lable at ScienceDirect

Plant Physiology and Biochemistry 49 (2011) 1100e1107

Contents lists avai

Plant Physiology and Biochemistry

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

Research article

The promoter of a plant defensin gene directs specific expressionin nematode-induced syncytia in Arabidopsis roots

Shahid Siddique a,1, Krzysztof Wieczorek a, Dagmar Szakasits a, David P. Kreil b, Holger Bohlmann a,*

a Institute of Plant Protection, Department of Applied Plant Sciences and Plant Biotechnology, University of Natural Resources and Applied Life Sciences,Konrad Lorenz Str. 24, 3430 Tulln, AustriabChair of Bioinformatics, Department of Biotechnology, University of Natural Resources and Applied Life Sciences, Vienna, Austria

a r t i c l e i n f o

Article history:Received 8 June 2011Accepted 11 July 2011Available online 20 July 2011

Keywords:Plant defensinAntimicrobial peptideSyncytiumHeterodera schachtiiArabidopsisRoots

Abbreviations: Dpi, days post infection.* Corresponding author. Tel.: þ43 1 47654 3360; fa

E-mail addresses: shahid.siddique@boku.ac.atwieczorek@boku.ac.at (K. Wieczorek), dagmar.szakasdavid.kreil@boku.ac.at (D.P. Kreil), holger.bohlmann@

1 Current address: INRES, Department of MoleculBonn Karlrobert-Kreiten-Str.13, 53115 Bonn, Germany

0981-9428/$ e see front matter � 2011 Elsevier Masdoi:10.1016/j.plaphy.2011.07.005

a b s t r a c t

The beet cyst nematode Heterodera schachtii induces a feeding site, called syncytium, in roots of hostplants. In Arabidopsis, one of the genes whose expression is strongly induced in these structures is Pdf2.1which codes for an antimicrobial plant defensin. Arabidopsis has 13 plant defensin genes. Besides Pdf2.1,the Pdf2.2 and Pdf2.3 geneswere strongly expressed in syncytia and therefore the expression of all three Pdfgenes was studied in detail. The promoter of the Pdf2.1 gene turned out to be an interesting candidate todrive a syncytium-specific expression of foreign genes as RT-PCR showed that apart from the feeding site itwas only expressed in siliques (seeds). The Pdf2.2 and Pdf2.3 geneswere in addition expressed in seedlings,roots, leaves, stems, and flowers. These results were supported by the analysis of promoter::GUS lines.After infection with H. schachtii all GUS lines showed a strong staining in syncytia at 5 and 15 dpi. Thisexpression pattern was confirmed by in situ RT-PCR.

� 2011 Elsevier Masson SAS. All rights reserved.

1. Introduction

Nematodes are a group of animals which include free-livingbacterial feeders such as the intensively studied worm Caeno-rhabditis elegans as well as many pathogens of animals and plants.Obligate biotrophic plant-parasitic nematodes attack mainly theroots of many plant species, often causing severe damage to cropplants either directly or as virus vectors. Some of the economicallymost important species are the cyst and root-knot nematodeswithinthe family Heteroderidae which enter the plant roots as secondstage juveniles (J2) and establish specialized feeding structures.The worldwide crop losses due to nematode damage have beenestimated at over $100 billion per year [1]. Root-knot nematodes(genus Meloidogyne) induce a feeding structure which is composedof several giant cells [2]. The feeding structure of cyst nematodes(genera Heterodera and Globodera) is a syncytium [3], which isinitiated from a single root cell and then expands by incorporating upto several hundred neighbouring cells by local cell wall dissolution.

x: þ43 1 47654 3359.(S. Siddique), krzysztof.

its@boku.ac.at (D. Szakasits),boku.ac.at (H. Bohlmann).ar Phytomedicine, University.

son SAS. All rights reserved.

The nematodes feed only from the syncytiumwhich is thus a severenutrient sink for the plant. Adult male cyst nematodes leave the rootto mate with females. The fertilized female cyst nematode continuesto feed but dies once egg development is completed, leaving severalhundred eggs contained within its enlarged body. It subsequentlyhardens to form a cyst, which protects the eggs until infective J2hatch in favourable conditions.

The development of the syncytium from the initial syncytial cellinside the central cylinder is probably initiated through secretionsfrom the nematode and a coordinated expression of plant genes.Such plant genes are, for instance, expansins and cellulases that areimportant for the degradation of cell walls leading to incorporationof new cells into the growing syncytium [4e6]. But syncytial cellwalls also undergo modifications which require the synthesis ofnew cell wall polysaccharides. This might involve the myo-inositoloxygenase pathway as all 4 Arabidopsis genes that code for myo-inositol oxygenase are strongly expressed in syncytia [7]. Protu-berances are produced at the interface between syncytia and xylemvessels and these are thought to be important for the transportof water and solutes [3]. The cells that are incorporated into thesyncytium undergo drastic changes in structure and activity. Thisincludes fragmentation of the central vacuole intomany small ones,accumulation of mitochondria and ribosomes in a dense granularcytoplasm and a proliferation of the endoplasmic reticulum [8,9].To cope with this high metabolic activity, nuclei and nucleoli areenlarged and contain endoreduplicated DNA [10].

Table 1Expression of defensin genes in syncytia and control root segments.

