Differential Regulation of Anthocyanin Synthesis in Apple Peel under Different Sunlight Intensities

Sunlight radiation is a main environmental factor which affects anthocyanin synthesis. To clarify the regulatory mechanism of sunlight on the synthesis of anthocyanin in apple peel, bagged apples were exposed to diverse intensities of sunlight through different shading treatments. Under an increased solar ultraviolet-B (UV-B) light intensity, the concentration of anthocyanin in apple peels was consistent with the Michaelis–Menten equation. Under lower sunlight intensities, diphenyleneiodonium chloride (DPI, an inhibitor of plasma membrane NAD(P)H oxidase) treatment increased both the concentration of cyanidin-3-glycoside and the activity of dihydroflavonol 4-reductase (DFR). However, under higher sunlight intensities, DPI treatment decreased the concentrations of cyanidin-3-glycoside and quercetin-3-glycoside, as well as the activities of DFR and UDP-glycose: flavonoid 3-O-glycosyltransferase (UFGT). These results indicate that, under low sunlight intensity, anthocyanin synthesis in apple peel was limited by the supply of the substrate cyanidin, which was regulated by the DFR activity. Nevertheless, after exposure to high sunlight intensity, the anthocyanin produced in the apple peel was dependent on UFGT activity.


Introduction
Apples are an important agricultural crop and anthocyanins are the visible sign of apple maturity [1]. Anthocyanins belong to the reddish secondary metabolites, which consist of anthocyanidin and glycoside. They can increase the commerciality of various health foods and extend their shelf life [2,3]. Anthocyanins act as visual signals to attract insects for pollination and agents for seed dispersal [4], and also function as anti-oxidants, scavenging free radicals and protecting plant tissues against both biotic and abiotic stresses [5]. These compounds also have a positive effect on human health, serving as vasodilators and reducing the risk of myocardial infarction [6,7]. Anthocyanins have potential anti-diabetic properties; in addition, anthocyanin-rich foods can reduce starch consumption and delay glucose absorption by inhibiting alpha-amylase and alpha-glucosidase [8]. Anthocyanins can also benefit our bodies in terms of neuroprotection, vision improvement, anti-inflammatory effects, antimicrobial activity, chemoprevention, and cancer protection [9][10][11].
The synthesis of anthocyanins is involved in the phenylpropanoid metabolic pathway. Recently, the regulatory mechanisms of anthocyanin synthesis were reported [12][13][14][15][16][17][18]. Sunlight is one of the environmental factors regulating gene expression and plant development, and ultraviolet-B (UV-B) is considered to be a major factor to increase the synthesis of anthocyanins and flavonoids in plants [12,15]. UV-B induced photomorphogenesis is initiated by the specific photoreceptor, UV RESISTANCE LOCUS 8 (UVR8), which utilizes its tryptophan residue as an internal chromophore to sense UV-B [18]. UVR8 regulates the light signaling factors E3 ubiquitin ligase CONSTITUTIVELY PHOTOMORPHOGENIC1

Anthocyanin Concentrations under Different Sunlight Intensities Fit with Michaelis-Menten Equation
Through different shading treatments, the apple fruits were exposed to different intensities of sunlight. With increased solar UV-B light intensity, the concentration of total anthocyanins (cyanidin-3-galacotoside plus cyanidin-3-glucoside) in both 'Fuji' and 'Red Delicious' apple peels initially increased in a linear pattern, and then gradually leveled off in either 2016 or 2017 ( Figure 1A  . , R 2 = 0.992 (C), respectively.

Analysis of Anthocyanin and Flavonol Concentrations under Different Sunlight Intensities
The concentrations of flavonol, quercetin-3-galactoside, and quercetin-3-glucoside changed in similar patterns to cyanidin-3-glycoside in 'Fuji' apple peel as the solar UV-B light intensity increased ( Figure 2C,D). When the light intensity was below 50%, DPI treatment increased the concentration of cyanidin-3-glycoside, but did not change the concentration of quercetin-3-glycoside in 'Fuji' apple peel. When the solar light intensity was over 50%, DPI treatment decreased the concentrations of both cyanidin-3-glycoside and quercetin-3-glycoside.

