Preharvest application of antagonistic yeasts for the control of fungal pathogens in organic peach and plum orchards

Abstract

BACKGROUND

The control of fungal diseases in organic fruit production remains a major challenge due to the limited availability of authorized phytosanitary tools. This study evaluates the efficacy of two antagonistic yeasts – Metschnikowia pulcherrima L672 and Hanseniaspora uvarum L793 – in organic peach ( Prunus persica ) and plum ( Prunus salicina ) orchards over two consecutive seasons. Four treatments were compared: an untreated control, a standard organic management program, and foliar applications of each yeast strain.

RESULTS

Microbiological monitoring revealed that Metschnikowia pulcherrima L672 demonstrated superior persistence and colonization, significantly reducing the abundance and colony size of filamentous fungi, including key pathogenic genera such as Alternaria, Cladosporium, and Penicillium spp. This biocontrol effect translated into improved fruit quality, notably a significant reduction in discarded plums at harvest, without compromising overall yield. In contrast, H. uvarum L793 exhibited limited persistence and a more transient antagonistic effect.

CONCLUSION

These findings support the integration of Metschnikowia pulcherrima L672 as a promising and effective biocontrol agent for managing fungal diseases in organic stone fruit production systems. © 2026 The Author(s). Journal of the Science of Food and Agriculture published by John Wiley & Sons Ltd on behalf of Society of Chemical Industry.

INTRODUCTION

Spain is the leading European producer of fruits and vegetables, accounting for approximately 25% of total production, followed by Italy and Poland. ¹ Among these, peaches and plums are key stone fruits. Spain produces over 39% of Europe's peaches. ² In the Extremadura region, plums are the dominant stone fruit crop, with a harvest of 98 545 t in 2022, followed by peaches with 42 901 t. ¹ The fruit sector is of significant socio‐economic importance in the region, representing 25.9% of the total value of vegetable production.

However, fruit production faces significant challenges. Fungal pathogens such as Monilinia spp., Alternaria alternata, Botrytis cinerea, and Cladosporium herbarum are responsible for major preharvest and postharvest losses and pose a food safety risk due to potential mycotoxin contamination.^{3, 4} While synthetic fungicides such as tebuconazole and fludioxonil are commonly used in conventional agriculture, their use is prohibited in organic farming under EU Regulation 2018/848, ⁵ which mandates sustainable practices that protect the environment.

This regulatory framework leaves organic fruit production particularly vulnerable due to the limited availability of effective phytosanitary tools, prompting the search for alternative strategies. These include the use of microbial bioproducts and biological control agents (BCAs). Recent reviews highlight that microbial inoculants are gaining prominence in sustainable agriculture for their roles in plant growth promotion, disease suppression, and enhancing plant tolerance to abiotic stress.^{6, 7}

Biological control, particularly using antagonistic yeasts, represents a promising strategy to manage fungal diseases in fruit crops. Yeasts such as Metschnikowia pulcherrima and Hanseniaspora uvarum exhibit multiple antagonistic mechanisms, including competition for nutrients, iron chelation via pulcherrimin production, and the release of antifungal volatile organic compounds (VOCs).^{8, 9} The efficacy of these mechanisms has been validated in both in vitro and postharvest applications, where they effectively control pathogens such as Penicillium expansum, B. cinerea, Monilinia laxa, and A. alternata.^{10, 11, 12, 13}

Despite their postharvest success, the translation of BCAs to field conditions remains a significant hurdle due to variable environmental conditions, formulation limitations, and regulatory challenges. Recent studies highlight the need for robust formulations and standardized application protocols to enhance their field effectiveness and scalability.^{14, 15}

Therefore, this study aimed to evaluate the preharvest application of M. pulcherrima L672 and H. uvarum L793 in organic peach and plum orchards over two seasons. Specifically, we assessed their colonization efficiency and persistence on epiphytic surfaces, their impact on the native fungal community structure, and the resulting effects on fruit yield and quality at harvest.

MATERIALS AND METHODS

Experimental design and climatic conditions

Field trials were conducted on the peach cultivar ‘Crimson Lady’ and the plum cultivar ‘Larry Ann’, planted at densities of 5 × 3 m and 5 × 4 m, respectively. Experimental plots were established in organic orchards managed by ‘Extremadura Sana’ (peach) and ‘Agrícola Ecológica’ (plum), both located in Badajoz, Spain. A randomized complete block design was implemented to minimize potential cross‐contamination between treatments. Each treatment plot covered an area of 417 m², with a planting framework of 5 m × 3 m and buffer zones established between plots. Four treatments were evaluated: BC1 (untreated control), BC2 (standard organic management), BC3 (foliar application of M. pulcherrima L672 ¹¹ ), and BC4 (foliar application of H. uvarum L793 ¹¹ ). The BC2 treatment, representing the standard organic management protocol, included applications of copper oxychloride (70%) at 2.5 kg ha⁻¹, tribasic copper sulfate (40%) at 4 kg ha⁻¹, paraffinic oil, azadirachtin (1 L ha⁻¹), pyrethrum extract (4%, 2 L ha⁻¹), potassium soap (6 L ha⁻¹), calcium polysulphide, and wettable sulfur (80%, 5 kg ha⁻¹).

