Entomopathogenic Fungi in Peri-Urban Green Spaces: A Reservoir for Seasonal Biological Control of Insect Pests

Abstract

Peri-urban ecosystems represent underexplored habitats rich in entomopathogenic fungi (EPF) that can serve as valuable resources for managing insect pests. This study characterized the EPF communities in two peri-urban sites near Patras, Greece (Dasyllio and Elos), during 2018–2019. Soil samples were collected seasonally, and fungi were isolated using insect baiting with Tribolium confusum Jacquelin du Val and Sitophilus zeamais Motsch., a selective method favoring generalist, fast-acting entomopathogens. A total of 814 isolates were recovered. Of a randomly selected subset (n = 177) subjected to molecular identification, 46.9% were characterized as known EPF, while 53.1% were classified as putative EPF based on taxonomic affiliation (ITS sequence similarity ≥ 99%), pending confirmation of pathogenicity. The Dasyllio site yielded more isolates (63.4%) than Elos (36.6%). Seasonal trends strongly influenced EPF occurrence, with infective fungi peaking in spring and summer (p < 0.001), while community diversity remained stable throughout the year, with the highest evenness (Evenness Index = 0.93) observed in autumn. These results highlight peri-urban green spaces as reservoirs of diverse and ecologically stable EPF, suggesting their potential as sources of biocontrol agents for future development and seasonal integration into pest management strategies.

1. Introduction

Sustainable agroforestry and urban ecosystems increasingly rely on alternatives to synthetic pesticides, with entomopathogenic fungi (EPF) offering an environmentally benign option for insect pest regulation [1,2]. Their host specificity and minimal impact on non-target organisms make them ideal components of Integrated Pest Management (IPM) programs [3,4].

Peri-urban ecosystems, situated at the interface of urban and rural landscapes, are increasingly recognized as biodiversity hotspots and ecological buffers. Their intermediate disturbance regimes, heterogeneous vegetation, and reduced agrochemical inputs may foster diverse and stable EPF communities compared to intensively managed agricultural soils [5,6]. Moreover, their proximity to agricultural zones enhances their relevance as potential reservoirs of biocontrol agents.

The soil serves as the primary ecological reservoir for most EPF species. Their persistence and distribution are influenced by a complex of abiotic and biotic factors, including soil type [7], microclimate, agricultural practices [8,9], and the plant community [10,11]. Among the most studied and commercially exploited EPF are genera like Beauveria and Metarhizium, renowned for their broad host range and environmental adaptability [12,13].

In Greece, EPF have been identified in agricultural settings, such as Beauveria. bassiana (Balsamo-Crivelli) Vuillemin and Metarhizium Anisopliae (Metschnikoff) Sorokin in Crete and Western Greece [14,15]. However, the potential of EPF communities in peri-urban ecosystems—transitional zones between urban and rural areas—remains unexplored. These habitats may host unique, locally adapted fungal strains that could be developed into effective biocontrol agents [4,14].

A critical factor for the successful deployment of EPF is understanding their seasonal dynamics, as temperature, humidity, and host availability fluctuate throughout the year [16,17]. These environmental variables directly influence EPF sporulation, infectivity, and persistence [18,19], making seasonal timing essential for optimizing biocontrol applications. However, data on the seasonal distribution of EPF in Mediterranean peri-urban contexts remain notably scarce.

This study aims to investigate the spatial and seasonal dynamics of EPF in two peri-urban habitats in Greece to inform seasonally optimized biocontrol strategies. Specifically, we assessed the influence of location, host insect species, and season on the isolation frequency, community composition, and diversity of these fungi to inform the development of seasonally targeted biocontrol strategies. Our objectives were to: (1) isolate and identify EPF using insect baiting, (2) compare the fungal communities between two different sites and two bait insect species, and (3) analyze the seasonal variation in EPF occurrence and diversity to inform the timing of biocontrol applications. Thus, there is a significant knowledge gap regarding the spatial and seasonal dynamics of EPF in peri-urban Mediterranean ecosystems, limiting the development of locally adapted, seasonally optimized biocontrol strategies.

2. Materials and Methods

2.1. Study Sites and Soil Sampling

Seasonal soil sampling was conducted over two consecutive years (2018–2019) at two peri-urban sites in the municipality of Patras, Western Greece: Dasyllio (38°15′12″ N, 21°44′18″ E) and Elos (38°18′05″ N, 21°48′22″ E). Both sites are characterized as transitional green spaces with mixed vegetation (herbaceous plants, shrubs, scattered trees) and are situated on clay-loam soils, with no recent history of pesticide application or tillage, and are minimally disturbed by anthropogenic activity. A stratified random sampling design was employed to ensure spatial representativeness within each site [20,21]. At each location, 22 fixed sampling points were established using a random coordinate generator within a 1-hectare area. The same points were revisited during each sampling season to allow for temporal comparison. Sampling was conducted four times per year, corresponding to the meteorological seasons: winter (December–February), spring (March–May), summer (June–August), and autumn (September–November).

