Comparative molecular analysis of Bursaphelenchus vallesianus, a wood-inhabiting nematode isolated from declining pine trees in the Czech Republic

Authors


Abstract

The Bursaphelenchus genus (Nematoda: Parasitaphelenchidae) comprises mostly wood-inhabiting nematodes that feed on various tree-colonizing fungi. One species of the genus, Bxylophilus, has been proven as an agent causing pine wilt disease (PWD). However, involvement of other Bursaphelenchus species in the PWD remains enigmatic. In the current paper, comparative molecular analysis is performed based on nuclear ribosomal DNA (rDNA) of Bvallesianus, a species that was recently isolated from pine trees (Pinus sylvestris) exhibiting wilting and declining symptoms in the Czech Republic. Sequencing of the nuclear-encoded ITS1–5·8S–ITS2 rDNA region confirmed previous taxonomic conclusions based on morphology. Evolutionary reconstructions resulted in a phylogenetic tree, where the Czech isolate of Bvallesianus occupied a common clade together with other species belonging to the so-called Bsexdentati group. Unexpectedly, comprehensive analysis of the sequence data revealed a genetic variation distinguishing the Czech isolate of Bvallesianus from all other species of the Bsexdentati group. This dissimilarity consists of the presence of a four nucleotide exchange found in the 5·8S rRNA-coding gene. The newly identified genetic variation appears to affect the 5·8S rRNA folding, as deduced from secondary structure models. Additionally, it is shown that for the first time, to the authors’ knowledge, both bursaphelenchid internal transcribed spacers (ITS1 and ITS2) fold into the multibranched closed loops. While the ITS2 closed loop is formed with help of canonical 5·8S-28S rRNA pairing, the ITS1 forms the thermodynamically stable closed loop with no support of flanking rRNA sequences. The current information on bursaphelenchid ITS rDNA sequence diversity and structure is further discussed.

Introduction

The genus Bursaphelenchus (Nematoda: Parasitaphelenchidae) comprises mostly wood-inhabiting, fungal-feeding nematodes that inhabit various trees (Rutherford et al., 1990). So far, more than 70 species have been described within this genus (Ryss et al., 2005). One species from the genus, Bxylophilus, has been proven as a causal agent of pine wilt disease (PWD; Rutherford et al., 1990). Bursaphelenchus xylophilus is native to North America, where it causes little or no damage to trees (Jones et al., 2008). Introduction of Bxylophilus has been reported in Portugal (Mota et al., 1999), Spain (Abelleira et al., 2011) and Madeira Island (Fonseca et al., 2012), where the nematode can cause huge damage. In contrast, involvement of a few other Bursaphelenchus species, for example Bmucronatus and Bsexdentati, in PWD development remains unclear. However, recent controlled experimental inoculations in greenhouse conditions showed that these species also possess a pathogenic potential (Polomski & Rigling, 2010; Dayi & Akbulut, 2012). Therefore, because of the detection of Bxylophilus in Europe, strategic programmes have been applied to monitor the distribution of Bursaphelenchus spp. throughout the European continent. One result which has emerged from these is the discovery and description of a new Bursaphelenchus species, Bvallesianus (Braasch et al., 2004).

The first occurrence of Bvallesianus was reported in Valais (Switzerland), where this species was isolated from declining Pinus sylvestris trees (Braasch et al., 2004; Polomski et al., 2006). Bursaphelenchus vallesianus is morphologically similar to Bsexdentati and Bborealis, and therefore clearly belongs to the so-called Bsexdentati group (Braasch et al., 2004). Further monitoring activities identified the occurrence of Bvallesianus in other European countries including the Czech Republic (Zouhar et al., 2006) and Germany (Schonfeld et al., 2006, 2008), and also in the Asian region, such as Turkey (Akbulut et al., 2008). Involvement of Bvallesianus in PWD is poorly understood. However, recent bioassays showed that experimental inoculation of 3-year-old Psylvestris trees led to similar, if not identical, external and internal symptoms that are caused by Bxylophilus (Polomski & Rigling, 2010). More recently, a pathogenic potential of Bvallesianus was also observed in greenhouse conditions, where 2-year-old seedlings of three different pine species, Pinus nigra, Pbrutia and Ppinea, were experimentally inoculated (Dayi & Akbulut, 2012). These findings supported the theory that overall mechanisms causing PWD are shared by all pathogenic Bursaphelenchus species, including Bvallesianus (Polomski & Rigling, 2010). Therefore, further investigations on Bvallesianus are needed to get broader insights into its biology, possible pathogenicity and evolutionary history, and thus to better understand its role in PWD development.

