«Joshua Frederick Coulcher UCL Submitted for the Degree of Doctor of Philosophy September 2011 Declaration I, Joshua Frederick Coulcher, confirm that ...»
Therefore creating an evolutionary scheme (like fig. 7.1 and 7.2) by placing appendages of stem group mandibulates through crown group mandibulates in a linear series of limb evolution is speculative. These hypotheses of serial homology are conditional on very particular arrangements of taxa in a linear series and are not robust to relatively minor revisions of arthropod phylogeny, or reinterpretations of particular fossils.
Confirmation of suggested homologous relationships requires evidence from more taxa, and for these characters to be mapped onto a phylogeny to determine homology or convergence. The study of PD domain genes together with the notch signalling pathway to mark developing segments complements morphological and palaeontological studies and could provide more objective criteria to establish homology.
7.3 Molecular development of the mandible
Conserved mandibular segment patterning genes in Drosophila and Tribolium Results from chapter four have shown that the mandibular differentiating function of cnc by repressing Dfd is conserved in Tribolium. Despite the obvious larval and adult morphological differences, there are numerous conserved genetic functions and interactions between Tribolium and Drosophila in the gnathocephalon of the developing embryo. This indicates that the function of genes that pattern the gnathocephalon of Tribolium and Drosophila are conserved, and is evidence that mandibular patterning mechanisms may be conserved in other mandibulates.
The similarities are evident in the expression of genes, shown in fig.7.3A-C. The Hox genes Dfd and pb are expressed in similar domains (fig. 7.3B), as are cnc and Dfd, with repression of Dfd from the mandibular appendage (fig. 7.3A). There is similarity in the expression of prd and Dll in the proximal and distal regions of the maxillary lobe. Tc cnc also represses Tc Dll and the Hox genes Dfd and mxp (fig. 7.3C).
Fig.7.3 Comparison of gene expression and gene regulation in the gnathocephalon, the mandibular, maxillary and labial segments of Tribolium and Drosophila. (A-C) Conserved expression of genes in the mandibular, maxillary and labial segments of Tribolium and Drosophila. Tribolium gnathal lobes are on top, Drosophila gnathal lobes are on the bottom. Expression patterns are very similar between both species, with similarities in both segmental expression domains and proximal-distal domains. Proximal (P) and distal (D) axis are indicated. (A) Expression of cnc and Dfd. (B) Expression of Dfd and pb. (C) Expression of Dll and prd. (D) Summarized genetic interactions of Tc Dfd and Tc cnc in Tribolium.
Interactions are similar to Drosophila, although Tc dac upregulates the proximal domain of Tc dac in the maxillary segment in Tribolium which does not occur in Drosophila. (E) Summarized genetic interactions of Dfd and cnc in Drosophila. Interactions are similar to Tribolium. Dfd activates ser and pb which does not occur in Tribolium.
There are several shared genetic interactions between Tribolium and Drosophila that are representative of their expression patterns (fig. 7.3D,E). In both species, Dfd patterns proximal structures that are derived from the gnathal lobes and regulates genes that have expression domains on the proximal part of the gnathal lobes. The proximal domain of Dll and the maxillary prd domain are regulated by Dfd.
There are some differences, like the regulation of ser by Dfd, and mxp by cnc which occur in Drosophila but not in Tribolium. More striking is the activation of cnc by Dfd which I have shown occurs in Tribolium but not Drosophila.
The phylogenetic relationship between these two members of the Holometabola has been examined recently in a phylogenomic study that showed that hymenopterans are more distantly related to Diptera than are the Coleoptera (Savard et al., 2006). This indicates that the genetic mechanisms patterning the mandibular and maxillary segments are common to the Coleoptera, Diptera and Lepidoptera but cannot necessarily be extrapolated to the more distantly related Hymenoptera.
Considering the conserved expression of cnc, Dfd and pb in diverse mandibulates, the mechanisms of patterning the mandibular and maxillary segments may also be conserved across mandibulates.
How Tc cnc and Tc Dfd pattern the mandible in Tribolium.
In this section, based upon evidence from chapters four and five, an outline of mandibular segment patterning in Tribolium will be presented. There is some consideration given to the role that cnc plays in patterning the mandibular segment in Drosophila as the experiments performed on the molecular developmental functions of cnc have revealed particular details that may be of significance for all taxa that have mandibular segments patterned by cnc. This hypothetical outline of mandible patterning may resemble the primitive functions of these genes in stem lineage mandibulates, with some differences concerning the presence of primitive mandibular palps. Following this section I discuss the likelihood of the conservation of this mechanism by comparing Hox gene expression in diverse mandibulates and nonmandibulates.
Here is a hypothetical scheme of the mandible patterning functions of Tc cnc
and Tc Dfd (see fig. 7.4):
1) Tc Dfd, as one of its roles as a Hox gene, is required in the mandibular segment to repress antennal development and establish the post-antennal appendage P/D axis, as defined by the PD domain genes (fig. 7.4A).
2) Tc Dfd activates Tc cnc in the mandibular segment. Tc cnc begins to differentiate the mandible from the maxilla. Tc cnc represses palp development in the mandibular segment by repressing the palp domain of mxp(pb) (fig. 7.4B).
3) Tc Dfd patterns the protopodite, including the endites of the mandible and maxillary segments (fig. 7.4C).
4) As soon as the endites start to develop, Tc cnc represses Tc Dfd from the mandibular endites, and Tc cnc differentiates the mandible endite from the maxillary endites (fig. 7.4D).
