Lipocalin 2 Diminishes Invasiveness and Metastasis of Ras-transformed Cells
2005; Elsevier BV; Volume: 280; Issue: 14 Linguagem: Inglês
10.1074/jbc.m413047200
ISSN1083-351X
AutoresJun‐ichi Hanai, Tadanori Mammoto, Pankaj Seth, Kiyoshi Mori, S. Ananth Karumanchi, Jonathan Barasch, Vikas P. Sukhatme,
Tópico(s)Hippo pathway signaling and YAP/TAZ
ResumoLipocalin 2, an iron-siderophore-binding protein, converts embryonic kidney mesenchyme to epithelia. We found that lipocalin 2 could also convert 4T1-Ras-transformed mesenchymal tumor cells to an epithelial phenotype, increase E-cadherin expression, and suppress cell invasiveness in vitro and tumor growth and lung metastases in vivo. The Ras-MAPK pathway mediated the epithelial to mesenchymal transition in part by increasing E-cadherin phosphorylation and degradation. Lipocalin 2 antagonized these effects at a point upstream of Raf activation. Lipocalin 2 action was enhanced by iron-siderophore. These data characterize lipocalin 2 as an epithelial inducer in Ras malignancy and a suppressor of metastasis. Lipocalin 2, an iron-siderophore-binding protein, converts embryonic kidney mesenchyme to epithelia. We found that lipocalin 2 could also convert 4T1-Ras-transformed mesenchymal tumor cells to an epithelial phenotype, increase E-cadherin expression, and suppress cell invasiveness in vitro and tumor growth and lung metastases in vivo. The Ras-MAPK pathway mediated the epithelial to mesenchymal transition in part by increasing E-cadherin phosphorylation and degradation. Lipocalin 2 antagonized these effects at a point upstream of Raf activation. Lipocalin 2 action was enhanced by iron-siderophore. These data characterize lipocalin 2 as an epithelial inducer in Ras malignancy and a suppressor of metastasis. Down-regulation of epithelial proteins and the induction of mesenchymal proteins (EMT) 1The abbreviations used are: EMT, epithelial to mesenchymal transition; MAPK, mitogen-activated protein kinase; ERK, extracellular signal-regulated kinase; HA, hemagglutinin; MEK, MAPK/ERK kinase; DMEM, Dulbecco's modified Eagle's medium; PI3K, phosphatidylinositol 3-kinase; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; FCS, fetal calf serum; RT, reverse transcription; SRE, serum-response element; DFO, deferoxamine mesylate. (1.Chambers A.F. Groom A.C. MacDonald I.C. Nat. Rev. Cancer. 2002; 2: 563-572Crossref PubMed Scopus (3095) Google Scholar, 2.Birchmeier W. Behrens J. Biochim. Biophys. Acta. 1994; 1198: 11-26Crossref PubMed Scopus (935) Google Scholar, 3.Hay E.D. Acta Anat. (Basel). 1995; 154: 8-20Crossref PubMed Scopus (1256) Google Scholar, 4.Grunert S. Jechlinger M. Beug H. Nat. Rev. Mol. Cell. Biol. 2003; 4: 657-665Crossref PubMed Scopus (571) Google Scholar, 5.Thiery J.P. Nat. Rev. Cancer. 2002; 2: 442-454Crossref PubMed Scopus (5531) Google Scholar, 6.Fidler I.J. Nat. Rev. Cancer. 2003; 3: 453-458Crossref PubMed Scopus (3592) Google Scholar) enhance the metastatic potential of epithelial tumors (7.Boussadia O. Kutsch S. Hierholzer A. Delmas V. Kemler R. Mech. Dev. 2002; 115: 53-62Crossref PubMed Scopus (208) Google Scholar, 8.Islam S. Kim J.B. Trendel J. Wheelock M.J. Johnson K.R. J. Cell. Biochem. 2000; 78: 141-150Crossref PubMed Scopus (28) Google Scholar, 9.Thiery J.P. Chopin D. Cancer Metastasis Rev. 1999; 18: 31-42Crossref PubMed Scopus (173) Google Scholar), whereas reactivation of epithelial genes reverses the malignant phenotype (MET) (10.Vanderburg C.R. Hay E.D. Acta Anat. (Basel). 