A study on Xenorhabdus and Photorhabdus isolates from Northeastern Thailand: Identification, antibacterial activity, and association with entomopathogenic nematode hosts

Xenorhabdus and Photorhabdus are gram negative bacteria that can produce several secondary metabolites, including antimicrobial compounds. They have a symbiotic association with entomopathogenic nematodes (EPNs). The aim of this study was to isolate and identify Xenorhabdus and Photorhabdus species and their associated nematode symbionts from Northeastern region of Thailand. We also evaluated the antibacterial activity of these symbiotic bacteria. The recovery rate of EPNs was 7.82% (113/1445). A total of 62 Xenorhabdus and 51 Photorhabdus strains were isolated from the EPNs. Based on recA sequencing and phylogeny, Xenorhabdus isolates were identified as X. stockiae (n = 60), X. indica (n = 1) and X. eapokensis (n = 1). Photorhabdus isolates were identified as P. luminescens subsp. akhurstii (n = 29), P. luminescens subsp. hainanensis (n = 18), P. luminescens subsp. laumondii (n = 2), and P. asymbiotica subsp. australis (n = 2). The EPNs based on 28S rDNA and internal transcribed spacer (ITS) analysis were identified as Steinernema surkhetense (n = 35), S. sangi (n = 1), unidentified Steinernema (n = 1), Heterorhabditis indica (n = 39), H. baujardi (n = 1), and Heterorhabditis sp. SGmg3 (n = 3). Antibacterial activity showed that X. stockiae (bMSK7.5_TH) extract inhibited several antibiotic-resistant bacterial strains. To the best of our knowledge, this is the first report on mutualistic association between P. luminescens subsp. laumondii and Heterorhabditis sp. SGmg3. This study could act as a platform for future studies focusing on the discovery of novel antimicrobial compounds from these bacterial isolates.


Introduction
Xenorhabdus and Photorhabdus are motile, gram-negative rods, facultative anaerobes, nonsporeforming, oxidase-negative, and chemoorganotrophic heterotrophs with respiratory and fermentative metabolism. These bacteria symbiotically inhabit the intestine of the infective juvenile (IJ) stage of entomopathogenic nematodes (EPNs) belonging to the Steinernematidae and Heterorhabditidae families [1]. The IJs of EPNs enter the digestive tract of the insect larvae, penetrate the hemocoel of the insect host, and release the bacteria into the hemolymph. Together, the IJs and bacteria rapidly kill the insect larvae within 24-48 h [2]. Otherwise, the nematodes or bacteria themselves make significant contributions to pathogenesis within the insect [3][4][5].
The objectives of this study to isolate and identify EPNs and their symbiotic bacteria Xenorhabdus and Photorhabdus from Northeastern Thailand; we also analyzed their phylogenetic diversity. The antibacterial activity of the extracts of the identified Xenorhabdus and Photorhabdus strains against antibiotic-resistant bacteria was also evaluated using the disk diffusion method, minimum inhibitory concentration (MIC), and minimal bactericidal concentration (MBC). This study will provide information at the molecular level that can assist in taxonomy of Xenorhabdus and Photorhabdus isolates, and their EPN hosts from Thailand. These bacteria may serve as a resource for discovery a novel bioactive compound.

Collection of soil samples
A total of 1,445 soil samples from 289 soil sites were collected from nine provinces in Northeastern Thailand. All soil sites belonged to public areas and no specific permission was required. For each soil site, five soil samples were randomly taken from an area of approximately 10 m 2 and at a depth of 10-15 cm using a spade. Approximately 500 g of each soil sample was placed in a plastic bag. Site location, latitude, longitude and altitude, soil temperature, pH, and moisture were recorded. Soil samples were maintained at 25-30˚C during transportation to the Department of Microbiology and Parasitology, Faculty of Medical Science, Naresuan University.

Isolation and identification of entomopathogenic nematodes
The IJs of EPNs were isolated by the baiting technique as previously described [17]. For each soil sample, five larvae of Galleria mellonella (greater wax moth) were placed on top of the soil sample stored in a plastic container. Subsequently, the container was covered with a lid, and it was turned upside down to let the larvae move into the soil. It was incubated in dark at 30˚C for 5 days. The dead larvae of G. mellonella were collected from the soil samples and then larval cadavers were placed on a White trap [23] to allow the IJs to emerge. The IJs were collected in a tissue culture flask, cleaned with sterile distilled water, and stored at 15˚C.

