In vitro germination and bulblet and shoot propagation for wild edible Eremurus spectabilis M.Bieb

Eremurus spectabilis M.Bieb is consumed as a vegetable because of its nutritious characteristics. The plants are also used for medicinal purposes, in the cut flower industry as an ornamental geophyte, and in industry as a natural adhesive. The aim of the present study was to improve the in vitro propagation protocol for germination and bulblet/shoot formation of E. spectabilis. For this purpose, E. spectabilis seeds were in vitro germinated in four different nutrient media: Murashige and Skoog (MS), Gamborg (B5), White (WH), and Shenk and Hildebrandt (SH). To stimulate bulblet and/or shoot regeneration, hypocotyls of 35-40-day-old in vitro-germinated plantlets were cut into 0.5-1.0 cm pieces, and the resultant explants were cultured in MS media containing 2,4-dichlorophenoxyacetic acid (2,4-D) (0.5, 1.0, 2.0, and 4.0 mg L) + kinetin (0.5 mg L), thidiazuron (TDZ) (0.5, 1.0, 2.0, and 4.0 mg L) + 1-naphthylacetic acid (NAA) (0, 0.1, 0.5, and 1.0 mg L), and 6-benzylaminopurine (BAP) (0.5, 1.0, 2.0 and 4.0 mg L) + 2,4-dichlorophenoxyacetic acid (2,4-D) (0, 0.1, 0.5, and 1.0 mg L). The best outcomes for germination ratio (57.5%) were obtained from the B5 medium. In the third set of in vitro propagation experiments, 100% bulblet formation was achieved in TDZ (0.5 mg L ) and NAA (0.5 and 0.1 mg L) combinations of MS media, and this was followed by 0.5 mg L BAPcontaining medium (81.3%). Shoot formation ratios with the same media combinations varied from 60-70%, and the number of shoots per explant varied from 1.4-2.4 shoots. Further in vitro propagation research is planned with larger bulb sizes to develop a protocol for rooting bulblets and/or shoots.

over the pedicle. The fruit is green and has three united carpels (Figure 1c). The seeds are 3-4 mm long, trigonal in shape, with sharp edges, narrow wings, and a brownish colour ( Figure 1d) (Matthews, 1986).
E. spectabilis is quite rich in antioxidants, phenolics, and minerals and thus, it is a highly nutritious plant (Tosun et al., 2012;Tuzcu et al., 2017). The shoots and leaves are cooked and consumed as vegetables. In addition, different sections of the plant are used for medicinal purposes, especially for the treatment of fungal diseases, diabetes, hepatitis, liver and stomach disorders and some cancers (Baytop, 1999;Tuzlacı and Doğan, 2010;Pourfarzad et al., 2014;Abubaker and Hidayat, 2015;Tuzcu et al., 2017). As well as medical significance, E. spectabilis is also used in cut flower production as a popular ornamental geophyte because of its long and alluring spikes (Schiappacasse et al., 2013;Ahmad et al., 2014). Plant roots are also dried and ground to produce a natural adhesive in industry (Dashti et al., 2005;Heshmatol Vaezin et al., 2010;Eghtedarnejad and Mansouri, 2016).
In vitro regeneration techniques are commonly used for the regeneration of valuable wild species, endangered species, endemic species, and hard-to-regenerate species. Only one study has been conducted on the in vitro regeneration of E. spectabilis worldwide (Tuncer, 2017). The researcher cultured E. spectabilis leaf and rhizome explants in BAP (1.0 and 2.0 mg L -1 ) and 2,4-D (0.5, 1.0, and 2.0 mg L -1 ) containing MS medium (Murashige and Skoog, 1962). The researcher reported that in vitro shoot regeneration was not achieved in either explant types. In vitro regeneration studies have also been conducted in different species of the Liliaceae family, including the Aloe (Molsaghi et al., 2014;Baskaran et al., 2015;Sharma et al., 2015;Suh et al., 2015;Razib et al., 2016), Lilium (Ault and Siqueira, 2008;Tang et al., 2009;Sun and Jin, 2011), Chlorophytum (Nurashikin et al., 2010;Pandey, 2016), Fritillaria (Çakmak et al., 2016), and Scilla (Kamaleswari et al., 2016) species. In these studies, different in vitro shoot and/or bulblet formation ratios were achieved based on the species and the nutrient media combinations.
The first stage of the protocol to be developed includes in vitro germination, in vitro bulblet and shoot formation. For this purpose, the aim of the present study was to investigate the possibilities of in vitro germination of E. spectabilis seeds in four different nutrient media: Murashige and Skoog (MS), Gamborg (B5), White (WH), and Shenk and Hildebrandt (SH), and in vitro bulblet and/or shoot regeneration from hypocotyl and bulblet explants developed under in vitro conditions with different plant growth regulator (PGR) combinations of MS medium. This study will be the first comprehensive report worldwide about the in vitro propagation of E. spectabilis.

