JIPB

Journal of Integrative Plant Biology

Association analysis of fiber quality traits and exploration of elite alleles in Upland cotton cultivars/accessions (Gossypium hirsutum L.) State Key Laboratory of Crop Genetics & Germplasm Enhancement, Hybrid Cotton R & D Engineering Research Center, Ministry of Education, Nanjing Agricultural University, Nanjing 210095, China. *Correspondence: [email protected]

Abstract Exploring the elite alleles and germplasm accessions related to fiber quality traits will accelerate the breeding of cotton for fiber quality improvement. In this study, 99 Gossypium hirsutum L. accessions with diverse origins were used to perform association analysis of fiber quality traits using 97 polymorphic microsatellite marker primer pairs. A total of 107 significant marker‐trait associations were detected for three fiber quality traits under three different environments, with 70 detected in two or three environments and 37 detected in only one environment. Among the 70 significant marker‐trait associations, 52.86% were reported previously, implying that these are stable loci for target traits. Furthermore, we detected a large number of elite alleles associated simultaneously with two or three traits. These elite alleles were mainly from accessions collected in China, introduced to China from the United States, or rare alleles with a frequency of less than

5%. No one cultivar contained more than half of the elite alleles, but 10 accessions were collected from China and the two introduced from the United States did contain more than half of these alleles. Therefore, there is great potential for mining elite alleles from germplasm accessions for use in fiber quality improvement in modern cotton breeding.

INTRODUCTION

ate the mining of novel genes, enable the quick and efficient pyramiding of non‐allelic QTLs by marker‐assisted selection (MAS), and enhance the breeding of elite commercial cotton cultivars. Many QTLs related to target traits have been tagged using family‐based QTL mapping methods by assembling intraspecific segregating populations of G. hirsutum with different target traits, such as fiber quality traits (Shen et al. 2005, 2006, 2007; An et al. 2010; Zhang et al. 2012), yield and its components (Guo et al. 2006, 2013; Wang et al. 2006, 2007a, 2007b, 2007c, 2007d; Qin et al. 2008; Liu et al. 2011), resistance traits (Yang et al. 2008; Jiang et al. 2009; Wang et al. 2009a; Wang et al. 2012), and leaf morphology, oil content, drought‐related, and other important agronomic traits (Levi et al. 2011; Alfred et al. 2012). In fact, family‐based linkage mapping can be regarded as a case of linkage disequilibrium (LD) in which LD is generated by establishing mapping populations. The precision of detecting QTL locations depends on the recombination fraction between the QTL for the traits of interest and adjacent markers (Mackay and Powell 2007). Association mapping is another effective approach for connecting phenotypes and genotypes in plants when information on population structure and LD is available (Thornsberry et al. 2001). However, while association mapping is not the same as LD, it actually represents an application of LD. According to Gupta et al. (2005), association mapping refers to the significant association of a marker locus with a phenotypic trait, while LD refers to the nonrandom association between two markers or two genes/QTLs. Breseghello and

www.jipb.net

Keywords: Association mapping; elite allele; fiber quality; population structure; Upland cotton Citation: Cai C, Ye W, Zhang T, Guo W (2014) Association analysis of fiber quality traits and exploration of elite alleles in Upland cotton cultivars/ accessions (Gossypium hirsutum L.). J Integr Plant Biol 56: 51–62. doi: 10.1111/jipb.12124 Edited by: Hai‐Chun Jing, Institute of Botany, CAS, China Received Jun. 1, 2013; Accepted Oct. 28, 2013 Available online on Oct. 29, 2013 at www.wileyonlinelibrary.com/ journal/jipb © 2013 Institute of Botany, Chinese Academy of Sciences

January 2014 | Volume 56 | Issue 1 | 51–62

Free Access

Cotton (Gossypium spp.) is the world’s most important natural textile fiber and a significant oilseed crop. Gossypium hirsutum and Gossypium barbadense L. are two cultivated tetraploid cotton species. G. hirsutum is characterized by its high yield, wide adaptability, and moderate fiber quality; G. barbadense is prized for its superior fiber quality. G. hirsutum is the most widely cultivated cotton species, accounting for 95% of annual worldwide cotton production, while G. barbadense accounts for less than 2% of overall cotton production (National Cotton Council, USA, 2012, http://www.cotton.org/econ/cropinfo/index.cfm). With the acceleration of spinning speeds, the demand for improving cotton fiber quality is increasing rapidly. Therefore, one of the goals of modern Upland cotton breeding is to improve fiber quality by breeding new cultivars with both high yield and elite fiber qualities. China is the largest cotton‐ production country in the world, with the yield of Chinese cotton cultivars equal to or a bit higher than those developed in the United States or Australia. However, the fiber qualities of the Chinese cultivars, especially fiber strength, are not as good as those of American or Australian varieties (Wang et al. 2009b). Therefore, it is important to elucidate the molecular genetics of Upland cotton fiber quality traits. Such information would enable the subsequent improvement of cotton cultivars by pyramiding multiple breeding target traits. Upland cotton is cultivated worldwide. Tagging economically important quantitative trait loci (QTLs) from G. hirsutum accessions in different genetic backgrounds will help acceler-

Research Article

Caiping Cai, Wenxue Ye, Tianzhen Zhang and Wangzhen Guo*

52

Cai et al. traits. This study demonstrates that association mapping and identifying elite alleles can provide genetic information that is complementary to that obtained from family‐based linkage mapping for exploring QTLs related to important fiber quality traits. This study provides molecular information, which can be used for MAS breeding for fiber quality improvement.

Sorrells (2006) demonstrated that association mapping in elite germplasms can enhance the information derived from QTL studies via the implementation of MAS. For complex traits, such as grain yield and abiotic stress tolerance, understanding the interactions between genotypes (markers/QTLs) and the environment is important for enabling the efficient utilization of MAS in plant breeding (Francia et al. 2005). In recent years, association mapping has been widely used in studies of different crops including cotton (Abdurakhmonov et al. 2008, 2009; Kantartzi and Stewart 2008; Zeng et al. 2009; Jena et al. 2011; Kalivas et al. 2011; Zhang et al. 2013). Abdurakhmonov et al. (2008) reported the extent of genome‐ wide LD mapping of fiber quality traits using a core set of 95 microsatellite markers in a total of 285 exotic G. hirsutum accessions, comprising 208 landrace stocks and 77 variety accessions of Mexican and African origin. Further, Abdurakhmonov et al. (2009) reported genetic diversity, population characteristics, the extent of linkage disequilibrium (LD), and association mapping of fiber quality traits using 202 microsatellite marker primer pairs in 335 G. hirsutum germplasm accessions. To mine QTLs related to fiber quality in G. arboreum, 56 G. arboreum accessions introduced from nine regions of Africa, Asia, and Europe were evaluated with 98 simple sequence repeat (SSR) markers (Kantartzi and Stewart 2008). Using a total of 260 lines derived from multiple crosses among tetraploid species in Gossypium and 86 SSR markers, association mapping was performed to analyze lint percent, boll weight, and fiber quality across three environments (Zeng et al. 2009). Recently, Zhang et al. (2013) performed association mapping of agronomic and fiber quality traits in seven diverse environments using 81 G. hirsutum cultivars with 121 SSR markers. Transmission and variation of QTL‐alleles mapped by LD association in three cotton breeding periods revealed that some of these alleles could be detected in almost all Chinese cotton cultivars. These important studies have laid the foundation for analyzing the genetic mechanisms underlying cotton fiber quality traits and the use of MAS in cotton fiber quality breeding. In this study, we used 101 cotton germplasm accessions, including 99 from G. hirsutum and two from G. barbadense, to perform association analysis of fiber quality traits using 97 polymorphic microsatellite marker primer pairs. Further, by combining the results of population‐based association mapping in the current study with previously reported results of family‐based QTL mapping downloaded from the web of science (http://en.wikipedia.org/wiki/Web_of_Science), we were able to identify stable and elite QTLs for fiber quality

