Potency of Titanium Dioxide Nanoparticles, Sodium Hydrogen Sulfide and Salicylic Acid in Ameliorating the Depressive Effects of Water Deficit on Periwinkle Ornamental Quality

Abstract

In this study, the optimal concentration of sodium hydrosulfide (NaSH), salicylic acid (SA), and titanium dioxide nanoparticles (TiO₂NPs), and their relative effectiveness on alleviating the adverse effects of water deficit on ornamental quality, were investigated in periwinkle. Plants were cultivated under three water deficit levels (80, 50, and 20% available water content) and received two foliar applications of TiO₂NPs (0, 0.5 and 1 mM), NaSH (0.5 and 1 mM), or SA (1 and 2 mM). Water deficit deteriorated ornamental quality, amplified the risk of buckling (lower stem strength) and suppressed longevity. It decreased both light interception (leaf area) and carbon assimilation. Besides impaired hydration status, water-stressed plants underwent oxidative damage as indicated by reduced chlorophyll content, elevated membrane degradation, and lipid peroxidation. Spray treatments improved all traits, besides stem strength and proline content. Additionally, they enhanced carotenoid content and the activities of catalase and peroxidase. Their relative effectiveness (TiO₂NPs > NaSH > SA) and optimal concentration (i.e., 0.5 mM (TiO₂NPs, NaSH), and 1 mM (SA)) was independent of water deficit level. In conclusion, this study provides practices for improved ornamental quality and longevity independently of water availability, with their positive effect being stronger under irregular or limited water supply.

1. Introduction

Water deficit is globally by far the main factor limiting plant growth and productivity [1]. The areas with limited water resources are continuously expanding, owing to the adverse impacts of on-going climate change, including reduced precipitation and increased evaporation [2]. In this perspective, developing mitigation strategies for enhancing yield under water limiting environments evidently remains in the spotlight for further endeavors [3,4].

The water deficit-induced impairment of plant growth and productivity is modulated via a broad range of underlying processes [5]. For example, water deficit typically downregulates carbon assimilation via reductions in both light capture (leaf area) and chlorophyll content [6]. Water deficit additionally causes osmotic stress, which without the respective adjustment (e.g., via proline formation) impairs enzymatic activity and triggers macromolecules deformity [3,7]. Moreover, the equilibrium between the generation and detoxification of reactive oxygen species (ROS) is commonly disturbed when water is limiting [3,5]. Catalase (CAT) and peroxidase (POD) are main ROS scavenging enzymes, while carotenoids are major non-enzymatic antioxidant metabolites [8,9]. Excessive generation of ROS stimulates a range of damaging effects, such as lipid peroxidation and coupled electrolyte leakage [10,11].

Several factors alleviating the adverse effects of water deficit have been previously documented, including exogenous application of sodium hydrosulfide (NaSH, [12]), salicylic acid (SA, [13]), and more recently titanium dioxide (TiO₂) nanoparticles (NPs, [4]). These effects vary depending on the species, and they are concentration dependent [14,15,16]. Although the mitigating effects of these compounds have been addressed, knowledge on their relative effectiveness remains scant. In this perspective, which intervention is most effective in masking the adverse effects of water deficit evidently stays understated. Moreover, the majority of previously conducted studies on these alleviating interventions has been focused on a single water deficit level [12,13]. In this perspective, it has not been adequately investigated whether or not the optimal concentrations, and their relative effectiveness, vary based on water deficit level.

The objectives of this study were (1) to determine the optimal concentration of NaSH, SA, and TiO₂ NPs for ornamental quality and (2) to compare their relative effectiveness across different water deficit levels. Alongside to biomass accumulation, carbon assimilation, stomatal anatomical features (index, density, size, shape), membrane stability and major antioxidant defense components were investigated. Periwinkle (Catharanthus roseus (L.) G. Don) was employed as model species, since it is popular as either pot or landscape plant [17]. Noteworthy, periwinkle also has an important medicinal value, as it represents a critical source of compounds employed for chronic diseases including cancer [3,17].

2. Materials and Methods

2.1. Plant Material and Growth Conditions

Seeds of periwinkle (Catharanthus roseus) ‘Pacifica XP Really Red’ were obtained from a commercial supplier (PanAmerican Seed Co., Chicago, IL, USA), and were selected from a single lot based on size homogeneity. They were sterilized (5 min) by soaking in sodium hypochlorite solution (2%, v/v), and then washed (3 min) with double-distilled water. Seeds were then planted in 2 L pots (diameter (top) × diameter (bottom) × height = 15 × 13 × 15 cm; Figures S1–S3)) containing a mixture of soil, sand (particle size of 0.8–2 mm) and composted livestock manure (1:1:1, v/v). Two seeds were planted at the central area of each pot. Prior to the appearance of the first leaf, seedlings were thinned by keeping the most vigorous one per pot.

Prior to potting, growth media had been sieved (6 mm). Pots were filled with equal weight of growth media (300 g per pot). At transplanting, substrate water content (2.66 ± 0.02 g g⁻¹) was uniform within and across pots. Substrate water content was computed by employing extra pots (not including experimental plants) based on the difference between saturated and dry substrate mass weights [5]. Saturation state was implemented by extra irrigation followed by a 48 h period of rest phase, where pots were enclosed in (non-perforated) polyethylene bags (size: 40 × 40 cm; film thickness: 0.17 mm). The dry weight was established after placing the substrate in an oven (105 °C) for 24 h.

Afterwards, pots were randomly placed in a greenhouse, located in the central-west part of Iran (Khorramabad, 33° N). A plant density of 20 pots m⁻² was implemented. Using a completely randomized design, twenty-one treatments (3 water deficit levels × 7 spray treatments) were laid out as a factorial experiment.

Three water deficit levels were realized, including irrigation to 80, 50, and 20% available water content. The available water content expresses the amount of water which is available for plant uptake, and corresponds to the difference between field capacity and permanent wilting point. The field capacity of the growth media was determined by using the retention curve method, and corresponds to the water content at a matric potential of −0.1 kPa. The permanent wilting point was also determined by the same method, and denotes the water content at a matric potential of −1.5 MPa. At this tension, growth media still contains some water, which cannot be absorbed by plant roots. Each water deficit level was sustained by daily regulating irrigation amount. Considering the regular irrigation mode and the large growth medium volume (2 L), day-to-day difference in substrate water content is expected to be rather low [5].

In each water deficit level, plants received exogenous application of TiO₂ NPs (0, 0.5 and 1 mM), NaSH (0.5 and 1 mM), or SA (1 and 2 mM) via foliar spray. The suitable concentration span was chosen on the basis of both a comprehensive literature survey [4,12,13,14,15,16], and a pre-experiment. Each time before application, TiO₂ NPs solution was ultrasonically homogenized using a sonicator (UP100H, Hielscher Ultrasonics, Germany) [10,11,18]. Two spray applications were performed. At the time of the first application, plants were at the four-leaf stage. The application was repeated one week later.

Plants were grown under naturally fluctuating light conditions with a mean daily light integral of 16.7 ± 0.3 mol m⁻² day⁻¹. Average air temperature was 21.2 ± 1.8 °C, while average relative air humidity was 65 ± 2%.

