Synergy in Efficacy of Artemisia sieversiana Crude Extract and Metarhizium anisopliae on Resistant Oedaleus asiaticus

In order to explore the synergistic control effect of crude extracts of Artemisia sieversiana and Metarhizium anisopliae on Oedaleus asiaticus, we used different doses of M. anisopliae and crude extracts of A. sieversiana singly and in combination, to determine their toxicities to fourth instar O. asiaticus. The results showed that the combination of 10% crude extract of A. sieversiana with 107 and 108 spores/g M. anisopliae concentrations and the combination of 20% crude extract of A. sieversiana with 107 and 108 spores/g M. anisopliae concentrations had significant effects on the mortality, body weight gain, body length gain, growth rate, and overall performance of O. asiaticus than those of the crude extract of A. sieversiana and M. anisopliae alone. Among them, the 20% A. sieversiana crude extract mixed with 108 spores/g M. anisopliae and 10% A. sieversiana crude extract combined with 107 spores/g M. anisopliae, had the best control efficacy. In order to clarify the biochemical mechanism underlying the immune responses of O. asiaticus to the pesticide treatments, we monitored the activities of four enzymes: superoxidase dismutase (SOD), peroxidase (POD), catalase (CAT), and polyphenol oxidase (PPO). The results showed that the activities of three enzymes (SOD, CAT, and PPO) were significantly increased from the treatment with the combination of M. anisopliae mixed with crude extract of A. sieversiana. Interestingly, compared to the crude extract, the combination treatment did not significantly induce the expression of POD enzyme activity, which may be a biochemical factor for increasing the control effect of the combination treatment. Our results showed that the combination treatment had synergistic and antagonistic effects on host mortality, growth, development, and enzyme activities in O. asiaticus.


