Co-Application of Arbuscular Mycorrhizal Fungi and Silicon Nanoparticles: A Strategy for Optimizing Volatile Profile, Phenolic Content, and Flower Yield in Rosa damascena Genotypes

Abstract

This study investigated the individual and synergistic impacts of arbuscular mycorrhizal fungi (AMF) inoculation and foliar-applied silicon nanoparticles (SiNPs) on the yield parameters, volatile profile, and phenolic composition of two Rosa damascena genotypes (D231 and C193). Experiments were conducted using a split–split–plot design, involving AMF inoculation (main plot), three SiNPs concentrations (subplot), and two rose genotypes (sub-subplot). The results demonstrated that AMF, SiNPs, and genotype all had significant and interactive effects on flower yield parameters. Foliar application of SiNPs, particularly when combined with AMF inoculation, consistently enhanced flowering parameters, including flower size, number, and weight across both genotypes. High-performance liquid chromatography (HPLC) further confirmed that phenolic acids (vanillic acid and rutin) increased following foliar application of SiNPs and AMF root colonization, particularly in genotype C193. SPME-Arrow analysis revealed that alcohols, ketones, and terpenes were the predominant volatile constituents. Phenethyl alcohol was the most abundant compound, accounting for approximately 84.69% of the total aroma content and contributing significantly to the ‘rose’ aroma. Other major volatiles included 2-undecanone (4.42%), benzyl alcohol (2.97%), and citronellol (1.95%); however, their levels varied depending on treatment and genotype. These findings highlight that the combined application of AMF and SiNPs offers a sustainable approach to enhancing both the quantitative yield and qualitative phytochemical composition (essential oil components and phenolic compounds) of R. damascena, providing a scientific foundation for optimizing its production in organic farming systems.

1. Introduction

Roses are one of the most diverse and economically important ornamental plants, with over 200 species and more than 18,000 varieties documented to date [1]. Among the Rosaceae family, four primary species (Rosa damascena Mill., R. moschata Hemsl., R. centifolia L., and R. gallica L.) are notable producers of valuable rose essential oil. The Damask rose (R. damascena), widely cultivated in countries such as Turkey, Bulgaria, India, and Iran, is often referred to as the “queen of flowers” due to its significant medicinal, aromatic, and ornamental properties [2]. Rose essential oil, often termed “liquid gold,” is extracted from large quantities (approximately 3 to 4 tons per liter of oil) of fresh flowers [3]. The majority of research on rose oil extraction (86.4%) focuses on R. damascena, underscoring its commercial and scientific importance.

Rose essential oil is extensively utilized in the perfumery and cosmetic industries, and it also serves as a natural flavoring agent in the food sector [4]. According to the International Organization for Standardization (ISO 9842:2003) [5], rose oil quality standards specify minimum concentrations of key constituents, including at least 20% citronellol, 15% geraniol, and 8% nonadecane [6].The perfume industry prefers oils with a citronellol/geraniol (C/G) ratio in the range of 1.25 and 1.30, which is considered characteristic of authentic rose oil [6].

The chemical composition of Damask rose essential oil can be affected by various factors, including extraction techniques, genotype, geographical origin, climate conditions, soil properties, harvest time, and post-harvest handling. Recent studies by Katekar et al. (2022) identified 46 and 15 aroma components from fresh oily rose petals using headspace solid-phase microextraction (HS-SPME) and hydro-distillation, respectively [4]. Since its introduction by Pawlyshin et al. [7] in the late 1980s, SPME has become a powerful and widely adopted technique for analyzing volatile compounds in plants [8]. SPME has been successfully applied to volatile extractions from rose species such as R. rugosa, R. roxburghii, and Chinese rose [8], and more recently, R. damascene [9]. Unlike conventional distillation methods, SPME offers several advantages, including the elimination of heat and organic solvents, direct adsorption of floral volatiles onto the fiber, and more accurate profiling of the aroma components [10]. Arbuscular mycorrhizal fungi (AMF), belonging to the phylum Glomeromycota, form symbiotic associations with approximately 90% of terrestrial plants; this relationship has evolved over 460 million years [11]. Inoculation with AMF has been shown to enhance flower yield, macronutrient content, chlorophyll levels, and overall nutrient uptake in Damask rose [12]. Moreover, AMF inoculation positively affects mineral nutrition, promoting plant growth and improving essential oil production in Rosmarinus officinalis [13].

Silicon nanoparticles (SiNPs) have recently gained attention in agriculture as components of fertilizers, biopesticides, and fungicides. SiNPs can be synthesized from various chemical, inorganic, and organic sources [14]. Although SiNPs have shown promising effects on plant growth and secondary metabolite accumulation, concerns regarding their potential environmental risks, such as soil accumulation and effects on non-target organisms, have been raised. Therefore, their application should be considered within the broader framework of sustainable agriculture [15]. Foliar application of silicon has been observed to alter the chemotype distribution of Damask rose essential oil, influencing key compounds such as linalool, citronellol, geraniol, and methyl eugenol [16]. Previous studies have documented improvements in flower yield and metabolite accumulation in Damask rose following individual treatments with SiNP application and AMF inoculation. However, most studies have focused on the phytochemical profiling of essential oils by gas chromatography–mass spectrometry (GC-MS), with limited attention paid to metabolite profiling using advanced methods such as SPME-Arrow. Furthermore, although volatile oils have been extensively studied, phenolic compounds (another class of bioactive metabolites) have received comparatively little investigation.

Several studies have demonstrated that AMF inoculation not only enhances plant growth and stress tolerance but also significantly alters secondary metabolite production, including phenolics and flavonoids, in various plant species under abiotic stress conditions [17,18]. In addition to nutrient uptake, AMF can induce reprogramming of host plant metabolic pathways, resulting in an elevated accumulation of antioxidants, osmoprotectants, and other bioactive compounds under stress [17,19]. Similarly, the application of nano-silicon has been shown to enhance secondary metabolism, including the accumulation of polyphenols and volatile oils, especially under environmental stress in plants such as Mentha pulegium and many crop plants [20,21]. The synergistic use of AMF and nano-silicon has also been reported to enhance both primary and secondary metabolites, including sugars, amino acids, phenolics, and aroma-related volatiles, in crops under abiotic stress [22,23].

The objectives of the present study were to (1) conduct a comprehensive evaluation of the combined effects of AMF inoculation and SiNP foliar application on the metabolite profile of Damask rose to identify the optimal treatment combinations, (2) characterize the phytochemical composition of rose volatiles using SPME-Arrow analysis, and (3) quantify the accumulation of phenolic compounds in rose flowers. The findings aim to support sustainable cultivation practices for Damask rose, particularly in organic farming systems, by enhancing the production of valuable metabolites through environmentally friendly and economically viable approaches such as AMF symbiosis and SiNPs foliar nutrition.

2. Materials and Methods

2.1. Experimental Site and Plant Material

The experiment was conducted at the Agricultural Research Farm of Isfahan University of Technology, Lavark, during 2022–2023. Prior to planting, the soil was amended with 50 tons per hectare of well-decomposed manure fertilizer to enhance soil fertility (Table S1). Soil physical and chemical analyses were performed according to the standard procedures described by Page et al. (1982) [24]. The meteorological data during the experimental period are summarized in Table S2. Two genotypes (D231 from Bardsir province and C193 from Kashan province) of Rosa damascena (Damask rose), propagated via tissue culture, were cultivated at a spacing of 2.5 m × 3 m between plants. The experiment was conducted on five-year-old R. damascena genotypes, with 120 plants per genotype. The plant material originated from elite Iranian genotypes with stable flower yield and essential oil content, and was propagated via tissue culture by a commercial biotechnology company (Urum Zist Tak, Urmia, Iran). As the propagation process was carried out externally using a commercial protocol, detailed tissue culture methods are not described in this paper. Flowering of R. damascena typically occurs between 15 °C and 25 °C; under the study conditions, flowering lasted for 30 days, from April 21 to May 20, with peak harvest on April 27, consistent with previous yield observations from the site. Therefore, plant material was collected in April 2023.

2.2. Mycorrhizal Inoculation and Silicon Nanoparticle Treatments

Half of the rose shrubs were inoculated with 50 g of a commercial arbuscular mycorrhizal fungi (AMF) inoculum comprising Funneliformis mosseae, Rhizophagus intraradices, and Claroideoglomus etunicatum per plant [25]. At planting, 60 plants per genotype were inoculated with a commercial AMF inoculum (MycoRoot^®; Zist Fanavar Pishtaz Varian, Karaj, Iran). For each plant, 50 g of the inoculum was placed directly in the root zone at the time of planting. To ensure sufficient establishment of AMF colonization, foliar application of silicon nanoparticles (SiNPs) was performed in the following growing season (2023). SiNP treatment was performed according to the designed protocol. To reduce environmental variation, flower sampling for all treatments was conducted during the same morning hours. To confirm the effectiveness of AMF inoculation, root colonization was evaluated following harvest. Root systems were carefully washed with deionized water to remove adhering soil particles and then cut into ~1 cm segments. Root fragments were cleared in 10% (w/v) potassium hydroxide (KOH) solution for 2 h at room temperature and subsequently rinsed five to six times with deionized water to eliminate foam residues. The samples were then bleached in alkaline hydrogen peroxide (H₂O₂) for 40 min, rinsed again with deionized water, and acidified in 10% hydrochloric acid (HCl) for 30 min to improve tissue transparency. Thereafter, the roots were stained with 0.05% trypan blue in lactophenol solution [12]. For each treatment, approximately 100 root segments (each about 1 cm long) were randomly selected from three plants per treatment for microscopic observation and examined under a stereomicroscope at 400× magnification to estimate the colonization frequency. Colonization percentage was calculated according to standard procedures. In this study, colonization percentage refers to the frequency of AMF colonization (F%), calculated as the proportion of root segments containing fungal structures (hyphae, arbuscules, or vesicles) following the standard procedure [26] described by Hashemi et al. (2024). As anticipated, AMF-inoculated plants exhibited clear intraradical structures (hyphae, arbuscules, and vesicles), whereas control plants showed very low levels of background colonization likely due to indigenous non-sterilized soil fungi (

Table 1. Leaf silicon concentration and AMF symbiosis percentage (mean ± standard deviation) in R. damascena genotypes under different SiNP treatments.

