Somatic embryo induction and plantlet regeneration of Canna × generalis from immature zygotic embryo

Somatic embryogenesis is a unique method of in vitro regeneration, which can be used in plant reproduction, germplasm conservation, and molecular-assisted breeding. The results showed that the optimum medium for embryogenic callus induction was MS + 6 mg L −1 6-BA + 1.5 mg L −1 TDZ + 0.5 mg L −1 NAA + 30 g L −1 sucrose + 7 g L −1 agar, and the induction rate was 47.45%. The best somatic differentiation medium was MS + 2 mg L −1 6-BA + 1.5 mg L −1 TDZ + 30 g L −1 sucrose + 7 g L −1 agar, and the induction rate of somatic embryos was 54.45%. The optimum medium for embryoid proliferation was MS + 6 mg L −1 6-BA + 1 mg L −1 NAA + 0.2 mg L −1 TDZ, and the proliferation rate and the multiplication coefficient reached 46.33% and 7.83, respectively. The mature somatic embryos were put into MS, B5, and 1/2MS medium for seedling culture. In MS medium, true leaves grew, complete plants were obtained, and the seedling rate was 88.00%. At the same time, the survival rate of transplanting seedlings in the mixed matrix (peat: organic fertilizer: soil = 1:1:1) was as high as 98%. Cytological observation showed that the somatic embryos underwent globular, heart-shaped, torpedo, and cotyledon stages. This study established a regeneration system of C. × generalis with excellent somatic embryos, and provided basic technical support for the large-scale commercial propagation and germplasm resources protection. It will lay a foundation for further research on gene function and breeding new varieties and ideal research materials for the study of somatic embryogenesis mechanism and genetic transformation of C. × generalis.


Introduction
Canna spp., also named canna lily, belongs to the Canna genus, Canna family, native to tropical America and Africa and is a perennial bulbous flower.It is one of the valuable ornamental plants with long-term blossoming (Mitrofanova et al. 2017) and is widely used in landscaping and purifying sewage, repairing and protecting ecological environment resulting from the advantages of strong growth, wide adaptability, easy cultivation, etc. (Dong et al. 2019;Zhang et al. 2021;Zhao et al. 2022).In addition, canna lily also has high economic value and medicinal value (Singh et al. 2019).Its flowers can be used as a natural source of anthocyanin pigments in the food ingredient industry (Srivastava and Vankar. 2010), and its roots can be used in Thai folk medicine to treat AIDS (Woradulayapinij et al. 2005).The rhizomes contain various chemical components and can treat hepatitis, stomachache, analgesia, hemostasis, and anthelmintic activity (Nirmal et al. 2007), prevent abscesses, wound healing cardiovascular and free radical-related diseases (Mahmoud et al. 2021).Currently, the propagation of the canna lily is mainly by splitting the rhizome, and breeding new canna cultivars rely on traditional breeding methods, which have brought many disadvantages.For example, the reproduction coefficient of conventional methods is extremely low, and long-term clonal propagation is prone to virus accumulation and variety degradation (Mishra et al. 2015).These traditional propagation methods cannot meet the commercial demand in the garden.In vitro culture of canna lily has important effects on genetic improvement and variety selection (Singh et al. 2019).Therefore, re-establishing a reliable regeneration system will lay a foundation for this plant's large-scale micropropagation and breeding engineering.
Plant tissue culture effectively produces and propagates plants rapidly and in large quantities under certain conditions.It plays an effective role in plant conservation, mass propagation, genetic improvement, and secondary metabolite production (Wang et al. 2021).Tissue culture and breeding technology have a high degree of genetic stability and can maintain the good characteristics of clonal mothers (Oh et al. 2010).In the tissue culture of the plant, there are two ways to regenerate plants in vitro.One is to obtain regenerated plants through organogenesis, and the other is to obtain regenerated plants by somatic embryogenesis.Somatic embryogenesis is a process in which a single cell develops into an embryo.In this process, somatic cells can differentiate into totipotent embryonic stem cells and then further differentiate and develop into complete plants (Horstman et al. 2017), which have the advantages of relatively stable genetic characteristics, large number, fast speed, complete structure, and high regeneration rate (Piovan et al. 2014).Somatic embryogenesis plays a vital role in plant regeneration and rapid reproduction (Tomiczak et al. 2019).The method has also been successfully applied to bananas (Shchukin et al. 1997;Natarajan et al. 2020), lilies (Yan et al. 2020), peonies (Du et al. 2020), mango ginger (Raju et al. 2013) and other ornamental plants.Moreover, it is of great value for the large-scale propagation of superior germplasm resources and has many potential application prospects (Egertsdotter et al. 2019).
So far, the rhizomes (Wafa et al. 2016;Purshottam et al. 2019), terminal bud (Sakai and Imai. 2007), and other explants have been used to regenerate plants through organogenesis in Canna propagation in vitro.Singh et al. (2019) identified the optimal hormone concentrations for adventitious bud regeneration of Canna rhizomes as 6-BA 2 mg L −1 , TDZ 1.5 mg L −1 , and IAA 0.1 mg L −1 , with the highest regeneration rate of 36%.Wafa et al. (2016) induced the regeneration of Canna plants on 6-BA 3 mg L −1 and NAA 1.5 mg L −1 medium, and the induction rate of adventitious buds was 73.3%, successfully establishing the regeneration system.However, the regeneration system produced by the traditional organotype method also had some defects.For example, the plants were seriously polluted in the process of culture, the browning rate was high, and the reproduction coefficient and the differentiation rate were low (Purshottam et al. 2019;Mitrofanova et al. 2017).Previous studies have reported that the leaves of C. indica L. were used to induce callus and regenerated plants by somatic embryogenesis.The callus was cultured in 2 mg L −1 6-BA medium for 10 to 15 days to form somatic embryos.Each culture flask produced an average of 4-5 individual cell embryos (Mishra et al. 2015).This is the only report about the canna regeneration by somatic embryogenesis, but the induction rate was lower than in the other plants.Somatic embryogenesis (SE) was found to be the best method for plant cultivation, germplasm conservation, polyploidy breeding, and genetic transformation (Martinez et al. 2019).Therefore, effectively improving the rate of somatic embryo acquisition and induction efficiency was the main problem of Canna by SE induction.The production of SE is strongly influenced by explant sources, genotypes, culture conditions, exogenous growth regulators, and the plant growth regulators (PGRs) added to the culture medium (Kanchanapoom et al. 2011).It is expected to provide new ideas and technical references for the study of gene function and genetic improvement of Canna.
C. × generalis is one of the two most widely used cannas in gardens (Bailey 1923) and is an important material for improving canna cultivars (Li 2022).The purpose of this study is to investigate the effects of different plant growth regulators, medium types, and culture conditions during the process of induction and germination of SE from the immature embryo of C. × generalis.Establish an efficient system for somatic embryo induction and regeneration and make clear the stages and the structural characteristics of SE formation and development by histomorphological observation.This study will provide an effective regeneration method for genetic improvement, germplasm innovation, and a new variety selection of C. × generalis.

Plant material
After anthesis for 30 days, the immature fruitwas taken from the germplasm resources garden of the canna lily (C.× generalis cv.'Mohong') at Guizhou University from July to October 2021.First, the fruits were washed with liquid detergent for 10 min and rinsed with distilled water 5 times.Next, the immature seeds were taken from the fruits on the ultra-clean workbench and placed in a beaker.Then the seeds were treated with 75% ethanol for 30 s, rinsed 5-6 times using sterile distilled water, disinfected with 0.1% (w/v) mercuric chloride for 7 min, rinsed 3-5 times with sterile distilled water, and dried on a sterile filter paper.After that, the zygotic embryos were separated aseptically from the seeds and inoculated into medium for embryogenic callus induction.

