Biotechnological approaches to enhance salidroside, rosin and its derivatives production in selected Rhodiola spp. in vitro cultures

Rhodiola (Crassulaceae) an arctic-alpine plant, is extensively used in traditional folk medicine in Asian and European countries. A number of investigations have demonstrated that Rhodiola preparations exhibit adaptogenic, neuroprotective, anti-tumour, cardioprotective, and anti-depressant effects. The main compounds responsible for these activities are believed to be salidroside, rosin and its derivatives which became the target of biotechnological investigations. This review summarizes the results of the diverse biotechnological approaches undertaken to enhance the production of salidroside, rosin and its derivatives in callus, cell suspension and organ in vitro cultures of selected Rhodiola species.


Introduction
Rhodiola spp., herbaceous perennial plants of the Crassulaceae family, are extensively used in traditional medicine in Asian and European countries (Tolonen et al. 2003;Yousef et al. 2006;Galambosi et al. 2010;Panossian et al. 2010). The Rhodiola species grows in regions of cool temperature, in the sub-arctic areas of the northern hemisphere, including North and Central Europe, Asia and North America (GBIF 2010 http://www.gbif.org ;Galambosi 2006;Guest and Allen 2014). The genus Rhodiola is believed to originate from the mountainous regions of Southwest China and the Himalayas and nowadays is distributed in mountainous as well as coastal habitats (Brown et al. 2002).
The current taxonomic status of the genus Rhodiola is quite complex due to the generally similar morphology (Brown et al. 2002;Liu et al. 2013). According to GBIF (2010) the genus Rhodiola comprises of 136 accepted species while the Plant List (http://www.theplantlist.org) includes 135 scientific plant names of species rank for the genus Rhodiola. Of these 61 are accepted species names, 16 have not been clarified.
The morphologicals of Rhodiola plants are as follows: stems dimorphic with usually very stout caudex or rhizome, usually with brown or blackish, membranous, scalelike leaves, sharply differentiated from much more slender, erect or ascending, leafy flowering stems ). The plant is dioecious with male and female flowers located on different plants, and rarely hermaphroditic (Tutin 1964). Among the distinguishing characters of the genus are two series of stamens totaling twice the number of petals; free or nearly free petals (not joined in a tube); a stout rhizome from whose axils the flowering stems rise (Flora of China 2001).
Recently, genotyping and chemotaxonomic markers have been used to identify Rhodiola species within the genus. On the basis of the chemical profiles of the 47 collected Rhodiola samples of R. crenulata, R. sachalinensis, R. himalensis, R. serrata, R. rosea, R. kirilowii and R. fastigiata it was demonstrated that different kinds of characteristics reference markers occurred at various concentrations in the different Rhodiola species. The eight compounds: rosarin, rosavin, and rosin, tyrosol and salidroside, catechin, rhodionin and gallic acid have been proposed as reference chemotaxonomic markers. Salidroside and gallic acid were found in all species while rosarin and rosin were detected in R. sachalinnsis, R. himalensis, R. rosea. Rosavin was characteristic only for R. himalensis, R. serrata, R. rosea ).
The roots and rhizomes of Rhodiola spp. have been reported to contain distinct groups of chemical compounds (Table 1). Initially, in the 1970s, the compound responsible for its unique pharmacological properties was believed to be salidroside (Saratikov and Krasnov 1987). According to the Russian Pharmacopeia (1989), the raw material of R. rosea should contain 0.8 % salidroside (Furmanowa et al. 1999). However, further studies revealed that not only salidroside but also rosin derivatives are important bioactive compounds (Sokolov et al. 1985(Sokolov et al. , 1990Wagner et al. 1994;Zapesochnaya et al. 1995). Wiedenfeld et al. (2007a) summarized the results of comparative studies on the activities of salidroside and rosin derivatives by Sokolov et al. (1985Sokolov et al. ( , 1990 which showed CNS, adaptogenic and immunostimulating properties, however he concluded that total Rhodiola rosea extracts are superior to the single components which indicates that the glycosides mentioned are not the only compounds responsible for the medicinal effect. In the subsequent studies it was reported that aqueous and hydroalcoholic extracts of R. rosea exhibited stronger neovascular reaction than rosavin applied alone (Skopińska-Ró _ zewska et al. 2008b). Rhodiola rosea extracts used in most pharmacological studies were standardized to a minimum 3 % of rosin and its derivatives and 0.8-1 % salidroside because the naturally occurring ratio of these compounds in R. rosea root is approximately 3:1 (Brown et al. 2002).
