Running head : Ethylene and apple fruitlet abscission

37 Apple (Malus x domestica) is increasingly being considered an interesting model species for studying 38 early fruit development, during which an extremely relevant phenomenon, i.e. fruitlet abscission, may 39 occur as a response to both endogenous and/or exogenous cues. Several studies were carried out 40 shedding light on the main physiological and molecular events leading to the selective release of 41 lateral fruitlets within a corymb, either occurring naturally or as a result of a thinning treatment. 42 Several studies pointed out a clear association between a rise of ethylene biosynthetic levels in the 43 fruitlet and its tendency to abscise. A direct mechanistic link, however, has not yet been established 44 between this gaseous hormone and the generation of the abscission signal within the fruit. In the 45 present research, the role of ethylene during the very early stages of abscission induction was 46 investigated in fruitlet populations with different abscission potentials due either to the natural 47 correlative inhibitions determining the so called “physiological fruit drop” or to a well tested thinning 48 treatment performed with the cytokinin 6-benzyladenine. A crucial role was ascribed to the ratio 49 between the ethylene produced by the cortex and expression of ethylene receptor genes in the seed. 50 This ratio would determine the final probability to abscise. A working model has been proposed 51 consistently with the differential distribution of four receptor transcripts within the seed, which 52 resembles a spatially progressive cell-specific immune-like mechanism evolved by apple to protect the 53 embryo from harmful ethylene. 54 55 www.plantphysiol.org on October 31, 2017 Published by Downloaded from Copyright © 2015 American Society of Plant Biologists. All rights reserved. Introduction 56 Vegetative and reproductive organs that are no longer needed, infected, damaged or senescent may 57 shed from the main plant body following a specific sequence of highly regulated events, i.e. abscission 58 (González-Carranza et al., 1998; Taylor and Whitelaw, 2001; Estornell et al., 2013). This process relies 59 upon a complex regulatory network, activated by the abscising organ, that leads to the activation of the 60 abscission zones (AZs) (Addicott, 1982; Taylor and Whitelaw, 2001; González-Carranza et al., 2002; 61 2007; Lashbrook and Cai, 2008; Zanchin et al., 1995; del Campillo and Bennett, 1996). 62 Fruit trees have set up a developmental strategy aimed at controlling fruit load according to nutrient 63 availability, thus making efficient use of resources. This strategy is achieved through the so-called 64 ‘physiological drop’ or ‘June drop’, involving the abscission of young developing fruits mainly due to a 65 ‘correlative dominance effect’ of adjacent fruit and/or nearby shoots (Bangerth, 2000). This process 66 differs from senescence-driven abscission, which consists in a developmentally programmed process 67 occurring at or after ripening. The correlative effect is mainly transduced in nutritional terms (i.e. 68 sugar starvation), thus generating intra-organ metabolic rearrangements and signals leading to AZ 69 activation. The currently accepted model for correlatively-driven abscission implies that auxin, 70 produced by the subtending organ and transported through the AZ, can reduce its sensitivity to 71 ethylene and delay its activation. Once that the auxin flow through the AZ decreases or its transport is 72 depolarized, the AZ becomes sensitive to ethylene and is activated (Dhanalakshmi et al., 2003; Blanusa 73 et al., 2005; Meir et al., 2006, 2010). This ‘downstream model’, however, describes only the events 74 occurring when abscission is induced, whereas it does not tell anything about how and why the auxin 75 flow changes, i.e. the origin of the abscission signal. 76 Within this context, apple revealed to be a good model system to study the generation of the 77 abscission signal in young developing fruits (Botton et al., 2011; Eccher et al., 2013; Eccher et al., 78 2014), as it develops corymbs with a clear gradient of correlative dominance related to the position 79 and size of the fruit. This dominance is naturally responsible for the physiological fruit drop, which is, 80 however, not able to guarantee high fruit quality and, on the other hand, a suitable return to flowering 81 in the following season. Fortunately, this dominance can be magnified by means of chemical 82 treatments, thus inducing a significantly higher rate of fruitlet abscission (Bangerth, 2000; Greene et 83 al., 1992). Benzyladenine (BA) is a widely known chemical thinner (Bangerth, 2000; Buban, 2000), 84 which can induce abscission in a controlled, inducible and selective way through the enhancement of 85 correlative inhibitions (Dal Cin et al., 2009a; Dal Cin et al., 2009b; Dal Cin et al., 2007; Dal Cin et al., 86 2005; Botton et al., 2011). A ‘model experiment’ with BA can provide as a result different populations 87 of fruitlets with clearly predictable abscission potentials, that are: i) small lateral fruitlets that abscise 88 spontaneously even upon the thinning treatment (L1), ii) big lateral fruitlets that would naturally 89 www.plantphysiol.org on October 31, 2017 Published by Downloaded from Copyright © 2015 American Society of Plant Biologists. All rights reserved. 6 persist (L3), iii) big lateral fruitlets that abscise upon the BA treatment (LB3), and iv) big central 90 fruitlets that would persist (C3) also upon the thinning treatment (CB3). This experimental system 91 allowed to shed light on the signalling pathways mediating the induction of apple fruitlet abscission, 92 which can be summarized in a hypothetical model describing the cortex as the primary sensor of the 93 nutritional stress occurring within the tree. In this tissue, the molecular mechanisms linking nutrient 94 starvation to hormone signalling are consequently activated, mainly involving abscisic acid (Eccher et 95 al., 2013) and ethylene, whose levels increase peaking at 2 and 4-5 days after abscission induction, 96 respectively. The seeds seem to perceive the situation at a later stage and with a less pronounced 97 transcriptional and metabolic reprogramming leading to its abortion (Botton et al., 2011). Failure of 98 embryo development is the critical irreversible step of abscission induction (Goldschmidt and Koch, 99 1996; Yuan and Greene, 2000), which is someway triggered by the primary reaction of the cortex. 100 Since the seed is the primary source of auxin within the fruit, its abortion decreases the supply of this 101 hormone to the AZ, leading to its activation according to the same model proposed for the leaf by 102 Sakamoto et al. (2008). 103 Despite its robustness, confirmed by physiological, metabolic, and transcriptional data, this model still 104 lacks a critical point: how is the reaction of the seed triggered? In other words, how does the cortex 105 communicate the critical situation to the seed? Altough the involvement of the gaseous hormone 106 ethylene in apple fruitlet abscission has already been pointed out and discussed in several studies (Dal 107 Cin et al., 2005, 2007, 2007; Botton et al., 2011), a direct mechanistic link with the physiology of this 108 process is still missing. In the present study, the relationship between ethylene and fruitlet shedding is 109 revisited under a new perspective, pointing out a role for ethylene perception in the seed tissues as a 110 possible main factor involved in transducing the signal generated in the cortex. Experimental data give 111 also some important indications about the molecular and cellular mechanisms which most likely 112 determine the final destiny of the fruit, discriminating from fruitlets destined to persist from those 113 designated to abscise. 114 115 www.plantphysiol.org on October 31, 2017 Published by Downloaded from Copyright © 2015 American Society of Plant Biologists. All rights reserved.