Efficacy of soil-borne entomopathogenic fungi against subterranean termite, Coptotermes curvignathus Holmgren (Isoptera: Rhinotermitidae)

Coptotermes curvignathus Holmgren (Isoptera: Rhinotermitidae) is a subterranean termite that poses serious damage to oil palm and rubber trees. Chemical pesticides could cause negative effect to human and the environment in long-term usage. The use of entomopathogenic fungi (EPF) to suppress the population of subterranean termites is in favour when compared to chemical pesticides because they do not harm to the environment and non-target organisms. The study aimed to isolate and identify the EPF from the soil using yellow mealworm larvae of Tenebrio molitor Linnaeus in the baiting method and assessed their efficacy against subterranean termite, C. curvignathus. Eleven EPF isolates were successfully isolated from the oil palm plantation in Universiti Putra Malaysia, namely: Aspergillus auricomus (UPM-A1C-1), A. caelatus (UPM-A1C-2), Metarhizium anisopliae var anisopliae (UPM-A2C-1, UPM-A3C-1, UPM-A3C-2, UPM-A5C-1 and UPM-A10C-1), Purpureocillium lilacinum (UPM-A2C-3 and UPM-A7C-1), Cordyceps javanica (UPM-A2C-5), and M. pinghaense (UPM-A13C-2). The identity of these EPF were confirmed by morphological and molecular characteristics. All EPF yielded 100% mortality in C. curvignathus in 10 days after inoculation (DAI), except UPM-A1C-1 and UPM-A1C-2 after exposure to 1 × 107 conidia ml−1. UPM-A2C-5 Cordyceps javanica yielded the highest mycelia formation (69%) after 6 DAI. The LT50 values varied from 3.90 to 7.75 days. UPM A2C-1 M. anisopliae var anisopliae showed the lowest LT50 (3.90 days), while UPM-A1C-1 Aspergillus auricomus showed the highest LT50 (7.75 days). The lowest LC50 value (1.49 × 105 conidia ml−1) was recorded in UPM A2C-1 M. anisopliae var anisopliae. The present study confirmed the soilborne EPF with potential insecticidal activity against C. curvignathus. UPM-A2C-1 M. anisopliae var anisopliae was a potential biological control agent against Subterranean termite, C. curvignathus due to its virulence score and high percentage of mycelia formation after 6 DAI. The data reported in the present study, particularly using P. lilacinum, M. pinghaense, Aspergillus auricomus, A. caelatus and C. javanica with potential insecticidal activity against C. curvignathus, are new records.


Background
The subterranean termite, Coptotermes curvignathus Holmgren (Isoptera: Rhinotermitidae) was recorded in 1927 in Malaya attacking oil palm, Elaeis guinensis Jacq. (Deasy 2008) and it proved to be the main pest of oil palm, particularly in the immature palm trees. They are the most aggressive and largest subterranean termites among the genus Coptotermes spp. (Wong et al. 2015). They feed vigorously on the fresh tissue of oil palm tree as the main diet rather than fed on wood-based materials. This pest attacks palm trees by creating earthy coloured mound and permanent working trails from the centre part of the trees, after which the affected trees gradually dry up and die (Kon et al. 2012).
The common insecticides used in controlling insect pest could cause negative effect to human and the environment in long term usage (Aktar et al. 2009). Development of resistance to chemical pesticides has become the major concern in the oil palm industry. One of the alternatives to overcome it is the application of EPF. EPF such as Metarhizium, Beauveria, Aschersonia, Lecanicillium, Aspergillus, Tolypocladium, and Hirsutella are members of phylum Ascomycota and have been commercially manufactured as biopesticides (Bischoff et al. 2006). The mode of infection is the same for all EPF. Fungal conidia will bind to the cuticle of the hosts by hydrophobic binding under desirable condition and geminate into germ tubes (Inglis et al. 2000). The germ tubes penetrate the cuticle by producing metalloid proteases and amino peptidases. EPF form hyphal during parasitic stage, and live as a saprophyte by obtaining nutrients from the carcass while maintaining hypha development upon death of their hosts. They will form fungal mycelia when the insect dies (Inglis et al. 2000). Most of the EPF produced secondary metabolites known as destruxins that consist of an a-hydroxy acid and 5 amino residues which could weaken the insect immune response (de Bekker et al. 2013). Some EPF have been successfully developed for commercial purposes. Beauveria bassiana and Metarhizium anisopliae were among the EPF commercialized as biopesticide to target for a wide range of insect hosts. Till date, majority of the EPF based biopesticides (approximately 50 products in the global market) contain Metarhizium spp. as the active agent. In Malaysia, the trend in using biopesticide against insect pest of oil palm plantation has become popular.
Several EPF have been isolated, identified and tested pathogenic to C. curvignathus in Malaysia. The pathogenicity of different species of EPF may be influenced by factors such as percentage of mortality (Lo Verde et al. 2015), effective lethal time (LT 50 ) (Sileshi et al. 2013) and the ability of spore to germinate fast on target pest (Montesinos-Matías et al. 2011). The above factors can be used as the criteria for selection of good EPF candidate for target pest. The present study was carried out to evaluate the efficacy of 11 soil borne EPF against subterranean termite, C. curvignathus under laboratory conditions.

