Exploring the Microbial Communities in Green Honey from Banggi Island, Sabah, Malaysia using Amplicon Sequencing Analysis
This article has been Reviewed by the following groups
Discuss this preprint
Start a discussion What are Sciety discussions?Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
The green honey food product from honeybee is produced in the forest's subsurface region on Banggi Island, Because of its uniqueness, the product attracts a great market price. Naturally the honey doesn’t acquire microorganisms due to strong chemical property and dormant nature. Major causes of microbial contamination are probable to include pollen, the digestive tracts of honeybees, sources that are extremely difficult to control include earth, air, dust, and nectar. This study was carried out to determine the microbial diversity of green honey using amplicon sequencing method, we targeted 16S rRNA V3-V4 region for bacterial species and ITS1 region for fungal diversity. Result revolved that; Bacteria include the largest number of identified taxa (256) and read (136856400), total of (118) different bacteria species identified and validated with seven distinct phyla, namely protobacteria, Firmicutes, Actinobacteriota, Bacteriodata, fusobacteriota, chloroflexota, and some unclassified phylum. Among other identified species only 11 bacterial species relative abundance is higher than one percent. Fungi is the second largest with 08 taxa identified and number of read (16217100). Six different species of fungi and only two species of yeast were identified, all the identified species were belong to two different phyla Ascomycota and Basidiomycota. Among other identified genera genus Zygosaccharomyces have relative abundance (99%) as compared to other species identified, it reflects the very high abundance of Zygosaccharomyces in green honey sample.
Article activity feed
-
Dear Habeebat Adekilekun Oyewusi, Following additional review, it is my opinion that the manuscript is currently not ready for publication. This is in part due to conclusions not being supported by the data and also, not sufficiently addressing reviewer concerns. While we welcome a resubmission in the future, in the mean time, please consider the additional review comments below. Best wishes, John.
-
Comments to Author
The authors have argued that their analysis pipeline excludes kitome contaminants or false positives, but based on their description of their methodology, this does not appear to be the case. (points 1-3 from my previous comments). The authors now argue either weakly, that "we presume that the utilization of this technique allowed for the identification of a diverse range of microorganisms, including rare or low-abundance species that may have eluded detection by other techniques" (125-127) or strongly, that "The amplicon sequencing method employed in this study allowed for the identification of all microorganisms, including bacteria and fungi, present in or interacting with the sample." (376-378). The authors should be careful not to overrepresent their results, which are, in fact, from a single …
Comments to Author
The authors have argued that their analysis pipeline excludes kitome contaminants or false positives, but based on their description of their methodology, this does not appear to be the case. (points 1-3 from my previous comments). The authors now argue either weakly, that "we presume that the utilization of this technique allowed for the identification of a diverse range of microorganisms, including rare or low-abundance species that may have eluded detection by other techniques" (125-127) or strongly, that "The amplicon sequencing method employed in this study allowed for the identification of all microorganisms, including bacteria and fungi, present in or interacting with the sample." (376-378). The authors should be careful not to overrepresent their results, which are, in fact, from a single sequencing experiment that lacked key controls. Additional points, which I hope will help the authors to improve their work: 1. line 80. 108 CFU or 10e8 CFU? 2. discussion of bee/honey microbiome is improved but still incoherent in places (lines 74-109) 3. a grossly oversimplified view of 16S rRNA sequencing (esp lines 121-123). Overstating the case does not bolster the authors' results and only harms their credibility. A more nuanced view here (including known bias and artefacts) would be much more appropriate. 4. line 125 - "access to the study results" - this is unclear 5. 126-7 - not clear that the authors' methodology is sufficiently rigorous to identify "rare or low-abundance species" (see prev. concerns) 6. cut-off of ASV showing more than 10 reads (193-6) appears to be arbitrary, and it is not clear that this is sufficient to eliminate false taxa from the authors' analysis 7. lines 203-204 - "a custom script" - details? 8. Contamination Identification (213-219) is not sufficiently detailed to allow for replication of the authors' analyses 9. 230-232 - the presence of a species does not necessarily imply that it has an impact on fermentation or spoilage. (this is based on sequencing - there is not even any evidence that these microbes are alive, let alone biochemically active) 10. 237-8: see prev. comment 11. 239-243 - do not appear to be supported by the authors' results, these statements are speculative 12. 263-272 - see my comments on prev. drafts regarding the inability to draw conclusions about abundance based on these data 13. Figures 2 and 3 - rather difficult to read, and likely inaccessible to anyone with red/green colourblindness 14. 310 - Bacillus is in fact Gram positive, not Gram negative. The authors cite a number of supposed health benefits (which may be true for certain species such as, for example, B. subtilis var. natto) - however, the genus also contains a number of pathogens (e.g., B. cereus and B. anthracis). The authors must be careful not to cherry-pick data that support a particular hypothesis (e.g. the health benefits of this honey) 15. Conclusions (372-389) often not substantiated; for example, the authors claim that they have identified "all microorganisms" "present in or interacting with" the sample, which seems unlikely. Claims that the honey is safe for consumption based on sequencing analysis (lines 384-5) appear equally unsubstantiated (there are plenty of substances unsafe for human consumption that wouldn't necessarily appear to be so based on the results of DNA sequencing).
