Zinc Nanoparticle Effects on the Green Leaf Volatiles and Phyllosphere Bacteriome in Capsicum annum Seedling

Abstract

The application of zinc oxide nanoparticles (ZnONPs) in agriculture is expanding due to their biostimulant potential; however, their influence on plant chemical communication and associated microbial communities remains not fully characterized. This study presents a multi-perspective analysis contrasting the effects of ZnONPs with those of conventional microparticulate ZnO (Bulk) on Capsicum annuum seedlings grown in substrate at 50 and 500 mg kg⁻¹. Results indicate that, at high doses, the bulk material (B500) led to higher foliar zinc accumulation (128.7 mg kg⁻¹) compared to ZnONPs (NP500, 119.7 mg kg⁻¹), a difference potentially linked to nanoparticle aggregation in the soil matrix limiting root uptake. At the physiological level, a distinct response was observed: while Bulk ZnO stimulated superoxide dismutase (SOD) activity, ZnONPs resulted in a marked reduction (93%), suggesting a shift in the antioxidant strategy toward non-enzymatic mechanisms, such as increased total phenol content. Regarding the volatilomic profile, ZnONPs induced specific metabolic alterations in the green leaf volatile (GLV) pathway, characterized by hexanal accumulation and reduced levels of hexanol and hexyl acetate. Additionally, ZnONPs were associated with lower methyl salicylate (MeSA) emissions, whereas the Bulk treatment increased its relative abundance to 41.7%. Finally, metagenomic analysis revealed that zinc treatments modulated the phyllosphere microbiota, favoring the proliferation of Actinobacteria while decreasing the abundance of sensitive taxa, such as Spirochaetes. Taken together, these findings suggest that ZnONPs act as a distinct metabolic modulator, altering internal physiology and chemical signaling.

1. Introduction

Chili pepper (Capsicum annuum) is a crop of high global economic importance [1,2,3]; however, its productivity is frequently limited by biotic and abiotic stresses. This challenge has catalyzed the search for efficient biofortification and protection strategies, positioning nanotechnology as a key tool to enhance crop yield and resilience [4].

Nanotechnology has emerged as a disruptive tool in plant physiology, with the potential to enable precision agriculture through nanofertilizers and nanoelicitors. Among these, zinc oxide nanoparticles (ZnONPs) have attracted particular attention. ZnONPs can function as a slow-release reservoir that, upon dissolution, provides ionic zinc (Zn²⁺). This ionic form is an essential micronutrient that serves as a catalytic or structural cofactor for more than 300 enzymes, including superoxide dismutase (Cu/Zn-SOD) and alcohol dehydrogenase, thereby regulating vital processes ranging from protein synthesis to cell membrane integrity and antioxidant defense [5]. Zn²⁺ acts as a catalytic and structural cofactor not only for nitrate reductase but also for enzymes like glutamate dehydrogenase (GDH), directly linking inorganic nitrogen uptake to amino acid assimilation. Furthermore, Zinc is crucial for ribosomal stability, ensuring the efficiency of protein synthesis under metabolic demand [6]. Zinc modulates hormonal homeostasis through the tryptophan-dependent pathway: it serves as an essential cofactor for tryptophan synthesis, the direct precursor of auxins (Indole-3-acetic acid, IAA) [7]. However, despite these benefits, the transition from biostimulation to phytotoxicity is complex. It is not only dose-dependent but also significantly modulated by particle size and the specific route of exposure.

In the case of chili pepper (Capsicum annuum), the literature identifies a broad effective dosage window depending on the application method. Reported concentrations for ZnONPs typically range from 50 to 1000 mg kg⁻¹ for soil amendments and equivalent ranges in mg L⁻¹ for liquid-based treatments (foliar spray, seed priming) [4,8,9,10,11,12,13,14]. Furthermore, ZnONPs have demonstrated efficacy in mitigating biotic stress through distinct pathways. These include ROS-mediated oxidative damage, Zn²⁺ ion release leading to membrane depolarization, and direct physical disruption of microbial structure. For instance, ZnONPs inhibit the growth of fungal pathogens such as Fusarium oxysporum and Colletotrichum capsica [15], and exhibit nematicidal activity against Meloidogyne incognita [16,17]. Notably, in the case of the Pepper Huasteco Yellow Vein Virus (PHYVV), foliar application of ZnONPs reduces disease severity probably via an indirect induction of systemic resistance in the host plant by modulating defense-related enzymes such as peroxidase (POD) and superoxide dismutase (SOD) [18]. However, direct action, similar to that reported for the Tobacco mosaic virus (TMV) in Nicotiana benthamiana, cannot be ruled out [19].

Beyond growth promotion under normal conditions, ZnONPs also enhances tolerance to abiotic stresses. Evidence from controlled greenhouse studies indicates that ZnONP application can effectively mitigate salt stress. For example, in chili plants exposed to varying salinity levels (0, 25, 50, and 75 mM NaCl), foliar treatments with 1000 ppm ZnONPs (applied three times at 14-day intervals) significantly reduced oxidative damage [20]. This protection was characterized by enhanced chlorophyll fluorescence parameters, increased K⁺ retention, and a marked reduction in electrolyte leakage and H₂O₂ accumulation, particularly under moderate salinity conditions (25–50 mM NaCl) [21,22].

However, the consistency of these positive effects across studies remains a challenge, as the boundary between biostimulation and phytotoxicity is strictly dose- and method-dependent. Several studies report that ZnONPs applied through seed imbibition enhance germination and early seedling growth in Capsicum spp. at moderate concentrations, typically between 100–300 mg L⁻¹ [12,14]. Conversely, concentrations exceeding 500 mg L⁻¹ often induce oxidative stress, inhibit root elongation, or reduce biomass [11,15]. Notably, the delivery method alters this threshold; for instance, foliar application of ZnONPs at 1000 mg L⁻¹ has been shown to improve growth and yield in C. chinense, whereas higher doses (2000 mg L⁻¹) shift the response toward ‘productive stress,’ enhancing capsaicinoid and phenolic content at the expense of vegetative development [13]. Therefore, in addition to concentration and genotype, the application route and exposure time play pivotal roles in determining the plant’s physiological outcome” [23].

