Comparing aryltetralin lignan accumulation patterns in four biotechnological systems of Linum album

Accepted Manuscript Title: Comparing aryltetralin lignan accumulation patterns in four biotechnological systems of Linum album Authors: Liliana Lalaleo, Rubén Alcazar, Javier Palazon, Elisabeth Moyano, Rosa M. Cusido, Mercedes Bonfill PII: DOI: Reference: S0176-1617(18)30291-8 https://doi.org/10.1016/j.jplph.2018.06.006 JPLPH 52797 To appear in: Received date: Revised date: Accepted date: 22-3-2018 6-6-2018 9-6-2018 Please cite this article as: Lalaleo L, Alcazar R, Palazon J, Moyano E, Cusido RM, Bonfill M, Comparing aryltetralin lignan accumulation patterns in four biotechnological systems of Linum album, Journal of Plant Physiology (2018), https://doi.org/10.1016/j.jplph.2018.06.006 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. Original Paper Comparing aryltetralin lignan accumulation patterns in four biotechnological systems of Linum album PT Liliana Lalaleo1, Rubén Alcazar2, Javier Palazon2, Elisabeth Moyano3, Rosa M Cusido2, Mercedes Bonfill2* Faculty of Agricultural Sciences, Technical University of Ambato, Ecuador 2Laboratory of Plant Physiology, Department of Biology, Health and Environment, University of Barcelona, 08028 Barcelona, Spain 3Departament de Ciències Experimentals i de la Salut, Universitat Pompeu Fabra, Avda. Dr. Aiguader 80, E-08003, Barcelona, Spain SC RI 1 N U * Correspondence Dr. Mercedes Bonfill, E-mail: mbonfill@ub.edu A Abstract A CC EP TE D M Linum album is a herbaceous plant with medical interest due to its content of podophyllotoxin (PTOX), an aryltetralin lignan with cytotoxic activity. Previous studies in our laboratory showed that cell suspension cultures of L. album produced more PTOX than methoxypodophyllotoxin (6-MPTOX), both lignans being formed from the same precursor after divergence close to the end of the biosynthetic pathway. In contrast, the hairy roots produced more 6-MPTOX than PTOX. Taking into account this variability, we were interested to know if the lignan profile of an in vitro PTOX-producing L. album plant changes according to the biotechnological system employed and, if so, if this is due to cell dedifferentiation and/or transformation events. With this aim, we established four biotechnological systems: (1) Wild type cell suspensions, (2) transformed cell suspensions, (3) adventitious roots and (4) hairy roots. We determined the production of four aryltetralin lignans: PTOX, 6-MPTOX, deoxypodophyllotoxin (dPTOX) and β-peltatin. The results show that in vitro plantlets, WT cells and transformed cells predominantly produced PTOX, production being 11-fold higher in the plantlets. Otherwise, the adventitious and hairy roots predominantly produced 6-MPTOX, the adventitious roots being the most productive, with MPTOX levels 1.58-fold higher than in transformed roots. We can infer from these results that in the studied plants, cell differentiation promoted the formation of 6-MPTOX over PTOX, while transformation did not influence the lignan pattern. Keywords: Podophyllotoxin, methoxypodophyllotoxin, cell suspension culture, coronatine, hairy roots, Linum album. 1 1. Introduction A CC EP TE D M A N U SC RI PT Plant secondary metabolism constitutes a network of chemical signals, known as secondary metabolites (SMs), which connect the plant with its environment. Among the wide diversity of plant SMs, lignans are plant phenols biosynthetically derived from phenylpropanoids (Suzuki et al., 2007). The arylnaphtalene lignans are of particular interest, as unlike other lignans they feature a privileged structure containing a tetrahydronaphtalene or tetralin molecule, the 1-aryltetralin skeleton. They include podophyllotoxin (PTOX), deoxypodophyllotoxin (dPTOX), -peltatin, and 6methoxypodophyllotoxin (6-MPTOX) (Figure 1). Lignans play an important role in plant defense, conferring protection against herbivores and microorganisms, and show antibacterial, antiviral and antifungal properties. They have a wide range of applications as components of food, textiles and medicine. PTOX in particular has attracted considerable attention due to its pharmacological properties and has been recently applied to control crop-threatening insect pests (Yang et al., 2017). It shows cytotoxic and antiviral activities and is used for the treatment of genital warts (Condylomata acuminata) caused by the human papilloma virus (Damayanthi and Lown, 1998). PTOX is also the natural source of the semi-synthetic derivatives etoposide, teniposide and etopophos, which are used for the treatment of several types of cancer (Gordaliza et al., 2001; Pujol et al., 2005; Liu et al., 2007). Linum album is a herbaceous medicinal plant that belongs to the Linaceae family. Found in Iran and surrounding countries, it bears white flowers and globose capsules containing 6-10 seeds. Within the Linum genus, L. album belongs to the section Syllinum, whose most representative species accumulate aryltetralin lignans as the major lignans (Schmidt et al., 2010; Konuklugil et al., 2016). While PTOX and 6-MPTOX are the main aryltetralin lignans produced by Linum plants, the accumulation pattern differs from one species to another. For instance, L. flavum and L. nodiflorum predominantly produce 6-MPTOX (Schmidt et al., 2010; Xia et al., 2000), whereas L. album is chiefly a PTOX producer (Malik et al., 2014). However, some L. album plants synthesize more 6-MPTOX than PTOX, indicating that the aryltetralin lignan profile can change according to the plant origin (Schmidt et al., 2010). We were interested to know if the lignan profile of in vitro PTOX-producing L. album plants changes according to the biotechnological culture system employed and, if so, if this is due to cell dedifferentiation and/or transformation events. With this aim, we established four biotechnological systems: (1) wild type cell suspension cultures, (2) transformed cell suspension cultures, (3) adventitious roots and (4) hairy roots. Also, with the aim of studying the effect of elicitation on the lignan profile, we treated the four systems with coronatine, an elicitor previously untried with Linum plants. In its mode of action, coronatine is closely related to (+)- 7-iso-jasmonoyl-L-isoleucine, the endogenous bioactive jasmonate, whose derivative, methyl jasmonate (MeJA), to date has proven to be the best elicitor for increasing the lignan production in L. album cell cultures (van Furden et al., 2005). Finally, in the four biotechnological systems, we compared the production of PTOX, 6MPTOX, dPTOX and -peltatin with the expression level of L. album genes encoding cinnamoyl-CoA reductase (CCR), cinnamylalcohol dehydrogenase (CAD) and pinoresinol-lariciresinol reductase (PLR), which are involved in the PTOX biosynthetic 2 pathway. Figure 1 shows the main steps in the PTOX biosynthetic pathway, adapted from Federolf et al. (2007) and Kumar et al. (2015). 