Please rate the manuscript for methodological rigour
Poor
Please rate the quality of the presentation and structure of the manuscript
Satisfactory
To what extent are the conclusions supported by the data?
Not at all
Do you have any concerns of possible image manipulation, plagiarism or any other unethical practices?
No
Is there a potential financial or other conflict of interest between yourself and the author(s)?
No
If this manuscript involves human and/or animal work, have the subjects been treated in an ethical manner and the authors complied with the appropriate guidelines?
Yes
-
-
Thank you for your returned manuscript. It has been well improved as noted by both reviewers. However, reviewer 2 still has substantial concerns around appropriate controls for the Metabarcoding experiment and these concerns were not suitably address in the returned manuscript. Please consider the points made in their review. ACMI fundamentally is a sound science platform. As a result methodological rigour is of key importance. Reviewer 2 has offered both an experimental and non-experimental option to address this, both are acceptable.
-
Comments to Author
This manuscript has been significantly improved from the first submission and is much closer to being ready for publication - my remaining suggested amends are much more minor and are provided below: Contractions should be avoided Line 75 - the paper could just be reference without naming the specific author and paper, e.g., this has been shown [REF] as opposed to this has been shown by name et al., [ref] Data should be referred to as OTUs or ASVs, whichever is correct - if they are clustered to ASVs then it is not appropriate to apply to them as OTUs. Most abundant may be a better way to refer to taxa as opposed to dominant, as dominance infers the role in the community. Figure three the taxonomic levels in the middle could be superimposed on the out of date phyla names to improve neatness and …
Comments to Author
This manuscript has been significantly improved from the first submission and is much closer to being ready for publication - my remaining suggested amends are much more minor and are provided below: Contractions should be avoided Line 75 - the paper could just be reference without naming the specific author and paper, e.g., this has been shown [REF] as opposed to this has been shown by name et al., [ref] Data should be referred to as OTUs or ASVs, whichever is correct - if they are clustered to ASVs then it is not appropriate to apply to them as OTUs. Most abundant may be a better way to refer to taxa as opposed to dominant, as dominance infers the role in the community. Figure three the taxonomic levels in the middle could be superimposed on the out of date phyla names to improve neatness and accuracy within the figure. Family names need italicising
Please rate the manuscript for methodological rigour
Good
Please rate the quality of the presentation and structure of the manuscript
Satisfactory
To what extent are the conclusions supported by the data?
Partially support
Do you have any concerns of possible image manipulation, plagiarism or any other unethical practices?
No
Is there a potential financial or other conflict of interest between yourself and the author(s)?
No
If this manuscript involves human and/or animal work, have the subjects been treated in an ethical manner and the authors complied with the appropriate guidelines?
