Anti-Inflammatory Effects of Marine-Derived Resorcylic Acid Lactone Derivatives in Ulcerative Colitis via the MAPK/ERK Pathway

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The paper describes the isolation and structure elucidation of three new resorcylic lactone derivatives from Penicillium sp obtained from a Chinese mangrove plant. The results presented are impressive, demonstrating that the anti-UC activity is effective in vitro and in vivo, and that the new compounds regulate specific inflammatory markers. The structure elucidation is sound, as supported by spectroscopic data and an appreciable degree of purity fit for the bioassay. Some comments to be addressed:

1. The fungus was characterized only by ITS. Why did the authors skip multi-gene phylogenetic analysis?
2. Please include details of the fungus in the SI - culture photos - macroscopic and microscopic.
3. Please provide optical purity of the compounds.
4. The elucidation of the planar structures is convincing except for the deduction of the absolute configuration of compound 3, especially that DP4+-aided calculations and ECD TDDFT were used. Based on my impressions on the ECD-TDDFT spectra, the negative Cotton shift centered at ca. 285 nm (expt) is missing in the calculated lines. Did the authors consider energy and conformational factors?
5. The compounds being phenolic may potentially hold antioxidant effects in the inflammatory pathways. Why was NrF2 activation not determined?

Comments on the Quality of English Language

pls check minor errors esp formatting

Author Response

Comment 1:

The fungus was characterized only by ITS. Why did the authors skip multi-gene phylogenetic analysis?

Response 1:

Thank you for your attention to our research methodology. We fully acknowledge that multi-gene phylogenetic analysis is crucial for the precise identification of fungi. The primary objective of this study was to systematically investigate the chemical diversity and bioactive secondary metabolites of this strain. Therefore, we employed the commonly used ITS region for preliminary identification within the field to quickly establish the research framework.

Comment 2:

Please include details of the fungus in the SI - culture photos - macroscopic and microscopic.

Response 2:

Thank you for your valuable suggestion. We have supplemented the Supplementary Information (Figs. S26, 27) with macroscopic and microscopic morphological characteristic photographs of the strain Penicillium sp. HN-20.

Comment 3:

Please provide optical purity of the compounds.

Response 3:

Thank you for your valuable feedback. We have performed HPLC-UV analysis on all isolated compounds to ensure their optical purity, and further confirmed their purity by combining nuclear magnetic resonance (NMR) spectroscopy. The corresponding chromatograms have been included in the Supplementary Information (Figs. S28–S37).

Comment 4:

The elucidation of the planar structures is convincing except for the deduction of the absolute configuration of compound 3, especially that DP4+-aided calculations and ECD TDDFT were used. Based on my impressions on the ECD-TDDFT spectra, the negative Cotton shift centered at ca. 285 nm (expt) is missing in the calculated lines. Did the authors consider energy and conformational factors?

Response 4:

We sincerely thank you for your in-depth attention and valuable comments on the absolute configuration analysis of the compounds. The absolute configuration of C-3 in compound 3 was determined based on the single-crystal X-ray diffraction data of compound 9, together with consideration of the plausibility of its biosynthetic pathway. We attempted to assign the configuration of C-9 using the Mosher’s method; however, due to the extremely limited remaining quantity of the compound (less than 1 mg), we were unable to successfully prepare the derivatives necessary for stereochemical assignment.

Based on the issue you observed regarding the missing negative Cotton effect around 285 nm in the experimental versus calculated spectra, we employed DP4+ probability analysis, and the results indicated a probability exceeding 99% in support of the assigned configuration, thereby providing statistical evidence for the stereochemical determination. (Page 4, lines 122-126).

Comment 5:

The compounds being phenolic may potentially hold antioxidant effects in the inflammatory pathways. Why was NrF2 activation not determined?

Response 5:

We thank the reviewer for this thoughtful suggestion. We agree that the phenolic nature of resorcylic acid lactones raises the possibility of antioxidant activity and potential involvement of the Nrf2 pathway. However, the present study was designed to focus on inflammatory signaling mechanisms that are directly linked to ulcerative colitis pathogenesis, particularly the MAPK/ERK pathway, which was identified through both transcriptomic analysis and functional validation. While Nrf2 activation may contribute to cytoprotective or antioxidant effects, it was not the primary focus of this work and was therefore not examined. Importantly, the observed suppression of pro-inflammatory cytokines and disease severity can be mechanistically explained by ERK pathway inhibition. We have now acknowledged the potential involvement of antioxidant pathways, including Nrf2 signaling, as an interesting direction for future investigation.