Moreover, assessments of nanomaterial effects have predominantly focused on physiological parameters such as biomass and antioxidant enzymes, often overlooking the broader metabolic implications. Recent integrated frameworks propose that environmental drivers regulate plant quality and functional outcomes through a cascade involving oxidative balance, membrane stability, and metabolic reprogramming [24]. In this context, preharvest stress perception extends beyond immediate physiological adjustments to include the modulation of secondary metabolites, particularly Volatile Organic Compounds (VOCs).

Specifically, Green Leaf Volatiles (GLVs) comprising aldehydes, alcohols, and six-carbon (C6) esters synthesized via the lipoxygenase (LOX) and hydroperoxide lyase (HPL) pathways—represent a critical, yet unexplored, dimension of this response [25]. Far from being mere metabolic byproducts, GLVs constitute a chemical “language” that enables plants to activate defenses and alert neighboring organisms to imminent threats (priming) [26,27]. Consequently, potential interference by ZnONPs in key enzymes of this pathway, such as lipoxygenase or alcohol dehydrogenase, could alter the metabolic products [28].

At the same time, the surface of plants (phyllosphere), represents a dynamic microbial habitat on the planet, colonized by a diverse community of bacteria known as the phyllosphere bacteriome [29]. These microorganisms form a biotic barrier that protects against pathogens, fixes atmospheric nitrogen, and modulates host physiology by producing phytohormones [30]. Given that ZnONPs possess intrinsic antimicrobial properties, capable of inducing oxidative stress and membrane disruption in prokaryotic cells [31], their accumulation in leaf tissues following root absorption and translocation could exert significant selective pressure on this ecosystem.

This study proposes a multi-perspective analysis to evaluate the effects of ZnONPs on Capsicum annuum, contrasting them with conventional microparticulate zinc oxide (Bulk ZnO). Although both materials release Zn²⁺ ions upon dissolution, their particle size differences lead to distinct dissolution kinetics and bioavailability [32], allowing for the differentiation of size-dependent mechanism despite common ion release, we examined bioaccumulation and enzymatic antioxidant responses alongside the volatilomic profile (specifically Green Leaf Volatiles and methyl salicylate) and the phyllosphere bacterial community structure.

2. Materials and Methods

2.1. Nanoparticle Characterization

Zinc oxide nanoparticles (ZnONPs) were synthesized in-house (Nanomaterials Synthesis Laboratory, CUCEI) via a hydrothermal method. A reaction mixture was prepared by dissolving 0.35 g of zinc acetate dihydrate (Sigma-Aldrich, St. Louis, MO, USA; ACS reagent, ≥98%) in 13 mL of methanol (Sigma-Aldrich, St. Louis, MO, USA; ACS reagent, ≥99.8%) and 0.24 g of citric acid monohydrate (Sigma-Aldrich, St. Louis, MO, USA; ACS reagent, ≥99.0%) in 67 mL of deionized water. The mixture was homogenized under constant stirring at 40 °C for 30 min, and the pH was adjusted to 13 using 1 M NaOH (prepared from pellets; Sigma-Aldrich, St. Louis, MO, USA, ≥97%). The solution was transferred to a 100 mL autoclave and heated at 150 °C for 19 h under autogenous pressure. The resulting precipitate was collected by filtration, washed thoroughly with deionized water and methanol, and dried at 80 °C for 12 h.

The morphology and size of the ZnONPs were characterized by transmission electron microscopy (TEM) using a JEOL 1010 instrument (JEOL, Tokyo, Japan) operated at 80 kV. Samples were prepared by dispersion in an organic solvent and deposited onto copper grids for imaging at 120,000× magnification. Further surface analysis was performed by scanning electron microscopy (SEM) using a FEI Helios Nanolab 600 (FEI Company, Hillsboro, OR, USA). The nanoparticles were mounted on double-sided carbon tape, and micrographs were acquired at an accelerating voltage of 5 kV using a secondary electron detector to enhance surface contrast and minimize potential electron beam damage. Finally, surface charge was assessed by zeta potential measurements using a Zetasizer Nano ZS (Malvern Panalytical, Malvern, UK). ZnONPs were dispersed in isopropanol to evaluate the effect of solvent polarity on the electrical double layer and electrostatic stabilization.

2.2. Experimental Design and Growth Conditions

The experiment was conducted using a randomized complete block design. Seeds of Capsicum annuum var. Jalapeño M. (Caloro^®, Jalisco, Mexico) were sown in a substrate mixture of agricultural soil and perlite (1:1, v/v). According to previous work, ZnONPs exhibit a hormetic effect: low or moderate doses (1–100 ppm) have a positive effect, whereas high doses (>100 ppm) have a negative effect on plant development [33,34,35]. Furthermore, the Zn requirement in most crops is 15 to 50 ppm in mature leaves [36]. Consequently, the 50 mg kg⁻¹ dose was selected as a soil enrichment strategy to promote biofortification and ensure tissue sufficiency. Thus, the selection of 50 and 500 mg kg⁻¹ was based on the reported hormetic thresholds for ZnONP and on the Zn requirements in crops. While 50 mg kg⁻¹ represents a plausible agronomical dosage for soil enrichment, the 500 mg kg⁻¹ dose was intentionally selected as an extreme boundary scenario to evaluate the plant’s physiological limits, rather than as a recommended field application. The experimental design was complemented by ZnO bulk controls at the same concentrations (B50 and B500) and a control group (Ctrl) without supplementation, allowing comparison of the effects of nanometric-scale materials with conventional materials. To ensure the homogeneous distribution of ZnONPs and bulk particles within the substrate, a geometric dilution method was employed. Briefly, the weighed amount of zinc was first manually homogenized with a small aliquot of substrate, which was then progressively mixed into larger soil volumes until the total mass for each treatment was reached. This procedure was performed individually for each batch to minimize heterogeneity. For each treatment, amended soil was distributed into 500 mL pots, and three seeds were sown in each pot. Pots were kept in darkness at 27 °C for 7 days to promote germination and subsequently transferred to a greenhouse at the Colegio de Postgraduados, Campus Montecillo, Mexico (19°27′40″ N, 98°54′13″ W). Seedlings were grown under greenhouse conditions (12 h light/12 h dark photoperiod, 25–35 °C). The onset of the zinc treatment occurred at sowing (BBCH 00), exposing the plants to the amended substrate throughout the germination and vegetative development phases. Plants were irrigated with a nutrient solution (120 ppm N, 24 ppm P, 155.2 ppm K) according to crop requirements. Final sampling for volatilomic and biochemical analyses was performed when the plants reached the BBCH 18 stage (eight true leaves unfolded on the main shoot) [37].

