (MKP-1) is important for the regulatory processes examined in this study. In response to this comment, we have revised Figure 1 to explicitly include MKP-1 in the schematic representation of dexamethasone-mediated TLR2 signaling regulation. Specifically, MKP-1 is now depicted as a glucocorticoid-inducible negative regulator of the p38 and JNK MAPK pathways, providing mechanistic context for the glucocorticoid-dependent modulation of TLR2 expression discussed in this manuscript.

 

We have added this description in Lines 66-68, 77-81.

 

Q4 Sentences regarding attribution of schematic in lines 62-63 can be moved to the figure legend.

 

Response

As suggested by the Reviewer, the sentence originally included in lines 62 and 63 (“The diagram shows the principal components of the MEK–ERK, MKK3/6–p38, and MKK4/7–JNK MAPK pathways, as well as the NF-κB signaling pathway.”) has been moved from the main text to the legend of Figure 1.

 

 

Q5 In Fig 1, the authors show the changes in TLR2 expression after exposure to the fungus and bacteria. On line 158 they state "However, co-culture with M. restricta significantly supressed this Dex-enchanced induction, with TLR2 expression remaining approximately 50-fold relative to the control, which was significantly lower than that of C. acnes stimulation alone (p <0.01)". However, according to panel A, the Dex+M.restricta response is not lower than the C. acnes response alone, so this sentences is either incorrect or requires re-wording.

 

Response

We thank the Reviewer for bringing our attention to this inconsistency. We agree that the original wording in Section 3.1 was misleading when compared directly with the data shown in Figure 2.

We have thus revised the text to clarify as follows.

 

We have added this description in Lines 176-177.

 

Q6 The use of ELISA to measure TLR2 is unusual for this type of study. This is mainly because the membrane-associated, extracellular TLR2 is necessary for signaling and response to the bacterium. As such, membrane associated TLR2 expressed on the outer surface of the cells would be a better measure of their capability of responding to C. acnes. Flow cytometry or even membrane-fraction western blot would be a more appropriate approach.

 

Response

We agree with the Reviewer that the functionally relevant TLR2 form for bacterial recognition and signaling is the membrane-associated, extracellular version, and that cell-surface TLR2 assessment is a more appropriate measure of keratinocyte responsiveness to C. acnes.

In fact, extracellular, membrane-localized TLR2 expression has already been evaluated in our study by immunofluorescent staining using an extracellular TLR2 domain-recognizing antibody (Figure 3). This analysis directly visualizes TLR2 on the outer surface of keratinocytes, thereby addressing the functional concern raised by the Reviewer.

In contrast, the ELISA-based quantification presented in Figure 2B reflects total cellular TLR2 protein levels and does not distinguish between membrane-associated and intracellular pools. Given the valid point of the Reviewer concerning functional relevance, we have removed Figure 2B and the corresponding description from the manuscript. Moreover, we have revised the Results section accordingly to focus on cell-surface TLR2 expression and downstream signaling events. We believe that this revision improves the methodological rigor of the study and better aligns the data with the biological interpretation of TLR2-mediated responses to C. acnes.

 

We have added this description in Lines 239-240.

 

Q7 The immunofluoresence images in Fig. 3 are difficult to interpret. There should be a control label for the nucleus (ex: DAPI) to differentiate cells. Also, the scale bars are too small to read. Higher magnification is necessary to fully visualize the distinct cells. Presumably, these keratinocytes express some constitutive TLR2 protein -- why is it not visible in panel A? This is missing the Dex alone and M. restricta alone control -- and the isotype control images should be presented as a supplementary figure.

 

Response

We thank the Reviewer for the constructive suggestions related to the interpretation and presentation of the immunofluorescent images in Figure 3. In accordance with these comments, we have revised the figure and the corresponding description as follows.

1. We have included an restricta-alone condition in Figure 3 to clearly distinguish its independent effect on TLR2 expression from that observed in the co-culture experiments.
2. We have incorporated nuclear counterstaining with DAPI in all panels to allow for the clear identification of individual keratinocytes and improve cellular context.
3. We have enlarged the scale bars and standardized them to 50 μm for readability.
4. We have included the images at higher magnification for the clearer visualization of distinct cells.

In addition, we have provided isotype control staining in Figure S1. Under the C. acnes + Dex condition, we have observed strong green fluorescence when stained with the anti-TLR2 antibody, whereas replacement with the corresponding IgG1 isotype control yielded minimal background fluorescence, confirming the specificity of the TLR2 immunofluorescent signal.

