Multidrug and toxic compound extrusion-type transporters implicated in vacuolar sequestration of nicotine in tobacco roots

Abstract

Nicotine is a major alkaloid accumulating in the vacuole of tobacco (Nicotiana tabacum), but the transporters involved in the vacuolar sequestration are not known. We here report that tobacco genes (NtMATE1 and NtMATE2) encoding transporters of the multidrug and toxic compound extrusion (MATE) family are coordinately regulated with structural genes for nicotine biosynthesis in the root, with respect to spatial expression patterns, regulation by NIC regulatory loci, and induction by methyl jasmonate. Subcellular fractionation, immunogold electron microscopy, and expression of a green fluorescent protein fusion protein all suggested that these transporters are localized to the vacuolar membrane. Reduced expression of the transporters rendered tobacco plants more sensitive to the application of nicotine. In contrast, overexpression of NtMATE1 in cultured tobacco cells induced strong acidification of the cytoplasm after jasmonate elicitation or after the addition of nicotine under nonelicited conditions. Expression of NtMATE1 in yeast (Saccharomyces cerevisiae) cells compromised the accumulation of exogenously supplied nicotine into the yeast cells. The results imply that these MATE-type proteins transport tobacco alkaloids from the cytosol into the vacuole in exchange for protons in alkaloid-synthesizing root cells.

Figure 1.

Comparison of the NtMATE1 and NtMATE2 amino acid sequences with MATE protein sequences. A, Amino acid sequences were aligned using ClustalW. Residues identical to the NtMATE1 sequence are shaded. Twelve transmembrane domains (TM1–TM12) were predicted by the TMHMM program version 2.0 and are shown in solid lines. An N-terminal hydrophilic region (dotted line) was used as an antigen to produce NtMATE antibodies. B, Phylogenetic relationship of NtMATE1/2 with representative or closely related MATE family members. The nearest-joining method was used to create the phylogenetic tree, and TREEVIEW was used to visualize the resulting tree. The scale indicates the average number of substitutions per site for each cladogram, and the numbers give the bootstrap values of each node (1,000 bootstrap trials). AtDTX1, ALF5, TT12, EDS5, FRD3, TT12, At1g61890, and At3g21690 are from Arabidopsis, MTP77 is from tomato, NorM is from V. parahaemolyticus, and hMATE1 is from humans.

Figure 2.

Expression patterns of NtMATE1/2. A to C, RNA gel-blot analysis. A, NtMATE1/2 expression in tobacco roots (cv Burley 21) with different NIC genotypes. B, Root-specific expression of NtMATE1/2 in the wild type and nic1nic2. C, Wound-induced gene expression. After tobacco leaves were wounded, expression levels of NtMATE1/2, PMT, and PI-II were monitored in the leaf and the root of wild-type plants. D to G, Histochemical GUS staining of P_NtMATE1∷GUS transgenic tobacco seedlings. D, Five-day-old seedling. E, Root tip. F and G, Cross section (F) and longitudinal section (G) of the root in the differentiation zone.

Figure 3.

Subcellular localization of NtMATE1/2. A, Confocal images of NtMATE1-GFP in transgenic BY-2 cells. Images for NtMATE1-GFP fluorescence (top left), FM4-64 fluorescence on the tonoplast (bottom left), and bright field (top right) are shown; a magnified image of an overlay between the fluorescence images (bottom right) corresponds to the boxed regions. A dotted line indicates the position of the plasma membrane. Note that NtMATE1-GFP and FM4-64 are not localized at the plasma membrane. B, Suc density gradient fractionation of endomembranes. A microsomal fraction from tobacco roots was separated on a Suc density gradient. Proteins were electrophoresed, blotted, and immunodetected using antibodies against NtMATE, P-ATPase (plasma membrane marker), BiP (endoplasmic reticulum marker), and V-ATPase (tonoplast marker). C and D, Immunoelectron microscopy. The presence of gold particles is indicated by arrows. Vac, Vacuole. C, Wild-type BY-2 cells treated with MeJA at 50 μm for 12 h. D, NtMATE1-overexpressing OX5 BY-2 cells.

