Analysis of SSR information in transcriptome and development of molecular markers in Torreya grandis

Abstract：【Objective】The genus Torreya is an important member of Taxaceae with only 9 species distributed in China, the United States of American, Japan and Korea. T. grandis Fort. ex Lindl. has been famousof its nut seeds with high oil for economic usage and as food for thousands years in China. Breeding of T.grandis is currently limited on selection of the wild resources derived from spontaneous mutation and occational seedling. The genetic characteristics of the wild resources are great concern for the efficiency of im⁃provement of T. grandis. To get insight of hereditary information for commercial utilization and cultivar discrimination of genetic resource of this species, microsatellite loci were exploited based on compared tran⁃scriptomic profiles de novo assembled for RNA-seq data of T. grandis.【Methods】Seeds of T. grandis‘Xifei’and T. grandis‘Dielsii’were sampled from a orchard (Fuyang, Zhejiang province, China). Based onthe biochemistry analysis of fatty acids biosynthesis, seeds at initiation stage (449 days after pollination,DAP), fastigium stage (469 DAP) and steady stage (512 DAP) were separately mixed as sample pools forRNA extraction; followed by comparative transcriptomic analysis through RNA-Seq. De novo transcrip⁃tome assembly was achieved using Trinity. Microsatellite mining was performed using the software MIcro⁃SAtellite (MISA). Primers (20 base pairs) were designed using Primer 3 were used. 49 pairs of primerswere randomly selected and were used for morphological amplification with product sizes ranging from 100to 300 bp. Ten individuals of each species, including T. nucifera, T. fargesii, T. jackii, T. grandis and T.ESTgrandisvar. jiulongshanensis, were selected for primary assess on the microsatellite polymorphism. Genom⁃ic DNA was extracted from young leaves by the cetyltrimethylammonium bromide (CTAB) modified meth⁃od.【Results】All high quality reads were assembled into 142 213 unigene. A total of 5 458 SSRs contain⁃ing inserts with various microsatellite motifs (di- to penta nucleotides) were identified from 22 976 unigenes (> 1 kb). The distribution density was 105.4 SSRs per Mb. Mono nucleotide was the most abundanttype of repeat motif with a frequency of 64.95%, followed by the tri-nucleotide (18.56%), di (14.77%), tet⁃ra (0.97%), hexa (0.42%) and pentanucleotides (0.33%). The AT/AT was the most frequent repeat motif(3.85%), followed by TA/TA (3.57%) and AG/CT (2.29%)；Among the tri-nucleotide repeats, the dominantrepeat motif in EST-SSRs was GAA/TTC (2.36%)，followed by CTT/AAG (1.32%) and AAT/ATT (1.26%).The frequency of tetra-and hexanucleotide were much lower than that of di- and tri-nucleotide. Numbersof each repetitive unit ranged from 5 to 21. Units repeated less than or equal to 10 times and greater than10 times were predominantly at mono nucleotide type with 1 990 (36.46%) and 1 555 (28.49%), and followed by 5 (12.90%), 6 (11.95%), 7 (5.42%) and 8 (2.53%) times among di- to hexanucleotide type. Repeat number was mainly 6, 7 and 8 in Di nucleotide unit as well as 5, 6 and 7 in tri-nucleotide unit. A totalof 4 633 prime pairs were designed, among them 49 primer pairs were selected randomly to test in Torreya.From these, 16 polymorphic SSRs were characterized. The number of alleles of each locus ranged from 2 to6, which was averagely 3 per locus. The average value of polymorphism information content (PIC), observed heterozygosity (Ho) and expected heterozygosity (He) were 0.464 4, 0.283 7 and 0.500 6 respective⁃ly.【Conclusion】The transcriptome sequencing data of Torreya can provide abundant SSR locus, which canbe used into develop SSR primer rapidly. The primer developed can provide marker selection for Torreyagenetic diversity, the construction of fingerprint pattern of variety and molecular mark assisted breeding.