AMF	Gen	SiNP	Si (mg/gDW)	AMF Root Colonization (%)	AMF	Genotype	SiNPs (mg/L)	Si (mg/gDW)	AMF Root Colonization (%)
0	D231	0	0.0026 ± 0.0004	6.00 ± 1.05	1	D231	0	0.0037 ± 0.0002	42.44 ± 1.41
		2	0.0033 ± 0.0002	7.00 ± 1.07			2	0.0044 ± 0.0003	45.21 ± 2.48
		4	0.0046 ± 0.0004	9.67 ± 0.58			4	0.0078 ± 0.0001	48.53 ± 1.47
	C193	0	0.0014 ± 0.0008	6.67 ± 1.15		C193	0	0.0019 ± 0.0006	54.89 ± 0.16
		2	0.0028 ± 0.0005	9.67 ± 0.61			2	0.0030 ± 0.0008	61.41 ± 2.19
		4	0.0049 ± 0.0002	11.33 ± 0.59			4	0.0052 ± 0.0002	63.72 ± 1.08
LSD (p ≤ 0.05)			0.0009	1.63	LSD (p ≤ 0.05)			0.0007	2.91

AMF: arbuscular mycorrhizal fungi; Gen: genotype; SiNP: silicon nanoparticles (mg/L); Si: silicon concentration in leaves; DW: dry weight; AMF root colonization (%) represents the percentage of root segments colonized by AMF.

).

Foliar applications of SiNPs were performed using a chelated nanosilicon fertilizer at concentrations of 0 (control), 2, and 4 mg L⁻¹. The SiNPs had an average particle size of 23–27 nm and purity of 90% (Figure 1). The nanosilicon fertilizer was supplied by Sodour Ahrar Shargh (Khazra^® Company, Tehran, Iran), a patented product registered under USPTO patent number US8288587 (2012) [26]. Fully expanded leaves were selected and all bushes were uniformly pruned to a consistent height. Foliar applications were conducted once weekly for three consecutive weeks, with the solution sprayed until approximately 80% of the leaf surface was covered. Distilled water was used as the control treatment. Throughout the growing season, all plants were drip irrigated and managed under uniform agronomic practices to ensure consistent growth conditions.

To quantify silicon accumulation in the leaves, 0.1 g of dried and finely ground leaf tissue from each treatment was digested with 10 mL of 2 N HCl using a closed-vessel digestion system. The digested samples were then filtered and diluted to a final volume of 50 mL with deionized water. Silicon concentrations were determined using Inductively Coupled Plasma Optical Emission Spectrometry (ICP-OES). Calibration curves were constructed using silicon standard solutions, and quality control was ensured by including reagent blanks and certified reference materials in each analytical batch. The measured silicon concentrations (initially obtained in parts per billion, ppb) were subsequently converted to milligrams per gram (mg g⁻¹) of dry weight (Table 1). The application of SiNPs led to a dose-dependent increase in leaf silicon concentration, which was accompanied by enhanced colonization of AMF in both genotypes. Moreover, AMF inoculation itself promoted higher AMF symbiosis rates and further improved leaf silicon accumulation, indicating a synergistic interaction between SiNP application and AMF association (Table 1).

2.3. Experimental Design

A split-plot experimental design was employed, with AMF inoculation as the main plot factor, nanosilicon concentrations (0, 2, and 4 mg/L) as subplot factors, and genotypes randomized within sub-subplots. For all biochemical and physiological measurements, three biological replicates were analyzed per treatment. Each replicate consisted of a composite sample prepared by pooling randomly collected plant materials (leaves, petals, or roots, depending on the analysis) from several plants within the same replicate to ensure representative sampling and minimize within-treatment variability.

2.4. Determination of Yield-Related Flower Traits

Open flowers were hand-harvested each morning from mid-June to the end of May, during the flowering season. The following parameters were recorded for each plant: pedicle length, pedicle diameter, florescence length, petal length, petal diameter, petal number per flower, and flower number per plant. Additionally, fresh and dry flower weights were measured, and flower diameter was recorded using calipers as flower size. All morphological traits were measured in triplicate (biological replicates), and the average values were reported.

2.5. Characterization of Volatile Compounds by HS-SPME-GC-MS

For volatile compound analysis, dried rose petals (0.5 g) were accurately weighed and placed into a 10 mL headspace glass vial, which was immediately sealed with a silicone/PTFE septum and aluminum cap to prevent compound loss. Samples were pre-incubated at 45 °C for 30 min to allow equilibration of the volatile compounds in the headspace. A 50/30 µm SPME fiber, suitable for a wide range of volatiles, was exposed to the headspace for 30 min to adsorb volatile organic compounds. After extraction, the fiber was immediately desorbed in the injection port of a gas chromatograph coupled with a mass spectrometer operated in electron impact (EI) ionization mode at 70 eV. Separation was performed using an HP-5MS capillary column (30 m × 0.25 mm i.d., 0.25 µm film thickness), with helium as the carrier gas at a constant flow rate of 1 mL/min. The GC oven temperature was programmed as follows: initially held at 40 °C for 2 min, increased to 200 °C at 5 °C/min, and finally to 250 °C at 10 °C/min and held for 5 min. The injector was maintained at 250 °C in splitless mode, and the desorption time was 5 min. Mass spectra were recorded in the range of m/z 40–400. Compounds were identified by comparing their mass spectra with those in the NIST 23 and Wiley libraries and by calculating retention indices (RI) and comparing them with literature values. Relative quantification was performed using peak area normalization, and compound abundance was expressed as the percentage of total ion current (TIC). The analysis was performed in triplicate for each sample. The procedure followed the protocol described by Behnamnia et al. (2024), with minor modifications [9].

2.6. Quantification of Phenolic Compounds by HPLC

Phenolic compounds were analyzed using an Agilent HPLC system equipped with a mass spectrometer and diode array detector (DAD). Chromatographic separation was performed on a Luna C18 column with a mobile phase consisting of solvent A (0.01% formic acid in water) and solvent B (0.01% formic acid in acetonitrile) at a flow rate of 1 mL/min. Dried petals were ground and passed through a 20-mesh sieve prior to extraction. Approximately 0.625 mg of the sample was extracted in 12.5 mL of methanol-water solution (70:30 v/v) by shaking for 48 h. Extracts were filtered, and 20 µL aliquots were injected into the HPLC system (each treatment was analyzed in three replicates). The DAD monitored absorbance at multiple wavelengths (266, 267, 270, 315, 328, 330, and 370 nm) to assist in the identification and quantification of phenolic compounds.

2.7. Determination of Total Phenolic Content (TPC)

Total phenolic content was determined using the Folin–Ciocalteu colorimetric assay on methanolic extracts of dried petals. Extracts were prepared by shaking samples at 150 rpm and 25 °C for 24 h and subsequently filtered through Whatman No. 4 paper. A 100 μL aliquot of the filtered extract was diluted with 1800 μL methanol and then mixed with 0.5 mL of 10% Folin–Ciocalteu reagent and 0.4 mL of 7.5% sodium carbonate (w/v). The mixture was incubated at room temperature for 30 min, and the absorbance was measured spectrophotometrically at 765 nm. TPC was expressed as mg of tannic acid equivalents per gram of dry weight (mg TAE/g DW) as described by [27]. TPC was measured in three biological replicates, and the average values were reported.

2.8. Determination of Total Flavonoid Content (TFC)

Total flavonoid content was assessed using the aluminum chloride colorimetric method with quercetin as a calibration standard. A 100 µL aliquot of the extract was mixed with 900 µL distilled water followed by 60 µL of NaNO2 solution and incubated in darkness at room temperature for 5 min. Subsequently, 120 µL of 10% aluminum chloride solution was added, followed by 400 µL of 1 M NaOH, and 400 µL of distilled water. The mixture was vortexed, and absorbance was measured at 510 nm using a spectrophotometer. Flavonoid content was calculated using a quercetin standard curve and expressed as mg quercetin equivalents per gram of dry weight (mg QE/g DW) [27]. TFC was measured in three biological replicates, and the average values were reported.

