Mesoamerican Cypripedium: Mycorrhizal Contributions to Promote Their Conservation as Critically Endangered Species

In the valuable orchid genus Cypripedium, the section Irapeana consists of a distinctive group of Mesoamerican species that is formed by Cypripedium dickinsonianum Hágsater, C. irapeanum Lex., and C. molle Lindl. All lady slipper orchids exhibit different distributions and abundances. Data analysis that used herbarium accessions and field investigations indicated that the habitats of these three species have been dramatically reduced. Prospecting for suitable habitats based on climatic, vegetation, and soil parameters allows us to predict potential distributions. Conservation strategies, such as ex situ propagation by asymbiotic and symbiotic approaches, have indicated that the culture media used are a determining factor for seedling development. Mycorrhizal isolates play a main role in the compatibility and further development of germinated seeds. The fungi isolated from adult plants belong to two different families, which makes it possible that widely distributed C. irapeanum populations will be fungal-specific as well as restricted for C. molle. Root mycorrhization patterns occur high on the secondary roots. In contrast with other species of the genus, in situ germination can occur over a short period of two months, but we have documented periods as long as ten years. Cypripedium is a highly problematic genus for ex situ conservation because the germination requirements and cultures are poorly documented, and there is great urgency for in situ conservation to develop strategies for identifying hotspot habitats and actualize the protection status to avoid extinction of this genus.


Introduction
The terrestrial genus Cypripedium is one of the most appreciated in the orchid family and is found in the natural environment, botanical gardens, natural parks, and scientific and private collections [1,2]. The origin of the name comes from the Greek root Cypris that refers to Aphrodite's sandal pedilon because of the globose-sac-shaped flower lips. Along with modifications of sepals and petals, columns with two anthers and the presence of staminodes make them unique relative to all other orchids. The genus contains approximately 50 species that are distributed in the Northern Hemisphere in mountain woodlands, grasslands, shrubs, or swamps, and is mainly associated with Quercus or Pinus forests [2][3][4][5][6][7][8][9]. All species are endangered by overcollection, anthropogenic activities, and climate change [10,11].
Mesoamerica has been postulated to be the origin center of the genus Cypripedium with the section Irapeana, which is a sister clade to the other members of this taxa [2,4,7,[12][13][14]. The section contains three species, namely, C. dickinsonianum Hágsater, C. irapeanum Lex., and C. molle Lindl. [8]. C. irapeanum from Irapeo, which is located in Michoacán State of Mexico, is the type species, but the type locality no longer exists [13][14][15]. There is remarkable morphological variability among C. irapeanum populations, which are likely to The Mesoamerican species have dissimilar distributions. C. dickinsonianum is restricted to small populations in the Mexican states of Queretaro and Chiapas and in Guatemala and grows on light slopes with Juniperus, and is sometimes sympatric with C. irapeanum [13,14,18]. In contrast, C. irapeanum is widely distributed in shrubs and grasslands or is associated with Quercus and Pinus in several Mexican states (e.g., Chiapas, Puebla, Mexico, Morelos, Michoacán, Guerrero, Colima, Nayarit, Sinaloa, and Veracruz), and is found in some locations in Guatemala. Cypripedium molle is restricted to the Mexican state of Oaxaca and grows in similar forest associations as C. irapeanum, but with a difference wherein it develops close to disturbed sites such as roadsides [19] (Figure 2). The Mesoamerican species have dissimilar distributions. C. dickinsonianum is restricted to small populations in the Mexican states of Queretaro and Chiapas and in Guatemala and grows on light slopes with Juniperus, and is sometimes sympatric with C. irapeanum [13,14,18]. In contrast, C. irapeanum is widely distributed in shrubs and grasslands or is associated with Quercus and Pinus in several Mexican states (e.g., Chiapas, Puebla, Mexico, Morelos, Michoacán, Guerrero, Colima, Nayarit, Sinaloa, and Veracruz), and is found in some locations in Guatemala. Cypripedium molle is restricted to the Mexican state of Oaxaca and grows in similar forest associations as C. irapeanum, but with a difference wherein it develops close to disturbed sites such as roadsides [19] (Figure 2).

