Plant Cell Rep DOI 10.1007/s00299-015-1776-y

ORIGINAL PAPER

Construction of a high-density linkage map and mapping quantitative trait loci for somatic embryogenesis using leaf petioles as explants in upland cotton (Gossypium hirsutum L.) Zhenzhen Xu • Chaojun Zhang • Xiaoyang Ge • Ni Wang • Kehai Zhou • Xiaojie Yang • Zhixia Wu • Xueyan Zhang • Chuanliang Liu • Zuoren Yang Changfeng Li • Kun Liu • Zhaoen Yang • Yuyuan Qian • Fuguang Li

•

Received: 11 December 2014 / Revised: 26 January 2015 / Accepted: 17 February 2015 Ó Springer-Verlag Berlin Heidelberg 2015

Abstract Key message The first high-density linkage map was constructed to identify quantitative trait loci (QTLs) for somatic embryogenesis (SE) in cotton (Gossypium hirsutum L.) using leaf petioles as explants. Abstract Cotton transformation is highly limited by only a few regenerable genotypes and the lack of understanding of the genetic and molecular basis of somatic embryogenesis (SE) in cotton (Gossypium hirsutum L.). To construct a more saturated linkage map and further identify quantitative trait loci (QTLs) for SE using leaf petioles as explants, a high embryogenesis frequency line (W10) from the commercial Chinese cotton cultivar CRI24 was crossed with TM-1, a genetic standard upland cotton with no embryogenesis frequency. The genetic map spanned 2300.41 cM in genetic distance and contained 411 polymorphic simple sequence repeat (SSR) loci. Of the 411 mapped loci, 25 were developed from unigenes identified Communicated by X. S. Zhang. Z. Xu, C. Zhang, and X. Ge contributed equally to this work.

Electronic supplementary material The online version of this article (doi:10.1007/s00299-015-1776-y) contains supplementary material, which is available to authorized users. Z. Xu
C. Zhang
X. Ge
N. Wang
K. Zhou
X. Yang
Z. Wu
X. Zhang
C. Liu
Z. Yang
C. Li
K. Liu
Z. Yang
Y. Qian
F. Li (&) State Key Laboratory of Cotton Biology, Institute of Cotton Research, Chinese Academy of Agricultural Sciences, Anyang 455000, China e-mail: [email protected] Z. Xu Institute of Industrial Crops, Jiangsu Academy of Agricultural Sciences, Nanjing 210014, China

for SE in our previous study. Six QTLs for SE were detected by composite interval mapping method, each explaining 6.88–37.07 % of the phenotypic variance. Single marker analysis was also performed to verify the reliability of QTLs detection, and the SSR markers NAU3325 and DPL0209 were detected by the two methods. Further studies on the relatively stable and anchoring QTLs/markers for SE in an advanced population of W10 9 TM-1 and other cross combinations with different SE abilities may shed light on the genetic and molecular mechanism of SE in cotton. Keywords Quantitative trait loci
Somatic embryogenesis
Genetic map
Simple sequence repeat
Cotton

Introduction Cotton (Gossypium spp.) is the most important source of natural commercial fiber of the world (Sunilkumar et al. 2006; Yang et al. 2014a, b). As such, cotton production plays a vital role in global agricultural economy. With steady world population increases, the demand for cotton has been increasing correspondingly. Thus, intense efforts are being made to improve broadening the range of attributes that can lead to improve yield and fiber quality. Over the years, conventional breeding has made many important contributions to cotton improvement. Nevertheless, far-reaching improvement has been limited by the narrow genetic diversity of cotton. To circumvent such constraint, genetic engineering is being used to broaden the cotton germplasm (Zeng et al. 2006). Agrobacterium-mediated transformation and particle gun bombardment are effective techniques for transferring genes to cotton, and

123

Plant Cell Rep

the former method has been widely used based on its excellent advantages, such as the clear principle and proven technique system (Wilkins et al. 2000). This has led to the development of herbicide- and pest-resistant cotton (Keller et al. 1997; ISAAA 2014). Somatic embryogenesis (SE) that converts transgenic cells to embryos is an effective tool for recovering transgenic cotton plants from Agrobacterium-mediated transformation (Jin et al. 2005; Li et al. 2006). Induction of embryogenic calli (EC) is the key step to achieve regeneration during cotton SE. Unfortunately, high-frequency cotton SE, especially EC induction, has been confined to limited varieties (Trolinder and Goodin 1987; Kumria et al. 2003; Zheng et al. 2014). This precludes using SE to recover transgenic plants from many desirable cotton lines which are recalcitrant to SE (Kumria et al. 2003). Apparently, addressing the recalcitrant SE problem will require elucidation of the genetic control of cotton SE competency. In most previous reports, explants used during cotton tissue culture were derived from hypocotyls, cotyledons, cotyledon nodes, and radicals of seedlings (Gawel and Robacker 1990a, b; Kumria et al. 2003; Mishra et al. 2003; Ganesan and Jayabalan 2004). This makes it difficult to provide enough explants from individuals for replicated tests and to track individual descendants when investigating the genetic basis of SE in cotton. This is largely because of the destruction of individuals during the tissue culture process. Fortunately, petioles from true leaves could provide possibilities for generation advancement and a practically unlimited supply of explants during cotton tissue culture (Gawel and Robacker 1990a, b). We previously used cotton leaf petioles to create 40 explants for studying SE inheritance on an individual plant basis from segregating populations (Zhang et al. 2011). Recently, molecular markers, especially DNA markers, have been used to construct genetic map to identify the number, significance, and location of quantitative trait loci (QTLs) associated with a variety of phenotypic characteristics (Kalia et al. 2011). Such DNA markers have been used to locate genes contributing to SE in other crops such as alfalfa (Yu and Pauls 1993), sunflower (Flores Berrios et al. 2000), soybean (Song et al. 2010), and rice (Nishimura et al. 2005; Li et al. 2013). Since the first allotetraploid cotton genetic map was constructed by Reinisch et al. (1994), molecular markers and QTL analysis have been employed to map genes or chromosomal regions for various phenotypic characteristics in cotton, including fiber quality and yield (Saeed et al. 2011; Wang et al. 2013; Yang et al. 2015). Molecular markers associated with SE during cotton tissue culture were only reported in our previous study (Zhang et al. 2011). Three QTLs underlying SE in cotton were detected on two linkage groups using simple

