New evidence of nematode-endosymbiont bacteria coevolution based on one new and one known dagger nematode species of Xiphinema americanum-group (Nematoda, Longidoridae)

Three populations of Xiphinema primum n. sp. and two populations of X. pachtaicum were recovered from natural forests and cultural regions of northern Iran. Both species belong to the X. americanum-group and were characterized by their morphological, morphometric and molecular data. The new species, which was recovered in three locations, belongs to the X. brevicolle-complex and is characterized by 2124–2981 μm long females with a widely rounded lip region separated from the rest of the body by a depression, 103–125 μm long odontostyle, two equally developed genital branches with endosymbiont bacteria inside the ovary, which are visible under light microscope (LM), vulva located at 51.8–58.0%, the tail is 26–37 μm long with a bluntly rounded end and four juvenile developmental stages. It was morphologically compared with nine similar species viz. X. brevicolle, X. diffusum, X. incognitum, X. himalayense, X. luci, X. parabrevicolle, X. paramonovi, X. parataylori and X. taylori. The second species, X. pachtaicum, was recovered in two geographically distant points close to city of Amol. Molecular phylogenetic studies of the new species were performed using partial sequences of the D2-D3 expansion segments of the large subunit ribosomal RNA gene (LSU rDNA D2-D3), the internal-transcribed spacer rDNA (ITS = ITS1+5.8S+ITS2), and the mitochondrial cytochrome c oxidase I gene (COI mtDNA) regions. The Iranian population of X. pachtaicum was also phylogenetically studied based upon its LSU rDNA D2-D3 sequences. Both species were also inspected for their putative endosymbiont bacteria. Candidatus Xiphinematobacter sp. was detected from two examined populations of the new species, whereas the second endosymbiont bacterium, detected from three examined isolates of X. pachtaicum, was related to the plant and fungal endosymbionts of the family Burkholderiaceae. The phylogenetic analyses of the two endosymbiont bacteria were performed using partial sequences of 16S rDNA. In cophylogenetic analyses, significant levels of cophylogenetic signal were observed using both LSU rDNA D2-D3 and COI mtDNA markers of the host nematodes and 16S rDNA marker of the endosymbiont bacteria.