At each sampling point, approximately 1 kg of soil was collected from the top 0–10 cm layer using a sterilized stainless-steel soil corer (5 cm diameter). The soil was immediately transferred into sterile polyethylene bags, labeled, and placed in a portable cooler at 4 °C for transport to the laboratory. All samples were processed within 24 h of collection to preserve fungal viability.

2.2. Insect Baiting and Fungal Isolation

Two insect species were utilized as baits: the confused flour beetle, Tribolium confusum (Coleoptera: Tenebrionidae), and the maize weevil, Sitophilus zeamais (Coleoptera: Curculionidae). These generalist stored-product pests were selected for their proven sensitivity to a wide range of soil-borne EPF and methodological reliability in baiting assays [15,22,23,24]. Insect colonies were maintained under controlled laboratory conditions (25 ± 1 °C, 60 ± 5% RH, 16:8 h L:D photoperiod) on sterilized diets, following established rearing protocols [22,23,24]. Healthy adults (7–14 days post-emergence) were used for baiting. Prior to baiting, soil samples were air-dried at room temperature (≈22 °C) for 48 h to reduce moisture and minimize interference from entomopathogenic nematodes [15]. While this may affect some moisture-sensitive fungi, previous studies indicate that air-drying does not significantly impair the recovery of common EPF, such as Beauveria and Metarhizium [15,23]. Each sample was homogenized and sieved through a 2 mm mesh.

For each soil sample, ten groups of ten adults per insect species (totaling 100 individuals per species) were placed on the soil surface in sterile 90 mm Petri dishes. Dishes were sealed with Parafilm M (Amcor, Neenah, WI, USA) and gently inverted every 48 h to ensure uniform insect–soil contact. All samples were incubated in complete darkness at 25 ± 1 °C for 14 days. Mortality was recorded daily; insects were considered dead if unresponsive to gentle prodding [15,23]. Dead insects were surface-sterilized by immersion in 6% sodium hypochlorite for 3 s—a standard exposure time shown to effectively surface-sterilize without significantly reducing recovery of cuticle-colonizing fungi [14,15]—followed by three rinses in sterile distilled water. Dead insects were placed on moistened filter paper in Petri dishes and incubated at 25 ± 1 °C in darkness. They were examined daily under a stereomicroscope Zeiss Stemi 2000 (Carl Zeiss Microscopy GmbH, Jena, Germany) for the emergence of fungal structures [14,15]. Fungal growth from cadavers was transferred to Sabouraud Dextrose Agar (Condalab, Torrejón de Ardoz, Madrid, Spain) plates supplemented with 0.1% streptomycin sulfate. Plates were incubated at 25 ± 1 °C for 7–14 days. Pure cultures were obtained through repeated subculturing and preserved in 20% glycerol at −80 °C [14].

2.3. Molecular Identification

Given the high number of recovered isolates (n = 814), a subset of 177 isolates was chosen for molecular identification. This subsampling strategy provides a cost-effective, statistically robust characterization of the taxonomic breadth of culturable fungi isolated via insect baiting [15,16]. To ensure representativeness, isolates were initially grouped into distinct morphotypes based on macroscopic and microscopic characteristics, and representatives from each major morphotype were included in the subset. Genomic DNA was extracted from 7-day-old cultures using a modified CTAB protocol [25]. Approximately 100 mg of mycelium was ground in liquid nitrogen, incubated in CTAB buffer at 65 °C for 60 min, and purified with chloroform-isoamyl alcohol. DNA was precipitated with isopropanol, washed, and resuspended in TE buffer.

The internal transcribed spacer (ITS) region was amplified using universal fungal primers ITS5 and ITS4 [25]. Each 25 µL PCR reaction contained Taq 2X Master Mix (New England Biolabs, Ipswich, MA, USA), 0.4 µM of each primer, and approximately 20 ng of template DNA. Cycling conditions were: 95 °C for 5 min; 33 cycles of 95 °C for 30 s, 55 °C for 40 s, 68 °C for 1 min; and final extension at 68 °C for 5 min. PCR products were visualized on 1% agarose gels, purified, and bidirectionally sequenced by Eurofins Genomics (Ebersberg, Germany). Sequences were assembled and edited using Geneious Prime 2024.0.3 and compared against the NCBI GenBank database via BLASTn algorithm (National Center for Biotechnology Information, Bethesda, MD, USA). Species-level identification was assigned based on ≥99% sequence identity and ≥95% query coverage [26]. Isolates were classified into three categories based on taxonomic affiliation and documented ecological function: Recorded EPF: Species belonging to genera with well-documented and widespread entomopathogenic activity (Beauveria, Metarhizium, Cordyceps, Purpureocillium). Putative EPF: Species from genera where some members have reported entomopathogenic activity, but the specific species’ pathogenicity is less established or requires confirmation. Other Fungi: Primarily saprophytic, plant-associated, or soil-borne species with no well-documented entomopathogenic activity. The remaining isolates that were not sequenced (n = 637) were categorized as “Unidentified Fungi.”

