Expression of OsFRDL1, a MATE gene family member, indicates its involvement in aluminum response in rice

In soils under acidic conditions, Aluminum (Al) is solubilized to its ionic form, which is toxic to plants. Al rapidly inhibits root elongation, water and nutrient uptake, resulting in crop yield reduction. Members of the MATE family are responsible for citrate transport and Al detoxification in different species. In rice, the OsFRDL1 gene (MATE family) is homologous to the HvAACT1 and SbMATE, which are involved in Al tolerance in barley and sorghum, respectively. Silencing OsFRDL1 showed that it is not involved in Al tolerance in rice. However, the OsFRDL1 expression was not accessed in rice genotypes contrasting for Al tolerance. Thus, in this study, four Brazilian rice genotypes were evaluated in response to Al treatment under different times of exposition and OsFRDL1 expression was analyzed. The cultivars displayed different responses to Al dose x time. Al affected root growth in all analyzed genotypes, however, a minor negative effect that only occurred after 72 and 48 hours of exposure was detected in Farroupilha and BRS Curinga cultivars, respectively. In contrast, BR-IRGA 410 and IAS 12-9 showed a negative effect in root growth from the first hours of exposure to Al. Two cultivars differing in Al tolerance were used for gene expression analysis. The expression of OsFRDL1 was highly increased in Al-tolerant cultivar Farroupilha when compared to the Al-sensitive cultivar BR IRGA 410. This result indicates that OsFRDL1 is regulated by Al. This finding suggests that OsFRDL1 is involved in Al stress response, however seems to be insufficient in controlling Al tolerance.

introduction Aluminum (Al) is the most abundant metal on earth crust. In acid soils, Al is solubilized to its ionic form (Al 3+ ), which is toxic to the plants (FOY, 1988;ARENHART et al., 2014). The presence of Al in soil solution quickly inhibits the root growth as well as water and nutrient uptake by the plant, resulting in a significant reduction in the crop production under these conditions (CHANG et al., 2015). It is estimated that acidic soils comprise 40 and 50% of the world's arable soil (KOCHIAN et al., 2015;RAHMAN et al., 2018). Some plant species evolved mechanisms to deal with Al toxicity KOCHIAN et al., 2005;ARENHART et al., 2014). The most studied mechanism of tolerance is related to the secretion of organic acid anions from roots (MA, 2000;MA et al., 2001;KOCHIAN et al., 2005;KOCHIAN et al., 2015). These anions include citrate, oxalate and/or malate (YOKOSHO et al., 2011;KOCHIAN et al., 2015) and are able to complex with Al, preventing Al from entering the plant (FURUKAWA et al., 2007). Many efforts have been applied to understand the nature of organic acid secretion induced by Al KOCHIAN et al., 2005;ZHANG et al., 2019a). Plants differ from each other related to the type of secreted organic acid, temporal secretion patterns, sensitivity to temperature and responses to different Al doses (MA, 2000;YANG et al., 2013).
Two different patterns (I and II) can be identified in organic acids release based on secretion rhythm. In type I, the secretion occurs almost immediately after the addition of Al, which suggests that the metal activates a pre-existing ion channel in the plasma membrane and that it is not necessary to induce gene expression. Plants that display type II, the release of organic acids is delayed for several hours after exposure to Al, suggesting that there is a need of gene expression induction. Some inducible proteins may be involved in the metabolism of organic acids or in the transport of their anions (MA, 2000;YANG et al., 2013). Previous studies demonstrated that the secretion of organic acids is performed through anionic channels or transporters. In maize plants, Al has been shown to activate Clefflux, as well as the citratepermeable anionic channel (PIÑEROS & KOCHIAN, 2001).
The molecular control of the anion secretion has been unveiled. The first gene directly related to Al tolerance in plants, ALMT1 (Al-Activated Malate Transporter 1), was identified to be responsible for malate release in wheat (SASAKI et al., 2004;RAHMAN et al., 2018). Some proteins belonging to the MATE (Multidrug and Toxic Compound Extrusion) family are involved in the transport of citrate into plants and required for iron (Fe) translocation or Al detoxification (YOKOSHO et al., 2009). Differential gene expression analysis in soybean demonstrate that MATE genes, specially GmMATE75, are involved in Al tolerance and increased transcript accumulation 12 and 24 hours after exposed to Al treatment (LIU et al., 2016). Two MATE members were also characterized in maize, ZmMATE1 and ZmMATE2, which co-localize to a major Al tolerance QTL (MARON et al., 2010). In addition, many transporters, including members of MATE and ABC families, were involved in the process of Alcitrate complex transport in Hydrangea macrophylla roots (CHEN et al., 2015). In A. thaliana, it has been shown that the FRD3 (Ferric Reductase Defective3) protein acts in citrate transport, which is required for translocation of Fe from roots to shoots (DURRETT et al., 2007). Two other studies showed that the release of citrate induced by Al in barley (Hordeum vulgare) and sorghum (Sorghum bicolor) is mediated by transporters from MATE family (FURUKAWA et al., 2007;MAGALHÃES et al., 2007). In barley, the gene HvAACT1 (Aluminum Citrate-Activated Transporter1), encodes the carrier protein placed in plasma membrane of the root epidermal cells and is able to realize citrate secretion under Al toxicity condition (FURUKAWA et al., 2007). In sorghum, the SbMATE gene is also involved in the citrate efflux leading to Al tolerance (MAGALHÃES et al., 2007).
The rice genome presents six AtFRD3, HvAACT and SbMATE homologous genes, which were identified as OsFRDL (Ferric Reductase Defective-like). OsFRDL1 (Os03g0216700), a citrate transporter, is close to HvAACT gene (FURUKAWA et al., 2007;YOKOSHO et al., 2009). The silencing of OsFRDL1 indicates that it is not involved with citrate secretion induced by Al, but with the efficient translocation of Fe to the shoot (YOKOSHO et al., 2009). In addition, the authors observed in a specific genotype that OsFRDL1 expression was not affected by Al treatment. However, the expression profile of OsFRDL1 in response to exposure to Al in rice genotypes with different levels of tolerance has not been evaluated. Since rice roots secrete citrate in response to Al, a difference in the expression of genes involved in this process can be expected between tolerant and sensitive genotypes. The identification of differences in gene expression may contribute to the elucidation of the mechanisms involved in rice Al response. In this sense, this work aimed to evaluate the OsFRDL1 expression in rice genotypes with contrasting Al response.

