Plant materials and growth conditions
Mature durian (Durio zibethinus L. ‘Chanee’) fruits were harvested 95 d after anthesis from a commercial orchard in Trat Province, Thailand in early April 2017. The samples were maintained at room temperature (30 °C) before being peeled on days 1, 2, and 4 (unripe, midripe, and ripe postharvest stages, respectively). Monthong served as a representative sample for gene expression analysis. Its three different ripening stages were also evaluated but their timings (days) slightly differed from those of Chanee. Monthong fruits were harvested at day 105 after anthesis. Unripe, midripe, and ripe Monthong fruit samples were peeled and analysed at 1 d, 3 d, and 5 d after storage at room temperature, respectively .
To investigate the association between DzLAP1 and durian fruit ripening, unripe Monthong samples were treated either with ethephon or 1-methylcyclopropene (1-MCP). These synthetic phytohormones have opposite modes of action. Ethephon is converted to ethylene which enhances ripening. In contrast, 1-MCP is an ethylene antagonist and delays ripening. The treated samples were compared with controls naturally ripened according to the method of Khaksar et al. (2019). Three biological replicates were used and each comprised a single durian fruit harvested from a separate tree.
Crude extract from durian pulp was used to determine Cys-Gly dipeptidase activity. Durian cultivar Chanee was obtained from a local market in Nonthaburi Province, Thailand. Postharvest samples were collected at the unripe and midripe stages. Three biological replicates were used and each consisted of a single lobe harvested from a separate durian fruit.
Nicotiana benthamiana plants were raised for agroinfiltration. Seeds were sown on peat moss and the seedlings were grown at 25 °C under a 16 h/8 h light/dark photoperiod and 4,500 lx (artificial light). Two-week-old plants were individually transplanted to pots and raised under the same conditions for another 2 wks.
Phylogenetic analysis and putative LAP identification in durian fruit
The protein sequences of LAPs harbouring Cys-Gly dipeptidase activity in Treponema denticola (accession no. WP_010698434.1)  and Arabidopsis thaliana (AtLAP1 and AtLAP3; accession nos. P30184.1 and Q8RX72.1, respectively)  served as queries for a BLAST search against the D. zibethinus cultivar Musang King NCBI database. The MaGenDB database  confirmed DzLAP isoforms.
To establish the phylogenetic relationships among LAPs, the amino acid sequences of the putative DzLAPs and other LAPs deposited in NCBI were subjected to ClustalW multiple alignment. A neighbour joining (NJ) tree was created with MEGA v. 7  using 1,000 bootstrap replicates.
Determination of tissue-specific DzLAP1 expression
A search of the DzLAPs against the genomic data for durian cultivar Musang King disclosed two candidate genes (accession nos. XM_022894525.1 (LOC111299369) and XM_022874012.1 (LOC111284913)) annotated as DzLAP1-like and named DzLAP1_MK and DzLAP2_MK, respectively. Attention was directed to the durian fruit pulp as it accumulated several sulphur volatiles. DzLAP1 expression was analysed in silico in various fruit tissues. To compare relative DzLAP expression in different tissues, normalised total read counts (RCs) derived from Illumina reads were obtained from the Sequence Read Archive (SRA) resource and processed according to the method of Khaksar et al. (2019). RNA-seq data were obtained for SRX3188225 (root), SRX3188222 (stem), SRX3188226 (leaf), and SRX3188223 (aril/pulp) . A heatmap based on the RCs was constructed using MetaboAnalyst v. 4.0 .
