Abstract
Adolescent Idiopathic Scoliosis is a disease of the spine, affecting between 1-3% of the population. This disease manifests as a lateral curvature of the spine that can lead to a range of symptoms such as back pain and respiratory issues. Recent advances in genomics have enabled researchers to identify variations in the genomes of affected individuals associated with AIS. These single nucleotide polymorphisms have been identified near the gene encoding Ladybird homeobox 1 (LBX1 ) transcription factor, which is important for development of limb muscles and the dorsal spinal cord in mice. These SNPs are located in regions that are likely to function as cis-regulatory elements and it has been proposed that a risk-associated SNP leads to increased Lbx1 gene expression. The SNP of interest in this investigation is SNP rs11190870.
The over-arching aim of this research is to understand the molecular basis that underlies the pathology of AIS associated variants, which will allow us to begin to establish a genetic model for Lbx1-associated AIS.
My project aimed to firstly identify and characterise the expression of putative regulators of Lbx1 gene expression, and secondly confirm that they bind at/near these AIS-SNP associated sites. Lastly, I quantified the expression of candidate Lbx1 target genes in a new Lbx1 deletion mouse line.
The SNP of interest to my investigation, AIS-SNP rs11190870, was predicted to alter a DNA binding motif for Pbx1/3 transcription factors. RT-qPCR was carried out to identify if putative regulators of Lbx1 were expressed in the neural tube. In situ analysis showed that spatial expression of these potential regulators appear to overlap. ChIP-qPCR was carried out using EZH2 and PBX antibodies, to see if these were binding at the regions of interest surrounding SNP rs11190870. EZH2 was seen to bind differentially in the brain, at the transcriptional start site of Lbx1 and the SNP rs11190870. I identified binding of PBX antibody to a positive control region but not for the SNP-AIS enhancer region, suggesting that at least at E12.5, PBX does not bind in vivo to the SNP-containing region. ChIP-qPCR analysis confirmed this, suggesting that PRC2 plays a role in repressing Lbx1 gene expression. The literature also predicted that rs11190870 will bind POU3f proteins, however little is known about these in the developing neural tube. I carried out RT qPCR analysis and identified Pou3f1 and Pouf3fb with high expression in the neural tube at E12.5 compared to the developing brain. I then cloned the Pou3f1 gene for probe synthesis, and using in situ hybridisation, found it to be strongly expressed in the dorsal neural tube.
To measure the consequence of altered Lbx1 function, a CRISPR-Cas9 gene edit of a truncated version of Lbx1 was used to create an Lbx1 deletion mouse model. A neuronal stain (nissl) was used to confirm a reduction of dorsal horn neurons in the Lbx1−/− mouse at E15.5. To investigate if there is also a change in expression of putative downstream target genes, RT-qPCR was carried out on wild-type (Lbx1+/+) and Lbx1−/− E12.5 neural tubes. RT-qPCR showed Lbx1 to be under-expressed in Lbx1−/− mouse neural tubes. A number of genes were found to be significantly under-expressed in the neural tubes of the Lbx1−/− mouse, namely Notch1, Zeb1, Robo2, Etv1, and Chmp2b. Lmx1b and Pax2 were found to be significantly over-expressed in the Lbx1−/− mouse.
In conclusion, I have visualised the expression of a number of genes hypothesised to be involved in Lbx1 expression, in the developing E12.5 mouse neural tube. I characterised the presence of potential regulators of Lbx1, and demonstrated some of the implications of Lbx1 knockout, in both structures of the developing neural tube and in gene expression. Future work includes more biological replicates of RT-qPCR to improve the significance of these findings, and investigating the knock-down of the genes I found to be over-expressed and under-expressed in the Lbx1−/− mouse.