(This is a sample cover image for this issue. The actual ...mrec.ifas.ufl.edu/grapes/genetics/Selected-Publications/B PlantSci... · plants via the cisgenic/intragenic/ingenic approach

(This is a sample cover image for this issue. The actual cover is not yet available at this time.)

This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institution

and sharing with colleagues.

Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies areencouraged to visit:

http://www.elsevier.com/copyright

http://www.elsevier.com/copyright

Author's personal copy

Plant Science 196 (2012) 132–142

Contents lists available at SciVerse ScienceDirect

Plant Science

j our nal ho me p age: www.elsev ier .com/ locate /p lantsc i

Large-scale characterization of promoters from grapevine (Vitis spp.) usingquantitative anthocyanin and GUS assay systems

Zhijian T. Li, Kyung-Hee Kim, Jonathon R. Jasinski, Matthew R. Creech, Dennis J. Gray ∗

Grape Biotechnology Core Laboratory, Mid-Florida Research and Education Center, University of Florida/IFAS, 2725 Binion Road, Apopka, FL 32703-8504, USA

a r t i c l e i n f o

Article history:Received 13 June 2012Received in revised form 10 August 2012Accepted 11 August 2012Available online 20 August 2012

Keywords:Promoter activityQuantitative reporter markerAnthocyaninColor histogramUbiquitin gene promoterVitis vinifera

a b s t r a c t

Successful implementation of cisgenic/intragenic/ingenic technology for crop improvement necessitatesa better understanding of the function of native promoters for driving desired gene expression in hostplant. Although the genome of grapevine (Vitis vinifera) has been determined, efforts to explore pro-moter resources for the development of cisgenics are still lacking. Particularly, there is a shortage ofconstitutive promoters for marker and/or target gene expression in this species. In this work, we uti-lized an anthocyanin-based color histogram analysis method to evaluate quantitatively a large numberof promoters for their ability to activate transgene expression. Promoter fragments corresponding toknown genes were amplified from various genotypes and used to drive the VvMybA1 gene of ‘Merlot’ foranthocyanin production in non-pigmented somatic embryo (SE) explants to infer transcriptional activity.Results revealed that among 15 tested promoters belonging to seven ubiquitin genes, at least three pro-moters generated constitutive activities reaching up to 100% value of the d35S promoter. In particular,the high activity levels of VvUb6-1 and VvUb7-2 promoters were verified by transient GUS quantitativeassay as well as stable anthocyanin expression in sepal and corolla of transgenic tobacco. Variations inpromoter activity of different ubiquitin genes in grapevine did not correlate with the presence and sizesof 5′ UTR intron, but seemed to be related positively and negatively to the number of positive cis-actingelements and root-specific elements respectively. In addition, several of the 13 promoters derived froma PR1 gene and a PAL gene produced a higher basal activity as compared to previously reported induciblepromoters and might be useful for further identification of strong inducible promoters. Our study con-tributed invaluable information on transcriptional activity of many previously uncharacterized nativepromoters that could be used for genetic engineering of grapevine.

© 2012 Elsevier Ireland Ltd. All rights reserved.

1. Introduction

In recent years, concerted efforts in genetic manipulation ofcrop plants via the cisgenic/intragenic/ingenic approach (cisgenicshereafter) have gained an accelerated momentum due to the rapidprogress in plant genomics and proteomics. As compared to trans-genics technology that primarily relies on the use of foreign geneticmaterials to modify recipient species, cisgenics technology specif-ically utilizes reengineered native genes and genetic elements toimprove host species. Cisgenic plants are considered to be essen-tially similar to plants bred through conventional hybridizationmethod. In addition, cisgenic products will require less stringentregulatory scrutiny designed to prevent unintended disseminationof transgenes to the environment while boosting consumer accep-tance and confidence [1–4].

∗ Corresponding author. Tel.: +1 407 884 2034x126; fax: +1 407 814 6186.E-mail address: [email protected] (D.J. Gray).

Grapevine (Vitis vinifera L.) is being grown worldwide withsignificant economic impact on many producing regions. On theother hand, traditional breeding programs for improving resistanceand agronomic performance of this crop are often hampered bymany limiting factors, including difficulties to achieve unique eno-logic quality and wine attributes consumers preferred, long lifecycle, self-incompatibility and highly heterologous genetic milieuassociated with vegetative reproduction [5]. The advances in therefinement of tissue culture and genetic transformation techniquesand the completion of grape genome provided an unparalleledopportunity for introduction of native genes and genetic elementsinto elite grape varieties for trait improvement without chang-ing other existing characteristics [6–8]. However, efforts are stillneeded to resolve obstacles to the implementation of such cisgenicstechnology in grapevine. For the time being, there is a critical short-age of known native promoters that can be readily used to directdesired expression of native genes in engineered grape plants.

Promoter is the single most important genetic element orches-trating gene expression both qualitatively and quantitatively. Overthe last three decades, a large number of promoters have been

0168-9452/$ – see front matter © 2012 Elsevier Ireland Ltd. All rights reserved.http://dx.doi.org/10.1016/j.plantsci.2012.08.009


Z.T. Li et al. / Plant Science 196 (2012) 132–142 133

Table 1Chromosome locations and predicted introns of 1 kb ubiquitin gene promoter fragments derived from the ‘Pinot Noir’ genome.

Promoter Chromosomea (12×) Promoter Intron Gene accession no.

Start End Orientationb 5′-Splicingc 3′-Splicing Size (bp)

VvUb1 4 4271375 4272374 Reverse 509 793 285 N/AVvUb2 7 4556007 4557006 Sense 637 755 119 GSVIVP00028185001VvUb3 8 4578661 4579660 Sense N/A N/A N/A GSVIVP00027151001VvUb4 8 17077647 17078646 Sense N/A N/A N/A N/AVvUb5 16 21450319 21451318 Reverse N/A N/A N/A AM434449VvUb6 19 5889905 5890904 Sense 566 832 267 CBI22354VvUb7 19 5914878 5915877 Sense 569 843 275 N/A

a Ubiquitin gene promoters were identified from corresponding chromosomes of the 12× assembly of the ‘Pinot Noir’ genome.b Orientation of the promoters is shown based on sequence orientations of the published chromosomes.c Intron splicing sites within each 1 kb promoter were identified using NetPlantGene server for predictions of splice sites in Arabidopsis thaliana DNA:

http://www.cbs.dtu.dk/services/NetPGene/. N/A denotes no matching splicing sites or gene accession hits.

isolated from a wide range of plant species and characterized ingreat detail. In general, promoters function in plants to support con-stitutive, inducible, tissue-specific and developmentally regulatedexpression [9]. In particular, constitutive promoters direct high lev-els of gene expression in all cell types throughout the entire periodof growth and development, and thus provide a broader applicationin genetic engineering programs [10]. In grapevine, the majorityof promoters used in transformation studies were mainly derivedfrom viral origin [11–15]. It has become more urgent to identify andcharacterize native promoters with constitutive activity and otherdesired regulatory control to replace viral promoters used in thisspecies.

In this study, we isolated a total of 31 promoters from variousgenotypes of grapevine and extensively characterized their tran-scription activity using transient expression of both anthocyaninand GUS reporter genes and associated quantitative analysis meth-ods. We showed for the first time that a number of ubiquitin genepromoters were capable of directing high-level of gene expressioncomparable to the double enhanced CaMV 35S promoter. Stabletransgene expression driven by these native promoters was alsodetermined in transgenic tobacco. These promoters provide aninvaluable tool for both promoter development and cisgenic engi-neering of grapevine.

2. Materials and methods

2.1. Plant culture preparation and transformation

Somatic embryos (SE) (V. vinifera cv. ‘Thompson Seedless’)were initiated from young leaves of in vitro-grown shoot tips andmaintained on X6 medium according to Scorza et al. [11]. SE atmid-cotyledonary stage of development were used for transforma-tion analysis. Transformation procedures and culture conditionsspecifically designed for anthocyanin expression were describedpreviously [16].

Tobacco seeds (N. tabacum cv. ‘Samsun’) were sterilized in 50%bleach solution for 10 min followed by three rinses with sterilizedwater. Seeds were plated on Petri dishes containing MS mediumand allowed to grow under light for one week. Cotyledonary leaveswere collected and used for transformation and plant regenerationaccording to Burrow et al. [17]. Transgenic tobacco plants wereestablished in potted soil in the greenhouse for examination ofanthocyanin expression.

