Biometric, Physiological, and Genetic Profile of Chrysanthemum Plants in Response to CdS, Co₃O₄, and Fe₃O₄@Co Nanoparticles Treatment

Introduction

In recent years, the convergence of nanotechnology and biotechnology in plant science has begun a new area of exploration, enabling unprecedented insights into the intricacies of plant growth and development.¹ Nanoparticles (NPs) are defined as atomic or molecular aggregates less than 100 nm in size, possessing unique and captivating physical and chemical properties compared to micrometer-sized materials. Nanoparticles, due to their small size, increased number of surface atoms, huge surface area-to-volume ratio, high particle energy, and catalytic reactivity, easily interact with biological systems, eg, plant cells.^2–4 Interactions between nanoparticles and living organisms are complex and often ambiguous. The same type of nanoparticles can cause different biological effects depending on the genotype, organ, age and developmental stage of the plant, cultivation conditions, time of exposure to nanoparticles, and depending on the physicochemical properties of nanoparticles, including their composition, functional groups on the surface, type of coating, concentration, size, shape, stability or method of plant treatment.^5–7

The application of nanotechnology in agriculture has greatly improved plant protection and nutrition, and significantly increased efficiency and productivity. Research in the realm of horticultural plants is still in its early stages. Among other applications, the effects of nanoparticles on in vitro plant regeneration have emerged as a fascinating area of study.^1,3 A review of scientific publications provides reports on the positive aspects of using various nanoparticles in plant in vitro cultures – most often for eliminating microbiological contamination, stimulating seed germination and seedling development, inducing caulogenesis, organogenesis and somatic embryogenesis, or producing metabolites. Nanoparticles can also negatively affect plant metabolism, growth, and development.^8,9 It is assumed that nanoparticles modify structural elements of cell membranes and cellular macromolecules, affect cell division, induce oxidative stress, and disrupt physiological and biochemical processes of plants.^10,11

The genotoxicity of nanoparticles has been especially studied in microbes and mammals, while it is little examined in plants. It has been proven that nanoparticles can induce genotoxicity either directly or indirectly. In the direct mechanism, nanoparticles pass through the cell and nuclear membrane and interact with DNA mechanically or chemically. Indirect genotoxicity of nanoparticles results from interaction with proteins (involved in replication, transcription, translation, etc) or induction of oxidative stress, which plays a critical role in nanotoxicity. Nanoparticle treatment induces chromosomal aberrations, affects the cell cycle and DNA repairing system, alters gene expression, causes DNA damage, and leads to epigenetic variations in the plant genome.^12,13 The 10, 20, 40, and 50 mg·L⁻¹ Ag NPs treated root tip cells of wheat (Triticum aestivum L.) exhibited various types of chromosomal aberrations, such as incorrect orientation at metaphase, chromosomal breakage, metaphasic plate distortion, spindle dysfunction, stickiness, aberrant movement at metaphase, fragmentation, scattering, unequal separation, chromosomal gaps, multipolar anaphase, erosion, as well as distributed and lagging chromosomes.¹⁴ Molecular markers can also be used to detect the nanoparticles’ effect on the plant genome and to determine potential mutations in the DNA sequence.¹⁵

Chrysanthemum × morifolium (Ramat.) Hemsl. (formerly Chrysanthemum × grandiflorum /Ramat./ Kitam.) (Asteraceae) is one of the most culturally and economically important ornamental plant species in the world.^16,17 The value of its inflorescence is primarily determined by the ligulate flowers, which exhibit a wide variety of colors and shapes. Inflorescences also contain biologically active metabolites and are a herbal material.¹⁸ The demand for cultivars with new inflorescence and/or leaf features, increased stress tolerance, or diversified plant habit is growing year by year. Moreover, improvements in large-scale cuttings production with the use of micropropagation are also of great importance.¹⁹

Exploring the genotoxic effects of nanoparticles on plants is particularly valuable for the development of breeding methods. The novelty and originality of our study lie in synthesizing potentially toxic and mutagenic metal-sulfide-based (CdS) and metal oxide-based (Co₃O₄ and Fe₃O₄@Co) nanoparticles and using them to induce molecular and phenotypic variability in chrysanthemum. Novel, user-friendly, easily accessible, and moderately expensive techniques of plant improvement are desirable. It is crucial for ornamental plants due to the huge market demand for new cultivars. So far, only the usefulness of silver nanoparticles (Ag NPs) in chrysanthemum breeding has been demonstrated.^20,21 Against, zinc oxide nanoparticles (ZnO NPs) and zinc oxide nanoparticles combined with silver nanoparticles were successfully used to improve the efficiency of adventitious shoots regeneration,²² and biometric parameters of chrysanthemum shoots and/or roots in meristematic explant cultures.²³

Cadmium sulfide nanoparticles (CdS NPs) have proven biomedical properties and are popular in biosensors, bioimaging, antibacterial, and anticancer applications.²⁴ Ullah et al²⁵ applied CdS NPs (3–100 mg·L⁻¹) on maize (Zea mays L.) seedlings, and observed chlorotic effects, metabolic disruption, as well as reduced shoot and root biomass. In plants, cadmium toxicity causes oxidative stress, destruction of cell membranes, biomolecules, and organelles. Cd also reduces uptake of Fe and Zn, resulting in leaf chlorosis, and interferes with the transport and uptake of Ca, P, Mg, K, and Mn.²⁶ Excessive concentrations of sulfur also harm plants, damaging the root system, foliage, and reducing growth.²⁷

Cobalt (Co) is one of the toxic metals for plants. When its level increases, it leads to severe toxicity. Cobalt stress inhibits plant growth and development through oxidative stress. It also reduces nutrient uptake and photosynthetic efficiency.²⁸ Cobalt oxide nanoparticles (Co₃O₄ NPs) are used in wastewater treatment, drug delivery, magnetic resonance imaging, electrochromic sensors, heterogeneous catalysis, and energy storage devices.^2,29 They also have various biomedical applications due to distinctive antioxidant, antimicrobial, antifungal, larvicidal, antileishmanial, and anticholinergic properties.³⁰ In the study conducted by Ogunyemi et al.³¹ Co₃O₄ NPs improved the agronomic parameters and biomass of rice (Oryza sativa L.) and increased the photosynthetic parameters of Arabidopsis thaliana (L.) Heynh. In rapeseed (Brassica napus L.), Co₃O₄ NPs applied at low concentrations stimulated growth; however, high concentrations caused extensive oxidative damage and mediated defense responses.³² Faisal et al³³ reported genotoxic effects of cobalt oxide nanoparticles on eggplant (Solanum melongena L).

