The Microbiome of Fertilization-Stage Maize Silks (Style) Encodes Genes and Expresses Traits That Potentially Promote Survival in Pollen/Style Niches and Host Reproduction.

Within flowers, the style channel receives pollen and transmits male gametes inside elongating pollen tubes to ovules. The styles of maize/corn are called silks. Fertilization-stage silks possess complex microbiomes, which may partially derive from pollen. These microbiomes lack functional analysis. We hypothesize that fertilization-stage silk microbiomes promote host fertilization to ensure their own vertical transmission. We further hypothesize that these microbes encode traits to survive stresses within the silk (water/nitrogen limitation) and pollen (dehydration/aluminum) habitats. Here, bacteria cultured from fertilization-stage silks of 14 North American maize genotypes underwent genome mining and functional testing, which revealed osmoprotection, nitrogen-fixation, and aluminum-tolerance traits. Bacteria contained auxin biosynthesis genes, and testing confirmed indole compound secretion, which is relevant, since pollen delivers auxin to silks to stimulate egg cell maturation. Some isolates encoded biosynthetic/transport compounds known to regulate pollen tube guidance/growth. The isolates encoded ACC deaminase, which degrades the precursor for ethylene that otherwise accelerates silk senescence. The findings suggest that members of the microbiome of fertilization-stage silks encode adaptations to survive the stress conditions of silk/pollen and have the potential to express signaling compounds known to impact reproduction. Overall, whereas these microbial traits have traditionally been assumed to primarily promote vegetative plant growth, this study proposes they may also play selfish roles during host reproduction.

Bacteria existing in pre-pollinated styles (silks) can defend the exposed male gamete fertilization channel of maize against an environmental Fusarium pathogen.

In flowering plants, fertilization requires exposing maternal style channels to the external environment to capture pollen and transmit its resident sperm nuclei to eggs. This results in progeny seed. However, environmental fungal pathogens invade developing seeds through the style. We hypothesized that prior to environmental exposure, style tissue already possesses bacteria that can protect styles and seed from such pathogens. We further hypothesized that farmers have been inadvertently selecting immature styles over many generations to have such bacteria. We tested these hypotheses in maize, a wind-pollinated crop, which has unusually long styles (silks) that are invaded by the economically-important fungal pathogen Fusarium graminearum (Fg). Here, unpollinated silk-associated bacteria were cultured from a wild teosinte ancestor of maize and diverse maize landraces selected by indigenous farmers across the Americas, grown in a common Canadian field for one season. The bacteria were taxonomically classified using 16S rRNA sequencing. In total, 201 bacteria were cultured, spanning 29 genera, 63 species, and 62 unique OTUs, dominated by Pseudomonas, Pantoea and Microbacterium. These bacteria were tested for their ability to suppress Fg in vitro which identified 10 strains belonging to 6 species: Rouxiella badensis, Pantoea ananatis, Pantoea dispersa, Pseudomonas koreensis, Rahnella aquatilis, and Ewingella americana. Two anti-Fg strains were sprayed onto silks before/after Fg inoculation, resulting in ≤90% reductions in disease (Gibberella ear rot) and 70-100% reductions in associated mycotoxins (deoxynivalenol and zearalenone) in progeny seeds. These strains also protected progeny seeds post-harvest. Confocal fluorescent imaging showed that one silk bacterium (Rouxiella AS112) colonized susceptible entry points of Fg on living silks including stigmatic trichomes, wounds, and epidermal surfaces where they formed thick biofilms. Post-infection, AS112 was associated with masses of dead Fg hyphae. These results suggest that the maize style (silk) is endowed with potent bacteria from the mother plant to protect itself and progeny from Fusarium. The evidence suggests this trait may have been selected by specific indigenous peoples, though this interpretation requires further study.

Maize pollen carry bacteria that suppress a fungal pathogen that enters through the male gamete fertilization route.