123

sequence repeat (SSR) marker analysis. These QTLs explained 8.40, 14.40, and 58.10 % of the phenotypic variation. However, chromosomal locations for these three QTLs were uncertain because of a low number of markers and unsaturated linkage map. Therefore, a more saturated linkage map is needed to further map QTLs underlying SE. Thus, the objectives of this study were: (1) to construct a high-density linkage map and (2) to further identify QTLs contributing to cotton SE.

Materials and methods Plant materials The mapping population was from a single-seed descent of a cross between W10 and TM-1. W10, an inbred line with high embryogenesis frequency, was selected from the commercial Chinese cotton cultivar CRI24. TM-1, a genetic standard upland cotton with a Deltapine 14 background, is a non-regenerable genotype. Leaf petiole explants of W10 had SE rate of [98 %, whereas explants of TM-1 did not undergo SE in our laboratory. The two parental lines were self-pollinated and selected based on SE generation potential using cotton leaf petioles on regeneration culture medium at the China Cotton Research Institute (CCRI), Anyang, Henan Province, China. This was performed for eight generations before making a hybrid combination. F1 hybrids were planted in Hainan Province in the winter of 2010 to produce F2 seeds by selfpollination. The parental lines and F2 plants were grown on the CCRI Experimental Farm in 2011. A total of 140 F2 plants were separately harvested, then the resulting F2:3 progenies were planted on the CCRI Experimental Farm in the summer of 2012. Tissue culture conditions Leaf petioles from two parents, F2, and F2:3 individuals were collected and sterilized as previously described (Zhang et al. 2011). The petioles were then transected into segments of 5 mm ([40 segments per plant) to serve as explants. The segments were cultured on the callus induction medium as previously described (Xu et al. 2013). After 40 days of tissue culture, calli were transferred onto the embryogenic callus (EC) induction medium according to Xu et al. (2013). In this study, all cultures were maintained at 28 °C and a 14/10 h (day/light) period rather than a 16/8 h (day/light) period as reported in our previous study (Zhang et al. 2011). The callus and EC induction media described by Xu et al. (2013) were optimized compared to our previous study (Zhang et al. 2011). Subsequently, calli were transferred and subcultured on to a

Plant Cell Rep

fresh medium every 50 days. Following three subcultures (about 8 months), SE ability was tested according to the EC rate of cultured explants per genotype. EC were assessed based on callus texture and color, especially embryos produced in that EC were friable, loose, primarily whitish, and could produce embryos. The percentage of EC induction was calculated as the number of EC-producing explants/total cultured explants (Zhang et al. 2011).

of EST-SSR primers were used to screen for polymorphisms between two parents. Then, the yielded polymorphic markers were used to genotype the F2 plants. PCR amplification and electrophoresis (10 % non-denatured polyacrylamide gel) were performed as described by Zhang et al. (2002).

EST-SSR marker development

A linkage map was constructed using JoinMap 3.0 with a logarithm of odds (LOD) threshold C3 and a maximal distance of 50 cM (Stam 1993). Based on the Kosambi mapping function, recombination frequencies were converted to genetic map distance (centiMorgen, cM). Linkage groups were assigned to chromosomes based on the genetic linkage map available from CMD and the published literature (Guo et al. 2007; Yu et al. 2012; Wang et al. 2013). QTLs associated with EC rates of petioles were detected using WinQTLCart 2.5 software by the composite interval mapping (CIM) method in the F2 and F2:3 populations. A QTL was considered significant if the corresponding LOD score was greater than 2.5. The linkage map and QTLs were presented using MapChart 2.2 software (Voorrips 2002). The QTL nomenclature follows that described by Zhang et al. (2011). The designation begins with ‘q’, followed by an abbreviation of the trait name, the name of the chromosome or linkage, and its serial number. To further verify the reliability of QTL detection, single marker analysis (SMA) was performed using Icimapping 3.0 software (Li et al. 2008).