Reviewer 2 Report

Comments and Suggestions for Authors

Dear authors and editor,

During the review of the manuscript, the following questions and comments arose:

1. The NMR spectra of compound 1 differ significantly from those of peniceminolide E, particularly for the signals of the carbon and proton atoms C-7-C-11. The signals for C-9, which is asymmetric, are particularly different. The authors of the manuscript then base their proof of the configuration of compound 1 on its affinity for compound 9, which is identified as Penicimenolide E. However, the spectra of compound 9 are not even provided in the supplementary material to verify this. This raises my doubts. Moreover, according to X-ray structural analysis, compound 9 has a different configuration of the C-9 asymmetric center from Penicimenolide E.

For Penicimenolide E, the configuration of the asymmetric centers was determined to be 3R9S using the Mosher’s method. The authors of this manuscript provide the configurations of their compounds 1 and 9 as 3R9R! Apparently, compound 9 has been incorrectly identified, probably compound 9 is an epimer of Penicimenolide E and is a new compound.

Moreover, the CD spectra of compounds 1, 9, and Penicimenolide E are identical. This also requires some comment.

NMR data for compound 9 must be provided. Furthermore, its X-ray crystallography data must be deposited at the Cambridge Crystallography Data Center (CCDC). The CCDC number has been provided in the experimental section, as well as the CIF file in the supplementary material.

2. The double bond configuration for compound 2 has not been definitively established. The authors of the manuscript claim that the double bond configuration between C-8 and C-9 is Z. They cite a spin-spin coupling constant of 9.6 Hz, but no such values are found in Table 1!!! Table 1 shows completely different constants for H-8 and H-9. The presence of a constant of 15.2 Hz more likely indicates an E configuration of the double bond. This issue needs clarification. It would also be helpful to provide an expanded picture of these signals in the supplementary material.
3. In my opinion, the absolute configuration for compound 3 has not been convincingly established. Quantum-chemical calculations of the ECD spectra are needed.

There are also a number of minor comments:

1. Line 90, the authors likely mean C-18, not C-8.
2. Line 101, the signal value for H-9 is incorrectly reported.
3. The table is incorrect: the first column of the table lists the carbon atom numbers! The α and β designations are used for proton atoms, and only if these signals are accurately determined from the ROESY (NOESY)spectra; otherwise, the letters a and b are used. It is necessary to move the alpha and beta symbols to the column where the proton shifts are indicated.

Following are some notes related to biological activity studies:

1. The authors began the study of anti-inflammatory activity immediately with a test for the production of NO in LPS-stimulated macrophages. However, the results of the assessment of the cytotoxic effect of these compounds on unstimulated macrophages are not shown. Thus, it is unclear whether the effect on the NO level is really anti-inflammatory or whether the NO level decreases due to cell death.
2. In Figure 6G, it is noteworthy that the level of ZO-1/b-actin, claudin-1/b-actin under the action of substance 4 is higher than in the control. This should be noted and discussed in the relevant section.
3. >The Discussion section is missing and the activity data has not been sufficiently discussed. It is necessary to analyze the structure-activity relationship and try to explain why compound 4 has the best activity. In addition, it is necessary to compare the data obtained with those already known about the anti-inflammatory activity of structurally similar compounds.
4. Since the authors claim an ERK-dependent mechanism, it is also necessary to discuss exactly how compound 4 can act in this way, which of the receptors or transcription factors may be the real target with which substance 4 interacts.

Despite the numerous comments, I believe the authors' research is interesting. The manuscript may be accepted for publication after major revisions. This particularly concerns the structures of compounds 1, 2, and 9.

Author Response

During the review of the manuscript, the following questions and comments arose:

Comment 1:

The NMR spectra of compound 1 differ significantly from those of peniceminolide E, particularly for the signals of the carbon and proton atoms C-7-C-11. The signals for C-9, which is asymmetric, are particularly different. The authors of the manuscript then base their proof of the configuration of compound 1 on its affinity for compound 9, which is identified as Penicimenolide E. However, the spectra of compound 9 are not even provided in the supplementary material to verify this. This raises my doubts. Moreover, according to X-ray structural analysis, compound 9 has a different configuration of the C-9 asymmetric center from Penicimenolide E.

For Penicimenolide E, the configuration of the asymmetric centers was determined to be 3R9S using the Mosher’s method. The authors of this manuscript provide the configurations of their compounds 1 and 9 as 3R9R! Apparently, compound 9 has been incorrectly identified, probably compound 9 is an epimer of Penicimenolide E and is a new compound.

Moreover, the CD spectra of compounds 1, 9, and Penicimenolide E are identical. This also requires some comment.

