Aberrant expression and/or activity of members of the Src family of nonreceptor protein tyrosine kinases (SFK) are commonly observed in progressive stages of human tumors. In prostate cancer, two SFKs (Src and Lyn) have been specifically implicated in tumor growth and progression. However, there are no data in preclinical models demonstrating potential efficacy of Src inhibitors against prostate cancer growth and/or metastasis. In this study, we used the small molecule SFK/Abl kinase inhibitor dasatinib, currently in clinical trials for solid tumors, to examine in vitro and in vivo effects of inhibiting SFKs in prostate tumor cells. In vitro, dasatinib inhibits both Src and Lyn activity, resulting in decreased cellular proliferation, migration, and invasion. In orthotopic nude mouse models, dasatinib treatment effectively inhibits expression of activated SFKs, resulting in inhibition of both tumor growth and development of lymph node metastases in both androgen-sensitive and androgen-resistant tumors. In primary tumors, SFK inhibition leads to decreased cellular proliferation (determined by immunohistochemistry for proliferating cell nuclear antigen). In vitro, small interfering RNA (siRNA)–mediated inhibition of Lyn affects cellular proliferation; siRNA inhibition of Src affects primarily cellular migration. Therefore, we conclude that SFKs are promising therapeutic targets for treatment of human prostate cancer and that Src and Lyn activities affect different cellular functions required for prostate tumor growth and progression. [Cancer Res 2008;68(9):3323–33]

Prostate cancer is the second leading cause of cancer-related deaths in men in the United States (1). Whereas the 5-year relative survival rate of local and regional stages of prostate cancer is 99.9%, the rate drops to 34% if a distant metastasis, including those to bone, is found at the time of diagnosis (1). Prostate cancer metastasis to the bone is the major cause of mortality in afflicted patients. However, due to increased screening for prostate-specific antigen, the majority of prostate cancer patients are diagnosed when the tumor is still confined to the prostate. Nevertheless, large retrospective clinical studies from multiple institutions showed that lymph node metastasis in clinically localized prostate cancer patients is a poor prognostic factor for recurrence and disease-free survival (25). In addition, extended pelvic lymph node dissection increases progression-free survival of the patients at high risk of lymph node involvement (6, 7). Thus, therapeutic strategies that would inhibit growth of existing lymph node metastases or prevent development of lymph node metastases would greatly improve survival of patients with lymph node–positive disease.

The nonreceptor protein tyrosine kinase Src and its family members (Src family kinases, SFKs), have been shown to be up-regulated in multiple types of human tumors, with Src activity increasing proportionally to the progressive stages of the disease (8, 9). Among the SFKs, Src itself is most frequently implicated in human cancer, and previous studies have shown that, in mouse models, Src activation is associated with progression and metastasis in pancreatic (10, 11) and colorectal (12, 13) carcinomas. In prostate cancer cells in vitro, inhibition of SFKs decreases proliferation (14) and, more profoundly, invasion and migration (15); the latter through selective inhibition of phosphorylation of Src substrates, such as focal adhesion kinase (FAK) and Crk-associated substrate (p130Cas; refs. 16, 17). However, Lyn, an SFK member, previously thought to be expressed primarily in B cells (18), has been shown more recently to be expressed in normal prostate epithelium and in the majority of primary human prostate cancer specimens (19). Mice, in which Lyn is functionally deleted, showed abnormal morphogenesis of the prostate gland (19). Moreover, administration of a Lyn-specific peptidomimetic inhibitor significantly reduced the primary tumor volume of human prostate cancer cells implanted s.c. into nude mice (19). Thus, prostate cancer may be unusual in that two SFKs, Src and Lyn, promote distinct stages of cancer progression (while the SFK Yes is expressed in prostate epithelial cells, there is no evidence that it plays a role in prostate development or tumorigenesis). A small molecule SFK/Abl selective inhibitor, dasatinib (BMS-354825; refs. 14, 20), is approved by the U.S. Food and Drug Administration for chronic myelogenous leukemia and Philadelphia chromosome-positive acute lymphoblastic leukemia and is now in phases I and II clinical trials for treatment of solid tumors (21, 22). Dasatinib has been shown to effectively inhibit autophosphorylation of SFKs, including Src, Yes, Lck, and Lyn (14, 15). In prostate cancer cells in culture, Nam et al. showed that dasatinib affects tumorigenic properties, inhibiting migration and invasion (15). However, studies in preclinical models demonstrating potential efficacy of Src-selective inhibitors against prostate cancer growth and/or metastasis had not heretofore been undertaken and which antitumor properties might be attributed to Src and Lyn are not known. We, therefore, examined the role of Src and Lyn individually by silencing them in cultured cells and examining their properties and then used dasatinib to conduct in vivo orthotopic mouse experiments to determine the effects of inhibition of both Src and Lyn on human prostate cancer growth and development of lymph node metastases. Here, we provide the first demonstration that inhibition of SFKs effectively inhibited both tumor growth and development of lymph node metastasis of human prostate cancer cells growing in nude mice.

Cells and cell culture. PC-3 and LNCaP human prostate tumor cells were originally purchased from the American Type Culture Collection. The highly metastatic variant of PC-3 (PC-3MM2) was established by several cycles of in vivo orthotopic implantation-metastatic selection, as previously described (23, 24). For in vivo bioluminescence imaging, the PC-3MM2GL cells were produced by stably transfecting a green fluorescent protein (GFP) and luciferase (GL) fusion gene using vesicular stomatitis virus G–pseudotyped retrovirus produced in 293GPG packaging cells, as described previously (25). Briefly, PC-3MM2 cells (2 × 104/cm2) were incubated with 1:1 mixture of retroviral supernatant and DMEM/Ham's F-12 media supplemented with 10% fetal bovine serum (FBS) for 48 h in the presence of polybrene (6 μg/mL). The monolayer cultures were then expanded into 15-cm dishes and GFP-expressing cells were selected by FACSAria cell sorter (BD Biosciences). Two rounds of cell sorting enhanced the percentage of GFP-positive cells to 98%. PC-3MM2GL cells were measured for luciferase activity by IVIS 200 bioluminescence imaging system (Xenogen Co.). PC-3MM2GL cells maintained the metastatic potentials of the parental cells determined by an in vivo orthotopic mouse model.

The androgen receptor (AR)–expressing variant of PC-3 cell line was created as previously described (26). The FLAG epitope tag was fused at the NH2 terminus of the human AR and subcloned into pcDNA3.1 to generate pcDNA-f:AR plasmid. PC-3 cells were transfected with 1 μg of pcDNA3.1 and selected with G418 (0.5 mg/mL). Expression of AR in the selected clones was confirmed by Western blotting using anti-FLAG monoclonal and anti-AR polyclonal antibodies.

Cells were maintained as monolayer cultures in DMEM/Ham's F-12 media supplemented with 10% FBS and 1× penicillin-streptomycin (Life Technologies Invitrogen Co.) and incubated in 5% CO2:95% air at 37°C. Cultures were free of Mycoplasma and murine viral pathogens (assayed by Whittaker M.A. Bioproducts).

Immunoblotting. Cultured cells were lysed in radioimmunoprecipitation assay-A (RIPA-A) buffer, and immunoblotting was done as previously described (27, 28). Primary antibodies and dilutions are as follows: anti–phospho-[Y416]-Src antibody (Cell Signaling Technology, Inc.; diluted 1:1,000), anti-Src monoclonal antibody (Calbiochem; diluted 1:1,000), anti–phospho-[Y861]-FAK antibody (Biosource Invitrogen Co.; diluted 1:1,000), anti-FAK monoclonal antibody (BD Transduction Laboratories; diluted 1:1,000), anti–phospho-[Y165]-p130 CAS antibody (Cell Signaling Technology, Inc.; diluted 1:1,000), and anti-p130 CAS antibody (BD Transduction Laboratories; diluted 1:1,000).

Immunoprecipitation. For detection of phospho-[Y397]-Lyn expression, cell lysates were prepared as described above and immunoprecipitation was done as previously described (11, 27). Briefly, protein samples (500 μg), in a total volume of 200 μL of RIPA-A buffer, were incubated with 2 μg of mouse monoclonal anti-Lyn antibody (Santa Cruz Biotechnology, Inc.) overnight at 4°C. Protein-antibody complex was precipitated with protein G agarose beads (Upstate USA, Inc.), subsequently followed by immunoblotting with rabbit anti–phospho-[Y416]-Src antibody (diluted 1:1,000; Cell Signaling Technology, Inc.) as described above.