Analysis of Anthocyanin and Flavonol Concentrations under Different Sunlight Intensities
The concentrations of flavonol, quercetin-3-galactoside, and quercetin-3-glucoside changed in similar patterns to cyanidin-3-glycoside in 'Fuji' apple peel as the solar UV-B light intensity increased ( Figure 2C,D). When the light intensity was below 50%, DPI treatment increased the concentration of cyanidin-3-glycoside, but did not change the concentration of quercetin-3-glycoside in 'Fuji' apple peel. When the solar light intensity was over 50%, DPI treatment decreased the concentrations of both cyanidin-3-glycoside and quercetin-3-glycoside. Figure 2. Concentrations of cyanidin-3-galactoside (A), cyanidin-3-glucoside (B), quercetin-3galactoside (C), and quercetin-3-glucoside (D) in 'Fuji' apple peels after exposing bagged fruits, with or without diphenyleneiodonium chloride (DPI) treatment, to diverse sunlight intensities. Each data point represents mean ± SE (n = 5). The asterisk indicates a significant difference between DPI treatment and no DPI treatment at p < 0.05 (t-test).

Analysis of Nucleotide Sugar Concentrations under Different Sunlight Intensities
As nucleotide sugar provides glycosyl group for anthocyanin synthesis, the concentration of UDP-galactose was also assayed. The concentration of UDP-galactose in 'Fuji' apple peel declined and then remained at a stable level along with the increasing of solar UV-B light intensity ( Figure 3). Treatment with DPI did not change the concentration of UDP-galactose. Concentrations of UDP-galactose in 'Fuji' apple peels after exposing bagged fruits, with or without DPI treatment, to diverse sunlight intensities. Each data point represents mean ± SE (n = 5).

Analysis of Nucleotide Sugar Concentrations under Different Sunlight Intensities
As nucleotide sugar provides glycosyl group for anthocyanin synthesis, the concentration of UDP-galactose was also assayed. The concentration of UDP-galactose in 'Fuji' apple peel declined and then remained at a stable level along with the increasing of solar UV-B light intensity ( Figure 3). Treatment with DPI did not change the concentration of UDP-galactose. Fuji' apple peels after exposing bagged fruits, with or without diphenyleneiodonium chloride (DPI) treatment, to diverse sunlight intensities. Each data point represents mean ± SE (n = 5). The asterisk indicates a significant difference between DPI treatment and no DPI treatment at p < 0.05 (t-test).

Analysis of Nucleotide Sugar Concentrations under Different Sunlight Intensities
As nucleotide sugar provides glycosyl group for anthocyanin synthesis, the concentration of UDP-galactose was also assayed. The concentration of UDP-galactose in 'Fuji' apple peel declined and then remained at a stable level along with the increasing of solar UV-B light intensity ( Figure 3). Treatment with DPI did not change the concentration of UDP-galactose. Concentrations of UDP-galactose in 'Fuji' apple peels after exposing bagged fruits, with or without DPI treatment, to diverse sunlight intensities. Each data point represents mean ± SE (n = 5).

Analyze Gene Expression Levels under Different Sunlight Intensities
Transcription levels of MdUVR8, MdHY5, MdCOP1, MdMYB10, MdCHS, MdCHI, MdDFR, MdANS, and MdUFGT in 'Fuji' apple peel initially increased and then remained unchanged with Concentrations of UDP-galactose in 'Fuji' apple peels after exposing bagged fruits, with or without DPI treatment, to diverse sunlight intensities. Each data point represents mean ± SE (n = 5).

Analyze Gene Expression Levels under Different Sunlight Intensities
Transcription levels of MdUVR8, MdHY5, MdCOP1, MdMYB10, MdCHS, MdCHI, MdDFR, MdANS, and MdUFGT in 'Fuji' apple peel initially increased and then remained unchanged with increasing solar UV-B light intensity, except for MdDFR and MdUFGT, which slightly decreased under higher light intensities ( Figure 4A-H and Figure S1). DPI treatment did not change the gene levels, regardless of sunlight intensity, except for MdHY5 and MdCOP1, which increased upon full exposure. increasing solar UV-B light intensity, except for MdDFR and MdUFGT, which slightly decreased under higher light intensities ( Figure 4A-H and Figure S1). DPI treatment did not change the gene levels, regardless of sunlight intensity, except for MdHY5 and MdCOP1, which increased upon full exposure. The asterisk indicates a significant difference between DPI treatment and no DPI treatment at p < 0.05 (t-test). The asterisk indicates a significant difference between DPI treatment and no DPI treatment at p < 0.05 (t-test).

