«Joshua Frederick Coulcher UCL Submitted for the Degree of Doctor of Philosophy September 2011 Declaration I, Joshua Frederick Coulcher, confirm that ...»
No morphological structure specific expression pattern was detected by in situ hybridization in any of the numerous in situ hybridization experiments that were performed with labelled RNA probes of the antisense At cnc sequence. Positive control in situ hybridization experiments showed that the in situ hybridization protocol used was successful in detecting the transcripts of At en, At Dll and At Dfd-1.
Comparison of the staining revealed by the antisense At cnc probe compared to the sense probe reveals stronger relative ubiquitous staining. Staining with the sense probe was limited to the yolk cells at the interior of the embryo which is common for in situ hybridizations with all genes, sense or antisense, in this species. It is not possible to determine whether the ubiquitous staining observed with the antisense probe is signal or background. Ubiquitous expression may indicate that At cnc may be performing a housekeeping role that results in ubiquitous expression in all germ tissue.
Lack of specific expression domains present in other arthropods, such as in the labrum, would be evidence of the loss of these expression domains. It is possible that the primitive function of cnc had no developmental role prior to acquisition of developmental roles in the mandibulate arthropod lineage. Although the function of a cnc homologue in C. elegans, skn-1, has roles in patterning mesodermal and endodermal structures during development, which suggests that developmental roles of cnc homologues may be ancestral (Bowerman et al., 1992; Walker et al., 2000).
The sense probe is not a quantitative control however, and the reasons for the non-specific background staining of At cnc antisense probes cannot be determined with any confidence based upon the in situ hybridization expression patterns alone.
Conserved housekeeping role of cnc in responses to environmental stresses cnc has been shown to play important roles in a non-developmental context.
Homologues of cnc have important roles in oxidative and xenobiotic stress responses that have shown to be conserved in diverse organisms.
The lack of an expression pattern in the in situ hybridization experiments described in this chapter (some representative embryos shown in fig.6.8) could represent the actual expression of At cnc. If this is the case, the developmental roles of cnc have been lost in Achaearanea and possible the chelicerate lineage. Alternatively, cnc had no developmental role prior to the acquisition of a developmental role in the mandibulate lineage. Much more work would have to be performed to show this is this case as it is very difficult to prove that a gene plays no role in development based upon a lack of gene expression.
CNC family members, Nrf2 in chordates and cnc in Drosophila, are activated in response to stress and are involved in oxidative stress responses in diverse organisms (Motohashi and Yamamoto, 2004; Sykiotis and Bohmann, 2008; Sykiotis and Bohmann, 2010). The Keap1/Nrf2 signalling pathway is activated in the fruit fly in response to oxidants and xenobiotics. Nrf2 acts in combination with Keap1 to promote organismal homeostasis, inducing anti-oxidant and detoxification responses. The Keap1/Nrf2 pathway is involved in aging. Antioxidant response elements (ARE) have been shown to be activated by Nrf transcription factors to regulate antioxidant stress responses. For example, in Drosophila, isoform C of cnc (CncC) has been shown to bind an ARE sequence upstream of the glutathione S-transferase gene (gstD) involved in the stress response (Sykiotis and Bohmann, 2008). Recently it has been shown that CncC in Drosophila has another role in regulating the 26S proteosome that is critically important for regulating the degradation of proteins and cell cycle control (Grimberg et al., 2011). CncC has also been shown to be necessary and sufficient for transcriptional responses to three xenobiotics: phenobarbital (PB), chlorpromazine, and caffeine (Misra et al., 2011).
Considering the involvement of CNC family members in these housekeeping roles in chordates such as mice and insects like Drosophila, it seems highly likely that At cnc will also perform such a role in Achaearanea.
In chapter four, it was observed that significantly increased mortality rates occurred in parental Tc cncRNAi experiments in Tribolium. As cnc is important for housekeeping roles in Drosophila, Tc cncRNAi may therefore affect organismal homeostasis, in particular the oxidative and xenobiotic stress responses, in adult beetles which could explain the significantly increased levels of mortality. It remains to be tested whether this is actually the case, but considering the conservation of the Keap1/Nrf2 signalling pathway between humans, mice and fruitflies, it is considered highly likely that Tc cnc is important for oxidative stress responses and proteolytic regulatory mechanisms.
What role does cnc perform in Chelicerates?
If the posterior collar domain of mandibulate cnc expression is missing from the first homologous walking leg segment of chelicerate groups, it would provide some evidence that the differentiation of the mandibular from the maxillary segment by cnc is specific to Mandibulata and is evidence that the mandible is a synapomorphy of that group. However, because there was no specific expression domain of At cnc observed, it is not possible to conclude that there is no L1 expression domain of At cnc, which would be indicative of a developmental role in patterning that segment.
It is reasonable to consider it unlikely that cnc will differentiate the first and second leg segments for two reasons. Firstly, the anterior boundary of Dfd homologues are expressed in the first leg segment and the anterior boundaries of the hox genes Scr and ftz are expressed in the second leg segment. Any L1 differentiating function of At cnc would probably be redundant. Secondly, the first and second leg appendages display little differentiation from one another.