2. Material and methods 2.1. Plant material U SC RI PT Linum album plants were grown in vitro from seeds kindly ceded by Dr M. Yousefzadi (Bander Abbas University, Iran) (Yousefzadi et al., 2012). Seeds were sterilized as follows: after immersion in ethanol 70% for 30 seconds, they were washed in sterile distilled water, dipped in 0.1% HgCl2 solution for 15 minutes within an ultrasonic bath, washed in sterile distilled water, immersed in 3% NaOCl with 0.1% Tween 20 for 15 min within an ultrasonic bath, and rinsed with distilled water repeatedly. To obtain in vitro L. album plantlets, sterilized L. album seeds were placed in Petri dishes with Murashige and Skoog (1962) (MS) medium supplemented with 500 μg/L gibberellic acid (GA3, Sigma-Aldrich) and maintained at 25°C with a photoperiod of 16h light/8h dark cycles. After 30 days, small plantlets developed and were subcultured in Magenta jars (SIGMA) with hormone-free MS medium, which was changed every month. N 2.2. Obtaining wild type and transformed cell suspension cultures. Elicitation. CC EP TE D M A Callus and cell suspension cultures of wild type L. album were developed and maintained as described elsewhere (Yousefzadi et al., 2012). To establish the transformed callus cultures, L. album hairy roots, obtained after infection with Agrobacterium rhizogenes LBA9402, were grown in MS medium supplemented with 2 mg/L N-(2-chloro-4-pyridyl)-N-phenylurea (4-CPPU) and 0.1 mg/L indole-3-butyric acid (IBA) in the same conditions as the WT callus until a non-differentiated mass of cells was obtained. The suspension cultures were established with 2 g in 20 ml of the corresponding liquid media. For WT calli, the MS liquid medium was supplemented with 2 mg/L 1naphthaleneacetic acid (NAA) and 0.4 mg/L kinetin (KIN), and for the transformed calli with 2 mg/L N-(2-chloro-4-pyridyl)-N-phenylurea (4-CPPU) and 0.1 mg/L indole-3butyric acid (IBA). The elicitation began at day 7 of culture in WT and transformed cell suspensions, coinciding with the growth phase in both systems. The elicitor assayed was coronatine at the concentration (1M) previously found to be the most effective in elicitation experiments with other plant species (Onrubia et al., 2013). The samples were taken every day after the elicitation to determine the production. A 2.3. Obtaining adventitious roots and hairy roots. Elicitation. The adventitious root cultures were established from the L. album in vitro plantlets. The roots were separated and cultured in half-strength MS supplemented with 0.5 mg/L indole-3-acetic acid (IAA) to obtain elongated roots, and were kept in the dark at 25°C for 30 days. After this period of time, the adventitious roots were subcultured in hormone-free half-strength MS medium and maintained in the same conditions as above for 30 days. 3 RI PT For the induction of hairy roots, leaf discs of the in vitro L. album plantlets were infected with the LBA9402 strain of Agrobacterium rhizogenes in hormone-free MS medium as described in Chashmi et al. (2013). Samples were kept in the dark at 25°C for 48 h to allow the bacterial infection. After this period of time, leaves were transferred to MS medium supplemented with 500 mg/L Claforan (Cefotaxime) to eliminate all the bacteria. After 30 days, roots from inoculated explants were separated from leaf discs and incubated in MS basal medium. The established root lines were maintained by subculturing every 20-30 days in fresh medium. Liquid root cultures were performed with 2 g of the line in 20 ml of the corresponding media. The elicitor coronatine was added for a final concentration of 1 M at day 23 of the experiment. The samples were taken every day after the elicitation with three repetitions and a control per treatment. 2.4. Growth measurements and cell viability M A N U SC Cell growth was measured as fresh weight (FW) and dry weight (DW). The FW was obtained by filtering with nylon filters of 35 m diameter. Filtered cells were lyophilized to obtain the DW. The cell viability was determined as the percentage of living cells in relation to the total cells using fluorescein diacetate (FDA) and propidium iodide (PI), both at 0.01% (w/v) (Pollard and Walker, 1990). The growth of the roots was measured as in Chashmi et al. (2013): 2g of FW root mass was transferred to a 250 mL Erlenmeyer flask containing 50 mL of hormone-free MS liquid medium at 25ºC for a period of 4 weeks. Flasks were harvested with three replicates every week. The growth of hairy root lines was measured every week. Hairy roots were dried by blotting paper to remove all external moisture and weighed to determine the final FW before being frozen for lignan extraction. D 2.5. Lignan extraction and determination by HPLC-ESI-MS analysis A CC EP TE The extraction was performed from powdered lyophilized cells and roots as described in Yousefzadi et al. (2010a) with some modifications. A solution of 0.2 M hydrochloric acid was added after the methanol extraction in order to hydrolyze the glucosides of the lignans to obtain the aglycons. From the culture medium, after addition of 0.2M hydrochloric acid, 20 mL of media was mixed and vortexed for 2 min with 5mL of dichloromethane (DCM), followed by 1 h sonication at 25 ◦ C. Once the organic