Virulence of Different Entomopathogenic Fungi Species and Strains against the Hazel Longhorn Beetle Oberea linearis (Coleoptera: Cerambycidae)

Abstract

The aim of this study was to investigate alternative methods to control longhorn beetle (Oberea linearis) infestations in walnut orchards. Wild isolates of entomopathogenic fungi obtained from soil samples from Greece and Cyprus were tested for their efficacy against adults and larvae of O. linearis. Insect populations were acquired from a heavily infested walnut orchard and individuals were placed in Petri dishes provided with ground walnut wood for larvae and fresh leaves for adults. The tested insects were subjected to 16 different wild isolates from the genera Beauveria, Cordyceps, Metarhizium, and Purpureocillium, where 10⁸ conidia/mL were applied by spraying, and insects were monitored daily for 16 days. The results showed that all the tested fungi resulted in a mortality rate of 66–100%, with Cordyceps fumosorosea exhibiting the highest virulence, causing complete mortality to both larvae and adults. These findings suggest that the management of O. linearis, which has traditionally relied on chemical applications, could transition to an organic approach by utilizing entomopathogenic fungi.

1. Introduction

Hazelnuts and walnuts are highly appreciated as nutritious food sources, and as a result they are massively produced in the northern hemisphere. They play a vital role in contributing to the national revenue of many countries like Turkey, USA, Spain, Italy, and Greece. For Turkey in particular, they are an extremely important agricultural product as Turkish walnuts and hazelnuts account for 80% of world production [1]. The longhorn beetle Oberea linearis L. (Coleoptera: Cerambycidae) is considered one of the most important pests of hazelnuts and walnuts in Europe [2,3]. However, the quantity of literature dedicated to this stem borer is not commensurate with its significance as a pest of natural forests and tree orchards. The geographical range of this insect is restricted to Europe. Official reports of its occurrence in Greece [4], Italy and Sardinia [5,6], Serbia [7], Turkey [8], Albania [9], and Georgia [10] have been published. However, there are also anecdotal reports of its occurrence in other parts of Europe. The insects’ primary hosts are hazel (Corylus sp.) and walnut trees (Juglans sp.), as well as the European hop hornbeam (Ostrya carponifolia), alder (Alnus sp.), and elm (Ulmus sp.) [11]. Furthermore, the proximity of walnut trees to wild hazel can have an impact on the cultivated trees [12].

The adult O. linearis has a very narrow body, with length size 11–16 mm. It is colored black with pale yellow legs. In the male, the antennae are slightly shorter than the body, whereas in the female they are even shorter. The larvae are apodous, with the characteristic shape of the Cerambycidae. They can reach a final length of 20–25 mm and are white to yellowish, with a brownish prothoracic plate [4]. The complete life cycle of O. linearis commonly lasts two years, although it can be prolonged to three years in areas exhibiting lower temperatures. However, Anagnostopoulos [4] claimed that in Greece the insect completes one generation in a year in walnut trees [4,13]. The females lay eggs separately near the tip of young shoots. On walnut trees, females may lay eggs on the pedicel of the fruit. The larvae hatch and create a gallery, which causes the apical part of the shoot to wither. As the tunnel elongates, the larvae open small holes to remove excrement. Galleries on hazel trees are longer than those on walnut trees. After overwintering for two years, the larvae pupate and adults emerge (May–June) and feed on the lateral veins of young leaves and mate. Affected twigs exhibit wilting, desiccation, and splitting at the level of egg-laying scars. Consequently, on walnut trees, there can be fruit depletion and loss of dormant floral buds and certain foliage sections. The severity of damage increases with a higher population density of the insects, potentially causing significant harm. In hazelnuts, damage is typically confined to shoots and twigs, usually not warranting control measures. Nevertheless, economic damage to the hazelnut crop may still occur in cases of excessively high insect population densities [12,13].

