A novel nuclear Src and p300 signaling axis controls migratory and invasive behavior in pancreatic cancer

The presence of Src in the nuclear compartment has been previously reported, although its significance has remained largely unknown. We sought to delineate the functions of the nuclear pool of Src within the context of malignant progression. Active Src is localized within the nuclei of human pancreatic cancer cells and mouse fibroblasts over-expressing c-Src where it is associated with p300. Nuclear Src additionally promotes the tyrosine phosphorylation of p300 in pancreatic cancer Panc-1 cells. Src, together with p300, is associated with the high-mobility group AT-hook (HMGA)2 and SET and MYND domain-containing protein (SMYD)3 gene promoters and regulates their expression in a Src-dependent manner. These nuclear Src-dependent events correlate with anchorage-independent soft-agar growth and the migratory properties in both pancreatic Panc-1 cells and mouse fibroblasts over-expressing Src. Moreover, analyses of human pancreatic ductal adenocarcinoma (PDAC) tumor tissues detected the association of nuclear Src with the HMGA2 and SMYD3 gene promoters. Our findings for the first time show the critical importance of nuclear Src and p300 function in the migratory properties of pancreatic cancer cells. Further, data together identify a previously unknown role of nuclear Src in the regulation of gene expression in association with p300 within the context of cells harboring activated or over-expressing Src. This novel mechanism of nuclear Src-p300 axis in PDAC invasiveness and metastasis may provide an opportunity for developing more effective early clinical interventions for this lethal disease. Active Src is complexed with and phosphorylates p300 in the nucleus, and the complex is bound to HMGA2 and SMYD3 genes, thereby regulating their expression to promote pancreatic tumor cell migration and invasiveness.


MATERIALS AND METHODS
digest RNA and accessible chromatin. Next, 7mL of ice-cold sucrose cushion buffer (30% sucrose, 20 mmol/L triethanolamine, pH 7.5, 0.1 mmol/L MgCl 2 , 1 mmol/L DTT, and protease and phosphatase inhibitors) was underlaid and the tube was centrifuged in a tabletop centrifuge at 750g for 10 minutes. The upper layer supernatant was removed and labeled as "DNase/RNase fraction," representative of the soluble nucleoplasmic material liberated by digestion of accessible chromatin and RNA. Next, the underlayment was removed and the pellet was resuspended in 0.5mL extraction buffer (10% sucrose, 20 mmol/L triethanolamine, pH 7.5, 0.1 mmol/L MgCl 2 , 1 mmol/L DTT, and protease and phosphatase inhibitors), to which was added 0.25mL extraction buffer containing 0.3mg/mL heparin dropwise, while gently vortexing. Seven milliliters of sucrose cushion buffer was again underlayed and the samples were centrifuged at 750g for 15 minutes. The supernatant was removed and labeled as "Heparin extract," representative of the DNAse-inaccessible nucleoplasmic heterochromatin. The pellet was lysed in radio immunoprecipitation assay (RIPA) buffer (25 mmol/L Tris-HCl, pH 7.5, 150 mmol/L NaCl, 1% NP-40, 1% sodium deoxycholate, 0.1% SDS) and labeled "Nuclear envelope." For "Crude nuclear envelope" preparation, the pellet following DNAse/RNase treatment was lysed in RIPA buffer and centrifuged at 17,000g for 15 minutes at 4°C.

Stable transfectant selection
Src, Yes, and Fyn-null MEF (SYF-/-) cells were first subjected to a Zeocin (Gibco) kill curve and a concentration of 200 μg/mL was determined to be optimal for selection. Cells were transfected with either pTracer-NLSvSrc or pTracer-NESvSrc using Lipofectamine 2000 (Invitrogen). After two days, medium containing 200 μg/ mL Zeocin was added freshly every three to five days for approximately two weeks until colonies were visible. Cells from these colonies were harvested by trypsinization and seeded into new culture dishes as pools of either NLSvSrc or NESvSrc stable transfectants. These cells were then cultured in media containing 50 μg/mL Zeocin to maintain selective pressure.

Chromatin immunoprecipitation (ChIP)
ChIP assay was performed according to the Affymetrix Chromatin Immunoprecipitation Assay protocol and as previously described [4]. Briefly, 1.2 x 10 7 cells were fixed in 1% formaldehyde at room temperature for seven minutes, and then fixation was quenched with 0.125M glycine for 10 minutes at room temperature. Cells were then washed three times with ice cold PBS and resuspended in lysis buffer for nuclear isolation. Nuclei were then sonicated (Omni International) at 50% power set to 50% pulse for shearing DNA to an average fragment size of 200-700bp. Lysates were then precleared with protein A/G agarose beads (Santa Cruz Biotechnology, Dallas, TX) for one hour at 4°C while rocking. Cleared lysates were incubated with anti-p300 (C-20), anti-Src (B12) antibody, or pre-immune mouse or rabbit antibody as control. Samples were incubated overnight at 4°C while rocking. Immune complexes were precipitated with protein A/G agarose beads and washed in buffers of increasing stringency. Beads were washed twice in Wash Buffer 1 (20 mmol/L Tris pH 8.0, 2 mmol/L EDTA, 1% Triton X-100, 150 mmol/L NaCl, and protease and phosphatase inhibitors), once each with Wash Buffer 2 (20 mmol/L Tris pH 8.0, 2 mmol/L EDTA, 1% Triton X-100, 0.1% SDS, 500 mmol/L NaCl, and protease and phosphatase inhibitors) and Wash Buffer 3 (10 mmol/L Tris pH 8.0, 1 mmol/L EDTA, 25 mmol/L LiCl, 0.5% NP-40, 0.5% Deoxycholate), and then twice in TE buffer (10 mmol/L Tris pH 8, 1 mmol/L EDTA). Samples were eluted and then treated with proteinase K (Pierce, Waltham, MA) at 55°C overnight to both digest protein and reverse crosslinks. DNA was purified with ChIP Spin Columns (Zymo Research Corp, Irvine, CA) and subjected to quantitative polymerase chain reaction (qPCR) analysis.

ChIP-on-chip assay and analysis
Purified DNA from ChIP samples was subjected to two rounds of manual linear amplification with adapter-linked random nonamer primers (GTTTCCCAGTCACGGTC(N) 9 ) using Sequenase (Affymetrix, Santa Clara, CA) reagents. After linear amplification, samples were exponentially amplified using Taq polymerase for 30 cycles with a 20%/80% ratio of dUTP/dTTP in order to incorporate uracil into the fragments. Amplified DNA was sent to the University of Hawaii Cancer Center Genomics Core Facility (Honolulu, HI) for uracil DNA Glycosylase fragmentation and labeling, and hybridization to the Human Promoter 1.0r Microarrays (Affymetrix). Once the arrays were hybridized and scanned, 'CEL' files containing raw intensity values were provided for analysis. 'CEL' files corresponding to each array were converted to 'TAG' files in the Affymetrix Tiling Analysis software suite. 'TAG' files were generated from raw data using quartile normalization with a target intensity of 250. Data were expressed with a Log2 signal scale and a -10Log10 p-value scale. Analysis of 'TAG' files was performed as a one-sided upper perfect match only using a bandwidth of 250. The p-value threshold was set to 23 (p=.005) with a maximum gap of 100bp and a minimum run of 200bp in order to exclude nonspecific hits from individual probes with artificially high signal. The analysis generated 'CHP' files, which could be viewed on the Affymetrix Integrated Genome Browser represented as p-value of enrichment of the specific IP vs the control IgG for each probe.

Cell viability assay
CyQuant cell proliferation assay was performed as previously described [4,5]. Cells were seeded in triplicate in 96-well plates and grown for either 72 or 48 hours in the presence of inhibitors. Initial cell density was determined such that the DMSO control would grow to < 90% confluence by the end time point.

Wound healing assay
Cells were seeded as confluent monolayers into 12well plates in triplicate. Monolayers were scratched with a p10 pipet tip, then fresh medium containing 100 nmol/L dasatinib or 20 μmol/L C646 was added. Plates were then marked with a thin-tipped permanent marker to denote the imaging field across the scratch. Cells were imaged in phase contrast with the Zeiss Axiovert 200 microscope at 10X magnification immediately after the addition of inhibitor-containing media and noted as the zero hour time point. Cells were allowed to migrate for the indicated amount of time and imaged again at the same field marked on the plate.