Sublethal concentrations of spinosad synergize the pathogenicity of fungi to larvae of Ephestia kuehniella (Lepidoptera: Pyralidae)

We evaluated the effi cacy of four entomopathogenic fungi (EPF) and their compatibility with the bioinsecticide spinosad for control of Ephestia kuehniella (Zeller) under laboratory conditions. Three EPF, including Beauveria bassiana (Balsamo-Criveili) Vuillemin isolates Z1 and Iran 1395C, Lecanicillium (= Verticillium) lecanii (Zimmerman) Zare & Gams, isolate Iran 229, and Purpureocillium (Paecilomyces) lilacinum (Thom) Luangs-ard, Hywel-Jones & Samson, isolate Iran 1026 were tested against third and fi fth larval instars of Ephestia kuehniella using a fi lter paper bioassay. Mortality caused by the EPF ranged from 63.3–72.5% for third instars and 50–65.5% for fi fth instars, with LT50 ranging from 8.4–10.5 d and 10.1–12.9 d, respectively. The effect of spinosad at LC10 (= 26.2 ppm) on EPF spore germination was evaluated and found to be negligible, ranging from 0% for B. bassiana Z1 to 5.7% for P. lilacinum. The LC50 values for spinosad against third and fi fth instar E. kuehniella larvae were 452.5 and 1446 ppm, respectively. Subsequently, spinosad at LC10 was applied to third instar E. kuehniella larvae 24 h before application of the EPFs at LC50. The addition of spinosad to applications of L. lecanii and B. bassiana Z1 and Iran1395C isolates synergized their pathogenicity to E. kuehniella larvae, whereas the effect was merely additive for P. lilacinum. Our results suggest that these EPF isolates can be used effectively in combination with spinosad for management of E. kuehniella in stored products.

Another promising alternative to synthetic chemical pesticides for protection of stored products are biopesticides of natural origin (Shishir et al., 2015). One such biopesticides is spinosad, a natural insecticide produced by the soil actinomycete bacterium Saccharopolyspora spinosa (Mertz

INTRODUCTION
The Mediterranean fl our moth, Ephestia kuehniella (Zeller) (Lepidoptera: Pyralidae), is a cosmopolitan pest of many stored products, especially grains and fl ours (Ben-Lalli et al., 2011). Feeding and web-spinning by moth larvae fouls stored products and causes signifi cant economic losses (Lynn & Ferkovich, 2004). Control of stored-product pests in commodites currently relies heavily on the use of chemical pesticides, especially fumigants, but this can result in toxic residues and contamination of food products. Given the high costs of these pesticides, and the inevitable evolution of resistance in target pests, there is an urgent need for alternative, more environmentally friendly, management tactics, such as biological control (Talukder, 2009;Pimentel et al., 2010).

Assays of EPF virulence against E. kuehniella larvae
The B. bassiana isolates Z1 and Iran 1395C, the L. lecanii Iran 229 isolate, and the P. lilacinum Iran 1026 isolate, were tested against third and fi fth instar E. kuehniella larvae at the full concentration of the isolated suspension (as above). Larvae were exposed to 0.7 ml of conidial suspension absorbed onto fi lter paper discs (60 mm diam) in plastic Petri dishes (n = 10 larvae per dish, three replicates per concentration) following the procedure of Draganova et al. (2007). Control larvae were exposed to paper discs treated with sterile distilled water containing 0.02% Tween 80. After 24 h of exposure, wheat fl our was added to Petri dishes as a food source. The dishes were then sealed with Parafi lm® and incubated at 26 ± 2°C in the dark. The numbers of live and dead larvae (discolored and/or with mycelial growth evident on the surface) were counted every other day for 14 d.

Assays of spinosad toxicity to E. kuehniella larvae
Spinosad (SP), brand name "Tracer® 24% SC", was obtained from Dow AgroSciences, UK and bioassays were performed on third and fi fth instar E. kuehniella larvae. Five concentrations were selected for testing, based on the results of preliminary trials. Serial dilutions of the formulated compound were prepared on the day of the bioassay using distilled water containing 0.02% Tween 80, plus a water control with Tween 80 only. Each concentration was assayed in three replicates, with 10 larvae per replicate and mortality was recorded 96 h after exposure. This bioassay was conducted twice with the same concentrations and the methodology and conditions used were the same as those described above to assess pathogenicity.

