Combined Application of Bacillus aryabhattai and Silicon Enhances Membrane Stability, Biochemical Attributes, and Soil Biological Quality in Yellow Passion Fruit Under Water Deficit

Abstract

Yellow passion fruit production is frequently limited by water scarcity, necessitating biotechnological strategies to ensure seedling quality. This study investigated the synergistic effects of Bacillus aryabhattai (Auras^®) and silicon (Si) as mitigators of water deficit in Passiflora edulis seedlings. The experiment was conducted in a greenhouse in Catolé do Rocha, PB, Brazil, using 4 dm³ plastic bags. A randomized block design was used with a 4 × 3 + 2 factorial scheme, testing four available water contents (AWC: 50, 60, 70, and 80%) combined with three mitigation strategies (Auras, Si, and Auras + Si), plus two additional controls (50% and 100% AWC). Water deficit severely compromised growth and soil biological activity; however, mitigation treatments significantly improved physiological and biochemical responses. When applied separately, B. aryabhattai inoculation enhanced the accumulation of photoprotective pigments (carotenoids) and secondary metabolites (flavonoids and anthocyanins) under severe drought, while individual Si application provided homeostatic stability to plant biomass, maintaining dry matter production at levels comparable to moderate irrigation. The Auras + Si combination was the most effective, promoting the highest membrane stability, pigment maintenance, and vigorous growth even under 50% AWC. Furthermore, this interaction optimized soil microbial biomass and reduced the metabolic quotient by 56.7% compared to the stress control. These findings demonstrate that the combined application of B. aryabhattai and Si effectively mitigates the negative impacts of water scarcity on the initial development of passion fruit seedlings and soil microbial activity.

1. Introduction

Yellow passion fruit (Passiflora edulis Sims) stands out as one of the most agronomically and nutritionally relevant fruit crops in Brazil, which accounts for approximately 70% of global production. The country is established as the leading global producer and consumer, with an annual output exceeding 700,000 tons [1,2]. This crop plays a vital socioeconomic role, particularly for small and medium-sized producers in the Brazilian Northeast. However, the expansion of cultivation in semi-arid environments faces challenges imposed by climate change, such as the scarcity of quality water and high evaporative demand [3,4]. These conditions trigger water deficit stress, the main limiting factor for seedling production and orchard establishment, thereby compromising future yield.

Water deficit in plants induces negative physiological responses, starting with the overproduction of reactive oxygen species (ROS), which leads to lipid peroxidation and the loss of cell membrane integrity [5]. This structural damage can be monitored via electrolyte leakage, a sensitive indicator of cellular stability. Furthermore, the restriction of water availability limits gas exchange by promoting stomatal closure, which directly reduces the net CO₂ assimilation rate and photosynthetic efficiency. Additionally, reduced turgor limits cell division and elongation, resulting in decreased growth and dry mass production [6]. This scenario compromises the uptake and redistribution of nutrients, such as nitrogen, and triggers metabolic adjustments in the synthesis of photosynthetic pigments (chlorophylls and carotenoids) and photoprotective phenolic compounds, such as flavonoids and anthocyanins [7,8,9].

In this context, the application of beneficial elements such as silicon (Si) emerges as a promising strategy. Although non-essential, silicon alleviates abiotic stresses through its deposition in the cell wall and under the cuticle as amorphous silica (SiO₂·nH₂O), increasing the structural rigidity of tissues and reducing cuticular transpiration [10]. This silicification process acts as a physical barrier that prevents excessive water loss while maintaining leaf temperature under high solar radiation. Beyond this physical barrier, Si acts in metabolic regulation, increasing antioxidant enzyme activity and favoring osmotic adjustment through the accumulation of compatible solutes [4]. In Passiflora, recent studies demonstrate that silicon supply mitigates the effects of water deficit and salinity by preserving photosynthetic efficiency and maintaining cell membrane stability under conditions of low water availability [11,12,13].

Concurrently, the use of plant growth-promoting rhizobacteria (PGPR), such as Bacillus aryabhattai, represents a sustainable biotechnological tool. B. aryabhattai was first isolated in 2009 in India [14]. In Brazil, the CMAA 1363 strain—derived from the rhizosphere of Cereus jamacaru, a native cactus—was isolated in 2013 and has demonstrated high agronomic potential [15,16]. Studies indicate that this rhizobacterium mitigates drought effects through phytohormonal regulation, ACC-deaminase production, activation of antioxidant defenses, and exopolysaccharide (EPS) synthesis [16,17,18,19]. EPS production is particularly critical, as it promotes soil particle aggregation around the roots, forming a rhizosheath that maintains a localized humid microenvironment and enhances water retention and nutrient diffusion under drought conditions [20]. However, most research still focuses on maize and soybean, highlighting the need to further investigate its use in fruit crops.

In addition to direct effects on the plant, it is believed that the use of these technologies can optimize soil biological quality. Indicators such as Microbial respiration (MR), Microbial biomass carbon (MBC), and the Metabolic quotient (qCO₂) are fundamental for diagnosing the mass, activity, and bioenergetic efficiency of the microbiota under stress [21]. These biological attributes are more sensitive than chemical or physical properties, providing “early warnings” about the ecological impact of drought and the effectiveness of bio-inoculants in restoring soil functionality. Thus, the integration of plant physiological components and soil biological indicators allows for a more comprehensive assessment of plant responses to water deficit and the application of stress attenuators.