In response to the growing demand, from consumers and farmers, for alternative, non-chemical methods of pest management, the utilization of entomopathogenic fungi (EPF) has been a promising option for the replacement or reduction of the excessive use of chemical insecticides [14,15,16]. These are naturally occurring microorganisms that are harmless to the environment and mammals [17]. Apart from that, they regrow on the dead insect bodies, thus introducing more inoculum into the ecosystem [18,19]. Various species of EPF have been successfully employed for pest control, not only for beetles [20] but also for other insect orders like moths and whiteflies [21,22]. Among the most commonly used species are Beauveria bassiana (Balsamo) Vuillemin (Hypocreales: Cordycipitaceae) and Metarhizium anisopliae (Metschn.) Sorokin (Hypocreales: Calvicipitaceae), which have already been applied as biological products for pest control in the commercial sector [23,24].

Despite the plethora of published studies on the insecticidal effects of many EPF, there is scarce data on their action against xylophagous beetles. Notable mortality rates of 51.1–86.3% and 69.6–81.8% were observed in Hedypathes betulinus (Klug) (Coleoptera: Cerambycidae) when infested by B. bassiana and M. anisopliae isolates, respectively [25]. Moreover, commercial EPF stains were tested on the devastating pest the Asian longhorned beetle, Anoplophora glabripennis (Motschulsky) (Coleoptera: Cerambycidae). Metarhizium isolates were found to be the most promising, causing complete adult mortality in less than 28 days [26], while Beauveria were observed to reduce adult longevity and oviposition rate [27]. Similar studies have reported that EPF exhibit moderate-to-high virulence in adults and larvae of beetle wood borers [28,29,30,31,32].

To the best of our knowledge, this study represents the first investigation into the impact of EPF on O. linearis. The objective of this experiment was to identify a biological control approach for the management of O. linearis and, more specifically, to evaluate the virulence of 16 wild strains of entomopathogenic fungi (EPF) isolated from orchard soils, namely Beauveria, Cordyceps, Metarhizium, and Purpureocillium genera, against the larvae and adults of O. linearis in a laboratory setting.

2. Materials and Methods

2.1. Entomopathogenic Fungi

2.1.1. Isolation and Identification of Entomopathogenic Fungi

The insect bait method was used to collect EPF from soil samples collected from five orchards in Greece (Achaia, Crete, Ileia, and Attiki) and one in Cyprus (Paramali). In total, there were 200 soil samples. The samples were from suburban green spaces in Achaia, Paramali, and Attiki, from tomato cultivation in Ileia, and, finally, from olive cultivation in Achaia and Crete. During sampling preparation, the surface litter was removed, and the soil was dug to a depth of 10 cm with a soil core borer. A total of 1000 g of soil from each point was placed in plastic bags and stored at 4 °C, until they were transferred to the laboratory for further processing. After drying the soil samples in air to avoid possible entomopathogenic nematode infestation [30], they were placed on rough cardboard on the laboratory counter for 24 h to reduce their humidity. Afterwards, the soil was sieved (2 mm × 1 mm, Aggelis Equipment, Athens, Greece) and 10 gr of it was placed in 9 cm Ø Petri dishes. Ten Petri dishes were prepared for each soil sample, and 10 individuals of the selected insect bait species (

Table 1. Isolates of various entomopathogenic fungal species (EPF) that were tested in the present study. All collected fungal isolates were lab cultured and stored at 25 °C on SDA plates.

