Introduction

In plants, ABC transporters constitute a large family of proteins exhibiting ATP hydrolyzing activity associated with the presence of the conservative ATP-binding cassette (ABC) domain. The family comprises different subfamilies separated according to the distinct order of their cytoplasmic and membrane-spanning domains. The full-size transporters contain two ABC (NBD) domains that can be differentially organized in relation to transmembrane domains (TMDs) of the proteins. They can be subdivided into three families with the following NBD-TMD composition: (TMD-NBD)2 in multidrug resistance proteins MDR/ABCB, TMD-(TMD-NBD)2 in multidrug resistance-associated proteins MRP/ABCC and (NBD-TMD)2 in pleiotropic drug resistance proteins PDR/ABCG (Crouzet et al. 2006). The half-size ABC proteins contain one TMD domain followed by one ABC domain (TAP, ATH, PMP and ATP subfamilies) with the exception of white-brown complexes (WBC), showing an ABC domain followed by TMD (Sanchez-Fernandez et al. 2001).

Of all the ABC subfamilies, PDR transporters have recently received particular attention since they have been found in plants and fungi but not in animal or prokaryote species. Besides PDRs, only WBC proteins show reverse topology of NBD and TMD domains, so a specific relation between the two subfamilies was somewhat expected. Indeed, sequence analyses of yeast and plant PDRs suggest that PDRs arose by duplication of genes encoding for the half-sized WBC proteins (Dassa and Bouige 2001; Crouzet et al. 2006). They also revealed that the diversity of the PDR family within yeast and plants appeared after the separation of fungi and plant ancestors; therefore, the individual yeast and plant PDR homologs can fulfill different functions (Crouzet et al. 2006). Detailed functional characterization of many of the yeast PDRs revealed that the proteins confer resistance to large number of diverse toxic compounds, including fungicides, herbicides, pesticides, antibiotics and detergents (Crouzet et al. 2006). Hence, yeast PDRs are now regarded as proteins involved predominantly in cell detoxification and cell resistance (Bauer et al. 1999; Rogers et al. 2001). Until recently, it was assumed that plant PDRs are also involved in the cell response to abiotic and biotic stress. Arabidopsis thaliana contains 15 genes encoding members of the PDR family but only a few have been characterized in detail (Table 1). Recently, growing reports reveal that plant PDR proteins might be important not only for the efflux of toxic and harmful compounds from the cell but also for the proper trafficking of the crucial signaling molecules, including hormones that are essential for the regulation of plant growth, development and response to environmental stimuli. It has been shown that AtPDR12/AtABCG40 mediates the uptake of phytohormone ABA in guard cells and other types of plant cells, contributing to the timely closure of stomata in response to drought stress, normal seed germination and lateral root development (Kang et al. 2010). Other plant PDRs, AtPDR9/AtABCG37 and its homolog AtPDR8/AtABCG36 transport a range of synthetic auxinic compounds as well as the endogenous auxin precursor indole-3-butyric acid (IBA) out of the cells (Strader and Bartel 2009; Růžička et al. 2010). In addition, transcriptional profiles of PDR genes in rice reveal that the expression of plant PDRs is significantly induced by jasmonates, salicylic acid and ABA (Moons 2008; Table 1). Also the expression of soybean homolog of AtPDR12/AtABCG40, GmPDR12, is rapidly induced by salicylic acid and methyl jasmonates (Eichhorn et al. 2006).

Table 1 The properties of so far characterized plant PDR genes and/or proteins
Full size table

Altogether, the data indicate a rather novel and significant role for plant PDR proteins in the transport and homeostasis of plant growth regulators and suggest a possible connection between the PDRs and jasmonates and salicylic acid in the plant response to pathogens. The questions arise whether PDR proteins contribute to the signaling pathways induced by other phytohormones essential for the proper function of plant cells and whether PDR proteins in other plants are functionally similar to their Arabidopsis homologs. The availability of PDR sequences in other plants is still limited though, due to the lack of full genomic resources.

