Original Research published: 03 December 2019 doi: 10.3389/fpls.2019.01581 Expression QTL (eQTLs) Analyses Reveal Candidate Genes Associated With Fruit Flesh Softening Rate in Peach [Prunus persica (L.) Batsch] Tomás Carrasco-Valenzuela 1†, Claudia Muñoz-Espinoza 1†, Aníbal Riveros 1, Romina Pedreschi 2, Pere Arús 3, Reinaldo Campos-Vargas 1 and Claudio Meneses 1,4* 1Centro de Biotecnología Vegetal, Facultad de Ciencias de la Vida, Universidad Andres Bello, Santiago, Chile, 2 Escuela de Agronomía, Pontificia Universidad Católica de Valparaíso, Quillota, Chile, 3 IRTA, Centre de Recerca en Agrigenòmica (CSIC- Edited by: IRTA-UAB-UB), Barcelona, Spain, 4 FONDAP Center for Genome Regulation, Santiago, Chile Marco Maccaferri, University of Bologna, Italy Significant differences in softening rate have been reported between melting flesh in peach Reviewed by: Tetyana Zhebentyayeva, and nectarine varieties. This trait seems to be controlled by several genes. We aimed to Pennsylvania State University (PSU), identify candidate genes involved in fruit softening rate by integrating quantitative trait loci United States Pedro Martinez-Gomez, (QTL) and expression QTL (eQTL) analyses, comparing siblings with contrasting softening Spanish National Research Council, rates. We used a segregating population derived from nectarine cv. ‘Venus’ selfing, which Spain was phenotyped for softening rate during three seasons. Six siblings with high (HSR) Maurizio Cocucci, University of Milan, Italy and six with low softening rate (LSR) were sequenced using RNA-Seq. A group of 5,041 Mauro Commisso, differentially expressed genes was identified. Also, we found a QTL with a LOD (logarithm University of Verona, Italy of odds) score of 9.7 on LG4 in all analyzed seasons. Furthermore, we detected 1,062 *Correspondence: Claudio Meneses eQTLs, of which 133 were found co-localizing with the identified QTL. Gene Ontology claudio.meneses@unab.cl (GO) analysis showed ‘Response to auxin’ as one the main over-represented categories. † These authors have contributed Our findings suggest over-expression of auxin biosynthetic related genes in the HSR equally to this work group, which implies a higher expression and/or accumulation of auxin, thereby triggering fast softening. Conversely, the LSR phenotype might be explained by an altered auxin- Specialty section: homeostasis associated with low auxin levels. This work will contribute to unraveling the This article was submitted to Plant Breeding, genetic mechanisms responsible for the softening rate in peaches and nectarines and a section of the journal lead to the development of molecular markers. Frontiers in Plant Science Received: 01 April 2019 Keywords: expression quantitative trait loci, Prunus persica, softening rate, flesh firmness, RNA-seq Accepted: 12 November 2019 Published: 03 December 2019 Citation: INTRODUCTION Carrasco-Valenzuela T, Muñoz-Espinoza C, Riveros A, Peach [Prunus persica (L.) Batsch] is the second most predominant temperate tree fruit species in Pedreschi R, Arús P, worldwide production. In fact, peach production rose to 24.9 million tons of fruits in 2016, representing Campos-Vargas R and Meneses C a cultivated area of 1,639,925 ha throughout the world (FAOSTAT, 2018). Peach is a self-compatible (2019) Expression QTL (eQTLs) fruit species, diploid, and exhibits a relatively short juvenile period (2–3 years) (Arús et al., 2012). Analyses Reveal Candidate Peach belongs to the Rosaceae family (Bassi and Monet 2008; Shulaev et al., 2008), and has become Genes Associated With Fruit Flesh Softening Rate in Peach the most economically important crop in Prunus (Abbott et al., 2002), a genus that also includes [Prunus persica (L.) Batsch]. nectarine, plum, apricot, cherry, and almond (Arús et al., 2012). In addition, peach is one of the best Front. Plant Sci. 10:1581. genetically characterized species in the Rosaceae family (Ogundiwin et al., 2009) and it has been used doi: 10.3389/fpls.2019.01581 as a model for genetics and genomics studies of tree fruit species (Byrne, 1990; Abbott et al., 2002; Frontiers in Plant Science | www.frontiersin.org 1 December 2019 | Volume 10 | Article 1581 Carrasco-Valenzuela et al. Expression QTL Analyses in Peach Shulaev et al., 2008; Zhebentyayeva et al., 2008; Arús et al., 2012). as peaches, significantly affecting softening (Ramina et al., 2008; Also, peach has a small genome (ca. 224.6 Mb) (Verde et al., 2013). Eduardo et al., 2011; Serra et al., 2017). Since the fruit is destined to Furthermore, peach breeding programs require a relatively distant markets, then quality traits such as firmer texture and very short time to obtain crossings and marketable products relative slow softening rates are the main targets of breeding programs, to other tree fruit crops, which is an advantage when developing according to consumer preferences (Eduardo et al., 2011). new cultivars for dynamic markets (Infante et al., 2008). In fact, Considering fruit firmness and texture at the ripe stage, two breeding efforts have been focused on the selection of traits main types have been characterized, both at the genetic and associated with fruit appearance and textural attributes (Cantín physiological level, known as non-melting flesh (NMF) and melting et al., 2010), as well as quality fruit traits including flavor, aroma, flesh (MF). NMF shows a very slow softening without a significant texture, and nutritional attributes, such as fruit size, color, reduction in flesh firmness (Tatsuki et al., 2013), although red firmness, resistance to postharvest handling, and an extended coloration is usually absent at ripening. However, NMF cultivars shelf life (Cirilli et al., 2016). Thus, a better understanding of the are mainly used for canning purposes, based on fruit integrity genetic and physiological basis of fruit quality traits could pave after heat processing (Ghiani et al., 2011). Conversely, MF exhibits the way for a more efficient new cultivar development process a rapid softening rate after harvest, resulting in fruit with a short (Infante et al., 2008). shelf life (Tatsuki et al., 2013). Nonetheless, most fresh market Fruit ripening encompasses several coordinated, genetically peaches belong to the MF group, showing a rapid softening at programmed events (biochemical and physiological). During ripening (Infante et al., 2008; Ghiani et al., 2011). ripening, the fruit texture changes, resulting in a decrease in Regarding fruit softening differences observed between MF firmness accompanied by an increase of ethylene release and and NMF peach cultivars, a sustained increase in soluble pectin up-regulated expression of genes, often triggered by plant is associated with the MF phenotype, which is associated with hormones (Tonutti et al., 1991; Trainotti et al., 2007; Zhang et al., progressive pectin depolymerization. In contrast, in the NMF 2017). Also, changes in the composition and ratio of fruit inclusion, cultivars, fruits maintain firmness at ripening as a result of the fruit color, sugar accumulation, aromatic compound production, scarce solubilization or depolymerization of pectin (Brummell and fruit softening are induced (Crisosto, 1994; Gapper et al., et al., 2004; Ghiani et al., 2011). 2013; Zhang et al., 2017). Modifications and remodeling of the cell It has been proposed that this is a result of the cell wall wall polysaccharide composition occur, triggering fruit softening remodeling enzymes [e.g., endo-polygalacturonase (endoPG)] (Hayama et al., 2006; Kumar et al., 2014). during ripening (Pressey and Avants, 1978; Morgutti et al., 2006). Peach is a climacteric fruit. The peach fruit respiration rate Evidence suggests that this trait is under the control of a major undergoes a measurable peak before the onset of the ripening gene (M/m) localized on linkage group 4 (LG4) (Peace et al., process, as well as an ethylene production peak (Trainotti et al., 2005). In fact, in the position of the M locus, there are two genes 2007). As a climacteric fruit, peach ripening is mainly controlled coding for endoPG, one of which (Prupe.4G261900) has been by ethylene (Liu et al., 2015). Concomitant with the ethylene proposed as a candidate for the determination of this trait (Lester production peak, auxin increases have also been detected in et al., 1996; Peace et al., 2005; Morgutti et al., 2006). climacteric fruits (Trainotti et al., 2007; Liu et al., 2015). Also, it Also, a high association between the pit/flesh adherence has been suggested that auxin has a role in ethylene biosynthesis trait and flesh texture in peach – resulting in the combination and signaling genes expression regulation in peach and tomato of freestone (F) melting (M) peaches or clingstone (f) NMF (m) (Trainotti et al., 2007; Pirrello et al., 2012; Liu et al., 2015). In fact, fruits – has been reported (Infante et al., 2008). Two other types several reports have suggested a possible relationship between have been recently reported, denominated ‘stony hard’ (Tatsuki fruit softening and auxin at ripening (Miller et al., 1987; Agusti et al., 2006; Pan et al., 2015) and ‘slow ripening’ (Brecht et al., et al., 1999; Ohmiya, 2000; Tatsuki et al., 2013). 