2.2. Genomic sequence source and promoter isolation

All promoter sequences were identified by searching avail-able public genomics databases including NCBI GenBank(http://www.ncbi.nlm.nih.gov/Genbank) that contains the genomeof V. vinifera ‘Pinot Noir’ clone ENTAV 115 and Genoscope

(http://www.genoscope.cns.fr/cgi-bin/blast server/projetML/blast.pl) with the genome of ‘Pinot Noir’ clone PN40024.

Retrieved DNA sequences from different sources were verifiedby alignment analysis using the Vector NTI AdvanceTM softwareversion 10.3.0 (InVitrogen, Carlsbad, CA, USA) prior to use as atemplate for PCR amplification.

Promoters corresponding to genes encoding ubiquitin (hereinnamed VvUb), a previously described PR1 protein XP 002273416(named VvPR1) [18], 2S albumin protein (named VvAlb1) [19],cytochrome P450 (named VvGP450), phenylalanine ammonia-lyase (PAL) (named VvPAL1) were tested. Since many ubiquitingenes contain an inconspicuous intron in the 5′-untranslated region(5′-UTR), we used the simple term promoter to describe the spec-ified sequence region upstream from the translation start site of acorresponding ubiquitin gene, which included upstream transcrip-tion regulatory sequence and a 5′ UTR region with or without anintron. A total of seven ubiquitin genes identified from the genomeof ‘Pinot Noir’ were used to retrieve promoters from genomic DNAof different varieties after PCR amplification. Chromosome loca-tions, promoter sequence orientations and intron splicing sites ofthese ubiquitin gene promoters were listed in Table 1.

Phylogenetic analysis and dendrogram reconstruction were car-ried out using MEGA5 software [20]. The Neighbor-Joining (NJ)method with 2000 replications and a bootstrap value of >70% assignificant branch limit was employed [21,22].

2.3. Template DNA isolation and PCR amplification

Total genomic DNA was isolated from leaf tissues of greenhouse-grown plants using the procedure of Lodhi et al. [23] with amodified extraction buffer containing an increased EDTA concen-tration of 100 mmol/l. Genotypes used to obtain target promotersequences included five V. vinifera genotypes (‘Merlot’, ‘Pinot Noir’,‘Pinotage’, ‘Remaily Seedless’ and ‘Superior Seedless’), a wild Vitisspecies ‘Haines City’ (Vitis shuttleworthii House) and a Florida bunchgrape hybrid ‘BN5-4′ [Vitis aestivalis Michx. ssp. simpsoni MunsonX ‘Remaily Seedless’ (V. vinifera X Vitis labrusca L.)] (developed atthe Florida Experiment Station, University of Florida).

Oligonucleotide primers were designed by using Vector NTIAdvanceTM. Primers were synthesized by Integrated DNA Tech-nologies, Inc. (Coralville, IA, USA). Gateway Clonase Technology(InVitrogen) was utilized to facilitate the cloning of PCR-amplifiedDNA fragments into transformation vectors. Accordingly, DNAsequences corresponding to both attB1 (5′-GGG GAC AAG TTT GTACAA AAA AGC AGG CT-3′) and attB2 (5′-GGG GAC CAC TTT GTACAA GAA AGC TGG GT-3′) sites for BP recombination reactionwere incorporated into the terminal region of forward and reverseprimers respectively (Table 2).

Chemical reagents and DNA polymerase were purchased fromPromega (Madison, WI, USA). PCR reactions were carried out in


134 Z.T. Li et al. / Plant Science 196 (2012) 132–142

Table 2Oligonucleotide primers for PCR amplification of test promoters from grapevine.

Promoter Primer pair DNA sequence (5′ to 3′ orientation)a

VvUb1 Ub1-51 ggggacaagtttgtacaaaaaagcaggcttatttatgataatattaaaaattgaggattt(1 kb) Ub1-32 ggggaccactttgtacaagaaagctgggtcttgagactttgagagagacgttctaaagcc

VvUb2 Ub2-51 ggggacaagtttgtacaaaaaagcaggctacaatcttaacttacttcaagttcaacttaa(1 kb) Ub2-32 ggggaccactttgtacaagaaagctgggtcttcgcgtcaggggaagaggggtaaggatta

VvUb3 Ub3-51 ggggacaagtttgtacaaaaaagcaggcttgtatactaattaaatataaaacaaatataa(1 kb) Ub3-32 ggggaccactttgtacaagaaagctgggtcttggctccttttctgtggagggccttcagg

VvUb4 Ub4-51 ggggacaagtttgtacaaaaaagcaggctaattttagcatcttatcctctatcagtattt(1 kb) Ub4-32 ggggaccactttgtacaagaaagctgggtcttctctctcgttatcttcctcttcttctgc

VvUb4-Long Ub4L-51 ggggacaagtttgtacaaaaaagcaggctgccatgcttgcagcacaagatgtt(2 kb) Ub4-32 ggggaccactttgtacaagaaagctgggtcttctctctcgttatcttcctcttcttctgc

VvUb5 Ub5-51 ggggacaagtttgtacaaaaaagcaggctacctcaatgaatggatcagagaacctccaga(1 kb) Ub5-32 ggggaccactttgtacaagaaagctgggttttcgccgccttctgtcgcttgctgaggatg

VvUb5-Long Ub5L-51 ggggacaagtttgtacaaaaaagcaggctcattctattgaatcgatgtatggt(2 kb) Ub5-32 ggggaccactttgtacaagaaagctgggttttcgccgccttctgtcgcttgctgaggatg

VvUb6 Ub6-51 ggggacaagtttgtacaaaaaagcaggctagaagtctgggtggttgcagacttgcagtta(1 kb) Ub6-32 ggggaccactttgtacaagaaagctgggtctgtttcaattaaaacaccacaaagatcagg

VvUb7 Ub7-51 ggggacaagtttgtacaaaaaagcaggcttatttttggaattgtttttctattgcgaaag(1 kb) Ub7-32 ggggaccactttgtacaagaaagctgggtctgtttcaattgaaaaaccacaaagatcagg

VvUb7-Long Ub7L-51 ggggacaagtttgtacaaaaaagcaggcttaaatatggtaaaaatgtataaat(2 kb) Ub7-32 ggggaccactttgtacaagaaagctgggtctgtttcaattgaaaaaccacaaagatcagg

VvPR1 3PS-51 ggggacaagtttgtacaaaaaagcaggctcattttatttattttaactcatgaatataaa(0.5 kb) 3PL-32 ggggaccactttgtacaagaaagctgggttttcagttgtgaagtttaatgtaattgatgt

VvPR1-Long 3PL-51 ggggacaagtttgtacaaaaaagcaggctttgtagaaaagttttgataattattagagca(1.5 kb) 3PL-32 ggggaccactttgtacaagaaagctgggttttcagttgtgaagtttaatgtaattgatgt

VvGP450 P4-51 ggggacaagtttgtacaaaaaagcaggctcacaatggacttcctgccataaaatgca(1 kb) P4-32 ggggaccactttgtacaagaaagctgggtggtgatatagatggtgataactgtg

VvAlb1-1 ABLS-51 ggggacaagtttgtacaaaaaagcaggctggaaaatctaaatttgcatgcac(1.8 kb) ABS-32 ggggaccactttgtacaagaaagctgggtgtaatgagagagagggatcga

VvAlb1-2 ABL-51 ggggacaagtttgtacaaaaaagcaggctcacagatcgatgacttgttccagtgggtgtt(2.1 kb) ABS-32 ggggaccactttgtacaagaaagctgggtgtaatgagagagagggatcga

VvPAL1 PAL-51 ggggacaagtttgtacaaaaaagcaggcttttatgaaacgtcaaagctaatattc(1.4 kb) PAL32 ggggaccactttgtacaagaaagctgggtacaaagaaacaactttaaaaagc

a Primer sequences were designed based on the genome of V. vinifera cv. ‘Pinot Noir’.

a PTC-100 thermocycler (MJ Research, Inc., Watertown, MA, USA)with a 10 �l reaction mixture that contains 30 ng template DNA,4 ng each primer, 1× Gotaq Flexi buffer, 1.5 mM MgCl2, 150 �Meach PCR nucleotide and 1 U GoTaq DNA polymerase. Thermocy-cling conditions were as follows: one cycle at 94 ◦C for 3 min; 40cycles at 93 ◦C for 30 s, 55 ◦C for 1 min and 72 ◦C for 2 min followedby one cycle at 72 ◦C for 8 min. To remove unused primers, amplifi-cation products were mixed with 1/2 volume of 30% (w/v) PEG 8000solution with 30 mM MgCl2 followed by centrifugation at 13 k rpmfor 20 min at 4 ◦C. DNA pellets were finally resuspended in 10 �l ofsterile water and subsequently used in gel electrophoresis analysisand clonase reaction.

