On the synonymy of Trophotylenchulus asoensis and T. okamotoi with T. arenarius, and intra-generic structure of Paratylenchus (Nematoda: Tylenchulidae)

DNA extraction

For molecular analysis, a single female nematode of each of the populations was transferred into a drop of distilled water on a microscopic slide and examined under the light microscope. Each specimen was washed three times in deionized water and then transferred into an Eppendorf tube with 25 µl distilled water. Posteriorly, 25 µl lysis buffer (23.75 µl NaCl 0.2 M and Tris-HCL 0.2 M, 1.0 µl β-mercaptoethanol and 0.25 µl proteinase K) was added to each Eppendorf tube. Nematode specimen was crushed with a microhomogeniser for 2 min. The tubes were incubated at 65°C (1 hr) and then at 95°C (15 min) (Tanha Maafi et al., 2003).

PCR amplification

DNA amplification was carried out based on the protocols described by Tanha Maafi et al. (2003). The D2–D3 expansion regions of the 28 S rRNA gene were amplified with the forward D2A (5′–ACAAGTACCGTGAGGGAAAGTTG–3′) and the reverse D3B (5′–TCGGAAGGAACCAGCTACTA–3′) primers (Nunn, 1992). The PCR was performed in a final volume of 30 µl containing: 3 µl DNA template, 1 µl of each PCR primer, 15 µl PCR mastermix (Amplicon, Denmark) and 10 µl deionized water. The following PCR program was used: denaturation 94°C for 5 min; 35 cycles × (denaturation 94°C, 30 sec; anneling 54°C, 30 sec; extension 72°C, 70 sec) and a final extension 72°C for 5 min.

The PCR products were purified using the QIAquick Gel Extraction Kit (Takapozist, Iran) according to the manufacturer’s instruction and used for direct sequencing. The PCR products were sequenced in both directions at BioNeer Inc. (Korea).

Phylogenetic analysis

The obtained sequences (Table 1) were compared with sequences of other taxa in GenBank with a BlastN homology search and then the closest sequences were selected for phylogenetic analyses. The sequences of D2–D3 segments of 28S rRNA gene were aligned with ClustalX 1.83 (Thompson et al., 1997), using default parameter values and were manually edited if necessary. Phylogenetic analyses of the sequence datasets were based on Bayesian inference (BI) using MrBayes 3.1.2 (Ronquist and Huelsenbeck, 2003). The best-fit model of DNA evolution was obtained using JModelTest V.2.1.7 (Darriba et al., 2012) with the Akaike Information Criterion (AIC). The best-fit model, the base frequency, the proportion of invariable sites, and the gamma distribution shape parameters and substitution rates in the AIC were then used in MrBayes for the phylogenetic analyses. The general time-reversible model with invariable sites and a gamma-shaped distribution (GTR + G) for the D2-D3 segments of 28S rRNA gene were run with four chains for 2 × 106 generations, respectively. The Markov chains were sampled at intervals of 100 generations. Two runs were conducted for analysis. After discarding burn-in samples and evaluating convergence, the remaining samples were retained for further analyses. The topologies were used to generate a 50% majority rule consensus tree. Posterior probabilities (PP) were given for appropriate clades. Pairwise divergences between taxa were computed as absolute distance values and as percentage mean distance valuesbased on whole alignment, with adjustment for missing data with PAUP* 4.0b 10 (Swofford, 2003). Trees were visualized using TreeView (Page, 1996).

Results

Description

Description

Description

Description

Description

Phylogenetic analysis

The D2–D3 alignment was 596 bp long and consisted of 88 sequences as ingroups and three sequences including Aglenchus agricola (De Man, 1884; Andrássy, 1954), Basiria gracilis (Thorne, 1949) and Coslenchus costatus (De Man, 1921; Siddiqi, 1978) as outgroups. Two new sequences for Trophotylenchulus species of the D2–D3 expansion fragments of 28 S rRNA gene were obtained in the present study. Phylogenetic relationships within Trophotylenchulus species and other representatives of the family Tylenchulidae inferred from the analysis of this partial 28 S rRNA gene sequences with collapsed branches, with PP less than 50%, are given in Figure 11.

The BI tree revealed good separation between the four subfamilies in trees inferred from 28 S rRNA gene sequences. Four distinct clades were formed, congruent with the morphological classification of the subfamilies in Ghaderi et al. (2016). Representatives of Paratylenchinae (Thorne, 1949) included several species of Paratylenchus sensu lato and Cacopaurus pestis (Thorne, 1943) formed clade I, with two distinct subclades 1A and 1B. Three populations of Trophotylenchulus arenarius, two ones of T. floridensis (Raski, 1957) and five Tylenchulus species (subfamily Tylenchulinae Skarbilovich, 1947) formed a well-supported clade II, T. floridensis having a basal position in this subfamily. The representatives of Sphaeronematinae (Raski and Sher, 1952) and Meloidoderitinae (Kirjanova and Poghossian, 1973) formed basal clades III and IV, respectively.

