Heregulin-induced activation of ErbB3 by EGFR ... - Wiley Online Library

Download PDF
15 downloads 5534 Views 464KB Size Report
Comment
Apr 8, 2008 - the foot pad were amputated at a similar size. These results .... tumor mass were measured using a digital gauge (Mitsutoyo,. Kanagawa ..... gene signature and clinical outcome in non-small-cell lung cancer. N. Engl J Med ...
Int. J. Cancer: 123, 340–347 (2008) ' 2008 Wiley-Liss, Inc.

Heregulin-induced activation of ErbB3 by EGFR tyrosine kinase activity promotes tumor growth and metastasis in melanoma cells Yoko Ueno1, Hiroaki Sakurai1,2*, Satoshi Tsunoda1, Min-Kyung Choo1,2, Mitsuhiro Matsuo1, Keiichi Koizumi1 and Ikuo Saiki1,2 1 Division of Pathogenic Biochemistry, Institute of Natural Medicine, University of Toyama, Toyama, Japan 2 21st century COE Program, University of Toyama, Toyama, Japan ErbB3 receptor tyrosine kinase has been shown to induce tumor progression in several types of cancer through heterodimerization with ErbB2. However, the role of ErbB3 and its ligand heregulin (HRG) in tumor metastasis remains poorly understood. In the present study, we tried to clarify their contributions to the metastasis of ErbB3-overexpressing B16-BL6 melanoma cells. Stimulation with HRG induced phosphorylation of ErbB3 and metastatic properties including MMP-9 expression, invasion, adhesion and experimental lung metastasis in vivo. These cellular responses were blocked by inhibiting the tyrosine kinase activity of EGFR with PD153035. In addition, phosphorylation of EGFR was rapidly induced by HRG, suggesting that EGFR is a possible heterodimeric counterpart of ErbB3. RNA interference demonstrated that subcutaneous tumor growth and angiogenesis was attenuated by inactivation of ErbB3 in cancer cells. Although experimental pulmonary metastasis was not affected by the knockdown of ErbB3, spontaneous metastasis was, even when primary tumors in the foot pad were amputated at a similar size. These results indicate that HRG-induced activation of ErbB3 via EGFR promotes tumor growth and metastasis of melanoma cells. ' 2008 Wiley-Liss, Inc. Key words: ErbB3; EGFR; heregulin; metastasis; melanoma

The receptor tyrosine kinase (RTK) family of ErbBs forms part of a complex signal cascade modulating cell proliferation, survival and adhesion. The family comprises 4 homologous receptors, EGFR, ErbB2, ErbB3 and ErbB4, which form homodimers or heterodimers to initiate intracellular signaling in response to their ligands. EGFR and ErbB2 have recently been focused on the molecular targeted therapy of cancer, because overexpression, amplification and mutations are involved in carcinogenesis and the progression of several types of cancer including metastasis.1–3 Tyrosine kinase inhibitors and neutralizing monoclonal antibodies against these receptors are clinically approved for the treatment of nonsmall cell lung cancer (NSCLC) and breast cancer. In addition to tumorigenesis, these receptors have also been shown to contribute to metastasis.4,5 In spite of no high affinity ligand for ErbB2 and the inactive kinase domain of ErbB3, the ErbB2/ErbB3 heterodimer is believed to be the most biologically active and tumorigenic of the possible heterodimer complexes of ErbB3.6–8 It has been reported that 20– 30% of human breast cancers overexpress ErbB2.9 ErbB3 expression is also reported to be closely associated with relapse-free and overall survival and is associated with a high risk of metastasis among patients with NSCLC.10 In breast cancer ErbB3-dependent signaling through the ErbB3/ErbB2 heterodimer contributes to metastasis by enhancing tumor cell invasion and intravasation.11 In comparison to the ErbB2/ErbB3 heterodimer, little is known about the tumorigenic and metastatic functions of the EGFR/ ErbB3 heterodimer. In addition, while the ErbB3 in other cancers is being studied, the role of ErbB3 in the metastasis of melanoma cells remains to be characterized. There are many different types of ligands for ErbB family receptors. Some ligands such as epidermal growth factor (EGF) specifically binds to the EGFR, whereas betacellulin (BTC) binds to both EGFR and ErbB4.12 The ligands for ErbB3 are different in that there are several splice variants. HRG1-b1 is the most studied ligand of the HRG family. The binding of HRG to ErbB3 triggers Publication of the International Union Against Cancer