Plant material and growth conditions
The rice genotypes BRS Curinga, Farroupilha, BR-IRGA 410 and IAS 12-9 Formosa were grown in hydroponic system, under controlled environmental conditions. Rice seeds were germinated on nylon screens adapted to pots containing complete nutrient solution (CAMARGO & OLIVEIRA, 1981) composed by 4mM Ca(NO 3 ) 2 ; 2mM MgSO 4 ; 4mM KNO 3 ; 0.435mM (NH 4 ) 2 SO 4 ; 0.5mM KH 2 PO 4 ; 2mM MnSO 4 ; 0.3µM CuSO 4 ; 0.8µM ZnSO 4 ; 30µM NaCl; 10µM Fe-EDTA; 0,10µM NaMoSO 4 and 10 µM H 3 BO 3 . After four days in the dark, the genotypes were subjected to a photoperiod of 12 hours of light / 12 hours of dark to the light intensity of 7,000 lux. A half part of the total of plants in V3 stage (SOSBAI, 2018) were transferred to aluminum excess treatment, which consisted of one-tenth of the total solution (without addition of phosphate to avoid possible precipitation of Al 3+ ) containing concentrations of 0 and 14 mg L -1 of aluminum, provided in the form of Al 2 (SO 4 ) 3 . The rest of the plants were kept in a standard solution (control condition). Control and Al stressed plants were kept in hydroponic system at 26°C. Plant solutions were continuously aerated and its pH adjusted to 4.0 ± 0.3 by addition of 1N HCl, as described by Camargo & Oliveira (1981). The main root length of ten plants in each treatment were morphologically evaluated at 0, 2, 6, 12, 24, 48, 72 and 96 hours after exposure to treatment. Root samples for gene expression analysis were collected at 0, 12, 24 and 48 hours after exposure to treatment. The samples were washed with autoclaved ultrapure water and stored at -80°C until RNA extraction.