NMR data for compound 9 must be provided. Furthermore, its X-ray crystallography data must be deposited at the Cambridge Crystallography Data Center (CCDC). The CCDC number has been provided in the experimental section, as well as the CIF file in the supplementary material.

Response 1:

We sincerely appreciate the reviewer's careful and valuable feedback. Due to our oversight, there are some errors in the manuscript. We have made corrections in the corresponding sections. (Page 4-5, Table 1). There are indeed significant differences in the NMR signals (particularly in the carbon and hydrogen spectra) of compound 1 compared to the known compound peniceminolide E in the C-9 region. This discrepancy may be attributed to the electron-withdrawing effect and spatial orientation of the ester carbonyl group attached to the C-9 position, which could alter the charge distribution and conformation around C-9, thereby affecting the chemical shifts of the carbon atoms. We have re-examined and meticulously verified the single-crystal X-ray diffraction data of compound 9. The crystallographic data (Flack parameter = 0.04(4)) unequivocally confirm that the absolute configurations of C-3 and C-9 of compound 9 are consistent with those of Peniceminolide E. (Page 3, lines 102-104).

Consequently, the statement in the manuscript, "compound 9 unambiguously assigned its absolute configuration as (3R,9R) (Figure 3)," has been revised to "compound 9 unambiguously assigned its absolute configuration as (3R,9S) (Figure 3)." (Page 3, lines 105).

Since the circular dichroism (CD) spectra of compound 1 and compound 9 are identical, this further confirms that they share the same absolute configuration. We have now fully uploaded the NMR spectra (¹H NMR, ¹³C NMR) of compound 9 and its crystallographic data (.cif file) as part of the supplementary material. (Figure S23, 24).

The The X-ray crystallographic data for compound 9 have been deposited with the Cambridge Crystallographic Data Centre (CCDC). The corresponding deposition number (CCDC No. [2524940]) is provided in the Experimental Section, and the crystallographic information file (CIF) is included in the Supplementary Information. (Page 13, lines 387-395).

Comment 2:

The double bond configuration for compound 2 has not been definitively established. The authors of the manuscript claim that the double bond configuration between C-8 and C-9 is Z. They cite a spin-spin coupling constant of 9.6 Hz, but no such values are found in Table 1!!! Table 1 shows completely different constants for H-8 and H-9. The presence of a constant of 15.2 Hz more likely indicates an E configuration of the double bond. This issue needs clarification. It would also be helpful to provide an expanded picture of these signals in the supplementary material.

Response 2:

We sincerely thank the reviewer for their meticulous and valuable feedback. Due to our oversight, the original assignment of the double bond configuration between C-8 and C-9 was incorrect. Upon re-examination of the raw data, the coupling constant between H-8 and H-9 is indeed 15.2 Hz, not the previously cited value of 9.6 Hz. This value unequivocally indicates an E configuration. We have accordingly corrected the relevant sections in the manuscript (including both the text and tables) and have included an expanded view of the corresponding NMR signals in the Supplementary Material for verification. We apologize for any inconvenience caused by this error and once again express our gratitude for the reviewer's rigorous assessment. (Page 3-4, lines 114 and 118).

Comment 3:

In my opinion, the absolute configuration for compound 3 has not been convincingly established. Quantum-chemical calculations of the ECD spectra are needed.

Response 3:

Thank you for your in-depth attention and valuable comments regarding the absolute configuration analysis of the compound. The absolute configuration of C-3 in compound 3 was determined based on the single-crystal X-ray diffraction data of compound 9, along with consideration of the plausibility of its biosynthetic pathway. In investigating the configuration of C-9, we attempted to apply the Mosher's method; however, due to the extremely limited amount of the compound (less than 1 mg), we were unable to successfully obtain the derivatives required for stereochemical analysis. Furthermore, we performed ECD calculations for the four possible configurations of compound 3. As the hydroxyl group at the C-9 position did not generate distinct Cotton effects with other structural moieties, the calculated spectra did not fully match the experimental data. Ultimately, DP4+ probability analysis provided a confidence level exceeding 99% in support of the assigned configuration, thereby offering statistical evidence for the determination of its stereochemistry.

There are also a number of minor comments:

Minor comment 1:

Line 90, the authors likely mean C-18, not C-8.

Response 1:

Thank you for your revision suggestion. We have changed " as well as the HMBC correlations from H₃-19 to C-8, and from H-9 to C-8 (Figure 2)" to " as well as the HMBC correlations from H₃-19 to C-18, and from H-9 to C-18 (Figure 2)". (Page 3, lines 100).

Minor comment 2:

Line 101, the signal value for H-9 is incorrectly reported.

