Introduction

Desertification and salinization of cultivated land are serious problems in viticulture, which affects the yield and fruit quality of grapevines. It is important to rapidly develop new grapevine genotypes that are tolerant to various abiotic stresses. Physical and chemical methods have been employed to mutate plant cells, and some Pro analogs are used as selection agents. Some variations show varied degrees of tolerance to abiotic stresses, usually accompanied by the ability to produce excessive Pro (Xin and Browse 1998; Tantau et al. 2004; Yang et al. 2009). Hydroxyproline, a toxic analog of Pro, was used as a selection agent to screen Pro overaccumulation cell lines eventually acquiring heritable and increased tolerance to salt stress (Fuller et al. 2006; Dörffling et al. 2009). By using Hyp as selection agent, Hyp-tolerant cell lines of crops have been obtained, such as Lucerne (Li et al. 2008). These cell lines exhibited improved stress tolerance ability, positively correlated to intracellular Pro content (Tantau and Dörffling 1991; Anjum and Villiers 1998; Tantau et al. 2004). This breeding strategy has not been applied in grapevines, where traditional hybridization breeding is the normal method to introduce new traits or create new germplasms. Therefore, we set to try this breeding approach in grapevines to rapidly select new salt- or drought-tolerant grapevine genotypes.

Several physiological and biochemical changes happen when plants undergo osmotic stress. Some compatible solutes, such as polyols and proline (Pro), are synthesized and accumulated in plant under abiotic stress (Delauney and Verma 2002). Plants respond to abiotic stresses, usually through intracellular Pro accumulation to protect protein structure and activities and prevent cell damage by forming hydrophilic colloids in aqueous media (Goring and Thien 1979; Parvanova et al. 2004a, b). Intracellular Pro can be synthesized from glutamate or ornithine (Orn) in plants (Verbruggen and Hermans 2008; Szabados and Savouré 2010). △1-Pyrroline-5-carboxylate synthase (P5CS, EC1.2.1.41) catalyzes glutamate to form pyrroline-5-carboxylate (P5C) and then reduced by △1-pyrroline-5-carboxylate reductase (P5CR) to form Pro (Hu et al. 1992). P5CS appears to catalyze the rate-limiting step in Pro biosynthesis pathway that the expression of P5CS gene plays a key role in Pro levels. The enzyme orn-δ-amino-transferase (OAT) can catalyze the transamination of Orn to synthesize Pro. Pro degradation is catalyzed by two sequential enzymes: Pro dehydrogenase (ProDH, EC1.5.99.8) that acts on P5C, and P5C dehydrogenase (P5CDH. EC1.5.1.12) catalyzes conversion of P5C to glutamate. Plants under abiotic stress have higher expression levels of P5CS gene while the ProDH expression is inhibited (Yoshiba et al. 1997). P5CDH also plays an important role in preventing cell damages from intermediary metabolites such as P5C (Deuschle et al. 2001; Rizzi et al. 2015). Therefore, Pro level in plants appears to be regulated mainly by P5CS and P5CDH.

Vitis vinifera Chardonnay, one of the most popular white wine grape cultivars, is sensitive to abiotic stresses (Cavagnaro et al. 2006; Vincent et al. 2007). The goal of this study was to develop grape plants with improved tolerance to salt stress. In this study, Hyp-resistant calli of Chardonnay were obtained through 60Co irradiation and Hyp selection, which contained higher Pro content than the normal calli. The intracellular Pro content of new plantlets was measured under salt stress. In addition, the expression levels of genes related in Pro biosynthesis pathway during Chardonnay subjected to salt stress was evaluated using qRT-PCR.

Materials and methods

Pre-embryogenic culture and growth conditions

Vitis vinifera Chardonnay pre-embryonic calli (PEs) were obtained from the Center for Viticulture and Small Fruit Research, Florida Agricultural and Mechanical University. The calli were cultured on Murashige and Skoog (MS) basal medium with vitamins (Murashige and Skoog 1962) for long-term maintenance (LTM), which contained 2.5 μM 2,4-D, 5 μM 6-benzylaminopurine, 0.1 mg L−1 inositol, 20 g L−1 sucrose, and 3 g L−1 phytagel. All calli were cultured in 9-cm Petri dishes at 25 °C and 16 h (light intensity 70 μmol s−1 m−2) photoperiod cycle.