Despite the individual benefits of Si and B. aryabhattai, studies on their synergistic effect on P. edulis in the semi-arid region are still scarce in the literature. It is hypothesized that the combination of Si and B. aryabhattai can provide a superior protective effect, maintaining cellular homeostasis and biomass while modulating biochemical and microbiological parameters. Therefore, the objective of this study was to evaluate growth, and physiological and biochemical performance of yellow passion fruit seedlings, as well as soil biological quality, under different irrigation depths associated with silicon application and B. aryabhattai inoculation.

2. Materials and Methods

2.1. Experimental Conditions

The experiment was conducted under protected conditions (greenhouse) at the Center for Human and Agrarian Sciences of the State University of Paraíba (UEPB), located in Catolé do Rocha, PB, Brazil (6°20′38″ S, 37°44′48″ W, 252 m a.s.l.), from October 2025 to January 2026. The maximum and minimum values of air temperature and relative humidity recorded daily during the experimental period are presented in Figure 1a and Figure 1b, respectively.

The plant material used consisted of yellow passion fruit (Passiflora edulis Sims) cv. ‘BRS Gigante Amarelo’, propagated by seeds. Initially, seeds were sown in polyethylene trays containing cells with a volume of 0.0125 dm³. Subsequently, the most vigorous seedlings with one pair of fully expanded leaves were selected and transplanted into 4 dm³ polyethylene bags filled with a substrate composed of soil, cattle manure, and sand (1:1:1, v/v/v), where they were maintained until the end of the experiment.

The soil used in the experiment was classified as an Entisol (Fluvent) according to USDA [22], and its physical and chemical characteristics, as well as those of the cattle manure used to prepare the substrate, are presented in

Table 1. Chemical and physical characteristics of the soil and cattle manure used in the experiment.

Chemical Features of the Soil
pH	OM			P			K+		Na+			Ca2+				Mg2+				Al3+			H+ + Al3+			SEB		CEC		V
1:2.5	g kg−1			mg dm−3			cmolc dm−3																							%
6.00	13.58			16.63			0.08		0.05			1.09				1.12				0.00			1.24			2.34		3.58		65.36
Physical features of the soil
Sand		Silt			Clay			BD			PD			TP				VWCFC				VWCPWP						Textural class
g kg−1								g cm−3						%														–
546.00		230.00			224.00			1.53			2.61			41.38				23.33				11.72						SCL
Chemical features of the cattle manure
pH			EC			OM				C			N				C/N				P				K+		Ca2+		Mg2+
1:2.5			dS m−1			dag kg−1				g kg−1							-				g kg−1
7.70			6.09			36.20				166.90			13.90				12.00				3.20				18.70		16.20		6.10
S			CEC			B				Fe					Cu				Mn					Zn			Si		Na+
g kg−1			mmolc dm−3			mg kg−1				mg kg−1																	g kg−1
2.50			133.90			14.80				11,129.90					19.30				491.40					65.30			12.50		3.50

pH—hydrogen potential; OM—organic matter; SEB—sum of exchangeable bases; CEC—cation exchange capacity; V—base saturation percentage; BD—bulk density; PD—particle density; TP—total porosity; VWC_FC—volumetric water content at field capacity (−0.033 MPa); VWC_PWP—volumetric water content at the permanent wilting point (−1.500 MPa); SCL—sandy clay loam; EC—electrical conductivity.

.

2.2. Treatments and Experimental Design

The treatments consisted of the combination of four irrigation depths (50, 60, 70, and 80% of available water content in soil—AWC) and three drought stress mitigation strategies: inoculation with Bacillus aryabhattai; silicon (Si) application; and the combined application of Si + B. aryabhattai, in addition to two absolute controls (100% and 50% AWC) without the application of mitigation strategies. The experimental design was completely randomized, arranged in a 4 × 3 + 2 factorial scheme, with five replicates.

Initially, the substrate moisture content was raised to field capacity by saturation followed by drainage. Irrigation was performed daily, replacing the water volume for each AWC level determined by the weighing method, with the irrigation depth calculated as the difference between the pot mass at field capacity (previous mass) and the current mass. Irrigation was carried out using local supply water with an electrical conductivity of 0.3 dS m⁻¹.

Silicon application was performed via fertigation using orthosilicic acid as the source, according to Almeida et al. [23]. The total dose of 0.55 g of Si per plant was divided into two equal applications (0.275 g of Si each), with the first applied at planting (basal fertilization) and the second at 30 days after transplanting. The commercial product Sifol^® was used as the silicon source, with a chemical composition of 92% SiO₂ and 42.9% Si, apparent density ranging from 40 to 80 g L⁻¹, particle size between 8 and 12 mesh, and pH ranging from 6.0 to 7.5.

For inoculation with B. aryabhattai, the commercial bioinput Auras^® was used, developed by the Brazilian Agricultural Research Corporation (Embrapa) in partnership with NOOA Biotecnologia. The product contains the strain B. aryabhattai CMAA 1363, humic substances, thickener, preservative, and water, with a density of 1.04 g mL⁻¹ at 20 °C. Bacterial inoculation consisted of seed soaking (20 μL per 50 seeds for 24 h) and the direct application of 100 mL per plant of a solution containing the bioinput and water (25 μL L⁻¹) to the substrate prior to transplanting the seedlings into polyethylene bags. These values were defined based on preliminary tests conducted by our research team.