In the present study, comparative molecular analysis of Bvallesianus isolated from declining pine trees in the Czech Republic was performed. The sequence of the nuclear ITS1–5·8S–ITS2 rDNA region was determined and phylogenetic relationships between the Czech isolate of Bvallesianus and all other species belonging to the Bsexdentati group were reconstructed. Comprehensive analysis of the sequence data revealed an unexpected 4-nucleotide substitution occurring in the 3′-end of the 5·8S rRNA which appears to influence the structural folding of 5·8S rRNA, but its biological implications remain unclear. Finally, the secondary structure of both bursaphelenchid internal transcribed spacer (ITS1 and ITS2) transcripts is predicted. The present knowledge of bursaphelenchid ITS rDNA sequence diversity is further discussed.

Materials and methods

Nematode sampling and extraction

Using a drill approach, numerous wood samples were collected from the trunks of pine trees (Pinus sylvestris) exhibiting wilting and declining symptoms in Rohatec (Moravia, Czech Republic; Fig. 1a), where occurrence of Bvallesianus was previously reported (Zouhar et al., 2006). Nematodes were extracted from wood pieces using a modified Baermann funnel method. Morphological determination and morphometric measurements were performed on specimens (10 males and 10 females) freshly isolated from wood samples, following the diagnostic key for the genus Bursaphelenchus (Ryss et al., 2005).

Figure 1.

(a) Geographic locality of Rohatec (GPS coordinates: 48°53′25·375″N, 17°11′13·386″E) shown on a map of the Czech Republic. (b) Typical wilting and declining symptoms of pine tree (Pinus sylvestris). (c) Presence of colonizing blue-fungi in the wood of the declined pine tree. White arrowheads point to debarked trunk uncovering massive blue-fungi colonization. (d) Light microscopy image of Bursaphelenchus vallesianus male. Inset 1 and 2 shows head region and tail, respectively.

Genomic DNA preparation

Total genomic DNA was separately extracted from nematodes eluted from eight independent wood samples from Rohatec, following a modified method previously described by Marek et al. (2005). Briefly, c. 20–30 individuals of Bvallesianus were transferred to a sterile 1·5 mL tube containing 50 μL lysis buffer (100 mm Tris–HCl [pH 8·0], 5 mm EDTA, 200 mm NaCl, 0·2% SDS and 4 μg mL−1 proteinase K). The mixture was incubated for 1 h at 37°C and then proteinase K was denatured for 5 min at 85°C. The DNA was extracted through phenol–chloroform extraction. Finally, DNA was precipitated with an equal volume of ice-cold isopropanol and washed with 80% ethanol. The final DNA was resuspended in 20 μL TE buffer (10 mm Tris, 0·5 mm EDTA [pH 8·0]). The DNA stocks were stored at −20°C.

PCR amplification and restriction assays

To amplify the nuclear rDNA region encompassing the 3′-end of 18S rRNA, complete sequences of ITS1, 5·8S rRNA and ITS2, and the 5′-end of 28S rRNA, universal primers F194 (5′- CGTAACAAGGTAGCTGTAG-3′) and 5368 (5′-TTTCACTCGCCGTTACTAAGG-3′) were used (Ferris et al., 1993). PCR reactions were performed in 0·2 mL tubes with 25 μL final volume of reaction mixture containing 2·5 μL 10 × buffer for DNA polymerase (Fermentas), 3 μL 25 mm MgCl2 (Fermentas), 0·25 μL 25 mm dNTP (Fermentas), 0·25 μL oligonucleotide primers (50 μm), 0·4 μL (2·5 U) LA Taq DNA polymerase (Fermentas) and 1 μL DNA (100 ng μL−1). The PCR protocol was as follows: initial denaturation at 95°C for 2 min, followed by 35 cycles of 94°C for 1 min, 45–5°C for 30 s, and 72°C for 1 min, and a final extension at 72°C for 7 min.

For restriction assays, the amplified DNA (15 μL) was digested with 3·5 U of restriction endonucleases (AluI, HaeIII, HinfI, MspI and RsaI) for 4 h at 37°C in a total volume of 30 μL. The cleaved PCR products were separated by electrophoresis in 1% (w/v) agarose gels with 1 × TBE buffer, stained with ethidium bromide and visualized under UV light. The lengths of DNA fragments were estimated by comparison with MassRuler Low Range 100 bp DNA ladder (Fermentas).

