Introduction

Camellia japonica is a broad-leaved evergreen woody species, widely distributed in Japan and Korea. Camellia contains the antioxidants quercetin, quercetin-3-O-glucoside, quercitrin, and kaempferol (Nakajima et al. 1984) and camelliagenin A, B, and C (Itokawa et al. 1969). Previously reported, it contains saponins in seeds (Yoshikawa et al. 1994), flavonol glycosides in leaves and triterpenes, flavonols and phenolic compounds in flowers (Nakajima et al. 1984) depending on the parts of the camellia. These secondary metabolites enhance the expression and activity of cellular antioxidant enzymes in the human body, thus helping to eliminate reactive oxygen species (Piao et al. 2011).

The production of secondary metabolites from C. japonica has been constrained by the limited distribution of this plant, changes in quality due to climate change and contamination of soil, triggering the search for alternative methods for their production (Kirakosyan et al. 2000;Sirvent and Gibson 2002;Walker et al. 2002). In vitro culture was developed to provide a stable production of secondary metabolites (Karuppusamy 2009). Various culture systems have been implemented for in vitro culture, such as shoot, callus, adventitious roots, and hairy root cultures.

Among the various in vitro culture systems for biomass production, calli and adventitious roots are often used for industrial production as non-genetically modified materials. Many studies have corroborated the applicability of these two culture systems in the production of useful secondary metabolites (Bourgaud et al. 2001;Grąbkowska et al. 2016;Khan et al. 2018). Callus is a mass of dedifferentiated cells that can be induced from any plant part. I t is characterized by a fast growth rate, which renders it superior to other culture systems such as adventitious root, shoot, and somatic embryogenesis (Dörnenburg and Knorr 1995). However, prolonged callus subculture in some species results in a loss o f regenerative capacity and o ccurrence of s omaclonal variation (Larkin and Scowcroft 1981). Adventitious roots are composed of differentiated tissues and have more bioactive compounds and less genetic variation than calli, depending on the species (Murthy et al.2008). Production of bioactive compounds through callus culture has been studied in Cistanche deserticola (Cheng et al. 2005) and Stevia rebaudiana (Tadhani et al. 2007).

Despite the numerous studies on different culture system s, i t is d ifficult t o com pare t he e fficiency of e ach system for production of bioactive compounds because little research w as c onducted t o com pare c ultures derived f rom the same genotype. Previous studies on in vitro culture of camellias investigated shoot tip culture (Thomas et al. 2010;Wojtania et al. 2011), somatic embryogenesis from roots (Vieitez et al. 1991), plantlet production through cotyledon culture (Kato 1986), and shoot multiplication (Samartin 1989). These studies were focused on micropropagation of camellia plantlets, whereas only a few studies examined the production of secondary metabolites, especially phenolic compounds, from callus and adventitious root cultures.

In the present study, two culture systems, callus and adventitious root, were compared to identify the most suitable culture system for production of biomass and phenolic compounds from Camellia japonica. In addition, the relationship between the origin of explants and their cultures were investigated in both culture systems.

Materials and Methods

Plant materials

Calli and adventitious roots derived from the leaf, root, petal, and ovary of Camellia japonica were used. Calli were maintained on Murashige and Skoog (MS) medium (Murashige and Skoog 1962) supplemented with 26.85 μM 1-Naphthaleneacetic acid (NAA), 0.44 μM 6-Benzylaminopurine (BA), 100 mg･L^-1 casein hydrolysate, 100 mg･L^-1 citric acid, 100 mg･L^-1 L-ascorbic acid, 30 g･L^-1 sucrose, and 2.4 g･L^-1 gelrite, whereas adventitious roots were cultured on 3/4 MS medium supplemented with 24.6 μM Indole-3-butyric acid (IBA), 2.4 g･L^-1 gelrite, and 30 g･L^-1 sucrose. Both culture system s were i ncubated a t 24 ± 1°C in t he d ark. F ive masses of calli per Petri dish (250 mg each mass) and 10 segments of adventitious roots per Petri dish were prepared in triplicates. Subculturing was conducted every 4 weeks for calli and 8 weeks for adventitious roots. For growth assessment and high-performance liquid chromatography (HPLC) analysis, samples were harvested at the end of each subculture period.

