Introduction

Carnation (Dianthus caryophyllus) is one of the major global floricultural crops. It is a member of the family Caryophyllaceae and belongs to the genus Dianthus which have been recorded more than 300 species (Muneer et al. 2016). Carnation is widely cultivated for the cut flower and bedding flower industries due to its attractive characteristics, including both single and multi-color florets, fragrance, flower size, and longevity (Ali et al. 2008). Tissue culture is a powerful technique that is used for the rapid propagation and production of carnation plantlets. However, the artificial in vitro environment used in the tissue culture process results in several problems, among which hyperhydricity is one of the most concerning (Gao et al. 2017). Hyperhydricity causes malformed in plantlet morphology, including leaves and stems that are thicker, wrinkled or curled, translucent or glassy, and brittle (Piqueras et al. 2002; Wang et al. 2007). These plantlets survive very poorly when they are subcultured or transplanted, which results in considerable commercial losses to the plant mass propagation industry. Establishing an appropriate method to reduce or control hyperhydricity in carnation is thus important.

Rare earth element (REE) is a term generally used to represent the chemical elements in the periodic table belonging to Lanthanide series elements, such as Lanthanum (La), Cerium (Ce), Neodymium (Nd) (Wang et al. 2012). Research has shown that REEs have many biological effects on plant growth, including accelerating cell growth, increasing secondary metabolite synthesis, and enhancing tolerance to adverse environmental conditions (Peng and Pang 2002; Yuan et al. 2002). The REEs La, Ce, and Nd have been reported to have beneficial effects on plants (Wu et al. 2001; Chen et al. 2004). For instance, Xu et al. (2016) recently reported that lanthanum nitrate La(NO₃)₃ and cerium nitrate Ce(NO₃)₃ in appropriate concentrations accelerated the regeneration of the valuable medicinal plant Anoectochilus roxburghii. In this study, we investigated the influence of induction media supplementation with REEs in reducing the rate of hyperhydricity of adventitious shoots during micropropagation in carnation.

Materials and Methods

Medium preparation and culture conditions

Stem segments (1 - 1.5 cm) taken from in vitro-grown 45-d-old carnation (Dianthus caryophyllus) cultivar ‘13827’ were used as the explants in this study. The culture medium consisted of Murashige and Skoog (MS) basal salts and vitamins (Murashige and Skoog 1962) supplemented with 3% (w/v) sucrose, and solidified with 0.8% (w/v) agar. The MS medium was fortified with 1.0 mg·L^-1 6-benzyladenine (BA) and 0.5 mg·L^-1 indole-3-acetic acid (IAA) as the control medium. Different concentrations of La(NO₃)₃, Ce(NO₃)₃, and neodymium chloride (NdCl3) were respectively added to the induction medium in the REE treatments (Table 1). All media were adjusted to pH 5.8 and autoclaved at 121°C for 15 min. In each treatment, 15 explants were used, and the experiments were repeated three times. All cultures were placed at 22 ± 1°C under a 16 h light photoperiod with 45 μmol·m-2·s^-1 photosynthetic photon flux density (PPFD) provided by cool white fluorescent light. Shoots induced from stem segments on induction medium lacking REE served as controls (C). After 28 d, the hyperhydricity rate and shoot induction rate were recorded as follows: hyperhydricity ratio (%) = (hyperhydric shoot number/total shoot number) × 100%, and shoot induction ratio (%) = (the stem segments induced adventitious shoots/total stem segments) × 100%.

Malondialdehyde (MDA) concentrations

The level of lipid peroxidation was measured as the amount of MDA determined by the thiobarbituric acid (TBA) reaction as described by Heath and Packer (1968). Frozen samples (0.5 g) were homogenized with 5 mL 0.1% (w/v) trichloroacetic acid (TCA) and centrifuged for 15 min at 12,000×g. The supernatant (1 mL) with 4 mL of 20% (w/v) TCA containing 0.5% (w/v) TBA made into solution was heated at 95°C for 30 min and then rapidly cooled on ice for 5 min. After centrifugation (12,000×g for 10 min at 4°C), the absorbance of the supernatant was assayed at 532 nm and 600 nm. The value of 600 nm was subtracted from 532 nm and the calculation was performed using the extinction coefficient of 155 mM^-1cm^-1.

