Ιnteractions between Beauveria bassiana and Isaria fumosorosea and Their Hosts Sitophilus granarius (L.) and Sitophilus oryzae (L.) (Coleoptera: Curculionidae)

The interactions between the entomopathogenic fungus Beauveria bassiana Balsamo (Vuillemin) (Hypocreales: Cordycipitaceae) and the entomopathogenic fungus Isaria fumosorosea (Wize) Brown and Smith (Hypocreales: Clavicipitaceae) were examined on young adults of Sitophilus granarius (L.) (Coleoptera: Curculionidae) and S. oryzae (L.) (Coleoptera: Curculionidae). Conidial suspensions of these entomopathogenic fungi were applied both separately and in combination, at three dosages, 104, 106, and 108 conidia/mL. Mortality of experimental adults was recorded daily for 15 days. An overall positive interaction between the pathogenic microorganisms was observed. Mean weevil mortality caused by the separate acting fungi, B. bassiana, ranged from 26.7% to 53.3% and from 36.6% to 63.3% for S. granarius and S. oryzae, respectively. The respective values for I. fumosorosea were 20.0%–53.3% and 46.7%–66.7%. The combined treatments showed a distinct interaction between the pathogens; for S. granarius, the interaction between the pathogens was additive in all combinations, whereas, for S. oryzae, the interaction was additive in seven and competitive in two of the combinations. Applying both entomopathogenic microorganisms may offer a method for weevil control that could be more effective than using each pathogen alone.


Introduction
Sitophilus granarius (L.) (Coleoptera: Curculionidae) and S. oryzae (L.) (Coleoptera: Curculionidae) are the most important storage pests of raw cereals throughout the world. To control these pests, synthetic insecticides are used during the storage of grains [1,2]. Considering the dire effects for both humans [3] and the environment [4], health authorities are wary of the use of chemical insecticides on grains [5]. The control of S. granarius and S. oryzae is currently based mainly on the use of two broad categories of insecticides: residual insecticides and fumigants. Sitophilus species have been reported to develop resistance to synthetic chemicals [6,7]. The development of resistance to these substances and the demands of consumers for residue-free food have led researchers to evaluate the use of alternative control methods that do not leave residues on the product and are generally safe for the environment. The conidial suspensions were stirred using a magnetic stirrer (Bande Stirrers magnetic stirrer MS300, Bante Instruments Inc., Sugar land, TX, USA) and filtered twice using a sterile cloth. Suspensions were adjusted according to Gurulingappa et al. [38] using a Neubauer hemocytometer (TIEFE 0, 100 mm 1/400 9 mm). Following Goettel and Inglis [39], the viability of conidia was determined after 24 h. The germination test was run for every stock suspension in order to ensure the constancy of the viability assessments. The average viability of conidia was for I. fumosorosea 98.7% and B. bassiana, 96.9%. Preparation of conidial suspensions and conidial germination took place in a laminar flow chamber (Equip Vertical Air Laminar Flow Cabinet Clean Bench, Mechanical Application Ltd. Athens, Greece).

Bioassays
The virulence of each fungus was investigated separately on S. granarius and S. oryzae adults which were treated with three different conidial concentrations from I. fumosorosea (If) and B. bassiana (Bb) (10 4 , 10 6 , and 10 8 conidia /mL). Each fungus was applied separately. Ten weevil adults (3-5 days old) were collected from lab cultures and transferred in sterile Petri dishes (9-cm diameter) with a single layer of hard wheat grains. Experimental adults were sprayed with 2 mL of the desired conidial suspension using a Potter spray tower (Burkard Manufacturing Co. Ltd., Rickmansworth, Hertfordshire, UK) at 1 kgf cm −2 . Following this, Petri dishes were kept in incubators set at 25 ± 1 • C and 65% relative humidity during the entire experimental period. Adults were observed daily, and mortality was recorded for 15 days. Adults that were sprayed simply with an aqueous solution with 0.01% Tween ® 80 (Sigma-Aldrich ® , Munich, Germany) were used as control. Each treatment (Petri dish with 10 adults) was replicated 10 times (n = 100 adults for every treatment). The applications of pathogens within each replicate were done at the same time.
The combined effect of the two Hypocreales fungi was tested on adults exposed to all nine different combinations of the three tested conidial concentrations. Experimental adults were initially sprayed with 2 mL conidial suspension of the one pathogen and, after 2 s, they were sprayed with 2 mL conidial suspension of the other pathogen. The spraying sequence (meaning which fungus was applied first) changed from a Petri dish to the next one. This was necessary so as to exclude any variation in our results because of the time of fungus colonization. The experimental procedure (number and age of experimental adults, number of replications, recording of mortality, etc.) were the same as in the case of the separate fungus study (described above).

Microscopic Fungal Identification Method
Isolates were sub-cultured several times to ensure that purity and monosporic cultures from all isolates were obtained, and they were then morphologically identified by a microscope ZEISS Primo Star (Carl Zeiss Microscopy GmbH, Jena, Germany) at 400× magnification.

