Screening and Virulence of the Entomopathogenic Fungi Associated with Chilo suppressalis Walker

The current study aimed to explore the entomopathogenic fungi associated with the larvae of Chilo suppressalis Walker, a serious pest of rice, in northern Iran. The collected specimens were cultured and identified through morphological and molecular methods. The 38 specimens were identified by microscopic examination and genetic sequencing of the ITS region as follows: twenty-one isolates of Beauveria bassiana, five isolates of Akanthomyces lecanii, four isolates of Akanthomyces muscarious, three isolates of Metarhizium anisopliae, two isolates of Hirsutella subulata, two isolates of Trichoderma sp. and one isolate of Aspergillus sp. All the identified isolates were treated on the larvae through bioassay, evaluating the amount of hydrophobin and the activities of proteases, chitinases and lipase to find their virulence. Moreover, the percentage of thermotolerant and cold activity of the isolates were tested to determine their environmental persistence. The overall results revealed the isolates of B. bassiana, including BBRR1, BBAL1 and BBLN1 as the most virulent and environmental adaptive isolates among the fungi associated with C. suppressalis.


Introduction
The rice striped stem borer, Chilo suppressalis Walker (Lepidoptera: Crambidae), is an economic pest of rice, annually causing significant losses in Asia, southern America and northern Africa [1]. The larvae feed intensively on rice stems and cause "whitehead" and "dead-heart" of the seedlings, which directly decrease the overall yield of rice [2]. The main control measure to suppress the C. suppressalis population is the wide-spraying of synthetic insecticides, including diazinon, Padan ® and Reagent ® . Nevertheless, C. suppressalis has developed resistance to these insecticides on one hand and resulted in environmental pollution, food residuals and toxicity on non-target organisms on the other hand [3,4]. These concerns should change the management strategies of chemical insecticides toward biocontrol agents like entomopathogens. Among the entomopathogens used to manage the population of insect pests, entomopathogenic fungi cause epizootics among insect pests and appear as the prevalent natural pathogens to regulate population fluctuations of pests and subsequent losses [5]. Their presence in almost all terrestrial and aquatic ecosystems, as well as way of infection by producing different extracellular enzymes and by releasing toxic secondary metabolites, has led to the success of entomopathogenic fungi to affect noxious arthropods in agriculture, forestry and livestock [5,6].
Exogenous isolates of the entomopathogenic fungi that were commercialized as pest biocontrol agents in different countries may be ineffective on some pests due to environmental suitability and strain differences related to the host [24]. Therefore, the application of local isolates may be a promising choice mainly in case of ecological suitability with pest species and lower hazards on non-target organisms compared to exotic strains [22,[25][26][27]. These points were verified by several studies that demonstrate the virulence of isolates belonging to the same fungal species could be different because of genetic variations occurring in a specific geographical distribution [28][29][30]. The provinces of Guilan and Mazandaran are located in the north of Iran with high humidity, moderate annual temperatures and heavy rainfall, in which these conditions are appropriate for entomopathogenic fungi [5]. The rice fields of northern Iran, known as a reservoir of C. suppressalis [31], can represent ideal sites to study the existence of entomopathogenic fungi with natural enzootics to C. suppressalis, so the aims of our study were to; (a) isolate and identify different entomopathogenic fungi from fungus-infected C. suppressalis larvae, (b) evaluate the virulence of these fungi against the larvae of C. suppressalis, (c) examine the infection process of these isolates by the production of extracellular secretions and (d) compare the conidial germination of the fungal isolates after exposure to heat and cold.

Collection and Morphological Identification
The collection sites were all the municipal regions of Guilan and Mazandaran provinces in the north of Iran (Mazandaran and Guilan, Iran) with the highly cultivated area of rice. In each site, the remained stems of rice within the paddy fields were opened, and the infected larvae of C. suppressalis were collected and kept in sterile centrifuge tubes. The infected larvae were recognized according to the mycelial growth outside the larval body. Once the samples were transferred to the laboratory, the larvae were surface disinfected with sodium hypochlorite (2%) for 3 min and rinsed three times in sterile distilled water [27]. The larvae were then transferred on potato dextrose agar (PDA, Merck) plates and incubated at 25 • C for 2-4 days for fungal development. Afterward, the fungal mycelia were picked up and transferred to fresh PDA plates for purification. Finally, single-spore cultures were gathered according to the method described by Fang [32] and cultured on PDA slants. All collected specimens were inoculated on PDA plates and incubated at 25 • C in the dark for 14 days. For microscopic examination, mycelia and conidia from fungal specimens were mounted on a sliding glass and observed at 100× magnification on a phase-contrast microscope (Canon INC DS126311, Taiwan). Morphological identification of the specimens was made based on conidial morphology, shape, color and size based on the following literature: Akanthomyces spp. isolates [28,33,34], Beauveria spp. Isolates [28,34,35], Hirsutella spp. isolates [34,[36][37][38] and Metarhizium spp. isolates [34,39].