Climatic data for the 2018 and 2019 growing seasons were obtained from the REDAREX network of meteorological stations operated by the Junta de Extremadura. Data for monthly maximum, mean, and minimum temperatures, along with total and effective precipitation, were recorded at the Arroyo de San Serván station. These data are summarized in Supporting Information, Table S1.

Foliar application of treatments

All treatments (BC1–BC4) were applied as foliar sprays using a standardized protocol to ensure uniform coverage across all plots. A total spray volume of 100 L was applied per treatment plot. The treatments consisted of: BC1 (control), sprayed with water only; BC2, sprayed with the standard organic phytosanitary products described above; BC3, a suspension of M. pulcherrima L672; and BC4, a suspension of H. uvarum L793.

Yeast inocula were prepared by culturing each strain in Yeast Extract Dextrose Peptone (YEDP) broth (Scharlab, Barcelona, Spain), at 25 °C for 72 h with constant agitation (90 rpm). The resulting cultures were centrifuged (5000 rpm, 5 min), washed twice with sterile distilled water, and the final cell concentration was adjusted to 1 × 10⁷ cells mL⁻¹ using a Neubauer chamber.

Six applications of antagonistic yeast treatments were carried out throughout the fruit development period in plum and peach orchards. The intervals between applications were adjusted according to the crop and its phenological progression, ranging from 8 to 14 days for peach and 14 to 29 days for plum (Supporting Information, Table S2). Treatments were scheduled to cover the main phenological stages, including flowering, full bloom/petal fall, fruit set, and all subsequent phases of fruit growth and maturation (Table S2). This strategy ensured not only early protection during the most vulnerable initial stages but also continuous coverage as the fruit developed, thereby maximizing the colonization, persistence, and efficacy of the biological control agent during preharvest. The specific timing of applications was dictated by the phenological development of each species and adjusted between the two growing seasons to account for interannual climatic variability, which influenced pathogen pressure and the optimal windows for treatment delivery (Table S2). To monitor the dynamics of the biocontrol agents (BC3 and BC4), samples were collected immediately before and after each of the six applications. The pre‐application sampling was designed to establish the baseline native population (prior to the first treatment) and to assess the persistence of the inoculated yeast from the previous application (for subsequent treatments).

Sampling and microbiological analysis

To evaluate the microbial dynamics on plant surfaces, leaf and fruit samples were collected immediately before and after each treatment application. For each sampling event, three composite samples were collected per plot, with each composite sample comprising material from five randomly selected trees.

Microbiological analysis was conducted by preparing six serial decimal dilutions (from 10⁻¹ to 10⁻⁶) from each sample in tryptone water (0.9% w/v). Initially, 10 g of sample were homogenized in 90 mL of diluent, and 100 μL aliquots from each dilution were spread‐plated on Potato Dextrose Agar (PDA) medium (Scharlab), acidified to pH 4.5 with tartaric acid to selectively inhibit bacterial growth. Plates were incubated at 25 °C for 5 days.

After incubation, colony‐forming units (CFUs) of yeasts and filamentous fungi were enumerated from plates containing between 30 and 300 colonies. Results were expressed as log₁₀ CFU per gram of fresh weight. All microbiological analyses were performed in triplicate.

Monitoring biocontrol agents and fungal populations

The persistence and colonization efficiency of the biocontrol yeasts M. pulcherrima L672 and H. uvarum L793 were assessed through morphological and molecular identification of yeast colonies recovered from treated phyllosphere (leaves) and carposphere (developing fruit surfaces) samples. Colonies displaying morphological characteristics consistent with the inoculated strains were selected from PDA plates for molecular characterization. Furthermore, colonies with distinct morphologies were also isolated to assess the impact of the treatments on the native yeast community.

The impact of biocontrol treatments on the resident filamentous fungal population was evaluated using two complementary approaches: (i) measurement of colony diameter to assess growth inhibition; and (ii) morphological grouping and molecular identification to determine shifts in fungal community composition.

Colony diameters were measured using a digital calliper and classified into three size categories: <3 mm, 3–8 mm, and > 8 mm. The proportion of colonies within each size range was calculated for each treatment, fruit type, and year. In parallel, filamentous fungal colonies were grouped based on their macroscopic morphology, and representative isolates (two to three per morphotype) were selected for molecular identification.