Mathematical Estimation
The interaction between the pathogens was estimated using the formula of Robertson and Preisler [40]: where P E is the expected mortality induced by the combination of the two pathogens; P 0 is the observed mortality of the control; P 1 is the observed mortality caused by the first pathogen (separate action); P 2 is the observed mortality caused by the second pathogen (separate action). Distribution was determined by the chi-square formula: where L 0 is the number of recorded live larvae of the control, D 0 is the number of recorded dead larvae of the control, L E is the expected number of live larvae, and D E is the expected number of dead larvae (estimated like P E with Equation (1)). The formula was used to test the hypothesis independent-simultaneous relationship (df = 1, p = 0.05). If χ 2 < 3.84, the ratio is defined as additive, if χ 2 > 3.84 and the observed mortality is higher than expected, the relationship is defined as synergistic.
On the contrary, if χ 2 > 3.84 and the observed mortality is less than expected, the relationship is defined as competitive [14].

Statistical Analysis
Prior to analysis, these mortality values were arcsine transformed. Mortality data were then analyzed by means of univariate ANOVA using the general linear model of the IBM (version 23.0, SPSS Inc., Armonk, NY, USA). In case of significant F values, means were compared using the Bonferroni test. The significance level was set at p < 0.05. Kaplan-Meier analysis was also selected to determine the median survival time of S. granarius and S. oryzae individuals following exposure to the pathogens which had been applied both separately and in combination. Comparison of median survival time was performed using one-way ANOVA (Treatment as Factor) (SPSS v.23.0).
As far as the combined action of the two entomopathogens is concerned, a total of nine combined treatments of B. bassiana and I. fumosorosea were applied against S. granarius and S. oryzae. Adult mortality of both S. granarius and S. oryzae varied significantly among the various dose combinations (S. granarius: F: 1.755; df: 8.270; p: 0.086; S. oryzae: F: 1.301; df: 8.270; p: 0.044) ( Table 2). On the contrary, no significant differences in mortality were noted between the two insect species (F: 1.540; df: 1.720; p: 0.214) when they were treated with the same combination ( Table 2).
The results of the combined treatments showed a distinct interaction between the pathogens. In the case of S. granarius, the interaction between the pathogens was additive in all combinations (Table 3). On the other hand, pathogens demonstrated an additive interaction when infecting S. oryzae, in seven combinations while, in two of the treatments, the interaction was characterized as competitive (Table 3). No synergistic relationship was recorded in any of the combinations.

Fungal Sporulation
The percentage of dead insects that did not show sporulation from any of the fungal pathogens varied from 3% to 70% (S. granarius) and from 2% to 68% (S. oryzae) ( Figure 1). The highest fungal sporulation occurred on the cadavers of the combinations B (10 6 Bb × 10 6 If) and C (10 4 Bb × 10 8 If) in both insects. The two pathogens proved to be equivalent in sporulation, given that cadavers infected with B. bassiana were more in half of the combined treatments, with I. fumosorosea sporulating more in the other half ( Figure 1). The adults that were inoculated with the single treatments of these fungal pathogens exhibited confirmed mortality with the typical fungal symptoms of I. fumosorosea and B. bassiana in 82% and 79% of the adults, respectively.

Fungal Sporulation
The percentage of dead insects that did not show sporulation from any of the fungal pathogens varied from 3% to 70% (S. granarius) and from 2% to 68% (S. oryzae) (Figure 1). The highest fungal sporulation occurred on the cadavers of the combinations B (10 6 Bb × 10 6 If) and C (10 4 Bb × 10 8 If) in both insects. The two pathogens proved to be equivalent in sporulation, given that cadavers infected with B. bassiana were more in half of the combined treatments, with I. fumosorosea sporulating more in the other half ( Figure 1). The adults that were inoculated with the single treatments of these fungal pathogens exhibited confirmed mortality with the typical fungal symptoms of I. fumosorosea and B. bassiana in 82% and 79% of the adults, respectively.

Medial Survival Time
Kaplan-Meier analysis showed that the median overall survival time for experimental adults was 11.557 ± 0.18 days for S. granarius and 10.692 ± 0.21 days for S. oryzae (Figure 2). The median survival time of S. granarius adults with the separate doses was between 9.13 and 12.68 days compared to S. oryzae adults whose median lethal time was between 7.97 and 12.65 days (Table 1). For the combined doses of the pathogens, the median survival time of S. granarius adults was between 8.00 and 11.8 days compared to S. oryzae adults whose median survival time was between 6.36 and 10.37 days in the same combinations (Table 2).

Medial Survival Time
Kaplan-Meier analysis showed that the median overall survival time for experimental adults was 11.557 ± 0.18 days for S. granarius and 10.692 ± 0.21 days for S. oryzae (Figure 2). The median survival time of S. granarius adults with the separate doses was between 9.13 and 12.68 days compared to S. oryzae adults whose median lethal time was between 7.97 and 12.65 days (Table 1). For the combined doses of the pathogens, the median survival time of S. granarius adults was between 8.00 and 11.8 days compared to S. oryzae adults whose median survival time was between 6.36 and 10.37 days in the same combinations (Table 2).

