Identification of major QTLS for fiber quality traits in cotton using interspecific backcross mapping population

Introduction

Cotton is an important natural fibre of global importance and generates high employment as it moves from cultivating field to textile mills to consumer. This crop has significant contributions in Indian economy by earning > 30 per cent of foreign exchange. Internationally, India ranks first in cotton area but occupies second position in production, next to China. Because of competition with synthetic fibres, textile manufacturers have adopted more efficient production machinery capable of generating more products per unit of time. Thus, the widespread use of high speed spinning technology has increased the demand for raw cotton fibre with higher strength. Besides fibre strength, length (2.5 per cent span length) and fineness (micronaire) are the other primary qualities that influence textile processing (Kohel, 1999) [1] . Thus, a long term challenge faced by cotton breeders is the simultaneous improvement of yield and fibre quality. Although conventional breeding methods have contributed much to the development of cotton cultivars, simultaneous improvement of fibre quality and yield is still difficult (Lin et al., 2005) [2] . Often, these attempts had resulted in poor agronomic qualities of the progeny, distorted segregation, sterility, motes formation and limited recombination (Reinisch et al., 1994) [3] . Recent developments in quantitative trait loci (QTL) mapping and marker assisted selection (MAS) offers new avenues to overcome the above said limitations. The identification of the stable QTLs affecting fibre traits across the genetic backgrounds would be more useful for efficient MAS. The development of DNA markers linked to the fibre quality QTLs would allow cotton breeders to trace this very important trait in early plant growing stages or early segregating generations (Boopathi et al., 2011) [4] . Among the DNA markers, advent of simple sequence repeat (SSR) or microsatellite markers have brought a new, user friendly and highly polymorphic class of genetic markers for cotton. SSRs are simple, usually codominant, well dispersed throughout the genome and easily shared between labs via flanking primer sequence. Reddy et al. (2001) [5] suggested that the total pool of SSRs present in the cotton genome is sufficiently abundant to satisfy the requirements of extensive genome mapping (~5000 cM), QTL analysis and MAS for the development of highly productive G. hirsutum cultivars with superior fibre quality and can facilitate efficient comparative QTL mapping. The first and foremost requirement in QTL analysis is development of mapping population to study the segregation of molecular markers and trait of interest. Permanent mapping populations such as recombinant inbred lines (RILs) permit multi-location experiments and thus they are expected to be superior over F 2 population with regard to the power of QTLs detection. But those QTLs with large overdominant effects would not be detected in RIL population because RIL express only additive effects. Further, genetics of F 2 progenies reveal all the allelic combinations and inter and intra genic interactions. In addition, development of RILs is a cost involving and time consuming process (Lacape et al., 2005) [6] . Consequently, F 2 and backcross (BC) progenies are the most preferred populations since they are easy to construct, require a short time for development and can be constructed with minimum efforts as compared to other populations (Reiter et al., 1992) [7] . The upland cotton (G. hirsutum) is the most extensively grown among the four cultivated Gossypium species. However, as revealed by all means of assessment, the level of genetic diversity is low in G. hirsutum especially among agriculturally elite types (Gutierrez et al., 2002) [8] . Existence of low polymorphism in the commonly cultivated and locally well adapted cultivars hinders the linkage map construction and QTL analysis of fibre quality traits (Paterson et al., 2003) [9] . Hence, use of interspecific cotton population offers significant level of polymorphism for efficient QTL mapping despite the fact that regional breeding program involves locally adapted intraspecific crosses. Identification of molecular markers linked to QTLs for fibre quality traits have already been reported using interspecific populations of G. hirsutum and G. barbadense (Lacape et al., 2005 [6] , Paterson et al., 2003 [9] , Lacape et al., 2003 [10] , Collard et al., 2005 [11] , Zhang and Stewart, 2000 [12] , Iwata and Ninomiya, 2006 [13] ). Ideally, QTL mapping studies should be independently confirmed due to several factors that influence the QTL detection and validated in the local elite lines (Collard et al., 2005) [11] . Furthermore, several studies have proposed that QTL positions and effects should be evaluated in independent populations, because QTL mapping based on typical population sizes results in a low power of QTL detection and a large bias of QTL effects (Collard et al., 2005) [11] . Unfortunately, due to constraints such as lack of research funding and time and possibly a lack of understanding of the need to confirm results, QTL mapping studies are rarely confirmed. Hence, the objective of this study was to develop interspecific BC 1 F 1 population derived from locally adapted G. hirsutum var. KC3 and G. barbadense var. Suvin for linkage map construction and QTL analysis of fibre quality traits using simple and anchored SSR markers.

