Previous PageTable Of ContentsNext Page

A Genomic Region Associated with Aluminium Tolerance in Barley

Harsh Raman1, S. Moroni1, R. Raman1, A. Karakousis2, B. Read1 K. Sato3 and B.J. Scott1

1Wagga Wagga Agricultural Institute NSW Agriculture, Wagga Wagga NSW 2650, Australia
2
University of Adelaide, Waite Campus, Adelaide
3
Research Institute for Bioresources, Okayama University, Kurashiki 710-0046, Japan

Abstract

To understand the genetic control of Al tolerance, molecular mapping of genes conferring Al tolerance has been initiated in various crops including barley. Previous work has shown that the Al tolerance gene (Alp2) in Yambla/WB229 population is located on the long arm of chromosome 4H and a similar genomic region is involved for tolerance (Alp) in Dayton/Harlan Hybrid population. We investigated whether Brindabella has the same locus conferring aluminium tolerance. An F2 population of Harrington/Brindabella was screened for aluminium tolerance and the results indicated that a major QTL, controlled by the Al tolerant parent Brindabella, was identified on 4H. Microsatellite markers Bmac186 and Bmac310 showed a highly significant association with the Al tolerance gene and therefore can be used as markers for selection of the Al tolerance gene from Brindabella (Alp3). It appears that homologous genomic regions may be involved in aluminium tolerance in barley. High PIC values (0.44 to 0.66) and % difference (45 to 89%) among 40 lines that are extensively used in Australian barley breeding program indicated that microsatellite markers can be used to select Al tolerance derived from Brindabella in the different backgrounds.

Introduction

Development of Al tolerant cultivars is a one of the strategies to enable cultivation of barley on acid soils. There is considerable genetic variation for Al tolerance in barley and this has been exploited by conventional breeding methods (Stølen 1972, Read et al 1988) to improve productivity and quality of barley growing on acidic soils. However, the genetic basis of Al tolerance is not well understood. The number of Al tolerance genes that may exist in Hordeum species is unclear, as is their uniqueness in relation to other species. It has been shown that a single major gene controls Al tolerance in barley populations developed from Yambla/WB229, Mimosa/WB229 (Raman et al in press), Dayton/Harlan Hybrid (Minella and Sorrells 1997, Tang et al 2000), Dayton/F6ant and Dayton/Kearney (Raman et al 2001).

To understand the genetic control of Al tolerance and to determine the homology of Al tolerance genes, molecular mapping has been initiated in various crops including barley (Tang et al 2000, Raman et al in press, Raman et al 2001), wheat (Reide and Anderson 1995), rye (Gallego et al 1998), rice (Nguyen et al 2001), maize (Rhue et al 1978) and soybean (Bianchi et al 2000). The Al tolerance gene in theYambla/WB229 population is located on the long arm of chromosome 4H and a similar genomic region is involved for tolerance in Dayton/Harlan Hybrid (Raman et al 2001). Recently, comparative molecular mapping of a major Al tolerance gene in Dayton has revealed that this gene is likely to be the functional ortholog of a major Al tolerance gene located on the long arm of chromosome 4D in wheat (Tang et al 2000). In rye, chromosome 4R confers tolerance to Al (Aniol and Gustafson 1984), and 4R is orthologous to wheat 4D. However, other genomic regions on 5AS, 2DL in wheat and 3R and 6RS in rye have been implicated in the control of Al tolerance (Aniol and Gustafson 1984, Aniol 1990).

Methods to select Al tolerant genotypes in the field, in glasshouses and in nutrient solution have been developed (Blum 1988, Scott and Fisher 1989, Raman et al 2001) and correlations between different screening methods and field performance have been made. In barley selection based on haematoxylin staining for Al tolerance was unsatisfactory. Selection of Al tolerant individuals using nutrient solution culture is not efficient for handling large breeding populations. The Al tolerance in Australian cultivar Brindabella is effective on acidic soils where the cultivar yields 3 times more than the Al sensitive malting quality cultivar Schooner (Read et al 1988). One of the key objectives of our Australian barley breeding program is to develop malting cultivars suitable for cultivation on acidic soils. Therefore, introgression of Al tolerance gene(s) into malting quality genotypes such as Harrington is being carried-out at NSW Agriculture. Indirect selection for Al tolerance using marker-assisted selection may prove to be an excellent substitute over traditional phenotypic selection and could improve the efficiency of the conventional plant breeding, especially as markers are not affected by environment and can be detected at all stages of plant development. Hence, an investigation was undertaken to (1) develop molecular markers linked to Al tolerance (2) establish the chromosomal location of Al tolerance gene and (3) determine the uniqueness of Al tolerance loci among different Al tolerant barleys.