1996; 157: 87-104Crossref PubMed Scopus (45) Google Scholar). We hypothesized that an endogenous epithelial inducer (21.Yang J. Goetz D. Li J.Y. Wang W. Mori K. Setlik D. Du T. Erdjument-Bromage H. Tempst P. Strong R. Barasch J. Mol. Cell. 2002; 10: 1045-1056Abstract Full Text Full Text PDF PubMed Scopus (531) Google Scholar), lipocalin 2 (also called siderocalin, Ngal, 24p3, uterocalin, and neu-related lipocalin), could stimulate the epithelial phenotype in Ras-transformed cells and reverse their metastatic potential. Lipocalin 2 is a member of a superfamily of carrier proteins (11.Flower D.R. Biochem. J. 1996; 318: 1-14Crossref PubMed Scopus (1397) Google Scholar) that is expressed in granulocytic precursors (12.Cowland J.B. Borregaard N. Genomics. 1997; 45: 17-23Crossref PubMed Scopus (489) Google Scholar) as well as in numerous epithelia cell types (13.Friedl A. Stoesz S.P. Buckley P. Gould M.N. Histochem. J. 1999; 31: 433-441Crossref PubMed Scopus (240) Google Scholar, 14.Nielsen B.S. Borregaard N. Bundgaard J.R. Timshel S. Sehested M. Kjeldsen L. Gut. 1996; 38: 414-420Crossref PubMed Scopus (348) Google Scholar). Crystallography showed that the protein is a carrier of iron bound to a siderophore (15.Strong R.K. Bratt T. Cowland J.B. Borregaard N. Wiberg F.C. Ewald A.J. Acta Crystallogr. Sect. D Biol. Crystallogr. 1998; 54: 93-95Crossref PubMed Scopus (29) Google Scholar), which is a small organic molecule produced by bacteria (20.Goetz D.H. Holmes M.A. Borregaard N. Bluhm M.E. Raymond K.N. Strong R.K. Mol. Cell. 2002; 10: 1033-1043Abstract Full Text Full Text PDF PubMed Scopus (1037) Google Scholar). Both recombinant and mammalian-expressed lipocalin 2 (16.Yang J. Mori K. Li J.Y. Barasch J. Am. J. Physiol. Renal Physiol. 2003; 285: F9-F18Crossref PubMed Scopus (121) Google Scholar) induce the de novo expression of E-cadherin, the formation of polarized epithelia, and the development of tubules in embryonic mesenchyme in an iron-dependent fashion (17.Li J.Y. Ram G. Gast K. Chen X. Barasch K. Mori K. Schmidt-Ott K. Wang J. Kuo H.C. Savage-Dunn C. Garrick M.D. Barasch J. Am. J. Physiol. Cell Physiol. 2004; 287: C1547-C1559Crossref PubMed Scopus (38) Google Scholar). Although lipocalin 2 is highly expressed upon polyoma, SV40 or neu transformation, and after malignant transformation of the breast, lung, colon, and pancreatic epithelia (12.Cowland J.B. Borregaard N. Genomics. 1997; 45: 17-23Crossref PubMed Scopus (489) Google Scholar, 13.Friedl A. Stoesz S.P. Buckley P. Gould M.N. Histochem. J. 1999; 31: 433-441Crossref PubMed Scopus (240) Google Scholar), the functional role of lipocalin 2 in this context is unknown. Here we suggest that the protein regulates the epithelial characteristics of malignant cells as it does for embryonic mesenchyme. This activity might result from iron transport or signaling through unknown receptors (18.Devireddy L.R. Teodoro J.G. Richard F.A. Green M.R. Science. 2001; 293: 829-834Crossref PubMed Scopus (331) Google Scholar). To test these hypotheses, we added purified lipocalin 2 or lipocalin 2 vectors to Ras-transformed 4T1 mouse mammary tumor cells. These cells are known to metastasize to bone, liver, and lung tissue in a pattern similar to that found in human breast cancer (19.Lin P. Buxton J.A. Acheson A. Radziejewski C. Maisonpierre P.C. Yancopoulos G.D. Channon K.M. Hale L.P. Dewhirst M.W. George S.E. Peters K.G. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 8829-8834Crossref PubMed Scopus (338) Google Scholar). However, introduction of lipocalin 2 reversed Ras-induced EMT, reduced tumor growth, and dramatically suppressed metastasis. In lipocalin 2-treated cells, E-cadherin was rescued from proteasomal degradation by inhibition of Ras-MAPK signaling. This protection was iron-dependent. Plasmids, Virus Constructs, Lipocalin 2 Proteins, Antibodies, and Signaling Inhibitors—Human lipocalin 2 cDNA (GenBank™ accession number BC033089) with a C terminus HA tag was PCR-amplified and subcloned into pcDNA3.1 (Invitrogen). The constitutively