Comparative Analysis of Chemical Composition and Antioxidant Activity in Conventional, Civet, and Elephant Coffees: Is There a Definitive Authentication Marker of Elephant Coffee?
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsDear Authors,
Review of paper:
Chemical composition and antioxidant activity of elephant coffee
This study contributes to the sparse literature on elephant coffee. It is very important because elephant coffee, which is currently the most expensive coffee in the world, is an attractive commodity for adulteration.
I recommend two corrections:
Line 67-75: Please add the number of elephant, civet and conventional coffee samples used in the study.
Line 218-219: Please add to the sentence "Elephant coffee was significantly different in DPPH, ABTS, TPC and TFC values (p≤0.01)" in comparison with which coffee.
Comments for author File: 
Comments.pdf
Author Response
Query: Line 67-75: Please add the number of elephant, civet and conventional coffee samples used in the study.
Response: It is now indicated that each sample was represented by a single commercial batch by the number N being added after each sample (i.e. N = 1; se lines 85-86).
Query: Line 218-219: Please add to the sentence "Elephant coffee was significantly different in DPPH, ABTS, TPC and TFC values (p≤0.01)" in comparison with which coffee.
Response: The sentence was changed to: “Elephant coffee significantly differed in values for DPPH, ABTS, TPC and TFC (p≤0.01) to both conventional and civet coffee” (see line 256).
Reviewer 2 Report
Comments and Suggestions for AuthorsThe paper presents interesting results.
The abstract and method are sufficiently clear to the paper overall.
The work is well described and elucidative of some principal characteristics of the coffee.
Comments and corrections.
1. Line 255. The error must be corrected in the final proofs.
2. Some other compounds may be characterized in Table 2 and the HPLC quantification. Can the author comment?
3. In Table 3, some compounds, such as pyridine, are not natural compounds that can be derived from roost coffee. It can be found in minimal quantities in roasted foods but is mainly known as a synthetic chemical. So, maybe the authors can comment on that case or others? Or can there be another discussion about the kind of roost and time?
Author Response
Query: Line 255. The error must be corrected in the final proofs.
Response: The broken figure reference has been repaired (see line 290).
Query: Some other compounds may be characterized in Table 2 and the HPLC quantification. Can the author comment?
Response: In the HPLC-UV analysis, approximately 20 standards of phenolic compounds, flavonoids, and caffeine were tested. Of these, only the five compounds listed in Table 1 were detected in the coffee samples; the remaining compounds were not present. As this was a targeted analysis and identification was based solely on UV spectra, additional compounds present in the samples could not be reliably identified and thus, it is impossible to include them in the table.
Query: In Table 3, some compounds, such as pyridine, are not natural compounds that can be derived from roost coffee. It can be found in minimal quantities in roasted foods but is mainly known as a synthetic chemical. So, maybe the authors can comment on that case or others? Or can there be another discussion about the kind of roost and time?
Response: Pyridines, pyrazines, pyrroles belong to the one of the most common groups of volatiles in roasted coffee. Pyridine is considered to be formed from the decomposition of trigonelline during the roasting (1). This pyridine alkaloid is present in high amounts in coffee. The pyridine content is influenced by roaster type and roasting profile. Higher levels of pyridine in roasted coffee indicates so called scorched roast defect that can be detectable as dark spots on the beans due to excessive heat. The levels of pyridine in coffee reported in literature range from 1.7 to 28.7 % RPA (2). In our study the roasted coffee contained from approx. 2.8 to 6.3 % RPA. In our opinion the content of pyridine in our finding does not require special emphasis.
- Yang N, Liu C, Liu X, Degn TK, Munchow M, Fisk I. Determination of volatile marker compounds of common coffee roast defects. Food Chemistry. 2016 Nov 15;211:206-14.
