Plants offer an enormous potential for humans in applications such as novel drugs, biopolymers, high-value chemical compounds, food and feed. Being ‘fully green,’ they possess the intrinsic capacity for environmentally friendly production processes and sustainability. Each plant cell has the whole genetic potential for differentiation into e.g., root, stem, leaf, or flower tissue. Besides whole plants, plant cells can be cultivated as cell cultures in synthetic growth media and this way they offer a great potential for various biotechnological applications. Plant cell cultures can be divided in two main classes, differentiated and undifferentiated cell cultures. The former consist of e.g., organs like shoots, roots or embryos, whereas callus and cell suspension cultures are referred to as undifferentiated cell cultures. Cell cultures offer advantages to produce cell biomass or metabolites in a controlled environment. Cultivation in laboratory or in bioreactors enables production processes which are independent of climatic conditions simultaneously allowing optimization of different production parameters.

Hairy root is caused by a plant disease through the infection of Agrobacterium rhizogenes carrying Ri (root inducing) plasmid. Agrobacterium (Rhizobiaceae) is a soil bacterium, which is able to deliver a part of its own plasmid-DNA (T-DNA) into the nuclear genome of the plant cell. As a result of very complex genetic machinery, hairy roots are then formed in the infection site. Finally they can be detached and grown as independent hairy root clones. Infection of a plant with A. rhizogenes is often associated with the induction of secondary metabolite biosynthesis (Sevón and Oksman-Caldentey, 2002). In the T-DNA there are four genetic loci, called rolA, rolB, rolC, and rolD, which are responsible for the hairy root phenotype and have been shown to positively affect the secondary metabolite production (Palazón et al., 1998; Bonhomme et al., 2000; Bulgakov et al., 2002). These genes can individually induce the secondary metabolite biosynthesis, with rolB probably being the most powerful inducer followed by rolC gene (Sharma et al., 2013). Although the exact mechanism of rol gene –mediated induction of secondary metabolite biosynthesis is not known, it is thought that induction involves the function of signal transduction pathways, including interaction with 14-3-3 proteins and tyrosine phosphatase (Moriuchi et al., 2004). In this work, H. muticus plant was infected with A. rhizogenes carrying hyoscyamine-6β-hydroxylase (h6h) from H. niger, a gene encoding for bifunctional enzyme responsible for tropane alkaloid hyoscyamine conversion into scopolamine (Hashimoto and Yamada, 1986). Hydroxylase activity of H6H leading to 6β-hydroxyhyoscyamine has commonly been observed to be much higher than the epoxidase activity leading to scopolamine (Yun et al., 1992).

Decades of research work has turned this hairy root disease into a valuable biotechnological application (Georgiev et al., 2007; Häkkinen et al., 2014; Vasilev et al., 2014). For a number of desired plants, hairy roots have been induced for the commercial scale production of metabolites, often yielding even higher amounts of metabolites than that of the parent plant (Jouhikainen et al., 1999; Häkkinen et al., 2005; Guillon et al., 2006; Srivastava and Srivastava, 2007). In addition, hairy roots gain biomass rather rapidly and have simple cultivation medium requirements.

