Phytohormone profiles in non-transformed and AtCKX transgenic centaury (Centaurium erythraea Rafn) shoots and roots in response to salinity stress in vitro

Plant hormones regulate numerous developmental and physiological processes. Abiotic stresses considerably affect production and distribution of phytohormones as the stress signal triggers. The homeostasis of plant hormones is controlled by their de novo synthesis and catabolism. The aim of this work was to analyse the contents of total and individual groups of endogenous cytokinins (CKs) as well as indole-3-acetic acid (IAA) in AtCKX overexpressing centaury plants grown in vitro on graded NaCl concentrations (0, 50, 100, 150, 200 mM). The levels of endogenous stress hormones including abscisic acid (ABA), salicylic acid (SA) and jasmonic acid (JA) were also detected. The elevated contents of total CKs were found in all analysed centaury shoots. Furthermore, increased amounts of all five CK groups, as well as enhanced total CKs were revealed on graded NaCl concentrations in non-transformed and AtCKX roots. All analysed AtCKX centaury lines exhibited decreased amounts of endogenous IAA in shoots and roots. Consequently, the IAA/bioactive CK forms ratios showed a significant variation in the shoots and roots of all AtCKX lines. In shoots and roots of both non-transformed and AtCKX transgenic centaury plants, salinity was associated with an increase of ABA and JA and a decrease of SA content.

www.nature.com/scientificreports/ SA, endogenous plant growth regulator with phenolic nature, is involved in the regulation of physiological processes such as growth, photosynthesis, and nitrogen metabolism 31 . It also plays an important role in protection against biotic and abiotic stresses including salinity 32 . The role of SA in plant defense from salt stress conditions is species-specific and has been well documented [33][34][35] . For example, the endogenous SA level was decreased under salt stress in rice seedlings 36 and salt-stressed Iris hexagona plants 37 . Salinity also strongly decreased SA concentration in glycophyte tomato S. lycopersicum but on the other hand, SA content increased in response to salt in the wild halophyte S. chilense 20 .
Jasmonic acid (JA) is an important cellular regulator involved in diverse developmental processes such as seed germination, root growth, fertility, fruit ripening, and senescence 37 . It is known that JA and its derivatives also strongly respond to salinity stress conditions 38 . It has been reported that JA levels in various tomato cultivars changed in response to the salt stress 39 . When exposed to salinisation, the salt-tolerant tomato cultivar considerably increased its JA level whereas the salt-sensitive cultivar decreased it. Concentration of jasmonates was more than two times higher in the halophyte S. chilense than in the glycophyte S. lycopersicum, and salinity increased its concentration by more than 100% in both species compared to control plants 20 .
Previous work described salinity stress response of non-transformed and AtCKX transgenic centaury shoots and roots grown in vitro 15 . It was shown that centaury roots showed higher salinity tolerance compared to shoots. Morphogenic potential of AtCKX1 transgenic line was similar to non-transformed line under increased NaCl concentrations in vitro. On the other hand, there were significant difference in morphogenic response of two investigated AtCKX2 transgenic lines, specially in root culture. Accordingly, in this followed work, we compared the profiles of endogenous of phytohormones in non-transformed and transgenic centaury AtCKX lines grown in vitro under salinity stress conditions. The contents of CKs, auxins, ABA, JA and SA were determined in shoots and roots cultured in vitro for eight weeks on increased NaCl concentrations. This work represents the first report on altered phytohormone levels in centaury, the significant medicinal plant species, during salt stress. The obtained results might be a useful tool for better understanding of salinity tolerance of non-transformed and CK-deficient transgenic centaury plants grown in vitro.

Results
Endogenous cytokinins. Endogenous CKs were detected and quantified in shoots and roots of nontransformed ( Fig. 1), one transgenic AtCKX1 (line 29, Fig. 2) and two AtCKX2 (lines 17 and 26) centaury lines (Figs. 3, 4) grown on media with graded NaCl concentrations. According to their physiological function and conjugation status, endogenous CKs were divided into five groups (as proposed by 40 ) including bioactive forms (CK nucleobases), transport forms (CK ribosides), storage forms (CK O-glucosides), deactivation forms (CK N-glucosides) and immediate biosynthetic precursors (CK phosphates). The abbreviations for CKs presented in this work were adopted and modified according to 41 . The levels of individual endogenous CK groups were determined and are shown on Figs. 1, 2, 3 and 4.
