Drought and symbiosis – why is abscisic acid necessary for arbuscular mycorrhiza?

Plant hormones are prime targets when addressing the regulation of such an intimate plant–microbe interaction as arbuscular mycorrhiza (AM), which is found in nearly 80% of all plant species. Besides the most important feature of the mutualistic symbiosis – AM fungal provision of mineral nutrients in return for plant carbohydrates – the colonization of a plant root by AM fungi often improves growth and stress tolerance of the whole plant (Linderman, 2000). Alterations in the homeostasis of plant hormones have been implicated in this process and there are a large number of publications showing that the levels of plant hormones such as cytokinin, jasmonate (JA), auxin, auxin-related compounds and abscisic acid (ABA) actually change upon the establishment of AM (Hause et al., 2007; Fig. 1). Although these findings suggest that the beneficial effects of AM are not restricted to an improved nutrition of the plant, in most cases they do not provide clear functional proof regarding the importance of the changes in hormone levels observed. In addition, they only address the hormonal and physiological consequences of an established AM symbiosis. The question of a possible involvement of plant hormones in AM establishment is rarely in their scope. In this issue of New Phytologist, however, Herrera-Medina et al. (pp. 554–564) address exactly this question. By analyzing AM colonization of the tomato mutant sitiens, in which ABA levels are reduced to only 8% of those in wild-type tomato plants, they were able to show that ABA is necessary for the proper formation of arbuscules (the key symbiotic interface of AM) and for a sustained colonization of the plant root. The tomato mutant sitiens exhibits a reduced AM formation, which can be restored by the application of ABA, correlating with increased mycorrhization by treatment of wild-type roots with ABA. In addition, the data presented suggest an antagonistic role for the plant hormone ethylene (ET). Schematic summary of nutritional and hormonal changes upon colonization of plants by arbuscular mycorrhiza (AM) fungi and of hormonal effects on AM colonization. The picture gives a model of a colonized root cortical cell; the cell wall and cytosol have been partially removed. The plant cell nucleus is depicted in the center of the schematic arbuscule. Nutritional changes are given in the upper part of the scheme (the plant receives mineral nutrients and water while providing carbohydrates to the fungus) and the hormonal changes are given in the lower right part. Plants colonized by AM fungi have been described to contain lower levels of abscisic acid (ABA), but increased levels of auxin (and auxin-related compounds such as indolbutyric acid), jasmonate and cytokinin. The lower left part gives the observed effects of plant hormones on establishment of the AM symbiosis. Apart from the positive effect of jasmonate described by Isayenkov et al. (2005), the effects of ABA and ethylene are described by Herrera-Medina et al. (this issue; pp. 554–564). ‘Once the fungus has entered the root our knowledge on the molecular communication is only fragmentary’ Despite the striking discoveries in recent years regarding the molecular communication between plants and AM fungi, it should be stressed that our knowledge of this communication is limited. It has been shown that a group of compounds (strigolactones) exude from the plant root and induce hyphal branching of AM fungi (Akiyama et al., 2005). Moreover, a plant signal-transduction cascade, which is initiated by a receptor-like kinase, clears the way for the entry of AM fungi into the plant root (Parniske, 2004). Interestingly, this signaling cascade is partially shared in the plant–rhizobial and the plant–nematode interactions (Weerasinghe et al., 2005). This initial signaling is followed by the formation of a highly specialized transient intracellular structure, designated the prepenetration apparatus, which is assembled by the epidermal cell with a novel cytoskeletal organization and plays a central role in the elaboration of the apoplastic interface compartment for fungal growth (Genre et al., 2005). However, once the fungus has entered the root, our knowledge on the molecular communication is only fragmentary. The publication by Herrera-Medina et al. now adds a new piece to this puzzle. Similar experiments using a transgenic approach in Medicago truncatula have shown that the plant hormone jasmonic acid plays a comparable role for the establishment of a functional AM symbiosis (Isayenkov et al., 2005). In addition, a number of plant mutants have been described to be disturbed in these later steps of the interaction. ABA is well known for its important signaling and regulatory roles that enable plants to survive a variety of abiotic stresses, such as drought, salinity and cold stress (Finkelstein & Rock, 2002). In recent years, ABA has also been implicated in the regulation of a number of biotic stresses, such as pathogen attack. Its role, however, seems to depend on the interacting organisms. An increase in the level of ABA causes an increased resistance to the bacterial leaf pathogen Pseudomonas syringae or against viral infection in tobacco, explained by the induction of stomatal closure or by the inhibition of callose degradation (Adie et al., 2007). This contrasts, however, with the reduced susceptibility (enhanced resistance) of ABA-deficient Arabidopsis or tomato mutants and soybean plants treated with an ABA biosynthesis inhibitor to various pathogens. The inverse relationship between the level of ABA and the induction of certain disease-resistance components led to the assumption that ABA has an antagonistic interaction with the JA/ET defense pathway that is necessary for resistance in Arabidopsis to the necrotrophic pathogen Fusarium oxysporum (Anderson et al. 2004). However, ABA also seems to interfere with signaling involving salicylic acid, leading to reduced plant resistance against biotrophic pathogens (Adie et al., 2007). Along these lines, the effect of ABA treatment of Arabidopsis plants on the accumulation of two key resistance components after inoculation with an avirulent strain of P. syringae pv. tomato demonstrated that ABA is a suppressor of SA and lignin production and therefore the increase in ABA levels results in an increased susceptibility to infection (Mohr & Cahill, 2007). Regarding the AM symbiosis, it appears at least questionable if mechanisms observed in pathogenic interactions are applicable. Elements of plant defense reactions are observed only in a very reduced form during this interaction (Garcia-Garrido & Ocampo, 2002). Nevertheless, cross-talk with other hormones – at least with JA and salicylate – cannot be excluded. It will be a great challenge to establish how different hormones act in AM and whether their synergistic and/or antagonistic effects determine the outcome of the interaction. AM fungi not only improve the mineral nutrition of colonized plants, they also increase their resistance to abiotic stresses. In particular, drought stress can be alleviated by AM fungi; indeed, a number of cases have been reported where the beneficial effects of AM fungi on plant performance became apparent only under drought stress conditions (Augé, 2001). Although reports are not unanimous on this point, the beneficial consequences of AM colonization regarding drought stress seem to contribute not only towards a better mineral nutrition of plants, but encompass a more direct improvement of the plant water status. In addition, reduced levels of ABA and, accordingly, improved photosynthetic parameters have been described for above-ground parts of AM plants (Augé, 2001). Given this context, it makes sense that root ABA is necessary for a sustained colonization by AM fungi. This might ensure that roots become colonized particularly strongly when it is most needed (i.e. under drought stress conditions, when ABA levels are high). A similar regulatory mechanism has already been hinted at for phosphate, the main mineral nutrient provided by AM fungi to plant roots. In this case it has been shown that symbiotic structures unable to provide this nutrient are quickly degraded (Maeda et al., 2006; Javot et al., 2007). Many more experiments will be necessary to prove that plants in the field really use such mechanisms to engage only in AM interactions when they actually obtain benefits by these interactions.