Somaclonal variation of pitaya (Hylocereus undatus) in vitro plants andidentification by molecular markers

Abstract:【Objective】Pitaya production has become a new, special and excellent agricultural project.Usually, propagation of pitaya is conducted by using cuttings from field plants. However, multiplicationrates are low and it is difficult to obtain enough true-to-type plants for both plantation and research use.Tissue culture is the main means of rapid propagation of plants, and also a good method to obtain soma⁃clonal variation. Currently, the shoots derived from a clone were used as the explants to study the rapidpropagation in vitro. After rapid propagation of several cycles, the plants growth and morphological traitswere different among some regenerated plants. To better apply the rapid propagation system and make fur⁃ther use of the variants, it is necessary to evaluate the genetic fidelity of the in vitro plants derived fromthe current rapid propagation system as well as to reveal the genetic relationship among the variants.【Methods】Pitaya shoots from single seed germination as explants were multiplied on Murashige andSkoog (MS) medium with 0.1 mmol·L^-1 naphthalene acetic acid (NAA) and 2 mmol·L^-1 6-benzyladenine(6-BA), successively in vitro shoots were subcultured for four cycles. Sixty-nine plantlets subcultured for1 and 4 cycles as well as their stock plant which primarily derived from a seed were used to analyze themorphological variations as well as propagation coefficient, and the genetic fidelity was identified byISSR, SRAP and IRAP markers. NTSYS 2.01 software was further used to illustrate the variation plants.【Results】The propagation coefficient of the rapid micropropagation system was 14.9. The 4 derived shootsregenerated from a seedling characterized in 4 arris. Six arris types, i.e. 3-, 4-, 5-, 6-, 7- and 8-arris, ofcluster buds were investigated in the regenerated plants successively subcultured for 4 cycles, amongwhich 5-arris buds were the most common, accounting for about 50.54% of the total, followed by the 4-ar⁃ris buds which accounted for 27.2% of the total, and the 8-arris buds was the least, only about 1.0%. Withthe increase of arris number, the shoot diameter became larger, especially for 6-, 7-, and 8-arris shoot,and the opposite tend was observed in shoot length. No difference in DNA markers obtained from 11 pairsof SRAP primers was investigated between the 3 and 4 arris buds, however an aberrant band was detectedin RNA level by primer me3em3. A total of 405 bands were scored from the 70 samples by 24 ISSR prim⁃ers, 11 pairs of SRAP primers and 17 IRAP primers, among which 28 were polymorphic, which includedpresence and absence of DNA bands in comparison with the stock plant. No polymorphic band was ob⁃served from the three plants which micropropagated in the first cycle, however 16 plants were detected asthe variants among the 66 plants which subcultured 4 cycles, accounting for 24.2% of the total. Among thevariants, 6 plants and 5 polymorphic loci were detected by the ISSR primers 811, 825 and 856; 9 plantsand 14 polymorphic loci were detected by two pair of SRAP primers (me9em2 and me3em3); 12 plantsand 9 polymorphie loci were detected by the IRAP primers 15f, 17r and 4r. The genetic similarity coeffi⁃cient among the 16 mutants and their stock plant varied from 0.77 to 0.97 based on statistical analysis ofgenetic diversity using NTSYSpc 2.10e software. Using unweighted pair group mathematics average (UPG⁃MA) analysis, 16 mutants might be divided into 4 groups taking the similarity coefficient 0.90 as thethreshold. Group I included the stock plant and 12 mutants, the genetic similarity coefficient in the rangeof 0.91-0.97; mutants 7 and 10 were grouped into group II; mutants 41 and 33 were singly clustered intogroup III and IV, respectively. Mutant 33 showed the least genetic relatedness with the stock plant. Afterfurther micropropagation for 3 cycles, mutant 33 appeared no growth in height and no multiplication, andthe regeneration plants were detected as the GA3-deficient dwarf mutant. The normal plant and 3 GA3-mutants could be distinguished by a pairs of SRAP primers me9em2.【Conclusion】The propagation coeffi⁃cient of the current micropropagation system in vitro was 14.9. However, the propagation coefficient,growth potential and genetic stability of somaclones demonstrated a decrease trend with increase of sub⁃culture cycles. Morphologically, the change of the shoot arris numbers, shoot length and diameter were al⁃so investigated herein. Further, the somaclonal variation rate after successively subculture for 4 cycles ac⁃counted for 24.2% of the total as revealed by molecular markers, and the variations mainly showed onpresent or absent amplified bands. The 16 mutants were divided into 4 groups, 12 of which and the stockplants were closely related, and the genetic similarity coefficient ranged from 0.91 to 0.97. Mutant 33 wasa GA3-deficient dwarf mutant. The difference of 3- and 4-arris shoots might be ascribed to the differen⁃tial expression of genes. The newly developed IRAP marker was more sensitive than ISSR and SRAPmarkers to detect the DNA mutation in different loci of the genome, therefore was an effective means forsomaclonal variation detection in pitaya. The obtained results facilitated the rapid propagation, germplasmconservation and creation, as well as mutant detection in this fruit.