heterodimerization with other ErbB receptors such as ErbB2 and EGFR, which results in effective tyrosine phosphorylation of the ErbB3.13 It has been reported that the blockage of HRG expression inhibits the tumorigenicity and metastasis of human breast cancer.14,15 In the present study, we have found ErbB3 to be overexpressed in metastatic B16-BL6 melanoma cells. While most studies focusing on ErbB3 have concluded that ErbB2 is its heterodimeric partner, EGFR is in fact responsible for the activation of ErbB3 in response to HRG in melanoma cells. In addition, the ErbB3/EGFR activation participates in tumor growth and metastasis in vivo. The targeting of ErbB3 and EGFR may be an effective treatment against melanomas with high ErbB3 levels. Material and methods Antibodies and reagents Antibodies against EGFR (No. 2232), phospho-EGFR (Tyr845; No. 2231), ErbB2 (No. 2242), phospho-ErbB2 (No. 2241), phospho-ErbB3 (No. 4791), phosphoextracellular signal-regulated kinase (ERK) (Thr202 and Tyr204; No. 9101), phospho-Akt (Ser473; No. 9271) and phospho-JNK (Thr183 and Tyr185; No. 9251) were purchased from Cell Signaling Technology (Danvers, MA). Anti-EGFR (sc-03), ErbB3 (sc-285), Akt (sc-1618) and antiPECAM-1 (sc-1506) antibodies were obtained from Santa Cruz Biotechnology (Santa Cruz, CA). Recombinant murine EGF was purchased from Upstate Biotechnology (Billerica, MA). Human HRG-1-b1 (HRG) and murine BTC were from R&D systems (Minneapolis, MN). The ErbB tyrosine kinase inhibitors PD153035 and AG825 were purchased from Calbiochem (Darmastadt, Germany). Cell culture Murine metastatic cell lines, including B16-BL6, Lewis lung carcinoma (LLC) and QR-32, were maintained in EMEM. Murine hepatocellular carcinoma CBO140C12 and Colon 38 cells were maintained in DMEM:F-12. Murine mammary carcinoma 4T1 cells were maintained in DMEM 1 10 mM HEPES. All media contained 8–10% FCS, 320 lg/l of L-glutamine, 100 U/ml of penicillin and 100 lg/ml of streptomycin, and cultures were kept at 37°C in a humidified atmosphere of 5% CO2/95% air. For immunoblotting assay they were cultured in medium containing 0.1% BSA. RNA interference A double stranded oligonucleotide was cloned into pSUPER.gfp/neo (OligoEngine, Seattle, WA) for the expression of short hairpin RNAs (shRNA). The sense strand sequence used was Grant sponsor: Ministry of Education, Culture, Sports, Science and Technology, Japan. *Correspondence to: Division of Pathogenic Biochemistry, Institute of Natural Medicine, University of Toyama, 2630 Sugitani, Toyama 9300194, Japan. Fax: 181-76-434-5058. E-mail: [email protected] Received 14 September 2007; Accepted after revision 10 January 2008 DOI 10.1002/ijc.23465 Published online 8 April 2008 in Wiley InterScience (www.interscience. wiley.com).

HEREGULIN-INDUCED METASTASIS VIA ErbB3/EGFR 0

5 -GATCCCCCAGTCTGCATTAAAGTCATCGAGGATTCAA GAGATCCTCGATGACTTTAATGCAGACTGTTTTTA-30 for ErbB3 and 50 -GATCCCCCATCACGTACGCGGAATACTTCA AGAGAGTATTCCGCGTACGTGATGTTTTTA-30 for firefly GL2 luciferase. B16-BL6 cells were transfected with the vectors using LipofectAMINE with Plus reagent (Invitrogen Life Technologies, Carlsbad, CA). Stable transfectants (sh-ErbB3 and shLuc cells) were selected with geneticin at 1 mg/ml (GIBCO, Carlsbad, CA). EGFP expression was confirmed by FACS and immunoblotting. Immunoblotting and immunoprecipitation Immunoprecipitation was performed as previously described.16 Briefly, cell lysates were incubated with anti-EGFR antibody (sc-03) for 2 hr and then rotated with Protein G-Sepharose at 4°C overnight. Samples were subjected to immunoblotting after washing. Immunoblotting was performed as previously described.17 Briefly cell lysates were subjected to SDS-PAGE and transferred to Immobilon-P membranes (Millipore, Billerica, MA). Blots were probed using the primary antibodies described above and horseradish peroxidase-conjugated secondary antibodies (DAKO, Glostrup, Denmark) visualized with the ECL system (Amersham Biosciences, Buckinghamshire, England). Reverse transcription-polymerase chain reaction and real-time polymerase chain reaction Reverse transcription-polymerase chain reaction (RT-PCR) was performed as previously described.17 The sequences of the primers were as follows: MMP-9, 50 -TTCTCTGGACGTCAAATGTGG-30 and 50 -CAAAGAAGGAGCCCTAGTTCAAGG-30 ; integrin b-1, 50 -TGCAGGTGTCGTGTTTGTGAATGC-30 and 50 -CAGCAGT CATCAATGTCCTTCTCC-30 ; ICAM-1, 50 -TTTTGCTCTGCC GCTCTGGAG-30 and 50 -TACACATTCCTGGTGACATTC-30 ; and GAPDH, 50 -GGTGAAGGTCGGTGTGAACGGATTT-30 and 50 -GATGCCAAAGTTGTCATGGATGACC-30 . Real-time RTPCR was performed as previously described.17 The sequences of the primers were as follows: VEGFA 50 -GTGCACTGGACCCT30 and 50 -GGTCTCAATCGGACGG-30 , GAPDH 50 -AAATGGTGAAGGTCGGTGTG-30 and 50 -TGAAGGGGTCGTTGATGG-30 . Gelatin zymography Gelatin zymography was performed as previously described17 with some modifications. Briefly, the conditioned media were concentrated using Centricon (Millipore) according to the manufacturer’s instructions and applied to 7.5% SDS-polyacrylamide gels copolymerized with gelatin (0.1% w/v) and incubated at 37°C for 24 hr. Enzyme-digested regions were quantified by ChemiDoc XRS (Bio-Rad). Invasion assay The invasion assay was performed using Transwell culture chambers (Corning Costar, Corning, NY) as previously described.17 The filter was precoated with 1 lg of fibronectin on the lower surface and 1 lg of Matrigel on the upper surface. Cells in 0.1% BSA-containing media were stimulated with HRG for 24 hr. Then, 2 3 104 cells were added to the upper compartment of the chamber and incubated for 6 hr. The cells stained with hematoxylin and eosin were counted under the microscope in 5 predetermined fields at a magnification of 4003. Adhesion assay The adhesion assay was performed as previously described17 with some modifications. Cells in 0.1% BSA medium were stimulated with HRG for 24 hr. The 4 3 104 cells were seeded in 96well plates precoated with 1 lg of fibronectin. After incubation for 15 min, attached cells were stained with 0.5% crystal violet. The cells were lysed with 30% acetic acid, and the absorbance at 590 nm was measured.

