Isolation, identification, and pathogenicity of Steinernema carpocapsae and its bacterial symbiont in Cauca-Colombia

Soil samples were taken in the Toribío and Tacueyó reserves of the municipality of Toribío, in the Department of Cauca, Colombia, where EPNs have never been applied. These rural areas are located between 1,757 and 2,963 meters above sea level (m.a.s.l) where temperatures range between 13.9° and 20°C. Following the sampling method developed by Varón de Agudelo and Castillo (2001), 10 sampling sites in eight types of vegetation cover were selected: natural grassland (NG), pasture bordering annual crops (PBAC), natural forest (NF), first-growth American bamboo (Guadua sp.) (FGAB), coffee cultivation (Coffea arabica) (CC), horticultural cultivation (HC), strawberry cultivation (Fragaria sp.) (SC), and combined lulo (Solanum quitoense) and papaya cultivation (Carica sp.) (CLPC). At each site, two composite soil samples (10 subsamples, ~1 kg) were taken at depths between 20 and 30 cm, after removing soil cover to avoid cross contamination. Each independent composite sample was taken in an area measuring approximately 16 m² with a zigzag distribution of 10 subsampling points. Samples were packed in plastic bags, labeled, and stored in a cooler (~15°C) (Orozco et al., 2014) in dark conditions for less than 24 hr until they arrived at the Biological Control laboratory of Pontificia Universidad Javeriana in Bogotá, Colombia to be immediately processed.

Composite soil samples taken from each site were mixed and stored at room temperature for 24 hr prior to testing for IJs using the insect baiting technique (Bedding and Akhurst, 1975). For this, six samples of 150 g of mixed soil were deposited in 200 cm³ plastic boxes. Then, five G. mellonella larvae and five Tenebrio molitor (Colepotera: Tenebrionidae) larvae were added per box at the soil surface. The boxes were labeled, inverted, and kept in darkness at 20°C. Four days later, the boxes were evaluated for dead larvae killed by EPNs. Sagging bodies and color changes were noted. Subsequently, dead larvae were placed individually in White traps (White, 1927) to obtain IJs. The IJs recovered were then multiplicated in G. mellonella larvae and maintained for later identification. To corroborate whether EPNs were absent or present in soil samples, samples were processed twice by repeating the process described above.

The IJs isolated were stored in polyurethane foams at 10°C in the Biological Control laboratory and deposited in the Entomology collection of the Javeriana Museum of Natural History of the Pontificia Universidad Javeriana in Bogotá, Colombia.

Abundance of EPNs at sampled sites (positive sites for EPNs/total sites) and the recovery frequency of EPNs (positive samples for EPNs/total samples analyzed) were determined according to the type of vegetation cover (Liu and Berry, 1995) and expressed as a percentage.

Genomic DNA was extracted individually from females following the protocol described by Çimen et al. (2016). After lysis of the nematodes, protein content was separated using NaCl at a final concentration of 1.7 M and centrifuging at 3,000 g for 15 min at room temperature. Finally, the supernatant was transferred to another tube for alcohol precipitation of DNA.

Sequences of the following three taxonomic markers commonly used for nematodes were amplified by PCR and used for molecular identification: a fragment of the 18S rRNA sequence using primers SSU18A-4F: 5′-GCTTGTCTCAAAGATTAAGCCATGCATG-3′ and SSU26Rplus4: 5′-AAGACATTCTTGGCAAATGCTTTCG-3′ (Morise et al., 2012); a fragment that contained the sequences ITS1, 5.8S, and ITS2 using primers 18S: 5′-TTGATTACGTCCCTGCCCTTT-3′ and 26S: 5′-TTTCACTCGCCGTTACTAAGG-3′ (Vrain et al., 1992); and a rRNA 28S fragment that contained the D2/D3 expansion sequence using D2F: 5′-CCTTAGTAACGGCGAGTGAAA-3′ (Nguyen et al., 2006) and 536: 5′-CAGCTATCCTGAGGAAAC-3′ as primers (Stock et al., 2001).

For fragment amplification, 50 µl of PCR mixture was prepared using 1X NH₄ Buffer (Bioline, England), 3 mM MgCl₂, 0.2 mM dNTPs, and 0.5 µg/µl Bovine Serum Albumin (New England Biolabs, United States), 0.5 µM forward primer, 0.5 µM reverse primer, Taq DNA polymerase 2U (Bioline, England), and 100 to 200 ng of template DNA (Hominick et al., 1997; Stock et al., 2001).

The PCR protocol for the 18S fragment consisted of one initial denaturation cycle at 94°C for 2 min followed by 35 cycles at 94°C for 10 sec, 55°C for 30 sec, 68°C for 1 min (Morise et al., 2012), and a final extension at 68°C for 7 min. For the ITS and D2/D3 fragments, the PCR protocol consisted of one cycle of initial denaturation at 94°C for 7 min followed by 35 cycles at 94°C for 1 min, 1 min at the annealing temperature (ITS: 50°C and D2/D3: 55°C), 72°C for 1 min and a final extension at 72°C for 7 min (Çimen et al., 2016).