Iron oxide nanoparticles (Fe₃O₄ NPs) are one of the most widely explored nanomaterials. Due to their unique properties (superparamagnetism, inherent biocompatibility), they have found numerous applications in medical diagnostics, controlled drug release, separation technologies, and environmental engineering.³⁴ Fe₃O₄ NPs promoted growth and chlorophyll accumulation in muskmelon (Cucumis melo L.)³⁵ or increased chlorophyll content and vegetative growth of chrysanthemum plants in greenhouse experiments.³⁶ Bare and citrate-stabilized (CA) iron oxide nanoparticles affected the germination and growth of chrysanthemum ‘Richmond’ propagated from synthetic seeds.³⁷ In this approach, nanoparticles in combination with indole acetic acid (IAA) were added to the alginate bead matrix containing a single axillary bud. It was found that shorter exposure to Fe₃O₄CA stimulated the elongation of shoots in vitro and improved the acclimatization efficiency of plants in the greenhouse, resulting in improved leaf growth metrics. Negative effects were evident after extended exposure. Therefore, it is interesting to explore the effects of other nanoparticle types, not only for reproduction purposes but also for mutation breeding, and to study other cultivars. The excess of iron ions in plants can cause oxidative stress, damage to membranes, lipids, proteins, and DNA, stunted growth, reduced root development, decreased water and nutrient uptake, and reduced photosynthetic activity.³⁸ To date, there are no studies on the use of the proposed heavy metal-based nanoparticles in mutation breeding of ornamental plants. Importantly, in addition to the well-known toxicity of heavy metal ions to plants, CdS, Co₃O₄, and Fe₃O₄@Co nanoparticles possess semiconducting properties that enable them to generate reactive oxygen species (ROS).^39–41 These, in turn, directly affect plant physiology by increasing oxidative stress.

Undoubtedly, interactions between nanoparticles and plants require further detailed research to better understand and use the potential that nanotechnology offers to modern agriculture and horticulture. The specific objectives of our study focused on: (1) synthesis and physicochemical characteristic of CdS, Co₃O₄, and cobalt doped Fe₃O₄ (Fe₃O₄@Co) NPs, (2) establishment of chrysanthemum in vitro cultures and treatment of internode explants with nanoparticles, (3) determination of the efficiency of in vitro adventitious shoot regeneration, (4) assessment of the extent of oxidative stress by analyzing the biochemical profile (pigment content, chlorophyll fluorescence, antioxidant capacity, and enzymatic activity) of plant tissues in vitro, (5) phenotype and genetic analyses of plants at the stage of ex vitro cultivation in the glasshouse; to further reveal effects of these nanoparticles on plants and study their use in plant mutation breeding. The research hypothesis assumed that selected types of nanoparticles can penetrate plant cells and induce changes at the physiological and molecular level, and the resulting genetic variability among regenerated plants can manifest in their phenotype and be valuable for breeding. The use of nanoparticles as a chemical mutagen added to the micropropagation medium can make chrysanthemum breeding relatively easy and effective. Therefore, it could be routinely performed in in vitro plant laboratories, without the need for specialist equipment or the use of expensive and advanced methods of genetic transformation and genome editing.

Materials and Methods

Chemicals and Methods for Nanoparticles

Cobalt (II) chloride hexahydrate (CoCl₂.6H₂O), iron (II) chloride tetrahydrate (FeCl₂.4H₂O), iron (III) chloride hexahydrate (FeCl₃.6H₂O), and sodium hydroxide (NaOH) were supplied from the Warchem Sp. z o.o., Warsaw, Poland. Cadmium acetate (Cd(CH₃COO)₂.2H₂O), sodium sulfide nonahydrate (Na₂S∙9H₂O), and sodium borohydride (NaBH₄) were supplied from Sigma-Aldrich, St. Louis, MO, USA. All the chemicals were used without further purification. Ethanol (EtOH) and hydrochloric acid (HCl) were purchased from POCH, Gliwice, Poland. Deionized (DI) water was obtained using the HYDROLAB water filtering system, Gliwice, Poland.

The morphology of nanoparticles was studied using Scanning Electron Microscopy (SEM), Merlin Zeiss, Carl Zeiss, Stuttgart, Germany, where the dried samples were placed on the conducting tape. The Attenuated Total Reflection Fourier Transform Infrared Spectroscopy (ATR-FTIR) was used to record the spectra in the 4000–400 cm⁻¹ infrared region that are characteristic of the specific vibrations in the obtained particles. Before the spectra recording, the samples were ground with a mortar and pestle. ATR-FTIR Spectrum Two Perkin Elmer was supplied by Pro-Environment Polska Sp. z o.o., Warsaw, Poland. The dynamic light scattering (DLS) method was used to measure the hydrodynamic size of the particles with Malvern Instruments Zetasizer Nano Sizer, Malvern, UK. The measurements were performed in the aqueous solution at pH 5.8, which is identical to the following in vitro conditions.

Procedures for Nanoparticles

Synthesis of CdS NPs

Two hundred and sixty-six milligrams of Cd(CH₃COO)₂ was dissolved in the 100 mL of 1:1 water-ethanol solution by magnetic stirring at 600 rpm at 80°C. Then, 240 mg of Na₂S was dissolved in 1:1 water-ethanol and added dropwise with continuous stirring and heating for 15 min. After the synthesis, the solution was cooled and washed several times with DI water and EtOH, with centrifuging. Last, the precipitate was dried overnight at 50°C.

Synthesis of Co₃O₄ NPs

First, 450 mg of CoCl₂.6H₂O was placed into a 50 mL beaker, where 20 mL of EtOH was added, and the solution was stirred with a magnetic stirrer at 600 rpm for 5 min at room temperature. Next, 500 mg of NaBH₄ was added with continuous stirring for 15 min, until the change in the color of the solution from blue to black. After synthesis, the powder was rinsed several times with water (min. 5 times) to remove excessive NaBH₄, and centrifuged at 4000 rpm for 4.5 min. Blackish precipitate was rinsed with EtOH twice and dried onto a Petri dish at 90°C for 60 min.

Synthesis of Fe₃O₄@Co NPs

To prepare the Co-doped Fe₃O₄ (labeled as Fe₃O₄@Co), 540 mg of FeCl₃6H₂O, 100 mg of FeCl₂4H₂O, and 100 mg of CoCl₂6H₂O were placed in a 50 mL volume beaker and dissolved in 20 mL of DI by a magnetic stirrer with a mixing rate of about 600 rpm. Next, the 1 M NaOH solution was added dropwise till pH 10.5 with controlled stirring at 600 rpm at 85°C for 15 min. Next, the suspension was cooled down, the precipitate was collected at the bottom of the beaker onto the magnet, and washed several times with DI water and EtOH by centrifuging. Next, the samples for analysis were dried in an oven overnight at 50°C.

Characteristics of Nanoparticles

Figure 1a clearly shows the formation of CdS granules sized ~50 nm. Following, Co₃O₄ also forms aggregates by the granules having a similar size (Figure 1b), while the cobalt-doped iron oxide Fe₃O₄@Co spherical nanoparticles are sized below 20 nm in diameter (Figure 1c).

Figure 1 Scanning Electron Microscopy (SEM) images of (a) CdS NPs, (b) Co3O4 NPs, (c) Fe3O4@Co NPs, (d) hydrodynamic diameter, and (e) Fourier Transform Infrared (FTIR) spectra for CdS, Co3O4, and Fe3O4@Co NPs.

