The Baseline Gut Microbiota Enterotype Directs Lifestyle-Induced Amelioration of Pollen Allergy Severity: A Self Controlled Case-Series Study

Round 1

Reviewer 1 Report

The first part of the article focuses on the study of the relationship between bacterial  enterotype, and some lifestyle factors diet related, and the prevalence and severity of pollen allergy (PA).

In this first part, the methodology is reasonably clear. Although the control group is much smaller than the group with pollen allergy, the latter suffers stratification by severity, resulting in smaller groups.

However, it would be very important for the interpretation of the results to know what medications were taken by those allergic to pollen. With the high severity of several individuals, the is a high probability of being medicated, for example with antihistamines. Only the non-ingestion of antibiotics is described. There may be a risk that the action of other drugs on the intestinal microflora may not be considered.

The negative association between “No weekly meat intake” and “Daily intake of fermented plants” and the severity of PA is interesting and could be better discussed.

The inclusion of Gender and Age in “Lifestyle factors” description is a bit strange.

In a second part of the article, an intervention in lifestyle is applied, different for each enterotype, and there is the intention to evaluate if this intervention changed the severity of PA. In this part, the methodology is unclear and I believe that it is not the best experimental design to achieve the objectives.

-          Why is the intervention unequal in the different enterotypes? What data allowed this strategy to be drawn?

-          The intervention lasted a month. Evaluation of allergic symptomatology occurs at the end of this time. The differences are made to the individual variations. However, nothing is said about the type of allergy of each individual. In pollen allergy, exposure to pollen type(s) to which each individual is allergic is determinant of the severity of symptoms. Pollen levels can rise, maintain or decrease within a month, which will influence symptomatology much more decisively than any lifestyle intervention. This aspect has not been discussed.

Author Response

We thank the reviewer for finding our work of interest. Please find below the point to point answers to comments.

Author Response File: Author Response.pdf

Reviewer 2 Report

The authors describe that by changing the diet, they can change the microbiome which correlates with allergic symptoms. Their diets reduce allergic symptoms.

The authors over sell the data. P<0.1 is not acceptable. Also for instance citing a  mice study as clinical is pointing to the need to oversell as the results are not so spectacular. Also the discussin is like a review, showing the impoirtance of diet and allergy, but distracts from their results. Why? Perhaos because their results are not so convincing

 

I think the paper will merit a lot by reducing the discussion to their own results, an only cite others when needed. Do not review diet and allergy too much.

 

specific comments:

 

Abstract: negatively correlated with no weekly meat. Is there a simpler way to say that?

 

First page: affects approx. 40% of children??? That are the data for allergic sensitization. But rhinitis is lower

 

respectivefeature. Write respective feature.

 

The Japanese PA questionnaire allows the participants to self-assess and score nasal symp- toms, sneezes, rhinorrhea, and nasal congestion based on a five-point scale (0 = none, 1 = mild, 2 = moderate, 3 = severe, and 4 = very severe).

Daily? If yes, please add. Or is it one score once for the whole study period?

 

2.6.3. Bonferroni correction?

 

Fromm which p value did you consider it to be significant?

 

Table 1. how can you say p<0.001 for allergy when mild allergy is zero for both groups?

 

In the legend of table 2 indicate where you find what BF and PF means, or explain again

 

In table 2 all p values are p=0.059? Why not list the p values per enterotype

Make colum 1 a bit larger so enteroptype BM is not on the next line

 

How was severity score measured? Daily or single point?

 

Table 4. What is fermented plant intake?

 

Poor sleep is negatively correlated. Yeah, allergic individual during the pollen season sleep worse

 

The study found males (coefficient = 0.160; p = 0.099) were more prone to PA : th ep value is abot 0.1. Thus not sstatistically significant. Thus not true

3.4. What is Natto?

Table 5. P0.092 is not statistically significant.

Where does the legend of table 5 end and the text continues?

 

3.6. the severity score was reduced 1 point. How big is the scale? (How much is one point)

 

Concordantly, a previous study demon- strated that fermented plant products can alleviate the clinical symptoms of Japanese PA without modifying systemic immunological characteristics [33].

This is a study in  mice. Can you say “clinical” in that case, implying you talk on humans?

mainly copmprising fruits : rewrite

 

explain why you have less butyrate when your bacteria produce acetate?

 

Furthermore, interaction analysis in our study revealed that PA se- verity in individuals consuming three meals per day was negatively correlated

Not true, that is p=0.1.

Author Response

We thank the reviewer for finding our work of interest. Please find below the point to point answers to comments.

Author Response File: Author Response.pdf

Reviewer 3 Report

1.     In 2.4, the authors performed the traditional 16S rRNA sequencing to classify the bacterial composition of fecal samples. People usually sequence V3/V4 region in 16S rRNA sequencing as in majority of the cases that these regions contain the maximum nucleotide heterogeneity. Could the authors explain why V1-V2 region was sequenced in this study?

2.     In 2.6.1, the authors used the centering values of the clusters to assign the sub-enterotypes.

Could the authors (1) explain why they wanted to define sub-enterotypes for the downstream association analysis? (2) provide more information about how they found Faecalibacterium and Megamonas significant based on the analysis?

3.     In 2.6.3, the authors defined and calculated Yitc, severity score based on “glm2” and the variable factors in their cohort. Could the authors provide the relevant references for this approach if there is any? If no references can be cited here, it would be better that the authors could provide evidence that the proposed model can be applied to analyze the hypothetical or known datasets that show the significant associations between variable factors and outcomes.

4.     In 3.3, the authors analyzed the associations between major enterotypes and severity of PA. In the previous reports, several different methods of collapsing enterotype variation into a few discrete clusters suggest that enterotype distribution is continuous and can vary widely within an individual. Therefore, the associations between major enterotypes and severity of PA here is limited. Is it possible that the authors could analyze the associations between individual bacterial species/genus and severity of PA?

5.     In Figure 2(a), please indicate the groups with enterotype information on the right as shown in (b).  

6.     In Figure 2, please try to simplify the labels. For instance, if G0 indicated control group, can the authors indirectly use “control” instead of G0. Similarly, if Dim1 means PCA1, why don’t you use PCA1 as x and y axis?

7.     Please try to place the text in the empty area to avoid covering the data. 

Author Response

We thank the reviewer for finding our work of interest. Please find below the point to point answers to comments.

Author Response File: Author Response.pdf