Acidic soils are widely distributed worldwide, predominantly in tropical and subtropical areas, reaching around 50% of the arable soil in the world (Von Uexküll and Mutert, 1995; Cançado, 2006; Kochian et al., 2015). This type of soil strongly reduces crop production, mainly because of the presence of aluminum therein. Aluminum (Al) is the most abundant metal on Earth, and conjugated with oxides, silicates, and hydroxides constitutes soil particles (Boff, 2006; Oliveira, 2012). In highly acidic soils (pH < 5), Al is solubilized into a phytotoxic form Al(H₂O)₆³⁺, usually referred to as Al³⁺. Aluminum is extremely rhizotoxic, affecting root growth and function and therefore limiting crop production and its toxicity, along with drought, is a primary food security concern (Von Uexküll and Mutert, 1995; Magalhães et al., 2007).

A well-known physiological mechanism used by plants to cope with Al stress involves the activation of plasma membrane transporters responsible for organic acid anions secretion from the root apex to the rhizosphere (Kochian et al., 2015). These anions form non-phytotoxic stable complexes with Al³⁺, preventing its absorption by the roots (Boff, 2006; Kochian et al., 2015). In sorghum (Sorghum bicolor), a membrane transporter gene belonging to the MATE family was identified and characterized as an aluminum-activated citrate transporter gene responsible for the Al tolerance in this crop, codifying for a proton-dependent transporter protein (Magalhães et al., 2007). Arabidopsis thaliana plants overexpressing the S. bicolor MATE gene (SbMATE) demonstrated an Al tolerant phenotype and increased root citrate exudation when compared to NT plants (Magalhães et al., 2007). These results support that a member of a MATE family from S. bicolor (SbMATE) is an Al-activated citrate secretion transporter conferring aluminum tolerance. In addition, it was demonstrated that MATE homologs are root citrate transporters in other monocot species such as maize (ZmMATE1, Maron et al., 2010, 2013), and rice (OsFRD1, Yokosho et al., 2011). Moreover, the expression of the citrate transporters SbMATE and FDR3 in barley enhanced the Al tolerance in this sensitive cereal (Zhou et al., 2014). Other study demonstrated that a MATE gene cloned and isolated from Vigna umbellata (VuMATE), when expressed in tomato, was able to increase Al resistance, accompanied by enhanced citrate efflux (Yang et al., 2011). Currently, some physiological and molecular mechanisms of organic acid anions exudation are described. Several studies have implicated that plasma membrane H⁺-ATPases are involved in Al tolerance mechanism by citrate exudation. For instance, in soybean and broad bean, the citrate exudation from the roots was associated with the up-regulation of the plasma membrane H⁺-ATPase, indicating that this protein modulates Al-induced citrate exudation (Shen et al., 2005; Chen et al., 2013, 2015). Wang et al. (2016) demonstrated that auxin is possibly involved in Al stress tolerance in soybean, since the concentration of IAA was increased by Al and exogenous application of IAA decreased Al concentration in roots by increasing citrate exudation through upregulation of GmMATE and activation of the plasma membrane H⁺-ATPase. In addition to the well described SbMATE, a Brassica oleracea MATE gene has also been shown to be involved in Al tolerance responses in A. thaliana. Transgenic Arabidopsis lines overexpressing BoMATE demonstrated Al tolerance and increased citrate exudation. Furthermore, these lines showed a lower K⁺ efflux and higher H⁺ efflux in the presence of Al, indicating that MATE is involved in the K⁺ and H⁺ fluxes during Al stress (Wu et al., 2014). Therefore, MATE, along with the anion channels ALMTs (Sasaki et al., 2004), are potential root organic acid anion transporters involved in Al resistance in plants and are potential targets for crop improvement under Al stress.

Setaria viridis is an emerging monocot plant model for molecular and genetic studies (Muthamilarasan and Prasad, 2015). It is a short, fast-growing, C4 plant with its genome sequence fully available (Bennetzen et al., 2012), making it a reliable model for genetic studies. In addition, S. viridis is amenable for genetic transformation through Agrobacterium tumefaciens, with well-established transformation protocol (Martins et al., 2015). In the present work, S. viridis was used as a model plant system to overexpress a new ortholog of Sorghum bicolor MATE (SbMATE), Brachypodium distachyon MATE (BdMATE), under the control of the maize ubiquitin constitutive promoter (ZmUbi-1) for aluminum tolerance. The BdMATE gene was chosen because it is up-regulated under Al conditions (Contreras et al., 2014), whereas S. viridis MATE is not (this study, data not shown). The rationale was to use a close homolog of SbMATE and a patent-free gene, in order to use as proof of concept to further transform C4 crops of economic importance such as sugarcane, wheat, and maize, which are difficult and time-consuming to transform. S. viridis plants overexpressing BdMATE showed an Al tolerance phenotype, characterized by sustained root growth under {20} μM Al³⁺, whereas NT plants showed RGI. Moreover, exclusion of aluminum from the root apex in transgenic plants was inferred based on results from hematoxylin assays. In addition, transgenic plants constitutively overexpressing BdMATE presented higher citrate secretion levels when compared to NT plants under Al stress, suggesting that the mechanism of organic acid anion exudation and consequent Al chelation by these compounds are involved in the Al toxicity amelioration in these plants. The expression of aluminum tolerance-related genes was evaluated, and the results obtained are discussed based on the known Al tolerance. Our results demonstrated that, in the future, the overexpression of BdMATE could be used to develop Al-tolerant crops closely related to S. viridis, such as maize, wheat, and sugarcane.

