Immune Functional Analysis of Chitin Deacetylase 3 from the Asian Citrus Psyllid Diaphorina citri

Round 1

Reviewer 1 Report

Yu et al. have identified a cDNA for chitin deacetylase from the Asian citrus psyllid, Diaphorina citri, which they have named DcCDA3.  RT-PCR was used to examine the corresponding gene expression in the different life stages and select tissues. The gene expression was also found to be upregulated by 20-hydroxyecdysone as well as bacterial challenge. The authors also present evidence that the recombinant protein can also inhibit growth of gram-positive bacteria, suggesting that it may have a role in immunity. Overall, the analysis is thorough and the manuscript well written. However, I have concerns if the proper substrate was used for enzymatic assays as well as the purity of the recombinant protein for these experiments (points 8 and 9 below).

1) Although the phylogenetic analysis suggests that the D. citri protein may be most similar to N. lugens CDA3, a bootstrap value of 50 is not very strong and all of the other Group III proteins are classified as CDA4, therefore, why not name the D. citri gene/protein CDA4 as well?

2) Lines 84-94: In the section titled "Bioinformatic analysis ..." I would like to see an analysis of residues important for catalytic activity to see whether or not DcCDA3 is capable of deacetylating chitin. (Also see points 8 and 9 below.)

3) Lines 96 and 97: the sequence shown in Figure 1 should be referred to as the cDNA and not the gene.

4) Line 133: Delete "were" from the sentence "To analyse whether DcCDA3 were expression...".

5) The statistical analysis shown in Fig. 3 is ambiguous as it is not known between which developmental phases the difference is significant. For example, although first instar nymphs have a significant difference at p < 0.01, it is not obvious which of the different developmental phases this difference applies to. Presentation of the statistical analysis should be similar to that of enzyme activity as shown in Fig. 7.

6) Section 2.3, lines 132 to 138: No mention of Fig. 4 is made in the text. Additionally, as significant differences were observed at 1, 12, and 18 hours post 20E application, the authors could comment on whether 20E acts directly (a direct target) or indirectly (a downstream target) on the CDA3 gene.

7) Section 2.6, lines 176-181: No reference to the SDS-PAGE figure is made in the text. Also, two figures are labeled "Figure 6" ("The expression levels of DcCDA3 in the midgut and fat body ..." and "Analysis of the recombinant DcCDA3 protein using SDS-PAGE").

8) Line 189: The authors should give a reference for the use of 4-nitroacetanilide as a valid substrate for deacetylase activity as only recently have insect CDAs been shown to have enzymatic activity and these reports have used various forms of chitin (for example see Liu et al, Journal of Insect Physiology 113 (2019) 42-48, and Liu et al, Journal of Biological Chemistry (2019) 294: 5774-5783).  If 4-nitroacetanilide has previously been used for CDA activity (for example with fungal or bacterial CDAs) it would be beneficial to reference that here.

9) The purified protein shown in the SDS-PAGE figure contains multiple contaminating proteins; the authors should perform a parallel expression/purification using an empty vector and subsequent activity assay to show that one of these proteins was not responsible for the activity described in the text.

10) Line 192: Figure 7 shows the maximum activity at 50 ^oC and not 55 ^oC.

11) In Figure 8: Bacillus is misspelled as "Bacilius".

12) Lines 294-295: I do not understand the statement "D. citri adult was precooled in the ice to escape".

13) Line 299: Given the small nature of the insects, a description of how the fat body was collected is appropriate.

14) Line 304: It is not clear here if the dilutions of 20E were made with DEPC water or acetone; presumably with DEPC water as this was used for the control but it should be explicitly stated to avoid confusion.  Additionally, the concentration of the 20E stock solution should be given so that the final concentration of acetone can be determined.

15) Line 321: What query sequence was used for the BLASTX search?

16) Lines 327-328: A description of the phylogenetic analysis should include a citation to Table S1.

17) Line 394: "induced up-regulated" should be either "induced" or "up-regulated" but not both.

