To initiate the in vitro culture of endangered species, the use of seeds is the preferred method because it avoids destroying the mother plants and preserves genetic diversity. In this work, seed disinfection with PPM proved to be efficient, and no contamination was observed (data not shown). This compound is a broad-spectrum biocide with no adverse effects on in vitro seed germination, callus proliferation or callus regeneration.
Although the germination is usually low in seeds with a hard coat (Rojas-Aréchiga and Vazquez-Yanez, 2000), as with T. mombergi, we obtained 80% of the germination at 45 days (Fig. 2a). Longer periods did not improve the response. The percentage of germination obtained in this study was higher than the data reported for T. laui with in vitro conditions (29 %) or under ex vitro conditions for T. laui (71%) and T. pseudopectinatus (8%) (Santos-Díaz et al. 2003b; Flores et al. 2005). In addition, the germination rate was higher than that reported for Turbinicarpus valdezianus and T. subterraneus, which presented 46% and 90% of germination at double time (Davila-Figueroa et al. 2005). The best response obtained in this work could be associated with the disinfection protocol. The asepsis of T. laui, T. valdezianus and T. subterraneus seeds was performed with sodium hypochlorite and ethanol, which are more aggressive agents than PPM, and therefore could damage the embryo structure. Other factors that affect germination in cacti are age, size, dormancy and origin of seeds (Rojas-Aréchega et al. 1997; Rojas-Aréchiga and Vázquez-Yanes, 2000).
After seed germination, well-defined epicotyls and roots were observed at 30 days, reaching 4 mm and 5.8 mm, respectively, at 90 days. The presence of spines was evident after 14 days and increased proportionally to time culture (Fig. 2b). Since T. mombergeri is a little known species, there is no information about the germination rate or growth parameters in the wild for comparison.
The epicotyls of T. mombergeri were cut longitudinally, and the segments were transferred to a medium with BA. Because the scarcity of material (14 germinated seeds) only B2 and B4 media were tested. These BA concentrations were selected because, in previous studies, they successfully induced the shoot formation in T. laui (Santos-Díaz et al. 2003b); it also has been reported that BA was efficient in propagating other Turbincarpus species (Pérez-Molphe et al. 2015). Data showed that 50% and 7% of the explants cultivated on B2 and B4 media, respectively, regenerated one shoot at 90 days, which were highly hydrated, and presented abundant callus formation (Table 1, Fig. 3a). This response was lower than that reported by Dávila-Figueroa et al. (2005), who obtained between 7.8 to 19.7 shoots per explant during the propagation of several Turbincarpus species. It has described that the heterosis in hybrids can affects the regenerative capacity. For example, the ability for generating in vitro shoots was higher in a tomato parental line than in their hybrids, and this difference was attributed to the heterosis and maternal effects (Ohki et al 1978). Additional genetic studies must be done to determine if this phenomena, is also present in the hybrid T. mombergeri.
The shoots were transferred to B2 to increase shoot number, and after a second subculture an average of 2.8 shoots per explant were obtained, still hydrated and with abundant callus. Hyperhydricity have been described during micropropagation of many cacti species, such as, Mammillaria gracillis, M. pectinifera, Escobaria minima and Pelecyphora aselliformis, among others (Giusti et al. 2002; Poljuha et al. 2003). This effect has often been considered a physiological response to simultaneous stress factors of the in vitro culture, which negatively impacts the micropropagation efficiency and survival of plants in ex vitro conditions (Debergh et al. 1992). Some biochemical characteristics present in hyperhydric tissues are reduced dry weight, and less lignin, cellulose and calcium content, as well as a low Ca+ 2/uronic acid ratio (Kevers et al. 2004).
Hyperhydricity can be reduced by improving ventilation and to decrease ethylene accumulation in vessels; by adding osmotic agents (mannitol, polyethylenglycol), to diminish the water potential of media and to low the water content in tissues; by decreasing the concentration of nutrients in the medium; or by increasing the Ca concentration (Thomas et al. 2002; Snyman et al. 2011; Nikam et al. 2019).
Therefore, to reduce the hyperhydricity in T. mombergeri shoots, the effect of culture media (MS, ½ MS, ¼ MS, ½ WPM media), osmotic agents (1 % PEG) and double calcium concentration (2Ca) were tested. The reduction in salts concentration in ½ MS medium generated 21% of compact shoots at 90 days. This percentage improved in ¼ MS medium or ½ WPM medium containing 2Ca concentration and PEG, generating between 80 to 90% of compact shoots or shoots with a very low degree of hyperhydricity (Table 2).
The beneficial calcium effects could be attributed to the cell walls strengthening, providing rigidity by reversibly cross-linking with the pectic chains. Its association with plasma membrane also helps to maintain its stability by bridging phosphate and carboxylate groups (White and Broadley 2003). Calcium was also important for vegetative buds formation, and the development of flowers and roots in tobacco pith explants (Capitani and Altamura 2004). Furthermore, it is an essential element for cactus nutrition, representing 85% of dry weight in some species (Gallaher 1975). As T. mombergeri grows in calcareous soil, high levels of Ca might be required for good shoot development.
