Haplotyping, linkage mapping and expression analysis of barley genes regulated by terminal drought stress influencing seed quality
BMC Plant Biology
Haplotyping, linkage mapping and expression analysis of barley genes regulated by terminal drought stress influencing seed quality
Sebastian Worch 0
Kalladan Rajesh 0
Vokkaliga T Harshavardhan 0
Christof Pietsch 2
Viktor Korzun 2
Lissy Kuntze 1
Andreas Brner 0
Ulrich Wobus 0
Marion S Rder 0
Nese Sreenivasulu 0
0 Leibniz-Institute of Plant Genetics and Crop Plant Research (IPK) , Corrensstr.3, 06466 Gatersleben , Germany
1 Nordsaat Saatzucht GmbH , Bohnshauser Strae 1, 38895 Langenstein , Germany
2 KWS LOCHOW GmbH , Ferdinand- von-Lochow-Str.5, 29303 Bergen , Germany
Background: The increasingly narrow genetic background characteristic of modern crop germplasm presents a challenge for the breeding of cultivars that require adaptation to the anticipated change in climate. Thus, high priority research aims at the identification of relevant allelic variation present both in the crop itself as well as in its progenitors. This study is based on the characterization of genetic variation in barley, with a view to enhancing its response to terminal drought stress. Results: The expression patterns of drought regulated genes were monitored during plant ontogeny, mapped and the location of these genes was incorporated into a comprehensive barley SNP linkage map. Haplotypes within a set of 17 starch biosynthesis/degradation genes were defined, and a particularly high level of haplotype variation was uncovered in the genes encoding sucrose synthase (types I and II) and starch synthase. The ability of a panel of 50 barley accessions to maintain grain starch content under terminal drought conditions was explored. Conclusion: The linkage/expression map is an informative resource in the context of characterizing the response of barley to drought stress. The high level of haplotype variation among starch biosynthesis/degradation genes in the progenitors of cultivated barley shows that domestication and breeding have greatly eroded their allelic diversity in current elite cultivars. Prospective association analysis based on core drought-regulated genes may simplify the process of identifying favourable alleles, and help to understand the genetic basis of the response to terminal drought.
Drought is one of the most serious abiotic stress factors
which occur throughout the development of the plant
and, if sufficiently severe and/or prolonged, results in
the modification of the plants physiology and severely
limit crop productivity. Plants have evolved a range of
defence and escape mechanisms , and these are
typically mediated by multiple rather than by single genes.
In barley, QTL underlying drought tolerance has been
mapped to almost every chromosome [2-6]. However,
little information has been gathered to date regarding
the genomic location of drought-regulated genes, either
expressed throughout plant development or at late
reproductive stages influencing seed yield and quality.
Of all the genetic marker types available, single
nucleotide polymorphisms (SNPs) are the most
abundant, and thus offer the greatest level of genetic
resolution. They are of potential functional relevance and they
are also well suited to high throughput analytical
methods . The representation of SNPs on the barley
linkage map has grown over recent years [8-10], and in
particular, a SNP-based map featuring gene sequences
expressed differentially in response to various abiotic
stresses has recently been developed . Here we
present a SNP-based genetic map of barley, specifically
focussing on nucleotide variation in ESTs demonstrated
to be involved in the response of barley to drought
stress occurring at early vegetative stages, during
anthesis and the grain filling process.
Table 1 Marker frequency and map length of the
individual mapping populations for deriving the
While the productivity of the cereals has risen greatly
since their domestication, in response to farmer
selection and methodical breeding, there are indications that
the increasing fixation of elite alleles in modern
breeding germplasm is already inhibiting further genetic gain.
In the face of potential climate change, these elite allele
combinations may become sub-optimal and will
necessitate a search for better adapted alleles among crop
landraces or wild materials . Population of wild barley
(Hordeum vulgare ssp. Spontaneum, hereafter referred
to as H. spontaneum) have been shown to possess
favourable genetic variation for a number of agronomic
traits [12,13] including biotic [14,15] and abiotic stress
We report haplotyping data for 17 starch biosynthesis/
degradation genes demonstrating the broad diversity
among H. spontaneum accessions and H. vulgare
landraces but rather limited genetic variance in the current
elite breeding germplasm by fixing certain haplotypes.
Similar observations were made for seed starch
accumulation during terminal drought for a diverse set of 50
Results and Discussion
SNP discovery in sequences responding to drought stress
The initial set of 613 drought-responsive ESTs (covering
20 functional categories; Additional file 1) was
determined from 5 or 21 day old seedlings, flag leaves-post
anthesis or developing grains. Suitable sequence
information from the four parents of mapping population
and the four advanced backcross (AB) population
parents were obtained for 327 genes (53.3%). The sequence
reads were assembled individually for each locus. A total
of 1,346 informative SNPs were dispersed through
263 of the sequences, giving a mean SNP density of
5.1 per kb (Additional file 2). The Oregon Wolfe parents
were the best differentiated (627 SNPs across 181 ESTs,
density 3.4 per kb), which is consistent with
comparisons made elsewhere between these two lines [7,20].
Some 30% of the loci were polymorphic between cvs.
Steptoe and Morex, as noted in the previous studies for
these cultivars [10,20]. The proportion of informative
loci in cv. Brenda versus HS584 was 33%, and between
cv. Scarlett and ISR42_8 39%. Note that a polymorphism
survey based on 400 microsatellite loci showed that 46%
were informative between cv. Brenda and HS584 ,
while 97 out of 220 (44%) were polymorphic between
cv. Scarlett and ISR42_8 .
Marker development and linkage mapping
The SNPs present in the 263 ESTs were converted into
31 pyrosequencing-based markers for Steptoe/Morex, 76
for Oregon Wolfe and 34 markers common to both
populations, for a total of 141 SNP markers (Table 1).
Of the 20 functional gene categories represented among
the 613 initially selected ESTs, 17 classes were retained
among the genes tagged by the 141 markers (Additional
file 1). Genes involved in carbohydrate, amino acid
metabolism, hormone signalling, storage protein
synthesis and the response to desiccation, as well as a number
of transcription factors were particularly common
(Additional file 1). Genotypic data associated with both
the 141 de novo SNP markers (GBS3120-GBS3260), and
with an established set of 140 GBS (GBS0001-GBS0921;
) and 71 BIN markers were then used to construct a
352 marker-based map (Figure 1), in which the BIN
markers were situated as expected [10,23]. The only
change in GBS marker order occurred on chromosome
arm 3HL, where GBS0538 mapped distal, rather than
proximal to ABC161 . The genetic length of each
chromosome ranged from 127.2 cM (4H) to 198.8 cM
(5H), and the overall map length was 1,072 cM (Table 1).
