Inheritance of seed dormancy in Tibetan semi-wild wheat accession Q1028

J Appl Genet 46(2), 2005, pp. 133-138 Inheritance of seed dormancy in Tibetan semi-wild wheat accession Q1028 Xiu-Jin Lan, Yu-Ming Wei, Deng-Cai Liu, Ze-Hong Yan, You-Liang Zheng Triticeae Research Institute, Sichuan Agricultural University, Dujiangyan City, Sichuan, P.R. China Abstract. Tibetan semi-wild wheat (Triticum aestivum ssp. tibetanum Shao) is one of the Chinese endemic hexaploid wheat genetic resources, distributed only in the Qinghai-Xizang Plateau of China. It has special characters, such as a hulled glume and spike disarticulation. However, seed dormancy, another important character for wheat resistance to pre-harvest sprouting, was rarely reported. Seed dormancy of more than 10 Tibetan semi-wild wheat accessions was evaluated, and their germinations were 0% or near 0% with both treatments of threshed seeds and intact spikes at hard dough stage. Tibetan semi-wild wheat accession Q1028 was investigated for its seed dormant characters by testing the seed germination percentages of intact spikes, seeds with bract powder, normal seeds, seeds with pierced coat, and sectioned embryos. It was observed that embryo dormancy of Q1028 accounted for its seed dormancy. Using threshed seeds and intact spikes at hard dough stage, the inheritance of seed dormancy was carried out using the F 1,F 2,F 3 and F 2 BC 1 populations of the cross between Q1028 and a wheat line 88 1643, susceptible to preharvest sprouting. The germinations of seeds and intact spikes in F 1 were 1.0% and 0.9%, respectively. It indicated that seed dormancy of Q1028 was inherited as a dominant trait. From the genetic analysis of the F 2,F 3 and F 2 BC 1 populations it was found that the strong seed dormancy of Q1028 was controlled by two dominant genes. Key words: inheritance, preharvest sprouting tolerance, seed dormancy, Triticum aestivum ssp. tibetianum Shao. Introduction Tibetan semi-wild wheat (T. aestivum ssp. tibetanum Shao, AABBDD) is one of the Chinese endemic hexaploid wheat resources, found only in the Qinghai-Tibet Plateau of China (Huang et at. 1987). It is mainly distributed in Longzihe, Lantsang and Chayuhe valleys of the Brahmaputra, and grows in primitive wheat fields as weeds (Shao et al. 1980). The typical characters of Tibetan semi-wild wheat are spontaneous spike disarticulation and hulled character, thus indicating that it is closer to the wild wheat than other representatives of T. aestivum (Shao et al. 1984). Tibetan semi-wild wheat has a primitive genomic structure (Yen et al. 1988). Due to its primitive characters, it has an important value in the phylogenetic evolution study of hexaploid wheat. More attention has been paid so far to its origin and evolution, as well as relationship with common wheat. The genetic variation among Tibetan semi-wild wheat has been investigated using C-banding (Chen et al. 1988; Chen et al. 1996), N-banding (Lu and Zhang 1983), esterase isozyme (Cui and Ma 1990; Lan et al. 2001), SDS-PAGE and APAGE (Wei et al. 2001, 2002), and molecular markers (Ni et al. 1997; Huang et al. 2002; Wei et al. 2003). The two typical primitive characters, spontaneous spike disarticulation and the hulled character of Tibetan semi-wild wheat, were investigated by various methods. The hulled character was controlled by a dominant gene on 2DS, which is different from the Tg gene of T. spelta (Chen et al. 1999). Cao et al. (1997) thought that spike disarticulation of T. aestivum ssp. tibetanum can Received: November 29, 2004. Accepted:April 4, 2005. Correspondence: YL Zheng, Triticeae Research Institute, Sichuan Agricultural University, Dujiangyan City 611830, Sichuan, P.R. China; e-mail: ylzheng@sicau.edu.cn