Mesoamerican Slipper Orchids: Unique and Critically Endangered
Many factors increase the risk of loss of these species and developing strategies for their conservation is extremely urgent. International treatment CITES (Convention on International Trade in Endangered Species of Wild Fauna and Flora) considers C. dickinsonianum as endangered (EN), C. irapeanum as vulnerable (VU), and C. molle as near threatened (NT) [20]. Mexican policies recognize C. dickinsonianum as protected and C. irapeanum as EN, while the status of C. molle is unconsidered [21]. However, the actual endangered status of all Mesoamerican species is more serious: by using georeferenced entries from Mexican National herbarium records from 1954 to 2013 (176 entries) (Salazar-Chávez, G., personal database), data from previous papers [18], and confirmed records from the website inaturalist.org [22], we obtained 11 records for C. dickinsonianum, 102 for C. irapeanum, and 57 records for C. molle. We determined by using geographic information system techniques, satellite images, georeferenced databases [23], and ecological niche modeling projections [24] that only 3, 43, and 27 of the records, respectively, maintained their original habitats, with a loss of 58% for both C. dickinsonianum and C. irapeanum and 53% for C. molle ( Figure 3). Most of the remnants confirmed that populations are endangered due to their proximity to urban sites or sites with probable habitat transformation: from field investigations conducted in 2018 and 2019, we visited 27 recorded C. molle populations, and only 3 of them could be found, while the others had been destroyed at the end of 2019 by Agave angustifolia plantations used for mezcal production, an alcoholic drink, whose consumption has popularized since the Appellation of Origin was obtained by only the Mexican state of Oaxaca [25]. In the case of C. irapeanum, two of five studied populations were subjected to pillage of flowers and plants. Additional threats come from habitat transformation due to road expansion and the establishment of illegal trash dumps. C. dickinsonianum is the most vulnerable, and the remaining four populations, which represent half of the total number of records, are near highways and urban centers.

Mesoamerican Slipper Orchids: Unique and Critically Endangered
Many factors increase the risk of loss of these species and developing strategies for their conservation is extremely urgent. International treatment CITES (Convention on International Trade in Endangered Species of Wild Fauna and Flora) considers C. dickinsonianum as endangered (EN), C. irapeanum as vulnerable (VU), and C. molle as near threatened (NT) [20]. Mexican policies recognize C. dickinsonianum as protected and C. irapeanum as EN, while the status of C. molle is unconsidered [21]. However, the actual endangered status of all Mesoamerican species is more serious: by using georeferenced entries from Mexican National herbarium records from 1954 to 2013 (176 entries) (Salazar-Chávez, G., personal database), data from previous papers [18], and confirmed records from the website inaturalist.org [22], we obtained 11 records for C. dickinsonianum, 102 for C. irapeanum, and 57 records for C. molle. We determined by using geographic information system techniques, satellite images, georeferenced databases [23], and ecological niche modeling projections [24] that only 3, 43, and 27 of the records, respectively, maintained their original habitats, with a loss of 58% for both C. dickinsonianum and C. irapeanum and 53% for C. molle ( Figure 3). Most of the remnants confirmed that populations are endangered due to their proximity to urban sites or sites with probable habitat transformation: from field investigations conducted in 2018 and 2019, we visited 27 recorded C. molle populations, and only 3 of them could be found, while the others had been destroyed at the end of 2019 by Agave angustifolia plantations used for mezcal production, an alcoholic drink, whose consumption has popularized since the Appellation of Origin was obtained by only the Mexican state of Oaxaca [25]. In the case of C. irapeanum, two of five studied populations were subjected to pillage of flowers and plants. Additional threats come from habitat transformation due to road expansion and the establishment of illegal trash dumps. C. dickinsonianum is the most vulnerable, and the remaining four populations, which represent half of the total number of records, are near highways and urban centers. Due to the swift loss of natural habitats for these Mesoamerican species, ide of potential habitats was achieved to identify possible new locations to find or r populations in the future. Some climatic factors may strongly determine the tions [18], so ecological preferences and ecological niche modeling have been using the WorldClim-Global Climate Data, which include the temperature s mean annual precipitation, annual temperature range, mean temperature in quarter, annual mean precipitation, and precipitation seasonality [26]. We obta thetical projections for the three species by using the maximum entropy algori lated with MAXENT software [27]. To estimate the model, we used herbarium records obtained from the website inaturalist.org [22], climatic variables, soil layers (e.g., total carbon and nitrogen, bulk density, water-holding capacity, mo tent, and wilting point) [28], altitudinal records, and vegetation types [23] (Tab Due to the swift loss of natural habitats for these Mesoamerican species, identification of potential habitats was achieved to identify possible new locations to find or reintroduce populations in the future. Some climatic factors may strongly determine their distributions [18], so ecological preferences and ecological niche modeling have been studied by using the WorldClim-Global Climate Data, which include the temperature seasonality, mean annual precipitation, annual temperature range, mean temperature in the coldest quarter, annual mean precipitation, and precipitation seasonality [26]. We obtained hypothetical projections for the three species by using the maximum entropy algorithm calculated with MAXENT software [27]. To estimate the model, we used herbarium and online records obtained from the website inaturalist.org [22], climatic variables, soil conditions' layers (e.g., total carbon and nitrogen, bulk density, water-holding capacity, moisture content, and wilting point) [28], altitudinal records, and vegetation types [23] (Table S1), and these data had previously been transformed to compatible formats in ArcGIS© [29].
