1

TITLE: Identification of RAPD Markers Linked to the Pm1, Pm2, Pm3a,

Pm3b, Pm3c, Pm4a Loci for Powdery Mildew Resistance in Wheat

AUTHORS: A. N. Shi, S. Leath*, and J. P. Murphy

ADDRESS: Steven Leath

Department of Plant Pathology

North Carolina State University

Raleigh, NC 27695-7616

A. N. Shi, Dep. of Plant Pathology, North Carolina State Univ., Raleigh, NC 27695-7616;

S. Leath, (USDA-ARS), Dep. of Plant Pathology, North Carolina State Univ., Raleigh, NC 27695-7616; J. P. Murphy, Dep. of Crop Science, North Carolina State Univ., Raleigh, NC 27695-7629. Received_________________

*Corresponding author ().

Abbreviations: Bgt, Blumeria graminis f. sp. tritici; Random amplified polymorphic DNA; NILs, Near-isogenic lines; CI, Confidence interval.


ABSTRACT

Random amplified polymorphic DNA (RAPD) markers linked to major genes for resistance to powdery mildew (Blumeria graminis f. sp. tritici = Erysiphe graminis f. sp. tritici) in wheat (Triticum aestivum L.) were identified using a series of 'Chancellor' near-isogenic lines (NILs) with the Pm1, Pm2, Pm3a, Pm3b, Pm3c, and Pm4a alleles. Forty-one markers revealed specific polymorphisms: sixteen for the susceptible recurrent parent Chancellor, four for Pm1, one for Pm2, three for Pm3, six for Pm3a, seven for Pm3b, three for Pm3c, and one for Pm4a. The marker OPU17750 was linked to Pm1 (2.2 ± 1.07 cM) in coupling phase with the resistant phenotype in the F2 population from the cross between NK-Coker 68-15 and CP4 (NK-Coker 68-15*6//CI13836/8*Cc). The marker OPAN71400 was linked to Pm3b (1.2 ± 0.85 cM ) in coupling phase with the resistant phenotype in the Chancellor//Chul/8*Cc F2 population. The markers, OPN09600, OPN091200, and OPS031400, were linked to the Pm3 locus in repulsion phase with the resistant phenotype in the Chancellor//Chul/ 8*Cc F2 population and in the NK-Coker 68-15/Saluda F2 population (0.00 - 0.15 cM). The three markers were linked (1.00 ± 0.718 cM) to a recessive allele resistant to isolate E215 in coupling phase with the resistant phenotype in the NK-Coker 68-15/Saluda F2 population.


Powdery mildew of wheat ( Blumeria graminis (DC.) E.O. Speer f. sp. tritici Em. Marchal, Bgt = Erysiphe graminis DC. ex Merat f. sp. tritici Em. Marchal) causes yield loss every year in many wheat producing regions. Single gene resistance is the primary method of control and genes for resistance are routinely overcome by new Bgt isolates (Leath & Murphy, 1985; Menzies et al, 1986; Heun, 1987). The common management strategy has been to replace cultivars when their major gene resistance is no longer effective ( Wolf, 1984; Leath & Heun, 1990). Molecular markers tightly linked to disease resistance genes can provide breeders with a tool for marker-assisted selection of resistance genes and gene pyramiding in plants (Stuber, 1992; Michelmore, 1995).

Identification of molecular markers linked to major genes can be based on a series of backcross-derived near-isogenic lines (NILs) or bulked segregant analysis (BSA) (Michelmore, 1995). Six ‘Chancellor’ near-isogenic lines carrying single powdery mildew (Pm) resistance alleles, Pm1, Pm2, Pm3a, Pm3b, Pm3c, and Pm4a, have been developed (Briggle, 1969). Such lines have greatly facilitated the identification of RFLP markers associated with the major Pm genes, Pm1, Pm2, Pm3a, Pm3b, Pm3c, and Pm4a (Hartl et al. 1993, 1995; Ma et al. 1994). Random amplified polymorphic DNA (RAPD) technology has advantages over RFLPs, such as simplicity, speed, cost and absence of radioactivity (Williams et al. 1990). The objective of this research was to identify RAPD markers linked to Pm1, Pm2, Pm3a, Pm3b, Pm3c, and Pm4a using a series of 'Chancellor' near-isogenic lines carrying these Pm genes. A preliminary report has been published (Shi et al. 1995).