2.3. Sample Processing and Preservation

For volatilomic and metagenomic analyses, each sample (biological replicate) consisted of fresh leaves from nine plants. From the remaining seedlings, true leaves and cotyledons were harvested and pooled to obtain three composite biological replicates per treatment (n = 3); each composite sample consisted of tissue from 15 seedlings. These samples were immediately frozen in liquid nitrogen and lyophilized in a FreeZone 4.5 freeze-dryer (LABCONCO^®, Kansas City, MO, USA) at −52 °C and 0.052 mBar for 72 h. Finally, the dried tissue was pulverized and stored at −80 °C.

2.4. Zinc Quantification in Plant Tissues

Total zinc concentration in seedling leaves was determined by atomic absorption spectroscopy (AAS). A 200 mg aliquot of lyophilized tissue was subjected to acid digestion using a mixture of 3 mL of 65% nitric acid and 3 mL of 70% perchloric acid at 300 °C for 40 min. After cooling, the digest was diluted to a final volume of 25 mL with 2% (v/v) nitric acid and filtered through a 0.45 µm nylon syringe filter. Zinc concentration was measured using a SpectraAAS 220 FS spectrophotometer (Varian, Palo Alto, CA, USA) at a wavelength of 213.9 nm. For each biological replicate (n = 3), seven technical replicates were performed and averaged for statistical analysis. Quantification was performed against a standard curve prepared from a 100 mg L⁻¹ stock solution, with concentrations ranging from 0 to 2.0 mg L⁻¹. Results were expressed as mg Zn per kg of dry weight (mg kg⁻¹ DW).

2.5. Non-Enzymatic Antioxidant Activity and Total Polyphenols

Methanolic extracts were prepared by macerating 20 mg of lyophilized tissue in 10 mL of 80% (v/v) methanol for 24 h with orbital shaking in darkness. The mixture was centrifuged at 4500 rpm for 10 min, and the supernatant was collected and stored at −20 °C. For each treatment, three independent biological replicates were processed, and all subsequent biochemical assays were performed in technical triplicate (n = 3). All absorbance measurements were performed on a Multiskan SkyHigh microplate spectrophotometer (Thermo Fisher Scientific^®, Waltham, MA, USA).

2.5.1. ABTS Assay

Radical scavenging activity was determined following the method of Stratil et al. [38]. The assay was performed at 734 nm, and results were expressed as mmol Trolox equivalents per gram of dry weight (mmol TE g⁻¹ DW).

2.5.2. DPPH Assay

Free radical scavenging activity was measured according to Brand-Williams et al. [39]. Absorbance was read at 515 nm, and results were expressed as mmol TE g⁻¹ DW.

2.5.3. FRAP Assay

Ferric reducing antioxidant power was assayed as described by Benzie and Strain [40]. Absorbance was measured at 593 nm, and results were expressed as mmol TE g⁻¹ DW.

2.5.4. Total Polyphenol Content

Total polyphenols were quantified using the Folin–Ciocalteu method [41]. Gallic acid (purity ≥ 97.5%, Cat. No. G7384, Sigma-Aldrich, St. Louis, MO, USA) was used as the standard. Absorbance was measured at 750 nm after 2 h of incubation. Results were expressed as mg gallic acid equivalents per gram of dry weight (mg GAE g⁻¹ DW).

2.6. Antioxidant Enzyme Activity

Enzyme extracts were prepared by homogenizing 20 mg of lyophilized tissue in 2 mL of an ice-cold extraction buffer [0.1 M potassium phosphate (pH 7.5), 0.5 mM EDTA, 2% (w/v) PVP, and 0.5% (v/v) Triton X-100] using an ultrasonic bath for 10 min. The homogenate was centrifuged at 14,000 rpm for 10 min, and the supernatant was stored at −80 °C. For each treatment, three independent biological replicates were prepared, and subsequent enzymatic assays were performed in technical triplicate.

2.6.1. Protein Quantification

Total protein content was determined using the Pierce™ BCA Protein Assay Kit (Thermo Fisher Scientific^®) with bovine serum albumin (BSA) as the standard. Absorbance was read at 562 nm.

2.6.2. Guaiacol Peroxidase (GPOX)

Guaiacol peroxidase (GPOX, EC 1.11.1.7) activity was determined spectrophotometrically following Castillo et al. (1984) [42]. The protocol was adapted to a microplate format by scaling down reaction volumes while maintaining standard concentrations. The reaction mixture consisted of 52.6 mM potassium phosphate buffer (pH 6.1), 16.8 mM guaiacol, and 2.1 mM H₂O₂. Reactions were initiated by adding 190 µL of the mixture to 10 µL of enzyme extract (n = 3). Tetraguaiacol formation was monitored at 470 nm every 15 s for 5 min using a Multiskan SkyHigh spectrophotometer (Thermo Scientific, Rockford, IL, USA). Activity was calculated using the molar extinction coefficient ε470 = 26.6 mM⁻¹ cm⁻¹ and expressed as µmol mg⁻¹ protein min⁻¹.

2.6.3. Ascorbate Peroxidase (APX)

Ascorbate peroxidase (APX; EC 1.11.1.11) activity was determined according to Nakano & Asada (1980) by monitoring ascorbate oxidation at 290 nm [43]. Enzyme extract (10 µL) was mixed with 140 µL of reaction medium (71.4 mM potassium phosphate pH 7.0, 0.14 mM EDTA, and 0.71 mM ascorbic acid) in a 96-well microplate. The reaction was initiated by adding 10 µL of 0.4 mM H₂O₂ (n = 3). Kinetics were monitored alongside an ascorbic acid calibration curve (0–1000 µmol/L) using a Multiskan SkyHigh spectrophotometer (Thermo Scientific, Rockford, IL, USA) at 15 s intervals for 5 min. Activity was calculated via linear regression against the standard curve and expressed as µmol oxidized ascorbate mg⁻¹ protein min⁻¹.

