Transcriptome Analysis of the Necrotrophic Pathogen Alternaria brassicae Reveals Insights into Its Pathogenesis in Brassica juncea

ABSTRACT Alternaria blight or leaf spot caused by Alternaria brassicae has an enormous economic impact on the Brassica crops grown worldwide. Although the genome of A. brassicae has been sequenced, little is known about the genes that play a role during the infection of the host species. In this study, the transcriptome expression profile of A. brassicae during growth and infection was determined. Differential expression analysis revealed that 4,430 genes were differentially expressed during infection. Weighted gene coexpression network analysis helped identify 10 modules, which were highly correlated with growth and infection. Subsequent gene ontology (GO) enrichment analysis of the modules highlighted the involvement of biological processes such as toxin metabolism, ribosome biogenesis, polysaccharide catabolism, copper ion transport, and vesicular trafficking during infection. Additionally, 200 carbohydrate-active enzymes (CAZymes) and 80 potential effectors were significantly upregulated during infection. Furthermore, 18 secondary metabolite gene clusters were also differentially expressed during infection. The clusters responsible for the production of destruxin B, brassicicene C, and HC-toxin were significantly upregulated during infection. Collectively, these results provide an overview of the critical pathways underlying the pathogenesis of A. brassicae and highlight the distinct gene networks that are temporally regulated. The study thus provides novel insights into the transcriptional plasticity of a necrotrophic pathogen during infection of its host. Additionally, the in planta expression evidence for many potential effectors provides a theoretical basis for further investigations into the effector biology of necrotrophic pathogens such as A. brassicae. IMPORTANCE Alternaria brassicae is a necrotrophic pathogen that can infect almost all members of the Brassicaceae family. A. brassicae causes extensive yield losses in oilseed mustard and has practically restricted the cultivation of oilseed brassicas in regions with cool and foggy climatic conditions (foothills and mountainous terrains) where the severity of the pathogen is the highest. In this study, I identified the differentially expressed genes associated with the pathogenicity of A. brassicae through transcriptome sequencing. Also, I have been able to delineate pathways that are active during the early and late stages of infection. Consequently, this study has provided crucial insights into the molecular mechanisms underlying the pathogenesis of A. brassicae, an important necrotrophic pathogen.

In this study, the author investigated the differentially expressed genes during growth and infection in the Alternaria blight or leaf spot pathogen Alternaria brassicae. The author listed the biological processes after GO and WGCNA analysis. Improving the understanding of the in planta expression is an important research area. However, a substantial amount of functional analysis should be accomplished. There is a general lack of description in approaches used and there are minimal details provided in the Methods or Figure Legends. Some specific issues are outlined below: 1 The author performed transcriptome sequencing at 2 or 4 days post inoculation, designated as the early and late time points of infection. But the author did not provide any evidence to prove the sampling time points represented as the early and late infection. In addition, the sampling should be described. The mycelia from PDA plate included the aerial and feeding hyphae, which introduced the variety of differentially expressed genes. 2 When analyzed the upregulated genes at 2 and 4 dpi, the auther did not mention the common genes up-or down-regulated at2 and 4 dpi. This should be addressed here and as the in planta expression needed. 3 The author state that "In contrast, three genes, ABRSC02.163, ABRSC01.1330, and ABRSC09.497, were significantly downregulated at 2 dpi but were upregulated at 4 dpi" However, it doesn't appear that this data has been indicated. 4 For GO and WGCNA analysis, the author only list the gene modules, but not described their differences at different stages and with different expression level. What are their expression patterns during infection? What are those genes function in infection? 5 For the candidate genes, is the GO and WGCNA analysis important to identify those genes? What is the relationship of your network analysis with the biological process and those pathogenesis genes? The list of these genes has been described in many papers, and there were nothing novel. What are the common or specific pathways /module in the necrotrophic pathgens? Comment: When analyzed the upregulated genes at 2 and 4 dpi, the auther did not mention the common genes up-or down-regulated at2 and 4 dpi. This should be addressed here and as the in planta expression needed.

