Stauratostoma shelleyi n. gen., n. sp. (Nematoda: Rhabditida: Thelastomatidae) from Appalachian Polydesmid Millipedes (Polydesmida: Xystodesmidae)
Article Category: research-article
Published Online: Jun 03, 2018
Page range: 133 - 146
DOI: https://doi.org/10.21307/jofnem-2018-023
Keywords
© 2018 Gary Phillips et al., published by Sciendo
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
The nematode fauna living within the gastrointestinal tract of North American millipedes is not well documented (Carreno, 2007). North American millipedes are understudied hosts of oxyuridomorph and rhigonematomorph nematodes, and records of their distribution in North America are scant. Most studies of nematodes that parasitize diplopods have originated from tropical areas (Carreno et al., 2013). Joseph Leidy, the father of American parasitology, was one of the first researchers to document the existence of nematodes living inside the intestine of millipedes in temperate North America (Leidy, 1849, 1850, 1851, 1853). The first known thelastomatid nematode recorded by Leidy was
In the course of a recent survey of Appalachian millipedes in Tennessee, North Carolina, and Alabama, as well as other areas from the southeastern and western United States, many species of Rhigonematomorpha and Oxyuridomorpha were collected from millipedes. Oxyuridomorphan nematodes are commensal parasites in the intestine of millipedes, scorpions, and several orders of insects, while rhigonematomoran nematodes are found only in millipedes. Among the thelastomatid and rhigonematid nematode taxa collected in the southern Appalachian region of North America was an unusual species that could not be placed in any described genus. A new genus and species of nematode,
Between February 2013 and July 2017, a total of 352 xystodesmid millipedes were collected in several locations in the southeastern United States, including several Tennessee and Alabama State Parks. Millipedes were transported back to the lab where they were maintained in enclosures with their natural substrate and fed pieces of cucumber fruit. Prior to dissection, morphometric data were recorded for each millipede, including weight, length, width, and sex. Millipedes with movement activity typical of their species were considered healthy.
Millipedes were dissected by severing the head and epiproct with a razor blade as described by Phillips et al. (2016). The intestine was pulled intact from the body cavity with fine-tip forceps and placed in a Syracuse watch glass containing distilled water. The intestine was sectioned into three parts: foregut, midgut, and hindgut (Crawford et al., 1983). The intestine was sketched and then dissected with the aid of Zeiss Stemi 2000 or Olympus SZ51 stereomicroscopes. Nematodes were removed from intestinal tissue and sorted. Males, females, and juveniles were segregated, counted, and grouped to family or genus-level taxa according to general features. Each dissected millipede was preserved as a voucher specimen in 70% or 95% ethanol.
Specimens were prepared for light microscopy, scanning electron microscopy (SEM) or molecular analysis. Most specimens were killed and fixed in hot (60‒70°C) 4% formalin, then later processed to anhydrous glycerin (Seinhorst, 1959) for permanent mounts on glass slides or long-term storage in vials. Slide-mounted specimens were examined with an Olympus BX-63 DIC microscope system and imaged with a 14-megapixel Q-camera. Measurements in Table 1 were made from glycerin-preserved specimens.
Morphometrics of female
Paratype females ( |
|||||
---|---|---|---|---|---|
Measurements (µm) | Holotype female | Mean | SD | Range | CV |
Length | 2,522 | 2,357 | 249 | 1,805‒2,683 | 10.6 |
Maximum width | 192 | 178 | 33.9 | 98‒237 | 19.1 |
First annule width | 11.1 | 10.3 | 0.89 | 8.1‒12.3 | 8.3 |
Second annule width | 8.4 | 7.8 | 1.2 | 5.3‒9.9 | 14.9 |
Esophagus length | 508 | 486 | 50.0 | 347‒548 | 10.3 |
Basal bulb length | 116 | 111 | 8.6 | 84‒126 | 7.8 |
Head to excretory pore | 404 | 389 | 43.9 | 274‒445 | 11.3 |
Vulva to anterior head | 991 | 935 | 105.3 | 676‒1066 | 11.3 |
Anus to anterior head | 1,570 | 1,517 | 174.1 | 1,105‒1,715 | 11.5 |
Egg length | 75.2 | 74.2 | 3.4 | 67.2‒79 | 4.5 |
Egg width | 48.4 | 49.1 | 2.7 | 43.9‒55.4 | 5.4 |
Tail length | 952 | 84.0 | 90.8 | 649‒973 | 10.8 |
Ratios | |||||
|
13.1 | 13.6 | 2.2 | 10.1–18.4 | 16.5 |
|
5.0 | 4.9 | 0.43 | 4.3‒5.9 | 8.9 |
|
2.6 | 2.8 | 0.2 | 2.4‒3.3 | 5.5 |
|
39.3 | 39.7 | 2.4 | 33.2‒44.4 | 6.1 |
|
0.63 | 0.62 | 0.03 | 0.53‒0.69 | 5.1 |
CV = Coefficient of Variation.
