- Email: [email protected]
Dasatinib Inhibits the Development of Metastases in a Mouse Model of Pancreatic Ductal Adenocarcinoma
GASTROENTEROLOGY 2010;139:292–303
Dasatinib Inhibits the Development of Metastases in a Mouse Model of Pancreatic Ductal Adenocarcinoma JENNIFER P. MORTON,*,‡ SAADIA A. KARIM,* KATHRYN GRAHAM,* PAUL TIMPSON,* NIGEL JAMIESON,‡ DIMITRIS ATHINEOS,*,‡ BRENDAN DOYLE,* COLIN MCKAY,§ MAN–YEUNG HEUNG,储 KARIN A. OIEN,‡ MARGARET C. FRAME,储 T. R. JEFFRY EVANS,*,‡ OWEN J. SANSOM,* and VALERIE G. BRUNTON储 *Beatson Institute for Cancer Research, Glasgow, United Kingdom; ‡Centre for Oncology and Applied Pharmacology, Division of Cancer Sciences and Molecular Pathology, University of Glasgow, Glasgow, United Kingdom; §West of Scotland Pancreatic Unit, Glasgow Royal Infirmary, Glasgow, United Kingdom; and 储Edinburgh Cancer Research Centre, University of Edinburgh, Edinburgh, United Kingdom
BASIC–LIVER, PANCREAS, AND BILIARY TRACT
BACKGROUND & AIMS: Pancreatic ductal adenocarcinoma (PDAC) is a highly invasive and metastatic disease for which conventional treatments are of limited efficacy. A number of agents in development are potential anti-invasive and antimetastatic agents, including the Src kinase inhibitor dasatinib. The aim of this study was to assess the importance of Src in human PDAC and to use a genetically engineered mouse model of PDAC to determine the effects of dasatinib on PDAC progression. METHODS: Src expression and activity was measured by immunohistochemistry in 114 human PDACs. Targeting expression of Trp53R172H and KrasG12D to the mouse pancreas results in the formation of invasive and metastatic PDAC. These mice were treated with dasatinib, and disease progression monitored. Cell lines were derived from mouse PDACs, and in vitro effects of dasatinib assessed. RESULTS: Src expression and activity were up-regulated in human PDAC and this correlated with reduced survival. Dasatinib inhibited the migration and invasion of PDAC cell lines, although no effects on proliferation were seen at concentrations that inhibited Src kinase activity. In addition, dasatinib significantly inhibited the development of metastases in Pdx1-Cre, Z/EGFP, LSL-KrasG12D/⫹, LSL-Trp53R172H/⫹ mice. However, there was no survival advantage in the dasatinib-treated animals owing to continued growth of the primary tumor. CONCLUSIONS: This study confirms the importance of Src in human PDAC and shows the usefulness of a genetically engineered mouse model of PDAC for assessing the activity of potential antimetastatic agents and suggests that dasatinib should be evaluated further as monotherapy after resection of localized invasive PDAC. Keywords: Pancreatic Cancer; Metastasis; Src Kinase; Dasatinib.
I
nfiltrating pancreatic ductal adenocarcinoma (PDAC) is the fifth most common cancer and the fourth most common cause of cancer deaths in the United Kingdom. Aggressive invasion and early metastases are characteristic of the disease, such that 90% of patients have surgi-
cally unresectable disease at the time of diagnosis. Furthermore, most systemic therapies are largely ineffective in advanced, inoperable disease, and the estimated 5-year survival rate is less than 5%.1 In addition, a majority of the selected patients who undergo potentially curative resection for small, localized lesions inevitably develop recurrent or metastatic disease,2 presumably because of the presence of distant micrometastases at initial diagnosis. Consequently, the development of more effective strategies to treat preinvasive pancreatic cancer, or micrometastatic disease, is of paramount importance. In human PDAC, development, progression, and metastases arise via the accumulation of multiple genetic and epigenetic changes, including inactivation of tumorsuppressor genes and activation or overexpression of proto-oncogenes.3,4 PDAC arises from preinvasive lesions called pancreatic intraepithelial neoplasms (PanINs) whose progression is driven by activating mutations in Kras, which are detected in almost 100% of advanced PDACs.5 In a recently developed mouse model of PDAC, KrasG12D and Trp53R172H are expressed in pancreatic cells, resulting in the formation of PanINs that develop into invasive and widely metastatic PDAC.6 The genetics of the mouse model mimics the human disease and importantly the histopathology, disease progression, and sites of metastases in the mice also recapitulate many aspects of the human disease, providing a good model with which to study potential therapies for PDAC.7 Currently, there is considerable interest in exploiting Src kinase as a novel therapeutic target in malignant disease, and in particular whether such an agent may be useful in inhibiting the development of metastatic disease.8 Src is a nonreceptor tyrosine kinase whose expression is increased in a number of epithelial tumors, most notably in colon and breast,9 and 2 studies have reported Abbreviations used in this paper: CI, confidence interval; FAK, focal adhesion kinase; IQR, interquartile range; LNR, lymph node ratio; MMP, matrix metalloproteinase; PanIN, pancreatic intraepithelial neoplasia; PDAC, pancreatic ductal adenocarcinoma. © 2010 by the AGA Institute 0016-5085/$36.00 doi:10.1053/j.gastro.2010.03.034
overexpression of Src in PDAC.10,11 Src is involved in many aspects of tumor cell behavior that impact on their metastatic capacity such as survival, adhesion, migration, and invasion.9 A number of small-molecule Src kinase inhibitors are now in clinical development including dasatinib, a highly potent, adenosine triphosphate– competitive inhibitor of Src family and Abl kinases12,13 and we have used dasatinib to determine how best to evaluate a potential antimetastatic agent in the clinical management of pancreatic cancer. We have shown that dasatinib, by inhibiting Src kinase activity, inhibits the development of metastases in the Pdx1-Cre, Z/EGFP, LSL-KrasG12D/⫹, LSL-Trp53R172H/⫹ mouse model of PDAC but has no effect on the proliferation of the primary tumor, supporting the hypothesis that Src kinase inhibitors may function primarily as antimetastatic agents. It remains to be determined if dasatinib could improve overall survival when used in the adjuvant (postoperative) setting after removal of the primary tumor burden.
Material and Methods Genetically Modified Mice and Animal Care Conditional LSL-KrasG12D/⫹ mice (strain 01XJ6, mouse models of human cancer consortium [MMHCC], NCI-Frederick, Frederick, MD)14 were mated to LSLTrp53R172H/⫹ mice (strain 01XL9, MMHCC, NCIFrederick),15 and Pdx1-Cre mice16 were mated to Z/EGFP mice.17 Progeny from these crosses then were interbred to obtain Pdx1-Cre, Z/EGFP, LSL-KrasG12D/⫹, LSL-Trp53R172H/⫹ mice on a mixed background. Mice were genotyped by polymerase chain reaction analysis as described previously.6 Experiments were performed in compliance with UK Home Office guidelines. Mice were dosed daily by oral gavage with 10 mg/kg dasatinib (Bristol-Myers Squibb, Princeton, NJ) in 80 mmol/L citrate buffer. Tumor and metastatic burden was assessed by gross pathology and histology and organs/tumors were removed and fixed in 10% buffered formalin. For histologic detection of hepatic metastases, serial sections of the whole liver were analyzed. The Olympus OV100 Whole Mouse Imaging System (Olympus Corp, Tokyo, Japan) was used for imaging in live mice. Anesthesia was induced and maintained with a mixture of isoflurane and oxygen.
Immunohistochemistry and Immunocytochemistry Immunohistochemistry of formalin-fixed, paraffin-embedded tissues was performed as described previously.18 Primary antibodies used were anti–phospho-Src Y418 1:200 and anti-Src 1:800 (both Cell Signaling, New England Biolabs, Hitchin, UK). For immunocytochemistry, cells were fixed in 10% neutral buffered formalin and permeabilized with 0.1% Triton X-100 before staining with an anti-Pdx1 antibody (Abcam, Cambridge, UK).
DASATINIB AND MOUSE PANCREATIC CANCER
293
Cell Culture and Dasatinib Treatment PDAC cell lines were generated from primary pancreatic tumors taken from Pdx1-Cre, Z/EGFP, LSLKrasG12D/⫹, LSL-Trp53R172H/⫹ mice and then passaged in growth media (Dulbecco’s modified Eagle medium containing 10% fetal bovine serum and 2 mmol/L L-glutamine). Dasatinib was prepared as a 100 mmol/L stock in dimethyl sulfoxide and diluted in growth media. Recombination of Trp53R172H in the cell lines was confirmed by polymerase chain reaction using forward: AGCCTGCCTAGCTTCCTCAGG, reverse: CTTGGAGACATAGCCACACTG primers.
Western Blot Analysis Western blot analysis was performed as described previously18 using anti–phospho-Src Y418, phosphop130Cas Y249 (Cell Signalling), phospho-focal adhesion kinase (FAK) Y861 (Becton Dickinson, Oxford, UK), or anti-Src (Cancer Research UK, London, UK) primary antibodies all at 1:1000.
Proliferation Assay A 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) proliferation assay was performed as described previously.18 Mean values were calculated from quadruplicate wells and plotted as the mean percentage of the untreated controls. The highest concentration of dimethyl sulfoxide (drug diluent) added to the cells had no effect on cell proliferation (data not shown).
Migration Assay Confluent monolayers of PDAC cells were scored with a fine pipette tip to produce a wound. Migration into the wound was monitored by time-lapse video microscopy over 48 hours in the presence or absence of dasatinib using 20⫻ magnification on a Zeiss Axiovert S100 microscope (Zeiss, Welwyn Garden City, UT) with AQM Advance Software (Kinetic Imaging, Nottingham, UK). Three representative areas were scored for each treatment and the area of the wound was calculated using ImageJ software (National Institutes of Health, Bethesda, MD). Cells also were transfected with 2 mol/L ON-TARGETplus src SMARTpool small interfering (si) RNA solution or a 2 mol/L ON-TARGETplus nontargeting siRNA solution according to the manufacturer’s protocol (Dharmacon, Fisher Scientific, Loughborough, UK), and after 24 hours, migration was monitored as described earlier.
Invasion Assay Cells were seeded on the bottom of Transwell inserts (Corning, Fisher Scientific, Loughborough, UK) containing polymerized collagen type I (Becton Dickinson). Transwell inserts then were placed in serum-free medium, and medium supplemented with 10% fetal calf
BASIC–LIVER, PANCREAS, AND BILIARY TRACT
July 2010
294
MORTON ET AL
serum and 50 ng/mL hepatocyte growth factor (R&D Systems, Abingdon, UK) was placed on top of the matrix in the absence or presence of dasatinib. Five days after seeding, invading cells were stained with Calcein-AM (Invitrogen, Paisley, UK) and visualized by confocal microscopy (Leica SP2, Leica, Milton Keynes, UK). Serial optical sections were captured at 10-m intervals and quantified using ImageJ software using the area analysis module. Invasion was calculated as cells that had moved more than 20 m into the collagen.