341

Cell proliferation assay For the 5-bromo-20 -deoxyuridine (BrdU) incorporation assay, 2 3 104 cells were seeded in 96-well plates and incubated in complete medium. The medium was changed to 0.1% BSA-containing medium and incubated for 24 hr. After a wash with PBS, cells were incubated with BrdU for another 2 hr. BrdU-labeled cells were detected using an enzyme-linked immunosorbent assay (ELISA)-based colorimetric kit (Roche, Indianapolis, IN) according to the procedure provided by the manufacturer. Animals Five-week-old specific pathogen-free female C57BL/6 mice for B16-BL6 and B6C3F1 mice for CBO140C12 were purchased from Japan SLC (Hamamatsu, Japan). The mice were maintained under specific pathogen-free conditions and used according to institutional guidelines. Preparation of tumor lysates B16-BL6 cells (5 3 105) or a CBO140C12 tumor fragment were subcutaneously inoculated into corresponding mice. Tumor tissues (20 mg) were homogenized in 1 ml of tissue lysis buffer (137 mM NaCl, 20 mM Tris-HCl (pH 7.5), 1% NP-40, 10% glycerol, 2 mM EDTA) using ultrasonic cell disruptor (Microson, NY). Protein concentration was equalized using DC protein assay kit (Bio-Rad, Hercules, CA). Samples were subjected to immunoblotting as described earlier. Metastasis models B16-BL6 cells were prestimulated with HRG in 0.1% BSA medium for 24 hr. Cells (4 3 104) suspended in PBS were inoculated intravenously. Fourteen days after the inoculation, mice were sacrificed and the metastatic colonies on the lung surface were counted macroscopically. For spontaneous metastasis, B16-BL6 cells (5 3 105) were inoculated into the right footpad using a Hamilton syringe and 25gauge needle, as described previously.18 The tumor volume was calculated using the following formulas, {(long length 3 short length 3 (thickness of left foot pad-thickness of right foot pad))/ 6} 3 3.14. The tumor mass in the footpad was excised surgically under anesthesia with ether. The number of metastatic colonies in the lungs was counted macroscopically on day 14 after the amputation. Tumor growth and angiogenesis in vivo B16-BL6 cells (5 3 105) were inoculated into subcutaneous tissue of the dorsal skin using a Hamilton syringe and 25-gauge needle as described previously.18 The large and short diameters of the tumor mass were measured using a digital gauge (Mitsutoyo, Kanagawa, Japan). The tumor volume was calculated using the formula; (large diameter 3 short diameter2)/2. For local administration of siRNA, when the tumors had reached an average diameter of 3–6 mm (on day 4), the tumor-bearing mouse was injected with 100 ll of siRNA solution with atelocollagen (AtelogeneTM, Koken, Tokyo, Japan). The final concentration of atelocollagen solution was 50% and that of siRNA, 10 lM. The siRNA sequences used were as follows: ErbB3, 50 -CAGUCUGCAUUAAA GUCAUCGAGGA-30 and Luc, 50 -CGUACGCGGAAUACU UCGA-30 . For the angiogenesis assay, the mice were infused with 0.1% evans blue (Wako, Osaka, Japan) and 0.5% gelatin (Wako, Osaka, Japan) into the left ventricle on day 7. The dorsal skin was collected and the vessels entering the tumor mass were counted macroscopically. Analysis of microvessel density Subcutaneously inoculated B16-BL6 tumors were paraffin-embedded after fixed in 10% formalin solution and stained as previously described.18 Briefly, sections were immunohistochemically

342

UENO ET AL.

FIGURE 2 – HRG signaling via ErbB3 enhances invasion and adhesion in vitro. (a) B16-BL6 cells were treated with HRG for the periods indicated, and then mRNA expression of MMP-9, integrin b1 and ICAM-1 was analyzed by RT-PCR. (b) The conditioned media of HRG-treated cells were analyzed for MMP2 and MMP9 activities using gelatin zymography. (c) Cell proliferation in vitro with or without HRG for 24 hr was determined by BrdU assay. (d and e) Effects of HRG on cell adhesion to fibronectin (d) and invasion (e) were determined. **p < 0.01.