Genomic DNA was extracted using the Quick‐DNA Fungal/Bacterial Miniprep Kit (Zymo Research, Irvine, CA, USA). The internal transcribed spacer (ITS) region was amplified using the ITS1 and ITS4 primers, ¹⁶ following the protocol of Gallardo et al. ¹⁷ Polymerase chain reaction (PCR) products were purified with the GeneJET PCR Purification Kit (Thermo Fisher Scientific, Waltham, MA, USA) and sequenced bidirectionally at the STAB facility of the University of Extremadura (Badajoz, Spain). Consensus sequences were assembled using Chromas Pro v1.49 beta (Technelysium) and compared against the EMBL and GenBank databases using the BLASTn algorithm. Taxonomic identification was assigned based on the highest sequence similarity (>98%).

Determination of yield

Fruit yield was assessed at harvest by manually collecting all fruit from individual trees. Each tree was considered a replicate. A total of 16 trees were sampled per fruit type, treatment, and year, with four trees per plot. The total fruit weight per tree was recorded and extrapolated to yield values expressed in tonnes per hectare.

Harvested fruit was classified into three categories: (i) marketable fruit (MF), expressed in tonnes per hectare; (ii) unripe fruit, defined as fruit that had not reached the optimal ripening stage, expressed as a percentage of the total yield; and (iii) discarded fruit, comprising pieces affected by microbial decay, pest damage, or overripening, also expressed as a percentage.

To determine average weight of the fruit, 30 fruits were randomly selected from each replicate tree and weighed individually.

Data analysis

Statistical analyses were performed using analysis of variance (ANOVA) to evaluate the effects of fruit type (peach or plum), treatment (BC1–BC4), and year (2018 or 2019) as main factors. For microbial data, sampling time (before vs. after application) was included as an additional factor, with data from the six individual application events pooled for the analysis.

Microbial count data were log‐transformed before analysis to meet the assumptions of normality. The relative frequency of fungal colonies of each size category (<3 mm, 3–8 mm, >8 mm) was also analysed.

Heatmaps were generated using R software (v. 4.5.1; R Core Team, Vienna, Austria) to visualize the clustering of samples based on the abundance of fungal morphotypes. Data were row‐normalized (z‐score) and analysed using hierarchical clustering with Euclidean distance and the complete linkage method.

Yield parameters, including total yield (t ha⁻¹), MF (t ha⁻¹), percentage of unripe fruit, percentage of discarded fruit, and average fruit weight, were analysed using ANOVA. The experimental unit for yield analysis was the individual tree (n = 16 per fruit type, treatment, and year). Where significant differences were detected (P < 0.05), Tukey's Honest Significant Difference (HSD) test was used for mean separation.

RESULTS

Impact of antagonistic yeast applications on epiphytic yeast populations

The application of antagonistic yeasts influenced the composition and dynamics of the native yeast communities present on the epiphytic surfaces (leaves and fruit) of peach and plum trees. These results, which represent the pooled data from all six application events, revealed no statistically significant differences in the overall abundance between treatments or fruit types (Fig. 1(A),(B)), with mean values ranging from 5.01 ± 1.12 to 5.60 ± 0.96 log₁₀ CFU g⁻¹ across seasons. However, a slight non‐significant increase in yeast counts was observed in treatments BC3 and BC4, particularly after biocontrol applications (Fig. 1(A)).

Figure 1.

Mean epiphytic yeast counts (log₁₀ CFU g⁻¹). (A) Comparison across treatments (BC1–BC4). For the biocontrol treatments (BC3 and BC4), data are separated into samples taken immediately before and after application. (B) Comparison of total, peach, and plum samples. Data represent the average of all sampling points from the 2018 and 2019 seasons. Error bars represent the standard deviation.

While quantitative shifts were minor, morphological changes in yeast colonies were pronounced, particularly in the BC3 treatment (M. pulcherrima). Visual inspection of PDA plates revealed a clear transition in colony morphology before and after treatment (Supporting Information, Figure S1), consistent with successful colonization by the inoculated strain.

In untreated control plots (BC1), the yeast community was dominated by native yeast‐like fungi, primarily Aureobasidium pullulans and Aureobasidium leucospermi. They are characterized by soft, moist colonies with pale‐pink pigmentation that darkened over time. These species were consistently detected across fruit types and seasons, with population levels ranging from 3.67 to 5.84 log₁₀ CFU g⁻¹. Pigmented yeasts such as Cystobasidium pinicola, Cystofilobasidium macerans, Cystobasidium lysinophilum, Symmetrospora symmetrica, Vishniacozyma carnescens, and Sporobolomyces roseus were sporadically detected, while non‐pigmented species, including Filobasidium magnum, F. stepposum, and F. wieringae, were present at lower concentrations (<4.62 log₁₀ CFU g⁻¹).