Methods
Research experiments were conducted at the Laboratory of Insect Pathology, Faculty of Agriculture, Universiti Putra Malaysia from 2019 to 2021.

Sampling of termites
Termites were collected from the infested oil palm trees at the oil palm plantation in Universiti Putra Malaysia, Serdang, Selangor. Termites were baited using termite trap with some modifications of the design developed by Evans and Gleeson (2006). The specimens of C. curvignathus were kept at room temperature at the Laboratory of Insect Pathology for 7 days for habituation before they were used in the pathogenicity study.

Sampling of soil
The soil samples were obtained from 30 different sampling sites at the oil palm plantation in Universiti Putra Malaysia, Serdang, Selangor, Malaysia. The distance between each sampling site was more than 10 m. The sampling locations were divided into 2 groups depending on the pesticide exposure. Group 1 consisted of area with pesticide (malathion) application, while group 2 consisted of area without pesticide application. Both areas were free from biopesticide exposure. Approximately 300 g of soil was collected by scrapping 10-15 cm deep into the soil and deposited separately in clean plastic containers. Five soil samples were collected from each sampling location. Soil samples were brought back to the Laboratory of Insect Pathology at the Department of Plant Protection, Faculty of Agriculture, Universiti Putra Malaysia. The containers containing the soil were covered tightly with lid containing small holes to allow gaseous exchange and air-dried at 27 ± 2 °C for 72 h.

Isolation of EPF
Yellow mealworms, T. molitor were used in baiting EPF from the soil samples. Mealworms were surface-sterilized with 0.1% (v/v) sodium hypochlorite and then heat-treated at 60 °C for 30 s prior to experiment. The mealworms may appear dead at first after the heat treatment. Three quarters of the plastic containers (4.0 cm × 7.5 cm) were filled with soil. The soil was rehydrated by spraying with distilled water. Ten larvae (alive) were placed on the surface of the soil. The plastic containers were kept at room temperature for 30 days in dark with the soil sample and the larvae were gently inverted every day during the first week of the baiting process. The dead larvae were harvested and placed on moist filter paper in a Petri dish and incubated for 5 days until EPF had fully grown on the larval body. The larvae infected with EPF were disinfected with 2% (v/v) sodium hypochlorite for 2 min, followed by rinsing 3 times with distilled water before incubated at 27 ± 2 °C on malt extract agar (MEA). EPF grown on MEA were sub-cultured to potato dextrose agar (PDA) a few times until pure cultures were obtained.

Morphological identification of EPF
Fungal discs (3 × 3 mm) of 7 days old culture were placed in the middle of fresh PDA and incubated at 27 ± 2 °C for 14 days, following the method of Bischoff et al. (2006). The culture near to the margin was harvested by using an inoculation needle for morphological study as this area contained mycelia, hyphae, young and matured conidiophores, while deeper toward the centre colony area contained older conidiophores that were heavily sporulated (Bischoff et al. 2006). The mycelia were mounted on a microscopic slide with a drop of lactophenol cotton blue mounting medium. A cover lip was positioned over the mount by placing one edge of the cover lip to the slide and slowly covered the whole mount so that no air bubble was trapped within the mount. Morphological characteristics of fungal culture such as the texture, form, and colour on upper and lower surfaces of media were recorded. The size, length, width and length/width ratio of conidia were also calculated. The micro-morphological features of all EPF isolates were observed under compound microscope (Olympus BX 41, Olympus Corporation, Tokyo, Japan) at 100X magnification and the photomicrographs of the EPF cultures were taken using Dino-Capture 2.0 Software.

