2023 Volume 64 Issue 2 Pages 55-62
Gentiana zollingeri (Gentianaceae) is an initial mycoheterotrophic plant that depends on a specific group of arbuscular mycorrhizal (AM) fungi for carbon source during underground growth after seed germination. In this study, a mycorrhizal fungus dominant in mycoheterotrophic seedlings of G. zollingeri was successfully isolated from a soil core collected from a point close to a flowering G. zollingeri. The AM fungal isolate was identified as conspecific or closely related to Dominikia aurea (Glomeraceae) by spore morphology and molecular phylogeny. Basic Local Alignment Search Tool (BLAST) searches against the MaarjAM database showed that the nuclear small subunit ribosomal DNA sequences of the isolate matched the AM fungal sequences obtained from a wide range of plants in various ecosystems, including several mycoheterotrophs. Thus, it is suggested that the AM fungal isolate is one of the cheating susceptible AM fungi. Furthermore, the sequences corresponded to those of a group of AM fungi dominantly detected in Japanese temperate forests. Accordingly, there is a possibility that mycoheterotrophic plants, including seedlings of G. zollingeri, may target AM fungi with a wide host range and ubiquitous distribution.
In the general mycorrhizal symbioses, host plants provide photosynthetic carbon to mycorrhizal fungi; in return, plants obtain soil nutrients, such as nitrogen and phosphorus, from fungi (van der Heijden, Martin, Selosse, & Sanders, 2015). However, some plants have lost their photosynthetic ability and depend on mycorrhizal fungi for carbon sources. Such a nutritional mode is called mycoheterotrophy, which has evolved independently at least 47 times in various plant lineages (Leake, 1994; Merckx, Mennes, Peay, & Geml, 2013). Mycorrhizal fungi for mycoheterotrophic plants are selected from Basidiomycota, Ascomycota, or Glomeromycotina depending on the plant taxa.
Many land plant species form arbuscular mycorrhiza (AM) with Glomeromycotina fungi (Spatafora et al., 2016). Morphologies of AM are divided into two types, i.e., Arum and Paris (Gallaud, 1905), although intermediate structures are often present (Dickson, Smith, & Smith, 2007). Arum-type AM characterized by intercellular hyphae and arbuscules is usually formed by plants living in sunny conditions, such as farmland and grassland (Dickson et al., 2007), whereas Paris-type AM characterized by the formation of intracellular hyphal coils is frequently found in understorey plants (Brundrett & Kendrick, 1990; Yamato & Iwasaki, 2002). There are >230 mycoheterotrophic AM plant species in nine families (Merckx, 2013), and Paris-type AM has been observed in all examined plants (Imhof, Massicotte, Melville, & Peterson, 2013). Recently, it has been suggested that partial mycoheterotrophy, having both photosynthesis and mycoheterotrophy, may be common among Paris-type AM plants (Giesemann, Rasmussen, Liebel, & Gebauer, 2020; Giesemann, Rasmussen, & Gebauer, 2021).
Mycoheterotrophic plants usually produce extremely small seeds with limited amounts of nutrients (Eriksson & Kainulainen 2011). These plants depend on mycorrhizal fungi for external organic carbon sources used in seed germination and subsequent seedling development. This feature is also known for plants with initial mycoheterotrophy, the mycoheterotrophy at the seedling stage of some photosynthetic plants, in Orchidaceae and Ericaceae (Jacquemyn & Merckx, 2019). Recently, Yamato, Suzuki, Matsumoto, Shiraishi, and Yukawa (2021) reported initial mycoheterotrophy in Gentiana zollingeri Fawc. (Gentianaceae), a small green herbaceous plant that stands about 5-10 cm in height and forms Paris-type AM. This plant is a spring-flowering species distributed in Japan, Korea Peninsula, China, Sakhalin, and Amour. Yamato et al. (2021) also reported that seedlings of G. zollingeri growing underground were dominantly colonized by a specific group of Glomeraceae fungi which are closely related to mycorrhizal fungi for a fully mycoheterotrophic plant Sciaphila tosaensis Makino (Yamato, Yagame, & Iwase, 2011a).
