Virulence and proteolytic activity of entomopathogenic fungi against the two-spotted spider mite, Tetranychus urticae Koch (Acari: Tetranychidae)

The virulence and proteolytic activity of some entomopathogenic fungi isolates, viz., Metarhizium anisopliae, Beauveria bassiana, Verticillium lecanii, and Trichoderma harzianum, against the two-spotted spider mite, Tetranychus urticae Koch (Acari: Tetranychidae), were evaluated. Common maize plants (Zea mays L.) infested with females of T. urticae were treated in vivo by spraying with suspensions of 1 × 108 conidia ml−1 concentration of selected isolates. Lethal effects of fungal isolates were assessed as percentages of daily mortalities of mites, compared to the mortality in control. Virulence of the fungi isolates was estimated based on the LC50 values calculated by probit analysis for the individuals treated by 1 × 105 conidia ml−1 concentration. Proteolytic activity of isolates was assayed on casein substrate to reflect their virulence towards T. urticae. The mite mortality rates increased with increasing conidial concentrations as well as days after treatment. The mortality rate caused by M. anisopliae isolate varied from 18.75 to 85%, with LC50 value of 4.6 × 105 conidia/ml and LC90 value of 2.4 × 108 conidia/ml during 7 days, respectively. The isolate of B. bassiana caused 15 to 70% mortality, and its LC50 and LC90 values estimated 3.3 × 106 and 7.8 × 109 conidia/ml, respectively. However, V. lecanii isolate caused 11.25 to 72.50% mortality with LC50 of 5.2 × 106 conidia/ml, while T. harzianum was potentially less virulent than other isolates causing 8.75 to 63.75% mortality rate to T. urticae with LC50 of 9.4 × 106 conidia/ml. M. anisopliae showed the highest proteolytic activity at all concentrations, followed by B. bassiana in 3rd, 5th, and 7th day post treatment. These findings recommend the selection of virulent fungal isolates for use as natural and environmentally safe agents in biological control programs to combat mite pests.

1. The two-spotted spider mites T. urticae were sensitive and affected by entomopathogenic fungi. 2. M. anisopliae isolate had the most lethal effect, but the isolates V. lecanii, B. bassiana and T. harzianum were less lethal.
3. M. anisopliae was the highest virulent isolate against T. urticae than the other isolates with LC 50 4.6 × 10 5 ppm. 4. Protease activities of fungal isolates were increased with increased conidia concentrations from 1 × 10 5 and up to 1 × 10 8 conidia ml −1 . 5. M. anisopliae gave the highest activities at all concentrations, compared with other isolates, followed by B. bassiana which exceeded V. lecanii and T. harzianum in its proteolytic activities.

Background
Spider mites (Acari: Prostigmata: Tetranychidae) are obligated predatory plant mites, which are unique in possessing highly elongated movable stylets used to puncture the individual cells of their host plants (Walter et al., 2009). Pests of known economic significance belonging to this group are the two-spotted mites, Tetranychus urticae Koch, which are considered as violent pest that affects vegetables, orchards, medicinal, and ornamental plants (Naher et al., 2006). The conventional control of this pest includes acaricides that have been widely used for mite control in greenhouses, orchards, and many other cropping systems (Van Leeuwen et al., 2005), but it can cause undesirable problems like the death of natural enemies such as predators or development of pesticide-resistant races of mites and residue concerns (Draganova and Simova, 2010). Thus, the search for an eligible biocontrol agent is therefore a step in developing new environmentally friendly strategies or improving existing strategies that provide alternatives to conventional pest control. Research and biological control strategies for spider mites have largely focused on preserving natural enemies and releasing predator mites (Zhang, 2003), and nevertheless, an additional spray of acaricides is needed. The entomopathogenic fungi (EPF), developed as mycoacaricides, can naturally manage mite populations and can be used in a useful control strategy as standalone to replace synthetic acaricides already used, or as an integrated component for mite management (Maniania et al., 2008). The fungus penetrates the pest cuticle with the help of hyphae protrusion germinated from the conidia, as soon as contact to the host (Revathi et al., 2011). The EPF rely on their enzymatic capacity to degrade the integument of insects or mites, consisting of lipases, proteases, and chitinases, furthermore, glutaminase, β-galactosidase, and catalase as well (Mondal et al., 2016). Fungal proteases seem to be particularly important in the penetrating process since the insect and mite cuticle comprises up to 70% protein (Charnley, 2003). The virulence of the EPF against their hosts can be verified through their production of cuticle-degrading enzymes (Pinto et al., 2002). The successful use of EPF as agents for biological control of mites ultimately depends upon the extent of selected fungal strains (Maniania et al., 2008).
The goal of this study was to determine the activity of protease and the virulence of 4 EPF isolates against the two-spotted spider mite, T. urticae.