Gene expression analysis by qRT-PCR
Total RNA was isolated from Chanee and Monthong durian cultivar pulps with PureLink® plant RNA reagent (Thermo Fisher Scientific, Waltham, MA, USA) following the manufacturer’s instructions. DNase-treated RNA sample quantity and integrity were assessed. Approximately 1 µg total RNA was reverse-transcribed to cDNA with a RevertAid first-strand cDNA synthesis kit (Thermo Fisher Scientific, Waltham, MA, USA). Gene-specific primers listed in Supplementary Table S1. The qRT-PCR elucidated DzLAP1 expression in unripe, midripe, and ripe durian fruit. The reactions were performed in 10 μL total volume in a 96-well PCR plate. The cDNA and primers were combined with Luna® universal qPCR master mix (New England Biolabs, Ipswich, MA, USA). PCR was run in a CFX Connect™ real-time PCR detection system coupled to CFX Manager™ (Bio-Rad Laboratories, Hercules, CA, USA). Single amplicon production was verified by melting curve analysis. Relative gene expression levels were calculated by the 2−ΔΔCt method  based on the cycle threshold (Ct) of the gene relative to the reference gene elongation factor 1 alpha (EF-1α). There were three independent biological replicates in the qRT-PCR. Gene expression analyses were also conducted on the ethephon- and 1-MCP-treated samples and the naturally un/ripened samples (controls).
In planta Cys-Gly dipeptidase activity assay
To determine Cys-Gly dipeptidase activity in durian fruit pulp, crude enzyme was extracted from it at the unripe and midripe stages. Pulp samples were separately collected, frozen in liquid nitrogen, and pulverised in the MM400 mixer mill (Retsch GmbH, Haan, Germany) at 30 Hz for 1 min. Then 250 mg of each sample was dissolved in 2.5 mL lysis buffer (50 mM K3PO4, pH 8.0) and gently mixed at 4 °C for 1 h. The samples were centrifuged at 14,000 × g and 4 °C for 5 min and the supernatant was collected. An enzyme activity assay was performed by incubating durian fruit pulp extract with 20 mM Cys-Gly in reaction buffer (50 mM K3PO4 buffer (pH 8.0) + 1 mM MgCl2) at 30 min, 1 h, and 3 h. The enzyme activity was measured by a modified acidic ninhydrin method . Briefly, 50 µL of each enzyme reaction system was terminated with 50 µL glacial acetic acid followed by 50 µL acidic ninhydrin solution (250 mg ninhydrin in 6 mL glacial acetic acid + 4 mL HCl), boiled for 9 min, and cooled with tap water. The pink endpoint indicated the reaction between the released Cys and ninhydrin under acidic conditions. Colour intensity was measured spectrophotometrically at A560 (BioTex, Winooski, VT, USA). Three independent biological replicates were used.
In planta Cys-Gly dipeptidase activity was also investigated. Full-length DzLAP1 was amplified with Phusion Hot Start II high-fidelity DNA polymerase (Thermo Fisher Scientific, Waltham, MA, USA) using Chanee cDNA as a template. The gene-specific primers (excluding the stop codon) listed in Supplementary Table S1 were used in the PCR. The PCR product was cloned into a pCR™8/GW/TOPO® TA cloning vector (Invitrogen, Carlsbad, CA, USA) and generated pTOPO-DzLAP1. The latter was then sequenced. DzLAP1 was transferred to the pEAQ3 destination vector and fused at the C-terminal with 6× His  via Gateway® LR Clonase® II (Invitrogen, Carlsbad, CA, USA). The pEAQ-DzLAP1 product was transformed into Agrobacterium tumefaciens GV3101 by electroporation.