Measurements were undertaken in whole-plant and leaf levels. For the latter, selected leaves were fully expanded, and had developed under direct light. Replicate leaves were sampled from different plants. Leaves were harvested at the onset of the light period (06:00–07:00 h). In growth and biomass allocation measurements, the time between sampling and the start of the evaluation was less than 15 min. In the remaining evaluations, samples were deposited in vials, flash-frozen in liquid nitrogen, and delivered to a freezer (−80 °C) for storage. Sampling was performed at the end of the cultivation period. In all determinations, four replicates were assessed per treatment.

2.2. Characterization of TiO₂ Nanoparticles

TiO₂ NPs in the anatase form (>99% purity) were obtained from an industrial supplier (Iranian Nanomaterial Pioneers Company, Mashhad, Iran). TiO₂ NPs were spherical in shape with a mean particle size of 15–20 nm. The specific surface area of TiO₂ NPs was 200–240 m² g⁻¹, while purity exceeded 99%. The structural properties of TiO₂ NPs were characterized by using transmission electron microscopy [18]. The crystal properties of TiO₂ NPs were investigated by X-ray diffraction, validating that all NPs were in the anatase form.

2.3. Flower Induction and Intact Flower Bud Longevity

Treatment effects on the initiation of flowering, flower bud opening and intact (on plant) flower bud longevity were documented. Time to flowering (also referred to as time to visible bud) was recorded as the period from sowing to the appearance of the flower bud (i.e., flower bud length ≈ 0.5 cm). Intact (on plant) flower bud longevity was computed as the period between opening and wilting (i.e., petal turgor loss) of the flower bud.

2.4. Plant Growth, Morphology, and Biomass Allocation

Determinations included number of lateral branches, lateral branch length, main stem length, main stem diameter (assessed midway along its length), flower diameter (the mean of the largest diameter and the one perpendicular to it), number of leaves, and leaf area. For measuring leaf area, leaves were scanned (HP Scanjet G4010, Irvine, CA, USA) and then assessed by employing the Digimizer software (version 4.1.1.0, MedCalc Software, Ostend, Belgium) [6].

Prior to the assessment of root traits, the pot was placed in a vessel containing water for 1 h. After removing the growth medium from the roots by gentle washing, root volume was determined by using the volume-displacement technique [19]. Roots were submerged in a cylinder containing water. Root volume was afterwards assessed by evaluating the volume of water which was relocated by the roots. Root diameter corresponded to the diameter of the largest sphere, which fitted into the root and contained it. Root length was regarded as the length from the shoot-to-root junction to the tip of the primary root.

Main stem, lateral branch, leaf, flower, and root (fresh and dry) masses were also recorded (±0.001 g; Mettler ME303TE, Giessen, Germany). For dry weight evaluation, samples were moved to an oven (80 °C) for 72 h [19]. By employing dry mass, specific leaf area (leaf area per leaf mass), flower mass ratio (flower mass per plant mass), leaf mass ratio (leaf mass per plant mass), and root to shoot ratio (root mass per aboveground mass) were computed. The ratio of plant dry weight of each treatment to plant dry weight of the control (referred to as stress tolerance index [7]), and water use efficiency (plant dry mass per water used [20] were calculated. The strength (mass per length) and tissue density (mass per volume) of the main stem were also calculated, as indices of its sensitivity to buckling [21,22].

2.5. Gas Exchanges Features

Leaf gas exchange features were in situ evaluated. Measurements were conducted via a portable photosynthesis system (CI-340; CID, Inc., Camas, WA, USA). During measurements, air temperature (22 °C), relative air humidity (50%), incoming air CO₂ concentration (400 μmol mol⁻¹) and light intensity (200 μmol m⁻² s⁻¹) were maintained stable. Measurements were undertaken 2 h following the onset of the light period to secure uniform pattern of stomatal conductance [23].

2.6. Chlorophyll and Carotenoid Contents

Leaf chlorophyll content is critical for photosynthesis, while carotenoids are important non-enzymatic antioxidants [8,9]. Following fine chopping, leaf material (0.1 g) was homogenized with 10 mL of 100% acetone. Next, the extract was centrifuged (14,000× g for 20 min), and the supernatant was collected. Owing to the light sensitivity of chlorophyll, the extraction was conducted under darkness [10,11]. The extract was then assessed by using a spectrophotometer (Mapada UV-1800; Shanghai Mapada Instruments Co., Ltd., Shanghai, China). Leaf chlorophyll and carotenoid contents were eventually computed as described by Lichtenthaler and Wellburn (1983) [24].

2.7. Stomatal and Epidermal Cell Anatomical Features

Stomatal and epidermal cell features were determined. The sampling area was assigned in the middle of the lamina base and apex, as well as halfway the midrib and margin [25,26]. The abaxial surface was covered with nail polish. Next, transparent tape was employed for peeling off the dried polish. The dried polish was placed on a microscopic slide, and observations were made with a light microscope. Images were obtained by the Omax software (ver. 3.2, Omax Corp., Kent, WA, USA). Stomatal length and width were assessed on 10 stomata (magnification × 100), while stomatal and epidermal cell densities were evaluated on five fields of view per replicate leaf [27]. Stomatal index (i.e., stomatal number per total (stomatal and epidermal cell) number) was calculated [27]. Stomatal size was defined as stomatal length multiplied by stomatal width [22,27]. Stomatal area per leaf area was calculated (stomatal size × stomatal density). Based on this, epidermal cell area per leaf area was computed (10⁶—stomatal area per leaf area). Mean epidermal cell size was then calculated by dividing the epidermal cell area per leaf area by epidermal cell density. Image processing was handled with ImageJ software (ver. 1.37, Wayne Rasband/NIH, Bethesda, MD, USA).

2.8. Water Status

Leaf water status was evaluated by determining relative water content (RWC). Sampling was performed 3 h following the onset of the photoperiod [26]. Fresh weight was obtained (±0.0001 g; Mettler AE 200, Giessen, Germany) immediately after excision. Next, leaves were placed on double-distilled water in a 9 cm Petri dish, which was closed with a lid. After incubation (24 h), the turgid (saturated) weight was assessed. Eventually, dry weight (72 h at 80 °C) was evaluated. RWC was computed as described by Taheri-Garavand et al. (2021) [23].

2.9. Proline Content

Proline participates in cell osmotic regulation via lowering water potential, and through this way, it protects both enzymatic activity and macromolecules’ structure [7]. On this basis, leaf proline content was determined. Leaf discs (0.5 g) were homogenized, and injected in 3% (w/v) aqueous sulphosalicylic acid (10 mL). The obtained extract was filtered (Whatman No. 2), and the filtrate (2 mL) was mixed with equal volumes of acid-ninhydrin (2 mL) and glacial acetic acid (2 mL). The resultant solution was exposed to 100 °C (1 h). The reaction mixture was extracted with toluene (4 mL), and the chromophore containing toluene was then aspirated from liquid phase. After reaching 25 °C, the absorbance was evaluated at 520 nm by using a spectrometer (Mapada UV-1800, Shanghai. Mapada Instruments Co., Ltd., Shanghai, China). Proline concentration was measured via a calibration curve [5].

