Indirect establishment increases the chances of in vitro propagation of mosses occurring in the Cerrado - a new method

Abstract The use of micropropagation techniques is crucial for the conservation of endangered moss species and their reestablishment in nature. This study aimed to establish in vitro cultures of gametophyte fragments of ten species of Cerrado mosses. After disinfestation with alcohol and commercial bleach, moss explants were grown in Petri dishes containing Knop medium. The species Bryum argenteum, B. coronatum, Isopterygium tenerifolium, Leucobryum crispum, Pogonatum pensilvanicum, and Vitalia cuspidifera were successively established with efficiency rate ranging from 1 to 31.2%. However, no aseptic cultures were obtained for the species Barbula indica, Bryum densifolium, Fissidens flaccidus, and Sphagnum platyphylloides. Even though, a few contaminated explants of these species were able to develop and grow. Thus, all ten species were submitted to rescue techniques to establish cultures in aseptic conditions, from partially contaminated explants (indirect establishment). Consequently, the indirect establishment resulted in higher percentages of explant development, which enhanced the establishment of in vitro cultures for most of the species tested. This fact is especially important for conservation purposes, mainly for species whose material is sensitive or scarce. Therefore, indirect establishment as a new in vitro culture methodology was a viable form of propagating the bryophyte species listed in this research. This fact is essential for conservation purpose, especially for species whose material is sensitive or scarcer.


Introduction
The Brazilian bryoflora comprises 1,524 species with 11 anthrocerus, 633 hepatic and 880 mosses. There are 478 cataloged bryophyte species in the Cerrado, from which 63 classified as endemic (Costa & Peralta 2015). The Brazilian Cerrado is a tropical savanna and has been considered a hotspot for conservation because of its immense biodiversity and an enormous anthropic pressure (Myers et al. 2000). It is important to register that this tropical savanna has been severely threatened by agriculture, disorderly urbanization and tourism, which have a negative impact on its flora, and, consequently, on native bryophytes (Söderström 2006). Thus, Cerrado bryophytes are losing their niches, which makes essential to identify the endangered species and work towards their conservation. Therefore, the improvement and implementation of specific strategies to ensure the long-term conservation of Cerrado bryophytes is of paramount importance (Costa et al. 2016;Söderström 2006).
Despite that, some conservation techniques have been used in bryophytes. For instance, in situ conservation is the most common strategy applied to this class of plants. However, the efficiency of this approach is not always suitable for bryophytes conservation (Söderström 2006). Nevertheless, ex situ techniques, through axenic cultures, can be successfully applied for long-and medium-term conservation. Hence, cryopreservation, tissue culture, as well as genetic and molecular studies are suitable techniques for ecological research and environmental reintroduction (Duckett et al. 2004). For example, Ros et al. (2013) tested the in vitro culture for drought-tolerant moss species with subsequent cryopreservation, resulting in high survival rate.
In fact, the use of tissue culture techniques in mosses is not unusual. Several studies show the efficiency of these techniques mainly in the conservation of rare and endangered species (Reski 1998;Duckett et al. 2004). Furthermore, tissue culture has also been applied to the development of more efficient methods for propagation, commercial and industrial exploitation of mosses (Duckett et al. 2004;Cuvertino-Santoni & Montenegro 2013). It is worth mentioning the large-scale cultivation of Sphagnum palustre L. in bioreactors, to produce standardized and sustainable material for biomonitoring air quality through a technique termed as "Moss Bag" (Beike et al. 2014;Capozzi et al. 2016). This technique has also been applied with similar efficiency to other species of Sphagnum (Hu et al. 2018), as well as to Hypnum cupressiforme Hedw. (Ares et al. 2012), and Pseudoscleropodium purum (Hedw.) Fleisch. (Capozzi et al. 2017). Additionally, Bryum argenteum Hedw. has been micropropagated as a means of improving its production and commercial use to create greener environments (Liang et al. 2010).
