Recovery, histological observations and genetic integrity in coconut (Cocos nucifera L.) embryogenic calli cryopreserved using encapsulation-dehydration procedure

This article is published under the Creative Commons CC-BY-ND License (http://creativecommons.org/licenses/by-nd/4.0/). This license permits use, distribution and reproduction, commercial and non-commercial, provided that the original work is properly cited and is not changed in anyway. Abstract: Encapsulation-dehydration protocol was evaluated for the long-term conservation of coconut embryogenic calli (EC). Encapsulated EC were pretreated in a sucrose medium (either 0.5 M or 0.75 M) for diff erent durations (24, 48 and 72 h) followed by desiccation in silica gel (8, 12, 16 and 20 h) prior to storage in liquid nitrogen. Survival and regrowth of EC following desiccation alone or desiccation and freezing were recorded after four and eight months, respectively. Histological studies and scanning electron microscopy (SEM) were carried out to evaluate the structural changes of cryopreserved samples. Recovered samples from cryopreserved and non-cryopreserved EC were tested for genetic fi delity at 11 simple sequence repeat (SSR) marker loci. Highest survival (45 %) and recovery (25 %) rates were achieved by pre-treating EC on 0.75 M sucrose for three days followed by dehydration for 20 hours, prior to liquid nitrogen immersion. Ultra-structural studies showed no aberrations or damages in the exterior regions of cryopreserved ECs. Massive damages at the interior regions after cryopreservation suggested that osmotic and cryo injuries occur in the internal regions of EC. DNA banding patterns at SSR marker loci showed no evidence of somaclonal variations in tested material confi rming that encapsulation-dehydration have not caused any genetic variation in EC.


INTRODUCTION
Conservation of coconut germplasm is important to facilitate future breeding programmes. Conventional storage of coconut as seeds is impossible because of the large size of the seed, recalcitrant nature and lack of a dormancy period (Engelmann, 1997). Currently, coconut germplasm is conserved in fi eld genebanks, which are frequently threatened by adverse weather conditions and pest and disease attacks (Nguyen et al., 2015). Moreover, fi eld gene banks require large areas for the maintenance of plantations and the process is highly labour intensive, thus very costly. Hence, development of in vitro culture techniques for conservation and exchange of coconut germplasm is highly desired.
Cryopreservation, which is a method of conserving plant tissues at an ultra-low temperature, usually in liquid nitrogen (LN; is an important technique for the long-term storage of plant germplasm in many problematic species such as non-orthodox seed species, vegetatively propagated plants and rare, endangered plant species. At this temperature, metabolic processes and growth activity of cells completely cease and thus plant materials remain stable for an unlimited period with minimal space and maintenance (Engelmann, 1997).
The early attempts of in vitro conservation of coconut germplasm focused on ex-situ conservation of zygotic embryos and pollen (Assy-Bah & Engelmann, 1992a;Frison et al., 1993) as the source of explant with subsequent eff orts using plumular tissues (Malaurie et al., 2006). Immature embryos have been cryopreserved June 2020 Journal of the National Science Foundation of Sri Lanka 48 (2) by pre-growth desiccation (Assy-Bah & Engelmann, 1992b), resulting in survival rates of 10 -43 %. It is reported that mature embryos can be cryopreserved by vitrifi cation (Sajini et al., 2011;Cueto et al., 2013), rapid dehydration (N'Nan et al., 2012), pre-growth desiccation (Assy-Bah & Engelmann, 1992a), and droplet vitrifi cation (Cueto et al., 2013) and a maximum of 40 % success rate (recovery) for the modifi ed desiccation protocol has been reported (Sisunandar et al., 2010). However, coconut is an open pollinated heterozygous plant, thus genetic characters of the zygotic tissues may vary from the mother palm. Conserving somatic embryos or EC will enable the conservation of superior palms with true-to-type genetic makeup. Furthermore, cell and callus line maintenance, continuous supply of standard stock cultures for physiological and biochemical experiments are considered as benefi cial outcomes of cryopreserved in-vitro cultures (Poobathy et al., 2013).
Cryopreservation of EC has experimented in several other perennial commercial crops such as rubber (Zhou et al., 2012), cocoa (Fang et al., 2004) and oil palm (Suranthran et al., 2012). Encapsulation-dehydration procedure, which is developed based on the technology for the production of artifi cial seeds can be successfully applied to calli because of the protected covering formed around the delicate calli tissue providing the protection against dehydration and freezing damage. This technique has been successfully applied in several species (Feng et al., 2013;Barraco et al., 2014) and reported for coconut plumule cryopreservation with 60 % survival and 20 % shoot development (reviewed in Welewanni et al., 2017).
No studies have been reported for coconut cryopreservation using EC, which may be indeed important for conservation of coconut germplasm. The objective of this study is to determine the amenability of coconut embryogenic calli (derived from unfertilised ovary explants) to cryopreservation technique, assessment of regenerated plants for their genetic fi delity and ultrastructural changes leading to the identifi cation of an optimum cryopreservation protocol for coconut EC by encapsulation dehydration method.

