Infection with mermithid nematodes causes the depriesteri morphology in Philodromus collinus (Araneae: Philodromidae)

Abstract. Two female Philodromus specimens were found with genital structures similar to those described for Philodromus depriesteri Braun, 1965, currently considered a nomen dubium. Molecular analysis revealed that the two specimens belong to the species Philodromus collinus C. L. Koch, 1835. The opisthosomas of both spiders each contained a parasitoid, and one of the worms was identified as a member of the nematode family Mermithidae. It is very likely that the parasitoids interfered with the development of the female genital organs, resulting in the characteristic depriesteri vulva structure. Consequently, we propose Philodromus depriesteri Braun, 1965 = Philodromus collinus C. L. Koch, 1835 syn. nov., thus releasing P. depriesteri from being considered a nomen dubium. Zusammenfassung. Zwei weibliche Philodromus-Exemplare wurden gefunden, deren Genitalstrukturen Philodromus depriesteri Braun, 1965 ähneln, ein Name, der aktuell als nomen dubium gilt. Eine molekulare Analyse ergab, dass die Exemplare zur Art Philodromus collinus C. L. Koch, 1835 gehören. Die Opisthosomata beider Spinnen enthielten je einen parasitoidischen Wurm. Die Analyse ergab, dass einer von diesen der Nematoden-Familie Mermithidae angehört. Es wird angenommen, dass diese Parasitoiden die Entwicklung der Vulven verändert und dabei das charakteristische depriesteri-Erscheinungsbild hervorgerufen haben. Folglich synonymisieren wir Philodromus depriesteri Braun, 1965 mit Philodromus collinus C. L. Koch, 1835 (syn. nov.), so dass P. depriesteri nicht mehr als nomen dubium betrachtet werden muss.

To identify spiders at the species level, genital morphologi cal characteristics are usually used. These are relatively con stant morphologically (Huber 2004). Aberrant appearances of the genital organs are rare ( Jocqué 2002) and mostly due to gynandromorphism and intersexuality (Holm 1941, Kas ton 1961. The latter are mostly asymmetrical. Symmetrical aberrations are extremely rare ( Jocqué 2002). Infestation with parasitoids has been discussed as a possible cause for such alterations (Martin 2013). Outside the Araneae, the phe nomenon of sexually abnormal individuals that coincide with the presence of a parasitoid, namely Mermithidae, occurs in the fly Simulium trangense (Ya´cob et al. 2021).
Mermithidae are a family of nematodes. They are very long and thin worms that mainly parasitize insects such as mosquitoes, grasshoppers, butterflies, damselflies or beetles, but also in other invertebrates such as crustaceans, leeches and spiders. Often the mermithid hosts are unknown (Nickle 1972). The occurrence of these nematodes in spiders seems to be a frequent phenomenon (Penney & Bennet 2006, Košulič & Mašová 2019 and could have an impact on their general morphological appearance, including the appearance of the genitalia. It is interesting to note that all parasitized spider species found generally feed on adult insects that have an aquatic larval stage, e.g. Chironomidae, Culicidae, Tricho ptera (Poinar 1985). Eggs of mermithids laid in or near aquatic habitats are ingested and hatch in the gut of the paratenic host, which may be one of various invertebrates (insect lar vae, turbellarians, annelids). The preparasitic larvae invade the tissue and enter a state of dormancy, which is only broken when the paratenic host is eaten by the developmental host (e.g. a spider). There, the parasite completes its development into a postparasitic juvenile. Such an infestation is thus indi rect (Poinar 1985).
The indirect cycle includes two additional variables com pared to the direct form. The paratenic host must fall prey to the correct predator and the parasitized host must return to the aquatic environment with the nematode when the latter is ready to emerge (Poinar 1987). The direct cycle involves a single developmental host. The preparasitic juvenile hatch ing from the egg laid in the environment enters a receptive host, develops there and hatches as a postparasitic juvenile. This stage returns to the environment to mature, mate and lay eggs (Poinar 1987). The indirect cycle, when mermithid para sitoids use insects as a paratenic host, which then get con sumed by, e.g. spiders, may explain the parasitism of spider species associated with different habitats (Poinar & Benton 1986). A potential explanation for how infestation works is that juvenile mermithids are quite small and live in the prey haemolymph, not encysted in tissues, and thus could be in gested along with the liquid diet of the spider (Poinar & Early 1990). Whether the mermithids exhibit defences that protect them against digestive enzymes is unknown. Also unknown is how they penetrate the gut to enter the cavity of the spi der body (Carl N. Keiser, pers. comm.). Parasitized spiders were found in a variety of habitats (on plants, in webs and on the ground). They were observed going into water where the nematodes hatched from the bodies of the hosts (Poinar 1987). It is likely that a number of different mermithid genera and species infest spiders and that some of them may also have a direct cycle (Poinar 1987).
Except for Mermithidae, wellknown parasites of spi ders include Diptera (Schlinger 1987) and Nematomorpha (Poinar 1987). The latter bear a superficial resemblance to Mermithidae and can best be distinguished by colour, when only morphological examination is possible. Mermithid nematodes are usually whitish or creamcoloured with shades of pink, yellow and green, while nematomorph gordiids are mostly brown or black (Poinar 1987). Overall, it can be said that despite the high diversity and ecological impact of both parasites and spiders, the understanding of parasitoidspider interactions is very limited compared to similar fields, e.g., parasites or parasitoids of insects (Durkin et al. 2021).
As already mentioned, infestations with mermithids can affect the development of the vulva or other sex-specific traits (Leech 1966, van den Berg & DippenaarSchoemann 2009). Consequently, an altered genital morphology could lead to the description of new species, as this an impor tant diagnostic character in spider taxonomy (Breitling et al. 2015). Some of these cases are now considered nomina dubia, and in this study the status of such a name, Philodromus depriesteri Braun, 1965, is investigated and discussed. This species was first described by Braun (1965) based on two specimens found in the early 1950s, one in Krimml (Austria, Oberpinzgau in Salzburg) near the great water falls and a second in Geisenheim (Germany: Rheingau in Hesse). Both sites are approximately 600 km apart. Despite this large distance and generally increased sampling activ ity since this time, it took more than 50 years until the next specimens were recorded in 2012 in the Allgäu region in Bavaria (Breitling et al. 2015). Due to the high similarity of the nongenital characters, P. depriesteri was considered closely related to P. collinus by Braun (1965). When the last specimen of P. depriesteri was found in 2012, a para sitic worm was also discovered in the opisthosoma, which already led to the consideration that the aberrant form of the P. depriesteri vulva was induced by this nematode (Brei tling et al. 2015). However, Breitling et al. (2015) did not synonymize P. depriesteri and P. collinus because of the similarity of females in the aureolus group. In the present study, new arguments supporting this synonymy based on gene sequence (cytochrome c oxidase subunit I) data are presented.

