Down-Regulating CsHT1, a Cucumber Pollen-Specific Hexose Transporter, Inhibits Pollen Germination, Tube Growth, and Seed Development

Abstract

Efficient sugar transport is needed to support the high metabolic activity of pollen tubes as they grow through the pistil. Failure of transport results in male sterility. Although sucrose transporters have been shown to play a role in pollen tube development, the role of hexoses and hexose transporters is not as well established. The pollen of some species can grow in vitro on hexose as well as on sucrose, but knockouts of individual hexose transporters have not been shown to impair fertilization, possibly due to transporter redundancy. Here, the functions of CsHT1, a hexose transporter from cucumber (Cucumis sativus), are studied using a combination of heterologous expression in yeast (Saccharomyces cerevisiae), histochemical and immunohistochemical localization, and reverse genetics. The results indicate that CsHT1 is a plasma membrane-localized hexose transporter with high affinity for glucose, exclusively transcribed in pollen development and expressed both at the levels of transcription and translation during pollen grain germination and pollen tube growth. Overexpression of CsHT1 in cucumber pollen results in a higher pollen germination ratio and longer pollen tube growth than wild-type pollen in glucose- or galactose-containing medium. By contrast, antisense suppression of CsHT1 leads to inhibition of pollen germination and pollen tube elongation in the same medium and results in a decrease of seed number per fruit and seed size when antisense transgenic pollen is used to fertilize wild-type or transgenic cucumber plants. The important role of CsHT1 in pollen germination, pollen tube growth, and seed development is discussed.

Figure 1.

Phylogenetic analysis of CsHT1. The first identified cucumber monosaccharide transporter CsHT1 shows high homology with plasma membrane monosaccharide transporters and is clearly separated from the three plastid localized monosaccharide transporters OepGlcT, McpGlcT, and SocTP1 and from the five tonoplast-localized monosaccharide transporters AtTMT1, AtTMT2, AtTMT3, OsTMT1, and OsTMT2. For the GenBank accession numbers of the sequences used for the analyses, see Supplemental Table S1. CkHUP3, Chlorella monosaccharide transporter3; MtST1, Medicago truncatula hexose transporter1; OsMST4, Oryza sativa monosaccharide transporter4; RCSTC, Ricinus communis sugar carrier protein; SspSGT2, sugar cane monosaccharide transporter2.

Figure 2.

Functional and kinetic characterization analysis of CsHT1 in yeast (Saccharomyces cerevisiae). A, Growth recovery on Glc of the hexose transporter-deficient yeast cells by CsHT1 expression. B, Time course of [¹⁴C]Glc uptake by yeast mutant EBY.VW4000 transformed with pDR196/CsHT1 (black circle) or pDR196 alone (white circle). Carrier [¹⁴C]Glc concentration was 100 μm. C, pH dependence of [¹⁴C]Glc uptake by EBY.VW4000 transformed with pDR196/CsHT1. Carrier [¹⁴C]Glc concentration was 100 μm. Time interval for [¹⁴C]Glc uptake was 4 min. D, Concentration-dependent [¹⁴C]Glc uptake. Eadie-Hofstee transformation of the data used to estimate K_m. The estimated K_m is 107.3 ± 3.78 μm. V_max = 35.7 ± 0.53 pmol min^–1 mg^–1 cells. E, Substrate specificity and effects of metabolic inhibitors on the activity of the CsHT1 expressed in the EBY.VW4000. Substrate specificity was determined by competitive inhibition of [¹⁴C]Glc uptake. Carrier [¹⁴C]Glc concentration was 100 μm (without competitor as one internal control, indicated by Control). Competing sugars were added 30 s prior to the addition of labeled [¹⁴C]Glc at a concentration of 1 mm (1 mm Glc as another control, indicated by Glc). The final concentration of metabolic inhibitors was 50 μm. The time interval for [¹⁴C]Glc uptake was 4 min. The values are presented as units relative to the values from the internal control taken as 100%. Values in B to E are means ± se (n = 3). FW, Fresh weight; OD₆₂₃, optical density at 623 nm.

Figure 3.