Given the unequal genomic distribution of the marker
loci, marker development was focussed on chromosomes
1 H (32 loci) and 2 H (28 loci), because these chromo
somes are known to harbour drought-related QTL
(unpublished data and [3,4,6]). For example, Teulat
et al.  identified a QTL for drought related traits at
the SSR marker Ebmac684 on 2 H analysing grain
material from field grown barley from an environment with
limited water availability especially during the grain
filling period. The marker Ebmac684 maps close to
ABC468 , in a chromosomal region where several
de novo markers representing putative candidate genes
were mapped. These genes encode transcription
regulators (GBS3215, GBS3217, GBS3224), a cytochrome
protein (GBS3138), a protein kinase (GBS3167) and the
starch branching enzyme (GBS3257). Chromosomes 4 H
(nine loci) and 6 H (ten loci) contained the least
de novo marker, while 21, 22 and 19 loci were mapped
to chromosomes 3 H, 5 H and 7 H, respectively. Each
member of the pairs of sequences GBS3141/GBS3216,
GBS3193/GBS3250, GBS3129/GBS3260, and GBS3150/
Taken the consensus transcript map in  five of the
GBS3223 was derived from the same EST, and thus
de novo SNP loci are represented there, namely GBS3178/
mapped to the same position (Additional files 3 and 4).
GBS0237 (chromosome 1H), GBS3158/GBS0400
(chroThe pairs GBS3230/GBS3231, GBS3172/GBS3173 and
mosome 2H), GBS3246/GBS0073 and GBS3170/GBS0043
ferent EST clusters but represent the same gene as they
7H). A further 14 GBR or GBM markers identified the
do not overlap due to shorter contigs, and mapped to a
same loci as the de novo SNPs, but two (GBS3139 on
single chromosome bin (Additional files 3 and 4).
Overlap with other barley SNP maps
chromosome 1 H, GBR1494 on chromosome 2H; and
GBS3207 on chromosome 1 H, GBR1571 on chromosome
2H) had a discrepant chromosome location. The pairs
Only seven of the previously mapped abiotic stress related
GBS3253/GBR0625 and GBS3185/GBM1405 all mapped
responsive 141 de novo SNP markers  (Additional file
4). Another high-density transcript linkage map based on
4). GBS3193 and GBS3250 belong to the same mapped
a total of 2890 SNP, CAPS and INDEL markers was
1 H in . On chromosome 2 H, GBS3244 is covered by markers show overlap with 28 loci of the present map. scsnp00592, GBS3138 by scsnp01644 and GBS3158 Finally, 67 of the 2,943 SNP loci present on the Close
by scsnp03343. GBS3198 (chromosome 4H) corresponds
et al.  map correspond to GBS marker(s), with no
disto scsnp06435, and GBS3247 (chromosome 5H) to
crepancies in terms of chromosomal location. Marker
scsnp14350. Six of the seven overlapping markers mapped
1_0686 (matching GBS3207 and GBR1571 ) was
to their expected chromosomal BIN, but GBS3244
located to chromosome 1 H, thereby confirming the
posiloci of drought-responsive genes represent novel means
for characterizing the genetic basis of drought tolerance in
barley and they may also provide useful information for
the construction of the barley physical map as the next
step towards genome sequencing.
The drought stress response of mapped transcripts over
To reveal the drought stress response of mapped
transcripts during various stages of development, we
normalized the expression data by utilizing the publicly
available expression data sets deposited in Gene
Expression Omnibus (GEO) from five (GEO accession series
number: GGSE3170) and 21 (GSE6990) day old
seedlings, flag leaves post anthesis (GSE15970), green spike
tissues (awn, lemma and palea, GSE17669) and own
data from developing grain during 20 days after
fertilization (DAF). A range of barley cultivars has been used to
generate these data, including the drought tolerant cv.
Martin and the susceptible cv. Moroc9-75, parents of
mapping and AB populations (OWB-D, OWB-R, Morex,
Brenda and Hs584). The clustering process identified
three major groups: groups 1 and 2 contained genes
which are up-regulated as a result of drought stress,
while the ones in group 3 were down-regulated (Figure 2).
While group 2 genes showed up-regulation mostly in early
vegetative tissues, group 1 members were up-regulated
across all developmental stages, and were expressed in a
range of organs (seedlings, flag leaf, lemma, palea, and
awn and to a lesser extent in the developing grain). Thus,
group 1 genes could be considered to represent a core set
of drought responsive genes. The functional groups
particularly overrepresented in groups 1 and 2 included
transcription regulators, genes induced by abiotic stress, genes
responsible for the synthesis of storage proteins and genes
related to amino acid and carbohydrate metabolism, and
ABA-induced hormone related genes, calculated by
Fishers exact test with a P-value cut off 0.01 (Figure 2 and
Additional file 3).
An ABA signalling gene (protein phosphatase 2C,
marker GBS3123), a bZIP ABA-responsive element binding
protein (GBS3212) were consistently up-regulated by
drought throughout development in barley (Figure 3). In
A. thaliana, protein phosphatase 2C regulates a
Snf1related kinase , and mediates signal transduction to
an ABF2 transcription factor . Thus in barley, it
seems likely that an ABA signalling pathway
orchestrates the adaptive response to drought, not just at the
seedling stage but also in the flag leaf, awn, lemma
and palea (Figure 3). In addition several Ras family
Gproteins (GBS3161, GBS3162, GBS3163, GBS3245)
thought to be involved in ABA signalling are found to
be induced in 21 day seedlings and flag leaf (Figure 3).
Several ABA-induced late embryogenesis abundant
proteins (GBS3120, GBS3121, GBS3248) were induced to
drought in seedlings (Figure 3), and these have been
shown previously to be involved in desiccation tolerance
. A number of ABA signalling related genes were
included in the genetic map (Additional file 3). Other
transcription factors were induced by drought in a
nonorgan specific manner; these included AP2/ERF II
(GBS3206), VII (GBS3208), VIII (GBS3207), bHLH
(GBS3210), bZIP (GBS3212, GBS3211), MYB (GBS3142,
GBS3145, GBS3219), NAC (GBS3146, GBS3147) and
several other unclassified factors (Figure 3). The specific
function(s) of most of these regulators remains unclear,
but their up-regulation by drought stress indicates that
they probably do play a role in the plants response to
Abiotic stress induced genes
Genes encoding dehydrin 9, universal stress proteins,
hydrophobic proteins and various classes of heat shock
proteins (HSPs) were induced by drought across all the
developmental stages (Figure 2 group 1). Among the
HSPs were HSP70 (GBS3180); HSP81-1 (GBS3182) and
HSP26 (GBS3181), which mapped, respectively, to
chromosomes 1 H, 2 H and 4 H (Additional file 3). Other
HSPs were not so generally up-regulated by drought.
The up-regulation of HSP is consistent with their
presumed protection of proteins from oxidative damage
induced by drought stress .
Drought response of mapped transcripts contributing to
Barley grain storage proteins comprise a mixture of four
distinct prolamin polypeptides: the B- and g-
(sulphurrich) hordeins, the C- (sulphur-poor) hordeins and the
high molecular weight D-hordeins. The hordein genes
are known to be organised in clusters encoding the
B-hordeins (Hor2 and Hor4), C-hordeins (Hor1),
g-hordein (Hor5) and D-hordein (Hor3) which are all located
on chromosome 1 H . The present genetic map
showed that GBS3200, a marker for B1-hordein, lay
near the telomere of chromosome 1 H, while GBS3205
(marking another B1-hordein) was linked closer to
GBS3202 (B3-hordein), around 11 cM distant from
GBS3201 (g1-hordein). A third B-hordein marker
(GBS3204) was placed further apart, closer to
g3-hordein. Thus the B-hordein gene family is represented by
at least three different loci on the short arm of
chromosome 1 H, while the g-hordein genes also map to two
distinct loci on the same chromosome arm (Figure 4).