Materials and Methods

Plant materials

Short duration upland cotton variety, G. hirsutum L. var. KC3, possessing short staple length but well adapted to rainfed tracts of Tamil Nadu, India was used as recurrent parent. A contrasting long duration G. barbadense L. var. Suvin, possessing exceptionally long and strong fibres and susceptible to sucking pests and drought, was used as donor parent. The two parents and 62 BC 1 F 1 individuals were grown in a field by following regular agronomic practices at Cotton Breeding Station, Tamil Nadu Agricultural University (TNAU), Coimbatore, India for evaluation of fiber quality traits and for providing leaf samples for linkage map construction.

Phenotyping

Observations on five important fibre quality traits were recorded on every replicated progenies of BC 1 F 1 population. Matured bolls were collected from each plant, ginned and the lint was used to assess the fibre quality traits using highvolume instrument (HVI900 classic). The tested fibre quality traits included 2.5 per cent span length (mm), bundle strength (g/tex), fibre fineness (micronaire), uniformity ratio and elongation per cent. From these data, broad sense heritability for all the fibre quality traits was calculated besides regular statistical analysis.

Genotyping and Segregation analysis

Genomic DNA was isolated from the fresh leaves of parents and the 62 BC 1 F 1 individuals and genotyping with SSR markers were done with established procedure (Zhang and Stewart, 2000) [12] and PCR products were separated on a 4% metaphor agarose gel. A total of 391 pair of SSR primer sequences that span all over the cotton genome were obtained from cotton marker database (CMD) (http://www.cottonmarker.org/Downloads). A sum of 141 polymorphic SSRs was used for genotyping the BC 1 F 1 mapping population. The segregation pattern of SSR markers for all the BC 1 F 1 individuals were scored as codominant fragments and this genotyping data was used for linkage map construction and QTL analysis for fibre quality traits. Initially, genotyping data of all the SSRs were used for Chi-square (χ 2 ) test using AntMap Ver. 1.2 for deviations from the expected Mendelian ratios with P < 0.05 (Iwata and Ninomiya, 2006) [13] . The SSRs, which have shown 1:1 segregation ratio for BC 1 F 1 progenies were used for linkage map construction.

Linkage map construction and QTL analysis

Preliminary linkage analysis was performed using both AntMap Ver. 1.2 (Iwata and Ninomiya, 2006) [13] and MAPMAKER/EXP 3.0 (Lander et al., 1987) [14] . Linkage groups were assigned to corresponding chromosomes using anchored SSR markers as per CMD. QTL analysis was performed using Windows QTL Cartographer V2.5 (Basten et al., 1999) [15] by employing single marker analysis (SMA), interval mapping (IM) and composite interval mapping (CIM) methods. As per the results of 1000 permutation tests, a LOD threshold of ≥2.5 was set to identify the putative QTL linked to fibre quality traits.

Results

Genetic Variation and Heritability

Segregation data for all fibre quality traits in BC 1 F 1 progenies was tested for genetic variation and heritability. The two parental lines differed significantly for all the fiber quality traits (Table 1). Transgressive segregation or continuous variation for all the investigated fibre quality traits showed the existence of wide genetic variation among the BC 1 F 1 population and represented both the extremes of parental values for all the investigated traits. The heritability estimates ranged from 54.10 to 72.30 per cent in BC 1 F 1 progenies (Table 1).