Response:
The common genes up-or down-regulated at 2 and 4 dpi have been depicted in Figure 3A as a Venn diagram and the top up-regulated genes in 2 and 4 dpi are depicted in Figure 3B as a heatmap and also in Table S2 and Table S3. Additionally, the common and unique genes are explained in Page 5, Line 120-127.

Comment:
The author state that "In contrast, three genes, ABRSC02.163, ABRSC01.1330, and ABRSC09.497, were significantly downregulated at 2 dpi but were upregulated at 4 dpi" However, it doesn't appear that this data has been indicated.

Response:
The data pertaining to the above statement is indicated in Supplementary Table S2 and referenced at the end of the statement on Page 5, Lines 128 and 130.
Comment: For GO and WGCNA analysis, the author only lists the gene modules, but not described their differences at different stages and with different expression level. What are their expression patterns during infection? What are those genes function in infection?

Response:
The WGCNA was performed only for the DEGs. The modules have been correlated to the time points of infection and have been listed in the same order. The modules M1 and M2 were positively correlated to the growth on PDA plates, whereas M6, M7, and M8 were positively correlated to the early time point of infection. The M9 module was positively correlated to the late time point of infection. This has been explained in the results section in Page 6, Line 156-166.
The GO analysis indicates the functions of the genes that are overrepresented in the respective modules. The specific biological processes of the genes in specific modules have also been described in Page 6 and 7, Line 167-179. The details of the GO enrichment analysis of the modules is also given in Figure 5 and Table S8. Additionally, the model described in Figure 7 lists the specific modules and GO processes that are active at the different stages of infection.

Comment:
For the candidate genes, is the GO and WGCNA analysis important to identify those genes? What is the relationship of your network analysis with the biological process and those pathogenesis genes? The list of these genes has been described in many papers, and there was nothing novel. What are the common or specific pathways /module in the necrotrophic pathgens?

Response:
The GO and WGCN analysis is independent of the list of candidate genes discussed. The candidate genes identified were classified into CAZymes, Effectors, and secondary metabolites. Although some of these genes may have been described in other papers, the functional expression of these genes in A. brassicae is shown here for the first time. Additionally, many of the effectors described in the study have no known domain or function and have not been reported elsewhere. Effectors of necrotrophic pathogens are not well described and the current study provides a foundation for further functional characterisation of these effectors. In case of the secondary metabolites, Destruxin B as well HC-toxin gene clusters are specifically present in A. brassicae and not in other infective Alternaria species. The current study thus provides proof of these secondary metabolite clusters to be activated during infection and may therefore encourage further studies on these important secondary metabolites.
1st Revision -Editorial Decision December 6, 2022 Dr. Sivasubramanian Rajarammohan National Agri-Food Biotechnology Institute MOHALI India Re: Spectrum02939-22R1-A (Transcriptome analysis of the necrotrophic pathogen Alternaria brassicae reveals a biphasic mode of pathogenesis in Brassica juncea) Dear Dr. Sivasubramanian Rajarammohan: Three other experts reviewed your manuscript. All of them raised the issue that more time points should have been included to make the paper more significant, but in Microbiology Spectrum the impact of a manuscript is not the main point of evaluation. Reviewer 2 raised a serious question about the title of your manuscript, because it seems that this biphasic mode of infection is not well highlighted in the text. I agree with him, and strongly suggest to change the title of your manuscript. Moreover, all reviewers together raised a number of comments that you can find in the email below, and I can accept your manuscript only if an esaustive answer will be provided to all their comments. Failing to address some comments will likely result in manuscript rejection, without the possibility of further appealing.
When submitting the revised version of your paper, please provide (1) point-by-point responses to the issues raised by the reviewers as file type "Response to Reviewers," not in your cover letter, and (2) a PDF file that indicates the changes from the original submission (by highlighting or underlining the changes) as file type "Marked Up Manuscript -For Review Only". Please use this link to submit your revised manuscript -we strongly recommend that you submit your paper within the next 60 days or reach out to me. Detailed instructions on submitting your revised paper are below.