2.4. Construction of transformation vectors

A BP-recombination plasmid pDONR221 was used to clonePCR-amplified DNA products into pEntry vector via site-specificrecombination catalyzed by BP ClonaseTM II enzyme mix (InVit-rogen). The identity of DNA inserts in the resulting pEntry colonieswas confirmed by comparison of restriction patterns with templatesequences after DNA digestion with appropriate enzymes and elec-trophoresis on 0.6% agarose gels. DNA sequencing of selected cloneswas conducted by the Interdisciplinary Center for BiotechnologyResearch (ICBR) at University of Florida.

To incorporate test promoters into pExpression transfor-mation vectors, a pDestination binary vector harboring an

anthocyanin-inducing VvMybA1 gene was developed based on apreviously described vector pCsVM [13]. Specifically, the EGFP-NPTII gene in pCsVM was first replaced with a VvMybA1 genethat was previously isolated from ‘Merlot’ [16]. Then the entireCsVMV promoter in the VvMybA1-harboring vector was excised andreplaced with a DNA fragment containing both LR-specific attR1and attR2 recombination sites and the selectable ccdB/Cam genecassette (InVitrogen). The resulting pDestination vector termedpDesAT-1 was then used to receive target promoters to form pEx-pression vectors via site-specific recombination catalyzed by LRClonaseTM II enzyme mix (InVitrogen). All pExpression binary vec-tors were subsequently introduced into Agrobacterium tumefaciensstrain EHA105 by the freeze–thaw method [17] and used for planttransformation.

A pDestination binary vector termed pDesEGG-1 containing anEGFP/GUS translational fusion gene was also constructed using thesimilar approach and used to develop pExpression constructs forpromoter activity analysis via quantitative GUS activity assay.

2.5. Agrobacterium-mediated transformation and geneexpression detection

Agrobacterium cultures were initiated in a 125 ml flask contain-ing 25 ml of MGL medium supplemented with 100 mg/l kanamycinand 20 mg/l rifampicin [6]. Bacterial cultures were maintained at25 ◦C on a rotary shaker at 200 rpm overnight. When cultures



reached an optical density of 0.8–1.0 at 620 nm (OD620), bacterialcells were collected by centrifugation at 6000 rpm for 3 min andthen resuspended in an equal volume of liquid X2 medium (X6medium modified to contain 20 g/l sucrose) [6]. SE at midcotyle-donary stage of development were submerged in Agrobacteriumsolution for 1 min. After removing excessive liquid, SE were placedon X2-prewetted filter paper in a Petri dish and kept in the darkfor 72 h. Transformed SE were subcultured onto agar-solidifiedX6cc medium (X6 medium supplemented with 7 g/l TC-agar and200 mg/l each of cefotaxime and carbenicillin) and maintained inthe dark at 25 ◦C for a total of 6 d prior to image acquisition or GUSactivity analysis.

2.6. Color histogram-based quantitative anthocyanin expressionanalysis

SE transformed with various promoter constructs were pho-tographed with a Leica MZFLIII dissecting microscope (LeicaMicroscopy Systems Ltd, Heerbrugg, Switzerland) under identicalmagnification and lighting conditions. For each image, a fixed areaof 36 pixels covering anthocyanin-expressing cells was selectedto acquire a red brightness reading via Photoshop program as asingle reading. Ten readings were obtained to arrive at the meanred brightness (MRB) value for each treatment sample. MRB val-ues were then used for evaluation of anthocyanin expression levelsaccording to previously described procedures [16].

Since non-pigmented SE showed a relatively high intensity ofyellowish color leading to a higher than expected background his-togram reading, two modifications were made to the calculationof histogram-calculated anthocyanin quantity (HCAQ): (1) a valueof 19 based on MRB from control SE was added as a normal-izing factor to arrive at a relative mean red brightness (RMRB)value for each treatment; (2) the RMRB data were then convertedto RMRB-calculated optical density (RMCOD) values at 492 nmbased on YA = −0.0131XA + 2.9436 where YA represents RMCOD andXA denotes RMRB value (Table 3, A). Finally histogram-calculatedanthocyanin quantity (HCAQ) values were determined using thepreviously described equation YB = 406.42XB − 180.76 where YBdenotes HCAQ, XB refers to RMCOD [16]. To bring data set values topositive position, the test sample that showed the lowest negativeHCAQ value was adjusted to zero by addition of an identical pos-itive value. Meanwhile all other samples were also treated in thesame fashion to arrive at an adjusted HCAQ (Table 3, Adj-B). Antho-cyanin expression levels were represented by the percentage rateof sample HCAQ over that from a control promoter of either DEATor DAT [14]. Experiments were repeated twice. Standard statisticalmethodology was applied for data analysis.

2.7. Quantitative GUS activity assay

SE were collected six d after transformation, ground into a slurryin GUS extraction buffer and subject to a fluorogenic assay withthe substrate 4-methylumbelliferyl glucuronide (MUG) followingthe procedure of Jefferson [24]. The relative GUS activity at pmolMU/min/mg protein observed from each test promoter was con-verted to a percentage value over the indicator vector pd35G witha d35S promoter for comparative analysis [14]. Experiments wererepeated three times. Standard statistical methodology was appliedfor data analysis.

3. Results

3.1. Isolation of promoter sequences from grapevine

A total of 31 promoters were isolated from various geno-types of grapevine (Vitis spp.) following PCR amplification. These

promoters included 15 variant VvUb promoters belonging toseven ubiquitin genes identified through sequence search ofthe ‘Pinot Noir’ genome (Tables 1 and 3), 10 VvPR1 promot-ers from a grapevine PR1 gene, three VvPAL1 promoters froma PAL gene, two VvAlb1 promoters from a 2S albumin geneand one VvGP450 promoter from a cytochrome P450 gene(Table 3).

The identity of all promoter fragments was confirmed pri-marily based on comparison of DNA size and restriction patternsof PCR-amplified promoters with known promoter sequencesof ‘Pinot Noir’ (data not shown). Due to the large numberof promoters employed, the relatively long sequence lengthwith each promoter and the less stringent functional require-ment for promoter sequences as compared to gene codingsequences, PCR-amplified promoters were not subject to sequenc-ing analysis prior to expression experiments. Instead, multiplepromoters from the same and/or different genotypes wereevaluated for transcriptional activity. In this study, only 1 kbubiquitin gene promoters that showed high levels of activitywere subsequently sequenced and analyzed for sequence fidelity(Table 4).

3.2. Analysis of transient anthocyanin expression driven byvarious promoters

Anthocyanin production as the result of the VvMybA1 geneexpression was used to infer transcriptional activity of test pro-moters [16]. Production and accumulation of red to purple colorpigment were observed in whitish SE 72 h after transformation.However, such rapid anthocyanin expression mostly appearedin SE transformed with viral promoters. Anthocyanin productionfrom grapevine promoters was largely delayed by 2–3 d (data notshown). Accordingly, expression analysis was performed after anextended period of time to further pigment development.

One week after transformation, anthocyanin pigment produc-tion in grape SE containing viral promoters was clearly intensified(Fig. 1, panels 1–4). In particular, the d35S-derived bi-directionalduplex promoter (BDDP) DEAT supported the highest intensity oftransient anthocyanin expression (Fig. 1, panel 1). Such observedexpression pattern is in good agreement with findings of our pre-vious report that provided great details of promoter sequencesources, vector construction, expression enhancement and possibleexplanations for this promoter [14]. A wide range of anthocyaninlevels were detected with the majority of promoters isolated fromgrapevine. Noticeably, three promoters failed to induce any antho-cyanin production, including VvUb1-1, VvUb3-1 and VvGP450-1(Fig. 1, panels 5, 7 and 30; Table 3). Several VvUb promotersshowed relatively higher levels of anthocyanin pigment content.In particular, VvUb5-2, VvUb7-2 and VvUb7-5 showed a pigmen-tation level comparable to that of viral control promoters (Fig. 1,panels 12, 16 and 19). On the other hand, relatively lower antho-cyanin pigmentation was detected in SE transformed with VvPR1and VvPAL1 promoters (Fig. 1, panels 20–29 and 33–35). Antho-cyanin expression at a relatively low intensity was strictly localizedin cotyledonary tissues in SE transformed with two VvAlb1 pro-moters (Fig. 1, panels 31 and 32), confirming the seed-specificexpression pattern of the 2S albumin gene promoter as previouslydemonstrated [19].

