Conflicting evidence questions whether melanoma follows a CD271-based stem cell model (Boiko (B) et al, Nature 466:133; Quintana (Q) et al, Cancer Cell 18:510; Civenni (C) et al, Cancer Res 71:3098). Comparisons of these studies are confounded by the performance of assays in different laboratories, by different investigators, using different methods and different melanomas. We have addressed this through side-by-side testing of the CD271-based stem cell model of melanoma progression, using the published methods above (B, Q, and C).
Cells were isolated from the same human melanomas using each method. Yields of viable melanoma cells were increased (p<0.05) by 2-3 fold if cells were isolated from patient melanomas using B and C methods. Isolated cells were separated by flow cytometry according to CD271 expression. No consistent differences relating to cell isolation method were observed in the proportions of CD271- and CD271+ cells. Accordingly, CD271 transcripts were expressed at similarly low levels in CD271- cells isolated according to each method. These findings do not support the possibility (Cancer Res 71:3098) that trypsin-including tissue digestion (i.e. method Q) impairs the ability to separate CD271- and CD271+ melanoma cells using flow cytometry.
CD271- and CD271+ cells derived through different methods were then transplanted to test the effects of different tumorigenesis conditions: male vs female recipients, NOD/SCID vs NOD/SCID IL2rgnull mice, premixture in different Matrigel formulations (e.g. standard vs high protein). Consistent tumor formation was observed from both CD271- and CD271+ cells, irrespective of xenograft conditions.
We thus find no evidence that melanoma follows a cancer stem cell model characterized by CD271 expression and a lack of effect on this conclusion of controlled variation in published assay conditions.