Link Not Available
Below you will find instructions from the Microbiology Spectrum editorial office and comments generated during the review.
ASM policy requires that data be available to the public upon online posting of the article, so please verify all links to sequence records, if present, and make sure that each number retrieves the full record of the data. If a new accession number is not linked or a link is broken, provide production staff with the correct URL for the record. If the accession numbers for new data are not publicly accessible before the expected online posting of the article, publication of your article may be delayed; please contact the ASM production staff immediately with the expected release date.
The ASM Journals program strives for constant improvement in our submission and publication process. Please tell us how we can improve your experience by taking this quick Author Survey. Reviewer #3 (Public repository details (Required)): The authors have produced an RNA-Seq dataset. They state it is uploaded to NCBI and provide an accession number, however it is not yet public so I am unable to verify that it will be available upon publication. The verification link provided in the previous response to reviewers does show files awaiting release, so on balance I think the author will release them, though I would prefer they were made public on submission rather than upon acceptance, particularly as a preprint has been submitted.
Reviewer #3 (Comments for the Author): The author has produced a RNA-Seq dataset of an infection time-course of Alternaria brassicae infecting Brassica juncea. The author has followed this with some initial analyses of the changes in expression and some basic functional annotations. Whilst this is an understudied pathosystem and I would follow similar initial exploratory analyses, I do think there are several improvements that can be made to the manuscript.

Major:
1) The author should explain which parameters were used for STAR as it is highly customisable. I would also like to know if a one or two pass method was used. Two-pass methods have previously been shown to improve the quantification accuracy of different splice variants of a gene.
2) I would like an explanation of why featurecounts was used. This is older software that simply pulls the number of reads aligning to a feature, whereas newer tools like stringtie feature newer algorithms that can further improve accuracy and suggest novel transcripts that could have been missed during annotation.
3) The author cites a note on the genome assembly (reference 16) of A. brassicae as being the source of the assembly and annotations, yet this note only describes the assembly. The author should provide the source of the annotations or describe how they were generated if this is novel work. -I think this is covered by reference 22, but it is not cited at line 380.
4) The author should clarify in the WGCNA methodology whether a false discovery rate was again used for multi-test correction when assessing over-representation.

Minor:
1) At line 70, the author has already explained that knowledge of infection is limited and yet does so again in the next sentence. This hurts the readability of the manuscript and could be rephrased.
2) The author frequently uses plural pronouns eg. we despite this being a single author manuscript. If others conducted the analyses they should receive appropriate credit. If not, this should be rephrased to avoid confusion to a reader.
3) I would like the author to explain in more detail how the PCA was performed and whether any parameters were passed to the analysis.
4) The author should reword mentions of effectors, CAZYs and Toxins to make clear these are previously identified rather than new in this paper.
Reviewer #4 (Public repository details (Required)): Data seems to have been deposited in NCBI Reviewer #4 (Comments for the Author): The manuscript entitled "Transcriptome analysis of the necrotrophic pathogen Alternaria brassicae reveals a biphasic mode of pathogenesis in Brassica juncea" aim to provides novel insights into the transcriptional plasticity of a necrotrophic pathogen during infection of its host.
To be honest I am a bit confused as the title clearly state: "biphasic mode of pathogenesis", therefore I was assuming that the main novelty of this work is the clear description of a new pathogenicity strategy in Alternaria brassicae. However, nothing is reported about a biphasic mode of pathogenesis in the abstract... moreover more I was reading the manuscript more I was consolidating my doubts about a biphasic mode of pathogenesis.
The author used a transcriptomic approach using 2 time points and comparing the expression profiles of Alternaria brassicae to the fungus growing on PDA. To describe a "biphasic mode of pathogenesis" I would have expected more timepoints. It's quite funny that the author found a biphasic mode using two time points... I am wondering if he could have found a triphasic mode using three time points or a fourphasic mode using four time points.
