www.elsevier.com/locate/plaphy

Plant Physiology and Biochemistry 44 (2006) 766–775

Research article

Isolation and characterization of a novel potatoAuxin/Indole-3-Acetic Acid family member (StIAA2)

that is involved in petiole hyponasty and shoot morphogenesis

B. Kloosterman*, R.G.F. Visser, C.W.B. Bachem

Laboratory of Plant Breeding, Department of Plant Sciences, Graduate School of Experimental Plant Sciences, Wageningen University and Research Center, P.O.Box 386, 6700 AJ Wageningen, The Netherlands

Received 7 March 2006; accepted 10 October 2006Available online 03 November 2006

Abstract

Auxin/indole-3-acetid acid (Aux/IAA) proteins are short-lived transcriptional regulators that mediate their response through interaction withauxin response factors (ARF). Although 29 Aux/IAA proteins have been identified in Arabidopsis thaliana, their individual functions are stillpoorly understood and are largely defined by observed growth defects in gain-of-function mutant alleles. Here we present the isolation andcharacterization of a novel Aux/IAA protein in potato (Solanum tuberosum) that is named StIAA2. Down regulation of StIAA2 results in dis-tinctive phenotypes that include, increased plant height, petiole hyponasty and extreme curvature of growing leaf primordia in the shoot apex.Gene expression analysis of transgenic plants with reduced StIAA2 transcript levels resulted in the identification of a number of genes withaltered expression profiles including another member of the Aux/IAA gene family (StIAA). The phenotypes that were observed in the StIAA2suppression clones can be associated with both common as well as unique functional roles among Aux/IAA family members indicating theimportance of analyzing Aux/IAA expression in different plant species.© 2006 Elsevier Masson SAS. All rights reserved.

Keywords: Aux/IAA protein; Solanum tuberosum; Hyponasty; Shoot morphogenesis

1. Introduction

It is well known that auxins, mainly represented by indole-3-acetic acid (IAA), are important in plant growth and devel-opmental processes through the regulation of auxin responsivegene expression [1]. In recent years, the regulatory componentsof auxin signaling have become more evident and haverevealed the existence of a highly complex system of bothearly and late auxin responses (reviewed in [2–4]). Severalgenes involved in the regulation of auxin dependent transcrip-tion have been identified in which primary roles have been

Abbreviations : ARF, auxin response factor ; Aux/IAA, auxin/indole-3-acetic acid ; LD, long day conditions ; NTS, non-tuberizing stolon tip ; qRT-PCR, quantitative reverse transcriptase-polymerase chain reaction ; SD, shortday conditions.

* Corresponding author. Tel.: +31 317 48 4875; fax: +31 317 48 3457.E-mail addresses: bjorn.kloosterman@wur.nl (B. Kloosterman),

richard.visser@wur.nl (R.G.F. Visser), christian.bachem@wur.nl(C.W.B. Bachem).

0981-9428/$ - see front matter © 2006 Elsevier Masson SAS. All rights reserved.doi:10.1016/j.plaphy.2006.10.026

given to the auxin responsive factors (ARF) [5] and Aux/IAAprotein family members [6].

ARF proteins can function as either transcriptional activa-tors or repressors that can bind to auxin responsive elements(AuxREs) found in promoters of auxin responsive genesthrough a specific DNA binding domain [7]. The Aux/IAAproteins are short-lived transcription factors that contain fourhighly conserved domains (referred to as domain I, II, III andIV). Aux/IAA proteins are thought to act as regulators ofauxin-induced gene expression through heterodimerization ofdomains III and IV with ARF proteins thereby modifyingARF activity [3,8,9]. In this model, auxin regulates transcrip-tion by stimulating the degradation of the Aux/IAA proteinsthrough interaction of the Aux/IAA protein with the ubiquitinprotein ligase SCFTIR1 complex [10].

Given that in Arabidopsis 29 different Aux/IAA and 23ARF proteins have been identified to date, the number ofpotential Aux/IAA–ARF interactions and the subsequentauxin responses possible in the different tissues of a plant areenormous.

B. Kloosterman et al. / Plant Physiology and Biochemistry 44 (2006) 766–775 767

Hence, elucidation of individual Aux/IAA gene functionshas proven difficult since different mutant lines often exhibitunique as well as overlapping auxin related plant growthdefects. Arabidopsis mutant screens have identified a numberof Aux/IAA ‘gain-of-function’ mutants where a mutation indomain II yields a more stable protein that is less sensitive todegradation [11,12]. These mutant lines confer dramatic auxin-related developmental defects, including altered gravitropismand apical dominance in axr2/IAA7 [13], and axr3/IAA17[12], lateral root defects in IAA28 [14] and IAA14 [15], andphotomorphogenic defects in shy2/IAA3 [16]. In contrast,Overvoorde et al. [17], analyzed insertion mutation mutants,representing a loss-of-function mutation, in 12 of the 29 iden-tified Aux/IAA proteins in Arabidopsis, but failed to show visi-ble developmental defects in comparison to the wild type.These findings suggest a broad functional redundancy amongthe Aux/IAA gene family members making identification ofbiological function complex.

However, the analysis of Aux/IAA gene family members inother plant species may prove valuable when attempting tounderstand both common as well as process specific auxindependent gene regulation. For example, the down regulationof a single Aux/IAA family member in tomato (IAA9) results ina dramatic pleiotropic phenotype, having simple leaves insteadof wild type compound leaves and early fruit development,giving rise to parthenocarpy [18].

Auxins have long been implicated to play a regulatory rolein potato (Solanum tuberosum L.) tuber development. Endo-genous auxin levels were found to be high just prior and duringstolon swelling after which auxin levels gradually decreased[19]. Furthermore, Xu et al. [20], observed earlier tuberizationwhen IAA was applied to single node cuttings in tuber-inducing medium. On the other hand, tuber formation wascompletely inhibited by high concentrations of IAA [21]. Inpotato, only a single Aux/IAA (StIAA, [22]) and ARF protein(ARF6, [23]) have been described to date. StIAA expressionlevels increased after fungal infection, wounding or applicationof auxin [22]. Arf6 expression levels are reduced in the apicalmeristem of the stolon tip at tuber onset and growth, andinduced during meristem activation in dormant tuber buds[23]. The degree of auxin dependent gene regulation of devel-opmental processes through the action of other Aux/IAA andARF family members in potato plants is however still largelyunknown. In this paper we describe the isolation and character-ization of a novel potato Aux/IAA using a reverse geneticsapproach and discuss its potential functions in potato plantgrowth and development.