ID Gene Control Syncytium(5 þ 15 dpi)

Control vssyncytium

q-value

At1G75830 Pdf1.1 2.7 2.8 0.2 0.33At5G44420 Pdf1.2a 2.5 3.0 0.5a 0.00At2G26020 Pdf1.2b 3.2 3.1 �0.1 0.49At1G19610 Pdf1.4 3.2 3.2 0.0 0.89At2G02120 Pdf2.1 3.3 11.0 7.7a 8.99E þ 04At2G02100 Pdf2.2 11.1 13.2 2.0a 0.00At2G02130 Pdf2.3 12.2 12.4 0.2 0.33At1G61070 Pdf2.4 3.9 3.9 0.0 0.89At5G63660 Pdf2.5 3.2 2.3 �0.9 0.00At2G02140 Pdf2.6 3.4 3.4 0.0 0.89

Data for microaspirated syncytia at 5 dpi and 15 dpi were combined and comparedwith control roots (elongation zone without root tip was used as control). Allexpression values have been normalized and are on a log2 scale (third and fourthcolumn) and the differences (fold changes) between the pairwise samples displayed(fifth column) are accordingly normalized log2 ratios (see Methods section fordetails). q-values indicate significance after correction for multiple testing control-ling the False Discovery Rate.

a Indicates significant up- or downregulation (false discovery rate < 5%).

Fig. 1. RT-PCR analysis of Pdf2.1, Pdf2.2 and Pdf2.3. Semi quantitative RT-PCR analysis ofPdf2.1, Pdf2.2, and Pdf2.3 in different Arabidopsis tissues. Pdf2.1 was only detected insiliques, whereas transcripts of Pdf2.2 and Pdf2.3 were detected in all plant tissues.Expression of both Pdf2.2 and Pdf2.3 was most intense in 5 days old root tissues.The Suc2 gene was used as a control.

S. Siddique et al. / Plant Physiology and Biochemistry 49 (2011) 1100e1107 1101

The sugar beet cyst nematode Heterodera schachtii completes itslife cycle on Arabidopsis roots in vitrowithin six weeks [11] and thisinteraction has been established as amodel system. The translucentArabidopsis roots growing on artificial media facilitate the studyof the development of this and other nematode species insidethe root [12]. Using this system, we have recently analyzed thetranscriptome of syncytia induced by H. schachtii at 5 and 15 dayspost infection (dpi) [13]. Our results revealed that the Pdf2.1 gene,coding for a plant defensin, was one of the genes that were stronglyupregulated in syncytia as compared to control root sectionsfrom uninfected plants. This was in contrast to the majority ofplant defense-related genes which were not expressed or evendownregulated in syncytia [13].

Plant defensins are a group of antimicrobial peptides witha molecular weight in the range of 5 kDa that have been found invirtually all plants [14,15]. Peptides with a similar structure havealso been found in a variety of animals, including humans [16] aswell as in fungi [17]. All defensins are relatively small, basic peptideswith a three-dimensional structure comprised of a triple-strandedb-sheet with a parallel a-helix [18e20]. In plant defensins thisstructure is usually stabilized by 4 disulfide bridges. All defensinshave repeatedly been shown to have antimicrobial activity in vitro[21,22] and anti-insect activities [23]. In addition, especially thoseplant defensins that were originally called g-thionins, inhibita-amylase activity [24] and protein synthesis [25].

Arabidopsis has 13 Pdf genes that can be divided into two groups[14] and 10 of these genes are represented on the ArabidopsisGeneChip. Three genes of group 1 (Pdf1.2a, Pdf1.2b, and Pdf1.2c)are closely related and encode the same defensin peptide. Pdf1.2 isgenerally regarded as a marker gene for the pathogen specificinduction through the ethylene and jasmonic acid pathways [26].The other Pdf genes are constitutively expressed in certain planttissues (Fig. S1) [27,28]. Group 1 Pdf genes, (Pdf1.1, Pdf1.2a, Pdf1.2b,Pdf1.2c, Pdf1.3 and to some extent Pdf1.4) are induced in thenon-host response of Arabidopsis to the barley powdery mildewfungus [29]. Overexpression of PDF1.1 resulted in enhanced resis-tance of Arabidopsis plants against Cercospora beticola [30]. Inaddition to a role in plant resistance, plant defensins have also beenshown to be involved in conferring zinc resistance in the zinchyper-accumulating plant Arabidopsis halleri [31]. Furthermore,defensin-like peptides were identified as the male determinant ofself-incompatibility in Brassica [32].

Cyst nematodes are a serious problem for a range of importantcrops and researchers are therefore testing several transgenicapproaches to enhance their resistance [33] by targeting thenematode feeding site. This includes for instance the expression ofprotease inhibitors or RNAi for the downregulation of genes whoseexpression is vital for the development of syncytia. In many casesthe CaMV 35S promoter is used which is active in most tissuesof many plant species and might therefore lead to unwantedside effects. There is thus a need for specific promoters that areas specific as possible for nematode feeding sites. The strongupregulation of Pdf2.1 in syncytia according to the GeneChip dataindicated that the Pdf2.1 promoter might be useful to drive theexpression of transgenes in syncytia. Using different techniques wehave therefore studied in detail the expression of Pdf2.1 and theclosely related genes Pdf2.2 and Pdf2.3, which are expressed insyncytia and control root sections.

