Skip to main content
Download PDF

Seed banking impacts native Acacia ulicifolia seed microbiome composition and function

Environmental Microbiome volume 20, Article number: 4 (2025) Cite this article

Abstract

Background

Seed banks are a vital resource for preserving plant species diversity globally. However, seedling establishment and survival rates from banked seeds can be poor. Despite a growing appreciation for the role of seed-associated microbiota in supporting seed quality and plant health, our understanding of the effects of conventional seed banking processes on seed microbiomes remains limited. In this study we investigated the composition and functional potential of seed-associated bacterial epiphytes associated with stored and freshly collected seeds of a native plant, Acacia ulicifolia, using both 16S rRNA gene sequencing and culture-based approaches.

Results

Seeds obtained from seed banking facilities were found to host significantly less diverse bacterial populations, with substantial reductions in both low-abundance taxa and in community members commonly identified in freshly collected A. ulicifolia seeds. Bacteria with key plant growth promoting traits including IAA production, ACC deaminase activity, phosphate solubilisation, siderophore activity, and nitrogen fixation were identified in seed epiphytic communities, but these beneficial traits were less prevalent in stored seed compared to fresh seeds.

Conclusion

Overall, these results suggest that epiphytic seed microbiomes may undergo significant changes during the storage process, selecting for bacteria tolerant to storage conditions, and potentially reducing the population of plant-growth promoting bacteria on seeds.

Introduction

Over the past three decades, there has been a global effort to preserve seeds from plant species to prevent extinction and support in-situ conservation initiatives. It is standard for restoration seed suppliers and seed banking facilities, such as the Millenium Seed Bank (MSB) which holds ~ 2.4 billion seeds, to reduce seed-moisture content to < 10% and store in airtight containers at low temperatures (− 18 °C to 15 °C) [1]. When appropriately stored, the viability of seeds from many species can be maintained for many years [2]. However, seeds from such facilities may in some cases suffer poor rates of establishment in the field, hindering the effectiveness of this resource in meeting increasing demands [3,4,5]. These inefficiencies have been attributed to a poor understanding of storage requirements, seed biology, and seed ecology [6, 7] but it is not known whether alterations to seed-associated microorganisms during storage may also be important.

Microbial interactions with plants are known to significantly influence germination success and growth, forming the basis of an essential relationship in seed ecology [8, 9]. Seeds host a diverse array of fungi, bacteria, and archaea, including both endophytes and epiphytes, collectively forming the seed microbiome [10, 11]. These microbes can either be acquired from the surrounding environment or be maternally transmitted [12, 13] and may include taxa with plant growth-promoting capacities [14]. Although the seed microbiome is typically less diverse than other mature plant organs [15], research indicates it serves as an advantageous reservoir of beneficial microbes that may contribute to the nascent seedling microbiome [16,17,18]. Several studies demonstrate that seed-associated microorganisms can mediate seed dormancy and germination, fix and mobilise nutrients from the environment, regulate host hormones, and improve resistance to biotic and abiotic stressors [19,20,21,22,23,24]. In light of this, the seed microbiome is anticipated to be a key determinant of host fitness during critical early stages through to maturity [25].

Contemporary understanding of seed microbiomes is mostly drawn from research in domesticated plant species, and knowledge of native seed microbiota remains limited [26]. As reviewed by Simonin and colleagues [11], seeds are known to host a core microbiome shared among individuals and populations of a given host species [12] which is suggested to be evidence of coevolved microbial symbioses fulfilling important biological and ecological functions in hosts [27]. Commonly reported core genera such as Pantoea, Pseudomonas, Sphingomonas, and Methylobacterium are found in many plants, including native species [11]. Evidence suggests both endophytes and epiphytes may constitute the core seed microbiome [9, 28, 29], however the epiphyte community is more likely to be influenced by local environmental factors and reflect the surrounding microbial diversity [30, 31]. Nevertheless, both fractions of the seed microbiome are functionally significant and may contribute to the developing seedling microbiome [9]. It has also been recognised that native seeds tend to host more diverse microbial communities compared to domesticated species, and host taxa with well-documented plant-growth-promoting capacities [26, 32,33,34].

Despite increasing recognition of the importance of the seed microbiome, the effect of storage practices remains largely unknown [14], particularly for non-agricultural species. Research to date has indicated that seed storage conditions applied in laboratory settings (seeds stored at temperatures between 4 to − 20 °C after drying) can result in the loss of bacterial diversity in seeds, but these findings are primarily derived from agricultural host species using surface sterilised seeds, which removes the epiphytic fraction prior to analysis. Previous work by Chandel and colleagues [35] on soybean [Glycine max (L.) Merr.] endophytes indicate the most significant reduction in diversity occurs during the initial drying stages, with further gradual losses occurring over time in storage. Additionally, temperature was identified to influence the rate of bacterial diversity loss during storage in both soybean and rice [Oryza sativa (L.)] seed endophytes [35, 36]. Overall, seed storage procedures appear to favour the retention of taxa tolerant to desiccation and low temperatures such as Pantoea and Bacillus, although this has not been reported for seed retrieved from seed-banking facilities.

To assess the impact of seed storage on the microbiome of species commonly used in ecological restoration, we compared the bacterial epiphytic seed microbiome from natural and stored populations of Acacia ulicifolia (Salisb.) Court across the Greater Sydney Basin (New South Wales, Australia). Specifically, we compared the seed-associated microbiome from undisturbed natural populations to seven seed samples of A. ulicifolia obtained from two different, independent seed banking organisations to identify differences in bacterial community composition, abundance, and functional potential. Analyses considered both bacterial community composition and load, using a combination 16S rRNA gene community sequencing, flow cytometry, and culture-based investigations. Cultured isolates were also subject to laboratory assays to assess trait profiles associated with plant growth-promoting functions. Given the exposure of seed surfaces to the environment, we hypothesised native seed epiphytic communities to vary between populations, and between stored and freshly collected seeds. Additionally, we anticipated epiphytes associated with stored seed to possess adaptive strategies toward low temperature and low humidity environments.

Materials and methods

Seed collection and processing

Seeds for this study originated from two sources; freshly collected seed pods and samples of stored seeds obtained from two seed banking organisations. Populations of A. ulicifolia were sourced by examining occurrence records for the species in the Atlas of Living Australia (https://www.ala.org.au/) (accessed September 2022). Freshly collected (hereafter fresh) seed pods were sampled from five separate locations (Table S1). From each location, three biological replicates of A. ulicifolia seed pods were collected while wearing gloves to avoid contamination. Each replicate consists of opened pods bearing seeds, collected from 3 to 5 individuals growing within a 10-m radius and replicates were spaced at least 100 m apart. All samples were collected under a Scientific Licence granted by the NSW Government. Stored A. ulicifolia seeds included in this study were sourced from two organisations routinely involved in seed banking and storage; Greening Australia (GA), and the Australian PlantBank (Botanic Gardens of Sydney, Mount Annan, NSW) (PB). Four accessions stored at 15 °C were obtained from GA, and three accessions stored at 4 °C were obtained from PB (original sampling locations and dates provided Table S1). Both groups of samples were stored according to standard methods, in airtight aluminium bags. All accessions were treated as independent samples in all subsequent experiments. Locations of seed sampling sites are indicated in Fig. 1.

Fig. 1
figure 1

Indicative image of A. ulicifolia (A) opened seed pod, (B) unripe unopened fruit and (C) map of locations from which seed samples were obtained in this study. Sample collection group indicated by colour. Yellow: Samples freshly collected from natural populations for this study. Green: sources of samples collected and later stored by Greening Australia. Green/Yellow: location where fresh collection and Greening Australia samples have shared geographic source. Blue: sources of samples collected and later stored by Australian PlantBank

Full size image

Seeds were extracted from freshly collected seed pods using flame-sterilised forceps. For each sample 0.2 ± 0.1 g of seed (approximately 20 seeds) was added to 1 mL of sterile phosphate buffered saline with 0.05% Tween-20 (PBST) in a sterile microfuge tube. Epiphytes were collected by vortexing for 2 min, followed by bath sonication at 40 kHz for 30 min, to detach microorganisms from seed surfaces to allow subsequent culturing and DNA-based investigations [37]. Plant debris was removed by filtering supernatants through a 100 µm cell strainer (Corning, USA). From each epiphyte suspension, a 600 µL aliquot was reserved and immediately stored at − 20 °C for DNA extraction, additionally, 200 µL of each epiphyte suspension was cryopreserved in 25% v/v glycerol and stored at − 80 °C, and the remainder of each sample was stored at 4 °C for culturing and flow cytometric analysis. Bacterial isolation and flow cytometric measurements from all epiphyte washes were performed within 24 h of preparation to ensure results were as representative of populations in planta as possible.

