Green Globular Body Induction and Plantlet Regeneration of Endangered Fern Adiantum reniforme var. sinense

As an endemic pteridophyte with considerable ornamental importance and medicinal value, Adiantum reniforme var. sinense is a dwarfish unifoliate evergreen fern, mainly distributed in Asia and first reported its existence along both sides of the upstream Yangtze River in China (Lin, 1980). Due to low reproductive ability through its spores and lack of cultivation studies, this fern is rarely grown commercially worldwide. Moreover, the wild population of A. reniforme var. sinense has gradually become scarce due to sparse distribution, extractivism and destruction of natural habitats by the building of the Three Gorges Dam; thus, this fern is highly threatened and has been categorized as an endangered plant (Liu, 2007).
       
GGBs are unique tissue emerging from explants in the tissue culture of pteridophytes. They are usually induced with runner tips, leaves and leaf primordia of sporophytes as explants In vitro (Bertrand et al., 1999; Higuchi et al., 1987; Liao and Wu, 2011; Li et al., 2015; Yu et al., 2017; Yu et al., 2021). GGBs exhibit great potential to generate many sporophytes due to the presence of several meristems within a single GGB. Consequently, GGBs are widely applied in the production and species conservation of pteridophytes (Higuchi et al., 1987; Fernández, 2018). In A. reniforme var. sinense, the gametophytic GGBs have been induced with aseptic prothalli. The resulting GGBs differentiated into gametophytes, which is inconsistent with the purpose of sporophyte regeneration (Huang et al., 2008). Regarding sporophytic GGBs originating from embryos and their subsequent multiplication and differentiation, they haven’t been recorded in previous studies of ferns.
       
In this study, GGBs were induced successfully with the embryos developing from prothalli and the effects of various components in the medium, including basal salts, sucrose concentrations, gelling agents and plant growth regulators (PGRs), on the GGB proliferation were experimented. Also, the histological characteristics of the embryo and GGB were analyzed by microtome sections. Through the induction of GGBs with an embryo, as well as the proliferation of GGBs, an in vitro reproduction system of A. reniforme var. sinense is established. This is the first report that GGB from the sporophytic embryo has been observed in A. reniforme var. sinense.

The present investigation was carried out during year 2018-2020 at Orchid Biotech Lab, Department of Horticulture, National Ilan University, Yilan, Taiwan.
 
Plant material
 
The leaves with Adiantum reniforme var. sinense sporangia were cut and washed with clean water, followed by sterilization with 75% alcohol and 0.6% sodium hypochlorite in sequence. These leaves were rinsed several times with sterile water and dried with filter paper. The leaves were sliced and sown in a basal medium containing MS (Murashige and Skoog, 1962) (Duchefa Biochemie, Haarlem, The Netherlands) with 30 g L^-1 sucrose (Sigma Aldrich, St. Louis, MI, USA), 2 g L^-1 Bacto-tryptone (BD, Biosciences, Franklin Lakes, NJ) and 4 g L^-1 gelrite (Kelco, Rahway, NJ) with the pH adjusted to 5.8. Sterile gametophytes were formed after spore germination within a month and small embryos were produced at the centre of several prothallium (Fig 1a).
 

 
GGB induced with embryos
 
Sterile prothallium (approximately 0.2 cm in length) with a single embryo was cultured in MS medium with 30 g L^-1 sucrose, 2 g L^-1 Bacto-tryptone and 4 g L^-1 gelrite and adjusted pH to 5.8. Different concentrations of BA (0, 0.1, 0.5 and 1 mg L^-1) and NAA (0, 0.15, 0.7 and 1.4 mg L^-1) were added for combinatorial experiments. Every treatment was repeated five times, each with the culturing of one explant. After 60 days, GGB number and weight produced by a single embryo under different PGRs were observed.
 
