The peach RGF/GLV signalling peptide pCTG134 is involved in a regulatory circuit that sustains auxin and ethylene actions

Peach is a climacteric species whose ripening is regulated by the plant hormone ethylene. A crosstalk mechanism with auxin is necessary to support climacteric ethylene synthesis. The homeostasis control of auxin is regulated also by the activity of peptide hormones (PHs), acting both as short and long distant ligands. In this work, we investigated the role of CTG134, a peach gene encoding a GOLVEN-like PH isolated in mesocarp at the onset of ripening. In peach fruit, CTG134 was expressed during the climacteric transition and its mRNA level was induced by auxin and 1-methylcyclopropene (1-MCP) treatments, whereas it was minimally affected by ethylene. To better elucidate its function, CTG134 was overexpressed in Arabidopsis and tobacco, which showed abnormal root hair growth, similar to wild-type plants treated with a synthetic form of the peptide. Molecular surveys demonstrated an impaired hormonal crosstalk, resulting in a re-modulated expression of a set of genes involved in both ethylene and auxin domains. In addition, the promoter of pCTG134 fused with GUS reporter highlighted gene activity in plant organs in which the auxin-ethylene interplay is known to occur. These data support the role of pCTG134 as mediator in an auxin-ethylene regulatory circuit. Highlight The role of the peach RGF/GLV peptide during root hair formation in Arabidopsis and tobacco supports its involvement in a cross-hormonal auxin-ethylene regulatory circuit.

In Angiosperms, fruits, besides providing essential and beneficial compounds to 80 the human diet, protect the seeds enabling their dispersion at the end of the 81 ripening phase. These organs originally develop from the ovary, with the 82 possible contribution of other flower parts. According to the physiological 83 regulation of ripening, fleshy fruits can be distinguished into climacteric (such as 84 tomato, peach and apple) and non climacteric (such as strawberry, grape and 85 citrus), depending on the presence of a burst in the production of the plant 86 hormone ethylene accompanied by a respiratory increase occurring at the late 87 stage of fruit ripening (Liu et al., 2015). Ethylene is produced through the 88 sequential activation of two biosynthetic systems (McMurchie et al., 1972;89 Oetiker et al., 1997). The auto-inhibitory system 1, found in both climacteric and 90 non-climacteric fruit, maintains a basal level of ethylene during the vegetative 91 growth of plants as well as in wound and stress response. The autocatalytic 92 system 2 produces, instead, a much larger amount of the hormone, typical of 93 climacteric fruits in full-ripening phase. In the tomato model, the switch between 94 the two systems is based upon the differential expression of 1-amino-95 cyclopropane-1-carboxylic acid (ACC) synthase (ACS) and ACC oxidase (ACO) 96 genes during the ripening process (Barry et al., 2000). The transition from the 97 system 1 to system 2, which represents a crucial point in the ripening process of 98 climacteric species (Cara and Giovannoni, 2008), is also regulated by genetic 99 (Vrebalov et al., 2002;Manning et al., 2006) and epigenetic factors (Zhong et 100 al., 2013), together with the surrounding environmental effect and interplay with 101 other plant hormones (Klee and Giovannoni, 2011). 102 Emerging evidence suggests that relative functions of plant hormones are not 103 restricted to a particular stage only. A complex network of more than one plant 104 hormone is therefore involved in controlling various aspects of fruit development 105 (Kumar et al., 2014). The knowledge of the hormonal network and crosstalk 106 relationship between different hormones during the stages of the fruit life cycle 107 is still far from being complete (Kumar et al., 2014) and relies almost entirely on 108 model species (Arabidopsis and tomato). The action of the phytohormones 109 depends not only on the cellular context, but also on the relationship 110 established among different hormones. To date, the hormonal crosstalk has 111 been mainly investigated in Arabidopsis, which shed light, among others, on the 112 crosstalk between auxin and ethylene (Poel et al., 2015). The first and most 113 evident effect of the interaction between these two hormones is on the 114 regulation of the root morphogenesis process. Indeed, in this organ it has been 115 demonstrated that root hair formation, elongation (Pitts et al., 1998;Dolan, 116 2001) and differentiation as well as the development of lateral roots are 117 regulated by the interplay occurring between auxin and ethylene (Zhang et al., 118 2016). On the other hand, ethylene can also modify the auxin patterning by 119 modulating IAA transport (Prayitno et al., 2006). Cellular and genetic evidences 120 have shown a physiological connection between hormones and peptide 121 hormones (PHs). ROOT GROWTH FACTOR/GOLVEN/CLE-Like 122 (RGF/GLV/CLEL) peptides can in fact alter auxin gradients by changing the 123 turnover of IAA carriers (Whitford et al., 2012). Despite the importance of this 124 regulatory mechanism, the biology of PHs is still in its infancy, especially in non-125 model but agronomically relevant species. 126 Auxin and ethylene have been described to interact at the level of ethylene 127 biosynthesis (Abel et al., 1995) not only in Arabidopsis roots but also during the 128 ripening of different fruit species, such as tomato (Abel et al., 1996), peach 129 (Trainotti et al., 2007) and apple (Shin et al., 2015). Although the molecular 130 mechanisms of the interplay between auxin and ethylene during fruit ripening 131 are still unknown, recent data suggest that PHs (Matsubayashi, 2014;132 Tavormina et al., 2015) could be the crossroads between the two hormones in 133 peach (Tadiello et al., 2016). One PH in particular, namely CTG134 GLV-like, 134 was identified through a comprehensive survey carried out with the µPeach1.0 135 (Tadiello et al. 2016). This gene was expressed at the transition step between 136 the preclimacteric and the climacteric stage. Moreover, while CTG134 was 137 induced by exogenous treatment of 1-methylcyclopropene (1-MCP), an 138 ethylene competitor largely used to delay the normal physiological ripening 139 progression (Watkins, 2006), its expression was also totally repressed in ripe 140 fruit of stony hard, a peach mutant showing impairment both in ethylene 141 production and cell wall metabolism (Pan et al., 2015). 142 In this work the peptide pCTG134, isolated from peach, was functionally 143 validated in Arabidopsis and tobacco, providing new evidence about its role as 144 a major regulator in the auxin-ethylene crosstalk. 145 146

