Guo, J. 2016. Rhizosphere metagenomics of three biofuel crops. Dissertation, Michigan State University, East Lansing, Michigan.

Citable PDF link: https://lter.kbs.msu.edu/pub/3671

Soil microbes form beneficial associations with crops in the rhizosphere and also play a major role in ecosystem functions, such as the N and C cycles. Thus large-scale cultivation of biofuel crops will have a significant impact on ecosystem functions at least regionally. In recent years, advances in high throughput sequencing technologies have enabled metagenomics, which in turn opens new ways to access the unknown majority in microbiology but poses great challenges for data analysis due to the large data size and short read length of sequence data sets.

We generated about 1 Tb of shotgun metagenomic data from rhizophere soil samples of three biofuel crops: corn, switchgrass, and Miscanthus. My central goal is to devise methods to extract meaning from this rhizosphere metagenomic data, with a focus on N cycle genes since N is the most limiting resource for sustainable for biofuel production. I initially provide a review of genetargeted methods for analyzing shotgun metagenomics.

In the second chapter I develop a method that improves the speed with which rRNA genes fragments can be found and analyzed in large shotgun metagenome data sets, thereby avoiding primer bias and chimeras that are problematic with PCR-based methods. I present a pipeline, SSUsearch, to achieve faster identification of short subunit rRNA gene fragments plus provides unsupervised community analysis. The pipeline also includes classification and copy number correction, and the output can be used in traditional amplicon downstream analysis platforms. Shotgun derived rhizosphere data from this pipeline yielded higher diversity estimates than amplicon data but retained the grouping of samples in ordination analysis. Our analysis confirmed the known bias against Verrucomicrobia in a commonly used V6-V8 primer set as well as discovered likely biases against Actinobacteria and for Verrucomicrobia in a commonly used V4 primer set.

In the third chapter, I explore an alternative phylogenetic marker to the widely used SSU rRNA gene, which has several limitations including multiple copies in the same genomes and low resolution for differentiating strains. I demonstrate that rplB, a single copy protein coding gene, provides finer resolution more akin to species and subspecies level and also finer scale (OTU) diversity analysis. The method requires shotgun sequence since the gene is not conserved enough for recovery by primers. When used on the rhizosphere sequence data, it revealed more microbial diversity and
better differentiated the communities among the three crops than the SSU rRNA gene analysis.

In the last chapter I address my central biological question on rhizosphere metagenomics: do they differ among the three crops and what does this information suggests about function? I compare the rhizosphere metagenomes for overall community structure (SSU rRNA gene), overall function (annotation from global assembly), and N cycle genes (using Xander, a targeted gene assembly tool). All three levels showed corn had a significantly different community from Miscanthus and switchgrass (except for ammonia-oxidizing Archaea), and that the two perennials showed a trend of separation. In terms of life history strategy, the corn rhizosphere was enriched in copiotrophs while the perennials were enriched in oligotrophs. This is further supported by higher abundance of genes in the “Carbohydrates” subsystem category and higher fungi/bacteria ratios. Additionally, the nitrogen fixing community of corn was dominated by nifH genes most closely affiliated to Rhizobium and Bradyrhizobium while the perennials had nifH sequences most related to Coraliomargarita, Novosphingobium and Azospirillum, indicating that the perennials independently selected beneficial members. Moreover, higher numbers of nitrogen fixation genes and lower number of nitrite reduction genes suggest better nitrogen sustainability of the perennials. These data indicate that perennial bioenergy crops have advantages over corn in higher microbial species richness and functional diversity as well as in selecting members with beneficial traits,
consistent with N use efficiency.

Associated Treatment Areas:

  • G5 Switchgrass
  • G6 Miscanthus
  • G1 Continuous Corn

Download citation to endnote bibtex

Sign in to download PDF back to index
Sign In