1982; Nuñez-Lillo et al., 2015). In the former, ethylene production Considering that the sensory and nutritional characteristics and the subsequent softening in mature fruits are absent, possibly defining fruit quality traits are determined at the ripening stage, a result of a mutation in the ethylene biosynthesis pathway (Haji the identification of genetic and molecular factors acting as key et al., 2001; Infante et al., 2008). In the latter, a mutation prevents drivers of ripening will allow improvements to be made to the the normal ripening process, which seems to be under control of overall fruit quality (Carrari and Fernie, 2006; García-Gómez a single gene (Sr/sr). Individuals exhibiting this phenotype are et al., 2019), as well as to preserve fruit quality attributes during often discarded from breeding programs (Meneses et al., 2017). the postharvest shelf life (Liu et al., 2015). Differences in the softening rate occur have been detected Flesh firmness is an important indicator of fruit quality and among MF varieties. One of these, described as slow-melting freshness for peach consumers (Infante et al., 2008). Hence, flesh (SMF) trait, is characterized by a significant decrease in softening constitutes a key aspect determining peach fruit shelf the postharvest softening rate, providing a longer shelf life and life. As a climacteric fruit, peach softening occurs in two stages: a delayed harvest-time, ultimately resulting in an improved fruit first, fruit firmness decreases continuously and slowly; in the quality (Serra et al., 2017). Hence, SMF varieties have become second stage, known as the ‘melting’ stage, the fruit firmness standard in the peach industry due to their higher fruit quality rapidly decreases (Lester et al., 1994; Ghiani et al., 2011). and better handling after harvest (Laurens et al., 2018). However, Fruits are harvested at physiological maturity, and ripening the underlying genetic mechanisms controlling characters related happens off the tree. Thus, responsiveness to postharvest to postharvest shelf life, including softening rate (SOR) and SMF, conditions represents a quality trait for climacteric fruits, such have only been partially elucidated (Serra et al., 2017). Frontiers in Plant Science | www.frontiersin.org 2 December 2019 | Volume 10 | Article 1581 Carrasco-Valenzuela et al. Expression QTL Analyses in Peach Several studies have focused on peach in efforts to identify (g), and firmness (Newtons, N). The total of siblings phenotypically associations between major genes, QTLs, and candidate genes evaluated per year varied from 104 to 113 plants, depending on for fruit quality traits (Etienne et al., 2002; Infante et al., 2008; plant vigor and fruit availability (Supplementary Table 1). Eduardo et al., 2011; Fresnedo-Ramírez et al., 2015), as well as in Peach mature fruits were harvest at the field in the morning other Prunus species, such as apricot (Salazar et al., 2014; García- and later transported to the laboratory. Fruits exhibiting optimum Gómez et al., 2019) and Japanese plum (Salazar et al., 2017); for a commercial maturity, considering fruit firmness values reached review, see Aranzana et al. (2019). 50.0 ± 2.0 Newtons and chlorophyll absorbance at harvest (IAD) The advent of Expression Quantitative Trait Loci (eQTLs) values between 0.8–1.5 (DA-Meter FR Turoni, Forlí, Italia) (Lurie analysis has strengthened ongoing efforts focused on improving et al., 2013), uniform size, and without defectives were selected for our understanding of how genetic variants affect gene expression phenotypic evaluations. Evaluations were performed using nine (Rockman and Kruglyak, 2006; Cookson et al., 2009). Hence, eQTL fruits per sibling, and fruit background color was considered as the analysis is based on the use of gene expressions as quantitative harvest index parameter (Nilo et al., 2012) (Supplementary Table 1). molecular phenotypes to identify genetic variants significantly Fruit firmness was evaluated on the two opposite cheeks of associated with gene expression modifications (Stranger et al., 2007). the fruit, after removing the fruit skin, using a fruit pressure Recently, high-throughput Next Generation Sequencing tester with an 8 mm cylindrical plunger (Effigi, Alfonsine, Italia). technologies (NGS), including microarrays and RNA-Seq, have Measurements were carried out at harvest and harvest plus 3 been applied to study genetic variation in gene expression, days at 20°C, under conventional atmosphere, and 12 hours of representing an important source for eQTL analysis. An advantage light and 8 h of dark, denominated as shelf life stage according of this methodology is its suitability to studying the genetic with Serra et al. (2017) and Giné-Bordonaba et al. (2016). Also, architecture of complex traits (Ye et al., 2014; Galpaz et al., 2018). the softening rate was estimated using five fruits per sibling, as Therefore, in order to determine candidate genes related the ratio between observed firmness at harvest and at shelf life, to targeted traits, transcript abundances had been used as according to the following equation: quantitative trait data in the QTL analysis (Druka et al., 2010; Galpaz et al., 2018; García-Gómez et al., 2019). Multiple authors Softening rate = 100 × [1 − (Fm SL )/ Fm HV] (1) have noted the usefulness of gene expression analysis based on qPCR for the experimental validation of QTLs in fruit crops Where Fm SL corresponds to the flesh firmness measurement (Huang et al., 2014; Sugiyama et al., 2014; García-Gómez et al., observed at shelf life, while Fm HV represents the flesh firmness 2019), confirming the correlation between genotype and gene at harvest, and both are expressed as Newton (N). expression levels (Conesa et al., 2016). The main aim of this work was to elucidate genetic factors Experimental Design and Sample associated with softening rate (SOR) during peach ripening. Collection Second, using transcriptomic analyses siblings with contrasting A group of 12 siblings of V×V with contrasting phenotypes softening rate phenotypes, we aim to identify QTLs and eQTLs. for softening rate, i.e., higher and lower softening rate, named Third, we aim to integrate findings from both approaches to as LSR-1 to LSR6 and HSR-1 to HSR6 were analyzed using select candidate genes involved in fruit softening rate during ANOVA and the post-hoc Tukey test (Supplementary Table 2). peach maturation and to validate these by qPCR. Subsequently, these groups of individuals were selected to develop an RNA-Seq experiment. Fruit samples were collected in MATERIALS AND METHODS the 2015 season at harvest and maintained at 20°C for 3 days to accomplish shelf life. Later, samples were frozen in liquid N2 and Plant Material stored at −80 °C until RNA extraction. The mapping population (‘Venus’ × ‘Venus’; V×V; N = 138) was obtained from self-pollination of Prunus persica cv. Venus, a freestone RNA Isolation, Library Construction, and nectarine with yellow melting flesh. This population is located at Rayentué Experimental Station of the Instituto de Investigaciones cDNA Sequencing Total RNA was extracted using a pool of five fruit per sibling, Agropecuarias (INIA), VI Region, Chile (34°24’S and 70°50’W). according to Gudenschwager et al. (2012). The total RNA The mapping population V×V consists of 9-year-old trees in an concentration was measured using a Qubit® 2.0 Fluorometer experimental orchard, which were grafted over cv. ‘Nemaguard’, and (Invitrogen, Paisley, UK), following the manufacturer’s planted at 1 × 3 m spacing. This population segregates for peach instruction. Sample integrity was evaluated using a Fragment fruit quality traits, including mealiness, solid soluble content, titrable Analyzer™ Automated CE System. acidity, and maturity date, among others (Nuñez-Lillo et al., 2015). To accomplish the library’s construction, 10 µg of total RNA Siblings were managed under standard conditions for watering, was used from each pool, and mRNA isolation was made using fertilization, pest and diseases control, and pruning. poly(A) beads. Subsequently, libraries were constructed using a TruSeq Stranded Total RNA Library Prep Kit (Illumina) and Fruit Phenotyping sequenced using a Hi-Seq 2000 Illumina platform (Macrogen, Individuals were sampled to evaluate fruit quality traits, such as Korea). Fruit pooled samples from 12 siblings were used as soluble solids content (°Brix), titratable acidity (%), fruit weight biological replicates for each group (six for HRS and six for Frontiers in Plant Science | www.frontiersin.org 3 December 2019 | Volume 10 | Article 1581 Carrasco-Valenzuela et al. Expression QTL Analyses in Peach LSR). Also, two technical replicates (two independent lanes of Gene Ontology Analysis HiSEq2000) were used per sample. The RNA-Seq data used in A gene ontology (GO) enrichment analysis was performed this study are available at the NCBI’s Sequence Read Achieve considering all differentially expressed genes, as well as DE genes (available at: http://www.ncbi.nlm.nih.gov/sra) with SRA that co-localized with eQTLs. The frequency of query genes was accession number SRP186384. compared with the complete reference genome for Prunus persica version 2.0 (http://www.rosaceae.org/), searching for possible enrichment in biological processes. Analyses were performed Sequencing Data Analysis using the agriGO tool (available at: http://bioinfo.cau.edu.cn/ Paired-end reads were analyzed using FastQC to evaluate read agriGO), with the singular enrichment analysis and complete GO quality, and we trimmed the raw data reads by Trim Galore options. Significant GO terms (p < 0.05) were calculated using Cutadapt (Martin, 2011), using -paired and -q 25 options. the hyper geometric distribution, and the Yekutieli multi-test Subsequently, trimmed reads were aligned to the reference adjustment method (Du et al., 2010). Subsequently, the main GO genome for peach version 2.0 [Verde et al., 2013; Genome enrichment results were graphically represented using libraries Database for Rosaceae (GDR)] using software STAR (Dobin ‘Clusterprofile’ (Yu et al., 2012) and ‘ggplot2’ (Wickham, 2016) of et al., 2013), and later annotated. SAMtools was used to sort.bam the statistical software R. files using the option -n (Li et al., 2009). Counts per read were calculated with HTseq-count software (Anders et al., 2015), later normalized as counts per million (CPM) (Glusman et al., 2013). Gene Expression Analysis by Quantitative Real-Time PCR Expression Analysis QTL and eQTL Analysis Quantitative real-time PCR expression analysis (qPCR) of six The QTL analysis was performed using the available linkage randomly selected DE genes was performed to validate the map of V×V (Nuñez-Lillo et al., 2015), the softening rate data observed in silico expression