Plant material
Unfertilised ovaries were excised from female fl owers of 4-stage (which is the infl orescence that will open in four months considering the most recently opened infl orescences as 0 stage; Perera et al., 2007) infl orescences from mature palms of the improved coconut hybrid CRIC 65.

Initiation and multiplication of embryogenic callus through ovary culture
Ovary derived calli were raised using the method described by Bandupriya et al. (2017) with the modifi cation that, instead of CRI 72 medium, modifi ed Eeuwens Y 3 medium (Eeuwens, 1976) was used. Welldeveloped embryogenic calli (Figure 1a) from the third or fourth cycle of sub-culturing were used for cryopreservation studies.

Cryopreservation by encapsulation-dehydration
Dissected embryogenic calli parts (2-3 mm; Figure 1b) were pre-cultured on the same modifi ed Eeuwens Y 3 medium for 3 d to screen for microbial contamination. Encapsulation was carried out by suspending the EC parts in Eeuwens Y 3 medium (devoid of CaCl 2 ) containing 3 % (w/v) Na-alginate (2 % viscosity, SIGMA) and 4 % sucrose. This was followed by dropping each EC using a sterile pipette into 0.1 M CaCl 2 .2H 2 O solution prepared in Eeuwens Y 3 medium supplemented with 4 % sucrose and allowed to polymerise for 45 min at room temperature to form beads (4 -5 mm in diameter, Figure 1c). Osmoprotection was carried out by pretreating the EC containing beads in liquid Y 3 medium containing either 0.5 M or 0.75 M sucrose for diff erent durations (24, 48 and 72 h). The beads were then subjected to dehydration for 8, 12, 16, or 20 h inside glass jars by placing 20 pretreated beads on top of 40.0 g of dried silica gel separated by a fi lter paper. Upon dehydration (Figure 1d), half of the beads were cultured into the standard recovery medium and the other half of the beads of each treatment was transferred to 2.0 mL sterilised cryo-tubes and directly plunged into LN at least for 2 h. Thawing was performed by immersing the cryotubes in a water bath at 40 °C for 3 min. The water content of the beads (on fresh weight basis) was determined by drying them in an oven at 80 °C until a constant weight was obtained.

Plant regeneration
Each alginate bead was cultured for regeneration in a small glass vial containing 15 mL somatic embryo induction medium [modifi ed Eeuwens Y 3 medium containing 95.0 μM 2,4-D and 2.7 gL -1 Glutamine (3G)] for 4 wk. Plant regeneration protocol described by Bandupriya et al. (2017) was followed for the regeneration of plantlets.
Journal of the National Science Foundation of Sri Lanka 48 (2) June 2020

Assessment of survival and regrowth
The survival was assessed after 4 months of in vitro culture as the percentage of calli manifesting new tissue growth as indicated by any sign of growth such as swelling, development of tissue and/or callus formation. The recovery of calli (indicated by the ability of calli to produce somatic embryos and/or grow into shoots or plantlets following desiccation alone or desiccation and freezing) was assessed at least after 8 months in culture.

