Identification of Jasmonic Acid Biosynthetic Genes in Sweet Cherry and Expression Analysis in Four Ancient Varieties from Tuscany

Sweet cherries are non-climacteric fruits whose early development is characterized by high levels of the phytohormone jasmonic acid (JA). Important parameters, such as firmness and susceptibility to cracking, can be affected by pre- and postharvest treatments of sweet cherries with JA. Despite the impact of JA on sweet cherry development and fruit characteristics, there are no studies (to the best of our knowledge) identifying the genes involved in the JA biosynthetic pathway in this species. We herein identify the sweet cherry members of the lipoxygenase family (13-LOX); allene oxide synthase, allene oxide cyclase and 12-oxo-phytodienoic acid reductase 3, as well as genes encoding the transcriptional master regulator MYC2. We analyze their expression pattern in four non-commercial Tuscan varieties (‘Carlotta’, ‘Maggiola’, ‘Morellona’, ‘Crognola’) having different levels of bioactives (namely phenolics). The highest differences are found in two genes encoding 13-LOX in the variety ‘Maggiola’ and one MYC2 isoform in ‘Morellona’. No statistically-significant variations are instead present in the allene oxide synthase, allene oxide cyclase and 12-oxo-phytodienoic acid reductase 3. Our data pave the way to follow-up studies on the JA signaling pathway in these ancient varieties, for example in relation to development and post-harvest storage.


Introduction
Prunus avium L. is an economically relevant tree of the Rosaceae family producing stone fruits. Its fruits are appreciated worldwide for their aroma, taste and richness in bioactive molecules. Sweet cherries are indeed rich in polyphenols and pentacyclic triterpenes [1,2]. They have a low content of calories. They are cholesterol-free and therefore valuable from a nutraceutical point of view. The ripening of sweet cherries is accompanied by an early increase in endogenous jasmonic acid (JA) levels [3]. This phytohormone is known to regulate different aspects of plant growth and development [4], as well as the response to exogenous stresses [5,6].
Given the physiological relevance of JA, many studies in the literature have specifically looked at the response of plants to this phytohormone in relation to flower and seed development [7], fruit ripening [8,9], senescence [10] and secondary metabolite production [11][12][13]. Concerning fruits, the majority of papers have dealt with the postharvest application of JA and its derivatives, which improved shelf life by protecting against pathogen attack and by enhancing the antioxidant content [14].
Based on the phylogenetic relationship with the thale cress 13-LOX, we could assign a nomenclature for the identified sweet cherry lipoxygenases ( Table 1). The gene PavLOX2.3 corresponds to two paralogs; however, it should be noted that Pav_sc0001040.1_g230.1.br is much shorter and may represent an incomplete sequence; indeed the best hit in NCBI corresponds to XP_021822570.1, which is 914 amino acids. The presence of a chloroplast signaling peptide was confirmed for all of the 13-LOX from P. avium (scores berween 0.509-0.565). All the other accessions in the tree are sequences from P. avium (see Table 1). Bootstrap values comprised of between 0.8 and 1 are displayed (the bigger the circle, the higher the bootstrap value). The tree was rooted with Pseudomonas aeruginosa lipoxygenase PAO1 (accession number AAG04558.1). Based on the phylogenetic relationship with the thale cress 13-LOX, we could assign a nomenclature for the identified sweet cherry lipoxygenases ( Table 1). The gene PavLOX2.3 corresponds to two paralogs; however, it should be noted that Pav_sc0001040.1_g230.1.br is much shorter and may represent an incomplete sequence; indeed the best hit in NCBI corresponds to XP_021822570.1, which is 914 amino acids. The presence of a chloroplast signaling peptide was confirmed for all of the 13-LOX from P. avium (scores berween 0.509-0.565).
Additionally, we identified two AOCs, one AOS and one thale cress ortholog of OPR3 (Table 1). Interestingly, three orthologs of the master regulator MYC2 were retrieved from the genome of sweet cherry ( Table 1): The gene named MYC2.2 showed the highest percentage identity (obtained when using the blastx algorithm, and as input, the GDR nucleotide sequences indicated in Table 1) with Arabidopsis MYC2 (85.71%), while MYC2 and MYC2.3 had lower scores (37.88% and 52.51%, respectively).
The MYC2 phylogenetic tree shows that PavMYC2.3 branches together with V. vinifera MYC2 (Figure 2), while the other sweet cherry MYC2 proteins do not show obvious similarities with MYC2 from other plant species [23]. Additionally, we identified two AOCs, one AOS and one thale cress ortholog of OPR3 (Table 1). Interestingly, three orthologs of the master regulator MYC2 were retrieved from the genome of sweet cherry ( Table 1): The gene named MYC2.2 showed the highest percentage identity (obtained when using the blastx algorithm, and as input, the GDR nucleotide sequences indicated in Table 1) with Arabidopsis MYC2 (85.71%), while MYC2 and MYC2.3 had lower scores (37.88% and 52.51%, respectively).
The MYC2 phylogenetic tree shows that PavMYC2.3 branches together with V. vinifera MYC2 (Figure 2), while the other sweet cherry MYC2 proteins do not show obvious similarities with MYC2 from other plant species [23].

