Plant material and climate conditions
The experiment was performed from June to October, 2016 in a Chinese solar greenhouse (CSG) and an artificial climate chamber (ACC, Zhejiang Qiushi Environment Co., Zhejiang, China) at the Horticultural Research Center, Shandong Agricultural University, P. R. China. Sweet pepper (Capsicum annuum L. cv. Hongqijian) seeds (Jinan Weili Seeds Co., Ltd., Shandong, China) were immersed in water for 15 min at 55 oC and then soaked in cold water (4 oC) for 24 h. The seeds were sown into 50-cell plug trays (54.0 × 30.0 × 4.4 cm) filled with a mixture of peat (Floragard Seed 2, Floragard Co., Oldenburg, Germany) and vermiculite (2:1, v/v) in the CSG. All seedlings were watered daily with half-strength Yamazaki’s pepper nutrient solution. Three weeks later, when their second true leaf had fully expanded, the seedlings were transplanted into plastic pots (8 cm long, 8 cm wide and 10 cm deep, one seedling per pot) containing the same substrate and watered with full-strength nutrient solution. Then, a total of 480 seedlings were selected, moved into the ACC, and cultured under four light quality treatments for 28 d. Each light treatment was repeated three times in the same ACC and there were 40 plants for per replication per treatment. Five plants were randomly sampled at 7, 14, 21, and 28 DAT from each replication each treatment and were subjected to morphological and biochemical analyses. There was ventilation in the controlled environment, so the CO2 level was the same as the CO2 level of atmosphere outside. The relative humidity (RH) was maintained at 70 ± 10 %, with a 12 h photoperiod and a temperature of 26 ± 1 oC during the daytime and 18 ± 1 oC at night.
All the mixed LEDs had a uniform spectrum for R and B light and were designed by Chunying Optoelectronics Technology Co., Ltd., Guangdong, China. The cultivation rack in the ACC was a steel frame structure with an LED light source placed at the top. The different treatments were insulated from one another by silver shading material. The plants were grown under the following light conditions: monochromatic B light with a maximum intensity at 457 nm, R light with a maximum intensity at 657 nm or mixed R and B light (3:1, RB: 75% R light at a wavelength of 657 nm and 25% B light at a wavelength of 457 nm). There was a multi-wavelength W light treatment as control (Supplementary Fig. 1). The light intensity, expressed as PPFD at the canopy level, was set at 300 μmol/m2·s, which was measured using a quantum sensor (LI-250, LI-COR Inc., Lincoln, NE, USA) and maintained by adjusting the distance of the LEDs from the canopies. The distance between the LEDs and the canopy was approximately 10 cm. The spectral photon flux density distributions (SPDs) of the LEDs were measured using a spectroradiometer (Unispec-SC Spectral Analysis System, PP Systems Inc., Haverhill, MA, USA).
Five seedlings, including leaves and roots, were removed from each replication each treatment at 28 DAT and dried in an oven at 105 °C for 30 min. The oven temperature was changed to 75 °C and the plants were dried to a constant weight. Then, the DWs of leaves and roots were measured using an electronic balance (precision: ± 0.1 g, Model LA16001S, Sartorius Co., Hamburg, Germany).
Leaf anatomy was measured on the fully expanded second leaves from five pepper seedlings at a similar position for each replication each treatment  on 28 DAT. Leaf segments of 5 mm × 5 mm were taken from the central leaf blade next to the main vein, fixed with formalin-acetic acid-alcohol (FAA) fixative, dehydrated in an alcohol and xylene series, embedded in paraffin, cross-sectioned to a thickness of 10 μm, and stained with red-solid green. The total thickness of the whole leaf and the thickness of the upper epidermis, lower epidermis, PT, and SPT were measured under a transmission light microscope (DP71, Olympus Inc., Tokyo, Japan). Images were collected using a digital camera (Camedia C4040, Olympus Inc., Tokyo, Japan) and analyzed by AnalySIS 5.0 (Olympus Inc., Tokyo, Japan).
Photosynthetic light- and CO2-response curves
The photosynthetic light-response curves and CO2-response curves were measured on the second fully expanded leaf between 09:00 am and 14:00 pm using a portable photosynthesis systems machine (LI-6400XT, Li-COR, Lincoln, NE, USA) at 28 DAT. The measurement technique was based on a modified method described by Pan et al. . The leaf chambers were set to temperature 26 ± 1 ∘C, air relative humidity 65 ± 5 %, and flow rate 300 μmol/s. The light-response curves were measured under a graded PPFD series of 1800, 1500, 1200, 1000, 800, 600, 400, 300, 200, 150, 100, 50, 20, and 0 μmol/m2·s. When the CO2-response curve measurements were taken, the light intensity and CO2 concentration of the leaf cuvette were set to 1000 μmol/m2·s and 400 μmol/mol, respectively, for 30 min. After it reached a steady state, a CO2 mixer was used to measure the CO2-response curves under a graded Ci value series of 400, 300, 200, 100, 50, 100, 200, 300, 400, 600, 800, 1000, 1200, 1500, and 1800 μmol·CO2/mol. It took about 120 to 180 s for the leaf chamber to adjust to its new microclimate for each measurement with one match. Each curve was measured three times and fitted with a non-linear regression equation, as previously reported [73, 74], so that the LCP, LSP, Pnmax, CCP, CSP, and the maximum RuBP regeneration rate. The AQY was the initial slope of the light-response curve and the CE was the initial slope of the CO2-response curve.
Chlorophyll fluorescence and chlorophyll fluorescence transients
The Chl fluorescence measurements were performed using a portable pulse modulation fluorometer (FMS-II, Hansatech Instruments Ltd., King’s Lynn, Norfolk, UK). The second fully expanded leaves of five seedlings from each replication each treatment were dark adapted for 20 min, and the Fo (original fluorescence yield) and Fm (maximum fluorescence yield) were determined. The leaves were then illuminated by natural light for 1 h, and the F'o, F'm, and Fs values were measured under 800 μmol/m2·s activating light. The saturation pulse intensity and duration were 3000 μmol/m2·s and 0.8 s, respectively. F'o and F'm represent the minimum and maximum fluorescence yields of an illuminated leaf, respectively, and were measured using the saturation pulse method. Fs represents the steady state fluorescence yield. The maximum photochemical efficiency of PSII was calculated using Fv/Fm = (Fm – Fo) / Fm, actual PSII photochemical efficiency was calculated using (ΦPSII) = (F'm – Fs) / F'm, and maximum photochemical efficiency of PSII under light adaptation was calculated using (F'v/F'm) = (F'm – F'o) / F'm.
The OJIP was measured on the second leaves by a plant efficiency analyzer (Handy PEA, Hansatech Instruments Ltd., King’s Lynn, Norfolk, UK). The JIP-test formulae and glossary of terms were calculated according to Strasser [75, 76]. The following derivative parameters were determined according to Lin et al.  and Miao et al. : RC/ABS, Sm, DIo/RC, TRo/RC, PIABS, PItotal, ΦRo, and δRo.
Calvin cycle enzymes activity
The second leaves from the top of 15 plants per treatment were sampled at 7, 14, 21, and 28 DAT to determine the enzyme activities. Leaf tissue (0.5 g) was homogenized in 4 mL of ice-cold extraction buffer: (25 mM Hepes (K+), pH 7.5, 10 mM MgSO4, 5 mM dithiothreitol (DTT), 1 mM Na2EDTA, 1 mM phenylmethanesulfonyl fluoride (PMSF), 5% (w/v) insoluble polyvinylpyrrolidone (PVP), and 0.05% (v/v) Triton X-100). The homogenate was filtered through muslin cloth and centrifuged at 14,000 × g for 5 min at 4 °C. The supernatant was used as the enzyme extract for the enzyme activity assays .
The Rubisco (EC 188.8.131.52), FBPase (EC 3.13.11), FBA (EC 184.108.40.206), GAPDH (EC 220.127.116.11), and TK (EC 18.104.22.168) activities were determined using an ELISA kit (Shanghai Yanji Biological Technology Ltd., Shanghai, China) and the extraction methods used for these enzymes were according to Rao and Terry  and Wang et al.  with some modification. The frozen leaf samples (0.5 g) were ground to a fine powder in liquid nitrogen with a mortar and pestle, transferred to a centrifuge tube, and then extracted in pre-chilled extraction buffer (5 mL). The enzyme extraction solution was centrifuged at 12,000 × g for 15 min at 4 oC. The supernatant was used for the Calvin cycle enzymes activity assay. Subsequently, the activities of the Calvin cycle enzymes were determined using a microplate absorbance reader (Bio-Tek ELX800, Bio-Tek Instruments, Winooski, VT, USA) at an absorbance of 450 nm according to the manufacturer’s instruction.
The protein concentration of each enzyme extraction solution was measured according to Bradford . The results were expressed as U/g of protein.
Total RNA was extracted using Quick RNA Isolation Kit according to the supplier’s instructions (Huayueyang Biotech Co., Ltd., Beijing, China). Reverse transcription was conducted using a ReverTra Ace qPCR RT-Kit (Toyobo Bio-Technology, Co., Ltd., Osaka, Japan). The gene expression analysis was conducted using real-time PCR, with 18S rRNA as an internal control. The thermal cycler program was one initial cycle of 94 °C for 2 min, followed by 40 cycles of 94 °C for 10 s, 60 °C for 20 s, and 72 °C for 30 s. Relative gene expressions were analyzed using the method described in Livak and Schmittgen . The specific gene primers used for real-time PCR analysis of the genes involved in the PS complexes are shown in Supplementary Table 1.
The experiment had a completely randomized design. Values presented are the mean ± standard deviation (SD) of three replicates. The data were analyzed by one-way analysis of variance (ANOVA) and the differences between the means were tested using Duncan’s multiple range test (P < 0.05). The charts were created using Origin (version 8.5, Microcal Software Inc., Northampton, MA, USA).