Analysis of endogenous CKs in non-transformed shoots showed that the levels of bioactive CK forms were significantly increased only on one NaCl concentration (150 mM) while decreased on two NaCl doses (50 and 200 mM). In non-transformed roots, the contents of bioactive CK forms were significantly enhanced on all NaCl concentrations (Fig. 1). Transport CK forms were reduced in non-transformed shoots on media supplemented with NaCl (except for 150 mM) but elevated on all media in non-transformed roots. Concentrations of storage CK forms were decreased in non-transformed shoots on all salt-treated media except for 100 mM NaCl while in roots they were considerably enhanced. Compared to control NaCl-free medium, the deactivation CK forms and CK phosphates (immediate biosynthetic CK precursors) were increased in non-transformed shoots grown on 50, 100 and 150 mM NaCl whereas on media supplemented with 200 mM NaCl their levels were decreased. In roots, deactivation CK forms and immediate biosynthetic CK precursors were increased on all applied NaCl concentrations. Increase in the total CK contents was recorded in non-transformed shoots on 50, 100 and 150 mM NaCl. On the other hand, elevated total CK levels were detected on all NaCl concentrations.
Transgenic AtCKX1-29 shoots showed increased levels of bioactive CKs on media containing 100 and 200 mM NaCl while their enhanced contents were found on all NaCl concentrations in roots (Fig. 2). Transport and deactivation CK forms as well as CK phosphates were increased in both AtCKX1-29 shoots and roots on all NaCl concentrations. The levels of storage CKs were elevated in AtCKX1-29 shoots only on lower, i.e. 50 and 100 mM NaCl, concentrations whereas increased storage CKs contents were recorded for all NaCl-treated variants in roots. On all NaCl concentrations, the AtCKX1-29 shoots and roots showed increased levels of total CKs in comparison to the control NaCl-free medium.
In shoots of transgenic AtCKX2-17 line, the levels of bioactive CKs were not much affected by NaCl treatments except for the medium with 200 mM NaCl where their significant decrease was recorded (Fig. 3). On the other hand, increased levels of bioactive CK forms were found in roots grown on 50, 100 and 150 mM NaCl. Transport, storage and deactivation CKs represent the three elevated groups of CKs in AtCKX2-17 shoots grown on all NaCl concentrations. Unlike shoots, the transport, storage and deactivation CK forms exhibited varying dependencies on applied NaCl concentrations in AtCKX2-17 roots. In AtCKX2-17 shoots, the immediate biosynthetic CK precursors decreased or remained unaffected on NaCl-treated media (except for their increase on 150 mM). On the other hand, the contents of CK phosphates were elevated in roots. In analogy to AtCKX1-29 transgenic line, AtCKX2-17 shoots and roots showed increased total CKs on all applied NaCl concentrations (with the exception of 200 mM NaCl for roots) in comparison to the control plants grown on NaCl-free medium.
Interestingly, another analysed AtCKX2 line, AtCKX2-26, showed different pattern compared with AtCKX2-17 (Fig. 4). Whereas the levels of bioactive CKs were increased on 50  In AtCKX2-26 roots, the contents of all particular CK forms as well as total CKs were increased on media containing more than 50 mM NaCl.
Quantification of endogenous IAA levels and determination of IAA/bioactive cytokinins ratio. Alterations in endogenous IAA contents were detected in non-transformed and AtCKX-overexpressing centaury plants grown in vitro on media with graded NaCl doses (Fig. 5). In non-transformed shoots, the IAA levels were decreased on all applied NaCl concentrations (Fig. 5a). Simultaneously, the contents of bioactive CKs were reduced on 50 and 200 mM NaCl and increased on 150 mM NaCl (Fig. 1). On the other hand, in non-transformed roots a decrease of IAA content was detected only on 100 mM NaCl while the amounts of  www.nature.com/scientificreports/ bioactive CK contents were enhanced on all NaCl concentrations. Regardless these differences, the IAA/bioactive CKs ratios were decreased in all non-transformed shoots as well as in roots on all applied NaCl concentrations (Fig. 5a). A similar pattern of the IAA contents was ascertained in AtCKX1-29 plants (Fig. 5b). In comparison to the NaCl-free medium, AtCKX1-29 shoots showed decrease in the IAA contents on 50, 100 and 150 mM NaCl. The levels of bioactive CK forms were elevated on two NaCl concentrations, 100 and 200 mM NaCl (Fig. 2). Unlike shoots, in the AtCKX1-29 roots increased IAA contents were detected only on 50 mM NaCl whereas elevation of bioactive CK amounts were recorded on all NaCl concentrations. In analogy to the non-transformed plants, the AtCKX1-29 shoots and roots showed decreased IAA/bioactive CKs ratios (Fig. 5b).