active form H-ras A12-pBabe retroviral vector and empty-pBabe were gifts from Dr. M. Ewen (Dana Farber Cancer Institute, Boston, MA). Another constitutively active form of ras plasmid (H-ras V12-pcDNA3.1) was purchased from the Guthrie cDNA Resource Center (Sayre, PA). The constitutively active form of MEK (MEK-DD) and Lac-Z adenoviral vectors were gifts from Dr. E. O'Leary (Harvard Institute of Medicine, Boston, MA). MEK-DD cDNA was a gift from Dr. H. Iba (Tokyo University, Tokyo, Japan). Recombinant mouse lipocalin 2 (accession number NM008491) was expressed as a glutathione S-transferase fusion protein in the BL21 strain of Escherichia coli (Stratagene, La Jolla, CA), which does not synthesize siderophore (20.Goetz D.H. Holmes M.A. Borregaard N. Bluhm M.E. Raymond K.N. Strong R.K. Mol. Cell. 2002; 10: 1033-1043Abstract Full Text Full Text PDF PubMed Scopus (1037) Google Scholar, 21.Yang J. Goetz D. Li J.Y. Wang W. Mori K. Setlik D. Du T. Erdjument-Bromage H. Tempst P. Strong R. Barasch J. Mol. Cell. 2002; 10: 1045-1056Abstract Full Text Full Text PDF PubMed Scopus (531) Google Scholar). Ferric sulfate (Sigma-Aldrich) was added in the culture medium at 50 μm. The protein was isolated using glutathione-Sepharose 4B beads (Amersham Biosciences), eluted with thrombin (Sigma-Aldrich), and further purified with gel filtration (Superdex 75; Amersham Biosciences). Iron-loaded lipocalin 2 (Lipo:Sid:Fe) and iron-unloaded lipocalin 2 (Lipo:Sid) were prepared by mixing the recombinant protein with iron-loaded and iron-unloaded forms of a bacterial siderophore enterochelin (EMC Microcollections, Tübingen, Germany) in phosphate-buffered saline at room temperature for 60 min. Unbound siderophore was removed with Microcon YM-10 (Millipore). The recombinant protein diluted in culture medium was sterilized before addition to the cells using 0.22-μm filters (Millipore). The following reagents were purchased from the respective companies: anti-Ras antibody, Oncogene Research Products (San Diego, CA); anti-Raf, anti-phospho-Raf, anti-MEK1/2, anti-phospho-MEK1/2, anti-ERK1/2, and anti-phospho-ERK1/2 antibodies and MEK (U0126) and PI3K inhibitors (LY294002), Cell Signaling Technologies (Beverly, MA); anti-E-cadherin and PY20 anti-P-Tyr monoclonal antibodies, BD Transduction Laboratories (Deerfield, IL); anti-vimentin monoclonal antibody and fluorescein isothiocyanate-conjugated goat anti-mouse IgG, Santa Cruz Biotechnology (Santa Cruz, CA); anti-GAPDH antibody, Chemicon International Inc. (Temecula, CA); anti-Hakai antibody, Zymed Laboratories (San Francisco, CA); proteasome inhibitor MG132, Boston Biochemistry (Cambridge, MA); and deferoxamine mesylate salt, Sigma-Aldrich. Stable Cell Lines—293T and 4T1 cells (American Type Culture Collection, Manassas, VA) were cultured in DMEM and 10% FCS and seeded (106 cells/100-mm dish) 12 h prior to transfection with FuGENE 6 reagent (32.5 μl; Roche Applied Science) and retroviral construct (10 μg, CA-H-ras-pBabe or empty-pBabe). 10 ml of condition medium were collected at 48 h and diluted 1:1 with DMEM and 10% FCS and added to the 4T1 cells (106 cells/100-mm dish) for 48 h, followed with selective medium containing hygromycin (Invitrogen). 