So far, many studies investigating AM fungi in roots of mycoheterotrophic plants have reported specificity toward particular lineages of fungi in the symbioses (Bidartondo et al., 2002; Franke, Beenken, Döring, Kocyan, & Agerer, 2006; Yamato et al., 2011b; Suetsugu, Kawakita, & Kato, 2014). These plants target more narrow lineages of fungi than green plants despite the larger fungal pool available in the soil (Gomes, Aguirre-Gutiérrez, Bidartondo, & Merckx, 2017). Also, various mycoheterotrophic plants with independent phylogenetic lineages often utilize common AM fungi, and many of them belong to Glomeraceae (Merckx et al., 2012; Perez-Lamarque, Selosse, Öpik, Morlon, & Martos, 2020). Such fungal overlapping among different mycoheterotrophic plant species suggests that several cheating susceptible AM fungi exist. Moreover, it was suggested that mycoheterotrophic plants preferentially target AM fungi that are highly connected to surrounding autotrophic plants (Gomes, Fortuna, Bascompte, & Merckx, 2022).
In this study, a mycorrhizal fungus specifically associated with mycoheterotrophic seedlings of G. zollingeri was isolated from a natural habitat of this plant to make a pot culture. Compared to the number of AM fungal species described, pot-cultured species are limited. Moreover, considering the inclusion of many undescribed fungi in AM fungal community analyses based on PCR amplified fungal DNA data (Öpik, Davison, Moora, & Zobel, 2014), the ratio of cultured fungi is further limited. AM fungi have been described based on spore morphologies, and AM fungal cultures have been established using single or multiple spores as inocula (Brundrett, Bougher, Dell, Grove, & Malajczuk, 1996). Thus, AM fungi with limited spore production have not been described or cultured. In this study, we successfully isolated a mycorrhizal fungus using a short fragment of a colonized root as an inoculum. The AM fungus was identified by spore morphology and molecular analyses. Moreover, fungal distribution was investigated along with potential host plants by searches against the MaarjAM database to infer the ecological characteristics of the fungal taxa.
In mycoheterotrophic seedlings of G. zollingeri, the specific symbiosis with a group of AM fungi was shown by Yamato et al. (2021). In this study, an AM fungus was isolated into a pot culture as follows (a schematic figure of the isolation process was shown in Supplementary Fig. S1). Four soil cores (5 cm diam and 10 cm in depth) containing roots of surrounding plant species were collected using a stainless tube at a 10 cm distance from each of four flowering shoots of G. zollingeri arbitrarily selected in a habitat in Tsukuba Botanical Garden (Tsukuba City, Ibaraki Prefecture; N36.10, E140.11) on Apr 23, 2017.
The soil core was buried into a pot containing 1 L of an autoclaved soil medium, i.e., a mixture of akadama soil (granules of volcanic soil) and river sand (1:1, v/v). Seeds of Medicago sativa L. sterilized by sodium hypochlorite with 1% effective chlorine were sown on Apr 24, 2017. Three seedlings were grown in each pot, and the pot cultures were fertilized with 500 mL Peters liquid fertilizer (25-5-20) at an N concentration of 100 mg/L every 3 mo. Seed packets with 40 × 60 mm rectangles of 100 μm nylon mesh containing ～100 seeds of G. zollingeri were prepared according to Yamato et al. (2021), and four seed packets were buried in each of the four pots on Jun 1, 2017, 38 d after the seed sowing. The pot cultures were cultivated on a cultivation shelf MLR-1546 (Sanyo, Osaka, Japan) in a condition of 16 h day length with a photon flux density of around 40 μmol m-2 s-1 at 25 °C.
On Aug 9, 2018, 434 d after burying the seed packets, all seed packets were checked for seedling development. From a pot culture with a G. zollingeri seedling development, a small soil core (15 mm diam) containing M. sativa root was collected. Root fragments of M. sativa with extraradical AM fungal mycelium were collected from the soil core and cut into 5 mm length. The small root fragment was inoculated to the tip of the root of another M. sativa seedling grown in the autoclaved soil medium for 3 wk. The seedling was then planted in a pot containing 100 mL of the autoclaved soil medium. After fertilizing with 100 mL of the liquid fertilizer described above, the pot cultures were cultivated on the cultivation shelf for 3 mo. From the pot cultures with AM fungal colonization, a sporocarp isolated by wet sieving (Brundrett et al., 1996) was used as an inoculum for the subsequent culture. In the subsequent culture, a M. sativa seedling was grown in 500 mL of the autoclaved soil medium by applying 200 mL of the liquid fertilizer every 3 mo.
