The cDNA Structures and Expression Profile of the Ascorbate Peroxidase Gene Family During Drought Stress in Wild Watermelon

Ascorbate peroxidase (APX) plays an important role in detoxifying reactive oxygen species under environmental stress. Although previous work in drought-tolerant wild watermelon has shown an increase in chloroplast APX enzyme activity under drought, molecular entities of APX have remained uncharacterized. In this study, structure and transcriptional regulation of the APX gene family in watermelon were characterized. Five APX genes, designated as CLAPX1 to CLAPX5, were identified from watermelon genome. The mRNA alternative splicing was suggested for CLAPX5, which generated two distinct deduced amino acid sequences at their C-terminus, in resemblance to a reported alternative splicing of chloroplast APXs in pumpkin. This observation suggests that two isoenzymes for stromal and thylakoid-bound APXs may be generated from the CLAPX5 gene. Phylogenetic analysis classified CLAPX isoenzymes into three clades, i.e., chloroplast, microbody, and cytosolic. Physiological analyses of wild watermelon under drought showed a decline in stomatal conductance and CO2 assimilation rate, and a significant increase in the enzyme activities of both chloroplast and cytosolic APXs. Profiles of mRNA abundance during drought were markedly different among CLAPX genes, suggesting distinct transcriptional regulation for the APX isoenzymes. Up-regulation of CLAPX5-I and CLAPX5-II was observed at the early phase of drought stress, which was temporally correlated with the observed increase in chloroplast APX enzyme activity, suggesting that transcriptional up-regulation of the CLAPX5 gene may contribute to the fortification of chloroplast APX activity under drought. Our study has provided an insight into the functional significance of the CLAPX gene family in the drought tolerance mechanism in this plant.


Introduction
Drought-associated water deficit is one of the major factors restricting plant productivity and crop yields worldwide.To sustain themselves when exposed to water deficit, plants activate various physiological and metabolic mechanisms that protect them from adverse physicochemical injuries.Some of the key molecules generated in large quantities in plant cells during environmental stress are the reactive oxygen species (ROS), such as hydrogen peroxide (H 2 O 2 ), superoxide radical (O 2 -), and hydroxyl radical (•OH) (Gill & Tuteja, 2010;You & Chan, 2015).These species oxidize various cellular components, such as nucleic acids, lipids, and proteins, which can cause lethal damage to plants.Plants are equipped with several enzymatic and non-enzymatic systems that decompose these ROS and maintain them at nontoxic levels (Mittler et al., 2004).The enzymes that scavenge ROS include superoxide dismutase (SOD), catalase (CAT), glutathione peroxidase (GPX), and ascorbate peroxidase (APX).Moreover, an array of non-enzymatic antioxidants, such as ascorbate, glutathione, carotenoids, and tocopherols, serve as defense agents for protecting plant cells from oxidative injuries.These enzymes and antioxidants are mutually related in the glutathione-ascorbate cycle, which plays an essential role in resisting water deficit and oxidative stress under drought (Li et al., 2013).Nevo, 2014;Brozynska et al., 2016).One of the most drought-tolerant wild plant species is wild watermelon (Citrullus lanatus), which inhabits very harsh conditions in the Kalahari Desert and has the capacity to thrive under high light and drought stress (Akashi et al., 2008;Yoshimura et al., 2008).Although the fruit of this plant is not palatable to humans, it is a very important source of water for wildlife inhabitants in the desert, especially during long dry spells.Unraveling the mechanisms responsible for drought tolerance in the wild watermelon will offer valuable insights for breeding other crops, which are currently susceptible to adverse drought effects, toward improved drought tolerance.
A previous study showed that the enzyme activity of chloroplast APX increased in the leaves of wild watermelon under drought stress (Nanasato et al., 2010).APX catalyzes the conversion of H 2 O 2 to water with concomitant oxidation of ascorbate to monodehydroascorbate, thereby playing a pivotal role in the detoxification of ROS under biotic or abiotic stresses (Asada, 2006;Miyake et al., 2006).Extensive research has shown that APX isoenzymes are localized in at least three different subcellular compartments in plant cells, namely, cytosol (cAPX), microbody (mAPX), and chloroplast (Ishikawa & Shigeoka, 2008).The chloroplastic APXs are further categorized into at least two isoenzymes according to their distinct microenvironments, i.e. stroma-soluble (sAPX) and thylakoid membrane-bound (tAPX) forms.Interestingly, plants appear to be divided into two groups according to their mode of biogenesis of sAPX and tAPX.In the first plant group (which includes Arabidopsis, rice, and tomato), sAPX and tAPX are encoded by distinct genes.In the second plant group (which includes spinach, tobacco, and pumpkin), on the other hand, sAPX and tAPX are encoded by a single gene, and their protein products are generated by post-transcriptional alternative splicing of the mRNA precursors (Ishikawa & Shigeoka, 2008).
Another feature of APX enzymes is their instability in the absence of ascorbate (Hiner et al., 2000;Kitajima et al., 2006).APX isoenzymes in chloroplasts are particularly sensitive to inactivation under ascorbate-deficient conditions, in comparison to their cAPX and mAPX counterparts (Yoshimura et al., 1998).Although increased expression of cAPX under various abiotic stresses is well documented in plants (Karpinski et al., 1997;Shigeoka et al., 2002), reports on the up-regulation of chloroplast APXs under drought and high light stress have been limited, and even a decrease in chloroplast APX activity was reported in spinach under high light stress (Yoshimura et al., 2000).These observations suggest that chloroplast APX may be one of the initial targets for oxidative injuries in plant leaves under drought-and light-related stress (Shikanai et al., 1998).In this regard, the observed increase in the chloroplast APX activity in wild watermelon under water deficit stress (Nanasato et al., 2010) is intriguing because it may suggest that this plant fortifies the activity of chloroplast APXs that are otherwise vulnerable to drought and high light conditions.However, to our best knowledge, no report has been found on the structure and mRNA expression profiles of the APX gene family in watermelon species.Therefore, in this study, we attempted to characterize the gene organization and transcriptional regulation of the putative APX gene family in wild watermelon, together with the physiological responses of this plant during drought and high light stress conditions.