Fungal Species	Isolate	Insect Bait	Collection Site	Percent Identity	Blast ID Number
Beauveria bassiana	BB K	Tribolium confusum	Moires—Crete, Greece	98.5	H9WJ96N9013
Beauveria bassiana	2	Sitophilus granarius	Elos Ayias—Patras, W. Greece	98.0	H9PGGF6301R
Beauveria bassiana	147	Sitophilus granarius	Dasylio—Patras, W. Greece	98.0	H9PPYWZ0013
Beauveria bassiana	65	Sitophilus granarius	Dasylio—Patras, W. Greece	98.2	H9PX08B6016
Beauveria bassiana	149	Sitophilus granarius	Dasylio—Patras, W. Greece	99.8	H9PZCCA601R
Beauveria bassiana	BB A	Galleria mellonella	Athens—Central Greece	98.9	20140422CS6P8_H02_2016-04-27
Beauveria bassiana	BB AC	Tribolium confusum	Glafkos—Patras, W. Greece	98.9	20170105CS6P2_F02_2017-01-11
Beauveria bassiana	BB Kar	Galleria mellonella	Hleia—Greece	98.6	20180422CS7P4_C08_2021-03-12
Beauveria varroae	23	Sitophilus granarius	Elos Ayias—Patras, W. Greece	99.6	1E2T99P1 GenBank accession MZ047310
Cordyceps blackwelliae	64	Sitophilus granarius	Dasylio—Patras, W. Greece	96.8	8E2M17WB01R
Cordyceps fumosorosea	IS	Galleria melonella	Ag. Stefanos—Central Greece	96.9	20170105CS10P1_G01_2017-01-11
Metarhizium anisopliae	MET A	Rhyzopertha dominica	Zavlani—Patras, W. Greece	99.1	20140422CS3P2_C02_2016-04-27
Metarhizium anisopliae	MET B	Plodia interpunctella	Glafkos—Patras, W. Greece	99.4	20140422CS3P3_C02_2016-04-27
Metarhizium anisopliae	MET S	Galleria mellonella	Paramali—Cyprus	96.7	20140422CS3P4_C02_2016-04-27
Metarhizium robertsii	MET K	Sitophilus granarius	Galia—Crete, Greece	99.1	H9VK23WY013
Purpureocillium lavendulum	101	Sitophilus granarius	Elos Ayias—Patras, W. Greece	99.0	1E6TE101P1 GenBank accession MZ047311

) were placed in each dish, meaning that 100 individuals per species were tested per soil sample. To sterilize dead insect baits (to prevent saprophytic fungi from growing), they were immersed in a 6% NaOCl₂ solution for 3 s. Mycelia appeared after they were placed in Petri dishes with filter paper impregnated with ddH₂O. Stemi 2000 stereomicroscopes (Carl Zeiss^®, Jena, Germany) were used to examine dead insects, showing external mycelia growth. In order to determine possible entomopathogenic fungi infestation, these samples were kept in special dark chambers at a temperature of 25 °C and observed daily using the stereomicroscope. The cultivation of fungal conidia, which had been removed from the infected baits, was imprinted on the in-Petri dishes on Sabouraud dextrose agar (SDA). Following this, the samples were placed in dark chambers at 25 ± 1 °C for 14 days. Any spore-like growth was inoculated onto new medium for the purification of fungal cultures. The culture purification was performed until the growth of a single colony on SDA plates.

The isolates, after being subcultured several times (at least three) on plates with SDA to ensure purity and monosporic cultures, were morphologically identified using a ZEISS Primo Star microscope (Carl Zeiss Microscopy GmbH, Jena, Germany) at ×400 magnification. At the end of the growth period, transparent tape was used to stick the sporulation structure down from the edge, followed by staining with phenol cotton blue reagent (Sigma-Aldrich 61335, St. Louis, MI, USA). The selected isolates were stored in the microorganism’s repository in the Department of Agriculture, University of Ioannina. The conidia were scraped from the surface of the dead insects using a sterile loop and were transferred to SDA plates [33]. Molecular identification of tested EPF was performed following the method by Rogers and Bendich [34]. A fragment of the ITS spacer region was expanded by applying universal primer sets ITS4 (5′-TCCTCCGCTTATTGATATGC-3′) and ITS5 (5′-GGAAGTAAAAGTCGTAAC AAGG-3′). PCR reactions (25 µL) were performed using Taq 2× Master Mix (M0270) and included working concentrations of 1× Master Mix, 0.5 µM of each primer, and 50 ng of template gDNA. The PCR protocol for amplification of the ITS regions included 33 cycles, at 95 °C for 30 s, 55 °C for 40 s, and 68 °C for 1 min, followed by a final elongation at 68 °C for 5 min. PCR products were kept at 4 °C. The quantity and quality of PCR products were determined by gel electrophoresis using 1% agarose gel, which was stained with SYBR Safe DNA Gel Stain (Invitrogen, Waltham, MA, USA) and visualized under UV light (BIO-RAD, Molecular Imager Gel Doc XR System, Hercules, CA, USA). The purification and the sequencing of the amplified products took place at Eurofins Genomics (Ebersberg, Germany). The similarity of the fungal DNA sequences in the present work with homologous sequences was matched using the Basic Local Alignment Search Tool (NCBI BLAST 2.15.0, https://blast.ncbi.nlm.nih.gov/Blast.cgi, accessed on 10 December 2023).