The roots of M. sativa from the pot culture of isolated fungus were cleared by autoclaving at 121 °C for 15 min in 10% KOH and stained using 0.05% trypan blue in lactic acid. The stained roots were squashed using a cover glass, observed under a microscope BX51 equipped with differential interference contrast optics (Olympus, Tokyo, Japan), and photographed with a CCD camera DP72 (Olympus).
2.2. Spore extractionAfter 1 year of pot culture, sporocarps were collected from the cultured soil by wet sieving using a stack of sieves of 2 mm, 250 µm, and 106 µm mesh sizes. Sporocarps on 250 and 106 µm sieves were used for morphological and molecular analyses of the AM fungus.
2.3. Morphological analyses of the sporocarpsIntact sporocarps were observed with a stereomicroscope SZX7 (Olympus) and photographed with a CCD camera (DS-Fi3-L4; Nikon, Tokyo, Japan). Spores derived from sporocarps were mounted on microscope slides in water, polyvinyl lactoglycerol (PVLG), or a mixture of PVLG and Melzer's reagent (1:1, v/v). Some spores were ultrasonically washed in water for 30 s before mounting. The resulting spore slides were examined under the microscope BX51 and photographed with the CCD camera DP72. Size measurements of sporocarps (n=13) and spores (n=100) were made using the image assessment software Lumina Vision version 3.7 (Mitani, Fukui, Japan).
2.4. Molecular analysisDNA was extracted from sporocarps using the DNeasy Plant Mini Kit (Qiagen, Hilden, Germany). To identify the fungal taxa, a DNA fragment, including partial small subunit (SSU), whole internal transcribed spacer (ITS), and partial large subunit (LSU) within nuclear ribosomal RNA gene (SSU-ITS-LSU rDNA), was amplified from the extracted DNA by polymerase chain reaction (PCR) with the primer pair SSUmCf/LSUmBr (Krüger, Stockinger, Krüger, & Schüßler, 2009) using TaKaRa Ex Taq Hot Start Version (Takara Bio, Inc., Kusatsu, Japan). The PCR mixture contained 1 μL extracted DNA, 0.075 μL Taq polymerase, 0.4 μM of each primer, 200 μM of each deoxynucleotide triphosphate, and 1.5 μL of the supplied PCR buffer in a total volume of 15 μL. The PCR program was performed on a Thermal Cycler Gene Atlas (ASTEC, Fukuoka, Japan) as follows: initial denaturation at 94 °C for 2 min, followed by 35 cycles at 94 °C for 30 s, 55 °C for 1 min, 72 °C for 1 min, then a final elongation step at 72 °C for 10 min. The largest subunit of RNA polymerase II gene (RPB1) was also amplified with the primer pair RPB1-HS_A1a/RPB1-DR1730r (Stockinger, Peyret-Guzzon, Koegel, Bouffaud, & Redecker, 2014). The PCR mixture contained 2 μL extracted DNA, 0.075 μL Taq polymerase, 0.25 μM of each primer, 200 μM of each deoxynucleotide triphosphate, and 1.5 μL of the supplied PCR buffer in a total volume of 15 μL. The PCR program was as follows: initial denaturation at 94 °C for 2 min, followed by 35 cycles at 94 °C for 30 s, 58 °C for 1 min, 72 °C for 2 min, then final extension step at 72 °C for 10 min.
To confirm whether the fungal isolate corresponded to the mycobiont of G. zollingeri detected by Yamato et al. (2021), SSU of nuclear rDNA was amplified from the extracted DNA obtained from the sporocarps with PCR primers NS1 and NS8 (White, Bruns, Lee, & Taylor, 1990). The composition of the PCR mixture was the same as RPB1 described above, except for the primers and amount of extracted DNA (1 μL). The PCR program was the same as SSU-ITS-LSU rDNA described above.
All PCR products were purified using a Gel/PCR DNA Fragments Extraction Kit (RBC Bioscience, Taipei, Taiwan) and cloned using the pGEM-T Easy Vector System I (Promega, Madison, WI, USA). Colonies with DNA inserts were randomly selected and sequenced by a commercial sequencing service (Takara Bio).
