This is the accepted version of the article: Giné Bordonaba, Jordi; Eduardo, Iban; Arús i Gorina, Pere; [et al.]. «Bio- chemical and genetic implications of the slow ripening phenotype in peach fruit». Scientia Horticulturae, Vol. 259 (January 2020), art. 108824. DOI 10.1016/j.scienta.2019.108824 This version is available at https://ddd.uab.cat/record/255784 under the terms of the license 1 Biochemical and genetic implications of the slow ripening phenotype in peach fruit 2 Jordi Giné-Bordonabaa, Iban Eduardob, Pere Arúsb and Celia M. Cantín*c,d,e a 3 IRTA XaRTA-Postharvest, Edifici Fruitcentre, Parc Científic i Tecnològic 4 Agroalimentari de Lleida, 25003, Lleida, Spain. b 5 IRTA, Centre de Recerca en Agrigenomica CSIC-IRTA-UAB-UB, Campus UAB, 6 Edifici CRAG, Cerdanyola del Vallès (Bellaterra), 08193 Barcelona, Spain. c 7 Aragon Agency for Research and Development (ARAID), E-50018, Zaragoza, Spain d 8 Horticulture Unit, Agrifood Research and Technology Centre of Aragon (CITA), 9 Aragon Government, 50059, Zaragoza, Spain. 10 e Instituto Agroalimentario de Aragón – IA2 (CITA-Universidad de Zaragoza), 50013, 11 Zaragoza, Spain. 12 *corresponding author: cmcantin@cita-aragon.es 13 14 Highlights 15 - Different allele combinations of the Sr gene influence peach ripening physiology and 16 biochemistry. 17 - The Sr gene affects ethylene synthesis, and therefore all the ethylene-dependent 18 ripening changes. 19 - Sr2sr hybrids showed a longer harvest window and improved postharvest behaviour 20 compared to Sr1sr and Sr1Sr2. 21 - Selection of Sr2sr hybrids could be an efficient strategy for improved shelf life in new 22 peach cultivars. 23 Abstract 24 The peach [Prunus persica L. (Batsch)] slow ripening (SR) trait is a mutation preventing 25 the normal fruit ripening process. It is determined by a single Mendelian gene (Sr/sr) 26 located on linkage group 4, where only homozygous individuals for a recessive allele (sr) 27 show the SR phenotype and are generally discarded from breeding programs. Ripening- 28 related traits such as fruit weight, firmness loss, ethylene production, ACO activity, 29 sugars and organic acids composition, malondialdehyde, antioxidant capacity and total 30 phenolic content were evaluated in a segregating population for the SR trait during two 1 31 consecutive harvest seasons and at different maturity stages. Although there is no 32 commercial value for the slow ripening (srsr) individuals, our results demonstrate that a 33 heterozygous combination involving sr and another allele at this locus (Sr2) showed 34 interesting traits including a longer harvest window and improved postharvest behaviour 35 if harvested at the appropriate maturity (IAD ≥ 2). All these traits seem to be linked to a 36 delayed ripening behaviour mediated, in turn, by a lower ethylene production capacity 37 and an altered sugar (mainly sucrose) and organic acid accumulation/utilisation on-tree. 38 The selection of this allelic combination could be an easy and efficient strategy to obtain 39 new peach cultivars with potentially improved shelf life. 40 Graphical abstract 41 42 Keywords 43 Breeding, climacteric, fruit quality, firmness, softening, texture. 44 45 Introduction 2 46 Peach is a typical climacteric fruit, with a rapid softening and short shelf life after 47 harvest that adversely affect its market value. Fruit maturation is a complex and highly 48 coordinated developmental process that affects colour, firmness, taste and flavour. Fruit 49 ethylene production rate has a clear effect on the firmness loss and senescence of the fruit, 50 and therefore in the fruit storage life (Barry and Giovannoni, 2007; Osorio et al., 2011; 51 Giné-Bordonaba et al., 2016). 52 Peach fruit showing the slow ripening (SR) phenotype fail to ripen. This character 53 was first described in some breeding populations as fruit that remained firm when mature, 54 with a very slow rate of flesh softening, a reduced CO2 and C2H4 production, poor sensory 55 qualities and a high susceptibility to internal breakdown (Brecht et al., 1984, 1982). The 56 same authors also reported that exogenous C2H4 application failed to induce ripening as 57 normally observed in other climacteric fruit. 58 Some years later (Ramming, 1991; Tataranni et al., 2010), the slow-ripening trait 59 was proposed to be controlled by a single gene (Sr/sr), and when the recessive sr allele 60 was in homozygosis, it prevented the fruit undergoing normal ripening. More recently, 61 the sr gene has been mapped on linkage group 4 (G4) and molecular markers for its 62 selection have been developed (Eduardo et al., 2015; Meneses et al., 2016; Nuñez-Lillo 63 et al., 2015). These markers are based on a large deletion of 26.6 kb containing two NAC 64 transcription factors that could be the causal mutation. In fact, genes of the NAC family 65 have been shown to be involved in the regulation of ethylene-mediated ripening in tomato 66 (Osorio et al., 2011) and banana (Chen et al., 2012). A major gene/QTL for maturity date 67 (Eduardo et al., 2015; Pirona et al., 2013) and a QTL for the chilling injury symptoms of 68 fruit fresh mealiness and bleeding (Martínez-García et al., 2013) have also been identified 69 at the Sr genomic region. More recently, Botton et al. (2016) proposed that the peach 70 HEC3-like gene FLESHY may be involved in the SR phenotype, having an important 3 71 role in fruit tissue patterning at early fruit development. Information also exists on the 72 relationship between the ripening behaviour and ethylene production of different SR 73 (srsr) fruit (Brecht et al., 1982), but no other studies have ever investigated and compared 74 the physiological and biochemical differences occurring during ripening among Srsr and 75 SrSr genotypes. 