True-to-type micropropagated plants of para rubber (Hevea brasiliensis Müll. Arg.) via somatic embryogenesis

Plant micropropagation via somatic embryogenesis is a powerful technique for rapid mass propagation, especially in para rubber (Hevea brasiliensis Müll. Arg.). However, somaclonal variations are the major limitation of this process. To date, DNA fingerprinting, i.e., RAPD (Randomly Amplified Polymorphic DNA), Star Codon Targeted (SCoT), and SSRs (Simple Sequence Repeats), is one of the most successful technologies to detect the genetic fidelity in the somatic embryos. The aim of present study was to induce somatic embryos from inner integument explants of para rubber cv. ‘RRIM 600’ at different developmental stages and subsequent acclimatization and transplantation (under greenhouse and field conditions) of the propagated seedlings. The genetic stability of the plants derived from somatic embryos was also analysed in comparison to the mother plant using RAPD, SCoT and SSRs markers. Somatic embryos derived from inner integuments of 5-week-old immature seeds after pollination were more efficient than older and younger seeds. In addition, para rubber mother plants cv. ‘RRIM600’ and plants derived from somatic embryogenesis demonstrated the same pattern of DNA fragments, as confirmed by three PCR-based techniques, RAPD, SCoT and SSRs, whereas these in the pattern were different from ‘RRIT 226’, ‘PB 235’, ‘PB 251’, ‘PB 255’ and ‘BMP 24’. Interestingly, T2 plant was found to possess somaclonal variations when compared with mother plant. Based on the results, we confirm that the plants derived from somatic embryogenesis of para rubber cv. ‘RRIM 600’ were true-to-type to that of ‘RRIM 600’ master stock.

report, global natural rubber consumption in the year 2019 was 12.119 megatons and it has been expected to increase at a rate of 16.5 megaton per year (IRSG, 2020). Maximum consumption of natural rubber has been reported in the tires of automobiles, which utilizes ~70% of the total production, and its demand has been predicted to increase by 30 megatons per year by the year 2030 (Cornish, 2017). The para rubber planting area is rapidly expanding into the Montane Mainland Southeast Asia including Grate Mekong Subregion (China, Vietnam, Thailand, Laos, Myanmar and Cambodia) (Golbon et al., 2018;Yang et al., 2019) and Peninsular Malaysia (Hazir et al., 2020).
Para rubber cv. 'RRIM600' is a good candidate for plantation in Thailand and China with adaptation to drought, chilling and frost injuries and moderate yield (Priyadarshan et al., 2005;Gonçalves et al., 2011;Pethin et al., 2015;Yang et al., 2019). Elite clones of para rubber with high yield of latex, rapid growth, longlifetime tapping (20-30 years) and disease-resistant traits being developed through breeding programs (Supriya et al., 2019). Rootstock grafting with elite scion genotype is a common protocol to generate master stock of para rubber for small holder farmers in both Thailand (>85% of the total area; Chantuma et al., 2011) and Malaysia (>93% of the total area; Hazir et al., 2020), which are the major plantation areas. Recently, the limitation of cross-talking (grafting efficiency) between rootstock and scion has been found to significantly affect the latex yield and growth rate of scion (Gonçalves and Martins, 2002;Cardinal et al., 2007;Yao et al., 2017) and to induce DNA methylation (epigenetic changes) in heterografted plants (Uthup et al., 2018). Selfrooted para rubber is an alternative way for elite clonal propagation (75% success rate from shoots to mature rooted plants) at the commercial scale to produce large number of healthy plants by cuttings (Masson and Monteuuis, 2016). Microcutting  and nodal segment micropropagation  via in vitro culture are the other alternatives. However, the vegetatively propagated plants lack taproot system, which makes it susceptible to wind disasters (Masson et al., 2013;Masson and Monteuuis, 2017). Somatic embryogenesis may overcome this problem, as it produces large number of plantlets containing tap root system, similar to the seedlings produced by seed germination (Zhao et al., 2015;Rahman et al., 2017;Wang et al., 2017). Recently, somatic embryos have been successfully developed from inner integuments of immature seeds and uninucleate anthers of male flower Lardet et al., 2008;Kouassi et al., 2018;Tisarum et al., 2018). However, success of somatic embryogenesis in para rubber depends on genotype, and there are risks associated with in vitro acclimatization, somaclonal variation and high cost of production (Masson and Monteuuis, 2016;Tisarum et al., 2018).
True to type somatic embryos of para rubber, without somaclonal variation, are required for large scale production of elite clones . Based on this requirement, genetic and epigenetic changes have been studied in para rubber using molecular markers [including randomly amplified polymorphic DNA (RAPD) and simple sequence repeats (SSRs)] and DNA methylation [including methylation-sensitive amplification polymorphism (MSAP)] (Karumamkandathil et al., 2015;Wang et al., 2017). RAPD and SSRs are a simple, rapid, reproducible, and costly techniques to identify the genetic similarity in both intergeneric and intrageneric diversity (Thakur et al., 2016;Rohela et al., 2019). SCoT (start codon targeted marker) is a novel technique to confirm homogeneity of micropropagated plantlets derived from organogenesis or embryogenesis. In addition, two or three molecular markers, confirming genetic stability, have been well established in many plant species (Bhattacharyya et al., 2017a,b;Sharma et al., 2019). The aim of present study was to induce somatic embryos from inner integuments of para rubber cv. 'RRIM 600' at different developmental stages and subsequent acclimatization and transplantation (under greenhouse and field conditions) of the propagated seedlings. The genetic stability of the plants derived from somatic embryos was also analysed in comparison to the mother plant using RAPD, SCoT and SSRs markers.