Isolation and identification of symbiotic bacteria
The infected dead larvae of the greater wax moth were surfaced sterilized with 95% ethanol before dissection. The hemolymph was collected by a sterile loop, and then streaked onto a nutrient bromothymol blue agar (NBTA). The plate was incubated at 28˚C in dark for 4 days. The bacterial isolates (blue or green colonies) were selected and stored in LB broth containing 50% glycerol (v/v) at -80˚C.
The genomic DNA of 113 isolates of the symbiotic bacteria was extracted with the Genomic DNA Mini Kit (Blood/Cultured cell) (Geneaid Biotech Ltd., Taiwan) according to the manufacturer's instructions. A partial sequence of recA gene was amplified from the genomic DNA by PCR using forward and reverse primers (5'-GCTATTGATGAAAATAAACA-3' and 5'-RATTTTRTCWCCRTTRTAGCT-3') to obtain an 890 bp amplicon (24). The PCR mixture (50μL) consisted of 10 μL of 5X buffer, 7 μL of 25 mM MgCl 2 , 1 μL of 200 mM dNTPs, 2 μL of 5 mM of each primer, 0.5 μL of 5 unit Taq polymerase (Sigma, USA), 2.5 μL of DNA template, and 25 μL distilled water. PCR cycling parameters for the recA gene of Xenorhabdus were as follows: an initial denaturing step of 94˚C for 5 min, followed by 30 cycles of denaturation at 94˚C for 1 min, annealing at 50˚C for 1 min, and extension at 72˚C for 2 min, and a final extension at 72˚C for 7 min. PCR parameters for Photorhabdus were as follows: an initial denature step of 94˚C for 5 min, followed by 30 cycles of denaturation at 94˚C for 1 min, annealing temperature of 50˚C for 45 s and extension of 72˚C for 1.5 min, and a final extension at 72˚C for 7 min. The PCR products were checked on 1.2% agarose gel by electrophoresis and purified using the Gel/PCR DNA Fragment Extraction Kit (Geneaid Biotech Ltd., Taiwan).

PCR for 16S rDNA, gyrB, dnaN gltX, and infB
recA analysis revealed that one Xenorhabdus (KK9.1_TH) isolate had lower than 96% similarity in the BLASTN search; this isolate was selected for further analysis, and sequencing of its additional nucleotide regions, including 16S rDNA, gyrB, dnaN, gltX, and infB, was performed.

Sequence and phylogenetic analysis
The sequencing of the PCR products was done at Macrogen Inc. Service (Korea) (http://www. macrogen.com). The nucleotide sequences were edited and merged with the SeqMan TM II software (DNASTAR Inc., Wisconsin, USA). The recA sequences of the bacteria from the present study were deposited in the NCBI database under the Genbank accession numbers KY809276 to KY809337, MT160765 to MT160768, and MT158222 for Xenorhabdus spp., and KY809338 to KY809388 for Photorhabdus spp. The 28S rDNA sequences of Steinernema isolates were deposited in the NCBI database under the Genbank accession numbers KY809389 to KY809425, and the ITS sequences of the Heterorhabditis isolates were deposited in NCBI database under the Genbank accession numbers KY809426 to KY809468.
The consensus sequences of each species were used for multiple sequence alignment using Clustal W [26] in the MEGA software version 6.0 [27]. Species identification was performed using BLASTN (http://blast.ncbi.nlm.nih.gov/Blast.cgi). Similarity � 97% was considered as the same species. The known nucleotide sequences of EPNs and their symbiotic bacteria in the NCBI database were downloaded and used as the reference species. For EPNs, maximum likelihood (ML) trees of the entire gene (28S rDNA, and ITS) were constructed based on Tamura 3-parameter with 1,000 bootstrap replicates model using MEGA 6.0 software [27]. For symbiotic bacteria, maximum likelihood (ML) trees of the entire gene (16S rDNA, recA, gyrB, dnaN gltX, and infB) and the concatenation of truncated sequences of recA, gyrB, dnaN gltX, and infB were constructed based on Tamura 3-parameter model using MEGA 6.0 software [27]. Also neighbor-joining trees (NJ) were constructed based on a Kimura 2-parameter with 1,000 bootstrap replicates using MEGA 6.0 software [27]. Bayesian analysis was performed based on Markov chain Monte Carlo method in MrBayes v3.2 [28].