Plant Materials and Growth Conditions
Wild watermelon (Citrullus lanatus Acc.No. 101117-1) (Yoshimura et al., 2008) were self-pollinated at least three times and their seeds were used in this study.The seeds were soaked overnight in water at 30 °C in the dark and planted in pots filled with a horticulture soil the following morning.The germinated seedlings were grown in a growth chamber under LED lights with a light intensity of 800-1,000 μmol photons m -2 s -1 under a 14 h light and 10 h dark photoperiod, with an air temperature of 30 °C and relative humidity of 50%, for the entire growth and monitoring periods.After germination the plants were watered daily with supplementation of a 1,000-fold diluted Hyponex nutrient solution (Hyponex Japan Corp., LTD, Osaka, Japan) twice a week.When they reached the stage when the fifth true leaf had fully expanded, drought stress was introduced by withholding watering.

Characterization of Gene and cDNA Structures for Watermelon APXs
Watermelon cDNA and genomic sequences for the putative APX genes were searched using TBLASTN against a genome sequence database of C. lanatus subsp.vulgaris cv.97103 in the Cucurbit Genomics Database (Guo et al., 2013), with known Arabidopsis APX sequences as the queries, using BLOSUM62 matrix and a setting of gap opening and extension penalties for 11 and 1, respectively, and a threshold E-value of 1e-10.The genome and unigene sequences, and chromosome locations of watermelon APX genes were obtained from the Cucurbit Genomics Database.
For PCR amplification of the whole coding DNA sequences (CDSs) of wild watermelon, a leaf of the watermelon that was exposed to water deficit stress for three days was used for total RNA extraction.The leaf sample was snap frozen into liquid nitrogen and stored at -80 °C until RNA isolation.Total RNA was extracted with the Spectrum Plant Total RNA Kit (Sigma Aldrich, St Louis, MO, USA), and trace amounts of genomic DNA were degraded using the On-Column DNase I Digestion Set (Sigma Aldrich).The cDNA syntheses were performed using the ReverTra-Ace- synthesis kit (Toyobo, Osaka, Japan) with an Oligo(dT) primer (Toyobo).To design primers for amplifying whole CDSs, sequences flanking the CDSs, i.e., upstream of the translation start codons and downstream of the stop codons (Table A1 in Appendix), in the unigene information from cv. 97103, were employed.The 3'-RACE was performed according to the instructions from a 3'-Full RACE Core Set (Takara, Shiga, Japan).The PCR amplification was performed by a KOD FX NEO high-fidelity proofreading enzyme (Toyobo).The PCR products were separated in an agarose gel electrophoresis and amplicons were purified from the agarose gel using a MinElute Gel Extraction Kit (QIAGEN, Germantown, MD, USA), and then sub-cloned into an Invitrogen TOPO-BLUNT vector (Life Technologies, Carlsbad, CA, USA).Sequence reactions were performed using a BigDye terminator v3.1 Cycle Sequencing Kit (Life Technology) and analyzed using a 3130xl DNA sequencer (Applied Biosystems, Foster City, CA, USA).
Intracellular locations of these gene products were predicted by the WoLF PSORT and DeepLoc-1.0programs (Horton et al., 2007;Armenteros et al., 2017).Visualization of the exon-intron structures of the genes was performed by the GSDS 2.0 gene feature visualization server (Hu et al., 2015).ClustalW and ETE3 programs in the GenomeNet server (Kanehisa et al., 2002) were used to generate an amino acid sequence alignment and phylogenetic tree.

