A Customized Gene Expression Microarray Reveals That the Brittle Stem Phenotype fs2 of Barley Is Attributable to a Retroelement in the HvCesA4 Cellulose Synthase Gene

A Customized Gene Expression Microarray Reveals That the Brittle Stem Phenotype fs2 of Barley Is Attributable to a Retroelement in the HvCesA4 Cellulose Synthase Gene
of 13
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  A Customized Gene Expression Microarray RevealsThat the Brittle Stem Phenotype  fs2  of Barley IsAttributable to a Retroelement in the  HvCesA4  CelluloseSynthase Gene 1[W][OA] Rachel A. Burton 2  , Gang Ma 2  , Ute Baumann, Andrew J. Harvey, Neil J. Shirley, Jillian Taylor,Filomena Pettolino, Antony Bacic, Mary Beatty, Carl R. Simmons, Kanwarpal S. Dhugga, J. Antoni Rafalski,Scott V. Tingey, and Geoffrey B. Fincher* Australian Centre for Plant Functional Genomics, School of Agriculture, Food, and Wine, University of Adelaide, Waite Campus, Glen Osmond, South Australia 5064, Australia (R.A.B., G.M., U.B., A.J.H., N.J.S., J.T., G.B.F.); Australian Centre for Plant Functional Genomics, School of Botany, University of Melbourne,Parkville, Victoria 3010, Australia (F.P., A.B.); Bioinformatics (M.B., C.R.S.) and Genetic (K.S.D.) DiscoveryGroups, DuPont Agricultural Biotechnology, Pioneer Hi-Bred International, Johnston, Iowa 50131–1004; andGenetic Discovery Group, DuPont Agricultural Biotechnology, Wilmington, Delaware 19880 (J.A.R., S.V.T.) The barley (  Hordeum vulgare ) brittle stem mutants,  fs2 , designated X054 and M245, have reduced levels of crystalline cellulosecompared with their parental lines Ohichi and Shiroseto. A custom-designed microarray, based on long oligonucleotidetechnology and including genes involved in cell wall metabolism, revealed that transcript levels of very few genes were alteredin the elongation zone of stem internodes, but these included a marked decrease in mRNA for the  HvCesA4  cellulose synthasegene of both mutants. In contrast, the abundance of several hundred transcripts changed in the upper, maturation zones of stem internodes, which presumably reflected pleiotropic responses to a weakened cell wall that resulted from the primarygenetic lesion. Sequencing of the  HvCesA4  genes revealed the presence of a 964-bp solo long terminal repeat of a  Copia -likeretroelement in the first intron of the  HvCesA4  genes of both mutant lines. The retroelement appears to interfere withtranscription of the  HvCesA4  gene or with processing of the mRNA, and this is likely to account for the lower crystallinecellulose content and lower stem strength of the mutants. The  HvCesA4  gene maps to a position on chromosome 1H of barleythat coincides with the previously reported position of   fs2 . The mechanical strength of stems of cereal cropspecies is an important agronomic trait, particularly just prior to harvest, when the stems are laden withdeveloping grain. Mechanical failure of stems at thisstage, which might be initiated by environmentalconditions such as high winds or heavy rain, canlead to substantial losses of yield. While crop specieswith brittle stems are undesirable in this context, brittle stem mutants are attracting attention in emerg-ing lignocellulosic bioethanol technologies, insofar asthey are likely to require lower energy input duringgrinding of the crop residues prior to the conversionsteps of bioethanol production. As a result, there has been considerable interest in identifying factors thatinfluence stem strength and how these factors might be manipulated, either through traditional breeding oremerging genetic engineering approaches, to generatecereal varieties with altered mechanical strength.Stems of plants are variously referred to as stems,stalks, or culms, depending on the species. Here, wewill use stems as a general term to cover each of thesepossibilities.Brittle stem mutants of rice ( Oryza sativa ), barley(  Hordeum vulgare ), and maize ( Zea mays ) have becomeimportant tools for the identification of factors thatcontribute to stem strength (Kokubo et al., 1989; Liet al., 2003; Ching et al., 2006). In some mutant lines,reduced stem strength has been correlated with de-creased cellulose content of the stems (Kokubo et al.,1991; Appenzeller et al., 2004) and with greatly re-duced numbers of terminal rosette cellulose synthasecomplexes in the plasma membrane (Kimura et al.,1999). The content of lignin, pectic polysaccharides,arabinoxylans, and (1,3;1,4)-  b - D -glucans in stems of the barley brittle culm mutants remain largely un-changed (Kokubo et al., 1991). In a few cases, the gene 1 This work was supported by the Australian Research Council. 