2.1.2. Lab Culture of Fungal Isolates

SDA was used as a medium for cultivation of EPF isolates in 9 cm Petri dishes for 15 days at 25 °C and 65°H relative humidity. To prevent contamination, Petri dishes were sealed with Parafilm^® (American National Can, Chicago, IL, USA). Fresh conidia were harvested after 15 days by scraping the Petri dish surface with a sterile loop and placing them in a 500 mL glass beaker containing 50 mL of sterile distilled water and 0.05% Tergitol^® NP9 (Sigma-Aldrich, St. Louis, MI, USA). A magnetic stirrer was used to mix the conidial suspension for 5 min after filtering through sterile cloth layers [35]. A Neubauer hemocytometer was used for the measurement of the fungus conidia concentration (WEBER SCIENTIFIC hemocytometer for cell counting, Hamilton Township, NJ, USA). Dilutions were made by adding 10 mL of the conidial suspension to the required amount of sterile water, resulting in a final concentration of 10⁸ conidia per ml for the fungal isolates. This specific concentration was chosen due to its widespread use in numerous relevant studies, and conidial viability exceeded 97% for all fungal isolates.

The viability of all the tested fungi was determined by spreading a 100 μL aliquot of a conidia suspension (1 × 10⁶ conidia mL⁻¹), prepared with a sterile surfactant solution (0.1% v/v) of Tween 80, on SDA medium in Petri dishes (90 × 15 mm) and incubating them in the dark at 25 ± 1 °C. SDA plates of the tested fungi were incubated for 18 h, prior to evaluation. Conidia were scored as viable if any germ tube was 2× longer that the diameter of the spore: a total of 100 conidia per sample under 400× magnification. Conidial viability was calculated based on the formula below:

Viability (%) = [G1/(G1 + G2)] × 100

where G1 refers to the number of germinated conidia, G2 is the number of non-germinated conidia, while the sum of G1 and G2 is equal to 100. Thus, the percentage of viable conidia was determined by counting a total of 100 conidia per fungal sample. Fungal strains presenting ≥ 95% viability were used in the insect bioassays.

2.2. Insect Rearing

The original population of O. linearis was collected during April–July 2022 from infested walnut trees from a walnut orchard located in Kalavryta, Achaia (NW. Greece). More specifically, adults were collected, and infested apical shoots were cut and transferred to the Plant Protection Institute of Patras, Achaia, Greece. Healthy new twigs were coarsely ground to be applied as a substrate for maintenance and rearing of the larvae. The larvae were very carefully removed from the infested twigs and placed in Petri dishes (9 cm Ø), with 10 gr of ground wood for the larvae and 10 gr fresh leaves for the adults (Figure 1A). Adults and larvae were monitored separately, 10 individuals were put on each dish, and the substrate was renewed weekly. No source of water was provided, and the lid was sealed to ensure some airflow. The petri dishes were maintained in complete darkness in a growth chamber (PHC Europe/Sanyo/Panasonic Biomedical MLR-352-PE), under controlled environmental conditions (25 ± 1 °C, 65 ± 5% relative humidity).

