Date of Award

January 2019

Document Type


Degree Name

Medical Doctor (MD)



First Advisor

Kristopher T. Kahle


The Vein of Galen Malformation (VOGM) is a specific subtype of arteriovenous malformation (AVM) that becomes evident in weeks 6-11 of embryonic development. VOGM comprise less than 1% of all vascular malformations, yet represent 30% of all pediatric intracranial vascular malformations. Depending on their specific characteristics, i.e. its feeding vessels, it can present clinically as devastating congestive heart failure in neonates, hydrocephalus in children, or seizures and headaches in young adults. Advances in treatment provide improved survival, primarily through endovascular surgery. The genetic and molecular etiology of VOGM remains relatively unknown, with the only associated genes being in the context of other syndromes, including seven mutations in p120-RasGAP (RASA1) in Cutaneous Malformations-Arteriovenous Malformations (CM-AVM), as well as one mutation each in activin A receptor type II-like 1 (ACVRL1) mutation and Endoglin (ENG) in Hereditary Hemorraghic Teleangiectasia.

Our limited knowledge of the molecular genetics of VOGM has hindered the development of novel therapies. We hypothesized that the apparent sporadic occurrence of VOGM may frequently be attributable to damaging de novo mutation events or incomplete penetrance of rare transmitted variants. Unbiased whole-exome sequencing (WES) can overcome these barriers for gene discovery. We recruited 55 patients, including 52 parent-offspring trios. WES revealed statistically significant rare, damaging de novo mutations in chromatin modifier genes involved in brain and vascular development (p=8.9 x 10-4). VOGM probands also had inherited, missense deleterious and loss of function, in ephrin signaling genes, specifically a whole exome significant mutation burden in EPHB4 (p=7.47 x 10−10). Finally, we observed a whole exome significant inherited mutation in Claudin 14 (p= 6.44 x 10−7). Inherited mutations demonstrated incomplete penetrance and variable expressivity with mutation carriers often exhibiting cutaneous vascular abnormalities suggesting a two-hit mechanism. The identified mutations account for 30% of studied VOGM cases.

To functionally validate candidate mutations and establish causality, we developed a screening platform using Xenopus tropicalis by using CRISP/Cas9 gene editing and observing the effects of candidate gene knockdown on vasculogenesis and brain vasculature.

CRISPR/Cas9 knockdown of EPHB4 and CLDN14 yielded significantly abnormal vasculogenesis (p=0.0028 and p=0.0001 respectively) observed by in situ hybridization. We implemented and modified established clearing techniques to allow visualization of the full thickness of the Xenopus brain vasculature. The evolutionary precursor to the vein of Galen, the posterior vascular plexus (PVP), and the mesencephalic veins (MSV), were imaged. EPHB4 and CLDN14 CRISPR/Cas9 knockdown revealed significant decreases in MSV length (p<0.0001 and p=0.0006 respectively) but there was no significant decrease in PVP area. Attempted rescue with wild-type vs. mutant human CLDN14 mRNA did not lead to significant improvement of the phenotype.

Together these findings are the first step in better understanding the mechanism and pathogenesis of VOGM and potential novel therapeutic targets. While Xenopus is not a perfect model system, it does show promise as a tool to assess candidate VOGM that merit further study in mammalian systems harboring a true vein of Galen.


This thesis is restricted to Yale network users only. It will be made publicly available on 06/01/2022