Plant material and sterilization processes
The E. spectabilis seeds to be used in the experiments were supplied from the Gürpınar (38° 8' 20.75'' N, 43° 30' 55.15'' E, 1730 m) in Van province of Turkey in August 2017 ( Figure 1d). The seeds were initially kept in 0.3% benomyl solution for an hour to remove fungal disease agents (Figure 2a), and then shaken and kept in distilled water for an hour. The seeds were then placed into a laminar flow sterile cabin, kept in 40% sodium hypochlorite (NaOCl) solution containing 1-2 drops of Tween-20 for 10 minutes, and then washed through sterile bidistilled water for five minutes three times (Figure 2b-c). The surface-sterilized seeds were placed over filter papers to remove excess water. The materials to be used in the germination and regeneration experiments (petri dishes, glass jars, forceps, bistoury, etc.) were sterilized in an autoclave at 121 °C for 1.5 hours, and the nutrient media were sterilized for 20 minutes.
In vitro germination trials Tip-cut seeds were germinated in MS (Murashige and Skoog, 1962), SH (Schenk and Hildebrandt, 1972), B5 (Gamborg et al., 1968), and WH (White, 1943) media under in vitro conditions (Figure 2d-e). All nutrient media were supplemented with 7 g L -1 agar, 0.75 g L -1 GA3, and 50 mg L -1 citric acid. The pH of all the media was adjusted to 5.8 (by adding 1 N NaOH or 1 N HCl). 20 g L -1 of sucrose was added to the B5 medium, while 30 g L -1 of sucrose was added to the MS and SH media. WH media did not contain sucrose. Following seeding, the petri dishes were wrapped in parafilm and preserved in a fridge at 4 °C until the initiation of germination to break the physiological dormancy of the seeds. Following the initiation of germination, the petri dishes were taken into an incubator set at 13-14 °C with dark conditions. The emergence of about 2 mm of radicle from the seed test was considered the criterion for germination ( Figure 2e). The following equations were used to calculate the germination parameters (Abdul- Baki and Anderson, 1973;Murillo-Amador et al., 2002;Keskiner and Tuncer, 2019): Germination ratio (%) = (G/T) × 100 Mean germination time (day) = [(1.day G × 1. day) + (2.day G × 2.day) +…+ (n.day G × n.day)] /Total G Germination speed (day) = n1/t1 + n2/t2 +………. + nn/tn n1, n2: number of germinated seeds, t1, t2: number of days for germination, G: germinated seed number; T: total number of seeds
In the second and third sets of experiments, bulblet segment explants developed under in vitro conditions were used to stimulate bulblet and shoot formation. In the second set of experiments, explants were cultured in 17 PGRs in combination with MS media selected from the first set of experiments. In the third set of experiments, explants were cultured in 9 PGRs combinations selected from the second set of experiments. All the cultures were incubated at 25 °C ± 2 °C for 16/8 h photoperiod with a light intensity of 3000 lux.

Statistical analysis
The data were statistically analyzed using analysis of variance (ANOVA) with Statgraphics statistical software, followed by Duncan's multiple range test comparisons for significant differences. The germination experiments were conducted in a completely randomized block design with four replications and five petri dishes (12 seeds/petri dish) in each replication; the in vitro bulblet and shoot regeneration experiments were also set up in a randomized blocks design with three replications with 4-6 explants in the first set of experiments, 5-7 explants in the second set of experiments, and 5-8 explants in the third set of experiments for each replication.