FIGURE 1 – Overexpression of functional ErbB3 in B16-BL6 melanoma cells. (a) Murine metastatic cells incubated under the serumstarved conditions for 24 hr in vitro and subcutaneously inoculated tumors in vivo were analyzed by immunoblotting for the expression of the ErbB family. (b and c) Serum-starved B16-BL6 melanoma (b) or CBO140C12 hepatocellular carcinoma (c) cells were stimulated with EGF, BTC or HRG for 5 and 10 min. Cell lysates were immunoblotted with phosphospecific and control antibodies for ErbB3, ERK, JNK and Akt.

stained with anti-PECAM-1 using Histofine SAB-PO kit according to the manufacturer’s instructions (Nichirei, Tokyo, Japan). The high density of microvessels stained positively with PECAM1 was observed with a microscope at 4003 power and analyzed as described by Weidner with some modifications.18,19

Results Overexpression of functional ErbB3 in B16-BL6 melanoma cells To determine the functional importance of ErbBs in metastasis, we first examined the expression of ErbBs in mouse metastatic cell lines by immunoblotting. Consistent with our previous study,3 hepatocellular carcinoma CBO140C12 cells overexpressed EGFR

and ErbB2, but not ErbB3 (Fig. 1a). In contrast, B16-BL6 melanoma cells were characterized by high levels of ErbB3. Interestingly, ErbB2, a major heterodimeric partner of ErbB3 in many cancer cells, was not found to be expressed although EGFR was slightly expressed. Moreover, RT-PCR analysis confirmed a similar expression of ErbBs at the mRNA level (data not shown). Similarly, B16-BL6 tumors inoculated subcutaneously into the back skin expressed ErbB3 and EGFR (Fig. 1a). The different expression profile correlated with the response to their specific ligands. B16-BL6 cells, but not CBO140C12 cells, specifically responded to the ErbB3 ligand HRG, where phosphorylation of ErbB3 as well as ERK, JNK and Akt downstream was rapidly induced (Figs. 1b and 1c). In contrast, EGF, a ligand for EGFR, and BTC, a ligand capable of activating both EGFR and ErbB4, induced no obvious phosphorylation of MAPKs and Akt in B16-BL6 cells (Fig. 1b), although these ligands strongly induced activation of EGFR and subsequent phosphorylation of the ERK in CBO140C12 cells (Fig. 1c). These results imply that B16-BL6 cells were predominantly activated both in cultured cell and in tumor by HRG. HRG signaling via ErbB3 enhances invasion and adhesion in vitro We next investigated the biological significance of the HRGinduced activation of ErbB3 by focusing on metastatic properties such as invasion and adhesion in vitro. HRG is a mitogenic factor in breast cancer cells14,15,20; however, the proliferation of B16BL6 cells was not stimulated by HRG (Fig. 2c). To elucidate the metastasis-promoting activity of ErbB3, we examined the effect of HRG on the mRNA expression of metastasis-related genes. HRG induced the expression of several mRNAs such as MMP-9, integrin b1 and ICAM-1 (Fig. 2a). This correlated with the increase in MMP-9 enzyme activity, whereas MMP-2 activity was constitutively detected (Fig. 2b). Reflecting the inducible gene expression, HGR significantly enhanced adhesion and invasion in vitro (Figs. 2d and 2e). These results indicate that HRG stimulation of ErbB3 triggers the metastatic potential of B16-BL6 melanoma cells.

HEREGULIN-INDUCED METASTASIS VIA ErbB3/EGFR

343

FIGURE 3 – HRG-induced phosphorylation of ErbB3 is mediated by EGFR. (a and b) B16-BL6 cells were treated with PD153035 (0.01, 0.1 or 1 lM) and AG825 (30, 60 or 100 lM) for 30 min, followed by stimulation with HRG for 5 min. Phosphorylation of ErbB3, ERK and EGFR was determined by immunoblotting. (c) Cells were stimulated with HRG for 5 min and immunoprecipitated with anti-EGFR antibody. The immunoprecipitates were immunoblotted with antiphospho-ErbB3, ErbB3 and EGFR antibodies. (d) Cells were pretreated with PD153035 (1 lM) and AG825 (100 lM) for 30 min, then stimulated with HRG for 24 hr. Adhesion to fibronectin was determined as described in Figure 2e. (e) Cells were pretreated with PD153035 and AG825 for 30 min, and then stimulated with HRG for 24 hr. Metastatic colonies in the lung were counted on day 14 after intravenous inoculation of the cells. Data represents the mean 6 SD for 8 mice. **p < 0.01.

HRG-induced phosphorylation of ErbB3 is mediated by EGFR in melanomas Because ErbB3 lacks intrinsic protein tyrosine kinase activity within the intracellular domain, heterodimerization with other ErbBs is essential to initiate intracellular signaling. Therefore, we next clarified the heterodimeric counterpart of ErbB3 in response to HRG. Of the many possible heterodimeric complexes, EerbB2/ ErbB3 is believed to be the most biologically active and protumorigenic.17 But, as shown in Figure 1a, we were unable to detect ErbB2 expression in B16-BL6 cells. This is correlated with the finding that the ErbB2 tyrosine kinase inhibitor AG825 did not inhibit HRG-induced phosphorylation of ErbB3 and ERK (Fig. 3a). In contrast, EGFR was effectively phosphorylated in response to HRG (Fig. 3b), and the HRG-induced signaling events were blocked by an EGFR tyrosine kinase inhibitor, PD153035 (Figs. 3a and 3b). Formation of heterodimer was evident from immunoprecipitation where HRG stimulation induced coimmunoprecipitaion of phosphorylated ErbB3 with EGFR (Fig. 3c). Consistent with the result of the immunoblotting, HRG-induced adhesion was abrogated by the inhibition of EGFR activation by PD153035 (Fig. 3d). Furthermore, the stimulation of cultured B16-BL6 cells with HRG for 24 hr promoted experimental metastasis to the lung, which was also blocked by PD153035 (Fig. 3e). These results suggest that HRG induces the formation of an EGFR/ErbB3 heterodimer to initiate the metastasis-promoting signaling in melanoma cells. ErbB3 is essential for HRG-induced cellular responses To further explore the importance of ErbB3 in metastasis, we isolated ErbB3-depleted cells stably expressing shRNA against ErbB3 mRNA (sh-ErbB3). The sh-Luc cells express a control