Materials and Methods

Plant Material: The response of barley to Al was investigated with an F2 population derived from a single F1 plant derived from Harrington/Brindabella. Harrington is a 2-rowed malting barley cultivar from Canada (Harvey and Rossnagel, 1984). Brindabella is a 2-rowed barley cultivar from Australia and is moderately tolerant to Al (Read and Oram 1995). The F2 population of 64 seedlings was phenotyped for Al tolerance in a solution culture assay.

Phenotyping for Al tolerance: Barley seedlings were screened for Al tolerance in a solution culture system using a modified Al pulse-recovery method (Berzonsky and Kimber 1986) as described earlier (Raman et al. in press; Method 3). Seeds were germinated overnight as described by Raman et al (2001, this issue). Four days after germination an Al-pulse treatment (40 μM) was superimposed over the basal nutrient solution and maintained for 24 hrs. Five days after germination the roots were stained with a haematoxylin solution (Polle et al 1978) prior to transferring them into the recovery solution containing Al (15μM). The root staining marked the position from which root re-growth occurred during the recovery period. The selected lower Al concentration used during the recovery stage allowed for maximum differences between parents and was determined from previously developed Al dose response curves (data not presented). All nutrient solutions were prepared with deionised water, constantly aerated and were adjusted to pH 4.3 with HCl. The F2 population and both parents were included during the assay as controls. During the recovery stage (6 to 8 days after germination) daily measurements of the root length were conducted consecutively for 3 days to determine the root extension rate (RER) of each individual seedling. Analyses were performed using PROC REG (regression procedure) of the statistical software SAS/STATS (SAS Institute, 1989). Ranking of RER was used to classify the F2 seedlings as tolerant, sensitive or intermediate based on response to Al.

Molecular Mapping and Linkage Analysis: Microsatellite markers were synthesised by Life Technology Gibco-BRL and were used to map the locus conferring Al tolerance in cv. Brindabella. DNA extraction, PCR amplification and electrophoresis were carried-out as described by Raman et al (2001). The Al tolerance loci Alp2 in WB229 and Alp in Dayton have been shown to be located on chromosome 4HL (Minella and Sorrells 1992), so microsatellite markers previously mapped on 4H were used to establish any linkage between the trait and marker. Linkage analysis was also performed on F2 segregation data for QTL mapping using the interval mapping function by a regression procedure (Harley and Knott 1992, Martinez and Curnow 1992) using Map Manger QTX011b software. Permutation tests (Doerge and Churchill 1996) were carried-out on associations that identified the QTLs for 1cM and 500 iterations and an association between the marker and Al tolerance was identified at 99.9% significance (LRS=17).

Allele Diversity of Microsatellites: The usefulness of closely linked markers was investigated using 40 barley lines (Sloop, WI2976, DH115, Gairdner, Pitcher, WI3102, TR129, TG-Harrington, WI3099, WI3140, WI3155, VB9728, VB9729, VB9613, WI3148, 117/50, Amaji Nijo, Chariot, WI3141, Kredit, Skiff, WI3139, Monarch, VB9622, AB271, Milka, WI3380, WI3385, Fanfare, WI3387, Gleam, WI3389, WB223, Tilga, Angora, Clarity, Riviera, Caminant, Cork, and Scarlett by using PIC values and % difference as criteria (% lines with different allele(s) than those linked to Al tolerance in Brindabella).

Results

The pulse-recovery assay enabled the discrimination of Al tolerant, sensitive and intermediate genotypes. During the recovery stage, the Al tolerant cv. Brindabella showed extensive seminal root re-growth as well as growth of lateral roots while the cv. Harrington showed very little root re-growth. Cultivar Brindabella was scored as Al tolerant (RER 4.25 ± SE 0.12) while Harrington was scored as Al sensitive (2.82 ± 0.05); thus, the assay was able to discriminate clearly between both parents as well as the F2 population for Al tolerance. The regression coefficients (R2) for the estimated RERs were greater than 0.95 indicating that root re-growth was linear and therefore data transformation for analysis was not necessary. Based on the estimated RERs, the 64 F2 plants of Harrington/Brindabella cross were classified as follows: 15 were scored as Al sensitive (RER 1.8 to 3.1 mm day-1), 31 as intermediate in tolerance to Al (RER 3.3 to 4.3 mm day-1) and 18 as tolerant to Al (RER 4.5 to 5.8 mm day-1). This distribution (Fig1) fitted a 1:2:1 Mendelian ratio for monogenic segregation (χ2 = 0.344 p=0.842). This indicated that a single gene controlled Al tolerance in the F2 progeny developed from the Harrington/Brindabella. In the mapping population, only one plant had RER 1.8, which could be perhaps due to the poor seed quality.