2.10. Electrolyte Leakage

The relative ion content in the apoplastic space, obtained as an index of membrane stability, was assessed by determining electrolyte leakage [10,11]. Leaf discs (1 cm²) were washed 3 times (3 min) with double-distilled water (to take away surface-adhered electrolytes), and then placed on double-distilled water (10 mL). Vials were placed on a shaker at room temperature for 24 h. Then, electrolyte leakage in the obtained solution was determined by using a conductimeter (Crison 522, Crison Instruments, S.A., Barcelona, Spain). Next, samples were autoclaved at 120 °C for 20 min. After reaching 25 °C, total conductivity was assessed. Per replicate leaf, four discs were evaluated.

2.11. Lipid Peroxidation

Malondialdehyde (MDA) content, obtained as an index of lipid peroxidation level, was assessed by using the thiobarbituric acid reactive substance assay [10,11]. Leaf discs (0.1 g) were homogenized, and then placed in 5 mL of 20% (w/v) trichloroacetic acid and 0.5% (w/v) thiobarbituric acid. The suspension was then centrifuged (6000× g for 15 min). The resultant solution was exposed to 100 °C (25 min). After reaching 25 °C, the precipitate was obtained by centrifugation (6000× g for 5 min). The amount of MDA was estimated by the absorbance at 532 nm after deducting the absorption at 450 and 600 nm (Mapada UV-1800; Shanghai Mapada Instruments Co., Ltd., Shanghai, China). For the computation, an extinction coefficient (156 mmol MDA L⁻¹ cm⁻¹) was employed [10,11]. Per replicate leaf, four discs were evaluated.

2.12. Enzymatic Activity

The activity of CAT was assayed as described by Ahmadi-Majd et al., 2022a [10]. Leaf material (0.3 g) was ground with a mortar and pestle in the presence of liquid nitrogen, homogenized by 1.5 mL of potassium phosphate buffer (including 1 mM EDTA and 2% polyvinylpyrrolidone), and eventually centrifuged (14,000× g for 20 min) at 4 °C. The activity of CAT in the supernatant was evaluated by recording the decline in absorbance at 240 nm for 2 min (intervals of 10 s) in a reaction mixture including potassium phosphate buffer and hydrogen peroxide. For the computation, an extinction coefficient (39.4 M⁻¹ cm⁻¹) was employed. The activity of CAT was expressed as μmol of hydrogen peroxide reduced min⁻¹ g⁻¹ tissue.

The activity of POD was determined as described by Ahmadi-Majd et al., 2022a [10]. Leaf material (0.3 g) were ground with a mortar and pestle in the presence of liquid nitrogen, homogenized by 1.5 mL of 50 mM potassium phosphate buffer (pH 7.0), and eventually centrifuged (14,000× g) and 4 °C for 20 min. The activity of POD in the supernatant was evaluated by recording the decline in absorbance at 470 nm for 2 min (intervals of 10 s) in a reaction mixture including potassium phosphate buffer, guaiacol, and hydrogen peroxide. For the computation, an extinction coefficient (26.6 mM⁻¹ cm⁻¹) was employed. The activity of POD was expressed as μmol of hydrogen peroxide reduced min⁻¹ g⁻¹ tissue.

2.13. Statistical Analysis

Data were submitted to analysis of variance by using SPSS 23 (SPSS Inc., Chicago, IL, USA). A two-way ANOVA was applied, with water deficit level serving as the main factor and spray treatment as the split factor. Data was initially checked for normality (Shapiro–Wilk test) and homogeneity of variances (Levene’s test). Eventually, estimated least significant differences (LSD) of treatment effects were computed (p = 0.05).

For the 21 experimental units, eigenvalues were computed and the most contributing variables for each dimension were determined. The first two eigenvalues accounted for 72% of the total variance, and were kept to create the principal components. A principal component analysis (PCA) was carried out to denote correlations across the water deficit levels, and spray treatments to principal components. Individuals were clustered (by distinct color) and variables by their involvement to the principal components (gradient colors). A correlation plot was also computed in order to depict positive and negative associations across the variables inquired. The “corrplot”, “FactoMineR”, “factoextra” and “readxl” libraries were employed under the R-studio integrated development environment (RStudio suite V 1.2.5033, Boston, MA, USA).

3. Results

3.1. Gas Exchange Features

As growth medium available water content decreased (i.e., water deficit became more intense), gas exchange features (transpiration rate, stomatal conductance, internal CO₂ concentration, and photosynthesis rate) declined (

Table 1. Gas exchange features of periwinkle ‘Pacifica XP Really Red’ plants receiving exogenous application of titanium dioxide nanoparticles (TiO₂ NPs; 0, 0.5 and 1 mM), sodium hydrosulfide (NaSH; 0.5 and 1 mM), or salicylic acid (SA; 1 and 2 mM) under different watering levels (80, 50 and 20% available water content) during cultivation. Means followed by different letters indicate significant differences (p = 0.05; comparison in columns). Values are the mean of four replications.

Available Water Content (%)	Compound	Concentration (mM)	Transpiration Rate (mmol m−2 s−1)	Stomatal Conductance (mol m−2 s−1)	Internal CO2 Concentration (ppm)	Photosynthesis Rate (μmol m−2 s−1)
80	–	0.0	3.21 e	0.33 i	428 fgh	3.21 k
	TiO2 NPs	0.5	5.81 a	1.15 a	650 a	10.4 a
		1.0	5.13 b	0.90 b	488 c	8.20 b
	NaSH	0.5	4.19 c	0.80 c	483 c	6.19 def
		1.0	3.80 d	0.59 d	435 efg	5.83 g
	SA	1.0	3.87 d	0.43 fg	446 e	5.90 efg
		2.0	3.21 e	0.34 i	441 ef	5.20 h
50	–	0.0	1.37 j	0.25 j	351 j	2.26 m
	TiO2 NPs	0.5	2.70 f	0.51 e	535 b	6.77 c
		1.0	2.21 g	0.45 f	482 cd	6.27 d
	NaSH	0.5	1.74 h	0.43 fg	425 gh	6.21 de
		1.0	1.52 i	0.41 g	394 i	5.58 fg
	SA	1.0	1.56 i	0.37 h	414 h	5.97 d–g
		2.0	1.40 j	0.34 hi	432 efg	5.21 h
20	–	0.0	0.30 o	0.15 m	291 k	1.23 n
	TiO2 NPs	0.5	1.22 k	0.25 j	479 cd	5.71 g
		1.0	0.68 l	0.22 jk	467 d	4.69 i
	NaSH	0.5	0.63 lm	0.20 kl	439 efg	4.15 j
		1.0	0.53 n	0.19 l	445 e	3.55 kl
	SA	1.0	0.58 mn	0.19 kl	424 gh	3.77 k
		2.0	0.49 n	0.18 lm	389 i	2.55 m
F-Value			119.85	163.59	41.49	36.86
p-Value			<0.001	<0.001	<0.001	<0.001

). Across water deficit levels (80, 50, and 20% available water content), spray treatments generally improved gas exchange features. In doing so, TiO₂ NPs and SA were the most and least effective, respectively. In all three compounds under study, the lowest employed concentration was the most effective (i.e., 0.5 mM (TiO₂ NPs, NaSH), and 1 mM (SA)).