shRNA against luciferase. Stable transfection was confirmed by GFP expression from the shRNA vectors using FACS (data not shown) and immunoblotting (Fig. 4a). ErbB3 expression and HRG-induced phosphorylation of ErbB3 and ERK were significantly reduced in sh-ErbB3 cells as compared with parent B16BL6 and sh-Luc cells. In addition, although adhesive and invasive abilities were not significantly affected by inactivation of ErbB3, the HRG-induced increase in abilities was abrogated in sh-ErbB3 cells (Figs. 4b and 4c). These results indicated that ErbB3 is essential for the HRG-induced cellular responses in B16-BL6 cells. Suppression of tumor growth and angiogenesis by inactivation of EbrB3 in vivo Further supporting the result in Figure 2, cell proliferation in vitro was not affected by the decrease in ErbB3 expression (Fig. 5a). In contrast, subcutaneous tumor growth of sh-ErbB3 cells was significantly decreased, especially from day 8 after tumor inoculation (Fig. 5b). Similarly, a local injection of siRNA against ErbB3 with the atolocollagen resulted in a decrease in the tumor growth of parent B16-BL6 cells (Fig. 5c). These results indicate that ErbB3 plays a crucial role in the maintenance of tumor growth after day 8. To explore the mechanism of this suppression of tumor growth, we compared the number of blood vessels entering the tumor on day 7. The number of vessels directed toward the shErbB3 tumor was reduced as compared with that of the sh-Luc tumor (Fig. 6a). In addition, microvessel density of these tumors on day 15 using anti-PECAM-1 antibody was significantly decreased in sh-ErbB3 tumors (Fig. 6b). VEGFA mRNA expression was evaluated as a factor which induces angiogenesis. VEGFA expression was enhanced by HRG in sh-luc cells, but not in sh-ErbB3 cells in vitro (data not shown). The activation of

344

UENO ET AL.

FIGURE 4 – HRG-induced metastatic properties were dependent on ErbB3. (a) Parent and shRNA (Luc and ErbB3)-expressing B16-BL6 cells were cultured with or without HRG for 10 min. Cell lysates were immunoblotted with the indicated antibodies. (b and c) HRG-induced adhesion (b) and invasion (c) was evaluated as described in Figures 2d and 2e, respectively. *p < 0.05, **p < 0.01.

FIGURE 5 – Suppression of tumor growth by inactivation of ErbB3. (a) Cell proliferation in vitro for 24 hr was determined by the BrdU assay. (b) sh-Luc and sh-ErbB3 cells were inoculated subcutaneously into the dorsal skin, and tumor volume on days 6, 8, 10, 12, 14 and 15 was measured. Data represents the mean 6 SD for 10 mice. (c) B16-BL6 cells were inoculated subcutaneously. On day 4, 100 ll of the 1:1 mixtures of 20 lM siRNAs (against ErbB3 and Luc) and atelocollagen was injected into the tumor region. Tumor volume on days 4, 6, 8, 10, 12 and 13 was measured. Data represents the mean 6 SD for 6 mice.

FIGURE 6 – Suppression of angiogenesis by inactivation of ErbB3. (a) The number of new vessels entering the tumor mass of sh-Luc and sh-ErbB3 cells was counted by visualizing with evans blue. Data represents the mean 6 SD for 6 mice. (b) Tumor tissue sections were immunohistochemically stained with anti-PECAM-1 antibody. The number of PECAM-1 positive microvessel in tumors was counted. (c) The relative levels of endogenous VEGFA mRNA after 2 hr of HRG stimulation were measured with real-time RT-PCR. *p < 0.05, **p < 0.01.

EGFR was also essential for induction of VEGFA expression as PD153035 inhibited HRG-induced VEGFA expression (Fig. 6c). Activation of ErbB3 is essential for tumor metastasis The metastatic potential of shRNA cells was evaluated using both experimental and spontaneous metastasis models. When intravenously inoculated, sh-ErbB3 cells normally metastasized to

the lung (Fig. 7a). However, the HRG-induced increase in metastasis was completely abrogated (Fig. 7a). This is consistent with the results from adhesion and invasion experiments in vitro (Figs. 7b and 7c), and indicates that HRG binds to ErbB3 to promote pulmonary metastasis. To quantify the effect of the tumor environment on the process of metastasis from the primary tumor, a foot pad spontaneous

HEREGULIN-INDUCED METASTASIS VIA ErbB3/EGFR

FIGURE 7 – Critical role of ErbB3 activation in tumor metastasis. (a) sh-Luc and sh-ErbB3 cells were stimulated with HRG for 24 hr prior to their intravenous injection. After 14 days, the number of tumor colonies in the lung was counted. Data represents the mean 6 SD for 8 mice. (b) Cells were injected into the right foot pad. The primary tumor was amputated when the group average of tumor volume exceeded 280 mm3 (day 19 for sh-luc and day 21 for sh-ErbB3). Twenty-one days after the amputation, the number of colonies in the lung was counted. Representative photographs of the lung are shown. Data represents the mean 6 SD for 10 mice. **p < 0.01.

