Insecticidal effect of the entomopathogenic nematode Heterorhabditis megidis (Nematoda: Heterorhabditidae) baited from the soil on the larvae of Monochamus alternatus (Coleoptera: Cerambycidae)

To scout for an entomopathogenic nematode (EPN) that effectively controls the pine sawyer beetle, Monochamus alternatus Hope (Coleoptera: Cerambycidae), we screened EPNs from soil samples using M. alternatus larvae as bait. A population of the nematode obtained was named isolate SOz01, and molecularly identified as Heterorhabditis megidis Poinar, Jackson and Klein (Nematoda: Heterorhabditidae). In the inoculation test of the infective juveniles (IJs) of SOz01 onto mature larvae of M. alternatus under laboratory conditions, a mortality of 86–100% occurred when more than 20 IJs were inoculated. Even when the larvae were inoculated with only five IJs, one-third of them died. This finding suggests that H. megidis SOz01 has a strong insecticidal effect on M. alternatus.


Introduction
The pine sawyer beetle, Monochamus alternatus Hope (Coleoptera: Cerambycidae), is the principal vector of the pinewood nematode, Bursaphelenchus xylophilus (Steiner and Buhrer) Nickle (Nematoda: Aphelenchoididae), the causal agent of pine wilt disease (PWD), in Japan, Korea, and China (Kishi 1995;Shin 2008;Zhao 2008). In Japan, PWD is usually controlled by killing the vector insects through fumigating damaged trees and spraying chemical insecticides onto live pine trees (Kamata 2008). On the other hand, the use of the natural enemies of vector insects has long been studied (Shimazu 2008). Thus far, no practical use of a biological agent has been achieved, except for the entomopathogenic fungus Beauveria bassiana (Balsamo) Vuillemin (Shimazu and Higuchi 2007).
Entomopathogenic nematodes (EPNs) primarily comprise the Heterorhabditidae and Steinernematidae families (Grewal et al. 2005) and possess a symbiotic relationship with entomopathogenic bacteria of the genera Photorhabdus and Xenorhabdus, respectively (Poinar 1979(Poinar , 1990. EPNs undergo an infective juvenile stage (IJ) in their life cycles. They invade the insect host through the mouth, anus, and tracheal system, and then release their symbiotic bacteria into the hemocoels (Goodrich-Blair and Clarke 2007). The toxins produced by the bacteria kill the insect and the propagated nematodes emerge from the cadaver as IJs to seek a new host (Poinar and Thomas 1966).
There have been several studies on the insecticidal activity of EPNs against M. alternatus (Katagiri et al. 1984;Mamiya and Shoji 1986;Mamiya 1989;Yamanaka 1993;Yu et al. 2016). High insecticidal rates of M. alternatus larvae treated with EPNs have been attained under experimental conditions; however, their practical use in the field has not been achieved. This is probably because of the insufficient mobility of the nematodes to penetrate under the bark and access the M. alternatus larvae living deep in the wood (Phan 2008). Furthermore, the EPNs employed in these studies were originally used to control other insect species. In bait trapping to screen EPNs, Galleria mellonella L. (Lepidoptera: Pyralidae) has been widely used as the bait insect (Bedding and Akhurst 1975), whereas there are attempts to use the insects to be controlled as bait (Kushida et al. 1986(Kushida et al. , 1987Mamiya 1988Mamiya , 1989. The screening of EPNs using M. alternatus as bait may yield an EPN that has a strong insecticidal effect on M. alternatus under natural conditions.
In this study, we screened EPNs from soil using M. alternatus larvae as bait. The EPN obtained was molecularly identified and its insecticidal effect against M. alternatus larvae was evaluated under laboratory conditions.

Insects used
The insects used were the mature M. alternatus larvae reared in our laboratory using the artificial diet reported by Maehara et al. (2018), which originated from the insects obtained in Oshu City, Iwate Prefecture, Japan. Mature larvae were kept at 10 °C in the dark until used.

