Stimulation of Lignan Production in Schisandra rubriflora In Vitro Cultures by Elicitation

The study investigated the effect of elicitation with: chitosan (CH) (200 mg/L), yeast extract (YeE) (3000 mg/L), ethephon (ETH) (25 µM/L), and methyl jasmonate (MeJA) (50 µM/L), on lignan accumulation in agitated and bioreactor (Plantform temporary immersion systems) microshoot cultures of female (F) and male (M) Schisandra rubriflora Rehd. et Wils. (Schisandraceae) lines. The elicitors were supplemented on the 10th day of culture. Biomasses were collected at 24 h and 48 h, and 4, 6, and 8 days after the addition of each elicitor. The 24 compounds from the dibenzocyclooctadiene, aryltetralin, dibenzylbutane, and tetrahydrofuran lignans and neolignans were determined qualitatively and quantitatively in biomass extracts using the UHPLC–MS/MS method. The highest total contents [mg/100 g DW] of lignans were: for CH-95.00 (F, day 6) and 323.30 (M, 48 h); for YeE 104.30 (F, day 8) and 353.17 (M, day 4); for ETH 124.50 (F, 48 h) and 334.90 (M, day 4); and for MeJA 89.70 (F, 48 h) and 368.50 (M, 24 h). In the biomass extracts of M cultures grown in bioreactors, the highest total lignan content was obtained after MeJA elicitation (153.20 mg/100 g DW). The maximum total lignan contents in the biomass extracts from agitated and bioreactor cultures were 3.29 and 1.13 times higher, respectively, than in the extracts from the non-elicited cultures. The poor understanding of the chemical composition and the lack of studies in the field of plant biotechnology of S. rubriflora emphasize the innovativeness of the research.


Introduction
The Schisandraceae family includes twenty-seven species of the genus Schisandra. In modern phytotherapy, only one of them is most often used-Schisandra chinensis (Turcz.) Baill.-Chinese magnolia vine. Knowledge of the medicinal and cosmetic properties and alimentary value of the raw material-Schisandrae chinensis fructus, came from traditional Chinese medicine (TCM) [1,2]. The European monograph, the raw material appeared for the first time in 2008 in the European Pharmacopoeia 6th [3]. The raw material is also known in North America [4]. It also has a monograph in the International Pharmacopoeia published by WHO [5]. S. chinensis fruit extract shows inter alia, hepatoprotective, adaptogenic, anti-inflammatory, antibacterial and antioxidant activities [2,6,7]. Knowledge about the phytochemical composition and medicinal or alimentary properties of other species of the Schisandra genus, compared to S. chinensis, is very poor [8].
The subject of this study is the endemic, dioecious species of the genus Schisandra that occurs naturally only in the western part of the Sichuan province (south-western part of China)-Schisandra rubriflora Rehd. et Wils. (Schisandra chinensis). It's a species known in TCM; mainly used as a tonic and sedative. It is also recommended in the treatment of hepatitis, chronic gastroenteritis, and neurasthenia [8][9][10].    Table 1. Comparison of biomass gains (Gi ± SD) of F and M lines of S. rubriflora agitated cultures depending on harvesting time and the elicitor used. Data expressed as the mean value ± SD (n = 3). Different superscript letters (a-c) within a row indicate significant differences between means (Duncan's multiple range test; p < 0.05). At 24 h after the addition of the elicitor, the biomass gains of elicited cultures, except for MeJA elicitation for lines M, were higher than those of the control cultures. The highest biomass gain of lines M at 24 h was recorded for YeE elicitation (max. Gi = 41.76), it was 2.06 times higher compared to the control. At 24 h after addition, elicitors did not affect the biomass gains of lines F except for YeE elicitation (max. Gi = 23.35). This gain was 1.28 times higher compared to the control (Table 1).

Time of Harvesting
At 48 h after the addition of the elicitor, higher biomass gains were recorded for lines M except for those elicited with ETH and MeJA. The highest biomass gain of lines M was found for CH elicitation at 48 h (max. Gi = 37.57), it was 1.43 times higher compared to the control. The elicitors at 48 h after their addition also affected the biomass gain of the female lines except for MeJA elicitation. A high value of Gi coefficient was also found after ETH elicitation (max. Gi = 27.47). This gain was 1.94 times higher compared to the control (Table 1).
