Comparative analysis of phenolic acids and flavonoids in shoot cultures of Eryngium alpinum L.: an endangered and protected species with medicinal value

Phenolic acids and flavonoids, important bioactive compounds of polyphenols, play a significant role in plants; their impact, mainly as antioxidants, on human health have been of great interest in recent years. The genetically uniform shoots of Eryngium alpinum L. cultured in vitro, developed via axillary buds and regenerated from callus tissue, maintained on the media supplemented with various plant growth regulators, were subjected to the phenolic acids and flavonoids quantitative analysis applying HPLC-DAD technique. In vitro cultures give the opportunity to perform the phytochemical studies on the protected species without harvesting the plant material from natural environment. Depending on the hormonal supplementation, the biomass from the shoot cultures accumulated from 11.41 to 25 times more phenolic acids [the total content ranged from 158.66 to 1817.96 mg/100 g of dry weight (DW)] and from 6.8 to 11.8 times more flavonoids (the total content ranged from 29.30 to 51.30 mg/100 g DW) than the shoots from the soil-grown plant. The polyphenols present in the shoot cultures include two phenolic acids: 3,4-dihydroxyphenylacetic and caffeic, four depsides: caftaric, neochlorogenic, chlorogenic, isochlorogenic, and rosmarinic acids, and flavonoids: aglycone—isoquercetin and glucoside—quercitrin. Most of them (apart from chlorogenic and rosmarinic acids) were detected for the first time in this species cultured in vitro. To our best knowledge, the present report is the first one that discusses establishment of Eryngium alpinum L. in vitro cultures and the shoot and callus biomass capacity to produce two subgroups of polyphenols i.e. phenolic acids and flavonoids. This is the first report on establishment of Eryngium alpinum in vitro cultures and the shoot and callus biomass capacity to produce polyphenols—phenolic acids and flavonoids.


Introduction
Eryngium alpinum L., a perennial species characterized by blue amethyst colored flowers, belongs to the Saniculoideae subfamily of the Apiaceae family (Cullen et al. 1997). It can be found across the European Alps, in open and sunny habitats, at altitudes between 1500 and 2000 m a. s. l. The population of the plant is in decline due to its cutting for ornamental purposes and collecting seeds; recreational activities, such as skiing, are degrading its habitat as well. Spring grazing, seasonal mowing, agricultural practices, mowing of meadows by farmers could have an enormous impact on the reproductive success (Andrello et al. 2017). The species is listed on Annex II of the Habitats Directive and under Appendix I of the Convention on the Conservation of European Wildlife and Natural Habitats (the Bern Convention); it is protected by the European Habitat Directive of Natura 2000, the national red lists of protected species and it is also considered vulnerable by International Union for Conservation of Nature (Gygax et al. 2013).
Due to the unavailability of the plant material, little research on this taxon was carried out. Only a few papers indicated the presence of phenolic acids, flavonoids and the essential oil in the organs of the soil-grown plant (Crowden et al. 1969;Le Claire et al. 2005;Dunkic et al. 2013). The identification of two flavonoids, namely quercetin and kaempferol, in the leaves was described by Crowden et al. (1969). The roots of alpine eryngo are known to contain chlorogenic acid, R-(+)-rosmarinic acid and its derivative-R-(+)-3′-Oβ-d-glucopyranosyl rosmarinic acid (Le Claire et al. 2005). Moreover, the crude extract of the roots demonstrated moderate radical scavenging activity toward 2,2-diphenyl-1-picrylhydrazyl (Le Claire et al. 2005). The dominant components identified in the essential oil obtained by hydrodistillation of the aerial parts were caryophyllene, bibyclogermacrene and germacrene (sesquiterpene hydrocarbons), cariophyllene oxide and α-bisabolol (oxygenated sesquiterpenes), and camphor (oxygenated monoterpene). The essential oil showed promising antiphytoviral effect, which is probably correlated with a high content of β-caryophyllene and caryophyllene oxidate (Dunkic et al. 2013).
The Global Strategy for Plant Conservation promotes the development of both in situ and ex situ conservation methods of rare and vulnerable species. The in vitro cultures of E. alpinum are a part of the collection of protected species of the Department of Pharmaceutical Botany and Plant Biotechnology of Poznan University of Medical Sciences (PUMS). The use of in vitro cultures of alpine eryngo allows for conducting the phytochemical analyses and testing further biological activities of this species without devastating its natural sites to obtain the plant material.
The aim of the study was to obtain the shoot biomass of E. alpinum under in vitro conditions and to comprehensively analyze the quality and quantity of phenolic acids and flavonoids. Our study involved the shoots from the selected media with various hormonal supplementation. The shoots were developed in two ways: from meristematic tissue of the lateral buds from the soil-grown plant and via indirect morphogenesis (from the callus). Then the shoots were multiplied by means of the axillary bud proliferation technique.
The protocol allows for harvesting the high-quality raw material from alpine eryngo without decreasing the medicinal quality and quantity of bioactive compounds.