Material and methods
Isolation of fungi, culture conditions, purification, and identification The fungi, Metarhizium anisopliae, Beauveria bassiana, Verticillium lecanii, and Trichoderma harzianum were collected from dead cadavers of the two-spotted spider mite individuals on leaf surface of maize plants (Zea mays L.). Samples containing fungi, wrapped in bags, were transferred to the Biological and Integrated Pest Control Laboratory, Plant Protection Institute, Zagazig, Egypt. Serial dilution method was used to isolate EPF according to the method of Goettel (1996). One ml of each soil dilution was poured onto Petri dishes containing Sabouraud's dextrose agar (SDAY) medium (Merck, Germany) containing 1% yeast extract. To isolate the fungi grown on the surface of the mite body, the mites were superficially sterilized with 70% ethanol for 1 min and 5% sodium hypochlorite for 3 min and rinsed three times with sterile distilled water; then, the fungi were transferred to SDAY medium. The dishes were incubated at 25°C for 15 days until the colonies appeared. Then, the colonies were selected, picked up, and maintained on PDA (potato dextrose agar) medium until identification. Fungal species were identified using the lacto-phenol cotton blue staining method according to Humber (2005). Single spore fungus was carried out according to Ho and Ko (1997). For preparation of the bioassay suspensions, the conidia of 10-day-old cultures, which were previously developed on SDAY medium, were harvested by washing with sterile distilled water containing 0.02% Tween 80, followed by decimal dilutions of the conidial suspensions obtained and their concentrations estimated using a hemocytometer under light microscope (Zeiss) (× 40 magnification).

Mite culture
Females of T. urticae were obtained from the Biological and Integrated Pest Control Laboratory, Plant Protection Institute, Zagazig, Egypt, and transferred to caged maize plants for subsequent culturing. A stock colony was established in ventilated jars (26 × 20 × 10 cm) at 32 ± 2°C, 55% R.H., and 16:8 (L:D) photoperiod. To avoid the excessive accumulation of excrements and exuviae of larvae, the growing media were updated three times weekly.

Protease assay
Casein substrate was used to determine protease enzyme activity (Vyas and Deshpande, 1989). A mixture of 2 ml reaction contained an aliquot of an appropriately diluted enzyme solution, 10 mg substrate (casein) in 0.2 mM sodium carbonate buffer, pH 9.7. The enzyme reaction mixture was incubated at 35°C for 20 min and then 3 ml trichloroacetic acid (TCA) was added (2.6 ml 5% TCA + 0.4 ml 3.3 N HCl). The mixture was centrifuged at 8000 rpm for 5 min, and the resultant was observed at 280 nm (Ultrospec II, LKB Biochrom, UK). One international unit (IU) is defined as an enzyme activity that produced 1 μmole of tyrosine per min.