To identify new SSR markers, 31,286 unigenes longer than 500 bp previously identified during cotton SE in different sister lines (Xu et al. 2013) were compared with the sequences in cotton marker database (CMD) (http://www. cottonmarker.org) using BLASTn (E-value cutoff 1e-06). Perl scripts based on MISA-MIcroSAtellite identification tool software package files (http://pgrc.ipk-gatersleben.de/ misa/) were written to identify SSRs, extract sequences, and design primers. SSR motifs, with over six repeat units in dinucleotides, five in trinucleotides, four in tetranucleotides, pentanucleotides, and hexanucleotides, and three in heptanucleotides and octanucleotides were used as the search criteria. If an SSR-containing EST sequence had sufficient length, then 200 bp sequences on both sides of the SSR repeat were extracted as candidates for SSR primer designing. Otherwise, all available sequences on both sides of the SSR were selected. The following parameters were used to design the SSR primers: primer length 20–25 bp, with 20 bp as the optimum; primer GC % = 40–65 %, with an optimum value of 50 %; primer Tm 55–63 °C, and product size range 100–500 bp. Other non-specified parameters were at the default settings. All the new SSR markers were filtered by electronic PCR (http://www.ncbi.nlm.nih.gov/tools/epcr/). Electronic PCRs were performed using the newly designed SSR primers as primers and the CMD sequences as templates. In parallel, we downloaded all the 752 ESTs for cotton SE including those reported by Zeng et al. (2006) from the National Center For Biotechnology Information (NCBI) database and compared them with the sequences in CMD and 31,286 unigenes used in this study using BLASTn (Evalue cutoff 1e-06). As a result, a total of 1023 pairs of new SSR primers were designed from this current study. SSR analysis Genomic DNA from the mapping parents and 140 F2 individuals were isolated as described by Paterson et al. (1993). A total of 18,187 pairs of SSR primers, including BNL, CIR, CM, DPL, GH, HAU, JESPR, MGHES, MUCS, MUSS, MON, NAU, PGML, TMB, DOW, and NBRI, were downloaded from CMD, and 1023 new pairs

Data and QTL analysis

Results Characterization of SSR sequences from unigenes associated with cotton somatic embryogenesis After comparing 31,286 unigenes with the sequences in CMD, 26,729 new unigenes were detected to identify SSR and design primers. A total of 1092 pairs of new SSR primers were identified and 1023 pairs (Table S1) were generated after filtering by electronic PCR and detailed manual inspection. Furthermore, the 1023 new SSR markers contained 1048 putative different types of SSR motifs. The EST-SSR repeat types are summarized in Table 1 and their motif types are shown in Fig. S1. Among these 1048 putative SSRs, the two most abundant types were dinucleotides (47.61 %) and trinucleotides (30.25 %) (Table 1). Other abundant SSRs such as tetranucleotide repeats (13.07 %) were also identified (Table 1). The remaining types, such as pentanucleotide, hexanucleotide, heptanucleotide, and octanucleotide repeats, accounted for a small percentage of all SSRs (Table 1). The AT/TA motif

123

Plant Cell Rep Table 1 Features of SSR markers identified in new EST-SSRs Motif length

No. of EST-SSRs

Frequency (%)

Dinucleotide repeats

499

47.61

Trinucleotide repeats

317

30.25

Tetranucleotide repeats

137

13.07

Pentanucleotide repeats

37

3.53

Hexanucleotide repeats

17

1.62

Heptanucleotide repeats

33

3.15

Octanucleotide repeats

8

0.76

(27.86 %) was the most frequent dinucleotide type, followed by AG/TC (7.06 %), CT/GA (6.49 %), and AC/TG (3.44 %) (Fig. S1). The dominant motif among trinucleotide SSRs was AAG/TTC (3.91 %), followed by CTT/ GAA (3.05 %) and AAT/TTA (1.91 %) (Fig. S1). Other important SSR motif types, including CA/GT (2.77 %) and AAAT/TTTA (2.10 %) were also detected but as a small proportion of the total (Fig. S1). To take advantage of the publicly released ESTs associated with cotton SE for EST-SSR marker development, 752 ESTs for cotton SE from the NCBI database were also used. However, only 39 new ESTs longer than 500 bp were detected, and no new EST-SSR marker was developed.

the At and Dt subgenomes was 1207.38 and 1093.03 cM, with 225 and 186 loci, respectively (Table 2). The average genetic distance between two adjacent markers on the At and Dt subgenomes was 5.37 and 5.88 cM, respectively (Table 2). The At subgenome was longer than the Dt subgenome and had more polymorphic markers assigned. Segregation distortion analysis of the mapped SSR markers Of the 411 mapped loci, 117 (28.47 %) showed segregation distortion (SD), with 65 loci favoring TM-1 alleles and 52 favoring W10 alleles (Table 2). The SD loci were unevenly distributed on the 24 chromosomes and four linkage groups with 0–13 loci (Fig. 1; Table 2). Sixty-two SD loci were located on the At subgenome and 55 on the Dt subgenome (Table 2). Most SD loci were located on Chr 8, Chr 5, Chr 2, Chr 14, Chr 16, and Chr 20 with 13, 12, 8, 6, 6, and 6 SD loci, respectively. In contrast, no SD loci were found on Chr 1, Chr 4, Chr 5, Chr 7, Chr 10, Chr 11, Chr 13, Chr 16, Chr 20, and LG4. The SD loci exhibited a phenomenon in which loci skewed toward the same species alleles on the same chromosome, e.g., all four loci skewed toward the TM-1 alleles on Chr1. QTLs mapped for somatic embryogenesis