Plant materials 148
Peach fruit s were collected from cv. 'Redhaven' (RH) as described in Tadiello 149 et al. (2016). The heterologous CTG134 overexpression was carried out in 150 Arabidopsis and tobacco plants. Seeds of Arabidopsis thaliana Columbia 151 accession (Col-0) were surface-sterilized, stratified overnight at 4°C and 152 germinated on plant growth medium (Murashige and Skoog, 1962)

Hormone treatments on Redhaven fruit 157
The auxin treatment was performed by dipping the whole fruit in 1-naphthalene 158 acetic acid [NAA, 2 mmol L -1 added with Silwet L-77 (200 µL L -1 ) as surfactant] 159 for 15 min; thereafter, fruit were sprayed with the NAA solution every 12 h over 160 a period of 48 h (NAA omitted in the mock control). The ethylene treatment was 161 instead carried out as previously described in (Tadiello et al., 2016). 162

1-MCP treatments on Stark Red Gold fruit 163
Treatment of cv. "Stark Red Gold" (SRG) peach fruits with 1-MCP was carried 164 out as described in Tadiello et al. (2016). 165

RNA extraction and expression analyses by quantitative Real time PCR 166 (qRT-PCR) 167
Peach RNA was prepared from a frozen powder obtained by grinding mesocarp 168 sectors from at least four different fruits. From four grams of this powder, total 169 RNA was extracted following a protocol previously described (Chang et al., 170 1993). Arabidopsis RNA was extracted from wild type and 35S::CTG134 mutant 171 seedlings, using the LiCl method (Verwoerd et al., 1989). Expression analyses 172 were performed using Power SYBR Green PCR Master Mix (Applied 173 Biosystems). Normalization was performed using UBIQUITIN10 (UBI10) and 174

ACTIN8
as internal standards for Arabidopsis and 175 Ppa009483m/Prupe.8G137600 for peach (Primers are listed in Table S1). qRT-176 PCR was performed and the obtained data analysed as previously described 177 (Tadiello et al., 2016). 178

