«Taxonomic Revision, Molecular Phylogeny and Zoogeography of the huntsman spider genus Eusparassus (Araneae: Sparassidae) Dissertation for attaining ...»
Within RTA-clade, four families: Sparassidae Bertkau, 1872 (giant crab spiders), Selenopidae Simon, 1897 (wall crab spiders), Philodromidae Thorell, 1870 (running crab spiders) and Thomisidae Sundevall, 1833 (―true‖ crab spiders) are grouped under name ‗Laterigradae‘. They share the character of laterigrade legs which characterize by a crab-like posture (Latreille 1802). As noted above, most of the spider classification dates back to 19th century and as mentioned by Foelix (2010) that the ―...natural system of classification is still very much a matter of controversy‖.
1.3. Family Sparassidae
The family of huntsman spiders or giant crab spiders, Sparassidae, are composed of small to very large hunting spiders occurring worldwide. Living in various habitats from humid rain forest of Amazon to arid sand dunes of Sahara (Jäger and Kunz 2005; Rheims 2010) and from sea level to high altitudes (~4000) (Moradmand and Jäger 2012a), they represents one of the highly diverse and successful groups of spiders. Sparassidae currently comprises 85 genera and 1132 described species (Platnick 2013). According to previous section on spider classification, Sparassidae are araneomorph, ecribellatae and entelegyne spiders. They are considered dionychans and placed in the RTA-clade but their systematic position and relationships to other families is unclear (Agnarsson et al. 2013).
Simon (1897, 1903), Järvi (1912-1914), Petrunkevitch (1928) and Roewer (1954) proposed classifications for Sparassidae. Simon (1897) classified Sparassidae in seven sub groups (Sparassinae, Heteropodinae, Palystinae, Staianinae, Sparianthinae, Clastinae and Chrosiodermatinae). Hogg (1903) placed most of the Australian endemic genera in his new group Deleninae. Simon (1903) proposed a new subfamily Tibellomatinae. Järvi (1912-1914) proposed Eusparassinae, Polybetinae and Micrommatinae using exclusively the female copulatory characters while previously just those of somatic characters for classification including eyes arrangement and shape of body were applied.
Petrunkevitch (1928) combined Polybetinae and Deleninae under Eusparassinae.
Croeser (1996) revised Palystinae and Jäger (1998) proposed synapomorphies for Heteropodinae and Sparianthinae. But, the status of the majority of Sparassidae subfamilies is still unknown. Moreover, the majority of the Sparassidae genera are not revised and consequently the systematic position of them within the family is quite unknown. The monophyly of the family and currently known subfamilies has never been tested excluding for the endemic Australian subfamily Deleninae (Agnarsson and Rayor 2013).
1.4. Genus Eusparassus and Eusparassinae, a historical review
The genus Eusparassus was erected by Simon (1903) and the type species is Eusparassus dufouri Simon, 1932. The designation of the type species of Eusparassus has a long and a relatively complicated history. It was due to the problematic generic name Sparassus Walckenaer, 1805 and specific name argelasius by Walckenaer (1805). The original description of the genus Sparassus was based on five species currently placed in three genera: three species now in the genus Micrommata Latreille, 1804 [sub S. samaragdulus (Fabricius, 1793), S. roseus (Clerck, 1757) and S. ornatus (Walckenaer, 1802)], a juvenile of Heteropoda venatoria (Linnaeus, 1758) [sub S. pallens (Fabricius, 1794)] and a single male of the genus Olios Walckenaer, 1837 (sub ‗Sparassus argelasius‘) from Bordeaux, France.
Walckenaer (1805) presented no description of ‗S. argelasius‘ which is therefore a nomen nudum. The following year, Walckenaer (1806) provided a description and illustration of this male under the name Sparassus argelasius. Although Simon (1903) doubted that Sparassus was a junior synonym of Micrommata, Jäger (1999) proposed the synonymy of Sparassus with Micrommata. Latreille (1818) examined two female specimens from Spain and described them under the name ‗Micromata argelasia‘ (Micromata is an incorrect original spelling of Micrommata in Latreille (1804)). Unfortunately he misidentified the species (Walckenaer‘s male was an Olios species and Latreille‘s female was an Eusparassus species). Simon (1903) established his new genus Eusparassus, and designated Latreille‘s (1818) misidentified Micrommata argelasia female specimens as the type species. The misidentification by Latreille (1818) was pointed out by Simon (1932) who described E. dufouri as the type species of the genus Eusparassus referring to the misidentified Spanish females of Latreille (1818). Simon (1932) was the first reviser of this case who also presented a description and illustration of Walckenaer‘s ‗Sparassus argelasius‘ under the generic name Olios. The type species of Eusparassus was misidentified under the name ‗E. argelasius‘, thus dufouri was selected as the valid specific name for the type species of the genus Eusparassus by Simon (1932).
