Steinernema sandneri n. sp. (Rhabditida: Steinernematidae), a new entomopathogenic nematode from Poland

Abstract A new species of entomopathogenic nematodes, Steinernema sandneri n. sp., was recovered by baiting from Poland. Its morphological traits indicate that the new species is a member of the feltiae-kraussei group. A body length of 843 (708–965) μm, a more anterior position of excretory pore (56 μm), and the lower D% value (40 vs > 46) discriminate this species from most of the other group members. The first-generation males of S. sandneri n. sp. can be distinguished from the other clade members by a 60 μm long spicule, a relatively long gubernaculum (GS% = 79), and the position of the excretory pore (80 μm). Phylogenetic analysis of the ITS rDNA, D2D3 of 28 S rDNA, and cox1 sequences confirmed that S. sandneri n. sp. is a new species of the feltiae-kraussei group, closely related to S. kraussei and S. silvaticum.


Morphological and morphometric studies
For light and scanning electron microscope observations, different life stages of S. sandneri were obtained from infected Galleria mellonella (Lepidoptera: Pyralidae) larvae exposed individually to ~50 infective juveniles in 0.5 ml Eppendorf test tubes for 18-24 h. Male and female nematodes of the first and second generation were obtained during dissections of insect cadavers in Ringer's solution after 5 or 10 days at 17.5°C, respectively. IJs were harvested with a modified White trap method (Stock and Goodrich-Blair, 2012) and collected in tap water for 5 days after initial migration. For light microscopy, all developmental stages of the nematodes were heat-relaxed in Ringer's solution (55°C, 5 min) and fixed in 2% formalin (48 h, room temperature). After fixation, the specimens were processed using the modified Seinhorst (1959) method and mounted in pure glycerin. All measurements were performed with a Leica 5500B microscope fitted with DIC optics, a digital camera (Leica 290HD), and the Leica Application Suite ver. 3.8.0 software. For SEM of IJs, first-generation males and females of the nematodes were prepared as described previously by Skrzypek et al., 2011 and observed with a scanning electron microscope (LEO 1430VP) at 15-kV accelerating voltage in a high-vacuum mode.

Hybridization test
Reproductive isolation of S. sandneri (isolate S17-050) and S. kraussei, S. silvaticum, S. feltiae, S. oregonense, S. ichnusae, S. weiseri, S. jollieti, and S. cholashanense was tested using the Nguyen and Duncan (2002) method. Simultaneously, negative (virginity/self-fertilization) and positive (crosses between females and males of the same species) controls were performed. All the treatments were replicated 30 times for each combination of the nematode species and observed for 20 consecutive days at 17.5°C.

Molecular characterization and phylogenetic analysis
DNA was extracted from three single virgin firstgeneration females of nematodes using a DNeasy Blood and Tissue Kit (Qiagen, Germany). PCR amplification of the internal transcribed spacer (ITS) region of rDNA, the D2D3 region of 28 S rDNA, and the mitochondrial cox1 gene encoding cytochrome c oxidase subunit was performed as described earlier by Lis et al. (2019). Three sets of primers (synthesized by Genomed, Poland) were used: 18 S and 26 S for ITS (Vrain et al., 1992), D2F and 536 F for D2D3 (Nguyen, 2007b;Stock et al., 2001), and 507 F and 588 R for cox1 gene (Nadler et al., 2006). The sequences obtained in this study were compared with those deposited in the GenBank using BLAST available on the NCBI website. Multiple sequence alignments were created using ClustalW (Higgins and Sharp, 1988) at the default configuration included in MEGA 6.06 (Tamura et al., 2013) and then optimized manually. Based on the aligned sequence datasets, phylogenetic trees of the studied nematode strains were inferred in MEGA 6.06 using the Maximum Likelihood method with best fit nucleotide base substitution models HKY + G for ITS, GTR + G for D2D3, and HKY + G + I for the cox1 gene (Hasegawa et al., 1985;Nei and Kumar, 2000). Caenorhabditis elegans was used as an outgroup. To determine the statistical support for the branches, bootstrapping with 1,000 replicates of the data were conducted (Felsenstein, 1985). Percentages of sequence identity were calculated from the multiple alignments using the SIAS (Sequence Identity and Similarity) application at the default configuration (Reche, 2008). Estimation of evolutionary divergence expressed by the number of base differences between the sequences was performed using Mega 6.06 at the pairwise deletion option. The number of unique positions in the sequences of S. sandneri S17-050 was computed using the same program. Accessions numbers of all sequences and details on nematode taxa used in the molecular study are presented in Table S1.