Construction of molecular linkage map and mapping of new EST-SSR markers In the present study, 19,210 pairs of SSR primers were used to screen polymorphisms between W10 and TM-1, and this resulted in 441 (2.30 %) pairs of polymorphic primers. Of the 1023 pairs of new EST-SSR primers designed in this study, 28 (2.74 %) pairs exhibited polymorphism. All of the 441 pairs of polymorphic primers were used to genotype the F2 population of 140 individuals. The linkage test (LOD [ 3) mapped the 411 polymorphic loci onto 42 linkage groups with an average genetic distance of 5.60 cM between two neighboring markers, and the linkage map spanned a total of 2300.41 cM in genetic distance (Fig. 1), covering approximately 45.72 % of the tetraploid cotton genome (Stelly 1993). Among these 411 mapped SSR loci, 25 were from 1023 newly designed markers and distributed on 11 chromosomes, i.e., Chr2, Chr5, Chr8, Chr9, Chr11, Chr12, Chr15, Chr19, Chr20, Chr24, and Chr25, each with 1–4 loci (Fig. 1). Furthermore, of these 42 linkage groups, 38 were assigned to 24 chromosomes based on the previously identified chromosome-anchored SSR markers, each with 2–44 loci (Fig. 1), and the remaining four linkage groups were unable to be associated with chromosomes and temporarily named as LG.U1, LG.U2, LG.U3, and LG.U4 (Fig. 1). The total genetic distance of

123

We used the EC rate of cultured explants per genotype in the F2 and F2:3 populations to test cotton SE ability as described by Zhang et al. (2011). Six QTLs for EC were detected by the CIM method and designated as qEC-C1-1, qEC-C5-1, qEC-C9-1, qEC-C11-1, qEC-C12-1, and qECC18-1 on six chromosomes (Chr 1, Chr 5, Chr 9, Chr 11, Chr 12, and Chr 18), explaining 6.88–37.07 % of the phenotypic variance of EC (Fig. 1; Table 3). The phenotype variation explained by qEC-C12-1 and qEC-C5-1 were the highest, 37.07 and 12.31 %, respectively, and qEC-C12-1 had a high dominance genetic effect (0.465) (Table 3). Among these six QTLs, qEC-C1-1 and qECC11-1 were detected in the F2 population, whereas qECC5-1, qEC-C9-1, qEC-C12-1, and qEC-C18-1 were detected in the F2:3 population (Table 3). In addition, the closest marker to qEC-C1-1, qEC-C5-1, qEC-C9-1, qECC11-1, qEC-C12-1, and qEC-C18-1 was NAU5163, NAU3325, MON_DPL0530, DPL0209, NBRI1622, and NAU748, respectively (Fig. 1; Table 3). Overall, the desirable alleles of these six QTLs came from different parents, in that qEC-C1-1, qEC-C11-1, and qEC-C18-1 were from W10, while qEC-C5-1, qEC-C9-1, and qEC-C12-1 were from TM-1 (Table 3). To verify the reliability of QTLs detection, the SMA method was applied to identify markers associated with EC. Three markers for EC were detected: NAU3325,

Plant Cell Rep

Fig. 1 Linkage map and location of QTLs. Deviating loci are shown in italics

CGR6829, and DPL0209, respectively (Table 4). The NAU3325 and CGR6829 markers were both mapped on chromosome 5, explaining 9.70 and 8.79 % of the phenotype variance of EC in the F2:3 population, respectively, and the CGR6829 marker displayed a high dominance genetic effect ([50 %) (Table 4). The DPL0209 marker was mapped on chromosome 11 and accounted for 8.68 %

of the phenotypic variation of EC in the F2 population (Table 4). In addition, the desirable alleles of these three markers also came from different parents, in that NAU3325 was from TM-1, while CGR6829 and DPL0209 were from W10 (Table 4). The SSR markers NAU3325 and DPL0209 could be detected by both the methods (CIM and SMA).

123

Plant Cell Rep

Fig. 1 continued

Discussion Genetic engineering is a powerful tool to broaden the narrow germplasm in upland cotton. Somatic embryogenesis can generate a large number of individual transformants, making it an ideal tool for recovery of cotton transformants (Kumria et al. 2003; Yang et al. 2014a). Unfortunately, only certain

123

cotton genotypes respond to SE and many desired genotypes do not. Identifying genes linked to cotton SE could provide insight into genetic control of SE, and this will help to develop strategies to overcome the regeneration difficulty of the recalcitrant genotypes. A high-density genetic map is an important tool for identifying QTLs associated with the main agronomic traits