Figure 2.
Unrooted maximum likelihood phylogenetic tree (number of bootstraps in ultrafast mode: 1000) of different MYC2 protein's full length sequences from P. avium (boxed) and other species taken from [23]. Bootstrap values comprised of between 0.8 and 1 are displayed (the bigger the circle, the higher the bootstrap value).

Total and Targeted Quantification of Phenolics in the Four Ancient Varieties
We previously characterized six ancient Tuscan varieties of P. avium sampled in 2017 from a nutraceutical point of view and ranked them according to their content in antioxidants, polyphenols, flavonoids and anthocyanins [1]. We here quantify the same parameters in a subset of four varieties known to show differences in the content of bioactives and sampled in the year 2018. We indeed wished to confirm the stability in the trend of the measured parameters over a different year of Figure 2. Unrooted maximum likelihood phylogenetic tree (number of bootstraps in ultrafast mode: 1000) of different MYC2 protein's full length sequences from P. avium (boxed) and other species taken from [23]. Bootstrap values comprised of between 0.8 and 1 are displayed (the bigger the circle, the higher the bootstrap value).

Total and Targeted Quantification of Phenolics in the Four Ancient Varieties
We previously characterized six ancient Tuscan varieties of P. avium sampled in 2017 from a nutraceutical point of view and ranked them according to their content in antioxidants, polyphenols, flavonoids and anthocyanins [1]. We here quantify the same parameters in a subset of four varieties known to show differences in the content of bioactives and sampled in the year 2018. We indeed wished to confirm the stability in the trend of the measured parameters over a different year of harvest. The results are shown in Table 2: Concerning the total antioxidants, the variety 'Crognola' ranks first, followed by 'Morellona', while 'Maggiola' is the lowest. The same ranking is observed for the polyphenols and flavonoids, while anthocyanins are highest in 'Morellona'. It should be noted that this variety is red fleshed, differently from all the others. Some differences are, however, observed with respect to the previously published data on fruits sampled in 2017 [1], where 'Carlotta' was the lowest producer of antioxidants, polyphenols and anthocyanins. In 2017, the highest amount of anthocyanins was observed for 'Crognola,' despite the absence of a red flesh. The targeted quantification of phenolics on the cherries sampled in 2018 reflects the spectrophotometric analyses, with the only exception of p-coumaric acid, whose abundance is significantly higher in 'Maggiola' as compared to 'Carlotta' and even 'Morellona' (Table 3). Despite being fruits sampled in an experimental field, the standard devations are overall low: This is partly due to our choice of pooling several fruits per biological replicate and to the grinding of the whole tissues (exocarp, mesocarp and skin). Eventual differences in the content of bioactive molecules in the pericarp and skin are thus levelled across replicates. These results differ from the values reported in 2017, where the values were the lowest for 'Maggiola' [1]. The samples were harvested from trees grown in an experimental field and therefore exposed to the natural environment-the differences observed may have been due to the varying climatic conditions between 2017 and 2018. It will be interesting to carry out high-throughput studies involving proteomics and transcriptomics to understand the reasons for the recorded differences in the two years of harvest.