RESULTS Genetic diversity SSR markers diversity Based on our previously published genetic linkage map information from G. hirsutum
G. barbadense (Guo et al. 2008) and SSR markers related to fiber quality QTLs reported in previous studies (Shen et al. 2005, 2006, 2007), 260 SSR markers with broad genome‐wide coverage were used to screen the polymorphisms of 12 randomly selected members of 99 G. hirsutum accessions. Of these markers, 97 SSR primer pairs showed polymorphism reproducibility and locus specificity, including 41 and 56 SSR loci in the A‐ and D‐subgenomes, respectively (Table S1). There were 18, eight, and seven SSRs detected in the D8, D2, and D9 chromosomes, respectively. No polymorphic loci were detected in the D3, D4, and D12 chromosomes. The 97 SSRs were used to further screen the 101 cotton accessions that we selected for analysis. A total of 376 alleles were obtained from the 97 SSR markers for the 101 cotton accessions, with an average of 3.876 alleles per locus (ranging from 2 to 10). The average genetic diversity was 0.457 (ranging from 0.058 to 0.807). The average polymorphism information content (PIC) was 0.409 (ranging from 0.056 to 0.785) (Tables 1, S1). There were 48 (12.77%) rare alleles with a frequency of less than 5%, indicating that many new alleles were present in the tested G. hirsutum collection populations. Among these, accessions from the United States contained nine rare alleles, accessions from China contained 41 rare alleles, and 63 varieties contained 25 rare alleles, with varieties from the Yellow River, the Yangtze River, Northern China, and the Northwestern inland containing 15, 13, two, and one rare allele, respectively. The 7,235 germplasm line contained the most rare alleles, with 30 in total, followed by two varieties released in the Yangtze River (Fangwu2 and Yapengmian), each containing four rare alleles. The average alleles per SSR marker for the 63 cotton varieties, 23 accessions collected from China, and 13 introduced accessions from the United States were 3.052, 2.979, and 2.557, respectively. The average gene diversity was 0.430, 0.419, and

Table 1. Genetic diversity of the tested cotton accessions revealed by simple sequence repeat (SSR) markers Accession origin

Sample size

No. loci

No. alleles

99 63 23 29 4 7 23 13 101

97 92 89 90 72 82 96 81 97

325 291 258 262 168 211 288 232 376

99 Gossypium hirsutum accessions 63 varieties The Yellow River The Yangtze River Northwestern inland Northern (Specific early maturation) Accessions from China Accessions from the United States 101 tested accessions January 2014 | Volume 56 | Issue 1 | 51–62

Mean allele No. (range) 3.351 3.052 2.742 2.773 1.990 2.330 2.979 2.557 3.876

(2–10) (2–9) (2–6) (2–6) (2–3) (2–4) (2–7) (2–5) (2–10)

Gene diversity (range) 0.442 0.430 0.398 0.406 0.357 0.411 0.419 0.393 0.457

(0.020–0.801) (0.031–0.733) (0.083–0.737) (0.067–0.773) (0.375–0.625) (0.245–0.735) (0.083–0.798) (0.142–0.746) (0.058–0.807) www.jipb.net

Fiber quality traits and elite alleles in Upland cotton 0.393, respectively (Table 1). We further analyzed the 63 cotton varieties in terms of ecological areas. The average gene diversity values were as high as 0.411 for Northern China, followed by 0.406 for the Yangtze River, 0.398 for the Yellow River, and 0.357 for Northwestern inland. Therefore, the gene diversity in the varieties from the Yangtze River was higher than that in the Yellow River. Phenotypic diversity The cotton accessions used in this study represent a broad variation in fiber quality traits. Among the 99 G. hirsutum accessions, the average fiber length was 28.55 mm (ranging from 24.12 to 32.93 mm), while the average fiber strength was 31.75 cN/tex (ranging from 22.69 to 49.20 cN/tex), and the average fiber fineness (micronaire) was 4.94 (ranging from 3.60 to 6.21). The highest coefficients of variation (CV%) of fiber length and fiber strength were found among accessions collected in China (FL: 6.930, FS: 16.520); the highest CV% of fiber fineness were from accessions from the Northwestern inland area (FF: 12.182). CV% for fiber length and fiber fineness were higher in the Yellow River accessions (FL: 4.569, FS: 9.782, FF: 7.263) than in the Yangtze River accessions (FL: 3.933, FS: 9.799, FF: 6.287). Phenotypic variation for 2 years and three fiber quality traits (fiber length, fiber strength, and fiber fineness) are shown in Table 2. Marker‐fiber quality trait association mapping Population structure A neighbor‐joining tree of the 101 cotton accessions/cultivars was constructed based on Nei’s genetic distances calculated by POWERMARKER v3.25 software. We found that G. hirsutum

53

accessions were not clustered together according to ecological area or pedigree origin (Figure S1). Further, we analyzed the population structure of 99 G. hirsutum accessions using STRUCTURE V2.3.3 software based on 61 unlinked SSR markers. The most likely number of subpopulations (K) was five, according to maximum LnP (D) values (Figure 1A). By contrast, the maximum DK value was observed for K ¼ 2, with a DK value of 251.09 (Figure 1B), which indicated that the entire population could be divided into two subpopulations. The 37 varieties, 17 accessions collected from China, and five accessions collected from the United States were assigned to subpopulation 1, and 26 varieties, six accessions collected from China, and eight accessions collected from the United States were assigned to subpopulation 2 (Figure 1C). When the number of K increased from two to five, some of the accessions in subpopulations 1 and 2 were assigned to subpopulation 3, and all accessions in subpopulations 4 and 5 were from subpopulation 3 (Figure 1C). Linkage disequilibrium and LD decay Linkage disequilibrium analysis of 97 SSR markers was estimated using the TASSEL 2.0.1 software package. The distribution of LD among 97 SSR markers in 99 G. hirsutum accessions is shown in Figure S2. Among the 4,656 SSR locus combinations, 1 017 (21.84%) showed significant (P < 0.05) LD for the 99 G. hirsutum accessions (Table 3). The interchromosomal LD was 896, which was higher than the intrachromosomal LD of 121. In subpopulations 1 and 2, the LD percentages were 11.23% and 6.49%, respectively. Considering the accession origins, in the Yellow River accessions, the Yangtze River accessions, accessions collected in China, and accessions collected in the United States, the LD percentages were

Table 2. Phenotype statistics of three fiber quality traits in the tested cotton cultivars FL (2.5% fiber span length, mm)

99 Gossypium hirsutum accessions 63 varieties The Yellow River The Yangtze River Northwestern inland Northern China area Accessions from China Accessions from the US 101 tested accessions

FS (fiber strength, cN/tex)

FF (fiber fineness, mm)

Mean (range)

SD

CV%

Mean (range)

SD

CV%

Mean (range)

SD

CV%

28.55 (24.12–32.93)

1.501

5.257

31.75 (22.69–49.2)

4.232

13.329

4.94 (3.6–6.21)

0.43

8.704

28.06 28.17 27.97 28.82 27.64 29.19 29.81 28.58

1.131 1.287 1.1 0.891 0.638 2.023 0.714 1.553

4.031 4.569 3.933 3.092 2.308 6.93 2.395 5.434

30.07 30.32 29.83 30.3 30.12 34.74 34.61 31.75

2.779 2.966 2.923 2.138 2.202 5.739 2.28 4.325

9.242 9.782 9.799 7.056 7.311 16.52 6.588 13.622

5.01 4.86 5.09 5.27 5.06 4.76 4.94 4.93

0.388 0.353 0.32 0.642 0.501 0.517 0.397 0.438

7.745 7.263 6.287 12.182 9.901 10.861 8.036 8.884

(25.21–32.31) (25.6–32.31) (25.21–29.87) (27.5–29.39) (26.57–28.48) (24.12–32.93) (28.7–30.96) (24.12–32.93)

(22.69–37.04) (22.69–37.04) (23.39–35.18) (27.77–32.93) (27.31–32.46) (26.6–49.2) (30.94–38.25) (22.69–49.2)

(4.21–6.21) (4.21–5.52) (4.53–6) (4.8–6.21) (4.3–5.75) (3.6–5.6) (4.15–5.45) (3.6–6.21)

CV%, coefficients of variation; SD, standard deviation.

