The trial was organized in a completely randomized design with three treatments and five repetitions. For the analysis of the results, an analysis of variance was applied by the F test, with the averages compared by the Tukey test at 5 % probability, using the statistical analysis system software (SAS) v. 9.2. The graphics were created with the aid of the 11.0 Sigma Plot v software.

Water, in plants, moves from areas with higher Ψ_w (less negative) to areas with lower Ψ_w (more negative). Plant cells seek thermodynamic balance with the surrounding environment by absorbing or losing water (Taiz and Zeiger, 2002). The strategies of plants to tolerate water restriction (WR), include several physiological and biochemical processes, including the reduction of Ψ_lw caused by the closure of stomata (Nardini et al., 2001), which was demonstrated in this test (Figure 3). Figure 1 show that the plants in sheet 1 maintained water potential values between -0.24 MPa and - 0.27 MPa during the performed evaluations. After the imposition of WR, the plants from irrigation sheets 2 and 3 showed values between -0.58 MPa and -0.65 MPa, and -0.80 MPa and -0.93 MPa, respectively. These results prove that, in bell pepper plants, the more negative the water potential is, the values of the components of gas exchange (Figures 2, 3 and 4) and fruit production (Table 5) are considerably reduced.

Figure 1

Figure 2

Figure 3

Figure 4

In figure 2, it can be observed that, when plants were submitted to WR, the photosynthesis gradually decreased as the water decreased in the soil. Similar results to this study, in which both the photosynthetic rate and the stomatal conductance decrease after the imposition of WR, were reported by Kulkarni and Phalke (2009). When working with different cultivars of hot pepper under conditions of protected environment and water restriction. Likewise, in seedlings (grafted and non-grafted) of mini-watermelon, these gas exchange components have decreased in parallel with the severity of water restriction (Rouphael et al., 2008).

The gs is a parameter that indicates the opening and closing levels of the stomata, depending on the availability of water for the plant. Thus, WR promotes the opening reduction of stomatal pores, since the amount of water for the guard cells is reduced. In this test it was proved that the gs decreased considerably due to the severity of the WR (Figure 3), in response to the adaptation mechanisms of the plant itself to avoid dehydration, thus regulating its water status. Similarly, Paiva et al. (2005) reported that, in bean leaves, the gs decreased significantly in the flowering and grain filling stages in the treatment with less water availability.

When plants have good water availability in the soil (field capacity) they have high transpiratory rates, facilitating water movement between the soil-plant-atmosphere. However, as the water content in the soil decreases as a result of the treatments, the plants decrease their transpiratory rate to reduce the loss of water in the form of steam. In this test (Figure 4) plants under water restriction showed lower values of E when compared to plants without water restriction. At 65 and 80 DAT, the plants of irrigation sheets 2 and 3 decreased their transpiratory rate respectively. At 95 DAT, the sheet 3 irrigation plants achieved transpiration rates similar to those of the irrigation sheet 2. This behavior, a result of WR, maintained a direct relationship with what happened in stomatal conductance (Figure 3), resulting in the closure of stomata in plants with less water availability as a strategy to avoid the excessive loss of water from plant tissues. Results similar to those in this trial were reported by Ferrara et al. (2011), who observed that when submitting bell pepper plants to WR the physiological variables, such as liquid CO₂ assimilation, stomatal conductance and transpiration significantly decreased in the vegetative and reproductive phases when compared to the control plants.

In bell pepper plants, irrigation sheets 2 and 3 caused a decrease in the isotopic discrimination values of δ¹³C to 80 DAT. In the last evaluation (95 DAT), only irrigation sheet 3 showed statistically less discrimination of δ¹³C in relation to other treatments. These results agree with those reported in cotton crops (Brito et al., 2014) and sunflower (Adiredjo et al., 2014), in which the isotopic discrimination of δ¹³C decreased in parallel with the severity of WR. That is, plants with scarce water availability in the soil have less isotopic discrimination of carbon. In this sense, the lower discrimination of δ¹³C in plants with less water availability in the soil after 80 DAT may has been influenced by the lower CO₂ pressure (Figure 2) due to stomatal closure (Figure 3). That is a factor contributes to the discrimination during diffusion and carboxylation (Farquahr et al., 1982).