In BC2 (standard organic management), yeast populations were similar in composition and abundance to those in BC1, indicating that copper‐based treatments did not significantly alter the native yeast community. As expected, isolates corresponding to the inoculated biocontrol strains were not found in either BC1 or BC2 plots.

The BC3 treatment resulted in a marked shift in yeast community structure. Colonies morphologically consistent with M. pulcherrima dominated the plates post application, and molecular identification confirmed its presence. Fifteen days after treatment, M. pulcherrima remained the predominant species, although Aureobasidium spp. reappeared in moderate numbers.

In contrast, the application of H. uvarum (BC4) resulted in lower persistence. Although H. uvarum was detected shortly after application, its population declined significantly within 15 days, allowing Aureobasidium spp. to regain dominance. These results suggest that M. pulcherrima exhibits superior ecological fitness and persistence in the phyllosphere compared to H. uvarum, making it a more promising candidate for preharvest biocontrol in organic fruit systems.

Impact of antagonistic yeast treatments on epiphytic filamentous fungal populations

The application of antagonistic yeasts influenced both the abundance and community structure of filamentous fungi on the epiphytic surfaces (leaves and fruit) (Fig. 2(A)). While fungal counts showed a slight, non‐significant reduction in 2018, a significant effect was observed in 2019 (Fig. 2(A)). In the 2019 season, the BC3 treatment (M. pulcherrima) significantly reduced filamentous fungi counts in both crops, while BC4 (H. uvarum) showed a moderate, non‐significant reduction (Fig. 2(B)).

Figure 2.

Mean filamentous fungal populations (log₁₀ CFU g⁻¹) on the epiphytic surfaces. (A) Overall counts per treatment in each season. (B) Counts separated by fruit type and season. Different letters indicate significant differences between treatments within a season or crop (Tukey's HSD, P ≤ 0.05). Error bars represent the standard deviation.

In addition to fungal abundance, the distribution of colony sizes was also affected by yeast treatments (Fig. 3). In BC3‐treated samples, small colonies (<3 mm) were predominant, accounting for ~78% of the total, a significantly higher proportion than in the control (BC1, ~53%) and standard organic (BC2, ~41%) treatments. Consequently, large colonies (>8 mm) were markedly reduced in the BC3 treatment (~4%). These results suggest that M. pulcherrima effectively inhibited fungal growth and hyphal expansion.

Figure 3.

Relative abundance of filamentous fungal colonies classified by size category (<3 mm, 3–8 mm, >8 mm). Data are shown for plum, peach, and the global average (Global) for each treatment. Treatments: BC1 (untreated control), BC2 (standard organic management), BC3 (Metschnikowia pulcherrima), and BC4 ( Hanseniaspora uvarum ). Different letters indicate significant differences (Tukey's HSD, P ≤ 0.05) between treatments for a specific size category.

To assess the impact of biocontrol treatments on population composition, filamentous fungi were classified into seven morphological groups (MGs). The dominant group (MG1), characterized by dark, irregular colonies, included species such as Cladosporium cladosporioides and Cladosporium oxysporum. Morphologies associated with this group were consistently detected in all samples. However, its abundance was generally lower in BC3‐treated samples compared to the other treatments (Table 1).

Table 1.

Mean abundance (log CFU g⁻¹ ± SD) of seven morphological groups (MG1–MG7) of filamentous fungi isolated from peach and plum epiphytic samples under four treatments (BC1–BC4) across two seasons (S1 and S2)