Response 2:

Thank you for your revision suggestion. We have changed " except for the loss of a ketocarbonyl group and the appearance of a pair of olefinic signals [δ_C 135.4, 131.3; δ_H 5.30 (dd, 15.2, 8.1), 5.40 (m)] at C-8/C-9 in 2" to " except for the loss of a ketocarbonyl group and the appearance of a pair of olefinic signals [δ_C 135.4, 131.3; δ_H 5.30 (ddd, 15.2, 8.1, 1.2), 5.38 (ddd, 15.2, 9.5, 3.5)] at C-8/C-9 in 2".(Page 3, lines 114 and 118， Table 1).

Minor comment 3:

The table is incorrect: the first column of the table lists the carbon atom numbers! The α and β designations are used for proton atoms, and only if these signals are accurately determined from the ROESY (NOESY) spectra; otherwise, the letters a and b are used. It is necessary to move the alpha and beta symbols to the column where the proton shifts are indicated.

Response 3:

We sincerely thank you for your meticulous review regarding the details of the table. Following the standard guidelines, we have uniformly used letters such as a, b, etc. for labeling. We have updated the relevant content in the revised manuscript, including Table 1 and the corresponding sections, to ensure the accuracy and standardization of the data annotations. Once again, we greatly appreciate your valuable feedback, which is crucial for enhancing the rigor of our work. (Page 4-5, Table 1).

Following are some notes related to biological activity studies:

Comment 1:

The authors began the study of anti-inflammatory activity immediately with a test for the production of NO in LPS-stimulated macrophages. However, the results of the assessment of the cytotoxic effect of these compounds on unstimulated macrophages are not shown. Thus, it is unclear whether the effect on the NO level is really anti-inflammatory or whether the NO level decreases due to cell death.

Response 1:

We thank the reviewer for raising this important concern. Cytotoxicity of compounds 4-7 toward unstimulated RAW 264.7 macrophages was evaluated prior to the anti-inflammatory assays, and the corresponding data have now been moved from the Supplementary Materials to the main text with the inclusion of explicit CC₅₀ values for the active compounds. As shown in Table 3, compound 4 displayed low cytotoxicity with CC₅₀ values well above their effective anti-inflammatory concentrations, confirming that the observed NO inhibition was not attributable to cell death (Page 6, lines 178-183 and Page 7, lines 190-191).

Comment 2:

In Figure 6G, it is noteworthy that the level of ZO-1/b-actin, claudin-1/b-actin under the action of substance 4 is higher than in the control. This should be noted and discussed in the relevant section.

Response 2:

We appreciate the reviewer’s careful observation. Indeed, the expression levels of ZO-1 and claudin-1 in the compound 4-treated groups slightly exceeded those of the control group. This likely reflects enhanced restoration and reinforcement of epithelial barrier integrity following DSS-induced injury, rather than a simple return to baseline levels. We have now noted and briefly discussed this observation in Section 2.3, emphasizing that compound 4 may promote tight junction recovery beyond basal expression under inflammatory conditions (Page 7, lines 204-207).

Comment 3:

The Discussion section is missing and the activity data has not been sufficiently discussed. It is necessary to analyze the structure-activity relationship and try to explain why compound 4 has the best activity. In addition, it is necessary to compare the data obtained with those already known about the anti-inflammatory activity of structurally similar compounds.

Response 3:

We thank the reviewer for this constructive suggestion. A dedicated Discussion section has now been added (Page 11, line 284). In the revised manuscript, we analyze the structure–activity relationship among compounds 1-10, highlighting the key structural features associated with anti-inflammatory activity and providing a rationale for the superior potency of compound 4 (Page 11, lines 298-308). In addition, we have compared our findings with previously reported resorcylic acid lactones and structurally related compounds, placing the present results in the context of existing literature on RALs with anti-inflammatory activity (Page 11, lines 319-325).

Comment 4:

Since the authors claim an ERK-dependent mechanism, it is also necessary to discuss exactly how compound 4 can act in this way, which of the receptors or transcription factors may be the real target with which substance 4 interacts.

Response 4:

We agree with the reviewer that clarification of this point is necessary. While our data demonstrate that ERK signaling is functionally required for the anti-inflammatory effects of compound 4, the direct molecular target(s) upstream of ERK have not yet been identified. In the revised Discussion, we now explicitly discuss plausible modes of action, including modulation of receptor-proximal signaling, direct effects on components of the RAF–MEK–ERK cascade, or enhancement of ERK dephosphorylation. We also clarify that identifying the precise binding target will require dedicated target-deconvolution studies, which we consider an important next step to pursue in future work (Page 12, lines 333-344).