Together with another sequence of T. arenarius (AY780971), the two sequences of this species from Iran (MK733978 and MK733979) formed a high-supported clade (PP = 100%) to other representatives of Tylenchulinae including Trophotylenchulus floridensis, Tylenchulus semipenetrans (Cobb, 1913), T. musicola (Tanha Maafi et al., 2012), T. graminis (Inserra et al., 1988), T. furcus (Van den Berg and Spaull, 1982), and T. palustris (Inserra et al., 1988).

The sequences of the four Paratylenchus species from Iran, P. bukowinensis, P. nawadus, P. sheri and P. similis (having a stylet under 40 µm, four lateral lines, and with cuticular vulval flaps) clustered among other Paratylenchus species bearing stylet shorter than 40 µm in subclade IA, but P. teres (having a stylet longer than 40 µm, four lateral lines, and without cuticular vulval flaps) formed a sister cluster with P. wuae in a basal group to C. pestis and other Paratylenchus species in subclade IB; this basal group containing only long-stylet bearing species (Fig. 11).

Discussion

Different populations of the three Trophotylenchulus closely related species, T. arenarius, T. asoensis and T. okamotoi, have been recovered from America, Asia and Europe so far. Trophotylenchulus arenarius has been described from USA and later has been recorded from Spain (Gomez-Barcina and Castillo, 1990) and Germany (Sturhan, 2014). Trophotylenchulus asoensis has been recovered from Japan (Minagawa, 1983) and Iran (Gharakhani et al., 2007), and T. okamotoi has been recovered from Japan (Minagawa, 1983), Venezuela, Costa Rica and USA (Inserra et al., 1993) and Poland (Brzeski, 1998).

Gomez-Barcina and Castillo (1990) synonymized T. asoensis and T. okamotoi with T. arenarius because they believed that most diagnostic features used for separating these species are minor quantitative differences. They further noted that the shape of female tail terminus is a variable intraspecific character and is not sufficient for separating these species. However, these nematodes were treated as separate valid species in the subsequent taxonomic works (Raski, 1991; Inserra et al., 1993; Brzeski, 1998; Siddiqi, 2000; Andrássy, 2007; Ghaderi et al., 2016).

Raski (1991) distinguished the three species based on the same characters mentioned by Minagawa (1983); they used morphometric differences of female tail, male gubernaculum and J2 body size for separating of T. asoensis from T. arenarius, and the shape of female tail end for differentiation of T. okamotoi from both species.

Inserra et al. (1993) found populations of T. okamotoi in Florida, Costa Rica and Venezuela which showed intraspecific morphometric variability. However, swollen females of their populations had tapering tails with bluntly rounded or small rounded termini like those of T. okamotoi paratypes from Japan and unlike the pointed or minutely digitate tail termini of T. arenarius paratypes from California. They accepted the strong arguments of Gomez-Barcina and Castillo (1990) to support synonymy of the three species, but finally preferred to maintain the validity of the described species until more information is available on the morphological variability of T. arenarius topotypes. Brzeski (1998) classified populations of Poland as T. okamotoi, but he admitted that controversy exists about the identification of these three species.

Two populations of Trophotylenchulus were recovered during the present study which one of them (Baghmalek population) had characteriztics of T. okamotoi (females with usually bluntly rounded tail end) and the other (Dehdasht population) had morphometric characters more similar to T. arenarius (c ratio in females  = 11–15, male gubernaculum 4.5 µm and second-stage juveniles longer than 320 µm) than to T. asoensis, but the shape of tail terminus in females or juveniles was more similar to the latter species. Furthermore, a reduced stylet was observed in a single male recovered during present study that this is in agreement with the original description of T. asoensis; males of the type population T. arenarius should be restudied for this character because they were described and illustrated without stylet. Swollen females of the population similar to T. okamotoi had a tail bearing usually bluntly rounded terminus; however, indented or finely rounded termini were also common in the population. Swollen females of the other population had a conoid tail with finely rounded terminus, and the only recovered male also had a conoid tail with a short terminal projection. Juveniles in both populations had similar tails with a usually slender acute projection at end; however, the projection was slightly shorter in the population similar to T. okamotoi. Finally, it may be concluded that all morphological (particularly the shape of female/juvenile tail terminus) or morphometric (c ratio of females, the length of gubernaculum in males, and the body size of second-stage juveniles) characters do not show any distinct border delimitating the three species. On the contrary, the two partial 28 S rRNA sequences of the populations from Iran had complete identity with each other and show only 4 bp (0.67%) difference with another sequence of T. arenarius in the GenBank (AY780971). They formed a clade with high probability support (PP = 100%) with T. floridensis and Tylenchulus spp. Considering current knowledge from morphological and molecular studies, we support here T. asoensis and T. okamotoi as junior synonyms of T. arenarius, as it has already been proposed based on morphological observations by Gomez-Barcina and Castillo (1990).