2.6.4. Catalase (CAT)

Catalase (CAT; EC 1.11.1.6) activity was quantified using the Catalase Colorimetric Activity Kit (Thermo Scientific, Rockford, IL, USA) according to the manufacturer’s instructions. Samples or bovine catalase standards (0–5 U/mL) (25 µL) were incubated with 25 µL of hydrogen peroxide for 30 min at room temperature. Subsequently, 25 µL of substrate solution and 25 µL of horseradish peroxidase were added, followed by a 15 min incubation. Absorbance was measured at 560 nm using a Multiskan SkyHigh spectrophotometer (Thermo Scientific, Rockford, IL, USA). Data were analyzed via four-parameter logistic (4PL) regression using SkanIt™ software v 7.1, and results were expressed as µmol decomposed H₂O₂ mg⁻¹ protein min⁻¹.

2.6.5. Superoxide Dismutase (SOD)

Superoxide dismutase (SOD; EC 1.15.1.1) activity was determined by measuring the inhibition of the photochemical reduction of nitroblue tetrazolium (NBT) described by Dhindsa et al. (1981) [44]. Enzyme extract (10 µL, 1:1 and 1:10 dilutions) was mixed with 180 µL of reaction buffer (55.6 mM potassium phosphate pH 7.8, 55.6 µM NBT, 0.11 mM EDTA) and 10 µL of 36.0 µM riboflavin. The reaction was initiated by illumination under a 15 W fluorescent lamp for 15 min. Absorbance was measured at 560 nm using a Multiskan SkyHigh spectrophotometer (Thermo Scientific, Rockford, IL, USA). One unit (U) was defined as the amount of enzyme required to cause 50% inhibition of NBT reduction relative to a no-enzyme control. Results were expressed as U mg⁻¹ protein.

All enzyme activities were normalized by protein content and expressed in their respective units per mg of protein.

2.7. Volatile Organic Compound (VOC) Analysis

Volatile organic compounds (VOCs) were analyzed using Headspace-Solid Phase Microextraction (HS-SPME) coupled to GC-MS. Two independent biological replicates (n = 2) were processed per treatment, each consisting of a 1 g composite sample pooled leaves from five seedlings in 20 mL of 0.4% (w/v) NaCl within 40 mL vials. Extraction and Chromatographic Conditions: Samples were equilibrated for 5 min at 60 °C, followed by a 30 min extraction at the same temperature using a DVB/CAR/PDMS fiber (80 µm; DVB/CAR/PDMS fiber (Supelco, Bellefonte, PA, USA), conditioned at 260 °C. These parameters were optimized for C. annuum by Mazida et al. [45] to maximize recovery. Desorption was performed at 240 °C (15 min, splitless) in an Agilent 7890B GC equipped with an HP-5ms Ultra Inert column (30 m × 250 µm × 0.25 µm). Helium was used as the carrier gas (1 mL/min). The thermal program consisted of: 40 °C (3 min), a 10 °C/min ramp to 280 °C, and a 2 min hold. MS detection (Agilent 5977A (Agilent Technologies, Santa Clara, CA, USA)) operated at 70 eV (scan range 30–550 amu; 13.8 spectra/s) with transfer line and source temperatures at 250 °C and 200 °C, respectively. Data Processing: Spectral analysis was performed using MassHunter Qualitative Analysis v.6.0. Empty vials and saline solution blanks were analyzed under the same conditions, and background compounds were removed from the dataset. Compound identification relied on the NIST 14 library and confirmation via Kovats Retention Indices (KRI) relative to C7–C40 n-alkanes, Lucero et al. [46]. Relative abundance was calculated as the average peak area percentage.

2.8. Phyllosphere Microbiome Analysis

Sampling and DNA Extraction: Total genomic DNA was extracted from 200 mg of fresh composite tissue samples (hypocotyl, cotyledons, true leaves, and apical buds). To ensure statistical reliability and correct biological variance estimation, three independent biological replicates were processed per treatment (n = 3). Each replicate was formed by pooling fresh tissue from three randomly selected seedlings, yielding a total of 9 plants sampled per treatment. Extraction was conducted using the DNeasy PowerSoil Pro Kit (QIAGEN Inc., Aarhus, Denmark), following the manufacturer’s protocol. Library Preparation and Amplification: The full-length V1-V9 hypervariable regions of the 16S rRNA gene were amplified by PCR using primers 27F and 1492R with Ultra DNA Polymerase (Jena Bioscience GmbH, Jena, Germany). Negative extraction controls (processing blanks) were included in the workflow; the complete absence of detectable amplification in these controls confirmed the lack of background contamination. The resulting amplicons from biological samples were purified using the PCR Purification Kit (Jena Bioscience GmbH). Sequencing and Bioinformatics: Purified amplicons were sequenced at Secoya Labs S.C. (Mexico City, Mexico). Sequencing libraries were prepared using the Ligation Sequencing Kit V14 (Oxford Nanopore Technologies, ONT, Oxford, UK). To minimize sample crosstalk, a dual-barcoding scheme was applied. Sequencing was performed on an R10.4.1 Flow Cell. Although a commercial mock community was not included, library integrity was validated using an internal 16S amplicon standard. Raw data were basecalled using Dorado v7.3.9 with a minimum Q-score of 10 and a minimum read length of 200 bp to exclude low-quality sequences. To ensure taxonomic reliability and filter out PCR chimeras/errors, reads were assembled into consensus contigs using Flye 2.9.4 and polished with Medaka. Taxonomic classification of the high-quality consensus sequences was performed against the NCBI 16S reference database using Minimap2.

2.9. Statistical Analysis

Statistical Analysis Data were subjected to descriptive statistical analysis and are presented as the mean ± standard deviation (SD) of the independent biological replicates. Prior to hypothesis testing, data distribution was checked for normality using the Shapiro–Wilk test, and homogeneity of variance was verified using Brown-Forsythe test. For physiological and biochemical datasets, differences among treatments were evaluated using a one-way Analysis of Variance (ANOVA). When significant differences were detected (p < 0.05), mean comparisons were performed using Tukey’s Honest Significant Difference (HSD) post hoc test. For the metagenomic analysis, alpha diversity indices were compared using ANOVA. To assess differences in microbial community structure (beta-diversity), a Permutational Multivariate Analysis of Variance (PERMANOVA) was performed using the adonis2 function from the vegan package in R. This analysis was based on Bray–Curtis dissimilarities calculated from Hellinger-transformed abundance data, utilizing 999 permutations to determine statistical significance. All statistical analyses were performed using R software v.4.3.1 or GraphPad Prism ^® v.9.2.0, and a p-value < 0.05 was considered statistically significant.

