UBC Faculty Research and Publications

RACK1 genes regulate plant development with unequal genetic redundancy in Arabidopsis Guo, Jianjun; Chen, Jin-Gui Oct 23, 2008

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
52383-12870_2008_Article_316.pdf [ 2.53MB ]
Metadata
JSON: 52383-1.0074637.json
JSON-LD: 52383-1.0074637-ld.json
RDF/XML (Pretty): 52383-1.0074637-rdf.xml
RDF/JSON: 52383-1.0074637-rdf.json
Turtle: 52383-1.0074637-turtle.txt
N-Triples: 52383-1.0074637-rdf-ntriples.txt
Original Record: 52383-1.0074637-source.json
Full Text
52383-1.0074637-fulltext.txt
Citation
52383-1.0074637.ris

Full Text

ralssBioMed CentBMC Plant BiologyOpen AcceResearch articleRACK1 genes regulate plant development with unequal genetic redundancy in ArabidopsisJianjun Guo and Jin-Gui Chen*Address: Department of Botany, University of British Columbia, Vancouver, BC V6T 1Z4, CanadaEmail: Jianjun Guo - jimguo@interchange.ubc.ca; Jin-Gui Chen* - jingui@interchange.ubc.ca* Corresponding author    AbstractBackground: RACK1 is a versatile scaffold protein in mammals, regulating diverse developmentalprocesses. Unlike in non-plant organisms where RACK1 is encoded by a single gene, Arabidopsisgenome contains three RACK1 homologous genes, designated as RACK1A, RACK1B and RACK1C,respectively. Previous studies indicated that the loss-of-function alleles of RACK1A displayedmultiple defects in plant development. However, the functions of RACK1B and RACK1C remainelusive. Further, the relationships between three RACK1 homologous genes are unknown.Results: We isolated mutant alleles with loss-of-function mutations in RACK1B and RACK1C, andexamined the impact of these mutations on plant development. We found that unlike in RACK1A,loss-of-function mutations in RACK1B or RACK1C do not confer apparent defects in plantdevelopment, including rosette leaf production and root development. Analyses of rack1a, rack1band rack1c double and triple mutants, however, revealed that rack1b and rack1c can enhance therack1a mutant's developmental defects, and an extreme developmental defect and lethality wereobserved in rack1a rack1b rack1c triple mutant. Complementation studies indicated that RACK1Band RACK1C are in principle functionally equivalent to RACK1A. Gene expression studies indicatedthat three RACK1 genes display similar expression patterns but are expressed at different levels.Further, RACK1 genes positively regulate each other's expression.Conclusion: These results suggested that RACK1 genes are critical regulators of plantdevelopment and that RACK1 genes function in an unequally redundant manner. Both the differencein RACK1 gene expression level and the cross-regulation are likely the molecular determinants oftheir unequal genetic redundancy.BackgroundReceptor for activated C kinase 1 (RACK1) is a seven tryp-tophan-aspartic acid-domain (WD40) repeat-containingprotein, and was originally identified as an anchoring pro-tein for protein kinase C (PKC) in mammals, shuttling theactivated enzyme to different subcellular sites [1,2]. Struc-[3,4]). Increasing evidence suggests that in addition tobinding the activated PKC, mammalian RACK1 functionsas a scaffold protein by physically interacting with manyother proteins and facilitating their interactions. It hasbeen shown that RACK1 plays regulatory roles in diversedevelopmental and physiological responses, includingPublished: 23 October 2008BMC Plant Biology 2008, 8:108 doi:10.1186/1471-2229-8-108Received: 12 August 2008Accepted: 23 October 2008This article is available from: http://www.biomedcentral.com/1471-2229/8/108© 2008 Guo and Chen; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.Page 1 of 11(page number not for citation purposes)turally, RACK1 is similar to the heterotrimeric G-protein βsubunit (Gβ) which has a seven-bladed propeller structurewith one WD40 unit constituting each blade (reviewed incell cycle control, cell movement and growth, immuneresponse, and neural responses in mammals (reviewed in[3,4]). Therefore, RACK1 is now viewed as a versatile scaf-BMC Plant Biology 2008, 8:108 http://www.biomedcentral.com/1471-2229/8/108fold protein, serving as a nexus for multiple signal trans-duction pathways.Although not recognized as such, the first plant RACK1gene was cloned from tobacco BY-2 cells as an auxin (2,4-dichlorophenoxyacetic acid, 2,4-D) inducible gene, arcA[5]. Subsequently, the amino acid sequence homologuesof RACK1 were found in all plant species examined(reviewed in [6]). Earlier studies based on gene expressionand induction analysis implied that plant RACK1 mayhave a role in hormone-mediated cell division [5,7], UVand salicylic acid responses [8]. In rice, RACK1, namedRWD [9], was found to be one of the seven proteins whoseexpressions were down-regulated in d1 mutant, a loss-of-function allele of rice heterotrimeric G-protein α subunit[10]. Further, rice RACK1 protein was induced by abscisicacid (ABA) in imbibed wild-type seeds, but not in d1mutant seeds. It was proposed that RACK1 may play a rolein rice embryogenesis and germination [10]. Furthermore,recently, it has been demonstrated that RACK1 proteinsare key regulators of innate immunity by interacting withmultiple proteins in the Rac1 immune complex in rice[11]. In Arabidopsis, RACK1 proteins have been found tobe associated with the subunits of ribosomes [12,13], butno signaling proteins have been identified to interact withArabidopsis RACK1 proteins.Structurally, RACK1 proteins in plants are similar tothose in mammals, containing a seven-bladed β-pro-peller [14]. However, analysis of RACK1 proteins inplants and in non-plant organisms revealed an impor-tant feature of plant RACK1 proteins: some plants havemore than one RACK1 genes, in contrast to the singlecopy of RACK1 gene in non-plant organisms. For exam-ple, the sequenced genomes of rice (Oryza sativa) andArabidopsis (Arabidopsis thaliana) contain two andthree RACK1 homologous genes, respectively (Figure1). The three RACK1 proteins encoded by the Arabidop-sis genome were designated as RACK1A, RACK1B andRACK1C, respectively [15]. Previously, we provided evi-dence that RACK1A mediates multiple hormoneresponses and developmental processes [15]. However,the functions of the other two Arabidopsis RACK1genes, RACK1B and RACK1C, and the