Complementary to the SEM analysis, the DLS method was used to determine the hydrodynamic diameter d_H. Figure 1d presents d_H for CdS above 150 nm, which can relate to the agglomerates or aggregates formation, while the d_H values decrease for the Co₃O₄ ~125 nm and below 100 nm for the Fe₃O₄@Co, which are in good agreement with the SEM results for the dry samples.

Subsequently, the presence of characteristic vibrations in the nanoparticles was studied using ATR-FTIR. As can be seen in the green curve in Figure 1e, the Co-O vibrations confirm the presence of cobalt oxide, where the band at 583 cm⁻¹ Co-O relates to the Co²⁺-O vibration in tetrahedral sites. The following band at 657 cm⁻¹ is the stretching vibration of Co³⁺-O in the octahedral sites in the spinel lattice.⁴² The weak band at 505 cm⁻¹ can be associated with the Co²⁺O₆ vibration that can suggest the formation of a non-stoichiometric structure for the unreduced Co³⁺ ions in the cobalt oxide structure.⁴³ The minor band at 1254 cm⁻¹ can relate to the CO₂ residues adsorbed onto the nanoparticles, while the following band at 1350 cm⁻¹ can be ascribed to the C-O stretching vibration from the ethanol residues. The following band at 1636 cm⁻¹ can correspond to the H-O-H bending vibration of water molecules and/or ethanol.

The blue curve in Figure 1e is characteristic of the iron oxide doped with cobalt. The minor band located at 421 cm⁻¹ appears for the Fe-O vibrations in the octahedral sites, while the vibrations at 471 cm⁻¹ and 522 cm⁻¹ come from the Fe-O in the tetrahedral sites of spinel structure.⁴⁴ The following band at 713 cm⁻¹ is characteristic of the Co-O in the iron oxide doped with cobalt.⁴⁵ The bands at 974 cm⁻¹ and 1013 cm⁻¹ can come from the organic residues from the sample washing, while bands at 1379⁻¹ and 1639 cm⁻¹ are characteristic of the C-O and water, respectively.

The black curve presents the spectrum recorded for CdS, where the bands at 613 cm⁻¹ and 685 cm⁻¹ are characteristic of the CdS stretching vibration,⁴⁶ while minor bands at 539 cm⁻¹ may relate to the S-S⁴⁷ or surface oxidation.⁴⁸ The following bands in the range between 950 cm⁻¹ and 1110 cm⁻¹ can be attributed to the carboxylates in the sample from the cadmium precursor. Next, bands at 1345 cm⁻¹, 1425 cm⁻¹, and 1542 cm⁻¹ can be ascribed to the C-O, C=O, and H-O-H stretching, respectively. The presence of the broad band in all samples above 3400 cm⁻¹ is caused by the O-H vibrations.⁴⁹

In vitro Culture Establishment and Nanoparticle Treatments

Chrysanthemum × moriflorium (Ramat). Hemsl. ‘Lilac Wonder’ was used in the experiment. The plant material was obtained from the in vitro gene bank of the Laboratory of Horticulture and Landscape Architecture, Bydgoszcz University of Science and Technology, Bydgoszcz, Poland. The plants were identified and authenticated by Dr Tomasz Stosik, a member of the Polish Botanical Society (PBS). A voucher specimen (no. 1594a) was deposited in the Herbarium of the Department of Plant Microbiology and Ecology, Bydgoszcz University of Science and Technology, Bydgoszcz, Poland.

Internode explants were inoculated on the modified MS medium,⁵⁰ in which the content of calcium and iron was increased by half, 30 g·L⁻¹ sucrose, and 8 g·L⁻¹ Plant Propagation LAB-AGAR^TM (BIOCORP, Warsaw, Poland) were added. The medium was supplemented with plant growth regulators: 0.6 mg·L⁻¹ BAP (6-benzylaminopurine) and 2.0 mg·L⁻¹ IAA (indole-3-acetic acid) (Sigma-Aldrich, St. Louis, MO, USA) to induce adventitious organogenesis. The pH was adjusted to 5.8, and next, 40 mL of the medium was poured into 350 mL glass jars sealed with plastic and autoclaved at 105 kPa and 121°C for 20 min.

Four internode explants were placed, in a horizontal position, in each culture jar. The nanoparticles suspensions: CdS, Co₃O₄, or Fe₃O₄@Co NPs, at the concentration of 75 mg·L⁻¹ were sterilized in an autoclave, and, before application, placed for 30 min in the Elmasonic S80(H) Ultrasonic Cleaner with the ultrasonic frequency of 37 kHz and the effective ultrasonic power of 150 W (Elma Schmidbauer GmbH, Singen, Germany) to disassemble potential aggregates. Then, nanoparticle suspensions were poured onto the explants and culture medium with an automatic pipette, 2 mL of each suspension per culture jar. The concentration of nanoparticles suspensions was determined based on the results of our previous research on chrysanthemum breeding.²¹ Explants inoculated on the medium without nanoparticles were used as the control object. Each experimental object consisted of 15 jars (60 explants in total). After eight culture weeks, the share of explants regenerating adventitious shoots, and the mean number of adventitious shoots per inoculated explant were estimated. Biochemical analysis was performed as described in detail below. Shoots were cut off from explants and transferred to the modified MS rooting medium with 2.0 mg·L⁻¹ IAA for two weeks. Additionally, 20 shoots propagated via the single-node method on the MS medium without PGRs and NPs were also rooted as a genotype/phenotype standard of ‘Lilac Wonder’ chrysanthemum. Rooted plantlets were then cultivated ex vitro in a greenhouse.

In vitro cultures were maintained in the growth room with the following conditions: 23±1°C, 16/8 h light/dark photoperiod, photosynthetic photon flux density of 35 μmol·m⁻²·s⁻¹ (Philips TLD 36W/54 fluorescent lamps emitting cool daylight; Koninklijke Philips Electronics N.V., Eindhoven, the Netherlands).

Biochemical Activity of in vitro Regenerated Plants

Pigments and Chlorophyll Fluorescence

The relative contents of flavonols, anthocyanins, chlorophyll, and Soil Plant Analysis Development (SPAD) leaf greenness index in the fresh leaves were estimated in 80 plants per experimental treatment using a portable MPM-100 multi-pigment meter (Opti-Sciences Inc., Hudson, NH, USA). The relative values of F₀ (minimal fluorescence), F_V (variable fluorescence), F_M (maximal fluorescence), F_V/F_M and F_V/F₀ ratios in the leaves were measured in 40 plants per experimental treatment using the OS-30p+ chlorophyll fluorimeter (Opti-Sciences Inc., Hudson, NH, USA).

Enzyme Activity

To determine the enzymatic activity, fresh leaves (100 mg) were homogenized in phosphate buffer (100 mM; pH 7.4) with 1 mM EDTA, 1 mM dithiothreitol (DTT), and 2% (w/v) polyvinylpyrrolidone (PVP) according to Homaee and Ehsanpour.⁵¹ The homogenates were next centrifuged (13,000× g for 20 min at 4°C; Centrifuge MPW-260R, MPW MED INSTRUMENTS, Warsaw, Poland). Supernatants were used for the determination of the total protein content⁵² and the activities of ascorbate peroxidase (APOX; EC 1.11.1.11),⁵³ guaiacol peroxidase (GPOX; EC 1.11.1.7),^54,55 and superoxide dismutase (SOD; EC 1.15.1.1).⁵⁶ The spectrophotometric analyses were performed using the SmartSpec PlusTM spectrophotometer (BioRad, Hercules, CA, USA) at specific wavelengths: for proteins at 595 nm, for APOX at 265 nm, for GPOX at 430 nm, and for SOD at 560 nm. Measurements were repeated six times for each experimental treatment. The enzymatic activity U (μmol·min⁻¹) was calculated per 1 mg of total protein content.