Source of plant material origin
The fruits of E. alpinum collected from the botanical gardens had very mature seeds. As it is known from literature, the seeds of this taxon become dormant soon after harvest (Njenga 1995). Therefore, the seedlings that could become the primary explants for the further biotechnological research were not obtained in vitro.

Establishment of in vitro cultures
The soft shoots with lateral buds were harvested; the collected explants were washed in double-distilled water (DDW) for 5 min, and 70% ethanol for 30 s, and then lightly disinfected with 200 mL of 50% (v/v) bleach solution (3.3% active calcium hypochlorite) with 2-3 drops of Tween 80 for 5 min. The surface disinfected shoots were taken to a laminar air-cabin and washed with autoclaved DDW to complete three separate washes of 3 min duration each. The shoots were also trimmed to remove all unwanted fragments. After the final wash, the individual explants (1-2 cm) were transferred to 250 cm 3 flasks with 50 cm 3 of a solidified basal MS medium (Murashige and Skoog 1962) with 3% sucrose and plant growth regulators (PGRs), namely 6-benzylaminopurine (BAP), indole-3-acetic acid (IAA) and GA 3 (gibberellic acid) at concentrations of 0.5 mg/L or 1.0 mg/L (Table 1), 0.76% agar and pH set to 5.8 prior to autoclaving at 121 °C, 105 kPa for 20 min. All PGRs as well as agar were purchased from Sigma-Aldrich (Saint Louis, USA). The culture vessels were placed in a growth chamber (21 °C; with a 16 h light/8 h dark photoperiod; 55 µmol/m 2 s photon flux density provided by cool-white fluorescent lamps) and subcultured every 5 weeks. Multiplication of the shoots was replicated three times for each hormonal treatment using at least ten explants (2-3 per flask).
The leaves (1 cm 2 segments) from the same individual were used for the callus initiation. The whole leaves were washed in DDW for 5 min and 70% ethanol for 30 s, and then disinfected with 300 mL of 70% (v/v) bleach solution (3.3% active calcium hypochlorite) with 2-3 drops of Tween 80 for 7 min. The surface disinfected leaves were washed three times for 3 min. After the final wash, the leaf explants were transferred to 150 cm 3 flasks containing 30 cm 3 of a solidified MS basal medium with 3,6-dichloro-2-metoxybenzoic acid (dicamba, Dic 1.0 mg/L) and thidiazuron (TDZ 0.5 mg/L). The callus clumps observed after three weeks of culture on the MS medium with Dic and TDZ were collected, divided into the clumps about the same size (1 cm 2 ) and distributed on the regeneration MS media (150 cm 3 flasks containing 30 cm 3 aliquots of medium) with the selected PGRs: BAP, IAA and 2,4-D (2,4-dichlorophenoxyacetic acid) at different concentrations. The number of shoots regenerated from a callus clump was recorded. The callus cultures and callus-derived shoots were grown under the same light and temperature conditions as the shoot cultures obtained from the lateral buds.