Plant Cell Rep

Fig. 1 continued

in cotton. SSR markers are valuable in the construction of genetic map. This is because of their useful properties such as co-dominance and high levels of polymorphism (Kalia et al. 2011). In recent years, many SSRs from different cotton genomes and tissues have been developed and used (Xiao et al. 2009; Kumpatla et al. 2009; Jena et al. 2012). SE-related ESTs/unigenes in cotton are very specific and

useful resources for identifying QTLs for SE; therefore, we firstly developed EST-SSR markers from ESTs including those reported by Zeng et al. (2006) and unigenes associated with cotton SE (Xu et al. 2013) to construct a more saturated linkage map and further identify QTLs for cotton SE in this study. As a result, 1023 new SSR markers containing 1048 putative types of SSR motifs were

123

Plant Cell Rep

Fig. 1 continued

generated. The findings in frequencies of different motifs in newly identified EST-SSRs are similar to those in G. arboretum L. and G. barbadense L. (Han et al. 2004; Zhang et al. 2007), but not with wheat, rice, and barley (Gao et al. 2004; Varshney et al. 2005). This difference may be due to fundamental differences between dicotyledons and monocotyledons (Frelichowski et al. 2006). Of the 19,210 SSR markers used to screen for polymorphisms between two parents, 2.30 % were informative, whereas in the 1023 new EST-SSR markers it was 2.74 %. However, only 28 of the newly developed 1023 EST-SSR markers exhibited polymorphisms, and 25 of them were mapped onto 11 linkage groups with a number of 1–4. No significant marker associated with cotton SE was identified in this study. However, these new EST-SSR markers and the linkage map have provided a good foundation for further studying of QTLs associated with cotton SE. Six QTLs for EC were identified by CIM analysis, and three markers were detected by SMA in this study (Tables 3, 4). A distribution analysis of these QTLs revealed that the majority were mapped to the At subgenome, except qEC-C18-1, which was mapped to Dt subgenome (Table 2). This suggests that the EC genes and QTLs may be mostly located at the At subgenome, and this result is similar to that reported by Zhang et al. (2011). The G. raimondii L. (D5; 2n = 26) and G. arboretum L. (A2; 2n = 26) genome, putative contributors of the Dt- and At- subgenomes of cotton species, have been sequenced and assembled recently (Wang et al. 2012; Li et al. 2014). This will facilitate evolutionary

123

analysis of SE-related genes in tetraploid cotton from their diploid ancestors. Although we previously identified three QTLs for SE on two linkage groups using SSR marker analysis, their chromosome locations remained undetermined and the genetic distance of their marker intervals was relatively large due to much smaller number of markers used (Zhang et al. 2011). Therefore, a comparison between the two studies could not be performed. However, in this study, we constructed a more saturated linkage map to further identify QTLs associated with cotton SE. Further studies on relatively stable and anchoring QTLs/markers for SE in an advanced population of W10 9 TM-1 and other cross combinations with different SE abilities may shed light on the genetic mechanism of SE in cotton. Somatic embryogenesis potential not only varies among cotton species, but also among genotypes within a species (Pradeep et al. 2010; Xu et al. 2013). In this study, W10 and TM-1 were continually self-pollinated for highly homogeneous genotypes in their SE potential before creating a new F2 segregation population; especially, TM-1 was selected for no SE ability, whereas TM-1 came directly from the commercial cultivars (Zhang et al. 2011). It also may not be a surprise if results from different studies varied due to conditions used in tissue culture and regeneration. To better evaluate genes/QTLs underlying cotton SE, the tissue culture time was extended to about 8 months in this study, while the previous study only spanned approximately 6 months (Zhang et al. 2011). Furthermore, the tissue culture media used in this study according to Xu

Plant Cell Rep Table 2 Genetic distance, number of marker loci, SD loci, and QTL distribution among chromosomes and linkage groups

