Host-Specific Fungal Assemblages, Dominated by Ophiostomatoid Taxa, in Scots Pine Bark Beetles from Slovakia Revealed by Metabarcoding

Abstract

Bark beetles (Coleoptera: Scolytinae) play a dual ecological role in forest ecosystems as disturbance agents and vectors of symbiotic fungi, including blue-stain taxa that affect wood quality and tree health. This study assessed fungal communities specific to four bark beetle species—Ips acuminatus (Gyllenhal, 1827), Ips sexdentatus (Börner, 1776), Ips typographus (Linnaeus, 1758), and Pityogenes chalcographus (Linnaeus, 1761)—colonizing Scots pine (Pinus sylvestris L.) in Slovakia. Fungal DNA was extracted from beetle surfaces and analyzed using ITS2 metabarcoding on the Illumina MiSeq platform to characterize the diversity and structure of associated mycobiota. Alpha- and beta-diversity analyses revealed a taxonomically and functionally rich fungal assemblage dominated by Ascomycota, comprising over one thousand operational taxonomic units. Fungal richness and diversity varied among beetle species: I. typographus and P. chalcographus supported the most diverse communities, I. sexdentatus harbored the least diverse assemblages, and I. acuminatus showed contrasting patterns depending on the index used. Beta-diversity analysis indicated that community composition was primarily structured by beetle species identity, with weaker effects of locality and sampling method. Ophiostomatoid fungi, particularly Geosmithia pallida (G. Sm.) M. Kolařík, Kubátová & Pažoutová, Ophiostoma distortum (R.W. Davidson) de Hoog & R.J. Scheff., and Ophiostoma minus (Hedgc.) Syd. & P. Syd., were consistently prevalent and formed the core mycobiome. Random forest classification and differential abundance analyses confirmed host-specific enrichment of several ophiostomatoid and yeast taxa. Yeasts (e.g., Kuraishia, Candida, Yamadazyma), saprotrophic molds (e.g., Penicillium, Davidiella), and the entomopathogen Beauveria bassiana (Bals.-Criv.) Vuill. also occurred frequently. These findings provide the first DNA-based evidence of host-specific fungal assemblages in Scots pine bark beetles in Slovakia and emphasize their ecological significance for beetle–fungus symbioses and pine forest health.

1. Introduction

Bark beetles (Coleoptera: Curculionidae: Scolytinae) are among the most significant disturbance agents in coniferous forests, causing extensive ecological and economic impacts through large-scale tree mortality and reduced forest productivity [1,2,3,4]. Outbreaks lead to decreased timber quality and market value, increased management costs, and reduced ecosystem services, such as carbon sequestration and recreation [3,5,6].

The success of bark beetles as forest pests is closely linked to their symbiotic associations with fungi, particularly those belonging to the ophiostomatoid (blue-stain) group. These fungi are not merely transported by the beetles—they play essential roles in overcoming host tree defenses, colonizing the sapwood, and influencing beetle fitness and outbreak potential [7,8,9]. Among them, Ophiostoma, Geosmithia, and Leptographium species are key players in the decline of coniferous trees, contributing to blue-stain discoloration and facilitating host colonization [10,11,12]. Understanding the diversity and specificity of these fungal symbionts is therefore crucial for explaining the ecological mechanisms underlying bark beetle infestations and their broader consequences for forest health.

In Central Europe, bark beetle–fungus interactions have been intensively studied in Norway spruce (Picea abies (L.) H.Karst.) forests [7,13,14], but similar systems involving Scots pine (Pinus sylvestris L.) remain poorly explored. The ecological consequences of fungal diversity and beetle-species specificity on a common host, and their contribution to infestation dynamics, are still not fully understood. Previous studies have reported several ophiostomatoid taxa associated with pine-infesting beetles [15,16,17], but broad, molecularly based surveys of fungal assemblages—particularly those associated with Ips and Pityogenes species—are still lacking. These knowledge gaps limit our ability to predict outbreak development and to design effective, ecologically based management strategies.

Pine forests in Slovakia, dominated by Scots pine (Pinus sylvestris) and Austrian pine (Pinus nigra Arnold), represent a valuable yet vulnerable component of forest ecosystems. Although they occupy only about 6.4% of the total forested area [8], they play essential ecological and economic roles, including soil protection, water management, biodiversity maintenance, and timber production. In recent years, these forests have experienced declining health due to a combination of prolonged droughts, wind disturbances, and secondary pests [9]. The most damaging bark beetles include Ips acuminatus (Gyllenhal, 1827), Ips sexdentatus (Börner, 1776), Tomicus piniperda (Linnaeus, 1758), Tomicus minor (Hartig, 1834), and Orthotomicus longicollis (Gyllenhal, 1827) [9].

Recent climatic trends have further intensified these pressures. Over the past three decades, the mean air temperature in Slovakia has increased by approximately +1.6 °C during the warm half of the year and by +1.8 °C annually [18], while the western part of the country has experienced more frequent and severe droughts (a decrease in the Standardized Precipitation–Evapotranspiration Index by 0.1–0.3) [19]. Such conditions weaken host trees and create favorable environments for bark beetle development and the growth of their fungal partners. These interlinked stressors may alter both the population dynamics of beetles and the composition of associated fungal communities, thereby reinforcing the importance of understanding beetle–fungus relationships in pine forests.

This study aims to fill these gaps by applying high-throughput DNA metabarcoding to Scots pine bark beetles in Slovakia to characterize the diversity and host specificity of associated fungal assemblages, with particular attention to ophiostomatoid taxa. We hypothesize that beetle species identity represents the main determinant shaping the structure of associated fungal communities, with each bark beetle species hosting a distinct assemblage of fungal symbionts. This beetle-species specificity on a shared host (Scots pine) is expected to be reflected in the prevalence of core ophiostomatoid fungi. We further hypothesize that the effect of the collecting site will not be detectable when beetles are collected from under the bark or from the surface of pheromone traps. The results are expected to reveal patterns of functional interaction that underpin host colonization success and have implications for forest health, outbreak modeling, and pest management.

