Introduction

Potato (Solanum tuberosum) is the fourth most important food crop in the world. Potato tubers are often stored at low temperature in processing industries to prevent sprouting, water loss, and disease infection for prolonging their storage period. However, cold storage leads to accumulation of reducing sugar (RS) in the tubers, known as cold-induced sweetening (CIS) (Greiner et al. 1999). During frying, RS interacts with free amino acids, making the products unsavory and unsightly (Thill and Peloquin 1995). This reaction also generates a neurotoxin and potential carcinogen acrylamide (Shepherd et al. 2010). Therefore, exploring the mechanism of CIS has both scientific and applied importance (Muttucumaru et al. 2008). Changes in carbohydrate metabolism (Blenkinsop et al. 2004) and membrane properties (Wismer et al. 1998) have been postulated to explain CIS, but its regulation mechanism remains incompletely understood.

Thioredoxin h (TRX h) in higher plants is involved in numerous biological processes. TRX h can regulate cereal seed quality by controlling the compounds delivery during seed filling, promoting the mobilization of stored nitrogen and carbon in the endosperm in imbibed-seed, and protection against oxidative stress in the living tissues during seed maturation and germination (Lozano et al. 1996; Yano et al. 2001; Abderrakib et al. 2008; Li et al. 2009). TRX h is also believed to be involved in other biological processes, such as self-incompatibility (Cabrillac et al. 2001), carbon and nitrogen metabolism (Kobrehel et al. 1991), defense responses (Sun et al. 2010) and seed sprouting (Guo et al. 2011). In the last 30 years, more than 400 potential thioredoxin target proteins were identified and some enzymes reported to contribute to CIS were included, such as Suc synthase (Blenkinsop et al. 2004), starch debranching enzyme (Cho et al. 1999) and alcohol dehydrogenase (Pinhero et al. 2007). However, information that directly indicates the association of thioredoxin to potato CIS is limited.

Based on 29 expressed sequence tags (ESTs) identified from a suppression subtractive hybridization (SSH) cDNA library that enriched the CIS-resistance genes of potato tubers (unpublished), an EST showing high similarity to thioredoxin was found to be up-regulated early in potato tubers stored at 4 °C. In present article, a novel TRX h gene, SbTRXh1, was cloned from a wild potato species Solanum berthaultii with CIS resistance and its involvement in CIS regulation of potato tubers was investigated.

Materials and methods

Cloning, sequencing and bioinformatics analysis of SbTRXh1

The SbTRXh1 cDNA was cloned by rapid-amplification of cDNA ends (RACE) approach using a SMART™ RACE cDNA Amplification Kit (Clontech, USA) following the operation manual. Primers 5′-TTCTGAGGAAGGACAAGTGAT-3′ (forward, for 3′-RACE) and 5′-TTAATGAAGACAAAGGTAGGCA-3′ (reverse, for 5′-RACE) designed according to the EST sequence identified from the SSH cDNA library were used. The 5′- and 3′-fragments were spliced with the primer 5′-TTACGAAGAGCTAAGGATTTG-3′ (forward) and Oligod (T)18 through the overlapping PCR way. The single-strand cDNA synthesized from the RNA of S. berthaultii tubers stored at 4 °C for 5 days was used as a template. The PCR products were gel-purified and cloned into the pMD18-T vector (TaKaRa, Japan) (the yielded plasmid was denoted as pMDTRX) for sequencing at Shanghai Sangon Biological Engineering Technology & Service Co., Ltd. (Sangon, China). The homologous sequences were analyzed using the National Center for Biotechnology Information (NCBI) blastn (nr/nt) (http://www.ncbi.nlm.nih.gov). The conserved domains of the deduced protein sequences were analyzed by the Pfam sequence search (http://pfam.sanger.ac.uk). Phylogenetic and molecular evolutionary analyses were conducted using the MEGA version 4. The protein theoretical molecular weight (MW) was predicted by the Compute pI/Mw (http://www.expasy.org/tools/pi_tool.html), and the signal peptide of putative proteins was forecasted by the Signal 3.0 server (http://www.cbs.dtu.dk/services/SinalP-3.0).

Subcellular localization of SbTRXh1

The tobacco BY-2 cell line (Nicotiana tabacum L. cv. Bright Yellow 2) was kindly supplied by professor Mengxiang Sun (Key Laboratory of Plant Developmental Biology, Wuhan University, Ministry of Education) and cultured in darkness in liquid Murashige and Skoog (MS) medium (pH 5.8) containing 1.0 mg/L 2,4-D and 3 % sucrose (Suc) at 26 °C on an incubator at 130 rpm, and sub-cultured every 5 days; 4-day-old cells were used in the experiment.