2. Results

A transcriptome analysis of syncytia induced by H. schachtii inArabidopsis roots [13] revealed several Pdf genes that were stronglyexpressed in syncytia. Analysis of these data (Table 1) showed thatfrom the 10 Pdf genes that were represented on the GeneChip (from

a total of 13 Pdf genes in the Arabidopsis genome), three werestrongly expressed in syncytia while the other Pdf genes wereexpressed at a very low level in both syncytia and in control rootsections. The three genes with a strong expression in syncytiawere Pdf2.1, Pdf2.2, and Pdf2.3. Pdf2.2 and Pdf2.3 were also stronglyexpressed in control root sections. Pdf2.2 had a significant fourfoldinduction in syncytia while Pdf2.3 was not significantly induced insyncytia (1.15fold). Pdf2.1, on the other hand, had a very low level ofexpression in control root sections andwas very strongly induced insyncytia (200 fold). The strong induction of Pdf2.1 in syncytia madeits promoter an interesting candidate to drive specific expression offoreign genes in nematode feeding sites. Therefore the expressionof all three Pdf genes with strong expression in syncytiawas studiedin detail using RT-PCR, promoter::GUS lines and in situ RT-PCR.

2.1. RT-PCR

The expression of Pdf2.1, Pdf2.2, and Pdf2.3 in the plant wasstudied by RT-PCR. RNA was isolated from flowers, siliques, stems,roots and leaves and RT-PCR was performed with specific primersfor all three genes as described in theMethods section. Expression ofPdf2.1 was only detected in siliques while expression of Pdf2.2 andPdf2.3 genes was detected in all analysed plant organs with thestrongest expression in 5 day old roots (Fig. 1). Primers for the Suc2gene were used as a control. GeneChip expression data available in

Fig. 2. GUS expression analysis pattern of the Pdf 2.1 promoter. No expression was observed for Pdf2.1 in seedlings (A), leaves (B), flowers (C) and roots (D, E). However, intense GUSexpression for Pdf2.1 was observed in siliques of these plants (F).

S. Siddique et al. / Plant Physiology and Biochemistry 49 (2011) 1100e11071102

Genevestigator [34] also show that the Pdf genes are expressedin different organs and tissues (Table S1). According to those dataPdf2.1 is especially expressed in seedswhile Pdf2.2 and Pdf2.3 are alsoexpressed in roots and leaveswhich is in agreementwith our results.

2.2. Promoter::GUS analysis

We produced promoter::GUS lines for Pdf2.1, Pdf2.2, and Pdf2.3.We selected representative homozygous lines and general GUSexpression for Pdf2.1, Pdf2.2 and Pdf2.3 was assessed in differentplant tissues of non-infected plants e.g. seedlings, leaves, roots,flowers and siliques. For Pdf2.1 (Fig. 2), expression in non-infectedplants was restricted to siliques, where expression was observedin seeds and at the base (Fig. 2F). Only occasionally we observeda faint GUS expression in leaves and roots (data not shown).

Fig. 3. GUS expression analysis pattern of the Pdf 2.2 promoter. GUS expression driven by tflowers (D), roots (E, F) and siliques (G) of transgenic Arabidopsis plants was observed. Wi

For Pdf2.2, (Fig. 3) there was GUS expression in tissues of non-infected plants which was especially strong in roots (Fig. 3EeF). Inseedlings, GUS expression was observed in radicles and in cotyle-dons (Fig. 3A). Similarly, GUS expression driven by the Pdf2.2promoter was observed in leaves, trichomes, flowers, roots, andsiliques. Within root tissues, expressionwas mostly restricted to thecentral cylinder and there was no GUS expression observed in theroot tip (Fig. 3E). The expression pattern of Pdf2.3 (Fig. 4) in non-infected tissues was similar to that of Pdf2.2, except that there wasalmost no expression in trichomes (Fig. 4C) but intense GUS stainingin thewhole root including root tips (Fig. 4EeF). Within root tissues,GUS expressionwas more intense in the central cylinder and in roottips (Fig. 4F). Twelve day old promoter::GUS lines of Pdf2.1, Pdf2.2,and Pdf2.3 were also infected with H. schachtii larvae. Roots of theinfected plants were stained at 5 and 15 dpi (Fig. 5). Syncytia

he promoter of Pdf2.2 in seedlings (cotyledons and roots) (A), leaves (B), trichomes (C),thin root tissues, there was no expression in the elongation zone and root tip.

Fig. 4. GUS expression analysis pattern of the Pdf 2.3 promoter. GUS expression driven by the promoter of Pdf2.3 in seedlings (cotyledons and roots) (A), leaves (B), flowers (D), roots(E, F), and siliques (G) of transgenic Arabidopsis plants was observed. In contrast to Pdf2.2, there was intense GUS expression in the elongation zone and the root tips. Trichomes didnot show any staining for Pdf2.3 (C).

S. Siddique et al. / Plant Physiology and Biochemistry 49 (2011) 1100e1107 1103

showed a strong staining at 5 dpi for all three genes (Fig. 5A, C, E).In the case of Pdf2.2 and Pdf2.3 staining was also detected in roottissues outside syncytia (Fig. 5C and E). This stainingwas very strongin the case of Pdf2.3 (Fig. 5E). At 15 dpi syncytia for all three genesalso showed a GUS staining, and again, staining for Pdf2.2 and Pdf2.3was also detected in root tissues outside syncytia (Fig. 5B, D, F).

