-
Research Collection
Doctoral Thesis
Impact of prolamin variation and 1BL.1RS translocation
onbread-making quality parameters of wheat (Triticum
aestivumL.)
Author(s): Gobaa, Samy
Publication Date: 2007
Permanent Link: https://doi.org/10.3929/ethz-a-005427073
Rights / License: In Copyright - Non-Commercial Use
Permitted
This page was generated automatically upon download from the ETH
Zurich Research Collection. For moreinformation please consult the
Terms of use.
ETH Library
https://doi.org/10.3929/ethz-a-005427073http://rightsstatements.org/page/InC-NC/1.0/https://www.research-collection.ethz.chhttps://www.research-collection.ethz.ch/terms-of-use
-
DISS. ETHNO. 17101
Impact of prolamin variation and 1BL.1RS translocation on
bread-making
quality parameters of wheat {Triticum aestivum L.)
A dissertation submitted to
ETH ZURICH
for the degree of
Doctor of Sciences
presented by
SAMY GOBAA
DEA, Université Paul Sabatier (Toulouse III)
born 08.09.1977
citizen of
France
accepted on the recommendation of
Prof. Dr. Peter Stamp, examiner
Dr. GeertKleijer, co-examiner
Dr. Gérard Branlard, co-examiner
Zurich, 2007
-
Table of contents
Table of contents I
Summary Ill
Résumé V
List of abbreviations VIII
1 General introduction 1
Why wheat? 1
The wheat kernel 1
The prolamins 2
The 1BL. 1 RS translocation 3
The aim of the study 5
2 Ax2", a new high molecular weight glutenin subunit coded by
Glu-Al: its predictedstructure and its impact on bread-making
quality 7
2.1 Introduction 7
2.2 Material and methods 8
2.3 Results 9
2.4 Discussion 12
3 Effect of the IBL.IRS translocation and of the Glu-B3
variation on bread-making qualitytests in a doubled haploid
population 15
3.1 Introduction 15
3.2 Materials and methods 16
3.2.1 Plant material and experimental design 16
3.2.2 Quality tests 17
3.2.3 SDS-PAGE and allelic variation 17
3.3 Results and discussion 18
3.3.1 Allelic variation in the DH population 18
3.3.2 Preliminary tests 20
3.3.3 Protein- and kernel-related traits 20
3.3.4 Starch-related tests 22
3.3.5 Rheology tests 22
3.3.6 Baking tests 24
3.4 Conclusions 24
4 Proteomic analysis of wheat recombinant inbred lines (Part I):
effect of the IBL.IRS
translocation on the wheat grain proteome 26
4.1 Introduction 26
4.2 Material and methods 27
4.2.1 Plant material 27
4.2.2 Translocation mapping 27
4.2.3 1-D electrophoresis and allelic variation 27
4.2.4 2D SDS-PAGE and quantitative variation 28
4.2.5 Protein identification 29
4.3 Results 30
4.4 Discussion 39
5 Proteomic analysis of wheat recombinant inbred lines (Part
II): variations in prolaminand dough rheology 43
I
-
5.1 Introduction 43
5.2 Material and methods 44
5.2.1 Plant material 44
5.2.2 Rheological tests and classification 44
5.2.3 Two-dimensional electrophoresis 45
5.2.4 Assignment of the prolamin fractions 45
5.2.5 Statistical analysis 46
5.2.6 Protein identification 47
5.3 Results 48
5.4 Discussion 55
6 Conclusions 60
7 References 64
Remerciements 72
Curriculum vitae 74
II
-
Summary
The suitability of wheat varieties for bread-making depends
largely on prolamins. The amino acid
composition of these gluten building blocks as well as their
quantitative regulation have a strong
influence on the rheology of the dough and, thus, on the
suitability of the variety for bread-
making. The wheat/rye 1BL.1RS translocation was reported to
drastically reduce rheological
properties. However, there are noticeable exceptions. This
chromosomal rearrangement was
designed to transfer resistance to pathogens such as Puccinia
recondita, Puccinia gramins,
Puccinia striiformis and Erysiphe graminis from rye to wheat.
Today this resistance is overcome
but 1BL.1RS translocation remains interesting because of the
improved yield it produces.
To better understand the impact of particular prolamins (Glu-Al
2 , Glu-BS j and Glu-DS c) on
quality and to gain better understanding on prolamin regulation
in general, a population of
doubled haploid was created and sown in a two-year field
experiment. The first part of this study
describes a new x-type high molecular weight glutenin subunit,
encoded at the locus Glu-Al and
named 2.The statistical analysis demonstrated that the subunit 2
was as favorable for quality as
the subunit 2*. This is in accordance with the results showing
that the 2 open reading frame had
the same number of cysteines as 2*. The small differences in the
length of the central domain had
no detectable effect on the elasticity, tenacity and baking
quality of the dough.
The effects of the rest of the polymorphic loci {Glu-BS and
Glu-D3) also produced innovative
results; allele Glu-BS], which marks the 1BL.1RS translocation,
was found to be associated with
lower rheology of the dough. The presence of this translocation
resulted in less tenacity, less
extensibility, less strength and greater softening. Allele
Glu-DS b was also detrimental to
rheology when compared to allele c. Its impact on the tenacity
of the dough was important and
globally it affected quality to a similar extent as the 1BL.1RS
translocation. The protein content
of the 1BL.1RS lines was significantly higher than that of the
regular IB lines. This led to loaves
with a greater volume. Furthermore, the softening produced by
the 1BL.1RS translocation and
Glu-DS b probably played a role in the greater bread volume by
reducing the resistance to the
deformation of the dough.
The comparison of the 2-DE proteomic profiles of 16 doubled
haploid lines, with or without the
1BL.1RS translocation, allowed an assessment of the impact of
this event on the prolamins;
quantitative and qualitative proteic variations, induced by the
1BL.1RS translocation, were
III
-
reported. Eight spots were found only in lines with the 1BL.1RS
translocation, 16 other spots
disappeared in the same lines. Twelve spots, present in both
types of genotypes, met the criteria
for up- or down-regulation. In translocated genotypes, a
y-gliadin-like LMW-GS, with nine
cysteine residues, was over-expressed by 801 %, suggesting that
the lack of LMW-GS was
counterbalanced by an over-expression of relatively similar
prolamins. Identification and
quantification of the prolamin fractions on the two-dimensional
electrophoresis gels
demonstrated that the HMW-GS were up-regulated by 25 % in
1BL.1RS DH lines, even though
the corresponding genes were not located on the missing IBS
chromosome. The y-gliadins were
also up-regulated by 36 %. Moreover, a significantly varying
spot, identified as a dimeric alpha-
amylase inhibitor, may be considered as a valuable candidate to
explain the starch gelatinization
defect observed in the 1BL.1RS lines of the DH population. To
investigate the impact of the
1BL.1RS translocation on dough strength and to understand how
1BL.1RS genotypes overcome
the loss of Glu-BS and Gli-Bl, the same 16 DH lines were
reclassified into high- or low-tenacity
and into high- or low-extensibility categories. The results
showed that 32 spots, mainly prolamins,
were differentially expressed and that five others were specific
to high-strength DH lines. The
polymeric prolamin fractions were also accumulating in
high-tenacity lines and decreasing in
high-extensibility lines, confirming the role of the interchain
disulfide bonds in the resistance to
deformation. In contrast, the monomeric fraction of a-gliadin,
in particular the Gli-Al allele of
the parent Toronit, favored extensibility and decreased tenacity
by higher accumulation (+12 %
of a-gliadins in high extensibility lines) when compared to the
Gli-Al allele of parent 211.12014.
The results will help the establishment of efficient breeding
strategies where translocated lines
are concerned. The accumulation of y-type HMW-GS and
good-quality y-gliadin can improve
quality. These results also lead to a reconsideration of the
mechanisms of prolamin evolution and
regulation.
IV
-
Résumé
Les prolamines déterminent en grande partie l'aptitude à la
panification du blé. La séquence
d'acide aminé de ces dernières ainsi que leurs variations
quantitatives déterminent, en grande
partie, les propriétés rhéologiques de la pâte. La translocation
blé/seigle 1BL.1RS a été démontré
comme ayant un effet négatif sur ces paramètres. Cependant des
exceptions notables existent et
certaines variétés de blé réussissent à atteindre un niveau
satisfaisant de qualité malgré la
présence de cette translocation. Ce réarrangement chromosomique
a été créé pour faire bénéficier
le blé de la résistance du seigle à des pathogènes tels que
Puccinia recondita, Puccinia gramins,
Puccinia striiformis et Erysiphe graminis. Aujourd'hui ces
résistances sont, pour la plupart,
surmontées mais la translocation 1BL.1RS reste employée pour les
meilleurs rendements qu'elle
induit. Une population de lignée haploïdes doublés a été créée
et évaluée sur deux années
d'expérimentation pour mieux saisir l'impact de certaines
prolamines particulières (Glu-Al 2..,
Glu-BS j et Glu-DS c) et pour apporter de nouveaux éléments de
compréhension sur le réseau de
régulation de ces protéines.
La première partie de cette étude rapporte l'existence d'une
nouvelle gluténine de haut poids
moléculaire codée par Glu-Al et nommée 2 . L'analyse a démontré
que cette sous unité gluténine
est aussi performante que la sous unité 2* en terme de rhéologie
et de qualité boulangère. Ceci
s'explique par une composition en acide aminé très semblable
entre les deux protéines. Seules
quelques petites différences existent au niveau de la longueur
du domaine répété de ces deux
gluténines.
L'analyse des effets du reste des loci polymorphes {Glu-BS et
Glu-D3) a aussi produit des
résultats originaux. L'allèle Glu-BS], qui signe la
translocation, s'est montré défavorable sur tous
les tests rhéologiques réalisés. Les lignées portant cet allele
ont eu de plus faibles ténacités, de
plus faibles extensibilités, de plus faibles forces et de plus
forts ramollissements. L'allèle Glu-DS
b c'est lui aussi montré défavorable par rapport à l'allèle
Glu-DS c. Cependant, il est à noter que
la variation à Glu-DS n'a eu d'effet que sur la résistance
opposée à la déformation de la pâte
contrairement à la variation à Glu-BS. Globalement les
variations sur Glu-BS et Glu-DS ont
produit des changements de rhéologie du même ordre de grandeur.
