In vitro response of vanilla (Vanilla planifolia Jacks. ex Andrews) to PEG-induced osmotic stress

Drought-induced water stress affects the productivity of the Vanilla planifolia Jacks. ex Andrews crop. In vitro culture technique is an effective tool for the study of water stress tolerance mechanisms. This study aimed to evaluate the morphological, physiological and biochemical response of V. planifolia under in vitro water stress conditions induced with polyethylene glycol (PEG). In vitro regenerated shoots of 2 cm in length were subjected to different concentrations of PEG 6000 (0, 1, 2 and 3% w/v) using Murashige and Skoog semi-solid culture medium. At 60 days of culture, different growth variables, dry matter (DM) content, chlorophyll (Chl), soluble proteins (SP), proline (Pro), glycine betaine (GB), stomatal index (SI) and open stomata (%) were evaluated. Results showed a reduction in growth, Chl content, SP, SI and open stomata (%) with increasing PEG concentration, whereas DM, Pro and GB contents rose with increasing PEG concentration. In conclusion, PEG-induced osmotic stress allowed describing physiological and biochemical mechanisms of response to water stress. Furthermore, the determination of compatible Pro and GB osmolytes can be used as biochemical markers in future breeding programs for the early selection of water stress tolerant genotypes.

Vanilla (Vanilla planifolia Jacks. ex Andrews) is an orchid species native to Mexico used for the extraction of vanillin, one of the most economically important aromatic compounds worldwide 1 . This species, which has a hemiepiphytic habit, requires water and abundant moisture for proper development 2 . It is estimated that one of the main factors affecting vanilla production is its high susceptibility to pests and diseases and its low tolerance to water stress conditions, due to its low genetic variability 3 . In addition, various climate change scenarios forecast prolonged droughts that will cause water stress in plants 4 . Vanilla is also categorized as endangered B2ab(iii, v) version 3.1 in the International Union for Conservation of Nature and Natural Resources (IUCN) red list of threatened plants (http:// www. iucnr edlist. org).
In plants, the main water stress tolerance mechanisms are the accumulation of reactive oxygen species, synthesis of antioxidant enzymes, synthesis of specific proteins (hydrophilins, dehydrins, chaperonins) and osmotic adjustment mediated by compatible osmolytes [5][6][7][8] . Water stress reduces the availability of water in the cytoplasm of cells, affecting their homeostasis. Lack of water in cells causes osmotic and oxidative stress, damaging physiological, biochemical and molecular processes 5,9 . In vitro osmotic stress simulation using osmotic agents is an efficient technique that provides a controlled and uniform environment for the study of different water stress response mechanisms 10 . High molecular weight polyethylene glycol (PEG) is highly soluble in water and non-penetrable in cells, produces a negative osmotic potential in the culture medium without generating toxicity and is one of the most widely used osmotic agents to study the effects of in vitro water stress in plants 11 . The use of PEG as an in vitro stress agent has been evaluated in crops such as sugar beet (Beta vulgaris L.) 12 , nopal (Opuntia ficus-indica (L.) Mill) 13 , wheat (Triticum durum Desf.) 14 , stevia (Stevia rebaudiana Bert.) 15,16 , and banana (Musa acuminata Colla) 17,18 among others. In the family Orchidaceae, it has been used in Acianthera teres Lindl., Octomeria crassifolia Lindl., O. gracilis Barb. Rodr. 19 (Table 1). Overall, in vitro growth of V. planifolia decreased with increasing PEG concentrations. The longest shoots (67.6 mm) were obtained at 0% of PEG, whereas the shortest shoots (26.6 mm) were obtained with 3% of PEG. For the variable number of leaves per shoot, the highest number of leaves (5.7) was observed in 0% of PEG, while the lowest number of leaves (2.0) was observed in 3% of PEG. For the variables number and length of roots, the shoots with the highest number (5.8) and length of roots (31.0 mm) were observed in 0% of PEG, while the smallest number (2.6) and size of roots (11.1 mm) was found in the treatment with 3% of PEG ( Fig. 1). For fresh weight and dry weight, the only significant decrease (0.44 mg) and (0.05 mg) in comparison with control was recorded in 3% of PEG, respectively. Regarding dry matter content, the highest percentages (10.01, 10.79 and 11.37%) were observed in the PEG treatments, while the lowest percentage (7.97%) was observed in the control treatment.