Expression analysis of the CsHT1 gene. A and B, Spatial expression of CsHT1 by RT-PCR. C to E, Temporal expression of CsHT1 in the male flower by RT-PCR (C) and histochemical analysis of pCsHT1:GUS plants (D and E) at different development stages of the male flower from 9 to 12. Inset of the picture of stage 12 in E shows the GUS expression is also in the germinated pollen and pollen tube. F to L, Immunolocalization of CsHT1 protein in pollen grains or pollen tubes of cucumber wild-type plants using affinity-purified anti-CsHT1 antiserum/IgG-alkaline phosphatase. F, No signal can be detected in the male flowers from stages 9 to 12 before the pollen germinated. G, I, and K, Strong signal can be detected in the germinating pollen and pollen tube in vitro (G) on the stigma (I) and in the ovule (K). H, J, and L, Same sections as in G, I, and K incubated with preimmune serum. The results in F to L represent at least five experimental replicates. R, Root; St, stem; L, leaf; mFl, male flower; fFl, female flower; Fr, fruit; C, carpopodium; Pe, petal; Se, sepal; An, anther; Po, pollen; PT, pollen tube. Bars = 4 mm (D), 100 μm (E), 20 μm (inset), 50 μm (G and H), and 100 μm (F and I–L).

Figure 4.

Localization of CsHT1 in cell plasma membrane. Transient expression of GFP (as control) and CsHT1:GFP under the control of 35S promoter in onion epidermal cells (A) and cucumber protoplast (B). Laser-scanning confocal microscopy images show fluorescence (GFP-) and merged images. Chlorophyll autofluorescence (Auto-), and the bright-field images are also presented. Arrows indicate the nucleus. Bars = 50 μm (A) and 10 μm (B).

Figure 5.

Effect of sugar composition and metabolic inhibitors on cucumber pollen germination in vitro. A, Light-microscopic photographs of wild-type cucumber pollen tubes grown in vitro with different carbon sources in different concentration after 4 h. B, The inhibition of pollen germination and tube growth by CCCP or PCMBS. Pollen germination 4 h after treatment of different concentrations of CCCP or PCMBS in the pollen germination medium with 15% (w/v) Suc as carbon source. Bars = 200 μm (A) and 500 μm (B).

Figure 6.

Characterization of pollen of CsHT1 sense and antisense transgenic T1 generation cucumber lines. A, The schematic structure of the sense expression (top) and antisense expression (bottom) vector carrying CsHT1. B, qRT-PCR analysis of the relative RNA transcription in the stage 12 male flowers of CsHT1-antisense expression lines (A) and sense expression lines (S). C, Raster electron microscopy images of pollen of wild-type (WT), A3, and S4 lines. D, No significant difference of pollen viability between the wild type, A3, and S4. E and F, Pollen grains of the wild type, A3, and S4 germinated in vitro in medium consisting of 3% (w/v) Glc (E) or 3% (w/v) Gal (F). G and H, Pollen germination ratio (G) and average pollen tube length (H) of wild-type, A3, and S4 lines after germination for 4 h. Only the pollen grains that germinated were calculated in H. Means of at least 1,000 pollen grains from three independent experiments. LB and RB indicate transfer DNA left and right borders, respectively. Npt II (Kan) indicates resistance genes encoding kanamycin. NOS-pro and NOS-ter indicate a promoter and terminator sequence, respectively. Bars = 200 μm.

Figure 7.

Pollen germination and pollen tube growth in vivo of cucumber wild-type (WT) and CsHT1 antisense transgenic line A3. A, A cucumber ovary/fruit (cv Xintaimici) shows the observation sketch. The ovaries/fruits of the wild type and A3 were cut into longitudinal slices, and the slices were fixed and stained with aniline blue. The red box indicated the approximated location that was photographed. B, The pollen of the wild type and A3 geminated, and pollen tubes grew in the stigma (top row) or in the ovaries/fruits (bottom row; arrow indicates the pollen tubes) 24 h after pollination. Bars = 500 μm.

Figure 8.

Phenotype analysis of CsHT1-antisense transgenic lines (T2 generation). A, Transcript levels of CsHT1 by qRT-PCR and semiquantitative RT-PCR in mature pollen of transgenic lines (T2). B, Immunolocalization of CsHT1 protein in pollen tubes. C to E, Phenotypic analysis of fruits and seeds of CsHT1 antisense transgenic lines. Longitudinal section of cucumber fruits 45 d after fertilization (C) and the figure of head-to-end (D) or side-to-side (E) setup of 20 seeds. F, The average fertile length (FL) of fruit (i.e. the region of the fruit with viable seeds) as shown in C. G to J, Total viable seed number per fruit (G), 10-seed length (H), 10-seed width (I), and 1,000-seed weight (J) of wild-type (WT) and CsHT1 antisense transgenic lines. X indicates the seeds of eight to 10 independent fruits from relative transgenic lines were counted. Bars = 100 μm.