Background

Ambrosia beetles (Coleoptera: Curculionidae: Platypodinae and Scolytinae) are very important pests on many trees (Hulcr and Dunn 2011). They are wood boring insects that make tunnels through sapwood of trees and feed primarily on symbiotic fungi living in the tunnels (Knížek and Beaver 2007). Adults and larvae feed solely with the fungi, except in some species where larvae also ingest fungus-infested wood (Biedermann 2007).

Turkey is the world’s biggest hazelnut producer, supplying nearly 80% of the total global production. Ambrosia beetles including Anisandrus dispar Fabricius, Xyleborinus saxesenii Ratzeburg, and Xylosandrus germanus Blandford are widespread pests on hazelnut trees in Turkey. Due to their damage, most of the hazelnut trees die finally, especially in orchards at Black Sea coastline in which the ground water level is high (Ak 2016). X. germanus was firstly detected in Turkish hazelnut orchards in 2011 (Knížek 2011) and became one of the major pests in hazelnut orchards with other ambrosia beetles. Afterwards, it was recorded as a pest of kiwi trees in Black Sea Region of Turkey (Ak et al. 2011).

Control methods against these ambrosia beetles, including X. germanus in hazelnut orchards, depend on the use of conventional insecticides against emerged adults, cultural methods (pruning and removing infested branches of trees), and mass trapping of emerged adults through red winged ethanol-baited sticky traps. When insecticide spraying is needed for control, it is necessary to use as many as 6–7 applications because of the beetle’s long prevalence periods in orchards. But, chemical control is obviously unpractical due to its detrimental effects, high cost, and low efficacy in such a large area of hazelnut plantations. For this reason, there is a need for more effective and ecologically safe alternative control methods. Using EPF may have a potential to suppress the ambrosia beetles (Arthurs and Bruck 2017).

EPF are common natural enemies of arthropods worldwide, attracting attention as potential biological control agents. They are more than 700 species (Roy et al. 2006). EPF such as Beauveria bassiana (Bals.) Vuill., Isaria fumosorosea Wize, Lecanicillium spp., and Metarhizium anisopliae (Metch) Sorokin play an important role in the regulation of insect populations (Zimmermann 2007a, 2007b and 2008). Previous studies have showed that they were effective on ambrosia beetles including, X. germanus, A. dispar, Xylosandrus crassiusculus Motschulsky, and Xyloborus glabratus Eichhoff (Castrillo et al. 2013; Carrillo et al. 2015; Kushiyev et al. 2018). Kushiyev et al. (2018) determined that I. fumosorosea at 1 × 108 conidia ml−1 against female adults of X. germanus applied on adults and on branches caused 100 and 90% mortality rates, respectively. EFP may affect not only adults outside the host trees but also larvae inside the beetle galleries (Castrillo et al. 2013).

The aim of the present was to evaluate the efficacy of two conidial concentrations (1 × 106 and 1 × 108 conidia ml−1) of M. anisopliae (TR-106) and B. bassiana (TR-217) against the females of X. germanus, using two different application methods.

Material and methods

Isolation of EPF

The EPF, M. anisopliae (TR-106) and B. bassiana (TR-217), isolates used in this study were obtained from female cadavers of X. germanus collected from hazelnut orchards in Samsun provinces in Black Sea Region in Turkey. The cadavers were observed under a Leica EZ4 stereomicroscope at ×40–70 magnification for fungal growth, and a piece of fungal mat was transferred to water agar (WA) by a sterile needle. The cultures were incubated for 3–5 days at 25 ± 1 °C and 75 ± 5% RH under 16: 8-h L:D photoperiods. Mycelial tips were cut from the edge of the 5-day-old fungal colonies, transferred to 2% malt extract agar (MEA) (Merck), and incubated for a period of 10–14 days under the same conditions. Single-spore isolates were obtained by serial dilution (Dhingra and Sinclair 1995), and the fungi were identified as M. anisopliae and B. bassiana according to Samson et al. (1988) and Humber (1997). The cultures were stored at 4 °C on Sabouraud Dextrose Agar (SDA) (Merck Ltd., Darmstadt, Germany) slants and − 80 °C in cryogenic tubes containing 15% glycerol and deposited in the fungal culture collection of the Mycology Laboratory at the Ondokuz Mayis University, Faculty of Agriculture’s Department of Plant Protection in Samsun, Turkey.