In vitro culture of bryophytes has the advantage of being able to initiate cultures from either sexual (spores) or vegetative structures such as gametophyte fragments and gemmae (Duckett et al. 2004). This broad application of tissue culture is essential for species that are not found in the fertile phase of their life cycle (Longton & Miles 1982;Löbel & Rydin 2010). Besides, under specific conditions, the use asexual propagules for in vitro propagation of bryophytes may promote the formation of specialized resting propagules capable of withstanding extreme environmental conditions such as desiccation stress Rowntree et al. 2007).
Concerning the establishment of axenic bryophyte cultures, sporophytes are usually more easily disinfested than gametophytes. This feature is very limiting for moss micropropagation because of the fragility of the tissues and the symbiotic interactions with diverse microorganisms, which is hard to replicate in vitro (Sabovljević et al. 2003;Duckett et al. 2004). Despite that, once the in vitro culture has been established, the cultivation of these plants becomes relatively easy (Rowntree et al. 2011). A further limiting factor to initiate in vitro cultures is the low availability of gametophytic material from endangered species, especially when spores are not available. Likewise, the protocols for in vitro culture are often specific for each species, which further complicates the use of plant tissue culture techniques in this class of plants (Sarasan et al. 2006;Tacoronte et al. 2009).
In addition, researches on the propagation of non-endangered species are also important for rare and endangered species because they may provide essential information that could be applied to the conservation of bryophytes with limited material available (Rowntree 2006;Sarasan et al. 2006). In this sense, the objective of current work was to establish in vitro cultures initiated from gametophyte fragments of ten mosses species occurring in the Cerrado.
The identification of the species was carried out in the Laboratory of Cryptogams of the University of Brasília -UnB. Gametophytes of each species were separated and washed in running water for 2 min, and then, in sterile water for 2 min. Preferably, the younger fragments of gametophytes, such as apices, leaves, parts of stems and branch leaves and new shoots were used as explants for micropropagation.
For Barbula indica, Bryum argenteum and Pogonatum pensilvanicum, due to their peculiarities, pre-cultures were carried out, aiming the production of gametophytes suitable for surface disinfestation. The precultivation procedure consisted of non-aseptic culture in closed containers, in which a portion of each sample, with part of the substrate, was packed in transparent pots of 250 mL containing either wet commercial substrate (Bioplant ® plus, Nova Ponte, Brazil) or vermiculite. Afterward, the pots were closed and kept under natural light and room temperature (25 ºC) until the development of gametophytes.
The pots were watered every 30 days. After the cultivation period, 0.5 to 1 cm gametophytes were collected and submitted to disinfestation and subsequent inoculation.

Culture medium and conditions
The inoculation of the explants, as well as the subcultures, were carried out in Knop medium (Knop 1865), with the addition of the following components: 0.8 g.L -1 of calcium nitrate [Ca (NO 3 ) 2 .4 H 2 O]; 0.2 g.L -1 potassium nitrate (KNO 3 ); 0.2 g.L -1 of monopotassium phosphate (KH 2 PO 4 ); 0.2 g.L -1 of magnesium sulfate (MgSO 4 .7 H 2 O) and sterile water. The solidification of the solutions was carried out with the addition of 7.5 g.L -1 of agar, after adjusting the pH to 5.7 ± 0.1. Then, the media was sterilized at 121 ºC, 1.3 atm for 20 min, and 15 mL was poured into Petri dishes (90 × 15 mm).
The explants were previously disinfested in 70% alcohol for 5 s and solution of Sodium hypochlorite (commercial bleach containing 2% active chlorine, NaOCl) in three treatments: 1% for 5 min, 1% for 10 min, and 2% for 2 min (1% -5', 1% -10' and 2% -2'), and then rinsed 3x in sterile water. The disinfestation treatments were applied to the species according to the availability and the sensitivity of the collected material. For the species with abundant material, all three disinfestation treatments were applied. For the species with limited material, only one treatment was performed. Additionally, eight explants per plate were inoculated. The number of inoculated explants was defined according to the quantity and quality of the material obtained in the collections. The total explants, as well as the treatments performed in each species are described in Table 1.
