Haplotyping and mapping a large cluster of downy mildew resistance gene candidates in sunflower using multilocus intron fragment length polymorphisms

Abstract

Downy mildew (Plasmopara halstedii (Farl.) Berlese et de Toni) is a serious foliar pathogen of cultivated sunflower (Helianthus annuus L.). Genetic resistance is conditioned by several linked downy mildew resistance gene specificities in the HaRGC1 cluster of TIR-NBS-LRR resistance gene candidates (RGCs) on linkage group 8. The complexity and diversity of the HaRGC1 cluster was assessed by multilocus intron fragment length polymorphism (IFLP) genotyping using a single pair of primers flanking a hypervariable intron located between the TIR and NBS domains. Two to 23 bands were amplified per germplasm accession. The size of the included intron ranged from 89 to 858 nucleotides. Forty-eight unique markers were distinguished among 24 elite inbred lines, six partially isogenic inbred lines, nine open-pollinated populations, four Native American land races, and 20 wild H. annuus populations. Nine haplotypes (based on 24 RGCs) were identified among elite inbred lines and were correlated with known downy mildew resistance specificities. Sixteen out of 39 RGCs identified in wild H. annuus populations were not observed in elite germplasm. Five partially isogenic downy mildew resistant lines developed from wild H. annuus and H. praecox donors carried eight RGCs not found in other elite inbred lines. Twenty-four HaRGC1 loci were mapped to a 2-4 cM segment of linkage group 8. The multilocus IFLP marker and duplicated, hypervariable microsatellite markers tightly linked to the HaRGC1 cluster are powerful tools for distinguishing downy mildew resistance gene specificities and identifying and introgressing new downy mildew resistance gene specificities from wild sunflowers.

Full text

Plant Biotechnology Journal (2003) 1, pp. 167–185
© 2003 Blackwell Publishing Ltd 167
Blackwell Publishing Ltd.
Haplotyping and mapping a large cluster of downy
mildew resistance gene candidates in sunflower using
multilocus intron fragment length polymorphisms
Mary B. Slabaugh1,*, Ju-Kyung Yu1,†, Shunxue Tang1, Adam Heesacker1, Xu Hu2, Guihua Lu2, Dennis Bidney2,
Feng Han2 and Steven J. Knapp1
1Department of Crop and Soil Science, Oregon State University, Corvallis, OR 97331, 2Pioneer Hi-Bred International, PO Box 1004, Johnston, IA 50131, USA
Summary
Downy mildew (Plasmopara halstedii (Farl.) Berlese et de Toni) is a serious foliar pathogen
of cultivated sunflower (Helianthus annuus L.). Genetic resistance is conditioned by several
linked downy mildew resistance gene specificities in the HaRGC1 cluster of TIR-NBS-LRR
resistance gene candidates (RGCs) on linkage group 8. The complexity and diversity of the
HaRGC1 cluster was assessed by multilocus intron fragment length polymorphism (IFLP)
genotyping using a single pair of primers flanking a hypervariable intron located between
the TIR and NBS domains. Two to 23 bands were amplified per germplasm accession. The
size of the included intron ranged from 89 to 858 nucleotides. Forty-eight unique markers
were distinguished among 24 elite inbred lines, six partially isogenic inbred lines, nine open-
pollinated populations, four Native American land races, and 20 wild H. annuus
populations. Nine haplotypes (based on 24 RGCs) were identified among elite inbred lines
and were correlated with known downy mildew resistance specificities. Sixteen out of 39
RGCs identified in wild H. annuus populations were not observed in elite germplasm. Five
partially isogenic downy mildew resistant lines developed from wild H. annuus and H.
praecox donors carried eight RGCs not found in other elite inbred lines. Twenty-four
HaRGC1 loci were mapped to a 2–4 cM segment of linkage group 8. The multilocus IFLP
marker and duplicated, hypervariable microsatellite markers tightly linked to the HaRGC1
cluster are powerful tools for distinguishing downy mildew resistance gene specificities and
identifying and introgressing new downy mildew resistance gene specificities from wild
sunflowers.
Introduction
Sunflower (Helianthus annuus L.) cultivars bred in Eastern
Europe were introduced to North American agriculture in the
first half of the 20th century and formed the elite genetic
base for the development of the crop (Putt, 1978; Rieseberg
and Seiler, 1990). Early on in the mid-20th century, following
an expansion in sunflower production in Canada and the
USA, the newly introduced cultivars were found to lack resist-
ance to downy mildew (Plasmopara halstedii (Farl.) Berl. et de
Toni), a serious foliar pathogen of sunflower (Zimmer and
Kinman, 1972). The outbreak of downy mildew exposed the
narrow genetic base of the crop and stimulated a search for
resistance in wild germplasm, a logical strategy because both
the host and pathogen are native to North America.
To date, 11 genes have been postulated to provide
resistance to one or more races of P. halstedii (Rahim et al.
2002). Among these, Pl1 and Pl2 confer resistance to race
1 (virulence phenotype 100) and races 1 and 2 (virulence
phenotype 300), respectively, and trace to early crosses to
‘Texas Wild’ (Putt and Sackston, 1957). The Pl5 gene from H.
tuberosus, a perennial hexaploid sunflower, protects against
race 3 (virulence phenotype 700) (Leclercq et al., 1970;
Pustovoit, 1966). Resistance to races 1, 2, 3 and 4 (virulence
phenotype 730) was introgressed from three sunflower
species, H. annuus (Pl6), H. praecox ssp. runyonii Heiser (Pl7)
Received 11 September 2002;
revised 23 December 2002;
accepted 6 January 2003.
Correspondence (fax +1 541 737 1334;
e-mail mary.b.slabaugh@orst.edu)
†Present address: Department of Plant
Breeding, Cornell University, Ithaca,
NY14853, USA
Keywords: Helianthus annuus,
disease resistance, downy mildew, TIR-
NBS-LRR, intron diversity, resistance
gene candidate

Mary B. Slabaugh et al.
© Blackwell Publishing Ltd, Plant Biotechnology Journal (2003), 1, 167–185
168
and H. argophyllus (Pl8) (Miller and Gulya, 1988, 1991).
Although these dominant genes confer functionally complete
resistance to one or more races of P. halstedii, new pathovars
continue to emerge. This threat, combined with the recent
development of resistance to chemical control (Gulya et al.,
1999), has spurred a vigorous search for determinants specific
to new races and for genes conferring broad resistance
specificities.
Genetic evidence suggests that at least three downy
mildew resistance genes, Pl1, Pl2 and Pl6, are clustered in the
sunflower genome (Mouzeyar et al., 1995; Roeckel-Drevet
et al., 1996; Vear et al., 1997). Linkage studies individually
placed these genes on linkage group 1 (LG1) of the CARTISOL
RFLP map of Gentzbittel et al. (1995), at approximately the
same distance from previously mapped RFLP markers. Pl6
was found to consist of at least two linked (0.6 cM) genetic
factors (Vear et al., 1997). Recently, a second linkage group
was found to contain two clustered resistance genes, Pl5 and
Pl8 (Bert et al., 2001). Neither the molecular structure nor the
mode of action of any of these determinants is known.
The mapping of resistance gene candidates (RGCs)
produced by degenerate PCR also points to a clustering of
disease resistance genes in the Pl1-Pl2-Pl6 region. Gedil et al.
(2001a) placed Pl1 and six RGCs of the NBS-LRR type on
the HA370 × HA372 public sunflower map. An RGC of the
TIR-NBS-LRR subclass, which was originally designated
Ha-4W2 but herein named HaRGC1, was linked to Pl1 on
linkage group 8 (LG8) of the public map (Gedil et al., 2001a).
LG8 corresponds to LG1 of the CARTISOL map (Gedil et al.,
2001b). When primers specific for HaRGC1 were used
to amplify DNA from the Pl6-containing lines HA335 and
HA336 and the Pl7-containing lines HA337, HA338 and
HA339, all five produced the same set of four to five frag-
ments, none of which was present in the downy mildew
susceptible recurrent parent HA89. Gentzbittel et al. (1998)
also mapped a PCR fragment with high homology to
HaRGC1 near the Pl1 region. Thus, the Pl1-Pl2-Pl6 region on
LG8 of sunflower is a complex locus populated with RGCs
of the TIR-NBS-LRR class and multiple disease resistance
determinants.
The clustering of resistance genes in plants has been
observed in Arabidopsis (Holub, 2001), soybean (Kanazin
et al., 1996), cereals (Leister et al., 1998), Solanaceae (Grube
et al., 2000) and lettuce (Meyers et al., 1998). Arrays of
homologous sequences provide genetic material from which
new variants can be readily generated by gene conversion
and nonallelic homologous recombination between para-
logues (Fluhr, 2001; Stankiewicz and Lupski, 2002), processes
that were recently documented in lettuce (Chin et al., 2001).
The deployment of resistance genes in structural arrange-
ments that favour their dynamic evolution is consistent with
their presumed role as sentinels that recognize invading
pathogens or pathogen–target complexes and activate plant
defence responses (Van der Hoorn et al., 2002).
The overall goals of the present study were: (i) to charac-
terize intron fragment length polymorphisms (IFLPs) for
the identification and mapping of HaRGC1 downy mildew
resistance gene candidates, and (ii) to use these markers to
assess the diversity of a resistance gene cluster in sunflower.
In this report, we describe the mapping of HaRGC1 genes in
three segregating populations and the integration of the
HaRGC1 maps into the simple sequence repeat (SSR) map of
cultivated sunflower (Tang et al., 2002; Yu et al., 2003). We
also report the use of IFLP fingerprinting to probe the
complexity of the HaRGC1 R gene cluster across a wide range
of domesticated and wild sunflower germplasm and assess
the utility of IFLP fingerprinting for predicting resistance gene
specificities in sunflower.
Results
Primers flanking an intron in TIR-NBS-LRR genes
amplify a large family of resistance gene candidates
Primers flanking an intron between the TIR- and NBS-
encoding regions of the HaRGC1 sequence (Figure 1)
produced evidence of RGC paralogues in which intron lengths
varied from ≈ 90 to > 800 nucleotides (Figure 2). PCR products
amplified using primer pair F97/R98 (Figure 2, upper panel)
appeared to be a subset of the bands produced with primer
Figure 1 Multilocus intron fragment length polymorphism RGC
amplification strategy. PCR primers flanking an intron site in sunflower
partial cDNA Ras 5–1 (top) and the PCR products amplified by primer pairs
F97/R98 and F97/R134 (centre) are shown relative to a canonical
TIR-NBS-LRR sequence (bottom). TIR, Toll Interleukin Repeat; NBS,
nucleotide binding site; LRR, leucine rich repeat.