Phytochemical investigations of Rhodiola plants pointed out that the content of salidroside, rosin and its derivatives depends on the morphological part of the plant-higher in rhizomes than in roots Węglarz et al. 2008;Tasheva and Kosturkova 2010). It also depends on the age and sex of the plantrhizomes originated from male plants accumulated higher amounts of salidroside (Revina et al. 1976;Platikanov and Evstatieva 2008;Węglarz et al. 2008;Kucharski et al. 2011). Salidroside content in plants is also connected with the place and time of harvesting (Revina et al. 1976;Kurkin et al. 1985;Bykov et al. 1999;Buchwald et al. 2006) as well as place of plant origin (Altantsetseg et al. 2007;Węglarz et al. 2008). To date, the results from the papers concerning chemical profiling of Rhodiola plants demonstrated the higher salidroside and rosin and its derivatives content in plants collected from natural sources than derived from field cultivation ( Table 2).
Research studies on Rhodiola species have been stimulated by their documented pharmacological properties. Recently a sharp increase in the demand for Rhodiola-based products has been observed worldwide which is connected with the necessity of a stable raw material supply and-at the same timeextensive collecting from natural sources. Rhodiola rosea is the most exploited species for commercial purposes. Due to the intensive collection, natural populations of R. rosea are highly threatened and Salidroside and tyrosol, 2-(4-hydroxyphenyl)-ethyl-Ob-D-glucopyranosyl-6-O-b-D-glucopyranoside, phydroxyphenacyl-b-D-glucopyranoside, icariside D2, rutin, picein, lotaustralin, rodiocyanoside A, crenulatin, rhodionin, rhodiosin, daucosterol, bsitosterol, hydroxycinnamic, gallic acid, creosides I, II, III, IV, and V Wang and Wang (1992), Du and Xie (1995), Peng et al. (1995), Su et al. (2007), Nakamura et al. (2008),  R. kirilowii (in total over 49 compounds) Salidroside and tyrosol, rosin, rosavin, rosarin, cinnamyl alcohol, herbacytrin, umbeliferon, esculetin, luteolin, tricetin, epigallocatechin, epigallocatechin gallate, lotaustralin, rodiocyanoside A, tannins, daucosterol and b-sitosterol, hydroxycinnamic, gallic acid, chlorogenic acid Kang et al. (1992), Krasnov et al. (1978), R. quadrifida Salidroside and tyrosol, rosin, rosavin, rosarin, cinnamyl alcohol, rhodiooctanoside, rhodiolin, mongrhoside, rhodiocyanosides A and B, rhodioflavonoside, rhodiooctanoside, tricetin, included in the list of endangered plant species in many countries (Russia, Great Britain, the Czech Republic, Bosnia and Herzegovina, vulnerable in Slovakia; in Bulgaria collection is strictly forbidden; in Poland R. rosea is only found in National Parks) (Galambosi 2006;Kołodziej and Sugier 2012). Sufficient yields of field-cultivated R. rosea roots/rhizomes could be obtained within 5-7 years of cultivation, however the costs are high (Galambosi 2006;Peschel et al. 2013). Another approach to provide a stable supply of the desired compound is its chemical synthesis. The chemical synthesis of rosavin, the accepted marker for genetically pure R. rosea and its extracts (Brown et al. 2002), was performed by Patov et al. 2006. In the case of salidroside the procedure of it synthesis in a multikilogram scale with 72 % overall yield and [98 % purity has been developed recently by Shi et al. (2011). Nevertheless chemical synthesis seems not to be the solution in the case of Rhodiola preparations as it is obvious on the basis of pharmacological studies that salidroside and rosavins are not the only biologically active ingredients responsible for its efficacy and health benefits (Brown et al. 2002;Nan et al. 2003;Zuo et al. 2007;Panossian et al. 2010;Gupta et al. 2010;Choe et al. 2012;Wang et al. 2012;. Biotechnological methods offer the possibilities to produce useful compounds under controlled conditions independent of environmental changes and elaboration of scale-up procedures would reduce labour costs and improve productivity of secondary metabolites. Moreover, the cells and organs involved directly with the biosynthesis of compounds of interests could be cultivated and manipulated to enhance their production. Numerous studies have revealed that the addition of an elicitor (microbial, physical, or chemical factors) or/and a precursor to the culture media significantly Rosarin-0.08-1.1 0.14-1.6 Rosin-0.36-0.53 Rosavin-0. 