PROPAGATION AND TISSUE CULTURE HORTSCIENCE 53(12):1855–1861. 2018. https://doi.org/10.21273/HORTSCI13376-18 Micropropagation Shortens the Time to Blooming of Begonia montaniformis 3 Begonia ningmingensis var. bella F1 Progeny I-Ling Lai Graduate Institute of Bioresources, National Pingtung University of Science and Technology, 1 Shuefu Road, Neipu, Pingtung 912, Taiwan Chih-Wan Lin Department of Forestry, National Pingtung University of Science and Technology, 1 Shuefu Road, Neipu, Pingtung 912, Taiwan Tsai-Yu Chen Department of Horticulture, National Chung Hsing University, 145 Xingda Road, South District, Taichung 402, Taiwan Wei-Hsin Hu1 Department of Biology, National Museum of Natural Science, 1 Guancian Road, Taichung 404, Taiwan Additional index words. plant growth regulator, light quality, shoot organogenesis, Begonia Abstract. Begonia montaniformis 3 Begonia ningmingensis var. bella hybrids have high ornamental potential. Hence, the aim of this study was to determine the optimal conditions for the micropropagation of a Begonia montaniformis 3 Begonia ningmingensis var. bella F1 progeny by using various concentrations of plant growth regulators (PGRs) and varying light spectra in half-strength Murashige and Skoog (1/2 MS) medium. The results showed that the explant regeneration was optimal when the lamina was incubated in a medium supplemented with 2.0 mM N6-benzylaminopurine and 0.8 mM a-naphthaleneacetic acid (NAA). Under such conditions, 98% of the explants regenerated adventitious shoots after 8 weeks, and 41 buds were produced per explant on average. The mean shoot length was 9.6 mm, and on average, 4.5 shoots per explant were more than 2 mm long. Subsequently, the induced adventitious shoots were transferred into rooting medium consisting of 1/2 MS and various NAA concentrations. After 4 weeks, the shoots subcultured in this medium showed ’93% root induction and an average of 3.5 adventitious roots per explant. Furthermore, the applied light spectrum significantly influenced shoot regeneration, and optimal results were achieved under an equal distribution of blue, red, and infrared light. The histological sections of shoots regenerated from direct organogenesis were observed through scanning electron microscopy (SEM). Afterward, the rooting adventitious shoots were subcultured in PGR-free medium for 8 weeks. The seedlings were successfully acclimated 4 weeks after being transferred to soil and bloomed after 11 months in a greenhouse. Thus, the PGR composition in micropropagation efficiently shortened the time to blooming from 25 to 16 months. Begonias are among the most popular and beautiful ornamental plants. They produce showy flowers and acclimate to shady environments such as those indoors. More Received for publication 10 July 2018. Accepted for publication 27 Sept. 2018. We gratefully acknowledge Yuan-I Lee for the technical support provided in histology, and the National Museum of Natural Science and Ministry of Science and Technology, Taiwan, for partly funding this project. We also thank Ching-I Peng, a famous Begonia taxonomist who passed away recently, for providing the experimental materials and information on the habitat of wild begonia cultivars used in this work. 1 Corresponding author. E-mail: bohu@mail.nmns. edu.tw. HORTSCIENCE VOL. 53(12) DECEMBER 2018 than 1600 species of the genus Begonia exist (Kiew et al., 2015). The species of section Coelocentrum are richly represented in limestone karst areas across the Sino-Vietnamese border region and comprise more than 60 species (Chung et al., 2014). Nearly half of the species in this section have been discovered in the past decade (Averyanov and Nguyen, 2012; Chung et al., 2014; Li et al., 2016). Some of the species of section Coelocentrum have beautiful maculation patterns and attractive leaf textures (Peng et al., 2015). The area from Southern China to Northern Vietnam harbors rich biodiversity (Sodhi et al., 2004). Begonia of section Coelocentrum is among the most characteristic limestone plants and is confined to cave-like microhabitats (i.e., caves, crevices, and fissures) of the Sino-Vietnamese karst region (Peng et al., 2015; Qin et al., 2017), with most species identified from a single or a several localities and differing from one another in leaf shape, pubescence, texture, and variegation (Gu et al., 2007). Both Begonia montaniformis and B. ningmingensis var. bella belong to section Coelocentrum and have special leaf variegation. B. montaniformis is an endemic lithophytic species found on limestone hills in North Vietnam. On the leaf surface of this plant dense conic bullae with a silvery green zone along the primary and secondary veins present an impressive stereoscopic shape, giving the plant high ornamental value. Furthermore, the yellowish-green color of its staminate and carpellate flowers is rare in the Begonia genus (Peng et al., 2015). However, B. montaniformis plants are difficult to grow. The germinated seedlings rarely survive and are extremely sensitive to environmental changes. However, if they grow satisfactorily, the leaves last for long periods. B. ningmingensis var. bella is also an endemic and rare species and is only distributed in the karst area of Southwestern