2.3. Experimental Analyses

At 80 days after sowing, plants were harvested, and electrolyte leakage (EL), leaf nitrogen and silicon contents, photosynthetic pigment concentrations (chlorophyll a, chlorophyll b, total chlorophyll, and carotenoids), flavonoids, and anthocyanins were evaluated, as well as seedling height, stem diameter, and shoot, root, and total dry mass. After plant collection, soil samples were obtained, and biological soil indicators were analyzed, including microbial respiration, microbial biomass carbon, and metabolic quotient.

2.3.1. Physiological Indicators Determination

For EL determination, five leaf discs per plant were incubated in 10 mL of deionized water for 90 min, and the initial electrical conductivity (ECi) was recorded. Subsequently, the samples were heated at 80 °C for 90 min, and the final electrical conductivity (ECf) was measured. Electrolyte leakage was calculated according to Equation (1).

$$\mathrm{E}\mathrm{L}={\displaystyle \frac{\mathrm{E}\mathrm{C}\mathrm{i}}{\mathrm{E}\mathrm{C}\mathrm{f}}}\times 100$$

(1)

For nutritional analyses, leaf samples were dried in a forced-air oven at 65 °C until constant mass, ground in a knife mill, and analyzed according to the methodology of Meneghetti [24]. Nitrogen content was determined by sulfuric digestion, followed by distillation and titration using the Kjeldahl method. Silicon was quantified by spectrophotometry using the molybdenum blue method adapted for this element.

Chlorophyll a, chlorophyll b, total chlorophyll, and carotenoids were quantified in the third leaf from the plant apex. Five leaf discs were macerated with 0.2 g of calcium carbonate and 5 mL of 80% acetone. The extract was centrifuged at 3000 rpm for 10 min at 10 °C, and the supernatant was transferred to a graduated cylinder, with the volume adjusted to 5 mL using the same extracting solution. An aliquot of the extract was then placed in a cuvette, and absorbance readings were taken at 470, 646, and 663 nm using a spectrophotometer (model IL-592-LC, Kasuaki, São Paulo, Brazil). Pigment contents were calculated based on the equations proposed by Lichtenthaler [25]. All procedures were carried out under controlled lighting conditions, in a dark environment, to prevent degradation of light-sensitive pigments.

Total flavonoids and anthocyanins were determined according to the method described by Francis [26]. Leaf samples (approx. 0.5 g) were macerated in a mortar with 5 mL of an extractive solution composed of 95% ethanol and 1.5 N HCl in an 85:15 (v/v) ratio. The extract was transferred to centrifuge tubes, and the mortar was washed with an additional 5 mL of the same solution to reach a final volume of 10 mL. The samples were stored at 5 °C in the dark for 24 h. Subsequently, the extract was centrifuged at 3000 rpm for 5 min at 5 °C and filtered through filter paper. Absorbance was measured using a spectrophotometer at 374 nm for total flavonoids and 535 nm for total anthocyanins.

2.3.2. Growth Indicators Determination

Seedling height (SH), stem diameter (SD), and plant dry mass (TDM), shoot dry mass (SDM), and root dry mass (RDM) were measured to evaluate plant growth. SH was measured from the substrate surface to the apex of the highest leaf, and SD was measured at 1 cm above the substrate surface. For dry mass determination, plants were dried in a forced-air oven at 65 °C until constant mass and weighed using a precision balance (0.01 g).

2.3.3. Soil Biological Indicators Determination

Microbial respiration (MR) was determined by the method based on CO₂ capture released during soil incubation, according to Alef and Nannipieri [27]. Microbial biomass carbon (MBC) was quantified using the irradiation–extraction method, according to Ferreira et al. [28]. The metabolic quotient (qCO₂) was calculated as the ratio between MR and MBC, following Anderson and Domsch [29], representing the metabolic efficiency of the soil microbiota (relationship between evolved CO₂ and total microbial C).

2.4. Statistical Analysis

The data were subjected to tests for normality of errors (Shapiro–Wilk) and homogeneity of variances (Bartlett), followed by ANOVA. For the irrigation depths factor, linear regression analysis was performed, whereas the means of the treatments with mitigation strategies were compared with the controls using Dunnett’s test (p ≤ 0.05), using R 4.5.1.

3. Results

3.1. Physiological Indicators

ANOVA revealed significant effects of the isolated factors, as well as interactions and additional controls (Table S1). Plants treated solely with Auras and Si exhibited linear reductions (Auras: R² = 0.93; Si: R² = 0.87) in electrolyte leakage (EL) as water availability increased, with decreases of 11.4% and 8.8%, respectively (Figure 2). Conversely, the Auras + Si combination showed a quadratic behavior (R² = 0.99), starting with the highest EL index under 50% AWC, but promoting the most pronounced reduction among the mitigators until reaching its maximum membrane stability point, estimated at 69.5% AWC (EL of 8.8%), before registering a slight increase at the 80% AWC level.

