Introduction

Peanut (Arachis hypogaea L.) is a vital oilseed crop. Oil extracted from peanut seeds is an important constituent of food oil supply in China. Endophytic and epiphytic microbiota, such as commensal, synergistic, and potentially pathogenic microbes, is present in crop seeds (Nicolaisen et al. 2014), which is related to seed storage, plant growth, and development (Links et al. 2014). Evidence presented that Aspergillus sp. producing carcinogenic mycotoxins was identified in peanut as symptomless endophytes, which suggests the potential for concern as pathogens and as food safety hazard (Palencia et al. 2010). However, further research suggested that mature peanut seeds from the experimental and control plants contained several species of nonpathogenic endophytic bacteria, and the Bacillus species in seeds demonstrated activity against Aspergillus flavus (Sobolev et al. 2013). So, the correlation between bacteria and fungi in peanut seeds should be further studied.

The peanut plant accommodates only selected bacterial species from diverse soil populations (Sobolev et al. 2013). Seed bacterization with phylloplane isolates promoted peanut growth indicating the possibility of isolating beneficial rhizosphere bacteria from different habitats (Kishore et al. 2005). The selected bacteria could potentially be utilized for the biocontrol of toxicogenic fungi (Wang et al. 2012). So, beneficial endophytic microbes promote peanut growth and control toxicogenic fungi could be isolated from soils (Ibañez et al. 2009, 2014; Taurian et al. 2010; Fabra et al. 2010). Agricultural environment may provide a way to reduce the toxigenic fungal contamination and mycotoxins in peanuts (Goncalez et al. 2008). However, the predominance of Aspergillus in hulls and kernels has been identified and 93.8% of A. flavus strains were producers of aflatoxin B1 and B2 (Nakai et al. 2008). Aflatoxin is commonly found presenting deeply in individual kernels and the aflatoxin contents of different portions of peanut kernels have been determined (Goldblatt 1971). So, the endophytic microbiota transmitted from seeds to seedlings would be helpful for control toxicogenic fungi and reduce the mycotoxin contents.

During seed germination, external environmental conditions affect dormant seeds and initiate dormancy-breaking signals. As seeds set out to germinate, seedling microbial communities begin to form gradually based on seed endophytes (Johnston-Monje and Raizada 2011). The transmission can choose beneficial endosymbionts, exclude pathogenic microbes, and benefit host plants by delivering beneficial endosymbionts to the next generation (Truyens et al. 2015). The peanut seeds consisted of germs and cotyledons; however, the difference of bacterial and fungal flora between peanut germs and cotyledons is unknown. Whether the flora in sprouts is derived from germs or cotyledons is still unknown. Next-generation sequencing is a robust approach that could be employed to decipher microbiome of plants, and the microorganisms that identified by microbiome analysis could be exploited to enhance host plant health and production (Mendes et al. 2013).

Herein, our objective was to identify the endophyte taxa of peanut seeds that can be transmitted to sprouts and their effects on sprout growth. Specifically, we isolated total DNA from seed germs, cotyledons, and sprouts of peanuts, and employed Illumina-based sequencing to analyze the conserved genomic regions of bacteria and fungi. Then, we compared the bacterial and fungal communities, identified the transmitted microbial taxa from seeds, and analyzed their effects on sprouts.

Material and methods

Seed germination

To surface disinfection, the peanut (Arachis hypogaea L., n = 5) seeds were immersed in 75% (v/v) ethanol for 5 min, sodium hypochlorite solution containing 5% active chlorine for 5 min, and then were rinsed with sterile distilled water for three times. The surface-disinfected seeds were germinated on sterilized moistened filer paper (Xinhua, Hangzhou, China) at 25 °C. Four days later, the genomic DNA was isolated from the processed sprouts. The surface-disinfected seed without germination were divided into germ and cotyledon parts for DNA isolation. The assay of each group (sprout, germ, and cotyledon) of samples was repeated three times.

DNA extraction

The genomic DNA was isolated from the peanut seed germs (designated as Pgerm), cotyledons (designated as Pcotyledon), and sprouts (designated as Psprout) with E.Z.N.A. HP Plant Miniprep DNA Kit (Omega Bio-Tek). The quality of genomic DNA was analyzed by 1% agarose gel electrophoresis.

Conservative region amplification and Illumina MiSeq sequencing

The 16S ribosomal RNA (rRNA) gene fragments and the ITS1 regions of fungal rRNA gene were amplified and amplicons were sequenced according to the procedure (Huang et al. 2016). Briefly, the triplicate PCR reactions for each sample were combined into a single volume and then purified using GeneJET Gel Extraction Kit (Thermo Scientific). The sequencing libraries were constructed according to the standard protocol for NEB Next® Ultra™ DNA Library Prep Kit (New England Biolabs). Sequencing was carried by Guangzhou Magigen Biotechnology Co., Ltd.

Data preprocessing and diversity analyses

Sequence processing, analysis, and community comparisons were performed as previously described (Huang et al. 2016). The sequencing data can be available in the NCBI Sequence Read Archive (SRA) database given the accession number SRP055979.

