Natural association of the entomopathogenic nematode Heterorhabditis indica (Rhabditida: Heterorhabditidae) from the Philippines with the non-symbiotic bacterium Ochrobactrum anthropi (Proteobacteria: Brucellaceae)

Entomopathogenic nematodes of the genera Heterorhabditis and Steinernema are widely used as biocontrol agents against diverse insect pests. Their association with symbiont bacteria is found to be the primary cause of insect mortality. The present study reported a comprehensive characterization of a local Philippine EPN isolate using congruent morpho-taxometrical and molecular data and the associated bacterium in the infective dauer juveniles. Using an insect-baiting technique, a nematode isolate labeled F2H was recovered from the sandy soils in the coastal area of southern Philippines. Based on morphology and morpho-taxometrics, the F2H isolate collected from Iligan City was identified as Heterorhabditis indica. Analysis of the ITS region of rDNA revealed the highest sequence match to H. indica. The sequences of the bacterial 16S rDNA gene showed that the symbiont of H. indica presented 100% similitude with the sequences of Ochrobactrum anthropi. Further, pairwise alignment and phylogenetic analysis demonstrated that O. anthropi, O. lupini and O. cytisi shared 100% similarity and form a monophyletic clade. These strains have not presented any differences in the phylogenetic and 16S rDNA data, and O. lupini and O. cytisi should be considered a later heterotypic synonym of O. anthropi. The description of H. indica-O. anthropi association was the first report in the Philippines. This provides additional account and collection of naturally isolated EPN-bacteria, contributing to limited knowledge on its diversity in the Philippines, which can be utilized in the biocontrol of pests after further comprehensive assessments.

Page 2 of 11 Leonar et al. Egyptian Journal of Biological Pest Control (2022) 32:83 Background The increasing need to reduce chemical pesticide use has driven the interest in utilizing entomopathogenic nematodes (EPNs) as biological control agents (BCA) against economic agricultural pests (Georgis et al. 2006). Unlike chemicals, EPNs are target-specific with a wide range, ecofriendly, and most notably, user safe (Ehlers 2003). Their effectiveness and success as BCAs depend on their adaptability to the environment. Local species or strains are of particular interest due to better adaptation to abiotic factors, local climate and other regulators (Campos-Herrera et al. 2012). The most successful EPN strains belong to the family of Heterorhabditidae. The genus Heterorhabditis (Nematoda: Rhabditida) has a worldwide natural occurrence with currently 19 valid species throughout the globe (Machado et al. 2021).
The two Heterorhabditis species, H. bacteriophora and H. indica, are suggested to have a global distribution, especially in the tropics and subtropics (Bhat et al. 2020). Heterorhabditis indica was first isolated from southern India as a type strain (Poinar et al. 1992) and later on in many other parts of the world with the most recent record in Fiji Islands (Kour et al. 2020) and in eastern Australia (Aryal et al. 2022). In the Philippines, new populations of H. indica from banana, rice (Navarez et al. 2021), peanut (Dichusa et al. 2021) and sugarcane fields (Ramos et al. 2022) were recently documented from Davao del Sur and Cotabato provinces in Mindanao island.
Morphological discrimination between EPN species starts with differences in cadaver coloration to morphometrics. However, Heterorhabditis have been shown to exhibit slight morphological differences between species (Dolinski et al. 2008), making species identification extremely difficult without the aid of a molecular technique (Darissa and Iraki 2014). Molecular characterization was done by the analysis of internal transcribed spacer (ITS) rDNA sequences, among many others (Nasmith et al. 1996). This gene is considered as informative marker for species identification, barcoding and phylogeographical studies of Heterorhabditis species (Dhakal et al. 2020). Even so, false assignment of sequences to a taxon can occur, creating risks in the genetic database. Thus, for an increased resolution and reliability, combining morphological, morpho-taxometrical, molecular and phylogenetic studies are suggested (Spiridonov 2017).
The EPNs form a mutualistic complex with specific symbiotic bacteria, of which the latter is the primary cause of insect mortality (Ciche 2007). They depend on the EPNs for penetration into insect hosts through natural openings while bacterial cells are carried in the infective juveniles' (IJs) gut. Once inside the host, they are released into the hemolymph, producing virulence factors like the toxin complex a (Tca) among many others produced by Photorhabdus (Proteobacteria: Morganellaceae) symbionts, killing the insect by septicemia within 48 h. (Waterfield et al. 2009). The bacteria then degrade host tissues, supplying nutrients for the growth and development of the nematode population (Hazir et al. 2003). It is known that symbiotic bacteria also produce compounds to antagonize competitors like other bacterial species, fungi, and scavengers that may disturb the nematode development (Waterfield et al. 2009). However, several non-symbiotic bacteria have also been isolated and reported in association with EPN species, for instance, Ochrobactrum anthropi, O. intermedium, and Pseudomonas protegens (Ogier et al. 2020). It was reported recently that the presence of non-symbiotic bacteria is likely to reduce the recycling ability of H. indica in other insect larvae (Upadhyay and Mohan 2021). Despite this, their functions, specific requirements, and relationships are by far still unclear.
Vast natural genetic diversity in EPN populations across the country is yet to be explored and exploited. The collection of local EPNs may bring higher infectivity to target insect pests as a result of better adaptation. In this study, characterizing a local EPN strain collected from Iligan City, as part of the leading agricultural producer in the second largest island in the Philippines, Northern Mindanao was aimed. By employing complementary morphological, morpho-taxometrical, and molecular approaches, a more comprehensive characterization of our local H. indica isolate was conducted, providing an additional account to existing knowledge of EPNs in the country. Furthermore, while symbionts have biocontrol potential, some non-symbiotic bacteria such as Ochrobactrum (Proteobacteria: Brucellaceae) were reported to be opportunistic and pose high risks to humans' health (Kettaneh et al. 2003). Hence prior to virulence assays and mass production, the non-symbiotic bacteria associated with our EPN species was also identified using the 16S rDNA region (Kämpfer et al. 2003). The outcomes of this study will be pivotal for the future establishment of biocontrol programs in the country.

