Microencapsulation of Yeast Cells and Its Potential Usage as a Post-Harvest Biocontrol Agent for Citrus Storage

Abstract

In this study, yeasts isolated from citrus groves, trees and leaves were identified, phylogenetic analyzes were performed and their antifungal effects were determined. Wickerhamomyces anamolus (M72), Meyerozyma guilliermondii (M77), and Pichia kudriavzevii (M74) species were identified and were found to have antifungal effects against P. digitatum and P. italicum. Pichia kudriavzevii (M74), which has the highest antifungal effect, showed 67% and 62% inhibition rates against P. digitatum and P. italicum, respectively. An encapsulation study was carried out using a microencapsulation process to ensure that the M74 strain, which has the most antifungal effect, is long-lived enough to be a biopesticide. The optimum spray drying process parameters as well as the optimum concentration values of wall materials were investigated in the spray drying process for the microencapsulation of Pichia kudriavzevii (M74) through the Taguchi methodology. The formulation containing 0.1% sodium alginate (SA) and 10% corn starch (CS) showed a good performance in the inhibition of P. digitatum, a mold that causes losses in orange, thanks to its high percentage of viability (73%). The inhibition percentages may indicate that this formulation may be a candidate to be considered as a potential alternative application to synthetic fungicides on orange fruits for the effective control of P. digitatum mold.

1. Introduction

The homeland of citrus is China, Southeast Asia and India, and today it can be grown in almost all countries with subtropical climates. It is a plant community that includes fruit species such as citrus, sour orange, tangerine, grapefruit and lemon [1]. According to USDA data, the 2022/23 production data of citrus fruits, which are preferred in many industries such as the fruit juice industry, amounted to a total of 101 million tons in the world, including 48 million tons of oranges, 37 million tons of tangerines, 9 million tons of lemons and 7 million tons of grapefruits. A total of 47% of this production belongs to oranges, 37% to tangerines, 9% to lemons and 7% to grapefruits [2].

Turkey has very important potential in world citrus production. Citrus agriculture developed rapidly after the Republic, and production amounts have increased until today. The citrus varieties commercially produced in Turkey are orange, tangerine, lemon and grapefruit. In 2022, 4.7 million tons of citrus fruit were produced, and 40% of the production was tangerine, 28% lemon, 28% orange and 4% grapefruit [2]. One of the most important problems in citrus production is post-harvest losses. If necessary precautions are not taken, serious wastes ranging from 30% to 60% occur after 30–60 days of storage [3]. The most important factors causing post-harvest losses are phytopathogens, such as mold, bacteria and viruses [4,5,6]. Phytofungi, especially, such as Penicillium digitatum, Penicillium italicum, Geotrichum citri-aurantii, Aspergillus niger and Aspergillus flavus, cause significant decreases in quality and marketable yield. The most common species encountered after harvest is P. digitatum [7,8].

However, if appropriate and effective preservation is carried out, post-harvest losses can be significantly reduced. Increased public concern due to the negative effects of synthetic chemicals on food safety and the environment has created a need for new environmentally and health-friendly alternatives. The most interesting application of biological control methods is the use of antagonist species as biopesticides [9,10,11].

Supported by different policies, the biopesticide market grew six times between 2005 and 2016 [12]. It represented 6% of the global pesticide market in 2016, with a CAGR of 14.1%. The microbial pesticide market, which was addressed as 3.48 B USD in 2018, reached a volume of 7.37 M USD in 2023 [13]. Mycopesticides accounted for only 10% of the global biopesticide market in 2016 [12].

Microorganisms are potential biocontrol agents (BCAs), and combat pests directly or indirectly with the metabolites they synthesize. They may contain bacteria, fungi, viruses, mushrooms, insects, nematodes, protozoa, etc. [14]. Only a few rare species of fungi have been developed as BCAs, so the technology is still in its infancy. In addition to their high effectiveness against agricultural pests and diseases, commercially available BCAs also have some disadvantages, such as rapid deactivation when exposed to light, a short activity period and sensitivity to day and night heat changes [15]. The success of BCAs lies in a suitable formulation, which depends on the characteristics of the agricultural pest and disease agent, its relationship with the formulation components, storage conditions and the surface or location of application.