Germination of EPF exposed to spinosad
In this experiment, we assayed the germination of EPF when exposed to spinosad at the LC 10 (26.2 ppm) for third instar E. kuehniella larvae. This concentration was selected on the assumption that additive mortality contributed by the fungi would permit use of a fractional dose of spinosad compared to what would be required to exert effective control alone. The insecticide was dissolved in sterile distilled water containing Tween 80 (0.02%) at the desired concentration and conidia of each EPF were suspended in the aqueous solution of insecticide. Then, 100 μl of each fungal suspension, containing about 1 × 10 7 conidia ml -1 , was spread onto a thin layer of 0.9% water-agar medium in a sterile plastic Petri dish (6 cm diam); conidia suspended in distilled water served as the control. The Petri dishes were incubated at 26 ± 1°C in the dark for 24 h, at which time one hundred conidia were selected at random on each Petri dish and the percentage germinated conidia was quantifi ed according to the methods of Marcuzzo & Eli (2016). The experiment was repeated twice with 7 replicates in each case. The percentage of conidial germination inhibition was calculated in comparison to the control using the formula of Hokkanen & Kotiluoto (1992): where I, C, and P are the percentage of conidial germination inhibition, conidial germination of fungus in the control, and conidial germination of fungus in pesticidal medium, respectively. & Yao, 1990). Spinosad kills insects by hyperexcitation of the nervous system (Snyder et al., 2007) and its effi cacy against E. kuehniella and other stored product pests has been demonstrated in previous studies (Mutambuki et al., 2003;Hertlein et al., 2011;Pozidi-Metaxa & Athanassiou, 2013). Interestingly, larvae of E. kuehniella show a preference for remaining on surfaces treated with spinosad, a response which might help improve its uptake and effi cacy (Athanassiou et al., 2018), and spinosad would appear to be compatible with the parasitoid Habrobracon hebetor, which can also be used for control of E. kuehniella (Mahdavi et al., 2015). Spinosad has a successful history of application against stored product pests (Subramanyam et al., 2014;Nayak & Daglish, 2017) and is often applied in combination with low doses of diatomaceous earth to improve its effi cacy (Machekano et al., 2017(Machekano et al., , 2019Gad et al., 2021) The combined use of this naturally-derived insecticide and an EPF could potentially increase the effi ciency of pest control while minimizing adverse chemical impacts (Paula et al., 2011;Sain et al., 2019). However, the possibility exists that certain insecticides could inhibit the germination or fungal growth of EPF, rendering them incompatible for joint application (da Silva et al., 2013). Therefore, the present study was conducted to evaluate the virulence of different species of EPF against E. kuehniella larvae, and their compatibility with spinosad, to determine the potential utility of combination applications of these agents for management of E. kuehniella in stored products. We hypothesized that the concentration of spinosad required to produce a given level of mortality would be higher for later stage larvae than for earlier stages, so we assayed toxicity for both third and fi fth instars.

Insect rearing
Eggs of E. kuehniella were obtained from the Prominent Insectarium in Ahvaz, Khuzestan Province, Iran. and placed in plastic containers (10 × 6 × 3 cm) containing wheat fl our and bran (10 : 1) and held at 26 ± 1°C, 65 ± 5% RH, in continuous darkness until the desired larval stages (third and fi fth instars) were harvested for use in the bioassays.

Fungal cultures
The EPF used in bioassays were B. bassiana isolates 'Z1' and 'Iran 1395C', the L. lecanii isolate 'Iran 229', and the Purpureocillium lilacinum (Thom) Luangs-ard, Houbraken, Hywel-Jones & Samson isolate 'Iran 1026'. The isolate B. bassiana 'Iran 1395C' was obtained from the Institute of Iranian Plant Protection, Tehran, Iran. Isolates of B. bassiana Z1, P. lilacinum Iran 1026, and L. lecanii Iran 229, were initially isolated from larvae of Spodoptera exigua Hubner (Lepidoptera: Noctuidae) in Nazlu, Urmia, Iran by Dr. Youbert Ghosta, at the University of Urmia. All fungal isolates were cultured in the laboratory on potato-dextrose agar (PDA), at 26 ± 1°C and 16L : 8D photoperiod for two weeks. Afterward, the conidia were scraped from the surface of the fungal cultures and placed in a glass bottle containing 0.02% Tween 80 (Merck, Germany). Subsequently, each suspension was vortexed for 2 min and fi ltered through a single layer of jaconet to