Flow cytometry analysis

The samples (50 mg) were freshly harvested after 3 days of culture. Cell division was analyzed with CytoFLEX (Beckman Coulter, USA). The samples were collected from each culture for cell cycle analysis. Data were calculated using the CytExpert program (Beckman Coulter, USA) for DNA index and cell division. Data are described as the ratio of the cells belonging to the stage gap 1 (G1), DNA synthesis (S), a nd g ap 2 a nd m etaphase (G2 + M).

Analysis of phenolic compounds

Extraction

Freeze-dried calli and adventitious roots were ground in a sterilized m ortar and pestle. Powered, d ry c alli a nd adventitious roots (250 mg) were soaked in 20 mL 80% methanol at 24°C for 1 h, sonicated by refluxing (LS-2050-S10, LS-TECH, Korea), and centrifuged twice at 9,358 xg for 10 min. The collected supernatant was filtered through filter paper (Advantec 110 mm; Toyo Roshi Kaisha Ltd., Japan).

Determination of total phenolic compounds

The Folin-Ciocalteu colorimetric method was used to quantify total phenols (Folin-Ciocalteu 1927). The methanolic extracts (0.05 mL) were mixed with distilled water (2.55 mL), followed by the addition of 0.1 mL of 2 N Folin-iocalteu reagent. A fter 5 m in, the reagent was com bined with 2 .5 mL of 20% Na₂CO₃ solution and incubated in the dark at room temperature. The absorbance (change in color) at 760 nm after 30 min was measured using a spectrophotometer (Optizen POP; Mecasys Co., Korea). A standard curve obtained with gallic acid (Sigma Chemical Co., USA) was used for data comparison and total phenolic content was expressed as milligrams of gallic acid equivalent per gram of dry weight (DW).

Determination of total flavonoids content

The colorimetrical method was used to measure total flavonoid content (Wu et al. 2006). Methanolic extracts and (+)-catechin (Sigma Chemical Co., USA) standard were prepared in 1.25 mL distilled water. NaNO₂ (0.075 mL, 5%) was added to the solution, followed by vigorous shaking. After 6 min of reaction time, 0.15 mL of 10% AlCl3 solution was added to the sample, followed by incubation for 5 min at room temperature. The absorbance at 510 nm was measured using a spectrophotometer (Optizen POP; Mecasys Co., Korea). The results were expressed as milligrams of (+)-catechin equivalents per gram of DW.

HPLC analysis

Powder samples (0.25 g) of calli and adventitious roots were sonicated (Sonicator; Mujigae, Korea) for 1 h in 80% methanol to ensure complete extraction. The extract was filtered through filter paper (110 mm; Advantec, Japan), and the solvent was evaporated. The dried residue was dissolved in 10% methanol and fractionated twice with 10 mL diethyl ether/ethyl acetate (1:1) before vacuum evaporation to dryness. The residues of both fractions were combined and dissolved in methanol before filtration through a membrane filter (0.2 μm pore size; Whatman, England). A photodiode array (PDA)-equipped HPLC (2690 Separations Module; Waters Chromatography, USA) system was used to measure phenolic compounds. Separation was performed using a Fortis C18 column (5 μL, 150 × 4.6 m m). A cetonitrile ( A) a nd 0 .1% aqueous a cetic acid ( v/v) (B) were used as the mobile phase with linear gradients of 8 - 10% A at 0 - 2 min, 10 - 30% A at 2 - 27 min, 30 - 90% A at 27 - 50 min, 90 - 100% A at 50 - 51 min, 100% A at 51 - 60 min, and 100 - 8% A at 60 - 70 min. The column was re-equilibrated for 10 min between injections at a 1.0 mL min^-1 flow rate, and 20 μL aliquots were injected into the HPLC at a t ime. Calibration plots were obtained by measuring the peak areas. UV absorption spectra and retention time were used as criteria for the identification of individual compounds.