In this study, we have identified and analyzed the expression profile of two cucumber genes, PDR8 and PDR12, in different organs at two stages of plant development and under various plant growth regulators. We also present the molecular and phylogenetic characterization of CsPDR8, CsPDR12 and 14 other cucumber PDR homologs of AtPDRs that have been identified through investigation of the recently sequenced cucumber genome (Huang et al. 2009).

Materials and methods

Plant material

Cucumber plants (Cucumis sativus, var. Krak) were grown hydroponically on 3-fold diluted Hoagland solution, pH 6.0, as described earlier (Migocka and Papierniak 2010). The nutrient solution was filter-sterilized, permanently aerated and exchanged twice a week. For the assays of organ expression analyses, roots, hypocotyls or stems, cotyledons, leaves, flowers and fruits were collected after 1 week or 8 weeks of plants cultivation. For the assays including short-term treatment of plants with plant growth regulators, 7-day-old seedlings were transferred on the fresh nutrient solutions containing ABA, IAA, 2,4-D, kinetin, salicylic acid, GA3, ACC or jasmonic acid and kept for 4 h. The final concentrations of each regulator are given in the figure captions. For each treatment, four root samples of 50 mg from four different plants were taken for RNA extraction and immediately frozen in liquid nitrogen before storage at −80 °C.

Semiquantitative and quantitative RT–PCR analysis

Total RNA was isolated from 50 mg of different cucumber tissues using the TRI Reagent (Sigma-Aldrich) following the manufacturer’s instruction. To remove contaminating genomic DNA, RNA samples were treated with DNAse I (Fermentas). DNase-treated RNA samples were used for the semiquantitative RT–PCR or reverse transcribed into cDNA for real-time PCR. Semiquantitative RT–PCR was performed using 200 ng of RNA and Titan One Step System (Roche) under the following conditions: 50 °C for 30 min (1 cycle), 94 °C for 2 min (1 cycle), 94 °C for 30 s, 57 °C for 30 s and 68 °C for 1 min (25 cycles), and 68 °C for 10 min (1 cycle). The reverse transcription of DNAse-treated RNA samples (2,000 ng) was performed using High-Capacity cDNA Reverse Transcription Kit (Applied Biosystem) with random primers following the manufacturer’s instructions. Quantitative RT–PCR was performed with a Lightcycler 2.0 system (Roche) using SYBR-Green (A&A Biotechnology) to detect gene expression abundance, and EFα gene was used as an internal control. The cDNA reaction mixture was diluted eight times, and 1 μL was used as a template in a 10 μL PCR. Amplifications were carried out in 20 μL capillaries (Roche) after pre-incubation at 95 °C for 30 s, followed by 45 cycles of denaturation at 95 °C for 10 s, annealing at 57 °C for 10 s and extension at 72 °C for 15 s, with final melting at 65 °C for 15 s. A negative control without cDNA template was included in the same PCR run for each primer pair. To confirm the specificity of amplification, melting curve analysis was performed allowing to identify putative unspecific PCR products (e.g., primer dimers, reaction mix contamination) and the RT–PCR products were sequenced. Successive dilutions of the sample with the lowest Cp were used as a standard curve. Amplification efficiency was around 2. For each of the two independent RNA extractions, measurements of gene expression were obtained in triplicate. The list of primers is shown in Table 2.