2.5. Phylogenetic analysisThe obtained SSU-ITS-LSU rDNA and RPB1 sequences were searched against the International Nucleotide Sequence Database (INSD) using the Basic Local Alignment Search Tool (BLAST; Altschul et al., 1997), and some phylogenetically related sequences of the described species were downloaded. For the set of sequenced and downloaded data, multiple sequence alignments by MUSCLE (Edgar, 2004) were performed using MEGA 11 (Tamura, Stecher, & Kumar, 2021). The alignments of SSU-ITS-LSU rDNA and RPB1 were deposited in figshare (https://doi.org/10.6084/m9.figshare.20688481). Maximum likelihood (ML) analyses with 1000 bootstrap (BS) replicates (Felsenstein, 1985) were carried out using MEGA 11. For the ML analysis of SSU-ITS-LSU rDNA and RPB1, general time reversible (GTR) and γ-distributed with invariable sites (+G+I), and Tamura 3-parameter (T92) +G were selected as the best-fit models, respectively. Phylogenetic trees were drawn using FigTree version 1.4.4 (http://tree.bio.ed.ac.uk/software/figtree/).
2.6. Calculation of genetic distancesTo evaluate the sequence similarity of SSU-ITS-LSU rDNA between the fungal isolate in this study and the most closely related species, pairwise genetic distances were calculated for the aligned data based on the Kimura two-parameter (K2P) model (Kimura, 1980) with pairwise deletion of gaps using MEGA 11.
2.7. Confirmation of the isolate as a mycobiont of G. zollingeriBecause Yamato et al. (2021) identified the mycobionts of G. zollingeri using partial SSU rDNA sequences amplified with primer pairs NS31/AML2 (Simon, Lalonde, & Bruns, 1992; Lee, Lee, & Young, 2008), partial SSU rDNA sequences were compared to confirm whether the AM fungus isolated in this study corresponded to the mycorrhizal fungi of G. zollingeri.
For the set of partial SSU rDNA sequences obtained in this study and Yamato et al. (2021), multiple sequence alignment by MUSCLE was performed using MEGA 11. The alignment of partial SSU rDNA was deposited in figshare (https://doi.org/10.6084/m9.figshare.20688481). ML analysis with 1000 BS replicates was carried out using MEGA 11, in which the T92 was selected as the best-fit model. Phylogenetic trees were drawn using FigTree version 1.4.4.
2.8. Investigation of the fungal distributionTo infer the distribution and the potential host plants of the AM fungus isolated in this study, the SSU rDNA sequences were searched against the MaarjAM database (Öpik et al., 2010), an online database consisting of AM fungal DNA sequences originating from ecological and taxonomic studies, in which partial SSU rDNA sequences are classified into virtual taxa (VTX). In the VTX corresponded with sequences of the fungal isolate, the number of registered sequences was counted according to the division of “Ecosystems” classification in MaarjAM: anthropogenic, culture, forest, grassland, shrubland, and successional. When several sequences were registered in the same sampling site of the same study, they were counted as one. The number of sequences were also counted in the VTX according to the division of “Continents” in MaarjAM: Asia, Europe, Africa, Oceania, North America, and South America. The host plants of each VTX registered in MaarjAM were also examined.
Among the 16 seed packets, four packets each from four pot cultures of M. sativa inoculated with soil cores collected from a G. zollingeri habitat, only one packet contained a G. zollingeri seedling that developed to the shooting stage (Fig. 1). Root fragments of M. sativa, 5 mm long, with AM fungal extraradical mycelium were cut off from this pot culture, and then the fragment was inoculated onto a root tip of another non-mycorrhizal M. sativa seedling. In total, 60 pot cultures were cultivated, from which AM colonization was confirmed in 20 pots after 3 mo. Since sporocarps were formed in most of the pot cultures with AM colonization, a sporocarp in a pot was arbitrarily chosen and used as an inoculum for the subsequent culture. The fungal pot culture (CI1701) was deposited to The Genetic Resource Center, National Agriculture and Food Research Organization, Japan, with accession number MAFF 329031.
Arbuscules and vesicles were observed in the roots of M. sativa colonized by the fungal isolate CI1701 after staining with 0.05% trypan blue (Supplementary Fig. S2).