Materials and Methods
In vitro culture and acclimatization of plantlets Immature seeds of Hevea brasiliensis cv. 'RRIM' 600 in different ages, 4, 5 and 6 weeks were collected after pollination. These were dipped in 95% ethanol and burned for surface sterilization in aseptic clean bench. Thereafter, inner integuments of each seed were dissected as initial explants and inoculated on MH-IN medium (Carron et al., 1989) under dark conditions for 25 d. Number of callus formed in MH-IN medium was recorded, and callus induction percentage in each treatment was calculated. Callus derived from inner integuments were proliferated on MH-EXP medium for 25 d (Carron et al., 1989) and then embryogenic callus were induced using MH-EXP for 25 d. Number of embryogenic callus in each treatment were counted.
Somatic embryos derived from embryogenic callus inoculating on MH-DEN for 25 d (Carron et al., 1989) were induced. Embryos and elongated embryos from single somatic embryos were induced by MH-MAT and MH-GER medium for 25 d (Carron et al., 1989), respectively. Para rubber plantlets derived from somatic embryos were transferred to test tube containing MH-PL medium with activated charcoal (1.5 g L -1 ) and kept for 25 d. The whole process of in vitro culture system of para rubber is demonstrated in Figure S1. In vitro acclimatization ( Figure S2) and ex vitro adaptation ( Figure S3) were practiced by following the method given by Tisarum et al. (2018) as along with 1.5 year-field trial ( Figure S3).

