Brent Schultz

Date of Award


Document Type

Open Access Thesis

Degree Name

Medical Doctor (MD)

First Advisor

Deepak Narayan


Purpose: To investigate the epi-genetic regulation of Insulin Like Growth Factor 2 (IGF2) and its reciprocally imprinted transcript H19 in infantile hemangiomas. Introduction: Infantile hemangiomas (IH) are the most common childhood tumor of the head and neck. Despite their prevalence and potentially morbid sequelae, little is known regarding the pathogenesis of this disease. However, a tumorigenic molecule, Insulin Like Growth Factor 2 (IGF2), has been implicated by microarray and confirmatory Real-Time PCR studies. There is substantial documentation that methylation abnormalities within the IGF2 and neighboring H19 loci are related to the overproduction of IGF2 in many distinct tumor types. An investigation of the methylation status, of this region, as well as the factors modifying methylation, may explain pathologic IGF2 overproduction in hemangiomas. Methods: Using bisulfite specific methylation sensitive PCR with quantitative pyrosequencing, confirmatory genomic southern analysis, and quantitative RT PCR the methylation status of multiple regions within the IGF2/H19 locus were correlated with two potential transcriptional consequences and/or causes of aberrant regulation. Results: This study identifies IH as the first non-malignant neoplasm expressing ectopic BORIS, an oncogene with expression normally limited to adult testes. The paradoxically benign nature of IH despite BORIS expression could be explained by a 13 fold increase in CTCF, BORIS only known antagonist, from proliferating to involuting IH. Interestingly, both proteins bind within IGF2 and H19. In the IH samples, as CTCF levels rose compared to BORIS, the IGF2 transcript decreased 6 fold. CTCF and BORIS likely regulate IGF2 by altering methylation of the region: The difference between CTCF and BORIS is most predictive of methylation levels at several imprinted sequences (R2 = .9.) Throughout the 130 KB regulatory region controlling IGF2 and H19, BORIS favored methylation specific activation of IGF2 and repression of H19, while CTCF favored the converse. The degree of these effects strongly correlated with a common C/T polymorphisn at the IGF2 imprinting control region. Here the T allele was strikingly more sensitive to CTCF and BORIS than the C allele. Hence, the C/T polymorphism may be an important disease modifier of IH. Conclusion: Identifying the aberrant expression of a known oncogene, BORIS in IH, suggests one factor driving early proliferation. Furthermore, the steady increase in the production of BORIS antagonist, CTCF, may support the involutionary process. The interplay between these two proteins likely takes place at the level of DNA imprinting, as the difference between CTCF and BORIS was highly predictive of methylation levels within key regulatory regions of the IGF2/H19 locus, a region previously demonstrated to control the relative expression of both genes. A potentially important disease modifier may be the C/T polymorphism within CTCF binding site six, which strongly affects methylation of the region relative to the CTCF BORIS difference. As 400,000 children are born each year in the US with IH, a clinical blood test resulting from this finding may be highly useful in predicting eventual tumor size and time to involution.


This is an Open Access Thesis.

Open Access

This Article is Open Access