Figure 1.
The transduction of OSK induced CSC properties in SW480 cells in vitro.
(A) qRT-PCR of previously reported markers related to colon CSCs and intestinal stem cells in the transduced SW480 cells. All the markers were upregulated in the OSK-SW480 cells. The mRNA expression levels were normalized to those of GAPDH. The relative expression levels compared to those of M-SW480 are shown. (B) The protein expression levels of CSC markers in the transduced SW480 cells as determined by a flow cytometric (FCM) analysis. The data from three independent experiments demonstrated that the numbers of CD133-, CD44-high-, CD26- and ABCG2-positive cells were significantly increased in the OSK-SW480 cells. The representative dot plots of each marker are shown in Figure S2. (C) The protein expression of ALDH1 in the transduced SW480 cells determined by a Western blot analysis. The panel shows representative data from three independent experiments. (D) The cell proliferation in vitro. A total of 3×105 cells were plated on six-well plates on day seven and were counted on day 11. The number of OSK-SW480 cells was lower than that of M-SW480 cells (n = 3). (E) The cells in the G1/0 phase were detected by a FCM analysis on day 11. The percentage of cells in the G1/0 phase significantly increased in the O-, K- and OSK-SW480 cells (n = 3). (F) 5-FU-chemoresitance analysis. The viability of OSK-SW480 cells in the presence of 5-FU was significantly higher than that of M-SW480 cells at both the 1 and 50 µg/ml concentrations of 5-FU (n = 3). The viability of the M-SW480 cells at each concentration was set to 100%. The error bars indicate the standard deviation: SD. *P<0.05, **P<0.01, Dunnett's test.
Figure 2.
The sphere formation ability in vitro and tumorigenicity in vivo.
(A) The spheroid formation assay. A total of 1×104 cells were plated on low attachment dishes on day 10 and cultured with serum-free medium for 10 days. In the O-, K- and OSK-SW480 cultures, the number of spheroids was significantly increased compared to that of the M-SW480 cells. The error bars indicate the SD (n = 3). Scale bars: 100 µm. (B) The tumorigenicity of the cells after implantation in the subcutaneous regions of immunodeficient nude mice. A total of 1×106 cells (left panel) or 3×105 cells (right panel) were subcutaneously injected into both flanks of immunodeficient nude mice on day 10. The volume of the tumors derived from the K- and OSK-SW480 cells were obviously higher than those of the Wt-, M-, O- and S-SW480 cells for both numbers of injected cells. The red bars indicate the median tumor volume. **P<0.01, N.S.: not significant, Dunnett's test (except of †: U-test).
Table 1.
Summary of tumor formation derived from transduced SW480 cells.
Figure 3.
Efflux activity for Hoechst33342.
(A) The OSK-SW480 cells included a population of cells unlabeled by 5 µg/ml of Hoechst33342 with the co-administration of 50 µM of verapamil (VM). We designated the cells unlabeled by Hoechst33342 without VM and with 50 µM of VM as V0-cells and V50-cells, respectively. The left panel shows representative dot plots of labeled and unlabeled cells at day 10 after transduction. The right panel shows the results of three independent experiments. The V0 subpopulation was increased in the O- and OSK-SW480 cells. The V50-cells were obviously seen in the OSK-SW480 cultures. The error bars indicate the SD (n = 3). *P<0.05, **P<0.01, Dunnett's test. (B) Dye efflux activity at day 10 (left panel) and day 28 (right panel) after transduction. The V50-cells disappeared under the treatment with the co-administration of 250 µM of VM even in the OSK-SW480 cells. The proportion of the OSK-V50 cells decreased with time.
Figure 4.
Characterization of the V50-cells in OSK-SW480 cells after FACS.