Fig1: Frequency Distribution of F2 plants Screened for Al Tolerance Using Nutrient Solution

Among the different microsatellite markers screened, HVM3, HVM67, Bmac310, Bmac186, EBmac906, and Bmag375 were polymorphic and segregated co-dominantly in the F2 mapping population into 1:2:1 ratio. Highly significant associations between microsatellite markers Bmac186 and Bmac310 and Al tolerance derived from Brindabella were identified. Thus microsatellite mapping in the Harrington/Brindabella population identifies the map location of the Al tolerance locus on the long arm of 4H, as reported previously in Dayton/Harlan Hybrid (Minella and Sorrells 1992, Tang et al 2000) and Yambla/WB229 (Raman et al, in press). The usefulness of an SSR marker for screening of this trait with respect to the donor parent is summarised in Table 1 and presented as % difference (with respect to the alleles of the 40 barley accessions).

A positive correlation between the number of alleles and PIC values was found. However, the same was not true with % difference. The results revealed that HVM68, Bmag353 and Bmac310 microsatellite that map very close to Bmag186 (Ramsay et al 2001) are highly divergent (Table1) among the lines screened and should detect QTL associated with Al tolerance from derivatives of these lines and cv. Brindabella.

Table 1: Allele Variations, and Polymorphic Informative Content (PIC) of Microsatellites Linked with Al Tolerance in Barley

Trait

Donor variety

SSR marker

Allele Range (bp)

No Alleles

PIC

% Difference

Aluminium Tolerance (4HL)

Brindabella

HVM68

187-223

5

0.48

89

   

Bmag353

106-126

4

0.44

75

   

Bmac310

143-182

6

0.66

45

Discussion

This work indicated that the control of an Al tolerance gene in cv. Brindabella is located in the long arm of 4H, as reported earlier in Yambla/WB229 (Raman et al, in press). The locus identified here may be the same as the Pht locus that confers tolerance to low soil pH in barley (Stφlen and Anderson 1978). The chromosome 4 of barley (4H) is homologous to wheat chromosomes 4B and 4D, as it differs by a paracentric inversion in the distal part of the long arm (Dubcovsky et al 1996). Since we do not know whether the loci Alp in Dayton and Alp2 in WB229 and the Al tolerance locus in Brindabella are the same, we designate this locus as Alp3. It is also possible that (a) there may be an allelic series of genes at one locus on chromosome 4H and (b) a single gene may operate a number of tolerance mechanisms. The Alp gene in Dayton may be different to that in the other Al tolerant genotypes (WB229 and Brindabella) if Al tolerance is conferred by an allelic series of this gene. These alleles could not be used to confer added tolerance if they map on same locus. Furthermore, the locus resides very close to the centromere, therefore the chances of recombination of these different alleles (if any) may be very low. We have not done any allelism testing so far; hence we do not know whether there is more than one locus for Al tolerance in barley.

Our results indicate that a homologous region may be involved in control of Al tolerance in barley and common linked molecular markers (RFLPs and microsatellites) developed in Yambla/WB229, Dayton/Harlan and Harrington/Brindabella populations can be used for indirect selection of aluminium tolerance in barley breeding programs. Marker-based selection (MAS) allows screening of germplasm at early stages of plant development and can effectively replace selection based on the phenotypic Al tolerance expression particularly in the crosses where there are not large detectable differences in root re-growth. In addition, environmental factors such as seed size, physiological maturity of seed and aluminium concentration could result in misclassification of the Al tolerance trait. Hence MAS can eliminate such inherent problems associated with phenotype based selection and can be used to transfer this gene for Al tolerance, avoiding the masking effect of environment. MAS performed in the early generations can increase the frequency of traits in the advanced breeding populations. The microsatellite markers closely linked with Al tolerance showed high polymorphic information content among different breeding lines and are co-dominant. Hence, they are able to detect heterozygotes which are very useful especially in backcrossing programs.

Acknowledgments

The work has been financed by the NSW Agriculture through Acid Soil Action, and the Grains Research and Development Corporation Australia. Harsh is thankful to Drs Peter Langridge, Andy Barr and Shoba Venkatanagappa for their support and comments.

References

1. Aniol A (1990) Plant Soil 123:223-227

2. Aniol A and Gustafson J.P. (1984). Can J Genet Cytol 26:701-705

3. Berzonsky WA, and Kimber G (1986) Plant Breed 97: 275-278

4. Minella E. and Sorrells M.E. (1992). Crop Sci 32:593-598

5. Raman H., Moroni J.S., Sato K., Read B.J., Scott, B.J. Theor and Appl Genet (in press).

6. Read B.J., and Oram R.N. (1995) Australian Journal of Experimental Agriculture 35:425

7. Stølen O. (1972) Royal Vet & Agric. In: Copenhagen Yearbook, pp 1-13

8. Tang, Y., Sorrells, M.E., Kochian, L.V. and Garvin D.F. (2000). Crop Sci.40,778-782

Previous PageTop Of PageNext Page