Supplementation with elicitors had a negative effect on the biomass gains of the experimental cultures harvested after 4 days. Only after CH elicitation for the biomass of line F, the Gi value (26.67) was 1.07 times higher compared to the control (Table 1).
For microshoots of agitated cultures of lines M 6 days after the addition of elicitors, only CH and MeJA showed a slight increase in Gi values compared to the control. The highest increase in biomass of line M was found after CH elicitation (max. Gi = 52.63), it was 1. 16 times higher compared to the control (Table 1). For line F, only two elicitors, CH and YeE, affected biomass growth after 6 days of addition. The highest Gi value was found after YeE elicitation (max. Gi = 30.00). This increase was 1.32 times higher compared to the control (Table 1).
After 8 days of elicitation application for line M cultures, only CH and MeJA were found to have a higher Gi value compared to the control cultures. The highest biomass gain for lines M was found for MeJA elicitation (max. Gi = 69.26), this was 1.5 times higher than for the control. On the other hand, for line F, the highest Gi value was found after MeJA elicitation (Gi = 42.99). This gain was 2.39 times higher compared to the control (Table 1).
The notable Gi values in the same time points are diverse. That is caused by the type of elicitor and the duration time of elicitor treatment. The elicitor addition often caused a decrease in in vitro culture biomass growth while the production of secondary metabolites increase. Such a fact is well known in plant biotechnology studies and has been described before, e.g., after CH elicitation in Hypericum perforatum root cultures [25], MeJA elicitation in Fagonia indica adventitious root cultures [26], YeE elicitation in Panax ginseng cell culture [27], or Aspergillus flavus fungus elicitation in Catharanthus roseus callus cultures [28].
Analysis of the literature allows a direct comparison of data on the effect of elicitation with two elicitors: YeE and CH on the value of the Gi index for in vitro cultures of S. chinensis species, and the experiment carried out in the present work. The highest recorded Gi index values for YeE elicitation of S. chinensis species were observed after YeE addition at a concentration of 1000 mg/L. The Gi value ranged from 33.37 to 42.89, where the value for the control sample was 40.74. The other, higher concentrations of YeE, 3000 mg/L and 5000 mg/L, caused a decrease in biomass growth especially when YeE was supplemented on the first day of culture [22]. In the course of our experiment, a decrease in the biomass growth of in vitro cultures of S. rubriflora was found 4 days after the addition of YeE at a concentration of 3000 mg/L. For line M it was also observable on day 6 and day 8 of culture. Line M of S. rubriflora was found to be much more sensitive to YeE compared to line F. Elicitation of S. chinensis with CH at concentrations in the range of 25-200 mg/L did not adversely affect microshoots growth. The Gi index was comparable to control cultures. The highest Gi value was found for CH supplementation at concentrations of 50 and 100 mg/L on the first day of culture. It was 57.39 and 56.87, respectively; where the Gi value for the control sample was 40.00 [22]. In the course of our experiment for in vitro cultures of S. rubriflora, the increase in CH elicited biomass for cultures of line F occurred only at 48 h after elicitor addition, it was the highest at day 8 after elicitor addition, while cultures of line M showed the highest increase at 24 h after elicitor addition (Table 1).

The Influence of Elicitation on Lignan Production
The study proved the significant effect of the applied elicitation on the accumulation of lignans in cultures of lines F and M of S. rubriflora. In biomass extracts, 22 compounds from four groups of lignans were qualitatively and quantitatively determined: dibenzocyclooctadiene lignans (schisantherin A and B, schisandrin, schisandrin C, gomisin A, D, G, J, N, O, 6-O-benzoylgomisin O, schisandrin A, rubrisandrin A, epigomisin O, schisanhenol, interiotherin C, angeloylgomisin H and O), aryltetralin lignans (wulignan A1), dibenzylbutane lignans (pregomisin, mesodihydroguaiaretic acid), and tetrahydrofuran lignans (fragransin A2). In addition, 2 compounds from the dihydrobenzofuran group of neolignans (licarin A and B) were also found in the analyzed extracts.
The study confirmed the effect of the elicitation schemes used on the accumulation of metabolites in the microshoot culture biomass extracts analyzed, while only trace amounts were found in the culture media (<5 mg/L).