In flowering plants, after being released from pollen grains, the male gametes use the style channel to migrate towards the ovary where they fertilize awaiting eggs. Environmental pathogens exploit the style passage, resulting in diseased progeny seed. The belief is that pollen also transmits pathogens into the style. By contrast, we hypothesized that pollen carries beneficial microbes that suppress environmental pathogens on the style passage. No prior studies have reported pollen-associated bacterial functions in any plant species. Here, bacteria were cultured from maize (corn) pollen encompassing wild ancestors and farmer-selected landraces from across the Americas, grown in a common field in Canada for one season. In total, 298 bacterial isolates were cultured, spanning 45 genera, 103 species, and 88 OTUs, dominated by Pantoea, Bacillus, Pseudomonas, Erwinia, and Microbacterium. Full-length 16S DNA-based taxonomic profiling showed that 78% of bacterial taxa from the major wild ancestor of maize (Parviglumis teosinte) were present in at least one cultivated landrace. The species names of the bacterial isolates were used to search the pathogen literature systematically; this preliminary evidence predicted that the vast majority of the pollen-associated bacteria analyzed are not maize pathogens. The pollen-associated bacteria were tested in vitro against a style-invading Fusarium pathogen shown to cause Gibberella ear rot (GER): 14 isolates inhibited this pathogen. Genome mining showed that all the anti-Fusarium bacterial species encode phzF, associated with biosynthesis of the natural fungicide, phenazine. To mimic the male gamete migration route, three pollen-associated bacterial strains were sprayed onto styles (silks), followed by Fusarium inoculation; these bacteria reduced GER symptoms and mycotoxin accumulation in progeny seed. Confocal microscopy was used to search for direct evidence that pollen-associated bacteria can defend living silks against Fusarium graminearum (Fg); bacterial strain AS541 (Kluyvera intermedia), isolated from pollen of ancestral Parviglumis, was observed to colonize the susceptible style/silk entry points of Fg (silk epidermis, trichomes, wounds). Furthermore, on style/silk tissue, AS541 colonized/aggregated on Fg hyphae, and was associated with Fg hyphal breaks. These results suggest that pollen has the potential to carry bacteria that can defend the style/silk passage against an environmental pathogen - a novel observation.

Conservation and diversity of the pollen microbiome of Pan-American maize using PacBio and MiSeq.

Pollen is a vector for diversification, fitness-selection, and transmission of plant genetic material. The extent to which the pollen microbiome may contribute to host diversification is largely unknown, because pollen microbiome diversity within a plant species has not been reported, and studies have been limited to conventional short-read 16S rRNA gene sequencing (e.g., V4-MiSeq) which suffers from poor taxonomic resolution. Here we report the pollen microbiomes of 16 primitive and traditional accessions of maize (corn) selected by indigenous peoples across the Americas, along with the modern U.S. inbred B73. The maize pollen microbiome has not previously been reported. The pollen microbiomes were identified using full-length (FL) 16S rRNA gene PacBio SMRT sequencing compared to V4-MiSeq. The Pan-American maize pollen microbiome encompasses 765 taxa spanning 39 genera and 46 species, including known plant growth promoters, insect-obligates, plant pathogens, nitrogen-fixers and biocontrol agents. Eleven genera and 13 species composed the core microbiome. Of 765 taxa, 63% belonged to only four genera: 28% were Pantoea, 15% were Lactococcus, 11% were Pseudomonas, and 10% were Erwinia. Interestingly, of the 215 Pantoea taxa, 180 belonged to a single species, P. ananatis. Surprisingly, the diversity within P. ananatis ranged nearly 10-fold amongst the maize accessions analyzed (those with ≥3 replicates), despite being grown in a common field. The highest diversity within P. ananatis occurred in accessions that originated near the center of diversity of domesticated maize, with reduced diversity associated with the north-south migration of maize. This sub-species diversity was revealed by FL-PacBio but missed by V4-MiSeq. V4-MiSeq also mis-identified some dominant genera captured by FL-PacBio. The study, though limited to a single season and common field, provides initial evidence that pollen microbiomes reflect evolutionary and migratory relationships of their host plants.