Across the 21 treatments, transpiration rate was positively correlated with stomatal conductance (R² = 0.81; see also Figure 1). The same was noted between internal CO₂ concentration and photosynthesis rate (R² = 0.70).

3.2. Flower Induction and Intact Flower Bud Longevity

As growth medium available water content decreased, time to flowering, time to open flower, time to wilting, and intact flower bud longevity declined (

Table 2. Time to flowering (also referred as time to visible bud), time to open flower, time to wilting (i.e., petal turgor loss), and intact (on plant) flower bud longevity [(time to wilting) − (time to open flower)] of periwinkle ‘Pacifica XP Really Red’ plants receiving exogenous application of titanium dioxide nanoparticles (TiO₂ NPs; 0, 0.5 and 1 mM), sodium hydrosulfide (NaSH; 0.5 and 1 mM), or salicylic acid (SA; 1 and 2 mM) under different watering levels (80, 50 and 20% available water content) during cultivation. Means followed by different letters indicate significant differences (p = 0.05; comparison in columns). Values are the mean of four replications.

Available Water Content (%)	Compound	Concentration (mM)	Time to			Intact Flower Bud Longevity (d)
			Flowering (d)	Open Flower (d)	Flower Wilting (d)
80	–	0.0	64.0 ef	77.0 ef	84.0 g	7.00 g
	TiO2 NPs	0.5	68.5 a	85.0 a	98.0 a	13.0 a
		1.0	67.0 bc	82.5 b	93.5 b	11.0 bc
	NaSH	0.5	66.3 bc	82.0 b	91.5 c	9.50 e
		1.0	65.0 de	79.0 cd	90.0 d	11.0 bc
	SA	1.0	66.3 bc	79.0 cd	90.3 cd	11.3 b
		2.0	66.0 cd	78.0 de	88.0 e	10.0 cde
50	–	0.0	62.3 g	74.0 h	78.0 i	4.00 h
	TiO2 NPs	0.5	67.3 b	80.0 c	89.5 d	9.50 e
		1.0	66.0 cd	77.0 ef	87.8 e	10.75 bcd
	NaSH	0.5	66.0 cd	78.0 de	87.8 e	9.75 de
		1.0	65.0 de	76.0 fg	86.0 f	10.0 cde
	SA	1.0	65.0 de	78.0 de	87.0 ef	9.00 ef
		2.0	64.3 ef	75.0 fg	86.0 f	10.0 cde
20	–	0.0	59.3 h	70.0 i	72.0 j	2.00 i
	TiO2 NPs	0.5	66.0 cd	78.0 de	86.0 f	8.00 fg
		1.0	65.0 de	76.5 f	84.0 g	7.50 g
	NaSH	0.5	65.0 de	75.0 gh	83.0 gh	8.00 fg
		1.0	63.5 f	75.0 gh	82.0 h	7.00 g
	SA	1.0	64.0 ef	76.0 fg	84.0 g	8.00 fg
		2.0	62.3 g	75.0 gh	83.0 gh	8.00 fg
F-Value			2.21	5.03	9.82	4.09
p-Value			0.02	<0.001	<0.001	<0.001

). Among these traits, intact flower bud longevity was most affected. For instance, as available water content decreased from 80 to 20%, time to flowering, time to open flower, time to flower wilting and intact flower bud longevity decreased by 7.3, 9.1, 14.3, and 71.4%, respectively. Across water deficit levels, spray treatments generally increased these four flower bud traits. In doing so, TiO₂ NPs and SA were the most and least effective, respectively. In most but not all cases, the lowest respective concentration was optimal.

3.3. Plant Growth, Morphology, and Biomass Allocation

Several morphological features were assessed in different organs including lateral branches (number, length), main stem (length, diameter), leaves (number, area), flower (diameter) and root (length, diameter, volume) (

Table 3. Morphological features of periwinkle ‘Pacifica XP Really Red’ plants receiving exogenous application of titanium dioxide nanoparticles (TiO₂ NPs; 0, 0.5 and 1 mM), sodium hydrosulfide (NaSH; 0.5 and 1 mM), or salicylic acid (SA; 1 and 2 mM) under different watering levels (80, 50 and 20 % available water content) during cultivation. Means followed by different letters indicate significant differences (p = 0.05; comparison in columns). Values are the mean of four replications.

Available Water Content (%)	Compound	Concentration (mM)	Lateral Branch		Main Stem				Number of Nodes	Leaf		Flower Diameter (mm)	Root
			Number	Length (cm)	Length (cm)	Diameter (mm)	Strength (mg cm−1)	Tissue Density (mg cm−3)		Number	Area (cm2)		Length (cm)	Diameter (mm)	Volume (cm3)
80	–	0.0	10.8 efg	18.4 c	13.4 def	3.36 gh	20.78 cd	2.39 ab	10.5 cde	98.0 bc	415 cd	37.3 def	19.8 h	4.13 b	9.00 ef
	TiO2 NPs	0.5	13.3 ab	27.4 a	16.0 a	4.77 a	28.38 a	1.77 c–f	12.3 a	115 a	488 a	43.3 a	25.6 bcd	5.65 a	13.8 a
		1.0	14.3 a	28.6 a	15.3 ab	4.68 a	28.37 a	1.85 c–f	11.5 abc	115 a	483 a	40.9 abc	24.3 def	5.41 a	12.5 b
	NaSH	0.5	12.3 bcd	27.4 a	14.6 bc	4.27 bc	25.23 b	2.03 a–e	12.0 a	112 a	485 a	42.7 a	24.0 ef	5.32 a	12.0 bc
		1.0	12.5 bc	24.1 b	15.4 ab	4.47 ab	25.25 b	1.81 c–f	11.5 abc	110 a	478 a	39.7 bcd	23.4 ef	5.17 a	10.8 d
	SA	1.0	11.0 d–g	19.5 c	14.1 cd	4.23 bc	23.37 bc	1.93 b–f	12.3 a	101 b	430 bc	42.0 ab	24.3 def	5.20 a	11.0 cd
		2.0	11.3 c–f	19.5 c	13.9 cde	4.29 bc	21.42 cd	1.67 c–f	10.8 bcd	100 b	405 d	38.1 de	21.1 gh	5.15 a	9.50 e
50	–	0.0	9.75 gh	10.0 ghi	10.0 j	3.36 gh	20.57 d	2.47 a	7.75 hi	77.0 fg	210 g	28.8 j	22.8 f	3.11 c–f	5.80 i
	TiO2 NPs	0.5	14.0 a	12.3 efg	13.1 efg	3.96 cd	19.91 de	1.62 def	12.0 a	91.8 d	447 b	39.3 cd	29.3 a	3.77 bcd	8.00 fg
		1.0	13.3 ab	15.5 d	12.6 fgh	3.79 def	18.71 d–g	1.66 c–f	11.3 abc	84.8 e	351 e	35.4 fgh	27.0 b	3.63 b–e	7.80 gh
	NaSH	0.5	12.25 bc	13.5 de	13.0 e–h	3.71 def	17.74 e–h	1.65 c–f	11.8 ab	96.0 bcd	360 e	37.7 def	26.0 bc	3.80 bc	7.30 gh
		1.0	11.3 c–f	11.4 efg	12.4 gh	3.80 de	20.78 cd	1.88 c–f	11.5 abc	82.3 ef	320 f	36.5 efg	23.4 ef	3.58 b–e	7.00 gh
	SA	1.0	12.5 bc	12.9 def	12.4 gh	3.71 def	18.87 d–g	1.76 c–f	12.0 a	93.5 cd	341 ef	34.5 ghi	24.0 ef	3.28 c–f	7.50 gh
		2.0	11.8 cde	10.8 fgh	12.1 h	3.69 def	16.93 fgh	1.58 ef	11.3 abc	80.5 efg	327 f	32.2 i	22.9 f	3.29 c–f	6.80 hi
20	–	0.0	9.00 h	0.01 f	9.00 k	3.24 h	17.63 e–h	2.14 abc	6.75 i	60.5 k	147 j	25.6 k	24.5 cde	2.61 f	3.00 l
	TiO2 NPs	0.5	10.8 efg	0.01 f	10.4 ij	3.59 efg	17.67 e–h	1.74 c–f	10.0 def	75.0 gh	202 gh	36.4 e–h	28.9 a	2.99 def	4.60 j
		1.0	10.0 fgh	0.01 f	10.3 ij	3.54 e–h	17.24 e–h	1.75 c–f	9.50 efg	70.5 hi	186 hi	34.1 ghi	26.0 bc	3.09 c–f	4.10 jk
	NaSH	0.5	10.3 fgh	0.01 f	10.4 ij	3.56 e–h	16.45 gh	1.66 c–f	9.50 efg	69.3 i	185 hi	35.9 e–h	24.6 cde	2.90 ef	4.00 jkl
		1.0	10.8 efg	0.01 f	10.3 ij	3.46 fgh	19.45 def	2.09 a–d	9.25 fg	67.0 ij	173 i	34.0 hi	24.0 ef	2.86 ef	3.90 jkl
	SA	1.0	9.25 h	0.01 f	11.13 i	3.63 efg	15.17 h	1.47 f	9.3 fg	65.8 ijk	176 i	35.6 fgh	24.3 def	2.92 ef	3.90 jkl
		2.0	11.0 d–g	0.01 f	10.2 ij	3.52 e–h	15.80 h	1.63 def	8.75 gh	63.5 jk	170 i	32.8 i	22.7 fg	3.04 c–f	3.40 kl
F-Value			2.66	5.42	2.41	2.80	3.37	0.70	2.11	4.03	19.74	2.47	3.18	2.05	3.28
p-Value			0.006	<0.001	0.01	0.004	0.001	0.749	0.02	<0.001	<0.001	0.01	0.001	0.03	0.001