metastasis model was used. Because the metastasis is known to correlate with the size of the primary tumor, amputation of the foot with the primary tumor was performed so that the average tumor size was similar in both groups. Figure 6b shows that the number of metastatic colonies in the lung 21 days after the amputation was significantly decreased in sh-ErbB3 cells. A similar result was obtained when the primary tumor was amputated individually when the tumor exceeded 570 mm3 (data not shown). These results demonstrated that depletion of ErbB3 reduces the metastatic ability of melanoma cells in a tumor growth-independent manner. Discussion Recently, EbrB3 has been characterized as an important factor contributing to the malignancy of tumors. A correlation of ErbB3 expression with survival and metastasis has been observed clinically in several tumors including ovarian, lung and breast cancers.21–23 It is no exception that the expression of ErbB3 in melanoma cells also leads to malignancy.10 Although studies concerning the function of ErbB3 have focused on ErbB2 as a heterodimeric partner, the role of EGFR in ErbB3 signaling is

345

largely unknown. In the present study, we provide the first direct evidence that HRG signaling via ErbB3 and EGFR can contribute to the metastasis of melanoma cells by promoting adhesion, invasion and angiogenesis. ErbB3 is a kinase dead receptor and relies on other members of the family for its phosphorylation. There are reports that ErbB3 is phosphorylated by ErbB2 especially in breast cancer cells with higher levels of ErbB2. However, ErbB2 is rarely overexpressed in melanoma cells,24 whereas ErbB3 has been found to be overexpressed in several studies.22 We demonstrated that HRG-induced activation of ErbB3/EGFR heterodimer led to the activation of MAPK and Akt and metastasis. Further proving the importance of EGFR and ErbB3 in melanoma, A375 human melanoma cells also displayed HRGinduced ErbB3 phosphorylation and invasion through EGFRdependent mechanisms (data not shown). Although the HRG used in this study is able to bind to both ErbB3 and ErbB4, the knockdown of ErbB3 largely canceled out HRG-induced cellular responses, indicating that ErbB4 is dispensable to the HRG signaling. This is well consistent with the observation that BTC, an ErbB4 ligand, induced no obvious phosphorylation of the downstream kinases. The binding of HRG to ErbB3 triggered the activation of EGFR by autophosphorylation; however, EGF did not induce the activation of MAPK or phosphorylation of ErbB3, suggesting that the EGFR/ErbB3 heterodimer preferentially formed with the ligand for ErbB3. As shown in Figure 1a, ErbB3 is highly expressed in B16-BL6 cells; therefore, trans-phosphorylation of ErbB3 by EGFR might amplify the signal for promoting tumor growth and metastasis.25 Although the stimulation of HRG expression and downregulation of ErbB3 expression did not affect cell proliferation in vitro, tumor growth in the skin on the back was significantly decreased by the shRNA-mediated downregulation of ErbB3. This dramatic change in tumor volume was also observed when ErbB3 siRNA was directly injected around the tumor. However, it is interesting that the early phase of tumor growth within a week was not affected by downregulation of ErbB3 in either the shRNA or siRNA experiment. A similar result was obtained in the growth of primary tumors inoculated into the foot pad. These results suggest stage-specific functions of ErbB3 in supporting and maintaining the tumor growth. This correlates with the findings that inactivation of ErbB3 decreased the number of vessels entering the tumor and the density of intratumoral microvessels. HRG is known to have an angiogenic effect by enhancing both the formation of tubes by vascular endothelial cells and VEGF expression in breast cancer cells.26–28 We also found that HRG-induced EGFR and ErbB3 activation enhanced VEGFA mRNA expression in B16-BL6 cells in vitro (Fig. 5e). These findings suggest that exposure of ErbB3-overexpressing melanoma to HRG is critical to support the tumor growth by, at least in part, promoting angiogenesis. On the other hand, it has recently been reported that inactivation of ErbB3 by siRNA promotes apoptosis and attenuates cell growth of A549 human lung adenocarcinoma cells in soft agar in vitro.29 Therefore, we can not rule out the possibility that a reduction of survival signaling such as to Akt from ErbB3 directly induces apoptosis of melanoma cells during the late phase of tumor growth. Characterization of the mechanism of reduction in tumor growth, including identification of the angiogenic factors released from cancer cells and apoptosis-related events, will provide insights into the molecular basis for the HRG/ErbB3induced tumor growth and metastasis. The importance of HRG in metastasis was highlighted in this study. HRG signaling via ErbB3 and EGFR promoted metastatic activities in vitro, including adhesion and invasion through the expression of metastasis-related genes. Stimulation of cultured melanoma cells with HRG enhanced metastasis to the lung in the intravenous inoculation model. In contrast,

346

UENO ET AL.

