Table Of ContentsNext Page

Sorghum ergot: revealing the genetic architecture of resistance

Dipal Parh1,2, David Jordan3, Elizabeth Aitken2,4, Lynne McIntyre2,5 and Ian Godwin1,2

1School of Land and Food Sciences, 2CRC for Tropical Plant Protection, University of Queensland, St Lucia, Qld 4072,
Email: d.parh@uq.edu.au
3Queensland Department of Primary Industries, Warwick, Qld 4370
4School of Life Sciences, University of Queensland, St Lucia, Qld 4072
5CSIRO, Plant Industry, Queensland Bioscience Precinct, St Lucia, QLD 4067

Abstract

Sorghum ergot, caused by Claviceps africana, is a major disease of the sorghum inflorescence. It is of great concern to the Australian sorghum industry because of its recent spread worldwide and its impact on seed production and animal health. The objective of this study was to identify molecular markers linked to the genomic region(s) underlying ergot resistance in sorghum for potential use in marker-assisted selection for this trait. One hundred and thirty four F5 recombinant inbred lines (RILs), developed from the cross 31945-2-2 x IS8525, were used to construct a genetic map of sorghum integrating 55 SSR, 230 AFLP and two morphological trait loci. Composite interval mapping identified five independent genomic regions for ergot resistance (pcergot). In addition, QTLs for two pollen traits, pollen quantity (pq) and pollen viability (pv) were identified, with five and three QTLs identified for each trait, respectively. A comparison of the map locations of the QTLs for all 3 traits identified only one linked QTL (28 cM away) for pcergot and pq; the remaining QTLs appear to be different. These results suggest that ergot resistance is a complex trait and that variation in pollen quantity and pollen viability do not explain the genetic variability for ergot resistance.

Media summary

Molecular markers associated with ergot resistance in sorghum have been identified and can be used for marker-assisted selection. The results demonstrate that pollen quality and pollen viability are not the only components involved in ergot resistance.

Key Words

Sorghum bicolor, Ergot resistance, Pollen quantity and viability, Linkage map, molecular markers

Introduction

Sorghum ergot caused by Claviceps africana is a great concern to the sorghum industry worldwide because of its impact on seed production and animal health. Hybrid seed production is particularly vulnerable to ergot damage if restorer lines do not produce adequate pollen when stigmas of the male-sterile seed lines are receptive. Sorghum grain contaminated with sclerotia (fungal bodies) can cause toxicity when fed to livestock particularly sows, dairy cattle and beef cattle in feedlots (Blaney et al. 2000).

The genetic and physiological bases for ergot resistance are poorly understood. In the past, a number of resistant lines have been reported but in most cases resistance appears to be a pollen-mediated disease escape. The first confirmed source of non-pollen-based ergot resistance was reported in IS8525 (Wang et al. 2000; Dahlberg et al. 2001). A recombinant inbred line population has been produced at QDPI using IS8525 and an elite sorghum parental line, 31945-2-2. Field studies of this population have confirmed that the population is segregating for ergot resistance. As severity of sorghum ergot varies with the environment, and given the importance of this disease, we have initiated a genetic study of ergot resistance in IS8525. The objectives of the present study were to identify molecular markers associated with ergot resistance for potential use in marker-assisted selection, and to analyse the relationship between ergot resistance and two pollen traits, pollen quality and pollen viability.

Methods

Plant material

A subset of 134 RILs of the cross 31945-2-2 x IS8525 were used for map construction and subsequent QTL analysis. 31945-2-2 is a commercially accepted restorer line developed by QDPI. It produces good quantities of viable pollen under favourable environmental conditions but its pollen production is severely hindered as the temperature becomes colder. IS8525 is a source of ergot resistance. It produces moderate quantity of highly viable pollen under both favourable and adverse environmental conditions.

Field trials

Two field trials were conducted at the Hermitage Research Station, Warwick, QLD, during the 2001 and 2002 growing seasons. In each year, there were two planting dates, sown 10 - 12 days apart, such that sufficient variation in flowering date occurred for each RIL to provide a ranges of inoculation dates for ergot ratings. These sowing dates were also expected to increase the possibility of weather conditions conducive to ergot development (Wang et al. 2000).