). As growth medium available water content decreased, all traits under study decreased, besides root length. Across water deficit levels, spray treatments generally increased these traits. In doing so, TiO₂ NPs and SA were generally the most and least effective, respectively. In most but not all cases, the lowest respective concentration was optimal. No symptoms of toxicity were apparent in the concentration range under study (Figures S1–S3).

As growth medium available water content decreased, both stem strength (mass per unit length) and tissue density (mass per unit volume) tended to decrease (Table 3). In most but not all cases, spray treatments tended to decrease these two stem traits.

As growth medium available water content decreased, plant dry weight also decreased (

Table 4. Plant growth, biomass allocation and morphology of periwinkle ‘Pacifica XP Really Red’ plants receiving exogenous application of titanium dioxide nanoparticles (TiO₂ NPs; 0, 0.5 and 1 mM), sodium hydrosulfide (NaSH; 0.5 and 1 mM), or salicylic acid (SA; 1 and 2 mM) under different watering levels (80, 50 and 20% available water content) during cultivation. Means followed by different letters indicate significant differences (p = 0.05; comparison in columns). Values are the mean of four replications.

Available Water Content (%)	Compound	Concentration (mM)	Main Stem	Lateral Branch	Leaf	Flower	Root	Plant	Water Use Efficiency (g Dry Mass L−1 H2O)	Stress Tolerance Index (%)	Root to Shoot Ratio (g g−1)	Specific Leaf Area (cm2 g−1)	Leaf Mass Ratio (g g−1)	Flower Mass Ratio (g g−1)
			Dry Weight (g)
80	–	0.0	0.28 de	0.10 c	1.15 d	0.19 f	0.59 ef	2.31 e	1.10 h	100 e	0.345 b–g	362 ab	0.50 bcd	0.083 a
	TiO2 NPs	0.5	0.45 a	0.16 a	1.60 a	0.30 a	1.07 a	3.58 a	1.63 a	149 a	0.427 a	305 cde	0.45 e	0.083 a
		1.0	0.43 a	0.15 a	1.58 ab	0.27 b	0.92 b	3.36 a	1.59 a	144 a	0.380 a–d	307 cd	0.74 de	0.081 ab
	NaSH	0.5	0.37 b	0.14 b	1.59 a	0.25 c	0.81 c	3.15 c	1.49 b	136 b	0.347 b–f	307 cd	0.50 bcd	0.078 a–d
		1.0	0.39 b	0.14 b	1.57 ab	0.24 cd	0.74 d	3.07 c	1.45 bc	132 b	0.319 fg	306 cde	0.51 bc	0.077 a–e
	SA	1.0	0.33 c	0.13 b	1.46 bc	0.23 d	0.73 d	2.88 d	1.36 cde	124 c	0.333 c–g	296 de	0.50 bc	0.081 abc
		2.0	0.30 cd	0.10 c	1.36 c	0.22 e	0.64 e	2.61 e	1.23 fg	112 d	0.325 efg	302 cde	0.52 b	0.082 ab
50	–	0.0	0.21 gh	0.05 fg	0.78 h	0.12 k	0.44 h	1.59 i	1.14 h	68.8 h	0.378 a–d	269 ef	0.49 bcd	0.073 def
	TiO2 NPs	0.5	0.26 def	0.09 d	1.13 de	0.17 g	0.65 e	2.29 f	1.64 a	99.0 e	0.392 ab	397 a	0.49 bcd	0.073 def
		1.0	0.24 fg	0.09 d	0.97 fg	0.15 i	0.55 fg	1.98 g	1.42 bc	85.8 f	0.388 abc	366 ab	0.48 bcd	0.073 def
	NaSH	0.5	0.23 fg	0.08 de	1.08 def	0.15 hi	0.55 fg	1.95 gh	1.49 b	90.4 f	0.356 b–f	334 bc	0.52 bc	0.072 def
		1.0	0.26 ef	0.06 f	0.95 g	0.16 gh	0.54 fg	1.95 fg	1.40 bcd	84.6 fg	0.378 a–d	339 bc	0.48 cd	0.081 abc
	SA	1.0	0.23 fg	0.07 e	1.01 efg	0.13 j	0.52 g	1.97 g	1.41 bcd	85.2 f	0.358 b–f	339 bc	0.51 bc	0.067 f
		2.0	0.21 gh	0.07 e	0.91 g	0.13 j	0.49 gh	1.81 h	1.29 ef	78.4 g	0.375 b–e	359 ab	0.50 bcd	0.072 def
20	–	0.0	0.16 i	0.023 k	0.58 j	0.06 n	0.22 j	1.04 l	1.12 h	48.3 k	0.268 h	254 f	0.55 a	0.060 g
	TiO2 NPs	0.5	0.18 hi	0.046 gh	0.70 hi	0.10 l	0.34 i	1.37 j	1.37 cde	59.5 i	0.334 d–g	287 def	0.51 bc	0.072 def
		1.0	0.18 hi	0.036 i	0.67 hij	0.10 l	0.33 i	1.31 jk	1.31 def	56.8 ij	0.341 c–g	278 def	0.51 bc	0.075 cde
	NaSH	0.5	0.17 hi	0.037 hi	0.67 hij	0.10 l	0.31 i	1.28 jk	1.28 ef	55.4 ij	0.318 fgh	280 def	0.52 b	0.076 b–e
		1.0	0.16 gh	0.033 ij	0.63 ij	0.09 lm	0.28 ij	1.23 jk	1.24 fg	53.2 ijk	0.296 gh	274 def	0.51 bc	0.073 def
	SA	1.0	0.18 hi	0.032 ij	0.62 ij	0.09 lm	0.33 i	1.24 jk	1.24 fg	53.7 ijk	0.365 b–f	286 def	0.50 bcd	0.071 ef
		2.0	0.18 i	0.026 jk	0.61 ij	0.08 m	0.30 i	1.17 kl	1.18 gh	50.9 jk	0.337 d–g	281 def	0.52 bc	0.072 def
F-Value			5.29	10.23	3.16	14.37	9.97	10.41	3.56	9.76	2.19	4.98	2.35	3.83
p-Value			<0.001	<0.001	0.001	<0.001	<0.001	<0.001	<0.001	<0.001	0.02	<0.001	0.01	<0.001