9-5 rosarin-0.16-0.3 Kurkin et al. (1985, Galambosi (2006) DW dry weight increases the production of secondary metabolites (Kolewe et al. 2008;Shilpa et al. 2010;Vijaya et al. 2010). This review discusses the ways to enhance salidroside and rosin and its derivatives production in various types of in vitro cultures of Rhodiola spp. To date, to the in vitro cultures have been introduced mainly for R. rosea and selected Asian species, such as R. crenulata, R. kirilowii, R. quadrifida and R. sachalinensis. Table 3 presents the distribution of the species.

Biosynthetic pathway of salidroside, rosin and its derivatives
The biosynthetic pathway of the salidroside precursor L-tyrosol and its regulation are not completely understood. There are two different views in the literature concerning the production of salidroside. Both are connected with the shikimate pathway producing L-phenylalanine or L-tyrosine (Yao et al. 1995). The first is that tyrosol is presumably produced by a decarboxylase from p-coumaric acid, which is mainly derived from L-phenylalanine (Xu and Su 1997). The second view states that the precursor of tyrosol may be tyramine, which is synthesized from tyrosine   (Fig. 1).
In order to elucidate the molecular pathways involved in tyrosol biosynthesis Ma et al. (2008) analyzed the effect of overexpressing the endogenous Galambosi (2006) PALrs1 gene in R. sachalinensis on the production of higher levels of p-coumaric acid and salidroside. The obtained results showed that overexpression of the PALrs1 gene resulted in a 3.3-fold increase of p-coumaric acid. However, the levels of tyrosine, tyrosol and salidroside were 2.6-fold, 4.7-fold and 7.7fold lower, respectively, in PALrs1 transgenic plants than in controls. Overexpression of the PALrs1 gene and accumulation of p-coumaric acid did not facilitate tyrosol biosynthesis. Thus, tyrosol probably is not derived from L-phenylalanine. Following the second view of salidroside biosynthesis Gyorgy et al. (2009) analyzed the expression of the R. rosea TyrDC gene encoding tyrosine decarboxylase. TyrDC expression was detected in leaves as well as in roots. The expression was significantly higher in roots, which coincides with the fact that salidroside accumulates preferentially in the underground parts of the plant. The expression was also higher in the line accumulating high levels of salidroside, compared to the line with lower salidroside content. Similar results were obtained by Zhang et al. (2011). In this study, a cDNA encoding TyrDC was isolated from R. sachalinensis and the effects of recombinant RsTyrDC and sense and antisense overexpression of endogenous RsTyrDC on tyrosol synthesis were investigated. For sense-RsTyrDC expression, RNA gel blotting indicated that RsTyrDC was over-expressed at transcriptional levels and TyrDC activity in the RsTyrDC transgenic lines was 74 and 127 % higher, respectively, than in the empty vector-transformed control plants. This overexpression resulted in the accumulation of tyrosol and salidroside. Simultaneously there was evidence of reduced accumulation of the endogenous TyrDC transcripts of R. sachalinensis in the antisense plant. Summarizing these research results, it is tempting to speculate that the biosynthesis of salidroside begins with the decarboxylation of tyrosine by TyrDC, which produces tyramine.
