16S rRNA metagenomic analysis of the bacterial community associated with turf grass seeds from low moisture and high moisture climates

Abstract

Turfgrass investigators have observed that plantings of grass seeds produced in moist climates produce seedling stands that show greater stand evenness with reduced disease compared to those grown from seeds produced in dry climates. Grass seeds carry microbes on their surfaces that become endophytic in seedlings and promote seedling growth. We hypothesize that incomplete development of the microbiome associated with the surface of seeds produced in dry climates reduces the performance of seeds. Little is known about the influence of moisture on the structure of this microbial community. We conducted metagenomic analysis of the bacterial communities associated with seeds of three turf species (Festuca rubra, Lolium arundinacea, and Lolium perenne) from low moisture (LM) and high moisture (HM) climates. The bacterial communities were characterized by Illumina high-throughput sequencing of 16S rRNA V3–V4 regions. We performed seed germination tests and analyzed the correlations between the abundance of different bacterial groups and seed germination at different taxonomy ranks. Climate appeared to structure the bacterial communities associated with seeds. LM seeds vectored mainly Proteobacteria (89%). HM seeds vectored a denser and more diverse bacterial community that included Proteobacteria (50%) and Bacteroides (39%). At the genus level, Pedobacter (20%), Sphingomonas (13%), Massilia (12%), Pantoea (12%) and Pseudomonas (11%) were the major genera in the bacterial communities regardless of climate conditions. Massilia, Pantoea and Pseudomonas dominated LM seeds, while Pedobacter and Sphingomonas dominated HM seeds. The species of turf seeds did not appear to influence bacterial community composition. The seeds of the three turf species showed a core microbiome consisting of 27 genera from phyla Actinobacteria, Bacteroidetes, Patescibacteria and Proteobacteria. Differences in seed-vectored microbes, in terms of diversity and density between high and LM climates, may result from effects of moisture level on the colonization of microbes and the development of microbe community on seed surface tissues (adherent paleas and lemmas). The greater diversity and density of seed vectored microbes in HM climates may benefit seedlings by helping them tolerate stress and fight disease organisms, but this dense microbial community may also compete with seedlings for nutrients, slowing or modulating seed germination and seedling growth.

Introduction

Plants bear numerous microbes that influence their nutrition, development, stress responses and phenotypes (Hardoim et al., 2015; Henning et al., 2016; White et al., 2014, 2018). Plant-associated microbes generally come from the surrounding environment; however, Johnston-Monje & Raizada (2011) showed that corn seeds vectored a diverse array of microbes. Further, Johnston-Monje et al. (2016) found that the rhizospheres of corn seedlings were composed of microbes that originated both from seeds and bacteria recruited from soils. These seed-vectored bacteria can influence germination and share a mutualistic association with the host seedlings (Cruz, Yañez-Ocampo & Wong-Villarreal, 2014; Shaik & Thomas, 2019; Somova et al., 2001; Zhu et al., 2017). Without seed-vectored bacteria, seedlings may lose gravitropic response of roots, fail to develop root hairs, and are more susceptible to soil-borne pathogens (Verma et al., 2017, 2018).

Some turf grasses possess fungal endophytes of ascomycete genus Epichloë that provide resistance to pathogens and insects, and increase abiotic stress tolerance in the host (Bultman & Bell, 2003; Clay, 1990; Meyer, Torres & White, 2012; White, 1987). Turf breeders have long been employing these environmentally safe endophytes to enhance turfgrass performance and stress tolerance (Meyer, Torres & White, 2012). Another important microbe resource, the seed-transmitted bacterial communities of turf grasses, are yet to be fully explored. Many of these bacteria are vectored on the surfaces of seeds and embedded within dried plant tissues (paleas and lemmas) that adhere tightly to seed surfaces (White et al., 2019). During seed germination some of these seed-surface microbes are activated and they externally and internally colonize seedling roots at the root tip meristems, becoming intercellular and intracellular endophytes in the emergent seedling roots (Verma et al., 2017, 2018; White et al., 2018). In this study, we employed Illumina HTS and 16S metagenomic analysis to investigate the bacterial community associated with cool-season turfgrass seeds produced in low moisture (LM) and high moisture (HM) climates (Pace et al., 1986; Riesenfeld, Schloss & Handelsman, 2004). We also evaluated the potential influence of the bacterial community on seed germination rates and seedling growth rates. The results showed that HM seeds vectored a denser and more diverse bacterial community than LM seeds. Also, bacterial groups at different taxonomic ranks correlated with the seed germination rate and time.

Materials and Methods Total DNA extraction from seeds of cool-season turfgrasses

Seeds of 27 cool-season turf cultivars were obtained from DLF Pickseed USA (Table S1). All varieties were produced from 2011 to 2015 at either Store Hedinge, Denmark or Les Alleuds, France. Based on the precipitation data collected from The National Oceanic and Atmospheric Administration, the seeds were classified into LM seeds (annual precipitation <750 mm) and HM seeds (annual precipitation >750 mm). With this classification, five samples were classified as LM seeds while 22 samples as HM seeds. A total of 100 mg of seeds of each turf cultivar were weighed out and washed with water for three times, 30 s each time to remove the dirt. The cleaned seeds were then ground into powder with a sterilized mortar and pestle for total DNA extraction. The DNA extraction was conducted with DNeasy^® PowerSoil^® Kit (QIAGEN, Hilden, Germany) following the manufacturer’s instructions. This PowerSoil^® Kit was chosen due to its versatility with diverse sample types. The concentration of extracted DNA was measured with The NanoDrop^® ND-1000 Spectrophotometer and normalized to five ng/µl for the library preparation.