Introduction
The nematode species of the genus Xiphinema Cobb, 1913 [1] are migratory ectoparasites of roots in a wide range of plants. They cause damage through direct feeding on the root cells of host plants and also through the transmission of some pathogenic plant nepoviruses [2]. Species of this genus are traditionally and historically divided into two groups, those of the americanum-group, and those of non-americanum-group [3][4][5][6].
The X. americanum-group species are generally small nematodes (compared to non-americanum species; usually, 1-3 mm long females) that are unified by the morphological similarities in the lip region, a spiral body after relaxation, and in genital tracts characters (for the revised grouping characters, see Lamberti et al. [5]). Historically, Lima [7] and Tarjan [8] noted the species with these morphological similarities as the X. americanum-complex. Species delimitation has been an issue due to their close morphology and usually overlapping morphometric ranges [9] and thus yielded a "complexity" of their identification, and designation of the X. americanum-complex. Mean values of morphometric data ranges have recently been proposed for primary clustering of populations [10]. The aforementioned "complex", is nowadays known as X. americanum-group, mainly after exploiting high-throughput molecular techniques for species delimitation [11], which are now routinely applied [12][13][14][15][16][17][18]. On the other hand, available molecular data is lacking for most species of the group. Lamberti et al. [5,19] listed 49 species in this group, including some species synonymized by Luc [4]. Since 2004, 10 extra species have been added to that group (Table 1).
From an agricultural vantage point, some species of the group are regarded as quarantine pests. Indeed, in Europe, the European Economic Union has ratified funding for projects dealing with the identification of virus vector species [5]. Some species are also rated as A1 and A2 quarantine nematodes by the European and Mediterranean Plant Protection Organizations  [10,20]. The species X. pachtaicum (Tulaganov, 1938 [22]) Kirjanova, 1951 [23] seemingly occurs worldwide, as it is common in in Europe [24][25][26][27] and North America [10]. According to Pedram [28], it is common in Iran.
The ability of these nematodes to retain viable plant viruses in absence of host plants adds to their importance as agriculture pests. Indeed, nine species of genus Xiphinema have been proven to transmit viruses (genus Nepovirus, family Secoviridae [27]), but this could be an underestimate as the capacity of several described Xiphinema species to transmit viruses has not examined. By 2000, seven species of the americanum-group Xiphinemas were known as capable of transmitting plant pathogenic nepoviruses in natural or laboratory conditions [5] and the aforementioned research measures and funding allocation, emphasize the need for reliable species identifications in this group.
Besides the biological ability of X. americanum-group members to transmit viruses, their relationships with endosymbiont Candidatus Xiphinematobacter spp. belonging to subdivision 2 of the Verrucomicrobia is another interesting area of research, already documented and discussed [29][30][31][32][33]. The nematode species in the group have prokaryotic endosymbionts and reproduce by thelytokous parthenogenesis; therefore, endosymbiont bacteria are maternally inherited [31]. The bacteria primarily occupy the gut of juveniles, while their final destination isfemale's ovaries for transovarial vertical transmission through thelytokous parthenogenesis [30,[32][33]. Long-term association of the nematode with its endosymbiont and species-specific associations are already hypothesized [34]. Recent studies have emphasized identification [16] and cophylogenetic analyses of these endosymbiont bacteria with their host nematodes [10,34], to corroborate the long-term association and significant cophylogeny between nematode host and endosymbiont. In one of the aforementioned studies [34], a new endosymbiont, belonging to the plant and fungal endosymbionts of the family Burkholderiaceae Garrity, Bell & Lilburn, 2005 [35] was discovered. The cophylogenetic analyses of this group and the host nematodes also demonstrated significant association between phylogenies including speciesspecific associations (for a few exceptions in species specificity of the endosymbionts, see [34]). In a more recent study, Howe et al. [36] also showed the congruence between the host and endosymbiont phylogeny using genetic markers of varying evolutionary rates.
Several members of the Xiphinema americanum-group have a global distribution [5,6,11,14]. Currently nine species of this group have been reported from Iran [28]. During our surveys in two Mazandaran and Tehran provinces conducted in 2015-2017, which followed our previous studies on Longidoridae Thorne, 1935 [37] in the country (the history within [38,39]), several populations of X. americanum-group were recovered. Primary morphological studies revealed three of them could be conspecific and do not fit known species of the group thus representing a new species, described herein as Xiphinema primum n. sp. The other populations belonged to X. pachtaicum. Study with light microscopes revealed endosymbiont bacteria occupying ovaries of several females of both species. Accordingly, the present study aims to (i) describe X. primum n. sp. and characterize the recovered populations of X. pachtaicum using morphological and molecular data, (ii) characterize the endosymbiont bacteria of the two nematode species, and (iii) perform cophylogenetic analyses using selected LSU rDNA D2-D3 and COI mtDNA sequences of host nematodes and 16S rDNA sequences of their endosymbiont bacteria.

Ethics statement
Specific permissions were not needed for collecting the nematodes studied. All soil samples were collected from public regions and natural forests, that were not under protection and no endangered animal species were under protection in those regions. The garden samples were collected after taking the permission from the landowner. There was no need for an animal research ethics committee to approve this research.

Nematode materials
The X. americanum-group species/populations studied in the present research were recovered from soil samples collected from cultivated fields and natural forests in two Mazandaran and Tehran provinces using the flotation method [40] and direct extraction of the nematodes using a series of 20, 60 and 270 mesh sieves (USA standard mesh numbers), equal to 850, 250, 53 μm openings size, respectively.

Morphological observations
The hand-picked specimens were killed with 4% formalin solution, transferred to anhydrous glycerin according to De Grisse [41], and mounted on permanent slides. Photographs were taken using an Olympus DP72 digital camera attached to an Olympus BX51 microscope powered with differential interference contrast (DIC). Drawings of the new species were made using a drawing tube attached to the microscope and were redrawn using CorelDRAW software version 16.