Yes
-
Comments to Author
Overall the manuscript by Ullah et al. has been much improved from the first submission; the authors' efforts to provide a more comprehensive introduction and describe their results more clearly is noted and much appreciated. However, I feel that some of my concerns raised in the initial review (detailed below, with the initial concern/authors' response/my reply) have not been adequately addressed. In particular, I am concerned about points 1 and 2: the biases and technical limitations present in a metabarcoding experiment that may mean that there is not a direct and accurate correlation between the abundance of sequencing reads and the abundance of species in the initial sample; and the lack of controls to eliminate "kitome" contaminants. Major concerns: 1. The authors present relative abundance …
Comments to Author
Overall the manuscript by Ullah et al. has been much improved from the first submission; the authors' efforts to provide a more comprehensive introduction and describe their results more clearly is noted and much appreciated. However, I feel that some of my concerns raised in the initial review (detailed below, with the initial concern/authors' response/my reply) have not been adequately addressed. In particular, I am concerned about points 1 and 2: the biases and technical limitations present in a metabarcoding experiment that may mean that there is not a direct and accurate correlation between the abundance of sequencing reads and the abundance of species in the initial sample; and the lack of controls to eliminate "kitome" contaminants. Major concerns: 1. The authors present relative abundance for different microbial taxa based on the results of a metabarcoding experiment, with the implicit assumption that the abundance of sequencing reads can be directly correlated with the abundance of species present in the initial sample. This is far from true, c.f. for example: Lamb PD, Hunter E, Pinnegar JK, Creer S, Davies RG, Taylor MI. How quantitative is metabarcoding: A meta-analytical approach. Mol Ecol. 2019 Jan;28(2):420-430. doi: 10.1111/mec.14920. Shelton AO, Gold ZJ, Jensen AJ, D Agnese E, Andruszkiewicz Allan E, Van Cise A, Gallego R, Ramón-Laca A, Garber-Yonts M, Parsons K, Kelly RP. Authors' response: Yes. This time we do analysis based on SILVAngs. My reply: The authors' response does not at all address point 1, which is that there may not be a direct correlation between the sequencing read abundance and species abundance in the honey sample. 2. The authors do not mention any controls for their metabarcoding experiment, for example, to rule out contaminating organisms that are part of the "kitome", or to rule out false positives, see e.g.: Salter SJ, Cox MJ, Turek EM, Calus ST, Cookson WO, Moffatt MF, Turner P, Parkhill J, Loman NJ, Walker AW. Reagent and laboratory contamination can critically impact sequencebased microbiome analyses. BMC Biol. 2014 Nov 12;12:87. doi: 10.1186/s12915-014-0087-z. Weiss S, Amir A, Hyde ER, Metcalf JL, Song SJ, Knight R. Tracking down the sources of experimental contamination in microbiome studies. Genome Biol. 2014 Dec 17;15(12):564. doi: 10.1186/s13059-014-0564-2. Ficetola GF, Taberlet P, Coissac E. How to limit false positives in environmental DNA and metabarcoding? Mol Ecol Resour. 2016 May;16(3):604-7. doi: 10.1111/1755-0998.12508. Authors' response: the microbe observe not part of kitome. N work was performed with ntc showing no band My reply: This reply appears to indicate that the authors included a step in their experiments where they ran the samples on an (agarose?) gel, including a no-template control (ntc) and were not able to see a band? This is 1) not detailed in their methods section, 2) not sufficient to rule out the presence of "kitome" reads - the next-generation sequencing methods having a much greater sensitivity than visualisation of DNA bands. 3. The introduction needs to be more detailed, giving a clearer understanding of what is known about the microbes found in honey samples, the physicochemical conditions, whether many honey-associated microbes are unculturable, etc. Authors' response: Points noted, and improvement was made accordingly. My reply: The improved introduction is noted, and it gives the reader much better context for understanding this paper. 4. Methods - A more detailed description of the sample is required. How/when/where was the honey collected? 5. Bioinformatics methods are not sufficiently detailed to permit replication Authors' response: Improvement was made accordingly My reply: The methods section is much improved. However, there now appear to be experiments which are not described in the methods section at all (e.g., the results presented in Figure 2 - it is not clear how this chart was generated or if this analysis could be reproduced by another scientist).
Please rate the manuscript for methodological rigour
Poor
Please rate the quality of the presentation and structure of the manuscript
Satisfactory
To what extent are the conclusions supported by the data?
Partially support
Do you have any concerns of possible image manipulation, plagiarism or any other unethical practices?
No
Is there a potential financial or other conflict of interest between yourself and the author(s)?