Gene Expression Analysis of JA Biosynthetic Genes in the Four Ancient Varieties
The qPCR analysis was performed on the same four varieties and on 8 targets whose expression was detectable in mature fruits, namely LOX6, AOC2, AOS, LOX3, LOX3.2, MYC2.2, MYC2.3 and OPR3. We reasoned that these four varieties would be the best samples to highlight an eventual correlation between bioactive content and the expression of genes related to JA biosynthesis. As can be seen in Figure 3, the four ancient varieties differ in the expression of LOX genes and MYC2.2. More specifically, the variety 'Maggiola' shows the highest expression of LOX3 and LOX3.2. LOXs code for non-haem iron-containing dioxygenases catalyzing the oxygenation of polyunsaturated fatty acids (PUFAs), such as α-linoleic or α-linolenic acid, and lead to the synthesis of molecules collectively known as oxylipins [24]. LOXs known as 13-LOXs catalyze the oxygenation of the C atom in position 13 in the PUFA α-linolenic acid which will then enter the JA biosynthetic pathway via the action of the enzymes AOS and AOC [25]. The higher expression of LOX3 and LOX3.2 in 'Maggiola,' coupled to the lack of any statistically significant differences in AOC2 and AOS expressions in the Tuscan fruits, suggest that the identified genes partake in other biosynthetic branches (at least in the developmental stage herein analyzed). For example, the two genes may be involved in the LOX branch linked with hydroperoxide lyases (HPLs) which leads to the formation of volatiles responsible for the aroma of fruits. Plant HPLs have been classified into those acting on 13-hydroperoxylinoleic acid and 13-hydroperoxy-α-linolenic acid and those cleaving 9-hydroperoxy isomers of linoleic and α-linolenic acids [26]. In the literature, it was indeed shown that, at maturity, the content of C6 aldehydes increased in Chinese varieties of sweet cherry [27]. developmental stage herein analyzed). For example, the two genes may be involved in the LOX branch linked with hydroperoxide lyases (HPLs) which leads to the formation of volatiles responsible for the aroma of fruits. Plant HPLs have been classified into those acting on 13-hydroperoxylinoleic acid and 13-hydroperoxy-α-linolenic acid and those cleaving 9-hydroperoxy isomers of linoleic and α-linolenic acids [26]. In the literature, it was indeed shown that, at maturity, the content of C6 aldehydes increased in Chinese varieties of sweet cherry [27].
LOX6 shows a different pattern (Figure 3): Even if slight, the higher expression in 'Crognola' and 'Morellona' is statistically significant and follows the same trend observed for the antioxidants, polyphenols, flavonoids and anthocyanins ( Table 2). The Pearson correlation analysis showed a strong positive correlation between LOX6 and total antioxidanty, polyphenols, flavonoids, and anthocyanins, as well as chlorogenic acid, catechin and cyanidin-3-glucoside (Supplementary File 1). Although JA is known to activate the late genes in the anthocyanin biosynthetic pathway via COI1 (Coronatine-Insensitive 1) in thale cress [28], a previous study comparing apple and sweet cherry fruits highlighted, in the latter, a lack of correlation between endogenous JA/MeJA content and anthocyanin accumulation at maturity [3]. Further experimental proofs are needed to understand the role of P. avium LOX6. We also measured the expression of two genes encoding members of the master regulator MYC2: MYC2.2 and MYC2.3, which show different trends, with the former expressed at the highest levels in 'Morellona' and the latter displaying no significant variations (Figure 3). MYC2 is known to be involved in flavonoid biosynthesis [29,30] and 'Morellona' is one of the highest producers; the gene is, however, expressed at lower levels in 'Crognola', despite its high levels of bioactives (Tables 2 and LOX6 shows a different pattern ( Figure 3): Even if slight, the higher expression in 'Crognola' and 'Morellona' is statistically significant and follows the same trend observed for the antioxidants, polyphenols, flavonoids and anthocyanins ( Table 2). The Pearson correlation analysis showed a strong positive correlation between LOX6 and total antioxidanty, polyphenols, flavonoids, and anthocyanins, as well as chlorogenic acid, catechin and cyanidin-3-glucoside (Supplementary File 1). Although JA is known to activate the late genes in the anthocyanin biosynthetic pathway via COI1 (Coronatine-Insensitive 1) in thale cress [28], a previous study comparing apple and sweet cherry fruits highlighted, in the latter, a lack of correlation between endogenous JA/MeJA content and anthocyanin accumulation at maturity [3]. Further experimental proofs are needed to understand the role of P. avium LOX6.
We also measured the expression of two genes encoding members of the master regulator MYC2: MYC2.2 and MYC2.3, which show different trends, with the former expressed at the highest levels in 'Morellona' and the latter displaying no significant variations (Figure 3). MYC2 is known to be involved in flavonoid biosynthesis [29,30] and 'Morellona' is one of the highest producers; the gene is, however, expressed at lower levels in 'Crognola', despite its high levels of bioactives (Tables 2 and 3). RNA was extracted from samples comprising both the skin and mesocarp. The variety 'Crognola' is characterized by fruits displaying an intense red-colored skin but a yellow pulp: This may be due to a lower expression of MYC2.2 with respect to 'Morellona'.