Treatment			Filamentous fungi morphological group
			MG1	MG2	MG3	MG4	MG5	MG6	MG7
			Mean ± SD	Mean ± SD	Mean ± SD	Mean ± SD	Mean ± SD	Mean ± SD	Mean ± SD
Peach	BC1	S1	3.53 ± 1.29	3.65ab ± 1.13	0.67 ± 1.15	0.00 ± 0.00	0.00 ± 0.00	0.00 ± 0.00	0.00 ± 0.00
		S2	3.39 ± 0.75	2.99ab ± 0.41	1.69 ± 1.15	0.00 ± 0.00	0.00 ± 0.00	0.50 ± 1.00	0.00 ± 0.00
	BC2	S1	4.24 ± 0.68	3.49ab ± 1.31	0.67 ± 1.15	1.59 ± 2.76	0.00 ± 0.00	1.33 ± 2.31	0.00 ± 0.00
		S2	3.20 ± 0.64	1.91ab ± 1.28	1.25 ± 1.50	0.00 ± 0.00	0.75 ± 1.50	0.00 ± 0.00	1.33 ± 1.62
	BC3	S1	4.17 ± 0.51	0.00b ± 0.00	0.00 ± 0.00	0.00 ± 0.00	0.67 ± 1.55	0.00 ± 0.00	1.33 ± 2.31
		S2	3.44 ± 0.37	2.41ab ± 0.47	0.58 ± 1.15	0.00 ± 0.00	0.50 ± 1.00	0.00 ± 0.00	0.00 ± 0.00
	BC4	S1	4.12 ± 0.98	3.43ab ± 0.75	2.00 ± 1.73	1.53 ± 2.66	0.00 ± 0.00	0.00 ± 0.00	0.00 ± 0.00
		S2	3.52 ± 0.46	2.57ab ± 0.34	1.83 ± 1.36	0.00 ± 0.00	1.12 ± 1.31	0.00 ± 0.00	0.00 ± 0.00
Plum	BC1	S1	5.30 ± 1.16a	3.65ab ± 2.33	2.65 ± 0.92	2.00 ± 2.83	1.00 ± 1.41	0.00 ± 0.00	0.00 ± 0.00
		S2	4.12 ± 0.67	2.49ab ± 0.50	2.00 ± 1.73	0.00 ± 0.00	0.77 ± 1.33	0.00 ± 0.00	0.00 ± 0.00
	BC2	S1	4.22 ± 0.37	1.00ab ± 1.41	1.65 ± 2.33	2.00 ± 2.83	0.00 ± 0.00	0.00 ± 0.00	0.00 ± 0.00
		S2	3.96 ± 0.56	2.26ab ± 1.96	2.83 ± 0.43	0.00 ± 0.00	0.67 ± 1.15	1.00 ± 1.73	0.00 ± 0.00
	BC3	S1	4.50 ± 1.13	2.50ab ± 3.54	0.00 ± 0.00	0.00 ± 0.00	0.00 ± 0.00	0.00 ± 0.00	0.00 ± 0.00
		S2	3.90 ± 0.28	1.77ab ± 1.57	0.00 ± 0.00	0.00 ± 0.00	1.67 ± 1.53	0.00 ± 0.00	0.00 ± 0.00
	BC4	S1	4.96 ± 1.58	4.30a ± 1.84	1.65 ± 2.33	0.00 ± 0.00	0.00 ± 0.00	0.00 ± 0.00	1.50 ± 2.12
		S2	3.94 ± 0.45	2.10ab ± 1.83	1.83 ± 1.75	0.00 ± 0.00	0.87 ± 1.50	0.00 ± 0.00	0.00 ± 0.00
Total	BC1		3.92 ± 1.08	3.14A ± 1.03	1.67A ± 1.30	0.33 ± 1.15	0.36 ± 0.84	0.17 ± 0.58	0.00 ± 0.00
	BC2		3.82 ± 0.70	2.24AB ± 1.54	1.56A ± 1.44	0.73 ± 1.72	0.42 ± 0.99	0.58 ± 1.38	0.44 ± 1.07
	BC3		3.91 ± 0.61	1.66B ± 1.65	0.19B ± 0.66	0.00 ± 0.00	0.75 ± 1.14	0.00 ± 0.00	0.33 ± 1.15
	BC4		4.01 ± 0.87	2.96AB ± 1.30	1.84A ± 1.45	0.38 ± 1.33	0.59 ± 1.08	0.00 ± 0.00	0.25 ± 0.87

^a Data are expressed as mean value ± standard deviation of CFU g⁻¹. Within the ‘Total’ section, different upper‐case letters in the same column indicate significant differences between treatments (P < 0.05). Within the main table, different lower‐case letters indicate significant differences for a specific morphological group across all conditions (fruit, treatment, season).

The second most prevalent group (MG2) comprised round, cottony, convex colonies with greenish‐brown mycelia, occasionally bordered by white crowns. Molecular identification confirmed the presence of several Alternaria species, including A. alternata, Alternaria infectoria, Alternaria tenuissima, Alternaria arborescens, and Alternaria eureka, as well as Didymella glomerata. The MG2 group was significantly affected by the biocontrol agents. The BC3 treatment consistently reduced MG2 counts to the lowest overall value (1.66 ± 1.65 log₁₀ CFU g⁻¹), a significant reduction compared to the controls (BC1, 3.14 ± 1.03 log₁₀ CFU g⁻¹; BC2, 2.24 ± 1.54 log₁₀ CFU g⁻¹). Notably, MG2 was completely absent in peach samples from the BC3 treatment during the first season (Table 1).

Other morphological groups were detected at lower frequencies. The third morphological group (MG3), exhibiting white mycelia or orange/red pigmentation, was identified as Epicoccum nigrum and Epicoccum layuense. These colonies occurred sporadically, with counts up to 2.83 ± 0.43 log₁₀ CFU g⁻¹ in plum samples under BC2, and 2.00 ± 1.73 log₁₀ CFU g⁻¹ in peach under BC4. The MG3 was significantly suppressed in the BC3 treatment (0.19 ± 0.66 log₁₀ CFU g⁻¹).