Linkage analysis and map construction

A total of 391 SSRs were used to screen for polymorphism between the parents KC3 and Suvin. Though, 162 SSRs showed polymorphism between parents, only 141 (41.43%) SSRs yielded unambiguous, scorable polymorphic loci and thus, they were chosen to genotype the BC 1 F 1 population. Genotypic data on all the 141 polymorphic loci were utilized to construct linkage map using AntMap Ver. 1.2. All the SSRs were assigned and arranged with their respective chromosome number as per CMD. Linkage analysis was carried out individually for all the 26 chromosomes of cotton and genetic map was constructed (Figure 1). The linkage map constructed using BC 1 F 1 population, among the 141 SSR loci, 19 loci deviated from an expected 1:1 segregation ratio (P≤0.05). Thirteen per cent of the markers showed distorted segregation, indicating that SSRs are likely to segregate normally, although the population is not large. Among the distorted loci, 9 loci skewed towards the 'KC3' parent and 10 loci skewed towards the heterozygote. The markers that are segregating normally were analyzed further and it was noticed that 57 loci were found to be linked and the 84 loci were remained unlinked at LOD of 2.5. A total of 19 linkage groups were formed which were assigned to 16 chromosomes of cotton. The 19 linkage groups contained two to seven markers with an average length of 47.98 cM. The map covered a total of 911.6 cM and the average distance between adjacent markers was 16 cM. Maximum interval length of 44.9 cM was observed between CIR185 and BNL3995 on chromosome 5 and a minimum inter marker distance of 5.1 cM has been noticed between CIR039 and CIR143 on chromosome 26 (Figure 1).

QTLs associated with fibre quality traits

QTL analysis was carried out with QTL Cartographer V.2.5 by performing SMA, IM and CIM to find out the QTLs linked to fibre quality traits in cotton. Interesting outcome of this study was identification of two major QTLs by SMA, IM and CIM. The marker interval NAU3935 and CIR219, on chromosome 19 has been found to be associated with 2.5 per cent span length and explained 17.7% of phenotypic variation. Another QTL has been found to be associated with micronaire which was located on chromosome 25 with the marker interval CIR150 and CIR407 and contributed 14.7 per cent of phenotypic variation in BC 1 F 1 progenies (Table 2).

Discussion

Heritability estimates are normally more helpful in predicting the gain under selection. In the present study, high heritability was observed for all the fiber quality traits which revealed that these traits were polygenic in nature and amenable for QTL analysis (Paterson et al., 2003) [9] . However, the estimates of heritability are subject to environmental conditions and therefore may be used with great care and caution. The present investigation was attempted to utilize BC 1 F 1 progenies for detecting the genomic regions associated with fibre quality traits. Though, similar kind of linkage maps using the same type of mapping populations derived from interspecific crosses of G. hirsutum L. and G. barbadense L. have already been reported, it is important to validate the identified QTLs in the locally adapted accessions for effective MAS (Boopathi et al., 2011) [4] . A population size of 62 BC 1 F 1 progenies were used for linkage map construction and QTL analysis in the present study. The earlier studies by Lacape et al. (2003) [10] , Nguyen et al. (2004) [16] and Lacape et al. (2005) [6] have used 75 individuals of BC 1 F 1 population. There is no specific study relating to the ideal number of individuals in a given population required to establish accurate genetic maps. A segregating population consisting of 50 to 250 individuals is required to construct the linkage map (Mohan et al., 1997) [17] , but high resolution fine mapping is possible only with a larger population size. Hence, it was believed that the population size used in this investigation was appropriate. SSR markers have been used for linkage map construction in this report. SSRs can be used as anchors to combine linkage map produced with different populations to generate more comprehensive genetic linkage map. Out of 391 markers, 141 SSRs revealed (41.43 per cent) polymorphism between KC3 and Suvin which was in accordance with recorded polymorphism of 38.31 per cent (Guo et al., 2007) [18] . This polymorphism rate provided by SSRs is generally regarded as sufficient for constructing a reasonably saturated linkage map. However, during linkage analysis it was found that >60 per cent of markers remained unlinked. This might be attributed to the existence of inter-parental incompatibility alleles, poor chromosomal rearrangements which could be due to narrow marker coverage and lack of informative recombinations (Lin et al., 2005) [2] . The current linkage maps covered approximately 16.5 per cent of genome, since the markers have sparsely distributed. Formation of small and a less number of total linkage groups also indicate that the marker coverage in the genome is not exhaustive and more number of markers needs to be added to the map to represent the entire genome. Similar kind of poor coverage of cotton genome has already been reported. For example, eleven linkage groups were identified from a trispecific F 2 population that spanned 521.7 cM with an average distance of 16.8 cM (Altaf et al., 1997) [19] . Han et al. (2006) [20] constructed a linkage map from BC 1 F 1 population which contains only six linkage groups. The earlier linkage studies also shown poor coverage of cotton genome due to sparse distribution of markers and less number of total linkage groups (Bolek et al., 2006 [21] , Wang et al., 2006 [22] , Abdurakhmonov et al., 2007 [23] , Guo et al., 2008 [24] , Lu et al., 2008 [25] ). The sparse distribution of the QTLs in this study is consistent with the observation that a significant proportion of cotton genome comprises of repetitive sequences (Reinisch et al., 1994) [3] . Hence, identification and incorporation of additional SSRs or other kinds of high-throughput markers are required to construct a sufficiently dense linkage map for reasonable QTL analysis. Lacape et al. (2005) [6] have published such a map with BC 1 and BC 2 populations.