The open reading frame (ORF) of SbTRXh1 (70–438 bp) was cloned into pBI121GFP vector (35S CaMV) via NruI/SmaI (TaKaRa, Japan), resulted in pBITRX-GFP plasmid. The vector was verified by sequencing (Sangon, China) and transformed into BY-2 cells by particle bombardment as previously described (Von 2007). The transformed cells were incubated at 28 °C in darkness for 24 h. After treating with 0.5 M NaCl for 10 min, fluorescence detection was carried out using a confocal laser scanning microscope (LSM510 Meta, Zeiss, Germany) with excitation 488 nm for green fluorescent protein (GFP).

Development of RNA interference (RNAi) and over-expression (OE) constructs

A 326-bp fragment of SbTRXh1 (286–611 bp) was selected for constructing the RNAi vector, and the 372 bp ORF of SbTRXh1 (70–441 bp) was used in constructing the OE vector. The plasmid pMDTRX was used as template in the DNA fragment cloning. The PCR products amplified using primers 5′-AAAAAGCAGGCTAAGAAATTGCAGAGGATTGGAATGT-3′ (forward) and 5′-AGAAAG CTGGGTTAGCCTTACACCAGTAAACACCATAT-3′ (reverse) were transformed into the pHGRV vector to produce RNAi-construct pHGTRXR (35S CaMV) by the BP clonase pathway using Gateway R BP clonase™ II Enzyme mix (Invitrogen) following the operation manual. The OE-recombinant pBITRX was constructed by cloning the PCR products amplified using primers 5′-TTGAGCTCGGATCCTGTGAGAATGGCTACTTCAT-3′ (forward) and 5′-TTGTCGACAGCA GTCACAATAGCAGGAG-3′ (reverse) into the pBI121 vector (35S CaMV) through double restriction digestion (BamHI/SacI; TaKaRa, Japan) and DNA ligation reaction (T4 DNA Ligase; TaKaRa, Japan). The inserts of both constructs were confirmed by sequencing (Sangon, China).

Transformation of SbTRXh1 into potato

The transformations of pHGTRXR (RNAi) and pBITRX (OE) into potato (S. tuberosum) cv. E-potato 3 (E3) microtubers were carried out via Agrobacterium-mediated transformation as described by Si et al. (2003), except that 50 mg/mL kanamycin was used for the selection of OE-transgenic plants. As described previously (Ni et al. 2010), the confirmation of transformations was performed by PCR analysis with specific primers of npt II gene and Southern blot hybridization using the DIG-labeled npt II fragment as probe. The transcripts of SbTRXh1 in three-week-old transgenic plants were detected by RT-PCR analysis using primers 5′-TGCTGCTCCTGCTATTGTTGA-3′ (forward) and 5′-TATACCACAAATGACCACATTCC-3′ (reverse). The transcripts of the elongation factor 1-α (ef1α) were used for normalization. Single-copy transgenic plants that showed expected transcripts of SbTRXh1 were selected for further analysis.

Storage of potato tubers and sugar content analysis

The RNAi- and OE-transgenic lines and the wild-type (WT) potato cv. E3 were grown in 24 cm × 24 cm plastic pots in a greenhouse at 20–25 °C. The mature tubers were harvested, placed at room temperature for 10 days for skin setting, and then divided into two groups for storage temperature treatments in darkness: one group was stored at 4 °C and the other at 20 °C. Tubers were sampled at time points of 0, 5, 15, and 30 days during storage. Sampled tubers were cut into small pieces, immediately frozen in liquid N2 and kept in a −70 °C freezer for the molecular and biochemical analyses as indicated bellow. Starch, RS and Suc contents were determined as described previously (Song et al. 2005; Liu et al. 2010).

PCR and qRT-PCR

The PCR programs for RACE, overlapping, and RNAi- and OE-recombinant construction were performed as follows: 94 °C for 3 min, 30 cycles of 94 °C for 45 s, 58 °C for 45 s and 72 °C for 1 min, and then 72 °C for 5 min. For the RT-PCR, the amplification cycles were reduced to 27.

