Both the integrin and insulin-like growth factor binding protein (IGFBP) families independently play important roles in modulating tumor cell growth and progression. We present evidence for a specific cell surface localization and a bimolecular interaction between the αvβ3 integrin and IGFBP-2. The interaction, which could be specifically perturbed using vitronectin and αvβ3 blocking antibodies, was shown to modulate IGF-mediated cellular migration responses. Moreover, this interaction was observed in vivo and correlated with reduced tumor size of the human breast cancer cells, MCF-7β3, which overexpressed the αvβ3 integrin. Collectively, these results indicate that αvβ3 and IGFBP-2 act cooperatively in a negative regulatory manner to reduce tumor growth and the migratory potential of breast cancer cells.

Integrins are a large family of αβ heterodimeric transmembrane glycoproteins that function as cell adhesion and signaling receptors regulating cell death, proliferation, migration, and tissue remodeling (1). In mammals, 1 of 18 α-subunits and one of eight β-subunits interact noncovalently to form ≥24 different heterodimeric receptors (2).

The αvβ3 integrin plays a significant role in a number of physiological and pathological processes, including bone resorption, wound healing, angiogenesis, tumor invasion, and metastasis (3, 4). The αvβ3 integrin binds, in an arg-gly-asp (RGD)-dependent manner, a large number of extracellular matrix proteins, including vitronectin (VN), fibronectin, fibrinogen, and osteopontin (5). In addition, αvβ3 binds to a number of proteins that are not classical extracellular matrix proteins, including fibroblast growth factor-2 (6), matrix metalloproteinase-2 (7), and Cyr61 (8); these interactions play important roles in tumor progression. Moreover, the finding that matrix metalloproteinase-2 interacts with αvβ3 in an RGD-independent manner (7) has led to the development of novel therapeutics that are proving effective in tumor models (9).

Although αvβ3 is expressed at low levels on normal epithelium, its expression is increased in a number of cancers and associated with reduced colon carcinoma survival, prostate tumor progression, enhancement of multiple myeloma invasiveness, melanoma invasion, growth, and metastasis (10, 11, 12, 13). One mechanism by which αvβ3 influences tumor cell progression is through the modulation of growth factor signaling (3). Integrins and growth factors often act synergistically on cell proliferation, differentiation, migration, and survival (14), e.g., αvβ3 can directly associate with a number of growth factor receptors resulting in enhanced cell proliferative signaling (15). Furthermore, αvβ3 is known to influence growth factor signaling when bound to its ligands, e.g., interaction of αvβ3 with tenascin-c modifies the epidermal growth factor growth response and results in enhanced epidermal growth factor receptor activation and downstream signaling (16).

The insulin-like growth factor (IGF) system, which has a profound role in the growth and differentiation of normal and malignant cells, is known to interact with αvβ3 (17), e.g., αvβ3 enhances IGF-I-mediated proliferation and migration when bound to vitronectin. Blocking this interaction reduces IGF-I signaling and downstream cellular responses, suggesting an important interplay of αvβ3, IGF-I receptor, and their signaling components (18, 19).

The components of the system include IGFs (IGF-I and IGF-II), type I and type II IGF receptors (IGF-IR, IGF-IIR), IGF-binding proteins (IGFBPs), and IGFBP proteases. Recently, high serum concentrations of IGF-I have been associated with increased risk of breast, prostate, colorectal, and lung cancers (20). Moreover, IGF-I and IGF-II are involved in tumor cell migration, invasion, and metastasis (17, 18, 19). IGF actions are determined by the availability of free IGFs to interact primarily with IGF-IR. The IGFBP family consists of six members (IGFBP-1–6) that bind IGFs with high affinity and regulate the amount of free IGFs in any given system. The rate of production, clearance, and level of binding of IGFBPs to IGFs determine the extent of IGF signaling and downstream biological consequences. All IGFBP members have been shown to be inhibitory of IGF signaling; however, a number of IGFBPs also mediate IGF-independent actions, including inhibition or enhancement of cell growth, migration, and induction of apoptosis (21, 22).

