Defence-related actin remodeling can be separated from defence-related cell-death
Actin remodelling has been recognised as early event heralding ensuing programmed cell death (reviewed in Gourlay & Ayscough, 2005; Franklin-Tong & Gourlay, 2008; Smertenko & Franklin-Tong, 2011). In grapevine cells, as well, the induction of programmed cell death, by either the bacterial elicitor harpin (Chang et al. , 2015; Qiao et al. , 2010), the stilbene aglycone resveratrol (Chang et al. , 2011), or the oxylipin derivative 3-cis -hexenal (Akaberi et al. , 2018) were preceded by actin remodelling, involving a depletion of the fine subcortical meshwork and massive bundling of transvacuolar actin cables. Thus, in these cases, actin remodelling was tightly correlated with programmed cell death. A correlation of two events is a necessary condition for claiming a causal relation between these two events. However, even if this correlation is tight, it cannot be considered as sufficient for causality.
In fact, actin remodelling of a similar type can be observed without any correlation with cell death. For instance, depletion of cortical actin and condensation of actin cables have been also observed in response to auxin depletion (Waller et al. , 2002), or in response to activation of the plant photoreceptor phytochrome (Waller & Nick, 1997), both events that occur in the absence of any cell death.
The same conclusion is reached by our observation that aluminium did not cause any significant increase of mortality (Suppl. Figure S4 ). Is actin remodelling just a side phenomenon, then? This seems not to be the case either, because in the same cell system (V. rupestris ), the induction of programmed cell death by harpin had been shown to be significantly mitigated by pretreatment with Latrunculin B (Chang et al. , 2015), which provides evidence that actin is necessary to activate cell death.
Thus, while actin remodelling is, usually, accompanied by defence-related cell death, and while suppression of actin remodelling will impair defence-related cell death, this apparently tight link is uncoupled in case of aluminium treatment. The most straightforward explanation would be that actin remodelling has to coincide with a second signal to activate programmed cell death. This second signal cannot be the apoplastic oxidative burst generated by RboH, because aluminium does trigger this response (Achary et al. , 2012). Moreover, we observed that actin remodelling in response to aluminium can be suppressed by diphenylene iodonium, indicative of RboH activity as transducing event (Figure 2 ). While in case of a real pathogen attack apoplastic oxidative burst is accompanied by calcium influx, this does not seem to be the case for aluminium (Suppl. Figure S3 ). A parsimonious hypothesis would therefore assume that calcium influx as a second signal is required to activate programmed cell death.