Seeds of date palm (Phoenix dactylifera L.), variety Khalas were washed, air-dried, and stored in plastic bags at room temperature. Seeds were germinated in vitro as described previously (Hanano et al., 2016). Seedlings were obtained 15 days after sowing. Seedlings with a radicle length of 0.5, 2.5, or 5 cm were referred as stage I, II, and III, respectively. Non-germinated seeds (stage 0). The 2,3,7,8-tetrachlorodibenzo-p-dioxin (2,3,7,8-TCDD dissolved in toluene at 10 μg mL⁻¹, purity 99%) was purchased from Supelco Inc., USA. To test the effect of TCDD, seeds were daily watered by the prepared solutions at various concentrations of TCDD (0, 10, 50 ng L⁻¹) as described previously (Hanano et al., 2016).

Escherichia coli strain TOP10 was used as host for plasmid cloning experiments. Bacteria were grown in Luria–Bertani medium supplemented with ampicillin (100 mg mL⁻¹) at 37°C. Saccharomyces cerevisiae Wa6 (ade, his7-2 leu2-3 leu2-112 ura3-52) the host for protein expression studies. Recombinant yeast was grown in S medium (7 g L⁻¹ yeast nitrogen base, 1 g L⁻¹ casamino acids, 20 g L⁻¹ glucose supplemented with 50 mg mL⁻¹ histidine, 200 mg mL⁻¹ adenine, and 50 mg mL⁻¹ leucine) for 2 days with shaking at 30°C. Culture media components and agarose were purchased from Sigma-Aldrich, USA and liquid-chromatography grade solvents and buffer components were supplied by Merck, Germany. Isopropyl-b-D-thiogalactopyranoside (IPTG), X-Gal, restriction enzymes, RNase A, molecular markers for DNA, and dNTPs were purchased from Fermentas (Vilnius, Lithuania). Oligonucleotides, aniline, thiobenzamide, cumene hydroperoxide, fatty acids, antibiotics, and amino acids were supplied by Sigma-Aldrich, USA.

LDs were isolated from seeds of date palm (Phoenix dactylifera L.) according to Hanano et al. (2016) with minor modifications. Seeds of date palm were firstly subjected to dry grinding using a high performance grinder (Cross Beater Mill SK, Retsch, Germany). Well-hulled and crushed grains were obtained and brief sieving was used to separate the woody cover particles of the seeds from their stony cores (0.5 mm). Five grams of the ground core were taken and milled in a brass mortar in the presence of liquid nitrogen until a fine powder was obtained. The powder (5 g) was immediately hydrated with 10 mL of buffer A (100 mM potassium pyrophosphate, 0.1 M sucrose, and pH 7.4). The mixture was gently homogenized for 5 min using an ultra-dispenser (T25 digital ULTRA-TURRAX, IKA laboratory, Germany) and centrifuged for 10 min at 10,000 × g. The resulting supernatant was subjected to a second centrifugation at 100,000 × g for 1 h and a floating white pad, consisting of LDs, was collected from the top of the tube. LDs were washed twice with 5 mL of buffer B (buffer A without sucrose). After a final centrifugation (100,000 × g for 1 h), the LD fraction was suspended in 2 mL of buffer B and stored at 4°C for further analysis.

Peroxygenase activity was measured by the oxygenation of aniline or thiobezamide as substrates (Blee and Durst, 1987). Hydroperoxide-reductase activity was measured by incubation of 9-HPOD or 13-HPOD overnight at 26°C with 50 μg of purified recombinant proteins in 500 μL of sodium acetate (0.1 M, pH 5.5) (Hanano et al., 2015b). Residual substrate and products were extracted in 3 × 2 mL of dichloromethane/ether (1:1, v/v). After drying under nitrogen flow, extracts were taken with 25 μL of acetonitrile/water/acetic acid (50/50/0.1, v/v/v). Extracts were analyzed using a Jasco LC-2000 plus series HPLC system (Jasco, USA) with a UV-detector (RF-10Axl, Shimadzu) (234 nm) and a C18 column (Eclipse XDB-C18 150 × 4.6 mm, 5 μm; Agilent, USA). The analysis was performed using a mobile phase of acetonitrile/water/acetic acid (50/50/0.1, v/v/v) at a flow rate of 0.6 mL min⁻¹.

For analysis of lipoxygenase (LOX) activities in date palm seedling tissues, two grams of fresh tissues (plumule, petiole, or radicle) were ground into a fine powder in liquid nitrogen with a mortar and a pestle, followed by resuspension in 100 μL of enzyme extraction buffer (50 mM sodium phosphate buffer, pH 7.5, 10 mM EDTA, 0.1% Triton X-100). The homogenate was centrifuged at 4°C, 16,000 × g for 10 min to remove cell debris. Total protein content was determined by a Bradford assay. LOX activity was determined according to Hanano et al. (2015b) in 500 μl of 50 mM sodium phosphate buffer (pH 7.0) containing 50 μg of proteins extract and 500 μM of linoleic acid (C18:2) at 25°C with constant shaking for 30 min. The reaction products were extracted in 5 mL n-hexane/2-propanol (3/2) (v/v) with 0.025% (w/v) butylated hydroxytoluene and the mixture immediately ultra-homogenized for 30 s on ice. A spiked sample with 100 mM of each hydroperoxide was used as a control. The extract was shaken for 10 min and centrifuged at 3000 × g at 4°C for 10 min. The upper phase was then dried under nitrogen. Hydroperoxides were dissolved in 25 μl acetonitrile/water/acetic acid (50/50/0.1) (v/v/v) and quantified by HPLC as described above. The analysis was performed using a mobile phase of acetonitrile/water/acetic acid (50/50/0.1, v/v/v) at a flow rate of 0.6 mL min⁻¹. FA-hydroperoxides were quantified using their respective standards. One unit of activity (U) corresponds to the amount of enzyme that converts 1 μmol of substrate per minute.

Nucleotide sequences of primers used in this study are shown in Table S1. The PCR amplification was performed in a 25 μl reaction final volume containing 3 mM MgSO₄, 200 μM each of the four dNTPs, 10 μM of each primers and 2.5 U Taq DNA polymerase. PCR conditions were: 94°C for 4 min followed by 35 cycles at 94°C for 30 s, 58°C for 30 s, 72°C for 1 min and 72°C for 10 min. PCR amplifications were verified by 1% agarose gel electrophoresis for 30 min at 100 V. The gel was visualized in an UV illuminator. If necessary, PCR fragments were purified using a QIAquick PCR purification kit (Qiagen, USA).

Full-length ORFs of PdCLO2 and PdCLO4 genes (720 and 711 bp, respectively) were amplified by PCR using date palm cDNA. For further purification of recombinant proteins, primers were designed to add a His tag at the N-terminal ends of the PdCLO2 and PdCLO4 genes (Table S1). Purified PdCLO2 and PdCLO4-PCR products were ligated to pGEM-T Easy vector (Promega, USA) following the manufacturer's manual. The recombinant plasmids pGEM-T/PdCLO2 and pGEM-T/PdCLO4 were transferred into E. coli Top10 competent cells. Recombinant plasmids were extracted from E. coli Top10 using a plasmid purification mini kit (Qiagen, Germany). The presence of PdCLO2 and PdCLO4 genes in the isolated plasmid was confirmed by PCR and SacI-EcoRI digestion. At least three clones of both genes were sequenced on both strands using T7 and SP6 primers on an ABI 310 Genetic Analyser (Applied Biosystems) using Big Dye Terminator kit (Applied Biosystems). Nucleotide sequence analysis and multiple alignments of the nucleotide and deduced amino acid sequences were performed with Vector NTI advance (Invitrogen).

PdCLO2 and PdCLO4 were subcloned into the yeast constitutive expression vector pVT102U (Vernet et al., 1987) using SacI and EcoRI restriction sites. The recombinant plasmids pVT102U/PdCLO2 and pVT102U/PdCLO4 were then introduced into the yeast Saccharomyces cerevisiae Wa6 (ade, his7-2 leu2-3 leu2-112 ura3-52) (Schiestl and Gietz, 1989). Expression of the recombinant proteins in transformed yeast cells was carried out as described previously (Hanano et al., 2006). The LD and microsomal fractions of recombinant yeast cells were extracted and resuspended in 10 mM potassium phosphate (pH 8) containing 10% glycerol (v/v) and treated with the non-polar detergent, emulphogene, (final concentration, 0.2%) for 45 min at 4°C. The mixture was then centrifuged at 100,000 × g for 2 h to removed non-solubilised material.

The LDs of recombinant yeast cultures resuspended in 10 mM potassium phosphate (pH 8) containing 10% glycerol (v/v) were treated separately with emulphogene (Sigma-Aldrich, USA) at final concentration of 0.2% (v/v) for 45 min. The mixture was then centrifuged at 100,000 × g for 45 min. The supernatant was taken and the concentration of solubilized proteins was estimated by Bradford assay and enzymatic activity. The solubilized fraction was used for protein purification. All steps were performed on ice (4°C). His-tagged PdCLO2 and PdCLO4 from in the supernatant were purified on a Ni-NTA Superflow column (Qiagen) under native (non-denaturing) conditions, according to the manufacturer's instructions. The purity of the recombinant proteins was confirmed by SDS-12% PAGE followed by Coomassie-Blue staining. For western blot experiment, proteins were fractionated by 12% SDS-PAGE and electroblotted to Immobilon-P membranes (Millipore Corp., Bedford, MA) using a mini-transblot transfer cell apparatus (Bio-Rad). For detection of the recombinant CLO:His fusion protein, a mouse monoclonal anti-His antibody and an anti-mouse antibody conjugated to peroxidase were used at 1:2000. Blots were developed using the ECL kit from Pierce.

The changes in relative transcriptional abundance of genes encoding PdCLO2, PdCLO4, 9-LOXs, and 13-LOXs in response to TCDD exposure were analyzed by reverse-transcription quantitative PCR (RT-qPCR). Briefly, frozen fine powder (1 g) samples from whole seedlings at stages 0, I, and II or specific tissues were used to extract total RNA using an RNeasy kit according to the manufacturer's instructions (Qiagen, Germany). After quality control, aliquots of 1 μg total RNA were used for first-strand cDNA synthesis according to Hanano et al. (2014a). Real-time PCR was performed in 48-well plates using a AriaMx Real-time PCR System from Agillent technologies, USA. In brief, 25 μL reaction mixtures contained 0.5 μM of each specific oligonucleotide primer for the target genes (PdCLO2, PdCLO4, 9-LOXs, and 13-LOXs) and the reference gene (TIP-41) (Table S1), 12.5 μL of SYBR Green PCR mix (Bio-Rad, USA) and 100 ng cDNA. Subsequently, the relative quantification RQ of target genes was calculated directly using the software installed in the qPCR system. The sequences of amplified regions were confirmed by sequencing on an ABI 310 Genetic Analyzer (Applied Biosystems) using a Big Dye Terminator kit (Applied Biosystems).

For sequence retrieval and identifying LOX proteins, eight CLO sequences of Arabidopsis thaliana were obtained from Arabidopsis information resources TAIR database (https://www.arabidopsis.org/, release 10) (Berardini et al., 2015). CLO sequences of Phoenix dactylifera, Elaeis guineensis, and Musa acuminate were retrieved from NCBI (http://www.ncbi.nlm.nih.gov/) (see Table 1 for accession numbers). Five CLO proteins have been identified in the Phoenix dactylifera and eight CLO proteins in the Arabidopsis thaliana genomes (Shen et al., 2014; Hanano et al., 2016). In this study, six CLO proteins in Elaeis guineensis and two CLO proteins in Musa acuminate were also identified. Using Arabidopsis LOXs as query sequences, LOX sequences were obtained for Phoenix dactylifera, Elaeis guineensis, Musa acuminate from NCBI via local BLAST+ search (Chen et al., 2015). These candidate LOX sequences were analyzed using the InterProScan program (http://www.ebi.ac.uk/interpro/) to confirm the presence of the major LOX domain and PLAT/LH2 (polycystin-1, lipoxygenase, lipoxygenase homology) domain (Jones et al., 2014). CLO and LOX sequences were visually inspected to confirm the presence of LOX and CLO domains.

In this study, we identified 34 LOX proteins among the four analyzed species and divided them into two groups: 9-LOX and 13-LOX. This division is supported by the phylogenetic tree analysis as discussed below. Net charge for each LOX protein was calculated using BACHEM (http://www.bachem.com/service-support/peptide-calculator/). Physical and chemical properties of LOX and CLO proteins (Molecular Weight (MW), isoelectric point (pI), amino acid length) were computed using ExPASY (http://www.expasy.org/protscale/) (Gasteiger et al., 2003).

Multiple sequence alignment and domain analysis were performed using ClustalOmega software version 1.2.2 for 9- LOX, 13-LOX, and CLO sequences using default parameters (Sievers et al., 2011). The alignments were inspected using CLC sequence viewer (Figures S1, S2) (http://www.clcbio.com). Complete alignments with RasMol color codes (Sayle and Milner-White, 1995), are available in the supplementary documents sections. The amino acid sequence alignments were used to construct a phylogenetic tree using ClustalW2 program version 2.1. The tree was generated following Bayesian Inference (BI), Neighbour-joining (NJ), and Unweighted Pair Group Method with Arithmetic Mean (UPGMA) methods. The tree topologies constructed using the three methods showed complete consistency. The NJ tree was constructed using ClustalW2 program version 2.1 (Larkin et al., 2007). The constructed tree was inspected using FigTree and is shown in Figure 8 (Rambaut and Drummond, 2009). Multiple sequence alignment of caleosin sequences from all four species revealed that the Ca²⁺ binding and EF hand motifs were well conserved, except for the single case of AtCLO7. This may indicate AtCLO7 has lost the ability to bind calcium and may explain its very low extent of expression according to transcriptome analysis (Shen et al., 2014, Figure 2). The CLO central hydrophobic domain and proline knot motifs were identified as present in all sequences except PdCLO5 (Figure 2). MEME software was used to investigate whether the 9-LOX and 13-LOX proteins contain the histidine rich motif (Figures S3, S4, Bailey and Elkan, 1994). The overall workflow of the computational methods is shown in Figure S5.

All data presented were expressed as means ± standard deviation (SD). Statistical analysis was performed using STATISTICA software, version10 (StatSoft Inc.). Comparisons between control and treatments were evaluated by ANOVA analysis and multiple comparisons using Duncan's multiple range test.

Caleosin/peroxygenases (CLO/PXGs) have been reported to be lipid-associated proteins in several plant species. We therefore tested the subcellular distribution of these proteins in date palm seedlings where we found that both microsomes and lipid droplets (LDs) were able to catalyze the same co-oxidation reactions reported for other plant PXGs as follows. Both subcellular fractions oxidized thiobenzamide to its corresponding sulfoxide (212 nmol min⁻¹ mg⁻¹ of protein in microsomes vs. 320 in LDs), hydroxylated aniline to nitrobezene (190 in microsomes versus 286 in LDs) and epoxidised oleic acid into 9,10-epoxy-stearate (138 in microsomes vs. 216 in LDs), as shown in Figure 1A. In contrast, the supernatant fraction (at 100,000 × g) was unable to catalyze such reactions. Control experiments showed that both microsomes and LDs were inactivated after heat treatment or in absence of the oxygen donor, cumene hydroperoxide (Figure 1A).

We recently reported that the expression of the CLO-encoding genes, CLO-like2 and CLO-like4, was upregulated when seeds were germinated in the presence of TCDD (Hanano et al., 2016). We therefore examined whether PXG activity was induced by TCDD at three different stages of seedling development (I, II, and III). Non TCCD-treated seedlings showed a basal PXG activity in stage I that increased about 3-fold at stage II before declining to its lower level at stage III (Figure 1B). In contrast, TCDD-treated seedlings showed a significant increase in the PXG activity (about 4- to 2.5-fold) in crude extracts from seedlings exposed to 10 or 50 ng L⁻¹ TCDD respectively, compared with controls at all three stages of germination (Figure 1B). The highest induction of the PXG activity by TCDD was detected in stage II. Interestingly, no activity was detected in non-germinated seeds at stage 0 (data not shown).

The properties and amino acid sequence alignment of predicted CLO-like proteins from date palm, oil palm, banana, and Arabidopsis are shown respectively in Table 1 and Figure 2. We previously showed that the genome of date palm contains five potential orthologous CLO-like sequences (Hanano et al., 2016). However, as noted in Figure 2, the predicted sequence of the PdCLO5 protein was significantly truncated compared to the other caleosins and several canonical caleosin domains were missing. Therefore, although the PdCLO5 gene was expressed in palm seedlings (albeit at low levels) it seems unlikely that this ortholog has the capacity for the range of enzymatic functions of previously described CLO/PXGs (Hanano et al., 2016). In contrast, the remaining four members of the date palm CLO gene family contained all four typical structural features of previously described plant caleosins, namely; a EF-hand motif containing 12 residues involved in Ca²⁺ binding, a central hydrophobic domain including a proline knot (PXXXPSPXXP), two histidine residues responsible for iron binding (heme axial ligand) and a serine kinase phosphorylation site near the C-terminal (Figure 2).

The accession numbers, number of amino acid residues, molecular weight, position of histidine residues responsible for iron binding, position of calcium binding residues and positions of EF-hand motif of cloned date palm caleosins and other plant caleosins are summarized in Table 1. Although all of these CLO genes were expressed in date palm seedlings (Hanano et al., 2016), only two of them, namely PdCLO2 and PdCLO4, were expressed at high levels and the same two genes were significantly induced following exposure to the dioxin TCDD. To gain more detailed information about the possible involvements of these CLO isoforms, we studied their spatiotemporal expression and enzymatic activities at various stages of post-germinative seedling growth and development.

Since PXG activity varied after seed germination (Figure 1B), we used qRT-PCR to assess whether this was due to differential expression of PdCLO2 or PdCLO4 genes after seed germination at stages 0, I, II, and III. Transcription of both PdCLO2 and PdCLO4 was undetectable at stage 0 but was activated by stage I. PdCLO4 transcript levels were highly elevated at stage II compared with those of PdCLO2. After this, there was a net decline in PdCLO4 transcripts, while PdCLO2 transcripts were maximal at stage III (Figure 3A). The increase in PdCLO4 and PdCLO2 transcripts in stage II and III, respectively, led us to investigate whether this was related to a specific tissue in the seedling. When transcripts of both genes were evaluated in the plumule, petiole and radicle tissues in stages II and III, the levels of PdCLO4 gene transcripts were approximately 5.6-fold higher in plumule than in radicle. In contrast, PdCLO2 was 12-fold more expressed in radicle tissues than in plumule in stage III (Figure 3B). Very small levels of transcripts for both genes were found in petioles.

Variation in PdCLO4 and PdCLO2 gene expression in seedling tissues was confirmed by measuring the corresponding enzymatic activities. The 13- and 9-HPOD reductase activity was estimated in crude extracts prepared from plumule, radicle, and petiole in stage I and II respectively. As Figure 3C shows, the maximal 13-HPOD reduction activity was measured in the plumule extract at stage II, where almost all of this hydroperoxide was reduced into 13-HOD (peaks 1 and 2 respectively, panel left-above). In contrast, about 75% of 13-HPOD was transformed into 13-HOD after incubation with radicle extracts (panel left-middle) and no activity was detected in petiole extract (panel left-down). In parallel, 9-HPOD was slightly reduced with plumule extracts but more so with radicle extracts compared with petiole extract (panels right-above middle and down, respectively). An inverse result was observed in stage III where the activity of 13-HPOD was much lower in plumule and was undetectable in radicle and petiole extracts (Figure 3D, panel on left). A brief, a high and a null 9-HPOD-reductase activity was measured in plumule, radicle, and petiole extracts (Figure 3D, panel in right). Together these results indicate that expression of PdCLO4 and its 13-HPOD-reductase activity is higher in less differentiated leaf tissues while expression of PdCLO2 and its 9-HPOD-reductase activity was higher in roots. Due to the interesting variation in tissue expression and catalytic activity these two CLO genes, we selected both of them for cloning and expression in a heterologous yeast cell system.

Following the bioinformatic analysis described above, sequences encoding the PdCLO2 and PdCLO4 isoforms were selected for cloning and separate heterologous expression in yeast in order to test for PXG activity and substrate specificity. The predicted proteins have calculated molecular masses of 26618 and 26732 Daltons respectively. Crude extracts of yeast expressing PdCLO2 and PdCLO4 genes efficiently catalyzed hydroxylation of aniline (132.6 vs. 136.2 nmol min⁻¹ mg⁻¹) and epoxidation of oleic acid (16.2 vs. 96.4 nmol min⁻¹ mg⁻¹) in the presence of cumene hydroperoxide (Figure 4A). No enzymatic activity was observed in the fractions isolated from yeast transformed with an empty vector (data not shown). When yeast crude extracts were sub-fractionated by differential centrifugation, only microsomes and LDs contained catalytic activity and the LDs from two recombinant yeast strains were more active than their microsomes (Figure 4A). However, the LDs isolated from yeast/pVT102U/CLO2 were approximately 6-fold less active in epoxidation reactions than LDs isolated from yeast/pVT102U/CLO4, while the hydroxylation activity in both was similar.