In-situ hybridizations 179
Prunus persica S3II and S4 fruits were fixed and embedded in 4% 180 paraformaldehyde. A CTG134 specific probe was amplified by PCR from S3II 181 and S4 fruit cDNAs (primers listed in Table S1) and further cloned in pGEM T-182 easy vector (Promega). The CTG134 transformed vector was further used as 183 template for the creation of sense and antisense probes by an in-vitro 184 transcription performed with SP6 and T7 polymerases. Sections of plant tissue 185 were probed with dioxigenin-labelled antisense RNA-probe as previously 186 described (Brambilla et al., 2007) and observed with a Zeiss Axiophot D1 light 187 microscope (http://www.zeiss.com). 188

pPR97-proCTG134:GUS construct design 189
To assess the CTG134 promoter activity, a fragment of 2679 bp located activity. The promoter was tested by cloning the upstream sequence and a GUS 200 reporter gene interrupted by a plant intron (Vancanneyt et al., 1990). To make 201 easier the cloning, a CC_rfA gateway cassette was inserted (SmaI) upstream of 202 the reporter gene and the antibiotic kanamycin was used to select resistant 203 successfully transformed plants. 204

pGreen-AmpR-KanNos-35S:CTG134 construct design 205
The CTG134 coding sequence (524 bp) was amplified by PCR from Prunus 206 persica (cv Red Haven, S4I development stage) cDNA and subsequently 207 cloned into the pCR8/GW/TOPO TA Cloning vector (Invitrogen, Carlsbad, CA, 208 USA). The CTG134 CDS was further inserted into a pGreen-derived vector 209 (Hellens et al., 2000) with the Gateway cloning system (LR Clonase II -210 Invitrogen, Carlsbad, CA, USA). The pGreen-derived vector was modified to 211 confer resistance to both kanamycin and ampicillin. Moreover, a CC_rfA 212 gateway cassette was inserted downstream of the 35S promoter in the EcoRV 213 site. As before, the selection of plants was carried out with kanamycin 214 ( Supplementary Fig. S1B). 215

Arabidopsis thaliana and tobacco transformation 216
Single PCR-positive Agrobacterium GV3101 colonies were used to grow liquid  In-vitro grown N. tabacum SNN plants were instead transformed following the 231 protocol reported by (Fisher and Guiltinan, 1995). As for Arabidopsis, plants 232 were screened for the presence of the transgene with PCR on genomic DNA 233 using specific primer pairs. 234

Regulation of CTG134 expression 261
Expression of CTG134 was assessed in peach mesocarp during the onset of 262 fruit ripening (i.e. at early stage 4 -S4I - Fig. 1A). CTG134 mRNA accumulated 263 in preclimacteric fruit (i.e. S3II) after auxin treatment, while exogenous ethylene 264 had no effect (Fig. 1B). Moreover, treatment with the ethylene inhibitor 1-MCP 265 induced CTG134 transcription at stages before (cl 0) and coincident (cl 1) with 266 the full climacteric (Fig. 1C). The peach mesocarp at ripening is mainly made up 267 of parenchymal cells and vascular tissue (Zanchin et al., 1994). To localize the 268 types of cells expressing CTG134 at ripening, in-situ hybridization experiments 269 were carried out with mesocarp sections prepared by peach fruit in S4 stage. 270 The CTG134 mRNA was localized in vascular bundles (Supplementary Fig.  271 S2C), most likely in the phloem or parenchymal cells (Fig. 1D). 272 Since peach is a recalcitrant species to transform, proCTG134:GUS lines were 273 generated in both tobacco and Arabidopsis model species. In tobacco, a slight 274 but evident GUS staining was detected in the apical meristem (RAM) of in-vitro 275 grown lateral roots ( Fig. 2A). Moreover, a dark staining was visible in lateral root 276 emergence ( Fig. 2B) as well as in leaf, mainly associated, but not limited to, the 277 vascular tissue (Fig. 2C). In the stem of one-week-old plantlets, GUS 278 expression was localized in phloem of cell layers closed to the cambium (Fig. 279 2D). GUS expression was also tested in reproductive organs, where it was 280 detected in the tips of both young sepals and petals (not shown) and in 281 capsules at the level of the dehiscence zone (Fig. 2E). The inner part of the fruit 282 was the part more significantly stained ( Figures 2F and 2G), with the highest 283 expression in the placenta (Fig. 2G). On the contrary, in all the transgenic lines 284 investigated in this study, the GUS colouration was never observed in ovule. In