Detailed results of quantitative analyses in the control sample and depending on the elicitation scheme used and on the time of tissue harvesting are presented in Supplementary Data (Tables S1-S5).
The highest obtained contents of the analyzed compounds in the course of the whole experiment were as follows: wulignan A1 (max. 0. 36 (Tables S1-S5 and Tables 2-6).       Total content of lignans was highest for line M elicited with MeJA, and was 368.50 mg/100 g DW, 3.29 times higher than in the control sample (Tables S1-S5 and 2-6).
The obtained results were compared with elicitation in in vitro cultures of wellestablished in phytotherapy S. chinensis which affected the production of dibenzocyclooctadiene lignans [22]. Cultures of S. chinensis were maintained on MS medium containing 3 mg/L BA and 1 mg/L NAA. Cultures were treated with elicitors at 10 and 20 days of culture. The elicitors tested were cadmium chloride (CdCl 2 ), YeE, CH, MeJA and the permeabilizing agent dimethyl sulphoxide (DMSO). The duration of culture was 30 days. In the biomass extracts of the experimental cultures, the content of dibenzocyclooctadiene lignans was determined by DAD-HPLC. The total content of lignans obtained in cultures elicited with CdCl 2 was twice as high as in non-elicited control samples. The content of secondary metabolites after the use of this elicitor was about 730 mg/100 g DW. Elicitation with CH increased lignan production by 1.35 times (500 mg/100 g DW). The use of YeE led to a 1.8-fold increase in the content of the tested compounds. The identified lignans differed from those determined in in vitro cultures of S. rubriflora species during the present experiment. In the S. chinensis species, only lignans from the dibenzocyclooctadiene lignan group were determined: schisandrin, gomisin A, gomisin G, schisantherin A, schisantherin B, Schisanhenol, deoxyschizandrin, γ-schisandrin, schisandrin C, angeloylgomisin H and Q, schisandrin B, benzoylgomisin P and schisantherin D. Only the effect of elicitation on the compounds found in both species, i.e., dibenzocyclooctadiene lignans, was evaluated comparatively in this study (compounds from the other labeled lignan groups in S. rubriflora were not considered): schisandrin, gomisin A, G, schisantherin A, B, schisanhenol, deoxychizandrin and the total content of these compounds (Table 7). Schisandrin, gomisin A and G, and schisantherin B showed higher content multiplicity relative to the control due to elicitation in S. rubriflora species (3.13, 3.31, 3.91, 5.03 times higher for MeJA, respectively, 50 µM, 24 h,) than in S. chinensis species (2.28, 2.80 times more for CdCl 2 , respectively, 1000 µM, 10 days, 3.03 times more for YeE, 3000 mg/L, day 10; 2.61 times more for YeE, 1000 mg/L, 20 days). The content of these compounds was lower in elicited cultures of S. rubriflora species. Schisantherin A, schisanhenol, deoxyschizandrin showed lower multiplicity of content relative to control upon elicitation in S. rubriflora species (1.99, 1.79 times more, respectively, ETH, 25 µM, 48 h, 1.65 times more, YeE, 3000 mg/L, 4 days) in cultures of S. rubriflora species than in those for S. chinensis (3.11 times more, YeE, 3000 mg/L, day 0; 4.44 times more, CdCl2, 1000 µM, 10 days; 1.95 times more, YeE, 5000 mg/L, days 10). The contents of schisantherin A and schisanhenol were lower in S. rubriflora cultures than in S. chinensis. The only compound that showed a higher content in S. rubriflora cultures was deoxyschizandrin (max. 94.86 mg/100 g DW). The total content of lignans was higher in S. chinensis cultures (max. 730.60 mg/100 g DW) than in S. rubriflora (max. 368.50 mg/100 g DW), while the multiplicity relative to the control was higher in S. rubriflora cultures (3.29 times higher) ( Table 7). Studies on the effect of elicitation on the production of specific compounds from the lignan group are a rather difficult subject of research, and that is probably why they are not a frequent object of study. Studies on a popular compound with anticancer activitypodophyllotoxin are of greatest interest. Kasparova  The highest content of podophyllotoxin was 1.6 mg/g DW, which was 1.14 times higher than in the control sample and was obtained in biomass cultured on medium with 20 g/L sucrose. The researchers found no significant effect of elicitation on the production of podophyllotoxin [32].