relationshipbetween Arabidopsis RACK1 genes remain unknown.Here we demonstrate that although RACK1B andRACK1C genes are likely dispensable, they still contrib-ute significantly to the RACK1A-regulated developmen-tal processes in Arabidopsis. We provide evidence thatthe difference in the gene expression level and thecross-regulation are likely the molecular determinantsof unequal genetic redundancy of RACK1 genes in reg-ulating plant development.Multiple amino acid sequence alignment of RACK1 in plants and in humansFigure 1Multiple amino acid sequence alignment of RACK1 in plants and in humans. The amino acid sequences were aligned by CLUSTALW multiple alignment of BioEdit Sequence Alignment Editor http://www.mbio.ncsu.edu/BioEdit/bioedit.html. Amino acids that are identical or similar are shaded with black or gray, respectively. Gaps are shown as dashed lines. The pro-teins aligned are (name of species and accession number in parentheses): RACK1A_At (Arabidopsis thaliana, NP_173248), RACK1B_At (Arabidopsis thaliana, NP_175296), RACK1C_At (Arabidopsis thaliana, NP_188441), RACK1A_Os (Oryza sativa, NP_001043910), RACK1B_Os (Oryza sativa, NP_001056254), RACK1_Pt (Populus trichocarpa, ABK92879), RACK1 _Vv (Vitis vinifera, CAN61810), and RACK1_Hs (Homo sapiens, NP_006089). The positions of GH and WD dipeptides in each WD40 repeat are indicated by triangles and asterisks, respectively, on the top of residues. The positions for WD repeat domains were Page 2 of 11(page number not for citation purposes)obtained from the SMART database http://smart.embl-heidelberg.de.BMC Plant Biology 2008, 8:108 http://www.biomedcentral.com/1471-2229/8/108ResultsT-DNA insertional mutants of RACK1B and RACK1CArabidopsis genome contains three RACK1 homologousgenes, designated as RACK1A, RACK1B and RACK1C,respectively [15]. Within the RACK1 gene family, mutantalleles for only RACK1A have been reported previously[15]. We report here the isolation and characterization ofrack1b and rack1c mutant alleles. By searching the SalkInstitute sequence-indexed insertion mutant collectionhttp://signal.salk.edu/cgi-bin/tdnaexpress, we obtainedtwo independent T-DNA insertional alleles for eachRACK1 gene. All alleles are in the Columbia (Col-0) eco-typic background. We designated the two mutant allelesfor RACK1B as rack1b-1 and rack1b-2, respectively. Inrack1b-1 allele, the T-DNA was inserted in the second exonof RACK1B gene, and in the rack1b-2 allele, the T-DNAwas inserted in the first intron (Figure 2A). RT-PCR analy-sis indicated that the full-length transcript of RACK1B wasabsent in both alleles (Figure 2B), implying that they arelikely loss-of-function alleles. Unlike rack1a mutants,rack1b mutants do not display any apparent developmen-tal defects (Figure 2C). We designated the two mutantalleles for RACK1C as rack1c-1 and rack1c-2, respectively(Figure 2D). In rack1c-1 allele, the T-DNA was inserted inthe second exon of RACK1C gene, and in the rack1c-2allele, the T-DNA was inserted in the 5'-UTR region. RT-PCR analysis indicated that the full-length transcript ofRACK1C was absent in both alleles (Figure 2E), implyingthat they are likely loss-of-function alleles. Similar torack1b mutants but unlike rack1a mutants, rack1c mutantsdo not display any apparent defects in plant development(Figure 2C).Loss-of-function mutations in RACK1B and RACK1C enhance the developmental defects in rosette leaf production of rack1a mutantPreviously, we showed that loss-of-function mutations inone member of Arabidopsis RACK1 gene family, RACK1A,resulted in multiple defects in plant development [15].Because loss-of-function alleles of RACK1B and RACK1Cdid not display apparent defects in plant development, wewanted to test if mutations in RACK1B or RACK1C canenhance the developmental defects of rack1a mutants.Therefore, we generated rack1a-1 rack1b-2 and rack1a-1rack1c-1 double mutants. One of the most dramatic phe-notypes observed in rack1a single mutants was thereduced number of rosette leaves [15]. Therefore, we grewsingle and double mutants together with wild-type (Col)under identical, short-day conditions with 10/14 h pho-toperiod, counted the number of rosette leaves in doublemutants, and compared it with Col and rack1a-1 singlemutant. We found that while rack1b-2 and rack1c-1 singlemutants produced wild-type number of rosette leaves,both rack1b-2 and rack1c-1 significantly enhanced thephenotype of reduced number of rosette leaves of rack1a-1 single mutants (Figure 3A, B). When plants were grownunder 10/14 h photoperiod for 48 days, wild-type pro-duced approximately 30 rosette leaves, whereas rack1a-1single mutant produced 22 rosette leaves. Under theseconditions, rack1a-1 rack1b-2 and rack1a-1 rack1c-1 dou-ble mutants only produced about 16 and 19 rosetteleaves, respectively (Figure 3B). The rate of rosette leafproduction was reduced approximately 27% and 14%,respectively, in rack1a-1 rack1b-2 and rack1a-1 rack1c-1double mutants, compared with rack1a-1 single mutant(Figure 3C). We also examined the rosette size by measur-ing the diameter of rosette of each genotype. Similar to thesituation of number of rosette leaves, the diameter ofrosette was significantly reduced in rack1a-1 singlemutant, compared with wild-type plants, and such reduc-tion was further enhanced in rack1a-1 rack1b-2 andrack1a-1 rack1c-1 double mutants (Figure 3D). Interest-ingly, no synergistic effect was observed between rack1b-2T-DNA insertional mutants of RACK1B and RACK1CFigure 2T-DNA insertional mutants of RACK1B and RACK1C. (A) A diagram to illustrate the T-DNA insertion sites in rack1b-1 and rack1b-2 mutants. (B) RT-PCR analysis of RACK1B transcript in rack1b mutants. RACK1B-specific prim-ers that amplify the full-length transcript of RACK1B in wild-type (Col) were used. (C) The rosette morphology of rack1b and rack1c mutants. Shown are plants grown 48 days under 10/14 h photoperiod. (D) A diagram to illustrate the T-DNA insertion sites in rack1c-1 and rack1c-2 mutants. (E) RT-PCR analysis of RACK1C transcript in rack1c mutants.RACK1C-spe-cific primers that amplify the full-length transcript of RACK1C in Col were used. Gray boxes in (A) and (D) represent cod-ing regions and white boxes represent 5'-UTR and 3'-UTR regions. The T-DNA inserts are not drawn to