134 X.J. Lan et al. be divided into two types, namely that of wedge modifying genes (Wm) and barrel modifying genes (Bm). They were codominant and independent of each other. The former was controlled by more than two complementary genes, and the latter was related to 1 3 dominant independent genes. Chen et al. (1998) found that the one of the spike disarticulation genes was controlled by one dominant gene on 3DS, which is different from the spike disarticulation gene of T. spelta on 5AL. However, Huang et al. (2002) suggested that spike disarticulation was probably controlled by genes on 5S, 5L, 6L, 7S, or their polygenic accumulation. Seed dormancy was evaluated in more than ten Tibetan semi-wild wheat accessions. It was found in this study that all the Tibetan semi-wild wheat accessions had strong seed dormancy with the germination of 0%. Strong seed dormancy of Tibetan semi-wild wheat should play an important role in both phylogenetic evolution and breeding improved preharvest sprouting tolerance in wheat. However, the inheritance of this character was rarely reported. The objective of this study is to investigate the inheritance of strong seed dormancy of Tibetan semi-wild wheat accession Q1028. Material and method Material Tibetan semi-wild type wheat Q1028, synthetic wheat line RSP and 8 wheat cultivars were used in this study. RSP, a synthetic wheat line (2n = 42; AABBDD) between Chinese tetraploid landrace Ailanmai (Triticum turgidum L., 2n = 28, AABB) and Aegilops tauschii (DD, 2n = 14), had high tolerance to preharvest sprouting, which derived from Ae. tauschii (Lan et al. 1992; 1997; 2002), used as tolerant controls. The 8 wheat cultivars included 4 red kernel lines (i.e. Mianyang11, Chuanyu12, Liaochuen10 and Chuanmai42) and 4 white kernel lines (i.e. Shanmai897, Zhengmai9023, Mianyang26 and 88 1643). Method In 2002, the F 1 and F 2 generations produced by crossing between Q1028 (female parent) and 88 1643 (male parent) with susceptivity to preharvest sprouting were examined for germinations of threshed seeds and intact spikes. To decide whether the germination of each plant was similar to that of Q1028, the t-test was used in the F 2 population. The chi-square test was carried out to decide befitting gene numbers according to the segregation ratio of the F 2 population. In 2003, the F 3 and F 2 BC 1 (88 1643 as the recurrent parent) populations from dormant F 2 were further tested for the threshed seed germinations. In 2004, seed germinations of Q1028 and RSP were carried out using intact spikes, normal seeds, seeds with destroyed glumes, seeds with pierced coats around embryos, and embryos sectioned from seeds. In addition, 100 spikes were harvested after maturity from each genotype. Germination tests of intact spikes and seeds were carried out at 0, 1, 2, 3 and 4 weeks after harvest, until more than 50% seeds germinated. The germinations were investigated after seven days. When the embryo was just revealed from the seed coat, germination was recorded. Results Seed dormancy length of Q1028 Wheat varieties with various preharvest sprouting tolerances were used as controls. The germinations of intact spikes and threshed seeds were Table 1. The germinations of intact spikes and seeds in the tests of various post-maturation stages Intact spike germination Seed germination 0(w) 1(w) 2(w) 3(w) 4(w) 0(w) 1(w) 2(w) 3(w) 4(w) Q1028 0 0.4 0 0 0 0 1 2 RSP 2.2 5.2 9.2 15.1 19 21 32 43 MY11 83 11.6 17.4 26.8 42.7 21 29 43 56 CY12 14.2 28.9 33.7 42.1 51.2 36 52 LC10 26.0 32.7 40.8 51.4 37 57 99-1572 37.6 47.1 62.3 45. 61 SM897 6.3 12.4 25.2 38.1 51.0 24 43 58 ZM9023 9.7 22.7 38.6 52.3 23 48 65 MY26 40.4 51.6 38 56 88-1643 53.3 64

