Rambutan (Nephelium lappaceum L.) Shell as a Source of Polyphenols: Chemical Characterization and Biological Activities

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This research study provides a comprehensive analysis of rambutan shell (peel) extracts, focusing on their biological properties and chemical composition. Particular emphasis was placed on antioxidant, antibacterial, antibiofilm, antitumor, and hemolytic properties. Overall, I found this to be an interesting and fairly comprehensive study. I believe that the manuscript meets the publication requirements for Molecules; however, I recommend that the authors address the following:

1. Line 124: Ultrasonication time?
2. Lines 183, 197, 241, 250: Please specify what type of rambutan extract was used: ethanol-citric acid mixture (85-15) or acidified methanol? Other?
3. The authors must include the following reference (https://doi.org/10.1016/j.foodres.2023.112574). Please highlight this manuscript's main contribution relative to the cited reference, since both studies describe some similar analyses.
4. Where necessary, please include the supplier of chemicals and equipment.
5. What is the ripeness stage of the rambutan samples used?
6. Line 347: The natural abundance of p-coumaric acid normally exceeds that of m-coumaric acid. How did the authors know the latter is the most abundant in this study?
7. Why did the authors decide to present the negative ionization data before the positive ionization data? Positive ions are usually described, or both types of fragmentation are presented because they complement each other.
8. Monoisotopic mass of corilagin is 634.0806, authors describe 633. The same for other dominant ions. Please, explain.
9. How many matched peaks were selected for determining the GNPS cosine score?
10. The GNPS cosine score method tends to generate high rates of false positives, particularly with spectra from complex matrices, which may be the case here. How did the authors address this possibility?
11. The authors mention in line 375 that they used a high-resolution mass spectrometer; however, the corresponding data are not expressed as high-resolution masses, i.e., monoisotopic or exact masses. Using high-resolution masses instead of nominal masses is recommended, as the former provide greater certainty in metabolite identification.

Author Response

This research study provides a comprehensive analysis of rambutan shell (peel) extracts, focusing on their biological properties and chemical composition. Particular emphasis was placed on antioxidant, antibacterial, antibiofilm, antitumor, and hemolytic properties. Overall, I found this to be an interesting and fairly comprehensive study. I believe that the manuscript meets the publication requirements for Molecules; however, I recommend that the authors address the following:

 

Line 124: Ultrasonication time? 

Response: We appreciate the reviewer´s suggestion. We have included the ultrasonic agitation time. 

Line 119: .. homocysteine, then homogenised and stirred for one minute in an ultrasonic bath. 

 

Lines 183, 197, 241, 250: Please specify what type of rambutan extract was used: ethanol-citric acid mixture (85-15) or acidified methanol? Other? 

Response: We sincerely thank the reviewer. A single optimized extract was evaluated for all biological and antioxidant activities, which was obtained using the acidified ethanol system (ethanol–citric acid mixture, 85:15 v/v). Methanol was strictly used as a reaction solvent within the analytical protocols of the antioxidant assays (e.g., to dissolve reagents), not as the extraction solvent for the plant material.

 

The authors must include the following reference (https://doi.org/10.1016/j.foodres.2023.112574). Please highlight this manuscript's main contribution relative to the cited reference, since both studies describe some similar analyses.

Where necessary, please include the supplier of chemicals and equipment. 

Response: We sincerely thank the reviewer for directing our attention to this relevant study (https://doi.org/10.1016/j.foodres.2023.112574). While both studies explore the valorization of N. lappaceum byproducts, our manuscript offers distinct biotechnological and phytochemical innovations: (1) we utilized a highly targeted acidified green-solvent extraction platform that preserves unstable oligomeric ellagitannins, (2) our HPLC-DAD-MS dereplication successfully annotated late-eluting triterpenoid saponins (hederagenin isomers) not discussed in the cited paper, and (3) we expanded the biological characterization to include specific antibiofilm mechanisms against clinical isolates and cytotoxicity evaluation.

 

What is the ripeness stage of the rambutan samples used? 

Response:

We sincerely thank the reviewer for this observation. In the modified version of the manuscript we clarified that rambutan fruits were collected at commercial maturity, corresponding to Stage 6 of the visual ripening scale (intense red skin and spines).

 

Line 347: The natural abundance of p-coumaric acid normally exceeds that of m-coumaric acid. How did the authors know the latter is the most abundant in this study? 

Response: We are grateful for the reviewer’s suggestion. We would like to clarify that the relevant values can be found in Table 2.

 

Why did the authors decide to present the negative ionization data before the positive ionization data? Positive ions are usually described, or both types of fragmentation are presented because they complement each other.

 

Response:

 

We thank the reviewer for this important methodological observation. To clarify, both positive (ESI⁺) and negative (ESI⁻) ionization modes were evaluated during method development for the chromatographic profiling of N. lappaceum shell extract. However, given the chemical nature of the dominant compound classes identified in this matrix, ESI⁻ mode provided substantially richer and more informative spectra, and was therefore selected as the sole acquisition mode for data reporting. 

 

The mechanistic basis for this outcome is well established in the literature. Ellagitannins carry multiple galloyl and hexahydroxydiphenoyl (HHDP) groups bearing phenolic hydroxyl functions with pKₐ values in the range of 5 – 9. These sites deprotonate efficiently in the ESI source, generating stable deprotonated molecular ions [M−H]⁻ or, for higher-molecular-weight oligomers, doubly charged [M−2H]²⁻ anions with high ionization efficiency. Critically, the diagnostic fragmentation series that defines ellagitannin class assignment are exclusively generated from deprotonated precursors and are absent or non-diagnostic under positive ionization conditions (10.1002/mas.70013).  

 

This approach is entirely consistent with all published HPLC-MS/MS studies of N. lappaceum peel phenolics as well as with the methodology applied to the most closely analogous ellagitannin-rich matrices in the literature, including pomegranate (https://pubmed.ncbi.nlm.nih.gov/23140740/), strawberry (https://pubs.acs.org/doi/10.1021/jf0702592), and Terminalia chebula https://www.mdpi.com/1420-3049/30/11/2451), all of which exclusively employ ESI⁻ for ellagitannin characterization. Positive ESI mode is considered analytically complementary only when the extract contains substantial quantities of compound classes preferentially ionized as cations. 

 

The use of positive ionization mode was initially considered because it could be potentially helpful in anthocyanin's identification. However, we did not identify any anthocyanin in negative or positive ionization mode. These observations are now reported in the corrected manuscript.  

 

Monoisotopic mass of corilagin is 634.0806, authors describe 633. The same for other dominant ions. Please, explain. 

Response: We thank the reviewer for this observation The m/z values reported in Table 3 correspond to deprotonated molecular ions [M−H]⁻, not to neutral monoisotopic masses. Moreover, as we are using an LTQ-XL instrument (ion tramp) we have not high-resolution mass determination. It was also emphasized in the reviewed version of the manuscript.

How many matched peaks were selected for determining the GNPS cosine score? 

Response: We wish to clarify that the fragment ions reported in Table 3 for IDs 14 and 15 represent only the most abundant ions selected for reporting purposes, following standard practice in ion trap MS/MS data reporting. The complete experimental MS² spectrum (mgf file) is analyzed in MzMine with the GNPS database. In this case, the compounds 14 and 15 showed 23 and 19 peaks matched respectively with their references spectrums, yielding a cosine score of 0.886 and 0.857. This number of matched signals together with the additional information that the cumulative intensity of the matched peaks is of 0.780 and 0.769 provides robust statistical confidence in the spectral match. By the way, the secondary fragments observed at m/z 394, 405, 427, and 440 correspond to MS³ fragmentation of the m/z 472 aglycone ion and were not included in the MS² cosine score calculation. Additionally, we sincerely apologize for a typographical inconsistency in the previously submitted version: the cosine score for ID 15 was inadvertently misreported, and the correct value is 0.857, as now stated above. This has already been corrected in the revised manuscript, and we thank the reviewer for the careful reading that allowed us to detect and amend this inconsistency.

The GNPS cosine score method tends to generate high rates of false positives, particularly with spectra from complex matrices, which may be the case here. How did the authors address this possibility? 

Response: The risk of false-positives in spectral matches is a very important topic and problematic as suggested by the reviewer, especially in complex plant matrices. We consider in GNPS, the cosine similarity, the number of matched peaks and also that the matches peaks cover the most intense peaks of the spectrum. However, GNPS spectral match was intentionally not used as the sole basis for identification. The tentative assignment of IDs 14 and 15 as hederagenin-type triterpenoid saponins rests on five independent and mutually consistent lines of evidence:

1. GNPS cosine score: A score of 0.886 and 0.857 based on 23 and 19 matched  signals against the library reference spectrum, exceeding the commonly applied threshold of 0.7 for tentative annotation.
2. Diagnostic neutral loss series: The precursor at m/z 882 undergoes sequential losses of 132 Da (pentose), 146 Da (deoxyhexose), and 132 Da (pentose), generating a pentose–deoxyhexose–pentose sugar release sequence that is chemically specific to glycosylated triterpenoids and structurally incompatible with any ellagitannin, gallotannin, or phenolic acid class.
3. Aglycone identity and MS³ confirmation: Both the neutral loss series and the GNPS-matched fragmentation independently converge on m/z 472, identified as the hederagenin aglycone. MS³ fragmentation of this ion yields the characteristic dehydration and decarboxylation series at m/z 394 (base peak), 405, 427, and 440, consistent with the hederagenin skeleton as documented in independent dereplication studies (https://doi.org/10.1039/C4AY02837F).
4. Chromatographic behavior: IDs 14 and 15 elute at 27.72 and 27.77 min respectively, in the late-eluting apolar region of the chromatogram, fully consistent with the physicochemical properties of triterpenoid saponins and inconsistent with any ellagitannin or phenolic acid class.
5. Chemotaxonomic and database corroboration: Hederagenin-type saponins with this glycosylation pattern have been previously isolated and structurally characterized from Nephelium lappaceum hull (https://link.springer.com/article/10.1007/s10600-012-0108-5) and are recorded in both the COCONUT and LOTUS open natural products databases for the Nephelium genus.

We additionally acknowledge that most of the GNPS library reference spectrum were recorded with high-resolution instrument, whereas our experimental spectra were acquired on a linear ion trap at unit mass resolution. This instrumental difference (plus many other issues inside the GNPS database itself) may account for minor discrepancies in relative fragment intensities between the query and library spectra, and represents an additional reason why the GNPS score alone was treated as supporting rather than definitive evidence.

In light of these comments, the corresponding passage in Section 3.2 has been revised in the manuscript to explicitly incorporate all lines of evidence discussed above. The revised text now reads:

The late-eluting region was characterized by triterpenoid saponins. IDs 14 and 15 (m/z 882) showed sequential neutral losses of 132, 146, and 132 Da, consistent with a pentose–deoxyhexose–pentose sugar release sequence, yielding the hederagenin aglycone at m/z 472. MS³ fragmentation of this aglycone ion produced characteristic fragments at m/z 394 (base peak), 405, 427, and 440, corresponding to sequential dehydration and decarboxylation losses from the hederagenin skeleton, consistent with previously reported fragmentation data [DOI: 10.1039/C4AY02837F]. Accordingly, IDs 14 and 15 are tentatively assigned as two isomers of 10-[3-[3,5-dihydroxy-6-methyl-4-(3,4,5-trihydroxyoxan-2-yl)oxyoxan-2-yl]oxy-4,5-dihydroxyoxan-2-yl]oxy-9-(hydroxymethyl)-2,2,6a,6b,9,12a-hexamethyl-1,3,4,5,6,6a,7,8,8a,10,11,12,13,14b-tetradecahydropicene-4a-carboxylic acid. This assignment is further supported by a GNPS cosine score of 0.886, based on 19 matched fragment ions between the experimental spectrum and the library reference, and by the previously documented occurrence of this compound in the Nephelium genus in both the COCONUT [39] and LOTUS [40] databases.

The authors mention in line 375 that they used a high-resolution mass spectrometer; however, the corresponding data are not expressed as high-resolution masses, i.e., monoisotopic or exact masses. Using high-resolution masses instead of nominal masses is recommended, as the former provide greater certainty in metabolite identification. 

Response: We thank the reviewer for identifying this inconsistency. The reviewer is entirely correct. The term "high-resolution" used in line 375 of the original manuscript was applied erroneously and we sincerely apologize for this inaccuracy.

The instrument used in this study was a Thermo Fisher Scientific LTQ-XL, which operates at unit mass resolution. This instrument is explicitly not a high-resolution mass spectrometer. The LTQ-XL does not provide this capability, and consequently all m/z values reported in Table 3 are nominal masses.

The term has been removed from the revised manuscript. Line 375 now reads:

"HPLC-DAD-MS/MS analysis revealed that the phytochemical signature of N. lappaceum shell extract is dominated by hydrolyzable tannins..."

Reviewer 2 Report

Comments and Suggestions for Authors

This manuscript investigates the phytochemical composition and multifunctional biological activities of phenolic and anthocyanin extracts from rambutan (Nephelium lappaceum L.) shell, an agro-industrial byproduct. Using HPLC‑DAD‑MS/MS, the authors identified ellagitannins as major components and evaluated antioxidant, antibacterial, antibiofilm, antitumoral, and hemolytic properties. The topic aligns well with circular economy and natural product research, and the dataset is generally supportive of the claimed bioactivities. However, the manuscript suffers from insufficient experimental details and missing critical data, which weaken the validity and reproducibility of the study.

1. Table 1 and Table 5, what is “*” mean?
2. Line 205, “a final concentration of 2.45 APS”? Line 259, why the absorbance was ~0.7 at 734 nm?
3. The details of the experimental part are too incomplete, such as the extraction time, ultrasonic power, etc.
4. Section 2.6 mentions MBIC₅₀, but there are no relevant data in the main text, please confirm.
5. The manuscript lacks extraction yield data, which is essential to assess the efficiency and practical value of the developed extraction procedure.
6. Table 3, what is “See notes” mean?
7. Reference, many article number of page range is missing.

Author Response

This manuscript investigates the phytochemical composition and multifunctional biological activities of phenolic and anthocyanin extracts from rambutan (Nephelium lappaceum L.) shell, an agro-industrial byproduct. Using HPLC‑DAD‑MS/MS, the authors identified ellagitannins as major components and evaluated antioxidant, antibacterial, antibiofilm, antitumoral, and hemolytic properties. The topic aligns well with circular economy and natural product research, and the dataset is generally supportive of the claimed bioactivities. However, the manuscript suffers from insufficient experimental details and missing critical data, which weaken the validity and reproducibility of the study.

 

Table 1 and Table 5, what is “*” mean?
Response: We thank the reviewer for this important observation. In the modified version of the manuscript we clarified that (*) means that these seven strains are clinical multidrug-resistant isolates.

Line 205, “a final concentration of 2.45 APS”? Line 259, why the absorbance was ~0.7 at 734 nm? 

Response: Thank you for your observation. The text has been corrected to better explain the ABTS radical cation solution preparation as follows: In brief, 7 mM ABTS solution was mixed with 245 mM ammonium persulfate (APS) solution so that the final APS concentration is 2.45 mM. Thereon, the solution was placed in the dark for 16 hours and then diluted in water until the absorbance was ~0.7 at 734 nm (ABTS radical solution).

