Introduction

Drought stress is one of the main abiotic stressors that limit plant growth and production (Yang et al. 2010; Wang et al. 2011; Maqbool et al. 2009). Water deficiency is a threat to agriculture because of the inability to control water availability without the use of irrigation strategies (Prabu et al. 2011). Polyols, with their low molecular weight, high solubility and non-reducibility, are suitable substrates for translocation enzymes and their accumulation results in improved tolerance to abiotic stress (Noiraud et al. 2001).

Sorbitol is the main polyol that is produced by the Rosaceae family, whose members include almond (Prunus dulcis), apple (M. domestica), cherry (P. cerasus), peach (P. persica) and pear (Pyrus communis) (Loescher and Everard 1996; Soria-Guerra et al. 2011). Numerous studies have shown that abiotic stress induces the accumulation of sorbitol. In mature apricot (P. armeniaca) leaves, up to 65–75% of translocated carbon is from sorbitol. In the phloem of apples, sorbitol comprises about 80% of translocated carbohydrates (Kühn et al. 1999; Lalonde et al. 2003). Osmotic stress promotes sorbitol accumulation in the leaves and roots of loquat (Eriobotrya japonica) trees, and sorbitol content diminishes gradually once the osmotic stress is alleviated (Cui et al. 2003). Low temperature and drought stress also led to an increased concentration of sorbitol in peach leaves (Deguchi et al. 2002; Cui et al. 2004).

Many studies focused on the regulation of sugar transporter levels in plants subjected to abiotic stress. It has been shown that different sugar transporters are up-regulated or down-regulated by wounding, elicitor treatment, pathogen infection (Truernit et al. 1996), salt stress (Pommerrenig et al. 2007; Alexandra et al. 2006), drought stress (Alexandra et al. 2006), light (Kϋhn et al. 1997; Stadler et al. 2003) and phytohormones (Ehneβ and Roitsch 1997; Fillion et al. 1999). mRNA levels of STP4, which encodes an enzyme that catalyzes monosaccharide import into classic sinks, increased when Arabidopsis was wounded. The function of increased STP4 concentration is to meet the increased carbohydrate demand of cells responding to environmental stress. When celery plants were subjected to salt stress conditions, they favored the accumulation of mannitol (Truernit et al. 1996). In Plantago major, the mRNA levels of vascular sorbitol transporters PmPLT1 and PmPLT2 were up-regulated in phloem sap following salt treatment (Pommerrenig et al. 2007). The studies cited above indicate that different sugar transporters each have their own unique response to environmental stress. For this reason, many studies are focusing on polyol transporters in response to different stressors. Little is known about the relationship of sorbitol transporters and sorbitol accumulation in members of the Rosaceae family subjected to drought stress. Sorbitol transporters have been identified in various plants including sour berry (P. cerasus), apple (M. domestica), common plantain (P. major) and Arabidopsis (Arabidopsis thaliana) (Gao et al. 2003, 2005; Watari et al. 2004; Fan et al. 2009). MdSOT1 and MdSOT2 were cloned from apples while MdSOT3, MdSOT4, MdSOT5 and MdSOT6 were isolated from apple source leaves. Expression of MdSOT1 and MdSOT2 was seen in all sink tissues except watercore-affected fruit tissues. The K m values of MdSOT1 and MdSOT2 were 1.0 and 7.8 mM, respectively (Gao et al. 2005). Apparent K m values of MdSOT3 and MdSOT5 for sorbitol were estimated to be 0.71 and 3.2 mM, respectively. MdSOT3, MdSOT4 and MdSOT5 were expressed mostly in vegetative organs while fruits showed little or only weak expression of MdSOTs. Interestingly, MdSOTs expression increased with leaf maturation (Watari et al. 2004).

To understand if the expression levels of sorbitol transporter genes contribute to sorbitol accumulation and drought tolerance, we investigated the relationship between MdSOT3, MdSOT4 and MdSOT5 transport levels and the sorbitol content in roots, phloem and leaves. These tissues were isolated from micropropagated apple plants and subjected to different drought intensities by regulating the osmotic potential of PEG 6000 solutions.

Materials and Methods

Plant Material, Growth and Experimental Conditions

Micropropagated ‘Fuji’ apple plants (Malus domestica Borkh. ‘Fuji’) were pre-cultured in pots with 1/2 Hoagland nutrient solution for 10 days and transferred into Hoagland solution. The culture conditions were 24–26°C, 60% relative humidity under 10 h/14 h (day/night). Plants with heights ranging from 25 to 30 cm and with uniform growth were selected for the study.