Waqar et al. studied the effect of CH elicitation on the accumulation of specific lignans and neolignans: secoisolariciresinol diglucoside, lariciresinol diglucoside, dehydrodiconiferyl alcohol glucoside and guaiacylglycerol-β-coniferyl alcohol ether glucoside in suspension cultures of Linum usitatissimum. Cultures were elicited with CH at a concentration of 10 mg/L on day 10 of culture. The duration of culture was 30 days. Material was harvested at 8, 24 and 48h. Maximum enhancements of 7.3-fold (28 mg/g DW) occurred for lariciresinol diglucoside, 3.5-fold (58.85 mg/g DW) in dehydrodiconiferyl alcohol glucoside and while the least enhancement of 2-fold (18.42 mg/g DW) for secoisolariciresinol diglucoside was observed in CH treated cell cultures than to controls [33]. Nadeem et al. also studied the effect of elicitation, but with a different elicitor, YeE, on the accumulation of lignans and neolignans in suspension cultures of L. usitatissinum species. YeE was added to the medium on day "0" (inoculation) at concentrations of 10, 50, 100, 200, 500 and 1000 mg/L. YeE at a concentration of 200 mg/L caused the highest increase in the accumulation of the compounds: secoisolariciresinol diglucoside (3.36-fold, max. 10.1 mg/g DW), lariciresinol diglucoside (1.3-fold, 11.0 mg/g DW) and dehydrodiconiferyl alcohol glucoside (4.26-fold, max. 21.3 mg/g DW) [34].
Wawrosch et al. studied the effects of elicitation with AgNO 3 , MeJA and YeE on the accumulation of specific furan-type lignans: leoligin and 5-methoxy-leoligin in hairy root cultures of Leontopodium nivale. Elicitors were added to 3-week-old cultures at concentrations of: 15, 30 and 60 µM AgNO 3 , 50, 100, 200, 300 µM MeJA and 1, 2, 5 g/L YeE. The duration of culture was 4 weeks. The highest content of leoligin was recorded after elicitation with 100 µM MeJA (max. 0.05 in %), it was 8.33 times higher than that of the control sample. The content of 5-methoxy-leoligin was highest after supplementation with 15 µM AgNO 3 (max. 0.026 in %), and was 6.5 times higher than that of the control sample [35].
Schmitt and Petersen examined the effect of MeJA elicitation on the accumulation of tetrahydrofuran lignans, pinoresinol and matairesinol, in Forsythia × intermedia suspension cultures. MeJA was added on day "0" at a concentration of 100 µM. Cultures were harvested every other day from the addition of the elicitor. Pinoresinol content increased 3-fold (max. 0.8 mg/g DW) and matairesinol content increased 7-fold (max. 2.7 mg/g DW) relative to the control sample [36].
Sanchez-Sampedro et al. conducted an experiment proving the effect of elicitation with YeE, SA, CH and chitin on the synthesis of flavonolignan: silymarin in suspension cultures of Silybum marianum. SA, CH and chitin did not stimulate silymarin production even at higher concentrations. YeE caused intense browning and significant loss of cell viability after 48 h (at concentrations of 100 and 200 µg/mL). A slight increase in silymarin content was observed following YeE supplementation. MeJA at a concentration of 10 µM/mL was ineffective, but at a concentration of 100 µM/mL caused significant accumulation of silymarin in cells. MeJA alone or in combination with YeE gave the best results. Threeday cultures were treated for 48 h with 50 g/mL YeE, 100 µM MeJA or both elicitors simultaneously. Silymarin content for control samples was max 2.01 mg/g DW. The combination of MeJA and YeE yielded nearly 600% higher accumulation of silymarin in biomass [37].