The experiment was completely randomized, with eight explants per plate, and each plate constituted a repetition. The number of repetitions per treatment was according to the quantity and quality of plant material obtained in the collections. The total number of explants, as well as the treatments performed in each species, are described in Table 1.
After inoculation, all plates were sealed with plastic PVC film and kept in a room, with the temperature set at 25 (± 2 ºC), photoperiod of 16 h and light intensity of approximately 41 μmol.m -2 .s -1 .
The explants were evaluated on the 5 th and 10 th day after inoculation and classified as follows: developed without contamination or developed with contamination. Explants that showed apical growth, budding or growth of protonemata were considered as developed. Explants that did not show any development of plant tissue in 10 days of cultivation were discarded.
In the first evaluation (5 th day of cultivation), all the explants that showed some development, contaminated or not, were transferred to new plates (subculture plates). Then, the plates were again sealed and cultivated for another 5 days. In the second evaluation (10 th day), the explants that presented development were transferred to subculture plates.
It was considered established those in vitro cultures that, after disinfestation, at least one explant had been developed without a visible presence of microorganisms. The subcultures were based on developed explants with no signs of contamination and formed colonies suitable for propagation in subsequent subcultures. Figure 1 depicts the efficiency of disinfestation treatments with a solution of Sodium hypochlorite, which varied from 1 to 19% (1% for 5 min), 2.5 to 9% (1% for 10 min), and 3.5 to 31.2% (2% for 2 min). The use of hypochlorite in the disinfestation of fragments of bryophyte gametophytes has been shown to vary with the species, and with other parameters, according Sabovljević et al. (2003) Table 1 -Disinfestation treatments with commercial bleach (2% active chlorine) -1% concentration for 5 min (1% -5'), 1% for 10 min (1% -10'), and 2% for 2 min (2% -2'), bryophyte species and respective collection data and number of explants per treatment.

Results and Discussion
(*) precultivated species as observed in Aloina aloides. This is consistent with the results obtained in this investigation. Note that almost 50% of the species tested were not successfully disinfested with lower concentrations of hypochlorite ( Fig. 1; Tab. 2). Therefore, Liang et al. (2010) observed that the sterilization efficiency is dependent on the concentration and exposure time of explants. Furthermore, it was shown that disinfestation of Bryum argenteum explants achieved the highest efficiency when concentrations of 5% were used (Liang et al. 2010). Table 2 shows that the explants from 5 out of 6 species tested (80%) were better disinfested with 2% NaClO, which represented an average of 7.7% of all developed explants with no visible contamination. The only exception was Bryum densifolium that did not show decontamination with 2% NaClO (Fig. 1). Also, the highest percentage of decontamination was seen with Isopterygium tenerifolium, which had 19% of explants developed with no apparent contamination in treatment with NaClO 1% for 5 min (Fig. 1). Additionally, Sabovljević et al. (2002) also observed that variation in disinfestation efficiency depends on the morphology of the gametophyte, as well as on the position and density of the leaves. Some of the morphological characteristics could also be verified as limiting for disinfestation, as observed in Barbula indica that presents compact leaves, curved when dry and often with the presence of axillary gemmae (Sharp et al. 1994). Likewise, in Bryum argenteum leaves were ovate, concave, closely appressed and overlapped on the stem, which, usually, negatively interfered with the disinfestation (Sharp et al. 1994). Also, disinfestation difficulties were seen in Fissidens flaccidus that has a vaginant lamina in the leaves and in Sphagnum platyphylloides that presents empty cells that contain small openings named leucocysts (Costa et al. 2010). These morphological aspects, together with the fact that the species were collected directly from the soil, cemented pavement, and rocks, also contributed to difficult the disinfestation procedure as it was not possible to entirely remove these substrates from the samples.