Statistical analysis
Treatments were arranged in a completely randomised design (CRD) including 10 experimental units per treatment with two replicates. Data analysis was conducted using STATA 13.1. Summary of the results was presented as mean, standard deviation (SD), median and interquartile range (IQR) for continuous variables as appropriate, whereas counts are presented as percentages for categorical variables. As normality assumptions were violated, non-parametric statistical tests were used in the analysis in order to eliminate potential bias (Vickers, 2005) and median was used as the measure of central tendency. To compare more than two categories of a continuous variable, Wilcoxon rank sum test was used.
Results are displayed as proportional odds ratios for a one unit increment; whereas one unit is the increment of categorised survival or recovery percentiles by one level to other. Signifi cance level of the treatment eff ects was indicated at the 95 % level of probability (p ≤ 0.05).

Histological analysis
Cryopreserved and non-cryopreserved EC, treated with 0.5 M and 0.75 M sucrose for 24, 48 and 72 h and dehydrated for 20 h, were subjected to histology studies. EC without any treatment were used as the control. Samples were fixed in FAA solution (50 % ethanol: 10 % formaldehyde: glacial acetic acid = 18:1:1, v/v/v) for 72 h. Dehydration was carried out through a graded alcohol series (50 -100 % ethanol) and clearing was done using pure xylene. Calli were embedded in paraffi n wax and blocks were prepared followed by obtaining 4.0 µm thick sections using a microtome (Sakura, Japan) with steel blades. The sections were double stained with periodic acid Schiff 's reagent (PAS) and proteinspecifi c naphthol blue black (NBB) (Fisher, 1968). The slides were observed under microscope and images were obtained by a camera-fi tted light microscope (ZEISS, USA).

Scanning electron microscopy
Samples were selected from cryopreserved and noncryopreserved EC, pretreated with 0.75 M sucrose for 72 h followed by 20 h dehydration, and osmoprotected with 0.75 M sucrose for 72 h without physical dehydration by silica gel. EC without any treatment were selected as the control with 3 biological replicates from each treatment. Fresh samples were obtained from the recovery medium, the alginate covering was removed, and fi xed on doublesided strips of tape that were affi xed onto the sample stub. Samples selected were neither treated nor coated prior to the viewing and the samples were gold coated using a sputter coater (Quorum, SC7620). The samples were viewed in their native state, using scanning electron microscope (ZEISS, Evo LSIS) at saturated humidity at 4 °C at an extra high tension (EHT) value of 10 kV.

SSR marker analysis
Somatic embryos and shoots recovered after cryopreservation together with non-cryopreserved and counterpart controls were used to assess the genetic integrity. DNA was extracted using DNeasy ® Plant Mini Kit (QIAGEN ® ) following the manufacturer's instructions. SSR primers (Table 1) were selected according to Rivera et al. (1999). PCR reaction mixtures were prepared as described by Bandupriya et al. (2017). Amplifi cations were performed in thermal cycler (BIO-RAD; MyCyclerTM) programmed at 94 °C for 5 min for initial denaturation, followed by 94 °C for 30 s, annealing temperature depending on the primer used (Table 1) for 30 s and 72 °C for 1 min for 35 cycles followed by a fi nal step of extension at 72 °C for 5 min. PCR samples were electrophoresed on 6 % (w/v) polyacrylamide gel and bands were visualised by silver staining. The genetic polymorphisms of the samples were scored and compared with the control samples and the donor palm.
June 2020 Journal of the National Science Foundation of Sri Lanka 48(2)