The regulation of hordein family gene transcription
includes DNA methylation [30,31] and the concerted
action of distinct transcription factor families [32,33].
The expression of all the sulphur-rich hordein genes
was promoted by drought in the awn, lemma and palea
amino acid metabolism
amino acid metabolism
amino acid metabolism
GBS3160: Contig7501_s_at: signalling.calcium
GBS3165: Contig4218_at: signalling.phosphinositides
GBS3164: Contig3562_at: signalling.phosphinositides
GBS3166: Contig13498_at: signalling.receptor kinases.DUF 26
GBS3123: Contig9585_at: hormone metabolism.abscisic acid.signal transduction
GBS3121: Contig6276_s_at: hormone metabolism.abscisic acid.induced-regulated
GBS3248: Contig1830_at: hormone metabolism.abscisic acid.induced-regulated
GBS3120: Contig8149_at: hormone metabolism.abscisic acid.induced-regulated
GBS3162: Contig3165_at: signalling.G-proteins
GBS3161: Contig5611_at: signalling.G-proteins
GBS3163: Contig10901_at: signalling.G-proteins
GBS3245: Contig3167_s_at: signalling.G-proteins
GBS3247: Contig14350_at: signalling.receptor kinases.Catharanthus roseus-like RLK1
GBS3206: HA11J15u_s_at: AP2/EREBP family
GBS3208: Contig3914_s_at: AP2/EREBP family
GBS3210: Contig13678_s_at: bHLH,Basic Helix-Loop-Helix family
GBS3212: Contig21149_s_at: bZIP transcription factor family
GBS3141: Contig8202_at: C3H zinc finger family
GBS3145: Contig8571_at: MYB domain transcription factor family
GBS3142: Contig9706_at: MYB-related transcription factor family
GBS3147: Contig3361_at: NAC domain transcription factor family
GBS3146: Contig5241_at: NAC domain transcription factor family
GBS3148: Contig7464_at: putative DNA-binding protein
GBS3151: HVSMEf0011I05r2_s_at: transcription factor unclassified
GBS3157: Contig10344_at: transcription factor unclassified
GBS3207: Contig6636_at: AP2/EREBP family
GBS3209: HVSMEh0086A12r2_s_at: Argonaute
GBS3211: Contig9253_at: bZIP transcription factor family
GBS3213: Contig13989_at: C2C2(Zn) DOF zinc finger family
GBS3214: Contig20418_at: C2C2(Zn) DOF zinc finger family
GBS3139: Contig13200_at: C2C2(Zn) GATA transcription factor family
GBS3140: Contig9333_s_at: C2H2 zinc finger family
GBS3217: Contig5444_s_at: GRAS transcription factor family
GBS3143: Contig17371_at: Histone acetyltransferases
GBS3219: Contig8132_at: MYB domain transcription factor family
GBS3222: Contig3819_at: putative DNA-binding protein
GBS3149: Contig6099_at: putative DNA-binding protein
GBS3153: Contig8947_at: transcription factor unclassified
(Figure 4). Hordein transcripts first appear in the
endosperm at 12 days post anthesis, peaking in expression by
16 days, and then maintaining this level until grain
maturity [34,35]. The B1-hordein genes were induced in
developing seeds by drought stress in cv. Brenda, but
less prominently so in HS584 (Figure 4), indicating
distinct differences in B-hordein gene expression between
cultivated barley and its wild relative. Correspondingly,
the seed nitrogen/protein content also increased under
drought in Brenda but not in HS584. However, the
GBS3200: X01778_x_at: hordein B1
GBS3205: Contig524_x_at: hordein B1
GBS3202: Contig209_s_at: gamma 3 hordein
GBS3201: Contig518_s_at: gamma 1 hordein
GBS3204: Contig585_x_at: hordein B
Pcr2 gamma 1 hordein
Figure 4 The cluster of sulphur-rich hordein genes on the
short-arm barley chromosome 1 H (left panel) and their
corresponding expression profiles during development. For
abbreviations, see Figure 2 legend. Expression data from individual
replications are given in Additional file 3. In the lower panel,
percent crude protein estimated based on seed nitrogen (N%) for
the two parents of introgression line population (H.vulgare Brenda
and H. spontaneum 584) from control and drought stress treatments
B B W W (B (H
iltr(ng_odughOW iltrng_doug_hOW lisyedeng8_3S% lisyedeng_19S% lisyeden_g7S%W lftr(a_ed1odghuM lftr(a_ed3dohguM lftr(a_ed5dohguM lftr(a_ed1dohguM lftr(a_e3dodhguM lftr(ea_5ddohguM tr(a_4ddohguM tr()_4ddoughM tr)(4ddoughM tr()4d_doughM tr02_doghuFAD tr02_doghuFAD
dee dee ad1 ad1 ad1 lga lga lga lga lga lga emm laea _nw dee dee dee
S S 2 2 2 F F F F F F L P A S S S
absolute levels remained high in the control plants
In contrast, the down-regulation of the gene family
members of key starch biosynthesis genes, sucrose
synthase, ADP-glucose pyrophosphorylase are
downregulated by terminal drought stress in the post anthesis
period during 20 DAF (Figure 5A). Several genes
associated with the activity of the starch branching enzyme
became activated by terminal drought stress, which has
implications for the synthesis of amylopectin. Certain
genes involved in starch degradation (e.g., those
encoding sd1--amylase and chloroplast-targeted -amylase)
were also induced by drought stress, which points to a
concerted fine tuning of starch biosynthesis and
degradation in impairing seed starch accumulation and seed
quality. However, many genes associated with
carbohydrate metabolism including the genes encoding sucrose
synthase type I (GBS3129), ADP-glucose
pyrophosphorylase large subunit (GBS3259) and starch branching
enzyme class II (GBS3257) were up-regulated by
drought stress in seedlings, the flag leaf, the awn, lemma
and palea (Figure 5A). The production of starch in
vegetative tissues of Arabidopsis thaliana has been found to
be negatively correlated with plant biomass .
Likewise, we might expect that starch accumulation in
vegetative tissues negatively affects plant growth under
Sucrose synthase I
H1 T A C C T InDel A G G C C C A A C C C G T
H2 T A C C T InDel A G G C C C A
H4 C A C T C InDel G C C T C C A
GBS3128: Contig4521_s_at: sucrose-1-fructosyltransferase
GBS3127: Contig101_at: fructokinase I
STn13: Contig4153_at: hexokinase
GBS3129: Contig361_s_at: sucrose synthase 1
STn16: Contig460_s_at: sucrose synthase 1
GBS3258: Contig481_at: sucrose synthase 2
STn10: Contig481_s_at: sucrose synthase 2
STn02: Contig823_at: sucrose synthase 3
GBS3259: Contig3390_at: ADP-glucose pyrophosphorylase large subunit
STn19: Contig2267_s_at: ADP-glucose pyrophosphorylase smal subunit A
GBS3256: Contig10765_at: ADP-glucose pyrophosphorylase smal subunit B
STn22: Contig1808_at: starch synthase I
STn17: Contig12208_at: granule bound starch synthase Ib
STn08: Contig3541_s_at: starch branching enzyme I
GBS3257: Contig3761_at: starch branching enzyme 2
GBS3235: Contig11648_at: limit dextrinase
GBS3246: Contig3114_at: triose phosphate translocator
STn20: Contig1411_s_at: beta-amylase
STn21: Contig1406_at: Sd1 beta-amylase 1
GBS3126: Contig11522_at: chloroplast-targetedbeta-amylase
GBS3125: Contig3952_at: alpha-amylase
Sucrose synthase II
Additional file 6.