2.3. Localization of Pdf gene expression by in situ RT-PCR

In situ RT-PCR analysis was performed with syncytia at 5 and 15dpi and uninfected control roots with specific primers for Pdf2.1,Pdf2.2, and Pdf2.3 (Fig. 6). Expression of all three genes was clearlydetected in syncytia at 5 dpi as well as 15 dpi. Transcripts of Pdf2.1

Fig. 5. Expression of Pdf2.1, Pdf2.2, and Pdf2.3 promoter::GUS fusions in syncytia induced by Hdpi and 15 dpi (A, B). Root tissues surrounding the syncytium did not show any GUS expreHowever, this expression was not limited to the syncytium but was also found in surroundintissues at 5 dpi and 15 dpi (E, F). This expression, especially at 5 dpi, was more intense in

were restricted to the syncytium (Fig. 6A and G), while transcriptsfor Pdf2.2 (Fig. 6B and H) and Pdf2.3 (Fig. 6C and I) were also detectedoutside the feeding site within the central cylinder which is inagreement to the results obtained with promoter::GUS lines. Inuninfected roots, transcripts for Pdf2.2 (Fig. 6N) and Pdf2.3 (Fig. 6O)accumulated in cells of the central cylinder, while expression ofPdf2.1 in these root sections was only very faint inside the centralcylinder (Fig. 6M). Control reactions that were performed withoutpolymerase always lacked any specific staining (Fig. 6DeF, JeL, PeS).

For gene expression in roots data are also available froma transcriptome analysis of FACS purified root cells (Fig. S1) [35].These data reveal a very weak expression of Pdf2.1 especially inolder root tissues and a strong expression of Pdf2.2 and Pdf2.3

. schachtii. GUS expression driven by the Pdf2.1 promoter in nematode feeding sites at 5ssion. For Pdf2.2, there was also GUS expression in syncytia at 5 dpi and 15 dpi (C, D).g tissues. For Pdf2.3, there was also GUS expression in syncytia as well as in surroundingsurrounding tissues than in syncytia. N, nematode; S, syncytium

Fig. 6. In situ RT-PCR. In situ RT-PCR analysis of Pdf2.1, Pdf2.2, and Pdf2.3 on sections of syncytia induced by Heterodera schachtii in roots of Arabidopsis as well as uninfected controlroots. A, B and C e transcripts of Pdf2.1, Pdf2.2, and Pdf2.3, respectively, accumulate in the 5 dpi syncytia. D, E and F e control sections of 5 dpi syncytia show lack of the specific signalof Pdf2.1, Pdf2.2, and Pdf2.3, respectively. G, H and I e specific staining of Pdf2.1, Pdf2.2, and Pdf2.3, respectively, in 15 dpi syncytia. J, K and L e control sections of 15 dpi syncytia showlack of the specific staining of Pdf2.1, Pdf2.2, and Pdf2.3, respectively. M e very weak staining of Pdf2.1 within the central cylinder of the uninfected root. N e transcripts of Pdf2.2accumulate in cells of the central cylinder of the uninfected root. O e transcripts of Pdf2.3 accumulate in cells of the central cylinder of the uninfected root. P, R and S e controlsections of uninfected control roots show lack of the specific signal of Pdf2.1, Pdf2.2 and Pdf2.3, respectively. AeF and MeS scale bar ¼ 20 mm; GeL scale bar ¼ 50 mm.

S. Siddique et al. / Plant Physiology and Biochemistry 49 (2011) 1100e1107 1105

genes. Expression of Pdf2.3 is stronger in root tips and the rootelongation zone as compared to Pdf2.2. These data are in agreementwith our studies.

3. Discussion

Arabidopsis contains 13 plant defensin genes, some of which areinduced by infection with fungi [27,29,30,36]. Pdf1.2, actually 3 veryclosely related genes that code for the same peptide, is frequentlyused as a marker gene for the ethylene and jasmonic acid pathway[26]. We have recently shown that this gene and other defense-related genes which are regulated through jasmonic acid, ethylene,or salicylic acid, are not expressed in syncytia [13]. It has, however,been reported that PR-3 and PR-4 genes, which are regulatedthrough jasmonic acid, are induced in leaves of Arabidopsis plantsinfected with H. schachtii [37]. It is not known if this inductionextends to other jasmonic acid regulated genes including Pdf1.2. Inaddition to Pdf1.2 also Pdf1.1, Pdf1.3, and to a lesser extent Pdf1.4, wereinduced during non-host interaction of Arabidopsis plants withthe powdery mildew fungus Blumeria graminis f. sp. hordei. Thisinduction is probably mediated by jasmonic acid for Pdf1.2, Pdf1.3,and Pdf1.4, while Pdf1.1 is only slightly inducible by methyle jasm-onate as shown by experiments using treatment of the plants withmethyle jasmonate [29]. The response of Pdf1.2 to jasmonic acid andethylene is mediated through the transcription factor ORA59 whichbinds to GCC boxes in the Pdf1.2 promoter [38].

We found that expression of the Pdf2.1 gene was stronglyupregulated in syncytia induced by the beet cyst nematodeH. schachtii in Arabidopsis roots [13]. In addition, Pdf2.2 and Pdf2.3were also strongly expressed in syncytia but contrary to the Pdf2.1gene, both genes have a strong expression in uninfected roots.The ethylene/jasmonic acid marker gene Pdf1.2, however, was notinduced in syncytia as were other genes that respond to ethylene,jasmonic acid, and salicylic acid [13]. The Pdf2.1, Pdf2.2, and Pdf2.3genes, which are the focus of this publication, as well as Pdf2.4,Pdf2.5, and Pdf2.6 are not inducible by ethylene, jasmonic acid, andsalicylic acid according to Genevestigator (data not shown).