Genetic stability of hairy roots is well recognized. They show high genetic stability as well as more stable metabolic production than that of undifferentiated cell cultures. In certain cases the production of secondary metabolites in plants is limited to specialized tissues or organs and the capability is lost in undifferentiated cells. This pattern is also recognized with tropane alkaloids, probably due to the root-specific localization of the tropane alkaloid biosynthetic pathway (Hashimoto and Yamada, 1986). Much of this stability has been related to chromosomic stability displayed by hairy roots (Weber et al., 2008, 2010; Dehghan et al., 2012). Hairy roots usually carry the same chromosomic number and karyotype as the parent plant. In addition, the fact that hairy roots are cultivated without additional plant growth regulators is likely to play a role, since when exposed to growth regulators, even organized tissues modify their chromosomic numbers and display somaclonal variation, as shown by Baíza et al. (1999). However, Guivarc’h et al. (1999) reported quite contradictory findings with hairy roots of Daucus carota, which were shown to be highly unstable both by phenotype as well as by their ability to express the transgenes during 2 years follow-up. Similarly, Mano et al. (1986) reported unstable production of tropane alkaloids in hairy roots of Scopolia japonica, although the follow-up was performed only for 2 months. However, this is a short period as the adaptation time to culture conditions is always required. Only few reports show the accumulation of metabolites produced by hairy roots during a long subculturing. Peebles et al. (2009) reported a 5-year study of hairy roots of Catharanthus roseus which were cultivated in laboratory and showed to be genetically and metabolically stable during this period. Similarly Maldonado-Mendoza et al. (1993) analyzed the tropane alkaloid production of hairy roots of Datura stramonium during 5 years and reported growth rates and alkaloid contents to be stable. In this work, we present, for the first time, the genetic and metabolic activities of engineered hairy roots maintained for 16 years by continuous subculturing. After the establishment of the hairy root line 16 years ago (Jouhikainen et al., 1999), it has been subjected to analysis of production capacity once in Häkkinen et al. (2005) before the current study.

Biotechnology relies to a great part on working with cell cultures and continues to offer a future beyond depletion of natural resources. Establishment and maintenance of cell cultures is laborious and time-consuming process and so far only few successful methods exist for long-term storage for these cultures. Plant cells growing in undifferentiated state are known to be genetically unstable. During continuous subculturing, various rearrangements in chromosomal or gene level may take place, which can be of genetic or epigenetic origin (Larkin and Scowcroft, 1981; Kaeppler et al., 2000). Cryopreservation is a method where cells or tissues are stored in ultra-low temperatures, usually in liquid nitrogen (-196°C), halting all metabolic processes and thus allowing cells to retain their properties unchanged. However, cryopreservation of plant cells still mainly relies on the fully empirical testing, which makes the method development very laborious.

Basically, cryopreservation methods are divided in three categories, those based on slow freezing of the cells by using temperature gradients -0.3…-1.5°C/min, vitrification and encapsulation. The two latter ones do not require special freezing equipment but cells are immersed directly into liquid nitrogen after treatment with vitrification solutions or encapsulated with e.g., inside alginate in order to protect cell organelles from freezing injury. Optimization of intracellular water content is critical for cryopreservation success as plant cells may have water content up to 90%. Intracellular water expands during crystallization and may lead to bursting of cellular membranes. In addition, ice formation increases the solute concentration of remaining liquid phase in intracellular space which may cause toxicity during slow freezing. Prevention of intracellular water concentration may be overcome by several ways. Osmotic pretreatment means using non-permeable agents such as sugars and sugar alcohols which cause dehydration of cells. Cryoprotection on the other hand covers the use of permeable agents which function through a colligative effect and usually result in the increase in membrane fluidity. Freeze-induced dehydration is performed by controlled-rate cooling until terminal temperature -35…40°C is reached before samples are immersed into liquid nitrogen. Alternatively, when working with alginate bead protection, desiccation by air flow or silica gel is often used. When using vitrification-based techniques, dehydration is done by strong concentrations of osmotic agents and/or desiccation and cryoprotectives are included in vitrification solutions. However, vitrification solutions cause sometimes toxicity to plant cells and therefore are not suitable in all cases. Even though successful cryopreservation methods for a number of plant species have been established, a considerable experimentation is always needed to optimize all the steps in cryopreservation procedure (Reed, 2008). So far, there exists no routine method available for different species and cultures. Even though earlier reports of cryopreservation of hairy roots of Beta vulgaris and Nicotiana rustica (Benson and Hamill, 1991), Artemisia annua (Teoh et al., 1996), Eruca sativa and Gentiana macrophylla (Xue et al., 2008), Panax ginseng (Yoshimatsu et al., 1996), horseradish (Phunchindawan et al., 1997; Hirata et al., 1998), Maesa lanceolata and Medicago truncatula (Lambert et al., 2009) have been published, so far there exists no methods reported for cryostorage of Hyoscyamus hairy roots. Here, we present an easy method for cryopreservation of H. muticus hairy roots by using a simple vitrification procedure, allowing the freezing without complicated and expensive equipmentation. Furthermore, we present for the first time, the metabolite production of high scopolamine producing transformed hairy root clone of Hyoscyamus muticus after 16 years’ constant subculturing.