2.9. Total Antioxidant Activity (TAA)

The antioxidant capacity was evaluated using the DPPH (2,2-diphenyl-1-picrylhydrazyl) radical scavenging method. The diluted ethanolic extract (500 µL) was mixed with 80 µM DPPH solution (0.5 mL) in methanol. The reaction mixture was then incubated in the dark at room temperature for 30 min. Absorbance was measured at 517 nm. A control sample was prepared identically without the extract. TAA was measured in three biological replicates, and the average values were reported. The percentage of radical scavenging activity was calculated as follows [27]:

$$TAA(\%)={\displaystyle \frac{(A_{\mathrm{c}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}}-A_{\mathrm{s}\mathrm{a}\mathrm{m}\mathrm{p}\mathrm{l}\mathrm{e}})}{A_{\mathrm{c}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}}}}\times 100$$

(1)

2.10. Essential Oil Extraction (EO)

Freshly harvested flowers (1 kg per genotype) were immediately subjected to hydro-distillation using a 5 L Clevenger-type apparatus for six hours to extract the essential oil. The resulting oil was dried over anhydrous sodium sulfate (Na₂SO₄) to remove residual water. Essential oil content was expressed as the percentage of fresh flower weight. EO was measured in three biological replicates, and the average values were reported.

2.11. Statistical Analysis

Data were analyzed using SAS software version 9.4. Analysis of variance (ANOVA) was performed, and mean comparisons were conducted using LSD test at a significance level of p ≤ 0.05.

3. Results

3.1. Morphological Traits

The analysis of variance (Table S3) revealed that both SiNPs and AMF had significant effects on several morphological traits of R. damascena. These treatments significantly influenced most of the evaluated traits (Table S3). A significant genotypic effect was also observed for pedicel diameter, petal length, petal number, flower number, and flower fresh and dry weights, indicating a clear morphological variation between the two genotypes.

The interaction effects further highlighted the complexity of trait regulation. The AMF × SiNP interaction was significant for several flower-related traits, suggesting a synergistic effect of these two factors on floral development. Similarly, the AMF × genotype interaction significantly affected pedicel length and diameter, inflorescence length, flower number, and flower biomass (fresh and dry weights). The SiNP × genotype interaction was also important, significantly influencing petal number, flower size, flower number, and flower biomass. Overall, AMF application resulted in marked increases in key floral traits (pedicle length by 35%, pedicle diameter by 33%, inflorescence length by 21%, petal length by 16%, and petal number by 25%) compared with the control. These findings highlight the positive role of AMF in enhancing the structural and reproductive attributes of Damask rose. The application of SiNPs also led to notable improvements in morphological traits (Table 2). Higher SiNP concentrations (4 mg/L) significantly increased pedicle length (16%), pedicle diameter (25%), inflorescence length (23%), petal length (10%), petal diameter (9%), petal number (33%), and flower number (27%) compared to the untreated controls. Interestingly, the lower SiNPs concentration (2 mg/L) showed a more pronounced effect on flower size (5%), fresh weight (37%), and dry weight (30%).

Among the genotypes, C193 generally outperformed D231 across most traits. It exhibited higher petal length (5%), petal number (5%), flower number (7%), and both flower fresh and dry weight (15% each), reflecting its superior morphological potential. The combination of AMF and 4 mg/L SiNPs produced the longest inflorescence (49% higher than the control), while 2 mg/L SiNPs in the presence of AMF significantly improved flower size (5% increase) and flower fresh weight (59% increase). Interestingly, the highest flower number was recorded under 4 mg/L SiNPs without AMF, indicating a 62% increase over the control, suggesting that SiNPs alone can effectively enhance reproductive output in the absence of AMF. Maximum flower dry weight was achieved under two conditions: (i) 4 mg/L SiNPs without AMF, and (ii) 2 mg/L SiNPs with AMF, both resulting in approximately a 61% increase compared to the control.

In the presence of AMF, C193 had significantly higher flower number, fresh, and dry weight, whereas in the absence of AMF, the same genotype showed superior inflorescence length, pedicle length, and pedicle diameter (Table 2). Interestingly, D231 genotype produced the highest number of petals under 4 mg/L SiNPs, while C193 exhibited the highest values for flower size, flower number, flower fresh weight, and flower dry weight under 2 mg/L SiNPs. The three-way interaction (AMF × SiNPs × Gen) revealed that under 4 mg/L SiNPs without AMF, D231 showed enhanced fresh flower weight, while the highest fresh flower weight was observed in C193 treated with 2 mg/L SiNPs in the presence of AMF. This treatment also yielded the highest flower number. Furthermore, the highest pedicle diameter was observed in both genotypes when treated with 4 mg/L SiNPs in combination with AMF (Table 2). These results underscore the complex but beneficial interplay between AMF, SiNPs, and genetic background in modulating floral yield and morphology in R. damascena. The findings suggest that optimizing both biotic (AMF) and abiotic (SiNPs) inputs, in relation to genotype, can significantly enhance the horticultural quality of Damask rose flowers.

The correlation analysis (Table S4) revealed significant associations among several morphological, biochemical, and volatile traits of R. damascena genotypes under AMF and SiNPs treatments. Pedicle length and diameter, inflorescence length, petal diameter, flower number, and flower weight exhibited strong positive correlations, suggesting that these traits are developmentally coordinated and may collectively contribute to enhanced flower yield.

With respect to phenolic compounds, syringic acid and chlorogenic acid displayed significant positive correlations with petal number, petal diameter, petal length, inflorescence length, and pedicle diameter. This pattern indicates a bidirectional relationship: on one hand, these phenolic acids appear to be influenced by floral morphological traits; on the other hand, they may also contribute to regulating floral development and morphogenesis, consistent with their reported roles in plant defense and growth regulation. In contrast, p-coumaric acid exhibited significant negative correlations with pedicle diameter and inflorescence length, suggesting that higher accumulation of this compound may be associated with restrictions in cell wall expansion processes, thereby constraining certain floral structural attributes.

Table 2. The Mean comparison (±standard deviation) of morphological traits of R. damascena under AMF and SiNP treatments.

AMF	SiNP	Gen	pedicle length (cm)	pedicle diameter (mm)	florescence length (cm)	petal length (mm)	petal diameter (mm)
0	0	D231	4.06 ± 0.29 g	9.17 ± 0.23 g	106.0 ± 0.90 fg	22.69 ± 1.78 g	19.59 ± 0.68 e
0	0	C193	5.05 ± 0.80 ef	9.66 ± 0.94 fg	102.2 ± 1.58 g	25.89 ± 0.96 ef	21.41 ± 1.41 cde
0	2	D231	4.11 ± 0.08 g	11.20 ± 0.21 de	110.8 ± 5.83 efg	23.89 ± 1.01 fg	20.32 ± 0.63 de
0	2	C193	5.02 ± 0.41 ef	9.71 ± 0.35 fg	111.1 ± 3.50 efg	26.06 ± 0.83 def	21.46 ± 1.52 cde
0	4	D231	4.64 ± 1.00 fg	13.33 ± 0.59 c	118.9 ± 5.39 de	26.26 ± 0.81 de	22.48 ± 1.07 a–d
0	4	C193	5.85 ± 0.06 cd	10.46 ± 0.89 ef	124.4 ± 1.54 cd	26.79 ± 2.70 cde	21.94 ± 1.02 b–d
1	0	D231	6.50 ± 0.17 bc	14.48 ± 0.49 b	124.5 ± 1.88 cd	27.00 ± 0.78 cde	22.70 ± 1.43 a–d
1	0	C193	5.63 ± 0.26 de	11.25 ± 0.38 de	116.8 ± 1.28 def	28.29 ± 1.04 bcd	22.46 ± 2.25 a–d
1	2	D231	6.57 ± 0.47 abc	14.73 ± 0.50 b	131.5 ± 0.79 c	28.67 ± 0.61 abc	23.15 ± 1.94 abc
1	2	C193	6.09 ± 0.14 bcd	11.61 ± 0.74 d	132.3 ± 3.45 c	30.22 ± 0.65 ab	23.06 ± 2.50 abc
1	4	D231	7.27 ± 0.28 a	16.24 ± 0.34 a	165.1 ± 19.47 a	30.81 ± 1.56 a	24.82 ± 1.87 a
1	4	C193	6.80 ± 0.19 ab	15.87 ± 0.54 a	144.5 ± 2.21 b	30.19 ± 1.39 ab	24.40 ± 1.13 ab
AMF	SiNP	Gen	petal numbers	flower size (cm)	flower number	flower fresh weight (g)	flower dry weight (g)
0	0	D231	34.67 ± 1.53 f	5.81 ± 0.12 b	27.18 ± 0.60 g	43.71 ± 2.55 h	9.16 ± 0.68 f
0	0	C193	38.33 ± 1.53 ef	5.92 ± 0.02 b	33.76 ± 2.64 f	59.26 ± 1.83 ef	11.49 ± 1.60 e
0	2	D231	36.00 ± 1.00 ef	5.79 ± 0.03 b	35.12 ± 2.23 ef	59.13 ± 1.71 ef	11.89 ± 0.92 de
0	2	C193	42.33 ± 2.52 cde	6.00 ± 0.14 ab	40.27 ± 0.48 d	73.07 ± 1.53 c	14.81 ± 0.56 bc
0	4	D231	47.33 ± 1.15 bcd	5.81 ± 0.11 b	54.09 ± 2.17 a	83.40 ± 1.23 b	17.90 ± 0.43 a
0	4	C193	46.67 ± 3.21 bcd	5.94 ± 0.25 b	44.45 ± 0.60 c	72.20 ± 1.43 dc	15.36 ± 1.00 b
1	0	D231	41.00 ± 3.61 efd	5.92 ± 0.42 b	35.36 ± 2.60 ef	51.82 ± 1.13 g	11.47 ± 0.82 e
1	0	C193	46.00 ± 7.94 bcd	5.40 ± 0.17 c	38.06 ± 1.00 de	60.79 ± 1.29 e	13.57 ± 0.27 cd
1	2	D231	48.00 ± 6.24 bc	6.05 ± 0.17 ab	40.48 ± 1.01 d	69.67 ± 1.03 d	13.25 ± 0.98 cd
1	2	C193	51.00 ± 5.29 b	6.29 ± 0.30 a	50.27 ± 1.97 b	94.48 ± 1.32 a	19.54 ± 0.91 a
1	4	D231	62.00 ± 3.61 a	5.91 ± 0.05 b	35.12 ± 3.68 ef	56.94 ± 2.26 f	11.94 ± 1.35 de
1	4	C193	58.00 ± 4.58 a	5.98 ± 0.17 ab	36.97 ± 0.61 ef	59.31 ± 1.13 ef	12.26 ± 1.54 de

In each column, means sharing the same letter do not differ significantly, as determined by the LSD test at p ≤ 0.05.