Hyp-resistant calli selection

PEs of Chardonnay were proliferated on a liquid maintenance (LM) medium consisting of MS basal medium supplemented with 18 g L−1 maltose, 0.37% (v/v) glycerol, and 5 μM β-naphthoxyacetic acid (pH = 5.8) and incubated in a constant temperature shaker maintained at 25 °C and 150 rpm. PEs were collected and irradiated with 60Co γ-rays. The irradiations were carried out at 10 Gy min−1 dosage rate, with a 2 × 2 cm2 irradiation area and 80 cm source skin distance of 10, 40, 60, 100, and 500 Gy from the cobalt source. After irradiation, PEs were immediately placed on LM medium and incubated for 14 days in darkness. Following this, these PEs were transferred to a semi-solid LTM selection medium with or without 10 mM Hyp (LTM-H) for 2-year recurrent selection. Some Hyp-based resistant (HR) calli were selected and further cultured on LTM medium to recover and grow. HR calli were regularly sub-cultured on LTM-H medium every 28 days. Stable HR calli were then transferred onto somatic embryo multiplication (SEM) medium for further differentiation.

Testing Hyp tolerance

To test HR calli tolerant to Hyp, normal Chardonnay calli and HR calli were both transferred to LTM medium supplemented with 1 or 4 mM Hyp in 50 ml flasks. Ten embryo clusters, with a diameter of 1 cm, were placed in each flask. Three replications of each treatment were maintained at 25 °C in darkness. After 0.5, 1, 1.5, 2, 4, and 7 days of Hyp treatment, 0.3 g fresh calli were sampled and transferred to liquid nitrogen to measure intracellular Pro content. The survival rate (number of viable clusters/total number of clusters × 100%) of embryogenetic calli was calculated after incubating the culture in dark for 28 days.

Regeneration

After culturing on SEM medium for 2 months, Vitis vinifera HR (VvHR) embryogenic calli became granular, and some somatic embryos appeared. For embryo differentiation and maturity, both normal and HR somatic embryos of Chardonnay were transferred onto Lloyd & McCown woody plant medium w/ vitamins (WP medium) supplemented with 600 mg L−1 casein hydrolysate, 100 mg L−1 glutamine, 100 mg L−1 asparagine, 100 mg·L−1 arginine, 0.5 μM NAA, and 2 μM BA (Lloyd and McCown 1981; Agüero et al. 2006). Embryos were cultured at 25 °C, 16 h photoperiod, and with a light intensity of 70 μmol s−1 m−2 and sub-cultured every 28 days. Small plantlets were separated from the embryo clusters and further grown on WP medium to develop into plants.

Salt treatment

Chardonnay normal and HR plantlets were cultured on WP medium supplemented with 5, 7.5, and 10 mM NaCl (salt solution) for 14 days when normal plantlets exhibited some signs of wilting. The aerial parts of plantlets were harvested and placed immediately in liquid nitrogen and stored at − 80 °C for subsequent total RNA isolation and the determination of Pro content.

Determination of Pro content

Intracellular content of free Pro was determined by colorimetry according to Bates et al. (Bates et al. 1973). In brief, 300 mg of the material was boiled in 5 ml of 3% sulfosalicylic acid for 10 min, filtered into a clean tube after chilling. The tube contained distilled water, 2 ml glacial acetic acid, and 2 ml 2.5% ninhydrin (acetic acid:water, 60:40, v/v). The mixture was further boiled for 30 min, and the chromophore was extracted with 4 ml toluene by centrifugation. The supernatant was collected and absorbance was measured by spectrophotometry at 520 nm. Commercially available, pure Pro (Sigma-Aldrich, MO, USA) was used as the standard. All the experiments were carried out using three different calli and plantlets.

qRT-PCR analysis

Total RNA of aerial parts of plantlets were extracted using improved CTAB method (Logemann et al. 1987). A UV spectrophotometer was used to determine the RNA content and quality. Samples were treated with DNase using the Turbo DNA-free kit (TAKARA, Japan). DNase-treated total RNA served as the template for the first-strand cDNA synthesis using M-MLV reverse transcriptase (Promega, USA). A control reaction for each sample was performed simultaneously, however, without the reverse transcriptase.

