Introduction

Rheumatoid arthritis, a chronic joint disease, affects about 1% of the world population (Kumar et al. 2016). The anti-arthritic treatment is frequently based on plant remedies, which are cheaper, less toxic and more available for patients than synthetic pharmaceuticals, but comparably effective (Kumar et al. 2016; Setty and Sigal 2005). One of the groups of phytocompounds modulating the pro-inflammatory signaling pathways are terpenes, the components of essential oils. Their effects on the arachidonic metabolism, cytokines production and pro-inflammatory gene expression were reported (Laev and Salakhutdinov 2015; Miguel 2010).

Rhododendron tomentosum Harmaja (formerly known as Ledum palustre L.), an aromatic swamp plant from the family Ericaceae, has been used for ages in folk medicine to treat joint disorders (Dampc and Luczkiewicz 2013). Modern researches confirm anti-inflammatory, antioxidant and analgesic activity of its volatile fraction, indicating its potential as therapeutic agent against arthritis (Ahmad et al. 2013; Baananou et al. 2015; Zhang et al. 2010). However, in central Europe R. tomentosum is an endangered species because of wetlands’ degradation as well as over-harvesting of above-ground shoots by humans, who traditionally use it as a repellent (Jesionek et al. 2016). A large intraspecific variety of the specimens is an additional problem due to the fact that chemical composition of the essential oil is considerably related to the natural habitat, the phase of the vegetation period and the age of the discussed plant (Dampc and Luczkiewicz 2013). Taking into account the mentioned factors, providing the alternative, constant source of this valuable plant material which produces stable amounts of terpenoids seems to be indispensable to conduct further studies on its medicinal properties.

In our previous work, we established stationary microshoot cultures of R. tomentosum, capable of accumulating essential oil (Jesionek et al. 2016). In contrast to other plant metabolites, isolation of the measurable quantity of the volatile fraction for its analysis requires large amount of plant biomass. This was particularly important in case of the obtained in vitro cultures where the synthesis of essential oil in the microshoots decreased threefold in comparison with aged shoots of the maternal plant (Jesionek et al. 2016). Therefore, the objective of the current study was at first to scale up in vitro biomass production and subsequently to obtain the higher essential oil yield. To reach the goal, R. tomentosum liquid shoot cultures were established to provide the best growth parameters and afterwards, the various bioreactors (continuous immersion, temporary immersion and spraying systems) were applied. As shown by literature data (Grzegorczyk and Wysokinska 2008; Watt 2012), liquid cultures have a better access to nutrients and growth regulators compared with stationary ones, which may result in higher production of primary and secondary metabolites, including essential oils. Moreover, the agitated liquid culture serves as a preparatory step for introducing microshoots to large-scale installations, allowing to investigate the influence of the miscellaneous cultivation factors on biomass accumulation (Grzegorczyk and Wysokinska 2008). According to many reports, the controlled microenvironment, in terms of nutrients and gas transfer in bioreactors, is associated with higher multiplication rate, acceleration of biomass growth and improvement of secondary metabolites productivity (Georgiev et al. 2014; Jang et al. 2016; Paek et al. 2005; Scheidt et al. 2009; Yan et al. 2011; Zobayed et al. 2004). However, it should be noted that so far there are only a few articles concerning essential oil accumulation in bioreactor-grown microshoots (Fulzele et al. 1995; Hilton et al. 1995; Tisserat and Vaughn 2008).

Materials and methods

Reagents and general procedures

All reagents used for plant in vitro culture experiments were from Sigma-Aldrich (St. Louis, US-MO). Water was purified with the Elix/Synergy system (Merck KGaA, Darmstadt, Germany). Unless otherwise stated, in vitro cultures were incubated at 24 ± 2 °C under white fluorescent light (16/24 h photoperiod, 40 μmol m− 2 s− 1, TLD 35 W/33 tubes, Philips, Amsterdam, the Netherlands).

