Magnesium Hydroxide Microparticle Treatment Potently Inhibits Venturia oleaginea Pathogenesis on Olives

Abstract

Olive trees worldwide suffer from a number of devastating fungal diseases that affect production. One such serious disease is olive leaf spot caused by Venturia oleaginea. Recently, we applied magnesium hydroxide porous micron-scale particles (PMPs) on tomatoes and observed potent antimicrobial activity, reducing the fungal load of the treated phyllosphere. To assess the effectiveness of the compound on olive fungal disease, we applied it for two consecutive seasons. One particular olive tree exhibited extreme manifestations of fungal disease and was destined for removal. A single application of Mg(OH)₂ PMP reversed all symptoms of the disease and eliminated the Venturia pathogen, curing the tree of disease. Venturia oleaginea appears to be exceptionally susceptible to treatment compared to other species in the fungal leaf community. The beneficial fungus Aureobasidium pullulans increased in relative abundance in all the sprayed trees. No toxicity and leaf loss were observed, and the compound retention exceeded 47 days. All trees sprayed showed drastic reductions in the total fungal load and compared favorably to the commercial copper compound. Spraying induced a moderate expression of key indicator genes associated with stress responses. No leaf chlorosis or shedding were observed. Overall, Mg(OH)₂ PMP treatment appears to be a highly promising tool for combating plant fungal disease.

1. Introduction

The olive tree is an iconic species in Mediterranean cultural history. Its overall presence over thousands of years in this territory and its multiple applications in everyday life have made this species an economic pillar and cornerstone of Mediterranean agriculture [1]. Over 750 million olive trees are cultivated worldwide, 95% of which are in the Mediterranean region. Several pests and diseases affect olive trees, threatening production. Changes towards intensive agricultural practices have exerted additional stress and rendered the crops more susceptible to disease than previously. The most impactful fungal diseases affecting olive trees in Mediterranean olive-growing countries are anthracnose, cercosporiosis, peacock spot, and Dalmatian disease caused by Colletotrichum spp., Pseudocercospora cladosporioides, Venturia oleaginea, and Botryosphaeria dothidea, respectively [2].

Olive leaf spot (OLS) or peacock spot disease is caused by the ascomycetous fungus Venturia oleaginea (also named Fusicladium oleagineum and Spilocea oleagina in the anamorph). The disease is widespread in the Mediterranean region and all olive-growing areas worldwide. The fungus thrives on mild to low temperatures and free moisture to germinate [3]. Typical symptoms of the disease are circular spots colored in uniform brown to green on the upper surface of the leaf. The defoliation caused by the fungus has a direct negative impact on the length of inflorescences, on fruit set, and, consequently, on plant growth and productivity. An average 20–30% yield loss was estimated in areas where OLS is recurrent [3]. Copper compound spraying is mostly used to control OLS wherever the olive is cultivated. However, according to regulation 2009/1107 of the European Commission, copper compounds are included in the list of substances that are candidates for substitution, as they are persistent in soil and toxic to the environment.

To reduce the environmental impact of agricultural production and minimize damage to the environment, it is crucial to optimize the use of fertilizers and pesticides, as well as transition to renewable energy sources instead of fossil fuels [4]. Recently, nanoparticles such as titanium dioxide (TiO₂), zinc oxide (ZnO), copper oxide (CuO), and magnesium oxide/hydroxide (MgO/Mg(OH)₂) have been shown to possess antibacterial properties and are being explored for use in biomedicine and agriculture [5,6,7,8,9,10]. Thus, there is an urgent need to identify and apply effective, non-persistent, and non bioaccumulative alternatives, and NPs are a class of such materials that can offer such [7]. NPs are organic, inorganic, or hybrid particles with at least one of their dimensions ranging from 1 to 100 nm (i.e., at the nanoscale). Nanoparticles are advantageous due to their large surface area, easy attachment, and fast mass transfer, allowing for the effective delivery of agrochemicals [11]. There are differing hypotheses regarding the mode of action of NPs, and these are not fully resolved. Increased stability of agrochemicals, antifungal and antibacterial properties possibly through the disruption of membrane stabilization and controlled ROS generation, pesticidal properties, and induction of stress tolerance are some of the NP modes of action proposed [12]. However, the small size of nanoparticles can cause them to permeate physiological barriers, leading to toxic reactions in living organisms. The level of toxicity of nanoparticles is dependent on factors such as the composition, size, surface functionality, crystallinity, and aggregation. Nanoparticles enter the human body through the lung, intestinal tract, or skin, and can be toxic to the brain and cause lung inflammation and cardiac problems [13]. In fact, certain nanoparticles have been found to cause permanent cell damage through organ injury and oxidative stress, due to their size and chemical composition [14].

The application of hydrated magnesium oxide (MgO) as a plant protection product significantly reduced the presence of Acinetobacter sp., Methylobacterium sp., and other Gram-negative bacteria in leaf bacterial communities, while leaving Gram-positive extremophiles and cyanobacteria, which were previously undetectable in the leaf bacterial community profile [15]. The fungal composition was not affected by PMP treatments, but the quantification of the fungal load on treated leaves showed a significant reduction in the density of these fungal pathogens, indicating a fungicidal or fungistatic effect. The expression analysis of indicator genes in PMP-sprayed plants revealed the induction of phenylalanine ammonia lyase (PAL), though at lower levels than those produced naturally by a plant under an actual pathogenic attack, consistent with previous findings on Mg-NP induction of the salicylic acid (SA) pathway [16]. PAL catalyzes the first step in the phenylpropanoid pathway and is involved in the biosynthesis of SA, an essential signal involved in plant systemic resistance [17]. PAL gene expression responds to various environmental stresses, including pathogen infection, wounding, nutrient depletion, UV irradiation, and extreme temperatures [18].

Magnesium plays a vital role as a crucial nutrient for all living organisms. In plants, the majority of magnesium ions (Mg²⁺) are linked with proteins, and within the cytosol, Mg²⁺ is typically found at concentrations around 0.4 mM. About 15–20% of magnesium is connected to chlorophyll pigments, primarily serving as a coenzyme for enzymes engaged in photosynthetic carbon fixation and metabolic processes [6,19]. When plants experience magnesium deficiency, it can lead to stunted root growth, smaller roots, and the development of necrotic spots on leaves. These effects are a result of disruptions in carbon metabolism and photosynthesis caused by the lack of sufficient magnesium in the plant system [20].