Analysis of Enzyme Activities under Different Sunlight Intensities
The activities of CHS, DFR, and ANS in 'Fuji' apple peel continuously increased in accordance with increasing solar UV-B light intensity, but the CHI activity remained unchanged ( Figure 5A-D). The UFGT activity increased linearly and remained steady with increasing intensity, but then slightly declined upon full exposure ( Figure 5E). When the solar UV-B light intensity was below 50%, DPI treatment significantly increased DFR activity. When the solar UV-B light intensity was over 50%, DPI treatment decreased the activities of both DFR and UFGT. DPI treatment did not affect the activities of CHS, CHI, or ANS in 'Fuji' apple peel. The activities of CHS, DFR, and ANS in 'Fuji' apple peel continuously increased in accordance with increasing solar UV-B light intensity, but the CHI activity remained unchanged ( Figure 5A-D). The UFGT activity increased linearly and remained steady with increasing intensity, but then slightly declined upon full exposure ( Figure 5E). When the solar UV-B light intensity was below 50%, DPI treatment significantly increased DFR activity. When the solar UV-B light intensity was over 50%, DPI treatment decreased the activities of both DFR and UFGT. DPI treatment did not affect the activities of CHS, CHI, or ANS in 'Fuji' apple peel. Figure 5. Activities of chalcone synthase (CHS, (A)), chalcone isomerase (CHI, (B)), dihydroflavonol 4-reductase (DFR, (C)), anthocyanidin synthase (ANS, (D)), and UDP-glycose: flavonoid 3-Oglycosyltransferase (UFGT, (E)) in 'Fuji' apple peels after exposing bagged fruits, with or without DPI treatment, to diverse sunlight intensities. Each data point represents mean ± SE (n = 5). The asterisk indicates a significant difference between DPI treatment and no DPI treatment at p < 0.05 (t-test). in 'Fuji' apple peels after exposing bagged fruits, with or without DPI treatment, to diverse sunlight intensities. Each data point represents mean ± SE (n = 5). The asterisk indicates a significant difference between DPI treatment and no DPI treatment at p < 0.05 (t-test).