The V50-cells in the OSK-SW480 (OSK-V50) cells were sorted by FACS. Non-V0-, V0- and non-V50-cells in the M-SW480 cells (M-nonV0, M-V0 and M-nonV50, respectively) and non-V50-cells in the OSK-SW480 cells (OSK-nonV50) were also sorted and used as controls. These cells were all subsequently cultured. (A) The morphologies of the cells cultured for 10 days after sorting. The morphology of the OSK-V50 cells was similar to that distinctively observed in OSK-SW480 cells (Fig. S1D, lined circle). In contrast, the morphology of OSK-nonV50 cells was similar to that of M-V0, M-nonV0 and M-nonV50 cells. Scale bars: 100 µm. (B) The cell proliferation in vitro. A total of 3×105 cells cultured for 14 to 18 days after sorting were seeded and counted 96 hr later. The number of cells was significantly lower in the OSK-V50 cells than that in all the other lines. The error bars indicate the SD (n = 3). **P<0.01, Scheff's test. (C) Tumorigenicity of the cells in immunodeficient mice. A total of 3×105 or 1×105 cells were subcutaneously injected into immunodeficient nude mice on day 18 after sorting. The tumor volume and incidence were measured eight weeks after injection. Only the OSK-V50 cells generated obvious tumors for both the injected cell numbers, whereas no tumors were obtained from the M-nonV0, M-V0, M-nonV50 and OSK-nonV50 cells. The incidence of tumor formation by OSK-V50 cells was 4/4 for both injected cell numbers. The red bars indicate the median tumor volume.
Figure 5.
Colonic lineage specificity and potency to produce phenotypical diversity in OSK-V50 cells.
(A) Immunocytochemistry for CDX2 and CK20 protein. OSK-V50 cells were positive for CDX2 and CK20, as well as M-V0 cells. The cells were counterstained with Hoechst33342. Scale bars: 100 µm. (B) The phenotypical diversity in morphology and mobility of derivatives of OSK-V50 cells. Photographs are time-lapse images of M-V0 and OSK-V50 cells. The sorted cells were subsequently cultured for five days and then observed every six hours for 42 hr. Time-lapse imaging revealed that there was a higher diversity in both the morphology and mobility in OSK-V50 cells (lower panel) compared to M-V0 cells (upper panel). Original magnification, ×20. Movies of M-V0 and OSK-V50 cells are shown in Video S1 and Video S2, respectively.
Figure 6.
The histology of the xenografts derived from M-SW480 and OSK-V50 cells.
(A) Hematoxylin and eosin staining (HE). The tumors derived from M-SW480 cells predominantly consisted of homogeneous cells. On the other hand, the tumors derived from OSK-V50 cells showed glandular structures. (B) Immunohistochemical analysis. The tumors derived from M-SW480 cells were negative for CK20 and CK7. The tumors derived from OSK-V50 cells were partially positive for CK20 and negative for CK7. Tumors derived from both M-SW480 and OSK-V50 cells were positive for CDX2, but CDX2-negative cells were obviously seen in the tumors of OSK-V50 cells. Scale bars: 50 µm.
Figure 7.
The self-renewal capacity of the OSK-V50 cells in vivo.
(A) A schematic representation of serial transplantation experiments. (B) The morphological features of a series of cultured cells in the serial transplantation experiments. The day shown under the panel indicates the culture period after sorting for the first cultured cells and after dissociation from tumors for the second, third and fourth cultured cells. In the case of OSK-V50 cells, the morphological features in each culture were the same over three passages in vivo, and the features were similar to those shown in Figures 4A and 5B. Scale bars: 100 µm. (C) The cell proliferation of the second cultured cells in vitro. A total of 2×105 cells cultured for four to six days after dissociation were seeded and counted 72 hr later. In the second cultured cells, the number of OSK-V50 cells was significantly lower than that of M-nonV50 cells (n = 3). **P<0.01, t-test. (D) The proportion of V50 cells in the second, third and fourth cultured cells. The panel shows the proportion of V50 cells in each culture from three independent experiments. The V50 cells were maintained over three passages in vivo, whereas V50 cells were not observed in the second M-nonV50 cultured cells. The representative dot plots obtained by FCM are shown in Figure S6. (E) The pathological findings of xenograft tumors derived from OSK-V50 cells. The panel shows the histological findings of a series of tumors obtained from serial transplantation experiments. The results of the other series of tumors are shown in Figure S7. The OSK-V50 cells serially showed the same pathological findings as those described in Figures 6A and 6B; there was cell diversity and glandular structures observed by HE staining, and CK20- and CDX2-positive, and CK7-negative findings for the immunostaining. exp: experiment. Scale bars: 50 µm.