Fruit Harvesting
The 18-year-old sweet cherry trees were grown in an experimental field of the National Research Council (CNR) in Follonica (Tuscany, Italy). The coordinates of the experimental field were previously reported [31]. The fruits were harvested by picking them all over the tree canopy at an average height of 1.70-1.90 m from the ground. Sampling took place in 2018 (sampling day: 18 May 2018, sampling time: Between 9:00 and 10:00 a.m., daily maximum temperature: 24 • C, daily minimum temperature: 8 • C, humidity: 74%). The sampled fruits were immediately placed in liquid nitrogen, then brought to the laboratory for long term storage at −80 • C.

Chemical Assays, HPLC Analyses and Statistics
For the determination of the total antioxidants, polyphenols, flavonoids and anthocyanins, we followed the methods previously described [1]. The analyses were performed in 3 technical replicates and 3 biological replicates (each biological replicate consisted of a pool of 3 sweet cherry fruits). The quantification of targeted secondary metabolites was performed on an HPLC-DAD system (PerkinElmer Series 200 HPLC Systems, Diode Array Detector, Norwalk, CT, USA) and by using curves generated with standard molecules, as previously reported [1]. The values obtained by averaging 3 independent technical replicates were log-transformed, and a one-way ANOVA with a Tukey's post-hoc test was performed with IBM SPSS Statistics v19 (IBM SPSS, Chicago, IL, USA) to determine the statistically significant differences among groups.

RNA Extraction, Primer Design and Real-Time PCR Data Analysis
The RNA extraction method, together with quantification, assessment of integrities, reverse transcription and real-time PCR conditions were as previously reported [31]. The primers used for the reference genes have been previously published [2,31]; those for the JA biosynthetic genes are reported in Table 4. They were designed with Primer3Plus (http://www.bioinformatics.nl/cgibin/primer3plus/primer3plus.cgi, last accessed on: 20 May 2019) and the presence of self-dimers, hetero-dimers, hairpins was checked with OligoAnalyzer 3.1 (http://eu.idtdna.com/calc/analyzer, last accessed on: 20 May 2019). The gene expression values were determined with qBase PLUS (version 3.2, Biogazelle, Ghent, Belgium) by using actin7 and PP2A as reference genes. Statistics were performed with IBM SPSS Statistics v19. A one-way ANOVA with a Tukey's post-hoc test was carried out after having transformed the NRQs (Normalized Relative Quantities) in log2 values and after having confirmed the normal distribution with a quantile-quantile plot (Q-Q plot). Four independent biological replicates, each consisting of a pool of 3 fruits were used for the gene expression study.

Bioinformatics
Sweet cherry LOX, MYC, AOS, AOC, OPD3 were obtained by blasting the thale cress protein sequences in NCBI and the GDR (Genome Data for Rosaceae) database (https://www.rosaceae.org/ blast/nucleotide/protein, last accessed on: 20 May 2019). The phylogenetic analysis was carried out by using the full-length LOX and MYC protein sequences identified in sweet cherry and thale cress, as well as other plant species. The pair-wise multiple alignment of LOX and MYC2 proteins from P. avium, Arabidopsis thaliana and the other plant species indicated in Figures 1 and 2 was performed with CLUSTAL-Ω (http://www.ebi.ac.uk/Tools/msa/clustalo/, last accessed on: 20 May 2019) [32], and the alignment was then used for the maximum likelihood phylogenetic tree using the online program W-IQ-TREE [33] (bootstraps in ultrafast mode: 1000; in auto mode, without selecting FreeRate Heterogeneity), available at http://iqtree.cibiv.univie.ac.at. The tree was visualized with iTOL (available at https://itol.embl.de/, last accessed on: 20 May 2019).

Conclusions
We have herein identified the JA biosynthetic genes in the economically relevant fruit tree P. avium, and measured their expression in four ancient varieties from Tuscany characterized by different levels of bioactives. The data suggest that LOX3.2 in 'Maggiola' fruits at maturity is not linked to JA production, but to other biochemical branches; e.g., volatile production. MYC2.2 is highest in 'Morellona,' a variety ranking first in terms of anthocyanins content. With this Communication, we wish to inspire future studies focusing on kinetics and postharvest treatment with JA (or its derivatives), to evaluate the relationship between bioactive contents and JA production, as well as to assess the stability of the Tuscan fruits during postharvest storage.