Chlorophyll (Chl) content. Significant differences were observed in Chl contents in leaves of V. planifolia seedlings exposed to different concentrations of PEG 6000. The highest total Chl (0.45 mg g −1 FW) and Chl a (0.32 mg g −1 FW) contents were observed in the 0% of PEG treatment, followed by the 1, 2 and 3% of PEG treatments, respectively. The highest Chl b contents (0.14 and 0.12 mg g −1 FW) were observed in the 0 and 1% PEG treatments, while lower Chl b contents (0.10 and 0.08 mg g −1 FW) were observed in the 2 and 3% PEG treatments ( Figure 2a).
Soluble protein (SP) content. Differences in SP content were observed at all PEG concentrations. The highest SP content (23 mg g −1 FW) was observed in the 1% of PEG treatment, while lower SP contents were observed (17, 14 and 3 mg g −1 FW) in the 2, 3 and 0% of PEG treatments, respectively (Fig. 2b).

Proline (Pro) content.
A gradual increase in Pro content was observed with increasing PEG concentration.
Glycine betaine (GB) content. A gradual increase in GB content was observed with increasing PEG concentration. The highest GB content (28 µmol g −1 DW) was observed at 3% of PEG, while the lowest contents (15, 12 and 4 µmol g −1 DW) were observed in the 2, 1 and 0% of PEG treatments, respectively (Fig. 2d).

Stomatal index (SI) and percentage of open stomata.
Significant differences were observed in the SI and percentage of open stomata of V. planifolia exposed to different PEG concentrations. The highest SI (4.8) was observed in the control treatment, followed by the 1, 2 and 3% of PEG treatments, respectively (Fig. 3). The highest percentage of open stomata (92%) was observed in the control treatment, followed by the PEG treatments (Fig. 4).

Discussion
Results obtained in this study demonstrate that in vitro growth of V. planifolia is affected under osmotic stress conditions induced by different PEG concentrations. The different PEG concentrations had an effect on the osmotic potential of the culture medium (Ψs = − 0.14, − 0.23, − 0.32 and − 0.43 MPa for 0, 1, 2 and 3% of PEG). The PEG 6000-induced osmotic stress affects the in vitro growth of V. planifolia. One of the most important strategies that plants have adopted against osmotic stress is to stop or slow their growth 22,23 . In fact, slow growth is an adaptive response for plant survival under stress conditions 24 .
Bello-Bello et al. 25 reported that, during in vitro preservation of V. planifolia using 3% PEG 6000, this osmotic agent acts as a growth inhibitor. Similar results were described by Gao et al. 21 who found significant differences in the growth of Dendrobium officinale shoots with 1, 3 and 5% of PEG 6000. Jácome-Blásquez et al. 20 reported a reduction in germination of Prosthechea vitellina (Lindl.) W. E. Higgins with an increasing concentration of  27 reported an increased dry matter content in shoots of eight genotypes of grass pea (Lathyrus sativus) with 5, 10 and 15% of PEG 6000. According to Rai et al. 28 , growth inhibition under water stress is associated with a reduction in the ability of plants to absorb water and nutrients caused by the decreased water potential of the culture medium. In this study, the inhibition of in vitro V. planifolia shoot growth in the presence of PEG could be explained by the synthesis of phytohormones responsible of stomatal closure, causing a reduction in transpiration and low carbon assimilation capacity. According to the authors of Ref. 29 , the stomatal closure under drought condition is due to the accumulation of compounds such as abscisic acid, methyl jasmonate, ethylene, and brassinosteroids. On the other hand, the increase in dry matter content in the presence of PEG could be explained by synthesis of specific proteins, compatible osmolytes and/or metabolites in response to water stress and probably due to accumulation of inorganic solutes in relation to osmotic adjustment.