- Gancarz M, Dobrzański Jr B, Malaga-Toboła U, Tabor S, Combrzyński M, Ćwikła D, Strobel WR, Oniszczuk A, Karami H, Darvishi Y, Żytek A. Impact of coffee bean roasting on the content of pyridines determined by analysis of volatile organic compounds. Molecules. 2022 Feb 25;27(5):1559.
Reviewer 3 Report
Comments and Suggestions for AuthorsThe review of the manuscript entitled „ Chemical composition and antioxidant activity of elephant coffee” written by Hájíček et al. (Manuscript ID: beverages-3606335)
The submitted manuscript addresses the chemical composition and antioxidant activity of so-called elephant dung coffee a rare and highly valued product processed biologically in the gastrointestinal tract of elephants. The authors take on a topic that is both scientifically relevant and of consumer interest, given the growing market for luxury coffees and the increasing demand for authenticity verification.
The manuscript has scientific value, particularly in terms of profiling volatile compounds and applying a range of antioxidant assays. However, the experimental design is limited, and in its current form, the study does not provide sufficient evidence to support conclusions regarding authenticity markers specific to this type of coffee.
The topic has clear potential, but the study requires further development and refinement, both in chemical analysis and scientific argumentation, to be considered a valuable contribution to the field of food authentication and coffee chemistry.
Below, I present the significant gaps that should be filled:
- The manuscript title is not adequately aligned with the content of the study. It implies that the investigation focuses solely on elephant coffee, whereas the actual study compares three coffee types (conventional, civet, and elephant). Furthermore, the stated goal in the abstract and introduction is to identify authenticity markers for elephant coffee. However, the authors explicitly acknowledge in the conclusions that such markers could not be identified and that differentiation remains inconclusive. The title should be revised to more accurately reflect the comparative nature and exploratory aim of the research.
- The abstract lacks precision. Important elements such as the use of PCA, sample composition and origin, and the main conclusion — namely, that no definitive authenticity marker was identified — are omitted. These points are essential for an accurate and transparent summary of the study.
- The research aim is overpromised in the abstract and introduction. The objective is stated as "to identify potential authenticity markers of elephant coffee," suggesting a strong applied focus. However, the results do not support this goal: no biomarkers were found, no clear distinction was achieved, and no predictive models were developed.
Suggested revision of the aim:“The aim of this study was to compare the chemical composition and antioxidant activity of elephant dung coffee with civet and conventional arabica coffee, and to explore the potential of selected phenolic and volatile compounds as discriminating features for authenticity assessment.”
- The introduction is generally appropriate but requires refinement in several areas:
- It is overly general, with disproportionate space devoted to basic coffee facts at the expense of the scientific problem.
- The research gap is not clearly defined. The authors do not explain why existing authentication methods are insufficient.
- Prior studies on coffee authentication are barely addressed. Mention of civet processing is superficial and lacks reference to relevant analytical, sensory, or metabolomic studies.
It belongs:
- Clearly define the research gap and limitations of previous authentication methods.
- Relate the study explicitly to the practical problem of coffee adulteration and the lack of verification tools.
- Justify the novelty of the current study by pointing to limitations in prior work (e.g., lack of comparative samples, absence of multivariate statistics).
- The “Materials and Methods” section requires essential additions to ensure reproducibility and scientific rigor:
- The number of analyzed samples and biological replicates is not reported. It is unclear whether each coffee type was represented by a single commercial batch or multiple samples.
- Critical information on roasting parameters, moisture content, and storage conditions is missing.
- Analytical methods (HPLC-UV, GC-SPME-MS) are described generally, but key validation data such as LOD, LOQ, linearity range, correlation coefficients, and precision (RSD) are absent. This information is necessary to assess the reliability of both quantitative and qualitative results.
- For antioxidant assays (DPPH, ABTS, ORAC, TPC, TFC), the choice of extract concentrations is not justified, matrix effects are not discussed, and methods for calculating results (e.g., ORAC AUC) are not detailed.