76 Understanding the effect of the sr allele on the softening of peach in combination 77 with the above-mentioned and available molecular marker (Eduardo et al., 2015; Meneses 78 et al., 2016; Nuñez-Lillo et al., 2015), would offer breeders the possibility to include this 79 character in their progenies to obtain new cultivars with potentially longer shelf life. For 80 instance, a slower softening would allow a wider harvesting window, as well as an 81 extended postharvest life. In addition, a better understanding on the biochemical 82 mechanisms underlying such regulatory control of ripening would open the possibility to 83 modify certain biochemical pathways aiming to obtain peaches with extended postharvest 84 life. 85 Accordingly, the objective of this work was to deep insight the effect of each sr 86 allele configuration in the peach maturation/ripening physiology. To do so, we analysed 87 a series of traits such as fruit firmness, ethylene production, IAD, weight, biochemical 88 compounds (glucose, sucrose, malic and citric acid) and antioxidant metabolism related 89 molecules (malondialdehyde, antioxidant capacity, and total phenolic content) along 90 ripening in a SR segregating population for two consecutive harvest seasons. 91 92 Material and Methods 93 Plant material 4 94 A segregating F1 progeny (BbxNl) from the cross between the white peach 95 cultivar ‘Belbinette’ (Bb) and the yellow nectarine ‘Nectalady’ (Nl) was used in this 96 study. Trees were planted in the fields at the IRTA Experimental Station in Gimenells 97 (Lleida, Spain) initially on their own roots (2007) and later (2008) grafted on ‘Cadaman’ 98 rootstock. 99 For fruit phenotyping, including physiological and biochemical measurements, a 100 subset of 25 individuals plus the parents was selected. Available information about the 101 allelic composition of the parents Belbinette (Sr2sr) and Nectalady (Sr1sr) and offspring 102 from this population (Eduardo et al., 2015; Meneses et al., 2016), was used to select the 103 subset of individuals analyzed. The difference between Sr1 and Sr2 could be established 104 by Eduardo et al. (2015) as they were associated with an ~10-day interval in maturity 105 dates of the parents, earlier for Sr1 and later for Sr2. The selection in the progeny was 106 made in order to have represented all the genotypic classes, and the most variability of 107 maturity dates within each class: 108 - Slow ripening individuals (srsr, N=3) 109 - Individuals heterozygous for the sr allele: (Sr1sr, N=7) and (Sr2sr, N=7) 110 - Individuals not carrying the sr allele (Sr1Sr2, N=8) 111 112 Fruit growth 113 On-tree fruit growth (size) and non-destructive evaluation of the fruit maturity stage (DA- 114 value) both from the parents and two hybrids from each allelic class (srsr, Sr1sr, Sr2sr 115 and Sr1Sr2) was periodically monitored (every 7-10 days) from fruit set to harvest on 10 116 fruit randomly selected. Fruit size was measured with a digital calliper, as the distance 117 between the suture and the opposite side at the equatorial zone. DA-values, also known 118 as the index of Absorbance difference (IAD), were measured with a commercial equipment 5 119 (DA-Meter, TR Turoni, Forli, Italy), which measures non-destructively Vis-spectroscopy 120 according to the IAD index (index of absorbance difference = A670–A720) (Ziosi et al., 121 2008). 122 123 Fruit ripening and ethylene measurements 124 Forty fruit were harvested from each tree at 3 different maturity stages based on fruit 125 firmness (60-80N (M1), 40-60N (M2), <30N (M3)), IAD values and historical data on 126 their ripening pattern. In the case of sr fruit (srsr), firmness did never reach values <30N, 127 and therefore fruit were harvested at different maturity stages based exclusively in days 128 after full bloom (DAFB). Upon each harvest, fruit were immediately transported to the 129 lab for quality measurements (t0). Fruit was individually weighted in a digital scale. Then, 130 IAD values and firmness were measured on both sides of each fruit using 8 131 fruits/tree/harvest (total of individuals = 25). IAD values were measured as described 132 above whereas flesh firmness was measured with a digital penetrometer (Model. 53205; 133 Turoni, Forlí, Italy) equipped with an 8-mm diameter plunger tip after the removal of a 1 134 mm thick slice of skin. Full bloom date for each tree was recorded to calculate days after 135 full bloom (DAFB) at the moment of harvest. In the selected individuals harvest window 136 ranged from early August to mid September. 137 Ethylene production and firmness were evaluated the same day of harvest (t0) and after 138 1, 3 and 6 days after harvest (t1, t3, t6). Firmness at t1, t3 and t6 was measured as 139 described for t0 using 8 fruits/tree/harvest. Remaining fruit were stored at 20ºC (70% RH) 140 for the following evaluation times. Ethylene production (uL kg−1h−1) was measured in an 141 acclimatized chamber at 20ºC. The fruit (2 or 3 fruit per replicate, and at least 3 replicates 142 per tree and harvest) were placed in airtight 1.5 L flasks. After 2 h incubation, gas samples 6 143 (1 mL) were taken from the headspace of the flasks, using a 1 mL syringe, and injected 144 into a gas chromatograph (GC; Agilent Technologies 6890, Wilmington, Germany) fitted 145 with a FID detector and an alumina column F1 80/100 (2 m × 1/8 × 2.1, Tecknokroma, 146 Barcelona, Spain) following the methodology described elsewhere (Giné-Bordonaba et 147 al., 2017). Samples were taken again from the headspace of the flasks following the same 148 methodology at t1, t3 and t6. 149 The fruit flesh from each sample (8 fruits/tree) at each maturity stage (M1, M2 and M3) 150 was frozen at the time of harvest and stored at -80ºC until being used for biochemical 151 measurements. 