Le taux de protéine contenu
dans les grains des lignées 1BL.1RS a été significativement plus
important que celui des lignées
IB. Ceci a sûrement participé à l'obtention de volume de pain
plus important chez les lignées
V
-
transloquées. De plus, les paramètres rhéologiques de toute la
population ayant été élevés de
manière générale, la perte de force induite par les alleles
Glu-BS j et Glu-DS b pourrait expliquer
en partie les meilleurs volumes de pain associé à ces derniers
ceci en s'opposant moins
fermement à la déformation de la pâte que leurs pendants
alléliques.
La comparaison des gels d'electrophoreses bidimensionnelles de
seize lignées représentatives de
la population, ayant ou non la translocation 1BL.1RS, a permis
d'observer les effets associés à ce
réarrangement chromosomique. Les variations qualitatives et
quantitatives liées à cette
translocation ont pu être observées. Huit spots ont été
retrouvés uniquement chez les lignées
1B1.1RS, seize autres n'y ont jamais été observés. Douze spots
ont montré des variations
quantitatives significatives. Il a aussi été possible de
démontrer qu'un spot identifié comme une
« y-gliadin LMW-like » avait subi une augmentation de volume de
801 % chez les lignées
1BL.1RS. Ceci suggère que, chez ces dernières, le manque de
gluténines de faible poids
moléculaire a été contrebalancé par la surexpression de
protéines relativement semblables
capables de remplir une mission similaire. La quantification des
différentes fractions de
prolamine a permis d'établir que les gluténines de haut poids
moléculaire et les y-gliadines
avaient augmenté respectivement de 25 et 26 % en réponse à la
translocation 1BL.1RS. De plus il
a aussi été démontré que cette translocation induisait la perte
d'un spot identifié comme un
inhibiteur dimeric d'alpha amylase. Cette protéine pourrait donc
être considérée comme un
candidat intéressant expliquant le défaut de gélatinisation de
l'amidon, lié à la translocation,
observé lors de l'évaluation au champ de la population. Pour
établir l'impact de la translocation
sur la force de la pâte et pour comprendre comment certains
génotypes arrivent à surmonter la
perte de Glu-BS et Gli-Bl, les mêmes seize lignées ont été «
reclassifiées » en lignées de forte ou
faible ténacité et en lignées de forte ou faible extensibilité.
Ceci a démontré que 32 spots,
majoritairement des prolamines, ont montré une variation
quantitative significative. Cinq spots
n'ont été retrouvés que chez les lignées à fort W. il a aussi
été démontré que les fractions de
prolamine polymérique augmentaient chez les lignées à forte
ténacité et diminuaient chez les
lignées à forte extensibilité. Confirmant ainsi que les ponts
disulfures jouent un rôle majeur dans
la résistance à la déformation de la pâte. La fraction
monomérique des a-gliadines, plus
précisément 1'allele Gli-Al du parent Toronit, a été trouvé
comme favorisant l'extensibilité via
une accumulation plus importante que celle produite par l'allèle
Gli-Al du parent 211.12014.
VI
-
Les résultats obtenu aideront les sélectionneurs à encore
améliorer la qualité des blés transloqués
en favorisant l'accumulation des gluténines de haut poids
moléculaire de type y et celle des y-
gliadines de bonne qualité. Ces mêmes résultats aideront aussi à
réévaluer la variabilité et la
régulation des prolamines.
VII
-
List of abbreviations
%Vol percentage of volume
1 -D mono dimensional
2-DE two dimensional gel electrophoresis
6-11NL non linear/?/ range of 6 to 11
aa amino acid.
A-PAGE acid Polyacrylamid gel electrophoresis
bp base pair
C crosslinking ratio
CHAPS 3 - [(3 -Cholamidopropyl)dimethylammonio] -1
-propanesulfonate
Chrs chromosome
CO2 carbon dioxide
CV coefficient of variation
DH doubled haploid.
DNA deoxyribonucleic acid
DTT dithiothreitol
Xgli sum of the volumes of the gliadin spots.
^Protamin sum of the volumes of the prolamin spots.
EST expressed sequence tag
Ext mixograph's extensibility
Fal falling number
GH grain hardness
H20 abs water absorption as measured with the Farinograph
HCL hierarchical clustering
HCl hydrochloric acid
HMW-GS high molecular weight glutenin subunits
HPLC high-performance liquid chromatography
ICC international association for cereal science and
technology
le swelling index as measured by the Alveograph
IEF isoelectric focusing
IPG isoelectric point gradient
-
L Extensibility as measured by the Alveograph
LEA late embryogenesis abundant proteins
LMW-GS low molecular weight glutenin subumts
MALDI-TOF matrix-assisted laser desorption/ionization-time of
flight
MS mass spectrometry
MS/MS tandem mass spectrometry
MW molecular weight
NanoLC nano-hquid chromatography
NCBI National Center for Bioinformatics Information
MRS near infrared reflectance spectroscopy
ns non significant
ORF open reading frame
P tenacity as measured by the Alveograph
PCR polymerase chain reaction
pH pondus hydrogenn I pouvoir hydrogène
pi isoelectric point
PR protein content
QTL quantitative trait loci
RES resistance as measured by the Fannograph
Rmax strength as measured by the Mixograph
RMT rapid mix test
SD standard deviation
SDS sodium dodecyl sulfate
SDS-PAGE sodium dodecyl sulfate ployacrylamid gel
electrophoresis
SOF softening as measured by the Fannograph
T Polyacrylamid content (w/v)
TKW thousand kernels weight
Tns tnshydroxymethylaminomethane
v/v volume to volume
W strength as measured by the Alveograph
w/v weight to volume
IX
-
Zel Zeleny test
X
-
1 General introduction
Why wheat?
Wheat (Triticum aestivum L.) has become a major source of energy
and protein for human beings
since agriculture developed in Mesopotamia. Such a success
relied on multiple advantages. The
most decisive advantage was probably the non-dehiscence of the
wheat caryopsis which enabled
the "human-assisted" multiplication and made selection easer. On
the level of evolution, the loss
of wheat's natural ability for seed dispersal was
counterbalanced by a total reliance on human
agriculture to ensure its survival. Today, modern wheat is sown
on one of the largest surface
attributed to a crop and the worldwide wheat harvest ranks
second in total production behind
maize, the third being rice (FAOSTAT 2004). Unique end-uses also
played a crucial role in the
success of wheat. The leavening of the dough occurs after mixing
flour, water and yeast enabling
the production of large-volume bread: one of the largest markets
for wheat production today.
The wheat kernel
The endosperm of wheat consists mainly of starch. More than 70 %
of the grain mass is
composed of this complex sugar. After degradation, starch is the
unique source of carbohydrates
available to the embryo during germination. The protein content
of the wheat kernel varies from
9 to 16 %. The wheat albumen contains three main groups of
proteins: prolamins, globulins and
albumins. The latter two groups were found to play only a minor
role in bread-making quality.
However, enzymes belonging to these two groups of protein, such
as alpha- and beta-amylase,
may favor sprouting and, thus, lower the bread-making quality.
Friabilin is a protein associated
with the starch granule in the endosperm of wheat. It determines
the class of hardness of the
different wheat cultivars. The presence of friabilin is
associated with a soft phenotype. Darlington
et al. (2000) demonstrated that this protein is controlled by
the Ha locus on chromosome 5D.
Pentosanes have surface-active properties. These sugars were
reported to influence the size of the
alveoli in the crumb. The wheat grain also contains 1.5 to 2 %
of lipids but their role in bread-
making is limited.
1
-
The prolaminsProtamins exist in a large number of wild or
cultivated grass species. Their initial role was to
provide the embryo with a source of amino acids during
germination. Their structure was shaped
by their biological goal and, thus, they acquired the capacity
to fold and polymerize in order to
stock up during grain filling. Human selection applied to
primitive prolamins took advantage of
their natural propensity to polymerize and to withstand
mechanical deformation, thus meeting the
requirement for human nutrition. Selection for high
polymerization of prolamins and high-
strength gluten maximized the retention of CO2 during
fermentation of the dough and resulted in
large volume bread. Variation in quality and in quantity of
prolamin has a direct impact on bread-
making and on the rheology of the dough.
Prolamins account for 45 to 80 % of the total proteins in the
grain of modern wheat cultivars. The
most widely accepted classification puts prolamins into three
structural/functional classes: high
molecular weight gluten subunits (HMW-GS), low molecular weight
glutenin subunits (LWM-
GS) and gliadins. The two types of glutenins are the only
prolamins that are able to form
interchain disulfide bridges, which make up the gluten backbone.
Gliadins produce only weak
hydrogen bonds with the other prolamins.
Anderson and Greene (1989) described the structure of HMW-GS.
Three domains are present in
these proteins: the N- and C-terminal domains plus a central
repetitive domain. The cysteines
tend to be located in the external domains and are responsible
first for the globular conformation
of these domains and second for the cross-linking ability of
HMW-GS. The number of cysteines
and the structure of the central domain differentiate between
y-type and x-type HMW-GS.
HMW-GS are encoded by the three homeologous loci: the Glu-1
group. Each locus generally
controls the production of one x-type and one y-type subunit.
However, because of the null
alleles, the average number HMW-GS subunits in the wheat
cultivars is three to five. The
observed genetic variability of HMW-GS in wheat is limited to 20
subunits, i.e. 21 according to
this study. Regulation of HMW-GS accumulation is unclear.
Guillaumie et al. (2004) reported
that a QTL controlling the amount of HMW-GS was co-localizing
with the locus Glu-Bl
encoding HMW-GS. In an association study, Ravel et al. (2006)
further demonstrated that the
amount of HMW-GS was allele-dependent. Allelic variability of
HMW-GS and bread-making
quality was investigated in multiple studies and a quality index
was established for the most
common alleles (Branlard et al., 1992). Quality can be explained
in part by the variation in
2
-
HMW-GS and was found to be highly dependent on the structure of
the HMW-GS alleles.
Anderson et al. (1989) reported the presence of an extra
cysteine residue in the central domain of
subunit 1Dx5. This explained the positive impact of this subunit
on quality and proved that cross-
linking is critical for dough rheology (Shewry et al.,
1992).