Virulence test
Bioassays were conducted under laboratory conditions (23 ± 1°C, photoperiod 12:12 L:D). Females of mites were taken from the cultured source and gently transferred onto caged maize plants with the help of a fine camel hair brush. T. urticae was used as a host for the EPF reared on maize plants in earthen pots. Population of mites was counted on individual plants (having 4-5 leaves) before performing sequential sprays of fungi. The plants were divided into 4 groups (4 isolates), each of which containing 5 plants (4 treatments + control). Five plants in each treated group were sprayed with a selective fungus with recommended concentration (1 × 10 5 , 1 × 10 6 , 1 × 10 7 , and 1 × 10 8 conidia ml −1 ). These concentrations were applied by foliar application with hand atomizers on potted plants. In all cases, 300 μl of each conidial concentration, with Tween 80 solution of 0.02%, were directly sprayed on 10 adult female mites. Control plants received same volume of water containing 0.02% Tween 80. The whole experiment was repeated 3 times. Mite mortality was estimated 3, 5, and 7 days after treatment. Plant leaves carrying dead mites were incubated in a humid chamber to allow the germination of fungus conidia and confirm that the death was due to mycoses. Smears were prepared from dead mites stained with methylene blue, then observed under a light microscope. Fungal lethal effects were assessed as percentages of daily cumulative mortality due to mycoses, compared to mortality in untreated control. Virulence of fungal isolates was calculated by probit analysis (Finney, 1971) depending on the median lethal concentration (LC 50 ) of the individuals treated with 10 5 conidia ml −1 suspensions. Confidence intervals of varying LC 50 values were calculated at P level < 0.05.

Statistical analysis
Data were corrected for control mortality using Abbott's formula (Abbott, 1925). The mortality of each treatment was analyzed by analysis of variance (ANOVA) and means compared by Tukey's HSD test (SPSS 16.0). The LC 50 and LC 90 and their 95% fiducial limits were calculated from probit analysis (Finney, 1971) using the Minitab 15 computer software.

Results and discussion
Virulence of entomopathogenic fungi against T. urticae The bioassay revealed that M. anisopliae, B. bassiana, V. lecanii, and T. harzianum can be biologically employed to control T. urticae. Fungal lethal effects were assessed as percentage of daily cumulative mortality due to mycoses, compared to mortality in control. In general, the mortality of T. urticae was found to be the highest with increasing conidial suspensions from 1 × 10 5 and up to 1 × 10 8 conidia ml −1 (Figs. 1, 2, 3, and 4). Also, data showed that the mortality rate of T. urticae females increased significantly after 7 days more than those after 3 days and/or 5 days of spraying at all concentrations. This trend of deaths over the days after inoculation pointed that the advanced infections and deaths appeared from the third day and continued until the seventh. These results suggested that the experiments with some EPF require a long incubation time before evaluating their results (Amjad et al., 2012). Tamai et al. (2002) reported that all isolates of M. anisopliae and B. bassiana were pathogenic to T. urticae with increased mortality rates after the 3rd day and peaking on the 4th or 5th day post treatment. Thus, the bioassay of some EPF against mites requires a long period of incubation before evaluating the results. This delay may be regarded to the time needed for infectivity process of adhesion, penetration, germination, and fungus growth (Ekesi, 2001).
The isolates V. lecanii and B. bassiana were found highly virulent to T. urticae individuals with mortality rates 72.50 and 70.00, at 1 × 10 8 conidial ml −1 concentration, respectively. This finding was consistent with Wekesa et al. (2005) who tested 2 isolates of B. bassiana and 17 isolates of M. anisopliae against T. evansi and observed 80% mortality rate in mite population at a high concentration (1 × 10 8 conidia ml −1 ) caused by the 2 isolates. Also, Shi and Feng (2009) used B. bassiana, Paecilomyces fumosoroseus, and M. anisopliae isolates 10 days after spraying where they caused 73.1, 75.4, and 67.9% mortality rates of mites, respectively. On the reverse, M. anisopliae isolate remained least effective against females of T. urticae (Amjad et al., 2012), while V. lecanii and P. fumosoroseus showed more than 75% mortality (76.25 and 79.16%, respectively) at the highest lethal concentration 1 × 10 8 conidia/ml, 8 days after fungal application. On the other hand, T. harzianum was less lethal where it showed potency reaching 63.75% of deaths, 7 days after inoculation. Regarding the virulence of the 4 isolates, there were non-significant differences at P level < 0.05.