Figure 1. Continued www.jipb.net

January 2014 | Volume 56 | Issue 1 | 51–62

54

Cai et al.

Figure 1. Analysis of population structure (A) LnP(D) for 99 Gossypium hirsutum accessions. (B) DK for 99 G. hirsutum accessions. (C) Bar plot showing distribution of genotypes within subpopulations, k ¼ 2–5. January 2014 | Volume 56 | Issue 1 | 51–62

www.jipb.net

Fiber quality traits and elite alleles in Upland cotton

55

Table 3. Percentage of simple sequence repeat (SSR) loci with significant linkage disequilibrium (LD) (P < 0.05) in the tested Gossypium hirsutum accessions Accession The Yellow River The Yangtze River Accessions from China Accessions from the US Subpopulation 1 Subpopulation 2 Total

Sample size 23 29 23 13 59 40 99

Intra Chr. LD% (LD/non‐LD) 9.03 9.38 21.88 5.9 30.56 11.46 42.01

4.83%, 6.23%, 11.83%, and 5.00%, respectively. We found that the observed percentage of locus combinations in LD within each subgroup was lower than that in the 99 G. hirsutum accessions. A similar phenomenon was reported in Li et al. (2008) and Vanniarajan et al. (2012). We also investigated the LD decay rate using the r2 values (P < 0.05) of the SSR loci on the same chromosomes. As shown in Figure 2, the intrachromosomal LD decay occurred at a genetic distance of approximately 12 cM (r2 < 0.1). In some regions, LD decay between SSR loci further extended to 70 cM. When r2 < 0.2 LD was reduced to approximately 1–2 cM, revealing the potential for association mapping. Association mapping Association mapping using the mixed linear model (MLM) and General linear models (GLM) was performed for both years of the study and for the three fiber quality traits data (fiber length, fiber strength, and fiber fineness). This analysis revealed 196 marker‐trait associations involving 56 SSR markers for three fiber qualities, of which 96.94% were common associations (P < 0.05) derived from both the MLM and GLM models (Table S2). Of these associations, 70 significant associations were stably detected in two or three environments (2004, 2007, and the 2‐year average of target traits; Table 4), including 30 for fiber length, 27 for fiber

Figure 2. Linkage disequilibrium (LD) decay plot of 99 Gossypium hirsutum accessions Inner fitted trend line is a non‐linear logarithmic regression curve of r2 versus genetic distance. www.jipb.net

(26/262) (27/261) (63/225) (17/271) (88/200) (33/255) (121/167)

Inter Chr. LD% (LD/non‐LD) 4.56 6.02 11.17 4.95 9.96 6.16 20.51

(199/4 169) (263/4 105) (488/3 880) (216/4 152) (435/3 933) (269/4 099) (896/3 472)

Total LD% (LD/non‐LD) 4.83 6.23 11.83 5 11.23 6.49 21.84

(225/4 431) (290/4 366) (551/4 105) (233/4 423) (523/4 133) (302/4 354) (1 017/3 639)

strength, and 13 for fiber fineness. A total of 37 associations were detected in only one environment. All 107 marker‐trait associations were employed to explore elite alleles. We also found that some SSR markers were simultaneously associated with two or three traits of interest. Among the 70 significant associations markers, five markers (BNL1317, BNL2634, JESPR295, NAU934, and TMO06) were significantly associated with all three tested fiber quality traits, while 15 markers (BNL1395, BNL1521, BNL3145, JESPR127, JESPR153, JESPR78, NAU1037, NAU1262, NAU1322, NAU1336, NAU2272, NAU2443, NAU2508, NAU780, and TMD05) were associated with both fiber length and strength, NAU1302 was associated with both fiber length and fineness, and BNL3436 was associated with both fiber strength and fineness. In the study, we found that some SSR loci associated with fiber quality traits were clustered on specific chromosomes or chromosome regions. A total of 18 SSR markers were selected for association analysis on the D8 chromosome, and markers in this region that were associated with fiber strength and length were further confirmed. In previous reports of family‐based QTL mapping, a major fiber strength QTL in this region was subjected to fine mapping (Chen et al. 2009) and used for MAS breeding for fiber quality improvement (Guo et al. 2003; Kumar et al. 2012). Our results further confirmed that there are important QTLs/ genes related to elite fiber quality traits on the D8 chromosome, which were detected by both population‐based association and family‐based QTL mapping detection. Therefore, this chromosomal region is an important candidate region for studying the molecular mechanisms underlying fiber quality and for use in breeding cotton cultivars for improved fiber quality. Exploration of elite alleles and germplasm accessions Further exploration of elite alleles for all 107 marker‐trait associations, and 176 elite alleles associated with three fiber qualities, was carried out (Table S3). Among the 176 elite alleles, 68 were associated with fiber length, 67 were associated with fiber strength, and 41 were associated with fiber fineness. We also found that some elite alleles were simultaneously associated with two or three traits. For example, 43 elite alleles were associated with fiber length and fiber strength, 21 associated with fiber length and fiber fineness, 22 with fiber strength and fiber fineness, and 15 with all three tested traits. It is worth noting that of the 48 rare alleles with a frequency of less than 5%, 29 were elite alleles, with 20 associated with fiber length, 23 with fiber strength, and seven with fiber fineness. Of these, three rare alleles (BNL1395‐C, NAU1302‐A, and NAU2272‐ E) were associated with all three traits, 16 were associated with both fiber length and strength, and five were associated with January 2014 | Volume 56 | Issue 1 | 51–62

56

Cai et al.

Table 4. Significant simple sequence repeat (SSR) marker‐fiber quality trait associations, explained phenotypic variation in Gossypium hirsutum accessions Trait

Marker loci

r2 (%)a

Position 2004

FL

FS

NAU2277 NAU1167 NAU1200 NAU934 NAU2156 NAU1037 NAU2354 BNL1317

A2 (110.156) A3 (105.325) A5 A5 (186.594) A6 (69.535) A8 (53.526) A9 (132.251) A9 (88.456)

NAU2508 NAU445 JESPR295 JESPR153 NAU2272 NAU1102 JESPR218 BNL2634 BNL1395 NAU780 NAU1302 BNL1521 JESPR78 NAU1336 TMD05 NAU1262 BNL3145 NAU1322 JESPR127 TMO06 NAU904 NAU2443 NAU934 NAU1043 NAU474 NAU1037

A10 (126.827) A12 (123.764) A12 (32.732) A13 (63.046) D2 (55.105) D5 D5 (144.059) D7 D7 (78.652) D8 D8 (72.101) D8 (73.477) D8 (75.81) D8 (78.13) D8 (82.452) D8 (88.599) D8 (89.76) D8(89.779) D8 (97.939) D9 (109.242) D10 (53.078) D13 (62.214) A5 (186.594) A7 A7 (30.627) A8 (53.526)

BNL1317

A9 (88.456)

NAU2508 JESPR295 JESPR153 NAU2272 BNL3436 NAU905 BNL2634 BNL1395 BNL1122

A10 (126.827) A12 (32.732) A13 (63.046) D2 (55.105) D6 (105.073) D6 (20.168) D7 D7 (78.652) D7 (88.995)

