{ label (or @symbol) needed for fn }

Study site

Serapias vomeracea is distributed in the coastal area of the Black Sea and Aegean regions of Turkey. The research area of this study is Ondokuz Mayıs University campus area. The habitats of the species are the open spaces located next to oak forests at about 50 m a.s.l. The species of the genera Hordeum, Avena and Melilotus are present in these habitats and over 100 S. vomeracea individuals are also represented in the research area. They bloom in May and the seeds maturate in July. In 2015, seasonal avarage temperatures and amount of precipitation were 12.2 °C and 176 mm in spring, 23.9 °C and 85 mm in summer, 16.8 °C and 134 mm in autumn. In 2016, they were: 13.8 °C and 171 mm in spring, 24.8 °C and 69 mm in summer, and 15.1 °C and 133 mm in autumn.

Physical and chemical properties of the soil

For soil analyses, an about 2 kg sample was taken from 0-15 cm depth of 1 m2 area after removing leaves and other debris from the surface of soil in research area where S. vomeracea is commonly distributed. The soil was dried at room temperature in laboratory and used for all analyses.

In the soil saturated with water, a direct pH reading was taken with a glass electrode pH meter (Aciego Pietri and Brookes 2008). Electrical conductivity was determined with an EC – meter in saturation extract (Rhoades 1996), organic matter using the Smith-Weldon method (Nelson and Sommers 1982). Total sand, silt, lime and clay were determined by the densimeter method (Bowman and Hutka 2002). Exchangeable cations (calcium, magnesium, and potassium) were determined with the Mehlich-3 method (Mehlich 1984). Available phosphorus was analysed using the molybdate blue method (Murphy and Ridley 1962).

Fungus isolation

Before root samples were collected, S. vomeracea phenological stages were followed for one year in their habitats (Campus area of Ondokuz Mayis University). It was determined that the leaves formed in February to March and the roots completely dried in July.

The roots for fungi isolation were taken through two years (2015-2016). Every month, complete roots of a plant were collected. Cross-sections were then taken from the roots and examined under a microscope for the presence of fungi. All the roots containing fungal coils were used for fungi isolation, performed according to Clements et al. (1986). The roots were sterilized in 1.5% NaOCl solution for 5 minutes and washed in sterile distilled water. Under aseptic conditions, root pieces (1-2 cm) were placed in petri dishes containing fungi isolation medium (FIM) (Clements et al. 1986) and incubated at 27 °C for 2 days in the dark. Following stereo microscopy examination, any fungal colonies were transferred to FIM and purified. Pure fungus cultures were stored at 4 °C.

Morphological and molecular identification of the fungi

Isolates for preliminary identification were grown in FIM plates and thin agar blocks of 48-h-old cultures were viewed under a microscope for the mycelial and branching characteristics of the isolates. The following distinct morphological characters were observed: mycelia branched at acute to right angles, constrictions at or near the point of branching and septum formation near the branching point. Isolates showing rhizomorph and asexual spore (conidia) structure were defined as non–Rhizoctonia. The isolates were grown in potato dextrose agar (PDA) at 25 ± 2 °C for 7-8 days to determine the color and texture of the colonies. The color of the colony was defined according to the color chart of the Royal Horticultural Society of London. The hypha diameter of the fungi was measured under light microscope. To determine the number of nuclei in the cell, isolates were grown for three days at 25 °C in Petri dishes containing water-agar (WA), stained with Safranine-O solution and nuclei were counted under the microscope (Bandoni 1979).

DNA isolation from fungal mycelia was performed using the CTAB (cetyltrimethyl ammonium bromide) method described by Pascual et al. (2000) with some modifications (e.g., 50 mg of mycelium was crushed in a sterile mortar in 1 mL extraction buffer and incubated for 30 min.) For each isolate, the internal transcribed spacer (ITS) region of the nuclear ribosomal rDNA was amplified by PCR. ITS1 and ITS2 regions, including the ribosomal 5.8S RNA gene, were amplified using the universal primers ITS-1 (5’- TCCGTAGGTGAACCTGCGG-3’) and ITS-4 (5′ TCCTCCGCTTATTGATATGC 3′) (White et al. 1990). PCR amplification reactions were performed in a 50 μl reaction containing 1 μl genomic DNA (1 ng μl-1), 1 μl (2.5 mM) dNTP mix (Sigma), 0.25 μl Taq DNA polymerase (5 U/μl) (Promega, Go-TaqFlexi DNA Polymerase), 1 μl each of primers (25 pmoles), 10 μl 5 × PCR buffer supplied by manufacturer (Promega, Go-Taq Green Buffer) and 3 μl MgCl2 (1.5 mM) (Sigma) and 32.75 μl sterile ddH20. PCR amplification was carried out as follows: an initial denaturation at 94 °C for 3 min followed by 30 cycles of 94 °C for 1 min, 49 °C for 2 min, 72 °C for 3 min and a final extension at 72 °C for 7 min (Salazar et al 1999).