Table 2 The list of primers used in all PCRs
Full size table

Identification of full cDNAs of CsPDR8 and CsPDR12

The partial sequences of CsPDR8 and CsPDR12 were amplified using primers targeting conservative regions of PDR genes (Table 2). Primers were designed manually basing on the multiple alignment of the known PDR ESTs from different plants. PCRs were run using cDNA synthesized from the total RNA isolated from 1-week-old roots of cucumber seedlings and Marathon polymerase (A&A Biotechnology). The obtained PCR products were subcloned into pGEM T-Easy (Promega) and sequenced at least four times. For amplification of unknown 3′and 5′ cDNA ends of CsPDR8 and CsPDR12, RACE-PCRs were performed according to the protocol provided with the GeneRacer Kit (Invitrogen) using 5 μg of total RNA isolated from 1-week-old cucumber roots. For the 3′ end, the first strand synthesis reaction was carried out at 50 °C using GeneRacer OligodT Primer (Table 2). The obtained cDNAs were used in the RACE3′-PCR containing 0.3 μM GSP primers (PDR8race3 or PDR12race3) and 0.9 μM GeneRacer 3′ Primer (Table 2). For the 5′ end, the RNA was ligated to GeneRacer RNA Oligo Sequence and reverse transcribed using GSP primers: PDR8race5 or PDR12race5 (Table 2). The obtained cDNAs were used in the RACE5′-PCR containing 0.3 μM GSP nested primers (PDR8race5n or PDR12race5n) and 0.9 μM GeneRacer 5′ Primer (Table 2). All PCRs contained 1 U of Marathon polymerase (A&A Biotechnology), 1× Marathon amplification buffer and dNTP solution (0.3 μM each). The RACE-PCR conditions were as follows: 94 °C for 2 min, followed by 5 cycles of 94 °C for 30 s and 72 °C for 1 min/1 kB cDNA, followed by 5 cycles of 94 °C for 30 s and 70 °C for 1 min/1 kB cDNA, followed by 25 cycles of 94 °C for 30 s, 65–68 °C for 30 s and 72 °C for 1 min/1 kB cDNA with final extension at 72 °C for 10 min. In all cases, the PCR products obtained were cloned into TOPO vector (Invitrogen) and sequenced (at least three times each). Full cDNAs were submitted to GenBank.

Database searching and sequence analyses

The complete genomic sequence from cucumber has recently been made available to the public in GenBank with accession code ACHR01000000 (Huang et al. 2009). NCBI database (whole-genome shotgun reads) was used for Blastn searches of cucumber sequences with homology to the previously annotated Arabidopsis thaliana PDR sequences. Full cDNAs as well as the protein sequences encoded by the newly identified cucumber PDRs were generated using FGENESH (Salamov and Solovyev 2000). All protein alignments were performed using ClustalW, and the phylogenetic tree was generated using MEGA5.0 software (Tamura et al. 2011) with bootstraps 1,000. The prediction of subcellular localization was performed using ProtComp v8.0 (softberry.com), whereas TMHMM method (Sonnhammer et al. 1998), based on a hidden Markov model (HMM) approach, was applied to predict membrane topology of two CsPDR proteins.

Accession numbers

Sequence data from this article can be found in the GenBank/EMBL databases under the following accessions numbers: GQ374243 (CsPDR8) and GQ374244 (CsPDR12). Additional accession numbers are provided in Table 3.

Table 3 Pleiotropic drug resistance (PDR) transporters genes identified in the whole cucumber genome shotgun reads database (GenBank)
Full size table

Statistical analyses

The qPCR data were analyzed by the ΔΔCT method using the LightCycler® Software 4.1 (Roche). Student’s t test and ANOVA (Excel) were used to confirm the statistical significance of difference in CsPDRs expression between control and hormone-treated plants.

Results

Identification of cucumber genes homologous to Arabidopsis thaliana PDR8 and PDR12

The full sequences of two CsPDRs were cloned using a RACE-PCR approach and designated CsPDR8 and CsPDR12, according to their homology with AtPDR genes. The entire cDNAs of CsPDR8 and CsPDR12 contain an uninterrupted open reading frame of 4,428 bp and 4,356 bp long, respectively; hence, the putative proteins encoded by the two cucumber genes contain 1,475 and 1,451 amino acids, respectively. Both proteins have all domains and motifs characteristic of plant pleiotropic drug resistance type ABC transporters (ABCG), for example, two hydrophilic nucleotide binding folds (NBFs) (ABC_PDR_domain1 and ABC_PDR_domain2) and two hydrophobic transmembrane domains (TMD), the first with six, the second with seven transmembrane spanning regions, according to TMHMM 2.0 Server (Fig. 1). The cytosolic loops containing NBFs precede the transmembrane domains, giving a typical reverse configuration, [NBF-TMD]2, specific to all PDR proteins (Fig. 1). Within the two NBF domains, Walker A, Walker B and ABC signature motifs have been identified, while the PDR-associated domain has been recognized within the sixth span of transmembrane domain 1 (Figs. 1, 2). Both CsPDR8 and CsPDR12 exhibit a high degree of homology at the nucleic acid (62 %) and amino acid (58 %) level. Nevertheless, CsPDR8 and CsPDR12 are more similar to their close homologs of Arabidopsis thaliana, AtPDR8/AtABCG36 and AtPDR12/AtABCG40, with identity of 69 % at the nucleic acid level (CsPDR8 with AtPDR8/AtABCG36 and CsPDR12 with AtPDR12/AtABCG40), and 72 % (CsPDR8 with AtPDR8/AtABCG36) and 70 % (CsPDR12 with AtPDR12/AtABCG40) at the amino acid level.

Fig. 1
figure 1

Topology prediction of CsPDR8 and CsPDR12 according to TMHMM 2.0 Server (Sonnhammer et al. 1998)

Full size image
Fig. 2
figure 2

The alignment of CsPDR8 and CsPDR12 protein sequences with ClustalW (Higgins et al. 1996). ABC signature, Walker A and Walker B motifs as well as PDR-associated domain are presented in boxes

Full size image

The PDR protein family of cucumber

The release of the cucumber genome sequence (Huang et al. 2009) allowed for a first inventory of all cucumber genes encoding for PDR proteins. Through the screening of the whole cucumber genome shotgun reads submitted to GenBank, we have identified 16 genes that were homologous to 15 AtPDRs available in the Aramemnon database (Table 3). To date, only two cDNAs of cucumber PDR genes remain cloned to their full length: CsPDR8 and CsPDR12, which are described here. Research on the cucumber genome revealed that CsPDR8 and CsPDR12 sequences can be found in the contigs ACHRO10014054 and ACHRO10010243, respectively, and allowed for a more profound analysis of the DNA sequence of both genes. It was found that CsPDR8 has 22 exons (each 54–369 bp long) and 21 introns, whereas the slightly longer CsPDR12 has 23 exons (each 54–393 bp long) and 22 introns (Table 3, Fig. 3A). The 5′ and 3′ untranslated regions in CsPDR8 have 249 and 302 bp, respectively, whereas CsPDR12 5′UTR and 3′UTR regions are 672 and 142 bp long, respectively. According to ProtComp v. 9.0 prediction, the proteins encoded by CsPDR8 and CsPDR12 localize to the plasma membrane (Fig. 3b). Both CsPDR8 and CsPDR12 as well as the other predicted CsPDR protein sequences were used to construct the updated phylogenetic tree of plant PDR proteins using the neighbor-joining method (Fig. 4). The arrangement of sequences in the phylogenetic tree suggests four main clusters. Cluster I includes the majority of Arabidopsis PDRs (PDR2, PDR5, PDR9-10, PDR13-15), four cucumber PDRs (PDR5, PDR9-10 and PDR13), 5 rice PDRs (PDR1-2, PDR14, PDR22-23) and PDR3 of Nicotiana tabacum (Fig. 4). Another five cucumber PDRs (PDR4, PDR6, PDR11 and PDR15-16) and three AtPDRs (PDR4, PDR6 and PDR11) belong to cluster II together with three PDRs from rice (OsPDR3, OsPDR5-6), two from Nicotiana plumbaginifolia (Np40786 and NpPDR2), one from Vitis vinifera and one from Triticum aestivum (Fig. 4). Cluster III contains four Arabidopsis PDRs (PDR1, PDR3 and PDR7-8) together with 4 homologous cucumber PDRs (PDR1, PDR3 and PDR7-8), six rice PDRs (PDR9-13 and PDR15), two PDRs of Populus and two PDRs of Vitis (Fig. 4). The remaining three cucumber PDR proteins (PDR2, PDR12 and PDR14) belong to cluster IV, which contains Arabidopsis PDR12, the majority (8) of rice PDRs (PDR4, PDR7-8, PDR16-17 and PDR19-21), two Populus PDRs, three Nicotiana PDRs and one PDR from Glycine max and Spirodela polyrhiza (Fig. 4). Since CsPDR8 and CsPDR12 belong to two different clusters, it may be suggested that both proteins are involved in different physiological reactions of cucumber cells. Indeed, further analyses of organ expression of the genes encoding for CsPDR8 and CsPDR12, as well as their transcriptional profile under different plant growth regulators, revealed differences in the function/regulation of both proteins within plant cells.

Fig. 3
figure 3

The genomic organization (a) of CsPDR8 and CsPDR12 and subcellular localization (b) of the proteins encoded by two cucumber genes. The cDNAs were previously sequenced and the whole gene structures were predicted using FGENESH software (softberry.com). Plasma membrane localization of two CsPDRs was predicted using ProtComp v. 9.0 (software (sofberry.com). CDSf first coding segment (starting with start codon), CDSi internal exon, CDSl last coding segment (ending with stop codon), TSS position of transcription start (TATA-box position)

Full size image
Fig. 4
figure 4

Phylogenetic relationships of the 16 predicted PDR-like transporters identified in cucumber and PDR-like representatives from other plants. The unrooted tree was constructed by the MEGA5.1 software (Tamura et al. 2011) using the neighbor-joining method with 1.000 bootstrap replicate trees. The accession numbers are as follows: Arabidopsis thaliana, AtPDR1 (DAA00870), AtPDR2 (DAA00869), AtPDR3 (DAA00871), AtPDR4 (DAA00872), AtPDR5 (DAA00873), AtPDR6 (DAA00874), AtPDR7 (DAA00875), AtPDR8 (DAA00876), AtPDR9 (DAA00877), AtPDR10 (DAA00878), AtPDR11 (DAA00879), AtPDR12 (DAA00880), AtPDR13 (DAA00881), AtPDR14 (DAA00882), AtPDR15 (DAA00883); Oryza sativa, OsPDR1 (CAD59576 or Q8GU82), OsPDR2 (CAD59575 or Q8GU83), OsPDR3 (CAD59574 or Q8GU84), OsPDR4 (DAA00887), OsPDR5 (Q8GU86), OsPDR6 (Q8GU87), OsPDR7 (Q8U88 or CAD59570), OsPDR8 (Q8GU89 or CAD59569), OsPDR9 (AAQ02685, Q8GU90 or CAD59568), OsPDR10 (Q7PC80 or DAA00884), OsPDR11 (Q8GU92 or CAD59566 or DAA00885), OsPDR12 (Q5Z9S8 or CAD59565), OsPDR13 (Q8S628 or CAD59564), OsPDR14 (AJ535214), OsPDR15 (Q7FMW4 or CAD59563), OsPDR16 (AAQ01165), OsPDR17 (Os08g29570), OsPDR19 (EAZ44308), OsPDR20 (EU682752 or EAZ44307), OsPDR21 (Os09g16290), OsPDR22 (Os12g13720), OsPDR23 (Os12g32820); Populus trichocarpa, Pt83045 (EEE83045), Pt8240 (EEE8240), Pt82928 (EEE82928), Pt8123 (XP_002298123); Nicotiana plumbaginifolia, NpPDR1 (CAC40990), NpPDR2 (AJ831424), Np40786 (CAH40786); Nicotiana tabacum, NtPDR1 (BAB92011), NtPDR2 (BAD07484), NtPDR3 (CAH39853); Triticum aestivum, (TaFJ185035); Glycine max, GmPDR12 (AM261476); Spirodela polyrhiza, SpTUR2 (CAA94437); Vitis vinifera, VV9155 (XP_002279155, VV4885(XP_002284885), VV8313 (XP_002278313)

Full size image

Transcriptional analysis of CsPDR8 and CsPDR12 in various cucumber organs at two different developmental stages

To address the questions whether CsPDR8 and CsPDR12 function in the same or different tissue types or whether they are associated with a broad range of tissues or are specifically expressed in a particular tissue type, we assessed their organ expression patterns using both vegetative organs and inflorescence (Fig. 5). According to RT–PCR analysis, the highest level of CsPDR8 expression was observed in roots of 1-week-old seedlings (Fig. 5). At this stage of development, a small amount of CsPDR8 mRNA was also detected in cotyledons and petioles of the first leaves (Fig. 5). Similar to CsPDR8, the expression of CsPDR12 in young seedlings was also enhanced in roots, and in addition, a low level of transcript was detectable in petioles but not in cotyledons (Fig. 5). In 8-week-old plants, the level of CsPDR8 mRNA was slightly detectable only in roots, whereas CsPDR12 mRNA was the most abundant in roots and significantly lower but clearly detectable in male perianth, female perianth and pistils (Fig. 5). Roots of 1-week-old seedlings were chosen for further analysis of the expression of CsPDR8 and CsPDR12 under plant growth regulators including mainly phytohormones.

Fig. 5
figure 5

Semiquantitative RT–PCR analysis of the organ expression pattern of CsPDR8 and CsPDR12 in 1-week-old and 8-week-old cucumber plants. Gene encoding for elongation factor (EFα) was used as an internal control. Rt roots, Hp hypocotyls, Cot cotyledons, Pt pistil, Lf leaf, S stem, OPt old pistil, OLf old leaf, YPt young pistil, YLf young leaf, Td tendril, Fr fruit, MP male perianth, St stamen, FP female perianth, Ps pistil

Full size image

The transcriptional profiling of CsPDR8 and CsPDR12 in roots in response to phytohormones and plant growth regulators

Using real-time PCR, we performed transcriptional profiling of CsPDR8 and CsPDR12 in roots of 1-week-old cucumbers in response to plant growth regulators. The regulators included hormones involved in the plant response to biotic (SA, JA) and abiotic (ABA) stress as well as molecules crucial for plant growth and development (IAA, kinetin, GA3), phytotoxic herbicide (2,4-D) and ethylene precursor (ACC). Jasmonic acid and salicylic acid proved to be the main stimulators of CsPDR8 and CsPDR12 expression, respectively. 5–20 μM JA increased the level of CsPDR8 and CsPDR12 mRNAs up to 4-fold and over 5-fold, respectively (Fig. 6a, b). The effect of SA on both genes was clearly dependent on phytohormone concentration. CsPDR8 mRNA level was markedly increased only at the highest (10–20 μM) SA concentrations, whereas CsPDR12 expression increased significantly in the SA range tested (5–20 μM). Nevertheless, CsPDR8 expression was enhanced over 2-fold by salicylic acid, whereas CsPDR12 mRNA increased up to 14-fold in response to the hormone (Fig. 6a, b). The expression of CsPDR8 was more enhanced by the 100–200 μM ethylene precursor ACC, which did not affect CsPDR12 transcript (Fig. 6a, b). ACC stimulated CsPDR8 expression over 3-fold (Fig. 6a). In contrast, the treatment of plants with ABA, another stress-related hormone, resulted in a significant decrease in CsPDR8 mRNA (up to 50 %) causing a concentration-dependent increase (up to twofold) in CsPDR12 transcription (Fig. 6a, b). Besides hormones associated with the plant response to stress, the synthetic auxin 2,4-D markedly (6-fold) upregulated CsPDR12 expression when used in the highest (5 μM) concentration (Fig. 7b). However, the auxinic compound had no influence on CsPDR8 mRNA level. Of all other hormones used, only 5–50 μM concentrations of kinetin caused a concentration-dependent, up to 4-fold, increase in CsPDR12 transcript, but similar to 2,4-D had no effect on CsPDR8 expression (Fig. 7a, b). Only auxin and GA3 did not significantly influence the transcription of both cucumber PDRs (Fig. 7a, b).