Here, we define a novel interaction between αvβ3 and the IGF system through an interaction with IGFBP-2. This interaction was found to mediate reduced IGF-I and IGF-II migration and was associated with reduced in vivo growth. Therefore, these findings identify a novel interaction of αvβ3 with IGFBP-2 that results in the IGFBP’s cell surface localization and its subsequent modulation of IGF actions.

Antibodies and Reagents.

Antibodies toward αvβ3(LM609), αv(L230), αvβ5(P1F6), αvβ6(E7P6), β1(TS2/16), β3, β5, and focal adhesion kinase (77) were obtained from Chemicon (Temecula, CA). Monoclonal antibody 23C6 toward αvβ3 was provided by M. A. Horton (Rayne Institute, London, United Kingdom). β-tubulin and IGFBP-2 antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA), and antibodies toward pan-actin were obtained from Biosource International (Camarillo, CA). Polyclonal antibodies toward IGFBP-1,-2,-3,-5, and -6 were provided by Dr. Sue Firth (Kolling Institute, Sydney, Australia); Alexa 488-conjugated goat antirabbit IgG and Alexa 568-conjugated donkey antigoat IgG were purchased from Molecular Probes (Eugene, OR). Growth factors IGF-I, IGF-II, and IGF-I (E3R) were obtained from UBI (Lake Placid, NY), whereas epidermal growth factor and VN were from Becton Dickinson (Bedford, MA). Recombinant IGFBP-2 was from R&D Systems (Minneapolis, MN); biotinylation of IGFBP-2 was achieved by NHS-biotin obtained from Pierce (Rockford, IL) and performed according to manufacturer’s protocol. Collagen I (Vitrogen 100) was obtained from Cohesion (Palo Alto, CA); collagen IV was obtained from Sigma (Sydney, NSW, Australia). The RGD peptides GRGDSP, GRGESP, and gPEN were purchased from Life Technologies, Inc. (Auckland, New Zealand).

Recombinant soluble form of αvβ3 (rsαvβ3) integrin was generated according to the protocol of Metha et al.(23).

Cell Lines and Solid Tumors.

The human breast cancer cell line, MCF-7, was cultured routinely in DMEM with 10% fetal bovine serum (JRH Biosciences, Lenexa, KS) and 100 μg/ml insulin (Novo Nordisk, Rud, Norway). A generation of MCF-7 cells that stably expressed αvβ3 was achieved by PCR amplification of human β3 cDNA that was subcloned into a PCRII shuttle vector (Invitrogen, Melbourne, VIC, Australia) and subsequently subcloned into the pBabe retroviral vector, derived from Moloney MLV. Amphotropic packaging cells, GP+AM12, were transfected with 10 μg of vector plasmid DNA using the calcium phosphate/DNA coprecipitation methodology. Stable colonies were generated through puromycin (1.25 μg/ml) selection. Viral supernatant was obtained from the producer cells, and MCF-7 cells were infected. MCF-7 cells were selected in medium containing 1.25 μg/ml puromycin; resistant colonies were pooled to give a mixed population of cells. No additional cell population sorting was necessary. This resulted in nonclonal populations of cells expressing β3 (MCF-7β3) and mock-transduced controls (MCF-7puro) and avoided problems associated with clonal selection. Tumors were generated by suspending 2 × 106 cells in a 1:1 mixture of Matrigel (Becton Dickinson) and inoculated s.c. adjacent to the mammary fat pad in BALB/c nu/nu mice supplemented with estrogen pellets (Innovative Research of America, Sarasota, FL). Tumors were removed and paraffin embedded. Animal protocols were approved by the animal ethics committee of St. Vincent’s hospital and were in accordance with the NIH Guide for the Care and Use of Laboratory Animals.

Immunofluorescence and Immunohistochemical Analysis.