Total Content of Polyphenols (TCP) and Antioxidant Capacity – Ferric Reducing Antioxidant Power (FRAP) and [2,2’-Azino-Bis(3-Ethylbenzothiazoline-6-Sulfonic Acid)] Assay (ABTS)

Fresh plant tissue samples were put in zip-lock foil bags, frozen, and stored at −20°C for 24 h (AFG 6402 E-B freezer, Whirlpool, Milan, Italy). Next, the material was lyophilized (Alpha 1–4 LSC, Martin Christ Gefriertrocknungsanlagen GmbH, Osterode am Harz, Germany) (moisture content <2%), and ground in a mill into a fine powder (0.3–0.5 mm) (Ultra-Centrifuge Retsch ZM 100, Retsch, Haan, Germany). The prepared samples were stored in the dark in air-tight bags in desiccators. For extraction, a 0.5 g sample was mixed with 30 mL of a methanol-water mixture (70:30 v/v), followed by shaking for 1 h (KS 130 basic shaker, IKA, Staufen im Breisgau, Germany). Next, the samples were centrifuged for 15 min, 3000 rpm, 4°C (ROTINA 420R centrifuge, Hettich, Tuttlingen, Germany). The resulting solution was decanted into a 100 mL flask. This procedure was repeated three times for each experimental treatment. Next, the flask was filled with the methanol-water mixture and mixed to obtain the final extract for the analyses.

The TCP was determined according to Keutgen and Pawelzik.⁵⁷ Absorbance was measured using the Shimadzu UV-1800 spectrophotometer (Shimadzu, Kyoto, Japan) at 735.5 nm. Results were calculated based on a calibration curve for gallic acid equivalent (GAE) and given as mg·g⁻¹ dry weight (DW), in four repetitions for each experimental treatment.

The FRAP was determined following the Benzie and Strain⁵⁸ protocol. A working FRAP solution was prepared by mixing 250 mL of acetate buffer (pH 3.6), 25 mL of a TPTZ (2,4,6-tripyridyl-s-triazine) solution (10 mM in 40 mM HCl), 25 mL of a hexahydrate iron (III) chloride solution (20 mM), and 30 mL of distilled water, at 37°C. A 0.1 mL sample of the plant extract was combined with 1 mL of the working FRAP solution and incubated in a water bath at 37°C for 4 min. Next, the mixture was cooled to room temperature, and absorbance was measured using a UV-1800 spectrophotometer at 593 nm. Measurements were repeated four times for each experimental treatment, and results were expressed in mg TEAC (trolox equivalent antioxidant capacity)·g⁻¹ DW.

The antioxidant capacity was determined using the ABTS radical cation method.⁵⁹ The ABTS radical cation was generated by incubating a mixture of 7 mM ABTS solution and 2.45 mM K₂S₂O₈ in a 1:0.5 ratio in the dark for 12 h. Immediately before measurement, the ABTS radical cation solution was diluted with phosphate buffer (PBS, pH 7.4) until an absorbance of 0.700 (±0.020) was achieved at a wavelength of 734 nm. Subsequently, 0.1 mL of the suitably diluted extract or standard solution was mixed with 3.9 mL of the ABTS working solution and kept in the dark at room temperature for 6 min. Next, absorbance was measured at 734 nm using the Shimadzu UV-1800 spectrophotometer. The rate of ABTS radical cation reduction for each sample was calculated using the equation: Inhibition rate = (A_ref - A_sample) / A_ref × 100% (A_ref and A_sample are the absorbance values for the standard sample and extract, respectively).EC₅₀ (expressed in mg·mL⁻¹), indicating the concentration required to inhibit 50% of the ABTS radical, was used to evaluate the activity of each sample. Each measurement was repeated four times for each experimental treatment.

All chemical reagents were provided by PHU Syl-Chem Sylwia Kunca, Bydgoszcz, Poland.

Phenotype and Genetic Variability Evaluation During Ex Vitro Plant Cultivation

The rooted plantlets were acclimatized to glasshouse conditions (42°9’6.768” N, 18°0’29.27195” E), in July under natural light conditions. They were planted in plastic multi-trays with a mixture of peat substrate (Hartman, Poznań, Poland) and perlite (Perlit, Šenov u Nového Jičína, Czech Republic) (2:1, v/v). Plants were sprayed with water and covered with perforated transparent foil. The acclimatization lasted two weeks. Next, the plants were planted into plastic pots (ø 29 cm, 10 L, five plants per pot) filled with peat substrate. Plants were cultivated vegetatively in 16 h long-day conditions until mid-October. Subsequently, cultivation was conducted in 10 h short-day conditions to induce generative development. Chrysanthemums were cultivated using the standard method, ie, to form plants with one stem and one inflorescence.

At the stage of full flowering, the plants were assessed by defining the color and shape of the inflorescences. The color of the inner and outer sides of ligulate flowers was established using the Royal Horticultural Society Colour Chart catalog (RHSCC).⁶⁰ Moreover, the analysis of anthocyanins content in ligulate flowers was performed. Total anthocyanins (with cyanidin-3-glucoside used as standard) were extracted according to Harborne⁶¹ protocol, in six repetitions for each experimental treatment, and expressed in mg·g⁻¹ fresh weight (FW). Stem height (cm) and number of leaves on the stem were also measured. Moreover, excised leaves (three per each plant taken form the lower, middle and upper part of the stem) were scanned (Epson Perfection V800 scanner, Suwa, Japan), and analyzed using imaging software WinFOLIA™ (Reagen Instruments, Quebec, Canada) to measure the leaf area (cm²), leaf perimeter (cm), leaf length (L) (cm), leaf width (W) (cm), and leaf aspect ratio (W/L).

Randomly Amplified Polymorphic DNA (RAPD)⁶² and Start Codon Targeted Polymorphism (SCoT)⁶³ marker systems were used for genetic analysis (8 RAPD and 8 SCoT primers; Genomed S.A., Warsaw, Poland). A total of 20 plants from each experimental object (control, CdS NPs, Co₃O₄ NPs, and Fe₃O₄@Co NPs) and six genotype/phenotype standard plants were included in each primer analysis. Total genomic DNA was extracted from fresh leaves using the Auto-Pure Mini device for automatic magnetic isolation of DNA and MagnifiQ™ 16 Plant DNA instant kit (A&A Biotechnology, Gdańsk, Poland). The DNA concentration was measured with the SmartSpec PlusTM spectrophotometer. PCR was performed using the BioRad C1000 Touch thermal cycler with a heated cover (Bio-Rad, Hercules, CA, USA) in the 25-µL reaction solution (PCR Master MixPlus, A&A Biotechnology, Gdańsk, Poland) as described in detail in our previous paper.²³ The PCR products were visualized on a UV transilluminator (GelDoc XR+ Gel Photodocumentation System with Image Lab 4.1 software, Bio-Rad, Hercules, CA, USA) after staining with SimplySafe™ (EURx, Sp. z o.o., Gdańsk, Poland). As a size reference, the Gene Ruler^TM Express DNA Ladder (Thermo Fisher Scientific, Waltham, MA, USA) was used. The banding patterns were recorded as binary matrices, where “1/0” indicates the presence/absence, respectively, of a given fragment. The numbers of monomorphic, polymorphic, and specific loci were counted.