2. Materials and Methods

2.1. Study Area and Bark Beetle Sampling

Three Scots pine forest stands were selected as study sites: Hendrichovce (49.0077329 N, 20.9911378 E; 698 m a.s.l.) and Spišský Hrhov (49.0104798 N, 20.6327277 E; 522 m a.s.l.) in the Spiš region in eastern Slovakia, and Malacky (48.4503747 N, 17.0592722 E; 177 m a.s.l.) in the Záhorie region in western Slovakia. The study stands were chosen to enable comparison of the fungal community composition associated with dominant bark beetle species infesting Scots pine (P. sylvestris) across regions with contrasting ecological conditions in Slovakia. In addition, selection criteria included the confirmed presence of active bark beetle infestations on Scots pine, ensuring the availability of suitable host material and target insect species for sampling.

Plantations of P. sylvestris (100%) in Malacky were 35–90 years old, whereas in Hendrichovce they were 75 years old, with occasional admixture of Tilia cordata Mill., Picea abies, Fagus sylvatica L., and Prunus avium (L.) L. In Spišský Hrhov, plantations of P. sylvestris (78%) were 110 years old, with admixture of Larix decidua Mill. (12%) and P. abies (10%).

Monthly mean air temperature and total precipitation during the study period (May–July 2023) were obtained from the nearest climatological stations of the Slovak Hydrometeorological Institute (SHMI): Kuchyňa–Nový Dvor for the Malacky site, and Spišské Vlachy for both Hendrichovce and Spišský Hrhov. In Malacky, the mean monthly air temperatures were 14.3 °C in May, 19.4 °C in June, and 22.9 °C in July, with corresponding total precipitation values of 82.2 mm, 35.1 mm, and 22.3 mm, respectively. In eastern Slovakia (Spišské Vlachy), mean monthly air temperatures were 12.9 °C in May, 17.1 °C in June, and 19.4 °C in July, with total precipitation of 66.0 mm, 87.3 mm, and 78.8 mm, respectively. These data characterize the local climatic conditions during the sampling period [20].

Bark beetle sampling was conducted in Malacky in May and June 2023 and in Hendrichovce and Spišský Hrhov in May and July 2023. Adult beetles were collected using sterilized tweezers (cleaned with 96% ethanol after each use) either directly from their galleries in the inner bark of P. sylvestris branches and trunks or by means of modified pheromone traps ECOTRAP (Fytofarm Ltd., Bratislava, Slovakia) installed at the study sites. At each site, two pheromone traps were placed 10 m apart, each containing either the pheromone preparation IAC-Ecolure for the target species of bark beetles Ips acuminatus and Ips sexdentatus or PCIT-Ecolure for the target species Pityogenes chalcographus (Linnaeus, 1761) and Ips typographus (Linnaeus, 1758). To capture individual beetles, 30 cm × 12 cm adhesive-coated foils (100% Polyolefins; Forestina Ltd., Mnichov, Czech Republic) were attached to the trap barrier panels. The collected live bark beetle adults were placed in sterilized plastic tubes separately. They were identified to the species level according to Pfeffer [21] and stored at −20 °C for DNA analysis.

2.2. DNA Extraction, PCR, and Sequencing Analysis

The PrepMan™ Ultra Sample Preparation Reagent (ThermoFisher Scientific, Warrington, UK) was utilized for the extraction of DNA from the surface microbiome of bark beetles. Individual bark beetles were placed in 1.5 mL microtubes, each containing 100 μL of PrepMan™ Ultra reagent. The samples were then vortexed for 10 min at 1000 rpm using a laboratory vortex mixer. After vortexing, a total volume of reagent was pipetted into a new 1.5 mL microtube and heated at 98 °C for 10 min on a dry heat block. Following heating, the tubes were cooled to room temperature and centrifuged at 14,000 rpm for 2 min. Up to 30 μL of supernatant was transferred to a new microtube, stored at −20 °C, and used for the PCR reaction.

For fungal community analysis, the ITS2 region was amplified using primers gITS7 [22] and ITS4 [23], with all primers containing 6 bp identification sequences (tags). PCR was prepared using Q5 polymerase (New England Biolabs, Ipswich, MA, USA) according to the manufacturer’s recommendation. Amplification was performed in a Bio-Rad T100 Thermal Cycler. Initial denaturation (90 s at 98 °C) was followed by 35 cycles of denaturation (15 s at 98 °C), annealing (15 s at 62 °C), and extension (15 s at 72 °C). PCR products were purified using AMPureXP magnetic beads (Beckman Coulter Life Science, Brea, CA, USA) and quantified on a Qubit fluorometer (Thermo Scientific, Waltham, MA, USA). PCR amplicons were diluted to the same concentration and then pooled. The sequencing library was prepared with the TruSeq LT PCR-free kit (Illumina, San Diego, CA, USA) without DNA fragmentation and size selection. The library was quantified by qPCR using the Kapa Illumina Library Quantification kit (Roche, Basel, Switzerland), diluted to 4 nM, and denatured prior to sequencing on an Illumina MiSeq instrument with the MiSeq Reagent Kit v3 (600-cycle).

2.3. Data Processing

The obtained basic raw sequencing data were initially processed in SEED2 version 2.14 [24]. Samples were demultiplexed by tag sequences, and primers were removed. Then, samples were clustered into operational taxonomic units OTUs at 97% sequence similarity using Vsearch [25] in R ver. 4.2.2 [26]. The most abundant sequence from each OTU was selected and identified using RDP Classifier v2.14 with its default fungal training set corresponding to the UNITE + INSDC combined dataset (v8.3) [27]. Plant ITS region sequences were removed from further analysis. Data were evaluated on the MicrobiomeAnalyst platform (https://www.microbiomeanalyst.ca, accessed on 27 August 2025) [28].

2.4. Statistical Analysis

To assess the diversity and composition of fungal communities associated with bark beetles, multiple statistical analyses were conducted using the MicrobiomeAnalyst platform (https://www.microbiomeanalyst.ca, accessed on 27 August 2025) and R software (version 4.2.2). Alpha diversity indices (Chao1 richness, ACE richness, Shannon, Simpson) were calculated to compare fungal diversity among the four bark beetle species. Statistical differences among bark beetle species were assessed with the non-parametric Kruskal–Wallis test, followed by post hoc pairwise Mann–Whitney U tests. Statistical significance was set at p < 0.05. Beta diversity was evaluated using Bray–Curtis dissimilarity matrices and visualized via non-metric multidimensional scaling (NMDS) ordination plots. Permutational multivariate analysis of variance (PERMANOVA) was applied to test for significant differences in fungal community structure among bark beetle species, study sites, and collection methods (manual sampling from bark vs. pheromone traps). To identify overlaps and unique components of fungal communities, Venn diagrams were constructed based on presence–absence data of operational taxonomic units (OTUs) across bark beetle species and sites. The core mycobiome was defined as the OTUs present in at least 20% of the total samples, and its composition was summarized based on prevalence values.