Primer pairs for VvP5CS, VvOAT, and VvP5CDH genes are shown in Table 1 and were used in the RT-PCR. An EF1-α gene cDNA fragment (110 bp) from Pinot Noir was used as the reference. The specificity of PCR products was validated by dissociation curve analysis and agarose gel electrophoresis. Three biological replicates for each sample were used for qRT-PCR analysis, and three technical replicates were analyzed for each biological replicate. Statistical analysis was performed using SPSS version 19 (ANOVA, SNK test). Statistical analysis (P < 0.05) of Pro content and the expression profiles were performed between HR plants and normal plants.

Table 1 List of the primers of Pro metabolism-related genes for real-time PCR analysis
Full size table

Results

Irradiation and selection of Hyp-resistant PEs

Pre-embryogenic calli of V. vinifera Chardonnay were firstly irradiated with different dosages of 60Co. Following irradiation, these calli were cultured on 10 mM Hyp-added medium for 28 days. We discovered that embryogenic calli irradiated with 100 Gy of 60Co survived on the 10 mM Hyp-added medium. Subsequently, a few HR calli were obtained after extended culture on Hyp-added and Hyp-free medium for 2 years (Fig. 1).

Fig. 1
figure 1

a Normal embryonic calli cultured on Hyp-free LTM medium. bHyp-resistant calli obtained after 100 dosages of 60Co irradiation followed by 10 mM Hyp selection (signified by black arrows)

Full size image

To further verify the tolerance of new HR embryogenic calli to Hyp, both HR and normal embryogenic calli were cultured on 1 and 4 mM Hyp medium for 28 days. The results showed that all HR embryogenic calli (100%) were healthy and survived both on the 1 and 4 mM Hyp medium for 28 days (Fig. 2b, d). However, only 56.67% of normal calli survived on 1 mM Hyp medium (Fig. 2a) and none of them survived on the 4 mM Hyp medium (Fig. 2c). These results indicated that the HR calli exhibited higher Hyp resistance ability.

Fig. 2
figure 2

Cells cultured on LTM medium with 1 and 4 mM Hyp for 28 days. a Normal calli on LTM medium with 1 mM Hyp for 28 days. b HR calli on LTM medium with 1 mM Hyp for 28 days. c Normal calli on LTM medium with 4 mM Hyp for 28 days. d HR calli on LTM medium with 4 mM Hyp for 28 days

Full size image

Effect of Hyp treatment on intracellular Pro content

The intracellular Pro content of both normal and HR calli was investigated after culturing on Hyp-supplemented LTM medium (Fig. 3). When cultured on Hyp-free LTM medium, Pro content in the normal calli was determined to be 7.37 μg g−1 Fw. When normal calli were cultured on 1 mM Hyp medium, Pro content increased rapidly and reached peak levels (sevenfold higher than the original level) within 0.5 day before a drastic reduction to reach the original level at 1.5 days and remained constant until 2 days. At 7 days, we observed a remarkable increase, reaching approximately ninefold higher than the original Pro level. Similarly, when normal calli were instead cultured on 4 mM Hyp medium, Pro content also rose rapidly to peak (twofold higher than the original level) at 1 day and then decreased at 1.5 days. However, it rapidly spiked to reach the second peak level (threefold higher than the original level) at 4 days. Then, it gradually decreased and remained constant twofold higher than the original Pro level at 7 days. At this stage, we observed that some normal calli became brown and infiltrating.

