at SciVerse ScienceDirect

Journal of Cereal Science 56 (2012) 161e170

Contents lists available

Journal of Cereal Science

journal homepage: www.elsevier .com/locate/ jcs

Significant down-regulation of g-gliadins has minor effect on gluten and starchproperties of bread wheat

Javier Gil-Humanes a, Fernando Pistón a, Cristina M. Rosell b, Francisco Barro a,*

a Instituto de Agricultura Sostenible, CSIC, Alameda del Obispo s/n, E-14080 Córdoba, Spainb Instituto de Agroquímica y Tecnología de Alimentos, CSIC, E-46100 Valencia, Spain

a r t i c l e i n f o

Article history:Received 7 July 2011Received in revised form18 January 2012Accepted 16 February 2012

Keywords:Wheat gliadinsGluten proteinsDough qualityBread-making qualityRNAiMixolab

Abbreviations: SDSS test, sodium dodecyl sulfatehigh molecular weight; LMW, low molecular weight;* Corresponding author. Tel.: þ34 957499240; fax:

E-mail address: fbarro@ias.csic.es (F. Barro).

0733-5210/$ e see front matter � 2012 Elsevier Ltd.doi:10.1016/j.jcs.2012.02.009

a b s t r a c t

The contribution of gliadins to the baking performance of wheat has been widely discussed. In thepresent study we report the properties of the flours of fifteen transgenic wheat lines with g-gliadinsdown-regulated by RNAi. Technological properties were assessed by sodium dodecyl sulfate sedimen-tation (SDSS) test, and rheological properties were characterized during mixing using the Mixolab�.Changes in the proportions of gluten proteins, with significant increases of glutenins, were observed intransgenic lines. These changes did not affect the SDSS test in most of the transgenic lines. The doughstrength of some transgenic lines was reduced as determined by the Mixolab�. Changes in the starchbehavior were also observed in the transgenic lines, in which lower values of torque were observed at theC3, C4 and C5 points of the mixing curve. The results reported here are important in understanding therole of g-gliadins in the bread-making quality of flours.

� 2012 Elsevier Ltd. All rights reserved.

1. Introduction

Dough formed from wheat flours possesses unique biome-chanical properties, which allow it to be processed into pasta andnoodles, and which contribute to the retention of carbon dioxideduring fermentation, causing expansion of the dough to giveleavened bread. These properties depend on the structures andinteractions of the grain storage proteins, which together form the‘gluten’ protein fraction (Shewry et al., 2009). Wheat glutenproteins account for about 80% of the total grain proteins in typicalEuropeanwheats (Shewry, 2009) and are classified into two proteinfamilies: glutenins and gliadins. The glutenins consist of twodifferent subunits, the high molecular weight (HMW) and the lowmolecular weight (LMW) types. Glutenins form complex polymersstabilized by inter-chain disulfide bonds and are mainly relatedwith dough elasticity. The gliadins are classed into three structuraltypes: a-, g- and u-gliadins. They are monomeric components thateither lack inter-chain disulfide bonds or contain only intra-chainbonds, and they contribute mainly to the extensibility andviscosity of dough (Shewry and Halford, 2003). Some authors havesuggested that the HMW glutenins are major determinants of

sedimentation test; HMW,RNAi, RNA interference.þ34 957499252.

All rights reserved.

bread quality as they are encoded by single genes, and allelicdifferences in the HMW composition influence the structure andproperties of the glutenin polymers (Payne, 1987; Shewry et al.,1989). The functional properties of individual HMW gluteninsubunits were further confirmed by genetic transformation withHMW genes (Altpeter et al., 1996; Barro et al., 2003, 1997; Blechlet al., 2007; León et al., 2009). Although gliadins are related toloaf volume (Hoseney et al., 1969; Weegels et al., 1994), the roles ofindividual gliadins are not well understood (Fido et al., 1997) asthey are encoded by large multigene families and inherited inblocks. In addition, LMWglutenin subunit genes are linked to somegliadin genes, so that the effects of gliadins on quality are difficult tointerpret (Payne et al., 1987). The purification and addition ofindividual groups of gliadins to flours provided contradictoryindications of their contribution to dough quality (Branlard andDardevet, 1985; Khatkar et al., 2002; MacRitchie, 1987), and nega-tive, positive and/or no correlations were reported betweenamounts and types of gliadins, and functionality of wheat flours infunctional tests (Khatkar et al., 2002; Nieto-Taladriz et al., 1994; vanLonkhuijsen et al., 1992). Likewise, the effects on technologicalproperties of the isolation of gliadins and the incorporationprocedures used in these tests have to be considered since theymay alter the native conformational structures of the proteins(Weegels et al., 1994). van den Broeck et al. (2009) evaluatedthe technological properties of bread wheat deletion lines and re-ported that deletions of the a-gliadin locus on the short arm of

J. Gil-Humanes et al. / Journal of Cereal Science 56 (2012) 161e170162

chromosome 6D resulted in a significant decrease in technologicalproperties, whereas deletions in the short arm of chromosome 1D(u-gliadin, g-gliadin and LMW loci) maintained technologicalproperties. In a subsequent work, the same authors (van den Broecket al., 2011) reported a positive effect on the bread-making qualitydue to the deletion of the locus Gli-D2 of the short arm of thechromosome 6D, which contains the a-gliadin genes. However,chromosome deletions that successfully remove groups of geneslinked in clusters, also remove parts of the genome that may haveimportant implications in the normal metabolism of the plant and/or in the regulation of other genes.

RNAi-mediated gene silencing is a very useful tool for studyinggene function as it allows the specific silencing of genes or geneclusters in polyploid species (Lawrence and Pikaard, 2003; Mikiet al., 2005; Travella et al., 2006). Using this technology, thecontents of a-gliadins (Becker et al., 2006), g-gliadins (Gil-Humanes et al., 2008) and all a-, g- and u-gliadins (Gil-Humaneset al., 2010) were strongly down-regulated in bread wheat. Asa result, the gluten protein content and the proportions of proteinfractions in the grain were significantly altered. The silencing of a-gliadins in bread wheat, leading to 63% reduction in the a-gliadincontent, resulted in increased dough strength but slightlydecreased loaf volume (Wieser et al., 2006), whereas the silencingof the g-gliadins slightly increased the SDS sedimentation volume(Gil-Humanes et al., 2010) in a few wheat lines. However, moredetailed studies are required to understand the effect that thespecific silencing of groups of gliadins has on the rheological andfunctional properties of bread wheat.

The Mixolab� is an instrument that can be used to characterizethe rheological behavior of dough subjected to a dual mixing andtemperature constraint. It allows the measurement of importantphysical dough properties, like dough development time andstability, andpastingproperties of starch in actual doughs, all of theseparameters being indicators of quality and potential end-use inproducts.AlthoughtheMixolab� is a relativelynewinstrument, ithasbeen already used in research works to determine the quality ofwheat flours to manufacture bread (Collar et al., 2007; Marco andRosell, 2008; Rosell et al., 2009, 2010), cakes (Kahraman et al.,2008), cookies (Ozturk et al., 2008); and also to assess the quality offlours derived fromother cereals such as rice (Torbica et al., 2010), rye(Banuet al., 2010) andoat (Huanget al., 2010). TheMixolab�hasbeensuccessfully compared with traditional methods for determiningdough quality like Farinograph, Extensograph, Amylograph, FallingNumber and Rapid Visco-Analyser (RVA), providing good predictorsfor all the parameters measured with these traditional methods(Cato and Mills, 2008; Codina et al., 2010; Koksel et al., 2009).