2 These authors contributed equally to the article.* Corresponding author; e-mail authorresponsible fordistribution of materialsintegral to thefindings presented in this article in accordance with the policydescribed in the Instructions for Authors ( is:Geoffrey B. Fincher ( [W] The online version of this article contains Web-only data. [OA] Open Access articles can be viewed online without a  Plant Physiology  ,  August 2010, Vol. 153, pp. 1716–1728,    2010 American Society of Plant Biologists  that carries a lesion associated with reduced strengthandcellulosecontenthasbeenidentified. Forexample,the  Brittle Stalk2  ( bk2 ) gene of maize and the  BrittleCulm1  ( bc1 ) gene of rice encode glycosyl phosphati-dylinositol-anchored Cobra-like proteins (Li et al.,2003; Ching et al., 2006; Brady et al., 2007; Sindhuetal.,2007).Althoughtheprecisebiologicalfunctionof Cobra-like proteins has not been defined, perturba-tionsofthegene resultinreducedlevelsandabnormalorientation of cellulose microfibrils and abnormalanisotropic cell expansion (Schindelman et al., 2001;Roudier et al., 2005). In another example, the  bc7  phe-notypeinricemutantsgeneratedby 60 Co g  -irradiationis believed to result from an aberrant cellulose synthase( CesA ) gene (Yan et al., 2007). Overall, a large numberof different genes, possibly more than 20, have beenimplicated in various brittle stem mutants or in mu-tants with reduced cellulose content (SupplementalTable S1), although in most cases the gene muta-tions have not been identified unequivocally or char-acterized. Identification of genes involved in brittlestem mutants in commercially important membersof the Triticeae has been further hampered becauseno genome sequence has been available for thesespecies.Here, we have examined two brittle stem mutantsof barley, for which the “brittle” loci were srcinallydesignated “fragile stem” (  fs ). The single-gene brittlestem mutants are spontaneous mutants derived fromthe parental barley landraces Ohichi (line J755) andShiroseto (line J156). The corresponding mutant lines, both of which carried mutations at the  fs2  locus, aredesignated X054 and M245, respectively, and have adifferent genetic background (Kimura et al., 1999). Thegrowth of the internodes of barley stems follows thesame pattern as in other cereals, where dividing andelongating cells in each sequential internode push theapical meristem upward (Evert, 2006). The number of divisions and the extent of elongation together dictatethe final length of the internode in the stage of primarygrowth, where cellwalls are thin and plastic.Oncecellshave ceased elongating, secondary thickening of thewalls may occur. During secondary thickening, thewalls acquire additional layers of material composedmainly of cellulose, heteroxylans, and lignin. This isseen particularly in specialized cells of the vascular bundles and in some cells of ground tissue, includingolder collenchyma and sclerenchyma fibers (Evert,2006). Accordingly, there is a continuum of develop-mentalstagesalongthelengthofeachinternode,wherethe younger cells at the base, or elongation zone, havethe thinnest walls and may be still elongating and evendividing. In contrast, older cells in the upper, matura-tion zone of the internode may already be undergoingsecondary thickening.Our overall experimental approach was to usemicroarray techniques to identify genes for whichtranscript abundance in stems of the mutant linesdiffered significantly from levels in the parental lines.Although a barley gene microarray was commerciallyavailable, it was not used here because examination of its constituent genes revealed that many genes fromthe large  CesA  and cellulose synthase-like ( Csl ) genefamilies of barley (Hazen et al., 2002; Fincher, 2009)were not represented. Therefore, a custom-designed“barley cell wall microarray” was generated, based onlong oligonucleotide (60-bp) technologies. Gene se-quences were assembled from various publicly avail-able EST and other databases and from our ownunpublished sequences of barley  HvCesA  and  HvCsl genes. Approximately 1,400 genes encoding sugarnucleotide-interconverting enzymes, polysaccharidesynthases, glycosyl transferases, expansins, cell wallstructural proteins, polysaccharide hydrolases andtransferases, and enzymes involved in lignin biosyn-thesis and degradation were selected from the data- bases.Special attention was also directed to the samplingof stem tissues for microarray analysis. Thus, compar-isons of parental and mutant lines in the younger cellsof the basal, elongation regions of the internode wereundertaken in attempts to identify the primary