NAU780 NAU816 BNL1521 JESPR78 NAU1197 NAU1336 TMD05

D8 D8 D8 (73.477) D8 (75.81) D8 (77.384) D8 (78.13) D8 (82.452)

2007

10.47 21.54 17.4 7.83 7.75 10.4 12.18

5.11 9.82 17.83 13.32 14.12 14.77

Previous reports QTLsb Mean 10.32 21.26 13.25 10.14 10.31 11.35 14.49 16.86

9.36 21.56 18.24 16.21 16.21 16.21 6.93 16.21 6.93 10.12 16.21 19.56 13.88 5.53 6.19 15.86 17.44 9.43

14.11 6.77g 19.25 7.52 11.88 13.85 9.65 8.81 10.94 13.2 12.83 11.12 11.12 11.11 4.6g 11.98 4.6g 9.26 11.11 22.52 11.19 7.56 11.87 14.03 13.52 7.7

7.74

12.58

13.48

10.56 23.3

23.99 14.19 9.25 10.48

13.62 14.19

15.25 18.78 13.81 7.85g 7.22 10.34 7.33 11.84 12.65

18.14 6.65 9.26 9.26 7.76 9.26 6.21

13.17 9.4 7.66 7.66 6.12 7.66 5.64

10.22 11.56 13.52

15.17

9.32

16.26 10.06 11.11

10.02

16.86 7.99 14.08 10.02

7.43 12.72 12.47

13.6

Zhang et al. 2013 Qin et al. 2009; Lu 2005 Shen et al. 2006, 2007

Zhang et al. 2012 Shen et al. 2005, 2006, 2007, Wang J et al. 2007, Qin et al. 2008 Shen et al. 2006, 2007 Shen et al. 2005, Sang 2003 Shen et al. 2006, 2007 Shen et al. 2006, 2007 Shen et al. 2005

Chen et al. 2009 Shen et al. 2006, 2007 Chen et al. 2009 Chen et al. 2009 Shen et al. 2005, Chen et al. 2009 Shen et al. 2005, Chen et al. 2009

Wang et al. 2006 Shen et al. 2005, 2006, 2007, Chen et al. 2009 Wang J et al. 2007, Qin et al. 2009, Kalivas et al. 2011

Shen et al. 2005 An et al. 2010, Jin 2003 Shen et al. 2005, Abdurakhmonov et al. 2009, Zeng et al. 2009 Shen et al. 2006, 2007, Chen et al. 2009 Shen et al. 2006, 2007 Chen et al. 2009 Shen et al. 2005, Chen et al. 2009 Shen et al. 2006, 2007 Shen et al. 2006, 2007, Sang 2003, Chen et al. 2009, Lu 2005 (Continued)

January 2014 | Volume 56 | Issue 1 | 51–62

www.jipb.net

Fiber quality traits and elite alleles in Upland cotton

57

Table 4. (Continued)

Trait

Marker loci

r2 (%)a

Position 2004

FF

NAU1262 BNL3145

D8 (88.599) D8(89.76)

NAU1322 JESPR127 TMO06 NAU2443 NAU895 NAU934 BNL1317 BNL3902 NAU1162 JESPR295 NAU803 BNL3436 BNL2634 NAU1369 NAU1302 NAU1004 TMO06

D8 (89.779) D8 (97.939) D9 (109.242) D13 (62.214) A2 (111.631) A5 (186.594) A9 (88.456) D1 (62.419) A11 (99.954) A12 (32.732) D2 (55.392) D6 (105.073) D7 D8 (57.434) D8 (72.101) D9 D9 (109.242)

9.22 23.66 22.85

18.74 11.43 12.95

Previous reports QTLsb

2007

Mean

9.26 6.21

7.66 5.64

10.82 9.26 19.23 13.02 9.17 12.62 9.71 4.61

9.22 8.33 15.25 17.55 11.39 10.75 16.61 5.98 14.73 17.86 7.93 9.41 6.45g 7.53 13.8 11.13 15

7.96 9.24

9.68 9.47 6.22 8.32 13.67 12.84

Shen et al. 2006, 2007, Chen et al. 2009, Kumar et al. 2012 Shen et al. 2005, Sang 2003, Chen et al. 2009 Shen et al. 2006, 2007

Shen et al. 2006, 2007, Kalivas et al. 2011 Shen et al. 2005 Shen et al. 2005 Shen et al. 2005 Shen et al. 2005

Shen et al. 2005, 2006, 2007

Explained phenotypic variation corresponding to association markers (P < 0.05), further, P < 0.01 is indicated in bold. bBold indicates quantitative trait loci consensus with that in this study are from previous association analysis. g means significance detected only in general linear model (GLM) test after 100 000‐time permutation test.

a

both fiber strength and fiber fineness. These results suggest that there are elite alleles associated with fiber quality among rare alleles. More attention should therefore be paid to elite rare alleles for discovery and utilization. We further examined elite alleles associated with fiber quality traits in different subgroups. We found that some elite alleles existed in almost all of the accessions that were examined, including alleles such as NAU1280‐D, NAU2190‐A, NAU2156‐A, NAU904‐E, and NAU2354‐D. Furthermore, there were some elite alleles that were commonly found in the 23 accessions collected from China but were uncommon in the 63 varieties examined. For example, JESPR153‐C, NAU2443‐B, and BNL3436‐C were detected in the majority of accessions collected from China, with frequencies of 56.52%, 52.17%, and 91.30%, respectively. However, the frequencies of these three alleles were only 1.59%, 4.76%, and 49.21% in the 63 varieties. On the other hand, some elite alleles were common in the 13 accessions introduced from the United States but were quite rare in the accessions collected from China and the 63 varieties, including alleles such as NAU2277‐E, NAU1043‐B, NAU1167‐H, JESPR295‐B, NAU474‐B, and NAU1200‐A. For the 15 elite alleles that were simultaneously associated with three fiber quality traits, only NAU1302‐D was found in 36 of the 63 varieties, while the other 14 elite alleles (NAU934‐D, BNL1317‐C, NAU1369‐B, JESPR295‐B, JESPR295‐C, JESPR153‐A, JESPR153‐C, NAU2272‐E, BNL1395‐A, BNL1395‐C, NAU1302‐A, TMO06‐C, BNL2634‐C, and NAU1004‐B) were present in no more than 25% of the 63 varieties. These results imply that many elite alleles for fiber quality improvement that exist in germplasm accessions have been lost during commercial Upland cotton cultivar breeding. We further mined the accessions carrying the elite alleles. The average numbers of elite alleles in the 99 G. hirsutum accessions was 38.081, ranging from 19 (I‐62437) to 83 (7235). www.jipb.net

Only 12 accessions, which included 10 accessions collected from China (7235, I‐62434, I‐62429, I‐62433, I‐62440, I‐62441, I‐62435, I‐62431, I‐62436, and Yumian 1) and two introduced from the United States (HS427 and PD93019), contained more than half of the elite alleles (>53.5 of the elite alleles). These results suggest that there are many elite alleles for fiber quality traits in the germplasm accessions, and there is great potential for improving fiber quality traits by introducing these elite alleles via MAS in Upland cotton cultivar breeding programs.