Analysis of free radical scavenging (DPPH activity)

Antioxidant capacity was measured using the 1, -diphenyl-2-picrylhydrazyl (DPPH; Sigma Chemical Co., USA) method (Hatano et al. 1998). DPPH radical solution (0.8 mL 200 μM) was added to 0.2 mL methanolic callus extract, while 40% methanol served as a control. The prepared solution was incubated for 5 min at room temperature, and the absorbance at 517 nm was measured using a spectrophotometer (Optizen POP, Mecasys Co., Korea). The results were presented as follows:

Statistical analysis

The results are presented as the mean values. Significant difference between the groups was evaluated using one-way analysis of variance (ANOVA). Statistical assessments of the differences between the mean values were performed using the Duncan's multiple range test. A p-value of 0.05 was considered to indicate statistical significance, and all the data were analyzed using the SAS program (SA 9.4, SAS Institute Inc., USA).

Results and Discussion

Biomass of different origin-derived call and adventitious roots

Growth characteristics of the callus and adventitious root cultures were investigated after 4 and 8 weeks of culture, respectively (Fig. 1A and Fig. 2). The highest fresh weight (FW 2.0 g･mass^-1) and DW (90.0 mg･mass^-1) of the callus derived from the ovary (CO) were 1.3-fold greater than those of the callus derived from petals (CP). CP was white and CL was yellow as determined by visual observations (Fig. 2). Dale and Deambrogio (1979) reported that the induction rate and the characteristics of callus derived from various explants differed between explants in Hordeum vulgare. This might be due to change in the morphological and biochemical characteristics of different explant type, which m ay a ffect c ytokinin u ptake a nd callus f ormation (Dhar and Joshi 2005).

Unlike calli, the FW of adventitious roots was the highest in those derived from leaves (AdL, 88.5 mg･ea^-1) and derived from roots (AdR, 82.3 mg･ea^-1) (Fig. 1C). Similarly, DW was h igher in AdL (15.7 mg･ea^-1) and AdR (12.6 mg･ ea^-1) than in other adventitious roots; DW of AdL was 1.8-fold higher than that of ovary-derived adventitious roots (AdO), which had the lowest DW. The number of secondary adventitious roots in AdL was the greatest (12.8 ea) and 7.5-fold larger than that of AdO. In AdO, the number of secondary adventitious roots was small, but one of the secondary adventitious roots was the longest (2.0 cm in length) (data not shown).

These results indicate that the characteristics of adventitious roots differ according to the origin of the explants used for their culture. Concentration of endogenous hormone and sensitivity to hormone are different in different plant parts (Davies 2013). Zhang et al. (2011) also reported that root-derived adventitious roots had a higher growth than leaf-derived adventitious roots, the reason for which might be the different effect of IBA on different explants. Therefore, even if the same plant growth regulators were used, it is considered that different responses are due to different origin of the explants or different species.

Cell division in calli and adventitious roots originating from four different explants

In both the calli and adventitious roots, the highest proportion of G2/M cells was detected in petal-derived calli (CP) and adventitious roots (AdP), indicating an active cell division (Fig. 3A and B). In present experiment, cell division was most active in CP, but biomass was the highest in CO (Fig.1A and 3A). The correlation between cell division and biomass increase has been reported previously. For example, Gonzalez et al. (2012) stated that the increase in leaf size results from active division of many cells. However, Harashima and Schnittger (2010) reported different results about the cell division and biomass increase: cell proliferation could be stimulated by A- or D-type cyclins or E2F and DP transcription factors overexpressed or the RBR pathway is inactivated. In all these cases, the size of the organ significantly decreases, indicating that active cell division does not increase growth.

