The Brazilian campos rupestres, a unique grassland ecosystem situated on ancient rocky outcrops in central and eastern Brazil, present a significant ecological challenge: extremely low concentrations of essential nutrients, particularly phosphorus (P) and nitrogen (N), severely limit plant growth. These nutrient-poor conditions arise from intense weathering and leaching of parent rock, leading to low P availability and binding to iron and aluminum due to low soil pH. Paradoxically, despite these harsh abiotic constraints, campos rupestres boast remarkable biodiversity, harboring a disproportionately high number of endemic plant species, particularly within specialized lineages such as the Velloziaceae family. This high level of biodiversity in a nutrient-deprived environment raises a critical question: how do plants in this ecosystem acquire sufficient nutrients for survival and growth? While plant adaptations to nutrient limitation have been extensively studied (e.g., formation of durable structures, efficient P remobilization, specialized root systems), the role of plant-microbe interactions in nutrient acquisition within this ecosystem remains largely unexplored. This study aims to address this gap in knowledge by investigating the composition and function of microbial communities associated with two dominant Velloziaceae species growing on contrasting substrates (soil and rock) to determine their contribution to nutrient turnover and plant growth. We hypothesize that these plant species have actively selected and established associations with beneficial microorganisms capable of enhancing nutrient availability, and that these associations are influenced by the type of substrate.

Previous research has highlighted the significant adaptations employed by campos rupestres plants to cope with nutrient scarcity. Members of the Velloziaceae family, for example, exhibit several strategies, including the development of robust, well-protected structures, efficient remobilization of P from senescent leaves, and specialized root systems that enhance nutrient uptake through the secretion of carboxylates. Studies have focused primarily on plant-centric mechanisms, neglecting the crucial role of plant-microbe interactions. However, a growing body of research emphasizes the fundamental role of plant microbiomes in nutrient acquisition, particularly in stressful environments. Plant microbiomes are not simply random assemblages; they are shaped by intricate interspecies relationships and can significantly influence plant speciation, geographic distribution, and overall diversity. In other nutrient-limited ecosystems like grasslands and boreal forests, nitrogen-fixing bacteria and mycorrhizal fungi have been shown to substantially contribute to the overall nitrogen and phosphorus acquired by plants. While previous work has suggested that many campos rupestres plants lack mycorrhizal associations, a comprehensive understanding of the diversity and functions of plant-associated microorganisms in this ecosystem is lacking. Therefore, this study seeks to provide a high-throughput investigation of these microbial communities and their functional potential in nutrient acquisition and cycling.

This study utilized a combination of amplicon sequencing and metagenomics to characterize the microbial communities associated with two Velloziaceae species, *Vellozia epidendroides* (soil) and *Barbacenia macrantha* (rock), growing on contrasting substrates in the campos rupestres. Samples of substrate (soil and rock) and plant tissues (roots, stems, leaves – both external and internal) were collected from six individuals of each species. Environmental DNA was extracted, and amplicon sequencing of the 16S rRNA gene (for prokaryotes) and the ITS2 region (for fungi) was performed using the Illumina MiSeq platform. Metagenomic libraries were also generated for the external root and substrate communities from three individuals of each species, sequenced using Illumina HiSeq, and assembled using MEGAHIT. A total of 16 metagenomes were assembled, resulting in 522 high-quality metagenome-assembled genomes (MAGs). Amplicon sequence variants (ASVs) were inferred using DADA2, and taxonomic assignment was performed using the GTDB and UNITE databases. Weighted average community identity (WACI) was calculated to quantify taxonomic novelty. Phylogenetic novelty was also assessed using the phylogenetic gain metric. Functional analysis was performed by annotating metagenomic assemblies and MAGs using various databases (KEGG, Pfam, TIGRFAM). The abundance of genes involved in various processes, including nutrient transport, carbon cycling, phosphorus turnover, and nitrogen cycling, was quantified using the RPKG (reads per kilobase per genome equivalent) metric. Statistical analyses including PERMANOVA, linear mixed-effects models (LMMs), phylogenetic regressions, and hypergeometric tests were employed to assess differences in community composition, diversity, and functional potential between the two plant species and their substrates. Additionally, specific genes and pathways were investigated using a read-level targeted gene finding approach with GraftM to assess the presence of fungal high-affinity phosphate transporters (PHO84) and nitrogenase genes (nifH). Siderophore biosynthetic gene clusters (BGCs) were identified using antiSMASH and clustered using BIG-SCAPE.