RESULTS AND DISCUSSION
Eff ect of sucrose pretreatment and dehydration on the water content of encapsulated EC Sucrose concentration showed a signifi cant eff ect on the water content of beads (fresh weight basis) after dehydration (p = 0.0001) recording a higher water content in 0.5 M (38-42 %) than in 0.75 M (27-30 %) sucrose ( Figure 2). A higher sucrose concentration induced greater osmotic dehydration and the values are comparable with the water content recorded for encapsulated coconut plumules after sucrose pretreatment (N'Nan et al., 2008). However, the eff ect of duration of sucrose pretreatment (24, 48 or 72 h) on the water content of beads after dehydration was not signifi cant (p = 0.7123). Generally, a 20 -25% water content of beads is optimal for successful cryopreservation (Bian et al., 2002). The optimum water content reduction up to 28 % at 0.75 M sucrose level indicated the need for further reduction of water content to achieve higher recovery rates in further studies. Although water content is an important factor in cryopreservation, it cannot be dissociated with many other parameters such as the material used, the cryoprotectant, and its concentration, pretreatment duration and the method of dehydration, thus it is not possible to detect a specifi c water content related to the desiccation damage (N'Nan et al.   The main cause of injury during cryopreservation is intracellular ice crystal formation. Thus, survival can be increased by reducing the intracellular water content of the cells. Pre-culturing in high sucrose concentrations has been found to cause many cellular alterations, which are favourable for the survival during cryostorage. In most cases, encapsulation-dehydration technique uses sucrose which is a non-toxic substance as a cryoprotectant. Sugars or most importantly sucrose decrease the amount of freezable water content in cells due to its osmotic eff ect and accumulation of soluble sugars, thus enhances the freezing tolerance (Suzuki et al., 2006;Feng et al., 2013). Increment of total soluble sugar and accumulation of sugar inside cells could help maintain plasma membrane integrity by substituting for water on the membrane surface, thus stabilising proteins under dry and freezing conditions (Crowe et al., 1988). Furthermore, total soluble protein levels in buds increase when the explants are pretreated with high sucrose concentrations. It is considered to be one of the earliest physiological responses of osmotically stressed cells, which may be related to the increased freezing tolerance (Suzuki et al., 2006). The use of high sucrose concentrations facilitate the penetration of a suffi cient quantity of sucrose into the cells, and accumulation inside the cells as starch.
These accumulated starches could be the remedy for maintaining water output; without producing harmful eff ects and making the tissue to withstand further desiccation and freezing (González-Arnao et al., 2003).

Eff ects of sucrose pretreatment on EC survival and recovery
Sucrose concentration was the overall signifi cant factor (p = 0.0245) contributing to the survival rate while the duration in sucrose pretreatment was not signifi cant (p = 0.2173). The ordinal logistic regression revealed a higher survival rate in 0.75 M than in 0.5 M sucrose concentration. Similarly, higher survival rates were observed with increased duration of sucrose pretreatment (24 h to 72 h) for both tested sucrose concentrations ( Figure 3 a and b). The eff ect of sucrose concentration was not signifi cant for recovery rate (p = 0.2594), while the eff ect of sucrose pretreatment duration was signifi cant (p = 0.0359) recording higher recovery rates when the duration is increased (Figure 3 c and d). 2007) through encapsulation-dehydration. Interestingly, date palm (Phoenix dactylifera) EC could not tolerate high sucrose concentrations due to osmotic shock and optimum sucrose concentrations reported for successful pretreatment are ranged between 0.1 -0.5 M (Subaih et al., 2007).