Resistance gene cluster in sunflower
© Blackwell Publishing Ltd, Plant Biotechnology Journal (2003), 1, 167–185
169
pair F97/R134 (Figure 2, lower panel), but longer by ≈ 90 bp,
close to the 86 bp distance between the R98 and R134 bind-
ing sites in the reference sequence (Figure 1). Primer pair F97/
R134 produced as many as 15 bands on ethidium bromide
stained agarose gels. Because F97 was a single-sequence
primer and R134 was moderately degenerate (16-fold), we
concluded that these primers amplified a large and diverse
but highly homologous resistance gene family.
The DNA fingerprints produced by the F97/R98 or F97/
R134 primer combinations suggested that the HaRGC1
haplotypes and downy mildew resistance phenotypes might
be correlated. Four fertility restorer lines resistant to race 2
(RHA801, RHA274, RHA377 and PHA) displayed the same
pattern (Figure 2, lanes 5, 6, 7 and 8), whereas HA370, which
carries the Pl1 gene for resistance to race 1, displayed a
different pattern (lane 3). Partially isogenic lines (PILs) HA335
and HA336 (homozygous for Pl6) and HA337, HA338, and
HA339 (homozygous for Pl7) produced identical patterns
that included several RGCs not observed in other inbred lines
(lanes 11–15), whereas the fingerprint for RHA340
(homozygous for Pl8) resembled the recurrent parent HA89
(lanes 10 and 9, respectively), in agreement with earlier work
using independent HaRGC1-specific primers targeted to
a different part of the NBS-encoding region (Gedil et al.,
2001a). These results indicated that selection for downy
mildew resistance in HA335-339 resulted in the isolation of
lines that had acquired a common set of RGC paralogues
from wild H. annuus and H. praecox donors and that the wild
donors harboured unique HaRGC1 resistance genes.
HaRGC1 amplicons encode TIR-NBS-LRR gene
fragments
The DNA sequences of five F97/R98 RGC amplicons,
Rgc1.2(i105) (accession no. AY186604), Rgc1.12(i270)
(accession no. AY186605), Rgc1.20(i330) (accession no.
AY186606), Rgc1.21(i360) (accession no. AY186607), and
Rgc1.24(i384) (accession no. AY186608) (see Table S1,
Supplementary Material, for nomenclature <http://
www.blackwell-science.com/products/journals / suppmat / PBI /
PBI016/PBI016sm.htm>), confirmed that they were derived
from resistance genes belonging to the TIR-NBS-LRR subclass
as defined by Meyers et al. (1999). The PCR products
contained intact open reading frames encoding peptides
highly similar to the original HaRGC1 clone (Gedil et al.,
2001a), and each contained an intron at the predicted
location, within a few nts of the corresponding intron in
Arabidopsis RPP1 and RPP5 genes (Botella et al., 1998; Noel
et al., 1999). Length differences among the HaRGC1
amplicons were due primarily to intron sequence variation,
but a six-nucleotide insertion just upstream of the intron
sequences was present in Rgc1.2, Rgc1.20 and Rgc1.24.
The deduced amino acid sequences (106–108 amino acids)
included an invariant motif, GVGGGGKT TLASAAY, surrounding
the P-loop sequence (underlined). By searching the data-
bases, this sequence was found to be unique to a group of
sunflower NBS-LRR proteins that includes the HaRGC1 clones
and a collection of sequences recently amplified from the
sunflower lines HA335 and H52 that map to the same region
of the genome (Bouzidi et al., 2002). Pair-wise amino acid
identities among the sunflower HaRGC1-type sequences
ranged from 76 to 95%. A lettuce TIR-NBS-LRR gene
fragment, LsRgc4 (accession no. AF017754) was 59–67%
identical to the five RGCs we sequenced. The LsRgc4 family
is estimated to consist of ≈ 15 clustered paralogues based on
Southern blots (Shen et al., 1998). Thus, the HaRGC1 and
LsRgc4 families are likely homologous clades from two tribes
of the Compositae.
To determine whether same-sized PCR products amplified
from different inbred lines were homologues, we sequenced
Rgc1.20 from HA370, HA372, and RHA280, Rgc1.21 from
HA370, RHA801, PHA, and RHA280, and Rgc1.24 from
HA370 and RHA280. RGCs of the same size exhibited 99.5%
nucleotide sequence identity, indicating that they were likely
alleles.
Figure 2 PCR products produced using DNA isolated from 15 selected
inbred lines and primer pairs F97/R98, upper panel, and F97/R134, lower
panel. PCR products were separated on a 1.5% agarose gel and stained
with ethidium bromide. Lane M, molecular size markers.

Mary B. Slabaugh et al.
© Blackwell Publishing Ltd, Plant Biotechnology Journal (2003), 1, 167–185
170
IFLP fingerprinting of domesticated and wild
germplasm reveals an extraordinarily large resistance
gene family
Based on genomic location (Bouzidi et al., 2002; Gedil et al.,
2001a; Gentzbittel et al., 1998) and the correlation between
RGC DNA fingerprints and downy mildew resistance
phenotypes (Figure 2), some of the HaRGC1 R genes are likely
to encode downy mildew resistance determinants. In order to
assess whether domesticated and wild sunflower germplasm
differed in HaRGC1 complexity and to determine how much
of the total diversity was represented in domesticated
germplasm, the HaRGC1 IFLP marker (F97-R134 primer
combination) was used to fingerprint 63 wild and domesti-
cated sunflower germplasm accessions (Table 1). Fluores-
cently labelled PCR products were amplified from the
germplasm accessions and separated on polyacrylamide gels
using the ABI Prism 377 genotyping platform (Figure 3).
Because of the complex multilocus nature of the IFLP finger-
prints, the allelism of bands could not be ascertained. Never-
theless, the number of bands in inbred lines was presumed to
be nearly equal to the number of loci, minus the percentage
of segregating loci, e.g. 93.75% of the bands are predicted
to be homozygous and nonallelic in F5 lines.
Forty-eight unique HaRGC1 bands ranging in size from
404 to 1173 bp were observed among the 63 accessions and
designated Rgc1.1(i89) to Rgc1.48(i858) (Table S1). Two to
16 HaRGC1 bands (loci) were amplified per inbred line.
Haplotypes were constructed by scoring for the presence or
absence of each of the 48 bands. Forty-three haplotypes
were identified among the 63 germplasm accessions
(Table S1). The gene diversity of HaRGC1 (Dh) across germ-
plasm accessions was 0.98 and the mean diversity of the 48
polymorphic bands (Dk) was 0.27 (Table 3); hence, there is a
98% chance of observing different HaRGC1 haplotypes
between any two randomly selected germplasm accessions in
sunflower. The probability is lower, of course, when sampling
elite inbred lines only (Table 3).
The 63 germplasm accessions were split into seven groups
(as described in the Experimental procedures) on the basis of
breeding, domesticaton or historical origin. We split the elite
B- and R-lines in the PIL group (HA335-339 and RHA340,
respectively) into a group separate from other B- and R-lines
because of their unique breeding background (Miller and
Gulya, 1988). The haplotype diversity between groups (tested
using AMOVA) was highly significant (P < 0.0001); however,
the within-germplasm group variance component (66.55)
was twofold greater than the between-group variance
component (33.45) because of extraordinarily high within-
group gene diversities (D). The four exotic germplasm groups
(open-pollinated [OP] populations, Native American land
races, randomly selected wild populations, and wild donors
of resistance genes used to develop the HA335-339 and
RHA340) had gene diversity estimates of 1.0 (each member
of each group had a unique haplotype) (Tables S1 and 3).
Moreover, 32 out of 33 germplasm accessions in the four
exotic germplasm groups had unique haplotypes. VNIIMK8931,
the lone germplasm accession sharing the haplotype of an elite
inbred line (RHA373), is an elite open-pollinated popula-
tion and progenitor of contemporary oilseed inbred lines.
Fi
gure
3G
enescan ana
l
ys
i
s o
f
fl
uorescent
l
y
labelled PCR products produced from 43
sunflower genotypes using primer pair F97/R134.
Green colour, TET-labelled resistance gene
candidates (RGCs); red colour, TAMRA-labelled
DNA size markers. Lane 1, HA292; 2, RHA280; 3,
RHA282; 4, HA89; 5, HA369; 6, HA370; 7,
HA371; 8, HA372; 9, HA383; 10, HA407; 11,
HA821; 12, RHA274; 13, RHA377; 14, RHA391;
15, RHA392; 16, RHA409; 17, RHA417; 18,
RHA801; 19, Arikara; 20, Havasupai; 21, Hopi; 22,
Seneca; 23, Tarahumara; 24, Jilin; 25,
Tchernianaka; 26, Abendsonne Red; 27,
Mennonite; 28, Peredovik; 29, VNIIMK8931; 30,
Pervenets; 31, H. annuus 1811; 32, Zambian; 33,
H. annuus CO; 34, H. annuus WA; 35, H. annuus
WY; 36, H. annuus ND; 37, H. annuus AZ; 38, H.
annuus OK; 39, H. annuus MX; 40, H. annuus OR;
41, H. annuus CA; 42, H. annuus SD; 43, H.
annuus UT. Selected RGCs are identified to the left
of the GENESCAN picture; inferred sizes
(nucleotides) of included introns are indicated in
parentheses.

Resistance gene cluster in sunflower
© Blackwell Publishing Ltd, Plant Biotechnology Journal (2003), 1, 167–185
171
Table 1 Description of sunflower germplasm used in this study
Common name Number* Germplasm category Origin
HA89 PI 599773 Oilseed B† line USDA-ARS
HA369 PI 534655 Oilseed B line USDA-ARS
HA821 PI 599984 Oilseed B line USDA-ARS
HA371 PI 534657 Oilseed B line USDA-ARS
HA383 PI 578872 Oilseed B Line USDA-ARS
HA407 PI 597371 Oilseed B line USDA-ARS
HA370 PI 534656 Oilseed B line USDA-ARS
HA372 PI 534658 Oilseed B line USDA-ARS
PHD‡ Oilseed B line Pioneer Hi-Bred
RHA801 PI 599768 Oilseed R§ line USDA-ARS
RHA391 PI 603987 Oilseed R line USDA-ARS
RHA274 PI 599759 Oilseed R line USDA-ARS
RHA377 PI 560145 Oilseed R line USDA-ARS
RHA409 PI 603990 Oilseed R line USDA-ARS
RHA417 PI 600000 Oilseed R line USDA-ARS
RHA373 PI 560141 Oilseed R line USDA-ARS
RHA392 PI 603988 Oilseed R line USDA-ARS
PHB‡ Oilseed R line Pioneer Hi-Bred
PHA‡ Oilseed R line Pioneer Hi-Bred
PHC‡ Oilseed R line Pioneer Hi-Bred
HA292 PI 552937 Confectionery B line USDA-ARS
HA287 PI 552933 Confectionery B line USDA-ARS
RHA280 PI 552943 Confectionery R line USDA-ARS
RHA282 PI 552944 Confectionery R line USDA-ARS
HA335 PI 518773 Oilseed B line, PIL¶ USDA-ARS
HA336 PI 518774 Oilseed B line, PIL USDA-ARS
HA337 PI 518775 Oilseed B line, PIL USDA-ARS
HA338 PI 518776 Oilseed B line, PIL USDA-ARS
HA339 PI 518777 Oilseed B line, PIL USDA-ARS
RHA340 PI 518778 Oilseed R line, PIL USDA-ARS
H. annuus ANN423 PI 435434 Wild donor population Texas
H. annuus ANN432 PI 435437 Wild donor population Texas
H. praecox PRA417 PI 435850 Wild donor population Texas
H. praecox PRA419 PI 435851 Wild donor population Texas
H. praecox PRA424 PI 435852 Wild donor population Texas
VNIIMK8931 PI 340790 OP** oilseed cultivar Russia
Peredovik Ames 1838 OP oilseed cultivar Russia
Pervenets PI 483077 OP oilseed cultivar Russia
Tchernianka Select W13 PI 343794 OP oilseed cultivar Russia
Zambian PI 500689 OP confectionery cultivar Zambia
Jilin Ames 10106 OP confectionery cultivar China
Mennonite Ames 7574 OP confectionery cultivar Canada
Tarahumara OP confectionery cultivar Mexico
Abendsonne Red PI 490316 OP ornamental Germany
Arikara PI 369357 Native American land race North Dakota
Havasupai PI 369358 Native American land race Arizona
Hopi PI 369359 Native American land race Arizona
Seneca PI 369360 Native American land race New York
H. annuus MT PI 531022 Wild population Montana
H. annuus NV PI 468596 Wild population Nevada
H. annuus ND PI 468439 Wild population North Dakota
H. annuus MX PI 413123 Wild population Mexico
H. annuus CO PI 468625 Wild population Colorado
H. annuus WY PI 413019 Wild population Wyoming
H. annuus UT PI 468619 Wild population Utah
H. annuus SD PI 413039 Wild population South Dakota
H. annuus AZ PI 468575 Wild population Arizona
H. annuus WA PI 531018 Wild population Washington