The hairy root clone KB7 (or formerly 13A7) of H. muticus (strain Cairo) was initiated by transforming with A. rhizogenes strain LBA9402 pLAL21 (carrying 35S-h6h) as described by Jouhikainen et al. (1999). The hairy roots were routinely grown and subcultured every 4 weeks in solid modified B50 medium (Oksman-Caldentey et al., 1991; Jouhikainen et al., 1999). In all the studies discussed in this paper, root inoculum was 100 ± 5 mg (FW) in 20 ml medium, and cultivation was performed in 100 ml erlenmeyer flasks, 100 rpm shaking, +25°C, 8 h/16h dark/light regime. Analysis for PCR, RT-PCR, and alkaloids were performed using 28 day-old roots cultivated in liquid culture medium. In order to verify the presence of the transgene as well as hairy root phenotype, the PCR analysis was performed for h6h (Häkkinen et al., 2005) and for rolB (Sevón et al., 1995). The absence of virD gene in hairy roots were verified using PCR primers described by Hamill et al. (1991) and amplification was performed as follows: initial denaturation at 95°C for 1 min, followed by 35 cycles of denaturation at 95°C for 30 s, annealing at 59°C for 1 min, and extension at 72°C for 1.5 min, followed by the final extension at 72°C for 7 min.

Hairy root samples for total RNA isolation were collected by immersing the roots in liquid nitrogen and storing the samples in -80°C before analysis. Total RNA from H. muticus hairy roots was isolated with the ARNzol kit (REAL, Valencia, Spain). cDNA was prepared by reverse-transcription with SuperScript II Reverse Transcriptase (Invitrogen, Carlsbad, CA, USA). The primer sequences for 18S gene were 5′-ATGATAACTCGACGGATCGC-3′ and 5′-CTTGGATGTGGTAGCCGTTT-3′ (Zhang et al., 2007). The primer sequences for h6h gene were 5′- ACATCTGTGAAGGACTTGGGGC-3′ and 5′-GAACTTGGGTCTGGGCATGG-3′ designed using BT1 Gene Tool Lite (1.0.0.1 version). Real-time PCR was performed using the SYBR Green PCR Mastermix (Roche Applied Science, Mannheim, Germany) in a 384-well platform system (LightCycler 480 Instrument; Roche Applied Science). The 18S gene was used as housekeeping control. Data were analyzed using the LightCycler480 Software release 1.5.0 SP3 (Roche). The statistical analysis was performed with the computing environment R (R Core Team, 2013). The student T-test was used for statistical comparisons. A p-value of less than 0.05 was set for significant differences.

