ABC transporter OsABCG18 controls the shootward transport of cytokinins and grain yield in rice

Abstract

Cytokinins are one of the most important phytohormones and play essential roles in multiple life processes in planta. Root-derived cytokinins are transported to the shoots via long-distance transport. The mechanisms of long-distance transport of root-derived cytokinins remain to be demonstrated. In this study, we report that OsABCG18, a half-size ATP-binding cassette transporter from rice (Oryza sativa L.), is essential for the long-distance transport of root-derived cytokinins. OsABCG18 encodes a plasma membrane protein and is primarily expressed in the vascular tissues of the root, stem, and leaf midribs. Cytokinin profiling, as well as [14C]trans-zeatin tracer, and xylem sap assays, demonstrated that the shootward transport of root-derived cytokinins was significantly suppressed in the osabcg18 mutants. Transport assays in tobacco (Nicotiana benthamiana) indicated that OsABCG18 exhibited efflux transport activities for various substrates of cytokinins. While the mutation reduced root-derived cytokinins in the shoot and grain yield, overexpression of OsABCG18 significantly increased cytokinins in the shoot and improved grain yield. The findings for OsABCG18 as a transporter for long-distance transport of cytokinin provide new insights into the cytokinin transport mechanism and a novel strategy to increase cytokinins in the shoot and promote grain yield.

Keywords: ABC transporter; cytokinin; efflux transporter; grain yield; long-distance translocation; rice (Oryza sativa L.).

Fig. 1.

The expression patterns of OsABCG18 in various tissues and in response to different hormone treatments. (A) qRT-PCR analysis of OsABCG18 expression in different tissues of rice. R, root; S, stem; YL, young leaves; OL, old leaves; YP-12, 12 cm length young panicle; YP-18, 18 cm long young panicle; and OP, panicle at the filling stage. Data are means ±SD (n=3). (B) Relative expression of OsABCG18 in rice seedlings treated with different types of cytokinins (tZ, iP, cZ, and 6-BA) and the auxin IAA. Mock indicates the mock treatment as detailed in the Materials and methods. (C) GUS staining of seedlings at 10 DAG of OsABCG18pro::GUS transgenic plants, indicating that the GUS gene was primarily expressed in the root and shoot tip. Scale bar=1 cm. (D) Enlarged image of the primary root of the seedlings in (C). Scale bar=100 μm. (E and F) Semi-thin section of the primary root of the seedlings at 10 DAG. Scale bar=20 μm. (G and H) Paraffin section of the stem of adult OsABCG18pro::GUS transgenic plants. Scale bar=100 μm. (I) GUS staining of leaves from adult OsABCG18pro::GUS transgenic plants, indicating that the GUS gene was primarily expressed in the midribs. Scale bars=1 mm. (J) Semi-thin section of the leaves in (I). Scale bar=20 μm. (K) Primary branch of the 12 cm panicle (left), primary branch of the 18 cm panicle (middle), and primary branch of the panicle at the filling stage (right). Scale bar=1 cm. (L) Immature stamen. Scale bar=1 mm. (M) Mature stamen. Scale bar=1 mm. The red and blue arrows indicate the xylem and phloem, respectively. Data are means ±SD (n=3). *P<0.05 and **P<0.01 (Student’s t-test).

Fig. 2.

Subcellular localization of the GFP–OsABCG18 fusion protein. Green fluorescence distribution when 35S::EGFP-OsABCG18 (A) or 35S::EGFP (E) was transiently expressed in tobacco leaf epidermal cells; red fluorescence distribution of the plasma membrane marker CBL–mCherry (B) and (F) and their merged images with (A) and (E) as (C) and (G). (D) An enlarged image of (C), which indicates that EGFP–OsABCG18 was completely merged with the plasma membrane marker CBL–mCherry, but the negative control EGFP was only partially merged with CBL–mCherry (G and H). Scale bar=20 µm. (I–L) The fluorescence distribution in the root tip region of OsABCG18_pro::EGFP-OsABCG18 transgenic rice; the DIC image of the root (I), green florescence of EGFP–OsABCG18 (J), and red florescence of FM4-64 (K) are merged in (L). Scale bars=20 µm. The roots were stained with 10 mM FM4-64 (Molecular Probes) for 10–30 min.

Fig. 3.