SNPs InDels Approx. sequence length
BLAST search result
HY09D18 Sucrose synthase 1
HF08A21 Sucrose synthase 2
HA31O14 Sucrose synthase 2
HY04O18 ADP-glucose pyrophosphorylase large subunit
HB16O10 ADP-glucose pyrophosphorylase small subunit (alternatively
HA31F12 ADP-glucose pyrophosphorylase small subunit
HB05N09 Starch synthase I
HF05C15 Starch synthase IV
Granule-bound starch synthase 1b
HB30O07 Starch branching enzyme I
HB21K16 Starch branching enzyme IIa
HF17A10 Beta amylase
HF11O03 Sd 1 beta-amylase
HZ60P11 Alpha glucosidase
HB20O07 Gamma 2 hordein
Haplotype analysis of carbohydrate metabolism genes
A detailed analysis of sequence variants within 17 starch
biosynthesis/degradation genes was conducted for a core
set of 32 accessions, which included landraces, elite
breeding lines, the mapping population parents and H.
spontaneum. This delivered 180 polymorphic sites
(SNPs and indels) across both intronic and exonic
sequence, and led to the recognition of 78 haplotypes
(Table 2). Overall the elite breeding lines, including cv.
Brenda, showed little haplotypic variation, but the
remaining materials fell into a number of haplotype
groups indicating broader genetic diversity. Figure 5B
summarizes the variation present within the genes
encoding sucrose synthase types I (CR-EST:HY09D18,
marker: GBS3129) and II (CR-EST:HA31O14, CR-EST:
HF08A21; GBS3258) whereas the haplotyping data for
the remaining genes are listed in Additional file 5.
Within the 360 bp re-sequenced region of the sucrose
synthase type I amplicon, 18 SNPs and a 3 bp indels
were found. Among the SNPs, 11 were situated within
an intron and seven (six synonymous) within an exon;
the single non-synonymous SNP was a transition variant
present in cv. Morex, which converted a glycine residue
to a serine. The accessions could be classified into five
haplotypic groups (H1-H5), the largest of which (H5)
included all the elite breeding lines and half of the
remaining H. vulgare accessions. H2 contained only one
entry (cv. Morex), as did H1 (HS584). H3 captured
several H. vulgare and the other H. spontaneum accessions,
as well as the Oregon Wolfe dominant parent. The
Oregon Wolfe recessive parent fell into H4 along with two
other H. vulgare lines (Additional file 5).
GBS3258 represented about 550 bp of the sucrose
synthase type II sequence, and the re-sequencing of 29
accessions generated 14 SNPs. These allowed the
recognition of seven haplotypes (H1-H7), of which H2, H3
and H7 each contained only one accession. The elite
breeding lines were split among the two major groups
H1 and H6, along with most of the H. vulgare
accessions, although H6 also included ISR42-8, an H.
spontaneum accession. Groups H4 and H5 each contained
three accessions, the former containing the remaining
H. spontaneum accessions, and the latter the remaining
H. vulgare ones.
The relatively high level of haplotype diversity in these
two sucrose synthase genes among non-elite lines
suggests that these genes have experienced selection
processes during the course of domestication and farmers
selection. However, for improving sink strength specific
haplotypes (H5 from sucrose synthase I, H1 and H6
from sucrose synthase II) were fixed in the elite lines
during the breeding. In maize, key starch biosynthesis
enzymes and soluble carbohydrates were measured from
field grown samples from hundred recombinant inbred
lines and revealed major QTLs close to the locus
sucrose synthase (Sh1) gene known to be linked to
improved starch accumulation . To confirm the
importance of Sh1 locus, sucrose synthase gene
polymorphisms was analyzed in 45 genetically unrelated
maize lines. Therein, the Sh1 locus was also found to
significantly associate with higher starch and amylase
content as well as grain matter from multi-location
field trials . In the present study also a high level
of allelic diversity was detected in the genes
encoding sucrose synthase I, sucrose synthase II, starch
branching enzyme I and a-glucosidase, while the genes
encoding both the small and large subunits of
ADPglucose pyrophosphorylase were rather non-polymorphic
(Additional file 5).
Haplotype variation was also used to estimate the
extent of the genetic separation between cv. Brenda and
HS584. Among the 13 informative sequences, three
harboured non-synonymous exonic SNPs. Two
neighbouring SNPs within the granule bound starch synthase Ib
gene [CR-EST:HY09J12] were present in both HS584
and a number of the barley accessions, while the SNPs
present in both the -amylase [CR-EST:HF11O03] and
the g-2 hordein [CR-EST:HB20O07] genes were unique
to HS584. Another four genes (sucrose synthase type I
[CR-EST:HY09D18] and type II [CR-EST:HA31O14,
CR-EST:HF08A21], ADP-glucose pyrophosphorylase
small subunit sequence [CR-EST:HB16O10], and starch
branching enzyme I [CR-EST: HB30O07]) were found
to contain synonymous exonic substitutions. Intronic
SNPs were also detected in most of the genes, including
the ADP-glucose pyrophosphorylase small subunit
sequence [CR-EST:HB16O10], a gene known to
undergo alternative splicing . These data confirm
that wild barley alleles own the capability to alter
protein sequences (non-synonymous SNPs), codon usage
(synonymous SNPs) and the splicing process (intronic
SNPs) and emphasize the potential of the Brenda/
HS584 introgression line population to serve as a
model for the investigation of favourable wild barley
Intraspecific variation of grain starch content under
Identifying the molecular basis of phenotypic variation
can provide improved insights into the mechanisms
responsible for key agronomic traits such as grain yield
stability. Thus patterns of starch accumulation during
terminal drought were monitored for a diverse set of
50 barley accessions. A high genetic variation for grain
starch content was observed (Figure 6). The starch
content of the non-stressed barley landraces varied from
450-680 mg/g dry weight, while among the elite breeding
lines, the range was 514-648 mg/g (Additional file 5 and
Figure 6). Within gene bank accessions of H. vulgare and
H. spontaneum, two major classes were found; one class
suffered a reduction of up to 45% in the amount of starch
accumulated under terminal drought conditions, whereas
the other performed well in both well-watered and
terminal drought conditions (Figure 6). Unlike the wild
barleys and the landraces, the sample of elite breeding lines
showed little variation for starch accumulation, although
seed starch content
vERCH roeVHM vDHOM tvpeoeSH rvednaBH lvndHG
x t eP
many performed well under terminal drought stress.