We believe that these revisions have substantially improved the clarity, interpretability, and rigor of Figure 3.

 

We have added this description in Lines 150-151, 193-197, and 198-203.

 

Q8 In Fig. 4, the data suggests that M. restricta promotes the phosphorylation of p38, without affecting expression, which would presumably inhibit TLR2 expression. At the same time, M. restricta is inhibiting Dex+C.acnes-induced NF-kB translocation to the nucleus. Does this occur when the cells are only stimulated with C. acnes? In other words, there should be a C. acnes+M. restricta control to determine whether the presence of Dex is necessary for this interaction.

 

Response

We thank the Reviewer for this important and insightful comment concerning the interpretation of p38 phosphorylation and NF-κB nuclear translocation in Figure 4 as well as for highlighting the need to determine whether Dex would be required for the observed fungal–bacterial interaction.

Our original intention with Figure 4 was to examine how M. restricta modulates C. acnes–induced signaling under glucocorticoid exposure. However, we agree that the clarification of the related effects in the absence of Dex is essential.

To address this point, we have included the C. acnes + M. restricta condition (in the absence of Dex) in Figure 4A and B. With these additions, the data demonstrate that M. restricta enhances p38 phosphorylation and suppresses NF-κB nuclear translocation not only upon Dex + C. acnes stimulation but also upon that of C. acnes alone. We have explicitly described this clarification in the Results section.

Furthermore, we have expanded the Discussion to distinguish Dex-dependent and independent effects. While M. restrictamodulates p38 and NF-κB signaling irrespective of glucocorticoid exposure, the suppression of TLR2 expression occurs only in a Dex-primed signaling context, indicating that M. restricta does not simply interfere with C. acnes-induced signaling per se, but rather selectively counteracts glucocorticoid-dependent transcriptional TLR2 regulation.

 

We have added this description in Lines 225-228, and 233-235.

 

Q9 Figure 5 is again missing key controls (as mentioned above) but also the effect of the drugs alone should be tested. The concentrations of the drugs should be stated in the figure legends and the rationale for choosing these concentrations should be explained in the Materials and Methods or a supplementary data showing some concentration curves.

 

Response

We thank the Reviewer for the valuable comment related to incorporating appropriate controls, clarifying drug concentrations, and justifying the selected doses in Figure 5. To address these concerns, we have included Figures S5 and S6 in the revised manuscript.

First, to clarify how the inhibitors and activators affected TLR2 expression, we have explicitly stated the concentrations of all compounds used in Figure 5 in the revised figure legends. In addition, Supplementary Figure S5 presents the concentration-dependent effects of the NF-κB inhibitors MG-132 and celastrol (0.5 μM and 1 μM), demonstrating that both compounds dose-dependently suppress TLR2 gene expression.

Second, as suggested by the Reviewer, we have evaluated the effects of each compound when applied alone, without microbial stimulation. Figure S6 present these data, demonstrating that none of the inhibitors or activators alone induced TLR2 gene expression, confirming that the observed effects in Figure 5 reflect signaling pathway modulations rather than nonspecific drug-induced TLR2 activation.

Concerning the rationale for concentration selection, we have selected the doses used (0.5–1 μM) based on previously published studies and manufacturer recommendations, reporting effective NF-κB and MAPK signaling pathway modulation in keratinocytes without causing cytotoxicity. Moreover, these concentrations are widely used in similar signaling studies and thus allow for direct comparison with previously published results.

Taken together, the inclusion of Figures S5 and S6, explicit description of drug concentrations, and clarification of the rationale behind the dose selection strengthen the interpretation of Figure 5 and address the concerns of the Reviewer related to experimental controls and data transparency.

 

We have added this description in Lines 125-128, 246-250, 255-257.

 

 

Q10 All of the experiments in this manuscrip were performed with 100 nM of Dex, 50 MOI of C. acnes and 10 MOI of M. restricta. Why were these concentrations and doses chosen? Were these conditions optimized and how?

 

Response

We thank the Reviewer for this important question.

First, we apologize for the typographical error in the manuscript. We incorrectly described Dex concentration as 100 nM. Correctly, the concentration we used in all experiments was 1 μM. We have rectified this error throughout the revised manuscript, including the Materials and Methods, Results, figure legends, and Supplementary figures.