scale. LB, T-DNA left border. Total RNA isolated from 10 d-old, light-grown seedlings was used for RT-PCR analysis in (B) and (E). RT-PCR was performed with 30 cycles. The expression of Page 3 of 11(page number not for citation purposes)and rack1c-1 mutations. Statistically, rack1b-2 rack1c-1double mutants phenocopied parental single mutantsACTIN2 was used as a control.BMC Plant Biology 2008, 8:108 http://www.biomedcentral.com/1471-2229/8/108and displayed wild-type traits of these phenotypes (Figure3A, B).Subsequently, we generated rack1a-1 rack1b-2 rack1c-1 tri-ple mutant. Very few triple mutants could survive in soil.For those survived, they were extremely slow in growthand development, and produced fewest rosette leaves andsmallest rosette size among all genotypes examined (Fig-ure 3A–D). Not surprisingly, the rate of rosette leaf pro-duction in the triple mutant was the slowest among allgenotypes examined (Figure 3C). Because rack1a-1 rack1b-2 rack1c-1 triple mutants could not survive to maturity toproduce seeds, these triple mutants were maintained inplants homozygous for the rack1b-2 and rack1c-1 loci andwhereas rack1a-1 rack1b-2 rack1c-1 had extreme pleio-tropic phenotype, rack1a-1 rack1b-2 rack1c-1 triplemutants can be readily picked up from the segregatingprogeny of plants homozygous for the rack1b-2 andrack1c-1 loci and heterozygous for the rack1a-1 locus.Loss-of-function mutations in RACK1B and RACK1C enhance the defects in root development of rack1a mutantGenetic analysis indicated that loss-of-function mutationsin RACK1A affect the production of rosette leaves, andthat the effect of rack1a-1 mutation can be enhanced bythe rack1b-2 or rack1c-1 mutation or both (Figure 3). Wewanted to extend our analysis to non-aerial organs byexamining the impact of these mutations on root develop-ment. We measured the length of primary root andcounted the number of lateral root and used them asparameters of root development and root architecture. Wefound that the length of primary root of rack1a-1 mutantwas slightly shorter than that of wild-type whereas rack1b-2 and rack1c-1 mutants had wild-type length of primaryroot (Figure 4A). The length of primary root was furthershortened in rack1a-1 rack1b-2 and rack1a-1 rack1c-1 dou-ble mutants, compared with that in rack1a-1 singlemutant (Figure 4A), indicating that rack1b-2 and rack1c-1mutations can also enhance the effect of rack1a-1 muta-tion on primary root growth. Similar to the situation ofprimary root, rack1a-1 mutant produced fewer lateralrack1b-2 and rack1c-1 mutations enhance the rosette leaf phenotype of rack1a mutan sFigure 3rack1b-2 and rack1c-1 mutations enhance the rosette leaf phenotype of rack1a mutants. (A) The phenotype of rack1 mutants. Shown are plants grown for 48 days under 10/14 h photoperiod. Scale bars, 2 cm. (B) The number of rosette leaves of rack1 mutants. (C) The rate of rosette leaf production of rack1 mutants. The rate of rosette leaf produc-tion is expressed as the number of rosette leaves divided by the age of plants. (D) The size of rosette of rack1 mutants. The number of rosette leaves, the rate of rosette leaf pro-duction and the size of rosette were measured from plants grown for 48 d under 10/14 h photoperiod. Shown in (B) to (D) are the averages of at least four plants ± S.E. The same experiment was repeated twice with similar trends and the data from one experiment were presented. *, significant dif-ference from Col, P < 0.05. #, significant difference from rack1a single mutant, P < 0,05. **, significant difference from rack1a-1 rack1b-2 double mutant, P < 0.05.rack1b-2 and rack1c-1 mutations enhance the root phenotype of rack1a mut ntsFigure 4rack1b-2 and rack1c-1 mutations enhance the root phenotype of rack1a mutants. (A) The length of primary root of rack1 mutants. (B) The number of lateral roots of rack1 mutants. The length of primary root and the number of lateral roots were measured from 10 d-old, light-grown seedlings (under 14/10 h photoperiod). Shown are the aver-ages of at least 15 seedlings ± S.E. *, significant difference from Col, P < 0.05. #, significant difference from rack1a single mutant, P < 0.05. **, significant difference from rack1a-1 Page 4 of 11(page number not for citation purposes)heterozygous for the rack1a-1 locus. Because rack1b-2rack1c-1 double mutants had wild-type morphology rack1b-2 double mutant, P < 0.05.BMC Plant Biology 2008, 8:108 http://www.biomedcentral.com/1471-2229/8/108roots than wild-type whereas rack1b-2 and rack1c-1mutants had wild-type number of lateral roots (Figure4B). As expected, rack1b-2 and rack1c-1 mutationsenhanced the lateral root phenotype of rack1a-1 mutant(Figure 4B). Among all genotypes examined, the rack1a-1rack1b-2 rack1c-1 triple mutant produced the shortest pri-mary root and did not produce any lateral root under ourassay conditions (Figure 4A, B).Genetic complementation of rack1a mutants by overexpressing RACK1 genesGenetic analyses indicated that there is unequal geneticredundancy among three Arabidopsis RACK1 genes inregulating rosette leaf production and root development,and that RACK1A is likely a non-dispensable gene in thissmall gene family. Although RACK1B and RACK1C arelikely dispensable, they still contribute significantly to theoverall activity of RACK1 genes in regulating plant devel-opment, as revealed by the phenotypes of double and tri-ple mutants. We wanted to further explore the mechanismof the unequal genetic redundancy of RACK1 genes.Firstly, because RACK1B and RACK1C are highly similar(about 90% identity) to RACK1A at the amino acid level(Figure 1), we wanted to test if RACK1B and RACK1C arein principle functionally equivalent to RACK1A. We rea-soned that if RACK1B and RACK1C are indeed function-ally equivalent to RACK1A, one would expect thatoverexpression of RACK1B or RACK1C complements thedevelopmental defects of rack1a mutants. Therefore, wegenerated transgenic lines overexpressing RACK1B orRACK1C in the rack1a mutant background using theCaMV 35S promoter. As a control, we generated trans-genic plants overexpressing RACK1A in rack1a mutantbackground. At least two independent transgenic lineswere analyzed for each transformation. Overexpression ofthe transgene in these lines was confirmed by RT-PCRanalysis (Figure 