Nematode and bacterial isolates
Heterorhabditis indica F2H was isolated from a sandy soil sample collected from a coastal area in Iligan City, Mindanao, Philippines at 15 cm depth (N 8°12′ 5.10″) in January 2019. Soil samples were moistened and placed in plastic containers, with the subsequent addition of five super worm larvae, Zophobas morio (Coleoptera: Tenebrionidae) in each container (Bedding and Akhurst 1975). Containers with the soil and insect baits were incubated in the dark at 25 ± 2 °C. Regular checking of the samples to obtain infected cadavers was conducted every 2-3 days until required. A modified White trap method was used to isolate IJs from cadavers. Dextrose saline solutions (NSS, 0.9% NaCl) were poured into the Petri plate (5 cm diam.) and sealed with parafilm. The pure culture was obtained by re-inoculating EPNs thrice in sand assays and isolating fresh nematodes a new with modified white traps after every infection (Hoy et al. 2008). This monoculture was used for characterization. For the bacterial isolate, Ochrobactrum anthropi F2B was extracted directly from a total of five clean monocultured IJs of H. indica F2H, following the description of Aryal et al. (2022). Briefly, IJs were surface sterilized by soaking in 2% NaOCl for 5 min to avoid secondary contaminations.

Morphological characterization of the nematode
For morpho-taxometrical characterization, IJs were collected directly from the White trap, while adults were obtained through the dissection of cadavers 3-6 days after reinfection. For mounting and fixation of IJs and adults, a modified method originally described by Seinhorst (1959) was followed. A stepwise transfer of heatkilled IJs, hermaphrodite, male and female nematodes to anhydrous glycerol were done. Three following solutions were prepared: Solution 1 (containing 50:50 ratio of formalin (4%) and glycerin), Solution 2 (containing 50:50 ratio of 96% ethanol and 4% formalin) and lastly, Solution 3 (containing pure glycerin). Nematodes were then picked and mounted in clean glass slides with wax rings within a drop of pure glycerin. A maximum of five nematodes were placed on each permanent slide. A Carl Zeiss inverted routine microscope (ZEISS AxioVert. A1) equipped with a camera and AxioVision imaging software were used for studying morphology and measuring the key characters. The isolates were classified using taxonomic keys from Nguyen and Hunt (2007) and Machado et al. (2021).

Molecular characterization of the nematode and bacterium
DNA extraction was performed following the method described by Spiridonov et al. (2004) and using a tissue genomic DNA extraction kit (Dongsheng, Biotech). Briefly, the sterilized IJs were crushed by a sterile scalpel in a 1.5-ml microcentrifuge tube containing buffer, followed by adding 20 µl of Proteinase K. This was incubated at 55 °C until a homogenous solution was obtained. Different solutions provided within the kit were added following the manufacturer's protocol after each brief vortexing and centrifugation process. After elution, the obtained purified DNA was then kept at − 20 °C in a deep freezer. A fragment of rDNA containing the ITS rDNA was amplified using the primers TW81 (5'-GTT TCC GTA GGT GAA CCT GC-3') as a forward primer and AB28 (5'-ATA TGC TTA AGT TCA GCG GGT-3') as a reverse primer (Joyce et al., 1994). The thermal cycler program for EPNs was as follows: 1 cycle of 94 °C for 7 min, followed by 35 cycles of amplification at 94 °C for 60 s, 54 °C for 60 s and 72 °C for 60 s and a final extension at 72 °C for 10 min. For the bacterium, the universal primers used to amplify 16S rDNA genes were 27F (5'-AGA GTT TGA TCC TGG CTC AG-3') as the forward primer and 1492R (5′-GGT TAC CTT GTT ACG ACT T-3′) as the reverse primer (Frank et al. 2008). The PCR was performed with the following conditions: 1 cycle at 94 °C for 2 min, followed by 35 cycles at 94 °C for 45 s, 55 °C for 60 s, 72 °C for 60 s and a final extension at 72 °C for 10 min. The amplified products were Sanger sequenced by Macrogen, Inc. (Seoul, South Korea). The sequencing results of ITS and 16S rDNA of nematodes and bacteria were submitted in the GenBank under accession numbers, MT023075 and MT012075, respectively.

