A Transgenic Study on Affecting Potato Tuber Yield by Expressing the Rice Sucrose Transporter Genes OsSUT5Z and OsSUT2M

Abstract

In many plants, sucrose transporters are essential for both sucrose exports from sources and imports into sinks, indicating a function in assimilate partitioning. To investigate whether sucrose transporters can improve the yield of starch plant, potato plants (Solanum tuberosum L. cv. Désirée) were transformed with cDNAs of the rice sucrose transporter genes OsSUT5Z and OsSUT2M under the control of a tuber-specific, class-I patatin promoter. Compared to the controls, the average fructose content of OsSUT5Z transgenic tubers significantly increased. However, the content of the sugars and starch in the OsSUT2M transgenic potato tubers showed no obvious difference. Correspondingly, the average tuber yield, average number of tubers per plant and average weight of single tuber showed no significant difference in OsSUT2M transgenic tubers with controls. In the OsSUT5Z transgenic lines, the average tuber yield per plant was 1.9-fold higher than the controls, and the average number of tubers per plant increased by more than 10 tubers on average, whereas the average weight of a single tuber did not increase significantly. These results suggested that the average number of tubers per plant showed more contribution than the average weight of a single tuber to the tuber yield per plant.

Introduction

In plants, the systemic distribution of sugars and amino acids from sites of primary assimilation (source tissue) to import-dependent tissues and organs (sinks) is termed assimilate partitioning (Ransom-Hodgkins et al. 2003), and it is a major factor influencing whole plant productivity and crop yield (Wardlaw 1990). Presumably due to its physicochemical properties, sucrose represents the major transport form of photosynthetically assimilated carbohydrates in many plants (Ward et al. 1998). Thus, sucrose transport across multicellular organisms from sources to sinks is the key process in assimilate partitioning pathway. Therefore, the sucrose transporter (SUT), which mediates sucrose transport across the plasma membrane of plant cells (Riesmeier et al. 1992; Bush 1993; Williams et al. 2000; Lalonde et al. 2004), may play important roles in assimilate partitioning.

The SUT family can be divided into three main clades by phylogenetic analysis: type I, II, III (Aoki et al. 2003). Differential expression patterns of members of each type of SUT showed the distinct function. Some sucrose transporters mRNA and protein have been found in the phloem of source leaves, such as potato StSUT1, StSUT2, StSUT4 (Riesmeier et al. 1993; Kuhn et al. 1997; Weise et al. 2000; Reinders et al. 2002), NtSUT1 (Burkle et al. 1998), rice OsSUT1 (Matsukura et al. 2000), Arabidopsis AtSUT2, AtSUT3, AtSUT4 (Truernit and Sauer 1995; Stadler and Sauer 1996; Schulze et al. 2000; Weise et al. 2000; Meyer et al. 2004; Stadler et al. 2005). Knockdown and knockout of these sucrose transporter genes’ expression provide genetic evidence that sucrose transporters are essential for sucrose transport in plants. For example, reduced StSUT1 expression in tubers did not affect above ground organs but led to reduced fresh weight accumulation during early stages of tuber development, indicating that in this phase SUT1 plays an important role for sugar transport (Kuhn et al. 2003). Antisense-transgenic potato plants of NtSUT1 under the control of the CaMV35S promoter showed curled downward leaves, and strongly affected plants causing chlorosis and necroses which led to death. The effects demonstrate the importance of NtSUT1 for sucrose loading into the phloem via an apoplastic route and possibly for intermesophyll transport as well (Burkle et al. 1998). In addition, antisense-transformation of OsSUT1 does not affect photosynthetic activity and this is the first attempt to analyze the role of sucrose transporter in monocot (Ishimaru et al. 2001). Homozygous mutant of AtSUT2 resulted in stunted growth, retarded development, and sterility. The source leaves of mutant plants contained a great excess of starch, and radiolabeled sugar failed to be transported efficiently to roots and inflorescences. These data provide genetic proof that apoplastic phloem loading is critical for growth, development and reproduction in Arabidopsis and that AtSUT2 is at least partially responsible for this step (Gottwald et al. 2000).