Author Response

Although the phylogenetic analysis suggests that the citri protein may be most similar to N. lugens CDA3, a bootstrap value of 50 is not very strong and all of the other Group III proteins are classified as CDA4, therefore, why not name the D. citri gene/protein CDA4 as well?

Reply: Thanks for reviewer’s valuable and thoughtful comments. In insect, CDAs can be divided into five groups (I-V) based on the presence and order of the domains and the overall degree of protein sequence similarity. Group I and Group Ⅱ CDAs contain three domains, including CBD domain, LDLa domain and CDA domain. Group Ⅲ and Group Ⅳ CDAs contain CBD and CDA domains, and Group V CDAs retained only the CDA domain. In this study, we found that CDA3 in Tribolium castaneum, Bombyx mori, Bactrocera dorsalis, Anopheles gambiae, and Dropsophila melanogaster are grouped into Group Ⅱ except for Nilaparvata lugens. Based on the species similarity and evolutionary relationship, genetic relationship of Diaphorina citri is the most closely related to N. lugens. In previous report, N. lugens CDA3 belonged to Group Ⅲ, and showed high homology with D. melanogaster CDA3 and T. castaneum CDA3 (Chitin deacetylase family genes in the brown planthopper, Nilaparvata lugens (Hemiptera: Delphacidae)). Therefore, we named the D. citri CDA3. In addition, we have identified the CDA4 from genome and transcriptome database, and the phylogenetic analysis revealed that D. citri CDA4 kept high homology with N. lugens CDA4 (unpublished data).

Lines 84-94: In the section titled "Bioinformatic analysis ..." I would like to see an analysis of residues important for catalytic activity to see whether or not DcCDA3 is capable of deacetylating chitin. (Also see points 8 and 9 below.)

Reply: Thanks for reviewer’s thoughtful comments. According to domain analysis by SMART software, DcCDA3 contained a polysaccharide deacetylase-like catalytic domain (amino residues 139-267). As described in Figure 1, the double underline indicates the CDA domain.

Lines 96 and 97: the sequence shown in Figure 1 should be referred to as the cDNA and not the gene.

Reply: Thanks for reviewer’s thoughtful suggestions. We have revised “gene” into “cDNA sequence”, seeing Line 97.

Line 133: Delete "were" from the sentence "To analyse whether DcCDA3 were expression...".

Reply: Thanks for reviewer’s valuable comments. We have deleted the “were” in previous manuscript, seeing Line 136.

The statistical analysis shown in Fig. 3 is ambiguous as it is not known between which developmental phases the difference is significant. For example, although first instar nymphs have a significant difference at p < 0.01, it is not obvious which of the different developmental phases this difference applies to. Presentation of the statistical analysis should be similar to that of enzyme activity as shown in Fig. 7.

Reply: Thanks for reviewer’s valuable suggestions. We have revised the Figure 3, seeing Line 125 to Line 126.

Section 2.3, lines 132 to 138: No mention of Fig. 4 is made in the text. Additionally, as significant differences were observed at 1, 12, and 18 hours post 20E application, the authors could comment on whether 20E acts directly (a direct target) or indirectly (a downstream target) on the CDA3 gene.

Reply: Thanks for reviewer’s thoughtful comments. We have added the Figure 4 in previous manuscript, seeing Line 138. 20-hydroxyecdysone (20-E) plays an important role in controlling development, metamorphosis, reproduction and diapause of insect. through a specific nuclear receptor complex, ecdysone receptor (EcR) and ultraspiracle (USP). The ligand-receptor complex 20E-EcR/USP directly activities a small set of early-response genes and a much larger set of late-response genes (20-hydroxyecdysone Reduces Insect Food Consumption Resulting in Fat Body Lipolysis During Molting and Pupation; Identification of 20-Hydroxyecdysone Late-Response Genes in the Chitin Biosynthesis Pathway). In addition, several studies reported that the expression of chitin degradation-related genes in some insects could be regulated by 20E, indicating chitin degradation-related genes are important 20E late-responsive genes (Functional characterization of chitin deacetylase 1 gene disrupting larval-pupal transition in the drugstore beetle using RNA interference). Therefore, we speculated that 20E might directly bind EcR and USP to form a receptor complex, and then regulate the expression of DcCDA3. We have added the related descriptions in discussion, seeing Line 254 to Line 260.    

