Tropane alkaloids GC/MS analysis and low dose elicitors’ effects on hyoscyamine biosynthetic pathway in hairy roots of Algerian Datura species

Plant secondary metabolites play a major role in plant adaptation to stress. Species belonging to Solanaceae family and Datura genus produce tropane alkaloids: D. stramonium, D. tatula and D. innoxia. These species are cultivated for their hyoscyamine (HS) content, whence the interest of this study to induce transformed roots of these species with strain A4 of Agrobacterium rhizogenes. Hairy roots (HRs) of Datura were established at high frequency by infecting vitroplants. All HRs (343 independent lines) were next employed to study the production of HS and growth. A screening of HRs alkaloid content by GC/MS is performed; it reveals, for the first time, the production of 13 alkaloids by the selected root lines. The selection of high productive line offers an interesting option to enhance the productivity. As HS is the dominant compound, the lines of Datura species were selected for their characteristics for biomass and HS production. The elicitors salicylic acid (SA) and acetyl salicylic acid (ASA) were also used to increase HS production. The results showed that the optimal concentration of the two elicitors (AS and ASA) was 0.1 mM. The highest HS content (17.94 ± 0.14 mg g−1 D.W.) obtained in HRs of D. tatula treated with 0.1 mM of acetyl salicylic acid.

The family of Solanaceae represents a very important source for secondary metabolites such as alkaloids. Used in medicine, alkaloids are considered to be anticholinergic drugs model. Pharmaceutical industry manufactures active pharmaceutical substances containing tropane moiety in their structure, which are applied as mydriatics, antiemetics, antispasmodics, anesthetics and bronchodilators 1 . From the Solanaceae, the Datura species are particularly rich for their tropane alkaloid content among which hyoscyamine (HS) and scopolamine were the main compounds because of their operation on the human nervous system 2 . Several species of genus Datura such as Datura tatula L., Datura stramonium L. and Datura innoxia Mill. show genetic diversity in relation to the alkaloid content 3 . Most of these products derived from wild plants. However, in most cases, the productivity is too low for an economically feasible production and the interest for HS is always in increase 4 .
Classical and most used strategies to improve secondary metabolites productivity of plant cell and tissue cultures are three. The first one is the screening and selection of high producing line; the second is the optimization of growth and production media and the last one is based on the induction of secondary metabolite production by abiotic or biotic elicitors and metabolic engineering.
Hairy roots (HRs) obtained by genetic transformation with Agrobacterium rhizogenes have been extensively used for in vitro production of HS and scopolamine 5 . They seem to be the most promising approach for feasible alkaloid production. They are characterized by a reasonable genetic and biosynthetic stability of selected transgenic line. It is thus of interest to be able to improve secondary metabolite production by means of genetic engineering 6 . So, after induction of HRs, the first step consists on screening of high producing root lines 7  Elicitation is a strategy for enhancing the metabolites in the adventitious root cultures of many plants. Since the biosynthesis of secondary metabolites is controlled during development and the metabolites are accumulated in response to abiotic stresses or attack of pathogens, stress-signaling molecules are frequently used in elicitation experiments. Elicitors are generally considered to modulate many morphological and physiological events in plants 8,9 . These remains a need to increase alkaloid production rates to favor their commercial exploitation 10 . Among the various elicitors, salicylic acid (SA) and acetyl salicylic acid (ASA) are used for improve the alkaloid production in transgenic roots 11,12 . SA is stress-signal molecules related to the activation of plant defense genes and to the biosynthesis of secondary metabolites 8 .
In this communication, we report the establishment of stable hairy root cultures of three species belonging to the genus Datura, obtained by infection with Agrobacterium rhizogenes strain A4. These transformed root cultures were screened for their high alkaloids production and their biomass.
GC/MS approach has been proved being an effective analysis method for alkaloid identification and quantification in Datura genus 13,14 . In this study, we report the tropane alkaloid spectrum of HRs of three Algerian Datura species.
In addition, works on elicitation of alkaloid production showed that the increase in the production was accompanied by a loss of biomass 11,15,16 . Present study was focused on the improvement of HS production by HRs using two elicitors (AS and ASA). Elicitors used at low concentrations and with relatively important elicitation time in order to increase HS production and to avoid biomass losses.
Finally, selected lines were analyzed after treatment with SA and ASA for their tropane alkaloid metabolism. Factors influencing root growth, biomass and HS production are evaluated and discussed.

