Increasing the functional quality of Crocus sativus L. by-product (tepals) by controlling spectral composition

Crocus sativus L. is a crop grown for spice production, and large amounts of residues from the flowers are produced during the process. The underutilized by-product from saffron spice production, the C. sativus tepals, was investigated as a promising raw material of natural bioactive compounds using light spectrum manipulation in controlled environments. The plants were grown under either light-emitting diodes (LEDs) or natural light (NL, greenhouse). LED experiments were performed in controlled-environment chambers (120 µmol m–2 s–1of photosynthetically active radiation, 18 °C, 16-h photoperiod). The LED treatments used were as follows: (i) red ʎ = 660 nm (62%) and blue ʎ = 450 nm (38%) (RB); and (ii) red ʎ = 660 nm (50%), green ʎ = 500–600 nm (12%), and blue ʎ = 4 50 nm (38%) (RGB). Flower growth parameters, total phenols, total flavonoids, flavonols, flavonol glycosides, and antioxidant properties were measured in harvested tepals. Floral by-products from plants grown under the two LED treatments accumulated higher amounts of antioxidant compounds compared to those of plants grown under NL. The total flavonoids content was significantly enhanced in the RGB LED treatment, while the corolla fresh weight significantly declined in the same treatments. The higher content of bioactive secondary metabolites in plants grown under both RB and RGB light environments resulted in increased antioxidant capacity measured by DPPH free-radical scavenging capacity and the ferric reducing antioxidant power method. These results indicate that manipulation of LED spectra could boost secondary metabolites and antioxidant capacity to obtain phytochemically enriched floral by-products with superior functional quality.