Cloning and DNA sequencing

The amplified DNA fragments were gel-purified using the MinElute Gel Extraction Kit (QIAGEN) and cloned into the pTZ57R/T plasmid (Fermentas) as previously described (Marek et al., 2005). The DNA sequencing was performed by Genomac International (Czech Republic). In total, nucleotide sequences of the ITS1–5·8S–ITS2 rDNA region, which encompassed the 3′-end of 18S rRNA, complete sequences of ITS1, 5·8S rRNA and ITS2, and 5′-end of 28S rRNA, were determined for eight independent Bvallesianus isolates from Rohatec.

Sequence analysis and phylogeny reconstruction

For comparative analysis, the nucleotide sequences of ITS1–5·8S–ITS2 rDNA of other Bursaphelenchus species were retrieved from the non-redundant GenBank database (Benson et al., 2011) at the National Center for Biotechnology Information (Sayers et al., 2011). Multiple sequence alignments were constructed using clustalX v. 2.0 (Larkin et al., 2007) with a gap-opening penalty of 10 and gap-extension penalty of 0·05. Minor manual adjustments were performed in BioEdit (Hall, 1999) in order to minimize occurrence of misaligned sequences. Final graphical visualization of aligned sequences was done in jalview (Waterhouse et al., 2009). Evolutionary divergence (p-distance) between nucleotide sequences were computed from pairwise analysis using the Maximum Composite Likelihood method implemented in mega 4 (Tamura et al., 2007).

For molecular phylogeny reconstructions, all columns containing gaps were excluded from multiple alignments. The phylogenetic trees were built on the basis of multiple alignments using the maximum likelihood (ML) method as implemented in the software mega 4 (Tamura et al., 2007), which was also used for bootstrap analysis (500 replicates) and graphical representation of the resulting trees.

RNA secondary structure predictions

RNA secondary structure models were inferred as previously described (Marek et al., 2010; Douda et al., 2013). Briefly, structural two-dimensional (2D) models for ITS1, 5·8S and ITS2 transcripts for a Czech isolate of Bvallesianus (GenBank accession HM756288), Bvallesianus (AM269920), Bsexdentati (AM269915), Bborealis (AM179511), Bpoligraphi (AM179512) and Bpinophilus (AM160664) were computed by the Mfold algorithm (Zuker, 2003). Screening for thermodynamically optimal and suboptimal secondary models was performed with the help of the RNAstructure v. 4.4 software (Reuter & Mathews, 2010). The default folding parameters were used with exception of temperature adjustment to 25°C. The predicted models were exported from RNAstructure in ct format for final refinements and visualization with RNAviz (De Rijk & De Wachter, 1997) software.

Results

Morphological determination

Microscopic observations and morphometrics measurements of Burspahelenchus spp. isolated from the wilted and declined pine trees (P. sylvestris) from Rohatec showed that it is Bursaphelenchus vallesianus. Detailed information on geographic locality, overview of pine trees exhibiting typical wilting and declining symptoms, presence of bluestain fungi in declined pine trees, and representative photomicrographs of isolated Bvallesianus specimens are shown in Figure 1. The results of morphological measurements are summarized in Table 1. The determined morphometric parameters of Bvallesianus from Rohatec are in perfect agreement with previously published descriptions by Braasch et al. (2004) and Akbulut et al. (2008) (Table 1).

Table 1. Morphometrics of a Bursaphelenchus vallesianus isolate from the Czech Republic
CharacterBvallesianus isolate from the Czech RepublicBvallesianus Akbulut et al. (2008)Bvallesianus Braasch et al. (2004)
FemaleMaleFemaleMaleFemaleMale
  1. n, number of specimens on which measurments are based; L, overall body length; a, body length/greatest body diameter; c, body length/tail length; V, % distance of vulva from anterior. All measurements in μm and in the form: mean ± SD (range).

n 101010101010
L

1000·3 ± 73·4

(866·0–1116·0)

967·3 ± 96·0

(752·0–1107·0)

796 ± 48·1

(736–850)

810 ± 80·5

(696–914)

753 ± 110

(613–918)

880 ± 110

(688–1063)

Stylet

13·0 ± 0·9

(11·0–14·0)

12·9 ± 0·7

(12·0–14·0)

13·6 ± 0·7

(12·8–14·4)

13·6 ± 0·8

(12·7–15·1)

13 ± 0·9

(12–15)

13 ± 1·1

(11–15)