Infective juvenile
Body straight or slightly abdominally curved when heat-relaxed, tapering gradually from the base of esophagus to the anterior end and from anus to

First-generation male
Body C-or J-shaped when heat-relaxed. Cuticle with faint transverse striation visible in SEM ( Fig. 3A-D). Lateral fields not observed. Cephalic region smooth, rounded, with four cephalic and six smaller labial papillae and slit-like amphid openings (Fig. 3A

Second-generation male
Similar to first-generation male but shorter and more slender. Excretory pore located more posteriorly. Tail relatively longer, with prominent mucron (Table 1; Fig. 4C,D).

First-generation female
Body C-shaped when heat-relaxed and fixed. Cuticle smooth when observed in a light microscope, with faint striation in SEM (Fig. 5B  anterior to basal bulb (Fig. 5A). Cardia prominent. Gonads amphidelphic, reflexed. Vulva in the form of transverse slit located slightly posterior to mid-body (Table 1). Vulval lips slightly protruding, asymmetrical, with larger posterior lip (Figs 1D and 5C,D). Tail length shorter than body anal diameter, with slight postanal swelling. Tail terminus with mucron (Figs 1C and 5E,F).

Second-generation female
Similar to first-generation female but smaller. Vulva more protruding, with distinct asymmetry between lips. Tail with mucron, but without pronounced postanal swelling (Table 1).

Life cycle
Steinernema sandneri n. sp. can be successfully reared on G. mellonella or Tenebrio molitor (Coleoptera: Tenebrionidae) larvae at a temperature in the range of 15-20°C. The life cycle of S. sandneri n. sp. is similar to that of other Steinernema species. G. mellonella larvae exposed to 50-100 IJs die within 3-4 days. Adults of the second generation can be found

Diagnosis and relationship
The new species was characterized by analysis of the morphology and morphometrics of IJs and adults (Table 1)

Cross-breeding tests
Mating attempts were observed between S. sandneri n. sp. and S. kraussei and S. silvaticum, but no fertile offspring was produced in any of the crosses. Hybridization tests with S. kraussei, S. silvaticum, S. feltiae, S. oregonense, S. ichnusae, S. weiseri, S. jollieti, and S. cholashanense showed that S. sandneri n. sp. was reproductively isolated. The positive control always yielded a progeny.  manubrium is almost as long as it is wide, the shaft is short, and the velum expands from the calomus to the end of the ventral rib. The tail of both generation males is mucronated. The first-and second-generation females of S. sandneri n. sp. have a slightly protruding vulva and a mucron at the posterior end.

Type locality and habitat
Natural host unknown. The nematode isolate S17-050 was obtained from sandy-loamy soil samples collected in eastern Poland (51°46'55"N 22°42'35", 147 m a.s.l.) in 2017. The soil samples were collected in a mixed forest from 0 to 20 cm depth. Nematodes were isolated using a modified live trap method (Bedding and Akhurst, 1975) with the use of G. mellonella larvae as a bait. Detailed studies were performed on a straight line of nematodes (offspring of 2 IJs) reproducing successfully in G. mellonella and maintained in our laboratory.

Type designation and deposition
Holotype male, paratype males, paratype infective juveniles, paratype females, and second-generation paratype males and females were deposited in the nematode collection of the Museum and Institute of Zoology, Polish Academy of Sciences, Wilcza 64, Warsaw, Poland (see Table S2 for deposition numbers).