The production of rosin and its derivatives in Rhodiola is restricted to only a few species. Biosynthesis of phenolic glycosides occurs spontaneously in Rhodiola roots and rhizomes (Krajewska-Patan et al. 2013). According to György (2006) the cinnamyl alcohol glycosides are products of phenylpropanoid metabolism, derived from L-phenylalanine. Phenylalanine ammonia lyase (PAL) converts L-phenylalanine to cinnamic acid. From cinnamic acid cinnamyl-CoA ester is formed through hydroxycinnamate: CoA ligase (4CL). This CoA ester is reduced to cinnamaldehyde by cinnamyl-CoA reductase (CCR). The cinnamaldehyde is further reduced by cinnamyl alcohol dehydrogenase (CAD) to cinnamyl alcohol (Fig. 2). The enzymes that take part in the formation of the glycosides of cinnamyl alcohol are not yet described. By one glucose transfer rosin is formed, which is the simplest cinnamyl alcohol glycoside. From rosin rosavin is formed by adding an arabinopyranose unit and rosarin by adding an arabinofuranose unit. Further glycosides may be formed depending on the sugar type and the connection site.
The enhancement of salidroside, rosin and its derivatives production in Rhodiola spp. in vitro cultures Biotechnology offers an opportunity to exploit cells, tissues, organs or entire plants by growing them in vitro and to genetically manipulate them to get the desired compounds (Rao and Ravishankar 2002). Till now the biotechnological production of salidroside using plant cell cultures derived from R. crenulata, R. rosea, R. kirilowii, R. quadrifida and R. sachalinensis have been investigated. The influence of external factors such as nutrient levels, stress factors, light and growth regulators on salidroside and rosavins accumulation is summarized in Table 4.

The effect of growth regulators
Enhanced salidroside accumulation in R. sachalinensis compact callus aggregate culture (CCA) was observed by Xu et al. (1998c) compared to dispersed cells. The CCA were maintained on MS medium supplemented with 3 % sucrose, 0.3 mg/l NAA, and 3 mg/l BA. The salidroside yield in CCA after 24 days of culture was 6.8 mg/g DW and, approximately six times as much as that of the dispersed cells; the growth rate of CCA was comparable to that of the dispersed cells. Furthermore, the suitability of airlift reactors for large-scale culture of CCA was investigated in a 5-l airlift reactor. The salidroside amount was similar to that of shake-flask cultures though the biomass was relatively low. Xu et al. (1999) estabished liquid cultures of compact callus aggregates (CCA) of R. sachalinensis derived from stems, and found that a higher cytokinin/auxin ratio was beneficial for both CCA growth and salidroside accumulation. The maximum salidroside yield of 4.7 mg/g DW was achieved when MS medium was supplemented with 0.2 mg/l NAA and 3.3 mg/l BA. Also a favourable effect on the metabolite productivity of raising the initial sucrose levels has been observed. The authors indicated that for growth and salidroside accumulation (up to 7.9 mg/g DW) the CCA culture should be cultivated in MS medium containing 40 g/l of sucrose, 0.2 mg/l NAA and 3.3 mg/l BA. A compact callus aggregate culture (initiated from callus established from cotyledons) of R. sachalinensis reached a high yield of salidroside under appropriate culture conditions. A significant increase in salidroside content (up to 10.78 mg/g DW) was obtained when the inoculum amount was 10 % and the concentration of BA and IBA added to the liquid medium was 5 and 2.5 mg/l, respectively. The addition of 2,4-D at a concentration of 2.0 mg/l boosted the salidroside accumulation in the treated cells to 41.94 mg/g DW while the addition of kinetin inhibited both salidroside accumulation and growth of the CCA suspension culture (Wu et al. 2003).
Callus tissue of R. crenulata was initiated from three types of explants: leaf, stem and root, that were maintained on half strength MS medium supplemented with 3.0 mg/l NAA and 0.5 mg/l BA. Callus initiation was noticeably faster on leafy explants (within 9 days from the beginning of culture) compared to stems and roots (13 and 15 days, respectively). For callus subcultures and subsequent cell suspension cultures full strength MS containing 0.5 mg/l TDZ (thidiazuron) and 0.5 mg/l NAA proved to be the best. The average salidroside content in cell suspension culture reached 26.48 mg/g DW (Shi et al. 2013).