geneticlesion. Analyses of the older cells in the upper, mat-uration zones of the internodes were expected toreveal pleiotropic responses that occur as the planttries to compensate for reduced stem strength.The microarray data pointed to a relatively smallnumberof genes that weredifferentially transcribed inthe lower, elongation regions of internodes of themutant lines, while transcript levels of many geneschanged in the more mature, upper regions of theinternode. Following confirmation of changes in tran-script levels by real-time quantitative (Q)-PCR, a  CesA gene that mapped to the same locus as the  fs2  gene on barley chromosome 1H was sequenced. A lesion in thegene that is likely to result in the lower crystallinecellulose content and stem strength of the mutant lineswas characterized. RESULTSStem Strength and Crystalline Cellulose Content AreGreatly Reduced in the Brittle Culm Mutant Lines To confirm and quantitate the differences in tissuestrength srcinally reported for the barley brittle stemmutants(Kokuboetal.,1989,1991),thetensilestrengthofdryleaves andtheflexuralstrengthoffreshstemsof the parental and mutant lines were measured. In allcases, the strength parameters of the mutant lines,including the three-point flexural strength of the sec-ond and fourth internodes and the “load-to-break” of dry leaves, were significantly lower than those of thecorresponding parental lines (Fig. 1, A and B). Thecrystalline cellulose contents of stems of the twomutant lines, as measured by the acetic acid-nitricacid method of Updegraff (1969), were also reduced to between approximately 60% and 40% of the corre-sponding parental line (Fig. 1C). The approximate Silenced  HvCesA4  in Barley  fs2  MutantsPlant Physiol. Vol. 153, 2010 1717  locations of tissues sampled for the subsequent micro-array analyses are shown in Figure 1D.Despite the lower levels of crystalline celluloseobserved in stems of the mutant lines (Fig. 1C), acomplete linkage analysis of cell wall polysaccharidesfrom leaf and leaf sheath material revealed only minordifferences between the mutants and their parentallines (Supplemental Table S2). In particular, levels of the 4-Glc  p  derivative, which are indicative of (1,4)-linked glucosyl residues, were slightly higher in wallsof the mutant lines (Supplemental Table S2). Levels of the other major partially methylated alditol acetatederivatives were also very similar. When we estimatedthe polysaccharide composition of the walls, based onthe structures of well-characterized wall polysaccha-rides from barley (Gibeaut et al., 2005), we could findno evidence for significant changes in other polysac-charides that contain 4-Glc  p  residues, namely the(1,3;1,4)-  b - D -glucans or xyloglucans (Supplemental Ta- ble S3).These data for wall composition in leaves andcrystalline cellulose content in stems suggest that thewalls of the barley brittle stem mutant lines containsimilar levels of “cellulose,” defined as (1,4)-  b - D -glu-can, but that the crystallinity or possibly the molecularsize of the cellulose is reduced. Given the data of Kokubo et al. (1991), who showed that the degree of polymerization of cellulose was similar between the brittle  fs2  lines and corresponding parental lines, it ismore likely that the analytical data presented abovereflect reduced crystallinity of cellulose in the mutantlines. Microarray Normalization Procedures Were Evaluated The barley cell wall microarray chip falls into thecategory of “boutique array,” because the majority of the features are long oligonucleotides, derived fromgenes orESTs,which correspond to fragments ofgenesinvolved in one particular function, namely cell wall biology. These oligonucleotide features are referredto here as their corresponding genes. As stated by Figure 1.  Comparisons of strength and cellulose content of barleybrittle culm mutants and their parental lines. A, Maximum “load-to-bend” of the second and fourth internodes of the barley lines Ohichi(J755) and Shiroseto (J156) compared with the mutant lines X054 andM245, respectively. In both mutant lines and in both internodes, thestrength of the stems of the mutant lines is significantly lower than thatof the parental lines. B, Maximum load-to-break of dry leaves fromthe barley lines Ohichi (J755) and Shiroseto (J156) compared with themutant lines X054 and M245, respectively. Again, the strength of thetissues from the mutant lines is lower than that of the parental lines. C,Crystallinecellulosedetermination(Updegraff,1969)ofstemcellwallsfrom brittle stem mutants, where E = lower, elongation zone, M =upper, maturation zone, and T = transition zone. Both