Imazalil is one of the most common mold fungicides currently in use. After the use of imazalil began in the 1980s, the first case of imazalil resistance was reported in 1987 [16]. Over the years, the use of imazalil and similar chemical pesticides has increased the population of resistant biotypes of pathogenic fungi [17]. It is envisaged that chemical pesticides will be completely banned within the scope of sustainability and green transformation frameworks due to their effects, such as developing resistance, leaving residues, containing microplastics [18] and harming beneficial species. This situation has increased the interest in biocontrol agents [19].

Encapsulation offers a great opportunity to ensure the effectiveness of natural preservatives and biocontrol agents, the bioavailability of their functional components, and the stability of the encapsulated material, thus increasing their use instead of synthetic agents [20]. For this reason, biopolymer-based microencapsulation processes and viability-preserving formulations have gained importance. The microencapsulation of microorganisms is a process in which microorganisms are coated or trapped in a matrix to protect them from adverse environmental conditions [21]. In the case of yeasts, the microencapsulation process allows them to withstand factors that reduce their viability, such as solar radiation, temperature, and relative humidity, reducing the doses and number of applications [22,23].

The aim of this study (Figure 1) is to characterize the Wickerhamomyces anomalus (M72), Meyerozyma guilliermondii (M77), and Pichia kudriavzevii (M74) isolated from citrus groves in the Mediterranean Region of Turkey by microencapsulating them with the spray dryer method, and to investigate its effectiveness as an in vitro and in vivo biocontrol agent against the citrus pathogenic mold P. digitatum. In this study, Pichia kudriavzevii was encapsulated for the first time using the spray dryer method, as far as we know. This method is cost-effective and one of the most suitable encapsulation methods at the industrial scale.

2. Materials and Method

2.1. Materials

Sodium alginate (SA), whey protein isolate (WPI), corn starch (CS) and malthodextrin (MDX) were obtained from Katkı Deposu (Istanbul, Turkey). Yeast extract, malt extract, peptone, glucose, agar and potato dextrose agar (PDA) were obtained from Merck (Darmstadt, Germany).

2.2. Isolation and Cultivation of the Yeast Cells

In this study, 120 yeast strains were isolated from citrus garden soils, trees, and moldy fruits in the Adana province of the Mediterranean Region of Turkey [21,22,23]. A total of 76 of the isolates consisted of ascomycetous strains and 44 were basidiomycetous strains. To obtain active cultures, yeasts were grown on yeast extract peptone dextrose agar (YPDA, 10 g/L, peptone 20 g/L, dextrose (glucose) 20 g/L) at 26 °C. P. digitatum, which causes green mold on orange fruits after harvest, was isolated from the same environments during the isolation of yeast. To obtain active culture, it was grown on potato dextrose agar (PDA; 4 g/L potato infusion, 20 g/L glucose and 15 g/L agar), respectively, for 14 days. Three of these yeast strains (Wickerhamomyces anomalus (M72), Meyerozyma guilliermondii (M77), Pichia kudriavzevii (M74)) with antifungal effects were selected and Pichia kudriavzevii (M74) was used for encapsulation.

2.2.1. Identification of Yeast Strains

The identification of isolated yeast strains was performed by DNA isolation, followed by PCR using ITS (internal transcribed spacer) primers. The primers used are ITS-F (5′-TCCGTAGGTGAACCTGCG-3′) and ITS-R (5′-TCCTCCGCTTATTGATATGC-3′), and the PCR conditions were as follows: the denaturation was set as 3 min at 94 °C, the annealing temperature was set as 45 s at 53 °C, the extension was 90 s at 72 °C, and the final extension was 10 min at 72 °C. The entire reaction was set for 30 cycles. PCR products were sequenced by the Sanger Sequence method.

2.2.2. Selection of Antagonistic Yeasts Capable of Inhibiting Fungal Pathogens Causing Post-Harvest Diseases

By dual-cultivating a yeast and a fungal pathogen on a PDA plate, the antagonistic activities of 120 yeast strains from citrus groves and trees in the Turkish Mediterranean Region that cause post-harvest illnesses in oranges were assessed. The antagonistic activity was carried out as described by Sukmawati et al., [24] with slight modifications. The PDA medium was inoculated by linear streaking with a colony of active yeast cells approximately 3 cm from one edge of the plate. After incubation of the plate for 48 h at 26 °C, a 4 mm diameter disk of an actively growing fungal pathogen (Penicillium italicum and Penicillium digitatum) was inoculated onto the opposite edge of the plate, approximately 3 cm from the yeast streak. Ten days of 26 °C incubation were spent on the plates. A plate that had only been infected with the fungus was used as the control. For every treatment, three duplicates were carried out. A zone of inhibition indicates the presence of antagonistic activity. Using the formula {(R1 − R2)/R1} × 100, the % inhibition of mycelium growth was determined. The fungal mycelium growth radius in the control treatment was measured as R1, and the mycelium growth radius was measured as R2.