Introduction
Agriculture and agro-industrial processing are responsible for producing massive amounts of agri-food wastes and byproducts. Disposal of agricultural wastes causes significant economic and environmental problems by contributing to greenhouse gas emissions (Girotto et al. 2015). However, many of these residues are generally considered sources of valuable bioactive compounds (such as micro-and macronutrients, dietary fibers, lipids, starch, proteins, vitamins, and secondary metabolites with antioxidant biological functions) that could find applications in the food, cosmetics, and pharmaceutical industries due to their antioxidant, antimicrobial, and health-promoting properties (Varzakas et al. 2016).
Therefore, the recovery and valorization of plant residues (waste) is a promising strategy to minimize the ecological impact through circular economy models and to create sustainable products.
C. sativus L. flowers are grown worldwide to produce the valuable saffron spice. Commercial saffron is made from the dried stigmas of these flowers, and it is considered to be the most expensive food spice in the world due to the meticulous manual operations needed for its production and the high labor cost (Giupponi et al. 2019). The flowers of C. sativus are collected early in the morning to avoid color and quality loss of stigmas, and then stigmas are separated from Communicated by Sanghyun Lee. the rest of the flowers (Giupponi et al. 2019). Moreover, this work is conducted only a few days a year and for only a few hours a day, and all of the other activities (field preparation, corms planting, weeding, etc.) are performed manually (Husaini et al. 2010). The principal producers and exporting countries (Iran, Afghanistan, and India) are located in Asia (Jouki et al. 2013a;OEC 2019). In recent years demand for this spice has increased due to Asian population growth (Arslanalp et al. 2019) and to the popularity of Asian cooking worldwide (Giupponi et al. 2019).
The phytochemical composition of saffron spice gives aroma, color, and flavor to food. Crocetin esters are responsible for saffron coloring, generating a range of yellowish-red hues (Moratalla-Lopez et al. 2019). Picrocrocin is the foremost compound responsible for the bitter taste, and safranal is the most important component of the volatile fraction contributing to saffron's distinctive aroma (Carmona et al. 2006). To avoid microbial growth in saffron stigmas, some studies used γ-irradiation and modified atmosphere packaging to prolong the shelf life without any significant quality deterioration (Jouki et al. 2011, Jouki andKhazaei, 2013b).
Saffron is not only a culinary condiment, but its flowers are also a rich source of health-promoting phytochemicals, including phenols, anthocyanins, flavonoids, and carotenoids. The stigmas have been reported to have high levels of carotenoids, and the petals are a rich source of flavonoids and anthocyanins (Hosseini et al. 2018, Jadouali et al. 2017. The positive effects on human health of these phytochemicals have been widely documented, including the antioxidant, anti-inflammatory, anti-depressant, and anti-carcinogenic benefits (Ahmad et al. 2005;Hosseinzadeh et al. 2007;Talaei et al. 2015;Bathaie et al. 2013;Kyriakoudi et al. 2015). The antioxidant activity is mainly linked to polyphenol compounds, such as kaempferol and quercetin (flavonoids) (Colombo et al. 2019;Riahi-Chebbi et al. 2019). Serrano-Díaz et al. (2014) demonstrated the absence of cytotoxicity in an aqueous extract, showing that this extract can be safely added to foods. Moreover, tepal extract has been reported to be a natural source of antioxidants and used as an ingredient to develop high-quality cosmetic products, since it plays a role in delaying the skin aging process (Acero de Mesa et al. 2018).
The traditional method of saffron production employs manual flower harvesting and stigma separation. After separating stigmas from flowers, the other flower components, including tepals, stamens, and styles, are dismantled. To produce 1 kg of saffron, dried pistils of approximately 200,000 flowers are needed with 63 kg of floral bio-residues (Serrano-Diaz 2013). The tepals account for a majority of the total weight of the whole saffron flower at 78.4%, followed by stamens at 13.4%, stigmas at 7.4%, and styles at 0.7% (Moratalla-Lopez et al. 2019). Previous studies reported that large quantities of saffron floral residues are not used and thrown away after harvesting, and these byproducts account for about 86 to 93% of the mass of saffron flowers (Moratalla-Lopez et al. 2019). In this context, the range of beneficial properties of C. sativus tepals make this by-product a sustainable source of promising phytochemicals (mainly polyphenols) to be used in food, nutraceutical, and cosmetic industries (Trivellini et al. 2016).
The content of polyphenols, in particular flavonoids, may be related to the response of plants to environmental conditions, such as light conditions (Tattini et al. 2005;Agati et al. 2011). The spectrum and direction of light-emitting diodes (LEDs) can be modulated to control light intensity and decouple lighting from heating (Singh et al. 2015). This allows not only control of the growth, development, and yield of plants, but also improvement of overall quality and energy use. The phytochemical content can be effectively modulated by controlling the spectral composition of LEDs to promote the most suitable composition of plant tissue for nutraceuticals. Several studies, recently reported and reviewed by Paradiso and Proietti (2021), investigated the influence of light-quality treatments, using red (R), blue (B), green (G) LEDs and their ratios, on secondary metabolism to regulate the amount of functional metabolites of several horticultural crops. In general, in response to select spectral bandwidths (R, B, G, RB, and RGB), functional quality was improved by promoting the accumulation of antioxidant compounds (i.e., phenolic acids, flavonoids, carotenoids, and antioxidant capacity).
In this study, the underutilized by-product from saffron spice production, the C. sativus tepals, was investigated as a promising raw material of natural bioactive compounds with possible applications in the food and cosmeceutical industries. C. sativus plants were grown in a growth chamber to control flower organ development and secondary metabolite accumulation using light spectrum manipulation to obtain a standard production of high-value bioactive compounds from floral residues (Fig. 1).

Plant materials
Saffron (Crocus sativus L.) bulbs were purchased from Floriana Bulbose (Monte Porzio Catone, Roma, Italy). The bulbs had a diameter of 3 cm and before transplanting were stored in a refrigerator at 4 °C with 80% humidity for 50 days to promote a good vegetative-flowering restart.