INTRODUCTION
As important biological control agents, botanical pesticides have received extensive research attention due to characteristics such as their low toxicities, easy degradation, and miscibility with other pesticides and against which insects are yet to develop resistance (Dubey et al., 2010). Crude plant extracts, the most notable of botanical pesticides, contain secondary metabolites such as alkaloids, terpenoids, phenols (mainly including flavonoids), tannins, lignin, nitrogenous organic compounds, etc., which have shown to be effective against insects (Wink, 1988). They have been extracted from plants species such as Melilotus suavcolen, Peganum harmala (zygophylllaceae), Ajuga iva (labiateae), Aristolochia baetica (Aristolochiaceae), Raphanus raphanistrum (Brassicaceae), etc. (Jbilou et al., 2006). These secondary metabolites are chemical compounds which have evolved in plants, to have specialized functions such as adaptation, survival of competition, defense against herbivorous insects, etc. They deter herbivorous insects from feeding on plants by poisoning them or interfering with digestion and the absorption of nutrients in herbivorous insects (Rattan, 2010). For instance, flavonoids can inhibit many insect species from feeding and oviposition (Simmonds, 2001). In addition, these substances also induce a rise in the detoxification mechanism of foreign compounds in insects (Bi and Felton, 1995). They can also stimulate and induce insects to feed or to produce sex hormones. Thus, they have mutually beneficial and co-evolutionary relationships (Kergoat et al., 2017).
China is rich in plant resources. There are more than 220 plant species that can be used as botanical insecticides. Compositae, one of the plant families which accounts for more than 20% of these species, is characterized by the most complex and diverse chemical compositions (Shang et al., 1997). Artemisia, a dominant genus in grasslands in Inner Mongolia, has shown promising insecticidal activities. In recent years, it has increasingly become one of the common plants, with the intensification of degradation of the grasslands in Inner Mongolia. The main chemical components of Artemisia are flavonoids, terpenoids, sesquiterpene lactones, cyanogenic glycides, flavonal, coumarins, etc. (Vander et al., 2008). Extensive studies have been conducted on the insecticidal activity of Artemisia plants (da Silva et al., 2005). For example, the crude extract of Anthemideae was shown to have strong anti-feedant properties against Tribolium castaneum Herbst, Tetranychus cinnabarinus, and Maruca testulalis. Also, Artemisia mongolica, Artemisia lavandulaefolia, and Acacia vestita showed excellent insecticidal effects against Sitophilus zeamais (Wu, 2015).
Artemisia sieversiana, a typical Artemisia plant, also known as Artemisia, A. alba, A. scoparia, etc., is one of the dominant species in grasslands in China. It is widely distributed in India, Pakistan, Afghanistan, Russia, Mongolia, and China's Heilongjiang, Shanxi, Hebei, Ningxia, Gansu, Xinjiang, Inner Mongolia, and other regions (Cui et al., 2020). A. sieversiana contains a variety of insecticidal and antibacterial chemical components, such as flavonoids, lignins, sesquiterpenes, and volatile oils (Liu et al., 2010;Cui et al., 2020). In addition, its essential oil was shown to have fumigation and killing activity on Callosobruchus chinensis, grubs, and other insects (Ayub et al., 2019). Further, the ethanol extract of A. sieversiana was shown to have significantly inhibited and killed the growth of human colon cancer cells (HT-29, HCT-15, and Colo-205;Tang et al., 2015).
Entomopathogenic fungi are another group of important biological control agents with low environmental risk, used in pest management. Insect pests are not easily resistant to entomopathogenic fungi. They have great potential for pest control in integrated pest management programs (Blanford et al., 2005;Sabbour and Sahab, 2005;Zimmermann, 2007). Metarhizium anisopliae, one of such entomopathogenic fungal species, is widely used globally, and has been studied for more than 100 years in China. It has been used on more than 200 insect species and has shown significant control effects against the larvae of Curculio chinensis Chevrolat, Monochamus alternatus, Polyphylla laticollis Lewis, Locusta migratoria manilensis, Plutella xylostella, O. asiaticus, etc. (Peng et al., 2008;Kaaya et al., 2011).
M. anisopliae infects insects through the insect cuticle or when insects feed (stomach poison; Mannino et al. 2019). M. anisopliae directly enters the midgut through the feeding of insects, and the mycelium propagates rapidly in the midgut cells, resulting in the loss of cellular functions, directly affecting cellular metabolism, and ultimately destroying the midgut tissue. These result in the death of insects. Mortality rates of insects from midgut infections are significantly higher than that from cuticle infections. M. anisopliae infections are very easy to spread in their colonies. However, the long lethal time of fungal insecticides lowers their efficacy (MeCoy, 1981;Dimbi et al., 2009).
All groups of insecticides, including botanical insecticides, fungicides, and chemical insecticides, etc., have advantages and disadvantages (Marican and Durán-Lara, 2018). For example, botanical insecticides are environment friendly, but also have unstable formulations, easy photodegradation and their active ingredients are difficult to extract (Pavela, 2016). Insects do not easily develop resistance against fungal insecticides, but their lethal times are longer, and their prevention and control of pest outbreaks is slower. Chemical insecticides show rapid effectiveness and good control effects, but their excessive usage generates residues which pollute the environment and also have negative impact on food safety for humans and animals (Sabarwal et al., 2018). However, the use of a combination of these insecticides can effectively solve these challenges by producing a synergistic efficacy to achieve better control of insect pest. It can also solve the challenge of excessive use of chemical agents (Gibson et al., 2014). For example, the use of a combination of M. anisopliae V275 and imidacloprid or fipronil had a significantly higher control effect on Otiorhynchus sulcatus than that from each insecticide alone (Shah et al., 2007). Also, neem oil effectively maintained the toxicity of the entomopathogenic fungus M. anisopliae, protected it from UV radiation, and increased the control effect on Aedes aegypti larvae (Paula et al., 2019). The use of a combination of the insecticide chlorantraniliprole and M. anisopliae showed excellent Frontiers in Physiology | www.frontiersin.org synergistic control on locusts, scarabaeidea, Empoasca pirisuga Matumura, rice planthoppers, and other agricultural pest (Farenhorst et al., 2010;Jia et al., 2016).
Oedaleus decorus asiaticus (Bey-Bienko), is a major pest in the grasslands of northern China. In recent years, its frequent large-scale outbreaks are partly responsible for the degradation and desertification of grasslands in Inner Mongolia, in China (Zhang et al., 2013). At present, the main control of O. asiaticus is based on biological control methods such as the use of poxvirus, M. anisopliae, Beauveria bassiana, and the introduction of insect natural enemies. Among them, M. anisopliae has shown a very significant control effect on O. asiaticus. It has the advantage of being green, safe, and has long duration, but has a slow effect on insects. Quercetin (a kind of flavonoids) was shown to reduce the survival rate, inhibit the growth and development of O. asiaticus (Cui et al., 2019). The crude extract of A. sieversiana contains flavonoids, which has been shown to have a toxic effect on insects, but its efficacy lasted for a short period of time. For practical applications, the simultaneous use of fungi and chemical agents for insect pest control, may result in higher mortality than from the use of each alone (Duarte et al., 2016;Jia et al., 2016). Therefore, we evaluated the synergetic control effect of the crude extract of A. sieversiana and M. anisopliae on O. asiaticus in this study, to provide a new direction for the research and development of new biological agents for the effective control of locust or grasshopper pest.