Based on our observations of the Mg(OH)₂ PMP effects on the tomato leaf phyllosphere, we aimed to apply the compound as a potential treatment for fungal disease in olives. To this end, we treated several OLS-infected olive trees, among which, one exhibited severe manifestations of the disease. A single application successfully reversed the symptoms of the disease, allowing the tree to fully recover and remain healthy in the following years. Analysis of the fungal microbiome confirmed the significant levels of the Venturia pathogen and the high amounts of fungal load on leaves, which were drastically reduced upon treatment. Fungal load reduction was observed in multiple applications in two consecutive years. The Venturia pathogen appeared to be especially susceptible to treatments compared to other fungal species, while the beneficial fungus Aureobasidium pullulans increased in relative abundance. Mg(OH)₂ PMP has no phytotoxicity and upon eventual absorption offers a source of magnesium to plant cells. Thus, it can offer attractive environmentally friendly alternatives in plant protection.

2. Materials and Methods

2.1. Materials

The Mg(OH)₂ PMP aqueous suspension was prepared by the addition of a high-porosity MgO PMP (Calix Ltd., Pymble NSW 2073, Australia, d_mean = 5.4 μm, BET surface area of 234 m²/g) in hot (90 °C) distilled water under continuous stirring/heating on suitable plates, for approximately 20 h. The suspensions were handled and transported in concentrated form (60 wt% solids) and were diluted just prior to spraying in the field at 2.3 wt%. Details on the particles, their textural characteristics, production method, and all relevant procedures from the raw source to the final suspensions are described in [15,21,22]. As explained and experimentally verified in [15], the resulting suspension had a Mg(OH)₂ concentration of approximately 2.9 wt% before its foliar application. The Mg(OH)₂ PMPs of the final suspension, as observed by TEM, are shown in the Results in Section 3.1.

Kocide^® 2000 35 WG (K+N Efthymiadis S.A., Industrial Area of Thessaloniki, 57022 Sindos, Greece) contains copper hydroxide with metallic copper at 35% w/w. In our applications, we followed the manufacturer’s recommendations for combating Cycloconium (OLS) at a dose of 340 g/100 L of water.

2.2. Plant Spraying and Sample Collection

In an effort to reveal the impact of active materials on olive trees’ pathogens, a small olive grove of the Chondroelia Chalkidiki variety was chosen to be sprayed in the Municipality of Thermi, near Thessaloniki, Greece. Most of the trees were infected with OLS, which was at various stages of progression (healthy to mild), with one of them (T1) exhibiting severe manifestations of OLS.

To assess disease severity in each tree, leaves from tree branches collected randomly were visually assessed for the presence of fungal infection and olive leaf spots. The percentage of affected leaves was enumerated and assigned to a scale: healthy 1–10%, mild 10–20%, moderate 20–40%, significant 40–60%, and severe 60–100%. In the case of the single severely affected tree, the percentage of affected leaves amounted to approximately 75%.

In May 2022, the trees were sprayed with the abovementioned Mg(OH)₂ PMP suspension. The application was such in order to achieve full coverage of the leaves. Leaf samples were collected just before and 18 days after spraying from T1 and two neighboring trees with healthy appearance (T2 and T3). In April 2023, and after the revival of T1 health status, the olive grove again showed signs of OLS infection. This year, the trees were sprayed again with Mg(OH)₂ PMP, and triplicate samples were collected from T1, T2, and T3 just before and 10 days after spraying. Samples from an untreated tree (T10) were also collected in the second year. Εach biological replicate comprised several leaves collected from a different branch, in order to obtain a representative picture of the whole tree.

To validate the results and compare the impact of Mg(OH)₂ PMP with that of a conventional copper fungicide, another experiment was conducted in an olive grove in the Chalkidiki area, with Chondroelia variety trees (40 km east of Thermi, Thessaloniki). Three groups were assigned for this experiment, each one comprising a row of three trees. The first group was sprayed with Mg(OH)₂ PMP, the second one with a common copper fungicide (Kocide^® 2000 35 WG), and the last one remained unsprayed. Each sample consisted of leaves from all three trees of the group. The procedure was repeated twice, in order to obtain duplicate samples (a complete list of samples is included in Table S3 in the Supplementary material document).

Different treatments and storage procedures for the leaves of each sample were followed, depending on the downstream applications that would be carried out. For the gene expression analysis, RNA was extracted from a single leaf from each sample. The leaves were placed in 1.5 mL tubes, immediately frozen in liquid nitrogen, and finally stored at −80 °C until extraction. On the other hand, approximately 8 to 10 leaves from each sample were used for microbial DNA extraction, which was subsequently used for metagenomics and fungal load analysis. The leaves were incubated in isotonic phosphate-buffered saline (PBS) for 1 h at room temperature in an orbital shaker, and after centrifugation, the supernatant was discarded, and the remaining material containing the microbial cells was collected. The procedure of preparing the samples for microbial DNA extraction is more analytically described in our previous study [15].