MATERIALS AND METHODS

Plant Materials

1. Initial screening for markers with near-isogenic lines: Six ‘Chancellor’ (Cc) near-isogenic lines (NILs): Axminster/8*Cc, Ulka/8*Cc, Asosan/8*Cc, Chul/8*Cc, Sonora/8*Cc, and Khapli/8*Cc, carrying Pm1, Pm2, Pm3a, Pm3b, Pm3c, and Pm4a, respectively (Briggle, 1969), plus the recurrent parent which contains no major Pm resistance gene, were used to screen primers for RAPD markers linked to these Pm genes.

2. Estimating linkage between RAPD markers and the Pm1 and Pm3b alleles and the Pm3 locus: An F2 population from the cross NK-Coker 68-15/CP4 was developed to determine linkage with Pm1. CP4 (NK-Coker 68-15*6//CI13838/8*Cc) is a component of the ‘NK-Coker 68-15’ near-isogenic line series currently under development. CI13836/8*Cc is a 'Chancellor' near-isogenic line with the Pm1 allele (Briggle, 1969).

F2 populations from the crosses Chancellor//Chul/8*Cc and NK-Coker 68-15/Saluda were developed to determine linkage between markers and the Pm3 locus and the Pm3b allele. Chul/8*Cc is a 'Chancellor' near-isogenic line with the Pm3b allele (Briggle 1969); Saluda has the Pm3a allele (Leath and Heun, 1990).

Three BC1F1 populations from the crosses Chancellor*2//Chul/8*Cc, NK-Coker 68-15*2/Saluda, and Saluda*2/NK-Coker 68-15 were developed to test the genetic expression of resistance to Bgt isolates, and verified that the Pm3b allele was in Chul/8*Cc and that the Pm3a allele was in Saluda,

3. Detection of RAPD markers for Pm1 and Pm3 alleles in wheat lines: the markers, which displayed specific polymorphic bands for Pm1, Pm3b, or other Pm3 alleles, were further tested in a series of wheat lines, which contained Pm1, Pm3b, or other Pm3 alleles (Pm3a, 3c, 3d, 3f), and their F1 progenies. The differential lines with other Pm alleles from Pm1 to Pm21 (except Pm10, Pm11, Pm14, and Pm15), and two susceptible lines without known Pm alleles, were included as controls.

All crosses and generation advances were conducted in the greenhouse during the 1993-94 and 1994-95 winter seasons.

Powdery Mildew Evaluation

Powdery mildew evaluations were performed using a detached leaf technique (Leath and Heun, 1990). Assessment of reaction was based on a descriptive scale of resistant (0-3), intermediate (4-6), and susceptible (7-9) reaction types (Leath and Heun, 1990). Two to four Bgt isolates, characterized for virulence, were used to test each segregating population.

RAPD Marker Analysis

DNA Extraction and RAPD Assay

Genomic DNA was extracted from fresh wheat leaves (Doyle and Doyle, 1990). Random 10-mer primers were obtained from Operon Technologies Inc. (Alameda CA). The PCR procedure described by Williams et al. (1990) was followed with minor modifications. Each reaction consisted of 2.4 ul reaction buffer mix, 1.2 ul dNTPs (2.5mM), 5 ul primer (4 or 5 ng/ul), 0.2 ul Taq polymerase (5u/ul), 1.2 ul unacetylated bovine serum albumen (BSA), and 5.0 ul (4 or 5 ng/ul) genomic DNA. A total of 41 cycles of PCR amplification were performed using a standard RAPD program with denaturation at 92 C for 1 min, annealing at 35 C for 1 min, and extension at 72 C for 2 min. The reactions were visualized on 1.2-1.5% agarose gels in 1X TBE.


Primer Screening

A total of 332 random Operon primers were used to screen for RAPD markers in the six ‘Chancellor’ near-isogenic lines with the Pm1, Pm2, Pm3a, Pm3b, Pm3c, Pm4a alleles and the recurrent parent Chancellor. An additional 104 random primers were used to screen RAPD markers in the four ‘Chancellor’ near-isogenic lines with the Pm2, Pm3a, Pm3c, and Pm4a alleles.