Statistical Analysis

The experiment was set up in a completely randomized design. One-way analysis of variance (ANOVA) and post hoc Fisher’s least significant difference (LSD) test (p ≤ 0.05) were performed using Statistica 13.3 software (StatSoft Polska, Cracow, Poland) to determine the effects of nanoparticles treatment on the variability of analyzed traits. To identify the relationship between the distinguishing qualitative features under study, Spearman’s rank correlation coefficients were determined at p ≤ 0.05. The mean values and standard deviations (SD) were calculated. For data expressed as a percentage, before ANOVA analysis, the Freeman–Tukey double-arcsine transformation was performed. GenAlex 6.5 software⁶⁴ was used to run the principal component analysis (PCoA). Population groups were distinguished based on the analysis of molecular variance (AMOVA) estimates. Heterozygosity index (H), polymorphic information content (PIC), effective multiplex ratio (E), discriminating power (D), and resolving power (R) were investigated for every primer and marker system used.⁶⁵

Results

In vitro Plant Regeneration in Response to NP-Treatment

Internodes treated with the tested nanoparticles presented a varied ability to regenerate adventitious shoots (Figures 2 and 3). The application of Co₃O₄ NPs resulted in the lowest share of regenerating explants (46.67%), in contrast to other experimental objects, where the values of this parameter ranged from 70% to 80.35%. Statistically, the control and Fe₃O₄@Co NP-treated internodes regenerated most shoots (3.59 and 4.22, respectively), whereas the use of CdS NPs (2.6) and, in particular, Co₃O₄ NPs (2.22), resulted in a strong limitation of shoot formation.

Figure 2 Adventitious shoots regeneration on C. morifolium ‘Lilac Wonder’ internodes after treatment with various types of nanoparticles, the eighth week of culture; bar = 1 cm.

Figure 3 Efficiency of adventitious shoots regeneration on C. morifolium ‘Lilac Wonder’ internodes after treatment with various types of nanoparticles. Means or means ± Standard Deviation; (a–c) means on graphs followed by the same letter do not differ significantly at p ≤ 0.05 (Fisher’s test).

Plant Biochemical Activity in Response to NP-Treatment

In the tissues of shoots regenerated on internodes treated with the tested nanoparticles, a decrease in the relative flavonols content (0.050–0.053), and an increase in the anthocyanins content (0.042–0.046) was reported (Table 1). The content of flavonols and anthocyanins in the control was 0.062 and 0.033, respectively (Table 1). The shoots treated with CdS NPs had the lowest chlorophyll content (9.93), followed by those treated with Co₃O₄ NPs (10.28), while shoots treated with Fe₃O₄@Co NPs (10.45) and the control (10.70) showed the highest values. No significant differences were noted in the SPAD leaf greenness index value between experimental objects (32.79–33.68).

Table 1 Relative Content of Flavonols, Anthocyanins, Chlorophyll, and SPAD Index in C. morifolium ‘Lilac Wonder’ Adventitious Shoots After Treatment with Various Types of Nanoparticles

The application of Co₃O₄ NPs and Fe₃O₄@Co NPs resulted in a significant decrease of the value of minimal fluorescence (F₀) compared to the control (Table 2). Moreover, shoots treated with Co₃O₄ NPs were characterized by the lowest value of variable (F_V) and maximal (F_M) chlorophyll fluorescence. The control and CdS NP-treated shoots had the lowest F_V/F_M and F_V/F₀ ratios, followed by Co₃O₄ NP-treated shoots, while the Fe₃O₄@Co NP-treated shoots had the highest values of these parameters.

Table 2 Chlorophyll Fluorescence Indices in C. morifolium ‘Lilac Wonder’ Adventitious Shoots After Treatment with Various Types of Nanoparticles

A significant effect of the tested nanoparticles on the stimulation of polyphenols biosynthesis was demonstrated (Table 3). The TCP was the highest in CdS NP- and Co₃O₄ NP-treated shoots (46.63 and 47.82 mg·g⁻¹ DW, respectively). The content of polyphenols in Fe₃O₄@Co NP-treated shoots (43.04 mg·g⁻¹ DW) was also statistically higher than in control shoots (38.74 mg·g⁻¹ DW). As for the antioxidant capacity assay, the FRAP/ABTS values were significantly increased/decreased after NP applications, respectively, particularly for Co₃O₄ NP and Fe₃O₄@Co NP treatments.

Table 3 TCP, FRAP, and ABTS Analysis in C. morifolium ‘Lilac Wonder’ Adventitious Shoots After Treatment with Various Types of Nanoparticles

Statistical analysis also showed a significant, more than two- to four-fold increase in the APOX and GPOX activity in shoots regenerated on the CdS NP-treated internodes compared to the other experimental objects, which did not differ in this respect (Table 4). Interestingly, the SOD activity was the lowest in control shoots (14.89 U), and significantly higher in most NPs-treated shoots (18.09–18.36 U), except for Co₃O₄ NPs (17.67 U).

Table 4 APOX, GPOX, and SOD Activity (U) in C. morifolium ‘Lilac Wonder’ Adventitious Shoots After Treatment with Various Types of Nanoparticles

Plant Phenotype Variability in Response to NP-Treatment

During glasshouse cultivation, the share of flowering plants varied depending on the experimental treatment (Table 5, Figure 4A and B). A vast majority of standard (90%) and control plants (83.33%) formed inflorescences. As for CdS NPs and Co₃O₄ NPs treatments, the share of flowering plants was lower, and amounted to 73.15% and 76.84%, respectively. Phenotypic evaluation of inflorescences did not show qualitative variability of ligulate florets’ color. The inflorescences were pink; however, quantitative changes related to the different content of anthocyanins and color intensity appeared. The highest content of these pigments was found in ligulate flowers of plants treated with Fe₃O₄@Co NPs, and the lowest in the standard plants.

Table 5 Flowering and Inflorescence Characteristics of C. morifolium ‘Lilac Wonder’ Plants During Glasshouse Cultivation: Standard, Control, and Adventitious Shoots After Treatment with Various Types of Nanoparticles

Figure 4 Plants of C. morifolium ‘Lilac Wonder’ during the vegetative growth stage (A); plants at full flowering (B); variegated mutant found among Co3O4 NP-treated plants (C); leaf sample scan of the variegated mutant (D).