Stating that a pathogen has a biphasic mode means a clear switch between two different phases. An example of a biphasic lifestyles is normally used for pathogens such as Colletotrichum and Magnaporthe, in those systems the fungus differentiates distinctive structures and activate different genes during different stages (However also in these cases the lifestyle is much more complex). That does not mean that necrotrophic pathogens have a unique expression profile during the interaction with the host, still necrotrophic pathogens do not differentiate distinctive structures and no clear switch can be detected. Assuming that necrotrophy is a simple "kill and eat" process with no molecular cross-talk, gene modulation, and complexity is quite simplistic and old fashion.
The classification of different trophic lifestyles (like any other biological classification) is a strategy that does not reflect the complexity of biology. If the author wants to have an alternative approach to classification, maybe he/she should look at the different shades that biology has... rather than increase the number of classes.
Having said that there are also technical issues in the manuscript that should be considered. Afew examples are: The author cites their work for the identification of selected gene families like the CAZy. I checked the published work and it state "Genes were then annotated using BLAST (version 2.7.1+) against UniProt, SWISS-PROT, CAZy, MEROPS, and PHI-BASE." Not considering that nothing is stated about how the BLAST was performed (score, e-value or similarity cutoffs) I would not trust such an approach for the characterization of CAZy and peptidases as the similarity may be misleading and usually a more complex approach (sequence alignments, hidden markov models (HMMs) and phylogenies of the signature domain regions). The characterization of both those gene families is much more complex.
The author analyzes candidate effectors, CAZY and SM clusters... why not all SM associated backbone genes and no peptidases were analyzed? How did he/she select specific gene families and not others?
The manuscript is also quite hard to read and follow, an example is provided by the numerous time the author state "during infection" in the abstract or the structure of the images.
Reviewer #5 (Public repository details (Required)): The author states that the transcriptome dataset has been deposited in the NCBI BioProject database.
Reviewer #5 (Comments for the Author): In this publication, the author performed transcriptomic analysis of Alternaria brassicae on Brassica juncea during two infection time points. These data should fill a gap in the field on molecular mechanisms of pathogenicity of A. brassicae. Previous reports have focused on secondary metabolite toxins and the host response to A. brassicae.
Major comments: 1, The author responded to the comments of the reviewers by including more information in the Methods and Results section. However, the author did not clearly state if there was a mixed population of A. brassicae on the 15-day-old PDA plate at the time of sampling. 2, Regarding the timing of the infection and collecting the samples, this manuscript would benefit from an earlier time point during infection as a control, instead of an artificial media at the negative control. One could envision that collecting the sample directly after inoculation and prior to penetration at day 2, would also make the populations more consistent. 3, In this discussion, the author states that "approximately 59% of the putative effectors were expressed at least at one of the time points. This indicates that proteinaceous effectors are vital during the pathogenesis of A. brassicae." A comparison to percentage of effectors in other necrotrophic pathogens would be helpful, as would some functional assays that provide evidence of the role of effector proteins in A. brassicae pathogenicity. However, it is understandable that the latter is outside of the scope of this paper.

Preparing Revision Guidelines
To submit your modified manuscript, log onto the eJP submission site at https://spectrum.msubmit.net/cgi-bin/main.plex. Go to Author Tasks and click the appropriate manuscript title to begin the revision process. The information that you entered when you first submitted the paper will be displayed. Please update the information as necessary. Here are a few examples of required updates that authors must address: • Point-by-point responses to the issues raised by the reviewers in a file named "Response to Reviewers," NOT IN YOUR COVER LETTER. • Upload a compare copy of the manuscript (without figures) as a "Marked-Up Manuscript" file. • Each figure must be uploaded as a separate file, and any multipanel figures must be assembled into one file. For complete guidelines on revision requirements, please see the journal Submission and Review Process requirements at https://journals.asm.org/journal/Spectrum/submission-review-process. Submissions of a paper that does not conform to Microbiology Spectrum guidelines will delay acceptance of your manuscript. " Please return the manuscript within 60 days; if you cannot complete the modification within this time period, please contact me. If you do not wish to modify the manuscript and prefer to submit it to another journal, please notify me of your decision immediately so that the manuscript may be formally withdrawn from consideration by Microbiology Spectrum.