Tail length

26 ± 2·3

(23·0–33·0)

29 ± 2·2

(26·0–35·0)

25·2 ± 4·6

(20–31)

20·9 ± 4·8

(17–27)

26 ± 5·1

(19–37)

26 ± 3·1

(19–31)

Anterior end to anus

974·3 ± 73·7

(839–1090)

938·3 ± 93·7

(723–1078)

Vulva to anus

732·2 ± 54·9

(629–820)

a

53·7 ± 4·8

(47·3–64·1)

61·0 ± 5·8

(47·0–68·9)

33·6 ± 5·02

(28–41)

40·6 ± 3·4

(36·5–46)

29 ± 7·6

(22–46)

38 ± 3·1

(31–42)

c

38·8 ± 4·6

(29·4–43·6)

33·5 ± 3·4

(25·9–38·2)

32·2 ± 6·3

(25–42)

42·0 ± 5·3

(37·6–49·8)

30 ± 5·3

(20–37)

30·4 ± 4·0

(28–42)

V

73·2 ± 1·0

(72·0–75·1)

75·3 ± 1·5

(73–77)

73 ± 1·6

(71–75)

Spicules

15·6 ± 2·3

(13·0–21·0)

16·2 ± 1·5

(14·4–18·0)

17 ± 1·5

(15–9)

Restriction mapping of the nuclear ITS1–5·8S–ITS2 rDNA region

To confirm morphological conclusions about B. vallesianus, a molecular approach was used. To address this, the nuclear ITS1–5·8S–ITS2 rDNA region was selected for genetic analysis because numerous homology sequences from Bursaphelenchus spp. have been determined and deposited in GenBank (Benson et al., 2011). The PCR amplification using universal primers, F194 and 5368, resulted in a single DNA amplicon of c. 980 bp that comprised the partial sequence (3′-end) of 18S rRNA, full lengths of ITS1, 5·8S rRNA and ITS2, and partial sequence (5′-end) of 28S rRNA (Fig. 2). The highest yield and purity of the PCR-amplified product was achieved using an annealing temperature of 56·5°C, as determined by gradient-temperature PCR screening (data not shown).

Figure 2.

Experimentally determined restriction patterns of the ITS1–5·8S–ITS2 rDNA region of the Czech isolate of Bursaphelenchus vallesianus (a). Referential virtual restriction patterns of ITS1–5·8S–ITS2 rDNA of Bsexdentati (b), Bvallesianus (c), Bmucronatus (d) and Bxylophilus (e). Note that restriction patterns of the Czech isolate of Bvallesianus (a) are absolutely consistent with those virtually derived from the originally described Bvallesianus species (c).

In the next step, a set of restriction endonucleases (AluI, HaeIII, HinfI, MspI and RsaI) was used to generate restriction profiles of the amplified ITS rDNA fragment in order to unambiguously determine the Bursaphelenchus isolate, an approach previously described by Burgermeister et al. (2009). Careful inspection and comparative analysis with virtually generated restriction patterns from some representative Bursaphelenchus species showed perfect agreement with Bvallesianus (Fig. 2). The restriction mapping data thus strongly supported conclusions based on morphological measurements, that the Bursaphelenchus isolate from Rohatec, Czech Republic, is indeed B. vallesianus.

Specific sequence features and phylogenetic relationships

Nucleotide sequences of the ITS1–5·8S–ITS2 rDNA region for eight independent Bvallesianus isolates from Rohatec were determined. Because no sequence variations were observed between these isolates, a single sequence was deposited in the GenBank database (Benson et al., 2011) under accession number HM756288. The sequencing of the cloned rDNA fragment revealed that the exact size of the amplified ITS1–5·8S–ITS2 rDNA segment was 980 bp, as expected for Bvallesianus species. The nucleotide composition of the amplified ITS rDNA segment from the Czech isolate of Bvallesianus was: 22·7% A, 18·9% C, 26·1% G and 32·4% T, and showed 45% GC content.

Using the determined nucleotide sequence as a query to search in GenBank, the highest similarities were observed for all deposited Bvallesianus isolates. The determined values of genetic distances between all currently known Bvallesianus isolates and some representative members of the Bsexdentati group are summarized in Table 2. The results show that all Bvallesianus ITS1 and ITS2 sequences are highly conserved, with the exception of one Bvallesianus isolate (GenBank accession: AM269921) that shows several unique single-point substitutions. Interestingly, the genetic distance matrixes show that the Czech isolate of Bvallesianus differs from all other species belonging to the Bsexdentati group in the 5·8S rRNA-coding sequence (Table 2).