Morphological phenotype of osabcg18 mutant lines in the vegetative stage. (A) Phenotype of seedlings at 7 DAG of the wild type and osabcg18 mutant lines g18-14, g18-32, and g18-24. Scale bar=1 cm. (B) Growth of the shoot (unfilled column) and root (filled column) was significantly delayed, as calculated from 4 to 7 DAG. Data are means ±SD (n=15). (C) The peeled shoot bases of wild type and osabcg18 mutant lines at 20 DAG. The wild type generated tiller buds that initiated in mutants but had not emerged at the time. Scale bar=1 mm. (D) Shoot bases of the wild type and osabcg18 mutant lines at 35 DAG. The wild type generated a few tillers that were invisible in the mutants. Scale bar=1 cm. (E) Phenotype of the wild type and osabcg18 mutant at 55DAG. Scale bars=10 cm. (F) Quantification of the tiller number and plant height in the osabcg18 mutant and wild type in (E). Data are means ±SD (n≥10). ***P<0.001 (Student’s t-test).

Fig. 4.

[¹⁴C]tZ tracer experiment and xylem sap assays in the wild type and osabcg18 mutant. Radioactivity in shoots (A) and roots (B) of the wild type and osabcg18 mutant 20 h after feeding [¹⁴C]tZ to the roots. Seedlings grown in hydroponic conditions were used. The transport of exogenously applied [¹⁴C]tZ from roots to shoots was significantly suppressed in osabcg18 mutant. Data are means ±SD (n=5). (C) Cytokinin profiling in xylem sap from the wild type and osabcg18 mutants. tZ, tZR, cZ, cZR, iP, and iPR were measured. Data are means ±SD (n=3). *P<0.05 and **P<0.01 (t-test in one-way ANOVA). tZ, trans-zeatin; tZR, trans-zeatin riboside; cZ, cis-zeatin; cZR, cis-zeatin riboside; iP, isopentenyladenine; iPR, isopentenyladenosine.

Fig. 5.

OsABCG18 mediated efflux transport of various cytokinins in tobacco. (A) Immunoblot analysis of EGFP–OsABCG18 fusion protein expression in tobacco leaves. M, marker; G18, total protein from tobacco leaves with transient expression of 35S::EGFP-OsABCG18; Ctrl, control, total protein from non-transgenic tobacco leaves; RbcL, Rubisco large subunit, serving as the protein loading control. The red arrow indicates EGFP–OsABCG18 protein. (B–H) Time course quantification of exported cytokinins tZ, tZR, DHZ, iP, iPR, cZR, and ABA (negative control) from tobacco leaves transformed with OsABCG18 or empty vector. The results are given as pmol g^–1 WT. (I) The ratios of exported cytokinins from tobacco leaves (with EGFP–OsABCG18 and free GFP expression) in the presence or absence of 1 mM sodium vanadate at the time point of 60 min. CKs, cytokinins. (J) Quantification of exported cytokinins tZ, tZR, DHZ, iP, iPR, cZR, and ABA (negative control) from protoplasts isolated from tobacco leaves transformed with OsABCG18 or empty vector at the time point of 40 min. (K) The ratios of exported cytokinins from the tobacco protoplast (with EGFP–OsABCG18 and free GFP expression) in the presence or absence of 1 mM sodium vanadate at the time point of 40 min. Data are means ±SE (n=4). *P<0.05, ** P<0.01, and *** P<0.001 from Student’s t-test. ND, not detected.

Fig. 6.

Mutation of osabcg18 affected multiple agronomic traits. (A) The phenotypes of osabcg18 mutant lines g18-14, g18-32, and g18-24 at the mature stage. Scale bar=10 cm. (B) Images of the panicles in the wild type and mutant lines in (A). Scale bar=2 cm. (C) Observation of the grain length (left) and width (right) in the wild type and mutant lines in (B). Scale bar=1 cm. (D) Effective tiller number in the wild type and mutant lines in (A). (E) The grain number per panicle in the wild type and mutant lines in (A). (F) The grain length and (G) width in the wild type and mutant lines. (H) The 1000-grain weight in the wild type and mutant lines. (I) The grain weight per plant in the wild type and mutant lines. Data are means ±SD (n=15); for the measurement of grain length and width, >150 seeds were measured for each plant. **P<0.01 and ***P<0.001 (Student’s t-test).

Fig. 7.

Overexpression of OsABCG18 increased the rice grain yield. (A) The phenotype of OsABCG18Pro::EGFP-OsABCG18 transgenic plant lines G18OE-10, 14, and 25 at the mature stage. Scale bar=10 cm. (B) Panicles of ZH11 and G18OE-14 lines in (A). Scale bar=2 cm. (C) Quantification of the expression of OsABCG18 in seedlings of the ZH11 and G18OE-10, -14, and -25 lines. The effective tiller number (D), panicle length (E), grain number per panicle (F), grain number per plant (G), and grain weight per plant (H) of the G18OE-10, -14, and -25 lines and the wild type in (A). Data are means ±SD (n≥12). *P<0.05, **P<0.01, and ***P<0.001 (Student’s t-test).