Genomic DNA Extraction and PCR
DNA extraction was done using the protocol of Montero-Pau et al. [40]. Briefly, the mass mycelia of the specimen grown in PDA media were transferred to the 1.5 mL tubes containing 100 µL of alkaline lysis buffer (0.2 mM disodium ethylene diamide tetraacetic acid, 25 mM NaOH, pH 8.0, Merck) and centrifuged for 30 min at 2000× g. Then, the tubes were incubated at 95 • C for 30 min and cooled on ice for five min. Finally, 100 µL of Tris-HCl solution (Sigma-Aldrich, Vienna, Austria; 40 mM, pH 5.0) was added to the tubes, vortexed and maintained at −20 • C. The extracted solution was used as a template for PCR.
To amplify the internal transcribed spacers (ITS5-5.8S-ITS4), ITS5 (5 GGAAGTAAAAG TCGTAACAAGG3 ) and ITS4 (5 -TCCTCCGCTTATTGATATGC-3 ) primers were synthesized as previously described [41]. The PCR reaction mixture consisted of 12.5 µL of master mix (Including 10× PCR buffer. MgCl2, dNTPS TaqPolymerase, CinnaClone, Tehran, Iran), 7.5 µL of double-distilled H 2 O, 1 µL of each primer and 3 µL of DNA solution. PCR was carried out using a thermal cycler (Eppendorf Personal, Darmstadt, Germany) with the following reaction parameters: an initial denaturation for 2 min at 94 • C, 30 cycles of 94 • C for 30 s, 53 • C for 30 s and 72 • C for 1 min and a final extension for 5 min at 72 • C. Amplified PCR products were visualized by electrophoresis on 1% agarose gels. The PCR products were sent to a sequencing service company (Royan Zistagene Co., Tehran, Iran) for purification and sequencing. Finally, sequences were compared with other fungi using the BLAST search tool in NCBI (https://blast.ncbi.nlm.nih.gov/Blast.cgi).

Insect Rearing
The stock population of C. suppressalis was established by collecting the egg patches from rice fields of Amol, northern Iran. The eggs were kept in a growth chamber at 25 ± 2 • C, 85 ± 5% R.H. and 16:8 (L:D) h of photoperiod. The newly hatched larvae were transferred to the tubes (20 × 15 cm) supplied by rice seedlings of Hashemi variety. Quality of food was monitored every day, and the old cutting stems were replaced by the fresh ones [42]. Rearing was continued for three generations at the same controlled conditions.

Bioassay
Conidia of the two-week-old PDA cultures of the identified isolates were removed by a scalpel then suspended in sterile distilled water containing Tween-80 (0.02%). The concentrations of 10 2 to 10 8 conidia/mL from each isolate were separately prepared based on the preliminary tests. Early fourth instar larvae of C. suppressalis were randomly selected and separately dipped in the serial concentrations of each isolate while the control larvae were dipped in an aqueous solution of 0.02% Tween-80 alone. The bioassays were done in three replicates with ten larvae per replication, and the larvae were maintained at the rearing condition for the whole bioassay period. Mortality was recorded within 7 days, and LC 50 values were determined using POLO-Plus software. For calculation of LT 50 , mortality was recorded until the death of all larvae at 10 8 conidia/mL concentration.