Table 2. Distance matrix based on the ITS1 (a), 5·8S rRNA (b) and ITS2 (c) sequence alignments for all currently known Bursaphelenchus vallesianus isolates and some selected members of the Bsexdentati group
 Bvallesianus (HM756288)Bvallesianus (AM160662)Bvallesianus (AM269920)Bvallesianus (AM269918)Bvallesianus (AM269919)Bvallesianus (AM269922)Bvallesianus (AM269921)Bpoligraphi (AM179512)Bpinophilus (AM160664)Bsexdentati (AM269915)Bborealis (AM179511)
  1. Note that for each matrix, the values are pairwise P-distances, which represent the divergence of all bases between two sequences. The number following species name at each nucleotide sequence represents GenBank accession number. The newly sequenced Czech isolate of Bvallesianus is indicated in bold.

(a) ITS1
 Bvallesianus (HM756288)          
 Bvallesianus (AM160662)0·000         
 Bvallesianus (AM269920)0·0000·000        
 Bvallesianus (AM269918)0·0000·0000·000       
 Bvallesianus (AM269919)0·0000·0000·0000·000      
 Bvallesianus (AM269922)0·0000·0000·0000·0000·000     
 Bvallesianus (AM269921)0·0070·0070·0070·0070·0070·007    
 Bpoligraphi (AM179512)0·0410·0410·0410·0410·0410·0410·041   
 Bpinophilus (AM160664)0·0300·0300·0300·0300·0300·0300·0370·037  
 Bsexdentati (AM269915)0·0340·0340·0340·0340·0340·0340·0410·0650·048 
 Bborealis (AM179511)0·0550·0550·0550·0550·0550·0550·0620·0270·0510·076
(b) 5·8S rRNA
 Bvallesianus (HM756288)          
 Bvallesianus (AM160662)0·026         
 Bvallesianus (AM269920)0·0260·000        
 Bvallesianus (AM269918)0·0260·0000·000       
 Bvallesianus (AM269919)0·0260·0000·0000·000      
 Bvallesianus (AM269922)0·0260·0000·0000·0000·000     
 Bvallesianus (AM269921)0·0260·0000·0000·0000·0000·000    
 Bpoligraphi (AM179512)0·0260·0000·0000·0000·0000·0000·000   
 Bpinophilus (AM160664)0·0260·0000·0000·0000·0000·0000·0000·000  
 Bsexdentati (AM269915)0·0260·0000·0000·0000·0000·0000·0000·0000·000 
 Bborealis (AM179511)0·0260·0000·0000·0000·0000·0000·0000·0000·0000·000
(c) ITS2
 Bvallesianus (HM756288)          
 Bvallesianus (AM160662)0·000         
 Bvallesianus (AM269920)0·0000·000        
 Bvallesianus (AM269918)0·0000·0000·000       
 Bvallesianus (AM269919)0·0000·0000·0000·000      
 Bvallesianus (AM269922)0·0000·0000·0000·0000·000     
 Bvallesianus (AM269921)0·0030·0030·0030·0030·0030·003    
 Bpoligraphi (AM179512)0·1790·1790·1790·1790·1790·1790·175   
 Bpinophilus (AM160664)0·0660·0660·0660·0660·0660·0660·0690·189  
 Bsexdentati (AM269915)0·0660·0660·0660·0660·0660·0660·0690·1720·093 
 Bborealis (AM179511)0·1820·1820·1820·1820·1820·1820·1790·0900·2000·200

To get insights into the phylogenetic relationships between the Czech isolate of Bvallesianus and other Bursaphelenchus species, a molecular phylogeny approach was applied. The phylogenetic tree based on ITS1–5·8S–ITS2 rDNA sequence data showed overall clustering of Bursaphelenchus species into several groups, as previously demonstrated (Beckenbach et al., 1999; Lange et al., 2006). The Czech isolate of Bvallesianus was positioned in the clade shared by all members of the Bsexdentati group including species Bborealis, Bpinophilus, Bpoligraphi, Bsexdentati and Bvallesianus (Fig. 3a). Unexpectedly, the Czech Bvallesianus did not occupy the same subclade as the other Bvallesianus isolates, but formed a separate sister clade (bootstrap value 98%) to all species occupying the Bsexdentati group (Fig. 3a).

Figure 3.