Random-forest (RF) classification was used to test whether fungal community profiles discriminate among bark beetle species and to identify the most informative taxa. RF models were fitted with 500 trees and out-of-bag (OOB) error was used as an internal estimate of generalization performance. The number of candidate predictors tried at each split (mtry) was tuned over 2–50.

Differential abundance of fungal taxa between bark beetle species was tested using DESeq2 [29]. Pairwise contrasts were tested for each pair of bark beetle species, and log₂ fold changes (LFCs) were computed. p-values were corrected for multiple testing using the Benjamini–Hochberg false discovery rate (FDR) procedure. Co-occurrence network was constructed at the genus level based on Spearman rank correlations among relative abundances (p < 0.01, ρ > 0.5).

In addition, the frequency of occurrence of ophiostomatoid fungi was calculated for each bark beetle species as the proportion of individuals in which a given taxon was detected. To obtain an overall estimate across all hosts, frequencies were derived from the pooled dataset of all beetle individuals. Although all statistical analyses were performed at the OTU level, OTUs affiliated with ophiostomatoid fungi were also summarized at the genus/species level to facilitate their taxonomic identification and frequency calculations.

3. Results

3.1. Bark Beetle Identification

Four species of bark beetles from the subfamily Scolytinae were identified across the study sites (

Table 1. Number of bark beetle samples collected from pine branches/trunks and traps at sampling sites in Slovakia.

Bark Beetles	Collection Method 1	Sampling Sites			Total
		Hendrichovce	Spišský Hrhov	Malacky
Ips acuminatus	24/9	11	4	18	33
Ips sexdentatus	13/0	0	0	13	13
Ips typographus	0/28	10	10	8	28
Pityogenes chalcographus	0/22	6	9	7	22
Total	37/59	27	23	46	96

¹ Collection method: manually from pine branches or trunks/from the adhesive plate on the trap.

): Ips acuminatus (n = 33), Ips typographus (n = 28), Pityogenes chalcographus (n = 22), and Ips sexdentatus (n = 13). The collection method varied among species: I. typographus and P. chalcographus were obtained exclusively from pheromone traps, whereas I. acuminatus and I. sexdentatus were primarily collected directly from infested branches or trunks. This difference in collection technique may have influenced the structure of the fungal communities detected on the beetle surface, as further explored in Section 3.4.

3.2. Bark Beetle Surface Mycobiome

Illumina MiSeq sequencing of the ITS2 region yielded 1,551,599 raw reads from the surface of 96 bark beetles. After quality filtering (Phred score ≥ 30), 1,400,729 high-quality reads were retained for downstream analysis. Rarefaction curves demonstrated sufficient sequencing depth across all host species, supporting downstream comparisons of alpha and beta diversity (Figure 1). Sequencing completeness was further confirmed by high Good’s coverage values (mean ± SE): Ips acuminatus = 99.44 ± 0.04%, Ips typographus = 99.41 ± 0.04%, Pityogenes chalcographus = 99.36 ± 0.10%, and Ips sexdentatus = 95.30 ± 4.16%. These results indicate that the sequencing effort adequately captured the majority of fungal diversity in each dataset.

Across hosts, the mycobiome was dominated by Ascomycota, followed by Basidiomycota, with minor contributions of Zygomycota, Glomeromycota, and Chytridiomycota. In the combined dataset (all beetles), the relative composition was Ascomycota 72.8%, Basidiomycota 10.4%, Zygomycota 0.7%, Glomeromycota 0.2%, and Chytridiomycota 0.1%. In addition, a substantial proportion of sequences (15.9%) belonged to the Fungi_unidentified group. Host-aggregated profiles showed consistent Ascomycota predominance but with species-specific shifts: Ips acuminatus 74.9%/14.2%/9.9% (Ascomycota/Basidiomycota/Fungi_unidentified), I. typographus 65.0%/7.4%/26.1%, I. sexdentatus 93.0%/3.7%/3.2%, and P. chalcographus 69.7%/14.3%/15.5%. Minor phyla (Chytridiomycota, Glomeromycota, and Zygomycota) together contributed < 1.5% per bark beetle species (Figure 2A).

At class resolution, the combined profile (all beetles) was led by Saccharomycetes (29.4%), followed by Sordariomycetes (13.8%), Dothideomycetes (11.6%), Eurotiomycetes (9.5%), Malasseziomycetes (5.1%), Leotiomycetes (3.6%), Tremellomycetes (2.1%), and Agaricomycetes (1.8%); with other classes summed to 2.6%. A considerable fraction of sequences remained unassigned, represented by Ascomycota_unidentified (4.4%) and Fungi_unidentified (15.9%) in the relative abundance plot (Figure 2B). Host-wise, Saccharomycetes were most prominent in I. sexdentatus (65.1%; compared to I. acuminatus 28.1%, I. typographus 30.5%, P. chalcographus 5.5%), while Dothideomycetes and Eurotiomycetes were comparatively more abundant in P. chalcographus (19.1% and 18.0%, respectively). Sordariomycetes were present in all beetle species at comparable levels (I. acuminatus 14.2%, I. typographus 14.0%, I. sexdentatus 8.6%, P. chalcographus 16.6%). A notable fraction remained unassigned at the class level (Fungi_unidentified: I. acuminatus 9.9%, I. typographus 26.1%, I. sexdentatus 3.2%, P. chalcographus 15.5%) (Figure 2B).

Family-level patterns further resolved the balance between yeast and filamentous lineages. In all beetles, the leading families were Pichiaceae (11.2%), Debaryomycetaceae (9.5%), Trichocomaceae (8.1%), Saccharomycetaceae (6.4%), Davidiellaceae (5.6%), Ophiostomataceae (5.1%), Malasseziaceae (4.6%), and Pleosporaceae (1.9%); the remainder summed to Other (27.4%). In addition, a substantial proportion of reads belonged to unidentified taxa, represented by Fungi_unidentified (15.9%) and Ascomycota_unidentified (4.4%) in Figure 2C. When examined by host, the most pronounced family-level differences reflected the same patterns observed at the class level: Debaryomycetaceae (I. sexdentatus 29.5%; vs. I. acuminatus 3.5%, I. typographus 11.7%, P. chalcographus 1.0%) and Pichiaceae (I. sexdentatus 21.4%; vs. 15.3%, 8.1%, 3.4%) showed higher proportions in I. sexdentatus, whereas Trichocomaceae were more represented in P. chalcographus (16.2%; vs. I. acuminatus 6.3%, I. typographus 4.5%, I. sexdentatus 9.9%). Ophiostomataceae were detected across hosts at comparable shares (I. acuminatus, 4.8%; I. typographus, 5.3%; I. sexdentatus, 4.8%; P. chalcographus, 5.1%), aligning with their known associations with bark beetles (Figure 2C).