Molecular profiles
For DNA extraction, each nematode was washed with distilled water, surface sterilized using sodium hypochlorite (0.5% v/v) for 5 min, or 5% (v/v) solution of commercial Dettol for 1 min., followed by washing in distilled water three times, transferred to an Eppendorf tube containing 1 μl proteinase K (CinnaGen, Tehran) (10 mg/ml) and 49 μl of extraction buffer (worm lysis buffer [42]; containing: 50 mM KCL, 10 mM Tris-Cl pH 8.3, 2.5 mM MgCl2, 0.45% NP40, and 0.45% Tween 20), frozen at −80˚C (20 min), incubated at 65˚C (2 h), and consequently treated at 95˚C (10 min) for inaction of the proteinase K enzyme. DNA samples were stored at −20˚C until their use as PCR templates. PCR was carried out in a total volume of 30 μl [19.2 μl distilled water, 3 μl 10× PCR buffer, 0.6 μl 10 mM dNTP mixture, 1.2 μl 50 mM MgCl2, 1.2 μl of each primer (10 pmol/μl), 0.6 μl of Taq DNA polymerase (5 unit/μl, CinnaGen, Tehran, Iran) and 3 μl of DNA template]. The thermal cycling program for amplifying three nematode genomic fragments (SSU, LSU D2-D3 and ITS rDNA) was as follows: denaturation at 95˚C for 4 min, followed by 35 cycles of denaturation at 94˚C for 30 s, annealing at 52˚C for 40 s, and extension at 72˚C for 80 s. A final extension was performed at 72˚C for 10 min. The COI mtDNA was amplified according to aforementioned PCR program, except the annealing temperature was set to 46˚C. The PCR mixture rations and the thermal cycling program for amplifying bacterial 16S rDNA was as already described for nematode genomic fragments amplifications.
The information of the primers used for amplification of aforementioned fragments are given in Table 2.
The PCR products were sequenced in both directions using the same primers with an ABI 3730XL sequencer (Applied Biosystems) at Macrogen (Seoul, South Korea). Newly obtained sequences of the studied nematode species and their endosymbiont bacteria were deposited into the GenBank database under the accession numbers given in Table 3.
The DNA sequences for reconstructing the phylogenetic trees were selected according to the recent studies [10,21,34,48]. The selected sequences were updated after BLAST search.
The LSU rDNA D2-D3 and COI mt DNA sequences (94 and 75 sequences respectively, including outgroup sequences) were aligned using the MUSCLE algorithm implemented in MEGA 6 [50]; and the resultant alignment was manually edited.
The nematode ITS sequences (53 sequences including outgroup sequences) and the bacterial 16S rDNA (42 sequences for Candidatus Xiphinamtobacter spp. and 25 sequences for Burkholderiaceae spp. phylogenies respectively, including outgroup sequences) were aligned using the Q-INS-i algorithm of online version of MAFFT version 7 (http://mafft.cbrc.jp/ alignment/server/) [51]. The Gblocks program (version 0.91b) with the three less stringent parameters, a server tool at the Castresana Lab (http://molevol.cmima.csic.es/castresana/ Gblocks_server.html), was used for post-editing of this alignment to eliminate poorly aligned regions or divergent positions. The most appropriate model of nucleotide substitution was selected using the Akaike information criterion (AIC) in MrModeltest 2 [52]. The general time-reversible model, including a gamma distribution for rates across sites and a proportion of invariant sites (GTR+G+I) was selected for all phylogenetic analyses in this study. Bayesian inference (BI) was performed using MrBayes v3.1.2 [53], running the chains for 5 × 10 6 generations (for all datasets). After discarding burn-in samples, the remaining samples were retained for further analyses. The Markov-chain Monte Carlo method within a Bayesian framework was used to estimate the posterior probabilities of the phylogenetic trees [54], using the 50% majority rule. Adequacy of the posterior sample size was evaluated using autocorrelation statistics as implemented in TRACER v.1.5 [55]. A maximum likelihood (ML) tree was reconstructed with RaxmlGUI 1.1 [56] software and the same nucleotide substitution model as that used for BI including 1000 bootstrap (BS) pseudoreplicates. For the LSU rDNA dataset, the species Tylencholaimus teres Thorne, 1939 [57] and Ecumenicus monohystera De Man, 1880 [58], (accession number EF207243, AY593013 respectively), for the ITS rDNA phylogeny, the species Heterodorus veletensis Guerrero, Leibanas & Pena-Sañtiago, 2007 [59] (accession number EU477380) and for COI phylogeny, Xiphinema diversicaudatum (Micoletzky, 1922 [60]) Thorne, 1939 [57] (KF292292) and X. index Thorne [64] (AJ251634) were used as outgroup species in accordance with previous studies [10,34].