If your manuscript is accepted for publication, you will be contacted separately about payment when the proofs are issued; please follow the instructions in that e-mail. Arrangements for payment must be made before your article is published. For a complete list of Publication Fees, including supplemental material costs, please visit our website.
Corresponding authors may join or renew ASM membership to obtain discounts on publication fees. Need to upgrade your membership level? Please contact Customer Service at Service@asmusa.org.
Thank you for submitting your paper to Microbiology Spectrum.
In this publication, the author performed transcriptomic analysis of Alternaria brassicae on Brassica juncea during two infection time points. These data should fill a gap in the field on molecular mechanisms of pathogenicity of A. brassicae. Previous reports have focused on secondary metabolite toxins and the host response to A. brassicae.
Major comments: 1, The author responded to the comments of the reviewers by including more information in the Methods and Results section. However, the author did not clearly state if there was a mixed population of A. brassicae on the 15-day-old PDA plate at the time of sampling.
2, Regarding the timing of the infection and collecting the samples, this manuscript would benefit from an earlier time point during infection as a control, instead of an artificial media at the negative control. One could envision that collecting the sample directly after inoculation and prior to penetration at day 2, would also make the populations more consistent.
3, In this discussion, the author states that "approximately 59% of the putative effectors were expressed at least at one of the time points. This indicates that proteinaceous effectors are vital during the pathogenesis of A. brassicae." A comparison to percentage of effectors in other necrotrophic pathogens would be helpful, as would some functional assays that provide evidence of the role of effector proteins in A. brassicae pathogenicity. However, it is understandable that the latter is outside of the scope of this paper.
In this publication, the author performed transcriptomic analysis of Alternaria brassicae on Brassica juncea during two infection time points. These data should fill a gap in the field on molecular mechanisms of pathogenicity of A. brassicae. Previous reports have focused on secondary metabolite toxins and the host response to A. brassicae.
Major comments: 1, The author responded to the comments of the reviewers by including more information in the Methods and Results section. However, the author did not clearly state if there was a mixed population of A. brassicae on the 15-day-old PDA plate at the time of sampling.
2, Regarding the timing of the infection and collecting the samples, this manuscript would benefit from an earlier time point during infection as a control, instead of an artificial media at the negative control. One could envision that collecting the sample directly after inoculation and prior to penetration at day 2, would also make the populations more consistent.
3, In this discussion, the author states that "approximately 59% of the putative effectors were expressed at least at one of the time points. This indicates that proteinaceous effectors are vital during the pathogenesis of A. brassicae." A comparison to percentage of effectors in other necrotrophic pathogens would be helpful, as would some functional assays that provide evidence of the role of effector proteins in A. brassicae pathogenicity. However, it is understandable that the latter is outside of the scope of this paper.

Response to reviewers
Reviewer #3 (Public repository details (Required)): Comment: The authors have produced an RNA-Seq dataset. They state it is uploaded to NCBI and provide an accession number, however it is not yet public so I am unable to verify that it will be available upon publication. The verification link provided in the previous response to reviewers does show files awaiting release, so on balance I think the author will release them, though I would prefer they were made public on submission rather than upon acceptance, particularly as a preprint has been submitted.
Response: As per the reviewer's suggestion, the dataset uploaded to NCBI has been made public now.

Reviewer #3 (Comments for the Author):
The author has produced a RNA-Seq dataset of an infection time-course of Alternaria brassicae infecting Brassica juncea. The author has followed this with some initial analyses of the changes in expression and some basic functional annotations. Whilst this is an understudied pathosystem and I would follow similar initial exploratory analyses, I do think there are several improvements that can be made to the manuscript.