). The same was noted in individual weights of different organs (main stem, lateral branches, leaves, flower, roots). Across water deficit levels, spray treatments generally increased plant and individual organ dry weight. In doing so, TiO₂ NPs and SA were generally the most and least effective, respectively. In most but not all cases, the lowest respective concentration was optimal.

Growth medium available water content did not affect water use efficiency (dry mass produced per unit of water used) (Table 4). Instead, as growth medium available water content decreased, the ratio of dry weight to dry weight of the control (i.e., stress tolerance index) decreased. Across water deficit levels, spray treatments generally increased water use efficiency and stress tolerance index. In doing so, TiO₂ NPs and SA were generally the most and least effective, respectively. In most cases, the lowest respective concentration was optimal.

Across the 21 treatments, larger plants exhibited larger leaf area and increased individual organ (main stem, lateral branches, leaf, flower, and roots) weight (Figure 1). Higher flower mass was translated to wider flower diameter (R² = 0.61; see also Figure 1). Root diameter and volume effectively reflected variation in root dry weight (R² of 0.82 and 0.87, respectively; see also Figure 1). Instead, root length and dry mass were not related (R² = 0.01; see also Figure 1).

As growth medium available water content decreased, leaf thickness (SLA) and flower mass ratio tended to decrease (Table 4). The respective effect on root-to-shoot ratio and leaf mass ratio varied depending on the initial available water content. The effect of spray treatments on root to shoot ratio and SLA also generally varied depending on the available water content under investigation. Spray treatments generally exerted limited effects on leaf mass ratio. The same was noted for flower mass ratio, besides the lowest available water content (20%), where an increase was apparent.

Across the 21 treatments, plant dry weight was not associated with leaf thickness (SLA), root-to-shoot ratio, leaf mass ratio or flower mass ratio (Figure 1).

3.4. Stomatal and Epidermal Cell Anatomical Features

As growth medium available water content decreased, stomatal and epidermal cell density increased, whereas stomatal index, as well as stomatal and epidermal cell size decreased (

Table 5. Epidermal cell and stomatal anatomical features of periwinkle ‘Pacifica XP Really Red’ plants receiving exogenous application of titanium dioxide nanoparticles (TiO₂ NPs; 0, 0.5 and 1 mM), sodium hydrosulfide (NaSH; 0.5 and 1 mM), or salicylic acid (SA; 1 and 2 mM) under different watering levels (80, 50 and 20% available water content) during cultivation. Means followed by different letters indicate significant differences (p = 0.05; comparison in columns). Values are the mean of four replications.

Available Water Content (%)	Compound	Concentration (mM)	Epidermal Cell			Stomatal
			Density (mm−2)	Size (µm2)	Area Per Leaf Area (µm2 mm−2)	Density (mm−2)	Index (%)	Length (µm)	Width (µm)	Size (µm2)	Length to Width Ratio	Area Per Leaf Area (µm2 mm−2)
80	–	0.0	575 g–j	1686 b–e	964040 d–g	155 e–h	21.3 bc	25.2 fg	9.20 fg	464 gh	2.74 bc	35960 c–f
	TiO2 NPs	0.5	585 f–j	1634 c–g	953965 i	159 def	21.3 bc	27.0 a	10.8 a	581 a	2.51 hi	46035 a
		1.0	639 de	1526 ghi	961633 e–h	148 f–i	18.9 e–h	25.7 de	10.1 bcd	521 cd	2.54 ghi	38367 b–e
	NaSH	0.5	663 cd	1457 h–k	959871 h	156 d–h	19.1 e–h	26.6 b	9.70 e	514 d	2.750 bc	40129 b
		1.0	561 hij	1723 b–e	966334 bcd	133 j	19.1 e–h	25.3 efg	10.0 cd	508 d	2.52 hi	33666 fgh
	SA	1.0	666 cd	1461 hij	960877 fgh	143 hij	17.7 gh	26.2 bc	10.5 ab	549 b	2.50 i	39123 bcd
		2.0	599 e–i	1619 d–g	969055 abc	144 g–j	19.4 d–g	24.2 h	8.86 gh	430 k	2.74 bcd	30945 ghi
50	–	0.0	729 b	1331 kl	968930 abc	184 ab	20.1 cde	22.3 k	7.57 i	338 l	2.95 a	31070 ghi
	TiO2 NPs	0.5	674 cd	1418 ijk	954722 i	168 cde	19.9 cde	26.1 cd	10.4 bc	541 bc	2.51 hi	45278 a
		1.0	708 bc	1360 jk	961707 e–h	158 d–g	18.2 fgh	25.2 fg	9.63 e	486 ef	2.62 d–h	38293 b–e
	NaSH	0.5	625 d–g	1549 fgh	967001 bcd	133 j	17.5 h	25.3 fg	9.88 de	499 def	2.56 ghi	32999 fgh
		1.0	631 def	1529 ghi	964326 def	149 f–i	19.1 e–h	25.2 fg	9.51 ef	479 fg	2.65 c–g	35673 def
	SA	1.0	604 e–h	1598 efg	965007 de	153 fgh	20.2 cde	25.3 fg	9.08 gh	459 gh	2.78 b	34993 ef
		2.0	706 bc	1367 jk	964990 de	170 bcd	19.4 def	23.6 j	8.72 h	412 jk	2.71 b–e	35010 ef
20	–	0.0	808 a	1204 l	972205 a	195 a	19.5 def	20.4 l	7.00 j	285 m	2.92 a	27795 i
	TiO2 NPs	0.5	541 j	1788 b	965290 cde	138 ij	20.3 cde	25.5 ef	9.88 de	505 de	2.59 f–i	34710 efg
		1.0	553 ij	1757 bc	969969 ab	145 f–j	20.8 cd	23.7 ij	8.75 h	414 j	2.70 b–f	30031 hi
	NaSH	0.5	559 hij	1719 b–e	960338 gh	174 bc	23.7 a	23.3 j	9.80 de	456 h	2.38 j	39662 bc
		1.0	460 k	2109 a	969823 ab	136 ij	22.9 ab	24.0 hi	9.21 fg	443 hi	2.61 e–i	30177 hi
	SA	1.0	558 hij	1734 bcd	966321 bcd	150 f–i	21.2 bc	25.0 g	8.97 gh	449 hi	2.79 b	33679 fgh
		2.0	581 f–j	1670 b–f	968948 abc	159 def	21.5 bc	22.4 k	8.75 h	391 k	2.56 ghi	31052 ghi
F-Value			14.99	13.53	5.94	7.67	6.24	24.27	11.43	12.17	8.93	5.97
p-Value			<0.001	<0.001	<0.001	<0.001	<0.001	<0.001	<0.001	<0.001	<0.001	<0.001