Nevertheless, in six species out of 10 species, the in vitro establishment was obtained from the 1 st subculture: Bryum argenteum, B. coronatum, Figure 1 -Explant development rate of ten Cerrado mosses species in different treatments (1% for 5 min; 1% for 10 min; and 2% for 2 min), after 10 days of in vitro cultivation. The bars represent the standard error. * Precultivated material.
I. tenerifolium, L. crispum, P. pensilvanicum, and V. cuspidifera. The percentage of explants established of the species mentioned above varied among them as well as treatments. Furthermore, the 2% solution for 2 min were more effective for five species (Fig. 1). Therefore, it has been considered established those in vitro cultures that at least one non-contaminated explant was obtained in the first subculture, which was termed as direct establishment (Fig. 2a-c).
The other four species: Barbula indica, Bryum densifolium, F. flaccidus, and S. platyphylloides did not development without visible explant contamination (Fig. 1). However, despite contamination, it was observed that some explants had developed, whose efficiency was 3.1% (F. flaccidus) to 28.4% (S. platyphylloides). Also, it is important to point out that in all ten species studied showed the development of contaminated explants (Fig. 1). Compared with non-contaminated explants, the development of contaminated explants was quite variable. However, contaminated explants had higher development rates than noncontaminated when the disinfecting treatment was with NaClO 1% for 10 min (1.5 to 86.7%) and with NaClO 2% for 2 min (5.2 to 43.5%). Differently, the NaClO 1% for 5 min treatment showed a lower developmental rate of contaminated explants (2 to 15%) than those without contamination (19%). Moreover, the development of noncontaminated and contaminated explants in cultures of S. platyphylloides, I. tenerifolium, and Bryum coronatum was 0 and 28.4%, 3.5% and 43.5%, and 7.1% and 86.7%, respectively (Fig. 1).
Although most of the species studied are considered to be of widespread occurrence in Brazil, explant rescue techniques were applied to all. According to Rowntree et al. (2011), contaminated explants that maintain their capacity to develop can be propagated in subsequent subcultures. Therefore, contaminated explants of all ten species were used to recover and establish in vitro cultures to obtain at least one explant under aseptic conditions. The rescue technique was initially introduced by Rowntree (2006), in which they described the recovery of plant material from partially contaminated cultures, using sodium dichloroisocyanurate (NaDCC) and PPMTM (Plant Cell Technology, Inc., Washington, USA) as sterilizing agents, increasing the success rate of 28 % to 52% in the in vitro establishment of rare plants. A similar method also appears to have been tested in other plants, such as Thamnobryum alopecurum and Hypnum cupressiform but was unsuccessful (Vujičić et al. 2011;Sabovljević et al. 2012).
The protocol for the rescue techniques is outlined in Fig. 2 (in the following order a, d-g, and then back to c), which was called indirect establishment, as it does not follow the regular protocol used in in vitro establishment of bryophytes of aseptic cultures. Thus, the indirect establishment uses techniques to recover explants developed after disinfestation that were not adequately decontaminated. Therefore, this procedure was applied to all species in this study. This procedure was constituted by successive subcultures of the explants developed with contamination, wherein the during the first two subcultures, the contaminated explants were transferred to new plates every 5-10 days. In the following subcultures (30 to 40 days each),  the explants formed colonies, and the regions of the cultures with no visible contamination were transferred to new plates. The level of explant contamination decreased after each subculture, which made possible to gradually achieve asepsis until the in vitro establishment of the species. Figure 3 shows the indirect establishment of Bryum argenteum and Isopterygium tenerifolium cultures. These species were submitted to the disinfestation treatment, which did not result in adequate decontamination. However, despite the contamination, the explants showed capacity to develop and grow ( Fig. 3a-f, i-j). Hence, the rescue techniques were applied, and after successive subcultures, protonemata (Fig. 3c,d), young gametophytes (Fig. 3e) or branches of elongated gametophytes (Fig. 3i) were carefully transferred to new plates and subcultivated until no contamination was observed. A similar procedure was repeated to reduce contamination (Fig. 3g,k) Figure 2 -a-g. Establishment and in vitro propagation of mosses. a-c. direct establishment (without contamination)a. developed explants were transferred to new plates; b. at the end of the 1 st subculture, the established explants were fragmented and transferred to new plates; c. propagation of the species. a, d-g and c. indirect establishment (with contamination) -a. developed explants were transferred to new plates; d-e. the explants were subcultivated until they had produced uncontaminated regions; f. sufficiently developed explants were fragmented and subcultivated; g. presence of established explants with no visible contamination in the plates; c. explants, free of contamination, were used to establish in vitro aseptic cultures and used for propagation of the species.  (Fig. 3h,l).
