Significant Genes
In my analysis I identified 260 statistcially significant genes, as described in the ‘Analysis’ section. Of the statistically significant genes identified, 137 were upregulated and 123 were downregulated.
It is important to note that statistical significance does not equate to biological significance, it simply means that in the statistics tests performed using this dataset we are 95% confident that the genes classed as up or down regulated are not due to chance. We have yet to prove the causal relationship between these genes and Archenteron development.
In order to categorize the genes identified I used a database of gene expression throughout development in the whole S. purpuratus embryo (http://legacy.echinobase.org/shiny/quantdev/) (Tu, Cameron & Davidson). This database contained information about the class, expression level, what cluster the gene was located on, the peptide sequence for the protein coded for etc. allowing me to quickly search through the significantly expressed genes to find those of interest.
The potential effector genes (those coding for proteins which actually change the morphology and function of the cell) were of various classes, such as nervous, adhesion, metabolism etc, however, as previously mentioned, it is the Transcription Factors and Signaling genes which form the Gene Regulatory Network, and thus these are the ones we are most interested in.
From my analysis I was able to identify 9 transcription factors and 5 signaling genes as being significantly expressed, these are shown below.
Upregulated: Red; Downregulated: Blue
Archenteron GRN
The GRN was modelled using BioTapestry (Version 7.1.2) (Paquette & Leinonen, 2016). Typically, for an accurate Gene regulatory network, there are many steps taken to interrogate each possible linkage between genes.
Firstly, differentially expressed genes of interest must be identified through RNA sequencing analysis as done in this research. These genes are then visually confirmed as differentially expressed through in situ hybridization.
After each gene has been identified and confirmed, they must be perturbed one at a time so that the consequence of gene knockout can be found. The results of gene perturbation experiments using morpholinos is used along with other data such as the gene expression profile for the whole embryo during development and the transcript elongation rate, RNA half-life and functional time to determine what order each gene acts in and what effect they have on each other.
While signaling pathways can be blocked at any point in the development of S. purpuratus it is not possible to do this for transcription factors, the morpholino must be injected at the start of development before the first cleavage to ensure that the specific gene is knocked out in every cell in the embryo. Therefore, there is a compiling effect of the initial gene knockout over time, as we cannot determine if the effect is a direct or an indirect consequence of the knockout, this is why the gene expression profile and the time an expressed gene takes to act must be considered. This also means it is difficult to determine the effects after the first wave of expression of the gene.
For this internship, as already stated, our objective was not to follow these steps and create an accurate GRN model which reflects the biological processes occurring in the embryo during Archenteron formation, but to familiarize myself with the research process and better my understanding of the field under the guidance of Dr. Barsi. For the GRN modelled in this internship, in order to form linkages between Transcription Factors and Signaling genes identified as significantly expressed, the expression profile for the whole embryo during development and other literature available was used.
The first step of creating this GRN was to look at the gene expression profile (Figure 1) for each transcription factor and signaling gene identified (Table 1), in order to see how expression levels, change over time and so that possible relationships can be inferred.
However, this correlation itself is not enough to determine a link between two genes, we had to use information available online in other research involving these genes and the endoderm development in S. purpuratus.
Using the resources I had, I created a model Gene Regulatory network (figure 2) containing the genes identified in my analysis. While we know this model is not accurate, putting it together has been essential for learning more about this field of research and how this process would normally be undertaken. While I was not able to collect experimental data myself for each gene, I did collect evidence from online open access journals to support each link I have made.
Top: Signaling genes; Bottom: Transcription Factors
Evidence
i.
“These data suggest that components of endoderm specification are lost when Wnt6 is knocked down”
FoxA is not expressed following knockout of Wnt6, this suggests that FoxA is enhanced directly or indirectly through the Wnt and β-Catenin signaling pathway by Wnt6.
As Wnt6 is upregulated in the Archenteron, it could be doing this through autocrine signaling.
(Croce, et al., 2011)
ii.
“Follistatin increased the Foxa2 positive endoderm fraction to 78%”
“Follistatin might interact with other pathways such as Wnt [61] and BMP [62] to induce differentiation”
This data suggests that Follistatin enhances FoxA expression directly or indirectly.
(Parashurama, et al., 2008)
iii.