152 153 Biochemical measurements 154 The enzyme 1-Aminocyclopropane-1-carboxylic acid oxidase (ACO) was 155 extracted as described by Lindo-Garcia et al. (2019) with some modifications. The sample 156 (0.5 g of frozen tissue) was homogenized in 1 mL of buffer containing 400 mM MOPS at 157 pH 7.2, 10% glycerol, 30 mM ascorbic acid sodium salt and PVP 40000 2%. The 158 homogenized was slightly shaken for 10 min at 1⁰C and centrifuged at 17,000 x g for 30 159 min at 4°C. Subsequently, the supernatant was stored at -80°C until analysis of the enzyme 160 activity using the methodology described by Giné-Bordonaba et al. (2017). 161 Sugars (glucose and sucrose) and organic acids (citric and malic acid) were extracted from 162 frozen tissue as described by Giné-Bordonaba et al. (2019) with some modifications. For 163 sugars determination, 2 g of sample were diluted in 5 mL of 62.5% (v/v) aqueous methanol 164 solvent and placed in a thermostatic bath at 55°C for 15 min, mixing the solution with a 165 vortex every 5 min to prevent layering. Then, the samples were centrifuged at 24,000 x g 166 for 15 min at 20 °C. The supernatants of each sample were recovered and used for enzyme 167 coupled spectrophotometric determination of glucose (hexokinase) and sucrose (β- 7 168 fructosidase) using commercial kits (BioSystems S.A., Barcelona, Spain) and following 169 the manufacturer instructions. 170 Citric and malic acid were extracted by dissolving 2 g of frozen tissue in 5 mL of distillate 171 water. The solution was slightly shaken for 10 min at room temperature and then 172 centrifuged at 24,000 x g for 7 min at 20°C. The resulting supernatant was recovered and 173 used to determine the concentration of citric (citrate lyase/malate dehydrogenase) and 174 malic acid (malate dehydrogenase) using commercial kits (BioSystems S.A., Barcelona, 175 Spain) and following the manufacturer instructions. 176 177 Statistical analysis 178 All data were subjected to analysis of variance (ANOVA) using JMP® 13.1.0 SAS 179 Institute Inc. Mean comparisons for the interaction genotype * maturity was evaluated 180 using HSD test at a significance level of p ≤ 0.05, while comparisons for specific traits 181 between genotypes along storage time or days after full bloom was done by least 182 significant difference values (LSD; p ≤ 0.05) using critical values of t for two-tailed tests. 183 Correlations between experimental variables were made using Spearman's Rank 184 Correlations and, if required, presented as Spearman's Correlation Coefficient (r) and P 185 value based on a two-tailed test. 186 Results 187 Fruit maturation 188 As described above, the different maturity stages (M1, M2 and M3) were selected based 189 on both firmness and IAD values and generally comprised fruit harvested from 140 to 188 190 DAFB yet depending on the genotypic class (Table 1). The fruit weight of all the genetic 191 classes increased progressively in the tree during maturation (Fig. 2). However, fruit 192 weight of the SR individuals (sr/sr) individuals increased very little, reaching very low 8 193 weights if compared to the other individuals. In all genotypes, the increase in fruit size 194 seemed to be intimately regulated by the fruit maturity since the increase in fruit size was 195 negatively correlated (R2 >0.85; Figure 8) to the loss of fruit firmness and the IAD value. 196 The IAD, which indicates the degree of chlorophyll degradation in the fruit, showed a 197 progressive decrease at different maturities for the genetic classes Sr1Sr2, Sr1sr and Sr2sr 198 (Table 1). In the slow ripening fruit (sr/sr), the IAD failed, as expected, to decrease 199 indicating that the chlorophyll levels remained very high and it was not degraded as it 200 occurs in a normal ripening process. Based on the IAD values, the selected maturity classes 201 (M1, M2 and M3) were comparable among the different genotypic classes except for SR 202 individuals that besides not growing (Table 1) did not show a substantial chlorophyll 203 degradation on-tree. 204 Indeed, fruit growth was similar for all the genetic classes from fruit set to approximately 205 135 days after full bloom (DAFB) (Fig. 3). However, after approximately 145 DAFB the 206 fruit growth of the SR (sr/sr) individuals started to slow down if compared to the rest of 207 individuals, reaching smaller sizes at the end of the fruit development (ca. 190 DAFB). 208 Similarly, the degradation of chlorophyll (IAD) in the SR individuals started to slow down 209 at around 155 DAFB compared to the rest of individuals. It can be also observed that the 210 genetic classes with one copy of the sr allele (Sr1sr and Sr2sr) had a slower degradation 211 of chlorophyll at these last stages of the fruit ripening than the rest of the non-slow 212 ripening (Sr1Sr2) individuals. 213 Firmness loss and ethylene production 214 The flesh firmness of the fruit from genetic class Sr1Sr2 rapidly began to decrease after 215 1 day of shelf life (Fig. 4), regardless the initial fruit firmness at harvest (M1, M2 and 216 M3). On the other side, and as expected, the SR genotypes (sr/sr) failed to soften during 9 217 the shelf life period (6d), and their firmness remained over 60N. The hybrids Sr1sr and 218 Sr2sr showed a different behaviour to the former genetic classes, showing mixed results 219 depending on the fruit maturity at harvest. In detail, these genotypes soften rapidly if 220 harvested at firmness ≤ 60N yet the loss of firmness was prevented if the fruit were 221 harvested at less mature ripening stages (firmness > 60N; IAD ≥ 2). In the case of the Sr2sr 222 individuals, fruit firmness decreased slightly after one day of shelf life (generally referred 223 as softening), but then it remained stable after 3 days of shelf life (non-melting phase) 224 contrasting with Sr1sr individuals, where it kept decreasing and hence experienced both 225 softening and melting. Differences on the on-tree flesh firmness reached by the 226 individuals from the four genetic classes were very clear after 190 DAFB (Fig. 4). As 227 expected, slow ripening individuals (srsr) showed the highest firmness, individuals 228 without the sr allele (Sr1Sr2) showed the lowest, whereas the heterozygous classes (Sr1sr 229 and Sr2sr) showed intermediate values. Fruit from Sr1Sr2 and Sr1sr individuals showed 230 the highest and similar firmness loss (FL) rates (-0.95 and -1.05 N/day, respectively), 231 whereas Sr2sr showed a similar FL rates than sr/sr individuals (-0.76 and -0.62 N/day, 232 respectively). 233 We observed clear differences in the ethylene production rates among the four genetic 234 classes in the BbxNl population (Fig. 5). Ethylene production at harvest for M1 was 235 negligible for all the individuals, independently of their genetic class. However, 236 significant differences were observed in the fruit after some days of shelf life at 20ºC. No 237 detectable or negligible levels were found for srsr and Sr2sr individuals during shelf life, 238 whereas fruit from Sr1Sr2 trees produced large amounts of ethylene after 12 days of shelf 239 life at 20ºC. Fruit from Sr1sr individuals started to produce some ethylene after 13 days 240 of shelf life. The fruit from Sr2sr individuals did not produce any ethylene during shelf 241 life when harvested at flesh firmness of 60N. However, they were able to produce 10 242 ethylene during shelf life when harvested in a more advanced maturity stage (M3, flesh 243 firmness <30N) (results not shown). On the other side, the slow ripening fruit (sr/sr) did 244 not produce ethylene during shelf life for any of the three maturity stages investigated 245 herein. 246 ACC oxidase activity 247 Similarly to what happened for ethylene production, no ACC oxidase (ACO) activity was 248 observed at harvest for any of the genetic classes when harvested at M1 (Fig. 6). 249 Significant differences were observed at more advanced maturity stages (M2 and M3). 250 At M2, ACO activity was detected in fruit from Sr1sr and Sr1Sr2 individuals, whereas it 251 was almost no detectable for Sr2sr and srsr individuals. The ACO activity was negligible 252 for any of the maturity stages in the fruit from the SR individuals (sr/sr). However, in the 253 case of the Sr2sr individuals, the ACO activity significantly increased in the more mature 254 fruit (M3), although reaching much lower levels than Sr1Sr2 and Sr1sr classes at similar 255 maturity stages. 256 Sugars and organic acids 257 The pattern of sucrose accumulation in the fruit was significantly different among 258 different genetic classes (Table 1). In general, sucrose increased as fruit ripened on the 259 tree. However, accumulation in fruit from Sr2sr and sr/sr individuals was slower, 260 reaching much lower levels than the rest of fruit, especially in the slow ripening 261 individuals (sr/sr). In contrast, glucose levels decreased throughout maturation in all the 262 genetic classes, except in the sr/sr individuals where it slightly increased (Table 1). 263 Malic acid increased during maturation on the tree for the four genetic classes (Table 1). 264 Thus said, final malate content was significantly lower in the sr/sr and Sr2sr genotypes. 265 Contrary, citric acid decreased over on-tree maturation in all cases. In this case, Sr2sr and 11 266 sr/sr genotypes showed a slower rate of citric acid catabolism, whereas Sr1Sr2 and Sr1sr 267 genotypes showed a rapid drop in the concentration of this acid in the first stages of 268 maturation. 269 Discussion 270 271 Peach ripening involves dramatic changes in the colour, firmness, texture, aroma, sugars 272 and organic acids composition. As for other climacteric fruit, these changes are triggered 273 by rapid changes in the rate of C2H4 production (Tonutti et al., 1991; Alexander, 2002; 274 Baró-Muntel et al., unpublished). However, SR peach fruit show a failure to ripen and 275 hence do not undergo such dramatic changes (Brecht et al., 1984; Ramming, 1991). This 276 behaviour may be caused by the inability to synthesize ethylene as in anti-sense tomato 277 (Picton, 1993) and melon mutants (Ayub et al., 1996), or due to problems in the ethylene 278 perception as in the Nr tomato mutant (Wilkinson et al., 1995). Non-climacteric 279 phenotypes may be also due to alterations upstream the ripening cascade as in rin, nor 280 and Cnr tomato mutants (Giovannoni, 2001). SR fruit do not respond to exogenous 281 ethylene treatment, as it occurs with Nr tomato and PI 161375 melon mutants (Wilkinson 282 et al., 1995; Périn et al., 2002), which demonstrates that SR mutation blocks not only 283 ethylene synthesis but also ethylene perception (Supplementary Fig. 1). 284 In this study, the effect of some alleles of the Sr gene in the ripening physiology and 285 biochemistry as well as in the fruit postharvest behaviour, even when present in 286 heterozygosis, have been investigated. Whereas Sr1Sr2 and Sr1sr allelic combinations 287 resulted in fruit undergoing a normal ripening process, both on- and off-tree, we observed 288 changes in several ripening related parameters in offspring with the Sr2sr genotype. 289 Individuals carrying the sr allele as Sr1sr, Sr2sr and srsr produced little or no ethylene 290 when harvested at IAD ≥ 2. The results observed at the firmness loss and ethylene 12 291 production rate in the individuals carrying the sr allele, and especially in Sr2sr 292 individuals, indicate changes at the biochemical or physiological level resulting in 293 inhibition of normal ripening. Such differences at the fruit ethylene production capacity 294 were not explained by ACO activity since all individuals at M1 (when most of the 295 differences on the postharvest ripening occurred among individuals) showed minimal 296 enzyme activity (Fig. 6). Thus said, our results clearly indicate that the Sr gene 297 undoubtedly affected ethylene synthesis (Fig. 5 and 6), and therefore altered all the 298 ethylene-dependent ripening changes. Such results are in agreement with Brecht et al. 299 (1984) who reported that SR fruit harvested and stored at 20ºC, showed delayed and 300 reduced climacteric peaks of respiration and ethylene production if compared to non-SR 301 ‘Fantasia’ fruit. As reported by other authors (Botton et al., 2016), SR fruit maintained 302 typical immaturity traits throughout development. However, these authors suggested a 303 disturbance at the level of fruit patterning as the cause for SR phenotype. 304 Previous work by Eduardo et al., (2015) mapped the sr gene in the same region of linkage 305 group 4 (G4), where a NAC transcription factor candidate gene (ppa008301m) is located. 306 This gene has been proposed as a candidate gene for the maturity date (MD) trait (Pirona 307 et al., 2013). Genes from the NAC family have been shown to be involved in the 308 regulation of ethylene mediated ripening in tomato (Osorio et al., 2011) and banana (Shan 309 et al., 2014, 2012). Rapid softening occurring at late ripening stages of peach requires 310 significant levels of ethylene (Hayama et al., 2006, 2003), and therefore decreases in the 311 ethylene production rates as observed in Sr2sr fruit had a direct effect on the fruit 312 softening rate. Recently, several candidate genes for MD and mealiness (M) traits have 313 been identified in the same region of G4 (Nuñez-Lillo et al., 2015), which would explain 314 the high susceptibility to internal breakdown found in the SR fruit. Accordingly, Giné- 13 315 Bordonaba et al. (2016) also reported that a greater capacity of the fruit to produce 316 ethylene after cold storage was associated to lower mealiness incidence in peach fruit. 317 Fruit from srsr individuals failed to develop normal ripening traits such as fruit size, color, 318 aroma and flavour. Whether other loci are also responsible for the altered fruit weight in 319 SR individuals is still unknown, what is clear is that the SR locus clearly altered this fruit 320 attribute. In the peach ripening model (Lü et al., 2018), ethylene-activated NAC 321 transcription factor binds to the promoter of key fruit ripening genes such as those 322 involved in pigment accumulation, volatile secondary metabolite production, cell wall 323 softening and sugar accumulation. Therefore, the srsr allelic combination in the SR fruit 324 would disrupt the activation cascade triggered by the NAC transcription factor in some 325 way. Both the peach and melon NAC loci are located in quantitative trait loci that are 326 associated with late ripening phenotypes (Perin, 2002; Pirona et al., 2013). However, we 327 could not discard the contribution of other transcription factors or regulatory mechanisms 328 such as post-translational regulations in the ripening inhibition in the SR fruit. Indeed, 329 Sr2sr fruit, contrary to what occurs in sr/sr fruit, developed normal fruit size, normal fruit 330 colour, aroma and flavour, which would demonstrate a normal function of the 331 transcriptional ripening feedback circuit, although with certain delay likely attributed to 332 a blockage of the auxin-ethylene hormonal crosstalk (Botton et al., 2016). This delay 333 could be an interesting trait in a new peach cultivar since it would extend the shelf life of 334 the fruit and the harvest window as detailed herein (Figure 3). 335 Ethylene is synthesized from S-adenosyl-Lmethionine (SAM) via the intermediate 1- 336 aminocyclopropane-1-carboxylic acid (ACC). The conversion of SAM to ACC is 337 catalysed by ACC synthase, and the subsequent oxidation of ACC to ethylene is catalysed 338 by ACC oxidase (ACO) (Ververidis and John, 1991; Yang and Hoffman, 1984). The 339 direct effect of ethylene production in ACO activity has been proven previously by the 14 340 strong depression of ACO expression by the treatment with 1-MCP, and its induction by 341 ethephon (Zhang et al., 2012). It has been demonstrated that ACO is one of the proteins 342 with the largest change in relative abundance during the fruit transition from the pre- 343 climacteric (“unripe”) to the climacteric (“ripe”) phase (Prinsi et al., 2011). The small 344 increase of ACO in the Sr2sr genotypes observed in this work demonstrated an altered 345 pattern of ripening, althoug those individuals did not show the SR phenotype. In the peach 346 ripening model (Lü et al., 2018), ethylene transcription factor EIN3 activates the NAC 347 transcription factor, which binds to the ACO and ACS promoters to activate the ethylene 348 synthesis. Therefore, the lower ACO activity observed in the Sr2sr individuals could also 349 explain their lower ethylene levels. The same effect of a delayed ripening due to a 350 reduction in the ethylene production has been reported by the expression of the tomato 351 ACO in the antisense orientation in transgenic plants (Ayub et al., 1996; Hamilton et al., 352 1990). RNA gel blot analyses in normal ripening fruit demonstrated that ACO transcripts 353 greatly increased at late stages of fruit development (Callahan et al., 2004). 