Isolation of Halothiobacillus from sprouts

The peanut sprouts under aseptic conditions were clip into about 0.2 × 0.2 cm pieces. Then, the pieces were put into mineral medium (MM) (Park et al. 2008) and incubated at 28 °C (Shi et al. 2011). Colonies formed on plates were transferred on mineral medium with 2 M NaCl (selective medium for Halothiobacillus) to obtain the pure culture (Sorokin et al. 2006).

Pot experiments

Soil with texture of sandy loam, which contains 70% sand, 15% silt, and 15% clay, was obtained from the suburbs of Guangzhou, China. The bacterial strains grown in MM medium were harvested by centrifugation, washed twice, and suspended in saline (0.85% NaCl) solution. Meanwhile, the peanut seeds were sterilized in NaClO solution, which contains 6.5% active chlorine, for 30 min and rinsed three times with sterile distilled water. The seeds were inoculated by soaking in bacterial suspension of 107 cfu ml−1 for 2 h. The seeds soaked in sterile water were used as a negative control. All seeds were sowed in the test soils amended with sulfur (50 mg kg−1 of soil). Ten seeds were sowed in each pot and each treatment contained three pots. The soil moisture was maintained at the saturated condition. The plants were grown at 25 °C and a 16/8 h day/night regime in a glasshouse. The shoot length (cm), shoot fresh weight (g), and root fresh weight (g) were evaluated 28 days after germination (Anandham et al. 2014). Each treatment contains three pots of seedlings.

Statistical analysis

For unweighted and weighted Unifrac distances were calculated by QIIME (V1.7.0) (Kõljalg et al. 2013). KRONA was used to visualize the phylogenetic relationship of different microbial taxa (Ondov et al. 2011). Estimates of species richness and relative diversity level were provided by Shannon, Simpson, Chao1, ACE, and coverage indices (Kemp and Aller 2004). Statistical analysis was performed by previous methods (Huang et al. 2016).

Results

Overview of fungal and bacterial community diversity

After low quality, chimeric reads were filtered out, 260,308 bacterial sequences were generated with an average length of 406 bp, and 183,204 fungal sequences were generated with an average length of 272 bp (Table 1). Based on their shared sequence similarity at a 97% threshold, sequences of each sample were clustered into an average of 66 bacterial OTUs ranging from 29 to 102 OTUs, and an average of 26 fungal OTUs ranging from 18 to 31 OTUs with coverage more than 0.99. The alpha diversity indices indicated that the more diverse bacterial OTUs were detected in the germs and cotyledons than those in sprouts; nevertheless, the sprouts contain similar diverse fungi with germs and cotyledons (Table 2).

Table 1 Summary of the features of valid tags of the samples of cotyledons (Pcotyledon), germs (Pgerm) of peanut seeds, and peanut sprouts (Psprout)
Table 2 Summary of species richness estimators of the samples of cotyledons (Pcotyledon), germs (Pgerm) of peanut seeds, and peanut sprouts (Psprout)

Although the germs and cotyledons were divided from surface sterilized seeds, and the sprouts are segregated from environmental bacteria and fungi during germination, diverse endophytes in these samples were detected. A total of 102 bacterial OTUs were obtained in cotyledons and 29 OTUs in germs, and 4 bacterial OTUs coexisted in germs and cotyledons simultaneously. With the germination, 14 bacterial OTUs of cotyledons were still found in sprouts, and 6 OTUs of germs were found in sprouts. Moreover, 4 OTUs were found in germs, cotyledons, and sprouts simultaneously (Fig. 1). Unlike endophytic bacteria, the sprouts and germs contained more fungal OTU than cotyledons, and 20 OTUs were found in germs and sprouts simultaneously; 17 OTUs of cotyledons were still detected in sprouts (Fig. 2), and 15 OTUs were detected in sprouts, germs, and cotyledons simultaneously.

Fig. 1
figure 1

Venn diagram illustration of the variation in bacterial OTUs of cotyledons (Pcotyledon), germs (Pgerm) of peanut seeds, and peanut sprouts (Psprout)

Fig. 2
figure 2

Venn diagram illustration of the variation in fungal OTUs of cotyledons (Pcotyledon), germs (Pgerm) of peanut seeds, and peanut sprouts (Psprout)

The community structure and composition of fungi and bacteria

Representative bacterial sequences of each OTU were assigned into the domains Bacteria (99.99% of the total data set), of which Halothiobacillus dominated the overall bacterial genera (31.0%) (Fig. 3). The Synechococcus were the second most dominant (5.2%). Representative fungal sequences of each OTU were assigned into the domains Fungi (100% of the total data set), of which the Rhizopus dominated the overall fungal genera (0.4%) (Fig. 4). Penicillium was the second most dominant (0.1%). The dominant fungal and bacterial taxa in three peanut samples were different from each other. Halothiobacillus, Synechococcus, Paracoccus, Agrobacterium, and Gallionella were the most abundant bacterial genera in cotyledons. The dominant bacteria in germs were Synechococcus, Halothiobacillus, Mycobacterium, and Rhodococcus among all the bacterial genera. Halothiobacillus, Synechococcus, Burkholderia, and Paracoccus were found to be the most abundant bacterial genera in sprouts. With the germination, the dominant bacterial genera Halothiobacillus and Synechococcus were still the most detected genera in spouts. Other bacterial genera Erwinia, Hyphomonas, and Devosia were detected in cotyledons and sprouts simultaneously.