Phylogenetic analyses of nematode and bacterium
The generated sequences were first edited using Chromas 2.6.6. A nucleotide basic local alignment search tool (BLASTn) was performed in order to compare to all sequences deposited in the National Centre for Biotechnology (NCBI) (Altschul et al. 1997). A cut of ≥ 97% identity was considered for the same nematode species ) and ≥ 99% identity for bacterial sequence (Wang et al. 2015). ITS and 16S rDNA sequences were aligned with their closest BLAST search matches, using ClustalW multiple alignment. Poorly aligned regions were removed from the alignments using MEGA-7. The base substitution model was evaluated using jModeltest 0.1.1 (Posada 2008). Phylogenetic trees were elaborated using the Bayesian inference method as implemented in the program MrBayes 3.2.7 (Ronquist et al. 2012). The HKY + (gamma distribution of rate variation with a proportion of invariable sites) model was selected. The selected model was initiated with a random starting tree and ran with the Markov chain Monte Carlo (MCMC) for 1 × 10 7 generations. The Bayesian tree was

Morphological characterization of the nematode
Key morphological characters were identified according to Nguyen and Hunt (Nguyen and Hunt 2007 Table S1). The sequencing of amplified product of 16S rDNA resulted in the production of a partial sequence of 1220 bp which is the size expected for 16S rDNAs. Using the BLASTn algorithm, the 16S rDNA gene of the present bacterial strain was identified as the most closely related to Ochrobactrum sequences available in the GenBank database. The highest levels of identity were obtained with O. anthropi: the 16S rDNA sequence of isolate F2B had 100% identity with O. anthropi reported  Table S2).

Phylogenetic analysis of the nematode and bacterium
Phylogenetic analyses performed based on ITS1-5.8S-ITS2 rDNA showed that the present Heterorhabditis isolate (F2H) formed a monophyletic clade with Heterorhabditis indica reported from different countries of the world, thus confirming its identification. This monophyletic clade formed a group with closely related species, H. noenieputensis (Fig. 1) and together formed a group with other known members of Indica-clade. Members of Indica-clade formed a sister clade with described species of Bacteriophora-and Megidis-clade (Fig. 1). Thus, based on the phylogeny of ITS rDNA sequences, the F2H Heterorhabditis isolate from the Philippines was conspecific to H. indica. Based on the 16S rDNA gene sequences, the bacterial isolate F2B was closely related to Ochrobactrum anthropi and shares 100% sequence similarity. Phylogenetic relationship reconstructions confirmed this observation and suggested that the bacterial isolate F2B belongs to the O. anthropi species (Fig. 2)