Table 2. The Mean comparison (±standard deviation) of morphological traits of R. damascena under AMF and SiNP treatments.

AMF	SiNP	Gen	pedicle length (cm)	pedicle diameter (mm)	florescence length (cm)	petal length (mm)	petal diameter (mm)
0	0	D231	4.06 ± 0.29 g	9.17 ± 0.23 g	106.0 ± 0.90 fg	22.69 ± 1.78 g	19.59 ± 0.68 e
0	0	C193	5.05 ± 0.80 ef	9.66 ± 0.94 fg	102.2 ± 1.58 g	25.89 ± 0.96 ef	21.41 ± 1.41 cde
0	2	D231	4.11 ± 0.08 g	11.20 ± 0.21 de	110.8 ± 5.83 efg	23.89 ± 1.01 fg	20.32 ± 0.63 de
0	2	C193	5.02 ± 0.41 ef	9.71 ± 0.35 fg	111.1 ± 3.50 efg	26.06 ± 0.83 def	21.46 ± 1.52 cde
0	4	D231	4.64 ± 1.00 fg	13.33 ± 0.59 c	118.9 ± 5.39 de	26.26 ± 0.81 de	22.48 ± 1.07 a–d
0	4	C193	5.85 ± 0.06 cd	10.46 ± 0.89 ef	124.4 ± 1.54 cd	26.79 ± 2.70 cde	21.94 ± 1.02 b–d
1	0	D231	6.50 ± 0.17 bc	14.48 ± 0.49 b	124.5 ± 1.88 cd	27.00 ± 0.78 cde	22.70 ± 1.43 a–d
1	0	C193	5.63 ± 0.26 de	11.25 ± 0.38 de	116.8 ± 1.28 def	28.29 ± 1.04 bcd	22.46 ± 2.25 a–d
1	2	D231	6.57 ± 0.47 abc	14.73 ± 0.50 b	131.5 ± 0.79 c	28.67 ± 0.61 abc	23.15 ± 1.94 abc
1	2	C193	6.09 ± 0.14 bcd	11.61 ± 0.74 d	132.3 ± 3.45 c	30.22 ± 0.65 ab	23.06 ± 2.50 abc
1	4	D231	7.27 ± 0.28 a	16.24 ± 0.34 a	165.1 ± 19.47 a	30.81 ± 1.56 a	24.82 ± 1.87 a
1	4	C193	6.80 ± 0.19 ab	15.87 ± 0.54 a	144.5 ± 2.21 b	30.19 ± 1.39 ab	24.40 ± 1.13 ab
AMF	SiNP	Gen	petal numbers	flower size (cm)	flower number	flower fresh weight (g)	flower dry weight (g)
0	0	D231	34.67 ± 1.53 f	5.81 ± 0.12 b	27.18 ± 0.60 g	43.71 ± 2.55 h	9.16 ± 0.68 f
0	0	C193	38.33 ± 1.53 ef	5.92 ± 0.02 b	33.76 ± 2.64 f	59.26 ± 1.83 ef	11.49 ± 1.60 e
0	2	D231	36.00 ± 1.00 ef	5.79 ± 0.03 b	35.12 ± 2.23 ef	59.13 ± 1.71 ef	11.89 ± 0.92 de
0	2	C193	42.33 ± 2.52 cde	6.00 ± 0.14 ab	40.27 ± 0.48 d	73.07 ± 1.53 c	14.81 ± 0.56 bc
0	4	D231	47.33 ± 1.15 bcd	5.81 ± 0.11 b	54.09 ± 2.17 a	83.40 ± 1.23 b	17.90 ± 0.43 a
0	4	C193	46.67 ± 3.21 bcd	5.94 ± 0.25 b	44.45 ± 0.60 c	72.20 ± 1.43 dc	15.36 ± 1.00 b
1	0	D231	41.00 ± 3.61 efd	5.92 ± 0.42 b	35.36 ± 2.60 ef	51.82 ± 1.13 g	11.47 ± 0.82 e
1	0	C193	46.00 ± 7.94 bcd	5.40 ± 0.17 c	38.06 ± 1.00 de	60.79 ± 1.29 e	13.57 ± 0.27 cd
1	2	D231	48.00 ± 6.24 bc	6.05 ± 0.17 ab	40.48 ± 1.01 d	69.67 ± 1.03 d	13.25 ± 0.98 cd
1	2	C193	51.00 ± 5.29 b	6.29 ± 0.30 a	50.27 ± 1.97 b	94.48 ± 1.32 a	19.54 ± 0.91 a
1	4	D231	62.00 ± 3.61 a	5.91 ± 0.05 b	35.12 ± 3.68 ef	56.94 ± 2.26 f	11.94 ± 1.35 de
1	4	C193	58.00 ± 4.58 a	5.98 ± 0.17 ab	36.97 ± 0.61 ef	59.31 ± 1.13 ef	12.26 ± 1.54 de

In each column, means sharing the same letter do not differ significantly, as determined by the LSD test at p ≤ 0.05.

In terms of volatile compounds, benzaldehyde and phenethyl formate showed significant negative correlations with pedicle diameter, implying that morphological characteristics can influence, or be influenced by, the biosynthesis and emission of specific aromatic volatiles. Conversely, pedicle length, petal length, and petal diameter exhibited strong positive correlations with α-geraniol, one of the key fragrance constituents of R. damascena. This relationship suggests that larger floral structures may provide greater metabolic flux toward monoterpene biosynthesis, thereby enhancing both fragrance intensity and commercial value.

Taken together, these correlation patterns highlight the intricate two-way interdependence between morphological attributes, phenolic accumulation, and volatile composition. The findings further emphasize that improvements in floral architecture—potentially stimulated by AMF and SiNPs co-application—may indirectly optimize the phenolic and volatile profiles, ultimately enhancing both flower yield and essential oil quality in R. damascena.

Generally, these results demonstrate the multifactorial nature of morphological regulation in R. damascena. Both AMF and SiNPs substantially improved floral yield and quality, and their effectiveness was strongly influenced by genotype. The findings suggest that careful optimization of biotic (AMF) and abiotic (SiNPs) inputs, tailored to genotype, can significantly enhance the horticultural performance of Damask rose.

3.2. Phenolic Components

The analysis of variance (Table S5) revealed that SiNPs, AMF, genotype, and their interactions had significant effects on most of the phenolic compounds studied. These findings underscore the complex regulatory influence of both genetic and environmental factors on phenolic biosynthesis in R. damascena. Among the phenolic constituents, the highest levels were generally recorded for rutin (42% increase compared to the control), followed by vanillic acid, syringic acid, gallic acid, and rosmarinic acid (

Table 3. Mean comparison (±standard deviation) of phenolic compounds (µg g⁻¹ DW) of R. damascena under AMF and SiNP treatments.

