Morphological and molecular characterisation, and phylogenetic position of X. browni sp. n., X. penevi sp. n. and two known species of Xiphinema americanum-group (Nematoda, Longidoridae)

Abstract Using ribosomal (18S, ITS1, ITS2, D2-D3 expansion segments of 28S rDNA) and mitochondrial (partial cox1 and nad4) DNA markers in a study of several populations of Xiphinema americanum-group from Europe and Morocco, two cryptic species Xiphinema browni sp. n. (formerly reported as Xiphinema pachtaicum) and Xiphinema penevi sp. n. were revealed. The species are described, illustrated and their phylogenetic relationships discussed. The first species is most similar to Xiphinema parasimile and is a member of Xiphinema simile species complex. The phylogenetic reconstructions inferred from three molecular markers (18S, D2-D3 28S rDNA and cox1) showed that Xiphinema penevi sp. n. is part of Xiphinema pachtaicum-subgroup and is closely related to Xiphinema incertum, Xiphinema pachtaicum, Xiphinema parapachydermum, Xiphinema plesiopachtaicum, Xiphinema astaregiense and Xiphinema pachydermum. Also, a separate “Xiphinema simile-subgroup”, outside the Xiphinema pachtaicum-subgroup and so far consisting only of the parthenogenetic species Xiphinema simile, Xiphinema parasimile, Xiphinema browni sp. n. and probably Xiphinema vallense was formed. New primers for amplification and sequencing of part of the nad4 mitochondrial gene were designed and used.


Introduction
The Xiphinema americanum-group is a well defined natural complex of species (Lamberti et al. 2000, Coomans et al. 2001, He et al. 2005b) with high significance to agriculture caused by the ability of several species to transmit economically important plant viruses (McFarlane et al. 2002), although there are controversial opinions defining the group (Archidona-Yuste et al. 2016). Even for experienced nematologists species delimitation within this group is challenging because they have rather similar morphology and metrics, and the existing keys (Lamberti et al. 2000 do not always allow species differentiation and identification. During the last decade wide usage of DNA sequencing in Xiphinema taxonomy including this group revealed the existence of a number of cryptic species (Gutiérrez-Gutiérrez et al. 2010, Archidona-Yuste et al. 2016. This was the case with several populations from the Czech Republic and Slovakia (Kumari et al. 2005(Kumari et al. , 2010b originally identified as X. pachtaicum (Tulaganov, 1938) and one population from Morocco provisionally also determined as X. pachtaicum. The objectives of the present study were: i) to characterise populations from the Czech Republic, Slovakia and Morocco both morphologically and genetically; ii) to sequence populations of X. pachtaicum and X. parasimile  from Bulgaria for comparison; iii) to clarify phylogenetic relationships of identified species using ribosomal and mithochondrial DNA.

Sampling, nematode isolation and processing
The Xiphinema specimens examined originated from various localities in the Czech Republic (Kurdějov, Mohyla míru and Sokolnice, grapevines), Slovakia (Moča, grapevine), Bulgaria (Balgarene village, pear tree, Vinogradets vicinity, vineyard) and Morocco (Ifrane, holm oak tree). Details of the soil sampling, nematode isolation and processing for Czech and Slovakian populations are given in Kumari et al. (2005Kumari et al. ( , 2010b. A decanting and sieving technique was used for extracting nematodes from soil samples from Bulgaria and Morocco. Xiphinema specimens recovered were heat killed at 55°C for two minutes, fixed in a 4% formalin, 1% glycerol solution, processed to anhydrous glycerol (Seinhorst 1959), and mounted on glass microscope slides. Drawings were prepared using an Olympus BX51 compound microscope with differential interference contrast (DIC). Photographs were taken using an Axio Imager.M2-Carl Zeiss compound microscope with a digital camera (ProgRes C7) and specialised software (CapturePro Software 2.8). Measurements were made using an Olympus BX41 light microscope, a digitising tablet (CalComp Drawing Board III, GTCO CalCom Peripherals, Scottsdale, AZ, USA), and computer Digitrak 1.0f programme (Philip Smith, Scottish Crop Research Institute, Dundee, UK).