Three accessions (LP101, LP107 and LP109) suffered a
slight reduction in grain starch content and,
consequently, thousand grain weight (TGW) when challenged
with terminal drought stress under both field and green
house conditions (Additional file 5). Interestingly, those
lines which showed dramatic reduction of starch content
under terminal drought in comparison to their respective
controls possess haplotypes H3 (Hv32), H4 (Hs3, Hs5,
Hv10) and H5 (OWB-DOM, Hv29, Hv30) from sucrose
synthase II gene (starch content of control versus stress
with low correlation of R2 = 0.4) and lines possessing
haplotype H6 (ISR42_8, Hv13, Hv20, Hv22, LP103,
LP104, LP106, LP107, LP110) from sucrose synthase II
gene correlate positively to optimum starch accumulation
under both control and drought treatments (with R2 =
0.9 at a significance level of a = 0.01 using Steigers
Z-test for Pearson correlation) [Additional file 6].
Similarly, we also noticed a higher genetic variation for TGW
of barley landraces not only under control conditions but
also under drought stress (Additional file 7). Moreover,
global correlation analysis between seed starch content
and an average of TGW obtained from multi-location
field trials from two consecutive years (2007 and 2008)
using both methods (water withhold and potassium
treatments) and green house screening for all genotypes
under drought stress conditions signifies correlation with
R2 = 0.72 at a significance level of a = 0.01 using Steigers
Z-test for Pearson correlation (Figure 7). The origin and
IG-number is provided for all 50 barley accessions in
Additional file 8.
The genetic mapping of 141 drought regulated ESTs has
extended the abiotic stress SNP map of barley  by a
further 134 novel markers. An extensive expression
analysis of these ESTs at various developmental stages for
drought response and across a range of barley
accessions resulted in creating an expression map for
genetically mapped markers. The mapped candidate genes
R2 = 0.7235
Seed starch content (mg/g)
Figure 7 Scatter plot and correlation analysis of seed starch
content and thousand grain weight (TGW) under terminal
drought stress. For further details refer Methods section.
have been reported to co-segregate with drought related
traits, which fall into diverse functional categories like
stress response (e.g. dehydrin [39,40]), transcription
factors (e.g. CBF ), carbohydrate metabolism (e.g.
sucrose synthase ) and many more [3,6,41,42]. The
map also disclosed an interesting correlation between
several clusters of sulphur-rich hordeins on the short
arm of chromosome 1 H and their co-expression,
potentially linked to methylation based regulation [30,31].
The haplotype structure of 17 starch biosynthesis/
degradation genes was explored, revealing that the genes
encoding sucrose synthase (both types I and II) and
starch synthase were surprisingly variable in wild barley
and landraces. Superior alleles related to haplotype H5
from sucrose synthase I and H6 from sucrose synthase
II were found to be present in the studied breeding lines
too, selected for improved performance. This
observation provides additional evidence that these alleles may
be causally associated with improved starch
accumulation under control as well as terminal drought stress
conditions. The gained knowledge represents a valuable
source for the development of functional markers to
assess larger collections of barley accessions for the
correlation of relevant haplotypes of starch biosynthesis/
degradation genes to seed starch content under drought
and, therefore, for further improvement of barley
cultivars in terms of improved grain weight.
Plant material, starch and DNA extraction
The eight barley accessions from which ESTs were
resequenced were the parents of mapping populations cvs.
Steptoe and Morex , the parents of the Oregon
Wolfe population  and the parents of AB
populations cv. Scarlett and ISR42_8 , and cv. Brenda and
the H. spontaneum accession 584 . The Steptoe/
Morex and Oregon Wolfe mapping populations
comprised 80 and 94 individuals, respectively. Total genomic
DNA was extracted from 4-6 g young leaf material,
using the protocol described in .
A set of 50 barley accessions was assembled from the
IPK Gatersleben and the ICARDA genebanks, and these,
along with cv. Brenda and H. spontaneum accession
584 (HS584), were grown till flowering under a 16 h
light/20C, 8 h dark/15C regime. Terminal drought
stress was imposed for a period of three weeks beginning
one week after fertilization (8 DAF) during the
postanthesis period. The automatic watering procedure was
monitored by a DL2e data logger (Delta T) with SM200
sensors connected to individual pots. This enabled to
maintain the control plants at 60% soil moisture and
drought stressed plants at 10% soil moisture. Mature
seeds were harvested from the mature plants of control
and drought stressed plants and estimated TGW using
MARVIN seed counter. For each line, three independent
replicates were maintained for both control and drought
stress treatments and for each replicate seeds were
pooled from five plants.
Starch was extracted from ground mature grains in
80% v/v ethanol at 60C, each followed by a
centrifugation (15 min, 14,000 g). The final supernatant was
discarded and the remaining pellet used for the
quantification of starch content . Starch content is
measured from three replicates.
Crude protein content (%) was obtained by
multiplying seed N% with the factor 5.83 . Total seed N%
was measured using elemental analyzer (Vario EL;
Elementar analyse system).
All 50 accessions were also subjected to drought stress
in the field at breeding station, Nordsaat Bhnshausen
over the two consecutive years (2007 and 2008) by
following two different strategies. Strategy I: All genotypes
were planted as two-row plots per entry with two
replications in randomized blocks, directly on soil in a closed
green house/rain shelter which completely protects rain
fall. Control plants were irrigated four times during the
period from seed set until seed filling. For imposing
terminal drought stress, ten days post anthesis staged
plants were subjected to water withhold by stopping
irrigation until the end of grain development. Strategy II: All
genotypes were planted as three-row plots per entry with
two replications in randomized blocks. Control plants
remained untreated. For mimicking drought stress
treatments 10% w/v potassium iodide is sprayed to whole
plant at ten days post anthesis. After reaching maturity,
all the genotypes of the two replicates from two strategies
were harvested by hand and TGW and seed quality was
determined in Nordsaat seed quality laboratory.
Correlation analysis was carried out between TGW data
and starch content under drought stress. To consider
seasonal variability, lines in each year were z-score
normalized, and then variance was calculated across the years.
The Z-score normalized TGW data for all accessions are
shown as heat maps (Additional file 7). Those lines with
too high variance levels were excluded from the
correlation analysis. For the remaining lines, average values of
TGW data were calculated across the years (2007 and
2008) from field and rain shelter as well as green house
experiments. These averaged TGW data were correlated
with the corresponding average values of green house
replicates of seed starch content from drought treatments
using the Pearson correlation measure. Statistical
significance of the calculated r2 values was assessed using
Steigers z-test at a significance level of a = 0.01 .