Discussion
Identification of these microscopic nematodes is often described to be rather difficult and requires meticulous efforts. At low taxonomic levels (i.e., order, family,  and genus level), systematics was reported to be unstable. This is mainly due to their scantiness of observable and informative characters as well as overlapping morphometrics (Holterman 2008). H. indica IJs of this study were separated from other Heterorhabditis species by their smaller body length (< 600 μm). The rest of the  et al. Egyptian Journal of Biological Pest Control (2022) 32:83 characters are within the range of the original descriptions of Poinar et al. (1992). Males were slightly small at the type strain but comparable to the H. indica H.TN48 populations in Vietnam (Phan et al. 2003). However, Heterorhabditis exhibited slight morphological differences between species (Dolinski et al. 2008), making species identification extremely difficult (Darissa and Iraki 2014). With the advent of molecular diagnostic tools, morphological data can be complemented with modern techniques (Spiridonov 2017). In the present study, ITS rDNA region was used to identify and confirm the collected Heterorhabditis species.
Of particular highlight in this study was the bacterium detected in the IJs gut. These IJs initiate the parasiticpathogenic life cycle of any EPNs as they carry bacteria in their guts or vesicle compartments while seeking their target insect hosts (Hazir et al., 2003). Thus, the natural bacteria present were determined in the IJs itself as it was rarely investigated in EPNs. Any phenotypic characterization, physiological, or biochemical profiling were not conducted to augment molecular analysis. Nonetheless, the conventionally used 16S rDNA gene analysis, which is the most used universal marker for identification of prokaryotes (Kämpfer et al. 2003) was used. For the first time, Ochrobactrum anthropi was reported as a  et al. Egyptian Journal of Biological Pest Control (2022) 32:83 non-symbiotic bacterium naturally associated with the IJs of H. indica F2H from the Philippines. Previous reports relating to existence of non-EPN bacterial symbionts are commonly recognized either as laboratory contaminants (Poinar 1966), symbiont's inability to antagonize other species (Jackson et al. 1995), symbiont's phase variations (Razia et al. 2011), or ephemeral associations (Gouge and Snyder 2006). Some findings on the natural occurrence of dixenic associations between the Ochrobactrum and Photorhabdus in H. indica were reported specifically inside the infected insect cadaver (Razia et al. 2011). For this reason, the previous idea of general contamination was dismissed as the identified Ochrobactrum bacteria were fresh harvests without laboratory transfer (Babić et al. 2000) and the method of bacterial isolation entailed surface disinfection on the nematodes (Fischer-Le Saux et al. 1998 Associated bacteria are usually isolated and identified from the insect host hemolymph and seldom from the IJ. With the traditional culture-dependent approach, several other bacterial associates from nematodes remain cryptic (Gouge and Snyder 2006). To investigate its community dynamics, a heterorhabditid-infected insect cadaver was profiled for bacterial communities using a culture-independent next-generation sequencing tool. Some studies have been also carried out in the model organism, C. elegans (Dirksen et al. 2016). This recent approach for EPN-microbe interactions revealed a gradual shift from Photorhabdus-dominant community toward one shared with non-symbiotic bacteria in the insect environment (Wollenberg et al. 2016). In order to elucidate findings, Ogier et al. (2020) employed a modern and systematic approach to directly examine for the first time the IJ microbiota of the Steinernema carpocapsae by metabarcoding. The authors found that instead of the standard monoxenic model of the EPN-bacteria complex, it proposes the pathobiome view. Ochrobactrum sp. was among many other bacteria described from Steinernema species. They further reported O. anthropi from H. bacteriophora microbiota, which validates our recent findings on the O. anthropi association with the IJs of H. indica. Although the traditional method was used, nevertheless it was complemented. As a future prospect, the IJ microbiota of the tested H. indica tropical isolates will be investigated.
Interestingly, only Ochrobactrum was identified in this study and neither symbiotic bacterium nor dixenic associations was detected inside the IJs gut. Directly, the bacterial DNA inside the IJ gut was extracted and then identified using the 16S rDNA. The lone detection could be attributed to the methods used or for some other reasons. For example, the possibility of the insensitivity and tolerance of this Ochrobactrum strain against antimicrobials produced by the symbiont (Jackson et al. 1995) or that the antimicrobial activity of the non-symbiotic species may have affected the primary symbiont and contributed to interspecies competition (Ogier et al. 2020). For this, it was assumed that this O. anthropi F2B nonsymbiont may have dominated and decimated the symbiotic bacteria, or that they persisted well inside the IJ gut. This kind of sustainable association of bacteria other than the symbiont was demonstrated in the Steinernema-Xenorhabdus model, validating the persistence of nonsymbionts like Ochrobactrum (Ogier et al. 2020). In a similar study, Ochrobactrum sp. was found persisting in the nematode gut of an EPN, Oscheius chongmingensis where it was also the most abundant bacterial genus at 59.82%, followed by Bacillus at 7.13% and others (Fu and Liu 2019). Ochrobactrum was further reported to be associated with the microbiome of other nematode species like the free-living cephalob nematode, Acrobeloides maximus (Baquiran et al. 2013), and in the model nematode C. elegans under stressful conditions (Dirksen et al. 2016), even indicated to coexist with several bacteria in nature. In general, Ochrobactrum is abundant in the soil, in the insect cadaver and in the IJ microbiota. As EPNs belong to the group of Rhabditida that feeds on microorganisms, they likely have retained this microbivorous behavior (Boemare et al. 1996), allowing non-symbionts to multiply in the nematode gut. Whether this bacterium is likewise maternally transmitted similar to a previously known elaborate pathway of Heterorhabditis symbionts (Ciche et al. 2008) is subject to further investigations.
The concern further is raised to health hazards. Although recently found not to be virulent on insects, the genus was earlier documented to be opportunistic pathogens to immune-compromised humans such as O. anthropi (Kettaneh et al. 2003). Nematodes then can be a reservoir of this opportunistic bacterium. As for the non-symbionts like Ochrobactrum, which can occur in laboratory culture, Boemare et al. (1996) emphasized that their association was not considered a risk because they cannot support long-term mass rearing of nematode populations. However, based on the documented persistence presented, risk cannot be ruled out. Thus, caution must always be practiced in the future outlook of applying EPN technology to the local agricultural industries for the farmers' benefit. When applied in the field, several