Medial Survival Time
Kaplan-Meier analysis showed that the median overall survival time for experimental adults was 11.557 ± 0.18 days for S. granarius and 10.692 ± 0.21 days for S. oryzae (Figure 2). The median survival time of S. granarius adults with the separate doses was between 9.13 and 12.68 days compared to S. oryzae adults whose median lethal time was between 7.97 and 12.65 days (Table 1). For the combined doses of the pathogens, the median survival time of S. granarius adults was between 8.00 and 11.8 days compared to S. oryzae adults whose median survival time was between 6.36 and 10.37 days in the same combinations (Table 2).
Significant differences in mortality caused by the separate action of each entomopathogen were detected not only between the different insects but also between the two fungal entomopathogens. Generally, I. fumosorosea was more virulent for S. oryzae and B. bassiana was more virulent for S. granarius. Although mortalities between different host and pathogen species, in our study, did not always differ significantly, it has been well established that these two factors (insect and pathogen species) play an important role in the insecticidal efficacy of entomopathogenic fungi [1,2,32,43,44].
An overall positive interaction between the two pathogens was noticed in terms of adult mortality, especially for S. oryzae. Based on our results, the interaction of I. fumosorosea with B. bassiana was additive for S. granarius in all combinations. On the contrary, in two treatments, the interaction between the pathogens was negative (competitive) for S. oryzae. In our combined treatments, the insect species did not have a significant effect on host mortality.
Competitive interaction was observed in two of the treatments, in the case of S. oryzae. A competitive interaction refers to the negative relationship between pathogens. The nature of competition between entomopathogenic fungi is not well known. The competitive interaction is predominately mediated by resources in the host and the extent to which these benefit the entomopathogens. Mietkiewski and Gorski [45], for instance, report that when the two entomopathogenic fungi are used concurrently with other biological insecticides, they exhibit synergy, competitiveness, or a neutral interaction. Staves and Knell's [46] findings suggest that if mixed infection persists, then the type of interaction (direct and/or indirect) within the host can play a major role in determining how an entomopathogen will evolve in terms of its virulence. When the combination of two entomopathogenic fungal isolates was tested, all possible effects (additive, synergistic, competitive) were recorded [31][32][33] depending on the host and pathogen species, temperature, and pathogen concentration.
The dead adult percentage showing pink muscardine of I. fumosorosea seemed to be higher in 9 of the 18 mixed fungal infections, while the white muscardine of B. bassiana on adult cadavers appeared at a higher percentage in the other 9 mixed infections. Our results support Staves and Knell [46] in their suggestion that the dominance outcome during a multiple entomopathogenic fungal infection is not defined solely by the virulence of the entomopathogenic fungi. Moreover, as is reported in an older study, in all the combination treatments with two fungal entomopathogens, only one of the fungi sporulated on the larval cadaver, never both [31].
It is possible to accomplish significant pest control results using a mixture of entomopathogens, either by combining systematically close microorganisms or systematically remote organisms [15]. Several studies have described interactions between different pathogens within the same species [14,18,[20][21][22][23][47][48][49][50]. These interactions may affect pest mortality antagonistically (reducing the observed mortality of hosts as compared to single pathogen infections [29]), synergistically (increasing host mortality in comparison to single pathogen infections [16,22,28]), or they may vary per genotype, dose, and order of infection [47,48,51]. Infections of insects by more than one pathogen usually lead to an increase in host mortality [15] as it was proven in most cases in the present study.
It was our hypothesis that in the interaction between entomopathogenic fungal isolates, one isolate would dominate as more virulent while the less virulent would still play an auxiliary role in the infection process. Insects would thus die from a reinforced infection and the cadavers would exhibit signs of the more virulent fungus. Our results support this theory.
Our experiment demonstrates that although B. bassiana and I. fumosorosea are each on their own able to reduce the population of S. granarius and S. oryzae, the combination of the two could increase host mortality, thus providing a more effective pest control method than using each pathogen alone.
The impact of fungal epizootics on host populations can be very dramatic and many attempts have been made to harness this potential for pest-control purposes.

Conclusions
We tested the hypothesis that the co-application of B. bassiana and I. fumosorosea could be used to increase the mortality, which each fungus alone would individually incite in Sitophilus weevils. We observed that the final mortality was greater for the combination treatments than for the separate treatments, in most cases. While this was a relatively simple simulation, it demonstrates the potential value of utilizing fungal "cocktails" as effective biological tools for pest management. As many entomopathogenic fungi will have some impact upon the host's immune system, understanding the nature of this impact will be critical for understanding the dynamics of the interaction. Nevertheless, further studies are needed to investigate the mechanism of toxicity of such combinations against serious stored pests.