8–10 single clones (4T1-Ras (referred to henceforth as R) or 4T1-EV (referred to henceforth as EV)) were selected. A single clone (clone 1) from the R group was used for additional studies. Similarly, a single clone (clone 1) from the EV group was selected. R cells (clone 1) were transfected with lipocalin 2-pcDNA3.1, selected with neomycin, and screened for HA-tagged lipocalin 2 using anti-HA antibody (Santa Cruz Biotechnology). RL (double transfectant) clone (clone 6), which showed the highest level of lipocalin 2 expression, was used for additional studies. Supplemental Fig. A shows that H-ras A12 DNA is present in RL cells and that the transcript is expressed in these cells, i.e. RL cells are indeed a double transfectant and have not merely lost expression of the mutant ras gene. Immunodetection—Cells were stained as described previously (22.Mammoto T. Mukai M. Mammoto A. Yamanaka Y. Hayashi Y. Mashimo T. Kishi Y. Nakamura H. Cancer Lett. 2002; 184: 165-170Crossref PubMed Scopus (136) Google Scholar), and images were acquired with a DeltaVision system (Applied Precision, Issaquah, WA) equipped with an Axiovert 100 microscope (Carl Zeiss MicroImaging Inc., Shelton, CT) and a Photometrics 300 series scientific-grade cooled charge-coupled device camera, reading 12-bit images and using the 63/1.4 NA plan-Neofluar objective. For immunoprecipitation and immunoblotting, tissues were weighed; diced; soaked in ice-cold radioimmune precipitation assay buffer with 1 mm phenylmethylsulfonyl fluoride, 1 μg/ml aprotinin, 1 mm Na3VO4, and 1 mm NaF; homogenized on ice; and centrifuged at 10,000 × g for 10 min at 4 °C. The supernatant fluid was collected as total cell lysate. Cultured cells were washed, scraped, and solubilized in a lysis buffer containing 20 mm Tris-HCl, pH 7.5, 150 mm NaCl, 0.5% Triton X-100, 1% aprotinin, and 1 mm phenylmethylsulfonyl fluoride. After 20 min on ice, the cells were pelleted by centrifugation, and the supernatants were used as a cell lysate. Cell lysates or immunoprecipitated cell lysates were separated by PAGE (NuPAGE® gels; Invitrogen), followed by electroblotting onto a polyvinylidene difluoride membrane. Protein bands were detected using SuperSignal® West Pico Chemiluminescent Substrate (Pierce) (23.Hanai J. Gloy J. Karumanchi S.A. Kale S. Tang J. Hu G. Chan B. Ramchandran R. Jha V. Sukhatme V.P. Sokol S. J. Cell Biol. 2002; 158: 529-539Crossref PubMed Scopus (143) Google Scholar). Luciferase Assay—After transient transfection of the plasmids, cells were incubated for 20 h in 10% FCS, and luciferase activity in the cell lysates was determined using a luminometer normalized by sea-pansy luciferase activity under the control of the thymidine kinase promoter. The Dual-Luciferase Reporter Assay System was purchased from Promega (24.Hanai J. Dhanabal M. Karumanchi S.A. Albanese C. Waterman M. Chan B. Ramchandran R. Pestell R. Sukhatme V.P. J. Biol. Chem. 2002; 277: 16464-16469Abstract Full Text Full Text PDF PubMed Scopus (217) Google Scholar). In Vitro Invasion Assay—Polycarbonate membranes (6.5-mm diameter, 8-μm pore size) of Transwells (Coster, NY) were coated with Matrigel® (BD Biosciences), and cells were seeded (106 cells/100 μl) with DMEM including 0.1% serum. 16 h later, cells were fixed and stained with Giemsa solution, and the upper surface of each membrane was scraped with a cotton swab. Cells that had reached the lower surface of the membrane (migrated cells) were counted in 20 random fields using a light microscope (×400). Semiquantitative Reverse Transcription (RT)-PCR—Total RNA was isolated from 4T1 cells in vitro using the SV Total RNA Isolation system (Promega). Tissue RNA was collected with TRIzol® (Invitrogen). RT-PCR was performed on the PerkinElmer Life Sciences GeneAmp PCR System 2400 using Omniscript (Qiagen) for reverse transcription reaction, and Taq DNA polymerase (Qiagen) and primers for mouse E-cadherin (5′-TGCCCAGAAAATGAAAAAGG-3′ and 5′-AATGGCAGGAATTTGCAATC-3′), GAPDH (5′-ACAGTCTTCTGAGTGGCA-3′ and 5′-CCCATCACCATCTTCCAG-3′), and HA-tagged lipocalin 2 (5′-GGAGTACTTCAAGATCAC-3′ and 5′-GAAAGCATAGTCTGGAACGTCATAG-3′) for DNA amplification. The PCR conditions were established for DNA amplification in the linear range. RT-PCR products were analyzed on 1% agarose gels. In Vivo Assay for Primary Tumor Growth and Pulmonary Metastases—107 4T1 (EV, R, and RL) cells were injected subcutaneously in Balb/c mice (25.Asai T. Ueda T. Itoh K. Yoshioka K. Aoki Y. Mori S. Yoshikawa H. Int. J. Cancer. 