The activity of the nitrate reductase enzyme has been used as an indicator of stress and changes linked to factors that modulate plant growth (Dubey et al., 2021; Carelli et al., 1996). Table 2 shows that NR activity was also reduced in pepper plants after the imposition of water treatments (65 and 80 DAT). In the last evaluation (95 DAT), the irrigation 3 sheet plants differed statistically from irrigation 1 and 2 sheets plants. These results of the lower NR values under WR treatments are possibly related to the decrease in the photosynthetic rate (Figure 2), with nitrogen participating in the formation of chlorophyll, which is essential in the process of photosynthesis. Similarly, cotton plants (Gossypium hirsutum L.) under water stress have been reported to decrease NR activity in parallel with the decrease in leaf water potential (Marur et al., 2000), a fact that was also evidenced in the present trial (Figure 1).

Table 1

Treatment	50 DAT	65 DAT	80 DAT	95 DAT
	Carbon isotopic discrimination (δ13C)
S1	-29.3 a	-30.1 a	-29.6 b	-29.3 b
S2	-29.3 a	-30.2 a	-28.4 a	-28.6 b
S3	-29.7 a	-29.8 a	-27.7 a	-27.1 a
P	0.4702	0.6332	0.0002	0.0020
CV (%)	2.0	2.4	1.7	2.7

Means followed by the same letter in the column do not differ statistically from each other by the Tukey test, at the level of 5 % probability. DAT = days after transplant; S1: Ψ_m = 10 to 15 kPa, S2: Ψ_m = 34 to 40 kPa and S3: Ψ_m = 54 to 60 kPa. CV (%) = Percentage of the coefficient of variation.

Table 2

Treatments	50 DAT	65 DAT	65 DAT	95 DAT
	Nitrate reductase enzyme activity (nM NO2 g-1 MF h-1)
S1	2309.2 a	2127.8 a	1878.4 a	1038.9 a
S2	2385.6 a	1686.2 b	1491.1 b	833.3 a
S3	2306.2 a	1554.8 b	1365.1 b	247.1 b
P	0.7219	0.0004	0.0098	0.0001
CV	7.45	9.42	14.37	17.31

Means followed by the same letter in the column do not differ statistically from each other by the Tukey test, at the level of 5% probability. DAT = days after transplant; S1: Ψ_m = 10 to 15 kPa, S2: Ψ_m = 34 to 40 kPa and S3: Ψ_m = 54 to 60 kPa. CV (%) = percentage of the coefficient of variation.

The increase in the enzymatic activity of superoxide dismutase is a defense response of plants under stress conditions, since it is the first response line of the antioxidative system (Del Río et al., 2018). In Table 3 can be observed that, at 80 DAT, the plants in sheet 3 showed the highest SOD activity. In the last evaluation (95 DAT) the plants of sheets 2 and 3 had a significantly similar behavior, but with higher values to plants without water restriction. On the other hand, Yuan et al. (2016) worked with tomatoes under four soil moisture levels during two cycles under protected environment conditions, noting that the SOD values increased according to the severity of WR in the growth stages.

Table 3

Treatment	50 DAT	65 DAT	80 DAT	95 DAT
	Superoxide dismutase enzyme activity (IU μg protein-1)
S1	1.2 a	1.2 a	1.3 b	1.6 b
S2	1.1 a	1.6 a	1.5 b	2.2 ab
S3	1.0 a	1.9 a	2.5 a	2.7 a
P	0.779	0.147	0.0003	0.0149
CV (%)	18	13	20	11

Means followed by the same letter in the column do not differ statistically from each other by the Tukey test, at the level of 5% probability. DAT = days after transplant; S1: Ψ_m = 10 to 15 kPa, S2: Ψ_m = 34 to 40 kPa and S3: Ψ_m = 54 to 60 kPa. CV (%) = percentage of the coefficient of variation.