DNA extraction
Leaf samples of three master stocks (mother plants) of para rubber cv. 'RRIM 600', nine in vitro acclimatized plantlets ( Figure 1A), ten plants derived from somatic embryogenesis transplanted in the field for 1.5 year ( Figure 1B) and an individual master stock of other cultivars, i.e., 'RRIT 226', 'RRIT 251', 'PB 235', 'PB 255' and 'BPM 24' were collected (Table 1). DNA was extracted using genomic DNA Mini Kit (Plant) (GP100, Geneaid™, Geneaid Biotech Ltd., South Korea) according to company protocol. In brief, 50 mg of leaf tissues were ground using liquid nitrogen, transferred to 1.5 mL plastic tube, mixed with 400 mL GPX1 buffer containing 5 mL RNase A (mixed by vortex) and incubated at 60 °C for 15 min (invert the tube 5 min interval). Thereafter, 100 mL GP2 buffer was added, mixed by vortex and then incubated on ice for 3 min. Solution was filtered through filter column, centrifuged at 12,000×g for 2 min. The filter column was discarded, and supernatant was collected and transferred to a new 1.5 mL plastic tube to which 1.5 volume GP3 buffer was added, and immediately mixed with vortex for 5s and filtered through GD column. The extracted solution was centrifuged at 12,000×g for 2 min, and added with 400 mL W1 buffer, then again centrifuged at 12,000×g for 30s, and added with 400 mL absolute ethanol (wash buffer), and again centrifuged at 12,000×g for 30s. After discarding the flow-through, the GD column was placed in the 1.5 mL collection tube and centrifuged at 12,000×g for 3 min to dry the column. Dry GD column was moved to a new 1.5 mL plastic tube, 150 mL pre-heated elution buffer (10 mM Tris-HCl, pH 8.5 at 60 °C) to the centre of the column matrix, incubated at room temperature for 5 min, centrifuged at 12,000×g for 1 min, and then the purified DNA was collected and incubated at -20 °C for further assay.