Reviewer 3 Report

Comments and Suggestions for Authors

Introduction section
1. Lines 45-47. Only a general list of RALs' biological activities (anticancer, antifungal, etc.) is presented. There is no direct logical connection between this spectrum of activity and chronic inflammation in UC.

Are there any preliminary data or structural features of RALs that make them particularly promising for the treatment of inflammatory bowel diseases, beyond their general anti-inflammatory activity in macrophages?

2. Lines 48-60. In the review, the authors point out the limitations of existing data on in vitro macrophage models. However, there is no hypothesis as to why selected marine RAL derivatives might overcome these limitations.

--

Results section

3. Lines 76-78, 96-98, and 113-15. In all three cases, HR-ESI-MS data is shown in both positive and negative modes. For compound 3, only the negative mode was shown. Why was this?

4. Line 100 lists the spin-spin coupling constants for H-8/H-9 as dd, 15.2, and 8.1, which are typical for trans-oriented protons. However, line 105 states that the constant J8.9 = 9.6 Hz indicates the Z-configuration (i.e., cis). Isn't there a contradiction here? There is likely a typo either in the numerical value or in its interpretation (?).

Please, check and clarify the value of the spin-spin coupling constant J8,9 for compound 2 (lines 100 and 105)

5. The authors miss an excellent opportunity for a preliminary structure-activity analysis based on the presented data. The text merely states the activity of compounds 4-7, but does not even offer a hypothesis as to why these compounds (including the well-known cis-resorcylide 4) are active, while their closest analogs (8, 9, 10) are not. Based on the structural formulas (Figure 1) and activity data, can you make a preliminary hypothesis about the key structural features required for anti-inflammatory activity in this series?

6. Lines 159-162: What are the CC50 values or, at least, the maximum non-toxic concentrations for compounds 4-7, and in particular 4? Present these key data either in the main text or refer to them directly so that the reader can estimate the therapeutic index of the lead compound in vitro.

7. Section 2.5. SEW2871 is a sphingosine-1-phosphate receptor 1 (S1PR1) agonist, not a direct ERK agonist. Why was it chosen for "ERK activation"?

--

M&M section

8. Line 331 lists a fixed concentration of SEW2871/Ulixertinib, but does not specify which one.

9. The age range of mice (6 to 8 weeks) and gender (Male) are indicated. What is the exact number of animals (n) in each group?

10. The molecular weight of DSS is not indicated.

11. Line 371 states: "Samples for the collection were obtained from three separate experiments." It is unclear whether this means 3 biological replicates (3 mice) per group or 3 technical replicates from a single experiment.

12. There is no mention of data deposition. Raw RNA-seq data should be made available in a public repository (SRA).

13. The software and parameters used for alignment, gene quantification, and differential expression analysis are not specified.

14. The list of antibodies includes P-P38 and P38, but the results (sections 2.4 and 2.5) do not present or discuss data on p38. So why were they used?

15. Lines 397-412. The results text (2.3) mentions the measurement of IL-1α, TNF-α, IL-1β, and IL-6 in serum. However, the methods for serum list IL-1α (line 401), but it does not appear in the results. This may be a typo or inconsistency.

Author Response

Introduction section：

Comment 1:

Lines 45-47. Only a general list of RALs' biological activities (anticancer, antifungal, etc.) is presented. There is no direct logical connection between this spectrum of activity and chronic inflammation in UC.

Are there any preliminary data or structural features of RALs that make them particularly promising for the treatment of inflammatory bowel diseases, beyond their general anti-inflammatory activity in macrophages?

Response 1:

We thank the reviewer for this insightful comment. In the revised Introduction, we strengthened the rationale linking RALs to ulcerative colitis by highlighting that several RALs suppress key inflammatory mediators, including NO, PGE₂, and pro-inflammatory cytokines, which are central to chronic intestinal inflammation and epithelial barrier damage in UC. Although current evidence is largely derived from in vitro macrophage models, macrophage-driven cytokine production plays a critical role in epithelial barrier disruption in UC (Page 2, lines 48-50). We therefore proposed that selected marine-derived RAL derivatives, owing to their macrocyclic lactone scaffold and characteristic oxygenation patterns, may modulate shared inflammatory signaling pathways such as MAPK that link immune activation to epithelial injury, supporting their evaluation in experimental colitis models (Page 2, lines 61-70).

Comment 2:

Lines 48-60. In the review, the authors point out the limitations of existing data on in vitro macrophage models. However, there is no hypothesis as to why selected marine RAL derivatives might overcome these limitations.