Material and methods
Material examined. GERMANY: Bavaria: Bayerischer Wald, Steinklamm Spiegelau, 48.90276°N, 13.37237°E, 2 (( beating from pine, 22. Jul. 2020. Sample availability see below. Morphological examination. The examination of the spiders and the further preparation of the epigynes and the worms was carried out using an Optika SZM2 stereoscope. Epi gynes were macerated with lactic acid at room temperature and were subsequently fixed in Euparal. Genital images were taken with a Zeiss Standard 15 microscope and a Canon EOS 80d. Stacks were taken in RAW format using Helicon Remote software and merged using Helicon Focus 7 soft ware. The resulting TIFF images were lightly edited in Ado be Photoshop and saved in JPEG format (photos: Viktoria Wegewitz).

Molecular analysis.
Total genomic DNA was extracted from ethanolpreserved tissue (spiders: legs; worm 2: body fragments, worm 1: whole specimen) using silica membrane columns of the Blood and Tissue kit by Qiagen (Hilden, Germany), following the manufacturer's specifications. We amplified 658 bp of spider DNA from the 5'end of the COI (cytochrome c oxidase subunit I) gene with primers HCO2198JJ and LCO1490JJ (Astrin & Stüben 2008) in reaction volumes of 20 μl, including 2.0 μl of DNA tem plate, and using the "Multiplex PCR Master Mix" (Qiagen). PCR conditions were as follows: first cycle set (15 repeats): 35 s denaturation at 94°C, 90 s annealing at 55°C (−1°C per cycle) and 90 s extension at 72°C. Second cycle set (25 repeats): 35 s denaturation at 94°C, 90 s annealing at 45°C, and 90 s extension at 72°C. PCR products were sent for bi directional Sanger sequencing to BGI (Hong Kong, China). For the nematodes, the ribosomal SSU (18S) Hyper Vari able Region I (HVR I) was amplified as described by Zhou et al. (2019). The primers SSU18A and SSU26R were used for PCR and SSU9R for sequencing. These primers are suit able for a wide range of nematodes (Floyd et al. 2002, Blax ter et al. 1998. DNA extracts and tissue samples (except for worm 1, which was processed destructively elsewhere) are available from the LIB Biobank at Museum Koenig, Bonn under the voucher IDs specified in Tab. 1. Figure 1 shows the dorsal view of a normally developed vulva from Philodromus collinus, beaten from a pine in the Harz Mountains on 30. Jul. 2019 (51.52505°N, 11.09100°E). The two dissected female reproductive organs of specimen 1 (Fig. 2, body size 5.0 mm) and specimen 2 (Fig. 3, body size 5.8 mm) from Spiegelau show incomplete vulva structures, in particular the receptacula seminis are barely developed. The epigyne/vulva of specimen 1 is somewhat more developed than in specimen 2. The structure of their epigyne/vulva is characterized by the following points: Missing fully developed receptacula seminis and overall un derdeveloped appearance. The structure of vulva 2 corresponds to the holotype of P. depriesteri (fig. 93 in Braun 1965) and the structure of vulva 1 corresponds to that of the paratype of P. depriesteri ( fig. 94 in Braun 1965). The sclerotized structures of the vulvae correspond to "Rezeptakularöhren" (which translates to receptacula tubes) by Braun (1965) and "copulatory ducts" by Muster & Thaler (2004), respectively. Holotype* P. depriesteri ZFMKDNAFD15676217 *: DNA of the type specimen of P. depriesteri (Senckenberg Museum Frankfurt) was too degraded to produce any PCR product, but has been deposited at LIB Biobank for potential future non-Sanger based approaches