2.2.3. Capacity of the Yeast Isolates Chosen as Potential BCAs for Production of Volatile Organic Compounds

The antifungal activity of the volatile organic compounds (VOCs) produced by yeast isolates against P. digitatum and P. italicum was tested using a twofold Petri dish experiment, according to the protocol [25,26]. PDA dishes containing only Penicillium species were utilized as a control. Three replicate plates were utilized for each treatment, and the experiment was repeated twice to measure the radial growth inhibition of the test fungi.

2.2.4. Hydrolitic Enzymes

β-1,3-glucanase, protease and chitinase were among the extracellular hydrolytic enzymes whose production was evaluated by the spot inoculation of candidate BCAs on appropriate media: the substrates of these enzymes were chitin, β-glucan (from barley) and skim milk, respectively, in the Petri dishes [26]. Using previously published methods, aliquots (10 μL) of 24-h yeast culture suspensions (10⁷ cells/mL) were superficially sprinkled onto solid medium (agar 15 g/L) containing the matching substrates. On a solid medium containing a 5 g/L laminarin and 6.7 g/L yeast nitrogen base, the activity of β-1,3-glucanase was evaluated. Plates were incubated for 72 h at 25 °C, stained with Congo Red (0.6 g/L), and then allowed to stand at room temperature for ninety minutes. A yellow-orange zone surrounding the colonies on the plates indicated the presence of glucan hydrolysis after the unabsorbed blot was decanted. Protease activity was assessed using 2% agar and 10% skim milk powder. After five minutes of autoclaving at 110 °C, the medium was transferred onto Petri dishes. For a week, inoculated plates were kept at 25 °C and checked every day. The presence of a bright halo around the inoculation site indicated the presence of enzymatic activity. Chitinase activity was evaluated on chitin at pH 7. After 1–6 days at 25 °C, the existence of a clear zone surrounding the inoculum site served as a daily indicator of extracellular chitinase activity. For every isolate, three duplicate dishes were used. The colony’s halo surrounding its development was seen as a positive result.

2.2.5. Biofilm Formation

Using the dye crystal violet, the microtiter dish assay was used to assess the biofilm development by the yeast isolates chosen as potential BCAs, as previously reported [25,26]. Yeast strains were grown at 10⁷ cells/mL and inoculated in triplicate into the wells of a 96-well polystyrene plate, and the plates were incubated at 75 rpm at 26 °C for up to 72 h. After incubation, the wells were washed with PBS and stained with 100 μL of 0.4% aqueous crystal violet solution for 45 min. Then, the wells were washed with sterile distilled water (SDW) and immediately destained with 200 μL 95% ethanol. After 45 min, the amount of crystal violet in the solution was measured at 590 nm. Results are given in OD.

2.3. Microencapsulation of Yeast Cells

2.3.1. Preparation of Polymer Solution and Spray Drying Process

To extract nutritional components from the starting medium, cells from the yeast culture were recovered by centrifugation at 6000× g for 10 min. They were then twice washed with sterile 0.1% (w/v) peptone water. Considering previous studies and the properties of biopolymers, different polymer concentrations (0.1–10% w/v) and their combinations were subjected to preliminary tests to determine whether they could be successfully spray dried. SA was stirred until completely dissolved in sterile distilled water at 25 °C, and then polymers (MDX, starch, WPI) were added in the amount corresponding to the determined ratios and stirred for 2 h until a homogeneous solution was obtained. The yeast concentration was adjusted to 1 × 10⁹ cells/mL and added to the polymer solution.

The Taguchi approach, prepared with Minitab Statistical Software Version 20.4 was used to optimize the composition of the bioformulation by the spray dryer process. The sodium alginate (SA) concentrations selected for the Taguchi approach of biopolymers were 0–0.1% (w/v), and the whey protein isolate (WPI), corn starch (CS) and malthodextrin (MDX) concentrations were 0–10% (w/v). The response variables for the experimental design were percentage survival (%) and yield of spray dryer powder produced. A total of 14 experimental studies were generated with the Taguchi approach, and each treatment was repeated three times.