Eff ect of dehydration duration on EC survival and recovery
Although diff erences in survival rate did not signifi cantly diff er (p = 0.1308), dehydration duration signifi cantly aff ected the recovery rate (p = 0.0219) after cryopreservation. Ordinal logistic regression for categorised data revealed higher recovery rates when dehydration durations are elevated (Figure 4 a and   Survival rates observed in the treatments after dehydration and freezing were very low compared to the untreated controls. Some of the cryopreserved and non-cryopreserved EC turned brown and some remained white without any growth, even after two months of culturing in the recovery media. These EC were not dead or vitrifi ed, but turned into rubber like form. Some of the cryopreserved and non-cryopreserved (but dehydrated) calli, which were sub-cultured into a new callus induction medium were able to produce new EC and somatic embryos (Figure 1 e, f and g). Shoots developed from noncryopreserved (but dehydrated) treatments (Figure 1h) showed no diff erence in morphology compared to the untreated controls ( Figure 1 i and j). Overall somatic embryo and shoot development after dehydration and freezing was very slow. Similar results have been reported in plantlets derived from cryopreserved coconut June 2020 Journal of the National Science Foundation of Sri Lanka 48(2) embryos and plumules (N'Nan et al., 2008;Sajini et al., 2011). N'Nan et al. (2008) explained that 20 % recovery is a commendable achievement in species like coconut which are recalcitrant. The highest recovery (25 %) recorded in this study for the best treatment combination is encouraging for further research to develop a reliable cryopreservation technique from coconut EC.
Histological structure of EC after osmoprotection, dehydration, LN application and comparison with untreated In the present study, cells in the untreated coconut EC ( Figure 5 a and b) showed similar characteristics to normal meristematic cells as per the description of Perera et al. (2007) for coconut and other species such as date palm (Bagniol et al., 1992) and yam (Barraco et al., 2014). The calli showed cellular heterogeneity upon dehydration. However, the intensity of cellular changes varied among samples. Cells were plasmolysed with deformed and retracted nuclei in majority of the samples and the nucleoli were no longer visible. Damaged plasma membrane and cell walls were indicated by the leakage of intracellular soluble proteins in certain cells (Figure 5e). Highly plasmolysed cells were characterised by condensed and poorly stained cytoplasm and slightly visible nuclei. Ration of the area of the nucleus and nucleocytoplasmic decreased signifi cantly (Figure 5f). These results were compatible with the results observed in the research by Bagniol et al., (1992) and Barraco et al., (2014). In cryopreserved samples, the plasma membranes of numerous cells were broken and leakage of intracellular soluble proteins was observed. The nuclei of cells were pyknotic without visible nucleolus ( Figure 5 f and g). Most of the treatments showed similar results as described previously, except cryopreserved EC which underwent osmoprotection for 3 days in 0.75 M sucrose concentration and dehydration for 20 h. These samples showed visible nuclei in numerous cells and slightly visible nucleoli in some cells even after cryopreservation (Figure 5h). Non-cryopreserved EC after sucrose pretreatement and dehydration showed accumulation of starch ( Figure 5d) as a result of the uptake of sucrose and its partial metabolism to form starch (glucose storage) (Bachiri et al., 2000) indicating the accumulation of sucrose in large quantities in cells subjected to encapsulation/dehydration. According to N'Nan et al. (2008) cryopreserved plumular cells after 8 h dehydration displayed a retracted cytoplasm, nucleus and the presence of starch grains. Furthermore, increased levels of starch have been recorded when the cells had undergone 16 h dehydration. Barraco et al. (2014) explained osmoprotection using encapsulation-dehydration method and recorded dramatic changes in cellular structure, such as plasmolysis of cells in whole tissue. Shrinkage and intense staining of nuclei can cause pyknotic nuclei and disappearance of nucleoli. This happens due to chromatin contraction and increased protein concentration, respectively which is a reaction to tolerate stress conditions (Barraco et al., 2014). Nevertheless, this is advantageous due to growth arrest of tissues leading to redirecting resources to growth after cryopreservation. However, it may be related to the duration of recovery in standard medium (Barraco et al., 2014).
In contrast, dehydration also causes severe damage to cells especially in outer cell layers (Bagniol et al., 1992;Barraco et al., 2014) and dehydration reported to be the major cause of structural damage rather than freezing during cryopreservation. Considering the results of histological studies, massive structural damages were observed at the interior regions after encapsulation and dehydration of the coconut EC. This might be due to low penetration of the osmoticum. These results were compatible with the observations made on meristems of raspberry (Wang et al., 2005) and shoot tips of yam (Dioscorea alata; Barraco et al., 2014). Barraco et al. (2014) explained the recovery of yam cells into their original morphology in one week after cryopreservation. Therefore, it is necessary to make histological observations regularly after cryopreservation in order to get a better idea on the degree of recovery of cells. However, in this study, cells of EC showed severe cryo injuries during the procedure preventing any growth or callus formation in most of the samples even after six weeks.