Experimental design and statistical analysis
The experiment was arranged as completely randomized design (CRD) with 4 biological replications. Data were subjected to Analysis of variance (ANOVA) and mean values obtained from three treatments were compared using Tukey's HSD at p ≤ 0.01 using SPSS software (version 11.5, SPSS for Window  ).
In our study, 20 SSR markers were used for genetic diversity assay (Table S6). Within the plantlets derived from somatic embryogenesis, low genetic diversity was noticed (<4%) except in T2, which was different from RRIM 600 original (by >20%). Based on UPGMA cluster analysis, genetic stability in plantlets derived from somatic embryogenesis was evaluated in the same group with 'RRIM 600' original, #1, #2 and #3 (Cluster
In para rubber, two RAPD makers, OPAA-07 and OPJ-19 were used to detect genetic viability of plant derived from zygotic polyembryony (Karumamkandathil et al., 2015). In African hyacinth (Ledebouria revoluta), 17 RAPD markers were used to confirm the genetic stability in plants derived from somatic embryogenesis (Haque and Ghosh, 2016). Ten RAPD markers were used to validate the genetic stability in plants derived from indirect regeneration of Rauwolfia tetraphylla L. (Rohela et al., 2019). Six selected lines of Echinacea purpurea derived from somatic embryogenesis were evaluated for their genetic stability using 10 RAPD markers (Lema-Rumińska et al., 2019). In contrast, about 20-30% genetic variation using only one OPD-20 RAPD primer was detected in rough lemon (Citrus jambhiri Lush.) plantlets derived from somatic embryogenesis (Savita et al., 2015). In Withania coagulans, 12 RAPD markers were used to confirm genetic stability in shoot regenerated plants derived from indirect organogenesis of leaf explants (Rathore et al., 2016). Genetic homogeneity in micropropagated plants of Pitosporumerio carpum Royle was detected by 10 RAPD markers and the plantlets were clustered in the same group as that of donor mother plant (97% similarity) (Thakur et al., 2016). Micropropagation via direct and indirect organogenesis in Rumex nepalensis has shown a less degree of variation within the clones when analyzed using 7 RAPD makers compared to 12 SCoT makers (Bhattacharyya et al., 2017a). Moreover, RAPD marker is suggested to be suitable for identifying the diversity in genetic resources of para rubber (Venkatachalam et al., 2004;Nakkanong et al., 2008;Liyanage et al., 2014). This is the first report to apply SCoT markers for assessing genetic fidelity in para rubber micropropagation. In a previous study, high degree of genetic variation in plantlets of Rumex nepalensis, micropropagated via direct and indirect shoot organogenesis was detected by 12 SCoT primers, whereas it was undetected using 7 RAPD markers (Bhattacharyya et al., 2017a). Genetic stability in 9 randomly propagated plants of Bauhinia racemosa using 18 SCoT makers was reported as true-to-type (Sharma et al., 2019).
Similarly, micropropagated plants free of somaclonal variations (100% similarity with mother plant), derived via indirect organogenesis were confirmed using SCoT markers in Helicteres isora (Muthukumar et al., 2016), Chrysanthemum morifolium and C. coccineum (Nasri et al., 2018) and Saccharum officinarum (Sathish et al., 2018). Similarity value ranged from 0.89-1.00 in 9 individual plants derived from nodal segment micropropagation of Pittosporum eriocarpum tested using 10 SCoT primers (Thakur et al., 2016). Moreover, genetic fidelity of plantlets derived from indirect organogenesis of Rauwolfia tetraphylla was confirmed using 10 SCoT primers as true-to-type without any polymorphism (Rohela et al., 2019). Genetic similarity between somatic embryos and mother plants of Coffea arabica detected using 12 SCoT makers was found to be 0.987 (Bychappa et al., 2019). In medicinal orchid (Ansellia africana), 16 reproducible SCoT primers were applied to identify the genetic stability in first, second, and third generation of subculture which ranged from 0.95 to 1.00 as per Jaccard's similarity matrix and were presented in the same cluster as that of mother plants as per UPGA cluster analysis (Bhattacharyya et al., 2017b). Genetic stability of plantlets of Morus alba cv. 'Chinese', derived from indirect organogenesis was confirmed to be similar to the mother plant using 10 SCoT primers (Rohela et al., 2020). In Nardostachys jatamansi, genetic variations in micropropagated plants was found to be dependent on the pathway of genesis of new plantlets as detected by 10 SCoT primers, with direct shoot organogenesis showing a variation of 4.38% from that of mother plant and indirect shoot organogenesis (showing a variation of 7.14% from that of mother plant) (Bose et al., 2016). Moreover, genetic stability of shoots proliferated from nodal segments of Alhagi maurorum was validated using 13 SCoT primers (Agarwal et al., 2015).
In para rubber clone CATAS7-33-97, variation of regenerated plants (2.61% of mother plant) derived from secondary embryogenesis using 12 EST-SSR markers was observed . Somaclonal variation in callus culture of para rubber was detected by hmct1 and hmc5 SSR primers, whereas these ignored the variations in somatic embryos from somatic embryogenesis (Srichuay and Te-chato, 2015). In addition, in vitro plantlets derived from nodal culture of para rubber were confirmed as true-to-type without somaclonal variation by 3 SSR primers . In guava, genetic homogeneity in plants derived from somatic embryogenesis was validated by 6 SSR markers (Rai et al., 2012). Absence of somaclonal variations in the sesame plantlets, derived via indirect organogenesis was validated using 10 SSR markers (Anandan et al., 2018). In contrast, somaclonal variations in somatic embryogenesis-regenerated Spanish grapevine (Vitis vinifera) cv. 'Torrontés' were detected in a mutant allele (231) instead of the normal allele (237) at the locus VVMD5 by SSR microsatellite analysis (Prado et al., 2010). Moreover, somaclonal variations in olive plants derived from somatic embryogenesis was slightly higher in T1 lines (youngest lines; identified by lowest similarity coefficients) when compared with T2 and T4 based on SSR makers (Bradï et al., 2019).

Conclusions
In the present study, somatic embryogenesis was successfully carried out from the inner integuments of 5-week-old immature para rubber fruits. The plantlets were subsequently acclimatized under in-vitro, greenhouse and field conditions. Genetic fidelity was confirmed using RAPD, SCoT and SSRs markers. It was concluded that the somaclonal variation is present only in one plant (T2) as detected by RAPD, SCoT and SSR markers (PCR-based rapid detection). Based on this investigation, a thorough research on the genetic variations in T2 plants compared to wild type 'RRIM 600' is suggested.

Authors' Contributions
Conceptualization, supervision, validation, visualization, writing -original draft, writing -review and editing: SC and WP; Methodology, data curation and analysis, RT, JM and TS.
All authors read and approved the final manuscript.