In the present study we report the technological properties offifteen transgenic lines of bread wheat with the g-gliadins down-regulated by RNAi (Gil-Humanes et al., 2008, 2010). The techno-logical properties of the flours of the transgenic lines were deter-mined by the sodium dodecyl sulfate sedimentation (SDSS) test andby using the Mixolab�. The results have been studied, along withthe chemical composition of the flour, to analyze the influence of g-gliadin silencing on the bread-making quality. The results reportedhere are important to help our understanding of the role of g-gliadins on the bread-making quality of flours and the effects of thegliadin genes down-regulation and the modification of the glutenprotein synthesis.

2. Materials and methods

2.1. Plant material

Eight transgenic lines of Triticum aestivum cv ‘Bobwhite 208’(BW208) and seven transgenic lines of T. aestivum cv ‘Bobwhite

2003’ (BW2003) with reduced levels of g-gliadins, and thecorresponding wild-type lines, were used in this study. LineBW2003 was found to carry the T1BL.1RS translocation. All trans-genic lines were previously reported or obtained as described byGil-Humanes et al. (2008, 2010) using two hairpin RNA (hpRNA)vectors: the pghpg8.1 vector containing the D-hordein promoter(Pistón et al., 2008) and the pGhpg8.1 vector harboring theg-gliadin promoter (Pistón et al., 2009). Lines A1152, A1158, C655,C657, D445, D623, C217 and D598 contained the pghpg8.1 vector,and lines D577, D682, D715, D716, D815, 22A and 22C containedthe pGhpg8.1 vector. Transgenic lines were self-pollinated forfour generations to obtain homozygous plants and assayed asdescribed below.

2.2. Polyacrylamide gel electrophoresis analysis

Two mature seeds from one block were crushed into a finepowder and used to extract the endosperm storage proteins. Glia-dins were extracted from 40 to 100 mg of flour with a 5:1(ml mg�1

flour) ratio solution of 1.5MN,N-Dimethylformamide, 20%(w/v) sucrose for 45 min at room temperature (RT) with shaking.Samples were centrifuged for 20min at 13,000�g and supernatantsloaded and separated in AePAGE gels for 1 h 15 min at 15 �C and40 mA per gel. Gel composition was as described in Khan et al.(1985) with some minor modifications: 3.8 � 10�4 ferrous sulfateinstead 2.1 � 10�4 and 0.125% (w/v) aluminum lactate instead ofsodium lactate. Pellets from the previous step were washed twotimes with a 50% (v/v) aqueous 1-propanol solution for 30 min at60 �C, and centrifuged for 10 min at 13,000�g. Supernatants werediscarded and glutenins extracted with a 5:1 (ml mg�1

flour) ratiosolution of 0.08 M TRISeHCl, pH 8.5, 50% (v/v) 1-propanol, 2% (w/v)dithiothreitol (DTT) during 30min at 60 �C with vortexing. Sampleswere centrifuged for 10 min at 13,000�g and cold acetone wasadded to the supernatants in a ratio of 5:1 (v/v) and incubatedovernight at �20 �C. Samples were centrifuged for 10 min at13,000�g and the precipitated glutenins resuspended in a 5:1(ml mg�1

flour) ratio solution of 0.125 M TRISeHCl, pH 6.8, 2% (w/v)SDS, 2% (w/v) DTT, 10% (v/v) glycerol, 0.002% (w/v) BromophenolBlue and incubated for 30 min at 60 �C with continuous vortexing.Glutenin extracts were loaded and separated in SDSePAGE gels for5 h and 30 min at 10 �C and 15 mA per gel. Tricine electrophoresisbuffer and acrylamide stacking and resolving (T ¼ 10%, C ¼ 1.28)gels were as described by Shewry et al. (1995). AePAGE andSDSePAGE gels were stained with a solution of 0.05% (w/v) Coo-massie Blue R-250, 5% (v/v) ethanol, 12% (w/v) trichloroacetic acidand then immersed in tap water overnight to remove excess ofstain.

2.3. Protein and starch analysis and SDS sedimentation test

The protein content of whole flour was calculated from theKjeldahl nitrogen content (%N � 5.7) according to the standard ICCmethod no. 105/2 (ICC, 1994a), and the starch content was deter-mined according to the standard ICCmethod no.123/1 (ICC,1994b).Both parameters were expressed on a 14% moisture basis. The SDSsedimentation volume was determined as described by Williamset al. (1988). Three technical replicates were carried out for eachbiological sample.

2.4. Reversed-phase high-performance liquid chromatography(RP-HPLC)

The gliadin fractionwas extracted from 100mg of flour stepwisethree times with 670 ml of 60% (v/v) ethanol, vortexing for 2 minand incubation at RT 10 min with shaking. Samples were

J. Gil-Humanes et al. / Journal of Cereal Science 56 (2012) 161e170 163

centrifuged at 6000�g for 20 min, supernatants were collected andmixed all together. The glutenin fraction was extracted from theinsoluble material of the previous step stepwise two times with500 ml of 50% (v/v) 1-propanol, 2 M urea, 0.05 M TriseHCl (pH 7.5)and 2% (w/v) DTT, vortexing for 2 min at RT and incubation for15 min at 60 �C with shaking. Samples were centrifuged at 6000�gfor 20 min; supernatants were collected and mixed all together.Finally, samples were filtered through a 0.45 mm nylon filter(Teknokroma, Barcelona, Spain). Gliadin (40 ml) and glutenin (40 ml)extracts were applied to a 300SB-C8 reverse phase analyticalcolumn (4.6 � 250 mm, 5 mm particle size, 300 Å pore size; AgilentTechnologies, Santa Clara, CA) using a 1200 Series Quaternary LCSystem liquid chromatograph (Agilent Technologies, Santa Clara,CA) with a DAD UVeV detector, as described byWieser et al. (1998).Absorbance was monitored with the DAD UVeVmodule at 210 nm.The integration procedure was handled automatically by the soft-ware with some minor manual adjustment. Absolute amounts ofgliadin and glutenin fractions were determined using bovine serumalbumin (BSA; BSA� 98%, fraction V., SigmaeAldrich, St. Louis, MO)as standard. Three independent repetitions were carried out foreach transgenic line and wild type. The ratio glutenins:gliadins(Glu:Gli) (Table 1) of each line was calculated as the mean of theGlu:Gli ratios of the three independent repetitions. Due to theoutlier data that were eliminated in the statistical analysis it ispossible that in some of the lines the Glu:Gli ratio does not fit withthe value obtained by using the means of total glutenins and totalgliadins.