RNA isolation, reverse transcription, and qRT-PCR were performed as described previously (Liu et al. 2010). The transcripts of ef1α were used as an internal control. The specific primers for the qRT-PCR were 5′-GAAGTGGATAGGGTGGTTGG-3′ (forward) and 5′-AGCAGCACCATGC TTCTCTA-3′ (reverse). The relative gene expression data were analyzed using a comparative Cp method, as described by Roche. The PCR program was set up as: 95 °C for 30 s, 40 cycles of 95 °C for 15 s, 55 °C for 30 s and 72 °C for 5 s.

Tricine-SDS-PAGE and Western blot hybridization

The total protein of sampled tubers was extracted using the plant total protein extraction kit (Applygen, China) following the operation manual. The total protein concentration was determined by Bradford analysis using the Protein Quantification Kit (Applygen, China) following the instructions of the manufacturer. The polyclonal antibody (rabbit) of SbTRXh1-PepEST “CHKVEDWEVQPQKGV” was produced by Hangzhou HuaAn Biotechnology Co., Ltd. (China). The antibody of β-tubulin (aC-18) was purchased from Santa Cruz Biotechnology, Inc., USA.

For each sample, approximately 30 μg (for RNAi-transgenic lines and WT) or 15 μg (for OE-transgenic lines and WT) total protein was subjected to 15 % Tricine-SDS-PAGE (12 % glycerine v/v, 1.83 M acrylamide, 0.023 M methylene diacrylamide, 1 M Tris–HCl pH 8.45, 0.1 % SDS w/v, 0.1 % AP w/v, and 0.05 % TEMED v/v), and then transferred onto a PVDF membrane (0.45 μm) in mini-PROTEAN electrophoresis cells (Bio-Rad). Western blot was performed using the PepEST-antibody as the primary antibody and the goat anti-rabbit IgG conjugated with horse radish peroxidase as the secondary antibody. The protein bands were visualized by enhanced chemiluminescence detection reagents (Applygen, China). The protein content of β-tubulin was taken as an internal control.

Results

Cloning and characterization of SbTRXh1 cDNA

SbTRXh1 was cloned from the cold-stored tubers of S. berthaultii. The 5′-RACE of SbTRXh1 resulted in a 340 bp fragment, while the 3′-RACE produced a 587 bp fragment. The 258 bp common fragment showed 100 % identity with the EST of SbTRXh1. Splicing of the two fragments yielded a 669 bp sequence of SbTRXh1 gene containing a 72 bp 5′-untranslated region and a 228 bp 3′-untranslated region (Fig. 1a). The ORF of SbTRXh1 encodes a protein of 123 amino acids (aa) which has an estimated MW of 13.56 kDa and isoelectric point of 5.54 without signal peptide. The SbTRXh1 gene was submitted to the GenBank database (GenBank ID: DQ413184.1).

Fig. 1
figure 1

Characterization of SbTRXh1 cDNA. a Nucleotide sequence of SbTRXh1 cDNA and deduced amino acid sequence. The deduced protein sequence is underneath the corresponding nucleotide sequence. The putative polyA signal (AATAA) is underlined. The conserved active site (WCGPC), characteristic tryptophan (W), and the beginning and stop codons are framed. b The evolutionary relationship of SbTRXh1 to isoforms of thioredoxin h group from Arabidopsis thaliana. The phylogenetic tree was constructed by neighbor-joining method with amino acid sequences. The GenBank accession numbers of the genes used in the construction of the phylogenetic tree are AtTrxh1 (P29448.1), AtTrxh2 (Q38879.2), AtTrxh3 (Q42403.1), AtTrxh4 (Q39239.2), AtTrxh5 (Q39241.1), AtTrxh7 (Q9XIF4.1), AtTrxh8 (Q9CAS1.1), AtTrxh9 (Q9C9Y6.1), AtTrxh10 (Q9LXZ8.2), AtCXXS1 (AF144390) and AtCXXS2 (ATU35639)

Full size image

The blastn (nr/nt) analysis of SbTRXh1 in NCBI showed that the full-length cDNA of the gene shared 91 % identity with an isoform of thioredoxin h group in Capsicum annuum (GenBank ID: EF371503.1) and 88 % identity with the thioredoxin from N. tabacum (GenBank ID: ×58527.1). The sequence search analysis in Pfam showed that the deduced protein sequence of SbTRXh1 contained a conserved region of thioredoxin family from the first to the 104th aa, with a characteristic tryptophan (W) of h1 subgroup at the 19th aa and a conserved active site (WCGPC) (Gelhaye et al. 2004) from the 41st aa (Fig. 1a). SbTRXh1 was assigned to the thioredoxin h1 subgroup by phylogenetic analysis with the TRX h isoforms from Arabidopsis thaliana, showing a close relationship with AtTrxh1 (P29448.1) (Fig. 1b).