To better visualize the lateral field, head region, and phasmid features, several females were fixed in 95% ethanol for 2 days and mounted directly on slides into Hoyer’s medium. Slides were placed in a 50°C oven for 3 days to expand the nematodes and harden the mounting medium, then ringed with red insulating varnish. Features were imaged with a 17-megapixel DP73 camera on an Olympus BX-53 phase-contrast microscope.
For SEM, methods used by Phillips et al. (2016) were followed. Formalin-fixed nematodes were washed in distilled water for 20 min then placed into a 12-mm × 30-µm microporous specimen capsule (Electron Microscopy Services, Hatfield, PA). Each capsule was placed in a 5-ml glass well and dehydrated with a graded ethanol series consisting of 25%, 50%, 75%, 95%, and 100% ethanol, each for 20 min. Following the 100% ethanol dehydration step, a 1:1 mixture of 100% ethanol and reagent grade hexamethyldisilazane (HMDS) was used in place of a critical point dryer. The HMDS series consisted of 75%, 100%, and a second 100% dehydration, each for 20 min. Nematodes were placed on carbon tape affixed to a 45°/90° aluminum stub and sputter-coated with gold for 10 sec at 20 μA in a SPI-Module Sputter Coater (West Chester, PA). Specimens were viewed with a Hitachi TM 3030 electron microscope at a voltage of 15 kV.
Total genomic DNA was extracted from representative single specimens with the Qiagen DNeasy Blood and Tissue Kit #69506 (Waltham, MA) differing from the manufacturer’s instructions only in reduction of the final elution volume to 70 μl (2 × 35 μl) from 400 μl (2 × 200 μl). The resulting gDNA was stored at −20°C. Polymerase chain reaction (PCR) was carried out with TaKaRa Ex Taq Hotstart DNA polymerase (Takara Bio, Shiga, Japan) per the manufacturer’s suggested protocol, plus 2 μL of DNA template and 3 μl of 20 μM working stocks of 28S LSU rDNA primers. Several primer pairs were employed, with the greatest success from LSU391F: 5′-AGCGGAGGAAAAGAAACTAA-3′ and LSU501R: 5′-TCRGARGGAACCAGCTACTA-3′ (Nadler et al., 2006). Two other custom internal primer pairs were also used with limited success: 28S 40F: 5′-GGAARCGGATAGAGTTGAC-3′ and 28S 41R: 5′-CTACTAGATGGTTCGATTAGTC-3′ and 28S 40NF: 5′-GAGTTCAAGAGGGCGTGAAAC-3′ and 28S 41NR: 5′-CCTCTAATCATTCGCTTTACC-3′. Cycling was done using a GenePro (Bioer Technology Co., Hangzhou, China) thermal cycler using the following PCR regime: initial 90 sec denaturing step at 94°C, then four cycles of 30 sec at 94°C, 30 sec at 56°C and 75 sec at 72°C, followed by four cycles of 30 sec at 94°C, 25 sec at 52°C, and 75 sec at 72°C, nine cycles of 30 sec at 94°C, 20 sec at 48°C and 75 sec at 72°C and finally, 38 cycles of 30 sec at 94°C, 20 sec at 45°C and 75 sec at 72°C.