For the reasons mentioned above, Eusparassus and Olios were not explicitly diagnosed to date and many misidentifications and misplacement are expected. Even some contributors used the generic name Sparassus (e.g. Levy 1989) for Eusparassus species following Bonnet (1958). Levy (1989) in a brief review on some Eusparassus species (sub Sparassus) re-described E. walckenaeri (Audouin, 1826) and proposed some diagnostic characters to identify Eusparassus species (e.g.
female vulva and colouration of ventral opisthosoma). Prior to current study, Eusparassus comprised 29 nominal species (Platnick 2013) (see Table 1) and most of them were known by a single sex and by their original description.
Moreover, the diagnostic characters within Eusparassus species were incomplete. Early diagnoses for Eusparassus species were based mostly on unreliable and variable somatic characters. Eusparassus is currently placed in Eusparassinae along with some proposed African genera. But, the systematic position of Eusparassus and Eusparassinae genera within Sparassidae is uncertain, since the majority of the Sparassidae genera are not revised to date. Simon (1897) placed Euparassus (sub Sparassus) in Sparassinae. After creating the genus Eusparassus, Simon (1903) placed it along with several other genera in Delenineae.
Järvi (1912) proposed the new subfamily Eusparassinae (sub ―Eusparasseae‖) for Eusparassus along with the genera Pseudomicrommata Järvi, 1914 and Rhitymna Simon, 1897. Jäger and Kunz (2003) proposed the re-establishment of Eusparassinae by noting some synapomorphies and supposed that some endemic African genera to be potentially included in this subfamily. The systematics of these taxa is obscure and no comprehensive taxonomic revision has been carried out so far.
1.5. Aims of my dissertation
The aims of this dissertation are:
First, to revise the genus Eusparassus Simon 1903 (Araneae: Sparassidae) in its entire geographical distribution, to define the genus and to clarify the status of its species and to propose species-groups, based on morphological characters of somatic and copulatory organs.
Second, to apply a wide range of molecular markers of the broadest possible sample of Eusparassus species and Sparassidae genera (with focus on Eusparassinae) to test the monophyly of the morphologically proposed species-groups, genus Eusparassus, subfamily Eusparassinae and family Sparassidae, and also to clarify the position of Eusparassus within Sparassidae and subsequently Sparassidae within RTA-clade by reconstructing phylogenetic trees.
Third, to explore the phylogenetic relationships associated with the distributional patterns and geological events to propose evolutionary scenarios for the origin of Eusparassus and its zoogeography.
Specimens for morphological investigation were mostly obtained from the large spider collections (public and private) in Europe, Africa and North America (see list of ―collections and curators‖ in chapters 3.1 and 3.2).
Examination, measurements and illustration of the specimens were performed using a Leica MZ 165C stereomicroscope equipped with a drawing tube.
Measurements included the prosoma, opithosoma and all leg joints as well as eyes and all eye distances. The diagnostic characters were illustrated and/or photographed including somatic characters: chelicerae (ventral view), anterior part of prosoma focusing on eyes (dorsal view) and copulatory characters: male palp (three views: prolateral, ventral, retrolateral), female epigyne (dorsal and ventral views) and vulvas (anterio-dorso-lateral view). Male palps were dissected from patella joint and were observed in 70% ethanol. Using forceps and fine needles, the hairs covering the bulb and the base of RTA were removed for a better view on the structures.
Female epigynes after dissection and cleaning from hairs and soft tissue surrounding vulva were submerged in 96% lactic acid for clear observation of the internal duct system. A Canon EOS 50D installed on the Microscope was used to photograph the specimens and their structures. Details of characters used for the descriptions and diagnoses of Eusparassus are summarized in Fig. 1.
Subsequent art works on the illustrated and photographed characters were carried out by computer programs Adobe Photoshop CS3, CorelDRAW X5 and Inkspace.
FIGURE 1. Schematic illustration of the structures used for examination, measurements and description of Eusparassus sp.
(a) specimen, dorsal view, Fe: femur, Pa: patella, Ti: tibia, Me: metatarsus, Ta: tarsus; (b, c) eye arrangement and measurement areas: AME: anterior median eyes, ALE: anterior lateral eyes, PME: posterior median eyes, PLE: posterior lateral eyes; 1 AME-AME, 2 AME-ALE, 3 PME-PME, 4 PME-PLE, 5 AME-PME, 6 ALE-PLE, 7 clypeus height at AME, 8 clypeus height at ALE (b dorsal view, c frontal view); (d) chelicera, AT: anterior teeth, PT: posterior teeth, Ide: intermarginal denticles; (e) male palp, ventral; (f) soft trilobate membrane; (g) opisthosoma of female, ventral; (h) female epigyne, ventral.