3. Results and Discussion

3.1. Material Characterization

Both ZnO materials were characterized using electron microscopy to evaluate their morphology and size (Figure 1). SEM micrographs revealed that the bulk form (Figure 1a) possessed a significantly larger particle size compared to the synthesized ZnONPs (Figure 1b). Higher magnification via TEM (Figure 1c) revealed that the ZnONPs predominantly exhibit a rectangular morphology, with a size range of 20–190 nm and an average size of 72 nm. Furthermore, the surface charge was assessed in isopropanol, yielding a zeta potential of −1.47 mV. This near-neutral value indicates a low surface charge and minimal electrostatic repulsion, consistent with the solvent’s low polarity, which hinders the formation of a stable electrical double layer and suggests that dispersion is primarily driven by attractive forces.

3.2. Zn Cuantification in Plant Tissue

The total Zn concentration in seedling leaves was determined using atomic absorption spectroscopy (AAS). One-way ANOVA revealed that both the source and the concentration of zinc significantly influenced foliar accumulation. Post hoc analysis using Tukey’s test confirmed that all treated groups (NP and Bulk) had significantly higher Zn levels than the control (p < 0.05). Specifically, NP50 and B50 treatments resulted in an 8.3–9.0% increase in Zn concentration, with no statistically significant difference between these two sources at low doses. However, at the higher concentration (500 mg kg⁻¹), a substantial increase was observed: 35.1% for NP500 and 45.3% for B500 compared to the control (

Table 1. Zn content on freeze-drive leaves from seedlings treated. Treatments were either bulk ZnO (B50 and B500) or ZnNPs (NP50 or NP500).

Treatment	[Zn], mg kg−1
Ctrl	88.55 ± 3.47 a
NP50	96.60 ± 2.98 b
B50	95.92 ± 2.51 b
NP500	119.7 ± 4.00 c
B500	128.7 ± 3.15 d

Values are expressed as mg kg^–1 of dry weight. [Zn] ± SD, n = 3. Different letters indicate statistically significant differences between treatments (p < 0.05) according to Tukey’s post hoc test.

)

Notably, the statistical analysis highlighted a significant difference (p < 0.05) between the NP500 and B500 treatments. This observation indicates that while Zn bioaccumulation followed a dose-dependent pattern, the efficiency of uptake became dependent on the zinc source (nano vs. bulk) specifically at the highest concentration (500 mg kg⁻¹). In this scenario, the bulk form (B500) resulted in a 7.5% higher Zn accumulation than the nanometric form (NP500).

Our results align with Hoon et al. [47], who observed that in Brassica chinense, Zn accumulation was comparable or higher when using ionic/bulk sources relative to ZnONPs at elevated doses (>500 mg kg⁻¹). This finding contrasts with the general assumption that smaller particle sizes invariably lead to higher bioavailability [48,49]. Based on current colloidal chemistry principles, this reduced efficiency of ZnONPs at high doses is likely attributable to rapid agglomeration in the soil matrix, where the formation of larger clusters reduces the effective surface area available for dissolution and root uptake [47,50]. Furthermore, potential complexation with organic matter may limit the free Zn²⁺ available from the nanoparticle source compared to the bulk material in this specific substrate [51].

3.3. Bacterial Community of the Capsicum annuum Phyllosphere

Unlike the extensively studied rhizosphere [52,53,54], the impact of ZnONPs on the phyllosphere bacteriome remains strictly limited. To address this, we evaluated the community shifts in C. annuum seedlings using nanopore sequencing. Analysis of alpha diversity indices revealed no significant impact of zinc treatments on species richness or evenness (

Table 2. Alpha-diversity indices of the phyllosphere microbiome across treatments.

Treatment	Shannon Diversity Index	Simpson’s Index	Berger-Parker Index
Ctrl	4.11	0.960	0.100
NP50	4.16	0.960	0.100
B50	4.05	0.960	0.110
NP500	4.23	0.963	0.097
B500	4.30	0.960	0.100

Ctrl: Control (no Zn supplementation); NP: Zinc oxide nanoparticles; B: Bulk zinc oxide. The numerical suffixes (50 and 500) indicate the concentration applied to the substrate in mg kg⁻¹.

). The Shannon and Simpson indices remained stable across all groups, indicating that neither the concentration (50 vs. 500 mg kg⁻¹) nor the source (NP vs. Bulk) compromised the overall bacterial diversity. Furthermore, the Berger-Parker index (0.097–0.110) confirmed a balanced community structure, where the most abundant species represented only ~10% of the total population, ruling out the dominance of any single opportunistic taxon under stress conditions. Microbiome analysis of shared operational taxonomic units (OTUs) defined a robust “core microbiome”: out of 937 identified species (cutoff ≥ 10 reads), 715 (76.3%) were conserved across all samples regardless of treatment. The Venn diagrams (Figure 2) illustrate that low zinc concentrations exerted negligible pressure on community composition. However, at high concentrations (500 mg kg⁻¹), a divergence was observed in the non-shared fraction of the microbiome. Contrary to the assumption that nanoparticles exert a unique selective pressure, the loss or gain of unique species was primarily dose-dependent, with both NP500 and B500 treatments showing a comparable deviation from the control.

Unlike the stability observed in alpha diversity, the PERMANOVA analysis confirmed that zinc treatments significantly altered the overall microbial community structure (F = 2.32, p = 0.027). Regarding taxonomic composition (Figure 3), Proteobacteria remained the dominant and most stable phylum across all groups, highlighting its resilience as a core component of the C. annuum phyllosphere. However, distinct lineage-specific responses were driven by the concentration applied: The abundance of Spirochaetes fluctuated across all ZnO treatments compared to the control, suggesting a general sensitivity of this phylum to zinc exposure regardless of the dosage. Cyanobacteria showed a specific, slight increase in relative abundance at low Zinc concentrations (50 mg kg⁻¹). Conversely, Aquificae and Firmicutes were enriched specifically at high concentrations (500 mg kg⁻¹).