5A). We examined the same parametersdescribed above, namely the number of rosette leaves, thelength of primary root and the number of lateral roots inthe transgenic lines overexpressing each RACK1 gene andcompared them with those in Col and rack1a singlemutants. As expected, overexpression of RACK1A fullycomplemented the mutant phenotype of rack1a mutant(Figure 5B–D). Similarly, we found that overexpression ofRACK1B or RACK1C fully restored rack1a mutant to wild-type morphology, evident by the wild-type number ofrosette leaves, wild-type length of primary root and wild-type number of lateral roots in transgenic lines (Figure5B–D).Expression of Arabidopsis RACK1 genesBecause constitutive expression of RACK1B or RACK1Ccould efficiently complement rack1a mutant's develop-RACK1A, and that the unequal genetic redundancy ofRACK1 genes is likely due to the difference in their expres-sion patterns or expression levels. Therefore, we soughtadditional evidence that would shed light on the relation-ship between RACK1 genes. We examined the expressionpatterns of RACK1A, RACK1B and RACK1C in various tis-sues and organs of young seedlings and mature plants byRT-PCR. We found that all three Arabidopsis RACK1 geneswere expressed widely in all tissues examined (Figure 6A).These results are largely consistent with the results of anal-ysis of RACK1 gene promoter:β-glucuronidase (GUS) tran-scriptional reporter lines [15]. By using RT-PCR, wenoticed that in any given tissues or organs examined, thetranscript level of three RACK1 genes were different, witha general trend of RACK1A > RACK1B > RACK1C (Figure6A).In order to quantify the difference in transcript level ofRACK1A, RACK1B and RACK1C genes, we used quantita-tive real-time PCR to more accurately compare the tran-script level of three RACK1 genes in different tissues andorgans of wild-type Col plants. We selected the samples ofshoots and roots of 4 d- and 7 d-old light-grown seedlingsand rosette leaves and roots of mature plants for quantita-tive real-time PCR analysis. We found that consistent withthe result of RT-PCR analysis, the transcript level ofRACK1C was the lowest and that of RACK1A was the high-est among three RACK1 genes, with a trend of RACK1A >RACK1B > RACK1C in all samples examined (Figure 6B).For example, the transcript level of RACK1A was about 5-fold higher than that of RACK1C in the roots of 4 d-old,light-grown seedlings (Figure 6B). In this sample, the tran-script level of RACK1B was approximately 2-fold higherthan that of RACK1C.Cross-regulation of RACK1 genes at the transcription levelThe analysis of the expression patterns and transcript levelof three RACK1 genes in various tissues and organs sup-ported the view that the unequal genetic redundancy ofRACK1 genes is likely due to the difference in the geneexpression level. However, other possibilities may alsoexist. For example, as reviewed by Briggs et al. (2006),cross-regulation is another mechanism that attributes tothe unequal genetic redundancy of some homologousgenes [16]. Because RACK1A, RACK1B and RACK1C areapproximately 90% identical to each other at the aminoacid level, we were unable to obtain antibodies that canspecifically recognize each RACK1 protein. Therefore, inthis study, we examined the impact of loss-of-functionmutations of each RACK1 gene on the transcription of theother two RACK1 genes. Further, we examined the impactof combination of loss-of-function mutations of twoPage 5 of 11(page number not for citation purposes)mental defects, these results implied that RACK1B andRACK1C are likely in principle functionally equivalent toRACK1 genes on the transcription of the other RACK1gene. Specifically, we examined the transcript level ofBMC Plant Biology 2008, 8:108 http://www.biomedcentral.com/1471-2229/8/108Page 6 of 11(page number not for citation purposes)The complementation of rack1a mutants by overexpression of RACK1 genesFigure 5The complementation of rack1a mutants by overexpression of RACK1 genes. (A) RT-PCR analysis of the expression of RACK1 genes in transgenic lines. The transgenic lines 2-7, 6-2, 8-3 and 25-3 are RACK1A overexpressors in rack1a-2 mutants. The transgenic lines 4-5 and 28-2 are RACK1B overexpressors in rack1a-1 mutants. The transgenic lines 4-3, 5-3, 8-3 and 9-6 are RACK1C overexpressors in rack1a-1 mutants. RT-PCR was performed at 28 cycles. The expression of ACTIN2 was used as a control. (B) The number of rosette leaves in transgenic plants overexpressing individual RACK1 gene in rack1a mutant back-ground. The number of rosette leaves was collected from plants grown for 37 d under 14/10 h photoperiod. Shown are the averages of number of rosette leaves from at least four plants ± S.E. (C) The length of primary root in transgenic plants over-expressing individual RACK1 gene in rack1a mutant background. The length of primary roots was measured from seedlings grown for 10 d under 14/10 h photoperiod. (D) The number of lateral roots in transgenic plants overexpressing individual RACK1 gene in rack1a mutant background. The number of lateral roots was counted from seedlings grown for 11 d under 14/10 h photoperiod. Shown in (C) and (D) are the averages of at least 20 seedlings ± S.E. *, significant difference from Col, P < 0.05.BMC Plant Biology 2008, 8:108 http://www.biomedcentral.com/1471-2229/8/108RACK1A in rack1b and rack1c single and double mutants,the transcript level of RACK1B in rack1a and rack1c singleand double mutants, and the transcript level of RACK1Cin rack1a and rack1b single and double mutants, and com-pared with their transcript levels in wild-type. For thisanalysis, we used the 4.5 d-old, light-grown whole seed-lings. By using RT-PCR, we noticed that the transcript levelof RACK1B was reduced in rack1a and rack1c single anddouble mutants (Figure 7A). Similarly, the transcript levelof RACK1C was reduced in rack1a and rack1b single anddouble mutants (Figure 7A). However, we did not observewith that in wild-type (Figure 7A). Because the transcriptlevel of RACK1A is the most abundant among threeRACK1 homologous genes and the conditions used forRT-PCR (e.g. PCR at 28 cycles) may not allow us to visual-ize any differences in RACK1A transcript level among dif-ferent samples, subsequently we used quantitative real-time PCR to