Inoculation and Data collection

A conidial suspension was prepared by washing infected sorghum heads containing fresh honeydew. Inoculation was performed on six different dates from 27 April to 15 May during 2001 and on ten different dates from 4 April to 1 May during 2002. Percent ergot infection of the inoculated plant was recorded at the soft dough stage of grain development. Pollen quantity was measured by flicking selected heads once and rated using a 1-10 scale, where 1 was no visible pollen and 10 was copious quantities of visible pollen. Pollen viability was measured by collecting flicked pollen on a water-agar plate, spraying with iodine solution and counting for viability under a compound microscope.

Map construction

A linkage map of sorghum was constructed using the computer program Map Manager QTX version 0.27 (Manly et al. 2001) and integretated 58 SSR and 234 AFLP markers and two morphological traits, coleoptile colour and red leaf. The map was constructed with a high LOD score (P = 0.001). A Kosambi (Kosambi 1944) mapping function was used to calculate the centimorgan (cM) values.

QTL analysis

QTLs were detected by Composite Interval Mapping (CIM) (Zeng 1994) using the computer software package QTL Cartographer (Basten et al. 2002). Model 6, forward and backward step-wise regression and a 10-cM scan window were used for the detection of QTL. The significant LOD threshold level, as determined by 1000 permutation test (Churchill and Doerge 1994) for pcergot, pq and pv were 2.95, 2.99 and 3.06, respectively. However, putative QTL exceeding a LOD threshold of >2.0 has also been reported to identify suggestive QTL for all traits.

Results

Phenotypic data for pcergot and pollen traits

The overall best linear unbiased predicted (BLUP) value for each RIL for pcergot and the two pollen traits, pq and pv were estimated using the factor analytic model for genotype by environment interaction (Smith et al. 2001). An approximately normal distribution of BLUP value (square-root transformed) among the 134 RILs was observed for all the traits studied (Figure 1). The overall mean of RILs and the two parents, 31945-2-2 and IS8525 were 14.86, 30.18, 4.08 for pcergot; 4.71, 3.09, 4.76 for pq and 86.99, 79.51, 92.63 for pv, respectively and the ranges of the RILs were 3.60-44.29, 2.64-6.39, 72.40-94.75 for pcergot, pq and pv, respectively.

Linkage map construction

The linkage map contained 287 markers grouped into sixteen linkage groups (LGs); all 16 LGs could be aligned to the published 10 LGs A-J of sorghum (Bhattramakki et al. 2000; Menz et al. 2002). This map was used for subsequent QTL analysis.

Figure 1 Frequency histogram of pcergot, pq and pv in the mapping population 31945-2-2 x IS8525. The square-root transformed values for each trait are shown on the top of each bar.

QTL analyses

The CIM analysis identified a number of genomic regions significant for QTL-marker associations for pcergot and the two pollen traits, pq and pv (Table 1). For pcergot, five genomic regions on LGs 4, 6, 7, 11 and 12 were identified with statistical significance (LOD >2.5). Three of these on LGs 6, 11 and 12 also met the permutation-based threshold of LOD 2.95. The

phenotypic variation explained by each of the loci varied from 5.5 –11.4% and collectively explained 39.2% of the total variation for pcergot. The allele effect of all the loci linked to QTLs

Table 1 QTLs for pcergot, pq and pv in the mapping population 31945-2-2 x IS8525

Trait

Linkage
group

Markera

QTL
position

Recomb-
in-ation x Lb

Recomb-
in-ation x Rc

LOD
score

Additive
effectd

R2e

pcergo

4

10 (TxP56)

74.22

0.0001

0.0404

2.72

-1.53

0.0550

6

37 (AGC+CTG4)

252.65

0.0201

0.0062

3.07

-1.68

0.0699

7

5 (ACA+CAA3)

27.32

0.0001

0.0513

2.91

-1.61

0.0656

11

22 (TxP274)

136.97

0.0201

0.0338

4.78

-2.12

0.1141

12

15 (AGC+CTA2)

128.23

0.1548

0.0340

3.32

-1.87

0.0878

pq

4

11 (ACC+CTC2)