phase was recovered, it was evaporated. Lignan (PTOX, 6-MPTOX, -peltatin and dPTOX) separation and quantification was done in a Brisa-LC2-C18 column (3 m 15 x 0.46 mm) from Teknokroma using an HPLC-ESI-MS platform (Varian IT 500MS). Twenty microliters of sample were injected into the column and eluted with acetonitrile (A) and water with formic acid at 0.1% (B) as the mobile phase, with the following gradient (min : % A): (00:40%), (10:67%), (12:100%), (17:100%), (18:40%), (24:40%). The flow rate was 0.8 mL/min, and a 1/3b split was done before the detection by MS. PTOX standard was supplied by Dr. R. Arroo from De Montfort University, Leicester (UK). The standard of 6-MPTOX was provided by Dr. M. Yousefzadi and Dr. M.H. Mirjalili from the Shahid Beheshti University, Tehran (Iran) after purification in his laboratory. dPTOX and -peltatin were purified and given by Dr M.A. Castro at the University of Salamanca (Spain). 4 The four standards were diluted in a matrix of absolute methanol. Every sample was assayed with three replicates. The lignan production was determined using a calibration curve. Lignan data were expressed as micrograms per gram DW. 2.6. Genomic DNA extraction and PCR analysis N U SC RI PT The transformed nature of hairy roots and cells obtained from the dedifferentiation of hairy roots was confirmed by detecting the presence of ROLC and AUX1 genes and the absence of the VIRD1 gene by PCR amplification. Total genomic DNA was extracted with the method proposed by Dellaporta et al. (1983). Around 200 μg of the initial material was used for the extraction. PCR analysis was carried out using PuReTaq Ready-To-Go PCR Beads (GE Healthcare) and the following primers: ROLC-F5’-TAACATGGCTGAAGACGACC3’, ROLC-R5’-AAACTTGCACTCGCCATGCC-3’, AUX1-F5’-TTCGAAGGAAGCTT GTCAGAA-3’, AUX1-R5’-CTTAAATCCGTGTGACCATAG-3’, VIRD1-F5’-ATGTC GCAAGGCAGTAAGCCCA-3’, VIRD1-R5’-GGAGTCTTTCAGCATGGAGCAA-3’. The cycling conditions of gene sequences were: ROLC (94ºC 5 min; 35 cycles of 94ºC 1 min, 60ºC 1 min, 72ºC 1 min; and 72ºC 5 min; AUX1 (94ºC 5 min; 35 cycles of 94ºC 1 min, 60ºC 1 min, 72ºC 1.30 min; and 72ºC 5 min), VIRD1 (94ºC 5 min; 35 cycles of 94ºC 1 min, 56ºC 1 min, 72ºC 1 min; and 72ºC 5 min).The amplicon size was analyzed in bromide agarose gel at 1%. A 2.7. qRT-PCR assay A CC EP TE D M The transformed nature of the induced hairy roots was checked. Transformed calli derived from the hairy roots were also confirmed by PCR. To determine the gene expression, the samples were collected after 6 hours on the first day of elicitation and then every day until the end of the experiment. According to the growth curves the first day of elicitation in the cell suspensions systems was day 7 and in the root systems day 23. Samples frozen in liquid N2 were used to evaluate the expression level of the following genes: cinnamoyl-CoA reductase (LaCCR), which is involved in the formation of coniferyl aldehyde, cinnamylalcohol dehydrogenase (LaCAD), which converts coniferyl aldehyde into coniferyl alcohol, and pinoresinol-lariciresinol reductase (LaPLR), which catalyzes the two previous steps before the direct precursor of matairesinol, a key lignan in the PTOX biosynthetic pathway (Figure 1). All the genes are from L. album. ACTIN1 (LaACT1) from L. album was used as the housekeeping gene. Gene sequences are available at GenBank under the accession numbers: LaCCR, AJ440712.1; LaCAD, AJ811963.1; LaPLR, AB525816.1; and LaACT1, AY857865.1. Oligonucleotides designed using PrimerBlast and MacVector 15.03.3, were tested for the absence of primer dimers and non-specific PCR products. Total RNA was isolated using TRIzol reagent (Invitrogen). Two micrograms of RNA was treated with DNAse I (Invitrogen) and first strand cDNA synthesized using Superscript II (Invitrogen) and a mixture of oligo dT and random hexamers (1:1). Quantitative real-time PCR was performed using the SYBR Green I dye and 1:2 dilution of cDNA as template. PCR was performed on the Roche LightCycler 480 II detector system using the following conditions: 95°C 2 min, 40 cycles (95°C, 15 s; 60°C, 10 s; 68°C, 20 s) followed by a melting curve. 5 Standard curves for each primer set were performed for quantification. qRT-PCR analyses were always performed on three replicates. 2.8. Statistical Analysis PT Statistical analysis was performed with Statgraphics (Centurion XV) and Excel software. The multi-factorial ANOVA analysis followed by the Tukey multiple comparison tests were used for statistical comparisons. A P 0.05 was considered for significant differences. 3. Results SC 3.1.1. Growth and PTOX production in non-elicited and elicited cells RI 3.1. Wild type and transformed cell suspension cultures A CC EP TE D M A N U Growth of wild-type (WT) and transformed cell suspensions, both with and without the elicitor coronatine, was determined over 14 days (Fig. 2). Growth was slightly higher in the WT than the transformed culture, and in both cases was reduced by elicitation. The highest growth was obtained in non-elicited WT cells (19 g FW/L) at day 9, which was the beginning of the stationary phase, as in Yousefzadi et al. (2010b). Elicitation of WT cells with coronatine at day 7 resulted in a transient decrease in growth, which recovered slightly at day 9 and reached a maximum at day 10, at the onset of the stationary phase. Under control conditions, the viability of WT cells was high, reaching values of 90% (day 7) and 78% (day 14). Consistent with its effect on growth, coronatine treatment reduced cell viability to 80% and 70% after 3 and 7 days of elicitation, respectively (Fig. 2A). In non-elicited transformed cells growth increased more gradually than in the nonelicited WT cells, reaching a maximum of 15 g FW/L at day 13, when the stationary phase began, four days later than in WT cells (Fig. 2B). When the transformed cells were elicited, the stationary phase began earlier, at day 10, as in elicited WT cells, but without a decrease in growth. Viability was slightly lower in transformed than in WT cells, with values of 80% (day 7) and 70% (day 14), and 70% (days 3 and 