In this study, a decrease in total Chl content was observed in V. planifolia shoots in relation to the increase in PEG concentration. In Orchidaceae in vitro, Gao et al. 21 reported a decrease in Chl content in D. officinale shoots exposed to 0, 1, 3 and 5% of PEG 6000. Suis et al. 30 found a reduction in Chl content in protocorm-like bodies of Aranda Broga Blue Bell with 15% of PEG 400. A reduction in Chl content in V. planifolia under water stress (− 0.9 MPa) has been reported under ex vitro conditions 31 . For chlorophyll content, according to the authors of Ref. 32 , a reduced chlorophyll content under water stress is caused by a decrease in photosynthetic activity due to stomatal closure. Also, the low photosynthetic activity is related to dry matter accumulation as a result of osmotic stress affecting carbohydrate transport 33,34 . In addition, with low photosynthetic activity, Chl degradation is a response mechanism to avoid energy wasting [35][36][37] . www.nature.com/scientificreports/ Results of biochemical analysis showed significant differences in SP, Pro and GB contents (Fig. 2). The SP content decreased with increasing PEG concentration in V. planifolia shoots; however, SP content was significantly higher in all PEG treatments with respect to the control treatment. Variation in SP content is a water stress tolerance response 37 related to synthesis, accumulation or degradation. The SP accumulation at 1% of PEG compared to control (without PEG stress) is probably related to the synthesis of hydrophilins 38 , aquaporins 39 , dehydrins 40 , antioxidant enzymes 41 , and chaperonins 42 , among others. Plants can synthesize this type of protein to neutralize oxidative stress and osmotic stress 41,43 . The SP reduction in 2 and 3% of PEG could be explained due to a degradation and/or synthesis reduction caused by a physiological wilting point in plants during in vitro osmotic stress.
Qayyum et al. 44 reported an increase in protein content in seedlings of five wheat (Tritricum aestivum L.) cultivars exposed to different osmotic potentials (− 0.2, − 0.4 and − 0.6 MPa) generated with PEG 6000. In contrast, Razavizadeh et al. 35 observed a decrease in protein content in thyme (Thymus vulgaris) seedlings grown under 2, 4, 6 and 8% of PEG 6000. Similarly, Gao et al. 21 observed a decrease in protein content in D. officinale shoots grown under water stress induced with 1, 3 and 5% of PEG 6000. Suis et al. 30 observed a reduction in SP using PEG 400 (0, 5, 10 and 15%) in protocorm-like bodies of the Aranda Broga Blue Bell orchid. Chandran and Puthur 31 note that the decrease in SP content in the V4 somaclone of V. planifolia under water stress (− 0.9 MPa) may be due to its degradation.
The results of this study suggest that under the in vitro conditions evaluated, protein accumulation in 1% of PEG treatments could be explained by the biosynthesis of specific proteins as a mechanism of tolerance to osmotic stress; whereas, the decrease in SP content at doses of 2 and 3% of PEG could be due to a reduction in biosynthesis and/or an increase in their degradation. Razavizadeh et al. 35 note that protein degradation under osmotic stress provides sources of nitrogen used in different metabolic processes. www.nature.com/scientificreports/ The Pro accumulation in plants is a physiological response to salt and water stress by reducing plant growth 45 . Under water stress conditions, Pro acts as a stabilizer of structural proteins, a compatible osmolyte and a catalyst of antioxidant enzyme activity 46 . In our study, a significant increase in Pro content was observed in shoots exposed to PEG; however, the 3% PEG treatment showed a twofold increase in Pro content compared to the control. This result suggest that Pro has a role in water stress tolerance mechanisms in V. planifolia.
