«Joshua Frederick Coulcher UCL Submitted for the Degree of Doctor of Philosophy September 2011 Declaration I, Joshua Frederick Coulcher, confirm that ...»
One of the first victims was the Articulata, which united all panarthropods together with segmented annelids. The Ecdysozoa has replaced it. Ecdysozoa includes panarthropods and the Cycloneuralia (Nematoda, Nematomorpha, Priapulida, Kinoryncha and Loricifera). The Ecdysozoa has strong support from molecular based phylogenies (Aguinaldo et al., 1997; Telford et al., 2008; Rota-Stabelli et al., 2010). The Ecdysozoans are characterized by their ability to moult (ecdysis).
Monophyly of the Arthropoda is strongly supported by molecular and morphological data (Mallatt et al., 2004; Mallatt and Giribet, 2006; Bourlat et al., 2008;
Dunn et al., 2008; Edgecombe, 2010). The extant sister group to the Euarthropoda, the true arthropods, is considered to be either the Onychophora or the Tardigrada (RotaStabelli et al., 2010). Onychophora is more commonly recognised as sister group to the arthropods and has some support from molecular phylogenies (Roeding et al., 2007;
Dunn et al., 2008). A more distant sister group to the Arthropoda are the priapulids and nematodes (Rota-Stabelli et al., 2010).
Tardigrada, the water bears, have been included with Euarthropoda to form the Tactopoda. Onychophora, the velvet worms, do not possess jointed appendages or a hardened cuticle. Arthropods and Tardigrades share a cuticle and jointed legs, although the segmented leg appendages of Tardigrades are telescopic in nature rather than resembling arthropod segmented appendages (Edgecombe, 2010).
Recent molecular phylogenetic analyses as well as morphological analyses have placed the Hexapods with crustaceans to form a monophyletic clade called the Pancrustacea/Tetraconata (Friedrich and Tautz, 1995; Telford and Thomas, 1995;
Boore et al., 1998; Cook et al., 2001). The Hexapoda originated from within the crustaceans, and the crustaceans are therefore paraphyletic. Previous classification schemes had united myriapods, hexapods and onychophorans in the group Uniramia (Manton, 1977), or more traditionally grouped myriapods and hexapods together in the Atelocerata/Tracheata (Snodgrass, 1935) whilst excluding the crustaceans.
Sister group relationships within Arthropoda
The monophyly of the Hexapods, Myriapods and Chelicerates is widely supported, as is the paraphyly of the crustaceans (monophyly of Pancrustacea). Much of the difficulty of constructing an arthropod phylogenetic tree concerns the relationships between these taxa (Edgecombe, 2010). There is no consensus for the crustacean sister group to the insects (Edgecombe, 2010; Jenner, 2010).
Sister group to the Pancrustacea: Mandibulata vs. Myriochelata
Accepting the validity of the Pancrustacea, more recent controversies lie in the position of the myriapods in relation to other arthropod groups, namely the Pancrustacea or Chelicerata (Caravas and Friedrich, 2010; Edgecombe, 2010). If myriapods are grouped with Pancrustacea they form the monophyletic group the Mandibulata. If Myriapods are sister group to the chelicerates then they comprise the monophyletic group the Myriochelata or Paradoxopoda. These two competing phylogenetic hypotheses are illustrated in fig. 1.5.
Acceptance of Mandibulata or Myriochelata has important implications regarding the evolution of the mandible. Acceptance of Mandibulata is compatible with, or even suggestive of, the homology of the mandible across mandibulates. If Myriochelata is valid, this would strongly imply that the mandible has evolved independently in the lineages that lead to the Myriapoda and the Pancrustacea (Mayer and Whitington, 2009) or has reversed into a leg in the chelicerate lineage. If it is found that the mandibular patterning mechanisms (the focus of the present study) are Fig.1.5. Two competing Arthropod phylogenetic relationships, the Mandibulata and Myriochelata hypotheses. Arthropoda are monophyletic and include the extant groups: Hexapoda, Crustacea, Myriapoda and Chelicerata. Onychophora are sister taxa to the Arthropoda. Hexapoda, Crustacea and Myriapoda are mandibulate. The difference between the two hypotheses concerns the position of the Myriapoda. (A) The Mandibulata includes the Myriapoda as sister group to the Hexapoda and Crustacea.
(B) The Myriochelata includes the Myriapoda as sister group to the Chelicerata.
conserved between diverse lineages of Mandibulates, this would provide additional support for the monophyly of mandibulates.
Molecular phylogenies of 28S and 18S ribosomal RNA have tended to support Myriochelata over Mandibulata (Friedrich and Tautz, 1995; Giribet et al., 1996; Mallatt et al., 2004). Phylogenies based on single nuclear genes such as hemocyanin (Kusche and Burmester, 2001) and Hox genes (Cook et al., 2001) have also favoured Myriochelata. However, these phylogenies based on alignment of single genes are highly susceptible to stochastic error artifacts due to the small size of the datasets.
Studies have therefore sought to overcome the limitations of single gene phylogenies by increasing the sample size of the datasets. Phylogenies which are made from concatenated mitochondrial genes support Myriochelata over Mandibulata (Hwang et al., 2001). It has been shown recently that the choice of outgroup will affect support for either Mandibulata or Myriochelata (Rota-Stabelli and Telford, 2008).
Selecting outgroups based upon similarity of nucleotide composition results in phylogenies that favour Mandibulata. An outgroup that is considered more appropriate is the Priapulida. Less appropriate outgroups are the Nematoda and the more closely related Onychophora. Combination of mitochondrial genes and nuclear genes to create a phylogeny have produced mixed results, supporting either Myriochelata (Pisani et al., 2004) or Mandibulata (Bourlat et al., 2008).