2. Results

2.1. Isolation of a novel potato Aux/IAA gene family member(StIAA2)

In a previous study we analyzed gene expression changesduring early potato tuber development [24]. Within this studya number of genes were identified that showed a dramatic

down regulation in transcript levels during early tuberizationevents. One of these sequences showed strong homology tomembers of the Aux/IAA gene family. Sequencing of potatocDNA clone (gb|BI179192), provided a full length sequencewith a predicted ORF of 213 amino acids and a molecularweight of 23.9 kDa (isoelectric point (pI); 9.14). Sequencecomparisons between previously identified and characterizedAux/IAA proteins from a number of plant species reveal theexistence of Aux/IAA subgroups (Fig. 1A). The identifiedpotato StIAA2 protein is closely positioned on a branchtogether with Aux/IAA proteins from Arabidopsis thalianaIAA14, IAA7, IAA17, IAA16; Nicotiana tabacum IAA28and poplar (Populus tremula × Populus tremuloides IAA6,and IAA7). Alignments of the StIAA2 protein sequence withthese highly homologous Aux/IAA proteins are presented inFig. 1B and clearly show the presence of conserved domainsI–IV. Overall sequence homology of the predicted protein withthe previously identified potato Aux/IAA protein (StIAA; 349amino acids) reaches only 43%, with a significant higherhomology in the C-terminal end spanning domains III and IV(77%). Based on its sequence homologies with other Aux/IAAproteins, the predicted protein was classified as a novel potatoAux/IAA family protein and was named StIAA2.

In order to assess the expression profile of StIAA2 in potatoplants, quantitative RT-PCR (qRT-PCR) was performed on aset of different potato tissues (Fig. 2). Transcript levels werefound to be highest in the petiole, non-tuberizing stolon(NTS) tip, shoot apex, leaf and stem. Lower levels of expres-sion were detected in root, tuber and stolon. Very low expres-sion of StIAA2 was detected in dormant tubers stored at 4 °C.

2.2. Differential expression during potato tuber development

Since auxins are thought to play a regulatory role in theformation of a tuber, StIAA2 expression levels were measuredduring eight tuber developmental stages (Section 4). A subsetof the different developmental stages that were harvested arerepresented in Fig. 3A and include; NTS tip grown underlong day conditions (LD; 16 h), NTS under tuber-inducingshort day conditions (SD; 8 h), tuber initiation and subsequenttuber growth stages. Expression levels were calculated andplotted relative to an internal standard and the first harvesttime point (ΔΔCt) as described in Section 4. After the firstharvest under LD conditions, plants were transferred to SDconditions (T = 0 days) and subsequent harvest stages arerepresented on the x-axis as days after the switch from LD toSD. A sharp decline of StIAA2 transcript levels was observed atthe time point of tuber organogenesis (Fig. 3B; StIAA2 stages 3and 4; ΔΔCt –2.4 and ΔΔCt –3.1, respectively) indicating stricttranscriptional control during these stages. To check for a com-mon expression pattern of Aux/IAA gene family members,gene specific primers were designed for the only other potatoAux/IAA gene characterized, StIAA. Transcript levels of StIAAat tuber organogenesis (Fig. 3B; StIAA stages 3 and 4) exhibit asmall increase (twofold) after which gene expression returns tobasal levels. The expression patterns of StIAA and StIAA2 have

Fig. 1. (A) Phylogenetic analysis of 58 Aux/IAA proteins including potato StIAA2 and StIAA (bold). The unrooted phylogenetic tree was visualized with theTreeview program (http://rana.lbl.gov/EisenSoftware.htm). (B) Sequence alignments of closely related Aux/IAA proteins reveals the presence of highly conserveddomains indicated with black bars (I–IV). Proteins included in the alignment are: S. tuberosum StIAA2 (gb|EF053504), A. thaliana AtIAA14 (gb|NP_193191),AtIAA17 (gb|NP_171921), AtIAA16 (gb|NP_187124), and AtIAA7 (gb|NP_974355), N. tabacum Nt-IAA28 (gb|AAD32146), poplar 1(P. tremula × P. tremuloides)PptIAA6 (gb|CAC84710), and PptIAA7 (gb|CAC84711).

B. Kloosterman et al. / Plant Physiology and Biochemistry 44 (2006) 766–775768

Fig. 2. Comparison of the relative expression levels of StIAA2 in differenttissues using qRT-PCR. Expression levels were calculated relative to the tissuesin which expression level was highest (petiole). Tissues studied include, petiole,NTS tip, shoot apex, leaf, stem, root, tuber, stolon and dormant tuber (stored for4 weeks at 4 °C).

Fig. 3. qRT-PCR measuring gene expression of two potato Aux/IAA proteins,StIAA and StIAA2 during potato tuber development. RNA derived from eighttuber developmental stages (1–8) harvested over a period of 25 days after aswitch from 16 to 8 h light period (t = 0). (A) Upper panel shows a subset of thedifferent growth stages that can be found during development. (B) Expressionlevels of StIAA and StIAA2 genes relative to internal standard and the firstharvest time point (ΔΔCt).

