«Joshua Frederick Coulcher UCL Submitted for the Degree of Doctor of Philosophy September 2011 Declaration I, Joshua Frederick Coulcher, confirm that ...»
The morphology of the maxilla palp is affected in Tc Dfd knock down embryos, although the precise details were not deciphered. The base of the transformed maxillary palp is larger in developing embryos, lacking the characteristic inverse taper as the diameter of the maxilla palp decreases proximally to a narrow base that attaches to the maxilla protopodite. In first instar larvae, the maxilla is smaller in DfdRNAi larvae than wild type larvae. There also appears to be a reduced number of segments in the larval palp. However, it was difficult to detect expression domains in the maxillary palp in general, and it was also difficult to detect segmentation in the larval cuticle preparations of both wild type and knock down larvae so more research would have to be performed on the role of Tc Dfd in the palp to be more certain.
The identity of the proximal maxilla appendage.
It has been shown previously by Brown et al. that the Hox gene necessary for proximal labial identity the Tribolium homologue of Scr, Cephalothorax (Cx), is not expressed in the maxillary segment of Tc Dfd mutant embryos (Brown et al., 2000).
This result shows that the affected proximal maxilla is not of labial identity, despite resemblance of the larval maxilla to half the labial appendage (which is made up of two separate appendages, left and right, which later fuse centrally during embryogenesis). It has been shown in Drosophila that Dfd activates pb. Expression of mxp, the Tribolium orthologue of pb, has not been tested14 in Tc Dfd knock down larvae or mutants. Therefore in order to confirm that mxp is still expressed in the maxillary appendage in Tc Dfd knock down embryos expression was detected by in situ hybridisation.
At least mxp expression has not been studied in Tc Dfd mutants in the published literature.
Without a Hox gene present in the base of the maxillary appendage in Tc DfdRNAi embryos, it is not immediately obvious what the identity of the base of the transformed maxillary appendage will be. It is likely that the presence of mxp will repress antennal development in the maxilla and activate the post-antennal P/D axis genetic pathway.
The affected maxillary protopodite could have a generic non-antennal proximal identity lacking the specific input of Hox genes. The resemblance of the Tc DfdRNAi maxilla to half of the labial appendage may represent the sharing of default P/D axis patterning mechanism at the base of the maxillary appendage as there is no Hox gene expressed at the base of the appendage. The maxilla is patterned in an additive fashion by two Hox genes, Tc Dfd and mxp. Expression of a Hox gene is required in the ectoderm for appendage specification. Without a Hox gene present, the appendage primordia patterns antennal appendage by default. Considering that in Tc DfdRNAi embryos, there is still a Hox gene present that can repress antennal identity, mxp, means that it is unlikely that the base of the maxilla has been transformed to antennal identity.
It is possible that another PD domain gene is affected by loss of Tc Dfd function.
Tc hth is expressed in the mandible and the protopodite of all post-antennal appendages. In Drosophila, the proximal coxal ring of ser expression is activated by hth (Rauskolb, 2001). Considering that the proximal domains of Tc ser are apparently unaffected by Tc Dfd knock down in Tribolium, if Tc hth has a conserved function to activate proximal domains of Tc ser, it is reasonable to assume that Tc hth will not be affected by Tc Dfd Knock down. This will have to be tested in order to prove this hypothesis.
Comparison of Tc Dfd function with Dfd in Drosophila
The function of Tc Dfd in Tribolium is very similar to the function of Dfd in Drosophila. Both genes pattern structures derived from the proximal part of the maxillary gnathal lobe. In Drosophila, Dfd is necessary to pattern mandible and proximal maxillary gnathal lobe derived structures such as the cirri and ventral organ of the maxillary gnathal lobe which are hypothesized to be homologous to the endites of the maxilla (Jurgens et al., 1986). Tc Dfd is necessary to pattern the maxilla protopodite, especially the endites, and the mandible. There are also similar genetic interactions present in both species. In both Drosophila and Tribolium, Dfd regulates the proximal domain of Dll and regulates prd in the maxillary gnathal lobe. There are however several differences between Dfd in Drosophila and Tribolium. In Tribolium, Tc Dfd regulates Tc dac and activates Tc cnc. In Drosophila, Dfd has been shown to regulate ser and mxp which does not occur in Tribolium as is shown above.
Dfd is necessary to activate three genes that are important for patterning the proximal part of the maxillary gnathal lobe in Drosophila. Dfd activates the proximal domain of Dll by a maxillary-specific enhancer (called ETD6) and is required for the formation of proximal maxillary lobe derived structures, the cirri (O'Hara et al., 1993).
Dfd activates the late expression domain of prd in the proximal region of the gnathal lobes (Gutjahr et al., 1993; Li et al., 1999). prd is necessary for proximal maxillary lobe derived structures (such as the cirri and the ventral organ) (Vanario-Alonso et al., 1995). ser is also a target of Dfd in the mandibular and maxillary lobes. ser is required for normal mouth hook development (Wiellette and McGinnis, 1999).
In Drosophila ser is regulated by Dfd. Tc ser does not appear to be regulated by Tc Dfd as Tc ser ring domains are present in knock down embryos. However, it was not possible to determine whether there is any effect on more distal domains of Tc ser¸ primarily because the staining of Tc ser is weak in the distal part of the maxilla.