In order to use anthocyanin pigment production as a quanti-tative attribute to infer promoter activity, a previously describedcolor histogram-based image analysis approach was applied [16].Among the viral promoters tested, a single d35S promoter DATand two dCsVMV-associated promoters CAT and CEAT supportedabout 60% anthocyanin quantity of the d35S-derived BDDP pro-moter DEAT (Fig. 2, bars 2–4 vs. 1).



Table 3Transformation vectors with promoters of various origins and VvMybA1 gene and color histogram-based estimates of anthocyanin expression in SE of grapevine (V. viniferacv. ‘Thompson Seedless’) obtained 6 d after transformation.

No. Promotera Size (bp) Origin 6 d-Anthb MRBc RMRBd Ae Bf Adj Bg SEh

1 DEAT 1594 d35S-BDDP Positive 109.19 128.19 1.26 333.10 712.10 1.792 DAT 419 d35S Positive 165.09 184.09 0.53 35.47 414.47 8.693 CAT 845 dCsVMV Positive 165.69 184.69 0.52 32.25 411.25 10.494 CEAT 1594 dCsVMV-BDDP Positive 158.65 177.65 0.62 69.76 448.76 8.175 VvUb1-1 1000 Merlot Negative 197.50 261.50 −0.48 −376.65 2.35 0.006 VvUb2-1 1000 Pinot Noir Positive 223.25 242.25 −0.23 −274.20 104.80 1.897 VvUb3-1 1000 Merlot Negative 191.32 261.76 −0.49 −378.06 0.94 0.008 VvUb4-1 1000 Merlot Positive 223.24 242.24 −0.23 −274.11 104.89 12.859 VvUb4-2 2000 Pinot Noir Positive 220.37 239.37 −0.19 −258.85 120.15 1.89

10 VvUb5-1 1000 Merlot Positive 198.37 217.37 0.10 −141.70 237.30 8.4711 VvUb5-2 1000 BN5-4 Positive 177.07 196.07 0.38 −28.32 350.68 9.6912 VvUb5-3 2000 Merlot Positive 196.64 215.64 0.12 −132.50 246.50 3.4813 VvUb5-4 2000 Pinot Noir Positive 212.94 231.94 −0.09 −219.29 159.71 9.6314 VvUb6-1 1000 Merlot Positive 195.82 214.82 0.13 −128.14 250.86 5.9615 VvUb7-1 1000 Merlot Positive 222.03 241.03 −0.21 −267.70 111.30 5.3316 VvUb7-2 1000 Merlot Positive 175.95 194.95 0.39 −22.38 356.62 4.8817 VvUb7-3 1000 Merlot Positive 220.96 239.96 −0.20 −261.99 117.01 10.3918 VvUb7-4 1000 BN5-4 Positive 219.27 238.27 −0.18 −253.00 126.00 8.4219 VvUb7-5 2000 Pinot Noir Positive 163.48 182.48 0.55 44.05 423.05 9.0820 VvPR1-1 555 BN5-4 Positive 220.15 239.15 −0.19 −257.70 121.30 5.3521 VvPR1-2 555 BN5-4 Positive 235.86 254.86 −0.40 −341.32 37.68 1.6322 VvPR1-3 555 Haines City Positive 221.94 240.94 −0.21 −267.23 111.77 1.0523 VvPR1-4 555 Merlot Positive 233.21 252.21 −0.36 −327.22 51.78 5.0724 VvPR1-5 555 Superior Seedless Positive 236.38 255.38 −0.40 −344.09 34.91 8.0725 VvPR1-6 555 Pinot Noir Positive 228.92 247.92 −0.30 −304.39 74.61 2.3226 VvPR1-7 1555 BN5-4 Positive 228.40 247.40 −0.30 −301.59 77.41 4.8327 VvPR1-8 1555 Merlot Positive 229.92 248.92 −0.32 −309.70 69.30 2.5928 VvPR1-9 1555 Pinotage Positive 226.46 245.46 −0.27 −291.26 87.74 2.3329 VvPR1-10 1555 Remaily Seedless Positive 230.53 249.53 −0.33 −312.92 66.08 1.0830 VvGP450-1 1070 Pinot Noir Negative 243.22 262.22 −0.49 −380.50 −1.50 0.0031 VvAlb1-1 1800 Merlot Positive 238.01 257.01 −0.42 −352.77 26.23 3.0632 VvAlb1-2 2100 Merlot Positive 240.18 259.18 −0.45 −364.32 14.68 4.9733 VvPAL1-1 1400 Merlot Positive 228.02 247.02 −0.29 −299.58 79.42 3.0634 VvPAL1-2 1400 Pinot Noir Positive 235.59 254.59 −0.39 −339.89 39.11 3.4235 VvPAL1-3 1400 BN5-4 Positive 210.82 229.82 −0.07 −208.03 170.97 12.22

a Promoters ending with a different number were derived from different PCR clones with either same or varied expected amplicon sizes.b Anthocyanin expression was visually scored 6 d after transformation.c Mean red brightness (MRB) data represent average values of two duplicated experiments. In each experiment, 10 anthocyanin-expressing foci each with a sample area

of 36 pixels were determined.d RMRB = (MRB + 19) for scale shift value derived from anthocyanin non-expressing SE background measurement.e A: RMRB-calculated optical density (RMCOD) value at 492 nm based on YA = −0.0131XA + 2.9436 (XA denotes RMRB value).f B: Histogram-calculated pigment quantity (HCPQ) value is derived from YB = 406.42XB − 180.76 (YB denotes HCPQ, XB refers to MCOD).g Adj-B: Adjusted HCPQ is derived from HCPQ + 379 (white background value).h SE represents standard error.

Quantitative analysis showed that up to three VvUb pro-moters generated activity levels reaching approximately 50%value of the DEAT promoter (Table 3 and Fig. 2, bars 11, 16and 19 vs. bar 1). Noticeably, two VvUb7 promoters, VvUb7-2and VvUb7-5 measuring 1 and 2 kb in size respectively showedthe highest expression levels (Fig. 2, bars 16 and 19), indicat-ing the adequacy of 1 kb promoter sequence for achieving astrong expression activity. However, two other 1 kb VvUb7 pro-moters from ‘Merlot’ (VvUb7-1 and VvUb7-3) and one from‘BN5-4’ (VvUb7-4) only produced less than 20% activity valueof the compared DEAT promoter (Fig. 2, bars 15, 17 and 18 vs.bar 1).

Among the four VvUb-5 promoters tested, a 1 kb pro-moter VvUb5-2 generated an activity up to 30–56% higherthan other three tested promoters (VvUb5-1, VvUb5-3 andVvUb5-4) (Fig. 2, bar 11 vs. bars 10, 12 and 13). In addi-tion, two 2 kb promoters (VvUb5-3 and VvUb5-4) both showeda lower activity level than the 1 kb version (VvUb5-2) (Fig. 2,bars 12 and 13 vs. bar 11), indicating the possible lack of

activity enhancement from sequences upstream of the 1 kbfragment.

Among other analyzed VvUb promoters, VvUb2-1 and VvUb4-1yielded about 15% activity value of the compared BDDP promoter(Fig. 2, bars 6 and 8 vs. bar 1), whereas VvUb6-1 produced an activitylevel of 35% of the compared DEAT promoter (Fig. 2, bar 14 vs. 1).