DNA extraction, amplification and sequencing
Individual nematodes from Bulgaria, Morocco (DESS-preserved), Czech  were mounted on temporary slides containing glass beads and after taking measurements and photomicrographs the slides were dismantled, individual nematodes removed, and added in 0.25 M NaOH to digest overnight and thereafter heated to 99°C for 3 min. Afterwards 10 μl of 0.25 M HCl, and 5 μl each of 0.5 M Tris-HCl (pH 8) and 2% Triton X-100 were added and the mixture was incubated for another 3 min at 99°C (Stanton et al. 1998). Finally, the DNA suspension was cooled and the DNA was either used directly for PCR or stored at -20°C until template was needed for PCR reactions. Genomic DNA which was prepared by Kumari et al. (2010b) was also used in this study.
Six regions (18S, ITS1, ITS2, D2-D3 expansion segments of 28S, cox1 and nad4) of ribosomal and mitochondrial DNA were amplified and sequenced. Primer sequences and references to the primers are given in Table 1. The 18S gene of the Czech population was amplified by using primers SSU_F_04+SSU_R_09 (first fragment), SSU_F_22+SSU_R_13 (second fragment) and SSU_F_23+SSU_R_81 (third fragment). The 18S gene of other populations was amplified by using primer combination 988F+1912R (first fragment) and 1813F+2646R (second fragment).
Initially partial nad4 gene was amplified with the primers CDF+CDR but only one specimen was amplified using these primers. A pair of new primers (nadpachF+nadpachR) was designed using online software PRIMER 3 (http://frodo. wi.mit.edu/) from the sequences which were amplified by (CDF+CDR). For final analysis all specimens and populations of X browni sp. n. from the Czech Republic and Slovakia were amplified and sequenced by using nadpachF + nadpachR primers.
The PCR reaction was performed in 25 μl total volume containing 1 PCR bead (GE Healthcare, Buckinghamshire, UK), 20.5 μl double distilled sterile water, 2.0 μl of each primer (10pmol/μl) (synthesized by Generi Biotech, Hradec Králové, Czech Republic), and 0.5 μl of DNA added as a template for PCR. A negative control (sterilized water) was included in all PCR experiments. The cycling profile for all ribosomal DNA and mtD-NA markers was as described by Kumari and Subbotin (2012) and by He et al. (2005a), respectively. All PCR reactions were performed in a DNA Engine PTC-1148 thermal cycler (Bio-Rad). Aliquots of PCR were analysed by gel electrophoresis and the remaining products were purified using High Pure Product Purification kit (Roche Diagnostics GmbH, Mannheim, Germany) and sequenced in both directions using each primer pair one forward and one reverse (Macrogen, Netherlands). Sequencher TM 4.8 (Genes codes. Corp., Ann Arbor, MI, USA) was used to assemble and view each sequence and check for base-calling errors. Accession numbers of all sequences are given in Table 2.

Sequence and phylogenetic analyses
A BLAST (Basic Local Alignment Search Tool) search at NCBI (National Center for Biotechnology Information) was performed using the obtained sequences as queries to confirm their nematode origin and to identify the most closely related nematode sequences. Sequences revealing high similarity to those obtained here were included in the phylogenetic analyses of both ribosomal and mitochondrial gene regions , He et al. 2005b, Gozel et al. 2006, Holterman et al. 2006Lazarova et al. 2006, Kumari et al. 2009, Gutiérrez-Gutiérrez et al. 2010, Kumari et al. 2010a, Kumari et al. 2010b, De Luca and Agostinelli 2011, Gutiérrez-Gutiérrez et al. 2011a, Gutiérrez-Gutiérrez et al. 2011b, Meza et al. 2011, Sakai et al. 2011, Gutiérrez-Gutiérrez et al. 2012, Kumari and Subbotin 2012, Sakai et al. 2012, Kumari and Cesare 2013, Tzortzakakis et al. 2014, Getaneh et al. 2015. Sequence numbers are presented in the trees. The multiple sequence alignments (MSA) of all datasets were performed using the GUIDANCE2 Server available at http://guidance.tau.ac.il/ (Sela et al. 2015). All three alignment algorithms (MAFFT, PRANK and ClustalW) were tested and the MSAs having highest alignment confidence scores were used for ITS phylogenetic reconstructions. Subsequently, the MSAs were manually optimised and trimmed using MEGA 6 (Tamura et al. 2013). The phylogenetic reconstructions were performed using the Bayesian Inference (BI) algorithm implemented in MrBayes 3.2.5. (Huelsenbeck and Ronquist 2001;Ronquist et al. 2012) using the General Time Reversible model plus Gamma distribution rates (GTR + G). The Bayesian MCMC tree searches were run using default heating parameters for 2 000 000 generations with a sample frequency of 1000 generations. The first 25% of the chains discarded as burning and the remaining 75% trees kept to summarise the tree topology, branch lengths,  Tables 3-5. Description. Females. Body slender C to open spiral shaped. Cuticle with fine transverse striae. Thickness of the cuticle at postlabial region 1-1.5 μm, 1.5 rarely 2 μm at mid-body and 2 μm at post-anal region. Labial region set-off from the rest of the body by a constriction, expanded, rounded laterally, 5.0±1.1 (4-7) μm high. Amphideal fovea hardly visible, funnel-shaped, its opening c. 5 μm (50%) wide visible posterior the constriction level. Distance between first and second guide ring in specimens with retracted odonostyle 5-10 μm long. Odontophore with moderately developed basal flanges 6.1±0.6 (5.5-7) μm wide. A small vestigium observed occasionally in slender part of pharynx. Pharyngeal characters presented at Table 4. Dorsal pharyngeal gland nucleus 2 μm diam. Ventrosublateral nuclei barely visible. Rectum 20.8 ± 1.5 (18-23) μm, n=7, or c 1.3 times anal body diameter. Reproductive system amphidelphic, symbiont bacteria present in the ovaries. Separate uteri and ovejector present (Table 5), oviduct 90.5±13.0 (68-101) μm; vagina bell-shaped 39.5% of the corresponding body width (33-50%, n=14), vulva post-equatorial. Numerous sperm observed in one female from Kurdějov (Figs 2B, 4B). Tail conical, dorsally convex, ventrally straight or slightly concave with narrowly rounded to pointed terminus. Two pairs of caudal pores.