Phylogenetic relationships and some unique sequence features of Bursaphelenchus vallesianus from the Czech Republic. (a) ML-tree built up from ITS1–5·8S–ITS2 rDNA sequence data. The sequences were aligned with the help of clustalX. All positions containing gaps and missing data were eliminated from the data set, and phylogenetic analysis was conducted in mega 4. Species Aphelenchoides besseyi and Abicaudatus were used as out-groups. Colour circles at branches denote bootstrap values (out of 500 replicates) as explained in legend. Numbers in the brackets indicate the GenBank accession number, and colour squares at each nematode species indicate taxonomical classification as explained in legend (top frame). The newly sequenced Czech isolate of Bvallesianus is indicated in bold. (b) Molecular features distinguishing the Czech isolate of Bvallesianus from all other species belonging to the Bsexdentati group. Schematic representation of the rDNA gene unit (top scheme) with a set of aligned partial 5·8S rRNA gene sequences. The alignment was constructed using clustalX, and its colouring was performed in jalview. Orange circles over the aligned sequences depict positions of specific substitutions.

Careful examination of the nucleotide data set used for phylogeny inference revealed several substitutions that distinguish the Czech isolate of Bvallesianus from other members of the Bsexdentati group including other Bvallesianus isolates. The presence of these substitutions resulted in the Czech isolate of Bvallesianus being positioned in a separate sister clade to other Bsexdentati group-occupying species in the phylogenetic reconstruction. When nucleotide positions with these substitutions were omitted from the phylogenetic analysis, the Bvallesianus perfectly clustered with other Bvallesianus isolates within the designated Bsexdentati group (Fig. 4). Interestingly, the identified substitutions distinguishing the Czech isolate of Bvallesianus from the others were not found in the evolutionary divergent, non-coding ITS regions as assumed, but in the 3′-end region of the evolutionary conserved 5·8S rRNA-coding gene (Fig. 3b). The data collected so far thus indicates that the Czech isolate of Bvallesianus exhibits a newly identified genetic variation in the species.

Figure 4.

Phylogenetic relationships within the Bursaphelenchus sexdentati group, when the sequence positions encompassing four-nucleotide exchanges in the 5·8S rRNA-coding sequence were omitted from the analysis. Maximum likelihood tree built up from ITS1–5·8S–ITS2 rDNA sequence data. Numbers at branches denote bootstrap values (out of 500 replicates). Numbers in the brackets indicate the GenBank accession number. The newly sequenced Czech isolate of Bvallesianus is indicated in bold font.

Structural implications of the nucleotide substitutions in the 5·8S rRNA

In general, the 5·8S rRNA gene sequence is highly conserved between related species due to its biological function in the ribosome structure, unlike for example non-coding ITS elements intercalated between rRNA-coding genes (Fig. 3b). Up to now, all known species belonging to the Bsexdentati group share an entirely identical sequence motif, GAN6TC, where N6 represents six nucleotide positions, in the 3′-end region of the 5·8S rRNA-coding gene (Fig. 3b). This sequence motif can also be found in some other Bursaphelenchus species such as Bxylophilus group-occupying members and some others (Fig. 3b). Unexpectedly, the Bvallesianus isolate from the Czech Republic contains a different sequence motif, CGN6GT, in the 3′-end of the 5·8S rRNA. This sequence motif is typical for most Bursaphelenchus species, which belong to neither the Bsexdentati nor the Bxylophilus groups (Fig. 3b).

Using an in silico RNA structure prediction approach, it was revealed that the observed substitutions in the 5·8S rRNA probably affect its folding. The Bursaphelenchus species with the CGN6GT motif in the 3′-end of the 5·8S rRNA, where the Czech isolate of Bvallesianus also belongs, form a 5·8S rRNA structure with three, intramolecular helical domains (I to III; Fig. 5a). In contrast, the other Bursaphelenchus species containing the motif GAN6TC, characteristic for the Bsexdentati group members, show 5·8S rRNA structures with two intramolecular helical domains (I and II; Fig. 5b). More interestingly, it has been previously shown that 3′-end region of 5·8S rRNA forms a stable duplex region with the 5′-end of 28S rRNA in the 5·8S–ITS2–28S pre-RNA molecule, preceding excision of the ITS2 segment (Ellis et al., 1986). Here, it is shown that the species from Bursaphelenchus with the sequence motif GAN6TC form the 5·8S–28S rRNA duplex with an internal 2-nucleotide mismatch (Fig. 5b), while perfect complementary RNA–RNA pairing was observed for the other Bursaphelenchus species (Fig. 5a). Biological implications of these structural differences remain unclear.