A callus culture of R. kirilowii obtained by Krajewska-Patan et al. (2009) also produced salidroside. Two lines of callus tissue were initiated. The first started from cotyledons on MS medium supplemented with BAP 2.0 mg/l, adenine chloride 1.0 mg/l and NAA 2.0 mg/l. The second started from hypocotyles The results of phytochemical analyses revealed that callus tissues cultivated on solid medium produced the complete spectrum of active compounds characteristic for the roots of the plant. The highest salidroside content of 0.166 mg/g DW was observed in the line obtained from cotyledones. An investigation concerning the effect of plant growth regulators on the induction and proliferation of callus of Rhodiola quadrifida was conducted on MS medium. The medium containing 1 mg/l 2,4-D, 2 mg/l NAA, 0.5 mg/l BAP and 0.1 mg/l kinetin proved to be the best for induction of the callus from R. quadrifida (the induction rate was 83.3 %). The optimized combination of plant growth regulators for callus subculture was 1 mg/l 2,4-D, 0.1 mg/l BAP and 0.5 mg/l kinetin. When the callus was cultured in the optimum medium for 30 days the salidroside content reached 2.8 mg/g DW (Sheng et al. 2005). Furmanowa et al. (1995) reported the micropropagation of R. rosea by apical and axillary meristems. Out of many combinations of growth regulators added to different basal media the best for shoot development were 0.1 mg/l kinetin and 0.1 mg/l IAA added to NN medium (Nitsch and Nitsch 1969). The shoots were rooted on the same medium. In the shoots growing in vitro traces of salidroside, rosin and its derivatives were found. In further studies Furmanowa et al. (1998) described the condition for induction and maintenance of R. rosea callus cultures. Green callus tissue was formed from leaf segments on MS medium containing 2 mg/l BAP, 2 mg/l NAA, 1 g/l casamino acids and 6 % sucrose. Following four passages, callus was subcultured on new media with various combinations of auxins and cytokinins to find the best growth promoting conditions. Two lines were obtained: deep green and yellow. The phytochemical analyses revealed that only one green line produced salidroside,  In vitro cultures for micropropagation and the reintroduction of R rosea were established by Tasheva and Kosturkova (2010). The authors reported that the best shoot induction was obtained from seedling explants on media containing 2 mg/l zeatin or BA each. Efficient rooting was induced on half-strength MS containing 2 mg/l IBA and stimulated by adding 0.2 mg/l IAA. Regenerants rooted in perlite, peat, and soil (1:1:2), adapted in a greenhouse, and transplanted in the mountains survived (70 %) and developed similarly to the wild plants. The salidroside content of these plants after one or 2 years reached 6.4 and 6.1 mg/g DW in rhizomes and 6.2 and 5.3 mg/g DW in roots.

Precursor feeding and elicitation
Plant cells are structures accumulating many chemical compounds that take part in biochemical pathways of growth, defense, and survival. However, plant secondary metabolites often occur in plants in very small amounts, less than 1 % of dry weight (Oksman-Caldentey and Inze 2004). Hence, biotechnology offers an opportunity to exploit the cells, tissues, organs or entire organisms by growing them in vitro and to genetically manipulate them to obtain the desired compounds (Ramachandra Rao and Ravishankar 2002).
Elicitation and precursor feeding in Rhodiola spp. were reported by Xu et al. (1998a), using R. sachalinensis cell suspension cultures. Among the examined elicitors, one extracted from Aspergillus niger was the most effective for salidroside production which reached up to 9.9 mg/g DW. In comparison untreated cultures produced only 1.7 mg/g DW of salidroside. The applied precursor feeding strategy revealed that a significantly high level of salidroside 14.4 mg/g DW was obtained when tyrosol was added to the medium. The combined application of both precursor and elicitor feeding resulted in a markedly high salidroside content of 17.3 mg/g DW (Xu et al. 1998a). Table 5 summarizes studies on elicitors and precursors applications for the production of phenolic glycosides in different Rhodiola spp.
