Bmi‐1 is related to proliferation, survival and poor prognosis in pancreatic cancer

Cell culture.  Five human pancreatic cancer cell lines, PANC‐1, BxPC‐3, ASPC‐1, CFPAC‐1, and SW1990, and a normal pancreas cell line, HPDE6c7, were maintained at 37°C and 5% CO₂ in RPMI‐1640 medium (Gibco, Gaithersburg, MD, USA) supplemented with 10% fetal bovine serum (Invitrogen, Carlsbad, CA, USA), penicillin (100 U mL/L), and streptomycin (0.1 mg mL/L).

Lentiviral vector construction and transduction.  Potential target sequences for RNA interference were chosen using the BLOCK‐iT RNAi Designer. Specific sequences for targeting BMI1 (accession no. NM_005180.5) included the following: si1 (5′‐GGAGGAGGTGAATGATAAA‐3′) and si2 (5′‐AGAAT‐TGGTTTCTTGGAAA‐3′). An invalid RNAi sequence (5′‐TT‐CTCCGAACGTGTCACGT‐3′) was used as a negative control. Short‐hairpin RNAs (shRNAs) were synthesized and cloned into the pENTR/U6 entry plasmid, and shRNAs were then recombined into the pLenti6/BLOCK‐iT expression vector. Inserted sequences were checked with restriction digests and sequencing. Lentivirus production, amplification, and titer were performed using BLOCK‐iT Lentiviral RNAi Expression System protocols (catalog no. K4944‐00; Invitrogen). CFPAC‐1 and PANC‐1 cells were transduced with recombinant lentiviruses harboring BMI1 shRNA or control lentiviral vectors, and pooled stable clones were selected using blasticidin (Sigma, St. Louis, MO, USA). Clones expressing BMI1si1 and BMI1si2 as well as control clones were obtained for both the PANC‐1 and CFPAC cell lines.

Cell growth assays.  Cell growth was assessed using MTT [3‐(4,5‐dimethylthiahiazol‐2‐y1)‐2,5‐diphenytetrazolium bromide; Sigma] assays. Briefly, 2000 cells/well were seeded into 96‐well plates, and cell viability was assayed on Days 1–4 following seeding. Absorption values were determined using an enzyme‐linked immunosorbent assay reader (Dasit, Milan, Italy) at 490 nm.

Colony formation assay and ex vivo tumor inhibition.  Approximately 3 × 10² cells from each stably transfected cell line were plated in six‐well dishes. After 2 weeks, cells were fixed with 20% methanol and stained with 1% crystal violet. Colonies consisting of more than 50 cells were counted per well, and each experiment was performed in triplicate. For in vivo tumorigenicity assays, stably transfected cells (3 × 10⁶ cells/mouse) were prepared in a 0.1 mL volume of PBS and were subcutaneously injected into 5‐week‐old male BALB/c nude mice (four mice per group; BiKai, Shanghai, China). Four weeks after the initial implantation, mice were sacrificed, and the tumors were recovered. Tumor size was monitored using micrometer calipers, and tumor volume was calculated as follows: Volume = length × (width)²/2. Animal experiments were performed in full accordance with the Medicine Institutional Guidelines of the Fourth Military Medical University.

Analysis of cell cycle and apoptosis.  Both cell cycle distribution and spontaneous apoptosis events were detected using a FACScaliber II sorter and Cell Quest FACS system (BD Biosciences, San Jose, CA, USA). To analyze cell‐cycle distribution, cells were synchronized using serum starvation for 24 h and stimulated with complete medium for 24 h before being harvested. Cells were fixed with 70% ethanol at 4°C overnight, washed twice with PBS, and resuspended in Staining Solution (50 μg/mL propidium iodide, 1 mg/mL RNase A, 0.1% Triton X‐100 in PBS) for 30 min at 37°C in the dark. To detect the extent of apoptosis under stress conditions, cells were grown in serum‐free medium for 72 h and stained with fluorescein isothiocyanate (FITC)‐conjugated Annexin V and propidium iodide (PI) using the Annexin V‐FITC Apoptosis Detection kit (Jingmei, Shanghai, China) according to the manufacturer’s protocol.

Western blot analysis.  Cells were solubilized in RIPA lysis buffer (50 mm Tris–HCl [pH 7.4], 1% Triton X‐100, 5 mm EDTA, 1 mm leupeptin, 1 mm phenylmethylsulfonyl fluoride, 10 mm NaF, 1 mm Na₃VO₄) and centrifuged at 20 000 g for 30 min at 4°C to remove debris. Protein concentrations were determined by a BCA assay (Pierce, Rockford, IL, USA). Protein extracts were separated using SDS–PAGE and transferred to nitrocellulose membranes (Amersham Biosciences, Piscataway, NJ, USA). Membranes were blocked with 5% nonfat dry milk in Tris‐buffered saline with Tween‐20 for 1 h, then incubated overnight at 4°C with primary antibodies raised against the following proteins: BMI1 (Upstate); cyclin D1, cyclin E, Cdk2, Cdk4, Cdk6, p21, p27, Akt, Ser⁴⁷³phospho‐Akt, Bcl‐2, and Bax (Santa Cruz Biotechnology, Santa Cruz, CA, USA); and GAPDH (Kangchen, Shanghai, China). Immune complexes were visualized using horseradish peroxidase‐conjugated secondary antibodies and an enhanced chemiluminescence detection system (Amersham Life Science, Piscataway, NJ, USA). Glyceraldehyde‐3‐phosphate dehydrogenase (GAPDH) was used as an internal loading control.