Endogenous IAA and bioactive CKs contents, as well as IAA/bioactive CKs ratios were considerably altered in transgenic AtCKX2-26 plants grown on graded NaCl concentrations (Fig. 5d). In comparison to NaCl-free medium, the AtCKX2-26 shoots were characterized by increased IAA content on 100 and 150 mM NaCl. Bioactive CK levels were increased on 50 and 100 mM NaCl and decreased on 150 and 200 mM NaCl concentrations According to physiological function and conjugation status cytokinins were divided into five groups including bioactive forms (DHZ, iP, tZ, cZ), transport forms (DHZ9R, iP9R, tZ9R, cZ9R), storage forms (DHZ9ROG, DHZOG, tZ9ROG, tZOG, cZ9ROG, cZOG), deactivation forms (DHZ7G, DHZ9G, iP7G, iP9G, tZ7G, tZ9G, cZ7G, cZ9G), immediate biosynthetic precursors (tZRMP, DHZRMP, cZRMP, iPRMP) and total CKs content (boja kolone). Data represent mean ± standard error. Means marked with asterisks are significantly different from corresponding control values (LSD test, p ≤ 0.05). The bigger asterisks represent the values significantly higher than control ones while smaller asterisks represent values significantly lower than control ones. www.nature.com/scientificreports/ Analysis of endogenous stress hormones. The levels of endogenous stress hormones, ABA, JA and SA, in non-transformed and AtCKX transgenic centaury shoots and roots cultured on media with graded NaCl concentrations were also determined (Figs. 6, 7, 8, 9). In all analysed shoot and root samples, the same increment/decrement tendency for particular stress hormone was noticed. The ABA and JA contents were elevated on  It was intersting to note that in non-transformed shoots JA and SA contents were higher than in roots on all applied NaCl concentrations while ABA content was higher in non-transformed roots than in shoots only on 200 mM NaCl (Fig. 6). On the other hand, transgenic AtCKX1-29, AtCKX2-17 and AtCKX2-26 shoots showed higher ABA and JA contents than roots (Figs. 7, 8, 9). Also, higher SA content were determined in AtCKX1-29 shoots than in roots whereas in both AtCKX2 lines the similar amounts of SA were detected in both organs.

Discussion
Phytohormones are produced in very low concentrations in plants where they act as chemical messengers in signalling mechanisms under stress conditions. In this study, non-transformed and AtCKX transgenic centaury plants were grown on graded NaCl concentrations in vitro with the aim to elucidate how salinity stress affects phytohormone profiles of this valuable medicinal plant species. Taking advantage of advanced high-performance liquid chromatography tandem mass spectrometry methodology, the patterns and levels of endogenous CKs as well as the amounts of auxin IAA, ABA, JA and SA in the shoots and roots of AtCKX1 and AtCKX2 centaury lines under salt stress were analyzed and are shown in the present work.