Chr. or LG

No. of loci

Length

Average interval

No. of SDs

No. of QTLs

Chr1-1

7

64.31

9.19

4

1

Chr1-2

2

8.16

4.08

0

0

Chr2-1

21

200.07

9.53

8

0

Chr2-2

2

9.15

4.57

2

0

Chr4-1

5

45.13

9.03

2

0

Chr4-2

4

15.81

3.95

0

0

Chr5-1

29

122.99

4.24

12

1

Chr5-2

6

40.85

6.81

0

0

Chr7-1

10

69.50

6.95

3

0

Chr7-2

4

24.05

6.01

0

0

Chr8

44

103.52

2.35

13

0

Chr9-1

13

61.66

4.74

3

0

Chr9-2

17

63.05

3.71

4

1

Chr10-1

4

23.86

5.96

0

0

Chr10-2 Chr11-1

2 12

7.91 69.17

3.96 5.76

2 3

0 1

Chr11-2

4

16.59

4.15

0

0

Chr12-1

12

91.86

7.66

3

1

Chr12-2

15

130.15

8.68

3

0

Chr13

12

39.61

3.30

0

0

225

1207.38

5.37

62

5

Chr14-1

18

71.54

3.97

6

0

Chr14-2

5

16.80

3.36

1

0

15

85.16

5.68

4

0

Chr16-1

6

53.84

8.97

6

0

Chr16-2

2

1.48

0.74

0

0

At subgenome

Chr15

Chr16-3

2

0.39

0.19

0

0

Chr17

2

23.49

11.74

2

0

Chr18

6

84.43

14.07

3

1

Chr19

11

105.01

9.55

5

0

Chr20-1 Chr20-2

24 4

101.78 7.12

4.24 1.78

6 0

0 0

Chr21

20

43.67

2.18

1

0

Chr22

8

92.41

11.55

2

0

Chr23

14

99.35

7.10

1

0

Chr24-1

4

26.70

6.67

2

0

Chr24-2

7

45.64

6.52

4

0

Chr25

25

93.88

3.76

4

0

Chr26

2

19.72

9.86

1

0

175

1093.03

5.88

48

1

LG1

4

48.64

12.16

3

0

LG2

2

8.00

4.00

2

0

LG3

3

53.24

17.75

2

0

LG4

2

10.78

5.39

0

0

Total

411

2300.41

5.60

117

6

Dt subgenome

et al. (2013) were optimized compared to our previous study (Zhang et al. 2011), and the light regime was also different from our previous study. However, it is a surprise

that no common QTLs identified in this study could be detected in both F2 and F2:3 generations. The small mapping population (140) used in the current study may have

123

Plant Cell Rep Table 3 QTLs detected for the EC trait in the F2 and F2:3 populations using composite interval mapping

QTL

Generation

Chr.

Nearest marker

LOD

Additive

Dominant

R2 (%)

Direction

qEC-C1-1

F2

1

NAU5163

2.84

0.056

-0.105

8.00

W10

qEC-C5-1

F2:3

5

NAU3325

4.39

-0.111

0.016

12.31

TM-1

qEC-C9-1

F2:3

9

MON_DPL0530

2.72

-0.087

0.049

7.32

TM-1

qEC-C11-1

F2

11

DPL0209

2.86

0.001

0.117

7.98

W10

qEC-C12-1

F2:3

12

NBRI1622

2.75

-0.073

0.465

37.07

TM-1

qEC-C18-1

F2:3

18

NAU748

2.53

0.075

0.062

6.88

W10

Table 4 Markers detected for the EC trait in the F2 and F2:3 populations using single marker analysis Marker name

Generation

Chr.

LOD

Additive

Dominant

R2 (%)

Direction

NAU3325

F2:3

5

3.06

-0.097

0.011

9.70

TM-1

CGR6829

F2:3

5

2.76

0.048

0.746

8.79

W10

DPL0209

F2

11

2.76

0.002

0.120

8.68

W10

led to the inconsistency of QTL identification between the two generations. Large mapping populations including permanent populations (such as recombinant inbred lines) should be developed with more markers used toward the identification and cloning of QTLs associated with SE in cotton. Author contribution statement F. L. conceived the work and revised the manuscript. Z. X. designed the research, performed the experiments, and drafted the manuscript. C. Z. and X. G. helped design the work, perform the experiments, and revise the manuscript. N. W. and K. Z. helped analyze the data. X.Y., Z. W., X. Z., and C. L. helped perform the analysis with constructive discussions. Z. Y., C. L., K. L., Z. Y., and Y. Q. helped perform the experiments. Acknowledgments We wish to thank Dr.Chee-kok Chin from Rutgers University, Jinfa Zhang from New Mexico State University, and Jiahe Wu from the Chinese Academy of Sciences for writing assistance and critical suggestions. This work was supported by the National Science Fund for Distinguished Young Scholars (31125020) and the Innovation Scientists and Technicians Troop Construction Projects of Henan Province. Conflict of interest of interest.

The authors declare that they have no conflict

References Flores Berrios E, Sarrafi A, Fabre F, Alibert G, Gentzbittel L (2000) Genotypic variation and chromosomal location of QTLs for somatic embryogenesis revealed by epidermal layers culture of recombinant inbred lines in the sunflower (Helianthus annuus L.). Theor Appl Genet 101:1307–1312 Frelichowski JE, Palmer MB, Main D, Tomkins JP, Cantrell RG, Stelly DM, Yu JZ, Kohel RJ, Ulloa M (2006) Cotton genome mapping with now microsatellites from Acala‘Maxxa’BACends. Mol Genet Genomics 275:479–491