7) after elicitation (Fig. 2B). The maximum PTOX production in non-elicited WT cells was observed at the beginning of the stationary phase at day 9 (47 g/g DW) (Fig. 3A). In the elicited WT cells, the PTOX production reached its maximum at day 11 (42 g/g DW), one day after the transition to the stationary phase and 4 days after elicitation. The PTOX yield was slightly lower than in the non-elicited WT cells, but differences between nonelicited and elicited WT cells were only significant at days 9 and 11 (P0.05). Although PTOX production was increased by elicitation at day 11, it remained lower than the maximum levels in the non-elicited cells. It should be mentioned that this is the first time the effect of the elicitor coronatine has been explored in cell suspension cultures of L. album. In transformed cells, the PTOX production (Fig. 3C) was similar in non-elicited and elicited cells. In contrast with the WT cells, there were no peaks of PTOX production, the levels remaining constant at an average of 36 g/g DW. All the cells produced 6 PT PTOX as the main aryltetralin lignan, and the WT cells were always more productive than the transformed cells, elicited or not (Fig. 3A, C), but without significant differences (P0.05). Elicitation had a slight enhancing effect on PTOX production over time in both systems (WT and transformed cells). However, in WT cells the positive effect was only observed at day 11 (Fig. 3A). We conclude that in WT cells the maximum production of PTOX occurs during the transition to the stationary phase, whereas transformed cells provide a more stable and constant system for PTOX production, independently of the growth curves. Interestingly, a temporary shift was observed at the beginning of the stationary phase in elicited WT cells, which exhibited a maximum of PTOX 4 days after elicitation (Fig. 3A). No excretion of PTOX to the culture medium was observed in any case. RI 3.1.2. 6-MPTOX, dPTOX and -peltatin production in non-elicited and elicited cells CC EP TE D M A N U SC All the cell suspension cultures produced more PTOX than 6-MPTOX, the other end product of the PTOX biosynthetic pathway (Fig. 3B, D). In non-elicited WT cells (Fig. 3B), the highest 6-MPTOX production was 9.5 g/g DW at day 10, compared to 5.8 g/g DW in the elicited WT cells at day 11, with significant differences (P0.05). Thus, as occurred with PTOX (Fig. 3A), coronatine treatment did not increase the overall yield of 6-MPTOX (Fig. 3B) compared to the non-elicited cells. Nevertheless, the elicitation did increase lignan production for several days, the enhancing effect being more noticeable for 6-MPTOX than PTOX, with significant differences between nontreated and treated WT cells. In transformed cells, elicited or not, the levels of 6MPTOX production were very low (approx. 2 g/g DW), with significant differences only at day 8 (Fig. 3D). -peltatin, a precursor of 6-MPTOX, was not detected in any type of cell suspension culture, before or after elicitation. dPTOX, a precursor of both PTOX and 6-MPTOX, was detected at a constant level of approximately 40 g/g DW in both WT and transformed cells, indicating that a high pool of this precursor was present throughout the experiment. Our results indicate that the levels of 6-MPTOX/dPTOX/PTOX were maintained at a ratio of 1:10:10 in both types of cell suspension cultures studied. No excretion of these lignans to the culture medium was observed in any case. 3.2. Adventitious and hairy root cultures 3.2.1. Growth and 6-MPTOX production in non-elicited and elicited roots A Fig. 4 shows the growth curves of adventitious and hairy roots, both control and elicited. The stationary phase began at day 24 in the adventitious roots and one day earlier in the hairy roots. The main lignan produced by all root systems was 6-MPTOX (Fig. 5). The adventitious roots were more productive than hairy roots (maximum values of 15 mg/g DW and 9.5 mg/g DW, respectively) with significant differences (P0.05) at days 28 and 25, respectively (Fig. 5B, D). Elicitation with coronatine increased 6-MPTOX production in relation to the controls in both adventitious and hairy roots, reaching peak values of 17 mg/g DW and 10 mg/g 7 DW, respectively (Fig. 5B, D), with significant differences (P0.05) at days 27 and 28, respectively (Fig. 5B, D). However, no significant differences between non-treated and treated adventitious roots were observed (Fig. 5B), as in non-treated and treated hairy roots (Fig. 5D). No excretion of 6-PTOX to the culture medium was observed in any case. 3.2.2. PTOX, dPTOX and peltatin production in non-elicited and elicited roots CC EP TE D M A N U SC RI PT All the root systems assayed (adventitious and hairy roots, non-elicited and elicited) showed a lower production of PTOX than MPTOX (approximately 100-fold less) (Fig. 5A, C). In the non-elicited adventitious roots, the PTOX production was higher than in the non-elicited hairy roots, with a maximum of 135 g/g DW at day 25, one day after the beginning of the stationary phase. This value increased to 150 g/g DW four days after elicitation with coronatine, but the differences between these two values were not significant (P0.05) (Fig. 5A). In contrast, significant differences (P0.05) between non-treated and treated adventitious roots were observed at days 25, 27 and 28 (Fig. 5A). In non-elicited hairy roots, PTOX production was lower than 60 g/g DW, which is about half the yield observed in adventitious roots, and it was not increased by elicitation. Significant differences (P0.05) between non-treated and treated hairy roots were observed from days 25 to 28 (Fig. 5C). Notably, in each system (cells and roots) the largest increase in production after elicitation corresponded to the compound that each system synthesized less. Thus, in the roots, the PTOX pathway was more elicited than that of 6-MPTOX (Fig. 5A, C), and in the cells, vice versa (Fig. 3B, D), as previously observed by van Furden et al. (2005). High values of -peltatin were found in both non-elicited and elicited adventitious roots, especially in the latter, with