Oct-4 expression maintained cancer stem-like properties in lung cancer-derived CD133-positive cells

Abstract

CD133 (prominin-1), a 5-transmembrane glycoprotein, has recently been considered to be an important marker that represents the subset population of cancer stem-like cells. Herein we report the isolation of CD133-positive cells (LC-CD133(+)) and CD133-negative cells (LC-CD133(-)) from tissue samples of ten patients with non-small cell lung cancer (LC) and five LC cell lines. LC-CD133(+) displayed higher Oct-4 expressions with the ability to self-renew and may represent a reservoir with proliferative potential for generating lung cancer cells. Furthermore, LC-CD133(+), unlike LC-CD133(-), highly co-expressed the multiple drug-resistant marker ABCG2 and showed significant resistance to chemotherapy agents (i.e., cisplatin, etoposide, doxorubicin, and paclitaxel) and radiotherapy. The treatment of Oct-4 siRNA with lentiviral vector can specifically block the capability of LC-CD133(+) to form spheres and can further facilitate LC-CD133(+) to differentiate into LC-CD133(-). In addition, knock-down of Oct-4 expression in LC-CD133(+) can significantly inhibit the abilities of tumor invasion and colony formation, and increase apoptotic activities of caspase 3 and poly (ADP-ribose) polymerase (PARP). Finally, in vitro and in vivo studies further confirm that the treatment effect of chemoradiotherapy for LC-CD133(+) can be improved by the treatment of Oct-4 siRNA. In conclusion, we demonstrated that Oct-4 expression plays a crucial role in maintaining the self-renewing, cancer stem-like, and chemoradioresistant properties of LC-CD133(+). Future research is warranted regarding the up-regulated expression of Oct-4 in LC-CD133(+) and malignant lung cancer.

Figure 1. Isolation and characterization of lung cancer-derived CD133⁺ (LC-CD133⁺).

(A) Using a magnetic bead method, we sorted CD133⁺ cells from tissue samples of patients with lung cancer (LC), and characterized them by FACS assay. (B) LC-CD133⁺ sorted from two patient with No.1 (PLC-CD133⁺) and No.2 (LLC-CD133⁺) were cultured in bFGF and EGF with DMEM serum-free medium. (C) Evaluation of the formation abilities of spheroid-like bodies (SB) from LC-CD133⁺ and LC-CD133⁻ under serum-free medium with bFGF & EGF. (D) The growth curves of LC-CD133⁺ and LC-CD133⁻ were measured by hemocytometer. Bar: 100 µm. Data shown here are the mean±SD of three experiments.

Figure 2. Detection of surface markers and the tumorigenicity in LC-CD133⁺ and LC-CD133⁻ in vitro.

(A) and (B) The expression levels of CD133, CD117 (c-Kit), and ABCG2 were analyzed by FACS assay in LC-CD133⁺ and LC-CD133⁻. The capabilities of (C) the migration/invasion and (D) the tumor foci (soft agar colony) formation in LC-CD133⁺ were significantly increased compared with LC-CD133⁻ (*p<0.001). Data shown here are the mean±SD of three experiments.

Figure 3. Evaluation of the tumorigenicity of LC-CD133⁺ and LC-CD133⁻ in vivo.

(A) The in vivo tumorigenicity of LC-CD133⁺ and LC-CD133⁻ in tail vein-injected mice was analyzed by macroscopic and histological examination. A1–3: LC-CD133⁻; arrows: normal alveolar structure of lung. A4–6: LC-CD133⁺; arrows: tumor formation. A7–9: LC-CD133⁺; arrows: neovascularity and thrombosis. Bar: 200 µm. (B) The in vivo tumor-restoration and proliferative ability of 10⁴ LC-CD133⁺, 10⁵ LC-CD133⁻ and 5×10⁵ total tumor cells from patient No. 1, 2, 4, and 7 were examined by xenotransplanted tumorigenicity analysis. (C) The tumor repopulation ability of LC-CD133⁺ was studied in transplanted SCID mice. The expression levels of CD133 were determined by FACS analysis from primary LC-CD133⁺, second tumor, and third tumor. Data shown here are the mean±SD of three experiments.

Figure 4. Chemodrugs and radiation sensitivity of LC-CD133⁺ and LC-CD133⁻.

(A) Both 10,000 LC-CD133⁺ and LC-CD133⁻ were plated in a 96-well plate and treated with various concentrations of cisplatin, VP-16, doxorubicin, and paclitaxel for 24 hours in 10% FBS/DMEM/F-12 medium. The survival rate was determined by MTT assay. (B) To determine the radiation effect on the tumor growth rate, an ionizing radiation (IR) dose from 0 to 10 Gy was used to treat LC-CD133⁺ and LC-CD133⁻. *p<0.01: LC-CD133⁺ compared with LC-CD133⁻. (C) The combined treatment effect of radiochemotherapy in LC-CD133⁺ and LC-CD133⁻ were further evaluated. The four protocols—radiation (2 Gy) only, radiation with cisplatin (10 µM), radiation with VP-16 (10 µM), and radiation with paclitaxel (10 nM)—were used. *p<0.01. Data shown here are the mean±SD of three experiments.