One individual (mutant) was phenotypically identified within Co₃O₄ NP-treated plants (Figure 4C and D). This plant developed leaves with variegation, ie, mosaic yellow-green discolorations on the leaf blades.

At the greenhouse cultivation stage, plants treated with Fe₃O₄@Co NPs (71.79 cm) and controls (69.28 cm) formed the longest stems. However, the stem length of plants treated with CdS NPs, and Co₃O₄ NPs did not differ statistically from the standard or control plants (Table 6). The Co₃O₄ NP-treated chrysanthemum developed more leaves (38.47) than standard plants (35.67). Control and NP-treated plants had a similar number of leaves. Leaf area of control and NP-treated plants (25.81–29.46 cm²), except for Co₃O₄ NPs (30.80 cm²), was lower than in the standard plants (33.95 cm²). Leaves of CdS NP-treated plants had the lowest perimeter, length, and width. The highest (0.64) and the lowest (0.58) leaf aspect ratios were found in the standard and Fe₃O₄@Co NP-treated plants, respectively.

Table 6 Biometric Parameters of C. morifolium ‘Lilac Wonder’ Plants During Glasshouse Cultivation: Standard, Control, and Adventitious Shoots After Treatment with Various Types of Nanoparticles

Plant Genotype Variability in Response to NP-Treatment

A total of 3707 and 6202 scorable bands were detected by eight RAPD and eight SCoT primers in 86 chrysanthemum plants, respectively (Table 7). Among the molecular marker systems tested, SCoTs generated more products and polymorphisms. Primer S-4 generated the highest number of bands (16 per sample), while primer S-2 produced seven amplicons per sample. Four primers detected 11 polymorphic loci (including five specific) in a total of six specimens (ie, 7.0% of all plants analyzed). The highest number of polymorphic plants (4) was obtained with the S-6 primer. As for the RAPD analysis, three primers detected three polymorphic and five specific bands in seven samples (8.1% of all analyzed plants). Primer R-4 generated the highest number of bands in this marker system (12 per sample), but R-2 was most effective in screening for variation. In contrast, primers R-5 and R-6 generated only one product (Table 7).

Table 7 Comparative Analysis of Molecular Products Obtained From C. morifolium ‘Lilac Wonder’ Plants Analyzed with RAPD and SCoT Marker Systems

Among the two marker systems used, higher mean H, PIC, E, D, and R values were reported with SCoTs (Table 8). When comparing individual primers, S-5 generated the highest heterozygosity index, discriminating power, and resolving power, while S-4 had the highest effective multiplex ratio. For PIC, R-2 was notable among RAPD primers (Table 8).

Table 8 Values of Heterozygosity Index (H), Polymorphic Information Content (PIC), Effective Multiplex Ratio (E), Marker Index (MI), Discriminating Power (D), and Resolving Power (R) of the Marker Systems Used in the Study

According to the PCoA analysis based on RAPD genotyping, three agglomerations could be distinguished (Figure 5). The largest group comprised 61 homogenous, monomorphic plants with no variation detected compared with the non-treated adventitious control or meristematic-derived standard. Polymorphic specimens were placed in two groups. The first included four plants treated with CdS NPs, one treated with Fe₃O₄@Co NPs, and one control. These plants had the same genotype. The population most distant from the predominant control and standard consisted of a single specimen treated with CdS NPs.

Figure 5 Graphs of principal coordinates analysis (PCoA) of C. morifolium ‘Lilac Wonder’ individuals after treatment with various types of nanoparticles; based on Randomly Amplified Polymorphic DNA (RAPD) and Start Codon Target Polymorphism (Scot) markers.

As for the PCoA analysis based on the SCoT marker system, the plants studied were grouped into four agglomerations. Due to the minor genetic variation detected in two specimens treated with Co₃O₄ NPs, these plants were placed in a single agglomeration with the control and other monomorphic plants (Figure 5). Three plants obtained from the Fe₃O₄@Co treatment were grouped into two distinct populations. The most altered genotype, treated with CdS NPs, was included in a separate group. This was the only plant considered polymorphic by both marker systems used. The remaining polymorphic genotypes did not overlap in the RAPD and SCoT analyses.

Following the AMOVA analysis, most of the genetic variation detected by the RAPD and SCoT markers in chrysanthemum ‘Lilac Wonder’ was considered intra-population origin. Nevertheless, 2–3% of the variation was caused by nanoparticles. Example band profiles are shown in Figure 6.

Figure 6 Example Randomly Amplified Polymorphic DNA (RAPD) band profiles of C. morifolium ‘Lilac Wonder’ received as a result of nanoparticles treatment (primer R-2). Outermost lanes are DNA base pair weight markers (wm), while inner lines represent control plants, plants treated with nanoparticles, and the standard. Arrows point to polymorphic genotypes.

Discussion

In the present study, the use of 75 mg·L⁻¹ CdS NPs and Co₃O₄ NPs reduced the efficiency of adventitious shoot regeneration compared to the control internodes. However, a statistically similar number of shoots was regenerated on the Fe₃O₄@Co NP-treated and control explants. In our previous experiments, the application of Ag NPs at concentrations of 20–100 mg·L⁻¹ also strongly limited the formation of adventitious shoots on internodes and leaf explants in C. morifolium, resembling the action of physical or chemical mutagens.^20,21 It is well known that nanoparticles alter biological architecture in many complex and not fully understood ways.⁴ Nanoparticles can lead to changes at the molecular level and consequently affect the overall plant physiology, growth, and development, including shoot and root formation.⁷ It is assumed that each type of nanoparticles imparts distinct effects on physiological processes via different signaling cascades. Possible pathways through which nanoparticles influence plant regeneration include modulation of hormonal levels, increased ROS production, and potential epigenetic modifications. Nanoparticles have been found to cause lipid membrane peroxidation in plant cells, leading to loss of membrane integrity, ultimately hampering the regeneration potential. They can also reduce cell survival by altering the expression of nucleic acids, causing DNA damage, disrupting chlorophyll synthesis, and causing electrolyte leakage.¹ Nanoparticles can cause phytotoxicity and negatively affect the overall metabolism of plants, but at the same time, their unique properties can also be used to improve plant functioning.⁸ The key factors determining the stimulation or inhibition of in vitro regeneration by nanoparticles seem to be their chemical composition, the concentration at which they are applied to the plants, and the plant genotype itself.