Since Pdf2.1 was among the most strongly upregulated genes insyncytia, we were interested to use the promoter of this gene forthe expression of foreign genes in feeding sites. We thereforestudied its expression as well as the expression of Pdf2.2 and Pdf2.3in detail. RT-PCR and GUS analysis showed that the Pdf2.2 andPdf2.3 genes were expressed throughout the plant with the stron-gest expression in roots and siliques while expression of Pdf2.1wasrestricted to siliques (especially seeds). GUS lines and in situ RT-PCRalso demonstrated the expression of all three genes in syncytia.These results confirmed that the promoter of the Pdf2.1 gene canbe used for a strong and specific expression in syncytia. Besidesfeeding sites, the Pfd2.1 gene was only expressed in seeds, however,this disadvantage has to be accepted as no gene is known to beexclusively expressed in syncytia but nowhere else in the plant.Thus, the promoter of the Pfd2.1 gene is a valuable tool forengineering plants with resistance against cyst nematodes whichwould involve a strong and specific expression in syncytia. Onemight envision that such resistance could be achieved by over-expression of genes that are downregulated in syncytia. We haveidentified a large number of such genes as compared to controlroot sections [13] which could be tested in such an approach. Astrong and specific expression in syncytia is also needed if miRNAs[39] would be used to downregulate genes that are essential forthe function and development of syncytia. Also a large numberof candidate genes in this category have been identified in ourprevious transcriptome analysis of syncytia. We have already usedthis promoter for a specific downregulation of some of these genesin the syncytium using artificial miRNAs (Siddique and Bohlmann,

unpublished results; Ali and Bohlmann, unpublished results).Furthermore, resistance against nematodes might also be achievedby expressing proteins or other compounds that lead to the death ofthe nematodes if taken up from the feeding sites [33]. Examplesfor the latter include proteinase inhibitors and dsRNA or miRNAswhich are targeted against essential nematode genes. For all theseapproaches promoters that direct a strong and specific geneexpression in feeding sites would be useful.

In contrast to cyst nematodes, root-knot nematodes, such asMeloidogyne incognita, induce several giant cells embedded in thegall tissue. The transcriptome of giant cells, induced byM. incognitain Arabidopsis roots, which were cut out from infected roots hasbeen studied using CATMA microarrays [40]. The available datashow that Pdf2.1 is, in contrast to syncytia, not induced in giantcells. No data are available for Pdf2.2 and Pdf2.3 but also the otherPdf genes that can be studied using CATMA microarrays did notshow any induction in giant cells.

Cyst nematodes and root-knot nematodes live only from nutri-ents that are taken up from syncytia or giant cells, respectively, withthe help of the stylet and specialized structures, called feedingtubes, that the nematodes produce within the feeding sites. Root-knot nematodes seem to have a higher size exclusion limit thanthe cyst nematodes. According to Urwin et al. [41], H. schachtiiwas able to take up the cystatin Oc-IDD86 (11.2 kDa) but not GFP(28 kDa).M. incognita, however, was able to take up GFP. Böckenhoffet al. [42] used microinjection of fluorescent probes to determinethe size exclusion limit of H. schachtii that was estimated to bebetween 20 and 40 kDa. Taking these data together, this meansthat the plant defensins PDF2.1, PDF2.2, and PDF2.3 (approximately5 kDa), which seem to be produced in large amounts in syncytia,could be easily taken up by the cyst nematodes. This would alsoindicate that these defensins have no toxic effect on the nematodeand that theymight rather be a rich source of nitrogen and sulfur forthem. However, it cannot be excluded that the plant defensins couldbe retained in subcellular compartments that are not accessed bythe feeding nematode or could be secreted to the apoplast outsidethe syncytium. To resolve these different possibilities would requirethe use of specific antibodies.

4. Conclusion

Our expression analysis of the Arabidopsis plant defensin genesPdf2.1, Pdf2.2, and Pdf2.3 showed that they are strongly expressed insyncytia with Pdf2.1 being one of the most strongly induced genes.Our results also demonstrated that the Pdf2.1 promoter couldbeused for syncytium-specific expression of nematicidal products orgene products that would inhibit the development of the syncytium.

5. Methods

5.1. Plant cultivation

Arabidopsis seeds were surface-sterilized for 20 min in 6% (w/v)sodium hypochlorite and subsequently washed three times withsterile water. Seeds were placed into sterile Petri dishes (9 cm) ona modified Knop medium with 2% sucrose [11]. Seeds were grownin a growth chamber at 25 �C in a 16 h light and 8 h dark cycle.

5.2. Statistical analysis of microarray data

Affymetrix CEL files were analyzed using packages of theBioconductor suite (www.bioconductor.org). Details are provided inSzakasits et al. [13]. For the statistical tests, individual gene vari-ances have been moderated using an Empirical Bayes approach asdescribed in Siddique et al. [7]. Tests were restricted to the subset of

Table 2Primer pairs used for PCR.

Forward primer Reverse primer

Pdf2.1 AAATGCGTGAGCGATACA ACACACTAAACACGCATACPdf2.2 ACATGCGTGAGTGCATCA TAGCTTTGTTATCAGAACATAGATTTTPdf2.3 CCATGTGTGAGCACACAC GACTCCGGTTATTAAAAACTTTAtSuc2 GCCTCTAAGAAGCTTTACAACGAC CCCATAGTAGCTTTGAAGGCAPromPdf2.1 ATTAGGGTACCTTTGGAGTGACAGATTC AGAGAACTCCATGGTTGGAGAAAGAGAAPromPdf2.2 GAGTGGTACCAGACACAAATCTCACTAGAT AGAGAGCTCCATGGCAAGAGAGATAAAGAPromPdf2.3 TCAAGGTACCAATATGAAGTATAAAAACGTTT AGAGAGCTCCATGGTTGAGAGGGATAGA

S. Siddique et al. / Plant Physiology and Biochemistry 49 (2011) 1100e11071106

10 Pdf genes (from a total of 13) that are included on the GeneChip.This considerably increases the statistical power of the testingprocedure as it reduces the necessary correction for massivemultiple testing.