1998; 76: 418-422Crossref PubMed Scopus (185) Google Scholar). Although this model is not the standard orthotopic model used, we have used it extensively in our laboratory to study metastases in lung. Primary tumor volume (V) = a × b × b/2, where a represents the minimum tumor diameter, and b represents the maximum tumor diameter. After 3 weeks, lung weight and the number of metastatic nodules on the lung surface were evaluated. Statistical Analysis—All values are expressed as mean ± S.E. A one-tailed Student's t test was used to identify significant differences in multiple comparisons. A level of p < 0.05 was considered statistically significant. Lipocalin 2 Reverses the Ras-transformed Phenotype—We chose a syngeneic spontaneously metastasizing murine breast cancer model (4T1 cell line) and accelerated its metastatic potential by introduction of constitutively active mouse H-ras mutant A12 using retrovirus. Whereas 4T1 cells infected with an empty vector (EV) grew in a cobblestone-shaped pattern (Fig. 1A, top left panel), 4T1-Ras (R) cells were spindle-shaped and did not form clusters at low confluence (Fig. 1A, top middle panel). To assess the effects of lipocalin 2 expression on Ras transformation, we generated stable clones of 4T1-Ras cells expressing lipocalin 2 (RL) by transfection of a lipocalin expression plasmid (lipocalin 2-pcDNA3.1). Compared with R cells, the RL cells (Fig. 1A, top right panel) reverted to an epithelial morphology and grew appositionally (similar to EV cells), reexpressed E-cadherin, and suppressed the expression of mesenchymal vimentin. (Fig. 1, A, bottom panels, and B). In contrast, E-cadherin mRNA remained unchanged (Fig. 1C), suggesting that the effects of Ras and lipocalin were post-transcriptional. Expression of E-cadherin in RL cells was dependent on the dose of lipocalin 2-pcDNA3.1 expression vector (transiently introduced in a population of R cells), on a conditioned medium containing lipocalin 2 (Fig. 1, D and E), and on recombinant lipocalin 2 protein. Indeed stable lipocalin 2 expression (RL) almost completely reversed (by ∼76%) Ras-induced invasiveness in vitro (Fig. 2).Fig. 2Invasion migration assay using each stable clone of 4T1. Polycarbonate membranes of Transwells were coated with Matrigel®, and cells were seeded. 16 h later, cells were fixed, stained with Giemsa solution, and counted for each of the stable clones EV, R, and RL.View Large Image Figure ViewerDownload Hi-res image Download (PPT) To determine whether lipocalin 2 could alter the growth of tumors in vivo, we injected EV, R, or RL cells subcutaneously into the backs of Balb/c mice and assessed primary and metastatic tumor size at 1, 2, and 3 weeks post-inoculation. Primary tumors of R cells were significantly larger than those of lipocalin 2 cells (RL) or control cells (EV) (Fig. 3A). Indeed, lipocalin 2 reversed the invasion of adjacent muscle seen in tumors derived from R cells (Fig. 3B). Just like control EV cells, RL tumors were solid, compact, and condensed (they could be "shelled out"). Indeed, RL tumors had more E-cadherin and less vimentin than R cells, making them similar to control tumors (EV cells; Fig. 3C). Most dramatically, the number of metastatic pulmonary nodules was reduced by 80% in RL cells compared with R cells (Fig. 3, E–G), and lung weights were lower. All of these effects were likely post-transcriptional; although mRNA for E-cadherin seemed down-regulated in the R versus EV tumors (Fig. 3D), loading differences (note the GAPDH "controls") make this effect less pronounced and more consistent with the