Additional groups were detected occasionally. The MG4 group, characterized by greyish/bluish sporulation which included Penicillium spp. and Talaromyces ramulosus, reached counts of 2.00 ± 2.83 log₁₀ CFU g⁻¹ in plum under the BC1 and BC2 treatments, and 1.59 ± 2.76 log₁₀ CFU g⁻¹ in peach in the BC2 treatment. The BC3 treatment showed a complete absence of the MG4 group across all samples. The MG5 group (cottony white mycelium, Arthrinium and Stemphylium) appeared sporadically, with low counts in the BC1 and BC4 treatments (≤1.12 ± 1.31 log₁₀ CFU g⁻¹), while the BC3 treatment maintained similar or lower levels (0.75 ± 1.14 log₁₀ CFU g⁻¹). The MG6 group (Aspergillus spp.) was absent in most treatments, except for isolated detections in the BC2 treatment in peach (1.33 ± 2.31 log₁₀ CFU g⁻¹). The MG7 group (B. cinerea) was detected at very low levels, never exceeding 1.50 ± 2.12 log₁₀ CFU g⁻¹, and was absent in the BC3‐treated samples.

These patterns were visualized using a heatmap analysis (Figure 4). The analysis confirmed that MG1 was the most abundant group across all treatments. In contrast, samples from the BC3 treatment consistently showed lower relative abundances of MG2 and the minor morphological groups (MG3–MG7), indicated by the prevalence of blue tones. Seasonal variation was evident, with higher overall abundance in season 1, yet the BC3 treatment maintained consistent suppression across both campaigns.

Figure 4.

Heatmap illustrating the relative abundance (z‐score normalized) of the seven fungal morphogroups (MG1–MG7). Rows represent sample combinations (fruit–treatment–season), and columns represent the morphological groups. Red tones indicate higher relative abundance, while blue tones indicate lower abundance. Hierarchical clustering was applied to rows and columns to group them based on similarity. Crops: peach (PEA) and plum (PLU); treatments (BC1–BC4); seasons: S1 and S2.

Overall, the BC3 treatment showed the most balanced performance, reducing the MG2 group and secondary morphological groups while limiting the MG1 group dominance. The BC3 treatment is the most promising biocontrol strategy for managing fungal diversity and minimizing pathogenic pressure in peach and plum production systems.

Effect of antagonistic yeast treatments on yield performance

The yield parameters of peach and plum orchards subjected to four treatments (BC1–BC4) over two consecutive seasons (2018 and 2019) are presented in Table 2.

Table 2.

Yield parameters of peach and plum orchards under four treatments in 2018 and 2019 seasons

Fruit	Treatment				Year		P‐value
Factor	BC1	BC2	BC3	BC4	2018	2019
	Mean ± SD	Mean ± SD	Mean ± SD	Mean ± SD	Mean ± SD	Mean ± SD	Year	Tr	Y* Tr
Peach
Total yield (t ha−1)	21.9 ± 6.33	20.5 ± 3.17	16.8 ± 4.26	13.6 ± 6.61	20.2 ± 5.29	16.2 ± 5.91	0.153	0.167	0.506
MF (t ha−1)	20.9 ± 6.54	18.7 ± 4.02	15.5 ± 4.08	12.6 ± 6.32	18.7 ± 5.40	15.2 ± 5.83	0.226	0.216	0.568
Unripe fruit (%)	1.46 ± 0.53	4.94 ± 6.38	3.27 ± 2.76	4.49 ± 2.72	4.42 ± 4.58	2.66 ± 2.33	0.363	0.565	0.367
Spoiled fruits (%)	3.43 ± 2.15	4.64 ± 2.29	4.92 ± 3.76	3.49 ± 2.00	3.51 ± 1.83	4.73 ± 2.97	0.447	0.855	0.937
AFW (g)	113.3 ± 13.7	116.6 ± 8.36	128.8 ± 12.2	137.4 ± 5.74	123.8 ± 15.2	124.2 ± 13.0	0.953	0.077	0.854
Plum
Total yield (t ha−1)	37.2 ± 11.8	37.3 ± 4.16	42.1 ± 12.6	46.4 ± 5.44	36.9 ± 9.91	44.7 ± 7.63	0.052	0.283	0.989
MF (t ha−1)	29.3 ± 7.93	33.6 ± 3.90	36.4 ± 11.2	41.2 ± 5.69	32.0 ± 8.85	38.3 ± 7.06	0.059	0.087	0.941
Unripe fruit (%)	10.8bc ± 3.77	7.34a ± 1.91	16.27c ± 4.06	8.97ab ± 2.12	8.38 ± 2.92	13.31 ± 4.54	0.000	0.000	0.026
Spoiled fruits (%)	7.79b± 2.13	3.56a ± 0.69	2.72a ± 0.70	4.07a ± 0.48	5.11 ± 2.52	3.96 ± 1.93	0.022	0.000	0.889
AFW (g)	116.3 ± 5.25	88.0 ± 28.1	91.2 ± 10.5	72.7 ± 6.59	90.4 ± 26.1	93.8 ± 16.9	0.624	0.003	0.866

MF, marketable fruit; AFW: average fruit weight. Within each row, mean values followed by different letters indicate statistically significant differences (P ≤ 0.05).