Discussion
In apple peel, the predominant anthocyanin component is cyanidin-3-galactoside [29,38,39]. The synthesis of cyanidin-3-galactoside was catalyzed by UFGT, namely, cyanidin + UDP-galactose where V is the rate of cyanidin-3-galactoside synthesis, and t is time. According to the Michaelis-Menten equation, In this study, apples were exposed to different sunlight intensities for the same period of time. Thus, A at different light intensities was determined by V. Under low The produced cyanidin-3-galactoside depends on the concentration of the substrate [S]. Under high light intensity, Vo = Vmax, namely, A = Vmax × t, and the produced cyanidin-3-galactoside depends on UFGT activity. Clearly, according to the Michaelis-Menten equation, the regulatory mechanism of anthocyanin synthesis in apple peel was different upon the exposure of bagged apples to diverse intensities of solar UV-B. When the solar UV-B light intensity was relatively low (below 50%), anthocyanin synthesis was limited by the supply of the substrate, whereas it was limited by UFGT activity when the solar UV-B light intensity was relatively high (over 50%, Figures 1 and 5).
In our previous studies, after exposing bagged 'Golden Delicious' apples to sunlight, DPI (an inhibitor of plasma membrane oxidase) treatment inhibited the production of reactive oxygen species (ROS) via plasma membrane NADPH oxidase and the signaling pathway, reducing the expressions of MdMYB10 and MdUFGT, the enzyme activity of UFGT, and the synthesis of anthocyanin in apple peel [12]. In this study, DPI treatment also inhibited anthocyanin synthesis in 'Fuji' apple peel when the bagged fruits were exposed to relatively high sunlight intensities ( Figure 2). However, DPI did not affect the expression levels of MdMYB10 and MdUFGT, but inhibited the enzyme activities of DFR and UFGT (Figures 4 and 5). The different gene expression and enzyme activity responses might be related to the different cultivar properties, as 'Golden Delicious' is a non-red cultivar and 'Fuji' is a red cultivar. UFGT catalyzes the glycosylation of both cyanidin and quercetin in apple peel [27,40], while DFR is only involved in the synthesis of cyanidin [30][31][32]. Therefore, under relatively high sunlight conditions, the lower concentrations of quercetin-3-glycoside in DPI-treated 'Fuji' peels suggest that DPI may inhibit the synthesis of anthocyanin mainly by affecting the activity of UFGT ( Figure 2). This is also consistent with the conclusion derived from the Michaelis-Menten equation, as mentioned above.
Interestingly, DPI treatment significantly increased the concentration of anthocyanin in 'Fuji' fruit peels after exposing the bagged fruits to relatively low sunlight conditions ( Figure 2). This indicates that DPI treatment might have different effects on the regulation of anthocyanin synthesis in apple peel under diverse sunlight intensities. Indeed, DPI treatment did not inhibit the enzyme activity of UFGT, but increased that of DFR under relatively low sunlight conditions ( Figure 5). DFR and ANS catalyze the synthesis of cyanidin, so the relatively higher DFR activity may provide more cyanidin for anthocyanin synthesis. According to the Michaelis-Menten equation, the synthesis of anthocyanin is limited by the substrate concentration under relatively low sunlight conditions (Figure 1). Since DPI treatment did not affect the concentration of UDP-Gal and the expression level and enzyme activity of ANS in apple peel (Figure 3, Figure 4G and Figure 5D), the supply of cyanidin catalyzed by DFR should be the limitation of anthocyanin synthesis under conditions of relatively low sunlight. The concentration of quercetin-3-glycoside was not affected by DPI treatment in relatively low sunlight conditions, indicating that other genes (enzymes) in the pathway shared by both anthocyanin and quercetin synthesis were not affected by DPI treatment. It is unclear why DPI treatment improved DFR activity under relatively low sunlight conditions. Wargent and Jordan (2013) suggested that anthocyanin synthesis is regulated through different signaling pathways under different UV-B irradiation conditions [37]. As anthocyanin was more sensitive to ROS than other flavonoid compounds in apple peel [41], it is perhaps that the more accumulated anthocyanins in DPI treated fruit peels were attributed to fewer ROS produced via plasma membrane NADPH oxidase. However, at first, under relatively low sunlight conditions, the produced ROS was also at low levels. Secondly, although ROS such as hydrogen peroxide may go through the cellular membrane, the enzymatic and non-enzymatic antioxidant systems in cytosol would be the first defense line against ROS prior to diffusion from the extracellular site of genesis to the vacuole where the anthocyanins are mainly located. In our previous studies, even at high light conditions, anthocyanin was not involved in the detoxification of ROS in pear peel [42].
In conclusion, the regulatory mechanisms of anthocyanin synthesis in apple peel were different under diverse sunlight intensities. This might be the reason that previous studies have shown different limiting factors for anthocyanin synthesis in apple peel [12,20,34]. Under relatively low sunlight intensities, anthocyanin synthesis in apple peel was limited by the supply of the substrate cyanidin, which was regulated by the activity of DFR. However, under relatively high sunlight intensities, the produced anthocyanin is dependent on UFGT activity ( Figure 6). These results might be useful for the development of new biotechnological strategies for improving the quality and market value of apples. enzymatic and non-enzymatic antioxidant systems in cytosol would be the first defense line against ROS prior to diffusion from the extracellular site of genesis to the vacuole where the anthocyanins are mainly located. In our previous studies, even at high light conditions, anthocyanin was not involved in the detoxification of ROS in pear peel [42].
In conclusion, the regulatory mechanisms of anthocyanin synthesis in apple peel were different under diverse sunlight intensities. This might be the reason that previous studies have shown different limiting factors for anthocyanin synthesis in apple peel [12,20,34]. Under relatively low sunlight intensities, anthocyanin synthesis in apple peel was limited by the supply of the substrate cyanidin, which was regulated by the activity of DFR. However, under relatively high sunlight intensities, the produced anthocyanin is dependent on UFGT activity ( Figure 6). These results might be useful for the development of new biotechnological strategies for improving the quality and market value of apples.

Plant Materials
Two apple cultivars (Malus domestica Borkh.), 'Fuji' and 'Red Delicious', were used in this study. 'Fuji' trees were field-grown at a spacing of 2.5 m × 3. (approximately 144 days after full bloom). Five replicates (ten fruits for each replicate) of the bagged fruits were sampled (three trees per replicate, 15 trees total) without removing the bags, to avoid exposure to light before chemical and sunlight exposure treatments.

Chemical and Diverse Intensities Solar UV-B Light Exposure Treatments
An external solar electric quantum meter (Spectrum Technologies Company, Chicago, Illinois, USA) and an ultraviolet radiometer (Instrument Factory of Beijing Normal University, Beijing, China) were used to measure visible and ultraviolet light transmittance, respectively. The data were collected at 07:00, 09:00, 11:00, 13:00, 15:00, 17:00, and 19:00 h in an open place on a sunny day. The strongest solar light intensity, recorded at 13:00, was 1800 ± 50 μmolm −2 s −1 photon flux density and set as 100%

Plant Materials
Two apple cultivars (Malus domestica Borkh.), 'Fuji' and 'Red Delicious', were used in this study. 'Fuji' trees were field-grown at a spacing of 2.5 m × 3.5 m in Qianxian (34. . Five replicates (ten fruits for each replicate) of the bagged fruits were sampled (three trees per replicate, 15 trees total) without removing the bags, to avoid exposure to light before chemical and sunlight exposure treatments.