Co-phylogenetic analyses
In order to understand how the new species identified in our study, and its endosymbiont, fit into the broader evolutionary picture of the Xiphinema americanum-group, we undertook a cophylogenetic analysis of the nematodes and their endosymbionts. Our analysis examined cophylogenetic signal between the endosymbiont 16S phylogenies and both nematode phylogenies (one based on the LSU sequences, the other based on COI sequences). Genetic data was available for nine pairs of host and endosymbiont for the LSU region and seven pairs of species for the COI region. Sequences for the host nematode species were selected when sequences for their endosymbiont bacteria were also available. Cophylogenetic signal in the associations of host and endosymbiont can allude to a parallel evolutionary history between the two as is expected from results elsewhere in similar systems [66][67][68].
We used the cophylogenetic tool, PACo [69,70], to perform our analyses. PACo uses a Procrustes approach to superimpose one phylogeny on the other based on the observed hostendosymbiont associations and estimates cophylogenetic signal as the sum of squared residuals (ss) from this superimposition [69]. Observed cophylogenetic signal is considered significant if the observed test statistic is significantly smaller than the test statistic observed when the matrix of host-endosymbiont associations is randomized and the analysis re-run [70]. In our analysis, we performed PACo analysis with the asymmetric statistic which assesses the goodness-of-fit of the endosymbiont phylogeny onto the host phylogeny (since we would expect host evolution to drive endosymbiont evolution). We compared our observed cophylogenetic signal to the same from 10,000 randomizations of the association matrix and considered cophylogenetic signal significant if the observed signal was smaller than the random expectation at α = 0.05. We also examined the results from the perspective of individual hostendosymbiont associations to examine how cophylogenetic signal of our newly identified species compared to that between the other host-endosymbiont pairs.

Nomenclatural acts
The electronic vision of this paper meets the requirements of the amended international code of zoological nomenclature (ICZN), and hence the new name contained herein is available under that code from the electronic vision of this paper. This published work and the contained nomenclatural acts have been registered in the online registration system for the ICZN in ZooBank. The ZooBank LSID (life science identifiers) for this publication is: urn:lsid:zoobank.org:pub:96D34294-16C8-45C2-BEFD-5A0A5C5AA148. The related LSID information can be viewed through any standard web browser by appending the LSID to the prefix "http:// www.zoobank.org/References/". The electronic edition of this work was published in a journal with an ISSN, and has been archived and is available from the following digital repositories: PubMed Central, LOCKSS. Description.

Xiphinema primum
Female. Body cylindrical, very gradually narrowing towards both extremities, assuming a loose spiral after fixation (Fig 1D and 1E). Cuticle two layered, smooth under light microscopy, fine transverse striae visible on outer layer mostly in tail. The hyaline part of tail 11-15 μm thick, occupying about 1/3 of tail length. Lateral cord 12-28 μm wide, occupying about 1/4 of corresponding body diam. Lip region widely rounded, separated from the rest body by a shallow depression (Fig 1C and Fig 2B-2D and Fig 3A-3D). Amphidial fovea cup-shaped; aperture 12-13 μm wide, located slightly anterior to lip region-body junction region. Odontostyle ca. 1.7 times longer than odontophore, the latter with well-developed flanges. Guiding ring double and guiding sheath 5-10 μm long depending on the degree of protraction/retraction of the stylet. Pharyngeal bulb 78-85 μm long and 19.5-26.0 μm wide. The larger dorsal gland nucleus (DN) located at 15.3-24.7% of pharyngeal bulb length, the two smaller ventrosublateral nuclei (S1N) located at about the same level, at 50.2-60.3% of terminal bulb length. Cardia 9-19 × 5--13 μm in size. Intestine simple, prerectum often not well seen, and rectum 0.6-0.9 times the anal body diameter. Reproductive system didelphic-amphidelphic, the branches about equal in size, the endosymbiont bacteria within the ovaries visible in most examined individuals ( Fig  1B and Fig 2F and 2J and Fig 3F-3H), uterus short ca 50 μm long, and the sphincter not clearly seen. Vagina perpendicular to body axis, about 1/2 to 1/3 of corresponding body diameter, the pars distalis vaginae 15-20 μm long, the pars proximalis vaginae about 1.5 times wider than tall. Tail short, conical, dorsally more convex, with bluntly rounded end (Fig 1K-1N and Fig  2M-2P and Fig 3I-3K). Measurements are listed in Table 4.
Male. Not found.
Juveniles. Recovered in one of the three populations of the new species (the type population with code D, from Ramsar). All four juvenile developmental stages were identified and separated from each other according to Robbins et al. [71]. The correlation between body length of juveniles and females, replacement odontostyle of juveniles and functional odontostyle of females and juveniles is given in Fig 4. The first juvenile (J1) was characterized by the replacement odontostyle tip close to base of functional odontostyle, located at the level of the odontophore, and three other stages had replacement odontostyle posterior to odontophore flanges in resting position of the spear. All stages had dorsally convex conical tails, with a bluntly rounded terminus, however tail in J1and J2 was narrower (Fig 1G-1J). Measurements are listed in Table 5.