“Erk Signaling regulates differentiation into the mesoderm and endoderm lineages…Grb2…delivers activation signals to Erk”
“Gene knockout experiments show that… [Grb2 deficient] … embryos fail to develop a primitive endoderm layer and die in blastocyst outgrowths or in vivo”
(Liu, et al., 2013)
Grb2 and FoxA are both essential for endoderm development indicating they could be linked in the archenteron. Grb2 expression could affect the Archenteron through the RTK and TGF-β pathways.
iv.
Chordin is downstream of nodal and is a BMP antagonist
(Lapraz, Besnardeau, & Lepage, 2009)
“Chd is a large, secreted protein that inhibits BMP from transducing signals by binding BMP and preventing its interaction with its receptor”
Redundancy of Chordin as Noggin and Follistatin also bind with BMPs.
Therefore, Chordin and Follistatin both repress the BMP2/4 signalling pathway
(Bradham, et al., 2009)
v.
Dlx is target gene of BMP2/4, Chordin antagonizes this pathway, repressing expression of Dlx
(Koop, et al., 2017) (Runcie, et al., 2013)
Dlx regulated by BMP and Wnt signaling, and it reinforces early ectodermal patterning.
(Woda, Pastagia, Mercola, & Artinger, 2003)
vi.
FoxA is an auto repressor
(Davidson, et al., 2002)
vii.
Ets1 activates alx1
(Ben-Tabou de-Leon & Davidson, 2009)
viii.
Alx1 is an auto repressor
(Ben-Tabou de-Leon & Davidson, 2009)
ix.
(Lamoille, Xu, & Derynck, 2014)
HGF signaling pathway involved in affect of FoxA on target genes
Fox family can stimulate Hgf signaling pathway
(Bach, et al., 2018)
x.
Lbx peaks at the same time as FoxA, suddenly increasing expression from 0% to 100% just after FoxA begins increasing between 10 and 24 hours.
xi.
Lbx affected by ets1 knockdown
(Rafiq, Shashikant, McManus, & Ettensohn, 2014)
Ets1 repressed in endoderm, therefore must be a repressor of Lbx
Ets1 also activates Dlx which Lbx represses.
xii.
“Endodermal neurogenesis is mediated by Six3 and Nkx3-2”
“Nkx3-2 is expressed initially throughout a significant fraction of the foregut endoderm but later in only a subset of these cells”
“SoxB1 and expression of Six3 and Nkx3-2 could specify the foregut as neuroendoderm”
(Wei, Angerer, & Angerer, 2011)
Nkx3-2 possibly is repressed/inactive at the stage of our research, repressor releases/it is activated soon after, as also indicted by the max expression of nkx3-2 being at 40 hpf on whole embryo expression database.
Grb2.1 and FoxA levels fall after 30 hpf, these could be responsible for repression of Nkx3-2, then allowing it to be activated by Soxb1 and Six3.
xiii.
Dlx is activated by ets1
(Rafiq, Shashikant, McManus, & Ettensohn, 2014)
Dlx not affected by C59, therefore not affected by knockout of Wnt signaling and not downstream of FoxA or Wnt6.
(Cui, Siriwon, Li, Davidson, & Peter, 2014)
xiv.
Dlx repressed by Lbx (Kioussi & Gross, 2008)
xv.
Nkx3-2 is also affected by ets1 expression. Both are downregulated in the Archenteron, therefore ets1 possibly activates Nkx3-2.
(Rafiq, Shashikant, McManus, & Ettensohn, 2014)
xvi.
E78 affected by Alx1 knockdown, E78 is upregulated, Alx1 is repressed, therefore Alx1 likely represses E78.
(Rafiq, Shashikant, McManus, & Ettensohn, 2014)
“E78 also has an early embryonic function unrelated to sex-specific development”
“Genes assigned to the “pattern” class may have roles in specifying…major embryonic events such as gastrulation (e.g., E78)”
(Bodofsky, Koitz, & Wightman, 2017)
On graph of expression, sudden jump to max at 18 hours could indicate the release of a repressor
There is nothing else to indicate what causes the expression of E78.
Unlabeled Links:
All other links, from the ‘Inputs not present in Archenteron’ and ‘Inputs present in Archenteron’ boxes, are present to illustrate activation/deactivation of genes for which we did not have experimental evidence or research available to support a link from another gene.
These links also represent the different combination of factors that we have not described but know are present (Such as the effect of β-Catenin) which will cause differential expression of genes in the respective tissues upstream of the GRN modelled.