Genome size estimation
The leaves of in vitro multiplied shoots, the shoots regenerated from the callus and callus biomass were used for the nuclear DNA content estimation. A nuclei isolation buffer  (Galbraith et al. 1983) with 1% (v/v) polyvinylpyrrolidone (PVP-10), propidium iodide (PI; 50 µg/cm 3 ) and ribonuclease A (50 µg/cm 3 ) was used. Petunia hybrida P × Pc6 (2.85 pg/2C; Marie and Brown 1993) was used as an internal standard. For each sample, 5000-8000 nuclei were analysed using a CyFlow SL Green flow cytometer (Partec GmbH, Münster, Germany) equipped with a high-grade solid-state laser with green light emission at 532 nm, a long-pass filter RG 590 E, DM 560 A as well as with side (SSC) and forward (FSC) scatters using linear amplification. FloMax software (Partec GmbH, Münster, Germany) was used. The analyses were replicated at least three times for each plant material. The coefficient of variation (CV) of G 0 /G 1 peak of E. alpinum ranged from 3.2 to 4.6%. The nuclear DNA content was calculated using the linear relationship between the ratio of the 2C peak positions E. alpinum/P. hybrida on a histogram of fluorescence intensities.

Statistical analysis
The data were analysed using a one-way analysis of variance (ANOVA) and the statistical significance was determined applying Duncan's POST-HOC test (P-value of 0.05). All the analyses were conducted employing STATISTICA v. 13 (StatSoft, Inc. 2015).

