Seeds of Ocimum basilicum L. of the variety “Italian classic basil” were obtained from a commercial nursery (Sementi Dotto – Italy) and sown in 2 L cylindrical pots (height 14 cm, lower Ø 11 cm, and upper Ø 16 cm) filled with a commercial sterilized soil-less substrate (mixture of peat, siliceous sand, and bark humus 1:2:1). Plants were then grown for 30 days in a growth chamber equipped with high-pressure sodium (HPS) lamps, at a soil level light intensity of 200 µmol m^-2s^-1, and a photoperiod of 14 h (DLI = 10.08 mol m^-2day^-1). The temperature was maintained as close as possible to 22°C, with an air humidity ranging between 50% and 70%, and watering till saturation with tap water. After this initial growth period, the plants were subjected to light treatments as described below. Mentha x piperita L. plants were obtained from a commercial nursery (Maison aromatique – Italy). Young shoots were separated from the rhizome and transplanted to 2 L pots filled with sterilized soil-less substrate. The new plants were grown for 15 days under the same growth conditions used for O. basilicum before being subjected to different light treatments.

After the initial growth period, we subjected our plants to three different light treatments (LTs), respectively named HPS200, HPS100, and CoeLux^®. For each LT and each species, 10 plants were grown for 38 days and subsequently sampled to perform morphological and essential oils analysis. Moreover, after the same initial growth conditions and period, ten additional plants of both species were moved outdoors under direct sunlight (SUN LT) as a natural reference for the composition of the essential oils only. For the CoeLux^® LT, plants were grown inside the sunbeam of this artificial sunlight. We used the highest light intensity achievable with this lighting technology, corresponding to 100 µmol m^-2s^-1 (DLI = 5.04 mol m^-2day^-1) at the soil level and up to 120-140 µmol m^-2s^-1 (DLI = 6.05-7.06 mol m^-2day^-1) at the top of the plants. The soil level had a distance of from the lighting system of 50 cm, while the top of the plants reached a 10 cm distance. The photoperiod was maintained at 14h as set during the initial seedlings’ growth. We chose HPS lamps as the control light type since they are historically considered ideal light sources for indoor plant growth (Islam et al., 2012) (Pinho et al., 2012). Two different light intensities were used, 100 µmol m^-2s^-1 (DLI = 5.04 mol m^-2day^-1; named HPS100 LT) and 200 µmol m^-2s^-1 (DLI = 10.08 mol m^-2day^-1; named HPS200 LT), both measured at the soil level. For these treatments, the light reached an intensity at the top of the plants of 120-140 µmol m^-2s^-1 (DLI = 6.05-7.06 mol m^-2day^-1) for the HPS100 LT and 240-280 µmol m^-2s^-1 (DLI = 12.10-14.11 mol m^-2day^-1) for the HPS200 LT. This setting was in line with the indoor production of leafy vegetables and herbs since the existing literature often reports values ranging from 100 µmol m^-2s^-1 to 300 µmol m^-2s^-1 (Pennisi et al., 2020; Balázs et al., 2022). In addition, the 100 µmol m^-2s^-1 light intensity used in the HPS100 LT corresponds to the maximum light intensity achievable under the CoeLux^® lighting systems. The 200 µmol m^-2s^-1 light intensity used in the HPS200 LT was chosen as a further control for assessing the sole effect of the light intensity within the same light spectra. Plants moved under direct sunlight were grown in an open field at the University of Insubria - Campus Bizzozero (45°47’53.4” N 8°51’17” E – 392m asl - Varese, Italy) and harvested during the balsamic period (July 2021). In the months of June and July, this region is characterized by an average daily light integral (DLI) of 50-55 mol m^-2day^-1 (ENEA, 2020) and a photoperiod of 14.8-15.7 h.