LMW-GS are also important for bread-making. The presence of
cysteine (for interchain cross-
linking), in even or uneven numbers, determines whether LMW-GS
extend the gluten network or
acts as chain terminators. Multiple studies reported the impact
of allelic variation ofLMW-GS on
a large number of quality parameters. Metakovsky et al, (1990)
established a list of favorable
alleles for Rmax and W. Cornish et al. (1993) found that the
combination of the allele b b b on
the three Glu-3 loci was favorable for extensibility. Gupta et
al. (1994) concluded that the LMW-
GS alleles, encoded at the IB chromosome, are more important for
Rmax than those encoded at
the 1A and ID chromosomes. Luo et al. (2001) established a list
of favorable alleles for
Farinograph parameters, protein content, SDS sedimentation and
Pelshenke tests. These findings
where often contradictory but they clearly established that
LMW-GS allelic variation affects
quality and that the effects of the different alleles were
mainly additive.
The structure of the gliadins is similar to that of the LMW-GS,
but unlike LMW-GS, gliadins are
soluble in ethanol. This property is due to the absence of
covalent cross-linking between the
gluten and the gliadins. The recent report that some gliadins
are also able to produce interchain
disulfide bonds reduces the gap between these two classes of
prolamins (Ferrante et al., 2006).
Gliadins are acknowledged to play the role of a solvent; their
accumulation usually favors
extensibility (Branlard et al, 2001). Because the loci: Glu-3
(LMW-GS) and Gli-1 (gliadin) are
closely linked it is difficult to assess the effect of each kind
of protein.
The 1BL.1RS translocation
The insertion of alien chromatin into the genetic background of
wheat was designed to take
advantage of the rusticity of other species such as rye, the
best example being the creation of
Triticale (x Triticosecale wittmack), which combines the A and B
genomes of wheat with the R
genome of rye. Nevertheless, smaller-scale genetic
rearrangements were created to limit the
disadvantages of the insertion of a full R genome. Zeller and
Hsam (1983) traced the first
1BL.1RS wheat/rye chromosomal translocations to breeding
programs using 1R (IB) substitution
lines in Germany. The most famous translocated wheat was
probably the cultivar Kavkaz which
was used as the main source of the 1BL.1RS chromosome in the
modern breeding programs.
3
-
Translocated cultivars were appreciated mainly for their
resistance to powdery mildew, green bug,
wheat curl mite and stem rust (Zeller, 1973; Zeller and Hsam,
1983). Nevertheless, other
advantages, such as improved grain yield (Villareal, 1995; Kim
2004) and grain protein content
(Lee et al, 1995) were also important. Henry et al., (1993) also
demonstrated that IBL.IRS
translocation greatly improved the in vitro regeneration of
microspore-derived haploid embryos.
In contrast, IBL.IRS translocation was largely associated with
end-use defects such as sticky
dough, poor rheology of the dough (strength, tenacity, and
extensibility), low resistance to over-
mixing and small volume of the bread. Such attributes were
explained first by the secalin
encoded by the 1RS chromosome, but it appeared that the quality
defect associated with
IBL.IRS was due to the loss of the LMW-GS and gliadins encoded
at Glu-BS and Gli-Bl
respectively (Figure 1.1). Since recombination is impossible
between IBS and 1RS, the IBL.IRS
translocation is inherited as a block. Thus both its advantages
and disadvantages are linked. To
break such an association, attempts to generate smaller
insertions of rye material were tested with
the 1D.1R recombinant lines (Rogowsky et al, 1993). The other
alternative would be to create a
genetic background in which the disadvantages of the IBL.IRS
translocation are suppressed or at
least greatly reduced.
GH-B1 -11 Sec'Glu-B3 I
Glu-B1 -ü Glu-B1
Chromosome Chromosome
1B1BL. 1RS
Figure 1.1 "regular" IB chromosome (left) and 1BL 1RS
translocated chromosome (right) of wheat and their
associated
prolamm loci The hatched material represents rye chromatin
4
-
The aim of the studyTo understand the basis of gluten rheology,
it is necessary to determine the rheology of its
elementary component: the HMW-GS. The number of cysteine
residues was proved to determine
the ability of each subunit to polymerize. Less is known about
the determinants of extensibility.
The role of the helicoidal central domain was investigated
unsuccessfully by evaluating
transgenic wheat lines, which carry a modified HMW-GS in the
central domain. The goal of the
first part of the present study was to quantify the effects of
small-sequence variations in HMW-
GS on the measured quality parameters. This approach was based
on the comparable HMW-GS
alleles Glu-Al 2 and Glu-Al 2*, which were present in the
studied DH population. Field
experiments were performed to analyze the impact of these
alleles on a broad range of quality
traits. Furthermore, the population also had two other variant
alleles of prolamin: Glu-BS and
Glu-D3. The presence of Glu-B3 j marked the presence of the
1BL.1RS translocation. The impact
of this chromosomal rearrangement is generally negative on the
bread-making quality. However,
translocation does occur in some high-quality cultivars.
Assessing how this translocation interacts
with the other prolamin loci and with the environment could lead
to a better appreciation of the
quality profile of the 1BL.1RS genotypes. Analysis of the impact
of the variation in the LMW-
GS of Glu-DS on quality was planned to address the relatively
contradictory results published in
the literature. Establishing the effect on quality of the
different alleles (at the polymorphic loci) of
HMW-GS and LMW-GS of the present population will provide
valuable information for breeders.
Knowing the precise impact of these alleles on a broad range of
quality traits will help in
achieving precise breeding objectives.
Fundamental insight will also be gained by observing how the
wheat proteome copes with the
replacement of chromosome IBS by the chromosome 1RS (rye). This
approach ought to prove
the existence of a complex prolamin regulation network. Previous
studies suggested such
mechanisms: Dumur et al. (2004) demonstrated that, in monosomic
lines of the cultivar Courtot,
the loss of prolamin loci boosted the production of other
prolamins and even favored the
expression of prolamins that remained silent in the "wild-type"
cultivar. Recent studies
demonstrated that HMW-GS accumulation is influenced by the
nature of the HMW-GS (Ravel et
al, 2006). Nevertheless, no clear regulation scheme is available
today for explaining the
regulation of prolamins.
5
-
As well as gaining a better understanding of the regulation
networks of prolamins, the goal is to
determine how wheat is able to reactivate silenced prolamins
when and if they are required. This
may lead to a re-evaluation of the variability available to
breeders. Several prolamin-coding
sequences are present at each locus but few proteins are
expressed and are accumulated in the
endosperm (Clarke and Appels 1999). Such observations lead to
the question as to why such a
large pool of variability was conserved in the wheat genome and
why it is kept silenced. It is
known that the accumulation of prolamins in the endosperm is
vital for the persistence of the
species. Thus, robust mechanisms of compensation should exist to
ensure the "packaging" and,
thus, the accumulation of prolamins in the endosperm even when a
major locus is lost.
6
-
2 Ax2",a new high molecular weight glutenin subunit
coded by Glu-Al: its predicted structure and its
impact on bread-making quality.
Published in Plant Breading (2007), volume 126, issue 1, pages
1-4.
2.1 Introduction
By conferring visco-elasticity to the dough (for review Shewry
et al, 1992), the high molecular
weight glutenin subunits (HMW-GS) determine the suitability of
the wheat (Triticum aestivum
L.) for bread-making. The amino acid composition of these
proteins explains their polymeric and
elastic behavior (Shewry and Tatham, 1997) and, consequently,
determines the rheology and the
baking quality of the dough (Veraverbeke and Delcour, 2002).
The HMW-GS proteins are coded by the complex Glul loci on the
long arm of the chromosomes
1A, IB and ID. Each locus codes for two structurally different
kinds of proteins: the x-type and
the y-type HMW-GS. Due to silencing phenomena and/or
pseudogenes, the total number of
HMW-GS in a cultivar ranges from three to five with some
exceptions reported for Swedish
bread wheat cultivars which have six HMW-GS (Margiotta et al.,
1996). To date, only three x-
type alleles at Glu-Al have been reported (Gianibelli et al,
2001).
The variability amongst the HMW-GS alleles has an impact on
quality. Anderson et al. (1989)
reported the existence of an "extra" cysteine residue in the
central domain of the HMW-GS Dx5
(allele Glu-Dld). In an in vitro system, Buoncore et al. (1998)
demonstrated that this "extra"
cysteine residue is responsible for the good quality associated
with this subunit. D'Ovidio et al.
(1996) established that the central domain of an HMW-GS coded by
Glu-Dl shows an insertion
of a 187 amino acids. He et al. (2005) suggest that, in
transformed plants, such an event may
increase the strength of the dough. No reports have been
published yet, which discuss the impact
on quality of smaller changes occurring in the central
repetitive domain.
The quantitative aspect also plays an important role in the
bread-making ability. Thus, the over-
expressed Bx7 subunit in cultivars such as 'Glenlea' or 'CD87'
leads to better quality (Marchylo
et al, 1992; Butow et al, 2003).
7
-
Here we report the sequence of a new HMW-GS x-type allele of
Glu-Al and the impact of
limited changes in the sequence of its central repetitive domain
on ten traits important for bread-
making.
2.2 Material and methods
Swiss genotypes of Triticum aestivum, from Agroscope
Changins-Wädenswil (Nyon,
Switzerland), were used. The breeding line '211.12014' was
utilized to clone and sequence the
allele 2 (at Glu-Al). As a control, we also used the cultivar
'Toronit' to clone and sequence the
allele 2* (at Glu-Al). A double haploid (DH) population was
created by crossing the two
genotypes ('211.12014' X 'Toronit'). The microspores of Fi
plants were isolated and cultivated
according to Kunz et al. (2000) to obtain 174 DH lines. This
population was evaluated in field
experiments in 2004 and 2005 at Nyon (430m above sea level) on
loamy cambisol soil. Average
precipitation was 1037mm in 2004 and 774mm in 2005.
Fertilization was done to meet the
plant's needs.
The following measurements were carried out for each line:
protein content (PR) and grain
hardness (GH) were measured by near infrared reflectance
spectroscopy (MRS) according to the
ICC standard N°159, the Zeleny sedimentation test (ZEL) was done
according to the ICC
standard N°l 16/1, the Alveograph parameters such as tenacity
(P), elasticity (L) and strength (W),
according to the ICC standard N°121 and Farinograph parameters,
such as water absorption
(H20), resistance (RES) and degree of softening (SOF) according
to the ICC standard N° 115/1.