3.3. Analysis of transient GUS activity driven by VvUb promoters

To further confirm the fidelity of anthocyanin expression forpromoter activity analysis, seven VvUb promoters of 1 kb in sizefrom ‘Merlot’ or ‘Pinot Noir’ were introduced into the binary vec-tor pDesEGG-1 containing the GUS gene. Analysis of transformed SErevealed that the mean GUS activity ranged from 7549, 0, 167, 1126,195, 664, 1395, 7085 to 8498 pmol MU/min/mg protein for DAT-G(d35S), non-transformed control CK-G, VvUb1-1-G to VvUb7-2-G, respectively. When expression activities were converted intopercentage values of the DAT promoter, at least 5 promoters includ-ing VvUb1-1, VvUb2-1, VvUb3-1, VvUb6-1 and VvUb7-2 produced



Fig. 1. Transient anthocyanin expression in SE of grapevine (V. vinifera cv. ‘Thompson Seedless’) driven by various promoters 6 d after transformation. The number and nameindicated in each image panel corresponds to a particular promoter listed in Table 3. Digital images were taken microscopically using an identical aperture and lightingsetting. Scale bar represent 3 mm.

comparative GUS activities that followed a similar activity trendas found in anthocyanin analysis with corresponding promoters(Fig. 3, promoter constructs with an end label G vs A). Noticeably,even with the use of the same promoter clones, the GUS activity lev-els achieved from VvUb6-1 and VvUb7-2 were 50% and 22% higherthan the activity levels derived from anthocyanin analysis (Fig. 3,VvUb6-1-G vs. VvUb6-1-A and VvUb7-2-G vs VvUb7-2-A). Also itis worth noting that promoter VvUb7-2 reached more than 100% ofthe activity from DAT promoter (Fig. 3, VvUb7-2-G vs. DAT-G). Theremaining two promoters VvUb4-1 and VvUb5-1 generated a rela-tively lower GUS activity than the anthocyanin estimates (Fig. 3). In

spite of the observed differences, the similar activity trend with dif-ferent assay systems validated the results of quantitative promoteranalysis using the non-destructive anthocyanin marker.

3.4. Stable anthocyanin expression in transgenic tobacco

Transgenic tobacco plants containing 1 kb VvUb promoters andanthocyanin reporter gene were recovered. Although a low levelof pigment accumulation occurred in leaves of transgenic tobaccoplants that were maintained in culture vessels (data not shown),none of the transgenic tobacco plants containing VvUb promoters

Fig. 2. Quantitative comparisons of transient anthocyanin expression activity driven by various promoters of grapevine (V. vinifera) with a d35S-derived BDDP promoterDEAT. Color histogram data were collected from SE images taken 6 d after transformation. Mean activity values of test promoters were converted into percentage data overthat of DEAT promoter. Bar values were derived from two repeated experiments with standard errors indicated on top of each data bar.



Fig. 3. Quantitative comparisons of transient anthocyanin and GUS expression activity driven by various ubiquitin gene promoters of grapevine (V. vinifera) with a d35Spromoter DAT. Transient expression data were collected from SE samples 6 d after transformation. Color histogram-derived activity data represent mean values from tworepeated experiments. GUS enzymatic activity data correspond to average values from three repeated experiments. Standard error values were indicated.

Fig. 4. Localized anthocyanin expression in sepal and corolla of transgenic tobacco transformed with various promoter constructs. CK refers to non-transformed control.Plants with anthocyanin expression pattern representative of the promoter type were shown.

showed stable anthocyanin expression in leaf tissues after beingtransplanted and established in the greenhouse. At later growthstages, anthocyanin pigment reappeared in tissues of the inflores-cence in some plants. However, among all test promoters, onlytransgenic plants transformed with VvUb6-1 and VvUb7-2 pro-duced anthocyanin pigment in sepal and corolla. In addition, thestable expression activity of these two promoters in transgenicplants corroborated results of transient anthocyanin analysis ingrape SE as described earlier (Fig. 1 vs. Fig. 4).

3.5. Analysis of cis-acting elements present in VvUb promoters

DNA sequences of all 1 kb VvUb promoters identified from the‘Pinot Noir’ genome (see Table 1), along with several previously-characterized ubiquitin gene promoters of other species wereanalyzed to determine sequence correlativeness. Phylogeneticanalysis indicated that VvUb6 and VvUb7 showed the highestcladal relatedness with 100% bootstrap support (Fig. 5). These twopromoters were also significantly related to UBQ3 promoter ofArabidopsis. On the other hand, other VvUb promoters seemed tohave a higher degree of cladal diversity (Fig. 5). Accordingly, phy-logenetic correlation corroborated genomic locations of analyzedVvUb promoters (Table 1). Both VvUb6 and VvUb7 were identifiedfrom chromosome 19, whereas VvUb-1, VvUb-2, VvUb-3, VvUb-4, and VvUb-5 were derived from chromosomes 4, 7, 8, 8 and 16,respectively.

Fig. 5. Phylogenetic analysis of 1 kb DNA sequences of ubiquitin gene promotersfrom grapevine (V. vinifera cv. ‘Pinot Noir’) and other plant species. DNA sequencesof grapevine promoters were retrieved from the genome of ‘Pinot Noir’. Promoters ofother species were described and referenced in the text. The Neighbor-Joining (NJ)method with 2000 replications was used for dendrogram reconstruction. Numbersassociated with each node represent bootstrap values. A scale of substitution ratewas shown.



Table 4Analyses of sequence structure and cis-acting elements for transcription up-regulation and root-specific expression from cloned 1 kb ubiquitin gene promoters of grapevine(V. vinifera) and rice (O. sativa).

Promoter VvUb1-1 VvUb2-1 VvUb3-1 VvUb4-1 VvUb5-1 VvUb6-1 VvUb7-2 Rice Ub1 Rice Ub2

Sequencing analysisSource Merlot Pinot N. Merlot Merlot Merlot Merlot Merlot Rice RiceSequenced promoter size (bp) 1058 1000 899 981 1007 1000 1000 1000 1000Single nucleotide substitution 7 2 60 5 7 8 11Deletion event (Total size in bp) 0 0 4 (106) 2 (22) 0 0 0Insertoin event (Total size in bp) 1 (58) 0 1 (7) 3 (3) 1 (7) 0 0

Analysis of cis-acting elementsa in cloned and genome-derived ubiquitin gene promotersb

ABRELATERD1 (ACGTG) 0 (0) 0 (0) 1 (1) 0 (1) 2 (2) 2 (2) 3 (3) 0 2LTRE1HVBLT49 (CCGAAA) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 1 (1) 1 (1) 0 0MYBCOREATCYCB1 (AACGG) 0 (0) 1 (1) 0 (0) 0 (0) 2 (2) 3 (3) 1 (1) 2 4PYRIMIDINEBOXOSRAMY1A (CCTTTT) 0 (0) 1 (1) 0 (0) 1 (1) 1 (1) 3 (3) 2 (2) 0 2WRKY71OS (TGAC) 2 (1) 2 (2) 1 (1) 1 (1) 4 (4) 2 (2) 2 (2) 1 3Total positive elements 2 (1) 4 (4) 2 (2) 2 (3) 9 (9) 11 (11) 9 (9) 3 11ROOTMOTIFTAPOX1 (ATATT) 7 (6) 6 (6) 21 (15) 1 (1) 1 (2) 1 (1) 2 (2) 4 0

a Sequence motif and functions of cis-acting elements are described by Higo et al. [25] and on website: http://www.dna.affrc.go.jp/PLACE/. Search for target motif sequenceswas performed using Vector NTI software program.

b Element numbers in parentheses are derived from 1 kb ubiquitin gene promoters from the ‘Pinot Noir’ genome.