Hydrophobin Protein Extraction and Estimation
Hydrophobin content was determined according to the method described by Ying and Feng [43]. Conidia from the two-week-old cultures were added to 1 mL of 2% SDS aqueous solution containing β-mercaptoethanol (5%, Merck) and incubated in a boiling water bath for 10 min before being centrifuged at 22,000× g and 4 • C for 10 min. The supernatant was removed, and conidia were rinsed twice in distilled water to eliminate SDS (Merck) soluble proteins. Samples were incubated in 1 mL formic acid at zero temperature for 2 h before being centrifuged at 22,000× g and 4 • C for 10 min. The supernatant was transferred into fresh tubes, and 0.5 mL of distilled water was added to the samples. Afterward, 0.75 mL of 45% NaOH solution added to the mixture and maintained at 4 • C overnight. The proteins were separated from the supernatant by centrifugation at 22,000× g and 4 • C for 10 min. The extractable proteins of formic acid were rinsed twice with ethanol solution (3:1, v/v) and then dissolved in 2% SDS to quantitatively determine the amount of protein as mg/mL of conidia using the procedure of Lowry et al. [44].

Liquid Culture for Enzyme Production of the Isolates
The liquid media used for biochemical production of the extracellular enzymes contained; 0.02% of KH 2 PO 4 ; 0.01% of CaCl 2 ; 0.01% of MgSO 4 ; 0.02% of Na 2 HPO 4 ; 0.01% of ZnCl 2 and 0.01% of yeast extract (Merck). The media were inoculated with 1 mL of 10 8 conidia/mL concentration of each isolate separately and 5% (weight) of larval cuticle was added to each flask containing liquid media. Then the flasks were kept on a rotatory shaker (70 rev/min) for 8 days at 25 ± 1 • C [3].

Sample Preparations for Enzymatic Assays
After 8 days, the culture flasks were harvested by centrifugation at 10,000× g for 30 min and washed in ice-cold Tris-HCl (25 mM, pH 8). Weighed mycelia were ground to a fine powder, suspended in DW, homogenized and centrifuged at 22,000× g and 4 • C for 30 min to obtain the supernatant of enzyme assay [12].

Lipase Assay
Lipase assay was done using the method of Tsujita et al. [45]. Fifty microliters of p-nitrophenyl butyrate (27 mM, Sigma-Aldrich, Co., Vienna, Austria) as substrate, 20 µL enzyme solution and 100 µL of Tris-HCl buffer (20 mM, pH 7) were incorporated and incubated at 37 • C for 5 min. Then, 100 µL of NaOH (1 N) was added to each tube, and the absorbance was recorded at 405 nm.

Endochitinase Assay
Twenty microliters of the enzyme solution were added to 50 µL of 0.5% colloidal chitin (Sigma-Aldrich, Co., Vienna, Austria) as substrate and 100 µL of Tris-HCl buffer (20 mM, pH 7). Then, the samples were incubated in a water bath of 30 • C for 60 min. Then, 100 µL of dinitrosalisylic acid (DNS, Sigma-Aldrich, Vienna, Austria) was added, the incubation was prolonged for 10 min at boiling water, and the absorbance was recorded at 545 nm [46].

Exochitinase Assay
The activity of exochitinase was assayed by 200 µL of p-nitrophenyl-N-acetyl-β-Dglucosaminide (pNPg; Sigma-Aldrich, Co., Vienna, Austria) solution (1 mg pNPg per mL of distilled water) as substrate, 25 µL of enzyme solution and 500 µL of Tris-HCl (25 mM, pH 7) which was incubated at 40 • C for 20 h. Then, the mixture was centrifuged at 13,000 rpm at 4 • C, and the supernatant was added to 200 µL of sodium tetraborate-NaOH buffer (125 mM, pH 10) before to read the absorbance at 400 nm. The extinction coefficient of 18.5 Mm −1 −cm −1 was considered for activity calculation based on the following formula: where, V t = total volume; Vs = sample volume; 18.5 = millimolar extinction coefficient of p-nitrophenol under the assay condition; 1.0 = lightpath length (cm); t = reaction time; and df = dilution factor [46].

Protein Assay
The method of Lowry et al. [44] was used to determine the content of protein in the provided samples. Twenty microliters of the enzyme solution were added into 100 µL of reagent (Ziest Chem. Co., Tehran, Iran) and incubated for 30 min before reading the absorbance at 545 nm.

Effects of Thermotolerance and Cold Activity on Conidial Germination
To measure thermotolerance for conidial germination, 100 µL of conidial suspensions (5 × 10 6 conidia/mL) from each isolate was transferred to 1.5 mL tubes and kept in a thermal cycler adjusted to 45 • C. After 1 h and 2 h, 20 µL of conidial suspensions were removed and plated (without spreading) on PDA. Finally, plates were maintained at 25 • C, and conidial germination was counted after 24 h by microscopic observation. Moreover, 20 µL of a conidial suspension (5 × 106 conidia/mL) was plated (without spreading) on PDA and kept at 5 • C to determine the germination after 7 and 14 days in cold condition.
In both experiments, the conidial control suspensions were inoculated on PDA at 25 • C. The relative percent germination was estimated by comparing conidial germination to untreated isolates, and at least 100 conidia were counted for each treatment in every test [27].