Section 2.6, lines 176-181: No reference to the SDS-PAGE figure is made in the text. Also, two figures are labeled "Figure 6" ("The expression levels of DcCDA3 in the midgut and fat body ..." and "Analysis of the recombinant DcCDA3 protein using SDS-PAGE").

Reply: Thanks for your valuable comments. We have revised the related descriptions in previous manuscript, seeing Line 182, Line 186, Line 194, Line 195, Line 198, Line 206 and Line 210.

Line 189: The authors should give a reference for the use of 4-nitroacetanilide as a valid substrate for deacetylase activity as only recently have insect CDAs been shown to have enzymatic activity and these reports have used various forms of chitin (for example see Liu et al, Journal of Insect Physiology 113 (2019) 42-48, and Liu et al, Journal of Biological Chemistry (2019) 294: 5774-5783). If 4-nitroacetanilide has previously been used for CDA activity (for example with fungal or bacterial CDAs) it would be beneficial to reference that here.

Reply: Thanks for reviewer’s valuable suggestions. We have added the related references in previous manuscript, seeing Line 516 to Line 520.  

The purified protein shown in the SDS-PAGE figure contains multiple contaminating proteins; the authors should perform a parallel expression/purification using an empty vector and subsequent activity assay to show that one of these proteins was not responsible for the activity described in the text.

Reply: Thanks for reviewer’s valuable and thoughtful comments. The recombinant DcCDA3 was expressed by using prokaryotic expression system. In order to further investigate the effect of contaminating proteins on subsequent activity assay, we have added a parallel expression/purification using empty pET-32 a vector. At the optimum temperature and pH, according to the calculation formula of enzyme activity (Enzyme activity=(A₄₀₀-A₀)/KT), and the results showed that the value of A₀ is more than A₄₀₀, indicating bacterial protein has no effect on enzyme activity.

Line 192: Figure 7 shows the maximum activity at 50 ℃ and not 55 ℃.

Reply: We are sorry for incorrect descriptions in previous manuscript. We have revised “55 ℃” into “55 ℃”, seeing Line 195.

In Figure 8: Bacillus is misspelled as "Bacilius".

Reply: Thanks for reviewer’s thoughtful comments. We have revised the Figure 9, seeing Line 207 to Line 208.

Lines 294-295: I do not understand the statement "D. citri adult was precooled in the ice to escape".

Reply: We are sorry for unclear descriptions in previous manuscript. D. citri adult is very active, and it is easy escape once touching. Therefore, before the injection, we placed the D. citri adult on the ice to freeze it for a while. After 5 min on the ice, the D. citri adult become inactive. In addition, we revised the related descriptions, seeing Line 309.     

Line 299: Given the small nature of the insects, a description of how the fat body was collected is appropriate.

Reply: Thanks for reviewer’s thoughtful comments. We have added the related descriptions to previous manuscript, seeing Line 313 and Line 317. Briefly, D. citri adult abdomen was dissected using microforceps and obtained the midgut, then removed D. citri reproductive organs and hemolymph. Due to it is very hard to collect D. citri fat body by using the microforceps. Therefore, the remaining cuticle tissues containing fat body were dipped into the precooled PBS solution to elute the fat body. The collected tissues were ground using liquid nitrogen and stored at -80 °C.  

Line 304: It is not clear here if the dilutions of 20E were made with DEPC water or acetone; presumably with DEPC water as this was used for the control but it should be explicitly stated to avoid confusion. Additionally, the concentration of the 20E stock solution should be given so that the final concentration of acetone can be determined.

Reply: Thanks for reviewer’s thoughtful and valuable comments. 20E was dissolved in acetone to prepare the stock solution at a final concentration of 500 μg/μL. And then, the stock solution was diluted to 10 μg/μL working concentrations with DEPC water. We have added the detailed descriptions in previous manuscript, seeing Line 322 to Line 324. 

Line 321: What query sequence was used for the BLASTX search?

