CALLUS INDUCTION AND PLANT REGENERATION FROM IMMATURE EMBRYOS CULTURE OF TROPICAL MAIZE

The development of protocols to overcome the current limitations of callus induction and in vitro regeneration of highly recalcitrant tropical maize is crucial for plant genetic transformation. The ability of embryogenic callus (EC) formation of 46 tropical maize hybrids and 14 inbred lines was tested using N6 medium with the following changes: medium A (N6 + 15 μM dicamba + 25 mM L-proline + 88 mM AgNO3), medium B (N6 + 30 μM dicamba + 25 mM L-proline + 88 mM AgNO3), medium C (N6+ 30 μM dicamba + 6 mM L-proline), and medium D (N6+ 30 μM dicamba + 25 mM L-proline). Compact (Type I) and friable (Type II) callus were induced in the four media. Twenty genotypes produced callus in all media, and four genotypes (CO32, AG8012, CMS477BC4F2, and CMS-HGZ10) produced the highest number of callus (114, 134, 131, and 126 calli, respectively). All immature embryos of ten genotypes produced callus in at least one medium, while eight genotypes were highly recalcitrant and they did not produce any callus. The frequencies of EC ranged from 0% to 38%, and the highest rate of EC was observed on medium B (0.40) with a total of 865 calli, and the lowest induction rate was obtained with medium D (0.29) with 555 calli (P= 0.05). From the seventy-two EC of 26 genotypes transferred to Murashige & Skoog regeneration medium, twenty-four (66.7%) differentiated into green plants which produced seeds in R0 and R1 generations, and twelve (33.3%) developed into albino plants. The results demonstrated that the problem of the recalcitrant genotypes can be, at least partially, overcome by using immature embryos as explants together with tissue culture media formulations.

Maize is the second largest crop in Brazil surpassed only by soybean.The crop accounts for about 30% of the Brazilian agricultural area, with production estimates of 80 million tons of grains per year in two growing seasons (Oliveira Neto, 2018).
However, nutrient-poor and acidic soils, insect pests, and diseases represent a serious threat to the Brazilian agriculture, which impose the adoption of new approaches to face this challenge.
The genetic transformation is a reliable strategy for introducing foreign genes into plant chromosomes in vitro (Quispe-Huamanquispe et al., 2017).However, maize plants are difficult to genetically engineer, mainly due their inherent recalcitrance toward the in vitro culture conditions (Altpeter et al., 2016).The identification of maize genotypes with high ability of callus production and high green-plant regeneration ability is a necessary initial step for the subsequent genetic transformation by any of the currently available methods, namely particle bombardment, protoplast transformation, Agrobacterium mediatedinfection and in planta transformation (Buggs et al., 2017;Yadava et al., 2016).
The majority of successful protocols for plant transformation using callus induction and plant regeneration in maize have been developed for non-commercial inbred lines with low agronomic performance and adapted to temperate climates (Lu et al., 1982).Thus, crosses through conventional plant breeding methods are required to transfer the foreign genes to elite lines.The use of highly regenerative tropical genotypes selected from highvalue commercial varieties can reduce the costly and time-consuming breeding programs to transfer the genes from one variety to another.
Many authors started to evaluate the potential of callus formation and plant regeneration of tropical-adapted maize thorough indirect somatic embryogenesis using immature embryos (Bohorova et al., 1995;Carvalho et al., 1997;Petrillo et al., 2008).In tropical maize, the ability of a given genotype to induce Type I (compact) or Type II (friable) callus and regenerate plants are genetically determined with frequencies ranging from 4% to 32.5% (Al-Abed et al., 2006).However, Akoyi et al. (2013) reported that embryogenic and non embryogenic callus induction was independent of the genotype when young immature embryos were cultured in medium supplemented with Dicamba.In addition, results of other studies have also shown that callogenesis and somatic embryogenesis in maize are also determined by biochemical parameters of the medium, such as the carbon and nitrogen sources, the type and concentration of growth regulators (auxins, gibberellins, cytokinins, abscisic acid and ethylene), L-proline, silver nitrate, and even the orientation of the embryos in the culture medium (Azad et al., 2015;Lu et al., 1982;Muli et al., 2017).Among the growth regulators, the auxin-like 2,4-dichlorophenoxyacetic (2,4-D) was considered the most efficient for embryogenic callus induction and plant regeneration in cereal crops.This is probably because it is metabolized more slowly compared to other auxins, and its role in activating the transcription of the somatic embryogenesis receptor kinase gene responsible for somatic embryogenesis in many plant species (Ombori et al., 2008).Differently, some authors reported that immature embryos showed better callus response and produced more somatic embryos in medium supplemented with the auxinlike dicamba herbicide than in the medium containing 2,4-D (Furini & Jewell, 1994;Rakshit et al., 2010).
In the present work were evaluated the effects of four culture medium formulations on callus induction and plant regeneration ability of sixty maize genotypes adapted to the Brazilian tropical climate.

