Effect of root exudation on microbial community
Roots determine the assembly and recruitment of plant-specific rhizosphere microbial communities by releasing exudates into the rhizosphere (Bulgarelli et al., 2013). Despite this important functional role in the plant-soil feedbacks, comprehensive and mechanistic evidence of the processes underlying the influence of root exudates towards rhizosphere microbiota is currently lacking. In this respect, rough correlations between the accumulation of root exudates and microbial communities (Chen et al., 2019) and few specific exudates included in soil (Badri et al., 2013; Zwetsloot et al., 2020; Gu et al., 2020) or growth medium (Zhalnina et al., 2018) indicated the effect(s) of root exudates on soil microbial community. Differently to these works, we applied a Multi-Omics Factor Analysis (MOFA+) using maize root exudome and microbiome dataset collected within the same growth unit to extract microbial taxa and maize root exudates that have a high probability to be involved in root-microbiota interaction.
The metabolites extracted were amino acids (L-serine, L-histidine, L-homocysteine), aminocyclopropane-1-carboxilic acid, phenolics (p-coumaric, eugenol, quercetin 3-O -(6-acetyl-galactoside) 7-O -rhamnoside, rhamnetin and cinnamoyl glucose) together with carnosic acid (supplemental material 1). Amino acids in root exudates have been proposed as osmotically active compounds (Vives-Peris, Gómez-Cadenas & Pérez-Clemente 2018), whereas phenolics play a direct antioxidant activity and chemotaxis activity that has been proposed to shape the rhizosphere community (Iannucci et al. 2021; Lucini et al. 2019; Zuluaga et al. 2021). Except carnosic acid, all these compounds were negatively correlated with most gram-negative bacterial genus (supplemental material 1) suggesting that the maize root exudates finely modulated the depletion or enrichment of most diderm (or Gram negative) lineages. Interestingly, the pattern of soil microbial compositional shifts towards the gram-positive bacterial community is a universal response to abiotic stress (Xu and Coleman-Derr, 2019). On the other side, the increase of the abundance of specific bacteria belong to thePaenibacillus , Sphingomonas , Mesorhizobium andBacillus was also observed in our work (supplemental material 1) which, altough being gram-negative, have been associated with the plant tolerance to abiotic stress (Luo et al., 2019; Figueiredo, Burity, Martínez & Chanway 2008; Khan, Mishra, Chauhan & Nautiyal 2011; Liu, Sikora & Park 2020b; Yadav, Yadav, Singh, Singh & Singh 2021; Kang et al. 2019; Khan et al. 2020 ).
An interesting finding of our work was that the root exudates affecting bacterial taxa were stress-specific. In drought-stressed plants, we found that p-coumaric acid ethyl ester and L-serine negatively correlate with a pool of OTUs representing nine bacterial genera (supplementary material 1). P-coumaric acid belongs to the phenolic acids, a class of compounds that is well-known to affect the activity and diversity of the rhizosphere microbial communities. Indeed, this compound has been demonstrated to sharply affect the microbial community of the rhizosphere of crops (Zhou and Wu, 2012; Zhou et al., 2018) and specifically to reduce the abundance of Lysobacter (Folman et al. 2003), Haliangium (Fudou et al. 2001), and Gymnoascus spp. (Liu et al. 2017b). These species pointed out plant-growth-promoting and/or plant pathogen-inhibiting effects. In our study, we also found that L-serine negatively correlated with different bacterial genera. However, it has been also observed that amino acids as well as sugars showed a lower impact than phenolic acids on the microbial community (Badri et al., 2013). Hence, maize roots reducing the exudation of p-coumaric and L-serine in presence of drought stress determined from one side a lower soil accumulation of beneficial-microbial-inhibited metabolites (p-coumaric) and, at the same time, the preservation of an useful metabolite (L-serine) for the within-plant tolerance mechanisms. Indeed, we found that p-coumaric is negatively correlated withMucilaginibacter that has been found to alleviate salt stress (Fan, Subramanian & Smith 2020), Sphingomonas andMesorhizobium which contributed to alleviate the effects of drought stress (Luo et al. , 2019; Yadav, Yadav, Singh, Singh & Singh 2021) and Paenibacillus that are also well known to be recruited by plants during drought stress episodes (Figueiredo, Burity, Martínez & Chanway 2008; Khan, Mishra, Chauhan & Nautiyal 2011; Liu, Sikora & Park 2020b).