One of the main harmful effects of nanoparticles on plant cells is the induction of oxidative stress, which is a complex physiological, biochemical, and molecular phenomenon resulting from the overproduction and accumulation of ROS.⁸ Reactive oxygen species can cause serious damage to cellular components, including proteins, lipids, carbohydrates, and DNA. Although plants have innate antioxidant defense mechanisms to mitigate the effects of ROS, these mechanisms often fail, leading to oxidative damage, stunted growth, and reduced productivity.⁶⁶ Stress conditions can reduce the content of chlorophyll, leading to reduced plant growth and development.¹¹ Polyphenols, including flavonols, are non-enzymatic antioxidant molecules produced by plants mainly to mitigate stress and participate in detoxification reactions, acting as metal chelators in the scavenging of ROS by peroxidases.^67,68 SOD, APOX, and GPOX belong to the group of enzymatic antioxidants that catalyze the degradation of ROS. Changes in the content of non-enzymatic antioxidants and the activity of the above-mentioned enzymes are biological markers of oxidative stress.⁵¹ Results of the present study revealed that the tested nanoparticles also influenced the biochemical activity of regenerated plants. The tissues of NP-treated shoots accumulated less flavonols, but more total polyphenols and anthocyanins, and showed increased antioxidant activity compared to the control shoots. The highest activity of oxidative stress enzymes (APOX, GPOX, and SOD), and at the same time, the lowest chlorophyll content were noted in the shoots treated with CdS NPs. Similarly, tomato (Solanum lycopersicum L.) plants treated with multiwalled carbon nanotubes produced less flavonoids and more anthocyanins.⁶⁹ Manganese nanoparticles (Mn NPs) used at the concentrations of 125 and 250 mg·mL⁻¹ also increased the activity of SOD and peroxidases (POX), and reduced the chlorophyll content in wheat (Triticum aestivum L.), which indicated the occurrence of stress.⁷⁰

There are many different methods for determining antioxidant activity.^71,72 In the present study, two methods with varying mechanisms of action were used. The ABTS test with 2,2’-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) is based on free radical scavenging, whereas the FRAP test is based on the reduction of the Fe³⁺TPTZ salt to its blue form Fe²⁺TPTZ.⁷² The tested nanoparticles increased the antioxidant activity measured by the FRAP method, which confirmed the increase in the reducing capacity of the NP-treated plants. The strongest effect was obtained after the application of Co₃O₄ and Fe₃O₄@Co NPs, indicating their potential properties, which induce defense mechanisms in the plant. A similar effect was observed for the ABTS assay, where a loss of color and a decrease in absorbance could be explained by the interaction of antioxidants with ABTS radicals. The highest ABTS quenching activity was obtained for Co₃O₄ and Fe₃O₄@Co NPs. Other studies also indicate a significant effect of nanoparticles on the non-enzymatic antioxidant capacity of plants.^73,74 Wei et al⁷⁴ observed an increase in the FRAP parameter in S. lycopersicum plants cultivated in soil with SiO₂ NPs. Likewise, Thakur et al⁷³ observed the FRAP increase in wheat (Triticum aestivum L.) seedlings after the application of ZnO and TiO₂ NPs at two concentrations (1.5 mg·L⁻¹ and 10 mg·L⁻¹). Owji et al⁷⁵ reported a decrease in the FRAP antioxidant potential of fenugreek (Trigonella foenum-graceum L.) treated with 100 μg·mL⁻¹ Al₂O₃ NPs. The mechanism of nanoparticles’ action on the non-enzymatic antioxidant potential of plants can be complex and depends on the type, concentration, and the method of application.^76–78 Nanoparticles can be applied to plants in various ways (eg, through roots, leaves, seeds) and, due to the ability to penetrate the cells, they interact with many cellular structures.^79,80 These interactions can initiate several physiological responses, eg, increased production of ROS or modulation of metabolic pathways. Probably, one of the main nanoparticles effects is the modification of the phenylpropanoid pathway, in which phenolic compounds with antioxidant properties are synthesized.⁸¹ However, the action of nanoparticles is biphasic. In appropriate concentrations, they can act as a mild stress factor („priming”), activating plant defense mechanisms, and in excessive concentrations, they can cause oxidative stress, leading to lipid peroxidation, membrane and DNA damage, and exceeding compensatory capacities, which can lead to the inhibition of secondary metabolism and reduced antioxidant potential.⁸² Our results also indicated a high correlation between the content of polyphenols and the antioxidant potential of the tested samples (ABTS/TCP: r = 0.61, FRAP/TCP: r = 0.75). Similarly, Patil et al⁸³ analyzed the antioxidant potential of 22 chrysanthemum cultivars. They obtained a correlation at the level of r = 0.40, while Cao et al⁸⁴ observed r = 0.64 in studies with five chrysanthemum cultivars.

In the present study, the nanoparticles treatment also had a subsequent effect on the further ex vitro growth and development of chrysanthemum shoots, and may be important as a factor modifying the growth force and habit of plants. Chrysanthemums treated with Fe₃O₄@Co NPs formed the longest shoots, whereas the application of CdS NPs and Co₃O₄ NPs had the opposite effect. In addition, shoots regenerated in the presence of CdS NPs developed the smallest leaves. This effect can be correlated with the sulphur bioavailability and toxicity. The observed effects are also consistent with the biometric results obtained for maize treated with CdS NPs²⁵ or for chrysanthemum treated with Fe₃O_4.³⁶ Iron oxide nanoparticles also promoted shoot elongation of chrysanthemum plants developed in vitro from synthetic seeds.⁸⁵ Iron is an essential micronutrient involved in many metabolic pathways (eg, DNA synthesis, respiration, and photosynthesis) in plants. It is a component of many vital enzymes, stimulates chloroplast synthesis, and is responsible for maintaining the structure and function of chloroplasts.⁸⁶ The physiological importance of iron could cause the observed stimulation of growth in contrast to cadmium, sulphur, and cobalt. However, this element can be potentially toxic and can promote ROS formation.⁸⁶ The effects of various mutagens, including radiation and nanoparticles, can cause varying effects in stimulating plant growth. Low doses can stimulate, while higher doses can limit shoot growth.^17,21,87

Biochemical analysis of inflorescences revealed quantitative changes in anthocyanins content. The highest amounts of these pigments were found in ligulate flowers of Fe₃O₄@Co NP-treated plants, and the lowest in plants treated with Co₃O₄ NPs and in standard plants. Cobalt is known to influence several enzymes in plants, acting as a cofactor or signaling agent at low concentrations. Hence, the different anthocyanins accumulation can be closely correlated with the variation in Co dosage and bioavailability between the two NP-based formulations. Enhanced anthocyanin biosynthesis in Fe₃O₄@Co NP-treated chrysanthemums can be stimulated by low cobalt dosage, while Co₃O₄ NPs can deliver cobalt faster or in a less regulated manner. At the same time, the Fe ions availability in Fe₃O₄@Co NPs may reduce the oxidative stress and indirectly influence pigment accumulation, resulting in more intense inflorescence pigmentation. A decrease or increase in the content of anthocyanins, along with the associated change in the intensity of inflorescence color, is often observed in plants exposed to X or gamma irradiation, or Ag NPs, as reported in our previous studies on induced mutagenesis in chrysanthemum.^20,21,88 Qualitative changes resulting from the appearance or disappearance of carotenoid pigments are also very desirable and manifest themselves as a change in inflorescence color compared to the mother cultivar.^17,20,21 No such change was observed in the present study, suggesting that the tested nanoparticles do not have a high potential to induce phenotypically manifested mutations of inflorescence traits; however, one individual (mutant) identified among Co₃O₄ NP-treated plants had variegated leaves. Variegated plants usually have leaves with green and white (or yellow) sectors. Cells in the green sectors contain typical chloroplasts, whereas chloroplasts in the white (or yellow) sectors lack chlorophyll and/or carotenoid pigments and appear to be blocked at different stages of biogenesis. Variegations may be caused by nuclear, chloroplast, or mitochondrial gene mutations. Variegated sectors may have different genotypes (chimeras) or the same (mutant) genotype. In mutagenesis experiments, variegations arise at a low, but variable, frequency.^89,90 In our study, variegation occurred with a frequency of 1.05%. In studies with Arabidopsis thaliana, the frequency of variegation was 0.04–0.2% in genetic transformation experiments, and 10% in experiments with chemical and physical mutagens.⁹⁰ Bajpay and Dwivedi⁸⁷ obtained variegated chrysanthemums with a frequency of 10–22.66% after irradiation with gamma rays at very high doses of 100, 150, 200, and 250 Gy. Treatment with gamma rays was also a source of variegated chrysanthemum plants, with a frequency of 5%, in a breeding experiment performed by Mandal et al.⁹¹ According to these authors, the novel chlorophyll variegated chrysanthemums were attractive, and variegation gives additional beauty to the plant throughout the vegetative stage of growth. It can be concluded that the frequency of variegation is highly influenced by the type and dose of mutagen used.