The study revealed a high degree of taxonomic novelty in the campos rupestres microbial communities, with a significant portion of ASVs unassigned at the family level. Despite the distinct substrates and resulting differences in microbial community composition between *V. epidendroides* and *B. macrantha*, a shared core microbiome was identified, composed of highly abundant and efficient colonizers. This shared microbiome was significantly enriched in bacterial families like Xanthobacteraceae and Bryobacteraceae, which displayed high abundances of genes involved in phosphorus turnover (transport, mineralization, and solubilization). The study found evidence for a full repertoire of nitrogen cycle-related genes in both microbiomes, including a notable discovery of nitrogen-fixing potential within a lineage of Isosphaeraceae acquired through horizontal gene transfer. Furthermore, there was evidence suggesting that nitrification may involve a metabolic handoff association between Binataceae (involved in ammonia oxidation) and Isosphaeraceae (involved in hydroxylamine oxidation), rather than the traditional two-step process within a single organism. The abundance of genes involved in phosphorus and nitrogen turnover was significantly higher in the rhizosphere compared to the adjacent substrates, indicating active microbial recruitment by the plants. The study also observed that the rock-dwelling *B. macrantha* microbiomes displayed higher abundance of siderophore-producing gene clusters, potentially reflecting a greater demand for iron scavenging in the iron-limited rock substrate. Although Velloziaceae are known to have limited mycorrhizal associations, evidence of fungal PHO84 phosphate transporters in the rhizosphere suggested a potential role for endophytic fungi in phosphorus nutrition. Analysis of nifH genes showed that nitrogen fixation was potentially driven by both Bradyrhizobium and Isosphaeraceae, and this varied based on plant species and substrate conditions. In particular, there was notable diversity in nifH groups I and II, suggesting multiple approaches to nitrogen fixation.

The findings of this study provide compelling evidence that plant-associated microbial communities play a critical role in nutrient cycling and plant growth in the nutrient-depleted campos rupestres ecosystem. The identification of a shared core microbiome enriched in genes involved in phosphorus and nitrogen turnover suggests a selective process by which plants recruit and maintain beneficial microbial partners. The high taxonomic novelty and the discovery of unique metabolic pathways highlight the importance of exploring understudied microbiomes for novel mechanisms of nutrient cycling. The observed differences in community composition between the two plant species, driven by substrate-specific factors (e.g., carbon availability), suggest that environmental factors influence the functional diversity of plant microbiomes. The potential metabolic handoff between Binataceae and Isosphaeraceae for nitrification expands our understanding of nitrogen cycling, suggesting the potential for previously unappreciated microbial interactions in nutrient-poor environments. While the study indicates potential for phosphorus acquisition through several microbial mechanisms and through a potentially significant, yet uncharacterized, fungal component, future experimental work is needed to directly confirm the mechanisms of nutrient uptake by the identified microorganisms and their contribution to plant nutrition.

This study provides significant insights into the functional diversity of plant microbiomes in a nutrient-limited ecosystem. The identification of novel microbial lineages and metabolic processes underscores the importance of exploring understudied biodiversity hotspots for uncovering novel mechanisms of nutrient turnover. The observed enrichment of genes related to nutrient cycling in the rhizosphere suggests a strong influence of plant-microbe interactions on nutrient acquisition. Future research should focus on experimental validation of the proposed metabolic pathways and microbial interactions, alongside exploration of other biodiversity hotspots to reveal more nuanced mechanisms driving plant fitness in nutrient-limited environments. This research also supports a holistic approach in ecosystem modeling that integrates the roles of plants and their microbial communities in nutrient acquisition and cycling.

While this study provides a comprehensive analysis of microbial community composition and function, further research is needed to confirm the functional roles of the identified microorganisms in nutrient cycling. Specifically, experimental validation of the proposed metabolic pathways and microbial interactions is necessary. Additionally, the study focused on two plant species and a limited number of samples, which may not fully represent the biodiversity of the campos rupestres. Future studies should expand the sampling scale to include a greater diversity of plant species and environmental conditions to enhance the generalization of findings.