Amplification of all PCR products and their respective negative amplification controls were verified by 1% (w/v) agarose gel electrophoresis in 1X TBE buffer stained with 0.5 X Hydragreen (ACTGene, United States). PCR products were purified for sequencing with Wizard SV Gel and PCR clean-up system (Promega, United States). Three purified amplicons obtained from three independent DNA extractions were sequenced for each molecular marker. Sequencing of each amplicon was performed in both directions.

The six sequences obtained for each marker were visualized, aligned, and edited manually using MEGA X software (Kumar et al., 2018) in order to obtain a consensus sequence. Its identity was initially verified by means of the Basic Local Alignment Search Tool (BLAST) (Altschul et al., 1990) on the basis of the non-redundant (nr) database of the NCBI. Consensus sequences of all markers were deposited in GenBank (NCBI) under the accession numbers listed in Table 1.

The consensus sequences of all markers were subsequently aligned with sequences of various species of EPNs deposited in GenBank (Table 1) using MUSCLE (Edgar, 2004) in MEGA X software (Kumar et al., 2018). The alignment obtained was edited so that all sequences used for analysis were of the same length and same genetic region.

Then, phylogenetic trees were independently constructed for each marker. The Tamura–Nei model of maximum likelihood estimation (Tamura and Nei, 1993) with gamma distribution and 1,000 bootstraps was used for the 18S fragment, and the General Time Reversible model of maximum likelihood estimation (Nei and Kumar, 2000) with gamma distribution and 1,000 bootstraps was used for the ITS fragment. Unweighted Pair Group Method with Arithmetic Mean (UPGMA) (Sneath and Sokal, 1973) with gamma distribution and 1,000 bootstraps was used for the D2/D3 fragment of rRNA 28S.

Because of the small number of sequences available for the 18S marker in GenBank, the additional phylogenetic analysis of concatenated sequences was made only with ITS (601 nts) and D2/D3 (754 nts) fragments. This phylogenetic tree was constructed using UPGMA (Sneath and Sokal, 1973) (Tamura–Nei model of maximum likelihood estimation (Tamura and Nei, 1993) with gamma distribution and 1,000 bootstraps).

Bacterial symbiont was isolated from hemolymph of G. mellonella larvae infected with the EPNs recovered, following the procedure described by Kazimierczak et al. (2017). After isolation, bacterial symbionts were stored at −80°C in LB Broth (Scharlab, Spain) with 20% (v/v) glycerol (Merck, United States) and deposited in the Microorganisms Collection of Pontificia Universidad Javeriana – CMPUJ.

The correspondence of bacterial isolates to what is expected for EPNs bacterial symbionts of the Xenorhabdus or Photorhabdus genus (Akhurst and Boemare, 2015; Boemare and Akhurst, 2015) was initially confirmed through microscopic and macroscopic morphological characteristics by Gram staining and, on NBTA (13 g/liter nutrient broth, 0.025 g/liter bromothymol blue, 0.04 g/liter chloride 2, 3,5-triphenyltetrazolium, and 15 g/liter agar), MacConkey (Scharlab, Spain) and blood agar (Becton, Dickinson and Company, United States). Samples were incubated at 26°C for 48 hr. To corroborate symptoms of G. mellonella larvae infected with the bacterial isolates, a colony was removed from NBTA medium after 48 hr growth at 26°C, and resuspended in saline solution (0.85% (w/v) NaCl [Merck, United States]). The suspension was serially diluted to 10⁻⁴. Then, 10 µl of the last dilution was injected into five larvae of G. mellonella with a syringe (Hamilton Company, United States). The surface of the larvae was disinfected with 1% NaClO (BNS S.A, Colombia) for 1 min, followed by three washes with sterile distilled water. As a negative control, 10 µl of sterile saline solution was injected into five larvae of G. mellonella.

Following injection, larvae were incubated at 26°C for 48 hr, after then, their coloration, mortality, and consistency were all checked.

To identify metabolic characteristics of the bacterial isolate, a colony of the bacteria was exposed to a 3% hydrogen peroxide solution (Sigma-Aldrich, United States) using sterile wooden sticks. Subsequently, a colony was taken from nutrient agar (Scharlab, Spain) after 48 hr growth at 26°C and then resuspended in 0.85% API NaCl medium (BioMérieux, France). The API 20E fast identification system (BioMérieux, France) was then used according to the manufacturer’s instructions and incubated at 26°C for 48 hr. After incubation, results were read and interpreted following the manufacturer’s instructions.