Measurement of Plant-Water Relations
Gravimetric soil moisture content was determined essentially as described by Reynolds (1970), with the following minor modification: after harvesting leaf samples, the entire aboveground plant tissues were cut out.The moist soil, together with the planting pot, were then weighed and recorded as the wet mass (WM).The soil was oven dried at 105 °C for 72 h, then weighed and recorded as the dry mass (DM).The soil moisture content (θg) was determined by the following formula (Reynolds, 1970): Leaf relative water content was measured essentially as described by Barrs (1968), with the following modification: the leaves were harvested and the fresh weight was quickly measured on a Unibloc AUX 120 balance (Shimadzu, Kyoto, Japan) and recorded as the fresh weight (FW).After the FW were recorded, the leaf samples were placed in zip-lock plastic bags, which were filled with distilled water and kept overnight at 25 °C.The next day, excess water was removed by blotting the leaves in paper towels.The water-saturated leaves were then weighed and recorded as turgid weight (TW).The turgid leaves were then oven dried at 80 °C for 3 d and their weights were recorded as the dry weights (DW).The leaf relative water content (LRWC) was calculated by the following formula (Barrs, 1968): LRWC = (FW -DW)/(TW -DW) × 100 (2)

Measurements of Photosynthetic Parameters
Leaf chlorophyll contents were measured using a SPAD-502plus meter (Konica Minolta, Tokyo, Japan).Leaf stomatal conductance was measured by an SC-1 leaf porometer (Decagon Devices, Pullman, WA, USA) 5 h after the onset of the light regime.CO 2 assimilation and chlorophyll fluorescence were measured in the third true leaves using an open gas exchange system LI6400XT photosynthesis meter (LI-COR Biosciences, Lincoln, NE, USA).A 2 cm radius IRGA gas chamber was used for all the measurements, with the chamber temperature set at 25 °C, CO 2 flow rate at 400 μmol mol -1 , light intensity at 1,000 μmol photons m -2 s -1 , and relative humidity at 50%.CO 2 assimilation was measured 3 h after the onset of the light regime, while the chlorophyll fluorescence of dark-adapted leaves was measured early morning, before the onset of the light regime, after the plants were kept in darkness overnight.

APX Enzyme Assay
Crude leaf extracts were prepared essentially as described (Nanasato et al., 2010), with the following minor modifications: approximately 200 mg of leaf tissues were ground to a fine powder using a pestle and mortar, with the aid of liquid nitrogen, in 1 ml of homogenization buffer containing 50 mM potassium phosphate, pH 7.0, , and 2% polyvinylpolypyrrolidone.The detergent CHAPS was included in the buffer to solubilize thylakoid-bound APX (Veljovic-Jovanovic et al., 2001).The homogenized samples were centrifuged at 12,000 × g for 20 min at 4 C and then the supernatant was collected in a new tube.The extracts were desalted by running them through an Amicon Ultracel 3K filter (Merck Millipore, Burlington, MA, USA), and their protein concentration was quantified using Protein Assay CBB Solution (Nacalai, Kyoto, Japan) and the Multiskan FC plate reader (Thermo Fisher Scientific, Waltham, MA, USA), using bovine serum albumin as the standard.
APX enzyme activity was measured essentially as described (Amako et al., 1994;Nanasato et al., 2010), with the following modifications: the reaction was performed in 1 ml of assaying buffer containing 50 mM potassium phosphate buffer, pH 7.0, 1 mM sodium ascorbate, 10 μl of the crude leaf extract, and 1 mM of H 2 O 2 .The assay was started by the addition of H 2 O 2 substrate and a solution without the H 2 O 2 substrate was used as a reference.The oxidation of ascorbate was continuously monitored by optical absorbance at 290 nm using a UH5300 spectrophotometer (Hitachi, Tokyo, Japan) and an absorption coefficient of 2.8 × 10 -3 M cm -1 (Nakano & Asada, 1981) was used for the calculation of reaction rates.To separately quantify the two activities of chloroplast and cytosolic isoenzymes, the plant protein was incubated in the assay mixture with 10 M of H 2 O 2 without ascorbate for 5 min, and then residual activity was assayed as the cytosolic isoenzyme.The total APX was quantified without any prior incubation with H 2 O 2 and used to calculate the value of chloroplast isoenzyme activity (Amako et al., 1994).