Agitated cultures

Initiation of agitated culture

The plant material obtained at the elongation stage of the micropropagation protocol of R. tomentosum (Jesionek et al. 2016) after 28 day growth on the stationary Schenk-Hildebrandt medium (Schenk and Hildebrandt 1972) modified with the addition of 0.5 g l−1 ammonium nitrate (SHRt) and supplemented with 24.60 μM 2-isopentenyladenine (2iP), was used for the liquid culture initiation. 3 g microshoots were transferred to 250 ml Erlenmeyer flasks closed with silicone foam stoppers (Carl-Roth, Karlsruhe, Germany) which contained 30 ml liquid medium with the same composition, but excluding agar. The shoot culture was agitated using the orbital shaker (120 rpm, 25.4 mm stroke, INNOVA 2300, New Brunswick Scientific, Enfield, CT, USA) and subcultured at 4-week intervals for 3 months prior to further experiments.

Optimization of culture conditions

For optimization of the culture medium composition, 3 g microshoots taken on 28 day of growth cycle from the initial agitated liquid culture were immersed in 250 ml Erlenmeyer flasks in 30 ml SHRt medium supplemented with different plant growth regulators (PGRs) (Table 1a). 2iP, indole-3-butyric acid (IBA), thidiazuron (TDZ) and adenine (AD) were added to the growth medium prior to autoclaving while zeatin (ZEA) and triacontanol (TRIA) were sterile-filtered (syringe filters, pore diameter 0.2 μm, Cronus, Maisemore, UK). Due to the limited solubility in water, TRIA was first dissolved in stock solution containing chloroform and Tween 20 and then diluted with water, according to the procedure described by Tantos et al. (2001). The pH of the studied media was adjusted to 5.6.

Table 1 The effect of various modifications of liquid SHRt medium composition in terms of PGRs (a), the application of BF as well as the change of pH and enrichment with PE (b) on the growth parameters and morphology of R. tomentosum microshoots after 28 days’ cultivation in Erlenmeyer flasks
Full size table

Further modifications of cultivation conditions referred to SHRt medium with 24.60 μM 2iP and 592.02 μM AD (Table 1b). The spectrum of daylight was modified by covering the Erlenmeyer flasks with the foil bluelight filter (BF) (119 dark blue, ca. 3% visible light transmission, Lee Filtres, Andover, UK). Furthermore, the mineral content of SHRt medium was changed by replacement of the pure water with the aqueous peat extract (peat: aqua 1:5), commercial one (PE) (garden peat AB, Hollas, Paslek, Poland) as well as the extract from the peat collected at the maternal plant’s natural habitat in Miszewko near Gdansk in Poland (PENH). One more modification was to adjust the pH of the liquid medium to 5.0 which is natural for PE.

The microshoots from the PGR-free SHRt medium as well as from the initial agitated liquid culture (SHRt medium with 24.60 μM 2iP) were the controls during the PGR optimization experiments while for the culture conditions study the plant material from SHRt medium with 24.60 μM 2iP and 592.02 μM AD served as the control group. Each experiment was conducted with three replicates. On the 28th day of the growth cycle the plant material was harvested and the growth parameters (fresh weight (FW), growth index (Gi), dry weight (DW)) as well as microshoots’ morphology (length, colour, vitality) were estimated (Table 1).

Optimization of the origin and quantity of plant inoculum

To assess the impact of the origin (agar-gelled or agitated cultures) and the quantity of plant inoculum on R. tomentosum biomass growth, 2 and 3 g microshoots from stationary and liquid SHRt media supplemented with 24.60 μM 2iP were immersed in 30 ml liquid SHRt medium with 24.60 μM 2iP and 592.02 μM AD and shaken in Erlenmeyer flasks (Table 2). The inoculum [g]/liquid medium [ml] ratio was 1:15 or 1:10, respectively.

Table 2 The effect of the origin and the quantity of plant inoculum on the growth parameters of R. tomentosum microshoots after 28 days’ cultivation in agitated Erlenmayer flasks, in liquid SHRt medium supplemented with 24.60 μM 2iP and 592.02 μM AD
Full size table

On the 28th day of the growth cycle the plant material was collected and the growth parameters (FW, Gi, DW) were compared (Table 2). Each experiment was conducted with three replicates.