DISCUSSION G. hirsutum is the most widely cultivated cotton, comprising 95% of cultivated cotton worldwide. Modern Upland cotton breeding has focused on breeding varieties with high yields and elite fiber quality due to the accelerated spinning speeds and the improved cotton quality expected by consumers. However, modern Upland cotton cultivars have an intrinsically narrow genetic base and low genetic diversity due to long‐term domestication and breeding performed with limited resources (Chen and Du, 2006). We therefore explored elite alleles for fiber quality traits in various germplasm accessions to increase molecular markers for MAS breeding for fiber quality improvement. Association mapping is an effective method for exploring significant markers‐traits or elite alleles Family‐based linkage mapping is a key tool for identifying the genetic basis of quantitative traits in plants. However, constructing segregation population is required for linkage mapping, and studies have sometimes been limited by low polymorphism, small population size, and poor stability. Association mapping is an approach that exploits naturally January 2014 | Volume 56 | Issue 1 | 51–62

58

Cai et al.

occurring haplotype blocks that are conserved in the germplasm (Malysheva‐Otto et al. 2006; Rostoks et al. 2006). Association mapping offers three main advantages over linkage mapping, providing a higher mapping resolution, an increased number of identified alleles, and a more rapid way to establish a marker‐trait association and to use this information in breeding programs (Flint‐Garcia et al. 2003). In the present study, 99 G. hirsutum accessions (including 63 varieties, 23 accessions collected from China, and 13 introduced from the United States) were used to carry out association analysis of fiber quality traits using 97 polymorphic microsatellite marker primer pairs. We detected 107 marker‐ trait associations for three fiber quality traits (Tables 3, S3), with 70 significant marker‐trait associations detected simultaneously in two or three environments. Among the 63 varieties tested, not one variety contained more than half of the elite alleles (>53.5 elite alleles). Therefore, there is great potential for breeding modern cotton cultivars with improved fiber quality traits. In the present study, we found that there were three types of elite alleles, including the following: (i) elite alleles associated simultaneously with two or three traits; (ii) elite alleles that are commonly present in accessions collected from China or introduced from the United States; and (iii) elite rare alleles (with a frequency of less than 5%). We hope that these elite alleles can be used to accelerate fiber quality trait improvement in modern Upland cotton cultivar breeding programs via MAS. Lines containing many elite alleles for fiber quality traits in the study were mainly introgression line accessions. These accessions, such as Yumian 1, 7235, HS427, PD93019, and others, contained more than half of all tested elite alleles (>53.5 elite alleles) and were mainly collected from China or introduced from the United States. Pedigree analysis showed that 7235 was derived from a G. anomalum introgression line and simultaneously hybridized with other elite fiber strains, such as Acala 3080 and PD4381, during the process of development (Qian et al. 1992; Zhou et al. 2006). Yumian 1 is an introgression line with G. barbadense, G. arboreum, and G. raimondii as putative donors, which is characterized by high lint yield and high fiber strength (Zhang et al. 2005). A detailed pedigree of HS427 has not been reported, but this line is known to be from an Acala background (S. Oakley, pers. comm.). Association mapping further suggested that allelic relationships of different elite fiber genes and possible common gene origins across different donors may have contributed to the cluster of QTLs, as well as the contribution from introgression strains of the PD or Acala series or from G. barbadense. Comparison of family‐based QTL mapping results with the results of other association mapping studies We compared the trait‐associated SSR loci detected in the current study with SSR loci for fiber quality traits identified in previous family‐based QTL and population‐based association mapping studies. Among the 70 significant marker‐trait associations that were detected in two or three environments, 36 SSR loci coincided with previously reported QTLs detected based on family‐based QTL mapping, including 15 for fiber length, 15 for fiber strength, and six for fiber fineness. An additional four marker‐trait associations corresponded to previously reported association mapping results (Table 4). These stably inherited QTLs/genes, which simultaneously were January 2014 | Volume 56 | Issue 1 | 51–62

detected by different segregating populations with different genetic backgrounds, can potentially be used in MAS breeding. The remaining 34 SSR loci did not coincide with previously reported QTLs, which are novel and previously unreported loci resources associated with fiber quality traits. Four marker‐trait associations corresponded to previously reported association mapping results. In the present study, BNL1122 was associated with fiber strength, while this QTL was reported to be associated with fiber strength by Abdurakhmonov et al. (2009) and Zeng et al. (2009), as well as fiber length (Kantartzi and Stewart 2008; Abdurakhmonov et al. 2009) and fiber fineness (Kantartzi and Stewart 2008). At the same time, a QTL for fiber strength (qFS‐16‐1b) was identified in the (TM‐1
HS427) F2 and F2:3 populations, at marker interval BNL1017–BNL1122, which explained 10.2% and 12.0% of phenotypic variation (PV) in F2 and F2:3, respectively (Shen et al. 2005). We found that BNL1317 was associated with fiber length, fiber strength, and fiber fineness, and this QTL was reported to be associated with fiber strength and fiber fineness by Kalivas et al. (2011). Near this marker, a QTL for fiber length, fiber strength, and fiber fineness was also previously detected by family‐based QTL mapping (Shen et al. 2005, 2006, 2007; Wang et al. 2007d; Qin et al. 2008, 2009). NAU2277 was associated with fiber length in this study and in Zhang et al. (2013). These QTLs, which were detected in different populations with different genetic backgrounds, were stably inherited and can be used in MAS breeding for fiber quality improvement. Elite alleles accelerate the process of fiber quality trait improvement in Upland cotton cultivar breeding programs via MAS In the modern textile industry, stronger, thinner, more uniform cotton fiber is needed to match higher spinning speeds. Therefore, the demand for improved cotton fiber quality has been increasing. The use of molecular markers that are tightly linked to major QTLs/genes for high quality traits in MAS will greatly accelerate the improvement of fiber quality in commercial cultivars. Furthermore, some markers that are simultaneously associated with several traits of interest may be used to improve targeted traits by MAS. Previous reports have revealed a region on the D8 chromosome that is stably associated with elite fiber quality traits. For example, in 7235, one QTL could explain 35%–53.8% of the phenotypic variation (PV) in F2 and F2:3, respectively (Zhang et al. 2003; Shen et al. 2005). This QTL was genetically stable in different genetic backgrounds and different environments (Shen et al. 2006). Furthermore, five fiber length QTLs accounted for 20.1%–32.7% of PV, while five fiber strength QTLs accounted for 28.8%–59.6% of PV, and four fiber fineness QTLs accounted for 17.9%–41.5% of PV, all of which were confirmed by fine mapping (Chen et al. 2009). Recently, Kumar et al. (2012) also confirmed the effects of the major QTL for fiber strength in 7235 by crossing 7235 with two US germplasm lines with different fiber strengths. In addition, 27 novel markers that are tightly linked in this region were tagged for the QTL and were deployed for breeding cultivars with improved fiber strength (Kumar et al. 2012). In this study, we also found that some SSR markers were associated simultaneously with two or three traits of interest, and some associated SSR loci clustered in specific chromosomes or chromosome regions. On the D8 chromosome, 18 SSR markers www.jipb.net

Fiber quality traits and elite alleles in Upland cotton were selected for association analysis, and major fiber strength and fiber length QTLs/genes in this region were further confirmed. These elite QTLs/genes may prove to be stable, powerful donors for fiber quality trait improvement in modern cotton breeding. Using MAS technology, cotton molecular breeding programs have already produced new cotton strains with improved fiber quality. In a previous study, the major QTL for fiber strength in 7235 was used to improve cotton fiber strength in China (Guo et al. 2005). For example, there was a significant difference in the mean of fiber strength in plants with or without SSR marker BNL1521 (36.32 cN/tex vs. 33.64 cN/ tex, respectively). This major QTL was expected to increase cotton fiber quality by 2–3cN/tex. As a result of this study, QTLs for stronger fiber strength and transgene cryIA have been rapidly pyramided, and a new strain characterized by elite fiber qualities, insect‐resistance, and high yield potential was bred (Guo et al. 2005). In the current study, we detected 70 significant marker‐trait associations in two or three environments, with 36 and four SSR markers associated with earlier reported SSR markers developed by family‐based QTL mapping and association mapping, respectively. Stably inherited QTLs/genes are present near these markers, which may be used in MAS and improving fiber quality via cotton breeding. Recently, a preliminary map of the whole‐genome scaffold sequence of diploid cotton G. raimondii, the closest living model of the ancestral genome donor for tetraploid cotton species, was released by two different groups (Wang et al. 2012b, Paterson et al. 2012). As an application, G. raimondii genome sequences have provided remarkable advantages to assemble the tetraploid transcriptome and mine candidate genes of interest (Zhu and Li, 2013). So, information from the publically available database in Gossypium will serve for elite gene discovery of traits with interest such as fiber quality. Furthermore, restriction‐site associated DNA (RAD) or single nucleotide polymorphisms (SNPs) developed from whole‐ genome sequences will be used for association mapping and application of improving important agronomic traits in Upland cotton cultivar breeding programs via MAS.