Integrin expression determined by flow cytometry was performed as described previously by Meyer et al.(24). Immunofluorescence was performed using standard techniques. Briefly, cells were fixed in 3% paraformaldehyde, washed in PBS/0.1% Tween 20, and blocked with 3% BSA/PBS for 30 min. Primary antibodies were incubated overnight at 4°C and washed with PBS/0.1% Tween 20. Alexa 488 conjugated to goat antirabbit IgG were incubated at room temperature for 1 h. Samples were washed as above and incubated with Toto-3 (Molecular Probes; 1:1000) and RNase (50 μg/ml) for 30 min at room temperature. Images were captured on a Leica MRC 1024 laser confocal microscope (Bio-Rad, Regents Park, NSW, Australia). Confocal imaging of αvβ3 and IGFBP-2 in tumors’ 5-μm sections were processed as above with the inclusion of Alexa 568-conjugated donkey antigoat IgG.

Western Blot and Immunoprecipitation Analysis.

Cell lysates were prepared in modified radioimmunoprecipitation assay buffer [50 mm Tris-HCl (pH 7.4), 1% NP40, 0.25% Na deoxycholate, and 150 mm NaCl] with EDTA-free Complete Protease Inhibitors (Roche, Kew, Vic, Australia). Resulting cell lysates were sonicated, and protein concentrations were determined by bicinchoninic acid protein assay following the manufacturer’s protocol (Pierce). For immunoprecipitations, cell lysates were precleared with protein G-Sepharose coupled to mouse IgG. Precleared lysates were incubated overnight at 4°C with primary antibodies, followed by a 1-h room temperature incubation with protein G-Sepharose. The complexes were washed in radioimmunoprecipitation assay buffer and eluted for electrophoresis in reducing SDS-PAGE sample buffer. Crude membrane extracts were isolated (25) with the following modifications. Cells were scraped, resuspended in PBS, and washed three times before cell lysis in 20 mm HEPES (pH 7.4) and 250 mm sucrose with EDTA-free Complete Protease Inhibitors. Lysates were freeze thawed and passed through a 25-gauge needle 10 times. For Western blot analysis, cell lysates and eluted immunocomplexes were electrophoresed on a 10% SDS-PAGE gel under reducing conditions. Proteins were transferred to polyvinylidene difluoride membranes, probed with appropriate antibodies, and incubated with horseradish peroxidase-conjugated secondary antibodies or horseradish peroxidase-conjugated streptavidin in respect to brIGFBP-2. Western blots were visualized by enhanced chemiluminescence detection system according to manufacturer’s instructions.

Analysis of Gene Expression Analysis By Quantitative Reverse Transcription-PCR.

Total RNA was isolated using Qiagen RNeasy midi-kit according to the manufacturer’s recommendations (Qiagen, Hilden, Germany). Total RNA (1 μg) was used to generate cDNA using SuperScript II reverse transcriptase (Invitrogen) and anchored oligo(dT) primers according to manufacturer’s instructions. Quantitative reverse transcription-PCR was performed on an ABI Prism 5700 Sequence Detection system (PE Applied Biosystems, Sydney, NSW, Australia) on cDNA generated from an equivalent of 20 ng of RNA in 10 mm Tris-HCl (pH 8.0), 2.5 mm MgCl2, 50 mm KCl, 200 μm deoxynucleoside triphosphates, 1/40,000 dilution of SYBR Green I (Molecular Probes), 1 μg/ml 6-carboxy-X-rhodamine (Molecular Probes), 8% DMSO, 200 nm primers, and 0.625 unit of AmpliTaq Gold polymerase (Applied Biosystems) per 25-μl reaction. Reaction conditions were 95°C for 10 min followed by 50 cycles of 95°C for 15 s and 60°C for 1 min. Melt curve analysis was performed at the end of each run from 60°C to 95°C. Primer sequences are available on request.

Cell Migration Analysis.