Guangxi, China. The upper surface of its leaf is dark green, brown, or dark brown, with white maculation along the major veins, and its lower surface is reddish or red (Fang et al., 2006). Conserving the superior F1 hybrid is potentially valuable for horticultural purposes, and in vitro regeneration techniques may be advantageous in coping with market demand for this hybrid. Therefore, in vitro hybrid regeneration techniques should be investigated and the corresponding control mechanism determined before this hybrid is promoted for commercial use. Furthermore, light quality influences adventitious shoot regeneration (Burritt and Leung 2003; Zhou et al., 2016). A novel technology has been used to examine the effect of the light spectrum on in vitro micropropagation (Fang et al., 2011; Lee et al., 2011). This study attempted to develop a micropropagation protocol for rapid in vitro propagation of the F1 hybridized progeny of B. montaniformis · B. ningmingensis var. bella (Novel F1), especially in terms of plant growth regulators (PGRs) and light quality. Materials and Methods Plant materials. Individual plants of B. montaniformis and B. ningmingensis var. bella were collected from their natural habitats and cultivated in the experimental greenhouse of the National Museum of Natural Science in Taiwan. The greenhouse was determined to have a natural photoperiod per day, 85% to 95% relative humidity, and 25 C/20 C day/night temperature. The aforementioned plants were artificially hybridized, and the seeds of their F1 hybrids were collected and sterilized with 0.8% sodium hypochlorite. Subsequently, the seeds were germinated in a dark cabinet at 25 ± 1 C with basal medium containing one-fourth-strength Murashige and Skoog (1962, 1/4 MS) salts 1855 supplemented with niacin (0.5 mg/L), pyridoxine HCl (0.5 mg/L), thiamine HCl (0.1 mg/L), myoinositol (100 mg/L), glycine (200 mg/L), and sucrose (2%, w/v) and solidified with 0.75% (w/v) agar. Most seeds germinated within 4 to 6 weeks. Subculture and culture medium. After 8 weeks, the germinated F1 seedlings were incubated in 617-mL erlenmeyer flasks containing 100 mL of half-strength (1/2 MS) medium supplemented with 3% (w/v) sucrose and 0.1% peptone (w/v). The other conditions were identical to those used for the aforementioned seed medium. After 3 months, there were 3–4 leaves present on the plants. The length of leaf was 4–5 cm. We randomly selected the sample leaves regardless of leaf position or age. The laminas were cut randomly from whole leaf and 10–12 pieces could be obtained from one leaf. About 8 · 8 mm2 lamina of plants with stereoscopic and white spotted leaf characteristics was obtained using a sharp scalpel. The explants were then placed individually with their adaxial sides upward in 18 · 150 mm2 test tubes (Pyrex No. 9820; Corning Life Sciences, Tewksbury, MA) containing 7.5 mL of 1/2 MS medium supplemented with N6-benzylaminopurine (BA), 6-(4-hydrooxy-3-methil-but-2-enylamino) purine (zeatin), or N6-(3-hydroxybenzyl)amino)purine (meta-topolin, mT) at different concentrations (0, 2, 4, or 8 mM) and without or with auxin, namely 0.8 mM a-naphthaleneacetic acid (NAA). After 8 weeks, the regeneration percentage of adventitious shoots, number of shoots and elongated shoots, and shoot length per explant were recorded. Subsequently, adventitious shoots with length greater than 5 mm were separated from the explants grown in medium containing 2 mM BA and 0.8 mM NAA and were transferred to 1/2 MS supplemented with 0.00, 2.69, or 5.38 mM NAA. After 4 weeks, the percentage of rooted shoots, number of roots, number of leaves, and death rate were recorded. Cultivation conditions. The pH of all media was adjusted to 5.8 with 0.1 M NaOH before autoclaving at 121 C for 15 min. The culture was incubated at 25 ± 1 C with a photoperiod of 40 mmol·m–2·s–1 (daylight fluorescent tubes FL-40D/38 and FL-40BR/38, 40 w, China Electric Co., Taipei, Taiwan) and light/dark cycle of 16/8 h. Spectral quality affects morphogenesis of explant plantlets during in vitro culture. A novel system equipped with light-emitting diodes (LEDs) as light sources for tissue culture (TC) plantlets (Nano Bio Light Technology Co., Ltd., Taiwan; Chen et al., 2016; Fang et al., 2011) was used to test the impact of spectral quality on the micropropagation of novel F1 hybrids. Eight sets of LED chips were configured through a combination of blue, green, red, and infrared (IR) (B/G/R/IR) LEDs and mounted on the lips of the TC vessels to produce the same light intensity. The LED peaks for B/G/R/IR were 450 ± 3, 525 ± 3, 660 ± 5, and 730 ± 5 nm, respectively. Two combinations produced cool white light and warm white light with a spectrum similar to that of daylight. The B/G/R/IR ratios in these two combinations were 26:26:26:2 and 10:45:51:4. Four sets had chip ratios of 3B:3R:3IR, 1B:7R:1IR, 1B:1G:7R, and 1B:8R. Two sets had only one band each, namely 9B and 9R. Five laminae of the F1 hybrid were transferred to one TC vessel holding 1/2 MS medium with 2 mM BA and 0.8 mM NAA. Three laminae were placed in each light quality set. The experiments were repeated thrice. The photosynthetic photon flux density was adjusted to 42 mmol·m–2·s–1 in each light quality set. The