Hormonal regulation of CTG134 in tobacco 305
To test whether the auxin responsiveness was due to the promoter regulatory 306 region, one-week old tobacco seedlings of line #2 were exposed to increasing 307 concentrations of IAA. The CTG134 promoter was responsive to IAA already at 308 0.5µM, with an activity pattern proportional to the hormone concentrations. The 309 system reached saturation at 50µM (Fig. 4A). The IAA induction kinetic was 310 assessed over a time course of 20 hours on tobacco seedlings of line #2 311 treated with 10µM IAA. An initial slight induction in both control and treated 312 samples was observed already after 30 minutes, after which the GUS activity 313 remained at a basal level in the control, while in the IAA treated samples a 314 significant burst was observed after 3 hours after the treatment (Fig. 4B). Since 315 in peach fruit the expression of CTG134 was insensitive to ethylene and 316 induced by 1-MCP ( Fig. 2B and 2C), the promoter responsiveness was tested 317 by treating ten-day-old tobacco seedlings for sixteen hours with ethylene (10µL 318 L -1 ), IAA (10μM) and 1-MCP (1µL L -1 ). 1-MCP induced the reporter activity 319 similarly to auxin (Fig. 4C), while treatment with ethylene did not change the 320 expression of the GUS reporter gene. 321

Over-expression of CTG134 in tobacco 322
To functionally investigate the role of the peptide CTG134 peptide, its full-length with respect to control wild type plants (Fig. 5B). The effect on root development 334 was also evident during adventitious roots formation in in-vitro plants 335 (Supplementary Fig. S3A and B). Indeed, root primordia emerged earlier in 336 35S:CTG134 scions than in wild type, although the root growth was slower, 337 resulting at the end in shorter roots ( Supplementary Fig. S3C). Within the 338 hypothesis of the auxin-ethylene crosstalk, the putative mediating role of 339 wrapping the root body (Fig. 5F). Subsequently, a Scanning Electron 351 Microscopy (SEM) analysis disclosed that the previously observed root hair 352 phenotype was due to an increase of their density in the 35S:CTG134 lines (Fig.  353 5H) with regards to control (Fig. 5G). Indeed, most of the root epidermal cells of 354 35S:CTG134 seedlings developed root hairs, while in WT trichoblasts were 355 arranged in alternating files with atrichoblasts along the root surface. 356 Since the CTG134 sequence was originally isolated from peach fruit, and