2.3. Bioassay

An examination of the virulence of each fungal isolate was carried out in Petri dishes with 10 individuals. To examine larval mortality, 10 3rd-instar larvae with 10 g of sterilized coarsely grounded (0.78 mm) walnut tree wood as food were placed on each dish (Figure 1B). Accordingly, dishes with 10 adults of mixed sex and 10 g fresh leaves, cut into small pieces including the lateral veins, were prepared for adult trials. The sterilization of coarsely ground walnut tree wood was made to avoid outside fungus contamination. Each treatment included five replicates. The insects were sprayed directly with 2.5 mL of conidial suspension (10⁸ conidia/mL) of each of the tested fungi, using a Potter spray tower (Burkard Manufacturing Co. Ltd., Rickmansworth, Hertfordshire, UK) at 1 kgf cm⁻². Larvae and adults treated as controls were sprayed with a solution of sterile distilled water and 0.05% Tergitol^® NP9. During the assay, all the insects were kept under controlled conditions (25 ± 1 °C, 65 ± 5% r.h., complete darkness), as described above.

In the experiments, larvae and the adults were inspected daily for 16 days after treatment. During each inspection, dead larvae and adults were counted, removed, and promptly immersed in a 6% NaOCl₂ solution for 3 s to prevent fungal saprophytic growth, rinsed in sterile distilled water for 5 min, allowed to air-dry, and positioned on moistened filter paper inside a Petri dish. The whole procedure was carried out in a laminar flow chamber (Equip Vertical Air Laminar Flow Cabinet Clean Bench, Mechanical Application Ltd., Athens, Greece). The dead insects were maintained at 25 °C and 65 ± 5% relative humidity in the dark for 5–7 days. Those exhibiting characteristic hyphal growth indicative of EPF were documented as infected. Larvae displaying external mycelial growth were scrutinized using a ZEISS Primo Star microscope (Carl Zeiss Microscopy GmbH, Jena, Germany) at 400× magnification, and the fungi were identified based on the characteristics of hyphal growth, including shape and size [36,37,38,39]. Apart from morphological identification, molecular methods (see Section 2.1.1) were applied to dead insects with fungal infection after treatment.

2.4. Statistical Analysis

Mean values of larval mortality were compared using analysis of variance, with the main factors being fungal isolate and exposure time (days after treatment). Where necessary, experimental data were arcsine-transformed to meet the requirements of parametric analysis for equal variation among treatments. To find statistically significant differences between factors, Tukey’s test was used with a significance level of 0.05. All statistical tests were performed using SPSS (SPSS, Inc., Chicago, IL, USA, version 24). Moreover, the Kaplan–Meier method was applied to determine the mean survival time of experimental insects.

3. Results

A total of 15 isolates were molecularly identified (Table 1). Mycelial and conidial growth on dead insects suggested that almost all deaths were related to the tested fungal species examined (Figure 1C–E). Observations of dead insects showed that external mycelium appeared within the first 72 h after placing them on moist filter paper. This was also validated by PCR. DNA sequences in the present investigation were matched with existing sequence data in GenBank, working with the Basic Local Alignment Search Tool (NCBI BLAST). The effectiveness of the entomopathogenic fungi was significant against O. linearis larvae and adults (

Table 2. ANOVA parameters for mortality levels of O. linearis larvae and adults exposed for 16 days to different strains of EPF in 10⁸ conidia per mL.

Factor	df	F	p Values
Insect biological stage	1	6.643	0.015
Treatment	15	85.373	0.000
Time	15	49.312	0.000
Insect biological stage × treatment	15	59.306	0.000
Insect biological stage × Time	15	9.717	0.004
Treatment × Time	225	8.733	0.000
Insect biological stage × Treatment × Time	225	5.450	0.028

). The three-way factor model of insect biological stage × treatment × time showed a significant effect in terms of the mortality of larvae and adults of the insect (Table 2).

The main effects and interactions for all factors on larval mortality proved to be significant (fungal isolates: F_{15, 959} = 15.232, p < 0.001; exposure time: F_{15, 959} = 31.844, p < 0.001; fungi × time: F_{224, 959} = 3.335, p < 0.001). This indicates that the several fungal isolates affected the survival time of the insect in diverse ways.