354 Glucose and sucrose contents observed in the NlxBb progeny are similar to those reported 355 in the flesh of other peach cultivars (Moing et al., 1998; Genard et al., 1999; Famiani et 356 al., 2016; Baró-Montel et al., unpublished). As observed in our work, glucose content has 357 been reported to decrease during peach ripening as a result, among others, of the large 358 increase in the fruit volume and therefore, the dilution of the glucose content within the 359 fruit (Famiani et al., 2016). The larger increase in the glucose concentration observed in 360 the slow ripening individuals (sr/sr; Table 1) through ripening could be explained by the 361 scarce increase in flesh volume as well as by the lower utilization of sugars as respiratory 362 substrates since ethylene production and respiration are largely depleted in these 363 individuals. In certain fruit species, evidence suggests that sucrose accumulation is linked 364 to the initiation of the ripening process on-tree (i.e. pears; Lindo-García et al., 2019). The 15 365 lower sucrose content and accumulation during on-tree ripening for sr/sr or Sr2sr 366 individuals agrees with the results from Botton et al. (2016) and further suggests a 367 putative role of this compound on triggering on-tree ripening also for peach fruit. 368 Malate together with citrate accounts for a large proportion of the organic acid content of 369 peach flesh (Chapman and Horvat, 1990; Byrne et al., 1991; Moing et al., 1998; Baró- 370 Montel et al., unpublished). As reported for other peach cultivars (Chapman and Horvat, 371 1990; Byrne et al., 1991; Moing et al., 1998; Famiani et al., 2016; Baró-Muntel et al., 372 unpublished), the concentration of malate in the NlxBb population increased throughout 373 ripening, whereas, that of citrate decreased. On the other hand, while sugars can be 374 synthesized both in fruit and leaves, acids are exclusively synthesized in leaves and then 375 translocated to sink (fruit). Alterations not only at the ethylene level but also in the 376 translocation pathways of these compounds, likely mediated by auxins (Daie et al., 1986) 377 may explain the lower levels of malic acid in sr/sr and Sr2sr individuals’ fruit. 378 Accordingly, our results also support previous findings from Famiani et al. (2016) who 379 pointed out that malate and citrate account only for negligible amounts of the respiratory 380 substrates during peach ripening since no association was found between these 381 compounds and the fruit respiration pattern throughout on-tree ripening (data not shown). 382 Some other authors have reported an anomalous accumulation pattern of phenolic 383 compounds at the flesh of SR fruit, more typical of lignifying endocarp (Masia et al., 384 1992). Our results do not support such findings and rather reveal that the fruit antioxidant 385 capacity was similar among individuals. Although there is not commercial value for the 386 slow ripening (sr/sr) individuals, the Sr2sr hybrids could be interesting due to their longer 387 shelf life and slower rate of firmness loss. Postharvest losses are not only a current 388 problem in developing countries but also in modern supply chains. Therefore, extending 389 shelf life of peach fruit without compromising flavour and texture quality attributes is a 16 390 desired feature in peach industry and could be accomplished by the selection of Sr2sr 391 allelic combination. Selection of this allelic combination through the already available 392 molecular markers (Meneses et al., 2016) could be an easy and efficient strategy to obtain 393 new peach cultivars with potentially improved shelf life. However, our results should be 394 confirmed in other populations with a different genetic background. 395 The presence of SR phenotypes in the progenies of commercial breeding programs is 396 frequent. Based on the data of Meneses et al. (2016) on a sample of 27 peach cultivars we 397 estimated that the frequency of the sr allele is 0.33. While this is probably an 398 overestimation of its real value because the sample of cultivars chosen included some 399 known to carry sr in heterozygosis, this indicates that the presence of this allele in the 400 peach elite breeding pool is very high, particularly considering that SR individuals are 401 systematically selected against. This scenario was already pointed out by Eduardo et al. 402 (2015) that proposed that the favourable selection of certain heterozygous combinations 403 of sr with other alleles at this locus could counteract the directional selection against sr, 404 leading to the maintenance of this allele at intermediate frequencies. 405 Acknowledgements 406 The financial support from the CERCA programme (Generalitat de Catalunya) and from 407 the Spanish Ministry of Economy and Competitiveness (MINECO/FEDER projects 408 AGL2015-68329-R and RTA2015-00050-00-00, Severo Ochoa Program for Centres of 409 Excellence in R&D 201-2019 SEV-2015-0533) is gratefully acknowledged. Thanks are 410 also given to Kaja Jensterle, Christian Fontich, Elisabet Duaigües and Dolors Ubach for 411 their technical assistance. 412 References 413 Alexander, L., 2002. Ethylene biosynthesis and action in tomato: a model for climacteric 414 fruit ripening. J. Exp. Bot. 53 (377), 2039–2055. https://doi.org/10.1093/jxb/erf072 17 415 Ayub, R., Guis, M., Amor, M. Ben, Gillot, L., Roustan, J.P., Latché, A., Bouzayen, M., 416 Pech, J.C., 1996. Expression of Selection of this allelic combination through the already 417 available molecular markers acc oxidase antisense gene inhibits ripening of cantaloupe 418 melon fruits. Nat. Biotechnol. 14, 862-866. https://doi.org/10.1038/nbt0796-862 419 Barry, C.S., Giovannoni, J.J., 2007. Ethylene and Fruit Ripening. J. Plant Growth Regul. 420 26, 143–159. https://doi.org/10.1007/s00344-007-9002-y 421 Botton, A., Rasori, A., Ziliotto, F., Moing, A., Maucourt, M., Bernillon, S., Deborde, C., 422 Petterle, A., Varotto, S., Bonghi, C., 2016. The peach HECATE3-like gene FLESHY 423 plays a double role during fruit development. Plant Mol. Biol. 91, 97–114. 424 https://doi.org/10.1007/s11103-016-0445-z 425 Brecht, J., Kader, A., Ramming, C., 1982. Postharvest physiology of non-ripening 426 nectarine selections. HortScience 17, 490. 