Preparation of antibiotic-resistant bacteria
Fifteen strains of antibiotic-resistant bacteria, including Acinetobacter baumannii (four clinical strains), Escherichia coli (two clinical strains), E. coli ATCC35218, Klebsiella pneumoniae (two clinical strains), K. pneumoniae ATCC700603, Enterococcus faecalis ATCC51299, Pseudomonas aeruginosa, Staphylococcus aureus (two clinical strains), and S. aureus ATCC20475, were used as the pathogens for testing the antibacterial activity of the extracts of the symbiotic bacteria. These bacteria were streaked on Mueller-Hinton agar (MHA) and incubated at 37˚C for 24 h. A single colony was resuspended in 0.85% sodium chloride (NaCl), and the turbidity was adjusted to 0.5 McFarland standards. Then, 100 μL of the bacterial suspension was swabbed on MHA plate for disk diffusion test [29].

Screening of Xenorhabdus and Photorhabdus isolates
Xenorhabdus and Photorhabdus isolates were cultured on NBTA at 28˚C in dark for four days. A single colony from each isolate was transferred to a 15-ml tube containing 5 mL of LB broth and incubated at room temperature for 48 h under shaking conditions. A paper disk (6 mm with diameter) with a 20 μL drop of the whole cell culture was placed on MHA plated with antibiotic-resistant bacteria. The plates were placed in an incubator at 37˚C for 24 h. The inhibition zone (clear zone) was checked and measured (millimeter). The most effective isolates of Xenorhabdus and Photorhabdus were selected for crude compound extraction.

Bacterial extracts
A single colony of Xenorhabdus and Photorhabdus on NBTA medium was inoculated to 1000 mL flask containing 500 mL of LB. The culture flask was shaken at 180 rpm for 72 h. The bacteria cultured was added with 1000 mL ethyl acetate and mixed well. All solvents were removed from bacterial extracts by a rotary vacuum evaporator (Buchi, Flawil, Switzerland). Dimethyl sulfoxide (DMSO) was added to bacterial extracts to make a final concentration of 500 mg/mL and stored at -20˚C until used.

Disk diffusion method
Cultured drug resistant bacteria were spread on MHA agar. A sterile 6 mm disc was put onto MHA agar plate and then 10 μL of each bacterial extract was dropped onto a sterile disc. Negative control was DMSO and Positive control was antibiotic disks. The plates were incubated at 37˚C for 24 h. The inhibition zone was measured in millimeter. The most effective results of bacterial extracts were further evaluated by MIC and MBC.

MIC and MBC assays
Bacterial extracts were diluted in two-fold serial dilutions in a 96-well micro titer plate. The suspension of drug resistant bacteria (1 × 10 8 cell/ml) was added into each well and mixed well. Cultured drug resistant bacteria, cultured drug resistant bacteria mixed with DMSO, and sterile Mueller-Hinton (MH) broth were used as controls. Plates were incubated at 37˚C for 24 h. No visible growth of drug resistant bacteria in the well was considered as MIC. In addition, 10 μL from each well from the MIC assay was dropped onto MHA plates. The plates were then incubated at 37˚C for 24 h. The lowest concentration of bacterial extract without growth of drug resistant bacteria was considered as MBC.