Type materials
Holotype, six paratype females and 14 juvenile paratypes (J1-J4) were deposited in the Nematode Collection at the Faculty of Agriculture, Tarbiat Modares University, Tehran, Iran. Seven females of Tehran population (the Dbana population) were deposited in the USDA Nematode Collection, Beltsville, MD, USA and 10 females of Amol population (the Dkond population) were deposited in Ghent University Museum, Zoology Collections, Ghent, Belgium. Specific sequences are deposited in GenBank with accession numbers given in Table 3.

Etymology
The specific epithet shows this is the first species of X. americanum-group described from Iran.
Diagnosis and relationships Xiphinema primum n. sp. belongs to the brevicolle-complex of the Xiphinema americanumgroup, mainly characterized by the lip region being offset from the rest of the body by a slight depression and conical tail, with a bluntly rounded tip. It is further characterized by 2124-2981 μm long females with a widely rounded lip region, separated from the rest of the body by a shallow depression, a 103-125 μm long odontostyle, two equally developed genital branches with visible endosymbiont bacteria under light microscopy, vulva located at 51.8-58.0%, 26-37 μm long conical tail with a more convex dorsal side, bluntly rounded tip and four juvenile developmental stages. The alpha-numeric identification codes of the new species according to Lamberti et al. [19] are: A56, B23, C1, D23, E3, F-, G1, H2, I123.
By having a widely rounded lip region, and conical tail with bluntly rounded tip, the new species comes close to nine species of the brevicolle-complex sensu Orlando et al. [ [75]. The comparisons with aforementioned species are as follows:  [14].
Differentiated from X. himalayense, the tentative cryptic species of the new species, besides distant placement in both LSU and ITS trees, by lip region separated from the body by a constriction (vs continuous), conical tail, dorsally more convex (vs conoid, ventrally straight or slightly concave with rounded terminus) and wider body at anus (35 (30-42)   One new and one known dagger nematode species and data on their endosymbionts Iranian population of X. pachtaicum from Amol (Fig 5) Two populations of the species were recovered from the rhizospheric soil samples of Citrus sp. and Alnus sp. trees, collected respectively in a garden and a forest close to the city of Amol One new and one known dagger nematode species and data on their endosymbionts in Mazandaran province, northern Iran (GPS coordinates 36˚26 0 39@N, 52˚19 0 8.8@E and 362 4 0 7.8@N, 52˚18 0 13.05@E, respectively). The recovered populations were morphologically compared with some other populations of the species from different parts of the world. Compared with the Spanish populations reported by Gutiérrez-Gutiérrez et al. [13], Hungarian populations reported by Repasi et al. [76], and Yugoslavian populations reported by Barsi & Lamberti [20], no remarkable differences in the presently studied population were observed. Compared to the American populations described by Orlando et al. [10], differences in vulva location (V = 55.9-60.5% in Iranian populations vs 48-57%) were observed. As mentioned by Orlando et al. [10], American populations show differences with other populations of the species from other parts of the world for this index. Compared to the Iranian populations of the species reported by Fadaei et al. [77], only minor intraspecies variation was observed for body width at anus (17-24 vs 13.5-19 μm). Measurements are listed in Table 6.

