Tulasnella calospora (UAMH 9824) retains its effectiveness at facilitating orchid symbiotic germination in vitro after two decades of subculturing

The technique of symbiotic germination—using mycorrhizal fungi to propagate orchids from seed in vitro—has been used as one method to cultivate orchids in North America and abroad for > 30 years. A long-held assumption is that mycorrhizal fungi used for this purpose lose their effectiveness at germinating seeds over time with repeated subculturing. We provide evidence for the lingering efficacy of one particular strain of Tulasnella calospora (266; UAMH 9824) to stimulate seed germination exemplified by the North American terrestrial orchid, Spiranthes cernua, as a case study. This fungus was originally acquired from roots from Spiranthes brevilabris in 1999 and sub-cultured during the two decades since. Seeds inoculated with the fungus in vitro developed to an advanced protocorm stage after 16 days, and leaf elongation was pronounced after 42 days. In a pilot study, seedlings co-cultured with Tulasnella calospora 266 were deflasked after 331 days and later transferred to soil under greenhouse conditions where they eventually initiated anthesis. During the course of two decades, seeds of 39 orchid species, cultivars and hybrids spanning 21 genera, germinated in vitro co-cultured with Tulasnella calospora 266. These orchids included temperate terrestrials and tropical epiphytes alike. The sustained effectiveness of this fungus is noteworthy because it argues against the concept of mycorrhizal fungi losing their symbiotic capability through prolonged subculturing. This study serves as an example of why in situ habitat preservation is essential for the conservation of orchids as a source of potentially useful mycorrhizal fungi.


Background
The technique of symbiotic germination has been used as one method to propagate orchids from seed for over 30 years (Clements et al. 1986). In 1999, an attempt was made to isolate fungi for this purpose to propagate Spiranthes brevilabris from seed in vitro-a rare species restricted to a single population consisting of 152 individuals clustered within a small (21,600 ha) state park in Florida, USA (Stewart et al. 2003). One fungus was subsequently isolated and provisionally identified as a member of the Rhizoctonia complex, the group of higher fungi (basidiomycetes) known for forming mycorrhizal associations with photosynthetic orchids worldwide. This fungus matched published descriptions for Epulorhiza repens (N. Bernard) R. T. Moore (1987), a common ubiquitous species. One year after its isolation, seeds collected from the donor population were inoculated with the fungus in vitro using standard protocols (Dixon 1987), and seed germination and seedling development rapidly ensued. Only 133 days after sowing, > 165 laboratory-grown seedlings were reintroduced to Florida, and 100% survived after one month, effectively doubling the size of the orchid population. Six months later, 17 of the reintroduced orchids initiated anthesis (Stewart et al. 2003). This fungus was then permanently retained and catalogued at Illinois College as Isolate 266 (Fig. 1) where it was then placed in cool (4 °C) storage for future use. In addition, the fungus was also deposited into the University of Alberta Microfungus Herbarium in Canada as UAMH 9824 for safekeeping under cryopreservation.
In the decades that followed, this fungal isolate was used in other symbiotic germination experiments involving a wide range of additional orchid species. Among the taxa successfully propagated to the leaf-bearing stage included epiphytes (Cyrtopodium punctatum and Epidendrum nocturnum) as well as terrestrials (e.g., Habenaria repens and Platanthera holochila) including five additional Spiranthes species (Massey and Zettler 2007). The use of UAMH 9824 in germination experiments extended into a second decade where it retained its effectiveness at facilitating advanced seedling development, earning it the nickname as the super-fungus. At the same time, modern molecular techniques, namely Sanger sequencing of barcode regions like Internal transcribed Spacer of rDNA (ITS), were being perfected that enhanced fungal identification within the Rhizoctonia complex to which members of the genus Epulorhiza belonged. These new techniques also coincided with the abandonment of the anamorphic classification system in higher fungi in favor of the teleomorphic binomial. Hence, members of the genus Epulorhiza R. T. Moore are now placed in the genus Tulasnella J. Schröt, and E. repens is now regarded as Tulasnella calospora (Bourdier) Juel, which we refer to herein as Tulasnella calospora 266.
The ability of Tulasnella calospora 266 to retain its efficacy during the span of two decades counters a long-held assumption that fungal endophytes lose their ability to stimulate seed germination through repeated subculturing (Bernard 1909;Alexander and Hadley 1983). At the time of this writing (2021), this fungus has facilitated in vitro symbiotic germination of two epiphytic orchids native to Ecuador (Quijia-Lamiña 2020) and is also now being used by at least one hobbyist to propagate several terrestrial orchids including one native to Europe (Dactylorhiza praetermissa; D. Martin, pers. com.). In this paper, we provide evidence for the sustained effectiveness of Tulasnella calospora 266 to stimulate seed germination spanning two decades, using Spiranthes cernua as a case study. The effectiveness of this fungus is compared to three other fungal isolates as well as two types of asymbiotic media. We conclude by presenting a comprehensive list of the orchid species propagated with this fungus during the past 20 years to the best of our knowledge.