Plant Material Treatments

PEG 6000 was used as the drought stress factor as reported by Hsiao (1973). The control plants received Hoagland solution with sterile water in place of PEG 6000 (CK). Nutrient solutions contributing mild stress (mild drought stress [MIS]) and severe stress (severe drought stress [SES]) were adjusted to osmotic potentials of −0.75 and −1.5 MPa, respectively. A randomized complete block design was used, with three replicates consisting of 108 plants per block. Samples were collected at 0, 2, 6, 12, 24 and 48 h following the experimental drought stress treatments and control treatment (CK). Plant materials were separated into leaves and phloem preps of stems and roots. Leaf numbers 9, 10 and 11 were collected from the shoot apex (leaf number 1 was the youngest and number 20 was the oldest). All leaves had an approximate mass of 0.1 g. The phloem preps of the stem were collected from leaf numbers 6 to 14 excluding the bark. Additionally, young roots were cut and collected. The samples, which were used for RT-qPCR, were immediately transferred into liquid nitrogen and stored at −80°C. Samples for HPLC and relative water content (RWCs were used as soon as they were collected.

Analysis of Relative Water Content

The RWC of the micropropagated apple leaves was measured as previously described by Yamasaki and Dillenburg (1999). Source leaves were collected and immediately weighed (fresh mass [FM]). The leaves were then submerged in distilled water inside a closed petri dish at room temperature until solutions became saturated (turgid mass, TM). Leaf samples were placed in a heated in oven at 80°C for 48 h in order to obtain the dry mass (DM). Values of FM, TM, and DM were used to calculate RWC using the following equation: \( {\text{RWC}}\left( \% \right) = \left[ {\left( {{\text{FM}} - {\text{DM}}} \right)/\left( {{\text{TM}} - {\text{DM}}} \right)} \right] \times 100 \).

Apple RNA Extraction and Cloning of Sorbitol Transporter Genes

For RT-qPCR, the RNA of samples from leaves, phloem and roots were extracted using the hot borate protocol, according to Wan and Wilkins (1994). cDNA sequences of the three sorbitol transporters were amplified from total RNA isolated from micropropagated apple leaves with gene-specific primers. The primers were designed with Primer 5.0 according to the reported sequences in NCBI. The primer sequences are as follows: SOT3f: 5′-AGAGCATGACGGCAGTGGAC-3′ SOT3r: 5′-ATCCTCTGGAGATTCACACACAA-3′, SOT4f: 5′-ATCGGCACCACTAACTTATCTCC-3′, SOT4r: 5′-TAACCTGTTCCGCTGTCTGC-3′, SOT5f: 5′-AAGATGGCTGACCGGACAACT-3′, SOT5r: 5′-AGCAGAGTAAGACGAGGAACATA-3′. PCR was performed at 94°C for 5 min, and then incubated using a multi-step program (94°C, 30 s; 60°C, 40 s; 72°C, 50 s) for 30 cycles.

Analysis of Sorbitol Content in Micropropagated Apple Plants

Sorbitol was extracted from micropropagated apple leaves, phloem and roots according to the protocol published by Li and Li (2007a), and sorbitol content was determined by HPLC. Separations were done by chromatography with an SCR-101 C column (7.9 × 30 mm ID, 5 μm) (Shimadzu Corporation, Kyoto, Japan). The column was placed in an oven set to 45°C. The injection volume was 20 μl for all the runs. The HPLC system consisted of a WATERS-410 series pumping system (Shimadzu Corporation) fitted with a WATERS-2996 PDA detector. The flow rate of the solvent was adjusted to 1 ml/min and consisted of degassed, distilled water.