In addition, some sucrose transporters are expressed in various sink organs and tissues, such as OsSUT1–5 in various sink organs of rice including developing grains (Furbank et al. 2001; Aoki et al. 2003; Sun et al. 2008); common plantain PmSUC1 (Gahrtz et al. 1996), barley HvSUT1 and HvSUT2 (Weschke et al. 2000; Endler et al. 2006) and fava bean VfSUT1 (Weber et al. 1997) in developing seeds; LjSUT4 in root nodules and also in uninfected roots and in green seedpods (Flemetakis et al. 2003); NtSUT3 in pollen of tobacco (Lemoine et al. 1999); AtSUC8 and AtSUC9 in various sink organs including floral tissues (Sauer et al. 2004; Sivitz et al. 2007). StSUT1 was expressed not only in the phloem of source leaves but also in sieve elements of sink tubers (Kuhn et al. 1997) and StSUT4 was strongly expressed in flowers and tubers (Chincinska et al. 2008), tomato LeSUT2 in fruit (Reinders et al. 2002). Transgenic and mutant study suggested these sucrose transporters play important roles in unloading sucrose from phloem into sinks. For example, antisense-inhibition of LeSUT2 gene expression markedly decreases in fruit and seed development and pollen germination (Hackel et al. 2006). Knockout mutant of AtSUC9 showed early flowering phenotype (Sivitz et al. 2007). Seed-specific overexpression of StSTU1 in pea increases sucrose uptake and the growth rate in developing cotyledons (Rosche et al. 2002). The phenotype of StSUT4-RNAi plants includes early flowering, higher tuber production, and reduced sensitivity toward light enriched in far-red wavelength (i.e. in canopy shade) (Chincinska et al. 2008). LjSUT4 was expressed in Xenopus oocytes and its transport activity assayed by two-electrode voltage clamping. Results suggested it functions in the proton-coupled transport of sucrose, and possibly other glucosides, from the vacuole into the cytoplasm (Reinders et al. 2008).

In our previous report, we cloned two sucrose transporter cDNAs from rice, OsSUT5Z and OsSUT2M. Expression pattern analysis suggested that OsSUT2M were higher in source leaf blades than other vegetative organs, whereas transcripts of OsSUT5Z were less traceable in vegetative organs. In reproductive organs, both transcripts of these two genes were high in panicles from the booting stage to 7 d after flowering (DAF). The different expression pattern between OsSUT5Z and OsSUT2M implied that the function of corresponding proteins OsSUT5Z and OsSUT2M might be different. OsSUT2M may not only load sucrose towards the phloem in source leaves, but also unload sucrose into floral tissues. However, OsSUT5Z may only primarily unload sucrose into floral tissues and developing caryopsis (Sun et al. 2008). In addition, another report tested the activity of OsSUT5 by expression in Xenpous oocytes and two-electrode voltage clamping (TEVC). Results showed OsSUT5 had a higher substrate affinity, less substrate specificity and less PH dependence compared with all type II SUTs tested to date. Unfortunately, expression of OsSUT2 did not produce substrate-dependent inward currents (Sun et al. 2010). To date, there is less data concerning the increase in plant yield by overexpressing sucrose transporter genes. In this study, we analyzed the function of the OsSUT5Z and OsSUT2M genes using transgenic potato driven by the tuber-specific potato class I patatin promoter (Paiva et al. 1983; Wenzler et al. 1989b). We detected increasing sugar and starch content, as well as changes in tuber yield in the transgenic potato plants.