2.5. Mixolab� analysis

Mixing and pasting properties of wheat flour samples werestudied using the Mixolab� analyzer (Chopin Technologies,Villeneuve-la-Garenne Cedex, France). For each assay, 50 g of wholemeal flour were loaded into the Mixolab� bowl and mixed underthe principle of constant water absorption of 70%. The standard‘Chopinþ’ protocol was used as follows: 80 rpm of constant speedduring the entire experiment, with 8 min of initial mixing at 30 �C,a temperature increase phase of around 15 min (2.3 �C min�1) until90 �C, 7 min at 90 �C, a cooling phase of around 10min (4 �Cmin�1)

Table 1Thousand seeds weight, starch and protein contents, SDS sedimentation test, and protei

Line HPLC (mg protein mg�1flour)

1000 seeds (g) Starch (%) Protein (%) g-glia

BW208 wild type 44.5 53.8 11.9 24.1A1152 47.3 52.8 13.4* 0.7*A1158 46 54.7 12.6 1.7*C655 43.4 53.5 12.7 2.8*C657 45.1 53.3 12.5 8.4*D445 34.5** 49.8* 15.3** 1.2*D623 44.8 54.5 12.9 3.2*D577 43.1 52.8 13.3* 4.5*D682 45.9 53.9 12.2 5.1*

Mean transgenics 43.8 53.7 13.1* 3.5*

BW2003 wild type 45.1 55.2 12.0 18.6C217 43.7 55.0 12.2 10.2*D598 42.7 55.6 12.4 6.1*D715 49.6* 55.3 12.8 4.4*D716 48.7* 53.3 13.2 2.0*D815 36.1** 51.6 12.9 2.1*22A 49.3* 54.4 12.7 3.2*22C 45.4 54.7 12.0 1.6*

Mean transgenics 45.1 54.4 12.5 4.2*

Means are significantly different to control (values in bold) as determined by Dunnett's

until 50 �C, and a final step of 5 min at 50 �C (Fig. 1). Three inde-pendent replicates of each line were analyzed. As occurred with theratio Glu:Gli, due to the eliminated outlier data, in some cases thesevalues may vary from those obtained by using the means of C3, C4and C5.

2.6. Experimental design and statistical analysis

The homozygous transgenic lines were assayed usinga randomized complete block design with three replicates of fiveplants each. Data were analyzed with the statistical software Rversion 2.12.1 (Ihaka and Gentleman, 1996) using the GraphicalUser Interface (GUI) R Commander, and the SPSS version 11.0statistical software package (SPSS Inc., Somers, NY). Majorassumptions of analysis of variance (ANOVA) were confirmed bythe KolmogoroveSmirnov’s test for normal distribution and by theLevene’s test for homogeneity of variances. Outlier data wereeliminated using the function outlierTest (package car). ANOVA andtwo-tailed Dunnett’s test for median multiple comparisons wereused to analyze the results and compare between transgenicand wild-type lines. P values lower than 0.05 were consideredsignificant, and lower than 0.01 were considered highly significant.Correlation matrix was obtained by Pearson’s correlation analysis.

3. Results

3.1. AePAGE and SDSePAGE analysis

Fig. 2 shows the gliadin (Fig. 2A) and glutenin (Fig. 2B) profile ofBW208 and BW2003 transgenic lines and the corresponding wild-type lines. In BW208 transgenic lines, twomajor bands were down-regulated in the g-gliadin region (Fig. 2A left). In BW2003 trans-genic lines, two bands in the g-gliadin fraction were down-regulated, and an additional band was reduced in the a-gliadinregion (Fig. 2A right). Possibly, more gliadin bands were down-regulated in the transgenic lines, but they were not detected inthe AePAGE gels due to the co-migration of proteins. This effectwas observed previously in these lines by Gil-Humanes et al. (2008)by comparing the AePAGE and MALDI-TOF silencing profiles. The

n composition of the flour samples.

dins Total gliadins Total glutenins Glu:Gli ratio SDS (ml)

74.5 24.5 0.33 12.9* 65.8 30 0.46 14.3* 118.1 27.7 0.31 12.2* 64.3 27 0.42 13.8

123.9 26.8 0.47 12.6* 157.6 33.4** 0.43 13.4* 80.9 29.2 0.37 12.2

102.3 30.3 0.34 13.3* 131 28.2 0.41 13.3

* 105.5 29.1* 0.40 13.1

63.3 26.3 0.42 12.4* 111.7 33.3* 0.45 13.9* 66.1 28 0.43 13.0* 71.2 31.5 0.45 13.7* 68.1 32.9* 0.48 15.0** 63.2 31.1 0.49 14.6* 109.8 37.0** 0.51 14.6* 63.1 33.0* 0.52 14.5

* 79 32.4** 0.48 14.2*

multiple comparison as follow: *P < 0.05; **P < 0.01.

Fig. 2. AePAGE of gliadins (A) and SDSePAGE of glutenins (B) from transgenic and wild-tysilenced bands in the a-gliadin region. HMW subunits (2*, 7 þ 9 and 5 þ 10) are labeled on g(B) are probably due to the translocation T1BL.1RS in genotype BW2003.

Fig. 1. Mixolab standard curve for wheat dough. DDT: dough development time.

J. Gil-Humanes et al. / Journal of Cereal Science 56 (2012) 161e170164

SDSePAGE analysis of transgenic and wild-type lines (Fig. 2B)showed that BW208 and BW2003 genotypes contain the sameHMW-GS 2*, 7 þ 9 and 5 þ 10, but show marked differences intheir LMW-GS profiles, probably due to the lack of the Glu-B3 locusof wheat in genotype BW2003 (that carries the translocationT1BL.1RS). As showed in Fig. 2B, HMW-GS and LMW-GS subunitswere not down-regulated by RNAi, and both transgenic and wildtypes had the same profile of bands.

3.2. Grain quality and protein characterization

Thousand seed weights of wild-type lines BW208 and BW2003were 44.5 g and 45.1 g, respectively (Table 1). Most of the BW208transgenic lines did not show significant differences with respect tothewild-type line for the thousand seed weight, and only line D445had a significant lower value with 34.5 g. However, in BW2003transgenic lines, four lines presented significant differences forthousand seed weight in comparison to that of the wild type: linesD715, D716 and 22A had higher values (49.6 g, 48.7 g and 49.3 g,respectively), whereas line D815 presented a lower value with

pe lines. Squares indicate silenced bands in the g-gliadin region and asterisks indicateels of panel B. Differences in the LMW profile between BW208 and BW2003 genotypes

J. Gil-Humanes et al. / Journal of Cereal Science 56 (2012) 161e170 165

36.1 g (Table 1). Wild-type lines BW208 and BW2003 showedstarch contents of 53.7% and 54.4%, respectively, and proteincontents of 11.9% and 12%, respectively (Table 1). The starch contentwas not affected in most of the transgenic lines, and only line D445presented a significantly lower content with respect to the wildtype BW208. In contrast, the percentage of protein was higher inmost of the transgenic lines, and lines A1152 (13.4%), D445 (15.3%)and D577 (13.3%), and the mean of BW208 transgenic lines (13.1%)showed significant increases with respect to the wild-type lineBW208 (Table 1). The protein composition of the flour wasevaluated in transgenic lines by using RP-HPLC. The content of g-gliadins of the wild types BW208 and BW2003 was 24.1 and18.6 mg protein mg�1

flour, respectively. A strong and significantreduction of g-gliadins was clearly observed in all the transgeniclines, independently of the genotype and the construct (Table 1).For the transgenic lines, the content of g-gliadins ranged from 0.7 to8.4 mg protein mg�1

flour in BW208 lines, and from 1.6 to10.2 mg protein mg�1

flour in BW2003 lines, which representsa reduction between 65e97% and 45e91%, respectively, for BW208and BW2003 transgenic lines. Despite this drastic reduction in g-gliadins, the total content of gliadins did not show significantdifferences in the transgenic lines with respect to the correspond-ing wild-type lines (Table 1). However, some lines had highercontents of gliadins than the wild types. Regarding the content of

Fig. 3. Mixolab curves of transgenic and wild-type lines of genotype BW208 (A) and BW2003the region surrounding the maximum torque C1. Wild types are represented by a wider bl

glutenins, both wild-type lines presented similar contents of glu-tenins: 24.5 and 26.3 mg protein mg�1

flour for BW208 andBW2003, respectively. All the transgenic lines had higher valuesthan those of the corresponding wild-type lines. In addition, fivetransgenic lines, namely D445, C217, D716, 22A and 22C, hadsignificantly higher values than their wild types, with glutenincontents of 33.4, 33.3, 32.9, 37.0 and 33.0 mg protein mg�1

flour,respectively. The glutenins:gliadins ratio was slightly affected,and the differences observed were not significant compared withthe wild types (Table 1). The SDS sedimentation test showed nodifferences in the BW208 transgenic lines compared with thewild type, whereas for BW2003 transgenic lines, only line D716,and the mean value of all BW2003 transgenic lines showedsignificant higher values than the wild type (Table 1).