PCR products were electrophoresed in 1% agarose gels at 110 V for 30 min. Bands were excised from the gel, cleaned using QiaQuick Gel Extraction Kits and eluted in 37 μL of elution buffer. Cleaned PCR products were used as templates for cycle sequencing (Sanger) reactions using 1.5 µl of the PCR primers at a working concentration of 5 µM. Sequencing was performed in both directions using BigDye v3.1 terminators (Applied Biosystems, Carlsbad, CA) in a 1/20th reaction using 0.4 µl BigDye and 3 to 5 µl of homemade 5X sequencing buffer cocktail in a 20 µl reaction volume. Centrisep columns (Princeton Separations, Adelphia, NJ) were used to clean the sequencing reactions, which were then dried in a Centrivap Concentrator (LABCONCO, Kansas City, MO). Dried samples were sent to the University of Tennessee Genomics Core for sequencing. Sequencher 4.7 (Gene Codes Corp., Ann Arbor, MI) was used to reconcile and verify opposing strands for accuracy. The 28S rDNA sequences of 11 females and two juveniles (
In order to ascertain the phylogenetic position of
Parsimony analysis was comprised of a heuristic search employing 1,000 random addition sequence replicates using tree bisection and reconnection (TBR) branch rearrangement. Of 1,279 aligned characters, 610 were parsimony informative. A single most parsimonious tree of 3,179 steps (CI: 0.506; RI: 0.625; HI: 0.494) was recovered in 900 of the 1,000 replicates conducted. Prior to Bayesian analysis, JModeltest v. 2.1.10 (Darriba et al., 2012) was used to determine the most appropriate evolutionary model. The best-fit model chosen was GTR + I + G: (-lnL: 16,894.3004). A Bayesian phylogeny was estimated using Markov Chain Monte Carlo methods implemented within MrBayes 3.2.2 (Ronquist et al., 2012) through the online CIPRES Science Gateway (Miller et al., 2010). No partitions within 28S LSU rDNA were recognized. Nucleotide substitution matrix, rate variation, gamma shape parameter, and base frequencies were estimated (nst = 6; rate = invgamma; unlink statefreq = (all); revmat = (all); tratio = (all); shape = (all); pinvar = (all); prset applyto = (all) ratepr = variable). Two runs with six chains each were run for a total of 10 million generations. Markov chains were sampled at intervals of 500 generations and the first 35% of trees discarded as burn-in prior to assembling a 50% majority rule consensus tree. Verification that stationarity had been reached was measured by the standard deviation of split frequencies being less than 0.1, the Potential Scale Reduction Factor approaching 1.0, and the MrBayes output overlay plot depicting no directional trend for either run. Resulting phylogenetic trees were modified using Canvas 8.0.5 (Deneba Systems) to produce publication-grade figures.
Millipede and nematode morphometric data were analyzed with mixed model analysis for a nested design, with nematode length as the response variable and millipede species as the categorical independent variable, and millipede length, width, and weight as numeric independent variables, respectively. Rank data transformation was applied when the diagnostics analysis showed non-normality and unequal variance on residuals. Significant effects were identified at
During this research, 972 millipedes spanning six orders, 16 families, and 53 species were collected from 20 middle Atlantic, southeastern, southwestern, and western states. A total of 66,685 nematodes were extracted and separated into morphotaxa. Of the 972 millipedes dissected, 352 (36.2%) were millipedes from the order Polydesmida. Within the order Polydesmida, only one family (Xystodesmidae) contained
Specimens in 15 genera of xystodesmid millipedes (Table 2) were dissected to determine the presence or absence of intestinal nematodes. Total nematode loads ranged from 0‒418 nematodes/specimen, primarily represented by genera in Thelastomatidae, Rhigonematidae, Aoruridae, and Coronostomatidae. Among these 15 genera, nine (60%) contained specimens of
Genera of xystodesmid millipedes dissected.
Species | Millipedes dissected ( |
Millipedes with |
Total |
Mean | Range |
---|---|---|---|---|---|
|
77 | 59 | 269 | 3.5 | 0‒19 |
|
4 | 0 | 0 | 0 | 0 |
|
56 | 35 | 276 | 4.9 | 0‒31 |
|
1 | 0 | 0 | 0 | 0 |
|
6 | 4 | 41 | 6.8 | 0‒21 |
|
3 | 2 | 45 | 15 | 0‒32 |
|
47 | 5 | 41 | 0.9 | 0‒23 |
|
2 | 0 | 0 | 0 | 0 |
|
29 | 0 | 0 | 0 | 0 |
|
56 | 0 | 0 | 0 | 0 |
|
44 | 0 | 0 | 0 | 0 |
|
2 | 2 | 20 | 10 | 5‒15 |
|
1 | 0 | 0 | 0 | 0 |
|
14 | 1 | 6 | 0.4 | 0‒6 |
|
10 | 2 | 2 | 0.2 | 0‒1 |
Total | 352 | 110 | 700 | 2.8 | 0‒32 |
Systematics
Thelastomatidae (Travassos, 1929)
Description
Obligatory commensal parasitic inhabitants of the hindgut and midgut of some xystodesmid millipedes.