SNP discovery and detection
For the sequences identified in the CR-EST database
(clustering project g03) http://pgrc.ipk-gatersleben.de/cr-est/,
GeneRunner software http://www.generunner.net was
applied to design PCR primer pairs each amplifying a
300600 bp fragment from an individual EST. Each 50 l PCR
contained 50-100 ng genomic DNA template, 1.5 mM
MgCl2, 0.2 mM dNTP, 10 M of each primer and 1U Taq
DNA polymerase. After an initial denaturation of 96C/2
min, the reactions were cycled 14 times through 94C/30
s, 72C (-1C/cycle)/20 s and 72C/90 s, and then a further
27 times through 94C/30 s, 58C/20 s and 72C/90 s,
before a final extension step of 72C/3 min. After checking
for correct amplification, each reaction was then purified
using a MinElute96 UF PCR Purification kit (Qiagen,
Hilden, Germany) according to the manufacturers
instructions, and subjected to cycle sequencing from both ends
using the relevant PCR primers. Cycle sequencing was
performed with the BigDye Terminator v3.1 ready reaction
cycle sequencing kit on an ABI 3730 1 sequencer
(Applied Biosystems). The re-sequenced ESTs were
aligned using the SeqMan tool within the Lasergene
software package http://www.dnastar.com to identify SNPs.
According to the annotation of polymorphisms, haplotype
groups were determined for a core set of starch
biosynthesis/degradation genes (Table 2).
SNP detection was carried out by pyrosequencing.
Corresponding assays were designed with the
quencingAssay Design Software Version 1.0.6 (Biotage
AB, Uppsala, Sweden). Genomic DNA was amplified as
above, except that one primer was biotinylated, and the
extension step was shortened from 90 s to 30 s.
vidin SepharoseHigh Performance (GE Healthcare
Biosciences, Uppsala, Sweden) was used to obtain single
stranded amplicons. SNP genotyping was performed
using the PSQ HS 96A System (Biotage AB, Uppsala,
Sweden). Further information on template preparation
and the pyrosequencing protocol can be found in .
JoinMap v4  was used to construct a genetic map
based on a combination of the de novo Steptoe/Morex
and Oregon Wolfe population SNP and already
published genotypic data . Recombination fractions were
converted to cM using the Kosambi mapping function,
with the following JoinMap settings: minimum LOD
score = 1.0, recombination threshold = 0.4, ripple value
= 0 1 and jump threshold = 0 5. For chromosomes 3 H
and 6 H, the marker order of the reference map 
was chosen as the starting order.
Affymetrix BarleyI GeneChip analysis
To identify drought regulated gene sets at various stages of
development, Affymetrix chip CEL files derived from both
control and drought-treated seedlings (Series GSE3170),
21 day old plants (Series GSE6990), flag leaves (Series
GSE15970) , awn, lemma, palea, and the early stages
of the developing grain (Series GSE17669)  were
downloaded and merged with in house expression data
obtained from developing grain 20 DAF from cv. Brenda
and HS584. RNA was obtained from two independent
replicates. For each replicate seeds were pooled from
five plants. The developing grain harvested from the
central part of the spike from both control and drought
treated plants of cv. Brenda and HS584 according to
. The RNA was isolated using the TRIzol reagent
(Invitrogen GmbH, Karlsruhe, Germany) and RNAeasy
columns (Qiagen, Hilden, Germany). Probe synthesis,
labelling and hybridization were performed according to
the manufacturers protocols (Affymetrix). The
expression of 22,000 genes extracted from all experiments was
subjected to RMA normalization, applying a linear
model via the limma package using R/Bioconductor
functions in Robin software . After normalization,
log2 expression values were derived to generate fold
differences between non-stressed and drought stressed
organs from independent experiments. A nested
multiple testing strategy was applied, using the
BenjaminiHochberg P-value correction (P-value cut-off 0.05) to
recognize significant differences in expression levels. A
selection of 141 mapped genes was made from the 613
genes identified, and analysed for expression differences
between watered and water withhold plants at various
developmental stages. These log fold-change expression
data is first subjected to hierarchical clustering and
obtained clusters groups was refined by applying a
K-means clustering method according to . Heat
maps were generated using Genesis software . The
differentially expressed genes were functionally assigned
according to . Functionally overrepresented gene
categories have been calculated by Fishers exact test
with a P-value cut off 0.01 .
Additional file 1: Classification of ESTs according to MapMan
functional categories .
Additional file 2: SNP frequency and the number of mapped
markers per population (shown in parenthesis). SM: Steptoe-Morex
population, OWB: Oregon Wolfe mapping population, Sc_ISR42_8: H.
vulgare cultivar Scarlett-H. spontaneum ISR48 introgression line
population, BHS584: H. vulgare cultivar Brenda- H. spontaneum584
introgression line population.
Additional file 3: Genes whose expression was induced/repressed
by drought stress imposed at various stages of development. Their
map location, putative function and normalized expression ratios (control
vs drought stressed) are indicated. Statistical significance is indicated as
0: non-significant, -1 significantly up-regulated, +1 significantly
downregulated under drought.
Additional file 4: De novo SNP markers. int. ID: internal identification
number, off. ID: official identification number, Chr.: chromosome location,
unigene A35: unigene number, according to HarvEST:Barley assembly 35
http://harvest.ucr.edu, EST: identifier taken either from the CR-EST
database, or from the Affymetrix Barley Contigs, alt. EST: identifier of
alternative (orthologous) ESTs used for primer design, Locusprimer: the
sequence of the PCR primers used for re-sequencing,
Pyrosequencingprimer: the sequence of the PCR primers used for
pyrosequencing. Redundancy to other SNP maps indicates that the
same gene target is present in one or more of four independent genetic
maps; Genotype refers to the population tested; Dispensation order
indicates the dispensation order applied in pyrosequencing.
Additional file 5: Haplotype groups based on observed variation in
a set of 17 starch biosynthesis/degradation genes are provided.
Additionally shown are the seed starch content values for each line
under drought as well as control conditions. A detailed list of accessions,
their origin and IG-number is also supplied.
Additional file 6: Correlations between seed starch content of
control and drought stress from the accessions pertaining specific
haplotypes in sucrose synthase.
Additional file 7: Heatmap of Z-score normalized thousand grain
weight (TGW) data from drought stress experiments of field-grown
(F), rain shelter (RS) from the two consecutive years (2007 and
2008). Red colour indicates higher TGW, yellow -medium and blue
Additional file 8: Detailed list of accessions, their origin and
IGnumber is provided.
List of abbreviations
AB: advanced backcross; DAF: days after fertilization; DW: dry weight; EST:
expressed sequence tag; GBS: Gatersleben barley SNP; OWB: Oregon Wolfe
Barleys; PCA: principal component analysis; RMA: robust multichip average;
SNP: single nucleotide polymorphism; SM: Steptoe Morex mapping
population; TF: transcription factor; TGW: thousand grain weight
We thank Anette Heber, Jana Lorenz, Gabriele Einert and Katrin Blaschek
for their excellent technical assistance and Prof. K. Pillen for providing
samples of ISR42-8 and cv. Scarlett. We greatly appreciate the help of Dr.
Marc Strickert for language improvement and for the help in bioinformatic
analysis of field phenotypic data and correlation analysis. This research was
financially supported by a grant from the German Ministry of Education
and Research (BMBF) (Project GABI-GRAIN: FKZ; 0315041A).
SW carried out the molecular genetic analysis, sequence alignment and
linkage mapping, and participated in the design of experiments and the
drafting of the marker part of the manuscript. CP designed the PCR primers,
while RK and VTH performed the glasshouse drought tolerance assessments,
measured starch content and isolated RNA. AB, VK and LK provided genetic
material and conducted the field-based drought screening. MSR monitored
the marker study and co-edited the part of manuscript, along with UW who
conceived the study. NS coordinated the work of the GABI-GRAIN
consortium, contributed to the development of concepts, conducted gene
expression analysis and critically revised the manuscript. All the authors have
read and approved the final manuscript.