Table 1Averagea plant height, internode length and petiole angle of StIAA2suppression clones and untransformed control plants

Average plantheight (cm)

Averageinternode length(cm)

Average petioleangle of third andsixth node (°)

Control 119 ± 10.4 7.0 ± 0.4 32.7 ± 5.7StIAA2_S6 158 ± 16.4 8.1 ± 0.6 21.3 ± 3.8StIAA2_S8 161 ± 12.7 8.5 ± 0.5 21.8 ± 3.6StIAA2_S11 160 ± 20.5 8.2 ± 0.7 15.8 ± 5.7StIAA2_S22 156 ± 14.8 7.6 ± 0.5 15.5 ± 5.5StIAA2_S25 162 ± 8.1 8.3 ± 0.3 21.0 ± 5.7a Averaged values and standard deviations were calculated from three inde-

pendent measurements.

B. Kloosterman et al. / Plant Physiology and Biochemistry 44 (2006) 766–775 769

an opposite pattern of expression during tuber organogenesis.However, StIAA2 shows a much larger difference in expressionlevels, indicating different mechanisms of transcriptional con-trol or tissue specificity for both Aux/IAA family membersduring early potato tuber development.

2.3. Suppression lines of StIAA2 exhibit several altered growthresponses

Potato plants were produced containing either the StIAA2over-expression or suppression constructs. Twenty-seven trans-genic clones harboring the suppression construct and fourteenover-expression clones were transferred to the greenhouse.Plants were grown in triplicate and scored for altered growthcharacteristics. For the over-expression clones, no visible dif-

ferences in growth patterns were observed in both the aerialparts of the plant and underground tuber formation in compar-ison to the untransformed control plants (data not shown).

In sharp contrast, over 30% of the suppression clones didshow several dramatic phenotypes. Plants were much tallerand contained on average two to three more internodes in com-parison to untransformed control plants (Table 1). The suppres-sion clones exhibited a lower petiole angle relative to the mainstem resulting in petioles that grow in a more vertical manner(hyponasty) (Fig. 4a, f and Table 1). The petiole angles arelowest around the upper nodes and comparable to controlplants in the basal regions. Furthermore, the shoot apexshowed exaggerated curvature of growing leaf primordia(Fig. 4b–d) and similar deformations were observed in theapices of axillary buds and lateral shoots. This curvature inyoung developing leaf primordia most likely has an effect onfurther leaf development and could partially explain theobserved hyponastic growth pattern of the petiole and subse-quent twisted leaf positioning in fully expanded leaves(Fig. 4e). Additionally, fully expanded leaves were curleddownwards. Another interesting phenotype that, however,occurs at low frequency in a subset of transgenic clones isthe presence of aborted leaf primordia (Fig. 4h). Surprisingly,no clear difference in the transgenic plants could be observedwith regard to growth characteristics of underground stolonsand tuber formation. Averaged stolon length and tuber yieldwas comparable to the untransformed control (data not shown).

For expression studies, the three clones showing the stron-gest overall phenotypes were selected and checked for the pre-sence of the suppression construct. Plants were grown in eightrepeats under controlled conditions in a growth cabinet (16 hlight at 20 °C/18 °C) to allow a more detailed comparisonbetween the transgenic and untransformed control plants. Phe-notypes observed under the controlled climate conditions weresimilar to the ones found in the greenhouse experimentalthough overall plant height was lower and differences inpetiole angle less dramatic (Fig. 4g). Different plant tissues(shoot apex, petiole, NTS) were harvested in replicates todetermine the level of StIAA2 gene suppression. Transcriptlevels were measured using qRT-PCR and were calculatedrelative to the untransformed control plants. In all silencingclones and tissues, StIAA2 transcript levels were significantlyreduced up to 10-fold as shown in Fig. 5.

Fig. 4. Growth phenotypes of StIAA2 down-regulated plants grown in the greenhouse (a–f) or in a growth cabinet (g, h). Increased plant height of 10-week-oldgreenhouse grown plants of StIAA2 suppression clones (S8, S25) compared to untransformed control plants (a). Exaggerated curvature of developing leaf primordiaaround the shoot apex of S8, S21 and S25 in comparison to control (b). Photograph of shoot apex of control (c) and suppression clone S21 (d). Downward curlingand twisted leaf positioning of leaves in the StIAA2 suppression clones (e). Hyponastic growth of young petioles and twisted leaf positioning (f). Comparison of plantheight in controlled climate conditions of StIAA2 suppression clones (S8, S21 and S25) and untransformed control (g). Aborted leaf primordia initiating from theprimary stem in StIAA2 suppression clone S21 (h).

B. Kloosterman et al. / Plant Physiology and Biochemistry 44 (2006) 766–775770

2.4. Downstream effects of reduced StIAA2 expression

Since Aux/IAA proteins are important regulators of auxin-mediated responses and some of the phenotypes found inStIAA2 suppression plants can be identified as enhanced

auxin responses, we have analyzed gene expression in theseclones to screen for altered gene expression profiles. For thisstudy we used the shoot apex to prepare RNA targets for hybri-dization on a dedicated potato tuber life cycle array containingaround 2000 elements. Transgenic clones S21 and S25 were

Fig. 5. qRT-PCR showing reduced expression levels of StIAA2 in differenttissues (NTS, petiole and shoot apex) of the suppression clones (S8, S21 andS25) relative to the untransformed control plants. Error bars indicate thestandard deviation between three independent measurements.

B. Kloosterman et al. / Plant Physiology and Biochemistry 44 (2006) 766–775 771

hybridized in repeats including a swap dye. After data normal-ization and statistical analysis genes that are significantly dif-ferentially expressed (> twofold) in both transgenic lines arelisted in Table 2.

Interestingly, an EST with high homology to StIAA, theonly other identified Aux/IAA family member in potato isdown-regulated 2.4- and 4.0-fold in clones S25 and S21,respectively. To confirm that StIAA is the gene that is actuallydown-regulated in both suppression clones, qRT-PCR was per-formed on the shoot apex and additional tissues, NTS andpetiole using gene specific primers. Lowered StIAA transcriptlevels were confirmed in all tissues analyzed (Fig. 6). Withinthe shoot apex, StIAA expression was down-regulated 4.6-foldand 5.6-fold for clones S25 and S21, respectively, indicatingan underestimation of differential expression in the microarraydata set. The finding of reduced transcript levels for StIAA inthe transgenic clones was unexpected since contrasting expres-sion profiles were observed for StIAA and StIAA2 during potatotuber development (Fig. 3).