Alpha diversity analysis revealed significant differences in mycobiome richness and diversity indices among the beetle species (Figure 3). Post hoc comparisons indicated that Ips typographus and Pityogenes chalcographus generally harbored the most diverse fungal communities, whereas Ips sexdentatus showed the lowest diversity. Ips acuminatus exhibited contrasting patterns depending on the metric: it showed the lowest richness (Chao1), intermediate values comparable to P. chalcographus (ACE, Shannon), but the highest evenness (Simpson index). These differences likely reflect species-specific microhabitat preferences, ecological roles, or associations with different fungal partners.

Beta diversity analysis based on non-metric multidimensional scaling (NMDS) revealed significant differences in fungal community composition among bark beetle species (PERMANOVA, F = 13.8; R² = 0.310; p < 0.001; NMDS stress = 0.199; Figure 4). The relatively high proportion of variance explained (31%) indicates that beetle host identity is a strong determinant of fungal community structure.

Pairwise PERMANOVA comparisons confirmed significant differences in fungal community composition among all bark beetle species (

Table 2. Pairwise PERMANOVA comparisons among bark beetle species based on Bray–Curtis dissimilarities of fungal communities. Lower triangle shows R² values (effect size), upper triangle shows adjusted p-values (Benjamini–Hochberg correction).

	I. sexdentatus	I. acuminatus	I. typographus	P. chalcographus
Ips sexdentatus	–	0.001	0.001	0.001
Ips acuminatus	0.195	–	0.001	0.001
Ips typographus	0.200	0.130	–	0.001
Pityogenes chalcographus	0.315	0.077	0.143	–

). The strongest separation was observed between I. sexdentatus and the remaining species (R² = 0.195–0.315, adjusted p < 0.01), whereas I. acuminatus and P. chalcographus showed relatively low but still significant dissimilarity (R² = 0.077, adjusted p < 0.01).

The Venn diagram (Figure 5) shows both shared and host-specific operational taxonomic units (OTUs). While certain OTUs, such as Geosmithia pallida (G. Sm.) M. Kolařík, Kubátová & Pažoutová, Ophiostoma minus (Hedgc.) Syd. & P. Syd., and Ophiostoma distortum (R.W. Davidson) de Hoog & R.J. Scheff., were widely distributed across beetle species, others showed a more restricted, host-associated occurrence.

The core mycobiome, defined as OTUs with >20% prevalence across samples, comprised 22 taxa in total. When evaluated separately by host species, the number of core taxa was highest in I. acuminatus (18) and P. chalcographus (18), followed by I. typographus (12), and I. sexdentatus (9). Ophiostomatoid fungi, particularly O. distortum, O. minus, and G. pallida, were consistently present across hosts, reflecting their central role in bark beetle–fungus associations. Yeasts represented a substantial part of the core community, with Kuraishia capsulata (Wick.) Y. Yamada, K. Maeda & Mikata, Candida sp., Yamadazyma scolyti (Phaff & Yoney.) Billon-Grand, Malassezia restricta E. Guého, J. Guillot & Midgley, and Malassezia sp. detected at high prevalence, especially in I. sexdentatus. Filamentous fungi such as Davidiella tassiana (De Not.) Crous & U. Braun, Penicillium bialowiezense K.W. Zaleski, and Penicillium citreonigrum Dierckx also belonged to the core assemblages, suggesting both endophytic and saprotrophic guilds contribute to beetle-associated microbiota. Moreover, the entomopathogen Beauveria bassiana (Bals.-Criv.) Vuill. was identified as part of the core community (overall prevalence 31.2%), indicating potential antagonistic interactions.

Table 3. Core mycobiome taxa prevalence across bark beetle species. Prevalence of fungal taxa (>20% threshold) associated with Scots pine bark beetles in Slovakia. Values are shown separately for each beetle species and for the combined dataset (all beetles). The table presents only those OTUs that were taxonomically assigned at the genus or species level.

Fungal OTUs Identified to Genus/Species Level	Ips acuminatus	Ips sexdentatus	Ips typographus	Pityogenes chalcographus	All Beetles
Alternaria eichhorniae Nag Raj & Ponnappa	0.545	0.538	0.0	0.0	0.292
Beauveria bassiana (Bals.-Criv.) Vuill.	0.394	0.0	0.0	0.591	0.312
Candida sp.	0.485	0.231	0.286	0.773	0.458
Davidiella tassiana (De Not.) Crous & U. Braun	0.970	0.231	0.821	0.955	0.823
Diplodia allocellula Jami, Gryzenh., Slippers & M.J. Wingf.	0.0	0.0	0.893	0.227	0.323
Geosmithia pallida (G. Sm.) M. Kolařík, Kubátová & Pažoutová	0.242	0.231	0.500	0.500	0.375
Kuraishia capsulata (Wick.) Y. Yamada, K. Maeda & Mikata	0.667	0.385	0.464	0.864	0.615
Malassezia restricta E. Guého, J. Guillot & Midgley	0.939	0.308	0.464	0.955	0.719
Malassezia sp.	0.303	0.0	0.0	0.636	0.292
Nakazawaea holstii (Wick.) Y. Yamada, K. Maeda & Mikata	0.455	0.0	0.0	0.591	0.354
Ophiostoma distortum (R.W. Davidson) de Hoog & R.J. Scheff.	0.455	0.0	0.0	0.682	0.354
Ophiostoma minus (Hedgc.) Syd. & P. Syd.	0.485	1.000	0.536	0.0	0.458
Penicillium bialowiezense K.W. Zaleski	0.879	0.308	0.536	0.909	0.708
Penicillium citreonigrum Dierckx	0.0	0.0	0.214	0.386	0.191
Yamadazyma scolyti (Phaff & Yoney.) Billon-Grand	0.394	0.0	0.250	0.682	0.385

lists only those core OTUs that could be taxonomically assigned at the genus or species level.