Analysis of MdSOT3, MdSOT4 and MdSOT5 by Real-Time Quantitative PCR

The probing primers for RT-qPCR were designed with Primer 5.0 software. The internal standard used was the apple 18S ribosomal RNA gene. The primer sequences used are as follows: MdSOT3f: 5′-CGTTGTTACTGACCAGCGTG-3′, MdSOT3r: 5′-CCAGGGGACTGAACCTTTGT-3′, MdSOT4f: 5′-GTGGAGAAAAGCCAATAAACTG-3′, MdSOT 4r: 5′-ATATTCGCTCACTAAGCCAAGA-3′, MdSOT5f: 5′-GATAATACAGTCTCCAGCCAACC-3′, MdSOT5r: 5′-CCAAAATAGCACAAGCAAAGG-3′, 18 s f: 5′-AAACGGCTACCACATCCA-3′, 18 s r: 5′-CACCAGACTTGCCCTCCA-3′. The total volume of each RT-qPCR reaction was 50 μl: 25 μl of 2× Hotstart Fluo-PCR mix, 1.0 μl of 3′ primers and 5′ primers (25 μM), 2 μl of cDNA and 20.7 μl of DEPC-treated ddH2O. All amplicons had varying lengths within 250–300 bp. PCR was performed with a FTC-2000 rapid thermal cycler (Funglyn, Toronto, Canada) and all reactions were prepared with the Shine Probe RT-qPCR Master Mix Kits (Shine, Shanghai). PCR reaction conditions are as follows: 94°C for 4 min and 40 cycles of 94°C for 30 s, 60°C for 30 s and 72°C for 30 s, 72°C for 10 min.

Statistical Analysis

All analyses were performed with version 13.0 of the SPSS software. Differences between treatments were analyzed by Duncan’s multiple range test at a probability level of 0.05.

Results

Changes in the RWC of Micropropagated Apple Leaves under Drought Stress

RWC remained at stable, high levels, between 89% and 92%, in CK leaves during the treatment period (Fig. 1) while the RWC of drought stress plants declined rapidly within 12 h (from 90% to 65% in MIS and from 90% to 48% in SES, respectively). Between 12 h and 48 h, the RWC declined only slightly. The RWC of apple leaves under MIS and SES remained at 60% and 42% at the end of the treatment, respectively.

Fig. 1
figure 1

Effects of drought stress on RWC of micropropagated apple plant leaves. Samples were collected at 0, 2, 6, 12, 24 and 48 h. Values are mean ± SE for three replicates. Statistically significant differences among treatments at each time are indicated by vertical bars (P < 0.05). Filled circle normal water conditions (CK), empty circle mild drought stress (MIS), inverted filled triangle severe drought stress (SES)

Full size image

Alterations in Sorbitol Content in Response to Drought Stress

The sorbitol content of micropropagated apple roots, phloem and leaves under the two experimental drought stress treatments increased rapidly (Fig. 2). In roots, this increase was significant within 2 h of the MIS and SES treatments and maintained a steady increased rate up to 12 h, particularly under SES. At the end of drought stress treatment, the sorbitol content of the MIS and SES groups in roots was 25 and 30.7 mg/g DW, respectively, while the content was 15.2 mg/g DW in CK roots (Fig. 2a). In phloem, significant differences compared to CK were observed at 6 h following drought stress treatments (Fig. 2b). Sorbitol content in phloem under MIS (50 mg/g DW) and SES (71 mg/g DW) were much higher than controls (38.6 mg/g DW) at the end of the treatments. Significantly higher contents of sorbitol were found in SES leaves compared to MIS leaves at 6 h after initiating the drought stress, and these differences were maintained throughout the course of the experiment. The sorbitol content of leaves under MIS and SES was much higher than that with CK treatment at 48 h (1.3- and 1.8-fold compared to CK levels, respectively) (Fig. 2c).

Fig. 2
figure 2

Sorbitol content in roots, phloem and leaves of micropropagated ‘Fuji’ apple plants subjected to varying degrees of drought stress. Values are mean ± SE for three replicates. Statistically significant differences among treatments at each time are indicated by vertical bars (P < 0.05). Filled circle normal water conditions (CK), empty circle mild drought stress (MIS), inverted filled triangle severe drought stress (SES)

Full size image

Changes in Sorbitol Transporter Gene Expression under Drought Stress

The sequences of the sorbitol transporter genes isolated from apple leaves were consistent with the reported gene sequences in NCBI: MdSOT3 (AB125646.1), MdSOT4 (AB125647.1), and MdSOT5 (AB125648.1). RT-qPCR results indicate that the expression levels of MdSOT3 during drought stress treatments increased rapidly and were significantly higher compared to CK in all organs (Fig. 3). This increased was more pronounced in plants treated with SES. The expression levels of MdSOT3 in roots remained at high levels after 6 h under SES and MIS, but in the CK treatment, these levels dropped to initial values after 24 h. The mRNA of MdSOT3 in roots achieved the highest level under SES at 12 h, and this was 4.6-fold higher than that in CK treatment. At the end of the drought stress treatments, MdSOT3 expression levels in MIS and SES were 2.3- and 11.9-fold compared to controls, respectively (Fig. 3a). The changes in MdSOT3 expression levels in apple phloem and leaves were similar to those in roots (Fig. 3b and c).

