Using molecular microbial ecology to define differential responses to the inoculation of barley silage

Previously, we investigated the impact of a mixed Lactobacillus buchneri, Lactobacillus plantarum, and Lactobacillus casei inoculant on fermentation and aerobic stability of barley silage over two years in 2009 and 2010. In 2009, a classical response to inoculation was obtained with an increase in acetic acid concentration of silage ensiled in both mini- and bag silos. In 2010, this classical response was not observed in mini-silos but was observed in bag silos. The objective of this study was to determine if molecular microbial ecology could explain the differential responses to the inoculation of barley silage between the two years. The Illumina MiSeq sequencing results showed that inoculation increased Lactobacillus and lowered Pediococcus, Weissella, and Leuconostoc in both types of silos in 2009. However, a similar trend was not observed in mini-silos, but was instead observed in bag silos in 2010. Inoculation did not alter the core fungal community in either silo type in either year. Cladosporium, Leptosphaeria, and Cryptococcus were abundant in fresh forage, but were superseded by Pichia and Kazachstania after ensiling. Our results suggest that changes in silage chemistry corresponded to differences observed in microbial ecology. Inoculation may have less impact when using more mature crops with shorter ensiling times.

Cell Envelope Integrity and Capsule Characterization of Rhodotorula mucilaginosa Strains from Clinical and Environmental Sources

Currently, there is very little known about the phenotypic variability within species of Rhodotorula strains and the role of their capsule. Cryptococcus neoformans has been considered the only encapsulated human fungal pathogen, but as more individuals come to live in states of immunocompromised health, they are more susceptible to fungal infections, including those by Rhodotorula. R. mucilaginosa species are some of those most commonly associated with clinical infections. We wanted to know if clinical and environmental strains of R. mucilaginosa demonstrated disparate capsule phenotypes. With limited antifungal options available and clinical Rhodotorula spp. often resistant to common antifungal drugs such as fluconazole, caspofungin (1, 2), and voriconazole (2), a better understanding of the fungal biology could inform the design and use of future antifungal drugs. The generation of an antibody specific to Rhodotorula fungi could be a useful diagnostic tool, and this work presents the first mention of such in the literature.

Rhodotorula yeasts are pink, encapsulated basidiomycetes isolated from a variety of environments and clinical settings. They are increasingly linked with disease, particularly central venous catheter infections and meningitis, in immunocompromised patients. Eight clinical and eight environmental strains molecularly typed as Rhodotorula mucilaginosa were compared to six Cryptococcus neoformans strains for phenotypic variability. Growth on cell integrity-challenging media suggested that R. mucilaginosa cells possess differences in signaling pathways, cell wall composition, or assembly and that their membranes are more susceptible to perturbations than those of C. neoformans. All 16 R. mucilaginosa strains produced urease, while none produced melanin with l-3,4-dihydroxyphenylalanine (l-DOPA) as a substrate. India ink staining reveals that clinical R. mucilaginosa capsules are larger than environmental capsules but that both are generally smaller than C. neoformans capsules. All R. mucilaginosa strains were resistant to fluconazole. Only two clinical strains were susceptible to voriconazole; all of the environmental strains were resistant. We generated an anticapsular antibody (Rh1) to R. mucilaginosa; Rh1 did not bind C. neoformans control strains, was specific to Rhodotorula species, and bound to all tested Rhodotorula strains. Binding assays performed with wheat germ agglutinin (WGA), concanavalin A (ConA), calcofluor white (CFW), and eosin Y dye (EY) cell surface probes suggested that chitin may be more accessible in R. mucilaginosa but that the total abundance of chitooligomers is less than in C. neoformans. This report describes a novel reagent that can be used to identify Rhodotorula species and lays the foundation for future cell envelope composition analysis.

IMPORTANCE Currently, there is very little known about the phenotypic variability within species of Rhodotorula strains and the role of their capsule. Cryptococcus neoformans has been considered the only encapsulated human fungal pathogen, but as more individuals come to live in states of immunocompromised health, they are more susceptible to fungal infections, including those by Rhodotorula. R. mucilaginosa species are some of those most commonly associated with clinical infections. We wanted to know if clinical and environmental strains of R. mucilaginosa demonstrated disparate capsule phenotypes. With limited antifungal options available and clinical Rhodotorula spp. often resistant to common antifungal drugs such as fluconazole, caspofungin (1, 2), and voriconazole (2), a better understanding of the fungal biology could inform the design and use of future antifungal drugs. The generation of an antibody specific to Rhodotorula fungi could be a useful diagnostic tool, and this work presents the first mention of such in the literature.