A. tumefaciens bearing the DzLAP1-6xHis or the empty pEAQ vector (control) was infiltrated into 4 wk plants. Briefly, cells from each culture were washed and suspended in MM buffer (10 mM MES buffer + 10 mM MgCl2; pH 5.6). The cell suspension was adjusted at A600 to OD = 0.6. Acetosyringone was added to make up 200 µM final concentration and the suspension was kept in the dark at 30 °C for 2 h before infiltration. At day 3 after agroinfiltration, the leaves were infiltrated with 15 mM Cys-Gly, incubated for 1 h , freeze-dried, and kept in a dry place at 30 °C until subsequent metabolite analysis by HPLC. To measure the foliar Cys-Gly levels, the co-infiltrated leaves were pulverised in the MM400 mixer mill (Retsch GmbH, Haan, Germany) at 30 Hz for 1 min. Each sample was suspended in 0.1 M HCl extraction buffer at 1 mg/50 µL. Then, 10 mM N-acetylcysteine (internal standard) was added and mixed by shaking at 250 rpm and 37 °C for 2.5 h. The samples were centrifuged at 14,000 × g and 30 °C for 5 min. The soluble fraction was transferred to a new microcentrifuge tube, combined with acetonitrile (1:1), and centrifuged at 14,000 × g and 30 °C for 5 min. The pellet was removed and the supernatant was dried with a CentriVap benchtop vacuum concentrator (Labconco Corp., Kansas City, MO, USA). The dried samples were re-dissolved in 200 µL deionised H2O, passed through a 0.22-µm syringe filter, and subjected to HPLC analysis. Ten microlitres of each sample was injected into a C18 column (250 mm × 4.6 mm; Phenomenex, Torrance, CA, USA) using acetonitrile (5%) in 2% perchloric acid as a mobile phase. The flow rate was 1 mL/min and the temperature was 40 °C. The Cys-Gly peak was detected at 210 nm . Three independent biological replicates were used and each consisted of a separate leaf.
DzLAP1 cloning and expression in Escherichia coli
The putative DzLAP1 was amplified with Phusion Hot Start II DNA polymerase (Thermo Fisher Scientific, Waltham, MA, USA) and midripe Chanee durian cultivar cDNA served as a template. The PCR temperature profile was as follows: initial denaturation at 98 °C for 30 s; 30 cycles of 98 °C for 10 s; 57 °C for 10 s; 72 °C for 1 min; and a final extension at 72 °C for 5 min. Gene-specific forward and reverse primers were designed according to the DzLAP1_MK sequence (Supplementary Table S1). The signal sequences predicted by the ChloroP1.1 and TargetP servers were excluded. The putative DzLAP1 PCR product was excised with restriction enzymes (FastDigest™; Thermo Fisher Scientific, Waltham, MA, USA) and cloned into a pET21b vector (Merck KGaA, Darmstadt, Germany). The product was pET21b-DzLAP1 in-frame with 18 nucleotides encoding 6× His residues at the C-terminus. It was transformed into E. coli TOP10 (K2500-20; Invitrogen, Carlsbad, CA, USA). Bacterial colonies were raised on LB agar supplemented with 1 mg mL-1 ampicillin and analysed by colony PCR. The nucleotide sequences of the positive clones were verified by 1st BASE DNA sequencing.
Recombinant DzLAP1 (rDzLAP1) production and purification
The pET21b-DzLAP1 was transformed into the T7 host E. coli SoluBL21 (DE3). Cells harbouring the recombinant plasmid were incubated overnight in LB broth supplemented with 1 mg mL-1 ampicillin. The starter culture was inoculated into fresh LB broth supplemented with 1 mg mL-1 ampicillin and incubated at 37 °C with shaking at 250 rpm until OD600 = 0.4 – 0.5. The rDzLAP1 was generated by induction with 1 mM isopropyl β-D-1-thiogalactopyranoside (IPTG) at 30 °C and shaking at 250 rpm for 17 h. The cells were harvested by centrifugation at 5,000 × g and 37 °C for 10 min, suspended in buffer A (25 mM K3PO4 buffer + 0.3 M NaCl; pH 7.2), and lysed by ultrasonication. Soluble intracellular proteins were collected by centrifugation at 7,500 × g and 4 °C for 30 min, analysed by western blot (6× His epitope tag antibody; Thermo Fisher Scientific, Waltham, MA, USA), and stored at 4 °C until purification.
The crude extract was loaded onto a HisTrapTM column (Merck, Darmstadt, GER) pre-equilibrated with buffer A. The column was washed with excess buffer A and the rDzLAP1 was eluted with buffer B (25 mM K3PO4 buffer + 0.3 M NaCl + 150 mM imidazole; pH 7.2). The samples were analysed by SDS-PAGE and western blot. The pooled purified rDzLAP1 fraction was dialysed against 25 mM K3PO4 buffer (pH 7.2). The protein concentration was determined by a modified Bradford assay . The reference protein standard was BSA.