Inheritance of seed dormancy in Tibetan semi-wild wheat accession Q1028 135 tested in Q1028. The results showed that Q1028 had the lowest germination percentages among all the genotypes in the same treatment. Except Q1028, the other genotypes germinated in the test beginning from the day of harvest (0 week of after-maturing), and the germination percentages increased gradually following the increment of post-maturation days. However, the germinations of intact spikes and threshed seeds of Q1028 were 0% or near 0% at 0 to 4 weeks of post-maturation. Compared with RSP, Q1028 has stronger seed dormancy (Table 1). Seed dormancy peculiarity of Q1028 Among all the treatments for RSP, germination percentages decreased gradually according to the order of the embryo, seed with a pierced coat, normal seed, seed with the destroyed glume and intact spike. It seemed that seed germination was affected by the pericarp, the contents of seed and glumes (Table 2). On the third day, germinations of normal seeds and seeds with destroyed glumes were 0% and 1%, respectively, whereas those of seeds with pierced coats and sectioned embryos were 15% and 23% (for embryos of the susceptible to preharvest sprouting line 88 1643 it was 57% after one day and 94% after three days), respectively. After 7 days, seeds with pierced coats had higher germination percentages than normal seeds. The pericarp of RSP accounted for its preharvest sprouting tolerance. However, the germinations of intact spikes, seeds with destroyed glumes and normal seeds of Q1028 were only 0%. Thus, Q1028 had stronger seed dormancy than RSP. After 7 days, the germinations of seeds with pierced coats of Q1028 and RSP were 21% and 72%, respectively, and the germinations of sectioned embryos of Q1028 and RSP were 19% and 82%, respectively. It indicated that the embryo dormancy of Q1028 accounted for its stronger seed dormancy. Inheritance of seed dormancy of Q1028 The female parent Q1028 with strong dormancy had 0% germination, and the susceptible male parent 88 1643 had 64.4% germinations. The germinations of seeds and intact spikes in F 1 were 1.0% and 0.9%, respectively. It indicated that seed dormancy of Q1028 was inherited as a dominant trait (Table 3). Among 180 F 2 single, the number of with 0% germination was 54, and Table 2. Comparison of germination percentages between Q1028 and RSP Embryo Seed with pierced Days coat Seed Seed with glume Intact spike Q1028 RSP Q1028 RSP Q1028 RSP Q1028 RSP Q1028 RSP 1d 0 4 0 3 0 0 0 0 2d 0 12 0 9 0 0 0 0 3d 0 23 0 15 0 1 0 0 4d 8 46 9 28 0 3 0 2 5d 12 66 14 41 0 6 0 2 6d 15 77 17 58 0 14 0 3 7d 19 82 21 72 0 19 0 6 0 2.2 the number of with over 10% germination was 33. With the t-test, germinations of 109 were not significantly different from that of Q1028, and they were treated as tolerant. Germinations of 71 were significantly different from that of Q1028, and they were treated as susceptive. The chi-square test value of separate proportion of 109:71, according with the ratio of 9:7 controlled by two dominant genes, was 2 = 1.187 < 2 0. 05; 1 = 3.84 (P > 0.25). Table 3. Germination percentages of seeds and intact spikes of the parents and F 1 from Q1028/88-1643 Threshed seed Intact spike Germinated Q1028 88-1643 F 1 Tested Germination (%) Germinated Tested Germination (%) Germinated Tested Germination (%) 0 250 0.0 161 250 64.4 1 100 1.0 0 317 0.0 152 283 53.7 2 217 0.9