other culture conditions were kept the same as in the aforementioned experiments. The regeneration percentage of adventitious shoots, number of shoots, and shoot length per explant were recorded after 8 weeks. Historesin sections and SEM. The adventitious shoots of a lamina placed in medium supplemented with 2 mM BA and 0.8 mM NAA were periodically removed for anatomic analysis. The shoots were cut into small sections and fixed at 4 C for 6–8 h under vacuum in 2% paraformaldehyde and 2.5% glutaraldehyde within 0.1 M sodium phosphate buffer (pH 6.8). These sections were then washed with 0.1 M sodium phosphate buffer and dehydrated using an ethanol series (15% to 100%). After dehydration, part of the tissue was infiltrated and embedded with increasing concentrations of Technovit 7100 resin (Kulzer, Hanau, Germany) under vacuum. After the final polymerization of the sample resin, serial transversal sections were cut using a rotary microtome (RM2245; Leica, Bensheim, Germany) and stained with Periodic Acid–Schiff (PAS) reaction for insoluble carbohydrate and amido black 10B for protein. These preparations were photographed and examined using a light microscope (AxioCam ERc 5s; Zeiss, Jena, Germany; Lee et al., 2006; Yeung, 1999). Other parts of the tissues were dehydrated in a critical point dryer. Subsequently, they were coated with gold using an ion sputter coater (E-1010; Hitachi Ltd., Tokyo, Japan) and observed through SEM (S-3000N; Hitachi Ltd.). Statistical analysis. In adventitious shoot induction experiments, 15 explants were used per treatment. In the rooting and light spectrum experiments, five explants were used per treatment. All experiments were repeated thrice. One-way ANOVA was performed using the Statistical Package for the Social Sciences, version 12.0.1 (IBM, Armonk, NY), and significant differences between the means were evaluated using Duncan’s multiple range test at P < 0.05. Table 1. Effect of different concentrations of plant growth regulators on Begonia montaniformis · B. ningmingensis var. bella shoot organogenesis after 8 weeks of culture. BA 0 2.0 4.0 8.0 2.0 4.0 8.0 Type of PGR (mM)z Zeatin mT 0 0 Adventitious shoots (%) Mean shoot length (mm) No. shoots/explants Mean no. of elongated shoots (>2 mm) Explant necrosis (%) NAA 0 51 ± 51 gy 0.0 ± 0.0 e 1.6 ± 2.2 f 0.0 ± 0.0 e 51 ± 50 a 86 ± 35 abcd 3.0 ± 4.9 cd 19.5 ± 13.9 c 1.3 ± 2.3 cd 10 ± 37 def 95 ± 22 ab 2.9 ± 4.2 cd 25.6 ± 15.4 c 1.4 ± 2.0 c 10 ± 30 def 87 ± 34 abc 1.7 ± 3.2 de 23.4 ± 20.5 c 1.2 ± 1.9 cde 18 ± 39 cde 0.8 98 ± 14 a 9.6 ± 9.6 a 41.2 ± 26.3 b 4.5 ± 4.7 a 4 ± 19 ef 0.8 98 ± 14 a 6.1 ± 8.8 b 50.1 ± 23.7 a 3.1 ± 4.4 b 0±0f 0.8 98 ± 14 a 5.0 ± 6.5 b 56.4 ± 30.9 a 2.4 ± 3.2 b 0±0f 2.0 66 ± 48 efg 0.8 ± 2.2 de 6.0 ± 7.5 def 0.4 ± 1.1 cde 25 ± 44 bcd 4.0 90 ± 31 ab 0.9 ± 2.3 de 5.5 ± 7.1 def 0.4 ± 1.1 cde 24 ± 43 bcd 8.0 84 ± 37 abcde 1.5 ± 3.2 de 10.3 ± 10.0 de 0.7 ± 1.4 cde 26 ± 44 bcd 2.0 0.8 67 ± 47 defg 0.5 ± 2.1 de 3.8 ± 4.8 ef 0.2 ± 0.9 cde 4 ± 20 ef 4.0 0.8 69 ± 47 cdefg 1.8 ± 4.0 de 5.0 ± 5.9 def 0.8 ± 1.9 cde 2 ± 14 ef 8.0 0.8 85 ± 36 abcd 1.6 ± 3.7 de 11.7 ± 9.9 d 0.7 ± 1.6 cde 6 ± 24 ef 2.0 76 ± 44 bcdef 2.5 ± 4.8 cde 3.5 ± 4.2 ef 1.0 ± 1.9 cde 52 ± 50 a 4.0 58 ± 50 fg 2.9 ± 1.2 cd 1.8 ± 2.8 f 0.2 ± 0.5 cd 54 ± 50 a 8.0 58 ± 50 fg 0.9 ± 2.2 de 3.1 ± 4.1 ef 0.5 ± 1.0 de 37 ± 49 b 2.0 0.8 17 ± 38 h 0.0 ± 0.0 e 0.8 ± 2.2 f 0.0 ± 0.0 e 33 ± 47 bc 4.0 0.8 12 ± 33 h 0.0 ± 0.0 e 0.2 ± 0.8 f 0.0 ± 0.0 e 18 ± 39 cde 8.0 0.8 28 ± 46 h 0.0 ± 0.0 e 1.6 ± 3.5 f 0.0 ± 0.0 e 10 ± 30 def z Lamina explants were cultured in half-strength MS medium supplemented with various cytokinins and NAA; PGRs: plant growth regulators. y Means ±SE followed by the same letters indicate that the values are not significantly different according to Duncan’s multiple test at P # 0.05. 1856 HORTSCIENCE VOL. 53(12) DECEMBER 2018 Results Adventitious shoot induction. Among the three cytokinins examined, BA was found to induce greater shoot regeneration than did zeatin and mT (Table 1). In the medium containing BA and 0.8 mM NAA, 98% of the explants regenerated adventitious shoots after 8 weeks, and more than 41 shoots were produced per explant. Although the addition of either 4.0 or 8.0 mM BA along with 0.8 mM NAA to the medium induced more shoots than those observed before the addition, supplementation with 2.0 mM BA and 0.8 mM NAA produced significantly more elongated shoots (>2 mm) than those observed before supplementation, and the mean shoot length achieved after 8 weeks was 9.6 mm (Table 1; Fig. 1A and B). Thus, the PGR composition was determined for further rooting and light spectrum experiments. The addition of 0.8 mM NAA efficiently reduced mortality irrespective of the cytokinin combination. However, compared with a low concentration of zeatin and the control set, a high concentration of zeatin induced slightly more shoots, but the results were much worse than those obtained using BA. Compared with the control set, mT exhibited no significant effect on shoot regeneration and produced fewer shoots when 0.8 mM NAA was added. Root induction, plantlet elongation, and acclimatization. To improve root induction, the adventitious shoots from the explants incubated in the medium supplemented with 2.0 mM BA