Library preparation and sequencing

The preparation of DNA libraries for each sample followed the Illumina guidelines. By using 12.5 ng of the normalized DNA from turf seeds as the template, V3–V4 hypervariable regions of bacterial 16S rRNA gene were amplified with the primer pair, S-D-Bact-0341-b-S-17 (5′-CCTACGGGNGGCWGCAG-3′) and S-D-Bact-0785-a-A-21 (5′-GACTACHVGGGTATCTAATCC-3′) fuzed with Illumina overhang forward adapters (5′-TCGTCGGCAGCGTCAGATGTGTATAAGAGACAG-3′) and reverse adapter (5′-GTCTCGTGGGCTCGGAGATGTGTATAAGAGACAG-3′), respectively (Klindworth et al., 2013). PCR clean-ups were conducted to purify the 16S V3–V4 amplicons away from free primers and primer dimers. Nextera XT index primers were then used for the index PCR and PCR clean-ups were performed again to generate the final library. The generated 16S V3–V4 region library was paired-end sequenced (2 × 300 bp) on an Illumina MiSeq platform in the Genome Cooperative Sequencing Facility, School of Environmental and Biological Sciences at Rutgers.

Bacterial community structure analysis

The collected sequencing data in FASTQ format was processed and analyzed with the QIIME2 software suite (Caporaso et al., 2010). The raw Illumina reads were imported into QIIME2 with “Casava 1.8 paired-end demultiplexed fastq” method, and then denoised and filtered with dada2 pipeline to remove noisy and chimeric sequences, construct denoised paired-end sequences, and dereplicate them (Callahan et al., 2016). De novo clustering was then carried out with VSEARCH plugin at 99% identity to generate Operational Taxonomic Units (OTUs) (Rognes et al., 2016). The taxonomy assignment of OTUs was performed by using feature-classifier against the SILVA 1.28 database (released 29 September 2016). After removing mitochondria and chloroplast sequences, the filtered data were aligned with mafft program and fasttree method to generate rooted and unrooted phylogenetic trees (Price, Dehal & Arkin, 2010). All core metrics used in alpha and beta diversity analysis were computed based on the rooted phylogenetic tree. Alpha diversity (intra group diversity) was calculated with the observed OTUs and Faith’s phylogenetic diversity (Faith, 1992) at the sample depth of 1,000 reads to normalize the variance and this excluded four samples (three HM samples and one LM sample), leaving four LM samples and 19 HM samples. The Kruskal–Wallis (pairwise) test was utilized to assess the statistical significance of alpha diversity. Beta diversity was performed with both qualitative (Jaccard and unweighted UniFrac) and quantitative (Bray–Curtis and weighted UniFrac) distance metrics at sample depth of 1,000 reads. In this process, QIIME2 diversity plugin was employed. Statistical significance among different groups was evaluated by permutation-based ANOVA (PerMANOVA) test (Anderson, 2005) with 999 permutations (beta-group-significance command in diversity plugin). Principal coordinates analysis plots (PCoA) were generated by Emperor tool of QIIME2 to explore the bacterial community structure. The bar plots showing taxonomy levels were generated by QIIME2 taxa plugin. The metagenomic analysis workflow is shown in Fig. 1.

Figure 1: Graphical workflow of metagenomic analysis in our study. DOI: 10.7717/peerj.8417/fig-1

The Venn diagram was generated with a WWW-based tool to calculate the intersection(s) of the list of elements that in this study was represented by the list of genera of bacteria found in each climate condition and species. The graphical output is in the form of a Venn/Euler diagram.

Seed germination test

Seeds of 19 cool-season cultivars were placed in Petri dishes containing 25 ml 1.5% agar. All Petri dishes were kept in a growth chamber at 28 °C. Seed germination was observed every 24 h until no more seed germinated. Seed germination rates and time for each sample were calculated and correlation analysis was performed and visualized with python-based libraries SciPy (V0.19.1), pandas (V0.22.0), seaborn (V0.9.0) and matplotlib (V2.2.3).

Results Sequence analysis

In total, 7,405,226 sequences (about 274,368 sequences per sample) were generated by Illumina MiSeq sequencing and imported into QIIME2 pipeline suite for analysis. After being denoised and dereplicated with dada2 pipeline, the remaining high-quality sequences were clustered into 310 OTUs that had an average length of 427 bp, ranging from 267 bp to 440 bp. After the removal of mitochondrial and chloroplast genomes, a total of 247 OTUs were used to represent the bacterial profile of turf seeds samples (Table S2).

Diversity of bacterial endophytes associated with turf seeds from LM and HM climates

The bacterial community associated with turf seed samples was composed of five phyla, eight classes, 21 orders, 37 families and 69 genera. The bacterial community vectored by seeds produced in HM climates covered all discovered taxonomies, with 6,644 sequences/sample. However, seeds from LM climate only hosted part of them, four phyla, eight classes, nine orders, 10 families and 15 genera, with 2,821 sequences/sample.

Regardless of the climate and turf species, bacterial communities at phylum level were dominated by Proteobacteria (51%) and Bacteroidetes (40%). Proteobacteria took 89% and 50% of the bacterial community on HM and LM climates seeds, respectively. Bacteroidetes was abundant in HM climate seeds (39%) but not LM climate seeds (2%). Actinobacteria (6%) and Firmicutes (3%) also comprised a portion of the bacterial community and exhibited no significant difference between the two climate types.