SEM observations of cryopreserved and noncryopreserved EC
SEM studies indicated the lack of dehydrating and freezing injuries at the exterior regions of the EC (Figure 6 b and d). Cryopreserved and non-cryopreserved EC displayed apparent epidermal layers without any signifi cant damages ( Figure 6b) and these EC were analogous in morphology to the regular untreated EC (Figure 6 a and c). However, there were shrunken cells after sucrose pretreatment (Figure 6e). Similar observations have been reported in scanning electron microscopy studies of the control and cryopreserved protocorm-like bodies of Dendrobium sonia-8, which is a hybrid orchid (Poobathy et al., 2013). The possible reason may be the thorough protection of outermost cell layers through osmoprotection compared to the inner cell layers.
Assessment of genetic stability in regenerants from cryopreserved and non-cryopreserved EC Eleven SSR markers used for the evaluation of genetic stability of cryopreserved shoots and somatic embryos were all informative and generated amplicons. Each tested SSR primer pair produced clear reproducible bands ranging in size from 100 -250 bp (Figure 7). All the samples scored identical alleles with their respective mother palm and the untreated controls at all SSR loci. The SSR loci CNZ10, CNZ29 and CAC65 scored heterozygous alleles while homozygous alleles were scored at all the remaining loci.
Cryopreservation imposes a series of stresses on plant material causing modifi cations or alterations in regenerated tissues resulting in possible genetic alterations and changes in allele frequencies (Harding, 2004). Majority of the studies showed genetic stability of cryopreserved material while some showed genetic variations. Studies carried out on Dendranthema grandifl ora Tzvelev (Martin & Gonzalez-Benito, 2005) and Rabdosia rubescens (Ai et al., 2012) showed genetic variation after cryopreservation, necessitating the attention for testing the genetic integrity in cryopreserved material.
True-to-type conformity of tissue cultured coconut plants derived from unfertilised ovaries was tested in a previous study with respect to the mother palm and genetic stability at SSR marker loci was confi rmed . Same SSR marker loci were evaluated for the genetic uniformity of the cryopreserved and non-cryopreserved EC and compared with the untreated controls and the mother palm. SSR analysis revealed genetic stability of coconut EC and showed no polymorphisms between the cryopreserved and noncryopreserved samples. The resulting bands were similar to with their untreated controls and respective mother palm. However, epigenetic changes due to genome methylation or transposable elements and mitotic June 2020 Journal of the National Science Foundation of Sri Lanka 48 (2) changes like aneuploidy could also occur during somatic embryogenesis and they cannot be detected using SSR markers.

CONCLUSIONS
Commendable survival (45 %) and recovery (25 %) rates were obtained in the treatment combination of 0.75 M sucrose pre-treatment for 72 hours followed by 20 hours dehydration prior to LN treatment. Water content on fresh weight basis ranged between 28 -30 % and attempts to reduce this further up to about 20 % may have a positive eff ect on recovery. Ultrastructural studies revealed massive structural damages at the interior regions after encapsulation dehydration of the coconut EC. Observations revealed that majority of the meristematic cells were injured either during the freezing/ thawing step or dehydration step. However, the treatment combination, 0.75 M sucrose pre-treatment for 72 hours followed by 20 hours dehydration and LN application showed better cell structure with less damage.
According to the amplifi cation patterns, SSR analysis did not reveal any polymorphisms between cryopreserved samples and untreated controls and with their respective mother palms, suggesting the genetic stability of coconut EC cryopreserved by encapsulation-dehydration method up to plant regeneration stage. It will generate more information when the same analysis is repeated in the fi eld planting the plants.
Journal of the National Science Foundation of Sri Lanka 48(2) June 2020