Comparison of the co-expression of PD domain genes with Tc ser shows that the three proximal domains of Tc ser in the affected maxilla are still present (fig.5.6.) Another significant difference relates to the regulation of dac. In Tribolium, Tc Dfd regulates the proximal domain of Tc dac, whereas in Drosophila dac is not expressed in the developing gnathal lobes of Drosophila and therefore is not be regulated by Tc Dfd (Tomancak et al., 2002; Tomancak et al., 2007).
In Drosophila, pb is activated by Dfd in the mandibular and maxillary segment (Rusch and Kaufman, 2000). In Tribolium, mxp is not activated by Tc Dfd. Knock down of Tc Dfd did not result in any decrease in mxp in the maxillary segment. In the mandibular segment mxp is expressed in the posterior of the transformed mandibular appendage.
As shown in chapter four, Tc Dfd activates the posterior collar domain of Tc cnc in the mandibular segment. This is in contrast to the Drosophila, where it has been shown that Tc cnc is activated by gap genes and pair rule genes and is activated independently of Hox genes (Mohler, 1993).
Experiments on other Mandibulates
Tc Dfd is important for patterning the protopodite of the maxilla in Tribolium. In the maxilla protopodite, Tc Dfd regulates two genes, Tc dac and Tc prd which also have pronounced expression domains in the mandibular endite. Therefore, it is likely that Tc Dfd will also activate Tc dac and Tc prd in a similar manner in the mandible. It is anticipated that the mandibular expression domains of these two genes are modified by Tc cnc after their initial activation by Tc Dfd, as part of the genetic program that specifies mandibular identity.
The conservation of the proximal patterning function of Dfd in Tribolium and Drosophila and the similarity of expression of Dfd orthologues in other mandibulates suggests that the proximal maxilla patterning function of Dfd may be conserved in other mandibulate organisms. If the maxilla-to-mandible differentiating function of cnc is found to be conserved in mandibulates, as comparative expression data suggest, then an understanding of how the mandible evolved from a maxilla-like precursor from a molecular developmental perspective can be obtained by comparing the development of both the mandible and the maxilla in diverse mandibulates.
In chapter seven, I will investigate expression the findings from the previous three chapters regarding the role that Tc cnc has been demonstrated to have in Tribolium. I will outline a hypothesis of how Tc cnc functions to pattern the mandibular appendage of Tribolium together with the protopodite patterning function of Tc Dfd as outlined in this chapter. I will then conclude as to what the ancestral patterning genes of the gnathocephalon in Mandibulata may have been based upon by comparing the expression of the genes necessary to pattern the gnathocephalon of Tribolium to the expression of these genes in other mandibulates, with particular focus on the evolution of the mandibular endite.
6.1 Introduction In order to study how conserved mandible patterning genes evolved in the ancestor to mandibulates it is necessary to study these genes in outgroups of mandibulates to try to determine their ancestral function. cnc and Dfd are two good candidate genes to have this role as shown in chapters four and five, therefore an investigation into the expression of the homologue of cnc in a non-mandibulate was undertaken. It was originally planned that the function of the spider cnc homologue and the two spider homologues of Dfd would be investigated. However, gene knockdown by parental RNAi was unsuccessful in positive controls and was therefore not continued.
It is the contention of this thesis that the arthropod mandible evolved from a leg through a maxilla-like precursor to form the mandible. Evidence that the mandible has evolved from a leg is evident from comparisons of the mandibular segment to the homologous segment in non-mandibulate arthropods and stem lineage arthropods (Waloszek et al., 2007; Chen, 2009). In all non-mandibulates such as chelicerates, there is a locomotory leg present on the homologous segment to the mandibular segment.
Non-mandibulate Cambrian arthropods such as the Lamellipedia (which includes trilobites), megacheirans and other stem group arthropods are characterised by numerous undifferentiated serially homologous biramous leg appendages, one of which is serially homologous to the arthropod mandible (Boxshall, 2004).
Evolution of the mandible from a maxilla-like precursor is evident in the structure of maxilla-like appendages present on numerous stem lineage crustaceamorph fossils present on the second post-antennal segment (Muller and Waloszek, 1986; Siveter et al., 2001; Waloszek et al., 2007). The Hox gene Dfd is expressed in the mandibular segment and patterns the mandible in combination with cnc in Tribolium and Drosophila. Without cnc, Dfd patterns maxillary structures. This genetic interaction may represent an important part of the evolutionary history of the mandible, with the original function of Dfd involved in a maxilla or biramous limb appendage specifying function and cnc recruited to differentiate the mandible protopodite from the maxilla or biramous limb protopodite.
cnc is a basic Leucine zipper (bZIP) member of which examples are present in the majority if not all Bilaterians (Mohler et al., 1991; Grimberg et al., 2011). bZIP proteins are characterized by a conserved domain that includes a basic DNA binding domain upstream of a Leucine zipper that consists of several heptad repeats. The repeating residue in the heptad repeat is typically a Leucine, or another hydrophobic residue such as Isoleucine or Valine. These hydrophobic residues are able to ‘interlock’ with other hydrophobic residues present on other Leucine zipper proteins to form dimers. Numerous CNC family members are only capable of forming heterodimers due to the presence of charged amino acid residues in the Leucine zipper domain that prevent homodimerization (Mohler et al., 1991). cnc present in Drosophila is one such example; it is unable to form homodimers and forms heterodimers with a small Maff protein MafS (Veraksa et al., 2000). The Caenorhabditis elegans CNC family member Skn-1 has lost the Leucine zipper region of the bZIP domain and consequently is able to function as a monomer (Walker et al., 2000).