Two ESTs showed high homology to a brassinosteroid bio-synthetic protein (DWARF1/DIMINUTO) which is intriguingsince Aux/IAA proteins have been suggested to play a role inbrassinosteroid sensitivity and signaling.

Other genes identified as being differentially expressed inthe StIAA2 suppression clones have a wide variety of functionsoften represented by multiple independent clones strengtheningtheir significance. For example, two different clones withhomology to UDP-glucose glycosyl transferases are severelydown-regulated in both transgenic clones. Similarly, two 2-oxo-glutarate dependent-dioxygenases, two arginine decarbox-ylase, two ent-kaurenoic acid oxidase, and two different clonesrepresenting phenylalanine ammonia-lyase (PAL) were identi-fied (Table 2). Surprisingly, no genes were detected thatshowed a significant increase in transcript levels in the sup-pression clones suggesting that StIAA2 may have primarily apromoting effect on auxin responsive gene expression.

3. Discussion

Based on its high sequence homology with other Aux/IAAfamily proteins and the presence of all four highly conserveddomains (Fig. 1), we have identified a second potato Aux/IAAfamily member protein that was named StIAA2. Plants withreduced StIAA2 expression exhibit a number of dramaticgrowth alterations in the aerial parts of the plant including,increased plant height, petiole hyponasty and curvature ofleaf primordia in the shoot apex. The curling of leaves anddifferences in plant height are common characteristics of differ-ential auxin responses identified in Arabidopsis Aux/IAAmutant lines (reviewed in [25]) and in tomato [18], suggestinga conserved role for Aux/IAA proteins in the regulation ofthese responses in plants.

A more unique phenotype that was observed in our StIAA2suppression clones in relation to Aux/IAA protein functionalstudies, is the hyponastic growth pattern of young petioles. Inan early study by Kazemi and Kefford [26], it was shown thatdecapitation of the main shoot of tomato induces hyponasty ofyoung leaves and the application of auxin to the cut apex pro-duces an epinastic growth response associated with ethyleneproduction. Similarly, Cox et al. [27], have shown that thehyponastic growth response leading to the upward movementand growth of young petioles in submerged Rumex palustris isa result of differential cell elongation across the petiole base.Within this response, differential auxin distribution is likely toplay a role in the maintenance of the hyponastic growth patternin which ethylene, ABA and gibberellin play regulatory andinducing roles. The observation that the hyponastic growth pat-tern in the suppression clones is highest in the upper nodes andlower in the basal part of the plant is consistent with a normalauxin distribution pattern. In this respect the shoot apex, theprimary site of auxin synthesis, exhibits one of the most severegrowth defects, the curvature of developing leaf primordia inthe shoot apex (Fig. 4b, d). Whether the hyponastic growthpattern of young petioles are a continuation of the observedcurvature of leaf primordia in the shoot apex remains to beinvestigated. Taken together, StIAA2 may play a role in theauxin-mediated growth and development of young petioles,supported by its high expression within these tissues.

In order to ascribe the observed downward curling of leavesin the transgenic plants to an increase in average cell size, wemeasured epidermal cell sizes on the adaxial surface of leafstaken from control and transgenic plants, however, no signifi-cant differences could be found (data not shown). The StIAA2plants exhibited increased internodes length as well as moreinternodes per plant analyzed. A similar type of increase ininternode length was observed in the tomato IAA9 anti-senselines [18] and slightly longer hypocotyls in the loss of functionmutant of IAA7 [13]. It is presently unclear whether theincrease in plant height, found in the StIAA2 suppressionclones, is a result of differential cell elongation, increased celldivision or a combination of factors. Auxin has been shown tobe able to regulate GA biosynthesis [28] as well as GA respon-siveness in Arabidopsis [29]. The observed down regulation of

Table 2Differential gene expression in the shoot apex of StIAA2 suppression lines (S21, S25) in comparison to untransformed control plants

Clone accessionnumber

Gene homology Fold down regulation

StIAA2_S25 StIAA2_S21BI178011 StIAA [S. tuberosum] –2.4 –4.0BG098056 DWARF1/DIMINUTO [S. lycopersicum] (Brassinosteroid biosynthetic protein) –3.8 –5.6BG596458 DWARF1/DIMINUTO [S. lycopersicum] (Brassinosteroid biosynthetic protein) –4.1 –5.1BE340263 Cellulose synthase catalytic subunit [A. thaliana] –4.2 –4.1BG095782 Cytochrome P450 [S. lycopersicom] –2.4 –3.2BG594226 Cytochrome P450 [S. tuberosum] –2.7 –2.9BE923430 Tyrosine/dopa decarboxylase [P. somniferum] –3.1 –2.7BE344117 Arginine decarboxylase 1 [D. stramonium] –2.8 –3.0BG600942 Arginine decarboxylase 1 [D. stramonium] –2.4 –3.3BG593608 PAL [S. lycopersicum] –2.4 –2.0BE344057 PAL 1 [S. tuberosum] –2.1 –2.9BG593871 Ent-kaurenoic acid oxidase [H. vulgare] –2.0 –2.9BI175804 Ent-kaurenoic acid oxidase [C. maxima] –3.6 –2.9BF188276 2-Oxoglutarate-dependent dioxygenase [S. tuberosum] –4.3 –4.6BG596984 2-Oxoglutarate-dependent dioxygenase [S. chacoense] –4.2 –5.5BG598323 UDP-glucose glucosyltransferase [S. tuberosum] –4.3 –8.3BE340276 UDP-glucose glucosyltransferase [S. tuberosum] –4.7 –3.9BI179665 Aspartic protease inhibitor 8 precursor [S. tuberosum] –2.5 –2.3BI179515 Major intrinsic protein 2 [S. tuberosum] –2.5 –4.0

Fig. 6. qRT-PCR confirmation of lowered StIAA expression in StIAA2suppression clones (S21 and S25) in harvested tissues (NTS, petiole and shootapex) relative to the untransformed control plants. Error bars indicate thestandard deviation between three independent measurements.