Cell migration was determined using a 48-well microchemotaxis chamber assay using collagen I, IV and vitronectin-coated, 8-mm polycarbonate membranes (Neuroprobe, Gaithersburg, MD) as described previously (26). Briefly, IGF-I, IGF-II, and IGF-I (E3R) were used as chemoattractants at the concentrations indicated and added to the bottom wells, whereas cells (1 × 106/ml) were resuspended in serum-free RPMI containing 0.1% BSA (RPMI/BSA) and added to the top chambers (56 μl/well). Chambers were incubated at 37°C in a humidified incubator in an atmosphere of 5% CO2/95% air for 6 h, after which, the filters were removed, fixed, and stained with Diff-Quik (Baxer Scientific, McGaw Park, IL) and mounted on glass slides. Nonmigrated cells were removed by wiping with a cotton swab. At least four random fields of vision per well (×20 objective) were counted for quantitation of cell migration. Triplicate wells were performed in each assay, and the assay was repeated at least three times.

Functional Expression of αvβ3 on MCF-7 Cells.

To examine the role of αvβ3 in the growth and progression of breast cancer, we used the estrogen-dependent human breast cancer cell line, MCF-7, as a model system that lacks expression of αvβ3 (24). Nonclonal populations of MCF-7β3 cells stably expressing full-length β3 and mock-infected counterparts, MCF-7puro, were generated. High expression of αvβ3 heterodimer in the MCF-7β3 cells was determined by fluorescence-activated cell sorting analysis with no detectable expression of the molecule in the MCF-7puro cells (Fig. 1,A). Western analysis showed increased levels of β3 and αv in the MCF-7β3 cells, confirming that β3 expression in MCF-7 cells increases the total amount of αv-subunit within the cell (Fig. 1 B). The ability to coimmunoprecipitate αv and β3 subunits with αv and β3 antibodies further confirmed the presence of heterodimer within the MCF-7β3 as observed previously by fluorescence-activated cell sorting analysis (data not shown).

Cell surface expression of other αv-containing heterodimers, αvβ5, αvβ6, and αvβ8 remained unaltered, as did the level of the β1 subunit (Fig. 1,A). Other integrin subunits that were examined showed no alteration in their respective levels and included α1, α2, α3, and α6 (data not shown). Although there was no overt alteration in cell morphology, the MCF-7β3 cells displayed a flattened morphology on tissue culture plastic when compared with the MCF-7puro cells (data not shown). The αvβ3 integrin localized to the cell periphery within focal adhesion-like structures and displayed a similar cellular distribution to that of focal adhesion kinase (Fig. 1,C). β1 localization and expression in the MCF-7puro and MCF-7β3 cells were similar (Fig. 1 C).

Expression of αvβ3 Results in Inhibition of IGF-Mediated Migration.

Examination of baseline cell proliferation between the MCF-7puro and MCF-7β3 demonstrated no significant difference between the cells; however, significant differences were observed on examination of cell migration. To examine the role of αvβ3 in cell migration, a standard microchemotaxis assay was used, using IGF-I and IGF-II as chemoattractants (27). Surprisingly, the expression of αvβ3 in the MCF7β3 cells significantly reduced the migration of the cells toward IGF-I (Fig. 2, A and B) and IGF-II (data not shown) on both collagen type I and IV at all growth factor concentrations tested. In contrast, the migratory profile of the cells was reversed when assays were performed on vitronectin, with the MCF7β3 cells displaying a significantly higher migratory rate than the MCF-7puro cells (Fig. 2 B). This observation suggested that a negative regulatory signal may be associated with αvβ3 in the MCF-7β3 cells that is sensitive to vitronectin competition.

Differential Sequestration of IGFBP-2 to MCF-7β3 Cells.