Fig. 6
figure 6

Quantitative real-time PCR analysis of CsPDR8 (a) and CsPDR12 (b) expression in roots of 1-week-old cucumbers treated with ABA (2, 5, 10, 50 μM), SA (5, 10, 20 μM), ACC (100, 150, 200 μM and JA (5, 10, 20 μM) for 4 h. Control plants were treated with the equivalent amount of methanol (the solvent of JA and ABA) or water. The expression data were analyzed using reference gene encoding for elongation factor (EFα) as an internal control. Asterisks indicate significant differences between control and hormone-treated plants (t test; *P < 0.05)

Full size image
Fig. 7
figure 7

Quantitative real-time PCR analysis of CsPDR8 (a) and CsPDR12 (b) expression in roots of 1-week-old cucumbers treated with IAA (5, 10, 30, 50 μM), 2,4-D (0,5, 2, 5 μM), kinetin (KIN, 5, 20, 50 μM) and GA3 (GA, 5, 20, 50 μM) for 4 h. Control plants were treated with the equivalent amount of methanol (the solvent of IAA and 2,4-D) or water. The expression data were analyzed using reference gene encoding for elongation factor (CsEF) as an internal control. Data are the mean ± SD of at least three independent RNA samples. The results were analyzed by the ΔΔCT method using the LightCycler® Software 4.1 (Roche). Asterisks indicate significant differences between control and hormone-treated plants (t test; *P < 0.05)

Full size image

Discussion

In this work, we present the first identification of PDR genes as well as predicted PDR proteins in cucumber. Two of the PDRs had been previously sequenced and submitted to GenBank as homologs of AtPDR8/AtABCG36 and AtPDR12/AtABCG40, according to their structural similarity to Arabidopsis PDRs. In addition to that, cucumber genome-wide screening revealed the presence of the remaining 14 homologs of Arabidopsis PDRs in cucumber (Table 3). The predicted proteins of the CsPDR family were subjected to phylogenetic analysis revealing that CsPDR8 and CsPDR12 belong to two different clusters, III and IV, respectively (Fig. 4). CsPDR8 clusters with its closest Arabidopsis homolog, AtPDR8/AtABCG36, which has been shown to contribute to the response to pathogen infection, detoxification of heavy metals (Cd and Pb) in plant cells and transmembrane transport of herbicides (2,4-D) and IBA (Kobae et al. 2006; Kim et al. 2007; Strader and Bartel 2009). The transcription of CsPDR8 was markedly upregulated by jasmonic acid, ethylene precursor and, to a lesser extent, by salicylic acid (Fig. 6a). In contrast, ABA significantly reduced the rate of gene expression (Fig. 6a). Since other plant growth regulators did not influence CsPDR8 transcript, it may be suggested that the function of CsPDR8 may be specifically related to the plant responses to biotic or abiotic stress which are mediated by salicylic acid, jasmonic acid, ethylene and/or ABA (Fig. 6a). JA and SA are known to control the local apoptotic hypersensitivity response (HR) and induce systemic acquired resistance (SAR) during pathogen invasion. JA is important in response to wounding and for development processes including growth inhibition, senescence, flower development and leaf abscission (Delker 2006). SA also contributes to the responses to some abiotic stresses such as low and high temperature, UV-B irradiation, ozone and heavy metals (Janda et al. 2007). Ethylene is another plant hormone associated with senescence and response to stress. Plants produce higher levels of ethylene under biotic or abiotic stress, such as pathogen attack, salt, wounding, drought, heat, flooding, low phosphorus (Borch et al. 1999) and low iron (Romera et al. 1999). It has been previously suggested that the compounds transported by PDRs (AtPDR6 and AtPDR11) include antifungal and antimicrobial chemicals that are required for plant protection against pathogen attack (Jasinski et al. 2001; Stukkens et al. 2005; Kobae et al. 2006; Stein et al. 2006; Badri et al. 2007). Like AtPDR6/AtABCG34 and AtPDR11/AtABCG39, cucumber PDR8 is constitutively expressed in root tissue (Fig. 5), which may suggest the possible involvement of the protein in the secretion of exudates in response to external stimuli. CsPDR8 clusters together with OsPDR9 (Fig. 4), which has been shown to be upregulated by phytohormones IAA, cytokinin and JA at the level of mRNA transcription (Moons 2003; Table 1). The different responses of rice and cucumber PDR to IAA and cytokinin suggest that structural homologs of PDRs in monocot and dicot plants are subjected to different modes of regulation by plant hormones.