Molecular identification of EPF
One week old actively grown EPF were sub-cultured by placing 3 mm mycelia plugs into Erlenmeyer flasks containing 100 ml of potato dextrose broth added with 1% (w/v) yeast extract. The broth cultures were agitated in a refrigerated orbital thermostat shaker at 160 rpm at 27 ± 2 °C for 5-7 days until a thin layer of mycelia formed on the broth. The mycelia were harvested, dried on a Whatman 2.0 filter paper and then ground with liquid nitrogen in a 1.5 ml microcentrifuge tube to a fine powder by using a sterile polypropylene disposable pestle. Approximately 500 mg of mycelial powder was used in the DNA extraction. DNA was extracted with EZ-10 Spin Column Fungal Genomic DNA Mini-Preps Kit (BIO BASIC INC) according to the manufacturer's protocol. The ITS region of rDNA was amplified using the primers ITS1 (5′-TCG GTA GGT GAA CCT GCG G-3′) and ITS4 (5′-TCC TCC GCT TAT TGA TAT GC-3′) (Gardes and Bruns, 1993), the Beta-Tubulin (βTub2) region with primers Bt2a (5′-GGT-AAC-CAA-ATC-GGT-GCT-GCT-TTC-3′) and Bt2b (5′-ACC-CTC-AGT-GTA-GTG-ACC-CTT-GGC-3′) (Glass and Donaldson, 1995), and the regions of translation elongation factor 1 alpha (TEF1-α), with primers EF1-1251R (5′-CCT-CGA-ACT-CAC-CAG-TAG-CG-3′) and EF1-668F (5′-CGG-TCA-CTT-GAT-CTA-CAA-GTG-C-3′) (O'Donnell et al.1998). Amplification reaction was prepared in a final volume of 25 μl. The amplification program for ITS region was performed with initial denaturation at 95 °C for 5 min, followed by 35 cycles with denaturation at 95 °C for 1 min, annealing at 52 °C for 30 s, and extension at 72 °C for 1 min, and a final extension at 72 °C for 10 min. The amplification program for βTub2 region was performed with initial denaturation at 94 °C for 1 min, followed by 35 cycles with denaturation at 94 °C for 30 s, annealing at 58 °C for 30 s, and extension at 72 °C for 5 min, and a final extension at 72 °C for 7 min. Lastly, the amplification program for TEF1-α region was performed with initial denaturation at 95 °C for 4 min, followed by 30 cycles with denaturation at 95 °C for 1 min, annealing at 57 °C for 75 s, and extension at 94 °C for 1 min, and a final extension at 72 °C for 10 min. PCR products were separated on a 1% (w/v) agarose gel with CSL-MDNA-1KbPLUS DNA Ladder RTU used as DNA marker. The PCR products were then sent out for sequencing services.

Sequence analysis
Upon receiving the DNA sequencing data, the noises of the sequences were removed using Biological Sequence Alignment Editor (BioEdit) version 7.2. The sequences were checked for their alignment by using Mega X software and corrected manually. The nucleotide sequences were blasted with sequences of others species deposited in the data bank at the National Center for Biotechnology and Information (NCBI) (http:// www. ncbi. nlm. nih. gov) using Basic Local Alignment Search Tool (BLAST) (Altschul et al. 1990). Phylogenetic tree was constructed using MEGA X software (Tamura et al. 2011) based on the Maximum Likelihood-Joining tree (Saitou and Nei 1987) by 1000 bootstrap value (Felsenstein 1985). Only branches with more than 30% bootstrap value are shown.

Preparation of conidia suspension
Conidia suspension of EPF isolates was prepared by adding 15 ml of sterile 0.05% (v/v) Tween 80 into the 4 weeks old EPF cultures and gently scraping the surface of the cultures with a sterile "L" shaped inoculation loop to dislodge the conidia from the surface of the PDA plates. The conidial suspensions were filtered through a 3 layer-sterile muslin cloth into 50 ml sterile plastic tubes to remove mycelium and residue of agar. The filtrate was vortexed for 5 min. The number of conidia was counted using a Neubauer haemocytometer with the help of a compound microscope (Olympus BX 41, Olympus Corporation, Tokyo, Japan) under 40X magnification. An initial stock suspension was prepared in 1 × 10 9 conidia ml −1 and Page 4 of 16 Kamarudin et al. Egyptian Journal of Biological Pest Control (2022) 32:44 then centrifuged at 8000 × g for 1 min. After washing for 5-10 times, the final pellet was then resuspended in 1 ml of double-autoclaved distilled water.