Embryogenic callus induction
MS medium was added with 30 g L −1 sucrose, and 7 g L −1 agar was used as the basic media (pH 5.8).Cultivation was conducted with different concentration plant growth regulators, namely, 6-BA (2.0, 4.0, 6.0 mg L −1 ), NAA (0.1, 0.2, 0.5 mg L −1 ), and TDZ (1.0, 1.5, 2.0 mg L −1 ).Nine treatments were selected for the three-factor and three-level orthogonal experimental design.The young embryo was inoculated with 30 explants per treatment, replicated three times.The growth of embryogenic callus was observed weekly,when callus was induced, the explants were transferred to a fresh induction medium every two weeks.Embryogenic callus induction rate was counted after inoculating for 30 days according to the following formula.
On the 15th day of induction, two kinds of calli of different shapes and forms were selected to observe the occurrence of embryonic callus microscopically.First, the MS medium on the surface was removed by gently washing it with sterile distilled water.The samples were then sucked dry, observed and photographed with an optical microscope (DM 750 LED), and the embryonic callus with distinct characteristics was selected for paraffin section observation.

Somatic embryogenesis induction
Embryogenic callus with uniform growth were cut into 0.5 cm × 0.5 cm in size, inoculated on somatic cell embryo Embryogenic callus induction rate (%) = number of explants induced to obtain embryogenic callus total number of inoculated explants × 100% induction medium.Based on previous studies (Mishra et al. 2015), the specific settings were designed and tested at 6-BA (0.0, 1.0, 1.5, 2.0, 2.5, 3 mg L −1 ) and TDZ (0.0, 0.5, 1, 1.5, 2, 2.5 mg L −1 ), which was showed in Table 2. Every treatment was repeated 3 times and 10 bottles per repeat.After culture for 30 days, the induction rate of somatic embryos was calculated according to the following formula.
In order to observe somatic embryogenesis microscopically, samples with obvious somatic embryo characteristics were collected every 5 days at different somatic embryo development stages and then gently washed with sterile distilled water to remove MS medium on the surface.The samples were then sucked dry, observed and photographed using a light microscope (LEICA ICC50W), and taken for paraffin section observation.

Somatic embryo proliferation culture
The embryoids at globular embryo stage were inoculated on the embryoid proliferation medium with different combinations of plant growth regulators.Design based on preliminary experimental results, and carried out with 6-BA (1.0, 3.0, 6.0 mg L −1 ), NAA (1.0, 1.5 mg L −1 ), and TDZ (0.2, 0.5 mg L −1 ), respectively.Each treatment was replicated three times and 10 bottles per repeat, and the growth status of embryoid proliferation was counted for 30 days.

Somatic embryo germination and seedling formation
The white or green normal somatic embryos about 1 cm long were inoculated into the somatic embryo germination medium (Table 4) for somatic embryo germination culture.Each treated material was inoculated with 25 individual cell embryos, and replicated three times.Germination rate and seedling rate were observed and counted after culturing for 30 days according to the following formula.

Acclimation and transplantation
Vigorous embryoid seedlings in bottles were selected and transferred to the greenhouse for 1 day to adapt to the greenhouse environment and then remove the bottle cap to train the seedlings for 2 days.Finally the training seedlings were transplanted to the matrix: organic fertilizer: peat: soil = 1:1:1 (v:v:v) for seedling culture.After transplanting the regenerated plants successfully, the leaf morphological differences between the regenerated plants and the mother plants were compared to test for genetic stability.

Histomorphology observation
Two kinds of calli and somatic embryos at different stages were fixed overnight with 75% FAA (formaldehyde, glacial acetic acid, 75% ethanol 1:1:18).After discarding the fixing solution, the somatic embryos were treated with 70, 80, 95 and 100% ethanol successively, and soaked in anhydrous ethanol twice to ensure complete dehydration.It is then treated with xylene for transparency.After paraffin embedding, the wax blocks were divided according to the material placement position, and the divided wax blocks were trimmed into trapezoids.It was placed at 4 °C for 24 h, and paraffin sections were prepared.The embedded wax blocks were cut into 8-10 μm slices using a microtome (Erma, Japan), and the slides were dried in a slide dryer at 38 ℃.The sections were observed and photographed under an optical microscope (LEICA ICC50W) after dewaxing, ethanol remixing, zorubin green staining, and neutral gum seal.The different stages of SE were recorded, and the dynamic changes in cells and their origin during SE were observed.