3.3. Mixing and pasting properties of flour

The two wheat lines used in this work contain the same HMW-GS 2*, 7þ 9 and 5þ10, but showmarked differences in their LMW-GS profile (Fig. 2). Fig. 3 shows a representative curve of theMixolab� from the whole meal flours of each of the transgenic andwild-type lines studied in this work. The Mixolab� measures in realtime the torque (Nm) produced by the passage of dough betweentwo kneading arms spinning at a constant speed. The instrument

(B). A representative curve of each genotype is shown. Insets show an amplification ofack line.

J. Gil-Humanes et al. / Journal of Cereal Science 56 (2012) 161e170166

allows the analysis of the quality of the protein network and starchproperties during heating and cooling. The typical Mixolab� curvecan be divided into two different phases: the first phase is deter-mined by the properties of the proteins and represents theconsistency (torque) (Nm) of the dough during mixing. Parametersrecorded in this phase are: initial consistency (Nm) or maximumtorque at 30 �C (C1), dough development time (min) or time toreach the maximum torque at 30 �C, stability (min) or elapsed timeat which the torque is maintained constant (torque � 90% C1),curve slope a and minimum torque (Nm) or the minimum value oftorque produced under mechanical and thermal constrains (C2).The second phase shows the properties of the starch and representsthe viscosity (torque) (Nm) of the dough under mixing and heatingconstraints. Parameters recorded in this phase are: curve slope b,peak torque/viscosity or the maximum torque produced during theheating stage (C3), the minimum torque during the heating stage(C4) and torque value obtained after cooling at 50 �C (C5). Cookingstability (C3eC4) and setback (C5eC4) were also calculated for eachline. Different curves were obtained for the BW208 and BW2003wild types, probably due to the different composition in gliadinsand LMW-GS, in part consequence of the 1BL.1RS translocation. Inaddition, different curves were also obtained for the transgeniclines compared with the corresponding wild types (Fig. 3). BW208transgenic lines showed differences in the first part of the Mixolab�

curve compared with the wild type, with a shorter developmenttime (time to peak C1) (Fig. 3A). In contrast, for BW2003 lines(Fig. 3B) the curves at the initial mixing part of the process werevery similar between transgenic and wild-type lines, with thelargest differences being observed as the temperature increased.Table 2 shows the data obtained from the Mixolab� curves. Valuesfor development time, stability and maximum torque (C1) weresimilar between the wild types BW208 and BW2003 and nosignificant differences were found. For BW208 lines, the meanvalue for development time of the transgenic lines was significantlylower, and two individual lines also had significantly lower valuesthan the wild type, namely A1152 and D445. Although stabilityvalues were not significantly different between transgenic andwild-type lines, the mean value of stability of all transgenic lineswas significantly lower when compared with the wild-type lineBW208. Development time and stability were very similar in all

Table 2Mixolab� parameters of the flour samples from wild types and transgenic lines of whea

Developmenttime (min)

Stability(min)

Maximumtorque (C1)(Nm)

Minimumtorque (C2)(Nm)

Temat C

BW208 wild type 3.54 2.76 1.00 0.40 53.8A1152 2.65* 1.74 0.99 0.36 55.4A1158 2.84 1.81 0.85 0.29** 56.2C655 3.17 2.34 1.02 0.41 54.4C657 2.86 2.36 1.05 0.41 54.3D445 2.38* 1.60 1.00 0.36 54.9D623 2.98 2.37 1.03 0.41 55.0D577 2.81 1.92 0.98 0.35 55.7D682 3.00 2.30 0.96 0.37 54.7

Mean transgenics 2.84** 2.06* 0.99 0.37 54.9

BW2003 wild type 3.87 2.65 0.95 0.34 57.4C217 3.60 2.36 0.94 0.33 57.3D598 3.23 2.16 0.91 0.30 57.0D715 3.67 2.33 1.05 0.39 55.6D716 3.41 2.45 1.00 0.32 58.5D815 3.28 3.01 0.86 0.29 57.122A 3.60 2.82 0.99 0.35 57.822C 3.88 2.64 0.97 0.32 57.8

Mean transgenics 3.52 2.54 0.96 0.33 57.3

Means are significantly different to control (values in bold) as determined by Dunnett's

BW2003 lines and no differences were observed between trans-genic and wild-type lines (Table 2). The values of maximum (C1)and minimum torques (C2) were unaffected in most of the trans-genic lines, and only line A1158 had a C2 value significantly lowerthan the wild type BW208 (Table 2). The torque at C3, C4 and C5was lower in most of the BW208 and BW2003 transgenic lines(Fig. 3). In BW208 genotypes, lines A1158 and D577 showedsignificantly reduced torque values at C3 and C5, but line D577 alsohad a significant reduction in torque at C4, whereas torque for lineC657 was reduced significantly only at C5 in comparison to thewildtype. However, values of torque between C3 and C5 for the mean ofBW208 transgenic lines did not differ significantly from the wild-type lines (Table 2). Although none of the BW2003 transgeniclines presented significant reductions of torque at C3, C4 or C5, themean value of all BW2003 transgenic lines was significantly lowerthan the wild-type lines (Table 2). The curve slopes a and b wereunaffected in all the transgenic lines. For the average of BW208transgenic lines, neither the cooking stability nor the setback pre-sented differences to the wild type, and only lines A1158, C657 andD577 had significantly lower values of setback than the wild type(Table 2). As a consequence of the differences observed in the tor-que at C3, C4 and C5 in transgenic BW2003 genotypes, the meanvalues of cooking stability and setback were significantly lowerthan the control line, and for two individual lines (D815 and 22A),they were also significantly reduced.

