Divergent expression of CD133 in different studies: The need for a consensus panel?письмо
Аннотация: We have red with great interest the article by Gemei et al. dealing with the different expression of CD133 reported in different studies for the same cell lines.1 More specifically, they focus on the HCT116 colon cancer cell line and note that different percentage of CD133+ cells have been reported, ranging from very low to very high, in different studies. In their analysis, Gemei et al. emphasize the importance of cross contamination between different cell lines as the major determinant of divergent findings existing in the literature. In the case of HCT116 cells, they report that the original cell line expresses more than 90% of positive cells while this value is significantly reduced in several studies. While we agree that using certified continuous cell lines is essential to obtain reproducible and reliable results, we believe that the question is extremely more complex when dealing with the CD133 molecule. In our laboratory, we have tested different colon cancer cell lines and have noticed a marked variability in the expression levels of the CD133 protein whose reasons remain unknown but, in our opinion, cannot be exclusively referred to cross contamination problems. As for the HCT116, different percentage of CD133+ cells have been also reported in the literature for the HT29 and CaCo2 colon cancer cell lines, ranging from values >95% for both cell lines2 to about 2% for HT293 and 5% for the CaCo2 cells.4 For both cell lines we observed that these values are highly variable and that cell culture conditions and serum availability can affect CD133 expression. Indeed, when expression of CD133 was evaluated in CaCo2 cells seeded at different densities, we observed a progressive increase in the percentage of CD133+ cells from sparse to dense cultures, which was confirmed both by flow cytometry (from 74 to 90%), using the well-characterized CD133 monoclonal antibody AC133-1, and by western blot using two different antibodies (Fig. 1). Similar results were also obtained with the HT29 cells (data not shown). Moreover, we found that serum concentration also influences the percentage of CD133+ cells, as assessed by flow cytometry (data not shown). These findings cannot be explained by cross contamination since they were performed in parallel using the same cells. We and others also previously reported that the CD133 molecule undergoes a complex post-translational regulation and that the AC133 epitope, but not the protein, is lost during differentiation of colon cancer cells.5, 6 This observation also implies that different results can be obtained using different antibodies when evaluating CD133 expression and this might contribute to explain the discrepancies reported in the literature. Finally, we used FACS to sort CaCo2 and HT29 cells into pure CD133+ and CD133− cultures and observed a reversion of both populations to the original mixed pattern after serial passages in vitro (data not shown). Data have been also reported suggesting that mycoplasma contamination, rather then cross contamination with other cell lines, can affect CD133 expression levels in cell cultures in vitro.7 Moreover, cell cycle,8 hypoxia9 and epigenetic events10 have been also shown to affect CD133 expression, at least in specific cell lines. Overall, these findings demonstrate that several factors contribute to the modulation of CD133 expression and warrant further studies to identify the underlying molecular mechanisms in order to verify how much relevant such mechanisms might also be in vivo. Increased CD133 expression in colon cancer cells with increasing cell density. Total cell extracts were prepared from CaCo2 cells plated in 10 cm dishes at a density of 0.3 (sparse), 1 (subconfluent), and 3 (dense) × 106 cells/dish (corresponding to about 0.5, 1.5 and 5 × 104 cells/cm2, respectively) and cultured for 48 hr. Proteins (50 μg) were resolved by SDS-PAGE, transferred to an Immobilon membrane and parallel blots were probed with the AC133 antibody recognizing an extracellular epitope (top panel) or an antibody directed against an intracellular C-terminal region (middle panel) of human CD133. In the bottom panel, the expression of β-actin in the same extracts is shown as a control for loading. These observations are also relevant for the proposed significance of CD133 as cancer stem cell (CSC) marker and for its use to isolate colon CSC. In fact, if cell culture conditions may cause changes in CD133 regulation and/or expression while cell culture phenotype (i.e., cell growth properties, tumorigenicity, etc.) remains unaffected this would suggest caution in considering CD133 a CSC marker and other factors, besides cell stemness, might influence its expression. The observation that the percentage of CD133+ cells in established colon cancer cell lines is usually higher than in the original tumor in vivo and in primary cultures newly established from tumors casts further doubts on the proposed significance of CD133 as a CSC marker, at least when referring to established cell lines. In conclusion, we believe that cell culture conditions and the type of antibodies and/or techniques (i.e., flow cytometry vs. western blotting) used to detect CD133 might significantly contribute to the discrepant results reported in the literature and should be kept in consideration when evaluating studies dealing with this molecule. This might also suggest the need for a consensus panel aimed to define standardized procedures and reagents to evaluate expression of this molecule especially in view of its potential clinical applications. Yours sincerely, Alessandro Sgambato, Federica Errico, Emanuele Caredda, Maria A. Puglisi, Achille Cittadini.
Год издания: 2010
Авторы: Alessandro Sgambato, Federica Errico, Emanuele Caredda, Maria Ausiliatrice Puglisi, Achille Cittadini
Издательство: Wiley
Источник: International Journal of Cancer
Ключевые слова: Cancer Cells and Metastasis, Mass Spectrometry Techniques and Applications, Cancer Genomics and Diagnostics
Другие ссылки: International Journal of Cancer (HTML)
PubMed (HTML)
PubMed (HTML)
Открытый доступ: closed
Том: 128
Выпуск: 9
Страницы: 2247–2249