Culture conditions
The basic medium used in this experiment was MS medium, and the reagents were purchased from Beijing Solarbio Biological Company.The medium pH of all experiments was 5.8 ± 0.2, sterilized at 121℃ for 25 min.Unless otherwise specified, MS solid medium was supplemented with 7 g L −1 agar as a coagulant and 30 g L −1 sucrose.All plants were cultured under a light intensity of 25-35 µmol m −2 s −1 with 16/8 h (light/dark) photoperiods at room temperature of (25 ± 2) ℃.

Statistical analysis
Excel software (Excel 2017) was used for statistical data.SPSS software (SPSS 20.0) was used for data processing and one-way analysis of variance (ANOVA), and Duncan's multiple tests were performed.All test treatments were set with three biological replicates.

Results and discussions
The effects of plant growth regulators on embryogenic callus induction  ability in long-term culture (Martins et al. 2022).To sum up, MS + 6 mg L −1 6-BA + 1.5 mg L −1 TDZ + 0.5 mg L −1 NAA is the best medium combination for inducing embryogenic calli from the young embryo of C. × generalis in this study.

Induction of somatic embryogenesis
Plant growth regulators are generally the most critical factors in somatic embryo induction and development (Feheir et al. 2003;Shekhawat et al. 2021).The most commonly used are 2,4-D, IBA, GA 3 and 6-BA (Stasolla and Yeung 2003;Miroshnichenko et al. 2017).Cytokinins are essential for somatic embryogenesis (Pereira et al. 2020), and 6-BA has a certain promotion effect on somatic cell embryogenesis (Martins et al. 2022).In this study, we have tested different cytokinin combinations for somatic embryogenesis.The number of somatic embryos induced on MS + 2 mg L −1 6-BA + 1.5 mg L −1 TDZ + 30 g L −1 sucrose + 7 g L −1 agar was the highest, and the induction rate of somatic embryos (SE) was 54.45% (Table 2).It was significantly higher than other combinations.At this time, the effect of 6-BA was significantly obvious on the induction rate of somatic cells (P = 0.01 < 0.05), while TDZ (P = 0.66 > 0.05) had no significant effect.On average, 10-15 somatic embryos produced per 0.5 g embryogenic callus with good condition, they were usually white at the early stage and then turn to light green or dark green (Fig. 2).Our induction somatic embryos were 4-5 more than previous research results (Mishra et al. 2015).The induced somatic embryo can be subcultured 5 times, and the activity  of the somatic embryo can be reduced with the increase of the subculture times to the sixth time, which may be due to the gradual aging of plant cells and the decline of totipotency (Murashig et al. 1965).
Morphological and histological studies were still crucial in the induction of plant embryogenic callus and somatic embryos (Islam et al. 2013).Previous studies found that embryogenesis and non-embryogenesis callus can be distinguished by tissue morphology (Maadon et al. 2016), and dynamic changes of cells at different stages of somatic embryogenesis can be observed (Ji et al. 2017).In this study, the observation results of the C. × generalis showed the regeneration mode of C. × generalis was somatic embryogenesis type (Fig. 3).During the process of callus redifferentiation, embryogenic calli have experienced a spherical embryos (Fig. 3a, e), heart-shaped embryos (Fig. 3b, f), torpedo embryos (Fig. 3c, g) and cotyledon embryos (Fig. 3d,  h) successively, with no vascular tissue connection to the parent.In the longitudinal section of the cotyledon embryo at the early stage of differentiation, it is evident that there are stem tip meristems and root meristems at the top and base, respectively.In the early stage of somatic embryo formation, there are apparent two-level embryogenic cell mass structures with prominent bipolar characteristics.
This study found that embryogenic calli developed into spherical embryos, heart-shaped embryos, torpedo embryos, and cotyledon embryos after two weeks of differentiation culture, which were similar to previous studies (Yang and Zhang 2010).In addition, previous studies showed that heart-shaped embryos were typical characteristics of dicotyledonous plant cell embryos (Kurczynska et al. 2007), and heart-shaped embryos were also observed at the monocotyledonous plant, such as lily, so some researchers thought that monocots evolved from extinct dicotyledones (Henslo 1911).The existences of a heart-shaped embryo of C. × generalis provided a strong evidence for this speculation.Moreover, the somatic embryogenesis and embryo maturation of C. × generalis was similar to other research results (Fang et al. 2022;Yang and Zhang 2010;Vahdati et al. 2008) and they had no vascular tissue connection with the maternal body, which was consistent with previous research and observation results (Zou et al. 2019;Quinga et al. 2018).