Results

Molecular detection of transgenic plants

To analyze the potential role of OsSUT5Z and OsSUT2M in sink tuber phloem, 82 pPatatin::OsSUT5Z seedlings harbored the construct containing cDNA of OsSUT5Z, 88 completely Kan-resistant seedlings transformed with pPatatin::OsSUT2M construct containing the OsSUT2M cDNA and 13 potato seedlings containing the empty vector pCNPT-II-2300 have been obtained. Randomly selected Kan-resistant plants transformed with OsSUT5Z and OsSUT2M construct respectively were analyzed by PCR analysis using specific primers for nptII amplification. Twenty-nine PCR-positive plants of the 48 pPatatin::OsSUT5Z transformants and 34 PCR-positive plants of the 58 pPatatin::OsSUT2M transformed plants were obtained, respectively (data not shown). PCR-positive plants were further detected using specific primers for OsSUT5Z or OsSUT2M, and a fragment of the expected size was detected in the OsSUT5Z or OsSUT2M transformed plants (data not shown). The expected nptII gene fragment was detected in 9 of the 13 plants transformed with the empty vector pCNPT-II-2300. No amplification was detected in these plants using the specific primers for OsSUT5Z or OsSUT2M amplification (data not shown). The results of the PCR analysis were further confirmed by a PCR Southern blot. Positive signals were detected in the OsSUT5Z or OsSUT2M transformed plants using the respective probe-specific coding regions of OsSUT5Z or OsSUT2M (data not shown).

The transgenic plants with OsSUT5Z and OsSUT2M were further analyzed respectively by RT-PCR. The results showed that the OsSUT5Z or OsSUT2M transcripts were readily detected when using cDNAs of potato tuber as the templates (Figure 1).

Figure 1. RT-PCR analysis of OsSUT2M(A) and OsSUT5Z(B) expression in their respective transgenic tubers. (A) RNA was isolated from tubers. Lanes: 1–9, OsSUT2M transgenic lines; 10, pPatatin::OsSUT2M (positive control); 11, wild type plant (negative control); 12, 1 kb ladder DNA marker. (B) Lanes: 1–9, OsSUT5Z transgenic lines; 10, plasmid pPatatin::OsSUT5Z (positive control); 11, wild type plant (negative control); 12, 1 kb ladder DNA marker.

Analysis of agronomic characters in transgenic plants

The transgenic plants expressing either OsSUT5Z or OsSUT2M were chosen for further analysis. From their appearance, no visible changes in tuber color or shape were observed in the OsSUT5Z and OsSUT2M transgenic plants (Figure 2). In this study, the average weight of a single tuber, the number of tubers per plant and the average tuber yield per plant were measured.

Figure 2. Phenotype analysis of transgenic potato tubers. From left to right: tubers of OsSUT5Z transgenic potato D-5Z-16-1, tubers of blank vector transgenic potato D-2300-5-2 and tubers of OsSUT2M transgenic potato D-2M-20-1. Tubers of single plants displayed here are representative of the average values in size, numbers and yield per plant in the three transgenic lines.

Compared to the controls, the average weight of a single tuber in the OsSUT5Z transgenic plants increased from 4.4 g to 5.7 g, i.e. by up to 131%, the average number of tubers per plant increased from 24 to 36, i.e. by up to 150.0% and the average tuber yield per plant increased from 98.4 g to 190.1 g, i.e. by up to 193.2%. F-test determinations demonstrated that the average weight of a single tuber was not significantly different, but the average number of tubers per plant showed a significant difference (P < 0.05), and the average tuber yield per plant displayed a highly significant difference (P < 0.01) (Table 1). Among 21 OsSUT5Z transgenic plants, line 9 showed the highest tuber yield per plant and increased form 98.4 g to 318.1 g, i.e. by up to 323.3%; the average tuber numbers per plant increased form 24 to 49, i.e. by up to 204.2%, and the average tuber weight of a single tuber increased from 4.4 g to 6.5 g, i.e. by up to 147.7%. Together with the average tuber yield and the average number of tubers per plant, the average tuber weight of a single tuber was also significantly different compared to the controls. In addition, 15 plants showed higher tuber yield than the highest yield measured in control plants, i.e. 164.4 g (Figure 3A); 14 plants showed a larger number of tubers than the biggest one in control plants, i.e. 29 (Figure 3B), whereas only seven showed a higher average tuber weight of a single tuber than the highest recorded in control plants, i.e. 6.7 g (Figure 3C).