AMF	SiNP	Gen	vanillic acid	apigenin	gallic acid	caffeic acid	4-hydroxy benzoic acid	syringic acid	p-coumaric acid
0	0	D231	11.7 ± 1.29 h	ND	4.65 ± 0.26 i	3.85 ± 0.58 h	ND	13.4 ± 0.55 f	3.25 ± 0.05 bc
0	0	C193	18.5 ± 1.19 de	1.65 ± 0.06 cd	8.76 ± 1.03 e	6.98 ± 0.31 bcd	ND	ND	3.83 ± 0.31 a
0	2	D231	15.7 ± 0.46 fg	1.07 ± 0.04 fg	5.40 ± 0.43 hi	5.80 ± 0.25 fg	ND	ND	3.02 ± 0.09 def
0	2	C193	20.5 ± 1.20 c	2.23 ± 0.09 b	16.6 ± 0.73 b	7.57 ± 0.47 b	ND	ND	3.72 ± 0.08 a
0	4	D231	16.2 ± 0.53 fg	1.31 ± 0.06 e	6.83 ± 0.52 fg	6.48 ± 0.32 de	ND	ND	2.98 ± 0.08 def
0	4	C193	25.9 ± 1.49 a	2.42 ± 0.09 a	24.2 ± 0.80 a	9.81 ± 0.20 a	15.4 ± 1.02 a	ND	3.05 ± 0.17 def
1	0	D231	11.3 ± 1.14 h	0.64 ± 0.10 h	5.15 ± 0.67 hi	4.04 ± 0.09 h	ND	14.3 ± 0.43 f	3.09 ± 0.07 cd
1	0	C193	17.1 ± 0.38 ef	1.54 ± 0.16 d	10.0 ± 0.39 d	5.63 ± 0.31 fg	ND	27.2 ± 0.94 c	3.41 ± 0.03 b
1	2	D231	14.6 ± 0.60 g	0.93 ± 0.07 g	5.77 ± 0.60 gh	5.24 ± 0.39 g	ND	23.3 ± 0.74 e	2.88 ± 0.02 ef
1	2	C193	19.3 ± 0.48 cd	1.74 ± 0.15 c	13.2 ± 0.50 c	6.77 ± 0.59 cd	ND	34.1 ± 1.68 b	2.98 ± 0.05 def
1	4	D231	15.5 ± 0.82 fg	1.21 ± 0.05 ef	7.63 ± 0.73 f	5.91 ± 0.17 ef	ND	25.5 ± 1.28 d	2.85 ± 0.002 f
1	4	C193	22.8 ± 1.63 b	2.15 ± 0.08 b	16.1 ± 0.47 b	7.23 ± 0.38 bc	11.6 ± 0.59 b	43.0 ± 0.54 a	2.87 ± 0.01 ef
AMF	SiNP	Gen	rutin		ferulic acid	chlorogenic acid	elagic acid	rosmarinic acid	quercetin
0	0	D231	18.7 ± 0.44 i		13.3 ± 1.66 b	ND	ND	7.23 ± 0.71 d	ND
0	0	C193	61.7 ± 0.42 d		ND	ND	3.87 ± 0.20 d	10.2 ± 0.62 c	1.22 ± 0.22 c
0	2	D231	40.5 ± 0.58 g		2.64 ± 0.69 d	ND	ND	7.34 ± 0.26 d	ND
0	2	C193	103 ± 2.56 b		ND	ND	5.24 ± 0.33 b	12.3 ± 1.48 b	1.35 ± 0.03 b
0	4	D231	52.9 ± 0.33 e		0.10 ± 0.03 e	ND	ND	9.98 ± 0.28 c	ND
0	4	C193	110 ± 0.58 a		ND	ND	6.78 ± 0.32 a	17.3 ± 0.57 a	1.68 ± 0.10 a
1	0	D231	19.3 ± 1.26 i		22.3 ± 2.98 a	0.80 ± 0.05 de	ND	4.00 ± 0.26 e	ND
1	0	C193	54.0 ± 1.56 e		ND	0.73 ± 0.06 e	2.73 ± 0.35 e	7.00 ± 0.54 d	1.17 ± 0.02 c
1	2	D231	30.2 ± 0.45 h		8.75 ± 2.15 c	1.00 ± 0.10 c	ND	6.40 ± 1.03 d	ND
1	2	C193	50.3 ± 0.15 f		ND	0.92 ± 0.09 cd	4.60 ± 0.34 c	10.1 ± 0.68 c	1.21 ± 0.03 c
1	4	D231	85.4 ± 0.83 c		ND	1.43 ± 0.16 a	ND	9.46 ± 0.54 c	ND
1	4	C193	86.5 ± 1.70 c		ND	1.15 ± 0.14 b	5.16 ± 0.22 b	13.1 ± 0.45 b	1.38 ± 0.08 b

ND: not detected. In each column, means sharing the same letter do not differ significantly, as determined by the LSD test at p ≤ 0.05.

). The presence and accumulation of some compounds were condition-dependent, indicating differential metabolic responses under specific treatment combinations.

Apigenin was absent under control conditions, indicating that its biosynthesis is strongly inducible and dependent on specific environmental cues. This pattern suggests that apigenin is not a constitutive metabolite in R. damascena but rather requires external elicitors for activation. Similarly, 4-hydroxybenzoic acid was only detected in genotype C193 treated with 4 mg/L SiNPs, with or without AMF, highlighting a genotype-specific metabolic pathway triggered under high SiNP levels. Syringic acid was observed exclusively in the presence of AMF across both genotypes, except for one case in C193 under 4 mg/L SiNPs, indicating that AMF is central in modulating its biosynthetic pathway. Interestingly, ferulic acid was detected only in D231 but was suppressed when treated with 4 mg/L SiNPs, suggesting that SiNPs may interfere with its accumulation (Table 3). In contrast, chlorogenic acid, ellagic acid, and quercetin were detected only in C193, regardless of the treatment, confirming the dominant role of genotype in shaping phenolic biosynthesis.

These findings collectively demonstrate that phenolic compound accumulation is primarily governed by genetic background, while environmental factors such as SiNPs and AMF act as modulators of metabolic expression. The three-way interaction among AMF, SiNPs, and genotype was significant for all phenolic traits, except vanillic acid (Table S5). For this compound, significant two-way interactions (AMF × SiNPs and SiNPs × Gen) revealed that the maximum accumulation occurred under 4 mg/L SiNPs without AMF, with C193 showing the highest content. This underscores the stronger regulatory role of SiNPs compared to AMF for vanillic acid while also reflecting the decisive impact of intrinsic genetic potential.

Overall, genotype C193 consistently accumulated higher levels of all phenolic compounds than D231, confirming its superior biochemical capacity. Notably, the highest apigenin levels in C193 were obtained under two conditions: 2 mg/L SiNPs without AMF (35% increase compared to the control) and 4 mg/L SiNPs with AMF (30% increase). This indicates that while AMF has a supplementary effect, SiNPs serve as the primary driver of apigenin biosynthesis. In D231, apigenin was absent under control conditions but appeared in considerable amounts under 4 mg/L SiNPs without AMF, further emphasizing the strong inducibility of this metabolite by SiNPs, even in a less responsive genotype.

Caffeic acid showed a clear genotype-dependent response. While C193 produced the highest absolute levels under 4 mg/L SiNPs without AMF, genotype D231 exhibited a stronger relative increase (68% vs. 41% in C193) compared to its control, highlighting its inducibility despite lower baseline levels.

Syringic acid was almost entirely dependent on AMF, with maximum accumulation under 4 mg/L SiNPs in both genotypes (34 µg/g DW in D231 and 43 µg/g DW in C193), pointing to a strong synergistic effect of AMF and SiNPs (Table 3).

In contrast, p-coumaric acid consistently decreased, particularly in C193 (25% reduction) compared to D231, suggesting a metabolic flux reallocation toward other phenylpropanoid derivatives under combined AMF and SiNPs treatment. Among all compounds, rutin was the most abundant, reaching its highest level in C193 under 4 mg/L SiNPs without AMF (79% increase). Interestingly, although D231 produced lower absolute levels, it showed a remarkable 357% relative increase compared to its control, indicating strong inducibility (Table 3). Rosmarinic acid followed a similar trend, with maximum accumulation in C193 under 4 mg/L SiNPs (69% increase). Collectively, these results reveal that C193 responds more efficiently to SiNPs alone, whereas D231 benefits more from the combined application of SiNPs and AMF. This divergence reflects distinct genotype-specific regulatory networks shaping secondary metabolism in R. damascene (Table 3).

Comparable trends have been reported in other medicinal plants: silicon enhanced phenolic acids in Ocimum basilicum under salinity stress [28]; AMF plus silicon boosted phenolic compounds in Glycyrrhiza glabra [29], and genotypic differences in metabolite accumulation were observed in R. damascena under stress [30]. In summary, SiNPs act as a potent elicitor in C193, while AMF has a more supportive role in D231. These findings highlight the metabolic plasticity of Damask rose and suggest that tailoring biostimulant strategies to genotype can optimize phenolic accumulation for medicinal and aromatic applications.

The correlation analysis (Table S4) revealed complex interactions among phenolic acids, flavonoids, and volatile compounds in Damask rose under AMF and SiNPs treatments. Vanillic acid and apigenin showed broad positive correlations with several phenolics (gallic, caffeic, 4-hydroxybenzoic, rutin, ellagic, rosmarinic, quercetin) and volatiles (para-cymene, rose oxide), but negative associations with ferulic acid and phenethyl alcohol, indicating their central role in linking phenylpropanoid metabolism with volatile biosynthesis. Gallic and caffeic acids were also extensively correlated with other phenolics and key volatiles such as linalool and β-geranial, while showing negative correlations with ferulic acid, reflecting metabolic competition within the phenylpropanoid pathway. Similarly, rutin was positively associated with ellagic, rosmarinic, and quercetin, but negatively with ferulic acid. Ferulic acid itself exhibited inverse correlations with rosmarinic acid, reinforcing its antagonistic role in phenolic–flavonoid flux. Certain metabolites acted as bridges between phenolics and volatiles. For example, chlorogenic acid correlated positively with α-geraniol, total phenolic and flavonoid contents, and essential oil yield, but negatively with benzaldehyde. Ellagic acid and quercetin showed extensive positive associations with monoterpenes and hydrocarbons (linalool, β-geranial, rose oxide, para-cymene, 2-undecanone, 1-hexanol), but negative correlations with phenethyl derivatives, highlighting their regulatory role in fragrance biosynthesis.

Overall, the pattern of positive and negative correlations demonstrates both cooperative and competitive interactions across phenolic and volatile pathways. These two-way relationships suggest that phenolic accumulation and volatile emission are dynamically interconnected, with AMF and SiNPs co-application potentially reshaping metabolic fluxes to enhance flower quality, fragrance, and essential oil yield in R. damascena.

3.3. Volatile Compounds

The effects of SiNPs, AMF, genotype, and their interactions were significant for most volatile phenolic compounds, with the exception of a few traits as shown in Table S6. Among aldehydes, benzaldehyde was the most abundant, followed by phenylacetaldehyde. Among the terpenes, citronellol and geraniol had the highest concentrations (

Table 4. Mean comparison (±standard deviation) of aromatic compounds (%) of R. damascena under AMF and SiNP treatments.

