Cheryl Gregory-Evans

The majority of ALS patients have a sporadic form of ALS (sALS), which lack evidence of a dominantly inherited genetic factor. sALS has long been associated with environmental factors such as toxicants, however a clear causal relationship between the environment and sALS has yet to be demonstrated. It is likely that susceptibility genes predispose individuals to develop sALS, and advances in genetics are now identifying sALS risk genes. The majority of single nucleotide polymorphisms (SNPs) are located in non-coding DNA, and the effort to understand the genetic basis of disease is shifting to focus on non-coding regions. Enhancer regulatory elements (eREs) are a type of regulatory element located in non-coding DNA that function to increase gene expression. However, the role of eREs in ALS remains unknown. Here I investigated if toxicants and/or genetic components can be considered as sALS risk factors. The overall hypothesis of my work is that variants in gene enhancer elements are sALS risk factors, which interact with environmental toxicants to drive motor neuron degeneration. I described the effect of lead (Pb) on motor neuron degeneration and further studied the pathogenesis underlying toxicant-induced degeneration. To screen low frequency variants for eREs, I used sALS summary statistics from a large genome wide association study and identified SNPs in 312 distinct eREs. I further prioritized 13 top candidate eRE target genes using RNA-seq data from laser captured motor neurons from sALS patients. Through functional analysis, I demonstrated that knockdown of nucleoporin 50 (nup50), a component of nuclear pore complex, in vivo in a zebrafish model results in motor neuron degeneration, proposing 4 eREs targeting NUP50 as novel sALS risk factors. Further, I investigated the effect of gene-toxicant interactions and demonstrated the synergistic effects on toxicity despite a lacking motor phenotype. These findings describe pathogenic features underlying toxicant-induced motor neuron degeneration, demonstrate a role of eREs as novel sALS risk factors and highlight the complicated nature of modeling gene-toxicant interactions.

View record

Aniridia and Usher syndrome 1D are rare congenital defects that lead to vision loss in childhood. Here we tested several different approaches to treat animal models of these diseases. Aniridia is a pan-ocular condition caused by deletion or mutation of the PAX6 gene itself or by downstream intragenic abnormalities. We tested two approaches to target the aniridic-glaucoma phenotype in the Pax6Sey/+ mouse model of aniridia. First, since Tgfβ2 is a direct downstream target of Pax6, we tested whether injection of Tgfβ2-secreting mesenchymal stem cells into the Pax6Sey/+ mouse eye could improve development of anterior segment tissue abnormalities. We observed complete formation of Schlemm’s canal (SC) and a partially repopulated trabecular meshwork (TM). Secondly, we tested whether nonsense suppression strategy could rescue the TM defect in the mouse model harboring a nonsense mutation. Either an aqueous suspension of Ataluren® was injected subcutaneously, or topical eyes drops were instilled twice daily from P5 - P45. We found improved structural anatomy, and increased levels of Tgfβ2, Pitx2 and Foxc1 proteins. Furthermore, nonsense suppression via the topical route rescued the developmental defects of TM and SC better than by systemic treatment. Usher syndrome 1D (USH1D) is an autosomal-recessive condition characterized by deafness, vestibular dysfunction and vision loss caused by absence of CDH23 protein. We obtained a mouse model Y2209X line (Cdh23mtblr+/-) which carries a Cdh23 nonsense mutation. Homozygous Cdh23 mice show profound head shaking, circling behavior, deafness, and reduced ERG response. Here, we used prenatal and postnatal nonsense suppression with Ataluren®. We observed a reduction in severity of phenotypic features in Ataluren-treated mice compared to mock-treated mice as well as corrected localization of photoreceptor proteins. We also studied ex-vivo nonsense suppression therapy in Usher patient-specific cells. Induced pluripotent stem cells (iPSCs) were derived from a patient’s blood cells and three-dimensional (3D) retinal eyecups were generated in culture. The retinal eye cups were treated with Ataluren® which restored CDH23 protein levels, as well as other photoreceptors proteins including arrestin, recoverin and S-opsin. These results suggest patient-derived retinal eyecups are a useful tool for preclinical testing of small molecule drugs.