more accurately compare the transcript levelof three RACK1 genes in wild-type and mutants. We foundthat the transcript level of any given RACK1 gene wasreduced in the loss-of-function alleles of each and both ofthe other two RACK1 genes (Figure 7B).DiscussionRoles of RACK1 genes in plant developmentRACK1 gene is evolutionarily conserved in diverse organ-isms. Although the research interest in RACK1 has grownexponentially since its discovery [1] and RACK1 is nowviewed as a multi-functional, versatile scaffold protein inmammals and in yeasts (reviewed in [3,4]), the functionThe expression of RACK1A, RACK1B and RACK1C genesFigure 6The expression of RACK1A, RACK1B and RACK1C genes. (A) RT-PCR analysis of the expression of RACK1 genes in various tissues and organs of young seedlings and mature plants. RT-PCR was performed at 30 cycles. The expression of ACTIN2 was used as a control. (B) Quantitative real-time PCR analysis of the transcript levels of RACK1 genes. The transcript level of each RACK1 gene was normal-ized against the transcript level of ACTIN2 in each sample. The relative transcript levels of RACK1 genes were compared to that of RACK1C in the roots of 4 d-old, light-grown seed-lings (set as 1). Shown are the averages of three replicates ± S.D.The expression of RACK1 genes in rack1a, rack1b and rack1c singl  and double mutantsFigure 7The expression of RACK1 genes in rack1a, rack1b and rack1c single and double mutants. (A) RT-PCR analysis of the expression of RACK1 genes in rack1a, rack1b and rack1c single and double mutants. RT-PCR was performed at 28 cycles. The expression of ACTIN2 was used as a control. (B) Quantitative real-time PCR analysis of the transcript level of RACK1 genes in rack1a, rack1b and rack1c single and dou-ble mutants. The transcript level of RACK1 genes was normal-ized against the transcript level of ACTIN2 in each sample. The relative transcript level of RACK1 genes in mutant back-grounds was compared with that in wild-type (Col) (set as 1). Page 7 of 11(page number not for citation purposes)a dramatic reduction of the transcript level of RACK1A inrack1b and rack1c single and double mutants, comparedShown are the averages of three replicates ± S.D.BMC Plant Biology 2008, 8:108 http://www.biomedcentral.com/1471-2229/8/108of RACK1 in plants remains poorly understood. We arejust starting to have some hints about its potential func-tions in plants. Preliminary analysis suggested RACK1may mediate multiple hormone responses and develop-mental processes in Arabidopsis [15]. In this study, wefocused on the two characteristic developmental defects ofrack1a mutants, namely the reduction in rosette leaf pro-duction and the reduction in primary root growth and lat-eral root formation, to study the function of RACK1B andRACK1C and the genetic relationship between RACK1homologous genes in plant development. We demon-strated that RACK1 genes are critical regulators of plantdevelopment and are essential for plant survival. Simulta-neous disruption of the function of all three RACK1 genesresults in lethality. Thanks to the unequal genetic redun-dancy of RACK1 genes, we are still able to study the roleof RACK1 genes in plant development. The rack1a singlemutants, rack1a rack1b and rack1a rack1c double mutantsall display developmental defects and are viable. There-fore, these mutants can be treated as "weak alleles" ofrack1 mutants. Now that we have identified RACK1 genesas critical regulators of plant development and all "weakalleles" of rack1 mutants are available, future studiesshould focus on the elucidation of the molecular mecha-nism by which RACK1 genes regulate plant development,including rosette leaf production, root growth and lateralroot formation. Because rack1a mutants have also beenshown to display altered responses to hormones [15], itremains unclear if the developmental defects observed inrack1 mutants are due to the altered responses to multiplehormones and if there is also unequal genetic redundancyof RACK1 genes in mediating hormone responses. This isa fertile area that is worth further investigation.Mechanism of unequal genetic redundancy of RACK1 genesGenetic redundancy of homologous genes is thought tobe due to gene duplication events during the evolution ofthe organism. Between homologous genes, genetic redun-dancy can be classified as full redundancy, partial redun-dancy, and unequal redundancy [16]. While fullredundancy and partial redundancy have been docu-mented in numerous cases, unequal genetic redundancyhas just begun to be recognized as a common phenome-non of genetic relationship of homologous genes [16].Unlike non-plant organisms whose genomes contain onlya single RACK1 gene, some plant genomes contain morethan one RACK1 genes (Figure 1). In particular, the Arabi-dopsis genome contains three RACK1 genes, which sharethe similar gene structure with two exons and one intron,and encode three highly similar proteins with approxi-mately 90% identity at the amino acid level [15]. How-ever, the relationship between three Arabidopsis RACK1Arabidopsis RACK1A genes, RACK1A, conferred multipledefects in plant development [15]. Here we show thatloss-of-function mutations in RACK1B or RACK1C do notconfer apparent developmental defects (Figure 2). Theseresults suggested that RACK1B and RACK1C are likely dis-pensable in plant development. However, we found thatalthough rack1b and rack1c mutants displayed wild-typemorphology, rack1b and rack1c can strongly enhance thedevelopmental defects of rack1a mutants (Figure 3, Figure4). These results suggested that RACK1B and RACK1C stillcontribute significantly to the overall activity of RACK1genes. Because the significance of the RACK1B andRACK1C is determined via the mutants, not directly in thewild-type plants, it is also possible that in the wild-typeplants, all the function of RACK1 genes is explicated byRACK1A with no contribution from RACK1B or RACK1Cand the these latter can play a role only if RACK1A is notpresent (e.g. in the rack1a mutant). Nonetheless, thebehaviors and relationship of rack1 mutants satisfy thekey criteria for RACK1 genes being unequally redundanthomologous genes [16].The unequal genetic redundancy is caused by many fac-tors. Among them, the difference in gene expression pat-tern, expression