The resulting projections of the potential habitats for each species indicate that C. irapeanum and C. dickinsonianum are sympatric in some habitats, whereas certain potential distribution habitats have not yet been recorded (Figure 4a,b). For C. molle, the estimated distribution is restricted to the gap located between the union of the eastern and south Sierra Madre located in the physiographic province of Oaxaca s Sierras Centrales (Figure 4c). The inclusion of soil conditions generates more robust models for suitable habitat projections; however, it is not clear whether soil conditions have a direct impact on plant requirements or the associated microorganisms, such as orchid mycorrhizal fungi (OMF), as we conduct further research [30].
Plants 2022, 11, x FOR PEER REVIEW 5 distribution habitats have not yet been recorded (Figure 4a,b). For C. molle, the estim distribution is restricted to the gap located between the union of the eastern and s Sierra Madre located in the physiographic province of Oaxaca´s Sierras Centrales (Fi 4c). The inclusion of soil conditions generates more robust models for suitable habitat jections; however, it is not clear whether soil conditions have a direct impact on p requirements or the associated microorganisms, such as orchid mycorrhizal fungi (OM as we conduct further research [30].

Underground Growth Pattern in Different Habitats
We analyzed the underground rhizome growth after conducting the field investigations for three habitats: two for C. irapeanum (Quercus forest in the states of Mexico, Morelos, and Puebla, and Tropical Deciduous Forest (TDF) in Veracruz State) and one for C. molle (Quercus forest in Oaxaca State). Both species show consistent patterns of underground root growth that are generally less than 10 cm-deep and are located between an undecomposed litter layer and a layer with high organic matter content that is very similar to the growth of other Cypripedium species [4,31] ( Figure 5). Mesoamerican species develop short rhizomes and an annual stem with several large roots that can persist for years by storing carbohydrates in the form of starch. For other species in the genus, the root systems may live for nearly 14 years, which allows these plants to remain dormant for long periods without aerial shots until adequate conditions for vegetative growth are met. Many years (e.g., ca. 7-16) are necessary to develop full-grown flowering plants from seedlings of lady slipper orchid species [4,[32][33][34]. For C. irapeanum, the populations in Quercus forest grow under layers rich in organic matter on volcanic-derived acidic soils (pH of 5.6), while populations in TDF grow in poor soils, with high levels of calcium, little organic matter, and a pH of approximately 7.5 [14,35]. Both populations develop under seasonally dry environments. It is worth mentioning that in five years of study on these populations, only one germination event was recorded for each population (Figure 6a,b). The incidence of in situ germination has been estimated to be extremely low (e.g., approximately 0.001%) [8] and can occur over a period of two months (Figure 6c-e) or ten years after seed dispersion (Figure 6f-h). It is not clear whether these scarce germination events depend on the potential of a habitat to provide a germination niche (for example, until colonization by compatible symbiotic fungi) or on the natural long-term dormancy of seeds as a survival mechanism. OMFs are needed for germination and carbon acquisition throughout life to support survival during dormant states, quite common in the genus Cypripedium [4,32,[36][37][38][39]. In general, the in situ germination requirements for this genus are poorly understood, and long-term studies are needed [40]. Both C. irapeanum and C. molle grow on different soils, such as reddish, clayey, lateritic, limestone soils or volcanic-derived soils, and this indicates that a broad range of diverse conditions can meet the requirements for germination and increases the potential habitats and probability of establishing new populations [41]. The genus is considered highly dependent on mycorrhizal fungi, and this could be the reason why asymbiotic propagation methods have not been completely established for many species [42][43][44].