in vitro data (Fig. 1C). Taken together, we find that lipocalin 2 enhanced the epithelial phenotype and inhibited metastasis of Ras-transformed cells. MAPK Signaling: Activation by Ras and Suppression by Lipocalin 2—Ras has multiple downstream effectors (26.Campbell P.M. Der C.J. Semin. Cancer Biol. 2004; 14: 105-114Crossref PubMed Scopus (219) Google Scholar). It activates Raf, which in turn activates MEK, leading to the phosphorylation of MAPK. Ras also activates PI3K. To clarify the Ras pathway of EMT, we assessed the effect of a MEK inhibitor (U0126) and a PI3K inhibitor (LY294002) on R cells. As shown in Fig. 4, the MEK inhibitor reversed Ras-induced EMT, but the effect of the PI3K inhibitor was partial, at best. Because U0126 can inhibit MEK5 in addition to MEK1/2 (referred to here as MEK), we infected R cells with an adenovirus carrying a dominant negative form of MEK1 and found the same results as those obtained with U0126 (data not shown). These data indicate that Ras-MEK signaling is essential for EMT. To determine whether lipocalin 2 reversed Ras-induced EMT by interfering with MEK signaling, we added purified lipocalin 2 protein (iron-loaded with siderophore, Lipo:Sid:Fe) to R cells and found that Ras-induced phosphorylation of Raf, MEK, and ERK1/2 was largely abrogated but that total Ras expression was unchanged (Fig. 5A). Signaling events downstream of ERK activation were then monitored with a multi-copy serum-response element (SRE)-luciferase construct introduced into EV, R, and RL cells (Fig. 5B). RL and EV cells gave comparable levels of luciferase activity, but this was only about one-half to two-thirds of the transcription found in R cells. Just like R cells treated with exogenous protein (Fig. 5A), infection of R cells with recombinant adenovirus carrying lipocalin 2, but not green fluorescent protein, reduced SRE-luciferase activity and MEK and ERK1/2 phosphorylation, without altering Ras expression (data not shown). These data indicate that Ras-MEK is modulated by lipocalin 2. To localize the effect of lipocalin on Ras-MAPK signaling, we utilized an adenovirus and expression plasmid encoding a constitutively active MEK (MEK-DD) (27.Murakami M. Ui M. Iba H. Cell Growth Differ. 1999; 10: 333-342PubMed Google Scholar). MEK-DD adenoviral infection of EV cells led to increased SRE-luciferase activity (increased MAPK activity; data not shown). Importantly, constitutively active MEK resulted in a concentration-dependent EMT, as ascertained by cell shape and colony morphology (Fig. 5C) and by expression of E-cadherin protein (Fig. 5D) in RL cells, indicating that MEK-DD was dominant over the effect of lipocalin 2. Consistent with this idea, MEK-DD also increased SRE-luciferase activity in EV cells, but lipocalin 2 protein (Lipo:Sid:Fe) was unable to inhibit this effect (Fig. 5E, lanes 1, 4, and 5). On the other hand, lipocalin 2 protein down-regulated SRE-luciferase activity resulting from transfection of a constitutively active form of H-ras V12 (CA-H-ras) (Fig. 5E, lanes 1, 2, and 3), as would be expected from the data with stable clones in Fig. 5B. Also, lipocalin 2 cDNA transfection induced E-cadherin expression in EV cells, but this effect was reversed by concomitant MEK-DD adenoviral infection (see Supplemental Fig. B). These data indicate that lipocalin 2 acts upstream of MEK activation. Given that lipocalin 2 down-regulated Raf phosphorylation (Fig. 5A) but did not alter the level of Ras expression, our data indicate that lipocalin 2 acts on Ras-MAPK signaling between Ras and Raf. Furthermore, events outside the Ras-MAPK pathway affected by lipocalin are not sufficient to inhibit Ras-mediated EMT. Lipocalin 2 Inhibits Ras-induced E-Cadherin Phosphorylation and Degradation—To determine