inactivation of ErbB3 by shRNA did not directly influence the formation of metastatic colonies in the experimental metastasis model, although HRG-induced enhancement was completely dependent on ErbB3. However, spontaneous metastasis by inoculating the cells without HRG stimulation into the foot pad was significantly decreased. Xue et al. recently demonstrated that spontaneous lung metastasis of ErbB2-overexpressing breast cancer cells injected into the mammary fat pads was reduced by ErbB3 shRNA. Similarly, increasing ErbB3 expression in breast cancer cells with higher levels of ErbB2 and lower levels of ErbB3 enhances chemotaxis, intravasation and spontaneous metastasis, but not experimental lung metastasis. In addition, HRG is known to be secreted by several tumor cells including breast cancer and melanoma cells.30–32 Taken together, these findings suggest that HRG-induced ErbB3 activation in the primary tumor is essential for intravasation and subsequent extravasation into the target organ, and that exposure to HRG at the primary site is essential for the acquisition of metastatic activity. The HRG gene family has 4 members; HRG-1, HRG-2, HRG-3 and HRG-4, of which a multitude of different isoforms are synthesized by alternative exon splicing, showing various tissue distribution and biological activities. Disruption of the physiological balance between HRG ligands and their ErbB receptors is implicated in the formation of a variety of human cancers. Therefore, identification

of the sources and subtypes of HRGs and regulation of their expression at the primary tumor site are necessary to elucidate the mechanisms of tumor progression by HRG-induced activation of ErbB3. Once melanoma has progressed to the metastatic stage, there is no effective long-term treatment. This has led to the realization that targeted therapeutics are needed which inhibit the activities of specific genes or signaling pathways involved in the development of this disease. In the present study, we provide the possibility that EGFR and ErbB3 are potential targets for the treatment of melanoma. Therefore, anti-EGFR therapy is a new therapeutic strategy; however, ErbB3 has recently been reported critical to developing resistance to gefitinib and erlotinib33–35, EGFR tyrosine kinase inhibitors. Therefore, we expect ErbB3 to be a more effective molecular target. Because ErbB3 is a kinase-dead receptor, development of a novel neutralizing anti-ErbB3 antibody will result in therapy for ErbB3-overexpressing cancers, possibly in combination with anti-EGFR therapy. Acknowledgements We are grateful to Dr. F. Okada for providing QR-32 cells. We also thank Drs. M. Tsuda and A. Tabuchi for generating the shRNA plasmid vectors.

References 1.

2.

3.

4.

5.

6.

7. 8.

9.

10.

11.

12. 13.

Lacroix H, Iglehart JD, Skinner MA, Kraus MH. Overexpression of erbB-2 or EGF receptor proteins present in early stage mammary carcinoma is detected simultaneously in matched primary tumors and regional metastases. Oncogene 1989;4:145–51. Lewis TB, Robison JE, Bastien R, Milash B, Boucher K, Samlowski WE, Leachman SA, Dirk Noyes R, Wittwer CT, Perreard L, Bernard PS. Molecular classification of melanoma using real-time quantitative reverse transcriptase-polymerase chain reaction. Cancer 2005;104: 1678–86. Matsuo M, Sakurai H, Saiki I. ZD1839, a selective epidermal growth factor receptor tyrosine kinase inhibitor, shows antimetastatic activity using a hepatocellular carcinoma model. Mol Cancer Ther 2003; 2:557–61. Palmieri D, Bronder JL, Herring JM, Yoneda T, Weil RJ, Stark AM, Kurek R, Vega-Valle E, Feigenbaum L, Halverson D, Vortmeyer AO, Steinberg SM, et al. Her-2 overexpression increases the metastatic outgrowth of breast cancer cells in the brain. Cancer Res 2007; 67:4190–8. M€ uller-Tidow C, Diederichs S, Bulk E, Pohle T, Steffen B, Schw€able J, Plewka S, Thomas M, Metzger R, Schneider PM, Brandts CH, Berdel WE, et al. Identification of metastasis-associated receptor tyrosine kinases in non-small cell lung cancer. Cancer Res 2005;65:1778–82. Holbro T, Beerli RR, Maurer F, Koziczak M, Barbas CF, III, Hynes NE. The ErbB2/ErbB3 heterodimer functions as an oncogenic unit: ErbB2 requires ErbB3 to drive breast tumor cell proliferation. Proc Natl Acad Sci USA 2003;100:8933–8. Duneau JP, Vegh AP, Sturgis JN. A dimerization hierarchy in the transmembrane domains of the HER receptor family. Biochemistry 2007;46:2010–19. Hudelist G, Singer CF, Manavi M, Pischinger K, Kubista E, Czerwenka K. Co-expression of ErbB-family members in human breast cancer: Her-2/neu is the preferred dimerization candidate in nodalpositive tumors. Breast Cancer Res Treat 2003;80:353–61. Slamon DJ, Clark GM, Wong SG, Levin WJ, Ullrich A, McGuire WL. Human breast cancer: correlation of relapse and survival with amplification of the HER-2/neu oncogene. Science 1987;235:177– 82. Chen HY, Yu SL, Chen CH, Chang GC, Chen CY, Yuan A, Cheng CL, Wang CH, Terng HJ, Kao SF, Chan WK, Li HN, et al. A fivegene signature and clinical outcome in non-small-cell lung cancer. N Engl J Med 2007;356:11–20. Xue C, Liang F, Mahmood R, Vuolo M, Wyckoff J, Qian H, Tsai KL, Kim M, Locker J, Zhang ZY, Segall JE. ErbB3-dependent motility and intravasation in breast cancer metastasis. Cancer Res 2006;66: 1418–26. Dunbar AJ, Goddard C. Structure-function and biological role of betacellulin. Int J Biochem Cell Biol 2000;32:805–15. Stove C, Bracke M. Roles for neuregulins in human cancer. Clin Exp Metastasis 2004;21:665–84.