Pathogenicity test
All isolates of M. anisopliae, B. bassiana, V. lecanii, and T. harzianum infected mite individuals causing mortality ( Fig. 5a-d). They demonstrated rapid external hyphal development and sporulation under moist condition. Firstly, filamentous hyphae were evolved from the anal region of the mite cadaver and then quickly covered the cadaver, followed by sporulation within 7 days. Investigation of smears of cadaver of mites confirmed mycosis infectivity as cause of death in individuals treated with conidial suspensions of the fungal isolates. Considering the deadly effect of the tested fungi to T. urticae was due to the secondary metabolites toxicity and the dyes they produced (Tamai et al., 2002), a variety of degrading enzymes were produced by EPF worked in a coordinated manner, popularly known as cuticle degrading enzymes, including chitinases, proteases, and lipases, making their entry through the massive barriers of mite cuticle easier (Krieger de Moraes et al., 2003;Zare et al., 2014).

Protease activity
Proteases are considered to be the most important in the infectious process (Mustafa and Kaur, 2009). The dissolved proteins were further degraded into amino acids by exopeptidases and amino peptidases in order to provide nutrients for entomopathogenic fungi (Wang et al., 2002). The present results showed that protease activities of fungal isolates were increased by conidia concentrations from 1 × 10 5 and up to 1 × 10 8 conidia ml −1 . M. anisopliae gave the highest activities at all concentrations than other isolates, yielding 1.032, 1.361, 1.977, and 2.111 U/ml at 1 × 10 5 , 1 × 10 6 , 1 × 10 7 , and 1 × 10 8 conidia ml −1 concentrations, respectively (Fig. 6). This is followed by B. bassiana, which exceeded V. lecanii and T. harzianum in its proteolytic activities at 1 × 10 5 , 1 × 10 6 , and 1 × 10 7 conidia ml −1 concentrations yielding 0.897, 0.964, and 1.121 U/ml, respectively. V. lecanii exceeded B. bassiana at the highest concentration (1 × 10 8 ) yielded 1.356 U/ml enzyme activity. T. harzianum was higher than V. lecanii at low concentrations (1 × 10 5 and 1 × 10 6 conidia ml −1 ) while lower at higher concentrations (1 × 10 7 and 1 × 10 8 conidia ml −1 ). In this respect, invasive fungi produce large quantities of serine-protease (Pr1), which degrades protein substances, after the epicuticle breakdown by lipases. The subtilisin-like serine-protease (Pr1) and trypsin-like protease (Pr2) are the most common studied protein decomposition enzymes. Pr1 and Pr2 activities have been detected in M. anisopliae, M. flavoviride, B. bassiana, Nomuraea rileyi, and Lecanicillium lecanni (Liu et al., 2007). Furthermore, the finding of protease Pr1 in EPF was considered as an indicator of virulence (Castellanos-Moguel et al., 2007). Recent evidence indicated that different EPF isolates may show some degrees of variety in cuticle degrading proteases production (Pinto et al., 2002;Revathi et al., 2011 andZare et al., 2014). These varieties may reflect differences observed in the virulence of isolates and species such that, isolates with high protease activity are expected to represent a high virulence towards their host. These findings were in accordance with obtained findings, which revealed a wide variation in the activities of proteases among the studied isolates. A positive correlation between proteases activity and virulence of isolates was found, so that the highest the protease activity of a particular isolate, the highest the virulence of that isolate against T. urticae. The present findings were also coincided with those obtained by Zare et al. (2014) who showed a wide diversity in proteolytic activity and virulence of studied isolates. Additionally, they found a positive correlation between the proteolytic activity on casein substrate and virulence of the isolates against the Khapra beetle Trogoderma granarium.
The present results revealed that the 2 virulent isolates M. anisopliae and B. bassiana showed the highest protease activities. It is well known that both the tested EPF produce higher enzymes of chitinases, proteases, and lipases than other fungi species (Revathi et al., 2011). Besides, M. anisopliae, B. bassiana, and V. lecanii secreted a diversity of extracellular hydrolytic enzymes in liquid cultures that contain locust cuticle as sole carbon source (St Leger et al., 1986).

Conclusion
Degrading enzymes might regulate many infectious processes and function as control agents. The finding has given biotechnology researchers an opportunity to implement bio-process engineering to produce these enzymes on a large scale, along with strategies to create applications for the enzymatic extracts, as well as potential applications such as mycoacaricides in a green technology-based environment.