PCR products of the rDNA-ITS region were sequenced by Macrogen (Macrogen Inc., Seoul, Republic of Korea) by using ABI 3730 XL DNA sequencer with ITS1 and ITS4 primers. For each PCR product, sequences of both strands were combined to generate a consensus sequence by using BioEdit version 7.2.5 software (Hall 1999). The consensus sequences of the rDNA-ITS region were compared with the sequence data in GenBank (National Center for Biotecnology Information) by using BLASTn tool. We used 97 to 100% sequence identity to delimit fungal species and genera. One sequence for each isolate was deposited in the GenBank database. DNA polymorphism was determined for the ITS gene sequences using DNA Sequence Polymorphism software (DNASP), version 6.0 (Rozas et al. 2017). Haplotype groups were formed according to the nucleotide polymorphism of the sequences. The data set formed for Tulasnella species including the sequence of 15 fungal isolates and the 24 reference sequences obtained from NCBI.

Genetic distances for data sets were calculated using the MEGA 6 software package (Tamura et al. 2013). The tree (Rhizoctonia-like) was constructed using Maximum Likelihood (ML) analysis with 1000 bootstrap replications using MEGA 6.

Collection of the seeds and seed viability test

Capsules produced by natural pollination were collected from the individuals grown in the meadows next to oak forests in 2015. The seeds were removed from the capsules, kept in room conditions for a few days in the laboratory to lose moisture, and then stored at 4 °C. Pre-treatments for viability test and TTC (2,3,5 triphenyl tetrazolium chloride) test were conducted according to Kömpe et al. (2020) and Van Waes and Deberg (1986), respectively. Approximately 100-150 seeds were placed in each pack. The seeds were incubated in moist cocopeat for 7 days so that the seed coat could crack. Thus, TTC would be allowed to enter into the embryo. The seeds were incubated in TTC (1%) solution for 12 hours at 28 °C, followed by 12 h incubation in sterile distilled water. The viability experiments were performed with six repetitions. The seeds with red-pink embryos were evaluated as live. Viability rates (%) = number of stained embryos /total seed number × 100.

Germination of the seeds and development of the seedlings under ex vitro condition

The soil samples were taken from 3-5 cm depth and 30 cm around the adult orchids individuals. A mixture of 2:1 soil: perlite was prepared and sterilized in autoclave at 121°C for 20 min. The sterile soil mixture was filled into pots (20 x 31 x 13 cm) to a depth of 15 cm. The seeds were placed between sheets of water-resistant nylon mesh (45 µm pore size). Approximately 400-500 seeds (10 mg) were placed in each pack. Ten packs of seeds were placed in each pot. Fifteen Rhizoctonia-like (Tulasnella) and four non-Rhizoctonia isolates (Fusarium, Aspergillus, Talaromyces) were used in germination tests. For each fungal isolate, two independent pots containing sterile soil mixture were prepared. Fungi were grown on FIM for 7 days and fungus discs (1-2 mm in diameter) obtained from then were placed in each corner of the pots. Seed packs were placed in control pots not inoculated with fungus. Ex vitro germination and growth experiments were performed with six repetitions. The pots were incubated at 25 ± 2 °C in the climate chamber with a 16/8h light/dark photoperiod and 33 PAR (photosynthetic active radiation). It was irrigated with sterile distilled water once a week. Three months after the seeds were embedded in the pots, 3 seed packs were randomly selected from each pot to calculate germination and seedling growth rates and evaluated according to the scale of Clements et al. (1986).