profiles. Hence, three DE genes determined during three consecutive seasons of evaluation derived from the group of 169 DE genes up-regulated in LSR (2014, 2015 and 2016) and the library ‘qtl’ of statistical software siblings (Prupe.1G332600, Prupe.3G255800 and Prupe.8G079500) R (Broman et al., 2003). Also, chlorophyll absorbance (IAD and three derived from a group of 71 DE genes up-regulated value) was used as co-variable. To identify conventional QTLs, in HSR individuals (Prupe.1G460100, Prupe.5G130800 and two strategies were used: Interval Mapping test (IM) and Prupe.7G216300) were selected for validation of their expression Haley–Knott (HK) (Collard et al., 2005; Arends et al., 2010; profiling in LSR and HSR siblings at harvest. Eduardo et al., 2013). RNA extraction was performed using approximately 3 g of In the case of eQTL analysis, the mRNA transcript abundances the powder obtained from fruit cheeks, previously removing were estimate as the Log2 of ratio between counts per million the exocarp, and homogenized in liquid nitrogen according to observed in Low Softening Rate siblings than counts per million Gudenschwager et al. (2012). The total RNA concentration was observed in High Softening Rate siblings [Log2 (CPMLSR/ measured using a Qubit® 2.0 Fluorometer (Invitrogen, Paisley, CPMHSR)], and treated as quantitative traits that were mapped as UK), following the manufacturer’s instruction. Subsequently, gene-expression QTL (eQTL) (Galpaz et al., 2018). cDNAs were obtained by reverse transcription reactions with Also, the eQTLs analysis was performed using the library 1 µg of total RNA as template, using RevertAid First Strand ‘qtl’ of statistical software R, which provides computationally cDNA (Thermos) following the manufacturer instructions intensive algorithms based on linear regression, and suitability and oligo dT primers according to standard procedures. for experimental crosses (Broman et al., 2003). Therefore, qPCRs were carried out using the Eco Real-Time PCR System conventional QTLs and eQTLs were analyzed by interval mapping (Illumina Inc.) equipment. The qPCR amplification reactions test, and candidate QTLs or eQTLs, respectively, were selected were performed in a total volume of 10 μl containing 1 μl cDNA according to LOD (logarithm of odds) score cut-off of 3 (Collard (50 ng/μl), 5 μl primer mix (10 μM), 5 μl Eva Green PCR Master et al., 2005; Van Ooijen, 2006; Arends et al., 2010; Eduardo et al., Mix (2×) (Applied Biosystems), and 3 μl nuclease-free water. 2013). Hence, eQTLs co-localizing with conventional QTLs were Reactions were performed using three technical replicates, and selected as candidate genes for softening rate. a total of 18 observations were represented per point for LSR and HSR siblings. The thermal cycling conditions were denaturation at 95°C Differential Expression Analysis for 10 min, followed by 40 cycles of template denaturation at To identify differentially expressed genes (DE), libraries derived 95°C for 15 s, primer annealing at 60°C for 15 s and extension from low softening rate siblings (LSR-1 to LSR-6) were compared at 72°C for 15 s, followed by a single cycle at 95°C for 15 s, at with libraries from high softening rate HSR siblings (HSR-1 55°C for 15 s, and finally, at 72°C for 15 s. The amplification to HSR-6) at the phenological stage of harvest. Individuals efficiency was calculated using LingRegPCR software. The exhibiting the same phenotype for softening rate were considered relative expression values were estimated according to the as biological replicates in the analysis. Differential expression Pfaffl equation (Pfaffl, 2001). Values were normalized based analysis was done using ‘edgeR’ and statistical software R on the housekeeping gene Translation elongation factor 2 (Robinson et al., 2010). Also, results were visualized using (TEF2), which has been selected as a reliable reference for ggplots2 and ‘heatmap.2’ function of statistical package R. peach (Tong et al., 2009). Frontiers in Plant Science | www.frontiersin.org 4 December 2019 | Volume 10 | Article 1581 Carrasco-Valenzuela et al. Expression QTL Analyses in Peach Primer Design QTL Analysis Specific primers for gene validation were designed using PRIMER Softening rate data from the V×V population was used to perform 3 software, according to parameters described by Taylor et al. QTL analysis to identify chromosomic regions associated with (2010), and checked in silico using the Operon software (available softening rate. One QTL (qP-SOR4) with LOD = 8.9–9.6 (Table 1) at: http://www.operon.com/tools/oligo-analysis-tool.aspx). was consistently found on LG4 with a maximum coinciding Primers were synthesized by Integrated DNA Technologies Inc. with markers SNP_IGA41265 (genome position 10,649,641 bp (Coralville, Iowa). in chromosome 4 of the peach v2.0a1) and SNP_IGA_410398 (10.696.489 bp), located at genetic distances of 34.4 and 34.7 cM from the origin of the chromosome in V×V, respectively. A RESULTS confidence interval of 27.1 to 45.2 cM was calculated using a LOD score cut-off of 5 (Nuñez-Lillo et al., 2015). qP-SOR4 showed a Phenotypic Evaluation high effect (Figure 2), explaining approximately 58% of the total The V×V mapping population (N = 138) was phenotyped for phenotypic variance for softening rate (Table 1). No additional fruit quality at harvest during three consecutive seasons (from significant QTLs were identified for this trait in any of the three 2014 to 2016). Physiological parameters, such as titratable evaluated seasons. The position of this QTL interval encompasses acidity, chlorophyll absorbance, fruit weight, soluble solids, that of the Sr gene, also mapped in this population, and that of firmness, and softening rate, were evaluated (Supplementary the NAC transcription factor gene (PrupeG4.186800) located at Table 1). 11,116,841–11,118,655 bp of the same chromosome. Results showed that the titratable acidity observed in the three seasons varied from 0.1 to 0.9, with an average of 0.5 (Supplementary Table 1). Also, chlorophyll absorbance (IAD) Transcriptomic Analysis and Gene varied from 0.8 to 1.5, with an average of 1.13 between the Differential Expression three seasons. Interestingly, in the case of both parameters, An RNA-Seq experiment based on 12 siblings from the V×V no significant differences between seasons were observed population with contrasting phenotypes for softening rate was (Supplementary Table 1). performed during 2015. Interestingly, a negative correlation between During the three analyzed seasons, fresh fruit weight (g) softening rate and harvest date was observed in the group of LSR and ranged from 69.5 to 253.4, with averages of 134.8, 148.9, and HSR siblings (Pearson correlation -0.86, p-value 0.0003) (Figure 1D). 175.5, respectively, in seasons 2014, 2015, and 2016. In fact, A total of 24 libraries consisting of 12 samples with two significant differences in average values were found between technical replicates were analyzed (Supplementary Figure 1; season 2016 versus seasons 2014 and 2015 (p-value < 0.05) Supplementary Table 3). A total of approximately 38 million (Supplementary Table 1). In the case of soluble solid content paired-end reads were obtained per library, with an average (SSC), values varied from 8.4 to 16.4 °Brix, with averages of length of 100 bp. Subsequently, 924 million paired-end reads 13.2, 11.6, and 11.3 °Brix, respectively, in seasons 2014, 2015 and were obtained, with a length of 100 bp and QC score over 37 2016. Significant differences in SSC average values were found (Supplementary Table 3). Thus, approximately 93.7% of total between season 2014 versus seasons 2015 and 2016 (p-value trimmed reads mapped to the reference genome of P. persica < 0.05) (Supplementary Table 1). Firmness evaluation varied (Supplementary Table 3). Also, 14,538 genes presented at least between 13.3 to 71.2 N, with averages of 46.7, 56.9, and 50.7 N, 1 cpm (counts by million) in HSR or LSR siblings, representing for the 2014, 2015, and 2016 seasons, respectively. Significant 54.1% of the total genes described for peach (26,874 genes; Verde differences in average values were detected when comparing et al., 2013). A principal components analysis was performed seasons 2014 and 2015 (p-value < 0.05) (Supplementary S1). considering the total of expressed genes in LSR and HSR libraries. The analysis of softening rate was performed using 12 siblings The results showed that component 1 (PC1) explained 50.5% with contrasting phenotypes: six with low softening rate (LSR) of the phenotypic variance, and two clusters were identified and six with high softening rate (HSR) values, considering the six according to the softening rate phenotype (Supplementary Figure siblings as biological replicates, and five fruits for each individual 1). In addition, a group of 5,041 differentially expressed (DE) genes were evaluated in the 2014, 2015, and 2016 seasons. Results between LSR and HSR individuals were identified (p-value < 0.05, are summarized in Supplementary Table 2. For LSR siblings, FDR < 0.05), corresponding to 3,721 up-regulated and 1,320 averages varied from 17.20 to 21.16, and for HSR siblings from down-regulated DE genes (Supplementary Table 4, Figure 3A). 