and 0.8 mM NAA were carefully separated and transferred to the rooting medium with 1/2 MS supplemented with three concentrations of NAA. The root induction had begun in all rooting media after 2 weeks. After 4 weeks, more than 66% root induction was observed in the adventitious shoot cultured in media without and with NAA (Table 2; Fig. 1C). Plantlets rooted in the 1/2 MS media containing different concentrations of NAA did not differ significantly in rooting frequency. When the NAA concentration was increased to 5.38 mM, the mean number of roots per plantlet decreased to 2.27. Plantlets in auxin-free medium had significantly more leaves (average 4.53) than those in medium containing 2.68– 5.38 mM NAA. The wounded lamina slightly thickened after 3 weeks of culture (Fig. 1A) and showed expanded areas at the excised edges. The optimum PGR combination for adventitious shoot regeneration was determined in this study to be 1/2 MS medium supplemented with 2 mM BA + 0.8 mM NAA. Adventitious shoot initiation occurred on the surface and cut margins of the lamina explants within 8 weeks of inoculation in culture medium (Fig. 1B). Root induction in media containing different concentrations of NAA was observed. After 4 weeks, the roots were well developed (Fig. 1C) and fit for subculturing in PGR-free medium for future plantlet production. Plantlets grown from rooting explants required up to 8 weeks in PRG-free Fig. 1. Adventitious shoot regeneration using laminae of in vitro plantlets of B. montaniformis · B. ningmingensis var. bella. (A) Growth from the excised edges of the lamina explants after 3 weeks in half-strength MS medium supplemented with 2 mM BA and 0.8 mM NAA. (B) Adventitious shoot induction from the lamina explants after 8 weeks in the regeneration medium. (C) Rooting of in vitro regenerated shoots after 4 weeks in PGR-free half-strength MS medium. (D) Explants at 11 months after transferring to soil in a greenhouse. (E) Well-developed plantlets after 8 weeks in PGR-free halfstrength MS medium. (F) Explants at 4 weeks after transferring to soil in a greenhouse. Scale bars: 2 mm (A), 2 mm (B), 1 cm (C), 4 cm (D), 3 cm (E), and 3.5 cm (F). Table 2. Effects of half-strength MS medium, with or without various NAA concentrations, on the in vitro rooting of regenerated shoots after 4 weeks of culture. NAA (mM) Rooted shoots (%) No. roots No. leaves Explants necrosis (%) 3.47 ± 0.84 ab 4.53 ± 0.84 a 0.00 ± 0.00 a 0 93.33 ± 6.67 az 2.69 86.67 ± 9.09 a 6.33 ± 1.77 a 1.87 ± 0.38 b 20.00 ± 10.69 a 5.38 66.67 ± 12.60 a 2.27 ± 0.94 b 2.13 ± 0.51 b 26.67 ± 11.82 a z Means ±SE followed by the same letters indicate that the values are not significantly different according to Duncan’s multiple test at P # 0.05. HORTSCIENCE VOL. 53(12) DECEMBER 2018 1857 medium for suitable shoot elongation and root development (Fig. 1E). The seedling successfully acclimated 4 weeks after transfer to soil (Fig. 1F) and produced slightly yellowish flowers after 11 months (Fig. 1D). The period from adventitious shoot induction to flowering was 16 to 17 months. Effect of light quality on micropropagation. The lamina of the explants incubated in 1/2 MS medium supplemented with 2.0 mM BA and 0.8 mM NAA grew optimally under an equal distribution of blue, red, and IR light, such as in 3B3R3IR (Table 3). The induction frequency of adventitious shoots, shoot length, and number of shoots were significantly greater under 3B3R3IR light than under other light sources. Under such conditions, the explant produced on average 30 shoots, an average shoot length of 2.4 mm, and an adventitious shoots percentage of 95%. The explant under warm white light (WW) and only blue light (9B) produced the second best results in terms of the percentage of adventitious shoots and number of shoots. The results were less significantly different from the explant cultured in 3B3R3IR light. However, the blue light slightly inhibited shoot elongation. The least optimal regeneration was recorded under cold white light (CW) and strong red light (1B8R). Historesin sections and SEM. Histological studies showed that morphogenesis gradually changed after the initiation of the Table 3. Effects of various light spectra on shoot regeneration. Medium containing half-strength MS supplemented with 2 mM BA and 0.8 mM NAA after 8 weeks of culture. Treatment Adventitious shoot (%) Shoot length (mm) No. shoots CW 76 ± 43 bcz 1.4 ± 1.0 d 18.4 ± 21.9 bc WW 88 ± 33 abc 2.0 ± 1.0 ab 26.8 ± 22.4 ab 9B 90 ± 30 ab 1.8 ± 1.0 bcd 26.5 ± 34.7 ab 3B3R3IR 95 ± 22 a 2.4 ± 0.8 a 30.4 ± 24.3 a 1B7R1IR 79 ± 42 abc 1.6 ± 1.1 bcd 21.2 ± 29.0 abc 1B1G7R 81 ± 40 abc 1.7 ± 1.1 bcd 18.5 ± 24.3 bc 1B8R 71 ± 46 c 1.4 ± 1.1 cd 15.7 ± 21.6 bc 9R 83 ± 38 abc 1.9 ± 1.0 bc 12.2 ± 12.3 c z Means ±SE followed by the same letters indicate that the values are not significantly different according to Duncan’s multiple test at P # 0.05. culture in 1/2 MS medium containing 2 mM BA and 0.8 mM NAA. After 13 d of culture, both periclinal and anticlinal division occurred in the areas within the outermost one to two epidermal cell layers. Cross-sections revealed cell activation or dedifferentiation, with small and dense cytoplasm in which a series of organized divisions continued (Fig. 