Molecular phylogenetic relationships. Molecular characterization of Xiphinema primum n. sp. and Iranian population of X. pachtaicum
The basic local alignment search tool (BLAST) results using newly obtained sequences given in Table 3 The BLAST search using partial LSU rDNA D2-D3 sequences of four isolates of the new species (MF372941, MF372942, MF372943, MF372947) revealed they have 99% identity with three accession numbers MH248814, MH248815 and AY601602 recorded in the GenBank as Xiphinema sp. Kra-BG, Xiphinema sp. Per-BG and X. taylori, respectively.
The BLAST search using ITS sequences of four isolates of the new species (MF372951, MF372952, MF372953, MF372954) revealed their identity with available sequences having a high coverage were never higher than 91%.
The BLAST search using COI mtDNA sequences of four isolates of the new species (MK202795, MK202796) revealed their identity with currently available sequences is at maximum 90%.
The BLAST search using the 16S rDNA of the endosymbiont bacterium from three isolates of X. pachtaicum, revealed they have 99-100% identity with several sequences of Burkholderiacea bacterium (KT735078, KT735074, KT735068, KU899555, KU899554, KU899551, KT735076, KT735082 and some others) already isolated from the same species. One new and one known dagger nematode species and data on their endosymbionts  One new and one known dagger nematode species and data on their endosymbionts The BLAST search using the 16S rDNA of the endosymbiont bacterium from two isolates of X. primum n. sp. revealed they have at maximum, 95% identity with several sequences of Candidatus Xiphinematobacter spp. (KJ614453, KJ614452, KJ614450, AF217462, KJ614451,  KT735101, KT735096).
The BLAST search using partial SSU rDNA of an isolate of X. pachtaicum sequenced during the present study (MF372940), revealed it has 99% identity with other sequences of the species (KU250139, KP407873, AM086682, MH484520) and an isolate of X. penevi Lazarova, Peneva & Kumari, 2016 [16] (KU250141) available in GenBank. No phylogenetic analyses were performed in the present study using this sequence.
The D2-D3 expansion segments of LSU rDNA phylogeny based on Xiphinema primum n. sp. and Iranian population of X. pachtaicum sequences. This dataset was composed of 727 characters of which 314 characters were variable. Fig 6 represents the Bayesian phylogenetic tree reconstructed using this dataset. The tree is divided to two major clades I and II, according to the nomenclature used in previous studies [10,18]. In this tree, the new species appeared as an independent lineage in a clade containing brevicolle-complex spp. However, the relationships between most species in this clade were not resolved due to polytomy. The Iranian populations of X. pachtaicum formed a clade with two other selected isolates of the species (KU250155, HM921390). The five isolates of X. pachtaicum clade had almost identical LSU rDNA D2-D3 sequences (only one indel in overlapping region was observed), however, the length of their sequences differed, yielding intraspecies cladogenesis and varied branch lengths inside this species clade. The topotype population of X. brevicolle is marked by asterisk in this tree, corroborating its restricted distribution [21].
The ITS phylogeny based on Xiphinema primum n. sp. and Iranian population of X. pachtaicum sequences. The ITS dataset was composed of 1077 characters of which 452 characters were variable. Fig 7 represents the Bayesian phylogenetic tree reconstructed using this dataset. In this tree, a similar pattern to that observed in LSU phylogeny was observed, i.e., the sequences in this tree were divided into two major clades I and II. However, the relationships between the clades inside group II were not resolved due to polytomy. The new species occupies a basal placement to the clade of the X. brevicolle-complex, mostly having similar morphology, and appeared as an independent lineage. Iranian isolates of X. pachtaicum are not included in this tree as attempts to get their ITS sequences failed. Again, the topotype population of X. brevicolle is marked by asterisk, corroborating its restricted distribution [21].
The COI mtDNA phylogeny based on Xiphinema primum n. sp. and Iranian population of X. pachtaicum sequences. The COI dataset was composed of 348 characters of which 201 characters were variable. Fig 8 represents the Bayesian phylogenetic tree reconstructed using this dataset. In this tree, the relationship between the new species and other species and several members of the X. brevicolle-complex is not resolved due to polytomy, and it is sister to X. parabrevicolle and X. peruvianum Lamberti & Bleve-Zacheo, 1979 [9] with poor clade support. Similar to two previous phylogenies, the topotype population of X. brevicolle is marked by asterisk, corroborating its restricted distribution [21].