Bioreactor systems

Five experimental in vitro culture systems were compared: (I) continuous immersion system: Magenta™ vessel GA-7 (MV) (Sigma-Aldrich, St. Louis, US-MO) modified with a stainless steel net (1 × 1 mm) at 1 cm above the bottom (Fig. 1c); (II) temporary immersion systems (TIS): (a) RITA® bioreactor (Vitropic, St. Mathieu de Treviers, France) (Georgiev et al. 2014; Watt 2012) (Fig. 1d), (b) PLANTFORM bioreactor (Plant Form AB, Sweden & TC propagation Ltd., Ireland) (Welander et al. 2014) (Fig. 1e), (c) temporary immersion glass bioreactor (TIGB) (Fig. 1f); (III) spray glass bioreactor (SGB) (according to Jaremicz et al. 2014, with modifications) (Fig. 1g). Systems I, IIa i IIb are commercially available while IIc and III are assembled by the authors of the paper (Fig. 2). The TIGB (Fig. 2b, c) and SGB (Fig. 2d) consisted of a cylindrical glass vessel (length 20 cm, diameter 12 cm), equipped with sterile air supply (Fig. 2a) and stainless steel net (1 × 1 mm) for microshoots immobilisation. Additionally, SGB was connected with a peristaltic pump (MasterFlex L/S, Cole-Parmer, Vernon Hills, USA). All bioreactors were steam sterilized (121 °C, 0.1 MPa, 22 min).

Fig. 1
figure 1

Rhododendron tomentosum microshoots in the studied in vitro systems, after 28 days of cultivation: on agar-gelled SHRt medium supplemented with 24.6 μM 2iP: a baby food culture jar (plant inoculum); on liquid SHRt medium supplemented with 24.6 μM 2iP and 592.02 μM AD: b Erlenmeyer flask; c Magenta™ vessel GA-7; d RITA® bioreactor; e PLANTFORM bioreactor; f TIGB; g SGB

Full size image
Fig. 2
figure 2

Schematic diagrams of custom-made bioreactor systems used in the study: a air supply (common for all systems), b TIGB (immersion phase), c TIGB (medium withdrawal phase), d SGB; 1 air pump, 2 air prefilter, 3 flowmeter, 4 bottle washer (air humidifier), 5 inlet PTFE filter (0.22 μm), 6 medium reservoir, 7 growth vessel with stainless steel mesh support, 8 outlet PTFE filter (0.22 μm), 9 polypropylene fiber filter, 10 peristaltic pump, 11 spray nozzle. White and black arrows indicate air and growth medium flow, respectively

Full size image

In all systems, R. tomentosum microshoots grown on stationary SHRt medium with 24.60 μM 2iP for 28 days (Fig. 1a) were inoculated into liquid SHRt medium with 24.60 μM 2iP and 592.02 μM AD at 1:15 (m/v) ratio. All TIS and spray bioreactor were attached to an air humidifier and an air pump (IPX4 ACO-9602, Hailea, China) via a filter disc (diameter 60 mm, pore diameter 0.22 μm, Cole Parmer, US-IL). Additionally, SGB was connected with a peristaltic pump (MasterFlex L/S, Cole-Parmer, Vernon Hills, USA). The microshoots in MV were constantly immersed in the medium while immersion in TIS took place every 85 min for 5 min (air flow 0.5 l min−1). The same cycle was maintained for SGB (medium dispersion rate 100 ml min−1), using the 1.5 size TN-type hydraulic nozzle (Spraying Systems Co, Wheaton, US-IL).

The microshoots from the agitated liquid culture were used as the control (Fig. 1b). In the control group, R. tomentosum microshoots grown on stationary SHRt medium with 24.60 μM 2iP for 28 days (Fig. 1a) were inoculated into Erlenmeyer flask containing liquid SHRt medium with 24.60 μM 2iP and 592.02 μM AD at 1:15 (m/v) ratio.