AMF	SiNP	Gen	hex-(2e)-enal	benzaldehyde	phenylacetaldehyde	para-cymene	linalool	rose oxide	citronellol	α-geraniol
0	0	D231	0.13 ± 0.02 c	1.26 ± 0.032 b	0.81 ± 0.09 d	0.09 ± 0.01 d	0.16 ± 0.01 d	0.19 ± 0.01 h	1.45 ± 0.03 f	0.73 ± 0.11 e
0	0	C193	0.16 ± 0.01 b	1.83 ± 0.38 a	1.41 ± 0.08 b	0.12 ± 0.01 c	0.17 ± 0.01 cd	0.66 ± 0.03 a	2.09 ± 0.02 bc	0.76 ± 0.10 e
0	2	D231	0.10 ± 0.01 d	1.37 ± 0.24 b	1.81 ± 0.04 a	0.09 ± 0.01 d	0.12 ± 0.01 f	0.26 ± 0.02 f	1.88 ± 0.12 de	0.77 ± 0.07 e
0	2	C193	0.14 ± 0.01 c	1.95 ± 0.16 a	1.16 ± 0.21 c	0.13 ± 0.02 c	0.17 ± 0.01 cd	0.57 ± 0.01 c	2.11 ± 0.06 abc	0.80 ± 0.08 e
0	4	D231	0.07 ± 0.01 e	1.13 ± 0.07 b	1.67 ± 0.17 a	0.08 ± 0.01 d	0.13 ± 0.01 ef	0.25 ± 0.01 f	1.85 ± 0.05 de	0.94± 0.04 d
0	4	C193	0.11 ± 0.02 d	1.32 ± 0.08 b	0.73 ± 0.11 de	0.12 ± 0.01 c	0.21 ± 0.02 b	0.46 ± 0.01 d	1.93 ± 0.09 d	1.17 ± 0.12 c
1	0	D231	0.08 ± 0.01 e	0.65 ± 0.05 c	0.56 ± 0.07 f	0.10 ± 0.01 d	0.12 ± 0.01 f	0.23 ± 0.02 g	1.88 ± 0.04 de	1.68 ± 0.09 a
1	0	C193	0.28 ± 0.01 a	1.35 ± 0.14 b	1.05 ± 0.02 c	0.17 ± 0.01 b	0.25 ± 0.01 a	0.63 ± 0.01 b	2.20 ± 0.01 ab	1.32 ± 0.01 b
1	2	D231	0.10 ± 0.02 d	1.10 ± 0.01 b	0.83 ± 0.02 d	0.08 ± 0.01 d	0.12 ± 0.01 f	0.25 ± 0.01 f	2.22 ± 0.09 a	0.94 ± 0.06 d
1	2	C193	0.14 ± 0.01 c	1.27 ± 0.11 b	0.63 ± 0.05 f	0.13 ± 0.01 c	0.20 ± 0.02 b	0.44 ± 0.02 e	1.78 ± 0.01 e	1.55 ± 0.10 a
1	4	D231	0.13 ± 0.01 c	1.12 ± 0.06 b	1.07 ± 0.01 c	0.09 ± 0.02 d	0.14 ± 0.01 e	0.23 ± 0.02 g	1.86 ± 0.13 de	0.95 ± 0.08 d
1	4	C193	0.11 ± 0.02 d	0.75 ± 0.10 c	1.03 ± 0.08 c	0.19 ± 0.01 a	0.17 ± 0.01 cd	0.43 ± 0.01 e	2.08 ± 0.03 c	1.63 ± 0.03 a
AMF	SiNP	Gen	β-geranial	phenethyl formate	2-phenethyl acetate	2-undecanone	pentanoic acid	1-hexanol	phenethyl alcohol
0	0	D231	0.09 ± 0.02 gh	0.16 ± 0.01 bcd	0.46 ± 0.04 bc	3.24 ± 0.56 d	0.10 ± 0.01 i	0.85 ± 0.07 ef	87.2 ± 0.78 a
0	0	C193	0.14 ± 0.01 cd	0.16 ± 0.01 bcd	0.40 ± 0.01 cde	4.30 ± 0.30 bc	0.12 ± 0.01 gh	1.12 ± 0.09 bc	83.7 ± 0.49 def
0	2	D231	0.10 ± 0.01 fg	0.15 ± 0.01 cde	0.41 ± 0.02 cde	3.64 ± 0.49 cd	0.12 ± 0.01 gh	0.82 ± 0.10 efg	85.5 ± 1.10 a–d
0	2	C193	0.16 ± 0.02 b	0.15 ± 0.01 cde	0.39 ± 0.03 de	4.60 ± 0.06 b	0.16 ± 0.01 e	1.30 ± 0.11 b	83.1 ± 0.30 ef
0	4	D231	0.11 ± 0.01 ef	0.18 ± 0.01 a	0.60 ± 0.06 a	3.18 ± 0.32 d	0.15 ± 0.01 ef	0.61 ± 0.13 g	85.9 ± 0.43 abc
0	4	C193	0.23 ± 0.02 a	0.14 ± 0.01 e	0.50 ± 0.07 b	4.89 ± 0.68 b	0.21 ± 0.01 c	1.00 ± 0.10 cde	83.8 ± 1.19 def
1	0	D231	0.14 ± 0.02 cd	0.17 ± 0.02 ab	0.50 ± 0.03 b	4.72 ± 0.50 b	0.24 ± 0.01 b	0.72 ± 0.03 fg	84.9 ± 2.17 b–e
1	0	C193	0.15 ± 0.01 bc	0.14 ± 0.01 e	0.37 ± 0.02 e	6.72 ± 0.22 a	0.23 ± 0.01 b	1.52 ± 0.17 a	80.1 ± 0.17 g
1	2	D231	0.12 ± 0.01 de	0.16 ± 0.01 bcd	0.46 ± 0.03 bc	3.37 ± 0.49 d	0.14 ± 0.01 f	1.00 ± 0.09 cde	86.1 ± 0.53 abc
1	2	C193	0.17 ± 0.01 b	0.15 ± 0.02 cde	0.41 ± 0.07 cde	4.30 ± 0.19 bc	0.18 ± 0.01 d	1.08 ± 0.17 cd	84.0 ± 0.85 cde
1	4	D231	0.11 ± 0.01 ef	0.18 ± 0.02 a	0.50 ± 0.04 b	3.46 ± 0.44 d	0.14 ± 0.01 f	0.97 ± 0.18 cde	86.3 ± 0.17 ab
1	4	C193	0.07 ± 0.001 h	0.17 ± 0.01 ab	0.39 ± 0.06 de	6.46 ± 0.60 a	0.34 ± 0.04 a	0.87 ± 0.12 def	81.8 ± 2.79 fg

In each column, means sharing the same letter do not differ significantly, as determined by the LSD test at p ≤ 0.05.

).

Alcohols constituted the major class of aromatic compounds in Damask rose. Within this group, phenethyl alcohol was the most abundant, followed by benzyl alcohol and 1-hexanol, respectively. Other detected components include esters, ketones, and straight-chain fatty acids, with ketones being the most abundant. Overall, the primary aromatic compounds identified in descending order of abundance were phenethyl alcohol, 2-undecanone, benzyl alcohol, citronellol, benzaldehyde, geraniol, phenylacetaldehyde, and 1-hexanol (Table 4). Both genotypes displayed distinct volatile component profiles. Genotype C193 generally had higher levels of most volatile compounds compared to D231.

Genotype D231 was characterized by relatively higher concentrations of phenylacetaldehyde, 2-phenethyl acetate, and 2-undecanone, whereas C193 showed greater accumulation of hex-(2E)-enal, p-cymene, linalool, geranial, pentanoic acid, and 1-hexanol under different treatments (Table 4). The presence of AMF clearly shifted the volatile profile, enhancing the levels of hex-(2E)-enal, p-cymene, citronellol, geraniol, and 1-hexanol, whereas its absence favored the accumulation of benzaldehyde, phenylacetaldehyde, rose oxide, geranial, and phenethyl alcohol. Application of SiNPs exerted a selective influence: 4 mg/L SiNPs reduced benzaldehyde, whereas 2 mg/L SiNPs promoted geraniol, 2-undecanone, and 1-hexanol (Table 4).

Genotype-specific patterns were also observed. In C193, rose oxide and phenylacetaldehyde consistently decreased across all treatments, whereas compounds such as hex-(2E)-enal, p-cymene, linalool, geranial, pentanoic acid, and 1-hexanol were markedly enhanced, especially under AMF with 4 mg/L SiNPs (Table 4). In contrast, D231 tended to accumulate benzaldehyde, citronellol, phenylacetaldehyde, rose oxide, and geraniol, with the maximum levels generally achieved under AMF inoculation in the absence of SiNPs. Notably, C193 responded more strongly to the combined AMF–SiNP treatment, whereas D231 reached its peak volatile content when AMF was applied alone.

These results indicate that the balance between aldehydes (benzaldehyde and phenylacetaldehyde), terpenes (citronellol, geraniol, and linalool), and alcohols (phenethyl alcohol, 1-hexanol, and benzyl alcohol) is strongly modulated by both genotype and treatment, with AMF being central in shaping the volatile bouquet of R. damascena.