59

were categorized according to the release years, including nine varieties released before 1960, 10 varieties released in the 1960s–1970s, and 44 varieties released in the 1980s– 1990s (Tables 1, S4). The names, origins, release years, and ecological areas of the 101 cotton accessions are listed in Table S4. In both 2004 and 2007, the 101 cotton accessions were planted in the Jiangpu experimental station of Nanjing Agricultural University, Nanjing, Jiangsu Province, China and grown using normal field practices. Fiber samples were tested in the Supervision, Inspection, and Test Center of Cotton Quality, Ministry of Agriculture in China using three biological replicates. The analysis of fiber quality traits focused mainly on 2.5% fiber span length (FL), fiber strength (FS), and fiber fineness (FF). Genomic DNA was isolated from all 101 cotton cultivars as described by Paterson et al. (1993). Simple‐sequence repeat polymerase chain reaction (SSR‐PCR) amplifications were performed using a Peltier Thermal Cycler‐225 (MJ Research), and electrophoresis of the products was performed as described by Zhang et al. (2000, 2002). Based on our previously published genetic linkage map information from G. hirsutum
G. barbadense (Guo et al. 2008), which contains 2,247 loci in 26 linkage groups, SSR markers at approximately 10 cM intervals in each chromosome were selected to ensure broad genome‐wide coverage of genotyping and a representative estimation of genetic distances. Furthermore, SSR markers related to fiber quality QTLs reported in previous studies (Shen et al. 2005, 2006, 2007) were also selected for association mapping. As a result, 260 primer pairs were used to screen 101 accessions. Of these, 97 showed polymorphisms in 23 chromosomes of the cotton genome, with no polymorphic loci detected in D3, D4, and D12. All SSR primer pair information can be downloaded from http://www.cottonmarker.org/ or http://www.CottonGen.com/. Amplification products were scored as either a present (1) or an absent (0) band. The alleles were coded A, B, C, and so on according to molecular weight size. Data analysis

MATERIALS AND METHODS Plant materials A total of 99 Gossypium hirsutum L. and two Gossypium barbadense L. germplasm accessions (Table S4) were selected for this experiment. These germplasm accessions were made available from the cotton germplasm collection in Hybrid Cotton R&D Engineering Research Center, Ministry of Education (Nanjing, China) and Institute of Cotton Research, Chinese Academy of Agricultural Science (CRI‐CAAS). Among these accessions, two G. barbadense accessions were used as controls in genetic diversity analysis, and 99 G. hirsutum accessions (including 63 varieties, 23 accessions collected from China, and 13 introduced from the United States) were used for association mapping and exploration of elite alleles. The 63 varieties were further divided into four different ecological areas according to the ecological characteristics of the cotton growing regions, including 23 varieties from the Yellow River, 29 varieties from the Yangtze River, four varieties from the Northwestern inland, and seven varieties from Northern China (specifically, the early maturation area). Furthermore, the 63 varieties www.jipb.net

Genetic diversity POWERMARKER v 3.25 (Liu and Muse 2005, http://statgen. ncsu.edu/powermarker/) software was used to estimate the number of alleles per marker, the gene diversity, and the polymorphism information content (PIC) of 97 SSR primer pairs for the tested cotton accessions. Nei’s (1983) genetic distances among the 101 tested cotton accessions were calculated, and a neighbor‐joining dendrogram was constructed with POWERMARKER v 3.25 software. The neighbor‐joining tree was edited using FigTree v1.3.1 (Rambaut 2009, http://tree.bio.ed.ac.uk/ software/figtree/). Association mapping Population structure analysis The model‐based software STRUCTURE v 2.3.3 (Pritchard and Wen 2004, http://pritch.bsd.uchicago.edu/software.html) was used to infer the population structure of 99 G. hirsutum accessions (K ¼ 1 to K ¼ 10, with three runs at each K) using a burn‐in of 10 000 and a run length of 100 000. An assumption of STRUCTURE software is that the loci are unlinked. Therefore, in January 2014 | Volume 56 | Issue 1 | 51–62

60

Cai et al.

this study, 36 markers (between two adjacent loci 0, such an allele is a positive allele, and if ai < 0, the allele is a negative allele.

ACKNOWLEDGEMENTS This program was financially supported in part by the National Science Foundation in China (30871558), the National High Technology Research and Development Program of China (863 January 2014 | Volume 56 | Issue 1 | 51–62

Program) (2012AA101108‐04‐04), Jiangsu Agriculture Science and Technology Innovation Fund (cx(13)3059), and a project funded by the Priority Academic Program Development of Jiangsu Higher Education Institutions.

REFERENCES Abdurakhmonov IY, Kohel RJ, Yu JZ, Pepper AE, Abdullaev AA, Kushanov FN, Salakhutdinov IB, Buriev ZT, Saha S, Scheffler BE, Jenkins JN, Abdukarimov A (2008) Molecular diversity and association mapping of fiber quality traits in exotic G. hirsutum L. germplasm. Genomics 92: 478–487 Abdurakhmonov IY, Saha S, Jenkins JN, Buriev ZT, Shermatov SE, Scheffler BE, Pepper AE, Yu JZ, Kohel RJ, Abdukarimov A (2009) Linkage disequilibrium based association mapping of fiber quality traits in G. hirsutum L. variety germplasm. Genetica 136: 401–417 Alfred Q, Liu HY, Xu HM, Li JR, Wu JG, Zhu SJ, Shi CH (2012) Mapping of quantitative trait loci for oil content in cottonseed kernel. J Genet 91: 289–295 An C, Jenkins JN, Wu J, Guo Y, McCarty JC (2010) Use of fiber and fuzz mutants to detect QTL for yield components, seed, and fiber traits of upland cotton. Euphytica 172: 21–34 Bradbury PJ, Zhang Z, Kroon DE, Casstevens RM, Ramdoss Y, Buckler ES (2007) TASSEL: Software for association mapping of complex traits in diverse samples. Bioinformatics 23: 2633–2635 Breseghello F, Sorrells ME (2006) Association mapping of kernel size and milling quality in wheat (Triticum aestivum L.) cultivars. Genetics 172: 1165–1177 Chen G, Du XM (2006) Genetic diversity of source germplasm of upland cotton in China as determined by SSR marker analysis. Acta Genetica Sin 33: 733–745 Chen H, Qian N, Guo W, Song Q, Li B, Deng F, Dong C, Zhang T (2009) Using three overlapped RILs to dissect genetically clustered QTL for fiber strength on Chro.D8 in upland cotton. Theor Appl Genet 119: 605–612 Evanno G, Regaut S, Goudet J (2005) Detecting the number of clusters of individuals using the software STRUCTURE: A simulation study. Mol Ecol 14: 2611–2620 Flint‐Garcia SA, Thornsberry JM, Bukler ES (2003) Structure of linkage disequilibrium in plants. Ann Rev Plant Biol 54: 357–374 Francia E, Tacconi G, Crosatti C, Barabaschi D, Bulgarelli D, Dall’Aglio E, Valè G (2005) Marker assisted selection in crop plants. Plant Cell Tiss Org 82: 317–342 Guo WZ, Cai CP, Wang CB, Zhao L, Wang L, Zhang TZ (2008) A preliminary analysis of genome structure and composition in Gossypium hirsutum. BMC Genomics 9: 314 Guo WZ, Ma GJ, Zhu YC, Yi CX, Yu J, Kohel RJ, Zhang TZ (2006) Molecular tagging and association of morphological markers and QTLs for lint percentage in molecular genetic map of Upland cotton. J Integr Plant Biol 48: 320–326 Guo WZ, Zhang TZ, Ding YZ, Zhu YC, Shen XL, Zhu XF (2005) Molecular marker assisted selection and pyramiding of two QTLs for fiber strength in upland cotton. Acta Genet Sin 32: 1275–1285 Guo WZ, Zhang TZ, Shen XL, Yu JZ, Kohel RJ (2003) Development of SCAR marker linked to a major QTL for high fiber strength and its usage in molecular–marker assisted selection in upland cotton. Crop Sci 43: 2252–2256 Guo X, Guo YP, Ma J, Wang F, Sun MZ, Gui LJ, Zhou JJ, Song XL, Sun XZ, Zhang TZ (2013) Mapping heterotic loci for yield and agronomic traits using chromosome segment introgression lines in cotton. J Integr Plant Biol 55: 759–774