Matrix Metalloproteinase Zymography Matrix metalloproteinase (MMP) activity was assessed from culture supernatants of PDAC cells. Cells were seeded onto collagen-coated dishes and allowed to adhere, were then washed, and serum free-media was added. After 24 hours in the presence or absence of dasatinib (100 nmol/L), culture supernatants were collected and equal volumes were subjected to gelatin zymography (Invitrogen) under nonreducing conditions according to the manufacturer’s instructions.
Subcutaneous Tumor Growth in Mice Mouse PDAC cell lines were injected subcutaneously into the right flank of 4-week-old to 6-week-old female CD-1 nude mice (Charles Rivers, Harlan, UK). When tumors were established dasatinib was administered by oral gavage. The mice were killed 2 hours later, and the tumors were formalin-fixed for paraffin-embedding or snap-frozen in liquid nitrogen.
Tissue Microarray BASIC–LIVER, PANCREAS, AND BILIARY TRACT
Src and phospho-Src expression levels in a human pancreaticobiliary tissue microarray were scored based on staining intensity and area of tumor cells using a weighted histoscore calculated from the sum of (1 ⫻ % weak staining) ⫹ (2 ⫻ % moderate staining) ⫹ (3 ⫻ % strong staining), providing a semiquantitative classification of staining intensity. Representative images are shown in Figure 1 and further details of the analysis and array are provided in the Supplementary Material section.
Results Src Expression and Activity in Human PDAC We looked at Src expression and also activity using an antibody against the autophosphorylation site of Src (tyrosine 418) in a tumor array of 114 human PDACs (Figure 1). Src was expressed in all tumors, with similarly high levels of the activated protein noted (94.1%). The median histoscore for membranous Src expression was 90.8 (interquartile range [IQR], 63.8 –137.1) whereas the median membranous phospho-Src expression was 80.0 (IQR, 38.3–112.9). Src expression was found in 84.7% of normal ducts (89 of 105) whereas phosphorylated Src was present in 82.6% of normal ducts (92 of 108). However, the levels of expression and activity
GASTROENTEROLOGY Vol. 139, No. 1
were very low in normal ducts, and when expression in the tumor and normal ducts from the same patient was compared we found that in 74.7% of cases there was an increase in Src expression in the tumors, whereas 60.1% of tumors had increased activated Src compared with the normal duct from the corresponding patient according to the highest scoring regions.
Src Expression With Clinicopathologic and Survival Analyses Low tumor grade was associated with significantly reduced expression of Src (median histoscore, 83.3; IQR, 61.7–122.5) compared with high-grade tumors (median histoscore, 112.7; IQR, 73.3–143.1; P ⬍ .039) (Figure 2A). Increasing lymph node ratio (LNR: number of involved nodes/total number of resected nodes) was associated with greater Src expression with a median histoscore of 105.1 (IQR, 87.1–163.1) for node-negative tumors (n ⫽ 24) compared with a median histoscore of 103 (IQR, 79.2–155.8) for tumors with an LNR of 0.1 or less (n ⫽ 52); 121.1 (IQR, 102.5–212.4) for an LNR of 0.1– 0.5 (n ⫽ 13), and 153 (IQR, 113.5–222.9) for an LNR of 0.5 or greater (n ⫽ 30). Src expression was significantly greater in tumors with an LNR of greater than 0.1 compared with an LNR of 0.1 or less or with no lymph node involvement (P ⬍ .009) (Figure 2B). The presence of vascular invasion within the tumor was associated with increased Src expression (median histoscore, 105.2; IQR, 75.4 –145.4) compared with tumors with no vascular invasion (median histoscore, 74.6; IQR, 59.2–137.1; P ⬍ .03) (Figure 2C). There was no association with tumor stage, tumor size, or resection margin status. Full details are provided in Supplementary Table 1. Univariate outcome assessment revealed a median overall survival time of 15.4 months (95% confidence interval [CI], 7.9 –22.9 mo) for patients with high-Srcexpressing tumors (median histoscore, ⬎108); compared with 17.9 months (95% CI, 13.4 –22.3 mo) for low-Srcexpressing tumors (median histoscore, ⬍108); however, this failed to reach significance (P ⫽ .121). Combining tumor grade with Src expression identified a particularly poor prognostic group, with high tumor grade and high Src expression (n ⫽ 22) associated with a median survival of 7.6 months (95% CI, 2.55–12.8 mo) after pancreaticoduodenectomy compared with 16.8 months (95% CI, 0.5–38.4 mo) for high-grade and low-Src-expressing tumors (log-rank, P ⬍ .003) (Figure 2D and Supplementary Figure 1).
Activated Src Expression With Clinicopathologic and Survival Analyses Phospho-Src Y418 membranous expression was not associated with clinicopathologic factors including tumor stage, lymph node invasion, tumor grade, vascular invasion, perineural invasion, or resection margin status. Univariate survival analysis was performed using the me-
DASATINIB AND MOUSE PANCREATIC CANCER
295
BASIC–LIVER, PANCREAS, AND BILIARY TRACT
July 2010
Figure 1. Src expression and activity in human ductal adenocarcinoma. Immunohistochemical analysis of Src (left panels) and phospho-Src Y418 (right panels) in a tumor array of human PDAC. Representative images are shown for absence of staining, ⫺; weak staining, ⫹; moderate staining, ⫹⫹; and strong staining, ⫹⫹⫹. Original magnification, 10⫻.
dian membranous expression value of 80 as a cut-off; however, there was no difference in outcome between those patients with tumors showing low or high phospho-Src expression (14.8 mo; 95% CI, 11.6 –17.9 mo;
compared with 19.6 mo; 95% CI, 16.2–23.1 mo) (log-rank, P ⫽ .621). We previously observed that there is a poor correlation between the levels of phospho-Src and Src expression in
296
MORTON ET AL
GASTROENTEROLOGY Vol. 139, No. 1
BASIC–LIVER, PANCREAS, AND BILIARY TRACT
Figure 2. High Src expression and activity in human PDAC correlates with reduced survival. (A) Box plot of Src histoscore vs tumor grade. (B) Box plot of Src histoscore vs LNR. (C) Box plot of Src histoscore vs vascular invasion. (D) Kaplan–Meier analyses for high-grade patients (n ⫽ 34) showing that high Src expression is associated with poorer outcome. (E) Kaplan–Meier analyses for resection margin–positive patients (n ⫽ 90) showing that cases with evidence of grade 3 phospho-Src expression have poorer outcome after pancreaticoduodenectomy compared with those with grade 1/2 phospho-Src expression.
mouse PDAC, with phospho-Src expression restricted to more invasive areas of the tumor. With this in mind, we performed survival analysis on patients with tumors where hotspots of phospho-Src expression were identified. Based on intensity of expression we identified a group of 23 patients with evidence of grade 3 phospho-
Src expression. Although this increased level of expression did not yield prognostic significance with all patients included, subgroup analysis of the 90 resection margin–positive patients did show that tumors with regions of very high phospho-Src expression had a significantly poorer outcome after pancreaticoduodenectomy
(median overall survival, 10.3 mo; 95% CI, 5.99 –14.5 mo) compared with tumors with no evidence of grade 3 phospho-Src expression (median overall survival, 15.6 mo; 95% CI, 12.7–18.5 mo) (log-rank, P ⫽ .038) (Figure 2E).
Src Activity and Expression in Pdx1-Cre, Z/EGFP, LSL-KrasG12D/ⴙ, LSL-Trp53R172H/ⴙ Mice Targeting endogenous expression of Trp53R172H and KrasG12D to the mouse pancreas via the Pdx1 pancreatic progenitor cell gene promoter results in the formation of preinvasive PanIN (grades 1–3), which develops into invasive and metastatic pancreatic cancer (Supplementary Figure 2).6 Src was expressed in normal ducts within the pancreas, and this expression was maintained in PanIN lesions and PDACs (Figure 3A). The activity of Src was very low in normal ducts and early PanIN lesions; however, in late, high-grade PanIN-3 there was a marked increase in staining intensity seen with the phospho-Src Y418 antibody (Figure 3B). These high levels of phosphoSrc staining were maintained in the invasive PDAC (Figure 3B). Thus, Src activity is up-regulated during progression to an invasive PDAC in the Pdx1-Cre, Z/EGFP, LSLKrasG12D/⫹, LSL-Trp53R172H/⫹ mice.
Generation of Cell Lines From PDAC in Pdx1-Cre, Z/EGFP, LSL-KrasG12D/ⴙ, LSL-Trp53R172H/ⴙ Mice Initial studies to determine the response of PDAC to dasatinib were performed in cell lines derived from PDACs taken from the Pdx1-Cre, Z/EGFP, LSL-KrasG12D/⫹, LSL-Trp53R172H/⫹ mice. PDACs were harvested from 3 mice, and cell lines were established in culture. The histology of the tumors from which the cell lines were established is shown in Supplementary Figure 3. Pdx-1 staining and recombination of the Trp53R172H allele confirmed that the cells originated from the PDAC (Figure 4A and B). Treatment with dasatinib resulted in a dosedependent inhibition of Src kinase activity as measured by Src autophosphorylation on Y418, with concentrations between 100 and 200 nmol/L required for inhibition (Figure 4C). Treatment with dasatinib at concentrations that inhibit Src kinase activity had no effect on the proliferation of the 3 cell lines, although with higher concentrations (up to 1 mol/L) between 25% and 40% inhibition of proliferation was observed (Figure 4D). This is most likely owing to inhibition of other tyrosine kinases by dasatinib that are required for proliferation.18
Dasatinib Inhibits the Migration and Invasion of PDAC Cells We next examined the ability of dasatinib to prevent migration of the PDAC cells into a wounded area in a confluent monolayer. Dasatinib inhibited the migration of all 3 cell lines (Figure 5A) and this was observed
DASATINIB AND MOUSE PANCREATIC CANCER
297
at concentrations that inhibited Src autophosphorylation (Figure 4C). There was around 80% inhibition in the 82739 and 83320 cell lines whereas only 50% inhibition was observed in 86119 cells. Representative images of the wound healing assay are shown in Supplementary Figure 4A. Dasatinib is known to inhibit additional kinases such as Abl. Therefore, to determine whether the migration of the PDAC cells was dependent on Src, we used siRNA to specifically down-regulate Src expression in the cells. Treatment with Src siRNA sequences reduced Src expression over 72 hours (Figure 5B), which resulted in a significant inhibition of cell migration as seen with dasatinib treatment (Figure 5C). The scrambled siRNA sequences had no effect on Src expression or cell migration (Figure 5B and C). Thus, reduced Src expression is sufficient to inhibit migration of the PDAC cells. Invasion into 3-dimensional collagen gels in response to a chemotactic gradient also was examined using a Transwell invasion assay. After 72 hours cells were stained with Calcein AM and their invasion into the collagen was visualized (Supplementary Figure 4B). Quantification showed that treatment with dasatinib inhibited invasion of the PDAC cells by around 60% (Figure 5D). Src is known to phosphorylate a number of proteins that are involved in migration and invasion including FAK and p130Cas. Dasatinib treatment resulted in a dose-dependent inhibition of FAK phosphorylation on Y861 and p130Cas phosphorylation on Y249, which are both reported Src phosphorylation sites (Figure 5E). A reduction in phosphorylation of both FAK and p130Cas also was seen in Src siRNA-treated cells (Supplementary Figure 4C). The FAK-p130Cas signaling complex is known to regulate invasion via modulation of MMP production, and treatment of cells with dasatinib reduced both MMP-2 and MMP-9 production as shown by gelatin zymography (Figure 5F). Src inhibitors therefore may have potential as antimetastatic agents by preventing the migration and invasion of tumor cells. Thus, we decided to use the Pdx1-Cre, Z/EGFP, LSL-KrasG12D/⫹, LSL-Trp53R172H/⫹ mice to address whether inhibition of Src kinase activity by dasatinib could affect the spread of PDAC in vivo.