Reply: Thanks for reviewer’s valuable comments. We have added the query sequence in previous manuscript, seeing Line 345 to Line 347.

Lines 327-328: A description of the phylogenetic analysis should include a citation to Table S1.

Reply: Thanks for reviewer’s thoughtful comments. We have added the citation to Table S1 in previous manuscript, seeing Line 353 to Line 355.

Line 394: "induced up-regulated" should be either "induced" or "up-regulated" but not both.

Reply: Thanks for reviewer’s thoughtful suggestions. We have deleted “up-regulated” in previous manuscript, seeing Line 421.

Reviewer 2 Report

This paper presents the results of a series of experiments interrogating the function of a chitin deacetylase in Diaphorina citri, vector of Candidatus Liberibacter asiaticus. The authors find tissue-specific and growth stage-dependent variation in the expression of theis protein, which is presumably a member of group III chitin deacetylases, and demonstrate the expression is inducible by 20E. Downstream RNAi experiments fail to show a phenotype, but purified protein appears to have antimicrobial activity.

Although these results have interesting and possibly impactful implications, it is not recommended for publication at this time. There are major problems, as critical information has been omitted, results not sufficiently explained, and in sections the logic is unclear. So much of the results rely on a robust RT-qPCR assay, so the failure to report sufficient data is a serious flaw in the manuscript.

Specifically, I am referring to the following:

Purification of the protein and downstream assays.

My immediate question was why would the protein be antimicrobial? What is the substrate in bacteria, which lack chitin? Has this been shown before? The authors give no reasoning or explanation of this result. Why is the expressed protein 20 kDa larger than the native form? His tags are generally <10 residues in length. The full-length protein sequence should be reported. The authors report “strong antibacterial activity” against the gram positive bacteria, but the zones of inhibition aren’t very large at all, and with Bacillus, there appears to be a lawn. This type of assay isn’t quantitative. Lines 274-275 report their hypothesis that soluble chitosan is inhibiting the growth of bacteria, but they are treating the cells with the enzyme, not chitosan.

RT-qPCR

The kit used for RNA extraction is not specifically named. Are the psyllids or digested tissue tested individually or pooled? If they were tested individually, I can’t imagine the RNA would be visible on a standard gel stained with ethidium bromide. If pooled, how many per pool? The gel image should be included as supplemental data. How much RNA was used for reverse transcription? Were the GAPDH primers verified to be stable in all life stages and tissue types? Are they published? For the delta delta Cq method, values are normalized to both a housekeeping gene and a control treatment. What control sample was used for normalization? I think that the authors actually used the delta Cq method. How did the authors verify that the DNAse treatment worked? What were the results of the melt curve? What was the amplification efficiency of the primers?

General comments

Statements made in lines 73-74 and 252-253 are in direct contradiction. Authors don’t indicate if or how many replicate experiments were performed. Group 3 chitin deacetylases should be introduced in the beginning. Are they known to affect insect moulting, etc. in any species of insect? What is the function of group 3 chitin deacetylases, specifically? The authors only report the function of group 1 CDAs in the discussion, which is not relevant to the work presented in this paper The authors report glycosylation sites with no explanation as to why this matters. Lines 259-261 are very confusing. Do the psyllids lack the pathway for defense against gram negatives? Are the authors implying that the chitin deacetylase is functioning as a substitute for immunity pathway? Controls are inconsistent - DEPC treated water, PBS, BSA...

Author Response

My immediate question was why would the protein be antimicrobial? What is the substrate in bacteria, which lack chitin? Has this been shown before? The authors give no reasoning or explanation of this result. Why is the expressed protein 20 kDa larger than the native form? His tags are generally <10 residues in length. The full-length protein sequence should be reported. The authors report “strong antibacterial activity” against the gram-positive bacteria, but the zones of inhibition aren’t very large at all, and with Bacillus, there appears to be a lawn. This type of assay isn’t quantitative. Lines 274-275 report their hypothesis that soluble chitosan is inhibiting the growth of bacteria, but they are treating the cells with the enzyme, not chitosan.