2.3.2. Morphological Analysis of Spray-Dried Powder via Optical Microscopy and SEM

Optical microscopy images were obtained using an optical microscope (CKX41; Olympus Corporation, Tokyo, Japan) to examine the overall distribution of powder samples obtained by spray drying. The powder’s surface morphology was examined using Scanning Electron Microscopy. Powder samples obtained by spray drying were fixed to aluminum stubs using double-sided adhesive tape. Excess dust was removed by directing dry air to the surface of the stubs. Samples were coated with gold on a SCD 005 sputter coater (Darmstadt, Germany) and were examined at different magnifications with a Zeiss Gemini 500 Field Emission Scanning Electron Microscope (Darmstadt, Germany) operating at an accelerating voltage of 10 kV.

2.3.3. Particle Dispersion and Size Analysis

Spray-dried powder (0.5–1 g) was mixed with ethanol, and a particle size analyzer (Mastersizer-2000, Malvern Instruments Ltd., Malvern, UK) was used to measure the diameters of the particles. A true diffraction index of 1.41 for protein and 1.33 for water was used for all measurements. Every sample was measured three times. The dv90 mean values are given.

2.3.4. Viability Assay after Spray Drying

The viable yeast cell counts after the spray drying process was determined by plating on agar plates. For this, a 1 g sample of spray-dried yeast powder was rehydrated in 9 mL of 0.1% (w/v) peptone water, vortexed for 1 min and left to dissolve for about 15 min. Serial dilutions were carried out, followed by plating on YPD agar and incubation for 48 h (26 °C). Viable yeast cells were calculated using the following formula (colony number in a plate × dilution factor/plated sample volume) [27], and the result was reported as viable yeast cells per 1 g of spray dryer product. The same was performed by diluting 1 mL of the spray dryer feed suspension to calculate percentage of viability. The experiment was carried out in triplicate. The percentage of the viability was calculated according to the following formula: % viability = Nr/Nf × 100, where Nr means the number of viable cells in spray dryer product and Nf means the number of viable cells in the feed before drying [28].

2.3.5. Antifungal Analysis of Spray-Dried Yeast Cell

The agar pour method was performed as described by Virgili R et al. [29] with modifications. This was carried out to determine the minimum inhibitory concentration (MIC) of yeast strains against the fungal strain (yeast cell/mL required to inhibit fungal growth). For this, a 1 g sample of spray-dried yeast powder was rehydrated in 9 mL of 0.1% peptone water, vortexed for 1 min and left to dissolve for about 15 min. The yeast suspension was added to Petri dishes in triplicates and mixed with malt extract agar (45–50 °C) to obtain final cell concentrations from 1 × 10⁰ up to 1 × 10⁸ cell/mL. After agar solidification, 10 µL of spore suspension (1 × 10⁶ spores/mL), prepared from 10-day old Penicillium italicum and Penicillium digitatum grown on malt extract agar, was added to the center of each dish. The dishes were incubated for 7 days (26 °C). The radius of the fungus growth was measured and the inhibition percentage was calculated. MIC was determined as the cell concentration which inhibited fungi growth.

2.3.6. Cell Viability (MTT Assay)

MTT analysis was performed to evaluate whether the mitochondrial activities of the yeasts continued after encapsulation and to examine their vitality. The viability of the spray-dried yeast was evaluated using MTT assay as described by dos Reis Almeida et al. with slight modifications [30]. For this, 1 g sample of spray-dried yeast powder (10th, 7th and 4th formulation) was rehydrated in 9 mL of 0.1% peptone water, vortexed for 1 min and left to dissolve for about 15 min. Centrifugation was carried out at 6000 rpm for 10 min and the supernatant was discarded and replaced with equal amount of YPD broth, followed by overnight incubation in rotary shaker (26 °C 120 rpm) to allow for cell revival. After that, 100 µL of yeast sample was added to 96-well plate, followed by the addition 10 µL of (5 mg/mL) MTT solution and incubation in the dark at 26 °C for 4 h. The suspension was discarded without disturbing formazan crystals. Finally, 100 µL of DMSO was added and incubated for 10 min in a rotary shaker at 26 °C, and the absorbance was measured at 595 nm using a plate reader (BioTek Epoch 2 Microplate Reader, Santa Clara, CA, USA). As a control group, free encapsulated M74 yeast was used.