Notably, the strongest positive response to high Zinc pressure was observed in Actinobacteria, which increased considerably in abundance in both NP500 and B500 treatments. This proliferation is consistent with the well-documented ecological traits of Actinobacteria, a phylum known for its intrinsic resistance to heavy metal stress and desiccation [55,56]. As Gram-positive bacteria, many members of this group possess specific metal-efflux pumps and the ability to produce protective secondary metabolites [57].

3.4. Metabolic Activities

3.4.1. Antioxidants

To evaluate the plant’s systemic defense against oxidative stress, we analyzed the antioxidant capacity using three complementary assays (ABTS, DPPH, and FRAP) and quantified Total Phenolic Content (TPC) see Figure 4a–c. Statistical analysis revealed that zinc treatments significantly enhanced the antioxidant capacity of C. annuum seedlings compared to the control across all three assays (ANOVA, p < 0.05). DPPH Assay: showed the most distinct response. While NP50, B50, and NP500 treatments resulted in a moderate increase of 20–22%, the high-concentration bulk treatment (B500) exhibited a significantly higher spike, increasing capacity by 50% compared to the control (Tukey’s HSD, p < 0.01). ABTS and FRAP: Both assays confirmed a general upregulation of antioxidant activity in treated plants (p < 0.05), with no statistically significant differences observed between the nano (NP) and bulk (B) forms at equivalent concentrations.

These results align with findings by Sánchez-Pérez et al. [58] and García-López et al. [11], who reported a concentration-dependent increase in antioxidant activity in Capsicum species. However, our observation that B500 elicited a stronger DPPH response than NP500 contrasts with studies on other species like Portulaca oleracea [59], where nanoparticles typically induce higher activity than bulk forms. This suggests that in our specific soil–plant system, the bulk material at high doses remains chemically active, triggering a robust scavenging response potentially due to sustained Zn²⁺ release.

Regarding total phenols, a different pattern emerged. While low/medium stress treatments (NP50, B50, B500) increased TPC by 13–16% relative to the control (p < 0.05), the highest nanoparticle concentration (NP500) showed no statistically significant difference from the control. Notably, a comparison between NP50 and NP500 reveals a 13% decrease in phenolic content as the nanoparticle dose increased. This creates a divergence: while NP500 maintained high antioxidant capacity (via ABTS/FRAP), its phenolic content dropped. The observed divergence between TPC and total antioxidant capacity in the NP500 treatment suggests a distinct physiological adjustment under high nanometric stress. While Adnan et al. [60] reported linear TPC increases in C. annuum, our data implies that at 500 mg kg⁻¹ of ZnONPs, the accumulation of phenolic compounds does not follow a linear trend. This raises the possibility that the sustained antioxidant capacity observed in NP500 (Figure 4a–c) may be supported by compensatory non-phenolic mechanisms, such as the mobilization of other low-molecular-weight antioxidants or distinct enzymatic responses (as discussed in Section 3.2). These findings point towards a complex, non-linear metabolic response that differentiates the high-dose nano-treatment (NP500) from the bulk counterpart (B500).

3.4.2. Antioxidant Enzymes

The enzymatic defense machinery was evaluated by quantifying the activity of Guaiacol peroxidase (GPOX), Ascorbate peroxidase (APX), Catalase (CAT), and Superoxide dismutase (SOD). Statistical analysis revealed that the zinc treatments induced highly specific and distinct enzymatic profiles (ANOVA, p < 0.05).

Regarding Guaiacol peroxidase (GPOX, EC 1.11.1.7), the results (Figure 5a) showed a general decrease in activity across all zinc-treated seedlings compared to the control (p < 0.05). However, a detailed analysis by concentration reveals a distinct difference dependent on particle size. At the lower dose (50 mg kg⁻¹), the nanometric form (NP50) resulted in a substantial inhibition of 46.4%, whereas the bulk equivalent (B50) showed a minor reduction of only 4.8%. At the highest concentration (500 mg kg⁻¹), both forms induced decreases in activity (33.5% for NP500 and 45.4% for B500). These findings indicate a higher specific activity of the nanomaterial: to achieve a ~45% inhibition in GPOX activity, only 50 mg kg⁻¹ of ZnONPs were required, whereas a ten-fold higher concentration (500 mg kg⁻¹) of bulk zinc oxide was necessary to produce a comparable effect.

As for Ascorbate peroxidase (APX, EC 1.11.1.1) enzyme activity, the results showed a distinct response pattern compared with GPOX (Figure 5b). Interestingly, there was no difference between treatments B50 and NP500 compared with the control. Only in the treatment with nanoparticles at 50 mg/kg (NP50) was a significant decrease in APX activity observed, with activity 12.2% lower than in the control group. In contrast, treatment with 500 mg/kg ZnO in bulk form (B500) increased enzyme activity by 13.5%.

Catalase (CAT, EC 1.11.1.6), activity decreased across all treatments independent of concentration (Figure 5c). However, the extent of inhibition varied by form: while ZnONPs reduced activity by a modest average of 11.2%, bulk treatments (B50 and B500) caused a significantly more pronounced suppression of 38.6% relative to the control.

Superoxide dismutase (SOD) exhibited the most striking divergence based on particle size (Figure 5d). While bulk treatments nearly doubled activity (+94.6%), ZnONPs caused a drastic reduction of up to 93.0%. This sharp inhibition aligns with reported mechanisms of nanoparticle-induced conformational changes or active site blocking [61,62]. Consequently, this enzymatic compromise likely necessitated the compensatory upregulation of non-enzymatic defenses (phenols/DPPH) described previously [58,63].

The observed enzymatic profiles underscore the multifactorial nature of plant-nanoparticle interactions. The generalized downregulation of GPOX and CAT in our study aligns with the “inflection point” phenomenon described in maize and tomato [58,63]. In those studies, antioxidant activities initially peak but eventually collapse once a toxicity threshold is exceeded; our data suggests that the concentrations used here (50–500 mg kg⁻¹) likely surpassed this physiological threshold for enzymatic maintenance.

Furthermore, the contrast with previous C. annuum studies reporting generalized enzymatic activation [20,60] highlights the critical role of the exposure pathway. Unlike foliar application, which often triggers immediate defense signaling, substrate administration involves root-complexation mechanisms that may alter systemic bioavailability and stress signaling. Finally, genotypic specificity remains a decisive factor. As evidenced by opposing enzymatic strategies in different Jalapeño varieties (e.g., ‘Don Pancho’ vs. ‘Don Benito’) [18], the plant’s response is not uniform but modulated by its specific genetic background.