Each experiment was conducted with three replicates. After 28 days, the growth parameters were evaluated (FW, Gi, DW) (Fig. 3), the content of the essential oil was determined in the dried biomasses (Fig. 4) and the obtained volatile fractions were subjected to GC/MS analysis (Table 3).

Fig. 3
figure 3

Growth parameters of R. tomentosum microshoots, cultivated in the studied bioreactor systems for 28 days. Values are the means of three replicates; data followed by different letters are significantly different at p < 0.05 (FW–DMRT, Gi and DW–Dunn’s test)

Full size image
Fig. 4
figure 4

Essential oil content in R. tomentosum biomasses obtained in the studied bioreactor systems, after 28 days of cultivation. Values are the means of three replicates; data followed by different letters are significantly different at p < 0.05 (DMRT)

Full size image
Table 3 The chemical composition of R. tomentosum essential oils obtained from various in vitro bioreactor systems and the agitated culture after 28 days’ cultivation in liquid SHRt medium supplemented with 24.60 μM 2iP and 592.02 μM AD
Full size table

Biomass growth and essential oil production profile in RITA® bioreactor compared with agitated culture

For R. tomentosum microshoot culture maintained in RITA® bioreactor, the biomass growth and the essential oil production profile were determined in a 63 day experiment. 13.3 g microshoots from the stationary agar culture (SHRt medium with 24.60 μM 2iP, 28 day of cultivation) were transferred to RITA® bioreactor and immersed every 85 min for 5 min in 200 ml liquid SHRt medium supplemented with 24.60 μM 2iP and 592.02 μM AD. The biomass was collected at 7 days intervals after total unloading of the culture vessel to determine the growth parameters (FW, Gi, DW) and to establish the volatile fraction content (starting from 14 day of the growth cycle). The obtained results were compared with the microshoots from the agitated liquid culture where 2 g inoculum from the stationary agar culture (SHRt medium with 24.60 μM 2iP, 28 day of cultivation) were immersed in 30 ml SHRt medium with 24.60 μM 2iP and 592.02 μM AD in Erlenmeyer flasks (Figs. 5, 6). For R. tomentosum agitated culture, the biomass was collected also at 7 days intervals during 63 days, but the essential oil content was determined only in 28 day of cultivation. Each experiment was conducted with three replicates. The GC/MS analysis of volatiles was performed (Fig. 7).

Fig. 5
figure 5

Growth profile of R. tomentosum microshoots, cultivated in RITA® bioreactor, in comparison with the agitated liquid culture, during 63 days of cultivation. Values are the means of three replicates

Full size image
Fig. 6
figure 6

Dry weight values and the essential oil production profile of R. tomentosum microshoots cultivated in RITA® bioreactor (a) in comparison with the agitated liquid culture (b), during 63 days of cultivation. Values are the means of three replicates

Full size image
Fig. 7
figure 7

Content of the selected terpenoid compounds in the essential oils of R. tomentosum microshoots, cultivated in RITA® bioreactor for 63 days. Values are the means of two replicates

Full size image

Determination of growth parameters

Fresh weight (FW) was measured by weighing the harvested plant material after cleaning it from the liquid medium. Gi was calculated with the formula: Gi = (FWx − FW0)/FW0 × 100 where Gi is the growth index and FW0 and FWx is the fresh weight of the inoculum and the fresh weight of the microshoots after x days of cultivation (usually 28 days unless otherwise stated), respectively. The obtained biomasses were dried for 24 h at 30 °C in the drying chamber with forced convection (FD 115, Binder, Tuttingen, Germany) to determine dry weight (DW).