2.4. Statistical Analysis

All analyses were conducted in Python 3.11 using pandas v2.2.3, SciPy v1.13.1, statsmodels v0.12.2, and matplotlib v3.9.1. Descriptive statistics were used to summarize isolation frequencies. Associations between categorical variables (site, host, season) and isolation categories were tested using chi-square or Fisher’s exact tests. Alpha diversity indices (Shannon H’, Simpson D, Pielou’s Evenness J’) were calculated for the molecularly characterized fungal community (n = 177 isolates) per season, site, and host. The Kruskal–Wallis test was used for multi-group comparisons, followed by pairwise Mann–Whitney U tests with Bonferroni correction. Two-group comparisons used the Mann–Whitney U test. Statistical significance was set at α = 0.05.

3. Results

3.1. Overall Isolation Success and Geographical Distribution

A total of 814 fungal isolates were recovered from the two peri-urban sites over the two-year study period. With equal sampling effort across sites (22 points per site per season), the Dasyllio site yielded a higher number of isolates (516, 63.4%) than Elos (298, 36.6%). The two bait insect species contributed almost equally to the isolations, with T. confusum accounting for 418 isolates (51.4%) and S. zeamais for 396 (48.6%). The majority of isolates (637, 78.3%) were not subjected to molecular analysis and were therefore classified as “Unidentified Fungi”, while the molecularly characterized subset (n = 177) was further assigned to Recorded EPF, Putative EPF, or Other Fungi (Figure 1).

3.2. Seasonal Variation in Isolation Frequency

Isolation frequency varied seasonally and between years, with consistently higher numbers in 2018 than in 2019 (statistical comparison not performed due to sampling design focused on seasonal rather than interannual trends). Spring 2018 had the highest number of isolations (188), while winter 2019 had the lowest (20). A chi-square test confirmed a highly significant seasonal effect on isolation frequency (χ² = 62.44, df = 6, p < 0.001).

The seasonal distribution of Recorded EPF, Putative EPF, and Other Fungi is provided in Figure 2.

3.3. Molecular Identification of Fungal Isolates

The 177 molecularly characterized isolates, 94 originated from Dasyllio and 83 from Elos, reflecting the proportional isolation success from each site. A total of 47 different fungal species were identified (Supplementary Table S1). Sequences were assembled and edited using Geneious Prime^® and compared against the NCBI GenBank database via BLASTn. Species-level identification was assigned based on ≥99% sequence identity and ≥95% query coverage. The most frequently encountered species were B. bassiana (39 isolates), Aspergillus aflatoxiformans Frisvad, Hubka & Samson (16), and Talaromyces annesophieae Sklenar, Hubka & Jurjevic (15).

3.4. Statistical Analysis of Factors Influencing Isolation

Chi-square analysis revealed no significant association between the sampling site and the isolation categories (χ² = 0.184, df = 2, p = 0.912), or between the host insect species and these categories (χ² = 75.89, df = 65, p = 0.167). In contrast, season had a highly significant influence on isolation frequency (χ² = 62.44, df = 6, p < 0.001), with a higher frequency of entomopathogenic isolates (both Recorded and Putative) recovered in spring and summer (Figure 2).

3.5. Analysis of Community Diversity

Analysis of diversity indices showed no statistically significant differences in fungal community diversity between the two sites (Mann–Whitney U = 0.0, p = 1.0) or between the two host insect species (Mann–Whitney U = 1.0, p = 1.0). A Kruskal–Wallis test found no significant overall effect of season on diversity indices (H = 2.15, p = 0.54). Subsequent pairwise Mann–Whitney U tests with Bonferroni correction confirmed no significant differences between any two seasons. Seasonal analysis showed that autumn exhibited the highest values for Simpson (0.93) and Evenness (0.93) indices (

Table 1. Diversity indices of fungal communities by season, region, and host insect species.

Category	Subcategory	Shannon Index	Simpson Index	Evenness Index
Season	Spring	2.77	0.88	0.84
	Summer	2.71	0.90	0.87
	Autumn	2.98	0.93	0.93
	Winter	2.88	0.93	0.91
Region	Dasyllio	3.24	0.92	0.83
	Elos	3.08	0.92	0.87
Host Insect	S. zeamais	3.32	0.92	0.83
	T. confusum	2.73	0.91	0.90

).