Mycorrhiza Studies for Conservation Purposes
Cypripedium species, like other orchids, have two main symbiotic relationships: with insects for pollination and seed production and with OMF for germination and nutrient transfer. For pollination, there are several genera of insects that can function as Cypripedium pollinators, and despite the different reports on these insects, few actual successful pollinia removal events have been recorded [45][46][47]. In the case of Mesoamerican C. irapeanum and C. molle, Halictideae wasps have been considered as pollinators for both, and the only pollinator for C. irapeanum is considered as Lasioglossum nyctere, while C. dickinsonianum has been recognized as self-pollinating [13,14].
However, mycorrhizal symbiotic associations have more effectively studied for several species in adult plants, and scarce and erratic patterns of root colonization [37,48]. We confirm two main patterns of mycorrhiza development for Mesoamerican C. irapeanum and C. molle: On the main roots, colonization is scarce and can remain for years with highly degraded hyphal coils and large numbers of starch granules [37,48,49]. In contrast, the secondary roots, which are usually short and numerous (ca. 2-3 cm), lack starch reserves because they are actively growing and are highly colonized by hyphal coils in diverse stages of digestion, while most of them are undigested (Figure 7). Seedlings that develop by in situ germination of both species are always colonized with the same pattern as the secondary roots.

Mycorrhiza Studies for Conservation Purposes
Cypripedium species, like other orchids, have two main symbiotic relationships: with insects for pollination and seed production and with OMF for germination and nutrient transfer. For pollination, there are several genera of insects that can function as Cypripedium pollinators, and despite the different reports on these insects, few actual successful pollinia removal events have been recorded [45][46][47]. In the case of Mesoamerican C. irapeanum and C. molle, Halictideae wasps have been considered as pollinators for both, and the only pollinator for C. irapeanum is considered as Lasioglossum nyctere, while C. dickinsonianum has been recognized as self-pollinating [13,14].
However, mycorrhizal symbiotic associations have more effectively studied for several species in adult plants, and scarce and erratic patterns of root colonization [37,48]. We confirm two main patterns of mycorrhiza development for Mesoamerican C. irapeanum and C. molle: On the main roots, colonization is scarce and can remain for years with highly degraded hyphal coils and large numbers of starch granules [37,48,49]. In contrast, the secondary roots, which are usually short and numerous (ca. 2-3 cm), lack starch reserves because they are actively growing and are highly colonized by hyphal coils in diverse stages of digestion, while most of them are undigested (Figure 7). Seedlings that develop by in situ germination of both species are always colonized with the same pattern as the secondary roots.
There have only been three successful attempts to isolate OMF on Cypripedium, for C. macranthos var. rebunense [43] and Mesoamerican C. irapeanum [35] and C. molle (Moreno-Camarena and Ortega-Larrocea, submitted). The mycorrhizal endophytes that were obtained from the roots of adult C. irapeanum plants from two habitats belong to the anamorphic genus Epulorhiza (hyphae less than 4 µm, pearly monilioid cells, creamy submerged colonies on PDA, and slow growth rates ca. 0.2 mm per day). Endophytes that were obtained from C. molle show characteristics of the anamorph Ceratorhiza (hyphae of more than 4 µm, barrel-shaped monilioid cells, brownish colonies on PDA, aerial mycelium, and growth rates of 0.5 mm/day) ( Figure 8, Table 1) [50]. The isolation and long-term conservation of these isolates was achieved only on Green Pea Agar medium [51], AWA (Acidic Water Agar), and FIM (Fungal Isolation Medium) [52]. In media such as PDA (potato dextrose agar) or OMA (oatmeal agar), which are usually used for OMF cultivation, no long-term growth was achieved, and the isolates lost viability in both cases. There have only been three successful attempts to isolate OMF on Cypripedium, for C. macranthos var. rebunense [43] and Mesoamerican C. irapeanum [35] and C. molle (Moreno-Camarena and Ortega-Larrocea, submitted). The mycorrhizal endophytes that were obtained from the roots of adult C. irapeanum plants from two habitats belong to the anamorphic genus Epulorhiza (hyphae less than 4 µm, pearly monilioid cells, creamy submerged colonies on PDA, and slow growth rates ca. 0.2 mm per day). Endophytes that were obtained from C. molle show characteristics of the anamorph Ceratorhiza (hyphae of more than 4 µm, barrel-shaped monilioid cells, brownish colonies on PDA, aerial mycelium, and growth rates of 0.5 mm/day) ( Figure 8, Table 1) [50]. The isolation and long-term conservation of these isolates was achieved only on Green Pea Agar medium [51], AWA (Acidic Water Agar), and FIM (Fungal Isolation Medium) [52]. In media such as PDA (potato dextrose agar) or OMA (oatmeal agar), which are usually used for OMF cultivation, no long-term growth was achieved, and the isolates lost viability in both cases.    