78.91 to 82.98; no significant differences between individuals Subsequently, a group of 240 DE genes with log2FC < -2, and seasons were identified in both cases. However, significant log2FC > 2 was selected, corresponding to 169 up-regulated DE differences between LSR and HSR siblings were observed in the genes and 71 down-regulated DE genes. Hierarchical clustering three analyzed seasons (p-value < 0.05). was performed using gene expression normalized values, and In addition, results from the characterization of the V×V two main clusters were observed (Figure 3B). population (N = 138) for softening rate showed similar average Also, a functional enrichment analysis (Gene Ontology) was values in the three seasons, fluctuating between 41.06 and 42.24 developed to assess the main processes over-represented in each (Supplementary Table 2), and followed a normal distribution group of DE genes using the agriGO platform. Concomitantly, GO (Shapiro-Wilk normality test W = 0.988, p-value = 0.514) analysis of the group of 169 DE genes up-regulated in LSR siblings (Figures 1A–C). showed two over-represented categories ‘Response to auxin’ and Frontiers in Plant Science | www.frontiersin.org 5 December 2019 | Volume 10 | Article 1581 Carrasco-Valenzuela et al. Expression QTL Analyses in Peach FIGURE 1 | (A–C) Softening rate distribution during the analyzed seasons. The X-axis represents the softening rate percentage, while the Y-axis shows the number of siblings. The segmented line represents the median, and the dotted line represents the average. (A–C) represent the season 2014, 2015, and 2016, respectively. (D) Correlation analysis between softening and harvest date, the latter represented as days after September 1st (DASF), was performed using data from season 2014 based on the group of the 12 siblings with contrasting phenotypes for softening rate. High softening rate (HSR) and Low softening rate (LSR) siblings are represented in red and blue circles, respectively. ‘Photosynthesis’ related to the Biological Process category (Figure PpendoPGF (Prupe.4G261900) were found differentially expressed 4A and Supplementary Figure 2). Results are concordant with between LSR and HSR individuals, being over-expressed in HSR the identification of nine DE genes, which presented GO term (Supplementary Table 4). Also, genes associated with cell wall ‘GO:0009733, Response to auxin’, all of them up-regulated in LSR remodeling coding for cellulose synthase-like B4 (Prupe.3G214400), siblings; four of them coding for auxin-induced proteins X10A expansin A1 (Prupe.1G276700) and xyloglucan endotransglucosylase/ (Prupe.8G081800, Prupe.8G078700 and Prupe.8G079600) and hydrolase 30 (Prupe.5G027800) were also identified in the group of 10A5 (Prupe.8G081400), as well as five coding for indole-3-acetic genes co-localizing with qSOR4 (Supplementary Table 5). acid-induced protein ARG7 (Prupe.8G081100, Prupe.8G080800, Prupe.8G079500, Prupe.8G081300, and Prupe.8G079200). eQTLs Analysis However, in the case of the group of 71 DE genes up-regulated An eQTL analysis was performed, and 2,413 eQTLs exhibited in HSR individuals, no over-represented category was identified. LOD scores between 3.0 and 19.9, associated with 1,496 genes, and Furthermore, DE genes coding for YUCCA flavin mono- distributed through the eight peach chromosomes. Of these eQTLs, oxygenase (PpYUC11) (Prupe.6G157400, Prupe.6G157500) 409 showed LODs >5.0. Subsequently, to identify candidate genes and NAC transcription factor (Prupe.4G186800), all of them associated with softening rate in peach, a possible co-localization of associated to the auxin biosynthesis pathway (Serra et al., 2017), QTLs and eQTLs was analyzed by the comparison of LSR and HSR were over-expressed in HSR siblings (Supplementary Table 4). siblings. Hence, a group of 1,062 eQTLs exhibiting a LOD >5.0 were Also, cell wall remodeling enzyme genes coding for found co-localizing at LG4 (27.1–45.2 cM). In relation to the linear endopolygalacturonases, PpendoPGM (Prupe.4G262200), and regression analyses implemented into R/qtl software to identify Frontiers in Plant Science | www.frontiersin.org 6 December 2019 | Volume 10 | Article 1581 Carrasco-Valenzuela et al. Expression QTL Analyses in Peach Table 1 | QTL significantly associated with softening rate identified on LG4 during the evaluated seasons (2014, 2015 and 2016). Chromosome location, genetic position (cM), LOD score, and total phenotypical variance explained (%) are showed. Marker qP-SOR4 2014 2015 2016 position (cM) LOD Phenotypic LOD Phenotypic LOD Phenotypic score Variance score Variance score Variance explained (%) explained (%) explained (%) SNP_IGA_403152 27.1 4.5 4.5 4.2 SNP_IGA_403613 27.4 5.8 6.0 6.1 SNP_IGA_403741 28.2 5.8 6.0 6.1 SNP_IGA_405773 29.3 5.8 6.0 6.1 SNP_IGA_407364 30.4 5.8 6.0 6.1 SNP_IGA_408059 30.7 5.8 6.0 6.1 SNP_IGA_408505 32.6 6.3 6.8 7.0 38 42 34 SNP_IGA_408981 33.3 6.3 6.8 7.0 SNP_IGA_409379 33.6 6.3 6.8 7.0 SNP_IGA_410265 34.4 8.9 9.6 9.1 SNP_IGA_410398 34.7 8.9 9.6 9.1 Slow Ripening+ 36.5 8.6 9.3 8.8 SNP_IGA_437516 45.2 5.8 5.9 5.3 SNP_IGA_438108 45.6 5.9 6.0 5.4 : Slow ripening is a morphological marker on LG4 described in Nuñez-Lillo et al., 2015. + differentially expressed genes overlapping at qP-SOR4, several algorithms were evaluated obtaining the same results. Furthermore, 133 DE genes were found co-localizing with qP-SOR4 identified on LG4 (Supplementary Table 5) exhibiting a LOD score greater than 5 (Figure 5), of which 13 correspond to cis QTLs (10%) and 120 to trans QTLs (90%). Also, a functional enrichment analysis (Gene Ontology) was performed considering this subset of genes. Thus, ‘Response to auxin’ was the main over- represented category related to the Biological Process category (Figure 4B; Supplementary Figure 3), in agreement with the identification of DE genes coding for proteins related to perception or response to auxin, such as indole-3-acetic acid (IAA)-inducible genes (Aux/IAA, SAUR) (Figure 4B; Supplementary Figure 3). Additionally, we identified five genes coding for SAUR-like auxin- responsive protein family (Prupe.8G081400, Prupe.8G079200, Prupe.8G081300, Prupe.8G078700, and Prupe.8G081100), and an early auxin response gene, capable of being induced by auxin and ethylene (Li et al., 2015) (Supplementary Figure 4). qPCR Expression Analysis of a Group of FIGURE 2 | Chromosome position of qP-SOR4 at LG4. The linkage map for Differentially Expressed Genes From a peach reported by Nuñez-Lillo et al. (2015) is represented in the center of the Comparison of Siblings With Contrasting figure. Black lines correspond to markers through the eight linkage groups. Phenotypes for Softening Rate LOD-score values were plotted, and results observed in analyzed seasons Specific primers were designed to validate six randomly are shown as a heatmap (values 0 to 10). selected DE genes. The primers used in real-time experiments Frontiers in Plant Science | www.frontiersin.org 7 December 2019 | Volume 10 | Article 1581 Carrasco-Valenzuela et al. Expression QTL Analyses in Peach FIGURE 3 | (A) Volcano plot of the total of 5,041 differentially expressed (DE genes) identified in the comparison between LSR and HSR siblings. (B) Hierarchical clustering of the selected group of 240 DE genes (logFC > 2 or logFC < -2), corresponding to 169 DE genes up-regulated in HSR siblings and 71 DE genes up-regulated in LSR siblings. Both schemes are based on gene expression normalized values. Frontiers in Plant Science | www.frontiersin.org 8 December 2019 | Volume 10 | Article 1581 Carrasco-Valenzuela et al. Expression QTL Analyses in Peach FIGURE 4 | Dot plot of enriched Gene Ontology (GO) terms of (A) 169 differential expressed (DE) genes (logFC > 2), up-regulated in LSR siblings; and (B) 133 differentially expressed (DE) genes with an eQTL peak over 5 and also co-localize with qP-SOR4, identified at chromosome 4. Y-axis indicates the GO term and X-axis shows the count of genes per GO term. Color gradient indicates adjusted p-value (0.05), using the Benjamini-Hochberg method. GO term ‘Response to auxin’ (GO:0009733) is marked in a red square. (qRT-PCR) are summarized in Supplementary Table 6. (Prupe.3G255800), Indole-3-acetic acid-induced protein Subsequently, the identified transcriptional expression ARG7 (Prupe.8G079500), two genes coding for an Unknown profiles were validated by real-time qPCR experiments at Protein Function (Prupe.1G460100, Prupe.5G130800), and harvest (Figure 6). Inositol oxygenase 4 (Prupe.7G216300). The obtained results The selected DE genes encompassed RNA-dependent RNA agreed with the in silico expression profiles observed in LSR polymerase (Prupe.1G332600), Cyclin-like family protein and HSR siblings at harvest (Figure 6). Frontiers in Plant Science | www.frontiersin.org 9 December 2019 | Volume 10 | Article 1581 Carrasco-Valenzuela et al. Expression QTL Analyses in Peach FIGURE 5 | Expression QTLs (eQTL) distribution on Venus × Venus genetic linkage map. Lines represent eight chromosomes of peach. The co-localization of eQTLs (LOD score >5) on qP-SOR4 is highlighted in yellow. In fact, the expression profile of the DE gene coding for the slow melting flesh character (SMF). To provide additional indole-3-acetic acid-induced protein ARG7 (Prupe.8G079500), information on this character, we studied the genetic factors a possible auxin homeostasis-related gene (Figure 6), was associated with fruit softening rate (SOR), integrating data confirmed by qPCR at harvest. This evidence suggests that auxin derived from conventional QTLs and eQTLs in an F2 mapping homeostasis is altered in LSR individuals and might be associated population of P. persica. We observed a normal distribution for with low auxin levels, which might activate a compensatory SOR in this progeny in the three years studied, suggesting that it mechanism to accomplish softening. is under polygenic control (Figures 1A–C). Firmness loss values were analyzed in a subset of siblings with contrasting phenotypes for softening rate, denominated as LSR and DISCUSSION HSR, and results for SOR were consistent for this sample among seasons (Supplementary Table 2). These results were similar to Peach Fruit Softening those previously reported, where SOR values were used to estimate Peaches and nectarines are frequently maintained for 3–4 weeks the SMF trait, considered as a slower fruit softening after the maturity after harvest, which could significantly affect softening. For this stage (Serra et al., 2017). In fact, the SOR values for the SMF trait in reason, the responsiveness to postharvest conditions represents a key LSR siblings (17.2–21.2) were similar to those previously reported quality trait for breeding programs (Ramina et al., 2008). However, by