Heterologous expression of the apple hexose transporter MdHT2.2 altered sugar concentration with increasing cell wall invertase activity in tomato fruit

Abstract

Sugar transporters are necessary to transfer hexose from cell wall spaces into parenchyma cells to boost hexose accumulation to high concentrations in fruit. Here, we have identified an apple hexose transporter (HTs), MdHT2.2, located in the plasma membrane, which is highly expressed in mature fruit. In a yeast system, the MdHT2.2 protein exhibited high ¹⁴ C-fructose and ¹⁴ C-glucose transport activity. In transgenic tomato heterologously expressing MdHT2.2, the levels of both fructose and glucose increased significantly in mature fruit, with sugar being unloaded via the apoplastic pathway, but the level of sucrose decreased significantly. Analysis of enzyme activity and the expression of genes related to sugar metabolism and transport revealed greatly up-regulated expression of SlLIN5, a key gene encoding cell wall invertase (CWINV), as well as increased CWINV activity in tomatoes transformed with MdHT2.2. Moreover, the levels of fructose, glucose and sucrose recovered nearly to those of the wild type in the sllin5-edited mutant of the MdHT2.2-expressing lines. However, the overexpression of MdHT2.2 decreased hexose levels and increased sucrose levels in mature leaves and young fruit, suggesting that the response pathway for the apoplastic hexose signal differs among tomato tissues. The present study identifies a new HTs in apple that is able to take up fructose and glucose into cells and confirms that the apoplastic hexose levels regulated by HT controls CWINV activity to alter carbohydrate partitioning and sugar content.

Keywords: apple; carbohydrate partitioning; cell wall invertase; hexose transporter; tomato; unloading.

Figure 1

Spatiotemporal expression level analysis of MdHT2.2. (a) Relative MdHT2.2 expression levels in different tissues from flower, young fruit, mature fruit, mature fruit peel and flesh, young leaf, mature leaves, old leaf, root, fruit pedicel, leaf pedicel, phloem and xylem of apple trees. The expression level in flower was set to 1. (b) Trends in relative expression of MdHT2.2 at different developmental stages of apple fruit. Samples from 0 days after bloom (DAB) were set to 1. (c) Change in the relative transcript level of MdHT2.2 after treating flesh samples for 6 h with 2% exogenous sugar (Sor: sorbitol; Glc: glucose; Fru: fructose; and Suc: sucrose, with ddH₂O as a negative control, set to 1). Expression levels were calculated relative to that of MdActin. Bars represent the mean value ± SE (n ≥ 3). Different letters indicate significant differences.

Figure 2

Subcellular localization of C/N‐terminal GFP fusion proteins in Arabidopsis protoplasts. (a,c) GFP expressed in the protoplast membrane, which is indicated by the green ring outside the tonoplast. (b,d) Bright‐field images of the left side images. (e,f) CaMV35S:GFP was widely expressed in the protoplast as the control. The tonoplast is indicated by white arrows.

Figure 3

Heterologous expression of MdHT2.2 in yeast. (a) Growth complementation of a yeast mutant strain. The HT‐deficient yeast mutant carrying the empty pYST2.0 vector or the MdHT2.2‐antisense recombinant plasmid could not grow normally on 4 monosaccharides, whereas the MdHT2.2‐sense mutant yeast grew normally on fructose and glucose but slowly on galactose and xylose. (b) Relative uptake rates of feeding with 100 μm [¹⁴C]‐labelled exogenous sugars. For the inhibition assay, 50 μm CCCP (proton pump uncoupler) was applied for 30 s before the addition of [¹⁴C] fructose. The fructose uptake rate was set to 100%. Different letters indicate significant differences. (c) Time course of the sugar uptake assay by the strain carrying pYST2.0‐MdHT2.2 (black spot) and pYST2.0 empty vector (black rings, as negative control) at 100 μm exogenous [¹⁴C] fructose at pH = 5.5. Uptake rates were determined after the addition of fructose. The inset image shows a typical Km = 220.91 μm. Bars represent the mean value ± SE (n ≥ 4).

Figure 4

Sugar concentrations in tomato fruit. Glucose (Glc), fructose (Fru), sucrose (Suc) and starch concentrations at 15 DAB (young fruit), 30 DAB (breaker‐stage fruit) and 45 DAB (ripening fruit), and SSC (soluble solid content) in 45‐DAB fruit. Bars of different colours represent different lines. Bars represent the mean value ± SE (n ≥ 4). The asterisk indicates P ≤ 0.05.

Figure 5

Assay of enzymes involved in sugar metabolism in tomato fruit. (a) Enzyme activity of CWINV (cell wall invertase), NINV (neutral invertase), SUSY (sucrose synthase), SPS (sucrose phosphate synthase) and VINV (vacuolar invertase) in tomato fruit at 15 DAB (young fruit), 30 DAB (breaker‐stage fruit) and 45 DAB (ripening fruit). Bars of different colours represent different lines. Bars represent the mean value ± SE (n ≥ 4). The asterisk indicates P ≤ 0.05. (b) Acid invertase activity staining of ripening tomato fruit in 45 DAB (ripening fruit). Black bar = 1 cm.

Figure 6

Relative expression levels of several enzymes and sugar transporter genes in the unloading pathway at different developmental stages of tomato fruit. Total RNA was isolated from fruit samples collected at 15 DAB (young fruit), 30 DAB (breaker‐stage fruit) and 45 DAB (ripening fruit), and expression levels were calculated relative to that of SlActin. The expression of WT at 15 DAB was set to 1. Bars of different colours represent different lines. Bars represent the mean value ± SE (n ≥ 3). The asterisk indicates P ≤ 0.05.

Figure 7

The sugar concentration and CWINV enzyme assay of lin5‐knockout mutant lines. (a) The DNA sequence of the target region; SlLIN5 was the target sequence chosen from CRISPR direct (http://crispr.dbcls.jp/) and used with the CRISPR‐Cas9 system in this study. The SlLIN5 sequence is the DNA sequence in the reference genome database of NCBI. Sequences were aligned using DNAMAN. The dark region is the target sequence, the blue region was the difference in sequence between those lines, the target sequences of WT and L5 were the same as the control, the edited lines WT‐sllin5‐1 and L5‐sllin5‐1/2 had 4 bases missing, and WT‐sllin5‐2 had 1 base insert. (b,c) CWINV activities of ripening fruit in different lines (WT, MdHT2.2‐heterologous expression line L5, WT‐sllin5‐knockout lines, L5‐sllin5‐knockout lines). (d) glucose (Glc), fructose (Fru) and sucrose (Suc) concentrations and SSC (soluble solid content) in ripening fruit. Bars represent the mean value ± SE (n ≥ 4). Different letters indicate significant differences at P ≤ 0.05.