Reply: Thanks for reviewer’s thoughtful and valuable comments. Chitin is found in bacteria, fungi, arthropods and nematodes (Chitin metabolism in insects: structure, function and regulation of chitin synthases and chitinases). Chitin deacetylase can catalyze the hydrolysis of N-acetamido binds of chitin, converting it to chitosan (Chitin deacetylases: new, versatile tools in biotechnology). In this study, after injection of E. coli and S. aureus, the expression level of DcCDA3 can be induced. In addition, recombinant DcCDA3 protein also showed antibacterial activity for gram-positive bacteria. Therefore, our results revealed that DcCDA3 might be involved in immune response in D. citri. In previous report, Sun et al. revealed that chitin deacetylase 1 (CDA1) from Exopalaemon carinicauda may play its biological activity in immune defense by deacetylation from chitin (Immune function against bacteria of chitin deacetylase 1 (EcCDA1) from Exopalaemon carinicauda). In Nilaparvata lugens and Tribolium castaneum, after silencing of CDA3, there was no observable morphological phenotype or interstructural abnormality. The authors also speculated that CDA3 may contribute to the defence of insects from pathogens (Chitin deacetylase family genes in the brown planthopper, Nilaparvata lugens (Hemiptera: Delphacidae); Analysis of functions of the chitin deacetylase gene family in Tribolium castaneum). However, the molecular mechanisms of anti-bacteria are unclear. We considered that DcCDA3 might be involved in the deacetylation of bacterial chitin to form chitosan. The soluble chitosan exhibited antibacterial properties.

In this study, the predicted molecular weight of DcCDA3 is 51.3 kDa. By prokaryotic expression system, we used pET-32a expression vector and expressed recombinant DcCDA3 protein. The pET-32a expression vector contains a His tag and a Thioredoxin tag. The total of molecular weight of tag is about 19 kDa. Therefore, the final molecular weight of DcCDA3 protein is about 70.3 kDa.

For Bacillus subtilis treatment group, we repeated the experiment of antibacterial activity of recombinant DcCDA3, and the results indicated that recombinant DcCDA3 protein showed weak antibacterial activity, seeing Line 207 to Line 208. In addition, we have performed quantitative determination of antibacterial activity, and the results were added to the Attachment 1.

The kit used for RNA extraction is not specifically named. Are the psyllids or digested tissue tested individually or pooled? If they were tested individually, I can’t imagine the RNA would be visible on a standard gel stained with ethidium bromide. If pooled, how many per pool? The gel image should be included as supplemental data. How much RNA was used for reverse transcription? Were the GAPDH primers verified to be stable in all life stages and tissue types? Are they published? For the delta delta Cq method, values are normalized to both a housekeeping gene and a control treatment. What control sample was used for normalization? I think that the authors actually used the delta Cq method. How did the authors verify that the DNAse treatment worked? What were the results of the melt curve? What was the amplification efficiency of the primers?

Reply: Thanks for reviewer’s thoughtful and valuable comments. In this study, TaKaRa MiniBEST Universal RNA Extraction Kit was used for RNA extraction. Because D. citri individuals are so tiny, it is difficult for a single adult or nymph to isolate complete RNA. Therefore, the psyllids or digested tissues tested pooled. A total of 100 D. citri adults were pooled for RNA extraction. Three biological replicates were performed for each sample. The results of RNA agarose gel electrophoresis were added into Attachment 2. A total of 1.0 μg of RNA for each sample was reverse-transcribed using a First Strand cDNA Synthesis kit. We have added the related descriptions in previous manuscript, seeing Line 337 to Line 342.  Based on the related reference (Selection of Reference Genes for Expression Studies in Diaphorina citri (Hemiptera: Liviidae)) and our previous study (Silencing of the Chitin Synthase Gene Is Lethal to the Asian Citrus Psyllid, Diaphorina citri), we found that the expression of GAPDH gene is stable in all life stages and tissue types. RT-qPCR data processing was performed using 2^-ΔΔCt method based on previous reference (Analysis of Relative Gene Expression Data Using Real-Time Quantitative PCR and the 2^-ΔΔCt Method). Therefore, GAPDH gene was used as housekeeping gene, and the sample with the lowest expression level was used for normalization. To improve the quality of the cDNA template, all RNA was treated using DNAse to remove genome DNA. We used First Strand cDNA Synthesis kit to perform RNA reverse transcription. The kit contains two steps, and the first step is to remove contaminated genome DNA. We have added the melt curve to the attachment. The primer efficiencies were determined according to making standard curves. The primer efficiency was 97.2% by the formula of 10^-1/slope-1. The related data was added to the Attachment 3.    