2A). Nineteen days after the beginning of induction, the cells of the meristematic zone had rapidly divided and had high cytoplasmic content with small vacuoles and abundant starch grains. The division of cells protruded from the leaf surface and had epidermal layers connected with the original lamina (Fig. 2B). The emerged cells had developed a well-defined meristem and two new leaf primordia after 28 d of initiation. The base of the explant epidermis had differentiated into parenchyma cells, indicating the early stages of shoot bud differentiation (Fig. 2C). The thoroughly developed bud had an obvious apical meristem with lively differentiation of cells after 33 d of culture (Fig. 2D). The leaf epidermis of novel F1 explants at the beginning of culture (0 d) exhibited flat orderly rows of cells and distinct raised trichomes in SEM observations (Fig. 3A). Well-developed buds with numerous leaf Fig. 2. Histological observation of the series of adventitious shoots developed from lamina explants of B. montaniformis · B. ningmingensis var. bella incubated in half-strength MS medium supplemented with 2 mM BA and 0.8 mM NAA. (A) After 13 d of culture, both periclinal and anticlinal divisions occurred in the areas within the outmost one to two epidermal cell layers. (B) After 19 d of culture, the cells of the globular meristematic zone formed and proliferated. (C) After 28 d, a bud with leaf primordia connected to the original lamina explant developed. (D) A fully developed apical meristem with leaf primordia after 33 d of culture. Scale bars: 100 mm (A–D). 1858 HORTSCIENCE VOL. 53(12) DECEMBER 2018 Fig. 3. Scanning electron micrographs of different stages of adventitious shoot development of B. montaniformis · B. ningmingensis var. bella leaf explants cultured in half-strength MS medium supplemented with 2 mM BA + 0.8 mM NAA. (A) Leaf epidermis at the beginning of explant. (B) Leaf epidermis after 33 d in induction medium, with bud primordia displayed. (C) Different stages of the regeneration structure with meristems from leaf explant after 50 d in induction medium. (D) Magnified image of elongated bud with leaf primordial after 50 d in the induction medium. Scale bars: 250 mm (A), 100 mm (B), 500 mm (C), and 250 mm (D). primordia were observed after 33 d in the induction medium (Fig. 3B). Different developmental stages of primordia that involved early swelling to completely formed buds with apical meristems and leaf primordia were observed after 50 d of induction (Fig. 3C and D). Discussion Effect of plant regulators on induction of adventitious shoots. The regeneration of adventitious shoots of novel F1 hybrids from lamina explants was dependent on the presence of both auxin and cytokinin in the medium. Our study findings suggest that BA produces more favorable results than zeatin and mT. Shoot multiplication of the novel F1 hybrid was most efficient when the medium was supplemented with 2.0 mM BA and 0.8 mM NAA. BA is one of the most effective and low-cost cytokinins and has thus been used in numerous plant micropropagation studies (Werbrouck et al., 1996). BA has also been commonly used in TC media to stimulate adventitious shoot induction in Begonia leaf explants (Espino et al., 2004; Kaviani et al., 2015; Kumari et al., 2017; Nakano et al., 1999). In this study, using BA alone resulted relatively HORTSCIENCE VOL. 53(12) DECEMBER 2018 unsatisfactory mean shoot elongation and high explant necrosis (Table 1), whereas combining BA with NAA improved the number of shoots and shoot length. A similar observation has also been reported in other Begonia species (Godo et al., 2008; Kumari et al., 2017; Mendi et al., 2009; Nada et al., 2011; Nakano et al., 1999), indicating that combining cytokinins at a high concentration with auxin is more effective for shoot multiplication compared with using cytokinins alone. The same situation was observed in a Begonia petiole transverse thin-cell-layer culture study (Nhut et al., 2005). Combining BA and a low concentration of auxin (NAA) has also promoted direct shoot regeneration and the formation of multiple shoots from leaf explants in various plants such as Lysimachia (Zheng et al., 2009) and Solanum (Ghimire et al., 2012). Meta-topolin has been used in adventitious shoot induction of B. semiparietalis (which belongs to section Coelocentrum) (Chung et al., 2016). Begonia sect. Coelocentrum is a highly diverse group and generally found in cave-like microhabitats of karst limestone (Chung et al., 2014; Hughes and Hollingsworth, 2008). Most sect. Coelocentrum species show isolated distribution patterns and are single-site endemic (Chung et al., 2014; Qin et al., 2017). Therefore, their genetic background variety may lead to different regeneration patterns. Rooting and acclimatization. Rooting is a critical step in plant TC. Adventitious shoots can be induced using high concentrations of cytokinin. However, such shoots inhibit rooting, leading to explants with inadequate development after acclimatization (Werbrouck et al., 1995). The balance of cytokinins and auxins is crucial when considering in vitro organ regeneration (Su et al., 2011). In this study, significant differences in the number of leaves were determined between the control and