level and cross-regulation of homologousgenes have been recognized as major determinants [16].The unequal genetic redundancy of some homologousgenes is mainly due to the difference in expression patternand/or expression level. For example, CAULIFLOWER(CAL) is closely related in sequence to APETALA1 (AP1),but AP1 and CAL regulate the formation of floral meris-tem in an unequally redundant manner because AP1 isexpressed at much higher level than CAL throughout sepaland petal development [17]. The unequal genetic redun-dancy of homologous genes can also be primarily due tothe cross-regulation. For example,LONG HYPOCOTYL 5(HY5) and its close homolog HY5 HOMOLOG (HYH),both of which are regulators of photomorphogenesis, area pair of unequally redundant genes with similar expres-sion patterns and levels [16,18], but a normal proteinexpression and activity of HYH was dependent on thepresence of a functional HY5 [18].In order to get insight into the mechanism of unequalgenetic redundancy of three RACK1 genes, we examinedeach of these possibilities. Firstly, we showed thatRACK1B and RACK1C are likely in principle functionallyequivalent to RACK1A, because overexpression of eitherRACK1B or RACK1C under the constitutive CaMV 35Spromoter fully complemented the developmental defectsof rack1a mutants (Figure 5). Ideally, it would be advanta-geous to use the native RACK1A promoter to assess theextent of functional equivalency. Nonetheless, resultsPage 8 of 11(page number not for citation purposes)homologous genes has been unknown. Previously, weshowed that loss-of-function mutation in one member offrom our complementation studies indicated that overex-pression of RACK1B or RACK1C can restore rack1a mutantBMC Plant Biology 2008, 8:108 http://www.biomedcentral.com/1471-2229/8/108to wild-type equally well as overexpression of RACK1A,supporting the view that RACK1B and RACK1C likelyfunction similarly as RACK1A. These results implied thatthe unequal genetic redundancy of RACK1 genes is likelydue to the difference in gene expression pattern and/orexpression level, rather than the difference in proteinsequence or activity. To examine this possibility directly,we found that three RACK1 genes are widely expressed invarious tissues and organs in young seedlings and inmature plants (Figure 6). However, RACK1 genes areexpressed at different levels with a general trend ofRACK1A > RACK1B > RACK1C in all tissues and organsexamined (Figure 6). These results supported the view thatthe difference in gene expression level attributes to theunequal genetic redundancy of RACK1 genes in plantdevelopment. However, these results cannot rule out thepossibility that the expression of each RACK1 gene mayalso be restricted to certain cell types. For example, BRL1and BRL3 are homologous to BRI1, a receptor for brassi-nosteroid (BR), and function as BR receptors in vasculardifferentiation in Arabidopsis [19]. It was found that BRI1is ubiquitously expressed in growing cells, but the expres-sion of BRL1 and BRL3 is restricted to non-overlappingsubsets of vascular cells. Future expression analysis at celllevel (e.g. by in situ hybridization and reporter GFP anal-yses) may help address the possibility of cell type-specificexpression of RACK1 genes.We also explored the possibility of cross-regulation byexamining the transcript level of each RACK1 gene in theloss-of-function alleles of each or both of the other twoRACK1 genes. We found that the transcript level of anygiven RACK1 gene was reduced in the single or doublemutants for the other two RACK1 genes (Figure 7). There-fore, both the difference in gene expression level and thecross-regulation contribute to the unequal genetic redun-dancy of RACK1 genes. Unlike HY5 and HYH, for whichthe expression of the duplicate gene (HYH) depends onthe presence of the ancestral gene (HY5) [18], RACK1homologous genes mutually depend on each other forreaching full expression, adding another level of complex-ity for the unequal genetic redundancy. The molecularbasis of such mutual cross-regulation of RACK1 genes ispresently unknown. It would be interesting to test ifRACK1 proteins can work together in a complex, forinstance, through homo- and hetero-dimerization.ConclusionAmong three RACK1 homologous genes in Arabidopsis,RACK1A is likely the ancestral gene whereas RACK1B andRACK1C are duplicate genes because RACK1A appears toretain most of the function of RACK1 gene family. RACK1genes regulate plant development in a continuous, quan-ence on the plant development (Figure 8). Because rack1band rack1c single mutants do not exhibit any defects inplant development whereas the rack1a rack1b and rack1arack1c double mutants display enhanced phenotypescompared with the rack1a single mutant, it is likely thatthe residual activities of RACK1B and RACK1C are abovethis threshold (Figure 8). Therefore, although bothRACK1B and RACK1C are likely dispensable, they stillcontribute significantly to the overall activity of RACK1genes. Both the difference in gene expression level and thecross-regulation are likely the molecular determinants ofunequal genetic redundancy of RACK1 genes in regulatingplant development.The model of unequal genetic redundancy of RACK1 genes in regulating plant developm ntFi re 8The model of unequal genetic redundancy of RACK1 genes in regulating plant development. Arabidopsis genome contains three RACK1 homologous genes, designated as RACK1A, RACK1B and RACK1C, respectively, which encode three highly similar proteins. RACK1 genes regulate plant development likely in a continuous quantitative manner. RACK1A is likely the ancestral gene whereas RACK1B and RACK1C are the duplicate genes, because RACK1A retains the most functions of RACK1 genes. The expression of RACK1 follows a general trend of RACK1A > RACK1B > RACK1C. A certain threshold of gene activity is likely required for the RACK1 genes to have any influence on plant development, and the gene activity can be saturated once an excess of gene activity is reached. Because the loss-of-function mutations in RACK1B or RACK1C or both do not confer any defects in plant development while enhancing the developmental defects of rack1a mutants, the residual activities of RACK1B and RACK1C are likely above this threshold but below the