- The number of replicates (N) used in ANOVA and PCA is not specified. For PCA, it should be clarified which variables were included (e.g., volatiles only or also phenolics), and whether data were scaled or normalized. Information about cross-validation is also missing.
- The results are mostly clearly structured, but several improvements are recommended:
- Antioxidant activity and phenolic content - adding selected bar charts (with standard error bars) would enhance clarity and comparability. The number of biological replicates should be specified in table captions.
- Caffeine and phenolics (HPLC-UV) - include RSD values and specify whether data represent technical or biological replicates. Comment on whether differences in chlorogenic acid content are statistically significant, as this is emphasized in the discussion.
- Volatile profile (GC-MS): Supplementary data should include a complete list of identified volatiles with CAS numbers, NIST IDs, and retention times. The current selection of compounds appears somewhat arbitrary — please clarify selection criteria (e.g., abundance, statistical relevance, literature).
- Statistical analysis (PCA) - the manuscript contains a broken figure reference (“Error! Reference source not found.”) that must be corrected. Clarify which variables were included in PCA, how many observations were analyzed, and whether data were standardized. The PCA plot should label key compounds driving the separation of groups.
- The discussion includes several insightful interpretations but requires better structure and critical reflection:
- Although the stated aim was to identify authenticity markers, the authors do not sufficiently address why this goal was not achieved. It is important to clearly state that no definitive markers for elephant coffee authentication were identified, and to explain potential reasons (e.g., limited sample size, high biological variability, insufficient sensitivity of methods).
- Some interpretations - such as the effects of digestive tract length, microbial fermentation, or compound degradation - are speculative and lack supporting microbiological, sensory, or cited analytical data. Greater caution is advised, and references to existing studies would strengthen these points.
- The conclusions are brief and do not fully reflect the findings. While the discussion suggests an applied outcome, the authors admit that no biomarkers were successfully identified. Therefore, the conclusion should state that although certain differences in volatile profiles were observed, no compound or group of compounds can currently serve as a reliable authenticity marker. The authors are encouraged to reframe the conclusions to emphasize the exploratory and comparative nature of the study and suggest future directions (e.g., metabolomics, larger sample sets, sensory profiling).
The manuscript is generally understandable, but the quality of English requires moderate revision to meet the standards of international scientific communication. Several sections - particularly the abstract, results, and discussion - contain grammatical inconsistencies, awkward phrasing, and repetitive expressions that affect clarity and readability. Additionally, the use of terminology is occasionally imprecise or overly casual for a scientific context.
It is recommended that the manuscript be reviewed by a native speaker or a professional scientific editor
Author Response
Reviewer #3
Query: The manuscript title is not adequately aligned with the content of the study. It implies that the investigation focuses solely on elephant coffee, whereas the actual study compares three coffee types (conventional, civet, and elephant). Furthermore, the stated goal in the abstract and introduction is to identify authenticity markers for elephant coffee. However, the authors explicitly acknowledge in the conclusions that such markers could not be identified and that differentiation remains inconclusive. The title should be revised to more accurately reflect the comparative nature and exploratory aim of the research.
Response: The title has been changed according to your comments to the following title: “Comparative Analysis of Chemical Composition and Antioxidant Activity in Conventional, Civet, and Elephant Coffees: Is There a Definitive Authentication Marker of Elephant Coffee?”
Query: The abstract lacks precision. Important elements such as the use of PCA, sample composition and origin, and the main conclusion — namely, that no definitive authenticity marker was identified — are omitted. These points are essential for an accurate and transparent summary of the study.
Response: The abstract was changed based on your comment. Now it contains information on which methods were used in this study (DPPH, ORAC, ABTS, TPC, TFC, HPLC-UV and GC-MS-SPME, including the use of PCA analysis to identify differences from samples were also mentioned). It also now contains information about the samples (e.g. origin). However, the information that “no definitive authenticity marker was identified” was already there (see lines 35-37. The word significant in this sentence was changed to definitive. Otherwise, no changes were made in this regard.