maximum levels at day 26 (1600 g/g DW and 2350 g/g DW, respectively) with significant differences (P0.05) (Fig. 6A). In elicited hairy roots, the maximum -peltatin production was 1000 g/g DW at day 28 with significant differences (P0.05) in relation to the other days (Fig. 6B). dPTOX was not detected in any type of root system. Our analysis indicates that 6-MPTOX/-peltatin/PTOX levels were maintained at a ratio of 100:10:1. No excretion of these lignans to the culture medium was observed in any case. 3.3. Expression levels of CCR, CAD and PLR genes 3.3.1. Expression levels in WT and transformed cell suspension cultures A The LaCCR, LaCAD and LaPLR genes were always more expressed in transformed than in WT cells (Figs. 7 and 8), especially the LaPLR gene, whose average expression level was 25-fold and 60-fold higher than for LaCCR and LaCAD genes, respectively, with significant differences (P0.05) (Fig. 8A, B, C). Elicitation enhanced the expression of the target genes in both cell systems (Figs. 7 and 8). In transformed cells, coronatine dramatically increased the expression level of the LaPLR gene at day 10, at the beginning of the stationary phase and at day 3 after elicitation (Figs. 2B and 8C), although the production of PTOX and dPTOX remained constant throughout the experiment (Fig. 3C). In contrast, at day 3 after elicitation there 8 SC RI PT was a significant increase in 6-MPTOX levels, which doubled, although production remained negligible (Fig. 3D). In WT cells, the expression level of the LaPLR gene was lower, peaking after 12 hours of elicitation (Fig. 7C). As in the transformed cells, PTOX and dPTOX production did not increase (Fig. 3A) and 6-MPTOX production increased at day 11, while remaining low (Fig. 3B). Maximum transcript accumulation was observed for LaCCR and LaCAD genes in WT and transformed cells (Figs. 7A, B and 8A, B) 6 hours after elicitation. However, as mentioned before, PTOX and dPTOX production was not affected (Fig. 3A, C), whereas an increase of 6-MPTOX (Fig. 3B, D) was observed. Transcriptional profiling of LaCCR, LaCAD and LaPLR in WT and transformed cells, elicited or not, revealed two maximum peaks of expression, one after elicitation, which coincided with the beginning of the growth phase, and the second during the stationary phase (Figs. 2, 7 and 8). This agrees with the pattern of PTOX and 6-MPTOX production, which also exhibited two peaks, although remaining very low (Fig. 3B, D). LaCCR and LaPLR expression levels were higher in cells than in root systems (Figs. 7, 8, 9 and 10), whereas LaCAD levels were highest in hairy roots (Fig. 10B). U 3.3.2. Expression levels in adventitious and hairy root cultures CC EP TE D M A N The expression level of LaCCR and LaPLR genes was similar in adventitious and hairy roots (Figs. 9 and 10), but lower in roots than in cell systems. LaCAD expression was much higher in hairy roots than in adventitious roots or cells (Fig. 10B). Elicitation induced an increase in the expression of LaCCR and LaPLR genes in both root systems (Figs. 9 and 10), whereas it affected the LaCAD gene expression only in hairy roots (Fig. 9B). Notably, adventitious roots reached the highest production of 6-MPTOX (17 mg/g DW) and -peltatin (2.35 mg/g DW) at days 27 and 26, respectively, after elicitation (Fig. 5B and Fig. 6A). In hairy roots, coronatine greatly increased the expression level of the LaCAD gene at day 26 (Fig. 10B); however, the production of 6MPTOX (10 mg/g DW) and -peltatin (1 mg /g DW) was lower than in adventitious roots (Fig. 5D and Fig. 6B). After elicitation, LaCCR and LaPLR mRNA levels in adventitious and hairy roots were similar, peaking at days 2 or 3 (Figs. 9A, C and 10A, C); compared with the nonelicited roots, LaPLR transcripts increased 18- and 60-fold, respectively (Figs. 9C and 10C), and LaCCR gene expression in hairy roots increased around 45-fold, although in both cases the total transcript levels remained low (Fig. 10A). 4. Discussion A In previous studies in our laboratory, callus lines obtained from predominantly PTOXproducing L. album in vitro plants were also PTOX producers (Yousefzadi et al., 2012; Yousefzadi et al., 2010b), whereas hairy roots from these plants mainly produced 6MPTOX (Chashmi et al., 2013). In order to know whether such a change in the lignan profile is due to differentiation or transformation, we obtained calli (WT cells) from L. album in vitro plants and determined if the PTOX profile was maintained. Transformed calli, obtained by dedifferentiation of hairy roots, were also generated to determine whether the lignan 9 A CC EP TE D M A N U SC RI PT production pattern was the same as in the WT cells, or had changed as a result of transformation. The study was carried out using cell suspension cultures established from the corresponding calli. Additionally, hairy roots were induced by infection of the plantlet with Agrobacterium rhizogenes to compare their lignan profile with that of the cells, and adventitious roots were obtained from the plantlets to compare their lignan profile with the other three biotechnological systems. The four systems were assayed for their PTOX/6-MPTOX profile, growth, bioproduction of PTOX, 6-MPTOX, dPTOX and -peltatin as well as the expression level of three L. album genes (LaCCR, LaCAD and LaPLR), in control conditions and after elicitation with coronatine. From our analyses we conclude that WT and transformed cells are predominantly PTOX producers, like the L. album plants, whereas hairy roots and adventitious roots mainly produce 6-MPTOX. The hairy roots reported by Chashmi et al. (2013), also obtained by Agrobacterium rhizogenes (LBA9402), were also 6-MPTOX producers, as mentioned before. They reported a growth curve with the stationary phase beginning at day 28, five days later than here. Regarding the adventitious roots, no comparative data were available for L. album because this is the first time that L. album adventitious roots have been assayed in in vitro experiments. Adventitious roots of Podophyllum peltatum in which only PTOX was determined have