NAA-treated explants after 4 weeks in the rooting medium (Table 2). Endogenous auxin may be produced when more shoot formation occurs (Song et al., 2011); however, adding exogenous auxins to the culture may inhibit shoot growth. Comparing the effect of different concentrations on root induction and development revealed that PGR-free medium was superior to all the other treatments. This is in agreement with the finding of previous studies on Begonia micropropagation, in which the recovery of intact plants was easy even when the regenerated shoots were rooted in PGR-free medium (Chung et al., 2016; Espino et al., 2004). Werbrouck et al. 1859 (1995) reported that the BA that accumulates on the basal portion of the plant is slowly released during acclimatization, which may interfere with the rooting and acclimatization of micropropagated plantlets. Thus, our results suggest that the effect of a low concentration of BA (2.0 mM) in combination with NAA (0.8 mM) is not limited to in vivo growth. In general, no morphological or bloom variations occurred in the acclimated seedlings after 11 months in a greenhouse environment (Fig. 1D). Spectral quality affects morphogenesis of explants during in vitro culture. Plant seedling germination and morphogenesis are significantly regulated by light quality (Lee et al., 1996). Plants’ adaptation to light quality is usually associated with their strategy of adapting to shade stress in the natural environment (Smith, 1982). In a study on B. erythrophylla, Burritt and Leung (2003) proposed that the R/IR light ratio regulated the response of phytochrome to stimulate adventitious shoot production in TC, but blue light independently regulated the response through another photoreceptor cryptochrome. In our study, the possible regulation of blue light and the R/IR light ratio were also independent. The adventitious shoots of the novel F1 hybrid regenerated under equal distribution of B/G/R/IR light had more shoot development when compared with other shoots (Table 3). When the proportion of red light was higher or the proportion of blue light and IR light was lower (e.g., in the combination 1B7R1IR), the number of roots was lower and the inhibition of morphogenesis induction and shoot lengthening was greater. Our results also reveal that adventitious shoot formation occurs directly from the epidermal cells of the explant (Fig. 2), and different spectra of light directly stimulate the onset of organogenesis. Zhou et al. (2016) discovered that the induction and growth of the adventitious shoots of Anoectochilus roxburghii were significantly promoted by adding IR or blue light to red light, but they revealed that a high proportion of blue light was not conducive to shoot growth. In our study, strong blue light promoted shoot initiation. Our results also differ from those in a study on B. erythrophylla, in which adventitious shoot regeneration was inhibited using IR and blue light (Burritt and Leung, 2003). Both parent species belong to the Coelocentrum section of the Begonia genus and are restricted to the limestone karst area. The observed light signaling was different from that observed for other shade species possibly due to their adaptation to the unique habitat of karst caves. The notches of karst caves— usually housing Coelocentrum section plants—are commonly shady for long periods and exposed to oblique direct sun light for only short durations (Coombes et al., 2015). Although these places are as shady as those under the forest canopy, the light spectrum of such a unique environment should be more equally proportioned to different active wavelengths without reducing much of the blue and red light for upper 1860 leaf photosynthesis. The adaptation to shade contributes to their great potential as valuable indoor ornamental plants. Organogenesis could be accurately examined through histological and electron micrography techniques. Both techniques have been used for observing organogenesis in various explant species (Burritt and Leung, 1996, 2003; Chlyah and Tran Thanh Van, 1984; Hunter and Burritt, 2004; Pickens et al., 2006; Vatankhah et al., 2014). Organogenesis can occur either/both directly or indirectly through callus formation (Burritt and Leung, 1996; Ma et al., 2011). Our SEM observation (Fig. 3) revealed that regenerated primordia formed directly on the leaf epidermis. The ability to regenerate shoots from epidermal or subepidermal layers has gained considerable attention in histological Begonia studies (Burritt and Leung, 1996). A similar differentiation pathway of leaf epidermis was reported by Chlyah and Tran Thanh Van (1984) for Begonia rex. Moreover, histochemical staining results have ascertained that energy and carbon sources are required for cell differentiation. Therefore, starch accumulation (Fig. 2C and D) could play a crucial role in the differentiation and formation of apical meristems during in vitro organogenesis (Yeh et al., 2017). Accordingly, the relationship between starch accumulation in tissues and regeneration potential has been reported for many species (Chen and Ziv, 2005; Fortes and Pais, 2000). Studies have indicated that higher starch levels are correlated with higher regeneration potentials (Chen and Ziv, 