Maintenance of Tulasnella calospora 266
Two types of agar media were employed in the care and maintenance of Tulasnella calospora 266 over the course of two decades. For short-term maintenance at ambient temperature, potato dextrose agar (PDA) was used (Difco ™ #213400, Difco Laboratories, Detroit, MI, USA). Cultures were maintained on PDA in 9 cm diam. petri dishes that were wrapped in Parafilm ® "M" (Menasha, WI, USA) to seal in moisture. At 3-6 month intervals, a 1 cm 3 block of fungal inoculum from the outer margin of the colony was subcultured to a new PDA plate using a sterile scalpel, and the process was repeated. In cases when the fungus failed to initiate growth usually > 6 months, a back-up culture of the fungus was restarted on PDA. Back-up cultures consisted of maintaining T. calospora 266 in refrigeration (4 °C) on oatmeal agar (OMA) slants in screw-cap test tubes for 1-2 years. OMA consisted of 2.5 g/L Quaker Oats ® , Chicago, IL, USA; 7.5 g/L Bacto ™ Agar (#214010), Becton, Dickinson and Co., Sparks, MD USA.

Seeds, mycorrhizal fungi, and agar media
Seeds of Spiranthes cernua (L.) Rich were collected by E. Esselman on 25 September 2018 from a natural population located in Bond Co., Illinois that contained ca. 50 individual orchids. Multiple mature capsules derived from natural pollination among 10 different individuals were removed from inflorescences just prior to shown within a 9 cm diam. petri dish containing Potato Dextrose Agar > 10 days after inoculation. The culture is recognizable by the smooth, cream-colored mycelium and smooth colony margins. Photo courtesy of Michael E. Kane dehiscence, cleaned of debris and immediately dried over CaSO 4 (Drierite, W.A. Hammond Co., Xenia, Ohio, USA) at ambient temperature (22 °C) for 10 days. Seeds were then removed from capsules, pooled into a single airtight vial, and stored at 4 °C for 61 days in darkness until use.
Four fungal isolates were tested for their ability to facilitate seed germination of S. cernua. Two were from Florida and assignable to the genus Tulasnella including 266, and the other two consisted of Ceratobasidium strains from prairie habitats similar to where S. cernua grew naturally (Table 1). Two types of asymbiotic media were chosen for comparative purposes: P723 Orchid Seed Sowing Medium, and B141 Terrestrial Orchid Medium (PhytoTechnology Laboratories ® , Shawnee Mission, KS, USA). The media chosen for symbiotic germination consisted of standard oatmeal agar or OMA (mentioned previously). All media were prepared using RO water with the pH adjusted to 5.6-5.8 prior to autoclaving using NaOH or HCl.