Bacterial isolation

Isolation of seed-associated bacteria

Epiphyte washes were serially diluted in sterile phosphate buffered saline (PBS) and plated onto nutrient agar (NA) plates amended with 2.5 µg mL−1 amphotericin B to suppress fungal growth. Plates were incubated at 28 °C for 4 days followed by 22 °C for 6 days. Colonies were counted, and those determined to be morphologically distinct were picked and subcultured onto fresh NA plates. Pure isolate cultures were grown in nutrient broth (NB), qualitatively screened for a range of putative plant growth-promoting phenotypes, and cryopreserved in 25% glycerol at − 80 °C. Prior to functional characterisation, isolates were grown in NB and washed in sterile PBS by centrifugation 3 times to remove residual media and create isolate suspensions for additional assays. For each assay 3 technical replicates were performed. All glassware used in media preparation was cleaned with 6 M HCl and rinsed thoroughly with ddH2O prior to use.

Screening for selected plant growth-promoting phenotypes

Phosphate solubilisation capacity

Inorganic phosphate mobilisation was assessed by culturing isolates on National Botanical Research Institute's phosphate growth medium (NBRIP) [38]. First, 2 µL of isolate suspension was spotted onto NBRIP agar and incubated at 22 °C. Phosphate solubilisation capacity was determined by the development of a clear halo surrounding colonies after 7 days (Fig. S1).

Nitrogen fixation

Biological nitrogen fixation was evaluated based on growth in nitrogen-free media. Liquid NFb media was prepared as described by Cordova-Rodriguez et al. [39]. Next, 5 µL of each isolate suspension was added to a 96-well plate containing 295 µL NFb and incubated at 28 °C for 7 days. A colour change from green to blue was inferred to indicate nitrogen fixing capacity.

Biosynthesis of auxin (indole-3-acetic acid, IAA)

Auxin biosynthesis was assessed by the colorimetric method described by Khalaf and Riazada [40]. First, IAA production was induced by adding 10 µL of each isolate suspension to 290 µL of Luria-Bretani (LB) broth (Difco Laboratories, USA) supplemented with 5 mM L-tryptophan and incubated at 28 °C for 4 days. Cell-free supernatants were collected by centrifugation at 4800×g for 15 min and transferred to a 96-well plate in triplicate. Equal volumes of Salkowski’s reagent (0.5 M FeCl3 in 35% HClO4) were added to each well and the reaction was incubated in the dark at 22 °C for 30 min. IAA production was measured by spectrophotometric measurement at 530 nm, relative to Escherichia coli DH5α and uninoculated culture controls.

Siderophore production

Siderophore production was estimated by liquid chrome azurol S (CAS) assay [41, 42]. Briefly, iron-limited modified King’s B medium (MKB) was prepared in acid washed glassware as outlined by Gu et al. [41]. Next, the assay solution was prepared by dissolving 121 mg CAS in 100 mL ddH2O and 20 mL of 1 mM FeCl3 solution prepared in 10 mM HCl. This was slowly added to a 20 mL solution of 5 mM hexadecyl trimethyl ammonium bromide (HDTMA) while stirring. Isolates were grown in MKB media for 3 days at 28 °C, and cell-free supernatants were collected by centrifugation. Equal volumes of supernatant and CAS-HDTMA solution were transferred to a 96-well plate and following incubation for 30 min at room temperature, absorbance at 630 nm was measured using a spectrophotometer. Siderophore activity was determined by measuring a decrease in A630 relative to uninoculated controls.

ACC deaminase activity

Isolates were screened for 1-aminocylopropane-1-carboxylic acid (ACC) deaminase activity by the ninhydrin method [43]. First, Dworkin and Foster salts (DF) medium was prepared following the protocol of Penrose and Glick [44]. Next, solutions of DF media were prepared containing either 3 mM ACC or (NH4)2SO4 and inoculated with isolate suspensions in a 96-well plate. Following incubation at 28 °C for 4 days, culture supernatants were collected by centrifugation, and diluted tenfold with fresh DF media. A ninhydrin reagent was prepared by dissolving 500 mg of ninhydrin and 15 mg of ascorbic acid in 60 mL of ethylene glycol and mixed with 60 mL of 1 M citric acid (pH 6.0) immediately prior to analysis. ACC deaminase activity was measured by adding equal volumes of dilute DF-ACC supernatant to the ninhydrin reagent followed by boiling for 30 min and measurement of absorbance at 570 nm, relative to uninoculated DF-ACC controls.

Taxonomic identification of isolates

Isolates with plant growth-promoting traits of interest were selected for taxonomic identification via polymerase chain reaction (PCR) and 16S rRNA gene sequencing using the universal primer set 27F (5′-AGAGTTTGATCMTGGCTCAG-3′) and 1492R (5′-TACGGYTACCTTGTTACGACTT-3′) which amplify close to the full length of the 16S rRNA gene [45]. Amplicons were generated in 25 µL reactions using GoTaq 2X master mix (Promega, USA), 0.5 µM of each primer, and 0.75 µL template DNA. PCR was performed at 98 °C for 2 min, followed by 30 cycles of 98 °C for 30 s, 55 °C for 30 s, and 72 °C for 1.5 min, and a final extension at 72 °C for 10 min. Amplicons were assessed for size and purity by electrophoresis on a 2.0% agarose gel, and once confirmed, purification and Sanger sequencing using 27F primers was carried out by Macrogen (Seoul, RoK). Sequences were trimmed based on basecalling quality using Geneious Prime (version 2023.2.1) and searched against the SILVA (v. 138) 16S rRNA gene database website using the Alignment, Classification and Tree (ACT) tool with a cut-off threshold of 0.97. Sequences were aligned and a phylogenetic tree was built based on maximum likelihood (RAxML).

Flow cytometric analysis of epiphytic cell densities

The microbial cell densities of seed epiphyte washes were estimated by flow cytometry following well-established methods with minor modifications [46]. Samples were stained by incubating a 50 µL aliquot of epiphyte wash with a 1:10,000 dilution of SYBR Green I (SYBR) (Invitrogen) at 22 °C for 1 h in the dark, PBS was passed through a 0.22 µm filter (Sartorius, Germany) and likewise stained to assess background noise. Stained samples were serially diluted 1:10 and 1:100 in 0.22 µm filtered PBS, and four technical replicates were prepared for each dilution. Sample analysis was performed using a Beckman CytoFLEX S flow cytometer (Beckman Coulter, USA) using a 488 nm laser for the analysis of forward scatter (FSC), side scatter (SSC), and SYBR fluorescence signal (525/40). The instrument was run using ddH2O as sheath fluid. Instrument calibration was performed prior to analysis using CytoFLEX Daily QC Fluorospheres (Beckman Coulter, U.S.A.). FSC trigger signal and thresholding were empirically determined by analysing a tube containing a mixture of Fluoresbrite yellow-green (YG) 1.00 µm polystyrene beads (Polysciences, USA) and CaliBRITE FITC 6 µm polymethacrylate beads (BD, USA). Based on the FSC signal intensity for YG beads (1050), the FSC trigger was set to 800 arbitrary units for all further analysis. Samples were acquired at a 10 µL min−1 flow rate for 2 min and gating was performed using the CytExpert (v 2.5) software. Gates were drawn to isolate singlet events within the 1–5 µm size range with SYBR fluorescence intensities greater than 100 au at 525 nm. Counts were imported to R (v 4.2.3) [47] for background correction and statistical analysis.

Bacterial community profiling

DNA sequencing and library construction

DNA was extracted from all seed epiphyte washes using the DNeasy PowerSoil DNA extraction kit (Qiagen, Germany) following the manufacturer’s specifications. DNA quality and yield were assessed on a NanoDrop ND 1000 and a Qubit 3 fluorometer, respectively.

Amplicons of the near full-length 16S rRNA gene were generated by PCR using universal primers 27F and 1492R with unique custom barcodes. Reactions consisted of 25 ng template DNA, 10 µL 5 × GC reaction buffer, 1 µL of 10 mM dNTP mix, 0.5 µL of each 50 µM primer, 1.5 µL DMSO, 0.25 µL Phusion polymerase (New England Biolabs, USA), and 30 µL NF-ddH2O (Invitrogen). PCR was performed at 98 °C for 2 min, followed by 35 cycles at 98 °C for 30 s, 60 °C for 30 s, and 72 °C for 1.5 min, and a final extension at 72 °C for 10 min. Reactions were prepared in triplicate and pooled. All samples were successfully amplified except for Kurrajong 1988; despite optimisation efforts, no product could be obtained for this sample due to low DNA yield, and so was not carried onto sequencing.