B. Kloosterman et al. / Plant Physiology and Biochemistry 44 (2006) 766–775772

an early GA biosynthesis gene (Table 2; ent-kaurenoic acidoxidase) in the apex of the StIAA2 transgenic clones wouldsuggest altered GA metabolism. However, no changes in tran-script levels were observed for genes active later in the GAbiosynthesis pathway, that are known to be under direct feed-back control of bioactive GA levels [30,31]. Similarly, thedown regulation of a brassinosteroid biosynthesis gene(Table 2; DWARF1/DIMINUTO [32]) in the shoot apex oftransgenic clones is interesting due to the interdependencybetween auxin and brassinosteroid signaling [33–36]. Auxinsand BR’s are known to act both independent as well as syner-gistically regulating transcription of a unique and similar subsetof genes [35,37]. Furthermore, Aux/IAA proteins function inBR signaling pathways modulating BR sensitivity in a mannerdependent on organ type [33]. Recently it was shown that bras-sinosteroids are able to regulate expression of several Aux/IAAgenes in Arabidopsis [36]. The differential expression of a BRbiosynthesis gene further supports the existence of crosstalkbetween Aux/IAA and brassinosteroid signaling and regulationas described in a number of research papers [33–35,38].Detecting a change in overall hormone concentrations, balanceor distribution could provide more insight into the biologicalmechanisms underlying altered plant height or the hyponasticgrowth response in StIAA2 transgenic plants.

A number of Aux/IAA proteins have been shown to func-tion as transcriptional repressors through interaction with ARFproteins that have the capacity to induce or suppress the tran-scription of auxin responsive genes [3,8,39]. Studying geneexpression in the shoot apex of both suppression clones incomparison to the untransformed control using a dedicatedcDNA microarray, we detected significant down regulation of19 different ESTs with often overlapping functional homolo-gies (Table 2). Although the dedicated potato array used inour studies contains a relative small number of genes(< 2000), it is enriched for containing hormone regulated

genes [24]. However, one could expect the total number ofdifferent genes being either down-regulated or up regulated inthe StIAA2 silencing clones to be larger.

Besides changes in gene expression of hormone relatedgenes as described above, we found that one of our down-regulated ESTs showed strong homology to another Aux/IAAfamily member in potato, StIAA. Gene specific qRT-PCR con-firmed the down regulation of StIAA in the shoot apex as wellas in the other tissues studied (Fig. 6). Differential gene expres-sion of other family members in Aux/IAA down-regulatedplants has been described for tomato, where down regulationof IAA9 results in increased transcript levels of IAA3 [18]. Inthis system IAA9 acts as a transcriptional repressor of IAA3. Inpotato plants therefore, one could hypothesize a role forStIAA2 as transcriptional inducer of StIAA expression. How-ever, during the early stages of potato tuber development,expression of StIAA2 is rapidly decreasing whilst StIAA expres-

B. Kloosterman et al. / Plant Physiology and Biochemistry 44 (2006) 766–775 773

sion shows a relative small transient increase. This apparentcontradiction may be explained if both genes are expressed indifferent regions or cell types within the growing potato tuberwhilst sharing similar expression specificity in the NTS tip andaerial tissues allowing transcriptional control. StIAA2 andStIAA do not share more than 61% homology at the nucleotidelevel and therefore cross-silencing of StIAA through the StIAA2suppression construct seems unlikely but can not be ruled outcompletely. Similarly, cross-silencing of other still unidentifiedAux/IAA proteins with higher homology to StIAA2 is possibledue to the presence of the conserved domains found within thisprotein family. One should therefore take great care beforeassigning functional roles for individual Aux/IAA family mem-bers.

Other down-regulated genes that were detected in the shootapex of both suppression clones have a wide variety of knownand undefined plant functions that may or may not contributeto the observed phenotypes. Whether these genes are down-regulated directly as a result of StIAA2 gene suppression orindirectly by the altered plant stature, hormone or metabolicstatus still remains to be determined.

Auxin is expected to play an important role in potato tuberdevelopment [19,21]. Therefore, StIAA2 was initially regardedas a good candidate for playing a role in auxin-mediated stolonand tuber development, based on its high expression in NTSand its strong down-regulated expression profile during tuberorganogenesis (Fig. 3). However, we were not able to detect avisible growth defect in any of the underground organs, sug-gesting StIAA2 may exert its function(s) more strongly in lightgrown tissues. In addition, a much suggested explanation forthe finding of no visible phenotypes in Aux/IAA mutant linesis the presence of overlapping functions of other Aux/IAAfamily members within specific plant tissues [17,25]. Thelack of visible phenotypes has become a recurring problem inArabidopsis Aux/IAA loss-of-function mutants as clearlyshown in a recent study by Overvoorde et al. [17]. Therefore,it was surprising to find such distinctive phenotypes in thetransgenic potato clones with reduced StIAA2 expressionresembling a ‘loss of function’ mutation. Similarly, in tomatothe down regulation of IAA9 results in dramatic alterations inleaf morphology and fruit development indicating a specificand unique role for IAA9 in this process [18]. These findingsunderline the importance of analyzing different AUX/IAA pro-teins in a number of different plant species. Furthermore, ana-lysis of Aux/IAA protein functions in other plant species withunique developmental processes (for example tuber formation),enhances the possibility of discovering novel pleiotropicgrowth defects mediated by members of the Aux/IAA genefamily.

4. Methods

4.1. Cloning of StIAA2 and sequence analysis

The full length sequence of the potato StIAA2 was obtainedby sequencing cDNA clone BI179192 deriving from an in

vitro grown microtuber EST library (cv. Bintje) using vectorbased primers T3 and T7. Sequence data of StIAA2 have beensubmitted to GenBank database under accession numbergb/EF053504. Homology searches were carried out using theblastX program [40]. Protein alignments of closely relatedAux/IAA genes identified in other plant species were carriedout using the ClustalX program [41; http://bips.u-strasbg.fr/en/Documentation/ClustalX/].