Other studies show that Pro is a biochemical indicator that contributes to water stress tolerance mechanisms. Piwowarczyk et al. 27 reported an increase in Pro content in grass pea (Lathyrus sativus) shoots at 5, 10 and 15% of PEG 6000. Razavizadeh et al. 35 observed an increase in Pro accumulation in thyme (Thymus vulgaris) shoots at 2, 4, 6 and 8% of PEG. In Orchidaceae, Jácome-Blásquez et al. 20 found that the Pro content in Prosthechea vitellina (Lindl.) W. E. Higgins increased by twice as much relative to the control treatment by increasing the PEG 8000 concentration. In contrast, a previous study reported 21 that the Pro content in Dendrobium officinale shoots decreased in the treatments with 1, 3 and 5% of PEG 6000. In addition to Pro, there are other osmolytes such as sugars and GB that accumulate in response to water stress 47 .
Glycine betaine content showed a significant increase when V. planifolia shoots were exposed to PEG. To date, GB accumulation in V. planifolia under in vitro water stress has not been reported. However, GB accumulation has been reported in other species. Hajihashemi and Ehsanpour 48 reported GB accumulation in S. rebaudiana shoots exposed to 2, 4 and 6% of PEG 6000. Datir and Inamdar 49 observed that GB accumulation increased in all wheat (T. aestivum L.) cultivars in response to water stress induced by 15% of PEG 6000. Jácome-Blásquez et al. 20 found increased GB content in P. vitellina (Lindl.) W. E. Higgins at 5-15% of PEG 8000. In addition, GB is a compatible osmolyte that acts as an osmoprotectant, maintains photosynthetic activity by protecting the thylakoid membrane, and neutralizes reactive oxygen species by regulating the expression of oxidative stress response genes during osmotic stress 37,[50][51][52] . The absence of GB in cells could result in cell membrane damage, protein degradation, DNA damage and loss of photosynthetic capacity 37,53,54 .
PEG-induced osmotic stress reduced the SI and percentage of open stomata in V. planifolia. Paletri et al. 55 reported a decrease in SI in Cattleya sp. Lindl. in vitro with 25% of PEG 6000. Chandran and Puthur 31 reported a reduced SI (1.65) in ex vitro somaclones of V. planifolia under water stress generated by water deficit (− 0.9 MPa). According to the authors of Ref. 56 , variation in the SI under water stress is different for each species. V. planifolia under normal conditions has a SI of 2.81 ± 0.27 31 . On the other hand, stomatal closure as a result of using PEG for the evaluation of osmotic stress in vitro has been reported. Another study 57 found a reduction in stomatal opening in Persian walnut (Juglans regia L.) of 43 and 58% at 3 and 5% of PEG 6000 relative to the control.
Plant stomata are normally open during in vitro conditions due to excess water in culture medium and high humidity within the culture vessels. However, the stomatal functionality is stimulated during in vitro PEGinduced osmotic stress. Stomatal functionality, through the opening and closing of stomata, is a mechanism that regulates transpiration and maintains water balance during hydric stress 58 . In this regard, it has been reported 59 that adequate stomatal functionality regulates the water balance in plants. Stomatal closure and a low SI generate a low transpiration rate to avoid dehydration due to water loss.  Osmotic stress induction with PEG. For in vitro selection pressure with PEG, 2 cm long shoots (without roots) obtained after four subcultures of 45 days each in multiplication medium were used. The shoots were grown in 22 × 150 mm test tubes containing 20 mL of MS semi-solid medium without growth regulators and different concentrations of PEG 6000 MW (Sigma-Aldrich) (0, 1, 2 and 3% w/v). The PEG was added directly into the culture medium. The culture medium pH and the sterilization and incubation conditions were the same as described above. Each treatment consisted of 10 explants, with one shoot per test tube. At 60 days of culture (without explant subcultures), the different morphological variables and percentage of dry matter (DM) were evaluated. The DM content was calculated using the formula: dry weight/fresh weight × 100. Chl, SP, Pro and GB contents were determined. Additionally, SI and percentage of open stomata were evaluated.