Statements made in lines 73-74 and 252-253 are in direct contradiction. Authors don’t indicate if or how many replicate experiments were performed. Group 3 chitin deacetylases should be introduced in the beginning. Are they known to affect insect moulting, etc. in any species of insect? What is the function of group 3 chitin deacetylases, specifically? The authors only report the function of group 1 CDAs in the discussion, which is not relevant to the work presented in this paper. The authors report glycosylation sites with no explanation as to why this matters. Lines 259-261 are very confusing. Do the psyllids lack the pathway for defense against gram negatives? Are the authors implying that the chitin deacetylase is functioning as a substitute for immunity pathway? Controls are inconsistent - DEPC treated water, PBS, BSA...

Reply: We are sorry for unclear descriptions in previous manuscript. In Tribolium castaneum, no visible phenotypic changes were observed after injection of dsRNAs for TcCDA3 to 9, whereas injection of dsRNA for TcCDA1 or TcCDA2 affected all types of molts (Analysis of functions of the chitin deacetylase gene family in Tribolium castaneum). We have revised the related descriptions, seeing Line 73 to Line 75. In this study, RNAi experiment was conducted containing three biological replicates, seeing Line 379. In addition, we have added the detailed descriptions in previous manuscript, seeing Line 233 to Line 239. According to NetNGlyc 1.0 software analysis, three potential N-glycosylation site were predicted. It indicated that DcCDA3 may belong to glycosylated protein. The specific glycosylated proteins are involved in the protein folding, quality control and assembly, and play an important role in the immune system (Glycosylation and the Immune System). Tsigos et al. revealed that chitin deacetylase from Colletotrichum lindemuthianum is an acidic glycoprotein with a carbohydrate content. It exhibits a very stringent specificity for β-(1, 4)-linked N-acetyl-D-glucosamine homopolymers (Purification and Characterization of Chitin Deacetylase from Colletotrichum lindemuthianum). Based on previous references, Arp et al. found that D. citri is capable of defending against E. coli, a Gram-negative bacterium, despite lacking the IMD defense pathway, most antimicrobial peptides, 1, 3-β-glucan recognition proteins (GNBPs) and complete peptidoglycan recognition proteins (Innate immune system capabilities of the Asian citrus psyllid, Diaphorina citri; Annotation of the Asian Citrus Psyllid Genome Reveals a Reduced Innate Immune System). In this study, combined bacterial challenge experiment and antibacterial activity, we speculated DcCDA3 might be involved in the D. citri immune response against bacterial infection. However, the specific reasons need to further research.        

Round 2

Reviewer 1 Report

1) The authors did not address my question if DcCDA3 contained the residues critical for catalytic activity, they simply repeated that SMART analysis identified a deactylase-like domain. The authors should see the analysis of five important motifs discussed in the N. lugens and the T. castaneum manuscripts (references 17 and 18).

2) The authors did not address my question of an appropriate citation for the use of 4-nitroacetanilide as a valid substrate for deacetylase activity. The citations they give, references 33 and 34, use various forms of chitin as substrate to measure deacetylase activity, not 4-nitroacetanilide. My request to the authors is to describe why it is an appropriate substrate for this class of enzyme and to give a relevant reference.

3) Line 98: "... amino acid sequence of the DcCDA gene" should be "... amino acid sequence of the DcCDA cDNA".

4) Line 193: Please confirm at what temperature the optimal pH assays were performed at. In section 2.7 it is listed as 55 C ("Then, we tested the optimal pH at 55 C, as shown in Figure 8B"), however, in section 4.9 it is listed as 50 C (The sample ... was pre-incubated in a 50 C, followed by incubation in a 50 C water bath").

