Machine learning for exploring microbial inter-kingdom associations in Cystic Fibrosis and Bronchiectasis

Background: Cystic fibrosis (CF) and non-CF bronchiectasis (BX) are lung diseases characterised by severe chronic infections. Fungal and bacterial components of infection are both recognized. Recent molecular investigation of sputum from patients with CF and BX has revealed a complex mycobiome. However, little is known about how fungal and bacterial organisms interact or whether the interactions impact on disease outcomes. Methods: Quantitative PCR and next generation sequencing of ITS2 and 16S rRNA gene was carried out on 107 patients with CF and BX and defined clinical fungal infection status. Fungal and bacterial communities were explored using supervised and unsupervised machine learning to understand associations between fungal and bacterial communities and their relationship to disease. Results: Fungal and bacterial communities both had significantly higher biomass and lower diversity in CF compared to BX patients. Random forest modelling demonstrated that the fungal and bacterial communities were distinct between CF and BX patients. Within the CF group, bacterial communities contained no predictive signal for fungal disease status. Neither bacterial nor fungal community composition were predictive of the presence of CF pulmonary exacerbation (CFPE). Intra-kingdom correlations were far stronger than those between the two kingdoms. Dirichlet mixture components analysis identified two distinct clusters of bacteria related to the relative abundance of Pseudomonas. Fungal community composition contained no predictive signal for bacterial clusters. Conclusions: Clear changes in diversity were observed between patients with different clinical disease status. Although our results demonstrate that bacterial community composition differs in the presence of fungal disease, no direct relationship between bacterial and fungal OTUs was found.

Fungal Perspective of Pine and Oak Colonization in Mediterranean Degraded Ecosystems

Forest restoration has become one of the most important challenges for restoration ecology in the recent years. In this regard, soil fungi are fundamental drivers of forest ecosystem processes, with significant implications for plant growth and survival. However, the post-disturbance recovery of belowground communities has been rarely assessed, especially in highly degraded systems such as mines. Our aim was to compare forests and mined systems for biomass and structure of fungal communities in soil during early stages of tree establishment after disturbance. We performed ergosterol analysis and PacBio and Illumina sequencing of internal transcribed spacer 2 amplicons across soil layers in P. sylvestris, Q. robur and Q. ilex (holm oak) forests and naturally revegetated mined sites. In pine forests, total fungal biomass was significantly higher in litter and humus compared to mineral layers, with dominance of the mycorrhizal genera Tomentella, Inocybe and Tricholoma. Conversely, in oak forests the most abundant mycorrhizal genera were Tomentella, Cortinarius and Sebacina, but the biomass of saprotrophic fungi was greater in the litter layer compared to mycorrhizal fungi, with the genus Preussia being the most abundant. In the revegetated mined sites, ectomycorrhizal fungi dominated in the humus and mineral layers, with the mycorrhizal genus Oidiodendron being dominant. In contrast, in holm oak forests saprotrophic fungi dominated both soil humus and mineral layers, with the genera of Alternaria, Bovista and Mycena dominating the soil humus forest layer, while the genus Cadophora dominated the mineral layer. The habitat-specific differences in soil fungal community composition and putative functions suggest that an understanding of soil–plant–microbial interactions for different tree species and use of specific soil/litter inoculum upon planting/seeding might help to increase the effectiveness of tree restoration strategies in Mediterranean degraded sites.

Colonic Mucosa-associated Mycobiota in Individuals With Normal Colons

To characterize the spatial variation of the mucosa-associated adherent mycobiota along the large intestine in individuals with a normal-colon, we performed eukaryotic rRNA operon’s internal transcribed spacer-2 sequencing to profile fungal community composition and structure in 70 mucosal biopsies taken from the cecum, ascending, transverse, descending colon, and rectum of 14 polyp-free individuals. The bacteriome of these samples was previously characterized by sequencing the V4 region of the 16S rRNA gene. We identified 64 amplicon sequence variants (ASVs) with the relative abundance no less than 0.05% from these colonic mucosa samples. Each individual has a unique community composition of the gut mycobiome (P = 0.001 for beta diversity). Alpha-diversity and beta-diversity did not differ significantly across the colon segments. The most common phyla (relative abundance) were Ascomycota (45.4%) and Basidiomycota (45.3%). The most common genera were Malassezia (28.2%) and Candida (13.4%). Malassezia was found in 13 of 14 individuals. Other fungi genera were sporadically found in the large intestine. The most common species were Malassezia restricta (22.7%), Candida albicans (11.9%), Malasseziales sp. (8.80%), unclassified fungi (7.80%), and Penicillium paneum (5.70%). Malasseziaceae was co-abundant with Enterobacteriaceae and co-exclusive with Barnesiellaceae, Rikenellaceae, and Acidaminococcaceae. Malassezia was widely colonized whereas other fungal genera were sporadically colonized in the large intestine. The physiologic and pathogenic functions of fungi in human gastrointestinal tract including Malasseziaceae that may interact with several bacterial families remain to be fully elucidated.

