| Purpose1. To investigate the genetic characterization of DMBA-induced pancreatic cancer2. To investigate the role of BMSCs in the carcinogenesis of pancreatic cancer3. To investigate the role of BMSCs in the development of pancreatic cancerMethods1. The model of DMBA-induced pancreatic cancer was established while histological pathology of lesions was evaluated. K-ras and p16 alterations were detected by PCR and DNA sequencing.2. Cellular surface markers of mouse GFP-BMSCs were detected by flow cytometry. Adipogenic and osteogenic differentiation were induced in corresponding medium. The model of DMBA-induced pancreatic cancer and GFP-BMSCs injection via tail vein was established. The migration and distribution of GFP-BMSCs were identified by the immunohistology and immunofluorescence of GFP. The contribution of GFP-BMSCs was uncovered by immunohistology and immunofluorescence confocal microscopy of CK-19, a-SMA and Desmin.3. Cellular surface markers of HMSCs (Human BMSCs) were detected by flow cytometry. Adipogenic and osteogenic differentiation were induced in corresponding medium. In vitro study:after indirect and direct co-culture of HMSCs and PCs (pancreatic cancer cells, Panc-1 or Bxpc-3), the effect of HMSCs on the proliferation of PCs was quantified with low power epifluorescence microscopy and a hemocytometer. After induction of apoptosis in PCs by hydrogen peroxide, the effect of HMSCs on the apoptosis of PCs was evaluated with Annexin V flow cytometric assays. The effect of HMSCs on the migration and invasion was analyzed using transwell chambers. In vivo study:subcutaneous xenograft model was established using injection of panc-1 alone or co-injection of panc-1 and HMSCs to examine the effect of HMSCs on tumor growth. Model of orthotopic pancreatic cancer was established using injection of Bxpc-3 alone or co-injection of Bxpc-3 and HMSCs to observe the effect of HMSCs on tumor metastasis in the spleen, liver, mesentery and retroperitonium. HE staining, masson’s trichrome staning, ki-67 immunohistology and TUNEL assay were performed to analyze the effect of HMSCs on the pathological histology, formation of the stroma, proliferation index and apoptosis index in vivo.Results1. No K-ras mutations were observed in those lesions. The overall frequency of p16 alteration was 42.86%(15/35) and all changes resulted from point mutations instead of homozygous deletions. P16 mutations were present in 30.77%(8/26) of PanINs and 77.78%(7/9) of carcinoma, which showed a statistically significant difference (P<0.05). Increasing p16 alterations were detected in 20.00%(1/5) of PanIN-1,28.57% (2/7) of PanIN-2 and 35.71%(5/14) of PanIN-3 lesions.11 transversion mutations and 10 transition mutations were identified in 35 cases. In addition, most samples contained mutations in two or more condons.2. GFP expression was found in both parenchyma and mesenchyma of pancreatic cancer in the experiment group. GFP+/CK-19+ subpopulation which accounted for 10.8±2.6% of GFP+ cells was found distributed through the tumor parenchyma. GFP+ /α-SMA+ and GFP+/Desmin+ subpopulations which accounted for less than 5% of GFP+ cells was present at the tumor mesenchyma.3. Panc-1 cell number in the conditioned medium group or co-culture group was statistically higher than that in the control group (7.83±0.65 vs.5.38±0.69×104 cells/well, p<0.05; 10.62±0.47 vs.5.38±0.69×104 cells/well, p<0.01). In addition, Bxpc-3 cell number in the conditioned medium group or co-culture group was statistically higher than that in the control group (24.57±1.33 vs.15.43±0.61×104 cells/well, p<0.05; 38.48±1.01 vs.15.43±0.61×104 cells/well, p<0.01). The ability of Panc-1 and Bxpc-3 cells to migrate toward the HMSCs conditioned medium was significantly greater than the ability of these cells to migrate toward the control medium (225.6±35.0 vs.140.6±18.5 cells/field, P<0.001; 101.9±15.1 vs.43.6±11.5 cells/field, P<0.001). Moreover, the invasive ability of Panc-1 and Bxpc-3 cells was significantly greater toward MSC-conditioned medium than toward control medium (101.8±16.8 vs.45.8±13.1, P<0.01; 114.4±21.8 vs.49.0±11.34, P<0.01). The growth rate for the Panc-1+HMSCs group was approximately 5 times greater than that of the Panc-1 group (P<0.0001). The tumor weight for the Panc-1+HMSCs group was statistically higher than that of the Panc-1 group (0.325±0.091 vs.0.063±0.017 g, P<0.05). HE and masson’s trichrome staining showed the desmoplasia surrounding panc-1 cells in the Panc-1+HMSCs group was significantly more than that of the Panc-1 group (27.1±4.3 vs.2.7±0.8%, P<0.001). The Ki-67 LI was significantly higher (68.6±8.1 vs.30.1±6.4%, P<0.01) and the AI was significantly lower in the Panc-1+HMSCs (10.0±2.5 vs.2.8±0.8%, P<0.01). Co-injection of Bxpc-3 cells with HMSCs resulted in a significantly greater number of spleen and liver metastases than that resulting from injection of Bxpc-3 cells alone (25.0 vs.75.0%, P<0.05; 25.0 vs.83.3%, P<0.05).Conclusions1. These findings indicate that point mutations in K-ras gene, at least in exons 1 and 2, are not involved in the development of pancreatic cancer in this model, which is different from those of human lesions. However, p16 alteration is a common event in the carcinogenesis of this model and the mutation pattern is almost similar to that of human lesions.2. BMSCs, which migrate to the tumor parenchyma and mesenchyma, are involved in the carcinogenesis of DMBA-induced pancreatic cancer by trans-differentiation. BMSCs are likey to be potential cellular origin of pancreatic cancer.3. BMSCs promote growth and metastasis of pancreatic cancer by enhancing proliferation, migration and invasion and by inhibiting apoptosis of tumor cells. BMSCs play a pro-tumorigenic role in the progression of pancreatic cancer, which shows that the security of using BMSCs as vehicles for cancer therapy needs further evaluation. |
PDF Full Text Request