2.3. DNA Extraction, Library Construction, and Sequencing

The ZymoBIOMICS DNA Miniprep Kit (ZYMO RESEARCH, Irvine, CA, USA) was used for microbial DNA extraction according to the manufacturer’s instructions and after the sample preparation previously described. DNA concentration was measured on a Qubit 4.0 Fluorometer using the Qubit^® dsDNA HS assay kit (Invitrogen, Carlsbad, CA, USA). Bacterial communities were investigated by sequencing the V3–V4 hypervariable region of the 16S rRNA gene while fungal communities were identified by sequencing the V7–V8 hypervariable region of the 18S rRNA gene as well as the ITS1 region. Illumina’s (Illumina Inc., San Diego, CA, USA) 16S Metagenomic Sequencing Library Preparation (15044223 B) protocol was used for library construction. For the amplification of the V3–V4 region of the 16S rRNA gene, gene-specific primers based on Klindworth et al. [23] were used; for the amplification of the V7–V8 region of the 18S rRNA gene, universal primers FR1 and FF390 were selected from Chemidlin et al. [24]; and finally, for the amplification of the ITS region, primers BITS and B58S3 were selected based on Bokulich and Mills [25]. All primers were modified by adding an Illumina (Illumina Inc., San Diego, CA, USA) overhang adapter nucleotide sequence at the 5′ end. The sequences of the primers are listed in Table S1. For the purification of PCR products and libraries from unincorporated primers and primer–dimer species Agencourt AMPure XP magnetic beads (Beckman Coulter—Life Sciences, Indianapolis, IN, USA) were used. All libraries were quantified with fluorometric quantification using the Qubit^® dsDNA BR assay kit and their size was evaluated on a Fragment Analyzer 5200 system (Agilent Technologies Inc., Santa Clara, CA, USA) using the DNF-474–0500 kit. Molarity of libraries was calculated by a quantitative PCR (qPCR), carried out on a Rotor-Gene Q thermocycler (Qiagen, Hilden, Germany) with the KAPA Library Quantification kit for Illumina sequencing platforms (Kapa Biosystems, Woburn, MA, USA). Libraries were sequenced on a MiSeq platform using the MiSeq^® reagent kit v3 (2 × 300 cycles) (Illumina, San Diego, CA, USA). Sequencing data have been deposited to NCBI under BioProject ID PRJNA1029802.

2.4. RNA Extraction and RT-qPCR

Total RNA was purified from the leaves of three different branches (n = 3 per group) of the olive tree exhibiting the most severe infection with OLS (T1) before spraying and eighteen days after spraying, for growing season 2022, and ten days after spraying, for growing season 2023, using the Spectrum Plant Total RNA kit (Sigma-Aldrich, St. Louis, MO, USA) and according to the manufacturer’s instructions. To assess the effects of Mg(OH)₂ PMP on plant stress responses, specific stress-related genes were selected for analysis. The selected genes are important for defense mechanisms and stress responses.

cDNA synthesis and RT-qPCR reactions were performed using the Luna Universal One-Step RT-qPCR kit (New England Biolabs, Ipswich, MA, USA) in an AriaMx Real-Time PCR System (Agilent Technologies Inc., Santa Clara, CA, USA). Primer sequences are listed in Table S2. The correct size of amplified RT-qPCR products was verified by electrophoresis in a 2% agarose gel. RT-qPCR reactions were performed in triplicates. Expression of target genes was normalized to the relevant housekeeping EF1a gene. For each gene, a 2^-ΔCt value was measured and used for the calculation of the mean value and standard deviation (STDEV) per group. Data visualization and statistical analysis were performed using a Student’s t-test in the GraphPad Prism software v8.0 (San Diego, CA, USA).

2.5. Bioinformatics and Data Analysis

FastQC was employed for quality control checks to ensure that the raw data were of sufficiently high quality and there were no problems or biases in the data.

The analysis of the bacterial communities was conducted using the Quantitative Insights Into Microbial Ecology 2 (QIIME2-v.2020.2) pipeline [26]. The imported data to QIIME2 were paired-end reads. Denoising, dereplication, chimera filtering, and joining of the sequences were performed using the DADA2 plugin [27]. The output of DADA2 was an Amplicon Sequence Variant (ASV) table [28]. For taxonomic classification, sequences were aligned against the SILVA138 database using a trained classifier, at 99% similarity [29]. Archaeal, chloroplastic, mitochondrial, eukaryotic, and unassigned sequences were filtered out of the dataset.

The 18S rRNA sequencing data obtained for the eukaryotic communities were analyzed with the same pipeline as the 16S rRNA data of the bacterial communities, with the exception of the taxonomic filtering step, during which archaeal, bacterial, chloroplastic, mitochondrial, and Magnoliophyte ASVs were removed from the data.

Analysis of the ITS sequencing data obtained for the eukaryotic communities was conducted in R (version 4.2.2) [30]. The sequences were trimmed using cutadapt [31]. Denoising, dereplication, merging of paired-end reads, and chimera filtering of the sequences were realized using the dada2 package [32]. The output of the pipeline was an ASV table. ASV sequences were aligned against the UNITE database [33] at 99% similarity. Taxonomic filtering was performed to filter out ASVs belonging to archaea, bacteria, chloroplasts, mitochondria, and Magnoliophyta as well as unassigned sequences.

The ASV tables were imported in R (version 4.2.2) to further process and visualize results. All plots were visualized by combining functions provided by the phyloseq [34], ggplot2 [35], and gplots [36] packages. All barplots were normalized to 100% as abundance estimations within each sample and percentages do not accurately represent the true biomass fraction in each sample. Principal component analysis (PCA) was applied to all samples in the various conditions. PCA visualization was realized with plotly and ggfortify packages [37,38].

2.6. Quantification of Fungal Load

To quantify the fungal load of olive leaves [39] before and after treatment with Mg(OH)₂ PMP, microbial DNA was isolated as described above and used in a quantitative PCR (qPCR) for the amplification of the whole eukaryotic internal transcribed spacer (ITS) region ITS1-5.8S-ITS2 with the ITS primers ITS4 and ITS5.

For the construction of the calibration curve for quantitative analysis, a fragment of 575 bases was amplified from microbial DNA extracted from symptomatic olive leaves, using the ITS4 and ITS5 primers. The amplicon was introduced into pCRII-TOPO (Invitrogen, Carlsbad, CA, USA) by TOPO TA cloning and verified by Sanger Sequencing. Plasmid DNA was extracted with the NucleoSpin Plasmid, Mini kit for plasmid DNA (MACHEREY-NAGEL, Düren, Germany) and quantified with Qubit^® dsDNA BR assay kit (Invitrogen, Carlsbad, CA, USA). Serial dilutions of the plasmid DNA, ranging from 1 ng to 0.001 pg, were used to construct an initial calibration curve. For the samples used in the analysis, since they all fell within the range of 100 pg to 0.01 pg, we chose five standards (10-fold dilutions between 100 pg and 0.1 pg) amplified in triplicate to generate the final calibration curve.