Bacterial and fungal core microbiomes associated with small grain silages during ensiling and aerobic spoilage

Background

Describing the microbial populations present in small grain silage and understanding their changes during ensiling is of interest for improving the nutrient value of these important forage crops. Barley, oat and triticale forages as well as an intercropped mixture of the 3 crops were harvested and ensiled in mini silos for a period of 90 days, followed by 14 days of aerobic exposure. Changes in fermentation characteristics and nutritive value were assessed in terminal silages and bacterial and fungal communities during ensiling and aerobic exposure were described using 16S and 18S rDNA sequencing, respectively.

Results

All small grain silages exhibited chemical traits that were associated with well ensiled forages, such as low pH value (4.09 ± 0.28) and high levels of lactic acid (59.8 ± 14.59 mg/g DM). The number of microbial core genome operational taxonomic units (OTUs) decreased with time of ensiling. Taxonomic bacterial community profiles were dominated by the Lactobacillales after fermentation, with a notable increase in Bacillales as a result of aerobic exposure. Diversity of the fungal core microbiome was shown to also be reduced during ensiling. Operational taxonomic units assigned to filamentous fungi were found in the core microbiome at ensiling and after aerobic exposure, whereas the Saccharomycetales were the dominate yeast population after 90 days of ensiling and aerobic exposure. Bacterial and fungal orders typically associated with silage spoilage were identified in the core microbiome after aerobic exposure.

Conclusion

Next Generation Sequencing was successfully used to describe bacterial communities and the first record of fungal communities throughout the process of ensiling and utilization. Adequately describing the microbial ecology of silages could lead to improved ensiling practices and the selection of silage inoculants that act synergistically with the natural forage microbiome.

Electronic supplementary material

The online version of this article (doi:10.1186/s12866-017-0947-0) contains supplementary material, which is available to authorized users.

Variable adhesion and diurnal population patterns of epiphytic yeasts on creeping bentgrass

Irrigation and an in vitro agitation assay were used to determine the percentage of the epiphytic yeast community (Cryptococcus, Pseudozyma, Rhodotorula, and Sporobolomyces) adhering to the phylloplane of creeping bentgrass (Agrostis palustris (Huds.) Pers.). Colony-forming units (cfu) of total epiphytic yeast populations (adherent and nonadherent cells) and of adherent populations (cells not removed by agitation) were determined by leaf washing and dilution plating. In an in vitro assay, 40.0% and 57.1% of the yeast adhered to the leaves, whereas, in initial field trials the percentage of adherent yeasts ranged from 40.0% to 71.9% of the total population. Adherent yeast cfu on leaves in the morning were significantly lower on bentgrass (8.0 × 103to 3.1 × 104cfu·cm–2) compared with total yeast cfu (1.4 × 104to 4.7 × 104cfu·cm–2) on the nonirrigated control. No differences in yeast populations were observed between irrigated and nonirrigated plots 2 h after the 0900 treatments. Yeast populations followed a diurnal pattern, with larger cfu recovered from bentgrass leaves in the morning and significantly lower populations recovered in the afternoon. At 1400 the adherent yeast were 83.1%–100% of the total yeast population recovered from the leaves. The relative adhesiveness of the epiphytic yeast community on bentgrass leaves is dynamic with nonadherent cells making up a larger percentage of the population in the mornings than the afternoons.Key words: adherence, Cryptococcus, leaf surface, Rhodotorula, turfgrass.

Effect of foliar disease on the epiphytic yeast communities of creeping bentgrass and tall fescue

The effect of mechanical wounding or foliar diseases caused by Sclerotinia homoeocarpa or Rhizoctonia solani on the epiphytic yeast communities on creeping bentgrass and tall fescue were determined by leaf washing and dilution plating. Total yeast communities on healthy bentgrass and tall fescue leaves ranged from 7.9 x 103 to 1.4 x 105 CFU.cm-2 and from 2.4 x 103 to 1.6 x 104 CFU.cm-2, respectively. Mechanically wounded leaves (1 of 2 trials) and leaves with disease lesions (11 of 12 trials) supported significantly larger communities of phylloplane yeasts. Total yeast communities on S. homoeocarpa infected or R. solani infected bentgrass leaves were 3.6-10.2 times and 6.2-6.4 times larger, respectively, than the communities on healthy leaves. In general, healthy and diseased bentgrass leaves supported larger yeast communities than healthy or diseased tall fescue leaves. We categorized the majority of yeasts as white-pigmented species, including Cryptococcus laurentii, Cryptococcus flavus, Pseudozyma antarctica, Pseudozyma aphidis, and Pseudozyma parantarctica. The percentage of pink yeasts in the total yeast community ranged from 2.6% to 9.9% on healthy leaves and increased to 32.0%-44.7% on S. homoeocarpa infected leaves. Pink-pigmented yeasts included Rhodotorula glutinis, Rhodotorula mucilaginosa, Sakaguchia dacryoidea, and Sporidiobolus pararoseus. Foliar disease significantly affected community size and composition of epiphytic yeasts on bentgrass and tall fescue.