Molecular characterization of endosymbiont bacterial species
Ca. Xiphinematobacter spp. phylogeny. The 16S rDNA dataset of Ca. Xiphinematobacter spp., including two newly amplified sequences of the bacterial 16S rDNA from the new species, included 526 total characters of which 158 characters were variable. Fig 9 represents the Bayesian phylogenetic tree reconstructed using this dataset. In this tree, the relation of the major clades was not resolved due to polytomy, and the newly generated sequences (MF372955, MF372956) were placed inside the clade including sequences of endosymbionts from X. brevicolle-complex (AF217462, KJ614448, KX263241, KJ614453). However, the clade received poor support (0.66/-) and different alignment, alignment post-editing, and reconstructing methods need to be applied to test any putative scenario for the relationship between these sequences.
The Burkholderiaceae spp. phylogeny. The 16S rDNA dataset of Burkholderiaceae spp. including three newly generated sequences of endosymbionts from X. pachtaicum had 1228 characters of which 214 characters were variable. In this tree (Fig 10), the newly generated sequences occupied a placement inside the clade of several sequences of Burkholderiaceae from X. pachtaicum.

Cophylogenetic analyses
We examined the cophylogenetic patterns between selected sequences of Xiphinema americanum-group spp. and Ca. Xiphinematobacter spp. using both COI and LSU sequences for nematodes. In both cases, we observed strong, significant cophylogenetic signal between the two groups (LSU: ss = 0.076, n = 10,000, p < 0.01, Fig 11A; COI: ss = 0.168, n = 10,000, p < 0.01, Fig 12A). In Figs 11A and 12A, we plot the phylogenies of endsymbionts and hosts (LSU and COI, respectively) and blue lines connect host with endosymbiont where line thickness represents cophylogenetic signal strength (thinner lines represent smaller discrepancy between host and endosymbiont and therefore stronger cophylogenetic signal, dashed lines represent the species identified in this study). Only 0.5% and 0.1% of randomized host-endosymbiont associations, respectively, generated a stronger cophylogenetic signal than the observed associations. When comparing host-endosymbiont associations to each other, we saw that the endosymbiont-host pairs identified in this study display a medium level of cophylogenetic association (Figs 11B and 12B) with some associations (e.g. pairs 4, 6, 7 in Fig 11A  and pair 6 in Fig 12A) showing stronger signal and others showing a weaker signal (e.g. pairs 3 & 9 in Fig 11A and pairs 1 -5 in Fig 12A). In Figs 11B and 12B, we plot the distribution of cophylogenetic signal between pairs of hosts and endosymbionts and mark the species/populations described in this study with vertical lines that demonstrate their medium level of cophylogenetic signal relative to the rest of the group for the LSU marker and relatively strong signal with the COI marker. Overall, these results confirm the established pattern of strong cophylogenetic signal between these two groups, identify average levels of cophylogenetic signal between the host and endosymbiont introduced in this study, and provide another example of parallel cladogenesis between endosymbionts and their hosts.