Barcoded amplicons were purified using AMPure XP beads (Beckman Coulter, USA), resuspended in 25 µL NF-ddH2O, and quantified using a Qubit fluorometer. An equimolar multiplexed 16S rRNA gene amplicon library was then generated using custom barcoded primers (Table S2) and the Oxford Nanopore Technologies (ONT) Ligation Sequencing Kit (SQK-LSK110) as per the manufacturer’s specifications. Sequencing was performed using a R9.4.1 flow cell and MinION MK1B device. Basecalling was performed with Guppy (v 6.3.2) using the high-accuracy basecalling model with default parameters resulting in only sequences > Q7 and most sequences > Q10. Raw reads were exported for each barcode in FASTQ format.

Bioinformatic analysis of community sequencing data

Demultiplexed raw reads for 16S rRNA amplicons were processed using the Spaghetti pipeline [48]. This process involved adapter removal by porechop, read filtering to retain reads with a size range of 1.2–1.8 kb using Nanofilt, quality control with Nanostat, chimera detection and removal using ycard, and read mapping against the SILVA (v. 138) database by mimimap2, to the species level. One sample from the STR sample group (STR1) was removed from the dataset due to the low number of reads recovered. Processed sequencing data were imported to R (v. 4. 2. 3) as taxonomy and abundance tables, and associated metadata. Host-associated mitochondria and chloroplast reads, and low abundance features (< 10 counts) were removed from the dataset, and samples were rarefied to the lowest number of read counts (31,140 reads) without replacement using the phyloseq R package (v 1.44.0) [49] (rarefaction curves shown in Fig. S2). Next, operational taxonomic units (OTUs) were agglomerated to the species level to remove artificial heterogeneity which can be introduced into Nanopore sequencing data, as per Latorre-Pérez et al. [48]. Data was exported to the microeco (v. 1.2.2) [50] package for all subsequent analyses using the file2meco package (v 0.5.0) [50].

Taxonomy-based functional prediction was performed using the Tax4Fun2 [51] function from the microeco package and BLAST command line package (v 2.13.0) (Camacho [52]. KEGG ortholog abundances obtained from Tax4Fun2 prediction were annotated using the PLaBAse plant growth promoting trait ontology (PGPTO) [53].

Statistical analysis

All statistical tests and data visualisations were performed in R using the microeco and ggplot2 packages [54]. Differences in Shannon diversity, Chao1 richness, and Inverse-Simpson (InvSimpson) evenness were tested for significance by the Kruskal–Wallis pairwise test. Beta diversity based on a Bray–Curtis dissimilarity matrix was assessed by principal coordinate analysis (PCoA), and differences in group distance was tested using the Wilcoxon rank-sum test and by analysis of similarity (ANOSIM) tests. Functional differential abundance was assessed by a linear discriminant analysis effect size (LEfSe) algorithm [55] as implemented in the microeco package [50]. False discovery rate (FDR) p-value correction was applied to all tests in this study.

Results

Bacterial community profiling

Comparison of seed epiphyte microbiomes of fresh and stored seed

A total of 2443 OTUs were identified in the dataset following sequence quality checking and filtering. All OTUs which were assigned a taxonomy were bacterial, with no archaea identified. Of these, the majority of OTUs were found to be unique to fresh seeds (42.5%), and few OTUs were identified to be unique to a specific location (3.1%). A set of highly abundant shared OTUs were observed that were present across all fresh samples, these taxa while only representing ~ 20% of recovered OTUs were important contributors in terms of overall relative abundance, with 71.8% of quality filtered reads belonging to these bacteria (Fig. 2). A large number of OTUs found in fresh samples were identified in all sampling locations (1014, 46%) (Fig. S3).

Fig. 2
figure 2

Acacia ulicifolia seed-associated epiphytic community alpha and beta diversity and number of shared OTUs observed in fresh and stored seed from different banking organisations. A Diversity of microbial communities was assessed based on Shannon diversity indices, Chao1 species richness, and Inverse Simpson evenness for fresh and stored seed (observed richness shown in Figure S5). B Principal coordinate analysis indicating differences in overall seed epiphyte bacterial community composition between fresh and stored seed (colour) and sampling location (points represented by different symbols). C Venn diagram indicating the number and proportion of unique and shared OTUs between the seed microbiomes of samples from different seed

Full size image

The epiphyte bacterial communities on A. ulicifolia seeds from freshly collected and stored seed samples were compared, looking at both bacterial diversity and community composition. Stored samples were found to be significantly lower in bacterial diversity (p < 0.001), species richness, and evenness (p < 0.001) than fresh samples (Fig. 2A). Provenance factors influencing seed microbiome composition were examined using Principal Coordinate Analysis (PCoA), which also showed that stored and fresh seed bacterial communities were distinct, clustering separately (Fig. 2B). A Wilcoxon rank sum test indicated significant differences in Bray–Curtis distance between stored and fresh samples (p < 0.001) (Figure S4). Similarly, ANOSIM results showed significant community dissimilarity between stored and fresh samples (R = 0.8383, p < 0.001). Sampling location also contributed to variation in community composition among fresh samples, but to a lesser extent (R = 0.2939. p = 0.0170). In contrast, within freshly collected samples, alpha diversity (Shannon) and species richness (Chao1) were comparable across all locations surveyed (p > 0.05) and no significant difference in evenness was identified (InvSimpson, p = 0.98).

Epiphytic composition of A. ulicifolia seed microbiomes

The 2443 OTUs identified in this study span 34 phyla, 257 families, and 594 genera. OTUs affiliated with the phyla Proteobacteria, Acidobacteria, and Actinobacteriota were the most prominent colonisers of fresh seeds, whereas GA and PB seeds were principally colonised by Firmicutes and Proteobacteria respectively (Fig. 3A). For fresh seeds there was considerable overlap in the families observed to contribute most to relative abundance across the different sampling locations (Fig. 3B). GA-sourced seeds also tended to host similar epiphytic taxa at the family level, however this differed from the observed family-level composition of both fresh seed and PB-sourced seed. At lower taxonomic ranks, a set of bacterial OTUs with high relative abundance was observed across freshly collected seed samples from different sampled locations (Fig. 3C). These OTUs were identified in every fresh sample and together accounted for approximately 50% of the total relative abundance per sample. These included taxa that are currently not well characterised below genus level, including an undescribed Rhizobiales genus within the Beijerinckiaceae family (1174-901-12), and representatives of Sphingomonas, Methylobacterium-Methylorubrum (hereafter Methylobacterium), Allorhizobium-Neorhizobium-Pararhizobium-Rhizobium (hereafter Rhizobium), Acidiphilium, and Terriglobus. The remainder of OTUs in fresh seed microbiota consisted of a diverse assortment of genera each of which had low individual abundance (< 5%). Interestingly, a sizeable proportion of these included OTUs assigned to genera for which there are currently no cultured representatives in the Silva v. 138 database (Fig. 3C). In contrast, microbiomes of stored seeds contained far fewer low-abundance taxa. Instead, the microbiome of GA seeds largely consisted of Bacillus and Paenibacillus, whereas PB seeds were predominantly comprised of Erwinia, Pantoea, and Pseudomonas. Additionally, Massilia and Curtobacterium were abundant genera common to both sources of stored seed.

