Recipient Organization
UNIV OF CALIFORNIA (VET-MED)
(N/A)
DAVIS,CA 95616
Performing Department
VM: Population Hlth & Reprod
Non Technical Summary
Plants grow in close association with microbial communities that influence plant traits related to nutrient acquisition, plant defense, plant morphology and abiotic stress tolerance. The co-evolution between plants and their associated microbial communities enables plants to produce an immune system-like response as well as provide nutrients and minerals for plant growth and health. Isolation of a collection of microbes from the associated corn microbiome in a unique oligosaccharide mixture that collects on aerial roots provided evidence that nitrogen can be fixed by many members of the microbiome. This unexpected result was replicated in 600 different isolates using molecular mechanisms that are found in other bacteria (~30%) in addition to novel mechanisms that do not use the same gene sets (~70%) of the isolates. This proposal seeks to understand novel mechanisms bacteria use within the microbiome to fix nitrogen.
Animal Health Component
20%
Research Effort Categories
Basic
80%
Applied
20%
Developmental
0%
Goals / Objectives
p.p1 {margin: 0.0px 0.0px 0.0px 0.0px; font: 12.0px Helvetica} span.s1 {font: 8.0px Helvetica}Nitrogen fixation in non-legumes has been considered a 'holy grail'for at least 50 years. The well-described system of Klebsiella with a core set of six genes has been a long-standing model. It was recently recognized that alternative molecular models for fixation are relatively common. The discovery of nitrogen fixation in maize by our group provides a launching pad to approach for thedetermination of the molecular mechanisms from themicrobiome. Weimer's group collected over 1500 isolates from this plant over three years that have varying nitrogen fixation capabilities. The most active fixation capacity is found in organisms that do not contain the model gene set for fixation. Considering the importance of nitrogen in theplant life cycle and the need to reduce dependence on fertilization to increase sustainability, this group has all the needed capability and materials in hand to examine the methods of fixation, the environmental conditions needed for fixation, and the molecules beyond ammonia that can be usedto deliver nitrogen to the plant. The overarching research objective of this proposal is to understand the functionalities of the Sierra Mixe microbiota that contribute to N2 fixation in association with the indigenous maize landrace and to identify the consortium of microbes that contribute thesefunctionalities. Achieving this would allow the design and reconstitution of microbial consortia with the potential to be transferred to other varieties of corn and potentially to other cereal crops.We hypothesize that this functional N-fixing microbiota is comprised of a consortium of microbes that collectively provide fixed nitrogen to the maize plant via multiple mechanisms beyond ammonia delivery by digesting mucilage polysaccharide to support fixation,reduce oxygen tension in the mucilage for optimum conditions, and fix atmospheric N2 via a collection of genes that go beyond the known set for Klebsiella. Here we propose to test these functionalities and identify candidate members of this N2-fixingfunctional consortium from a collection of >1500 microbes that we have isolated from this indigenous landrace of maize. While it is possible that these functionalities may reside in a singlemicrobe, we expect that this is a community function comprising a minimum of three individual microbial species and potentially many more.We hypothesize that this functional N-fixing microbiota identified in association with SierraMixe maize is comprised of a consortium of microbes that collectively:1) release monosaccharides 10 from a complex mucilage polysaccharide,2) reduce oxygen tension in the mucilage,3) utilize released monosaccharides4) fix atmospheric N2.We willtest these functionalities and identify candidate members exhibiting these functionalities from a collection of over 1500microbes that we have isolated from this indigenous landrace of maize.
Project Methods
p.p1 {margin: 0.0px 0.0px 0.0px 0.0px; font: 12.0px Helvetica} span.s1 {font: 8.0px Helvetica}Whole genome sequence (WGS) will be done as described for the 100K Pathogen Genome Project. Genomic DNA will be extracted from each isolate, quality assessed, and sequencing libraries will be constructed using the KAPA HyperPlus DNA library kit. The gDNA libraries will be pooled for multiplex sequencing on the Illumina HiSeq XTen using paired end 150 bp with a target of 60x genome coverage for each individual to produce draft genomes with a target of <50 contigs. For a selection of isolates that are novel or unique, we will produce complete genomes using nanopore sequencing (Oxford Nanopore; Weimer unpublished). The paired fastq files for each isolate will be screened for Phi-X sequence contamination and trimmed using Trimmomatic. The resulting reads from each library will be assembled into contiguous DNA sequences using the graph-based de novo assembler MEGAhit (62) followed by taxonomic classification and distribution using k-mers (31mers) with Kraken. For taxonomic diversity estimates, Sourmash will be used to generate minhash sketches, as well as define new genes within clusters that are linked to novelty and candidate N2 fixation gene set diversification (53, 54). These sketches will be queried against a database of similar sketches derived from all microbial genomes (refseqcomplete) at NCBI. The assemblies of prokaryotic isolate genomes (Aim 1) will be annotated using Prokka for prokaryotic genome annotation (68). Fungal