Quantification of APX mRNA Expression
Pairs of specific primers used for a RT-qPCR analysis of wild watermelon APX genes (Table A1) were designed using the Primer3 online tool (Untergasser et al., 2012).Total RNAs were extracted from the leaves of wild watermelon stressed for 0, 3, 5, 7, 9, and 11 days, and cDNA synthesis were performed as described in Section 2.2.The mRNA abundance of the APX genes was measured by a Light-Cycler 480 (Roche Diagnostics, Mannheim, Germany), using a LightCycler 480 SYBR Green I Master Kit (Roche), according to the manufacturer's instruction.As reference genes, three sets of primers for γ-actin (ylsACT), α-tubulin (ylsTUB), and glyceraldehyde-3-phosphate dehydrogenase (ylsGAPDH), which showed highly homogeneous expression in a wide range of tissue types, developmental stages and environmental stimuli in watermelon (Kong et al., 2014; Table S1), were used as controls, and their normalized value (Vandesompele et al., 2002) was used to calculate relative abundance of the APX mRNAs.The profiling of mRNA quantification was run with three biological replications, each consisting of an average of three technical replications.

Structures of the Putative APX Genes in the Watermelon Genome
Using the protein sequences deduced from all APX genes in Arabidopsis thaliana (Panchuk et al., 2002; Table A2) as the queries, we identified five homologous genes in the whole genome sequence of cultivar watermelon (Citrullus lanatus subsp.vulgaris cv.97103) in the Cucurbit Genomics Database.These five putative APX genes were designated as CLAPX1 to CLAPX5, according to the order of highest sequence similarity with the translated sequence of Arabidopsis AtAPX1 (Table 1).These five genes were located on four different chromosomes, i.e., 1, 2, 3, and 8. Two genes for CLAPX1 and CLAPX3 on chromosome 2 were approximately 27 Mbp apart, showing that the watermelon APX genes were not clustered but rather scattered in the genome.From the comparison between the genomic sequence and assembled EST sequences, or unigenes, 9-12 exons were predicted in these CLAPX genes (Figure 1).A nucleotide sequence alignment between CLAPX5 and the pumpkin chloroplast APX gene in their C-terminal region is shown.The region spanning exon 11 and exon 12 is presented.Identical nucleotides between the two sequences are labeled as vertical lines and different nucleotides are labeled as #.Deduced amino acid sequences for CLAPX5 and pumpkin chloroplast APX are presented on top of and below the nucleotide sequences, respectively.Terminator codons are indicated by asterisk (*).A conserved "gt" motif at the beginning of intron 11 is boxed and labeled as (i), while the conserved "ag" motif for the termination of intron 11 for CLAPX5-II and CLAPX5-I are also boxed and labeled as (ii) and (iii), respectively.

Copyrights
Copyright for this article is retained by the author (s), with first publication rights granted to the journal.
This is an open-access article distributed under the terms and conditions of the Creative Commons Attribution license (http://creativecommons.org/licenses/by/4.0/).
Figur Note.Chan (d) during (a), the va mean±SD differences

Figure
Figure A1.C-terminus amino acid alignment between CLAPX5 and pumpkin chloroplast APXsNote.Amino acid alignments of C-terminal region for a pair of CLAPX5-I and pumpkin thylakoid Cka.tAPX (a), and a pair of CLAPX5-II and pumpkin stromal Cka.sAPX (b).The region spanning exon 11 and exon 12 is shown for each pair.Identical amino acids are labeled by vertical lines, similar amino acid residues between the pair are labeled as +, and amino acids with different chemical properties are labeled as #.

Figure A3 .
Figure A3.The observed cleavage sites for addition of poly(A) tails, and putative cis-acting poly(A) signals in the CLAPX5 gene Note.The positions of observed proximal and distal cleavage sites (CSs) for the addition of poly(A) tails were shown by the downward arrows, and the conserved CS motifs were indicated with asterisks on top of the dinucleotides.The hexanucleotides for the putative near upstream elements (NUEs) of the poly(A) signals (Loke et al., 2005) are shown by the underlines.

Table A2 .
Accession numbers for plant APX genes that were used in this study.
Note. *1 For Arabidpsis, AGI locus identifiers were shown.For spinach and pumpkin, NCBI/EMBL/DDBJ accession numbers were shown.