Fig. 3
figure 3

Composition of epiphytic bacterial communities from A. ulicifolia seeds at different taxonomic levels. A Relative abundance of the most predominant bacterial taxa at phylum and B family taxonomic levels for each sample location and stored seed lot, with less abundant taxa combined and entitled ‘others’. C Heatmap and dendrogram of the 75 most abundant genera in A. ulicifolia seed-associated bacterial communities showing clustering of stored (red) and fresh (aqua) samples

Full size image

Functional profiling of culturable epiphytic bacteria from fresh and stored A. ulicifolia seeds

A total of 243 morphologically distinct bacteria were isolated from A. ulicifolia seed surfaces (224 from fresh samples and 19 from stored samples). Screening for three plant growth-promoting (PGP) phenotypes: nitrogen fixing, phosphate mobilisation and IAA production, indicated most isolates possessed at least of one these traits (72.8% of fresh samples 36.7% of stored samples). The main bacterial morphotypes from this collection (70 from fresh seed accessions, 10 from GA seed and only 1 isolate from PB seed, Bacillus (2E04)) were subject to further analysis to determine taxonomy and extend the PGP trait profiling. The taxonomic profiling, performed via 16S rRNA gene sequencing, showed that this isolate collection included representatives of 28 different genera, with Curtobacterium, Pseudomonas, Microbacterium, Luteibacter, Xanthomonas, and Methylobacterium among the most prevalent (Fig. 4). Of these 81 bacterial isolates, 40.7% possessed ACC deaminase activity, 55.6% produced siderophores, 48.1% produced IAA, 14.8% solubilised inorganic phosphate, while 60.4% displayed biological nitrogen fixation capacity (Table S3). Most isolates (87.7%) possessed at least two traits (Fig. 4). Notably, one isolate recovered from fresh seed, a representative of the genus Leclercia, possessed all traits screened.

Fig. 4
figure 4

Phylogenetic analysis of 16S rRNA gene sequences from bacterial isolates recovered from A. ulicifolia seeds. Phylogenetic tree inferred from maximum likelihood (RAxML) analysis and taxonomic assignments based on classification in the SILVA (v. 138) database using the Alignment, Classification and Tree (ACT) tool. Putative plant growth promoting functions for each isolate determined in vitro indicated by green circles for each trait tested (BNF = biological nitrogen fixation, Phos. = phosphate solubilisation, IAA = Indole-3-acetic acid production, Sid. = Siderophore production, ACC = 1-aminocyclopropane carboxylic acid deaminase activity). Isolates recovered from stored and fresh seeds are indicated by red and blue circles respectively

Full size image

Predicted functional capacity of the bacterial seed epiphytic community

Microbiome functional predictions were performed using 16S rRNA gene community sequencing data and Tax4Fun2 function in the microeco package to generate trait profiles. Predicted functional profiles of stored and fresh seed were compared to identify potential differences, focusing on traits of relevance to microbe-plant interactions, using the PLaBAse PGPTO database to identify pathways relevant to plant hosts. A set of 40 functions were found to be significantly enriched/depleted between stored and fresh seeds (p < 0.05) based on LEfSe differential abundance analysis. Fresh seed microbiota were found to be enriched for functions associated with nitrogen fixation, phosphate enrichment, bacterial antagonism, plant resource utilisation, and resistance to mercury and arsenic, compared to stored seeds. In contrast, the functional profile of stored seeds was enriched with traits associated with fungal antagonism, potassium enrichment, thiamin biosynthesis, and the ter operon (a set of genes associated with tellurite resistance) (Fig. 5). No significant difference in trait profiles was identified between different fresh sampling locations, or different seed storage facilities (p > 0.05) following FDR adjustment.

Fig. 5
figure 5

Predicted set of differentially-abundant plant growth-promoting functions inferred using Tax4fun2 and applying linear discriminant analysis (LDA). The Functional classification are taken from the PLaBAse plant growth promoting trait ontology [53]

Full size image

Estimates of seed-associated epiphytic cell densities

To determine if seed surfaces from different samples hosted bacterial communities of different density, flow cytometric counts of DNA-stained cells 1.0–6.0 µm in size (the size range of most culturable bacterial cells) were calculated per gram of seed and log transformed (Fig. 6A). For all freshly collected samples, as well as GA-stored seed, a sizable cell population was observed in the seed wash (> 107 cells per gram washed seed). However, stored seed from PB samples had significantly (p < 0.001) lower cell counts than all other sample groups, with considerable variation between counts for specific PB samples. Flow cytometric estimates of bacterial populations within each seed wash were also observed to be fairly consistent for samples from most locations, with the exception of PB samples.

Fig. 6
figure 6

Microbial cell counts of A. ulicifolia seed epiphytes determined by (A) flow cytometry and (B) colony forming unit counts. Counts are reported as mean (n = 12) values for each fresh seed sampling location (aqua) or stored sample group (red). Significant differences, as determined by Kruskal–Wallis tests, are denoted unshared letters

Full size image

The number of culturable bacterial colonies recovered from each seed surface (expressed as CFU per gram of seed) was also counted following 10 days of incubation, as another measure of bacterial density (Fig. 6B). Bacterial colonies were recovered from all samples except for the stored sample PB1988 where no growth was observed following the incubation period. When compared to flow cytometric counts, CFU estimates displayed considerable variation within and between sample groups, including a notable outlier within the GA sample set (from HXT2005) containing similar CFU counts to fresh seed. Counts of CFU were also consistently several-fold lower than flow cytometric cell counts, likely reflecting the high proportion of environmental bacteria that are unculturable under standard laboratory conditions. Culture analysis reflected flow cytometry results in that the number of bacterial colonies recovered from seeds tended to be lower for stored than for freshly collected samples, with PB samples in particular having very few bacterial colonies recovered. Pairwise analysis was performed and showed there was a significant difference in average CFU per gram of seed between some sample groups (p < 0.001, Fig. 6B), with PB samples significantly lower than any of the fresh samples. Overall, estimates of cell densities from both methods indicate that stored seeds generally contained fewer bacterial epiphytes, and that significantly fewer bacteria were observed on PB seeds compared to GA.

Discussion

High quality seed from native species is an essential resource for restoration and conservation of plant diversity the world over. Seed-associated bacteria can play a demonstrated role in ensuring seed quality in relation to germinability and viability [14, 21, 26] but seed storage under standard seed bank conditions may affect the load and composition of seed microbiomes, impacting quality. We show that the epiphytic microbiome of seedbank-stored native A. ulicifolia seeds hosted less diverse communities of bacteria, coinciding with a significantly reduced total bacterial population relative to fresh, field-collected seeds. Bacterial community composition also differed substantially between stored and field-collected seeds, with many taxa abundant in fresh seeds, including members with plant-beneficial functional traits, absent in stored seeds. Our results have implications for the successful downstream use of seeds in nature repair programs, particularly where altered or reduced bacterial composition affects germination and early-stage growth.

Understanding native seed microbiomes

While much of the research on seed microbiomes has focussed on agricultural species, those of native seeds are beginning to receive research attention. Cross-taxon studies have identified bacterial groups common across seed microbiomes and have shown that native seeds tend to host more diverse microbiomes compared to seeds from agricultural plant species [32, 34]. Many of the genera observed to be abundant in seed microbiome of our study species are common to a diverse suite of plant species. For instance, reviews of seed microbiome research emphasise Pantoea, Sphingomonas, Methylobacterium, Bacillus, Curtobacterium and Pseudomonas as prominent seed endophytes and epiphytes in both native and domesticated plant lineages [10, 11, 32].

We show that Sphingomonas, Methylobacterium, Rhizobium, Curtobacterium, and 1174-901-12 (a member of Rhizobiales belonging to an undescribed genus) were consistently key contributors in terms of relative abundance in fresh seed, with Pseudomonas also important in seeds from specific sites based on 16S rRNA sequencing. Culture-based work also regularly recovered isolates of Microbacterium, Luteibacter, Staphylococcus and Cellulomonas, but did not recover all highly abundant microbiome components, such as 1174-901-12. Interestingly A. ulicifolia seeds hosted significant populations of subdivision 1 Acidobacteria including Terriglobus, Granulicella, and Bryocella. Their presence on seeds is unusual, as these genera, in addition to Acidiphilium, typically reside in mineral or soil environments and are only reported to colonise aboveground plant surfaces in a small number of studies [56, 57]. Previous research has noted the abundance of these taxa in phyllosphere microbiomes correlates with seasonal changes in precipitation [58] with suggestions they may be deposited on aerial plant surfaces by rain without necessarily colonising the host [59]. Whilst other prominent members of the A. ulicifolia seed microbiome including 1174-901-12, Sphingomonas, Pseudomonas, and Methylobacterium have also been reported to be transmitted by rain, these genera are prominent colonisers of mature aerial and internal compartments in diverse plant species [60,61,62,63,64].