From 60% AWC onwards, all three mitigation strategies presented EL values lower than the untreated control (Ctrl. AWC 50%), which maintained a constant damage level near 19.5% (Figure 2), according to Dunnett’s test (Table S2). However, with the combined treatment (Auras + Si), from 60% AWC onwards, EL did not differ statistically from the irrigated control (Ctrl. AWC 100%), while for the isolated Auras and Si treatments, this equivalence to Ctrl. AWC 100% (which remained below 9%) was only observed from 80% AWC onwards.

The isolated application of Auras promoted a linear increase (R² = 0.96) in leaf nitrogen content as water availability increased, raising levels from 32.14 g mg⁻¹ (under 50% AWC) to 41.79 g mg⁻¹ (under 80% AWC), representing a 30% increase in the accumulation of this nutrient (Figure 3a). Conversely, treatments containing silicon exhibited a quadratic response (R² = 0.70). The isolated application of Si resulted in an initial decrease in N levels until the minimum point estimated at 66.7% AWC (32.96 g mg⁻¹), before increasing again at higher irrigation levels. A similar behavior was observed for the Auras + Si combination (R² = 0.99), which reached its minimum concentration point at 58.4% AWC (35.11 g mg⁻¹), followed by a rise until reaching 41.68 g mg⁻¹ at the 80% AWC level.

Under conditions of lower water availability (50% AWC), plants treated with isolated Si and the Auras + Si combination presented N levels higher than the irrigated control (Ctrl. AWC 100%) (Figure 3a). In contrast, Ctrl. AWC 50% showed the lowest N levels, stabilizing at 31.4 g mg⁻¹. Upon reaching 80% AWC, both the Auras treatment and the Auras + Si combination significantly outperformed both controls (Table S2).

The isolated application of silicic acid promoted a linear increase in leaf Si levels as water availability increased (R² = 0.98), raising levels from 1220.4 mg kg⁻¹ (under 50% AWC) to 2951.5 mg kg⁻¹ (under 80% AWC) (Figure 3b). On the other hand, treatments involving the biostimulant and the combination exhibited quadratic effects (Auras: R² = 0.99; Auras + Si: R² = 0.77). The Si content with the isolated use of Auras reached its maximum accumulation point estimated at 67.9% AWC (2486.1 mg kg⁻¹), followed by a decline at higher irrigation levels. A similar behavior was observed for the Auras + Si combination, which reached its peak concentration at 66.1% AWC (3281.4 mg kg⁻¹), outperforming the isolated treatments under intermediate water availability conditions.

At AWC levels between 60% and 80%, all mitigation strategies maintained leaf Si levels higher than the stress control (Figure 3b), according to Dunnett’s test (Table S2). However, even with the increase provided by the mitigators, no treatment reached the Si levels observed in the fully irrigated control with 100% AWC. It is noteworthy that under the greatest water deficit (50% AWC), the Auras + Si combination was the most efficient strategy, ensuring an initial silicon accumulation approximately 66% higher than the isolated Si treatment.

Regarding the total chlorophyll content in passion fruit leaves, all mitigation strategies promoted quadratic behavior (Auras: R² = 0.87; Si: R² = 0.88; Auras + Si: R² = 0.99) in response to varying water availability (Figure 4a). The isolated application of Auras showed an initial decrease followed by an increase, reaching its highest observed value (53.5 mg 100 g⁻¹ FM) under the 80% AWC level. In contrast, treatments containing silicon presented downward-facing curves. The isolated use of Si reached its maximum total chlorophyll efficiency estimated at 62.0% AWC (41.2 mg 100 g⁻¹ FM), while the Auras + Si combination reached its peak at 66.0% AWC (43.3 mg 100 g⁻¹ FM).

At all tested AWC levels, plants receiving the mitigators maintained total chlorophyll levels higher than the stress control (Table S2). Furthermore, from 60% AWC onwards, all mitigation strategies significantly outperformed the fully irrigated control, evidencing that supplementation with Auras and Silicon, whether isolated or combined, enhanced the accumulation of chlorophyll pigments, even under conditions of moderate water limitation.

The dynamics of chlorophyll a in the leaves were also altered by the interaction between water supply and mitigating methods (Table S1). Unlike the other strategies, exclusive inoculation with Auras resulted in a positive linear progression (R² = 0.64) in (Figure 4b), in which increasing water availability from 50% to 80% AWC raised the chlorophyll a content by approximately 63.5%. In contrast, the inclusion of silicon induced quadratic responses (Si: R² = 0.86; Auras + Si: R² = 0.99). While the isolated use of Si reached its accumulation ceiling at 62.9% AWC, the Auras + Si combination shifted this point of maximum efficiency to a slightly higher level of 66.5% AWC.

Notably, under all AWC levels, both Auras and Si were able to sustain chlorophyll a levels higher than the negative control (50% AWC). Conversely, at 60% and 70% AWC levels, the chlorophyll a contents promoted by all mitigating treatments exceeded the fully irrigated control (Table S2).