DNA content of calli originating from four different explants

The flow cytom etry a nalysis of c alli b y origin s howed that the mean value of the first peak corresponding to the G1 phase in CP, CL, CR, and CO (Table 1). Long-term subculture increases the possibility of genetic variation (Anandan et al. 2018), which can be attributed to the physiological conditions of the explant as determined by genetic factors (Dhar and Joshi 2005).

In the p resent e xperim ent, t he DNA c ontent o f all calli (except CP) was about 2 times higher than that of CP, and the increase in DNA content was considered to be somaclonal variation (Table 1). Therefore, when callus derives from a petal, it is considered that the culture can be continued for an extended period because the cell division is stable. However, there was no change in DNA content of ARs according to origin explants (data not shown).

Phenolic compounds of calli and adventitious roots derived from four different explants

The total phenolic content was the highest in CP (7.67 mg･g^-1 DW), followed by that in CO (6.40 mg･g^-1 DW), and there was no significant difference in their content between CL and CR. The contents of flavonoids were high in CO (3.16 mg･g^-1 DW) and CP (2.77 mg･g^-1 DW) (Fig. 4A). Therefore, the content of phenols and flavonoids in calli derived from four explants seems to have varied.

The amount of accumulated secondary plant metabolites differs between different plant parts (Izhaki 2002). Thus, the content of total phenols and flavonoids varies in different parts of ginger plants (Lee et al. 2014), whereas the content of stilbene in callus changes depending on the type of explants in grape, suggesting that the synthesis potential is retained to some extent in grape cells of different origin (Liu et al. 2010).

The HPLC analysis of individual phenol substances (Fig. 5) revealed that homogentisic acid was characteristically analyzed in CO, and the content of caffeic acid was higher in CO than in other callus lines. Furthermore, homogentisic acid was analyzed only in the AdO. In adventitious roots, the level of quercetin was greater compared with that in calli in all cell lines. Both calli and adventitious roots showed higher phenolic content in a line derived from the ovary than in other lines.

The phenol content in adventitious root lines was higher than that in callus lines. Especially, AdO (1,012.6 μg･g^-1 DW) had 1.13-fold higher level of phenols than did CO (899.8 μg･g^-1 DW) (Fig. 5).

Secondary metabolites are closely related to the morphological and cellular status of the cells-high content of phenols is found in highly differentiated plant parts (Rekha and Rakhi 2013). However, in this experiment, morphological changes were not founded in ARs and calli. Phenolics content in adventitious roots are higher than that of calli, because callus is a dedifferentiated cell, but adventitious roots are composed of differentiated cells (Dörnenburg and Knorr 1995;Murthy et al. 2008).

The phenolics content per unit gram was higher in adventitious roots, although the production per 1 L of culture medium was 6.0-fold greater in calli (33.26 mg･g^-1) than in adventitious roots (5.58 mg･g^-1) considering culture period and biomass (Fig. 6). In both the calli and adventitious roots, the total phenolics content per gram of grams was significantly different in all lines, but there was no significant difference in productivity between any of the lines, except in CO and AdL. The results of the present experiment reveal that calli are a more suitable culture system than adventitious roots for scaled-up production of phenolic compounds.

Antioxidant activity

DPPH radical scavenging activity showed the highest value of 88.6% in CO (Fig. 4B), followed by that in CP (69.7%), CR (33.7%), and CL (22.9%). Phenolic acid and flavonoids have antioxidant properties, and the higher the content of these substances, the greater the antioxidant activity (Shahidi et al. 1992;Rice-Evans et al. 1996). Therefore, the high antioxidant activity detected in CP and CO was related to the high content of total phenols and flavonoids in the two cell lines.

Conclusion

In this experiment, we compared the culture systems of callus and adventitious root in terms of biomass and phenolic compounds. It revealed that more phenolics could be obtained from calli in terms of productivity (per liter) than t hat of A Rs. When c om pared by o rigin, i t was analyzed that the phenolic content was the highest in callus and adventitious root derived from the ovary. The results abstained here showed that the calli was suitable culture system for the production of phenolics while ARs contained various phenolics, and it can be used as one of the antioxidant cosmetic substances.