However, in the fruit maturation stage, the plants of the control treatment (75 to 80 % of field capacity) and those of the treatment considered as mild water deficiency (55 to 60 % of the field capacity), showed significantly similar SOD values. Thus, according to the results reported by Yuan et al. (2016) and those of this test in bell pepper plants, the SOD values depend on the time of application of WR, severity, and phenological stage of the plants.

Several studies show that plants under stress conditions (biotic and abiotic) present accelerated production of reactive oxygen species (ROS), which can be toxic to the metabolism of plants or inhibit their growth. The dismutation of reactive oxygen species is performed by the antioxidative system, which comprises numerous enzymes and low molecular weight compounds (Banerjee and Roychoudhury, 2017). The catalase enzyme is the main enzyme that acts in the decomposition of hydrogen peroxide (H₂O₂) to water (H₂O) and molecular oxygen (O₂). In this trial, CAT activity (Table 4) was activated at 80 and 95 DAT in plants under water restriction, which showed significantly similar values between treatments, however, with values higher than those of irrigation sheet 1. Similar results to those obtained in this study were reported by Abid et al. (2017) in two greenhouse cultivars of fava beans (Vicia faba L.) in the winter period under three irrigation sheets (90, 60 and 30 % of field capacity). The results obtained for both cultivars (tolerant and susceptible) under the treatment considered as severe water restriction, showed significantly higher CAT activity when compared to the plants of the control treatment.

Table 4

Treatments	50 DAT	65 DAT	80 DAT	95 DAT
	Activity of the catalase enzyme (μKat μg protein-1)
S1	44.8 a	47.5 a	50.6 b	56.9 b
S2	46.7 a	60.3 a	64.5 ab	70.7 ab
S3	45.7 a	71.5 a	75.7 a	84.5 a
P	0.983	0.093	0.002	0.001
VC (%)	20	16	12	13

Means followed by the same letter in the column do not differ statistically from each other by the Tukey test, at the level of 5% probability. DAT = days after transplant; S1: Ψ_m = 10 to 15 kPa, S2: Ψ_m = 34 to 40 kPa and S3: Ψ_m = 54 to 60 kPa. CV (%) = percentage of the coefficient of variation.

On average, 7.4 fruits were collected per plant for irrigation sheet 1, with commercial productivity of 31.1 t ha^-1 (Table 5). Plants of sheet 2 and 3 presented lower commercial productivity, in the order of 11.2 and 2.7 t ha^-1, respectively. It should be noted that the harvest period lasted 28 days. On the other hand, Carvalho et al. (2016) worked with different soil water tensions in the reproductive phase of red pepper cv. Konan, the authors reported production for the control treatment of 5.85 fruits per plant, equivalent to commercial production of 12.03 t ha^-1 in protected environment conditions, with a 42-day harvest period. Thus, and according to the results obtained in this trial and those reported by other authors (Frizzone et al., 2001; Dermitas and Ayas, 2009), it is concluded that the decrease in water in the soil strongly affects the yield of peppers. In addition, it is emphasized that this difference in fruit productivity may vary according to the cultivar, season and climatic conditions.

Table 5

Treatments	NFP	AFM (kg)	TFMP (kg)	TP (ha-1)	% NMF	PC (t ha-1)
S1	7.4	0.183	1.354	33.9	8	31.1
S2	5.4	0.128	0.691	17.3	35	11.2
S3	4.0	0.70	0.280	7.0	62	2.7

NFP = number of fruits per plant; AFM = average fruit mass; TFMP = total fruit mass per plant; TP = total production; % NMF = percentage of non-marketable fruits; CP = commercial production; S1: Ψ_m = 10 to 15 kPa, S2: Ψ_m = 34 to 40 kPa and S3: Ψ_m = 54 to 60 kPa.