3.5. Volatiles Analysis

For the analysis of volatile organic compounds (VOCs), solid phase microextraction (SPME) was performed according to the conditions detailed in the methodology.

To perform an initial screening and compare the diversity and abundance of VOCs, a global analysis was conducted. Those compounds with an abundance of at least 1300 arbitrary units (A.U.) in the chromatograms and showing distinct fragmentation patterns were considered. We analyzed the emission of Volatile Organic Compounds (VOCs). A global screening revealed that high-dose nanoparticle stress (NP500) elicited the most diverse metabolic response, generating 61 distinct compounds, 10 of which were unique to this treatment (Figure 6). In contrast, the bulk treatments and lower nano-doses maintained a basal diversity similar to the control (45–47 compounds). The increase in diversity observed in the NP500 treatment is consistent with previous studies, such as that of Piesik et al. [64] who observed increased production and higher concentrations of VOCs in Brassica napus plants treated with iron nanoparticles and elicitors, compared to those treated only with elicitors, demonstrating a response in VOC concentration dependent on nanoparticle concentration.

Subsequently, a targeted screening filtered for high-confidence compounds based on strict identification (Kovats index deviation ≤ 10) and abundance thresholds (≥1.0%). The resulting profile, detailed in

Table 3. Relative abundance of volatile compounds identified in C. annuum seedlings under the different treatments.

			% Area
Name	FM (1)	MM (2)	Ctrl	NP50	B50	NP500	B500	IK(exp) (3)	IK(lit) (4)
(Z)-3-Hexenal	C6H10O	98.1	3.5	7.5	12.7	-	1.2	792.7	799.5
n-Hexanal	C6H12O	100.2	5.7	7.9	4.4	6.9	12.7	794.3	793.0
(E)-2-Hexenal	C6H10O	98.1	11.4	11.9	12.9	8.1	6.1	854.2	854.0
(Z)-3-Hexenol	C6H12O	100.2	7.7	11.1	9.2	5.7	3.3	858.6	858.0
(E)-2-Hexenol	C6H12O	100.2	3.9	3.6	2.0	0.8	0.5	869.3	869.0
p-Xylene	C8H10	106.2	-	-	-	2.1	0.8	869.7	865.0
n-Hexanol	C6H14O	102.2	8.0	4.3	6.7	-	2.2	871.3	871.0
(Z)-3-Hexenyl Acetate	C8H14O2	142.2	15.9	14.4	7.3	9.0	0.8	1005.2	1007.0
n-Hexyl Acetate	C8H16O2	144.2	2.7	1.9	0.8	1.0	-	1012.4	1012.0
2-Hexenyl Acetate	C8H14O2	142.2	3.5	3.7	0.5	1.4	-	1015.4	1017.0
Indene	C9H8	116.2	0.1	1.3	-	0.03	-	1046.5	1051.0
Linalool	C10H18O	154.2	1.4	-	0.05	-	-	1097.8	1098.0
n-Nonanal	C9H18O	142.2	1.1	1.7	3.4	4.9	3.8	1102.2	1102.0
Methyl Salicylate	C8H8O3	152.1	24.4	12.0	0.2	16.6	41.7	1196.9	1196.0
n-Decanal	C10H20O	156.3	0.5	1.2	2.8	3.5	2.6	1204.0	1204.0
2-Ethyl-3-hydroxyhexyl-2-methylpropanoate	C12H24O3	216.3	0.5	1.3	0.1	2.8	2.3	1377.3	1373.0
n-Decyl Acetate	C12H24O2	200.3	-	1.4	-	0.5	-	1406.2	1408.0
β-Farnesene	C15H24	204.4	-	-	17.0	-	1.3	1457.1	1457.0
1-Dodecanol	C12H26O	186.3	0.9	3.0	1.0	3.4	3.4	1472.3	1472.0
2,2,4-Trimethyl-1,3-pentanediol diisobutyrate	C16H30O4	286.4	0.6	1.7	-	3.9	1.9	1597.4	1587.5
α-Bisabolol	C15H26O	222.4	0.3	-	-	2.9	1.1	1675.8	1685.0
2-Ethylhexyl salicylate	C15H22O3	250.3	0.2	0.4	-	1.0	0.8	1810.9	1812.0
Isopropyl myristate	C17H34O2	270.5	-	-	0.1	1.2	0.3	1821.4	1823.0

⁽¹⁾ FM: Molecular formula. ⁽²⁾ MM: Molecular mass (Atomic mass units). ⁽³⁾ IK_exp: Kovats indices calculated from retention times on a HP-5ms Ultra Inert column. ⁽⁴⁾ IK_lit: Kovats index from the NIST Chemistry Web Book [65]. Ctrl: Control (no Zn supplementation); NP: Zinc oxide nanoparticles; B: Bulk zinc oxide. The numerical suffixes (50 and 500) indicate the concentration applied to the substrate in mg kg⁻¹.

, comprises diverse chemical classes including aldehydes, C6-alcohols, esters, terpenes, and aromatic derivatives.

Methyl salicylate (MeSA) constituted the predominant volatile fraction across most treatments, accounting for 24.4% of the relative abundance in the control group. However, zinc supplementation induced a divergent response dependent on particle form: while ZnONPs (NP50/NP500) downregulated MeSA emissions (12.0–16.6%), the high-dose bulk treatment (B500) triggered a substantial upregulation to 41.7%. Given MeSA’s established role as a systemic stress signal and modulator of downstream volatile synthesis [66,67,68,69], these data suggest that Bulk ZnO amplifies this salicylic acid-mediated pathway, whereas nanomaterials tend to suppress it.

A distinct pattern was observed in the B50 treatment. In this group, MeSA relative abundance decreased to 0.2%, coinciding with the presence of β-farnesene (17.0%), a sesquiterpene primarily detected in bulk treatments. This inverse relationship suggests a shift in the volatile profile, favoring terpenoid accumulation over phenolic signaling at this specific dosage. To visualize the broader distribution of these differential profiles, a heat map was generated (Figure 7).