3.4. Correlations between flour composition and technologicalparameters

Table 3 shows the correlations between main flour componentsshown in Table 1 and the technological parameters of the flourshown in Table 2. The SDS sedimentation test was found to bepositively and highly correlated with total glutenin content(r ¼ 0.70) and with the glutenins:gliadins ratio (r ¼ 0.73), andnegatively correlatedwith g-gliadin content (r¼�0.42). The doughdevelopment time did not show strong correlations, but significantnegative correlations were found with total protein (r¼�0.69) andtotal gliadin (r ¼ �0.55) contents, whereas significant positivecorrelations were found with starch content (r ¼ 0.65) and g-gliadin content (r ¼ 0.43) (Table 3). Stability was negatively

t.

perature2 (�C)

C3(Nm)

C4(Nm)

C5(Nm)

Alpha Beta Cookingstability(C3eC4)

Setback(C5eC4)

3 1.52 1.19 1.71 �0.04 0.43 0.33 0.520 1.40* 1.25 1.85 �0.04 0.38 0.20 0.600 1.34** 1.08 1.45* �0.03 0.40 0.27 0.37**3 1.50 1.15 1.60 �0.04 0.43 0.35 0.453 1.50 1.08 1.48* �0.04 0.43 0.42 0.40*0 1.43 1.25 1.83 �0.04 0.43 0.19 0.580 1.49 1.14 1.60 �0.04 0.47 0.35 0.460 1.41* 1.04* 1.45* �0.04 0.42 0.37 0.41*7 1.49 1.16 1.61 �0.04 0.47 0.32 0.45

5 1.44 1.13 1.59 �0.04 0.43 0.31 0.46

7 1.38 1.05 1.39 �0.04 0.47 0.33 0.350 1.34 0.88 1.16 �0.04 0.48 0.46 0.283 1.34 0.92 1.19 �0.04 0.51 0.42 0.280 1.26 0.79 1.09 �0.05 0.30 0.47 0.300 1.22 0.76 1.03 �0.05 0.48 0.46 0.273 1.26 0.87 1.12 �0.04 0.42 0.39 0.25*0 1.20 0.65 0.90 �0.04 0.37 0.55 0.25*0 1.28 0.91 1.18 �0.04 0.53 0.37 0.27

3 1.27* 0.83* 1.10** �0.04 0.44 0.43* 0.27**

multiple comparison as follow: *P < 0.05; **P < 0.01.

Table 3Pearson’s correlation coefficients between Mixolab� parameters and flour composition (starch and proteins). SDSS: sodium dodecyl sulfate sedimentation test; DDT: doughdevelopment time.

SDSS DDT Stability C1 C2 C3 C4 C5 a b Cooking Setback

Starch NS 0.65** NS NS NS NS NS �0.43* NS NS 0.49* �0.49*Total proteins NS �0.69** �0.61* NS NS NS NS NS NS NS �0.43* 0.44*g-gliadins �0.42* 0.43* NS NS NS NS NS NS NS NS NS NSTotal gliadins NS �0.55* �0.46* NS NS NS NS NS NS NS NS NSTotal glutenins 0.70** NS NS NS NS �0.72* �0.61** �0.53* NS NS NS NSGlu:Gli 0.73*** NS NS NS NS �0.57* �0.56* �0.53* �0.41* NS 0.42* �0.46*

Correlations are significant at: *P < 0.1; **P < 0.01; ***P < 0.001; NS: no significant.

J. Gil-Humanes et al. / Journal of Cereal Science 56 (2012) 161e170 167

correlated with total protein content (r ¼ �0.61) and total gliadincontent (r ¼ �0.46). Total glutenins and the ratio of glu-tenins:gliadins were also negatively correlated with the torquevalues of the last part of the Mixolab� plot: C3, C4 and C5. Amongthem, the highest correlation was found between total glutenincontent and C3 (r ¼ �0.72). In addition, the starch content was alsonegatively correlated with the C5 value (r ¼ �0.43). The derivedvalues, cooking stability (C3eC4) and setback (C5eC4) were mainlycorrelated with starch content, but also with total proteins and theratio of glutenins:gliadins (Table 3).

4. Discussion

RNAi is a useful tool for gene silencing, and specifically for thestudy of gene function by the down-regulation of groups of proteinsencoded by multigene families. Gil-Humanes et al. (2008) reportedthe specific reduction of g-gliadins in transgenic bread wheat linesby using RNAi techniques, with no off-target effects or changes ingrain morphology being observed. These transgenic lines areexcellent materials to study the influence of g-gliadins on thetechnological properties of the flour and also to understand thecompensatory processes that operate in the grain in response togene silencing.

Compensatory effects of B-hordein and glutelins (Lange et al.,2007), and B- and g-hordein (Hansen et al., 2007) were reportedin transgenic lines of barley with reduced levels of C-hordein. Thisup-regulation of barley protein fractions was linked to the up-regulation of the gene encoding the barley prolamin-bindingfactor (BPBF) (Hansen et al., 2007), indicating that these compen-satory effects may be transcriptionally regulated. In the presentwork, modifications in the expression of storage proteins wereobserved in all the transgenic lines as a consequence of thesuppression of the g-gliadins (Table 1). The reduction of g-gliadinswas significant in all the 15 transgenic lines, but the total content ofgliadins was not significantly different to the wild type due to theup-regulation of the other groups of gliadins. Total glutenins werealso up-regulated and were significantly higher in five of thetransgenic lines and in the mean values for BW208 and BW2003transgenic lines (Table 1). Interestingly, the ratio glutenins:gliadinswas not significantly affected in transgenic lines in comparisonwith their wild types (Table 1), suggesting that the changes in thestorage proteins in response to the suppression of g-gliadins couldbe aimed at maintaining the balance of glutenins:gliadins in thegrain. Most of the transgenic lines had SDS sedimentation valueshigher than those of the wild types, and the mean value of SDSsedimentation for transgenic lines was significantly affected inBW2003 lines. The SDS sedimentation test was significantlycorrelated with the glutenin content and the ratio of glu-tenins:gliadins (Table 3). The SDS sedimentation test is a small-scale test used in bread wheat breeding programs to predictgluten strength and baking quality (Carter et al., 1999). The sedi-mentation in the SDS solution theoretically results from theswelling of the glutenin strands (Eckert et al., 1993), and high

SDS sedimentation volumes have been associated with strongergluten and superior bread-making quality (Ayoub et al., 1993;Lorenzo and Kronstad, 1987).

Dough quality analysis was carried out by using the Mixolab�,which allows the characterization of the physicoechemical prop-erties of dough when submitted to dual mixing and temperatureconstraints, recording the mechanical changes due to mixing andheating, simulating the mechanical work and heat conditions thatmight be expected during the baking process (Rosell et al., 2007).There are two different parts in theMixolab� curve (Fig. 1). The firstpart is determined by dough consistency and represents theproperties of dough proteins, and the second part that involves heattreatment, is determined by dough viscosity and is closely relatedwith the starch properties. In the first stage, the parameters ob-tained from the Mixolab� were development time, stability,maximum torque at C1 and minimum torque at C2 (Fig. 1). Devel-opment time and stability are tightly related to dough strength (thestronger the dough, the longer the development time and thestability). Results reported here showed that for BW208 transgeniclines, the mean values of development time and stability weresignificantly lower than in the wild-type line, indicating weakerflours (Table 2 and Fig. 3A). Gliadins have been reported to decreasethe overall dough strength and stability when added to base flours(Branlard and Dardevet,1985; Fido et al., 2002; Khatkar et al., 2002;MacRitchie, 1992). Our results agree with this since significantnegative correlations between total gliadin content and develop-ment time (r ¼ �0.55) and stability (r ¼ �0.46) were found(Table 3). However, the g-gliadin content was positively correlatedwith development time (Table 3), and consequently couldcontribute to the flour strength. This agrees with results reported inother works, in which g-gliadins were positively correlated withdough strength (Branlard and Dardevet, 1985) and mixing toler-ance (Dong et al., 1992). However, Khatkar et al. (2002) found thatall the gliadin fractions, including the g-gliadins, produceda weakening effect on the flours. Similar results were obtained byFido et al. (1997), but with the g-gliadins showing the least weak-ening effect. Considering these observations, the g-gliadins mayhave a different role than the rest of the gliadins in the formation ofpolymers with other gluten proteins. The a- and g-gliadins have sixand eight cysteine residues, respectively; as a result three and fourintra molecular disulfide bonds are formed (Anderson et al., 2001;Müller and Wieser, 1995), whereas in the u-gliadins no cysteine ormethionine are present (Hsia and Anderson, 2001; Kasarda et al.,1983) and therefore no disulfide bonds exist. Anderson et al.(2001) reported that in the g-gliadins, after the formation of intramolecular disulfide bonds, the free-SH left might form intermolecular disulfide bonds, contributing to the glutenin polymerformation. In addition, the repetitive domain of g-gliadins is rich inb-reverse turns andmight form an extended structure, whereas thenon-repetitive domain is rich in a-helices. The feature is also pre-sented in a-gliadins, but the b-turns have an irregular distributionwhen compared with g-gliadins (Gianibelli et al., 2001), thus g-gliadins may contribute to the formation of the extensive gluten

J. Gil-Humanes et al. / Journal of Cereal Science 56 (2012) 161e170168

film network improving gas retention and dough strength. Starchcontent was positively correlated with development time.However, this may be a consequence of the high negative correla-tion (r ¼ �0.79) between protein content and starch content, andnot of a direct effect of starch on the dough strength.