Transient expression of plasmids pBI121GFP and pBITRX-GFP in BY-2 cell was carried out to figure out the subcellular localization of SbTRXh1. As shown in Fig. 2, fluorescent signals could be detected in whole of the cells for GFP alone, while SbTRXh1 were mainly localized in cytoplasm and nucleus. However, whether the GFP signal in nucleus is due to the transient expression or the physiological significance of SbTRXh1 will be further studied.

Fig. 2
figure 2

Subcellular localization of SbTRXh1 BY-2 cell line was transformed by particle bombardment with pBI121-GFP and pBITRX-GFP plasmids respectively. a1–a3 Transient expression of pBI121-GFP shows green fluorescent protein (GFP) signals in whole cell. b1–b3 Transient expression of SbTRXh1-GFP shows GFP signals mainly in nucleus and cytoplasm. The scale bars stand for 10 μm (color figure online)

Full size image

Development and selection of transgenic plants

Five RNAi-transgenic and 33 OE-transgenic lines were obtained from 20 and 170 regenerated plants, respectively. According to the results of the RT-PCR (Fig. 3a) and Southern blot analysis (Fig. 3b), the RNAi-transgenic lines SbTRXh1-Ri-E3-1 (TR-1) and SbTRXh1-Ri-E3-5 (TR-5), and the OE-transgenic lines SbTRXh1-OE-E3-105 (T-105) and SbTRXh1-OE-E3-105 (T-156) with single copy of the insertion were chosen for further study. Among them, the transcripts of SbTRXh1 were decreased by 66.1 % in TR-1 and by 88.2 % in TR-5, whereas the transcripts were increased to 6.3- and 8.5-fold in T-105 and T-156, respectively (Fig. 3a), representing expected transcriptional expression of SbTRXh1.

Fig. 3
figure 3

Selection of transgenic lines for further study. a RT-PCR-based transcript analysis of SbTRXh1 gene in transgenic and WT potato plants. RNA from 3-week-old transgenic plants was used in the analysis. The transcripts of the elongation factor-1α were used for normalization. The electrophoretic bands were quantized using Quantity One 1-D analysis software (Bio-Rad, USA), and the SbTRXh1/ef1α ratio was calculated by dividing the value of SbTRXh1 bands by that of ef1α in each sample. The error bar indicates the SE of the three replications. b Southern blot analysis on genomic DNA from transgenic and WT potato plants. The genomic DNA was restriction digested with EcoRI. The DIG-labeled 676 bp npt II fragment was taken as probe. Single-copy transgenic lines TR-1, TR-5, T-105 and T-156, which showed expected transcriptional expression of SbTRXh1, were chosen for further study. E3 WT of E-potato 3, TR-1 SbTRXh1-Ri-E3-1, TR-2 SbTRXh1-Ri-E3-2, TR-4 SbTRXh1-Ri-E3-4, TR-5 SbTRXh1-Ri-E3-5, T-105 SbTRXh1-OE-E3-105, T-156 SbTRXh1-OE-E3-156

Full size image

Transcripts of SbTRXh1 and SbTRXh1 protein contents in stored potato tubers

During the storage, both transcripts and protein contents of SbTRXh1 in the RNAi-transgenic tubers were significantly lower than those in the WT tubers at each time point, the reverse was true for the OE-transgenic tubers (Fig. 4).

Fig. 4
figure 4

Analysis of transcripts of SbTRXh1 and SbTRXh1 protein content of stored potato tubers. a, b Expression patterns of SbTRXh1 in stored tubers analyzed by RT-qPCR approach. The transcripts of the elongation factor 1-α was set to be 100 and used as normalization in RT-qPCR analysis. The error bar indicates the SE of the three replications. c Analysis of SbTRXh1 protein content in tubers stored at 4 and 20 ºC by Western blot. The PepEST-specific polyclonal antibody and β-tubulin-specific monoclonal antibody were used for the detection of SbTRXh1 and β-tubulin, respectively. d Relationship between relative expression of SbTRXh1 and relative SbTRXh1 protein content. Protein bands in Fig. 3c were quantized using Quantity One 1-D analysis software (Bio-Rad, USA), and protein content in E3 tubers stored at 4 ºC for 5 days (E3-4-5) was set to be 1, the relative SbTRXh1 protein content was calculated by dividing the value of protein bands by that of E3-4-5 in the same membrane. **Significant at the p < 0.01 level. E3 WT of E-potato 3, TR-1 SbTRXh1-Ri-E3-1, TR-5 SbTRXh1-Ri-E3-5, T-105 SbTRXh1-OE-E3-105, T-156 SbTRXh1-OE-E3-156

