Enhanced growth and cardenolides production in Digitalis purpurea under the influence of different LED exposures in the plant factory

In this report, we have investigated the influence of different light qualities on Digitalis purpurea under a controlled environment. For this purpose, red (R), blue (B), fluorescent lamp (FL, control), along with combined red and blue (R:B) LEDs were used. Interestingly, the plant growth parameters such as number of leaf, longest root, width of leaf, width of stomata, width of trichome, leaf area, leaf or root fresh weight (FW), weight (DW) as well as length of trichome were maximum under R:B (8:2), and significantly larger than control plants. The stomatal conductance or anthocyanin was maximum under B LED than those under FL, however the photosynthesis rate was greater under FL. RuBisCO activity was maximum under R:B (1:1) LEDs while the quantity of the UV absorbing substances was highest under R LED than under FL. The maximum amount of cardenolides were obtained from leaf tissue under R:B (2:8) LED than those under FL. The R:B LEDs light was suitable for Digitalis plant growth, development, micro- and macro-elements, as well as cardenolides accumulation in the plant factory system. The adaptation of the growth strategy developed in this study would be useful for the production of optimized secondary metabolites in Digitalis spp.

commercial potential of the crop 3 . In vitro screening methods can be useful to ascertain the selection of plant extracts that possess potentially useful components, for further chemical and pharmacological explorations 12 .
Use of PFS ensures a year-round production of plants via the optimization of aerial and root environments 13 . A PFS can also be efficiently used in order to increase bioactive compounds or phytochemicals production in the plants 14 . Optimization, standardization, and absolute regulation of the environment for growth and development of plants have a positive contribution in achieving an improved production of quality crops 15 . In addition, with the application of PFS, a uniform growth can be achieved, production planning and scheduling may be made possible, and contamination of crops (by diseases, insect, metals, and other detrimental elements) can be considerably lessened or completely eradicated 16 . Therefore, growing or cultivation of plants under PFS (controlled environment) can be considered as an convenient and alternative way for a hassle-free and efficient plant production 17 . It is also desirable that the production of pharmacologically important secondary metabolites should take place under controlled optimal conditions. Plants are photoautotrophic and sessile in nature. A whole life cycle of plants is greatly affected by the continuous change in light environment 18,19 . Light is one of the most important factors that affects a plant photosynthesis rate and the amount of phytochemicals produced 20 . Artificial light sources; especially light-emitting diodes (LEDs) have been used in a PFS, where controlled-environmental conditions are needed. The application of narrow-waveband LEDs with the best chosen combination of wavelength makes it possible to optimize the light quality for experimental purpose 21 . Initially, the plant technologists had mostly used the red LEDs as a light source to promote photosynthesis. That scenario has transformed gradually since certain evolutionary changes occurred in the greenhouse plants that were adopted to use a much wider spectrum of light 22 . Optimal development of plants cannot be achieved using red (R) light alone, but it needs blue (B) light as well, to regulate processes at variance with photosynthesis [23][24][25] . B light has been documented to influence vegetative growth, photo-morphogenesis, stomatal opening, chlorophyll synthesis, and secondary metabolite production 26,27 . These reports have encouraged the authors to undertake a systematic investigation on the influence of different light qualities on D. purpurea growth and cardenolides accumulation, under controlled environment, using a PFS with LEDs.
In the current study, we set upon the idea that optimizing the different light qualities and their combinations can increase plant biomass and cardenolides production in the D. purpurea plants. Thus, the aim of the study includes (1) development of an efficient plant growth system using different types of LEDs (alone and in combinations) in PFS, (2) examination of the influence of LEDs on the plant growth parameters, photosynthesis, stomatal conductions, and RuBisCO activity, (3) analysis of the variations of cardenolides (digoxin and digitoxin) in leaf samples. This study is a step forward in exploring plant growth and cardenolides profiles, present in this medicinal plant D. purpurea. The inferences drawn are expected to be helpful in formulating herbal remedies for cancer and related diseases.