mutant linesshowed reduced cellulose levels relative to their corresponding paren-tal line. For each line, the lower, elongation zone showed the lowestcellulose content of the three zones tested. D, Fourth internode from abarley stem, showing the approximate positions of tissue sampling. Burton et al.1718 Plant Physiol. Vol. 153, 2010  Oshlack et al. (2007), “there is as yet no widely ac-cepted standardmethod for normalization ofboutiquearrays” (p. R2.2). Furthermore, most of the investiga-tions of normalization methods for boutique arrayshave focused on two-colorcDNAmicroarrays,not longoligonucleotide arrays (Wilson et al., 2003; Oshlacketal.,2007),andmaynotbeapplicabletotheone-color,long oligonucleotide array used here. Therefore, threenormalization methods were evaluated to accountfor systematic errors in the data, including the cyclicLoess normalization method (Ballman et al., 2004),75th percentile scaling, and normalization to the 44housekeeping genes present on the array. Loess nor-malization assumes that the majority of the genes arenotdifferentiallyexpressed(Dudoitetal.,2002),andina study on two-color microarrays, Oshlack et al. (2007)showed that Loess normalization is only robust if lessthan 20% of genes show asymmetric differential ex-pression.It appeared from the MA plots of the barley cell wallmicroarray used here that the number of potentiallydifferentially expressed genes was too large to justifynormalizing the data using the fast linear Loessprocedure or 75th percentile scaling. Normalization,therefore, was performed using the series of controlgenes. Microarray Analyses Reveal DifferentiallyTranscribed Genes The results of the statistical analysis of the micro-array data showed that transcription levels of up toseveral hundred genes differed between the mutantlines and their near parental lines in the upper, mat-urationzonesofthesteminternode(TableI).However,only two features on the microarray were stronglydown-regulated in the lower, elongating zones of theinternodes in the X054 and the M245 mutants (Table I).The two features, in fact, were both  HvCesA4 , whichhad been monitored on the microarray with twoseparate oligonucleotides. The log 2  fold changes of the  HvCesA4  transcripts in the mutant lines are shownin Table II, where it is apparent that transcript abun-dance is approximately 16-fold lower in the mutantlines. Also shown in Table II are the up-regulatedgenes in the lower, elongation zone, namely a Sasandaretrotransposon of the type I  Copia  group and genesencoding a putative aquaporin/silicon transporterand a putative ascorbate oxidase. In the case of the Copia  retrotransposon, this means that the transcriptabundance in the mutant lines is between 48-foldand 630-fold higher than in the parental lines. The se-quence of the oligonucleotide probe of the type I Copia  retrotransposon corresponded to the  gag  region,which encodes proteins that form the nucleocapsidcore and protease components of the retrotransposon(Grandbastien, 1998).Transcript abundance of differentially transcribedgenes in the upper, maturation zones of the barleyinternodes are shown in Table III,wheresimilar trendsare observed for the  HvCesA4  gene and the retrotrans-poson. However, transcripts for a larger number of othergenes arealso alteredin the mutant lines in theseupper regions of the internode.When the levels of   HvCesA4  transcripts were com-pared in the elongation and maturation zones of theparental lines, the  HvCesA4  mRNA abundance wasgenerally slightly higher in the maturation zone, con-sistent with the synthesis of secondary cell walls inthat zone (data not shown).Forexample, ininternodesfrom the Ohichi parent, signal intensities of 14.12 and12.38 (log 2 ) were observed for the upper, maturationzone and the lower, elongation zone, respectively. Q-PCR Confirms the Lower Levels of  HvCesA4 Transcripts in the Mutant Lines When transcript levels of the  HvCesA4  gene werecompared quantitatively by Q-PCR in stems of theparentallinesandthemutants,10-to12-folddecreaseswereobservedinthemutantlines(Fig.2A).Thebarley  HvCesA4  gene was shown by Burton et al. (2004) to becoexpressedwiththe  HvCesA7 and  HvCesA8 genesina broad range of tissues. However, the lower transcriptlevels of the  HvCesA4  gene in the mutant lines werenot matched by lower levels of   HvCesA7  and  HvCesA8 gene transcripts, which were slightly up-regulated inthe mutant lines (Fig. 2B). A Retroelement Has Been Inserted into the  HvCesA4 Genes of the Mutants The full genomic sequence of the  HvCesA4  gene wasobtained from the Morex bacterial artificial chromo-some P453O19 clone and used as the basis for primerdesign during the analysis of the mutant lines. Forsequence analysis, several attempts were made toamplify full-length  HvCesA4  genes from both theparent and mutant genomic DNA using primers Table I.  