Test Insects
O. asiaticus were caught from the wild, in the grasslands of Xilinhot City, Inner Mongolia, China. They were reared on their host plant, Stipa krylovii, in a cage (1 m × 1 m × 1 m) at room temperature (about 27°C) for 1 week.

Test Strain
M. anisopliae IMI330-189 was provided by the Grassland Pest Group, Institute of Plant Protection, Chinese Academy of Agricultural Sciences (IPPCAAS), which was introduced from the International Centre for Biological Control, United Kingdom, and preserved in the IPPCAAS.

Preparation of Crude Plant Extract
The crude plant extract was extracted with water and two concentrations (10 and 20%) of the extracts were prepared ( Table 1). Briefly, the fresh plant materials of A. sieversiana were cut into small pieces of about 1 cm and put into a conical flask. Distilled water was added and then autoclaved at 120°C for 90 min. They were taken out after heating and cooled to room temperature to filter plant residues.

Separate Treatments With A. sieversiana Extract or M. anisopliae
Healthy 4th-instar O. asiaticus were selected from the insect cage in the laboratory, and raised in a 26 × 13 × 7 cm 3 breeding frame with a glass cover (30 × 50 × 0.3 cm). Each cage contained 10 females and 10 males (20 heads in each frame). One frame served as a replicate, and five repetitions were treated as a group. In the A. sieversiana extract treatment, the 10 and 20% A. sieversiana extracts were spread separately on leaves of S. krylovii with a brush and fed to O. asiaticus in the cages. Leaves on which distilled water was spread, served as the control. The experiment lasted for 7 days and data on mortality were recorded.
In the M. anisopliae treatment, 1 g each of the three concentrations of M. anisopliae (10 6 , 10 7 , and 10 8 spores/g) was weighed and separately mixed with a carrier (10 g vegetable oil and 100 g wheat bran) as poison bait. The mixture was divided into five petri dishes (dm = 7 cm) and placed in the insect rearing frame containing 20 healthy 4th instar O. asiaticus which had been prior fed with fresh S. kirnii for 24 h. The experiment lasted for 7 days and data on mortality were recorded.

Treatment With the Combination of A. sieversiana Extract + M. anisopliae
In this treatment, mixtures which were made up of each concentration of A. sieversiana extract and each concentration of M. anisopliae were prepared ( Table 2). These combinations were then added to a mixture made up of 10 g vegetable oil

Concentration (%)
Amount of distilled water (g) Amount of A. sieversiana (g) 10 500 50 g 20 500 100 g mixed into 100 g wheat bran. They were mixed evenly, and then dispensed into glass culture dishes to feed O. asiaticus in the cages. After feeding for 24 h, the diet was changed to fresh S. kirnii for continuous feeding for 7 days. Data on mortality were recorded during this period.

Determination of Body Weight and Body Length of O. asiaticus
Female and male 4th instar O. asiaticus caught from the Xilinguole grassland were used to determine the effects of the treatments on body weight. We randomly selected 30 healthy 4th females and 4th males, respectively, from the captured 4th instar O. asiaticus to detect their initial body weight and body length. We use a vernier caliper to measure these selected females and males, and then put them in the oven at 90°C for 24 h after which the dry weights were measured using an electronic balance. We also selected enough healthy 4th instar females and 4th instar males from the captured O. asiaticus for experiments. We conducted the experiment twice, each with five biological replicates, each cage contained 10 females and 10 males (20 heads in each frame). The survival and developmental time of O. asiatica in the cages were recorded every day to ensure that there were enough males and females for body weight and body length measurement at the end of the experiment. Dead individuals were removed in time when we change fresh S. krylovii. At the end of 7 days, the body lengths of 30 locusts were measured using the vernier caliper. They were then dried in the oven at 90°C for 24 h after which the dry weights were measured using an electronic balance.

Determination of Activities of Antioxidant Enzymes in O. asiaticus
The effect of the pesticide treatments on activities of four antioxidant enzymes [superoxidase dismutase (SOD), peroxidase (POD), catalase (CAT), and polyphenol oxidase (PPO)] in O. asiaticus were measured. To extract crude enzyme solutions, about 0.1 g of insect tissue was weighed, to which 1 ml of an extraction solution was added (purchased from Suzhou Keming Biotechnology Co., Ltd.). They were then placed in an ice bath for homogenization. The supernatant was then centrifuged at 8000 g at 4°C for 10 min and placed on ice for the measurement of enzyme activities with detection kits of each antioxidant enzymes (Suzhou Keming Biotechnology Co., Ltd.). Preparation of the test samples followed the manufacturer's instructions. The absorbance wavelength of each enzyme was measured on a microplate reader (DNM-9602G).
For POD, the wavelength was adjusted to 470 nm, and that of distilled water adjusted to zero. The initial absorbance value, A1, was recorded at 470 nm at 1 min and that of A2 at 2 min. The final absorbance value was given by ΔA = A2−A1. For CAT, the wavelength was adjusted to 240 nm, and that of distilled water was adjusted to zero. The initial absorbance value A1 was recorded at 240 nm and the 2nd absorbance value A2, was recorded after 1 min. The final absorbance value was given by ΔA = A1−A2. For PPO, the wavelength was adjusted to 525 nm, and that of distilled water was adjusted to zero. The absorbance of the determination tube and reference tube was detected at 525 nm. The final absorbance value was given by ΔA = A measured-A control. For SOD, the absorbance was measured at 560 nm. The final absorbance value was given by the formula, Inhibition percentage = (A control tube-control tube) ÷ A control tube × 100%.
The activities of the enzymes were determined using the following formulae;Paragraph text. (1) (V total reaction : total volume of reaction; V sample : volume of added sample; V sample total : volume of added extract; T: reaction time; W: sample mass; ε: molar extinction coefficient of H 2 O 2 , 4.36 × 104 L/mol/cm) The experimental data were all analyzed using Excel 2010, the means were evaluated via the one-way ANOVAs at the 5% significance level, and differences among means were compared by using the Turkey at p < 0.05 . And Prism6 was used for making the figure, and the error bars is the SE.