been described (Anbazhagan et al., 2008). It is notable that the PTOX-producing transformed cells were derived from 6-MPTOXproducing hairy roots, showing that the lignan profile changed after differentiation. To date, only one report has appeared on L. album transformed cells (Baldi et al., 2010), which were not derived from hairy roots but obtained by co-culture with A. rhizogenes. These transformed cells were also PTOX producers, although the authors did not analyze the lignan profile of the corresponding WT cells and L. album plants, or the hairy roots induced, and only PTOX was determined. Regarding lignan production, the roots produced more 6-MPTOX and PTOX than the cells. Of the four systems tested, the most productive was adventitious roots (Fig. 5A, B), which yielded 1.5-fold more 6-MPTOX than hairy roots, and 2-fold and 2.8-fold more PTOX than hairy roots and cells, respectively. Adventitious roots had a slightly higher growth rate than transformed roots (Fig. 4A), which could be a reason for this greater production. Likewise, transformed cells showed the lowest growth and the lowest lignan production (Fig. 2B and Fig. 3C, D). Accordingly, we can infer that in our biotechnological systems, lignan production was positively related to growth. Interestingly, -peltatin and dPTOX were only found in the systems predominantly producing the respective lignan for which they are precursors. Thus, -peltatin, being a precursor of 6-MPTOX, was only found in the roots, and dPTOX, the immediate precursor of PTOX, only in the cells. Comparing the total lignan production in the four systems and the L. album plantlets, adventitious roots were the most productive (16735μg/gDW), followed by the hairy roots (9650μg/gDW), plantlets (620.22μg/gDW), WT cells (96.5μg/gDW) and transformed cells (78μg/gDW). This is the first time that coronatine has been tested as an elicitor in in vitro L. album cells and roots. It was chosen because of its very similar mode of action to MeJA, which has proved to be the most effective elicitor for increasing lignan production in different L. album in vitro assays (Malik et al., 2014; Satake et al., 2015). Moreover, elicitation with coronatine enhanced taxane production in Taxus media cells more than MeJA 10 A CC EP TE D M A N U SC RI PT (Onrubia et al., 2013), as well as in Corylus avellana cell suspension cultures (Gallego et al., 2015). As mentioned in the Introduction, although our main purpose in using elicitation was to verify if it caused a change in the lignan profile in the four studied systems, we also thought that coronatine could increase the lignan production when used at the concentration found optimum in other plant cells (1M). In the current study, elicitation did not change the lignan profile, and increased lignan production mainly in the roots, especially adventitious (Fig. 5A, B). Significant differences were also observed between non-treated and treated hairy roots (Fig. 5C). Notably, the minority lignan in each system was the most elicited one, that is, PTOX in the roots, and 6-MPTOX in the cells, although in the latter the production levels remained negligible. The growth rate decreased after elicitation with coronatine in all four biotechnological systems, in contrast with coronatine-elicited T. media or salicylic acid-elicited L. album cell suspension cultures, in which the growth rate was not reduced (Yousefzadi et al., 2010b; Onrubia et al., 2013). However, growth in T. media cell suspension cultures was lower after elicitation with MeJA (Onrubia et al., 2013). Overall, in our assays, the enhancing effect of elicitation was more pronounced on the metabolite with the lowest yield (Fig. 3B, D and Fig. 5A, C). Regarding the effect of coronatine on the expression level of the three studied genes, only LaPLR increased its expression in all four systems, above all in transformed cells, where the values were very high (Fig. 8C). However, transformed cells were the least productive system (Fig. 3C, D), in which elicitation barely increased PTOX production. In previous studies with L. album WT cell suspensions, the LaPLR expression level did not change after elicitation with salicylic acid (Yousefzadi et al., 2010b), although PTOX production increased. On the contrary, also in L. album WT cell suspensions, after elicitation with blue light, the expression level of the LaPLR gene increased greatly, although PTOX production improved only slightly (Yousefzadi et al., 2012). Curiously, in our transformed cells, LaPLR expression peaked at day 3 after elicitation and an increase in 6-MPTOX was observed on the same day (Figs. 8C and Fig. 3D), which was not found in the other systems. LaCCR and LaCAD genes were more expressed in transformed than in WT cells, the LaCAD gene being the least expressed in both systems. In roots, the LaCAD gene was strongly expressed only in hairy roots (Fig. 10B), whereas the LaCCR and LaPLR genes achieved similar expression levels in adventitious and hairy roots. Although the levels of all three genes increased in hairy roots (Figure 10A, B, C), adventitious roots were more productive. Kumar et al. (2015), working with PTOX-producing Podophyllum hexandrum adventitious roots, found a very low expression level of PhCCR, PhCAD and PhPLR genes, especially PhPLR. On the other hand, Hazra et al. (2017) found a high expression level of the PhCAD3 and PhCAD4 genes in cell suspension cultures of P. hexandrum after elicitation with MeJA. Thus, coronatine, like MeJA, can up-regulate the genes upstream in the PTOX biosynthetic pathway, such as CAD and CCR, although in our cell and root cultures these genes were apparently not directly involved in the PTOX and 6-MPTOX production. In our cell cultures, the LaCAD gene even had a lower expression than LaCCR and LaPLR. Similarly, it would seem that increases in LaPLR gene levels are not directly related with PTOX or 6-MPTOX production, although this gene is more strongly expressed when the lignan biosynthetic pathway is activated. 