Determination of essential oil content

For the determination of volatile fraction content, the dried biomasses obtained in the examined bioreactors and in RITA® system during the essential oil production profile study, as well as the microshoots from agitated liquid culture grown for 28 days in the optimized SHRt medium with 24.60 μM 2iP and 592.02 μM AD were subjected to hydrodistillation (5 g, 100 ml of distilled water, 3 h) in the Deryng apparatus (Polish Pharmacopoeia 2002). The isolated volatile fractions were collected in 1.0 ml of hexane:diethyl ether mixture (1:1 v/v) (Butkienė et al. 2008) which was afterwards evaporated from the samples under a stream of nitrogen gas. The essential oil content in the studied plant materials was presented as a mean value from the results of three hydrodistillations (Figs. 4, 6).

GC/MS analysis of essential oils

For GC/MS analysis, Gas Chromatograph 7890A coupled with Mass Selective Detector 5977A (Agilent Technologies, Santa Clara, US-CA) was used. The essential oils obtained from R. tomentosum in vitro cultures were diluted with ethyl acetate (1:80) and 1.0 μl of the prepared volatile fractions solutions was introduced on the DB-5ms 30 m × 0.25 mm × 0.25 µm capillary column (Agilent J&W) with Agilent Autosampler Systems 7693 (a split ratio 1:10, the injector temperature 250 °C). The carrier gas was helium and the constant flow rate was applied (1.1 ml min−1). The oven temperature increased from 50 to 280 °C, at a rate 7 °C min−1. The final GC temperature was held for 20 min. The single GC run time was 53 min. The GC/MS analysis results were compared with the data from NIST Library 11.0. and are presented in Table 3 and Fig. 7 as a mean values from two replicates of the experiment.

Statistical analysis

The collected data were subjected to one-way analysis of variance (ANOVA) followed by Duncan’s Multiple Range Test (DMRT, p < 0.05) or Dunn’s Test (p < 0.05), when equal variance test failed. The analyses were conducted with SigmaPlot 11.00 (Systat Software, San Jose, US-CA).

Results and discussion

Agitated cultures

In this study the R. tomentosum agitated liquid cultures were established from microshoot clusters growing on stationary SHRt medium with 24.60 μM 2iP by their immersion in the medium with the same composition, excluding agar. As shown by previous research (Jesionek et al. 2016), the SHRt medium enriched with 24.60 μM 2iP gave the best results in multiplication and elongation of R. tomentosum microshoots, providing their high vitality. However, the growth parameters obtained in the initial agitated liquid culture were too low to receive sufficiently high amount of plant material for essential oil extraction (Table 1). Therefore, attempts were undertaken to stimulate growth of the studied biomass by addition of other PGRs to the medium, simultaneously maintaining the content of 2iP as the basic cytokinin responsible for the proper microshoots’ morphology.

The PGRs chosen for the experiment included 2iP at doubled concentration, AD, ZEA, TDZ, TRIA and IBA (Table 1a). 2iP is widely applied to control the growth of Rhododendron sp. (Eeckhaut et al. 2010; Mao et al. 2011; Tomsone and Gertnere 2003) whereas AD is the constituent of Anderson’s medium which is commonly used in the micropropagation of the mentioned genus (Anderson 1978). According to the literature data, addition of ZEA as PGR usually intensifies shoots proliferation in Ericaceae family (Almeida et al. 2005; Eeckhaut et al. 2010). Small amounts of TDZ and TRIA are known to increase multiplication rate, and thus are frequently used in woody plant cultures (Guo et al. 2011; Huetteman and Preece 1993; Tomsone and Gertnere 2003; Tantos et al. 2001). Auxins, like IBA, are usually applied in combination with cytokinins to balance their effects and to elongate microshoots (Almeida et al. 2005; Mao et al. 2011; Tomsone and Gertnere 2003; Vejsadova 2008).

As presented in Table 1a, the enrichment of SHRt medium with the mixture of 2iP and above-mentioned PGRs improves growth parameters of the R. tomentosum liquid shoot culture. The highest values of FW, Gi and DW were determined for AD, ZEA and TDZ supplementation. However, TDZ caused undesirable adverse changes in the microshoots’ morphology. Taking into account the comparable effects of AD and ZEA, the first one seems to be more suitable for large-scale cultivation systems due to its much lower price and the lack of callus-inducing effect. Surprisingly, the addition of IBA did not influence the length of microshoots. It is also worth noting that doubling 2iP concentration from 24.60 to 49.20 μM did not correlate with the biomass accumulation. It is in accordance with the research of Mao et al. (2011) which indicated that 2iP increased the number of Rhododendron sp. shoots only at concentrations up to 40 μM, whereas concentrations over 60 μM inhibited shoot formation.