). This decrease in stomatal size was mediated via decreases in both length and width. This was more prominent in stomatal width as compared to length, resulting in higher length to width ratio (i.e., more elliptic (less rounded) stomata).

Spray treatments tended to decrease stomatal density and increase stomatal size (Table 5). This increase in stomatal size was mostly mediated via increases in width, resulting in lower length to width ratio (i.e., more rounded stomata).

3.5. Water Status, as Well as Chlorophyll and Carotenoid Contents

As growth medium available water content decreased, leaf hydration status (RWC) as well as leaf chlorophyll and carotenoid contents decreased (

Table 6. Relative water content, electrolyte leakage, content of chlorophyll, carotenoids, and malondialdehyde (MDA), as well as peroxidase and ascorbate peroxidase activity of periwinkle ‘Pacifica XP Really Red’ plants receiving exogenous application of titanium dioxide nanoparticles (TiO₂ NPs; 0, 0.5 and 1 mM), sodium hydrosulfide (NaSH; 0.5 and 1 mM), or salicylic acid (SA; 1 and 2 mM) under different watering levels (80, 50 and 20% available water content) during cultivation. Means followed by different letters indicate significant differences (p = 0.05; comparison in columns). Values are the mean of four replications. FW, fresh weight.

Available Water Content (%)	Compound	Concentration (mM)	Relative Water Content (%)	Chlorophyll	Carotenoid	Electrolyte Leakage (%)	MDA	Proline	Catalase	Peroxidase
				Content (mg g−1 FW)			Content (µmol g−1 FW)		Activity (µmol min−1 g−1 FW)
80	–	0.0	82.2 de	10.4 f	3.01 a–d	22.1 fgh	0.343 j	4.23 i	0.00104 l	0.0099 m
	TiO2 NPs	0.5	91.5 a	15.1 a	3.11 ab	17.2 m	0.255 l	1.58 m	0.00114 kl	0.0088 m
		1.0	88.1 b	12.7 b	3.15 a	17.9 lm	0.288 kl	2.38 l	0.00114 kl	0.0120 kl
	NaSH	0.5	85.9 c	11.9 c	3.04 abc	18.7 klm	0.298 k	2.58 kl	0.00112 kl	0.0110 l
		1.0	82.6 de	11.6 cd	2.90 cde	19.0 j–m	0.295 k	2.99 jk	0.00137 j	0.0130 k
	SA	1.0	83.1 d	11.8 c	2.95 b–e	19.8 i–l	0.308 jk	2.75 kl	0.00116 k	0.0096 m
		2.0	82.2 de	10.6 f	2.84 def	20.0 ijk	0.340 j	3.40 j	0.00163 i	0.0157 j
50	–	0.0	72.3 j	8.25 i	2.23 kl	37.0 b	0.570 bc	8.33 f	0.00195 h	0.0270 i
	TiO2 NPs	0.5	81.5 def	11.3 de	2.76 efg	20.3 h–k	0.398 i	5.31 h	0.00252 e	0.0386 f
		1.0	80.9 efg	10.8 ef	2.69 fgh	20.8 g–j	0.435 ghi	7.21 g	0.00233 f	0.0399 f
	NaSH	0.5	80.7 efg	9.74 g	2.59 ghi	22.0 fgh	0.450 fgh	7.28 g	0.00213 g	0.0293 h
		1.0	79.2 ghi	9.16 h	2.52 hi	21.7 f–i	0.473 efg	7.35 g	0.00226 f	0.0300 h
	SA	1.0	79.5 fgh	9.78 g	2.54 hi	23.1 ef	0.473 efg	7.39 g	0.00205 g	0.0303 h
		2.0	78.4 hi	9.05 h	2.47 ij	20.9 ghi	0.498 de	9.19 e	0.00212 g	0.0342 g
20	–	0.0	57.7 k	5.89 l	1.22 o	42.8 a	0.643 a	17.5 a	0.00384 d	0.0647 a
	TiO2 NPs	0.5	78.1 hi	8.96 h	2.43 ij	22.6 efg	0.428 hi	10.0 d	0.00484 a	0.0517 c
		1.0	77.3 i	8.39 i	2.28 jk	23.1 ef	0.440 gh	14.5 c	0.00484 a	0.0535 b
	NaSH	0.5	77.8 hi	8.02 i	2.19 klm	24.1 de	0.480 ef	14.6 c	0.00436 b	0.0454 e
		1.0	77.7 hi	6.79 k	2.01 m	24.6 cde	0.520 d	14.7 c	0.00436 b	0.0466 de
	SA	1.0	77.4 i	7.41 j	2.09 lm	25.2 cd	0.533 cd	14.6 c	0.00419 c	0.0474 d
		2.0	72.5 j	6.17 l	1.75 n	26.3 c	0.573 b	16.8 b	0.00420 c	0.0478 d
F-Value			20.96	6.98	7.47	19.93	3.62	23.36	41.85	118.39
p-Value			<0.001	<0.001	<0.001	<0.001	<0.001	<0.001	<0.001	<0.001

). Across water deficit levels, spray treatments improved leaf hydration status, as well as leaf chlorophyll and carotenoid contents. In doing so, TiO₂ NPs and SA were generally the most and least effective, respectively. In all cases, the lowest employed concentration was the most effective.

Across the 21 treatments, chlorophyll and carotenoid contents were positively correlated (R² = 0.77; see also Figure 1).

3.6. Leaf Proline Content

As growth medium available water content decreased, leaf proline content increased (Table 6). Spray treatments tended to decrease leaf proline content.

3.7. Electrolyte Leakage and Lipid Peroxidation

As growth medium available water content decreased, leaf electrolyte leakage and MDA content (an index of lipid peroxidation) increased (Table 6). Across water deficit levels, spray treatments decreased leaf electrolyte leakage and MDA content. In doing so, TiO₂ NPs and SA were the most and least effective, respectively. In all cases, the lowest employed concentration was the most effective.

Across the 21 treatments, electrolyte leakage and MDA content were positively correlated (R² = 0.59; see also Figure 1).