Measurements. See
Male. Very rare. One specimen found in Sokolnice population. Male similar to the female with posterior region more strongly curved. Lip region and tail shape as in females, differences were observed within body width and tail length, which reflected a and c' values. Spicules robust, slightly curved, lateral guiding piece 7 μm long. Adanal pair preceeded by a row of 5 irregularly spaced supplements, the two anteriormost weakly developed. Tail conoid, ventrally straight, dorsally convex with pointed terminus, caudal pores not visible. The slide of the only male specimen, described by Kumari et al. (2005), was subsequently damaged.
Juveniles. The scatter diagram based on functional and replacement odontostyle, and body length revealed the presence of four juvenile stages (Fig. 8). Tail shape and length similar in all stages and females with c' slightly decreasing in successive stages (Kumari 2005, Fig. 3, Table 3).
Type material. The holotype, 9 paratype females and juveniles from all stages are deposited in the nematode collection of the Institute of Biodiversity and Ecosystem Research, Sofia, Bulgaria. Other paratypes deposited as follows: 15 females in the Crop Research Institute, Prague, the Czech Republic; 5 females in the USDA Nematode Collection, Beltsville, Maryland, USA; 5 females in the Nematode Collection of the Institute of Plant Protection, Bari, Italy; 5 females in the Wageningen Nematode Collection (WANECO), Wageningen, the Netherlands. The ribosomal and mtDNA sequences (18S rDNA, ITS1, ITS2, D2-D3, cox1, nad4) of X. browni sp. n. are deposited in GenBank (for accession numbers see Table 2).
Sequence and phylogenetic analyses. There was no sequence variation between populations for 18S and D2-D3, ITS1 and ITS2 rDNA regions of X. browni sp. n. Of all four populations studied cox1 region of three population from the Czech Republic (Kurdějov, Mohyla Míru, Sokolnice) were sequenced by Kumari et al. (2010b) and all populations were identical therefore only one population was submitted to Gen-Bank (accession number GU222424). The Slovakian population was sequenced in this study and it was identical to previously published sequence of Kurdějov the population identified as X. pachtaicum (GU222424, Kumari et al. 2010b). All four sequenced populations were also identical for nad4 part.
BLAST at NCBI using 18S and D2-D3 region sequences as queries revealed highest similarity (99 and 87%) to the corresponding sequences of X. simile Lamberti, Choleva & Agostinelli, 1983 from Serbia (AM086681) and two Spanish populations of X. opisthohysterum Siddiqi, 1961 (JQ990040 and KP268967), respectively. The es-  timated divergences (p-distance) between the 18S rDNA sequences of the new species and the closest species, X. parasimile from Bulgaria (this study) and X. simile from Serbia (AM086681) were 0.3 (6 nt) and 1.2% (21 nt), respectively. Again, the new D2-D3      sequence of X. parasimile from Bulgaria was most similar (p-distance = 4.6%), followed by the Serbian populations of X. parasimile (p-distance = 7.6-7.9%, calculated for D2 region only) and various populations of X. simile (14.1-14.7%). The partial cox1 sequences of X. browni sp. n revealed highest similarity to X. simile from Slovakia (AM086708). Surprisingly, these two species showed very high similarity 99% (2 nts difference) in cox1 sequences and