Development of real-time PCR using Minor Groove Binding probe to monitor the biological control agent Candida oleophila (strain O)

A real-time PCR assay using a 3'-Minor Groove Binding (MGB) probe was developed for specific detection and monitoring of Candida oleophila (strain O), a biocontrol agent against Botrytis cinerea and Penicillium expansum, on harvested apples. The application of the RAPD technique on C. oleophila strains followed by reproducible sequence characterized amplified region (SCAR) amplifications allowed the identification of a semi-specific fragment of 244 bp, observed in the profiles of strain O and three other C. oleophila strains. After sequencing, polymorphisms (3%) were observed between the strain O sequence and the three other sequences. A 3'-Minor Groove Binding probe was designed to specifically match a region of the strain O sequence and was able to discriminate a single base mutation or a two-base difference in the corresponding sequences of the non-target strains. This specific detection method was applied to monitor strain O population, recovered by a washing buffer, from harvested apples. Population densities were calculated using an external standard curve consisting in a serial dilution of strain O cells in the washing buffer from untreated apples. Linearity in the standard curve was kept between 1.64 x 10(2) and 1.64 x 10(5) cfu cm(-2) of apple surface. During a first practical experiment, the calculated population densities were similar to those obtained by plating on semi-selective media. This new real-time PCR method is a promising tool to monitor quickly and specifically strain O population on apple surface in middle- or large-scale experiments.

In vitro attachment of phylloplane yeasts toBotrytis cinerea,Rhizoctonia solani, andSclerotinia homoeocarpa

The ability of yeasts to attach to hyphae or conidia of phytopathogenic fungi has been speculated to contribute to biocontrol activity on plant surfaces. Attachment of phylloplane yeasts to Botrytis cinerea, Rhizoctonia solani, and Sclerotinia homoeocarpa was determined using in vitro attachment assays. Yeasts were incubated for 2 d on potato dextrose agar (PDA) prior to experimentation. A total of 292 yeasts cultured on PDA were screened for their ability to attach to conidia of B. cinerea; 260 isolates (89.1%) attached to conidia forming large aggregates of cells, and 22 isolates (7.5%) weakly attached to conidia with 1 or 2 yeast cells attached to a few conidia. Ten yeasts (3.4%), including 8 isolates of Cryptococcus laurentii, 1 isolate of Cryptococcus flavescens, and an unidentified species of Cryptococcus, failed to attach to conidia. All non-attaching yeasts produced copious extracellular polysaccharide (EPS) on PDA. Seventeen yeast isolates did not attach to hyphal fragments of B. cinerea, R. solani, and S. homoeocarpa after a 1 h incubation, but attachment was observed after 24 h. Culture medium, but not culture age, significantly affected the attachment of yeast cells to conidia of B. cinerea. The 10 yeast isolates that did not attach to conidia when grown on agar did attach to conidia (20%–57% of conidia with attached yeast cells) when cultured in liquid medium. Attachment of the biocontrol yeast Rhodotorula glutinis PM4 to conidia of B. cinerea was significantly greater at 1 × 107yeast cells·mL–1than at lower concentrations of yeast cells. The ability of yeast cells to attach to fungal conidia or hyphae appears to be a common phenotype among phylloplane yeasts.Key words: adhesion, biological control, Cryptococcus laurentii, Rhodotorula glutinis.