Both HM and LM seeds shared some of the most abundant bacterial classes, that is, Actinobacteria, Bacteroidia, Bacilli, Alphaproteobacteria and Gammaproteobacteria (Fig. 2). Compared to LM seeds, seeds from HM environment was richer in terms of Faith’s phylogenetic diversity (Fig. 3; Table S3). At class level, Actinobacteria, Bacteroidia and Bacilli had the same portion as phyla Actinobacteria, Bacteroidetes and Firmicutes, respectively (Table 1). Alphaproteobacteria and Gammaproteobacteria were the two classes within phylum Proteobacteria. Alphaproteobacteria took 6% and 18% of the bacterial community of LM and HM climate seeds, respectively. Gammaproteobacteria was 83% and 32% for LM and HM climate, respectively.

Figure 2: Bar plot analysis illustrating the relative abundance and distribution of the OTUs assigned to class-level taxonomy. LM, low moisture; HM, high moisture. DOI: 10.7717/peerj.8417/fig-2

Figure 3: Box plots depicting the Faith’s phylogenetic diversity for different climate conditions (A), different genera (B) and different species (C). DOI: 10.7717/peerj.8417/fig-3

Table 1: Composition of bacterial community from LM and HM climate seeds.

Taxa level	Taxa name	Average percentage
LM (%)	HM (%)	Combined (%)
Phylum	Actinobacteria**	6	7	6
	Bacteroidetes†	2	39	40
	Firmicutes**	3	4	3
	Patescibacteria	–	<1	<1
	Proteobacteria†	89	50	51
Class	Actinobacteria**	6	6	6
	Bacteroidia	2	39	40
	Bacilli**	3	4	3
	Saccharimonadia	–	<1	<1
	Alphaproteobacteria†	6	18	15
	Gammaproteobacteria†	83	32	36
Order	Micrococcales**	6	5	5
	Cytophagales	<1	2	2
	Flavobacteriales†	<1	11	9
	Sphingobacteriales†	1	26	21
	Bacillales**	3	4	4
	Saccharimonadales	–	<1	<1
	Rhizobiales	–	3	2
	Sphingomonadales†	6	15	13
	Betaproteobacteriales†	30	13	17
	Enterobacteriales†	24	12	15
	Pseudomonadales†	29	6	11
Family	Microbacteriaceae**	6	5	5
	Hymenobacteraceae	<1	1	1
	Flavobacteriaceae†	<1	3	3
	Weeksellaceae†	<1	7	6
	Sphingobacteriaceae†	1	26	21
	Paenibacillaceae**	3	4	4
	Rhizobiaceae†	–	2	1
	Sphingomonadaceae†	6	15	13
	Burkholderiaceae†	30	13	17
	Enterobacteriaceae†	24	12	15
	Pseudomonadaceae†	29	6	11
Genus	Curtobacterium**	2	2	2
	Hymenobacter	<1	1	1
	Flavobacterium†	<1	3	3
	Chryseobacterium†	<1	7	6
	Mucilaginibacter	–	<1	<1
	Pedobacter†	1	25	20
	Paenibacillus**	3	4	4
	Rhizobium*	–	2	1
	Sphingomonas†	6	15	13
	Duganella†	5	3	3
	Massilia†	25	9	12
	Erwinia†	<1	3	3
	Pantoea†	23	9	12
	Pseudomonas†	29	6	11
Unassigned		<1	<1	<1

DOI: 10.7717/peerj.8417/table-1

Notes:

*Rhizobium group also includes Allorhizobium, Neorhizobium, and Pararhizobium.

**Bacterial groups without significant difference between LM and HM.

†Bacterial groups with significant difference between LM and HM.

At genus level, LM seeds harbored a significantly higher percentage of Massilia (p = 0.013), Pantoea (p = 0.060) and Pseudomonas (p = 0.045) compared to HM seeds (Table 1). In contrast, HM seeds harbored more of Flavobacterium (p < 0.001), Chryseobacterium (p < 0.001), Pedobacter (p < 0.001), Sphingomonas (p = 0.035) and Erwinia (p = 0.122) (Table 1).

Principal coordinates analysis plots separated bacterial communities associated with turf seeds by climate (Fig. 4). Also, the PERMANOVA test showed a significant difference between the two groups (p = 0.002, Table S4). However, no significant correlation was detected between different species of turf seeds and their bacterial profile (Festuca rubra vs. Lolium arundinacea, p = 0.101; F. rubra vs. L. perenne, p = 0.109; L. arundinacea vs. L. perenne, p = 0.204, Table S4).

Figure 4: PCoA Emperor plots based on Bray–Cuitis diversity matrix. Samples are scattered concerning their bacterial community. Climates are represented by different colors: red-low moisture; blue-high moisture. Species were represented by different shapes: ring-Loium arudinacea; sphere-Lolium perenne; square-Festuca rubra. DOI: 10.7717/peerj.8417/fig-4

LM seeds didn’t bear any unique bacteria genus that HM seed didn’t (Fig. 5A; Table S5). However, HM seeds harbored 55 genera that LM seeds didn’t. Seeds of three turf species shared four phyla and 27 genera, including five genera that were uncultured or unknown species from either Bacterioidetes or Patescibacteria (Fig. 5B; Table 2).

Figure 5: Venn diagrams showing the number of shared bacterial genera between different climate (A) and among different turf species (B). LM, low moisture; HM, high moisture; Fr, Festuca rubra; Lp, Lolium perenne; La, Loium arudinacea. DOI: 10.7717/peerj.8417/fig-5

Table 2: Seed-vectored bacterial genera shared Loium arudinacea, Lolium perenne, and Festuca rubra.