Morphological analysis
In both specimens the opisthosoma was filled with a worm (Fig. 4). The worm of specimen 2 had a cream colour immediately after dissection. Gradually it turned brown, pro bably due to contact with oxygen. The worm of specimen 1 was darker (picture not shown). In general appearance, the infected spiders resembled specimens with the typical mor phology of the Philodromus aureolus group, with specimen 1 looking slightly damaged.

Molecular analysis of spiders
Both sequences of specimens 1 and 2 from Spiegelau could be unambiguously assigned to the species Philodromus collinus. In addition to BLAST searches in GenBank and BOLD, the sequences were also compared with the Philodromus speci mens sequenced in the German Barcode of Life (GBOL) data release (Astrin et al. 2016). Embedded in this dataset, the sequences of the two spider specimens (which differed from each other in one base) had pairwise genetic distances (uncorrected pdistances) of 0.2% to 1.1% with respect to Philodromus collinus sequences. The attempt to sequence one leg of the holotype of Philodromus depriesteri failed.

Molecular analysis of nematodes
Primer pair SSU HVR1 is the standard for amplification of the18s rDNA hypervariable region. Both worms yielded an identical band in the expected range. Unfortunately, the band from worm 1 turned out to be host derived and also in several rounds of new PCR with different conditions and different primers, we were unable to amplify the worm sequence, pre sumably due to high contamination with host derived DNA. The sequence of worm 2 was very clear. In a BLAST search (on 3. May 2021 and repeated on 13. Jun. 2022) with the sequence of worm 2 as bait, the 10 top hits were all sequences from various mermithids with identities between 94.17% and 97.51% and evalues of 0.0. The fragment considered for the BLAST search corresponds to positions 51 to 531 in AB647219.1 (unclassified Mermithidae), which is the best hit. From this, we conclude that worm 2 (and presumably also worm 1) is a molecularly not yet characterized species of Mermithidae.

Discussion
The status and identity of the mysterious Philodromus depriesteri has caused longlasting confusion and discussion. When Braun described P. depriesteri as the sister species to P. collinus in 1965, he was hesitant because the epigyne had an aberrant appearance. The depriesteri vulva with its reduced receptacula would be in the aureolus group, where voluminous receptacula are otherwise present (Breitling et al. 2015). In later per sonal conversations between Rudolf Braun und Peter Jäger, Braun assumed that he had described a juvenile specimen, most probably belonging to P. collinus, based on preepigynes of juvenile specimens (P. Jäger, pers. comm.). However, this can be excluded, since Philodromus preepigynes show a weakly developed septum without sclerotizations (S. Indzhov, pers. comm.). Expressed in detail, the preepigyne occupies about half the distance between the epigastral fold and the petiolus. It is not sclerotised and looks like the surrounding cuticula. A weakly developed septum is defined by two shallow, semicir cular furrows (S. Indzhov, pers. comm.). Instead, the repeated observation of this aberrant vulva phenotype in specimens that carry parasitic nematodes in the opisthosoma suggests a causal relationship. Most probably, the absence of fully devel oped receptacula is caused by a general developmental delay due to the mermithid infestation. Other explanations, e.g. that the parasites consumed the initially developing receptacula or that the receptacula broke off are less likely, since the recep tacula are missing on both sides and the specimens appeared otherwise uninjured. Currently we do not know whether the "depriesteri phenomenon" may also occur in related Philodromus species as consequences of worm infections. However, the structure of the copulatory ducts in the holotype of P. depriesteri is highly specific for P. collinus. Thus, regardless of the fact that the sequencing of the holotype failed, we propose the new synonymy P. depriesteri Braun, 1965 = Philodromus collinus C. L. Koch, 1835 syn. nov., instead of considering it a nomen dubium. Most probably the holotype also contains a worm in its opisthosoma. Thus, it is evident that an unidenti fied nematode of the family Mermithidae induces a degener ate appearance of the vulva in P. collinus which falsely led to P. depriesteri being described as a new species.