4.2. Plant transformation

The complete coding sequence of StIAA2 derived from cul-tivar Bintjes and was amplified by PCR (forward primer 5′-CACCATGGACTTGAATCTCAAGGA-3′ and reverse primer5′-TGCTTTCATTCTTCATCAAC-3′). Forward primer con-tains a 5′-CACC partial overhang to facilitate cloning of thePCR product in pENTR/SD/D-TOPO vector (Invitrogen, theNetherlands). StIAA2 was further sub cloned by a LR recombi-nation reaction into Gateway plant transformation destinationvector pk7WG2 and pk7GWIWG2 (Plant Systems Biology,University of Ghent, Belgium) [42]. The two resulting expres-sion constructs were named pk7StIAA2_O and pk7StIAA2_S.pk7StIAA2_O expression results in heterologous expression ofStIAA2 under control of the 35S-CaMV promoter. ThePk7StIAA2_S expression construct consists of a full lengthStIAA2 inverted repeat harboring an intron behind the 35S-CaMV promoter that when expressed produces a hairpin struc-ture resulting in post transcriptional gene silencing (PTGS).pk7StIAA2_O and pk7StIAA_S were transformed into Agro-bacterium tumefaciens strain Agl0 using electroporation. Invitro shoots of the S. tuberosum cv. Karnico were used forA. tumefaciens mediated transformation [43]. After regenera-tion of in vitro shoots on selective kanamycin MS medium(100 mg/l) [44], the shoots were transferred to the greenhouseto produce mature plants. Plants regenerated from untrans-formed in vitro shoots were used as wild type controls.

4.3. Plant material and measurements

Regenerated transformed plants were grown in 5 l soil-filledpots in the greenhouse and regularly scored for growth altera-tions in comparison to control plants. Due to low light condi-tions within the greenhouse, plants were provided with addi-tional lighting through high pressure sodium lamps (Philips,the Netherlands) to maintain a 16 h light period. From a subsetof plants exhibiting increased growth and the control plants,plant height was measured as the distance from the shootapex to the pot rim and average plant height and internodelength was calculated from three independent replicates after12 weeks of plant growth. From the same set of plants averagepetiole angles were calculated by measuring the angle betweenthe petiole base and main stem axis from the third and sixthnode counted from the first node having a fully expanded leaf.

The three suppression clones exhibiting the strongest overallphenotypes were selected for more detailed analysis and geneexpression studies. Integration of at least one copy of either

B. Kloosterman et al. / Plant Physiology and Biochemistry 44 (2006) 766–775774

construct in the selected transgenic plants was confirmed byPCR using 35S-CaMV promoter sequence specific primers(forward primer 5′-GCACCTACAAATGCCATCA-3′ andreverse primer 5′-GATAGTGGGATTGTGCGTCA-3′). Sixreplicates of in vitro untransformed control plants and of thethree selected suppression clones were transferred to soil andgrown in a climate chamber under controlled conditions (16 hlight 20 °C and 8 h dark at 18 °C). After 6 weeks of plantgrowth, tissues including, leaf, petiole and shoot apex wereharvested of two independent replicates of the three suppres-sion clones and untransformed control plant. The collectedmaterial was immediately frozen in liquid nitrogen and storedat –80 °C till further analysis.

4.4. RNA isolation and qRT-PCR

For expression studies within the different transgenic clonesand control plants, total RNA (tRNA) was isolated as describedby Bachem et al. [45]. For expression studies of the StIAA2 andStIAA genes during the potato tuber developmental time seriesand potato tissues, tRNA was used from a previous study [24].One additional tissue sample was included, namely the shootapex, and was harvested from 8-week-old greenhouse grownplants (S. tuberosum cv. Karnico). Relative expression levelsof genes were determined by qRT-PCR on a Perkin ElmerAbi Prism 7700 Sequence detector (Perkin Elmer, Niewerkerk,the Netherlands) following the protocol described in Klooster-man et al. [24]. Potato ubiquitin primers (ubi3) were used as acontrol. Relative quantification of the target RNA expressionlevel and standard deviation was performed using the compara-tive Ct method according to the User Bulletin #2 (ABI Prism7700 Sequence Detection System, December 1997, AppliedBiosystems). The primer sequences for the genes studied areas follows: StIAA2 (gb|EF053504) forward primer 5′-TGATTCATGCAAGCGTTTACG-3′, reverse primer 5′-TGCAAGTCCAATGGCTTCTG-3′; StIAA (gb|AY098938) forward primer5′-CCTTGGGAGATGTTTATTGACAC-3′, reverse primer 5′-TCCGACACTTTTCCATAGCC-3′ and ubi3 (gb|L22576) for-ward primer 5′-TTCCGACACCATCGACAATGT-3′, reverseprimer 5′-CGACCATCCTCAAGCTGCTT-3′.

4.5. Microarray hybridization and data analysis

mRNA was purified from isolated tRNA from the shootapex of suppression clones S21, S25 and untransformed con-trol plants using the GenElute™ mRNA miniprep kit (SigmaAldrich, Zwijnberg, the Netherlands). First strand cDNAsynthesis followed by target labeling was performed using theSuperScript™ Indirect cDNA Labeling System (Invitrogen, theNetherlands), according to the manufactures protocol. Cy3 orCy5 labeled targets from both suppression clones were hybri-dized independently against the labeled untransformed controltarget. A swap dye experiment was included for both suppres-sion clones. Potato tuber life cycle cDNA microarray slideswere pre-hybridized and processed as described in [24]. Hybri-dization of the target samples was performed in the HybAr-

ray12™ (Perkin Elmer, Niewerkerk, the Netherlands) hybridi-zation station at 42 °C over a period of 20 hours. Followinghybridization, slides were washed as described in van Doorn etal. [46]. Slides were immediately scanned using aScanarray®ExpressHT scanner according to the manufacturesspecifications (Perkin Elmer, Niewerkerk, the Netherlands).Spotfinding, data extraction, LOWESS normalization and sev-eral quality control filters (spot quality, low intensity threshold,signal to noise ratio) were performed with the ScanArray®

(Perkin Elmer, Niewerkerk, the Netherlands) and the MicrosoftExcel software packages. Probes on the potato cDNA micro-array were spotted in triplicate and were analyzed separately toproduce a potential number of six independent measurementsfor each suppression clone. Detection of significantly differen-tially expressed genes shared between both suppression cloneswas performed using the SAM software [47] with a set FDR of0. The identified differentially expressed gene set was importedinto the SPSS software package and were subjected to furtherT-testing (P<0.005) for significant expression change>twofoldin both directions.

