Up-regulation of a magnesium transporter gene OsMGT1 is required for conferring aluminum tolerance in rice

Abstract

Magnesium (Mg)-mediated alleviation of aluminum (Al) toxicity has been observed in a number of plant species, but the mechanisms underlying the alleviation are still poorly understood. When a putative rice (Oryza sativa) Mg transporter gene, Oryza sativa MAGNESIUM TRANSPORTER1 (OsMGT1), was knocked out, the tolerance to Al, but not to cadmium and lanthanum, was decreased. However, this inhibition could be rescued by addition of 10 μm Mg, but not by the same concentration of barium or strontium. OsMGT1 was expressed in both the roots and shoots in the absence of Al, but the expression only in the roots was rapidly up-regulated by Al. Furthermore, the expression did not respond to low pH and other metals including cadmium and lanthanum, and was regulated by an Al-responsive transcription factor, AL RESISTANCE TRANSCRIPTION FACTOR1. An investigation of subcellular localization showed that OsMGT1 was localized to the plasma membrane. A short-term (30 min) uptake experiment with stable isotope (25)Mg showed that knockout of OsMGT1 resulted in decreased Mg uptake, but that the uptake in the wild type was enhanced by Al. Mg concentration in the cell sap of the root tips was also increased in the wild-type rice, but not in the knockout lines in the presence of Al. A microarray analysis showed that transcripts of genes related to stress were more up- and down-regulated in the knockout lines. Taken together, our results indicate that OsMGT1 is a transporter for Mg uptake in the roots and that up-regulation of this gene is required for conferring Al tolerance in rice by increasing Mg concentration in the cell.

Figure 1.

Expression pattern of OsMGT1 in rice. A, Expression of OsMGT1 in the roots and shoots. Rice seedlings were exposed to a solution containing 50 μm Al at pH 4.5 for 6 h and the roots and shoots were sampled for analysis. B, Root spatial expression. After exposing to 50 μm AlCl₃ for 6 h, root segments (0–1 cm and 1–2 cm) were excised with a razor. C, Expression of OsMGT1 in the art1 mutant. Both wild-type (WT) rice and the art1 mutant were exposed to 20 μm AlCl₃ for 4 h. The expression level was determined by real-time RT-PCR. Histone H3 was used as an internal standard. Expression level relative to WT root (−Al) in A, C, and to root tip (−Al) in B is shown. Data are means of three biological replicates. The asterisk shows a significant difference (P < 0.05 by Tukey’s test).

Figure 2.

Time-dependent and metal-specific expression of OsMGT1 in rice roots. A, Time-dependent expression of OsMGT1. Rice seedlings were exposed to a solution containing 50 μm AlCl₃ for different time. B, Metal- and pH-dependent expression. Rice seedlings were exposed to a solution containing 0, 30 μm Cd, 10 μm La, or 50 μm Al at pH 4.5 or containing 0 μm Al at pH 5.6 for 6 h. The expression level was determined by real-time RT-PCR. Histone H3 was used as an internal standard. Expression level relative to 0 h in A and to pH 4.5 (−Al) in B is shown. Data are means of three biological replicates. The asterisk shows a significant difference (P < 0.05 by Tukey’s test).

Figure 3.

Subcellular localization of OsMGT1. OsMGT1-GFP (A), GFP-OsMGT1 (B) constructs, or GFP alone (C) were transformed into rice leaf protoplasts by polyethylene glycol method. Magenta color shows chloroplast autofluorescence. Scale bar = 10 μm.

Figure 4.

Phenotypic analysis of OsMGT1 knockout lines. A, Sensitivity of OsMGT1 knockout lines to Al. Seedlings of wild-type rice (WT) and two OsMGT1 knockout lines (NF0595 and NE4528) were exposed to a solution containing different concentration of Al (0, 10, 30, and 50 μm) for 24 h. The root length was measured before and after the treatment. Data are means ± sd (n = 10). The asterisk shows a significant difference at P < 0.05 by Tukey’s test. B, Sensitivity of OsMGT1 knockout lines to other metals. WT and two knockout lines were exposed to a solution containing 50 μm Al, 10 μm Cd, or 5 μm La for 24 h and root elongation relative to the root growth without metals is shown. Data are means ± sd (n = 10). The asterisk shows a significant difference (P < 0.05 by Tukey’s test). C, Growth on acidic soil. Germinated seeds were sowed on acidic soil (Andsol, pH 4.5) or neutral soil (pH 6.5) and grown for 1 week.

Figure 5.

Alleviation of Mg on Al toxicity in OsMGT1 knockout lines. A and B, Alleviative effect on Al toxicity by low (A) and high concentrations (B) of Mg. Seedlings of wild-type and two OsMGT1 knockout lines were exposed to a solution containing 50 μm Al in the presence of different concentrations of Mg for 24 h. The root length was measured before and after the treatment and elongation relative to the root growth without Al was shown. Data are means ± sd (n = 10). C, Alleviation of Al toxicity in OsMGT1 knockout lines by divalent metals. Seedlings of wild-type and knockout lines were exposed to a solution containing 50 μm Al with 10 μm of Mg, Ba, or Sr for 24 h. Data are means ± sd (n = 10). Means with different letters are significantly different (P < 0.05 by Tukey’s test).

Figure 6.

Al-induced citrate secretion in OsMGT1 knockout line. Seedlings (21-d-old) of wild-type rice (WT) and a knockout line (NF0595) were exposed to a solution containing 50 μm Al. Root exudates were collected for 24 h after Al treatment. Citrate was determined by an enzymatic method. Data are means ± sd (n = 3). DW, Dry weight. Means with different letters are significantly different (P < 0.05 by Tukey’s test).

Figure 7.

²⁵Mg uptake in OsMGT1 knockout lines. A, A time-dependent uptake of ²⁵Mg. Seedlings of wild-type and knockout lines were exposed to a solution containing 10 μm ²⁵Mg. At different time points, the roots and shoots were sampled for determination of ²⁵Mg with ICP-MS. B, Effect of Al on ²⁵Mg uptake. Seedlings of wild-type and knockout lines were pretreated with or without Al (0 or 50 μm, pH 4.5) for 6 h and subsequently subjected to an uptake solution containing different concentrations of ²⁵Mg. After 30 min, the roots were sampled for determination of ²⁵Mg. Data are means ± sd (n = 3). C, OsMGT1-mediated Mg uptake. The net uptake was calculated by subtracting Mg uptake of knockout lines from that of wild-type rice.

Figure 8.

Effect of Al on Mg concentration in the cell sap and cell wall. Mg concentration in the cell sap (A) and cell wall (B) of root tips (0–1 cm). Seedlings of both wild-type rice (WT) and two knockout lines were exposed to a solution containing Al (0 or 50 μm) for 8 h. The root tips (0–1 cm) were excised and fractionated into cell sap and cell wall. The Mg concentration was determined by ICP-MS. Data are means ± sd (n = 3). The asterisk shows a significant difference (P < 0.05 by Tukey’s test).