Fig. 3
figure 3

Effect of drought stress on MdSOT3 gene expression in micropropagated ‘Fuji’ apple plant roots (a), phloem (b) and leaves (c). Values are mean ± SE for three replicates. Statistically significant differences among treatments at each time are indicated by vertical bars (P < 0.05). Filled circle normal water conditions (CK), open circle mild drought stress (MIS), inverted filled triangle severe drought stress (SES)

Full size image

MdSOT4 expression in roots was paralleled by that of MdSOT3 (Fig. 4a), but the mRNA levels of MdSOT4 were much higher than those of MdSOT3 and MdSOT5. MdSOT4 expression, however, declined in phloem under MIS throughout the course of the experiment. At 6 and 12 h, MdSOT4 in phloem was reduced to 91% and 71% of CK treatment levels under SES (Fig. 4b). The most striking alteration was noted in phloem: the expression was dramatically increased at 24 and 48 h (about 1.6-fold compared to the CK treatment). No obvious change between MdSOT4 expression levels of phloem under SES at 24 and 48 h was observed (Fig. 4b).

Fig. 4
figure 4

Effect of drought stress on MdSOT4 gene expression in micropropagated ‘Fuji’ apple plant roots (a), phloem (b) and leaves (c). Values are mean ± SE for three replicates. Statistically significant differences among treatments at each time are indicated by vertical bars (P < 0.05). Filled circle normal water conditions (CK), open circle mild drought stress (MIS), inverted filled triangle severe drought stress (SES)

Full size image

The expression levels of MdSOT4 in leaves were reduced under MIS at 6 h and 12 h but was increased at all other time points. As opposed to the MdSOT4 expression levels in phloem, the transcription levels of MdSOT4 in leaves had no significant changes prior to 12 h but increased at 24 and 48 h under MIS (Fig. 4c). MdSOT4 mRNA levels started to increase 2 h post-SES treatment until the end of the experimental run (Fig. 4c). The expression levels of MdSOT5 (Fig. 5) under MIS paralleled the changes in MdSOT3 expression. MdSOT5 expression levels decreased under MIS after 2 h in phloem and leaves, then gradually increased with persistent drought stress (Fig. 5b). The expression pattern of MdSOT5 in SES (Fig. 5) was similar to that of MdSOT3.

Fig. 5
figure 5

Effect of drought stress on MdSOT5 gene expression in micropropagated ‘Fuji’ apple plant roots (a), phloem (b) and leaves (c). Values are mean ± SE for three replicates. Statistically significant differences among treatments at each time are indicated by vertical bars (P < 0.05). Filled circle normal water conditions (CK), empty circle mild drought stress (MIS), inverted filled triangle severe drought stress (SES)

Full size image

The results indicate that the expression level of sorbtiol transporter genes, MdSOT3, MdSOT4 and MdSOT5, is enhanced under conditions of SES with the exception of MdSOT4 transcript levels in phloem at 6 h and 12 h (Fig. 4b). Additionally, the expression levels of MdSOT3 and MdSOT5 were induced by MIS at 6 h until 48 h. Only the expression level of MdSOT4 in phloem decreased under MIS at all times compared to the CK treatment. Initially, under SES, MdSOT4 expression was slightly elevated (1.1-fold compared to the CK treatment). The expression decreased from 6 to 24 h and remained elevated after 24 h when the SES treatment persisted (Fig. 4b).

Discussion

Annually, yield losses due to drought stress exceed the sum of other abiotic stresses (Su et al. 2010). Drought is one of the main environmental factors limiting crop production and plant distribution worldwide (Stolf-Moreira et al. 2011). Polyols are efficient osmolytes, and their accumulation is related to resistance to abiotic stressors (drought, salt and cold) (Hu et al. 2005; Rejskova et al. 2007) and biotic stresses (Stoop et al. 1996). Drought stress potentially disrupts the sorbitol distribution in different organs as a defense mechanism to a disadvantageous environment (Li and Li 2007a). Many plant physiological processes are influenced by abiotic stressors (Gao et al. 2010; Xu et al. 2001; Zhang et al. 2011).