Over-expression of CTG134 in Arabidopsis 366
Similarly to tobacco, the same construct was further employed to transform 367 Arabidopsis. T2 CTG134 overexpressing lines were easily identified for their 368 root phenotype when grown on horizontal plates. The primary root of five-day-369 old 35S:CTG134 seedlings had indeed longer hairs than WT ones (Fig. 6A). 370 Moreover, root hairs developed closer to the apex that in WT roots. To quantify 371 the latter effect, the hairless portion of the root was about half (ANOVA, F = 372 101.1, df = 23, p < 0.001) of that in the WT (Fig. 6B). As regards to root hair 373 length, being not uniform along the root and clearly depending on age, sizes 374 were taken at given distances from the root-stem transition zone and in a region 375 of the tip that was determined to be, based on growth rate, four-day old. Both 376 measures clearly indicated that the root hairs in the overexpressing lines were 377 longer (ANOVA, F = 95.07, df = 342, p < 0.001; ANOVA, F = 98.31, df = 342, p 378 < 0.001, respectively) than wild type (Fig. 6C). Members of the RGF/GLV family 379 in Arabidopsis are known to induce developmental defects in roots when over-380 expressing seedlings were grown on tilted plates, as reported by (Whitford et al., 381 2012;Fernandez et al., 2013). Accordingly, in this work Arabidopsis 382 35S:CTG134 seedlings produced roots with larger and more irregular waves 383 than the WT (Fig. 6D). This effect could be phenocopied by the WT when the 384 synthetic CTG134 peptide (pCGT134) was added to the medium, with the 385 sulfated form being more active than the non-sulfated one (Fig. 6D). Albeit the 386 hairless portion of the root was shorter in overexpressing seedlings, the 387 meristematic region of the root was longer. Moreover, both 35S:CTG134 lines 388 and WT seedlings grown in a medium supplemented with pCTG134 had an 389 increase in root meristem size ( Figures 6E and F). The effect on the root 390 meristem size was saturable, as overexpressing lines did not respond to 391 exogenous pCTG134 as the WT (Fig. 6F). 392 The effect of CTG134 overexpression at the transcriptional level was tested on 393 five-day-old seedling roots (Fig. 7). Alteration in root hairs morphology and 394 quantity was accompanied with a reduction of GLABRA2 (GL2) and a slight  (Kieber et al., 1993) was unaffected ( Supplementary Fig. S5). About 407 auxin, both TAA1 and YUC3 and 6 genes involved in the indole-3-pyruvic acid 408 branch of the hormone synthesis pathway (Tivendale et al., 2014) were induced 409 in CTG134 overexpressing seedlings, while AMI1, involved in the indole-3-410 acetamide branch of the pathway, seemed unaffected (Figures 7 and S5). Free 411 auxin levels depend not only on hormone synthesis but also on its release from 412 storage compartments and transport. The expression of IAR3, a gene encoding 413 an IAA-Ala hydrolase (Davies et al., 1999), decreased in CTG134 414 overexpressing plants, while PIN2, encoding an auxin efflux carrier (Müller et al., 415 1998) was induced (Fig. 7). 416

Discussion 436
Peptide hormones participate in both proximal and distal cell-to-cell 437 communication processes necessary during growth as well as to cope with 438 biotic and abiotic stimuli (reviewed in (Matsubayashi, 2014;Tavormina et al., 439 2015;Wang et al., 2016). Despite the growing interest in peptide hormones, 440 their possible role during fleshy fruit ripening remains almost unexplored (Zhang 441 et al., 2014). In peach fruit, gene expression profiling suggested that CTG134, 442 encoding a peptide belonging to the RGF/GLV family, could be involved in the 443 crosstalk between auxin and ethylene occurring at the onset of fruit ripening 444 (Tadiello et al., 2016). 445

CTG134 expression is ripening specific and affected by auxin and 446 ethylene perception 447
Extensive RNA profiling confirmed that CTG134 is expressed almost exclusively 448 at the onset of ripening, during the transition stage from system 1 to 2 (Fig. 1), 449 as initially suggested by Tadiello et al. (2016). 450 Considering the difficulties typical of Prunus species during the in vitro 451 regeneration phase, tobacco and Arabidopsis transgenic lines expressing the 452 GUS reporter gene driven by the CTG134 promoter sequence, were created. 453 The cis-regulatory elements present in the peach CTG134 promoter drive GUS 454 gene expression in cell/tissue types where the crosstalk between auxin and 455 ethylene was described both in tobacco (Fig. 2) and Arabidopsis (Fig. 3). These 456 comprise both cells undergoing separation processes, like abscission, 457 dehiscence zones, lateral root primordia (Roberts et al., 2002;Kumpf et al., 458 2013), cambium associated cells (Love et al., 2009;Sanchez et al., 2012) and 459 placenta cells (De Martinis and Mariani, 1999;Pattison et al., 2015). The 460 specificity of the GUS staining pattern obtained in heterologous systems was 461 validated by in-situ hybridization in peach mesocarp, where CTG134 expression 462 was more abundant in bundle associated cells (Fig. 1d). It is noteworthy that 463 also regulatory regions of tomato (Blume and Grierson, 1997), apple (Atkinson 464 et al., 1998) and peach (Moon and Callahan, 2004) ACO genes drove GUS 465 expression more abundantly in bundle than parenchyma cells of tomato 466 pericarp. Besides spatial regulation, also hormone responsiveness within 467 CTG134 regulatory regions supported the role in the crosstalk between auxin 468 and ethylene ( Fig. 1 and 4). Indeed, both on ripening mesocarp and tobacco 469 seedlings, not only IAA had an inductive effect, probably due to the presence of 470 AREs, but also the altered perception of ethylene (due to 1-MCP treatment) 471 stimulated both CTG134 transcription in ripening fruit and GUS accumulation in 472 tobacco seedlings. In ripening peaches 1-MCP induced auxin synthesis 473 (Tadiello et al., 2016), and this might be the reason of the CTG134 induction. 1-474 MCP treatment might have induced IAA synthesis, and thus GUS expression, 475 also in tobacco seedlings. In roots of Arabidopsis treated with silver (also 476 blocking the perception of ethylene; Negi et al. 2008) the exogenous application 477 of 1-MCP might have altered the distribution of IAA, leading to GUS induction. 478