The correlation analysis (Table S4) revealed complex interactions among volatile constituents of R. damascena under AMF and SiNPs treatments. Benzaldehyde and phenylacetaldehyde were negatively correlated with α-geraniol, indicating a trade-off between benzenoid aldehydes and monoterpene alcohols. Para-cymene, linalool, rose oxide, and citronellol showed broad positive correlations with each other and with aliphatic volatiles (2-undecanone, pentanoic acid, 1-hexanol), but were generally negatively correlated with phenethyl derivatives, highlighting a competitive relationship between terpenoid/aliphatic and phenylpropanoid pathways. Rose oxide and linalool emerged as central nodes, positively linked with multiple terpenes and aliphatics, but negatively associated with phenethyl alcohol, phenethyl formate, and 2-phenethyl acetate. Phenethyl alcohol itself displayed consistent negative correlations with terpenoids and aliphatics, confirming its antagonistic role in balancing fragrance composition. In contrast, α-geraniol correlated positively with 2-undecanone, pentanoic acid, and essential oil yield, while β-geranial was positively associated with antioxidant activity, suggesting dual roles in fragrance and stress protection.

Overall, the pattern of correlations demonstrates a coordinated yet competitive network where positive associations reflect cooperative biosynthetic fluxes and negative ones indicate trade-offs across metabolic branches. These two-way relationships suggest that phenylpropanoid, terpenoid, and aliphatic-derived volatiles are dynamically interconnected, and that AMF and SiNPs co-application may shift this balance toward enhanced fragrance quality and essential oil productivity in R. damascena.

3.4. Essential Oil Content and Antioxidant Capacity

Analysis of variance (Table S7) revealed significant main effects of AMF, SiNPs, and genotype on the studied traits, whereas their interactions were not significant. Genotype C193 exhibited higher essential oil content than D231. Both AMF inoculation and SiNPs application significantly enhanced EO content relative to the control. Total antioxidant activity was markedly increased by both SiNPs and AMF treatments. The highest TAA was observed in genotype D231 under the combined treatment of SiNPs and AMF, followed by genotype C193 under AMF alone, highlighting the strong effect of AMF on antioxidant enhancement (

Table 5. Mean comparison (±standard deviation) of antioxidant parameters and essential oil content of R. damascena under AMF and SiNP treatments.

AMF	SiNP	Gen	TPC (mg/gDW)	TFC (mg/gDW)	TAA (mmol/gDW)	EO (%)
0	0	D231	29.23 ± 0.67 g	4.98 ± 0.21 f	0.518 ± 0.09 f	0.036 ± 0.002 c
0	0	C193	48.24 ± 2.30 e	4.52 ± 0.29 f	0.589 ± 0.01 c–f	0.040 ± 0.003 bc
0	2	D231	47.98 ± 1.20 e	4.99 ± 0.35 f	0.664 ± 0.06 a–d	0.042 ± 0.004 b
0	2	C193	43.71 ± 2.15 f	6.18 ± 0.27 de	0.625 ± 0.08 b–e	0.041 ± 0.005 bc
0	4	D231	51.66 ± 2.14 d	4.64 ± 0.24 f	0.575 ± 0.03 def	0.041 ± 0.005 bc
0	4	C193	43.61 ± 2.35 f	6.85 ± 0.13 d	0.717 ± 0.05 ab	0.048 ± 0.003 a
1	0	D231	56.34 ± 1.35 c	10.63 ± 0.39 c	0.675 ± 0.01 abc	0.047 ± 0.002 a
1	0	C193	69.54 ± 1.16 b	6.00 ± 0.32 e	0.737 ± 0.04 a	0.049 ± 0.002 a
1	2	D231	55.14 ± 3.55 c	6.23 ± 0.77 de	0.559 ± 0.03 ef	0.048 ± 0.002 a
1	2	C193	56.58 ± 0.65 c	5.97 ± 0.57 e	0.733 ± 0.10 a	0.050 ± 002 a
1	4	D231	74.31 ± 3.07 a	23.24 ± 0.93 a	0.755 ± 0.6 a	0.049 ± 0.004 a
1	4	C193	46.63 ± 1.04 ef	12.46 ± 0.42 b	0.529 ± 0.03 f	0.052 ± 0.004 a

TPC: total phenolic content; TFC: total flavonoid content; TAA: total antioxidant activity; EO: essential oil percentage. In each column, means sharing the same letter do not differ significantly, as determined by the LSD test at p ≤ 0.05.

).

For secondary metabolites, the maximum TPC and TFC were recorded for genotype D231 when treated with both SiNPs and AMF, demonstrating a synergistic effect. In genotype C193, AMF alone maximized TPC, while SiNPs alone led to the highest TFC, indicating trait- and genotype-specific responses to the treatments (Table 5). The correlation analysis revealed a significant positive relationship among TPC, TAA, and EO (Table S4).

These results demonstrate that both AMF and SiNPs are effective biostimulants for enhancing the essential oil yield and antioxidant capacity of R. damascena. While C193 displayed a stronger capacity for EO production, D231 showed a greater potential for the accumulation of phenolic and flavonoid compounds, particularly under combined AMF and SiNPs treatment. This genotype-specific response highlights the importance of tailoring biotic and abiotic inputs to maximize the targeted biochemical and functional traits. Optimizing such strategies could significantly improve the pharmaceutical and industrial value of Damask rose.

4. Discussion

The morphological improvements observed in R. damascena could be mechanistically linked to the well-established roles of AMF and SiNPs in plant physiology. AMF symbiosis enhances nutrient and water uptake, particularly phosphorus, and promotes root architecture, which facilitates greater assimilation and allocation to reproductive organs [31,32]. This enhanced nutrient and water availability likely underlies the increased pedicle length and flower diameter, because improved turgor and cellular expansion are directly dependent on sufficient nutrient supply and hydration. However, caution is warranted when attributing these effects solely to increased water and nutrient uptake because these parameters were not directly measured in this study. Moreover, AMF can modulate phytohormone levels, including auxins and cytokinins, which regulate cell division and elongation, thus providing an additional explanation for the observed structural changes in floral organs [32].

Similarly, SiNPs contribute to morphological enhancement through their ability to strengthen cell walls, improve photosynthetic efficiency, and mitigate biotic and abiotic stresses, thereby facilitating better assimilate partitioning toward flowers [33]. By enhancing the structural integrity and physiological resilience of tissues, SiNPs support the development of larger and more robust floral organs, thus complementing the nutrient-driven effects of AMF [34].

A key insight from this study is the synergistic interaction between AMF and SiNPs. This co-application likely operates through multiple complementary mechanisms; AMF may enhance Si uptake and translocation, improving its availability in plant tissues, while SiNPs may stabilize AMF structures or improve fungal colonization efficiency, amplifying the benefits of mycorrhization. This bidirectional facilitation could explain the pronounced enhancement in morphological traits under the combined treatments. Such synergistic interactions have also been reported in other crops, including baby corn, where AMF and SiNPs together produced greater biomass and secondary metabolite accumulation than either treatment alone [23].

Genotype is another critical determinant of morphological performance, indicating a stronger genetic potential for resource allocation to reproductive structures. This finding supports the importance of genotype-specific responses in breeding programs targeting floral productivity. Similar genotypic variability was reported by Chettri et al. (2024) for Rosa spp., where significant differences in floral yield traits were attributed to genetic diversity among landraces [35]. Interestingly, 2 mg/L SiNPs was more effective than 4 mg/L Si in enhancing flower weight, especially when combined with AMF. This suggests a possible hormetic effect, where lower concentrations of SiNPs optimize metabolic processes while avoiding potential toxicity or metabolic burden at higher doses. Such dose-dependent responses have been reported in other horticultural species, including in the work of Tofighi et al. (2021) [36], where moderate SiNPs levels improved the flower quality of gerbera, while excessive SiNPs did not yield further benefits. Overall, the morphological enhancements of pedicle length, flower diameter, and other reproductive structures in R. damascena can be interpreted as the outcome of enhanced nutrient acquisition, hormonal modulation, tissue fortification, and stress mitigation, with the AMF–SiNPs synergy representing a promising strategy for optimizing floral productivity and quality in both horticultural and commercial contexts [37].

The observed enhancement in secondary metabolite biosynthesis can be mechanistically linked to the well-established role of AMF in stimulating plant metabolic networks. AMF colonization is known to influence phenylpropanoid and shikimate pathways, leading to increased synthesis of phenolic and flavonoid compounds, which function as key antioxidants and stress mitigators [32,38,39]. These effects are mediated in part through AMF-induced modulation of plant signaling cascades and phytohormones, which prime the plant’s metabolic machinery to enhance secondary metabolite production [39,40]. Moreover, AMF are known to positively influence essential oil biosynthesis in aromatic plants, likely by regulating enzyme activity and precursor allocation within terpene biosynthesis pathways [19,41,42,43].

These findings confirm that SiNPs act as beneficial modulators of plant metabolism, enhancing stress tolerance, hormonal balance, and ROS detoxification [44,45]. SiNPs can induce the activity of phenylalanine ammonia-lyase (PAL), a key enzyme in the phenylpropanoid pathway, which is directly responsible for phenolic and flavonoid accumulation [46]. The modest yet significant effect on EO content further supports the regulatory impact of SiNPs on terpene biosynthesis pathways, possibly by modulating gene expression and precursor flux through the MEP or mevalonate pathways [47]. These interpretations highlight that AMF and SiNPs function through complementary physiological, biochemical, and signaling pathways to enhance secondary metabolite production. Genotype-dependent responses indicate that biostimulant strategies should be customized for each cultivar to maximize phytochemical quality in aromatic and medicinal plants [41,48,49,50,51].