Figure 5. Up-regulated expressions of Oct-4 mRNA and protein in LC-CD133⁺.

(A) The amounts of Oct-4 transcripts of isolated LC-CD133⁺ were significantly increased compared with those of LC-CD133⁻ by real-time RT-PCR analyses. (B) Western blott data showed that the protein levels of Oct-4 in LC-CD133⁺ isolated from PLC and LLC were also significantly upregulated compared with those of LC-CD133⁻. (C) The protein expression of Oct-4 in LC-CD133⁺ were effectively blocked by Oct-4 siRNA. Treatment of Oct-4 siRNA in LC-CD133⁺ can impede the capabilities of SB formation and further facilitate SB to differentiate into attached epithelial-like cells. Bar: 100 µm. (D) By using immunofluorescent staining, we showed that the protein expression levels of both Oct-4 and CD133 in LC-CD133⁺ were significantly diminished after Oct-4 siRNA treatment. Bar: 30 µm. (E) The proportion of LC-CD133⁻ were significantly increased in Oct-4 siRNA-treated LC-CD133⁺ by FACS assay. *p<0.001. Data shown here are the mean±SD of three experiments.

Figure 6. Evaluation of the chemoradiosensitivity and apoptotic activity in the knockdown Oct-4 expression of LC-CD133⁺.

(A) Migratory invasion ability and colony formation of LC-CD133⁺ treated by Oct-4 siRNA was significantly decreased compared with non-Oct-4 siRNA-treated LC-CD133⁺ or LC-CD133⁺ treated with scramble-siRNA (control; p<0.001). (B) The combined treatment effect of radiochemotherapy in non-Oct-4 siRNA-treated LC-CD133⁺, scramble-siRNA, and Oct-4 siRNA-treated LC-CD133⁺ were further evaluated. Three groups of cells were exposed to IR (2 Gy) only or radiation plus cisplatin. The cell survival rate was determined by MTT assay. After 72 hours of Oct-4 siRNA treatment, (C) the percentage of Annexin V-positive cells and (D) the activities of caspase 3 (detected by ELISA assay) in Oct-4 siRNA-treated LC-CD133⁺ were significantly increased. (D) The western blot data further showed that PARP was significantly induced in Oct-4 siRNA-treated LC-CD133⁺ when exposed to IR alone, cisplatin alone, or IR combined with cisplatin. (*p<0.05: LC-CD133⁺ with Oct-4 siRNA plus cisplatin vs. LC-CD133⁺ with Oct-4 siRNA only. #p<0.05: LC-CD133⁺ with Oct-4 siRNA plus cisplatin and IR vs. LC-CD133⁺ with Oct-4 siRNA plus IR). Bar: 50 µm. Data shown here are the mean±SD of three experiments.

Figure 7. Inhibition of tumorgeneic activity in Oct-4 siRNA-treated LC-CD133⁺ with chemoradiotherapy.

LC-CD133⁺-GFP (10⁴) cells of different treatment groups were injected into the subcutaneous sites of nude mice. (A) The tumor volume and (B) The number of pulmonary tumor nodules were significantly decreased in Oct-4 siRNA-treated LC-CD133⁺ when exposed to IR alone, cisplatin alone, or IR combined with cisplatin. Cis: cisplatin; Oct-4i: Oct-4 siRNA. (#p<0.01: LC-CD133⁺ treated with Oct-4 siRNA vs. LC-CD133⁺. *p<0.01: LC-CD133⁺ treated with Oct-4 siRNA plus cisplatin vs. LC-CD133⁺ treated with cisplatin only. **p<0.01: LC-CD133⁺ treated with Oct-4 siRNA plus IR vs. LC-CD133⁺ treated with IR only. ##p<0.01: LC-CD133⁺ treated with Oct-4 siRNA plus cisplatin and IR vs. LC-CD133⁺ with Oct-4 siRNA plus IR). (C) The in vivo tumorgeneic activities of LC-CD133⁺- and LC-CD133⁻-injected SCID mice were evaluated by histological review (left upper part). The expression levels of Oct-4 in the lung lesions of LC-CD133⁺-injected SCID mice with the different treatment protocol were studied by immunohistochemistry (IHC). The black arrows indicate the positive signals for Oct-4 expression in the tumors detected by IHC. Bar: 50 µm. (D) Survival analysis of SCID mice injected by LC-CD133⁺, LC-CD133⁺ treated with cisplatin, LC-CD133⁺ treated with IR, and Oct-4 siRNA-treated LC-CD133⁺ groups (IR alone, cisplatin alone, or IR combined with cisplatin). Each group tested six mice (n = 6). Data shown here are the mean±SD of three experiments.