Screening of EPNs from soil
Based on the Galleria baiting method by Bedding and Akhurst (1975), we conducted a screening of EPNs from the soil using M. alternatus larvae as bait. Soil samples were collected from four forest sites in Iwate and Miyagi Prefectures, Japan (Table 1). We took 500-700 mL of soil, excluding leaf litter, from the forest floor at a depth of 20-30 cm, and packed it in zippered plastic bags (220 × 170 mm). The following day, approximately 350 mL of soil was transferred into a plastic cup (φ130 × 60 mm high) and two M. alternatus larvae were placed on the soil surface. The cup was then covered with a lid and placed under dark conditions at 25 °C. The larvae were observed daily for 27-37 days to monitor their survival. When a larva was immobile and the body color turned black or red, it was regarded as dead.
Each dead larva was washed with distilled water (DW) and individually placed on a 55-mm filter paper (ADVAN-TEC, No. 1) in a 6-cm plastic Petri dish. The Petri dish was wrapped with ParafilmM® (Bemis Flexible Packaging) to avoid drying and incubated in the dark at 25 °C. Thereafter, we checked the larval body color and the presence of nematodes inside the larval body once every 2-3 days. A larva that showed a reddish body color and had nematodes inside its body was transferred to the White trap (White 1927) to collect the IJs of the EPN according to the following procedure: a lid of 6-cm plastic Petri dish (8.8 mm high) was placed in the center of an unsterilized 9-cm glass Petri dish (21 mm deep); three pieces of 55-mm filter paper were placed so that the outermost edges were soaked in DW poured in the 9-cm Petri dish; the dead larva was placed on the filter paper; the nematodes that emerged from the larval body were trapped in the water.
The EPN obtained was inoculated on M. alternatus larvae. We placed a larva on DW-moistened filter paper in a 6-cm plastic Petri dish and applied a 0.5-mL suspension containing 25 IJs of the EPN directly onto the larval body. Once the larva died, it was transferred to the White trap to collect IJs propagated in the dead larva. The culture population of the EPN was maintained by repeating this procedure.

Molecular identification of the EPN
The genomic DNA of the EPN was extracted with extraction buffer supplied with the B. xylophilus Detection Kit (Nippon Gene Co., Ltd.).
We used primers no. 93 and no. 94 for PCR amplification of ITS rDNA; this primer pair was designed by Stock et al. (2001) and was used to determine the phylogenetic relationships of Heterorhabditis nematodes (Maneesakorn et al. 2011). PCR amplification was performed according to the protocol for GoTaq® G2 Hot Start Green Master Mix 2X (Promega), where the primer annealing temperature was 47.8 °C. The singly amplified PCR product of approximately 800 bp was purified by the QIAquick PCR Purification Kit (QIAGEN). The purified sample was sequenced with ABI The sequence obtained was compared for homology using the NCBI nucleotide BLAST. Subsequently, we conducted a phylogenetic analysis using the sequences of H. megidis Poinar, Jackson, and Klein, including the one obtained in this study and five closely related species (Fig. 1) obtained from the NCBI nucleotide collection database.
The ClustalW Multiple alignment using BioEdit v.7.0.5.3 was employed to compare sequences (Thompson et al. 1994;Hall 1999). A phylogenetic tree was constructed from the maximum likelihood method and the Kimura 2-Parameter Model (Kimura 1980) using Molecular Evolutionary Genetics Analysis v.6.0 (MEGA6) software (Tamura et al. 2013). The bootstrap value was set at 1000. Fig. 1 Phylogenetic tree of Heterorhabditis megidis and closely related species to H. megidis using sequences containing ITS-1, ITS-2, and the 5.8S ribosomal RNA gene. The phylogenetic tree was constructed by the maximum likelihood method using 59 sequences. The final data set included 386 positions. The numbers at each branch represent the bootstrap probabilities of the branch points. The species name and the accession number of each Heterorhabditis nematode are indicated at the terminal of each branch. H. zealandica (EF530041) was set as an outgroup in the phylogenetic tree. For H. megidis, the name of the country of origin was given if available; and designated "N/A" if the country was unknown. The arrow shows the nematode species obtained in this study (SOz01)