Table 1. Measurement of tuber yield of transgenic potato plants

Genotype	Line number	Average number of tubers per plant	Average weight of single tuber (g)	Average yield per plant (g)	(%) yield
Note: D-2M represents the OsSUT2M transgenic Désirée lines; D-5Z represents the OsSUT5Z transgenic Désirée lines; D-2300 represents the blank vector transgenic Désirée lines. Data are presented as the means ± SD. *, represents values that were determined by the F-test to be significantly different from the controls (P < 0.05). **, represent values that were determined by the F-test to be extremely different from the controls (P < 0.01).
D-2300	8	24 ± 5	4.4 ± 2.0	98.4 ± 36.9	100
D-2M	22	21 ± 12	4.0 ± 1.8	84.7 ± 62.7	86.1
D-5Z	21	36 ± 13*	5.7 ± 2.1	190.1 ± 67.4**	193.2

Figure 3. Measurement of tuber yield of transgenic potato. (A) Tuber yield of single transgenic potato plants. The genotypes are as follows: D-2M, OsSUT2M transgenic Désirée lines; D-5Z, OsSUT5Z transgenic Désirée lines; D-2300, blank vector transgenic Désirée lines. (B) Number of tubers in single transgenic potato plants. The genotypes are as follows: D-2M, OsSUT2M transgenic lines; D-5Z, OsSUT5Z transgenic lines; D-2300, blank vector transgenic control lines. (C) Average weight of a single tuber of a transgenic potato. The genotypes are as follows: D-2M, OsSUT2M transgenic lines; D-5Z, OsSUT5Z transgenic lines; D-2300, blank vector transgenic control lines.

However, in the OsSUT2M transgenic plants, the average weight of a single tuber, the average number of tubers per plant and the average tuber yield per plant all decreased slightly compared to control plants, but no significant differences were observed (Table 1). Among the 22 plants tested, only three showed a higher tuber yield than the highest measured in control plants (Figure 3A); four plants displayed a greater number of tubers than the highest in control plants (Figure 3B), and two plants exhibited a higher than average weight of a single tuber than the highest weight recorded in control plants (Figure 3C).

Detection of starch and sugar contents in transgenic tubers

To investigate the contents of starch and sugar (fructose, glucose and sucrose) in the transgenic plants mentioned above, tubers of similar size were selected for HPLC analysis. Three parallel experiments were carried out and gave similar results. In the OsSUT5Z transgenic tubers, the average starch content increased from 142.560 mg/g (dry weight, DW) to 162.653 mg/g DW, i.e. by up to 114.094%. The average glucose content increased from 0.166 mg/g (DW) to 0.498 mg/g DW, i.e. by up to 300.000%. Interestingly, the average fructose content increased from 0.221 mg/g (DW) to 0.611 mg/g DW, i.e. by up to 276.471% and was significantly different than the controls (P < 0.05). In the OsSUT2M transgenic plants, other that the average sucrose content decreasing slightly (from 8.651 mg/g to 8.865 mg/g of (dry weight), i.e. by up to 1024.737%), the starch, fructose and glucose contents were all higher than those in controls but showed no significant differences (Table 2).

Table 2. Sugar and starch content of transgenic potato tubers

Genotype	Number of line	Fructose (mg/g DW)	Glucose (mg/g DW)	Sucrose (mg/g DW)	Starch (mg/g DW)
Note: D-2M represents the OsSUT2M transgenic Désirée lines; D-5Z represents the OsSUT5Z transgenic Désirée lines; D-2300 represents the blank vector transgenic Désirée lines; DW represents the dry weight; *, represents values that were determined by the F-test to be significantly different from the controls (P < 0.05).
D-2300	5	0.221 ± 0.495	0.166 ± 0.370	8.865 ± 2.926	142.560 ± 13,403
D-2M	22	1.945 ± 4.744	1.244 ± 1.580	8.651 ± 7.396	195.633 ± 89.681
D-5Z	21	0.611 ± 0.293*	0.498 ± 0.462	14.540 ± 16.774	162.653 ± 32.067