136 X.J. Lan et al. Table 4. Separation of seed germination percentages of F 2 from Q1028/88-1643 No. of Total of tested 0% 0.1~ 2.0% 2.1~ 5.0% 5.1~ 10.0% >10.0 % Tolerant Separation regions of germination percentage Susceptible 2 (9:7) Threshing seed 180 54 33 22 38 33 109 71 1.187* Intact spike 180 62 47 29 28 14 126 54 2 0. 05; 1 = 3.84, P 0.25 = 1.32 Table 5. Germination percentages of seeds in F 3 and F 2 BC 1 populations from 10 F 2 with 0% germination Population 1 2 3 4 5 Mean tion 1 2 3 4 5 Mean F 3 Popula- F 2 BC 1 F 3-1 2 2 6 36 12 11.6 B-1 18 2 30 0 16 13.2 F 3-2 10 32 32 34 22 26 B-2 16 18 36 64 38 34.4 F 3-3 14 12 10 0 2 7.6 B-3 22 20 28 28 30 25.6 F 3-4 2 0 18 6 0 5.2 B-4 8 24 44 12 10 19.6 F 3-5 4 10 12 6 0 6.4 B-5 12 0 6 4 22 8.8 F 3-6 2 4 2 4 2 2.8 B-6 2 4 0 14 8 5.6 F 3-7 8 2 2 0 4 3.2 B-7 2 8 0 12 2 4.8 F 3-8 0 2 2 0 0 0.8 B-8 2 0 2 4 0 1.6 F 3-9 2 0 0 2 0 0.8 B-9 0 6 2 4 4 3.2 F 3-10 2 2 16 32 10 12.4 B-10 36 2 6 10 18 14.4 Mean 7.68 Mean 13.12 60 50 No. of plant 40 30 20 10 0 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 Germination percentage [%] Figure 1. Plant distribution of seed germination percentages in F 2 The result indicated that seed dormancy of Q1028 was controlled by two dominant genes (Table 4). Two peak values could be found from the distribution figure of the 180 F 2, at 6% and 16%. They seemed to be the areas of A_bb and aab_ genotypes respectively, and the in the 0 4% ranges could be considered as the A_B_ genotype. It was consistent with the above statistical conclusion by Chi 2. Furthermore, the two dominant genes with controlling seed dormancy differed in strength from each other. If susceptible were partitioned into 18% or 20%, both plant numbers (aabb) answered for 1/16 in approximation. The F 3 and F 2 BC 1 (88-1643 as the recurrent parent) populations derived from ten F 2 with 0% germination. The average germinations of the F 3 and F 2 BC 1 populations were 7.68% and 12.92%, respectively (Table 5). Among the 10 F 3 populations, F 3-1, F 3-2, F 3-3, F 3-4, F 3-5 and F 3-10 had higher sprouting variance, and germinations of some were found significantly different from that of Q1028 by the t-test. It indicated that they were heterogeneous. There were fewer variations in the germinations of single of each line from F 3-6, F 3-7, F 3-8 and F 3-9. Among the corresponding backcross lines F 2 BC 1-6, F 2 BC 1-7, F 2 BC 1-8 and F 2 BC 1-9, only of

Inheritance of seed dormancy in Tibetan semi-wild wheat accession Q1028 137 both F 3 and F 2 BC 1 from No. 8 and No. 9 F 2 were not significantly different from Q1028 by the t-test. These results indicated that only two were homogeneous, and 8 were heterogeneous among the ten F 2. According to two dominant genes ratio of 9:7,there should be one homogeneous plant in nine tolerant. The chi-square test of tolerance/susceptibility of 2/8 according with 1:8was 2 = 0.005 < 2 0. 05; 1 = 3.84 (P > 0.9). It gave further support to the assumption that the dormancy of Q1028 was controlled by two dominant genes. Discussion In this study, Tibetan semi-wild wheat accession Q1028 had a very strong seed dormance ability with seed germination of 0%. Seed dormancy of Q1028 was different from the preharvest sprouting tolerance of RSP. The pericarp accounted for the preharvest sprouting tolerance of RSP, whereas embryo dormancy accounted for the dormancy of Q1028, the strength of which is rarely found among cultivars of wheat. It has been reported that preharvest sprouting tolerance in Ae. tauschii could be expressed in artificial synthetic wheat (Lan et al. 1992; 1997; Gatford et al. 2002b). Seed dormancy in Ae. tauschii was affected by the caryopsis structure, embryo dormancy and inhibitor within the glume (Gatford et al. 2002a). The preharvest sprouting tolerance of RSP is controlled by a single recessive gene derived from the 2D chromosome of Ae. tauschii (Lan et al. 1997, 2002). However, the