GGB proliferation under different basal medium
 
The GGBs induced with the Adiantum reniforme var. sinense embryo were used as explants, each with a weight of approximately 0.02 g. The basal media were prepared by supplementing MS medium with 30 g L^-1 sucrose, 2 g L^-1 Bacto-tryptone and 4 g L^-1 gelrite and adjusting the pH to 5.8. Then, the basal salts of the media were adjusted to 1/2 MS, full MS and 3/2 MS (2.2, 4.4 and 6.6 g L^-1, respectively), as well as 1/2 Hyponex (Hyponex Corporation, Marysville, OH), full Hyponex and 3/2 Hyponex (1, 2 and 3 g L^-1) for the experiments. Every treatment was repeated twelve times. After 60 days, the total fresh weight of GGBs and protruding shoot number were observed and recorded separately.
 
GGB proliferation under different sucrose concentrations
 
The GGBs each with a weight of approximately 0.02 g were used as explants. The basal media were prepared by supplementing MS medium with 2 g L^-1 Bacto-tryptone and 4 g L^-1 gelrite and adjusting the pH to 5.8. The sucrose content of the basal medium was adjusted to 0, 15, 20, 30 and 45 g L^-1 for the experiments. Every treatment was repeated twelve times. After 60 days, the total fresh weight of GGBs and protruding shoot numbers were recorded separately.

GGB proliferation under different gelling agents
 
The GGBs each with a weight of approximately 0.02 g were used as explants. The basal media were prepared by supplementing MS medium with 30 g L^-1 sucrose, 2 g L^-1 Bacto-tryptone and adjusting the pH to 5.8. The gelling agents in the basal medium were adjusted to the gelrite (3, 4 and 5 g L^-1) and agar (6, 7 and 8 g L^-1) (Hispanagar, Burgos, Spain) for the experiments. Every treatment was repeated four times. After 60 days, the total fresh weight of GGBs and protruding shoot number were observed and recorded separately.
 
GGB proliferation under different PGRs
 
The GGBs were used as explants, each with a weight of approximately 0.02 g. The basal media consisted of 3/2 Hyponex, 30 g L^-1 sucrose, 2 g L^-1 Bacto-tryptone and 4 g L^-1 gelrite, with the pH adjusted to 5.8. Different types and concentrations (1, 5 and 10 mg L^-1) of PGRs, including 2,4-D (2,4-Dichlorophenoxyacetic acid), BA, KT, NAA and thidiazuron, were added. A basal medium without PGRs was used as the control. Every treatment was repeated five times. After 60 days, the total fresh weight of GGBs and protruding shoot number were observed and recorded separately.
 
Preparation of paraffin sections of embryos and GGBs
 
The embryos and GGBs of Adiantum reniforme var. sinense at different growth stages were cut into several pieces at appropriate sizes and placed in a mixed solution of F.P.A. (formalin 5%, propionic acid 5% and ethyl alcohol 50-70%) for fixing, which was followed by 3-7 days of air extraction based on the tissue size. A series of dehydration procedures were then performed with the mixture of tertiary butanol (t-butanol) and ethanol and 0.5% safranin staining was finally performed. The dehydrated material was placed in an oven at 60-62°C, thereby gradually increasing the concentration of paraffin in the solution. After paraffin penetration, the block was cut into thin sections of 10-20 μm using a rotary microtome (Tominga, Taiwan). The paraffin sections were spread onto glass slides and air-dried for 2 days. Finally, the paraffin strips were dissolved in xylene and stained with 0.5% safranin and 0.1% fast green stain, followed by mounting and air drying. Then, under an optical microscope, the histological examination of the embryo and GGBs at different stages was performed (Cason, 1974).
 
Statistical analysis
 
The data calculations and statistics for the experiments were performed with Microsoft Excel 2013 software and the Statistic Analysis System (SAS) software package was applied to Duncan’s multiple range test. The level of significant difference between treatments was defined at 5%.

GGB induction with embryo
 
The sterilized spores of A. reniforme var. sinense germinated and produced many prothallia on the basal medium, which sporadically generated a yellowish embryo to grow into a single sporophyte as in the natural life cycle of the fern (Fig 1a-b). When these embryos were isolated and cultured in the medium containing low amounts of BA or NAA, GGBs were induced from the embryos within 14 days, which subsequently differentiated into multiple ferns on the surface (Fig 1c-d). As seen in Table 1, there is no GGB formation from embryos cultured in the PGR-free medium. When media contained NAA and/or BA, embryos produced GGBs at every treatment with various efficiency. Specifically, the media supplemented with the lowest amounts of BA (0.1 mg L^-1) and NAA (0.15 mg L^-1) had the highest GGB induction from embryos with a maximum fresh weight of 0.29 g. By contrast, the growth of A. reniforme var. sinense gametophytes had the highest fresh weight of 0.67 g in the medium containing the highest BA (1 mg L^-1) in this experiment.
 