Type species:
Description
The collection localities are located in the southern Appalachian Mountains from North Carolina and Tennessee to Alabama, and in the adjacent Ridge and Valley Province and Cumberland Plateau in Tennessee. This nematode was not found in any xystodesmid millipedes collected elsewhere.
Basal layer of cuticle composed of bands of pebble-like sectors (Fig. 4A,C) and more amorphous sectors (Fig. 4B,D) in each annule, with pebbly sectors more apparent in muscle-free zones (Fig. 4A,B). Pseudocoelom in esophageal region with numerous crisscrossed connective fibers anchored at hypodermis and esophageal wall (Fig. 4D); fibers sparse along most of intestinal length, more abundant in posterior intestine-rectal region (Fig. 1F).
Head end with two pairs of flattened, bifurcated flaps forming a cruciate oral opening (Fig. 2A–D). Amphid apertures inconspicuous, small, linear, each with a minute associated guard spine (Fig. 2A–D). Stoma wide; cheilostom thin, sinuous in profile; gymnostom appearing as a thick, curved rod wider at base than at apex; telostom thick, angled, bearing three pairs of pointed teeth, larger teeth appearing less sclerotized than smaller teeth (Fig. 1B). Esophagus with long cylindrical corpus, short isthmus, and pyriform basal bulb containing a grinding valve; terminus of corpus with distinct esophago-intestinal valve. Anterior region of esophagus surrounded by six glands (dorsal, ventral, four sublateral), each with prominent granules and 1 or 2 nuclei (Fig. 1B). Anterior end of intestine swollen (Fig. 1A).
Nerve ring encircling esophagus at middle of procorpus. Secretory-excretory system consisting of massive ampulla and large oval pore located level with anterior end of basal bulb at about the 34th annule (Figs. 1C; 3B,C), canals not seen. Reproductive system amphidelphic, vulva located just anterior to midbody, transverse, anterior lip flap-like; vagina strongly muscular, directed anteriorly; both uteri on right side of body, gonads doubly reflexed in mature females, stretching from anterior intestinal region nearly to anus; anterior gonad without spermatheca, posterior gonad with an axial spermatheca containing small, broadly oval sperm (Fig. 1A,D-G). Egg (Fig. 3D) broadly oval, shell thin, minutely roughened. Newly formed eggs with cytoplasm distributed uniformly inside egg, older eggs with cytoplasm contracted (Fig. 1F,G). Four ventral coelomocytes present, most anterior coelomocyte associated with anterior flexures of gonads, other three coelomocytes in region of anterior uterus (Fig. 1A,D–F). Intestine composed of single layer of polygonal cells (Fig. 1D; rectum about two body annules long, straight, not inflated (Fig. 1F). Anus without prominent flap. Phasmid aperture a minute pore about 80 μm posterior to anus (Fig. 1F).
Males not known.
Juveniles similar to females except for development of reproductive system (Fig. 1E); four coelomocytes distributed as in female: one at anteriormost part of gonads, the other three closer to developing vagina.
Phylogenetic analysis of 28s rDNA was conducted using two different methods: maximum parsimony and Bayesian Inference. Both analyses yielded largely congruent trees, differing chiefly in the relationships among the basal-most clades, which were weakly supported. The Bayesian inference is presented as Fig. 5. Both trees recovered
The length of each specimen of





Molecular tree of millipede-parasitic nematodes based on partial 28S rDNA analysis.


Correlation of nematode length and millipede weight for two
Thelastomatidae has been considered a paraphyletic group (Adamson and van Waerebeke, 1992) without a unifying synapomorphy, a hypothesis supported by a limited molecular survey of thelastomatid species from a cockroach (Jex et al., 2005). The family as currently circumscribed contains about 30 genera following additions and subtractions (Adamson and van Waerebeke, 1992; Jex et al., 2005; Phillips et al., 2016). None of these genera have the cephalic development seen in
One other species shows a possible slight convergence with
Appalachian representatives of Thelastomatidae have several morphological characteristics in common with
The length of each specimen of