Fig. 3
figure 3

Phylogenetic diagram of bacterial taxa OTUs of cotyledons (Pcotyledon), germs (Pgerm) of peanut seeds, and peanut sprouts (Psprout)

Fig. 4
figure 4

Phylogenetic diagram of fungal taxa OTUs of cotyledons (Pcotyledon), germs (Pgerm) of peanut seeds, and peanut sprouts (Psprout)

For the dominant fungal genera in the peanut plants, Rhizopus, Emericella, and Penicillium were found the most in all samples (Fig. 4). Humicola, Emericella, and Penicillium were the dominant fungal genera in germs, and cotyledons contained Emericella as the dominant genus. The dominant fungal genera of germs, such as Humicola, Emericella, and Penicillium, were still detected in sprouts. With the germination, the sprouts obtained the Rhizopus as dominant fungal genera. The Emericella was detected in sprouts, germs, and cotyledons, simultaneously, whereas the Emericella detected in cotyledons were at low abundance.

Pot experiments

A pure culture of Halothiobacillus was obtained and designated as SOB4. The pot experiments demonstrated the stimulation effects of sulfur oxidation bacteria strain SOB4 in peanut sprouts. Compared with the negative controls, the growth of peanut seedlings inoculated with strain SOB4 was facilitated (P < 0.05). Root fresh weight, shoot length, and shoot fresh weight were promoted by 73, 35, and 84%, respectively (Table 3).

Table 3 The growth promotion effects of Halothiobacillus SOB4 on peanut seedlings inoculated

Discussion

Peanut is a widespread leguminous plant of great agricultural and economic significance, and nodulation by native bacteria is usually assumed to be adequate (Taurian et al. 2010; Angelini et al. 2011). The rhizosphere and endophytic bacteria that enhanced plant yield had been isolated and detected in the previous studies (Ibañez et al. 2009; Haldar et al. 2011). However, the bacterial flora transmitted from peanut seed was not reported. Our results suggested that the more diverse bacteria were detected in germs and cotyledons than those in sprouts; nevertheless, the sprouts contain more diverse fungi than germs and cotyledons. The seed’s physiological state might influence the microbial community during the germination process. In the study, the surface-disinfected seeds were germinated on sterilized moistened filter paper, the sprouts did not obtain bacteria and fungi from environments, and the assorted bacteria and fungi in sprouts might come from germs and cotyledons. In the study, 4 bacterial OTUs and 15 fungal OTUs were found in germs, cotyledons, and sprouts simultaneously.

With the germination, the dominant bacterial genera Halothiobacillus and Synechococcus were still the most detected genera in spouts. Other bacterial genera Erwinia, Hyphomonas, and Devosia were detected in cotyledons and sprouts simultaneously. However, the bacterial genera except for Hyphomonas were not detected in peanut rhizosphere soils (Haldar et al. 2011). Some portions of endophytes might transmit from seeds, not from rhizosphere soils. The dominant fungal genera of germs, such as Humicola, Emericella, and Penicillium, were still detected in sprouts, and the Emericella was detected in sprouts, germs, and cotyledons, simultaneously. The Humicola and Emericella were still isolated from symptomless leaves of peanut (Suryanarayanan and Murali 2006), and the fungi might transmit from seeds because the leaves did have not contact with soils.

Co-inoculation of sulfur-oxidizing bacteria and Rhizobium could increase the dry weight, nodule number, plant biomass, and pod yield of peanut (Anandham et al. 2007). The fungal genera Humicola and Emericella in roots could solubilize soil phosphate and promote plant growth under salt stress or control plant disease (Sibounnavong et al. 2010). The Halothiobacillus could oxidize thiosulfate and solubilize soil phosphate, and are effective early plant growth-promoting rhizobacteria (Anandham et al. 2014). The Halothiobacillus as dominant bacterial genera was detected in germs, cotyledons, and sprouts, so the peanut plants contain core Halothiobacillus. In the study, the pot experiments showed that the isolated Halothiobacillus stain could promote the growth of peanut seedlings.

The previous studies have reported seed endophytes for many plants. However, still little is known on the persistence or existence of seed-borne endophytes across germination (Ferreira et al. 2008). It appeared that peanut has core mycobiome and bacteriome that is conserved across the germination from cotyledons and germ to sprouts and can be important sources of endophytes that colonize the mature plant. Based on the endophytic bacterial and fungal community in peanut cotyledons, germs, and sprouts, Halothiobacillus strain with growth promotion was isolated. Some novel species should be isolated by different methods. The procedure for beneficial endophytic strains is more feasible than traditional isolation and selection in greenhouse (Huang et al. 2016). The Illumina-based sequencing is suitable for rational isolation of beneficial endophytic strains to promote plant growth.