Biological control of Fusarium head blight of wheat and deoxynivalenol levels in grain via use of microbial antagonists

Efforts to reduce mycotoxin contamination in food logically start with minimizing plant infection by mycotoxin producing pathogens. Fusarium graminearum (perfect state, Gibberella zeae) infects wheat heads at flowering, causing the disease Fusarium head blight (FHB) and losses of over 2.6 billion dollars in the U.S. during the last 10 years. The pathogen often produces deoxynivalenol (DON) resulting in grain size and quality reduction. Highly resistant wheat cultivars currently are not available for reducing FHB, and labeled fungicides are not consistently effective. The feasibility of biologically controlling FHB is currently being evaluated. Microbial isolates obtained from wheat anthers were screened for their ability to utilize tartaric acid, a compound that is poorly utilized by F. graminearum and could be utilized in formulations of biological control agents. Four strains that utilized tartaric acid and three that did not were effective in reducing FHB disease severity by up to 95% in greenhouse and 56% in field trials. Additional research programs around the globe have identified other antagonist strains with potential for biologically controlling FHB. Though a considerable body of research remains to be completed, strategies and microorganisms for biologically controlling FHB have reached an advanced stage of development and offer the promise of being an effective tool that could soon contribute to the reduction of FHB severity and DON contamination of grain in commercial agriculture.

In vitro antagonism of Botrytis cinerea by phylloplane yeasts

The influence of inoculum and nutrient concentrations on the in vitro antagonism of Botrytis cinerea by phylloplane yeasts was investigated with detached leaf disk and conidial germination assays. Rhodosporidium toruloides Y-1091 significantly reduced lesion development after 96 h on geranium leaf disks when co-inoculated at 1 × 106 or 1 × 107 yeast cells/mL with B. cinerea conidia at 1 × 104 or 1 × 105 conidia/mL. No effect on lesion development was observed when yeast and fungus were co-inoculated at a 1:1 ratio. Biocontrol activity of R. toruloides was greatest in 20 mM glucose and 20-fold dilute yeast nitrogen base. Twenty-five phylloplane yeasts exhibited a wide range of biocontrol activity when screened for antagonism of B. cinerea on geranium leaf disks (1 × 106 yeast cells/mL, 1 × 105 conidia/mL). Lesion development was significantly reduced by yeasts initially identified as poor antagonists when inoculated at higher concentrations (5 × 106 or 1 × 107 yeast cells/mL) with B. cinerea. Both poor and good antagonists significantly reduced in vitro germination of B. cinerea conidia. The presence of B. cinerea conidia had a greater effect on the growth of two poor antagonists compared with two good antagonists on leaf disks. These data suggest that many phylloplane yeasts will antagonize B. cinerea under conditions of low nutrient availability and with high antagonist to pathogen ratios.Key words: yeast, biological control, competition, nutrients, germination, Rhodosporidium toruloides.

The effects of fungicides on the phylloplane yeast populations of creeping bentgrass

The effects of fungicides on population size and the development of fungicide resistance in the phylloplane yeast flora of bentgrass was investigated. In the spring of 2001, azoxystrobin, chlorothalonil, flutolanil, and propiconazole were applied separately over a 6-week period to creeping bentgrass (Agrostis palustris Huds.). Total and fungicide-resistant yeast populations were assessed by dilution plating onto either potato dextrose agar or potato dextrose agar amended with the test fungicides. Total yeast populations in the fungicide-treated plots were significantly lower than the check plots on three out of four sample dates. In the fall, azoxystrobin or propiconazole were applied twice to the bentgrass over 3 weeks. Significantly larger total yeast populations were observed compared with resistant or highly resistant populations for each treatment on every sample date. Total yeast populations were significantly higher in the check plots compared with either the propiconazole- or azoxystrobin-treated plots on the first three of five sample dates. A collection of yeasts (N = 114) with no prior exposure to fungicides were more sensitive to chlorothalonil, propiconazole, flutolanil, and iprodione than a second group (N = 115) isolated from fungicide-treated turfgrass. These results suggest that fungicide resistance among phylloplane yeasts is widespread and could be an important factor in the development of biological control agents for turfgrass diseases.

The Ecology and Biogeography of Microorganisms on Plant Surfaces

The vast surface of the plant axis, stretching from root tips occasionally buried deeply in anoxic sediment, to apical meristems held far aloft, provides an extraordinarily diverse habitat for microorganisms. Each zone has to a greater or lesser extent its own cohort of microorganisms, in aggregate comprising representatives from all three primary domains of life-Bacteria, Archaea, and Eucarya. While the plant sets the stage for its microbial inhabitants, they, in turn, have established varied relationships with their large partner. These associations range from relatively inconsequential (transient epiphytic saprophytes) to substantial (epiphytic commensals, mutualistic symbionts, endophytes, or pathogens). Through recent technological breakthroughs, a much better perspective is beginning to emerge on the nature of these relationships, but still relatively little is known about the role of epiphytic microbial associations in the life of the plant.