The extent of seed germination and development was divided into stages: S0, S1, S2, S3 and S4. These stages are represented as follows: S0: No germination (seed); S1: Protocorm; S2: Leaf premordium; S3: The first photosynthetic leaf; S4: Seedling with advanced leaves (and/or rooted).

Germination percentage was calculated as number of seeds in stages 1-4 divided by the total number of seeds (Clements et al. 1986). Germination rates (%) = number of germinated seeds /total seed number X 100.

Microscopic observations of mycorrhizal associations

For showing that mycorrhizal association had been established, transverse sections were taken from experimental protocorms produced ex vitro, from the roots of experimental seedling and from the roots of adult plants in natural habitat, and 0.1% lactophenol (RAL Diagnostic) was dropped and incubated for 10 min at 25 °C. The sections were then washed with distilled water to remove residual staining. Each section was examined under a microscope for presence of fungal coil and photographed.

Statistical analysis

For showing that mycorrhizal association had been established, transverse sections were taken from experimental protocorms produced ex vitro, from the roots of experimental seedling and from the roots of adult plants in natural habitat, and 0.1% lactophenol (RAL Diagnostic) was dropped and incubated for 10 min at 25 °C. The sections were then washed with distilled water to remove residual staining. Each section was examined under a microscope for presence of fungal coil and photographed.

The fungi of Serapias vomeracea

In the first year of the experiment, the phenological stages continued from March to July. In the second year they took place between February and June. Fungal isolations were made from March to July in 2015 and from February to June in 2016. From the roots of S. vomeracaea 18 isolates (Svl 1-18) were found in the first year and 16 in the second year (Svl 19-34). Applying the pre-identification procedure, two main isolate groups were determined: binucleate Rhizoctonia-like group with 30 isolates and multinucleate non-Rhizoctonia group with 4 isolates. Rhizoctonia-like (Tulasnella sp.) isolates were found every month when fungal isolation was carried out. Among non-Rhizoctonia isolates, two isolates of Fusarium sp. were isolated in March 2015, one isolate of Aspergillus sp. in June 2015 and one isolate of Talaromyces sp. in February 2016 (Fig. 1). Morphological features of the isolated fungi in both groups (hyphae diameter, number of nuclei, colony color, colony appearance) were determined. Three different colony colors of Rhizoctonia-like isolates were determined after growth on PDA medium at 25 ± 2 °C in the dark for two weeks. The most frequently observed colony color was grayish yellow, while orange-white and yellowish white were observed less frequently. Colony appearance was frequently submerged and powdery. Vegetative hyphae color of all the isolates was transparent. Hyphae diameter varied between 2.5 and 4.4 µm. Vegetative hyphae color of all isolates was transparent. Colony color of the non-Rhizoctonia isolates were determined as red and orange (Svl 10, Svl 11, respectively), white (Svl 13), grayed yellow (Svl 20), while the colony appearance was determined as submerged aerial hyphae and powdery. Vegetative hyphae color of all isolates was transparent. Hyphae diameter varied between 2.5 and 4.4 µm.

Fig. 1 Numbers of endophytic root fungi of Serapias vomeracea during the isolation months at 2015 and 2016. Rhizoctonia-like (Tulasnella sp.) isolates were obtained every month when fungal isolation was carried out. Among non-Rhizoctonia isolates, two isolates of Fusarium sp. were isolated in March 2015, one isolate of Aspergillus sp. in June 2015 and one isolate of Talaromyces sp. in February 2016.

The PCR products of the ITS regions of all the fungal isolates were sequenced and aligned. Obtained consensus sequences were uploaded to GenBank and compared with other sequences from that database. Accession numbers for our fungal ITS regions and identity percentages with most closely related sequences from GenBank are given inTab. 1. According to BLASTn results, 2 of the 4 non-Rhizoctonia isolates were found to show 100% identity to Fusarium tritinctum (Svl 10 – Svl 11) (GenBank acession number MK250655, MK250515 respectively), while one isolate was found to show 99% identity to Aspergillus spelaeus (Svl 13) (GenBank accession number MK2505615) and one isolate was found to show 99% identity to Talaromyces pinophilus (Svl 20) (GenBank acession number MK255324). All of the Rhizoctonia-like isolates had identity rates to uncultured Tulasnella (e.g. Svl 1, Svl 3, Svl 4 etc.) of between 97-100% (GenBank acession number MK249887, MK250064, MK250062, respectively) (Tab. 1). A phylogenetic tree was constracted to reveal the phylogenetic relationship between Tulasnella isolates of S. vomeracea and Tulasnella sp. from other orchids (Fig. 2).