102.28

0.2232

0.0371

2.76

0.235

0.0770

5

2 (AG+CTC3)

33.62

0.1901

0.0851

3.39

-0.292

0.1321

6

42 (ACT+CAT3)

302.30

0.0987

0.0174

3.18

-0.231

0.0823

11

11 (CC)

62.25

0.0001

0.0506

5.49

0.301

0.1419

12

8 (ACC+CTC3)

54.63

0.0988

0.0558

2.38

0.214

0.0710

pv

3

23 (AAC+CAA2)

85.47

0.0201

0.0134

2.22

-1.99

0.0646

6

19 (AAG+CTG3)

160.10

0.0400

0.0150

3.44

1.16

0.0951

12

10 (ACA+CAG3

70.72

0.0001

0.0120

2.84

1.06

0.0763

aOnly marker with highest LOD score is shown
bRecombination with left marker
cRecombination with right marker
dEffect of substitution of AA (IS8525) allele by BB (31945-2-2) allele
e% of total phenotypic variation explained was negative indicating that the region linked to QTLs derived the beneficial allele from IS8525, the resistant parent of the cross.

For pq, a total of five QTLs on LGs 4, 5, 6, 11 and 12 were identified. The phenotypic variation explained by each of the loci varied from 7.10-14.19% and in total explained 50% of the total variation for pq. The direction of individual allele effect for pq was not consistent and were both positive and negative. Thus both parents contributed beneficial alleles for QTLs of pq.

For pv, three QTLs on LGs 3, 6 and 12 were identified. The phenotypic variation explained by each of the loci varied from 6.46 to 9.51 and together, these three QTLs explained 23.60% of the total phenotypic variation for pv. The allele effect was both positive and negative. Thus both parents contributed beneficial alleles for QTLs of pv.

Conclusion

The main objective of the present study was to study the genetic basis of ergot resistance in sorghum. We have identified QTLs associated with ergot resistance and two components of ergot resistance, pollen quantity and pollen viability. While only one QTL of pcergot and pq appears to be linked, the majority do not, supporting the earlier findings that ergot resistance in IS8525 is not pollen-based. The markers identified in the present study can be used in marker-assisted selection programs for this important trait.

References

Basten C, Weir B, Zeng Z-B (2002). QTL Cartographer, Version 1.16. Department of Statistics, North Carolina State University, Raleigh, NC.

Bhattramakki D, Dong JM, Chhabra AK, Hart GE (2000). An integrated SSR and RFLP linkage map of Sorghum bicolor (L.) Moench. Genome 43, 988-1002.

Blaney B, McKenzie R, Walters J, Taylor L, Bewg W, Ryley M, Maryam R (2000). Sorghum ergot (Claviceps africana) associated with agalactia and feed refusal in pigs and dairy cattle. Australian Veterinary Journal 78, 102-107.

Churchill G, Doerge R (1994). Empirical threshold values for quantitative trait mapping. Genetics 138, 963-971.

Dahlberg JA, Bandyopadhyay R, Rooney WL, Odvody GN, Madera-Torres P (2001) Evaluation of sorghum germplasm used in US breeding programmes for sources of sugary disease resistance. Plant Pathology 50, 681-689.

Kosambi D (1944). The estimation of map distances from recombination values. Annal Eugenetics 12, 172-175.

Manly K, Cudmore R, Meer J (2001). Map manager QTX, cross-platform software for genetic mapping. Mammalian Genome 12, 930-932.

Menz MA, Klein RR, Mullet JE, Obert JA, Unruh NC, Klein PE (2002). A high-density genetic map of Sorghum bicolor (L.) Moench based on 2926 AFLP (R), RFLP

and SSR markers. Plant Molecular Biology 48, 483-499.

Smith A, Cullis B, Thompson R (2001). Analyzing variety by environment data using multiplicative mixed models and adjustments for spatial field trend. Biometrics 57, 1138-1147.

Wang E, Meinke H, Ryley M (2000) Event frequency and severity of sorghum ergot in Australia. Australian Journal of Agricultural Research 51, 457-66.

Zeng Z-B (1994). Precision mapping of quantitative trait loci. Genetics 136, 1457-1465.

Top Of PageNext Page

...

Search Site