Full size image

Compared with WT tubers at each time point, transcripts of SbTRXh1 were suppressed by 62.5–89.4 % in TR-1 and by 65.7–92.7 % in TR-5 (Fig. 4a), while they were increased to 3.4- to 39.4-fold in T-105 and to 13.2- to 166.4-fold in T-156 (Fig. 4b). The results also showed that low temperature (4 °C) induced a dramatically higher expression of SbTRXh1 when compared to the tubers stored at 20 °C, suggesting SbTRXh1 is cold-inducible.

Similar tendency was found in SbTRXh1 protein content. As shown in Fig. 4c, the RNAi-transgenic tubers had remarkably lower protein contents, whereas the OE-transgenic tubers yielded higher protein contents in comparison with the WT tubers. The storage temperature had obvious effect on SbTRXh1 protein content; a relatively higher protein content was detected at each time point of 4 °C storage compared with the tubers stored at 20 °C. The analysis further indicated that the protein content (y) showed a high correlation with the abundance of the transcripts (x) represented by a function of y = 0.064 + 0.512 ln x (r 2 = 0.821, n = 56) (Fig. 4d), implying that the function of SbTRXh1 could be mainly regulated at transcriptional level.

Content changes of RS, Suc and starch in stored tubers

It is expected that the low temperature (4 °C) storage led to a continuous increase in RS and Suc content, whereas they were not obviously changed in the tubers stored at 20 °C (Fig. 5a, b). Interestingly, there were no significant shifts in starch content observed in the tubers stored either at 4 or 20 °C (Fig. 5c), leading to a speculation that Suc might be more responsive to CIS than starch in potato tubers.

Fig. 5
figure 5

Sugar contents and correlations between sugar contents and transcripts of SbTRXh1 of potato tubers in storage. a–c Contents of reducing sugar (RS), sucrose (Suc) and starch of potato tubers during the storage at 4 and 20 °C. The error bar indicates the SE of the three replications. d–f Correlations between sugar contents and transcripts of SbTRXh1 in RNAi-transgenic and WT potato E-potato-3 tubers stored at 4 and 20 °C, and between RS and Suc contents in all stored tubers. **Significant at the p < 0.01 level

Full size image

In comparison with the WT tubers, the RNAi-transgenic tubers exhibited a lower RS content showing 33.7–65.2 % reduction (Fig. 5a) when tubers are stored at 4 °C. Similar results were obtained when measuring the Suc content of these tubers (Fig. 5b), demonstrating that suppression of the SbTRXh1 gene had impact on accumulation of Suc and RS in potato tubers subjected to low temperature. However, much difference in RS content between the OE- and WT tubers was not detected at 4 °C after 15 days but a lower Suc content in the former than in the latter was observed, which may reflect a high functional potential of the endogenous SbTRXh1 gene or its homologue in the recipient plant.

In stored RNAi-transgenic tubers, the RS (or Suc) contents were parallel to the transcripts of SbTRXh1 and could be shown by a linear function (Fig. 5d, e). These results indicated a close relationship between SbTRXh1 expression and RS (or Suc) accumulation in WT and RNAi-transgenic tubers. However, when the data of OE-transgenic tubers were added, the correlations between the transcripts of SbTRXh1 and RS (or Suc) contents lost their significance because magnificent abundance of SbTRXh1 transcripts in OE tubers were accompanied by limited accumulation of RS and Suc.

Our results also demonstrated a very significant positive linear correlation between RS and Suc content (Fig. 5f), implying that the degradation of Suc may be a main source of RS accumulation in cold-stored potato tubers.

Discussion

In present study, the 5′- and 3′-fragments of the SbTRXh1 gene were cloned separately. The overlapping sequence showed 100 % identity with the corresponding EST from the SSH library, which confirmed correct cloning of the gene. Results of blastn (nr/nt) analysis in NCBI, sequence search in Pfam, and phylogenetic analysis (Fig. 1) demonstrate that SbTRXh1 is a gene that encodes a new protein of thioredoxin h1 subgroup (Gelhaye et al. 2004) in potato.