point of saturation. RACK1 genes mutually regulate each other's transcription. Both the difference in gene expression and the cross-regulation are likely the molecular determi-nants of unequal genetic redundancy of RACK1 genes in regu-lating plant development. The model is schematically based on the possible explanations for unequal genetic redundancy Page 9 of 11(page number not for citation purposes)titative manner. It is likely that a certain threshold of geneactivity is required for the RACK1 genes to have any influ-provided by Briggs et al. (2006) [16].BMC Plant Biology 2008, 8:108 http://www.biomedcentral.com/1471-2229/8/108MethodsPlant materials and growth conditionsAll mutants are in the Arabidopsis Columbia (Col-0) eco-type background. The rack1a-1 and rack1a-2 mutants havebeen reported previously [15]. Plants were grown in 5 × 5cm pots containing moistened 1 : 3 mixture of SunshineMix #1 (Sun Gro Horticulture Canada Ltd., http://www.sungro.com) and Metro-Mix 220 (W.R. Grace & Co.,http://www.grace.com) with 10/14 h (short-day condi-tions) or 14/10 h (long-day conditions) photoperiod atapproximately 120 μmol m-2 s-1 at 23°C.Isolation of rack1b and rack1c T-DNA insertional mutantsThe T-DNA insertion mutants of RACK1B (At1g48630),rack1b-1 (SALK_117422) and rack1b-2 (SALK_145920),and the T-DNA insertion mutants of RACK1C(At3g18130), rack1c-1 (SAIL_199_A04) and rack1c-2(SALK_017913), were identified from the SALK T-DNAExpress database http://signal.salk.edu/cgi-bin/tdnaexpress. For the SALK T-DNA insertional mutants [20], theinsertion was confirmed by PCR using RACK1B-specificprimers (5'-TCTCGACCTCAAACCCTG-3' and 5'-GAGAA-GACTTTAGAGTCGATGGA-3') or RACK1C-specific prim-ers (5'-ATCTCTCGCTCTGTTACGC-3' and 5'-ACAATACTGACGCAGTCTGG-3') and a T-DNA left bor-der-specific primer JMLB1 (5'-GGCAATCAGCTGTT-GCCCGTCTCACTGGTG-3'). For the SAIL T-DNAinsertion mutants [21], a different T-DNA left border-spe-cific primer, GarlicLB3 (5'-TAGCATCTGAATTTCATAAC-CAATCTCGATACAC-3'), was used. The absence of full-length transcript of RACK1B or RACK1C in these alleleswas confirmed by RT-PCR.Generation of rack1a, rack1b and rack1c double and triple mutantsDouble mutants between rack1a-1 and rack1b-2 or rack1c-1 were generated by crossing rack1b-2 or rack1c-1 intorack1a-1 single mutant and isolated in the F2 progeny byPCR genotyping. Similarly, double mutants betweenrack1b-2 and rack1c-1 were generated by crossing rack1c-1into rack1b-2 single mutant and isolated in the F2 progenyby PCR genotyping. For simplicity, the rack1a rack1b,rack1a rack1c and rack1b rack1c double mutant nomencla-tures in this report refer specifically to the rack1a-1 rack1b-2, rack1a-1 rack1c-1 and rack1b-2 rack1c-1 mutants, respec-tively.Triple mutant among rack1a-1, rack1b-2 and rack1c-1 wasgenerated by crossing rack1b-2 rack1c-1 into rack1a-1rack1b-2 double mutants. Because rack1a-1 rack1b-2rack1c-1 triple mutants cannot survive in soil to maturity,they are maintained in plants homozygous for the rack1b-2 and rack1c-1 loci and heterozygous for the rack1a-1Genetic complementationThe full-length open-reading frames of RACK1A(At1g18080), RACK1B and RACK1C were amplified froma cDNA library made from seedlings grown in light for 10d, cloned into the pENTR/D-TOPO vector (Invitrogen,http://www.invitrogen.com), and then subcloned intoGateway plant transformation destination binary vectorpB2GW7 [22] by LR recombination reactions. In theseconstructs, the expression of RACK1A, RACK1B orRACK1C was driven by the 35S promoter of the Cauli-flower mosaic virus. Binary vectors were transformed intorack1a-1 or rack1a-2 mutants by Agrobacterium-mediatedtransformation [23]. At least 16 independent transgeniclines were selected from each transformation, and two tofour representative lines were used for further studies. Theexpression of transgene was examined by RT-PCR.RNA isolation, RT-PCR and quantitative real-time PCR analysesFor tissue/organ expression pattern analysis, total RNAwas isolated from different parts of seedlings or matureplants, using the TRIzol reagent (Invitrogen). cDNA wassynthesized from 1 μg total RNA by oligo(dT)20-primedreverse transcription, using THERMOSCRIPT RT (Invitro-gen). RACK1A-specific primers (5'-GGCATCTCCA-GACACCGAAA-3' and 5'-GCAGAGAGCAACGACAGC-3'), RACK1B-specific primers (5'-TCTCGACCTCAAAC-CCTG-3' and 5'-GAGAAGACTTTAGAGTCGATGGA-3'),and RACK1C-specific primers (5'-ATCTCTCGCTCTGT-TACGC-3' and 5'-ACAATACTGACGCAGTCTGG-3') wereused to amplify the transcripts of these three genes,respectively. The expression of ACTIN2 (amplified byprimers 5'-GTTGGGATGAACCAGAAGGA-3' and 5'-GAACCACCGATCCAGACACT-3') was used as a controlin PCR reactions. For the examination of the transcriptlevel of RACK1A, RACK1B and RACK1C in the T-DNAinsertional mutants or in the transgenic lines, 10 d-old,light-grown seedlings were used for total RNA isolation.For the quantitative analysis of RACK1A, RACK1B andRACK1C transcript levels in the different tissues/organs ofwild-type Col plants or in the rack1a-1, rack1b-2 andrack1c-1 single and double mutants, real-time PCR wasperformed. RACK1A-specific real-time PCR primers (5'-CTGAGGCTGAAAAGGCTGACAACAG-3' and 5'-CTAG-TAACGACCAATACCCCAAACTC-3'), RACK1B-specificreal-time PCR primers (5'-GGTTCTACTGGAATCG-GAAACAAGACC-3' and 5'-CTAGTAACGACCAATAC-CCCAGACCC-3'), and RACK1C-specific real-time PCRprimers (5'-GCAGAGAAGAATGAAGGTGGTGT-3' and 5'-CTAGTAACGACCAATACCCCAGACCC-3') were used.The expression of ACTIN2 (amplified by real-time PCRprimers 5'-CCAGAAGGATGCATATGTTGGTGA-3'and 5'-Page 10 of 11(page number not for citation purposes)locus. The status of triple mutant was confirmed by PCRgenotyping.GAGGAGCCTCGGTAAGAAGA-3') was used to normalizethe expression of each gene. The quantitative real-timeBMC Plant Biology 2008, 8:108 http://www.biomedcentral.com/1471-2229/8/108PCR was performed using the MJ MiniOpticon real-timePCR system (Bio-Rad, http://www.biorad.com) and IQSYBR Green Supermix (Bio-Rad).Rosette leaf production assayThe number of rosette leaves was collected from wild-typeCol and mutant plants grown under 10/14 h or 14/10 hphotoperiod with approximately 120 μmol m-2 s-1 at23°C. At least four plants from each genotype were usedin each experiment, and the experiment was repeatedtwice. The rate of rosette