Date of Award

Winter 12-15-2019

Author's School

Graduate School of Arts and Sciences

Author's Department

Biology & Biomedical Sciences (Molecular Genetics & Genomics)

Degree Name

Doctor of Philosophy (PhD)

Degree Type



The epidermis is the outermost tissue of the skin and provides the body’s first line of defense against external assaults. The epidermis is primarily composed of keratinocytes that terminally differentiate and rise apically toward the surface to form the semipermeable barrier of the skin. A hallmark of keratinocyte terminal differentiation is the expression of genes from the Epidermal Differentiation Complex (EDC) locus. Many of the EDC protein products contribute to the structural integrity of the skin barrier, evidenced by several gene knockouts such as loricrin, and even genetic variation within gene coding sequences, that modulate the integrity of the skin barrier. Many of the genes in the EDC are coordinately expressed but the mechanism(s) to coordinate this expression is poorly understood. Our lab previously identified the conserved non-coding element 923 in the EDC as an epidermal-specific enhancer. I further hypothesized a role for the 923 enhancer to activate EDC gene expression. To test this hypothesis, we generated independent deletions of the 923 enhancer using CRISPR/Cas9 genome editing in mice. The deletion alleles included a specific 1.2kb deletion of 923 (923del) and a large deletion that included the 923 enhancer and an additional deletion of 40kb downstream (923large). Both 923del/del and 923large/large deletion mice appeared phenotypically normal under homeostatic conditions. However, both 923del/del and 923large/large newborn skin transcriptomes revealed decreased Ivl and Rps3a1, and increased Rps3a3, Hist1h2a1, and Gm16011 gene expression changes in contrast to wild-type. Transcriptomic analysis of involucrin deficient (Ivl-/-) mice also revealed the same gene expression changes for these 5 genes and demonstrated that the shared gene expression changes are due to loss of Ivl rather than 923. Ivl is the most proximal gene to 923. Together, I identified Ivl as the primary target gene of 923 enhancer regulation. I further identified an additional requirement for the 923large allele for late cornified envelope 6a (Lce6a) expression suggesting the deletion of an additional regulatory element in the 923large allele. Allele-specific assays for Ivl and Lce6a gene expressions in the skins of 923del and 923large C57Bl/6 and BALB/cBYJ mouse hybrids allowed us to further determine that the 923 enhancer regulates gene expression in cis. ATAC-seq on 923del and 923large homozygous epidermis compared to wild-type mice revealed a shared enrichment of Differentially Accessible Regions that were less accessible within the EDC. This suggests the 923 enhancer affects local chromatin accessibility in keratinocytes. Together, my RNA-seq and ATAC-seq experiments identify a cis regulatory module for the 923 enhancer and Ivl target gene expression in the epidermis, as well as a functional role for the 923 enhancer to maintain local chromatin accessibility.Filaggrin (FLG) is an EDC gene with population-specific genetic variation identified in individuals of European and Asian ancestries. FLG loss-of function (LOF) variants are the most well-known risk factors for the common inflammatory skin disease atopic dermatitis (AD). Despite our understanding of FLG LOF for AD in populations of European descent, we have a poor understanding of the prevalence of FLG LOF variants in African Americans (AA) populations, who are disproportionately affected with AD. I hypothesized the presence of FLG LOF variants specific to African ancestry in a well-characterized AA AD pediatric cohort. To discover FLG LOF, we utilized a tiled array based sequencing approach to sequence the highly repetitive FLG gene, with further validation by Sanger sequencing. Our targeted sequencing identified five FLG LOF variants (c.488delG, p.R501*, p.R826*, p.S3101*, and p.S3316*) in nine AA AD patients. Two variants, c.488delG and p.S3101*, were novel as they were not previously reported in the dbSNP, ExAC, or the ESP variant databases. Additionally p.S3316* has only been previously reported in individuals of African ancestry, suggesting a population-specific African variant. Moreover, a significantly higher frequency of FLG LOF was observed in our AA AD group (11.5%) compared to the frequency for Africans reported in ExAC (2.5%) and ESP (1.7%) indicating an enrichment for FLG LOF in the context of AD in this shared ancestral population. Our results demonstrate a prevalence of population-specific FLG LOF variants in AA AD patients at higher rates than previously reported and justify future precision medicine investigations to best address this AD health disparity.In summary, my dissertation functionally demonstrates the existence of a cis-regulatory module in the epidermis, namely the 923 enhancer:Ivl expression module, and the 923 enhancer’s remodeling effect on local EDC chromatin accessibility. Furthermore, by using a tiled sequencing approach, I discovered FLG LOF variants in AA AD patients, including p.S3316* that was African-specific, highlighting a higher prevalence of FLG LOF in contrast to previous reports. These two seminal studies together expand our existing knowledge of gene regulation and variation in the epidermal skin barrier.


English (en)

Chair and Committee

Cristina de Guzman Strong

Committee Members

Shining Chen, Jacqueline Payton, Celia Santi, John Edwards,


Permanent URL: https://doi.org/10.7936/e0gf-wf54