All reactions were performed in triplicates, using the KAPA HiFi HotStart PCR Kit (Kapa Biosystems, Woburn, MA, USA) and Syto9 fluorescent dye, on the StepOnePlus™ Real-Time PCR System (Applied Biosystems™, Waltham, MA, USA). The Applied Biosystems StepOnePlus™ Real-Time PCR Software v2.3 was used for the analysis and the quantification of the samples based on the calibration curve. The number of copies of target DNA was calculated by DNA quantities considering that the average weight of a base pair is 650 Da bp (Number of copies of DNA template per µL = ng × Avogadro’s number/length of template (bp) × conversion factor to ng × average weight of a base pair (Da)). Finally, DNA copies were calculated per ng of microbial DNA of each sample, considering that each reaction contained 2 μL of microbial DNA. The concentration of the microbial DNA extracted from the samples was previously quantified with the Qubit^® dsDNA HS assay kit (Invitrogen, Carlsbad, CA, USA). Data visualization and statistical analysis were performed using a Student’s t-test in the GraphPad Prism software v8.0 (San Diego, CA, USA).

2.7. Scanning Electron Microscopy (SEM)

To qualitatively assess both the coating of Mg(OH)₂ PMP and the fungus population on the sprayed leaves, Scanning Electron Microscopy (SEM) was employed. Additionally, with the aid of Energy-Dispersive X-ray Spectroscopy (EDS), aggregate average values of C, O, and Mg elements were determined at several spots of the observed samples. The associated equipment was a JEOL JSM-IT500 scanning electron microscope operated at high vacuum mode and at 20 kV operating voltage coupled with an EDS system by Oxford Instruments Inca X-Act (Oxford Instruments, High Wycombe, UK).

3. Results

3.1. Application of Mg(OH)₂ PMP on Olives

Figure 1 shows the Mg(OH)₂ particles of the suspension applied to the olive trees studied in this work. The high BET surface area of the initial particles is due to the much smaller particles (Figure 1B) bound in larger aggregates (Figure 1A). On the 12 May 2022, we located a severely infested tree (75% affected leaves) in an olive grove exhibiting typical symptoms of olive leaf spot disease. The extent of damage from the fungal disease had progressed to the young leaves and stems, and parts of the bark had taken a black appearance. To test the previously observed antifungal activity of the Mg(OH)₂ PMP, we sprayed the sick Tree 1 (T1) along with two apparently healthy neighboring trees (T2, T3). Leaf samples were collected just prior to spraying (day 0) and at regular intervals after spraying for analysis and microscopic observation.

Eighteen days later, observation of T1 showed dramatic changes in appearance. All newly grown leaves were free of OLS and the leaves with moderate infection had cleared. In addition, the stems and bark where the PMP had been attached were cleared of the dark coloration of the fungal growth (Figure 2).

In order to visualize the fungal infection and the effect of the treatment on the leaves, a Scanning Electron Microscopy (SEM) study was carried out to study the morphological changes on the leaf surface and to compare the structural differences between the infected leaves before treatment and the ones after treatment.

A significant number of leaves at the same growth stage were selected per sampling day to ensure a statistically significant number of samples for each group for reproducibility and reliability and stored at −20 °C. A few days before the SEM measurement, indicative leaves were selected from the entire batch and allowed to unfreeze under low-humidity conditions in a desiccator. The dried leaves were mounted on SEM stubs using carbon tape and then coated with a thin layer of gold as conductive material, to prevent charging under the electron beam.

In Figure 3, it can be seen that the surface area of the untreated leaf is covered with a high density of conidia, whereas on the treated surface, there is a noticeable reduction. The leaf is partially covered with a layer of the Mg(OH)₂ PMP.

To obtain information on the presence of the Mg(OH)₂ PMPs over time on the leaf surface, we conducted elemental analysis by Energy-Dispersive X-ray Spectroscopy (EDS), which was coupled to the SEM apparatus. The EDS analysis was applied to several spots (equivalent area per sample equal to ~1 × 7.3 cm) from two representative leaves collected from proximal areas of the tree at days 0 (i.e., prior to spraying), 18, 32, and 47. The average aggregate values from the EDS spectra are shown in Figure 4. In all cases and as expected, EDS identified high concentrations of carbon and oxygen (>60 wt% and 25 wt%, respectively). Regarding inorganic elements, Mg, Si, Al, K, and Ca were detectable. Evidently, the Mg-based layer on the leaf surface was clearly detectable until even the 47th day after spraying. Naturally, the average content of Mg decreased with time due to the gradual removal of the PMP layer.

3.2. Olive Leaf Epiphytic Microbiome Analysis

Using 18S rRNA gene and ITS as taxonomic identification markers, amplicon metabarcoding analysis was carried out to assess the leaf fungal microbiome community after Mg(OH)₂ PMP spraying. The results of the 18S rRNA and ITS analysis for the leaf fungal microbiome community before and after Mg(OH)₂ PMP spraying are shown in Figure 5.

The sequencing of the 18S rRNA amplicon resulted in 64,494, 67,007, 96,722, 70,623, 66,991, 89,532, 81,836, and 65,187 raw reads for unsprayed T1 (1, 2, 3), sprayed T1 (1, 2, 3), T2, and T3, respectively. After quality and chimera filtering, 37.09%, 38.63%, 36.77%, 33.76%, 37.36%, 34.32%, 43.58%, and 36.19% of sequences were removed from the samples. In the end, 62.91%, 61.37%, 63.23%, 66.24%, 62.64%, 65.68%, 56.42%, and 63.81% of the raw reads of the respective samples were assigned to ASVs and remained for taxonomic classification. In the unsprayed T1, Aureobasidium and a non-assigned genus (red bar), of the Dothideomycetes class, were the two main genera that represented, respectively, 26% and 34% of the total leaf fungal microbiome for the unsprayed T1.1, 24% and 36% for unsprayed T1.2, and 46% and 15% for unsprayed T1.3 samples. Cladosporium, a common fungus responsible for indoor and outdoor mold, represented 8% of the total fungal community for unsprayed T1.1 and T1.2, and 5% for unsprayed T1.3. An uncultured genus, part of the Arthoniomycetes (light blue-green bar), represented, respectively, 7%, 8%, and 10% of the total leaf fungal microbiome for Unsprayed T1 (1, 2, and 3).