To further evaluate sequence structure-function characteristics,the distribution of cis-acting regulatory elements within VvUb pro-moters derived from both sequenced selected clones and the ‘PinotNoir’ genome were determined using a plant specific cis-actingelement database PLACE version 30.0 [25] (last update February,2007). As summarized in Table 4, the number of cis-acting elementsthat are often associated with general expression enhancement orup-regulation in response to stress showed a high to low order fromsequenced promoters VvUb6-1 (11), VvUb7-2 (9), VvUb5-1 (9),VvUb2-1 (4), VvUb4-1 (2), VvUb3-1 (2) to VvUb1-1 (2). This orderrange coincided with the high to low range of expression activityamong these promoters (Figs. 2 and 3). On the other hand, an oppo-site high to low number range of a root-specific cis-acting elementwas found from promoters VvUb3-1 (21), VvUb1-1 (7), VvUb2-1 (6),VvUb7-2 (2), VvUb4-1 (1), VvUb5-1 (1), to VvUb6-1 (1). Thus, theincrease in the number of root-specific cis-acting elements seemedassociated with a reduced expression activity for tested VvUb pro-moters. Noticeably, the profile of cis-activing elements was nearlyidentical between sequenced promoters and corresponding pro-moters derived from the ‘Pinot Noir’ (Table 4, lower section). Inaddition, analysis of similar cis-acting elements revealed that aubiquitin gene promoter UB2 of rice (Orysa sativa) with report-edly a 15–48-fold increase of GUS activity over a sinlge CaMV35S promoter in transgenic plants also contained a cis-acting ele-ment profile similar to that of strong VvUb6-1 promoter, whereasa relatively weaker rice UB1 promoter showed a profile closelyresembling that of a relatively weaker VvUb2-1 promoter (Table 4)[26].

4. Discussion

Over the years, progress in exploitation of native promoters forgenetic manipulation of grapevine has been slow. A limited num-ber of grapevine promoters were described previously but theywere all associated with developmental and environmental regula-tion and tissue-specific expression. For instance, a 2S albumin genepromoter was isolated and demonstrated to confer seed-specificexpression with responses to ABA up-regulation [19]. Two alco-hol dehydrogenase gene promoters along with leader sequenceswere shown to have expression activities higher than a CaMV 35Spromoter in response to various conditions of anaerobiosis and atstages of fruit ripening [27]. A stilbene synthase gene promoterwas found to contain both ozone and pathogen-responsive regions[28,29]. To our knowledge, there has been no report about the isola-tion and characterization of constitutive promoters from grapevine.

In this study, a larger number of grapevine promoters were clonedand examined for transcription activity using both anthocyaninand GUS-based reporter systems in transient and stable expres-sion experiments. These promoters were derived from 5′ upstreamsequences of known genes with various size ranges relative to thetranscription start site and isolated from different varieties. Ouranalysis indicated that the majority of these promoters were capa-ble of activating transcription in all tissue types of SE with a widerange of activity levels.

A number of VvUb promoters showed relatively high levels oftranscription activity. Ubiquitin functions as a carrier molecule inthe ubiquitin-proteasome pathway for protein degradation and isencoded by a multigene family with identical amino acid sequencein duplicated units in eukaryotes [30]. Each ubiquitin gene is con-trolled by an independent and primarily constitutive promoter.Evolution of the ubiquitin multigene family is thought to follow abirth-and-death process, meaning that new members are evolvedvia repeated duplications of the ancestor gene and that amongnewly evolved members some remain functional while othersmay become nonfunctional or pseudogenes [31]. In addition, DNAmutations within the promoter region can also lead to loss of tran-scriptional activation and silence of a gene. This might explain whysome of the VvUb promoters including VvUb1-1 and VvUb3-1 areinactive. Accordingly, it is imperative to perform expression analy-sis to determine the activity of each ubiquitin gene promoter priorto utilization in transformation studies.

To gain insight into the sequence relation, phylogenetic analy-sis was conducted with ubiquitin gene promoters from ‘Pinot Noir’of grapevine and other species. Results indicated that the majorityof the compared VvUb promoters did not show a clear cladal rela-tion with each other whereas VvUb6 and VvUb7 formed a singleclade at 100% confidence level. In addition, these two promoterswere also found to be syntenic on the same chromosome. We spec-ulate that these two highly active promoters were recently evolvedvia gene duplication and might be responsible for the constitutiveexpression of ubiquitin proteins in grape cells. On the other hand,other VvUb gene members might have evolved much earlier andsubjected to increased diversification [32].

Sequencing analysis indicated that all corresponding VvUb pro-moters derived from ‘Merlot’ harbored a half dozen of sporadicsingle nucleotide substitutions. However, no sequence deletionor insertion events occurred in strong promoters VvUb6-1 andVvUb7-2, whereas other relatively weaker or inactive promoterscontained deletion/insertion mutations (Table 4). Noticeably, theinactive promoter VvUb3-1 contained the highest single nucleotide



polymorphism (SNP) frequency and longest combined deletionsequences, suggesting the relative ancestral status of the promoter.Thus, our DNA sequence analysis supported the premise that theVvUb6 and VvUb7 promoters identified herein are likely to beassociated with recently evolved gene members and playing animportant role in directing active expression of ubiquitin proteins.

Sequencing analysis could also shed light on the possiblecause(s) for expression activity variation among different PCRclones of the same promoter type. For instance, the VvUb7-4promoter with an activity level of about 35% of VvUb7-2 wassequenced and found to harbor sequence mutations leading to theloss of a unique MYBCOREATCYCB1 (AACGG) element (data notshown). Whether similar sequence mutations associated with loss-of-function might also occur in other tested promoters needs tobe substantiated with more sequencing efforts. In addition, cau-tion must be exercised before concluding that these mutationsare derived from PCR amplification or preexisting in the templategenome.

It should be pointed out that VvUb1-1 harbored a singleinsertion of 58 bp as compared to the ‘Pinot Noir’ genome (12×)-derived counterpart (Table 4). However, sequence homology searchshowed that a sequence contig VV78X192609.45 or GenBankaccession AM479097.1 of ‘Pinot Noir’ clone ENTAV 115 actuallycontained the identical insertion sequence. The sequence discrep-ancy between the grape genome assembly and the contig fragmentremains to be determined.

Promoter activity is the culmination of intricate interplaybetween cis-acting elements and corresponding transcription fac-tors. Sequence analysis of ubiquitin gene promoters from grapevineand rice revealed a positive correlation of higher promoter activitywith increased number of cis-acting elements commonly associ-ated with expression activation and enhancement. Interestingly,the lack of constitutive activity was found to be negatively cor-related with increased number of root-specific cis-acting element[33–35]. Whether inactive VvUb promoters are actually involvedin expression activation in root tissues need to be substantiated.Nevertheless, the distribution of cis-acting elements within pro-moter region might facilitate the delineation of functional regionand provide a better predictor of expression activity of ubiquitingene promoters than simple sequence alignment analysis.

A number of ubiquitin genes in various species contained aunique intron in the 5′ UTR region. These introns were previouslyfound to play an important role in transcription enhancement.In Arabidopsis, intron-containing 5′ UTR sequences from at leastthree ubiquitin genes provided up to a 3-fold increase in transientmarker gene expression [36]. In rice, a 1140 bp 5′ UTR intron ofthe rubi3 gene was fused to a number of relative weak promoters.The resulting constructs led to up to 20-fold increase of transientGUS expression [37]. Although it remains to be elucidated in plants,such transcriptional enhancement by 5′ UTR introns of ubiquitingenes might be related to the formation of exon junction com-plex that enhances translation efficacy as observed in animal cells[38]. Our sequence analysis results using the ‘Pinot Noir’ genomeindicated VvUb genes harbored 5′UTR intron structures similar tothat of Arabidopsis. All VvUb1, VvUb2, VvUb6 and VvUb7 genescontain a 5′ UTR intron in relatively small sizes based on splicingsite analysis and cDNA information [36]. However, the expressionenhancement function of VvUb 5′UTR introns remained unclear.On the other hand, all VvUb promoters displayed expression activ-ities that seemed non-related to the presence of 5′UTR introns. Forinstance, promoter VvUb3-1 harbored an appropriately-positionedconserved TATA box (TATATAA) and a transcription initiation site(CTCACT) 50 and 31 nucleotides respectively upstream from thetranslation start codon. Yet this promoter failed to activate antho-cyanin and GUS expression. Also, the intronless promoter VvUb5-2supported an expression level similar to that of intron-containing

promoters VvUb6 and VvUb7. Thus, the expression activity of VvUbpromoters appeared to be dictated primarily by genetic elementsin upstream sequences of the intron.

Two promoters VvUb6-1 and VvUb7-2 were highly active, sup-porting transient anthocyanin and GUS expression at levels as highas that of a double enhanced 35S promoter (DAT). These two pro-moters were also capable of activating anthocyanin expression infloral tissues of stably transformed tobacco plants in the similar pat-tern conferred by the d35S-derived BDDP promoter DEAT [16]. Ourfindings are in excellent agreement with previous studies wherehighly active ubiquitin gene promoters were also identified fromvarious monocotyledonary and dicotyledonary species [26,39–42].Some of these promoters were successfully utilized to direct hightransgene expression in transgenic plants [43,44]. Thus, the cur-rent study provides a good starting point for further exploitationand utilization of promoters VvUb6 and VvUb7 to support consti-tutive expression of native genes in the development of cisgenicsof grapevine.