Combined application of EPF and spinosad against E. kuehniella
This experiment tested the effi cacy of fungal isolates for the control of E. kuehniella when combined with a low concentration of spinosad (LC 10 = 26.20 ppm). The bioassay was conducted with third instar E. kuehniella larvae which are signifi cantly more sensitive than fi fth instar larvae to the EPF isolates we tested. Larvae were fi rst exposed to spinosad following the same methodology described above. After 24 h, the larvae were exposed to B. bassiana isolates Z1 and Iran 1395C, L. lecanii Iran 229, and P. lilacinum Iran 1026 at concentrations of 3.6 × 10 9 , 2.63 × 10 9 , 3.81 × 10 9 , and 0.83 × 10 9 conidia ml -1 , respectively using the bioassay method described above. Additional treatments exposed larvae to the fungus alone, the spinosad alone, or a water control. Mortality of larvae was recorded daily for 14 days following the fungal treatment; each treatment was replicated six times with 10 larvae per replicate and larval mortality was corrected for control mortality using Abbott's formula (Abbott, 1925). The corrected mortality data were subjected to one-way ANOVA and then analyzed as a randomized complete block design using the GLM procedure of SAS (SAS Institute, 2003), with means separated by Fisher's LSD test ( = 0.05).

Statistical analyses
Cumulative mortality of E. kuehniella larvae in the EPF virulence assay was fi rst corrected for control mortality (Abbott, 1925) and then analyzed by 2-way ANOVA with 'treatment' and 'larval stage' as independent factors after passing tests for homogeneity of variance (Levine's test) and homoscedasticity (Bartlett's test). The time necessary to produce 50% mortality (LT 50 ) was estimated by probit analysis (SAS Institute, 2003). The data on toxicity of spinosad to E. kuehniella larvae were subjected to probit analysis using SAS software (SAS Institute, 2003) to estimate the median lethal concentration (LC 50 ) and its corresponding 95% confi dence intervals (95% CI).

Assays of EPF virulence against E. kuehniella larvae
All EPF isolates were pathogenic to third and fi fth instar E. kuehniella larvae, with mortality rates ranging from 50 to 72.5% (Table 1). Cummulative mortality was not affected by either EPF treatment (F = 1.06; df = 1,71; p = 0.374) or larval stage (F = 2.45; df = 1,71; p = 0.122), and interaction between these factors is not signifi cant (F = 0.14; df = 3,71; p = 0.934). The shortest estimated LT 50 value was for third instar larvae exposed to B. bassiana Iran 1395C, but this was not signifi cantly different from LT 50 values obtained for the three other EPFs.

Assays of spinosad toxicity to E. kuehniella larvae
A signifi cantly lower LC 50 value was obtained for third instar E. kuehniella larvae than for fi fth instars (Table 2). This confi rmed our hypothesis that later stage larvae would require exposure to a higher concentration than earlier stage larvae to obtain a similar level of mortality.

Combined application of EPF and spinosad against E. kuehniella
The mortality of third instar E. kuehniella larvae was signifi cantly different among treatments (F = 17.34; df = 8,45; p < 0.001). Spinosad alone at LC 10 caused less than 20% mortality of larvae, compared to 60-70% mortality for the various fungal isolates (Fig. 1). However, when combined with spinosad at LC 10 , B. bassiana Z1, B. bassiana Iran 1395C, and L. lecanii all produced mortalities approaching 100%, with the combination of spinsoad plus P. lilacinum producing ca. 80% mortality. Thus, the application of SP in combination with either of the B. bassiana isolates or the L. lecanii isolate had a synergistic effect on E. kuehniella mortality (i.e., more than additive), whereas in combination with P. lilacinum it had merely an additive effect on mortality.   (3) * LC 10 = lethal concentration that killed 10% of the tested population, with 95% confi dence intervals (CI). ** LC 50 = lethal concentration that killed 50% of the tested population, with 95% confi dence intervals (CI).