In vitro proliferation assays. PC-3MM2GL and LNCaP cells were plated at a density of 6,000 per well in 96-well plates in 100 μL of DMEM/Ham's F-12 media supplemented with 10% FBS. After 24 h of incubation, 100 μL of complete media containing 2× indicated amount of dasatinib (BMS-354825; Bristol-Myers Squibb Co.) dissolved in DMSO were added to each well. At 24, 48, and 72 h time points, the number of viable cells was counted with hemacytometer.

For in vitro proliferation assay of small interfering RNA (siRNA)–expressing PC-3MM2GL cells, cells were plated at a density of 25,000 per well in 24-well plates in 1.5 mL of selecting antibiotics-containing media. The number of viable cells was counted with hemacytometer at 24, 48, 72, and 96 h time points. All proliferation assays were performed in triplicate and repeated twice.

Migration assays. Effects of dasatinib or Src-targeted siRNA on migration of prostate cells were determined by the modified Boyden chamber migration assay as described by Lesslie et al. (29) Briefly, parental or Src-targeted siRNA-expressing PC-3MM2GL cells (2.0 × 105) were suspended in the upper well of the 8.0-μm pore size polyethylene terepthalate membrane culture inserts for 24-well plates (BD Biosciences) in 500 μL DMEM/Ham's F-12 containing 1% FBS and specified concentrations of dasatinib. The lower chamber was filled with 750 μL of DMEM/Ham's F-12 media supplemented with 10% FBS as a chemoattractant. After 48 h incubation, the culture inserts were removed, and nonmigratory cells in the upper membrane surface were scraped with a cotton swab. Cells that had migrated to the lower membrane surface were fixed and stained with HEMA-3 (Biochemical Sciences) according to the manufacturer's instruction. Migratory cells were counted under a microscope at 100× magnification in five random fields per inserts in triplicate.

Creation of Src-targeted or Lyn-targeted siRNA expression plasmid and stable transfection. Src-targeted siRNA expression plasmids were created using pSilencer1.0 U6 (Ambion, Inc.), as previously described (28). Briefly, two c-src–specific target sequences used were 5′-AACAAGAGCAAGCCCAAGGAT-3′ and 5′-AAGCTGTTCGGAGGCTTCAAC-3′. PC-3MM2GL cells, 80% confluent on 10-cm dish, were transfected with 1 μg of each siRNA expression plasmid and 1 μg of pcDNA G418-resistant promoterless plasmid for selection of transfectants, using FuGene 6 transfection reagent (Roche Diagnostics Co.). Cells were then grown in selective antibiotics media (600 μg/mL G418-containing DMEM/Ham's F-12 media supplemented with 10% FBS), and resistant clones were expanded.

Lyn-targeted siRNA expression vector was created in the laboratory of Dr. Eugenie S. Kleinerman, using pSilencer2.1 U6 Hygro (Ambion, Inc.), as previously described (30). Lyn-specific target sequence used was 5′-AAUGGUGGAAAGCAAAGUCCC-3′. PC-3MM2GL cells were transfected, selected, and expanded as described above, except for using DMEM/Ham's F-12 media containing 120 μg/mL hygromycin B as selective media. The control vector was constructed by inserting a sequence that expresses an siRNA with limited homology to sequences in the human and mouse genomes.

Orthotopic implantation of tumor cells. PC-3MM2GL or PC-3AR-A1 cells were detached from subconfluent cultures, and a desired number of cells were centrifuged and resuspended with Ca2+-free and Mg2+-free HBSS (Life Technologies Invitrogen Co.) For implantation of cells into the prostate, the procedure of Kim et al. (31) was followed. Male athymic nude mice (NCr-nu/nu; ages 8–12 wk; the National Cancer Institute-Frederick Animal Production Area) were anesthetized with pentobarbital sodium i.p. (0.5 mg/1 g body weight; Nembutal, Abott Laboratories) and placed in a supine position. A midline incision was made on the lower abdomen, and the prostate was exteriorized. Fifty microliters of HBSS containing PC-3MM2GL (5 × 104) or PC-3AR-A1 (1 × 106) cells were injected into the dorsolateral side of the prostate. The incision was closed with surgical metal clips (Braintree Scientific, Inc.).

Drug formulation and administration. Three days after xenograft injection, mice were randomized to receive drug or control vehicle (six mice per group). For p.o. administration, dasatinib (Bristol-Myers Squibb Co.) was dissolved in 80 mmol/L citrate buffer (pH 3.1) according to the manufacturer's instructions. A concentration of 15 mg/kg body weight/d was given p.o., at 24-h intervals, using 20-gauge gavage needle. The control group of mice was given an equal volume of diluent buffer by the same gavage technique. Mice were treated for 28 d. For in vitro assays, dasatinib stock solution (10 mmol/L; molecular weight, 488.0) was prepared and further diluted in DMSO.

Necropsy and tissue preparation. At the end of 4-wk treatment, mice were euthanized by pentobarbital sodium overdose (1 mg/1 g body weight) 4 h after the last drug dose or control diluent was given. Lymph node metastasis was assessed macroscopically, and enlarged lymph nodes were harvested for pathologic examination. Tumors were surgically excised and weighed, followed by fixation in phosphate-buffered 10% formaldehyde. A part of tumor tissue was embedded in OCT compound (Sakura Finetek), snap-frozen in liquid nitrogen, and stored at −80°C. Three medium-sized tumors from each group were chosen for further immunohistochemical analysis.

Immunohistochemical staining of total Src, Lyn, FAK, and autophosphorylated SFKs, and FAK-phospho-[Y861]. Paraffin-embedded tumor tissues were sectioned 8 to 10 μm thickness and mounted on positively charged microscope slides. Tissue slides were preheated at 60°C for 30 min and dewaxed by immersion in xylene followed by successively diluted solutions of ethanol. Antigen retrieval was accomplished either by boiling the slides in pressure cooker at 125°C for 5 min, immersed in Borg decloaker solution (Biocare Medical, Inc.) for Src, autophosphorylated SFKs and Lyn staining or by boiling the slides in 0.1mol/L EDTA buffer for 5 min using a microwave oven and subsequently incubating the slides for 1 h in Dako target retrieval solution (Dako North America, Inc.) for FAK and FAK-phospho-[Y861] staining, respectively. Endogenous peroxidase activity was blocked by incubating in 3% H2O2 in PBS for 12 min. After rinsing with PBS thrice for 3 min each, nonspecific tissue binding was blocked by 1-h incubation in protein block solution (Cyto Q immunodiluent buffer; Innovex). Primary antibody was diluted in protein block solution and incubated overnight at 4°C. Dilution of primary antibodies are as follows: anti-Src antibody (1:100; Cell Signaling Technology, Inc.), anti–phospho-[Y416]-Src antibody (1:100; Cell Signaling Technology, Inc.), anti-Lyn antibody (1:100; Santa Cruz Biotechnology, Inc.), anti-FAK antibody (1:100; Cell Signaling Technology, Inc.), and anti–phospho-[Y861]-FAK (1:100; Biosource Invitrogen Co.). Slides were washed with PBS thrice for 3 min each followed by Mach 4 Universal horseradish peroxidase polymer (Biocare Medical, Inc.) application for 20 min as a secondary antibody. The stain was visualized by incubation in 3,3′-diaminobenzidine (DAB) and counterstained with Gill's no. 3 hematoxylin. Internal negative control samples were exposed to protein block solution instead of the primary antibodies and showed no specific signaling. Slides were dried and mounted with Universal Mount solution (Research Genetics, Invitrogen Co.).

Quantification of immunohistochemical stain intensity. To quantify signal intensity of phosphorylated and total Src and FAK stainings, 5 to 10 randomly selected brightfield microscope images (magnification 100×, area 0.14 mm2) per sample were captured with Sony DXC-990 three-chip charged-coupled device color video camera (Sony Co.) mounted on Nikon Microphot-FX microscope (Nikon Co.). Images were then processed and quantified with ImageJ, a public domain Java image processing program (U.S. NIH6

). Briefly, brown-colored images specific for DAB stain (red = 0.26814753, green = 0.57031375, blue = 0.77642715) were extracted by color deconvolution macro (32), inversed and measured for intensity using ImageJ internal commands. All intensity values within the same group were averaged to calculate ratios of phosphorylated SFKs to total Src stain intensity and phosphorylated FAK to total FAK staining intensity.