Statistical comparisons (Kruskal–Wallis and Mann–Whitney U tests) indicated no significant differences in diversity indices between sites, hosts, or seasons (all p > 0.05), suggesting stable community structure despite seasonal fluctuations in abundance

4. Discussion

Our findings confirm that peri-urban ecosystems in Western Greece represent significant reservoirs of entomopathogenic and putatively entomopathogenic fungi (EPF). The higher isolation rate from the Dasyllio site suggests more favorable ecological conditions for fungal activity, while the comparable diversity indices within the characterized subset between Dasyllio and Elos indicate that both sites serve as valuable sources of fungal biodiversity with biocontrol potential [4,11]. The near-equal efficacy of T. confusum and S. zeamais as bait insects validates their utility as generalist baits for capturing diverse soil-borne entomopathogens, supporting established methodological approaches in fungal ecology [15,23].

The pronounced seasonal fluctuation in EPF abundance—with clear peaks in spring and summer—carries direct implications for the timing of biocontrol interventions in Mediterranean agroecosystems. This temporal pattern aligns with peak activity periods of many key pests, including lepidopteran, coleopteran, and hemipteran species [16]. Consequently, the observed seasonal peaks suggest that late spring to early summer may be an optimal window for inundative applications, aligning with periods of high natural fungal abundance and pest activity, though field validation is needed to confirm efficacy [1,16].

Conversely, the observed stability in the characterized fungal community’s diversity, particularly the high evenness recorded in autumn, highlights opportunities for conservation biological control. Autumn exhibited the highest community evenness within the characterized subset, suggesting a balanced and potentially resilient fungal assemblage, even though total EPF abundance was lower. This period may therefore be critical for conservation practices, such as reduced tillage, maintenance of organic matter, and preservation of plant diversity, to protect the resident community structure and promote inoculum buildup for the following season [8,10,27].

The substantial recovery of “Putative EPF” (53.1% of the molecularly identified isolates) underscores the untapped taxonomic and functional diversity within peri-urban soils. The molecular characterization (Supplementary Table S1) revealed a diverse assemblage of 47 species, including multiple known entomopathogens from the genera Beauveria, Cordyceps, and Purpureocillium, alongside a wide range of putative and non-entomopathogenic fungi, underscoring the taxonomic richness of these soils. Species such as A. aflatoxiformans and T. annesophieae warrant further investigation through targeted pathogenicity screening against economically important pests to assess their biocontrol potential [4]. The frequent isolation of B. bassiana—a commercially established biocontrol agent—validates the applied relevance of these habitats and suggests that peri-urban ecosystems may harbor strains of established biocontrol agents [12,14]. This supports the strategic value of such transitional zones as sources of novel or improved microbial agents for integrated pest management (IPM). While insect baiting is effective for isolating generalist pathogens, it may overlook specialist or slow-acting fungi [3]. Future studies should integrate direct soil plating and molecular metabarcoding to capture a broader spectrum of fungal diversity. To advance from ecological observation to predictive management, research must incorporate edaphic factor analysis—examining relationships between EPF communities and soil properties such as texture, pH, organic content, and moisture [5,7,11]. Additionally, in situ pathogenicity trials are needed to evaluate the field efficacy of promising isolates under realistic agronomic conditions.

From a plant protection perspective, this work demonstrates that peri-urban green spaces near agricultural land can serve as valuable, natural sources of EPF for integrated pest management, especially during spring–summer peaks. The consistent communities across sites suggest that biocontrol strategies could be broadly applied within a region. The observed seasonal variation in EPF abundance offers a preliminary ecological basis for timing interventions: spring and summer may be suitable for augmentative releases, while autumn could be targeted for conservation practices to support fungal overwintering. We recommend further research to develop the most promising putative EPF isolates into commercially viable biopesticides.

5. Conclusions

This study demonstrates that peri-urban green spaces in Greece serve as significant reservoirs of entomopathogenic and putatively entomopathogenic fungi, with distinct seasonal patterns in their abundance and community stability. The higher isolation rates of infective fungi in spring and summer align with peak pest activity periods, providing a scientific basis for timing inundative biocontrol applications. The consistent fungal diversity across sites suggests community stability and underscores the broad value of these habitats as sources of biocontrol candidates. The abundance of putative EPF highlights the untapped potential of these ecosystems for discovering novel biocontrol strains. Future research should focus on pathogenicity screening of these isolates against key agricultural pests and on investigating the influence of soil edaphic factors on EPF communities. From an applied perspective, our findings support the integration of peri-urban EPF resources into regional IPM strategies, with seasonal timing as a critical factor for maximizing efficacy. We recommend further research to evaluate the pathogenicity and field efficacy of promising isolates, with the long-term goal of developing commercially viable biopesticides and informed conservation strategies.