35S:CTG134 plants show phenotypes related to auxin and ethylene action 479
When CTG134 was permanently overexpressed in tobacco and Arabidopsis 480 plants ( Figures 5 and 6), the most striking effect was related to the length and 481 number of root hairs, mimicking the effect of exogenous treatments with auxin 482 or ethylene (Pitts et al., 1998). Adventitious root formation and elongation in 483 tobacco were also affected, as well as capsule size, further supporting the 484 interplay between auxin and ethylene actions. Besides the well-known effect on 485 root hair number and morphology reported for RGF/GLV/CLEL (Whitford et al., 486 2012;Fernandez et al., 2013) and CLE peptides (Fiers et al., 2005), CTG134 487 had an impact also on tobacco capsule size. In fact, at maturity, tobacco 488 capsules were 16% larger than WT on average, similarly to carnation flowers 489 treated with ethylene (Nichols, 1976). Ethylene synthesis is necessary for 490 normal ovule development which impacts flower size (De Martinis and Mariani, 491 1999). The GUS staining in tobacco placenta and the larger capsules in 492 CTG134 overexpressing plants allow therefore to hypothesize that CTG134 493 may corroborate auxin inductive and ethylene repressive actions during fruit 494 setting (Martínez et al., 2013;Shinozaki et al., 2015). 495