It was also observed that the time and number of subcultures required to obtain the indirect establishment of aseptic cultures varied among the species. Besides, none the studied species showed any correlation between the number and time of subcultures to establish in vitro cultures, as well as with factors such as plant growth habit, type of explant, and natural growth conditions of the species (Tab. 3).
Furthermore, Barbula indica, Bryum argenteum, and P. pensilvanicum were submitted to a precultivation procedure, a stage before disinfestation with the purpose of reducing contamination and increasing the production of explants. For Barbula indica and Bryum argenteum, the precultivation treatment resulted in neither enhancement of the number of explants nor better decontamination (Fig. 1). Nevertheless, the precultivation of P. pensilvanicum and subsequent disinfestation (2% bleach for 2 min) yielded 31.2% of explants with no visible contamination (Fig. 1).
In precultivated Bryum argenteum with subsequent disinfestation disinfested with 1% bleach for 5 min, which yielded 1.5% of explants with no visible contamination. Moreover, in this condition was also observed the development of four additional contaminated explants. It appears that, compared with non-precultivated samples, there was no improvement in explant production and decontamination. Conversely, Rowntree (2006) found that precultivated bryophytes can improve disinfestation, and, thereby, reduce the Figure 3 -a-l. Indirect establishment of Bryum argenteum (a-h) and Isopterygium tenerifolium (i-l) -a. contaminated colony of explants in the 3 rd subculture (arrow); b. partial contamination (arrow) in the 4 th subculture; c. partial contamination (arrow) in the 5 th subculture after 24 days of cultivation; d. partial contamination (arrow) in the 5 th subculture after 35 days, showing the growth of protonemata external to the medium with contamination (asterisk); e. growth erect gametophytes, externally to the medium with contamination (arrow) in the 5 th subculture after 24 days; f. growth of gametophytes near contaminated areas (arrows); g. apical fragment of gametophyte without contamination in the 6 th subculture after 11 days of cultivation; h. established Bryum argenteum culture during the 6 th subculture; i. Isopterygium tenerifolium with contamination (arrow) in the 3 rd subculture with 40 days; j. growth of gametophyte near the contamination (arrows); k. apical fragment of gametophyte without contamination in the 4 th subculture after 40 days; l. established Isopterygium tenerifolium culture free of contamination in the 5 th subculture after 40 days. Bars: k. = 0.1 mm; f, g, j, l = 0.5 mm; c, d, e, h, i = 1 mm; a, b = 2 mm.      level of contamination, compared with bryophytes disinfested after field collection. Hence, the precultivation treatment must be tested individually because it may yield different results depending on the species. Concerning Pogonatum pensilvanicum explants, which had limited biological material, and reduced gametophytes size (2-3 mm), rudimentary leaves and persistent protonemata (Sharp et al. 1994) were disinfested only after precultivation (Tab. 3). The explants used in the disinfestation treatment were at least 8 mm long, which was achieved after nine months of precultivation. The precultivation treatment was very efficient for this species and increased gametophyte production, as well as a higher rate of disinfestation. Thus, the percentage of explants developed without contamination for P. pensilvanicum was relatively high (31.2%). In fact, this was the highest number of explants developed without contamination among all species studied (Fig. 1). Additionally, the explants that development with contamination (9.4%) required fewer subcultures and less time to establish aseptic cultures (Tab. 3).