Materials and Methods
Reagents and chemicals. All the reagents and chemicals used were of hydroponic and HPLC grades, which were purchased from Sigma-Aldrich (St. Louis, MO, USA), unless stated otherwise.
Plant material. D. purpurea seeds (purple foxglove) were purchased from Aramseed Co. Ltd., South Korea. Seeds germination. Initially, rockwool pellets (Grodan, Netherlands) were fixed in plug trays [240-cell (60 cm × 41 cm × 5 cm)] and then the seeds were sown. The germinated seeds were grown for two months, under non-stress conditions at 22 °C daytime and 18 °C nighttime greenhouse temperatures, and nutrient solution was fed once a week after one month of germination, as well as watered as and when needed. After two months, the well-developed seedlings were transplanted in plant factory system (  (Fig. 1). In the controlled environment, the LEDs were placed horizontally, above the bench top, at a height of 20 cm. The average photosynthetic photon flux density (PPFD) (LI-250A, LI-COR Inc., USA) was adjusted to 150 µmol m −2 s −1 provided by the fluorescent lamps and bar-type LEDs. Light spectral distribution (Fig. S2, Supplementary Information) was scanned using a spectroradiometer (RPS-900R, International Light Co. Ltd., USA). In the PFS, plants were grown on rockwool medium for 35 days under the following conditions: 21 ± 1 °C, 70 ± 10% relative humidity, photoperiod of 18/6 (light/dark), and CO 2 concentration of 500 µmol mol −1 . The details of the nutrient solution used, have been provided in Table S1 (Supplementary Information).
Growth measurements and water potential (Ψw). Hydroponic-grown plants were uprooted with care from trays with rockwool pellets and the same were dehydrated with lint-free wipes, before separating the roots and leaves from each plant. Finally, the separated roots and leaves were kept in blotting paper sheets for further examination of their biomass. Dry biomass was determined after oven drying the leaf and root samples at 65 °C for two days. After determining the fresh-and dry-biomass of the leaf samples, the following formula was used for calculating leaf Ψw potential. The photographs of mounted specimens were taken with a scanning electron microscope (JXA-8530F, JEOL, USA), operated at 15 kV. Stomatal and trichome length, density, and width of the leaves were examined and measured with three replicates, respectively. Stomatal and trichome were observed using a DM4000 light microscope (Leica, Wetzlar, Germany) at different magnifications. For this purpose, fresh leaves were first harvested from PFS, and then very fine layers of leaf tissues were peeled off and transferred to glass slides. Drops of glycerine solution were added to the slides before coverslips were placed onto the surface.
Estimation of chlorophyll and carotenoid pigments. Chlorophyll (Chl) a, (Chl) b, and carotenoids were analyzed and calculated, as described by Lichtenthaler and Welburn 28 . Briefly, fully expanded young leaves of 30-days-old plants were collected and the leaf samples were ground to a fine powder and then transferred to 2-mL Eppendorf tube. Further 1 mL of acetone (80%; w/w) was added and homogenized for 10 min at 4 °C. The absorbance was measured by a UV-Vis spectrophotometer (Biochrom Libra S22) at 479, 649, and 665 nm.
UV absorbing substances (UAS). UAS was extracted and determined following the previously published spectrophotometric protocol 29 . For this, one 0.5 diameter leaf disc was incubated with 5-mL of methanol (99): HCL (1) mix and allowed digestion at −4 °C for 48 h. UAS was measured from leaf extract at 305 nm. Absorbance was expressed on the basis of the leaf FW.

RuBisCO determination by sodium dodecyl polyacrylamide gel electrophoresis (SDS-PAGE).
For the extraction of RuBisCO, freshly harvested leaves were homogenized in 100 mM Tris buffer (pH = 7.5) containing 5% (v/v) glycerol, 5 mM of DTT, and 2 mM iodoacetate; a leaf (1 g):buffer (10-mL) ratio was used for further extraction. A buffer without potassium (K + ) or sodium (Na + ) ions was recommended for RuBisCO analysis by SDS-PAGE, it is because those cations reduce the solubility of dodecyl sulfate (DS). A Trition X 100 was added, before centrifugation (5,000 × g) at 4 °C for 3 min. The supernatant was thoroughly mixed with 2-mercaptoethanol and lithium DS solution (25%, w/v) to final concentrations of 1% (v/v) and 1% (w/v), respectively. The final extraction was rapidly treated at 100 °C for 1 min and then stored at −30 °C, until it was being used for SDS-PAGE analysis. The samples were loaded on a 12% polyacrylamide gel. After completion of electrophoretic run, the gels were stained with silver stain. The stained bands corresponding to larger and smaller subunits of RuBisCO were then cut out from the gels with a razor blade and were eluted in 1-2.5 mL of formamide in s-stoppered amber test tubes and evaluated in the spectrophotometer. RuBisCO content was determined by using the standard curve calculated from the absorbance of a known amount of purified RuBisCO. Assessment of different elements contents in harvested leaves. For determination of the concentrations of ten elements (B, Ca, Cu, Fe, K, Mg, Mn, P, S, and Zn), around 1 g of fresh leaf sample was oven-dried at 45 °C and then digested with concentrated H 2 SO 4 and perchloric acid (50%, v/v) for 2-5 h at 100-300 °C. After digestion, the samples were then filtered with filter paper (Whatman) and finally diluted up to 100 mL with distilled water. The elemental contents were estimated by inductively coupled plasma optical emission spectrometry (ICP-OPTIMA 4300DV/5300DV/Perkin Elmer, Waltham, MA, USA).