Following this first stage of Mixolab� analysis (and still in thefirst part of the curve), the combination of mechanical stress andtemperature constraints produces protein destabilization andunfolding. In this phase, the parameters measured were the slopea (that represents the speed of protein weakening due to heating)and C2 (Rosell et al., 2007, 2010) (Fig. 1). The greater the decrease inconsistency (torque) at this phase, the lower the protein quality. Inour experiments, no significant differences were found in thesevalues for the transgenic lines, and only one of the lines hada significantly different value for C2 (Table 2; Fig. 3). This resultindicates that the g-gliadins have little contribution to doughconsistency. In addition, no significant differences were found forthe value of temperature at C2 (Table 2). The second part of theMixolab� curve, which involves heat treatment, represents thedough viscosity and is mainly influenced by the starch properties.This part can be divided in three phases: in the first phase (starchgelatinization), the starch granules absorb water and swell andamylose molecules leach out, producing an increase in the torque(Fig. 1). In this phase, the values of the slope b (starching speed ofthe dough) and C3weremeasured (Fig.1) (Rosell et al., 2007). In thesecond phase, the viscosity decreases due to the physical break-down of the swollen granules, determined by the amylolyticactivity. The intensity of the decrease depends on amylase activity.Minimum torque in this phase (C4) was determined by theMixolab� (Fig. 1). The third phase (starch retrogradation) is char-acterized by a decrease of the temperature that causes an increasein the viscosity due to re-crystallization of starch and gel formation,associated with amylose re-crystallization. A higher C5 value is anindication of a higher level of starch retrogradation. Transgeniclines did not show differences in the starching speed of the dough(b); however, the torque values C3, C4 and C5 presented significantdifferences for some BW208 transgenic lines and for the mean ofBW2003 transgenic lines with respect to their respective wild-typelines (Table 2; Fig. 3). The C3, C4 and C5 torque values weredecreased in both BW208 and BW2003 transgenic lines withrespect to the wild types. Nelles et al. (2000) reported lower valuesof viscosity in the pasting curve of the RVA as the starch concen-tration decreased in maize flour samples. They attributed this effectto fewer inter-granule and inter-molecule interactions that led tolower requirements of energy inputs. However, in our material,C3eC5 values were mostly correlated with the glutenin contentand the glutenins:gliadins ratio, and only C5 showed a negativecorrelation with the starch content (Table 3). It has been also re-ported that a higher lipid and protein content limits the swelling ofstarch granules, and consequently produces a lower peak viscosity(Debet and Gidley, 2006). Both lipids and proteins are known to beassociated with the surface and interior of starch granules(Baldwin, 2001). Surface proteins in wheat starches are oftendominated by endosperm storage proteins (HMW and LMW glu-tenins, and gliadins) and proteins associated with protecting thegrain from biotic and abiotic stresses (albumins and globulins)(Kasarda et al., 2008). With this information, one could hypothesizethat the lower peak torque (C3) and lower values of torque at C4and C5 obtained in the transgenic lines with reduced levels of g-gliadins could be caused by a limitation in the normal swelling ofstarch granules produced by the higher total protein content(especially glutenins) and the rearrangement of the differentprotein fractions. León et al. (2010) reported that starch propertiesdetermined by the RVA were affected in transgenic wheat linesover-expressing HMW-GS, with most transgenic lines presenting

lower values of peak viscosity, final viscosity, and breakdownviscosity. In those lines, the content of total protein, starch andamylose were not significantly affected with respect to theirrespective wild types, but presented significantly higher ratio ofHMW-GS:LMW-GS, suggesting that changes in protein composi-tion may be associated with changes in protein:starch interactionsand in turn, with the pasting properties of the transgenic wheatlines.

5. Conclusions

The down-regulation of g-gliadins by RNAi in wheat linesproduced a compensatory effect in the rest of the gluten proteins,with no statistically significant changes in the total content ofgliadins, whereas the glutenin content was increased. As a conse-quence, the total protein content was slightly increased in most ofthe transgenic lines. The suppression of g-gliadins, together withthe increase of the other fractions of gliadins and glutenins, haddifferent effects on the quality in each of the two phases thatdescribe the Mixolab curve. Thus, the mean values of BW208transgenic lines were significantly lower than the wild type fordevelopment time and stability (the first phase of the curve) butnot for the other parameters, with aweakening effect on the dough.In contrast, the mean values of BW2003 transgenic lines weresignificantly lower for C3, C4, C5, C3eC4 and C5eC4 parameters,the second phase of the curve, but not for the other parameters,indicating changes in the starch behavior. However, not all lineswere equally affected, and for many transgenic lines, the down-regulation of g-gliadins did not affect the dough properties, sug-gesting that g-gliadins are not amain determinant of dough quality,as their role on dough quality can be compensated by other glia-dins. However, they may contribute to dough strength as deter-mined in this work by the SDS sedimentation test and doughdevelopment time in the Mixolab�.

Acknowledgments

The authors thank Dr. Paul A. Lazzeri for his critical review of themanuscript. The technical assistance of Ana García is alsoacknowledged. This work was supported by the Spanish ComisiónInterministerial de Ciencia y Tecnología (AGL2010-19643-C02-02and TRA2009_0047), the European Regional Development Fund(FEDER) and Junta de Andalucía (Project P09AGR-4783). Javier Gil-Humanes acknowledges financial support from the I3P Programfrom the Agencia Estatal Consejo Superior de InvestigacionesCientíficas, which is co-financed by the European Social Fund.

References

Altpeter, F., Vasil, V., Srivastava, V., Vasil, I.K., 1996. Integration and expression of thehigh-molecular-weight glutenin subunit 1Ax1 gene into wheat. NatureBiotechnology 14, 1155e1159.

Anderson, O.D., Hsia, C.C., Torres, V., 2001. The wheat g-gliadin genes: character-ization of ten new sequences and further understanding of g-gliadin genefamily structure. Theoretical and Applied Genetics 103, 323e330.

Ayoub, M., Fregeaureid, J., Smith, D.L., 1993. Evaluation of the SDS-sedimentationtest for the assessment of eastern Canadian bread wheat quality. CanadianJournal of Plant Science 73, 995e999.

Baldwin, P.M., 2001. Starch granule-associated proteins and polypeptides: a review.Starch-Starke 53, 475e503.

Banu, I., Vasilean, I., Aprodu, I., 2010. Evaluation of rheological behaviour of wholerye and buckwheat blends with whole wheat flour using Mixolab. ItalianJournal of Food Science 22, 83e89.