 
Influence of PGRs, salts, sucrose and gelling agents on the GGB proliferation
 
Fig 2 shows the PGRs exerted an evident influence on the proliferation of GGBs. In terms of auxin, the increasing amount of NAA promoted the proliferation of GGBs and the fresh weight of GGBs was highest in the case of 10 mg L^-1 NAA. Though 2, 4 D played a promotive effect on proliferation at 1 mg L^-1, the fresh weight of GGBs declined with the increasing amount of 2, 4-D. However, the growth of GGBs was greatly inhibited by all kinds of cytokinin tested in this study. Eventually, it caused the death of GGBs by BA and KT greater than 5 mg L^-1 (Fig 2).
 

       
As seen in Fig 3, Hyponex salts significantly outperformed MS salts in terms of GGB proliferation and the number of emerging shoots. The highest fresh weight of GGBs (1.72 g) was obtained in media containing 2 g L^-1 Hyponex salts, which was 3.8 times greater than the maximum fresh weight (0.45 g) obtained in media containing 4.4 g L^-1 MS salts. When GGBs were cultured on a medium containing 2 or 3 g L^-1 Hyponex salts, all of them regenerated plantlets (Fig 3). The highest average number of shoots was obtained in 3 g L^-1 Hyponex salts, wherein 7.25 shoots were produced. In contrast, no plantlet regenerated from GGBs in the MS medium (Fig 3).
 

       
The results of the experiment on the influence of sucrose concentrations are shown in Fig 4. The best result was obtained in the medium containing 15 g L^-1 of sucrose, with the average fresh weight and number of shoots of GGBs reaching the highest of 1.01 g and 2 shoots respectively. In contrast, GGBs were subject to slow growth and no plantlets regeneration in the sucrose-free medium. However, both the fresh weight and the number of plantlets declined with the increasing concentration of sucrose in the medium (Fig 4). Regarding the influence of gelling agents on the GGB proliferation, gelrite distinctly outperformed agar. Specifically, after 60 days of cultivation, the average fresh weight of GGBs in medium containing 4 g L^-1 gelrite was approximately twice as high as that in medium with agar (Fig 5). In addition, GGBs failed to differentiate into shoots in medium with either gelrite or agar (data not shown).
 

 

 
Paraffin sections of an embryo and GGB of Adiantum reniforme var. sinense
 
Both embryo and GGB were sectioned and the internal structure was observed microscopically. As revealed in the histological examination of the embryo, a single meristem was observed in the centre of the embryo (Fig 6a). Each embryo continues to grow into a single fern as in natural conditions. However, the internal structure of the GGB revealed the existence of many meristems of different developing stages connected with the annular ring. With the development of meristems in GGB, many regenerated plantlets were formed on the surface of the swollen GGB (Fig 6b). The observation showed that GGB could be distinguished from the embryo.
 

       
GGBs are unique structures induced by sporophyte tissue during in vitro culture. In 1987, Higuchi et al. first described GGBs in the tissue culture of Nephrolepis cordifolia Prsel. Since then, GGBs have been induced with runner tips, leaves and shoot primordia in several ferns (Higuchi et al., 1987; Liao and Wu, 2011; Li et al., 2015; Yu et al., 2017). In this study, GGBs were induced for the first time with an Adiantum reniforme var. sinense embryo.
       