2.2. Molecular studies Tissues of the fresh sparassid specimens used for molecular studies were mostly obtained from the ―spider tissue collection for DNA analysis‖ (SD) deposited in the Arachnology section, Senckenberg Research Institute. Additional specimens were sampled by the author in Ethiopia (2011) and by colleagues from different parts of the distribution range (see Chapter 3.3: Table 1). The majority of outgroup sequences were extracted from GenBank (see Chapter 3.3: Table 2). For details see chapter 3.3: Material and methods.
Gene selection. Four gene markers were analysed to reconstruct the phylogeny including two mitochondrial genes cytochrome c oxidase subunit I (COI, 648 bp;
barcoding region), 16S rRNA (16S, ~500–510 bp) and two nuclear gene 28S rRNA (28S; ~780 bp) and Histon 3 (H3; 327 bp). These markers represent a comprehensive selection of data from both mitochondrial and nuclear protein-coding and ribosomal genes: COI as the DNA-barcoding region to address species-species relationships (after Barrett and Hebert 2005), 16S to address deeper phylogeny among genera and finally 28S and H3 both nuclear genes with slower gene evolution to address the relationships at the deeper nodes of the phylogenetic trees (after Hausdorf 1999; Dimitrov et al. 2012).
DNA Isolation. Genomic DNA was isolated using CTAB method after Wallace (1987;
Bayer and Schönhofer 2013). Portions of muscle tissues of the legs were cut into small pieces and dried at room temperature, and were subsequently additionally air dried using a heater (at 40°C). The dried tissues were homogenised in 753 µl homogenisation solution [750 µl CTAB (Cetyltrimethylammoniumbromid) (2%), 0.1 M Tris-HCl (pH 8), 1.4 M NaCl (Natriumchlorid), 2.5 mM EDTA (Ethylendiamintetraacetic acid), 2% SDS (Sodium dodecylsulfat, Natriumsalt), 1.5 µl ß-Mercaptoethanol (14.3 M); 1.5 µl Proteinase K (15 mg/ml)]. The mixture left at 60°C for 2–3 hrs (or overnight) on a heating block with shaker for a full digestion. By following centrifugation (13000 rpm, 10 min), the precipitate components and upper foams were removed and supernatant liquid recovered. Using a standard phenolchloroform-isoamylalcohol solution (after Sambrook and Russell 2001), the supernatant was recovered up to three times with 1.5 volumes, followed by centrifugation (12000 rpm, 12 min). During the final chloroform-isoamylalcohol extraction step, the DNA was precipitated using 1/10 volume 3 M Na-acetate, pH 5.2 and 2.5 volumes of ice cold absolute ethanol at –20C. The solution incubated at minus 20C over night. Following centrifugation (12000 rpm, 20 min), supernatant discarded and DNA pellet washed with 500 µl Ethanol 70% (-20C) in ice box followed by final centrifugation (12000 rpm, 5 min, at 4C). The air dried DNA pellet was dissolved in 20 µl ultrapure, sterile H2O. Spectrophotometry was applied to determine the concentration of the extracted DNA.
PCR amplification. For the PCR, the partial fragments of mitochondrial genes COI and 16S and the nuclear genes 28S and H3 were amplified using the primer pairs and PCR termocycling details presented in chapter 3.3: Table 3. The amplification was performed in 25 µl final volume containing 13.95 µl of ultra pure water (dd H 2O),
2.5 µl of 10*Polymerase-buffer, 0.4 µl of each primer (100 pmol/µl), 1.5 µl of dNTPs (2.5 mM), 3.5 µl of MgCl2 (25 mM), 2.5 µl of the genomic spider DNA templates (30– 35 ng/ µl) and 0.25 µl of Taq DNA polymerase. PCR products were purified using the QIAquick PCR purification kit (Qiagen).
Sequencing and editing. The purified fragments were sequenced using BigDye Terminator Cycle Sequencing Kit v. 3.1 using primers as mentioned above.
Sequencing was carried out from both forward and reverse for better evaluation and easier editing. Sequences were edited manually by viewing the chromatographs in BioEdit (Hall 1999) and CodonCode Aligner (v. 4.1.1, Codon Code Corporation). All newly sequenced markers (chapter 3.3: Table 1) will be deposited in the Genbank after acceptance of the manuscript (chapter 3.3). Because of the relatively huge number of samples, most of the DNA extractions, PCR and all sequencing were carried out by Scientific Research and Development GmbH (SRD), Bad Homburg, Germany.