Linkage Analysis

Linkage of loci between resistance alleles or between markers and resistance alleles segregating in the F2 population from the crosses NK-Coker 68-15/CP4, Chancellor// Chul/8*Cc, or NK-Coker 68-15/Saluda was analyzed based on the maximum likelihood estimator (Allard, 1956; Liu, 1997), and calculated by use of a SAS program (SAS Institute Inc., 1990), kindly provided by Dr. Ben-Hui Liu, Department of Forestry, North Carolina State University, Raleigh (unpublished).

The recombination frequency was transformed according to the Kosambi function using the formula cAB = ¼ ln[(1+2r)/(1-2r)] x 100 (Weir, 1996), where cAB is the map distance (cM). r is estimated recombination frequency. The standard deviation Sr was calculated using Sr = Ö [r(1-r)(1-2r+2r2)/2N(1-3r+3r2)] in a F2 population, and the confidence interval (CI) of the estimated recombination frequency was obtained using a log-likelihood approach (Liu, 1997). Goodness-of-fit and independence tests were carried out using an c2 (Liu, 1997).


RESULTS

RAPD Markers From NILs Analysis

Thirty-five out of 436 primers revealed forty-one specific polymorphisms: sixteen for the susceptible recurrent parent Chancellor, four for the Pm1 allele, one for the Pm2 allele, three for the Pm3 locus, six for the Pm3a allele, seven for the Pm3b allele, three for the Pm3c allele, and one for the Pm4a allele (Table 1).

The polymorphic bands produced by seven primers, OPE11, OPM06, OPS02, OPS04, OPAG08, OPAK12, and OPAK18, were present in the susceptible recurrent parent Chancellor only, and polymorphic bands produced by another seven primers, OPH01, OPT11, OPU09, OPAE06, OPAF12, OPAF16, and OPAG10, were present in the six NIL’s, but were absent in the susceptible recurrent parent (Table 1, Fig. 1a, 1b). Four primers, OPU17, OPAC08, OPAI02, and OPAL4, revealed polymorphisms associated with the Pm1 allele (Table 1, Fig. 1c). A marker OPT16850 was found to be associated with the Pm2 allele in Ulka/8*Cc among the seven NILs (Table 1, Fig. 1d). Two primers, OPN09 and OPS03, revealed three polymorphisms for alleles at the Pm3 locus (Table 1, Fig. 1e). Four primers, OPB19, OPB20, OPS11, and OPS13, revealed six polymorphisms for Pm3a (Table 1, Fig. 1f). Seven primers, OPD01, OPH19, OPS15, OPT15, OPAD16, OPAJ06, and OPAN07, displayed polymorphisms associated with the Pm3b allele (Table 1, Fig. 1g). Two primers, OPT09 and OPT12, exhibited three polymorphisms for Pm3c (Table 1, Fig. 1h). A primer, OPH10, revealed polymorphism for Pm4a with the band at about 1200-bp (Table 1, Fig. 1i). The RAPD markers specific for Pm2, Pm3a, Pm3c, and Pm4a need be further analyzed fully to characterize their position and reliability.

Linked Markers for Pm1

The segregation of Pm1 resistance to four Bgt isolates, E314, E215, 209a2, and Yuma, was tested in the parental, F1, and F2 generations from the cross NK-Coker 68-15/CP4. The plants in the segregating F2 population were divided into resistant and susceptible groups, and the segregation ratios of resistance to all four isolates fit a 3R:1S expected ratio for single locus with dominant gene action (Table 2).

The OPU17750 marker was stable and reliable not only for Pm1 in the NILs, but also in the F2 population (Fig. 2). OPU17750 was linked to Pm1 in coupling phase (2.2 ± 1.07 %), and the 95% confidence interval (CI) ranged from 0.34 to 6.78 % (Table 3).

Linked Markers for Pm3b and other Pm3 alleles

The segregation of resistance to four isolates E314, E215, 216a, and Yuma of Bgt was tested in the parental, F1, F2, and BC1F1 generations of the cross Chancellor// Chul/8*Cc. The segregation ratio of resistance to isolate E314 fit a 3R:1S in the F2 and a 1R:1S in the BC1F1 populations consisted with a single locus with dominant gene action, but was not in accordance with a single gene action for resistance to other isolates (Table 2).