Molecular targets of CTG134 and its role as mediator in the auxin/ethylene 496 crosstalk 497
The Arabidopsis root model was moreover exploited to gain insights into the 498 regulatory circuit associating CTG134 with auxin and ethylene (Figures 6 and 7). 499 The wavy root phenotype and the increase in meristem size were observed in 500 both overexpressing and peptide treated seedlings, confirming previous findings 501 (Matsuzaki et al., 2010;Whitford et al., 2012). The observed increase in the 502 meristem size was also supported by the induced expression of CYCB1;1 (Fig.  503   7), while the down-regulation of GL2 was in agreement with its repressing role 504 in root hair development (Ishida et al., 2008). More interestingly, genes of both 505 auxin and ethylene synthesis, transport and transduction pathways were 506 upregulated in CTG134 overexpressing roots, assigning to this RGF/GLV 507 peptide a role in the auxin/ethylene crosstalk (Stepanova et al., 2007). Although 508 we did not carry out a detailed analysis on the effects caused by the local 509 application of CTG134 peptide (that in Arabidopsis controlled the PIN2 510 abundance in the root meristem by a post-transcriptional mechanism, thus 511 guiding auxin distribution; Whitford et al., 2012), we showed that the 512 heterologous overexpression of the peach CTG134 peptide could be sensed in 513 the portion of the root where receptors initiate the signalling cascade 514 (Shinohara et al., 2016;Ou et al., 2016;Song et al., 2016). As for Peps 515 signalling in Arabidopsis , aequorin-based Ca 2+ measurement 516 assays (Fig. 8) demonstrated the induction by the sulfated peptide CTG134 of a 517 remarkable cytosolic Ca 2+ change, suggesting the likely involvement of Ca 2+ as 518 intracellular messenger in the transduction pathway activated by this signal 519 peptide. The role of Ca 2+ is supported also by the downregulation of several 520 CALCINEURIN B-LIKE PROTEIN (CBL) genes in roots of CTG134 521 overexpressing seedlings, in agreement with the downregulation of a CBL gene 522 in 1-MCP-treated peaches (Tadiello et al., 2016). Sensing the peptide also 523 induced the transcription of key genes of ethylene and auxin biosynthesis 524 pathways and thus, reasonably, the levels of these two hormones, which 525 eventually led to the observed phenotypes. While the response in the ethylene 526 pathway is somewhat straightforward investigating the induction of key genes in 527 its synthesis (ACS2), perception (ETR1) and signal transduction (EIN3), the 528 action on the auxin pathway is more intricate. Indeed, while the increased 529 transcription of TAA1, YUC3 and YUC6 sustains the induction of the two-step 530 IPA pathway, the unchanged levels of AMI1 seemed to exclude the conversion 531 of indole-3-acetamide (IAM) to IAA (Enders and Strader, 2015). Moreover, 532 although only IAR3 was tested, the contribution of conjugated forms of IAA 533 (Sanchez Carranza et al., 2016) seemed negligible in Arabidopsis, while the 534 expression of its peach homolog CTG475 was supposed to participate to the 535 free auxin increase measured before the climacteric production of ethylene in 536 peach (Tadiello et al., 2016), thus complementing the role of PpYUC11 (Pan et 537 al., 2015). However, the induced transcription of PIN genes in overexpressing 538 Arabidopsis seedlings (Fig. 7) and in climacteric peaches (Tadiello et al., 2016) 539 supported a key role of these peptides in regulating auxin distribution (Whitford 540 et al., 2012). 541 The comprehensive expression profiling data carried out in peach (Tadiello et 542 al., 2016) and the knowledge here achieved about CTG134 in tobacco and 543 Arabidopsis provide evidence on the involvement of this RGF/GVL secreted 544 peptide in a regulatory circuit that sustains auxin and ethylene actions. The 545 same circuit, working in both rosids (Arabidopsis) and asterids (tobacco) might 546 have appeared early during evolution of eudicots to participate in the control of 547 root hair development and later it could have been recruited in peach to 548 regulate the switch from system 1 to system 2 ethylene synthesis (Fig. 9). 549 Further research will be necessary to clarify the molecular details by which 550 CTG134 acts to either regulate auxin and ethylene synthesis or modify their 551 distribution and perception, or both. The kinase nature of GLVs receptors 552 (Shinohara et al., 2016;Ou et al., 2016;Song et al., 2016) agrees with the 553 measured Ca 2+ perturbations. 554 The unique mechanism that switches ethylene synthesis from system 1 to 555 system 2 in peach probably relies on the use of a single ACS gene for both 556 kinds of syntheses (Tadiello et al., 2016), thus differing from tomato (Barry et al., 557 2000) and apple (Wang et al., 2009). In these two latter fruits, the expression of 558 LeACS4 and MdACS3 (system 1) is necessary to start LeACS2 and MdACS1 559 transcription (system 2), respectively. During peach ripening, expression of 560 other ACS genes is, if present, several orders of magnitude lower than that of 561 ACS1 (Tadiello et al., 2016). The different amount of ethylene released by 562 system 1 and system 2 could be achieved by modulating system 1 ACS1 563 activity, thus leading to system 2 ACS1 increased transcription. ACS1 belongs 564 to type-1 ACS proteins, which are stabilized by phosphorylation mediated by 565 mitogen-activated protein kinases (MAPKs) (Liu and Zhang, 2004). 566 Phosphorylation cascades have been shown to start upon binding of peptide 567 signals (e.g. IDA) with their receptors (e.g. HAE/HSL2) (Cho et al., 2008). Given 568 the transcriptional regulation of CTG134, the nature of pCTG134 and of the 569 Arabidopsis receptors of its homologous RGF/GLV peptides (Shinohara et al., 570 2016;Ou et al., 2016;Song et al., 2016) and of the ability of pCTG134 to trigger 571 a cytosolic Ca 2+ signal, we hypothesized that the transition of ethylene 572 synthesis from system 1 to system 2 in peach could be controlled by ACS1, 573 whose activity might be therefore modulated through the action of pCTG134.     is barely detectable in RAM (C). In the reproductive part, expression was detected in abscission zones before (F) and after (G) organ shedding.
Expression was detectable also in maturing siliques mainly associated with vascular bundles (H). Scale bar in the panels B, C and F = 500 µm, in A = 200 µm, in D = 100 µm and in E = 1000 µm.     ACT8 was used as reference gene.