Isolation of EPNs
A total of 1,445 soil samples from 289 sites were collected from the Northeastern region of Thailand, including Kalasin [30], Khon Kaen, Chaiyaphum, Nakhon Ratchasima, Maha Sarakham, Loei, Nong Khai, Nhong Bua Lamphu, and Udon Thani provinces. The recovery rate of EPNs was 7.82% (113/1,445) of the total soil samples collected. We isolated 62 strains belonging to Xenorhabdus spp. and 51 strains belonging to Photorhabdus spp. from the EPNs ( Table 1). Most of the soil samples were positive with only one of the two genera of EPNs (Steinernema spp. and Heterorhabditis spp.). In contrast, few soil samples (two samples from Maha Sarakham province and one sample from Nong Khai province) were positive with both Steinernema and Heterorhabditis. Most of the EPNs were isolated from loam, and the mean pH, temperature, and moisture of the soil samples were 6.6, 28.4˚C, and 1.5%, respectively ( Table 2). These soil parameters were not significantly different between soil samples with and without EPNs (Mann-Whitney test).
The ML tree derived from all the sequences of recA among the Thai Xenorhabdus isolates and reference strains from GenBank database is shown in Fig 1. The Thai Xenorhabdus isolates were distributed in three groups. Group 1 was the majority group (60 isolates), which was closely related to X. stockiae. Group 2 contained one isolate (bKK26.2_TH), which was closely related to X. indica. Group 3 also contained one isolate (bKK9.1_TH), which fell in the clade of X. thuongxuanensis, X. ishibashii, and X. eapokensis. The ML tree derived from 16S rRNA, gyrB, dnaN, gltX, and infB genes are shown in S1-S5 Figs. The ML tree of concatenation of the five truncated genes (recA, gyrB, dnaN, gltX, and infB) is shown in Fig 2. All phylogenetic trees supported that Xenorhabdus bKK9.1_TH was closely related to X. eapokensis.
Fifty-one isolates of Photorhabdus were identified using BLASTN search of partial sequences of the recA gene. Twenty-nine isolates were identified as P. luminescens subsp. akhurstii (97-100% identity) and 18 isolates were identified as P. luminescens subsp. hainanensis (98-100% identity). Two isolates were identified as P. asymbiotica subsp. australis (99-100% identity). The remaining two Photorhabdus isolates were identified as P. luminescens subsp. laumondii (98% identity). ML analysis of 51 recA sequences of Photorhabdus distributed the isolates in three groups. Group 1 contained 47 isolates closely related to P. luminescens subsp. akhurstii and P. luminescens subsp. Group 2 contained two Photorhabdus isolates, which were closely related to P. luminescens subsp. laumondii. Group 3 contained the remaining two isolates of Photorhabdus, which were most closely related to P. asymbiotica subsp. australis (Fig 3).

Identification and phylogenetic tree of entomopathogenic nematodes
A total of 113 EPNs were isolated from the soil samples. EPNs (80 isolates) were identified using BLASTN search of a partial sequence of 28S rDNA for Steinernema and internal transcribed spacer for Heterorhabditis. The remaining 33 isolates of EPNs were lost due fungal contamination. Thirty-seven isolates were identified as Steinernema and the remaining 43 isolates were identified as Heterorhabditis. Steinernema (35 isolates) were identified as S. surkhetense (97-99% identity). One isolate was identified as S. sangi with 98% similarity. Species of the remaining one isolate Steinernema eKK26.2_TH was unidentified due to its low identity with S. abbasi (90%). The phylogenetic relationships among the Steinernema isolates and reference strains from GenBank database are shown in Fig 4. The ML analysis of the 37 sequences distributed them into three groups. Group 1 contained 35 isolates of Steinernema, which were closely related to S. surkhetense and S. anatoliense. Group 2 contained only one isolate, which was closely related to S. abbasi. Group 3 contained one isolate of Steinernema, which was closely related to S. sangi.
For Heterorhabditis nematodes, 39 isolates were identified as H. indica (98-100% identity), one isolate was identified as H. baujardi (99% identity), and three isolates were identified as Heterorhabditis sp. SGmg3 (97-99% identity). ML analysis of the 43 sequences of Heterorhabditis distributed them into three groups (Fig 5). Group 1 was the majority group (39 isolates), which was closely related to the clade of H. indica. Group 2 contained only one isolate, which was closely related to H. baujardi. Group 3 contained three isolates, which were closely related to Heterorhabditis sp. SGmg3.
Maximum association was observed between X. stockiae and the nematode host S. surkhetense (35 isolates). A single isolate of X. indica was associated with Steinernema sp., and one isolate of Xenorhabdus sp. (bKK9.1_TH), closely related to X. eapokensis, was associated with S. sangi. In addition, 39 isolates of P. luminescens were associated with H. indica. A single isolate of P. luminescens subsp. akhurstii was associated with H. baujardi. Two isolates of P. asymbiotica subsp. australis were associated with H. indica. A single isolate of P. luminescens subsp. luamondii was associated with Heterorhabditis sp. SGmg3.