For chlorophyll b levels (Figure 4c), the results reveal distinct response patterns among the mitigators according to variations in irrigation levels. Isolated Si supplementation resulted in a linear decline (R² = 0.96), reducing the pigment concentration from 9.04 to 7.18 mg 100 g⁻¹ FM as water availability increased from 50% to 80% AWC. In contrast, the other strategies provided quadratic adjustments (Auras: R² = 0.99; Auras + Si: R² = 0.75). Inoculation with Bacillus aryabhattai (Auras) promoted an initial reduction in chlorophyll b until the minimum point estimated at 58.7% AWC (6.67 mg 100 g⁻¹ FM), followed by a sharp increase that culminated in the highest content observed in the experiment (11.63 mg 100 g⁻¹ FM) under 80% AWC. On the other hand, the chlorophyll b of plants under the Auras + Si combination showed a quadratic effect, reaching its peak accumulation at 63.1% AWC (9.97 mg 100 g⁻¹ FM) (Figure 4c). Under all AWC levels, all mitigation methods provided chlorophyll b contents higher than the stress control (Ctrl. AWC 50%).

For carotenoid content (Figure 4d), the results demonstrate that all mitigation strategies followed quadratic models as a function of AWC levels (Auras: R² = 0.89; Si: R² = 0.60; Auras + Si: R² = 0.69). Isolated inoculation with Auras provided the highest accumulation of these photoprotective pigments, reaching its maximum estimated efficiency point at 67.1% AWC (5.06 mg 100 g⁻¹ FM). In contrast, strategies involving silicon presented curves with distinct concavities; while the isolated Si treatment exhibited a minimum point at 65.3% AWC (2.23 mg 100 g⁻¹ FM), the Auras + Si combination reached its peak concentration at 65.6% AWC (2.33 mg 100 g⁻¹ FM).

Regardless of the mitigation treatment, under 50% AWC, carotenoid levels were substantially higher than those observed in the stress control (Ctrl. AWC 50%), according to Dunnett’s test (Table S2). Notably, at the 70% AWC irrigation range, the use of Auras even outperformed the fully irrigated control by 42% (Figure 4d).

Total flavonoid and anthocyanin contents were significantly altered by the interaction between water availability and mitigating methods (Table S1). For total flavonoids (Figure 5a), the isolated application of Auras promoted a linear decrease as irrigation increased (R² = 0.85), with values ranging from 23.98 mg 100 g⁻¹ FM (under 50% AWC) to 21.84 mg 100 g⁻¹ FM (under 80% AWC). In contrast, strategies involving silicon exhibited quadratic behavior (Si: R² = 0.72; Auras + Si: R² = 0.88); both isolated Si and the Auras + Si combination reached their minimum accumulation points at intermediate irrigation levels, estimated at 67.1% and 61.1% AWC, respectively, before trending upward again.

Regarding anthocyanins (Figure 5b), the treatments promoted heterogeneous responses to the water supply. Exclusive inoculation with Auras followed a quadratic model (R² = 0.88), recording a minimum concentration point at 63.5% AWC (2.19 mg 100 g⁻¹ FM), followed by an increase up to the 80% AWC level. The combined Auras + Si treatment, in turn, promoted a positive linear progression (R² = 0.69), while isolated Si showed a linear decreasing (R² = 0.66) trend in anthocyanins as the irrigation level increased.

Regardless of the biochemical trait, under the greatest water deficiency (50% AWC), all mitigation treatments maintained flavonoid and anthocyanin levels higher than the stress control (Ctrl. AWC 50%) (Figure 5). Notably, the mitigating strategies ensured the accumulation of these secondary metabolites at levels equivalent to or higher than those of the fully irrigated control.

3.2. Growth Indicators

The growth traits of the passion fruit seedlings were significantly affected by the interaction between water availability and the mitigating methods (Table S1). For seedling height (SH) (Figure 6a), it was observed that the strategies promoted distinct responses according to the irrigation volume. The isolated application of Auras caused a quadratic model in SH (R² = 0.83), reaching its maximum estimated height point at 71.3% AWC (50.33 cm), a level after which there was a slight downward trend. In contrast, treatments involving silicon exhibited continuous linear increases (Si: R² = 0.70; Auras + Si: R² = 0.97) in SH as irrigation levels rose. The isolated use of Si raised SH from 41.36 cm to 56.71 cm, while plants under the Auras + Si combination showed the most vigorous growth in the experiment, jumping from 38.68 cm (under 50% AWC) to 63.99 cm (under 80% AWC). Notably, all treated plants maintained heights significantly higher than the severe stress control (Ctrl. AWC 50%), which remained at 13.2 cm.

Regarding stem diameter (Figure 6b), the results were less pronounced. Inoculation with Auras provided a quadratic behavior (R² = 0.98) for this trait, with a maximum thickness point at 64.8% AWC (4.15 mm). The isolated use of Si caused a linear reduction (R² = 0.69) in diameter as irrigation increased, while the Auras + Si combination promoted a discrete linear increase (R² = 0.50). As with height, all mitigation strategies ensured diameters superior to the stressed control (2.9 mm), although no treatment reached the vigor of the fully irrigated control (4.9 mm).

The total plant dry mass (PDM) (Figure 7a) was significantly influenced by the interaction of factors (Table S1), with a highlight on the stability provided by silicon. The isolated Si treatment presented a constant average of 3.64 g, regardless of the AWC level. Conversely, the PDM of plants under Auras inoculation followed a quadratic model (R² = 0.97), with a maximum efficiency point estimated at 68.1% AWC (3.76 g). The Auras + Si association exhibited a linear increase (R² = 0.76) in PDM, reaching its highest value (3.74 g) at 80% AWC.