Collection of insects

Hazelnut orchards in Samsun province (Turkey) were inspected to collect hazelnut branches infested with X. germanus during May 2016. Infested branches with beetles were cut (40-cm long) in the field, placed in plastic containers (20 × 25 × 40 cm), and then brought to the laboratory. Infested branches were dissected by pruning scissors, and females were collected from galleries. Collected females were inspected under the Leica EZ4 stereomicroscope at × 40–70 magnification to separate the healthy X. germanus females in order to use them in bioassays. Like other ambrosia beetles, males of X. germanus are in very low sex ratio (about 10:1), flightless, and rarely seen outside the gallery. Thus, females were used in the tests of the study (Ranger et al. 2016).

Inoculum of entomopathogenic fungi isolates

The isolates of M. anisopliae (TR-106) and B. bassiana (TR-217) were propagated on SDA at 25 ± 1 °C and 75 ± 5% RH for 15 days. Conidia were harvested by sterile distilled water containing 0.02% Tween 20. Mycelia were removed by filtering conidia suspensions through four layers of sterile cheesecloth. The concentrations of conidia were adjusted to both 1 × 106 and 1 × 108 conidia ml−1, using a Neubauer hemocytometer under Olympus CX31 compound microscope (Olympus America Inc., Lake Success, NY, USA) for each isolate (Kushiyev et al. 2018).

Assessment of conidial viability

The viability of conidia of the 2 isolates (TR-106 and TR-217) was determined by spreading 0.1 ml of concentration of 1 × 104 conidia ml−1 on 3 separate 6-cm-diameter Petri dishes containing Potato Dextrose Agar (PDA: Oxoid Ltd., Basingstoke, UK) and incubating them at 25 ± 1 °C for 24 h under a 14:10-h L:D photoperiod. After 24 h of incubation, germination was determined by counting 200 conidia/dishes with an Olympus CX-31 compound microscope at × 400 magnification. Conidia were regarded as germinated when they produced a germ tube at least half of the conidial length (Kushiyev et al. 2018).

Virulence of the EPF versus X. germanus females

The experiments were carried out in 2 main tests. In the first, the conidial concentrations (1 × 106 and 1 × 108 conidia ml−1) of TR-106 and TR-217 isolates were applied separately versus the females of X. germanus (2 ml per dish), using a Potter spray tower (Burkard, Rickmansworth, Hertz, UK) and one sterilized hazelnut branch (5 cm length and 1 cm in diameter) was added in each Petri dish. In the second test, the same concentrations of the isolates were sprayed on sterilized hazelnut branches in Petri dishes, and afterwards, the beetles were released on them. Another group of sterilized hazelnut branches was sprayed with sterile distilled water containing 0.02% Tween 20 (2 ml) for the control experiments. Petri dishes were loosely covered by a Parafilm after applications. Females of X. germanus were surface sterilized with 70% ethanol for 10 s before the experiments (Castrillo et al. 2013). The hazelnut branches used in the experiments were autoclaved in autoclavable polyethylene bags (30 × 30 cm) for 1 h at 121 °C on 2 successive days. The Petri dishes lined with moist filter paper (9 cm in diameter) at the bottom were used in all experiments. The spray tower was cleaned with 70% ethanol and sterile distilled water after each application of the fungus concentrations for disinfecting the apparatus. For 8 successive days of experimental observations, different groups of insects were used in 5 replications by 5 insects per Petri dish (n = 25 insects/day/concentration/application method) and so provided independence of the observations on the mortality on over time (Robertson et al. 2007). The same procedure was followed up at the control groups. Different control groups were set for each observation daily. Petri dishes were incubated at 25 ± 1 °C and 95% RH, 16:8-h light:dark photoperiod for 8 days in a BINDER incubator (Model KBWF 240, Germany). All dishes of each observation were inspected daily during 8 days, percent mortality was calculated on each observation day independently, and alive insects were discarded. Dead individuals in which the fungal sporulation is observed were counted under the Leica EZ4 stereo microscope at × 40–70 magnification.