Among the ITS sequence data set, 15 haplotypes were detected based on sequence analysis of 30 Tulasnella isolates. ITS sequences exhibited a haplotype diversity of 87%. Haplotype 5 exhibited the highest frequency within the ITS sequences, being generated from 12 sequences. Haplotype 4 and haplotype 7 exhibited low frequencies within the ITS sequences, and were generated respectively from 4 and 2 sequences. Other haplotypes exhibited the lowest frequencies among the ITS sequences, and were generated from only one sequence each. Tulasnella ITS sequences from our study grouped with T. calospora clade. In the phylogenetic tree, Tulasnella isolates (e.g. Svl 3, Svl4, Svl5, Svl9 etc.) isolated from S. vomeracea roots were therefore found to be closely associated with T. calospora. (GenBank accession no: AY373298.1, FJ613176.1, GU166421.1, HQ889722.1 etc.) (Fig. 2).

Fig. 2 Maximum-Likelihood (ML) tree based on an alignment of ITS-5.8 sequences, showing relationship of fifteen Tulasnella species. Node tips show NCBI accession numbers followed by fungal species name, host species and country name. The tree was rooted with the sequence of Xylaria polymorpha (accesion number EU272539.1) Numbers above branches are maximum likelihood bootstrap probalities (>50%).

The seed viability test

The embryos of seeds incubated in cocopeat for 1, 2 and 3 days were not stained. The viability rates of the seeds incubated for 5 days and 7 days were found to be 40.00% ± 9.27 and 90.32% ± 1.30, respectively.

Ex vitro symbiotic germination and symbiotic association with the seeds

Ex vitro germination tests were evaluated for total germination and development stages after three months of incubation. Developmental stages (S1-4) are shown inFig. 3A. Fusarium tricinctum (Svl 10, Svl 11), Aspergillus spelaeus (Svl 13) and Talaromyces (Svl 20) isolates did not promote germination.

Fig. 3 From the seed to the seedlings of Serapias vomeracea. A – developmental stages (S1-4). Arrows and numbers show the developmental stages. B – control (no fungal isolate). Seed coats ruptured but, no advanced development (S0). C – protocorms and leaf primordiums, D – leafy plantlet, E – the seedling in natural area, F – the first tuberous adult plant. Scale bars: A – 10 mm, B – 0.3 mm, C – 0.2 mm, D – 10 mm, E – 10 mm, F – 10 mm.

There was no germination in the control pots without fungi (Fig. 3B). All the Tulasnella isolates on the phylogenetic tree promoted germination and growth at varying rates (Tab. 2,Fig. 3C-E). According to the counts performed at the end of three months, 98% of seeds germinated in the packages inoculated with the isolate Svl 21 and the seeds inoculated with this isolate developed until to S3. The seeds placed in the pots inoculated with Svl 4, Svl 14 and Svl 34 germinated at the rates of 93.2%, 94% and 90% respectively, and these isolates supported the development of the seedling up to the S4 (13.67%, 10.8%, 9.8%, respectively) (Fig. 3D, E). The percentage of seedlings that reached the S4 of development (13.67%) inoculated with the isolate Svl 4 was found to be statistically significantly high when compared with seedlings reaching the same stage of development inoculated with other fungi.

Thirty seedlings grown in pots (S4- seedlings with advanced leaves (and/or rooted)) were planted in their natural habitats and the first real tubers occurred at 5 months after the seedlings were transferred to the soil. All of the seedlings transferred to soil developed tubers (Fig. 3F).

The presence of symbiotic association was shown in cross-sections from the roots of adult plants, the ex vitro experimental protocorms and the ex vitro experimental seedling roots (Fig. 4A, B, C, respectively).

Fig. 4 Fungal pelotons stained with lactophenol cotton blue. A – in the cortical cells of adult Serapias vomeracea roots, B – in the ex vitro protocorm cells, C – in the cortical cells of the roots of ex vitro seedlings. The arrows indicate fungal coils. Scale bars: 50 µm.