Somatic embryo proliferation culture
The results of somatic embryo proliferation culture can be seen from Table 3.When the 6-BA concentration was 1 mg L −1 , there is no embryoid proliferation.When the 6-BA concentration was 3-6 mg L −1 , the somatic cell embryos produced secondary embryo proliferation.When the concentrations of 6-BA, NAA and TDZ were 6 mg L −1 , 1 mg L −1 and 0.2 mg L −1 , the somatic embryo proliferation rate was to 46.33%, and the proliferation coefficient could reach 7.83, which was significantly higher than other combinations.
When the concentration of 6-BA was 3 mg L −1 , the proliferation of somatic embryos was white or light green (Fig. 4B, C), and the generation of secondary embryos was 5 d earlier than the concentration of 6-BA at 6 mg L −1 .When the concentration of 6-BA was 6 mg L −1 , the somatic embryos were good, and the color was dark green.In addition, some somatic embryos did not proliferate under low concentrations of cytokinin, and somatic embryo maturation occurred (Fig. 4D), which may be due to the inhibitory effect on cell division at the lower concentrations of cytokinin and affect the regulation of cell cycle, making somatic cell stay at a specific division stage, thus limiting cell proliferation (Soundar Raju et al. 2013).In conclusion, MS + 6 mg L −1 6-BA + 1 mg L −1 NAA + 0.2 mg L −1 TDZ was the best treatment for somatic embryos proliferation.

Somatic embryo germination and seedling formation
It could be seen from Table 4 that the mature embryoid developed into a plant in several culture media.The germination and seedling rates at MS media were 94.67% and 88.00%, respectively, which were significantly higher than that at the B5 and 1/2MS media.And the germination and seedling rates in B5 were only 23.33% and 13.33%, respectively, and the seedling rate was relatively low.Therefore, the medium suitable for embryoid germination and seedling formation was MS without adding plant growth regulator, which was similar to the previous study that it was beneficial to developing SE in a hormone-free medium or low hormone concentration for chestnut (Lu et al. 2017).The reasons may be that high hormone concentrations interfered with the normal development of the apical meristem (Halperin and Wetherll 1964).At the same time, part of the embryoid has a proliferation phenomenon of a secondary embryo at MS media.
As depicted in Fig. 5, it was observed that somatic embryos induced from embryogenic calli could germinate into complete plants at the MS medium (Fig. 5A).When the somatic embryos were inoculated into the MS medium for 15 days, the plants began to root, bud germinating and growing at the same time (Fig. 5B).After 20 days, the plants grew luxuriantly and the leaves were green (Fig. 5C, D), which could induce robust small plants (Fig. 5E).The medium used at this stage did not contain growth regulator, the results  were similar to other plants that have also observed the germination of somatic embryos in MS medium without regulators and the development of complete plants (Ferreira et al. 2022).In addition, somatic embryos with bipolar nature can directly develop into complete plantlets without the rooting stage (Arnold et al. 2002), which can shorten the seedling period and save costs.