In further studies, Xu et al. (1998b) showed that the biotransformation reaction depends not only on precursor addition, but also on the enzyme activities present in the cell suspension culture of R. sachalinensis. The relevant enzymes are classified as glucosyltransferases. The authors presented the effect of the tyrosol-glucosyltransferase (TGase) activity on tyrosol conversion to salidroside. The studies revealed that the low yield of salidroside in the cell culture of R. sachalinensis was connected with the low efficiency of glucosylation and non-synchronization between the TGase activity and tyrosol accumulation in cells. The highest TGase activity was found to be in the exponential growth phase on day 8. By repeated addition 3 mM tyrosol at 24 h intervals, starting on the day of the highest TGase activity, a significant salidroside yield of 154.5 mg/g DW was obtained (Xu et al. 1998b). To date, that investigation has produced the highest intracellular content of salidroside obtained through biotransformation by R. sachalinensis in vitro cultures.
The low efficiency of glucosylation and the lack of synchronization between TGase activity with tyrosol appearance are believed to be the main cause for low salidroside yield in cell suspension cultures of R. sachalinensis (Xu et al. 1998b). These findings prompted Ma et al. (2007) to research on glucosyltransferases on the molecular level. The putative UDP-glucose:tyrosol glucosyltransferase (UGT) cDNA UGT73B6 was isolated and it was also indicated that it exists in a single copy in the R. sachalinensis genome. Moreover UGT73B6 transcripts were more abundant in roots and callus than in stems and barely detected in leaves. The detected transcript levels were well correlated with the salidroside content where higher amounts of salidroside were found in roots and callus than in stems and leaves. Next UGT73B6 was successfully transferred into R. sachalinensis by Agrobacterium tumefaciens. Overexpression of the UGT73B6 gene resulted in an evident increase of salidroside content up to 2-and 2.4-fold in transgenic plants and transgenic calli, respectively, compared with the untransformed controls. These results suggested the role of UGT73B6 in the conversion of tyrosol to salidroside in R. sachalinensis. However, UGT73B6 expression was scarcely detectable in leaves with unchanged salidroside content which indicated the possibility that other UDPglucosyltransferases might exist.
Further examinations lead to the isolation of two glucosyltransferase genes/cDNAs from roots and cells Phytochem Rev (2015) 14:657-674 667 of R. sachalinensis cultured in the presence of methyl jasmonate: UGT72B14 and UGT74R1 . The transcripts of the former glucosyltransferase was more abundant in roots, while the latter glucosyltransferase was highly expressed in callus tissue. The relative activity of recombinant UGT72B14, UGT74R1 and UGT73B6 was investigated revealing the highest relative rates towards tyrosol, salicylic acid and p-coumaric acid, respectively. The overexpression of UGT72B14, UGT74R1 and UGT73B6 in hairy roots established with A. rhizogenes resulted in salidroside contents of about 19.81, 5.72 and 8.76 mg/g DW, respectively. Moreover, as expected the UGT73B6 hairy roots were capable of producing higher amounts of salidroside in comparison to UGT73B6 transgenic plants or callus tissue reported by Ma et al. (2007). The activity of tyrosol-glucosyltransferase (TGase) and the efficiency of salidroside production through biotransformation in natural and transformed root cultures of R. kirilowii were investigated by Grech- Baran et al. (2013). To induce the TGase activity, tyrosol 2.5 mM was added to natural and transformed root cultures on the day of inoculation. The first peak in TGase activity of 0.23 U/lg was detected on day 9 in the natural root culture, which correlated with the highest salidroside content of 15.8 mg/g DW. However, the highest TGase activity of 0.26 U/lg was noted on day 15. In the transformed root culture, on day 18 the highest TGase activity of 0.15 U/lg, was observed and coincided with the highest salidroside content of 2.4 mg/g DW. These findings prompted us to carry out supplementation of cultures with tyrosol on the days of the previously detected highest activity of TGase. The applied strategy resulted in a notable increase of salidroside production of up to 21.9 mg/g DW and 7.5 mg/g for the natural and transformed root cultures, respectively. Wu et al. (2003) described the feeding of salicylic acid (SA) (a signaling compound acting in response to stress factors), L-phenylalanine (Phe), L-tyrosine (Tyr), L-tryptophan (Trp), and L-tyrosol in suspensions of compact callus aggregate (CCA) cultures of R. sachalinensis. Each precursor or SA was added during the callus inoculation time. The results implied that feeding of Trp did not increase the salidroside yield; nevertheless, the addition of SA, Phe, and Tyr at a concentration of below 15 mg/L increased the salidroside content slightly. Supplementation with tyrosol at a concentration of 4 mM resulted in a boosted production of salidroside up to 57.7 mg/g DW.
The approach of tyrosol addition into CCA cultures of R. rosea was also described by Krajewska-Patan et al. (2007a). The most pronounced results seen after the biotransformation were with the use of 5 mM tyrosol. The salidroside content obtained was comparable to the results described above for the R. sachalinensis CCA culture.
Upon the challenge for increasing the secondary metabolite accumulation, Agrobacterium rhizogenes has been used for the induction of transgenic root cultures. Transformed root cultures are gaining increasing attention because of their fast growth, and genetic and biochemical stabilities (Giri et al. 2001). Thus, the biotransformation possibility in transformed root cultures of R. sachalinensis was studied by Zhou et al. (2007). The use of elicitors (Aspergillus niger, Coriolus versicolor, Ganoderma lucidum) and precursors (L-tyrosol, L-tyrosine, and L-phenylalanine) resulted in a significant increase of salidroside content, from 4.2 mg/g DW in the control to as much as 6.5 mg/g DW and 6.8 mg/g DW after treatment with the Ganoderma lucidum or L-tyrosine, respectively. The optimal concentration for elicitor and precursor in the liquid medium was 0.05 mg/l and 1.0 mM, respectively.
With the aim of enhancing the level of rosin and its derivatives through biotransformation in R. rosea cell suspension cultures, Furmanowa et al. (1999) measured the glycosylation level of exogenously applied trans-cinnamyl alcohol (CA) 2.5 mM. More than 90 % of the trans-cinnamyl alcohol, used as the substrate for biotransformation, was converted by cells of R. rosea to a number of constituents, including rosavin of up to 10.01 mg/g DW (1 %) of cell DW within 72 h. In studies involving R. rosea compact callus aggregate liquid culture, György et al. (2004) observed that optimal concentration of CA was 2 mM for rosin production with a yield of 12.5 mg/g DW. The highest rosavin amount of 0.83 mg/g DW was determined at the presence of 0.1 mM CA. Moreover, no rosarin was detected. Furthermore, the authors investigated the possibility of increasing the yield of biotransformation products by adding glucose to the culture medium (György et al. 2005). When the CCA liquid cultures were supplemented with 10 g glucose and 2 mM CA, the rosin content increased to 5.7 mg/g DW, and 4 new unexpected compounds related to rosin derivatives were detected. Nevertheless, the rosavin yield was still very low and reached a top concentration of only 0.02 % DW.
In order to investigate the biotransformation capacity of R. rosea CCA on solid medium CA was added at a concentration of 2.5 mM or 5 mM (Krajewska-Patan et al. 2007b). It is worth pointing out that the CCA maintained on solid medium produced the whole spectrum of active compounds present in the intact plant: salidroside, tyrosol, rosin and its derivatives, chlorogenic and gallic acids. The highest obtained rosin amount was 10.6 mg/g DW. Similar to prior studies, rosavin and rosarin were produced in low amounts.
During our investigation, we compared the rosin and its derivatives production by biotransformation in R. kirilowii natural and hairy root cultures (Grech- Baran et al. 2014). The obtained results confirmed that the addition of 2.5 mM exogenous CA enabled rosarin, rosavin, and rosin production in natural root culture and rosarin and rosavin production in hairy root culture. Approximately 80 and 95 % of the glycosides were released into the medium for natural and hairy root cultures, respectively. The highest intracellular rosavin yield 9.8 mg/g DW and total (intracellular ? extracellular) 505 mg/l, was observed in hairy root culture when cinnamyl alcohol was applied on the day of inoculation with the addition of sucrose on the 14th day of culture. For rosin production, supplementation with CA alone on day 14 was more favourable with the highest intracellular amount 5 mg/g DW in natural root culture. Only traces of rosarin were detected.