Reduction on salt concentration also seem to influence the T. mombergeri shoots compaction since the use of ½ MS, ¼ MS or ½ WPM media generated a higher number of compact shoots than the use of MS medium. Better results were obtained in ½ WPM medium compared to ½ MS medium. The major differences in macronutrients among these media are in ammonium and nitrate ion concentrations, as well as, total ion concentration. Full-strength MS is high in ammonium (20.6 mM) and nitrate ions (39.4 mM), while WPM contains lower concentrations of both ammonium (5 mM) and nitrate (9.7 mM) ions. It has been reported that the ratio of NH4:NO3 affects the levels of hyperhydricity in several species, such as, Aloe polyphylla (Ivannova and Van Staden 2008) and date palm (El-Dawayati and Zayed 2017). Thus, a reduction in the NH4:NO3 ratio could also contribute to reducing the hyperhydricity of T. mombergeri shoots.
On the other hand, 100% of shooting was observed in ½ WPM-2Ca-P medium (Table 2) generating two shoots per explant. Although the formation of compact shoots was attained, the presence of callus was still very high as shown in Fig. 3b.
Several approaches have been used to reduce callus formation, including cytokinin elimination or the employment of auxin transport inhibitors, such as TIBA. This compound enhanced somatic embryogenesis in groundnut and shoot formation in Morus alba (Venkatesh et al. 2009; Bhau and Wakhlu 2001) and enhanced Rosa hybrida micropropagation (Singh and Syamal 2000). Thus, we cultivated the T. mombergeri shoots in ½ WPM-2Ca-P added with 0.5, 1 and 2 mg L− 1 TIBA. The callogenesis was reduced in the presence of TIBA proportionally to the concentration (Table 3). This result suggests that the T. mombergi shoots must synthesize high levels of endogenous auxins that are responsible for callus generation. Figure 3c shows the aspect of T. mombergeri shoots without callus at 90 days of culture.
Root formation and transfer to soil
The compact shoots (2 to 3 cm high) were transferred to media ½ WPM-2Ca-P medium with 1 mg L− 1 TIBA (named WCPT) alone or in combination with 5.7 µM IAA (WCPT-1) or 0.5 mg L− 1 urea (WCPT-2) to induce the rooting of shoots (Table 4). After 90 days in the WCPT medium, 21.7% of the explants developed roots of approximately 3.8 mm long. In WCPT-1 medium, the percentage of rooting lightly increased and longer roots (4.6 mm) were generated at 90 days. A reduction on compact shoots, however, was observed according to time, probably because of the presence of the auxin in the medium, which induced an incipient callus formation.
The shoots cultivated in media WCPT-2 generated the lowest percentage of rooting, the root length was similar to that obtained in the WCPT-1 medium, but the shoots duplicated their diameter at 90 days (Fig. 3d).
Taking in account these results, an additional experiment was performed (WCP-3). The shoots were maintained in the medium WCPT-2 for 90 days to generate wide and thick shoots. The plant material was then transferred to medium WCPT-1 for 60 days, to induce a vigorous radical system, and was finally maintained in the WCP medium for 120 days (Table 4). Using this strategy, the callus formation was avoided completely, and at the end of experiment 96% of rooted shoots were generated with well-defined roots from with an average length of 13 mm. Figure 3e shows the aspect of rooted shoots after 1 year in culture. These results shows that T. mombergeri requires a long period to develop strong roots. In wild conditions, most Turbinicarpus species exhibit a very thick primary root, which represents 80% of the plant body, and acts like an anchor, and more importantly, as water storage for dry periods. The root growth is therefore a time- consuming event.
The beneficial effect of urea in growth and rooting of T. mombergeri shoots is attributed to a higher availability and better absorption of organic nitrogen. It is well known that nitrogen is required for the synthesis of chlorophyll and for amino acid metabolism, which are essentials for plant growth and development. Several urea transporters have been identified across different cellular membranes. For example, in Arabidopsis, a symporter, that cotransports urea with protons at high affinity, has been described. In the tonoplast, various tonoplast intrinsic proteins (TIPs), a subfamily of aquaporins, transport urea in a channel-like manner. These transporters seem to optimize the nitrogen intake and compartmentation in dependence of the nitrogen forms being available in the medium (Kojima et al. 2006). Further studies must be done to identify the putative urea transporters in Turbinicarpus species.
The T. mombergeri plants were transferred to soil, and 85% survived after 1 year. At this period, the plants showed the characteristic spines pattern observed in mature plants (Fig. 3f).
In summary, this work shows that reduction of salt medium concentration, high level of calcium concentration and presence of PEG reduced the shoots hyperhydricity. The addition of TIBA decreased caulogenesis and the presence of urea promoted the development of thick shoots. The protocol developed allowed the successful micropropagation of the critically endangered cacti T. mombergeri, contributing to its conservation.