Immunohistochemical staining of proliferating cell nuclear antigen. To determine the percentage of proliferating cells, paraffin-embedded sections were stained for proliferating cell nuclear antigen (PCNA), as previously described (33). For quantification, three randomly selected brightfield microscope images (magnification 40×, area 0.89 mm2) per sample were obtained as described above. The total cell number in each image was calculated by counting hematoxylin-positive cells using ImageJ particle count command, and DAB-positive cells were counted as PCNA-positive cells in each image.

Analysis of apoptotic tumor cells by terminal deoxynucleotidyl transferase biotin uUTP nick-end labeling. To quantify numbers of the apoptotic tumor cells, snap-frozen tumor tissue sections were analyzed with a commercially available terminal deoxynucleotidyl transferase biotin uUTP nick-end labeling (TUNEL) kit (DeadEnd Fluorometric TUNEL kit, Promega Corp.), and counting was performed as detailed previously (33).

Immunohistochemical staining of CD31/PECAM and quantification of microvessel density. To determine tumor microvessel density, snap-frozen tumor tissues were sectioned and stained for CD31/PECAM, as previously described (33). Briefly, the slides were fixed and incubated with anti-mouse CD31/PECAM rat antibody (BD PharMingen BD Bioscience; 1:800 diluted in protein block solution) overnight at 4°C, followed by secondary antibody reaction with peroxidase-conjugated goat anti-rat antibody (Jackson ImmunoResearch Laboratories, Inc.; diluted 1:200). The stain was visualized by incubation in DAB followed by counterstaining with Gill's no. 3 hematoxylin. Five randomly selected brightfield microscope images (magnification 40×; area 0.89 mm2) per sample were obtained as described above, and positively stained microvessels were counted using ImageJ program.

Statistics. All statistical analyses were performed in SPSS 12.0 for Windows (SPSS, Inc.). The Mann-Whitney U test was conducted to compare differences in tumor weight. Incidences of tumors and lymph node metastases were compared between groups with Fisher's exact test. Migratory cell numbers in modified Boyden chamber migration assay, microvessel density, PCNA-positive cells, and TUNEL-positive cells were compared by Student's t test. All statistical tests were two-sided.

Dasatinib treatment decreases expression of activated Src and Lyn and the Src substrates FAK-[Y861] and p130Cas-[Y165] in vitro. To directly determine the effect of dasatinib on Src autophosphorylation and phosphorylation of Src substrates, time course and dose-dependence experiments were performed. PC-3MM2GL cells were treated with increasing doses of dasatinib for 1 hour (Fig. 1A,, left) or with 100 nmol/L dasatinib (Fig. 1A,, right) for the times indicated. Cell lysates were prepared, and immunoblotting was performed as described in Materials and Methods. Dasatinib inhibited expression of Src-phospho-[Y419] in a dose-dependent manner, with inhibition apparent at 10 nmol/L (Fig. 1A,, left), results similar to those observed by Lombardo et al. (14). In contrast, but as expected for a competitive ATP inhibitor, expression of total Src was not changed. Inhibition of Src-[Y419] phosphorylation expression by dasatinib was observed as early as 30 minutes after addition of 100 nmol/L dasatinib, and the effects were sustained until 72 hours after treatment (Fig. 1A,, right). Phosphorylation of FAK and p130Cas were decreased as well, but with different kinetics. Expression of FAK-phospho-[Y861] was consistently (but transiently) increased at lower concentrations (<150 nmol/L; 1 hour treatment; Fig. 1A , left), but subsequently decreased at higher drug concentrations. Similar results were observed with expression of expression of p130Cas-phospho-[Y165]. Minimal effects on expression of total FAK and p130Cas were observed. The reason for the transient increase in FAK and p130Cas phosphorylation is unclear, but may reflect the requirement that Src inhibition must precede inhibition of phosphorylation of its substrates.

Figure 1.

In vitro effects of dasatinib on Src and Src substrates. A, dose dependency (left) and time course (right) of dasatinib's effects on inhibition of Src, FAK, and p130 Cas phosphorylation were determined by immunoblotting. PC-3MM2GL cells were grown to 90% confluency in 10% FBS-supplemented DMEM/F-12 media, then treated with increasing doses of dasatinib for 1 h (left), or treated with 100 nmol/L dasatinib (right) for the times indicated. DMSO-treated or empty-treated (No Tx) cells were used as controls (left). B, effects of dasatinib on sustained Src and FAK phosphorylation in LNCaP and PC-3 parental cells were examined by Western blotting. Cells (90% confluent) were treated with dasatinib (50 or 100 nmol/L) for 1 or 4 h. C and D, dose dependency (C) and time course (D) of dasatinib's effect on Lyn-phospho-[Y397] expression was determined by immunoprecipitation with anti-Lyn antibody followed by immunoblotting with anti–Src-phospho-[Y416] antibody due to its cross-reactivity with Lyn-phospho-[Y397]. PC-3MM2GL cells (90% confluent) were treated with increasing doses of dasatinib for 1 h (C) or treated with 100 nmol/L dasatinib (D) for the times indicated. Dasatinib treatment (>100 nmol/L) effectively reduced expression of Lyn-phospho-[Y397] in a dose-dependent manner (C). Inhibition of Lyn-phospho-[Y397] expression was apparent as early as 30 min and was sustained to 72 h after treatment (D). Dasatinib had minimal effects on total Lyn protein level determined by immunoblotting of whole cell lysates with anti-Lyn antibody.

Figure 1.

In vitro effects of dasatinib on Src and Src substrates. A, dose dependency (left) and time course (right) of dasatinib's effects on inhibition of Src, FAK, and p130 Cas phosphorylation were determined by immunoblotting. PC-3MM2GL cells were grown to 90% confluency in 10% FBS-supplemented DMEM/F-12 media, then treated with increasing doses of dasatinib for 1 h (left), or treated with 100 nmol/L dasatinib (right) for the times indicated. DMSO-treated or empty-treated (No Tx) cells were used as controls (left). B, effects of dasatinib on sustained Src and FAK phosphorylation in LNCaP and PC-3 parental cells were examined by Western blotting. Cells (90% confluent) were treated with dasatinib (50 or 100 nmol/L) for 1 or 4 h. C and D, dose dependency (C) and time course (D) of dasatinib's effect on Lyn-phospho-[Y397] expression was determined by immunoprecipitation with anti-Lyn antibody followed by immunoblotting with anti–Src-phospho-[Y416] antibody due to its cross-reactivity with Lyn-phospho-[Y397]. PC-3MM2GL cells (90% confluent) were treated with increasing doses of dasatinib for 1 h (C) or treated with 100 nmol/L dasatinib (D) for the times indicated. Dasatinib treatment (>100 nmol/L) effectively reduced expression of Lyn-phospho-[Y397] in a dose-dependent manner (C). Inhibition of Lyn-phospho-[Y397] expression was apparent as early as 30 min and was sustained to 72 h after treatment (D). Dasatinib had minimal effects on total Lyn protein level determined by immunoblotting of whole cell lysates with anti-Lyn antibody.

Close modal

Because PC-3MM2GL cells were genetically altered to express both GFP and luciferase, we examined the effects of dasatinib on phosphorylation of Src and FAK in two additional human prostate cancer cell lines (LNCaP and the parental PC-3 from which PC-3MM2GL cells were derived). As shown in Fig. 1B, cells were treated with dasatinib (50 and 100 nmol/L) for 1 or 4 hours, and immunoblotting was performed as described above. Inhibition of phosphorylated FAK and Src, but not total FAK or Src, was similar to that observed in PC-3MM2GL cells, suggesting that dasatinib can effectively inhibit Src autophosphorylation and phosphorylation of key Src substrates in vitro, and expression of luciferase and GFP did not affect SFK inhibition.

As Lyn is also expressed in prostate cancer cells and has been shown to contribute to tumorigenic growth (19), we next tested the effects of dasatinib on autophosphorylation of Lyn. PC-3MM2GL cells were treated with increasing doses of dasatinib for 1 hour (Fig. 1C) or 100 nmol/L of dasatinib for indicated times (Fig. 1D). Cell lysates were prepared, immunoprecipitated with anti-Lyn antibody. followed by immunoblotting with anti-chicken Src-phospho-[Y416] antibody, due to its cross-reactivity with the autophosphorylation site (Tyr397) of Lyn. As shown in Fig. 1C, dasatinib treatment effectively inhibited expression of Lyn-phospho-[Y397] in a dose-dependent manner, although a higher concentration was required to achieve a similar extent of inhibition to that of Src-phospho-[Y419] inhibition as shown in Fig. 1A (left). Inhibition of Lyn-phospho-[Y397] expression was apparent as early as the 30-minute time point and was sustained to 72 hours after treatment (Fig. 1D).

Dasatinib treatment inhibits proliferation of human prostate cancer cells in vitro. We next determined the effects of dasatinib on proliferation of PC-3MM2GL and LNCaP cells. For these experiments, cells were plated in 96-well plates (6,000 per well), followed by dasatinib treatment after 24 hours as described in Materials and Methods. Viable cells were counted after 24, 48, and 72 hours. Dasatinib treatment inhibits proliferation of both PC3-MM2GL cells (Fig. 2A) and LNCaP cells (Fig. 2B) in a dose-dependent manner, with significant inhibition occurring at concentrations above 100 nmol/L (P < 0.01; Student's t test), concentrations very close to the IC50 for Lyn inhibition.

Figure 2.

Effects of dasatinib on proliferation and migration of cultured prostate tumor cells. PC-3MM2GL (A) and LNCaP (B) cells were seeded (6,000 per well in 96-well plates) 24 h before treatment with indicated amount of dasatinib. Numbers of viable cells from triplicate wells were counted under a microscope at indicated time points with the aide of a hemacytometer. *, P < 0.01 by Student's t test compared with the control group. C, PC-3MM2GL cells (2.0 × 105 cells, suspended in 1% FBS supplemented media) were seeded in the upper well of a modified Boyden chamber and then treated with control (DMSO) or 50 or 100 nmol/L of dasatinib. Complete media containing 10% FBS was used as chemoattractant in the lower chamber. After 48-h incubation, migratory cells were stained and counted. Columns, average number of migratory cells (triplicate wells) in five random fields in 100× magnification per inserts; error bars, SD. *, P < 0.001 by Student's t test compared with the control group.

Figure 2.

Effects of dasatinib on proliferation and migration of cultured prostate tumor cells. PC-3MM2GL (A) and LNCaP (B) cells were seeded (6,000 per well in 96-well plates) 24 h before treatment with indicated amount of dasatinib. Numbers of viable cells from triplicate wells were counted under a microscope at indicated time points with the aide of a hemacytometer. *, P < 0.01 by Student's t test compared with the control group. C, PC-3MM2GL cells (2.0 × 105 cells, suspended in 1% FBS supplemented media) were seeded in the upper well of a modified Boyden chamber and then treated with control (DMSO) or 50 or 100 nmol/L of dasatinib. Complete media containing 10% FBS was used as chemoattractant in the lower chamber. After 48-h incubation, migratory cells were stained and counted. Columns, average number of migratory cells (triplicate wells) in five random fields in 100× magnification per inserts; error bars, SD. *, P < 0.001 by Student's t test compared with the control group.

Close modal

Dasatinib treatment inhibits tumor cell migration in vitro. Previous studies have shown that dasatinib is an effective inhibitor of prostate tumor cell migration in vitro (15). To confirm and extend these results, migration of PC-3MM2GL cells was examined, using a modified Boyden chamber as described in Materials and Methods. As shown in Fig. 2C, a significant reduction of migration was observed in cells treated with either 50 or 100 nmol/L dasatinib (P < 0.001, Student's t test).

Specific knockdown of src in cultured prostate tumor cells inhibits migration, whereas lyn-specific knockdown affects in vitro proliferation. As stated above, reduction of Src itself rarely affects proliferation. Therefore, to determine whether Src and Lyn affected different properties of prostate cancer cells, we generated src-specific or lyn-specific knockdown clones of PC-3MM2GL cells by stably transfecting siRNA-expressing vectors. As shown in Fig. 3C, src-targeted siRNA-expressing clones 23 and 24 were reduced in Src protein levels compared with the parental PC-3MM2GL cells while they maintained “wild-type” Lyn protein levels. Correspondingly, lyn-targeted siRNA expression clones 8 and 10 had reduced Lyn protein levels while maintaining “wild-type” Src protein levels. Scrambled sequence-targeted RNA transfection had no effect on either Src or Lyn protein levels.

Figure 3.

In vivo effects of dasatinib on growth and lymph node metastases and effects of src-specific or lyn-specific knockdown on proliferation and migration. A, cells were implanted into the prostates of nude mice. Three days later, the treatment group received dasatinib at 15 mg/kg/d in two doses. Tumors were surgically excised and kept on ice until photography. B, mice carrying similar size tumors were selected from each group on necropsy. The margin of the primary tumor was circled with a black dotted line (Tu). A solid, opaque, and enlarged tumor-positive lymph node (circled with a solid yellow line; +LN) was observed in the control mouse, whereas the dasatinib-treated mouse (right) had a clear and transparent lymph node (−LN). Intestines were displaced for photography. All metastatic lymph nodes were harvested and pathologically confirmed. C, PC-3MM2GL cells were stably transfected with siRNA-expressing vectors targeting Src or Lyn (denoted as siSrc and siLyn, respectively). SiRNA expression vector targeting scrambled sequence was used as siRNA control (siControl). Two clones from each transfection were selected and expanded under selective antibiotics media. Total protein levels of Src and Lyn were determined by Western blotting. src-targeted siRNA expression clones 23 and 24 had reduced Src expression while maintaining Lyn expression level. Similarly, lyn-targeted clones 8 and 10 had reduced Lyn expression while maintaining Src protein level. Scrambled-sequence control siRNA had no effect on Src or Lyn. D, PC-3MM2GL parental; Src or Lyn knockdown cells were seeded (2.5 × 105 per well in 24-well plates) in triplicate. Numbers of viable cells from triplicate wells were counted under a microscope at indicated time points with the aide of a hemacytometer. *, P < 0.05 by Student's t test compared with the parental cell line. Error bars, SD. E, PC-3MM2GL parental and Src knockdown cells (2.0 × 105; suspended in 1% FBS supplemented media) were seeded in the upper well of a modified Boyden chamber. Complete media containing 10% FBS was used as chemoattractant in the lower chamber. After 48 h incubation, migratory cells were stained and counted. Columns, average number of migratory cells (triplicate wells) in five random fields in 100× magnification per insert; error bars, SD. *, P < 0.001 by Student's t test compared with the parental cell line.

Figure 3.

In vivo effects of dasatinib on growth and lymph node metastases and effects of src-specific or lyn-specific knockdown on proliferation and migration. A, cells were implanted into the prostates of nude mice. Three days later, the treatment group received dasatinib at 15 mg/kg/d in two doses. Tumors were surgically excised and kept on ice until photography. B, mice carrying similar size tumors were selected from each group on necropsy. The margin of the primary tumor was circled with a black dotted line (Tu). A solid, opaque, and enlarged tumor-positive lymph node (circled with a solid yellow line; +LN) was observed in the control mouse, whereas the dasatinib-treated mouse (right) had a clear and transparent lymph node (−LN). Intestines were displaced for photography. All metastatic lymph nodes were harvested and pathologically confirmed. C, PC-3MM2GL cells were stably transfected with siRNA-expressing vectors targeting Src or Lyn (denoted as siSrc and siLyn, respectively). SiRNA expression vector targeting scrambled sequence was used as siRNA control (siControl). Two clones from each transfection were selected and expanded under selective antibiotics media. Total protein levels of Src and Lyn were determined by Western blotting. src-targeted siRNA expression clones 23 and 24 had reduced Src expression while maintaining Lyn expression level. Similarly, lyn-targeted clones 8 and 10 had reduced Lyn expression while maintaining Src protein level. Scrambled-sequence control siRNA had no effect on Src or Lyn. D, PC-3MM2GL parental; Src or Lyn knockdown cells were seeded (2.5 × 105 per well in 24-well plates) in triplicate. Numbers of viable cells from triplicate wells were counted under a microscope at indicated time points with the aide of a hemacytometer. *, P < 0.05 by Student's t test compared with the parental cell line. Error bars, SD. E, PC-3MM2GL parental and Src knockdown cells (2.0 × 105; suspended in 1% FBS supplemented media) were seeded in the upper well of a modified Boyden chamber. Complete media containing 10% FBS was used as chemoattractant in the lower chamber. After 48 h incubation, migratory cells were stained and counted. Columns, average number of migratory cells (triplicate wells) in five random fields in 100× magnification per insert; error bars, SD. *, P < 0.001 by Student's t test compared with the parental cell line.