The key finding of this study was identification of two major QTLs linked to fibre quality traits in the interspecific backcross mapping population. The genomic region flanked by CIR150 and CIR407 was associated with micronaire on chromosome 25 has already been reported with different genetic backgrounds. Contribution of this QTL to phenotypic variance was noticed as 14.7 per cent in this study. This can be considered as major QTL since it represents more than 10 per cent of phenotypic variance (Collard et al., 2005) [11] . Hence, this QTL has its potential to be used in future cotton improvement programme. Yet another appealing outcome of this study was identification of a major QTL region for 2.5 per cent span length located on chromosome 19, which was not reported by any of the earlier studies. This QTL explained 17.7 per cent of phenotypic variation. This may be a unique QTL of this population. However, use of this QTL in MAS has to be confirmed with replicated field trials across the environments with the employment of immortal populations such as RILs and NILs.

Conclusion

The two major QTLs reported in this study indicate their usefulness to breeders to trace this important fibre quality trait in early growing stage and early segregating generations and thus enormous cost on raising the experiments and manpower can be saved. Hence, markers identified in this study could be useful in future breeding programme to improve fibre quality traits in cotton.

Acknowledgement

AcknowledgementThis work is supported by Department of Biotechnology, Ministry of Science and Technology, Government of India.

References

1. Kohel RJ. Cotton germplasm resources and the potential for improved fibre production and quality. In: Basra AS, editor. Cotton fibres. New York: The Haworth Press; 1999. p. 167-182.
2. Lin ZD, He X, Zhang Y, Nie X, Guo C, Feng C, Stewart J. Linkage map construction and mapping QTL for cotton fibre quality using SRAP, SSR and RAPD. Plant Breeding. 2005;124(2):180-187.
3. Reinisch AJ, Dong J, Brubaker CL, Stelly DM, Wendelt JF, Paterson AH. A detailed RFLP map of cotton Gossypium hirsutum × Gossypium barbadense: chromosome organization and evolution in a disomic polyploid genome. Genetics. 1994;138:829-847.
4. Boopathi NM, Thiyagu K, Urbi Banerji, Magadum S, Gopikrishnan A, Aravind S, Gat Swapnashri, Ravikesavan R. Marker assisted breeding as next generation strategy for genetic improvement of productivity and quality: can it be realized in cotton? International Journal of Plant Genomics. 2011;2011:670104. doi:10.1155/2011/670104
5. Reddy OUK, Pepper AE, Abdurakhmonov I, Saha S, Jenkins JN, Brooks T, Bolek Y, El-Zik KM. New dinucleotide and trinucleotide microsatellite marker resources for cotton genome research. Journal of Cotton Science. 2001;5:103-113.
6. Lacape JM, Nguyen T, Courtois B, Belot J, Giband M, Gourlot J, Gawryziak G, Roques S, Hau B. QTL analysis of cotton fiber quality using multiple Gossypium hirsutum × Gossypium barbadense backcross generations. Crop Science. 2005;45(1):123140.
7. Reiter RS, Williams JGK, Feldman KA, Rafalski A, Tingey SV, Scolnik PA. Global and local genome mapping in Arabidopsis thaliana by using recombinant inbred lines and random amplified polymorphic DNAs. Proceedings of National Academy of Sciences USA. 1992;89:1477-1481.
8. Gutierrez OA, Basu S, Saha S, Jenkins JN, Shoemaker DB, Cheatham CL, McCarty JC Jr. Genetic distance among selected cotton genotypes and its relationship with F2 performance. Crop Science. 2002;42:18411847.
9. Paterson AH, Saranga Y, Menz M, Jiang CM, Wright RJ. QTL analysis of genotype × environment interactions affecting cotton fibre quality. Theoretical and Applied Genetics. 2003;106(3):384-396.
10. Lacape JM, Nguyen T, Thibivilliers S, Bojinov B, Courtois B, Cantrell R, Burr B, Hau B. A combined RFLP-SSR-AFLP map of tetraploid cotton based on a Gossypium hirsutum × Gossypium barbadense backcross population. Genome. 2003;46(4):612-626.
11. Collard BCY, Jahufer MZZ, Brouwer JB, Pang ECK. An introduction to markers, quantitative trait loci (QTL) mapping and marker assisted selection for crop improvement: the basic concepts. Euphytica. 2005;142:169-196.
12. Zhang J, Stewart JM. Economical and rapid method for extracting cotton genomic DNA. Journal of Cotton Science. 2000;4:193-201.
13. Iwata H, Ninomiya S. AntMap: constructing genetic linkage maps using an ant colony optimization algorithm. Breeding Science. 2006;56:371-377.
14. Lander ES, Green P, Abrahmson J, Barlow A, Daly MJ, Lincoln SE, Newburg L. Mapmaker: an interactive computer package for constructing primary genetic linkage maps of experimental and natural populations. Genomics. 1987;1:174-181.
15. Basten C, Weir B, Beng ZB. QTL Cartographer, Version 1.13. Raleigh: Dept. of Statistics, North Carolina State; 1999.
16. Nguyen T, Giband M, Brottier P, Risterucci AM, Lacape JM. Wide coverage of the tetraploid cotton genome using newly developed microsatellite markers. Theoretical and Applied Genetics. 2004;109(1):167175.
17. Mohan M, Bhagwat A, Krishna TG, Yano M, Bhatia CR, Sasaki T. Genome mapping, molecular markers and marker assisted selection in crop plants. Molecular Breeding. 1997;3:87-103.
18. Guo W, Cai C, Wang C, Han Z, Song X, Wang W, Niu X, Wang C, Lu K, Shi B, Zhang T. A microsatellite based gene rich linkage map reveals genome structure, function and evolution in Gossypium. Genetics. 2007;176(1):527-541.
19. Altaf MK, Stewart MJ, Wajahatullah MK, Zhang J. Molecular and morphological genetics of a trispecies F2 population of cotton. Proceedings of Beltwide Cotton Improvement Conference, USA. 1997;448-452.
20. Han Z, Wang C, Song X, Guo W, Gou J, Li C, Chen X, Zhang T. Characteristics development and mapping of Gossypium hirsutum derived EST-SSRs in allotetraploid cotton. Theoretical and Applied Genetics. 2006;112:430-439.
21. Bolek Y, El-Zik K, Pepper A, Bell A, Magill C, Thaxton P, Reddy O. Mapping of verticillium wilt resistance genes in cotton. Plant Science. 2005;168(6):1581-1590.
22. Wang B, Wu Y, Huang N, Zhu X, Guo W, Zhang T. QTL mapping for plant architecture traits in upland cotton using RILs and SSR markers. Acta Genetica Sinica. 2006;33(2):161-170.
23. Abdurakhmonov I, Buriev Z, Saha S, Pepper A, Musaev J, Almatov A, Shermatov S, Kushanov F, Mavlonov G. Microsatellite markers associated with lint percentage trait in cotton (Gossypium hirsutum L.). Euphytica. 2007;156(1):141-156.
24. Guo Y, McCarty JC, Jenkins JN, Saha S. QTLs for node of first fruiting branch in a cross of an upland cotton, Gossypium hirsutum L., cultivar with primitive accession Texas 701. Euphytica. 2008;163:113-122.
25. Lu Y, Curtiss J, Miranda D, Hughs E, Zhang J. ATGanchored AFLP (ATG-AFLP) analysis in cotton. Plant Cell Reporter. 2008;27:1645-1653.