The contents of individual compounds ranged from trace amounts <0.05 mg/100 g DW to 41.01 mg/100 g DW (gomisin A, MeJA, 24 h). The highest contents of the individual compounds analyzed in the course of the entire experiment were as follows: schisandrin (max. 37.60 mg/100 g DW; MeJA, 1.25 times higher than the control sample), gomisin A (max. 41.01 mg/100 g DW; MeJA, 1.03 times higher than the control sample), and deoxyschisandrin (max. 35.00 mg/100 g DW; MeJA, 1.06 times higher than the control sample) ( Table 8).
follows: MeJA at 24 h, CH at 48 h, YeE at day 4, ETH at day 4, control at day 4. The effect of the elicitors used on the appearance of microshoots was observed only in the case of YeE and MeJA elicitation, where the tissue was slightly browned after elicitor supplementation (Figure 2
The contents of individual compounds ranged from trace amounts <0.05 mg/100 g DW to 41.01 mg/100 g DW (gomisin A, MeJA, 24 h). The highest contents of the individual compounds analyzed in the course of the entire experiment were as follows: schisandrin (max. 37.60 mg/100 g DW; MeJA, 1.25 times higher than the control sample), gomisin A (max. 41.01 mg/100 g DW; MeJA, 1.03 times higher than the control sample), and deoxyschisandrin (max. 35.00 mg/100 g DW; MeJA, 1.06 times higher than the control sample) ( Table 8).
Total lignan contents ranged from 114.80 mg/100 g DW (ETH, 4 days) to 153.20 mg/100 g DW (MeJA, 24h) ( Table 8). The maximum total lignan content obtained was 1.13 times higher than for the control sample.
Trace amounts of lignans were found in culture media (<0.05 mg/100 g DW).
In general, for cultures grown in Plantform bioreactors, the most effective elicitor for which the highest individual and total lignan contents were obtained, was MeJA.  Total lignan contents ranged from 114.80 mg/100 g DW (ETH, 4 days) to 153.20 mg/100 g DW (MeJA, 24h) ( Table 8). The maximum total lignan content obtained was 1.13 times higher than for the control sample.
Trace amounts of lignans were found in culture media (<0.05 mg/100 g DW).
In general, for cultures grown in Plantform bioreactors, the most effective elicitor for which the highest individual and total lignan contents were obtained, was MeJA.
The accumulation of active compounds in plant cultures grown in bioreactors is less frequently studied due to the fact that for these experiments a much larger amount of plant tissue is required to initiate the experiment, as well as the availability of special bioreactor-like structures being limited [38]. Studies on the effect of YeE elicitation on dibenzocyclooctadiene lignan production in S. chinensis microshoot cultures maintained in Plantform bioreactors have been studied by us before [22]. Results proved that the supplementation with 1000 mg/L YeE on the 20th day of the growth cycle was the optimal. Through this elicitation scheme the total content of the estimated dibenzocyclooctadiene lignans was equal to 831.60 mg/100 g DW. The dominant dibenzocyclooctadiene lignans were schisandrin-186.8 mg/100 g DW, angeoyl/tigloyl-gomisin Q-183.4 mg/100 g DW and deoxyschisandrin-100.00 mg/100 g DW. In the cultures of S. rubriflora we tested, maintained in Plantform bioreactors, different lignans were proven to be dominant and their amounts were of lower order ( Table 8).
The effect of elicitation treatments on the production of other lignan groups in the biomass of plant in vitro cultures grown in bioreactors is a new and little-exploited research direction [38]. Dougué Kentsop et al. recently tested production of arylnaphthalene lignanjusticidin B, in L. lewisii adventitious and hairy-roots cultures maintained in the stirred tank bioreactor. Both of the culture types were grown in the bioreactor for 3 weeks and then elicited with 100 µM MeJA and grown for a further one week. The justicidin B content in both cultures after treatment with MeJA doubled in comparison to the control and was to equal 99.2 and 132.6 mg/g DW, respectively [39].

Biotechnological Evaluation of Elicitation Results
The compounds obtained in the course of the experiment were analyzed, and their highest contents were extracted taking into account the elicitation conditions. The collected data were compared in relation to the multiplicity of the content increase in relation to the control and compared with the results of the analysis of the material obtained from the parent plants (Table 9).
For a number of compounds: rubrisandrin A, interiotherin C, schisandrin, gomisin D, J, N and A, schisantherin A, licarin A, and schisandrin C, their contents in extracts from in vitro cultures were higher than in extracts from fruits of parent plant. This is important, as the fruit is widely recognized as Schisandra's raw material. Our study also proved higher values for individual compounds in extracts from in vitro cultures compared to the extracts from leaves and stems of parent plant material (Table 9). In this context, the results obtained have a potential applied nature.