To confirm and further characterize the enzymatic activity related to PdCLO2 and PdCLO4, we purified each His-tagged isoform separately. LD proteins were solubilized (about 65%) by 0.2% emulphogene and purified by affinity chromatography on a Ni²⁺ column with a purification factor of 11.3 to 11.9 (Table 2). When analyzed by SDS-PAGE, the purified fraction for each protein showed one major band at 26.5 kDa with a degree of purity of >98% as estimated by scanning densitometry and western blotting of the purified proteins with an anti-His antibody confirm their fusion with poly-His tail (Figure 4B). The purified recombinant PdCLO2 and PdCLO4 proteins catalyzed the oxidation of aniline (1.52 versus 1.61 μmol min⁻¹ mg⁻¹ of protein, respectively) and oleic acid (0.2 vs. 1.05 μmol min⁻¹ mg⁻¹ of protein, respectively) in the presence of cumene hydroperoxide as co-substrate. Heat-inactivated fractions were unable to catalyze such reactions (Figure 4C). These data strongly suggest that peroxygenase activity in the date palm tissues analyzed is associated with LDs, and to a slightly lesser extent with microsomal membranes.

PdCLO2 and PdCLO4 catalyzed epoxidation of oleic acid, a monounsaturated fatty acid (C18:1) at different rates (Figure 4C). Although the majority of plant oxylipins derive from linoleic and linolenic acids (C18:2 and C18:3) (Calvo et al., 1999), these fatty acids were only present in small amounts in date palm seeds (< 10%) while oleic acid (C18:1) constituted about 45% of the total fatty acid fraction (Basuny and Al-Marzooq, 2011). Therefore, the ability of PdCLO2 and PdCLO4 to epoxide mono- and/or polyunsaturated fatty acids was evaluated. [¹⁴C]-labeled palmitic acid (C16:1), oleic acid (C18:1), linoleic acid (C18:2), or linolenic acid (C18:3) were individually incubated with recombinant PdCLO2 and PdCLO4 in the presence of cumene hydroperoxide, H₂O₂ or 9-HPOD and 13-HPOD and their further metabolism followed by radio-TLC. Regardless of the hydroperoxide used, PdCLO2 preferentially epoxidised C16:1 and C18:1 but had much lower activities with C18:2 and C18:3. Conversely, C18:2 > C16:1 > C18:1 were epoxidised by PdCLO4. The most efficient co-substrates for PdCLO2 were 9-HPOD followed by 13-HPOD, H₂O_2, and cumene hydroperoxide. In contrast, PdCLO4 preferred 13-HPOD as a co-substrate. Linolenic acid (C18:3) was poorly epoxidised by both isoforms (Figure 5A).

Next the metabolism of fatty acid hydroperoxides in the presence of the purified recombinant proteins PdCLO2 and PdCLO4 was analyzed by HPLC. After 2 h of incubation, PdCLO2 poorly reduced 13-HPOD but more actively reduced 9-HPOD, with about 81 and 16% of the substrates remaining intact respectively (peaks 1 and 2 in Figure 5B, panel I). Conversely, PdCLO4 was more active in the reduction of 13-HPOD rather than 9-HPOD where about 10 and 85% of the substrates remained intact, respectively (peaks 1 and 2 in Figure 5C, panel I). Peak 3 and peak 4 were identified respectively as the alcohol products 13-HOD and 9-HOD, by co-elution with standards (panel II). No metabolites were detected from incubations of 13-HPOD and 9-HPOD with heat-inactivated PdCLO2 and PdCLO4 (panel III). These data indicate that PdCLO2 and PdCLO4 are differentiated by their substrate specificities, with PdCLO2 showing more specificity to reduce 9-HPOD, while PdCLO4 more actively reduced 13-HPOD.