Fig. 3
figure 3

Pro content of normal calli and HR calli cultured on LTM medium supplemented with 1 mM and 4 mM Hyp. Data are expressed as mean ± SD, n = 3. Independent sample t test of variance, *P < 0.05

Full size image

Different patterns of Pro accumulation were observed in HR calli cultured on Hyp-added LTM medium. When HR calli were cultured on Hyp-free LTM medium, Pro content was only 10.31 μg g−1 Fw, slightly higher than that in the normal calli (7.37 μg g−1 Fw). When HR calli were cultured on 1 mM Hyp LTM medium, Pro content increased slowly before 0.5 day, but increased rapidly until 7 days, considerably higher than that in the normal calli. Similarly, when HR calli were cultured on 4 mM Hyp medium, Pro accumulation was at low levels until 1.5 days, after which it increased rapidly to 124.56 μg g−1 Fw at 7 days, which was approximately ninefold higher than that observed in the normal calli. Furthermore, HR calli remained healthy at this stage.

Whether 1 or 4 mM Hyp was added to the medium, Pro content in HR calli generally increased and responded quickly by changing concentration and remained at higher levels over time. This contrasts with the accumulation pattern observed in the normal calli that showed an increase in Pro content at first, but then decreased over time, and slowly increased again. Both normal calli and HR calli responded faster on 1 mM Hyp medium than on 4 mM Hyp medium. One possible explanation for this rapid response with low Hyp content is that the calli may need more time to adapt to the stress caused by 4 mM Hyp medium.

Regeneration into plantlets

When Chardonnay normal calli and HR calli were both cultured in the germination medium for differentiation (Fig. 4a, b), normal calli regenerated into small plantlets with two to three leaves at 28 days (Fig. 4c). However, it is approximately 180 days for HR calli to reach the same stage as that of the normal embryogenic calli at 28 days (Fig. 4d), which may be probably due to 60Co irradiation. Similar effects of 60Co irradiation decreasing growth rate have been reported in tall fescue grass (Xiong and Fei 2013). Nevertheless, no obvious morphological differences were observed between plantlets generated from normal embryogenic calli and HR embryogenic calli.

Fig. 4
figure 4

Regeneration of HR embryogenic calli. a HR embryogenic calli cultured on differentiation medium. b Early regeneration of HR embryogenic calli cultured on WP medium. c, d Regenerated HR plantlets cultured on WP medium

Full size image

Effect of salt stress on Pro content

To study tolerance ability of HR plantlets to salt stress, Pro content of normal and HR plantlets was determined when plantlets were treated with 5, 7.5, and 10 mM NaCl for 14 days (Fig. 5). Pro content of HR plantlets remained at approximately constant levels when subjected to different NaCl treatments, whereas Pro content of normal plantlets increased with increase in NaCl concentration. Pro content of normal plantlets was found to be 130.87 μg g−1 Fw before NaCl treatment, which decreased to 56.94 μg g−1 Fw after 5 mM NaCl treatment. This increased to 209.43 μg g−1 Fw with 10 mM NaCl treatment. Notably, Pro content of HR plantlets was significantly lower than that of normal plantlets under 10 mM NaCl treatment. Morphologically, normal plantlets displayed wilting symptoms under 10 mM NaCl treatment for 14 days (Fig. 6c), the roots brown and subsequently the whole plantlets dead. HR plantlets remained healthy throughout the treatment period (Fig. 6d).

Fig. 5
figure 5

Pro content of normal and HR plants cultured on medium containing 5, 7.5, and 10 mM NaCl for 14 days. Data are expressed as mean ± SD, n = 3. Independent sample t test of variance, *P < 0.05

Full size image
Fig. 6
figure 6

The morphology of normal and HR plantlets under salt stress. a Normal plantlets cultured on WP medium without NaCl for 14 days. b HR plantlets cultured on WP medium without NaCl for 14 days. c Normal plantlets cultured on WP medium with 10 mM NaCl for 14 days. d HR plantlets cultured on WP medium with 10 mM NaCl for 14 days

Full size image

Effects of salt stress on the expression of VvP5CS, VvOAT, and VvP5CDH

To determine the relationship between Pro accumulation and transcriptional levels of a few related genes under salt stress, expression levels of VvP5CS, VvOAT, and VvP5CDH genes were assessed by real-time qRT-PCR. As shown in Fig. 7, the expression level of these genes in normal and HR plantlets varied significantly.