5.3. RT-PCR

Total RNA of different plant tissues was extracted with theRNeasy Plant Mini Kit (Qiagen). Superscript III (Invitrogen) wasused to transcribe total RNA into cDNA. Afterwards, 1 ml of cDNAwas used to perform RT-PCR with forward and reverse primers asgiven in Table 2. AtSUC2was used as control. Primers were designedby aligning all PDF genes using ClustalW and choosing the regionswhere these genes show maximum dissimilarity (see Fig S2).

5.4. GUS reporter analysis

Promoter regions (approx. 1000 bp) upstream the start codon ofPdf2.1, Pdf2.2, and Pdf2.3were amplified using Arabidopsis genomicDNA as template. Primer pairs are given in Table 2. Forward andreverse primers included restriction sites for Kpn1 and Nco1,respectively. These restriction sites were subsequently used forcloning into pPZP3425 [43]. Promoter::GUS constructs wereintroduced into Agrobacterium tumefaciens (GV3101), which wasthen used for transformation of Arabidopsis Col-0 using the floraldip method [44]. Transformed plants were selected on MS mediumcontaining 50 mg ml�1 kanamycin.

For analysis of GUS expression, Arabidopsis seeds were surface-sterilized for 5 min in 75% ethanol followed by 5 min in 10% (v:v)commercial bleach and subsequently washed three times in sterilewater. Plants were grown on Knop medium [11] and infectedwith nematodes as described below (see nematode infection). TheGUS expression was analysed at 5 and 15 dpi. For GUS expressionanalysis of aerial plant parts, plants were grown in soil in a climatechamber under long day light conditions. Plant tissues were sepa-rated and submerged in 100mmol NaPO4 buffer (pH 7.0) containing10 mmol EDTA, 0.01% Triton X-100, 0.5 mmol K3(Fe(CN)6),0.5 mmol K4(Fe(CN)6) and 1 mg ml�1 5-bromo-4-chloro-3-indolylglucuronide. Tissues were vacuum infiltrated for 5 min and thenincubated in the dark at 37 �C for 5e6 h for Pdf2.2 and Pdf2.3 andovernight in case of Pdf2.1. For analysis of GUS expression insyncytia, roots were always incubated for 8 h.

5.5. In situ RT-PCR

This analysis was carried out according to the protocolsdescribed in Koltai and Bird [45] and Urbanczyk-Wlochniak et al.[46]. 5 and 15 dpi syncytia as well as 5 dpi uninfected controlroots were dissected from the plates and immediately put into coldfixative (63% ethanol, v/v; 2% formalin, v/v). After a short incubationin vacuum they were placed on a horizontal shaker at 4 �C for 2days. Root samples were washed and embedded in 4% low-meltingagarose and 20 mm thick sections were prepared using a vibratom

(VT100, Leica, http://www.leica.com/). Subsequently, DNasedigestion and RT-PCR with digoxigenin-labelled dUTP were carriedout. For the staining reaction nitro blue tetrazolium substrate(NBT/BCIP) was used and sections were photographed using aninverted microscope (Axiovert 200M, Zeiss, http://www.zeiss.com/) with an integrated camera (AxioCam MRc5, Zeiss, http://www.zeiss.com/). For a detailed description see [4].

5.6. Nematode infection

H. schachtii cysts were harvested from in vitro stock cultures onmustard (Sinapis alba cv. Albatros) roots growing on Knop mediumsupplemented with 2% sucrose [11]. Hatching of J2 was stimulatedby adding 3mMZnCl2. The J2were resuspended in 0.5% (w/v) gelrite(Duchefa, Haarlem The Netherlands) and 12 day old Arabidopsisroots were inoculated under sterile conditions with approximately80e90 J2 per plant.

Acknowledgements

We appreciate the excellent technical assistance of SabineDaxböck-Horvath. This research was supported by grants P16296-B06, P21067-B12, and P20471-B11 of the Austrian Science Fund(FWF). Shahid Masood Siddique was supported by Higher Educa-tion Commission (HEC) of Pakistan. DPK gratefully acknowledgessupport by the Vienna Science and Technology Fund (WWTF),Baxter AG, Austrian Research Centres (ARC) Seibersdorf, and theAustrian Centre of Biopharmaceutical Technology (ACBT).

Appendix. Supplementary data

Supplementary data associated with this article can be found inthe online version, at doi:10.1016/j.plaphy.2011.07.005.

References

[1] J.N. Sasser, D.W. Freckman, J.A. Veech, D.W. Dickson, A World Perspective onNematology: the Role of the Society, Vistas on Nematology. Society ofNematologists, Hyatssville, 1987, pp. 7e14.

[2] M.G.K. Jones, H.L. Payne, Early stages of nematode-induced giant-cell forma-tion in roots of Impatiens balsamina, Journal of Nematology 10 (1978) 71e84.

[3] M.G.K. Jones, D.H. Northcote, Nematode-induced syncytium-A multinucleatetransfer cell, Journal of Cell Science 10 (1972) 789e809.

[4] K. Wieczorek, B. Golecki, L. Gerdes, P. Heinen, D. Szakasits, D.M. Durachko,D.J. Cosgrove, D.P. Kreil, P.S. Puzio, H. Bohlmann, F.M. Grundler, Expansinsare involved in the formation of nematode-induced syncytia in roots ofArabidopsis thaliana, Plant Journal 48 (2006) 98e112.