The present study aimed to assess the applicability of nanoparticles as potential mutagens in chrysanthemum breeding by evaluating the genetic variation induced by different nanoparticle types using RAPD and SCoT markers. The results indicate that nanoparticles can induce genetic variation in chrysanthemum, although to a limited extent. SCoT markers generated more amplicons and polymorphisms compared to RAPDs, confirming previous findings that SCoTs are more efficient in detecting genetic variation in plants.^92,93 RAPD genotyping identified three polymorphic and five specific bands in seven samples, while SCoTs identified 11 polymorphic loci (including five specific) in six specimens, with only one specimen overlapping between the two marker systems. This indicates that various marker systems capture different aspects of the genetic changes induced by nanoparticles. The AMOVA results suggest that nanoparticles do contribute to mutation occurrence in chrysanthemum. This is consistent with studies showing that Ag NPs can induce genetic, biochemical, and phenotypic variation in chrysanthemum.²⁰ The present study, however, proves that different types of nanoparticles have divergent mutagenic potential. In the current analysis, only one specimen treated with CdS NPs was identified as polymorphic by both marker systems (five primers). This suggests that this particular nanoparticle type is more useful in breeding than Fe₃O₄@Co NPs or Co₃O₄ NPs. On the other hand, CdS NPs are still less effective than Ag NPs as the latter nanoparticles induced genetic variation that manifested in the form of a change in the inflorescence colour,^20,21 which was not reported in the present study. The varied effects of CdS, Co₃O₄, and Fe₃O₄@Co nanoparticles on mutation induction in chrysanthemum can be related to their differences in physicochemical properties (size, solubility, and surface chemistry), metal ion release, and redox activity that affect cellular uptake and bioavailability. CdS nanoparticles easily release Cd²⁺ ions, which are highly genotoxic and interact directly with DNA and proteins.⁹⁴ On the other hand, Co₃O₄, and Fe₃O₄@Co NPs act mostly as redox catalysts, generating ROS.⁹⁵ Das and Mohanty⁹⁶ evaluated the toxicological impacts of cadmium, cobalt, and copper in radish (Raphanus sativus L.), lettuce (Lactuca sativa L.), and wheat (Triticum aestivum L.). They found that Cd inhibited seed germination and seedling growth more potently than Co or Cu. Co was toxic at a higher concentration (50 µM), but Cd was even more deleterious at a lower (10 µM) dose. These findings corroborate our results and explain why only this particular treatment caused the most excessive mutation.

Traditional mutation breeding in chrysanthemum involves physical mutagens like ionizing radiation or chemical mutagens such as ethyl methanesulfonate (EMS).¹⁷ These methods can induce significant changes in flower color, shape, and size.⁹⁷ Previous studies suggest that gamma and X-rays have a 5–30% or 3–20% mutation efficiency, respectively, depending on dose and cultivar.^19,98 According to some researchers, chemical mutagens are even more effective.⁹⁹ While nanoparticles offer a potentially easier method for inducing mutations, as they can be added directly to the culture medium without specialized equipment, the extent of genetic variation caused by nanoparticles in this study appears to be more limited compared to conventional techniques. Nonetheless, a yield of 7–8.1% mutated plants confirms their utility in breeding. Begum and Dasgupta⁹⁹ and Shahwar et al¹⁰⁰ reported that the lowest dose of γ-rays (200 Gy) and lower concentration of EMS (0.1%), caffeine (0.1%), Pb(NO₃)₂ (20 mg·L⁻¹), and Cd(NO₃)₂ (20 mg·L⁻¹) showed the highest mutagenic efficiency in all genotypes studied. Perhaps also in chrysanthemum, modifying the concentration of nanoparticles could improve the results.

In summary, the present study suggests that although nanoparticles can induce biochemical, genetic, and phenotypic changes, careful selection of marker systems and screening methods is crucial for this variation identification. Further research is needed to optimize nanoparticle treatments and explore their potential in combination with other breeding techniques to increase genetic diversity and create new chrysanthemum cultivars.

Conclusion

Our results contribute to a better understanding of the multifaceted effects of various nanoparticles on plants in vitro and ex vitro at the biochemical, genetic, and phenotypic levels. They are of scientific and practical importance for the development of modern horticulture. We elaborated an innovative approach in plant breeding, proposing the use of nanoparticles as a source of variability induction. Ease of use and lower toxicity compared to conventional mutagens make nanoparticles a valuable addition to breeding methods for the semiconducting properties of the demonstrated particles, which, besides indirect oxidative stress generation by the disruption of metabolic and/or electron transfer processes, can catalyze ROS, directly affecting the plant. The results are certainly interesting for scientists and chrysanthemum breeders.

The use of CdS NPs and Co₃O₄ NPs limited the efficiency of chrysanthemum adventitious organogenesis. Nanoparticles also modulated the biochemical activity of regenerating shoots. The induction of oxidative stress was most evident for CdS NPs application. Nanoparticles also affected the ex-vitro growth of chrysanthemums by modifying the habit of plants and leaf architecture. One mutant with variegated leaves was phenotypically identified within Co₃O₄ NP-treated plants. The use of nanoparticles also induced genetic variation, with CdS NPs causing the most distinct polymorphic changes, as confirmed by both RAPD and SCoT markers. While Fe₃O₄@Co NPs and Co₃O₄ NPs also generated polymorphic genotypes, their effects were less evident, suggesting that nanoparticles vary in mutagenic potential, but could serve as a novel tool for plant breeding. The proposed use of nanoparticles for variability induction is also an innovative approach in plant biotechnology.

The plant in vitro culture system has tremendous potential in breeding, which is still not fully exploited. The addition of nanoparticles to the culture medium can be a relatively cost-effective, easy, and effective way to induce variation for breeding. To improve the frequency of observed mutations in future studies, it is recommended to use the tested nanoparticles in even higher concentrations, especially 100, 150, and 200 mg·L⁻¹, as well as to focus the research on testing other types of nanoparticles containing cobalt and cadmium. The genotoxic effect of nanoparticles should also be experimentally tested in other plant genotypes.