Antibacterial activity
We found that whole cell extracts of four (X. stockiae, n = 3 and X. indica, n = 1) out of 113 isolates could inhibit the growth of at least one antibiotic-resistant bacterial strain. The extract from these bacterial isolates was tested against the antibiotic-resistant bacteria by disk diffusion method. Two isolates of X. stockiae (bMSK7.5_TH and bKS8.5_TH) and one isolate of X. indica (bKK26.2_TH) showed potential inhibition of the growth of the antibiotic-resistant bacteria (Table 3)  (ESBL and CRE). X. stockiae (bKS8.5_TH) and X. indica (bKK26.2_TH) could inhibit S. aureus strain PB36 (MRSA). However, X. stockiae (bUDT18.2_TH) was unable to inhibit any antibiotic-resistant bacteria by the disk diffusion method.

Discussion
The overall recovery rate of the EPNs (Steinernema and Heterorhabditis) from soil samples of Northeastern region of Thailand was 7.82%. This result was similar to that reported by Brodie [31] from Fiji Islands (7.3%), Valadas [32] from Portugal (6.7%), and Hatting [33] from South Africa (5%). However, this rate was higher than those reported by Caoili [34] from the Philippines (2.5%), Majić [35] from Croatia (2.0%), and Noujeim [36] from Lebanon (1%). Higher prevalence of EPNs in soil from that determined in the present study was observed by Kanga [37] from Southern Cameroon (10.4%), Khatri-Chhetri [38] in Nepal (10.5%), and Malan [39] in South Africa (17%). This suggests that global prevalence of EPNs is variable. Distribution of Steinernema and Heterorhabditis has been reported from several ecological niches in USA, Australia, Europe, and Asia, including Thailand [8, 15-18, 24, 40, 41]. Biotic and abiotic characteristics influence the distribution of the EPNs; however, in our study, soil temperature, moisture, and pH of the soil samples with and without EPNs were not significantly different. Nevertheless, our data supported previous reports from Thailand, which showed that EPNs were able to survive in a diverse soil environment and various soil types with a wide range of pH (3.2-6.9), temperature (20˚C-32˚C), and moisture (0-8%) [17,18,22,23]. Soil moisture, temperature, and rainfall affect the distribution of the insects that could be probable hosts for the EPNs [42]. This could also affect the distribution of EPNs.
Identification and phylogenetic analysis of 62 Xenorhabdus isolates revealed that X. stockiae was the predominant species. This bacterium was hosted by S. surkhetense, which has been previously described from India [43]. X. stockiae has also been isolated from S. siamkayai and S. minutum in Thailand [16,24,44,45]. It was reported as a bacterial symbiont with S. huense in Vietnam [46]. One isolate of X. indica was found to be associated with Steinernema sp. (90% similar with S. abbasi). X. indica was first reported to be associated with S. thermophilum [47]. Subsequently, the association between X. indica and S. abbasi was reported from Taiwan [48]. In a previous study, X. indica was associated with S. yirgalemense [49], and in the current study, it was found that X. indica, an Indian isolate, was associated with S. pakistanense [50]. This suggests that X. indica may be symbiotically associated with a wide range of EPN hosts.
A single isolate of Xenorhabdus (bKK9.1_TH) showed low similarity with X. thuongxuanensis (95% identity) and X. eapokensis (95% identity) by recA sequence analysis. In contrast, higher similarity of this isolate was found with X. eapokensis when 16S rDNA and infB (99% identity), and gltX and dnaN (98% identity) sequences were analyzed. In addition, multilocus sequence analysis (MLSA) based on concatenated partial gene sequences of recA, gyrB, dnaN, gltX, and infB revealed that Xenorhabdus (bKK9.1_TH) was closely related to X. eapokensis. Therefore, we identified Xenorhabdus (bKK9.1_TH) as X. eapokensis. This suggests that analysis of multiple genes may aid in the identification of this bacterium. Also, whole genome sequencing of this bacterium may assist in confirmation of its identity at the species level. We found that X. eapokensis was associated with S. sangi, which has been reported as a host for X. vietnamensis and X. thuongxuanensis [25,51].
For the genus Photorhabdus, in the current study, the following four species were identified: P. luminescens subsp. akhurstii (n = 29), P. luminescens subsp. hainanensis (18 isolates), P. luminescens subsp. luamondii (n = 2), and P. asymbiotica subsp. australis (n = 2). P. luminescens subsp. akhurstii and P. luminescens subsp. hainanensis were associated with H. indica and H. sp. SGmg3. These associations have been previously reported from Thailand [15,17,18]. In addition, P. luminescens subsp. hainanensis has also been isolated from H. baujardi in Thailand [8]; however, P. luminescens subsp. akhurstii has been found in association with H. bacteriophora in Iran, Hungary, Argentina, USA, and in association with H. indica in China [19]. To the best of our knowledge, this is the first report on mutualistic association between P. luminescens subsp. luamondii and Heterorhabditis sp. SGmg3. However, P. luminescens subsp. luamondii has also been reported to be associated with H. bacteriophora in Thailand, USA, and Argentina [17,52] and with H. safricana from South Africa [53]. P. asymbiotica subsp. australis was also found in the present study, which was in association with H. indica. This association has been found in Thailand previously [17,54]. P. asymbiotica is an emerging pathogen that has been reported to cause locally invasive soft tissue infection and disseminated bacteremia; clinical cases have been identified in both Australia and USA [55,56]. This suggests that P. asymbiotica could also cause these diseases in the residents of Thailand. Although no clinical case of P. asymbiotica infection has been reported in the country, management and healthcare strategies should be prepared in advance.
In the current study, X. stockiae (bMSK7.5_TH) extract showed the highest inhibitory effect against several antibiotic-resistant bacteria. Previous studies have shown that Xenorhabdus produces xenocoumacin [57] and amicoumacin derivatives [58], which are potent against S. aureus [59]. All Photorhabdus extracts from Mae Wong national park could inhibit S. aureus ATCC20475, S. aureus strain PB36 (MRSA), and S. aureus strain PB57 (MRSA) (8). In addition, P. luminescens subsp. akhurstii (bSBR11.1_TH) extract from Saraburi province could inhibit up to 10 antibiotic-resistant bacterial strains, and all Photorhabdus isolates showed the potential to inhibit the growth of S. aureus strain PB36 (MRSA) [41]. P. luminescens has been reported to produce isopropylstilbene [44,60], which has multiple biological activities, including antibiotic activity against E. coli, B. subtilis, S. pyogenes, and S. aureus RN4220 (drug resistant and clinical isolate) [61,62]. The bio-activity of isopropylstilbene has been extended to Caenorhabditis elegans was used as an out-group. Bootstrap values are reported out of 1000 replicates. Numbers shown above the branches are bootstrap percentages for clades supported above the 50% level. The bar indicates 5% sequence divergence.
https://doi.org/10.1371/journal.pone.0255943.g005  inhibit the growth of fungi [10]. We found that the bioactivity (MIC and MBC) of the crude extracts emphasized that the X. stockiae (bMSK7.5_TH) extracts were active against both gram-positive and gram-negative bacteria. The MIC values exhibited by all the extracts in this study ranged between 3.90-62.5 mg/mL, and the MBC ranged between 7.81-15.62 mg/ml. This may be due to the ability of each symbiotic bacterial isolate to produce different effective metabolites to kill drug resistant bacteria. This suggests that Xenorhabdus and Photorhabdus isolates are potential agents for the inhibition of the growth of MDR bacteria. Therefore, both Xenorhabdus and Photorhabdus isolates are a potential source for novel antibiotics.

Conclusions
In summary, 113 isolates of EPNs were obtained from a total of 1,445 soil samples collected from 289 sites in Northeastern region of Thailand. S. surkhetense and H. indica were the two most common EPN species found in the soil samples. For symbiotic bacteria, X. stockiae, X. indica, X. eapokensis, P. luminescens subsp. akhurstii, P. luminescens subsp. hainanensis, and P. asymbiotica subsp. australis were found in the studied area, and X. stockiae and P. luminescens subsp. akhurstii were found to be predominant. The common associations observed between EPN hosts and their symbiotic bacteria were S. surkhetense-X. stockiae and H. indica-P. luminescens. EPN host of X. eapokensis was S. sangi and that of X. indica was unidentified Steinernema. In addition, the crude extract from X. stockiae (bMSK7.5_TH) showed a broadspectrum inhibitory activity against several antibiotic-resistant bacterial strains. Thus, this study will be useful in further drug discovery from natural resources.