Sample preparation and cardenolide extraction. Shoots of well-developed plants, under closed
type PFS, were collected and then freeze-dried at −56 °C, and later used for digitoxin and digoxin extraction. Cardenolides were extracted following the method described by Wiegrebe and Wichtl 30 . Approximately 50 mg dry leaf powder was transferred to 2-mL centrifuge tube and 1 mL of 70% methanol was added. The mixture was kept in an ultrasonic bath for 30 min at 65-70 °C, following which the mixture was cooled on ice for 3-5 min and centrifuged at 13000 rpm for 10 min. The supernatant was collected and mixed with 0.5 mL of 4% (w/v) monosodium phosphate solution and 0.25 mL of 15% (w/v) lead acetate solution. The resultant extract was transferred and diluted with water up to 2 mL and then centrifuged at 12,000 rpm for 8 min at room temperature. The supernatant was collected and mixed with 0.5 mL isopropanol: chloroform (2:3) and centrifuged at 12,000 rpm for 5 min at room temperature. The lower phase was transferred in the new tube as the 'first extraction' . The remaining methanolic solution was used for the 'second extraction' by adding: isopropanol:chloroform and centrifuging at 13,000 rpm for 5 min at room temperature. The first and second extractions were mixed and evaporated under high air for 3 h and finally dissolved in HPLC grade methanol (500 µL). Experimental design, data collection, and statistical analysis. In the controlled environment experiment, a completely randomized design with three replications of each LED treatment (to minimize position effects) with total thirty seedlings per each treatment, were used. Data were collected and statistical analysis was performed with SPSS Version 18 (SPSS Inc., Chicago, IL, USA). The experimental results were subjected to an analysis of variance (ANOVA) and Duncan's multiple range tests 31 . The mean ± SE (standard error) were subjected to Duncan's multiple range test at p < 0.05 level.

Results and Discussion
D. purpurea was selected as a model plant since it is a rich source of cardiac glycoside or cardenolides. For the potential biological effects of LED in the plant factory system, the physiological, morphological, and biochemical investigation of LED on Digitalis growth could be of enormous benefit. In this study, a complex ecosystem in a PFS was constructed to create a model growth environment for Digitalis to further assess the effects of LEDs on plant growth (Fig. 1).
Plant growth parameter measurements and Ψw potential. The effect of LEDs on the shoot and root growth of D. purpurea have been presented in Fig. 2, with the plant growth data (in Table 1 Changes in stomatal and trichome characters. Stomatal density, length, width, and a length-to-width ratio of the Digitalis leaves were measured using the scanning electron microscopic images (Figs 4 and 5). It was recorded that the stomatal density and stomatal length:width ratio were the highest in the plants grown under    (Table 2). It is well known that trichome number and size is partly regulated by light 40 . In this case, stomatal and trichome density may increase or decrease in response to the environmental variations caused by the light intensity, quality or duration.
Influence on photosynthesis rates and stomatal conductance. The photosynthesis rates in the leaves of D. purpurea, under different light treatments have been presented in Fig. 6. Photosynthesis rate increased during the FL treatment, followed by treatment with R:B (8:2). The highest photosynthesis rate was recorded to be 7.6 ± 0.12 µmol CO 2 m −2 s −1 at 1,500 µmol mol −1 of CO 2 concentration in FL treatment, while 7.1 ± 1.3 µmol CO 2 m −2 s −1 at 1,200 µmol mol −1 of CO 2 concentration was recorded in the R:B (8:2) treated plants, respectively (Fig. 6a). This result is consistent with the findings of Goins et al. 41 who reported that wheat plants when grown under R:B LEDs had higher photosynthesis rates. These photosynthesis rates were higher due to increased stomatal conductance (stomata opening) under more B light 42 . Light-saturated maximum stomatal conductance and photosynthesis are closely associated in many plant species 43,44 . It is well known that a higher B light amount is mostly related to the development of 'sun-type' leaves, which are characterized by a higher leaf mass per unit leaf area and a high photosynthetic capacity [45][46][47][48] . Stomatal conductance in leaves was the highest under B (861.5 ± 108.5 at 300 mmol H 2 O m −2 S −1 ) LED treatments (Fig. 6b). Terfa et al. 42 and van Ieperen 49 reported that higher stomatal conductance in LED-grown plants might at least partly be due to a higher number and frequency of stomata per area of epidermal cells. As reported in some studies, stomatal conductance in cucumber plants proportionally increased with increasing B light and it was related to both the aperture of stomata and a higher number of stomata 50 . In another study, Wang et al. 51 observed that stomatal conductance increased in cucumber plants that were grown under B monochromatic light, when compared to plants that were grown under white, R, G, and yellow monochromatic lights.

