Biology Ph.D. Dissertations


Targeting Cancer Stem-LIike Cells in Human Esophageal Squamous Carcinoma Cell Lines by Curcumin

Date of Award


Document Type


Degree Name

Doctor of Philosophy (Ph.D.)


Biological Sciences

First Advisor

Roudabeh Jamasbi

Second Advisor

Deborah Wooldridge

Third Advisor

Michael Geusz (Committee Member)

Fourth Advisor

Carol Heckman (Committee Member)

Fifth Advisor

Vipa Phuntumart (Committee Member)


Numerous cancers contain cell subpopulations that exhibit stem cell characteristics. These putative cancer stem cells (CSCs) seem to resist chemotherapy and radiation therapy and initiate tumor recurrence. Therefore, it is important to isolate and further characterize CSC subpopulations, as well as develop therapeutic agents that target them. Among alternative therapeutic agents that target CSCs, recent laboratory experiments and clinical trials have tested the effectiveness of curcumin, a phytochemical agent and functional component of the spice turmeric, against various cancers. Still, the effect of curcumin on CSCs is not well established and new methods for identifying CSCs are needed to help clarify these effects. In addition to the use of aldehyde dehydrogenase-1A1 (ALDH1A1), CD44, and nuclear factor kappa-light-chainenhancer of activated B cells (NF-κB) as markers for CSCs, Aldefluor has also been used with flow cytometry to isolate CSCs. However, new techniques are needed to locate and identify CSCs in culture for live-cell analyses. I conducted the two studies presented here to determine the effect of curcumin on CSCs in human esophageal squamous cancer cells (ESCCs) using multiple CSC markers, including ALDH1A1, CD44, and Aldefluor, as well as NF-κB. In the first study, I evaluated curcumin-induced cell death in six human ESCC lines (KY-5, KY-10, TE-1, TE-8, YES-1, YES-2). In order to determine the effect of curcumin treatments on CSCs in ESCC cell lines, the six ESCC lines were exposed to 20-80 μM curcumin for 30 hrs, from which cell lines surviving 40 or 60 μM curcumin were established. ALDH1A1 and CD44, as well as NF-κB, were used as CSC markers to compare CSC-like subpopulations within and among the original lines as well as the curcumin-surviving lines. Additionally, YES-2 was tested for tumorsphere-forming capabilities. Finally, the surviving lines were treated with 40 and 60 μM curcumin to determine whether their sensitivity was different from the original lines. After the curcumin treatment, cell loss increased in a dose dependent manner in all cell lines. The curcumin-surviving lines showed a significant loss in the high staining ALDH1A1 and CD44 cell populations and tumorspheres were smaller and fewer in the YES-2 surviving line than the original. These results suggest that curcumin not only eliminates cancer cells but also targets CSCs, and therefore may be an effective treatment for esophageal and possibly other cancers in which CSCs can cause tumor recurrence. In the second study, I developed and validated a new method for identifying CSCs called the attached-cell Aldefluor method (ACAM), the first of its kind to apply Aldefluor staining to attached rather than suspended cells. In order to validate the ACAM technique, I created a cell line enriched in CSCs by isolating CSCs from the original YES-2 parental line using standard Aldefluor flow cytometry. This line showed significantly greater ACAM staining and higher CD44 levels than YES-2. ACAM also showed significantly higher ALDH activity in YES-2 CSC than the cell line that has a diminished CSC subpopulation after having survived the curcumin treatment (YES-2S). ACAM consistently stained the cells within tumorspheres made from the CSC-enriched line but not the differentiating cells derived from the tumorspheres. This study demonstrated the validity of this new method for generating and growing tumorspheres without the growth factor supplements routinely used.