The CoeLux^® lighting systems are engineered to perfectly resemble the natural light and sun appearance (Di Trapani and Magatti, 2014); the light is sourced by full-spectrum white LEDs, subsequently filtered to obtain the desired natural effect ( Figure 1 ) (Di Trapani and Magatti, 2017). Both the CoeLux^® and HPS light types were characterized in a previous study (Beatrice et al., 2021), using the HD 2302.0 Light Meter (Delta Ohm) to measure the light intensity and the SpectraScan PR655 (Photo Research) to measure the spectra every 4 nm in the range between 380 nm and 780 nm ( Figure 2 ). Spectra measurements were divided into color components, blue is the integral between 400-490 nm, green is the integral between 490-560 nm, yellow is the integral between 560-590 nm, red is the integral between 590-700 nm, and far-red is the integral between 700-780 nm. To calculate the red-to-far-red ratio (R/FR), red and far-red light was integrated between 650 and 670 nm and between 720 and 740 nm, while to calculate the blue-to-green ratio (B/G), blue and green light were integrated between 420 and 490 nm and between 500 and 570 nm, according to (Sellaro et al., 2010). Statistically significant differences were assessed using the post hoc Dunnett’s test. The HPS light type has a higher blue component, while the CoeLux^® light type has more yellow and red components. Green and far-red components showed no statistically significant differences. Despite similar values of FR light, the R/FR is higher under the CoeLux^® light type (4.68 ± 0.07) compared to the HPS light type (2.43 ± 0.02). While the B/G is higher under the HPS light type (0.83 ± 0.02) rather than under the CoeLux^® light type (0.50 ± 0.02). Additional spectra measurements were collected during a cloudless day in the location where the plants subjected to the SUN LT were grown. With respect to natural sunlight and considering equal light intensities, we observed a higher green, yellow, and red component in the CoeLux^® and HPS light types. The blue component was higher in the HPS light type and lower in the CoeLux^® light type, while the SUN light type had significantly higher levels of far-red. Under natural sunlight, we calculated an R/FR of 2.44 ± 0.02 and a B/G of 0.82 ± 0.02.

All leaves were detached from the stems and both organs were separately scanned at 800 dpi with the Epson Expression 12000XL instrument and then oven-dried at 70°C until constant weight. The roots were freed from the soil media by carefully washing them under running water and subsequently oven-dried at 70°C until constant weight. The dry organs were then weighed separately on an analytical scale (Orma AL220S) to obtain the leaves, shoot (leaves + stem), and root biomass (LB, SB, and RB) and calculate the shoot-to-root ratio (S/R). The scanned images were processed with WinRhizo (Regent Instrument Inc., Quebec, Canada) to measure the leaf area (LA) and with ImageJ software (National Institute of Health, USA) to measure the lamina and petiole length of three leaves for each plant. Subsequently, the leaf mass per area (LMA) and the lamina-to-petiole length ratio (L/P) were calculated.

For each plant, the central part of the fully expanded younger leaf was fixed and conserved at 4°C in formaldehyde-glutaraldehyde fixative (Karnovsky, 1965) for analysis at the microscope. To analyze the cross-section anatomy, leaves were dehydrated in a graded series of ethanol solutions and embedded in the Technovit^® 7100 resin (Kulzer GmbH, Germany). Samples were sliced with a Leica SM 2400 microtome (Leica Biosystems, Germany) at a thickness of 10 µm, colored with 0.5% toluidine blue and photographed using an Olympus BX63 optical microscope equipped with an Olympus DP72 camera (Olympus scientific solutions, Japan). Digital images ( Figure 3 ) were analyzed with the embedded cellSens imaging software (Olympus) to measure the whole leaf thickness (WLT) and the palisade thickness (PT). The ratio between WLT and PT was calculated. Five observation fields were randomly selected for each of the six biological replicates for a total of 30 fields per treatment.

To count the number of trichomes on the surface of the leaves, the samples conserved in the fixative solution were washed twice with phosphate-buffered saline (PBS) and immersed in a solution 1:5 v/v composed of osmium tetroxide 4% and PBS for 1.5 h. Samples were subsequently washed twice with PBS, dehydrated in a graded series of ethanol solutions, and incubated in hexamethyldisilazane (two steps of 20 min each). Two samples for each leaf were processed and mounted on aluminum stubs to analyze both the abaxial and adaxial surfaces of the leaf. The Emitech K550 gold sputter coater was used to coat the samples with a 10 nm gold layer. Finally, the samples were observed with an XL30 FEG (Philips) scanning electron microscope (SEM) at 7 kV and images of five randomly selected observation fields per leaf side were acquired at 50X and 150X magnification to count both peltate and capitate glands ( Figure 4 ). The total trichomes number, for the whole leaf area or per square centimeter, was calculated by summing the glands counted on the abaxial and adaxial surfaces of the leaf sample.