Molecular characterization and phylogenetic relationships
S. sandneri n. sp. was characterized genetically by the sequences of the ITS rDNA, D2D3 of 28 S rDNA, and the mitochondrial cox1 gene. No variation in the sequences of these genes was found between the analyzed individuals. The D2D3, ITS, and cox1

JOURNAL OF NEMATOLOGY
sequences of S. sandneri n. sp. were deposited in the GenBank with accession numbers MW078535, MW078536, and MW078544, respectively. As it is known that the molecular diversity in the group of nematodes assigned as S. kraussei is relatively high, we included multiple S. kraussei sequences in the molecular analysis, also these of S. kraussei from the Lublin region, which are sympatric to the new species (Table S1). Compared to other species of the genus Steinernema with ITS sequences available in the GenBank, S. sandneri S17-050 showed the highest ITS sequence identity with S. kraussei strains, i.e. 96.0-97.7%, corresponding to 16-28 nucleotide substitutions (Table S3). It was also noted that the GeneBank sequence AY171250, attributed to S. kraussei from Belgium, displayed 99.7% identity and 2 bp difference from this of S17-050 isolate, which implies that this nematode is a conspecific to S. sandneri n. sp and should be considered as misidentification. Among the other Steinernema species, the most similar sequence of the ITS region with S. sandneri n. sp. was displayed by S. silvaticum (94.5-95.0% identity, 34-37 substitutions) and S. xinbinense (94.5% identity, 35 nucleotide substitutions). The ITS sequences of the other species of the Steinernema genus were more divergent from that of S. sandneri n. sp., showing identity ≤ 94% and at least 41 nucleotide substitutions (Tables 4 and S3).
The highest sequence identity of the D2D3 region of the new species was 98.2%, corresponding to 15 nucleotide substitution, in respect to the analyzed S. kraussei strains. The new species differs from other species from the feltiae-kraussei group by at least 21 bp, showing ≤ 97.5% nucleotide identity (Table 5).
The alignment of the analyzed ITS sequences resulted in 870 positions, in which 266 positions were conserved, while 573 positions were variable, including 426 parsimony-informative and 136 singleton ones. The phylogenetic tree based on the ITS sequences shows that S. sandneri n. sp., S. kraussei, and S. silvaticum form a monophyletic cluster with 95% bootstrap support (BS) within the feltiae-kraussei group. It is also noted that S. sandneri n. sp. clusters with S. kraussei isolates as a sister group with 94% BS (Fig. 6). The ITS rDNA of S. sandneri n. sp. differs from that of the other species of the feltiae-kraussei group by four unique traits (present in the sequence alignment only in the new species but not in the  Liu and Berry (1996) Notes: Measurements are given in μm and in the form: mean (range). a abbreviations as in Table 1, NA = data not available. Notes: Measurements are given in μm and in the form: mean (range). a abbreviations as in Table 1. b MUC = mucron; P = present, A = absent, NA = data not available.
others from the group) in the following positions: 530, 608, 656, and 713. In addition, S. sandneri n. sp. contains a unique stretch of eight adenine nucleotides in position 480-487 of the ITS sequence.
In the case of the D2D3 region sequences, the alignment resulted in 899 positions, in which 611 positions were constant, while 247 positions were variable, including 130 parsimony-informative and 117 singleton ones. The phylogenetic tree based on D2D3 sequences shows that S. sandneri n. sp. and S. kraussei isolates form a monophyletic group with BS 56%, which is a part of the clade comprising S. silvaticum, S. xinbiense, S. cholashanense, S. tielingense, and S. oregonense with BS 68% (Fig. 7). S. sandneri n. sp. differs from the other species of the feltiae-kraussei group in the D2D3 region by six diagnostic traits in the following sequence positions: 52, 155, 374, 444, 456, and 463. The alignment in the cox1 gene sequences resulted in 567 positions, in which 376 positions were conserved, while 191 positions were variable, including 109 parsimony-informative and 82 singleton ones. The analysis involved only nine steinernematid nematode species, as the number of cox1 sequences available in the GeneBank is still limited. The phylogram based on cox1 gene sequences shows a    clade separating S. sandneri n. sp., S. kraussei, and S. silvaticum with BS 88% (Fig. 8). In this clade, the new species and S. kraussei strains form a sister branch with 100% BS. Steinernema sandneri n. sp. differs from the other species of the feltiae-kraussei group in the cox1 gene by five diagnostic traits in positions 21, 138, 165, 225, and 263.