Cytokinins are present in all plant tissues regulating numerous developmental processes such as cell division, apical dominance, leaf senescence, vascular and shoot differentiation and anthocyanin production 3 . It is known that CKs play both positive and negative roles in the stress tolerance. According to various reports, endogenous CK levels were decreased in the course of enhanced stressed conditions 13,42 while some observations indicated increased CKs content, whether it was short-or long-term exposure to stress 43,44 . The observed changes in the endogenous CK contents clearly indicated an involvement of these hormones in plant stress responses. Plant resistance to stress conditions is a complex process and the role of phytohormones can be investigated e.g. on plants with altered endogenous CK levels. Considering that CKs have significant roles in plant stress responses, www.nature.com/scientificreports/ genes encoding two key enzymes in CK metabolism, ipt (responsible for CK biosynthesis) and ckx (responsible for CK degradation), were proposed as possible target genes involved in stress resistance 13,45 . Altered endogenous CK levels in Arabidopsis, Nicotiana tabacum and Hordeum vulgare plants, modified either by loss of the ipt genes expression or by overexpression of the ckx encoding genes, showed that CKs played a negative role in response to heat and/or drought stress 10,13,[45][46][47][48][49] . Also, there are literature data describing that reduced CK contents improved drought and salt stress tolerance in Arabidopsis ckx overexpressing lines and ipt1, ipt3, ipt5 and ipt7 mutants 13,45,50 . Previous investigation showed that overexpression of the AtCKX1 and AtCKX2 genes in transgenic centaury plants resulted in altered CKs profile leading to a decline of bioactive CK levels and, at the same time, increased contents of storage CK forms, inactive CK forms and/or CK nucleotides 51 . Previous investigation showed that overexpression of CKX genes caused hyposensitivity to salt stress and drought stress in arabidopsis 13 and Physcomitrella patens 52 . Our current results generally indicated that all groups of endogenous CKs were considerably altered in non-transformed and AtCKX centaury shoots on all applied NaCl concentrations and that the total CK contents were elevated comparing to the corresponding control. Furthermore, increased levels of all five CK groups as well as of total CKs were recorded on used NaCl concentrations in non-transformed and AtCKX centaury roots. Furthermore, analyses of endogenous CKs showed that the levels of bioactive CK forms varied considerably among non-transformed and all analysed AtCKX shoots depending on NaCl concentration. On the other hand, increased levels of bioactive CK forms were found in roots of all analysed centaury lines, on almost all applied NaCl concentration. These observations indicated that salinity triggered different changes in the CK contents in centaury shoots and roots. Considering that CKs are produced mainly in roots which are directly involved in responses to nutrients it can be concluded that CKs represent an important signal traveling from roots to the shoots. The transport of CK from roots to shoots is considered as the key factor in response to salt stress 53 . Therefore, the salt-induced changes in CK profiles may disturb homeostasis on the subcellular level www.nature.com/scientificreports/ and also further alter the CK content in other compartments. Furthermore, it is important to note that altered CK levels in centaury tissues may also be the result of their prolonged exposure to salinity during eight weeks. Any strong conclusions about CK content is very difficult because the complexity of the cytokinin profiles has not been fully explored 53 . It is known that CKs affect antioxidant systems and related processes under drought and salt stress. Recent biochemical analyses showed lower free proline, MDA content and H 2 O 2 production as well as reduced SOD and POX activities in transgenic AtCKX centaury plants under prolonged salinity conditions 15 . Similarly, the activity of SOD was considerably decreased in leaves and roots of transgenic AtCKX2 tobacco plants during the salt stress 24 . Previous investigations showed that ectopic expression of ckx genes in barley activated genes involved in biosynthesis of flavonoids 49 and also in drought tolerance 54 . On the other hand, increased level of antioxidants during the stress could protect plant cells from reactive oxygen species (ROS) accumulation and reduce electrolyte leakage and/or rises in malondialdehyde levels 55,56 . All of these results indicated the effects of CKs in stress tolerance and involvement of antioxidant systems in response to manipulation of CK homeostasis. It was also reported that AtCKX transgenic centaury roots showed higher salinity tolerance compared to shoots 15 . These results are in accordance with findings describing that the ckx overexpression enhanced root growth, nutrient uptake and drought tolerance 54,57 . Additionally, several literature data indicate that the growth and development of root tissues could be important in CK-regulated responses to water limiting conditions, including the differentiation of vascular tissue 58 and lignification 48 .