Query: The research aim is overpromised in the abstract and introduction. The objective is stated as "to identify potential authenticity markers of elephant coffee," suggesting a strong applied focus. However, the results do not support this goal: no biomarkers were found, no clear distinction was achieved, and no predictive models were developed.
Response: The last paragraph of the introduction part was changed according to your suggestion. On top of that, the sentence “These results could be used to verify the authenticity of elephant coffee” was erased from the abstract.
Introduction
Query: It is overly general, with disproportionate space devoted to basic coffee facts at the expense of the scientific problem.
Response: Information on the specificity of coffee processing involving the animal gastrointestinal tract was added to further elucidate the scientific problem.
Query: The research gap is not clearly defined. The authors do not explain why existing authentication methods are insufficient.
Response: The introduction was revised based on your comment. We have clarified the shortcomings of current approaches and emphasized the need for alternatives.
Query: Prior studies on coffee authentication are barely addressed. Mention of civet processing is superficial and lacks reference to relevant analytical, sensory, or metabolomic studies.
Response: Additional information on previous studies involving civet coffee were added to the introduction for better context.
Query: Clearly define the research gap and limitations of previous authentication methods. Relate the study explicitly to the practical problem of coffee adulteration and the lack of verification tools. Justify the novelty of the current study by pointing to limitations in prior work (e.g., lack of comparative samples, absence of multivariate statistics).
Response: The complexity and limitations of both targeted and untargeted methods of coffee authentication related to coffee processing are explained. The fact that most of the methods are focused on the differences between Arabica and Robusta or place of origin is mentioned.
Materials and methods
Query: The number of analyzed samples and biological replicates is not reported. It is unclear whether each coffee type was represented by a single commercial batch or multiple samples.
Response: Each sample was represented by a single commercial batch. This is now indicated by the number N being added after each sample (i.e. N = 1; se lines 85-86).
Query: Critical information on roasting parameters, moisture content, and storage conditions is missing.
Response: The samples did not include any specific information about their roasting parameters. We attempted to contact the respective companies for more details but were informed that the roasting profiles are considered internal secrets. However, they did confirm that the coffees were roasted to a medium level. Based on comparison with coffee roast levels chart, the roasting degrees were estimated as follows: Full City for elephant coffee, City for arabica coffee, and between Full City and Vienna roast for civet coffee. Moisture content and storage conditions were added based on your comment.
Query: Analytical methods (HPLC-UV, GC-SPME-MS) are described generally, but key validation data such as LOD, LOQ, linearity range, correlation coefficients, and precision (RSD) are absent. This information is necessary to assess the reliability of both quantitative and qualitative results.
Response: Validation parameters were calculated for the HPLC-UV method and are now included in the form of Supplementary Table S1. There is now also a reference to this Table in the text. Since GC-MS-SPME was used for non-targeted analysis and no quantification of analytes was performed, validation parameters cannot be calculated for this method.
Query: For antioxidant assays (DPPH, ABTS, ORAC, TPC, TFC), the choice of extract concentrations is not justified, matrix effects are not discussed, and methods for calculating results (e.g., ORAC AUC) are not detailed.
Response: Methods for calculating results are now included in each of the antioxidant assay. The specific sample concentrations were used to fit within the lower and upper detection limits of the given method (so that the results for the samples were not too weak or strong). We think that this reason is obvious and does not need to be specifically mentioned in the material & method section. The matrix effect within the antioxidant activity methodology is not reported and present in any scientific literature. Even this point was thus (with all due respect) omitted.
Query: The number of replicates (N) used in ANOVA and PCA is not specified. For PCA, it should be clarified which variables were included (e.g., volatiles only or also phenolics), and whether data were scaled or normalized. Information about cross-validation is also missing.
Response: The number of replicates used in both the ANOVA and PCA analyses has been added to the main text. The explanation of the variables used in the PCA, as well as details regarding their normalization and the cross-validation procedure, has also been provided there.