Reviewer 2 Report

Thanks for reviewer’s thoughtful and valuable comments. Chitin is found in bacteria, fungi, arthropods and nematodes (Chitin metabolism in insects: structure, function and regulation of chitin synthases and chitinases). Chitin deacetylase can catalyze the hydrolysis of N-acetamido binds of chitin, converting it to chitosan (Chitin deacetylases: new, versatile tools in biotechnology).

Response: I tried but could not find any examples of bacteria that synthesize chitin, with the only exception of lipochitin oligosaccharides secreted by Rhizobial species. It's possible that the chitin deacetylase is acting on a part of the peptidoglycan that is also found in chitin (N-acetyl-D-glucosamine) in a non-specific manner (refer to Substrate recognition and specificity of chitin deacetylases and related family 4 carbohydrate esterases by Aragunde et al). Regardless, the authors need to be more careful in wording the discussion, since the idea that chitosan is released after treatment with the purified protein in these species of bacteria does not make sense. 

In this study, after injection of E. coli and S. aureus, the expression level of DcCDA3 can be induced. In addition, recombinant DcCDA3 protein also showed antibacterial activity for gram-positive bacteria. Therefore, our results revealed that DcCDA3 might be involved in immune response in D. citri. In previous report, Sun et al. revealed that chitin deacetylase 1 (CDA1) from Exopalaemon carinicauda may play its biological activity in immune defense by deacetylation from chitin (Immune function against bacteria of chitin deacetylase 1 (EcCDA1) from Exopalaemon carinicauda). In Nilaparvata lugens and Tribolium castaneum, after silencing of CDA3, there was no observable morphological phenotype or interstructural abnormality. The authors also speculated that CDA3 may contribute to the defence of insects from pathogens (Chitin deacetylase family genes in the brown planthopper, Nilaparvata lugens (Hemiptera: Delphacidae); Analysis of functions of the chitin deacetylase gene family in Tribolium castaneum). However, the molecular mechanisms of anti-bacteria are unclear. We considered that DcCDA3 might be involved in the deacetylation of bacterial chitin to form chitosan. The soluble chitosan exhibited antibacterial properties.

Response: Again, I can find no evidence of chitin in the bacteria used in this study. Please revise lines 283-284 accordingly.

In this study, the predicted molecular weight of DcCDA3 is 51.3 kDa. By prokaryotic expression system, we used pET-32a expression vector and expressed recombinant DcCDA3 protein. The pET-32a expression vector contains a His tag and a Thioredoxin tag. The total of molecular weight of tag is about 19 kDa. Therefore, the final molecular weight of DcCDA3 protein is about 70.3 kDa.

For Bacillus subtilis treatment group, we repeated the experiment of antibacterial activity of recombinant DcCDA3, and the results indicated that recombinant DcCDA3 protein showed weak antibacterial activity, seeing Line 207 to Line 208. In addition, we have performed quantitative determination of antibacterial activity, and the results were added to the Attachment 1.

Response: Please modify line 200 to replace the word "strong" with something that fits these results better.

Thanks for reviewer’s thoughtful and valuable comments. In this study, TaKaRa MiniBEST Universal RNA Extraction Kit was used for RNA extraction. Because D. citri individuals are so tiny, it is difficult for a single adult or nymph to isolate complete RNA. Therefore, the psyllids or digested tissues tested pooled. A total of 100 D. citri adults were pooled for RNA extraction. Three biological replicates were performed for each sample. The results of RNA agarose gel electrophoresis were added into Attachment 2. A total of 1.0 μg of RNA for each sample was reverse-transcribed using a First Strand cDNA Synthesis kit. We have added the related descriptions in previous manuscript, seeing Line 337 to Line 342.

Response: Please add to the methods the number of insects or dissected insect tissues used per extraction.

Therefore, GAPDH gene was used as housekeeping gene, and the sample with the lowest expression level was used for normalization. 

Response: Please indicate in the methods which sample was used to normalize the data.