Together with CKs, auxins play a crucial role in regulation of plant growth and development. They are also involved in plant tolerance against various environmental stresses like salinity, however, mechanisms of salt stress regulation by these phytohormones are not yet clear enough. Several studies revealed IAA accumulation under salt treatments in Arabidopsis, maize and Prosopis strombulifera 18,59,60 . A cross-talk between auxins and CKs has been widely studied and possible antagonistic roles in plant defense responses were revealed. Specific mechanisms that modulate this cross-talk are, however, still unknown 61,62 . Our results showed altered endogenous www.nature.com/scientificreports/ IAA content in shoots and roots of non-transformed as well as AtCKX-overexpressing centaury plants grown on graded NaCl concentrations in vitro. Interestingly, non-transformed and AtCKX1-29 shoots and roots showed decreased IAA/bioactive ratio in response to salinity whereas this ratio in both AtCKX2 transgenic lines varied considerably. A close connection between IAA levels and CK contents has been previously reported in AtCKX1 and AtCKX2 centaury transformants 51 that contained lower amounts of endogenous IAA in shoots as well as in roots. Consequently, the IAA/bioactive CKs ratios showed a significant variation in the shoots and roots of all analysed AtCKX transformed centaury plants. Previously it was shown that auxins affect CK degradation by downregulating ckx genes 63 . Jones et al. 64 showed that elevated CK level increased auxin biosynthesis in young, developing root and leaf tissues while decreased endogenous CK levels reduced auxin biosynthesis in seedlings and root meristem tissues of Arabidopsis. Literature data also demonstrated reduced IAA contents in AtCKX1 and AtCKX2 Arabidopsis seedlings 10 . On the other hand, overexpression of AtCKX3 transgenes did not affect endogenous IAA content in tobacco leaves 65 . Varying endogenous IAA and CKs contents and ratios may be explained by the fact that auxin-cytokinin cross-talk pathways are dependent on the developmental stage and tissue specific.
Considering that the plant responses to stress conditions involve numerous physiological and biochemical changes, including alterations in concentrations and ratios of endogenous stress phytohormones, it was observed that increasing salinity in centaury plants is associated with an increment of ABA and JA as well as a decrement of SA contents. In non-transformed and transgenic AtCKX centaury shoots and roots, ABA and JA levels were elevated on increased NaCl concentrations, reaching the maxima on 200 mM NaCl. ABA is well known to represent an endogenous signal molecule that helps plants to survive unfavorable environmental conditions. Accordingly, the exposure of plants to drought and high salt stress resulted in a proportional enhancement of ABA and JA levels in salt-sensitive and salt-tolerant plant species 66 . Increased endogenous ABA and JA contents were also observed in Iris hexagona as a response to salinity 37 . All these reports indicated that higher ABA and JA levels could act as effective protectors against salt stress conditions. Also, there are literature data describing exogenous ABA application in raising the plant salt tolerance [67][68][69] . Likewise, exogenous application of JA was www.nature.com/scientificreports/ successfully applied in order to recover the salt-induced defects on plants 6,70,71 . Exogenously amended stress hormones during salt treatment may change the balance of endogenous hormones, and thus provide an important trait for understanding the protection mechanisms against the salt stress. All of these results clearly demonstrate that ABA and JA have positive roles not only during wounding and biotic stress but also as effective protection agents against high salinity. SA is generally associated with plant biotic stress. However, numerous reports regarding SA participation in plant responses to abiotic stresses, salinity especially, have been documented recently. Our results showed enhanced SA content in non-transformed shoots and roots in comparison to centaury AtCKX lines. Some plant species contain very high basal levels of SA and consequently, SA does not appear to act as an effective signal molecule for defense genes and inducing the disease resistance 72 . It is known that CKs enhance the SA response which results in increased transcription of defense-related genes such as SID2 for salicylic acid biosynthesis and PR1, a marker gene for salicylic acid responses 73 . Accoedingly, it was expected that AtCKX centaury plants with altered endogenous CK content show different SA content comparing to non-transformed. Our results also showed declined SA contents in shoots and roots of non-transformed as well as transgenic AtCKX centaury plants grown on increased NaCl concentrations in vitro. The lowest endogenous SA content was observed on 200 mM NaCl. These results correlate with a reported decline of endogenous SA levels in Iris hexagona 37 and rice seedlings 36 under salt stress. However, some studies showed that SA accumulation is a signal of injury of plant tissues under adverse environmental conditions 74,75 . Just because some studies have reported the improved stress tolerance by SA accumulation whereas others have demonstrated less stress tolerance when the endogenous SA level remain high, the role of SA in plant responses to abiotic stress is not clear yet. It is also reported that SA considerable cross-talk with other plant hormones is implicated in drought and salinity tolerance mechanisms. In general, antagonistic interactions between SA and ABA are well known 76,77 . Beside, numerous literature data are describing that exogenous application of SA elevated salt stress resistance. The ameliorative effects of SA have been well documented including salt tolerance in crops such as bean 78 and mustard 79 . All of these described www.nature.com/scientificreports/ studies demonstrated that endogenous SA level has a plastic response that is modified depending on the plant organ, particular species or exposed type of stress. Considering that plants are continuously subjected to exposure to multiple stresses under natural conditions, endogenous hormones need to be interconnected in order to allow them to coordinate the reactions leading to survival using often limited resources. Exploring the interaction of endogenous phytohormones still remains a quite issue considering that all phytohormones excessively involved in cross-talk with each other and also affect and alter each other biosynthesis.