Mary B. Slabaugh et al.
© Blackwell Publishing Ltd, Plant Biotechnology Journal (2003), 1, 167–185
172
H. annuus OR PI 531015 Wild population Oregon
H. annuus CA PI 435593 Wild population California
H. annuus OK PI 435619 Wild population Oklahoma
H. annuus ANN1811 PI 494567 Wild population Texas
H. annuus ANN1238 Wild population Nebraska
*USDA-ARS National Plant Germplasm System plant introduction number.
†B, fertility maintainer line.
‡Proprietary lines developed by Pioneer Hi-Bred, International.
§R, fertility restorer line.
¶PIL, partially inbred line.
**OP, open-pollinated.
Common name Number* Germplasm category Origin
Wild H. annuus populations were members of the ‘wild
population’ and ‘wild donor’ groups. The latter, however,
was comprised of wild H. annuus and H. praecox popula-
tions identified by extensive screening of the sunflower
germplasm collection as carrying novel downy mildew
resistance specificities (Miller and Gulya, 1988), whereas
the former was comprised of geographically diverse but
otherwise randomly selected wild H. annuus populations.
The mean diversity of individual polymorphic bands (Dk)
was slightly greater for the wild donors (0.31) than the wild
populations (0.26) per se (Table 3). The three elite inbred
line groups had lower gene diversity estimates than the exotic
germplasm groups (0.72 for R-lines, 0.87 for B-lines, and
0.60 for the PILs), but were still remarkably diverse
(Table 3).
Downy mildew resistance is confined to haplotypes 6, 9
and 10/11
The 43 HaRGC1 haplotypes are displayed in a minimum
spanning tree constructed from a genetic distance matrix
(Figure 4). The nodes of the tree correspond to each
haplotype, e.g. HA335, 337, 338 and 339, share haplotype
10 (Table S1) and comprise one node. Inbred lines at the left
edge of the tree (haplotype 1) had the fewest HaRGC1 bands
(two). The lengths of the branches in the tree correspond
to the number of HaRGC1 band differences between the
node and the previous edge, e.g. the branch connecting the
HA336 node to the HA335, 337, 338 and 339 nodes is one
‘band’ long and corresponds to the extra RGC (Rgc1.21)
found in HA336. The PILs (HA335-339) and their wild H.
annuus and H. praecox donors (ANN423, ANN432, PRA417,
PRA419 and PRA424) stand out in this respect, having longer
branches than most other nodes in the tree (Figure 4).
Known downy mildew resistance loci were associated with
three well-separated nodes in the minimum spanning tree.
Seven elite oilseed R-lines homozygous for Pl2 (RHA274,
RHA377, RHA409, RHA417, RHA801, PHA and PHC) (Fick
Figure 4 Minimum spanning tree constructed from a genetic distance
matrix calculated from 63 HaRGC1 haplotypes. The nodes of the tree
correspond to each of the 43 haplotypes and the branch lengths
correspond to the number of HaRGC1 band differences between the
node and the previous edge. The scale bar indicates the length
corresponding to a single band difference.
Table 1 Continued

Resistance gene cluster in sunflower
© Blackwell Publishing Ltd, Plant Biotechnology Journal (2003), 1, 167–185
173
et al., 1975; Miller and Gulya, 1999; Miller, 1992; Roath
et al., 1981) formed one node (haplotype 6, Table S1). RAPD
markers linked to Pl2, OPAA14750 and OPAA111008, and a
SCAR marker developed from OPAA111008 (Brahm et al.,
2000), were amplified exclusively by the seven inbreds
comprising this node, strengthening the correlation between
Pl2 and haplotype 6. Compared to susceptible inbred lines,
the only RGC uniquely associated with this haplotype was
Rgc1.27. Rgc1.27 was also detected in HA335–339, three
land races (Arikara, Havasupai and Hopi), and one wild
H. annuus population (ANN1238). Lines HA335–339 are also
resistant to race 2 of downy mildew; the resistance status of
the land races and ANN1238 are not known.
Two confectionery R-lines (RHA280 and RHA282), a
confectionery B line (HA287), and one elite oilseed B-line
(HA370) formed haplotype 9 (Table S1), which clustered as
a single node distinct from the haplotype 6 node. HA370
is known to carry resistance to the European race of
downy mildew conferred by the Pl1 gene (Miller and Gulya,
1990). Although the downy mildew resistance status of
RHA280, RHA282 and HA287 has not been reported, the
three confectionery lines have rust resistant cultivars in
their pedigrees and are likely to have inherited Pl1 as well
(see below). Among elite inbred lines, seven HaRGC1 IFLP
markers were uniquely associated with haplotype 9:
Rgc1.9, Rgc1.26, Rgc1.31, Rgc1.36, Rgc1.41, Rgc1.44 and
Rgc1.46.
HA335 and HA336 (Pl6) and HA337–339 (Pl7) clustered
in the tree as a node independent of haplotypes 6 and 9.
Resistant to multiple races of downy mildew (Miller and
Gulya, 1991), these lines produced identical sets of 16
HaRGC1 bands, apart from one additional band produced
only by HA336 (haplotypes 10 and 11, Table S1). Half of the
RGCs common to HA335–339 were not observed in any
other inbred lines we surveyed. The HaRGC1 haplotype of
RHA340, which carries resistance determinants introgressed
from H. argophyllus (Pl8), resembled the recurrent parent
HA89. This result is consistent with the recent localization
of Pl8 to LG13 (Bert et al., 2001) and with the lack of
introgressed DNA on LG8 in RHA340 (see below).
RHA373 was reported as resistant to race 2 of downy
mildew (Miller, 1992). However, this line clustered with
the susceptible open-pollinated Russian oilseed cultivar
VNIIMK8931 as a separate node (haplotype 4). RHA373 and
VNIIMK8931 had no RGCs in common with haplotype 6,
other than the ubiquitous Rgc1.29. Unlike the haplotype
6 inbreds, RHA373 did not amplify either of the Pl2-linked
markers OPAA14750 and OPAA111008. RHA373 was derived
from a cross between RHA274 and a Romanian sunflower
line (Miller, 1992). We concluded that either a recombination
event had separated Pl2 from the RAPD markers during
development of RHA373, or that resistance to race 2 in
this line was due to another gene. Remaining inbred lines in
the germplasm panel were susceptible to all races of downy
mildew, based on published registration notes and
germplasm accession information in the GRIN database
<http://www.ars-grin.gov/npgs/searchgrin.html>.
HaRGC1 paralogues map to the Pl1-Pl2-Pl6 region of
LG8
Twenty-two of the 24 RGC paralogues amplified from elite
inbred lines were mapped using two recombinant inbred
line populations and an F2 population. The HaRGC1 R gene
marker loci clustered on the upper end of LG8 and spanned
2–4 cM on the three maps (Figure 5). Sixteen HaRGC1 IFLP
bands were scored in the RHA280 x RHA801 population of
Tang et al. (2002); 12 were scored in the HA370 x HA372
population of Gedil et al. (2001a); and 12 were scored in the
PHA x PHB population of Yu et al. (2003). Two additional
markers, Rgc1.49 and Rgc1.50, were mapped within the
HaRGC1 cluster in two populations using sequence-specific
primers developed from partial cDNA sequences. Rgc1.49
and Rgc1.50 had a high homology to HaRGC1, but were
not amplified by F97/R134 or F97/R98 primer combinations.
Three simple sequence repeat (SSR) markers (ORS166,
ORS299, and ORS1043) segregated within the cluster
(Figure 5).
The Pl1 locus and Pl2-linked RAPD markers OPAA14750 and
OPAA111008 (Brahm et al., 2000) were added to the genetic
maps (Figure 5). Pl1 was scored in the HA370 (resistant) x
HA372 (susceptible) cross (Gedil et al., 2001a) and co-
segregated with Rgc1.9, Rgc1.21 and SSR marker ORS1043.
In the RHA280 x RHA801 population, the Pl2-linked marker
OPAA14750 co-segregated with Rgc1.9, Rgc1.20 and Rgc1.49,
whereas OPAA111008 mapped distal to the RGC cluster. These
results are consistent with both Pl1 and Pl2 residing within
the HaRGC1 cluster, considering that OPAA14750 and
OPAA111008 were found to be ≈ 2 cM and ≈ 6 cM distal to
Pl2 (Brahm et al., 2000).
Rust resistance gene marker and the HaRGC1 RGC
cluster
Vranceanu and Stoenescu (1970) were the first to report an
association between resistance to rust and downy mildew
(Pl1) in sunflower. Fick and Zimmer (1975) found that a rust
resistance gene in RHA266, a line that carries Pl1 as well as

Mary B. Slabaugh et al.
© Blackwell Publishing Ltd, Plant Biotechnology Journal (2003), 1, 167–185
174
rust resistance, and the Pl2 gene in RHA273 were tightly
linked in repulsion (no recombinants were observed among
543 F3 lines). To investigate the relationship of rust resistance
to the HaRGC1 R gene cluster, we utilized a SCAR marker
(SCT06950) reported by Lawson et al. (1998) to co-segregate
with the rust R1 gene in a population of 76 F2 progeny. The
rust resistance marker mapped centromeric to the HaRGC1
cluster at a distance of 2–3 cM in the RHA280 x RHA801 and
HA370 x HA372 populations, and approximately 8 cM in the
PHA x PHB population, thereby substantiating the linkage
of rust resistance and downy mildew resistance on LG8
(Figure 5).
Two alleles of the SCT06950 SCAR marker were detected, a
‘high Tm’ allele designated XA in Table S1, and a ‘moderate
Tm’ allele of the same size designated XB. The B allele was
amplified at 55 °C annealing temperature but not at 67 °C.
When assayed as described by Lawson et al. (1998), the A
allele was present in all of the inbred lines exhibiting HaRGC1
haplotype 6 (Pl2) as well as in downy mildew susceptible lines
HA292, RHA282 and PHD. RHA274 and RHA801 (haplotype
6), HA292 and HA282 are reportedly resistant to race 1 of
rust or have rust-resistant Mennonite RR lines in their
pedigrees (Fick and Zimmer, 1979; Fick et al., 1974, 1975;
Roath et al., 1981). Twelve wild H. annuus populations also
produced the A allele, as did three Native American land races
and three OP cultivars (Table S1). Thus, we found a strong
association between the A allele and rust resistance in elite
inbred lines, but the rust marker was not predictive of downy
mildew resistance except within HaRGC1 haplotype 6 lines.
Additionally, the data suggest that the rust resistance gene
linked to the A allele of SCT06950 is widespread in wild H.
annuus populations. The B allele was amplified from HA370
(Pl1), a line reported to be resistant to race 1 of rust (Miller
and Gulya, 1990), but was also amplified from several inbred
lines not reported to carry any rust or downy mildew
resistance. The SCT06950 null allele (N) was observed in only
four inbred lines (HA371, HA372, HA821 and RHA391), all
reported to be susceptible to rust race 1.
Wild downy mildew resistance donor populations
amplified the largest number of HaRGC1 markers
The resistance genes found in HA335–339 were introgressed
by Miller and Gulya (1988) from wild sunflowers using
crosses produced by bulking pollen from several individuals
per donor population. Because transient individuals were
used in the original crosses, randomly selected individuals
from the donor populations (ANN423 and 432, PRA417, 419
and 424) were used for the HaRGC1 IFLP fingerprinting
reported here. Five ‘surrogate donor’ DNA samples produced
36 different HaRGC1 bands (Table S1), among which were
all but two of the HA335–339 HaRGC1 bands. Four of the
Figure 5 Genetic maps of the HaRGC1/Pl1 region on the upper end of sunflower linkage group 8 as determined in three mapping populations. A
reference chromosome from the public sunflower map of Yu et al. (2002) is shown at the left. SSR markers (bold type), HaRGC1 IFLP markers, SCAR
marker SCT06950, and RAPD markers OPAA111008 and OPAA14750 were scored on mapping populations of RHA280 × RHA801 (RIL), HA370 × HA372
(F2), and PHA × PHB (RIL).