Phylum	Genus
Actinobacteria	Curtobacterium
	Sanguibacter
Bacteroidetes	Chryseobacterium
	Dyadobacter
	Flavobacterium
	Mucilaginibacter
	Pedobacter
	Sphingobacterium
	Spirosoma
	Uncultured Sphingobacteriaceae
	Unknown Sphingobacteriaceae
Patescibacteria	Uncultured bacterium
	Uncultured Sphingobium sp.
	Unknown Saccharimonadales
Proteobacteria	Rhizobium*
	Aureimonas
	Brevundimonas
	Devosia
	Duganella
	Erwinia
	Massilia
	Novosphingobium
	Pantoea
	Pigmentiphaga
	Pseudomonas
	Sphingomonas
	Verticia

DOI: 10.7717/peerj.8417/table-2

Note:

*Rhizobium group also includes Allorhizobium, Neorhizobium, and Pararhizobium.

Correlation of seed germination and bacterial endophyte composition associated with turf seeds

Bacterial groups at different taxonomic ranks correlated with the seed germination rate and time (Figs. 6 and 7). Among the five phyla that we discovered through diversity analysis, Proteobacteria correlated positively with the seed germination rate (p = 0.028) and negatively with the seed germination time (p = 0.016). Phylum Actinobacteria also showed a negative correlation with the seed germination time (p = 0.040) but not a significant correlation with germination rate (p = 0.120). Another phylum, Firmicutes, showed correlation with germination (p = 0.109) rate and germination time (p = 0.069), but this was not statistically significant. However, the abundance of Bacteroidetes was negatively associated with the seed germination rate (p = 0.008), and positively associated with the seed germination time (p = 0.002).

Figure 6: Correlation of seed germination rate with abundance of bacteria groups at different taxonomy levels. (A) and (B) Phylum level. (C) Class level. (D) and (E) Order level. (F)–(H) Family level. (I)–(K) Genus level. (L) Species level. DOI: 10.7717/peerj.8417/fig-6

Figure 7: Correlation of average seed germination time with abundance of bacteria groups at different taxonomy levels. (A)–(C) Phylum level. (D)–(F) Class level. (G)–(K) Order level. (L)–(P) Family level. (Q)–(T) Genus level. (U) Species level. DOI: 10.7717/peerj.8417/fig-7

At class level, Bacilli and Gammaproteobacteria were groups showing exactly the same correlation as phyla Firmicutes and Proteobacteria, respectively (Figs. 6 and 7). Also, the abundance of Bacteroidia and Gammaproteobacteria showed a similar correlation to phylum Bacteroidetes and Proteobacteria.

At family level, seed germination rate was positively related to the abundance of bacteria from families Microbacteriaceae (p = 0.090), Paneibacillaceae (p = 0.109) and Pseudomonadaceae (p = 0.138), and negatively associated with the abundance of bacteria from Rhizobiaceae (p = 0.014), Sphingobacteriaceae (p = 0.005), and Weeksellaceae (p = 0.033). As expected, seed germination time also correlated negatively with the abundance of these bacterial families. Seed germination time was positively associated with Rhizobiaceae (p = 0.004), Sphingobacteriaceae (p = 0.002), and Weeksellaceae (p = 0.008), but negatively with Microbacteriaceae (p = 0.032) and Paneibacillaceae (0.069) and Pseudomonadaceae (p = 0.049).

At genus level, the abundance of Rhizobium, Chryseobacterium and Pedobacter was negatively associated with germination rate (p-value 0.041, 0.033 and 0.004, respectively) but positively with germination time (p-value 0.015, 0.008 and 0.001, respectively). But the abundance of Pseudomonas was positively related with germination rate (p = 0.138) but negatively associated with the germination time (p = 0.049), although the correlation was not significant.

Discussion A complex bacterial community associated with turf seeds

Compared to LM seeds, HM seeds harbored a more diverse bacterial community with many more bacterial cells (i.e., a higher bacterial load), as there were more sequences associated with HM seeds. The more individuals hypothesis, which predicts that communities with more individuals will have more species (Storch, Bohdalková & Okie, 2018), can explain the higher diversity associated with HM seeds. This result is similar to previous studies on soils indicating that moisture controls the structure and function of the soil microbial community (Brockett, Prescott & Grayston, 2012; Griffiths et al., 2003; Steven et al., 2013). Water availability in soil controls bacterial composition (Zeglin et al., 2011). Similarly, relatively HM will favor growth and replication of bacteria on seeds, while LM conditions will suppress development of the bacterial community associated with seeds.

Both LM and HM seeds vectored a large number of bacteria and shared some groups. For example, the abundance of Curtobacterium spp. was similar in both LM and HM seeds (LM 2%, HM 2%). Curtobacterium is a Gram-positive endophytic bacterial genus in rice seeds (Oryza sativa), field-grown tall fescue (L. arundinacea) and Noccaea goesingensis (De Los Santos et al., 2015; Mano et al., 2006; Ruiz et al., 2011). Some Curtobacterium strains provide host growth promotion and pathogen antagonistic effects (De Los Santos et al., 2015; Ruiz et al., 2011). Paenibacillus was the only genus from phylum Firmicutes in both LM and HM seeds (LM 3%, HM 4%). Paenibacillus have been isolated from many plants and shown to produce IAA, solubilize phosphate and inhibit the growth of phytopathogens (Aswathy et al., 2013; Diaz Herrera et al., 2016; Ruiz et al., 2011; Rybakova et al., 2015).