Statistical analysis

Independent-time mortality data from bioassays were analyzed by probity analysis program (POLO-PLUS Ver.2.0) to calculate 50% lethal time (LT50) and 90% lethal time (LT90). In lethal time analysis, Log-Probit analysis calculations were considered in the program. Slopes of regression lines were compared to each other by their standard errors. LT50 and LT90 values of two application methods and two conidial concentrations were compared based on overlapping of 95% fiducial limits.

Results and discussion

Conidial viability of M. anisopliae (TR-106) and B. bassiana (TR-217) was assessed before bioassays, and almost 100% of the conidia of the isolates germinated at the end of the 24 h. The results showed that the 2 isolates were pathogenic to females of X. germanus in the 2 different tested concentrations (1 × 106 and 1 × 108 conidia ml−1) and application methods (on branch and direct on insect).

Considering the application method, a significant difference was observed between the 2 concentrations of the isolate (TR-106) applied to X. germanus (P < 0.05). The concentration (1 × 108 conidia ml−1) was more effective than the other on X. germanus at both application methods (branch and insect). Following application with 1 × 106 conidia ml−1 of the TR-106 isolate of M. anisopliae, the LT50 value for X. germanus was 5.45 and 6.67 days, respectively, for branch and insect applications. Respective values at 1 × 108 conidia ml−1 were 3.97 and 4.43 days, respectively. However, the LT90 values for 1 × 106 conidia ml−1 of TR-106 isolate were 10.41 and 11.25 days, while the same values for 1 × 108 conidia ml−1 of the isolate were 5.68 and 6.01 days for application on branches and direct on insects, respectively (Table 1).

Table 1 Probit analysis data on mortality time of Xylosandrus germanus females after applications of two conidial concentrations of Metarhizium anisopliae TR-106 isolate on hazelnut branches and beetle females
Full size table

On the 1st day after treatment, there was no mortality for TR-106 isolate in the 2 tested concentrations applied in the 2 methods (Fig. 1). In the 2 methods, mortality rates were low during the 1st 3 days but death increased after the 3rd day (Fig. 1). Meanwhile, the mortality rates of X. germanus treated with 1 × 106 conidia ml−1 of the isolate reached 80 and 64% at the end of the 8th day on branches and directly on insects, respectively. The isolate caused 100% mortality rate at the end of the 8th day in the 2 application methods with 1 × 108 conidia ml−1. Moreover, all applications of TR-106 isolate caused approximately 100% mycosis rate on dead adults of X. germanus.

Fig. 1
figure 1

Mortality rates of Xylosandrus germanus treated with Metarhizium anisopliae TR-106 at different methods and conidial concentrations. a 1 × 106 conidia ml–1 on the branch. b 1 × 106 conidia ml–1 on the insect. c 1 × 108 conidia ml–1 on the branch. d 1 × 108 conidia ml–1 on the insect

Full size image

The efficacy of B. bassiana (TR-217) used in this study was found to be lower than that of M. anisopliae (TR-106). The LT50 values of TR-217 isolate at 1 × 106 conidia ml−1 were 8.38 and 9.57 days, while the LT90 values were 16.78 and 20.30 days for treated branches and for application directly on the insects, respectively. Respective LT50 values for 1 × 108 conidia ml−1 of the isolate were 5.96 and 6.03 days, while they were 11.79 and 10.80 days for LT90 (Table 2).