Sugarcane is a very important economic crop for Brazil nowadays. Brazil is the largest sugarcane producer worldwide, and its production is intended mainly for ethanol and sugar production. In addition, the biomass released from the sugarcane process can be used as lignocellulosic material to be degraded by microorganisms to generate renewable fuels and added-value products. Despite of its great importance, the regions for sugarcane growth in Brazil are becoming scarce. An alternative for growing sugarcane could be the Cerrado region, characterized by its soil with acidic pH. Therefore, a great effort has been made by the scientific community in order to improve sugarcane traits for its growth in poor soils. Setaria viridis is an emerging model for C4 species and it is an important model to validate some genes for further sugarcane transformation (Muthamilarasan and Prasad, 2015). In the present work, a close homolog of SbMATE gene, the B. distachyon MATE gene (BdMATE) was constitutively overexpressed in Setaria viridis (Brutnell et al., 2010), in order to verify aluminum tolerance in transgenic plants. The BdMATE gene was chosen instead of the S. viridis MATE gene (SvMATE) because BdMATE is a close homolog of SbMATE expressed under Al conditions (Contreras et al., 2014), while SvMATE is not (this study, data not shown). The results obtained with the overexpression of BdMATE in S. viridis plants clearly showed an Al tolerance phenotype in these plants (Figure 2). The main symptom of aluminum toxicity is inhibition of root growth, which is observed consistently in plants sensitive to this metal (Kochian et al., 2015). In the presence of Al, the two different homozygous transgenic lines demonstrated higher root length (Figure 2A) and RNRG (Figures 2B,C) when compared to NT plants. On the other hand, the percentage of RGI of transgenic plants under Al stress was lower when compared to the control (Figure 2D). However, it is worth to mention that the root growth of transgenic lines was lower in the presence of Al than in the absence of this metal, indicating that BdMATE ameliorates but do not completely eliminates the phytotoxicity effect of this cation.

Furthermore, hematoxylin staining, a common method used to monitor the location and aluminum accumulation in plants (Illéš et al., 2006), demonstrated that BdMATE plants did not accumulate Al in the root apex, a phenotype demonstrated by NT plants (Figure 3B). The primary site of aluminum accumulation and damage is still a matter of debate. In many species, the root apex was identified as the primary site of perception of aluminum toxicity, and consequently the expression of tolerance (Kollmeier et al., 2000; Yang et al., 2008; Horst et al., 2010; Motoda et al., 2011; Souza et al., 2016). Sivaguru et al. (2013) demonstrated that Al affects approximately the 1–3 mm portion of the roots in sorghum lines sensitive to the metal. Although the effects of aluminum may occur in all parts of the root growing regions, ruptures above pea (Yamamoto et al., 2001), maize (Jones et al., 2006), and bean roots (Kopittke et al., 2008) occurred predominantly in regions within about 1–2 mm of the root tip, in the distal portion known by the DTZ (Sivaguru and Horst, 1998). In order to effectively confer Al resistance via Al-activated citrate exudation, it is logical to assume that the highest density of root citrate transporters should be specifically located in the root regions most affected by Al toxicity (Sivaguru et al., 2013). Our hematoxylin staining results also showed that the portion of Al accumulation in roots of S. viridis NT plants occurs in the DTZ (Figure 3A). Thus, it is worth to verify the gene expression profile of target genes related to Al tolerance in different portions of S. viridis roots in non-transformed or transgenic plants submitted to Al stress, an approach that is currently being performed by our group.