Results
Query: Antioxidant activity and phenolic content - adding selected bar charts (with standard error bars) would enhance clarity and comparability. The number of biological replicates should be specified in table captions.
Response: Table 1 is now also re-illustrated in the form of a graph. This graph contains error bar charts and has been included (due to duplication of results) to the Supplementary materials (as supplementary figure S1). This fact is now also stated in the text. The number of biological and technical replicates are also mentioned in the graph.
Query: Caffeine and phenolics (HPLC-UV) - include RSD values and specify whether data represent technical or biological replicates. Comment on whether differences in chlorogenic acid content are statistically significant, as this is emphasized in the discussion.
Response: For each value, the relative standard deviation is now also listed in Table 2, calculated as SD/mean*100. It is also indicated here that the data represented technical replicates, i.e. those prepared from three separate extracts. It was also stressed out, that statistically different result are indicated in bold.
Query: Volatile profile (GC-MS): Supplementary data should include a complete list of identified volatiles with CAS numbers, NIST IDs, and retention times. The current selection of compounds appears somewhat arbitrary — please clarify selection criteria (e.g., abundance, statistical relevance, literature).
Response: According to your comment, a table with the volatile profiles of the samples has been added to the supplementary data as Supplementary Table S2. Retention times and CAS numbers have been included. We are not certain what is meant by “NIST IDs,” but the compounds were identified based on a comparison of calculated retention indices, derived from the retention times of n-alkanes, with reference values and fragmentation spectra from the NIST Standard Reference Database (as mentioned in the Methodology). The selection criteria for the reported compounds were based on their relative abundance and relevance according to previous literature investigating coffee composition.
Query: Statistical analysis (PCA) - the manuscript contains a broken figure reference (“Error! Reference source not found.”) that must be corrected. Clarify which variables were included in PCA, how many observations were analyzed, and whether data were standardized. The PCA plot should label key compounds driving the separation of groups.
Response: The reference has been corrected. The variables, number of samples (observations), and the standardization procedure are now described in Section 2.6 “Statistical Analysis” in response to the previous comment and are also mentioned in the figure caption in the main text. The figure has been updated to a biplot of PC2 and PC3 (which provides the best separation of the tested coffee types), which is built on the 10 of the most significant compounds contributing to the separation of the sample groups. The PC2xPC3 biplot with the complete variable set as well as 2D score plots of the first three PCs are added in Supplementary data.
Discussion
Query: Although the stated aim was to identify authenticity markers, the authors do not sufficiently address why this goal was not achieved. It is important to clearly state that no definitive markers for elephant coffee authentication were identified, and to explain potential reasons (e.g., limited sample size, high biological variability, insufficient sensitivity of methods).
Response: This is now addressed and clearly stated in the final paragraph of the discussion (see lines 397-406).
Query: Some interpretations - such as the effects of digestive tract length, microbial fermentation, or compound degradation - are speculative and lack supporting microbiological, sensory, or cited analytical data. Greater caution is advised, and references to existing studies would strengthen these points.
Response: Parts of the text containing speculations of digestive tract length, microbial fermentation, or compound degradation have been revised and various references have been added here (lines 358-360, 372-377, and 390-392). These are listed at the end of this document in the other changes section
Conclusion
Query: The conclusions are brief and do not fully reflect the findings. While the discussion suggests an applied outcome, the authors admit that no biomarkers were successfully identified. Therefore, the conclusion should state that although certain differences in volatile profiles were observed, no compound or group of compounds can currently serve as a reliable authenticity marker. The authors are encouraged to reframe the conclusions to emphasize the exploratory and comparative nature of the study and suggest future directions (e.g., metabolomics, larger sample sets, sensory profiling).
Response: conclusion was changed based on the above comments to specifically emphasize that no biomarkers were successfully identified and to emphasize future directions that subsequent research should take (see lines 425-442).