3.2. Species identification based on spore morphologyMorphological features of the isolated AM fungus CI1701 fit the description of Dominikia aurea (Oehl & Sieverd.) Błaszk, Chwat, G.A. Silva & Oehl, as described below.
Sporocarps, somewhat compacted, irregular in shape, 219-714 × 361-1114 μm diam, without peridium (Fig. 2A). Spores and hyphae showing pastel red to orange-red in Melzer's reagent (Fig. 2B). Parts of the sporocarps can be covered by loosely interwoven hyphae; spores and hyphae embedded in gelatinous material (Fig. 2B, C) that stains pastel red to orange-red in Melzer's reagent (Fig. 2B). Spores formed at the apex of repeatedly dichotomously branched hyphae (Fig. 2C, D) or directly on secondary hyphae perpendicularly; spores yellowish brown to golden brown, subglobose to ellipsoidal, 45.6-77.9 × 33.6-55.3 μm, 58.7 × 43.7 μm on average (Fig. 2D), or rarely pyriform 94.0 × 48.7 μm diam. Spore wall consisted of a single layer (w), yellow golden to golden brown, laminated, 2.2-5.2 μm thick, 3.5 μm on average, with external gelatinous material (egm) that is hyaline, evanescent, usually absent in mature spores (Fig. 2C-E). At the spore base, spore wall is thickened (Fig. 2D), subtending hypha cylindrical or slightly funnel-shaped, 6.4-14.1 μm in diam, 10.6 μm on average, the wall of subtending hypha continuous to the spore wall, yellow golden to golden brown, 2.1-7.7 μm thick, 4.6 μm on average, tapering with distance from the spore base, then becoming hyaline. The pore of the subtending hypha is straight, 1.0-3.0 μm wide, 1.7 μm on average, often closed by a septum formed up to 7 μm from the spore base (Fig. 2F).
Four sequences were obtained for each of the SSU-ITS-LSU rDNA and RPB1 regions. The DNA sequences were deposited in the INSD under accession numbers LC723711 to LC723718. BLAST searches showed that SSU-ITS-LSU rDNA and RPB1 sequences were most closely related to D. aurea among the described Dominikia species. Phylogenetic analysis in SSU-ITS-LSU rDNA and RPB1 regions showed that sequences obtained from CI1701 formed a sister clade to the D. aurea sequences (Figs. 3, 4). The K2P distances of SSU-ITS-LSU rDNA sequences ranged from 0.066% to 5.5% within CI1701 and from 2.5% to 5.1% between CI1701 and D. aurea (Supplementary Table S1).
Two SSU rDNA sequences (“SSU-A” and “SSU-B” hereafter), 1744 and 1745 bp long, obtained from CI1701 were deposited in the INSD under accession numbers LC723719 and LC723720. Compared to operational taxonomic units (OTUs) of the mycorrhizal fungi of G. zollingeri clustered with 97% sequence similarity in Yamato et al. (2021), the partial SSU-A (521 bp) was identical to the sequences (e.g., LC590278) belonging to the most dominant OTU (Glo1) of mycorrhizal fungi, and the partial SSU-B (522 bp) was 99.81% matched to the sequences (e.g., LC590316) belonging to the second dominant OTU (Glo2). Phylogenetic analysis also showed that partial SSU-A and SSU-B were clustered with sequences belonging to Glo1 and Glo2, respectively (Fig. 5). Therefore, it was confirmed that the AM fungus CI1701 isolated in this study corresponded to the mycorrhizal fungi of G. zollingeri.
BLAST searches against the MaarjAM revealed that SSU-A and SSU-B corresponded to VTX166 and VTX159, respectively. Because SSU-A and SSU-B were obtained from the same fungal culture, the two VTXs might be conspecific. Both VTXs had been detected in various ecosystems (Table 1) and in all six continents: Asia, Europe, Africa, Oceania, North America, and South America. Host plants of VTX166 and VTX159 belonging to 77 families included liverworts, hornworts, lycopods, ferns, gymnosperms, and angiosperms, which included mycoheterotrophic plants, such as Burmannia championii, B. nepalensis, Gymnosiphon divaricatus (Burmanniaceae), Sciaphila ledermannii, and S. tosaensis (Triuridaceae).
| Ecosystems a | AM fungal taxa | ||
| VTX166 (SSU-A) | VTX159 (SSU-B) | Total b | |
| Anthropogenic | 24 | 4 | 24 |
| Culture | 0 | 0 | 0 |
| Forest | 37 | 12 | 37 |
| Grassland | 21 | 15 | 22 |
| Shrubland | 13 | 1 | 13 |
| Successional | 8 | 4 | 12 |
| Total | 103 | 36 | 104 |
a The ecosystems are classified according to MaarjAM.
b Total number of registrations excluding duplicates of VTX166 and VTX159.