The fresh leaves of 10 different plants for each LT were pooled together and submitted to a 3 h hydrodistillation to extract the essential oils (EOs) according to the standard procedure described in the European Pharmacopoeia (Council of Europe et al., 2004). The EOs were dried over anhydrous sodium sulfate to remove traces of water and then stored in dark vials at 4°C before gas chromatography-mass spectrometry (GC-MS) analysis. The EOs yield was calculated and expressed as percentage on the dry weight of the extracted leaves. The characterization of all essential oils samples was determined with a gas chromatography system GC 86.10 Expander (Dani) equipped with an FID detector, an Rtx^®-5 Restek capillary column (30 m × 0.25 mm i.d., 0.25 μm film thickness) (diphenyl-dimethyl polysiloxane), a split/splitless injector heated to 250°C and a flame ionization detector (FID) heated to 280°C. The column temperature was maintained at 40°C for 5 min, then programmed to increase to 250°C at a rate of 3°C/min and held, using an isothermal process, for 10 min; the carrier gas was He (1.0 mL/min); 1μL of each sample was dissolved in n-hexane (1:250 n-hexane solution) and injected. The experiment was repeated three times. The GC-MS analyses were performed on a Trace GC Ultra (Thermo Fisher Scientific) gas chromatography instrument equipped with an Rtx^®-5 Restek capillary column (30 m × 0.25 mm i.d., 0.25 µm film thickness) and coupled with an ion-trap (IT) mass spectrometry (MS) detector Polaris Q (Thermo Fisher Scientific, Waltham, MA, USA). A Programmed Temperature Vaporizer (PTV) injector and a PC with a chromatography station Xcalibur (Thermo Fisher Scientific) were used. The ionization voltage was 70 eV; the source temperature was 250°C; full scan acquisition in positive chemical ionization was from m/z 40 up to 400 a.m.u. at 0.43 scan s⁻¹. The GC conditions were the same as those described above for the gas chromatography (GC-FID) analysis. The identification of the essential oil components was based on the comparison of their Kovats retention indices (RIs) and LRI (linear retention indices) (Kovats, 1965), determined in relation to the t_R values of a homologous series of n-alkanes (C8–C20) injected under the same operating conditions as those described in the literature (Doll and Kratz, 1963). The MS fragmentation patterns of a single compound were those from the NIST 02, Adams and Wiley 275 mass spectral libraries (NIST/EPA/NIH, 2005) (McLafferty, 1998). The relative contents (%) of the sample components were computed as the average of the GC peak areas obtained in triplicate without any corrections (Grob and Barry, 2004).

The statistical analysis was carried out with SPSS Statistics 25.0 (IBM) using the post hoc Dunnett’s test for multiple comparisons. In box plots, statistically significant differences (p< 0.05) between the means were marked with the letters a, b, and c. To characterize the levels of diversity in the chemical composition of the essential oils extracted from both M. piperita and O. basilicum (data in Supplementary Tables 1, 2 respectively), we considered the classic Shannon entropy (H) (Pielou, 1975), with p_i the proportion of the i-th of k compounds detected in the sample; and its relative version (J), the Pielou index (Mahmoud, 2021). The Shannon entropy can be used to compare the EOs diversity between treatments among the same plant species, while the Pielou index allows for a comparison of the diversity also between the two different plant species.

Furthermore, to compare the compositions of different samples and to assess the levels of dissimilarities between light treatments, we calculated the percent model affinity (PMA) index as reported in (Ärje et al., 2016), where A and B denote two generic samples.

In Mentha piperita the biomass of both shoot (SB) and root (RB) was respectively the highest and the lowest in plants grown under the HPS200 and CoeLux^® LTs, while plants grown under the HPS100 LT had intermediate values ( Figure 5A ). The shoot-to-root ratio (S/R) was respectively the highest and the lowest in plants grown under the CoeLux^® and HPS200 LTs, while plants grown under the HPS100 LT had intermediate values ( Figure 5C ). The lamina-to-petiole length ratio (L/P) was respectively the highest and the lowest in plants grown under the HPS200 and CoeLux^® LTs, while plants grown under HPS100 had an intermediate median, but no statistically significant differences were observed with the other LTs ( Figure 5E ).

A similar pattern was observed in Ocimum basilicum. However, no statistically significant differences in the SB were observed between plants grown under the HPS100 and HPS200 LTs ( Figure 5B ) and in the S/R between plants grown under the HPS100 and CoeLux^® LTs ( Figure 5D ). The L/P showed no statistically significant differences between all light treatments tested ( Figure 5F ).

In Mentha piperita the leaf area (LA) was the highest in plants grown under the HPS100 and CoeLux^® LTs and the lowest in plants grown under the HPS200 LT ( Figure 6A ). The leaf biomass (LB) was the highest in plants grown under the HPS200 and HPS100 LTs and the lowest in plants grown under the CoeLux^® LT ( Figure 6C ). The leaf mass per area (LMA) was respectively the highest and the lowest in plants grown under the HPS200 and CoeLux^® LTs, while plants grown under the HPS100 LT had intermediate values ( Figure 6E ).

A similar pattern was observed in Ocimum basilicum for the LA and the LB ( Figures 6B, D ), while the LMA was the highest in plants grown under the HPS200 and HPS100 LTs and the lowest in plants grown under the CoeLux^® LT ( Figure 6F ).

In Mentha piperita the whole leaf thickness (WLT) was respectively the highest and the lowest in plants grown under the HPS200 and CoeLux^® LTs, while plants grown under the HPS100 LT had intermediate values ( Figure 7A ). The palisade thickness (PT) was respectively the highest and the lowest in plants grown under the HPS200 and CoeLux^® LTs, while plants grown under the HPS100 LT had intermediate values ( Figure 7C ). The ratio between WLT and PT showed no statistically significant differences in all light treatments tested ( Figure 7E ).

In Ocimum basilicum the WLT was the highest in plants grown under the HPS200 and HPS100 LTs and the lowest in plants grown under the CoeLux^® LT ( Figure 7D ). The PT was respectively the highest and the lowest in plants grown under the HPS200 and CoeLux^® LTs, while plants grown under HPS100 had an intermediate median, but no statistically significant differences were observed with the other LTs ( Figure 7D ). The ratio between WLT and PT showed no statistically significant differences in all light treatments tested ( Figure 7F ).

In both plant species, the number of peltate and capitate glands measured for the whole leaf area or calculated per square centimeter of leaf surface did not show differences in all light treatments tested ( Supplementary Figures 1 and 2 ).

Mentha piperita leaves showed a higher essential oils (EOs) yield (0.67-1.34%) when compared to Ocimum basilicum leaves (0.17-0.40%). In M. piperita the highest and lowest yields were reported in the SUN and HPS200 LTs respectively, while in O. basilicum the highest and lowest yields were reported in the HPS200 and SUN LTs ( Table 1 ). A whole of 57 different molecular compounds was identified in the EOs of M. piperita leaves grown under different LTs ( Supplementary Table 1 ). The highest number of compounds was observed in plants grown under the SUN LTs, while the lowest number of compounds was observed under the HPS200 LT. The CoeLux^® and HPS100 LTs had an intermediate number of molecules. Among the different LTs, the GC-MS TIC chromatogram coverage area ranged from 96.8% to 99.3% ( Table 1 ). In O. basilicum a whole of 47 different molecular compounds was identified ( Supplementary Table 2 ). The highest number of compounds was observed in plants grown under the SUN LTs, while the lowest number of compounds was observed under the CoeLux^® LT. The HPS200 and HPS100 LTs had an intermediate number of molecules. Among the different LTs, the GC-MS TIC chromatogram coverage area ranged from 96.8% to 99.1% ( Table 1 ).

In M. piperita the EOs of plants grown under all three LTs under study (i.e., the CoeLux^®, HPS200 and HPS100 LTs) were richer in p-menthone and poorer in menthol with respect to the SUN LT ( Figures 8A-C ). The EO of plants grown under the HPS100 LT showed a chemical composition comparable to that of CoeLux^® plants ( Figure 8D ), while plants grown under the HPS200 LT showed to have more menthol and less p-menthone with respect to the CoeLux^® and HPS100 LTs ( Figures 8E, F ). Furthermore, cis-beta-Guaiene and Eptadecane are produced almost exclusively under the CoeLux^® LT; and Pulegone, Terpinene-4-ol, α-Terpineol, Himachelen-ol, Cadinol-epi-α, and Cedren-13-ol<8> are produced only under artificial lighting ( Supplementary Table 1 ). In contrast, 3-Octanol, trans Mentha-2,8-dien-1-ol, β-bourbonene, α-Muurolene, Guaiene<trans-b->, Selina-3,11-dien-6-a-ol, Himachalol, Eudesm-7(11)-en-4-ol are produced only under the SUN LT ( Supplementary Table 1 ).

In O. basilicum the EOs of plants grown under all three LTs under study (i.e., the CoeLux^®, HPS200 and HPS100 LTs) were richer in Eugenol and poorer in Cubenol with respect to the SUN LT ( Figures 9A-C ). The EO of plants grown under the CoeLux^® LT were richer in Estragole, Borneol and Terpinen-4-ol, and poorer in Linalool with respect to the other LTs ( Figures 9A, D, E ). The EOs of plants grown under the HPS LTs showed to have more 1,8-cineole with respect to the CoeLux^® and SUN LTs ( Figures 9B-E ). Furthermore, plants grown under the CoeLux^® LT produced no α-Terpineol, which was reported for the other LTs ( Supplementary Table 2 ).

Both in M. piperita and O. basilicum, the EOs composition assessed in plants grown under the CoeLux^® LT showed respectively the highest and lowest similarity with plants grown under the HPS100 LT and the SUN LT ( Table 2 ). The SUN LT showed the highest and lowest similarity in both species with the HPS200 and the CoeLux^® LTs, respectively ( Table 2 ). For M. piperita the highest similarity index value (0.899) was reported between the HPS200 and HPS100 LTs; for O. basilicum the highest similarity index value (0.893) was found between the CoeLux^® and HPS100 LTs.

In general, the O. basilicum EOs compositions had a higher level of diversity (Pielou index ranging between 0.589 and 0.644) than those observed for M. piperita (ranging between 0.459 and 0.536) ( Table 2 ). For both species, the EOs composition of plants grown under the SUN LT resulted in the highest entropy and diversity index values ( Table 2 ). In M. piperita the lowest diversity values resulted in the HPS100 LT, while in O. basilicum the lowest diversity values resulted in the HPS200 LT ( Table 2 ).

Previous studies have described the impact on the model plant Arabidopsis thaliana of the LED-sourced biophilic lighting systems CoeLux^® (Beatrice et al., 2021), characterized by low light intensity levels and low blue and high red light spectrum composition. In A. thaliana, the phenotype was characterized by low values of biomass, leaf area, and lamina-to-petiole ratio, symptomatic of the onset of an intense shade avoidance syndrome (SAS). In particular, plants grown at lower light intensities showed a delayed life cycle and were significantly smaller than plants grown with higher light intensities. Furthermore, within the same light intensity plants grown under the CoeLux^® light type showed an additional deficit when compared to control plants grown under the HPS light type (Beatrice et al., 2021). However, responses to such a peculiar light type might change among different plant species that differently modulate the response to light characteristics (Larsen et al., 2020). Therefore, we analyzed the morphological traits of two common aromatic plant species, Ocimum basilicum and Mentha x piperita, aiming to understand to which extent the CoeLux^® light type may enable their growth and development in closed environments. In addition, through the application of GC-MS analysis, we investigated the essential oils (EOs) composition as a possible indicator of the activation of diverse metabolic pathways.

Multiple morphological changes were observed in response to the different light types and intensities, which underscored a slightly different response among the two plant species analyzed. M. piperita showed a general suppression of both above and belowground biomass due to the lowering of the control light intensity and to the difference in the light spectrum occurring between the control (HPS) and the CoeLux^® light type. In the case of O. basilicum, the same pattern was observed for the root biomass (RB) while the shoot biomass (SB) was significantly lower only when comparing plants grown under the two light types and independently of the light intensity. These findings are in contrast with what was found in O. basilicum by (Jensen et al., 2018a) and (Larsen et al., 2020), who found no differences in SB between LTs with different proportions of red and blue light. Our results suggest that a higher proportion of blue light (24% instead of 14%) and a lower proportion of red light (35% instead of 41%), as in the case of the HPS light type, could increase the SB. Another difference between the two species was highlighted by the shoot-to-root ratio (S/R) values. Indeed, in the case of M. piperita S/R values were higher with the lowering of the light intensity (within the HPS light type) and between the HPS and the CoeLux^® light type (within the same light intensity). Under unfavorable light conditions, the plants increased the allocation of biomass to the shoots in an attempt to enhance the uptake of the most limiting factor, light (Poorter et al., 2012). On the contrary, in the case of O. basilicum, the only difference was detected with the decrease in light intensity of the HPS light type, while no differences were observed among different light spectra originating from the HPS and the CoeLux^® light types. These findings highlight that the growth of O. basilicum, especially for the aboveground organs, is influenced to a lower extent by both light intensity and spectra.

The leaf area of both M. piperita and O. basilicum was higher with the lowering of the light intensity (within the HPS light type), while no difference was detected among the two different light spectra (within the same light intensity). Tabbert et al. (2022) attributed the greater leaf expansion observed in M. piperita to the low blue-to-green ratio (B/G) of their LTs, while Larsen et al. (2020) described a decrease in leaf area with the increase of the fraction of blue light in O. basilicum. However, we found no significant differences in the leaf area between the two light spectra of our LTs, despite having a higher B/G in the HPS light type (0.83) with respect to the CoeLux^® light type (0.50) due to a higher blue fraction composing the HPS light spectra (24%) in respect to the CoeLux^® light spectra (14%) and equal green fractions (24%). In both species, a lowering in the leaves’ biomass was detected in plants grown under the CoeLux^® light type when compared to plants grown under the HPS light type (within the same light intensity), but no differences were observed with the lowering of the light intensity. However, when the leaf mass per area (LMA) was calculated, further differences between the two plant species were detected. Indeed, in the case of M. piperita values were lower with the lowering of the light intensity (within the HPS light type) and between the HPS and the CoeLux^® light type (within the same light intensity), while in O. basilicum a lower value was observed only in the case of the CoeLux^® light type (characterized by a higher red and lower blue fraction). These data are in contrast with what was observed by Jensen et al. (2018), who found no differences in the LMA of O. basilicum grown under LTs with different fractions of red and blue light.

Further investigations showed that a lower whole leaf thickness (WLT) could be the reason for the lower LMA observed in both species ( Figure 6 ). Indeed, the reduction in leaf thickness is a typical response observed during the SAS onset (Smith and Whitelam, 1997) for a plant dealing with a reduction in light intensity or a change in the light spectrum (e.g., lowering in the red/far-red ratio) (Franklin et al., 2005). In M. piperita WLT values were lower with the lowering of the light intensity (within the HPS light type) and between the HPS and the CoeLux^® light type (within the same light intensity). On the contrary, in O. basilicum a lower value was observed only in the case of the CoeLux^® light type, showing that a lowering in light intensity is not significantly affecting the leaf thickness in this species. When the leaf palisade was investigated, it was observed that for both species its structure remains intact also at the lowest light intensity ( Figure 3 ), but the thickness (PT) reduced significantly with different magnitudes among the two plant species. Indeed, in the case of M. piperita, the PT values were lower with the lowering of the light intensity (within the HPS light type) and between the HPS and the CoeLux^® light type (within the same light intensity). On the contrary, in the case of O. basilicum, a significant PT reduction was detected only when the interplay between light intensity decrease and light spectrum change is considered (i.e., between the HPS200 and CoeLux^® LTs). These trends seem consistent with the WLT reduction, as no statistically significant differences were observed in the ratio between whole leaf thickness and palisade thickness. Apparently, all leaf organs undergo an even thickness reduction in response to the biophilic light conditions tested.

In A. thaliana the lamina-to-petiole length ratio (L/P) reduction is considered a hallmark response to unfavorable light (Keller et al., 2011). A low lamina-to-petiole ratio and a high shoot-to-root ratio are characteristic responses for improving light collection under unfavorable light conditions (Poorter et al., 2012). Indeed, under low light conditions, A. thaliana develops a longer petiole and a shorter lamina in an effort to better align the leaves toward the light source and collect more light for photosynthetic activity (Beatrice et al., 2021). However, in M. piperita an L/P reduction was detected only with the interplay between light intensity decrease and light spectra change (i.e., between the HPS200 and CoeLux^® LTs), while no differences were observed within plants grown under the HPS light type with a reduction in light intensity or within plants grown at the same light intensity but with a different spectrum. Furthermore, O. basilicum showed no statistically significant differences under all LTs tested, demonstrating that the responses to biophilic lighting of this plant species are more focused on strategies like leaf area expansion, instead of leaf movement toward better light conditions.