| Stem-like cells may be integral to the development and maintenance of human cancers. Direct proof in SCLC is still lacking.In this study, we sought to identify and characterize a subgroup of cancer stem-like cells from SCLC cell line H446.Objective:1. To optimize the method to culture the tumor cell spheres in SCLC cell line H446 in vitro;2. To test whether stem cell-like populations exist in SCLC H446.Methods:1. With the defined serum-free culture, the formation of the H446 tumor spheres was observed under different culture conditions. The cells were inoculated to ultra low adherent plates and non-ultra low adherent plates. The cell density was adjusted to 1×104/mL, 1×105/mL, 1×106/mL and 1×107/mL. The spheres were blowed into single cells through micropipette blow method and pinhead blow method in order to further passages.2. After the tumor sphere formation, cell tracing analysis were used to identify the clonal origin of each sphere. Furthermore, to determine whether a population of self-renewing tumor stem cells exists in SCLC cells line, tumor sphere generation analysis in vitro, in vivo tumorigenesis analysis, limiting dilution analysis, differentiation experiments and cell toxicity experiments to prove that the enrichment of tumor stem-like cells in tumor spheres.3. By flow cytometry, immunofluorescence staining and confocal imaging,we compared the expression of uPAR, CD133, CD90, CD56, Cytokeratin, CD44, AnnexinsⅡ, EGFR, CD24, Nestin, CEA, CD34 and MDR1 in the spheric cells and adherent cells.Results:1. The relevant factors affecting the forming of tumor spheresThe effect of different types of culture plates, cell densities and blow methods on the formation and proliferation of tumor spheres:(1)In ultra low adherent plates, the tumor spheres appeared clearly at the second day, while in the tenth day in non-ultra low adherent plates. (2)At the concentration of 1×106/mL, there were about seven spheres per high power field and dozens of cells even hundreds of cells for each sphere. (3)In the pinhead blow method, the tumor spheres showed better activity and proliferative ability in second generation.2. The generation and characteristics of tumor spheres(1) The morphological characteristics of tumor spheres:The parental H446 cells were inoculated in serum-free medium at 105 cells/ml. We observed that after the 4-10 days cell culture, suspended tumor spheres formed. These spheres weres spherical. The spheres obtained ranged in size from 50～300 um in diameter, with the majority of spheres having a diameter of 120 um. For a sphere of 300 mm diameter, the number of cells varied between 400 and 450 cells, while a sphere of 50 mm contained 30 cells. The number of cells was about 250 cells in the majority of spheres. Cells inside the spheres with a few protrusions attached with each other tightly. And the spheres were blowed to single cells difficultly. The cells in spheres were observed with uniform shape and good refraction.(2) The clonality of the sphere:In order to distinguish between cell aggregation and clonal origin of tumor spheres under our experimental conditions, we carried out cell tracing experiments. Tumor spheres were collected, and made into single cell suspension. Single cell suspensions (A tube and B tube) were stained with two different dyes in separate reactions, for 15-45 minutes at 37℃, then mixed in equivalent amounts and cultured in serum-free medium. On the third day we observed that the vast majority of the spheres have only one type of fluorescence, indicating that they have a single cell origin under the microscope.(3) The serial passages and forming rate of tumor spheres:The tumor spheres had capacity of self-renewal, and the sphere cultures could be maintained for>40 passages. In order to study the foming rate we carried out tumor sphere generation analysis that the cells were inoculated at 103 cells/ml. These date showed that 19.50%±0.22% parental H446 cells formed into primary spheres (n=5). 30.00%±1.00% dissociated first-generation spheres generated the second-generation (n=5)and 49.02%±1.50% dissociated second-generation spheres generated third-generation spheres (n=5).There was significance between any two of the three generations (p<0.05).3. Limiting dilution analysisThe third-generation spheres were collected, and dissociated into single cell suspension for limited dilution experiments. After 2 to 3 weeks, the new spheres formed from single cells, and the formation rate was 77.00%±1.00%; and the colony forming rate inoculated with the parental H446 cells was 17.67%±1.53%(n=3,p< 0.001). In addition, we also found that the new spheres from single cells could be passage for more than 40 times in vitro.4. The results of immunofluorescence staining and flow cytometryThe percentage of uPAR, CD 133 and CD90 was increased in spheres. Approximately 92% of cells of the tumor spheres expressed uPAR, uP AR+ population was remarkably enriched in spheres compared to that of parental cells that was around 20%. The percentage of CD133+ and CD90+ population were less than 30% in spheres. On the contrary, for the expression of differentiative markers CD56 and Cytokeratin, the percentage of both population was decreased in spheres with a low level of expression.CD44, AnnexinsⅡand EGFR were positive in vast majority of both sphere-derived and parental cells, whereas CD24 were only positive in a small number of above cells. Moreover, CEA, CD34, Nestin and MDR1 expression was undetectable in both tumor spheres and parental cells.5. The results of in-vitro differentiationWe placed the third-generation spheres in the differentiation medium containing 10% serum, and observed that the spheres quickly adhered to the plates and differentiated. The differentiated cells showed significant heterogeneity. The immunohistochemical staining results of differentiated cells showed that:the differentiated cells lossed the expression of uPAR and CD133, acquired the high expression of Cytokeratin and CD56.6. Drug resistance of sphere-forming cells and SP analysisCisplatin (CDDP) results:MTT results showed that the survival rate of spheric cells was 91.60%±2.30%, while 38.80%±1.92% in parental H446 cells (n=5, p< 0.001). Etoposide (VP16) results:MTT results showed that the survival rate of spheric cells was 91.00%±4.12%, while 51.20%±1.92% in parenta H446 cells (n= 5, p<0.001). SP analysis results:the proportion of SP cells in spheric and parental H446 cells were 1.35%±0.03% and 0.14%±0.02% respectively (n=5,p<0.01).7. In vivo tumorigenesis analysisTumors could be formed with only 1×103 sphere-derived cells but needed at least 1 x 106 parental cells. By immunohistochemistry, the tumors were characterized by the expression of differentiate markers CD56, Cytokeratin and Chromogranin A.Conclusion:1. The data suggested spheric cells have clonality, the ability of self-renewal, in vitro differention, unlimited proliferation and tumorigenesis, and the enrichment of stem cell-like cells in tumor spheres, indicating that the presence of cancer stem cells in SCLC H446.2. Tumor spheres are composed by majority of undifferentiated cells.The majority expressed uPAR. Compared to H446, the tumor spheres were enriched with uPAR+ sub-population.3. Cytotoxicity experiments showed that the spheric cells have a stronger drug resistance than the parental H446 cells. SP analysis showed that the tumor spheres have more SP cells than the parental H446 cells. These results proved that the spheres were enriched in drug-resistant cells indirectly and directly.4. The tumor spheres can be separated effectively by ultra low adherent plate, the inoculum density of 1×106/mL and the pinhead blow method.These data indicate that a population of tumor sphere-generating cells with stem cell properties is present in human SCLC cell line. Tumor spheres have an enrichment of uPAR expression.
|
PDF Full Text Request