Acknowledgements

We thank Beatrix Horváth for the helpful discussions andcritically reading of the manuscript. The authors acknowledgefinancial support from EPS (Graduate School of ExperimentalPlant Sciences), Wageningen University and Plant ResearchInternational through its twinning project program.

References

[1] A. Theologis, Auxin-regulated gene expression in plants, Biotechnology(N. Y.) 12 (1989) 229–243.

[2] A.W. Woodward, B. Bartel, Auxin: regulation, action, and interaction,Ann. Bot. (Lond.) 95 (2005) 707–735.

[3] T. Guilfoyle, G. Hagen, T. Ulmasov, J. Murfett, How does auxin turn ongenes, Plant Physiol. 118 (1998) 341–347.

[4] G. Hagen, T. Guilfoyle, Auxin-responsive gene expression: genes, pro-moters and regulatory factors, Plant Mol. Biol. 49 (2002) 373–385.

[5] T. Ulmasov, J. Murfett, G. Hagen, T.J. Guilfoyle, Aux/IAA proteinsrepress expression of reporter genes containing natural and highly activesynthetic auxin response elements, Plant Cell 9 (1997) 1963–1971.

[6] S. Abel, P.W. Oeller, A. Theologis, Early auxin-induced genes encodeshort-lived nuclear proteins, Proc. Natl. Acad. Sci. USA 91 (1994) 326–330.

[7] T. Ulmasov, G. Hagen, T.J. Guilfoyle, Activation and repression of tran-scription by auxin-response factors, Proc. Natl. Acad. Sci. USA 96(1999) 5844–5849.

[8] S.B. Tiwari, X.J. Wang, G. Hagen, T.J. Guilfoyle, AUX/IAA proteins areactive repressors, and their stability and activity are modulated by auxin,Plant Cell 13 (2001) 2809–2822.

[9] S.B. Tiwari, G. Hagen, T.J. Guilfoyle, Aux/IAA proteins contain a potenttranscriptional repression domain, Plant Cell 16 (2004) 533–543.

[10] W.M. Gray, S. Kepinski, D. Rouse, O. Leyser, M. Estelle, Auxin regu-lates SCF(TIR1)-dependent degradation of AUX/IAA proteins, Nature414 (2001) 271–276.

[11] C.K. Worley, N. Zenser, J. Ramos, D. Rouse, O. Leyser, A. Theologis, J.Callis, Degradation of Aux/IAA proteins is essential for normal auxinsignalling, Plant J. 21 (2000) 553–562.

[12

[13

[14

[15

[16

[17

[18

[19

[20

[21

[22

[23

[24

[25

[26

[27

[28

[29

[30

[31

[32

[33

[34

[35

[36

[37

[38

[39

[40

[41

[42

[43

[44

[45

[46

[47

B. Kloosterman et al. / Plant Physiology and Biochemistry 44 (2006) 766–775 775

] D. Rouse, P. Mackay, P. Stirnberg, M. Estelle, O. Leyser, Changes inauxin response from mutations in an AUX/IAA gene, Science 279(1998) 1371–1373.

] P. Nagpal, L.M. Walker, J.C. Young, A. Sonawala, C. Timpte, M.Estelle, J.W. Reed, AXR2 encodes a member of the Aux/IAA proteinfamily, Plant Physiol. 123 (2000) 563–574.

] L.E. Rogg, J. Lasswell, B. Bartel, A gain-of-function mutation in IAA28suppresses lateral root development, Plant Cell 13 (2001) 465–480.

] H. Fukaki, S. Tameda, H. Masuda, M. Tasaka, Lateral root formation isblocked by a gain-of-function mutation in the SOLITARY-ROOT/IAA14gene of Arabidopsis, Plant J. 29 (2002) 153–168.

] Q. Tian, N.J. Uhlir, J.W. Reed, Arabidopsis SHY2/IAA3 inhibits auxin-regulated gene expression, Plant Cell 14 (2002) 301–319.

] P.J. Overvoorde, Y. Okushima, J.M. Alonso, A. Chan, C. Chang, J.R.Ecker, B. Hughes, A. Liu, C. Onodera, H. Quach, A. Smith, G. Yu, A.Theologis, Functional genomic analysis of the AUXIN/INDOLE-3-ACETIC ACID gene family members in Arabidopsis thaliana, PlantCell 17 (2005) 3282–3300.

] H. Wang, B. Jones, Z. Li, P. Frasse, C. Delalande, F. Regad, S. Chaa-bouni, A. Latche, J.C. Pech, M. Bouzayen, The tomato Aux/IAA tran-scription factor IAA9 is involved in fruit development and leaf morpho-genesis, Plant Cell 17 (2005) 2676–2692.

] Y. Koda, Y. Okazawa, Characteristic changes in the levels of endogen-ous plant hormones in relation to the onset of potato tuberization, Jpn. J.Crop. Sci. 52 (1983) 592–597.

] X. Xu, A.A. van Lammeren, E. Vermeer, D. Vreugdenhil, The role ofgibberellin, abscisic acid, and sucrose in the regulation of potato tuberformation in vitro, Plant Physiol. 117 (1998) 575–584.

] D. Kumar, P.F. Wareing, Studies on tuberization of Solanum andigena.II. Growth hormones and tuberization, New Phytol. 73 (1974) 833–840.