Dex concentration selection:
We have determined Dex concentration based on preliminary dose–response experiments. In the presence of C. acnes (MOI = 10), Dex enhanced TLR2 expression both at 0.1 and 1 μM, whereas a higher concentration (10 μM) resulted in a reduced effect. Based on these results and in agreement with those of previous studies (References 15,16) reporting synergistic effects of glucocorticoids with inflammatory stimuli at 1 μM, we have selected 1 μM Dex for our subsequent experiments.

50. acnes MOI selection:
    Previous studies investigating keratinocyte responses to C. acnes commonly used MOIs ranging from 10 to 50. In our preliminary experiments (MOI = 3–200), IL-6 expression increased dose-dependently up to MOI 50 and plateaued beyond. LDH assays indicated increased cytotoxicity at MOI >100. Therefore, we have selected MOI = 50 as the optimal condition to induce inflammation without causing cytotoxicity.
51. restricta MOI selection:
    Most previous keratinocyte–Malassezia co-culture studies have used MOIs between 10 and 20. In our preliminary experiments, M. restricta induced cytotoxicity at MOI ≥20, whereas MOI ≤ 5 minimally affected TLR2 suppression and p38 activation. Based on our results, we have selected MOI = 10 as the optimal non-cytotoxic condition with sufficient biological activity. Taken together, we have selected all experimental conditions based on combined preliminary optimization experiments and consistency with previous studies [18–24].

 

We have added this description in Lines 112, 116-119, 173-175, and 181

 

Q11 Dexamethasone was added to the cultures 20 min prior to infection with the bacteria and fungi. Why was this time point chosen? If the hypothesis is that Dex increases TLR2 expression on its own, this is insufficient time to allow for changes in TLR2 protein expression. 

 

Response

We thank the Reviewer for this important comment about Dex pretreatment timing.

We added Dex 20 min prior to microbial stimulation to allow for sufficient cellular uptake and glucocorticoid receptor (GR)-dependent signaling activation. Importantly, in this study, we did not aim to assess whether Dex alone would induce TLR2 protein expression, but rather to investigate how keratinocyte responses to C. acnes and Malassezia are modulated under an active glucocorticoid signaling context.

Therefore, we mostly designed the 20-min pretreatment to establish a GR-primed state at the time of microbial stimulation, rather than to induce de novo TLR2 protein expression by Dex alone, which would have required longer incubation.

In addition, we added all pharmacological inhibitors 1 h before stimulation and applied Dex after inhibitor pretreatment to evaluate Dex-dependent signaling pathway modulations under inhibitor-controlled conditions. Within this experimental framework, we considered a 20-min Dex pretreatment optimal.

Furthermore, previous studies described that a short (i.e., approximately 20 min) Dex pretreatment is sufficient to potentiate C. acnes-induced TLR2 responses, particularly at the transcriptional and signaling levels. In the present study, we adopted this established condition in order to introduce Malassezia as an additional variable and allow for direct comparison with the results of prior studies.

Consistent with these studies, we have confirmed that a 20-min Dex pretreatment was sufficient to modulate C. acnes-induced TLR2 expression at the mRNA and signaling levels in our experimental system.

 

We have added this description in Lines 112-114.

 

 

Q12 M. restricta on its own has been shown to upregulate TLR2 expression in previous studies. How do the authors reconcile their observations with these previous published reports? The discussion does not adequately adress this misalignment. Why would the addition of Dex, which also enhances TLR2 expression, cause this drop in TLR2 expression that they observed in Fig. 2 and Fig. 3?

 

Response

We thank the Reviewer for raising this important point regarding the apparent discrepancy between our findings and those of previous studies describing TLR2 upregulation by Malassezia species.

Previous studies reporting Malassezia-induced TLR2 expression remain relatively limited and have primarily focused on M. furfur, often at the mRNA level. In line with these reports, we have observed that M. restricta alone induced only a modest increase in TLR2 mRNA expression in NHEKs (Figure S2). Even at MOI = 20, the induction magnitude was limited (~2–3-fold) and substantially weaker than that induced by C. acnes or Dex. Therefore, our results do not contradict previous reports but indicate that M. restricta is a relatively weak TLR2 inducer under our experimental conditions.

We have revised the Results and Discussion sections to clarify this glucocorticoid-dependent regulatory role of M. restricta and explicitly address how it relates to previous reports.

 

We have added this description in Lines 207–210, 286-290

 

Q13 The Author contributions section indicates that funding was obtained but the funding sections states that "This research recieved no external funding."