Two of the genera, Mucilaginibacter and Rhizobium, were found in HM seeds but not LM seeds. Mucilaginibacter spp. can promote plant growth and produce extracellular polysaccharides (An et al., 2009; Lee et al., 2013; Madhaiyan et al., 2010; Mannisto et al., 2010). Rhizobium together with Allorhizobium, Neorhizobium, and Pararhizobium, composed 2% of the bacterial community on HM seeds. These genera comprise well-studied bacteria that promote growth of plants and nodulate legumes to fix nitrogen (Datta & Basu, 2000; Gutierrez-Zamora & Martınez-Romero, 2001; Kiers et al., 2003; Yanni et al., 1997).

At the phylum level, LM seeds hosted more Gammaproteobacteria than HM seeds. Several genera within Gammaproteobacteria contributed to these results, that is, Duganella, Massilia, Pantoea, and Pseudomonas. However, these bacteria were still found on a large portion of the HM seeds. Duganella spp. can suppress the growth of plant pathogens (Cretoiu et al., 2013; Haack et al., 2016). Massilia is a root-colonizing bacterial genus with the ability to degrade chitin (Adrangi et al., 2010; Faramarzi et al., 2009; Ofek, Hadar & Minz, 2012). Pantoea spp. promote plant growth and tolerance of environmental stresses (Chen et al., 2017; Feng, Shen & Song, 2006; Ferreira et al., 2008; Gond et al., 2015). Pseudomonas contains many endophytic bacterial strains that benefit hosts by producing IAA, producing biocontrol lipopeptides, and solubilizing phosphate (Oteino et al., 2015; Prieto & Mercado-Blanco, 2008; Suzuki, He & Oyaizu, 2003).

Some bacterial genera were more abundant in HM seeds than LM seeds, including Flavobacterium, Chryseobacterium, Pedobacter, Sphingomonas and Erwinia. Most of the bacteria comprised a very small portion of the bacterial community of LM seeds, but Sphingomonas made up 6%. Flavobacterium sp. has been found to promote plant growth and provide biocontrol activity to the hosts (Kolton et al., 2016; Soltani et al., 2010). Chryseobacterium spp. were also shown to be plant growth promoting bacteria (Dardanelli et al., 2009; Gutiérrez Mañero et al., 2003). Although Pedobacter has not been found to promote growth of plants, it can induce the production of antimicrobial compounds by Pseudomonas fluorescens Pf0-1 (Garbeva et al., 2011). Sphingomonas is an alphaproteobacterial genus containing strains that produce IAA and provide nutrients to hosts (Okunishi et al., 2005; Ruiz et al., 2011). Erwinia spp. have also been identified as endophytes in some plant species (Verma, 2019). However, genus Erwinia is well-known to contain many plant pathogenic species.

In total, the above genera together comprised 89% and 95% in HM and LM seeds, respectively. Some of the bacterial genera include plant pathogens, for example, Ewrinia and Pseudomonas. However, most of the bacteria are known to contain mainly plant growth promoting rhizobacteria, for example, Paenibacillus, Pseudomonas, Rhizobium, Pantoea. The seed microbes are important because they stimulate seedling development, increase stress tolerance in seedlings and protect seedlings from disease (Verma et al., 2017, 2018; White et al., 2018, 2019). Thus, a more diverse bacterial community on HM seeds may provide hosts with more microbial resources to utilize for plant development and stress tolerance.

In the study, the samples of LM and HM seeds had unequal sizes (LM: 4; HM: 19), which could create a bias in our final result. However, HM seeds vectored a more diverse bacterial community with significantly more bacteria cells (Fig. 3; Table S3). Also, the different abundances between bacterial groups were statistically significant.

The bacterial community affected seed germination and growth

Seeds from HM climates tended to show slower germination and reduced seedling growth rates. These seeds vectored a denser and more diverse community of bacteria, which may benefit seedlings but not without a cost. We hypothesize that the higher microbial load competes with seedlings, which slows germination and development of the host. This nutritional cost may result in slower seed germination and seedling development rates. Seed growers have observed that seed from HM climates seems to establish better with reduced damping-off disease compared to seed from LM climates (W. Meyer, 2016, unpublished data). While, seeds with richer and denser microbiomes grow slower initially, they may be better protected from soil borne pathogens than seeds with less developed microbiomes.

Seeds that have formed in LM situations, or where the natural microbiome has otherwise been damaged, could be remediated through application of microbes in seed coatings (Pedrini et al., 2017). Coating formulations with the correct microbes at the optimal concentrations could result in better fitness of seeds and seedlings.

Conclusions

We surveyed the bacterial community associated with seeds of several species of cool-season turfgrasses and identified the dominant bacterial groups of the communities at different taxonomic levels. Regardless of the moisture level during seed production and species of seeds, the core bacterial community included many PGPB strains. Seeds produced in HM conditions maintained a denser and more diverse bacterial community than seeds produced in LM conditions. This seed microbiome may help seedlings tolerate stress but may also compete with seedlings for nutrients and slow early seedling growth.

Supplemental Information

Information of seed resources.

For each seed sample, sample number, species, variety, production year, production location, and the annual rainfall are reported.

DOI: 10.7717/peerj.8417/supp-1

Download

OTUs Assignment.

For each unique OTU, Feature ID, assigned Taxonomy and the confidence value of the assignment are reported.

DOI: 10.7717/peerj.8417/supp-2

Download

Krustal-Wallis pairwise test with measure of Faith PD metric as response variable.

Climate, species, and genus were listed in separate worksheet.

DOI: 10.7717/peerj.8417/supp-3

Download

PERMANOVA pairwise results considering as response variable Bray-Curtis dissimilarity matrix.

Climate, species, and genus were listed in separate worksheet.

DOI: 10.7717/peerj.8417/supp-4

Download

Venn diagram results.

LM: low moisture; HM: high moisture; Fr: Festuca rubra; Lp: Lolium perenne; La: Loium arudinacea.

DOI: 10.7717/peerj.8417/supp-5

Download

Additional Information and Declarations

Competing Interests

The authors declare that they have no competing interests.

Author Contributions

Qiang Chen conceived and designed the experiments, performed the experiments, analyzed the data, prepared figures and/or tables, authored or reviewed drafts of the paper, and approved the final draft.

William A. Meyer conceived and designed the experiments, authored or reviewed drafts of the paper, and approved the final draft.

Qiuwei Zhang analyzed the data, prepared figures and/or tables, and approved the final draft.

James F. White conceived and designed the experiments, authored or reviewed drafts of the paper, and approved the final draft.

Data Availability

The following information was supplied regarding data availability:

The raw data is available on at FigShare: Chen, Qiang; Meyer, William A.; Zhang, Qiuwei; White, James F. (2019): turf_seeds_raw_data. figshare. DOI 10.6084/m9.figshare.9745061.v1.

Funding

This work was supported by USDA-NIFA Multistate Project W4147, the Rutgers Turf Science Center, and the New Jersey Agricultural Experiment Station. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

References

Adrangi S, Faramarzi MA, Shahverdi AR, Sepehrizadeh Z. 2010. Purification and characterization of two extracellular endochitinases from Massilia timonae. Carbohydrate Research 345(3):402-407

An DS, Yin CR, Lee ST, Cho CH. 2009. Mucilaginibacter daejeonensis sp. nov., isolated from dried rice straw. International Journal of Systematic and Evolutionary Microbiology 59(5):1122-1125

Anderson MJ. 2005. PERMANOVA: a FORTRAN computer program for permutational multivariate analysis of variance. New Zealand: Department of Statistics, University of Auckland. 24

Aswathy AJ, Jasim B, Jyothis M, Radhakrishnan EK. 2013. Identification of two strains of Paenibacillus sp. as indole 3 acetic acid-producing rhizome-associated endophytic bacteria from Curcuma longa. 3 Biotech 3(3):219-224

Brockett BFT, Prescott CE, Grayston SJ. 2012. Soil moisture is the major factor influencing microbial community structure and enzyme activities across seven biogeoclimatic zones in western Canada. Soil Biology and Biochemistry 44(1):9-20

Bultman TL, Bell GD. 2003. Interaction between fungal endophytes and environmental stressors influences plant resistance to insects. Oikos 103(1):182-190

Callahan BJ, McMurdie PJ, Rosen MJ, Han AW, Johnson AJ, Holmes SP. 2016. DADA2: high-resolution sample inference from Illumina amplicon data. Nature Methods 13(7):581-583

Caporaso JG, Kuczynski J, Stombaugh J, Bittinger K, Bushman FD, Costello EK, Fierer N, Peña AG, Goodrich JK, Gordon JI. 2010. QIIME allows analysis of high-throughput community sequencing data. Nature Methods 7(5):335-336

Chen C, Xin K, Liu H, Cheng J, Shen X, Wang Y, Zhang L. 2017. Pantoea alhagi, a novel endophytic bacterium with ability to improve growth and drought tolerance in wheat. Scientific Reports 7(1):41564

Clay K. 1990. Fungal endophytes of grasses. Annual Review of Ecology and Systematics 21(1):275-297

Cretoiu MS, Korthals GW, Visser JH, Van Elsas JD. 2013. Chitin amendment increases soil suppressiveness toward plant pathogens and modulates the actinobacterial and oxalobacteraceal communities in an experimental agricultural field. Applied and Environmental Microbiology 79(17):5291-5301

Cruz RS, Yañez-Ocampo G, Wong-Villarreal A. 2014. Effect of nodulating bacteria on the seed germination of Capsicum spp. African Journal of Microbiology Research 8(7):659-663

Dardanelli MS, Manyani H, González-Barroso S, Rodríguez-Carvajal MA, Gil-Serrano AM, Espuny MR, López-Baena FJ, Bellogín RA, Megías M, Ollero FJ. 2009. Effect of the presence of the plant growth promoting rhizobacterium (PGPR) Chryseobacterium balustinum Aur9 and salt stress in the pattern of flavonoids exuded by soybean roots. Plant and Soil 328(1–2):483-493

Datta C, Basu PS. 2000. Indole acetic acid production by a Rhizobium species from root nodules of a leguminous shrub, Cajanus cajan. Microbiological Research 155(2):123-127

De Los Santos MC, Taulé C, Mareque C, Beracochea M, Battistoni F. 2015. Identification and characterization of the part of the bacterial community associated with field-grown tall fescue (Festuca arundinacea) cv. SFRO Don Tomás in Uruguay. Annals of Microbiology 66(1):329-342

Diaz Herrera S, Grossi C, Zawoznik M, Groppa MD. 2016. Wheat seeds harbour bacterial endophytes with potential as plant growth promoters and biocontrol agents of Fusarium graminearum. Microbiological Research 186-187:37-43

Faith DP. 1992. Conservation evaluation and phylogenetic diversity. Biological Conservation 61(1):1-10

Faramarzi M, Fazeli M, Yazdi MT, Adrangi S, Al-Ahmadi KJ, Tasharrofi N, Mohseni FA. 2009. Optimization of cultural conditions for production of chitinase by a soil isolate of Massilia timonae. Biotechnology 8(1):93-99

Feng Y, Shen D, Song W. 2006. Rice endophyte Pantoea agglomerans YS19 promotes host plant growth and affects allocations of host photosynthates. Journal of Applied Microbiology 100(5):938-945