It was shown that all embryos produced GGBs easily in a medium containing low concentrations of BA (0.1 mg L^-1) and NAA (0.15 mg L^-1). The induction rate of GGBs was not increased by raising the amounts of BA, which is different from the findings for GGB induction with higher concentrations of cytokinin BA (1 mg L^-1) in Cibotium barometz and Pteris aspericaulis var. tricolor (Li et al., 2015; Yu et al., 2017). The combinations of auxin and cytokinin at various ratios were used successfully in previous studies to promote GGB proliferation (Bertrand et al., 1999; Li et al., 2015; Amaki and Toda, 2010). However, in this study, only auxin of 2-4, D (1 mg L^-1) or NAA (10 mg L^-1) played the promotive role in the GGB proliferation, while cytokinins (BA, KT and TDZ) inhibited proliferation and eventually leading to the death of the GGBs. Specifically, a low amount of NAA (0.15 mg L^-1) is needed for GGB induction while a high amount of NAA (10 mg L^-1) is needed for subsequent proliferation. The optimum amount of NAA for GGB proliferation is approximately 66 times higher than that for GGB induction.
       
As reported in previous studies, most GGB related investigations were conducted in a medium containing the inorganic salts and vitamins of MS (Amaki and Toda, 2010). However, our results showed that GGBs were subjected to low proliferation and no plantlet regeneration in the MS medium, while the Hyponex medium markedly increased the fresh weight and plantlet number of GGBs. The promoting effects of Hyponex medium surpassed that of PGRs, sucrose and gelling agents, which were the most influential factors in this study. The promoting effect of the Hyponex medium on the GGB proliferation and differentiation might be correlated to the relatively high content of trace elements and microelements in the medium. Despite the wide application of Hyponex in the tissue culture of orchids (An et al., 2021; Chen et al., 2015; Ket et al., 2004), there have been no reports on the use of the Hyponex medium in the tissue culture of ferns.
       
Sucrose serves as a carbon source, as well as a crucial component for maintaining osmotic potential in the medium. In the experiments concerned with sucrose concentrations, the best result was obtained in a medium containing 15 g L^-1 of sucrose. The GGB proliferation and plantlets regeneration was greatly promoted by the presence of sucrose in the medium, however, the promotive effect was reduced in a medium containing excessively high concentrations (> 15 g L^-1) of sucrose. Similar results have been obtained for other fern species (Amaki and Kadokura, 2009).
       
From histological observation, embryos and GGBs of Adiantum reniforme var. sinense differed in their internal structures. The embryo had only one meristem and subsequently differentiated into a single seedling. In contrast, the GGB had many meristems and multiple shoots are directly differentiated from these meristems and regenerated plantlets on the surface of the swollen GGB. It was shown that the vascular supply of shoots was found to be continuous with the vasculature of the GGB. The appearance of a vascular tissue connection between the regenerating structure and the explant is a major feature of organogenesis (Haensch, 2004; Wicart et al.,1984). Similar results were obtained from paraffin sections of GGBs of Nephrolepis cordifolia Prsel, Pteris aspericaulis var. tricolor, Asplenium nidus L. and Cibolium baromelz (Bertrand et al., 1999; Li et al., 2015; Yu et al., 2017; Higuchi and Amaki, 1989).

This paper reported a highly efficient plant regeneration through GGB-mediated protocol in Adiantum reniforme var. sinense, which was the first study on embryos as the explant in fern. Specifically, the experiments revealed that a 100% induction rate could be obtained under BA (0.1 mg L^-1) and NAA (0.15 mg L^-1). For the maximum proliferation of GGBs the medium containing 2 g L^-1 Hyponex, 15 g L^-1 sucrose, 4 g L^-1 gelrite and 10 mg L^-1 NAA were most favourable. However, cytokinin such as BA, KT and TDZ have adverse effects on GGB proliferation. In addition, paraffin sections were performed to explore the difference between embryos and GGBs in Adiantum reniforme var. sinense.

All authors contributed to the study’s conception and design. Cheng-Wei Lin, Chia-Hung Shih and Chien-Yuan Kao performed material preparation, data collection and analysis. Chien-Yuan Kao wrote the first draft of the manuscript and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

This research received no external funding.

Not applicable.

Our research complies with the Convention on Biological Diversity and the Convention on the Trade in Endangered Species of Wild Fauna and Flora. The plant material “Adiantum reniforme var. sinense” was bought from a local seller legally.

The authors declare no conflict of interest.