Overall, our results suggest that some fraction of the A. ulicifolia seed microbiome is likely dynamic, with fresh seed microbiomes comprised of transient or locally adapted bacterial taxa in addition to specifically adapted A. ulicifolia seed microbiota. Undoubtedly, repeated analysis of samples of seed at the same locations through time would contribute to a clearer picture of seed epiphytic microbiome dynamics. Such work is important given knowledge of epiphytic microbial communities on seed remains limited. Their composition may shift considerably through time and is determined by the local environment, contrasting with endophytes which are thought to establish long-term symbioses with hosts [65]. Significant variation in epiphyte communities along spatial and temporal gradients has been identified in seed of agricultural Brassica and Triticum cultivars, suggesting that the epiphytic microbiome is determined almost exclusively by environmental factors [31, 66, 67]. However, other studies report consistent populations of shared microbiota within the same hosts between distant populations [28, 29].

Differences in stored and freshly collected seed microbiomes

Although seed processing, drying, and low-temperature storage remain necessary to preserve seed viability, our results raise the possibility that important microbial symbionts may be lost in stored seed. We identified substantial reductions in bacterial load and community diversity associated with stored seeds, driven primarily by the absence of rare and low-abundance taxa compared to freshly collected counterparts. Notably, this was found both across multiple stored seed accessions from several locations, and seed collected from a single site over successive harvest years. These results are consistent with previous controlled laboratory studies in stored soybeans with larger sample sizes [35], suggesting similar selection processes are occurring on bacteria in native seed banking facilities.

Although studies examining the effects of postharvest seed storage on the seed microbiome remain scarce, available evidence suggests that bacteria with specific adaptive strategies to desiccation and temperature stress are selected for during the storage process. The present study and previous literature examining stored seed microbiota indicate stored seed microbiomes may be enriched in Bacilli and Gammaproteobacteria groups possessing traits such as sporulation, peptidoglycan cell walls, or secretion of extracellular polysaccharides, known to improve tolerance to temperature and desiccation stresses [68]. This most commonly includes strains of Pantoea, Erwinia, and Bacillus, reported as abundant members of both endophytic and epiphytic stored seed communities [35, 36], likely a consequence of reduced moisture content also occurring within stored seed tissues.

Certain members of the epiphytic community common in fresh A. ulicifolia seed, such as 1174-901-12, Sphingomonas, and Methylobacterium, were found to persist in stored seed samples, albeit in reduced relative abundance. These genera have been reported to together form pioneer biofilms on a range of surfaces susceptible to desiccation stress such as photovoltaic panels and roof tiles [69, 70], which possibly also contributes some degree of protection during seed storage. However, it should be acknowledged that, as few isolates of Sphingomonas or Methylobacterium were recovered from stored seed, it is possible 16S rRNA gene-based community profiling reflects dormant, non-viable or otherwise unculturable bacteria.

Differences in bacterial load and composition were also observed between each of the two storage facilities from which samples were sourced, suggesting that differences in handling practices may also shape stored seed microbiomes. It is worth noting that there were differences in temperature preferences of the dominant taxa from different locations (psychrotolerant in PlantBank and mesophilic in Greening Australia) consistent with 4 °C storage vs room temperature storage respectively (Fig. 3B, C). For example, the most abundant genera in GA seeds such as Paenibacillus and Bacillus generally consist of mesophilic strains, whereas dominant PB genera including Pantoea, Erwinia, and Pseudomonas include taxa capable of psychrotolerant growth.

Although past work has shown temperature can influence the composition of stored seed microbiota [35], it is likely that multiple environmental parameters, including temperature, desiccation, and variations in seed handling together with storage time all contribute to observed differences between organizations. To dissect the contributions of each of these factors to stored seed microbiomes, manipulative experiments examining each independently are required. However, the limited availability of seed accessions that can be obtained from seedbanks and matched to intact populations of wild plants remains a significant challenge to carrying out such studies. While we cannot comment on the microbiome of seeds before storage in this study, the work presented here supports previous assertions that storage may result in substantial reductions in diversity in the seed microbiome.

Ecological significance of changes in seed microbiota

In this study, we isolated numerous bacteria from freshly collected A. ulicifolia seed surfaces possessing well-established plant growth-promoting (PGP) phenotypes. This included evidence of ACC deaminase activity and IAA production, traits that have been shown to improve tolerance to drought and salinity stress across a wide range of plant species [22, 71]. Many of our isolated strains also show capacity for N-fixation, siderophore production and phosphate solubilisation, mechanisms which have been demonstrated to overcome nutrient deficiency [72, 73], pathogenic invasion [23], and more generally enhance host productivity [20, 21, 65]. Bacterial isolates possessing these traits, such as those identified in this study, may also synergistically interact leading to emergent properties with improved plant growth-promoting capacity [74], although we did not examine this. While stored seeds were not entirely lacking in isolates with PGP traits, only a small number of such isolates were recovered from stored seed, and few of these showed multiple beneficial traits. Stored seeds were also found to host bacterial communities possessing fewer traits relating to the utilisation of plant resources compared to fresh seed, functions known to influence colonisation success during seedling emergence [75]. As microbes on seed surfaces are important in shaping seedling microbiomes [18, 76], the reduced diversity of PGP bacteria associated with stored seeds may reflect microbiome changes which have the potential to influence the health and survivorship of seedlings produced from stored seed. However, the true impact of storage-induced changes to the microbiome cannot be adequately assessed without further studies in planta.

Although our understanding of conserved symbionts in A. ulicifolia is limited, the presence of set of shared genera which showed high relative abundance across all or most samples implies these taxa may play an important ecological role in host functioning. Additionally, research indicates that changes in low-abundance taxa may in some cases also lead to substantial changes in plant growth-promoting activity in microbial communities [73, 77, 78]. Most significantly, Rhizobia and related members of Beijerinckiaceae comprise taxa shown to stimulate nodulation as well as engage in associative symbiotic N-fixation with host plants [72]. In many species of Acacia they are known to be essential to healthy functioning [79, 80], and so their absence in stored seeds is concerning.

It is important to continue investigations on whether seed banking processes reduce seed-associated bacterial diversity, as this could limit the availability of beneficial microbes to developing seedlings. This is particularly relevant for restoration practitioners using stored seed in degraded ecosystems, where soils are often nutrient-poor and lack microbial diversity [81]. The loss of bacterial diversity during storage may reduce PGP functionality in the seed microbiome [73, 75, 78] potentially leading to poor seedling vigour in microbially depleted soils [21] and affecting restoration outcomes.

Our findings support calls to consider the seed-associated microbiome during processing and storage [14, 26]. One potential strategy for conserving seed microbiota is to collect and catalogue native seed-associated bacteria before storage. Native species are underrepresented in seed microbiome studies [11], and expanding our understanding of their structure and function could help identify important microbial taxa [26].This knowledge could guide the development of microbial inoculants or specialised preservation methods ensuring essential microbes are retained in stored seeds.

Studying microbiome dynamics in native seeds presents specific challenges, particularly the difficulty obtaining sufficient seed for research. Native seeds typically have short collection windows, with variable yields that can be harvested from fragmented populations [82]. The design of this study was constrained by seed availability, with limited seed available from seed banking organisations for research, due to need for such organisations to fulfil restoration and other goals. Despite these constraints, our approach offered the advantage of examining microbiome changes following relatively long term seed storage (10–20 + years), which would likely be unfeasible in controlled laboratory experiments. However, further research is needed, including sampling from geographically diverse wild populations and seed banks, as well as well-replicated laboratory aging experiments. Such studies could also explore how storage-related changes in seed microbiomes impact plant fitness.

Conclusion

Synthesising data from 16S rRNA gene-based community characterisation, flow cytometric cell counts, culture-based screening and phenotypic assays we have characterised the structure, function and load of the epiphytic seed microbiome of A. ulicifolia sourced fresh from native populations and from seed banking organizations. We demonstrate the epiphytic microbiome of freshly collected A. ulicifolia seeds consist of a diverse range of bacterial taxa including many strains with plant growth-promoting capacity. While the full significance of the seed microbiome to native species’ health and success in restoration work requires further investigation, results presented in this study suggest that some of these potentially beneficial microbes may be lost during storage. We demonstrate that the surfaces of A. ulicifolia seeds obtained from seedbanks support fewer and less diverse bacterial populations when compared to natural seed populations. Although our study was limited by the few stored seed accessions available for analysis, changes to the microbiome composition of stored seeds observed in this work reflect what has been reported in the small number of manipulative experiments published previously on seed storage and microbial composition for other plant species. Further research is required to determine the impact of these changes of the microbiome in planta, and the flow-on effects to restoration success.

Availability of data and materials

The sequencing data from this study have been deposited under NCBI BioProject ID PRJNA1134362.