Statistical analysis.  Correlations between categorical variables were analyzed using Pearson’s chi‐square tests, and two‐tailed t‐tests were used for continuous variables. Survival curves were plotted using the Kaplan–Meier method and were compared using the log‐rank test. All statistical analyses were performed using the SPSS software package (SPSS, Chicago, IL, USA). A P‐value <0.05 was considered statistically significant.

B‐cell‐specific Moloney murine leukemia virus insertion site 1 (BMI1) overexpression is associated with the progression and adverse prognosis of PDAC.  To investigate the clinical role of BMI1 during pancreatic carcinogenesis, BMI1 expression was compared with clinical pathological features of patients with PDAC. As shown in Table 1, the nuclear accumulation of BMI1 in PDAC was significantly associated with lymph node metastases (P < 0.05), indicating a correlation between BMI1 expression and PDAC invasion and metastasis. However, there was no difference between the expression of BMI1 and other clinical features such as patient age, sex, pathologic stage, and histology grade (P > 0.05 for all comparisons). Further, Kaplan–Meier survival analysis demonstrated that patients harboring BMI1 overexpression had a significantly shorter overall survival than patients with lower levels of BMI1 expression (P = 0.007, log‐rank test; Fig. 1e). These observations indicate that increased BMI1 is associated with PDAC clinical progression.

B‐cell‐specific Moloney murine leukemia virus insertion site 1 (BMI1) knockdown inhibited the growth of PDAC cells both in vitro and in vivo.  To determine whether BMI1 effects tumor cell growth, two BMI1‐specific shRNAs were constructed and inserted in lentiviral vectors. PANC‐1 and CFPAN‐1 cells, two cell lines with high levels of BMI1 expression, were infected with the lentiviral constructs and selected with blasticidin for 10 days. Expression of BMI1 was subsequently determined by western blot analysis. As shown in Figure 2(a), the shRNA BMI1si2 more effectively down‐regulated the expression of BMI1 in both PANC‐1 and CFPAC‐1 cells than did BMI1si1. Furthermore, the control vectors had no effect on the expression of BMI1 in either cell line. As a result, PANC‐1‐BMI1si2 cells and CFPAC‐1‐BMI1si2 cells were further characterized.

We first examined the effect of BMI1 depletion on cell viability using MTT assays. As shown in Figure 2(b), BMI1 depletion significantly suppressed the growth of PANC‐1 and CFPAC‐1 cells compared with control cells (P < 0.05). In contrast, BMI1 depletion did not affect the viability of HPDE6c7 cells (data not shown). Colony formation analysis showed that BMI1‐depleted tumor cells had a greatly reduced capacity to form colonies compared with the control cells (P < 0.05; Fig. 2c). To confirm these findings in vivo, xenograft tumor growth assays were performed in nude mice. Compared with the control cells, injection of PANC‐1 (CFPAC‐1) BMI1si2 cells led to dramatically decreased tumor volume (P < 0.05; Fig. 2d). Taken together, these data show that decreased expression of BMI1 in PDAC cells effectively suppressed the growth of PDAC cells both in vitro and in vivo.

B‐cell‐specific Moloney murine leukemia virus insertion site 1 (BMI1) knockdown arrested PDAC cells in the G0‐G1 phase.  Our flow cytometry analysis showed that the cell‐cycle distribution of PANC‐1 and CFPAC‐1 cells was significantly affected following BMI1 depletion. As shown in Figure 3, there was a higher proportion of G0‐G1 phase cells (56.92%) in PANC‐1‐BMI1si2 cells than in the control (42.03%). A compensatory decrease in the S (35.73%) and G2/M phase (7.35%) proportions was also detected as compared with the control in S (41.34%) and G2/M phases (16.63%). Similar results were obtained in CFPAC‐1 cells (Fig. 3). These results suggest that BMI1 silencing inhibited the entry of cells into the S phase and therefore suppressed cell growth. Furthermore, western blotting was performed to explore the cell cycle regulatory role of BMI1. Following BMI1 silencing, the expression of cyclin D1, cyclin‐dependent kinase (CDK)‐2, and CDK4 were decreased, while the level of p21 was increased (Fig. 4). Additionally, the levels of cyclin A, cyclin E, and p27 were also detected but were not significantly altered (data not shown).

B‐cell‐specific Moloney murine leukemia virus insertion site 1 (BMI1) knockdown increased apoptosis of PDAC cells under serum depletion.  To further explore the functional mechanisms by which BMI1 promotes PDAC tumor growth, BMI1si2‐infected PANC‐1 and CFPAC‐1 cells were maintained in serum‐free media for 72 h. A dramatic increase in apoptosis was detected in both cell lines compared to controls (Fig. 5). The percentage of apoptotic cells (Annexin‐V+/PI− and Annexin‐V+/PI+) in PANC‐1‐BMI1si2 (CFPAC‐1‐BMIsi2) cells was 28.38% (30.24%), while the percentage of apoptosis was 10.2% (12.84%) in controls. These results suggest that BMI1 protects PDAC cells from stress‐induced death. Furthermore, the expression of several cell survival‐related proteins was evaluated by western blot (Fig. 6). Following BMI1 silencing, the expression of phospho‐Akt and Bcl‐2 were decreased, while the level of Bax was increased. Additionally, there was no significant change in Bcl‐2 protein levels between PANC‐1‐BMI1si2 cells and control cells. This lack of difference may be due to the absence or low levels of expression of Bcl‐2 in PANC‐1 cells^{( 31} , ³² , ^{33 )} that could have impeded our examination of the effects of BMI1 silencing.