The 13- and 9-HPOD-reductase activities in plumule, petiole, and radicle extracts from seedlings in stage II and III were analyzed after exposure to various doses of TCDD. 13-HPOD-reductase activity was significantly increased about 3-fold and more than 5-fold in plumule extract of stage II and III, respectively, as a response to the highest dose of TCDD-exposure (Figure 6). The relatively small 9-HPOD-reductase activity of plumule extracts did not change after TCDD-treatments. Interestingly, the enzymatic activity of radicle extracts in transforming either 13- or 9-HPOD as a function of TCDD treatment increased progressively in stage II and stage III and was maximal in stage III radicle extracts at the highest dose of TCDD (Figure 6). These data show a possible induction of PXG activity in leaves and more particularly in roots of seedlings exposed to TCDD.

In the oxylipin pathway immediately upstream of the PXG activities characterized above are 9-LOX and 13-LOX enzymes that catalyze the hydroperoxidation of unsaturated fatty acids (Feussner et al., 2001). The quantification of transcripts levels for 9- and 13-LOX-like genes, (Figure 7) showed that transcripts of 9-LOX-like genes were specifically increased in root tissues that had been exposed to TCDD. This increase occurred progressively in the roots of both stage II and III and reached its maximum (about 18-flod) in the roots of seedling at stage III after administration of 50 ng L⁻¹ TCDD (Figure 7A). While the transcripts of 13-LOX-like genes also increased in plumule and in roots of both stages as a function of TCDD-exposure, the increase of 13-LOX-like transcripts was higher in the plumule than in the root (~16-fold vs. ~12-fold). Similar to their transcriptional induction, 9-LOX and 13-LOX activities were significantly and differentially augmented in seedling tissues after treatment with TCDD (Figure 7B). C18:2 was more actively transformed into its corresponding hydroperoxide (9-HPOD) under the action of the root extract and the maximal 9-LOX activity was detected in root extracts from both stages II and III (~ 2.5-fold) compared the root extracts of control plants. In contrast, the activity of 13-LOX in hydroperoxidation of C18:2 was similarly stimulated in the plumule as well as in the radicle of TCDD-treated-seedlings (~2-fold) (Figure 7B). Altogether these data indicate that 9- and 13-LOX activities are induced in a tissue specific manner by TCDD. While 9-LOX activity was uniquely stimulated in roots, similar activation in 13-LOX activity was found in both plumule and roots.

Bioinformatic analysis showed that the genome of date palm harbors nine potential orthologs of lipoxygenase-like (LOXs-like) genes, six of which encode putative 13-LOX proteins and three of which encode putative 9-LOX proteins as summarized in Table 3. This Table also shows comparative data from two closely related monocot tree crop plants, i.e., oil palm and banana, for which full genomic sequence data are available, plus the model dicot species, Arabidopsis thaliana. In general the 9-LOX proteins are slightly smaller than the 13-LOX proteins with lengths ranging from 847 to 885 and 865 to 926 residues respectively. In the particular case of date palm sequences, the 9-LOX and 13-LOX proteins ranged from 850 to 885 and 896 to 916 residues respectively. With only one exception, all of the 9-LOX sequences lacked a predicted plastid signal peptide as determined by several independent algorithms. The anomalous Arabidopsis sequence, At_9-LOX2 is annotated on the GenomeNet database as a chloroplastic protein

A phylogenetic analysis of the 38 LOX sequences from the four selected plant species is presented in Figure 8. This clearly shows the clustering of the three LOX groups, namely 9-LOX, 13-LOX type I, and 13-LOX type II with each of the four analyzed plant species having at least one isoform of each of the three LOX groups. Fully aligned sequences of the 15 identified 9-LOX and 19 identified 13-LOX proteins from the four selected plant species are presented in Figures S1, S2. In both the 9-LOX and 13-LOX proteins there are clusters of highly conserved amino acids located throughout these large proteins, which total between 847 and 926 residues in length. In Figures S3, S4 more detailed alignments are shown in the central region that includes the canonical histidine-rich domain that is a feature of all such lipoxygenases. This 38-residue motif has the following structure: [-H-X₄-H-X₄-H-X₁₇-H-X₈-H-] and is located at approximately residues 550–590 in 9-LOX and residues 610–650 in 13-LOX sequences. In addition to the five invariant histidines seen in all LOX proteins, well over half of the other residues in this canonical plant 9-LOX and 13-LOX domain were absolutely conserved even between species as divergent as oil palm and Arabidopsis. This indicates that not only the metal- and heme- coordinating histidines have important catalytic and structural functions but also most of the residues depicted above as “X.”

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.