Among the promoters analyzed, two ‘Merlot’-derived VvAlb1promoters with a DNA size 1.8 kb and 2.1 kb, respectively, con-ferred anthocyanin expression strictly localized in cotyledonarytissues. This expression pattern is similar to the previous find-ing that a 0.6 kb fragment of the same promoter directed EGFPexpression in cotyledonary tissue [19]. Accordingly, regardless ofthe promoter sizes used, gene expression was limited to cotyle-donary cells. However, expression levels from all tested promoterfragments remained relatively low as compared to viral promoterseven though production of anthocyanin was readily detectable. Onthe other hand, we observed previously that high levels of antho-cyanin accumulation in vegetative tissues of transgenic grapevinecould lead to reduced vigor and abnormal plant development [16].Thus, the VvAlb1 promoters could offer an alternative tool todrive visible/selectable marker expression in desired explant tissue.For instance, this promoter when linked to a native anthocyanin-promoting gene VvMybA1 can be used in studies where cotyledonsare used as target tissue for transformation and recovery of cisgen-ics. Transformed plants, once recovered, will expectedly have nogrowth-interfering anthocyanin overexpression and accumulationin vegetative tissues.

We also tested a number of PR1 gene promoters for their con-stitutive activity. Many PR1 genes in plants are often activatedby elicitors associated with pathogens, abiotic and environmentalstresses [45–47]. Although it was reported that a 300 bp upstreamfragment of a corresponding PR1 gene was sufficient to conferinducible expression, the basal activity somewhat was often cor-related with inducible activity and the promoter sizes tested [48].For instance, Eyal et al. [49] demonstrated that a 553 bp PRB-1b promoter fragment produced a significantly lower basal andinduced expression activity than an 863 bp fragment of the samepromoter. Both constitutive and inducible cis-elements were iden-tified in certain PR1 promoters using unique DNA-binding analysismethods [50]. Thus, the basal expression activity of PR1 gene pro-moters appears to be positively correlated with and could be usedto predict the inducible expression activity. In the current study, anumber of VvPR1 promoters with a DNA size of 555 bp or 1555 bpwere analyzed to reveal basal activity. Two 555 bp promotersVvPR1-1 and VvPR1-3 generated a basal activity higher than othertested promoters. It needs to be substantiated whether such ele-vated basal activity reflects a possible higher inducible expressionactivity among these promoters. Promoters with strong inducibleexpression activity could be used to drive native resistance-relatedgenes.

Quantitative analysis of promoter activity was traditionallyperformed by using reporter genes such as GUS and LUC. Theseassay systems, however, often require the destruction and harshtreatments of explants and carry many extraneous variables that



are difficult to identify. For instance, factors including explanthomogeneity, cell growth and expression capacity, processingtreatments and homogenate conditions, interference by endoge-nous activity, assay sample dilution and measurement, and soforth can affect the accuracy of activity estimates [51,52]. Inaddition, enzymatic assay procedures are often laborious and time-consuming and can severely limit the number of samples to beanalyzed simultaneously. In this report, we demonstrated theversatile utility of a previously developed non-destructive colorhistogram-based image analysis of anthocyanin expression for thesimultaneous activity quantification of a large number of promot-ers [16]. Activity levels of VvUb promoters determined based on theanthocyanin reporter system were comparable with results derivedfrom GUS activity assay, indicating the sensitivity and reliabilityof anthocyanin method. Hence, the non-destructive anthocyaninreporter system should offer a useful tool for high-throughput eval-uation of plant promoters.

Acknowledgements

This research was supported in part by the Florida AgriculturalExperiment Station, the USDA/NIFA Specialty Crops Research Ini-tiative and the Florida Department of Agriculture and ConsumerServices’ Viticulture Trust Fund. The authors thank Mr. MatthewMcKinley and Ms. Angela Adamitis, Emma Grubb, Marlene Saldivar,and Natalia Cimino of MREC for their excellent technical support.

References

[1] A. Konig, A framework for designing transgenic crops-science, safety and citi-zen’s concerns, Nat. Biotechnol. 21 (2003) 1274–1279.

[2] J.L. Lusk, A. Rozan, Consumer acceptance of ingenic foods, Biotechnol. J. 1 (2006)1433–1434.

[3] H. Schouten, F.A. Krens, E. Jacobsen, Cisgenic plants are similar to traditionallybred plants, EMBO Rep. 7 (2006) 750–753.

[4] C.M. Rommens, M.A. Haring, K. Swords, H.V. Davies, W.R. Belknap, The intra-genic approach as a new extension to traditional plant breeding, Trends PlantSci. 12 (2007) 397–403.

[5] D.J. Gray, S. Jayasankar, Z. Li, Vitaceae (Grape Vitis spp.), in: R.E. Litz (Ed.),Biotechnology of Fruit and Nut Crops, Biotechnology in Agriculture Series, No.29, CAB International Wallingford, UK, 2005, pp. 672–706 (Chapter 22).

[6] Z.T. Li, S.A. Dhekney, M. Dutt, D.J. Gray, An improved protocol for Agrobacterium-mediated transformation of grapevine (Vitis vinifera L.), Plant Cell Tissue OrganCult. 93 (2008) 311–321.

[7] O. Jaillon, J.M. Aury, B. Noel, et al., The grapevine genome sequence sug-gests ancestral hexaploidization in major angiosperm phyla, Nature 449 (2007)463–467.

[8] S.A. Dhekney, Z.T. Li, M.E. Compton, D.J. Gray, Optimizing initiation and main-tenance of Vitis embryogenic cultures, HortScience 44 (2009) 1400–1406.

[9] T.J. Guilfoyle, The structure of plant promoters, Genet. Eng.: Princ. Methods 19(1997) 15–48.

[10] C. Potenza, L. Aleman, C. Sengupta-Gopalan, Targeting transgene expressionin research, agricultural, and environmental applications: promoters used inplant transformation, In Vitro Cell. Dev. Biol. – Plant 40 (2004) 1–22.

[11] R. Scorza, J.M. Cordts, D.J. Gray, D. Gonsalves, R.L. Emershad, D.W. Ramming,Producing transgenic ‘Thompson Seedless’ grape (Vitis vinifera L.) plants, J. Am.Soc. Hortic. Sci. 121 (1996) 616–619.

[12] T. Yamamoto, H. Iketani, H. Ieki, Y. Nishizawa, K. Notsuka, T. Hibi, T. Hayashi, N.Matsuta, Transgenic grapevine plants expressing a rice chitinase with enhancedresistance to fungal pathogens, Plant Cell Rep. 19 (2000) 639–646.

[13] Z. Li, S. Jayasankar, D.J. Gray, Expression of a bifunctional green fluorescent pro-tein (GFP) fusion marker under the control of three constitutive promoters andenhanced derivatives in transgenic grape (Vitis vinifera), Plant Sci. 160 (2001)877–887.

[14] Z.T. Li, S. Jayasankar, D.J. Gray, Bi-directional duplex promoters with duplicatedenhancers significantly increase transgene expression in grape and tobacco,Transgenic Res. 13 (2004) 143–154.

[15] S. Dhekney, Z.T. Li, D.J. Gray, Grapevine engineered to express cisgenic Vitisvinifera thaumatin-like protein exhibit fungal disease resistance, In Vitro Cell.Dev. Biol. – Plant 47 (2011) 458–466.

[16] Z.T. Li, S.A. Dhekney, D.J. Gray, Use of the VvMybA1 gene for non-destructivequantification of promoter activity via color histogram analysis in grapevine(Vitis vinifera) and tobacco, Transgenic Res. 20 (2011) 1087–1097.

[17] M.D. Burrow, C.A. Chlan, P. Sen, N. Murai, High frequency generation of trans-genic tobacco plants after modified leaf disk cocultivation with Agrobacteriumtumefaciens, Plant Mol. Biol. Rep. 8 (1990) 124–139.

