Seven seedlings for two cultivars of C. intybus, Tardiva and Precoce (both provided by Franchi Sementi s.p.a.), were grown in 25 cc cells of polystyrene germination trays, containing a mixture of sand:peat (4:1, v:v), in a growth chamber with a temperature of 25°C and 60% of RH, for four weeks. Three groups of plants per cultivar were grown under three different light conditions, provided by a multi-channel LEDs lighting system (ENFIS Ltd, UK): i. White LEDs (complete visible spectrum LEDs), ii. Blue LEDs (monochromatic channel radiance, with the wavelength peak at 461 nm, set up to be 70% higher than the other light components) and iii. Red LEDs (631 wavelength peak radiance set to 70% higher than the other channels) ( Supplementary Figure S1 ). In all the light treatments, the total photosynthetic photon flux density (PPFD) was 600 μmol m^–2 s^–1.

After four weeks of growth under different light conditions, the photosystem II efficiency - φ_PSII - was measured on two leaves per seedling (i.e., the two broadest leaves) with a handheld porometer-fluorometer (Li-600, LICOR Biosciences, USA). Measurements have been collected under white light at 600 μmol m^–2 PPFD. Chlorophyll leaf content has been assessed at three and four weeks of growth using a handheld optical analyzer (Dualex, Force One, France).

Two leaves per plant from four plants were collected, frozen in liquid nitrogen and stored at −80°C until the moment of the extractions for biochemical analyses. Leaves from the same plants were pooled together to make individual replicates (n=4). Four replicates per treatment were used. Leaf fresh material (100 mg) was ground in a mortar with liquid nitrogen and then extracted with 3 × 1 mL ethanol 75% solution (pH 2.5 adjusted with formic acid) using an ultrasonic bath (BioClass^® CP104, Pistoia, Italy) at a constant frequency of 39 kHz and power of 100 W, during 30 min, at 5°C. After that, extracts were centrifuged (5 min, 9000 rpm, ALC^® 4239R, Milan, Italy), and the supernatants were partitioned with 3 × 2 mL of n-hexane to remove lipophilic compounds that could interfere with the analysis. The hydroethanolic phase was reduced to dryness using a rotavapor (BUCHI^® P12, Cornaredo, Italy; coupled to a vacuum controller V-855), and the residue was resuspended with 250 µL of MeOH: Milli-QH₂O solution (1:1 v/v, pH 2.5 adjusted with formic acid). After the extraction procedure, the solution was characterized and quantified by HPLC-DAD/Q-TOF.

The characterization and subsequent quantification of polyphenols was made by LC-QTOF (Agilent 6530C, Agilent Technologies SpA, Milan, Italy) utilizing a quadrupole mass spectrometer operating in the electrospray ionization (ESI) negative mode coupled to a diode array detector (DAD). The applied ESI parameters were as follows: capillary voltage, 4000 V; fragmentor 180 V; skimmer 60 V; OCT 1 RF Vpp 750 V; pressure of nebulizer 20 psi; drying gas temperature 325°C; sheath gas temperature 400°C. Compounds separation was performed using an Agilent Poroshell 120 Aq-C18 column (2.7 μm) applying a 40-minute linear gradient solvent passing from 97% of water acidified with 0,1% formic acid (solvent A) to 97% of acetonitrile acidified with 0,1% formic acid (solvent B). The flow rate was of 0.30 mL min^-1 and the injection volume was 1 μL. The quantification was performed in DAD at the specific wavelengths of the different compounds using five-point calibration curves of the following standards (all from Extrasynthese, Lyon, France): caftaric acid, chlorogenic acid, chicoric acid, quercetin 7-O-glucoside, apigenin 7-O-glucoside, kaempferol 7-O-glucoside, isorhamnetin 7-O-rhamnoside, luteolin 7-O-glucoside, ferulic acid, epigallocatechin, cyanidin 3-O-glucoside chloride and cyanidin 3,5-diglucoside chloride.

The data were analyzed for normality of distribution (Shapiro-Wilk test) and homoscedasticity (Levine test) and when at least one of these two assumptions was missing, the non-parametric Kruskal-Wallis test for multiple comparisons, followed by the Mann-Whitney U test for pairwise comparison, was performed. For normally distributed data, a one-way ANOVA, followed by Tukey’s multiple comparison test, was performed within the same sampling time (i.e., 3 or 4 weeks). For the comparison between the same treatment supplied to the two cultivars, an unpaired Student’s t-test was run. Differences were considered statistically significant at p <0.05.

The values of the photosystem II efficiency (φ PS II) measured for plants grown under red LEDs were significantly lower than those of blue and white LEDs treatments. This result was registered in both cultivars. Regarding the comparison of the two cultivars, no significant differences were observed within each single treatment (e.g. white/blue/red LEDs) ( Figures 1A, B ).

The chlorophyll content measured in Tardiva after three weeks of irradiance with white, blue and red LEDs did not change depending on the different light, as no significant differences were detected among the treatments ( Figure 2A ). A significant increase was registered for plants exposed to blue LEDs for four weeks, with this value being significantly higher than those observed for white and red LEDs treatments. The overtime comparison of each single treatment revealed that a significantly lower chlorophyll content was registered after four weeks of irradiance with red LEDs, whereas no significant time-dependent differences were observed for white and blue LEDs treatments ( Figure 2A ).

Considering the Precoce, significantly higher chlorophyll contents were measured after three weeks of irradiance with both blue and red LEDs with respect to white LEDs, the blue and red values being comparable to one another ( Figure 2B ). After four weeks of irradiance, plants exposed to white and red LEDs showed similar chlorophyll contents, whereas a significantly higher value was observed for blue LEDs treatment. Considering the single treatment, white and blue LEDs induced a significant increase in chlorophyll content over time, as values after four weeks were significantly higher than those registered after three weeks, whereas no time-dependent changes were observed for red treatment ( Figure 2B ).

Comparing the two cultivars, a significantly higher content of chlorophyll was registered for Precoce exposed to blue and red LEDs for four weeks, whereas the Tardiva showed a significantly higher value after three weeks of white LEDs irradiance ( Supplementary Figure S2 ).

The total content of polyphenols measured in Tardiva was significantly increased by the exposure to blue and red LEDs for three weeks, showing comparable values in the two treatments ( Figure 3A ). The highest content of polyphenols was measured in plants exposed to blue LEDs after four weeks of irradiation. In Tardiva, this value was significantly higher than in the red treatment, which in turn, significantly differed from the control (white LEDs). Blue LEDs also induced a significant increase in the total polyphenols content over time: after four weeks this content was significantly higher than that measured after three weeks ( Figure 3A ).

Looking at the Precoce, significantly higher levels of polyphenols were registered in response to blue LEDs irradiation compared to red one, after three weeks of exposure ( Figure 3B ). Red treatment also induced a significant accumulation of polyphenols compared to the control (white LEDs). After four weeks of treatment, the highest content of polyphenols was observed for blue LEDs treatment, inducing a significant accumulation of these compounds when compared to white and red LEDs. Regarding the overtime effect, a significant accumulation of polyphenols was registered in response to white and blue LEDs, whereas no significant differences were found between three and four weeks of red LEDs irradiance ( Figure 3B ).

The comparison of the two cultivars showed that Precoce has a significantly higher polyphenol content than Tardiva, regardless of the sampling time. This result was observed for all the treatments except for the blue LEDs at four weeks ( Supplementary Figure S3 ).

The HPLC-DAD-MS analysis allowed the identification of the polyphenols in the two cultivars ( Table 1 ) and revealed differences in the composition depending on the light treatment supplied and, to a lesser extent, on the considered cultivar. In both cultivars, a similar composition was observed in plants grown under white and blue LEDs, and a time-dependent accumulation was observed for all compounds. Under white and blue LEDs, 12 polyphenolic compounds were detected in the Precoce ( Supplementary Table S1 , Figure 4A ). The same composition was observed in Tardiva, excepting for the absence of kuromanin and three cyanidin derivatives, and the presence of isorhamnetin-7-O-glucuronide ( Supplementary Table S2 , Figures 5A , 6A ).

The red LEDs induced an accumulation of nine compounds in both cultivars ( Supplementary Tables S1 , S2 ). The Precoce also showed the presence of two additional compounds, i.e., kuromanin and cyanin 3,5 diglucoside ( Figure 4B ).

Chicoric acid, followed by quercetin derivative and quercetin 7-O-glucoside, were the main polyphenolic compounds detected in both cultivars exposed to white and blue LEDs. Kaempferol 3-O-glucuronide was the most represented compound found in plants exposed to red LEDs, and in the case of Tardiva cultivar, also high levels of 5-O-feruloylquinic acid were observed after four weeks of treatment ( Figures 5B , 6B ).

PS II efficiency was negatively affected by the red LEDs treatment in both cultivars, whereas the performances of the PS II did not vary in response to white and blue LEDs exposure. It is worth noting that longer wavelengths, such as those in the red and far-red region of the light spectrum, preferentially excite PS I and when a PS is overexcited the efficiency of the other one is limited (Zhen et al., 2019). An overexcitation of the PS I caused by the red treatment could explain the lower efficiency values observed in both C. intybus cultivars with respect to the other treatments. Nevertheless, it has been observed in other species, such as potato (Chen et al., 2021) and Artemisia argyi (Su et al., 2024) that plants grown under monochromatic red light develop a reduced potential photosynthetic capability in PS II and an increased light energy dissipation via non-photochemical quenching, compared to plants grown under white or blue light.

Concerning the chlorophyll content, the highest accumulation was induced by blue LEDs in both cultivars. Particularly, in the Precoce, chlorophyll accumulation was directly proportional to the time, also highlighting a clear distinction between the effect of blue LEDs and that of the other two irradiance systems at a more mature stage of development of the leaf (i.e., 4 weeks).

The blue LEDs also determined the greatest accumulation of polyphenols in both cultivars, at the early and mature state of leaf development. Similar increases in total chlorophyll content and polyphenols levels induced by blue LEDs light were previously observed in other herbaceous species (Johkan et al., 2010; Manivannan et al., 2015; Lobiuc et al., 2017; Park et al., 2019; Azad et al., 2020). Particularly, under blue LEDs, the polyphenols content in the Tardiva was more than doubled at four weeks, pointing out the correlation between the accumulation of these metabolites and leaf development. This is in agreement with previous studies demonstrating that blue LEDs-irradiation constitutes a powerful tool to produce high-quality red leaf vegetables, leading to an improved content of bioprotective pigments (Son and Oh, 2013; Carvalho and Folta, 2014).

Both blue and red lights are efficiently absorbed by plant photosynthetic pigments, but the blue wavelength contains more energy (Ouzounis et al., 2015; Gómez and Izzo, 2018; Sytar et al., 2019). Among all monochromatic radiations, the red one is the most efficiently converted into chemical energy, hence resulting in very photosynthetically active radiation (Gómez and Izzo, 2018). Tardiva in fact positively reacted also to the treatment with red LED, exhibiting a significantly higher polyphenols content than plants exposed to white LEDs. This increase was stable over time, although considerably lower compared to that observed for blue light. Research on LEDs manipulation already showed that the responses to light quality are species-specific, requiring unique protocols to obtain high-productive and high-nutritive vegetables, since a great variability of responses can be observed depending on the lighting conditions (i.e., light wavelength, intensity, photoperiod) (Mitchell and Stutte, 2015; Gómez and Izzo, 2018; Appolloni et al., 2022; Orlando et al., 2022; da Cristina Bungala et al., 2024). In some cases, responses can be different in diverse cultivars within the same species (Gómez and Izzo, 2018). In the case of chicory, our results suggested that the main differences seemed not to be correlated to the cultivars, but rather to the type of light provided during plants growth. Overall, the best performances in terms of both chlorophyll and polyphenols accumulation were obtained under blue LEDs. In terms of quantities, the Precoce showed the greatest content of polyphenolic compounds, regardless of the type of LEDs used for its cultivation. Most of the studies on microgreens, including those conducted on chicory (Pennisi et al., 2020), explore the effect of blue and red LEDs combined in different ratios, but very different results have been reported, especially regarding metabolite accumulation (Appolloni et al., 2022). In some cases, the best performances were obtained with monochromatic light only (Zhang et al., 2019), suggesting that the lighting system must be carefully tuned according to the species responses, in order to optimize the production.

The polyphenolic composition of the two C. intybus cultivars was strongly influenced by the type of light used for their cultivation. For plants grown under white and blue LEDs, only quantitative differences were found in the leaf polyphenols, whereas a completely different composition characterized the leaf extracts of those plants grown under red LEDs. This pattern was consistent across both cultivars ( Supplementary Figure S4 ). When exposed to white/blue LEDs, the leaves accumulated polyphenolic compounds with high antioxidant potential, such as quercetin derivatives and chicoric acid (Lesjak et al., 2018). This aligns with the increased efficiency exhibited by the PS II in response to white/blue LEDs and the higher content of chlorophyll as well, suggesting an intense photosynthetic activity. In such conditions, the disposal of strong antioxidant compounds ensures the prevention of ROS-related damages, being ROS massively produced during photosynthesis (Foyer, 2018; Singh et al., 2021).

The biosynthesis of quercetin is catalyzed by flavonol synthase (FLS), an enzyme that is affected by light spectral composition (Singh et al., 2021). For example, it was observed that the expression of LsFTS gene coding for the FLS in lettuce, was considerably enhanced by the combined irradiation with UV-B plus blue light compared to UV-B radiation only. This corresponded to a significant increase in the leaf quercetin (Ebisawa et al., 2008). Flavonol accumulation induced by blue LEDs has also been reported in other species (Li et al., 2018; Wang et al., 2022). A similar effect could be hypothesized for the two C. intybus cultivars considered in this study, as the metabolic pathway for the biosynthesis of flavonols is very ancient and highly conserved (Pollastri and Tattini, 2011). Particularly, in our study, it is likely that the enhancement of both FLS and F3’H (Flavonoid 3’-hydroxylase) activity induced by blue LEDs led to the accumulation of quercetin derivatives rather than kaempferol derivatives (Shi et al., 2014). In particular, blue LEDs enhanced the accumulation of three different quercetin derivatives (quercetin derivative, quercetin 7-O-glucoside, quercetin-7-O-(6’’-O-malonyl) glucoside), all present in the control (white LEDs), but in lower amounts. By enhancing the activity of several enzymes involved in the polyphenols biosynthesis, and in particular of the above-mentioned FLS and F3’H, blue light induces the accumulation of functional compounds with nutraceutical properties in many horticultural crops (Heo et al., 2012; Alrifai et al., 2019; Orlando et al., 2022).

In contrast, red LEDs drove the flavonoid biosynthetic pathway towards the accumulation of kaempferols, with kaempferol 3-O-glucuronide being predominant in both cultivars ( Supplementary Figure S4 ). Due to their structure (i.e. lower number of hydroxyl groups and lack of catechol in the molecule), these compounds are less antioxidant than quercetins (Rice-Evans et al., 1996; Agati and Tattini, 2010; Dueñas et al., 2011). This structural difference supports the hypothesis that red LEDs irradiance may induce lower stress levels in plants, as also confirmed by a reduced amount of caffeic acid derivatives in favor of ferulic acid derivatives which have a more structural property for cell walls rather than antioxidant activity (Harris and Trethewey, 2010). Accordingly, only few anthocyanins, which are also known to play a protective role as antioxidants and ROS scavengers (Quina et al., 2009; Tena et al., 2020; Agati et al., 2021), were found in Precoce seedlings treated with red LEDs compared to those exposed to blue LEDs. This result is in accordance with the presence of epicatechin derivatives detected only under red light, suggesting an induction of ANR (anthocyanidin reductase) by red light (Zhang et al., 2018). Monochromatic red light was found to prevent the synthesis of anthocyanins in red curly lettuce (Heo et al., 2012), whereas there is good evidence that blue LEDs typically induces an accumulation of anthocyanins in many horticultural plants, even at the postharvest phase (Johkan et al., 2010; Xu et al., 2014; Sytar et al., 2019). For instance, Stutte et al., 2009 evaluated the effect of different LEDs on red leaf lettuce, finding that after exposure to blue LEDs, the content of bioprotective anthocyanins was more than doubled compared to what observed under red LEDs.