In the treated T1 samples, the results showed a significant decrease in the abundance of the non-assigned genus (red) part of the Dothideomycetes class for the sprayed T1 samples (1, 2, and 3) (from 34 to 1%, from 36 to 1%, and from 15 to 1% abundance, respectively). Cladosporium genus increased significantly to cover the void for the three sprayed samples (from 8 to 20% for sprayed T1.1, from 8 to 34% for sprayed T1.2, and from 5 to 29% for sprayed T1.3). The parasite and saprophyte fungus Alternaria mildly increased for sprayed T1 (1, 2, and 3) (from 2 to 5%, from 2 to 4%, and from 2 to 10% abundance, respectively). The results (Figure 5) also showed the similarity of the fungal microbiome composition between sprayed T1 and the two controls, T2 and T3.

To better differentiate fungal diversity at the genus level and improve the microbiome analysis accuracy, the ITS marker was also used. The sequencing of the ITS gene resulted in 63,342, 63,671, 60,367, 67,913, 59,270, 55,612, 63,109, and 68,320 raw reads for unsprayed T1 (1, 2, 3), sprayed T1 (1, 2, 3), T2, and T3, respectively. After quality and chimera filtering, 7.25%, 6.86%, 6.76%, 5.85%, 5.92%, 5.87%, 6.29%, and 6.30% of sequences were removed from the data. In the end, 92.75%, 93.15%, 93.24%, 94.14%, 94.08%, 94.13%, 93.71%, and 93.70% of reads remained for ASV assignment and taxonomic classification.

The yeast-like fungus Aureobasidium and the fungal pathogen Venturia, causative agents of olive leaf spot disease, dominated all unsprayed T1 samples (respectively, 28%, 27%, and 35% abundance and 26%, 25%, and 20% for unsprayed T1 (1, 2, and 3)), thus assigning the genus Venturia to the previously unidentified, based on 18S rRNA analysis, species of the Dothideomycetes class. The previously identified Cladosporium and Alternaria genera represented, respectively, 7%, 8%, and 5% and 3%, 4%, and 2% of the total fungal community in the unsprayed T1 (1, 2, and 3) samples, respectively. The foliar plant pathogenic fungus Mycosphaerella corresponded to 4%, 3%, and 4% of the total abundance in the unsprayed T1 (1), (2), and (3) samples, respectively. The genus Buckleyzyma, a member of the Basidiomycota phylum, comprised 6%, 6%, and 12% abundance for unsprayed T1 (1, 2, and 3), respectively.

In the Mg(OH)₂ PMP-treated T1 sample, a dramatic decrease in Venturia sp. was observed (from 26 to 2%, from 25 to 1%, and from 20 to 1% for sprayed T1 (1, 2, and 3), respectively) (Figure 5), confirming the 18S rRNA results. The species Venturia oleaginea is the causal agent of olive peacock spot disease. This was the only genus with a major abundance decrease. Mycosphaerella, Alternaria, and Cladosporium doubled (tripled for sprayed T1.3) their abundance percentage to cover the void. Aureobasidium still dominated and lightly increased for sprayed T1.1 and T1.2 (from 28 to 37% and from 27 to 28% abundance, respectively) and decreased for sprayed T1.3 (from 35 to 30% abundance). Buckleyzyma increased from 6 to 8% abundance for sprayed T1.1 and T1.2 and decreased from 12 to 6% abundance for sprayed T1.3. In the controls T2 and T3, Alternaria, Aureobasidium, and Mycosphaerella dominated the microbiome with, respectively, 20%, 22%, and 19% abundance. Another fungal plant pathogen, Stemphylium, was present at 6% abundance in the control unsprayed trees (T2 and T3).

To assess the bacterial leaf community in our leaf samples, the 16S rRNA gene was used as a taxonomic identification marker and amplicon metabarcoding analysis was carried out. The results from the 16S rRNA gene analysis for the leaf bacterial community before and after Mg(OH)₂ PMP spraying are shown in Figure 6. The sequencing of the 16S rRNA gene resulted in 61,507, 66,141, 42,380, 49,195, 56,530, 40,235, 70,551, and 39,837 raw reads for unsprayed T1 (1, 2, and 3), sprayed T1 (1, 2, and 3), control T2, and control T3, respectively. After quality and chimera filtering, 30.35%, 30.36%, 27.97%, 35.62%, 31.92%, 25.01%, 32.64%, and 25.74% of sequences were removed from the data. In the end, 69.65%, 69.64%, 72.03%, 64.38%, 68.08%, 74.99%, 67.36%, and 74.26% of sequences were assigned to ASVs and proceeded for taxonomic classification.

A total of 285 unique genera were identified in the three unsprayed trees, but 15 of them comprised more than 95% of the bacterial community. At the genus level, Sphingomonas and Hymenobacter, which are Gram-negative soil bacteria with low virulence, dominated (on average 35% and 30% abundance, respectively) the unsprayed trees’ bacterial communities. In a smaller proportion, Kineococcus (10% abundance for unsprayed T1.1 and Τ1.2, 4% for unsprayed T1.3) represented Gram-positive cocci, and Methylobacterium (10% abundance for unsprayed Τ1.1, 8% for unsprayed Τ1.2, and 1% for unsprayed Τ1.3) and others represented Gram-negative soil bacteria. In unsprayed T1.3, Pseudomonas and Hydrotalea genera represented, respectively, 23% and 16% of the total abundance. Also, Bradyrhizodium genus was found in unsprayed Τ1.3 at 8% abundance. The Mg(OH)₂ PMP-treated samples showed a significant decrease in the abundance of Gram-negative soil bacterial genus Sphingomonas (from 35 to 17% abundance for sprayed T1.1 and Τ1.2, from 8 to 4% for sprayed T1.3). Methylobacterium genus also decreased (from 10 to 4% abundance for sprayed T1.1, from 8 to 4% for sprayed T1.2, and from 1 to 0% for sprayed T1.3). The Kineococcus Gram-positive genus surprisingly also decreased (from 10 to 3% abundance for sprayed T1.1, from 10% to 0% abundance for sprayed T1.2, and from 4 to 0% abundance for sprayed T1.3), whereas the Gram-negative Pseudomonas sp. increased significantly (from 1 to 7% abundance for sprayed T1.1 and Τ1.2, and from 23 to 34% for sprayed T1.3). Also, the Hydrotalea genus increased (from 16 to 23% abundance for sprayed T1.3). Firmicutes, which were hardly detectable in the unsprayed tree, were represented by the genus Bacillus, which reached 2% for sprayed T1.1 and 4% for sprayed T1.2. The “Remainder” category, which aggregates all other detected genera aside from the top 15, increased significantly in abundance for sprayed T1.1 (from 5 to 15%) and for sprayed T1.2 (from 5 to 40%), enriching the species diversity on the leaf surface.