Enzyme extraction and assay

For antioxidant enzyme extraction, 0.1 g frozen leaves were homogenized in 1.5 mL of 50 mM sodium phosphate buffer (pH 7.0) for guaiacol peroxidase (GPX) activity assay, and 50 mM sodium phosphate buffer with 1 M EDTA, 0.05% Triton X 100, and 2% polyvinylpyrrolidone for superoxide dismutase (SOD), ascorbate peroxidase (APX), and catalase (CAT) activity assays. The homogenate was then centrifuged at 15,000×g at 4°C for 20 min, and the clear supernatants were obtained for enzyme activity analysis and protein estimation. The protein content of the enzyme aliquot was determined using the Bradford (1976) method. Protein estimation and enzyme assay absorbance were recorded using a UV spectrophotometer (Uvikon 992, Kontron Instrumentals, Milano, Italy).

SOD activity

SOD activity was determined according to the nitro blue tetrazolium (NBT) inhibition method (Giannopolitis and Ries 1977). The 3 mL reaction mixture contained 50 mM phosphate buffer, 75 μM NBT, 13 mM L-methionine, and 0.1 mM EDTA. After the addition of 0 .1 mL of the enzyme extract sample and 4 μM of riboflavin to the buffer, the reaction was initiated by illumination under cool fluorescent white light for approximately 30 min. Tubes containing reaction mixtures lacking sample constituted the control and were maintained in the dark, and the absorbance was monitored at 560 nm as a reference. One unit of SOD activity was defined as the amount of enzyme that results in 50% inhibition of the photochemical reduction of NBT.

CAT activity

CAT enzyme activity was estimated in 50 mM phosphate buffer with 100 μL enzyme extract and 15 mM H₂O₂. The H₂O₂ scavenging activity was recorded at 240 nm for 180 sec at 30 sec intervals (extinction coefficient of 39.4 mM^-1 cm^-1) (Cakmak and Marschner 1992).

GPX activity

GPX activity was measured according to Shah et al. (2001) u sing 0 .1 M p hosphate b uffer containing a 1 00 μ L aliquot, 9 mM guaiacol, and 19 mM H₂O₂. The absorbance at 470 nm was recorded for a duration of 180 sec at 30 sec intervals. One unit of peroxidase was defined as the amount of enzyme that caused the formation of 1 mM tetraguaiacol per min [extinction coefficient (E) = 26.6 mM^-1 cm^-1].

APX activity

For APX enzyme activity, ascorbate (0.5 mM) was used as t he s ubstrate i n a 3 mL m ixture ( containing 5 0 mM phosphate buffer and 15 mM H₂O₂. The absorbance was measured at 290 nm at 30 sec intervals for 180 sec (extinction coefficient of 2.8 mM^-1 cm^-1) (Nakano and Asada 1981).

Superoxide radical (O_2-) staining

The O_2- staining method of Ramel et al. (2009) was modified as follows. The leaves of non-hyperhydric plants and hyperhydric plants were immersed and infiltrated under vacuum with NBT (0.1%) in phosphate buffer (50 mM, pH = 7.2) containing 4 mL riboflavin (2 mM), and then incubated with shaking (100 rpm/min) at room temperature for 4 h in the dark. Following three wash steps with distilled water to remove the NBT solution, the leaves were then de-colorized by soaking in ethanol (100%) at 95°C for 15 min. The O_2- was visualized as the blue color produced by NBT precipitation.

Results

Effect of REEs on the hyperhydricity ratio and induction rates of carnation

After the stage 2 nodal explants were subcultured in REE medium for four weeks, the effect of the REEs on the hyperhydricity rate and shoot induction rate were assessed (Table 1). Hyperhydricity was observed based on the typical ‘glassy’ appearance of the shoots, which were characterized as being elongated, thick, translucent, and brittle (Fig. 1). Hyperhydricity in the medium without REEs was observed in 68.9% of the cultures. With the addition of 0.05-0.15 mM La³⁺, Ce³⁺, and Nd³⁺ to the MS + BA 1.0 mg L^-1 + IAA 0.5 mg L^-1 media, hyperhydricity was reduced, with the lowest percentage (42.2%) observed in the medium supplemented with 0.05 mM Ce(NO₃)₃. The shoot induction rate was slightly reduced at REE concentrations ranging from 0.05 mM to 0.15 mM, with the exception of Nd³⁺ at 0.15 mM. The shoot induction rate was not significant between 0.05 mM and 0.1 mM La³⁺, Ce³⁺, Nd³⁺, and the control, but decreased in the 0.15 mM La³⁺, and Nd³⁺ treatments, indicating that REEs were influenced at high concentrations.