Abbreviations

MSB:

Millenium Seed Bank

STR:

Strickland State Forest

KRG:

Ku-ring-gai National Park

AB:

Agnes Banks Nature Reserve

GBK:

Glenbrook

BB:

Burnum Burnum Sanctuary

GA:

Greening Australia

PB:

Australian PlantBank

PBST:

Phosphate buffered saline 0.05% tween 20

PBS:

Phosphate buffered saline

NA:

Nutrient agar

NB:

Nutrient broth

NBRIP:

National Botanical Research Institute’s phosphate growth medium

IAA:

Indole-3-acetic acid

LB:

Luria-Bretani

CAS:

Chrome azurol S

MKB:

Modified King’s B medium

HDTMA:

Hexadecyl trimethyl ammonium bromide

ACC:

1-Aminocylopropane-1-carboxylic acid

DF:

Dworkin and foster salts

PCR:

Polymerase chain reaction

SYBR:

SYBR green I

FSC:

Forward scatter

SSC:

Side scatter

OTU:

Operational taxonomic unit

PGPTO:

Plant growth promoting trait ontology

PGP:

Plant growth promoting trait

PCoA:

Principal coordinate analysis

ANOSIM:

Analysis of similarity

LEfSe:

Linear discriminant analysis effect size

FDR:

False discovery rate

CFU:

Colony forming unit

References

  1. Frischie S, Miller AL, Pedrini S, Kildisheva OA. Ensuring seed quality in ecological restoration: native seed cleaning and testing. Restor Ecol. 2020;28(S3):S239–48.

    Article  Google Scholar 

  2. De Vitis M, Hay FR, Dickie JB, Trivedi C, Choi J, Fiegener R. Seed storage: maintaining seed viability and vigor for restoration use. Restor Ecol. 2020;28(S3):S249–55.

    Article  Google Scholar 

  3. Merritt DJ, Dixon KW. Conservation: restoration seed banks–a matter of scale. Science. 2011;332(6028):424–5.

    Article  PubMed  CAS  Google Scholar 

  4. Ceccon E, González EJ, Martorell C. Is direct seeding a biologically viable strategy for restoring forest ecosystems? Evidences from a meta-analysis. Land Degrad Dev. 2016;27(3):511–20.

    Article  Google Scholar 

  5. Chapman T, Miles S, Trivedi C. Capturing, protecting and restoring plant diversity in the UK: RBG Kew and the Millennium Seed Bank. Plant Divers. 2019;41(2):124–31.

    Article  PubMed  Google Scholar 

  6. Streczynski R, Clark H, Whelehan LM, Ang S-T, Hardstaff LK, Funnekotter B, Bunn E, Offord CA, Sommerville KD, Mancera RL. Current issues in plant cryopreservation and importance for ex situ conservation of threatened Australian native species. Aust J Bot. 2019;67(1):1–15.

    Article  Google Scholar 

  7. Turner SR, Cross AT, Just M, Newton V, Pedrini S, Tomlinson S, Dixon K. Restoration seedbanks for mined land restoration. Restor Ecol. 2022;30(S1):e13667.

    Article  Google Scholar 

  8. Vandenkoornhuyse P, Quaiser A, Duhamel M, Le Van A, Dufresne A. The importance of the microbiome of the plant holobiont. New Phytol. 2015;206(4):1196–206.

    Article  PubMed  Google Scholar 

  9. Nelson EB. The seed microbiome: origins, interactions, and impacts. Plant Soil. 2018;422(1):7–34.

    Article  CAS  Google Scholar 

  10. Truyens S, Weyens N, Cuypers A, Vangronsveld J. Bacterial seed endophytes: genera, vertical transmission and interaction with plants. Environ Microbiol Rep. 2015;7(1):40–50.

    Article  Google Scholar 

  11. Simonin M, Briand M, Chesneau G, Rochefort A, Marais C, Sarniguet A, Barret M. Seed microbiota revealed by a large-scale meta-analysis including 50 plant species. New Phytol. 2022;234(4):1448–63.

    Article  PubMed  Google Scholar 

  12. Shade A, Jacques MA, Barret M. Ecological patterns of seed microbiome diversity, transmission, and assembly. Curr Opin Microbiol. 2017;37:15–22.

    Article  PubMed  Google Scholar 

  13. Abdelfattah A, Tack AJM, Lobato C, Wassermann B, Berg G. From seed to seed: the role of microbial inheritance in the assembly of the plant microbiome. Trends Microbiol. 2023;31(4):346–55.

    Article  PubMed  CAS  Google Scholar 

  14. Berg G, Raaijmakers JM. Saving seed microbiomes. Isme J. 2018;12(5):1167–70.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  15. War AF, Bashir I, Reshi ZA, Kardol P, Rashid I. Insights into the seed microbiome and its ecological significance in plant life. Microbiol Res. 2023;269:127318.

    Article  PubMed  CAS  Google Scholar 

  16. Barret M, Briand M, Bonneau S, Préveaux A, Valière S, Bouchez O, Hunault G, Simoneau P, Jacquesa MA. Emergence shapes the structure of the seed microbiota. Appl Environ Microbiol. 2015;81(4):1257–66.

    Article  PubMed  PubMed Central  Google Scholar 

  17. Walsh CM, Becker-Uncapher I, Carlson M, Fierer N. Variable influences of soil and seed-associated bacterial communities on the assembly of seedling microbiomes. Isme J. 2021;15(9):2748–62.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  18. Arnault G, Marais C, Préveaux A, Briand M, Poisson AS, Sarniguet A, Barret M, Simonin M. Seedling microbiota engineering using bacterial synthetic community inoculation on seeds. FEMS Microbiol Ecol. 2024. https://doiorg.publicaciones.saludcastillayleon.es/10.1093/femsec/fiae027.

    Article  PubMed  PubMed Central  Google Scholar 

  19. Goggin DE, Emery RJ, Kurepin LV, Powles SB. A potential role for endogenous microflora in dormancy release, cytokinin metabolism and the response to fluridone in Lolium rigidum seeds. Ann Bot. 2015;115(2):293–301.

    Article  PubMed  CAS  Google Scholar 

  20. Verma SK, Kharwar RN, White JF. The role of seed-vectored endophytes in seedling development and establishment. Symbiosis. 2019;78(2):107–13.

    Article  Google Scholar 

  21. Rodríguez CE, Antonielli L, Mitter B, Trognitz F, Sessitsch A. Heritability and Functional Importance of the Setaria viridis Bacterial Seed Microbiome. Phytobiomes J. 2019;4(1):40–52.

    Article  Google Scholar 

  22. Hone H, Mann R, Yang G, Kaur J, Tannenbaum I, Li T, Spangenberg G, Sawbridge T. Profiling, isolation and characterisation of beneficial microbes from the seed microbiomes of drought tolerant wheat. Sci Rep. 2021;11(1):11916.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  23. Matsumoto H, Fan X, Wang Y, Kusstatscher P, Duan J, Wu S, Chen S, Qiao K, Wang Y, Ma B, et al. Bacterial seed endophyte shapes disease resistance in rice. Nat Plants. 2021;7(1):60–72.

    Article  PubMed  CAS  Google Scholar 

  24. Chimwamurombe PM, Grönemeyer JL, Reinhold-Hurek B. Isolation and characterization of culturable seed-associated bacterial endophytes from gnotobiotically grown Marama bean seedlings. FEMS Microbiol Ecol. 2016;92(6):fiw083.

    Article  PubMed  Google Scholar 

  25. Chesneau G, Torres-Cortes G, Briand M, Darrasse A, Preveaux A, Marais C, Jacques MA, Shade A, Barret M: Temporal dynamics of bacterial communities during seed development and maturation. FEMS Microbiol Ecol. 2020; 96(12).

  26. Mertin AA, Philpott M, Blackall LL, French K, Liew ECY, van der Merwe MM. Integrating seed microbiome knowledge into restoration and ex situ conservation of native Australian plants. Aust J Bot. 2023;71(7):379–94.

    Article  Google Scholar 

  27. Wassermann B, Abdelfattah A, Wicaksono WA, Kusstatscher P, Müller H, Cernava T, Goertz S, Rietz S, Abbadi A, Berg G. The Brassica napus seed microbiota is cultivar-specific and transmitted via paternal breeding lines. Microb Biotechnol. 2022;15(9):2379–90.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  28. Links MG, Demeke T, Gräfenhan T, Hill JE, Hemmingsen SM, Dumonceaux TJ. Simultaneous profiling of seed-associated bacteria and fungi reveals antagonistic interactions between microorganisms within a shared epiphytic microbiome on Triticum and Brassica seeds. New Phytol. 2014;202(2):542–53.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  29. Morales Moreira ZP, Helgason BL, Germida JJ. Crop, genotype, and field environmental conditions shape bacterial and fungal seed epiphytic microbiomes. Can J Microbiol. 2021;67(2):161–73.