Major comments -Comment: The author should explain which parameters were used for STAR as it is highly customisable. I would also like to know if a one or two pass method was used. Two-pass methods have previously been shown to improve the quantification accuracy of different splice variants of a gene.
Response: I had initially compared the one and two pass methods in STAR. However, I did not find any major improvement in the number of reads that mapped to the transcripts. The marginal increase in the reads that mapped using the two-pass method resulted in the increase in the transcript count for the highly abundant transcripts only and did not map to any transcript that was not assigned any reads earlier. However, in response to the reviewer suggestion in point no. 2 I have carried out the analysis using HISAT2 + stringtie2 and this has been explained in the response to the below comment. Comment: I would like an explanation of why featurecounts was used. This is older software that simply pulls the number of reads aligning to a feature, whereas newer tools like stringtie feature newer algorithms that can further improve accuracy and suggest novel transcripts that could have been missed during annotation.
Response: I thank the reviewer for this suggestion. Even though featurecounts is an older software it is still being extensively used to generate raw counts from RNASeq data and was therefore used in this study. However, as per the reviewer's suggestion, the analysis was carried out using Stringtie2, which resulted in the discovery of transcripts that were not predicted/annotated in the previous study. Therefore, a re-annotation of the genome was carried out using this transcriptome dataset to obtain a revised version of the genome annotation of A. brassicae (V 2.0). The revised version of the annotations was used for all further analyses. Further, the transcript quantification was also carried out using stringtie2 and the transcript counts obtained from stringtie2 were used in DESeq2 for differential gene expression analysis. Although the new analysis revealed more DEGs, the functional classification, and GO enrichment did not change drastically.
Comment: The author cites a note on the genome assembly (reference 16) of A. brassicae as being the source of the assembly and annotations, yet this note only describes the assembly. The author should provide the source of the annotations or describe how they were generated if this is novel work. -I think this is covered by reference 22, but it is not cited at line 380.
Response: I apologise for this omission. This has been rectified in the revised manuscript.
Comment: The author should clarify in the WGCNA methodology whether a false discovery rate was again used for multi-test correction when assessing over-representation.
Response: The genes identified in each module in the WGCNA methodology was subject to GO enrichment analysis using a standard hypergeometric test as implemented in the GOStats package. This information has now been included in the revised manuscript. One of the underlying assumptions of the hypergeometric is that the individual terms/genes being chosen are independent. Unfortunately, this is not true of genes or GO terms. Both the genes and GO terms are interdependent (expression of one gene affecting other(s) and GO terms are set up as a directed acyclic graph). Due to this interdependency, carrying out multiple testing on GO datasets are conceptually challenging. The use of FDR is useful but it does not solve the issue of dependency. Additionally, the sampling of genes used in the analysis has already been subjected to statistical testing and multiple correction (only the DEGs and not all the genes were used for network construction using WGCNA) and the correlation of genes to their functional attributes (as determined by GO terms) is deterministic. I have also only attributed functional categories that are represented by multiple genes (in the same module) in the manuscript. Therefore, FDR correction was not applied when assessing the overrepresentation.
Minor comments -Comment: At line 70, the author has already explained that knowledge of infection is limited and yet does so again in the next sentence. This hurts the readability of the manuscript and could be rephrased.
Response: I had referred to two different pathogens at Line 68-71 viz. A. brassicicola and A. brassicae. However, as per the reviewer's suggestion, I have rephrased the sentences for better readability in the revised manuscript.
Comment: The author frequently uses plural pronouns eg. we despite this being a single author manuscript. If others conducted the analyses they should receive appropriate credit. If not, this should be rephrased to avoid confusion to a reader.
Response: I apologise for this error. All the instances of "we" has been changed to avoid confusion to the reader.
Comment: I would like the author to explain in more detail how the PCA was performed and whether any parameters were passed to the analysis.
Response: The PCA was performed as is implemented in the DESeq2 package. The raw counts were transformed using the variance stabilising transformation and the transformed values were used to generate the PCA plot using default parameters. This has been now included in the revised manuscript.