Resistance gene cluster in sunflower
© Blackwell Publishing Ltd, Plant Biotechnology Journal (2003), 1, 167–185
175
surrogate donor plants produced more than 20 HaRGC1
bands. Miller and Gulya (1988) reported that the original
donors were heterozygous for race 2 resistance genes. More-
over, because the donors were outbred, the individual plants
we sampled could have been heterozygous for RGC genes.
Regardless, the mean number of HaRGC1 bands in the
surrogate donors (20.8) was significantly higher than the
mean in the 15 outbred wild H. annuus populations not
previously selected for downy mildew resistance (10.8)
(P < 0.0001). Similarly, the mean number of HaRGC1 bands
in the PILs (15.4) was significantly greater (P < 0.037) than in
the unselected wild populations. These data suggest a
correlation between high HaRGC1 paralogue copy number
and functional disease resistance.
Graphical genotypes of HA335 –339 and RHA340 reveal
wild donor introgressions on linkage groups 8 and 13
Partially inbred lines HA335–339 each carried an introgres-
sion of donor DNA on LG8 that encompassed the HaRGC1/
Pl region (Figure 6). A ≈ 15 cM segment was detected in
HA335, HA338 and HA339, whereas much longer donor
DNA segments persist in HA336 (≈ 60 cM) and HA337
(≈ 40 cM). HA335–339 carried identical alleles from the
wild donors for SSR markers within the ≈ 15 cM segment
common to all five lines (ORS166 and ORS328-8). DNA telo-
meric to the Pl region was apparently derived from recurrent
parent HA89 in HA335–339. Concurrent selection for disease
resistance and HA89 agronomic characteristics (Miller
and Gulya, 1988) may have profoundly restricted the recom-
binants recovered in this region, because the likelihood of
this occurring by chance is very small. However, we cannot
exclude the possibility that actual (as compared to ‘surrogate’)
wild donors carried alleles identical to HA89 in this region.
Only HA89 alleles were detected in RHA340 DNA throughout
LG8.
On linkage group 13, HA336, HA339 and RHA340 carried
wild donor introgressions, whereas HA335, 337 and 338
did not (Figure 6). RHA340 harboured ≈ 30 cM of foreign
DNA at the distal end. The introgressed fragment in RHA340
included Ha-1W23 and Ha-1W41, two non-TIR-NBS-LRR
RGCs previously mapped to LG13 by RFLP analysis (Gedil
et al., 2001a). The PCR-based assays used to detect Ha-1W23
and Ha-1W41 in the present study amplified multiple co-
segregating sequences per assay, as determined by SSCP
analysis (data not shown). These RGCs are candidates for the
Pl8 gene, which was recently mapped to LG13 (Bert et al.,
2001). In HA336 and HA339, the introgressed segments did
not overlap the donor segment present in RHA340.
HaRGC1 markers in non-inbred domesticated
germplasm and wild populations
Each of the nine OP sunflower cultivars had a unique
HaRGC1 haplotype (Table S1). Only four of 21 paralogues
detected in OP material were not present in elite inbred
haplotypes (19%), consistent with the history of contemporary
inbred lines whose pedigrees trace directly to Russian oilseed
and confectionery cultivars (Cheres and Knapp, 1998; Korell
et al., 1992). The four Native American land races we
investigated produced 17 HaRGC1 paralogues. While only
two were not present in inbred lines (12%), none of the land
race haplotypes matched any inbred haplotype.
Figure 6 Linkage group 8 and 13 graphical genotypes for partially inbred
lines HA335, HA336, HA337, HA338, HA339 and RHA340. Recurrent
parent (HA89) segments are white, introgressed segments are black, and
segments encompassing recombination events are stippled. Composite
reference chromosomes for linkage groups 8 and 13 are shown to the left
of the graphical genotypes.

Mary B. Slabaugh et al.
© Blackwell Publishing Ltd, Plant Biotechnology Journal (2003), 1, 167–185
176
Thirty-nine HaRGC1 paralogues were scored in the wild H.
annuus accessions not selected for disease resistance. Of
these, 16 (41%) were not detected in inbred lines. Between
5 and 19 different bands were amplified per individual in the
wild populations. With two exceptions, all of the paralogues
present in elite inbred lines, OP cultivars, and land races were
present in at least one of the wild accessions. Rgc1.36 from
inbred haplotype 6 and Rgc1.6 from OP ornamental cultivar
Abendsonne Red were not amplified from any of the 15
wild H. annuus DNAs. Additionally, seven RGCs amplified
from the wild surrogate donors were not observed in the
unselected wild H. annuus populations.
Thirty-one HaRGC1 markers were detected in both dome-
sticated and wild germplasm, but only four of these, Rgc1.29,
Rgc1.24, Rgc1.38 and Rgc1.20, were present at greater than
50% frequency in both groups. Rgc1.29 and Rgc1.24 were
notable in two respects. First, they were nearly universal, with
overall frequencies of 0.98 and 0.85, respectively. Second,
they were the only RGCs detected in inbred haplotype 1
(HA369, RHA392 and RHA340). These results imply a
selection pressure favouring the retention of specific RGCs.
In sum, the HaRGC1 region in OP cultivars and land races
was populated primarily with the same RGCs present within
the elite inbred gene pool. The two wild germplasm groups
and downy mildew resistant PILs carrying wild donor intro-
gressions on LG8, on the other hand, harboured almost
as many unique HaRGC1 markers as they did paralogues
present in elite inbred lines. Twenty-four RGC paralogues
found in elite inbred lines segregated and were mapped to
a telomeric region on LG8. Thus, it is likely that all or most
members of this resistance gene class reside in various com-
binations within one large cluster in the sunflower genome.
The total number of HaRGC1 paralogues is undoubtedly
larger than the 48 reported here. Several of the IFLP markers
represented multilocus targets, detected either as broad
peaks that contained closely migrating amplicons differing by
1–2 nts in length (Rgc1.31, Rgc1.33, Rgc1.38 and Rgc1.39),
or as a strong fluorescent band that was separated into weaker
signals by recombination events (Rgc1.24 in the RHA280 x
RHA801 population). Indeed, the inter-TIR-NBS intron in
Rgc1.38 corresponded in size to the two markers described
by Bouzidi et al. (2002), HaNBS4 and HaNBS12, and Rgc1.35
had the same intron size as HaNBS1, 3 and 6 in their study.
Intron hypervariability and evolution of the HaRGC1
cluster
The intron that separates the TIR- and NBS-encoding regions
in the HaRGC1 family of paralogues varied in size from 89 to
1046 nt (Table S1; Bouzidi et al., 2002). To investigate the
nature of this variability, a multiple sequence alignment was
constructed using introns from five sequences in this study
and 13 sequences from the analysis of Bouzidi et al. (2002)
(Figure 7a). The 1046 nt intron in NBS11 (Bouzidi et al., 2002)
was homologous to the others only at the extreme 3′ end.
The remaining sequences, however, aligned well with the
662 nt NBS5 intron. Within aligned blocks, 50–70% of the
Figure 7 Comparative analysis of HaRGC1 inter-TIR-NBS intron structure
and NBS amino acid sequences. Rgc1.2, Rgc1.12, Rgc1.20, Rgc1.21 and
Rgc1.24 sequences were from the present study; NBS1, 2, 3, 4, 5, 6, 7,
8, 9, 11 and 12 were described by Bouzidi et al. (2002). The intron
comparison (a) was constructed using multiple sequence alignment
software and manually optimized. For the distance tree (b), deduced
amino acid sequences were trimmed to a common segment of 105
residues and analysed with CLUSTALW using default parameters. The
phylogram was drawn with TREEVIEW from the CLUSTALW output. Branch
lengths are proportional to sequence divergence and can be measured
relative to the bar shown.