Plant growth conditions
Saffron bulbs were transplanted in plastic perforated boxes (50 cm × 30 cm × 25 cm) covered by non-woven 1 3 polypropylene fabric to control irrigation runoff. The growth medium used was peat-perlite (70:30 v/v) with a bed of pumice (5 cm high) to improve water drainage as saffron prefers soils with a high stone-skeleton. For each box, 20 bulbs were planted in five rows with four bulbs in each row. The transplanting took place on October 31, 2019 at the Department of Agriculture, Food and Environment of the University of Pisa, Italy (43°70′ N 10°43′ E). The plants were cultivated in a growth chamber at constant temperature of 18 °C and photoperiod of 16:8 h. Two different LED lamps were tested: (i) red ʎ = 660 nm (62%) and blue ʎ = 450 nm (38%) (RB); and (ii) red ʎ = 660 nm (50%), -green ʎ = 500-600 nm (12%), and blue ʎ = 450 nm (38%) (RGB). The lighting area was 0.5 m 2 (1.0 m × 0.5 m). The plant photosynthetic quantum flux density, measured 10 cm above the plants, was 120 µmol m −2 s −1 . The control plants were placed in plastic perforated boxes (50 cm × 30 cm × 25 cm) covered by non-woven polypropylene fabric and grown in a glass greenhouse under natural light (NL). The growing conditions were as follows: 18 °C average temperature, 70-80% humidity, approximately 15:9 (light/dark)-h photoperiod, and 70-120 µmol m −2 s −1 light intensity. In both growing conditions, surface irrigation was employed by applying 3000 g of water per box.

Saffron flower growth parameters
C. sativus flowers were collected the day after anthesis (to evaluate the metabolic response to the corolla's LED radiation), and the corolla was frozen in liquid nitrogen and stored at − 80 °C until biochemical analysis. The influence of the different combinations of light treatments were monitored on C. sativus harvested flowers by measuring the length (cm) and weight (g) of the (i) whole flower, (ii) pistil, and (iii) corolla.

Bioactive molecule extraction from tepals
Acidified 80% methanol (containing 1% hydrochloric acid) was used for the extraction of flavonolglycosides and total anthocyanins; pure methanol solution was used for the extraction of all the other bioactive components. The extraction protocol reported by Maggini et al. (2013Maggini et al. ( , 2018 was used with modifications. The saffron corolla samples (0.2 g) were soaked with 2.5 mL extraction solvent in 10-mL test tubes. The samples were sonicated in an ice bath for 30 min. The sonication was repeated four times, and the tubes were stored overnight at −20 °C. Then, the supernatant liquid was separated by centrifugation for 4 min at 2500 rpm. The pellet of each sample was extracted again with 2.5 mL of extraction solvent. The supernatants were pooled together before subsequent analysis. The results were calculated on a fresh weight (FW) basis.

Total flavonoids
Total flavonoids were determined spectrophotometrically as described by Kim et al. (2003): 240 µL of NaNO 2 (5%) was added to 400 µL of methanolic extract, and after 5 min and stirring, 160 µL of AlCl 3 (10%) was added. The samples were shaken again and let stand for 5 min. Finally, 1.6 mL of NaOH (1 M) and 800 µL of water were added. The spectrophotometer reading was at ʎ 510 nm. The results obtained were expressed as milligrams of quercetin equivalents per gram of sample.

Flavonol glycosides and total anthocyanins
Flavonol glycosides and total anthocyanins were measured as reported by Hrazdina et al. (1982). The amount of total flavonol glycosides was measured spectrophotometrically on the acidic methanol extracts. Absorbance was read at 360 nm. Total flavonol glycoside concentration was expressed as quercetin equivalent in milligrams per gram of sample. For the quantification of total anthocyanins, the absorbance of the same extract was read at 530 nm, and the results were expressed as milligrams of cyanidin-3-glucoside equivalent per gram of fresh weight (g −1 FW), using the value 38,000 M −1 cm −1 for the molar absorptivity.

Flavonols
Flavonols were determined according to Radovanovićet al. (2010). The sample extracted in methanol was diluted 1:10 with ethanol (10% in water). Subsequently, 500 µL of this dilution was added to 500 µL of HCl (0.1% in ethanol) and to 2 mL of HCl (2% in water). The spectrophotometric reading was at ʎ 360 nm, and the results obtained were expressed as milligram equivalents of quercetin per gram of sample.

Total phenols
The amount of total phenols was determined by the Folin-Ciocalteu phenol reagent, according to the method reported by Kang and Saltveit (2002) with minor modifications (Maggini et al. 2018). An aliquot of 100 µl methanol extract was mixed with 2.0 mL of distilled water and 300 μL of Folin-Ciocalteu phenol reagent. The mixture was incubated for 4 min at room temperature (20-24 °C) and then 7.5% sodium carbonate (1.6 mL) was added. Then, the mixture was stored in the dark at room temperature for 2 h. Absorbance was then measured at 765 nm. For calibration, standard solutions of gallic acid (0-500 mg L −1 ) were prepared. The concentration of total phenols was expressed in terms of gallic acid equivalent as milligrams of gallic acid per gram of fresh weight (g −1 FW).

Antioxidant capacity
The ferric reducing antioxidant power (FRAP) assay was performed following the method described by Benzie and Strain (1996). For this assay, 2.0 mL of acetate buffer (0.25 M, pH 3.6), 900 μL of freshly prepared FRAP reagent (2 mM ferric chloride and 1 mM 2,4,6-tris(2-pyridyl)-s-triazine in acetate buffer), and 100 μL plant methanolic extract were used. Calibration was performed using a standard curve prepared with ammonium ferrous sulfate standard solutions, containing 0-1000 µM ferrous ion. The absorbance at 593 nm was determined, and the results were expressed as micromole (µmol) of Fe 2+ equivalents per gram of fresh weight (µmol Fe 2+ (g −1 FW).
The DPPH (2,2-diphenyl-1-picrylhydrazyl) radical scavenging activity assay was conducted according to Dudonné et al. (2009) with slight modifications as reported by Maggini et al. (2018). A 100-μL aliquot of methanol extract was mixed with 2.97 mL of methanol DPPH solution (20 mg L −1 ). A blank solution was also prepared by replacing the plant extract with methanol. The mixture was incubated at room temperature in the dark for 45 min; then, absorbance was measured at 515 nm. The percentage inhibition of the DPPH radical per gram tissue was calculated from the absorbance values of the blank (Ablank) and from the sample (Asample) as follows: where Ablank is the absorbance of the blank solution (absorbance of DPPH solution), and Asample is the absorbance of the testing sample solution.

Statistical analysis
Statistical analysis was performed using PRISM 9 software (GraphPad Software, San Diego, CA, USA). One-way ANOVA was used to analyze the effects of two multispectral LEDs, as the only light sources, and a natural greenhouse light condition, on flower organ development and secondary metabolite accumulation. Significant differences among means values were determined by Tukey's post-test at p < 0.05. Values are means of at least 10 independent biological samples. The whole experiment was independently repeated twice.

Effects of light spectrum manipulation on flower growth
The plants grown under the RB-LED treatment showed the highest whole flower and corolla fresh weights (FWs), greater than those of the control (natural light, NL) and RGB-LED groups (Table 1). Plants grown under RB-LED and RGB-LED conditions showed a significant increase in both pistil FW and dry weight (DW) compared to the NL group (Table 1). There were no significant differences in these parameters (pistil FW and DW) between the RB-LED and RGB-LED groups. Both flower and pistil lengths of plants grown under RB-LED and RGB-LED conditions were greater than that of the NL plant, which showed the lowest level of all groups (Table 1). There were no significant differences in flower and pistil lengths between the RB-LED and RGB-LED groups.

Effects of light spectrum manipulation on total flavonoid, flavonol glycoside, and flavonol accumulation
The combination of red, green, and blue light in RGB significantly increased the content of total flavonoids in C. sativus corolla, about 14.92 mg/g compared to 11.72 and 9.24 mg/g in the RB-LED and NL groups, respectively (Fig. 2a). The corolla of plants grown under RB-LED and RGB-LED conditions had significantly increased flavonol glycosides (7.64 and 8.23 mg/g) compared to the control (5.01 mg/g) (Fig. 2b). Similarly, the content of flavonols in both LED treatments (2.89 mg/g under RB and 2.54 mg/g under RGB) was significantly enhanced compared to the NL group (0.93 mg/g) (Fig. 2c). No significant differences in flavonol glycosides and flavonol concentrations were observed between LED treatments. Overall, these results showed that the LEDs positively and significantly influenced the total flavonoid, flavonol glycoside, and flavonol contents in tepals compared to the NL treatment.

Effects of light spectrum manipulation on total phenols
The concentration of total phenols in the corolla of plants grown under RB-LED (8.81 mg/g) and RGB-LED (9.46 mg/g) conditions was significantly higher than that of the NL plants, which showed the lowest level (6.77 mg/g) of all groups (Fig. 3). There were no significant differences in total phenols content between the RB-LED and RGB-LED groups.  Fig. 2 The concentration of total flavonoids (a), flavonol glycosides (b), and flavonols (c) in the tepals of Crocus sativus affected by different light environments. Data are shown as means with at least 10 independent biological replicates, and error bars indicate standard error (SE). Data were subjected to one-way analysis of variance, and differences were analyzed by Tukey's post-test. Different letters denote significant differences at p < 0.05. NL, natural light; RB, redblue LED combination; RGB, red-green-blue light combination

Effects of light spectrum manipulation on total anthocyanins
The concentration of total anthocyanins tended to be higher (but not statistically significant) in the tepals of plants grown under the RGB light environment than in the other light treatments (Fig. 4).

Light spectrum manipulation affects antioxidant capacity
The antioxidant power of C. sativus corolla from plants grown under RB-LED and RGB-LED conditions was determined using both FRAP and DPPH assays (Fig. 5), and the results obtained with the two independent methods were significantly correlated (p value < 0.05). In both LED treatments, the ferric-reducing capacity of the antioxidants (Fig. 5a) and the ability of antioxidants to scavenge the DPPH radical (Fig. 5b) were significantly higher compared to those observed in NL group.

Discussion
Light quality, which refers to the spectral composition of light, is a key regulator of plant growth through various photoreceptors, which are involved in light sensing and downstream signal transduction (Ward et al. 2005;Weller and Kendrick 2008). Different wavelengths of light are perceived by a complex network of photosensory pathways that enable plants to selectively activate individual pathways by inducing the expression of genes related to several physiological and metabolic functions (Whitelam and Halliday 2007;Li et al. 2012a, b). In this study, C. sativus, a high-value culinary and medicinal plant, was selected to investigate the potential Fig. 3 Total phenol content in the tepals of Crocus sativus affected by different light environments. Data are shown as means with at least 10 independent biological replicates, and error bars indicate standard error (SE). Data were subjected to one-way analysis of variance, and differences were analyzed by Tukey's post-test. Different letters denote significant differences at p < 0.05. NL, natural light; RB, redblue LED combination; RGB, red-green-blue light combination Fig. 4 Total anthocyanin content in the tepals of Crocus sativus affected by different light environments. Data are shown as means with at least 10 independent biological replicates, and error bars indicate standard error (SE). Data were subjected to one-way analysis of variance, and differences were analyzed by Tukey's post-test. Different letters denote significant differences at p < 0.05. ns, not significant. NL, natural light; RB, red-blue LED combination; RGB, redgreen-blue light combination  . (b) Radical-scavenging activity of extracts from tepals of Crocus sativus affected by different light environments determined by a DPPH test. Data are shown as means with at least 10 independent biological replicates, and error bars indicate standard error (SE). Data were subjected to one-way analysis of variance, and differences were analyzed by Tukey's post-test. Different letters denote significant differences at p < 0.05. NL, natural light; RB, red-blue LED combination; RGB, red-green-blue light combination of two different LED treatments for targeted plant physiological responses on its by-product, the corolla, proposing this residue as a novel source of enriched functional ingredients or raw material for the food and cosmeceutical industries. The combined LED treatments were accurately designed to uniformly provide the following light environment: (i) 50% red (660 nm)/12% green (500-600 nm)/38% blue (450 nm) at 120 µmolm −2 s −1 (RGB); and (ii) 62% red (660 nm)/38% blue (450 nm) at 120 µmolm −2 s −1 (RB). Plants grown under natural light (NL) were used as the control. Blue and red wavelengths are important for photosynthesis and have the greatest impact on plant growth due to the absorption peaks of chlorophyll molecules by increasing stomatal opening, electron transport, Rubisco activity, polyphenol concentration, and pigment production (Ouzounis et al. 2016;Olle and Virsile 2013;Taulavuori et al. 2013). Whole-plant studies suggest that supplemental G light (ranging from 24 to 10%) may improve plant growth in combination with R and B by increasing total photosynthesis in the individual leaves, as well as by transmitting to lower leaf layers (Broadersen and Vogelmann 2010;Johkan et al. 2012). In this study, the whole flower and corolla of C. sativus had the highest fresh weight when grown under RB than under RGB or NL light environment conditions. R and B wavelengths correspond to the absorption spectra of chlorophyll a and b, and in general, a combination of these wavelengths with a high R:B ratio have been shown to promote plant growth by achieving greater biomass production compared with R or B wavelengths alone (Amoozgar et al. 2017;Wollaeger and Runkle 2014;Fan et al. 2013;Li and Kubota 2009). Compared with R and B wavelengths, G light has not been considered to be effective for plant treatments due to its low capacity for chlorophyll absorption (Kopsell et al. 2014;Paradiso and Proietti 2021). However, G wavelengths have high transmittance and reflectance, and penetrate deeper in the plant canopy, which could potentially increase plant photosynthesis (plant yield) and regulate secondary metabolism, when appropriately combined with R and B wavelengths (Dou et al. 2019a;Dou et al. 2017;Wang and Folta 2013).The inclusion or not of G wavelengths in combination with RB resulted in greater pistil and flower length and FW compared to plants grown under NL conditions. However, G light reversed the effects of RB wavelengths on corolla biomass accumulation. Similar to that observed in basil and brassica (Dou et al. 2020), in C. sativus the inclusion of G wavelengths decreased corolla biomass (on FW basis) compared with that of plants grown under R and B wavelength combinations with a similar B percentage (38%). The addition of G light to a RB background has previously been shown to reverse B light-induced stomatal opening, resulting in lower plant photosynthesis and plant biomass, which can be used to make small adjustments in plant growth and secondary metabolism to best exploit prevailing conditions (Dou et al. 2020;Folta and Maruhnich 2007;Talbott et al. 2006).
Previous research has demonstrated that light quality is not only involved in plant photomorphogenesis but also stimulates plant secondary metabolism via the photosensory network driven by photoreceptor pathways (Jones 2018;Dou et al. 2017;Zhang and Folta 2012). Specific wavelengths, such as R, B, and G, enhanced the concentration of certain phytochemicals in several horticultural plants compared with white light or sunlight, and the level of enhancement is dependent on species, compounds, and light treatments (Dou et al. 2020;Taulavuori et al. 2018Taulavuori et al. , 2016Lobiuc et al. 2017;Arena et al. 2016). Polyphenols are secondary metabolites found in plants with various structures including phenolic acids, flavonoids, anthocyanins, flavonols, and numerous derivates (Cheynier 2015). These compounds are generally synthesized to cope with environmental challenges (i.e., UV radiation) and biotic stress (Mosadegh et al. 2021;Sharma et al. 2019). Moreover, these metabolites have extremely diverse biological properties, which make them unique and promising natural ingredients to be used in the food, pesticide, pharmaceutical, and cosmeceutical industries (Trivellini et al. 2016). Polyphenols have been reported to be affected by specific spectral bandwidths in the light environment in species-dependent and phytochemical-dependent manners, providing evidence that the effect of light can target the modulation of plant secondary metabolism to produce phytochemically enriched plants of high functional quality (Gam et al. 2020;Kyriacou et al. 2019;Mosadegh et al. 2018;Taulavuori et al. 2013;Stutte et al. 2009;Li and Kubota 2009). For example, phytochemicals such as phenolics, flavonoids, and anthocyanins were highest in a B light environment in Eruca sativa, Rehmannia glutinosa, and Perilla frutescens plants (Taulavuori et al. 2018;Lee et al. 2014;Manivannan et al. 2015), while rosmarinic acid (a phenolic acid) concentration in leaf tissues of Ocinum basilicum was enhanced under R light (Shiga et al. 2009). Most of the health benefits of saffron have been recognized since ancient times, and the corolla of C. sativus possesses high phenolic content and excellent antioxidant properties (Serrano-Diaz et al. 2012;Menghini et al. 2018;Caser et al. 2020). In agreement with Caser et al. (2020), this study shows that C. sativus tepals are rich in total phenols. LED treatments of 50% R (660 nm)/12% G (500-600 nm)/38% B (450 nm) (RGB) and 62% R (660 nm)/38% B (450 nm) at 120 µmol m −2 s −1 (RB) cause significantly higher total flavonoid, flavonol, and flavonol glycoside content than saffron plants grown under natural light. Interestingly, the concentration of total flavonoids in corolla increased significantly only by the addition of G light to a RB light environmental background. Also, the total anthocyanin content tended to be higher under the RGB-LED treatment, although it was not significantly different among the different light environments. In parallel with a higher concentration of total phenol bioactive compounds, poor biomass accumulation occurred in the corolla, which is in agreement with growth-defense trade-offs in which defense activation against adverse conditions through the activation of secondary metabolism generally comes at the expense of plant growth (Taulavuori et al. 2013;Huot et al. 2014). R and B light or combined spectra of these wavelengths are widely known to enhance the concentration of phenols, up-regulating the expression of genes or key enzymes involved in their biosynthetic pathway (Cuong et al., 2019;Li et al. 2010;Meng et al. 2004). However, only a few studies were conducted on the interactions between G and RB light combinations. In general, the inclusion of G wavelengths tends to reverse the B wavelength-induced phytochemical accumulation in a range of plant species such as Arabidopsis, basil, and lettuce (Zhang and Folta 2012;Dou et al. 2019;Pennisi et al. 2019). In C. sativus corolla, the inclusion of G light had a positive effect on secondary metabolites. Moreover, like the concentrations of key secondary metabolites (total phenols, total flavonoids, flavonols, and flavonol glycosides), the antioxidant capacity of C. sativus tepals also was favored by both RGB and RB light environments compared to natural light. In agreement with our findings, Dou et al. (2020) reported an increase in concentration and total amounts of bioactive molecules, the anthocyanins, in green kale with the addition of G wavelengths to a RB background, suggesting a species-specific effect on phytochemical accumulation by G light, which is dependent on the RB ratio.

Conclusion
Overall, the results reported here demonstrate the ability to influence the growth, development, and functional quality of C. sativus agricultural by-products through the manipulation of the light spectrum.
(i)Both LED treatments (RB and RGB) increased total flavonoids, flavonols, flavonol glycosides, and antioxidant capacity; (ii)Plants grown under RB light obtained the highest tepals biomass while reducing total phenol concentration compared to the NL (control) light environment; (iii)The inclusion of G wavelengths had positive effects on phenolics accumulation (i.e.; total flavonoids) and no effect on biomass accumulation compared to the NL (control) light environment.
Therefore, the use of LED treatments, specifically RGB, might lead to greater accumulation of bioactive components and represent a promising strategy for the valorization of floral residues as a functional-enriched ingredient for the food and cosmeceutical industries.