] M.E. Zanetti, M.C. Terrile, A.V. Godoy, B.S. Segundo, C.A. Casalon-gue, Molecular cloning and characterization of a potato cDNA encodinga stress regulated Aux/IAA protein, Plant Physiol. Biochem. 41 (2003)760–775.

] O. Faivre-Rampant, L. Cardle, D. Marshall, R. Viola, M.A. Taylor,Changes in gene expression during meristem activation processes inSolanum tuberosum with a focus on the regulation of an auxin responsefactor gene, J. Exp. Bot. 55 (2004) 613–622.

] B. Kloosterman, O. Vorst, R.D. Hall, R.G.F. Visser, C.W. Bachem,Tuber on a chip: differential gene expression during potato tuber devel-opment, Plant Biotechnol. J. 3 (2005) 505–519.

] E. Liscum, J.W. Reed, Genetics of Aux/IAA and ARF action in plantgrowth and development, Plant Mol. Biol. 49 (2002) 387–400.

] S. Kazemi, N.P. Kefford, Apical correlative effects in leaf epinasty oftomato, Plant Physiol. 54 (1974) 512–519.

] M.C. Cox, J.J. Benschop, R.A. Vreeburg, C.A. Wagemaker, T. Moritz,A.J. Peeters, L.A. Voesenek, The roles of ethylene, auxin, abscisic acid,and gibberellin in the hyponastic growth of submerged Rumex palustrispetioles, Plant Physiol. 136 (2004) 2948–2960.

] J.J. Ross, D.P. O’Neill, J.J. Smith, L.H. Kerckhoffs, R.C. Elliott, Evi-dence that auxin promotes gibberellin A1 biosynthesis in pea, Plant J.21 (2000) 547–552.

] X. Fu, N.P. Harberd, Auxin promotes Arabidopsis root growth by mod-ulating gibberellin response, Nature 421 (2003) 740–743.

] E. Carrera, S.D. Jackson, S. Prat, Feedback control and diurnal regulationof gibberellin 20-oxidase transcript levels in potato, Plant Physiol. 119(1999) 765–774.

] P. Hedden, Y. Kamiya, Gibberellin biosynthesis, enzymes genes theirregul, Annu. Rev. Plant Physiol. Plant Mol. Biol. 48 (1997) 431–460.

] U. Klahre, T. Noguchi, S. Fujioka, S. Takatsuto, T. Yokota, T. Nomura,S. Yoshida, N.H. Chua, The Arabidopsis DIMINUTO/DWARF1 geneencodes a protein involved in steroid synthesis, Plant Cell 10 (1998)1677–1690.

] A. Nakamura, N. Nakajima, H. Goda, Y. Shimada, K. Hayashi, H.Nozaki, T. Asami, S. Yoshida, S. Fujioka, Arabidopsis Aux/IAA genesare involved in brassinosteroid-mediated growth responses in a mannerdependent on organ type, Plant J. 45 (2006) 193–205.

] K.J. Halliday, Plant hormones: the interplay of brassinosteroids andauxin, Curr. Biol. 14 (2004) R1008–R1010.

] J.L. Nemhauser, T.C. Mockler, J. Chory, Interdependency of brassinos-teroid and auxin signaling in Arabidopsis, PLoS Biol. 2 (2004) E258.

] H. Kim, P.J. Park, H.J. Hwang, S.Y. Lee, M.H. Oh, S.G. Kim, Brassi-nosteroid signals control expression of the AXR3/IAA17 gene in thecross-talk point with auxin in root development, Biosci. Biotechnol. Bio-chem. 70 (2006) 768–773.

] H. Goda, Y. Shimada, T. Asami, S. Fujioka, S. Yoshida, Microarray ana-lysis of brassinosteroid-regulated genes in Arabidopsis, Plant Physiol.130 (2002) 1319–1334.

] H. Goda, S. Sawa, T. Asami, S. Fujioka, Y. Shimada, S. Yoshida, Com-prehensive comparison of auxin-regulated and brassinosteroid-regulatedgenes in Arabidopsis, Plant Physiol. 134 (2004) 1555–1573.

] T.J. Guilfoyle, T. Ulmasov, G. Hagen, The ARF family of transcriptionfactors and their role in plant hormone-responsive transcription, Cell.Mol. Life Sci. 54 (1998) 619–627.

] S.F. Altschul, W. Gish, W. Miller, E.W. Myers, D.J. Lipman, Basic localalignment search tool, J. Mol. Biol. 215 (1990) 403–410.

] J.D. Thompson, T.J. Gibson, F. Plewniak, F. Jeanmougin, D.G. Higgins,The CLUSTAL_X windows interface: flexible strategies for multiplesequence alignment aided by quality analysis tools, Nucleic Acids Res.25 (1997) 4876–4882.

] M. Karimi, D. Inze, A. Depicker, GATEWAY vectors forAgrobacterium-mediated plant transformation, Trends Plant Sci. 7(2002) 193–195.

] R.G. Visser, A. Stolte, E. Jacobsen, Expression of a chimaeric granule-bound starch synthase-GUS gene in transgenic potato plants, Plant Mol.Biol. 17 (1991) 691–699.

] T. Murashige, F. Skoog, A revised medium for rapid growth and bio-assays with tobacco tissue cultures, Physiol. Plant. 15 (1962) 473–497.

] C.W. Bachem, R.S. van der Hoeven, S.M. de Bruijn, D. Vreugdenhil, M.Zabeau, R.G. Visser, Visualization of differential gene expression usinga novel method of RNA fingerprinting based on AFLP: analysis of geneexpression during potato tuber development, Plant J. 9 (1996) 745–753.

] W.G. van Doorn, P.A. Balk, A.M. van Houwelingen, F.A. Hoeberichts,R.D. Hall, O. Vorst, C. van der Schoot, M.F. van Wordragen, Geneexpression during anthesis and senescence in Iris flowers, Plant Mol.Biol. 53 (2003) 845–863.

] V.G. Tusher, R. Tibshirani, G. Chu, Significance analysis of microarraysapplied to the ionizing radiation response, Proc. Natl. Acad. Sci. USA 98(2001) 5116–5121.