2.4. Evaluation of the Efficacy of Antagonistic Yeasts in Controlling the Post-Harvest Diseases of Oranges

Orange fruits were used to test the in vivo activity of microencapsulated yeasts against Penicillium digitatum. Oranges (Citrus sinensis) were chosen at random for the project from a nearby grocery store, taking into account factors like age, homogeneity, and condition without defects. After two minutes of disinfection with a 2% (w/v) sodium hypochlorite (NaOCl) solution, the oranges were rinsed twice with clean water and allowed to air dry at 25 °C.

To observe the preventive effect on fruits, free yeasts (adjusted to 1 × 10⁸ cell/mL using a hematocytometer) were suspended and a dipping solution was prepared after dissolving the encapsulated yeast. This was done by dissolving the required amount of the spray dryer product in 0.1% (w/v) peptone water to obtain a final concentration of yeast cells of 1 × 10⁸ cell/mL. The cell count present in 1 g of spray dryer product was determined in the vitality analysis. First, the fruits were immersed in the yeast suspension (1 × 10⁶ cells/mL) and the fruit was left to dry for 1 h at room temperature. Then, the fruits were sprayed with the mold suspension so that they were completely covered. Fruits atomized with sterile distilled water comprised the negative control, whereas the fruits soaked in Imazalil^® (Azoxystrobin, 0.8 g/L; AMISTAR^® Syngenta, Frankfurt, Germany) constituted the positive control. Spore suspension was used to inoculate both treatments.

Storage condition was as follows: 25 °C (75% RH) for 21 days, and the disease incidence was calculated by measuring the pathogen growth [31]. Each group consists of 6 fruits, and imazalil-treated fruits were used as the positive control group and only pathogen-treated groups were used as the negative control. The experiment was repeated three times and it consisted of three replicates per treatment, each replicate containing ten orange fruits.

2.5. Statistical Analysis

Every experiment was conducted in triplicate, and the mean ± standard deviation was reported for each. The statistical study was carried out using IBM version 20 of SPSS Statistics Software, located in Armonk, NY, USA. Tukey’s post hoc test was conducted after a one-way analysis of variance (ANOVA) was used to assess the differences between the mean values. If p was less than 0.05, differences were determined to be significant.

3. Result and Discussion

3.1. Selection of Antagonistic Yeasts Capable of Inhibiting Fungal Pathogens Causing Post-Harvest Diseases

In this study, 3 out of a total of 120 yeast and bacterial isolates obtained from the peels of citrus fruits in Turkey were selected due to their high antagonistic antifungal ability against P. digitatum and P. italicum, the causative agents of green and blue molds. The reason for selecting candidate BCAs from the microorganism populations in Turkey living on the peel of citrus fruit was the assumption that endemic isolates would naturally be better adapted to local environmental conditions and would provide a better performance as BCAs [26]. Three yeast strains (Wickerhamomyces anamolus (M72), Meyerozyma guilliermondii (M77), Pichia kudriavzevii (M74) inhibited the growth of P. italicum, which causes fruit rot disease in oranges, by 3%, 43% and 62%, respectively, and P. digitatum by 20%, 38% and 67%, respectively (Figure 2). Pichia kudriavzevii (M74) was the yeast strain that showed the most antifungal effect against both fungal species. Therefore, this strain was chosen for encapsulation. The results of previous studies revealed that some yeast strains have been reported to inhibit the growth of these two fungal pathogens. For example, Delali et al. [32] showed in their study that Pichia kudriavzevii, among the many yeast species they isolated from kimchi, had the most effective antifungal effect against green mold on citrus. Strains M72, M74, and M77 appear to produce different levels of antifungal compounds due to the variety of components and effects they possess (Figure 3). These antifungal compounds include aldehydes, esters, ketones, alcohols, organic acids, etc. [32,33]. Delali et al. [32] determined the antagonist abilities of some yeast species they isolated in their study, such as Metschnikowia fructicola, Pichia kudriavzevii, Kluyveromyces marxianus and Yarrowia lipolytica, against green mold on citrus. Hammami et al. [26], in their study with more than 180 species of bacteria and swimsuit isolates, reported that Candida oleophila and Debaryomyces hansenii isolates showed antagonistic effects against P. italicum and P. digitatum in citrus.

3.2. Isolation and Cultivation of the Yeast Cell

The Neighbor-Joining approach was used to infer the evolutionary history [34]. Next to the branches are the proportion of duplicate trees where the related taxa are clustered together in the bootstrap test [35].