 

The ABTS radical cation is dark blue-green solution whose target absorbance is 0.70 ± 0.02 at 734 nm allowing for a linear, sensitive decrease in absorbance when antioxidants are added (Zheleva-Dimitrova et al. 2010)

 

Zheleva-Dimitrova D, Nedialkov P, Kitanov G. Radical scavenging and antioxidant activities of methanolic extracts from Hypericum species growing in Bulgaria. Pharmacogn Mag. 2010 Apr;6(22):74-8. doi: 10.4103/0973-1296.62889. Epub 2010 May 5. PMID: 20668569; PMCID: PMC2900065.  . 

 

The details of the experimental part are too incomplete, such as the extraction time, ultrasonic power, etc.

Response: We appreciate the reviewer´s suggestion. We have included the ultrasonic extraction time.

Line x: homogenised and stirred for one minute in an ultrasonic bath.

 

Section 2.6 mentions MBIC50, but there are no relevant data in the main text, please confirm. 

Response: We appreciate this observation. The results concerning the MBIC50 assessment are described in Section 3.5, and the text has been slightly modified to avoid confussion.

 

The manuscript lacks extraction yield data, which is essential to assess the efficiency and practical value of the developed extraction procedure. 

Response:

We appreciate the reviewer´s observation. In the current version of the manuscript, all the details of the extraction and yield are included.

Table 3, what is “See notes” mean? 

Response:
We thank the reviewer for this observation. The notation "See notes" in the Resource column of Table 3 for IDs 9, 10, and 11 was intended to direct the reader to an explanatory passage in Section 3.2 of the original manuscript, where these compounds were discussed. We acknowledge that this notation was insufficiently clear and did not meet standard editorial conventions for table footnotes.

In the revised manuscript, the following corrections have been made:

1. Table 3: The notation "See notes" has been replaced by the formal footnote marker ᵃ for IDs 9, 10, and 11. The following footnote has been added to the table caption:

ᵃ No specific literature match available. Tentatively classified as ellagitannins based on the diagnostic reporter ion at m/z 301 and secondary fragmentation at m/z 257, 229, and 185. See Section 3.2 for full discussion.

1. Section 3.2: The corresponding text has been revised and expanded for clarity:

"ID7 (m/z 433) was tentatively characterized as an ellagic acid pentoside based on loss of a pentose (−132 Da) yielding m/z 301, whose identity was confirmed by characteristic fragments at m/z 257, 229, and 185 [37,38]. IDs 9–11 (m/z 997, 979, and 923) did not match any known fragmentation patterns in available databases or literature. However, all three displayed an intense signal at m/z 301, consistent with the ellagic acid / lactonized HHDP anion, together with the characteristic secondary fragmentation cascade at m/z 257, 229, and 185. Given the confirmed diagnostic value of this ion series for ellagitannin class assignment, these compounds are tentatively classified as ellagitannins."

1. A misplaced citation [29] that appeared erroneously at the end of the IDs 9–11 paragraph in the original manuscript has been removed in the revised version. Reference [29] corresponds to the characterization of corilagin (ID3) and was inadvertently displaced during manuscript editing.

Reference, many article number of page range is missing. 

Response:
We sincerely apologize for the formatting omissions in the bibliography. We have thoroughly audited the reference list in the revised manuscript, ensuring that all complete information is present and accurately formatted according to the Molecules editorial guidelines.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Having read the revised version of the manuscript, I can confirm that the authors have adequately addressed the comments made by this reviewer. I believe the manuscript now meets the publication requirements for Molecules. I have one small recommendation: please ensure that all species names are written in italics. 

Reviewer 2 Report

Comments and Suggestions for Authors

The manuscript has been corrected.