Effect on Weight
The 10% crude extract of A. sieversiana had significant inhibitory effect on the body weight of female ( Figure 1A) and male 4th instar (Figure 1B) O. asiatica. However, the different concentrations (10 6 , 10 7 , and 10 8 spores/g) of M. anisopliae promoted weight gain in female ( Figure 1A) and male 4th instar (Figure 1B) O. asiatica. Among them, the 10 7 and 10 8 spores/g M. anisopliae concentrations had the strongest effect on the weight gain of female (29.5 ± 1.63 g) and male O. asiatica (26.9 ± 2.00 g), respectively. The combined treatment of 10% crude extract of A. sieversiana and each of the different concentration of M. anisopliae, had a significant inhibitory effect on the weight gain of O. asiatica. Among them, the combination of 10% A. sieversiana crude extract + 10 7 spores/g of M. anisopliae had the most significant inhibitory effect on the weight gain of the female ( Figure 1A) and male 4th instar (Figure 1B) O. asiatica (F = 233.829, DF = 7, p = 0), with the weight gain of 129.73 ± 1.97 g and −7.2 ± 1.07, respectively.

Effects on Body Length
The 10% crude extract of A. sieversiana and M. anisopliae alone or in combination had effects on the body length of female (Figure 2A)

Effects on Growth Rate
The 10% crude extract of A. sieversiana and M. anisopliae alone or in combination had effects on the growth rate of female ( Figure 3A) and male ( Figure 3B) O. asiatica. The growth rate of O. asiatica was significantly inhibited by the 10% crude extract of A. sieversiana, However, all the concentrations of M. anisopliae (10 6 ,10 7 , and 10 8 spores/g) promoted the growth rate of female and male grasshoppers. For female, the 10 7 spores/g concentration had the highest effect on growth rate. For male, the 10 8 spores/g concentration had the highest effect. The results showed that the combination of the 10% crude extract of A. sieversiana and the different concentrations of M. anisopliae (10 6 , 10 7 , and 10 8 spores/g) had significant inhibitory effect on the growth rate of O. asiatica. The combination of 10% crude extract + 10 8 spores/g M. anisopliae and 10% crude extract + 10 7 spores/g M. anisopliae had the most significant inhibitory effect on the growth rate of female and male O. asiatica, respectively.

Effects on Body Length
The 20% crude extract of A. sieversiana and M. anisopliae alone or in combination had effect on the body length of the female ( Figure 6A) and male (Figure 6B)

Effects on Growth Rate
The 20%

Effect of Crude Extracts of A. sieversiana, Different Concentrations of M. anisopliae and Their Combinations on the Mortality of O. asiatica
When A. sieversiana was used alone, the lethal effect of 20% crude extract was greater than that of 10% crude extract. The lethal effect of the different concentrations of M. anisopliae on O. asiatica followed a decreasing order as follows: 10 8 spores/g > 10 6 spores/g > 10 7 spores/g, but there was no significant difference between the 10 6 and 10 7 spores/g concentration. The lethal effect of combined treatment on O. asiatica also followed a decreasing order as follows: 20% As + 10 8 spores/g M. anisopliae > 20% As + 10 7 spores/g M. anisopliae > 10% As + 10 7 spores/g M. anisopliae > 10% As + 10 6 spores/g M. anisopliae > 20% As + 10 6 spores/g M. anisopliae. Among them, 10 8 spores/g M. anisopliae had the best efficacy, resulting into a mortality rate of 85 ± 5.08%. The results showed that the combination of 10% crude extract of A. sieversiana + 10 7 and 10 8 spores/g M. anisopliae, the combination of 20% crude extract + 10 7 and 10 8 spores/g M. anisopliae had significantly better control effect on O. asiatica than that of using only crude extract of A. sieversiana (F = 14.23, DF = 11, p = 0; Figure 9).