Discussion

In the present study, class-I patatin promoter (Rochasosa et al. 1989) was used to control the expression of both OsSUT5Z and OsSUT2M. It was an expected result, that OsSUT5Z and OsSUT2M with high level transcripts in the transgenic tubers, was obtained by RT-PCR experiment (Figure 1). The previous studies indicated patatin is an abundant storage protein that accounts for 30–40% of the soluble protein in potato (Solanum tuberrosum) tubers (Racusen and Foote 1980; Paiva et al. 1983; Racusen 1983), and is mainly composed of patatin class I (up to 99%) (Blundy et al. 1991). A transgenic experiment showed when the patatin I promoter fused to the GUS reporter gene, the maximum GUS activity was detected in developing tubers and in stolon tips, and an increasing GUS activity with 10–30% is observed in growing tubers during tuberization (Wenzler et al. 1989a). Patatin promoter was widely used to regulate endogenous or foreign genes specific expression in sink or storage organ, such as potato tuber etc. by transgenic technology (Purcell et al. 1998; Yamada et al. 2002; Morris et al. 2006; Diretto et al. 2007).

In OsSUT5Z transgenic plants, the average tuber yield per plant significantly increased in comparison to control plants (Table 1), mainly as result of the variation in the number of tubers per plant rather than the average tuber size. However, the average weight of a single tuber was significantly different in line 9 relative to the control (Figure 3C), suggesting that the average weight of a single tuber may also have showed more contributions to the tuber yield per plant. Surprisingly, in OsSUT2M transgenic plants, the average weight of a single tuber, the average number of tubers per plant and the average tuber yield per plant only decreased slightly when compared to control plants (Table 1).

It was previously demonstrated that the patatin protein is highly expressed under control of its promoter in both vascular and parenchyma tissue during later stages of tuber development and also expressed in stolons before tuber formation (Liu et al. 1991). As mentioned above, OsSUT5Z transgenic potato showed high average tuber yield mainly due to variation in number of tubers per plant instead of average weight of a single tuber as mentioned above. A reasonable deduction can be addressed that the expression of OsSUT5Z under the control of class-I patatin promoter promoted formation of effective stolons and finally resulted in increasing numbers of tubers during stolon developmental stage and early growing tubers during tuberization, and the heterologous expression of OsSUT5Z might execute a function for phloem unloading in stolons which is followed by sucrose storage in the vacuole and starch synthesis in amyloplasts (Oparka and Prior 1992). Previous report indicated a high supply of carbohydrates, such as sucrose, to the developing stolons has been a favoring condition for tuber induction. In addition, the most pronounced change observed during very early stages of tuber initiation and enlargement is the massive formation of starch, which in the mature tuber typically represents 20% of the fresh weight (Fernie and Willmitzer 2001).

As for the slight reduction in tuber yield of the OsSUT2M transgenic plants, it may be interpreted by the different function hypothesis of sucrose transporter (Kuhn et al. 2003), which indicated that the sucrose transporter can not only act as a sucrose importer, but also in the inverse orientation. Another possibility is the existence of a different mechanism between OsSUT5Z and OsSUT2M. Research in the recent decades has revealed that the genes coding for the sucrose transporter proteins belong to multi-member gene families, implying that different SUT proteins may have their own dominating functions with respect to coordination (Kuhn 2003). For example, in Arabidopsis, nine SUT members have been classified into the three clades, SUT1-like, SUT2 and SUT4. These members are capable of interacting as either homo- or heteromers (Schulze et al. 2003). A model has been proposed for the function of the three SUT proteins, where SUT2 functions as a receptor for extracellular sucrose and regulates the relative activities of the high-affinity SUT1 transporter and the low-affinity SUT4 transporter, either by controlling protein turnover or via signal transduction, resulting in transcriptional activation/repression in the companion cell (Weise et al. 2000).