Minor edit: Line 248 "Wang et al..." should actually be "Yang et al...." based on the references cited.

Author Response

Reviewer 1

1. The authors did not address my question if DcCDA3 contained the residues critical for catalytic activity, they simply repeated that SMART analysis identified a deacetylase-like domain. The authors should see the analysis of five important motifs discussed in the N. lugens and the T. castaneum manuscript (references 17 and 18)

Reply: Thanks for reviewer’s thoughtful and valuable comments. We have added the detailed descriptions in previous manuscript, seeing Line 98 to Line 100. The motif analysis of DcCDA3 was added to the Figure S1.

2. The authors did not address my questions of an appropriate citation for the use of 4-nitroacetanilide as a valid substrate for deacetylase activity. The citations they give, references 33 and 34, use various forms of chitin as substrate to measure deacetylase activity, not 4-nitroacetanilide. My request to the authors is to describe why it is an appropriate substrate for the class of enzyme and to give a relevant reference.

Reply: Thanks for reviewer’s valuable and thoughtful comments. The chitin is insoluble in water, so it needs to be prepared into water-soluble derivatives. The while process is very cumbersome. In recent years, determination of CDA activity mainly contains three methods, including radiometric labeling, NADH developing and benzpyrole/hydrochloric acid colorometric method. Zhong et al. expressed Bombyx mori chitin deacetylase 7 (BmCDA7) using eukaryotic expression system, and then assayed the activity of BmCDA7 using 4-nitroacetanilide as substrate. We have added the related reference, seeing Line 523 to Line 529.

3. Line 98: “…amino acid sequence of the DcCDA gene” should be “…amino acid sequence of the DcCDA cDNA”.

Reply: Thanks for reviewer’s valuable comments. We have revised “the DcCDA gene” into “the DcCDA cDNA” in previous manuscript, seeing Line 103.

4. Line 193: Please confirm at what temperature the optimal pH assays were performed at. In section 2.7 it is listed as 55 ℃ (“Then, we tested the optimal pH at 55 ℃, as shown in Figure 8B”), however, in section 4.9 it is listed as 50 ℃ (The sample…was pre-incubated in a 50 ℃, followed by incubation in a 50 ℃ water bath”).

Reply: Thanks for reviewer’s thoughtful comments. We have revised the incorrect descriptions, seeing Line 200.

Reviewer 2

I tried but could not find any examples of bacteria that synthesize chitin, with the only exception of lipochitin oligosaccharides secreted by Rhizobial species. It's possible that the chitin deacetylase is acting on a part of the peptidoglycan that is also found in chitin (N-acetyl-D-glucosamine) in a non-specific manner (refer to Substrate recognition and specificity of chitin deacetylases and related family 4 carbohydrate esterases by Aragunde et al). Regardless, the authors need to be more careful in wording the discussion, since the idea that chitosan is released after treatment with the purified protein in these species of bacteria does not make sense.

Reply: Thanks for reviewer’s valuable and thoughtful suggestions. We have revised the related descriptions in the discussion, seeing Line 290 to Line 292.

Again, I can find no evidence of chitin in the bacteria used in this study. Please revise lines 283-284 accordingly.

Reply: Thanks for reviewer’s valuable and thoughtful comments. We have revised the related descriptions, seeing Line 290 to Line 292.

Please modify line 200 to replace the word "strong" with something that fits these results better.

Reply: Thanks for reviewer’s valuable suggestions, we have revised “strong” into “obvious” to fit these results, seeing Line 208.

Please add to the methods the number of insects or dissected insect tissues used per extraction.

Reply: Thanks for reviewer’s valuable comments. We have added the number of insects or dissected insect tissues used per extractions, seeing Line 300 and Line 315.

Please indicate in the methods which sample was used to normalize the data.

Reply: Thanks for reviewer’s thoughtful comments. We have added the related descriptions in the methods, seeing Line 371.

Line 248 "Wang et al..." should actually be "Yang et al...." based on the references cited.

Reply: Thanks for reviewer’s valuable comments, we have revised “Wang et al…” into “Yang et al…”, seeing Line 256.