The Rapid Mix Test (RMT) was carried out as described by
Pelshenke et al. (1970). The ICC
standards are referenced in Standard Methods of the
International Association for Cereal Science
and Technology (1999).
The statistical analysis of the data was performed with the
program SAS 8.2 (SAS Institut, Cary,
NC, USA). The distribution of the residuals from the analysis of
each of the parameters was
evaluated by the Tukey-Anscombe and the QQ-plots methods. Since
the assumptions of the
ANOVA (homoscedasticity, normality) were not met for PR, RES,
SOF, P, W and GH, standard
data transformation for continuous (log), counts (square root)
and percentages/proportions
(arcsines of the square root) were tested. The analysis was
performed with the Mixed procedure.
Proteins were extracted from whole meal flour of 'Toronit',
'211.12014' and all the DH lines
according to the protocol of Singh et al. (1991). The proteins
were separated (230 minutes at
8
-
30mA) on 12 5% acrylamide SDS-PAGE gels (16X18 cm) The bands
were revealed after
Coomassie blue staining to obtain the glutemn profile of the
different genotypes
Genomic DNA was extracted from the leaves of young seedlings
using the Clontech®
NucleoSpin® Plant Kit The polymerase chain reaction (PCR) of the
Glu-Al gene was performed
with a Biometra® T3Thermocycler in 50ul volumes using the Qiagen
HotStartTaq® The
primers PI (5'-TAGCCAACCTTCACAATCTCT-3') and P2 (5*-
ATAGCTAANGTGCATGCATGCC-3') were used to amplify the whole open
reading frame
(ORF) of 2* and 2 , the primers P3
(5'-5-ATCAATCCCGCACATCCTCTC-3') and P4 (5*-
GCAAAAAGAACCAACCGCTTAGT-3') were used to amplify the upstream
region of the 2
and 2* ORF The PCR products were cloned with the Quiagen PCR
Cloning Kit® The obtained
Ax2 sequence was deposited in GenBank® (http //www ncbi nlm nih
gov/Genbank/index html)
under the accession number DQ533690 The ORF detection was
carried out with the ORF Finder
software at http //www ncbi nlm nih gov/gorf/gorf html The
sequence alignment was performed
with Multahgn (Corpet, 1988) (http //prodes toulouse inra
fr/multalin/multahn html) The
identification of the cis-acting elements in the upstream region
of ORF was performed with the
PlantCARE software (http //bioinformatics psb ugent
be/webtools/plantcare/html/)
2.3 Results
The SDS-PAGE experiment showed that the wheat genotype '211
12014' had an atypical x-type
Glu-Al allele In this cultivar, the HMW-GS coded by Glu-Al
showed a small mobility shift
compared to a classical 2* subunit (NCBI reference N° m22208) as
shown in Figure 2 1
The 2 ORF was detected with the ORF Finder program in NCBI This
predicted ORF was 2475
bp long The cloned upstream region of the ORF was 2173 bp
long
The sequence alignment of the upstream region of the 2 and 2*
ORFs showed 17 single
nucleotide substitutions, three single-base insertions, two
single-base deletions and one triple-
base deletion These sequences were analyzed with the PlantCARE
program to identify
potentially different cis-elements that may play a role in the
regulation of the expression of the
two alleles Compared to the 2* ORF upstream region, the 2 ORF
upstream region did not show
a difference in the predicted cis-elements, which are known to
be involved in prolamin
regulation, one prolamin box at -216 plus two CGN4-hke motifs
(-568 and -547) were identical
in both promoters
9
-
The predicted protein sequence translated from the ORF of 2 had
a theoretical molecular weight
of 89,47kDa (the theoretical mass predicted from the 2* sequence
is 88,47kDa) and a length of
824 amino acids (2* is nine amino acids shorter). The predicted
amino acid composition ofN and
C termini domains were identical in 2 and 2*. Thus, all the 2*
cysteines were conserved in 2 .
No extra cysteine appeared in the central repetitive domain of 2
. Nevertheless, 21 single amino
acid substitutions, two insertions (six and nine amino acids)
and one deletion (six amino acids)
were observed in the 2 central repetitive domain (Figure 2.2).
The amino acid substitutions
occurred over the whole length of the repetitive domain. The
insertions started at positions +134
and +507 respectively. The deletion occurred at the end of the
repetitive domain (+715). These
mutations led to some changes in the tripeptide, hexapeptide and
nonapeptide 2 central domain
composition compared with 2*. In 2 , 36 GQQ motifs were observed
instead of 37 in 2*, 8
B
Glu-B1
GIU-D1
Figure 2.1 SDS-PAGE separation of the HMW-GS of '211 14014'
(A) in and 'Toromt'(B) The blanc arrowhead points to the new
Glu-
Al allele 2 and the other one points to 2* The digits on the
right
correspond to the HMW-GS of the Glu-Bl 7+9 and Glu-Dl 5+10
alleles (common to both genotypes).
10
-
PGQGQQ motifs instead of 6, and 8 GYYPTSPQQ motifs instead of 9.
The two extra
hexapeptides were created by the two insertions at +134 and +507
and the missing nonapeptide
was the result of the substitution of Gly by an Arg at position
+566.
The statistical analysis of the results of the quality
measurements (Table 2.1) shows that the
impact of the year on the measured traits was highly significant
(P < 0.001, except RMT P <
0.01). There were no significant variations (at 5 % error
probability) due to 2 on the measured
traits in comparison to 2*, with the exception of the Zeleny
sedimentation test showing that the 2
subunit induced an average increase of 3.46 % in 2004 and 6.98 %
in 2005 compared to 2* (P <
0.05). The effect of the Glu-Al*Year was also non significant.
It was also demonstrated, that 2
did not interact significantly with the other glutenin loci
(data not shown). This implies that both
Glu-Al alleles, 2 and 2*, influenced the measured rheological
parameters in an identical manner.
1 130
PBLLRlffLSVTSPOJVSYYPGaflS---
PflLLRRifLSVTSPQaVSYYPGaflSPQIl
131 260
Rx2*
—SQRPGQGQQEYYLTSPQQSGQMQQrGQGQSGïyPTSPQQSGQKQPGYYPTBPMQPEQLQQPTQGQQRQQPGQGQQLRQGQQGQQSGOGQPRYyPTSSQQPGQLQQLnQGQQGQQPEKGOQGQQSGQÛ8x2"
PGQGQ8PGQGQQEYYLTSPQQSGQUQQPGQGQSGYYPTSPQQ5GQEQPGYYPTSPMQP8QLQQPTQGQQRQQPGQGQQLRQGQQGQQSGQGQPRYYPTSSQQPGQLQQLRQGQQGQQPEÜGQQGQQSGQG
261 390
I I
11x2*
QQLGQGQQGQQPGQKQQSGQGQQGYYPISPQQLGQGqQSGQGQLGYYPTSPQQSGQGQSGYYPTSHQQPGQLQQSTQEQQLGQEQQDQQSGQGRQGQQSGQRQQDQQSGQGQQPGQRQPGYYSTSPQQLG8x2"
QQLGQGQQGQQPGQKQQSGQGQQGYYPISPQQLGQGQQSGQGQLGYYPTSPQQLGQGQSGYYPTSHQQPGQLQQSTQEQQLGQEQQDQDPGQGROGQQLGQROQDQQSGaGQQPGORQPGYYSTSPQQLG
391 520
I I
(1x2*
QGQPRYYPTSPQQPGQEQQPRQLQQPEQGQQGQQPEQGQQGQQQRQGEQGQQPG0GQQG«QPGQGQPGYYPTSPQQSGQGQPGYYPTSPQQS6QLQQPRQGOQPGQEQQGQQPGQGQQ
PGQ8x2"
QGOPRYYPTSPQQPGQEQQPRQLQQPEQGQQGQQPEQGQQGQQPaQGEQGQQPGQGQQGKOPGQGQPGYYPTSPQOSGQGQPGYYPTSPeQSEQLBQPflQGQQPGQEQQGQQPGQGQQGQQPGOGQQPGO
521 G50
I I
11x2*
OPPGYYPTSPQQSGQEQQLEQMQQSGQGQPGHYPTSPLQPGQGQPSYYPTSPQBIGQGQQPGQLQQPTQGQQGQQPGQGQQGQQPGEGCQGQQPGQGQQPGQOQPGYYPTSLQQSGQGQQPGqMQQPGCG8x2"
RQPGYYPTSPQQSGQEQQLE0HQQSGQG0PGHYPTSPLQPGOGQPKYYPTSPQaiGQGQQPGQLQQPTQGQQ8QQPGQGQQGQQffGQGDQGQQPGt)GQQPGQGgPGYYPTSLQQSGQGQQPGQUQQPGQG
651 780
I I
flx2*
QPGYYPTSSLQPEQGQQGYYPTSQQQPGQGPQPGQHQ0BGQGQQGYYPTSPQQ5GQGQQPG0ULQPGQHLQGGYYLTGPQQLGQGQQPRQULQPRQGQQGYYPTSPQQSGQGQQLGQGQQGYYPTGPQQS8x2"
QPGYYPTSSL01GQGQQGYYPÏSQQQPGQGPQPGQMQQ1.GQGQQGYYPTSPQQSGQGQQPG0ULQ
SGYYLTSPQQLGQGQQPROHLQPRQGQQGYYPTSPQQSGQGQQLGQGQQGYYPTSPQQS
781 830
I i
Rx2*
GQGQQGYDSPYHVSREHQflflSLKVnKflQPLflflQLPflNCRtEGGDflttflS!!8x2-'GQGQQGYDSPYHVSflEHQHflSLKVRKRQQLflHQLPflrtCRLEGGDflLLflSi)Figure2.2AlignmentofthepredictedprimarystructureofAx2andAx2*HMW-GSThegrayshadingindicatesammoacidsubstitutionThepercentageofhomologybetweenthetwosequencesis945%(CLUSTALW18)11
-
2.4 Discussion
Shewry et al. (1992) and D'Ovidio et al. (1997) suggested that
the repetitive ß-turns motives in
the central domain of HMW-GS explain their anomalously slow
migration in SDS-PAGE. In our
SDS-PAGE experiments we also observed a difference between the
subunits 2* and 2 , the
former migrating faster than the latter. This mobility shift is
probably due to the combination of
two elements: the lkDa difference in the theoretical size of the
two proteins, as well as
modifications in the protein's secondary structure produced by a
different repetitive domain.