Conclusions
The discovery of multiple pharmacological activities of Rhodiola plants, in the majority R. rosea, and the increasing demand for Rhodiola preparations led to extensive research on the chemical profiling of the Rhodiola extracts and searching for a stable supply for plant material. The biological properties of Rhodiola extracts were initially attributed to the presence of salidroside which was first isolated from Salix triandra in 1926 (Brown et al. 2002). Salidroside could be produced synthetically on the multikilogram scale  or could be obtained in amounts which often exceeded its content in Rhodiola plants from other botanical families i.e. Salicaceae, Ericaeceae. But salidroside proved not to be the only active component of the extracts. Apart from salidroside: rosavin, rosin and rosarin were believed to be important for the pharmacological effects of Rhodiola extracts. Further detailed chemical examination of Rhodiola extracts revealed approximately 140 compounds from different chemical groups exerting a variety of pharmacological activities (Panossian et al. 2010). Moreover the superior efficacy of total Rhodiola extract to its single ingredients was demonstrated (Sokolov et al. 1985(Sokolov et al. , 1990Skopińska-Ró _ zewska et al. 2008b).
Nowadays nearly all raw plant material comes from natural resources, areas with short vegetative seasons and low temperatures. The intensive collection of plants for commercial purposes is connected with the excessive exploitation of natural populations which have become threatened with extinction (Brown et al. 2002;Galambosi 2006;Kolodziej and Sugier 2012;Peschel et al. 2013).
To meet the increasing need for Rhodiola plants R. rosea was introduced to the field cultivation in Russia, Poland, Germany, Sweden and Finland (Galambosi 2006; Kolodziej and Sugier 2012), R. kirilowii in Poland (Krajewska-Patan et al. 2009) and R. sachalinens in northern China, however in the latter case it was reported that cultivated plants were very sensitive to root rotting diseases (Meng et al. 1994;Xu et al. 1999). Moreover, plants cultivated in the fields require at least 5-7 years to be ready for harvest and to achieve a composition comparable to the plants gathered from natural populations (Galambosi 2006;Kołodziej and Sugier 2012).
On the basis of available data in plants of R. crenulata, R. kirilowii, R. quadrifida, R. rosea and R. sachalinensis higher salidroside content was determined in plants collected from natural sources than those derived from field cultivation. The highest salidroside amount was detected in wild plants of R. kirilowii (2.5 %). However, there is no available data on R. crenulata and R quadrifida field cultivation.
Plant biotechnology offers the opportunity to obtain plant material containing the desired compounds, often in higher amounts than the mother plant. Moreover plant collection to initiate in vitro cell and/ or organ cultures could be performed with respect to the actual laws protecting endangered species. The specific organ or tissue connected with the biosynthesis could be maintained in in vitro cultures to achieve the highest production of compounds of interest. Among the various biotechnological strategies applied to enhance salidroside accumulation in Rhodiola cell, and organ cultures the precursor feeding approach seemed to be the most efficient. In R. sachalinensis cell suspension cultures the salidroside content was ten times higher than in plants collected from natural sources. While the addition of growth regulators alone to the in vitro cultures did not result in a significant increase in salidroside production.
The presence of rosin and its derivatives were affirmed to wild plants of R. rosea, R. quadrifida and R. sachalinensis. In cultivated plants their presence was confirmed only in R. kirilowii and R. rosea, with higher amounts detected in R. rosea. In the in vitro cultures a significant enhancement in the production of rosin and its derivatives was observed when the cultures were fed with the precursor cinnamyl alcohol. In R. rosea compact callus aggregate cultures the observed rosin and rosarin content was even higher than in field cultivated plants while the rosavin level was five times lower. The developed hairy root cultures of R. kirilowii supplemented with cinnamyl alcohol addition exhibited a multiplied potential for the production of rosin and its derivatives in comparison to field cultivated plants.
The results presented in the review point out that biotechnological methods could be an attractive approach to produce salidroside, rosin and its derivatives at higher or at least comparable levels to wild or cultivated plants.
Open Access This article is distributed under the terms of the Creative Commons Attribution License which permits any use, distribution, and reproduction in any medium, provided the original author(s) and the source are credited.