The Neighbor-Joining approach was used to infer the evolutionary history [34]. The ideal tree is displayed (beside the limbs). The evolutionary distances are expressed in base substitutions per site and were calculated using the Maximum Composite Likelihood approach [36]. There were 11 nucleotide sequences in this investigation. For every pair of sequences, all the unclear locations were eliminated (pairwise deletion option). The final dataset contained 1862 locations in total. In MEGA11, evolutionary analyses were carried out [37]. The ITS and 18S rDNA sequences were used to identify the yeast isolates, respectively (Figure 3 and

Table 1. Molecular identification of isolated yeast strains by BLAST algorithm compared to the genomic sequences in the GenBank database. The ingroup represents the isolated yeast strains and the outgroup represents the closely related species. The strains listed by accession numbers correspond to a sequence found in GenBank.

	Isolate Code	Source	Culture	Accession Number	Identity (%)
Ingroup	M72	Orange (Citrus sinensis)	Wickerhamomyces anomalus	NR_111210.1	98.72%
				NR_155000.1	91.73%
				NR_073352.1	91.56%
				NR_138219.1	90.98%
				NR_111798.1	90.89%
	M74	Orange (Citrus sinensis)	Pichia kudriavzevii	NR_131315.1	99.55%
				NR_153293.1	98.35%
				NR_168173.1	98.22%
				NR_077085.1	97.65%
				NR_111358.1	97.63%
	M77	Orange (Citrus sinensis)	Meyerozyma guilliermondii	NR_111247.1	99.63%
				NR_152984.1	98.46%
				NR_149348.1	98.42%
				NR_111339.1	97.37%
				NR_111340.1	96.25%
Outgroup	-	-	Schizosaccharomyces pombe MUCL 30245	NG_070697.1	-
	-	-	Saccharomyces boulardii voucher URCS6	KT000037.1	-

). The outcomes demonstrated that the M72 yeast isolate was closely linked to Wickerhamomyces anomalus CBS5759 with 98.72% similarity, the M74 isolate was closely related to Pichia kudriavzevii ATCC6258 with 99.55% similarity, and the M77 yeast isolate was closely related to Meyerozyma guilliermondii CBS2030 with 99.63% similarity.

3.3. The Mechanism of Effect That Yeast Isolates Chose as a Potential BCA Function

Yeasts produce a number of antifungal protein compounds that act as antifungal hydrolytic enzymes, such as chitinase, glucanase, and proteases [26]. The antifungal activity of candidate BCAs has been linked to the production of at least two lytic enzymes, specifically chitinase, protease, and glucanase, by all of the chosen yeast isolates [11,38]. The primary constituents of the cell walls of the majority of the fungi are chitin and β-1,3-glucan, the substrates of the enzymes chitinase and β-1,3-glucanase. When taken separately, pure β-1,3-glucanases and chitinases inhibit certain fungi, but not the majority of them. On agar plates or in liquid media, combinations of the two enzymes are known to suppress a wide variety of saprophytic and pathogenic fungus [39]. In this study, it was observed that three isolates, M72 Wickerhamomyces anomalus, M74 Pichia kudriavzevii and M77, had chitinase and β-1,3-glucanase activities (

Table 2. Hydrolytic enzyme activity and biofilm forming capacity of selected yeasts isolates as candidate BCAs.

Isolate Code	Culture	Enzymatic Activity			Biofilm Capacity	VOCs Inhibition Rate (%)
		Chitinase	β-1,3-Glucanase	Protease	Optical Density (OD)	P. digitatum	P. italicum
M72	Wickerhamomyces anomalus	+	+	−	0.45 ± 0.12	38 ± 1.21	27 ± 1.13
M74	Pichia kudriavzevii	+	+	+	0.51 ± 0.09	55 ± 2.23	46 ± 2.14
M77	Meyerozyma guilliermondii	+	+	−	0.48 ± 0.05	35 ± 1.87	21 ± 1.01

The ability of yeast and bacterial cells to adhere to the polystyrene dishes is directly connected with OD values. VOCs: volatile organic compounds.

).

The generation of antifungal volatile organic compounds (VOCs) is another mechanism that BCAs commonly use to hinder mycelium growth. This is in line with other findings by different writers that suggest certain yeast species generate volatile organic compounds (VOCs) that have antifungal properties. Zhang et al. [40] showed that VOCs produced by Pichia spp. inhibit fungi emerging in the fermentation process. Similarly, Delali et al. [32] examined the effects of three yeast species isolated from kimchi on P. digitatum by screening them for extracellular lytic enzyme activity and evaluating their volatile organic compounds (VOCs). They reported that Pichia kudriavzevii reduced the incidence of green mold caused by Penicillium digitatum through biofilm formation, nutrient competition, and volatile release.