Osmotic potential measurements of the culture medium. The osmotic potential (Ψs) of the semisolid culture media with different PEG concentrations was determined using a vapor pressure osmometer (5520 Vapro, Wescor Inc., Logan, UT, USA) at the beginning of osmotic stress induction.
Determination of chlorophyll (Chl) content. The Chl content was determined according to the methodology proposed by Ref. 61 and with slight modifications by Ref. 62 . The absorbance readings were measured in a spectrophotometer (Genesys 10S, Thermo Scientific, Waltham, MA, USA) at 663 nm and 645 for Chl a and b, respectively. The Chl content was calculated using the following formulas: where A 663 and A 645 : Absorbance; C: Concentration (mg g −1 FW); V: volume graduation in mL −1 ; W: sample weight in g; 1000: conversion factor.
Determination of soluble protein (SP) content. Soluble protein estimation was carried out with the colorimetric method proposed by Bradford 63 . First, 100 mg of fresh leaf tissue were weighed and macerated in a mortar in 25 mL of cold acetone for 5 s. The macerated tissue was vacuum filtered and 12.5 mL of acetone was added. Then 5 mL of 0.1M TRIS-HCl Buffer (pH 7.1) was added to the extract obtained and the samples were stored on ice bath. Subsequently, they were centrifuged at 3100×g for 20 min at 4 °C. A 100 µL aliquot of the supernatant was acquired and added to 4.9 mL of Bradford reagent. After 12 min of reaction, the chromophore was read at an absorbance of 595 nm in a spectrophotometer (Genesys 10S). Quantification was done using a calibration curve with bovine albumin standard (Hoffmann-La Roche Ltd., Grenzacherstrasse, BS, SWI).

Determination of proline (Pro) content.
Proline determination was carried out by the method described by Ref. 64 . First, 125 mg of fresh leaf tissue were macerated in a mortar and homogenized with 5 mL of 3% (w/v) sulfosalicylic acid. The resulting solution was filtered through Whatman #2 filter paper. Then 1 mL aliquot was acquired, 1 mL of glacial acetic acid and 1 mL of ninhydrin were added. Incubation of 1 mL of the extract was carried out in a thermoregulated bath for one hour at 100 °C. The tubes were removed and the reaction terminated in an ice bath. Then 2 mL of toluene was added and mixed for 30 s for phase separation. The resulting chromophore was read at an absorbance of 520 nm in the spectrophotometer (Genesys 10S) using toluene as a blank. Quantification was done using a calibration curve with l-proline standard (Sigma-Aldrich).
Determination of glycine betaine (GB) content. Glycine betaine determination was carried out by the method proposed by Ref. 65 . For this, 100 mg of dry leaf tissue were weighed and reacted with water and 2N H 2 SO 4 after the addition of 0.2 mL of periodide complex (KI-I 2 ). Extraction was performed by centrifugation at 3100×g for 20 min at 4 °C and subsequent addition of 9 mL of 1,2-Dichloroethane. Absorbance was measured at 365 nm in a spectrophotometer (Genesys 10S) using 1,2-Dichloroethane as a blank. Quantification was done using a calibration curve with betaine standard (Sigma-Aldrich). www.nature.com/scientificreports/ obtained were tested with an analysis of variance (ANOVA) followed by Tukey's test (p ≤ 0.05), performed using IBM SPSS statistical software version 22 (https:// www. ibm. com/ suppo rt/ pages/ node/ 313621).

Conclusions
Polyethylene glycol induced osmotic stress under in vitro conditions had effects on growth, physiology and biochemical determinations evaluated in V. planifolia. In this study, a reduction in growth, chlorophyll content, soluble proteins, stomatal index and open stomata (%) was observed with increasing PEG concentration, whereas dry matter, proline and glycine betaine contents rose with increasing PEG concentration. The compatible proline and glycine betaine osmolytes will be helpful mechanism of water stress tolerance in V. planifolia genotypes. In addition, ROS production and antioxidant enzyme activities could be evaluated in a future study for a better understanding of water stress tolerance. Further research is recommended to study others species in Vanilla genus in these in vitro conditions.