    Article  PubMed  CAS  Google Scholar 

  30. Rochefort A, Simonin M, Marais C, Guillerm-Erckelboudt AY, Barret M, Sarniguet A. Transmission of seed and soil microbiota to seedling. mSystems. 2021;6(3):e0044621.

    Article  PubMed  Google Scholar 

  31. Klaedtke S, Jacques MA, Raggi L, Préveaux A, Bonneau S, Negri V, Chable V, Barret M. Terroir is a key driver of seed-associated microbial assemblages. Environ Microbiol. 2016;18(6):1792–804.

    Article  PubMed  CAS  Google Scholar 

  32. Wassermann B, Cernava T, Müller H, Berg C, Berg G. Seeds of native alpine plants host unique microbial communities embedded in cross-kingdom networks. Microbiome. 2019;7(1):108.

    Article  PubMed  PubMed Central  Google Scholar 

  33. Kim H, Lee KK, Jeon J, Harris WA, Lee YH. Domestication of Oryza species eco-evolutionarily shapes bacterial and fungal communities in rice seed. Microbiome. 2020;8(1):20.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  34. Chandel A, Mann R, Kaur J, Tannenbaum I, Norton S, Edwards J, Spangenberg G, Sawbridge T. Australian native Glycine clandestina seed microbiota hosts a more diverse bacterial community than the domesticated soybean Glycine max. Environ Microbiome. 2022;17(1):56.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  35. Chandel A, Mann R, Kaur J, Norton S, Edwards J, Spangenberg G, Sawbridge T. Implications of seed vault storage strategies for conservation of seed bacterial microbiomes. Front Microbiol. 2021;12:784796.

    Article  PubMed  PubMed Central  Google Scholar 

  36. Dutta S, Choi SY, Lee YH. Temporal dynamics of endogenous bacterial composition in rice seeds during maturation and storage, and spatial dynamics of the bacteria during seedling growth. Front Microbiol. 2022;13:877781.

    Article  PubMed  PubMed Central  Google Scholar 

  37. White LJ, Brözel VS, Subramanian S. Isolation of rhizosphere bacterial communities from soil. Bio Protoc. 2015;5(16):e1569.

    Article  Google Scholar 

  38. Nautiyal CS. An efficient microbiological growth medium for screening phosphate solubilizing microorganisms. FEMS Microbiol Lett. 1999;170(1):265–70.

    Article  PubMed  CAS  Google Scholar 

  39. Cordova-Rodriguez A, Rentería-Martínez ME, López-Miranda CA, Guzmán-Ortíz JM, Moreno-Salazar SF. Simple and sensitive spectrophotometric method for estimating the nitrogen-fixing capacity of bacterial cultures. MethodsX. 2022;9:101917.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  40. Khalaf EM, Raizada MN. Taxonomic and functional diversity of cultured seed associated microbes of the cucurbit family. BMC Microbiol. 2016;16(1):131.

    Article  PubMed  PubMed Central  Google Scholar 

  41. Gu S, Wan W, Shao Z, Zhong W. High-throughput method for detecting siderophore production by rhizosphere bacteria. Bio Protoc. 2021;11(9):e4001.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  42. Arora NK, Verma M. Modified microplate method for rapid and efficient estimation of siderophore produced by bacteria. 3 Biotech. 2017;7(6):381.

    Article  PubMed  PubMed Central  Google Scholar 

  43. Li Z, Chang S, Lin L, Li Y, An Q. A colorimetric assay of 1-aminocyclopropane-1-carboxylate (ACC) based on ninhydrin reaction for rapid screening of bacteria containing ACC deaminase. Lett Appl Microbiol. 2011;53(2):178–85.

    Article  PubMed  CAS  Google Scholar 

  44. Penrose DM, Glick BR. Methods for isolating and characterizing ACC deaminase-containing plant growth-promoting rhizobacteria. Physiol Plant. 2003;118(1):10–5.

    Article  PubMed  CAS  Google Scholar 

  45. Frank JA, Reich CI, Sharma S, Weisbaum JS, Wilson BA, Olsen GJ. Critical evaluation of two primers commonly used for amplification of bacterial 16S rRNA genes. Appl Environ Microbiol. 2008;74(8):2461–70.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  46. Marie D, Partensky F, Jacquet S, Vaulot D. Enumeration and cell cycle analysis of natural populations of marine picoplankton by flow cytometry using the nucleic acid stain SYBR green I. Appl Environ Microbiol. 1997;63(1):186–93.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  47. Team RC: A language and environment for statistical computing. (No Title) 2021.

  48. Latorre-Pérez A, Gimeno-Valero H, Tanner K, Pascual J, Vilanova C, Porcar M. A round trip to the desert: in situ nanopore sequencing informs targeted bioprospecting. Front Microbiol. 2021;12:768240.

    Article  PubMed  PubMed Central  Google Scholar 

  49. McMurdie PJ, Holmes S. phyloseq: an R package for reproducible interactive analysis and graphics of microbiome census data. PLoS ONE. 2013;8(4):e61217.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  50. Liu C, Cui Y, Li X, Yao M. microeco: an R package for data mining in microbial community ecology. FEMS Microbiol Ecol. 2021. https://doiorg.publicaciones.saludcastillayleon.es/10.1093/femsec/fiaa255.

    Article  PubMed  Google Scholar 

  51. Wemheuer F, Taylor JA, Daniel R, Johnston E, Meinicke P, Thomas T, Wemheuer B. Tax4Fun2: prediction of habitat-specific functional profiles and functional redundancy based on 16S rRNA gene sequences. Environ Microb. 2020;15(1):11.

    Article  CAS  Google Scholar 

  52. Camacho C, Coulouris G, Avagyan V, Ma N, Papadopoulos J, Bealer K, Madden TL. BLAST+: architecture and applications. BMC Bioinform. 2009;10:421.

    Article  Google Scholar 

  53. Patz S, Gautam A, Becker M, Ruppel S, Rodríguez-Palenzuela P, Huson DH: PLaBAse: a comprehensive web resource for analyzing the plant growth-promoting potential of plant-associated bacteria. bioRxiv 2021:2021.2012.2013.472471.

  54. Wickham H, Chang W, Wickham MH. Package ‘ggplot2.’ Create elegant data Visualisations using the Grammar of Graphics Version. 2016;2(1):1–189.

    Google Scholar 

  55. Segata N, Izard J, Waldron L, Gevers D, Miropolsky L, Garrett WS, Huttenhower C. Metagenomic biomarker discovery and explanation. Genome Biol. 2011;12(6):R60.

    Article  PubMed  PubMed Central  Google Scholar 

  56. Noble AS, Noe S, Clearwater MJ, Lee CK. A core phyllosphere microbiome exists across distant populations of a tree species indigenous to New Zealand. PLoS ONE. 2020;15(8):e0237079.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  57. Hakobyan A, Velte S, Sickel W, Quandt D, Stoll A, Knief C. Tillandsia landbeckii phyllosphere and laimosphere as refugia for bacterial life in a hyperarid desert environment. Microbiome. 2023;11(1):246.

    Article  PubMed  PubMed Central  Google Scholar 

  58. Li M, Hong L, Ye W, Wang Z, Shen H. Phyllosphere bacterial and fungal communities vary with host species identity, plant traits and seasonality in a subtropical forest. Environ Microbiome. 2022;17(1):29.

    Article  PubMed  PubMed Central  Google Scholar 

  59. Mechan Llontop ME, Tian L, Sharma P, Heflin L, Bernal-Galeano V, Haak DC, Clarke CR, Vinatzer BA. Experimental evidence pointing to rain as a reservoir of tomato phyllosphere microbiota. Phytobiomes J. 2021;5(4):382–99.

    Article  Google Scholar 

  60. Espenshade J, Thijs S, Gawronski S, Bové H, Weyens N, Vangronsveld J. Influence of urbanization on epiphytic bacterial communities of the platanus × hispanica tree leaves in a biennial study. Front Microbiol. 2019;10:675.

    Article  PubMed  PubMed Central  Google Scholar 

  61. Ares A, Pereira J, Garcia E, Costa J, Tiago I. The leaf bacterial microbiota of female and male kiwifruit plants in distinct seasons: assessing the impact of pseudomonas syringae pv. actinidiae. Phytobiomes J. 2021;5(3):275–87.

