Discussion
Here we document the occurrence of p-type ABA dynamics in an angiosperm species outside of Fabaceae, the evergreen Lauraceae species U. californica . In U. californica foliage levels of ABA during long term drought showed a highly similar pattern as Ψldeclined to that of the classical model species that initially characterized p-type responses of foliage ABA levels during long term drought, the conifer species C. rhomboidea (Brodribb & McAdam, 2013). This result, coupled with previous reports of p-type ABA dynamics in species from two genera in Fabaceae (Nolan et al. , 2017; Yaoet al. , 2021a; Yao et al. , 2021b), all adapted to seasonally dry or arid environments and with highly embolism resistant xylem (no leaf death was seen in our drought experiment in either species experienced a Ψl of -6 MPa), suggests that the evolution of the p-type ABA response to long-term drought is linked to the evolution of highly resistant xylem, and is not just a conifer and Fabaceae specific phenomenon. Highly resistant xylem has evolved frequently across angiosperm species (McAdam & Cardoso, 2018) suggesting that if the two are linked, highly embolism resistant xylem and the p-type ABA dynamic (Brodribb et al. , 2014), this ABA response to drought could be quite commonly observed across angiosperms. Taken together our results demonstrate the occurrence of a p-type ABA response now in two highly divergent angiosperm families, the early diverging Lauraceae and the Fabaceae (Nolan et al. , 2017; Yaoet al. , 2021a; Yao et al. , 2021b), future studies are needed to investigate whether this response is common across angiosperms with highly resistant xylem. The absence of high levels of ABA under long term drought in angiosperm implies that, like in conifers, the stomata of all p-type species may be closed passively by low Ψl under long term drought (Brodribb & McAdam, 2013; McAdam & Brodribb, 2015). This is a controversial hypothesis for angiosperm stomatal biology (Franks, 2013; Merilo et al. , 2017), especially given that angiosperm ABA biosynthetic and signalling mutants have stomata that are insensitive to changes in leaf water status (McAdam et al. , 2016; Cernusak et al. , 2019; Brodribbet al. , 2021). It has been suggested that passive regulation of stomatal aperture in response to changes in leaf water status is absent from this group of land plants (McAdam & Sussmilch, 2020). Recently in characterizing p-type ABA dynamics in Caragana , Yao et al.(2021a) suggested that ethylene might be closing stomata during drought and on recovery from drought when ABA levels are low but stomata are not yet open to maximum apertures. Further work is required to address whether the stomata of p-type angiosperms are closed at low Ψl passively via low cell turgor pressure, or via an alternative metabolic signal such as ethylene (Hasan et al. , 2021).
Similarities in the dynamics of ABA, catabolite and conjugate levels, as well as the inhibition of dehydration-induced ABA biosynthesis during drought between the p-type angiosperm and conifer species suggests that there is a shared mechanism driving the decline in ABA levels under long term drought stress in seed plants. By rapidly dehydrating branches on the bench we could assess the ability of leaf tissue to rapidly synthesize ABA in response to dehydration (Wright & Hiron, 1969). This technique allowed us to study ABA biosynthetic capacity without the need to quantify the expression of key ABA biosynthetic genes, which can be costly and time consuming and requires a detailed understanding of the homologues of key genes (Sussmilch et al. , 2019). Despite considerable investment in genome sequencing in the past decade (Kresset al. , 2022), we still do not have for any species with highly resistant xylem to embolism. We show that ABA biosynthesis is highly active and rapid in unstressed branches that are rapidly dehydrated on the bench, like numerous early studies into ABA biosynthesis in plants (Wright & Hiron, 1969; Pierce & Raschke, 1980; Davies et al. , 1981). This ability is eliminated in branches that are taken from plants when ABA levels are low under long-term drought and rehydrated overnight on the bench before dehydration (Figure 4). A key limitation to this method is that it relies on quantifying a change in ABA level which