leaf production was expressed asthe number of rosette leaves divided by the age of plant.Root growth assaySeedlings were grown on MS/G plates consisting of 1/2Murashige & Skoog (MS) basal medium supplementedwith vitamins (Plantmedia, http://www.plantmedia.com), 1% (w/v) sucrose and 0.6% (w/v) phytoagar(Plantmedia), with pH adjusted to 5.7 with 1N KOH. Theplates were placed under 14/10 h photoperiod withapproximately 120 μmol m-2 s-1 at 23°C with a verticalorientation for monitoring root growth. The length of pri-mary and the number of lateral roots were collected fromat least 15 seedlings each genotype.Authors' contributionsJG isolated the rack1 single, double and triple mutants,and conducted all experiments. JGC conceived of thestudy and participated in its design and coordination. Allauthors participated in drafting and editing the manu-script, and read and approved the final manuscript.AcknowledgementsWe thank the Salk Institute Genomic Analysis Laboratory (La Jolla, CA), the Syngenta Biotechnology, Inc. (Research Triangle Park, NC), and the Arabi-dopsis Biological Resources Center (Columbus, Ohio) for providing the rack1b and rack1c T-DNA insertional mutants. This work was supported by grants from the Natural Sciences and Engineering Research Council of Can-ada (grant No. RGPIN311651-05) and the Canada Foundation for Innova-tion (grant No. 10496).References1. Mochly-Rosen D, Khaner H, Lopez J: Identification of intracellu-lar receptor proteins for activated protein kinase C.  Proc NatlAcad Sci USA 1991, 88:3997-4000.2. Ron D, Chen CH, Caldwell J, Jamieson L, Orr E, Mochly-Rosen D:Cloning of an intracellular receptor for protein kinase C – Ahomolog of the β-subunit of G-proteins.  Proc Natl Acad of SciUSA 1994, 91:839-843.3. McCahill A, Warwicker J, Bolger GB, Houslay MD, Yarwood SJ: TheRACK1 scaffold protein: A dynamic cog in cell responsemechanisms.  Mol Pharmacol 2002, 62:1261-1273.4. Sklan EH, Podoly E, Soreq H: RACK1 has the nerve to act: struc-ture meets function in the nervous system.  Prog Neurobiol2006, 78:117-134.5. Ishida S, Takahashi Y, Nagata T: Isolation of cDNA of an auxin-regulated gene encoding a G-protein β-subunit-like proteinfrom tobacco BY-2-cells.  Proc Natl Acad Sci USA 1993,90:11152-11156.7. McKhann HI, Frugier F, Petrovics G, delaPena TC, Jurkevitch E,Brown S, Kondorosi E, Kondorosi A, Crespi M: Cloning of a WD-repeat-containing gene from alfalfa (Medicago sativa): a rolein hormone-mediated cell division?  Plant Mol Biol 1997,34:771-780.8. Perennes C, Glab N, Guglieni B, Doutriaux MP, Phan TH, Planchais S,Bergounioux C: Is arcA3 a possible mediator in the signaltransduction pathway during agonist cell cycle arrest by sal-icylic acid and UV irradiation?  J Cell Sci 1999, 112:1181-1190.9. Iwasaki Y, Komano M, Ishikawa A, Sasaki T, Asahi T: Molecularcloning and characterization of cDNA for a rice protein thatcontains seven repetitive segments of the Trp-Asp forty-amino-acid repeat (WD-40 repeat).  Plant Cell Physiol 1995,36:505-510.10. Komatsu S, Abbasi F, Kobori E, Fujisawa Y, Kato H, Iwasaki Y: Pro-teomic analysis of rice embryo: an approach for investigatingGα protein-regulated proteins.  Proteomics 2005, 5:3932-3941.11. Nakashima A, Chen L, Thao NP, Fujiwara M, Wong HL, Kuwano M,Umemura K, Shirasu K, Kawasaki T, Shimamoto K: RACK1 func-tions in rice innate immunity by interacting with the Rac1immune complex.  Plant Cell 2008, 20:2265-2279.12. Chang IF, Szick-Miranda K, Pan S, Bailey-Serres J: Proteomic char-acterization of evolutionarily conserved and variable pro-teins of Arabidopsis cytosolic ribosomes.  Plant Physiol 2005,137:848-862.13. Giavalisco P, Wilson D, Kreitler T, Lehrach H, Klose J, Gobom J,Fucini P: High heterogeneity within the ribosomal proteins ofthe Arabidopsis thaliana 80S ribosome.  Plant Mol Biol 2005,57:577-591.14. Ullah H, Scappini EL, Moon AF, Williams LV, Armstrong DL, PedersenLC: Structure of a signal transduction regulator, RACK1,from Arabidopsis thaliana.  Protein Sci 2008, 17:1771-1780.15. Chen JG, Ullah H, Temple B, Liang J, Guo J, Alonso JM, Ecker JR, JonesAM: RACK1 mediates multiple hormone responsiveness anddevelopmental processes in Arabidopsis.  J Exp Bot 2006,57:2697-2708.16. Briggs GC, Osmont KS, Shindo C, Sibout R, Hardtke CS: Unequalgenetic redundancies in Arabidopsis – a neglected phenom-enon?  Trends Plant Sci 2006, 11:492-498.17. Kempin SA, Savidge B, Yanofsky MF: Molecular basis of the cauli-flower phenotype in Arabidopsis.  Science 1995, 267:522-525.18. Holm M, Ma LG, Qu LJ, Deng XW: Two interacting bZIP pro-teins are direct targets of COP1-mediated control of light-dependent gene expression in Arabidopsis.  Genes Dev 2002,16:1247-1259.19. Caño-Delgado A, Yin Y, Yu C, Vafeados D, Mora-García S, Cheng JC,Nam KH, Li J, Chory J: BRL1 and BRL3 are novel brassinoster-oid receptors that function in vascular differentiation in Ara-bidopsis.  Development 2004, 131:5341-5351.20. Alonso JM, Stepanova AN, Leisse TJ, Kim CJ, Chen H, Shinn P, Ste-venson DK, Zimmerman J, Barajas P, Cheuk R, Gadrinab C, Heller C,Jeske A, Koesema E, Meyers CC, Parker H, Prednis L, Ansari Y, ChoyN, Deen H, Geralt M, Hazari N, Hom E, Karnes M, Mulholland C,Ndubaku R, Schmidt I, Guzman P, Aguilar-Henonin L, Schmid M,Weigel D, Carter DE, Marchand T, Risseeuw E, Brogden D, Zeko A,Crosby WL, Berry CC, Ecker JR: Genome-wide insertionalmutagenesis of Arabidopsis thaliana.  Science 2003, 301:653-657.21. Sessions A, Burke E, Presting G, Aux G, McElver J, Patton D, DietrichB, Ho P, Bacwaden J, Ko C, Clarke JD, Cotton D, Bullis D, Snell J,Miguel T, Hutchison D, Kimmerly B, Mitzel T, Katagiri F, GlazebrookJ, Law M, Goff SA: A high-throughput Arabidopsis reversegenetics system.  Plant Cell 2002, 14:2985-2994.22. Karimi M, Inze D, Depicker A: GATEWAY((TM)) vectors forAgrobacterium-mediated plant transformation.  Trends Plant Sci2002, 7:193-195.23. Clough SJ, Bent AF: Floral dip: a simplified method for Agrobac-terium-mediated transformation of Arabidopsis thaliana.  PlantJ 1998, 16:735-743.Page 11 of 11(page number not for citation purposes)6. Guo J, Liang J, Chen JG: RACK1: a versatile scaffold protein inplants?  Int J Plant Dev Biol 2007, 1:95-105.

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.52383.1-0074637/manifest

Comment

Related Items