Resistance gene cluster in sunflower
© Blackwell Publishing Ltd, Plant Biotechnology Journal (2003), 1, 167–185
177
nucleotides were identical across all 17 sequences. The 5′ end
of the intron was characterized by poly T runs interrupted
by (TC)n repeats (n = 2–4). A two-nt difference in length
between presumptive alleles of Rgc1.21 was due to a
haplotype-specific variation in this dinucleotide repeat.
Deletions were interspersed throughout the intron and
extended to within ≈ 6 nt of the 5′ end (Rgc1.2). Multiple
sequence alignment of the inter TIR-NBS intron suggested a
hypothesis that evolution of the HaRGC1 cluster proceeded
from a founder sequence with a large intron in this position
that underwent multiple duplication/deletion events, and in
the case of NBS11, a large insertional event.
Three distinct subclades were evident in both the multiple
sequence alignment of the introns (Figure 7a) and the
analysis of the amino acid sequences downstream of the
intron (Figure 7b). In general, HaRGC1 paralogues with
common intron deletion endpoints clustered on distinct
branches of the tree constructed from the amino acid
alignment. One subclade comprised RGCs with introns of
614–650 nt (NBS1, NBS3, NBS4, NBS6 and NBS12). All of
these gene fragments were isolated from a downy mildew
susceptible cultivar (Bouzidi et al., 2002). Rgc1.20, Rgc1.21,
Rgc1.24, NBS8 and NBS13 exhibited introns of 294–384 nt
length and formed a distinct subclade. These paralogues
were isolated from both resistant and susceptible cultivars.
Six RGCs (Rgc1.2, Rgc1.12, NBS2, NBS7, NBS9 and NBS14)
with the shortest introns (105–273 bp) were amplified from
lines containing introgressed regions from wild sunflowers
(HA335 and haplotype 6 inbreds, see below). These RGCs are
associated with downy mildew resistance. Based on the
amino acid alignment, NBS11 is a member of this subclade.
Fine structure map of the HaRGC1 genomic region
A probable gene order for the HaRGC1 region was deduced
(Figure 8). Twenty-four RGC paralogues, two RAPD markers
linked to Pl2, and a SCAR marker linked to R1 were first
organized into co-segregating groups as determined in three
mapping populations (Figure 5). Loci within co-segregating
groups were then arranged to maximize contiguous loci in all
domesticated and wild accessions for which IFLP haplotypes
were scored (Table S1). The proposed gene order illustrates
the similarities between progenitor open-pollinated
accessions (e.g. VNIIMK8931, Peredovik, Mennonite) and
elite inbred line haplotypes 1, 2, 3, 4, 5 and 8. Rgc1.29,
detected in 62 of 63 accessions, occupies a central position
in the cluster. A marker of this size was also amplified from
10 out of 10 wild Helianthus species other than H. annuus,
indicating an origin prior to speciation (unpublished data).
Sunflower lines with downy mildew resistance genes Pl1
and Pl2 (haplotypes 9 and 6, respectively) carried groups of
co-segregating RGC paralogues (Figure 8, boxes with thick
outlines) that were not observed in progenitor lines such as
VNIIMK8931 or in contemporary downy mildew susceptible
elite lines. Although none of the 20 wild accessions we
examined carried either of these RGC groups in their entirety,
each of the markers was present in one or more of the wild
accessions, consistent with the acquisition of downy mildew
resistance determinants from wild H. annuus. One of the
haplotype 6-specific markers, Rgc1.27, was amplified
from HA335–339 as well as from haplotype 6 inbreds. Inter-
estingly, this RGC was not amplified from HA292, a suscep-
tible line whose haplotype suggests a recombination within
the HaRGC1 cluster. Hence, Rgc1.27 is a candidate for the
Pl2 structural gene. Two markers that co-segregated with Pl1
in the HA370 x HA372 population, Rgc1.9 and Rgc1.21,
were not observed in susceptible inbreds or OP lines, but one
or both of these RGCs was present in haplotype 6 inbreds,
haplotype 9 inbreds, and the PILs (HA335–339). Hence, these
RGCs are candidates for the Pl1 structural gene.
More than half the paralogues detected using the HaRGC1
primers could not be placed in the composite gene order
because they have not been mapped. Nevertheless, the gene
order in Figure 8 provides a preliminary framework for the
organization of this important disease resistance region.
SSR marker ORS166 identifies HaRGC1 haplotype
groups
Two SSR markers that mapped within the HaRGC1 cluster,
ORS166 and ORS1043, produced complex patterns consisting
of multiple bands. ORS166 was screened for polymorphisms
among the 63 germplasm accessions and produced 16
different bands ranging in length from 311 to 355 nt (Table S2,
Supplementary Material, <http://www.blackwell-science.com/
products/journals/suppmat/ PBI/PBI016 /PBI016sm.htm>).
Because of the complex multiband (multilocus) nature of the
fingerprints, 34 ORS166 haplotypes were constructed using
the presence and absence of the 16 bands, as was done using
genotypes produced by the HaRGC1 IFLP marker. One to four
co-segregating bands were produced from elite inbred lines,
suggesting that the ORS166 SSR primer pair amplified one to
four duplicated loci per line. The 20 wild H. annuus and H.
praecox germplasm accessions each had a unique ORS166
haplotype, making this one of the most polymorphic markers
we have characterized (Table S2; Tang and Knapp, 2003).
We concluded that the ORS166 SSR was present in DNA that
had been duplicated along with resistance genes during the

Mary B. Slabaugh et al.
© Blackwell Publishing Ltd, Plant Biotechnology Journal (2003), 1, 167–185
178
course of the evolution of HaRGC1 paralogues. Supporting
this idea, two additional SSR-containing clones that differed
in the number of (CA)n repeats but had flanking sequences
that were 98% identical to ORS166 were identified by
searching sunflower SSR sequence databases for sequences
related to ORS166. These sequences had been discarded as
redundant in the process of SSR marker development, and
were isolated from independently produced SSR-enriched
genomic DNA libraries (Tang et al., 2002; Yu et al., 2003).
ORS166 SSR haplotypes were diagnostic for specific downy
mildew resistance genes, e.g. ORS166 haplotype 4 for Pl2 (IFLP
haplotype 6), ORS166 haplotype 5 for Pl1 (IFLP haplotype 9),
and ORS166 haplotype 6 for Pl6/7 (IFLP haplotypes 10 and
11). ORS166 haplotypes can be used in marker assisted
selection for associated disease resistance genes.
Discussion
The ability of plants to repel host-specific pathogens is medi-
ated by R genes that recognize the early stages of infection
and trigger localized cell suicide. Classical resistance breeding
has focused on identifying and transferring useful R genes
into crop plants. The deployment of such genes in crop
monocultures has repeatedly led to ‘boom and bust’ cycles as
Figure 8 Gene order in the HaRGC1 region of
sunflower linkage group 8. Twenty-two mapped
HaRGC1 paralogues, two partial cDNA
sequences (Rgc1.49 and Rgc1.50), a SCAR
marker linked to rust resistance (SCT06950), and
RAPD markers linked to resistance to race 2 of
downy mildew (OPAA111008 and OPAA14750)
were arranged in an order that was consistent
with segregation patterns in three mapping
populations and maximized adjacent positioning
of resistance gene sequences in 63 accessions.
Markers are coloured as follows: yellow, elite
inbred lines and partially isogenic lines (PILs); light
orange, open-pollinated accessions; dark
orange, Native American land races; red, wild
accessions. Markers with thicker outlines
highlight blocks of genes that were presumably
acquired from wild sunflowers (haplotypes 6 and
9).

Resistance gene cluster in sunflower
© Blackwell Publishing Ltd, Plant Biotechnology Journal (2003), 1, 167–185
179
pathogens eventually defeat the introduced defence system,
presumably due to an intense selection for variants carrying
rare or mutated avirulence genes. This phenomenon has
raised the question of how non-domesticated plants manage
to keep up the arms race with more genetically nimble path-
ogens. The answer seems to lie in having a large number of
R genes, deploying them in ways that promote a continued
evolution of specificities, and maintaining natural popula-
tions with a diversity of resistance traits (Holub, 2001; The
Arabisopsis Genome Initiative, 2000; Van der Hoorn et al.,
2002). A better understanding of variation in R genes and
their mechanisms of evolution in natural plant/pathogen
systems will assist in designing more durable genetic resistance
in domesticated crops.
Downy mildew and rust are important diseases of sun-
flower that for several decades have been controlled using
genetic resistance. The primary goal of this project was to
investigate, in both domesticated and wild germplasm, the
genomic organization of a region on LG8 where genes active
against downy mildew and rust pathogens are known to
reside (Bouzidi et al., 2002; Gedil et al., 2001a; Gentzbittel
et al., 1998; Mouzeyar et al., 1995; Vear et al., 1997). By a
judicious choice of PCR primers, we targeted a specific
subfamily of TIR-NBS-LRR genes which were previously
associated with the Pl1 gene (Gedil et al., 2001a; Gentzbittel
et al., 1998). The physical evidence of a large cluster of resist-
ance genes in the Pl1-Pl2-Pl6 region was discovered by using
a powerful and sensitive assay that detected differences in
the length of a specific intron. The paralogous array spanned
2–4 cM and encompassed classic phenotypic loci that
conferred resistance to several races of downy mildew. An
extraordinary diversity in the Pl1-Pl2-Pl6 region was found
in wild sunflowers using the HaRGC1 IFLP marker and SSR
markers. Only half of the total haplotypic diversity was present
in domesticated germplasm.
The assay developed for this study was a variant of the
degenerate PCR method widely used in numerous plants to
simultaneously amplify segments of genes encoding the
nucleotide binding site and leucine rich repeat domains (Aarts
et al., 1998; Kanazin et al., 1996; Yu et al., 1996). Rather
than detecting different gene fragments by cloning and
sequencing PCR products amplified from the highly conserved
NBS region, primers were positioned on either side of an
intron expected to be upstream of the NBS region, based on
the intron/exon structure of the Arabidopsis RPP1 and RPP5
genes. PCR products were differentiated by precisely sizing
fluorescently labelled amplicons using automated gel electro-
phoresis and fragment analysis software. This method takes
advantage of intron hypervariability and utilizes modern
genotyping methods that can reliably detect sequence length
differences as small as 1–2 bp. Amplification was restricted
to a specific subfamily of TIR-NBS-LRR genes by designing
a downstream primer in the NBS region such that only
HaRGC1-type sequences would be amplified. We expected
to find a multigene family because the HaRGC1 clone, which
was initially produced by degenerate PCR, detected multiple
bands in sunflower DNA when used as an RFLP probe (Gedil
et al., 2001a). The fluorescent IFLP assays were performed
using the universal M13-tail method to affix fluorescent
labels. Because the development of DNA markers such as
those described in this paper can be costly and is typically
done through a trial and error primer testing process, the
universal M13-tailed method permits rapid screening of
multiple primer pairs in fluorescent assay systems without the
added expense of fluorescently end-labelling one of the
primers in each primer pair.
The conclusion that different sized amplicons represent
unique TIR-NBS-LRR loci rests on the assumptions that all the
PCR products were actually derived from genes or pseudo-
genes in the targeted class and that differences in amplicon
lengths detected sequence variation among paralogues, not
allelic variation between homologues. We addressed these
issues by cloning and sequencing five amplicons of different
lengths from several inbred lines. The results validated our
original assumptions, with the caveat that very small length
differences may indicate alleles or more recently diverged
paralogues.
BLAST searches using the cloned sequences as queries
returned high-scoring hits to 13 genomic fragments from
sunflower lines HA335 and H52 (Bouzidi et al., 2002) and to
RGCs from RHA266 (Ha-NBS3, Gentzbittel et al. 1998) and
HA89 (HaRGC1, Gedil et al. 2001a). These sequences are
highly homologous and all have been mapped to the Pl1-Pl2-
Pl6 region of sunflower LG8. Although Bouzidi et al. (2002)
used several different PCR primers, the eight genomic
fragments they cloned from HA335 included the intron
utilized in this study. Thus, we were able to directly compare
the sets of HA335 paralogues produced in the two studies.
The single primer pair we used amplified 16 RGCs from
HA335, including all but two of the genomic sequences
identified by Bouzidi et al. (2002). One of the undetected
RGCs contained an intron that was too large to be detected
by GENESCAN analysis (> 1000 bp), and the other had several
mismatches in the F97 primer binding site and was not
amplified in our system. An examination of the sequences of
the 13 genomic fragments, however, showed that F97 and
R134 successfully amplified fragments with one or two
mismatched nucleotides in the primer binding sites.