Determination of Biologically Active Dose of Dasatinib Because we had shown that higher doses of dasatinib above those required to inhibit Src kinase activity can result in additional biological effects in vitro, initial experiments were performed to determine the biologically active dose of dasatinib required to inhibit Src kinase activity in vivo. The mouse PDAC cell lines were grown as subcutaneous tumors, and the animals were treated with a range of dasatinib concentrations. Tumors were harvested and immunoblot analysis showed that treatment with 5 mg/kg dasatinib had no effect on Src kinase activity whereas activity was inhibited in tumors
BASIC–LIVER, PANCREAS, AND BILIARY TRACT
July 2010
298
MORTON ET AL
GASTROENTEROLOGY Vol. 139, No. 1
BASIC–LIVER, PANCREAS, AND BILIARY TRACT
Figure 3. Src expression and activity during development of PDAC in Pdx1-Cre, Z/EGFP, LSL-KrasG12D/⫹, LSL-Trp53R172H/⫹ mice. Immunohistochemical analysis of (A) Src and (B) phospho-Src Y418 in normal pancreatic ducts, PanINs, and PDAC. Images taken at magnifications of 40⫻ for normal ducts and 20⫻ for PanINs and PDACs.
harvested from mice treated with 10 mg/kg dasatinib (Figure 6A). Inhibition of Src activity in tumors from mice treated with 10 mg/kg was confirmed by immunohistochemistry (Figure 6B). A dose of 10 mg/kg was used as an Src kinase inhibitory dose in subsequent in vivo studies.
Dasatinib Inhibits the Development of Metastases in Pdx1-Cre, Z/EGFP, LSL-KrasG12D/ⴙ, LSL-Trp53R172H/ⴙ Mice Analysis of disease progression in the Pdx1-Cre, Z/EGFP, LSL-KrasG12D/⫹, LSL-Trp53R172H/⫹ mice showed that preneoplastic PanIN lesions develop by 6 weeks of
DASATINIB AND MOUSE PANCREATIC CANCER
299
Figure 4. Dasatinib inhibits Src activity in mouse PDAC cells. (A) Cell lines were established from 3 PDACs (82739, 83320, and 86119) harvested from Pdx1-Cre, Z/EGFP, LSL-KrasG12D/⫹, LSL-Trp53R172H/⫹ mice. Phase contrast images are shown for each (left panels). Expression of Pdx1 in the cell lines confirms their pancreatic origin (right panels). Scale bars, 0.2 m. (B) Recombination of Trp53R172H in cell lines. (C) PDAC cells were treated with a range of dasatinib concentrations for 24 hours before Western blot analysis using anti–phospho-Src Y418 and anti-Src antibodies. (D) PDAC cells were treated with a range of dasatinib concentrations for 96 hours and an MTT proliferation assay then was performed. Values are mean ⫾ standard deviation of quadruplicate wells taken from a representative experiment in a series of 3.
age (n ⬎10 mice killed) and that by 10 weeks mice have an observable mass in the pancreas as detected by wholebody in vivo green fluorescent protein (GFP) imaging. Full pathologic details are provided in Supplementary Tables 2 and 3. Histologic sections of pancreata from mice killed at this time showed multifocal and often widespread neoplastic change, with some invasive PDAC, cystic dilatation of neoplastic ducts, and varying grades of PanIN involving the majority of ducts, with little normal pancreas remaining (Supplementary Figure 5). This model therefore allows us to assess the effects of dasatinib at very early stages of tumor development and at later stages of disease progression. Cohorts of 20 Pdx1-Cre, Z/EGFP, LSL-KrasG12D/⫹, LSL-Trp53R172H/⫹ mice were treated with vehicle or dasatinib (10 mg/kg) from 6 or 10 weeks of age. Mice were allowed to age and were monitored for signs of disease and euthanized when they became symptomatic of disease progression. There was no significant difference in survival between the different
treatment groups (Kaplan–Meier, log-rank, P ⫽ .827): the median survival of vehicle-treated animals was 131 days compared with 127 days and 130 days for animals treated with dasatinib from 6 weeks and 10 weeks of age, respectively (Figure 7A). Analysis of tumor burden in the mice showed that all mice had invasive PDAC; however, the number of mice with metastases was reduced significantly in dasatinib-treated animals (Figure 7B). The incidence of metastases was 61.1% in vehicle-treated animals compared with 26.7% in mice treated with dasatinib from 6 weeks (2 test, P ⫽ .048) and 23.1% in mice treated with dasatinib from 10 weeks (2 test, P ⫽ .036). Full details are provided in Supplementary Table 4. Inhibition of Src activity in tumors from mice treated with dasatinib was confirmed by Western blotting and immunohistochemistry using the anti–phospho-Src Y418 antibody (Figure 7C and D). To assess the effect of dasatinib on the proliferation of PanINs and PDACs, mice were injected with bromodeoxyuridine 2 hours before death. Immunohistochemical
BASIC–LIVER, PANCREAS, AND BILIARY TRACT
July 2010
300
MORTON ET AL
GASTROENTEROLOGY Vol. 139, No. 1
BASIC–LIVER, PANCREAS, AND BILIARY TRACT
Figure 5. Dasatinib inhibits PDAC cell migration and invasion. (A) Migration of PDAC cells was determined in the absence or presence of dasatinib (100 or 200 nmol/L) using a wound healing assay. Values represent the wound area at the end of the experiment expressed as a percentage of the original wound area. Values are mean ⫾ standard deviation of 3 areas of the wound and are taken from a representative experiment in a series of 3. (B) Western blot analysis of Src expression in PDAC cells treated with scrambled or Src siRNA sequences. -tubulin is used as a loading control. (C) Migration of PDAC cells treated with scrambled or Src siRNA sequences calculated as in panel A. (D) Invasion of PDAC cells into collagen gels was determined in the absence or presence of 100 nmol/L dasatinib. Quantification of fold invasion greater than 20 m relative to control (untreated cells) is presented from at least 3 independent experiments. Values are mean ⫾ standard error. (E) PDAC cells were treated with a range of dasatinib concentrations for 24 hours before Western blot analysis using anti–phospho-FAK Y861 and anti–phospho-p130Cas Y249 antibodies. -tubulin is used as a loading control. (F) MMP gelatin zymography was performed on supernatants of untreated and dasatinib-treated (100 nmol/L) cells.
analysis of bromodeoxyuridine staining was performed, and a number of bromodeoxyuridine-positive nuclei were scored. In both PanIN and PDAC there was no significant difference in proliferation between the vehicle- and dasatinib-treated animals (Figure 7E and F). These observations are consistent with continued growth of the primary tumor within the pancreas in mice treated with dasatinib and the lack of survival advantage in these mice, which develop symptoms as a result of primary tumor burden.
Discussion Most early phase clinical trials are performed in patients with advanced refractory disease, and the con-
ventional paradigm for efficacy is dependent on showing a reduction in tumor dimensions using anatomic imaging. However, this approach, and this patient population, is not likely to be appropriate for agents that inhibit cancer cell invasion, migration, and the development of metastases. This is particularly relevant for patients with PDAC, in whom cytotoxic chemotherapy agents have limited efficacy in advanced, inoperable, disease. The mainstay of preclinical in vivo evaluation of novel cancer therapeutics has been performed using human cancer cell lines grown as xenografts in immune-compromised mice but these have been suboptimal in predicting efficacy in subsequent clinical trials in human beings with pancreatic cancer. One drawback with this approach
July 2010
DASATINIB AND MOUSE PANCREATIC CANCER
301
is that the tumor microenvironment and inflammatory response, which play pivotal roles in pancreatic cancer development, are disrupted.19,20 Furthermore, these mouse models are not applicable to evaluate antimetastatic agents, which may not reduce primary tumor bulk but inhibit spread of the disease. These issues can, in part, be overcome by the use of orthotopic models in which injection of human pancreatic tumor cells directly into the mouse pancreas results in metastasis to the liver and lymph node, although this approach still relies on the use of immune-compromised animals. The use of appropriate genetically engineered mouse models of cancer may overcome these problems and can be an alternative tool for preclinical evaluation of cancer therapeutics.21 In this study we have used dasatinib to determine the effects of inhibiting Src kinase activity as an antimetastatic therapeutic strategy in vivo using the Pdx1-Cre, Z/EGFP, LSL-KrasG12D/⫹, LSL-Trp53R172H/⫹ mouse model. Many studies have shown that interference with Src activity inhibits the migration and invasion of epithelial tumor cells, which may contribute to the potential antimetastatic activity of inhibitors such as dasatinib.8 This has been attributed to the well-documented role of Src in regulating integrin adhesions, cadherin-mediated cell– cell adhesions, and MMP expression, all of which influence the ability of cells to break away from the primary tumor mass and colonize at distant sites in the body. Src-dependent phosphorylation of FAK and p130Cas regulates cell motility and invasion, and our data suggest that disruption of the Src-FAK-p130Cas-DOCK180 complex, which controls MMP production, contributes to the effects of dasatinib in PDAC.22 Evidence from human pancreatic cell lines suggests that Src inhibition also blocks tumor angiogenesis, which may contribute to the reduction in metastases seen in orthotopic models of pancreatic cancer.23,24 Here we have shown that inhibition of Src kinase activity by dasatinib can inhibit the development of metastases in vivo. Interestingly, the inhibition of develop-
ment of metastases was similar when dasatinib administration was commenced at either 6 weeks (when mice have preneoplastic PanIN) or 10 weeks (when mice have more advanced disease but not overt metastases). All of the mice in whom dosing commenced at 6 weeks of age ultimately developed invasive PDAC, indicating that dasatinib does not inhibit the conversion of PanINs to invasive PDAC, but has effects at later stages of disease progression. This is potentially significant in the human population in that there is currently no phenotype of PanIN in human beings that is clinically detectable. However, we speculate that administration of dasatinib after potentially curative resection of localized disease could reduce the risk of metastases and improve overall survival. Dasatinib did not improve survival when administered to the Pdx1-Cre, Z/EGFP, LSL-KrasG12D/⫹, LSL-Trp53R172H/⫹ mice. This is most likely owing to the continued growth of the primary tumor because the mice primarily die of tumor burden within the pancreas. The inability of dasatinib to inhibit PDAC proliferation is in agreement with reports in other tumor types that indicate that Src kinase activity is not required for the sustained growth of several types of epithelial tumors.18,25 Interestingly, in contrast to our findings, dasatinib inhibited primary tumor growth in an orthotopic model of pancreatic cancer, which also contributed to the reduction in metastatic burden.24 It may be that in this model different factors are controlling growth of the primary tumor, and it is possible that higher (ie, antiproliferative) doses of dasatinib, which inhibit other pathways in addition to Src kinase activity, also could have inhibited proliferation in our mouse model. In conclusion, we have shown that Src kinase activity increases with progression from early PanIN through to late PanIN and invasive PDAC in a genetically engineered mouse model that recapitulates the human disease. Indeed, we show that the levels of Src and phospho-Src are important indicators of vascular invasion, lymph node
BASIC–LIVER, PANCREAS, AND BILIARY TRACT
Figure 6. Determination of biologically active dose of dasatinib in mice. PDAC cells were grown as subcutaneous tumors, and the mice were treated with dasatinib at 5 mg/kg or 10 mg/kg. After 2 hours the mice were killed and tumors were collected. (A) Western blot analysis using anti–phosphoSrc Y418 and anti-Src antibodies. (B) Immunohistochemical analysis of tumors stained with anti–phospho-Src Y418 antibody from vehicle and dasatinib-treated (10 mg/kg) animals. Images taken at 20⫻ magnification. Representative results are shown for the 83320 cell line. Similar results were obtained for the 86119 and 82739 cells.