1. Pennisi E : Plant genetics . The blue revolution, drop by drop, gene by gene . Science 2008 , 320 : 171 - 173 .
2. Chen G , Krugman T , Fahima T , Chen K , Hu Y , Rder M , Nevo E , Korol A : Chromosomal regions controlling seedling drought resistance in Israeli wild barley , Hordeum spontaneum C. Koch. Genet Resour Crop Evol 2010 , 57 : 85 - 99 .
3. Diab AA , Teulat-Merah B , This D , Ozturk NZ , Benscher D , Sorrells ME : Identification of drought-inducible genes and differentially expressed sequence tags in barley . Theor Appl Genet 2004 , 109 : 1417 - 1425 .
4. Teulat B , Merah O , Sirault X , Borries C , Waugh R , This D : QTLs for grain carbon isotope discrimination in field-grown barley . Theor Appl Genet 2002 , 106 : 118 - 126 .
5. Tondelli A , Francia E , Barabaschi D , Aprile A , Skinner JS , Stockinger EJ , Stanca AM , Pecchioni N : Mapping regulatory genes as candidates for cold and drought stress tolerance in barley . Theor Appl Genet 2006 , 112 : 445 - 454 .
6. von Korff M , Grando S , Del Greco A , This D , Baum M , Ceccarelli S : Quantitative trait loci associated with adaptation to Mediterranean dryland conditions in barley . Theor Appl Genet 2008 , 117 : 653 - 669 .
7. Rostoks N , Mudie S , Cardle L , Russell J , Ramsay L , Booth A , Svensson JT , Wanamaker SI , Walia H , Rodriguez EM , et al: Genome-wide SNP discovery and linkage analysis in barley based on genes responsive to abiotic stress . Mol Genet Genomics 2005 , 274 : 515 - 527 .
8. Close TJ , Bhat PR , Lonardi S , Wu Y , Rostoks N , Ramsay L , Druka A , Stein N , Svensson JT , Wanamaker S , et al: Development and implementation of high-throughput SNP genotyping in barley . BMC Genomics 2009 , 10 : 582 .
9. Sato K , Nankaku N , Takeda K : A high-density transcript linkage map of barley derived from a single population . Heredity 2009 , 103 : 110 - 117 .
10. Stein N , Prasad M , Scholz U , Thiel T , Zhang H , Wolf M , Kota R , Varshney RK , Perovic D , Grosse I , Graner A : A 1, 000 - loci transcript map of the barley genome: new anchoring points for integrative grass genomics . Theor Appl Genet 2007 , 114 : 823 - 839 .
11. Feuillet C , Langridge P , Waugh R : Cereal breeding takes a walk on the wild side . Trends Genet 2008 , 24 : 24 - 32 .
12. Schmalenbach I , Pillen K : Detection and verification of malting quality QTLs using wild barley introgression lines . Theor Appl Genet 2009 , 118 : 1411 - 1427 .
13. von Korff M , Wang H , Leon J , Pillen K : AB-QTL analysis in spring barley: II. Detection of favourable exotic alleles for agronomic traits introgressed from wild barley (H. vulgare ssp . spontaneum). Theor Appl Genet 2006 , 112 : 1221 - 1231 .
14. Backes G , Madsen LH , Jaiser H , Stougaard J , Herz M , Mohler V , Jahoor A : Localisation of genes for resistance against Blumeria graminis f .sp. hordei and Puccinia graminis in a cross between a barley cultivar and a wild barley (Hordeum vulgare ssp . spontaneum) line. Theor Appl Genet 2003 , 106 : 353 - 362 .
15. von Korff M , Wang H , Leon J , Pillen K : AB-QTL analysis in spring barley . I. Detection of resistance genes against powdery mildew, leaf rust and scald introgressed from wild barley . Theor Appl Genet 2005 , 111 : 583 - 590 .
16. Shavrukov Y , Gupta NK , Miyazaki J , Baho MN , Chalmers KJ , Tester M , Langridge P , Collins NC : HvNax3-a locus controlling shoot sodium exclusion derived from wild barley (Hordeum vulgare ssp . spontaneum). Funct Integr Genomics 2010 , 10 : 277 - 291 .
17. Guo P , Baum M , Grando S , Ceccarelli S , Bai G , Li R , von Korff M , Varshney RK , Graner A , Valkoun J : Differentially expressed genes between drought-tolerant and drought-sensitive barley genotypes in response to drought stress during the reproductive stage . J Exp Bot 2009 , 60 : 3531 - 3544 .
18. Hubner S , Hoffken M , Oren E , Haseneyer G , Stein N , Graner A , Schmid K , Fridman E : Strong correlation of wild barley (Hordeum spontaneum) population structure with temperature and precipitation variation . Mol Ecol 2009 , 18 : 1523 - 1536 .
19. James VA , Neibaur I , Altpeter F : Stress inducible expression of the DREB1A transcription factor from xeric, Hordeum spontaneum L. in turf and forage grass (Paspalum notatum Flugge) enhances abiotic stress tolerance . Transgenic Res 2008 , 17 : 93 - 104 .
20. Pietsch C , Sreenivasulu N , Wobus U , Rder MS : Linkage mapping of putative regulator genes of barley grain development characterized by expression profiling . BMC Plant Biol 2009 , 9 : 4 .
21. Li JZ , Huang XQ , Heinrichs F , Ganal MW , Rder MS : Analysis of QTLs for yield components, agronomic traits, and disease resistance in an advanced backcross population of spring barley . Genome 2006 , 49 : 454 - 466 .
22. von Korff M , Wang H , Leon J , Pillen K : Development of candidate introgression lines using an exotic barley accession (Hordeum vulgare ssp. spontaneum) as donor . Theor Appl Genet 2004 , 109 : 1736 - 1745 .
23. Marcel TC , Varshney RK , Barbieri M , Jafary H , de Kock MJ , Graner A , Niks RE : A high-density consensus map of barley to compare the distribution of QTLs for partial resistance to Puccinia hordei and of defence gene homologues . Theor Appl Genet 2007 , 114 : 487 - 500 .
24. Varshney RK , Marcel TC , Ramsay L , Russell J , Rder MS , Stein N , Waugh R , Langridge P , Niks RE , Graner A : A high density barley microsatellite consensus map with 775 SSR loci . Theor Appl Genet 2007 , 114 : 1091 - 1103 .
25. Vlad F , Rubio S , Rodrigues A , Sirichandra C , Belin C , Robert N , Leung J , Rodriguez PL , Lauriere C , Merlot S : Protein phosphatases 2C regulate the activation of the Snf1-related kinase OST1 by abscisic acid in Arabidopsis . Plant Cell 2009 , 21 : 3170 - 3184 .
26. Fujii H , Chinnusamy V , Rodrigues A , Rubio S , Antoni R , Park SY , Cutler SR , Sheen J , Rodriguez PL , Zhu JK : In vitro reconstitution of an abscisic acid signalling pathway . Nature 2009 , 462 : 660 - 664 .