Close modal

First, proliferation rates of clones reduced in Src and Lyn were determined. For these experiments, cells were plated in 24-well plates (25,000 per well), and viable cells were counted at 24, 48, 72, and 96 hour time points. As shown in Fig. 3D, src-targeted siRNA expression has modest effects on proliferation of PC-3MM2GL cells (P = 0.047, Student's t test). However, in sharp contrast, PC-3MM2GL cell proliferation was extensively reduced by lyn-targeted siRNA expression compared with parental cells (96 hour time point, P < 0.001, Student's t test).

Migration of Src-knockdown PC-3MM2GL cells were next examined using a modified Boyden chamber, as described in Materials and Methods. As shown in Fig. 3E, a significant reduction of migration was observed in Src-knockdown cells compared with parental cells, consistent with numerous studies that Src affects cellular properties more associated with metastasis than cellular proliferation (reviewed by Gallick and Summy; ref. 8). Because of the very slow proliferating rate, migration of clones with reduced Lyn was not determined.

Oral administration of dasatinib reduces in vivo orthotopic tumor weight and incidence of lymph node metastases. We next examined whether dasatinib administration affects growth of the primary tumor and/or development of lymph node metastases in an orthotopic nude mouse model for prostate cancer. For these studies, two variants of PC-3 cell line, PC-3MM2GL (highly metastatic variant) and PC-3AR-A1 (transfected with a functional AR; ref. 26), were chosen to compare the effects of dasatinib on isogenic models with and without functional AR. Cells (1 × 105 for PC-3MM2GL and 1 × 106 cells for PC-3AR-A1) were injected into the prostates of nude mice. Different initial inocula were chosen as PC-3AR-A1 tumors grow more slowly than PC-3 variants not expressing AR. All mice developed primary prostate tumors (Table 1). For studies with dasatinib, mice (n = 6 for PC-3MM2GL tumors and n = 11 for PC-3AR-A1 tumors) were treated with dasatinib 3 (PC-3MM2GL) or 5 (PC-3AR-A1) days after injection (once daily, 15 mg/kg/d by p.o. gavage). Mice were sacrificed after 28 days of treatment. Although, under these experimental conditions, tumors from the untreated PC-3AR-A1 were still significantly smaller than those from untreated PC3-MM2, tumors from mice receiving dasatinib were significantly reduced in weight for both types of cells inoculated (P = 0.015 for PC-3MM2GL and P = 0.008 for PC-3AR-A1, Mann-Whitney U test) compared with the control diluent-treated mice. In addition, dasatinib administration significantly reduced the incidence of metastases to the iliac lymph nodes (P = 0.015 for PC-3MM2GL and P < 0.001 for PC-3AR-A1, Fisher's exact test). Lymph node metastases were assessed by identifying solid, opaque, and enlarged iliac lymph node, as represented in Fig. 3B. Metastatic lymph nodes were harvested and pathologically confirmed by H&E staining.

Table 1.

Effects of dasatinib on growth and lymph node metastasis of PC-3MM2GL or PC-3AR-A1 cells in an orthotopic nude mouse model

Cell lineTreatment groupTumor incidenceTumor weight (g)
LN metastases incidence*
MedianIQR
PC-3MM2GL Control 6/6 1.25 (1.58–1.00) 6/6 
 Dasatinib 6/6 0.55 (0.75–0.46) 1/6§ 
PC-3AR-A1 Control 10/10 0.51 (0.72–0.30) 10/10 
 Dasatinib 11/11 0.30 (0.30–0.15) 2/11§ 
Cell lineTreatment groupTumor incidenceTumor weight (g)
LN metastases incidence*
MedianIQR
PC-3MM2GL Control 6/6 1.25 (1.58–1.00) 6/6 
 Dasatinib 6/6 0.55 (0.75–0.46) 1/6§ 
PC-3AR-A1 Control 10/10 0.51 (0.72–0.30) 10/10 
 Dasatinib 11/11 0.30 (0.30–0.15) 2/11§ 

Abbreviations: IQR, interquartile range; LN, lymph node.

*

Incidence of lymph node metastases was determined by identifying solid, opaque, and enlarged iliac lymph node(s).

Interquartile range represents a range of 75th percentile and 25th percentile of the tumor weight in the group.

Median tumor weight in the dasatinib-treated group of PC-3MM2GL tumors was significantly reduced (P = 0.015, by Mann-Whitney U test).

§

Incidence of lymph node metastasis was significantly decreased (P = 0.015 for PC-3MM2GL tumors and P < 0.001 for PC-3AR-A1 tumors, by Fisher's exact test) in dasatinib-treated groups compared with the respective control groups.

Median tumor weight in the dasatinib-treated group of PC-3AR-A1 tumors was significantly reduced (P = 0.008, by Mann-Whitney U test).

We repeated the same in vivo tumor experiment, but with a different administration schedule. Orthotopic implantation was accomplished as described above (n = 10), but one group of animals (n = 5) received 15 mg/kg/d dasatinib divided by two doses at 12-hour intervals by the same p.o. gavage technique for 28 days. Figure 3A shows that dasatinib effectively reduced the median tumor weight (2.27 g for control group versus 0.43 g for dasatinib-treated group; P = 0.028, Mann-Whitney U test). Thus, we conclude that dasatinib significantly decreased prostate tumor growth and lymph node metastasis in this model.

Primary tumor size does not affect development of lymph node metastases. Because of the ability of dasatinib to inhibit primary tumor growth, we designed experiments to determine if inhibition of lymph node metastases simply reflected lack of time for tumor progression to occur or whether additional intrinsic properties of dasatinib independently inhibited development and/or outgrowth of lymph node metastases. As shown in Fig. 3B, there was no evidence of lymph node metastases in a dasatinib-treated mouse with a similar size primary tumor to that of metastatic control tumors. To more rigorously assess whether inhibition of lymph node metastases in dasatinib-treated group resulted from decreased primary tumor growth, the same orthotopic prostate tumor model was established by injecting PC-3MM2GL cells (1 × 105) into the prostates of nude mice (n = 18). Mice in the control group were sacrificed after 25-day treatment of diluent buffer only. All mice developed both primary tumors (median weight, 1.13 g) and lymph node metastases (Table 2). In the dasatinib-treated group (n = 6; once daily, 15 mg/kg/d by p.o. gavage), mice were treated for 35 days (compared with 28-day treatment in the previous experiments) to allow the primary tumor to grow to similar size as in the control group. This strategy was successful, as tumor weights of the extended dasatinib treatment group were not statistically different from tumors in the control animals (P = 0.240, Mann-Whitney U test). However, in sharp contrast, dasatinib-treated mice were significantly reduced in the incidence of lymph node metastases (Table 2; P = 0.015, Fisher's exact test). Moreover, in the untreated control group sacrificed at only 16 days and possessing a median tumor weight of 0.32 g, lymph node metastases developed in six of six animals (Table 2). Therefore, dasatinib treatment seems to have independent effects on tumor cell proliferation and development of lymph node metastases.

Table 2.

Primary tumor size does not affect development of lymph node metastases

TreatmentTreatment time (d)Tumor incidenceTumor weight (g)
LN metastases Incidence*
MedianIQR
Control 16 6/6 0.32 (0.40–0.30) 6/6 
 25 6/6 1.13§ (1.45–0.70) 6/6 
Dasatinib 35 6/6 0.97§ (1.35–0.97) 1/6 
TreatmentTreatment time (d)Tumor incidenceTumor weight (g)
LN metastases Incidence*
MedianIQR
Control 16 6/6 0.32 (0.40–0.30) 6/6 
 25 6/6 1.13§ (1.45–0.70) 6/6 
Dasatinib 35 6/6 0.97§ (1.35–0.97) 1/6 
*

Incidence of lymph node metastases was determined by identifying solid, opaque, and enlarged iliac lymph node(s).

Interquartile range represents a range of 75th percentile and 25th percentile of the tumor weight in the group.

Lymph node metastases were observed in all control diluent-treated mice carrying reduced tumor weight (median tumor weight, 0.32 g).

§

Median tumor weight in the control group (treated for 25 d) and in the dasatinib-treated group (treated for 35 d) were not statistically different (P = 0.240, by Mann-Whitney U test).

Dasatinib-treated mice group had statistically reduced incidence of lymph node metastases (P = 0.015, by Fisher's exact test) compared with the control-treated group.

Dasatinib treatment decreases expression of autophosphorylated SFKs and FAK-phospho-[Y861] in tumors. To determine whether dasatinib effectively inhibited activation of SFKs in the primary tumors in vivo, tumor tissues were harvested, fixed, and prepared for immunohistochemical analysis as described in Materials and Methods. As represented in Fig. 4A, dasatinib treatment effectively inhibited expression of SFK phosphorylation (the anti–Src-phospho-[Y416] antibody detects autophosphorylation of both Src and Lyn; Fig. 4A,, middle), whereas the treatment had minimal effects on expression of total Src (Fig. 4A,, right) or total Lyn (Fig. 4B). Next, the downstream Src substrate FAK was analyzed for activity in the tumor tissues. Dasatinib treatment reduced expression of FAK-phospho-[Y861] (a Src phosphorylation site), whereas total FAK levels were not changed significantly (Fig. 4C). Staining intensities specific for phospho-SFKs and total Src, respectively, were quantified as described in Materials and Methods. Dasatinib-treated tumor tissues showed significantly (P < 0.001, Student's t test) reduced ratios of activated SFKs to total Src and phospho-[Y861]-FAK to total FAK.

Figure 4.

Immunohistochemistry of tumor tissue for Src, autophosphorylated SFKs, Lyn, FAK, and FAK phospho-[Y861]. Three tumors that weighed close to the median tumor weight in each group were selected for immunohistochemical staining. Serially sectioned slides were stained for H&E, autophosphorylated SFKs, and total Src. A, area-matched images are represented in the panel. Staining intensities were quantified to calculate phosphorylated SFKs to total Src ratios (0.28 in dasatinib-treated group versus 0.83 in control group). B, staining for total Lyn was performed on identical specimens. C, staining for FAK and FAK phospho-[Y861]. Phosphorylated FAK to total FAK ratios were 1.44 in control group and 0.23 in treated group. Magnification for all images, 100×. D, tumor tissues were analyzed for expression of Lyn-phospho-[Y397] or Src-phospho-[Y419] by immunoprecipitation and immunoblotting. Three tumor tissues from dasatinib-treated or control-treated groups were lysed. Whole tissue lysates (1 mg of total protein) were immunoprecipitated with mouse anti-Src or mouse anti-Lyn antibodies, then immunoblotted with rabbit anti–Src-phospho-[Y416] antibody. Membranes were reprobed with anti-Src or anti-Lyn antibodies to show equal loading.

Figure 4.

Immunohistochemistry of tumor tissue for Src, autophosphorylated SFKs, Lyn, FAK, and FAK phospho-[Y861]. Three tumors that weighed close to the median tumor weight in each group were selected for immunohistochemical staining. Serially sectioned slides were stained for H&E, autophosphorylated SFKs, and total Src. A, area-matched images are represented in the panel. Staining intensities were quantified to calculate phosphorylated SFKs to total Src ratios (0.28 in dasatinib-treated group versus 0.83 in control group). B, staining for total Lyn was performed on identical specimens. C, staining for FAK and FAK phospho-[Y861]. Phosphorylated FAK to total FAK ratios were 1.44 in control group and 0.23 in treated group. Magnification for all images, 100×. D, tumor tissues were analyzed for expression of Lyn-phospho-[Y397] or Src-phospho-[Y419] by immunoprecipitation and immunoblotting. Three tumor tissues from dasatinib-treated or control-treated groups were lysed. Whole tissue lysates (1 mg of total protein) were immunoprecipitated with mouse anti-Src or mouse anti-Lyn antibodies, then immunoblotted with rabbit anti–Src-phospho-[Y416] antibody. Membranes were reprobed with anti-Src or anti-Lyn antibodies to show equal loading.

Close modal

Dasatinib administration reduces number of proliferating cells in tumor tissues and increases number of apoptotic tumor cells. As shown in Fig. 2B and C, dasatinib had antiproliferative effects on human prostate cancer cells in vitro; thus, we determined the effects of dasatinib on tumor cell proliferation in vivo by proliferating cell nuclear antigen staining. As shown in Fig. 5A, dasatinib treatment significantly decreased percentage of PCNA-positive tumor cells, compared with the control treatment (P = 0.007, Student's t test).

Figure 5.

Immunohistochemistry of tumor tissue for CD31, TUNEL, and PCNA. A, frozen tumor sections were stained for PCNA. Brightfield microscope images (magnification 40×, area 0.89 mm2) were photographed and analyzed to quantify PCNA-positive cell percentage. Columns, average percentage of PCNA-positive cells counted in three randomly selected fields in three tumor samples from each group; error bars, SD. B, tumor sections were analyzed for percentage of apoptotic cells by TUNEL assay. Nuclei were counterstained with Hoechst 33342 (blue stain). Fluorescent microscope images (magnification 40×, area 1.14 mm2) were taken and analyzed to quantify TUNEL-positive cell (green stain) number. Columns, average number of TUNEL-positive cells counted in three randomly selected fields in three tumor samples from each group; error bars, SD. Scale bars, 100 μm. C, frozen tumor sections were stained for mouse CD31 to quantify microvessel density of tumor tissues. Brightfield microscope images (magnification 40×, area 0.89 mm2) were taken, and positively stained microvessels were counted. Columns, average numbers of microvessels in 0.89 mm2 field; error bars, SD. Representative images (magnification 80×, area 0.225 mm2) were selected.

Figure 5.

Immunohistochemistry of tumor tissue for CD31, TUNEL, and PCNA. A, frozen tumor sections were stained for PCNA. Brightfield microscope images (magnification 40×, area 0.89 mm2) were photographed and analyzed to quantify PCNA-positive cell percentage. Columns, average percentage of PCNA-positive cells counted in three randomly selected fields in three tumor samples from each group; error bars, SD. B, tumor sections were analyzed for percentage of apoptotic cells by TUNEL assay. Nuclei were counterstained with Hoechst 33342 (blue stain). Fluorescent microscope images (magnification 40×, area 1.14 mm2) were taken and analyzed to quantify TUNEL-positive cell (green stain) number. Columns, average number of TUNEL-positive cells counted in three randomly selected fields in three tumor samples from each group; error bars, SD. Scale bars, 100 μm. C, frozen tumor sections were stained for mouse CD31 to quantify microvessel density of tumor tissues. Brightfield microscope images (magnification 40×, area 0.89 mm2) were taken, and positively stained microvessels were counted. Columns, average numbers of microvessels in 0.89 mm2 field; error bars, SD. Representative images (magnification 80×, area 0.225 mm2) were selected.

Close modal

To determine whether tumor growth inhibition of dasatinib treatment could also be attributed to inhibition of angiogenesis, microvessel densities were measured by counting CD31/PECAM-positive microvessels. Microvessel densities between dasatinib-treated and control groups were not significantly changed (Fig. 5C, P = 0.58, Student's t test). In addition, we further determined the number of apoptotic endothelial cells by CD31/TUNEL costaining of frozen tumor tissue samples, and there was no significant difference between groups (data not shown). However, we could observe a significant increase of apoptotic tumor cells in the dasatinib-treated group (P < 0.001, Student's t test; Fig. 5B). In summary, our results show that dasatinib treatment of mice bearing prostate tumors results both in decreased growth rate and decreased metastatic potential to the lymph nodes.