dormancy of Q1028 was controlled by two dominant genes. This indicated that seed dormancy of Tibetan semi-wild wheat was genetically different from that of RSP. It was indicated that the genetic basis of seed dormancy or preharvest sprouting tolerance in common wheat is quite complicated. There are different tolerance bases among various wheat cultivars with different tolerance genes (Xiao et al. 2002). It was found that genes controlling seed dormancy or sprouting tolerance of common wheat were associated to the following chromosomes, i.e. 6B, 7D (Roy et al. 1999), 5AL, 6AL, 3B, 7B (Zanetti et al. 2000); 2AL, 2DL, 4AL (Mares and Mrva 2001); 2A, 3A, 4A, 6A, 3B, 2D, 4D, 7D (Miura et al. 2002); 2AL, 4AL, 2DL, 3D (Mares et al. 2002); 3AL, 3BL, 3DL (linked to seed color genes) and 5AS (Groos et al. 2002). Tibetan semi-wild wheat has the same chromosome constitution as the other common wheat (T. aestivum L., AABBDD), and has been treated as a subspecies of T. aestivum L. Due to its annual and brittle traits, its mature spikelets break off naturally on the fields. The seeds have strong dormancy in favor of maintaining its reproduction. Genes controlling the brittle rachis and hulled character are located on D chromosomes, and it is presumed that they were derived from Ae. tauschii, the donor species of the D genome (Chen et al. 1998, 1999, 2001). It is unknown whether dormancy genes of Tibetan semi-wild wheat derived from the D chromosomes of Ae. tauschii. Acknowledgements. The authors are thankful to the National Natural Science Foundation of China (30471088; 30370883), the Science and Technology Committee, and the Education Committee of the Sichuan Province, China for their financial support. Y.-M. Wei and Z.-H. Yan were supported by the Foundation for the Author of National Excellent Doctoral Dissertation of China. Y.-L. Zheng was supported by Program for Changjiang Scholars and Innovative Research Team in University (PCSIRT), China. REFERENCES Cao WG, Scoles GJ, Hucl P, 1997. The genetic of rachis fragility and glume tenacity in semi-wild wheat. Euphytica 94: 119 124 Chen JM, Ren ZL, Gustafson JP, 1996. Heterochromatin differentiation in Xizang cultivated and semi-wild wheat revealed by C-banding technique. Acta Agron Sin 22: 525 529. Chen PD, Liu DJ, Pei GI, 1988. The chromosome constitution of three endemic hexaploid wheat in western China. Proc 7th Int Wheat Genet Symp. Institute of Plant Science Research Cambridge England: 75 80. Chen QF, Yen C, Yang JL, 1998. Chromosomal location of the genes for brittle rachis in the Tibetan weedrace of common wheat. Genet Reas Crop Evol 45: 407 410. Chen QF, Yen C, Yang JL, 1999. Chromosome location of the gene for the hulled character in the Tibetan weedrace of common wheat. Genet Res Crop Evol 46: 543 546. Chen QF, 2001. Inheritance of disarticulation derived from some hexaploid brittle rachis wheat. Genet Res Crop Evol 48: 21 25. Cui YX, Ma YS, 1990. Esterase isozyme of indigenous wheat to China. Acta Bot Sin 32: 39 44.

138 X.J. Lan et al. Gatford KT, Eastwood RF, Halloran GM, 2002a. Germination inhibitors in bracts surrounding the grain of Triticum tauschii. Functional Plant Biology 29: 881 890. Gatford KT, Hearnden P, Ogbonnaya F, Eastwood RF, Halloran GM, 2002b. Novel resistance to pre-harvest sprouting in Australian wheat from the wild relative Triticum tauschii. Euphytica 126: 67 76. Groos C, Gay G, Perretant MR, Gervais L, Bernard M, Dedryver F, Charmet G, 2002. Study of the relationship between pre-harvest sprouting and grain color by quantitative trait loci analysis in a white red grain bread-wheat cross. Theor Appl Genet 104: 39 47. Huang HL, Ma DQ, Zhou RH, 1987. Primary study on and