Insecticidal effect of the EPN
The IJs of the EPN for the inoculation test were taken from the culture population. The IJs were washed three times with DW and kept for 14 days at 5 °C before the inoculation test according to the manual of EPNs by Yoshiga (2014).
The inoculation test was repeated three times. The IJs used in the trials were each prepared whenever the tests were done. We set five levels of inoculum-density (i.e., number of IJs inoculated per M. alternatus larva) treatments to test the insecticidal effect: 320, 80, 20, 5, and 0 IJs (control). The M. alternatus larvae used in the test were weighed the day before EPN inoculation, and divided into five treatment groups in a manner that minimized the body weight bias between the groups (Online Resource 1). The larvae were washed with DW, immersed in 70% ethanol for 5 s, and washed with sterilized distilled water (SDW) before they were individually placed in a 6-cm plastic Petri dish with 55-mm filter paper moistened with SDW. An EPN suspension containing 320, 80, 20, 5, or 0 IJs in 500 µL of SDW was dropped directly onto the insect's body surface. A Pipetman (P-200, GILSON) attached to a transparent glass Pasteur pipette at the tip was used, making the inoculum visible and ensuring the correct number of nematodes was inoculated. The Petri dishes were wrapped with ParafilmM® and incubated in the dark at 25 °C. The larvae were observed every 24 h to check their survival. A larva was regarded as dead when it stopped moving and its body color turned red (including partial discoloration) or black. The observation was terminated when all of the larvae either died or pupated (Fig. 2).
Dead larva was washed with SDW and transferred to a different 6-cm plastic Petri dish with SDW-moistened filter paper. The inside of the original Petri dish and both sides of the filter paper removed from the Petri dish were washed with DW. The number of nematodes detected in the washing water was the lowest estimated number of IJs that did not invade the larva. The dead larva was dissected within 3-4 days after expiring, and the developmental stages of the EPNs (i.e., infective juvenile, fourth-stage juvenile, hermaphrodite adult, and next-generation small juvenile) detected in the larval body were determined and counted. For larvae in which no nematode was found by dissection, we determined the cause of death based on the body color and tissue conditions of the cadavers. It is known that an insect infected with H. megidis shows a unique red coloration (Poinar et al. 1987;Yoshiga 2014) and a dense mucilaginous appearance of its disintegrated tissues, referred to as "gummy consistency" by Poinar (1979).

Statistical analysis
The fresh weight of M. alternatus larvae used in the inoculation test was compared by one-way ANOVA among the insect groups of the five different densities of inoculum treatments. We used the Kruskal-Wallis test to compare the percentage of EPNs detected in the dead M. alternatus larvae to the total number of inoculated nematodes. All analyses were performed by R statistical software (v.3.6.1).

Screening of EPNs from soil
A total of 20 living larvae of M. alternatus were placed on 10 soil samples (Table 1), and 16 larvae died. Among Fig. 2 Survivorship of the Monochamus alternatus larvae in the inoculation test of Heterorhabditis megidis SOz01. The upper, middle, and lower graphs correspond to the first, second, and third trials, respectively. Different symbols indicate different inoculum-density treatments (i.e., the number of nematodes inoculated per insect larva): white circles indicate zero nematodes (control), black squares indicate five nematodes, black triangles indicate 20 nematodes, white inverted triangles indicate 80 nematodes, and black diamonds indicate 320 nematodes per larva them, one dead larva placed on soil sample b-3 taken in a forest stand of Japanese red pine (Pinus densiflora Sieb. and Zucc.) turned its body color red, and nematodes were observed moving inside the body. The nematodes derived from this larva were inoculated onto living M. alternatus larvae, which were consequently infected with the nematodes and died. Therefore, the nematode was confirmed to be an EPN. We named the nematode isolate obtained "SOz01". No EPN was detected in the other 15 dead larvae during the screening period.

Molecular identification of the EPN
The nucleotide sequence of 793 bp ITS rDNA for our isolate SOz01 was deposited in the NCBI GenBank (http:// www. ncbi. nlm. nih. gov/ genba nk/) with accession number MZ675644. A BLAST search showed that the ITS rDNA sequence of SOz01 had a 100% homology to H. megidis AB698759.
The phylogenetic tree indicated a species-specific grouping (Fig. 1). SOz01 was located in the branch formed by the Korean samples in the group of H. megidis. Thus, we classified SOz01 into H. megidis.