Since OMF isolation has been difficult for Cypripedium species, using molecular tools to identify the relevant fungi is usually conducted by using the fungal coils of adult roots. The genomic regions studied are the ITS (internal transcribed spacer), LSU (large subunit), and SSU (short subunit) [53][54][55]. Molecular identifications indicate that the genera are sometimes associated with distant phylogenetic fungal groups (Table S2) [56]. In addition, some of the identified fungal partners may be incidental inhabitants, pathogens, or temporal successors [36]. The most common OMFs associated with Cypripedium belong to the family Tulasnellaceae: Tulasnella cystidiophora, T. calospora, and T. deliquescens, which were found in 28, 12, and 5 species, respectively. The family Ceratobasidiaceae (Ceratobasidium cornigerum) has also been identified in C. californicum [37,44,48,57]. However, most of these identifications have been conducted on adult plants, since symbiotic germination is poorly documented and the identification of fungal germination promoters is needed [42,43,58].
To analyze the relationships among fungi that were isolated from the adult roots of C. irapeanum (Quercus forest in Mexico State and TDF in Veracruz State) and plantlets of C. molle (Quercus forest in Oaxaca State), we amplified their ITS regions by using the primer combination ITS 1/ITS 4 [36,37,48,51,57,59], and the resulting sequences were assembled and edited with Geneious (2021.0.3). The sequences are deposited in the GenBank-NCBI database. A search for the most similar sequences was conducted by using the BLAST algorithm [60,61], and alignments were conducted using MAFFT [62]. Both algorithms are contained in the Geneious software. Phylogenetic reconstruction was carried out using PHYML plugin [60,63] by using the maximum likelihood method and the Tamura-Nei model with bootstrap support of 1000 replicates (Figure 9).
Significant results were observed from the phylogenetic reconstruction. The isolates from C. molle plantlets belong to Ceratobasidiaceae in a clade that consists of terrestrial endophytes of Vanilla spp. from Puerto Rico and Cephalanthera rubra from France [64] (Figure 9). Conversely, the isolates of the widespread C. irapeanum were recovered from different habitats (e.g., Quercus forest and TDF) and belong to Tulasnellaceae in a clade that includes a terrestrial endophyte from Vanilla [50], which indicates a probable high level of specificity to a widely distributed soil generalist fungal clade (Figure 9). This behavior is opposite to that of the North American C. californicum, a soil specialist (always grows on serpentine soils) but is associated with various species of the genus Tulasnella [48]. The same is true for the terrestrial Dichromanthus that form specific associations with a particular mycorrhizal endophyte when the species are widely distributed in several habitats and soil conditions. However, in one restricted habitat, the species were associated with several clades of the same fungal family (López-Reyes and Ortega-Larrocea, personal communication). Significant results were observed from the phylogenetic reconstruction. The isol from C. molle plantlets belong to Ceratobasidiaceae in a clade that consists of terres endophytes of Vanilla spp. from Puerto Rico and Cephalanthera rubra from France [64] ( ure 9). Conversely, the isolates of the widespread C. irapeanum were recovered from ferent habitats (e.g., Quercus forest and TDF) and belong to Tulasnellaceae in a clade includes a terrestrial endophyte from Vanilla [50], which indicates a probable high lev specificity to a widely distributed soil generalist fungal clade (Figure 9). This behavio opposite to that of the North American C. californicum, a soil specialist (always grow serpentine soils) but is associated with various species of the genus Tulasnella [48]. The fact that plantlets of C. molle associate with Ceratobasidiaceae may indicate that the fungi from this family are germination promoters, as was proposed for C. calceolus [65]. This could explain why endophytes isolated from adult plants do not promote good germination, as has been observed in C. irapeanum [35] and previously in C. macranthos var. rebunense [51]. The isolation and identification of fungi from other populations and plant stages together with cultivation by asymbiotic and symbiotic methods would provide a better understanding of germination, specificity, fungal succession during plant lifetimes, distribution of populations, and microhabitat requirements to develop conservation strategies for these unique species.

Symbiotic and Asymbiotic Germination and Development
Seeds of Cypripedium spp. have tests that consist of a hydrophobic double-layer rich in lignin that allows them to float on water [66,67]. At maturity, these seeds have high concentrations of abscisic acid (ABA) that can delay germination [68,69] which are probably involved in long-term survival through the years and persist for at least four years to form seed banks in soils [4,42,70]. In situ germination relies on mycorrhizal fungi that can penetrate the testa and induce germination [42,43]. Symbiotic germination under in situ conditions is poorly studied in Cypripedium and probably occurs in spring or early summer at 2-5 cm soil depths in moisture-stable sites within a pH range from 5.3 to 8.1 [31]. In some species of the genus, germination is stimulated by cold (e.g., C. calceolus, C. lentiginosum, C. macranthos var. rebunense) because they grow in temperate regions [42,43,68,71].
In vitro germination of Mesoamerican species has been studied using symbiotic and asymbiotic approaches. Asymbiotic germination was tested on Phytamax TM (SIGMA) with sucrose 2% [72], Murashige and Skoog [73], Norstog [74], and oatmeal agar [52] at pH 5.6 after the seeds were stimulated by cold storage (4 • C) for four months to break dormancy (Moreno-Camarena and Ortega-Larrocea, in process). However, seedlings developed only on Norstog medium ( Figure 10). It is worth mentioning that while Asian or European Cypripedium species germinate at 4 to 54 months after sowing [42,58], C. irapeanum can germinate at 12-14 days after sowing (das) [35]. Germination development, in vitro or in situ, begins when the testa splits and the embryo swells, which gives rise to a protocorm with a promeristematic zone (Figure 10a-c). From this, the apex begins to grow, which is followed by root formation on the opposite point of the protocorm (Figure 10d-f). The protocorms are greenish and usually have long rhizoids [31,42], except for C. irapeanum, which does not develop rhizoids under asymbiotic in vitro conditions (Figure 10d-g). Roots emerge after shoot differentiation of two foliar sheets (Figure 10g). This root system continues to grow faster over the shoot leaves and develops into several small sheets and lateral shoots (Figure 10h-k).
Symbiotic germination was tested on C. irapeanum seeds that were obtained from a population located in an oak forest in Puebla State. Two mycorrhizal endophytes (e.g., Epulorhiza spp.) that were isolated from adult plants from two different sites (e.g., a Quercus forest from Mexico State and a TDF from Veracruz State) were tested, and they exhibited 35% and 20% germination rates, respectively (Table 2). While the isolate from Mexico State promoted a higher germination percentage, the Epulorhiza isolate from Veracruz promoted a more compatible and advanced developmental stage ( Figure 11). Shimura and Koda [43] state that symbiotic germination begins after some weeks of incubation in C. macranthos var. rebunense after cool storage, at which time the embryo is imbibed and develops rhizoids that serve as the entry points for fungal colonization. In the case of C. irapeanum, rhizoids did not develop under either asymbiotic or symbiotic conditions (Figure 11a-f), with mycorrhizal colonization likely beginning from the micropylar end (Figure 11d). This colonization is also suitable for water and nutrient absorption [75,76], even if suspensor cells are degraded in the C. calceolus and Mesoamerican hybrid Cypripedium × fred-mulleri to form spherical embryos [18,67]. However, neither of the two identified isolates allow protocorms to develop seedlings (stage 6), as in asymbiotic germination ( Figure 10). This incompatibility after symbiotic germination is observed when complete cells that form the protocorm are invaded by fungi, which prevents the apical meristematic cells from undergoing division and growth (Figure 11f-h). This phenomenon could be due to the origin of the isolates, since all were obtained from adult plants [51] or because Cypripedium compatibility with mycorrhizal fungi is low under natural conditions and resulted in the low recruitment observed. Shimura and Koda [43] achieved seedling formation with shoots after cultivation of symbiotic germinated protocorms in an antifungal medium, which indicated that autotrophic plants cannot regulate the symbiotic balance. Other attempts to obtain symbiotic plants have been reported and were without success [65,77]. Symbiotic germination was tested on C. irapeanum seeds that were obtained from a population located in an oak forest in Puebla State. Two mycorrhizal endophytes (e.g., Epulorhiza spp.) that were isolated from adult plants from two different sites (e.g., a Quercus forest from Mexico State and a TDF from Veracruz State) were tested, and