Serra et al. (2017) in the SMF variety ‘Big Top’ and other siblings the genetic basis of postharvest life in peaches is hampered by the considered SMF (< 20.0). In fact, SMF fruit characterization also difficulties associated with its phenotyping (Bassi and Monet, 2008). presents some difficulties and evidence suggests that it is under Previous studies developed by Nuñez-Lillo et al. (2015) and oligogenic or polygenic control, with at least two QTLs involved, Serra et al. (2017), which were performed to understand fruit one on LG4 and another on LG5 (Serra et al., 2017). quality traits related to softening, mainly focused on the maturity Firmness loss values observed in MF varieties ‘Armking’ date or the identification of conventional QTLs associated with (62.5–75.3) and ‘Nectaross’ (64.2–65.0) followed a similar pattern Frontiers in Plant Science | www.frontiersin.org 10 December 2019 | Volume 10 | Article 1581 Carrasco-Valenzuela et al. Expression QTL Analyses in Peach FIGURE 6 | Validation of six differentially expressed (DE) genes randomly selected; identified in the comparison between LSR and HSR siblings; using real-time (qPCR) experiments at harvest. Genes are (A) RNA-dependent RNA polymerase (Prupe.1G332600); (B) Cyclin-like family protein (Prupe.3G255800); (C) Indole-3- acetic acid-induced protein ARG7 (Prupe.8G079500); (D) Unknown Protein Function (Prupe.1G460100); (E) Unknown Protein Function (Prupe.5G130800); and (F) Inositol oxygenase 4 (Prupe.7G216300). The TEF2 gene (possible Translation elongation factor 2 gene) was used as a reference gene, and gene expression was expressed as relative expression. The relative abundance based on qPCR (Y-left axis) was compared with the relative abundance of normalized gene expression determined by RNA-Seq experiment (Y-right axis); both values are respectively represented as bar and point, and the standard deviation was also plotted. as HSR siblings (78.9–82.98) (Serra et al., 2017). Furthermore, as fruit weight (Da Silva Linge et al., 2015; Hernandez Mora segregating populations derived from the cross of ‘Big Top’ and et al., 2017), sugar contents (Zeballos et al., 2016; Hernandez ‘Armking’ (Bt×Ak) or ‘Nectaross’ (Bt×Nr) exhibited average Mora et al., 2017), mesocarp thickness (Donoso et al., 2016), values of firmness loss of 43.62 and 40.25, similar to those and mealiness (Cantín et al., 2010; Nuñez-Lillo et al., 2015), as obtained in the V×V mapping population in analyzed seasons well as disease resistance genes, including brown rot (Martínez- (41.06–42.24). Overall, these data suggest that the populations García et al., 2013; Pacheco et al., 2014) and bacterial spot (Yang previously analyzed had a similar fruit softening behavior than et al., 2013). V×V and that firmness loss is a robust approach to measure fruit Also, QTLs related to phenology and fruit quality traits have SOR as a component of postharvest shelf life of peach fruits. been reported at LG4 in apricot, such as firmness and soluble solids (Salazar et al., 2014), as well as ripening time and skin color in Japanese plum (Salazar et al., 2017; García-Gómez et al., QTL Analysis 2019). While some of these traits might result from pleiotropic A single major QTL (qP-SOR4), explaining 34–42% of the effects associated with the same gene or genes (Eduardo et al., phenotypic variance for the trait, was found in the progeny 2011; Eduardo et al., 2015; Serra et al., 2017), the importance of studied. This QTL maps at a different position than the M locus, this genomic region is evident and deserves additional study. and co-locates with SOR-related traits such as the (Sr/sr) gene, Furthermore, one QTL for firmness in apple was identified using responsible for slow ripening (SMF), that segregates in the a phenotyping method able to discriminate between the acoustic V×V progeny (Eduardo et al., 2015; Nuñez-Lillo et al., 2015), and mechanical components of the texture. The QTL is located on and the (MD/md) gene that determines maturity date (Cantín chromosome 10 (Longhi et al., 2012). Considering the macrosynteny et al., 2010; Eduardo et al., 2011; Dirlewanger et al., 2012; Pirona in members of the Rosaceae family, the study of Malus domestica, et al., 2013; Donoso et al., 2016; Serra et al., 2017). Hence, a NAC Fragaria vesca, and P. persica showed a high level of synteny among transcription factor located within LG4 interval (encoded by chromosomes 4 of Prunus and 10 of Malus (Illa et al., 2011; Jung Prupe.4G186800) has been proposed as the causal gene for MD et al., 2012). This evidence agreed with our finding of a QTL and Sr (Pirona et al., 2013; Eduardo et al., 2015). Also, qP-SOR4 associated with softening rate on chromosome 4, and might support maps with a QTL previously reported for firmness loss and the hypothesis that fruit texture is determined by common genes in maturity date in peach cultivars ‘Armking’ and ‘Nectaross’, both different members of the Rosaceae family (Illa et al., 2011). MF types (Serra et al., 2017). The results obtained in this work could lead to the Interestingly, this region of LG4 seems to be a chromosomal development of molecular markers associated with quality hotspot where major genes or major QTLs explaining important traits, such as SOR, which could be integrated into marker- traits have been identified in peach and other Prunus crops. In assisted selection (MAS), based on their transferability along addition to traits related to SOR, such as maturity date and Prunus family, and also contribute to breeding efforts (Salazar et SMF, others related to fruit quality have been described, such al., 2017; García-Gómez et al., 2019). Frontiers in Plant Science | www.frontiersin.org 11 December 2019 | Volume 10 | Article 1581 Carrasco-Valenzuela et al. Expression QTL Analyses in Peach Candidate Genes Associated With SMF Trait This result could be supported by previous reports from It has been proposed that IAA concentration might control comparisons made during the fruit ripening stage of melting ethylene production (Tatsuki et al., 2013) and the fruit softening and stony hard (SH) peach varieties, in which the latter was process in peaches (Pan et al., 2015). It seems that auxin is characterized by the absence of ethylene production and involved in a complex interplay with ethylene during peach fruit softening. In fact, differential expression of several ripening (Trainotti et al., 2007; Liu et al., 2015). The expression auxin-homeostasis-related genes, as well as a differential of ethylene biosynthesis and signaling genes is regulated in fleshy IAA accumulation, were observed, associated to low IAA fruits (e.g., tomato and peach) by auxin, where their texture is concentration in the SH peach variety (Haji et al., 2003; Haji, greatly modified during ripening (Gillaspy et al., 1993; Jones 2014; Pan et al., 2015). et al., 2002; Trainotti et al., 2007; Pirrello et al., 2012; Liu et al., Additionally, the process of cell wall disassembly has been 2015). Hence, the concentration of auxin increases before considered responsible for softening and textural changes on ripening in peach fruit coinciding with the climacteric ethylene fruit during ripening (Brummell and Harpster, 2001). The widely peak (Tatsuki et al., 2013). reported involvement of polygalacturonase (PG), an important Concomitantly, auxin regulates the expression of Auxin hydrolytic enzyme involved in pectin degradation during fruit Response Factors (ARF) and Aux/IAA encoding genes, softening (Qian et al., 2016), likely plays a key role as a positive associated to the auxin signaling pathway, whose expression regulator mediated by ethylene (Hiwasa et al., 2003; Asif and increases in the peach mesocarp during ripening (Jones et al., Nath, 2005). 2002; Trainotti et al., 2007). Additionally, tomato fruit firmness Moreover, we found that cell wall remodeling enzyme is partially modulated by a transcription factor related to auxin- genes coding for endopolygalacturonases, PpendoPGM and responses (SlARF4), reinforcing its possible role in ripening PpendoPGF – previously reported as associated to the MF/ control in climacteric fruits (Liu et al., 2015). NMF and clingstone/freestone, respectively (Gu et al., 2016) – Moreover, auxin induces the expression of genes encoding were over-expressed in HSR individuals (Supplementary Table 1-aminocyclopropane-1-carboxylic acid (ACC) synthase, 4). Furthermore, based on the comparison of peach cultivars associated with the ethylene biosynthesis (Tatsuki et al., 2013). presenting fast (‘Qian Jian Bai’) and slow softening patterns Also, exogenous application of ethylene on peach fruit has (‘Qin Wang’), it has been proposed that candidate genes coding been shown to promote postharvest softening (Haji et al., 2001; for PGs might play major roles in peach fruit softening (Qian Hayama et al., 2006). et al., 2016). Our results show that DE genes coding for YUCCA flavin mono- Also, genes associated with cell wall remodeling coding oxygenase (PpYUC11) (Prupe.6G157400, Prupe.6G157500) for cellulose synthase-like B4 (Prupe.3G214400), expansin and NAC transcription factor (Prupe.4G186800) – all of them A1 (Prupe.1G276700), and xyloglucan endotransglucosylase/ associated to the auxin biosynthesis way (Mashiguchi et al., hydrolase 30 (Prupe.5G027800) were also identified in the group 2011; Serra et al., 2017) – were over-expressed in HSR siblings of genes co-localizing with qP-SOR4 at LG4 (Supplementary (Supplementary Table 4). In fact, the Prupe.4G186800 was Table 5). In fact, expansin and XTH are active drivers of cell proposed as a candidate for a maturity date gene (Pirona et al., wall remodeling, both of which are sensitive to auxin induction 2013) and for the slow ripening gene (Eduardo et al., 2015) that (Labavitch and Ray, 1974; Nishitani and Masuda, 1981). co-locates with the position of qP-SOR4 identified in this work. By an integrated analysis based on QTLs, eQTLs, and Additionally, the expression profile of the DE gene coding for