Mary B. Slabaugh et al.
© Blackwell Publishing Ltd, Plant Biotechnology Journal (2003), 1, 167–185
180
Of the sequences described by Bouzidi et al. (2002), several
unique genomic sequences had introns of the same length.
This illustrates a limitation to the use of IFLP fingerprints alone
for RGC discovery. By combining information from both
approaches we concluded that the number of HaRGC1
paralogues present in HA335 numbers at least 20, making
this sunflower LG8 resistance gene cluster (TIR-NBS-LRR R
gene subclass) similar in complexity to the major lettuce
downy mildew resistance gene cluster (non-TIR-NBS-LRR R
gene subclass) (Meyers et al., 1998), and as large as any plant
R gene cluster described thus far.
The two SSR markers we identified in genomic DNA in the
LG8 TIR-NBS-LRR cluster produced signature multi-peak
fingerprints that we subsequently explained by gene duplica-
tion. This was substantiated by searching the genomic DNA
sequences produced in the process of developing SSR
markers (Tang et al., 2002; Yu et al., 2003). SSR-containing
DNA sequences that we had previously discarded as redundant
were in fact discovered to be paralogous. Moreover, three
SSR markers in the genomic region on LG13 introgressed
from H. argophyllus were found to either produce the telltale
multi-peak fingerprint or were present as paralogous clones
in our SSR marker database. These genomic DNA sequences
have been deposited in public databases and are a potential
source of paralogous SSR markers residing in NBS-LRR
resistance gene clusters elsewhere in the genome
<http://www.css.orst.edu/knapp-lab and http://www.com-
positdb.ucdavis.edu>.
Sunflowers are produced commercially using single-
cross hybrid seed. The possibility exists for simultaneously
deploying two to four different resistance gene specificities
in a single hybrid by fixing different alleles in the parents. The
IFLP and SSR markers identified on LG8 are ideal for pyramid-
ing downy mildew resistance genes, a virtual impossibility in
practice using phenotypic analysis alone. DNA markers
analogous to the IFLP marker have not yet been developed for
downy mildew resistance genes on LG13 or rust resistance
genes on either linkage group; however, excellent SSR maps
have been developed for both linkage groups, and several
polymorphic SSR markers have been identified in the regions
flanking the resistance genes (Tang et al., 2002; Yu et al.,
2003).
Based on the published pedigrees, all of the downy mildew
resistance genes in elite inbred lines of sunflower were
introgressed from wild relatives. The earliest known sources
of Pl1 (lines AD 66, HA265 and HA266), and Pl2 (lines HA61
and HA62), can be traced to crosses to Canadian rust
resistant lines CM953-102 and CM953-88, respectively (Fick,
1978; Korell et al., 1992). The closely linked rust and downy
mildew resistances in CM953-102 and CM953-88 apparently
resulted from accidental crosses to resistant wild H. annuus
populations in Texas between 1949 and 1953 (Putt and
Sackston, 1957). Lines tracing to both CM953-102 and
CM953-88 contributed to the complex parentage of the
USDA oilseed R-line pool, from which dominant rust and
downy mildew resistant breeding materials were selected
(Cheres and Knapp, 1998; Korell et al., 1992). The RGC pat-
tern we labelled haplotype 6 (Pl2) was most likely inherited
from CM953-88 via HA62. Independent of these events, rust
resistance and linked downy mildew resistance were intro-
duced into confectionery breeding lines following the chance
acquisition of rust resistance by a few individuals of the
cultivar Commander (Jerry Miller, personal communication).
Because RHA280, RHA282 and HA287 all have rust-resistant
confectionery cultivars in their pedigrees, the RGC pattern we
labelled haplotype 9 was most likely derived from these
sources.
The incidence of downy mildew R genes in natural popu-
lations of sunflower has not been rigorously examined. Miller
and Gulya (1988) found that 17% of the 136 wild accessions
they screened had some resistance to race 2 of downy
mildew. Tan et al. (1992) screened 60 wild H. annuus acces-
sions and found four new sources of resistance to race 4
(7%). Combined with the physical evidence for large num-
bers of RGCs detected by IFLP markers (this study), these
data indicate that wild sunflowers represent a rich source
of untapped R genes. The high multiplex IFLP fingerprinting
approach could be combined with disease screening in wild
populations to identify potentially useful new R genes in the
HaRGC1 cluster, and the same approach can be extended
to additional R gene families provided that sequence informa-
tion flanking polymorphic introns is available.
Experimental procedures
Plant material
Twenty-four elite inbred sunflower lines (nine oilseed fertility
maintainer B lines, 11 oilseed fertility restorer R lines, two
confectionery B lines, and two confectionery R lines), nine
open-pollinated (OP) cultivars, four North American land
races, and 15 wild H. annuus accessions, were chosen for
the study of genetic diversity in the Pl region (Table 1). Four
proprietary lines, PHA, PHB, PHC and PHD, were developed
by Pioneer Hi-Bred International (Woodland, CA). For analysis
of partially isogenic interspecific germplasm derived from
wild annual species with selection for disease resistance,
HA335-339 and RHA340 were used. HA335-339 and

Resistance gene cluster in sunflower
© Blackwell Publishing Ltd, Plant Biotechnology Journal (2003), 1, 167–185
181
RHA340 were developed by Miller and Gulya (1988, 1991) by
introgressing the resistance to race 2 of downy mildew from
wild accessions H. annuus 423 (HA335), H. annuus 432
(HA336), H. praecox 417 (HA337), H. praecox 419 (HA338),
H. praecox 424 (HA339), and H. argophyllus 415 (RHA340)
into the susceptible cultivar HA89. Because DNA samples
from actual donor plants used in the original crosses were not
available, an individual plant was grown from each accession
and its DNA served as a ‘surrogate donor’ in the IFLP analysis.
The donor accession for RHA340, H. argophyllus 415, was
not available. Sunflower genomic DNA was extracted from young
leaves using a modified cetyltrimethylammonium bromide
method (Webb and Knapp, 1990). DNA was prepared
from a single plant from each OP, land race, or wild accession.
RGC marker development
We aligned nine sunflower RGCs that had been amplified
from HA89 using degenerate primers (Gedil et al., 2001a),
four HaRGC1-related RGC sequences amplified from HA335
and HA370 (Gedil et al., 2001a), Ha-NBS-R3 (Gentzbittel
et al., 1998), and three partial cDNA sequences with
homology to HaRGC1 isolated by Pioneer Hi-Bred (Johnston,
IA) (Ras 5-1, Rs 6-8 and Ras 7-4, see US Patent Application
Serial no. 09/602 472). (HaRGC1 is the designation given to
the Ha-4W2 resistance gene candidate marker described by
Gedil et al., 2001a). All of the HaRGC1-related sequences
were assigned to the TIR-NBS-LRR subclass of the NBS
superfamily based on nucleotide binding site (NBS) sequence
motifs (Meyers et al., 1999). Of the aligned sequences, only
partial cDNA Ras 5-1 included the TIR domain. Using genomic
sequence information from Arabidopsis RPP1 (Botella et al.,
1998) and RPP5 (Noel et al., 1999) TIR-NBS-LRR gene
families, a probable intron site in the sunflower sequences
between the TIR- and NBS-encoding domains, was identified
and a non-degenerate primer just upstream of this site
was designed (F97) based on the Ras 5-1 sequence (Figure 1).
Partially degenerate downstream primers R98 and R134
were designed to discriminate against non-TIR-NBS-LRR
and non-HaRGC1-type TIR-NBS-LRR genes. The primer pairs
F101/R102 and F103/R104 were designed to amplify
fragments of partial cDNAs Rs 6-8 and Ras 7-4, respectively.
The allele-specific primers R131 and R132 were designed to
discriminate between the HA370 and HA372 alleles of
Rgc1.20(i330) using a 3′ mismatch at a single nucleotide
polymorphism (SNP) site. To produce fluorescently labelled
PCR products, 19 nucleotides of the M13 sequence were
added to the 5′ end of primers F97, F101 and F103. An M13
primer tagged with the fluorophore TET (M13tet) was included
in the three-primer PCR reactions, essentially according to the
methods of Schuelke (2000). The sequences of oligonucle-
otide primers used in this study are shown in Table 2.
Preliminary experiments established that incorporation
of the fluorescently labelled M13 universal primer was
enhanced if the concentrations of F97, F101 and F103 were
one-eighth the concentration of the M13 universal primer,
and that ethidium bromide-stained band patterns produced
with or without addition of the M13 universal primer were
identical. PCR reactions (15 µL) included 1 × Qiagen PCR
buffer, 2 mM MgCl2, 0.2 mM dNTPs, 0.05 µM F97, 0.4 µM
M13tet, 0.8 µM R134 or 0.4 µM R98, R131 or R132, 0.1% Tween-
20, 15 ng genomic DNA template, and 0.3 U HotStart Taq
Primer Sequence Purpose
F97 5′-GTGTAAAACGACGGCCAGTT T TCTGGATGGGTCATCAATG-3′* Rgc1.1–1.48
R98 5′-GACTCCCTACCACAACATCTT T-3′Rgc1.1–1.48
F101 5′-GTGTAAAACGACGGCCAGTGTTACAATCGGT TT TGGATCGT-3′* Rgc1.49
R102 5′-GAGCTCTTCCAAATCT TCCCTA-3′Rgc1.49
F103 5′-GTGTAAAACGACGGCCAGTCTAGAGTTGCATAAGAGCCAA-3′* Rgc1.50
R104 5′-GAGGCTTGAACAGTAGCTCAC-3′Rgc1.50
R131 5′-TCTCGCAAGCGAGTCTCTATT-3′Rgc1.20-T
R132 5′-TCGCAAGCGAGTCTCTATG-3′Rgc1.20-G
R134 5′-CACG(AG)ATATT T T(GC)AAGAA(AG)(AG)CA-3′Rgc1.1–1.48
F140 5′-GTATGGCTAGAATATTGTATAACGAT-3′Ha-1W23
R144 5′-GTT T TCCCAGTCGTCATAGTT TTCAT-3′Ha-1W23
F139 5′-GTCTAGCTAGACTT T TGTT TAATGAT-3′Ha-1W41
R143 5′GTT T TCCCAATTGTCATAGCTCTCGG-3′Ha-1W41
F141 5′-GTCTAGCAAGACTATTGTACAATGAA-3′Ha-1W53
R145 5′-GTT T TCCAATCCTCAGGACTCTCAC-3′Ha-1W53
M13tet 5′-TET-GTGTAAAACGACGGCCAGT-3′TET label
*Underlined nucleotides are M13 sequence added to enable three-primer PCR.
Table 2 Oligonuclotide primers used in this
study