 

Response

We thank the Reviewer for noticing this inconsistency.

This study did not receive any external funding from agencies or organizations outside the affiliated institution of the authors. Therefore, the statement “This research received no external funding” is correct and remains unchanged in the Funding section.

The term “funding acquisition” in the Author Contributions section refers to internal institutional research funds that supported this study, rather than to external grants. To avoid any ambiguity, we have removed the term “funding acquisition” from the Author Contributions section and revised it to reflect project administration supported by institutional resources.

 

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

This study developed a co-culture model of C. acnes and Malassezia restricta in the presence of dexamethasone to investigate the impact of microbial interactions on TLR2 expression and its downstream signaling pathways. The findings contribute to a deeper understanding of the role of skin commensal interactions in the pathogenesis of steroid-induced acne. However, several minor revisions are necessary for the manuscript to align with publication standards:

（1）It is recommended that the copyright statement for Figure 1, currently located on lines 62–63, be moved to the legend of Figure 1.

（2）The primer sequences listed on lines 113–116 should be incorporated into the preceding text as a single cohesive paragraph.

（3）It is suggested that Figure 3 include bright-field images to illustrate consistent cell density and condition across different groups.

（4）Figure 4 should be annotated to indicate the molecular weight of each band.

（5）In Figure 5, the significance of the asterisk marking on the "SP600125" bar graph is unclear and requires clarification.

Author Response

To Reviewer 2

Thank you very much for your valuable suggestions and comments on our manuscript. We have carefully addressed each comment point by point, and all revisions have been clearly highlighted in the revised manuscript.

Q1 It is recommended that the copyright statement for Figure 1, currently located on lines 62–63, be moved to the legend of Figure 1.

 

Response

We thank the Reviewer for this suggestion. We have moved the copyright statement previously located in lines 62 and 63 to the legends of Figure 1.

 

Q2 The primer sequences listed on lines 113–116 should be incorporated into the preceding text as a single cohesive paragraph.

 

Response

We thank the Reviewer for this helpful suggestion.

We have revised the Methods section (lines 113–116) to incorporate the primer sequences into the preceding text as a single cohesive paragraph, rather than listing them as separate lines. This change improves readability and formatting consistency.

 

We have added this description in Lines 134-137.

 

Q3 It is suggested that Figure 3 include bright-field images to illustrate consistent cell density and condition across different groups.

 

Response

We thank the Reviewer for this helpful suggestion.

Due to technical limitations in obtaining high-quality bright-field images under the fluorescence imaging conditions used, we have evaluated cell density and cellular condition instead, using DAPI nuclear staining. DAPI staining allows for a reliable cell number and distribution assessment (Figure 3) demonstrating comparable nuclear density across all experimental groups.

Therefore, the observed differences in TLR2 fluorescent intensities are not attributable to variations in cell density or general cellular condition.

 

We have added this description in Lines 188-189.

 

Q4 Figure 4 should be annotated to indicate the molecular weight of each band.

 

Response

We thank the Reviewer for this suggestion. We have revised Figure 4 to include annotations indicating the molecular weight of each band.

 

Q5 In Figure 5, the significance of the asterisk marking on the "SP600125" bar graph is unclear and requires clarification.

 

Response

We thank the Reviewer for bringing our attention to this lack of clarity.

In Figure 5, the asterisk (*) indicates statistically significant differences (p < 0.05) between the Dex + C. acnes- and Dex + C. acnes-treated groups in the presence of SP600125. We have now clearly specified these details in the revised figure legends.

 

 

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

The manuscript addresses a relevant subject, the interplay between bacteria, fungi, and epithelial cells in the activation of a proinflammatory response at the skin level. The positive aspect of the study is the use of primary human cells. The main shortcoming is the simplistic mechanism proposed. It is hard to think of a proinflammatory response involving a sole PRR, in this case, TLR2. This receptor is known to collaborate with other TLRs; the heterodimers TLR1/TLR2 and TLR2/TLR6 are well known. So, the mechanism proposed by the authors is likely incomplete. In addition, bacteria and fungal cells have different PAMPS  that will participate while coincubated with the human cells. So, the author should explore the participation of other PRRs besides TLR2. Another main shortcoming is related to the need for an active bacterial/fungal metabolism to trigger this response. Experiments using inactivated bacteria and yeast-like cells will address this aspect.

Other technical issues:

Why did the authors use different MOI for bacteria and fungal cells?