The capacity to generate biofilms is another method of action that could enhance the efficacy of bacteria and yeasts as BCAs [41]. All three of the yeast isolates selected in this study (M72, M74, and M77) showed a high capacity to form biofilms. This could account for the three yeast strains’ antagonistic activity in vitro—particularly M74, which had the strongest activity—against P. italicum and P. digitatum, as well as their ability to effectively inhibit Penicillium rot on experimentally inoculated citrus fruits. Liu et al. [42] observed a comparable relationship between citrus fruits’ ability to develop biofilms and their efficacy in avoiding Penicillium rot.

3.4. Microencapsulation of Yeast Cells

The highest experimental values of the viability percentage of microencapsulated M74 yeast were 71.42 ± 1.74% (

Table 3. Powder formulations determined by Taguchi design, and the powder quality, viability, and particle size values of the spray-dried encapsulation formulations of M74 yeast.

Samples	SA (w/v, %)	CS (w/v, %)	MDX (w/v, %)	WPI (w/v, %)	Powder Quality (0–5 Range)	Viability (%)	Particle Size (Dv90, µm)
1	0	5	0	0	3	36.78 ± 1.14	32.41 ± 0.22
2	0	0	5	0	2	10.32 ± 0.98	26.69 ± 0.34
3	0	0	0	5	2	9.41 ± 0.79	22.65 ± 0.12
4	0	10	0	0	3	54.41 ± 1.17	36.78 ± 0.18
5	0	0	10	0	2	21.41 ± 0.74	41.45 ± 0.26
6	0	0	0	10	2	11.69 ± 0.09	35.74 ± 0.31
7	0.1	5	0	0	3	65.48 ± 1.19	35.62 ± 0.19
8	0.1	0	5	0	3	35.42 ± 0.67	24.71 ± 0.09
9	0.1	0	0	5	2	16.35 ± 0.08	18.59 ± 0.11
10	0.1	10	0	0	4	71.42 ± 1.74	37.54 ± 0.23
11	0.1	0	10	0	3	22.65 ± 0.12	38.46 ± 0.25
12	0.1	0	0	10	3	14.28 ± 0.10	22.45 ± 0.08

SA: Sodium alginate, CS: Corn starch, MDX: Maltodextrin and WPI: Whey protein isolate.

) in the 10th formulation, containing 0.1% (w/v) SA and 10% (w/v) CS. As a result of the preliminary studies, the spray dryer parameters were determined as 90 °C inlet temperature, 10 mL/min flow rate and 70 °C outlet temperature. The exposure of microorganisms to high temperatures causes cell membrane damage, the denaturation of proteins and enzymes, and causes cell death [43]. Therefore, cell viability was determined through process optimization and formulation optimization. Huang et al. [44] created microspheres by immobilizing Pichia kudriavzevii in a gel network cross-linked with chitosan and sodium alginate, and found an encapsulation efficiency in the range of 42–74%. Martins et al. [45], microencapsulated the yeasts Saccharomyces cerevisiae CCMA 0543, Torulaspora delbrueckii CCMA 0684 and Meyerozyma caribbica CCMA 1738, employing a spray dryer and wall components consisting of whey powder (WP), high maltose (MA), and maltodextrin DE10 (MD). They found that the use of whey protein as a wall material protected yeast viability more than the use of MDX for all three yeast strains. Fundamental structural and functional changes occur in practically all organelles and cellular components during the drying, dehydration, and rehydration of yeast using a spray drier, and can result in cell death [46]. The addition of certain ingredients with protective properties, such as maltodextrin, starch and whey powder, used to improve cell protection, may have led to greater viability by stabilizing parts of the cell membrane [45].

Table 3 displays the values of particle size for the samples of dry powder. The table illustrates how the kind of carrier material utilized greatly affects the size of the encapsulated yeast cells’ particles. Dried powders range in particle size from 20 to 40 μm. The particle size of the formulation containing SA and SC was measured as 37.54 ± 0.23. The distinct film-forming and gelling capabilities of the wall materials employed for encapsulation can be used to explain the discrepancies in particle size and viability across the samples of yeast encapsulation dried with various carrier materials. Some studies in the literature state that sodium alginate/corn starch combinations enable the formation of much more uniform and smaller-sized capsules [47].