2005; StoyanovaKoleva et al., 2012). Both B. montaniformis and B. ningmingensis var. bella are rare and difficult to regenerate under natural conditions. B. ningmingensis var. bella leaves fall easily, and B. montaniformis does not flower easily and is difficult to breed; therefore, they were artificially hybridized in this study. The patterns of leaf variegation of the F1 hybrid were closer to those of B. ningmingensis var. bella. The leaf variegation was richer along the veins, manifesting as white bands, and was sparsely distributed on the leaf surface. It contrasted sharply between the mesophyll and veins. The unique tipped conic bulla of the leaf surface of the B. montaniformis (Peng et al., 2015) was lacking in the hybrid, but the summit-shape remained to give the leaf a more stereoscopic shape. The selected F1 hybrids inherited the yellowish flowers and the long-lasting attribute from B. montaniformis (Fig. 1D). Under natural conditions, the period from seed germination to flowering of the F1 hybrid is 25 months. Our micropropagation protocol reduced this period to 16–17 months. Micropropagation not only preserved the unique hybrid characteristics but also shortened the breeding duration. In the presence of suitable amounts of PGRs, BA, and NAA, shoot and root productivity levels higher than 80% could be achieved. Moreover, the leaves of the F1 hybrids grew faster and lasted for 8–9 months. Our study successfully devised a method for massively breeding of stable hybrid begonia; this method therefore increases their potential for commercial application. Literature Cited Averyanov, L.V. and H.Q. Nguyen. 2012. Eleven new species of Begonia L. (Begoniaceae) from Laos and Vietnam. Turczaninowia 15:5–32. Burritt, D.J. and D.W.M. Leung. 1996. Organogenesis in cultured petiole explants of Begonia · erythrophylla: The timing and specificity of the inductive stimuli. J. Expt. Bot. 47:557–567. Burritt, D.J. and D.W.M. Leung. 2003. Adventitious shoot regeneration from Begonia · erythrophylla petiole sections is developmentally sensitive to light quality. Physiol. Plant. 118:289–296. Chen, C.C., D.C. Agrawal, M.R. Lee, R.J. Lee, C.L. Kuo, C.R. Wu, H.S. Tsay, and H.C. Chang. 2016. Influence of LED light spectra on in vitro somatic embryogenesis and LC–MS analysis of chlorogenic acid and rutin in Peucedanum japonicum Thunb.: A medicinal herb. Bot. Stud. 57:e9. Chen, J.X. and M. Ziv. 2005. The effects of storage condition on starch metabolism and regeneration potentials of twin-scales and inflorescence stem explants of Narcissus tazetta. In Vitro Cell. Dev. Biol. Plant 41:816–821. Chlyah, A. and M. Tran Thanh Van. 1984. Histological changes in epidermal and subepidermal cell layers of Begonia rex induced to form de novo unicellular hairs, buds, and roots. Bot. Gaz. 145:55–59. Chung, K.F., W.C. Leong, R.R. Rubite, R. Repin, R. Kiew, Y. Liu, and C.I. Peng. 2014. Phylogenetic analyses of Begonia sect. Coelocentrum and allied limestone species of China shed light on the evolution of Sino-Vietnamese karst flora. Bot. Stud. 55:e1. Chung, T.Y., L.P. Hsu, and W.H. Hu. 2016. Effect of meta-topolin on adventitious shoot regeneration in Begonia semiparietalis Y. Liu, S. M. Ku & C. I Peng. Propag. Ornam. Plants 16:79–83. Coombes, M.A., E.C. La Marca, L.A. Naylor, L. Piccini, J. De Waele, and F. Sauro. 2015. The influence of light attenuation on the biogeomorphology of a marine karst cave: A case study of Puerto Princesa Underground River, Palawan, the Philippines. Geomorphology 229:125–133. Espino, F.J., R. Linacero, J. Rueda, and A.M. V
azquez. 2004. Shoot regeneration in four Begonia genotypes. Biol. Plant. 48:101–104. Fang, D., S.M. Ku, Y.G. Wei, D.H. Qin, and C.I. Peng. 2006. Three new taxa of Begonia (sect. Coelocentrum, Begoniaceae) from limestone areas in Guangxi, China. Bot. Stud. 47:97–110. Fang, W., C.C. Chen, Y.I. Lee, and M.Y. Chang. 2011. Development of LED lids for tissue culture lighting. Acta Hort. 907:397–402. Fortes, A.M. and M.S. Pais. 2000. Organogenesis from internode-derived nodules of Humulus lupulus var. Nugget (Cannabinaceae): Histological studies and changes in the starch content. Amer. J. Bot. 87:971–979. Ghimire, B.K., C.Y. Yu, and I.M. Chung. 2012. Direct shoot organogenesis and assessment of genetic stability in regenerants of Solanum aculeatissimum Jacq. Plant Cell Tissue Organ Cult. 108:455–464. Godo, T., Y.X. Lu, J.X. Li, and K.Y. Guan. 2008. Comparisons of response for plant growth regulators in tissue culture of Begonia native to Yunnan, China. Bull. Bot. Gard. Toyama 13:41–46. HORTSCIENCE VOL. 53(12) DECEMBER 2018 Gu, C.Z., C.I. Peng, and N.J. Turland. 2007. Begoniaceae, p. 153–207. In: Z.Y. Wu, P.H. Raven, and D.Y. Hong (eds.). Flora of China, Vol. 13. Sci. Press, Beijing and Missouri Botanical Garden Press, St. Louis, MO. Hughes, M. and P.M. Hollingsworth. 2008. Population genetic divergence corresponds with species-level biodiversity patterns in the large genus Begonia. Mol. Ecol. 17:2643–2651. Hunter, D.C. and D.J. Burritt. 2004. Light quality influences adventitious shoot production from cotyledon explants of lettuce (Lactuca sativa L.). In Vitro Cell. Dev. Biol. Plant 40:215–220. Kaviani, B., D. Hashemabadi, H. Khodabakhsh, R. Onsinejad, M.H. Ansari, and N. Haghighat. 