302
MORTON ET AL
GASTROENTEROLOGY Vol. 139, No. 1
BASIC–LIVER, PANCREAS, AND BILIARY TRACT
Figure 7. Dasatinib does not increase survival but reduces the incidence of metastasis in Pdx1-Cre, Z/EGFP, LSL-KrasG12D/⫹, LSL-Trp53R172H/⫹ mice. (A) Kaplan–Meier survival curves of mice treated with vehicle or dasatinib (10 mg/kg) from 6 or 10 weeks of age. (B) The number of mice with metastases was counted at the time of death, and confirmed by histology, and results are presented as the percentage of the total number of mice in each cohort. Black columns represent mice with metastases and grey columns represent those without metastases. 2 test: 6-week cohort vs vehicle, P ⫽ .048 and 10-week cohort vs vehicle, P ⫽ .036. (C) Lysates were prepared from tumors harvested from vehicle or dasatinib-treated (10 mg/kg) mice, and Western blot analysis was performed with anti–phospho-Src Y418 and anti-Src antibodies. (D) Immunohistochemical analysis of tumors harvested from vehicle and dasatinib-treated animals, stained with anti–phospho-Src Y418 antibody. Images taken at 20⫻ magnification. In each case representative examples are shown from 2 individual tumors. (E and F) Cell proliferation was assessed by labeling cells for 2 hours with bromodeoxyuridine (BrdU) before killing the mice. Immunohistochemical analysis of BrdU was performed, and the number of positive nuclei was determined. (E) Ten PanINs were scored from 4 mice in each treatment group (n ⫽ 40; Mann–Whitney U test, P ⫽ .665). (F) PDAC. Ten high-powered frames were scored from 4 mice in each treatment group (n ⫽ 40; Mann–Whitney U test, P ⫽ .557).
positivity, and prognosis in human PDAC. We also have shown that inhibition of Src kinase activity by dasatinib can inhibit PDAC cell invasion and migration in vitro, and can inhibit the development of metastases in vivo. We propose that dasatinib should be evaluated further in vivo, either as monotherapy after resection of localized invasive PDAC or in combination with gemcitabine for invasive, nonmetastatic, irresectable PDAC, with the ultimate aim of translating these observations through to clinical evaluation in patients with pancreatic cancer.
Supplementary Material Note: To access the supplementary material accompanying this article, visit the online version of Gastroenterology at www.gastrojournal.org, and at doi: 10.1053/j.gastro.2010.03.034. References 1. Li D, Xie K, Wolff R, et al. Pancreatic cancer. Lancet 2004;363: 1049 –1057.
July 2010
19.
20.
21.
22. 23.
24.
25.
303
in colon cancer cells following treatment with dasatinib. Mol Cancer Ther 2006;5:3014 –3022. Farrow B, Albo D, Berger DH. The role of the tumor microenvironment in the progression of pancreatic cancer. J Surg Res 2008; 149:319 –328. Mahadevan D, Von Hoff DD. Tumor-stroma interactions in pancreatic ductal adenocarcinoma. Mol Cancer Ther 2007;6:1186 – 1197. Olive KP, Jacobetz MA, Davidson CJ, et al. Inhibition of Hedgehog signaling enhances delivery of chemotherapy in a mouse model of pancreatic cancer. Science 2009;324:1457–1461. Hsia DA, Mitra SK, Hauck CR, et al. Differential regulation of cell motility and invasion by FAK. J Cell Biol 2003;160:753–767. Ischenko I, Guba M, Yezhelyev M, et al. Effect of Src kinase inhibition on metastasis and tumor angiogenesis in human pancreatic cancer. Angiogenesis 2007;10:167–182. Trevino JG, Summy JM, Lesslie DP, et al. Inhibition of SRC expression and activity inhibits tumor progression and metastasis of human pancreatic adenocarcinoma cells in an orthotopic nude mouse model. Am J Pathol 2006;168:962–972. Buettner R, Mesa T, Vultur A, et al. Inhibition of Src family kinases with dasatinib blocks migration and invasion of human melanoma cells. Mol Cancer Res 2008;6:1766 –1774.
Received June 30, 2009. Accepted March 8, 2010. Reprint requests Address requests for reprints to: Valerie Brunton, PhD, Edinburgh Cancer Research Centre, Edinburgh University, Crewe Road South, Edinburgh, EH4 2XR, United Kingdom. e-mail: [email protected]; fax: (44) 131-777-3520. Acknowledgments The authors are grateful for support from Colin Nixon and colleagues in Histology services; and from Stephen Bell, Derek Miller, and Tom Hamilton in Biological Research Services at the Beatson Institute of Cancer Research; Jane Hair from NHSGGC Biorepository; and Think Pink for the purchase of the slide scanner. Conflicts of interest These authors disclose the following: Margaret Frame and T. R. Jeffry Evans have received research funding and honoraria from Bristol-Myers Squibb, and have performed consultancy work for Bristol-Myers Squibb. The remaining authors disclose no conflicts. Funding This work was funded by Cancer Research UK grants C2193/A7603 and C157/A9148 and Chief Science Officer, Scottish Executive (N.J.).
BASIC–LIVER, PANCREAS, AND BILIARY TRACT
2. Yeo CJ, Cameron JL, Lillemoe KD, et al. Pancreaticoduodenectomy with or without distal gastrectomy and extended retroperitoneal lymphadenectomy for periampullary adenocarcinoma, part 2: randomized controlled trial evaluating survival, morbidity, and mortality. Ann Surg 2002;236:355–368. 3. Hezel AF, Kimmelman AC, Stanger BZ, et al. Genetics and biology of pancreatic ductal adenocarcinoma. Genes Dev 2006;20: 1218 –1249. 4. Hruban RH, Goggins M, Parsons J, et al. Progression model for pancreatic cancer. Clin Cancer Res 2000;6:2969 –2972. 5. Almoguera C, Shibata D, Forrester K, et al. Most human carcinomas of the exocrine pancreas contain mutant c-K-ras genes. Cell 1988;53:549 –554. 6. Hingorani SR, Wang L, Multani AS, et al. Trp53R172H and KrasG12D cooperate to promote chromosomal instability and widely metastatic pancreatic ductal adenocarcinoma in mice. Cancer Cell 2005;7:469 – 483. 7. Olive KP, Tuveson DA. The use of targeted mouse models for preclinical testing of novel cancer therapeutics. Clin Cancer Res 2006;12:5277–5287. 8. Brunton VG, Frame MC. Src and focal adhesion kinase as therapeutic targets in cancer. Curr Opin Pharmacol 2008;8:427– 432. 9. Frame MC. Src in cancer: deregulation and consequences for cell behaviour. Biochim Biophys Acta 2002;1602:114 –130. 10. Hakam A, Fang Q, Karl R, et al. Coexpression of IGF-1R and c-Src proteins in human pancreatic ductal adenocarcinoma. Dig Dis Sci 2003;48:1972–1978. 11. Lutz MP, Esser IB, Flossmann-Kast BB, et al. Overexpression and activation of the tyrosine kinase Src in human pancreatic carcinoma. Biochem Biophys Res Commun 1998;243:503–508. 12. Demetri GD, Lo Russo P, MacPherson IR, et al. Phase I doseescalation and pharmacokinetic study of dasatinib in patients with advanced solid tumors. Clin Cancer Res 2009;15:6232– 6240. 13. Khoury HJ, Guilhot F, Hughes TP, et al. Dasatinib treatment for Philadelphia chromosome-positive leukemias: practical considerations. Cancer 2009;115:1381–1394. 14. Jackson EL, Willis N, Mercer K, et al. Analysis of lung tumor initiation and progression using conditional expression of oncogenic K-ras. Genes Dev 2001;15:3243–3248. 15. Olive KP, Tuveson DA, Ruhe ZC, et al. Mutant p53 gain of function in two mouse models of Li-Fraumeni syndrome. Cell 2004;119: 847– 860. 16. Gu G, Dubauskaite J, Melton DA. Direct evidence for the pancreatic lineage: NGN3⫹ cells are islet progenitors and are distinct from duct progenitors. Development 2002;129:2447–2457. 17. Novak A, Guo C, Yang W, et al. Z/EG, a double reporter mouse line that expresses enhanced green fluorescent protein upon Cre-mediated excision. Genesis 2000;28:147–155. 18. Serrels A, Macpherson IR, Evans TR, et al. Identification of potential biomarkers for measuring inhibition of Src kinase activity
DASATINIB AND MOUSE PANCREATIC CANCER
Dasatinib Inhibits the Development of Metastases in a Mouse Model of Pancreatic Ductal Adenocarcinoma JENNIFER P. MORTON,*,‡ SAADIA A. KARIM,* KATHRYN GRAHAM,* PAUL TIMPSON,* NIGEL JAMIESON,‡ DIMITRIS ATHINEOS,*,‡ BRENDAN DOYLE,* COLIN MCKAY,§ MAN–YEUNG HEUNG,储 KARIN A. OIEN,‡ MARGARET C. FRAME,储 T. R. JEFFRY EVANS,*,‡ OWEN J. SANSOM,* and VALERIE G. BRUNTON储 *Beatson Institute for Cancer Research, Glasgow, United Kingdom; ‡Centre for Oncology and Applied Pharmacology, Division of Cancer Sciences and Molecular Pathology, University of Glasgow, Glasgow, United Kingdom; §West of Scotland Pancreatic Unit, Glasgow Royal Infirmary, Glasgow, United Kingdom; and 储Edinburgh Cancer Research Centre, University of Edinburgh, Edinburgh, United Kingdom
BASIC–LIVER, PANCREAS, AND BILIARY TRACT
BACKGROUND & AIMS: Pancreatic ductal adenocarcinoma (PDAC) is a highly invasive and metastatic disease for which conventional treatments are of limited efficacy. A number of agents in development are potential anti-invasive and antimetastatic agents, including the Src kinase inhibitor dasatinib. The aim of this study was to assess the importance of Src in human PDAC and to use a genetically engineered mouse model of PDAC to determine the effects of dasatinib on PDAC progression. METHODS: Src expression and activity was measured by immunohistochemistry in 114 human PDACs. Targeting expression of Trp53R172H and KrasG12D to the mouse pancreas results in the formation of invasive and metastatic PDAC. These mice were treated with dasatinib, and disease progression monitored. Cell lines were derived from mouse PDACs, and in vitro effects of dasatinib assessed. RESULTS: Src expression and activity were up-regulated in human PDAC and this correlated with reduced survival. Dasatinib inhibited the migration and invasion of PDAC cell lines, although no effects on proliferation were seen at concentrations that inhibited Src kinase activity. In addition, dasatinib significantly inhibited the development of metastases in Pdx1-Cre, Z/EGFP, LSL-KrasG12D/⫹, LSL-Trp53R172H/⫹ mice. However, there was no survival advantage in the dasatinib-treated animals owing to continued growth of the primary tumor. CONCLUSIONS: This study confirms the importance of Src in human PDAC and shows the usefulness of a genetically engineered mouse model of PDAC for assessing the activity of potential antimetastatic agents and suggests that dasatinib should be evaluated further as monotherapy after resection of localized invasive PDAC. Keywords: Pancreatic Cancer; Metastasis; Src Kinase; Dasatinib.