Statistical Analysis
Probit analysis was done to determine LC 50 and LT 50 values at the corresponding 95% confidence interval (CI) values by using POLO-Plus software. Biochemical data and germination of conidia were compared by one-way analysis of variance (ANOVA) followed by Tukey's test. Differences among control and treatments were statistically considered at a probability of less than 5% and marked by different letters.

Screening and Identification of Fungi
A total of 38 fungal specimens were collected from C. suppressalis larvae, which were naturally infected by fungi in the rice fields of northern Iran. The specimens were morphologically identified as Akanthomyces lecanii (×5 isolates), Akanthomyces muscarius (×4 isolates), Aspergillus sp. (×1 isolate), Beauveria bassiana (×21 isolates), Hirutella subulata (×2 isolates), Metarhizium anisopliae complex (×3 isolates) and Trichoderma sp. (×2 isolates) (Table 1, Figure 1). Among all specimens, it was the first report of the natural occurrence of H. subulata in Iran. In A. lecanii, conidiogenous cells were phialidic, phialides approximately small, length size of 11-16 µm and width size of 1.4-2 µm, aculeate and strongly tapering, solitary or in whorls 3-6, conidial shape ellipsoidal-cylindrical, length size 4.1-5.2 µm and width size of 1.3-2.1 µm (Table 1, Figure 1a). In A. muscarious, conidiogenous cells were phialidic; phialides burned straight on prostrate hyphae or on secondary branches, phialides generally tall and slender and longer than A. lecanii, length size of 28-35 µm and width size of 1.6-2 µm, conidial shape cylindrical and longer than A. lecanii, length size 6.5-9.5 µm and width size of 1.5-1.9 µm (Table 1, Figure 1b). The major difference between the species of Akanthomyces is the shape and size of phialides that our specimens matched accurately match the description given by Zare and Gams [33]. In Aspergillus sp. conidial shape was globose, length size 1.8-2.3 µm and width size of 1.8-2.3 µm (Table 1, Figure 1c). In addition, the Conidial shape of B. bassiana isolates was globose with a length size of 2.1-3.2 µm and width size of 2-3 µm; conidiogenous cells were phialidic; the phialides were flask-shaped, swollen at the base or near the base and tapering at the apex. In addition, the conidiogenous cells were usually solitary or in a cluster of up to five (Table 1, Figure 1d), which appeared typical of those described by other researchers [28,34,35]. The major difference between the species of Beauveria is the shape and size of conidia [28]. In H. subulata, conidiogenous cells were phialidic, phialides scattered, and the lower phialides were narrow ellipsoid; the conidial shape was ovoid and in a chain, length size 5.5-6.9 µm and width size of 3.9-5.1 µm (Table 1, Figure 1e). Our isolates were compared with Yoon et al. [37], and significant differences were not observed among them. The conidial shape of M. anisopliae isolates was oblong oval with a length size of 6.8-7.8 µm and a width size of 2.6-3.7 µm (Table 1, Figure 1f). Conidia were the only morphological particular that reliably distinguishes several Metarhizium species [39]. In Trichoderma sp. conidial shape was globose, length size 2.2-3.1 µm and width size of 2.1-2.9 µm (Table 1, Figure 1g). Note: final tested isolates were renamed after the submission of the ITS sequence to the GenBank database. The ITS5-5.8S-ITS4 region was used for molecular analysis, confirming identifications of the fungal isolates. The amplified ITS region from all specimens showed approximately 600 bp-sized fragments, and the samples were sequenced and compared in the GenBank database. Results of the ITS sequence data were consistent with those obtained using morphological studies. After submission to the GenBank database, the fungal isolates were renamed based on the given code (Table 1). In the current study, B. bassiana was the most frequent detected fungus in rice collected The ITS5-5.8S-ITS4 region was used for molecular analysis, confirming identifications of the fungal isolates. The amplified ITS region from all specimens showed approximately 600 bp-sized fragments, and the samples were sequenced and compared in the GenBank database. Results of the ITS sequence data were consistent with those obtained using morphological studies. After submission to the GenBank database, the fungal isolates were renamed based on the given code (Table 1). In the current study, B. bassiana was the most frequent detected fungus in rice collected from fields similar to others studies that have shown B. bassiana as the most widespread entomopathogenic fungi in the endemic Moroccan forests of Argania spinosa, Switzerland, Spain, China and Southern Italy [22,[47][48][49][50]. Moreover, B. bassiana has been reported as the major pathogen of insects in more than 200 species that have been identified from the soil and dead insects in nature [24,25,51,52].