Results and discussion
To our best knowledge Eryngium alpinum L. was introduced into the in vitro cultures for the first time and the shoot cultures were established in our laboratory to study their capability of producing desired polyphenols, i.e. phenolic acids and flavonoids. All the tested hormonal treatments had a significant impact on shoot multiplication of alpine eryngo via axillary bud development (Fig. 1a-d). However, the induction percentage of the new shoots was relatively low for the explants maintained on the MS medium without PGRs (58%) comparing with those cultured on the enriched media (86-100%, Table 1). The highest multiplication rate (6.18 ± 0.45) was achieved for the explants placed on MS with BAP 1.0 mg/L, IAA 0.5 mg/L and GA 3 1.0 mg/L (Fig. 1c, d). It is worth noticing that those shoots were also the longest (3.5 cm; Table 1). The regenerated shoots were vigorous and welldeveloped. In case of E. planum, the highest mean number of shoots developed from axillary buds was higher, i.e. 15.58 ± 0.54 − 17.10 ± 0.60 shoots per explant, depending on the culture media: MS + BAP 1.0 mg/L + IAA 1.0 mg/L or MS + BAP 1.0 mg/L + IAA 0.1 mg/L (Thiem et al. 2013). For E. campestre, more shoots (13.30 ± 3.73) were obtained when cultured on the same media composition (Kikowska et al. 2016). These two species were relatively easy to culture and exhibited a great potential of morphogenetic response. A completely different pattern of morphogenetic response was observed for E. maritimum (Kikowska et al. 2014) and E. alpinum (present studies)-rare and protected species living in extremely difficult climatic and soil conditions and exhibiting problems with seed germination. Although the fruits were subjected to scarification and stratification, the seedlings form the seeds could not be obtained for both species. The efficiency of shoot multiplication for E. maritimum varied between 1.2 ± 0.20 and 4.4 ± 0.24 shoots per explant on different media variants (Kikowska et al. 2014).
In our studies, the first sights of the callus induction were observed after 3 weeks of culture on MS with Dic 1.0 mg/L and TDZ 0.5 mg/L (Fig. 1e-f). The callus arising from the leaf explants was green, compact and hard. The first shoots arrived on the MS regeneration media with BAP/IAA/2,4-D after 2 weeks of culture. The callus-derived shoots were not highly-multiplied, relatively low, and with leaves of narrow blades (Fig. 1g). The regeneration media proved to be effective at converting callus tissue into the shoots with the efficiency ranging from 2.60 ± 0.31 to 4.11 ± 0.54 shoots per callus clump; the best result was observed for the explants cultured applying BAP 1.0 mg/L and 2,4-D 1.0 mg/L. The MS medium without PGRs turned out to be the least effective (Table 2). There are no reports on callus induction and indirect organogenesis in Eryngium species. The influence of BAP and 2,4-D on callus mediated shoot regeneration was evaluated for some medicinal plants: Melicope lunu-ankenda (Gaertn.) Merr. (Rahman et al. 2015), Eclipta alba (L.) Hassk (Sharma et al. 2013) or Arnica montana L. (Petrova et al. 2011).
The confirmation of genetic fidelity is of importance in plants used as a source of biological active metabolites. In our study the 2C DNA content of E. alpinum biomass-the leaves from the shoots developed from the axillary buds (on the media with and without PGRs) as well as regenerated from the callus and in the callus was similar and ranged from 2.342 ± 0.017 to 2.363 ± 0.011. There were no significant differences in DNA content between the plant material ( Table 3). The results concerning the genome size stability were notably consistent with our previous studies on other Eryngium species (Thiem et al. 2013;Kikowska et al. 2014Kikowska et al. , 2016. The harvested shoots and callus biomass as well as the shoots from the soil-grown plants were subjected to the phytochemical analysis by HPLC-DAD method of bioactive polyphenolic compounds, i.e. phenolic acids (PAs) (23 compounds tested) and flavonoids (Fs) (14 compounds tested) (Table 4; Figs. 2, 3).
The presence of 3,4-dihydroxyphenylacetic (13.2 mg/100 g DW) and small amounts (below 0.4 mg/100 g DW) of depsides: chlorogenic, isochlorogenic and rosmarinic acids was confirmed in the extracts from the shoots of the soil-grown plants from PAs (Table 4). In these extracts the low amount of isoquercetin (4.47 mg/100 g DW) was also detected ( Table 5). The presence of seven compounds including two phenolic acids i.e. 3,4-dihydroxyphenylacetic and caffeic as well as five depsides i.e. caftaric, neochlorogenic, chlorogenic, isochlorogenic, and rosmarinic acids was confirmed in the extracts from the in vitro shoots and callus cultures from PAs (Table 4). Aglycone-isoquercetin and glucoside-quercitrin were confirmed in the shoot cultures form Fs. Fs were not detected in the callus cultures (Table 5).
