Previous PageTable Of ContentsNext Page

Chromosome substitution lines from Gossypium barbadense L. as sources for G. hirsutum L. improvement

Johnie N. Jenkins1, Jack C. McCarty1, Jixiang Wu2, Sukumar Saha1 and David Stelly3

1USDA-ARS, Crop Science Research Laboratory, Mississippi State, MS, USA
2
Department of Plant and Soil Sciences, Mississippi State University, Mississippi State, MS, USA
3
Department of Soils and Crop Sciences, Texas A & M University, College Station, TX, USA

Abstract

Gossypium barbadense line 3-79 has exceptionally good fiber length, fineness, and strength. Chromosome substitution (CS-B) lines with individual 3-79 chromosomes or chromosome arms backcrossed into TM-1, G. hirsutum, were studied for combining ability following a top cross to five elite cultivars. General Combining Ability (GCA) effects predominated. Among the CS-B lines, CS-B25 had the highest GCA for fiber length and strength. Among the cultivars PCS 355 and FM 966 were the best in GCA for fiber length and FM 966 for fiber strength. Line 3-79 was the highest in GCA for fiber length and strength; however it also had very high negative GCA for yield and lint percent. These data suggest that these CS-B lines are useful for introgressing fiber quality genes from 3-79 into G. hirsutum; however, these CS-B lines do not solve all the problems with interspecific hybridization at the whole genome level commonly seen when crossing the two species.

Media summary

Chromosome substitution (CS-B) lines offer an effective way to introgress fiber quality genes from Gossypium barbadense, line 3-79 into upland cotton, G. hirsutum..

Key Words

Cotton, genetic diversity, species, fiber, breeding.

Introduction

Gossypium barbadense L.line 3-79 has exceptionally good fiber length and strength. Interspecific crosses of upland cotton, G. hirsutum L. and G. barbadense. have been used in attempts to transfer genes into upland cotton. Only limited success has been achieved. This is probably related to abnormal chromosome pairing in the interspecific hybrids resulting in partial sterility and/or abnormal segregation due to incompatability between the two species at the whole genome level. Chromosome substitution (CS-B) lines, in which single G. barbadense chromosomes or chromosome arms are introgressed into upland, using hypoaneuploid mediated backcross substitution, should offer relief from some problems encountered in interspecific crosses. Euploid plants of CS-B lines can be used in crosses with upland and subsequent hybrids will primarily have one G. bardadense chromosome and its genes segregating in the cross whereas full interspecific crosses have genes of all 26 chromosomes segregating with one member of each chromosome pair from upland and one member from G. barbadense. A series of backcrossed chromosome substitution lines have been developed in upland cotton germplasm line TM-1. Each of these lines has one chromosome or chromosome arm from G. barbadense, line 3-79 present with the 25 chromosomes from TM-1.

Methods

Development of Plant Materials

Five elite cultivars ‘Deltapine 90’ (DP90); ‘FiberMax 966’(FM 966); ‘Stoneville 474’ (ST 474); ‘Phytogen 355’(PSC 355); and ‘SureGrow 747’ (SG 747), from diverse commercial breeding programs, were crossed as females with thirteen chromosome substitution lines plus TM-1 and G. barbadense line 3-79, Table 1. The chromosome substitution lines represent 5 with subgenome A chromosomes and 8 with subgenome D chromosomes from G. barbadense line 3-79 substituted into TM-1.

Field Design and Procedures

The resulting 75 F2 of topcrossed lines, the five cultivar female parents, TM-1, line 3-79, and the 13 chromosome substitution lines were planted at two locations in 2003 at the Plant Science Research Center at Mississippi State, MS. Soil type for location one was a Marietta loam (Fine-loamy, siliceous, active, fluvaquentic Eutrudepts) and for location two it was a Leeper silty clay loam (Fine, smectitic, nonacid, thermic Vertic Epiaquept). Plots were planted in a plant two skip one row pattern on 28 May and harvested on 3 November, 2003 at location one and 31 October 31 at location two. Standard cultural practices were followed for the season and the growing season was above average at each location. A 25 boll sample was hand harvested from first position bolls near the middle nodes of plants in each plot. This sample was ginned on a 10 saw laboratory gin to determine boll weight and lint percentage. Lint samples were sent to a commercial lab, StarLab, for single instrument fiber measurements. After the boll sample was harvested, all plots were harvested with a commercial cotton picker modified to harvest seedcotton from each one row plot.