rnA Extraction, cdnA synthesis and real-time quantitative reverse transcription-Pcr (rt-qPcr) analysis
To evaluate the expression of OsFRDL1 (Os03g0216700) in response to aluminum treatment in rice roots, two genotypes, one tolerant (Farroupilha) and one sensitive (BR-IRGA 410), were used. For gene expression analyses, the total RNA was extracted from root samples using TRIzol reagent (Invitrogen, CA, USA). The RNA quality and integrity were assessed by spectrophotometry (Hitachi spectrophotometer, model U-1800) and electrophoresis in agarose gel. Subsequently, the RNA samples were treated with DNase I (Amplification Grade Dnase I, Invitrogen) in order to remove remaining genomic DNA. The cDNA synthesis was performed using SuperScript II RT (Invitrogen) and Oligo(dT) according to the manufactured recommendations. The RT-qPCR experiment was performed according to MIQE guidelines (BUSTIN et al., 2009). Oligonucleotides for the target gene OsFRDL1 (Forward primer -5'-TGCTGAAAAGACCAGGAAGACA-3' and Reverse primer -5'-GTTGGCTCATTTCTTGGGCTAC-3') were designed from sequences deposited in The Rice Annotation Project Data Base (RAPDB), using Primer3Plus (http://www.bioinformatics.nl/cgibin/primer3plus/primer3plus.cgi). Oligonucleotides for the housekeeping gene Ubiquitin5 (UBQ5) (Forward primer -5'-ACCACTTCGACCGCCACTACT-3' and Reverse primer 5'-ACGCCTAAGCCTGCTGGTT-3') were obtained from JAIN et al. (2006). The RT-qPCR assay was conducted in triplicate in an ABI RT PCR 7500 (Applied Biosystems) thermocycler using SYBR Green (Applied Biosystems, California, USA) detection system. The relative expression of the target gene was calculated through the ∆∆Ct method (LIVAK & SCHIMITTGEN, 2001).

Experimental design and statistical analysis
Three replicates in a random design were used. Morphological data from root evaluation were subjected to analysis of variance (ANOVA) and a regression analysis was performed since interaction between dose and exposure time was detected in ANOVA. Both analyses were performed using SAS statistical software (SAS, 2013). Expression data are shown in bar graphics and error bars represent standard deviation from three independent biological replicates.

results and discussion farroupilha roots are less affected by aluminum excess
Al toxicity is the major factor limiting plant growth in acid soils. Small Al concentrations (micromolar) can inhibit root elongation in minutes or hours, inhibiting the water and nutrient uptake, resulting in reduced growth and yield (MA & FURUKAWA, 2003;RAHMAN et al., 2018). Since roots are strongly affected by Al, many reports have shown the evaluation of traits related to the growth of the root system (CHANG et al., 2015). Here, to understand the Al toxicity effects on rice Brazilian genotypes, we evaluated the root length trait in BR-IRGA 410, BRS Curinga, IAS 12-9 Formosa and Farroupilha genotypes in response to 14mg L -1 of Al during 96 hours ( Figure  1). To understand the effect of Al on root growth, an analysis of variance was performed (Table 1) and interactions between the treatments and exposure time was detected. In this sense, a regression analysis was performed. The increase in exposure time to Al resulted in a reduction in root length of the BR-IRGA 410 cultivar when compared to the control. The root length was highly reduced after 24 hours of exposure, showing a 50% reduction at 96 hours ( Figure 1A). The negative effects in the root development of BRS Curinga were observed at 48 hours after Al treatment and were intensified at 72 hours ( Figure 1B). Al treatment negatively affected Farroupilha roots only after 72 hours of exposure ( Figure  1C). The presence of Al in the growing media was also harmful to IAS 12-9 Formosa root growth ( Figure 1D) as well as to BR IRGA 410, demonstrating sensitivity since the first hours of Al treatment. It can be also noticed that the Al effect in IAS 12-9 Formosa was less intense as the time of exposure to the metal increased.
One of the major constraints to evaluate plants related to Al tolerance is the correct setting of the stress level, which needs to achieve a significant reduction in root growth in the sensitive and a limited effect in the tolerant genotype. On top of this, the exposure time is also an important factor to be considered (MACEDO et al., 1997). The Al dose used here (14mg L -1 ) as well as the exposure time, seems to be useful to characterize different genetic materials. In addition, the morphological difference found here showed a negative effect of Al over all genotypes analyzed at 96 hours, although in different magnitudes, indicating that 96 hours is not a suitable treatment for gene expression analysis, since the molecular signaling responsible for the phenotype was activated before 96 hours. Taking into account the phenotype observed in the morphological analysis, we chose two contrasting genotypes as models for Al tolerance and sensitivity to perform gene expression analyses. In this sense, Farroupilha, as a tolerant and BR IRGA 410 as a sensitive genotype were chosen for OsFRDL1 transcriptional analyses at 0, 12, 24 and 48 hours.