how lipocalin might affect Ras-mediated EMT, we focused on the expression of E-cadherin and its relationship to MAPK signaling. Lipocalin 2 is likely to modulate E-cadherin expression on a post-transcriptional level because it did not affect E-cadherin mRNA levels (Figs. 1C and 3D), nor did it enhance E-cadherin promoter transcriptional activity (data not shown). Indeed, we found that E-cadherin is powerfully regulated by proteasome-mediated degradation because treatment with proteasome inhibitor MG132 (0.5 nm) for 2 days increased E-cadherin protein in R cells (Fig. 6B, lanes 3 and 4) and in EV cells (Fig. 6B, lanes 1 and 2). In contrast, MG132 only slightly increased E-cadherin in RL cells (Fig. 6B, lanes 5 and 6), suggesting that E-cadherin degradation was already inhibited and implicating lipocalin 2 in the process. There was also no significant difference in GAPDH protein expression, showing specificity and lack of toxicity of MG132. Furthermore, it is likely that regulation of E-cadherin by proteasomal degradation is relevant to Ras-mediated EMT because MG132 reverted R cells to an epithelial phenotype (Fig. 6A). E-cadherin degradation is mediated by phosphorylation at the binding site for p120 and then recognition by Hakai (28.Fujita Y. Krause G. Scheffner M. Zechner D. Leddy H.E. Behrens J. Sommer T. Birchmeier W. Nat. Cell Biol. 2002; 4: 222-231Crossref PubMed Scopus (689) Google Scholar), which targets the protein for ubiquitination and proteasomal degradation. However, Hakai expression was unchanged by Ras transformation or by lipocalin 2 expression (Fig. 6C). However, we found that E-cadherin phosphorylation was higher in R cells than in either EV or RL cells or in R cells treated with the MEK inhibitor U0126 (Fig. 6D, top panel), in a pattern inversely correlated with E-cadherin protein levels (Fig. 6D, second panel), but was unaccounted for by changes in E-cadherin mRNA levels (Fig. 6D, third panel). Hence, E-cadherin phosphorylation is a target of Ras signaling in 4T1 cells; MEK activation, which is critical for EMT, is also responsible (directly or indirectly) for E-cadherin phosphorylation; and lipocalin 2 impinges on the Ras-MAPK pathway, suppressing E-cadherin phosphorylation and presumably decreasing its turnover. Role of Iron—Because the inductive activity of lipocalin 2 is markedly enhanced by loading the protein with iron (17.Li J.Y. Ram G. Gast K. Chen X. Barasch K. Mori K. Schmidt-Ott K. Wang J. Kuo H.C. Savage-Dunn C. Garrick M.D. Barasch J. Am. J. Physiol. Cell Physiol. 2004; 287: C1547-C1559Crossref PubMed Scopus (38) Google Scholar), we tested the effect of iron on E-cadherin expression and MAPK signaling. Deferoxamine mesylate (2–5 μm; DFO), an iron chelating agent that can deplete iron from the intracellular pool (29.Paller M.S. Hedlund B.E. Kidney Int. 1988; 34: 474-480Abstract Full Text PDF PubMed Scopus (182) Google Scholar), changed the morphology of RL cells to a mesenchymal phenotype and suppressed E-cadherin expression (Fig. 7A), indicating that iron was necessary for E-cadherin expression. Indeed the effect of lipocalin 2 preparations on R cell epithelial morphology (see supplemental Fig. B) and E-cadherin expression correlated with iron carriage (Lipo:Sid:Fe > Lipo:Sid > lipocalin 2; Fig. 7B and supplemental Fig. C) and was dose-dependent (it should be noted that because the affinity of the siderophore for iron is so high (Kd = 10–49) (30.Loomis L.D. Raymond K.N. Inorg. Chem. 1991; 30: 906-911Crossref Scopus (273) Google Scholar), it is likely that the unloaded siderophore partially loaded with iron from the culture media). The same rank order was found the phosphorylation state of ERK1/2 (Fig. 7C) in cells treated with the lipocalins. In contrast to these results, simply