isolate genomes assemblies will be annotated using FunGAP in conjunction with the relevant transcriptome and protein datasets that will be downloaded from the SRA.p.p1 {margin: 0.0px 0.0px 0.0px 0.0px; font: 12.0px Helvetica} span.s1 {font: 8.0px Helvetica}Mucilage digestion: From the WGS and metagenome (Aim 4) we will annotate the microbial genes for glycosyl hydrolases (GH) and assay each microbial isolate for its ability to grow on a medium with the mucilage polysaccharide presented as the sole carbon source and tocatalyze the release of monosaccharides from the complex polysaccharide. The annotations will be done as described (Aim 1). The multi-protein fasta amino acid (.faa) files of each isolate genome that contain the annotated protein coding features will then be queried against the most current version of the dbCAN library of profile hidden markov models (pHMMs) for each carbohydrate active enzyme family in the CAZy database using the hmmscan feature of HMMER v.3.1. Hmmscan output files will be parsed using custom shell scripts, and all matches will be filtered using the R package dplyr to eliminate matches with E-values >1e-18 and model coverage <85%. Furthermore, the remaining dbCAN query records for each isolate will be analyzed and visualized in R to generate individual profiles of GH genes for each mucilage-derived bacterial isolate that can be compared.p.p1 {margin: 0.0px 0.0px 0.0px 0.0px; font: 12.0px Helvetica} span.s1 {font: 8.0px Helvetica}Cytochrome bd oxidase gene discovery approach: Similar to the method described in Aims1 and 2 for the annotation of GH activity, annotated protein sequences for each isolate genome will be queried against pHMMs for the following genes that comprise the cytochrome bd oxidasesystem of the model organism Escherichia coli (37, 81, 82) using the hmmscan function of the program HMMER v3.1 for CydA (Pfam - PF01654), CydB (Pfam - PF02322 and TIGRFAM - TIGR00203), CydC (TIGRFAM - TIGR02868), CydD (TIGRFAM - TIGR02857), CydX/YbgT(Pfam - PF08173; TIGRFAM - TIGR02106), and YbgE (TIGRFAM - TIGR02112). Output files from each genome scan will also be analyzed in R to compare all isolates based on their cytochrome bd oxidase gene profiles following stringent filtering of recorded gene model hits with90% pHMM coverage and a maximum e-value of 1e-18.O2 reduction assay approach: Microbial isolates will be assayed for their capacity to deplete O2 in the growth medium (83) using the GreenLight probe, whose phosphorescence is quenched by O2 so that O2 depletion results in enhanced phosphorescence. The probe has been shown to reliably report O2 levels in diverse and complex media. Microbial isolates will be grown in Luria's broth (LB) at 28ÅãC to OD600 = 0.8, centrifuged and the pellets washed and resuspended in NM minimal salt media containing fucose, arabinose and glucose as carbon sources. GreenLight probe will be reconstituted in PBS and added to the medium to a final concentration of 100 nmol/L. The resulting culture (90 μl) will be transferred to wells of a 384 well plate in triplicate, sealed with heavy mineral oil and monitored by time-resolved fluorescence with 340nm excitation and 642nmemission filters at 10 min intervals. Growth (OD600) will also be monitored and used to normalize samples since respiration may be accelerated simply related to the growth rate of the microbial isolate.p.p1 {margin: 0.0px 0.0px 0.0px 0.0px; font: 12.0px Helvetica} span.s1 {font: 8.0px Helvetica}Metagenome sequencing: Shotgun DNA and metaRNAseq metagenomics using aerial root mucilage will begin with samples collected from three different sources: 1) non-fixing field of Sierra Mixe region, 2) fixing field of the Sierra Mixe region, and 3) aerial root mucilage produced in Davis, CA. Samples from the Sierra Mixe region have been collected, frozen, and are ready for total DNA extraction using (Weimer, unpublished data). This comprises ~50 samples to select from for production of the DNA metagenome. The Weimer lab has demonstrated that although both DNA and RNA from the same sample will provide the necessary information to define the community, DNA overestimates the fraction of active bacteria in the microbiome. However, samples from the Sierra Mixe region can only be used for DNA analysis. The samples from Davis will be used for DNA and metaRNAseq. Performing RNAseq allows examination of the microbiome membership, as well as metabolic functional analysis (63, 64, 92). The sequencing libraries will be generated in the Weimer lab and sequenced on an Illumina NovaSeq with a minimum of ~150 million reads/sample using indexing. Bacterial identification will be done using 14 Kraken and Refseq complete as the reference database (64) that has proven accuracy, in which the maize reads will be removed before the microbial ID will be done to the genus level using Kraken and a maize-specific database. The RNA sequence will be used to examine specific gene sets to estimate functional pathways; for example, N2 fixation, oligosaccharide digestion, amino acidmetabolism, and others. We validated this analytical technique using food samples to examine pathogens and were successful in using the exact informatics pipeline and in robustly isolating transcripts within the samples to refine the bacterial phylogenomics. We will determine the specific genera that are unique across N-fixation and isolates in the collection to examine genera that are coordinately enriched or repressed with N-fixation capability. Using the metaRNAseq metagenome, microbiome metabolic mapping will be performed using established gene expression analysis methods (93) to examine the metabolic options to produce compounds, other than ammonia, as in this proposal.