Acclimation transplantation and genetic stability
As showed in Fig. 6, it was observed that tube seedlings from embryoid were obtained after strong seedling culture (Fig. 6A), domesticated by opening the bottle at green culture (Fig. 6B), and directly transplanted the domesticated seedlings into a pot with a matrix (peat: organic fertilizer: soil = 1:1:1, v:v:v).After culturing for 7 days, it was observed that C. × generalis seedlings started to grow new green leaves (Fig. 6-C).After 15 days, the seedlings developed into normal healthy small plants and grew vigorously with dark green leaves (Fig. 6-D).The results indicated that the addition of seedling substrate and peat had provided nutrients for its growth and development in the process of domestication and transplanting, and the survive rate was 98%, which was higher than that of the previous research (Mishra et al. 2015).Therefore, the mixed substrate of peat: organic fertilizer: soil = 1:1:1 (v:v:v) was the best cultivation substrate for C. × generalis.
It can be seen from Fig. 7, the leaf of regenerated plants was oval to lanceolate, gradually pointed leaves, dark green, wedge-shaped base, and green stem sheath, which was consistent with the Canna × generalis population, but the explants used in this study were zygotic embryos, so there must be genetic variation.However, this study laid a foundation for the next step of using asexual materials as explants to establish embryoid regeneration system of Canna × generalis.

Conclusions
In this study, the regeneration system of C. × generalis was successfully established through cell embryogenesis by using the immature embryosas materials.It took about 10 weeks to obtain complete small plants.In addition, microscopic and histological analysis showed that typical morphogenesis periods such as globular, heart-shaped, torpedo-shaped, and cotyledon-shaped were experienced successively during the development of somatic embryos and it belonged to the indirect type.The establishment of the regeneration system could not only to promote the genetic improvement and breeding process of C. × generalis by transgenic or genome editing technology, but also to help broaden our understanding the origin and development of somatic embryos and provide reference for somatic embryo culture of other varieties of Canna.

Fig. 1
Fig. 1 Callus induction from immature embryo of C. × generalis.a Explant status after inoculation for 0 days.b White granular embryogenic calli after 15 days inoculation.c White massive non-embryonic callus after 15 days inoculation.d Embryogenic calli and nonembryonic callus after 15 days inoculation.(The red arrow indicates

Fig. 3
Fig. 3 Phenotype (a-d) and microscopic observation (e-h) of four developmental stages of somatic embryogenesis.a, e Globular embryo.b, f Heart-shaped embryo.c, g Fish-shaped embryo.d, h cotyledon embryo

Fig. 4
Fig. 4 Proliferation of somatic embryoids under different culture days.a Proliferation and growth of somatic embryos after inoculation for 0 days.b, c Proliferation and growth of secondary embryos inocu-

Fig. 5 Fig. 6
Fig. 5 Somatic embryo germination and plant regeneration of C. × generalis.a Somatic embryo inoculated on the germination media.b Seedling status after germinating for 15d.c Seedling status

Table 1
*Data in the table represent mean ± standard error, and different letters (a, b,…) in the same column represent statistically significant differences at p < 0.05 (Duncan's test), NS indicate non-significant; * indicate significant at P < 0.05; ** indicate significant at P < 0.01; *** indicate significant at < 0.001

Table 2
Effect of plant growth regulators (PGRs) on the somatic embryo induction from Canna × generalis *Data in the table represent mean ± standard error, and different letters (a, b, )in the same column represent statistically significant differences at p < 0.05 (Duncan's test).NS indicate non-significant; * indicate significant at P < 0.05; ** indicate significant at P < 0.01; *** indicate significant at < 0.001

Table 4
Effects of different medium formulations on seedling culture of Canna × generalis embryoids Data in the table represent mean ± standard error, and different letters (a, b,…) in the same column represent statistically significant differences at p < 0.05 (Duncan's test) *