In the present study, the shoots propagated under in vitro condition produced from 11.41 to 25.10 times more PAs than the shoots from the ground plant (158.66-349.12 mg/100 g  DW and 13.91 mg/100 g DW, respectively). At the same time, their composition was richer in caftaric acid, neochlorogenic acid and caffeic acid in comparison with the soilgrown plant material. Generally, in the in vitro material, 3,4-dihydroxyphenylacetic acid, chlorogenic acid (CGA) and rosmarinic acid (RA) were the quantitatively dominant PAs. Surprisingly, the content of CGA and RA was disproportionately lower in the shoots of the ground plant than in the shoots cultured in vitro. The content of the selected phenolic acids has been already measured in the different species of Eryngium including E. planum (Kikowska et al. 2012(Kikowska et al. , 2015Thiem et al. 2013), E. maritimum (Kikowska et al. 2014) and E. campestre (Kikowska et al. 2016). The three phenolic acids, i.e. rosmarinic acid (RA), chlorogenic acid (CGA) and caffeic acid (CA), were detected in the plant material of E. planum. The highest content of the sum of RA, CGA and CA was found in the biomass from the in vitro cultures (shoots and roots) and was considerably higher than in the shoots and roots of the ground plant (Kikowska et al. 2012;Thiem et al. 2013). For E. maritimum the content of RA and CGA was higher in the shoots and roots from the in vitro cultures than in the analogous organs of the ground plant (6.75 and 12 times, respectively). In case of E. campestre, the sum of RA, its hexoside and CGA in the biomass from the shoot culture was lower than in the shoots of the ground plant (Kikowska et al. 2016).
In this study, the best source of CGA and RA was admittedly the callus with the impressive amount of both, especially of RA (1429.28 ± 74.29 mg/100 g DW and 546.92 ± 34.90, depending on the tested medium). Moreover, the callus maintained on induction medium contained a higher amount of PAs (1817.96 mg/100 g DW) than the callus cultured on regeneration medium (777.55 mg/100 g DW), but the shoots regenerated from it were characterized by the lowest content of PAs (158.66 mg/100 g DW), which was probably related to the influence of different PGRs (2,4-D instead of IAA and GA 3 ). As it turned out, the tissue (callus), which was used only for the regeneration of shoots, was the richest source of PAs (Table 4). The high production of PAs in the callus cultures was proved for some other species. The studies on Schisandra chinensis were performed on tissue at different stages of organogenesis in vitro on a few variants of the MS Table 3 Nuclear DNA content in leaves from in vitro-derived/regenerated shoots and callus of Eryngium alpinum L.     medium, containing PGRs (BAP and NAA). The total contents of the examined metabolites in the biomass of shootdifferentiating callus ranged from 14.90 to 60.05 mg/100 g DW. The total content of the compounds in the biomass from the callus culture was higher and amounted to 74.54 and 78.24 mg/100 g DW. The maximum total contents of phenolic acids in both types of the in vitro cultures were greater than in the leaves (4.55 mg/100 g DW) of the parent plants (Szopa and Ekiert 2012). In the studies on the callus of Anethum graveolenes (Szopa and Ekiert 2015) maintained on the LS medium (Linsmaier and Skoog 1965), variants differing in the PGRs content, the high amounts of PAs, the maximum total amounts of phenolic acids obtained in the in vitro grown biomass were slightly higher than their amounts in the fruits of the parent plant analyzed for comparison ( (Szopa et al. 2013;Szopa and Ekiert 2014). The total amounts of the estimated compounds for A. arbutifolia were from 1.12 to 3.07 times and for A. × prunifolia from 2.71 to 9.11 times higher in the shoots than in the calluses growing on identical MS variants (Szopa et al. 2018). The content of flavonoids in E. alpinum was significantly higher (from 6.8 to 11.48 times) in the shoots cultured in vitro than in the shoots from the ground plant and depended on a medium hormonal variant. The flavonoid content increased with higher GA 3 concentration. The qualitative differences were confirmed: quercitrin was absent in the control shoots (the soil-grown plant), but the callus did not produce flavonoids at all. The shoots regenerated from the callus were able to produce flavonoids in contrast to callus tissue, from which they derived (Table 5). The absence of flavonoids in the callus cultures of some plants, including a related species-Eryngium planum, was recorded (Thiem et al. 2001;Budzianowska et al. 2004;Kikowska et al. 2012).

Conclusion
In vitro cultures of Eryngium alpinum L. were initiated and established for the first time. Moreover, in vitro culture of genetically uniform shoots developed via axillary buds may be the method of ex situ conservation of this endangered and protected species. The shoot and callus biomass can be considered as a valuable alternative source of desired polyphenols.
Author contributions M.K. is the author of conception, who design the study, initiated and maintained the in vitro cultures of Eryngium alpinum L., and wrote the manuscript. M.K. and B.T. collected the plant material. M.R. and E.S. estimated the genome size of the plant material. A.S., M.K.-S. performed and elaborated the phytochemical analyses. H.E. corrected and improved the manuscript. All the authors contributed to data interpretation, critical revision of the manuscript and its final approval.

Compliance with ethical standards
Conflict of interest The authors declare that they have no conflict of interest.
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creat iveco mmons .org/licen ses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.