Plant material and experimental conditions
The experiment was conducted at fields of the Tropical Products Company of Castanhal (Tropoc), located in the municipality of Castanhal, PA, Brazil (01º 17' 50” S, 47º 55' 20” W, 40 m altitude). The study was carried out in partnership with the Study Group on Water and Soil Engineering in the Amazon (GEEASA) from the Universidade Federal Rural da Amazônia (UFRA). The field is composed of 2-year-old seedlings from two cultivars of Piper nigrum L., namely cv. Clonada and cv. Uthirankotta. They were arranged in an area of 1930 m2. The experiment is set up in a randomized blocks design, in which each plot consisted of four plants per cultivar (Clonada and Uthirankotta), in a double row, with a spacing of 4.0 m between rows and 2.20 x 2.20 m among plants. The cv. Clonada was obtained from seed germination and in vitro of the Kuthiravally cultivar, an early fruit maturation material, which normally sprouts from December to February, with a period of pollination to fruit maturation of six months. It displays medium-sized foliage, purpure-colored sprouts, long ears with good filling, and medium-sized fruits, with an average production of 3 kg plant-1 (Fig. S1). On the other hand, cv. Uthirankotta is a well-established cultivar in the region. It holds a late fruit maturation cycle, normally the flowering from January to March, with a phase of six months from pollination to fruit maturation. It presents wide leaves, long ears containing several rows of fruits, dark-purple-colored sprouts, and medium-sized fruits, with an average production of 3 kg plant-1 (Fig. S1).
The soil classification is a dystrophic Yellow Argisol (medium texture), with a predominance of secondary vegetation (Cardoso Júnior et al. 2007). The soil chemical and acidity corrections were carried out by applying liming (3.7 t ha-1) and both organic and mineral planting fertilizers as previously described by Oliveira and Nakayama (2007). Plants were cultivated under global radiation (GR) of 1303 W/m2 (nearly 700 µmol de photons m-2 s-1), and daytime temperatures of 25.3 to 26.2 ± 2 °C (day/night), with relative air humidity between 60.5 and 86%. A total precipitation of 1,520 mm, with rainfall occurring in January (489 mm) and May (388 mm) and the lowest in February (99.8 mm) (Fig. S2). To confirm that all plants were kept at field capacity over the course of this experiment, the soil moisture was monitored with tensiometers installed at a depth of 20 cm-30cm. The tensiometers were positioned in line with the culture, 15 cm from the drippers. Readings were taken daily at 8:30 am. The analyses of gas exchange and water status were performed 24 months after the experiment was initiated. The yield was determined at the end of the harvest season. The entire experiment period comprised 36 months.
Yield and water-use efficiency
Fruits were harvested during their respective seasons for each cultivar (between July and October for cv. Clonada, and August to September for cv. Uthirankotta). Based on pepper fruits fresh weight, the yield (Y) was calculated as Y=P/A, where Y, as yield (kg ha-1); P, as production (kg); and A, as area (ha). The water balance over the experiment period was monitored and expressed in mm. Thus, the water-use efficiency (WUEyield) was accomplished through the relationship between black pepper fruit productivity (kg ha-1) and water consumption (mm) (Doorenbos and Kassam 1994), as WUEyield=Y/w, where: WUEyield, as water-use efficiency, kg ha-1 mm-1; Y, as total yield, kg ha-1; w, as volume of water applied, mm.
Gas exchange
Gas exchange parameters were determined using the infrared gas analyzer (LCpro-SD, ADC BioScientific Ltd, United Kingdom). Analyzes were performed on the third to fourth fully expanded leaf of the twelfth branch from the base. The analyses included net carbon assimilation rate (A), stomatal conductance (gs), transpiration (E), and intercellular CO2 concentration (Ci), which were measured under photosynthetically active radiation (PAR) of 1000 µmol (photons) m–2 s–1 and 400 ppm CO2 at leaf level (Yin et al. 2009). Instantaneous (WUEE) and intrinsic water-use efficiency (WUEgs) were estimated based on the ratio between A/E and A/gs, respectively. Response curves of net photosynthesis (A) to photosynthetically active radiation (PAR) were also carried out. For this, leaves were submitted to 12 points of light intensity, varying the photosynthetically active radiation from 0 to 1400 µmol photons m-2 s-1 via a light source (halogen bulb with 50W reflector) under a fixed concentration of 400 µmol mol-1 air (Ca) of CO2 in the chamber. The photosynthetic light-response curve (A/PAR curve) were adjusted as described by Lobo et al. (2013). From the A/PAR curve plots, light compensation point (Ic), light saturation point (Is), photosynthesis rate in saturating light (Asat), and maximum gross photosynthesis rate (Amax) were calculated by fitting the mechanistic model (Lobo et al. 2013).