transcriptomic profiling, we identified genes involved in the indole-3-acetic acid-induced protein ARG7 (Prupe.8G079500) metabolic pathways affecting fruit softening rate in peach. Some was experimentally confirmed by qPCR, and was also of these genes were selected as candidates possibly acting as key up-regulated in LSR siblings. Our findings were concordant drivers of this trait in peach cultivars. These results are of interest with previous reports of high levels of auxin-induced cell wall to efforts to better understand the process of fruit maturation in modifications promoting fast softening rate in peach fruits peach and to estimate the effects of the natural allelic variation (Tatsuki et al., 2013). for these genes. Also, stony hard (SH) peach cultivars, such as ‘Yumyeong’, This information can be used to design appropriate breeding ‘Xia cui’, and ‘CN16’, showed a deficient auxin biosynthesis, strategies to improve the postharvest performance of commercial probably resulting from low levels of PpYUC11 gene expression, peach cultivars, considering the high synteny described in which might reduce the level of IAA in the stony hard phenotype Prunus, which could facilitate a successfully transferability of (Pan et al., 2015). The flavin mono-oxygenase gene PpYUC11 has molecular markers suitable to be integrated in breeding selection also been proposed as a regulator of the concentration of IAA programs (Dirlewanger et al., 2004; Infante et al., 2008). at the late-ripening stage in peach (Pan et al., 2015). In fact, a higher expression level and stronger up-regulation of PpYUC11 during maturation to the ripening stage had been correlated with CONCLUSIONS IAA concentration as well as with the activation of the ethylene biosynthesis mediated by the PpACS1 gene; concomitantly, We have developed an integrative analysis involving a rapid fruit softening at the late-ripening stage in MF peach conventional QTLs, eQTLs, and transcriptome profiling cultivars is observed (Tatsuki et al., 2013; Pan et al., 2015). analysis focused on siblings with contrasting phenotypes Frontiers in Plant Science | www.frontiersin.org 12 December 2019 | Volume 10 | Article 1581 Carrasco-Valenzuela et al. Expression QTL Analyses in Peach for flesh softening in peach fruit. A conventional QTL was AUTHOR CONTRIBUTIONS identified at LG4 co-localizing with a previously reported QTL associated with fruit firmness and the slow melting flesh TC-V performed QTL analysis, eQTL analysis, and RNA-Seq character (Serra et al., 2017). analysis. CM-E contributed with RNA-Seq, GO, and KEGG The integration of differential gene expression, as well as analysis and writing the manuscript. AR carried out the gene conventional and expression QTL analysis, provided a wide validation by RT-qPCR and KEGG analysis. RC-V, RP, and PA characterization of possible genetic factors affecting the collaborated to writing-review and editing of the manuscript. regulation of softening rate character in peach. Our findings CM supervised the bioinformatic analysis and worked on suggest that genes associated with the auxin biosynthetic writing-original draft. pathway were up-regulated in HSR siblings, reinforcing the importance of auxin as a key driver of rapid fruit softening at the late-ripening stage in melting flesh peach cultivars. FUNDING Conversely, the LSR phenotype might be explained by an altered auxin-homeostasis associated with low auxin levels. This work was supported by Fondo Nacional de Desarrollo This work contributes to unraveling the genetic mechanisms Científico y Tecnológico (FONDECYT) 1160584, FONDEF responsible for the softening rate in peaches and nectarines. D13i10005 and CORFO Biofrutales 13CTI21520-SP04. Furthermore, findings derived from this study could lead to the development of molecular markers associated with softening rate. Considering the high synteny described in ACKNOWLEDGMENTS Prunus, this might facilitate their successful transference The authors would like to thank Centro de Biotecnología Vegetal and integration into MAS, which would be useful for at Universidad Andrés Bello. breeding purposes. SUPPLEMENTARY MATERIAL DATA AVAILABILITY STATEMENT The Supplementary Material for this article can be found online Publicly available datasets were analyzed in this study. This data can be at: https://www.frontiersin.org/articles/10.3389/fpls.2019.01581/ found here: https://www.ncbi.nlm.nih.gov/sra/?term=SRP186384. full#supplementary-material REFERENCES Broman, K., Wu, H., Sen, Ś., and Churchill, G. (2003). R/qtl: QTL mapping in experimental crosses. Bioinformatics 19, 889–890. doi: 10.1093/bioinformatics/btg112 Abbott, A. G., Georgi, L., Inigo, M., Sosinski, B., Yvergniaux, D., Wang, Y., et al. Brummell, D. A., and Harpster, M. H. (2001). Cell wall metabolism in fruit (2002). Peach: the model genome for Rosaceae. Acta Hort. 575, 145–155. doi: softening and quality and its manipulation in transgenic plants. Plant Mol. Biol. 10.17660/ActaHortic.2002.575.14 47, 311–340. doi: 10.1023/A:1010656104304 agriGO tool. Available at: http://bioinfo.cau.edu.cn/agriGO. Accessed 20 Brummell, D. A., Dal Cin, V., Crisosto, C. H., and Labavitch, J. M. (2004). Cell wall September. metabolism during maturation, ripening and senescence of peach fruit. J. Exp. Agusti, M., Almela, V., Andreu, I., Juan, M., and Zacarias, L. (1999). Bot. 55, 2029–2039. doi: 10.1093/jxb/erh227 Synthetic auxin 3,5,6-TPA promotes fruit development and climacteric Byrne, D. H. (1990). Isozyme variability in four diploid stone fruits compared with in Prunus persica L. Batsch. J. Hortic. Sci. Biotechnol. 74, 556–560. doi: other woody perennial plants. J. Hered. 81, 68–71. doi: 10.1093/oxfordjournals. 10.1080/14620316.1999.11511152 jhered.a110927 Anders, S., Pyl, P. T., and Huber, W. (2015). HTSeq- a Python framework to work Cantín, C., Crisosto, C., Ogundiwin, E., Gradziel, T., Torrents, J., Moreno, M., with high-throughput sequencing data. Bioinformatics 31 (2), 166–169. doi: et al. (2010). Chilling injury susceptibility in an intra-specific peach [Prunus 10.1093/bioinformatics/btu638 persica (L.) Batsch] progeny. Postharvest. Biol. Tec. 58, 79–87. doi: 10.1016/j. Arús, P., Verde, I., Sosinski, B., Zhebentyayeva, T., and Abbott, A. G. (2012). postharvbio.2010.06.002 The peach genome. Tree Genet. Genomes 8, 531–547. doi: 10.1007/ Carrari, F., and Fernie, A. R. (2006). Metabolic regulation underlying tomato fruit s11295-012-0493-8 development. J. Exp. Bot. 57, 1883–1897. doi: 10.1093/jxb/erj020 Aranzana, M. J., Decroocq, V., Dirlewanger, E., Eduardo, I., Gao, Z. S., Gasic, K., et al. Cirilli, M., Bassi, D., and Ciacciulli, A. (2016). Sugars in peach fruit: a breeding (2019). Prunus genetics and applications after de novo genome sequencing: perspective. Hortic. Res. 3, 15067. doi: 10.1038/hortres.2015.67 achievements and prospects. Hortic. Res. 6, 58. doi: 10.1038/s41438-019-0140-8 Collard, B., Jaufer, M., Brouwer, J., and Pang, E. (2005). An introduction to Arends, D., Prins, P., Jansen, R. C., and Broman, K. W. (2010). R/qtl: high- markers, quantitative trait loci (QTL) mapping and marker-assisted selection throughput multiple QTL mapping. Bioinformatics 26 (23), 2990–2992. doi: for crop improvement: The basic concepts. Euphytica 142, 169–196. doi: 10.1093/bioinformatics/btq565 10.1007/s10681-005-1681-5 Asif, M. H., and Nath, P. (2005). Expression of multiple forms of polygalacturonase Conesa, A., Madrigal, P., Tarazona, S., Gomez-Cabrero, D., Cervera, A., McPherson, A., gene during ripening in banana fruit. Plant Physiol. Biochem. 43, 177–184. doi: et al. (2016). A survey of best practices for RNAseq data analysis. Genome Biol. 10.1016/j.plaphy.2005.01.011 17, 13. doi: 10.1186/s13059-016-1047-4 Bassi, D., and Monet, R. (2008). “Classical genetics and breeding,” in The Peach: Cookson, W., Liang, L., Abecasis, G., Moffatt, M., and Lathrop, M. (2009). Botany, Production and Uses. Eds. D. R. Layne, and D. Bassi (Wallingford, UK: Mapping complex disease traits with global gene expression. Nat. Rev. Genet. CAB International), 61–84. doi: 10.1079/9781845933869.0001 10, 184– 194. doi: 10.1038/nrg2537 Brecht, J., Kader, A., and Ramming, D. (1982). Postharvest physiology of “non- Crisosto, C. H. (1994). Stone fruit maturity indices: a descriptive review. ripening” nectarine selections. Hortic. Sci. 17, 490–501. Postharvest. News Inform. 5, 65–68. Frontiers in Plant Science | www.frontiersin.org 13 December 2019 | Volume 10 | Article 1581 Carrasco-Valenzuela et al. Expression QTL Analyses in Peach Da Silva Linge, C., Bassi, D., Bianco, L., Pacheco, I., Pirona, R., and Rossini, L. acceptance of different Prunus persica cultivars. Postharvest. Biol. Tec. 115, (2015). Genetic dissection of fruit weight and size in an F2 peach (Prunus persica 38–47. doi: 10.1016/j.postharvbio.2015.12.006 (L.) Batsch) progeny. Mol. Breed. 35 (2), 71. doi: 10.1007/s11032-015-0271-z Glusman, G., Caballero, J., Robinson, M., Kutlu, B., and Hood, H. (2013). Optimal Dirlewanger, E., Graziano, E., Joobeur, T., Garriga-Caldré, F., Cosson, P., scaling of digital transcriptomes. PloS One 8 (11), e77885. doi: 10.1371/journal. Howad, W., et al. (2004). Comparative mapping and marker assisted selection pone.0077885 in Rosaceae fruit crops. Proc. Nat. Ac. Sci. U.S.A. 101, 9891–9896. doi: 10.1073/ Gu, C., Wang, L., Wang, W., Zhou, H., Ma, B. Q., Zheng, H. Y., et al. (2016). Copy pnas.0307937101 number variation of a gene cluster encoding endopolygalacturonase mediates Dirlewanger, E., Quero-Garcia, J., Le Dantec, L., Lambert, P., Ruiz, D., Dondini, L., flesh texture and stone adhesion in peach. J. Exp. Bot. 67, 1993–2005. doi: et al. (2012). Comparison of the genetic determinism of two key phenological 10.1093/jxb/erw021 traits, flowering and maturity dates, in three Prunus species: peach, apricot and Gudenschwager, O., González-Ag̈ero, M., and Defilippi, B. G. (2012). A general sweet cherry. Heredity 109, 280–292. doi: 10.1038/hdy.2012.38 method for high-quality RNA isolation from metabolite-rich fruits. S. Afr. J. Dobin, A., Davis, C. A., Schlesinger, F., Drenkow, J., Zaleski, C., Jha, S., et al. (2013). Bot. 83, 186–192. doi: 10.1016/j.sajb.2012.08.004 STAR: ultrafast universal RNA-seq aligner. Bioinformatics 29 (1), 15–21. doi: Haji, T., Yaegaki, H., and Yamaguchi, M. (2001). Changes in ethylene 10.1093/bioinformatics/bts635 production and flesh firmness of melting, nonmelting and stony hard Donoso, J. M., Picañol, R., Serra, O., Howad, W., Alegre, S., Arús, P., et al. (2016). peaches after harvest. J. Japanese Soc. Hortic. Sci. 70, 458–459. doi: 10.2503/ Exploring almond genetic variability useful for peach improvement: mapping jjshs.70.458 major genes and QTLs in two interspecific almond x peach populations. Mol. Haji, T., Yaegaki, H., and Yamaguchi, M. (2003). Softening of stony hard peach by Breed. 