In shoot dry mass (SDM) (Figure 7b), the response pattern was similar. While isolated Si kept the SDM stable at 2.71 g, the application of Auras described a parabola (R² = 0.97) with a peak at 67.8% AWC, resulting in a maximum accumulation of 2.79 g. The Auras + Si combination recorded a progressive linear (R² = 0.73) gain, increasing shoot biomass as water availability expanded.

The root system was also favored by the mitigation strategies. For root dry mass (Figure 7c), the isolated use of Auras followed a quadratic adjustment (R² = 0.99), with a maximum point at 67.8% AWC (0.84 g), declining subtly under full irrigation. In contrast, both isolated Si and the Auras + Si combination presented linear increases (Si: R² = 0.56; Auras + Si: R² = 0.85). In all dry mass variables, the severe stress control (Ctrl. AWC 50%) exhibited the lowest average values. It is also important to highlight that, even under 50% AWC, seedlings treated with isolated silicon maintained PDM (Figure 7a), SDM (Figure 7b), and RDM (Figure 7c) levels very close to those of plants under moderate irrigation.

The qualitative analysis of the visual appearance of the passion fruit seedlings (Figure 8) corroborates the quantitative data on growth and biomass accumulation. It is observed that severe water restriction in the control without mitigators (Control 50% AWC) resulted in seedlings with the lowest vegetative growth in the experiment. In contrast, the application of mitigation strategies promoted a clear change in plant architecture, even under water deficit conditions. In the Auras strategy (Figure 8a), inoculation allowed the seedlings to maintain vertical growth and tendril emission starting from the 60% AWC level; however, plants inoculated with Auras and under 50% AWC appear more vigorous than the negative control.

The isolated use of Si (Figure 8b) conferred greater stem robustness and constant leaf expansion along the water gradient. The combination of strategies (Auras + Si) resulted in the most balanced and vigorous visual appearance in the study (Figure 8c). Notably, under 50% AWC, the seedlings treated with the association showed a significantly larger size than the stressed control, maintaining turgidity and developing support structures more prematurely. Overall, the photographic record confirms that the mitigators not only increased biomass but also preserved the morphological integrity and scaling potential of the seedlings under water limitation.

3.3. Soil Biological Indicators

Results for soil microbial respiration (MR) indicate that microbial metabolic activity was significantly influenced by the interaction between water availability and the applied mitigation strategies (Table S1). Strategies based on Auras and the Auras + Si combination showed linear decreases (Auras: R² = 0.87; Auras + Si: R² = 0.87) in MR (Figure 9a) as available water content (AWC) levels increased. With the isolated application of Auras, respiratory activity reduced from 25.45 mg C-CO₂ 100 cm⁻³ soil d⁻¹ (under 50% AWC) to 13.25 mg C-CO₂ 100 cm⁻³ soil d⁻¹ (under 80% AWC). A similar trend was observed for the Auras + Si combination, which showed a reduction of approximately 45% between the lowest and highest irrigation levels tested.

In contrast, the isolated Si treatment exhibited quadratic behavior (R² = 0.90), with the minimum respiratory activity point estimated at 66.8% AWC (16.19 mg C-CO₂ 100 cm⁻³ soil d⁻¹), increasing again at higher water supply levels (Figure 9a). Under 50% AWC conditions, all mitigation strategies induced higher respiration rates than the controls at the 50% and 60% AWC levels, indicating greater microbial metabolic activity in response to treatments in dry soil.

Comparatively, the stress control (Ctrl. AWC 50%) maintained a constant rate of 15.8 mg C-CO₂ 100 cm⁻³ soil d⁻¹, while the irrigated control (Ctrl. AWC 100%) showed the lowest stable respiratory activity (12.5 mg C-CO₂ 100 cm⁻³ soil d⁻¹) (Figure 9a). At the 80% AWC level, MR levels for treatments containing the biostimulant (Auras and Auras + Si) approached the values observed in the fully irrigated control.

Microbial biomass carbon (MBC) levels were significantly affected by the interaction between mitigation strategies and water availability levels (Table S1). Unlike the pattern observed for respiration, the Auras + Si combination promoted a positive linear increase (R² = 0.79) as irrigation increased, raising the microbial carbon stock from 651.2 μg C g⁻¹ soil (50% AWC) to 903.3 μg C g⁻¹ soil (80% AWC), even surpassing the fully irrigated control (Ctrl. AWC 100%) at the highest water level tested (Figure 9b).

The isolated strategies, in turn, promoted quadratic adjustments (Auras: R² = 0.56; Si: R² = 0.99) in MBC (Figure 9b). The application of Auras resulted in a maximum accumulation point estimated at 65.1% AWC (673.5 μg C g⁻¹ soil), maintaining values higher than the stress control (Ctrl. AWC 50%) throughout the interval. A more significant behavior was recorded for isolated Si, which reached its biomass peak at 65.9% AWC (851.2 μg C g⁻¹ soil). However, under 80% AWC, the silicon treatment promoted a decline in MBC, approaching the levels of the stressed control.