To assess the fungal phyllosphere dynamics of the same three olive trees (T1, T2, and T3) in another cultivation season, we collected samples from them the following year, in April 2023, and subsequently sprayed the trees with Mg(OH)₂ PMP. Prior to spraying, all trees had a healthy appearance (10% affected leaves) and only isolated leaves bearing peacock spots could be seen. Subsequent to spraying, we collected leaf samples ten days later, being cautious for rain. To differentiate fungal diversity at the genus level, we used only the ITS marker this time. The sequencing of the ITS gene resulted in 61,387, 54,857, 72,485, 74,502, 41,980, 71,520, 66,765, 67,205, and 63,793 and 74,474, 80,878, 64,454, 71,977, 53,458, 80,215, 76,805, 76,469, and 60,000 filtered reads for unsprayed (T1.1, T1.2, T1.3, T2.1, T2.2, T2.3, T3.1, T3.2, T3.3) and sprayed trees, respectively. In total, 393 genera were identified in the 18 samples; however, the top 15 genera accounted for >92% of the total abundance (Figure 7).

Surprisingly, the fungal pathogen Venturia dominated the fungal community in all unsprayed olive trees, T1 (23, 37, and 22%), T2 (23, 37, and 22%), and T3 (42, 38, and 24%). Treatment with Mg(OH)₂ PMP drastically reduced Venturia abundance after ten days, T1 (48, 0, and 2%), T2 (0, 0, and 8%), and T3 (0, 0, and 5%). Buckleyzyma was also found to be more susceptible to treatment. Its abundance was 10, 21, and 14% for unsprayed T1; 21, 26, and 13% for unsprayed T2; and 16, 23, and 9% for unsprayed T3; whereas for the sprayed samples, it was 16, 1, and 7% for T1; 3, 1, and 3% for T2; 5, 4, and 4% for T3. The void was covered by a significant increase in the relative abundance of the beneficial fungus Aureobasidium. In the unsprayed trees, its abundance was 16, 6, and 4% for T1; 9, 11, and 5% for T2; 6, 7, and 11% for T3; whereas for the sprayed trees, it was 4, 58, and 5% for T1; 55, 56, and 25% for T2; 39, 46, and 6% for T3. The results validate our previous observations on the extreme susceptibility of the Venturia pathogen to Mg(OH)₂ PMP treatment.

Principal component analysis (PCA) was applied to classify and cluster fungal leaf microbiome samples according to similarities of their identified ASVs for the three plants before and after spraying. Figure 8 shows that the first two principal components (PC1 + PC2) accounted for 90.9% of the total genetic variance. Samples before treatment clustered at a great distance from most of their counterpart-treated samples.

To assess further the effect of treatment with Mg(OH)₂ PMP in the leaf fungal microbiome community of olive trees in other locations and compare the effect to the one of a conventional copper fungicide, a region in Chalkidiki located 40 km away was chosen. The experiment was conducted in May 2023 and samples were collected for each group of trees before and 20 days after treatment. Each sample consisted of olive leaves collected from the three trees of each row. The experiment was repeated twice and in total, four replicates for each treatment were sampled. The results from the ITS analysis before and after Mg(OH)₂ PMP spraying are shown in Figure 9. The profile of the leaf epiphytic communities showed that the majority of fungal organisms present belonged to nine genera with four of them consistently present in all samples analyzed (Aureobasidium 18–63%, Symmetrospora 2–27%, Mycosphaerella 2–23%, Backleyzyma 1–21%) and five others only in distinct samples (Venturia 0–23%, Yarrowia 0–40%, Alternaria 0–6%, Ochroconis 0–15%, Pseudocercospora 0–18%). The most abundant genera were commensal species and have been considered as antagonistic to pathogens. In the case of the pathogen Venturia oleaginea, its presence was evident in patches of some trees morphologically. For the Kocide treatment, we did not observe a decrease in the abundance of Venturia, while in the case of Mg(OH)₂ PMP, the initial samples were relatively free of Venturia, and thus no conclusions can be made. Focusing on patches of leaves infected with OLS, Mg(OH)₂ PMP treatment reduced the presence of conidia, as previously seen (Figure S1).

3.3. Olive Leaf Epiphytic Fungal Load

Previously, when analyzing the effects of Mg(OH)₂ PMP on tomato leaves, a reduction in the fungal load was evident [15]. We therefore undertook quantifying the fungal load in the samples of the present study and related it to the abundance profile. For the quantification of the fungal load of olive leaves, microbial DNA was used in a quantitative PCR (qPCR) for the amplification of the whole eukaryotic internal transcribed spacer (ITS) region ITS1-5.8S-ITS2. Το construct the calibration curve, serial 10-fold dilutions of the plasmid DNA, ranging from 100 pg to 0.01 pg, were used, as explained in Section 2.6. The reaction efficiency was 101.12% and the linear regression equation (y = −3.295x + 19.72) related the DNA quantity to the Cq values with a regression coefficient (R²) of 0.995 (Figure S2). All the samples analyzed for fungal load in this work fell well within the calibration curve.

