Date of Award

Spring 5-20-2019

Document Type


Degree Name

Doctor of Philosophy (PhD)


Biological Science

First Advisor

Valerie Odero-Marah, Ph.D.

Second Advisor

Shafiq Khan, Ph.D.

Third Advisor

Cimona Hinton, Ph.D.


High mobility group A2 (HMGA2) is a non-histone protein highly expressed during the development but is low or absent in most adult tissues. Epithelial-mesenchymal transition (EMT) plays a critical role in prostate cancer progression and metastasis. HMGA2 has been shown to promote EMT in separate studies. Interestingly, wild-type HMGA2 and truncated (lacking the 3’UTR) HMGA2 isoforms are overexpressed in many cancers. However, there are no studies on the role of each isoform in prostate cancer progression. We hypothesized that wild-type and truncated HMGA2 promotes prostate cancer progression by different mechanisms. We analyzed the expression of HMGA2 in the prostate panel by western blot analysis and the localization in prostate tissue microarray by immunohistochemistry. We stably overexpressed wild-type and truncated HMGA2 cDNA in LNCaP cells and measured the expression and the localization of HMGA2 as well as EMT markers. We also performed the migration and cell viability assays.

We analyzed phospho-ERK in cells overexpressing HMGA2 as well as inhibition with U0126 (MAPK inhibitor). To explore the role of truncated HMGA2, we measured the reactive oxygen species (ROS) concentration by DCFDA dye, as well as analyzing Jun-D as a putative downstream effector of HMGA2. Additionally, we knocked down Jun-D and performed the migration and cell viability assays. We treated ARCaP-M mesenchymal cells with camalexin, a 3-thizol-2-yl-indole (a natural product, as a candidate to target HMGA2) in vitro and in vivo in nude mice. Our results showed an increase in nuclear HMGA2 expression with prostate cancer progression as compared to normal tissue. LNCaP cells overexpressing wild-type but not truncated HMGA2 displayed nuclear localization and induced EMT via the ERK1/2 pathway, and this effect could be reversed by treating the cells with U0126. Conversely, truncated HMGA2 displayed cytoplasmic expression and increased prostate cancer migration via increasing Jun-D expression and ROS; this could be antagonized by Jun-D knockdown. Finally, treating ARCaP-M aggressive prostate cancer cells with camalexin reduce its expression in vitro and in vivo.

In conclusion, both wild-type and truncated HMGA2 induce prostate cancer progression by different mechanisms which may be targeted by camalexin.