Mortality test of EPF against subterranean termite, C. curvignathus
A screening test was carried out on C. curvignathus using the conidia concentration of 1 × 10 7 conidia ml −1 containing 0.05% (v/v) Tween 80. Treatment was carried out by inoculating each individual termite with 20 μl conidia suspension (topical application) onto the dorsal abdominal segment of the termite. C. curvignathus were surface-sterilized with 0.01% (v/v) sodium hypochlorite and washed 3 times with double-autoclaved distilled water. The treated C. curvignathus were dried for 2 min. 10 termites per treatment with ten replications were accomplished. C. curvignathus treated with 0.05% (v/v) Tween 80 were served as negative control. All treated and untreated C. curvignathus were then placed on a wet Whatman No. 1 filter paper located inside a 90 mm × 15 mm non-treated polystyrene petri dish at room temperature for 24 h in dark. The signs and symptoms of infection, and the mortality rate were recorded daily until 100% mortality was achieved. Mortality data were corrected using Abbott's formula (Abbott 1925). Graph of mean percentage of cumulative mortality against days was plotted using Prism GraphPad for Mac version 8.1.1. The effective lethal time (LT 50 ) was determined by using Probit analysis method in the SPPS software version 27.0 (SPSS Inc., Chicago, IL, USA).

Assessment of mycoses of C. curvignathus
C. curvignathus that showed mycelial growth on their carcass (mycoses) were placed on PDA and maintained at 27 ± 2 °C for fungal growth. The mean percentage of mycoses was recorded and the mean separation test using Tukey HSD at alpha 0.05 (SAS Institute Inc., USA, version 9.4).

Virulent score
The virulent score assessment consisted of 3 criteria, namely the mean mortality at 6 DAI, LT 50 value and mean percentage of mycelia growth (Table 1). The virulent score consists of 5 levels as stated in Table 1.

Concentration dependent bioassay against C. curvignathus
EPF with virulent score of 3.1-5.0 were chosen for concentration dependent mortality test. The C. curvignathus were applied with five different conidia concentrations (1 × 10 9 , 1 × 10 8 , 1 × 10 7 , 1 × 10 6 and 1 × 10 5 conidia ml −1 ) prepared by serial dilution. C. curvignathus was surface sterilized with 0.01% (v/v) sodium hypochlorite and washed 3 times with double-autoclaved distilled water. After air-dried, each individual termite was inoculated with 20 μl conidia suspension containing 0.05% (v/v) Tween 80 on the dorsal abdomen. The inoculated C. curvignathus were dried for 2 min. C. curvignathus treated with 0.05% (v/v) Tween 80 were served as negative control. Ten termites per treatment with 10 replications were accomplished. All treated and untreated C. curvignathus were then placed on wet Whatman No.1 filter paper inside the 90 mm × 15 mm Petri dish at room temperature for 24 h in dark. Mortality rate was recorded daily until 100% mortality achieved, and the mortality data were corrected using Abbott's formula (Abbott 1925).

Statistical analysis
Means of treatment were compared using Tukey's test at α < 0.05 using Statistical Analysis Software (SAS Institute Inc., USA) version 9.4. The percentage of mycoses C. curvignathus was calculated by comparing the mortality in the control experiment. The lethal concentration (LC 50 ) value was determined by using Probit analysis method in SPSS software version 27.0 (SPSS Inc., Chicago, IL, USA).

Isolation of EPF
Eleven EPF were successfully isolated from the soil in the oil palm plantation in UPM using insect bait method. Larvae of T. molitor showed different types of mycoses symptoms (Fig. 1).

Morphological and molecular identification of EPF
EPF can be differentiated by colony texture, colony shape and colony colour at upper and lower surfaces on PDA, and conidia size and shape (Tables 2, 3). The EPF were molecularly analysed via the amplification and sequencing of their ITS, TEF1-α and βTub2 regions to confirm the morphological identification. The nucleotide sequence of ITS, TEF1-α and βTub2 genes were 500-650, 600-900 and 300-700 bp, respectively. ITS, TEF1-α and βTub2 nucleotide sequences were submitted to NCBI GenBank and accession numbers were obtained.