I
nfiltrating pancreatic ductal adenocarcinoma (PDAC) is the fifth most common cancer and the fourth most common cause of cancer deaths in the United Kingdom. Aggressive invasion and early metastases are characteristic of the disease, such that 90% of patients have surgi-
cally unresectable disease at the time of diagnosis. Furthermore, most systemic therapies are largely ineffective in advanced, inoperable disease, and the estimated 5-year survival rate is less than 5%.1 In addition, a majority of the selected patients who undergo potentially curative resection for small, localized lesions inevitably develop recurrent or metastatic disease,2 presumably because of the presence of distant micrometastases at initial diagnosis. Consequently, the development of more effective strategies to treat preinvasive pancreatic cancer, or micrometastatic disease, is of paramount importance. In human PDAC, development, progression, and metastases arise via the accumulation of multiple genetic and epigenetic changes, including inactivation of tumorsuppressor genes and activation or overexpression of proto-oncogenes.3,4 PDAC arises from preinvasive lesions called pancreatic intraepithelial neoplasms (PanINs) whose progression is driven by activating mutations in Kras, which are detected in almost 100% of advanced PDACs.5 In a recently developed mouse model of PDAC, KrasG12D and Trp53R172H are expressed in pancreatic cells, resulting in the formation of PanINs that develop into invasive and widely metastatic PDAC.6 The genetics of the mouse model mimics the human disease and importantly the histopathology, disease progression, and sites of metastases in the mice also recapitulate many aspects of the human disease, providing a good model with which to study potential therapies for PDAC.7 Currently, there is considerable interest in exploiting Src kinase as a novel therapeutic target in malignant disease, and in particular whether such an agent may be useful in inhibiting the development of metastatic disease.8 Src is a nonreceptor tyrosine kinase whose expression is increased in a number of epithelial tumors, most notably in colon and breast,9 and 2 studies have reported Abbreviations used in this paper: CI, confidence interval; FAK, focal adhesion kinase; IQR, interquartile range; LNR, lymph node ratio; MMP, matrix metalloproteinase; PanIN, pancreatic intraepithelial neoplasia; PDAC, pancreatic ductal adenocarcinoma. © 2010 by the AGA Institute 0016-5085/$36.00 doi:10.1053/j.gastro.2010.03.034
overexpression of Src in PDAC.10,11 Src is involved in many aspects of tumor cell behavior that impact on their metastatic capacity such as survival, adhesion, migration, and invasion.9 A number of small-molecule Src kinase inhibitors are now in clinical development including dasatinib, a highly potent, adenosine triphosphate– competitive inhibitor of Src family and Abl kinases12,13 and we have used dasatinib to determine how best to evaluate a potential antimetastatic agent in the clinical management of pancreatic cancer. We have shown that dasatinib, by inhibiting Src kinase activity, inhibits the development of metastases in the Pdx1-Cre, Z/EGFP, LSL-KrasG12D/⫹, LSL-Trp53R172H/⫹ mouse model of PDAC but has no effect on the proliferation of the primary tumor, supporting the hypothesis that Src kinase inhibitors may function primarily as antimetastatic agents. It remains to be determined if dasatinib could improve overall survival when used in the adjuvant (postoperative) setting after removal of the primary tumor burden.
Material and Methods Genetically Modified Mice and Animal Care Conditional LSL-KrasG12D/⫹ mice (strain 01XJ6, mouse models of human cancer consortium [MMHCC], NCI-Frederick, Frederick, MD)14 were mated to LSLTrp53R172H/⫹ mice (strain 01XL9, MMHCC, NCIFrederick),15 and Pdx1-Cre mice16 were mated to Z/EGFP mice.17 Progeny from these crosses then were interbred to obtain Pdx1-Cre, Z/EGFP, LSL-KrasG12D/⫹, LSL-Trp53R172H/⫹ mice on a mixed background. Mice were genotyped by polymerase chain reaction analysis as described previously.6 Experiments were performed in compliance with UK Home Office guidelines. Mice were dosed daily by oral gavage with 10 mg/kg dasatinib (Bristol-Myers Squibb, Princeton, NJ) in 80 mmol/L citrate buffer. Tumor and metastatic burden was assessed by gross pathology and histology and organs/tumors were removed and fixed in 10% buffered formalin. For histologic detection of hepatic metastases, serial sections of the whole liver were analyzed. The Olympus OV100 Whole Mouse Imaging System (Olympus Corp, Tokyo, Japan) was used for imaging in live mice. Anesthesia was induced and maintained with a mixture of isoflurane and oxygen.
Immunohistochemistry and Immunocytochemistry Immunohistochemistry of formalin-fixed, paraffin-embedded tissues was performed as described previously.18 Primary antibodies used were anti–phospho-Src Y418 1:200 and anti-Src 1:800 (both Cell Signaling, New England Biolabs, Hitchin, UK). For immunocytochemistry, cells were fixed in 10% neutral buffered formalin and permeabilized with 0.1% Triton X-100 before staining with an anti-Pdx1 antibody (Abcam, Cambridge, UK).
DASATINIB AND MOUSE PANCREATIC CANCER
293
Cell Culture and Dasatinib Treatment PDAC cell lines were generated from primary pancreatic tumors taken from Pdx1-Cre, Z/EGFP, LSLKrasG12D/⫹, LSL-Trp53R172H/⫹ mice and then passaged in growth media (Dulbecco’s modified Eagle medium containing 10% fetal bovine serum and 2 mmol/L L-glutamine). Dasatinib was prepared as a 100 mmol/L stock in dimethyl sulfoxide and diluted in growth media. Recombination of Trp53R172H in the cell lines was confirmed by polymerase chain reaction using forward: AGCCTGCCTAGCTTCCTCAGG, reverse: CTTGGAGACATAGCCACACTG primers.
Western Blot Analysis Western blot analysis was performed as described previously18 using anti–phospho-Src Y418, phosphop130Cas Y249 (Cell Signalling), phospho-focal adhesion kinase (FAK) Y861 (Becton Dickinson, Oxford, UK), or anti-Src (Cancer Research UK, London, UK) primary antibodies all at 1:1000.
Proliferation Assay A 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) proliferation assay was performed as described previously.18 Mean values were calculated from quadruplicate wells and plotted as the mean percentage of the untreated controls. The highest concentration of dimethyl sulfoxide (drug diluent) added to the cells had no effect on cell proliferation (data not shown).
Migration Assay Confluent monolayers of PDAC cells were scored with a fine pipette tip to produce a wound. Migration into the wound was monitored by time-lapse video microscopy over 48 hours in the presence or absence of dasatinib using 20⫻ magnification on a Zeiss Axiovert S100 microscope (Zeiss, Welwyn Garden City, UT) with AQM Advance Software (Kinetic Imaging, Nottingham, UK). Three representative areas were scored for each treatment and the area of the wound was calculated using ImageJ software (National Institutes of Health, Bethesda, MD). Cells also were transfected with 2 mol/L ON-TARGETplus src SMARTpool small interfering (si) RNA solution or a 2 mol/L ON-TARGETplus nontargeting siRNA solution according to the manufacturer’s protocol (Dharmacon, Fisher Scientific, Loughborough, UK), and after 24 hours, migration was monitored as described earlier.
Invasion Assay Cells were seeded on the bottom of Transwell inserts (Corning, Fisher Scientific, Loughborough, UK) containing polymerized collagen type I (Becton Dickinson). Transwell inserts then were placed in serum-free medium, and medium supplemented with 10% fetal calf
BASIC–LIVER, PANCREAS, AND BILIARY TRACT
July 2010
294
MORTON ET AL
serum and 50 ng/mL hepatocyte growth factor (R&D Systems, Abingdon, UK) was placed on top of the matrix in the absence or presence of dasatinib. Five days after seeding, invading cells were stained with Calcein-AM (Invitrogen, Paisley, UK) and visualized by confocal microscopy (Leica SP2, Leica, Milton Keynes, UK). Serial optical sections were captured at 10-m intervals and quantified using ImageJ software using the area analysis module. Invasion was calculated as cells that had moved more than 20 m into the collagen.