In heat-stressed plants, we identified five compounds (1-aminocyclopropane-1-carboxilic, L-histidine, quercetin, eugenol, and rhamnetin) that explain most of the variation in the pattern of the root exudates, and that significantly correlate with microbial taxa (supplemental material 1). Also in this case, we observed an interaction between these compounds and several bacterial taxa that have been previously reported to have a beneficial influence on stressed plants. In particular, species of Bacillus are widely recognized to be beneficial microorganisms, helping plants to cope with a variety of stressors including heat (Kang et al. 2019; Khan et al.2020). While Chitinophaga, Mesorhizobium and Rhizobacter are all well-known plant-growth promoting rhizobacteria (Kour, Rana, Yadav & Yadav 2019); no previous study reports a link betweenChitinophaga and abiotic stress.
Interestingly, in the combined-stressed plants, we identified three metabolites (L-homocysteine, cinnamoyl glucose and carnosic acid) that likely have a major role in the interaction between maize roots and soil microbiota and are completely different to that obtained in the single stress treatments (supplemental material 1). Similarly to what we observed for the single stressors, also these exudates correlated with taxa that have been previously reported to promote plant growth or alleviate the effect of plant stress. Among these genera, we observed (a) the gram negative Stenotrophomonas , including several species that have been reported as plant growth promoters and elicitors of plant resistance against biotic and abiotic stress (Singh and Jha, 2017), as well as the suppression of pathogens in the rhizosphere (Schmidt et al., 2012); (b) Microbacterium which improved the growth of sugarcane (Pereira et al., 2019) and pepper in presence of drought stress (Vílchez et al., 2018); (c) the Clavibacter that was observed in the rhizosphere of the halophyte Salicornia europaea (Hrynkiewicz et al., 2019); (d) the Dietzia whose specific strain, D. cinnamea 55, was isolated in abiotic stress environment and promoted the growth of maize plants (Khan et al., 2020) or the Dietzia natronolimnaea which improved the tolerance of the wheat plants to salt stress (Bharti et al., 2016).
Finally, within the same stress treatment, the identified metabolites interact with the same group of bacterial taxa (supplemental material 1) suggesting that, mechanistically, maize plants might modulate the exudation of a specific blend of molecules acting in concert to shape rhizosphere microbial communities.
Here we show that each individual stress (single and combined stress) produces a specific signature in the composition of root exudates targeting specific microbial taxa. In this respect, our results support the existence of a mechanism by which some root exudates are used by plants to recruit beneficial microorganisms (Naylor & Coleman-Derr 2018; Williams & de Vries 2020; Liu, Brettell, Qiu & Singh 2020a). Root exudates are known to directly affect the availability of nutrients in the rhizosphere (Canarini, Merchant & Dijkstra 2016). However, our findings indicate that indirect effects related to the need to recruit specific microbial communities is pivotal under drought, high temperature or their combination. In fact, our findings highlight a role of root exudates in the complex dynamics occurring in plant-microbe interactions during single or combined stress. The correlations observed confirm that specialized metabolites in root exudates may interfere with the mutualistic interactions between roots and defined phylogroups, as a well-known strategy to overcome stress conditions (O’Banion et al. 2020; Zuluaga et al. 2021). Edaphoclimatic conditions are known to determine plants’ ability to exude compounds able to recruit and sustain a defined rhizosphere microbial community (Karlowsky etal. 2018) in a genotype-related manner (Iannucci et al. 2021). Notwithstanding, such modulation at rhizosphere community level is rather stress-specific and cannot be generalized, even across relatively related abiotic stresses.