Comment: The author should reword mentions of effectors, CAZYs and Toxins to make clear these are previously identified rather than new in this paper.
Response: I thank the reviewer for pointing this out. This has been rectified in the revised manuscript.
Reviewer #4 (Public repository details (Required)): Data seems to have been deposited in NCBI Reviewer #4 (Comments for the Author): The manuscript entitled "Transcriptome analysis of the necrotrophic pathogen Alternaria brassicae reveals a biphasic mode of pathogenesis in Brassica juncea" aim to provides novel insights into the transcriptional plasticity of a necrotrophic pathogen during infection of its host.
Comment: To be honest I am a bit confused as the title clearly state: "biphasic mode of pathogenesis", therefore I was assuming that the main novelty of this work is the clear description of a new pathogenicity strategy in Alternaria brassicae. However, nothing is reported about a biphasic mode of pathogenesis in the abstract... moreover more I was reading the manuscript more I was consolidating my doubts about a biphasic mode of pathogenesis. The author used a transcriptomic approach using 2 time points and comparing the expression profiles of Alternaria brassicae to the fungus growing on PDA. To describe a "biphasic mode of pathogenesis" I would have expected more timepoints. It's quite funny that the author found a biphasic mode using two time points... I am wondering if he could have found a triphasic mode using three time points or a fourphasic mode using four time points. Stating that a pathogen has a biphasic mode means a clear switch between two different phases. An example of a biphasic lifestyles is normally used for pathogens such as Colletotrichum and Magnaporthe, in those systems the fungus differentiates distinctive structures and activate different genes during different stages (However also in these cases the lifestyle is much more complex). That does not mean that necrotrophic pathogens have a unique expression profile during the interaction with the host, still necrotrophic pathogens do not differentiate distinctive structures and no clear switch can be detected. Assuming that necrotrophy is a simple "kill and eat" process with no molecular cross-talk, gene modulation, and complexity is quite simplistic and old fashion. The classification of different trophic lifestyles (like any other biological classification) is a strategy that does not reflect the complexity of biology. If the author wants to have an alternative approach to classification, maybe he/she should look at the different shades that biology has... rather than increase the number of classes.
Response: I thank the reviewer for his/her suggestions. I agree that more time points are required to claim a "biphasic" mode of pathogenesis and have hence changed the title of the manuscript and removed all the instances of a biphasic mode in the revised manuscript. I proposed a "biphasic" mode of pathogenesis for A. brassicae not only from the transcriptome data generated in this study but also previous detailed histopathological studies that clearly showed that the pathogen had a short "biotrophic" phase where it grew within the host (intercellularly) without causing cell death. Apart from A. brassicae, this has also been reported for the broad host-range necrotrophic pathogen Sclerotinia sclerotiorum by other research groups. Although, A. brassicae does not develop very distinct structures during pathogenesis, the histopathological studies did show the formation of penetration pegs or appresoria-like structures that may perhaps indicate the development stage during which the "switch" may occur. However, this is speculative with the current information and would warrant further fine-scale transcriptome studies. I also agree with the reviewer that any type of classification does not fully reflect the complexity of biology. My intention was to draw attention to the aforementioned complexities during interactions of A. brassicae with its host and move away from the assumption of "kill & eat" process that the reviewer rightly calls as simplistic and old fashioned.
Comment: The author cites their work for the identification of selected gene families like the CAZy. I checked the published work and it state "Genes were then annotated using BLAST (version 2.7.1+) against UniProt, SWISS-PROT, CAZy, MEROPS, and PHI-BASE." Not considering that nothing is stated about how the BLAST was performed (score, e-value or similarity cutoffs) I would not trust such an approach for the characterization of CAZy and peptidases as the similarity may be misleading and usually a more complex approach (sequence alignments, hidden markov models (HMMs) and phylogenies of the signature domain regions). The characterization of both those gene families is much more complex.