Ferreira A, Quecine MC, Lacava PT, Oda S, Azevedo JL, Araujo WL. 2008. Diversity of endophytic bacteria from Eucalyptus species seeds and colonization of seedlings by Pantoea agglomerans. FEMS Microbiology Letters 287(1):8-14

Garbeva P, Silby MW, Raaijmakers JM, Levy SB, Boer W. 2011. Transcriptional and antagonistic responses of Pseudomonas fluorescens Pf0-1 to phylogenetically different bacterial competitors. ISME Journal 5(6):973-985

Gond SK, Torres MS, Bergen MS, Helsel Z, White JF. 2015. Induction of salt tolerance and up-regulation of aquaporin genes in tropical corn by rhizobacterium Pantoea agglomerans. Letters in Applied Microbiology 60(4):392-399

Griffiths RI, Whiteley AS, O’Donnell AG, Bailey MJ. 2003. Physiological and community responses of established grassland bacterial populations to water stress. Applied and Environmental Microbiology 69(12):6961-6968

Gutiérrez Mañero FJ, Probanza A, Ramos B, Colón Flores JJ, Lucas García JA. 2003. Effects of culture filtrates of rhizobacteria isolated from wild lupine on germination, growth, and biological nitrogen fixation of lupine seedlings. Journal of Plant Nutrition 26(5):1101-1115

Gutierrez-Zamora M, Martınez-Romero E. 2001. Natural endophytic association between Rhizobium etli and maize (Zea mays L.) Journal of Biotechnology 91(2–3):117-126

Haack FS, Poehlein A, Kroger C, Voigt CA, Piepenbring M, Bode HB, Daniel R, Schafer W, Streit WR. 2016. Molecular keys to the Janthinobacterium and Duganella spp. interaction with the plant pathogen Fusarium graminearum. Frontiers in Microbiology 7:1668

Hardoim PR, Van Overbeek LS, Berg G, Pirttila AM, Compant S, Campisano A, Doring M, Sessitsch A. 2015. The hidden world within plants: ecological and evolutionary considerations for defining functioning of microbial endophytes. Microbiology and Molecular Biology Reviews 79(3):293-320

Henning JA, Weston DJ, Pelletier DA, Timm CM, Jawdy SS, Classen AT. 2016. Root bacterial endophytes alter plant phenotype, but not physiology. PeerJ 4(8):e2606

Johnston-Monje D, Lundberg DS, Lazarovits G, Reis VM, Raizada MN. 2016. Bacterial populations in juvenile maize rhizospheres originate from both seed and soil. Plant and Soil 405(1–2):337-355

Johnston-Monje D, Raizada MN. 2011. Conservation and diversity of seed associated endophyes in Zea across boundaries of evolution, ethnography and ecology. PLOS ONE 6(6):e20396

Kiers ET, Rousseau RA, West SA, Denison RF. 2003. Host sanctions and the legume–rhizobium mutualism. Nature 425:78-81

Klindworth A, Pruesse E, Schweer T, Peplies J, Quast C, Horn M, Glöckner FO. 2013. Evaluation of general 16S ribosomal RNA gene PCR primers for classical and next-generation sequencing-based diversity studies. Nucleic Acids Research 41(1):e1

Kolton M, Erlacher A, Berg G, Cytryn E. 2016. The Flavobacterium genus in the plant holobiont: ecological, physiological, and applicative insights. In: Castro-Sowinski S, ed. Microbial Models: From Environmental to Industrial Sustainability. Singapore: Springer. 189-207

Lee HR, Han SI, Rhee KH, Whang KS. 2013. Mucilaginibacter herbaticus sp. nov., isolated from the rhizosphere of the medicinal plant Angelica sinensis. International Journal of Systematic and Evolutionary Microbiology 63:2787-2793

Madhaiyan M, Poonguzhali S, Lee JS, Senthilkumar M, Lee KC, Sundaram S. 2010. Mucilaginibacter gossypii sp. nov. and Mucilaginibacter gossypiicola sp. nov., plant-growth-promoting bacteria isolated from cotton rhizosphere soils. International Journal of Systematic and Evolutionary Microbiology 60(10):2451-2457

Mannisto MK, Tiirola M, McConnell J, Haggblom MM. 2010. Mucilaginibacter frigoritolerans sp. nov., Mucilaginibacter lappiensis sp. nov. and Mucilaginibacter mallensis sp. nov., isolated from soil and lichen samples. International Journal of Systematic and Evolutionary Microbiology 60(12):2849-2856

Mano H, Tanaka F, Watanabe A, Kaga H, Okunishi S, Morisaki H. 2006. Culturable surface and endophytic bacterial flora of the maturing seeds of rice plants (Oryza sativa) cultivated in a paddy field. Microbes and Environments 21(2):86-100

Meyer WA, Torres MS, White JF. 2012. Biology and applications of fungal endophytes in turfgrasses. In: Stier J, Horgan B, Bonos S, eds. Turfgrass: Biology, Use and Management. Madison: American Society of Agronomy. 713-731

Ofek M, Hadar Y, Minz D. 2012. Ecology of root colonizing Massilia (Oxalobacteraceae) PLOS ONE 7(7):e40117

Okunishi S, Sako K, Mano H, Imamura A, Morisaki H. 2005. Bacterial flora of endophytes in the maturing seed of cultivated rice (Oryza sativa) Microbes and Environments 20(3):168-177