Effect of REEs on soluble protein concentration

The soluble protein concentration of the non-hyperhydric shoots was higher than the hyperhydric shoots among the different treatments (Fig. 2). In the non-hyperhydric shoots, the protein content in the control was 9.57 mg×g^-1 FW, which was lower than that observed at 0.05 mM La³⁺ and Ce³⁺ (10.38 and 10.47 mg×g^-1 FW) and 0.1 mM and 0.15 mM La³⁺ (10.05 and 10.15 mg×g^-1 FW, respectively), but did not differ significantly between the treatments. However, the soluble protein was lower than the control when 0.05 mM Nd³⁺ or 0.1 mM Ce³⁺ was added to the medium. In hyperhydric shoots, the protein content at 0.05 mM Ce³⁺ (9.66 mg×g^-1 FW) and 0.1 mM Nd³⁺ (8.47 mg×g^-1 FW) was higher than that the control (4.5 mg×g^-1 FW), but similar protein contents to the control were observed when the REE concentrations were 0.15 mM.

Effect of REE on MDA concentrations

The MDA concentrations in the hyperhydric shoots were higher than the normal shoots (Fig. 3). The addition of different concentrations of La³⁺, Ce³⁺, and Nd³⁺ to the induction medium reduced the MDA concentrations in comparison to the control in both the non-hyperhydric and hyperhydric shoots. The MDA concentration was significantly decreased at a REE concentration of 0.05 mM in the hyperhydric plants. Decreased MDA content in the REE treatment signified oxidative stress reduction. Additionally, the O2 - staining levels of the hyperhydric leaves were higher than the non-hyperhydric leaves across the different treatments (Fig. 4).

Effect of REE on activities of antioxidative enzymes

The activities of antioxidative enzymes, such as SOD, CAT, GPX, and APX in the hyperhydric shoots were much higher than in the non-hyperhydric shoots (Fig. 5). SOD activity did not differ among the different treatments in the non-hyperhydric shoots, except in the 0.1 mM La³⁺ treatment. However, the SOD activity increased among the different hyperhydric plants treatments, particularly at 0.05 mM Nd³⁺, 0.1 mM Ce³⁺ and 0 .1 m M Nd³⁺, and 0.15 mM La³⁺, Ce³⁺, and Nd³⁺. CAT enzymes activity was decreased at 0 .05 mM Ce³⁺, 0.1 mM La³⁺ and 0 .1 mM Nd³⁺, and 0.15 mM La³⁺, Ce³⁺, and Nd³⁺ in the non-hyperhydric shoots. In the hyperhydric shoots, CAT activities decreased at 0.05 mM Ce³⁺, 0.1 mM REE, and 0.15 mM Ce³⁺ and 0 .15 mM Nd³⁺. The GPX enzymes activity decreased at 0.1 mM Ce³⁺ and Nd³⁺, and 0.15 mM Ce³⁺ in non-hyperhydric shoots. In hyperhydric shoots, the GPX activity decreased at 0.05 mM Ce³⁺, 0.1 mM Nd³⁺, and 0.15 mM Ce³⁺, and 0.15 mM Nd³⁺. APX activity significant decreased among the different treatments in the non-hyperhydric shoots, and significantly decreased at 0.05 mM Ce³⁺ and 0.05 mM Nd³⁺, 0.1 mM La³⁺ and 0.1 mM Nd³⁺ and 0.15 mM Nd³⁺ in hyperhydric plants.

Discussion

REE supplementation results in reduced hyperhydricity rate during the tissue culture