Discussion
Sequence analysis of ITS rDNA and D2D3 expansion segment of 28 S rDNA have been proved useful for estimation of EPN species, by supporting morphological data (Nadler et al., 2006;Nguyen, 2007a, b;Stock et al., 2001). The phylogenetic trees based on ITS, D2D3, and cox1 gene sequences presented in this paper show that S. sandneri n. sp. has a unique position in the feltiae-kraussei group and is evolutionarily very close to S. kraussei and S. silvaticum. A number of studies highlighted also the suitability of sequence divergence of these two regions as a good indication of lineage independence (e.g. Spiridonov et al., 2004a, b). The pairwise distances of sequences of the three studied genes clearly differentiate the new species from other nematodes in the feltiae-kraussei group. Nevertheless, so far, there is no defined threshold of the ITS or D2D3 rDNA similarity that may indicate whether the studied nematode is a new species or not. Nguyen (2007a, b) suggested an ITS threshold of 95% for Steinernema species; however, many closely related species of this genus do not meet this threshold -the difference in ITS sequences between closely related species of Steinernema is often ~3% (Spiridonov et al., 2004a, b). The sequence of the ITS region of S. sandneri n. sp. shows 2.3-4.0% difference from that of S. kraussei isolates (or 3.5-6.0% according to the other sequence identity definition). In fact, the main limitation of using the ITS sequence for estimation of the evolutionary relationships of EPN is their intraspecies and intra-individual sequence variability, making sequence aligning dubious and varying the estimation of the sequence identity (Půža et al., 2015). In turn, phylogenetic analyses of D2D3 have pro vided evidence that this region has fever ambiguously aligned positions than ITS rDNA, nevertheless it is too conservative to be informative of the relationships between closely related species of the feltiae-kraussei group (Nadler et al., 2006). The assessment of the amount of phylogenetic information by determination of the number of variable sites in the sequence alignments used in this study demonstrated that the D2D3 region had a substantially lower number of such positions, compared to ITS rDNA, i.e. 27.5 vs 66.2%. The D2D3 sequence of S. sandneri n. sp. shows 1.8% difference from S. kraussei and ≥ 2.5% divergence from the other species of the group.
We also analyzed the sequence of the mito chondrial cox1 gene of S17-050 nematode. The analysis revealed the highest level of its genetic divergence (6.2-7.1%) from sequences of S. kraussei strains, compared to the other molecular markers used. Data have shown that the cox1 gene undergoes fast evolution within the feltiae-kraussei group, inferring well the phylogenetic relationships among closely related species of this clade (Peat et al., 2009;Szalanski et al., 2000). However, the suitability of this gene to Steinernema species delimitation is still limited since a low number of sequences are available for comparison. In the case of the cox1 phylogram presented in this study, some uncertainty occurs due to the low number of sequences included; therefore, this is only an approach to resolving evolutionary relationships steinernematid nematode species related to S. sandneri n. sp. before more cox1 sequences appear.
In addition, the new species is well supported by the molecular diagnostic traits. Current evidence suggests that finding autapomorphies is useful in delimitation of nematode species for better indication of lineage independence . The sequence alignments of S. sandneri n. sp. show that it has four, six, and five diagnostic traits for ITS, D2D3, and cox1, respectively. S. sandneri n. sp. can also be easily differentiated from the other species from the group by the unique stretch of adenine nucleotides in the sequence of ITS rDNA.
In conclusion, the molecular analysis based on ITS rDNA, D2D3 of 28 S rDNA, and cox1 gene sequences confirms the status of S. sandneri n. sp. as a new species according to the phylogenetic and evolutionary species concept (Adams, 1998).