The tropane alkaloids scopolamine, hyoscyamine and 6β-hydroxyhyoscyamine were extracted from 28 days cultivated H. muticus hairy roots as described earlier (Häkkinen et al., 2005). Briefly, lyophilized roots were weighed (50 mg) and the lipids were removed with 2 ml petroleum ether. After vortexing and centrifugation (3000 rpm, 10 min) the solvent phase was discarded and the sample residue was dried under nitrogen flow. After adding H₂O (2 ml) and 50 μg internal standard homatropine (Sigma), the samples were made alkaline (pH 9) and the alkaloids were extracted twice with 2 ml Cl₂CH₂. Before GC-MS analysis, the dried residue was dissolved in 100 μl Cl₂CH₂ and derivatized with MSTFA (N-methyl-N-trifluoroacetamide, Pierce). The tropane alkaloids from the culture medium were extracted correspondingly from 2 ml medium. Analyses were performed with an Agilent 7890A GC combined with a 5975C mass selective detector (MSD) using an Rtx^®-5MS silica capillary column (15 m, 0.25 mm i.d., 0.25 μm phase thickness; Restek, Bellefonte, PA, USA). The oven temperature was increased from 70°C to 270°C at a rate of 10°C min^-1. Helium was used as the carrier gas on constant flow mode at 1.2 mL min^-1. The injector, ion source and interface temperatures were 250, 230, and 240°C, respectively, and the split ratio was 25:1. Samples (1 μL) were injected by a Gerstel Maestro MPS2 sampling system (Gerstel GmbH & Co. KG, Mülheim an der Ruhr, Germany). MSD was operated at electron-impact (EI) mode at 70 eV and the full scan data (m/z 40–600) was collected at a scan rate of 2.6 scans s^-1. Identification of the compounds was based on retention times, GC-MS library comparisons and literature data. Chromatographic data were collected and evaluated using Chemstation Software (Agilent Technologies, USA).

Several cryopreservation methods were tested for long-term storage of H. muticus hairy roots including slow freezing and vitrification with different freezing profiles. Experimental layout is shown in Figure 1.

PCR was performed using h6h gene-specific primers, and it was shown that both transgene and root phenotype responsible rolB region were present in H. muticus hairy roots after 16-year continuous subculturing (Figure 2). As expected, virD amplification, indicating the presence of the bacterial genome was negative.

RT-PCR of H6H expression was performed from original KB7 line and of that after cryopreservation (Figure 3). There was no significant difference in expression level of H6H in original and cryopreserved line (p = 0.8538).

Hyoscyamus muticus KB7 hairy root clone was established in 1998 (Jouhikainen et al., 1999) and the clone has been cultivated since then in the laboratory by subculturing on fresh media every 4 weeks. As described 16 years ago, the hairy roots have preserved their light color and fast growth rate, multiplying the biomass 65–67 fold in 28 days.

In Table 1, the accumulation of tropane alkaloids and growth during the 16-year follow-up is presented. First report of the KB7 line was published in Jouhikainen et al. (1999), with hyoscyamine and scopolamine production 125 mg/l and 14 mg/l, respectively. At that time, the hyoscyamine was determined by RIA (radio immuno assay) and scopolamine by ELISA (enzyme-linked immunosorbent assay), and the intermediate 6-β-hydroxyhyoscyamine was not measured. Six years later, we published a paper where conversion capacity of h6h-expressing Nicotiana and Hyoscyamus hairy roots were evaluated and in that study the endogenous production of the original KB7 was also measured. Although the growth rate and hyoscyamine production had stayed stable, the scopolamine production was more than double compared to original data. This result indicates that enzymatic activity of H6H at that time was for some unknown reason higher than before. Finally, last year the overall yields of alkaloids and biomass was somewhat lower than earlier (Table 1). Compared to the original data, hyoscyamine levels were 80% of the first reported values, whereas scopolamine content remained almost the same. However, when only cellular production capacity is taken into account (measured as mg/g DW) there are no differences between first and last measurement (Figure 4), and it is clear that the lower productivity shown in Table 1 is a result of the lower biomass production, which reflects to the (mg/l) values.

Hairy roots cryopreserved with slow freezing I (section “Slow freezing I”) showed post-thaw recovery. Root tips thawed with the fourth method displayed post-thaw growth 2 weeks after thawing, however, with a relatively low recovery frequency (10%). Similarily, vitrification (section “Vitrification”) showed post-thaw recovery 2 days after thawing (Figure 5), but with even lower recovery frequency (∼2%). The recovered hairy root was then cultivated on a solid medium and H6H expression was analyzed after 21 days. As anticipated, cryopreserved root showed unaltered H6H expression levels as shown in Figure 3 proofing the cryopreservation as a powerful tool for storing engineered biological material.