In the upstream region of the ORF, no differences in prolamin
boxes, CGN4-like or other cis-
acting elements important for endosperm expression (Thomas and
Flavell, 1990; Norre et al,
2002) were found between the two alleles. Such a high similarity
between the two promoters is
not in favor of a differential level of expression between the
two alleles. Moreover, the
rheological tests (discussion follows) do not show significant
variation in quality due to one of
the alleles. The quantification of the transcripts would address
this issue but we can already state
that we found no evidence in favor of a differential level of
expression between the subunits 2*
and 2 after comparing their ORF upstream region.
Branlard et al. (2001) found that the alleles Axl and Ax2* have
an identical impact on the
strength and extensibility of the dough as well as on the
Pelshenke test. Other studies (Cornish et
al, 2001) confirmed a similar impact on other quality
parameters. It was also demonstrated that
the null allele at the same locus has a negative effect on all
the quality traits (Branlard et al,
1992; Cornish et al, 2001). The quality tests showed no
significant differences between Ax2 and
Ax2* in our DH population with the exception of the Zeleny test
(Table 2.1). This is in
accordance with the complete conservation of cysteines in the
two alleles, indicating that the
patterns of intra- and inter-chain disulfide bonds are identical
for both proteins. Indeed, visco-
elastic parameters of dough are reported to be affected strongly
by additional cysteines (reviewed
by Shwery et al, 2000). Therefore it is coherent that the
measured visco-elastic parameters in our
DH population are not significantly different between the two
alleles. D'Ovidio et al. (1996)
reported a large insertion in the central domain of the Dx2.2*
HMW-GS and He et al. (2005)
tried to demonstrate that larger repetitive domains may affect
the strength of the dough. In this
study we demonstrated that small insertions (six plus nine amino
acids) or deletions (six amino
acids) in the central domain -inducing minor modifications in
the tri, hexa and nona-peptide
composition- do not produce a significant variation in the
rheology of dough. The slightly higher
12
-
volume of protein sedimentation (ZEL) observed in genotypes
carrying the Ax2 allele is
probably due to the same factors that explain the mobility
shift; a small conformational
modification plus an increase of IkDa in mass. These differences
might cause greater swelling of
the 2 subunit under the conditions of the Zeleny test (lactic
acid solution). The fact that only the
glutenins are capable of swelling in the Zeleny test (Eckert et
al. 1993) supports our observations.
Nevertheless, the differences observed in the Zeleny
sedimentation test between subunits 2 and
2* are probably too small to produce differences in the other,
more direct, quality parameters
such as RES, W or RMT.
By reporting a new (the fourth) x-type allele at Glu-Al we
increased the variability available for
selection, even if the effects of the 2 subunit are
indistinguishable from those of the 2* subunit
on quality parameters. Our study demonstrates that the alleles
Ax2 , Ax2* (and so Axl) can be
considered as having the same positive impact on quality
compared to the null allele (at Glu-Al).
Detecting the Ax2 allele in a breeding program predicts no
detrimental impact of this glutenin on
quality.
13
-
Table
2.1Analysis
oftheimpactofthe
alle
licvariationon
qual
ityparameters
indoubledha
ploi
dlinescontaining
eitherAx2
orAx2*HMW-GS
PR
Kernel'spr
otei
ncontent,ZEL
Zeleny
test,H20
water
absorption,RES
resistance
ofdo
ugh,
SOF
softeningofdo
ugh,RMT
RapidMix
Test,P
tenacity,L
elasticity,W
strength,GH
gram
hard
ness
,BU
Brabender®
units,DF
degreesoffreedom
PR(%)
ZEL
(ml)
H20(%)
RES
(min)
SOF(BU)
RMT
(ml)
P(mmH20)
L(mm)
W(1
04J)
GH
(%)
Fixedeffects
Sourceofvariation
DF
Sign
ific
ance
Year
1***
***
***
***
***
***
***
***
***
***
Glu-Al
1ns
*ns
ns
ns
ns
Ns
ns
ns
ns
Year*Glu-Al
1ns
ns
ns
ns
ns
ns
Ns
ns
ns
ns
Random
effects
Covarianceparameters
Estimates
Geno
type
(G/w-^47)
068
542
171
046
795
1283
152
513
2628
106
Residual
021
209
122
092
302
588
44
1588
1470
279
Comparisonofmeans
Alleles
2004
Ax2
1562±0
10
6156±096
6027±044
647±0
14
8813±368
6233±78
5736±156
1687±3
92480±7
32209±023
Ax2*
1558±0
11
5950±098
6031±046
631±0
14
8912±377 2005
6325±72
5747±1
58
1566±40
2416±7
52200±024
Ax2
1386±0
10
5041±091
5917±040
491±0
13
1123±3
56066±46
6393±147
1141±3
51972±68
2020±021
Ax2*
1380±0
10
4712±092
5856±040
478±0
13
1144±3
56150±47
6480±148
1125±36
1974±68
1987±021
Signific
ant
atP<005,P<001andP<0001
respectively, ns
non
sign
ific
ant
14
-
3 Effect of the 1BL.1RS translocation and of the Glu-
B3 variation on bread-making quality tests in a
doubled haploid population.
Submitted for publication to Journal ofCereal Science
3.1 Introduction
Wheat (Triticum aestivum) is the most suitable cereal for
bread-making. This unique feature is
determined by the gluten polymer that gives its rheological
property to the dough. The gluten
enables the retention of CO2 in the dough during fermentation,
giving its volume to the bread.
The gluten polymer is an assemblage of different prolamin
subunits, namely the high molecular
weight glutenin subunits (HMW-GS), the low molecular weight
glutenin subunits (LMW-GS)
and the gliadins. Polymerization of the different subunits is
achieved through interchain disulfide
or hydrogen bonds (Veraverbeke and Delcour, 2002). The nature
and the amount of these
elements in the endosperm, as well as their allelic variations,
determine most of the variation in
the bread-making quality (Branlard et al, 2001). Several
rheological studies demonstrated that
glutenins play a large role in resistance to deformation (for
review, see Shewry et al, 2002) due
to the interchain disulfide bonds that they produce. Gliadins
mainly affect extensibility by acting
as 'solvent' agents (Wieser and Kieffer, 2001). In an effort to
summarize the known effects of
HMW-GS and LMW-GS, quality indexes were created and the most
common alleles were
scored (Branlard et al, 1992).
The substitution of the chromosome arm IBS of wheat by the 1RS
arm of rye (1BL.1RS
translocation) was designed to improve the resistance of wheat
to several pathogens such as
Puccinia recondita, Puccinia graminis, Puccinia striiformis and
Blumeria graminis (Zeller,
1973; Zeller and Hsam, 1983). This chromosomal rearrangement had
also important impact on
yield and rheology. The loss of Glu-BS and Gli-Bl loci, encoding
LMW-GS and gliadins
respectively, was shown to be detrimental to parameters related
to end-use quality (Burnett et al.,
1995a; Wieser et al. 2000). However, the persistence of the
1BL.1RS translocation in the
breeding material is certainly due to its positive impact on
yield (Villareal et al, 1991; Kim et al,
2004).
15
-
Evaluation of bread-making quality is achieved by performing
direct baking tests (Pelshenke et
al, 1970). Nevertheless, rheology tests (Alveograph and
Farinograph) and analysis of the grain
constituents (protein content, grain hardness) provide more
detailed information about the
different quality components and, thus, are used to predict
bread-making quality in breeding
programs. In the present study 15 highly informative quality
parameters, covering the whole
range of the bread-making process, were investigated to achieve
a better understanding of the
impact of 1BL. 1RS translocation and of the allelic variation at
Glu-DS. The doubled haploid
(DH) population used here shows rather limited prolamin
polymorphism. Analysing this
population in a two-year experiment enabled a thorough
quantification of the contribution to the
quality of the 1BL.1RS translocation and of the b and c alleles
encoded at Glu-DS to quality. The
results should allow a better understanding of wheat
bread-making quality and should be of
interest to breeders.
3.2 Materials and methods
3.2.1 Plant material and experimental design
The doubled haploid population was created by crossing two
genotypes of the Swiss spring
wheat breeding program: line 211.12014 and cultivar Toronit
(Brabant et al, 2006). The
microspores of the Fi plants were isolated and cultivated
according to Kunz et al. (2000). The
population was multiplied in the green house by selfing the DH
lines. For the quality assessment
tests, the population was sown in the field at Nyon in 2004 and
2005 (430 m above sea level) on
loamy cambisol soil. The sums of the daily mean temperature and
precipitation were measured
during the vegetative period. According to the local recommended
practice, 110 and 120 kg/ha of
nitrogen were applied to the fields in 2004 and 2005
respectively. Assessment of phosphate,
potassium and magnesium were performed before seedling to ensure
that sufficient amounts of
these elements were available to the plants.
The experiments were performed in fully randomized blocks.
Because of the limited amount of
seeds obtained after multiplication in 2004, 138 DH lines were
sown in one to five micro-plots of
1.5 m2. In 2005 174 DH lines were sown in four micro-plots of
4.77 m2. A total of 126 lines were
used in the trials in both years. To asses the homogeneity of
the experimental field in 2004, both
parents were sown on 60 micro-plots and were randomly
distributed over the experimental field.
16
-
In 2005, no different number of replications biased the results
because all the DH lines were
sown on four micro-plots.
3.2.2 Quality tests
Protein content (PR) and grain hardness (GH) were measured on
intact kernels by means of a
Foss® Feed & Forage Analyzer (FOSS Analytical A/S,
Hilleroed, Denmark) near infrared
reflectance spectroscopy (MRS) system according to the ICC
standard N°159. Thousand kernel
weight (TKW) was calculated by counting the number of kernels in
a representative 10 g sample.