Bioassay
Only the isolates of B. bassiana, M. anisopliae and H. subulata caused mortality against the larvae of C. suppressalis. A comparison of LC 50 and LT 50 values indicated the significant differences among the isolates. The most virulent, isolate BBLN1 (1 × 10 4 conidia/mL), had the least LC 50 value, followed by BBAL1 (2.1 × 10 4 conidia/mL), BBRR1 (2.2 × 10 4 conidia/mL), BBLN2 (5.4 × 10 4 conidia/mL), BBAL3 (5.6 × 10 4 conidia/mL) and MASA (7.1 × 10 4 conidia/mL), while HSBL (1.6 × 10 6 conidia/mL), HSAL (7.9 × 10 5 conidia/mL) and BBLD5 (4.4 × 10 5 conidia/mL) showed the comparatively high LC 50 values (Table 2). Moreover, the least LT 50 values were obtained to be 2.71, 3.15, 3.45, 3.66 and 3.69 days for the larvae treated by BBRR1, BBLN1, BBAL1, BBAL4 and MASA, respectively (Table 3). These results revealed that BBRR1, BBLN1 and BBAL1 isolates of B. bassiana had higher efficacy than the other isolates on C. suppressalis larvae with a lesser concentration of conidia with a shorter time (days) to kill 50% of the larval population. Jandricic et al. [53] reported the higher virulence of B. bassiana isolates against the Myzus persicae Sulzer, Aphis gossypii Glover and Aulacorthum solani Kaltenbach (Hemiptera: Aphididae) compared to M. anisopliae isolates. Ramzi and Zibaee [12] showed that the two commercial isolates of B. bassiana and B. bassiana (BB1 and BB2) had the higher virulence against C. suppressalis larvae compared to A. lecanii, I. fumosoroseus and M. anisopliae. In addition, the higher virulence of the different isolates of B. bassiana and M. anisopliae has been observed on the boll weevil Anthonomus grandis Boheman (Coleoptera: Curculionidae) [29]. In our study, the least virulence of HSAL and HSBL as the two isolates of H. subulata were obtained compared to B. bassiana and M. anisopliae isolates, which may be correlated with low germination and sporulation rates in addition to the low activities of the extracellular enzymes of these isolates (see below) [54]. Finally, the isolates of A. lecanii and A. muscarious showed no mortality against C. suppressalis larvae. This case may be attributed to host-pathogen interaction between these isolates and the larvae of C. suppressalis, such as efficient attachment of conidia to the integument, negative impacts of integument composition with penetration tube of the fungi and immune responses of the larvae toward conidia. All these phenomena deserve detailed experiments to precisely elucidate the case.

Hydrophobin
The highest amounts of hydrophobin recorded in BBAL1, BBLD5, BBLD1, BBSI, BBBL1, BBLN1, BBLN2, HSAL and MASA, respectively (Table 4), while the least amounts of hydrophobin were in TSRT, ASAI and TSAH (Table 4). Entomopathogenic fungi achieve the nutrients at host bodies through the cuticle, so the first step of pathogenesis is adhesion to the integument. Therefore, the external surface of conidia has a fundamental protein with a hydrophobic rodlet layer that connects to the insect epicuticle [6]. Hydrophobins are a class of unique fungal proteins important in sporulation, pathogenesis, thermotolerant, growth and development of fungi [43,55,56]. Some studies reported that inhibition of hydrophobin gene expression negatively affected pigmentation, conidiation, hydrophobicity and virulence of entomopathogenic fungi [24,55,57]. Our findings revealed that the lesser amount of hydrophobin could be one of the reasons for no mortality of some isolates against the larvae of C. suppressalis. In fact, the proper attachment of the conidia to the host cuticle and subsequent germinations are the primarily important steps to effective infection by entomopathogenic fungi. The higher amounts of hydrophobin were obtained in the isolates with the more virulence-like BBLN1 and BBLN2. Although the higher amounts of hydrophobin were also found in HSAL with the least virulence, it should be noted that hydrophobin is not necessarily the main factor in the virulence of a fungus, but it only shows the better interaction with the host cuticle.