Response: The CAZymes were identified by searching against the CAZy database maintained at www.cazy.org. The CAZymes and peptidases (from MEROPS) were identified using BLAST with a cutoff of percent identity > 50% and a query coverage > 50% to rule out false positive matches. Additionally, the CAZymes were also annotated by using the dbCAN web server, which uses a signature domain-based annotation. The signature domain for every CAZyme family is derived based on the CDD (conserved domain database) search and literature curation. Additionally, dbCAN also contains HMMs of the signature domains, which are used to annotate a given set of proteins. The results from the BLAST search against CAZy database and the annotations from the dbCAN server were in concordance with a few extra CAZymes that were identified in the dbCAN approach. This set of CAZymes were only investigated in the current study. A similar approach was also used for the peptidases, wherein the BLAST search was carried out using only the signature domains of the peptidases as the reference. Comment: The author analyzes candidate effectors, CAZY and SM clusters... why not all SM associated backbone genes and no peptidases were analyzed? How did he/she select specific gene families and not others?
Response: The SM-associated backbone genes were analysed and their details are given in Supplementary Table S11. Specific genes and gene families that were already reported to have a role in pathogenesis were selected and discussed. For example, pectate lyases have been reported to have a role in pathogenesis of related pathogen A. brassicicola. Even though specific gene families have been discussed, the data on expression of all CAZymes, effectors, and SM clusters have been recorded in the Supplementary Tables S9-S12. Peptidases were not analysed earlier, but as per the reviewer's suggestion this has been carried out and is included in the revised manuscript (Line nos. 237-246; Table S12).
Comment: The manuscript is also quite hard to read and follow, an example is provided by the numerous time the author state "during infection" in the abstract or the structure of the images.
Response: The manuscript has been revised for structural changes in the language and repetitions have been avoided wherever possible.
Reviewer #5 (Public repository details (Required)): The author states that the transcriptome dataset has been deposited in the NCBI BioProject database.
Reviewer #5 (Comments for the Author): In this publication, the author performed transcriptomic analysis of Alternaria brassicae on Brassica juncea during two infection time points. These data should fill a gap in the field on molecular mechanisms of pathogenicity of A. brassicae. Previous reports have focused on secondary metabolite toxins and the host response to A. brassicae.
Major comments -Comment: The author responded to the comments of the reviewers by including more information in the Methods and Results section. However, the author did not clearly state if there was a mixed population of A. brassicae on the 15-day-old PDA plate at the time of sampling.
Response: A. brassicae collected from 15-day-old PDA plate had a mixed population of growing mycelia and spores at the time of sampling. This information has now been included in the revised manuscript.
Comment: Regarding the timing of the infection and collecting the samples, this manuscript would benefit from an earlier time point during infection as a control, instead of an artificial media at the negative control. One could envision that collecting the sample directly after inoculation and prior to penetration at day 2, would also make the populations more consistent.
Response: I thank the reviewer for this suggestion. However, there are two major hindrances in collecting an earlier time point. The A. brassicae spores start to adhere strongly to the surface of the host tissue only after 24 hours of inoculation and are maximally adhered and start to penetrate the tissue at 2 dpi. I did collect samples from 24 hours and 36 hours post inoculation, however, the representation of fungal RNA from these samples was negligible and therefore could not be used for further analysis. Enrichment of fungal RNA from such samples has proved to be challenging and we are exploring other methodological improvements to sample and assay the gene expression changes in the early time points of infection.
Comment: In this discussion, the author states that "approximately 59% of the putative effectors were expressed at least at one of the time points. This indicates that proteinaceous effectors are vital during the pathogenesis of A. brassicae." A comparison to percentage of effectors in other necrotrophic pathogens would be helpful, as would some functional assays that provide evidence of the role of effector proteins in A. brassicae pathogenicity. However, it is understandable that the latter is outside of the scope of this paper.
Response: I thank the reviewer for this suggestion. The comparison of effectors (expressing during infection) from other necrotrophic pathogens has now been added to the discussion. As the reviewer pointed out the evidence of role of effectors in A. brassicae is out of scope of this paper but will surely be followed up in subsequent studies from the lab.