Fig. 7
figure 7

Effect of salt stress on VvP5CS, VvOAT, and VvP5CDH expression in normal plantlets and HR plantlets with 5, 7.5, and 10 mM NaCl treatment for 14 days. Data are expressed as mean ± SD, n = 3. Independent sample t test of variance, *P < 0.05

Full size image

When normal plantlets were subjected to varying levels of NaCl treatments, we noticed a positive correlation between VvP5CS upregulated expression levels and NaCl concentration. On the contrary, VvOAT upregulated expression level had a negative correlation with NaCl concentration, indicating that Pro synthesis in normal plantlets was mainly channeled through the OAT pathway under low NaCl content and PSCS pathway under high NaCl concentration. The expression levels of VvP5CDH gene in normal plantlets were upregulated with concomitant increase in NaCl concentration. Combined with the above results on Pro content of normal plantlets under NaCl stress, the results of gene expression reveal that the upregulated expressions of VvP5CS and VvOAT genes promoted Pro accumulation to overcome NaCl stress. Expression of Pro metabolism-related gene VvP5CDH was upregulated and in turn decreased Pro accumulation when Pro was excessive, through the Pro feedback mechanism.

Compared to normal plantlets, the expression levels of VvP5CS and VvOAT genes in HR plantlets were considerably upregulated under 10 mM NaCl at 14 days, whereas they were not significantly affected by low NaCl concentrations (Fig. 7). The expression levels of VvP5CDH in HR plantlets were upregulated under low NaCl contents, whereas the upregulation levels of VvP5CDH expression were inhibited by high NaCl contents. Taken together, these results of Pro content in HR plantlets under NaCl treatment reveal that when Pro accumulation was insufficient to help defense severe NaCl stress, the expression of Pro synthesis-related genes VvP5CS and VvOAT was upregulated and the expression levels of VvP5CDH gene remained unchanged probably to favor Pro accumulation and subsequently increase the stress tolerance levels of HR plantlets under high salt treatment.

Discussion

With more and more agricultural land salinization, the development of high-yielding and salt-tolerant cultivars has become an important breeding objective. Mutagenesis can produce new variants with new traits which are not found in the nature and can thus enrich the genetic and breeding resources available for salinization resistance and other desirable traits (Liu and Zhen 2004; Jain 2001; Predieri 2001). 60Co-γ as a mutagen has direct mutagenic effects on DNA, as it can cause DNA breakage at high concentrations and interfere with DNA replication and synthesis at low concentrations by inhibiting DNA polymerases and ligases activity. Owing to its safety and high efficiency, 60Co-γ is being widely used in crop development programs (Weng et al. 2005; Xu et al. 2016). Given these advantages, we selected 60Co-γ to create mutants of Chardonnay pre-embryonic calli for breeding new genotypes with abiotic stress resistance.

To breed new germplasm resources, the combination of in vitro mutagenesis and directed screening for mutants via tissue culture can save labor and other resources (Ahloowalia et al. 1998). The proline analog Hyp has been used in tissue culture to select for stress-tolerant cell lines of many crops (Li et al. 2008; Zhang et al. 2008; Sui et al. 2015). Hydroxyproline-resistant calli could be obtained with higher stress tolerance ability, for example, Hyp was successfully used to select Pro-overproducing lines in potato and spring wheat to improve frost tolerance (Van Swaaij et al. 1986; Tantau and Dörffling 1991). Additionally, researchers used somaclonal variations to screen Hyp-tolerant mutants and finally to obtain regenerated plants with increased cold resistance (Zhang et al. 2008). The use of pingyangmycin-based in vitro mutagenesis in combination with directed screening with Hyp is effective for the creation of potential drought-tolerant mutants of peanut (Sui et al. 2015). In our work, grape somatic embryogenic calli after irradiation with 60Co rays were screened with Hyp, and HR plantlets with higher salt-tolerance ability were obtained.