Total lignan content was highest for the line M elicited with MeJA and was 368.50 mg/100 g DW and was 3.29 times higher than that of the control-non-elicited cultures. The content was 1.3 times higher than in the shoots, but 2.6-times lower than in the leaves and 2.1-times lower than in the fruit of the parent plant (Table 9). Table 9. Comparison of maximum lignan contents (mg/100 g DW) obtained from elicitation experiments on microshoot cultures of lines F and M with their contents in the fruits, leaves and shoots of the parent plant S. rubriflora [11].

Plant Material and Microshoot Culture Initiation
Plant material for in vitro culture initiation was obtained as part of cooperation with Clematis-Źródło Dobrych Pnączy (Pruszków, Poland) [40]. Moreover, the fruits, leaves and stems of the parent plant material were obtained from this arboretum. Plant species were identified by Dr. Szczepan Marczyński and Dr. Agnieszka Szopa. For these purposes the leaf buds of about 10 years old female (F) (100 individuals) and male (M) (50 individuals) S. rubriflora (Franch.) Rehd. et Wils specimens were collected in May 2018. Leaf buds were defatted with 70% ethanol (30 s) and then sterilized for 7 min with 0.1% HgCl 2 (mercuric chloride II). Sterile buds were rinsed three times with sterile redistilled water and transferred to agar medium according to Murashige and Skoog (1962) (MS) [41] containing 1 mg/L BA (6-benzyladenine) and 0.5 mg/L NAA (1-naphtaleneacetic acid).
Lignans were determined by ultra-high performance liquid chromatography coupled to a tandem mass spectrometer (UHPLC-MS/MS) technique. The apparatus consisted of a UHPLC Infinity 1260 (Agilent, Wolbrom, Germany) and a 6410 MG/100 G DWQ LC/MS tandem quadrupole mass spectrometer (Agilent, Santa Clara, CA, USA). Samples in a volume of 2 µL were injected onto an analytical column (Kinetex TM C18: 150 × 4.6 mm, 2.7 µm). The analytes were eluted in a gradient of 50% water in methanol (A) and 100% methanol (B) with the addition of 0.1% formic acid in both phases, from 20% to 65% solvent B for 22 min, at a mobile phase flow rate of 0.5 mL/min at 60 • C acc. to [11,18]. In addition, a DAD spectrophotometric detector was connected in the chromatographic system, and the tested compounds were monitored at a wavelength (λ) of 225 nm. For targeted profiling of lignans, a tandem quadrupole mass analyzer with electrospray ionization (ESI) was used in the positive atomic mass-to-charge ratio (m/z) ion monitoring mode. In order to obtain the greatest possible confidence in the identity of the compounds under study and the greatest possible sensitivity of the determination, the MRM (multiple reaction monitoring) technique was used, which involves selecting a single parent ion characteristic of the substance under study and then monitoring the progeny ions formed after collision with inert gas particles in a collision chamber. Standard lignan substances were purchased from ChemFaces Biochemical Co., Ltd. (Wuhan, China).

Statistical Analysis
Quantitative results are expressed in mg/100 g DW (dry weight) as the mean ± SD (standard deviation) of three replicates (n = 3). The influence of elicitor treatment was evaluated by one-way ANOVA. Differences between means were calculated using Duncan's multiple range test (p < 0.05) using the statistical package STATISTICA 13.0 (Stat-Soft, Inc., Tulsa, OK, USA).

Conclusions
The present study is the first such complex biotechnological study aimed at wide elicitation protocol elaboration using biotic elicitors-chitosan (CH) and yeast extract (YeE), as well as abiotic elicitors-methyl jasmonate (JaMe) and ethephon (ETH), for boosting the production of unique compounds from lignan groups. These metabolites are characteristic of S. rubriflora endemic Chinese species. Through our study we proved, for the first time, possibilities for increasing their production in biomass cultures under in vitro conditions (in independence on environmental factors). Our study also described for the first time the influence of elicitation on lignan compounds' production in the S. rubriflora Platform TIS bioreactors grown microshoot cultures. An important aspect of the research performed was also to compare the biosynthetic capabilities of the F and M lines, and thus the sexes on lignan production.
The results revealed a high competitiveness of S. rubriflora in vitro cultures in relation to soil grown plants. Our results showed new perspectives of potential in vitro cultures utilization as an alternative for a rare, hard-to-find plant material that is difficult to produce on an industrial scale.