No
If this manuscript involves human and/or animal work, have the subjects been treated in an ethical manner and the authors complied with the appropriate guidelines?
Yes
-
-
The reviewers have highlighted major concerns with the work presented. Please ensure that you address their comments. Please provide more detail in the Methods section and ensure that software is consistently cited and its version and parameters included. The reviewers raise concerns regarding the scientific rigour and experimental design of the work. The reviewers believe the results shown in the manuscript do not support the conclusions presented.
-
Comments to Author
Ullah et al use a metabarcoding approach to analyse the bacterial and fungal communities present in a sample of green honey. While it may be of general ecological interest, or specific economic interest, to describe the microbes present in green honey, the study as conducted has several major methodological flaws that limit the conclusions that can be correctly drawn from the authors' experiment. 1. Methodological rigour, reproducibility and availability of underlying data I have major concerns about the rigor and reproducibility of this work (outlined in points 1, 2, 4 and 6 below). The authors have deposited the underlying data in the SRA and it is accessible there. 2. Presentation of results The presentation of results could be improved (see major point 6 for improving the discussion of the …
Comments to Author
Ullah et al use a metabarcoding approach to analyse the bacterial and fungal communities present in a sample of green honey. While it may be of general ecological interest, or specific economic interest, to describe the microbes present in green honey, the study as conducted has several major methodological flaws that limit the conclusions that can be correctly drawn from the authors' experiment. 1. Methodological rigour, reproducibility and availability of underlying data I have major concerns about the rigor and reproducibility of this work (outlined in points 1, 2, 4 and 6 below). The authors have deposited the underlying data in the SRA and it is accessible there. 2. Presentation of results The presentation of results could be improved (see major point 6 for improving the discussion of the results, and minor point 6 for making the figures more accessible to colour blind readers.) 3. How the style and organization of the paper communicates and represents key findings The style and organisation of the paper is adequate to communicate the findings of the work, although the language in some places is not clear or precise enough to follow easily (minor point 1). The authors sometimes present conclusions that are not supported by the results that they present (e.g. minor points 3, 7, 8 11-14). The authors should be more careful and rigorous in describing their results. 4. Literature analysis or discussion The authors could do a better job of putting the study in the context of the relevant literature (major points 3 and 6 below). 5. Any other relevant comments Please see the points described below. Major concerns: 1. the authors present relative abundance for different microbial taxa based on the results of a metabarcoding experiment, with the implicit assumption that the abundance of sequencing reads can be directly correlated with the abundance of species present in the initial sample. This is far from true, c.f. for example: Lamb PD, Hunter E, Pinnegar JK, Creer S, Davies RG, Taylor MI. How quantitative is metabarcoding: A meta-analytical approach. Mol Ecol. 2019 Jan;28(2):420-430. doi: 10.1111/mec.14920. Shelton AO, Gold ZJ, Jensen AJ, D Agnese E, Andruszkiewicz Allan E, Van Cise A, Gallego R, Ramón-Laca A, Garber-Yonts M, Parsons K, Kelly RP. Toward quantitative metabarcoding. Ecology. 2023 Feb;104(2):e3906. doi: 10.1002/ecy.3906. Gold Z, Shelton AO, Casendino HR, Duprey J, Gallego R, Van Cise A, Fisher M, Jensen AJ, D'Agnese E, Andruszkiewicz Allan E, Ramón-Laca A, Garber-Yonts M, Labare M, Parsons KM, Kelly RP. Signal and noise in metabarcoding data. PLoS One. 2023 May 11;18(5):e0285674. doi: 10.1371/journal.pone.0285674. 2. The authors do not mention any controls for their metabarcoding experiment, for example to rule out contaminating organisms that are part of the "kitome", or to rule out false positives, see e.g.: Salter SJ, Cox MJ, Turek EM, Calus ST, Cookson WO, Moffatt MF, Turner P, Parkhill J, Loman NJ, Walker AW. Reagent and laboratory contamination can critically impact sequence-based microbiome analyses. BMC Biol. 2014 Nov 12;12:87. doi: 10.1186/s12915-014-0087-z. Weiss S, Amir A, Hyde ER, Metcalf JL, Song SJ, Knight R. Tracking down the sources of experimental contamination in microbiome studies. Genome Biol. 2014 Dec 17;15(12):564. doi: 10.1186/s13059-014-0564-2. Ficetola GF, Taberlet P, Coissac E. How to limit false positives in environmental DNA and metabarcoding? Mol Ecol Resour. 