Preferential survival and growth of MACS-sorted CD133(+) cells in mixed population xenografts, and sustained CD133 expression after reprogramming. GFP- CD133(+) and DsRed-CD133(−) cells were isolated by MACS (a), injected into the flanks of nude mice, and tumors were visualized and measured by Maestro in vivo imaging after indicated periods of time. p-values of <0.05 were considered statistically significant and represented with a single asterisk, while p < 0.01 are labeled with two asterisks. (b). (c) CD133 expression of reprogrammed BAK-R MIC and BAK-P populations assessed by immunofluorescent staining with anti-CD133-PE for 3 weeks show that BAK-R, but not BAK-P melanoma cells, exhibit sustained CD133 expression. Representative images after immunofluorescent staining of BAK-P and BAK-R cells with anti-CD133-PE (d, right panels) as well as qRT-PCR (e) and immunoblot (f) analyses showing elevated CD133 expression in BAK-R compared to BAK-P cells. CD133(+) Caco-2 colon cancer cells and CD133(−) 1205Lu melanoma cells served as positive and negative controls, respectively (d; left panels).

Preferential survival and growth of MACS-sorted CD133(+) cells in mixed population xenografts, and sustained CD133 expression after reprogramming. GFP- CD133(+) and DsRed-CD133(−) cells were isolated by MACS (a), injected into the flanks of nude mice, and tumors were visualized and measured by Maestro in vivo imaging after indicated periods of time. p-values of <0.05 were considered statistically significant and represented with a single asterisk, while p < 0.01 are labeled with two asterisks. (b). (c) CD133 expression of reprogrammed BAK-R MIC and BAK-P populations assessed by immunofluorescent staining with anti-CD133-PE for 3 weeks show that BAK-R, but not BAK-P melanoma cells, exhibit sustained CD133 expression. Representative images after immunofluorescent staining of BAK-P and BAK-R cells with anti-CD133-PE (d, right panels) as well as qRT-PCR (e) and immunoblot (f) analyses showing elevated CD133 expression in BAK-R compared to BAK-P cells. CD133(+) Caco-2 colon cancer cells and CD133(−) 1205Lu melanoma cells served as positive and negative controls, respectively (d; left panels).

Compared to BAK-P, BAK-R cells strongly express markers of cancer stem cells (Oct4, Nanog), EMT (vimentin), and invasion (MMP2 and MMP9) as shown by immunoblot analysis (a), melanosphere formation (b–d) and resistance to kinase inhibitors (e). BAK-P and BAK-R cells were cultured separately in melanosphere conditions, imaged (b) and melanosphere sizes as well as average number of non-aggregated cells were quantified by Image J analysis (c). CD133(+) DsRed and CD133(−) GFP BAK-R cells were recombined in a 1:1 ratio and grown in low attachment plates (d). CD133(+) cells derived by MACS sorting (left column) and BAK-R (right column) both exhibit increased resistance to dabrafenib (top) or trametinib (bottom), compared to CD133(−) cells (left panel) or parental cells (BAK-P; right panel). p-values < 0.0001 are shown as four asterisks (c,e).

BAK-R cells exhibit increased invasion and metastasis in transwell invasion and zebrafish assays, compared to BAK-P (b,c left panels), while CD133 siRNA knockdown reduces CD133 and MMP2 expression (a) and inhibits invasion (b,c right panels) *** represent p < 0.001. (c) Transwell cell invasion assays were performed for 48 h. Cells that have invaded to the bottom of the transwell were stained with toluidine blue for imaging. Percentage of cell invasion was quantified and presented as a bar graph. Zebrafish were injected with the BAK-P (d) and BAK-R cells alone (e) or in combination (f), and imaged at day 1 (left panels) and day 5 (right panels). Quantification revealed that whereas parental cells showed no metastasis (g), reprogrammed CD133(+) BAK-R cells grafted alone (h), or in combination with parental BAK-P cells (i), exhibited significant metastasis to the tail (j).

(a) CD133-depleted partial CRISPR-Cas9 BAK-R T3 cells show decreased CD133 expression and are less invasive and metastatic than BAK-R-SC controls. CD133 positivity was determined by immunofluorescence staining in BAK-P, BAK-R, partial CRISPR-Cas9 knockdowns using three target sgRNAs and a scrambled sgRNA control. CD133⁺ Caco-2 colon cancer cells and 1205Lu cells served as positive and negative controls, respectively. Images were taken at 10x magnification. RT-PCR (b) and immunoblot analysis (c) show depletion of CD133 RNA and protein in BAK-R T3 compared to BAK-R SC or BAK-R cells. (d) Transwell cell invasion assays showed decreased cell invasion in CD133-depleted BAK-R T3 cells, compared to BAK-R, control SC sgRNA, or T1 and T2 cells. (e) Injection of zebrafish with a 1:1 mixture of BAK-R SC (red) and BAK-R CD133 knockdown T3 cells (green) and representative images after 5 days; (f) quantification of % metastasis to the zebrafish tail. *, **, *** represent p < 0.05 and p < 0.01, and p < 0.001, respectively.

(a) Immunoblot analysis showing protein levels of CD133, as well as secreted MMP2 and MMP9, in different CRISPR-Cas9 melanoma cell lines and BAKP; β-Actin verified equal protein loading. Immunoblot analysis shows decreased levels of CD133 and secreted MMPs in CRISPR Cas-9 CD133 knockdown melanoma pooled clone T1 and T3, compared to BAKP and scrambled control cells. (b) Representative images of crystal violet stained invading cells after 48 hours in transwell invasion chambers; HT1080 and MCF-7 served as positive and negative controls, respectively. (c) Percent invasion of CRISPR Cas-9 CD133 knockdown cell lines (T1, T2, T3) compared to BAKP or scrambled (SC) control cells; *, **, *** represent p < 0.05, p < 0.01, or p < 0.001 respectively compared to T3 clone.

Dox-inducible CD133 expression in BAK-P cells as verified by qPCR (a) and immunoblot analysis (b). (c) Dox-inducible CD133 expression significantly increases cell invasion in transwell assays. (d) Immunoblot analysis with antibodies to CD133, MMP9, MMP2, or β-Actin for loading controls, show increased levels of CD133 as well as MP2 and MMP9 secreted by Dox-induced cells. (e) Zebrafish assays reveal enhanced metastasis in Dox-induced cells. (f,g left panels) Invasion assay results with other melanoma cell lines POT (f) and SK-MEL2 (g) CD133 CRISPR-cas9 SC versus T3 lines are consistent with results with BAK-P cells, showing that knockdown of CD133 (f,g right panels) results in decreased cell invasion. *, **, *** represent p < 0.05 and p < 0.01, and p < 0.001, respectively.

Dox-inducible CD133 expression in BAK-P cells as verified by qPCR (a) and immunoblot analysis (b). (c) Dox-inducible CD133 expression significantly increases cell invasion in transwell assays. (d) Immunoblot analysis with antibodies to CD133, MMP9, MMP2, or β-Actin for loading controls, show increased levels of CD133 as well as MP2 and MMP9 secreted by Dox-induced cells. (e) Zebrafish assays reveal enhanced metastasis in Dox-induced cells. (f,g left panels) Invasion assay results with other melanoma cell lines POT (f) and SK-MEL2 (g) CD133 CRISPR-cas9 SC versus T3 lines are consistent with results with BAK-P cells, showing that knockdown of CD133 (f,g right panels) results in decreased cell invasion. *, **, *** represent p < 0.05 and p < 0.01, and p < 0.001, respectively.