The co-occurrence network was highly modular, forming several host-specific clusters composed of genera primarily associated with individual bark beetle species (Figure S1). Within these modules, correlations were exclusively positive, indicating the co-occurrence of fungi that share similar ecological niches or environmental preferences on the beetle surface. In contrast, a few negative correlations were detected almost exclusively among the dominant yeast genera (Candida, Kuraishia, Malassezia, and Pichia), which also represented the largest and most abundant nodes in the network. These inverse relationships likely reflect competitive exclusion or resource partitioning among ecologically similar yeasts occupying overlapping niches on the beetle cuticle.

The possibility to distinguish bark beetle species according to their fungal communities was analyzed using random forest classification (Figure 6). The model reached an overall OOB error of 10.4% meaning that most beetle species were reliably distinguished, with low misclassification rates for I. typographus (7.1%), I. sexdentatus (7.7%), and I. acuminatus (9.1%). In contrast, P. chalcographus showed a higher error (18.2%), being occasionally misclassified as I. acuminatus, consistent with their overlapping habitats and similar fungal assemblages also indicated by PERMANOVA.

The highest-ranked predictors were multiple OTUs identified as Kuraishia capsulata (8), together with other yeasts such as Candida sp. (4), Ogataea wangdongensis Limtong, Kaewwich. & M. Groenew. (1), Nakazawaea holstii (Wick.) Y. Yamada, K. Maeda & Mikata (2), Pichia amylophila Kurtzman, M.J. Smiley, C.J. Johnson, Wick. & Fuson (1), and Yamadazyma scolyti (1). In addition, well-described ophiostomatoid fungi, including Ophiostoma ainoae H. Solheim (1), Ceratocystis polonica (Siemaszko) C. Moreau (1), and Ambrosiella sp. (1), were among the top-ranked features, confirming their central role in bark beetle–fungus symbioses.

Differential abundance analysis (Table S1) at the genus level confirmed the importance of random forest predictors and revealed strong host-associated shifts in fungal communities. Yeasts were the most frequently enriched taxa, including Kuraishia, Candida, Ogataea, Nakazawaea, and Pichia, which differed significantly across multiple beetle species (log₂FC up to ±11, FDR < 0.01). Several ophiostomatoid fungi (Ophiostoma, Ceratocystis, Ceratocystiopsis, Ambrosiella, and Graphium) also showed significant host associations, confirming their role as core bark beetle symbionts. In addition, saprotrophic and pathogenic genera such as Penicillium, Aspergillus, Alternaria, and Davidiella contributed to the differentiation. A complete list of all detected taxa, including both significant and non-significant contrasts across all pairwise comparisons, is provided in Table S2.

While the diversity analyses were based on OTUs, those affiliated with ophiostomatoid fungi were subsequently presented at the genus/species level (

Table 4. Frequency of occurrence of ophiostomatoid fungal taxa in pine bark beetle samples collected in Slovakia.

Taxa	Family	Frequency (%)
		IA (n = 33)	IS (n = 13)	IT (n = 28)	PC (n = 22)	All Beetles (n = 96)
Ambrosiella sp. 3PG4P A2	Ceratocystidaceae	75.76	7.69	57.14	45.45	54.17
Ceratocystidaceae sp.	Ceratocystidaceae	18.18	–	–	–	6.25
Ceratocystiopsis minuta (Siemaszko) H.P. Upadhyay & W.B. Kendr.	Ophiostomataceae	18.18	30.77	64.29	18.18	33.33
Ceratocystis polonica (Siemaszko) C. Moreau	Ceratocystidaceae	3.03	–	67.86	4.55	21.88
Ceratocystis rufipenni M.J. Wingf., T.C. Harr. & H. Solheim	Ceratocystidaceae	–	–	7.14	–	2.08
Ceratocystis sp. CspXger3	Ceratocystidaceae	21.21	23.08	7.14	9.09	14.58
Geosmithia langdonii M. Kolařík, Kubátová & Pažoutová	Bionectriaceae	9.09	–	14.29	13.64	10.42
Geosmithia pallida (G. Sm.) M. Kolařík, Kubátová & Pažoutová	Bionectriaceae	100.00	76.92	100.00	100.00	96.88
Geosmithia rufescens M. Kolařík	Bionectriaceae	3.03	7.69	3.57	4.55	4.17
Geosmithia sp. CCF3645	Bionectriaceae	–	–	–	4.55	1.04
Geosmithia sp. RJ0258	Bionectriaceae	–	–	–	4.55	1.04
Gondwanamyces sp. JAL 2011b	Ceratocystidaceae	–	–	3.57	–	1.04
Graphium fimbriasporum (M. Morelet) K. Jacobs, Kirisits & M.J. Wingf.	Microascaceae	9.09	–	57.14	9.09	21.88
Graphium pseudormiticum M. Mouton & M.J. Wingf.	Microascaceae	6.06	7.69	32.14	–	12.50
Graphium sp. JJK 2009a	Microascaceae	–	–	3.57	–	1.04
Hyalorhinocladiella sp. 2YT4P H2	Ophiostomataceae	–	7.69	21.43	–	7.29
Leptographium cf. truncatum CMW 22857	Ophiostomataceae	3.03	46.15	–	4.55	8.33
Ophiostoma ainoae H. Solheim	Ophiostomataceae	9.09	7.69	71.43	18.18	29.17
Ophiostoma bicolor R.W. Davidson & D.E. Wells	Ophiostomataceae	6.06	–	50.00	4.55	17.71
Ophiostoma brunneociliatum Math.-Käärik	Ophiostomataceae	30.30	76.92	46.43	22.73	39.58
Ophiostoma distortum (R.W. Davidson) de Hoog & R.J. Scheff.	Ophiostomataceae	96.97	92.31	100.00	100.00	97.92
Ophiostoma floccosum Math.-Käärik	Ophiostomataceae	3.03	7.69	3.57	–	3.13
Ophiostoma ips (Rumbold) Nannf.	Ophiostomataceae	18.18	69.23	14.29	9.09	21.88
Ophiostoma minus (Hedgc.) Syd. & P. Syd.	Ophiostomataceae	93.94	76.92	100.00	100.00	94.79
Ophiostoma piliferum (Fr.) Syd. & P. Syd.	Ophiostomataceae	–	–	3.57	–	1.04
Ophiostoma pluriannulatum (Hedgc.) Syd. & P. Syd.	Ophiostomataceae	3.03	7.69	–	18.18	6.25
Ophiostoma rectangulosporium Ohtaka, Masuya & Yamaoka	Ophiostomataceae	–	–	3.57	–	1.04
Ophiostoma cf. rectangulosporium 1039RJP	Ophiostomataceae	51.52	15.38	42.86	45.45	42.71
Ophiostoma cf. rectangulosporium CMW 26258	Ophiostomataceae	54.55	–	17.86	22.73	29.17
Ophiostoma saponiodorum Linnak., Z.W. de Beer & M.J. Wingf.	Ophiostomataceae	–	23.08	–	–	3.13
Ophiostoma tapionis Linnak., Z.W. de Beer & M.J. Wingf.	Ophiostomataceae	–	–	7.14	–	2.08
Ophiostomataceae sp.	Ophiostomataceae	45.45	23.08	78.57	63.64	56.25
Pesotum australi Kamg. Nkuek., K. Jacobs & M.J. Wingf.	Ophiostomataceae	18.18	15.38	14.29	27.27	18.75
Sporothrix sp. 2 ML 2011	Ophiostomataceae	–	–	7.14	4.55	3.13
Thielaviopsis basicola (Berk. & Broome) Ferraris	Ceratocystidaceae	–	–	–	4.55	1.04