11 PT As mentioned, despite showing high expression levels of the tested genes, hairy roots were less productive than adventitious roots and had slightly lower growth (Fig. 4B). This suggests that growth was a crucial factor in lignan production in our systems, whereas the three targeted genes do not seem to have a key role in the PTOX metabolic pathway. Considering the transformation effect, the WT systems showed the highest values of PTOX, 6-MPTOX and -peltatin; after elicitation, the production increased very little although the expression of the LaPLR gene was enhanced. In the transformed systems, as well as a lower PTOX, 6-MPTOX and -peltatin production, a notable increase in the expression of the LaCAD gene was observed in hairy roots. RI 5. Conclusions M A N U SC To sum up, transformation did not increase PTOX and 6-MPTOX production, neither in cells nor in roots, and did not alter the lignan pattern. In contrast, differentiation resulted in a change in the lignan pattern and higher PTOX and 6-MPTOX levels, particularly in the adventitious roots. Elicitation enhanced the expression level of the LaPLR gene in cells and of the LaCAD gene in roots, both more in the transformed than the non-transformed systems (Figs 8C and 10B), although the latter were the most productive. This suggests that the targeted genes, which are upstream in the lignan biosynthetic pathway, are not key genes in lignan production. Future studies should be directed to testing more downstream genes and elucidating the incomplete PTOX pathway in Linum album. Funding TE D This work was partially supported by grants from the Spanish MINECO (BIO201782374-R) and the Generalitat de Catalunya, Spain (2017SGR242). Liliana Lalaleo is grateful for her research grant from SENESCYT. Acknowledgements A CC EP We are grateful to Dr. R. Arroo at De Montfort University, Leicester (UK); Dr. M. Yousefzadi and Dr. M.H. Mirjalili at the Shahid Beheshti University, Tehran (Iran) and Dr. M.A. Castro at the University of Salamanca (Spain) for supplying the lignan standars. We also thank the Scientific and Technological Centers of the University of Barcelona (CCiTUB) for their help. 12 References Anbazhagan, V.R., Ahn, C.H., Harada, E., Kim, Y.S., Choi, Y.E., 2008. Podophyllotoxin production via cell and adventitious root cultures of Podophyllum peltatum. In Vitro Cell Dev. Biol-Plant. 44, 494-501. PT Baldi, A., Srivastava, A.K., Bisaria, V.S., 2008. Improved podophyllotoxin production by transformed cultures of Linum album. Biotechnol. J. 3, 1256–1263. RI Baldi, A., Farkya, S., Jain, A., Gupta, N., Mehra, R., Datta, V., Srivastava, A.K., Bisaria, V.S., 2010. Enhanced production of podophyllotoxins by co-culture of transformed Linum album cells with plant growth-promoting fungi. Pure Appl. Chem. 82, 227–241. SC Chashmi, N.A., Sharifi, M., Yousefzadi, M., Behmanesh, M., Rezadoost, H, Cardillo A, Palazon, J., 2013. Analysis of 6-methoxy podophyllotoxin and podophyllotoxin in hairy root cultures of Linum album Kotschy ex Boiss. Med. Chem. Res. 22, 745–52. N U Damayanthi, Y., Lown, J.W., 1998. Podophyllotoxins: current status and recent developments. Curr. Med. Chem. 5, 205–252. A Dellaporta, S., Wood, J. Hicks, J., 1983. A plant DNA minipreparation: Version II. Plant Molecular Biology Reporter. 1, 19–21. D M Farkya, S., Bisaria, V.S., Srivastava, A.K., 200. Biotechnological aspects of the production of the anticancer drug podophyllotoxin. Appl. Microbiol. Biotechnol. 65, 504–519. CC EP TE Federolf, K., Alfermann, A.W., Fuss, E., 2007. Aryltetralin-lignan formation in two different cell suspension cultures of Linum album: deoxypodophyllotoxin 6hydroxylase, a key enzyme for the formation of 6-methoxypodophyllotoxin. Phytochemistry. 68, 1397-1406. Gallego, A., Imseng, N., Bonfill, M., Cusido, R.M., Palazon, J., Eibl, R., Moyano, E., 2015. Development of a hazel cell culture-based paclitaxel and baccatin III production process on a benchtop scale. J. Biotechnol. 195, 93-102. A Gordaliza, M., Miguel del Corral, J. M., Castro, M.A., García-García, P.A., Feliciano, A., 2001. Cytotoxic cyclolignans related to podophyllotoxin. IL FARMACO. 56, 297304. Hazra, S., Bhattacharyya, D., Chattopadhyay, S., 2017. Methyl jasmonate regulates podophyllotoxin accumulation in Podophyllum hexandrum by altering the ROSresponsive podophyllotoxin pathway gene expression additionally through the down regulation of few interfering miRNAs. Frontiers in Plant Sci. 8. https://doi.org/10.3389/fpls.2017.00164 13 Konuklugil, B., Ionkova, I., Vasilev, N., Schmidt, T.J., Windhovel, J., Fuss, E., Alfermann, A.W., 2016. Lignans from Linum species of sections Syllinum and Linum. Nat. Prod. Res. 1, 1-6. Kumar, P., Pal, T., Sharma, N., Kumar, V., Soot, H., Chauhan, R. S., 2015. Expression analysis of biosynthetic pathway genes vis-à-vis podophyllotoxin content in Podophyllum hexandrum Royle. Protoplasma. 252, 1253-1262. PT Malik, S., Bíba, O., Grúz, J., Arroo, R.R.J., Strnad, M., 2014. Biotechnological approaches for producing aryltetralin lignans from Linum species. Phytochem. Rev. 13, 893–913. RI Murashige, T., Skoog, F., 1962. A revised medium for rapid growth and bioassays with tobacco tissue cultures. Physiol. Plant. 15, 473–497. SC Liu, Y., Yang, L., Tian, X., 2007. Podophyllotoxin: current perspectives. Curr. Bioact. Comp. 3, 37–66. N U Onrubia, M., Moyano, E., Bonfill, M., Cusido, R.M., Goossens, A., Palazon, J., 2013. Coronatine, a more powerful elicitor for inducing taxane biosynthesis in Taxus media cell cultures than methyljasmonate. J. Plant. Physiol. 170, 211-219. A Pollard, J. W., Walker, J. M., 1990. Plant Cell and Tissue Culture. Methods in molecular biology, vol 6, Humana Press, Clifton, New Jersey. M Pujol, M. D., Romero, M., Sanchez, I., 2005. Synthesis and Biological Activity of New Class of Dioxygenated Anticancer Agents. Curr. Med. Chem. 5, 215-237. TE D Satake, H., Koyama, T., Bahabadi, S.E., Matsumoto, E., Ono, E., Murata, J., 2015. Essences in Metabolic Engineering of Lignan Biosynthesis. Metabolites. 