In vitro germination
The in vitro germination trial results are summarized in Table 1. The greatest germination ratio (57.5%) and the shortest germination time (3.7 day) were obtained from the B5 medium. The germination ratios of the other nutrient media (MS, WH, and SH) varied between 35.0-38.7%, and mean germination times varied between 4.4-5.7 days; they were placed into the same statistical group (Table 1).
According to the results of the first set of experiments, 17 PGR combinations in MS media with more successful outcomes were selected (Table 3), and the second set of experiments were set up (35-40 days after the first set of experiments). Bulblet pieces developed under in vitro conditions were used as explants. The differences in the investigated parameters of the 17 PGR combinations in MS media were found to be significant (p < 0.01). The best outcomes in terms of bulblet formation ratios were obtained, respectively, from the 21 st (0.5 mg L -1 BAP) (83.3%), the 7 th (0.5 mg L -1 TDZ + 0.5 mg L -1 NAA) (81.6%), the 14 th (2 mg L -1 TDZ + 0.1 mg L -1 NAA) (75.7%), and the 17 th (4 mg L -1 TDZ) (75.7%) media. Except for a few media, (3 rd , 16 th , and 17 th media), browning was not observed in the explants (Table 3). In terms of shoot formation ratios, the 21 st (0.5 mg L -1 BAP) (83.3%) and the 7 th (0.5 mg L -1 TDZ + 0.5 mg L -1 NAA) media (63.3%) were the most successful. The greatest number of shoots per explant was obtained from the 21 st and the 17 th media, and they were followed by the 20 th (4 mg L -1 TDZ + 1 mg L -1 NAA), the 6 th (0.5 mg L -1 TDZ + 0.1 mg L -1 NAA), and the 7 th (0.5 mg L -1 TDZ + 0.5 mg L -1 NAA) media, which were placed into the same statistical group (Table 3).
For the third set of experiments, 9 PGR combinations with better outcomes in terms of bulblet and shoot formation were selected. Bulblet pieces developed under in vitro conditions were used as explants, and the resulting data are provided in Table 4. In the third set of experiments, a 100% bulblet formation ratio was achieved in TDZ (0.5 mg L -1 ) and NAA (0.5 and 1.0 mg L -1 ) combinations of MS media. These PGR combinations were followed by the MS medium containing only 0.5 mg L -1 BAP (81.3%) ( Table 4). The shoot formation ratios of PGR combinations varied from 10.0-70.0%, and the number of shoots per explant varied from 0.5-2.4 shoots. The greatest shoot formation ratio (70%) was obtained from the MS medium containing 0.5 mg L -1 TDZ and 1.0 mg L -1 NAA and the greatest number of shoots per explant (2.4 shoots) was obtained from the MS medium containing 0.5 mg L -1 TDZ and 0.5 mg L -1 NAA (Table 4). Figure 3 presents the in vitro bulblet and shoot regeneration in different PGR combinations of MS media.