In agreement with the different effects on tuber yield by OsSUT5Z and OsSUT2M, both transgenic lines displayed different effects on sugar contents in the tubers (Table 2). OsSUT5Z transgenic tubers showed significant increase of fructose content compared to controls. These data were in consistent with previous reports (Leggewie et al. 2003). The heterologous expression of a spinach (Spinacia oleracea L.) sucrose transporter (SoSUT1) in potato (Solanum tuberosum L.) led to increased sugar levels within the tubers (Leggewie et al. 2003). However, OsSUT2M transgenic tubers showed no obvious change in starch and sugars contents compared to controls. One explanation is that different SUT proteins may have their own dominating functions.

In plants, assimilate transport occurs from source to sink tissues. Sucrose is the main form of assimilate used for long distance transport in many plant species. After its synthesis in the leaf mesophyll cells, sucrose is loaded into phloem sieve element-companion cell complex (SE-CCC) either symplastically via plasmodesmata and/or apoplastically (Van Bel 2003). The unloading of sucrose from the SE-CCC into sink cells might also occur either symplastically via plasmodesmata or apoplastically (Williams et al. 2000). In apoplastic loading or unloading, sucrose transporters are responsible for sucrose transport (Riesmeier et al. 1992; Williams et al. 2000). A previous report showed that sucrose was available in the apoplastic space throughout tuber development (Hajirezaei et al. 2000). Another report showed that the endogenous potato sucrose transporter StSUT1 was either directly involved in phloem unloading in potato tubers or indirectly regulating the apoplasmic osmolarity via its retrieval function. Therefore, both processes would keep the osmolarity low in the apoplasmic space (Kuhn et al. 2003). Plasmodesmal opening and closure were suggested to be pressure dependent (Oparka and Prior 1992). Therefore, high osmolarity in the apoplasmic space would result in the closure of plasmodesmata, thus inhibiting symplasmic phloem unloading; however, low osmolarity in the apoplasmic space would trigger the opening of plasmodesmata and would promote symplasmic phloem unloading (Kuhn et al. 2003). Moreover, tubers have been suggested to switch from apoplasmic to symplasmic phloem unloading during the stolon-to-tuber-transition of potato (Viola et al. 2001). Therefore, if both the rice sucrose transporter genes expressed in potato tubers would maintain a low osmolarity in the apoplasmic space, like the endogenous sucrose transporter StSUT1, they would trigger or maintain the opening of plasmodesmata, which can carry out symplasmic phloem unloading and would therefore increase the rate of stolon-to-tuber-transition at early stages of tuberization.

Much progress had been made in transgenic research focused on increasing the yields of crop storage organs, such as seed, grain, tuber and root tuber, as well as improving the nutrition component, such as protein, fat and starch. Starch is the most important carbohydrate storage form that is produced predominantly in plant green tissues through photosynthesis and is used for food and feed purposes. Various attempts have been described to increase starch biosynthesis in the storage organ of crops (Smith 2008). For example, a study to increase grain production reported the overexpression of the rice GIF1 (GRAIN INCOMPLETE FILLING 1) gene encoding a cell-wall invertase, which was required for sucrose unloading in the vascular tissues for starch synthesis (Wang et al. 2008). The heterologous expression of a spinach (Spinacia oleracea L.) sucrose transporter (SoSUT1) in potato (Solanum tuberosum L.) led to increasing sugar levels within the tubers (Leggewie et al. 2003). As expected, the reduced expression of the potato endogenous gene StSUT1 by antisense techniques led to decreased starch content and fresh weight accumulation during early stages of potato tuber development (Kuhn et al. 2003). However, the expression of some sucrose transporters in transgenic plants did not affect the starch content or increase yield. For instance, the antisense expression of a rice sucrose transporter OsSUT1 in rice did not show differential starch and sucrose content among transgenic and wild type plants (Ishimaru et al. 2001). It is possible that other members of the rice sucrose transporter family complemented its function. Our results also showed that the overexpression of the rice sucrose transporter OsSUT2M in potato tubers did not change starch content, sugar content and tuber yield per plant. In this study, the heterologous expression of the rice sucrose transporter gene OsSUT5Z in potato plants showed a notably increased starch content and tuber yield. This suggests that OsSUT5Z could be particularly useful for breeding high yield starch crops such as, maize, wheat, rice, potato, sweet potato and cassava via the generation of transgenic plants.