Plant materials
In this study, a total number of 60 tropicaladapted maize genotypes were evaluated (Table 1).
Forty-four sib-pollinated hybrids of tropical maize from seeds companies and assigned into three groups according to cycle length: 12 early-maturity, 22 intermediate and 10 late-maturity hybrids; eleven inbred lines selected for high levels of gamma-zein accumulation in the endosperm, and five inbred lines from the Embrapa Maize and Sorghum breeding programs.

Explant preparation and culture conditions
Sixteen to twenty days after pollination, ears were harvested and surface-sterilized by immersion in Table 1 -List of maize genotypes and their genetic origins.
The average percentage of callus formation was scored for all genotypes and each medium tested.
Embryogenic calli were maintained by subculturing at every 21 days on the same callus initiation medium.
Both embryogenic calluses, Type I (hard, compact, nodulated, and yellowish tissue) and Type II (soft, friable, and white-to-brown) were used for plant regeneration.
After four weeks of incubation, the plantlets were transferred to the rooting medium, which consisted of the regeneration medium supplemented with 1.0 mg/L a-naphthalene acetic acid (NAA) and 8g/L agar.After rooting, the plantlets were transferred to plastic pots containing soil/vermiculite (2:1) mixture and kept in the greenhouse under high humidity level.

Statistical analysis
Statistical analysis to determine frequencies (%) of callus induction, somatic embryogenesis, and plant regeneration was performed using SAS (Statistical Analysis System) version 9.2 (2008).
Tukey's HSD (Honestly Significant Difference) test at 95% confidence interval was used to determine significant differences among the parameters under study.
Table 2 -Media composition for callus induction from maize immature embryos.