The terpenoid profile displayed a compound shift dependent on stress severity (Figure 7). While the production of linalool (a monoterpene) was suppressed in most zinc treatments, appearing only in the control (1.4%) and B50 (0.05%), high concentrations of ZnO triggered the synthesis of α-bisabolol. This sesquiterpene, absent at lower doses, increased significantly in NP500 (2.9%) and B500 (1.1%) treatments, serving as a distinct marker of high-stress conditions.

Concurrently, a dose-dependent accumulation of long-chain aldehydes (nonanal and decanal) was observed across all treatments. Since these volatiles originate from the fragmentation of lipoperoxides [68], their increasing relative abundance. Specific stress-associated metabolites emerged exclusively at high concentrations: p-Xylene and isopropyl myristate were detected primarily in NP500 and B500, while decyl acetate was identified uniquely in nanoparticle-treated seedlings. The presence of these metabolites could indicate their induction in response to stress conditions associated with high ZnO concentrations in the growing soil.

To complete the volatile characterization, we focused on Green Leaf Volatiles (GLVs), a group of compounds critical for stress signaling and plant defense [69]. As illustrated in Figure 8, their biosynthesis begins with the oxidation of fatty acids (specifically linoleic and α-linolenic by Lipoxygenases (LOX) to form hydroperoxides. These intermediates are subsequently cleaved by Hydroperoxide Lyases (HPL) into 6-carbon aldehydes. Finally, these aldehydes can be isomerized, reduced to alcohols, and acetylated to form esters, generating the diverse GLV profile [70]

Analyzing the derivatives of linoleic acid, a distinct trend was observed for hexanal (the aldehyde precursor). Its relative abundance increased in most treatments (reaching 12.7% in B500 vs. 5.7% in Control), with the exception of the B50 group. In contrast, the downstream metabolites: hexanol and hexyl acetate, exhibited a marked reduction or suppression, particularly in the high-concentration treatments (NP500 and B500). This pattern (accumulation of the upstream aldehyde coupled with depletion of the downstream alcohol) suggests a potential interference with the reductive step, possibly involving the Cinnamaldehyde/Hexenal Reductase (CHR) or Alcohol Dehydrogenase (ADH) enzymes, leading to the observed accumulation of hexanal.

A similar, albeit dose-dependent, behavior was noted in the α-linolenic acid pathway. The production of aldehydes (Z)-3-hexenal and (E)-2-hexenal appeared stimulated at low concentrations (NP50 and B50) but decreased at higher zinc loads (NP500 and B500). This suggests a biphasic modulation of the pathway: activation at lower stress levels and potential suppression at higher concentrations.

Regarding the downstream alcohols, (Z)-3-hexenol mirrored the trend of its aldehyde precursor (increasing at low doses, decreasing at high doses). However, its isomer (E)-2-hexenol showed a generalized, concentration-dependent decline across all treatments. Finally, the terminal esters: (Z)-3-hexenyl acetate and (E)-2-hexenyl acetate, were consistently lower than control levels in nearly all cases. Collectively, these results align with the hypothesis of a enzymatic bottleneck downstream of the aldehyde formation, particularly under high-concentration zinc exposure.

Currently, the literature regarding nanomaterial-mediated GLV emission remains limited. Unlike the linear, concentration-dependent stimulation observed with iron nanoparticles in Brassica napus [71], C. annuum exhibits a more complex, heterogeneous response to ZnO. Our data points specifically to a modulation of the linoleic acid degradation pathway. Our data suggest that high concentrations of ZnO may modulate the linoleic acid degradation pathway. Notably, the observed accumulation of hexanal coupled with a decrease in hexanol and hexyl acetate points toward a potential interference with Alcohol Dehydrogenase (ADH) activity downstream of the lipoxygenase (LOX) pathway [72]. Although zinc is an ADH cofactor, supraphysiological levels or direct nanoparticle interactions might alter enzymatic efficiency [73,74].

It is essential to contextualize these findings within the specific genetic background of Capsicum annuum var. Jalapeño M under controlled conditions. Since volatile biosynthesis, redox regulation, and phyllosphere assembly are traits tightly coupled to host genetics, the transferability of these specific metabolic shifts to other cultivars warrants caution. Consequently, to establish a generalized model of “nano-stress,” future research should adopt a comparative multi-genotype approach, aiming to distinguish between conserved physiological responses and cultivar-dependent variations.

Finally, our findings highlight that ZnONP bioactivity extends beyond simple ionic dissolution. Even at concentrations where macroscopic markers (biomass, germination) remained unaffected, ZnONPs induced specific shifts in chemical signaling (GLV modulation) and enzyme activities that diverged qualitatively from the Bulk treatments. This implies that current risk assessments, often limited to acute endpoints, may underestimate the ecological footprint of nanomaterials. As agricultural nanotechnology matures, developing safety frameworks that incorporate these subtler informational and metabolic dimensions will be crucial for ensuring their sustainable deployment.

4. Conclusions

This research demonstrates that the application of zinc oxide nanoparticles (ZnONPs) in Capsicum annuum elicits divergent physiological and molecular responses compared with their micrometric counterpart (Bulk ZnO). Specifically, ZnONPs exhibited significantly lower foliar accumulation at 500 mg kg⁻¹ than the bulk treatment, yet caused severe inhibition of SOD enzyme activity and a distinct metabolic shift in Green Leaf Volatiles (GLVs), characterized by the accumulation of hexanal and the suppression of downstream alcohols. Finally, zinc exposure—irrespective of particle form—induced a restructuring of the phyllosphere composition, favoring the proliferation of Actinobacteria.

5. Perspectives

The interaction between nanomaterials and plant biological systems represents a multifaceted toxicological challenge. By contrasting ZnONPs against bulk ZnO, this study highlighted divergent physiological and ecological responses that imply distinct, scale-dependent mechanisms of action. From a practical standpoint, while high experimental dosages (e.g., 500 mg kg⁻¹,) serve as valuable models for stress physiology, they are economically and environmentally impractical for broad-scale agriculture. Instead, the sustainable commercialization of nano-agrochemicals must pivot toward precision applications, typically utilizing concentrations below 100 mg kg⁻¹ to maximize nutrient use efficiency without triggering any adverse phenomena. Ultimately, as regulatory frameworks for nanomaterials in edible crops remain under development, future deployment depends on establishing rigorous risk assessments. These must extend beyond efficacy to guarantee food safety and long-term environmental compatibility, ensuring that the transition from laboratory promise to field application is both safe and sustainable.