Mortality of O. linearis larvae as recorded in our study is presented in Figure 2, indicating high efficacy (>80%) for most of the fungal isolates. Three isolates reached 100% mortality on the tested larvae: Beauveria varroae Vuill. (23), Cordyceps fumosorosea (Wize) (Hypocreales: Cordycipitacae) (Is), and Purpureocillium lavendulum Perdomo, Gené, Cano & Guarro (Hypocreales: Ophiocordycipitaceae) (101). The other thirteen isolates recorded mortality between 66.2% (B. bassiana BB AC) and 98.2% (M. robertsii MET K). The control mortality of O. linearis larvae reached 3.3% at the end of the experiment.

The main effects and interactions for all factors on adult mortality proved to be significant (fungal isolates: F_{15, 959} = 19.112, p < 0.001; exposure time: F_{15, 959} = 42.154, p < 0.001; fungi × time: F_{224, 959} = 1.298, p < 0.001). This indicates that the several fungal isolates affected the survival time of O. linearis adults in diverse ways. Mortality of O. linearis adults as recorded in our study is presented in Figure 3, indicating high efficacy for many isolates. The three isolates that induced the highest mortality on the tested adults were M. anisopliae (MET A) (96.6%), C. fumosorosea (Is) (92%), and M. anisopliae (MET S) (94.4%). The other 13 isolates recorded mortality between 66.6% (M. anisopliae (MET B)) and 90% (B. varroae (23)—M. robertsii MET K). The control mortality of O. linearis adults did not exceed 1.2% until the end of the experiment.

The lowest larval survival time was estimated at 6.42 days for B. varroae (23), 7.64 days for P. lavendulum (101), and 9.12 days for M. anisopliae (MET B) [Wilcoxon (Gehan) statistic = 92.695, df = 16, p < 0.001]. The survival time of the control (untreated) larvae exceeded 15 days (

Table 3. Survival time (days) for the larvae and for adults of O. linearis. Ten larvae/adults were used per replicates for five replicates per isolate. Values with the same letter within a column are not significantly different (p < 0.05).

Treatment	Larvae				Adults
	Estimate	Standard Deviation	95% Confidence Interval		Estimate	Standard Deviation	95% Confidence Interval
			Lower Bound	Upper Bound			Lower Bound	Upper Bound
Control	15.560 a	0.278	15.316	16.104	15.890 a	0.428	15.716	16.096
B. bassiana (147)	14.000 c	0.367	13.281	14.719	13.250 c	0.321	12.771	14.629
B. bassiana (149)	13.180 c	0.387	12.922	13.938	11.290 b	0.355	11.305	12.695
B. bassiana (2)	11.120 b	0.311	11.509	12.731	11.230 b	0.300	10.912	12.588
B. bassiana (BB AC)	14.160 c	0.261	13.448	14.472	13.260 c	0.480	12.059	14.441
B. bassiana (BB K)	11.680 b	0.428	10.802	12.558	11.860 b	0.620	10.784	12.216
B. bassiana (BB)	11.960 b	0.411	11.154	12.766	13.740 c	0.993	13.054	14.946
B. bassiana (Kar)	13.680 cc	0.448	12.802	14.558	13.110 c	0.320	12.784	14.216
B. varroe (23)	6.440 d	0.201	6.047	7.833	11.010 b	0.312	10.388	11.612
C. blackwelliae	13.000 bc	0.409	12.198	13.802	13.590 c	0.384	12.947	14.753
C. fumosorosea (IS)	10.640 b	0.346	9.962	11.318	12.670 bc	0.306	11.900	12.900
M. anisopliae (MET A)	12.040 bc	0.324	12.405	13.675	10.640 b	0.389	10.238	11.762
M. anisopliae (MET B)	9.120 e	0.367	8.401	10.839	13.900 c	0.359	13.297	14.703
M. anisopliae (MET S)	12.640 bc	0.346	11.962	13.318	8.710 d	0.306	7.400	9.600
M. robertsii (MET K)	12.280 bc	0.482	11.336	13.224	11.990 b	0.612	11.617	12.200
P. lavendulum (101)	7.640 d	0.346	6.962	8.318	13.140 c	0.306	12.800	14.600