www.jipb.net

Fiber quality traits and elite alleles in Upland cotton Gupta PK, Rustgi S, Kulwal PL (2005) Linkage disequilibrium and association studies in higher plants: Present status and future prospects. Plant Mol Biol 57: 461–485 Hedrick PW (1987) Gametic disequilibrium measures: Proceed with caution. Genetics 117: 331–341 Jena SN, Srivastava A, Singh UM, Roy S, Banerjee N, Rai KM, Singh SK, Kumar V, Chaudhary LB, Roy JK, Tuli R, Sawant S (2011) Analysis of genetic diversity, population structure and linkage disequilibrium in elite cotton (Gossypium L.) germplasm in India. Crop Pasture Sci 62: 859–875 Jiang F, Zhao J, Zhou L, Guo W, Zhang T (2009) Molecular mapping of Verticillium wilt resistance QTL clustered on chromosomes D7 and D9 in upland cotton. Sci China C Life Sci 52: 872–884 Jin JP (2003) Genetie Analysis of Quantitative Traits and Recurrent Selection of Yield Traits in Upland Cotton. A Dissertation for the Doctorate Degree Nanjing Agricultural University, Supervisor: Zhang TZ, Nanjing, China Kalivas A, Xanthopoulos F, Kehagia O, Tsaftaris AS (2011) Agronomic characterization, genetic diversity and association analysis of cotton cultivars using simple sequence repeat molecular markers. Genet Mol Res 10: 208–217 Kantartzi SK, Stewart JM (2008) Association analysis of fibre traits in Gossypium arboreum accessions. Plant Breed 127: 173–179 Kumar P, Singh R, Lubbers EL, Shen XL, Paterson AH, Campbell BT, Jones DC, Chee P (2012) Mapping and validation of fiber strength quantitative trait loci on Chromosome 24 in upland cotton. Crop Sci 52: 1115–1122 Levi A, Paterson AH, Cakmak I, Saranga Y (2011) Metabolite and mineral analyses of cotton near‐isogenic lines introgressed with QTLs for productivity and drought‐related traits. Physiol Plant 141: 265–275 Li Y, Guan R, Liu Z, Ma Y, Wang L, Li L, Lin F, Luan W, Chen P, Yan Z, Guan Y, Zhu L, Ning X, Smulders MJ, Li W, Piao R, Cui Y, Yu Z, Guan M, Chang R, Hou A, Shi A, Zhang B, Zhu S, Qiu L (2008) Genetic structure and diversity of cultivated soybean (Glycine max (L.) Merr.) landraces in China. Theor Appl Genet 117: 857–871 Liu K, Muse SV (2005) PowerMarker: Integrated analysis environment for genetic marker data. Bioinformatics 21: 2128–2129 Liu R, Wang B, Guo W, Wang L, Zhang T (2011) Differential gene expression and associated QTL mapping for cotton yield based on a cDNA‐AFLP transcriptome map in an immortalized F2. Theor Appl Genet 123: 439–454 Lu QX (2005) Molecular Tagging, Location of QTLs for Fiber Properties and Differential Expression of Fiber Strength Relative Genes. A Dissertation for the Master Degree Nanjing Agricultural University, Supervisor: Zhang TZ, Nanjing, China. Mackay L, Powell W (2007) Methods for linkage disequilibrium mapping in crops. Trends Plant Sci 12: 57–63 Malysheva‐Otto LV, Ganal MW, Röder MS (2006) Analysis of molecular diversity, population structure and linkage disequilibrium in a worldwide survey of cultivated barley germplasm (Hordeum vulgare L.). BMC Genet 7: 6 Nei M, Tajima FA, Tateno Y (1983) Accuracy of estimated phylogenetic trees from molecular data. J Mol Evol 19: 153–170 Paterson AH, Brubaker C, Wendel JF (1993) A rapid method for extraction of cotton (Gossypium spp.) genomic DNA suitable for RFLP or PCR analysis. Plant Mol Biol Rep 11: 122–127 Paterson AH, Wendel JF, Gundlach H, Guo H, Jenkins J, Jin D, Llewellyn D, Showmaker KC, Shu S, Udall J, Yoo MJ, Byers R, Chen W, Doron‐ Faigenboim A, Duke MV, Gong L, Grimwood J, Grover C, Grupp K, Hu G, Lee TH, Li J, Lin L, Liu T, Marler BS, Page JT, Roberts AW, Romanel E, Sanders WS, Szadkowski E, Tan X, Tang H, Xu C, Wang J, Wang Z, Zhang D, Zhang L, Ashrafi H, Bedon F, Bowers JE, Brubaker CL, Chee PW, Das S, Gingle AR, Haigler CH, Harker D,