adding iron (50 μm ferric ammonium sulfate) to R cells did not change their phenotype. Hence, the data demonstrate that lipocalin 2 inhibits Ras-mediated transformation by up-regulating E-cadherin through an inhibition of MAPK signaling in an iron-dependent manner, but iron alone is insufficient to reverse EMT. In this report we demonstrate that lipocalin 2 can alter the invasive and metastatic behavior of Ras-transformed breast cancer cells in vitro and in vivo by reversing the EMT-inducing activity of Ras, through restoration of E-cadherin expression, via effects on the Ras-MAPK signaling pathway (Fig. 8). The data are consistent with overexpression models of E-cadherin, which prevents invasiveness of human carcinoma cell lines (4.Grunert S. Jechlinger M. Beug H. Nat. Rev. Mol. Cell. Biol. 2003; 4: 657-665Crossref PubMed Scopus (571) Google Scholar, 10.Vanderburg C.R. Hay E.D. Acta Anat. (Basel). 1996; 157: 87-104Crossref PubMed Scopus (45) Google Scholar, 31.Steinberg M.S. McNutt P.M. Curr. Opin. Cell Biol. 1999; 11: 554-560Crossref PubMed Scopus (248) Google Scholar, 32.Adams C.L. Nelson W.J. Curr. Opin. Cell Biol. 1998; 10: 572-577Crossref PubMed Scopus (238) Google Scholar). However, to the best of our knowledge, there has never been a soluble factor that can up-regulate E-cadherin and reverse the metastatic phenotype in vitro and in vivo. Lipocalins, Cancer, and Effects of Lipocalin 2 on EMT in Tumor Cells—Increased expression of lipocalin 2 accompanies numerous transformations (induction by polyoma, SV40, phorbol ester, and the neu oncogene) and human carcinomas (colorectal, hepatic, pancreatic, and breast carcinomas), but the action of the protein has been obscure (reviewed in Ref. 33.Bratt T. Biochim. Biophys. Acta. 2000; 1482: 318-326Crossref PubMed Scopus (118) Google Scholar), with the exception of α2μ-globulin in inducing renal cancer (34.Lehman-McKeeman L.D. Caudill D. Toxicol. Appl. Pharmacol. 1992; 116: 170-176Crossref PubMed Scopus (116) Google Scholar). One report using antisense RNA in an esophageal cancer cell line implanted in an animal suggests that lipocalins are tumor promoters in vivo (35.Li E.M. Xu L.Y. Cai W.J. Xiong H.Q. Shen Z.Y. Zeng Y. Sheng Wu Hua Xue Yu Sheng Wu Wu Li Xue Bao (Shanghai). 2003; 35: 247-254PubMed Google Scholar), and lipocalin 2 may slightly promote the proliferation of estrogen receptor-negative mammary cells in vitro (36.Seth P. Porter D. Lahti-Domenici J. Geng Y. Richardson A. Polyak K. Cancer Res. 2002; 62: 4540-4544PubMed Google Scholar). However, using a large variety of assays, we find a protective role for lipocalin 2 during Ras-mediated transformation and metastasis in vitro and in vivo. Indeed, lipocalin 2 produced smaller, more coherent tumors of higher density (similar weight but different cell types), with less regional invasion and dramatically fewer metastases in vivo as assessed by lung weight, the number of nodules on the lung surface, and histology. Consistent with this antimetastatic action of lipocalin 2, two reports have noted the loss of lipocalin 2 expression at metastatic (colon) carcinoma sites (13.Friedl A. Stoesz S.P. Buckley P. Gould M.N. Histochem. J. 1999; 31: 433-441Crossref PubMed Scopus (240) Google Scholar, 14.Nielsen B.S. Borregaard N. Bundgaard J.R. Timshel S. Sehested M. Kjeldsen L. Gut. 1996; 38: 414-420Crossref PubMed Scopus (348) Google Scholar), in contrast to abundant expression in the primary location. Moreover, our data are consistent with the actions of lipocalin 2 on embryonic mesenchyme. Lipocalin 2 Signaling—Lipocalins may stimulate cell growth and development by binding to cell surface receptors (37.Wojnar P. Lechner M. Merschak P. Redl B. J. Biol. Chem. 2001; 276: 20206-20212Abstract Full Tex
Referência(s)