Results
Screening of hairy root lines. In total 343 HR lines were generated and three lines were selected, one for each species. The first selection criterion was lateral and longitudinal growth. Root's length, ramification's degree (number of secondary and tertiary roots relative to the main root), biomass ( Fig. 1) and HS production were observed and measured. The characteristics of the three selected root lines after 20 days of culture on B5 17 liquid medium were described by Harfi et al. 16 .

GC/MS analysis for no elicited root lines. In our study, GC/MS analysis of the three non-treated
Datura selected roots lines detect the presence of 13 tropane alkaloids ( Effects of SA and ASA elicitation on hyoscyamine production. As HS the dominant compound, the study of elicitation focused on the possible ways to stimulate its production in the three selected root lines using different concentrations of the two elicitors. Table 2 shows the results obtained after elicitation of HR lines previously selected for the three species of Datura. The variance analysis shows a significant effect of salicylic acid (SA) and acetyl-salicylic acid (ASA) on HR's HS content. The Tukey's test revealed 5 homogenous plots for L DS , 6 for L DT and 7 for L DI .
Addition of 0.1 mM ASA resulted in gave the maximum yield of HS for the selected line of D. stramonium (L DS ) after 24 h of exposure (11.58 ± 0.17 mg g −1 D.W.) and increased the HS content up to 1.42-fold compared with the control over the same period (8.16 ± 0.61 mg g −1 D.W.). The best results with SA were obtained with always concentration of 0.1 mM. However, the transgenic root growth of D. stramonium was not affected significantly by SA and ASA treatments at the concentrations of 0.1 and 0.01 mM (Table 2). Nevertheless, when the cultures were treated with 10 mM ASA, the root biomass decreased to 38% compared to the control culture.
The HS content of HRs of the selected line with D. tatula (L DT ) reached maximum (improvement factor > 1) with 0.1 mM of SA or ASA with respectively 16.78 ± 0.21 mg g −1 D.W. and 17.94 ± 0.14 mg g −1 D.W. Also ASA at 10 mM leads to a decrease of HS content and reduced HS to 4.01 ± 0.23 mg g −1 D.W. given an improvement factor In the present study, the effect of SA or ASA on the production of HS was optimal at 0.1 mM without HRs biomass loss. The addition of the elicitors SA and ASA at 10 mM affected significantly the root growth of the three selected lines.

Discussion
To obtain high alkaloid-producing root clones, systematic selection, based on growth rate and alkaloid content, is necessary because the induced HRs lines exhibited different growth and individual biochemical properties in this study. The majority of HRs showed a fast growth. Nevertheless, some transgenic roots were characterized by a weak growth and/or callus formation. These root lines were systematically eliminated.
Published studies on tropane alkaloids content of Datura HRs report the production (detection) of only two alkaloids, namely, hyoscyamine and scopolamine 2,3,11,15 . In this study, GC/MS analysis of no treated root lines shows the presence of 13 tropane alkaloids, HS being the major compound. El-Bazaoui et al. 18 reported the identification of nine tropane alkaloids in the whole plant of D. stramonium by GC/MS. This study is the first to reveal the biosynthesis of 13 alkaloids in HRs of the three species of Datura genus.
SA and ASA are involved in signal transduction and induce the transcription of biosynthetic enzymes involved in the formation of defense compounds in plants 19 . The general cellular process and regulatory principle for activation of plant secondary metabolite biosynthesis is that, an extracellular or intracellular signal is perceived by a receptor on the surface of the plasma membrane or endomembrane; the elicitor signal perception initiates a signal transduction network that leads to activation biosynthesis of transcription factors, which regulate the expression of biosynthetic genes involved in plant secondary metabolism 8 . The HRs of Datura sp. were treated with SA or ASA for 24 h with low different concentrations, after incubation, the biomass and HS content were determined and compared to the untreated control.
The effectiveness of ASA and SA elicitors used as a tool to enhance the production of alkaloids has been described for different species 20,21 . Elicitation has been developed as a promising alternative for in vitro producing metabolites by plant cell cultures 8 . SA is also a signal molecule in so-called induced systemic resistance (ISR) but SA is not referred as an universal inductor in the formation of defense compounds in plants. Nevertheless    When the HR cultures were treated with methyl jasmonate, yeast extract or oligogalactoronids, the improvement factor of tropane alkaloids in HRs is function of elicitor, with the following order: methyl jasmonate, yeast extract and oligogalactorunids. The same study showed that methyl jasmonate affected the HS rate in HRs; compared to control without elicitation the authors recorded an improvement factor of HS content of 1.82-fold 11 .