427 Brecht, J., Kader, A., Ramming, D., 1984. Description and postharvest physiology of 428 some slow-ripening nectarine genotypes. J. Amer. Soc. Hort. Sci 109, 596–600. 429 Callahan, A.M., Scorza, R., Bassett, C., Nickerson, M., Abeles, F.B., 2004. Deletions in 430 an endopolygalacturonase gene cluster correlate with non-melting flesh texture in peach. 431 Funct. Plant Biol. 31, 159–168. https://doi.org/10.1071/FP03131 432 Chen, J., Chen, J. ye, Wang, J. ning, Kuang, J. fei, Shan, W., Lu, W. jin, 2012. Molecular 433 characterization and expression profiles of MaCOL1, a CONSTANS-like gene in banana 434 fruit. Gene 496 (2), 110-117. https://doi.org/10.1016/j.gene.2012.01.008 435 Daie, J., Watts, M., Aloni, B., Wyse, R.E., 1986. In vitro and in vivo modification of 436 sugar transport and translocation in celery by phytohormones. Plant Sci. 46 (1), 35-41. 437 https://doi.org/10.1016/0168-9452(86)90129-9 438 Eduardo, I., Picañol, R., Rojas, E., Batlle, I., Howad, W., Aranzana, M.J., Arús, P., 2015. 439 Mapping of a major gene for the slow ripening character in peach: co-location with the 440 maturity date gene and development of a candidate gene-based diagnostic marker for its 441 selection. Euphytica 205, 627–636. https://doi.org/10.1007/s10681-015-1445-9 442 Giné-Bordonaba, J., Cantín, C.M., Echeverría, G., Ubach, D., Larrigaudière, C., 2016. 443 The effect of chilling injury-inducing storage conditions on quality and consumer 444 acceptance of different Prunus persica cultivars. Postharvest Biol. Technol. 115, 38–47. 445 https://doi.org/10.1016/j.postharvbio.2015.12.006 446 Giné-Bordonaba, J., Echeverria, G., Ubach, D., Aguiló-Aguayo, I., López, M.L., 447 Larrigaudière, C., 2017. Biochemical and physiological changes during fruit development 448 and ripening of two sweet cherry varieties with different levels of cracking tolerance. 449 Plant Physiol. Biochem. 111, 216-225. https://doi.org/10.1016/j.plaphy.2016.12.002 450 Hamilton, A.J., Lycett, G.W., Grierson, D., 1990. Antisense gene that inhibits synthesis 451 of the hormone ethylene in transgenic plants. Nature 346, 284–287. 452 https://doi.org/10.1038/346284a0 18 453 Hayama, H., Ito, A., Moriguchi, T., Kashimura, Y., 2003. Identification of a new expansin 454 gene closely associated with peach fruit softening. Postharvest Biol. Technol. 29(1):1-10. 455 https://doi.org/10.1016/S0925-5214(02)00216-8 456 Hayama, H., Tatsuki, M., Ito, A., Kashimura, Y., 2006. Ethylene and fruit softening in 457 the stony hard mutation in peach. Postharvest Biol. Technol. 41, 16–21. 458 https://doi.org/10.1016/j.postharvbio.2006.03.006 459 Lü, P., Yu, S., Zhu, N., Chen, Y.R., Zhou, B., Pan, Y., Tzeng, D., Fabi, J.P., Argyris, J., 460 Garcia-Mas, J., Ye, N., Zhang, J., Grierson, D., Xiang, J., Fei, Z., Giovannoni, J., Zhong, 461 S., 2018. Genome encode analyses reveal the basis of convergent evolution of fleshy fruit 462 ripening. Nat. Plants. https://doi.org/10.1038/s41477-018-0249-z 463 Martínez-García, P.J., Parfitt, D.E., Bostock, R.M., Fresnedo-Ramírez, J., Vazquez- 464 Lobo, A., Ogundiwin, E.A., Gradziel, T.M., Crisosto, C.H., 2013. Application of genomic 465 and quantitative genetic tools to identify candidate resistance genes for brown rot 466 resistance in peach. PLoS One 8. https://doi.org/10.1371/journal.pone.0078634 467 Masia, A., Zanchin, A., Rascio, N., Ramina, A., 1992. Some Biochemical and 468 Ultrastructural Aspects of Peach Fruit Development. J. Am. Soc. Hortic. Sci. 117: 808– 469 815. 470 Meneses, C., Ulloa-Zepeda, L., Cifuentes-Esquivel, A., Infante, R., Cantin, C.M., Batlle, 471 I., Arús, P., Eduardo, I., 2016. A codominant diagnostic marker for the slow ripening trait 472 in peach. Mol. Breed. 36. https://doi.org/10.1007/s11032-016-0506-7 473 Nuñez-Lillo, G., Cifuentes-Esquivel, A., Troggio, M., Micheletti, D., Infante, R., 474 Campos-Vargas, R., Orellana, A., Blanco-Herrera, F., Meneses, C., Blanco-Herrera 475 mblanco, F., 2015. Identification of candidate genes associated with mealiness and 476 maturity date in peach [Prunus persica (L.) Batsch] using QTL analysis and deep 477 sequencing. Tree Genet. Genomes 11, 86. https://doi.org/10.1007/s11295-015-0911-9 478 Osorio, S., Alba, R., Damasceno, C.M.B., Lopez-Casado, G., Lohse, M., Zanor, M.I., 479 Tohge, T., Usadel, B., Rose, J.K.C., Fei, Z., Giovannoni, J.J., Fernie, A.R., 2011. Systems 480 Biology of Tomato Fruit Development: Combined Transcript, Protein, and Metabolite 481 Analysis of Tomato Transcription Factor (nor, rin) and Ethylene Receptor (Nr) Mutants 482 Reveals Novel Regulatory Interactions. Plant Physiol. 157(1):405-25. 483 https://doi.org/10.1104/pp.111.175463 484 Perin, C., 2002. Molecular and Genetic Characterization of a Non-Climacteric Phenotype 485 in Melon Reveals Two Loci Conferring Altered Ethylene Response in Fruit. Plant 486 Physiol. 129(1): 300–309. https://doi.org/10.1104/pp.010613 487 Pirona, R., Eduardo, I., Pacheco, I., Da Silva Linge, C., Miculan, M., Verde, I., Tartarini, 488 S., Dondini, L., Pea, G., Bassi, D., Rossini, L., 2013. Fine mapping and identification of 489 a candidate gene for a major locus controlling maturity date in peach. BMC Plant Biol. 490 13, 1. https://doi.org/10.1186/1471-2229-13-166 491 Prinsi, B., Negri, A.S., Fedeli, C., Morgutti, S., Negrini, N., Cocucci, M., Espen, L., 2011. 492 Peach fruit ripening: A proteomic comparative analysis of the mesocarp of two cultivars 19 493 with different flesh firmness at two ripening stages. Phytochemistry 72, 1251–1262. 494 https://doi.org/10.1016/J.PHYTOCHEM.2011.01.012 495 Ramming, D.W., 1991. Genetic control of SR fruit trait. Can. J. Plant Sci. 71, 601–603. 