The plant circadian clock influences rhizosphere community structure and function

Abstract

Plants alter chemical and physical properties of soil, and thereby influence rhizosphere microbial community structure. The structure of microbial communities may in turn affect plant performance. Yet, outside of simple systems with pairwise interacting partners, the plant genetic pathways that influence microbial community structure remain largely unknown, as are the performance feedbacks of microbial communities selected by the host plant genotype. We investigated the role of the plant circadian clock in shaping rhizosphere community structure and function. We performed 16S ribosomal RNA gene sequencing to characterize rhizosphere bacterial communities of Arabidopsis thaliana between day and night time points, and tested for differences in community structure between wild-type (Ws) vs clock mutant (toc1-21, ztl-30) genotypes. We then characterized microbial community function, by growing wild-type plants in soils with an overstory history of Ws, toc1-21 or ztl-30 and measuring plant performance. We observed that rhizosphere community structure varied between day and night time points, and clock misfunction significantly altered rhizosphere communities. Finally, wild-type plants germinated earlier and were larger when inoculated with soils having an overstory history of wild-type in comparison with clock mutant genotypes. Our findings suggest the circadian clock of the plant host influences rhizosphere community structure and function.

Figure 1

Day vs night timepoint influences rhizosphere community structure. (a) Principal coordinate analysis of Bray–Curtis dissimilarities (n=17). Differences between the 1800 hours (day, orange) and 06 00 hours (night, blue) time points were significant at P=0.001. (b) Bivariate relationship between percent change in OTU abundance from 0600 to 1800 hours 22 July and percent change in OTU abundance from 0600 hours on 21 July to 1800 hours (r=0.423; P<0.001).

Figure 2

Clock function in A. thaliana alters rhizosphere community composition, abundances and diversity. (a) Principal coordinate analysis of Jaccard dissimilarities, where rhizosphere communities of Ws are represented by orange circles, toc1-21: blue circles, and ztl-30: gray circles (n=23). Rhizosphere community composition differs significantly between clock genotypes (P=0.001). (b) Principal coordinate analysis of Bray–Curtis dissimilarities (n=23). OTU abundances differ significantly between toc1-21 and both the Ws and ztl-30 genotypes (P=0.001). (c) Mean Shannon diversity index. The top and bottom of boxes represent the 75th and 25th percentiles, respectively. Whiskers represent 1.5 times the interquartile range. One-way analysis of variance and Tukey’s post hoc comparisons indicate significant differences between toc1-21 and both the Ws and ztl-30 genotypes (P=0.03).

Figure 3

Plant genotype influences both common (> 500 reads) and rare (< 500 reads) rhizosphere taxa. (a) Principal coordinate analysis of Jaccard dissimilarities of common taxa, where rhizosphere communities of Ws are represented by orange circles, toc1-21: blue circles and ztl-30: gray circles (n=23). Rhizosphere community composition differs significantly between all clock genotypes (P=0.001). (b) Principal coordinate analysis of Bray–Curtis dissimilarities of common taxa (n=23). OTU abundances differ significantly between toc1-21 and both the Ws and ztl-30 genotypes (P=0.001). (c) Principal coordinate analysis of Jaccard dissimilarities of rare taxa (n=23). Rhizosphere community composition differs significantly between all clock genotypes (P=0.001). (d) Principal coordinate analysis of Bray–Curtis dissimilarities of rare taxa (n=23). OTU abundances differ significantly between Ws and both toc1-21 and ztl-30 genotypes (P=0.001).

Figure 4

Indicator taxa based on (a) IVA and (b) Lefse.

Figure 5

Effects of soil overstory history on plant growth. Letters denote significant differences between soil treatments based on Tukey’s Honest Significant Differences post hoc comparisons. (a) At week 1 (n=60), plants grown in a soil with a history of Ws had significantly larger rosette diameters than plants grown in soils with a history of toc1-21 or ztl-30 (P=0.002). (b) In a separate experiment where seeds were not germinated synchronously (n=35), seeds sowed onto a soil with a history of Ws occurrence germinated significantly earlier than seeds sown into soils with a history of toc1-21 or ztl-30 genotypes (P<0.001).