For the Zeleny sedimentation test (ZEL) seeds were round in a
Quadrumat® Junior grinder
(Brabender, Duisburg, Germany). The Zeleny test was performed
according to the ICC standard
N° 116/1. The falling number (FN) was measured on the wholemeal
of grains ground in a
Perten® grinder (Perten Instruments AB, Huddinge, Sweden). The
test was performed according
to ICC standard N° 107/1. To test the rest of the quality
parameters, flour was obtained by
grinding the wheat grains in a MLU-202 grinder (Bühler, Uzwil,
Switzerland). The ash content
was adjusted to approximately 0.55 % by adding the corresponding
higher ash fraction if
necessary. Farinograph parameters, such as water absorption
(H20), resistance (RES) and degree
of softening (SOF), were measured on 50 g of flour according to
the ICC standard N° 115/1.
Chopin's Alveograph (CHOPIN Technologies, Villeneuve-la-Garenne,
France) parameters were
measured according to the ICC standard N°121; 250 g of flour
were used for each test. The
amount of water added was adjusted according to the water
content of the flour. The
representative Alveograph curve of each sample was deduced from
the best four of five curves.
Tenacity (P), extensibility (L) and strength (W), elasticity
index (le) and P/L were evaluated in
this test. All the ICC standards cited here are referenced in
Standard Methods of the International
Association for Cereal Science and Technology (1999). To
evaluate the baking behaviour of
each DH line in a direct test, the rapid mix test (RMT) was
carried out as described by Pelshenke
et al. (1970). The amount of water and flour used for this test
were adjusted for each DH line
according to the water absorption measured with the Farinograph
and to the water content of the
flour measured by MRS.
3.2.3 SDS-PAGE and allelic variation
Proteins were extracted from single grains of the cultivar
Toronit, 211.12014 and from all the
DH lines according to the protocol of Singh et al. (1991). The
embryos were excised from the
17
-
kernels that were used to establish the prolamin profiles The
extracted glutenins were separated
on 12 5 %T 1-D SDS-PAGE (18X16 cm) gels (3 h 30 min at 35 mA)
The ro-ghadins were
separated on 10 %T 1-D SDS-PAGE gels (3 h at 35 mA) The bands
were visible after staining
with Coomassie blue The HMW-GS, LMW-GS and co-ghadin profiles
were established for all
the genotypes in the experiment based on the obtained gels
3 2 4 Statistic analysis
The statistical analysis of the data was performed with the SAS
8 2 program (SAS Institute, Cary,
NC, USA) The distribution of the residuals of the analysis of
each of the parameters was
evaluated by the Tukey-Anscombe and the QQ-plots methods Since
the assumptions of the
ANOVA (homoscedasticity, normality) were not met for PR, GH,
RES, SOF, P and W, standard
data transformation for continuous (log) counts (square root)
and percentages/proportions
(arcsines of the square root) were tested The analysis was
performed with the MIXED procedure
The allelic variation nested inside the genotypes were
considered to be random factors The best
fit for the models was determined with the stepAIC procedure in
R 2 0 1 (http //www r-
project org/)
3.3 Results and discussion
3 3 1 Allelic variation in the DH population
SDS-PAGE analysis of the two parents and of the DH offspring
demonstrated that the prolamin
variation was limited to four loci Glu-Al, Glu-B3, Glu-D3 and
Gh-Bl (Table 3 1) The HMW-
GSs of the DH lines were similar with the exception of those
encoded at Glu-Al with either
allele 2* or 2 was observed Both have the same impact on the
rheology of the dough and on
bread-making quality (Gobaa et al, 2007a) The presence of the
LMW-GS null allele j at Glu-BS
indicated the presence of the IBL 1RS translocation, it was
always associated with the secahn
allele 1 at Gh-Bl The % test demonstrated that the IBL 1RS
translocation was inherited in a 1 1
ratio (at P < 0 001) No recombination was found between the
alleles of Glu-BS and Gh-Bl,
suggesting that the IBL 1RS translocation was inherited as a
block Additional tests, based on
four PCR markers, confirmed that the IBL 1RS translocation of
parent Toronit replaced the
whole IB short arm of wheat by the 1R short arm of rye (Gobaa et
al, 2007b)
18
-
Table 3.1 Prolamm allelic profiles (as revealed by SDS-PAGE)
and
means of 15 quality parameters of the parental lines used to
generate
the doubled haploid lines, as measured in 2004 and 2005
Traits 211.12014 Toronit
HMW-GS
Glu-Al / Glu-Bl / Glu-Dl 2 / 7-9/5-10 2*/7-9/5-10
LMW-GS
Glu-A3 / Glu-B3 / Glu-D3 a/c/c a/j/b
co-gliadinsGh-Al /Gh-Bl /Gh-Dl a/b/b a/l/b
Protein and kernels
Zeleny (mL) 60 5±148 53 75±10 2
PC (%) 14 43±1 33 13 96±1 22
GH (%) 19 96±3 20 42±1 16
TKW (g) 39 8±0 57 39 4±1 27
Starch
Amylo (BU) 1265±117 798±96
Fal (s) 397±52 353±37
Rheology
H2Oabs (%) 61 5±0 57 58±2 55
RES (mm) 5 2±1 77 5 3±2 05
SOF (BU) 97±23 3 65±46 0
P (mmH20) 79±22 6 48±1 4
L (mm) 126±60 153±14
P/L 0 75±0 54 0 31±0 01
W (10"4J) 233±33 94 185±25 46
Baking test
RMT(mL) 627±34 3 573±53 7
BU Brabender units, PC protein content, GH grain hardness, TKW
thousand
kernels weight, FN falling number, H20 abs water absorption, RES
resistance,SOF softening, P tenacity, L extensibility, W strength,
le swelling index, RMT
rapid mix test
19
-
3.3.2 Preliminary tests
In 2004 the DH lines were grown in a different number of
replications. To estimate the bias due
to the pool of a different number of replications, the protein
content was measured of the 120
parental lines sown in micro-plots and randomly distributed
across the five replications of the
experiment. The effect of replication was not found significant
(at P < 0.05) for the protein
content of the kernels (data not shown). This observation proved
that the environmental
conditions tended to be homogeneous in 2004 (data not shown). In
2005, all the DH lines were
sown in four replications. Thus, no bias due to the pool of a
different number of repetitions is
possible. In a separate comparison of the results of 2004 and
2005, a comparison of the means
proved that the variation in the 15 quality parameters was
always concurrent for all the
polymorphic loci, although there was some degree of difference
in the range of the variation
(data not shown).
3.3.3 Protein- and kernel-related traits
All the protein- and kernel-related traits (Table 3.2) were
highly influenced by year. The higher
temperature in July 2004 may explain the higher amount of
protein and, thus, the greater Zeleny
values and grain hardness in that year (Spiertz et al, 2006).
Nevertheless, the 1BL.1RS
translocation increased the protein content of the endosperm by
an average of 4.5 % across both
years of the experiment. This increase was not due to a
difference in grain size because the
variation in TKW was not significant (Table 3.2). According to
Ehdaie et al. (2003), 1BL.1RS
genotypes mature later and, thus, allow longer photosynthate
accumulation and produce higher
yield. Since the accumulation of protein riches a plateau in the
late grain filling stage (Carceller
and Aussenac, 1999), delayed maturity probably favoured the
accumulation of protein over the
accumulation of starch. The variation in kernel hardness was
influenced mainly by year and less
by the chromosome IB status; the variation was always within the
boundaries of the hard wheat
class of both parental lines. Thus, identical settings of the
grinder were used to produce the flour
used in the experiments. The allele Glu-Al 2 induced slightly
higher Zeleny values than the
allele Glu-Al 2* (Gobaa et al, 2007a). Moreover, 1BL.1RS DH
lines had significantly lower
sedimentation volumes. Allele b of Glu-DS was also associated
with an average decrease of
11 %when compared to allele c of the same locus predicting on
average better quality parameters
for the c alleles encoded at Glu-B3 and Glu-D3.
20
-
Table3.2
Adjusted meanscomparisonbetweenthepolymorphic
alle
lesofthedoubledha
ploi
dpo
pula
tion
forfifteenqualityparameters
QualityTraits
Glu-Al
2Variation
Glu-B3
jVariation
Glu-D3
bVariation
2004
Year
2005
Variation
Number
ofDH
lines
153
159
162
150
135
177
138
174
Proteinandkernels
Zeleny(mL)
540±058
565±058
+480%
b599±058
506±059
+155%"
585+063
520+054
-111%"
610+052
495+047
+233%"
Proteincontent(%)
146+008
147±009
-143±009
15±009
-46%"
147+009
146+008
-156+007
138+006
+113%"
Grainhardness(%)
208±02
211±02
-206±02
213±02
-37%b
21+02
209+02
-22+02
20+0
1+93%a
TKW
(g)
388±037
377±038
-290%c
383±037
382±038
ns
381+04
384+035
ns
3801+0
3385+0
3ns
Starch
Amylograph
(BU)
1017±487
860±49
0-1550%c
1117±48
1761±496
+319%"
960+52
917+45
ns
959+38
918+36
ns
Fallingnumber
(s)
342±6
6323±6
6-585%c
352±6
6313±67
+113%a
329+7
1335+62
ns
339+5
6326+52
+42%c
Rheology
H20
abso
rpti
on(%)
597±022
597±022
ns
592±022
602±022
-17%b
597+024
598+0
21
ns
603+0
18
592+0
17
+19%a
Resistance(mm)
541±0
1560±0
1-
538±0
1563±0
11
579+0
11
522+009
-98%"
631+0
11
48+007
+239%a
Soft
enin
g(BU)
98±2
3953±2
3-
807±2
1114±2
5-412%"
878+24
1059+22
+206%"
837+1
81105+19
-32%"
Tenacity(mmH20)
604±1
2604±1
2-
625±1
3583±1
2+68%b
648+14
562+1
-132%a
57+09
639+1
-121%"
Extensibility(mm)
137±27
143±27
ns
151±27
129±27
+149%"
144+29
137+2
5ns
163+243
118+217
+381%"
P/L
049±002
049±002
ns
046±002
052±002
-12%c
--
-037+002
06+002
-384%"
Strength
(lO^
J)2182±47
224±4
8-
2458±5
1977±46
+196%"
2541+5
51903+4
1-251%"
2447+4
51986+3
6189%"
le
491±048
495±049
ns
506±048
480±049
+51%"
523+052
463+045
-115%"
523+043
463+038
+131%"
Baking
test
RMT
(mL)
624±4
8612±5
1ns
606±44
630±4
8-4%
"
608+5
628+42
+33%
b627+5
1609+3
1+29%
"
a,bandc
sign
ific
ant
atP<0001,P<001andP<005respectively,TKW
thousandkernelweight
,P/L
tenacity/extensibility,le
elasticity
inde
x,
RMT
rapid mix
test,nsnon
signif
ican
t,BU
Brabender
units,
-
non
avai
labl
e,theco
rres
pond
ingfactorswereeliminatedfromthemodelbythe
steptAIC
proc
edur
ensnon
sign
ific
ant
AllelesGlu-Al
2*,Glu-B3
c,Glu-D3
candyear2005werereferencesinthecalculationofthevariations
Standarderrorsforpr
otei
ncontent,gramhardness,resistance,softeningtenacity
and
strength wereap
prox
imat
edformback-transformedvalues
21
-
3.3.4 Starch-related tests
With respect to the gelatinization potential of the flour,
1BL.1RS lines were clearly at a
disadvantage with regard to the quality of starch. The
Amylograph and falling number values of
these lines were 31 % and 11 % lower than those of the IB lines,
respectively (Table 3.2). Year
did not have an effect in either FN or Amylograph test.