Extracellular Enzymes
The conidia of entomopathogenic fungi attach to the cuticle of host insects, germinate and penetrate to the hemocoel with the assistance of extracellular enzymes, such as chitinases, proteases and lipases [58]. Trypsin (Pr1) and subtilisin-like (Pr2) proteases are the primitive synthesized enzymes to simplify penetration of the hyphae into the host body. Then, synthesis of the chitinases increases the penetration efficiency [59], and finally, lipases involved in hydrolyzing lipid derivatives within the cuticle and facilitating the infection of host cells [60]. Our results revealed differences in the activities of extracellular enzymes between the fungal isolates. Isolates BBAL1, BBRR1, BBLN2, BBLN1 and BBLD2 demonstrated the highest activity of Pr1 while the least activity was observed in ASAI, TSRT and TSAH isolates (Figure 2). In the case of Pr2, BBRR1, BBLN2, BBLD2 and BBLN1, isolates showed the highest activity ( Figure 2). The highest activity of lipase was recorded in BBLD4, BBSI, BBLN2 and BBRR1 isolates (Figure 3). BBRR1, BBLN2, BBAL1 and BBLN1 isolates showed the highest activity of exochitinase (Figure 4). In the case of endochitinase, the highest activity was obtained in BBRR1, BBLN2, BBBL1, BBAL1 and BBLN1 isolates (Figure 4). The higher Pr1 activity in the given isolates indicates the capability of protein digestion by these isolates in the initial stages of infection, so the efficiency of this enzyme may ensure the success of other enzymes to feasible penetration through insect cuticle. Charnley and St. Leger [61] believe in facilitating the cuticle infiltration by the proteases produced during invasion prior to chitinases during later steps. They concluded the major role of proteases in cuticle penetration compared to chitinases. Ramzi and Zibaee [12] demonstrated the different levels of proteinases, chitinase and lipase produced by B. bassiana, M. anisopliae, L. lecanii and I. fumosoroseus in the larvae C. suppressalis in which the isolates with the highest enzymatic activity led to the higher mortality Lu et al. [1] showed, the higher levels of protease and chitinase produced by ZJLSP09 isolate of Lecanicilium sp. in comparison with ZJLA07 and ZJLP08 isolates which were related to mortality in Diaphorina citri Kuwayana (Hemiptera: Psyllidae). Maqsoudi et al. [62] reported that the isolate of B. bassiana with the higher activity of proteases and chitinases led to the lower LC50 and LT50 values against Pseudococcus viburni Signoret (Hemiptera: Pseudococcidae). In our study, no clear correlation was obtained between lipase production and virulence of isolates, similar to earlier studies [12,23,63,64]. This conclusion on lipase may be more obvious in the case of BBLN2, which is the only isolate with higher virulence and lipase activity. Other isolates with higher virulence showed lower lipase activity. It seems that lipases are more important in the utilization of integument lipids for fungal development, not necessarily penetration. In contrast, the isolates with the higher virulence demonstrated the higher activity of proteases and chitinases, mainly BBLN1, BBLN2 and BBRR1. These findings apparently disclosure the correlation between efficiency of extracellular enzymes and higher virulence of the entomopathogenic fungi. Such isolates properly or rapidly penetrate through host cuticle with efficient cleavage of polypeptide and carbohydrate bonds then achieve hemocoel to continue the latter steps of infection. It should be mentioned that this process is accompanied by better production of blastospores and secondary metabolites within the host hemocoel to impose virulence on infected individuals. penetration through insect cuticle. Charnley and St. Leger [61] believe in facilitating the cuticle infiltration by the proteases produced during invasion prior to chitinases during later steps. They concluded the major role of proteases in cuticle penetration compared to chitinases. Ramzi and Zibaee [12] demonstrated the different levels of proteinases, chitinase and lipase produced by B. bassiana, M. anisopliae, L. lecanii and I. fumosoroseus in the larvae C. suppressalis in which the isolates with the highest enzymatic activity led to the higher mortality Lu et al. [1] showed, the higher levels of protease and chitinase produced by ZJLSP09 isolate of Lecanicilium sp. in comparison with ZJLA07 and ZJLP08 isolates which were related to mortality in Diaphorina citri Kuwayana (Hemiptera: Psyllidae). Maqsoudi et al. [62] reported that the isolate of B. bassiana with the higher activity of proteases and chitinases led to the lower LC50 and LT50 values against Pseudococcus viburni Signoret (Hemiptera: Pseudococcidae). In our study, no clear correlation was obtained between lipase production and virulence of isolates, similar to earlier studies [12,23,63,64]. This conclusion on lipase may be more obvious in the case of BBLN2, which is the only isolate with higher virulence and lipase activity. Other isolates with higher virulence showed lower lipase activity. It seems that lipases are more important in the utilization of integument lipids for fungal development, not necessarily penetration. In contrast, the isolates with the higher virulence demonstrated the higher activity of proteases and chitinases, mainly BBLN1, BBLN2 and BBRR1. These findings apparently disclosure the correlation between efficiency of extracellular enzymes and higher virulence of the entomopathogenic fungi. Such isolates properly or rapidly penetrate through host cuticle with efficient cleavage of polypeptide and carbohydrate bonds then achieve hemocoel to continue the latter steps of infection. It should be mentioned that this process is accompanied by better production of blastospores and secondary metabolites within the host hemocoel to impose virulence on infected individuals.           ALRT  ALLN  AMRT  AMAI  AMAL  AMBL  ASAI  BBAL1  BBAL2  BBAL3  BBAL4  BBBL1  BBBL2  BBLN1  BBLN2  BBLN3  BBLD1  BBLD2  BBLD3  BBLD4  BBLD5  BBRT1  BBRT2  BBRR1  BBRR2  BBSI  BBFN  BBAI  HSAL  HSBL  MASA  MAAI  MAAL  TSRT