Assessment of C. curvignathus with mycoses
The percentage of C. curvignathus with mycoses was correlated to those of mortality test (Table 5). Among the EPF tested, UPM-A1C-1 A. auricomus and UPM-A1C-2 A. caelatus were the slowest EPF in mycelia formation on the carcass, while the UPM-A2C-5 C. javanica, UPM-A2C-1 M. anisopliae var anisopliae and UPM-A13C-2 M. pinghaense scored the highest percentage of C. curvignathus with mycoses after 6 DAI. UPM-A2C-3 P. lilacinum and UPM-A5C-1 M. anisopliae var anisopliae did not show promising mycelia formation on C. curvignathus with their high percentage of mortality after 6 DAI.

Symptom of C. curvignathus with mycoses
The C. curvignathus infected with the EPF became inactive and showed abnormal behaviour such as avoiding others caste, changing in dietary and less feeding as early as 1DAI. The dead C. curvignathus become harden and reduced in body size. A thin layer of mycelia slowly Table 4 Percentage of sequence similarity based on ITS, βTub2 and TEF1-α genes of EPF morphospecies isolated from UPM GAN GenBank Accession Number appeared on the carcass. Different EPF showed different colours of conidia formed on the dead C. curvignathus. Metarhizium spp. produced white conidia on the dorsal abdomen of dead C. curvignathus and gradually covered the whole body of the carcass. The white conidia turned to green colour after 3DAI (Fig. 5A-F). UPM-A2C-3 P. lilacinum and UPM-A7C-1 P. lilacinum produced vinaceous-brown conidia on the carcass of C. curvignathus ( Fig. 5G, H). C. curvignathus infected with UPM-A2C-5 C. javanica were covered with greyish conidia starting from the dorsal abdomen and then spread to the whole body of the infected C. curvignathus (Fig. 5I). C. curvignathus infected with Aspergillus spp. had a dark integument after death before formation of conidia on their body. The conidia were scattered all over the body. The UPM-A1C-2 A. caelatus produced yellowish green  et al. Egyptian Journal of Biological Pest Control (2022) 32:44 conidia (Fig. 5J), while the UPM-A1C-1 A. auricomus produced yellow conidia (Fig. 5K).

Discussion
Soil is a natural habitat for microorganisms such as entomopathogenic fungi and other beneficial microorganisms and pathogens. It protects the soil-borne microorganisms from the environmental stress such as UV radiation and humidity, as well as providing micro-and macro-nutrients to them (Ignoffo and García 1995). The baiting method used in the present study was able to isolate 11 EPF of 4 different fungi genera. Several studies reported that insect baiting method by using T. molitor to isolate soil borne EPF have been shown to acquire many EPF from various genus (Kim et al. 2018 Visagie et al. (2014). Mongkolsamrit et al. (2020) concluded that specimens with 100% sequence homology can be identified as the same strain, 99% sequence homology as the same species, and 89-99% sequence homology as the same genus. The UPM EPF showed more than 99% sequence homology to those published sequences in GenBank. Both morphological and molecular results have confirmed the UPM EPF into species level. Construction of a phylogenetic tree depends on the availability of reference sequences. Reference sequences of TEF-α gene were lacking for A. auricomus and A. caelatus in the present study. The phylogenetic trees had grouped the UPM EPF into different monophyletic clusters. βTub2 gene showed a better grouping of UPM EPF into cluster compared to ITS gene and TEF1-α gene.
Symptoms and signs shown by the infected C. curvignathus varied among the UPM EPF, however, the formation of mycelia and colour of conidia were in line with other published EPF either on termites or other insects. UPM EPF such as C. javanica and M. anisopliae var anisopliae induced more than 65% mycoses at 6 DAI. Other EPF were slow in mycoses formation. Beside the data on the pathogenicity study, mycelium and spore formation is also an important factor in the selection of potential candidate as a biological control agent (Ansari et al. 2004). Grooming behaviour is a defensive mechanism in termites' colony in order to protect themselves against disease (Yanagawa et al. 2008). This behaviour was observed throughout the mycoses formation on the C. curvignathus infected with different isolates of EPF. Dead infected C. curvignathus were covered with pieces of filter paper in Petri dish by other alive C. curvignathus to protect the spread of fungi to other members. Infected members were observed having social distancing among the alive and infected members during the