Oteino N, Lally RD, Kiwanuka S, Lloyd A, Ryan D, Germaine KJ, Dowling DN. 2015. Plant growth promotion induced by phosphate solubilizing endophytic Pseudomonas isolates. Frontiers in Microbiology 6:745

Pace NR, Stahl DA, Lane DJ, Olsen GJ. 1986. The analysis of natural microbial populations by ribosomal RNA sequences. In: Marshall KC, ed. Advances in Microbial Ecology. Berlin: Springer. 1-55

Pedrini S, Merritt DJ, Stevens J, Dixon K. 2017. Seed coating: science or marketing spin? Trends in Plant Science 22(2):106-116

Price MN, Dehal PS, Arkin AP. 2010. FastTree 2–approximately maximum-likelihood trees for large alignments. PLOS ONE 5(3):e9490

Prieto P, Mercado-Blanco J. 2008. Endophytic colonization of olive roots by the biocontrol strain Pseudomonas fluorescens PICF7. FEMS Microbiology Ecology 64(2):297-306

Riesenfeld CS, Schloss PD, Handelsman J. 2004. Metagenomics: genomic analysis of microbial communities. Annual Review of Genetics 38(1):525-552

Rognes T, Flouri T, Nichols B, Quince C, Mahé F. 2016. VSEARCH: a versatile open source tool for metagenomics. PeerJ 4(17):e2584

Ruiz D, Agaras B, De Werra P, Wall LG, Valverde C. 2011. Characterization and screening of plant probiotic traits of bacteria isolated from rice seeds cultivated in Argentina. Journal of Microbiology 49(6):902-912

Rybakova D, Cernava T, Köberl M, Liebminger S, Etemadi M, Berg G. 2015. Endophytes-assisted biocontrol: novel insights in ecology and the mode of action of Paenibacillus. Plant and Soil 405(1–2):125-140

Shaik SP, Thomas P. 2019. In vitro activation of seed-transmitted cultivation-recalcitrant endophytic bacteria in tomato and host-endophyte mutualism. Microorganisms 7(5):132

Soltani A-A, Khavazi K, Asadi-Rahmani H, Omidvari M, Abaszadeh Dahaji P, Mirhoseyni H. 2010. Plant growth promoting characteristics in some Flavobacterium spp. isolated from soils of Iran. Journal of Agricultural Science 2(4):106

Somova L, Pechurkin N, Sarangova A, Pisman T. 2001. Effect of bacterial population density on germination wheat seeds and dynamics of simple artificial ecosystems. Advances in Space Research 27(9):1611-1615

Steven B, Lionard M, Kuske CR, Vincent WF. 2013. High bacterial diversity of biological soil crusts in water tracks over permafrost in the high arctic polar desert. PLOS ONE 8(8):e71489

Storch D, Bohdalková E, Okie J. 2018. The more-individuals hypothesis revisited: the role of community abundance in species richness regulation and the productivity-diversity relationship. Ecology Letters 21(6):920-937

Suzuki S, He Y, Oyaizu H. 2003. Indole-3-acetic acid production in Pseudomonas fluorescens HP72 and its association with suppression of creeping bentgrass brown patch. Current Microbiology 47(2):138-143

Verma P. 2019. Seed endophytes in crop plants: metagenomic approaches to study the functional roles and interactions. In: Verma SK, White JF, eds. Seed Endophytes. Berlin: Springer. 483-507

Verma S, Kingsley K, Irizarry I, Bergen M, Kharwar R, White JF. 2017. Seed vectored endophytic bacteria modulate development of rice seedlings. Journal of Applied Microbiology 122(6):1680-1691

Verma SK, Kingsley K, Bergen M, English C, Elmore M, Kharwar RN, White JF. 2018. Bacterial endophytes from rice cut grass (Leersia oryzoides L.) increase growth, promote root gravitropic response, stimulate root hair formation, and protect rice seedlings from disease. Plant and Soil 422(1–2):223-238

White JF. 1987. Widespread distribution of endophytes in the Poaceae. Plant Disease 71(4):340-342

White JF, Kingsley KL, Butterworth S, Brindisi L, Gatei JW, Elmore MT, Verma SK, Yao X, Kowalski KP. 2019. Seed-vectored microbes: their roles in improving seedling fitness and competitor plant suppression. In: Verma SK, White JF, eds. Seed Endophytes. Berlin: Springer. 3-20

White JF, Kingsley KL, Verma SK, Kowalski KP. 2018. Rhizophagy cycle: an oxidative process in plants for nutrient extraction from symbiotic microbes. Microorganisms 6(3):95

White JF, Torres MS, Johnson H, Irizarry I, Tadych M. 2014. A functional view of plant microbiomes: endosymbiotic systems that enhance plant growth and survival. In: Verma VC, Gange AC, eds. Advances in Endophytic Research. Berlin: Springer. 425-439

Yanni YG, Rizk R, Corich V, Squartini A, Ninke K, Philip-Hollingsworth S, Orgambide G, De Bruijn F, Stoltzfus J, Buckley D. 1997. Natural endophytic association between Rhizobium leguminosarum bv. trifolii and rice roots and assessment of its potential to promote rice growth. Plant and Soil 194(1–2):99-114

Zeglin LH, Dahm CN, Barrett JE, Gooseff MN, Fitpatrick SK, Takacs-Vesbach CD. 2011. Bacterial community structure along moisture gradients in the parafluvial sediments of two ephemeral desert streams. Microbial Ecology 61(3):543-556

Zhu YL, She XP, Wang JS, Lv HY. 2017. Endophytic bacterial effects on seed germination and mobilization of reserves in Ammodendron biofolium. Pakistan Journal of Botany 49:2029-2035