Barro, F., Barcelo, P., Lazzeri, P.A., Shewry, P.R., Martin, A., Ballesteros, J., 2003.Functional properties and agronomic performance of transgenic tritordeumexpressing high molecular weight glutenin subunit genes 1Ax1 and 1Dx5.Journal of Cereal Science 37, 65e70.

Barro, F., Rooke, L., Bekes, F., Gras, P., Tatham, A.S., Fido, R., Lazzeri, P.A., Shewry, P.R.,Barcelo, P., 1997. Transformation of wheat with high molecular weight subunit

J. Gil-Humanes et al. / Journal of Cereal Science 56 (2012) 161e170 169

genes results in improved functional properties. Nature Biotechnology 15,1295e1299.

Becker, D., Folck, A., Knies, P., Lörz, H., Wieser, H., 2006. Silencing the a-gliadins inhexaploid bread wheat. In: Lookhart, G.L., Ng, P.K.W. (Eds.), Gluten Proteins.AACC, St Paul, MN, pp. 86e89.

Blechl, A., Lin, J., Nguyen, S., Chan, R., Anderson, O.D., Dupont, F.M., 2007. Transgenicwheats with elevated levels of Dx5 and/or Dy10 high-molecular-weight glu-tenin subunits yield doughs with increased mixing strength and tolerance.Journal of Cereal Science 45, 172e183.

Branlard, G., Dardevet, M., 1985. Diversity of grain proteins and bread wheat quality.I. Correlation between gliadin bands and flour quality characteristics. Journal ofCereal Science 3, 329e343.

Carter, B.P., Morris, C.F., Anderson, J.A., 1999. Optimizing the SDS sedimentation testfor end-use quality selection in a soft white and club wheat breeding program.Cereal Chemistry 76, 907e911.

Cato, L., Mills, C., 2008. Evaluation of the MixoLab for assessment of flour quality.Food Australia 60, 577e581.

Codina, G.G., Mironeasa, S., Bordei, D., Leahu, A., 2010. Mixolab versus Alveographand Falling Number. Czech Journal of Food Sciences 28, 185e191.

Collar, C., Bollain, C., Rosell, C.M., 2007. Rheological behaviour of formulated breaddoughs during mixing and heating. Food Science and Technology International13, 99e107.

Debet, M.R., Gidley, M.J., 2006. Three classes of starch granule swelling: influence ofsurface proteins and lipids. Carbohydrate Polymers 64, 452e465.

Dong, H., Sears, R.G., Cox, T.S., Hoseney, R.C., Lookhart, G.L., Shogren, M.D., 1992.Relationships between protein composition and mixograph and loaf charac-teristics in wheat. Cereal Chemistry 69, 132e136.

Eckert, B., Amend, T., Belitz, H.D., 1993. The course of the SDS and zeleny sedi-mentation tests for gluten quality and related phenomena studied using thelight-microscope. Zeitschrift Fur Lebensmittel-Untersuchung Und-Forschung196, 122e125.

Fido, R., Tatham, A., Schofield, J., Khatkar, B., 2002. Functional properties of wheatgliadins. I. Effects on mixing characteristics and bread making quality. Journal ofCereal Science 35, 299e306.

Fido, R.J., Bekes, F., Gras, P.W., Tatham, A.S., 1997. Effects of a-, b-, g- and u-gliadinson the dough mixing properties of wheat flour. Journal of Cereal Science 26,271e277.

Gianibelli, M.C., Larroque, O.R., MacRitchie, F., Wrigley, C.W., 2001. Biochemical,genetic, and molecular characterization of wheat glutenin and its componentsubunits. Cereal Chemistry 78, 635e646.

Gil-Humanes, J., Piston, F., Hernando, A., Álvarez, J.B., Shewry, P.R., Barro, F., 2008.Silencing of g-gliadins by RNA interference (RNAi) in bread wheat. Journal ofCereal Science 48, 565e568.

Gil-Humanes, J., Pistón, F., Tollefsen, S., Sollid, L.M., Barro, F., 2010. Effective shut-down in the expression of celiac disease-related wheat gliadin T-cell epitopesby RNA interference. Proceedings of the National Academy of Sciences of theUnited States of America 107, 17023e17028.

Hansen, M., Lange, M., Friis, C., Dionisio, G., Holm, P.B., Vincze, E., 2007. Antisense-mediated suppression of C-hordein biosynthesis in the barley grain results incorrelated changes in the transcriptome, protein profile, and amino acidcomposition. Journal of Experimental Botany 58, 3987e3995.

Hoseney, R.C., Finney, K.F., Shogren, M.D., Pomeranz, Y., 1969. Functional (breadmaking) and biochemical properties of wheat flour components. II. Role ofwater-solubles. Cereal Chemistry 46, 117e125.

Hsia, C.C., Anderson, O.D., 2001. Isolation and characterization of wheat u-gliadingenes. Theoretical and Applied Genetics 103, 37e44.

Huang, W.N., Li, L.L., Wang, F., Wan, J.J., Tilley, M., Ren, C.Z., Wu, S.Q., 2010. Effects oftransglutaminase on the rheological and Mixolab thermomechanical charac-teristics of oat dough. Food Chemistry 121, 934e939.

ICC, 1994a. Determination of Crude Protein in Cereals and Cereal Products for Foodand for Feed. International Association for Cereal Science and Technology.Method No. 105/2.

ICC, 1994b. Determination of Starch Content by Hydrochloric Acid Dissolution.International Association for Cereal Science and Technology. Method No. 123/1.

Ihaka, R., Gentleman, R., 1996. R: a language for data analysis and graphics. Journalof Computational and Graphical Statistics 5, 299e314.

Kahraman, K., Sakiyan, O., Ozturk, S., Koksel, H., Sumnu, G., Dubat, A., 2008. Utili-zation of Mixolab� to predict the suitability of flours in terms of cake quality.European Food Research and Technology 227, 565e570.

Kasarda, D.D., Autran, J.C., Lew, E.J.L., Nimmo, C.C., Shewry, P.R., 1983. N-terminalamino-acid-sequences of u-gliadins and u-secalins: implications for theevolution of prolamin genes. Biochimica et Biophysica Acta 747, 138e150.

Kasarda, D.D., Dupont, F.M., Vensel, W.H., Altenbach, S.B., Lopez, R., Tanaka, C.K.,Hurkman, W.J., 2008. Surface-associated proteins of wheat starch granules:suitability of wheat starch for celiac patients. Journal of Agricultural and FoodChemistry 56, 10292e10302.

Khan, K., Hamada, A., Patek, J., 1985. Polyacrylamide-gel electrophoresis for wheatvariety identification: rffect of variables on gel properties. Cereal Chemistry 62,310e313.

Khatkar, B.S., Fido, R.J., Tatham, A.S., Schofield, J.D., 2002. Functional properties ofwheat gliadins. I. Effects on mixing characteristics and bread making quality.Journal of Cereal Science 35, 299e306.

Koksel, H., Kahraman, K., Sanal, T., Ozay, D.S., Dubat, A., 2009. Potential utilization ofMixolab for quality evaluation of bread wheat genotypes. Cereal Chemistry 86,522e526.

Lange, M., Vincze, E., Wieser, H., Schjoerring, J.K., Holm, P.B., 2007. Suppression of C-hordein synthesis in barley by antisense constructs results in a more balancedamino acid composition. Journal of Agricultural and Food Chemistry 55,6074e6081.