[5] K. Wieczorek, J. Hofmann, A. Blochl, D. Szakasits, H. Bohlmann, F.M. Grundler,Arabidopsis endo-1,4-beta-glucanases are involved in the formation of rootsyncytia induced by Heterodera schachtii, Plant Journal 53 (2008) 336e351.

[6] M. Goellner, X. Wang, E.L. Davis, Endo-beta-1,4-glucanase expression incompatible plantenematode interactions, Plant Cell 13 (2001) 2241e2255.

[7] S. Siddique, S. Endres, J.M. Atkins, D. Szakasits, K. Wieczorek, J. Hofmann,C. Blaukopf, P.E. Urwin, R. Tenhaken, F.M. Grundler, D.P. Kreil, H. Bohlmann,Myo-inositol oxygenase genes are involved in the development of syncytiainduced by Heterodera schachtii in Arabidopsis roots, New Phytologist 184(2009) 457e472.

S. Siddique et al. / Plant Physiology and Biochemistry 49 (2011) 1100e1107 1107

[8] W. Golinowski, F.M.W. Grundler, M. Sobczak, Changes in the structure ofArabidopsis thaliana induced during development of females of the plantparasitic nematode Heterodera schachtii, Protoplasma 194 (1996) 103e116.

[9] M. Sobczak, W. Golinowski, F.M.W. Grundler, Changes in the structure ofArabidopsis thaliana roots induced during development of males of the plantparasitic nematode Heterodera schachtii, European Journal of Plant Pathology103 (1997) 113e124.

[10] A. Niebel, J. De Almeida Engler, A. Hemerly, P. Ferreira, D. Inz, M. Van Mon-tagu, G. Gheysen, Induction of cdc2a and cyc1At expression in Arabidopsisthaliana during early phases of nematode-induced feeding cell formation,Plant Journal 10 (1996) 1037e1043.

[11] P.C. Sijmons, F.M.W. Grundler, N. von Mende, P.R. Burrows, U. Wyss, Arabi-dopsis thaliana as a new model host for plant-parasitic nematodes, PlantJournal 1 (1991) 245e254.

[12] U. Wyss, F.M.W. Grundler, Heterodera schachtii and Arabidopsis thaliana,a model host-parasite interaction, Nematologica 38 (1992) 488e493.

[13] D. Szakasits, P. Heinen, K. Wieczorek, J. Hofmann, F. Wagner, D.P. Kreil,P. Sykacek, F.M. Grundler, H. Bohlmann, The transcriptome of syncytiainduced by the cyst nematode Heterodera schachtii in Arabidopsis roots, PlantJournal 57 (2009) 771e784.

[14] B.P. Thomma, B.P. Cammue, K. Thevissen, Plant defensins, Planta 216 (2002)193e202.

[15] F.T. Lay, M.A. Anderson, Defensinsecomponents of the innate immune systemin plants, Current Protein and Peptide Science 6 (2005) 85e101.

[16] T. Ganz, Defensins: antimicrobial peptides of innate immunity, NatureReviews Immunology 3 (2003) 710e720.

[17] P.H. Mygind, R.L. Fischer, K.M. Schnorr, M.T. Hansen, C.P. Sonksen,S. Ludvigsen, D. Raventos, S. Buskov, B. Christensen, L. De Maria, O. Taboureau,D. Yaver, S.G. Elvig-Jorgensen, M.V. Sorensen, B.E. Christensen, S. Kjaerulff,N. Frimodt-Moller, R.I. Lehrer, M. Zasloff, H.H. Kristensen, Plectasin isa peptide antibiotic with therapeutic potential from a saprophytic fungus,Nature 437 (2005) 975e980.

[18] C. Bloch Jr., S.U. Patel, F. Baud, M.J. Zvelebil, M.D. Carr, P.J. Sadler,J.M. Thornton, 1H NMR structure of an antifungal gamma-thionin proteinSIalpha1: similarity to scorpion toxins, Proteins 32 (1998) 334e349.

[19] F. Fant, W. Vranken, W. Broekaert, F. Borremans, Determination of thethree-dimensional solution structure of Raphanus sativus antifungal protein 1by 1H NMR, Journal of Molecular Biology 279 (1998) 257e270.

[20] M. Bruix, M.A. Jimenez, J. Santoro, C. Gonzalez, F.J. Colilla, E. Mendez, M. Rico,Solution structure of gamma 1-H and gamma 1-P thionins from barley andwheat endosperm determined by 1H NMR: a structural motif common to toxicarthropod proteins, Biochemistry 32 (1993) 715e724.

[21] A. Segura, M. Moreno, A. Molina, F. Garcia-Olmedo, Novel defensin subfamilyfrom spinach (Spinacia oleracea), FEBS Letters 435 (1998) 159e162.

[22] M. Moreno, A. Segura, F. Garcia-Olmedo, Pseudothionin-St1, a potato peptideactive against potato pathogens, European Journal of Biochemistry 223 (1994)135e139.

[23] K.C. Chen, C.Y. Lin, C.C. Kuan, H.Y. Sung, C.S. Chen, A novel defensin encodedby a mungbean cDNA exhibits insecticidal activity against bruchid, Journal ofAgricultural and Food Chemistry 50 (2002) 7258e7263.

[24] P.B. Pelegrini, F.T. Lay, A.M. Murad, M.A. Anderson, O.L. Franco, Novel insightson the mechanism of action of alpha-amylase inhibitors from the plantdefensin family, Proteins 73 (2008) 719e729.

[25] E. Mendez, A. Moreno, F. Colilla, F. Pelaez, G.G. Limas, R. Mendez, F. Soriano,M. Salinas, C. de Haro, Primary structure and inhibition of protein synthesis ineukaryotic cell-free system of a novel thionin, gamma-hordothionin, frombarley endosperm, European Journal of Biochemistry 194 (1990) 533e539.