123

Ganesan M, Jayabalan N (2004) Evaluation of haemoglobin (erythrogen) for improved somatic embryogenesis and plant regeneration in cotton (Gossypium hirsutum L. cv. SVPR 2). Plant Cell Rep 23:181–187 Gao LF, Jing RL, Huo NX, Li Y, Li XP, Zhou RH, Chang XP, Tang JF, Ma ZY, Jia JZ (2004) One hundred and one new microsatellite loci derived from EST (EST-SSRs) in bread wheat. Theor Appl Genet 108:1392–1400 Gawel NJ, Robacker CD (1990a) Genetic control of somatic embryogenesis in cotton petiole callus cultures. Euphytica 49:249–253 Gawel NJ, Robacker CD (1990b) Somatic embryogenesis in two Gossypium hirsutum genotypes on semi-solid versus liquid proliferation media. Plant Cell Tiss Org Cult 23:201–204 Guo WZ, Cai CP, Wang CB, Han ZG, Song XL, Wang K, Niu XW, Lu KY, Shi B, Zhang TZ (2007) A microsatellite-based, generich linkage map reveals genome structure, function and evolution in Gossypium. Genetics 176:527–541 Han ZG, Guo WZ, Song XL, Zhang TZ (2004) Genetic mapping of EST-derived microsatellites from the diploid Gossypium arboretum in allotetraploid cotton. Mol Genet Genomics 272:308–327 ISAAA (2014) Global Status of Commercialized Biotech/GM Crops in 2013. Pocket K No. 16, Ithaca, NY, July 2014. http://www. isaaa.org/resources/publications/pocketk/16/ Jena SN, Srivastava A, Rai KM, Ranjan A, Singh SK, Nisar T, Srivastava M, Bag SK, Mantri S, Asif MH, Yadav HK, Tuli R, Sawant SV (2012) Development and characterization of genomic and expressed SSRs for levant cotton (Gossypium herbaceum L.). Theor Appl Genet 124:565–576 Jin SX, Zhang XL, Liang SG, Nie YC, Guo YP, Huang C (2005) Factors affecting transformation efficiency of embryogenic callus of upland cotton (Gossypium hirsutum L.) with Agrobacterium tumefaciens. Plant Cell Tiss Org Cult 81:229–237 Kalia RK, Rai MK, Kalia S, Singh R, Dhawan AK (2011) Microsatellite markers: an overview of the recent progress in plants. Euphytica 117:309–334 Keller G, Spatola L, McCabe D, Martinell B, Swain W, John ME (1997) Transgenic cotton resistant to herbicide bialaphos. Transgenic Res 6:385–392 Kumpatla SP, Shah MR, Mukhopadhyay S, Ren R, Thompson SA, Greene TW (2009) High-throughput development of SSR and SNP markers in plants by parallel implementation of multiple in vitro and in silico methods. In Plant and Animal Genome XVII Conference, San Diego, CA

Plant Cell Rep Kumria R, Sunnichan VG, Das DK, Gupta SK, Reddy VS, Bhatnaga RK, Leelavath S (2003) High-frequency somatic embryo production and maturation into normal plants in cotton (Gossypium hirsutum L.) through metabolic stress. Plant Cell Rep 21:635–639 Li ZT, Dhekeny S, Dutt M, van Anan M, Tattersll J, Kelley KT, Gray DJ (2006) Optimizing Agrobacterium-mediated transformation of grapevine. In Vitro Cell Dev Biol-Plant 42:220–227 Li HH, Ribaut JM, Li ZL, Wang JK (2008) Inclusive composite interval mapping (ICIM) for digenic epistasis of quantitative traits in biparental populations. Theor Appl Genet 116:243–260 Li H, Yang YC, Duan YB, Li J, Cong XH, Ni DH, Song FS, Li L, Wei PC, Yang JB (2013) Mapping QTLs for the tissue culture performance of rice mature embryo using indica-japonica recombinant inbred lines. AJCS 7:440–445 Li FG, Fan GY, Wang KB, Sun FM, Yuan YL, Song GL, Li Q, Ma ZY, Lu CR, Zou CS, Chen WB, Liang XM, Shang HH, Liu WQ, Shi CC, Xiao GH, Gou CY, Ye WW, Xu X, Zhang XY, Wei HL, Li ZF, Zhang GY, Wang JY, Liu Kohel RJ, Percy RG, Yu JZ, Zhu YX, Wang J, Yu SX (2014) Genome sequence of the cultivated cotton Gossypium arboretum. Nat Genet 46:567–572 Mishra R, Wang HY, Yadav N, Wilkins TA (2003) Development of highly regenerable elite Acala cotton (Gossypium hirsutum L.)— a step towards genotype-independent regeneration. Plant Cell Tiss Org Cult 73:21–35 Nishimura A, Ashikari M, Lin S, Takashi T, Angeles ER, Yamamoto T, Matsuoka M (2005) Isolation of a rice regeneration quantitative trait loci gene and its application to transformation systems. P Natl Acad Sci USA 102:11940–11944 Paterson AH, Brubaker CL, 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 Pradeep CD, Anand PT, Mary T, Rob H, Doug B (2010) Factors affecting somatic embryogenesis and transformation in modern plant breeding. South Pacific J Nat Appl Sci 28:27–40 Reinisch AJ, Dong JM, Brubaker CL, Stelly DM, Wendel JF, Paterson AH (1994) A detailed RFLP map of cotton, Gossypium hirsutum x Gossypium barbadense: chromosome organization and evolution in a disomic polyploidy genome. Genetics 138:829–847 Saeed M, Guo WZ, Ullah I, Tabbasam N, Zafar Y, ur-Rahman M, Zhang TZ (2011) QTL mapping for physiology, yield and plant architecture traits in cotton (Gossypium hirsutum L.) grown under well-watered versus water-stress conditions. Electron J Biotechn 14 ISSN: 0717-3458 Song XH, Han YP, Teng WL, Sun GL, Li WB (2010) Identification of QTL underlying somatic embryogenesis capacity of immature embryos in soybean (Glycine max (L.) Merr.). Plant Cell Rep 29:125–131 Stam P (1993) Construction of integrated genetic linkage maps by means of a new computer package: JoinMap. Plant J 3:739–744 Stelly DM (1993) Interfacing cytogenetics with the cotton genome mapping effort. Pro Beltwide Cotton Conf 3:1545–1550 Sunilkumar G, Campbell LM, Puckhaber L, Stipanovic RD, Rathore KS (2006) Engineering cotton seed for use in human nutrition by tissue-specific reduction of toxic gossypol. P Natl Acad Sci USA 103:18054–18059 Trolinder NL, Goodin JR (1987) Somatic embryogenesis and plant regeneration in cotton (Gossypium hirsutum L.). Plant Cell Rep 6:231–234 Varshney RK, Graner A, Sorrells ME (2005) Genic microsatellite markers in plants: features and applications. Trends Biotechnol 23:48–55 Voorrips R (2002) MapChart: software for the graphical presentation of linkage maps and QTLs. J Hered 93:77–78