Regarding the olive trees in the Thermi area, samples from the 2022 and 2023 experiments were examined before and after treatment with Mg(OH)₂ PMP. Three biological replicates were used from the heavily infected tree T1 in 2022 and from the same tree in 2023 as well as two neighbor trees, T2 and T3, and an unsprayed tree T10 (Figure 10).

Strikingly, the fungal load of the infested T1 tree was substantially higher than all other samples tested and was quantified at 3.6 million copies/ng. Spraying with Mg(OH)₂ PMP dramatically reduced the fungal load 18 days later (Figure 10). All other samples had a substantially lower fungal load prior to spraying, which was consistently further reduced upon treatment. This also explains their relatively healthy morphological appearance. Considering also that Venturia almost disappeared from the metagenomics profile after spaying, the pathogen seems to have exceptional sensitivity to the treatment. The fungal species that covered the void in abundance (i.e., Aureobasidium and Cladosporium) were also susceptible but to a lesser degree.

We additionally tested the leaf samples from Chalkidiki. In this case, aggregate samples from three trees (with three biological replicates for each one) treated in the same manner were pooled prior to microbial DNA extraction, which may explain the higher variance seen in the fungal load before treatments (Figure 11). Mg(OH)₂ PMP treatment was benchmarked to commercial copper fungicide treatment (Kocide), with nearly equivalent effectiveness in fungal reduction. Overall, the unsprayed replicates showed a wide variance in fungal copy number. This is not surprising because visually, a similar picture was evident. The unsprayed trees also exhibited a moderate decrease in the fungal load at the second timepoint, which is likely due to the fact that spraying and sampling were performed at a late period due to the extended rainy spring season, and thus represents the seasonal reduction due to weather changes. Treatment with Kocide and Mg(OH)₂ PMP caused significant reductions in the fungal load and the variance was reduced in both cases. The treated copy numbers were similar to those from the Thermi grove.

3.4. Assessment of Plant Stress Responses

Extending on previous data on tomatoes pointing to moderate induction of plant stress response, we assessed the effect of Mg(OH)₂ PMP on the infected olive tree T1. Total RNA was isolated as previously described and used in a quantitative PCR (qPCR) to measure the expression of key indicator genes associated with responses to biotic and abiotic stress [e.g., beta-1,3-glucanase (β-Glu), chitinase (CHI), and phenylalanine ammonia lyase (PAL) genes] pre- and post-PMP spraying in two different growing seasons (2022 and 2023). The expression of the PAL gene was measured only in samples from the 2023 growing season. The CHI gene was significantly upregulated in the sprayed samples compared to the unsprayed (Figure 12A,B). Similarly, the expression of the PAL gene was significantly higher in the sprayed T1 compared to the unsprayed one (Figure 12C). These findings confirm that treatment with Mg(OH)₂ PMP may trigger defense responses in olive trees.

In contrast, the expression of the β-Glu gene was clearly downregulated in both growing seasons (Figure 13) after spraying.

4. Discussion

Previous investigation of Mg(OH)₂ PMPs had shown them to possess antimicrobial properties on the phyllosphere of tomato plants. Currently, we aimed to evaluate the bioactivity of Mg(OH)₂ PMPs on olive trees under field conditions, where some of the trees had been affected by fungal disease. Olive leaf spot or peacock spot disease is very prevalent in the Mediterranean region, especially in areas with higher humidity and mild temperatures, such as in northern Greece. One particular olive tree in 2022 (Thermi) with extreme OLS symptoms, in an otherwise healthy grove, was initially included in the Mg(OH)₂ PMP treatments with the neighboring healthy trees. The infected tree, besides the typical symptoms of peacock rings on mature leaves, showed fungal attack on new leaf stalks and shoots, and large parts of the bark were darkened. The extent of fungal colonization was evident from the SEM images. The combination of 18S rRNA and ITS amplicon metabarcoding and fungal load quantitation by qPCR provided an illuminating picture of the status of fungal pathogen prevalence, consistent with visual assessments of olive tree health. It is evident that the number of fungal microorganisms in combination with the relative abundance of the pathogens are determining plant disease severity. The presence of Venturia in the unsprayed infected tree, at 18 days after Mg(OH)₂ PMP spraying, was drastically reduced while at the same time, the fungal load decreased to levels comparable to healthy trees. The results are validated during the second season 2023 (Thermi), where even though Venturia is detectable in all tested trees, the fungal load is at normal levels and the trees exhibit a relatively healthy appearance. Treatment with Mg(OH)₂ PMP again reduces Venturia presence disproportionally compared to other species.

The effects of Mg(OH)₂ PMP on the fungal load were benchmarked with Kocide copper hydroxide treatments and showed comparable levels of reduction in fungal load for both treatments. Metagenomics data are suggestive that the exceptional sensitivity may be more specific to Mg(OH)₂ PMP, perhaps due to the layer formed on the leaf surface, but more experimentation is required.

The management strategies to control OLS disease include a combination of cultural practices, such as the choice of cultivation site, tree density, pruning for improved air circulation to minimize moisture on the leaf surface, etc., and chemical applications for prevention and treatment. Chemical control of OLS disease is mostly carried out with contact copper compounds. Copper is thought to penetrate leaves through wounds caused by erupting conidia during the evasion and being copper phytotoxic, thereby causing the premature drop of leaves [40]. This is postulated to be causing a reduction in inoculum available for new infections. However, this appears to be a partial effect, since in our quantitation of the fungal load in leaves sprayed with the commercial copper hydroxide compound, a significant reduction in the total fungal load is observed on the remaining leaves. In the case of the Mg(OH)₂ PMP, where no leaf loss is evident subsequent to spraying, we observed equivalent reductions in the fungal load. Our data show that the compound is exceptionally effective against Venturia, possibly being in a sensitive stage in its lifecycle. The modes of action of copper hydroxide and Mg(OH)₂ PMPs may differ enough and complement each other. It would be interesting to assess combinations of them in the future.