higher dissimilarity in 18S rDNA (p-distance = 1.2%, 21 nts). Other authors (Gutiérrez-Gutiérrez et al. 2012) have also reported similar observation namely, 100% identity in cox1 part of two different species X. duriense Lamberti, Lemos, Agostinelli & D'Addabo, 1993 (JQ990053) and X. opistohysterum (JQ990054) and clear separation in D2-D3 28S sequences (or 96 % identity). Further, the cox1 sequences of X. browni sp. n. and the closest species X. parasimile, X. simile (GU222425, Czech Republic) and X. pachtaicum (HM921369, Spain) were translated to amino acids and aligned (Fig. 9). The estimated p-distances between X. browni sp. n. and the three species were 10.1%, 21.7% and 23.3%, respectively. In all three phylogeny reconstructions (18S, D2-D3 and cox1) X. parasimile from Bulgaria was a sister species of X. browni sp. n. and both species were part of a well supported clade with other European populations of X. simile (Figs 10-12). The recently described species X. vallense Archidona-Yuste, Navas-Cortes, Cantalapiedra-Navarrete, Palomares-Rius & Castillo, 2016 presented only with D2-D3 and ITS1 rDNA sequences seems also to be evolutionary very closely related (Figs 11 and 13), however amplifying additional sequences for other molecular markers (e.g. 18S and cox1) could help to better clarify its relationships. The position of the new species in the phylogeny trees based on ITS1 and ITS2 sequences was unstable (Figs 13 and  14). The analyses resulted in various tree topologies when using different alignment algorithms and reconstruction methods (ML and BI) and because of the absence of homologous sequences from closely related species. In most cases X. browni sp. n. was part of a clade of European X. americanum-group species considered as group II in a previous publication (Archidona-Yuste et al. 2016). Due to insufficient number of nad4 sequences of species belonging to the X. americanum-group at NCBI no phylogenetic reconstructions are presented.
Etymology. The species is named after Prof Derek JF Brown, an outstanding nematologist, for his significant contributions to the knowledge of plant parasitic nematodes and the development of nematology in Bulgaria. Description.
Male. Not found.
Juveniles. The scatter diagram based on functional and replacement odontostyle, and body length revealed the presence of four juvenile stages (Fig. 23). As in most species of the X. americanum-group there is a gradually decreasing of c' values with successive stages which reflects increasing body width while the tail length is more or less similar in juveniles and adults.
Type material. The holotype, 7 paratype females and juveniles from all stages are deposited in the nematode collection of the Institute of Biodiversity and Ecosys-   Males (number of supplements) rare or absent 5 frequent 4-5 rare or absent 5-6 rare or absent 5 males abundant 5 rare or absent 3-5 Males abundant Rare or absent 6, 7 tem Research, Sofia, Bulgaria. Other paratypes deposited as follows: 2 females in the USDA Nematode Collection, Beltsville, Maryland, USA; 2 females in the Nematode Collection of the Institute of Plant Protection, Bari, Italy; 1 female in the Wageningen Nematode Collection (WANECO), Wageningen, the Netherlands. Three ribosomal sequences (18S, ITS2 and D2-D3) of X. penevi sp. n. are deposited in GenBank (for accession numbers see Table 2).