www.jipb.net

61

Hoffmann LV, Hovav R, Jones DC, Lemke C, Mansoor S, ur Rahman M, Rainville LN, Rambani A, Reddy UK, Rong JK, Saranga Y, Scheffler BE, Scheffler JA, Stelly DM, Triplett BA, Van Deynze A, Vaslin MF, Waghmare VN, Walford SA, Wright RJ, Zaki EA, Zhang T, Dennis ES, Mayer KF, Peterson DG, Rokhsar DS, Wang X, Schmutz J (2012) Repeated polyploidization of Gossypium genomes and the evolution of spinnable cotton fibres. Nature 492: 423–427 Pritchard JK, Wen W (2004) Documentation for STRUCTURE software, The University of Chicago Press, Chicago. Qian SY, Huang JQ, Peng YJ, Zhou BL, Ying MC, Shen DZ, Liu GL, Hu TX, Xu YJ, Gu LM, Ni WC, Chen S (1992) Studies on the hybrid of G. hirsutum L. and G. anomalum Wawr. & Peyr. and application in breeding. Sci Agric Sin 25: 44–51 (in Chinese with an English Abstract). Qin H, Guo W, Zhang YM, Zhang T (2008) QTL mapping of yield and fiber traits based on a four‐way cross population in Gossypium hirsutum L. Theor Appl Genet 117: 883–894 Qin YS, Ye WX, Liu RZ, Zhang TZ, Guo WZ (2009) QTL mapping for fiber quality properties in upland cotton (Gossypium hirsutum L.). Sci Agric Sin 42: 4145–4154 (in Chinese with an English Abstract). Rambaut A (2009) FigTree v1.2.3. Institute of Evolutionary Biology, University of Edinburgh, Scotland, UK. Rostoks N, Ramsay L, MacKenzie K, Cardle L, Bhat PR, Roose ML, Svensson JT, Stein N, Varshney RK, Marshall DF, Graner A, Close TJ, Waugh R (2006) Recent history of artificial outcrossing facilitates whole‐genome association mapping in elite inbred crop varieties. Proc Natl Acad Sci USA 103: 18656–18661 Sang ZQ (2003) QTL Mapping for Elife Fiber Quality Properties in Upland Cotton. A Dissertation for the Master Degree Nanjing Agricultural University, Supervisor: Guo WZ, Nanjing, China. Shen XL, Guo WZ, Zhu XF, Yuan YL, Yu JZ, Kohel RJ, Zhang TZ (2005) Molecular mapping of QTLs for qualities in three diverse lines in upland cotton using SSR markers. Mol Breed 15: 169–181 Shen XL, Guo WZ, Zhu XF, Yuan YL, Zhang TZ (2007) Genetic mapping of quantitative trait loci for fiber quality and yield trait by RIL approach in upland cotton. Euphytica 155: 371–380 Shen XL, Zhang TZ, Guo WZ, Zhu XF, Zhang XY (2006) Mapping QTLs with main effects, epistatic effects and QTL
environment interaction for fiber and yield traits in RILs of upland cotton. Crop Sci 46: 61–66 Stich B, Melchinger AE, Frisch M, Maurer HP, Heckenberger M, Reif JC (2005) Linkage disequilibrium in European elite maize germplasm investigated with SSRs. Theor Appl Genet 111: 723–730 Thornsberry JM, Goodman MM, Doebley J, Kresovich S, Nielsen D, Buckler ESIV (2001) Dwarf8 polymorphisms associate with variation in flowering time. Nat Genet 28: 286–289 Vanniarajan C, Vinod KK, Pereira A (2012) Molecular evaluation of genetic diversity and association studies in rice (Oryza sativa L.). J Genet 91: 9–19 Wang BH, Guo WZ, Zhu XF, Wu YT, Huang NT, Zhang TZ (2006) QTL mapping of fiber quality in an elite hybrid derived‐RILs in upland cotton. Euphytica 152: 367–378 Wang BH, Guo WZ, Zhu XF, Wu YT, Huang NT, Zhang TZ (2007a) QTL mapping of yield and yield components for elite hybrid derived‐RILs in upland cotton. J Genet Genomics 34: 35–45 Wang BH, Wu YT, Guo WZ, Zhu XF, Huang NT, Zhang TZ (2007c) Genetic dissection of heterosis for fiber qualities in an elite cotton hybrid grown in second‐generation. Crop Sci 47: 1384–1392 Wang BH, Wu YT, Huang NT, Guo WZ, Zhu XF, Zhang TZ (2007b) QTL analysis of epistatic effects on yield and yield component traits for elite hybrid derived‐RILs in upland cotton. Acta Agron Sin 33: 1755–1762 Wang C, Ulloa M, Mullens TR, Yu JZ, Roberts PA (2012) QTL analysis for transgressive resistance to root‐knot nematode in interspecific

January 2014 | Volume 56 | Issue 1 | 51–62

62

Cai et al. cotton (Gossypium spp.) progeny derived from susceptible parents. PLoS ONE 7: e34874

Wang J, Guo WZ, Zhang TZ (2007d) QTL mapping for fiber quality properties in cotton cultivar Yumian1. Acta Agron Sin 33: 1915–1921 (Chinese with Abstract in English). Wang K, Wang Z, Li F, Ye W, Wang J, Song G, Yue Z, Cong L, Shang H, Zhu S, Zou C, Li Q, Yuan Y, Lu C, Wei H, Gou C, Zheng Z, Yin Y, Zhang X, Liu K, Wang B, Song C, Shi N, Kohel RJ, Percy RG, Yu JZ, Zhu YX, Wang J, Yu S (2012) The draft genome of a diploid cotton Gossypium raimondii. Nat Genet 44: 1098–1103 Wang P, Su L, Qin L, Hu B, Guo W, Zhang T (2009a) Identification and molecular mapping of a Fusarium wilt resistant gene in upland cotton. Theor Appl Genet 119: 733–739 Wang YQ, Yang WH, Xu HX, Zhou DY, Feng XA, Kuang M (2009b) The main problems and recommendations in Chinese cotton production. Chin Agr Sci Bull 25: 86–90 (in Chinese with an English Abstract). Yang C, Guo WZ, Li GY, Gao F, Lin SS, Zhang TZ (2008) QTLs mapping for Verticillium wilt resistance at seedling and maturity stages for in G. barbadense L. Plant Sci 174: 290–298 Yu J, Pressoir G, Briggs WH, Vroh BI, Yamasaki M, Doebley JF, McMullen MD, Gaut BS, Nielsen DM, Holland JB, Kresovich S, Buckler ES (2006) A unified mixed‐model method for association mapping that accounts for multiple levels of relatedness. Nat Genet 38: 203–208 Zeng L, Meredith WR, Jr, Gutiérrez OA, Boykin DL (2009) Identification of associations between SSR markers and fiber traits in an exotic germplasm derived from multiple crosses among Gossypium tetraploid species. Theor Appl Genet 119: 93–103 Zhang J, Guo WZ, Zhang TZ (2002) Molecular linkage map of allotetraploid cotton (Gossypium hirsutum L.
Gossypium barbadense L.) with a haploid population. Theor Appl Genet 105: 1166–1174 Zhang J, Wu YT, Guo WZ, Zhang TZ (2000) Fast screening of microsatellite markers in cotton with PAGE/silver staining. Cotton Sci 12: 267–269 Zhang K, Zhang J, Ma J, Tang S, Liu D, Teng Z, Liu D, Zhang Z (2012) Genetic mapping and quantitative trait locus analysis of fiber quality traits using a three‐parent composite population in upland cotton (Gossypium hirsutum L.). Mol Breed 29: 335–348 Zhang T, Qian N, Zhu X, Chen H, Wang S, Mei H, Zhang Y (2013) Variations and transmission of QTL alleles for yield and fiber qualities in upland cotton cultivars developed in China. PLoS ONE 8: e57220

January 2014 | Volume 56 | Issue 1 | 51–62

Zhang T, Yuan Y, Yu J, Guo W, Kohel RJ (2003) Molecular tagging of a major QTL for fiber strength in Upland cotton and its marker‐ assisted selection. Theor Appl Genet 106: 262–268 Zhang ZS, Xiao YH, Luo M, Li XB, Luo XY, Hou L, Li DM, Pei Y (2005) Construction of a genetic linkage map and QTL analysis of fiber‐ related traits in upland cotton (Gossypium hirsutum L.) Euphytica 144: 91–99 Zhao K, Aranzana MJ, Kim S, Lister C, Shindo C, Tang C, Toomajian C, Zheng H, Dean C, Marjoram P, Nordborg M (2007) An Arabidopsis example of association mapping in structured samples. PLoS Genet 31: e4 Zhou BL, Zhu XF, Guo WZ, Zhang TZ (2006) Major gene plus poly‐gene inheritance of super quality fiber properties in upland cotton introgressed from Gossypium anomalum Wawra & Peyritsch. Cotton Sci 18: 60–62 Zhu YX, Li FG (2013) The Gossypium raimondii genome, a huge leap forward in cotton genomics. J Integr Plant Biol 55: 570–571

SUPPORTING INFORMATION Additional Supporting Information may be found in the online version of this article: Figure S1. Neighbor‐joining (N‐J) tree of 99 Gossypium hirsutum accessions (A) Dendrogram based on genetic distance. Each color represents a different accession origin: blue indicating from America, green from Northern (Specific early maturation), red from North Western inland, yellow from The Yellow River, brilliant rose from The Yangtze River, and black from Local lines. (B) Population stratification for K ¼ 2. Figure S2. The distribution of LD among 97 simple sequence repeat (SSR) loci in 99 Gossypium hirsutum accessions Table S1. Diversity information of the 97 simple sequence repeat (SSR) primer pairs Table S2. 196 marker‐trait association mapping using mixed linear model (MLM) and general linear models (GLM) Table S3. The distribution frequency of elite alleles in different ecological area Table S4. Basic information and elite allele number of 101 tested accessions

www.jipb.net