Mary B. Slabaugh et al.
© Blackwell Publishing Ltd, Plant Biotechnology Journal (2003), 1, 167–185
182
polymerase (Qiagen Inc., Valencia, CA). Following 15 min incu-
bation at 95 °C to activate the polymerase, 35 cycles consisting
of 94 °C for 20 s, 55 °C for 30 s, and 72 °C for 1.5 min ensued.
Five µl of product was analysed on a 1.5% agarose/TBE gel,
and 0.5 µL was combined with gs2500 TAMRA size standard
and electrophoresed on an ABI Prism 377 DNA Sequencer
(Applied Biosystems, Foster City, CA). PCR product sizes
were determined using GENOTYPER 2.0 software (Applied
Biosystems, Foster City, CA) and confirmed by a manual
examination of the chromatograms. Intron sizes were estimated
by subtracting 316 bp (297 bp from the Ras 5–1 sequence
plus 19 bp M13 universal primer) from each PCR product.
Cloning and sequence analysis of Rgc1.2(i105),
Rgc1.12(i270), Rgc1.20(i330), Rgc1.21(i360) and
Rgc1.24(i384)
PCR products were amplified from the genomic DNA of
HA370, HA372, RHA280, RHA801 and PHA using primer pair
F97/R98 and a high fidelity polymerase (Clontech Laboratories
Inc., Palo Alto, CA). Products from each genotype were
cloned into a TOPO-TA cloning vector (Invitrogen, Carlsbad,
CA), and transformed bacteria were screened by colony PCR
for inserts of different sizes. Selected clones were sequenced
using BigDye chemistry and an ABI 3700 DNA Analyser
(Applied Biosystems, Foster City, CA). Sequences were
compared and aligned using the BESTFIT and PILEUP functions,
respectively, of the GCG sequence analysis software (GCG,
Madison, WI).
Haplotype and diversity analysis
HaRGC1 haplotypes were estimated from IFLP fingerprints of
the 63 germplasm accessions (Table 1) by scoring for the
presence (1) or absence (0) of each of the 48 bands produced
by the F97/R134 primer combination. Pair-wise genetic
distances between germplasm accessions were estimated
using the weighted FST estimator of Weir and Cockerham
(1984). The calculations were performed using ARLEQUIN
(Schneider et al., 2000). FST is the count of the number of
0–1 allele mismatches between haplotypes. A minimum
spanning tree (Kruskal, 1956) was constructed from the
genetic distance matrix (G) using ARLEQUIN. TREEVIEW (Page,
2001) was used to draw the minimum spanning tree from
coordinates output by ARLEQUIN.
The 63 germplasm accessions were split into seven groups
for analyses of gene diversity within and between groups:
elite B-lines, elite R-lines, elite and exotic OP populations,
Native American land races, wild H. annuus populations, wild
H. annuus and H. praecox donors of downy mildew
resistance genes, and partially isogenic lines (PILs) developed
from wild H. annuus, H. praecox and H. argophyllus donors
of downy mildew resistance genes (Miller and Gulya, 1988,
1991). We performed an analysis of molecular variance
(AMOVA) (Excoffier et al., 1992) on the G matrix using ARLEQUIN
(Schneider et al., 2000). Gene diversities were estimated for
haplotypes (Dh) and individual bands (Dk). Dh is the probability
that the haplotypes for two randomly chosen germplasm
accessions are different, and Dk is the probability that the
genotypes (0 or 1) for two randomly chosen germplasm
accessions are different for the nth band, where n = 1, 2, …,
48. The mean of individual band diversities (Dk) was estimated
from the 48 Dk estimates. Dh and Dk were estimated for each
of the seven germplasm groups.
Mapping
Ninety-four F2 plants from the HA370 × HA372 population,
94 F6 recombinant inbred lines (RILs) from PHA × PHB, and
94 F7 RILs from RHA280 × RHA801 were used to map RGC
markers. These populations have been described previously
(Gedil et al., 2001b; Tang et al., 2002; Yu et al., 2003). RGC
markers were scored as dominant based on either the
presence/absence of the PCR product, or strong/weak
fluorescent signal, except for marker Rgc1.20(i330) which
was scored using a co-dominant allele-specific single nucle-
otide polymorphism (SNP) assay. Maps were constructed
using MAPMAKER 3.0 (Lander et al., 1987) and G-MENDEL 3.0
(Holloway and Knapp, 1993). Loci were assembled using
Table 3 The number of germplasm accessions (n), number of
haplotypes (h), number of HaRGC1 bands (p), gene diversity (Dh),
and mean gene diversity of individual bands (Dk) for HaRGC1
haplotypes in sunflower (Helianthus annuus L.) estimated from 48
intron fragment length polymorphism bands among elite B-lines,
elite R-lines, elite and exotic open-pollinated (OP) populations, Native
American land races, wild H. annuus populations, wild H. annuus
and H. praecox donors of downy mildew (Plasmopara halstedii)
resistance genes, and partially isogenic lines (PILs) developed using
wild H. annuus, H. praecox and H. argophyllus downy mildew
resistance gene donors
Group nhpD
hDk
Elite B-Lines 11 6 21 0.87 ± 0.07 0.15 ± 0.09
Elite R-Lines 13 6 22 0.72 ± 0.13 0.18 ± 0.10
OP populations 9 8 14 1.00 ± 0.06 0.12 ± 0.07
Land races 4 5 21 1.00 ± 0.13 0.21 ± 0.13
Wild populations 15 15 39 1.00 ± 0.02 0.26 ± 0.14
Wild donors 5 5 29 1.00 ± 0.13 0.31 ± 0.20
PILs 6 3 15 0.60 ± 0.22 0.11 ± 0.07
Total 63 43 48 0.98 ± 0.09 0.27 ± 0.14

Resistance gene cluster in sunflower
© Blackwell Publishing Ltd, Plant Biotechnology Journal (2003), 1, 167–185
183
likelihood odds (LOD) ratios with a LOD threshold of 3.0
and a maximum recombination frequency threshold of 0.4.
Map distances (cM) were calculated from recombination
frequency estimates using the Kosambi (1944) mapping
function.
Graphical genotype analysis of HA335-339 and RHA340
Primers for 12 SSR markers localized to LG8 (ORS70, ORS154,
ORS166, ORS185-8, ORS243, ORS299, ORS328-8, ORS399,
ORS418-8, ORS536, ORS599 and ORS780) and 16 SSR
markers localized to LG13 (ORS191-13, ORS215, ORS224,
ORS317, ORS418, ORS503, ORS511, ORS534, ORS536,
ORS581, ORS596, ORS630-13, ORS673-13, ORS730,
ORS781-13 and ORS799) of sunflower (Tang et al., 2002; Yu
et al., 2003) were used to amplify DNA from the partially
inbred lines HA335-339 and RHA340, the recurrent parent
HA89, and individuals from donor accessions ANN423 and
432, and PRA417, 419, and 424. SSR marker assays were
performed as described by Tang et al. (2002). SSR alleles
amplified from HA335-339 and RHA340 DNA that were
different than HA89 alleles were scored as ‘non-HA89’ and
presumed to be introgressed DNA. In five cases (ORS299 and
ORS399 on LG8, and ORS503, ORS536, and ORS730 on
LG13), the HA89 SSR allele was also observed in one or more
of the surrogate donors. These markers were judged to be
inconclusive and were not used. PCR-based markers for
Ha-1W23, Ha-1W41 and Ha-1W53 were developed and
scored on the graphical genotype panel by single strand
conformational polymorphism (SSCP) using primers indicated
in Table 2 and methods previously described (Slabaugh et al.,
1997). ZVG33-8 was scored on the panel using a proprietary
single nucleotide polymorphism (Advanta Semillas, Argen-
tina) and a fluorescence polarization assay (Chen et al.,
1999). Graphical genotypes were assembled manually, and
aligned beside composite reference maps of LG8 and LG13.
SSR marker locations shown on the reference chromosomes
are as shown in Yu et al. (2003), except that Ha-1W23
and Ha-1W53 were placed on the LG8 reference map by
genotyping 94 RHA280 × RHA801 RILs using SSCP markers,
and the positions of ZVG33-8 (LG8) and Ha-1W41 (LG13)
were estimated from their segregation as RFLP markers in the
HA370 × HA372 population (Gedil et al., 2001a).
Sequence data
Sequence data from this article have been deposited with
the EMBL/GENBANK Data Libraries under accession nos.
AY186604–AY186608.
Acknowledgements
We are grateful to Jerry Miller (USDA-ARS, Fargo, ND) for
helpful discussions during the course of this research, and
Judith Kolkman (Department of Crop and Soil Science,
Oregon State University) for critical reading of the manu-
script. This research was funded by grants to S. J. Knapp
from Pioneer Hi-Bred International (Johnston, IA), the USDA
National Research Initiative Competitive Grants Program
(#98-35300-6166), and USDA CSREES Initiative for Future
Agriculture and Food Systems Plant Genome Program
(#2000-04292). Oregon Agricultural Experiment Station
Technical Paper no. 11943.
Supplementary material
The following material is available from http://www.black-
wellpublishing.com/products/journals/suppmat/PBI/PBI016/
PBI016sm.htm. Table S1. Linkage group 8 resistence gene
candidate marker dataset including 50 HaRGC1 markers and a
SCAR marker linked to rust resistance. The sizes (in nucleotides)
of the HaRGC1 introns amplified with primers F97 and R134
are indicated in parentheses adjzcent to the marker names.
Table S2. ORS166 simple sequence repeat marker genotypes.
References
Aarts, M., te Lintel Hekkert, B., Holub, E., Beynon, J., Stiekema, W.
and Pereira, A. (1998) Identification of R-gene homologous DNA
fragments genetically linked to disease resistance loci in
Arabidopsis thaliana. Mol. Plant-Microbe Interact. 11, 251–258.
Bert, P., Tourvieille de Labrouhe, D., Philippon, J., Mouzeyar, S.,
Jouan, I., Nicolas, P. and Vear, F. (2001) Identification of a second
linkage group carrying genes controlling resistance to downy
mildew (Plasmopara halstedii) in sunflower (Helianthus annuus
L.). Theor. Appl. Genet. 103, 992–997.
Botella, M., Parker, J., Frost, L., Bittner-Eddy, P., Beynon, J., Daniels, M.,
Holub, E. and Jones, J. (1998) Three genes of the Arabidopsis
RPP1 complex resistance locus recognize distinct Peronospora
parasitica avirulence determinants. Plant Cell, 10, 847–1860.
Bouzidi, M.F., Badaoui, S., Cambon, F., Vear, F., Tourvieille de
Labrouhe, D., Nicolas, P. and Mouzeyar, S. (2002) Molecular analysis
of a major locus for resistance to downy mildew in sunflower with
specific PCR-based markers. Theor. Appl. Genet. 104, 592–600.
Brahm, L., Rocher, T. and Friedt, W. (2000) PCR-based markers
facilitating marker assisted selection in sunflower for resistance to
downy mildew. Crop Sci. 40, 676–682.
Chen, X., Levine, L. and Kwok, P.-Y. (1999) Fluorescence polarization
in homogeneous nucleic acid analysis. Genome Res. 9, 492–498.
Cheres, M.T. and Knapp, S.J. (1998) Ancestral origins and genetic
diversity of cultivated sunflower: coancestry analysis of public
germplasm. Crop Sci. 38, 1476 –1482.
Chin, D.B., Arroyo-Garcia, R., Ochoa, O.E., Kesseli, R.V., Lavelle, D.O.
and Michelmore, R.W. (2001) Recombination and spontaneous