Table 2 Probit analysis data on mortality time of Xylosandrus germanus females after applications of two conidial concentrations of Beauveria bassiana TR-217 isolate on hazelnut branches and beetle females
Full size table

The mortality rates when applied at 1 × 108 conidia ml−1 of B. bassiana recorded 64 and 80% at the end of the 8th day on treated branches and insects, respectively. However, the efficacy of 1 × 106 conidia ml−1 B. bassiana (TR-217 isolate) against the beetle was lower than those of the highest concentration used in the study of the 2 application methods (Fig. 2). The mortality rates of X. germanus females at 1 × 106 conidia ml−1 of B. bassiana remained around 40.0 and 45.0%, even on the 8th day after treatment. Similar to M. anisopliae, B. bassiana isolate caused about 100% mycosis rate on females in all treatments.

Fig. 2
figure 2

Mortality rates of Xylosandrus germanus treated with Beauveria bassiana TR-217 at different methods and conidial concentrations. a 1 × 106 conidia ml–1 on the branch. b 1 × 106 conidia ml–1 on the insect. c 1 × 108 conidia ml–1 on the branch. d 1 × 108 conidia ml–1 on the insect

Full size image

LT50 and LT90 values of M. anisopliae (TR-106) and B. bassiana (TR-217) isolates decreased with increasing the conidial concentration at all applications. However, comparisons of the confidence intervals indicated insignificant differences between LT90 values at both concentrations of B. bassiana TR-217 isolate in the 2 application methods. The results of this study are in line with those of previous studies as many researchers have reported that efficacy of EPF increases according to concentration increase. Therefore, the LT50 and LT90 values of the fungi decreased (Kocaçevik et al. 2016 and Kushiyev et al. 2017, 2018). Kushiyev et al. (2017) found that LT50 and LT90 values of M. anisopliae TR-106 and B. bassiana TR-217 isolates applied to females of A. dispar decreased by increasing conidial concentration.

In the present study, there was insignificant difference between both application methods (P > 0.05) with regard to infectivity. Similarly, some researchers found insignificant differences in the mortality rates of X. glabratus females, treated with B. bassiana strain (GHA) and I. fumosorosea strains (Ifr 3581 and PFR), by dipping in fungal concentration or placing on treated avocado branches (Carrillo et al. 2015). In another study, Kushiyev et al. (2018) investigated the efficacy of I. fumosorosea TR-78-3 at 1 × 106 and 1 × 108 conidia ml−1 against females of A. dispar by treating the plant branches and direct spray on insects and found insignificant differences between the two application methods.

Liu et al. (2017) stated that B. bassiana (12,108 isolates) was more effective than M. anisopliae (3.4607 isolate) on adults of Curculio nucum (Coleoptera: Curculionidae). The difference between obtained results and that of Liu et al. (2017) may be derived from using different isolates. Consequently, the efficacy of EPF may vary between species, within a genus, and even between isolates of the same species (Goettel et al. 2005).

EPF can be effective when applied directly on insect pests or when they are later acquired by the insects from treated surface; therefore, ambrosia beetles, which spent the majority of their life cycles inside the host trees, are ideal target for EPF (Castrillo et al. 2013) and it can be observed lately that studies on ambrosia beetle-EPF interactions were increased. Commercial strains of B. bassiana (GHA and Naturalis) and M. brunneum (F52) applied on females of X. crassiusculus caused 77.96 and 79% mortality rate, respectively, on the 5th day after application under laboratory conditions, and also, the beetle females exposed to peach stems treated with the fungi had low survival capacity (Castrillo et al. 2013).

Conclusion

The present findings determined that EPF, M. anisopliae (TR-106) and B. bassiana (TR-217), seemed to be promising one of biological control agent against the invasive ambrosia beetle, X. germanus. However, further studies are necessary to evaluate the efficacy of the isolates on the treated pest under field conditions.