Materials and Methods

Plant materials

Potato plantlets grown in tissue culture (Solanum tuberosum L. cv. Désirée) were kindly provided by Prof. Liu Dehu (Biotechnology Research Institute, Chinese Academy of Agricultural Sciences) and were maintained on 1/2 MS medium (Murashige and Skoog 1962) which contained 2% (w/v) sucrose in a greenhouse at 25 °C (with 60% humidity and 16-h light/8-h dark). The plantlets were subcultured by inoculating the stem with 3–4 leaves onto new medium every 4 weeks.

Construction of plant expression vectors

A 706 bp EcoRI/SalI fragment containing the potato class I patatin promoter from pBCPatatin (kindly provided by Prof. Lin Zhongping; School of Life Science, Peking University), was cloned into the HindIII/SalI restriction site of the plant expression vector pCNPT-II-2300 (Li et al. 2001). The obtained p2300P at derivative was digested by SmaI and BamHI ligated with the 1531 bp fragment of OsSUT2M (DQ07259) cDNA and was followed by the nopaline synthesis terminator. The resultant plant expression vector was named after pPatatin::OsSUT2M. Thus the OsSUT2M cDNA containing the full length ORF was in the sense orientation under the control of class-I patatin promoter. A 700 bp BamHI/HindIII fragment containing the potato class-I patatin promoter from pBCPatatin together with a 1940 bp HindIII/EcoRV fragment containing the 1635 bp fragment of OsSUT5Z (DQ0725932) cDNA followed by the nopaline synthesis terminator were ligated concomitantly into the SamI/BamH restriction site in pCNPT-II-2300 to produce plant expression vector pPatatin::OsSUT5Z, in which the OsSUT5Z cDNA was driven by a class-I patatin promoter and terminated with the nopaline synthesis terminator. Both the plant expression vectors pPatatin::OsSUT5Z and pPatatin::OsSUT2M contain the plant selectable marker gene, neomycin phosphotransferase II gene (nptII), which conferring plant with resistance to kanamycin (Kan). These two plant expression vectors were then introduced into Agrobacterium tumefaciens strain LBA4404 (Ooms et al. 1982) by electroporation.

Plant transformation and plantlet regeneration

Agrobacterium-mediated transformation was performed with the Agrobacterium tumefaciens strain LBA4404 (Ooms et al. 1982) harboring the blank vector pCNPT-II-2300 and the plant expression vectors pPatatin::OsSUT5Z and pPatatin:: OsSUT2M, respectively. Leaves with their petioles from 4-week-old in vitro grown plantlets were used for the genetic transformation. The transformation procedure was a modification according to previously described methods (Beaujean et al. 1998). The cut explants were immersed in the Agrobacterium suspension for 5 min, blotted dry on filter paper and co-cultured for 3 d in the dark on MS medium supplemented with 0.25 mg/L of N-acetyl-aspartate (NAA), 2.5 mg/L of 6-benzylaminopurine (6-BA), 10 mg/L of GA₃ and 10 uM of acetosyringone (AS). After co-culturing, the explants were washed in 500 mg/L of cefotaxime (Cef) to kill the Agrobacterium. The explants were then cultured on a series of selective media. The selection of transgenic plants was first carried out by adding Kanamycin into the medium. Then inducement on callus selective medium (MS medium supplemented with 0.25 mg/ L of NAA, 2.5 mg/L of 6-BA, 10 mg/L of GA_3, 500 mg/L of Cef and 60 mg/L of Kan) for 1 month, and the explants with Kanamycin resistant calli were subsequently cultured on regeneration selective medium (MS medium supplemented with 2.5 mg/L of 6-BA, 10 mg/L of GA_3, 250 mg/L of Cef and 60 mg/L of Kan) for 1 month. Finally, 2–3 cm shoots with Kanamycin resistance were cut and cultured on rooting medium (MS medium supplemented with 250 mg/L of Cef and 60 mg/L of Kan) for 3–4 weeks to induce root formation. Seedlings with Kanamycin resistance were transferred into the field until potato tubers were harvested.