Lawrence, R.J., Pikaard, C.S., 2003. Transgene-induced RNA interference: a strategyfor overcoming gene redundancy in polyploids to generate loss-of-functionmutations. Plant Journal 36, 114e121.

León, E., Marin, S., Gimenez, M.J., Piston, F., Rodriguez-Quijano, M., Shewry, P.,Barro, F., 2009. Mixing properties and dough functionality of transgenic lines ofa commercial wheat cultivar expressing the 1Ax1, 1Dx5 and 1Dy10 HMWglutenin subunit genes. Journal of Cereal Science 49, 148e156.

León, E., Pistón, F., Aouni, R., Shewry, P.R., Rosell, C.M., Martín, A., Barro, F., 2010.Pasting properties of transgenic lines of a commercial bread wheat expressingcombinations of HMW glutenin subunit genes. Journal of Cereal Science 51,344e349.

Lorenzo, A., Kronstad, W.E., 1987. Reliability of two laboratory techniques to predictbread wheat-protein quality in nontraditional growing areas. Crop Science 27,247e252.

MacRitchie, F., 1987. Evaluation of contributions from wheat-protein fractions todough mixing and bread making. Journal of Cereal Science 6, 259e268.

MacRitchie, F., 1992. Physicochemical properties of wheat proteins in relation tofunctionality. Advances in Food and Nutrition Research 36, 1e87.

Marco, C., Rosell, C.M., 2008. Breadmaking performance of protein enriched, gluten-free breads. European Food Research and Technology 227, 1205e1213.

Miki, D., Itoh, R., Shimamoto, K., 2005. RNA silencing of single and multiplemembers in a gene family of rice. Plant Physiology 138, 1903e1913.

Müller, S., Wieser, H., 1995. The location of disulphide bonds in a-type gliadins.Journal of Cereal Science 22, 21e27.

Nelles, E.M., Dewar, J., Bason, M.L., Taylor, J.R.N., 2000. Maize starch biphasic pastingcurves. Journal of Cereal Science 31, 287e294.

Nieto-Taladriz, M.T., Perretant, M.R., Rousset, M., 1994. Effect of gliadins and HMWand LMW subunits of glutenin on dough properties in the F6 recombinantinbred lines from bread wheat cross. Theoretical and Applied Genetics 88,81e88.

Ozturk, S., Kahraman, K., Tiftik, B., Koksel, H., 2008. Predicting the cookie quality offlours by using Mixolab�. European Food Research and Technology 227,1549e1554.

Payne, P.I., 1987. Genetics of wheat storage proteins and the effect of allelic variationon bread-making quality. Annual Review of Plant Physiology and PlantMolecular Biology 38, 141e153.

Payne, P.I., Seekings, J.A., Worland, A.J., Jarvis, M.G., Holt, L.M., 1987. Allelic variationof glutenin subunits and gliadins and its effect on breadmaking quality inwheat: analysis of F5 progeny from ‘Chinese spring’ x ‘Chinese spring’ (Hope-1A). Journal of Cereal Science 6, 103e118.

Pistón, F., León, E., Lazzeri, P., Barro, F., 2008. Isolation of two storage proteinpromoters from Hordeum chilense and characterization of their expressionpatterns in transgenic wheat. Euphytica 162, 371e379.

Pistón, F., Marín, S., Hernando, A., Barro, F., 2009. Analysis of the activity of a g-gliadin promoter in transgenic wheat and characterization of gliadin synthesisin wheat by MALDI-TOF during grain development. Molecular Breeding 23,655e667.

Rosell, C.M., Collar, C., Haros, M., 2007. Assessment of hydrocolloid effects on thethermo-mechanical properties of wheat using the Mixolab. Food Hydrocolloids21, 452e462.

Rosell, C.M., Cortez, G., Repo-Carrasco, R., 2009. Bread making use of Andean cropsquinoa, kaniwa, kiwicha, and tarwi. Cereal Chemistry 86, 386e392.

Rosell, C.M., Santos, E., Collar, C., 2010. Physical characterization of fiber-enrichedbread doughs by dual mixing and temperature constraint using the Mixolab�.European Food Research and Technology 231, 535e544.

Shewry, P.R., 2009. Wheat. Journal of Experimental Botany 60, 1537e1553.Shewry, P.R., D’Ovidio, R., Lafiandra, D., Jenkins, J.A., Mills, E.N.C., Bekes, F., 2009.

Wheat Grain Proteins, fourth ed.. AACC International Inc., St. Paul, MN.Shewry, P.R., Halford, N.G., 2003. Genetics of Wheat Gluten Proteins Advances in

Genetics, vol 49. Academic Press Inc, San Diego, pp. 111e184.Shewry, P.R., Halford, N.G., Tatham, A.S., 1989. The high molecular weight subunits

of wheat, barley and rye: genetics, molecular biology, chemistry and role inwheat gluten structure and functionality. Oxford Surveys of Plant Molecularand Cell Biology 6, 163e219.

Shewry, P.R., Tatham, A.S., Fido, R.J., 1995. Separation of plant proteins by elec-trophoresis. In: Jones, H. (Ed.), Methods in Molecular Biology: Plant GeneTransfer and Expression Protocols. Humana Press, Totowa, New Jersey,pp. 399e422.

Torbica, A., Hadnadev, M., Dapcevic, T., 2010. Rheological, textural and sensoryproperties of gluten-free bread formulations based on rice and buckwheat flour.Food Hydrocolloids 24, 626e632.

Travella, S., Klimm, T.E., Keller, B., 2006. RNA interference-based gene silencing asan efficient tool for functional genomics in hexaploid bread wheat. PlantPhysiology 142, 6e20.

van den Broeck, H.C., Gilissen, L., Smulders, M.J.M., van der Meer, I.M., Hamer, R.J.,2011. Dough quality of bread wheat lacking a-gliadins with celiac diseaseepitopes and addition of celiac-safe avenins to improve dough quality. Journalof Cereal Science 53 (2), 206e216.

van den Broeck, H.C., van Herpen, T., Schuit, C., Salentijn, E.M.J., Dekking, L.,Bosch, D., Hamer, R.J., Smulders, M.J.M., Gilissen, L., van der Meer, I.M., 2009.Removing celiac disease-related gluten proteins from bread wheat while

J. Gil-Humanes et al. / Journal of Cereal Science 56 (2012) 161e170170

retaining technological properties: a study with Chinese Spring deletion lines.BMC Plant Biology 9, 12.

van Lonkhuijsen, H.J., Hamer, R.J., Schreuder, C., 1992. Influence of specificgliadins on the breadmaking quality of wheat. Cereal Chemistry 69,174e177.

Weegels, P.L., Marseille, J.P., Bosveld, P., Hamer, R.J., 1994. Large-scale separation ofgliadins and their breadmaking quality. Journal of Cereal Science 20,253e264.

Wieser, H., Antes, S., Seilmeier, W., 1998. Quantitative determination of glutenprotein types in wheat flour by reversed-phase high-performance liquidchromatography. Cereal Chemistry 75, 644e650.

Wieser, H., Koehler, P., Folck, A., Becker, D., 2006. Characterization of wheat withstrongly reduced a-gliadin content. In: Lookhart, G.L., Ng, P.K.W. (Eds.), GlutenProteins. AACC, St Paul, MN, pp. 13e16.

Williams, R., El-Haramein, F.J., Nakkoul, H., Rihawi, S., 1988. Crop Quality Evaluation.Methods and Guidelines. ICARDA, Aleppo. Syria. Technical Manual n� 4.