Bacteria Respond Stronger Than Fungi Across a Steep Wood Ash-Driven pH Gradient

Soil pH is probably the most important variable explaining bacterial richness and community composition locally as well as globally. In contrast, pH effects on fungi appear to be less pronounced, but also less studied. Here we analyze the community responses of bacteria and fungi in parallel over a local extreme pH gradient ranging from 4 to 8. We established the pH gradient by applying strongly alkaline wood ash in dosages of 0, 3, 9, 15, 30, and 90 t ha–1 to replicated plots in a Picea abies plantation and assessed bacterial and fungal community composition using high throughput amplicon sequencing 1 year after ash application. At the same time, the experiment investigated if returning wood ash to plantation forests pose any immediate threats for the microbial communities. Among the measured environmental parameters, pH was by far the major driver of the microbial communities, however, bacterial and fungal communities responded differently to the pH increment. Whereas both bacterial and fungal communities showed directional changes correlated with the wood ash-induced increase in pH, the bacterial community displayed large changes at wood ash dosages of 9 and 15 t ha–1 while only higher dosages (>30 t ha–1) significantly changed the fungal community. The results confirm that fungi are less sensitive to pH changes than bacteria but also that fertilizing plantation forests with wood ash, viewed through the lens of microbial community changes, is a safe management at standard dosages (typically 3 t ha–1).

Fungal Communities in the Native New Zealand Medicinal Plant Pseudowintera colorata (Horopito) Are Determined by Plant Organ Type and Host Maturity with Key Members Promoting Plant Growth

The plant Pseudowintera colorata is well known for its antimicrobial and medicinal properties and is endemic to New Zealand. Using PCR-Denaturing gradient gel electrophoresis (DGGE), we investigated the factors influencing the composition of endophytic fungal communities in P. colorata from ten distinct sites across New Zealand. Our results showed that plant organs of P. colorata influenced the diversity and richness of endophytic fungi (PERMANOVA, p < 0.05). In addition, plant maturity and its interactions revealed that endophytic fungal communities formed discrete clusters in leaves, stems, and roots of mature and immature P. colorata plants (PERMANOVA; p = 0.002, p = 0.001 and p = 0.039, respectively). For identifying isolates with biocontrol potential, dual culture tests were set up against four different phytopathogenic fungi. Isolates with high activity (zone of inhibition >10 mm) were sequenced and identified as Trichoderma harzianum, Pezicula neosporulosa, Fusarium tricinctum, Metarhizium sp., and Chaetomium sp. Applying selected endophytic fungi (n = 7) as soil drenchers significantly increased the growth of P. colorata seedlings and produced more internodes. Seedling shoots treated with Trichoderma sp. PRY2BA21 were 2.2 × longer (8.36 cm) than the untreated controls (3.72 cm). Our results elucidate the main plant factors influencing fungal community composition and demonstrate a role for endophytic fungi in P. colorata growth and further demonstrate that medicinal plants are a rich source of endophytes with potential as biocontrol agents.

Nitrogen Fertiliser Immobilisation and Uptake in the Rhizospheres of Wheat and Canola

Immobilisation of fertiliser nitrogen (N) by soil microorganisms can reduce N availability to crops, decreasing growth and yield. To date, few studies have focussed on the effect of different plant species on immobilisation of fertiliser N. Canola (Brassica napus) is known to influence the soil microbiome and increase mineral N in soil for future crops compared with cereals. We tested the hypothesis that canola can reduce immobilisation of fertiliser N by influencing the composition of the rhizosphere microbiome. To investigate this, we conducted a glasshouse soil column experiment comparing N fertiliser uptake between canola and wheat (Triticum aestivium) and partitioning of fertiliser N between plants and microorganisms. Plants were grown in soil to which high C:N ratio wheat residues and 15N-labelled urea fertiliser were applied. There was no difference between wheat and canola in fertiliser N uptake despite differences in fungal community composition and the carbon metabolising enzyme alpha-glucosidase in the rhizosphere. Canola obtained more soil-derived N than wheat. There was no significant difference in the rhizosphere bacterial communities present between wheat and canola and unplanted controls. Our results highlight the capacity of canola to increase mineralisation of soil N compared with wheat although the study could not describe the microbial community which facilitated this increase.