Scanning electron microscope (SEM) images of the 10th microcapsule formulation produced under optimum conditions showed slightly rough surfaces (Figure 4(B1–B3)) but no fractures or cracks. It was also confirmed by the optical microscope and SEM images that the microcapsules had a distribution ranging from 30 to 40 µm, in parallel with the size analysis. Similar morphological features were also confirmed in studies using yeast encapsulation with a spray dryer [48,49].

In the MTT results (Figure 5), the MTT results of the 10th, 7th, and 4th formulations were found to be 85, 68, and 59%, respectively. MTT analysis was performed for spray dryer formulations with a % encapsulation efficiency of more than 50%. The combination with the highest cell viability is number 10. The microencapsulation of Pichia kudriavzevii (M74) under optimum conditions by spray drying was effective against P. digitatum, which causes disease in orange fruit (Figure 6). The 10th formulation containing M74 yeast showed a 74% inhibition rate at 5 log cfu/mL. For in vivo fruit trials, the 10th formulation with the highest cell viability and vitality was chosen to be used.

3.5. Evaluation of the Efficacy of Antagonistic Yeasts in Controlling the Post-Harvest Diseases of Oranges

Since green mold is more prevalent in ambient temperatures, it usually results in higher losses during commercial operations. Blue mold (P. italicum) grows faster than P. digitatum in cold storage conditions below 10 °C [50]. Since ambient temperature conditions are more important in the commercialization of citrus, P. digitatum (green mold) was chosen as the model organism in the study. Furthermore, the application of microbial antagonists like BCAs offers the benefit of being a safe, environmentally friendly, and toxicologically sound method of managing diseases in fruit crops [26]. After 21 days of formulation, only 32% of the oranges were moldy compared to the unencapsulated yeast (98%). This can be explained by the fact that free yeasts do not remain alive for 21 days, and microencapsulated yeasts remain alive and can be released and have an effect over time. With encapsulation, a 68% reduction in disease incidence was achieved in the oranges compared to the control. Lima et al. [51] reported that the encapsulation of M. guilliermondii reduced the severity of papaya anthracnose by 50% (Figure 7).

The M74 isolate selected in this study showed a high in vitro inhibitory activity on the mycelial growth of P. digitatum, and was found to be effective by encapsulation in preventing Penicillium digitatum mold in orange fruits stored at ambient temperature (25 °C). The fact that the antifungal activity of encapsulated M74 is higher than that of free M74 may be due to the fact that the encapsulation system enabled the yeast to maintain its viability and continue to show activity during the storage period. Grambusch et al. [52] encapsulated Saccharomyces spp. with a spray dryer and showed that yeast viability could remain up to 75% for up to 90 days. It is also possible that encapsulation increases the antifungal activity of the bioactive component it contains by interacting with the cell wall of pathogens such as molds and bacteria, depending on the type of wall material. In parallel with this, Erarslan et al. [20] encapsulated sage essential oil and showed that its antifungal activity against Aspergillus niger and Botrytis cinerea could be significantly increased after encapsulation into PVA/Chi NPs.

The results of this study may indicate that the antifungal activity of the M74 yeast isolate selected in this study depends on more than one mode of action. In particular, the inhibition shown by free and encapsulated M74 against P. digitatum in in vitro and in vivo experiments may indicate that it is due to extracellular diffusible metabolites, such as hydrolytic enzymes, VOCs and biofilm-forming capacity, and thus may be a candidate BCA.

4. Conclusions

In this study, yeasts isolated from citrus gardens, trees and leaves were identified, phylogenetic analyses were performed, and their antifungal effects were determined. The present study allowed us to obtain optimum spray drying process parameters, as well as the optimum concentration values of polymers in the spray drying process for the microencapsulation of Pichia kudriavzevii (M74) through the Taguchi methodology. The formulation of Pichia kudriavzevii (M74), containing 0.1% SA and 10% CS, showed a good performance in the inhibition of P. digitatum, a mold that causes loss in orange, thanks to its high percentage of viability (73%). The inhibition percentages may be sufficient to consider this formulation as a potential alternative application to synthetic fungicides on orange fruits for the effective control of P. digitatum mold. In further studies, field studies on orange trees and experiments on other fruits can be conducted.