Molecular analysis of transgenic plants

For PCR analysis, genomic DNA isolated from leaves was used as template according to the method described by (Stewart and Via 1993). The PCR was first performed by amplifying the nptII marker gene in all transgenic plants; next, random PCRs were performed by amplifying the target genes OsSUT5Z and OsSUT2M on the nptII-positive plants. The nptII -PCR was carried out in a total reaction volume of 25 μL containing 1.0 μL of DNA template (1–2 μg), 1.0 μL of each primer (0.04 μM), 2.0 μL of each dNTP (0.2 mM), 2.5 μL of 10 × buffer (1×), 17.5 μL of sterilized double distilled water (ddH₂O) and 1.0 U of Taq polymerase (Tian Wei Shi Dai). The PCR conditions used were as follows: pre-denaturation at 94 °C for 10 min, followed by 35 cycles of denaturation at 94 °C for 45 s, annealing at 60 °C for 30 s and a final extension step at 72 °C for 1 minute. Both the target gene PCR and RT-PCR analyses were carried out as we described previously (Sun et al. 2008).

The primers for nptII-PCR were as follows:

• nptII-F: 5’-GAACAAGATGGATTGCACGCAGG-3’

• nptII-R: 5’-AAGAAGGCGATAGAAGGCGATGC-3’

• OsSUT2M-F: 5’-GGAGGCAAACACGAGAAGAAGATG-3’

• OsSUT2M -R: 5’- ATTCTTATCGGTGACTCTCCTCCTT-3’

• OsSUT5Z -F: 5’-TACGACGGCAATGGAGGAAGG -3’

• OsSUT5Z -R: 5’-TTTGATGGAGTTTCGCACTAGTG-3’

Analysis of starch and sugar content

Sugar and starch were isolated from transgenic potato tubers either transformed with the target genes or with a blank vector control according to the methods of (Trethewey et al. 1998) and (Wang et al. 1993). The tubers used for the isolation of sugars and starch were in similar size.

The analysis of sugars was carried out by high performance liquid chromatography (HPLC) according to a previously described method (Wang et al. 2004) and by following the instructions of instrument (LC-10Atvp, Shimadzu). The system controller was SCL-10Avp; the auto injector was SIL-10ADvp, the analytical column was CLC-NH₂ (6.0 mm × 150 mm); the differential refractive index detector was RID-10A; the high pressure pump was LC-10ATvp; the mobile phase was acetonitrile–water (75:25); the velocity of flow was 1.0 mL/min; the column temperature was 40 °C and the data analysis system was class-up. For the analysis of sugars, the concentration gradient of fructose, glucose and sucrose in the sugar mix was 0.06 mg/mL, 0.12 mg/mL and 0.60 mg/mL. The standard curves were expressed as follows:

• 

For starch analysis, the concentration gradient of standard glucose was 0.6 mg/mL, 3.0 mg/mL and 6.0 mg/mL. The standard curve was:

• 

χ= apex area

Starch was expressed as glucose equivalents multiplied by 0.9 (Li et al. 1996)

(Co-Editor: Jianmin Wan)

Acknowledgements

We thank Liu Dehu for kindly providing the potato plantlets grown in tissue culture and Lin Zhongping for kindly providing the class I patatin promoter. This research was supported by the State Key Basic Research and Development Plan of China, the Innovation Foundation of the Chinese Academy of Science, the Program Strategic Scientific Alliances and the China Exchange Program between China and The Netherlands.