Fig.7.4 A model of Tc cnc and Tc Dfd mandibular and maxillary patterning functions in Tribolium castaneum. (A) Tc Dfd represses antennal development and establish the post-antennal appendage P/D axis, as defined by the PD domain genes. (B) Tc cnc is activated by Tc Dfd in the mandibular segment. Tc cnc begins to differentiate the mandible from the maxilla, and represses the ectodermal telopodite domain mxp. (C) Tc Dfd patterns the protopodite, including the endites of the mandible and maxillary segments. As soon as the endites start to develop, Tc cnc represses Tc Dfd from the mandibular endites, and Tc cnc differentiates the mandible endite from the maxillary endites. (D) Tc Dfd expression is repressed from the majority of the mandibular limb bud.
Hox genes have been shown to repress antennal identity (Brown et al., 2000). It can be inferred that the activation of the post-antennal appendage P/D axis pathway is achieved by Hox genes. In the maxillary segment this role is performed by the presence of Tc Dfd and mxp. In the mandible, Tc Dfd would perform this function.
It was shown in chapter four that Tc cnc is activated in the mandibular segment by Tc Dfd. It was also shown that Tc cnc represses mxp expression in the mandibular segment and represses telopodite development which indicates that at an early stage Tc cnc is functioning to differentiate the mandibular segment from the maxillary segment, as mxp is expressed in limbs at an early stage of development.
Tc Dfd promotes protopodite development and patterns the endites in the maxilla, probably by up-regulating or activating gene expression such as the proximal domain of Tc dac and Tc prd, and the proximal domain of Tc Dll. These results were described in chapter five. The proximal domain of Tc dac is expressed early in the mandibular and maxillary limb buds, as soon as they start to form. It is assumed, that Tc Dfd functions in a similar manner in the mandible however this has to be tested.
As soon as the endites of the mandible and maxilla start to develop, Tc cnc represses Tc Dfd in the mandibular endite. This is evident in comparison of Tc cnc and Tc Dfd expression, as Tc Dfd expression is repressed from the position of the endite during early limb stages of embryogenesis shown in chapter four.
Tc cnc at this stage differentiates the mandibular endite from the maxillary endite. Tc cnc promotes an alternative developmental pathway of the mandibular endite, which is evident in the differential expression of Tc prd in the mandible compared to the maxilla. Knock down of Tc cnc resulted in a reiteration of the maxilla domains of Tc prd expression.
One important piece of evidence lacking from the framework above is evidence that shows Tc cnc is an activator of mandibular patterning genes in Tribolium, and whether Tc cnc is sufficient to pattern mandibular structures. cnc may play such a role in mandibulate arthropods like Tribolium. Tc cnc clearly functions to differentiate a mandible from a maxillary appendage, and it has been shown to have activator functions in Drosophila. The mandibular appendage obviously requires genes to pattern it. It is highly unlikely that repression of Tc Dfd by itself will result in the patterning of a mandibular appendage. Other genes will be activated to pattern the appendage. If Tc cnc is not directly involved in the activation of these mandibular patterning genes, then somehow Tc cnc indirectly activates them as Tc cnc is required to differentiate the mandible from a maxilla.
Numerous lines of evidence suggest that cnc in Drosophila functions as an activator and that any repression of Hox genes by cnc is likely to be indirect (Veraksa et al., 2000). Cnc possesses a strong activating domain. The cnc binding sequence, for example, when placed next to Dfd response element results in increased Dfd activity, not a reduction of activity which would be expected from a repressor.
cnc has been shown to repress both Dfd transcription and Dfd activity even when Dfd transcripts are present. It is hypothesized that Cnc represses Dfd protein which prevents the activation of the Dfd autoactivation circuit. This would result in repression of transcription. Cnc has been shown to physically interact with Dfd protein, although the significance of this is not known.
It has been noted that cnc has a function in patterning some mandibular segment derived structures independently of Dfd (Veraksa et al., 2000). Ectopic activation of cnc in Drosophila embryos results in the production of hypopharyngeal structures (the T-ribs), which are derived from the hypopharyngeal lobes, which are part of the mandibular segment (Economou and Telford, 2009). These ectopic mandibular segment derived structures are present along the ventral midline of the Drosophila embryo where there is no Dfd expression. This result indicates that cnc is necessary and sufficient to pattern some mandibular segment derived structures, independently of Dfd.
An instructive experiment would be to test whether Tc cnc is sufficient to pattern mandibular structures by ectopically activating Tc cnc throughout the developing Tribolium embryo. If ectopic mandibular structures were present ubiquitously, it would prove that Tc cnc was a master controller gene sufficient for at least some aspects of mandible development. If ectopic mandibular structures were formed in the maxilla and labial endites only, it would show that endite formation is required prior to Tc cnc mandible patterning function. If mandibular structures were only present in the maxillary segment then it would show that Dfd is required for cnc to pattern mandibular segment structures. This experiment would answer the question of whether Tc cnc is directly responsible for activating the alternative endite developmental pathway of the mandibular endite.
Another experiment that could show similarities between cnc function in Drosophila and Tribolium would be to ectopically express Tc cnc in Drosophila embryos to see if it could rescue the wild type phenotype in cnc mutants and also to see if ubiquitous ectopic activation resulted in the production of ectopic hypopharyngeal lobe derived structures.
The more that is understood about the molecular mechanism of mandibular patterning, the more that is likely to be understood about the evolution of the mandible by comparison to known genetic functions. Comparison to other taxa would also be easier and more informative as a result.
7.4 The role of cnc in the ancestor to all mandibulates