Discussion
The nematodes of the Xiphinema americanum-group are infamous for their difficulty in species delimitation using traditional criteria. The conserved morphology of even phylogenetically distant species or the close morphometric data ranges further complicate their identification. Using the mean values of the morphometric data as proposed by Orlando et al. [10] and applied in this study (see Differential and Diagnosis), could help primary species comparisons.
In this study, a new species in the Xiphinema americanum-group was described from Iran, as the first Iranian representative of the group, using an integrative approach. It belongs to X. brevicolle-complex, mainly characterized by a bluntly conoid tail, and the lip region not Bayesian 50% majority rule consensus tree inferred using LSU rDNA D2-D3 sequences of X. primum n. sp. and Iranian populations of X. pachtaicum under the GTR+G+I model. Bayesian posterior probabilities and maximum likelihood bootstrap values are given for appropriate clades in the shape BPP/ML BS. The newly generated sequences are in bold. According to Lazarova et al. [21], the accession number assigned to X. brevicolle marked by asterisk belongs to the topotype population of the species. https://doi.org/10.1371/journal.pone.0217506.g006 One new and one known dagger nematode species and data on their endosymbionts  [21], the accession number assigned to X. brevicolle marked by asterisk belongs to the topotype population of the species.  [21], the accession number assigned to X. brevicolle marked by asterisk belongs to the topotype population of the species.
https://doi.org/10.1371/journal.pone.0217506.g008  from the body by a sharp constriction. The species X. himalayense was regarded as its tentative cryptic species that besides morphological traits, and especially the tail characters, could be separated from it based on molecular data.
The molecular phylogenetic studies of the group were performed in some previous studies [10,21,34]. Using LSU rDNA, He et al. [12] found that the species of this group are divided into two subclades: the X. americanum and the X. brevicolle subgroups [12]. That latter was recently considered as "X. brevicolle species complex" [10]. The former phylogenetic studies based upon LSU rDNA D2-D3 and ITS markers, usually yielded congruent phylogenies [14], dividing the species into two main subclades I and II [10, 14, present study]. Two aforementioned studies, the present study, and the most recent study by Lazarova et al. [21] show that the COI mtDNA marker has acceptable interspecific variation to be used as an alternative approach that is helpful in overcoming the low divergence of LSU rDNA D2-D3 [10]. X. primum n. sp. was however separated from its morphologically close species and its cryptic species by both LSU and ITS markers in the corresponding phylogenetic trees.
The cryptic nature of some species and the misidentification of some species in GenBank remains an issue. The observed intraspecific divergence of D2-D3 marker in X. brevicolle s. l. [10,14] documents the cryptic nature of nematodes of this group well. Sequencing of the topotype population of X. brevicolle s. str. by Lazarova et al. [21] revealed several populations are indeed misidentified and are erroneously assigned to this species. As a result, any observed paraphyletic or polyphyletic status of a given species, should be checked by the correct identification of that species first, and more preferably, topotype individuals of the studied species should be included in sequencing experiments whenever possible.
The recovered Iranian populations of X. pachtaicum and the new species were inspected for the presence of endosymbiont bacteria inside their ovaries. The rod-shaped endosymbiont One new and one known dagger nematode species and data on their endosymbionts bacteria were observed inside the ovaries of both recovered species under light microscopy ( Fig 5G and 5H). The amplification of 16S rDNA and sequencing of the amplified fragments, revealed that the endosymbiont bacterium of the new species belongs to Ca. Xiphinematobacter sp., forming an independent lineage in the corresponding phylogenetic tree, and was not identical to other available sequences of this genus. The endosymbiont relation of Ca. Xiphinematobacter spp. with X. americanum-group members has recently been studied by several authors [34,48,10]. The endosymbiont of X. pachtaicum had an affinity with plant-and fungiassociated species of Burkholderiaceae, and was placed in the clade of species already isolated from X. pachtaicum. According to Palomares-Rius et al. [34], X. pachtaicum could harbor both endosymbionts; either a Ca. Xiphinematobacter sp. or a Burkholderiaceae sp.
Significant cophylogenetic signal has already been observed between nematodes of this group and both groups of the endosymbiont bacteria phylogenies, albeit with variation in signal strength [10,34]. In this study, significant cophylogenetic signal was observed between both recovered nematode species and their endosymbionts phylogeny. The pair of species introduced in this study showed similar cophylogenetic signal to the rest of the group and do not disrupt the overall signal within the group. These results confirm the generally held expectation that host-endosymbiont evolution often occurs in parallel and indeed the evolution in the dependent group, parallels that of the host [66][67][68] as previously documented for X. americanum-group members and their endosymbionts [10,34]. Indeed, the vertical transmission of endosymbionts via parthenogenesis means a significant cophylogenetic signal is expected. Cophylogenetic signal is displayed on the x-axis where smaller values represent stronger cophylogenetic signal. The density profile (grey) describes the cophylogenetic signal between the all pairs of species and the two pairs identified in this study are also represented by the blue vertical lines. https://doi.org/10.1371/journal.pone.0217506.g011 One new and one known dagger nematode species and data on their endosymbionts

Conclusion
Hereby, we studied two Iranian populations of Xiphinema pachtaicum and described a new species, both of which belonged to the X. americanum-group, using an integrated approach. Significant cophylogeny between the newly generated sequences of nematode LSU rDNA D2-D3 and COI mtDNA sequences and the 16S rDNA sequences of the endosymbiont bacteria were also observed. The phylogenies of endosymbiont bacteria alongside that of nematode host (Xiphinema americanum spp.) Blue lines connect host and endosymbiont and the thickness of each line represents the strength of cophylogenetic signal between host and endosymbiont. Thinner lines reflect stronger cophylogenetic signal (i.e. line thickness is the level of discordance between host and endosymbiont). Dashed blue lines and red numbers distinguish the species described in this study from the others used in the analysis. (B) The cophylogenetic signal of individual host-endosymbiont associations. Cophylogenetic signal is displayed on the x-axis where smaller values represent stronger cophylogenetic signal. The density profile (grey) describes the cophylogenetic signal between the all pairs of species and the pair identified in this study are also represented by the blue vertical line. https://doi.org/10.1371/journal.pone.0217506.g012 One new and one known dagger nematode species and data on their endosymbionts