IA—Ips acuminatus, IS—Ips sexdentatus, IT—Ips typographus, PC—Pityogenes chalcographus.

) to highlight their ecological composition and frequency across beetle hosts. A total of 35 ophiostomatoid taxa were detected across the 96 bark beetle individuals analyzed. These included members of the families Ophiostomataceae, Ceratocystidaceae, Bionectriaceae, and Microascaceae. The dataset comprised both well-identified species and several taxa classified at the genus or family level (e.g., Ceratocystidaceae sp., Geosmithia spp., Ophiostomataceae sp.). When considering only taxa confidently identified to the species level, approximately 20 distinct species were recorded. The most prevalent species across all beetle hosts were G. pallida (96.88%), O. distortum (97.92%), and O. minus (94.79%), each occurring in nearly all samples. Other frequently detected taxa included Ambrosiella sp. 3PG4P A2 (54.17%), Ophiostoma cf. rectangulosporium Ohtaka, Masuya & Yamaoka (42.71%), O. ainoae (29.17%), Ceratocystiopsis minuta (Siemaszko) H.P. Upadhyay & W.B. Kendr. (33.33%), and Ophiostoma brunneociliatum Math.-Käärik (39.58%).

Patterns of occurrence differed among the bark beetle hosts. In I. acuminatus, the top three species were G. pallida (100%), O. distortum (96.97%), and O. minus (93.94%). Additional frequent associates included Ambrosiella sp. 3PG4P A2 (75.76%) and O. cf. rectangulosporium (54.55%). In I. sexdentatus, the most frequent associates were O. distortum (92.31%), O. brunneociliatum (76.92%), and G. pallida (76.92%). O. minus (76.92%) and Ophiostoma ips (Rumbold) Nannf. (69.23%) also reached high frequencies. In I. typographus, the dominant species were G. pallida (100%), O. distortum (100%), and O. minus (100%), accompanied by high frequencies of O. ainoae (71.43%), C. polonica (67.86%), and C. minuta (64.29%). In P. chalcographus, the most common taxa were again G. pallida (100%), O. distortum (100%), and O. minus (100%), while Ophiostomataceae sp. (63.64%) and Ambrosiella sp. 3PG4P A2 (45.45%) were also frequently detected. Overall, while a small group of fungi (G. pallida, O. distortum, and O. minus) was nearly ubiquitous across all beetle hosts, each bark beetle species was associated with a distinct subset of additional taxa. This indicates the coexistence of generalist species widely shared among hosts and host-associated fungi that contribute to the variability in community composition.

The results show that the bark beetle surface supports a taxonomically diverse and functionally structured fungal community composed of blue-stain fungi, yeasts, environmental molds, and fungi associated with insects. This diversity reflects complex ecological interactions and possible functional complementarity within the beetle holobiont.

3.3. Habitat-Specific Fungal Communities

Non-metric multidimensional scaling (NMDS) analysis revealed statistically significant but weak differences in fungal community composition among beetles from different sites (PERMANOVA, F = 2.681; R² = 0.063; p < 0.001; stress = 0.193; Figure 7). With the site explaining only about 6% of the variance, the effect was modest; however, some separation of points in NMDS space suggests that geographic site contributed slightly to shaping fungal communities. This pattern may reflect differences in forest composition, elevation, or climate across study sites. Nevertheless, several fungal taxa were shared among all sites, including the dominant ophiostomatoid species O. minus, O. distortum and G. pallida.

3.4. Impact of Bark Beetles Collection Method on Fungal Communities

A comparison of fungal communities associated with I. acuminatus collected by two methods showed statistically significant but modest differences (PERMANOVA, F = 3.754; R² = 0.108; p < 0.001; NMDS stress = 0.179; Figure 8). The method of collection explained about 11% of the variance, with partial clustering of samples according to collection method and some overlap, especially along the NMDS2 axis. The main separation occurred along NMDS1, indicating that trap-caught and bark-extracted beetles harbored partially distinct fungal profiles. Trap-caught individuals tended to carry a more heterogeneous set of fungi, potentially due to exposure to airborne spores or surface contamination during flight, whereas beetles extracted from galleries displayed a more cohesive fungal community, likely reflecting direct contact with host tissues and gallery-associated microhabitats.

These results indicate that the collection method is a contributing, but not dominant, factor influencing the surface-associated mycobiota. Careful consideration of sampling strategy is therefore essential for accurate ecological interpretation, especially when comparing beetle–fungus associations across habitats or host species.

4. Discussion

This study provides one of the first comprehensive insights into the fungal communities associated with Scots pine-infesting bark beetles in Slovakia, using high-throughput sequencing of the fungal ITS2 region. Our results reveal a taxonomically and functionally diverse fungal assemblage dominated by members of the Ascomycota, especially within the classes Sordariomycetes, Dothideomycetes, and Eurotiomycetes, in agreement with patterns observed in similar studies [10,30].