Water status
Leaf water potential measurements were carried out at both predawn (Ψpd) and midday (Ψmd) using the Scholander pressure chamber (Scholander et al. 1965), model M 1505D (Pressure Chamber Instruments, PMS). The Ψpd was performed between 2:30 am and 5:30 am, while the Ψmd was carried out between 12:00 pm and 3:00 pm. Afterwards, the plant's hydraulic conductance (Kplant) was estimated as described by Kramer and Boyer (1995), and Avila et al. (2020), in which Kplant=E/([-(Ψmd – Ψpd)]); where Kplant, as hydraulic conductance of the plant (mmol H2O m-2 s-1 MPa-1); E, plant leaf transpiration (mmol H2O m-2 s-1); Ψmd, midday water potential (MPa); Ψpd, predawn water potential (MPa).
Parameters based on the pressure-volume (PV) curve (Ψw x RWC) were determined on the third to fourth fully expanded leaf of the twelfth branch from the base. Two leaves were collected per plot and block, totaling 6 leaves per cultivar. Leaves were cut under water and rehydrated overnight until Ψw reached nearly −0.1 MPa. Both leaf weight and Ψw were recorded using the Scholander pressure chamber and precision scale (de 0,001 g) over time during desiccation on the laboratory bench until Ψw stabilization. At least four to five points were collected before and after the turgor loss point for each leaf (Tyree and Hammel 1972). Except for the first weighing (considered as FW1) of each leaf per cultivar, the others were considered as fresh weight (FW2). The leaves were dried in an oven at 65 ºC, and the dry mass (DW) was recorded. Then, the relative water content (RWC) at each point was calculated as RWC (%) = (FW1− DW)∕FW2− DW) × 100. Finally, the osmotic potential at full turgor (Ψs100%), osmotic potential at turgor loss point (ΨsTLP), bulk elastic module (Ꜫv), relative water content at field capacity (RWC100%) and relative water content at turgor loss point (RWCTLP) were estimated based on recommendations of Cardoso et al. (2018) and Cardoso et al. (2020).
Growth Parameters
Total leaf area (TLA) was determined as described by Rhoads and Bloodworth (1964). For this, leaves were collected in the portion corresponding to ¼ of the plant height. The leaves were placed in paper bags and transferred to an oven at 65º C for 72 h. Afterwards, the leaves were weighed on an analytical scale and then TLA was estimated according to Reddy et al. (1989): TLA=FWleaf x SLA x 4. TLA, total average leaf area of a plant (m2); FWleaf, dry weight of leaves corresponding to ¼ of the plant height (kg); SLA, specific leaf area (g m2); 4, as a constant term. The specific leaf area (SLA) was determined as described by Hunt (1982).
Morpho-anatomical analyses
Morpho-anatomical analyses were carried out in the median region of the fourth fully expanded leaf of the twelfth branch (from base to apex) in 3-year-old plants. Samples were immediately fixed in FAA70 (70% formaldehyde-acetic acid-ethanol) in a ratio of 1:1:18 (Johansen 1940) for 24 h and subsequently stored in 70% alcohol. This material was embedded in methacrylate (Historesin-Leica) following the manufacturer's recommendations. Subsequently, samples were transversely sectioned (5 μm thick) with an automatic feeding rotary microtome (model RM2155, Leica Microsystems Inc., Heidelberg, Germany) and stained with toluidine blue (O'Brien et al. 1964).
To assess leaf thickness (LT) mesophyll thickness (MT), adaxial epidermis thickness (AdET), and abaxial epidermis thickness (AbET), the fragments of the median region of the leaves were immersed in 5% sodium hydroxide for 48h. After 24 h, fragments were washed with water and transferred to lactic acid before immersing in a water bath at 95ºC until they became translucent. Layers were then photographed with an optical microscope (Motic model) coupled to a digital camera. Finally, stomatal density (SD), stomatal index (SI), stomatal polar diameter (SPD), stomatal equatorial diameter (SED), and stomatal functionality (SF) were analyzed based on procedures previously described by Batista-Silva et al. (2019) using ImageJ software (Schneider et al. 2012), in at least 10 different fields of 0.053 mm2 per leaf.
Data analysis
Data were collected from experiments using a randomized block design, with each plot containing four plants per cultivar. The homoscedasticity and normality of the data were verified to meet assumptions surrounding ANOVA. If ANOVA showed significant effects, a Tukey test (P < 0.05) was used to determine differences between cultivars. Moreover, a principal component analysis (PCA) was carried out considering both cultivars and variables. Statistical procedures were performed using R software (v. 4.3.2; R Core Team 2023).