Under 50% AWC conditions, treatments with the biostimulant (Auras and Auras + Si) ensured a higher microbial biomass than soil treated only with Si (Figure 9b), an indication of greater resilience of microbial communities to drought when associated with Bacillus aryabhattai. The stress control (Ctrl. AWC 50%) presented the lowest constant average (448.5 μg C g⁻¹ soil), reinforcing the role of the tested strategies in maintaining and stimulating soil microbial biomass.

Results for the metabolic quotient (qCO₂) reveal distinct patterns of microbial stress as a function of water management and the applied mitigation strategies (Figure 9c). Interventions based on the isolated application of Auras and the Auras + Si combination resulted in significant linear decreases (Auras: R² = 0.75; Auras + Si: R² = 0.74) as water availability increased. In the strategy with the isolated biostimulant, a progressive reduction in the metabolic stress of the microbiota was observed, which, when comparing the 50% AWC drought scenario with 80% AWC, resulted in a 52.4% decrease in energy expenditure per unit of biomass. The Auras + Si association demonstrated an even more pronounced adaptive efficiency between the extreme water levels, providing a 56.7% reduction in qCO₂, which brought microbial activity closer to the equilibrium levels observed in the irrigated control.

The isolated Si treatment, although recording the highest qCO₂ under 50% AWC, promoted quadratic behavior (R² = 0.99) with the sharpest drop until its point of maximum efficiency, estimated at 69.0% AWC (Figure 9c). In this specific interval, the reduction in microbial energy expenditure reached 79.5%. Notably, under the most severe water deficit (50% AWC), the seedlings receiving the biostimulant presented stress indices substantially lower than those treated only with Si. Overall, from 70% AWC onwards, the interventions with B. aryabhattai were able to mitigate microbial stress to levels lower than those of the stressed control (Ctrl. AWC 50%).

4. Discussion

4.1. Plant Physiology

The results consistently demonstrate that water deficit compromised the physiological integrity of the passion fruit seedlings; however, inoculation with B. aryabhattai (via the Auras bioinput), silicon application, and especially their combination, promoted clear mitigating responses dependent on water availability levels. This interaction was particularly evident in electrolyte leakage (Figure 2), where the 50% AWC control maintained high and constant values, while the mitigating treatments progressively reduced membrane damage. Notably, the Auras + Si combination reached its maximum stability point at approximately 69.5% AWC, with EL values close to the fully irrigated control, anticipating this equivalence compared to the isolated treatments, which only reached this level under higher irrigation. This response indicates greater efficiency of the combined strategy in maintaining cellular integrity under intermediate water stress conditions, corroborating evidence that silicon reduces oxidative damage and stabilizes membranes [30], while plant growth-promoting microorganisms contribute to cellular homeostasis through the induction of systemic tolerance mechanisms and physiological modulation [16,17,18,19,31]. Furthermore, this internal cellular protection directly prevented severe structural damage, resulting in the maintenance of turgidity and preservation of seedling architecture, even under severe water restriction, as qualitatively evidenced in Figure 8.

Regarding nutritional status, leaf nitrogen levels revealed distinct responses among treatments, with an evident advantage for the mitigators under higher stress (Figure 3a). While the 50% AWC control presented the lowest N values, the Si and Auras + Si treatments under 70% AWC even outperformed the irrigated control in that condition, evidencing greater nutrient absorption or retention efficiency under water deficit. The positive linear response observed for Auras indicates a continuous effect of the biostimulant on N assimilation, whereas the quadratic behaviors with Si and Auras + Si suggest greater physiological efficiency in intermediate AWC ranges. It is important to note that although B. aryabhattai is not known as a direct N-fixer, it is hypothesized that the observed increase may be attributed to changes in root architecture induced by phytohormones, such as indole-acetic acid (IAA), which increase the soil exploration area and root hair density [32,33,34]. Additionally, microbial activity in the rhizosphere can enhance the mineralization of native organic matter, aiding in nutritional supply even under water stress conditions [35,36].

For leaf silicon, the data show that while no treatment reached the levels of the fully irrigated control, the Auras + Si combination promoted the highest accumulation under intermediate AWC conditions, surpassing both isolated Si and Auras use. Under 50% AWC, this same combination ensured accumulation approximately 66% higher than isolated Si, evidencing a positive interaction between factors, especially under greater water limitation. This result reinforces the hypothesis that the presence of the biostimulant may favor silicon absorption or redistribution, enhancing its physiological effects on reducing cuticular transpiration and mechanical tissue reinforcement. In a study with maize under salt stress [37], the co-application of silicon and Pseudomonas psychrotolerans CS51 increased Si content in plant tissues up to fourfold, indicating greater efficiency in silicon acquisition and use in the soil–plant system. Additionally, this treatment promoted significant increments in plant growth and physiology, including an increase of up to 109.2% in root length and 74.06% in chlorophyll content under stress, highlighting the synergistic role between PGPR and Si in modulating mineral nutrition and stress tolerance.

Photosynthetic pigments showed responses strongly modulated by the treatments (Figure 4), highlighting that from 60% AWC onwards, all mitigators outperformed the fully irrigated control in total chlorophyll (Figure 4a). This result is particularly relevant, as it indicates that the strategies not only compensated for the effects of drought but also enhanced photosynthetic capacity even under suboptimal water conditions. The Auras + Si combination presented an optimal point near 66% AWC for total chlorophyll and chlorophyll a (Figure 4b), suggesting that this range represents a balance between water availability and metabolic efficiency.