Plant thioredoxins were reported to be involved in responses to various biotic and abiotic stresses (Kocsy et al. 2004). In our research, the transcripts of SbTRXh1 and SbTRXh1 protein accumulated in both WT and transgenic tubers stored at 4 °C, indicating that SbTRXh1 is associated with the cold response of potato tubers (Fig. 4). This finding is in accordance with previous reports that thioredoxins could be induced by cold treatment in maize (Kocsy et al. 2004) and rice (Xie et al. 2009).

During cold storage, the starch in potato tubers is converted into hexosephosphate, which can generate Suc 6-phosphate and then Suc (Halford et al. 2011). The accumulation of Suc leads to an increase of cytochylema concentration to cope with the cold. Furthermore, Suc can be degraded into glucose and fructose (Malone et al. 2006) resulting in CIS. In the present research, RS content was increased in the tubers stored at 4 °C (Fig. 5a). More importantly, this increase was significantly correlated to the expression of SbTRXh1 (Fig. 5d), strongly suggesting that the SbTRXh1 gene is involved in regulation of CIS in potato tubers.

Further analysis of present research showed a logarithmic relationship between transcripts and its protein content of the SbTRXh1 gene (Fig. 4d), implying that the function of SbTRXh1 in potato CIS may be mainly regulated at the transcriptional level. Protein content of PsTRXh1, an isoform thioredoxin h group in Pea (Pisum sativum) L. cv Lincoln, showed a positive correlation with transcripts of PsTRXh1 in leaves and roots under oxidative stress (Traverso et al. 2007). In potato, protein content of CDSP 32, a drought-induced thioredoxin, was increased with the accumulation of transcripts when plants were exposed to high light/low temperature (8 °C) or γ-irradiation (Broin et al. 2000). However, Broin and colleagues also found that treatment with 10 μM methyl viologen led to a sharp increase of both transcripts and protein content of CDSP 32 at 6 h, but subsequently, the protein content continued to rise whereas the transcripts showed a consecutive drop. These results may suggest different regulation models of thioredoxin in response to varied stimulations.

Decreasing in RS accumulation in RNAi-transgenic tubers was accompanied by a declined Suc (Fig. 5a, b) and a stable starch content (Fig. 5c), demonstrating that the CIS of potato tubers might mainly result from Suc degradation. This relationship was further confirmed by analyzing all type of tubers in present research which still showed a significant relationship between RS and Suc (Fig. 5f) but not statistically significant between RS and starch. Therefore, it is reasonable to speculate that SbTRXh1 may function in Suc–RS pathway to regulate potato CIS. It is known that thioredoxin functions by reducing disulfide bridge of target protein with a CXXC active site. Some enzymes involved in Suc–RS pathway have been identified to be potential target proteins of thioredoxin, such as ADP-glucose pyrophosphorylase and phosphoglucomutase (Lindahl and Florencio 2003), UDP-glucose pyrophosphorylase (Wong et al. 2004; Alkhalfioui et al. 2007) and sucrose synthase 1 (Hägglund et al. 2008). There could be more target proteins not revealed so far that are worthy of further investigation to elucidate the mechanism by which SbTRXh1 implements its function in the CIS process of potato tubers.

When stored at 4 °C after 15 days, there was no significant difference in RS content observed between OE-transgenic and WT tubers (Fig. 5a), and the Suc content in the former was slightly lower than that in the latter (Fig. 5b). This phenomenon might have resulted from a function potential of endogenous SbTRXh1 gene and its homologues. Although little information is available for potato TRX h isoforms, their complex composition in other plant species have been reported, for example, there are 9 TRX h isoforms in rice (Nuruzzaman et al. 2008), 11 in A. thaliana (Meyer et al. 2008) and 12 in Medicago truncatula (Renard et al. 2011). Another possible explanation could be that SbTRXh1 may not play a direct role in CIS regulation; looking into its target proteins could be necessary to reveal this speculation. Characterization of molecular modifications (e.g., accumulation of sucrose precursors, fructose and glucose) in RNAi-transgenic lines would be helpful in the investigation of target proteins of SbTRXh1 in the regulation of potato CIS. In barley grain, over-expression of a TRX h with a signal peptide sequence for targeting to the protein body using an endosperm-specific B1-hordein promoter led to release of starch-hydrolyzing enzymes and reducing storage proteins (Cho et al. 1999; Wong et al. 2002). In our research, there was no significant difference in starch content found between OE-transgenic and WT tubers, suggesting that the functions of TRX h isoforms may be isoform-specific.