Seed sowing, inoculation, and incubation in vitro
Seeds were surface-sterilized within packets prepared from AeroPress coffee filters (AeroPress ® , Palo Alto, CA, USA). Seeds were placed in the center of each coffee filter, which was folded around the seeds and stapled shut into a packet. The number of seeds delivered into each packet was carried out volumetrically with a micro spatula. The volume was estimated at 4 mm 3 based on a small strip of seeds on the end of the spatula measuring ca. 4 mm long, 1 mm deep, and 1 mm high. Considerable care was given to adding an equal number of seeds to each packet. A total of 70 packets were constructed. The packets were pre-soaked by immersing in reverse osmosis (RO) water with 1 drop of Tween ® 20 per 100 mL, and agitated on an orbital shaker at 130 rpm for 10 min. Packets were then decontaminated by immersing them in an RO water solution containing 0.5% NaDCC + 1 drop Tween ® 20 per 100 mL that was agitated on the shaker at 130 rpm for 30 min. Packets were removed from the solution under a laminar flow hood and transferred to a container of sterile water to remove traces of the decontamination solution for ca. 5 min. Packets were then removed from the sterile water and cut open with a scalpel under a laminar flow hood and seeds from each packet were spread onto the agar surface within 9 cm diam. petri dishes (25 mL agar/dish). Because some seeds lacked embryos, only those that contained embryos were counted and included in our data set. This yielded a range of 37-125 seeds per dish. With the exception of controls, all oatmeal agar plates were inoculated with a given fungus by adding a 1 cm 3 block of inoculum to the center of each petri plate. Seven treatments were tested: OMA control, OMA + Tulasnella calospora 266, OMA + Tulasnella 427, OMA + Ceratobasidium PP4, OMA + Ceratobasidium EE465, PhytoTech P723, and PhytoTech B141.
Following sowing/inoculation, plates were sealed in polyethylene film, stacked into a Styrofoam box with a lid, and incubated in darkness within a growth chamber at 25 °C. After 16 days, plates were inspected visually for contamination and early signs of germination, and promptly returned to the growth chamber. After 42 days (post-sowing), all plates were removed from growth chambers for data collection and observation. Seed germination and seedling development were scored on a scale of 0-5 in accordance with Zettler et al. (1995) where: Stage 0 = no germination; 1 = rupture of testa by enlarged embryo (= germination); 2 = rhizoid formation; 3 = elongation of protocorm (shoot formation); 4 = appearance of first leaf within shoot region; and 5 = elongation of first leaf. Data were analyzed using general linear model procedures multivariate analysis of variance (MANOVA) and mean separation at α = 0.05 by IBM SPSS Statistics for Windows, Version 26.0 for Windows subprogram (Armonk, NY: IBM Corp.). Plates that harbored leaf-bearing seedlings were transferred to a second growth chamber and illuminated under an 18 h photoperiod (L:D 18:6 h) to induce photosynthesis in the shoot region (Fig. 2). Illumination was provided by full

Seedling establishment ex vitro
In a pilot study that led to the in vitro techniques described above, seedlings with sizable leaves were removed from petri dishes and transferred to taller 8 oz.
(ca. 250 mL) polypropylene vessels containing 80 ml OMA, 134 days after sowing. These seedlings, which were obtained using two of the fungal isolates (Tulasnella calospora 266 and Ceratobasidium PP4), were placed onto the surface of OMA in the vessel instead of a mineral-enriched salt medium. This decision was based on concerns that higher levels of mineral salts such as nitrogen may have a negative impact on plant-fungal symbiotic relationships. Each vessel contained 12 seedlings (seedlings from different treatments were not combined into the same vessel). Vessels were placed back into the growth chamber for further growth. After 331 days post-sowing, seedlings were removed from their vessels and the media was rinsed from the roots. They were placed into sealed plastic bags containing damp sphagnum (peat) moss and vernalized in a refrigerator at 4 ℃ for 3 months. After vernalization, seedlings were planted in potting mix consisting of aged pine bark, Canadian sphagnum moss, and perlite (Ball Horticultural Co, West Chicago, IL, USA) and placed in a greenhouse to continue growing leading to anthesis.

Results
Seed germination was observed in all seven treatments 16 days after sowing followed by incubation in darkness.
Of the seven treatments, seeds inoculated with Tulasnella calospora 266 progressed to Stage 4-the highest growth stage amongst the treatments at that point in time. After 42 days of incubation, seeds inoculated with Tulasnella calospora 266 and Ceratobasidium PP4 resulted in the highest percent germination observed (33.7 and 34.6%, respectively) followed by Tulasnella 427 (29.6%). The MANOVA revealed germination across treatments was significant, F (6,62) = 72.65, p < 0.01 (Table 2). After 42 days of incubation, 5.8% of seedlings co-cultured with Tulasnella calospora 266 developed to Stage 5 (leaf elongation) compared to only 1.2% of seedlings in the Ceratobasidium PP4 treatment (Fig. 3). None of the other five treatments, including OMA + Tulasnella 427 developed beyond Stage 2 at the 42 day mark (Table 2).