[26] I.A. Penninckx, B.P. Thomma, A. Buchala, J.P. Metraux, W.F. Broekaert,Concomitant activation of jasmonate and ethylene response pathways isrequired for induction of a plant defensin gene in Arabidopsis, Plant Cell 10(1998) 2103e2113.

[27] P. Epple, K. Apel, H. Bohlmann, ESTs reveal a multigene family for plantdefensins in Arabidopsis thaliana, FEBS Letters 400 (1997) 168e172.

[28] B.P.H.J. Thomma, W.F. Broekaert, Tissue-specific expression of plant defensingenes PDF2.1 and PDF2.2 in Arabidopsis thaliana, Plant Physiology andBiochemistry 36 (1998) 533e537.

[29] L. Zimmerli, M. Stein, V. Lipka, P. Schulze-Lefert, S. Somerville, Host andnon-host pathogens elicit different jasmonate/ethylene responses inArabidopsis, Plant Journal 40 (2004) 633e646.

[30] B.M. De Coninck, J. Sels, E. Venmans, W. Thys, I.J. Goderis, D. Carron,S.L. Delaure, B.P. Cammue, M.F. De Bolle, J Mathys, Arabidopsis thaliana plantdefensin AtPDF1.1 is involved in the plant response to biotic stress, NewPhytologist 187 (2010) 1075e1088.

[31] M. Mirouze, J. Sels, O. Richard, P. Czernic, S. Loubet, A. Jacquier, I.E. Francois,B.P. Cammue, M. Lebrun, P. Berthomieu, L. Marques, A putative novel rolefor plant defensins: a defensin from the zinc hyper-accumulating plant,Arabidopsis halleri, confers zinc tolerance, Plant Journal 47 (2006) 329e342.

[32] C.R. Schopfer, M.E. Nasrallah, J.B. Nasrallah, The male determinant ofself-incompatibility in Brassica, Science 286 (1999) 1697e1700.

[33] V.L. Fuller, C.J. Lilley, P.E. Urwin, Nematode resistance, New Phytologist 180(2008) 27e44.

[34] P. Zimmermann, M. Hirsch-Hoffmann, L. Hennig, W. Gruissem, Genevesti-gator. Arabidopsis microarray database and analysis toolbox, Plant Physiology136 (2004) 2621e2632.

[35] K. Birnbaum, D.E. Shasha, J.Y. Wang, J.W. Jung, G.M. Lambert, D.W. Galbraith,P.N. Benfey, A gene expression map of the Arabidopsis root, Science 302(2003) 1956e1960.

[36] B.P. Thomma, K. Eggermont, I.A. Penninckx, B. Mauch-Mani, R. Vogelsang,B.P. Cammue, W.F. Broekaert, Separate jasmonate-dependent and salicylate-dependent defense-response pathways in arabidopsis are essential for resis-tance to distinct microbial pathogens, Proceedings of the National Academy ofSciences U S A 95 (1998) 15107e15111.

[37] N. Hamamouch, C. Li, P.J. Seo, C.M. Park, E.L. Davis, Expression of Arabidopsispathogenesis-related genes during nematode infection, Molecular PlantPathology 12 (2011) 355e364.

[38] M. Pre, M. Atallah, A. Champion, M. De Vos, C.M. Pieterse, J. Memelink,The AP2/ERF domain transcription factor ORA59 integrates jasmonic acid andethylene signals in plant defense, Plant Physiology 147 (2008) 1347e1357.

[39] R. Schwab, S. Ossowski, M. Riester, N. Warthmann, D. Weigel, Highly specificgene silencing by artificial microRNAs in Arabidopsis, Plant Cell 18 (2006)1121e1133.

[40] F. Jammes, P. Lecomte, J. de Almeida-Engler, F. Bitton, M.L. Martin-Magniette,J.P. Renou, P. Abad, B. Favery, Genome-wide expression profiling of the hostresponse to root-knot nematode infection in Arabidopsis, Plant Journal 44(2005) 447e458.

[41] P.E. Urwin, S.G. Moller, C.J. Lilley, M.J. McPherson, H.J. Atkinson, Continualgreen-fluorescent protein monitoring of cauliflower mosaic virus 35Spromoter activity in nematode-induced feeding cells in Arabidopsis thaliana,Molecular Plant Microbe Interactions 10 (1997) 394e400.

[42] A. Böckenhoff, F.M.W. Grundler, Studies on the nutrient uptake by the beet cystnematodeHeterodera schachtii by in situmicroinjection offluorescent probes intothe feeding structures in Arabidopsis thaliana, Parasitology 109 (1994) 249e254.

[43] D. Szakasits, S. Siddique, H. Bohlmann, An improved pPZP vector forAgrobacterium-mediated plant transformation, Plant Molecular BiologyReporter 25 (2007) 115e120.

[44] S.J. Clough, A.F. Bent, Floral dip: a simplifiedmethod for Agrobacterium-mediatedtransformation of Arabidopsis thaliana, Plant Journal 16 (1998) 735e743.

[45] H. Koltai, D.M. Bird, High throughput cellular localization of specific plantmRNAs by liquid-phase in situ reverse transcription-polymerase chainreaction of tissue sections, Plant Physiology 123 (2000) 1203e1212.

[46] E. Urbanczyk-Wochniak, M. Filipecki, Z. Przybecki, A useful protocol for in situRT-PCR on plant tissues, Cellular and Molecular Biology Letters 7 (2002) 7e18.