View record

The fovea is a small retinal indentation packed with specialized cone photoreceptors. Despite its key-role in central vision, little is known about foveal pathologies and development at the cellular and molecular levels. Therefore, no treatment is yet available for vision loss resulting from underdeveloped-fovea (foveal hypoplasia (FH)).First, I used aniridia as a disease model to better understand FH at the cellular and molecular levels. Thirty-three aniridia subjects from British Columbia underwent a thorough ophthalmic examination with in-vivo imaging of foveal structure. Molecular investigations include sequencing of PAX6, candidate genes, in addition to 11p chromosomal analysis. In those in whom imaging was possible, FH was seen in the majority (80%) of cases. Best corrected visual acuities in the cohort ranged from normal vision to no light-perception. Molecular genetic defects involving PAX6 were identified in 30 participants, including 4 novel PAX6 mutations and 4 novel chromosome 11p deletions inclusive of PAX6 or its regulatory region. Then as a proof-of-principle, we employed the SMaRT (spliceosome-mediated RNA trans-splicing) method to rescue Pax6 defects in homozygous-mutant mouse embryonic-fibroblasts and then in-vivo in a naturally occurring Pax6 mouse model. We showed that by using SMaRT technology we were able to rescue Pax6 expression in-vitro and in-vivo, paving the way for potential future therapies for FH. Finally, we tested the feasibility of using Anolis carolinensis (green anole lizard) as a novel foveated model. With its complete published genome, bioinformatic analysis revealed that 85% of human candidate FH genes had an orthologous gene or DNA sequence in the anole. Eyes were collected at various stages of prehatching development for histological analysis, immunofluorescence, and apoptosis analysis. We demonstrated that embryonic foveal development in green anoles resembles human foveal development during infancy. Additionally, at embryonic stage (ES) 14 Pax6 was localized across the entire retina. However, at ES17 Pax6 expression in the ganglion cells of the central retina was markedly reduced. These findings provide the first insight into foveal morphogenesis in the green anole and suggest that it could be an ideal model for perturbing the molecular signals driving foveal development, thus informing on human foveal development and disease.

View record

Small gains and losses of chromosomal DNA, called copy number variants (CNVs), are the cause of many human developmental abnormalities detected before or after birth. Clinically-significant CNVs are found in 2-6% of developmentally arrested embryos and fetuses (termed miscarriage) and in ~15% of children with postnatal developmental abnormalities, typically including abnormal brain function and leading to neuro-developmental delay (NDD). The overall goal of my PhD project was to characterize CNVs found in both miscarriages and in children with NDD in order to identify candidate genes that cause these two aspects of abnormal development. I used a multi-faceted approach consisting of bioinformatics, human cell-line analysis and transgenic animal model investigations. I characterized CNVs reported in miscarriages from literature as well as from our laboratory by using bioinformatics approaches to determine the CNVs size, gene content, gene density and function, known gene knockout murine phenotype, and biological pathway enrichment for all miscarriage CNV genes. My analysis identified several genes from miscarriage CNVs with important functions during prenatal development and pregnancy (e.g. CDKN1C and TIMP2) and enrichment of genes from miscarriage CNVs in biological pathways and processes relevant to embryo/fetal development and feto-maternal interaction (e.g. immune response). For discovery of candidate genes responsible for childhood NDD, I characterized CNVs mapping to a chromosome region, 2p15p16.1, which are known to be associated with multiple postnatal developmental abnormalities and NDD (termed 2p15p16.1 microdeletion syndrome). I performed detailed phenotype and CNV analysis of 33 patients with 2p15p16.1 microdeletions and identified 3 candidate genes (XPO1, REL, and BCL11A) for the developmental problems. By studying their expression in patient cell-lines as well as phenotypic consequences of the loss or gain of their expression in zebrafish, I confirmed their role in developmental abnormalities associated with this syndrome. I have also explored the role of non-coding sequences from this CNV in regulation of one of the candidate genes, BCL11A. The results of my study provide a blueprint for identification of genes with a role in abnormal development by characterizing CNVs. Understanding the cause of the developmental abnormalities opens paths for exploring possibilities for their improved diagnosis, prevention, and potential cure.

View record