Bacterial genomic DNA was extracted following the CTAB-based DNA extraction protocol described by Feil et al. (2004). From the extracted DNA, a Multilocus Sequence Typing (MLST) analysis was performed with the partial sequences of five genes used as taxonomic markers. The following five sequences and primers were used: 16S rRNA using 16SP1 5′-GAAGAGTTTGATCATGGCTC-3′ and 16SP2 5′-AAGGAGGTGATCCAGCCGCA-3′ (Tailliez et al., 2006); dnaN (DNA polymerase III beta chain) using dnaN1 5′-GAAATTYATCATTGAACGWG-3′ and dnaN2 5′-CGCATWGGCATMACRAC-3′; recA (DNA recombinase) using recA1 5′-GCTATTGATGAAAATAAACA-3′ and recA2 5′-RATTTTRTCWCCRTTRTAGCT-3′; gltX (glutamyl-tRNA synthetase) using gltX1 5′-GCACCAAGTCCTACTGGCTA-3′ and gltX2 5′- GGCATRCCSACTTTACCCATA-3′; and rplB (50S subunit of ribosomal protein L2) using rplB1 5′-GGCAATTGTTAAATGTAAACC-3′ and rpblBXeno2 5′-GCGGCGTACGATGTATTGAT-3′ (Tailliez et al., 2010).

For molecular markers amplification, 25 µl of PCR mixture was prepared using 1X NH₄ Buffer, 3 mM MgCl₂, 0.2 mM dNTPs (Bioline, England), 0.5 µg/µl Bovine Serum Albumin (New England Biolabs, United States), forward primer and reverse primer for each 0.5 µM of taxonomic marker, Taq DNApolimerase 2U (Bioline, England) and 20 to 100 ng of bacterial DNA (Hominick et al., 1997; Tailliez et al., 2006).

The PCR protocol consisted of an initial denaturation at 94°C for 5 min, 30 cycles at 94°C for 30 sec, 30 sec at the annealing temperature of each pair of primers (16S: 64.9°C, dnaN: 45°C, recA: 45°C, gltx: 64.9°C and rplB: 58.3°C), extension at 72°C for 1 min, and a final extension at 72°C for 7 min. The specific amplification of all PCR products was verified by electrophoresis in 1% (w/v) agarose gel in 1X TBE buffer stained with 0.5 X Hydragreen (ACTGene, United States).

PCR products were purified for sequencing with Wizard SV Gel and PCR clean-up system (Promega, United States). Three independent amplicons were sequenced in both directions for each molecular marker.

The six sequences obtained for each marker were visualized, aligned, and edited manually using the MUSCLE tool (Edgar, 2004) in the MEGA X software (Kumar et al., 2018) in order to obtain the respective consensus sequence. The consensus sequence for 16S marker was initially identified by means of EzBioCloud (Yoon et al., 2017), while BLAST (Altschul et al., 1990) was used for initial identification of other markers. The consensus sequences of all markers were deposited in GenBank (NCBI) under the accession numbers registered in Table 1.

The consensus sequences of each marker were aligned with sequences of species deposited in GenBank (NCBI) using MUSCLE (Edgar, 2004) in MEGA X software (Kumar et al., 2018). The alignment obtained was edited so that all the sequences used in the analysis were of the same length and genetic region, and phylogenetic trees were constructed independently for each marker. Maximum likelihood estimation (Tamura–Nei model [Tamura and Nei, 1993] with gamma distribution, invariant sites (G + I), and 100 bootstraps) was used for the 16S fragment; maximum likelihood estimation (Kimura 2 model [Kimura, 1980] with gamma distribution, invariant sites (G + I), and 100 bootstraps) was used for the dnaN and rplB fragments; and maximum likelihood estimation (Tamura 3 model [Tamura, 1992] with gamma distribution, invariant sites (G + I), and 100 bootstraps) was used for the gltX and recA fragments.

In addition, the dnaN (815 nts), gltX (654 nts), recA (592 nts), and rplB (678 nts) markers were manually concatenated for construction of the phylogenetic tree using maximum likelihood estimation (General Time Reversible model (Nei and Kumar, 2000) with gamma distribution, invariant sites [G + I], and 100 bootstraps). The accession numbers of all the sequences used are found in Table 1.

As observed in the API biochemical gallery and under incubation conditions, the bacterial symbiont was negative for β-galactosidase, arginine-dihydrolase, lysine decarboxylase, ornithine decarboxylase, citrate assimilation, H2S production, urease, tryptophan deaminase, production of indole, acetoin production (Voges – Proskauer test), gelatinase, production of acid from glucose, mannitol, inositol, sorbitol, rhamnose, sucrose, melibiose, amygdalin, and arabinose.

The initial identification, using EzBioCloud for the 16S marker and BLAST searches of the nr database of the NCBI for other markers, showed greatest similarity to the respective Xenorhabdus nematophila markers: 99.47% identity and 63.9% coverage were obtained for the 16S marker; 99% identity and 100% coverage were obtained for dnaN; 100% identity and coverage were obtained for recA; 99% identity and 100% coverage were obtained for gltX; and 100% identity and coverage were obtained for rplB. The phylogenetic analysis showed that the bacterial isolate was grouped at the same level with X. nematophila for the 16S marker as well as for the concatenated markers, thus confirming its taxonomic identity (Fig. 4A, B).

Figure 4: