Figure 1.
Met knockdown reduces BxPC-3 and ASPC-1 cell survival.
BxPC-3 (A) or ASPC-1 (B) cells infected with recombinant lentivirus expressing Met knockdown shRNAs (2 or 5) or a non targeting (NT) shRNA were treated without (-) or with (+) HGF and examined by Western analysis for pMet (Y1234/1235), Met, pErk1/2, Erk1/2, pAKT, AKT and β-actin levels (n = 3). Flow cytometry detected reduced Met surface expression in BxPC-3 (C) and ASPC-1 (D) MetKD cells relative to NT cells. Values are expressed as the average mode +/− SEM (***p<0.001; ANOVA, n = 3). Reduced anchorage independent growth of BxPC-3 (E) and ASPC-1 (F) MetKD cells in soft agar relative to NT cells (***p<0.001; ANOVA, n = 3). Met knockdown had minimal effect on the growth of BxPC-3 (G) and ASPC-1 (H) MetKD cells relative to NT cells.
Figure 2.
Decreased cell motility and invasion in vitro by Met knockdown.
Serum-depleted, confluent NT or MetKD BxPC-3 (A) or ASPC-1 (B) cells were wounded with a scratch, six regions marked, and immediately imaged (0 hr) prior to treatment with HGF. Identical regions along the scratch were imaged after 3, 9, 18, and 24 hr and the values were normalized to the initial size of the scratch at 0 hr. Cell migration is reported as the % gap closure at each time point (***p<0.001; ANOVA, n = 3). MetKD reduces BxPC-3 cell invasion in response to HGF relative to control cells (C). Values represent the fold change in the number of cells/field and are reported as the mean +/− SEM (***p<0.001; ANOVA, n = 3). Live cell imaging of ASPC-1 MetKD cells plated on collagen I-coated glass bottom cell culture dishes show decreased cell migration in response to 100 ng/mL HGF (D). Images were collected every 2 min for a total of 120 min and data is expressed as the mean path length +/− SEM per condition (***p<0.001; ANOVA, n>30 cells).
Figure 3.
Met impairs in vivo tumor burden.
Western analysis confirmed that murine (mHGF) and human HGF (hHGF) activate human Met in BxPC-3 (A) and ASPC-1 (B) NT cell lines. No Met activation was observed in the absence of ligand. Values are expressed as mean relative density of pMet/Met +/− SEM (**p<0.01 ***p<0.001; ANOVA, two separate experiments). Tumor size was measured every 10 days using in vivo bioluminescent imaging for BxPC-3 (C) or ASPC-1 (D) cells post injection and are reported as mean tumor size in photons/sec/cm2 (*p<0.05, **p<0.01; ANOVA). Representative bioluminescent images are shown immediately prior to necropsy at 30 days (BxPC-3) or 45 days (ASPC-1). The weight and overall incidence (%) of BxPC-3 primary tumors (E) and ASPC-1 (F) primary and metastatic tumors are listed. Stable knockdown of Met mRNA in BxPC-3 (G) and ASPC-1 (H) MetKD xenografts was confirmed using qRT-PCR. Data were normalized to GADPH and are reported relative to the respective NT control (**<p0.01; ***p<0.001; ANOVA, n = 2).
Figure 4.
Decreased cell survival and tumor angiogenesis in ASPC-1 MetKD tumors.
Immunoprecipitation and Western analysis detected strong phosphorylated Met (pMet) in NT tumors relative to low pMet levels in MetKD tumors (A). Serial sections of paraffin-embedded tumors were either H&E stained (B) or processed for Ki67 (C), active caspase-3 (D) and frozen sections for CD31 (E) staining. Morphometric analysis indicates a significant increase in necrosis and cleaved caspase-3 staining with a corresponding decrease in Ki67 and CD31 staining per high powered field (HPF) in MetKD tumors relative to NT tumors (**p<0.01, ***p<0.001; ANOVA). Representative field images are shown.
Figure 5.
Met knockdown alters secretion of angiogenic factors.
Conditioned media collected from ASPC-1 NT or MetKD cells treated with or without HGF (48 hr) was used to probe angiogenesis antibody arrays in duplicate. Data are presented as the fold difference of the indicated factor in response to HGF +/− SEM (*p<0.05, ***p<0.001; ANOVA) (A). VEGF ELISA shows decreased VEGF levels in BxPC-3 (B) and ASPC-1 (C) MetKD xenografts relative to NT control tumors (***p<0.001; ANOVA). IL-8 ELISA detected reduced IL-8 levels in BxPC-3 (D) and ASPC-1 (E) derived MetKD pancreatic xenografts compared to NT xenografts (*p<0.05, **p<0.01; ANOVA).
Figure 6.
Met kinase activity regulates VEGF and IL-8 secretion.
Treatment of ASPC-1 (A & C) or MIA PaCa-2 without (-Met) or stably expressing exogenous Met (+Met) (B & D) with SU11274 or PF2341066 blocked HGF-induced VEGF (A & B) and IL-8 (C & D) secretion (**p<0.01, ***p<0.001; ANOVA, n = 4). RT-PCR confirmed that ASPC-1 and BxPC-3 cells do not express the IL-8 receptors CXCR1 or CXCR2. GADPH was used as a loading control. The specificity of primer sets for human CXCR1 versus CXCR2 primers was confirmed using plasmids encoding human CXCR1 or CXCR2 respectively.