496 Shan, W., Kuang, J.F., Chen, L., Xie, H., Peng, H.H., Xiao, Y.Y., Li, X.P., Chen, W.X., 497 He, Q.G., Chen, J.Y., Lu, W.J., 2012. Molecular characterization of banana NAC 498 transcription factors and their interactions with ethylene signalling component EIL during 499 fruit ripening. J. Exp. Bot. https://doi.org/10.1093/jxb/ers178 500 Shan, W., Kuang, J.F., Lu, W.J., Chen, J.Y., 2014. Banana fruit NAC transcription factor 501 MaNAC1 is a direct target of MaICE1 and involved in cold stress through interacting 502 with MaCBF1. Plant, Cell Environ. 37(9). https://doi.org/10.1111/pce.12303 503 Tataranni, G., Spada, A., Fuzz, C., Bassi, D., 2010. AFLP-based bulk segregant analysis 504 for tagging the slow-ripening trait in peach [Prunus persica (L.) Batsch]. J. Hortic. Sci. 505 Biotechnol. 85, 78–82. https://doi.org/10.1080/14620316.2010.11512634 506 Tonutti, P., Casson, P., Ramina, A., 1991. Ethylene Biosynthesis during Peach Fruit 507 Development. J. Am. Soc. Hortic. Sci. 79 (4): 315–326. 508 Ververidis, P., John, P., 1991. Complete recovery in vitro of ethylene-forming enzyme 509 activity. Phytochemistry 30 (3), 725-727. https://doi.org/10.1016/0031-9422(91)85241- 510 Q 511 Yang, S.F., Hoffman, N.E., 1984. Ethylene Biosynthesis and its Regulation in Higher 512 Plants. Annu. Rev. Plant Physiol. 35, 155-189. 513 https://doi.org/10.1146/annurev.pp.35.060184.001103 514 Zhang, L., Jiang, L., Shi, Y., Luo, H., Kang, R., Yu, Z., 2012. Post-harvest 1- 515 methylcyclopropene and ethephon treatments differently modify protein profiles of peach 516 fruit during ripening. Food Res. Int. 48, 609–619. 517 https://doi.org/10.1016/J.FOODRES.2012.05.022 20 518 519 Tables 520 Table 1. Main organic acids (malic and citric acid), sugars (glucose and sucrose) and malonaldehyde (MDA) content of peach fruit at harvest from 521 the different genetic classes in the parents and progeny from the ‘Belbinette x Nectalady’ F1 population. Fruit harvested at different maturity stages 522 according to flesh firmness and DA-values (M1, M2 and M3) expressed in days after full bloom (DAFB). Data shown are means ± Standard 523 deviation. Malic acid Citric acid Glucose Sucrose MDA AC Genotype Maturity DAFB IAD -1 -1 -1 -1 -1 -1 mg g mg g mg g mg g nmol g AU g Sr1Sr2 M1 140.4±6.80 1.9±0.43ab 2.4±0.65 bcd 2.0±0.81 ab 8.1±0.03 ab 33.6±4.19 cdef 8.6±2.23 1.9±0.09 M2 155.4±10.13 1.2±0.44cd 4.0±0.87 ab 0.5±0.12 c 7.1±1.32 bc 48.2±6.67 bc 11.5±1.40 1.7±0.09 M3 169.8±10.52 0.7±0.32d 3.9±0.51 abc 0.6±0.16 c 5.9±0.59 bc 52.3±11.08 b 15.0±1.88 1.3±0.10 Sr1sr M1 145.1±7.82 1.8±0.23ab 2.2±0.57 bcd 2.1±0.59 a 8.7±0.55 ab 35.5±5.62 cde 10.6±2.12 1.5±0.08 M2 162.7±10.42 1.0±0.46cd 5.2±1.12 a 0.7±0.17 c 7.6±0.13 abc 72.5±5.86 a 12.6±3.63 1.4±0.40 M3 181.3±8.54 0.3±0.19d 5.1±1.57 a 0.6±0.22 c 6.1±0.38 bc 53.22±4.29 b 15.7±2.24 1.2±0.20 Sr2sr M1 143.1±5.81 2.0±0.18ab 1.5±0.52 cd 1.4±0.18 abc 8.7±0.45 ab 22.1±1.27 efg 5.2±2.11 1.6±0.06 M2 162.6±8.04 1.3±0.47cd 3.3±0.73 abcd 1.4±0.25 abc 7.2±0.43 bc 35.5±1.39 cde 8.9±1.56 1.2±0.09 M3 188.0±6.26 0.6±0.41d 2.6±0.12 bcd 0.9±0.28 bc 5.2±0.74 c 43.7±2.55 bcd 10.2±0.07 0.7±0.05 srsr M1 146.0±0.00 2.0±0.03a 1.0±0.37 d 1.1±0.44 abc 8.9±0.19 ab 14.5±4.96 g 6.2±0.68 1.6±0.15 M2 169.0±0.00 1.9±0.04b 1.7±0.55 bcd 1.3±0.42 abc 10.2±0.62 a 18.3±1.82 fg 6.2±1.25 1.4±0.14 M3 188.0±0.00 1.7±0.06c 1.8±0.55 bcd 0.7±0.41 c 10.4±0.36 a 30.4±6.46 defg 7.1±0.70 1.2±0.03 p (genotype x maturity) - <0.0001 <0.0001 0.0005 <0.001 0.0002 ns ns 524 21 525 Figures 526 527 Figure 1. Visual appearance of peach fruit from the different genetic classes in the 528 progeny from the ‘Belbinette x Nectalady’ F1 population at the maturity stage 2 (M2). 22 529 530 Figure 2. Fruit weight (g), at different maturity stages (M1, M2 and M3), of peach fruit 531 from the different genetic classes in the progeny from the ‘Belbinette x Nectalady’ F1 532 population. Data shown are means ± Std. dev. Means with the same letter are not 533 significantly different according to analysis of variance (ANOVA) and Tukey’s HSD test 534 (p < 0.05). 535 536 537 23 538 539 Figure 3. Changes in fruit diameter (mm; A) and IAD values (B) from fruit set to harvest 540 in the four genetic classes of the progeny from the ‘Belbinette x Nectalady’ F1 population. 541 Data shown are means ± Std. dev. LSD values for the interaction genetic class x DAFB 542 are 10.1 and 0.69 for (A) and (B), respectively. (C) Flesh firmness at 190 days after full 543 bloom (DAFB) for the four genetic classes of the progeny from the ‘Belbinette x 544 Nectalady’ F1 population. Values represent the mean mean ± Std. dev and means with 545 the same letter are not significantly different according to analysis of variance (ANOVA) 546 and Tukey’s HSD test (p < 0.05). 24 547 548 549 Figure 4. Changes in flesh firmness of peach fruit from the different genetic classes in 550 the progeny from the ‘Belbinette x Nectalady’ F1 population, and harvested at different 551 maturity stages (M1, M2 and M3), during shelf life at 20ºC (70% RH). Data represents 552 the mean ± Std. dev. 553 554 555 556 25 557 558 559 Figure 5. Ethylene production of early harvested peach fruit (M1) from the different 560 genetic classes in the progeny from the ‘Belbinette x Nectalady’ F1 population during 561 shelf life. Data represents the mean ± Std. dev. LSD value for the interaction genotype x 562 days at 20ºC is 0.87. 563 26 564 565 Figure 6. ACC oxidase activity, at different maturity stages (M1, M2 and M3), of peach 566 fruit from the different genetic classes in the progeny from the ‘Belbinette x Nectalady’ 567 F1 population. Data shown are means ± Std. dev. Means with the same letter are not 568 significantly different according to analysis of variance (ANOVA) and Tukey’s HSD test 569 (p < 0.05). 570 571 27 572 573 574 Figure 7. Visualization of Spearman’s rank correlation matrix (significance level p<0.05) 575 between the different biochemical and quality traits analysed in the progeny from the 576 ‘Belbinette x Nectalady’ F1 population. 28