In peaches, no statistically significant differences were found in total yield, MF, or average fruit weight (AFW) across treatments or years (P > 0.05). However, notable trends emerged that support the potential of biocontrol agents. Specifically, M. pulcherrima (BC3) appeared to maintain yield quality while reducing fungal incidence.

Total peach yield ranged from 13.6 to 21.9 t ha⁻¹, with BC1 (untreated control) and BC2 (standard organic treatment) showing the highest values. Although BC3 and BC4 (yeast treatments) yielded slightly lower totals, the MF values remained comparable across treatments, suggesting that the biocontrol applications did not compromise marketable output. The percentage of unripe and discarded fruits was also statistically similar among treatments. Notably, BC1 exhibited the lowest proportion of unripe fruit (1.46%), and BC3 showed the highest AFW (137.4 g), suggesting a potential positive effect of M. pulcherrima on fruit size.

In plums, total yield ranged from 37.2 to 46.4 t ha⁻¹, with BC4 achieving the highest yield. Significantly, BC3 resulted in the lowest percentage of discarded fruits (2.72%) and a moderate proportion of unripe fruits (16.27%), suggesting a protective effect of M. pulcherrima against preharvest decay. The AFW values were consistent across treatments, with no significant differences observed.

DISCUSSION

The application of antagonistic yeasts in preharvest stages of fruit production represents a promising but less established strategy for sustainable disease management, particularly in organic systems where synthetic fungicides are restricted. While biological control has gained traction in postharvest contexts,^{18, 19} its implementation in field conditions, especially in perennial woody crops, remains limited due to biological, ecological, and logistical challenges. ²⁰ Woody plants, such as stone fruit trees, present unique constraints for biocontrol strategies: their perennial nature, complex anatomy, and large biomass complicate the establishment and persistence of microbial antagonists. Furthermore, the absence of crop rotation and the long‐term survival of pathogens in plant tissues reduce the efficacy of many control methods. In addition to these agronomic limitations, microbial pesticides face intrinsic challenges: their high specificity often requires the use of multiple strains to target different pathogens; their efficacy is highly variable under fluctuating environmental conditions; and many antagonists struggle to colonize and persist in the epiphytic environment, especially in perennial systems. ²¹ In this context, the use of antagonistic yeasts such as M. pulcherrima L672 and H. uvarum L793 offers a promising alternative, particularly suited to organic production systems where chemical inputs are restricted by legislation (e.g., EC Regulation 2018/848). ⁵

In the present study, the application of four different treatments (BC1–BC4) to peach and plum trees revealed distinct patterns in yeast population structure and dynamics. Although overall yeast counts remained statistically similar across treatments and seasons (ranging from 5.01 to 5.60 log₁₀ CFU g⁻¹), the composition and dominance of specific taxa varied notably. In untreated trees (BC1), the epiphytic surfaces were dominated by native yeast‐like fungi, particularly A. pullulans and A. leucospermi, consistent with previous reports of their prevalence in temperate orchards.^{22, 23} These species were accompanied by sporadic detection of pigmented genera such as Cystobasidium, Symmetrospora, and Vishniacozyma, as well as non‐pigmented Filobasidium spp., ²⁴ highlighting the natural diversity of the stone fruit epiphytic surfaces. In BC2, where standard organic treatments were applied, yeast populations mirrored those of BC1. The most pronounced shift occurred in BC3, where M. pulcherrima was applied. Post‐treatment samples showed a homogeneous dominance of colonies morphologically and molecularly identified as M. pulcherrima, confirming its strong colonization capacity under field conditions and its ability to outcompete native yeasts.^{23, 25} Conversely, BC4, which involved the H. uvarum application, resulted in lower colonization efficiency. Although H. uvarum was detected, its population declined significantly within 15 days post application, and Aureobasidium spp. regained dominance, consistent with previous findings in postharvest applications. ²⁶ These findings reinforce that successful biocontrol depends not only on the antagonistic properties of the applied strains but also on their ecological adaptability and persistence. Therefore, the superior establishment of M. pulcherrima in BC3 highlights its potential as a robust biocontrol agent in organic stone fruit production systems.