2015. Micropropagation of Begonia rex Putz. by 6-benzyladenine and a-naphthalene acetic acid. Intl. J. Biosci. 6:8–15. Kiew, R., J. Sang, R. Repin, and J.A. Ahmad. 2015. A guide to Begonias of Borneo. Nat. Hist. Publ. (Borneo), Sabah, Malaysia. Kumari, A., P. Baskaran, and J. Van Staden. 2017. In vitro regeneration of Begonia homonyma— A threatened plant. S. Afr. J. Bot. 109:174–177. Lee, D.W., K. Baskaran, M. Mansor, H. Mohamad, and S.K. Yap. 1996. Irradiance and spectral quality affect Asian tropical rain forest tree seedling development. Ecology 77:568–580. Lee, Y.I., W. Fang, and C.C. Chen. 2011. Effect of six different LED light qualities on the seedling growth of Paphiopedium orchid in vitro. Acta Hort. 907:389–391. Lee, Y.I., E.C. Yeung, N. Lee, and M.C. Chung. 2006. Embryo development in the lady’s slipper orchid, Paphiopedilum delenatii, with emphasis on the ultrastructure of the suspensor. Ann. Bot. 98:1311–1319. Li, C., L.H. Yang, D.K. Tian, Y. Chen, R.J. Wu, and N.F. Fu. 2016. Begonia leipingensis (Begoniaceae), a new compound-leaved species with unique petiolule pattern from Guangxi of China. Phytotaxa 244:45–56. Ma, G., J. L€ u, J.A. Teixeira da Silva, X. Zhang, and J. Zhao. 2011. Shoot organogenesis and somatic embryogenesis from leaf and shoot explants of Ochna integerrima (Lour). Plant Cell Tissue Organ Cult. 104:157–162. HORTSCIENCE VOL. 53(12) DECEMBER 2018 Mendi, Y.Y., P. Curuk, E. Kocaman, C. Unek, S. Eldogan, G. Gencel, and S. Cetiner. 2009. Regeneration of begonia plantlets by direct organogenesis. Afr. J. Biotechnol. 8:1860–1863. Murashige, T. and F. Skoog. 1962. A revised medium for rapid growth and bioassays with tobacco tissue cultures. Physiol. Plant. 15:473– 497. Nada, S., S. Chennareddy, S. Goldman, S. Rudrabhatla, S.D. Potlakayala, P. Josekutty, and K. Deepkamal. 2011. Direct shoot bud differentiation and plantlet regeneration from leaf and petiole explants of Begonia tuberhybrida. HortScience 46:759–764. Nakano, M., Y. Niimi, D. Kobayashi, and A. Watanabe. 1999. Adventitious shoot regeneration and micropropagation of hybrid tuberous begonia (Begonia · tuberhybrida Voss). Scientia Hort. 79:245–251. Nhut, D.T., N.T. Hai, P.X. Huyen, D.T.Q. Huong, N.T.T. Hang, and J.A. Teixeira da Silva. 2005. Thidiazuron induces high frequency shoot bud formation from Begonia petiole transverse thin cell layer culture. Propag. Ornam. Plants 5:151–157. Peng, C.I., C.W. Lin, H.A. Yang, Y. Kono, and H.Q. Nguyen. 2015. Six new species of Begonia (Begoniaceae) from limestone areas in Northern Vietnam. Bot. Stud. 56:e9. Pickens, K.A., J. Wolf, J.M. Affolter, and H.Y. Wetzstein. 2006. Adventitious bud development and regeneration in Tillandsia eizii. In Vitro Cell. Dev. Biol. Plant 42:348–353. Qin, Y.H., Y.Y. Liang, W.B. Xu, C.W. Lin, and C.I. Peng. 2017. Begonia ufoides (sect. Coelocentrum, Begoniaceae), a new species from limestone areas in central Guangxi, China. Phytotaxa 316(3):279–284. Smith, H. 1982. Light quality, photoperception, and plant strategy. Annu. Rev. Plant Physiol. 33:481–518. Sodhi, N.S., L.P. Koh, B.W. Brook, and P.K.L. Ng. 2004. Southeast Asian biodiversity: An impending disaster. Trends Ecol. Evol. 19:654–660. Song, J.Y., N.S. Mattson, and B.B. Jeong. 2011. Efficiency of shoot regeneration from leaf, stem, petiole and petal explants of six cultivars of Chrysanthemum morifolium. Plant Cell Tissue Organ Cult. 107:295–304. Stoyanova-Koleva, D., M. Stefanova, M. Zhiponova, and V. Kapchina-Toteva. 2012. Effect of N6benzyladenine and indole-3-butyric acid on photosynthetic apparatus of Orthosiphon stamineusplants grown in vitro. Biol. Plant. 56:607– 612. Su, Y.H., Y.B. Liu, and X.S. Zhang. 2011. Auxin– cytokinin interaction regulates meristem development. Mol. Plant 4:616–625. Vatankhah, E., V. Niknam, and H. Ebrahimzadeh. 2014. Histological and biochemical parameters of Crocus sativus during in vitro root and shoot organogenesis. Biol. Plant. 58:201–208. Werbrouck, S.P.O., B. Van der Jeugt, W. Dewitte, E. Prinsen, H.A. Van Onckelen, and P.C. Debergh. 1995. The metabolism of benzyladenine in Spathiphyllum floribundum schott ‘petite’ in relation to acclimatization problems. Plant Cell Rpt. 14:662–665. Werbrouck, S.P.O., M. Strnad, H.A. Van Onckelen, and P.C. Debergh. 1996. Meta-topolin, an alternative to benzyladenine in tissue culture? Physiol. Plant. 81:291–297. Yeh, C.H., F.S. Liao, K.L. Huang, I. Miyajima, and Y.I. Lee. 2017. An efficient protocol of protocorm-like bodies regeneration from callus cultures of gastrodia elata blume and the further associations with mycorrhizal fungi. J. Fac. Agr. Kyushu Univ. 62(1):39–46. Yeung, E.C. 1999. The use of histology in the study of plant tissue culture systems—Some practical comments. In Vitro Cell. Dev. Biol. Plant 35:137–143. Zheng, W., X.D. Xu, H. Dai, and L.Q. Chen. 2009. Direct regeneration of plants derived from in vitro cultured shoot tips and leaves of three Lysimachia species. Scientia Hort. 122:138– 141. Zhou, S.W., R.N. Li, W.W. Huan, Z.G. Xu, X.Y. Liu, and X.L. Jiao. 2016. Effects of the spectral energy distribution of red, blue, and far-red light on the induction of Anoectochillus rozburghii (Wall.) Lindl. adventitous shoots. Propag. Ornam. Plants 16:39–46. 1861