36, 16. doi: 10.1007/s11032-016-0441-7 ethylene and the induction of endogenous ethylene by 1-aminocyclopropane- Druka, A., Potokina, E., Luo, Z., Jiang, N., Chen, X., Kearsy, M., et al. (2010). 1-carboxylic acid (ACC). J. Japanese Soc. Hortic. Sci. 72, 212–217. doi: Expression quantitative trait loci analysis in plants. Plant Biotech. J. 8, 10–27. 10.2503/jjshs.72.212 doi: 10.1111/j.1467-7652.2009.00460.x Haji, T. (2014). Inheritance of flesh texture in peach and effects of ethylene Du, Z., Zhou, X., Ling, Y., Zhang, Z., and Su, Z. (2010). ). agriGO: a GO analysis treatment on softening of the stony hard peach. Japan Agric. Res. Q. 48, 57–61. toolkit for the agricultural community. Nucleic Acids Res. 38, W64–W70. doi: doi: 10.6090/jarq.48.57 10.1093/nar/gkq310 Hayama, H., Tatsuki, M., Ito, A., and Kashimura, Y. (2006). Ethylene and fruit Eduardo, I., Pacheco, I., Chietera, G., Bassi, S., Pozzi, C., Vecchietti, A., et al. softening in the stony hard mutation in peach. Postharvest. Biol. Technol. 41, (2011). QTL analysis of fruit quality traits in two peach intraspecific population 16–21. doi: 10.1016/j.postharvbio.2006.03.006 and importance of maturity date pleiotropic effect. Tree Genet. Genomes 7, Hernandez Mora, J. R., Micheletti, D., Bink, M., Van de Weg, E., Cantin, C., 323–335. doi: 10.1007/s11295-010-0334-6 Nazzicari, N., et al. (2017). Integrated QTL detection for key breeding traits Eduardo, I., Chietera, G., Pirona, R., Pacheco, I., Troggio, M., Banchi, E., et al. in multiple peach progenies. BMC Genomics 18 (1), 404–415. doi: 10.1186/ (2013). Genetic dissection of aroma volatile compounds from the essential oil s12864-017-3783-6 of peach fruit: QTL analysis and identification of candidate genes using dense Hiwasa, K., Kinugasa, Y., Amano, S., Hashimoto, A., Nakano, R., Inaba, A., et al. SNP maps. Tree Genet. Genomes 9, 189–204. doi: 10.1007/s11295-012-0546-z (2003). Ethylene is required for both the initiation and progression of softening Eduardo, I., Picañol, R., Rojas, E., Batlle, I., Howad, W., Aranzana, M. J., et al. in pear. J. Exp. Bot. 54, 771–779. doi: 10.1093/jxb/erg073 (2015). Mapping of a major gene for the slow ripening character in peach: Huang, Y. F., Vialet, S., Guiraud, J. L., Torregrosa, L., Bertrand, Y., Cheynier, V., co-location with the maturity date gene and development of a candidate gene- et al. (2014). A negative MYB regulator of proanthocyanidin accumulation, based diagnostic marker for its selection. Euphytica 205, 627–636. doi: 10.1007/ identified through expression quantitative locus mapping in the grape berry. s10681-015-1445-9 New Phytol. 201 (3), 795–809. doi: 10.1111/nph.12557 Etienne, C., Rothan, C., Moing, A., Plomion, C., Bodenes, C., Svanella-Dumas, L., Illa, E., Sargent, D. J., Lopez Girona, E., Bushakra, J., Cestaro, A., Crowhurst, R., et al. (2002). Candidate genes and QTLs for sugar and organic content in peach et al. (2011). Comparative analysis of rosaceous genomes and the reconstruction [Prunus persica (L.) Batsch]. Theor. Appl. Genet. 105, 145–159. doi: 10.1007/ of a putative ancestral genome for the family. BMC Evol. Biol. 11, 9. doi: s00122-001-0841-9 10.1186/1471-2148-11-9 FAOSTAT (2018). Food and Agriculture Organization of United Nations. Statistics Infante, R., Martínez-Gómez, P., and Predieri, S. (2008). Quality oriented fruit division. http://faostat3.fao.org. Accessed 15 January 2019. breeding: Peach [Prunus persica (L.) Batsch]. J. Food Agric. Environ. 6 (2), Fresnedo-Ramírez, J., Bink, M. C., van de Weg, E., Famula, T. R., and Crisosto, 342–356. doi: 10.1234/4.2008.1239 C. H. (2015). QTL mapping of pomological traits in peach and related species Jones, B., Frasse, P., Olmos, E., Zegzouti, H., Li, Z. G., Latché, A., et al. (2002). breeding germplasm. Mol. Breed. 35, 1–19. doi: 10.1007/s11032-015-0357-7 Down-regulation of DR12, an auxin-response-factor homolog, in the tomato Galpaz, N., Gonda, I., Shem-Tov, D., Barad, O., Tzuri, G., Lev, S., et al. (2018). results in a pleiotropic phenotype including dark green and blotchy ripening Deciphering genetic factors that determine melon fruit-quality traits using fruit. Plant J. 32, 603–613. doi: 10.1046/j.1365-313X.2002.01450.x RNA-Seq-based high-resolution QTL and eQTL mapping. Plant J. 94, 169–191. Jung, S., Cestaro, A., Troggio, M., Main, D., Zheng, P., Cho, I., et al. (2012). Whole doi: 10.1111/tpj.13838 genome comparisons of Fragaria, Prunus and Malus reveal different modes Gapper, N. E., McQuinn, R. P., and Giovannoni, J. J. (2013). Molecular and genetic of evolution between Rosaceous subfamilies. BMC Genomics 13, 129. doi: regulation of fruit ripening. Plant Mol. Biol. 82 (6), 575–591. doi: 10.1007/ 10.1186/1471-2164-13-129 s11103-013-0050-3 Kumar, R., Khurana, A., and Sharma, A. K. (2014). Role of plant hormones and García-Gómez, B. E., Salazar, J. A., Dondini, L., Martínez-Gómez, P., and Ruiz, D. their interplay in development and ripening of fleshy fruits. J. Exp. Bot. 65 (16), (2019). Identification of QTLs linked to fruit quality traits in apricot (Prunus 4561–4575. doi: 10.1093/jxb/eru277 armeniaca L) and biological validation through gene expression analysis using Labavitch, J. M., and Ray, P. M. (1974). Relationship between promotion of qPCR. Mol. Breed. 39 (2), 28. doi: 10.1007/s11032-018-0926-7 xyloglucan metabolism and induction of elongation by IAA. Plant Physiol. 54, Genome Database for Rosaceae (GDR). Available at: http://www.rosaceae.org/. 499–502. doi: 10.1104/pp.54.4.499 Accessed 24 May. Laurens, F., Aranzana, M. J., Arús, P., Bassi, D., Bink, M., Bonany, J., et al. (2018). Ghiani, A., Onelli, E., Aina, R., Cocucci, M., and Citterio, S. (2011). A comparative An integrated approach for increasing breeding efficiency in apple and peach study of melting and non-melting flesh peach cultivars reveals that during fruit in Europe. Hortic. Res. 5, 11. doi: 10.1038/s41438-018-0016-3 ripening endopolygalacturonase (endo-PG) is mainly involved in pericarp Lester, D. R., Speirs, G., Orr, G., and Brady, C. J. (1994). Peach (Prunus textural changes, not in firmness reduction. J. Exp. Bot. 62 (11), 4043–4054. persica) endo-PG cDNA isolation and mRNA analysis in melting and doi: 10.1093/jxb/err109 non-melting peach cultivars. Plant Physiol. 105, 225–231. doi: 10.1104/ Gillaspy, G., Ben-David, H., and Gruissem, W. (1993). Fruits: a developmental pp.105.1.225 perspective. Plant Cell 5, 1439–1451. doi: 10.2307/3869794 Lester, D. R., Sherman, W. B., and Atwell, B. J. (1996). Endopolygalacturonase and Giné-Bordonaba, J., Cantín, C. M., Echeverría, G., Ubach, D., and Larrigaudière, C. the melting flesh (M) locus in peach. J. Am. Soc. Hortic. Sci. 121, 231–235. doi: (2016). The effect of chilling injury-inducing storage on quality and consumer 10.21273/JASHS.121.2.231 Frontiers in Plant Science | www.frontiersin.org 14 December 2019 | Volume 10 | Article 1581 Carrasco-Valenzuela et al. Expression QTL Analyses in Peach Li, H., Handsaker, B., Wysoker, A., Fennell, T., Ruan, J., Homer, N., et al. (2009). Pfaffl, M. W. (2001). A new mathematical model for relative quantification in real- The Sequence Alignment/Map format and SAMtools. Bioinformatics 25 (16), time RT-PCR. Nucleic Acids Res. 29 (9), e45. doi: 10.1093/nar/29.9.e45 2078–2079. doi: 10.1093/bioinformatics/btp352 Pirona, R., Eduardo, I., Pacheco, I., Da Silva Linge, C., Miculan, M., Verde, I., Li, Z. G., Chen, H. W., Li, Q. T., Tao, J. J., Bian, X. H., Ma, B., et al. (2015). Three et al. (2013). Fine mapping and identification of a candidate gene for a SAUR proteins SAUR76, SAUR77 and SAUR78 promote plant growth in major locus controlling maturity date in peach. BMC Plant Biol. 13, 166. doi: Arabidopsis. Sci. Rep. 5, 12477. doi: 10.1038/srep12477 10.1186/1471-2229-13-166 Liu, M., Pirrello, J., Chervin, C., Roustan, J. P., and Bouzayen, M. (2015). Ethylene Pirrello, J., Prasad, B. C. N., Zhang, W., Chen, K., Mila, I., Zouine, M., et al. Control of Fruit Ripening: Revisiting the Complex Network of Transcriptional (2012). Functional analysis and binding affinity of tomato ethylene response Regulation. Plant Physiol. 169, 2380–2390. doi: 10.1104/pp.15.01361 factors provide insight on the molecular bases of plant differential responses to Longhi, S., Moretto, M., Viola, R., Velasco, R., and Costa, F. (2012). Comprehensive ethylene. BMC Plant Biol. 12, 190. doi: 10.1186/1471-2229-12-190 QTL mapping survey dissects the complex fruit texture physiology in Apple Pressey, R., and Avants, J. K. (1978). Difference in polygalacturonase composition (Malus x domestica Borkh.). J. Exp. Bot. 63, 1107–1121. doi: 10.1093/jxb/ of clingstone and freestone peaches. J. Food Sci. 43, 1415–1423. doi: 10.1111/ err326 j.1365-2621.1978.tb02507.x Lurie, S., Friedman, H., Weksler, A., Dagar, A., and Zerbini, P. E. (2013). Qian, M., Zhang, Y., Yan, X., Han, M., Li, J., Li, F., et al. (2016). Identification and Maturity assessment at harvest and prediction of softening in an early and Expression Analysis of Polygalacturonase Family Members during Peach Fruit late season melting peach. Postharvest. Bio Technol. 