To identify if the negative regulatory signal was associated with any members of the IGFBP family, we screened levels of IGFBPs in MCF-7puro and MCF-7β3 cells. Western blot analysis revealed that levels of IGFBP-2 and IGFBP–6 were elevated in the MCF-7β3 cells (Fig. 3,A), whereas no difference was observed in IGFBP-1, IGFBP-3, and IGFBP-5 (data not shown). Given that the greatest differential was observed for IGFBP-2, we focused our investigations on this member. Examination of IGFBP gene expression by real time quantitative reverse transcription-PCR demonstrated that increased levels of IGFBP-2 and IGFBP-6 in the MCF-7β3 cells were not caused by increased gene expression (Fig. 3,B). Moreover, altered responses of the cells toward the IGFs could not be ascribed to altered IGF-IR expression, because this was also similar between the cells (Fig. 3,B). Isolation of crude membrane extracts demonstrated that the differential protein levels of IGFBP-2 were associated with the membrane fraction, suggesting a differential localization of IGFBP-2 between the cell types (Fig. 3 C).

IGFBP-2 Localization Inhibits IGF-Mediated MCF-7 Migration.

To determine whether IGFBP-2 was involved in αvβ3 inhibition of IGF-mediated migration, we used the mutant form of IGF-I, IGF-I (ER3) that has a reduced binding profile toward IGFBPs. Analysis of cellular migration of MCF-7β3 cells demonstrated that the attenuated migration profile toward IGF-I was lost when IGF-I (ER3) was used as the chemoattractant (Fig. 4,A). Moreover, the MCF-7β3 cells showed an increased capacity to migrate toward this mutant form of IGF-I compared with that of the MCF-7puro cells (Fig. 4,A), indicative of IGFBP involvement in the αvβ3-mediated inhibition of migration. To identify whether IGFBP-2 was specifically involved in this inhibition, MCF-7β3 cells were incubated with recombinant IGFBP-2 before examining the cells’ ability to migrate toward IGF-I. IGFBP-2 significantly inhibited cell migration in the MCF-7β3 cells (Fig. 4 B).

VN and 23C6 Reduce Cell-Associated IGFBP-2 in MCF-7β3 Cells.

As vitronectin overcame the suppressive effects of αvβ3 on IGF-mediated migration (Fig. 2,B), we tested whether the presence of vitronectin resulted in altered sequestration of IGFBP-2 in the MCF-7β3 cells. Western blot analysis of cell membrane extract of cells cultured on vitronectin under serum-free conditions demonstrated a marked decrease in cell-associated IGFBP-2 when compared with those cultured on tissue culture plastic (Fig. 5,A). Thus, one mechanism by which vitronectin can overcome the suppressive effects of IGFBP-2 on IGF-mediated effects is through displacement of IGFBP-2 from the cell surface. Moreover, displacement of cell-associated IGFBP-2 by vitronectin suggests an interaction of IGFBP-2 with αvβ3. To test this, brIGFBP-2 was added to MCF-7puro and MCF-7β3 cells. Before the addition of brIGFBP-2, cells were incubated with the αvβ3-blocking monoclonal antibody 23C6 or mouse IgG. Western blot analysis of total cell lysates demonstrated that blockade of αvβ3 with 23C6 had no effect on the low levels of brIGFBP-2 recruitment to the MCF7puro cells (Fig. 5,B). However, there was a marked decrease (4–5-fold) in brIGFBP-2 recruitment to the MCF7β3 cells, suggesting an interaction between the two molecules (Fig. 5 B).

IGFBP-2 Interacts with αvβ3 in a Bimolecular Manner.

Use of streptavidin-coated agarose beads to precipitate brIGFBP-2 from cell lysates resulted in coprecipitation of αvβ3 in the MCF-7β3 cells, demonstrating the association of the two molecules (Fig. 6,A). Moreover, use of a of rsαvβ3, conjugated to NHS-Sepharose beads, was able to bind IGFBP-2 from MCF-7-conditioned media (Fig. 6,B). This interaction could be reduced by preincubation of rsαvβ3 with 23C6 antibody (Fig. 6,B, Lanes 5, 7, and 9). The presence of IGF-I and IGF-II in the conditioned media of the MCF-7β3 had no significant effect on the interaction of rsαvβ3 and IGFBP-2 (Fig. 6,B). Use of the RGD-containing peptide GRGDSP and gPEN were ineffective at inhibiting the interaction of the two molecules, suggesting the interaction to be one that was RGD independent in nature (data not shown). To determine whether the two molecules interacted in isolation, i.e., in a bimolecular manner, we used NHS-Sepharose-coupled rsαvβ3 and biotinylated recombinant IGFBP-2. Incubation of the molecules resulted in formation of a bimolecular complex demonstrating that no accessory molecules are required for the interaction of αvβ3 and IGFBP-2 to occur (Fig. 6 C).