The experimental setting requires a time-course and dose-dependent experimentation.

In the qPCR assays, please include the primer efficiency of each primer pair and add a declaration about the number of peaks obtained in the melting curves.

The Dunnett’s multiple comparison test is a post-ANOVA test. Please add details about the ANOVA test used. How did the authors analyze data normality?

Please provide unedited and uncropped images of the western blots. This could be as supplementary material. Figure 4 would benefit from a densitometric analysis. 

Author Response

To Reviewer 3

Thank you very much for your valuable suggestions and comments on our manuscript. We have carefully addressed each comment point by point, and all revisions have been clearly highlighted in the revised manuscript.

Q1  The manuscript addresses a relevant subject, the interplay between bacteria, fungi, and epithelial cells in the activation of a proinflammatory response at the skin level. The positive aspect of the study is the use of primary human cells. The main shortcoming is the simplistic mechanism proposed. It is hard to think of a proinflammatory response involving a sole PRR, in this case, TLR2. This receptor is known to collaborate with other TLRs; the heterodimers TLR1/TLR2 and TLR2/TLR6 are well known. So, the mechanism proposed by the authors is likely incomplete. In addition, bacteria and fungal cells have different PAMPS  that will participate while coincubated with the human cells. So, the author should explore the participation of other PRRs besides TLR2. Another main shortcoming is related to the need for an active bacterial/fungal metabolism to trigger this response. Experiments using inactivated bacteria and yeast-like cells will address this aspect.

 

Response

We thank the Reviewer for this thoughtful and constructive critique.

We fully agree with the Reviewer that a proinflammatory response in epithelial cells would unlikely be mediated by a single pattern recognition receptor (PRR) alone. Indeed, TLR2 is well known to function in cooperation with other Toll-like receptors, particularly through heterodimer formation with TLR1 or TLR6, and innate immune responses are generally shaped by the integration of multiple PRR signals.

However, in the present study, our analysis intentionally focused on TLR2, as our primary objective was to determine how Malassezia modulates TLR2 expression and downstream signaling in the context of C. acnes stimulation under glucocorticoid exposure. While we recognize that other PRRs, including TLR1 and TLR6, likely contribute to the overall inflammatory response, their detailed involvement lies beyond the scope of the current study and represents an important direction for future studies.

Consistent with previous reports, the Dex- and C. acnes-induced synergistic inflammatory response involved TLR2-containing receptor complexes, including the TLR2/TLR1 heterodimer. In our experimental system, pharmacological TLR2 inhibition markedly attenuated downstream signaling events, supporting the role of TLR2 as a key regulatory node in this context. As our main goal was to elucidate how Malassezia suppresses the C. acnes + Dex-induced inflammatory response, we thus focused on TLR2-centered signaling mechanisms.

In addition, to address the concern of the Reviewer regarding the requirement for active microbial metabolism, we have performed stimulation experiments using heat-inactivated Malassezia as well as Transwell-based non-contact co-culture assays (Figure S7). Under both conditions, the suppressive effect of Malassezia on TLR2 expression and downstream signaling was lost, indicating that the modulatory effect of Malassezia requires viable fungal cells and is not mediated solely by static pathogen-associated molecular patterns (PAMPs), suggesting an essential role for fungal metabolic activity and direct cell–cell interactions.

In summary, these results support the conclusion that while TLR2 is not the sole PRR involved in skin inflammatory responses, it represents a critical regulatory node in the C. acnes + Dex context, and that Malassezia modulates this pathway through fungal viability-dependent mechanisms. We have clarified these points in the Discussion to better position our findings within the broader framework of PRR-mediated host–microbe interactions.

 

We have added this description in Lines 281-286, and 290-293

 

Q2 Why did the authors use different MOI for bacteria and fungal cells?

 

Response

We thank the Reviewer for this important question.

The use of different MOIs for C. acnes and M. restricta reflects fundamental biological differences between bacterial and fungal cells as well as those in their interactions with keratinocytes. MOI values are not directly comparable across microbial species, particularly between bacteria and yeasts, as they differ substantially in cell size, surface composition, growth characteristics, and the magnitude of the host cell stimulation they induce. For C. acnes, previous studies examining keratinocyte inflammatory responses have commonly used MOIs ranging from 10 to 50. In our preliminary experiments, C. acnes induced IL-6 expression dose-dependently up to MOI 50, while higher MOIs yielded a plateau of inflammatory responses and increased cytotoxicity. Therefore, we have selected MOI 50 as an optimal condition to robustly induce inflammation without compromising cell viability.