English language
Query: The manuscript is generally understandable, but the quality of English requires moderate revision to meet the standards of international scientific communication. Several sections - particularly the abstract, results, and discussion - contain grammatical inconsistencies, awkward phrasing, and repetitive expressions that affect clarity and readability. Additionally, the use of terminology is occasionally imprecise or overly casual for a scientific context.
It is recommended that the manuscript be reviewed by a native speaker or a professional scientific editor
Response: Due to time constraints, the English throughout the manuscript was revised using Grammarly. Changes were not highlighted in the manuscript.
Other changes:
Following references were added to the text:
Adadi, P.; Mensah, E.O.; Blay, B.; Ahmmed, M.K.; Sumaiya, K.; Agyei, D.; Kebede, B. Advancements in Civet Coffee Production and Analytical Techniques: From Aroma Profiling to Market Dynamics and Ethical Considerations. Trends in Food Science and Technology 2024, 154, 104772, doi:10.1016/j.tifs.2024.104772.
Jumhawan, U.; Putri, S.P.; Bamba, T.; Fukusaki, E. Quantification of Coffee Blends for Authentication of Asian Palm Civet Coffee (Kopi Luwak) via Metabolomics: A Proof of Concept. Journal of Bioscience and Bioengineering 2016, 122, 79–84, doi:10.1016/j.jbiosc.2015.12.008
Wang, X.; Lim, L.-T.; Fu, Y. Review of Analytical Methods to Detect Adulteration in Coffee. Journal of AOAC International 2020, 103, 295–305, doi:10.1093/JAOCINT/QSZ019.
Downey, G. Advances in Food Authenticity Testing; 1st ed.; Woodhead Publishing: Cambridge, UK, 2016;
Pennington, T.; Eshima, J.; Smith, B.S. Identification of Volatile Metabolites Produced from Levodopa Metabolism by Different Bacteria Strains of the Gut Microbiome. BMC Microbiology 2024, 24, doi:10.1186/s12866-024-03373-7.
Elhalis, H.; Cox, J.; Frank, D.; Zhao, J. Microbiological and Chemical Characteristics of Wet Coffee Fermentation Inoculated With Hansinaspora Uvarum and Pichia Kudriavzevii and Their Impact on Coffee Sensory Quality. Frontiers in Microbiology 2021, 12, doi:10.3389/fmicb.2021.713969.
Rappert, S.; Botsch, K.C.; Nagorny, S.; Francke, W.; Müller, R. Degradation of 2,3-Diethyl-5-Methylpyrazine by a Newly Discovered Bacterium, Mycobacterium Sp. Strain DM-11. Applied and Environmental Microbiology 2006, 72, 1437–1444, doi:10.1128/AEM.72.2.1437-1444.2006.
Shen, X.; Wang, Q.; Wang, H.; Fang, G.; Li, Y.; Zhang, J.; Liu, K. Microbial Characteristics and Functions in Coffee Fermentation: A Review. Fermentation 2025, 11, doi:10.3390/fermentation11010005.
Feng, X.; Hua, R.; Zhang, W.; Liu, Y.; Luo, C.; Li, T.; Chen, X.; Zhu, H.; Wang, Y.; Lu, Y. Comparison of the Gut Microbiome and Resistome in Captive African and Asian Elephants on the Same Diet. Frontiers in Veterinary Science 2023, 10, doi:10.3389/fvets.2023.986382.
Nagana Gowda, G.A.; Raftery, D. NMR-Based Metabolomics. Advances in Experimental Medicine and Biology 2021, 1280, 19–37, doi:10.1007/978-3-030-51652-9_2.
Round 2
Reviewer 3 Report
Comments and Suggestions for AuthorsThe Authors took into account most of the reviewer reservations, significantly improved the quality of the article and responded to all comments.
On this basis, I propose to publish a revised version of the manuscript in the journal Beverages.