Response 2:

We appreciate the reviewer’s valuable suggestion. To address this point, we have added a clear hypothesis in the Introduction proposing that selected marine-derived RAL derivatives may overcome the limitations of macrophage-based assays by modulating inflammatory signaling pathways involved in both immune activation and epithelial injury. In particular, the macrocyclic lactone scaffold and characteristic oxygenation patterns of marine RALs may promote interactions with inflammation-related signaling pathways, such as MAPK, which links macrophage activation to intestinal epithelial dysfunction in UC. These additions provide a mechanistic basis for extending macrophage-based observations to in vivo colitis models (Page 2, lines 56-70).

Results section：

Comment 3:

Lines 76-78, 96-98, and 113-15. In all three cases, HR-ESI-MS data is shown in both positive and negative modes. For compound 3, only the negative mode was shown. Why was this?

Response 3:

Thank you for the reviewer's careful review and inquiry. The HR-ESI-MS data for compound 3 is presented only in negative ion mode because no clear molecular ion peak was detected in the positive ion mode. The signal intensity was too low, or background interference was significant, making it impossible to obtain accurate high-resolution mass spectrometry information.

Comment 4:

Line 100 lists the spin-spin coupling constants for H-8/H-9 as dd, 15.2, and 8.1, which are typical for trans-oriented protons. However, line 105 states that the constant J8.9 = 9.6 Hz indicates the Z-configuration (i.e., cis). Isn't there a contradiction here? There is likely a typo either in the numerical value or in its interpretation (?).

Please, check and clarify the value of the spin-spin coupling constant J8,9 for compound 2 (lines 100 and 105)

Response 4:

We sincerely appreciate the reviewer's thorough and valuable feedback. Due to our oversight, the original assignment of the double‑bond configuration between C‑8 and C‑9 was incorrect. After re‑examining the raw data, the coupling constant between H‑8 and H‑9 is confirmed to be 15.2 Hz, not the previously cited value of 9.6 Hz. This value unequivocally supports an E‑configuration. Accordingly, we have corrected the relevant sections in the manuscript (both in the text and tables) and have included an expanded view of the corresponding NMR signals in the Supplementary Material for verification. (Page 3-4, lines 114 and 118, Table 1).

Comment 5:

The authors miss an excellent opportunity for a preliminary structure-activity analysis based on the presented data. The text merely states the activity of compounds 4-7, but does not even offer a hypothesis as to why these compounds (including the well-known cis-resorcylide 4) are active, while their closest analogs (8, 9, 10) are not. Based on the structural formulas (Figure 1) and activity data, can you make a preliminary hypothesis about the key structural features required for anti-inflammatory activity in this series?

Response 5:

We sincerely appreciate your highly constructive suggestion. Following your guidance and based on the structural formulas presented in Figure 1 along with the activity data, we have incorporated a preliminary structure-activity relationship analysis into the "4. Conclusion" section of the manuscript. (Page 11, lines 298-308).

Comment 6:

Lines 159-162: What are the CC50 values or, at least, the maximum non-toxic concentrations for compounds 4-7, and in particular 4? Present these key data either in the main text or refer to them directly so that the reader can estimate the therapeutic index of the lead compound in vitro.

Response 6:

We thank the reviewer for this important point. Cytotoxicity of compounds 4-7 toward unstimulated RAW 264.7 macrophages was evaluated prior to the anti-inflammatory assays, and the corresponding data have now been moved from the Supplementary Materials to the main text with the inclusion of explicit CC₅₀ values for the active compounds. As shown in Table 3, compound 4 displayed low cytotoxicity with CC₅₀ values well above their effective anti-inflammatory concentrations. We have now explicitly referred to these data in Section 2.2 to allow readers to estimate the therapeutic index of the lead compound in vitro (Page 6, lines 178-183 and Page 7, lines 190-191).

Comment 7:

Section 2.5. SEW2871 is a sphingosine-1-phosphate receptor 1 (S1PR1) agonist, not a direct ERK agonist. Why was it chosen for "ERK activation"?

Response 7:

We appreciate the reviewer’s clarification. SEW2871 was selected because activation of S1PR1 is well established to induce downstream ERK phosphorylation in macrophages. In this study, SEW2871 was used as an indirect but physiologically relevant tool to enhance ERK signaling rather than as a direct ERK agonist. We have revised the text in Section 2.5 to clarify this rationale and avoid potential misunderstanding (Page 10, lines 260-263).

Comment 8:

Line 331 lists a fixed concentration of SEW2871/Ulixertinib, but does not specify which one.

Response 8:

We thank the reviewer for pointing out this ambiguity. The concentrations used were 500 nM for SEW2871 and 1 μM for Ulixertinib. This information has now been explicitly stated in Section 4.6 to ensure clarity (Page 13, line 422).