    Article  Google Scholar 

  62. Anguita-Maeso M, Ares-Yebra A, Haro C, Román-Écija M, Olivares-García C, Costa J, Marco-Noales E, Ferrer A, Navas-Cortés JA, Landa BB. Xylella fastidiosa infection reshapes microbial composition and network associations in the xylem of almond trees. Front Microbiol. 2022;13:866085.

    Article  PubMed  PubMed Central  Google Scholar 

  63. Allard SM, Ottesen AR, Micallef SA. Rain induces temporary shifts in epiphytic bacterial communities of cucumber and tomato fruit. Sci Rep. 2020;10(1):1765.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  64. Dreyling L, Schmitt I, Dal Grande F. Tree size drives diversity and community structure of microbial communities on the bark of beech (Fagus sylvatica). Front For Glob Change. 2022. https://doiorg.publicaciones.saludcastillayleon.es/10.3389/ffgc.2022.858382.

    Article  Google Scholar 

  65. Abdelfattah A, Wisniewski M, Schena L, Tack AJM. Experimental evidence of microbial inheritance in plants and transmission routes from seed to phyllosphere and root. Environ Microbiol. 2021;23(4):2199–214.

    Article  PubMed  CAS  Google Scholar 

  66. Morales Moreira ZP, Helgason BL, Germida JJ. Environment has a stronger effect than host plant genotype in shaping spring Brassica napus seed microbiomes. Phytobiomes J. 2021;5(2):220–30.

    Article  Google Scholar 

  67. Wassermann B, Cernava T, Goertz S, Zur J, Rietz S, Kögl I, Abbadi A, Berg G. Low nitrogen fertilization enriches nitrogen-fixing bacteria in the Brassica seed microbiome of subsequent generations. J Sustain Agric Environ. 2023;2(2):87–98.

    Article  CAS  Google Scholar 

  68. Berninger T, González López Ó, Bejarano A, Preininger C, Sessitsch A. Maintenance and assessment of cell viability in formulation of non-sporulating bacterial inoculants. Microb Biotechnol. 2018;11(2):277–301.

    Article  PubMed  CAS  Google Scholar 

  69. Moura JB, Delforno TP, do Prado PF, Duarte IC. Extremophilic taxa predominate in a microbial community of photovoltaic panels in a tropical region. FEMS Microbiol Lett. 2021. https://doiorg.publicaciones.saludcastillayleon.es/10.1093/femsle/fnab105.

    Article  PubMed  Google Scholar 

  70. Romani M, Carrion C, Fernandez F, Intertaglia L, Pecqueur D, Lebaron P, Lami R. High bacterial diversity in pioneer biofilms colonizing ceramic roof tiles. Int Biodeterior Biodegrad. 2019;144:104745.

    Article  CAS  Google Scholar 

  71. Ratnaningsih HR, Noviana Z, Dewi TK, Loekito S, Wiyono S, Gafur A, Antonius S. IAA and ACC deaminase producing-bacteria isolated from the rhizosphere of pineapple plants grown under different abiotic and biotic stresses. Heliyon. 2023;9(6):e16306.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  72. Borken W, Horn MA, Geimer S, Aguilar NA, Knorr KH. Associative nitrogen fixation in nodules of the conifer Lepidothamnus fonkii (Podocarpaceae) inhabiting ombrotrophic bogs in southern Patagonia. Sci Rep. 2016;6:39072.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  73. Shao J, Miao Y, Liu K, Ren Y, Xu Z, Zhang N, Feng H, Shen Q, Zhang R, Xun W. Rhizosphere microbiome assembly involves seed-borne bacteria in compensatory phosphate solubilization. Soil Biol Biochem. 2021;159:108273.

    Article  CAS  Google Scholar 

  74. Yang N, Nesme J, Røder HL, Li X, Zuo Z, Petersen M, Burmølle M, Sørensen SJ. Emergent bacterial community properties induce enhanced drought tolerance in Arabidopsis. npj Biofilms Microb. 2021;7(1):82.

    Article  CAS  Google Scholar 

  75. Torres-Cortés G, Bonneau S, Bouchez O, Genthon C, Briand M, Jacques MA, Barret M. Functional microbial features driving community assembly during seed germination and emergence. Front Plant Sci. 2018;9:902.

    Article  PubMed  PubMed Central  Google Scholar 

  76. Moroenyane I, Tremblay J, Yergeau É. Soybean microbiome recovery after disruption is modulated by the seed and not the soil microbiome. Phytobiomes J. 2021;5(4):418–31.

    Article  Google Scholar 

  77. Tom LM, Aulitto M, Wu YW, Deng K, Gao Y, Xiao N, Rodriguez BG, Louime C, Northen TR, Eudes A, et al. Low-abundance populations distinguish microbiome performance in plant cell wall deconstruction. Microbiome. 2022;10(1):183.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  78. Pan C, Sun C, Qu X, Yu W, Guo J, Yu Y, Li X. Microbial community interactions determine the mineralization of soil organic phosphorus in subtropical forest ecosystems. Microbiol Spectr. 2024;12(3):e0135523.

    Article  PubMed  Google Scholar 

  79. Barrett LG, Broadhurst LM, Thrall PH. Geographic adaptation in plant–soil mutualisms: tests using Acacia spp. and rhizobial bacteria. Funct Ecol. 2012;26(2):457–68.

    Article  Google Scholar 

  80. Thrall PH, Millsom DA, Jeavons AC, Waayers M, Harvey GR, Bagnall DJ, Brockwell J. Seed inoculation with effective root-nodule bacteria enhances revegetation success. J Appl Ecol. 2005;42(4):740–51.

    Article  Google Scholar 

  81. Hart MM, Cross AT, D’Agui HM, Dixon KW, Van der Heyde M, Mickan B, Horst C, Grez BM, Valliere JM, Rossel RV, et al. Examining assumptions of soil microbial ecology in the monitoring of ecological restoration. Ecol Solut Evid. 2020;1(2):e12031.

    Article  Google Scholar 

  82. Hay FR, Probert RJ. Advances in seed conservation of wild plant species: a review of recent research. Conserv Physiol. 2013;1(1):cot030.

    Article  PubMed  PubMed Central  Google Scholar 

Download references

Acknowledgements

This work was funded by the Australian Research Council Linkage Project grant LP200200688 to RVG, SGT and MMM. The authors would like to acknowledge the contribution of Paden Willson from Greening Australia, and Katherine Thompson from the Australian PlantBank for supplying seeds used in this study and Charlotte Mills from Airseed Technologies Australia for helpful discussions.

Funding

This work was funded by the Australian Research Council Linkage Project grant LP200200688 to RVG, SGT and MMM.

Author information

Authors and Affiliations

  1. School of Natural Sciences, Macquarie University, Sydney, NSW, Australia

    Dylan Russell, Vaheesan Rajabal, Matthew Alfonzetti & Sasha G. Tetu

  2. ARC Centre of Excellence for Synthetic Biology, Macquarie University, Sydney, NSW, Australia

    Vaheesan Rajabal & Sasha G. Tetu

  3. Research Centre for Ecosystem Resilience, Australian Institute of Botanical Science, Royal Botanic Gardens and Domain Trust, Sydney, NSW, Australia

    Marlien M. van der Merwe

  4. Hawkesbury Institute for the Environment, Western Sydney University, Sydney, NSW, Australia

    Rachael V. Gallagher

Authors
  1. Dylan Russell
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Vaheesan Rajabal
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Matthew Alfonzetti
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Marlien M. van der Merwe
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Rachael V. Gallagher
    View author publications

    You can also search for this author inPubMed Google Scholar

  6. Sasha G. Tetu
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

DHR, RVG, MMM, and SGT conceptualised the study and contributed to sampling design. DHR and SGT designed the experiment. RVG and SGT supervised the study. DHR performed field sampling, sample processing, and laboratory experiments. DHR performed amplicon sequencing and analysis with the assistance of VR. DHR wrote the first draft of the manuscript. DHR, MA, RVG, MMM, and SGT reviewed and edited the manuscript. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Sasha G. Tetu.

Ethics declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Russell, D., Rajabal, V., Alfonzetti, M. et al. Seed banking impacts native Acacia ulicifolia seed microbiome composition and function. Environmental Microbiome 20, 4 (2025). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s40793-024-00657-3

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s40793-024-00657-3

Keywords

Environmental Microbiome

ISSN: 2524-6372