Differentially transcribed genes (P   #  0.05, FDR adjusted) in internodes from barley stems  Genotypes (Wild Type - Mutant) Position inInternodeTotal No. of Differentially Expressed GenesDown-Regulated inMutant Ohichi (J755) - X054 Upper zone 466 212Lower zone 4 2Shiroseto (J156) - M245 Upper zone 555 176Lower zone 4 2 Silenced  HvCesA4  in Barley  fs2  MutantsPlant Physiol. Vol. 153, 2010 1719  5UFN and 3UR (Table IV). In each attempt, full-lengthfragments were generated from the parent lines butnever from the mutant genomic DNA preparations. Toovercome this, the genes were amplified in two largeoverlapping fragments from the mutant genomicDNAs, using the primer sets 4MIDF with 3UR or4REV4 with 5UFN (Table IV). This approach yieldedthe 3 #  sections of the mutant genes, but not the 5 # fragments, and therefore indicated that the problempreventing amplification lay in the 5 #  region of thegenes.Aseriesofprimersflankingeachofthefirstthreeintronsinthe5 # regionofthegeneweredesigned(EX1Fto EX4R; Table IV), and through systematic PCR acrossthese regions, it became apparent that the region of thegenethatwasresistanttoamplificationwasclosetoandwithin the first intron of the  HvCesA4  gene in both theX054 and M245 mutant lines.A nested PCR approach across the first intron in thepresence of dimethyl sulfoxide eventually yielded aproduct that was larger than the fragment predictedfrom the parental sequence. Sequencing of this prod-uct revealed a 964-bp insertion in the first intron; theintron was 125 bp in length in the parental lines (Fig.3). The insertion starts 10 bp inside the 5 #  border of theintron and consisted of a 957-bp region that showed76% sequence identity with the long terminal repeat(LTR)ofa Copia -likeretrotransposonknownasSasanda(Triticeae Repeat Sequence database, GrainGenes; A 7-bprepeat of the upstream 9-bp intron sequence flankingthe 5 #  end of the insertion was found at the 3 #  end of the insert. The sequences and the positions of theinsert were identical in the  HvCesA4  genes from bothX054 and M245.At the mRNA transcript level, PCR amplificationacross this region of cDNAs synthesized from the two Table II.  Differentially transcribed genes in the lower, elongation zones of internodes of stems from parental lines and the fs2 brittle stem mutants (P  #  0.05, FDR adjusted)  Functional AnnotationFold Change(log 2 ) a (Parent – Mutant) P  Shiroseto (J156) versus mutant M245 Ty1-Copia  retrotransposon  2 5.84 1.02E-07 HvCesA4  (oligonucleotide 1) 4.28 1.79E-05 HvCesA4  (oligonucleotide 2) 4.55 2.57E-05Putative  L -ascorbate oxidase  2 3.00 0.03Ohichi (J755) versus mutant X054 Ty1-Copia  retrotransposon  2 6.27 2.31E HvCesA4  (oligonucleotide 1) 3.89 9.45E HvCesA4  (oligonucleotide 2) 4.06 0.00017Putative aquaporin/silicon transporter  2 6.97 0.00017 a A negative number here indicates up-regulation, while a positivenumber indicates down-regulation. Table III.  Selected differentially transcribed genes in the upper zones of internodes from stems of Ohichi (J755) and mutant X054 (P   #  0.05, FDR adjusted)  The 12 genes with the highest absolute fold change are listed here. Functional Annotation Fold Change (log 2 )(J755 – X054)  P  Down-regulated in mutantPutative cinnamyl alcohol dehydrogenase 2.96 0.0159Tonoplast membrane integral protein  ZmTIP4-2  3.17 0.0262Solute carrier family 2, Glc transporter 3.26 0.0078Putative receptor protein kinase  CRINKLY4  3.38 0.0004Suc phosphate synthase 3.51 5.42E-05Cold-regulated protein, complete 3.72 0.0139Putative peroxidase 3.95 0.0175Putative cellulose synthase catalytic subunit 4.23 6.81E-06Putative galactoside 2- a - L -fucosyltransferase 4.64 0.0021Cellulose synthase catalytic subunit 10 4.66 4.00E-07Putative peroxidase 4.83 0.00063Peroxidase 6 7.72 0.00027Up-regulated in mutantPutative Pro-rich protein  2 8.46 0.019Endoxyloglucan transferase  2 7.89 1.29E-05Putative thaumatin protein  2 7.33 2.18E-05Floral organ regulator 1  2 7.06 0.0132Putative P450  2 7.03 9.00E-06Retrotransposon protein, putative,  Ty1-Copia  subclass -6.79 3.52E-10Putative peroxidase  ATP6a  2 6.79 8.98E-06Floral organ regulator 1  2 6.47 1.10E-05Expressed protein  2 6.39 3.51E-05Putative endo-1,4-  b -glucanase  2 6.32 3.25E-05Putative (1,4)-  b -mannan endohydrolase  2 6.22 2.51E-06Putative Pro-rich protein  2 6.02 0.00049 Burton et al.1720 Plant Physiol. Vol. 153, 2010
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks