Lettuce plants (Lactuca sativa L. cv. Romana Lentissima a Montare 4, FOUR-BLUMEN s.r.l., Piacenza, PC, Italy) were cultivated under the greenhouse of Agronomy and Crop Sciences Research and Education Center, University of Teramo, Mosciano Sant' Angelo (42° 53′ N and 13° 55′ E, 15 m; above sea level) from June to July 2013. The greenhouse is covered with a single layer of ethylene-vinyl acetate (PATILUX) provided by P.A.T.I. S.p.A. (San Zenone degli Ezzelini, TV, Italy); it has a natural ventilation system, and it is not provided of artificial lights, fans, and heaters. The % of reduction with respect to outdoors conditions in terms of total (W m ^-2) and PAR (µmol m^-2 s^-1) radiation amounted to 7.4% and 12.6%, respectively. Moreover, the plastic film, as expected, causes a reduction of the irradiance (µmol m^-2 s^-1) by 64, 32 and 24% on average in the ultraviolet, PAR and near infrared regions, respectively. Starting from transplanting air temperature was constantly monitored with sensors connected to a data logger (EM50 Data Collection System, Decagon Devices Inc., Pullman, WA, USA) .Seeds were sown on a nursery potting soil (Huminsubstrat N3, Neuhaus, Klasmann-Deilmann, Geeste, Germany), composed of 90% peat, 10% clay; pH 6; NPK 14:16:18, 1.3 kg m^-3; conductivity 35 mS m^-1. On 18 June, uniform sized seedlings of lettuce at the 3-leaf stage were transplanted into individual plastic pots (14 × 14 cm) filled with peat-based compost (peat:vermiculite:perilte 1:1:1, v/v); the composition of the peat moss is given as follows (percentage on dry matter): organic carbon 40%, organic nitrogen 0.1%, organic matter 80%. At 8 and 10-leaf stages, a treatment with fungicide Ortiva (a.i. Azoxystrobin 23.2%, Syngenta Crop Protection S.p.A., Milano, Italy) at the dose of 0.08 mL m^-2 was applied.The experiment was arranged on a complete randomized block design. Two nutrient-deficiency conditions and two abiotic stressful conditions were imposed starting from 4 days after transplanting (DAT), i.e. no phosphorus fertilization (named 0_P), no nitrogen fertilization (0_N), limitation of the photosynthetic active radiation (PAR, range from 400 to 700 nm) (LR) and water availability constraint (WR), plus one unstressed CONTROL. Each treatment was replicated three times and each replication consisted in 49 pots (7 rows, 7 pots per row) for a total of 147 pots per treatment; to avoid edge effects, plants were daily re-randomized with the accuracy in maintaining the same sun orientation. All the plots with no-phosphorus limitation (CONTROL, 0_N, LR and WR) were fertilized with simple superphosphate at the rate of 40 kg P₂O₅ ha^-1. All the plots with no-nitrogen limitation (CONTROL, 0_P, LR and WR) were fertilized with two applications, at 4 and 7 DAT, at the total rate of 90 kg N ha^-1 with calcium nitrate (Ca(NO₃)2). In order to standardize the amount of calcium to the plants, 0_N treatment was fertilized with 156 kg ha^-1 of calcium oxide (as the commercial product Brexil Ca, Valagro S.p.a., Piazzano di Atessa, CH, Italy). At transplanting plants were fertilized with potassium chloride at the dose of 100 kg K₂O ha^-1, and KSC Mix (Timac Agro Italia, Milano, Italy) at the dose of 0.02 kg ha^-1 composed as follows: 15% water-soluble magnesium oxide (MgO); 28% water-soluble sulphurous anhydride (SO₃); 0.5% water-soluble boron (B); 0.5% water-soluble copper (Cu) chelated by EDTA; 2.5% water-soluble iron (Fe) chelated by EDTA; 2% water-soluble manganese (Mn) chelated by EDTA; 0.2% water-soluble molybdenum (Mo); 1.5% water-soluble zinc (Zn) chelated by EDTA.Shade treatments (LR) were accomplished using a shade net in order to obtain a 85% of reduction in PAR wavelengths. PAR intensity was hourly measured with a PAR Photon Flux Sensor (Decagon Devices Inc., Pullman, WA, USA), connected to a data logger (EM50 Data Collection system, Decagon Devices Inc., Pullman, WA, USA). Per cent of shading was determined by comparing the average PAR values of net with the average PAR values of the un-shaded treatments. Nets were wrapped around a rigid and removable structure placed above the vegetation so that they covered the incoming light from the top and sides to 5 cm below the bottom of the pots. To allow air circulation, light was not limited from below.Water stress (WR) was imposed by maintaining soil volumetric water content at 30% of water holding capacity (WHC) which corresponded to 0.240 and 0.072 m³ m^-3 for no-water stressed and water stressed plants, respectively. Water loss, due to evapotranspiration, was constantly monitored with soil moisture sensors installed in 5 randomly selected pots, for each replicates (EC-5, Decagon Devices, Inc., Pullman, WA, USA); the sensors were connected to a data logger - EM50 Data Collection system (Decagon Devices Inc., Pullman, WA, USA). The pots were manually re-watered with tap water (pH 7.2, EC 0.23 mS cm^-1) every day at 18:00 h.

With the exception of light reduction, the other kind of limitation negatively influenced lettuce fresh yield; nevertheless, the reduction of PAR availability induced a decrease in the content of the main investigated phenolic compounds resulting in a strong reduction of total phenolic content as well as antiradical activity. Conversely, the scarcity of N nutrition allowed to obtain the highest total polyphenols content (TPC) and TEAC (Trolox Equivalent Antioxidant Capacity), although no differences were found in terms of the main phenolic compounds. Drought seems to improve the accumulation of caffeic, caftaric and chicoric acids in the bound forms as well as TPC and antiradical activity of the same fractions, while the reduction in P fertilization did not significantly influence lettuce leaves composition in terms of phytochemicals.