A number of genetic and epigenetic events affecting tyrosine kinase expression occur during progression of many solid tumors. In prostate cancer, overexpression of HER-1, HER-2, and HER-4 (3436), c-Met (31, 37), FAK (38), and mutations in c-Kit (39) are relatively common occurrences as prostate tumors develop and progress. In turn, each of these protein tyrosine kinases signals through nonreceptor protein tyrosine kinases of the Src family, resulting in deregulation of many important tumorigenic phenotypes, including proliferation, invasion, migration, epithelial-mesenchymal transition (40), apoptosis, survival, angiogenesis, etc. Thus, activity of SFKs increases in progressive stages of tumors, with highest activity in metastatic lesions, including prostate cancer.7

7

G.E. Gallick and N.U. Parikh, unpublished data.

Increasing evidence from molecular and pharmacologic approaches suggests that inhibition of Src, the prototype SFK member, inhibits tumor functions associated with metastasis, including migration, invasion, and expression of the proangiogenic molecules, such as interleukin-8 (28, 41) and vascular endothelial growth factor (42, 43). In addition, recent studies indicate that Src plays critical roles in host cells in the tumor microenvironment, as well as in the tumor cells that contribute to metastasis (9). Several studies have shown that Src-mediated phosphorylation of VE-cadherin, a cell adhesion molecule that is essential to the vascular cell-to-cell junctional integrity, directly leads to increased vascular permeability, thus facilitating intravasation and extravasation of migratory tumor cells (44, 45). For these reasons, selective inhibitors of SFKs may be promising drugs for cancer therapy, and a large number of clinical trials to test their efficacy are now under way (22). However, whether SFK activity functions primarily to promote tumor progression or metastasis or contributes to proliferation of the tumor at the primary site remains unclear in some tumors, especially in prostate cancer. In addition, whether specific SFK members play overlapping or distinct roles in tumor growth has not been studied.

Prostate cancer is unique in that, besides the expected inhibition of Src activity affecting migration and invasion of cultured cells (15), a second Src family member, Lyn, has been shown to contribute to normal prostate development and tumor growth; and inhibition of Lyn decreases growth of primary tumors (19). Thus, inhibitors of the SFK family, such as dasatinib, might be predicted to have greater efficacy in prostate cancer than in other tumors in which activation of Src alone predominates.

In this study, we provide the first direct experimental evidence that dasatinib has efficacy in a preclinical model of orthotopic growth of human prostate cells in nude mice and does so by inhibiting both tumor cell proliferation at the primary site, as well as the development of lymph node metastases. In contrast, previous studies from this laboratory have shown that, whereas dasatinib has modest effect on growth inhibition of primary pancreatic tumors in an orthotopic nude mouse model (11), these effects were not due to changes in proliferation at the growing portion of the tumor and the major effect was in inhibition of development of lymph node and hepatic metastases. Others also showed in vitro that Src activation does not contribute to proliferation of colon tumor cells (46). In the current study, we show that in prostate cancer, like many other tumors, dasatinib inhibits development or outgrowth of metastases. This property is especially evident as small tumors in untreated mice universally metastasize, whereas few metastases are found associated with much larger tumors in dasatinib-treated animals. As decreased expression of Src in prostate tumor cells in vitro by siRNA has little effect on proliferation, Src activation may be playing a similar role in prostate cancer as it is in other tumors.

However, the SFK Lyn has been shown to contribute to normal prostate development and tumor growth, and inhibition of Lyn decreases growth of primary tumors (19). Thus, Lyn seems to affect processes very distinct from Src. Also in contradistinction from Src, Lyn is not expressed in endothelial cells and thus is not directly activated by vascular endothelial growth factor. Therefore, the striking reduction in tumor cell proliferation, as observed by PCNA staining (Fig. 5A) and knockdown of Lyn in cultured prostate cells in vitro is thus far unique to prostate tumor cells in preclinical models, suggesting that Lyn affects proliferation; Src affects migration and invasion. Alternative explanations might account for the growth inhibitory effects of dasatinib. As no inhibitor is perfectly specific, there may be additional targets of the drug in prostate cancer that are not present or not critical to growth of other tumor types. Indeed, dasatinib does show some activity against c-Kit, which is mutated in a subset of prostate cancers (39), and the Tec family member Bmx (47), which plays a role downstream of Src and FAK in mediating neurotrophic-mediated growth of human prostate cancer cells (48). We note that both Kit and Bmx function through Src; thus, the siRNA experiments demonstrating effects of Lyn and Src offer a compelling case that these kinases are important targets of dasatinib.

Finally, studies with dasatinib in combination with other agents are now reaching the clinic for metastatic prostate cancer (22). As yet, it cannot be predicted if inhibition of Src will have effects on preexisting metastases, but the effects shown in this study on prostate tumor proliferation are encouraging in this regard. Longstanding studies have shown that Src activity is critical in osteoclast function; thus, numerous investigators have recognized the potential for Src inhibitors in the treatment of bone metastases. Further studies will be required to understand if additional SFKs contribute to bone metastasis and how Src inhibitors might best be used in treatment of this stage of the disease to which patients succumb.

Grant support: National Cancer Institute grant CA-16672 (G.E. Gallick), Lockton Foundation (G.E. Gallick), Gillson Longenbaugh Foundation (G.E. Gallick), NIH grant T32 CA-09599 (K.A. Phillips), and Sowell-Huggins Fellowship (J. Zhang). G.E. Gallick is a Sowell-Huggins Professor of Cancer Biology.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

We thank Dr. Christopher J. Logothetis for his critical review and comments on this manuscript; Nila U. Parikh and Donna M. Reynolds for all their excellent technical assistance and help with immunohistochemistry; Marjorie Johnson for her help in Src-knockdown cells; and Dr. Eugenie S. Kleinerman for providing Lyn-targeted siRNA expression vector.