3.8. Enzymatic Activity

CAT and POD are two key antioxidant enzymes. As growth medium available water content decreased, the activity of both enzymes increased (Table 6). With a single exception (CAT activity at 20% available water content), spray treatments tended to increase CAT and POD activities.

3.9. Principal Component Analysis

In order to identify and quantify the components that regulate the connections across treatments, a PCA was conducted (Figure 2). Eigenvalues were examined to determine the number of considered principal components. The first two dimensions explained 70% of the total variance (Figure S4). The level of significant contribution of each trait to the PCA was estimated by using the cos2 index (Figure S5). Among these descriptors, transpiration rate, leaf chlorophyll content, plant dry weight and stress tolerance index left a profound imprint on the categorization of the experimental units. PCA based on the first two components revealed the complex relationships among treatments (Figure 2A). The first axis revealed that the most significant factor was based on water deficit level within treatment groups (TiO₂ NPs, NaSH, and SA), as well as the employed concentration. The second axis allowed the differentiation of secluded treatment groups, while the lowest affinity was found for plants under 20% available water content and without spray treatment. This indicates that the application of the above-mentioned metabolites can provide a ‘rescue’ phenotype for water stressed plants. Furthermore, positive and negative correlations across traits were evident (Figure 2B). Most indices were highly homogenous (indicating co-regulation), while others were negatively correlated, such as MDA level with stress tolerance index and plant dry weight.

4. Discussion

In this study and for the first time, the optimal concentration of NaSH, SA, and TiO₂ NPs, as well as their relative effectiveness on determining plant growth and visual-perceived quality, was investigated across different water deficit levels (irrigation to 80, 50, and 20% available water content) in periwinkle, a plant of high ornamental and medicinal value.

Water deficit decreased main stem length, number of lateral branches and flower size (Table 3), which are critical ornamental (external quality) traits [19]. Moreover, the intact flower longevity was considerably shorter when irrigation was limited (Table 2). Water deficit was further related to lower stem strength (mass per unit length) (Table 3) and thus to a higher risk of buckling [21,22]. All these negative effects were stronger when water deficit was more intense. Collectively, these results indicate that water deficit adversely influences visually perceived quality, stem bending incidence and flower bud longevity, with this effect being water deficit intensity dependent.

Spray treatments (NaSH, SA, and TiO₂ NPs) improved visually perceived quality traits, and flower bud longevity (Table 2 and Table 3). By assessing these compounds individually, earlier studies also reported that their exogenous application exerted a positive effect on plant growth in other taxa [12,13,14,15,16]. This investigation additionally shows that both their relative effectiveness (TiO₂ NPs > NaSH > SA) and optimal concentration (i.e., 0.5 mM (TiO₂ NPs, NaSH), and 1 mM (SA)) was rather consistent across water deficit levels. Instead, spray treatments tended to decrease stem strength (Table 3). Although spray treatments may increase stem bending incidence, they consistently improved visually perceived quality traits, and flower bud longevity, with this effect being stronger as water deficit becomes more intense. Therefore, their application will improve ornamental plant quality and longevity independently of the watering regime, with their promotive effect being more evident in environments with irregular or limited water supply.

Leaf greenness is typically employed as an index of ornamental plant quality, health and vigor along the production, distribution, and sales processes [19,28]. As water deficit became more intense, leaf chlorophyll content decreased (Table 6). This water deficit-induced decrease in chlorophyll content has also been earlier documented in other species [5]. Spray treatments improved leaf chlorophyll content in control plants and largely alleviated the water deficit-induced reduction. Under drought, the mitigating effect of these compounds on leaf chlorophyll content has been earlier suggested in other taxa [12,13,14,15,16]. Therefore, water deficit additionally downgrades ornamental plant value by impairing leaf coloration, and this effect is largely reversed when employing the spray treatments under study.

Water deficit adversely affected whole plant and individual organ mass, especially when it was more intense (Table 4). The water deficit-imposed decline in leaf area (thus light interception) (Table 3), chlorophyll content (Table 6) and internal CO₂ concentration (Table 1) were factors contributing to the documented plant growth impairment. Similar findings have been earlier reported in other species [5,6]. Spray treatments generally improved these underlying processes and, in this way, both stimulated plant growth in control plants and mitigated the water deficit-induced decrease.

Plant water status during cultivation shapes stomatal anatomical traits [27]. Water deficit increased the density and decreased the size of stomata (Table 5). Similar effects have been earlier reported in other species [20]. Spray treatments tended to reserve the water deficit-induced effects on stomatal density and size. This mitigation is most probably related to the spray treatments-induced improvement in leaf hydration level (Table 6).

A range of processes underlying the negative effect of water deficit on plant growth was further investigated. Under adverse conditions, for example, proline is biosynthesized, which actively participates in both cell osmotic regulation and ROS scavenging [5,7]. Indeed, leaf proline content was more enhanced as water deficit became more severe (Table 6). Notably, spray treatments tended to decrease leaf proline content. Therefore, the promotive effect of spray treatments on plant growth and vigor was not mediated by the proline route.

In un-treated plants, water deficit was associated with decreased carotenoid content (non-enzymatic antioxidant) and increased activity of two critical antioxidant enzymes (CAT, POD) (Table 6). In addition, when water deficit became more intense, electrolyte leakage and MDA content (indicative of lipid peroxidation) increased, suggesting a more extensive cellular damage (Table 6). Comparable effects have been shown previously in other taxa [5,6]. Our results denote that spray treatments stimulated carotenoid content and the activity of the two antioxidant enzymes under study. Therefore, spray treatments supported plant ability to detoxify ROS and, in this way, mitigated the water deficit-induced oxidative stress and improved water deficit tolerance. Yet, a more extensive analysis of the oxidative protection networks, including a wider range of enzymes and metabolites, is essential for a deeper understanding of the underlying processes and the identification of more critical antioxidant elements mediating the alleviating response of the spray treatments under investigation.

5. Conclusions

Water deficit adversely affects plant growth and ornamental value. The optimal concentration of sodium hydrosulfide (NaSH), salicylic acid (SA), and titanium dioxide (TiO₂) nanoparticles (NPs), along with their relative effectiveness on mitigating the negative impact water deprivation on ornamental quality were evaluated in periwinkle. Water deficit downgraded several elements of ornamental quality (stem length, number of lateral branches, flower size, leaf coloration), shortened longevity, and raised the risk of buckling (lower stem strength). Water deficit also impaired plant growth, owing to reduced light interception (leaf area) and carbon assimilation. It also induced oxidative damage as expressed by lower chlorophyll content, as well as a higher degree of membrane degradation and lipid peroxidation. These adverse effects were stronger as water availability decreased. Spray treatments enhanced all traits under study, except for stem strength and proline content. Spray treatments further improved leaf carotenoid content, as well as catalase and peroxidase activities. The relative effectiveness (TiO₂ NPs > NaSH > SA) and optimal concentration (i.e., 0.5 mM (TiO₂ NPs, NaSH), and 1 mM (SA)) of spray treatments was independent of the water availability. Therefore, the practices under investigation generally improve ornamental quality and longevity, while this effect becomes increasingly important as water availability becomes limited.