Date of Award

January 2018

Document Type


Degree Name

Master of Science (MS)



First Advisor

Seema Somji

Second Advisor

Scott H. Garrett


Urinary carcinoma is one of the most prevalent cancers across the United States and is among the leading cause of cancer related morbidities. With muscle-invasive urinary carcinoma having only 15% survival at stage T4, diagnostic tools of this type of cancer is necessary. This work sought to characterize a model of urothelial carcinoma into a specific muscle-invasive bladder cancer subtype, which has important implications in for prognosis, development of targeted therapeutic agents, and disease management. Our laboratory has developed an immortalized uroepithelial cell line, UROtsa, which can be malignantly transformed by arsenite (As+3) and Cd2+ and has been shown to produce colonies on soft agar and tumors in nude mice. We therefore sought to determine if which subtype of muscle-invasive bladder cancer in our six As3+ and seven Cd2+-transformed cell lines and heterotransplant tumors. Real-time reverse transcriptase polymerase chain reaction (RT-PCR) was used to determine mRNA levels of 25 marker genes of basal and luminal subtypes of muscle-invasive bladder cancer. Western blotting and immunohistocemical analysis of select markers was used to evaluate protein expression. Data indicated that the expression of the 25 marker genes in the As3+ and Cd2+-transformed tumors was increased in the basal subtype compared to the luminal subtype. This same expression pattern was also seen in the transformed As3+ and Cd2+ cell lines and spheroids, or putative cancer stem cells, derived from our transformed cell lines via culture in serum-free media in low attachment flasks. Next, we sought to determine the mRNA and protein expression in our non-tumorigenic parental UROtsa cell line. We determined the parental UROtsa cell line had increased expression of the basal marker genes and this was further increased in the spheroids, while expression of the luminal marker genes was seen to be further decreased in expression in the spheroids.

Due to the observations seen in the mRNA expression of the spheroids, we became interested in the ability of these cells to maintain their gene signature when replaced back into the cell culture conditions of the original cell line. To do this, one As3+ cell line was grown in serum media and cell culture flasks that promote attachment. From there, the cells were placed into low attachments flasks with serum-free media to produce spheroids. Once grown, the cells were reintroduced into cell culture flasks that promote cell attachment with serum-containing media and harvested once reaching confluency (P1) and passages 4 and 8 (P4 and P8). Microarray analysis of these samples found that there were 4,415 differentially expressed genes when comparing the original cell line to the spheroid. This number decreased at each comparison of the passages to the cell line with 1,290, 452, and 191 differentially expressed genes compared to the cell line at P1, P4, and P8 respectively. This data indicated that the replacement of our spheroids into cell culture conditions identical to the original cell line promoted the reversal of the spheroids into our original cell line cultures. We next sought to understand if the gene signature of the spheroids would be maintained if they were cultured in conditions of flasks that promote cell attachment with serum-free media in the same As3+ transformed cell line, but also a Cd2+ cell line and our original parental UROtsa cell line. By performing real-time RT-PCR analysis of the 25 muscle-invasive bladder cancer marker genes, we found the number of differentially expressed genes compared to the cell line and either the spheroid or subsequent passages was relatively stable with at least 68% of the genes being differentially expressed by P8. This data demonstrates we have a unique gene signature that does not align with the original cell line.