Results and Discussion
Callus formation directly from the scutellum of maize immature embryos was observed by approximately forty days after culture initiation (Figure 1A).Callus production was observed on the four media tested (Table 2), and was highly dependent on the genotype (Figure 1B, C and D).The capacity of maize to produce regenerable callus from immature embryos is determined by quantitative trait loci (QTL) with epistatic interactions, and additive genes with positive heterotic effects (Tomes & Smith, 1985).Two types of embryogenic calli (type I and type II) were reported in maize tissue culture by Tomes and Smith (1985).In the present study, the number of regenerated plants from both types of embryogenic callus was very similar, thus they were considered together.The total percentage of callus formed ranged from 29% on media D to 40% on medium B (Table 3).The type I callus was compact, nodular, and white to cream colored, with slow growth and proliferating as a mixture of complex tissues showing somatic embryos.Callus Type II was light yellow, highly friable, soft and characterized by fast growth.
The type II callus has a high embryogenic activity rate, which permits its cultivation for a long period (over a year) as reported by Tomes and Smith (1985).According to Al-Abed et al. ( 2006) and Ombori et al. (2008), this characteristic of callus type II was due to the existence of meristematic cells in the scutellum of maize embryos.
The higher average percentage (40%) of callus formation for all genotypes tested was observed in the medium B (Table 3).The medium D, which was similar to medium A but lacking AgNO 3 , was the less efficient for callus induction (29%).This data stresses the favorable effects of silver cation (Ag + ) on callus induction and plant regeneration potential in tropical genotypes of maize (Bohorova et al., 1995;Rakshit et al., 2010).However, the cellular role of Ag + on cultured tissues remains unknown.Beyer (1976) speculated that silver ions may act as inhibitors of the ethylene action, which in turn inhibits auxin transport in plants.
In the present study, the influence of the auxin-like dicamba on callus induction could be observed comparing the data of Medium A (15 µM dicamba) and Medium B (30 µM dicamba) (Table 3).The average of embryos producing callus in the Medium B and Medium A were about 40% and 37%, respectively.Although the difference between these two treatments was not statistically significant, calli formed on medium B were more friable and embryogenic than those of Medium A. This result may be attributed to differences in the concentration of dicamba in the media.In fact, dicamba has been postulated as the best regulator for callus initiation and maintenance in tropical maize (Bohorova et al., 1995).Contrary, Anami et al. (2010) found that the 2,4-dichlorophenoxyacetic acid (2,4-D) concentration at the callus initiation stage from immature embryos is determinant for shoots regeneration in tropical maize.
In relation to the addition of the amino acid L-proline in the medium, studies reported that the presence of this amino acid in the culture promoted embryogenesis and plant regeneration in somatic tissues of maize (Armstrong & Green, 1985) and other species (Ozawa & Komamine, 1989).In our study, however, we could not observe any improvement on callus formation due the addition of L-proline to the media.In fact, different concentrations of L-proline in medium C (6 mM) and medium D (25 mM) showed similar results with averages of 32% and 29% of callus formation, respectively.Among genotypes with high content of gamma-zein, only three (CMS477, CMSHGZ08 and CMSHGZ18) showed embryogenic type II callus formation.However, none of these three genotypes regenerate plants.Among all 60 genotypes evaluated, only 26 formed embryogenic calli.
In our study, dark embryos without any callus were discarded after 45 days of cultivation and subculturing at three week intervals.Seventy-two embryogenic calli with somatic embryos transferred to regeneration medium turned green and developed vigorous shoots.The commercial hybrids C909 and Dina 887 also produced a good root system after 30 days on regeneration medium (Figure 1E and 1F), and some somatic embryos did not grow further to form plantlets.As previously postulated by Ikeda-Iwai et al. (2003) and Ombori et al. (2008), the down-regulation of genes responsible for plant regeneration from immature embryos-derived callus could explain the differences in tissue culture responses among genotypes.This may also explain the absence of plant differentiation of some somatic calli in our study.In addition, shoot formation from embryos-derived callus was significantly higher in hybrids C909 and Dina 887 compared to those formed by the other genotypes.This result showed that plant regeneration was genotype-dependent, as previously demonstrated in many cereal species (Anami et al., 2010;Bohorova et al., 1995;Ombori et al., 2008;Tomes & Smith, 1985).
In vitro regenerated plants (R0) with welldeveloped roots were transferred to pots with soil and vermiculite mixture for further growth in the greenhouse (Figure 1G and Table 4).Seeds of R0 plants grown to maturity in the green house and R1 seeds generated new plants (Figure 1H).
Some genotypes generated only plants with small shoots, and they were not able to develop roots, even after 30 days on rooting medium (1.0 mg L -1 NAA).
At the end of the process, nineteen plants were transferred to greenhouse and all of them produced seeds normally when transferred to green house.The efficiency of the process was of 26.39% (19 plants from 72 embryoids).

Conclusions
The present results indicated that somatic embryogenesis of recalcitrant maize genotypes and plant regeneration can be, at least partially, overcome by using immature embryos as explants and culture media formulation.
In this study, we have implemented and optimized a reliable protocol for callus induction and plant regeneration through somatic embryogenesis of immature maize embryos and culture media formulation.The efficiency of the process was of 26.39%.
70% ethanol, followed by 2.5% sodium hypochlorite for 40 min and rinsed three times with sterile deionized water.Immature embryos 1.0-1.5 mm long were excised from the kernels under aseptic conditions with a scalpel and spatula.Twelve to fourteen embryos per plate were placed, flat surface down, with the scutellum side up on four distinct callus initiation media, and the cultures were incubated at 27 o C, in the dark for two to three weeks.

Figure 1 .
Figure 1.Somatic embryogenesis and plant regeneration from immature embryos of tropical maize hybrid C909.A and B callus initiation on medium B. Embryos explants showing callus initiation after 14 days in culture (A).Embryogenic callus (Friable calli Type II) at the globular stage (B).Differentiation of heart-shaped somatic embryos in N6 medium supplemented with 6% sucrose and 1mg L -1 NAA (C and D).Shoot development from the embryogenic callus on MS medium without growth regulators (E).Root development in regenerated plantlets (F).Embryogenic R0 plants acclimatization in the greenhouse 4 months after culture initiation (G), and mature R1 plants (H).

Table 1 -
List of maize genotypes and their genetic origins.

Table 2 -
Media composition for callus induction from maize immature embryos.

Table 3 -
Average percentage (%) of callus formation from maize immature embryos in four different culture media.

Table 4 -
Morphogenic responses of maize callus in the regeneration medium.

Table 4 -
Morphogenic responses of maize callus on the regeneration medium.