Pentosanes assessment was performed on
the flour samples of 2004. No significant effects of the 1BL.1RS
translocation or of other
polymorphic loci were observed (data not shown). Similar
observations were made by Burnett et
al. (1995b), who demonstrated that starch from 1BL.1RS genotypes
was less viscous and more
soluble than starch from "regular" IB wheat. These results
demonstrated that the 1BL.1RS lines
produce functionally different starch. The proteomic analysis of
16 DH lines representative of
the present population revealed the absence of a dimeric
alpha-amylase inhibitor in the 1BL.1RS
lines (ExPASy database: Q4U1A2) suggesting a higher
alpha-amylase activity (Gobaa et al.,
2007b). Whether this different quality of starch is involved in
the dough stickiness phenotype,
associated with the 1BL.1RS genotypes, is under
investigation.
Subunit Ax2* of Glu-Al was also associated with slightly higher
viscosity (Table 3.2). As a
direct involvement of the glutenin subunits in the Amylograph
viscosity is unlikely, the observed
effect was probably due to unidentified elements located on the
long arm of chromosome 1A.
This is supported by the recent mapping of a QTL, in the
vicinity of Glu-Al, for starch quality
(McCartney et al, 2006).
3.3.5 Rheology tests
Table 3.2 shows that translocated DH lines had significantly
lower P, L, W and le values and
higher SOF and P/L values. This decrease of the rheological
parameters is certainly the
consequence of the loss of the LMW-GS encoded at Glu-BS and
located on the short arm of
chromosome IB (Graybosch, 2001; Gobaa et al., 2007c). Such loss
of polymeric glutenin
subunits was reported to produce weaker gluten and thus reduced
Rnax (Farinograph) and
extensibility (Wieser et al., 2000).
In the Farinograph test, necessary hydration to reach 500
Brabender® units of strength was
significantly higher in 1BL.1RS lines. This was probably due to
the poorer starch gelatinization
of translocated lines discussed above. Variation at Glu-DS did
not produce significant changes in
this trait.
22
-
DH lines with the allele Glu-D3 b had significantly lower RES,
P, W and le values and higher
SOF values than DH lines with the allele Glu-DS c. However,
unlike 1BL.1RS translocation,
Glu-D3 din not have a direct impact on extensibility (L and
P/L). The variation at Glu-D3
seemed to preferentially affect resistance to deformation. These
results contradict those of
Branlard et al. (2001), who found that Glu-D3 c and Glu-D3 b
have exactly the same positive
effect on W. The presented results contradict also with those of
Gupta et al. (1994), who
demonstrated that the c and b alleles in Australian cultivars
were equivalent for Rmax but that
higher extensibility was associated with allele b rather than
allele c. However, correlations
between particular allelic forms of LMW-GS and quality
parameters of bread wheat can differ,
possibly due to the genetic background, gene interactions and
environmental effects (D'Ovidio
and Masci, 2004). According to Xu et al. (2006) an LMW-GS
encoded at Glu-D3 with extra
cysteine residues could improve the rheological parameters of
the dough. Whether or not the
allele Glu-D3 c of parent 211.12014 had such this LMW-GS was not
investigated.
When the unweighted mean variation in the seven rheology
parameters were measured, both
alleles Glu-B3 j and Glu-D3 b decrease quality by an average of
16 %. This demonstrates the
equal importance of the variation at Glu-B3 and Glu-D3 for the
rheological parameters.
Interactions of glutenins and year, of variable significance,
were also observed for the parameters
Zeleny, SOF, RES, L, W, le and RMT (Table 3.3). The association
of the two detrimental alleles
(Glu-B3 j and Glu-D3 b) had, of course, a strong negative effect
on the quality parameters.
Martin et al. (2001) reported a comparable effect when 1BL.1RS
was associated with the allele
Glu-Dl 2+12. Therefore, the production of high-quality cultivars
combining the 1BL. 1RS
translocation with a detrimental glutenin is unlikely. The
interaction of Glu-Al with the other
prolamins was limited; the Glu-Al 2 IGlu-D3 c combination had
better W and le values than the
other combinations. The interaction between Glu-B3 and year was
highly significant for the
Zeleny test, RES and W. In 2005 the lower protein content
associated with the 1BL.1RS
translocation seemed to drastically reduce the strength of the
dough. None of the other prolamin
loci showed a similar Environment X Genotype interaction;
effects of the glutenin alleles
encoded at Glu-Al and Glu-D3 were stable in both years for all
of the measured quality
parameters.
23
-
Table 3.3 F values for the observed significant interactions
between the studied sources of variation
Glu-A1*YR ns
Glu-B3*YR 20 59*** 12 84***
Glu-D3*YR ns ns
Glu-A1*Glu-B3 3 97* 451* ns
Glu-A1*Glu-D3 ns ns ns
Glu-B3*Glu-D3 7 71** 6 16* 14 56***
Source Zeleny Resistance Softening Extensibility Strength le
RMT
ns ns
9 91**
ns
ns 5 2* ns
412* 4 2* 4 25*
5_8£ 5 05* 8 25** nsThe empty cases correspond to the factors
eliminated from the model by the stepAIC procedure ns non
significant, le elasticity index, *, **, *** significant at P
< 0 05, P < 0 01 and P < 0 001 respectively
3.3.6 Baking tests
The volumes of the loaves, measured by the RMT test, were
significantly lower in regular IB
and in Glu-DS c lines, despite better results in the rheology
tests (Table 3.2). This apparent
contradiction could be due to several factors. First, the amount
of water used for the RMT test is
adjusted according to the water absorption as measured by the
Farinograph (Pelshenke et al.,
1970). Since H20 absorption is significantly higher in 1BL.1RS
lines, the greater amount of
water used in the RMT test for the 1BL.1RS lines may explain a
small part of the larger volumes
associated with IBLIRS lines. Second, loaf volume is highly
correlated with the protein content
of the flour (Wieser and Kieffer, 2001) but only weakly
correlated with other rheological
parameters such as the Rmax of the Mixograph (Kieffer et al,
1998). In the present study, the
average protein content was quite high (14 %) and it was
significantly higher (P < 0.001) in
1BL. 1RS lines. Seemingly, the resistance to deformation opposed
to the dough during
fermentation was limiting the volume of the loaves above a
certain value of P. These results can
be compared to those obtained in experiments with transgenic
wheat genotypes over-expressing
a glutenin subunit (1Dx5). Such lines exhibited very high
tenacity and, thus, abnormal
rheological behaviour (Rooke et al, 1999; Popineau et al,
2001).
3.4 Conclusions
The study of DH lines contrasting for three alleles of
prolamins, with and without the 1BL.1RS
translocation, determined their impact on 15 important quality
traits. For the first time an
important defect in the gelatinization of starch, revealed by
the Amylograph test, was associated
with the 1BL.1RS translocation in a DH population. However, the
mechanisms leading to such a
24
-
defect remain unknown. More investigations are needed to
determine whether IBL.IRS wheat
had a naturally "sprouted-like" starch or whether the defect in
gelatinization was caused by an
original starch branching.
The negative effect of IBL.IRS translocation on rheology was
confirmed by the present study,
but it was limited. IBL.IRS translocation produced a change in
rheology comparable to that
produced by "regular" polymorphism at another LMW-GS-encoding
locus. Allelic variation at
Glu-DS caused even greater variation in tenacity and strength.
This indicates, that in breeding
programs, the lower quality predicted by the detection of the
IBL.IRS translocation can be
counterbalanced by accumulating favourable alleles at Glu-AS and
Glu-DS and/or by selecting
genotypes with a high protein content. However, care should be
taken translocated material is
included in breeding programs, because the importance of the
impact of IBL.IRS translocation
on rheology seemed to be modulated by the environmental factor.
The final baking test
demonstrated that, in the present case, lowering rheological
parameters can lead to an increase in
the volume of the loaves. It is important to consider that this
occurred in a population that
accumulates the beneficial alleles for quality at Glu-Al, Glu-Bl
and Glu-Dl and in which the
mean protein content was high. The observation that very
high-strength genotypes may produce
lower bread volumes and that they need a high energy input
during mixing can seriously
handicap their direct use in bakery. They may be more suitable
for improving the quality, in
mixture with poor quality genotypes.
25
-
4 Proteomic analysis of wheat recombinant inbred
lines (Part I): effect of the 1BL.1RS translocation
on the wheat grain proteome.
Proteomics (2007), in press.
4.1 Introduction
In Triticum aestivum cultivars, the replacement of the IB
short-arm chromosome by the 1R short-
arm chromosome of rye was originally designed to contribute to
resistance to diseases such as
leaf rust, stem rust, stripe rust and powdery mildew (Zeller,
1973; Zeller and Hsam, 1983). Even
though, the corresponding pathogens (Puccinia recondita,
Puccinia gramins, Puccinia striiformis
and Blumeria graminis respectively) have since overcome this
resistance, 1BL.1RS translocation
is still used in breeding programs, mainly for its beneficial
effect on grain yield (Villareal et al.,
1995; Kim et al, 2004). Nevertheless, severe end-use defects
(sticky dough, low tolerance to
over-mixing, dough tenacity, low SDS-sedimentation volume and
low specific loaf volume) of
flour obtained from translocated genotypes have been reported
(reviewed by Graybosch, 2001).