Matrix Metalloproteinase Zymography Matrix metalloproteinase (MMP) activity was assessed from culture supernatants of PDAC cells. Cells were seeded onto collagen-coated dishes and allowed to adhere, were then washed, and serum free-media was added. After 24 hours in the presence or absence of dasatinib (100 nmol/L), culture supernatants were collected and equal volumes were subjected to gelatin zymography (Invitrogen) under nonreducing conditions according to the manufacturer’s instructions.
Subcutaneous Tumor Growth in Mice Mouse PDAC cell lines were injected subcutaneously into the right flank of 4-week-old to 6-week-old female CD-1 nude mice (Charles Rivers, Harlan, UK). When tumors were established dasatinib was administered by oral gavage. The mice were killed 2 hours later, and the tumors were formalin-fixed for paraffin-embedding or snap-frozen in liquid nitrogen.
Tissue Microarray BASIC–LIVER, PANCREAS, AND BILIARY TRACT
Src and phospho-Src expression levels in a human pancreaticobiliary tissue microarray were scored based on staining intensity and area of tumor cells using a weighted histoscore calculated from the sum of (1 ⫻ % weak staining) ⫹ (2 ⫻ % moderate staining) ⫹ (3 ⫻ % strong staining), providing a semiquantitative classification of staining intensity. Representative images are shown in Figure 1 and further details of the analysis and array are provided in the Supplementary Material section.
Results Src Expression and Activity in Human PDAC We looked at Src expression and also activity using an antibody against the autophosphorylation site of Src (tyrosine 418) in a tumor array of 114 human PDACs (Figure 1). Src was expressed in all tumors, with similarly high levels of the activated protein noted (94.1%). The median histoscore for membranous Src expression was 90.8 (interquartile range [IQR], 63.8 –137.1) whereas the median membranous phospho-Src expression was 80.0 (IQR, 38.3–112.9). Src expression was found in 84.7% of normal ducts (89 of 105) whereas phosphorylated Src was present in 82.6% of normal ducts (92 of 108). However, the levels of expression and activity
GASTROENTEROLOGY Vol. 139, No. 1
were very low in normal ducts, and when expression in the tumor and normal ducts from the same patient was compared we found that in 74.7% of cases there was an increase in Src expression in the tumors, whereas 60.1% of tumors had increased activated Src compared with the normal duct from the corresponding patient according to the highest scoring regions.
Src Expression With Clinicopathologic and Survival Analyses Low tumor grade was associated with significantly reduced expression of Src (median histoscore, 83.3; IQR, 61.7–122.5) compared with high-grade tumors (median histoscore, 112.7; IQR, 73.3–143.1; P ⬍ .039) (Figure 2A). Increasing lymph node ratio (LNR: number of involved nodes/total number of resected nodes) was associated with greater Src expression with a median histoscore of 105.1 (IQR, 87.1–163.1) for node-negative tumors (n ⫽ 24) compared with a median histoscore of 103 (IQR, 79.2–155.8) for tumors with an LNR of 0.1 or less (n ⫽ 52); 121.1 (IQR, 102.5–212.4) for an LNR of 0.1– 0.5 (n ⫽ 13), and 153 (IQR, 113.5–222.9) for an LNR of 0.5 or greater (n ⫽ 30). Src expression was significantly greater in tumors with an LNR of greater than 0.1 compared with an LNR of 0.1 or less or with no lymph node involvement (P ⬍ .009) (Figure 2B). The presence of vascular invasion within the tumor was associated with increased Src expression (median histoscore, 105.2; IQR, 75.4 –145.4) compared with tumors with no vascular invasion (median histoscore, 74.6; IQR, 59.2–137.1; P ⬍ .03) (Figure 2C). There was no association with tumor stage, tumor size, or resection margin status. Full details are provided in Supplementary Table 1. Univariate outcome assessment revealed a median overall survival time of 15.4 months (95% confidence interval [CI], 7.9 –22.9 mo) for patients with high-Srcexpressing tumors (median histoscore, ⬎108); compared with 17.9 months (95% CI, 13.4 –22.3 mo) for low-Srcexpressing tumors (median histoscore, ⬍108); however, this failed to reach significance (P ⫽ .121). Combining tumor grade with Src expression identified a particularly poor prognostic group, with high tumor grade and high Src expression (n ⫽ 22) associated with a median survival of 7.6 months (95% CI, 2.55–12.8 mo) after pancreaticoduodenectomy compared with 16.8 months (95% CI, 0.5–38.4 mo) for high-grade and low-Src-expressing tumors (log-rank, P ⬍ .003) (Figure 2D and Supplementary Figure 1).
Activated Src Expression With Clinicopathologic and Survival Analyses Phospho-Src Y418 membranous expression was not associated with clinicopathologic factors including tumor stage, lymph node invasion, tumor grade, vascular invasion, perineural invasion, or resection margin status. Univariate survival analysis was performed using the me-
DASATINIB AND MOUSE PANCREATIC CANCER
295
BASIC–LIVER, PANCREAS, AND BILIARY TRACT
July 2010
Figure 1. Src expression and activity in human ductal adenocarcinoma. Immunohistochemical analysis of Src (left panels) and phospho-Src Y418 (right panels) in a tumor array of human PDAC. Representative images are shown for absence of staining, ⫺; weak staining, ⫹; moderate staining, ⫹⫹; and strong staining, ⫹⫹⫹. Original magnification, 10⫻.
dian membranous expression value of 80 as a cut-off; however, there was no difference in outcome between those patients with tumors showing low or high phospho-Src expression (14.8 mo; 95% CI, 11.6 –17.9 mo;
compared with 19.6 mo; 95% CI, 16.2–23.1 mo) (log-rank, P ⫽ .621). We previously observed that there is a poor correlation between the levels of phospho-Src and Src expression in
296
MORTON ET AL
GASTROENTEROLOGY Vol. 139, No. 1
BASIC–LIVER, PANCREAS, AND BILIARY TRACT
Figure 2. High Src expression and activity in human PDAC correlates with reduced survival. (A) Box plot of Src histoscore vs tumor grade. (B) Box plot of Src histoscore vs LNR. (C) Box plot of Src histoscore vs vascular invasion. (D) Kaplan–Meier analyses for high-grade patients (n ⫽ 34) showing that high Src expression is associated with poorer outcome. (E) Kaplan–Meier analyses for resection margin–positive patients (n ⫽ 90) showing that cases with evidence of grade 3 phospho-Src expression have poorer outcome after pancreaticoduodenectomy compared with those with grade 1/2 phospho-Src expression.
mouse PDAC, with phospho-Src expression restricted to more invasive areas of the tumor. With this in mind, we performed survival analysis on patients with tumors where hotspots of phospho-Src expression were identified. Based on intensity of expression we identified a group of 23 patients with evidence of grade 3 phospho-
Src expression. Although this increased level of expression did not yield prognostic significance with all patients included, subgroup analysis of the 90 resection margin–positive patients did show that tumors with regions of very high phospho-Src expression had a significantly poorer outcome after pancreaticoduodenectomy
(median overall survival, 10.3 mo; 95% CI, 5.99 –14.5 mo) compared with tumors with no evidence of grade 3 phospho-Src expression (median overall survival, 15.6 mo; 95% CI, 12.7–18.5 mo) (log-rank, P ⫽ .038) (Figure 2E).
Src Activity and Expression in Pdx1-Cre, Z/EGFP, LSL-KrasG12D/ⴙ, LSL-Trp53R172H/ⴙ Mice Targeting endogenous expression of Trp53R172H and KrasG12D to the mouse pancreas via the Pdx1 pancreatic progenitor cell gene promoter results in the formation of preinvasive PanIN (grades 1–3), which develops into invasive and metastatic pancreatic cancer (Supplementary Figure 2).6 Src was expressed in normal ducts within the pancreas, and this expression was maintained in PanIN lesions and PDACs (Figure 3A). The activity of Src was very low in normal ducts and early PanIN lesions; however, in late, high-grade PanIN-3 there was a marked increase in staining intensity seen with the phospho-Src Y418 antibody (Figure 3B). These high levels of phosphoSrc staining were maintained in the invasive PDAC (Figure 3B). Thus, Src activity is up-regulated during progression to an invasive PDAC in the Pdx1-Cre, Z/EGFP, LSLKrasG12D/⫹, LSL-Trp53R172H/⫹ mice.
Generation of Cell Lines From PDAC in Pdx1-Cre, Z/EGFP, LSL-KrasG12D/ⴙ, LSL-Trp53R172H/ⴙ Mice Initial studies to determine the response of PDAC to dasatinib were performed in cell lines derived from PDACs taken from the Pdx1-Cre, Z/EGFP, LSL-KrasG12D/⫹, LSL-Trp53R172H/⫹ mice. PDACs were harvested from 3 mice, and cell lines were established in culture. The histology of the tumors from which the cell lines were established is shown in Supplementary Figure 3. Pdx-1 staining and recombination of the Trp53R172H allele confirmed that the cells originated from the PDAC (Figure 4A and B). Treatment with dasatinib resulted in a dosedependent inhibition of Src kinase activity as measured by Src autophosphorylation on Y418, with concentrations between 100 and 200 nmol/L required for inhibition (Figure 4C). Treatment with dasatinib at concentrations that inhibit Src kinase activity had no effect on the proliferation of the 3 cell lines, although with higher concentrations (up to 1 mol/L) between 25% and 40% inhibition of proliferation was observed (Figure 4D). This is most likely owing to inhibition of other tyrosine kinases by dasatinib that are required for proliferation.18
Dasatinib Inhibits the Migration and Invasion of PDAC Cells We next examined the ability of dasatinib to prevent migration of the PDAC cells into a wounded area in a confluent monolayer. Dasatinib inhibited the migration of all 3 cell lines (Figure 5A) and this was observed
DASATINIB AND MOUSE PANCREATIC CANCER
297
at concentrations that inhibited Src autophosphorylation (Figure 4C). There was around 80% inhibition in the 82739 and 83320 cell lines whereas only 50% inhibition was observed in 86119 cells. Representative images of the wound healing assay are shown in Supplementary Figure 4A. Dasatinib is known to inhibit additional kinases such as Abl. Therefore, to determine whether the migration of the PDAC cells was dependent on Src, we used siRNA to specifically down-regulate Src expression in the cells. Treatment with Src siRNA sequences reduced Src expression over 72 hours (Figure 5B), which resulted in a significant inhibition of cell migration as seen with dasatinib treatment (Figure 5C). The scrambled siRNA sequences had no effect on Src expression or cell migration (Figure 5B and C). Thus, reduced Src expression is sufficient to inhibit migration of the PDAC cells. Invasion into 3-dimensional collagen gels in response to a chemotactic gradient also was examined using a Transwell invasion assay. After 72 hours cells were stained with Calcein AM and their invasion into the collagen was visualized (Supplementary Figure 4B). Quantification showed that treatment with dasatinib inhibited invasion of the PDAC cells by around 60% (Figure 5D). Src is known to phosphorylate a number of proteins that are involved in migration and invasion including FAK and p130Cas. Dasatinib treatment resulted in a dose-dependent inhibition of FAK phosphorylation on Y861 and p130Cas phosphorylation on Y249, which are both reported Src phosphorylation sites (Figure 5E). A reduction in phosphorylation of both FAK and p130Cas also was seen in Src siRNA-treated cells (Supplementary Figure 4C). The FAK-p130Cas signaling complex is known to regulate invasion via modulation of MMP production, and treatment of cells with dasatinib reduced both MMP-2 and MMP-9 production as shown by gelatin zymography (Figure 5F). Src inhibitors therefore may have potential as antimetastatic agents by preventing the migration and invasion of tumor cells. Thus, we decided to use the Pdx1-Cre, Z/EGFP, LSL-KrasG12D/⫹, LSL-Trp53R172H/⫹ mice to address whether inhibition of Src kinase activity by dasatinib could affect the spread of PDAC in vivo.