14. Tsai MS, Shamon-Taylor LA, Mehmi I, Tang CK, Lupu R. Blockage of heregulin expression inhibits tumorigenicity and metastasis of breast cancer. Oncogene 2003;22:761–8. 15. Atlas E, Cardillo M, Mehmi I, Zahedkargaran H, Tang C, Lupu R. Heregulin is sufficient for the promotion of tumorigenicity and metastasis of breast cancer cells in vivo. Mol Cancer Res 2003;1:165–75. 16. Choo MK, Sakurai H, Koizumi K, Saiki I. TAK1-mediated stress signaling pathways are essential for TNF-alpha-promoted pulmonary metastasis of murine colon cancer cells. Int J Cancer 2006;118:2758– 64. 17. Ueno Y, Sakurai H, Matsuo M, Choo MK, Koizumi K, Saiki I. Selective inhibition of TNF-alpha-induced activation of mitogen-activated protein kinases and metastatic activities by gefitinib. Br J Cancer 2005;92:1690–5. 18. Tsunoda S, Nakamura T, Sakurai H, Saiki I. Fibroblast growth factor2-induced host stroma reaction during initial tumor growth promotes progression of mouse melanoma via vascular endothelial growth factor A-dependent neovascularization. Cancer Sci 2007;98:541–8. 19. Weidner N. Current pathologic methods for measuring intratumoral microvessel density within breast carcinoma and other solid tumors. Breast Cancer Res Treat 1995;36:169–80. 20. Meiners S, Brinkmann V, Naundorf H, Birchmeier W. Role of morphogenetic factors in metastasis of mammary carcinoma cells. Oncogene 1998;16:9–20. 21. Rajkumar T, Stamp GW, Hughes CM, Gullick WJ. c-erbB3 protein expression in ovarian cancer. Clin Mol Pathol 1996;49:M199– M202. 22. Gy€orffy B, Lage H. A web-based data warehouse on gene expression in human malignant melanoma. Invest Dermatol 2007;127:394–9. 23. Tovey SM, Witton CJ, Bartlett JM, Stanton PD, Reeves JR, Cooke TG. Outcome and human epidermal growth factor receptor (HER) 1– 4 status in invasive breast carcinomas with proliferation indices evaluated by bromodeoxyuridine labelling. Breast Cancer Res 2004; 6:R246–R251. 24. Inman JL, Kute T, White W, Pettenati M, Levine EA. Absence of HER2 overexpression in metastatic malignant melanoma. J Surg Oncol 2003;84:82–8. 25. Li Z, Mei Y, Liu X, Zhou M. Neuregulin-1 only induces trans-phosphorylation between ErbB receptor heterodimer partners. Cell Signal 2007;19:466–71. 26. Xiong S, Grijalva R, Zhang L, Nguyen NT, Pisters PW, Pollock RE, Yu D. Up-regulation of vascular endothelial growth factor in breast cancer cells by the heregulin-beta1-activated p38 signaling pathway enhances endothelial cell migration. Cancer Res 2001;61:1727– 32. 27. Yen L, Benlimame N, Nie ZR, Xiao D, Wang T, Al Moustafa AE, Esumi H, Milanini J, Hynes NE, Pages G, Alaoui-Jamali MA. Differential regulation of tumor angiogenesis by distinct ErbB homo- and heterodimers. Mol Biol Cell 2002;13:4029–44.

HEREGULIN-INDUCED METASTASIS VIA ErbB3/EGFR

28. Russell KS, Stern DF, Polverini PJ, Bender JR. Neuregulin activation of ErbB receptors in vascular endothelium leads to angiogenesis. Am J Physiol 1999;277:H2205–H2211. 29. Sithanandam G, Fornwald LW, Fields J, Anderson LM. Inactivation of ErbB3 by siRNA promotes apoptosis and attenuates growth and invasiveness of human lung adenocarcinoma cell line A549. Oncogene 2005;24:1847–59. 30. Gollamudi M, Nethery D, Liu J, Kern JA. Autocrine activation of ErbB2/ErbB3 receptor complex by NRG-1 in non-small cell lung cancer cell lines. Lung Cancer 2004;43:135–43. 31. Li Q, Ahmed S, Loeb JA. Development of an autocrine neuregulin signaling loop with malignant transformation of human breast epithelial cells. Cancer Res 2004;64:7078–85. 32. Stove C, Boterberg T, Van Marck V, Mareel M, Bracke M. Bowes melanoma cells secrete heregulin, which can promote aggregation and

347

counteract invasion of human mammary cancer cells. Int J Cancer 2005;114:572–8. 33. Frolov A, Schuller K, Tzeng CW, Cannon EE, Ku BC, Howard JH, Vickers SM, Heslin MJ, Buchsbaum DJ, Arnoletti JP. ErbB3 expression and dimerization with EGFR influence pancreatic cancer cell sensitivity to erlotinib. Cancer Biol Ther 2007;6:548–54. 34. Sergina NV, Rausch M, Wang D, Blair J, Hann B, Shokat KM, Moasser MM. Escape from HER-family tyrosine kinase inhibitor therapy by the kinase-inactive HER3. Nature 2007;445:437– 41. 35. Engelman JA, Zejnullahu K, Mitsudomi T, Song Y, Hyland C, Park JO, Lindeman N, Gale CM, Zhao X, Christensen J, Kosaka T, Holmes AJ, et al. MET amplification leads to gefitinib resistance in lung cancer by activating ERBB3 signaling. Science 2007;316:1039– 43.