Figure 4.

Down-regulation of NtMATE1/2 renders tobacco roots more sensitive to exogenous application of nicotine. A, RNA gel-blot analysis of tobacco plant roots. In the NtMATE1-RNAi lines (R18 and R24), transcript levels of NtMATE1 and NtMATE2 were highly reduced compared to the level in the vector-transformed control line (VC). Equal loading of total RNA was confirmed with ethidium bromide staining. B, Alkaloid contents in leaves and roots of the 8-week-old tobacco plants. The data shown are the mean values (±sd) for more than four individual plants. C, Exogenous supply of nicotine retards root growth of tobacco seedlings. Seven-day-old seedlings that had been grown in a standard culture medium were transferred to a medium containing nicotine at 2 mm and grown for an additional 7 d. Root growth after the transfer was measured. The nic1nic2 double mutants in two different genetic backgrounds (cv Burley 21 or NC95) had highly reduced levels of NtMATE1/2 (see Fig. 2) and were included in the assay. The presented data are representative of three independent experiments. The data shown are the mean values (±sd) for more than 20 seedlings. Significant differences (P < 0.01) among the control and RNAi lines were determined by one-way ANOVA followed by Tukey-Kramer test and are indicated by different letters, while those (P < 0.01) between the nic1nic2 mutant and corresponding wild type were determined by Student's t test and are indicated by asterisks. D, Tobacco seedlings grown in the nicotine-containing medium for 7 d. Arrows indicate the positions of the root tips at the time of the transfer.

Figure 5.

MeJA elicitation or nicotine addition strongly acidifies the cytoplasm of NtMATE1-expressing tobacco cells. A, RNA gel-blot analysis of cultured tobacco BY-2 cells. When 4-d-old BY-2 cells were treated with MeJA at 100 μm in auxin-free medium for 1 d, expression of the endogenous NtMATE1/2 genes was induced. Compared to wild-type cells, NtMATE1-overexpressing cells (OX lines) had much higher levels of NtMATE1/2 transcripts, whereas NtMATE1-suppressed cells (RNAi lines) showed much lower expression levels. Gel staining with ethidium bromide confirmed equal loading of total RNA onto each lane. B, Cytoplasmic and vacuolar pH in cultured tobacco BY-2 cells. In vivo ³¹P-NMR spectroscopy was used to measure pH values of the cytoplasm and vacuole. Nicotine was supplied to the culture medium at 4 mm for 1 d. The data shown are the mean (±sd) of three NMR measurements of each sample. Significant differences among the lines at P < 0.05 were determined by one-way ANOVA followed by Tukey-Kramer test and are indicated by different letters. WT, Wild-type tobacco cells.

Figure 6.

Uptake of nicotine in NtMATE1-expressing yeast cells. A, Intracellular localization of the NtMATE1-GFP protein expressed in yeast. B, Time course of nicotine uptake into whole yeast cells. The culture medium contained nicotine at 1 mm. C, Reduction of nicotine uptake in NtMATE1- or NtMATE1-GFP-expressing yeast cells is blocked by ΔpH-disrupting drugs, NH₄Cl (10 mm) and gramicidin D (5 μm). D, Uptake of hyocyamine or scopolamine (each supplied at 0.5 mm) in yeast cells and competitive inhibition of their uptake by 2.5 mm nicotine. The data shown are the mean values (±sd) for more than three separate samples. For each treatment, significant differences (P < 0.01) among empty vector, NtMATE1, and NtMATE1-GFP were determined by one-way ANOVA followed by Tukey-Kramer test and are indicated by different letters in C, while those between empty vector and NtMATE1 were determined by Student's t test and are indicated by asterisks (*P < 0.05, **P < 0.01) in D. [See online article for color version of this figure.]