Tissue culture constitutes an indispensable technology for vegetative propagation and breeding in agricultural and horticultural industries. However, the tissue culture conditions are artificial and often extreme environments for plants, and may lead to physiological disorders, such as browning and hyperhydricity (Debergh et al. 1992). Hyperhydricity seriously impacts the quality of micropropagated plantlets and results in serious economic losses in horticultural industries (Soundararajan et al. 2017). In some cases, hyperhydricity has been reduced by changing water relations in vitro, increasing the agar concentration, incorporating additives to the culture medium (Debergh 1983; Sato et al. 1993; Hassannejad et al. 2012), or enhancing vessel ventilation (Thomas et al. 2000; Lai et al. 2005). Recently, Soundararajan et al. (2017) demonstrated that exogenous supplementation with silicon to the MS medium improved the recovery of vitrified carnation shoots. Gao et al. (2017) reported that supplementation with a definite concentration of silver nitrate (AgNO3) to the MS-based medium could reverse the hyperhydricity of pinks (Dianthus chinensis). In watermelon, Vinoth and Ravindhran (2015) showed that supplementation of the induction medium with AgNO3 reduced hyperhydricity and increased adventitious shoots. Hassannejad et al. (2012) found that the salicylic acid could reverse hyperhydricity in Thymus daenensis. In Cotoneaster, hyperhydricity was reduced by the addition of silicon to the culture medium (Sivanesan et al. 2011). Wang et al. (2007) showed that the addition of REEs to the induction media significantly reduced the hyperhydricity rate of Lepidium meyenii. In this study, the addition of La³⁺, Ce³⁺, and Nd³⁺ to the carnation shoot induction media reduced the hyperhydricity rate of the adventitious shoots. When the REE concentrations were 0 .05 mM and 0 .1 m M , the hyperhydricity r ate was significantly reduced. However, the addition of 0.1 mM and 0.15 mM Nd³⁺ resulted in a hyperhydricity rate that was only slightly lower than the control, and was not statistically significant. REEs have a number of biological activities in plants, although the mechanisms by which REEs influence plant is not clear (Hu and Ye 1996). As heavy metals, low doses of REEs could have positive effects, such as increasing the uptake of mineral elements and promoting photosynthesis in plants, while high doses of REEs could be harmful (Wang et al. 2005; Peralta-Videa et al. 2014; Xu et al. 2016). These results suggested that REE on carnation shoot induction and hyperhydricity rate in the media at around 0.05 mM.

Regulation of antioxidant enzymes by REEs in response to the oxidative stress

Hyperhydricity seriously impedes regeneration in plants. Some stress conditions in vitro resulted in hyperhydricity, such as high humidity, high levels of plant hormones, and gas accumulation in the atmosphere of the culture vessels (Saher et al. 2004). These different stress may result in a continued burst of reactive oxygen species (ROS), which can induce lipid peroxidation, cause DNA damage, and plant cell death (Cassells and Curry 2001; Ochatt et al. 2002). Lipid peroxidation is used as an indicator of stress-induced oxidative damage (Cassells and Curry 2001). MDA content has been used as a measurement for assessing lipid peroxidation and oxidative damage (Zhou and Zhao 2004; Sivanesan et al. 2011), MDA content has been observed to increase in carnation shoots exhibiting hyperhydricity (Piqueras et al. 2002; Saher et al. 2004). We also detected that there were higher concentrations of MDA in the hyperhydric plants than the non-hyperhydric plantlets (Fig. 3). Additionally, the antioxidative enzymes activity was higher in hyperhydric shoots in comparison to non-hyperhydric shoots (Fig. 4, 5). Much research has shown that the accumulation of ROS results in hyperhydricity in various plant species (Saher et al. 2004; Dewir et al. 2006; van den Dries et al. 2013; Tian et al. 2015). Dewir et al. (2006) found that SOD, POD, and CAT activity in the hyperhydric tissue of regenerated Euphorbia remillii plantlets were significantly increased compared to non-hyperhydric plantlets. These results suggest that antioxidative enzymes play a crucial role in preventing hyperhydricity in regenerated plantlets.

In this study, the addition of 0.05 mM La³⁺, Ce³⁺, and Nd³⁺ to the shoots induction media resulted in a significant decline in MDA contents in hyperhydric plants compared to the control, although at 0.05 mM La³⁺, and 0.05 mM Ce³⁺, and 0.1 mM La³⁺ and 0.1 mM Ce³⁺ the M DA c ontent o f non-hyperhydric plantlets was still lower than the control. The results of this study suggested that supplementation of the culture medium with REEs reduce oxidative stress. These results are similar to that of research on L. meyenii that the oxidant stress was alleviated partly by the antioxidative properties of La³⁺, Ce³⁺, and Nd³⁺ (Wang et al. 2007). Meanwhile, the activity of antioxidative enzymes such as CAT, GPX, and APX were most decreased in the non-hyperhydric and hyperhydic plantlets at different concentration of REEs, while SOD activity was only significantly reduced with the addition of 0.1 mM La³⁺ to the media. These results suggest that REEs may have the ability to remove redundant ROS when they are added to the media and absorbed by the plantlets. Wang et al. (1998) reported that Ce³⁺ and Ce4+ have the ability to remove ROS, as they have a similar function to SOD. Liu et al. (2008) reported that La³⁺ induced plant resistance in unfavorable environments. In general, an appropriate amount of REEs improves plant growth and plant resistance against stress (Peng et al. 2013). In the present study, the soluble protein of the plantlets was increased and oxidative stress was slightly alleviated when suitable concentrations of REEs were added to the carnation induction media. It is suggested that REEs were absorbed by the shoots, resulting in the promotion of plant growth and reduction in the hyperhydricity rate in the micropropagation of carnation.