Insecticidal effect of the EPN
In the treatment where 320 IJs were inoculated, all larvae died within 2-5 days after inoculation (Fig. 2). The overall trend of larval mortality in the treatment with 80 IJs was similar to that with 320 IJs (93.3%, Table 2). In the treatment with 20 IJs, the occurrence of larval mortality was delayed (in the second trial) or retarded (in the first and third trials) compared with the treatments with 320 and 80 IJs. The overall mortality was 86.7% (Table 2). In the treatment with five IJs, 5 out of 15 larvae tested died (Table 2, Fig. 2).
All of the dead M. alternatus larvae in the inoculation test showed systemic red body color, except for one larva inoculated with 320 IJs in the second trial, which showed partial black immediately after its death and then turned to partial red.
We dissected a total of 47 larvae that died in the inoculation test (Table 2). They all displayed the "gummy consistency" of the body tissue (Poinar 1979) characteristic of insects killed by EPNs. In 44 among 47 larvae dissected, we detected IJs of H. megidis and fourth-stage juveniles and/or adults that had developed from the inoculated IJs (Online Resource 2). No male adults were found in the detected nematodes, and the fourth-stage juveniles did not show morphological characteristics corresponding to the male spicule. Heterorhabditis megidis has been reported to develop into hermaphrodites from IJs (Poinar et al. 1987). Therefore, we considered the adult nematodes detected in the dead larvae as hermaphrodites. Additionally, we found small juveniles, which were clearly distinguishable from IJs ( Table 2).
The total number of the IJs, fourth-stage juveniles, and hermaphrodite adults detected in a dead larva was regarded as the estimated minimum number of nematodes that successfully invaded the larva, because some of the invading IJs may have died and disappeared before dissection. The number of invading nematodes increased with the number of inoculated nematodes (Table 2), although only 9-17% of the inoculated nematodes invaded the larvae. There was no significant difference in the percentage of invading nematodes among the inoculum-density treatments with 320, 80, 20, Table 2 Mortality of Monochamus alternatus larvae after inoculation with Heterorhabditis megidis SOz01 and the nematodes detected from the insect body or the inner surface of the container a The sum of the numbers of nematodes recovered from the surface of the insect body and the inside of the container used in the test was calculated b There was no significant difference in the percentage of invading nematodes among the treatments (Kruskal-Wallis test: χ 2 = 3.85, df = 3, p = 0.28) and 5 IJs (χ 2 = 3.85, df = 3, p = 0.28). The maximum number of invading nematodes detected in the dead larvae was 43 even when the larvae were inoculated with 320 IJs (Online Resource 2). In the treatment with five IJs, we detected one-two invading nematodes in three out of five dissected larvae (Online Resource 2). In a case among them (Online Resource 2: Code 5 IJs-21 in the second trial), we found one nematode invading the larval body and three outside the insect body (insect body surface + inside of the Petri dish), indicating that the number of invading nematode was two at most.

Discussion
Heterorhabditis megidis SOz01 had a potent insecticidal effect on M. alternatus larvae. Inoculation with more than 20 IJs resulted in 86-100% larval mortality (  Fig. 2). We recovered four nematodes from one of the dead M. alternatus larvae inoculated with five IJs: one successfully invaded the larval body, and three were found on the larval body surface and inside the Petri dish (Online Resource 2: Code 5 IJs-21). Although we did not find the remaining one IJ of the five inoculated nematodes, we concluded that one or possibly two IJs of SOz01 could cause larval mortality. The detection of the various stages of nematodes (i.e., fourth-stage juveniles, hermaphrodites, and next-generation small juveniles, as well as the inoculated IJs) indicates that the inoculated IJs could propagate in the insect body within 1 week or so, as most of the host larvae died within 3-4 days after inoculation (Online Resource 2) and they were dissected within 3-4 days after larval death. No nematodes were detected in three dead M. alternatus larvae that were inoculated with 20 or 5 IJs and showed the symptom of H. megidis infection. The inoculated IJs would have killed the larvae by their symbiotic bacteria but failed to develop in the larval body, and hence could not be detected.
Heterorhabditis megidis is a hermaphrodite (Strauch et al. 1994;Nguyen and Smart 1998), which can propagate through self-fertilization. This may be advantageous as a biocontrol agent. For example, S. carpocapsae, as an amphimixis, needs both sexes to propagate and may not be able to reproduce because of a bias in the sex ratio when a small number of IJs invade an insect. In fact, Yu et al. (2016) showed that none of the IJs emerged from M. alternatus larvae inoculated with five IJs of S. carpocapsae, whereas a mean number of 60,000 IJs was obtained from the larvae inoculated with 80 IJs. In the case of H. bacteriophora, the number of IJs obtained from the dead larvae of the rosaceae longhorned beetle, Osphranteria coerulescens, inoculated with 5 and 25 IJs was four-to-seven times greater than that when S. carpocapsae was inoculated (Sharifi et al. 2014). In the case of SOz01, we observed that more than 40,000 IJs emerged from M. alternatus larvae inoculated with five IJs of H. megidis SOz01 (unpublished data).
In conclusion, SOz01 has two advantages in controlling the larvae of M. alternatus, i.e., the strong insecticidal effect on them and the high reproduction potential based on a hermaphrodite.
Data availability Data are available up on request with the corresponding author.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.