). The other EPF isolates recorded moderate survival times (>10 days). On the other hand, the lowest adult survival time was estimated for M. anisopliae (MET S) at 8.71 days [Wilcoxon (Gehan) statistic = 61.895, df = 16, p < 0.001], with the rest of the isolates exceeding 10 days. Untreated adults recorded significantly higher survival times (>15 days) (Table 3).

4. Discussion

The practical difficulties in effectively pruning infested trees have led to the decision to use chemical agents to control O. linearis. Spraying with chemical insecticides has been the only route to efficiently manage this pest in walnut and hazelnut orchards [8]. However, the obvious negative impacts of chemicals (risks to the environment and humans, development of resistance, etc.) and the EU’s promoted policy (Farm to Fork strategy) of phasing out chemicals in sensitive environments (forests, cities, etc.) make chemical control increasingly unfeasible [40]. It has, therefore, become evident that biological control agents can play a pivotal role in protecting forest and agricultural ecosystems from wood borers such as O. linearis.

The isolate that stood out was C. fumosorosea (Is) as it induced 100% mortality on both larvae and adults of O. linearis. In contrast, other isolates caused the same effect but only on one of the insect stages. The great variation in the lethal action of some EPF isolates on larvae in comparison with adults is a very interesting outcome. For instance, P. lavendulum (101) demonstrated efficacy against larvae (100%) yet exhibited limited activity against adults (72.8%). Similarly, a screening of 27 EPF isolates on larvae and adults of the onion maggot Delia antiqua (Meigen) (Diptera: Muscidae: Anthomyiidae), revealed that only three isolates controlled larvae effectively, while twelve isolates controlled adults [41]. This discrepancy can be attributed to the role of insect microbiota, which can vary between the developmental stages of the insect [42]. The importance of gut microbiota in regulating the interactions between EPF and host insects has been well documented [43].

The only published work investigating biological methods of managing O. linearis is the one of Bahar and Demirbağ in 2007 [44], in which the bacterial flora of the insect was investigated, and the findings were used to assess the bacteria’s impact. The authors found that Serratia marcescens Bizio (Enterobacterales: Yersiniaceae) induced the highest mortality (65%). Conversely, a spore and crystal mixture of Bacillus thuringiensis Berliner (Bacillales: Bacillaceae), isolated from the European cockchafer Melolontha melolontha (L.) (Coleoptera: Scarabeidae), achieved a remarkable 90% mortality against the pest within a four-day period. There is a paucity of information on the biological control of O. linearis with the majority of the existing literature pertaining to hazel tree infestation, while our study was conducted on walnut. To the best of our knowledge, this is the first study to examine the vulnerability of O. linearis to different EPF isolates, in the respect of searching for non-chemical tools for controlling a serious pest that has become a nuisance for walnut producers. The screening process involved 15 isolates from four EPF genera namely, Beauveria, Cordyceps, Metarhizium, and Purpureocillium. All of them demonstrated a substantial insecticidal effect on both larvae and adults, with mortality rates ranging from 66 to 100%.

The results indicated that all the isolates of B. bassiana had a lethal effect, although this was not as pronounced as in other cases. However, B. varroae (23) resulted in 100% mortality on larvae and 90% on adults. This suggests that fungal virulence is not related to the genus, as observed in the case M. anisopliae, nor is it related to the species. It has been well established that EPF are not universally effective and different strains and isolates even of the same species vary in virulence, possibly due to genetic variability [45]. To elaborate, two M. anisopliae isolates (MET S and MET A) were found to be highly potent against adults, whereas the other M. anisopliae isolate (MET B) was not. The comparative analysis of Beauveria and Metarhizium demonstrated that the latter exhibited greater efficacy against larvae and adults of O. linearis.