With the advance of genome sequencing technologies, phylogenetic studies have started to incorporate numerous genetic loci. Two notable early examples of phylogenomic studies strongly favoured Myriochelata (Dunn et al., 2008; Philippe et al., 2009). These studies were concerned with the phylogeny of the Metazoa and, as a result, sampling within the Arthropoda was limited. More recent phylogenomic studies that have focused on the relationships of Arthropods, has favoured Mandibulata (Regier et al., 2008; Regier et al., 2010; Rota-Stabelli et al., 2011).
There is support for Mandibulata from the presence of characters of rare genomic change, such as micro RNAs (miRNA). miRNA sequences are highly conserved, and are rarely lost from the genome once acquired (Hertel et al., 2006). The presence of particular miRNAs, miR-965 and miR282, in all sampled mandibulates but not chelicerates supports Mandibulata (Rota-Stabelli et al., 2011).
It is clear then that some molecular phylogenies, particularly earlier studies, have favoured the Myriochelata, a clade which draws little support from morphological data. More recent phylogenies that include larger datasets, reduce the presence of long branch attraction, use appropriate models and choice of outgroup and incorporate rare phylogenetic changes are favouring Mandibulata over Myriochelata (Regier et al., 2008; Rota-Stabelli and Telford, 2008; Regier et al., 2010;
Rota-Stabelli et al., 2011). It is apparent that the internal branch lengths leading to either group are short, indicating that presence of a poor phylogenetic signal which is difficult to resolve (Rota-Stabelli and Telford, 2008).
In view of the recent molecular phylogenies in support of Myriochelata, there has been some attempt to identify morphological characters in support of Myriochelata. For example, there are similarities in the development of the nervous system and neurogenesis between spiders and myriapods (Stollewerk and Chipman, 2006; Mayer and Whitington, 2009).
Considering the strong level of support for Mandibulata from morphological data, and the increasing level of support from more recent molecular phylogenetic data, Mandibulata has more overall support than Myriochelata. As more recent phylogenies with larger datasets and more consideration to sources of potential bias favour Mandibulata, on balance, Mandibulata must be considered as the more favourable hypothesis.
1.5 Arthropod Fossil Record in the Cambrian An understanding of the evolution of the mandible requires an examination of the fossil record, to show us when the mandible was likely to have evolved and what kind of appendage it evolved from. Crown group arthropods evolved in the Cambrian.
Supposed fossil representatives of two major extant groups of arthropods, chelicerates and crustaceans, are present during this period.
The fossil record for arthropods during the Cambrian is rather good both in terms of level of preservation (complete specimens, embryos and larval stages) and the variety of organisms found. Numerous fossil arthropods have been discovered at several locations representing the lower Cambrian such as the Chengjiang (Chen, 2009) and the middle Cambrian as represented by the Burgess Shale fauna (505 million years ago). Numerous small, soft-bodied fossils are preserved from the lower and upper Cambrian, known as Orsten fauna (Waloszek and Maas, 2005; Budd and Telford, 2009).
Orsten fossils are very small, preserved specimens - typically larval forms less than two millimetres in size (Maas et al., 2006).
Despite the abundance of Cambrian arthropod fossils, the assignment of these fossils to particular clades of arthropods has not been easy, primarily because of the presence of stem lineage arthropods which display an enormous diversity of forms.
This is in contrast to extant arthropods, which whilst diverse, are more easily organised into groups that possess easily distinguishable body tagma and synapomorphies. The variety of Cambrian stem lineage forms is likely to represent a collection of paraphyletic groups which differ from the crown group, lacking significant apomorphies present in the crown group (Grimaldi and Engel, 2005). Arranging these stem lineage forms which share some characters but not others of the crown group makes assigning taxonomic relationships a complex task. However, increasingly palaeontologists are arriving at a consensus (Budd and Telford, 2009).
The earliest fossil example of a true insect (Ectognatha) is a probably pterygote mandible, Rhyniognatha hirsti that was discovered in the Rhynie chert (dating to about 396-407Mya) which points to the possible presence of wingless insects present on land along with myriapods and chelicerates during the Silurian. The earliest likely hexapod fossil is Rhyniella praecursor that dates from the same period (Entognatha, Collembola) (Whalley and Jarzembowski, 1981; Engel and Grimaldi, 2004).
Cambrian crustaceamorph fossils Identification of an ancestor to all mandibulate arthropods could help in characterizing the primitive mandible. It could provide evidence for particular characters present on the primitive mandible such as the number of segments present on the protopodite, the presence of a gnathal edge with incisor and molar processes, the number of palps and location of the gnathal edge on a particular segment.
Currently there is disagreement on the placing of particular Cambrian arthropods that are obviously ancestral to the Pancrustacea but share numerous characters with crustaceans. These fossils have been described as crustaceamorphs due to their resemblance to crustaceans (Waloszek et al., 2007). One such example is Martinssonia elongata (see fig.1.6).
If either Myriochelata or the Mandibulata hypothesis is true, one challenging problem to explain is the fact that there is no obvious stem lineage myriapod fossil for one hundred million years after the first appearance of crustaceans in the Cambrian.
The lack of any fossils of a particular lineage when it is expected that there should be is known as a ghost lineage. The earliest myriapods that appear in the fossil record in the Silurian about 420 million years ago are Cowiedesmus eroticopodis and Archaedesmus macnicoli from Scotland (Wilson, 2004; Shear and Edgecombe, 2010).
A myriapod ancestor should be present in the Cambrian as there are fossil representatives of crown group crustacean taxa present such as branchiopods like Rehbachiella kinekullensis (Olesen, 2007; Waloszek et al., 2007), Castracollis baltica (Olesen, 2007) and the maxillopods Bredocaris and Dala peilertae (Muller, 1983).