Figure 5.

Inferred secondary structures of the 5·8S rRNA for Bursaphelenchus vallesianus from the Czech Republic (a) and other species of the Bsexdentati group including Bsexdentati, Bvallesianus, B. pinophilus, Bborealis and Bpoligraphi (b). Note that the presence of the four-nucleotide exchanges (marked with asterisks) in the 3′-end of the 5·8S rRNA appears to affect the 5·8S rRNA folding and its interaction with the 28S rRNA. The models were computed by an energy-minimization approach using the Mfold and final adjustments of folded structures were performed with RNAviz. Bonds between C:G and U:A nucleotides indicated by a straight line; bonds between U:G indicated by a closed circle.

Both ITS1 and ITS2 transcripts appear to fold into the multibranched closed loop

Previously, it has been proven in eukaryotic organisms including nematodes that the ITS2 transcript, which is part of the 5·8S–28S pre-RNA molecule, tends to fold into a multibranched closed loop (Subbotin et al., 2005; Marek et al., 2010). Using an energy-minimization approach, this study shows that the ITS2 transcript of the Czech isolate of Bvallesianus also folds into the multibranched loop with three helical domains (I to III) and with the stem (basal) helix that is formed by perfect complementary pairing of 5·8S and 28S rRNA molecules (Fig. 6a). The helical domains I and II are relatively large and further manifold branched. On the other hand, the helical domain III is the shortest from the three paired regions, and it is important to note that the 5′-end of 28S rRNA is involved in its formation (Fig. 6a). In addition, within the domain II several U-U mismatches can be found (Fig. 6a). Comparative analysis of ITS2 structures between all species belonging to the Bsexdentati group showed that the Czech isolate of Bvallesianus shares similar structural topology of pairing domains with B. vallesianus, Bsexdentati and Bpoligraphi (Fig. 6). The only substantial difference is the symmetric bulge (2-nucleotide mismatches) within the stem 5·8S–28S duplex. The other two species, Bborealis and Bpinophilus, probably fold ITS2 transcripts in different structural topologies (Fig. 6d,e). The reconstructions predicted six helical domains within the ITS2 structure for Bborealis and seven helical domains in Bpinophilus ITS2 (Fig. 6d,e).

Figure 6.

Secondary structure models of the internal transribed spacer 2 (ITS2) in 5·8S-28S pre-rRNA molecule for Bursaphelenchus vallesianus from the Czech Republic (a), originally described Bvallesianus (b), Bsexdentati (c), Bborealis (d), Bpinophilus (e) and Bpoligraphi (f). The models were computed by an energy-minimization approach using the Mfold program and final adjustments of folded structures were performed with RNAviz. Bonds between C:G and U:A nucleotides indicated by a straight line; bonds between U:G indicated by a closed circle. The U-U mismatches (asterisks) are highlighted. Black arrowheads indicate position of internal bulge (mismatch region) in the stem helix occurring in all Bsexdentati group members (b–f) with the exception of Bvallesianus from the Czech Republic (a).

The same approach was also applied to predict the secondary structure of the ITS1 transcript. Surprisingly, the ITS1 appears to fold into the multibranched closed loop similar to the ITS2 transcript (Fig. 7). While the ITS2 closed loop is formed with the help of the 5·8S and 28S rRNA sequences, the ITS1 forms the closed loop without any help of flanking 18S and 5·8S rRNA sequences. Structural comparison revealed that Bvallesianus, B. sexdentati, Bpinophilus and Bborealis form the closed loop with five helical domains (I to V; Fig. 7). Fusion of domains IV and V can explain why only four helical domains (I to IV) are predicted in Bpoligraphi (Fig. 7).

Figure 7.

Secondary structure models of the internal transribed spacer 1 (ITS1) for Bursaphelenchus vallesianus from the Czech Republic (a), originally described Bvallesianus (b), Bsexdentati (c), Bpinophilus (d), Bborealis (e) and Bpoligraphi (f). The models were computed by an energy-minimization approach using the Mfold program and final adjustments of folded structures were performed with RNAviz. Bonds between C:G and U:A nucleotides indicated by a straight line; bonds between U:G indicated by a closed circle.

Discussion

This paper determines the nucleotide sequence of the nuclear ITS1–5·8S–ITS2 region of Bvallesianus, a recently identified species in the Czech Republic. Comparative molecular analysis based on the sequence data was performed between the Czech isolate of Bvallesianus and other closely related nematode species belonging to the Bsexdentati group.