Determination of Biologically Active Dose of Dasatinib Because we had shown that higher doses of dasatinib above those required to inhibit Src kinase activity can result in additional biological effects in vitro, initial experiments were performed to determine the biologically active dose of dasatinib required to inhibit Src kinase activity in vivo. The mouse PDAC cell lines were grown as subcutaneous tumors, and the animals were treated with a range of dasatinib concentrations. Tumors were harvested and immunoblot analysis showed that treatment with 5 mg/kg dasatinib had no effect on Src kinase activity whereas activity was inhibited in tumors
BASIC–LIVER, PANCREAS, AND BILIARY TRACT
July 2010
298
MORTON ET AL
GASTROENTEROLOGY Vol. 139, No. 1
BASIC–LIVER, PANCREAS, AND BILIARY TRACT
Figure 3. Src expression and activity during development of PDAC in Pdx1-Cre, Z/EGFP, LSL-KrasG12D/⫹, LSL-Trp53R172H/⫹ mice. Immunohistochemical analysis of (A) Src and (B) phospho-Src Y418 in normal pancreatic ducts, PanINs, and PDAC. Images taken at magnifications of 40⫻ for normal ducts and 20⫻ for PanINs and PDACs.
harvested from mice treated with 10 mg/kg dasatinib (Figure 6A). Inhibition of Src activity in tumors from mice treated with 10 mg/kg was confirmed by immunohistochemistry (Figure 6B). A dose of 10 mg/kg was used as an Src kinase inhibitory dose in subsequent in vivo studies.
Dasatinib Inhibits the Development of Metastases in Pdx1-Cre, Z/EGFP, LSL-KrasG12D/ⴙ, LSL-Trp53R172H/ⴙ Mice Analysis of disease progression in the Pdx1-Cre, Z/EGFP, LSL-KrasG12D/⫹, LSL-Trp53R172H/⫹ mice showed that preneoplastic PanIN lesions develop by 6 weeks of
DASATINIB AND MOUSE PANCREATIC CANCER
299
Figure 4. Dasatinib inhibits Src activity in mouse PDAC cells. (A) Cell lines were established from 3 PDACs (82739, 83320, and 86119) harvested from Pdx1-Cre, Z/EGFP, LSL-KrasG12D/⫹, LSL-Trp53R172H/⫹ mice. Phase contrast images are shown for each (left panels). Expression of Pdx1 in the cell lines confirms their pancreatic origin (right panels). Scale bars, 0.2 m. (B) Recombination of Trp53R172H in cell lines. (C) PDAC cells were treated with a range of dasatinib concentrations for 24 hours before Western blot analysis using anti–phospho-Src Y418 and anti-Src antibodies. (D) PDAC cells were treated with a range of dasatinib concentrations for 96 hours and an MTT proliferation assay then was performed. Values are mean ⫾ standard deviation of quadruplicate wells taken from a representative experiment in a series of 3.
age (n ⬎10 mice killed) and that by 10 weeks mice have an observable mass in the pancreas as detected by wholebody in vivo green fluorescent protein (GFP) imaging. Full pathologic details are provided in Supplementary Tables 2 and 3. Histologic sections of pancreata from mice killed at this time showed multifocal and often widespread neoplastic change, with some invasive PDAC, cystic dilatation of neoplastic ducts, and varying grades of PanIN involving the majority of ducts, with little normal pancreas remaining (Supplementary Figure 5). This model therefore allows us to assess the effects of dasatinib at very early stages of tumor development and at later stages of disease progression. Cohorts of 20 Pdx1-Cre, Z/EGFP, LSL-KrasG12D/⫹, LSL-Trp53R172H/⫹ mice were treated with vehicle or dasatinib (10 mg/kg) from 6 or 10 weeks of age. Mice were allowed to age and were monitored for signs of disease and euthanized when they became symptomatic of disease progression. There was no significant difference in survival between the different
treatment groups (Kaplan–Meier, log-rank, P ⫽ .827): the median survival of vehicle-treated animals was 131 days compared with 127 days and 130 days for animals treated with dasatinib from 6 weeks and 10 weeks of age, respectively (Figure 7A). Analysis of tumor burden in the mice showed that all mice had invasive PDAC; however, the number of mice with metastases was reduced significantly in dasatinib-treated animals (Figure 7B). The incidence of metastases was 61.1% in vehicle-treated animals compared with 26.7% in mice treated with dasatinib from 6 weeks (2 test, P ⫽ .048) and 23.1% in mice treated with dasatinib from 10 weeks (2 test, P ⫽ .036). Full details are provided in Supplementary Table 4. Inhibition of Src activity in tumors from mice treated with dasatinib was confirmed by Western blotting and immunohistochemistry using the anti–phospho-Src Y418 antibody (Figure 7C and D). To assess the effect of dasatinib on the proliferation of PanINs and PDACs, mice were injected with bromodeoxyuridine 2 hours before death. Immunohistochemical
BASIC–LIVER, PANCREAS, AND BILIARY TRACT
July 2010
300
MORTON ET AL
GASTROENTEROLOGY Vol. 139, No. 1
BASIC–LIVER, PANCREAS, AND BILIARY TRACT
Figure 5. Dasatinib inhibits PDAC cell migration and invasion. (A) Migration of PDAC cells was determined in the absence or presence of dasatinib (100 or 200 nmol/L) using a wound healing assay. Values represent the wound area at the end of the experiment expressed as a percentage of the original wound area. Values are mean ⫾ standard deviation of 3 areas of the wound and are taken from a representative experiment in a series of 3. (B) Western blot analysis of Src expression in PDAC cells treated with scrambled or Src siRNA sequences. -tubulin is used as a loading control. (C) Migration of PDAC cells treated with scrambled or Src siRNA sequences calculated as in panel A. (D) Invasion of PDAC cells into collagen gels was determined in the absence or presence of 100 nmol/L dasatinib. Quantification of fold invasion greater than 20 m relative to control (untreated cells) is presented from at least 3 independent experiments. Values are mean ⫾ standard error. (E) PDAC cells were treated with a range of dasatinib concentrations for 24 hours before Western blot analysis using anti–phospho-FAK Y861 and anti–phospho-p130Cas Y249 antibodies. -tubulin is used as a loading control. (F) MMP gelatin zymography was performed on supernatants of untreated and dasatinib-treated (100 nmol/L) cells.
analysis of bromodeoxyuridine staining was performed, and a number of bromodeoxyuridine-positive nuclei were scored. In both PanIN and PDAC there was no significant difference in proliferation between the vehicle- and dasatinib-treated animals (Figure 7E and F). These observations are consistent with continued growth of the primary tumor within the pancreas in mice treated with dasatinib and the lack of survival advantage in these mice, which develop symptoms as a result of primary tumor burden.
Discussion Most early phase clinical trials are performed in patients with advanced refractory disease, and the con-
ventional paradigm for efficacy is dependent on showing a reduction in tumor dimensions using anatomic imaging. However, this approach, and this patient population, is not likely to be appropriate for agents that inhibit cancer cell invasion, migration, and the development of metastases. This is particularly relevant for patients with PDAC, in whom cytotoxic chemotherapy agents have limited efficacy in advanced, inoperable, disease. The mainstay of preclinical in vivo evaluation of novel cancer therapeutics has been performed using human cancer cell lines grown as xenografts in immune-compromised mice but these have been suboptimal in predicting efficacy in subsequent clinical trials in human beings with pancreatic cancer. One drawback with this approach
July 2010
DASATINIB AND MOUSE PANCREATIC CANCER
301
is that the tumor microenvironment and inflammatory response, which play pivotal roles in pancreatic cancer development, are disrupted.19,20 Furthermore, these mouse models are not applicable to evaluate antimetastatic agents, which may not reduce primary tumor bulk but inhibit spread of the disease. These issues can, in part, be overcome by the use of orthotopic models in which injection of human pancreatic tumor cells directly into the mouse pancreas results in metastasis to the liver and lymph node, although this approach still relies on the use of immune-compromised animals. The use of appropriate genetically engineered mouse models of cancer may overcome these problems and can be an alternative tool for preclinical evaluation of cancer therapeutics.21 In this study we have used dasatinib to determine the effects of inhibiting Src kinase activity as an antimetastatic therapeutic strategy in vivo using the Pdx1-Cre, Z/EGFP, LSL-KrasG12D/⫹, LSL-Trp53R172H/⫹ mouse model. Many studies have shown that interference with Src activity inhibits the migration and invasion of epithelial tumor cells, which may contribute to the potential antimetastatic activity of inhibitors such as dasatinib.8 This has been attributed to the well-documented role of Src in regulating integrin adhesions, cadherin-mediated cell– cell adhesions, and MMP expression, all of which influence the ability of cells to break away from the primary tumor mass and colonize at distant sites in the body. Src-dependent phosphorylation of FAK and p130Cas regulates cell motility and invasion, and our data suggest that disruption of the Src-FAK-p130Cas-DOCK180 complex, which controls MMP production, contributes to the effects of dasatinib in PDAC.22 Evidence from human pancreatic cell lines suggests that Src inhibition also blocks tumor angiogenesis, which may contribute to the reduction in metastases seen in orthotopic models of pancreatic cancer.23,24 Here we have shown that inhibition of Src kinase activity by dasatinib can inhibit the development of metastases in vivo. Interestingly, the inhibition of develop-
ment of metastases was similar when dasatinib administration was commenced at either 6 weeks (when mice have preneoplastic PanIN) or 10 weeks (when mice have more advanced disease but not overt metastases). All of the mice in whom dosing commenced at 6 weeks of age ultimately developed invasive PDAC, indicating that dasatinib does not inhibit the conversion of PanINs to invasive PDAC, but has effects at later stages of disease progression. This is potentially significant in the human population in that there is currently no phenotype of PanIN in human beings that is clinically detectable. However, we speculate that administration of dasatinib after potentially curative resection of localized disease could reduce the risk of metastases and improve overall survival. Dasatinib did not improve survival when administered to the Pdx1-Cre, Z/EGFP, LSL-KrasG12D/⫹, LSL-Trp53R172H/⫹ mice. This is most likely owing to the continued growth of the primary tumor because the mice primarily die of tumor burden within the pancreas. The inability of dasatinib to inhibit PDAC proliferation is in agreement with reports in other tumor types that indicate that Src kinase activity is not required for the sustained growth of several types of epithelial tumors.18,25 Interestingly, in contrast to our findings, dasatinib inhibited primary tumor growth in an orthotopic model of pancreatic cancer, which also contributed to the reduction in metastatic burden.24 It may be that in this model different factors are controlling growth of the primary tumor, and it is possible that higher (ie, antiproliferative) doses of dasatinib, which inhibit other pathways in addition to Src kinase activity, also could have inhibited proliferation in our mouse model. In conclusion, we have shown that Src kinase activity increases with progression from early PanIN through to late PanIN and invasive PDAC in a genetically engineered mouse model that recapitulates the human disease. Indeed, we show that the levels of Src and phospho-Src are important indicators of vascular invasion, lymph node
BASIC–LIVER, PANCREAS, AND BILIARY TRACT
Figure 6. Determination of biologically active dose of dasatinib in mice. PDAC cells were grown as subcutaneous tumors, and the mice were treated with dasatinib at 5 mg/kg or 10 mg/kg. After 2 hours the mice were killed and tumors were collected. (A) Western blot analysis using anti–phosphoSrc Y418 and anti-Src antibodies. (B) Immunohistochemical analysis of tumors stained with anti–phospho-Src Y418 antibody from vehicle and dasatinib-treated (10 mg/kg) animals. Images taken at 20⫻ magnification. Representative results are shown for the 83320 cell line. Similar results were obtained for the 86119 and 82739 cells.