OsFDRL1 is activated in response to Al toxicity in different backgrounds
To evaluate the effects of the increased Al exposure (0, 12, 24 and 48 hours) on the expression of OsFRDL1, a RT-qPCR assay was performed. Treatment with 14mg L -1 of Al for 12 hours did not cause increases in the expression of OsFRDL1 in BR-IRGA 410, however, resulted in a 9-fold increase in Farroupilha. On the other hand, for 24 and 48 hours of treatment, 1.5 and 3.6-fold increases in expression were observed for BR-IRGA 410, respectively (Figure 2A). In Farroupilha, 5.2 and 6.1-fold increases were detected at 24 and 48 hours, respectively ( Figure 2B). Overall, Farroupilha showed a higher increase and a rapid activation of OsFRDL1 expression in response to Al.
Barley and sorghum, two other members of Poaceae, as is rice, present OsFRDL1 homolog genes, HvAACT1 in barley and SbMATE in sorghum (FURUKAWA et al., 2007;MAGALHÃES et al., 2007). In barley exposed to 0 and 30µM of Al for 6 hours, an increased expression of HvAACT1, that encodes a citrate carrier membrane protein, were detected in roots and shoots. However, higher transcript accumulation was detected in roots (FURUKAWA et al., 2007). The amount of transcript was 26 times higher in the Al tolerant cultivar than in the Al sensitive, but the level of expression was not induced by Al in none of them. The authors suggested that HvAACT1 is constitutively expressed in roots of barley and that secretion of citrate is mediated by the activation of HvAACT1 protein. Expression of SbMATE gene in sorghum, which is also related to citrate secretion, was increased in roots of resistant Al plants and was induced by the Al treatment. An increased expression was also detected with the increase of exposure time (MAGALHÃES et al., 2007).
When considering the amino acid sequence homology, OsFRDL1 shows 87% identity with HvAACT1 and 57% with AtFRD3 (present in Arabidopsis) (YOKOSHO et al., 2009). HvAACT1 is involved in the citrate secretion induced by Al (FURUKAWA et al., 2007), while AtFRD3 releases citrate that participates in the transport of iron from the roots to the shoots (DURRETT et al., 2007). Therefore, it is expected that this protein in rice membrane is functionally related to citrate release to extracellular environment in response to Al, which represents one of the major mechanisms of plant tolerance to this stress (KOCHIAN et al., 2005;ZHANG et al., 2019a). In the rice genome, there are six homologous genes close to AtFRD3, HvAACT1 and SbMATE. Previous reports showed that OsFRDL1, closed related to HvAACT1, was not affected by exposure to 50µM Al for 3 hours (YOKOSHO et al., 2009). In addition, no difference in  citrate secretion was detected between the knockout OsFRDL1 line and the cultivar Nipponbare in the presence of Al. On the other hand, here we identified differences in expression levels of OsFRDL1 in both BR-IRGA 410 (sensitive) and Farroupilha (tolerant) when control and Al treatments are compared (Figure 2).
Farroupilha showed a higher increase in OsFRDL1 expression in response to Al. As OsFRDL1 expression was observed at 12 hours after the treatment, probably the expression was initiated before, in a period between 0 and 12 hours. It may be associated to the absence of difference in root growth observed when comparing the control condition and the shorter times of Al exposure ( Figure 1C). BR IRGA 410 showed a different profile, an increase in OsFRDL1 transcripts was observed after 48 hours of Al treatment, and at lower levels when compared to Farroupilha. A tendency in root length reduction was observed in the first hours of Al treatment in BR IRGA 410. That reduction is probably related to the non-activation of OsFRDL1, responsible for citrate transport to extracellular medium. However, other genes can be involved in this process. Citrate displays a chelating role and neutralizes Al 3+ , the most toxic form of Al, preventing Al entering in root cells which can have negative effect on root growth in low pH conditions (KOCHIAN, 1995;ZHANG et al., 2019b). On the other hand, it was verified that both genotypes showed OsFRDL1 expression in absence of Al (data not shown), agreeing with the response to HvAACT1 gene in barley (FURUKAWA et al., 2007). However, when exposed to Al, the OsFRDL1 transcriptional activation was more efficient in Farroupilha, which may explain in part, the observed phenotype, whereas BR IRGA 410 seems not to be able to overcome the Al presence through this mechanism. conclusion -Farroupilha (Al tolerant) showed an increased expression of OsFRDL1 when compared to BR-IRGA 410 (Al sensitive).
-The differences found in expression levels may be associated with the morphological responses observed in genotypes in response to Al exposure, suggesting that OsFRDL1 is involved in response to Al in rice.