27. Sreenivasulu N , Sopory SK , Kavi Kishor PB : Deciphering the regulatory mechanisms of abiotic stress tolerance in plants by genomic approaches . Gene 2007 , 388 : 1 - 13 .
28. Miller G , Mittler R : Could heat shock transcription factors function as hydrogen peroxide sensors in plants? Ann Bot 2006 , 98 : 279 - 288 .
29. Pelger S , Sll T , Bengtsson BO : Evolution of hordein gene organization in three Hordeurn species . Hereditas 1993 , 119 : 219 - 231 .
30. Radchuk VV , Sreenivasulu N , Radchuk RI , Wobus U , Weschke W : The methylation cycle and its possible functions in barley endosperm development . Plant Mol Biol 2005 , 59 : 289 - 307 .
31. Sorensen MB : Methylation of B-hordein genes in barley endosperm is inversely correlated with gene activity and affected by the regulatory gene Lys3 . Proc Natl Acad Sci USA 1992 , 89 : 4119 - 4123 .
32. Haseneyer G , Stracke S , Piepho HP , Sauer S , Geiger HH , Graner A : DNA polymorphisms and haplotype patterns of transcription factors involved in barley endosperm development are associated with key agronomic traits . BMC Plant Biol 2010 , 10 : 5 .
33. Sreenivasulu N , Borisjuk L , Junker B , Mock HP , Rolletschek H , Seiffert U , Weschke W , Wobus U : Barley grain development: towards an integrative view . International Review of Cell and Molecular Biology 2010 , 281 : 49 - 89 .
34. Sreenivasulu N , Radchuk V , Strickert M , Miersch O , Weschke W , Wobus U : Gene expression patterns reveal tissue-specific signaling networks controlling programmed cell death and ABA- regulated maturation in developing barley seeds . Plant J 2006 , 47 : 310 - 327 .
35. Sreenivasulu N , Usadel B , Winter A , Radchuk V , Scholz U , Stein N , Weschke W , Strickert M , Close TJ , Stitt M , et al: Barley grain maturation and germination: metabolic pathway and regulatory network commonalities and differences highlighted by new MapMan/PageMan profiling tools . Plant Physiol 2008 , 146 : 1738 - 1758 .
36. Sulpice R , Pyl ET , Ishihara H , Trenkamp S , Steinfath M , Witucka-Wall H , Gibon Y , Usadel B , Poree F , Piques MC , et al: Starch as a major integrator in the regulation of plant growth . Proc Natl Acad Sci USA 2009 , 106 : 10348 - 10353 .
37. Thvenot C , Simond-Cote E , Reyss A , Manicacci D , Trouverie J , Le Guilloux M , Ginhoux V , Sidicina F , Prioul JL : QTLs for enzyme activities and soluble carbohydrates involved in starch accumulation during grain filling in maize . J Exp Bot 2005 , 56 : 945 - 958 .
38. Radchuk VV , Borisjuk L , Sreenivasulu N , Merx K , Mock HP , Rolletschek H , Wobus U , Weschke W : Spatiotemporal profiling of starch biosynthesis and degradation in the developing barley grain . Plant Physiol 2009 , 150 : 190 - 204 .
39. Pan A , Hayes PM , Chen F , Chen THH , Blake T , Wright S , Karsai I , Bed Z : Genetic analysis of the components of winterhardiness in barley ( Hordeum vulgare L.). Theor Appl Genet 1994 , 89 : 900 - 910 .
40. Teulat B , Zoumarou-Wallis N , Rotter B , Ben Salem M , Bahri H , This D : QTL for relative water content in field-grown barley and their stability across Mediterranean environments . Theor Appl Genet 2003 , 108 : 181 - 188 .
41. Chen G , Komatsuda T , Pourkheirandish M , Sameri M , Sato K , Krugman T , Fahima T , Korol AB , Nevo E : Mapping of the eibi1 gene responsible for the drought hypersensitive cuticle in wild barley (Hordeum spontaneum) . Breeding Science 2009 , 59 : 21 - 26 .
42. Chen G , Pourkheirandish M , Sameri M , Wang N , Nair S , Shi Y , Li C , Nevo E , Komatsuda T : Genetic targeting of candidate genes for drought sensitive gene eibi1 of wild barley (Hordeum spontaneum) . Breeding Science 2009 , 59 : 637 - 644 .
43. Kleinhofs A , Kilian A , Saghai Maroof MA , Biyashev RM , Hayes P , Chen FQ , Lapitan N , Fenwick A , Blake TK , Kanazin V , et al: A molecular, isozyme and morphological map of the barley (Hordeum vulgare) genome . Theor Appl Genet 1993 , 86 : 705 - 712 .
44. Costa JM , Corey A , Hayes PM , Jobet C , Kleinhofs A , Kopisch-Obusch A , Kramer SF , Kudrna D , Li M , Riera-Lizarazu O , et al: Molecular mapping of the Oregon Wolfe Barleys: a phenotypically polymorphic doubledhaploid population . Theor Appl Genet 2001 , 103 : 415 - 424 .
45. Plaschke J , Ganal MW , Rder MS : Detection of genetic diversity in closely related bread wheat using microsatellite markers . Theor Appl Genet 1995 , 91 : 1001 - 1007 .
46. Rolletschek H , Koch K , Wobus U , Borisjuk L : Positional cues for the starch/ lipid balance in maize kernels and resource partitioning to the embryo . Plant J 2005 , 42 : 69 - 83 .
47. Merrill A , Watt BK : Energy value of foods, basis and derivation . United States Department of Agriculture Handbook 74 USDA, Washington, DC ; 1973 .
48. Steiger JH : Tests for comparing elements of a correlation matrix . Psychological Bulletin 1980 , 87 : 245 - 251 .
49. Huang XQ , Rder MS : Development of SNP assays for genotyping the puroindoline b gene for grain hardness in wheat using pyrosequencing . J Agric Food Chem 2005 , 53 : 2070 - 2075 .
50. Van Ooijen JW : JoinMap 4, Software for the calculation of genetic linkage maps in experimental populations Wageningen , Netherlands: Kyazma B. V; 2006 .
51. Abebe T , Melmaiee K , Berg V , Wise RP : Drought response in the spikes of barley: gene expression in the lemma , palea, awn, and seed. Funct Integr Genomics 2010 , 10 : 191 - 205 .
52. Sreenivasulu N , Altschmied L , Radchuk V , Gubatz S , Wobus U , Weschke W : Transcript profiles and deduced changes of metabolic pathways in maternal and filial tissues of developing barley grains . Plant J 2004 , 37 : 539 - 553 .
53. Lohse M , Nunes-Nesi A , Krger P , Nagel A , Hannemann J , Giorgi FM , Childs L , Osorio S , Walther D , Selbig J , et al: Robin: An intuitive wizard application for R-based expression microarray quality assessment and analysis . Plant Physiol 2010 , 153 : 642 - 651 .
54. Sturn A , Quackenbush J , Trajanoski Z : Genesis: cluster analysis of microarray data . Bioinformatics 2002 , 18 : 207 - 208 .