Etymology.
The new species is named after Dr Lyubomir Penev, an internationally recognised publisher and authority in entomology and ecology as acknowledgement of his invaluable help and support provided to one of the authors (VP) in her research activities. Figures 15-18 Measurements. Tables 3-6.

Xiphinema pachtaicum (Tulaganov, 1938) Kirjanova, 1951
Note. Xiphinema pachtaicum has been recorded from Bulgaria and data on its morphology are available in previous studies (Lamberti et al. 1983;Peneva and Choleva 1992); here we present additional morhometric data only for the population from Balgarene together with illustrations, LM micrographs and sequence data (Table 2). It is common and associated with a wide spectrum of cultivated and wild plants (Lamberti and Siddiqi 1977). Barsi & Lamberti, 2004 Figures 15, 17, 18 Morphometric data and detailed description of X. parasimile from Bulgaria are reported previously (Lazarova et al. 2008). For the Vinogradets population two ribosomal and one mitochondrial DNA sequences were obtained (Table 2). Xiphinema parasimile has a limited distribution in Bulgaria (Lazarova et al. 2008).

Sequence and phylogenetic analyses
Three rDNA sequences were obtained for the Bulgarian X. pachtaicum population (18S, D2-D3 and ITS2) with BLAST showing identity or very high similarity to other X. pachtaicum populations available at NCBI (100% for 18S, 99/100% for D2-D3 and 98% for ITS2). Further, the DNA sequences of X. parasimile from Vinogradets (18S, D2-D3 and cox1) showed highest similarity to X. simile from Serbia (99% for 18S), various other populations of X. simile and X. opisthohysterum (88%, D2-D3) and 78% two cox1 sequences -X. pachtaicum from the Czech Republic (GU222424) and X. simile from Slovakia (AM086708). The first one is the previously published sequence of X. browni sp. n. identified as X. pachtaicum (Kumari et al. 2010b). The D2 28S rDNA region was further compared to the Serbian population of X. parasimile (D2 part of sequences AM490214, AM490217, Barsi and De Luca 2008) and the alignment showing the different nucleotides is presented (Fig. 24). The p-distance calculated for D2 part only was 1.8-2.1% that might indicate that X. parasimile population from Bulgaria could represent a cryptic species.
Based on the phylogenetic analyses performed (Figs 11-15) both new species described are members of two well-supported species complexes -X. simile and X. pachtaicum. The first subgroup includes X. simile, X. parasimile X. browni sp. n. and probably X. vallense. All occur in Europe and X. simile has also been reported from Central Africa (Liškova and Brown 1996, Coomans and Heyns 1997, Kumari 2006, Repasi et al. 2008, Lazarova et al. 2008, Bontă et al. 2012. Whether some of these records represent X. simile or closely related species requires new investigations using morphological discrimination and molecular markers. So far, X. parasimile has been recorded from the Balkan region , Lazarova et al. 2008, Bontă et al. 2012. Xiphinema browni sp. n. (previously reported as X. pachtaicum) seems to occur in central European countries. The second group of closely related species consists of X. pachtaicum, X. penevi sp. n., X. incertum, X, parapachydermum, X. plesiopachtaicum, X. astaregiense and X. pachydermum. Again, one of these species (X. pachtaicum) has a much wider distribution in Europe, Asia and Africa (Lamberti and Siddiqi 1977, Fadaei et al. 2003, Getaneh et al. 2015. Xiphinema incertum has been reported from Bulgaria, Serbia, Croaita and Spain, all other species have limited distributions -X. plesiopachtaicum, X. pachydermum, X. parapachydermum, X. astaregiense, reported only from Spain, the latter three species being amphimictic, and X. penevi sp. n. so far found only in north-western Africa (Sturhan 1983, Lamberti et al. 1983, Lamberti 2002, Gutiérrez-Gutiérrez, 2012).
Based on a hierarchical cluster analysis of morphometrics Lamberti and Ciancio (1993) distinguished five species subgroups, among them the X. pachtaicum-subgroup (IV) consisted of 8 species with five being described from Europe (X. fortuitum Roca, Lamberti & Agostinelli, 1987, X. incertum, X. madeirense, X. pachydermum and X. simile), one from North America (X. utahense Lamberti & Bleve-Zacheo, 1979), and one from Asia (X. opisthohysterum). Our analyses using ribosomal and mitochondrial DNA sequences currently available in GenBank and the two new species described in this study supports the delimitation of the "X. pachtaicum-subgroup", however it also includes X. incertum, X. pachtaicum, X. pachydermum and the recently described species X. parapachydermum, X. astaregiense, X. plesiopachtaicum and X. penevi sp. n. Phylogenetic reconstructions showed that X. madeirense, X. opisthohysterum, X. simile and X. uthahense are not part of this group, for X. fortutium no sequences are available. These results are in line with the findings of other recent studies on the X. americanum-group (Gutiérrez-Gutiérrez et al. 2012, Archidona-Yuste et al. 2016. Xiphinema simile (presented by two types of sequences for populations from Serbia and the Czech Republic in 18S rDNA and cox1 trees), X. parasimile and X. browni sp. n. formed a separate subgroup outside the X. pachtaicum-subgroup, so far consisting only of parthenogenetic species. Therefore we proposed this clade to be referred as the X. simile-subgroup. The recently described species X. vallense seems also evolutionary very closely related to this subgroup because of its high morphometric and DNA similarity, however amplifying additional sequences for other molecular markers (e.g. 18S and cox1) could help to clarify its relationships.