Generally, elicitation influences the root growth and the biomass in the transformed root culture. When the cultures of adventitious roots of Scopolia parviflora were treated with 2.0 mM methyl jasmonate, the root growth index (ratio between the dry weight of the harvested root and dry weight of initial inoculum) decreased to 63% of the control culture after 72 h of exposure 10 . Zabetakis et al. 11 reported that the treatment of Datura stramonium HRs, with 0.1 µM methyl jasmonate, reduced fresh weight to 11.3% of the control. Similarly results were reported with jasmonic acid in Datura stramonium HRs 22 . The elicitation with NaCl (from 17.24 to 172.4 mM) and Na 2 SO 4 (from 7 to 70.42 mM) increasing concentrations improves HS and scopolamine accumulation into adventitious roots of D. metel but decreases significantly the growth index of root cultures 15 . However, Kang et al. 10 reported that SA at low levels did not have a negative effect on growth of adventitious roots of Scopolia parviflora and did not stimulate root browning unlike methyl jasmonate.
Ours results confirmed these conclusions. Treatments with 0.1 and 0.01 mM SA and ASA enhanced HS production without effect on HRs biomass compared to the control. Nevertheless, high concentrations of SA and ASA (10 and 1 mM) lead to a biomass decrease to 38% for Datura stramonium line. This result agrees with those reached for some other plants, e.g. adventitious root of Scopolia parviflora elicited with 2 mM of methyl jasmonate, a loss of a third of root biomass (dry weight) was recorded, however, treatment with 0.1 mM SA or 0.1 mM of methyl jasmonate does not affect root growth 10 .
Amdoun et al. 22 reported HRs of Datura stramonium, after 28 days of culture with HS production of 2.1 and 3.8 mg g −1 D.W. into B5 medium and optimized B5 medium respectively. In the case of jasmonic acid elicitation, the HS content was 4.2 and 8.5 mg/g D.W. in the HRs grown respectively in B5 medium and optimized B5 medium but the HR biomass was decreased. Our results show that the HS content of Datura stramonium HRs (control) is 8.16 mg g −1 D.W. after 20 days of culture into B5 medium. The HS production of this line, after ASA elicitation, is 11.58 mg g −1 D.W. Nevertheless, the highest content is 17.94 mg g −1 D.W. for HR line of Datura tatula after elicitation with ASA (0.1 mM) without biomass inhibition.
Demand for natural products has been continuously increasing because of their significant value, mainly as pharmaceuticals. Aseptic systems for biotechnological and large-scale production of plant-derived secondary metabolites are significantly improved. These molecule of interest bio-production systems are attractive tools because of their stability. Plant secondary metabolism evolved in the context of highly organized and differentiated cells and tissues, featuring chemical complexity operating under tight environmental 4 .
The results obtained in Table 3 show the significant effects of ASA and SA elicitation on HS production. For the three species of Datura, elicited HR abundances (mg g −1 D.W.) were compared to those not elicited and to the bibliographic data by Shimomura et al. 23 , Boitel-Conti et al. 24 and Harfi et al. 3 . We noted that the ASA elicitation (0.1 mM) which has been used with the Datura tatula HRs enhanced HS accumulation 6.08-fold compared to bibliography. The highest improvement factor (2.82-fold) was observed with the HRs elicited with ASA (0.1 mM) for the Datura innoxia lines compared to the control. Our experiments with ASA (0.1 mM) are another example for the significant role of elicitation to induce the production of HS. The present study shows that the highest HS improvement rate after elicitation (1.32 to 2.82 -fold) corresponds to literature. Nevertheless, the HS contents of our selected HR lines are already high, so the improvement rates are between 1.89 to 6.08 fold.
The HRs are generally a heterogeneous population which the individual lines show various genetic and physiologic characteristics. Therefore, the screening of HRs is a prerequisite to obtain high producing lines. Effectively, Yukimune et al. 25 and Harfi et al. 16 demonstrated that the systematic selection could be a tool for the production of high producing lines of secondary metabolites. This can be explained by the fact that obtaining a large number of root lines increases the probability of retaining, after selection, an interesting HR with good secondary metabolite and biomass production.