Similar to CsPDR8, the expression of CsPDR12 was also stimulated by salicylic acid and jasmonic acid. However, SA caused a considerably high increase in the gene transcription when compared with the other growth regulators (Fig. 6b). GmPDR12 and OsPDR20, other proteins that cluster together with CsPDR12 and AtPDR12/AtABCG40, were also markedly induced by both JA and SA or SA and pathogen stress, respectively (Moons 2003; Eichhorn et al. 2006), confirming that PDR12-like proteins fulfill similar functions in different plants. NpPDR1 and NtPDR1, clustering with CsPDR12, have been implicated in pathogen defense (Table 1), possibly through the efflux of antimicrobial and antifungal metabolites out of plant cells (Jasinski et al. 2001; Sasabe et al. 2002; Stukkens et al. 2005). Hence, PDR12-like proteins seem to be key players in the pathogen-induced response. Furthermore, ABA also upregulated CsPDR12 expression. ABA is known to participate in many physiological processes in plants, including seed germination, stomata closure, seedling growth and lateral root development, resistance to drought and other biotic and abiotic stresses, including salinity and pathogen infection. It has been recently demonstrated that the CsPDR12 homolog AtPDR12/AtABCG40 mediates the closure of stomata in response to water deficit through the active ABA transport into mesophyll cells (Kang et al. 2010). In addition, Spirodela PDR clustering together with AtPDR12/AtABCG40 and CsPDR12 (Fig. 4) was also significantly induced by ABA (Smart and Fleming 1996). The SpTUR2 transcript was also elevated upon salinity and low temperature (Smart and Fleming 1996), suggesting the involvement of Spirodela PDR in the ABA-mediated response to these abiotic stresses. The response of CsPDR12 to hormones mediating the stress response suggests that cucumber protein is functionally similar to its Arabidopsis and Spirodela homologs.

Although CsPDR12 seems to be involved in the ABA-mediated response to stress, contrary to CsPDR8, it may not contribute to the ethylene-mediated reaction to environmental stimuli, since ACC did not influence the level of CsPDR12 mRNA (Fig. 6a, b). Unlike CsPDR8, CsPDR12 mRNA was markedly elevated by auxinic herbicide 2,4-D (Fig. 7b). This may suggest that besides ABA, the protein may be capable of transporting auxinic compounds. Indeed, it has been recently demonstrated that PDR proteins may be involved in the transmembrane movement of different auxinic compounds. The direct efflux of IBA or IBA and 2,4-D has been demonstrated using Schizosaccharomyces pombe cells and mammalian cells, respectively, expressing AtPDR9 (Růžička et al. 2010).

The strong stimulation of CsPDR12 by kinetin was in line with the frequent induction of rice PDRs in roots by another cytokinin, 6-benzylaminopurine, and confirmed the need for some of the PDR proteins in root meristematic tissues (Moons 2003, 2008). OsPDR8 and OsPDR17, clustering together with CsPDR12 and AtPDR12/AtABCG40 (Fig. 4), are two of the six OsPDRs significantly stimulated by cytokinin in roots (Table 1). However, all the implications regarding the function of cucumber PDRs require confirmation in further research.

Summing up, we present the first depiction of cucumber PDRs, revealing the presence of at least 16 genes encoding PDR proteins in Cucumis sativus. In addition, we provide evidence for the phytohormone-mediated regulation of at least two CsPDRs, CsPDR8 and CsPDR12. The organ expression pattern of CsPDR8 and CsPDR12 suggests that the proteins are specific to roots and, in the case of PDR12, for flowers. The results confirm the previous data demonstrating the involvement of phytohormones (ABA, ethylene, jasmonates, salicylic acid, cytokinins) and plant growth regulators (2,4-D, auxinic compounds) in the regulation of PDR protein synthesis and/or function in plants. Nevertheless, the precise mechanism of interaction between PDRs and phytohormones or their precursors still remains to be clarified. Interestingly, the genes encoding the two cucumber proteins respond to phytohormones in a distinct way, suggesting dissimilar functions of CsPDR8 and CsPDR12 in the response of plant cells to stimuli involving hormone-mediated signaling pathways.