This does not rule out the possibility of an appendage patterning role for At cnc. Indeed, the manner in which At cnc may function in the developing leg appendages, or how it interacts with Hox genes in these segments could provide clues as to how cnc evolved to pattern the mandible. For example in Tribolium, Tc cnc represses Tc Dfd and mxp in the mandibular segment and differentiates the mandible, which is protopodal in origin. Therefore, if At cnc is shown to interact with the Hox genes in the limbs or if cnc is expressed in the protopodite of the limbs the path of evolution of cnc mandibular patterning function can be understood. At cnc could be involved in patterning the protopodite of the arthropod ancestor for example, and its function modified to pattern the mandibular protopodite.
There are non-mandibular cnc expression domains in mandibulates that are expressed in structures that are present in non-mandibulates, such as the labrum and the stomodeum. cnc performs a gap gene like role in the labrum of Tribolium and is responsible for patterning labral structures in Drosophila. The labrum is a conserved structure present in all arthropods. It was anticipated that At cnc might be expressed in the labrum. There was however no expression detected in the labrum.
A more distant outgroup to the mandibulates are the onychophorans.
Degenerate PCR was performed to try to clone the cnc homologue present in an Onychophoran Euperipatoides to examine expression in this species. However, these experiments were not successful (data not shown). It would have been interesting to study cnc in the onychophoran, particularly regarding the conservation of the anterior expression domain of cnc in the labrum of insects and myriapods. The labrum has been hypothesized to be homologous to the primary antennae, or at least part of it, of Onychophorans (Budd and Telford, 2009; Eriksson et al., 2010). The labrum is appendage like in that it forms for two paired lobes that express leg patterning genes such as wg and the PD domain genes. The homeobox gene six3 is expressed in the labrum and the primary antennae of onychophorans (Haas et al., 2001; Kimm and Prpic, 2006; Posnien et al., 2009). Expression of cnc in an onychophoran could also have been informative about the role of cnc homologues in the ancestor to arthropods.
It is something of a disappointment that I was unable to determine a specific expression pattern of the spider cnc homologue as it means that I was unable to demonstrate conclusively that At cnc had no developmental function in the L1 segment, and that At cnc has no expression pattern in the labrum. The labrum is a structure of uncertain origin and uncertain homology to other structures. The labrum is purported to be homologous to the antennae of onychophorans and of an appendage nature, it is of immense interest to arthropodologists (Budd and Telford, 2009; Posnien et al., 2009). Tc cnc was shown in chapter four to be necessary for the formation of the labrum, and it would have been a significant result if it was expressed in the labrum of the spider indicating that it has a primitive function in patterning this interesting arthropod structure.
More research would need to be done both to determine if At cnc has lost all developmental functions, and whether it has retained a role in responses to xenobiotics and oxidative stress. An avenue of research that could be pursued is the detection of At cnc protein expression with a cnc specific antibody, which may be more sensitive to detect expression or may reveal specific protein expression domains.
It is entirely conceivable that At cnc has lost all developmental functions in the spider, but it is unfortunately difficult to test this hypothesis. One means of testing whether At cnc has lost developmental functions and is performing a housekeeping role, is to test relative expression levels throughout embryogenesis. Housekeeping genes are typically expressed at a constant level, in contrast to developmental genes which are up-regulated when activated to pattern embryonic structures. Detection of the relative abundance of transcripts by Quantitative PCR (QPCR) could show that At cnc is expressed at uniform levels at all stages of embryogenesis. Alternatively, QPCR could also determine at precisely which stage At cnc is upregulated during embryogenesis. This would facilitate the quest to obtain a tissue specific developmental expression pattern.
8 Chapter 7:Discussion
7.1 General overview of results The primary aim of this research thesis was to study the development of a primitive mandible in a mandibulate arthropod. If the mandible is homologous between insects, crustaceans and myriapods (as hypothesized), it suggests that there may be mandible patterning genes that are shared between these groups. The mandible has probably evolved from a maxilla-like biramous limb present in crustaceamorphs like Martinssonia by modification of an endite into the characteristic mandibular gnathal edge. The Tribolium mandible retains the gnathal edge, with an incisor and molar process, which is the primitive characteristic that distinguishes the mandible from all other appendage types. It was therefore hoped that by studying the Tribolium mandible that three aspects of mandible evolution and development could be explored: i) Discover conserved mandible patterning genes. ii) See if the manner in which the mandible is patterned recapitulates mandible evolution from a maxilla-like precursor. iii) Homologize mandibular sub-structures (like endites and segments) to sub-structures on other appendage types.
The experiments performed in this study have provided some answers to the questions outlined above. The Tribolium homologue of cnc is required to pattern the mandibular segment and differentiates the mandible from maxillary identity. Tc cnc achieves this in part by repression of the Hox genes that are responsible for patterning the maxilla, Tc Dfd and mxp. Tc Dfd patterns protopodite structures in the maxilla and activates protopodite specific domains of expression, which may be similar to what is required to pattern the mandible (a protopodite). All of these results are similar to those observed in Drosophila. In Tribolium however, Tc cnc is activated by Tc Dfd in the mandibular segment whereas in Drosophila, cnc is activated independently of Dfd.