classification of brittle wheat in Tibet, Paper Collection about Crop Resources Investigation in Tibet. Beijing. Chin Agric Sci & Technol Press 33: 39. Huang HL, Weng YJ, Zhang XZ, 2002. Progress of genetic diversity study and protocols for in-situ conservation on semi-wild wheat. J Plant Gen Res 3: 28 33. Lan XJ, Yen C, 1992. An amphidiploid derived from a Chinese landrace of tetraploid wheat, Ailanmai crossed with Aegilops tauschii native to China and with references to its utilization in wheat breeding. J Sich Agr Univer 10: 581 585. Lan XJ, Liu DC, Wang ZR, 1997. Inheritance in synthetic hexaploid wheat RSP of sprouting tolerance derived from Aegilops tauschii Cosson. Euphytica 95: 321 323. Lan XJ, Wei YM, Zheng YL, Liu DC, Zhou YH, 2001. Esterase isozyme of Chinese semiwild types in wheat. J Sich Agric Univer 19: 122 125. Lan XJ, Zheng YL, Wei YM, Liu DC, Yan ZH, Zhou YH, 2002. Tolerant mechanism and chromosome location of gene controlling sprouting tolerance in Aegilops tauschii Cosson. Agric Sci in China 1: 265 268. Liu DC, Lan XJ, Wang ZR, Zheng YL, Zhou YH, Yang JL, Yen C, 1998. Evaluation of Aegilops tauschii Cosson for preharvest sprouting tolerance. Genet Res Crop Evol 45: 495 498. Lu P, Zhang ZL, 1983. Karyotypes and banding patterns of T. aestivum ssp. Yunnanes King and T. aestivum ssp Tibetanum. Hereditas 5: 20 22. Mares DJ, Mrva K, 2001. Mapping quantitative trait loci associated with variation in grain dormancy in Australian wheat. Australian J Agric Res 52: 1257 1265. Mares DJ, Mrva K, Tan MK, Sharp P, Tan MK, Cowan AK, 2002. Dormancy in white-grained wheat: progress towards identification of genes and molecular markers. Euphytica 126: 47 53. Miura H, Sato N, Kato K, Amano Y, 2002. Detection of chromosomes carrying genes for seed dormancy of wheat using the backcross reciprocal monosomic method. Plant Breeding 121: 394 399. Ni ZF, Sun QX, Liu ZR, 1997. Study on wheat heterotic group I. Genetic diversity revealed by RAPD in Triticum aestivum ss. Tibetanum and Triticum spelta. J Agric Biotech 5: 103 111 Roy JK, Prasad M, Varshney RK, Balyan HS, Blake TK, Dhaliwal HS, Singh H, Edwards KJ, Gupta PK, 1999. Identification of a microsatellite on chromosomes 6B and a STS on 7D of bread wheat showing an association with preharvest sprouting tolerance. Theor Appl Genet 99: 336 340. Shao QQ, Li CS, Basang CR, 1980. Semi-wild wheat from Xizang (Tibet). Acta Gen Sin 7: 149 156. Shao QQ, Li CS, Basang CR, 1984. The Study on Semi-wild Type Wheat in Tibet. Crop in Tibet (by Qinghai-Tibet Plateau Crop Resources Collection Groups of Chinese Academy of Science). Beijing. Chin Sci Press 108 113. Wei YM, Zheng YL, Liu DC, Zhou YH, Lan XJ, 2002. HMW-glutenin and Gliadin variation in Tibetan weedrace, Xinjiang rice wheat and Yunnan hulled wheat. Genet Res Crop Evol 49: 327 330 Wei YM, Zheng YL, Zhou YH, Liu DC, Lan XJ, Yan ZH, Zhang ZQ, 2001. Genetic diversity of Gli-1, Gli-2 and Glu-1 alleles among Chinese endemic wheats. Acta Bot Sin 43: 834 839 Wei YM, Zheng YL, Yan ZH, Wu W, Zhang ZQ, Lan XJ, 2003. Genetic diversity in Chinese endemic wheats based on STS and SSR markers. Wheat Inf Serv 97: 9 15 Xiao SH, Run CS, Zhang HP, Sun GZ, 2002. Studies on preharvest sprouting in wheat. Chin Agric Sci & Technol Press 185 211. Yen C, Luo MC, Yang JL, 1988. The origin of the Tibetan weedrace of hexaploid wheat Chinese spring cheng-du-guang-tou and other landraces of the white wheat complex from China. Proceedings of the 7th International Wheat Genetic Symposium, Cambridge England: 175 179. Zanetti S, Winzeler M, Keller M, Keller B, Messmer M, 2000. Genetic analysis of pre-harvest sprouting resistance in a wheat spelt cross. Crop Sci 40: 1406 1417.