Furthermore, the distinct behavior of chlorophyll b (Figure 4c), with a significant increase under Auras at higher AWC levels, reinforces the differential action of the treatments on light-harvesting complexes. The concomitant increase in carotenoids (Figure 4d), primarily with Auras, and elevated levels of flavonoids (Figure 5a) and anthocyanins (Figure 5b) under 50% AWC indicate the activation of photoprotective and antioxidant mechanisms, with these effects being more pronounced in the mitigating treatments compared to the stress control. These findings are in line with the literature associating the accumulation of these metabolites with water stress tolerance, as carotenoids act in surplus energy dissipation and protection against reactive oxygen species, while flavonoids and anthocyanins perform antioxidant and cellular stabilization functions, contributing to the maintenance of the photosynthetic apparatus integrity under adverse conditions [38,39].

4.2. Plant Growth

We believe that the maintenance of nitrogen levels, combined with the robustness of the photopigmentary apparatus and the activation of the antioxidant system, created the metabolic support necessary for the efficient conversion of photoassimilates into plant tissues. This physiological adjustment prevented severe water stress from causing cell cycle arrest, allowing the energy preserved by biochemical protection to be directed toward cell expansion and elongation processes. As a direct consequence of this homeostasis, morphological growth variables reflected the systemic benefit of the applied strategies (Figure 6, Figure 7 and Figure 8), making the differences between treatments even more evident in seedling development.

In this sense, in seedling growth, the differences between treatments become even more evident. While the 50% AWC control showed severely limited growth (Figure 6a), the mitigating treatments promoted significant increments, even under this condition. The Auras + Si combination stood out by presenting the highest absolute gain in height along the water gradient, in addition to showing superior performance to the stress control even at 50% AWC. This result demonstrates that the association was more efficient in sustaining growth under water limitation. Visually, this superiority translated into plants with elongated internodes and early tendril emission (Figure 8c), essential structures for the growth habit of passion fruit. Similar results were observed for dry mass (Figure 7), where isolated silicon promoted stability, while Auras and Auras + Si presented more dynamic responses, highlighted by the linear increment in the combination. The maintenance of biomass values under 50% AWC, close to those observed at intermediate irrigation levels, reinforces the mitigating effect of the treatments, especially silicon, which acted as a structural support and metabolic stability element.

4.3. Soil Biological Indicators

Soil biological indicators confirm that these effects were not restricted to the plant (Figure 9). Microbial respiration (MR) was higher in the Auras and Auras + Si treatments under 50% AWC compared to the controls, indicating intense metabolic activity even in dry soil. Although this high respiratory rate, associated with lower microbial biomass (MBC) values under water deficit, might initially suggest a high maintenance metabolic cost, in the context of soils inoculated with B. aryabhattai, it is hypothesized that this phenomenon may indicate strategic “metabolic activation”. In this scenario, it is believed that the microbiota accelerates nutrient mineralization and carbon cycling at the moment of highest nutritional demand by the plant under stress [32,34], acting as a biological trigger for system resilience. Interestingly, as water availability increased, this respiratory activity was reduced (Figure 9a), approaching the levels of the irrigated control, which signals a transition from survival to biomass accumulation. This pattern is corroborated by the behavior of MBC (Figure 9b), which was maximized by the Auras + Si combination, especially under 80% AWC, even surpassing the fully irrigated control, suggesting that the association promoted not only a necessary initial catabolic activity but also the subsequent channeling of carbon for anabolism, favoring the long-term stability of the soil community.

The metabolic quotient (qCO₂) (Figure 9c) reinforces this interpretation by demonstrating that under 50% AWC, microbial stress was more pronounced, especially in the isolated Si treatment. Conversely, strategies containing the biostimulant promoted drastic reductions in qCO₂ with increasing irrigation, reaching levels close to the irrigated control from 70% AWC onwards. Reductions of 52.4% (Auras) and 56.7% (Auras + Si) suggest that bacterial inoculation was the determining factor in reducing the metabolic cost of the microbiota. Together, these indicators reveal that the presence of B. aryabhattai possibly optimized rhizosphere efficiency, making the system more effective in converting organic carbon into microbial biomass and assisting soil biological functionality under water limitation.

In an integrated context, the results demonstrate that, although all mitigators promoted improvements compared to the stress control, the Auras + Si combination showed superior performance in most variables, especially under moderate to severe water deficit conditions. This superiority was evidenced by the anticipation of membrane stability, higher silicon accumulation, pigment maintenance, greater growth, and robustness of soil microbial biomass. These findings confirm that the joint action of PGPR and silicon enhances water stress tolerance mechanisms. Despite these advances, this study was limited to the initial growth phase of seedlings in a controlled environment, which precludes direct extrapolation of the results to final productivity and field dynamics. Therefore, further research is needed to evaluate whether this synergy is maintained during the reproductive phases of the crop and under different soil and climatic conditions. Furthermore, transcriptomic and proteomic studies could elucidate the specific molecular pathways activated by the interaction between Bacillus and silicon.