In contrast, Malassezia species are substantially larger than bacteria and possess a lipid-rich cell wall that could strongly affect host cells even at lower numerical ratios. Consistent with previous keratinocyte–Malassezia co-culture studies, we found that M. restricta induced cytotoxicity at MOIs ≥ 20, whereas MOI ≤ 5 elicited minimal biological effects. Based on these results, we have selected MOI 10 as the highest non-cytotoxic condition that retained sufficient biological activity to modulate host signaling.

Therefore, we have determined the different MOIs used for bacterial and fungal cells independently for each organism based on preliminary optimization experiments and prior literature, with the aim of achieving biologically relevant stimulation without inducing cytotoxicity. This approach allows for the meaningful comparison of host responses under physiologically appropriate conditions for each microorganism.

 

 

Q3 The experimental setting requires a time-course and dose-dependent experimentation.

 

Response

We thank the Reviewer for this comment concerninh the need for time-course and dose-dependent experimentation.

In this study, we performed both time-course and dose-dependent analyses during the experimental optimization phase. For the time-course analysis, we examined TLR2 mRNA expression at 4 and 6 h after stimulation. We revealed that the C. acnes-induced TLR2 expression was the most pronounced at 6 h, we thus selected this time point for all subsequent experiments (Figure S3).

Moreover, we conducted dose-dependent experiments for each stimulus. For Dex, we tested concentrations of 0.1, 1, and 10 μM, and observed that 1 μM consistently induced the most robust TLR2 expression upregulation. Accordingly, we selected this concentration for further analyses.

For C. acnes, we evaluated a wide range (3–200) of MOIs. TLR2 and IL-6 expression increased dose-dependently up to MOI 50, whereas higher MOIs resulted in plateaued responses and increased cytotoxicity. In addition, we confirmed synergistic effects between Dex and C. acnes using three independent C. acnes strains.

For M. restricta, dose-dependent effects were assessed by analyzing p38 MAPK activation at MOIs of 5, 10, and 15, revealing a clear MOI-dependent increase in p38 phosphorylation without inducing cytotoxicity.

Taken together, these time-course and dose-dependent optimization experiments ensured the biological relevance, non-cytotoxicity, and suitability of the experimental conditions of this study for robust analysis of the host inflammatory responses.

 

Q4 In the qPCR assays, please include the primer efficiency of each primer pair and add a declaration about the number of peaks obtained in the melting curves.

 

Response

We thank the Reviewer for this important technical comment.

In the present study, we confirmed primer specificity by melting curve analysis for all qPCR assays. Each primer pair generated a single, well-defined melting peak, indicating specific amplification without detectable non-specific products or primer-dimers. We have included this information into the revised Methods section.

Regarding amplification efficiency, we adopted the primer sets used in this study from previously published reports, in which we validated their efficiencies. In addition, we confirmed that all primer pairs showed comparable amplification performance under our experimental conditions, allowing for reliable relative quantification using the ΔΔCt method. We have incorporated a clarification regarding primer efficiency to the Methods section.

 

We have added this description in Lines 138-139.

 

Q5 The Dunnett’s multiple comparison test is a post-ANOVA test. Please add details about the ANOVA test used. How did the authors analyze data normality?

 

Response

We thank the Reviewer for this important comment.

Prior to statistical analysis, we assessed data normality using the Shapiro–Wilk test. As the ΔCt values satisfied the assumption of normal distribution, we performed statistical comparisons using one-way analysis of variance (ANOVA) on ΔCt values. Following ANOVA, we applied Dunnett’s multiple comparison test as a post hoc test for comparisons against the control group.

We have now clarified these details in the Materials and Methods section.

 

We have added this description in Lines 166-169.

 

Q6 Please provide unedited and uncropped images of the western blots. This could be as supplementary material. Figure 4 would benefit from a densitometric analysis. 

 

Response

We thank the Reviewer for this valuable suggestion.

Unedited and uncropped Western blot images have now been provided as Figure S8. In addition, we have included densitometric analyses to Figure 4 (new panels Figure 4C and Fig. 4D) to quantitatively support the Western blot results.

 

We have added this description in Lines 228-229 and 239-240.

 

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report

Comments and Suggestions for Authors

The authors addressed my comments, and the revised manuscript is suitable for publication.