For Barbula indica, the precultivation treatment was not relevant because occurred a decrease from 12.5% to 3.4% in explant production submitted to non-precultivation and precultivation treatment, respectively (Fig. 1). Besides, only contaminated explants developed in both conditions. Even with low success rates for in vitro establishment of B. indica, it is crucial to improving the cultivation techniques for this species. Viet et al. (2010) state that this species is reliable for atmospheric biomonitoring, such as heavy metals air contamination of subtropical and tropical regions. However, to use B. indica for air biomonitoring a few obstacles must be overcome as it is an acrocarpic plant, about 1 cm high with strong adhesion to the substrate, which can influence the analysis on the level metal contamination of the air. Thus, it is essential to use B. indica for that purpose is necessary plants with no contact with external environment, which makes in vitro propagation the best choice to provide plant free of contaminants for biomonitoring. Furthermore, a high multiplication rate can be achieved through micropropagation techniques to provide plant material and place in more suitable substrates such as house walls, roofs, and tree trunks, as suggested by Viet et al. (2010).
All the species had protonema as the initial form of development. Besides, the species Bryum coronatum, Bryum coronatum, I. tenerifolium, V. cuspidifera, and S. platyphylloides also showed the development of shoots from inoculated explants (Fig. 4a). For most species, the initial development was from structures such as gametophyte apex (Fig. 4b), and, or, from the axil of the leaf, attached or not to the stem. Also, Bryum argenteum, Bryum coronatum and V. cuspidifera developed protonemata from stem sectioned regions (Fig. 4c). It was also observed the development of protonemata on sexual structures, such as in Bryum argenteum, in the female gametangium (archegonium) and the male gametangium (antheridium) of I. tenerifolium (Fig. 4d-e). Bryum argenteum, Bryum coronatum and S. platyphylloides showed apical growth (Fig.  4f). Only L. crispum developed protonemata directly from the base and apex of the leaf (Fig. 4g; Tab. 3).
With few exceptions, the usual form of growth and colony formation occurred from protonemata (Fig. 4h). Moreover, among the species, there was a considerable variation in the development of gametophytic buds, which usually took place within 30 days of cultivation. After this period of cultivation, most species presented more quantity and developed gametophytes, as seen in P. pensilvanicum (Fig. 4i). As well as V. cuspidifera, where gametophytic buds emerged the gametophytic buds appeared after 1 month of cultivation, and in 3 months, gametophytes were 1.5 mm long, and in 6 months they developed reproductive structures (archegonia and antheridia). In Bryum argenteum, the formation of reproductive structures was also observed after the 4 th month of cultivation. Only Barbula indica did not develop gametophytic buds at any moment.
I s o p t e r y g i u m t e n e r i f o l i u m a n d S . platyphylloides (Fig. 4j) showed the formation of colonies from gametophytes as a result of apical growth and free branching. The in vitro development of plants with similar morphological characteristics to that occurring in nature, with formation and predominance of their respective structures (leaves and stems) throughout the growing period, presents excellent advantages in studies related to the morphogenesis of mosses. The use of plant hormones can also be effective in inducing growth and differentiation of mosses. Sabovljević et al. (2010), investigated the influence of gibberellins (GA 3 and GA 7 ) on in vitro growth and multiplication of B. argenteum protonemata and confirmed their positive effects on the morphogenesis of this species. However, they also raised a question regarding the origin of the increase in protonemata growth of the studied species, whether cell expansion or division is the main factor in the growth of protonemata.
Thus, mosses that maintain their morphology during in vitro cultivation can help to clarify problems in such morphogenetic events, and may be very useful in studies of interest to conservation or biomonitoring of air quality, as has already been done at an international level, through the "Moss Bag" technique, with the species Pseudoscleropodium purum (Hedw.) M. Fleisch. native (Capozzi et al. 2016) and Sphagnum palustre L. cloned (Capozzi et al. 2017).