A relatively high proportion of sequences could not be classified beyond the fungal kingdom using the UNITE + INSDC database [31]. This reflects the incomplete taxonomic coverage of current ITS reference datasets rather than sequencing artifacts. UNITE species hypotheses contain numerous environmental lineages lacking formal taxonomic names, which are therefore reported as “Fungi sp.” Some of these, including the most abundant OTU in our dataset, were assigned to Wickerhamomyces bisporus (O. Beck) Kurtzman, Robnett & Bas.-Powers in alternative databases; however, due to inconsistent annotations across repositories, we retained a single reference system to ensure comparability. Because diversity analyses were based on relative abundances, unidentified taxa did not affect the interpretation of community patterns.

A key focus of this study was the identification and distribution of ophiostomatoid fungi, which act as well-documented obligate symbionts forming mutualistic or commensal associations with bark beetles. Multiple ophiostomatoid genera were detected—Ophiostoma, Geosmithia, Ceratocystis, Ceratocystiopsis, and Graphium—with O. distortum and G. pallida consistently ranking among the most frequent taxa across beetle species and sites (Table 4). The high frequency of O. minus, O. brunneociliatum, and C. polonica further highlights the potential of bark beetles as vectors of pathogenic or sapstaining fungi in Scots pine forests [32]. The potential pathogenic roles of detected ophiostomatoid fungi align with broader concerns about bark beetle-mediated fungal dispersal in European forests [33], particularly as climate change may alter both beetle dynamics and fungal virulence patterns. Importantly, the differentiation of fungal assemblages among beetle species reflects true vector-driven specificity rather than local environmental effects. This finding is consistent with studies from Central Europe showing that particular fungal symbionts are strongly linked to their insect vectors, even when these share the same host tree species [34]. Moreover, Six and Klepzig [35] emphasized that beetle behavioral ecology and gallery microhabitat act as key drivers of symbiotic assembly, shaping both the taxonomic and functional composition of associated fungi. Some taxa, such as Ophiostoma cf. rectangulosporium and several Geosmithia spp., may even represent novel or cryptic species, underscoring the significance of molecular methods in revealing hidden fungal diversity and highlighting the need for integrative approaches that combine metabarcoding with culture-based isolation and phylogenetic analyses [36].

Beyond ophiostomatoid taxa, we also detected a substantial presence of yeasts, such as Kuraishia capsulata, Candida spp., Yamadazyma scolyti, and Nakazawaea holstii, which formed part of the core mycobiome (Table 3). The high prevalence of Saccharomycetales, together with the frequent detection of Malassezia species, some of which may act as transient environmental fungi, suggests that yeasts may play important roles on the beetle cuticle, potentially linked to nutrient acquisition, detoxification processes, and microbial interactions within beetle galleries. Functionally, these yeasts contribute to metabolic complementation and chemical communication, reflecting the mechanistic integration within beetle life cycles [35]. This pattern aligns with the concept of multifunctional mycobiomes, where primary ophiostomatoid symbionts (e.g., Geosmithia spp.) facilitate host colonization and defense suppression, yeasts (e.g., Yamadazyma spp.) participate in nutrient cycling and volatile transformation, and secondary molds dominate later successional stages [32,37]. Some yeast species may also be involved in pheromone production or metabolite signaling, potentially influencing aggregation or reproduction of beetles [37,38,39]. Recent genomic work has shown that dominant yeast symbionts of bark beetles often possess full pathways for the synthesis of essential amino acids and vitamin B6, which may help compensate for nutrient deficiencies in phloem-based diets [40].

The modular architecture of the network (Figure S1), dominated by positive correlations within beetle-species-specific clusters, indicates that fungal assemblages are structured primarily by beetle identity and its associated microhabitat, with competition playing a secondary role. The exception lies in the highly connected yeast genera (Candida, Kuraishia, Malassezia, and Pichia), whose weakly negative associations suggest niche overlap and potential competition for surface substrates or metabolic resources. Such modularity supports the concept of vector-driven specificity, where distinct beetle species maintain functionally cohesive fungal consortia that reflect their behavioral ecology and gallery environment [35]. Similar patterns were reported by Kolařík and Jankowiak [34], who demonstrated that fungal guilds—particularly Geosmithia and Ophiostoma species—are consistently partitioned among bark beetle vectors even when inhabiting the same pine host. Together, these findings suggest that dominant yeasts and ophiostomatoid fungi act as keystone taxa stabilizing the network through a balance of cooperation and competition within vector-defined assemblages.

Among the filamentous fungi, Davidiella tassiana—the most prevalent taxon (82.3%)—is known as a common endophyte and epiphyte of conifers, which may influence host defense responses and contribute to shaping microbial communities within galleries. This species has been reported as an endophyte in Pinus halepensis Mill. in Spain [41], suggesting that it is well adapted to pine hosts and may play a broader role in conifer–fungus interactions. Saprotrophic molds such as Penicillium bialowiezense, P. citreonigrum, and Aspergillus flavus Link were also frequently detected as facultative associates in beetle galleries. Although traditionally viewed as environmental contaminants, their consistent detection across beetle species suggests they may play functional roles in degrading plant material or outcompeting more aggressive fungal invaders [42]. Their presence may also reflect airborne fungal spores adhering to bark beetle cuticles during dispersal or flight.