Conclusion
Altered contents and ratio of endogenous phytohormones such as CKs, auxin IAA, ABA, JA and SA in nontransformed and AtCKX transgenic lines indicated their involvement in responses of centaury plants to salinity. Considering that AtCKX transgenic centaury plants grow on medium supplemented with graded NaCl concentrations but without exogenous application of plant growth regulators, it can be concluded that endogenous phytohormones act as a main determinant for the salt tolerance. The majority of plant hormonal investigations are based on their exogenous applications but the knowledge of changing endogenous phytohormone profiles in response to salt stress during plant growth still remains limited. Although it is known that different plant hormones cross-talk mutually, the mechanisms underlying in these complex signal pathways have not yet been described thoroughly. Unravelling the mechanisms of different phytohormone actions undoubtedly represents a promising tool for future investigations.

Methods
Plant material and culture conditions. All experiments were performed with plant material originating from C. erythraea Rafn. seeds obtained from Jelitto Staudensamen GmbH, Schwarmstedt, Germany. Transgenic AtCKX centaury (Centaurium erythraea Rafn.) plants were obtained as previously described 80 . Shoots and roots of non-transformed, one transgenic AtCKX1 (AtCKX1-29) and two AtCKX2 (AtCKX2-17 and AtCKX2-26) centaury lines (≈ 10 mm long) were separately cultured in vitro, on half-strength MS medium (½MS 81 ). Shoot explants (10) were cultured in 1 l bottles with 100 ml of culture media, while root explants (20)(21)(22)(23)(24)(25) were grown in petri dishes (∅ 10 cm) filled with 25 ml of culture media. The selection of three transgenic lines was based on overexpression of AtCKX genes in centaury shoots and roots which resulted in an altered CKs profile leading to a decline of bioactive CK levels and, at the same time, increased contents of storage CK forms, inactive CK forms and/or CK nucleotides 51 . The medium was solidified with 0.7% agar and supplemented with 3% sucrose and 100 mg l −1 myo-inositol. All explants were cultured on ½MS hormone free or ½MS medium supplemented with graded NaCl concentrations ( in 70% methanol) containing hormones of basic character (CKs). The SPE eluates were evaporated to dryness and pellets reconstituted in 30 μl of 5% methanol in water. An aliquot (10 µL) of the extract was injected on an Ultimate 3000 high-performance liquid chromatograph (Dionex, Bannockburn, IL, U.S.A.) coupled to a 3200 Q TRAP hybrid triple quadrupole/linear ion trap mass spectrometer (Applied Biosystems, Foster City, CA, U.S.A.) set in selected reaction monitoring mode. Quantitative analysis was done using isotope dilution method with multilevel calibration curves (r2 > 0.99). Data processing was carried out with Analyst 1.5 software (Applied Biosystems). The concentration of all analysed phytohormones are presented as pmol/g of sample fresh weight.
Statistical analysis. The effect of graded NaCl concentrations (0-200 mM) on endogenous phytohormones content of 8-week-old non-transformed and transgenic AtCKX shoots and roots were evaluated using standard two-factor analysis of variance (ANOVA). Determination of endogenous phytohormones in three biological samples (for shoots and roots) per non-transformed and each transgenic line was repeated twice. Results are presented as mean ± SE. The comparisons between the mean values were made using a Fisher LSD (the least significant difference) post-hoc test calculated at a confidence level of p ≤ 0.05.