After the selection of the most beneficial PGR composition for the R. tomentosum liquid culture, further modifications in the cultivation conditions were made (Table 1b). Based on the reports that blue light promotes microshoots formation (Rout et al. 2000), the daylight spectrum was changed with the foil bluelight filter, which decreased the light intensity and its range (Table 1b). Although in natural environment R. tomentosum prefers shaded forest or wetland habitats (Sellmer et al. 2003), in the liquid culture under low-intensity light the significant reduction in the biomass accumulation was observed.

Furthermore, it is well known that the establishment of R. tomentosum field crops is limited by the plant’s specific requirements in terms of such soil factors as good drainage, moisture, high organic matter, low pH and sufficient nutrient availability (Sellmer et al. 2003; Shaver 1983). Moreover, the importance of the Ericaceae symbiosis with ericoid mycorrhizal fungi in utilizing mineral compounds as nitrogen and phosphorus is commonly emphasized (Perotto et al. 2002). Therefore, the effect of the supplementation of SHRt medium with undefined nutrients derived from the aqueous peat extracts was examined, using the horticultural soil (PE) as well as the peat collected from the natural habitat of maternal plant (PENH), which was additionally regarded as a source of microorganisms and the products of their metabolism, characteristic of the area. The pH of the medium was also decreased to provide more acidic environment, similar to natural conditions (pH 5.0) (Table 1b). The enrichment of the liquid medium with the peat extract improved the Gi parameter by about 7%, regardless of its origin. However, though noteworthy, this result was not statistically significant. For that reason as well as because of the expected difficulties in maintaining constant composition of the peat extract during all stages of the research, the SHRt medium with 24.60 μM 2iP and 592.02 μM AD without any alterations to its mineral composition was selected to further experiments. As demonstrated, the change of pH did not influence the growth parameters of in vitro culture (Table 1b).

Finally, the optimal inoculation procedure (with respect to biomass origin and shoot/medium ratio) was specified (Table 2). Although initial immersion of 3 g microshoots in 30 ml liquid SHRt medium with 24.60 μM 2iP and 592.02 μM AD (1:10) resulted in the similar growth parameters regardless of the biomass origin (agar or liquid shoot culture), reducing the amount of inoculum to 2 g (1:15) led to a different situation. Microshoots taken from the liquid culture showed strong tendency to drown under the medium surface after subculturing, probably as a consequence of the hindered stem verticalization under mechanical stress. In turn, along with lower amount of inocular biomass from stationary culture, Gi of the obtained plant material increased almost twice (from 163.41 to 323.64). The ratio of mass to volume 1:15, using the plant inoculum from agar-gelled culture, was thus selected to further experiments.

Bioreactor systems

For scaling up the R. tomentosum shoot culture, the various bioreactor types were used: MV as an continuous immersion system, RITA®, PLANTFORM and TIGB as temporary immersion systems (TIS) and SGB as a gas-phase system (Fig. 1). The largest accumulation of biomass was achieved in RITA® and SGB (FW = 250 and 240 g l−1, Gi = 280 and 250, respectively), probably due to the high moisture content inside the culture vessel (Fig. 3). However, the obtained FW and Gi values were slightly lower than in the microshoots shaken in the Erlenmayer flasks. It can be explained by the tissue hyperhydricity in the liquid agitated culture. This assumption can be confirmed by the DW values obtained in RITA® and SGB bioreactor systems (DW = ca. 20 g l−1) which were the highest from the studied biomasses, even in comparison with the shaken microshoots, presumably thanks to the less hydrated cells (Fig. 3). From the two systems mentioned above, application of RITA® seems to be more profitable for conducting the large-scale studies because it is relatively cheap, light and easy to handle in contrast to expensive, heavy and prone to breakage bioreactors made of glass. Moreover, RITA® containers can be connected in series for scaling up purposes and occupy little space in a growth room.