2016 May;16(3):604-7. doi: 10.1111/1755-0998.12508. 3. the introduction needs to be more detailed, giving a clearer understanding of what is known about the microbes found in honey samples, the physicochemical conditions, whether many honey-associated microbes are unculturable, etc. 4. Methods - more detailed description of the sample is required. How/when/where was the honey collected? 5. bioinformatics methods are not sufficiently detailed to permit replication 6. Results: the listing of organisms detected in the sample, followed by a seemingly random list of properties of selected organisms, is not very informative or helpful. Perhaps it would be more helpful to frame the results by discussing the different species present in the authors' honey sample, in terms of what is known about the resistance of these species to the osmotic conditions in honey (and other environmental stresses), and what is known about the environmental niches of each organism (how they might have gotten into the honey) Minor concerns: 1. Needs checking for grammar and spelling throughout, and could be written a little more clearly, e.g.: -"including its ability to eliminate leeches" (line 57) -not clear if this is a medicinal treatment or the plant itself being free of leeches -"the study results" (line 79) - not clear if this refers to the present study or another? 2. line 77 - reference needed (widely acknowledged by whom?) 3. The data presented do not support the conclusion "These findings suggest that both bacteria and fungi may play important roles in honey production and preservation" (lines 154-5): simply finding the bacteria/fungi present does not necessarily suggest that they play a role in production or preservation 4. I am not sure that taking an average GC % for all bacteria and all fungi is useful here (lines 155-9) 5. I am not sure that the majority of proteobacteria are pathogenic (lines 175-6) or that the authors have clearly shown the relevance of this statement to their work. Are some of the bacteria they have identified in their sample known to be pathogens? 6. Figures 1-4: suggest using another colour scale for the figures, as the green/red colour combination may not be accessible to some readers. 7. The authors say that "Surprisingly, most of the genera belong to uncultured bacteria (36.74%) and the second highest are unclassified bacteria (15.15%)." (lines 199-200), but it is not clear why this result is surprising 8. Unclear whether the medical benefits of Bacillus are relevant here (and reference 16 seems to be about plant endophytes) (lines 203-205) 9. Pseudomonadaceae a diverse genus, I am not sure that the author's description (lines 206-7) does justice to the wide variety of pseudomonads 10. line 235: reference needed for previous detection of Zygosaccharomyces in green honey samples 11. the authors state that "All possible organisms could be identified using the amplicon sequencing method used in this study" but I am not sure this is true - some organisms may have been missed (false negatives). 12. "In fungal amplicon sequencing (ITS1) the sample was highly abundant with genus Zygosaccharomyces" (258-9): to reiterate from the above points, it is not clear to me that claims about abundance can be drawn based on the authors' methods/data. 13. "So, by looking at pictures of microbes in Green Honey sample we can conclude the Green Honey natural source of beneficial bacteria and other microbes and safe for human usage" (260-261). I am not sure this conclusion is well supported by the data presented by the authors. 14. The authors write, "We also have the highest number of uncultured and unclassified species, so it reflects that the amplicon sequencing could not be able to present the whole functional structure, or the sample have some species which unable to culture yet." (262-264) - firstly, this is not written very clearly, and secondly, the authors should not draw conclusions about culturability based on their data (they have made no attempt to culture these organisms and do not know if they are culturable or not) 15. The references need to be carefully checked: for example, the reference for the barcodes used in the PCR primers is to Glenn, C.R., et al. Evidence base update of psychosocial treatments for self-injurious thoughts and behaviors in youth. Journal of Clinical Child & Adolescent Psychology 2019; 48(3): 357-392.
Please rate the manuscript for methodological rigour
Poor
Please rate the quality of the presentation and structure of the manuscript
Satisfactory
To what extent are the conclusions supported by the data?
Not at all
Do you have any concerns of possible image manipulation, plagiarism or any other unethical practices?
No
Is there a potential financial or other conflict of interest between yourself and the author(s)?