Discussion
This case study demonstrated the effectiveness of Tulasnella calospora 266 (UAMH 9824) at facilitating seed germination and seedling development of Spiranthes cernua despite repeated subculturing over the course of 20 years. This efficacy is evidenced by the rapid (42 day) development of seedlings to Stage 5 in symbiotic culture (Table 2) compared to two other fungal isolates acquired more recently (2016 and 2018). This study is just one example of the many different types of orchids that have been propagated to date spanning North America and abroad, even to the blooming stage such as Spiranthes sinensis ( Fig. 4; Table 3). The sustained effectiveness of Tulasnella calospora 266 to propagate a broad range of species, commercial hybrids (e.g., Cattleya; Fig. 5) and species under cultivation (e.g., Gongora grossa; Fig. 6) is noteworthy because it argues against the concept of mycorrhizal fungi losing their symbiotic capability through prolonged subculturing (Bernard 1909;Alexander and Hadley 1983). Consequently, this study may have merit for improving conservation and horticulture alike. The use of this fungus to germinate seeds of a North American hybrid (Platanthera integrilabia x ciliaris) by at least one hobbyist (D. Martin) suggests that Tulasnella calospora 266 could also be utilized in developing new varieties for horticulture, as well as answer fundamental ecological questions.
In the pilot study, we questioned whether the transfer of symbiotically-grown S. cernua seedlings on OMA to a more mineral salt-enriched medium might lead to mineral toxicity, and for this reason we chose to use the same OMA medium instead. This resulted in seedlings that appeared to display symptoms of nitrogen deficiency evidenced by chlorosis in the apical leaves and loss of most basal leaves closest to the crown. The subsequent transfer of these seedlings to potting mix after vernalization resulted in rapid growth suggesting that  the minerals in the mix provided an adequate supply of nutrients (e.g., nitrogen). In follow-up experiments (ongoing), seedlings on OMA transferred directly to pasteurized potting mix instead of OMA showed no signs of nitrogen deficiency (C. Dvorak, pers. obs.).
Consequently, we recommend that other researchers consider transferring seedlings on OMA directly to pasteurized potting soil, and that the use of media containing mineral-salts may not be necessary. Another highlight of this study concerns the origin of Tulasnella calospora 266, specifically from the donor orchid (Spiranthes brevilabris) and the location (Dunnellon, Levy Co., FL) from which it was originally obtained. According to Stewart et al. (2003), this fungus was acquired from the last known population of S. brevilabris that was reduced to just 152 individuals under legal protection in a small (21,600 ha) state park in Florida. The region bordering the park has experienced high population growth and habitat loss during the past 30 years. While artificial boundaries and legal protection do not always guarantee the security of rare plants such as S. brevilabris, it is conceivable that Tulasnella calospora 266 may never have been isolated had this area not been designated as a nature preserve prior. As such, this study exemplifies why in situ habitat preservation is essential for the conservation of orchids and other life forms even in areas restricted in size.
For the purposes of ecological restoration, i.e., reintroducing laboratory-grown seedlings into the wild, we urge researchers and land managers to use discretion in the widespread use of Tulasnella calospora 266 until more information becomes available on the natural range of this particular fungus. Zettler et al. (2001Zettler et al. ( , 2005, for example, opted not to reintroduce laboratory-grown seedlings of the U.S. Federally endangered Hawaiian endemic, Platanthera (Peristylus) holochila, because the fungus originated from Florida-a distance separated by > 7500 km. Although Tulasnella calospora is reported to be global in distribution, it apparently constitutes a diverse fungal assemblage in need of further study. The use of modern molecular techniques currently available is expected to resolve this complex in the coming years which should shed more light into the distribution of Tulasnella calospora 266 and its nearest relatives. Recently, Unruh et al. (2019), identified 32 orchid mycorrhizal fungi in North America with the internal transcribed spacer 47 region, and used shallow genome sequencing to functionally annotate these isolates, one of which was Tulasnella calospora Growth stages noted as in progress are symbiotic germination trials that are currently underway. Seed origin listed as MBOT were from collections made by the Missouri Botanical Garden under permit

Conclusions
For ex situ conservation and educational purposes, using Tulasnella calospora 266 may have an immediate impact for use in botanical gardens frequented by the general public. The ongoing use of this fungus at the Missouri Botanical Garden has already generated orchids to the blooming stage which are currently on display. For critically rare species propagated in this manner, cross-pollinating flowers by hand between orchids in captivity could potentially generate a new source of seed for conservation purposes, reducing or circumventing the need to collect seed from wild populations.