It is noteworthy to point out that the protonema phase is considered the most important for micropropagation. Protonemata increase the chances of in vitro establishment of bryophytes, and it is the most useful form for environmental and economic applications, such as bioprospecting, bioremediation, and biomonitoring because it is easy to propagate vegetatively, favors biomass production, and has excellent stability over time (Cuker et al. 2004;Duckett et al. 2004;Beike et al. 2014).
Concerning Bryaceae family, all species developed asexual propagules in cultures with approximately 120 days of age. In Bryum coronatum and B. densifolium the development of tubercle-like structures in the axil of the leaves, (Fig. 4k) and in the caulonemata was observed (Fig. 4l). The vegetative propagules formation has not been described in the literature for B. densifolium, and, therefore, to best of our knowledge, it is the first time that these structures are observed in this species. According to Spence (2014) formation of these propagules. The production of vegetative gemmae is a crucial mechanism for successful dispersal, especially for species that show limitations in the development of sporophytes, as well as for the dispersion and germination of spores in natural conditions, as demonstrated for B. coronatum (Egunyomi 1982).
In Bryum argenteum, the formation of spherical cells from protonemata filaments, termed as protonemic gemmae as described by . These structures can be produced in response to dehydration of the culture medium and constitute an important survival strategy to drought. Nevertheless, in plants grown under natural conditions, these structures are difficult to observe (Duckett et al. 2001).
According to Ares et al. (2014), the formation of protonemic gemmae was also verified in liquid culture media (only in contaminated media). The cultivation in liquid media has the ecological advantage of reducing the time of cultivation in about 1 month (Ares et al. 2014). has the ecological advantage of reducing the time of cultivation to about 1 month (Ares et al. 2014). In fact, in our studies, these gemmae were also observed in liquid cultures of B. argenteum; however, only in 4-month-old aseptic cultures (unpublished results). Furthermore, in both solid and liquid cultures, protonemic gemmae were observed in segmentation (Fig. 4m), without Tmena formation (abscission cells). Also, when these structures were transferred to new culture media, the germination occurred with the formation of protonema filaments, and, consequently, the formation of colonies of protonemata (Fig. 4n). The regeneration of protonemic gemmae in several Bryum species has been reported by Pressel et al. (2007); however, in B. argenteum, these results have not yet been reported.
Another species that also developed vegetative propagules was Barbula indica, where it was observed the formation of gemmae directly from the protonema apex (Fig. 4o). This species did not produce gametophytes under any circumstance or condition tested. It appears that in either natural or cultivation conditions, the production of gemmae is related to the availability of nutrients, whereas tubers-like structures are produced in response to the absence of nutrients (Duckett et al. 2004).
Some species were able to maintain viability for more than a year without the need of a medium exchange. Although some species had a dry or dead aspect after months of cultivation in the same medium, the regeneration from shoots or protonemata occurred after the transplantation to a new medium. This phenomenon was observed in Leucobryum crispum, which developed protonemata from the apex of the leaves (Fig. 4p).

Conclusions
The indirect establishment was possible for all species tested and could be an alternative for species with low availability of material. Therefore, the indirect establishment is proposed as a viable methodology for the establishment of in vitro cultures of bryophytes, because it has higher rates of development and can be used to cultivate bryophytes that are more sensitive or scarce. This new method will help expand the commercial, conservational, and biotechnological uses of bryophytes of ecological interest, which includes endangered species.
Thus, this technique can also be applied to achieve the establishment of species of conservational interest, which may help to obtain micropropagation in large scale to reintroduce them to their natural environment.
The improvement and structuring of strategies that increase the success rate in the in vitro establishment of bryophyte species, such as those that need to be cultivated from the gametophyte, is a fundamental practice and of inestimable significance to stimulate the development of new research on this group of promising plants.