1
Jemal A, Siegel R, Ward E, Murray T, Xu JQ, Thun MJ. Cancer statistics, 2007.
CA Cancer J Clin
2007
;
57
:
43
–66.
2
Bianco FJ, Scardino PT, Eastham JA. Radical prostatectomy: long-term cancer control and recovery of sexual and urinary function (“trifecta”).
Urology
2005
;
66
:
83
–94.
3
Cheng L, Zincke H, Blute ML, Bergstralh EJ, Scherer D, Bostwick DG. Risk of prostate carcinoma death in patients with lymph node metastasis.
Cancer Cytopathol
2001
;
91
:
66
–73.
4
Daneshmand S, Quek ML, Stein JP, et al. Prognosis of patients with lymph node positive prostate cancer following radical prostatectomy: long-term results.
J Urol
2004
;
172
:
2252
–5.
5
Masterson TA, Bianco FJ, Vickers AJ, et al. The association between total and positive lymph node counts, and disease progression in clinically localized prostate cancer.
J Urol
2006
;
175
:
1320
–4.
6
Bader P, Burkhard FC, Markwalder R, Studer UE. Disease progression and survival of patients with positive lymph nodes after radical prostatectomy. Is there a chance of cure?
J Urol
2003
;
169
:
849
–54.
7
Heidenreich A, Varga Z, Von Knobloch R. Extended pelvic lymphadenectomy in patients undergoing radical prostatectomy: high incidence of lymph node metastasis.
J Urol
2002
;
167
:
1681
–6.
8
Summy JM, Gallick GE. Src family kinases in tumor progression and metastasis.
Cancer Metastasis Rev
2003
;
22
:
337
–58.
9
Park SI, Shah AN, Zhang J, Gallick GE. Regulation of angiogenesis and vascular permeability by Src family kinases: opportunities for therapeutic treatment of solid tumors.
Expert Opin Ther Targets
2007
;
11
:
1207
–17.
10
Ito H, Gardner-Thorpe J, Zinner MJ, Ashley SW, Whang EE. Inhibition of tyrosine kinase Src suppresses pancreatic cancer invasiveness.
Surgery
2003
;
134
:
221
–6.
11
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
–72.
12
Cartwright CA, Meisler AI, Eckhart W. Activation of the Pp60C-Src protein-kinase is an early event in colonic carcinogenesis.
Proc Natl Acad Sci U S A
1990
;
87
:
558
–62.
13
Talamonti MS, Roh MS, Curley SA, Gallick GE. Increase in activity and level of Pp60C-Src in progressive stages of human colorectal-cancer.
J Clin Invest
1993
;
91
:
53
–60.
14
Lombardo LJ, Lee FY, Chen P, et al. Discovery of N-(2-chloro-6-methyl-phenyl)-2-(6-(4-(2-hydroxyethyl)-piperazin-1-yl)-2-methylpyrimidin-4-ylamino)thiazole-5-carboxamide (BMS-354825), a dual Src/Abl kinase inhibitor with potent antitumor activity in preclinical assays.
J Med Chem
2004
;
47
:
6658
–61.
15
Nam S, Kim DW, Cheng JQ, et al. Action of the Src family kinase inhibitor, dasatinib (BMS-354825), on human prostate cancer cells.
Cancer Res
2005
;
65
:
9185
–9.
16
Brabek J, Constancio SS, Siesser PF, Shin NY, Pozzi A, Hanks SK. Crk-associated substrate tyrosine phosphorylation sites are critical for invasion and metastasis of SRC-transformed cells.
Mol Cancer Res
2005
;
3
:
307
–15.
17
Parsons JT. Focal adhesion kinase: the first ten years.
J Cell Sci
2003
;
116
:
1409
–16.
18
Hibbs ML, Tarlinton DM, Armes J, et al. Multiple defects in the immune-system of lyn-deficient mice, culminating in autoimmune-disease.
Cell
1995
;
83
:
301
–11.
19
Goldenberg-Furmanov M, Stein I, Pikarsky E, et al. Lyn is a target gene for prostate cancer: sequence-based inhibition induces regression of human tumor xenografts.
Cancer Res
2004
;
64
:
1058
–66.
20
Shah NP, Tran C, Lee FY, Chen P, Norris D, Sawyers CL. Overriding imatinib resistance with a novel ABL kinase inhibitor.
Science
2004
;
305
:
399
–401.
21
Summy JM, Gallick GE. Treatment for advanced tumors: Src reclaims center stage.
Clin Cancer Res
2006
;
12
:
1398
–401.
22
Kopetz S, Shah AN, Gallick GE. Src continues aging: current and future clinical directions.
Clin Cancer Res
2007
;
13
:
7232
–6.
23
Kim SJ, Uehara H, Karashima T, Shepherd DL, Killion JJ, Fidler IJ. Blockade of epidermal growth factor receptor signaling in tumor cells and tumor-associated endothelial cells for therapy of androgen-independent human prostate cancer growing in the bone of nude mice.
Clin Cancer Res
2003
;
9
:
1200
–10.
24
Pettaway CA, Pathak S, Greene G, et al. Selection of highly metastatic variants of different human prostatic carcinomas using orthotopic implantation in nude mice.
Clin Cancer Res
1996
;
2
:
1627
–36.
25
Ory DS, Neugeboren BA, Mulligan RC. A stable human-derived packaging cell line for production of high titer retrovirus/vesicular stomatitis virus G pseudotypes.
Proc Natl Acad Sci U S A
1996
;
93
:
11400
–6.
26
Zhou LR, Wu H, Lee P, Wang ZX. Roles of the androgen receptor cofactor p44 in the growth of prostate epithelial cells.
J Mol Endocrinol
2006
;
37
:
283
–300.
27
Summy JM, Trevino JG, Lesslie DP, et al. AP23846, a novel and highly potent Src family kinase inhibitor, reduces vascular endothelial growth factor and interleukin-8 expression in human solid tumor cell lines and abrogates downstream angiogenic processes.
Mol Cancer Ther
2005
;
4
:
1900
–11.
28
Trevino JG, Summy JM, Gray MJ, et al. Expression and activity of Src regulate interleukin-8 expression in pancreatic adenocarcinoma cells: implications for anglogenesis.
Cancer Res
2005
;
65
:
7214
–22.
29
Lesslie DP, Summy JM, Parikh NU, et al. Vascular endothelial growth factor receptor-1 mediates migration of human colorectal carcinoma cells by activation of Src family kinases.
Br J Cancer
2006
;
94
:
1710
–7.
30
Ding Q, Stewart J, Olman MA, Klobe MR, Gladson CL. The pattern of enhancement of Src kinase activity on platelet-derived growth factor stimulation of glioblastoma cells is affected by the integrin engaged.
J Biol Chem
2003
;
278
:
39882
–91.
31
Kim SJ, Johnson M, Koterba K, Herynk MH, Uehara H, Gallick GE. Reduced c-Met expression by an adenovirus expressing a c-Met ribozyme inhibits tumorigenic growth and lymph node metastases of PC3–4 prostate tumor cells in an orthotopic nude mouse model.
Clin Cancer Res
2003
;
9
:
5161
–70.
32
Ruifrok AC, Johnston DA. Quantification of histochemical staining by color deconvolution.
Anal Quant Cytol Histol
2001
;
23
:
291
–9.
33
Landen CN, Lu C, Han LY, et al. Efficacy and antivascular effects of EphA2 reduction with an agonistic antibody in ovarian cancer.
J Natl Cancer Inst
2006
;
98
:
1558
–70.
34
Ben Yosef R, Starr A, Karaush V, et al. ErbB-4 may control behavior of prostate cancer cells and serve as a target for molecular therapy.
Prostate
2007
;
67
:
871
–80.
35
Di Lorenzo G, Tortora G, D'Armiento FP, et al. Expression of epidermal growth factor receptor correlates with disease relapse and progression to androgen-independence in human prostate cancer.
Clin Cancer Res
2002
;
8
:
3438
–44.
36
Scher HI, Sarkis A, Reuter V, et al. Changing pattern of expression of the epidermal growth factor receptor and transforming growth factor α in the progression of prostatic neoplasms.
Clin Cancer Res
1995
;
1
:
545
–50.
37
Pisters LL, Troncoso P, Zhau HE, Li W, Voneschenbach AC, Chung LWK. C-met protooncogene expression in benign and malignant human prostate tissues.
J Urol
1995
;
154
:
293
–8.
38
Slack JK, Adams RB, Rovin JD, Bissonette EA, Stoker CE, Parsons JT. Alterations in the focal adhesion kinase/Src signal transduction pathway correlate with increased migratory capacity of prostate carcinoma cells.
Oncogene
2001
;
20
:
1152
–63.
39
Paronetto MP, Farini D, Sammarco I, et al. Expression of a truncated form of the c-kit tyrosine kinase receptor and activation of Src kinase in human prostatic cancer.
Am J Pathol
2004
;
164
:
1243
–51.
40
Shah AN, Gallick GE. Src, chemoresistance and epithelial to mesenchymal transition: are they related?
Anticancer Drug
2007
;
18
:
371
–5.
41
Trevino JG, Gray MJ, Nawrocki ST, et al. Src activation of Stat3 is an independent requirement from NF-κB activation for constitutive IL-8 expression in human pancreatic adenocarcinoma cells.
Angiogenesis
2006
;
9
:
101
–10.
42
Ellis LM, Staley CA, Liu W, et al. Down-regulation of vascular endothelial growth factor in a human colon carcinoma cell line transfected with an antisense expression vector specific for c-src.
J Biol Chem
1998
;
273
:
1052
–7.
43
Gray MJ, Zhang J, Ellis LM, et al. HIF-1 α, STAT3, CBP/p300 and Ref-1/APE are components of a transcriptional complexthat regulates Src-dependent hypoxia-induced expression of VEGF in pancreatic and prostate carcinomas.
Oncogene
2005
;
24
:
3110
–20.
44
Criscuoli ML, Nguyen M, Eliceiri BP. Tumor metastasis but not tumor growth is dependent on Src-mediated vascular permeability.
Blood
2005
;
105
:
1508
–14.
45
Weis S, Shintani S, Weber A, et al. Src blockade stabilizes a Flk/cadherin complex, reducing edema and tissue injury following myocardial infarction.
J Clin Invest
2004
;
113
:
885
–94.
46
Brunton VG, Ozanne BW, Paraskeva C, Frame MC. A role for epidermal growth factor receptor, c-Src and focal adhesion kinase in an in vitro model for the progression of colon cancer.
Oncogene
1997
;
14
:
283
–93.
47
Hantschel O, Rix U, Schmidt U, et al. The Btk tyrosine kinase is a major target of the Bcr-Abi inhibitor dasatinib.
Proc Natl Acad Sci U S A
2007
;
104
:
13283
–8.
48
Lee LF, Guan JL, Qiu Y, Kung HJ. Neuropeptide-induced androgen independence in prostate cancer cells: roles of nonreceptor tyrosine kinases Etk/Bmx, Src, and focal adhesion kinase.
Mol Cell Biol
2001
;
21
:
8385
–97.