Mary B. Slabaugh et al.
© Blackwell Publishing Ltd, Plant Biotechnology Journal (2003), 1, 167–185
184
mutation at the major cluster of resistance genes in lettuce
(Lactuca sativa). Genetics, 157, 831–849.
Excoffier, L., Smouse, P. and Quattro, J. (1992) Analysis of molecular
variance inferred from metric distances among DNA haplotypes:
Application to human mitochondrial DNA restriction data.
Genetics, 131, 479–491.
Fick, G.N. (1978) Breeding and Genetics. In: Sunflower Science and
Technology, Agronomy 19. (Carter, J.F., ed.), pp. 279–338.
Madison, WI: American Society of Agronomy, Crop Science
Society of America, and Soil Science Society of America.
Fick, G.N., Kinman, M.L. and Zimmer, D.E. (1975) Registration of
‘RHA273’ and ‘RHA274’ sunflower parental lines. Crop Sci. 15,
106.
Fick, G.N. and Zimmer, D.E. (1975) Linkage tests among genes for
six qualitative characters in sunflowers. Crop Sci. 15, 777–779.
Fick, G.N. and Zimmer, D.E. (1979) Registration of six nonoilseed
sunflower parental lines. Crop Sci. 19, 422.
Fick, G.N., Zimmer, D.E. and Kinman, M.L. (1974) Registration of six
sunflower parental lines. Crop Sci. 14, 912.
Fluhr, R. (2001) Sentinels of disease. Plant resistance genes. Plant
Physiol. 127, 1367–1374.
Gedil, M.A., Slabaugh, M.B., Berry, S., Johnson, R., Michelmore, R.,
Miller, J., Gulya, T. and Knapp, S.J. (2001a) Candidate disease
resistance genes in sunflower cloned using conserved nucleotide-
binding site motifs: Genetic mapping and linkage to the downy
mildew resistance gene. Plant Genome, 44, 205–212.
Gedil, M.A., Wye, C., Berry, S., Segers, B., Peleman, J., Jones, R.,
Leon, A., Slabaugh, M. and Knapp, S.J. (2001b) An integrated
restriction fragment length polymorphism-amplified fragment
length polymorphism linkage map for cultivated sunflower.
Genome, 44, 213–221.
Gentzbittel, L., Mouzeyar, S., Badaoui, S., Mestries, E., Vear, F., De
Labrouhe, D. and Nicolas, P. (1998) Cloning of molecular markers
for disease resistance in sunflower, Helianthus annuus L. Theor.
Appl. Genet. 96, 519–525.
Gentzbittel, L., Vear, F., Zhang, Y.-X., Berville, A. and Nicolas, P.
(1995) Development of a consensus linkage RFLP map of
cultivated sunflower (Helianthus annuus L.). Theor. Appl. Genet.
90, 1079–1086.
Grube, R., Radwanski, E. and Jahn, M. (2000) Comparative genetics of
disease resistance within the Solanaceae. Genetics, 155, 873–887.
Gulya, T.J., Marty, D., Harbour, J., Holen, C., Knodel, J., Lamey, A.
and Mason, P. (1999) Metalaxyl resistance in sunflower downy
mildew in North America. In: Proceedings Sunflower Research
Workshop, pp. 118–123, 14–15 January 1999. Fargo, ND.
Holloway, J.L. and Knapp, S.J. (1993) G-MENDEL 3.0 User’s Guide,
pp. 1–130. Corvallis, OR: Oregon State University.
Holub, E.B. (2001) The arms race is ancient history in Arabidopsis,
the wildflower. Nat. Rev. Genet. 2, 516–527.
Kanazin, V., Marek, L.F. and Shoemaker, R.C. (1996) Resistance
gene analogues are conserved and clustered in soybean. Proc.
Natl. Acad. Sci. USA, 93, 11746–11750.
Korell, M., Moosges, G. and Friedt, W. (1992) Construction of a
sunflower pedigree map. Helia, 15, 7–16.
Kosambi, D.D. (1944) The estimation of map distance from
recombination values. Ann. Eugen. 12, 172–175.
Kruskal, J.B. (1956) On the shortest spanning subtree of a graph and
the traveling salesman problem. Proc. Am. Math. Soc. 7, 48–50.
Lander, E.S., Green, P., Abrahamson, J., Barlow, A., Daly, M.J.,
Lincoln, S.E. and Newburg, L. (1987) MAPMAKER: an interactive
computer package for constructing primary genetic linkage maps
of experimental and natural populations. Genomics, 1, 174–181.
Lawson, W.R., Goulter, K.C., Henry, R.J., Kong, G.A. and Kochman, J.K.
(1998) Marker-assisted selection for two rust resistance genes in
sunflower. Mol. Breed. 4, 227–234.
Leclercq, P., Cauderon, Y. and Dauge, M. (1970) Selection pour la
resistance au mildious du Tournesol a partir d’hybrides Topinam-
bour x Tournesol. Ann. Amelior. Plantes, 20, 363–373.
Leister, D., Kurth, J., Laurie, A., Yano, M., Sasaki, T., Devos, K.,
Graner, A. and Schulze-Lefert, P. (1998) Rapid reorganization of
resistance gene homologues in cereal genomes. Proc. Natl. Acad.
Sci. USA, 95, 370–375.
Meyers, B., Chin, D., Shen, K., Sivaramakrishnam, S., Lavelle, D.,
Zhang, Z. and Michelmore, R. (1998) The major resistance gene
cluster in lettuce is highly duplicated and spans several megabases.
Plant Cell, 10, 1817–1832.
Meyers, B.C., Dickerman, A.W., Michelmore, R.W., Sivaramakrishnan, S.,
Sobral, B.W. and Young, N. (1999) Plant disease resistance genes
encode members of an ancient and diverse protein family within
the nucleotide-binding superfamily. Plant J. 20, 317–332.
Miller, J.F. (1992) Registration of five oilseed sunflower germplasm
restorer lines (RHA373–377) and two nuclear male-sterile
populations (nms 274 and 801). Crop Sci. 32, 1298.
Miller, J.F. and Gulya, T.J. (1988) Registration of six downy mildew
resistant sunflower germplasm lines. Crop Sci. 28, 1040–1041.
Miller, J.F. and Gulya, T.J. (1990) Registration of ten oilseed
sunflower germplasm lines. Crop Sci. 30, 430–431.
Miller, J.F. and Gulya, T.J. (1991) Inheritance of resistance to race 4
of downy mildew derived from interspecific crosses in sunflower.
Crop Sci. 31, 40–43.
Miller, J.F. and Gulya, T.J. (1999) Registration of eight schlerotinia-
tolerant sunflower germplasm lines. Crop Sci. 39, 301–302.
Mouzeyar, S., Roeckel-Drevet, P., Gentzbittel, L., Philippon, J.,
Tourvieille de Labrouhe, D., Vear, F. and Nicolas, P. (1995) RFLP
and RAPD mapping of the sunflower Pl1 locus for resistance to
Plasmopara halstedii race 1. Theor. Appl. Genet. 91, 733–737.
Noel, L., Moores, T.L., van der Biezen, E.A., Parniske, M., Daniels,
M.J., Parker, J.E. and Jones, J.D.G. (1999) Pronounced intraspe-
cific haplotype divergence at the RPP5 complex disease resistance
locus of Arabidopsos. Plant Cell, 11, 2099–2111.
Page, R.D.M. (2001) TreeView <http://taxonomy.zoology.gla.ac.uk/
rod>.
Pustovoit, G.T. (1966) Interspecies hybridization as a method of
sunflower selection on group immunity. Genetika, 2, 59–69.
Putt, E.D. (1978) History and present world status. In: Sunflower
Science and Technology (Carter, J.F., ed.), pp. 1–29. Madison,
WI: American Society of Agronomy, Crop Science Society of
America, and Soil Science Society of America (in Russian).
Putt, E.D. and Sackston, W.E. (1957) Studies on sunflower rust. I.
Some sources of rust resistance. Can. J. Plant Sci. 37, 43–54.
Rahim, M., Jan, C.C. and Gulya, T.J. (2002) Inheritance of resistance
to sunflower downy mildew races 1, 2 and 3 in cultivated
sunflower. Plant Breed. 121, 57–60.
Rieseberg, L.H. and Seiler, G.J. (1990) Molecular evidence and the
origin and development of the domesticated sunflower (Helian-
thus annuus, Asteraceae). Econ. Bot. 44, 79–91.
Roath, W.W., Miller, J.F. and Gulya, T.J. (1981) Registration of
RHA801 sunflower germplasm. Crop Sci. 21, 479.
Roeckel-Drevet, P., Gagne, G., Mouzeyar, S., Gentzbittel, L., Philippon, J.,
Nicolas, P., Tourvieille de Labrouhe, D. and Vear, F. (1996)

Resistance gene cluster in sunflower
© Blackwell Publishing Ltd, Plant Biotechnology Journal (2003), 1, 167–185
185
Colocation of downy mildew (Plasmopara halstedii) resistance
genes in sunflower (Helianthus annuus L.). Euphytica, 91,
225–228.
Schneider, S., Roessli, D. and Excoffier, L. (2000) Arlequin software
for population genetic analysis <http://anthro.unige.ch/arlequin>.
Schuelke, M. (2000) An economic method for the fluorescent
labeling of PCR fragments. Nat. Biotechnol. 18, 233–234.
Shen, K., Meyers, B., Islam-Faridi, M., Chin, D., Stelly, D. and
Michelmore, R. (1998) Resistance gene candidates identified by
PCR with degenerate oligonucleotide primers map to clusters
of resistance genes in lettuce. Mol. Plant-Microbe Interact. 8,
815–823.
Slabaugh, M.B., Huestis, G.M., Leonard, J., Holloway, J.L., Rosato,
C., Hongtrakul, V., Martini, N., Toepfer, R., Voetz, M., Schell, J.
and Knapp, S.J. (1997) Sequence-based genetic markers for
genes and gene families: Single-strand conformational polymor-
phisms for the fatty acid synthesis genes of Cuphea. Theor. Appl.
Genet. 94, 400–408.
Stankiewicz, P. and Lupski, J. (2002) Genome architecture, rear-
rangements and genomic disorders. Trends Genet. 18, 74–82.
Tan, A.S., Jan, C.-C. and Gulya, T.J. (1992) Inheritance of resistance
to race 4 of sunflower downy mildew in wild sunflower acces-
sions. Crop Sci. 32, 949–952.
Tang, S., Yu, J.K., Slabaugh, M., Shintani, D.K. and Knapp, S.J.
(2002) Simple sequence repeat map of the sunflower genome.
Theor. Appl. Genet. 105, 1124–1136.
Tang, S. and Knapp, S.J. (2003) Microsatellites uncover extraordinary
dsiversity in Native American land races and wild populations of
cultivated sunflower. Theor. Appl. Genet., in press.
The Arabisopsis Genome Initiative. (2000) Analysis of the genome
sequence of the flowering plant Arabidopsis thaliana. Nature,
408, 796–815.
Van der Hoorn, A.L., de Wit, J. and Joosten, M. (2002) Balancing
selection favors guarding resistance proteins. Trends. Plant Sci. 7,
67–71.
Vear, F., Gentzbittel, L., Philippon, J., Mouzeyar, S., Mestries, E.,
Roeckel-Drevet, P., Tourvieille de Labrouhe, D. and Nicolas, P.
(1997) The genetics of resistance to five races of downy mildew
(Plasmopara halstedii) in sunflower (Helianthus annuus L.). Theor.
Appl. Genet. 95, 584–589.
Vranceanu, V. and Stoenescu, F. (1970) Imunitate la mana florri-
soarelui, conditionata monogenic. Probleme Aricole, 2, 34–40.
Webb, D.M. and Knapp, S.J. (1990) DNA extraction for a previously
recalcitrant plant genus. Mol. Biol. Rep. 8, 180–185.
Weir, B.S. and Cockerham, C.C. (1984) Estimating F-statistics for the
analysis of population structure. Evolution, 38, 1358–1370.
Yu, Y., Buss, G. and Saghai Maroof, M. (1996) Isolation of a super-
family of candidate disease-resistance genes in soybean based on
a conserved nucleotide-binding site. Proc. Natl. Acad. Sci. USA,
93, 11751–11756.
Yu, J.-K., Tang, S., Slabaugh, M.B., Heesacker, A., Cole, G., Herring, M.,
Soper, J., Han, F., Chu, W.-C., Webb, D.M., Thompson, L.,
Edwards, K.J., Berry, S., Leon, A.J., Olungu, C., Maes, N. and
Knapp, S.J. (2003) Towards a saturated molecular genetic linkage
map for cultivated sunflower. Crop Sci. 43, 367–387.
Zimmer, D.E. and Kinman, M.L. (1972) Downy mildew resistance
in cultivated sunflower and its inheritance. Crop Sci. 12, 749–
751.