This study successfully isolated an AM fungus for the mycoheterotrophic G. zollingeri seedling. Isolation of AM fungi is usually carried out using single or multiple spores as inocula, but a single small root fragment with extraradical AM fungal hyphae was used as the inoculum in the present study. Since many vesicles were observed in roots of M. sativa inoculated with this fungus, the success of the AM formation to the new host plant was possibly due to the regrowth of the mycelium from the vesicles. To our knowledge, this is the first report to use a single small root fragment as an inoculum source for specific AM fungal species isolation into pot culture. Using this method, it would be possible to isolate AM fungus regardless of spore production.
The phylogenetic analysis for both SSU-ITS-LSU rDNA and RPB1 showed that the fungal isolate CI1701 formed a clade closely related to Dominikia aurea. This species was first described as Glomus aureum with the holotype specimen originated from a grassland in Switzerland (Oehl, Wiemken, & Sieverding 2003). It was then recombined with several other species in Dominikia as D. aurea according to the phylogenetic analyses by Oehl, Sanchez-Castro, Silva, & Palenzuela (2015).
Although CI1701 and D. aurea formed independent clades in both phylogenies in this study, the analyzed samples were limited. Furthermore, the K2P distances of SSU-ITS-LSU rDNA sequences between CI1701 and D. aurea were 2.5%-5.1%, which was equivalent to the intra-isolate variation of CI1701 (0.066%-5.5%) or the intraspecific variation of AM fungi (0.47%-10.8%) in Stockinger, Krüger, and Schüßler (2010). We therefore avoided describing a novel species for the fungal isolate obtained in this study.
SSU-A and SSU-B obtained from the isolate CI1701 corresponded to VTX166 and VTX159 in MaarjAM, respectively. Each VTX was clustered at a 97% sequence similarity threshold for partial SSU rDNA, which is roughly corresponded to species-level taxa (Öpik et al., 2014). Thus, high intragenomic variation is suggested for the SSU rDNA of this isolate.
BLAST searches against the MaarjAM revealed that closely related AM fungi of the isolate CI1701 are the mycorrhizal fungi of various mycoheterotrophic plants not only in Gentianaceae but also in Burmanniaceae and Triuridaceae. Furthermore, VTX166 was recently detected in mycoheterotrophic Oxygyne yamashitae (Thismiaceae; Suetsugu, Okada, Hirota, & Suyama, 2022). Perez-Lamarque et al. (2020) found that some closely related AM fungi were shared by independently emerged mycoheterotrophic plant lineages. Thus, it is suggested that CI1701, i.e., D. aurea or its closely related AM fungus, is a part of the cheating susceptible AM fungi by mycoheterotrophic plants.
According to the MaarjAM, the VTX166 and VTX159 have been detected in various host plants from liverworts to seed plants on six continents. It has also been reported that VTX166 corresponded to one of the dominant OTUs in Japanese temperate forests (Miyake, Ishitsuka, Taniguchi, & Yamato, 2020; Matsuda, Kita, Kitagami, & Tanikawa, 2021). It was suggested that mycoheterotrophic plants preferentially target AM fungi that are well connected to many surrounding autotrophic plants (Gomes et al., 2022). In fact, in a 10 × 10 m plot established in an Estonian forest, VTX166 was detected in the roots of all 10 plant species examined (Öpik, Metsis, Daniell, Zobel, & Moora, 2009). Hence, mycoheterotrophic plants may target AM fungi with a wide host range and ubiquitous distribution.
The authors declare no conflicts of interest. All the experiments undertaken in this study comply with the current laws of the country where they were performed.
This study was supported by JSPS KAKENHI grant number 19K22269. We thank Dr. Tomohisa Yukawa for his help in the soil core sampling. We also thank Enago (www.enago.jp) for English language review.