Our results cannot be directly compared with other studies due to the complete lack of previous research on the effect of EPF on O. linearis. Nevertheless, some information concerning the impact of EPF on wood-boring beetles has been published. More specifically, A. glabripennis has suffered high adult mortality when exposed to Beauveria [27] and Metarhizium [26]. The same species recorded reduced survival times (ST₅₀) from 5.0 (M. anisopliae and B. brongniartii) to 24.5 (I. farinose) days [29]. In similar screening laboratory bioassays, three isolates (one B. bassiana BbPf VRI 1198 and two M. anisopliae MaPf VRI 0198 and MaPf NRCC 98) proved highly virulent on larvae of the cashew tree borer, Plocaederus ferrugineus L. (Coleoptera: Cerambycidae) causing mortality > 90% at 21 days after treatment [28]. Similar conclusions were reached when the European spruce bark beetle Ips typographus (L.) (Coleoptera: Curculionidae) was treated in the laboratory with strains of B. bassiana and M. anisopliae, where 100% mortality at four days post treatment was recorded [31]. Recently, 23 B. bassiana strains were tested against adults of the Japanese pine sawyer Monochamus alternatus Hope (Coleoptera: Cerambycidae) and three of them showed great virulence given that all treated adults died by the 12th day of the experiment [32]. The same fungal entomopathogen was found to be effective against both adults and eggs of Xylotrechus basifuliginosus Heller (Coleoptera: Cerambycidae). After 12 days of exposure under laboratory conditions, mortality was 71.34% in adults and 86% inhibition in the eggs, while in field conditions adult mortality reached 38.4% [46]. On the other hand, larvae of Osphranteria coerulescens Redtenbacher (Coleoptera: Cerambycidae) were quite resistant (34% mortality) when exposed for 18 days to conidial solutions of Cladosporium sp. [47].

As previously stated, the most effective method of controlling the pest would be the removal and destruction of infested dry shoots from autumn to early spring, before the emergence of adults. However, the excessive height of many hazel and walnut trees makes this practice practically impossible. Additionally, farmers may lack adequate information about this pest, making it challenging for them to differentiate between the damage caused by this pest and the natural drying process of twigs. Conversely, Loru et al. [13] reported that in Sardinia, despite the evident impact of O. linearis on hazel trees, the loss could be mitigated through appropriate pruning. However, this assertion does not apply to Greek walnut trees, as observed in the field. It is evident that for large-scale producers, the removal of infected branches on hundreds of hectares of trees is impractical, given the time and financial constraints involved.

Although the abovementioned data cannot be directly compared with our results, as they involve different insect species, fungal strains, and experimental protocols, they are indicative of the value of EPF as biological control agents against wood-boring beetles. There are several other factors that influence the insecticidal efficacy of EPF, including the insect’s behavior, population density, age, nutrition, and genetic information, as well as the host’s physiology and morphology that affect its sensitivity to biological control agents like EPF [48]. Therefore, the differences in insect susceptibility to EPF cannot be explained solely as a function of the applied conidial concentration [49].

Ιt should also be noted that the host specificity, and especially possible lethal effects of EPF on non-target organisms, is another parameter that must be considered before the practical application of EPF in the field. It has been shown by several studies that many EPF species may have a significant negative effect on pollinators (bees, bumblebees, etc.) and entomophagous insects [50,51,52,53] This obstacle can be overcome by selecting strains that are harmless to beneficial insects. It has been demonstrated that certain EPF isolates did not cause any harm to non-target insects [54].

Given the specificity of the ecological characteristics of these pests, in order to draw safe conclusions, field studies are necessary, in addition to laboratory tests, where the virulence of EPF is very likely to vary with respect to laboratory data. The results should also be useful in selecting natural fungi strains for use against this insect pest. Even though this is only a preliminary investigation into the use of EPF, the fungal isolates we tested showed encouraging insecticidal effects; however, they need to be extensively followed up.