[18] Z.T. Li, S.A. Dhekney, D.J. Gray, PR-1 gene family of grapevine: a uniquely dupli-cated PR-1 gene from a Vitis interspecific hybrid confers high level resistanceto bacterial disease in transgenic tobacco, Plant Cell Rep. 30 (2011) 1–11.

[19] Z.T. Li, D.J. Gray, Isolation by improved thermal asymmetric interlaced PCR andcharacterization of a seed-specific 2S albumin gene and its promoter from grape(Vitis vinifera L.), Genome 48 (2005) 312–320.

[20] K. Tamura, D. Peterson, N. Peterson, G. Stecher, M. Nei, S. Kumar, MEGA5:molecular evolutionary genetics analysis using maximum likelihood, evolu-tionary distance, and maximum parsimony methods, Mol. Biol. Evol. 28 (2011)2731–2739.

[21] S.B. Hedges, The number of replications needed for accurate estimation of thebootstrap P value in phylogenetic studies, Mol. Biol. Evol. 9 (1992) 366–369.

[22] J. Felsenstein, Confidence limits on phylogenies: an approach using the boot-strap, Evolution 39 (1985) 783–791.

[23] M.A. Lodhi, G.N. Ye, N.F. Weeden, B.J. Reisch, A simple and efficient method forDNA extraction from grapevine cultivars and Vitis species, Plant Mol. Biol. Rep.12 (1994) 6–13.

[24] R.A. Jefferson, Assaying chimeric genes in plants: the GUS gene fusion system,Plant Mol. Biol. Rep. 5 (1987) 387–405.

[25] K. Higo, Y. Ugawa, M. Iwamoto, T. Korenaga, Plant cis-acting regulatory DNAelments (PLACE) database: 1999, Nucleic Acids Res. 27 (1999) 297–300.

[26] J. Wang, J. Jiang, J.H. Oard, Structure, expression and promoter activity of twopolyubiquitin genes from rice (Oryza sativa L.), Plant Sci. 156 (2000) 201–211.

[27] C. Verries, M. Pradal, P. Chatelet, L. Torregrosa, C. Tesniere, Isolation and analysisof the promoter of VvAdh2, a grapevine (Vitis vinifera L.) ripening-related gene,Plant Sci. 167 (2004) 1067–1074.

[28] R. Schubert, R. Fischer, R. Hain, P.H. Schreier, G. Bahnweg, D. Ernst, H. San-dermann Jr., An ozone-responsive region of the grapevine resveratrol synthasepromoter differs from the basal pathogen-responsive sequence, Plant Mol. Biol.34 (1997) 417–426.

[29] B. Grimmig, M.N. Gonzalez-Perez, G. Welzl, J. Penuelas, R. Shubert, R. Hain, B.Heidenreich, C. Betz, C. Langebartels, D. Ernst, H. Sandermann Jr., Enthylene-and ozone-induced regulation of a grapevine resveratrol synthase gene: differ-ent responsive promoter regions, Plant Physiol. Biochem. 40 (2003) 865–870.

[30] M. Nei, I.B. Rogozin, H. Piontkivska, Purifying selection and birth-and-deathevolution in the ubiquitin gene family, Proc. Nat. Acad. Sci. U.S.A. 97 (2000)10866–10871.

[31] M. Nei, X. Gu, T. Sitnikova, Evolution by the birth-and-death process in multi-gene families of the vertebrate immune system, Proc. Nat. Acad. Sci. U.S.A. 94(1997) 7799–7806.

[32] M. Nei, A.P. Rooney, Concerted and birth-and-death evolution of multigenefamily, Annu. Rev. Genet. 39 (2005) 121–152.

[33] C. Hertig, G. Rebmann, J. Bull, F. Mauch, R. Dudler, Sequence and tissue-specificexpression of a putative peroxidase gene from wheat (Triticum aestivum L.),Plant Mol. Biol. 16 (1991) 171–174.

[34] T. Elmayan, M. Tepfer, Evaluation in tobacco of the organ specificity andstrength of the rolD promoter, domain A of the 35S promoter and the 35S2

promoter, Transgenic Res. 4 (1995) 388–396.[35] K. Xiao, J. Liu, G. Dewbre, M. Harrison, Z.Y. Wang, Isolation and characterization

of root-specific phosphate transport promoters from Medicago truncatula, PlantBiol. 8 (2006) 439–449.

[36] S.R. Norris, S.E. Meyer, J. Callis, The intron of Arabidopsis thaliana polyubiquitingenes is conserved in location and is a quantitative determinant of chimericgene expression, Plant Mol. Biol. 21 (1993) 895–906.

[37] E. Sivamani, R. Qu, Expression enhancement of a rice polyubiquitin gene pro-moter, Plant Mol. Biol. 60 (2006) 225–239.

[38] A. Nott, H. Le Hir, M.J. Moore, Splicing enhances translation in mammalian cells:an additional function of the exon junction complex, Genes Dev. 18 (2004)210–222.

[39] J. Callis, J.A. Raasch, R.D. Vierstra, Ubiquitin extension proteins of Arabidopsisthaliana structure, localization, and expression of their promoters in transgenictobacco, J. Biol. Chem. 265 (1990) 12486–12493.

[40] J.E. Garbarino, W.R. Belknap, Isolation of a ubiquitin-ribosomal protein gene(Ubi3) from potato and expression of its promoter in transgenic plants, PlantMol. Biol. 24 (1994) 119–127.

[41] Y.H. Joung, K. Kamo, Expression of a polyubiquitin promoter isolated formGladiolus, Plant Cell Rep. 25 (2006) 1081–1088.

[42] J. Lu, E. Sivamani, X. Li, Activity of the 5’ regulatory regions of the rice polyu-biquitin rubi3 gene in transgenic rice plants as analyzed by both GUS and GFPreporter genes, Plant Cell Rep. 27 (2008) 1587–1600.

[43] L. Rooke, D. Byrne, S. Salgueiro, Marker gene expression driven by themaize ubiquitin promoter in transgenic wheat, Ann. Appl. Biol. 136 (2000)167–172.

[44] D. Liu, S.V. Oard, J.H. Oard, High transgene expression levels in sugarcane (Sac-charum officinarum L.) driven by the rice ubiquitin promoter RUBQ2, Plant Sci.165 (2003) 743–750.

[45] Y. Xu, P.L. Chang, D. Liu, M.L. Narasimhan, K.G. Raghothama, P.M. Hasegawa,R.A. Bressan, Plant defense genes are synergistically induced by ethylene andmethyl jasmonate, Plant Cell 6 (1994) 1077–1085.

[46] M. Santamaria, C.J. Thomson, N.D. Read, G.J. Loake, The promoter of a basic PR1-like gene AtPRB1, from Arabidopsis established an organ-specific expressionpattern and responsiveness to ethylene and methyl jasmonate, Plant Mol. Biol.47 (2001) 641–652.

[47] J.K. Hong, S.C. Lee, B.K. Hwang, Activation of pepper basic PR-1 gene promoterduring defense signaling to pathogen, abiotic and environmental stresses, Gene356 (2005) 169–180.



[48] M. Ohshima, H. Itoh, M. Matsuoka, T. Murakami, Y. Ohashi, Analysis of stress-induced or salicylic acid-induced expression of the pathogenesis-related 1aprotein gene in transgenic tobacco, Plant Cell 2 (1990) 95–106.

[49] Y. Eyal, Y. Meller, S. Lev-Yadun, R. Fluhr, A basic-type PR-1 promoter directsethylene responsiveness, vascular and abscission zone-specific expression,Plant J. 4 (1993) 225–234.

[50] I. Meier, K. Hahlbrock, I.E. Somssich, Elicitor-inducible and constitutive in vivoDNA footprints indicate novel cis-acting elements in the promoter of a

parsley gene encoding pathogenesis-related protein 1, Plant Cell 3 (1991)309–315.

[51] S. Vitha, K. Benes, M. Michalova, M. Ongrej, Quantitative, �-glucuronidase assay in transgenic plants, Biol. Plant. 35 (1993)151–155.

[52] C. Cote, R.G. Rutledge, An improved MUG fluorescent assay for the determina-tion of GUS activity within transgenic tissue of woody plants, Plant Cell Rep. 21(2003) 619–624.

Documents

(This is a sample cover image for this issue. The actual ...mrec.ifas.ufl.edu/grapes/genetics/Selected-Publications/B PlantSci... · plants via the cisgenic/intragenic/ingenic approach