Rhizospheric Fungal Diversities and Soil Biochemical Factors of Fritillaria taipaiensis over Five Cultivation Years

Fritillaria taipaiensis is a valuable traditional Chinese medicinal plant that has experienced continuous decline over its cropping area. The present study aims to explore the reasons for the quality and growth decline. The fungal diversities and biochemical factors in its rhizospheric soils with cultivation duration from 1 up to 5 years were analyzed and compared. The results showed that rhizospheric fungi of F. taipaiensis belong to six phyla, including Neocallimastigomycota, Glomeromycota, Basidiomycota, Chytridiomycota, Zygomycota and Ascomycota. Thirteen genera (Pseudogymnoascus, Fusarium, Mortierella, Colletotrichum, Laetinaevia, Gibberella, Synchytrium, Lysurus, Trichocladium, Volutella, Monoblepharis, Aquamyces and Trichoderma) constituted the “core community” in the rhizosphere of F. taipaiensis. The dominant fungal genera varied significantly in rhizospheric soils with different cultivation years. The abundance of fungal species in the soil declined with the cultivation year generally. The pH, available P, organic matter and urease activity were the primary factors determining the fungal community composition in the rhizosphere. The content of organic matter, available N, P and K and the activities of urease and alkaline phosphatase decreased with cultivation years. The soil pH increased with cultivation years and was unsuitable for F. taipaiensis growth. These features suggested that long-term single planting altered the fungal community structure, fertility conditions and soil enzyme activities in F. taipaiensis rhizospheric soils, which could be detrimental for plant growth and quality.

Metabarcoding of Soil Fungal Communities from Perturbed Areas Reveal Drive Role of Environmental Factors in Microbial Diversity of Colombian Andosols

Changes in soil fungal community caused by land use have not been sufficiently studied in South-American Andosols, considered globally as important food production areas. This study analyzed 26 soil samples of Andosols collected from locations devoted to conservation, agriculture and mining activities in the southeastern region of Antioquia, Colombia, to establish differences between fungal communities as indicators of the degree of soil perturbation. The study developed a novel heminested PCR with primers SSUmCf Mix, ITS4 and fITS7 to assess Arbuscular Mycorrhizal Fungi detection in a Illumina MiSeq metabarcoding on nuclear ribosomal ITS2 region. A non-metric multidimensional scaling allowed exploring driver factors of fungal community changes, while fitted Dirichlet-multinomial models and PERMANOVA tests allowed identifying the correlations between alpha diversity indexes and community dissimilarities, as well as the significance of land use effects on fungal community composition. Furthermore, response ratios were determined to assess effect size by land use over relevant taxa. Results suggest a good coverage of fungal diversity with a detection of 10,529 high-quality ITS2 sequences belonged to phylum Glomeromycota. The analysis shows strong correlations of Shannon and Fisher indexes with dissimilarities on fungal communities among land uses (r=0.94), related to variations in temperature, air humidity and organic matter contents that lead to significant responses in abundances of relevant orders (such as Wallemiales and Trichosporonales). The study highlights the rich fungal biodiversity of the tropical Andosols, their specific sensitivities to environmental perturbation factors, and the useful range of a metabarcoding approach to characterize soil fungal communities.

Changes in Soil-Borne Communities of Arbuscular Mycorrhizal Fungi during Natural Regrowth of Abandoned Cattle Pastures Are Indicative of Ecosystem Restoration

Natural restoration of ecosystems includes the restoration of plant-microbial associations; however, few studies had documented those changes in tropical ecosystems. With the aim to contribute to understand soil microbial changes in a natural regrowth succession of degraded pastures that were left for natural restoration, we studied changes in arbuscular mycorrhizal (AM) fungal communities. Arbuscular mycorrhizal fungi (AMF) establish a mutualistic symbiosis with plants, improving plant nutrition. Amplification of the small subunit rRNA with specific primers and subsequent Illumina sequencing were used to search soil-borne AM fungal communities in four successional natural regrowth stages in two landscapes (hill and mountain) with soil differences, located in the Andean-Amazonian transition. Molecular results corroborated the results obtained previously by spores-dependent approaches. More abundance and virtual taxa of AMF exist in the soil of degraded pastures and early natural regrowth stages than in old-growth or mature forest soils. Although changes in AM fungal communities occurred similarly over the natural regrowth chronosequence, differences in soil texture between landscapes was an important soil feature differentiating AM fungal community composition and richness. Changes in soil-borne AM fungal communities reflect some signals of environmental restoration that had not been described before, such as the reduction of Glomus dominance and the increase of Paraglomus representativeness in the AM fungal community during the natural regrowth chronosequence.