Effects of Thermotolerance and Cold Activity on Conidial Germination
The inactivation and delay of conidia germination caused by heat and cold as the most important environmental factors reduce the efficiency of the entomopathogenic fungi as the biocontrol agents from both the virulence against host and persistence in ecosystems. Selection of the entomopathogenic fungi that tolerate thermal fluctuations te is necessary before field application [21,27,65]. The effect of thermotolerance on the germination rate of the conidia of the collected isolates in the current study have shown in Table 5. Fifteen isolates demonstrated a germination rate of more than 50% after 1 h exposure to 45 • C, while only one isolate exhibited a high tolerance after 2 h (Table 5). After 2 h of exposure, the isolate thermotolerance could be divided into three classes: low (below 30%), moderate (between 30% and 60%) and high tolerance (above 60%). Among the isolates, Aspergillus sp. (ASAI) showed high tolerance. Moderate tolerances were observed in four isolates of B. bassiana (BBLN1, BBAL1, BBLN2 and BBLD2), three isolates of M. anisopliae (MASA, MAAI and MAAL) and two isolates of Trichoderma sp. (TSAH and TSRT). The other isolates were low tolerance to the heat of 45 • C. Similar results have been reported by Lee et al. [27] as the rate of the conidial germination in B. bassiana, M. anisopliae and Lecanicillium attenuatum significantly reduced after 2 h at 45 • C. Rivas et al. [66] demonstrated the significant lower conidial germination of Lecanicillium isolates after incubation at 32 • C. The susceptibility of Metarhizium isolates to high temperatures (45 • C) was demonstrated by Rangel et al. [65]. Exposure to 35 • C for 10 min harmed the conidial germination of B. bassiana, but the M. anisopliae isolate germinated readily at this temperature [67]. Generally, the optimal temperature for conidia germination and growth of entomopathogenic fungi is between 23 and 28 • C. The growth was reduced above 30 • C, and it was totally inhibited above 34 • C [20,27,[65][66][67]. Our results imply only Aspergillus sp. isolates as the highly thermotolerant isolate has although it had no virulence against the larvae of C. suppressalis. Finally, the cold activity of fungal isolates was examined through the treatment of the conidia at 5 • C for one and two weeks. All the isolates showed high activity (above 80%) at 5 • C for both time intervals except for Aspergillus sp. (ASAI) ( Table 5). Lee et al. [27] reported the high cold activity at 7-14 days for almost all collected isolates. Based on earlier reports, most entomopathogenic fungi have high cold activity, although germination and sporulation may be delayed or stopped at a cold temperature [21,27,68,69]. Such a property may be important in the survival of entomopathogenic fungi in cold periods of the year.