The segregation of Pm3a resistance to two Bgt isolates was tested in the parental, F1, F2, and BC1F1 generations of the cross NK-Coker 68-15/Saluda. The segregation ratio of resistance to isolate E314 of Bgt fit a 3R:1S in the F2, and a 1R:1S in the susceptible parent BC1F1 generations, and a 1R:0S in the resistant parent BC1F1 population consistent with a single locus with dominant gene action (Table 2). The segregation ratio of resistance to E215 fit a 1R:3S in the F2, and a 1R:1S in the resistant parent BC1F1 population consistent with a single locus and single recessive gene action, but did not fit a all susceptible expected segregation ratio in the susceptible parent BC1F1 population (Table 2).

The marker OPAN071400 was stable and reliably revealed polymorphisms, not only for Pm3b alone in the NILs, but also in the Chancellor//Chul/8*Cc F2 population (Fig.3). OPAN71400 was closely linked to the Pm3b resistance allele in coupling phase in the Chancellor//Chul/8*Cc F2 population. The recombination frequency between Pm3b and the marker was 1.2 ± 0.85 %, and the 95% CI ranged from 0.06 to 5.36 % (Table 3).

The three markers, OPS031400, OPN09600, and OPN091200, were stable and reliably revealed polymorphisms, not only between the seven NILs for the Pm3 alleles (Pm3a, 3b, 3c), but also in the Coker 68-15/Saluda F2 population (Fig. 4, 5). The three markers, OPN09600, OPN091200, and OPS031400, were found to be linked to the Pm3b allele (0.00 - 0.12 %) in repulsion phase with the phenotype for resistance to isolate E314 in the Chancellor//Chul/8*Cc F2 population, and the 95% CI ranged from 0.00 to 26.72 % (Table 3). The three markers also were found to be linked to the Pm3a allele (0.00 - 0.10 %) in repulsion phase with the phenotype of resistance to isolate E314 in NK-Coker 68-15/Saluda F2 population. The 95% CI for the Pm3a allele and either OPN09600, or OPN09600 was 0 to 23.61 %, and between the allele and OPS031400 was 0 to 24.10 % (Table 4). The three markers were also found to be linked in coupling phase to a recessive allele resistant to isolate E215 in the NK-Coker 68-15/Saluda F2 population (1.00 ± 0.718 %). The 95% CI was 0.05 - 4.44 % (Table 4).

Detection of RAPD markers for Pm1 and Pm3b alleles, and the Pm3 locus

Thirty-one wheat lines and five F1 hybrids were tested to detect the reliability of OPU17750. All six lines, Axminster/8*Cc, ASII/8*Cc, CI13836/8*Cc, CP4 (NK-Coker 68-15*6//CI13836/8*Cc), Normandie and Mephisto, with the Pm1 resistant allele alone, plus two lines with Pm1 resistant in combination with Pm2 and Pm9 alleles contained polymorphic bands of OPU17750. Line #31 containing Pm12 and BRG 3N/76 containing Pm16 also exhibited the OPU17750 marker. Twenty-three lines containing other Pm genes did not displayed the marker (data not shown). Five F1 progenies, NK-Coker 68-15/CP4, NK-Coker 68-15/Normandie, Chancellor/Normandie, Ulka/8*Cc//Normandie, and Normandie//Axminster/8*Cc, containing the resistant Pm1 allele in either heterozygous or homozygous condition scored positive for the marker.

The OPAN071400 marker was useful for detection of Pm3b. The two lines, Chul/8*Cc and T.sphaerococcum/8*Cc, with Pm3b alone and three F1 progenies, Chul/8*Cc // Asosan/8*Cc, Chul/8*Cc // Sonora/8*Cc, and Chancellor // Chul/8*Cc, containing Pm3b in the heterozygous condition scored positive for the marker in the twenty-four wheat lines with and without Pm3b and fourteen F1 progenies (data not shown).

Three markers, OPN09600, OPN091200, and OPS031400, were useful for detecting alleles at the Pm3 locus. Only the lines with Pm3 alleles (Pm3a, 3b, 3c,3d, 3f) and F1 progenies with homozygous Pm3 alleles displayed the markers in the twenty-four wheat lines with and without Pm3 resistant alleles and in fourteen F1 progenies (data not shown).