Wang KB, Wang ZW, Li FG, Ye WW, Wang JY, Song GL, Yue Z, Cong L, Shang HH, Zhu SL, Zou CS, Li Q, Yuan YL, Lu CR, Wei HL, Gou CY, Zheng ZQ, Yin Y, Zhang XY, Liu K, Wang B, Song C, Shi N, Kohel RJ, Percy RG, Yu JZ, Zhu YX, Wang J, Yu SX (2012) The draft genome of a diploid cotton Gossypium raimondii. Nat Genet 44:1098–1103 Wang FR, Xu ZZ, Sun R, Gong YC, Liu GD, Zhang JX, Wang LM, Zhang CY, Fan SJ, Zhang J (2013) Genetic dissection of the introgressive genomic components from Gossypium barbadense L. that contribute to improved fiber quality in Gossypium hirsutum L. Mol Breed 32:547–562 Wilkins T, Rajasekaran K, Anderson DM (2000) Cotton biotechnology. Crit Rev Plant Sci 19:511–550 Xiao J, Wu K, Fang DD, Stelly DM, Yu J, Cantrell RG (2009) New SSR markers for use in cotton (Gossypium spp.) improvement. J Cotton Sci 13:75–157 Xu ZZ, Zhang CJ, Zhang XY, Liu CL, Wu ZX, Yang ZR, Zhou KH, Yang XJ, Li FG (2013) Transcriptome profiling reveals auxin and cytokinin regulating somatic embryogenesis in different sister lines of cotton cultivar CCRI24. J Integr Plant Biol 55:631–642 Yang ZR, Li CF, Wang Y, Zhang CJ, Wu ZX, Zhang XY, Liu CL, Li FG (2014a) GhAGL15 s, preferentially expressed during somatic embryogenesis, promote embryogenic callus formation in cotton (Gossypium hirsutum L.). Mol Genet Genomics 289:873–883 Yang ZR, Zhang CJ, Yang XJ, Liu K, Wu ZX, Zhang XY, Zheng W, Xun QQ, Liu CL, Lu LL, Yang ZE, Qian YY, Xu ZZ, Li CF, Li J, Li FG (2014b) PAG1, a cotton brassinosteroid catabolism gene, modulates fiber elongation. New Phytol 203:437–448 Yang XL, Zhou XD, Wang XF, Li ZK, Zhang Y, Liu HW, Wu LQ, Zhang GY, Yan GJ, Ma ZY (2015) Mapping QTL for cotton fiber quality traits using simple sequence repeat markers, conserved intron-scanning primers, and transcript-derived fragments. Euphytica 201:215–230 Yu KF, Pauls KP (1993) Segregation of random amplified polymorphic DNA markers and strategies for molecular mapping in tetraploid alfalfa. Genome 36:844–851 Yu JZ, Kohel RJ, Fang DD, Cho J, Van Deynze A, Ulloa M, Hoffman SM, Pepper AE, Stelly DM, Jenkins JN, Saha S, Kumpatla SP, Shah MR, Hugie WV, Percy RG (2012) A high-density simple sequence repeat and single nucleotide polymorphism genetic map of the tetraploid cotton genome. G3-Genes Genom Genet 2:43–58 Zeng FC, Zhang XL, Zhu LF, Tu LL, Guo XP, Nie YC (2006) Isolation and characterization of genes associated to cotton somatic embryogenesis by suppression subtractive hybridization and macroarray. Plant Mol Biol 60:167–183 Zhang J, Guo WZ, Zhang TZ (2002) Molecular linkage map of allotetraploid cotton (Gossypium hirsutum L. 9 Gossypium barbadense L.) with a haploid population. Theor Appl Genet 105:1166–1174 Zhang YX, Lin ZX, Li W, Tu LL, Nie YC, Zhang XL (2007) Studies of new EST-SSRs derived from Gossypium barbadense. Chinese Sci Bull 52:2522–2531 Zhang CJ, Yu SX, Fan SL, Zhang JF, Li FG (2011) Inheritance of somatic embryogenesis using leaf petioles as explants in upland cotton. Euphytica 181:55–63 Zheng W, Zhang XY, Yang ZR, Wu JH, Li FL, Duan LL, Liu CL, Lu LL, Zhang CJ, Li FG (2014) AtWuschel promotes formation of the embryogenic callus in Gossypium hirsutum. PLoS One 9:e87502

123