The prevalence of B. bassiana detected on I. acuminatus and P. chalcographus exoskeletons was notably high (39.4% and 59.1%, respectively). Beauveria bassiana is a ubiquitous entomopathogenic fungus, frequently encountered in bark beetle populations and forest habitats (e.g., [43,44,45,46,47]). Although this fungus can parasitize susceptible hosts via direct penetration of the cuticle [48,49] and is extensively studied as a biological control agent for a variety of arthropods [50], including bark beetles (e.g., [47,51,52]), our field collections revealed no cadavers symptomatic of active infection. This suggests that the high prevalence likely reflects the ability of Beauveria conidia to adhere passively to the beetle cuticle via electrostatic and hydrophobic interactions [53]. They are then passively present on the exoskeleton, rather than indicating ongoing disease outbreaks within the beetle populations. The presence of B. bassiana spores on the exoskeleton does not guarantee infection, because favorable environmental conditions are required to initiate pathogenicity [54]. DNA metabarcoding methods cannot discriminate between viable entomopathogenic propagules and non-viable or phoretic spores, potentially overestimating true infection rates. Many studies in Central Europe demonstrate that fungal infection by Beauveria spp. appears in bark beetle populations at low, but constant, prevalence levels (e.g., [43,44,45,55]). Besides abiotic factors, such as microclimatic conditions (e.g., high humidity and temperature), the efficacy of B. bassiana for controlling bark beetle populations is limited by a complex set of biotic factors inherent to forest environments [54,56]. Interactions with the resident cuticular microbiota can also affect Beauveria establishment and its pathogenic activity. Laboratory studies have demonstrated that ophiostomatoid fungi, such as Leptographium abietinum (Peck) M.J. Wingf., can outcompete B. bassiana through resource competition and volatile-mediated antagonism [56,57]. These blue-stain symbionts may thereby provide a form of defensive mutualism, reducing the likelihood of successful entomopathogen colonization. Abiotic factors, along with biotic interactions from cuticular microbiota, create a dynamic equilibrium in which B. bassiana is often detected, but persistent infections are rare. This demonstrates that the fungal community on bark beetles is shaped not just by exposure to spores, but also by complex relationships, such as antagonism or tolerance, among different fungal taxa. Interestingly, B. bassiana was not detected on I. typographus and I. sexdentatus in our metabarcoding survey, despite extensive literature demonstrating that both species are susceptible hosts for this entomopathogen (e.g., [43,44,45,58,59,60]). This absence does not reflect beetle host specificity or resistance, but rather indicates limitations of our dataset, and the underlying cause for this pattern remains unresolved within the scope of the present study.

Beta diversity analyses and random forest classification confirmed that fungal community composition was primarily driven by beetle-species identity, followed by site and collection method. Collection method explained approximately 11% of the variation in fungal community composition, a magnitude that was quantitatively smaller than the effect of beetle species identity (R² = 0.31) but still statistically significant. This moderate influence indicates that methodological differences, such as whether beetles were collected from galleries or captured in pheromone traps, contributed less to mycobiome variation than beetle species identity. These findings align with bark beetle mycobiome studies demonstrating that while sampling methodology may introduce some variation, the core mycobiome, particularly ophiostomatoid symbionts, remains consistently detectable across different collection approaches, reflecting stable beetle–fungus associations rather than methodological artifacts [61,62,63,64,65,66]. Trap-caught individuals may show greater heterogeneity in fungal profiles, likely reflecting exposure to environmental spores during flight or passive adhesion of airborne propagules to the cuticle [66], whereas gallery-collected beetles typically display more cohesive fungal assemblages shaped by direct contact with host tissues and gallery microhabitats [67]. Importantly, the core ophiostomatoid fungi in our study, G. pallida, O. distortum, and O. minus, remained consistently prevalent across both collection methods, confirming their stable association with bark beetles regardless of sampling approach. This quantitative assessment of sampling bias underscores that while methodological factors should be considered in study design and interpretation, the dominant driver of mycobiome structure in our dataset was the biological differences among beetle species.

Although all laboratory steps were performed under sterile conditions and negative PCR controls did not show any amplification, we cannot entirely exclude the possibility of low-level environmental or extraction-associated contamination. Genera such as Malassezia or Penicillium are known to occur as recurrent airborne or laboratory-associated taxa in fungal metabarcoding datasets [68,69]. Therefore, their occurrence in our samples should be interpreted cautiously and verified by future targeted analyses.

This study highlights the multifunctional fungal communities associated with Scots pine bark beetles in Slovakia, with a strong representation of ophiostomatoid fungi of economic and ecological significance, as well as a broader suite of yeasts, molds, and entomopathogens. By demonstrating that fungal assemblages differ mainly among beetle species sharing the same host tree, our findings clarify the ecological mechanisms driving bark beetle–fungus associations and emphasize the role of vector behavior and microhabitat in shaping these symbioses [34,35]. The discovery of a beetle-specific core assemblage dominated by Ophiostoma, Geosmithia, and yeasts further suggests that these fungi may influence beetle success and pine health under climate-driven stress. Environmental factors such as increasing temperature and drought have been shown to alter the diversity and pathogenic potential of ophiostomatoid fungi [32,38], potentially contributing to the decline of Scots pine forests across Europe. Comparable bark beetle–fungus systems in other conifer hosts and regions exhibit similar ecological dynamics and associated risks. In boreal Pinus sylvestris forests in Fennoscandia and western Russia, ophiostomatoid associates of pine- and spruce-infesting bark beetles are frequently recovered from stressed or dying trees and are considered contributors to physiological decline under drought and thermal stress [67]. Similarly, the introduction of the pathogenic ophiostomatoid fungus Leptographium wingfieldii M. Morelet to North America with the pine shoot beetle Tomicus piniperda has been associated with pine mortality in non-native ecosystems, demonstrating that bark beetle–fungus complexes can pose significant risks to pine stands on broader geographic scales [70]. Moreover, yeast-dominated microbiota can mediate semiochemical production and influence beetle aggregation and colonization efficiency [71]. Together, these findings advance our understanding of bark beetle–fungus symbioses and provide a baseline for future research on the role of microbial associates in forest pest dynamics and pine ecosystem health. Future research should integrate findings from diverse European forest systems (e.g., [33]) to develop comprehensive management strategies for bark beetle–fungus complexes across biogeographic gradients.

5. Conclusions

This study provides the first DNA metabarcoding-based insight into the fungal communities associated with bark beetles infesting Scots pine (P. sylvestris) in Slovakia. Both alpha and beta diversity analyses showed that beetle species identity was the main factor shaping fungal communities, while collection method and locality played weaker but still significant roles. Alpha diversity was generally highest in I. typographus and P. chalcographus and lowest in I. sexdentatus, whereas I. acuminatus exhibited contrasting patterns depending on the diversity index. Ophiostomatoid fungi, particularly O. distortum, G. pallida, and O. minus, were dominant and widespread across all beetle taxa, confirming their consistent and efficient transmission via beetle vectors and their central role in the bark beetle–fungus symbiosis. In addition to these well-known blue-stain fungi, the detection of yeasts, saprotrophic molds, and entomopathogenic fungi highlights the functional diversity of bark beetle-associated fungal assemblages. Such vector-mediated fungal dispersal underscores the ecological role of bark beetles as key agents shaping microbial dynamics and influencing host colonization success and forest health. These findings emphasize the importance of fungal communities in the life cycle and behavior of bark beetles and highlight the need to integrate molecular approaches with traditional culturing and functional assays to better understand the ecological roles of both common and rare fungal associates.