Effect of Crude Extracts of A. sieversiana, Different Concentrations of M. anisopliae and Their Combinations on the Activities of Antioxidant Enzymes in O. asiatica
There was no significant difference in the lethal effect of the 20% As + 10 7 spores/g M. anisopliae and 10% As + 10 7 spores/g M. anisopliae on O. asiatica (Figure 9). Likewise the effect of the increase in concentration of A. sieversiana extracts, on the mortality of O. asiatica was not significantly different. Therefore, in this experiment, we tested the activities of antioxidant enzymes in O. asiatica from the following treatments: 10% A. sieversiana extract + 10 7 spores/g M. anisopliae, 10% A. sieversiana extract alone, 10 7 spores/g M. anisopliae alone, and distilled water (control). The activities of catalase (CAT; 414.94 ± 16.94 U/g) and polyphenol oxidase (PPO; 88.20 ± 5.49 U/g) were significantly increased by the 10% A. sieversiana extract + 10 7 spores/g M. anisopliae (Figure 10). The activities of SOD and POD were also activated by this treatment. The highest activity of SOD (2564.10 ± 218.08 U/g) was recorded under the 10 7 spores/g M. anisopliae treatment,  and that of POD (368.40 ± 47.45 U/g) was recorded under the 10% A. sieversiana extract treatment (Figure 10).