A critical aspect of this study was the synergistic interaction between AMF and SiNPs, which can be interpreted through several complementary mechanisms. AMF may enhance the uptake and translocation of silicon, increasing its bioavailability within plant tissues, whereas SiNPs may facilitate fungal colonization or activity, thereby amplifying the metabolic priming effects of AMF. Additionally, the combined presence of AMF and SiNPs may act as a co-elicitor, reinforcing signaling pathways that drive secondary metabolite biosynthesis, including phenolics, flavonoids, and essential oils [23]. This synergy is particularly important in genotype-specific contexts, as differential responsiveness to AMF and SiNPs may be mediated by genetic variation in metabolic capacity and receptor-mediated elicitation [52].

The predominance of specific phenolic compounds such as rutin, vanillic acid, syringic acid, gallic acid, and rosmarinic acid can be explained by the selective modulation of secondary metabolite pathways by AMF and SiNPs. These findings are in agreement with previous studies that reported rutin as the dominant flavonoid in Damask rose petals, particularly under stress or elicitor treatments [53,54]. AMF colonization is recognized to activate key biosynthetic pathways in plants, including the shikimate and phenylpropanoid pathways, through hormonal signaling and metabolic priming, which can selectively enhance compounds such as syringic acid [32]. In contrast, SiNPs act as effective biochemical elicitors, modulating enzyme activities and influencing precursor flux toward flavonoid and phenolic biosynthesis [46,47]. This can explain the induction of compounds such as apigenin under SiNPs treatments, which are otherwise absent under non-stimulated conditions. Genotypic differences further shape the phenolic profile, reflecting intrinsic metabolic capacity and responsiveness to elicitors. Certain genotypes, such as C193, may possess higher baseline metabolic activity, enabling a broader and stronger accumulation of phenolics, whereas others, such as D231, may exhibit more selective responses, particularly to AMF colonization. These genotype-specific patterns highlight the importance of matching biostimulant strategies to cultivar identity in order to optimize secondary metabolite production [52,55,56]. Our results align with the findings of Zare et al. (2023), who reported the synergistic effects of nanoparticles and AMF in enhancing gallic acid in aromatic plants [56]. The synergistic interaction between AMF and SiNPs provides a plausible mechanism for enhanced phenolic accumulation. AMF may improve silicon uptake or signaling efficiency, while SiNPs may facilitate fungal establishment or metabolic activity, resulting in the co-elicitation of biosynthetic pathways. Such complementary effects can enhance both the diversity and concentration of phenolic compounds, depending on genotype-specific metabolic predispositions [23,55].

Interestingly, p-coumaric acid levels declined in both genotypes under treatment, a trend also noted by Tak et al. (2023), possibly due to a metabolic shift toward other phenolic biosynthetic pathways [57]. Overall, these interpretations emphasize that AMF and SiNPs act through complementary physiological and biochemical mechanisms, with genotype-dependent responsiveness, to regulate phenolic biosynthesis in Damask rose, offering potential strategies for targeted metabolic enhancement in aromatic and medicinal plants [32,46,57].

The composition of volatile and semi-volatile compounds in R. damascena is strongly influenced by the complementary effects of AMF and SiNPs, reflecting complex metabolic regulation. AMF colonization can modulate monoterpene and phenylpropanoid pathways, selectively enhancing key aroma compounds such as geraniol and phenethyl alcohol [32,58,59].

Interestingly, rose oxide, another important volatile known for its sweet and floral aroma, decreased by 7.92% under the AMF treatment. This reduction may be due to a shift in precursor allocation or enzymatic competition within the monoterpene biosynthetic pathway. Overall, these results highlight the selective influence of AMF on aroma compound composition, potentially enhancing the fragrance qualities of R. damascena. These findings are consistent with prior studies highlighting that AMF symbiosis promotes the biosynthesis of monoterpenes and sesquiterpenes by modulating primary metabolism and enhancing nutrient and water uptake [32]. In this study, the citronellol-to-geraniol ratio decreased under AMF treatment. Both geraniol and citronellol are key monoterpenoid alcohols that contribute to the characteristic floral and citrus aromas of R. damascena and are vital in the sensory quality of aromatic plants. The observed shift in their ratio suggests that AMF selectively enhances geraniol biosynthesis to a greater extent than citronellol. This modulation in volatile composition could lead to a perceptible change in fragrance balance, potentially improving the aromatic quality toward more pronounced floral notes, which are highly valued in the perfumery and flavor industries [59]. The increase in phenethyl alcohol and phenylacetaldehyde in mycorrhizal plants is also noteworthy, consistent with reports by Kumar et al. (2025) [58] and suggesting AMF-induced modulation of the phenylpropanoid and shikimate pathways. Interestingly, while some compounds, such as benzaldehyde and phenylacetaldehyde, were higher in AMF-free conditions, this could reflect stress-induced metabolite accumulation or a shift in carbon allocation due to reduced symbiotic demand [60]. Silica acts as a biochemical modulator, influencing secondary metabolism by affecting enzyme activities, precursor availability, and hormonal signaling pathways, such as jasmonic acid and salicylic acid [44,45]. SiNPs can enhance the biosynthesis of terpenoids and alcohols, including geranial, citronellol, and phenylacetaldehyde, possibly through dose-dependent effects on stress mitigation and antioxidant status [47]. These mechanisms align with the known role of Si in modulating secondary metabolism under both optimal and stress conditions. These compounds play major roles in plant–insect and plant–human interactions, and their elevation indicates the potential of SiNPs to improve both biotic resistance and flavor/aroma profiles [47].

The synergistic interaction between AMF and SiNPs appears to amplify the biosynthesis of volatile compounds. Potential mechanisms include enhanced silicon uptake facilitated by AMF, which may improve fungal activity or metabolic signaling, whereas SiNPs may support AMF establishment and effectiveness. Together, these complementary effects can optimize flux through the terpenoid and phenylpropanoid pathways, increasing both the diversity and concentration of aroma compounds in a genotype-specific manner [23,52]. This variation emphasizes the necessity of genotype selection in breeding programs aimed at improving flavor and nutritional quality through sustainable practices. Such interactions also support the notion that volatile compound biosynthesis is under complex genetic and environmental regulation, as indicated by studies in tomatoes [48] and peppers [49]. Overall, these insights highlight that AMF and SiNPs act through complementary physiological and biochemical mechanisms, enabling fine-tuning of the volatile profile in Damask rose. This integrative approach offers a promising strategy for optimizing fragrance quality and secondary metabolite content in aromatic crops, aligning with sustainable and precise horticultural practices. Although the present study clearly demonstrated the beneficial effects of SiNPs and AMF on the morphological and biochemical traits of R. damascena, it did not address the potential environmental concerns related to the use of SiNPs. Previous reports suggest that SiNPs may persist in the soil, interact with microbial communities, or potentially enter the food chain, raising questions about their long-term ecological safety [15,61,62]. However, in our experiment, the concentrations of SiNPs applied did not exceed the natural capacity of R. damascena leaves to accumulate silica (0.1–0.5 mg g⁻¹ DW in dicotyledonous species). Indeed, ICP analysis of leaf tissues revealed that residual Si levels under different treatments ranged from 0.0019 to 0.0078 mg g⁻¹ DW, which falls within the natural non-toxic range reported for dicotyledonous plants. Moreover, the synergistic application of AMF and SiNPs may enhance plant metabolic efficiency and nutrient utilization without increasing the environmental load, reflecting a complementary effect that supports both plant performance and ecological safety. Additionally, SiNP application has been associated with improved nutrient-use efficiency, reduced uptake of toxic heavy metals, and enhanced photosynthetic capacity, thereby minimizing potential environmental risks. Collectively, these observations indicate that the application of SiNPs, either at appropriate doses alone or synergistically with AMF, aligns with the principles of sustainable agriculture, and can simultaneously enhance soil health, crop productivity, and ecological balance.

5. Conclusions

This study provides a comprehensive evaluation of the effects of AMF inoculation and SiNPs application on the growth, phytochemical composition, and metabolic response of R. damascena. These findings highlight the multifactorial nature of morphological and biochemical regulation, demonstrating that both biotic (AMF) and abiotic (SiNP) inputs significantly enhance floral yield, essential oil content, antioxidant activity, and secondary metabolite accumulation, with responses strongly modulated by genotype.

The results indicated that genotype C193 exhibited a higher intrinsic capacity for essential oil production, whereas genotype D231 showed greater inducibility of phenolic and flavonoid compounds, particularly under the combined AMF and SiNP treatments. This genotype-specific response underscores the importance of tailoring biostimulant applications to optimize both yield and phytochemical quality.

Analysis of volatile and phenolic profiles revealed that AMF and SiNPs could modulate the composition of aroma-related metabolites and bioactive compounds, providing a mechanistic basis for improving essential oil quality and antioxidant potential. The combined application of these biostimulants demonstrated synergistic effects, enhancing both quantitative yield traits and metabolic outputs, thereby offering a practical strategy for improving the commercial, medicinal, and aromatic value of Damask rose.

Overall, this research establishes a scientific framework for integrating AMF and SiNPs into rose cultivation, supporting sustainable and environmentally friendly practices. These findings not only advance our understanding of the metabolic plasticity of R. damascena but also provide actionable insights for genotype-specific cultivation strategies aimed at maximizing both horticultural performance and phytochemical enrichment.