Comment 9:

The age range of mice (6 to 8 weeks) and gender (Male) are indicated. What is the exact number of animals (n) in each group?

Response 9:

We apologize for the omission. The exact number of animals used in each experimental group was six (n = 6 per group), and this information has now been clarified in Section 4.8 of the Methods (Page 14, line 438).

Comment 10:

The molecular weight of DSS is not indicated.

Response 10:

We thank the reviewer for pointing this out. The DSS used in this study has an average molecular weight of 40 kDa, as provided by the manufacturer (Adamas). This information has now been added to Section 4.8 (Page 14, line 439).

Comment 11:

Line 371 states: "Samples for the collection were obtained from three separate experiments." It is unclear whether this means 3 biological replicates (3 mice) per group or 3 technical replicates from a single experiment.

Response 11:

We appreciate this request for clarification. The RNA-seq analysis was performed using three independent biological replicates (three mice) per group, not technical replicates. We have revised Section 4.10 to explicitly state this (Page 14, lines 464-465).

Comment 12:

There is no mention of data deposition. Raw RNA-seq data should be made available in a public repository (SRA).

Response 12:

We agree with the reviewer. The raw RNA-seq data have now been deposited in the NCBI Sequence Read Archive (SRA). The corresponding accession number PRJNA1406744 has been added to the Data Availability Statement in the revised manuscript (Page 16, lines 540-541).

Comment 13:

The software and parameters used for alignment, gene quantification, and differential expression analysis are not specified.\

Response 13:

We thank the reviewer for this helpful suggestion. We have now revised Section 4.10 of the Methods to specify the software packages and key parameters used for RNA-seq data analysis. Briefly, raw sequencing reads were quality filtered using fastp, aligned to the Mus musculus reference genome using HISAT2, and gene expression levels were quantified using RSEM. Differential expression analysis was performed using DESeq2, with genes showing |log₂ fold change| ≥ 1 and FDR < 0.05 considered significantly differentially expressed (Page 14, lines 457-473).

Comment 14:

The list of antibodies includes P-P38 and P38, but the results (sections 2.4 and 2.5) do not present or discuss data on p38. So why were they used?

Response 14:

We appreciate the reviewer’s careful reading. p38 phosphorylation was initially examined during exploratory analyses but did not show significant changes upon compound 4 treatment. As these data were not central to the conclusions, they were not included in the Results. To avoid confusion, we have now removed P-P38 and P38 antibodies from the antibody list in Section 4.12 (Page 15, lines 489-490).

Comment 15:

Lines 397-412. The results text (2.3) mentions the measurement of IL-1α, TNF-α, IL-1β, and IL-6 in serum. However, the methods for serum list IL-1α (line 401), but it does not appear in the results. This may be a typo or inconsistency.

Response 15:

We thank the reviewer for identifying this inconsistency. Measurement of IL-1α was initially planned but was not included in the final data analysis. The mention of IL-1α in the serum ELISA Methods section was therefore a typographical oversight. This has now been corrected to ensure consistency between the Methods and Results (Page 15, lines 500-501).

Reviewer 4 Report

Comments and Suggestions for Authors

Dear Authors,

This manuscript describes a solid natural products research on fungus Penicillium sp. HN20 isolated from Chinese plant root Lumnitzera littorea, which lead to the purification of three new analogues and seven known compounds. The structures of newly identified compounds were characterized by 2D NMR, x-ray crystallography and ECD calculation. Finally, all isolated compounds were subjected to anti-inflammatory assays. 

Figure 2, 2D NMR analysis of compound 1, indication for the COSY correlation between H-7 and H-8 in compound 1 is missing. 

Table 1, coupling constant value between two olefinic protons H-8 and H-9 is J = 15.2 Hz, but the illustrated structure 2 indicated a cis configuration at double bond H-8/H-9. The correct geometry for this double bond should be trans based on the coupling constant values. 

Figure 1, compounds 1 - 3 should be colored in black. 

Lines 91-93, x-ray crystallography was performed on compound 9, but there is no explanation. Why compound 9 was selected for x-ray crystallography instead of compound 1? 

Author Response

Comment 1:

Figure 2, 2D NMR analysis of compound 1, indication for the COSY correlation between H-7 and H-8 in compound 1 is missing.

Response 1:

We sincerely appreciate your meticulous review of this manuscript. Following careful verification, we have now clearly indicated the correlation signal between H-7 and H-8 in the COSY correlation network diagram presented in Figure 2. (Page 5, lines 154-155, Figure 2).

Comment 2:

Table 1, coupling constant value between two olefinic protons H-8 and H-9 is J = 15.2 Hz, but the illustrated structure 2 indicated a cis configuration at double bond H-8/H-9. The correct geometry for this double bond should be trans based on the coupling constant values. 

