Rice Plant–Soil Microbiome Interactions Driven by Root and Shoot Biomass

Round 1

Reviewer 1 Report

This is interesting study to examine microbial community related to shoot and root biomass using RILs. The results contain valuable information on interaction of plant – microbes in rice and its transition along stages. The following points are to be considered to improve the paper.

P442, P468 and other parts: Reproductive stage starts not from heading stage but from panicle initiation stage. After panicle initiation stage, physiological status changes in shoot and root, in particular, with growing sink strength by panicle.  

P441: Nitrate formed on root surface can be an important N source for rice grown in submerged conditions and AOB is involved in this.

e.g. Kirk G (2001) Plant-mediated processess to acquire nutrients: nitrogen uptake by rice plants. Plant Soil 232: 129–134.

Kirk G, Kronzucker H (2005) The potential for nitrification and nitrate uptake in the rhizosphere of wetland plants: a modelling study. Ann Bot 96: 639.

Table 1: The VIP should be considered clearly in Discussion that enable deeper evaluation on bacterial species in relation to RN and SB.

Author Response

This is interesting study to examine microbial community related to shoot and root biomass using RILs. The results contain valuable information on interaction of plant – microbes in rice and its transition along stages. The following points are to be considered to improve the paper.

P442, P468 and other parts: Reproductive stage starts not from heading stage but from panicle initiation stage. After panicle initiation stage, physiological status changes in shoot and root, in particular, with growing sink strength by panicle.  

Thank you for pointing out this error in wording, we have changed these lines and all mentions heading being the beginning of the reproductive phase to being the ending of the reproductive phase. (lines 86, 142-143, 480-481, 523).

P441: Nitrate formed on root surface can be an important N source for rice grown in submerged conditions and AOB is involved in this.

e.g. Kirk G (2001) Plant-mediated processess to acquire nutrients: nitrogen uptake by rice plants. Plant Soil 232: 129–134.

Kirk G, Kronzucker H (2005) The potential for nitrification and nitrate uptake in the rhizosphere of wetland plants: a modelling study. Ann Bot 96: 639.

We agree that N is an important source for rice, thank you for providing these references, we have incorporated them into the discussion and added them to the reference list (lines 475-478).

Table 1: The VIP should be considered clearly in Discussion that enable deeper evaluation on bacterial species in relation to RN and SB.

Thank you for this comment, for ease of identifying and comparing the species described in the discussion we have added both the ID numbers and the VIP scores from Table 1 throughout the discussion. Additionally, we added a line in the discussion (lines 443-446) indicating the discussion section is focused on significant communities defined by PLS VIP scores >0.8 in the clusters showing the greatest differences by trait.

Reviewer 2 Report

This is an interesting topic in general: how might the plant biomass can be modulate the rhizobacterial communities? This information might be important for understanding the plant-microbe interactions in the context of plant growth and food security. However, this paper still has several defects as follows, and need to be made clear:

 (1)The hypothesis of the manuscript it’s not clear, the authors suggest  correlations between rhizobacterial community compositions and biomass in diverse variety of rice plant at two developmental stage, please incorporate the hypothesis in the introduction sections.

(2) The soil/plant physiochemical properties, such as the contents of TN, TP, pH, are missing. These data need to be added. Do the substrate physiochemical properties have effects on bacterial communities in rhizosphere? The correlation between bacterial community and substrate  parameters and the related discussion are complete missing.

(3) Significant differences in diversity , the presence of specific keystone taxa and dissimilarities in the bacterial community composition should be included as a results. 

(4) MM section, in particular the  metagenomic library should be explain more deeper, please incorporate the primers used in this study.

 (5) The section "Discussion" is considerably weak, and some important problems need to be discussed. For example, the keystone taxa in bacterial communities were the same specific taxa correlated with the biomass (PLS-selected candidate list)? Do plant niche have effects on the suggest species and functions? 

Author Response

This is an interesting topic in general: how might the plant biomass can be modulate the rhizobacterial communities? This information might be important for understanding the plant-microbe interactions in the context of plant growth and food security. However, this paper still has several defects as follows, and need to be made clear:

(1) The hypothesis of the manuscript it’s not clear, the authors suggest correlations between rhizobacterial community compositions and biomass in diverse variety of rice plant at two developmental stage, please incorporate the hypothesis in the introduction sections.

Thank you for your feedback, the authors have clarified the hypothesis and stated clearly there are two developmental stages in lines 103-107.

(2) The soil/plant physiochemical properties, such as the contents of TN, TP, pH, are missing. These data need to be added. Do the substrate physiochemical properties have effects on bacterial communities in rhizosphere? The correlation between bacterial community and substrate parameters and the related discussion are complete missing.

The soil total C (0.66%), total N (0.085%) and pH (5.6) are reported in lines 120-122 for the soil used in this study. We did not collect the total C and N for the plant tissue thus we are not able to report these values. We did measure total C and N of the soil for each sample (add in lines 160-162) at heading and maturity and found they did not vary much across the samples, the average total %C across samples was 0.73 ± 0.02 and average total %N across all samples was 0.083 ± 0.002 (added in lines 260-261). We also found their influence on the microbial community structure was minimal compared to the biomass traits thus we did not include them in the original figure; however, we have added them to the PCoA plot (Figure 1) and in the text (lines 258-261).

(3) Significant differences in diversity, the presence of specific keystone taxa and dissimilarities in the bacterial community composition should be included as a results. 

Thank you for this feedback, we have added Welch’s t-test to test the significant differences in diversity across developmental stages by genotype which has been added in Supplemental Figure 3 and in the main text (lines 198-199, 265-267). Additionally, we have added the top phyla (>1% relative abundance) and top species (>2% relative abundance) as supplemental figures 4 and 5, respectively and in the main text in lines 267-271.

Thank you also for your comment on the dissimilarities specifically. From your comment we realized we had not updated our text and had mistakenly called our principal coordinate analysis (PCoA) a principal component analysis (PCA). This was due to changing our analysis/figure before submission without updating the text accordingly. We have now updated the text to reflect Figure 1 is a PCoA using Bray-Curtis distances throughout the manuscript. This is an analysis of the dissimilarities of the bacterial community composition.

(4) MM section, in particular the metagenomic library should be explain more deeper, please incorporate the primers used in this study.

We have added the reason shotgun metagenomic sequencing was chosen library in the materials and methods (lines 165-167). We have also included the full library preparation and index kit names to make it easier for the reader to find information on the index adapter and primers which are published on the Illumina website (line 167-169), because these are proprietary mixes we did not include the sequences in the manuscript. Additionally, in the sequence processing section we clarified the exclusive use of the SEED database for our functional gene assignments analysis (lines 185-186).

(5) The section “Discussion” is considerably weak, and some important problems need to be discussed. For example, the keystone taxa in bacterial communities were the same specific taxa correlated with the biomass (PLS-selected candidate list)? Do plant niche have effects on the suggest species and functions? 

Thank you for your feedback, we added supplemental figures 4 and 5 to show the taxa most abundant during heading and at maturity for each genotype to make comparisons between the most abundant taxa and those identified by PCoA and PLS analysis. We also have added identifiers (ID numbers) and the VIP scores from Table 1 throughout the discussion to help with comparisons between the results. For the discussion we added a line indicating we focused on those species that showed the most differentiation by trait and had VIP scores >0.8 (lines 443-446). Discussion on the overlap between the high abundance taxa and those associated with biomass traits or developmental stage has been added in lines 483-498.