DISCUSSION
The frequent large-scale outbreaks of O. asiatica in the grasslands of northern China, are partly responsible for the recent degradation of grasslands and desertification in Inner Mongolia in China. The entomopathogenic fungus M. anisopliae, a recognized non-polluting and non-toxic pesticide, has been widely studied for pest control and has shown potential as a chemical substitute for locust and grasshopper control (Gillespie et al., 2000;Peng et al., 2008). However, insect pathogenic fungi have shortcomings such as a longer time of attacking pest and unstable control effect (Pilz et al., 2009). In addition, external conditions such as temperature, humidity, density, and height of vegetation also affect the control efficacy of M. anisopliae on locust. In dense vegetation areas, the reduced infectivity of M. anisopliae conidia on the vegetation surface also reduces its efficacy (Hunter et al., 2001). Therefore, the use of appropriate adjuvants can increase the persistence of fungi. Many entomopathogenic fungi are compatible with a range of chemical and natural insecticides (Santos et al., 2007;Pelizza et al., 2015). One possible candidate is crude plant extracts. For example, the combination of neem oil and M. anisopliae significantly increased the mortalities of A. aegypti and locusts. By this, the efficacy of this biological control agent was increased (Haroona et al., 2011;Paula et al., 2019). However, neem oil inhibited the vegetative growth of Beauveria bassiana and reduced the yield and viability of conidia (Depieri et al., 2005).
The secondary metabolites of Artemisia plants include terpenoids, flavonoids, coumarins, glycosides, steroids, and polyacetylenes (Tan et al., 1998). The crude extract of A. sieversiana was shown to have insecticidal activity (Liu et al., 2010). In this regard, we also found that the treatment with A. sieversiana alone significantly reduced the mortality of O. asiatica. Correspondingly, it caused a series of immune and physiological responses in the host insect after M. anisopliae infected the host (Wang et al., 2013;Ling et al., 2016). Therefore, the high concentration of conidia may enhance host fungal infections and cause high mortality in a relatively short time (Kirkland et al., 2004;Shrestha et al., 2015). This study also indirectly confirmed that a higher concentration of M. anisopliae enhances its lethality.
Fungal insecticides have good potential for pest control, as there is currently no record of development of insect resistance against them (Behie and Bidochka, 2013;Erler and Ates, 2015). In order to reduce the risk of resistance, the use of fungal agents should be minimized. The efficacy of a combination of fungi and chemicals was shown to be higher under laboratory or field conditions, and has been reported in several studies involving Anomala cuprea Hope and Locusta migratoria (Hiromori and Nishigaki, 1998;Jia et al., 2016). Although there was no significant difference between the mortality of O. asiatica treated with 20% crude extract of A. sieversiana combined with 10 8 spores/g M. anisopliae and 10 8 spores/g M. anisopliae alone in this study, the overall performance of O. asiatica was decreased. This indicates that the efficacy of the crude extract of A. sieversiana was significantly improved.
The death time of pest is an important factor affecting the control efficacy of insecticides for pest management (Asi et al., 2008;Gupta et al., 2009). The results of a field application of 6 × 10 7 spore M. anisopliae bait on locusts, showed that the peak infection of the locust population was reached on the 15th day after infection, with a mortality rate of 6.5%, and the mortality rate increased with time (Dong et al., 2011). Compared to this experiment, the mortality rate was much lower than that in the treatment with crude extract of A. sieversiana and M. anisopliae, and the peak time of infection was three times longer than that under the mixture of crude extract of A. sieversiana and M. anisopliae. The results showed that treatment with the combination of the crude extract of A. sieversiana and M. anisopliae significantly shortened the lethal time, and its efficacy was significantly better than that of other groups and any single agent. This shows that the synergetic control effect from the combination treatment was faster and better than the control effect of each pesticide alone. The effect of 10% A. sieversiana crude extract mixed with 10 7 spores/g M. anisopliae on the mortality of O. asiatica reached the effect of 10 8 spores/g M. anisopliae alone in this study; therefore, the dosage of M. anisopliae was reduced, and may largely prevent the development of resistance in insects against it.
Although it has not been reported that insects resistance to M. anisopliae (Dubovskiy et al., 2013), the risk of resistance to M. anisopliae can be reduced by observing the effects of A. sieversiana and M. anisopliae on the protective enzyme activities of O. asiatica. We found that the enzymatic activities of catalase and polyphenol oxidase of O. asiatica were over-expressed under the mixed treatment of the combination of the crude extract of A. sieversiana and M. anisopliae. And some studies have found that the high expression of catalase could shorten the germination time and increase the pathogenicity. of M. anisopliae (Hernandez et al., 2010). Therefore, we found that the crude extract of A. sieversiana can be used as a synergist of M. anisopliae to improve its pathogenicity. Under the conditions of external factors, such as plants and biological pesticides, the slower growth and development of insects are often associated with the expression of enzyme activity in their bodies (Serebrov et al., 2006). Our results also confirmed the conclusion that the mortality, body weight, growth rate, and overall performance of O. asiatica were significantly reduced under the treatment with crude extract of A. sieversiana and M. anisopliae, while the activity of its protective enzymes was significantly increased.
The use of M. anisopliae in combination with chemical agents have caused varied mortality rates of locusts, and even showed antagonistic effect . In our research, similar results of different concentration ratios causing varied mortality of O. asiatica were recorded. Therefore, in the considerations for the use of a combination of new biopesticides and fungal preparations for pest control, the compatibility of the two should be carefully evaluated. A number of studies have comprehensively evaluated the compatibility of fungi and chemicals agents under laboratory and field conditions (Cuthbertson et al., 2005;Asi et al., 2010).
We found an interesting phenomenon in this study. Metarhizium anisopliae can promote the increase of body weight and body length of O. asiatica. However, when M. anisopliae was mixed with the crude extract of A. sieversiana, it inhibited the growth of body weight and length, especially the growth of body weight, and the effect of the mixture of M. anisopliae and the crude extract of A. sieversiana on the development of female was greater than that of male. However, the reasons for this phenomenon need to be further studied. Our work was carried out under laboratory conditions. We, therefore, propose that further studies under field conditions characterized by many complicated and uncontrollable factors, be carried out to determine the best compound formula which shows promising efficacy. Additionally, other parameters such as the toxicity of fungi, the influence of the mixture of crude extract of A. sieversiana and M. anisopliae on natural enemies and non-target organisms, etc. should be considered in the evaluation of the combination of fungi and new biopesticides.
In summary, our research showed that M. anisopliae was a good biological control agent for on O. asiatica. The crude extract of A. sieversiana increased the virulence of M. anisopliae and shortened the mortality time of the pest. These results indicate that the crude extract of A. sieversiana could be used as a synergist for M. anisopliae, and this compound formulation can become an alternative biological control agent to control insect pest.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding authors.

AUTHOR CONTRIBUTIONS
ZZ and XT designed the experiments. SL and XT performed the experiments. SL and ZZ wrote the manuscript. SL, CX, and GD collected the data and analyzed it. ZZ and GW provided technical and material support. All authors contributed to the article and approved the submitted version.