Influence on anthocyanin content and UV absorbing substances (UAS). A significantly higher
concentration of anthocyanin was measured in leaf tissues of the plants grown under B LED (0.0028 μmol mL −1 ), followed by (0.0018 μmol mL −1 ) R LED. Anthocyanin content was not significantly affected, neither under FL nor under the combinations of R:B LEDs (1:1, 8:2, and 2:8); and as a result, anthocyanin production was recorded to be lower than the plants grown under B and R LEDs individually. It was previously reported that B LEDs is effective in anthocyanin production in most of the plant species, including tomato and cabbage seedlings 56 , strawberry cells 57 , roses 58 , and Chinese bayberry fruit 59 . With regard to UAS absorbing substances in Digitalis leaves, a decrease of 50% was observed in the plants grown under R:B LED combinations (1:1, 8:2, and 2:8), as well as a 75% decrease was observed in B or FL LED-grown plants (Fig. 7). It is widely known that if the level of UV-B    0.313 mg g −1 DW and 0.100 mg g −1 DW, respectively. Among the different types and combinations of LEDs i.e. B, R, R:B (1:1), R:B (8:2), and R:B (2:8), R:B (8:2) exhibited higher shoot fresh weight (FW) and dry weight (DW) (Fig. 3). However, R:B (2:8) produced less shoot FW and DW, with highest digitoxin and digoxin accumulation. The highest cardenolide accumulation have been reported in D. lanata by Ohlsson et al. 10 , using light in the blue region. Similarly other authors observed that irradiation with large doses of B or R-light produced higher contents of cardenolides than irradiation with yellow-green light in somatic embryos of D. lanata 8 . In the present study, this is a crucial factor that favours cardenolides synthesis in D. purpurea, when grown in a PFS under R:B combination or B alone LEDs light. It is well known that a blue light photoreceptor and protochlorophllide-holochrome, or phytochrome, are involved in the regulation of cardenolides biosynthesis and accumulation 8 . On the other hand, it is a commonly agreed fact that light intensity can positively influence the accumulation of phytochemicals; accordingly, the influence of light quality are observed to be more complicated and usually reported with variable outcomes 11,64,65 . A previous study indicated that variations in the light spectrum could cause fluctuations in secondary metabolite content 66 . In our case, the light spectrum not only modulated the amount but also showed variations in cardenolides accumulations in Digitalis.
In conclusion, an important finding of the present study is that the combination of R:B (8:2) LEDs promoted plant growth, while micro-and macro-elements and cardenolides accumulation was enhanced under R:B (2:8) LEDs in the PFS system. These should be taken into consideration as a beneficial component and the plant extract may successfully cure various human diseases. In addition, the present study shows that it is possible to modify the growth parameters and secondary metabolites accumulation and their composition, in D. purpurea, by applying appropriate light quality in a PFS system. The current study also reveals that growing D. purpurea plants under-regulated environments could be regarded as an alternative approach to improve the production of biomass and secondary metabolites, which would aid in an extensive production of plant-based medicine. This research would be another step to provide high quality, fast-growing, uniform growth of plants, for pharmaceutical or alternative propagation purpose. Further studies are required in order to understand the exact role of light sources, with regard to particular glandular trichomes, the biochemical pathways of the compounds they produce and secrete; and thereby advance our understanding of the secondary metabolites in plants.