In a wide range of environmental stresses, free proline accumulates in plant cells. Proline is a highly water soluble amino acid (Manafi et al. 2015) and is involved in imparting sustainable tolerance in plants (Parvanova et al. 2004a). Although Pro accumulation in plants plays an important role in osmotic balance, overaccumulation of Pro could result in slow plant growth and prolonged the florescence (Mattiol et al. 2009). Pro toxicity could cause ultra-structural alterations in chloroplast and mitochondria as well as several features of programmed cell death (Hare et al. 2002; Deuschle et al. 2001; Lehmann et al. 2010). Therefore, intracellular free Pro content can be used as an index to assess the tolerance capacity of plants to abiotic stress. Our results showed that normal calli under Hyp stress and normal plantlets under NaCl treatment were both sensitive to the environment and accumulated Pro at a faster rate, whereas HR calli and plantlets could sustain a stable physiological status under abiotic stresses, better than normal calli and plantlets. These results revealed that it was important for plants to maintain the internal environment homeostasis. The accumulation of Pro may play an important role in adjusting the osmotic potential to tolerate abiotic stress (Ashraf and McNeilly 2004).

Pro content of HR plantlets under NaCl treatments remained relatively stable under low salt stress, whereas Pro content of normal plants was significantly higher than that of HR plantlets under 10 mM salt stress. We speculated the reasons that caused different responses between normal plantlets and HR plantlets. Under salt treatment which simulated the natural environment, plantlets suffered salt stress not only from penetration imbalances but also from toxic effects of Na+/Cl− ions. In a saline medium, cells need to acquire essential nutrients from a milieu with a preponderance of ions that are potentially toxic (Chowdhury et al. 1995). To some extent, Pro ameliorated the deleterious effect of NaCl and favored better growth of plants. Some normal plantlets cultured on NaCl-added medium showed higher intracellular Pro content to maintain the osmotic balance. As HR plantlets may do not have the same sensitivity to salt as the normal plantlets, Pro content in HR plantlets was lower than that in normal plantlets.

Most of the research work concerning Pro metabolism in plants has been focused on its accumulation in vegetative tissues in response to abiotic stresses such as drought and salinity. Stress-induced Pro accumulation occurs predominantly through enhancing its biosynthesis from glutamate via the pathway catalyzed by enzymes P5CS and P5CR (Boggess et al. 1976; Rhodes and Bressan 1986). In our study, under low contents of NaCl, the expression levels of both VvP5CS and VvOAT genes in HR plantlets were maintained at constant levels, whereas the expression of VvP5CDH was upregulated, which showed that the increase in Pro degradation was more prominent in HR plantlets, maybe to maintain the stabilization of internal environment under low salt conditions. However, under high salt conditions, the stabilization of internal environment in HR plantlets was achieved through the upregulation of Pro synthesis genes VvP5CS and VvOAT and the inhibition of VvP5CDH upregulation expression. When normal plantlets accumulated sufficient amounts of Pro under low NaCl conditions for 14 days, the expression of VvP5CDH increased to trigger Pro degradation, through Pro feedback inhibition mechanism. This phenomenon was slow or delayed under high salt conditions, which illustrates that normal plantlets are sensitive to salt stress. High NaCl content causes plant etiolation and death. It is possible that Pro degradation pathway could be suppressed when homeostasis of Pro cycle is disturbed, leading to a distorted physiological state and damage to tissues.

Conclusion

Selection for Hyp resistance has been investigated as a possible alternative strategy for developing salt tolerance in plants (Dix et al. 2013). HR plantlets resistant to 10 mM Hyp showed improved resistance to NaCl. Proline accumulation as a basis for their Hyp resistance is yet to be confirmed. Tolerance to proline analog associated with proline accumulation is one possible approach but could be associated with agronomically undesirable traits such as loss of vigor (Hanson et al. 1979) and attraction of pests (Haglund 1980). However, the latter possibility has been excluded in proline-accumulating barley mutants (Bright et al. 1982).

In summary, we obtained several Hyp-resistant grape plants which could accumulate Pro without feedback inhibition. We observed strong correlation between Pro accumulation and VvP5CDH gene upregulation under salt stress. In addition, Pro could cause native effects on growth and regeneration of cells, though playing a critical role in protecting physiological activity during stress. Finally, 60Co irradiated HR plants showed higher tolerance to abiotic stress than normal plants. Further research is necessary to fully understand the stress tolerance mechanism in HR plantlets and the effects of Pro on plant growth at a molecular level.