Storage proteins, mainly prolamin (glutenin and gliadins), play
a major role in baking quality and
in the rheology of dough through their allelic or quantitative
variation (Shewry et al, 1992;
Veraverbeke and Delcour, 2002). It has been suggested that the
impact of 1BL.1RS translocation
on quality can be explained in part by the loss of the Glu-BS
locus, coding for one third of the
low molecular weight glutenin subunits (LMW-GS), and by the
substitution of the co-gliadins
(encoded at Gli-Bl) by co-secalins (encoded at Sec-1). These
changes drastically reduce the
polymeric fraction of gluten and increase the monomeric
fraction. However, it is not understood
how the 1BL.1RS translocation acts on the proteome of the wheat
grain.
Previous studies, on the proteome of wheat grain, demonstrated
that the presence of one or two
doses of chromosomes 1 A, IB or ID or the lack thereof has an
impact on the amount and nature
of several proteins that are not even located on the pair of
chromosomes affected by the
aneuploidy (Dumur et al., 2004). This demonstrated that storage
proteins are under the control of
complex regulation mechanisms and that compensation may occur
when loci coding for
prolamins are lost. The effects of heat stress (Majoul et al,
2003) and nitrogen fertilization
26
-
(Bahrman et al, 2004) were also analyzed. However, to the best
of our knowledge, no reports
exist on the impact of fragments of rye chromosomes on the wheat
grain proteome.
The goal of this study was to report on the effects of 1BL.1RS
translocation on the mature grain
proteome in contrasting doubled haploid (DH) lines of Triticum
aestivum. The comparison
between proteomic profiles characterizing translocated genotypes
and proteic profile
characterizing regular IB genotypes can contribute to a better
understanding of the mechanisms
that are affected by 1BL.1RS translocation, such as the
synthesis and accumulation of storage
proteins and the metabolism of starch.
4.2 Material and methods
4.2.1 Plant material
A doubled haploid population was produced by crossing two Swiss
genotypes of Triticum
aestivum, one carrying the 1BL.1RS translocation (Toronit) and
the other having a regular IB
chromosome (the breeding line 211.12014). The microspores of Fi
plants were isolated and
cultivated according to Kunz et al. (2000) to obtain 174 DH
lines. These lines were grown on
loamy Cambisol soil at Agroscope Changins-Wädenswil (430m above
sea level, Nyon,
Switzerland) in 2004. The average precipitation was 1037 mm.
Fertilization was performed
according to the local recommended practice (110 kg of nitrogen
per hectare).
4.2.2 Translocation mapping
Genomic DNA was extracted from the leaves of young seedlings
using the Clontech®
NucleoSpin® Plant Kit (Clontech Laboratories Inc., Mountain
View, CA, USA). To estimate the
size of the 1RS material in the genome of the cultivar Toronit
and the breeding line 211.12014, a
polymerase chain reaction (PCR) was performed with four markers:
5S, RIS, TEL and NOR
(Koebner, 1995). The annealing temperature was set at 65°C. PCRs
were performed with a
Biometra® T3Thermocycler (Biometra GmbH, Goettingen, Germany) in
50 \A volumes and with
Qiagen HotStartTaq® (Quiagen GmbH, Hilden, Germany). The
amplification products were
visualized on 1 % w/v standard agarose gels and co-migrated with
the Bio-Rad® 20 bp
Molecular Ruler (Bio-Rad Laboratoires, Hercules, CA, USA) to
determine their sizes.
4.2.3 1-D electrophoresis and allelic variation
Proteins were extracted from single grains of the cultivar
Toronit, 211.12014 and all the DH lines
according to the protocol of Singh et al. (1991). The glutenins
were separated on 12.5 %T 1-D
27
-
SDS-PAGE (18X16 cm) gels (3 h 30 min at 35 mA) The ro-ghadins
were separated on 10 %T 1-
D SDS-PAGE gels (3 h at 35 mA) After Coomassie blue staining the
bands became visible,
giving the glutenin and co-ghadin profiles of the different
genotypes
Eight DH lines carrying the IBL 1RS chromosome and eight DH
lines carrying the regular IB
chromosome were selected from the 174 available DH lines The DH
lines were selected on the
basis of the glutenin and ghadin profiles established by 1-D
SDS-PAGE
4 2 4 2D SDS-PAGE and quantitative variation
The 2-D SDS-PAGE experiment was performed on 16 selected DH
lines Wheat kernels, from
which the embryo had been excised, were ground in a Cyclotec
14920 grinder (Hillerod,
Denmark) The proteins were extracted from 100 mg of flour
according to Branlard and Bancel
(2006) IPG buffer 6-11 (2% v v) from GE Healthcare (Uppsala,
Sweden) was used as ampholyte
carrier and DTT 20 mM was added in the solubilization solution
The concentration in protein of
the extracts was assayed with a PlusOne 2-D Quant® Kit (GE
Healthcare) in a 5 \il aliquot of the
supernatant collected after the first centnfugation 120 ug of
extracted proteins were dissolved in
250 (il of extraction solution to passively réhydrate
immobilized pH gradient strips (Immobihne
DryStrip®, pH 6-11, 13 cm) Isoelectric focusing (IEF) and 2-DE
migration were performed
according to Dumur et al (2004) Fixation and Coomassie blue
staining were done according to
the method described by Neuhoff et al (1988) The resulting gels
were scanned with an
ImageScanner II (GE Healthcare) and the images were analyzed
with the ImageMaster 2D
Platinum v5 0 software (GE Healthcare) The percentage of volume
(%Vol) was measured for
each spot The obtained protein profiles of each selected DH line
were the result of duplicated
protein extractions used in four 2-DE replicates The spots were
considered only if they were
detected in, at least, three out of the four replicates
The statistical analysis was performed with the SAS v8 2
Software (SAS Institut, Cary, NC,
USA) The GLM procedure was used to detect the spots showing a
significant quantitative
variation (P < 0 001) in their %Vol between the conditions
"DH lines with IBL 1RS" and "DH
lines with the normal IB" Spots considered to be up- or
down-regulated showed at least a
twofold variation in their mean %Vol when both sets of
experimental conditions were compared
28
-
4 2 5 Protein identification
After image and statistical analyses, the relevant spots were
cut out and transferred to Eppendorf
low-binding micro tubes for mass spectrometry First, Coomassie
blue was eliminated with
ammonium bicarbonate-acetomtnle buffers The spots were incubated
for 30 mm in 100 ul of
buffer 1 containing 25 mM NH4HCO3 and 5 % v/v acetonitnle at
room temperature Then the
spots were washed twice with 100 ul of buffer 2 containing 25 mM
NH4HCO3 and 50 % v/v
acetonitnle (30 min incubation at room temperature) Finally, the
spots were washed for 10 mm
with 200 ul of pure acetonitnle The tubes were dried in a
Heto-Vac system (VR-1 Jovan Nordic,
Allerod, Denmark) For in-gel digestion with trypsin plus
chymotrypsin digestion, 15 ul of
trypsin (V5111, Promega, Madison, WI, USA), 10 ng/ul in 25 mM
NH4HCO3, 15 ul of
chymotrypsin (C6423, Sigma, St Louis, MO, USA), 10 ng/uL in 50
mM NH4HCO3 and 10 mM
CaCl2 were added to the dry gel Digestion was performed
overnight at 37°C The pieces of gel
were centnfuged and 10 ul acetonitnle were added to extract the
peptides The mixture was
sonicated for 5 mm and centnfuged at 5,000 g for 5 mm For
MALDI-TOF MS analysis, 1 (il of
supernatant was loaded directly onto the MALDI target One
microliter of the matrix solution (5
mg ml"1 a-cyano-4-hydroxycinnamic acid in 50% acetonitnle / 0 1%
tnfluoroacetic acid) was
added immediately and the mixture was left to dry at room
temperature A Voyager DE-Pro
model of the MALDI-TOF mass spectrometer (Perseptive BioSystems,
Farmingham, MA, USA)
was used in positive-ion reflector mode for peptide mass
fingerprinting External calibration was
performed with a standard peptide solution (Proteomix, LaserBio
Labs, Sophia-Antipohs,
France) Monoisotopic peptide masses were assigned and used from
NCBI or from a local
version of poaceae uniprot or from EST database searches with
the Mascot and ProFound
softwares (http //www matrrxscience com and http //prowl
rockefeller edu) Matches to protein
sequences from the Viridiplantae taxon were considered
acceptable if at least four peptide masses
from the PMF matched and a Z score of 1 00 or higher was
obtained from ProFound or a
significant score was obtained from MASCOT, which rates scores
as significant if they are above
the 95 % significance threshold (P < 0 05)
For NanoLC-MS/MS analysis, FIPLC was performed with an Ultimate
LC system combined with
a Famos autosample and a Switchos II microcolumn switching for
preconcentration (LC
Packings, Amsterdam, The Netherlands) The samples after
hydrolysis were loaded onto the
column (PEPMAP CI8, 5 urn, 75 urn ID, 15 cm, LC Packings) using
a preconcentration step in a
29
-
microprecolumn cartridge (300 \im ID, 1 mm) The sample (6 (iL)
was loaded onto the pre-
column at 40 uL/min After 3 mm, the pre-column was connected to
the separating column and
the gradient was started at 200 nL/min The solvents with 0 5 %
formic acid contained 95 %
water / 5 % acetomtnle (A) and 95 % acetomtnle / 5 % water (B) A
linear gradient of 10 to 90 %
B was applied for 45 mm For ion trap-MS, a LCQ deca with a nano
electrospray interface
(ThermoElectron, Les Ulis, France) was used Ionization (18 kV
ionization potential) was
performed with a liquid junction and an uncoated capillary probe
(New Objective, Cambridge,
MA, USA) Peptide ions were analyzed by the data-dependent
"triple-play" method as follows
(l) full MS scan (m/z 400-2000), (n) ZoomScan (scan of the major
ion with higher resolution),
(in) MS/MS of the later ion Mass data, collected during a
LC-MS