Bibliography
Bach, D., Long, N., Luu, T., Anh, N., Kwon, S., & Lee, S. (2018). The Dominant Role of Forkhead Box Proteins in Cancer. Int J Mol Sci. doi:10.3390/ijms19103279
Ben-Tabou de-Leon, S., & Davidson, E. (2009). Experimentally based sea urchin gene regulatory network and the causal explanation of developmental phenomenology. Wiley Interdiscip Rev Syst Biol Med. doi:doi:10.1002/wsbm.24
Bodofsky, S., Koitz, F., & Wightman, B. (2017). Conserved and Exapted functions of Nuclear Receptors in Animal Development. Nucl Receptor Res. doi:10.11131/2017/101305
Bradham, C., Oikonomou, C., Kuhn, A., Core, A., Modell, J., McClay, D., & Poustka, A. (2009). Chordin is required for neural but not axial development in sea urchin embryos. Dev Biol.
Croce, J., Range, R., Wu, S., Miranda, E., Lhomond, G., Peng, J., . . . McClay, D. (2011). Wnt6 activates endoderm in the sea urchin gene regulatory network. Development, 138(15), 3297-3306. doi: 10.1242/dev.058792
Cui, M., Siriwon, N., Li, E., Davidson, E., & Peter, I. (2014). Specific Functions of the Wnt signaling system in gene regulatory networks throughout the early sea urchin embryo. Proceedings of the National Academy of Sciences, 111(47), 5029-5038. doi:10.1073/pnas.1419141111
Davidson, E., Rast, J., Oliveri, P., Ransick, A., Calestani, C., Yuh, C., . . . Cameron, R. (2002). A Provisional Regulatory Gene Netwrok for Specification of Endomesoderm in the Sea Urchin Embryo. Developmental Biology, 162-190. doi:10.1006/dbio.2002.0635
Kioussi, C., & Gross, M. (2008). How to Build Transcriptional Network Models of Mammalian Pattern Formation. PLoS ONE. doi:10.1371/Journal.pone.0002179
Koop, D., Cisternas, P., Morris, V., Strbenac, D., Yang, J., Wray, G., & Byrne, M. (2017). Nodal and BMP expression during the transition to pentamery in the sea urchin Heliocidaris erythrogramma: insights into patterning the enigmatic echinoderm body plan. BMC Developmental Biology.
Lamoille, S., Xu, J., & Derynck, R. (2014). Molecular Mechanisms of epithelial-mesenchymal transition. Nat Rev Mol Cell Biol., 15(3), 178-196. doi: 10.1038/nrm3758
Lapraz, F., Besnardeau, L., & Lepage, T. (2009). Patterning of the Dorsal-Ventral Axis in Echinoderms: Insights into the evolution of the BMP-Chordin Signaling Network. PLoS Biol.
Li, E., Cui, M., Peter, I., & Davidson, E. (2014). Encoding regulatory state boundaries in the pregastrular oral ectoderm of the sea urchin embryo. Proc Natl Acad Sci USA.
Liu, Y., Liu, Q., Jia, W., Chen, J., Wang, J., Ye, D., . . . Kang, J. (2013). MicroRNA-200a Regulates Grb2 and Suppresses Differentiation of Mouse Embryonic Stem Cells into Endoderm and Mesoderm. PLoS ONE.
Parashurama, N., Nahmias, Y., Cho, C. H., van Poll, D., Tilles, A. W., Berthiaume, F., & Yarmush, M. L. (2008). Activin alters the kinetics of endoderm induction in embryonic stem cells cultured on collagen gels. Stem Cells, 26(2), 474-484. doi: 10.1634/stemcells.2007-0303
Rafiq, K., Shashikant, T., McManus, C., & Ettensohn, C. (2014). Genome Wide analysis of the skeletogenic gene regulatory network of sea urchins. Development.
Runcie, D., Garfield, D., Babbitt, C., Haygood, R., Nielson, W., & Wray, G. (2013). The Impact of Gene Expression Variation on the Robustness and Evolvability of a Developmental Gene Regulatory Network. PLoS Biology.
Wei, Z., Angerer, R., & Angerer, L. (2011). Direct Development of neurons within the foregt endoderm of sea urchin embryos. Developmental Biology, 9143-9147.
Woda, J., Pastagia, J., Mercola, M., & Artinger, K. (2003). Dlx proteins position the neural plate border and determine adjacent cell fates. Development, 331-342.
BIOS & Barsi Lab Internship 2021 