DISCUSSION
All fungal isolates tested were equally effective against E. kuehniella larvae with no signifi cant differences among them when applied alone. The effi cacy of EPF for control of E. kuehniella and other stored product moths has been reported previously (Bahmani et al., 2012(Bahmani et al., , 2020Sabour et al., 2012;Sohrabi et al., 2019). Draganova & Markova (2006) tested one isolate of L. lecanii (V. lecanii), four isolates of B. bassiana, and two isolates of M. anisopliae against E. kuehniella and observed more lethal effects from the B. bassiana isolates, with only one (B. bassiana 383) producing mortality over 70% with an average LT 50 value of fi ve days. Faraji et al. (2013) reported more than 80% mortality of third instar E. kuehniella larvae when treated with fi ve B. bassiana isolates at 1 × 10 8 conidia ml -1 , with LT 50 values ranging from 107 to 154 h. These different results are likely attributable to multiple causes, including the genetic diversity of the isolates, the origin of collections, differences in methodology used, and possibly differential susceptibility of E. kuehniella source populations. Based on our fi ndings, younger (third instar) larva were more susceptible to spinosad than were older (fi fth instar) larvae. This result suggests that the timing of any application relative to the age demography of the target pest population will affect control effi cacy, and that applications should be made as earlier as possible in an infestation when most larvae are still young and susceptible. Although larval stage did not affect EPF susceptibility signifi cantly in this study, a generally greater susceptibility of early instar larvae to EPF has been reported by other researchers (Navon & Ascher, 2000;Erler & Ates, 2015).
The toxicity of spinosad was signifi cantly greater against third instar E. kuehniella larvae when compared to fi fth instars. Mollaie et al. (2011) reported that spinosad at 0.1-1 mg/kg completely prevented larval survival and adult emergence of E. kuehniella. In another study, Pozidi-Metaxa & Athanassiou (2013) reported 89-94% mortality of E. kuehniella larvae after 25 days of exposure to a 1 ppm concentration of spinosad at three temperatures.
Conidial germination in the presence of an insecticide is an important criteria of their potential compatibility for joint application (Oliviera et al., 2003). In the present study, spinosad at LC 10 (26.2 ppm) did not inhibit germination of the EPF tested, and was thus judged to be safe for the various EPF. Similarly, Asi et al. (2010) found that spinosad was compatible with M. anisopliae and Isaria (Paecilomysces) fumosoroseus and was less inhibitory to conidial germination and mycelia growth of these fungi compared to various other insecticides that spanned a wide range of modes of action. Ericsson et al. (2007) detected a nonsignifi cant increase in the growth rate of M. anisopliae at low concentrations of spinosad, but reduced growth rates of the fungus at concentrations of 192 ppm or higher.
Previous studies have reported additive mortality of insect pests with combinations of EPF and various bioinsecticides. Shakarami et al. (2015) reported that a mixture of B. bassiana (6.3 × 10 4 conidia ml -1 ) and essential oil of Citrus vulgaris (111 l l -1 ) had a synergistic effect in controlling third instar larvae of E. kuehniella. Similarly, Bahmani et al. (2020) found that a mixture of B. bassiana and the microbial insecticide Bacillus thuringiensis kurstaki (BtK), each applied at their LC 50 concentration, resulted in superior control of E. kuehniella larvae compared to separate applications. In another study, combined applications of sublethal concentrations of spinosad (1.5-6.0 ppm g -1 sand) and M. anisopliae (10 4 conidia g -1 sand) caused high mortality and reduced feeding in two wireworm species, Agriotes lineatus (L.), and Agriotes obscurus (L.) (Ericsson et al., 2007). Spinosad has also been applied in combination with B. bassiana against Tribolium confusum, although the results suggested effi cacy similar to the same products applied alone (Athanassiou et al., 2016). In the present study, applications of sublethal concentrations of spinosad signifi cantly increased the susceptibility of E. kuehniella larvae to infection by the two B. bassiana isolates and the L. lecanii isolate, although the infectivity of P. lilacinum appeared unaffected. It has been proposed that additive or synergistic interactions with EPF maybe arise because the insecticide inhibits detoxifying mechanisms an infected insect might otherwise use to clear fungal toxins, thus accelerating its death (Ericsson et al., 2007).
Based on the results of this study, we concluded that these isolates of B. bassiana, L. lecanii, and P. lilacinum have the potential to be integrated with spinosad for management of E. kuehniella in stored products, and that these EPF should be applied while larvae are still in early stages of development. Low concentration spinosad (LC 10 ) synergized the pathogenicity of B. bassiana isolates Z1 and Iran 1395C, and the L. lecanii isolate Iran 229, as they caused greater than additive mortality to E. kuehniella larvae when considering the effects of these treatments separately. Such combined treatments could reduce the amount of spinosad needed to control E. kuehniella and diminish selection pressure on the pest to evolve resistance to the insecticide. Further studies are warranted to determine optimal combinations of EPF and spinosad for controlling E. kuehniella infestations on specifi c products under actual storage conditions.