Copper-based antimicrobial compounds are the current standard practice due to their high toxicity to plant pathogens, low cost, low mammalian toxicity of the fixed Cu compounds, chemical stability that prevents them from being washed away quickly from plant surfaces, and the long residual periods, which are among the most important advantages of these compounds. This has led to their widespread use to control foliar plant pathogens with satisfactory levels of disease management. Consequently, they have become an important component of integrated pest management (IPM) system seeking to provide long-term solutions for disease management [41]. However, in recent years, the systematic and long-term use of copper compounds in agriculture has raised concerns for the high level of accumulation in the soil, the risk of surface water contamination, and the potential public health problems due to such forced entry of copper in the food chain. High levels of Cu in agricultural soil may cause plant stress and phytotoxicity and reduce soil fertility, with adverse effects on crop yield and quality. Consequently, the European Union has restricted the use of copper compounds, prescribing a maximum of 28 kg/ha of metallic copper over a period of 7 years (on average 4 kg/ha/year) and further restrictions are expected in the near future.

The increasing pressure to reduce the amounts of copper compounds used in agriculture has encouraged research efforts to identify viable alternatives in agriculture. Nanoparticle-based metallic formulations and biological control agents including resistance inducers are the most promising alternatives. Nanoparticles offer the advantage of the effective delivery of agrochemicals due to their large surface area, high particle number per unit mass cf. the respective micron-sized compounds, easy attachment, and fast mass transfer. Metal oxide nanomaterials, such as CuO, ZnO, MgO, and aluminum oxides, have been proven in laboratory tests to be effective against foliar and soilborne plant diseases caused by Botrytis cinerea, Alternaria alternata, Monilinia fructicola, Colletotrichum gloeosporioides, Fusarium solani, Fusarium oxysporum f.sp. radicis-lycopersici, Verticillium dahliae, Phytophthora infestans, and Ralstonia solanacearum in many plant species [9,12,42,43,44].

The smaller size and higher surface-to-volume ratio of the nano-sized compounds allow metallic particles to penetrate microbial membranes and release metal ions into solution more efficiently than the micron-sized compounds, conferring the nanometer-sized metallic compounds with a higher antibacterial activity [45]. However, the toxicological hazards and risks associated with nanomaterial exposure are not well understood, and environmental and human exposure due to nanomaterial residues in soil and crops may lead to adverse health outcomes. Laser diffraction analysis on the MgO starting material [15] showed a mean particle size of 5.4 microns, which is orders of magnitude greater than that of commercial NPs (≤100 nm) used in studies of other research groups [15]. SEM and TEM analysis images [15] revealed an irregular shape and a composite polycrystalline nature of the particles. The high porosity of the substance yielded a BET specific surface area of 234 m²/g (a measurement of the surface area of a solid material), which is equivalent to that of 13 nm dense NPs.

Systemic acquired resistance (SAR) is a plant defense mechanism that provides long-lasting protection against a broad spectrum of pathogens. The systemic acquired resistance response is dependent on the plant hormone, salicylic acid (SA), and is associated with the accumulation of pathogenesis-related (PR) proteins, which contribute to resistance. In the SAR state, plants are primed (sensitized) to quickly and effectively activate defense responses upon subsequent pathogen attack. This priming allows the plant to mount a faster and stronger defense, leading to enhanced resistance. SAR is characterized by the accumulation of plant metabolites and the expression of a broad range of defense-related genes [46]. Previous investigations indicated that MgO NPs induce systemic resistance against the pathogen by activating SA, JA, and ethylene signaling pathways in tomato plants [16]. The expression analysis of indicator genes in Mg(OH)₂ PMP-sprayed tomato plants revealed a moderate increase in PR2 and PR3 and PAL induction in all Mg PMP treatments, though at lower levels than those produced naturally by a plant under actual pathogenic attack [15]. In the olive tree tested, PAL exhibited a significant induction upon spraying as well, and chitinase exhibited a modest induction, whereas β-1,3-glucanase exhibited a modest reduction. The defense response may be important for the long-term olive resistance to pathogens. The reduction in β-GLU could be due to the reduction in the fungal load on leaves. Future examination of the global responses of olive leaves to Mg(OH)₂ PMP treatment as well as those of the colonizing microbes will be highly informative.

Interestingly, in all spraying applications, the yeast-like fungus Aureobasidium pullulans appeared to be relatively resistant compared to the pathogen, increasing its relative abundance. The organism has been utilized for plant protection due to its biocontrol and plant defense-inducing properties. Its use in agriculture has shown promising results in protecting plants from various pathogens and in enhancing their resistance to diseases. Some of the key applications and effects of Aureobasidium pullulans for plant protection include reducing the incidence of diseases, such as late blight (Phytophthora infestans) and grey mold in strawberries, and inducing plant defense responses, including the expression of pathogenesis-related proteins (PR proteins), which are associated with enhanced plant resistance to pathogens [47,48]. Mg(OH)₂ PMPs could be integrated into the existing olive disease management programs and proportionally decrease copper hydroxide sprays during fall and spring periods. The compound could offer additional, unaccounted benefits to the trees, such as growth promotion due to Mg⁺² absorption by the leaves, and insect-repellent properties documented in other plant species, due to its mechanical properties.

5. Conclusions

Focusing on olive leaf spot, an important fungal disease greatly affecting plant health and production, applications of Mg(OH)₂ PMP decimated the Venturia oleaginea fungal pathogen and reduced the overall fungal load at levels equivalent to those achieved with copper hydroxide spraying. Retention of the compound on olive leaves exceeded 47 days post-spraying. No toxicity and leaf loss were observed. Chitinase and PAL gene upregulation indicate the induction of systemic acquired resistance response upon spraying. Beneficial microbes such as Aureobasidium pollulans are more resistant to treatment, thus covering the void of pathogen elimination. Taken together, the results point to the exceptional effectiveness of Mg(OH)₂ PMP against olive leaf spot disease, which can be valuable in plant protection efforts and IPM. Future work will aim to examine combinations of Mg(OH)₂ PMP and low doses of copper hydroxide for synergies and the addition of natural biopolymers for increased retention on the leaf surfaces. Additionally, the compound can be evaluated for its insect pest-repellent properties, previously established for tomatoes, on olives and other important plant species.