In summary, our study confirmed the role of SA and ASA into HS induction in HRs of three Datura species without biomass decrease. The role of SA on tropane alkaloids elicitation maybe explained by the fact that SA is a     3 with Agrobacterium rhizogenes. The used A4 strain of A. rhizogenes is an agropine-type plasmid. After the inoculation of bacteria on solid medium (activation of A. rhizogenes conserved à 4 °C), they were put in suspension in the YEM (Yeast Extract Medium) liquid culture medium for 72 hours (approximately 10 6 germs/ml). The co-culture (bacteria -explant) was done by simple deposit using a syringe on the level of the basal section of the hypocotyls fragments, from 0.5 to 1 cm, taken on the Datura sp vitro-seedling two months old. The culture of hypocotyls is operated on MS solid medium containing 250 mg L −1 cefotaxime. Only emergent roots from the infection site were retained. Four or five weeks after inoculation, the first HRs appeared at the infection sites of explants. Root segments were excised when they reach approximately 2 cm length and transferred into MS solid medium containing 250 mg L −1 cefotaxime. After several passages onto B5 medium for elimination of the bacteria, the HRs were placed on fresh B5 solid medium without antibiotic and were cut off in 2 cm long segments.
Induction HRs was estimated by the induction rate of HRs corresponding to a percentage of vitro-plants with transgenic roots at the infection site, the time of the first hairy root apparition and the mean of transgenic roots for each vitro-plant.
Screening of effective hairy root lines. The selection of efficient hairy root lines realized according to the protocol described by Harfi et al. 16 . Hairy roots that exhibited good growth (length of the main root greater than 15 cm after 20 days of culture on solid B5 medium) were retained and selected according to three main criteria: average speed of growth (cm per day), hairy root dry weight (D.W. [g]), and average levels of hyoscyamine (mg g −1 D.W.) after 20 days of culture. Selection of efficient root lines was carried out in Petri dishes containing hormone-free B5 medium supplemented with 7 g L −1 agar and 20 g L −1 sucrose placed in culture chamber at a temperature of 26 ± 1 °C and in total darkness.

Alkaloids extraction.
The extraction for all samples carried out on 50 mg of dry matter 22 . HRs were dried for 48 h at 40 °C, powdered and then, 50 mg were extracted with 6 mL of hexane for 5 min. The hexane phase containing fat compounds (but not alkaloids) was then discarded. 12 mL of HCl 0.1 N were then added for 10 min and after centrifugation, NH 4 OH (28%) was used to reach pH 10. Aqueous phase was filtered and extracted three times with an equal volume of chloroform. Organic phase was then dried with anhydrous Na 2 SO 4 . After another filtration step, the organic phase was evaporated and the residue was suspended again in 5 ml of dichloromethane, filtrated at 0.2 µm 22 .
GC/MS (screening of tropane alkaloids content) and GC (determination of hyoscyamine for treated root lines) analysis. For tropane alkaloids screening of selected root lines, the filtrates were analyzed by GC/MS according to the protocols of Nguyen et al. 26 .
GC/MS analysis was conducted on a Trace GC Ultra instrument, equipped with a Triplus autosampler and coupled to a DSQ II mass spectrometer (Thermo Fischer Scientific, Waltham, USA) as described in Nguyen et al. 26 . Alkaloids were analyzed using a low-bleed VF-5 MS column (30 m × 0.25 mm × 0.25 µm) (Varian Inc., Grenoble, France). One microliter was injected in splitless mode at 250 °C. The temperature of the oven was kept at 40 °C for 1 min, increased by 30 °C min −1 to 130 °C and then by 10 °C min −1 to 280 °C and kept at 280 °C for 5 min. Helium was used as carrier gas at a flow of one mL min −1 . Temperatures of transfer line and ion source were 300 °C and 200 °C. Mass spectra were recorded with a scanning range of 30-600 m/z, at 4.7 scans s −1 26 .
Peak finding, peak integration and retention time correction of the GC/MS chromatograms were performed with the XCMS Rpackage. Integrated peaks of the mass (m/z) fragments were normalized across all samples by expressing the peak areas relative to both an internal standard in each sample (ribitol or homatropine), and the exact dry weight of tissue used in each extraction. The XCMS output of integrated peaks was tested for robust integration 26 .
Each tropane alkaloid was identified by comparison of its mass spectrum to the NIST05 database and to the internal database of the laboratory issuing from Nguyen et al. 25 . The retention index of each compound was checked before final annotation. The level of each compound was estimated as % area based on the surface of peaks measured in total ion current 26 .
For treated root lines, the content of HS was analyzed by Gas Chromatography. The GC was performed on a CHROMPACK CP 9002 (Chrompack International B.V., Middelburg, The Netherlands), equipped with a FID detector. The dry residue was dissolved with CH 2 Cl 2 and filtered. A fractionated sample (2 µL) was injected into a capillary DB1 column (Agilent, 30 cm × 0.32 mm internal diameter). The GC conditions were as follows;