Enzymatic rDzLAP1 assay
The metal ion dependency of rDzLAP1 was evaluated. The enzyme was incubated at 37 °C for 15 min with 7.5 mM Cys-Gly in 25 mM K3PO4 buffer (pH 7.2) in the presence of 0 mM or 1 mM Ca2+, Zn2+, Mg2+, Ni2+, or Mn2+. The total reaction volume was 50 µL. Enzyme activity was measured by a modified acidic ninhydrin method . To determine the optimum enzyme pH, 50-µL reaction systems were prepared by incubating rDzLAP1 with 7.5 mM Cys-Gly and 1 mM Mg2+ at various pH (acetate buffer, pH 4 – 6; phosphate buffer, pH 6 – 8; Tris-HCl buffer, pH 8 – 9.5; glycine-NaOH buffer, pH 9.5 – 11) at 37 °C for 15 min. Maximum enzyme activity was defined as 100% relative activity.
To assess enzyme kinetics, 3.5 µg purified rDzLAP1 was incubated with 0–10 mM Cys-Gly, γ-Glu-Cys, GSH, Met-Gly, or Leu-Gly in the presence of 1 mM MgCl2 in 25 mM K3PO4 buffer (pH 8.0) at 37 °C for a specific length of time. The reactions were terminated with 0.13 N HCl. For γ-Glu-Cys and GSH, the reactions were evaluated by a modified acidic ninhydrin assay. For Cys-Gly, Met-Gly, and Leu-Gly, the reactions were analysed by monitoring the decrease in absorbance of the free peptides at A230, with minor modifications . Enzyme kinetics were measured with OriginPro® 2017 (OriginLab Corp., Northampton, MA, USA).
Full-length DzLAP1 was amplified and cloned into a pCR™8/GW/TOPO® TA vector (Invitrogen, Carlsbad, CA, USA) according to the in planta Cys-Gly dipeptidase activity assay. The pTOPO-DzLAP1 product was then sequenced. The DzLAP1 was transferred to the pGWB5 destination vector and fused at the C-terminal with green fluorescent protein (GFP)  via Gateway® LR Clonase® II (Invitrogen, Carlsbad, CA, USA). The pGWB5-DzLAP1 product was transformed into A. tumefaciens GV3101 by electroporation.
Agrobacterium tumefaciens bearing the DzLAP1-GFP construct and A. tumefaciens with the silencing suppressor p19 gene  were co-infiltrated into 4-wk plants as previously described, with some modifications. Briefly, cells from each culture were washed and suspended in MM buffer (10 mM MES buffer + 10 mM MgCl2; pH 5.6). Cell suspensions harbouring DzLAP1 and p19 were adjusted at A600 to OD = 0.8 and 0.6, respectively, and combined in a 1:1 ratio. Acetosyringone was added to the mixture to a final concentration of 200 µM and the suspension was maintained in the dark at room temperature for 2 h before infiltration. At day 3 after infiltration, autofluorescence was visualised under a FluoView® FV10i-DOC confocal laser scanning microscope (Olympus Corp., Tokyo, Japan). GFP, chloroplast autofluorescence, and phase-contrast detection excitation/emission were recorded at 473/510 nm, 559/600 nm, and 559/600 nm, respectively.
All data were analysed with SPSS Statistics® v. 22.0 (IBM Corp., Armonk, NY, USA). One-way ANOVA identified significant differences among mean enzyme activity levels in the absence and presence of metal ions. Cys-Gly dipeptidase gene expression levels and enzyme activity at various ripening stages and in response to different ripening regulators were analysed by Tukey’s HSD multiple comparisons test (p < 0.05). Student’s t-test (p < 0.05) identified significant differences between N. benthamiana leaves overexpressing DzLAP1 and the control in terms of in planta Cys-Gly dipeptidase activity.