The application of antagonistic yeasts, particularly M. pulcherrima L672 (BC3), significantly altered the epiphytic fungal community in both peach and plum orchards. While total fungal counts were not consistently reduced, BC3 consistently suppressed the abundance and colony size of filamentous fungi, especially during the 2019 season, which could be associated with low rainfall (Supporting Information, Table S1). This aligns with previous studies demonstrating the antifungal efficacy of M. pulcherrima and H. uvarum against a broad spectrum of phytopathogens.^{11, 13, 27}

The reduction in dominant morphological groups such as Cladosporium and Alternaria spp. under BC3 is particularly relevant, as the Alternaria genus is among the most prevalent and damaging pathogens in these fruit crops.^{3, 4} The ability of M. pulcherrima L672 to suppress these taxa and other key postharvest pathogens such as P. expansum, C. cladosporioides, B. cinerea and M. laxa ^{10, 11, 12, 27, 28} is likely due to its production of pulcherrimin, a pigment that chelates iron and limits fungal growth, as well as other mechanism as secretion of antifungal volatile organic compounds (VOCs) and direct parasitism.^{8, 11, 29, 30} These mechanisms have been validated in postharvest contexts, where M. pulcherrima extended the shelf life of cherries and plums.^{10, 27, 28}

In contrast, H. uvarum (BC4) exhibited a more limited and transient effect. The reduced persistence may be attributed to its lower ecological competitiveness in the epiphytic environment, as previously reported.^{13, 26} Nevertheless, H. uvarum has demonstrated antifungal activity through VOC production, particularly against Aspergillus flavus and P. expansum, suggesting its potential utility in integrated biocontrol strategies.^{12, 13, 31}

Beyond the direct impact on microbial communities, a key objective of preharvest biocontrol is to maintain or enhance fruit yield and quality. In this study, while the treatments did not lead to statistically significant differences in total or marketable yield in peaches, a notable trend was the higher average fruit weight observed in the BC3 treatment. This suggests a potential indirect benefit of M. pulcherrima, possibly by reducing subclinical pathogenic pressure and allowing for better resource allocation to fruit development. In plums, the protective effect was more evident: the BC3 treatment resulted in a significantly lower percentage of discarded fruit compared to the untreated control. This finding directly correlates with the observed suppression of fungal populations, translating the microbiological effect into a tangible improvement in marketable quality. Crucially, these benefits were achieved without negatively impacting the total yield, demonstrating that the application of these yeasts is safe for the crop and beneficial for the final product quality.

The observed reduction in fungal colony size, especially under BC3, supports the hypothesis that antagonistic yeasts can inhibit hyphal extension and sporulation, thereby limiting pathogen virulence. ³² This effect is particularly valuable in organic systems, where sustainable alternatives are needed.^{33, 34} Moreover, the heatmap analysis revealed that BC3 consistently suppressed both dominant and minor fungal groups across seasons and fruit types, indicating a stable and broad‐spectrum biocontrol effect. This is consistent with the ecological capacity of antagonistic yeasts, which can shift microbial communities toward less pathogenic profiles. ²⁴ In summary, the field application of M. pulcherrima (BC3) demonstrated robust and consistent suppression of filamentous fungi that translated into improved fruit quality, confirming its potential as a preharvest biocontrol agent and highlighting the promise of integrating antagonistic yeasts into sustainable disease management programs for organic stone fruit production.

CONCLUSIONS

This study demonstrates that preharvest application of antagonistic yeasts can significantly reshape the epiphytic fungal community in organic stone fruit orchards without compromising yield and, in some cases, improving fruit quality. Among the tested biocontrol agents, M. pulcherrima (BC3) exhibited superior ecological fitness and persistence. This resulted in a marked reduction in both the abundance and colony size of filamentous fungi, including key pathogenic genera such as Alternaria, Cladosporium, and Penicillium. These effects were consistent across two growing seasons and were accompanied by a notable suppression of fungal groups associated with high pathogenic risk. In contrast, H. uvarum (BC4) showed limited and transient activity, highlighting the importance of strain selection for field efficacy.

The demonstrated ability of M. pulcherrima to colonize the epiphytic surfaces and exert broad‐spectrum antagonism underscores its potential as a cornerstone of sustainable disease management strategies in organic fruit production. By reducing fungal pressure during critical preharvest stages, this approach offers a viable alternative to copper‐based fungicides, aligning with regulatory frameworks and consumer demand for residue‐free produce. Future research should focus on optimizing formulation stability, application timing, and integration with other biocontrol agents to enhance robustness under variable field conditions.

Supporting information

Figure S1. Representative yeast colonies isolated from peach leaf samples under BC3 (Metschnikowia pulcherrima) treatment. (A) Native yeast community before application (third decimal dilution), dominated by Aureobasidium‐like morphologies. (B) Dominance of Metschnikowia pulcherrima colonies, characterized by their reddish‐brown pigmentation due to pulcherrimin production, after antagonistic yeast application (fourth decimal dilution).

Table S1. Climatic data of the Arroyo de San Serván station (2018 and 2019).

Table S2. Schedule of the six foliar applications performed on peach and plum orchards during the 2018 and 2019 growing seasons.

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.