No
If this manuscript involves human and/or animal work, have the subjects been treated in an ethical manner and the authors complied with the appropriate guidelines?
Yes
-
Comments to Author
1. Methodological rigour, reproducibility and availability of underlying data The paper as presented represents a solid methodological framework for the characterisation of the honey microbiome. Key changes to enhance the methodology would be: - Expansion on the details of the DNA extraction - if this is a novel method then this should be given more weight as this is an exciting development to help future researchers. - Use of an additional/alternate bioinformatics pipeline to try and improve the classification of the unknown sequences. 2. Presentation of results The results would benefit from the following changes: - Updating of the phyla names to the most recent changes (e.g., Firmicutes as Bacilliota) - Combining the graphs showing the bacterial diversity into one major figure, as it may help …
Comments to Author
1. Methodological rigour, reproducibility and availability of underlying data The paper as presented represents a solid methodological framework for the characterisation of the honey microbiome. Key changes to enhance the methodology would be: - Expansion on the details of the DNA extraction - if this is a novel method then this should be given more weight as this is an exciting development to help future researchers. - Use of an additional/alternate bioinformatics pipeline to try and improve the classification of the unknown sequences. 2. Presentation of results The results would benefit from the following changes: - Updating of the phyla names to the most recent changes (e.g., Firmicutes as Bacilliota) - Combining the graphs showing the bacterial diversity into one major figure, as it may help with identifying the members of which taxa are present as the individual taxonomic levels are identified. - Table 3 is a good representation of the fungus data - inclusion of read numbers would advance this and then using this design in the bacterial diversity table would also be beneficial. 3. Literature analysis or discussion The discussion would benefit from expansion in its consideration of the different abundant taxa that the amplicon sequencing has identified. It would also benefit from less generalisation of taxa at the phylum/higher taxonomic levels 5. Any other relevant comments Minor edits with specific lines are provided below: 36 revolved may not be the correct word here 37 reads (n), with a total of Consistent capitalisation for taxa and new taxa names e.g., Pseudomonodata instead of Proteobacteria Inclusion of a summary sentence at the end of the abstract Line 62 - the sentence order would benefit from rearranging line 59 - inclusion of algae seems a bit unconnected from the other text line 64 - low water activity also relevant to mention here Lines 66-69 and lines 70 - 73 could possibly be combined as both contain relevant but similair information line 77 than culture based methodologies as opposed to traditional Line 74 - fungal diversity could also be mentioned here Line 121 - to clarify, the OTUs were identified via Blast searches against the NCBI datbase? As this has resulted in many of the sequences being classified as uncultured or unclassifiable, would an alternate system of classification be possible, such as comparison to a taxonomic database such as SILVA? As the abundance of these sequences does bias the characterisation of the community slightly. NCBI blast will also randomly order sequences of the same score, which can complicate classification. Results Line 146 - most abundant line 148 - would be a key focus for future research? Line 152 - what is meant by spots? Line 155 - This discussion of the GC content would possibly benefit from expansion? Line 176 - this is too much of an overgeneralisation at the phylum level - discussions like this are most effective at lower taxonomic levels Figure 1-3 could these possibly be combined into one figure for ease of analysis? even if inserted alongside each other? Line 192 - brackets not needed Genus names need italics 206 - this is not resolved to genus level, which is fine, but needs describing as an OTU that could only be classified to family level The genus discussion section would benefit from expansion and more discussion of the most abundant genera. Acidovorax is also very closely related to Comamomas, so it is hard to describe the entire genus as pathogenic. Line 218 Only two species of Ascoymycota were identified Table 3 - this is good, the bacteria would benefit from a similair table and both would benefit from the number of reads for each group. Line 252 - taxonomic composition of different species
Please rate the manuscript for methodological rigour
Satisfactory
Please rate the quality of the presentation and structure of the manuscript
Satisfactory
To what extent are the conclusions supported by the data?
Partially support
Do you have any concerns of possible image manipulation, plagiarism or any other unethical practices?
No
Is there a potential financial or other conflict of interest between yourself and the author(s)?
No
If this manuscript involves human and/or animal work, have the subjects been treated in an ethical manner and the authors complied with the appropriate guidelines?
Yes
-