is only possible if there are low levels of the hormone at the start of the experiment, hence we are unable to use this method to determine precisely when dehydration induced ABA biosynthesis was deactivated during drought, because for much of a drought ABA levels were high (Figure 1). We could speculate that the point at which ABA biosynthesis was deactivated corresponded to the Ψl close to when ABA levels peaked during drought. Coincidently, we found that the Ψl at which peak ABA levels occurred was very similar to the Ψtlp (Figure 8). Work is required to confirm that the expression of key ABA biosynthetic genes such as Nine-cis-epoxycarotenoid deoxygenase (NCED ) genes are no longer upregulated on rapid dehydration in branches that do not rapidly synthesize ABA when dehydrated on the bench (Hasan et al. , 2021). As cells loose turgor or lose volume ABA biosynthesis is triggered (Pierce & Raschke, 1980; Davies et al. , 1981; Creelman & Mullet, 1991; McAdam & Brodribb, 2016), yet there has been very little work conducted on plant tissue that has been dehydrated to a Ψl more negative than Ψtlp.. Consequently, the potential causes of ABA biosynthesis cessation at a Ψl more negative than Ψtlp are highly speculative. Explanations range from an absent trigger for NCED expression once membrane pressure on the cell wall ceases (Bacete et al. , 2022); cellular processes such as transcription and translation of RNA ceasing at a Ψl more negative than Ψtlp (Dhindsa & Cleland, 1975); or carotenoid precursors for ABA biosynthesis, often sorted in chloroplasts, may be depleted because of increased in the de-epoxidation state of the xanthophyll cycle (Munné-Bosch & Alegre, 2000), reducing availability for conversion to ABA. Munné-Bosch and Alegre (2000) found in the extremely drought resistant Rosmarinus officinalis , in which 50% of the xylem experiences embolism at a Ψl at -8 MPa (Brodribb et al. , 2017) during a severe summer drought the levels of ABA carotenoid precursors violaxanthin and neoxanthin declined by more than 85% maximum levels. We show that the ability to recover ABA biosynthesis can occur, with ABA levels accumulating, but not to levels in never-before stressed branches 4 days after rewatering. This recovery might reflect the rapid recovery of carotenoid levels upon rehydration (Munné-Bosch & Alegre, 2000). Future studies in herbaceous species at a Ψl more negative than Ψtlp could help elucidate whether the mechanism driving ABA biosynthesis cessation after Ψtlpis universal to all plants. These experiments can only be done in strictly controlled environments since a slight decrease in Ψl once turgor is lost could trigger lethal embolism as the two occur at very similar Ψ in herbaceous plants (Skelton et al. , 2017). There is a hypothesis that ABA biosynthesis is triggered by a loss of cell volume and not a change in cell turgor (Sack et al. , 2018), our results refute the hypothesis ABA biosynthesis ceases in p-type species yet cell volume presumably continues to decline at a fairly linear rate during drought. This corroborates recent data that indicates an intact cell wall is essential for the biosynthesis of ABA when cells are exposed to solutions of high osmotic potential (Bacete et al. , 2022).
Once ABA biosynthesis ceases at a Ψl more negative than Ψtlp our results suggest that continual conjugation of ABA, presumably into ABA-GE, which is the primary, if not only, conjugate for ABA (Milborrow, 1970), and not catabolism of the remaining is the main driver for a decline in foliage ABA levels once ABA biosynthesis ceases. It is believed that phloem flux is greatly reduced during drought (Hartmann et al. , 2013; Sevanto, 2014), and our results from a girdling experiment in C. rhomboidea demonstrate that girdling the phloem does not change the p-type ABA dynamic during drought or the accumulation of ABA conjugates. By quantifying the levels of PA during long term drought, we are able to categorically rule out catabolism was the primary driver of a decrease in ABA levels after Ψtlp. Our results demonstrate that catabolism of ABA into PA did not significantly increase after peak ABA levels. Interestingly, the dynamic of PA levels and ABA levels during drought suggests that PA levels and the rate of ABA catabolism could be a simple function of current ABA level.