Compared to CK treatment, the sorbitol content of micropropagated apple plants increased under MIS and SES to different extent (Fig. 2), and these observations were in agreement with previous studies (Wang and Stutte 1992; Escobar-Gutiérrez et al. 1998; Li and Li 2005; Meyer et al. 2004). The stress-induced changes in sorbitol content were related to the intensity and duration of the experimental drought stress. Sorbitol accumulated first in roots after a 2-h stress treatment, which was earlier than the response seen in leaves and phloem. This result indicates that micropropagated apple roots may be the most sensitive organ to drought stress, and adjust cell turgor by increasing sorbitol content.

A study in peach fruits (Escobar-Gutiérrez et al. 1998) has shown that a linear increase in S6PDH activity with drought stress correlates with a significant accumulation of sorbitol in phloem sap of severely stressed plants. The results in peach indicate that an up-regulation in sorbitol synthesis allowed sorbitol to accumulate in phloem. The activity of aldose-6-phosphate reductase (A6PR), which is the key enzyme in sorbitol synthesis, increased in expanded leaves of micropropagated apple plants subjected to different intensities of drought stress. The opposite was true for sorbitol dehydrogenase (SDH) (Bianco et al. 2000). The increased activity of A6PR and the decrease of SDH are important factors that contribute to the accumulation of sorbitol (Li and Li 2005).

Many physiological processes in plants are influenced by stress (Gao et al. 2010; Xu et al. 2001; Zhang et al. 2011). Plant responses to drought are complex, involving both coordinated gene expression and the integration of multiple biochemical pathways (Stolf-Moreira et al. 2011). The transport of sorbitol is a key factor that may be attributed to sorbitol accumulation under drought stress.

RT-qPCR results showed that mRNA levels of MdSOT3, MdSOT4 and MdSOT5 in roots increased rapidly as soon as the micropropagated apple plants were subjected to SES, and these elevated expression levels persisted until the end of treatment (Figs. 3a, 4a and 5a). This result could be explained by certain characteristics of A6PR. The activity of A6PR was not detectable in roots of micropropagated apple plants, suggesting that sorbitol could not be synthetized in roots (Li and Li 2007b). The sorbitol change in roots can also be attributed to the rapid increase in sorbitol transporter genes. Increased sorbitol transporters load more sorbitol onto phloem and roots. It is likely that apple plants adapt their carbohydrate production capability in order to resist drought stress and this is mediated by the enhancement of sorbitol transporter expression. This result was similar to the mRNA expression change of PmPLT1 and PmPLT2 in Plantago under salt stress (Pommerrenig et al. 2007).

The general change in sorbitol transporter expression levels of MdSOT3 and MdSOT5 in all organs and MdSOT4 in roots under drought stress is consistent with the fact that sorbitol content increased between 0 and 12 h. This shows that the expression of transport genes was increased with continued drought stress time and degree of drought stress.

A study on Mesembryanthemum crystallinum shows that the expression of the putative myo-inositol transporter, Mitr1, increased in all organs, but most dramatically in roots under salt stress (Chauhan et al. 2000). Conversely, the expression of AtSUC3 (a sucrose transporter from Arabidopsis sieve elements and sink tissues) was strongly induced upon wounding of Arabidopsis tissue (Meyer et al. 2004). In Figs. 3 and 5, the high expression levels maintained during the entire treatment process coincides with the responses of the monosaccharide transporter STP4 in Arabidopsis (Gao et al. 2005).

Regulated expression of MdSOT4 differs from that of MdSOT3 and MdSOT5 in phloem and leaves. This observation may be accounted for by the variable activity of the three sorbitol transporters. In yeast, MdSOT3- or MdSOT5-dependent sorbitol uptake is 62 or 17 times higher compared to controls, respectively. Transport activity of MdSOT4, however, is only 1.5 times higher than in controls (Ehneβ and Roitsch 1997). The strong up-regulation of MdSOT3 and MdSOT5 is important for increasing the sorbitol content distribution in different organs. Therefore, the resistant ability of apple plants is improved under drought stress. A similar response also occurs in studies of the tonoplast monosaccharide transporters (TMTs). TMT1 and TMT2 expression is induced by drought, salt, cold treatment and sugar. Environment stress and sugar do not result in the accumulation of TMT3 mRNA (Alexandra et al. 2006).

In conclusion, the mRNA levels of sorbitol transporters were influenced by MIS and SES. All three transporter genes were induced by drought stress with the exception of MdSOT4 in phloem and leaves. MdSOT3 and MdSOT5 play an important role in resisting drought stress and more research is needed to characterize the function of MdSOT4 under drought stress.