they exhibited 35% and 20% germination rates, respectively (Table 2). While the isolate from Mexico State promoted a higher germination percentage, the Epulorhiza isolate from Veracruz promoted a more compatible and advanced developmental stage (Figure 11). Shimura and Koda [43] state that symbiotic germination begins after some weeks of incubation in C. macranthos var. rebunense after cool storage, at which time the embryo is imbibed and de- By comparing asymbiotic and symbiotic development under in vitro conditions, imbibition, rupture of the testa, and tissue differentiation (polarization) occur at similar times after sowing. The multiple lateral shoot induction observed in asymbiotic seedlings produced in vitro could be a consequence of growth under an artificial environment because in situ seedlings developed only one shoot. Unfortunately, there is not yet a successful protocol for propagating lady slipper orchids under either approach. irapeanum, rhizoids did not develop under either asymbiotic or symbiotic conditions (Figure 11a-f), with mycorrhizal colonization likely beginning from the micropylar end (Figure 11d). This colonization is also suitable for water and nutrient absorption [75,76], even if suspensor cells are degraded in the C. calceolus and Mesoamerican hybrid Cypripedium × fred-mulleri to form spherical embryos [18,67]. However, neither of the two identified isolates allow protocorms to develop seedlings (stage 6), as in asymbiotic germination ( Figure 10). This incompatibility after symbiotic germination is observed when complete cells that form the protocorm are invaded by fungi, which prevents the apical meristematic cells from undergoing division and growth (Figure 11f-h). This phenomenon could be due to the origin of the isolates, since all were obtained from adult plants [51] or because Cypripedium compatibility with mycorrhizal fungi is low under natural conditions and resulted in the low recruitment observed. Shimura and Koda [43] achieved seedling formation with shoots after cultivation of symbiotic germinated protocorms in an antifungal medium, which indicated that autotrophic plants cannot regulate the symbiotic balance. Other attempts to obtain symbiotic plants have been reported and were without success [65,77].

Final Remarks
Mesoamerican Cypripedium species represent an interesting group for study due to their phylogenetic importance as a sister ancestral group to the rest of the genus and because all Cypripedium species are endangered and have some degree of importance. As seen here, their mycorrhizal preferences differ from other species in the genus, and it is probable that they have ecological preferences and, consequently, their morphological variations could be derived from these. The high endangered status of the remaining populations prompts the generation of information on the OMF that promotes in situ germination to develop priority mechanisms for the conservation of hotspot habitats.
Since orchids depend on symbiotic associations, conservation of these species depends on an understanding of the biological and ecological factors that drive the distributions of mainly fungal partners [78,79]. Some aspects may be more determinant than others in species establishment, such as the distributions of suitable mycorrhizal germinationpromoting fungi [80]. This contribution provides a partial view of some aspects of the biology of Mesoamerican Cypripedium; however, the successful pollinators, in situ symbiotic germination, and OMF specificity remain unknown. The destruction of habitats urges ex situ conservation strategies that must include not only seed collections but also mycorrhizal fungi that can promote compatible germination or facilitate adaptation. Clarifying the structures of the mycorrhizal fungal communities and isolating germination-promoting fungi would help to determine suitable habitats to conserve or re-establish the studied species and the viability and persistence of in situ seed banks. Asymbiotic propagation in suitable media is a promising tool, but little is known about asymbiotic plant ex vitro adaptation. Both approaches would ensure long-term conservation of these highly endangered species before most of their habitats disappear.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/plants11121554/s1, Supplementary Table S1. Habitat characteristics of Cypripedium spp. remnant populations used for projection of niche ecological model. Supplementary Table S2. Some studies in Cypripedium endophytes using metabarcoding or traditional approaches, entries are ordered in chronological descendent order.

Data Availability Statement:
The data presented in this study are available in the text and supplemental data. The data presented in this study are available on request from the corresponding author.