76, 10–16. doi: 10.1016/j. Softening. Int. J. Mol. Sci. 17, 1933. doi: 10.3390/ijms17111933 postharvbio.2012.08.007 Ramina, A., Tonutti, P., and McGlasson, B. (2008). “Ripening, nutrition and Martínez-García, P. J., Parfitt, D. E., Bostock, R. E., Fresnedo-Ramírez, J., postharvest physiology,” in The peach: botany, production and uses. Eds. D. R. Vazquez-Lobo, A., Ogundiwin, E. A., et al. (2013). Application of genomic Layne, and D. Bassi (Wallingford, UK: CAB International Press), 550–574. doi: and quantitative genetic tools to identify candidate resistance genes for 10.1079/9781845933869.0550 brown rot resistance in peach. PloS One 8 (11), e78634. doi: 10.1371/ Robinson, M. D., McCarthy, D. J., and Smyth, G. K. (2010). edgeR: a Bioconductor journal.pone.0078634 package for differential expression analysis of digital gene expression data. Martin, M. (2011). Cutadapt removes adapter sequences from high-throughput Bioinformatics 26 (1), 139–140. doi: 10.1093/bioinformatics/btp616 sequencing reads. EMBnet. J. 17 (1), 10–12. doi: 10.14806/ej.17.1.200 Rockman, M. V., and Kruglyak, L. (2006). Genetics of global gene expression. Nat. Mashiguchi, K., Tanaka, K., Sakai, T., Sugawara, S., Kawaide, H., Natsume, M., Rev. Genet. 7, 862–872. doi: 10.1038/nrg1964 et al. (2011). The main auxin biosynthesis pathway in Arabidopsis. Proc. Natl. Salazar, J. A., Ruiz, D., Campoy, J. A., Sánchez-Pérez, R., Crisosto, C. H., Martínez- Acad. Sci. U.S.A. 108, 18512–18517. doi: 10.1073/pnas.1108434108 García, P. J., et al. (2014). Quantitative trait loci (QTL) and Mendelian trait loci Meneses, C., Ulloa-Zepeda, L., Cifuentes-Esquivel, A., Infante, R., Cantin, C. M., (MTL) analysis in Prunus: a breeding perspective and beyond. Plant Mol. Biol. Batlle, I., et al. (2017). A codominant diagnostic marker for the slow ripening Rep. 32, 1–18. doi: 10.1007/s11105-013-0643-7 trait in peach. Mol. Breed. 36, 77. doi: 10.1007/s11032-016-0506-7 Salazar, J. A., Pacheco, I., Shinya, P., Zapata, P., Silva, C., Ruiz, D., et al. (2017). Miller, A. N., Walsh, C. S., and Cohen, J. D. (1987). Measurement of indole-3- Genotyping by sequencing for SNP-based linkage analysis and identification acetic acid in peach fruits (Prunus persica L. Batsch cv. Redhaven) during of QTLs linked to fruit quality traits in Japanese plum (Prunus salicina Lindl.). development. Plant Physiol. 84, 491–494. doi: 10.1104/pp.84.2.491 Front. Plant Sci. 8, 476. doi: 10.3389/fpls.2017.00476 Morgutti, S., Negrini, N., Nocito, E. F., Ghiani, A., Bassi, D., and Cocucci, M. Serra, O., Giné-Bordonaba, J., Eduardo, I., Bonany, J., Echeverria, G., (2006). Changes in endopolygalacturonase levels and characterization of Larrigaudière, C., et al. (2017). Genetic analysis of the slow-melting a putative endo-PG gene during fruit softening in peach genotypes with flesh character in peach. Tree Genet. Genomes 13, 77. doi: 10.1007/ nonmelting and melting flesh fruit phenotype. New Phytol. 171, 315–325. doi: s11295-017-1160-x 10.1111/j.1469-8137.2006.01763.x Shulaev, V., Korban, S. S., Sosinski, B., Abbott, A. G., Aldwinckle, H. S., Folta, K. M., NCBI’s Sequence Read Achieve. Available at: http://www.ncbi.nlm.nih.gov/sra. et al. (2008). Multiple models for Rosaceae genomics. Plant Physiol. 157, 985– Nilo, R., Campos-Vargas, R., and Orellana, A. (2012). Assessment of Prunus persica 1003. doi: 10.1104/pp.107.115618 fruit softening using a proteomics approach. J. Proteomics 75, 1618–1638. doi: Stranger, B. E., Nica, A. C., Forrest, M. S., Dimas, A., Bird, C. P., Beazley, C., et al. 10.1016/j.jprot.2011.11.037 (2007). Population genomics of human gene expression. Nat. Genet. 39, 1217– Nishitani, K., and Masuda, Y. (1981). Auxin-induced changes in the cell wall 1224. doi: 10.1038/ng2142 structure: changes in the intrinsic viscosity and molecular weight distribution Sugiyama, A., Omura, M., Shimoda, T., Fuji, H., Endo, T., Shimizu, T., et al. (2014). of matrix polysaccharides of the epicotyl cell wall of Vigna angularis. Physiol. Expression quantitative trait loci analysis of carotenoid metabolism-related Plant 52, 482–494. doi: 10.1111/j.1399-3054.1981.tb02720.x genes in Citrus. J. Japan Soc. Hort. Sci. 83, 32–43. doi: 10.2503/jjshs1.CH-054 Nuñez-Lillo, G., Cifuentes-Esquivel, A., Troggio, M., Micheletti, D., Infante, R., Campos- Tatsuki, M., Haji, T., and Yamaguchi, M. (2006). The involvement of Vargas, R., et al. (2015). Identification of candidate genes associated with mealiness 1-aminocyclopropane-1-carboxylic acid synthase isogene, Pp-ACS1, in peach and maturity date in peach [Prunus persica (L.) Batsch] using QTL analysis and deep fruit softening. J. Exp. Bot. 57, 1281–1289. doi: 10.1093/jxb/erj097 sequencing. Tree Genet. Genomes 11 (4), 1–13. doi: 10.1007/s11295-015-0911-9 Tatsuki, M., Nakajima, N., Fujii, H., Shimada, T., Nakano, M., Hayashi, K. I., et al. Ogundiwin, E. A., Peace, C. P., Gradziel, T. M., Parfitt, D. E., Bliss, F. A., and (2013). Increased levels of IAA are required for system 2 ethylene synthesis Crisosto, C. H. (2009). A fruit quality gene map of Prunus. BMC Genom. 10, causing fruit softening in peach (Prunus persica L. Batsch). J. Exp. Bot. 64 (4), 587. doi: 10.1186/1471-2164-10-587 1049–1059. doi: 10.1093/jxb/ers381 Ohmiya, A. (2000). Effects of auxin on growth and ripening of mesocarp discs of Taylor, S., Wakem, M., Dijkman, G., Alsarraj, M., and Nguyen, M. (2010). A peach fruit. Sci. Hortic. 84, 309–319. doi: 10.1016/S0304-4238(99)00137-5 practical approach to RT-qPCR-Publishing data that conform to the MIQE Operon software. Available at: http://www.operon.com/tools/oligo-analysis-tool. guidelines. Methods. 50, S1–S5. doi: 10.1016/j.ymeth.2010.01.005 aspx. Accessed 01 December. Tong, Z. G., Gao, Z. H., Wang, F., Zhou, J., and Zhang, Z. (2009). Selection of Pacheco, I., Bassi, D., Eduardo, I., Ciacciulli, A., Pirona, R., Rossini, L., et al. reliable reference genes for gene expression studies in peach using real-time (2014). QTL mapping for brown rot (Monilinia fructigena) resistance in an PCR. BMC Mol. Biol. 10, 71. doi: 10.1186/1471-2199-10-71 intraspecific peach (Prunus persica L. Batsch) F1 progeny. Tree Genet. Genomes Tonutti, P., Casson, P., and Ramina, A. (1991). Ethylene biosynthesis during 10 (5), 1223–1242. doi: 10.1007/s11295-014-0756-7 peach fruit-development. J. Am. Soc. Hort. Sci. 116, 274–279. doi: 10.21273/ Pan, L., Zeng, W., Niu, L., Lu, Z., Wang, X., Liu, H., et al. (2015). PpYUC11, a JASHS.116.2.274 strong candidate gene for the stony hard phenotype in peach (Prunus persica L. Trainotti, L., Tadiello, A., and Casadoro, G. (2007). The involvement of auxin in the Batsch), participates in IAA biosynthesis during fruit ripening. J. Exp. Bot. 66 ripening of climacteric fruits comes of age: the hormone plays a role of its own (22), 7031–7044. doi: 10.1093/jxb/erv400 and has an intense interplay with ethylene in ripening peaches. J. Exp. Bot. 58, Peace, C., Crisosto, C., and Gradziel, T. (2005). Endopolygalacturonase: a 3299–3308. doi: 10.1093/jxb/erm178 candidate gene for Freestone and Melting flesh in peach. Mol. Breed. 16, 21–31. Van Ooijen, J. (2006). JoinMap 4. Software for the calculation of genetic linkage doi: 10.1007/s11032-005-0828-3 maps in experimental populations. Wageningen Kyazma BV Frontiers in Plant Science | www.frontiersin.org 15 December 2019 | Volume 10 | Article 1581 Carrasco-Valenzuela et al. Expression QTL Analyses in Peach Verde, I., Abbott, A., Scalabrin, S., Jung, S., Schu, Sh., Marroni, F., et al. (2013). The Zhang, B., Peng, B., Zhang, C., Song, Z., and Ma, R. (2017). Determination of fruit high-quality draft genome of peach (Prunus persica) identifies unique patterns maturity and its prediction model based on the pericarp index of absorbance of genetic diversity, domestication and genome evolution. Nat. Genet. 45, 487– difference (IAD) for peaches. PloS One 12 (5), e0177511. doi: 10.1371/journal. 494. doi: 10.1038/ng.2586 pone.0177511 Wickham, H. (2016). Data Analysis. In: ggplot2. Use R!. Cham, Switzerland: Zhebentyayeva, T. N., Swire-Clark, G., Georgi, L. L., Garay, L., Jung, S., Forrest, S., Springer. doi: 10.1007/978-3-319-24277-4_9 et al. (2008). A framework physical map for peach, a model Rosaceae species. Yang, N., Reighard, G., Ritchie, D., Okie, W., and Gasic, K. (2013). Mapping Tree Genet. Genome 4, 745–756. doi: 10.1007/s11295-008-0147-z quantitative trait loci associated with resistance to bacterial spot (Xanthomonas arboricola pv. pruni) in peach. Tree Genet. Genomes 9 (2), 573–586. doi: Conflict of Interest: The authors declare that the research was conducted in the 10.1007/s11295-012-0580-x absence of any personal, professional or financial relationship that could constitute Ye, J., Liu, P., Zhu, C., Qu, J., Wang, X., Sun, Y., et al. (2014). Identification of a conflict of interest. candidate genes JcARF19 and JcIAA9 associated with seed size traits in Jatropha. Funct. Integr. Genomics 14 (4), 757–766. doi: 10.1007/s10142-014-0400-5 Copyright © 2019 Carrasco-Valenzuela, Muñoz-Espinoza, Riveros, Pedreschi, Yu, G., Wang, L., Han, Y., and He, Q. (2012). clusterProfiler: an R package for Arús, Campos-Vargas and Meneses. This is an open-access article distributed comparing biological themes among gene clusters. OMICS. A. J. Integr. Biol. 16 under the terms of the Creative Commons Attribution License (CC BY). The (5), 284–287. doi: 10.1089/omi.2011.0118 use, distribution or reproduction in other forums is permitted, provided the Zeballos, J. L., Abidi, W., Gimenez, R., Monforte, A. J., Angeles Moreno, M., and original author(s) and the copyright owner(s) are credited and that the original Gogorcena, Y. (2016). Mapping QTLs associated with fruit quality traits in publication in this journal is cited, in accordance with accepted academic peach [Prunus persica (L.) Batsch] using SNP maps. Tree Genet. Genomes 12 practice. No use, distribution or reproduction is permitted which does not comply (3), 37. doi: 10.1007/s11295-016-0996-9 with these terms. Frontiers in Plant Science | www.frontiersin.org 16 December 2019 | Volume 10 | Article 1581