302
MORTON ET AL
GASTROENTEROLOGY Vol. 139, No. 1
BASIC–LIVER, PANCREAS, AND BILIARY TRACT
Figure 7. Dasatinib does not increase survival but reduces the incidence of metastasis in Pdx1-Cre, Z/EGFP, LSL-KrasG12D/⫹, LSL-Trp53R172H/⫹ mice. (A) Kaplan–Meier survival curves of mice treated with vehicle or dasatinib (10 mg/kg) from 6 or 10 weeks of age. (B) The number of mice with metastases was counted at the time of death, and confirmed by histology, and results are presented as the percentage of the total number of mice in each cohort. Black columns represent mice with metastases and grey columns represent those without metastases. 2 test: 6-week cohort vs vehicle, P ⫽ .048 and 10-week cohort vs vehicle, P ⫽ .036. (C) Lysates were prepared from tumors harvested from vehicle or dasatinib-treated (10 mg/kg) mice, and Western blot analysis was performed with anti–phospho-Src Y418 and anti-Src antibodies. (D) Immunohistochemical analysis of tumors harvested from vehicle and dasatinib-treated animals, stained with anti–phospho-Src Y418 antibody. Images taken at 20⫻ magnification. In each case representative examples are shown from 2 individual tumors. (E and F) Cell proliferation was assessed by labeling cells for 2 hours with bromodeoxyuridine (BrdU) before killing the mice. Immunohistochemical analysis of BrdU was performed, and the number of positive nuclei was determined. (E) Ten PanINs were scored from 4 mice in each treatment group (n ⫽ 40; Mann–Whitney U test, P ⫽ .665). (F) PDAC. Ten high-powered frames were scored from 4 mice in each treatment group (n ⫽ 40; Mann–Whitney U test, P ⫽ .557).
positivity, and prognosis in human PDAC. We also have shown that inhibition of Src kinase activity by dasatinib can inhibit PDAC cell invasion and migration in vitro, and can inhibit the development of metastases in vivo. We propose that dasatinib should be evaluated further in vivo, either as monotherapy after resection of localized invasive PDAC or in combination with gemcitabine for invasive, nonmetastatic, irresectable PDAC, with the ultimate aim of translating these observations through to clinical evaluation in patients with pancreatic cancer.
Supplementary Material Note: To access the supplementary material accompanying this article, visit the online version of Gastroenterology at www.gastrojournal.org, and at doi: 10.1053/j.gastro.2010.03.034. References 1. Li D, Xie K, Wolff R, et al. Pancreatic cancer. Lancet 2004;363: 1049 –1057.
July 2010
19.
20.
21.
22. 23.
24.
25.
303
in colon cancer cells following treatment with dasatinib. Mol Cancer Ther 2006;5:3014 –3022. Farrow B, Albo D, Berger DH. The role of the tumor microenvironment in the progression of pancreatic cancer. J Surg Res 2008; 149:319 –328. Mahadevan D, Von Hoff DD. Tumor-stroma interactions in pancreatic ductal adenocarcinoma. Mol Cancer Ther 2007;6:1186 – 1197. Olive KP, Jacobetz MA, Davidson CJ, et al. Inhibition of Hedgehog signaling enhances delivery of chemotherapy in a mouse model of pancreatic cancer. Science 2009;324:1457–1461. Hsia DA, Mitra SK, Hauck CR, et al. Differential regulation of cell motility and invasion by FAK. J Cell Biol 2003;160:753–767. Ischenko I, Guba M, Yezhelyev M, et al. Effect of Src kinase inhibition on metastasis and tumor angiogenesis in human pancreatic cancer. Angiogenesis 2007;10:167–182. Trevino JG, Summy JM, Lesslie DP, et al. Inhibition of SRC expression and activity inhibits tumor progression and metastasis of human pancreatic adenocarcinoma cells in an orthotopic nude mouse model. Am J Pathol 2006;168:962–972. Buettner R, Mesa T, Vultur A, et al. Inhibition of Src family kinases with dasatinib blocks migration and invasion of human melanoma cells. Mol Cancer Res 2008;6:1766 –1774.
Received June 30, 2009. Accepted March 8, 2010. Reprint requests Address requests for reprints to: Valerie Brunton, PhD, Edinburgh Cancer Research Centre, Edinburgh University, Crewe Road South, Edinburgh, EH4 2XR, United Kingdom. e-mail: [email protected]; fax: (44) 131-777-3520. Acknowledgments The authors are grateful for support from Colin Nixon and colleagues in Histology services; and from Stephen Bell, Derek Miller, and Tom Hamilton in Biological Research Services at the Beatson Institute of Cancer Research; Jane Hair from NHSGGC Biorepository; and Think Pink for the purchase of the slide scanner. Conflicts of interest These authors disclose the following: Margaret Frame and T. R. Jeffry Evans have received research funding and honoraria from Bristol-Myers Squibb, and have performed consultancy work for Bristol-Myers Squibb. The remaining authors disclose no conflicts. Funding This work was funded by Cancer Research UK grants C2193/A7603 and C157/A9148 and Chief Science Officer, Scottish Executive (N.J.).
BASIC–LIVER, PANCREAS, AND BILIARY TRACT
2. Yeo CJ, Cameron JL, Lillemoe KD, et al. Pancreaticoduodenectomy with or without distal gastrectomy and extended retroperitoneal lymphadenectomy for periampullary adenocarcinoma, part 2: randomized controlled trial evaluating survival, morbidity, and mortality. Ann Surg 2002;236:355–368. 3. Hezel AF, Kimmelman AC, Stanger BZ, et al. Genetics and biology of pancreatic ductal adenocarcinoma. Genes Dev 2006;20: 1218 –1249. 4. Hruban RH, Goggins M, Parsons J, et al. Progression model for pancreatic cancer. Clin Cancer Res 2000;6:2969 –2972. 5. Almoguera C, Shibata D, Forrester K, et al. Most human carcinomas of the exocrine pancreas contain mutant c-K-ras genes. Cell 1988;53:549 –554. 6. Hingorani SR, Wang L, Multani AS, et al. Trp53R172H and KrasG12D cooperate to promote chromosomal instability and widely metastatic pancreatic ductal adenocarcinoma in mice. Cancer Cell 2005;7:469 – 483. 7. Olive KP, Tuveson DA. The use of targeted mouse models for preclinical testing of novel cancer therapeutics. Clin Cancer Res 2006;12:5277–5287. 8. Brunton VG, Frame MC. Src and focal adhesion kinase as therapeutic targets in cancer. Curr Opin Pharmacol 2008;8:427– 432. 9. Frame MC. Src in cancer: deregulation and consequences for cell behaviour. Biochim Biophys Acta 2002;1602:114 –130. 10. Hakam A, Fang Q, Karl R, et al. Coexpression of IGF-1R and c-Src proteins in human pancreatic ductal adenocarcinoma. Dig Dis Sci 2003;48:1972–1978. 11. Lutz MP, Esser IB, Flossmann-Kast BB, et al. Overexpression and activation of the tyrosine kinase Src in human pancreatic carcinoma. Biochem Biophys Res Commun 1998;243:503–508. 12. Demetri GD, Lo Russo P, MacPherson IR, et al. Phase I doseescalation and pharmacokinetic study of dasatinib in patients with advanced solid tumors. Clin Cancer Res 2009;15:6232– 6240. 13. Khoury HJ, Guilhot F, Hughes TP, et al. Dasatinib treatment for Philadelphia chromosome-positive leukemias: practical considerations. Cancer 2009;115:1381–1394. 14. Jackson EL, Willis N, Mercer K, et al. Analysis of lung tumor initiation and progression using conditional expression of oncogenic K-ras. Genes Dev 2001;15:3243–3248. 15. Olive KP, Tuveson DA, Ruhe ZC, et al. Mutant p53 gain of function in two mouse models of Li-Fraumeni syndrome. Cell 2004;119: 847– 860. 16. Gu G, Dubauskaite J, Melton DA. Direct evidence for the pancreatic lineage: NGN3⫹ cells are islet progenitors and are distinct from duct progenitors. Development 2002;129:2447–2457. 17. Novak A, Guo C, Yang W, et al. Z/EG, a double reporter mouse line that expresses enhanced green fluorescent protein upon Cre-mediated excision. Genesis 2000;28:147–155. 18. Serrels A, Macpherson IR, Evans TR, et al. Identification of potential biomarkers for measuring inhibition of Src kinase activity
DASATINIB AND MOUSE PANCREATIC CANCER