As described above, the exudation of organic acid anions by the root apex is a common mechanism to confer Al tolerance in plants, as these compounds are able to chelate this metal in the rhizosphere, avoiding root absorption. It was demonstrated that MATE homologs were root citrate transporters in different monocot species such as maize and rice (Maron et al., 2010, 2013; Yokosho et al., 2011). Here, it was demonstrated that a new MATE homolog, BdMATE, when overexpressed in S. viridis plants, was able to increase root citrate exudation in transgenic plants when compared to NT plants. We verified the exudation of citrate, the major organic acid anion secreted through the MATE transporter (Magalhães et al., 2007; Kochian et al., 2015) in S. viridis plants in the absence or presence of Al. Our results confirmed that transgenic plants overexpressing the BdMATE gene secreted more citrate than NT plants in both conditions (Figure 3B). Organic acids are metabolically active solutes with a diverse array of functions in many organisms. These acids have a potential role in the osmotic adjustment and in the balance of cation concentration of the cells, for instance. As described above, OAs anions are secreted by the root apex as an aluminum resistance mechanism in most plants. Citrate, malate, and oxalate are the main OAs anions secreted to the rhizosphere in many plant species (Kochian et al., 2015). Organic acids (OAs) are produced mainly in the mitochondria, by the TCA cycle and, in a lesser extent, in the glyoxysome (López-Bucio et al., 2000). The accumulation of OAs produced by TCA cycle is dependent upon the plant species, age of the plant and tissue type. In addition, the exudation of OAs anions by roots is dependent on the route of involvement of these OAs in specific physiological functions (Rees, 1990; López-Bucio et al., 2000). Recently, Nunes-Nesi et al. (2014) discussed the involvement of enzymes of the TCA cycle in response to the stress of aluminum, since OAs participate as key components in the mechanisms that some plants use to cope with nutritional deficiency and metal tolerance. Since citrate and malate are the main OAs anions secreted to the rhizosphere under Al stress, we aimed to investigate the expression levels of genes related to CS and MD, responsible for citrate and malate production, respectively, in the transgenic plants overexpressing BdMATE. It could be possible that the tolerance phenotype presented by transgenic plants under Al stress was due to an increased expression of genes responsible for the production of citrate or malate by these plants. However, we were not able to verify any difference in the gene expression levels between transgenic and NT plants, in the absence or presence of Al (Figure 4).

The relationship between increased CS activity and Al tolerance is still inconclusive. In some species such as Triticum aestivum L. (wheat) and Secale cereale L. (rye), the evaluation of CS production in aluminum stress conditions have been performed aluminum (Li et al., 2000). Plantlets of cv. King of the S. cereale demonstrated an increased CS activity when exposed to aluminum for more than 6 h, while in cv. Atlas 66 of the T. aestivum L. there was no difference in enzyme activity in plantlets subjected to aluminum (Ryan et al., 1993; Li et al., 2000). Our results demonstrated that S. viridis transgenic plants overexpressing BdMATE with Al tolerance phenotype do not increase the levels of CS gene expression. As mentioned above, the OAs that the plant uses in order to tolerate the aluminum through exudation by the root apex are metabolites originated in the TCA cycle, a key metabolic pathway to all organisms. Thus, it is plausible that the tolerance mechanism to aluminum in plants has a significant carbon cost, being highly regulated (Ryan et al., 1993; Sivaguru and Horst, 1998; Magalhães et al., 2007; Sivaguru et al., 2013). The CS gene is constantly being expressed to maintain the balance between citrate production and consumption by the cell. The product of the CS gene, citrate, is a molecule that is converted rapidly in the TCA cycle and/or can be degraded in other cellular compartments to ensure that the metabolic system may be constantly supplied (negative feedback). Thus, it could be very difficult to observe drastic changes in the expression levels of a gene that is involved in a fine-tune regulation of pivotal metabolic processes in the cell. Therefore, more studies are necessary in order to elucidate the relationship between the levels of citrate produced by the TCA cycle with Al tolerance in plants. We were not able to measure if the transformed plants had a carbon loss penalty compared to NT plants, especially because the experiments were performed in hydroponic conditions during a short time. We are currently establishing Al stress experiments in soil to further study the biomass of plants transformed with BdMATE. In addition, root-specific and/or Al-induced promoters are being currently studied by our group, to minimize the putative carbon loss caused by the constitutive expression of BdMATE.

Altogether, our results suggest that a C4 plant overexpressing a B. distachyon MATE transporter gene (BdMATE) was more tolerant to aluminum stress than wild plants through the exudation of citrate into the rhizosphere, with subsequent avoidance of Al absorption and accumulation in the DTZ of the roots. As consequence, the transgenic plants demonstrated sustainable root growth in the presence of Al when compared to NT plants. Our results suggested that, in the future, the overexpression of BdMATE could be used to develop Al tolerant crops closely related to S. viridis such as maize, wheat, and sugarcane.

HM, AK, AR, PM, WRS, and AA conceived and designed the experiments. PM and AR performed the in silico analysis. AR, WRS, FV, and KD performed the abiotic stress assays. AR, BC, and FV carried out the RT-qPCR assays and analyzed the data. AR, PM, WRS, and AK performed hematoxylin assay. PAO, RC, DC, WRS, MB, and HM performed and analyzed the organic acids. WRS and AR wrote the manuscript. AR, WRS, GC, JVM, CS, MB, and AA contributed to the discussion of the results. HM, AK, and GC provided intellectual input and revised the manuscript. All authors read and approved the final manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.