As many authors demonstrated, TIS, in which microshoots are covered by the capillary medium film after periodic flooding, have positive effect on the proliferation rate, providing the most natural environment, low mechanical stress, improved gas exchange and nutrients transfer (Alvard et al. 1993; Georgiev et al. 2014; McAlister et al. 2005; Zobayed et al. 2004). Notwithstanding, two of the studied TIS had some disadvantages which presumably contributed to their less effectiveness compared to RITA®: largely unused headspace volume despite higher medium consumption per vessel in PLANTFORM bioreactor and the separation of the medium reservoir from the culture vessel in TIGB, which hindered its handling and probably decreased to some extent the level of moisture inside. This was not in the case SGB, where the medium remained on the bottom of the culture vessel and was temporarily sprayed as a mist onto the biomass surface.

MV, in which microshoots were permanently submerged in the liquid medium without forced air supply and lack of mechanical stress, was characterized with the lowest growth parameters, probably due to asphyxia which outweighed the benefits of better tissue-medium contact (as compared to agar culture) (Watad et al. 1996).

The essential oil content in biomasses obtained in RITA® and SGB was clearly higher than in other bioreactor systems and the control groups, reaching 0.5% (v/m) (Fig. 4). This amount was comparable with the volatile fraction content in the aged shoots of maternal plant (Jesionek et al. 2016). It indicates that R. tomentosum has great potential to be used in large-scale installations for production of terpenoid compounds. There are few reports which refer to successful accumulation of secondary metabolites in shoot cultures cultivated in various types of bioreactors: dibenzocyclooctadiene lignans (Szopa et al. 2017), cardiotonic glycosides (Perez-Alonso et al. 2009), flavonoids (wogonin, baicalein, baicalin) (Zobayed et al. 2004), polyphenols (Jain et al. 2012), xanthones and benzophenone derivatives (Kokotkiewicz et al. 2015), indole alkaloids (vindoline, catharanthine) (Yingjin and Zongding 1994), isoquinoline alkaloids (galanthamine) (Georgiev et al. 2012), artemisinin (Liu et al. 2003), phenolics and flavonoids (Jang et al. 2016). However, only few semi-technical sources of volatile oils based on bioreactor-grown microshoots have so far been reported. These include in vitro shoot cultures of Artemisia annua maintained in a bubble column bioreactor, Mentha spicata microshoots grown in TIS, as well as transformed shoots of M. piperita and M. citrata in a mist bioreactor (Fulzele et al. 1995; Hilton et al. 1995; Tisserat and Vaughn 2008).

The volatile fractions obtained from the bioreactor-cultured R. tomentosum microshoots were characterized with the unique chemical composition (Table 3). About 70% of total compounds were identified. Methyl everninate and ledene oxide (II), revealed in a large amount (2.8–7.0% and 9.0–14.7%, respectively) in the studied in vitro biomasses, were absent in the maternal plant according to previous research (Jesionek et al. 2016). On the other hand, major volatile constituents of the intact plant were either not found (ledol) or confirmed only in trace quantities (palustrol) in the microshoots. It can be assumed that some metabolic pathways of the aromadendrane-related sesquiterpenes were not fully developed at this stage of the biomass cultivation under in vitro conditions. Additionally, it was shown that the essential oils from TIS-cultivated microshoots were characterized with the similar qualitative and quantitative terpenoids content which did not differ significantly from the volatile fraction isolated from the agitated liquid culture. Oxygenated sesquiterpenes (23.1–25.1%) and sesquiterpenes with aromadendrane skeleton (13.9–19.2%) constituted the majority of total compounds. The most abundant terpenes in TIS (>5%) were p-cymene, alloaromadendrene, shyobunone and ledene oxide (II). The volatile fractions obtained from immersion and spray systems included two times less compounds (22 and 23, respectively) than TIS (44). Monoterpene hydrocarbons predominated (31.9%) in the SGB-derived essential oil which contained substantial amounts of p-cymene (17.9%) and sabinene (5.1%). However, the alloaromadendrene level significantly decreased in SGB-grown shoots. The highest percentage of alloaromadendrene (8.1%), shyobunone (15.7%) and δ-2-carene (4.5%) were observed in MV.

