Microbiome succession during ammonification in eelgrass bed sediments
- Published
- Accepted
- Subject Areas
- Marine Biology, Microbiology, Plant Science
- Keywords
- succession, microbiome, ammonification, sulfur cycling, eelgrass, decomposition, seagrass
- Copyright
- © 2017 Ettinger et al.
- Licence
- This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. For attribution, the original author(s), title, publication source (PeerJ Preprints) and either DOI or URL of the article must be cited.
- Cite this article
- 2017. Microbiome succession during ammonification in eelgrass bed sediments. PeerJ Preprints 5:e2956v1 https://doi.org/10.7287/peerj.preprints.2956v1
Abstract
Background. Eelgrass (Zostera marina) is a marine angiosperm and foundation species that plays an important ecological role in primary production, food web support, and elemental cycling in coastal ecosystems. As with other plants, the microbial communities living in, on, and near eelgrass are thought to be intimately connected to the ecology and biology of eelgrass. Here we characterized the microbial communities in eelgrass sediments throughout an experiment to quantify the rate of ammonification, the first step in early remineralization of organic matter, or diagenesis, from plots at a field site in Bodega Bay, CA.
Methods. Sediment was collected from 72 plots from a 15 month long field experiment in which eelgrass genotypic richness and relatedness were manipulated. In the laboratory, we placed sediment samples (n= 4 per plot) under a N2 atmosphere, incubated them at in situ temperatures (15 oC) and sampled them initially and after 4, 7, 13, and 19 days to determine the ammonification rate. Comparative microbiome analysis using high throughput sequencing of 16S rRNA genes was performed on sediment samples taken initially and at 7, 13 and 19 days to characterize the relative abundances of microbial taxa and how they changed throughout early diagenesis.
Results. Within-sample diversity of the sediment microbial communities across all plots decreased after the initial timepoint using both richness based (observed number of OTUs, Chao1) and richness and evenness based diversity metrics (Shannon, Inverse Simpson). Additionally, microbial community composition changed across the different timepoints. Many of the observed changes in relative abundance of taxonomic groups between timepoints appeared driven by sulfur cycling with observed decreases in sulfur reducers (Desulfobacterales) and corresponding increases in sulfide oxidizers (Alteromonadales and Thiotrichales). None of these changes in composition or richness were associated with ammonification rates.
Discussion. Overall, our results showed that the microbiome of sediment from different plots followed similar successional patterns, which we surmise to be due to changes related to sulfur metabolism. These large changes likely overwhelmed any potential changes in sediment microbiome related to ammonification rate. We found no relationship between eelgrass presence or genetic composition and the microbiome. This was likely due to our sampling of bulk sediments to measure ammonification rates rather than sampling microbes in sediment directly in contact with the plants and suggests that eelgrass influence on the sediment microbiome may be limited in spatial extent. More in-depth functional studies associated with eelgrass microbiome will be required in order to fully understand the implications of these microbial communities in broader host-plant and ecosystem functions (e.g. elemental cycling and eelgrass-microbe interactions).
Author Comment
This is a submission to PeerJ for review.
Supplemental Information
Kruskal-Wallis tests on alpha diversity metrics
We used Kruskal-Wallis tests with 9999 permutations to assess whether alpha diversity was significantly different between categories. We used four different measurements of alpha diversity (observed number of OTUs, Chao1, Shannon Inverse Simpson). Categories examined included timepoint, eelgrass status (one genotype, multiple genotypes or none present), eelgrass initial relatedness (low, medium, high), eelgrass final richness and plot location.
Post-hoc Dunn tests assessing alpha diversity over time
Alpha diversity was determined to be significantly different across timepoints (Table 1).We examined which timepoint comparisons were stochastically dominant using the Dunn test on four different measurements of alpha diversity (observed number of OTUs, Chao1, Shannon Inverse Simpson). Timepoint 1 (initial samples), 2 (7 days), 3 (13 days), and 4 (19 days).
PERMANOVA results of beta diversity during microbial succession
PERMANOVA tests were performed to find significant differences in microbial beta diversity, calculated as the Weighted Unifrac distance metric, between different categorical variables including initial plot treatment (number of genotypes x level related), eelgrass plot richness, eelgrass initial level related (low, medium, high), eelgrass genotypic evenness, eelgrass status (one genotype, multiple genotypes or none present), timepoint, block (A-L), eelgrass richness, spot (1-6) and plot location (block x spot).
Pair-wise PERMANOVA results of beta diversity over time
Comparing microbial community structure between pair-wise timepoints using multiple beta diversity metrics (Weighted Unifrac, Unweighted Unifrac, Bray Curtis) to assess at which timepoints, the communities differed significantly. Timepoint 1 (initial samples), 2 (7 days), 3 (13 days), and 4 (19 days).
PERMANOVA tests on initial beta diversity
PERMANOVA tests were used to look for significant differences in microbial beta diversity, calculated as the Weighted Unifrac distance metric, between different categorical variables at timepoint #1. The categorical variables tested included initial plot treatment (number of genotypes x level related), eelgrass plot richness, eelgrass initial level related (low, medium, high), eelgrass genotypic evenness, eelgrass status (one genotype, multiple genotypes or none present), block (A-L), eelgrass richness and spot (1-6).
Mantel test results correlating microbial beta diversity throughout succession with measured variables
Mantel tests were used to identify significant correlations between microbial beta diversity, calculated as Bray Curtis dissimilarities, and different quantitative variables including ammonification rate (µmol NH4-N/L sediment/d), total belowground biomass (g/plot), total aboveground biomass (g/plot) and total biomass (g/plot).
Mantel test results correlating initial microbial beta diversity with measured variables
Mantel tests were used to identify significant correlations between microbial beta diversity, calculated as Bray Curtis dissimilarities, and different quantitative variables at timepoint #1. The quantitative variables tested include ammonification rate (µmol NH4-N/L sediment/d), total belowground biomass (g/plot), total aboveground biomass (g/plot), total biomass (g/plot), rhizome biomass (g/plot), root biomass (g/plot), Rao’s Q, eelgrass genotypic evenness, eelgrass Shannon Diversity, eelgrass average relatedness, plot detritus standing stock (g/plot) from prior months (June, July, August) and plot decomposition rate.
Kruskal-Wallis tests of mean relative abundance of taxonomic orders over time
The average relative abundance of taxonomic orders was compared between timepoints using Bonferroni corrected Kruskal-Wallis tests.
Post-hoc Dunn tests of mean relative abundance of taxonomic orders over time
Post-hoc Dunn tests were were performed on taxonomic orders that were found to have significantly different mean relative abundances across timepoints using Kruskal-Wallis tests (Table S8). These tests were used to identify which timepoint comparisons showed stochastic dominance. Only sequential timepoint comparisons are shown here. Timepoint 1 (initial samples), 2 (7 days), 3 (13 days), and 4 (19 days).
Mean, standard deviation and standard error of the relative abundances of taxonomic orders over time
Only orders with a mean relative abundance of greater than or equal to 2 percent are show here.