Discussion
There are only two studies about in vitro germination of E. spectabilis; these were conducted by Keskiner (2017) and Akdağ (2019). Keskiner (2017) germinated 1,500 ppm GA3-supplemented and unsupplemented E. spectabilis seeds in three different nutrient media (MS, WH, and B5) and reported quite low germination ratios (0.83%-1.50%) only in the WH medium. Akdağ (2019) germinated E. spectabilis seeds in three different nutrient media (MS, WH, and B5) under in vitro conditions, preserved the seeded petri dishes in a fridge (4 °C) for different durations (30, 50, 80, and 100 days), and reported the greatest germination ratio (5.88%) for 100-day cold storage with MS medium. However, it is remarkable that both Keskiner (2017) and Akdağ (2019) reported quite low germination ratios under in vitro conditions.
Other studies (not under in vitro conditions) have investigated the germination of E. spectabilis. Rahmanpour et al. (2005) subjected E. spectabilis seeds to mechanical abrasion followed by immersion into 0.01 M GA3 solution for 45 minutes and reported a 53% germination ratio. Güngör (2002) reported the greatest germination ratio (65.3%) for 90-day moist-cold stratification treatments, and Keskiner and Tuncer (2019) reported the greatest germination ratio (73.3%) for 100-day moist-cold stratification treatments. Despite the high germination ratios, quite long moist-cold stratification durations (90 and 100 days) were used, and long mean germination times were observed in those studies. Akdağ (2019) cut the tips of E. spectabilis seeds, subjected them to moist-cold stratification treatments for different durations (30, 50, 80, and 100 days) and reported germination ratios from 45.4-60.6%; high germination ratios were achieved even with a short duration (< 30 days) of moist-cold stratification treatments of the tip-cut seeds. Akdağ's (2019) results were taken into consideration in this study, and thus the in vitro germination experiments were set up with tip-cut E. spectabilis seeds. Therefore, the present findings comply with the results of Akdağ (2019).
In previous studies conducted with other species of the Eremurus genus under non-in vitro conditions, Rahmanpour et al. (2007) reported the greatest germination ratio (80%) and germination speed (1.6 day) of E. olgae for seeds immersed in water for 24-48 hours, tip cut, seed testa abraded, and immersed into a 0.08 M GA3 solution for 45 minutes; it was reported that tip cutting and abrasion of seed testa improved germination ratios and shortened mean germination times in E. spectabilis (Rahmanpour et al., 2005;Akdağ, 2019), in E. olgae (Rahmanpour et al., 2007). Present findings comply with the results of Rahmanpour et al. (2005), Rahmanpour et al. (2007) and Akdağ (2019). Sufficient germination ratios were achieved from tip-cut seeds preserved in a fridge (4 °C) for short durations (< 30 days) under in vitro conditions.
Only one study worldwide has investigated the in vitro propagation of E. spectabilis. Tuncer (2017) cultured leaf and rhizome explants of E. spectabilis in BAP (1.0 and 2.0 mg L -1 ) and 2,4-D (0.5, 1.0, and 2.0 mg L -1 ) combinations of MS media, but was not able to achieve any bulblet and shoot formations. In the present study, several different PGR combinations were experimented with in MS media, and sterile hypocotyl and bulblet explants obtained under in vitro conditions were used; thus, the findings of the present study were quite successful compared with Tuncer (2017). In the first set of experiments, in vitro direct bulblet formation was achieved from the hypocotyl explants; in the second and third set of experiments, both bulblet and shoot formations were achieved from bulblet segments developed under in vitro conditions with different PGR combinations on MS medium. In the third set of experiments, a bulblet formation ratio of 81.3-100% was achieved with TDZ (0.5 mg L -1 ) + NAA (0.5 and 1.0 mg L -1 ) combinations and 0.5 mg L -1 BAP-containing MS media; shoot formation ratios of 60-70% were also achieved with the same nutrient media combinations (Table  4).
In vitro bulblet and/or shoot formations were also reported in previous studies in which bulblet segments were used as explants. Haibin and Jiajun (2006) reported the greatest shoot regeneration of Lilium lancifolium in MS + BA (1.5 mg L -1 ) + NAA (0.2 mg L -1 ) media. Yang et al. (2010) reported the greatest shoot regeneration of L. tsingtauense in MS + 0.1 mg L -1 NAA + 2.0 mg L -1 BA combination. Sun and Jin (2011) reported the greatest shoot regeneration of L. longiflorum in MS + 2 mg L -1 BA + 0.2 mg L -1 NAAsupplemented media. Zhang and Jia (2014) reported the greatest shoot regeneration (88.91%) of Lily (cv. Siberia) in MS + 1.07 μM NAA + 4.44 μM BA-supplemented media. Çakmak et al. (2016) worked with bulblet explants taken from in vitro-developed Fritillaria persica seedlings and obtained the maximum bulblet regeneration ratio and number of bulblets per explant from 2.0 mg L -1 TDZ-supplemented MS medium. In another study, maximum number of shoots (13.2 shoots) of bulblet pieces of Scilla hyacinthina was obtained from 2 mg L -1 TDZ-supplemented MS medium (Kamaleswari et al., 2016). Complying with the findings of the above-mentioned reports (Haibin and Jiajun, 2006;Yang et al., 2010;Sun and Jin, 2011;Zhang and Jia, 2014;Çakmak et al., 2016;Kamaleswari et al., 2016), in the second and third set of my experiments (Table 3 and 4), bulblet and shoot regeneration were achieved from in vitro-developed bulblet segments at varying ratios based on PGR combinations.

Conclusions
This study is the first comprehensive report worldwide about the in vitro propagation of E. spectabilis. In the present study, bulblet and shoot regeneration from in vitro-developed hypocotyl explants and bulblet segments were investigated in several PGR combinations of MS media. The most successful outcomes in terms of bulblet and shoot formation were achieved in only BAP (0.5 mg L -1 )-containing and in TDZ (0.5 mg L -1 ) + NAA (0.5 and 1.0 mg L -1 ) combinations of MS media. In further studies of in vitro propagation of this species, besides TDZ + NAA combinations, BAP + NAA combinations could also be experimented with, bulb size could be increased, and a protocol for rooting the resultant bulblets and/or shoots could be developed.