5, 270-290. CC EP Schmidt, T.J., Hemmati, S., Klaes, M., Konuklugil, B., Mohagheghzadeh, A., Ionkova, I., Fuss, E., Alfermann, A.W., 2010. Lignans in flowering aerial parts of Linum species – Chemodiversity in the light of systematics and phylogeny. Phytochemistry. 71, 17141728. Suzuki, S., Unmezawa, T., 2007. Biosynthesis of lignans and norlignans. J. Wood Sci. 53, 273-284. A Van Fürden, B., Humburg, A., Fuss, E., 2005. Influence of methyl jasmonate on podophyllotoxin and 6- methoxypodophyllotoxin accumulation in Linum album cell suspension cultures. Plant Cell Rep. 24, 312–317. Xia, Z.Q., Costa, M.A., Proctor, J., Davin, L.B., Lewis, N.G., 2000. Dirigent-mediated podophyllotoxin biosynthesis in Linum flavum and Podophyllum peltatum. Phytochemistry. 55, 537–549. 14 Yang, R., Huang, X., Che, Z., Zhang, Y., Xu, H., 2017. Application of sustainable natural resources in crop protection: Podophyllotoxin-based botanical pesticides derived from Podophyllum hexandrum for controlling crop-threatening insect pests. Industrial Crops and Products 107, 45-53. Yousefzadi, M., Sharifi, M., Chashmi, A., Behmanesh, M., Ghasempour, A., 2010a. Optimization of podophyllotoxin extraction method from Linum album cell cultures. Pharm. Biol. 48, 1421-1425. RI PT Yousefzadi, M., Sharifi, M., Chashmi, A., Behmanesh, M., Ghasempour, A., Moyano, E., Palazon, J., 2010b. Salicylic acid improves podophyllotoxin production in cell cultures of Linum album by increasing the expression of genes related with its biosynthesis. Biotechnol. Lett. 32, 1739–1743. A CC EP TE D M A N U SC Yousefzadi, M., Sharifi, M., Behmanesh, M., Ghasempour, A., Moyano, E., Palazon, J., 2012. The effect of light on gene expression and podophyllotoxin biosynthesis in Linum album cell culture. Plant Physiol. Biochem. 56, 41–46. 15 Figure legends RI PT Figure 1. The main steps in the podophyllotoxin biosynthetic pathway, adapted from Federolf et al. (2007) and Kumar et al. (2015). Solid arrows indicate the steps already known. Written in capitals and framed are the target enzymes of this study. PAL, phenylalanine ammonia-lyase; C4H, cinnamic acid 4-hydroxylase; CCR, cinnamoylCoA reductase; CAD, cinnamylalcohol dehydrogenase; PS, pinoresinol synthase; PLR, pinoresinol-lariciresinol reductase; DOP7H, deoxypodophyllotoxin 7-hydroxylase; DOP6H, deoxypodophyllotoxin 6-hydroxylase; P6OMT, -peltatin 6-Omethyltransferase; PAM7H, -peltatin-A-methylether 7-hydroxylase. Lignans whose production has been determined in this study appear in capitals. SC Figure 2. Growth curves and viability (%) of wild type (A) and transformed cells (B) of Linum album with and without elicitation with 1  coronatine. Each value is the mean of three biological replicates  SD. WT, wild type. A N U Figure 3. PTOX and 6-MPTOX production in wild type (A, B,) and transformed (C, D,) Linum album cell suspension cultures with and without elicitation with 1 M coronatine. Each value is the mean of three biological replicates  SD. Different letters indicate significant differences between means (P0.05). WT, wild type. M Figure 4. Growth curves of adventitious (A) and hairy roots (B) of Linum album with and without elicitation with 1 M coronatine. Each value is the mean of three biological replicates  SD. TE D Figure 5. PTOX, 6-MPTOX production in adventitious (A, B) and hairy roots (C, D) of Linum album, with and without elicitation with 1 M coronatine. Each value is the mean of three biological replicates  SD. Different letters indicate significant differences between means (P0.05). CC EP Figure 6. -peltatin production in adventitious roots (A) and hairy roots (B) of Linum album, with and without elicitation with 1 M coronatine. Each value is the mean of three biological replicates  SD. Means with different letters are significant at P0.05 A Figure 7. Expression level of L. album genes encoding cinnamoyl-CoA reductase (CCR), cinnamylalcohol dehydrogenase (CAD), pinoresinol-lariciresinol reductase (PLR) and actin (ACT1) as the housekeeping gene in wild type (A, B, C) L. album cell suspension cultures, with and without 1 M coronatine. Each value is the mean of three biological replicates  SD. P  0.05 was accepted as level of significance; *** highly significant P < 0.001; ** significant P < 0.01; * less significant P  0.05; NS not significant for P > 0.05. (WT, wild type; d, days; h, hours). Figure 8. Expression level of L. album genes encoding cinnamoyl-CoA reductase (CCR), cinnamylalcohol dehydrogenase (CAD), pinoresinol-lariciresinol reductase (PLR) and actin (ACT1) as the housekeeping gene in transformed (A,B,C) L. album cell 16 suspension cultures, with and without 1 M coronatine. Each value is the mean of three biological replicates  SD. P  0.05 was accepted as level of significance; *** highly significant P < 0.001; ** significant P < 0.01; * less significant P  0.05; NS not significant for P > 0.05. (WT, wild type; d, days; h, hours). RI PT Figure 9. Expression level of L. album genes encoding cinnamoyl-CoA reductase (CCR), cinnamylalcohol dehydrogenase (CAD), pinoresinol-lariciresinol reductase (PLR) and actin (ACT1) as the housekeeping gene in adventitious roots (A, B, C) of Linum album, with and without 1 M coronatine. Each value is the mean of three biological replicates  SD. P  0.05 was accepted as level of significance; *** highly significant P < 0.001; ** significant P < 0.01; * less significant P  0.05; NS not significant for P > 0.05. (d, days). A CC EP TE D M A N U SC Figure 10. Expression level of L. album genes encoding cinnamoyl-CoA reductase (CCR), cinnamylalcohol dehydrogenase (CAD), pinoresinol-lariciresinol reductase (PLR) and actin (ACT1) as the housekeeping gene in hairy roots (A, B, C) of Linum album, with and without 1 M coronatine. Each value is the mean of three biological replicates  SD. P  0.05 was accepted as level of significance; *** highly significant P < 0.001; ** significant P < 0.01; * less significant P  0.05; NS not significant for P > 0.05. (d, days). 17 18 TE CC EP A D PT RI SC U N A M 19 TE CC EP A D PT RI SC U N A M 20 TE CC EP A D PT RI SC U N A M 21 TE CC EP A D PT RI SC U N A M 22 TE CC EP A D PT RI SC U N A M 23 TE CC EP A D PT RI SC U N A M 24 TE CC EP A D PT RI SC U N A M