Chemical and Diverse Intensities Solar UV-B Light Exposure Treatments
An external solar electric quantum meter (Spectrum Technologies Company, Chicago, Illinois, USA) and an ultraviolet radiometer (Instrument Factory of Beijing Normal University, Beijing, China) were used to measure visible and ultraviolet light transmittance, respectively. The data were collected at 07:00, 09:00, 11:00, 13:00, 15:00, 17:00, and 19:00 h in an open place on a sunny day. The strongest solar light intensity, recorded at 13:00, was 1800 ± 50 µmolm −2 s −1 photon flux density and set as 100% (Figure 7). The air temperature and humidity were 28 ± 1 • C and 45%, respectively, at midday during the treatments.  Figure 7). The air temperature and humidity were 28 ± 1 °C and 45%, respectively, at midday during the treatments. Apple peel samples were collected after one week of irradiation. The peels were immediately frozen in liquid nitrogen, then ground to powder and mixed in liquid nitrogen circumstance with an A11 grinder from IKA ® Works (VWR, Radnor, PA, USA), and finally stored at −80 °C until further analysis.

Analysis of Flavonoid Compounds
The extraction and analysis of flavonoid compounds were carried out as described by Li et al. [29]. Briefly, the frozen tissue powder (0.5 g) was ground in 1.5 mL phenolic compound, extracting a solution containing 70% methanol and 2% formic acid at 0-4 °C. After centrifugation at 10,000 g for 20 min, the supernatant was passed through a 0.22-μm syringe filter prior to analysis.

Analysis of Nucleotide Sugar
The extraction and analysis of nucleotide sugar were carried out as described by Li et al. [29]. Briefly, frozen tissues (0.5 g) were extracted with 1.2 mL of 6% HClO4 and 5% insoluble polyvinylpolypyrrolidone (PVPP) at 0-4 °C. After centrifugation at 12,000× g for 10 min, 0.8 mL of the supernatant were transferred to another Eppendorf tube and then neutralized with 55 μL of 5 M K2CO3. The resulting potassium chlorate was removed by 5 min centrifugation at 12,000× g. The supernatant was used to measure metabolites. Apple peel samples were collected after one week of irradiation. The peels were immediately frozen in liquid nitrogen, then ground to powder and mixed in liquid nitrogen circumstance with an A11 grinder from IKA ® Works (VWR, Radnor, PA, USA), and finally stored at −80 • C until further analysis.

Analysis of Flavonoid Compounds
The extraction and analysis of flavonoid compounds were carried out as described by Li et al. [29]. Briefly, the frozen tissue powder (0.5 g) was ground in 1.5 mL phenolic compound, extracting a solution containing 70% methanol and 2% formic acid at 0-4 • C. After centrifugation at 10,000 g for 20 min, the supernatant was passed through a 0.22-µm syringe filter prior to analysis.

Analysis of Nucleotide Sugar
The extraction and analysis of nucleotide sugar were carried out as described by Li et al. [29]. Briefly, frozen tissues (0.5 g) were extracted with 1.2 mL of 6% HClO 4 and 5% insoluble polyvinylpolypyrrolidone (PVPP) at 0-4 • C. After centrifugation at 12,000× g for 10 min, 0.8 mL of the supernatant were transferred to another Eppendorf tube and then neutralized with 55 µL of 5 M K 2 CO 3 .
The resulting potassium chlorate was removed by 5 min centrifugation at 12,000× g. The supernatant was used to measure metabolites.

Quantitative Real-Time Polymerase Chain Reaction (qRT-PCR) Expression Analysis
Total RNA was isolated using the SDS-phenol method according to Malnoy et al. [43]. First-strand cDNA was synthesized using the PrimeScript TM RT reagent Kit (Takara, Dalian, China), according to the manufacturer's protocol. All qRT-PCR experiments were performed with the Bio-Rad CFX96 system (Bio-Rad Laboratories, Hercules, CA, USA) with 0.1-mL 8-tube strips, using SYBR Premix Ex TaqTM II (Takara, Dalian, China). MdActin was used as the internal reference gene. The PCR amplification program was 95 • C for 3 min, 39 cycles of 95 • C for 10 s, and 57 • C for 30 s, followed by a melting curve analysis program. The primers for MdActin, MdUVR8, MdHY5, MdCOP1, MdMYB10, MdCHS, MdCHI, MdDFR, MdANS, and MdUFGT are shown in Table S1.

Conflicts of Interest:
The authors declare that there is no conflict of interest.