Given the obtained results, RITA® –TIS which stimulated to the largest extent the R. tomentosum microshoots growth and production of essential oil rich in potentially bioactive terpenes—was selected for further experiments.

Biomass growth and essential oil production profile

The biomass growth and the essential oil production profiles in RITA® bioreactor were determined during the 2-month cultivation, in parallel to the agitated liquid culture (Figs. 5, 6). The growth profiles of both cultures were similar and were characterized with the presence of the particular development phases. The phase of logarithmic and linear growth lasted until 28 day of cultivation, changing then into plateau phase. Although the growth rate of the agitated liquid culture was faster and both FW and Gi values achieved after 4 weeks were higher than in RITA®, the largest biomass accumulation defined by DW was obtained in TIS (ca. 20 g l−1). This difference was caused probably by the excessive hydration of the shaken microshoots. The above could also contribute to the more visible decay phase in the case of the agitated liquid culture (Fig. 6a), while the clear decrease in DW of in vitro cultures in RITA® was noticed only after 2 months and FW and Gi were still high (Fig. 6b).

The content of the essential oil in RITA® microshoots rised along with the biomass growth till 28 day of cultivation when it reached 500 µl 100 g−1 DW, exceeding five times the volatile fraction level in the agitated culture (Fig. 6). Although after 2 months of cultivation DW started to decline, the production of terpenoid compounds was still stable. This phenomenon is beneficial from the technological perspective, enabling the use of the proposed system as an alternative source of R. tomentosum essential oil for biological activity studies. However, it should be taken into account that chemical composition of the obtained volatile fraction may be subject to change over a period of days.

The percentage content of the most abundant components in the essential oil from RITA® bioreactor were determined every 7 days during 2 months, starting from 14 day of cultivation (Fig. 7). Some of the selected compounds were reported to possess bioactive properties: anti-inflammatory (p-cymene: Bonjardim et al. 2012; sabinene: Valente et al. 2013), nociceptive (p-cymene: Bonjardim et al. 2012), cytotoxic (γ-terpineol: Wu et al. 2014; alloaromadendrene; Mah et al. 2013) and insecticidal (shyobunone: Chen et al. 2015). Although shyobunone and alloaromadendrene content despite small fluctuations remained at a fairly constant level during the entire biomass cultivation period, the concentrations of other compounds varied depending on the stage of experiment. The amount of ledene oxide (II) declined after 42 day of the growth cycle in parallel with γ-terpineol and sabinene increase. The results of our previous research suggested that γ-terpineol content can be correlated with the age of plant material (Jesionek et al. 2016). The highest amount of methyl everninate was observed at the beginning and at the end of cultivation period which indicates that this compound might be produced in response to stress associated with initiation and dying of the in vitro culture. The biosynthesis of p-cymene was shown to be intensified only after 21 day of microshoots cultivation.

Due to the changes in the concentration of the volatile compounds during cultivation, it is recommended to select the proper biomass harvesting time to obtain the optimal content of the terpenoids in the R. tomentosum in vitro essential oil.

Conclusions

In conclusion, liquid shoot cultures of R. tomentosum were established for the first time, providing the alternative, renewable and continuous source of in vitro plant material for essential oil isolation. Scaling up to bioreactors improved the growth parameters and stimulated the essential oil production. It was proven that the type of in vitro cultivation system influences the primary and secondary metabolism of the studied biomass. The best effects were observed with application of RITA® bioreactor. Since the amounts of volatiles found in bioreactor-grown microshoots were comparable with maternal plant, it is possible that the developed system will enable the large-scale production of R. tomentosum essential oil rich in bioactive compounds.