Data analyses

Data was subjected to the analysis of variance using SAS version 8.0 (SAS Institute 1999). Means were separated using Fishers protected least significant difference (LSD) at the 0.05 level. Genetic effects were calculated using an additive-dominance model (AD) (Cockerham 1980). Locations were considered as environments for the data analyses. The following model was used for analyses:

Parents: Yhiik = μ + Eh +2Ai +Dii +2AEhi +DEhii +Bk(h) + ehiik

F2: Yhijk(2) = μ + Eh + Ai + Aj + 1/4Dii + 1/4Djj + 1/2Dij + AEhi +AEhj +1/4DEhjj + 1/4DEhii

+ 1/2DEhij + Bk(h) + ehijk

where μ= grand population mean, h = environment, i = male parent, j = female parent, and k = block within environment. The variance components for each trait were estimated utilizing minimum unbiased estimation (MINQUE) approach (Rao, 1971). The adjusted unbiased prediction (AUP) approaches (Zhu 2001) were used to make predictions of genetic effects and genetic x environment effects. Jackknifing over blocks within each environment was used to calculate standard errors for each parameter (Miller 1974). A one-tailed t-test was used to determine significance of variance components from zero, while a two-tailed t-test was used to determine significant difference among genetic effects.

Results

Yield and fiber properties of the cultivars and chromosome substitution lines are shown in Table 1. Boll weight among CS-B lines ranged from less than to equal to TM-1 or some cultivars. Lint percentage among CS-B lines ranged from less than to greater than TM-1 with none equal to cultivars. Lint yields among CS-B lines ranged from greater than to equal to TM-1and lower than cultivars. Micronaire ranged from less than to equal to TM-1. Fiber length ranged from less than to greater than TM-1 with none equal to 3-79. Fiber strength ranged from greater than to equal to TM-1 with none as high as line 3-79. Sufficient range existed for all traits to make this a useful group of lines to study.

Variance components in Table 2 show that General Combining Ability (GCA or additive effects) predominated for each trait; however, most traits also exhibited significant dominance effects. Data in Table 3 show a range of GCA effects among CS-B lines from positive to negative for most traits. CS-B25, line 3-79, and FM 966 had significant positive GCA for fiber strength and length. All cultivars had positive GCA for yield and lint percent. GCA for CS-B25, 3-79, and DP 90 indicated these should reduce micronaire. GCA among chromosome substitution lines for boll weight was positive for nine and negative for three. Only CS-B22sh and CS-B22Lo had major and positive GCA for lint percent and lint yield. Only CS-B25 had a negative GCA for micronaire.

Eight CS-B lines had a negative and two a positive GCA for fiber length. Nine were negative and one was positive for fiber strength. CS-B25 had positive GCA for fiber length and strength and negative GCA for micronaire. This appeared to be the best general combiner for improving fiber quality; however, it had a negative GCA for lint yield. GCA for fiber length and strength was the highest in line 3-79; however, 3-79 had a very large significant negative GCA for yield. GCA for each cultivar was positive for lint percent and lint yield. FM 966 exhibited positive GCA for fiber length and strength and PSC 355 showed positive GCA for fiber length.

Table 1. Yield and fiber data of parental lines averaged over two locations in 2003.

Entry Name

Boll Wt.

g

Lint

Percent

Lint Yield

Kg/ha

Micronaire

2.5% SL

mm

Strength

kNm/kg

CS-B02

5.96

33.60

1146

4.89

28.31

209

CS-B04

5.84

33.07

1022

4.38

29.02

190

CS-B06

6.05

33.17

1223

4.58

28.04

194

CS-B07

5.50

32.27

1087

4.66

28.43

193

CS-B16

6.19

35.66

460

4.66

27.31

191

CS-B17

5.44

30.18

686

4.10

27.62

194

CS-B18

5.38

34.28

785

4.71

26.10

194

CS-B25

4.93

31.35

831

4.31

30.13

212

CS-B05sh

5.31

33.71

1324

4.83

27.78

187

CS-B14sh

4.72

32.02

885

4.53

29.70

206

CS-B15sh

5.71

31.44

800

4.46

29.16

199

CS-B22sh

5.52

35.45

1032

4.90

26.96

196

CS-B22Lo

4.71

36.69

1189

4.90

27.05

194

TM-1

6.06

34.20

1097

4.65

28.08

192

Line 3-79

3.90

35.00

680

3.85

35.67

272

DP 90

5.03

38.80

1612

4.53

29.27

216

SG 747

5.52

41.06

1850

4.78

28.53

180

PSC 355

5.18

40.85

2082

4.95

29.05

199

ST 474

5.08

42.19

2008

4.85

28.21

192

FM 966

6.27

41.84

2165

4.64

30.45

238

LSD(P=0.05)

0.25

0.89

230

0.22

0.89

10

Table 2. Variance components and proportions for yield and fiber properties averaged over two locations in 2003.

Variance
Component

Boll Wt.