Response 2:

We sincerely appreciate your valuable feedback. We apologize for the oversight in our original manuscript. The configuration of the H-8/H-9 double bond in compound 2 has been revised from "cis" in the original text to "trans (or E)" to ensure full consistency with the coupling constant data. The relevant sections in the Results and Discussion have been updated accordingly, explicitly stating the trans configuration of this double bond, with the coupling constant J = 15.2 Hz cited as the key supporting evidence. (Page 3-4, lines 114 and 118, Table 1).

Comment 3:

Figure 1, compounds 1 - 3 should be colored in black.

Response 3:

We sincerely appreciate your careful attention to the details of the figures. We have followed your suggestion by uniformly adjusting the structures of compounds 1–3 in Figure 1 to black. (Page 3, lines 82-83, Figure 1).

Comment 4:

Lines 91-93, x-ray crystallography was performed on compound 9, but there is no explanation. Why compound 9 was selected for x-ray crystallography instead of compound 1?

Response 4:

We sincerely appreciate your thorough review of the manuscript. While we attempted to cultivate single crystals for compounds 1–3, none of them yielded crystals of sufficient size and quality for X-ray diffraction analysis under various conditions. In contrast, compound 9 successfully formed high-quality single crystals in a methanol solvent system, providing direct experimental evidence for structural elucidation. The molecular structure of compound 9 contains key chiral centers and functional groups (particularly the configurations at C-3 and C-9) that are common to this series of compounds. By determining the absolute configuration of compound 9 and correlating it with the CD spectroscopic data and biosynthetic pathways of compound 1, we were able to reasonably deduce and support the assignment of the absolute configuration for compound 1. We have added corresponding explanations near lines 92–95 to clearly articulate the rationale for selecting compound 9 for single-crystal analysis and to emphasize its relevance and supporting role in confirming the configuration of compound 1. (Page 3, lines 102-104).

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The authors did not satisfactorily answered the concern regarding the optical purity. It must be measured using chiral HPLC or other equivalent method.

Comments on the Quality of English Language

pls check minor errors esp formatting

Author Response

Comment 1:

The authors did not satisfactorily answered the concern regarding the optical purity. It must be measured using chiral HPLC or other equivalent method.

Response 1:

Thank you very much for reviewing our manuscript again and for your important comments regarding the optical purity issue. We have followed your suggestion and determined the optical purity of the product using chiral HPLC. The relevant data and spectra have been submitted as supporting materials. (Figs. S28–S37)

Reviewer 2 Report

Comments and Suggestions for Authors

Dear authors, I am very pleased that all my comments were analyzed and taken into account, which allowed me to improve your manuscript. However, I still insist on changing Table 1. Column 1 indicates the carbon atom number. The use of the symbols a and b in column 1 is incorrect. These symbols are correctly used for proton atoms. Accordingly, they should be moved to columns 3, 5, and 7, which provide the hydrogen atom signal values. However, in your specific case, using letter expressions for protons is completely pointless; they carry no semantic meaning and are not used in the manuscript.

Author Response

Comment 1:

Dear authors, I am very pleased that all my comments were analyzed and taken into account, which allowed me to improve your manuscript. However, I still insist on changing Table 1. Column 1 indicates the carbon atom number. The use of the symbols a and b in column 1 is incorrect. These symbols are correctly used for proton atoms. Accordingly, they should be moved to columns 3, 5, and 7, which provide the hydrogen atom signal values. However, in your specific case, using letter expressions for protons is completely pointless; they carry no semantic meaning and are not used in the manuscript.

Response 1:

Thank you sincerely for your meticulous review and highly valuable guidance. Regarding the format of Table 1, we fully agree with your expert opinion and have revised it according to your request: all alphabetical markers in the first column have been removed to ensure the scientificity and standardization of the table format. (Page 4-5, Table 1)

Reviewer 3 Report

Comments and Suggestions for Authors

The authors did an excellent job of correcting and responding to comments. I am completely satisfied.

Author Response

Comment 1:

The authors did an excellent job of correcting and responding to comments. I am completely satisfied.

Response 1:

Thank you for your positive feedback. We are pleased to hear that you are satisfied with the revisions. We truly appreciate your acknowledgment of our efforts.

Reviewer 4 Report

Comments and Suggestions for Authors

Dear Authors,

All comments have been addressed positively. I have no further comment.

Author Response

Comment 1:

All comments have been addressed positively. I have no further comment.

Response 1:

Thank you for your positive feedback. We are pleased to hear that you are satisfied with the revisions. We truly appreciate your acknowledgment of our efforts.