Morphological observations and morphometric measurements of Bvallesianus specimens isolated from declining pine trees in Rohatec (Czech Republic) showed perfect agreement with previously published descriptions of the species by Braasch et al. (2004) and Akbulut et al. (2008) (Fig. 1; Table 1). As shown in the restriction analysis of the amplified ITS rDNA region, no molecular features were detected which distinguished the Czech isolate of Bvallesianus from other currently known Bvallesianus isolates (Fig. 2). However, comprehensive sequence analysis identified a four-nucleotide exchange specific to the Czech isolate of Bvallesianus (Fig. 3b). This specific sequence motif (CGN6GT), which surprisingly was found in the 3′-end of the 5·8S rRNA-coding sequence, is unique for the Czech isolate of Bvallesianus as a member of the Bsexdentati group. All other species belonging to the Bsexdentati group contain a different sequence motif (GAN6TC) that can also be found in the Bxylophilus group members. Due to the presence of this four-nucleotide exchange, molecular evolutionary reconstructions resulted in a phylogenetic tree where the Czech isolate of Bvallesianus was separated as a sister clade to the Bsexdentati group members (Fig. 3a). When the nucleotide positions carrying these exchanges were omitted from phylogenetic analysis, the Czech isolate of Bvallesianus was grouped together with other Bvallesianus isolates into the Bsexdentati group (Fig. 4), as expected. Thus, the sequence characterization results provide new evidence about genetic variation distinguishing the Czech isolate of Bvallesianus from other previously described isolates in Switzerland (Polomski et al., 2006) and China (Gu et al., 2006).

Structural modelling showed that the presence of the four-nucleotide exchange in the 3′-end of the 5·8S rRNA-coding gene probably affects the 5·8S rRNA folding (Fig. 5). The bursaphelenchid 5·8S rRNA molecules with the GAN6UC motif in the 3′-end region, including the Czech isolate of Bvallesianus, fold into three intramolecular domains (I to III) and one intermolecular helix with 28S rRNA (Figs 5 & 6). In contrast, the other species including all members of the Bsexdentati group fold their 5·8S rRNA molecules into two main intramolecular domains (I and II) and one intermolecular 5·8S–28S helix (Figs 5 & 6).

Previously, it has been shown that the ITS1 structure is conserved at higher systematic levels, and that it mostly consists of an open multibranched loop with several helical domains (Mullineux & Hausner, 2009; Ullrich et al., 2010; Subbotin et al., 2011). Despite these facts, this study shows for the first time, as far as is known, that the bursaphelenchid ITS1 transcript tends to fold into the multibranched, self-closing loop (Fig. 7). While the number and topology of internal helical domains varies between individual Bursaphelenchus species, the stem helix mediating self-closing of the ITS1 structure is highly conserved (Fig. 7). Secondary structure predictions showed that the ITS1 5′- and 3′-end sequences do not seem to participate in intramolecular interactions; however, they can be involved in tertiary folding and/or in pairing with flanking rRNA molecules. However, these interactions remain to be discovered.

In agreement with previous reports on nematode ITS2 structures (Marek et al., 2010; Subbotin et al., 2011), this paper shows that the bursaphelenchid ITS2 transcript also folds into a multibranched closed loop. Unlike ITS1, the ITS2 structure is closed by the stem helix formed by a canonical 5·8S and 28S rRNA pairing (Fig. 6). More precisely, while the ITS2 stem helix (5·8S-28S duplex) of the Czech isolate of Bvallesianus exhibits perfect complementary pairing, the stem helices of other species belonging to the Bsexdentati contain an internal bulge (Fig. 6). This bulge, distinguishing the Czech isolate of Bvallesianus from other closely related species, is caused by the presence of the identified sequence variation in the 3′-end of the 5·8S rRNA (Figs 3 & 5).

Collectively, the comparative studies here provide new insights into the genetic variability of pine tree-inhabiting nematodes belonging to the Bsexdentati group. The results obtained on primary and secondary structures of the 5·8S rRNA and non-coding ITS segments give novel basic data, both for reconstruction of molecular evolution within the genus Bursaphelenchus, and also for analysing functions in the ribosome genesis.

Acknowledgements

This research was financed by a research grant from the Ministry of Agriculture of the Czech Republic (MZE-0002700603), and by a research grant from the Ministry of Education of the Czech Republic (MSM 604607901).

Ancillary