The other interesting result inferred from the phylogenetic analysis of the present study, distant position of the three sequences of T. arenarius (as representatives of Trophonema sensu; Raski, 1957) from the two sequences of T. floridensis (as a representative of Trophotylenchulus sensu; Raski, 1957). The two isolates of T. floridensis (JN112253 and JN112254) had 19 bp (3.18%) difference with each other, but their differences with the three isolates of T. arenarius were 130 to 138 bp (22.2–23.6%). Considering such a relatively high divergence, the possibility of Trophonema as a valid genus cannot be excluded with the present data. However, future taxonomic studies coupled with DNA sequencing on additional representatives of Trophonema and Trophotylenchulus will help to clarify relationships within and between these two genera.

From the five sequenced Paratylenchus sensu lato during the present study, three species (P. bukowinensis, P. nawadus and P. similis) belong to the genus Paratylenchus sensu stricto, but P. teres belongs to the Gracilacus sensu (Raski, 1962), and P. sheri belongs to the Paratylenchoides sensu (Raski, 1973). Although all species of Gracilacus formed a basal cluster in subclade IB to the species with stylet shorter than 40 µm (Paratylenchus sensu stricto) in the subclades IA and IB, P. straeleni (the other species with a stylet longer than 40 µm) occupied an unresolved position among short-stylet bearing species of the subclade IA.

Raski (1973) distinguished Paratylenchoides (Raski, 1973) from Paratylenchus based on stronger cephalic sclerotization, head dorso-ventrally narrower, and in having a small, narrow, rounded protrusion on the anterior surface of the conoid lip region. The only representative of this genus, P. sheri, occupied an unresolved position among other Paratylenchus species (subclade IA) including P. nanus, P. similis, P. bukowinensis, P. conicephalus and P. hamatus. Therefore, the validity of Paratylenchoides cannot be supported with the present phylogenetic analysis. It should be considered as a junior synonym of Paratylenchus, as already been stated by several studies (Raski and Luc, 1987; Raski, 1991; Brzeski, 1998; Siddiqi, 2000; Andrássy, 2007; Decraemer and Hunt, 2013; Ghaderi et al., 2014, 2016) based on information obtained from morphological observations.

Ghaderi et al. (2014, 2016) provided a grouping for all Paratylenchus sensu lato species, falling them into 11 groups based on stylet length, the number of lateral lines and the presence of advulval flaps. Groups 1 to 5 included short-stylet (<40 µm) species (Pratylenchus spp.), and groups 6 to 11 contain representatives of long-stylet (>40 µm) species (Gracilacus spp.). These groupings are sometimes in congruence with our results. In the tree (Fig. 11), clade IA mainly contains representatives of Group 3 (stylet < 40 µm, lateral lines = 4, advulval flaps present) but also has four isolates of P. straeleni from Group 10 (stylet > 40 µm, lateral lines = 4, advulval flaps present) and several unknown species (without any available morphological characters). With the present information, we cannot explain the clustering of P. straeleni with short-bearing stylet species. A more detailed morphological and molecular study on different populations of this species is recommended to clarify the unusual position of P. straeleni in the 28 S rRNA and ITS (present study; Van den Berg et al., 2014; Mokaram Hesar et al., 2019) phylogenetic trees.

The first cluster of subclade IB contains C. pestis, P. shenzhenesis (Wang et al., 2013) from Group 3, three species of Group 2 (stylet < 40 µm, lateral lines  =  3, advulval flaps present), and P. jasminae Phani et al., 2019 from Group 4 (stylet < 40 µm, lateral lines = 4, advulval flaps absent); the other cluster of the subclade IB can be identified as Gracilacus species (with stylet > 40 µm),with three (Groups 8 and 9) or four (Groups 10 and 11) lateral lines and either with (Groups 8 and 10) or without (Groups 9 and 11) advulval flaps.

As a starting point for a detailed analysis and molecular characterization of the genus Paratylenchus, Van den Berg et al. (2014) found some species complexes containing several species in P. aquaticus (Merny, 1966), P. hamatus (Thorne and Allen, 1950) and P. nanus (Cobb, 1923). They further noticed that the validity of Gracilacus cannot be rejected with their data sets. Our phylogenetic analysis revealed that such species complexes may be present also for P. similis because two isolates of the species and also an isolate of the very closely related species, P. labiosus Anderson and Kimpinski, 1967, occupied different positions in the tree. Regardless to the position of P. straeleni, the validity of Gracilacus is supported here as seven representatives of Gracilacus formed a distinct basal cluster from other Paratylenchus species. Nevertheless, it seems that the perfect outline of the Paratylenchus sensu lato still needs the inclusion of more sequences of known or probably undescribed species.