 

Lint
Percent

 

Lint Yield

 

Micronaire

 

2.5% SL

 

T1

 

VA

0.46

**

7.53

**

234851

**

0.113

**

1.23

**

230

**

VD

0.23

**

4.41

**

104406

**

0.049

**

0.35

*

42

 

VA*E

0.00

 

0.00

 

0.00

 

0.002

 

0.00

 

6

*

VD*E

0.01

 

0.72

*

39470

*

0.013

*

0.23

*

0

 

VRESIDUAL

0.07

**

0.82

**

54395

**

0.050

**

0.82

*

111

**

VPHENOTYPIC

0.76

**

13.48

**

433122

**

0.228

**

2.63

**

387

**

                         

Variance
proportion

                       

VA/VP

60.27

**

55.88

**

54

**

49.66

**

46.83

**

59

**

VD/VP

29.56

**

32.72

**

21

**

21.85

**

13.14

*

11

*

VA*E/VP

0.00

 

0.00

 

0

 

1.01

 

0.00

 

2

 

VD*E/VP

1.65

 

5.35

**

9

*

5.47

*

8.83

*

0

 

VRESIDUAL

8.52

**

6.05

**

13

*

22.02

**

31.20

**

28

**

Table 3. Genearal Combining Ability for parents averaged over two locations in 2003.

Entry Name

Boll Wt.
g

SE

Lint
%

SE

Lint Yield
Kg/ ha

SE

Mic

SE

2.5% SL
mm

SE

T1
kNm/kg

SE

Grand Mean

5.68

 

35.61

 

1311

 


4.57

 


29.10

 


201

 

CS-B02

0.28

0.05

-0.63

0.08

34

20

0.15

0.02

-0.31

0.06

2.2

1.3

CS-B04

0.19

0.04

-0.54

0.10

-98

15

0.00

0.02

0.19

0.11

-4.2

2.3

CS-B06

0.12

0.03

-1.03

0.15

-3

44

-0.05

0.02

-0.17

0.11

-8.3

1.2

CS-B07

0.14

0.05

-0.34

0.18

58

36

0.13

0.04

-0.15

0.14

-6.4

1.5

CS-B16

0.34

0.04

0.03

0.16

-320

24

0.07

0.04

-0.59

0.11

-5.2

1.7

CS-B17

0.18

0.02

-1.02

0.09

-102

28

-0.09

0.04

-0.33

0.12

-3.6

1.8

CS-B18

0.02

0.03

0.17

0.16

-142

28

0.15

0.03

-0.25

0.17

-1.8

1.2

CS-B25

-0.07

0.03

-1.58

0.18

-89

43

-0.31

0.02

0.47

0.11

9.0

1.4

CS-B05sh

-0.06

0.04

-0.22

0.12

47

54

0.23

0.06

-0.80

0.20

-6.2

1.3

CS-B14sh

-0.27

0.03

-1.68

0.12

-105

24

-0.02

0.05

0.20

0.22

-2.3

0.7

CS-B15sh

0.20

0.04

-1.63

0.18

-113

27

-0.09

0.05

0.22

0.08

-1.1

1.3

CS-B22sh

0.14

0.04

1.53

0.18

95

33

0.28

0.02

-1.46

0.08

-6.1

1.4

CS-B22Lo

-0.13

0.03

0.57

0.16

95

48

0.20

0.05

-0.63

0.17

-2.9

0.8

TM-1

0.23

0.05

-0.93

0.06

-20

21

-0.03

0.04

-0.21

0.12

-5.5

1.4

Line 3-79

-1.86

0.04

-4.97

0.29

-1148

31

-0.83

0.05

2.26

0.14

39.3

3.5

DP 90

-0.14

0.03

1.48

0.13

131

31

-0.15

0.02

0.05

0.14

-1.3

1.5

SG 747

0.06

0.04

2.35

0.15

411

32

0.09

0.02

-0.00

0.12

-11.3

1.0

PSC 355

0.04

0.02

2.86

0.09

468

26

0.14

0.02

0.47

0.14

2.5

2.0

ST 474

-0.05

0.04

1.92

0.17

301

24

0.14

0.03

-0.46

0.12

0.11

2.0

FM 966

0.66

0.06

3.70

0.19

500

31

0.00

0.03

1.53

0.14

13.7

2.1

Conclusion

These data suggest that chromosome substitution lines are useful for introgressing genes from G. barbadense, line 3-79 into G. hirsutum; however, these lines do not completely solve the negative genetic associations between yield and fiber quality or problems with interspecific hybridization at the whole genome level commonly seen when crossing the two species. CS-B25 was the best of the 13 CS-B lines for GCA for fiber quality traits of length, micronaire, and strength.

References

Cockerham CC (1980). Random and fixed effects in plant genetics. Theor. Appl. Genet. 56, 119-131

Miller RG (1974). The jackknife; a review. Biometrika 61, 1-15

Rao CR (1971). Estimation of variance and covariance components MINQUE theory. J. Multivariate Analysis 1, 257-275.

SAS Institute (1999). SAS software Version 8.0. SAS Institute, Cary, NC.

Zhu J (2001). New approaches for analyzing quantitative traits and their application in cotton, pp 43-63. In Genetic Improvements of Cotton: Emerging Technologies, J. N. Jenkins and S. Saha (eds). Science Publishers, Inc. Enfield, NH.

Previous PageTop Of PageNext Page