IGFBP-2 Interaction with αvβ3 in MCFβ3 Tumors Correlates with Reduced Tumor Growth.

To identify whether the αvβ3/ IGFBP-2 interaction may also occur in vivo and influence tumor growth, we inoculated MCF-7puro and MCF-7β3 cells into the mammary fat pad of BALB/c nu/nu mice. Tumor growth was monitored over a 50-day period. The MCF-7puro tumors were found to be significantly larger in size than the MCF-7β3 tumors (Fig. 7,A). The histology of the tumors showed no significant differences, with the differential expression of αvβ3 being maintained in vivo between the cell types (Fig. 7,B). Interestingly, in correlation with reduced tumor growth, MCF-7β3 tumors were found to efficiently localize IGFBP-2 that was not as evident in the MCF-7puro tumors (Fig. 7,B). Confocal microscopy of tumor tissue from the MCF-7β3 tumors demonstrated a clear association between αvβ3 and IGFBP-2 (Fig. 7 C). Therefore, the association of αvβ3 with IGFBP-2 not only occurred in vitro but also persisted in an in vivo model and was associated with reduced tumor growth.

Our results clearly demonstrate that expression of αvβ3 in a human breast cancer cell line inhibits IGF-I- and IGF-II-mediated migration through an interaction with IGFBP-2. This interaction is reversed by vitronectin and the αvβ3-specific blocking antibody, 23C6 (Fig. 6, A and B), resulting in enhanced IGF-mediated migration. In addition, the interaction was not perturbed by GRGDSP peptide and gPEN, suggesting the interaction to be RGD independent in nature. Moreover, in addition to the in vitro association of the molecules (Fig. 6,C), the interaction was also observed to persist in vivo (Fig. 7 C) with IGFBP-2 localizing more efficiently to αvβ3-expressing tumors and associated with reduced tumor growth.

The interplay between the IGF axis and αvβ3 is well established with respect to the integrin positively modulating IGF signaling and biological actions, e.g., maximal IGF signaling is achieved when αvβ3 interacts with vitronectin, resulting in increased migration and proliferation (18, 19, 28). In contrast, we have identified that αvβ3 plays a role in abrogating IGF-I- and IGF-II-mediated migration through its interaction with IGFBP-2. The exact mechanism by which vitronectin/αvβ3 interaction enhances IGF actions is most likely achieved by enhancement of a positive or removal of a negative signal in relation to IGF signaling. We propose that IGFBP-2 represents the negative signal that VN displaces.

Two models can be proposed by which αvβ3 interacting with IGFBP-2 influences IGF action. Firstly, sequestration of IGFBP-2 to the cell surface by αvβ3 may result in the formation of a decoy receptor that competes for free IGF-I and IGF-II with IGF-R. The binding of free IGFs will ultimately reduce IGF-I and IGF-II bioavailability and subsequent IGF signaling and action. Consistent with this hypothesis, Reeve et al.(29) has shown that cell surface bound IGFBP-2 was able to sequester free IGF-I away from the IGF-I receptor. Moreover, we can conclude that this model exists within the MCF-7 system, because use of mutant IGF-I with reduced IGFBP binding capacity, IGF-I (ER3), overcomes the observed inhibition by αvβ3 expression in the MCF-7 cells.

A second model is whereby binding of IGFBP-2 may directly influence integrin-mediated signaling, such as outside-in signaling, to regulate IGF biological actions, e.g., Maile and Clemmons (30) has demonstrated that interaction of echistatin, an inhibitor of αvβ3 function, enhanced the recruitment of the phosphatase SHP-2 to the membrane, resulting in reduced phosphorylation of IGF-IR and the subsequent reduction of downstream signaling from the receptor. Among the IGFBPs, IGFBP-1 and IGFBP-2 both possess an RGD sequence. Jones et al.(31) have shown previously that IGFBP-1 can directly bind α5β1 integrin, resulting in stimulation of cell migration in an RGD-dependent manner. In a similar manner, IGFBP-2 has been shown to have an RGD-dependent cell surface association in the breast cancer cell line Hs578T (32). The addition of exogenous IGFBP-2 was shown to bind in an α5β1-dependent manner and result in the reduced phosphorylation of focal adhesion kinase (32). In contrast to these findings, we have determined that the interaction of IGFBP-2 with αvβ3 can be RGD independent. In agreement with our observation, Hoeflich et al.(33) recently demonstrated that mutation of the RGD sequence in IGFBP-2 had no effect on plasma membrane association and growth inhibitory effects in vivo. It will be important to determine whether IGFBP-2’s interaction with αvβ3, although RGD independent in nature, may also influence focal adhesion kinase phosphorylation and its downstream signaling (34).

The IGF system influences tumor growth and development, with IGF-IR being associated with decreased survival and the presence of its ligand, IGF-I in serum, associated with tumor development (17, 20). Previous work has highlighted that IGF-I plays a major role in the in vitro and in vivo growth of the MCF-7 cell line used in this study (35, 36, 37). Moreover, MCF-7 cells have been shown previously to grow in athymic nude mice via an IGF-dependent mechanism (38, 39). The observation that the interaction of αvβ3 and IGFBP-2 not only effects cell migration in vitro but also correlates with the inhibition of MCF-7 mammary fat pad tumor growth indicates that this interaction may play an important role in tumor growth and progression. Therefore, αvβ3 expression in breast tumors may act in concert with IGFBP-2 to suppress IGF-mediated proliferation and migration, subverting tumor progression and metastases. In agreement with this, a number of studies has shown reduced expression of αvβ3 in breast tumors when compared with that of non-neoplastic breast tissue and benign tumors (40, 41). Thus, down-regulation of αvβ3 may represent a mechanism by which the suppressive effects of the integrin on the IGF axis may be overcome in the tumor microenvironment.

In contrast to the reduced expression of αvβ3 in primary breast tumors, αvβ3 is highly expressed in bone-residing breast tumor metastases (42). This suggests that these tumor cells have overcome the suppressive effects of αvβ3 on the IGF axis by alternative mechanisms, such as proteolytic modification of IGFBP-2, or the presence of molecules in the bone environment that can interfere with IGFBP-2 binding to αvβ3, such as vitronectin, osteopontin, or bone sialoprotein.

IGFBP-2 expression has been observed to be significantly higher in breast tumors when compared with normal adjacent tissue, suggesting that increased expression of IGFBPs is a feature associated with the malignant transformation of breast tissue (43, 44). Recently, the expression of IGFBP-2 was shown to have an inverse correlation to breast cancer risk in premeopausal women (45). Observations demonstrating that IGFBPs can act to suppress IGF signaling and biological actions in cancer have led to the postulation that IGFBPs may prove efficacious in cancer therapy (46). Our findings demonstrate that IGFBP-2 binding to αvβ3 may aid in the inhibitory actions of the binding protein, and therefore, identification of patients with αvβ3-positive tumors may result in better therapeutic outcomes.

However, in contrast to its documented role in suppressing IGF signaling pathways and downstream effects, a number of studies has shown that IGFBP-2 can also have a role in tumor growth and progression. Recently, IGFBP-2 has been shown to be associated with prostate and glioma progression (47, 48, 49), and it will be important to identify whether the actions of IGFBP-2 on growth and progression are linked with expression of αvβ3 and whether dissociation of this interaction could prevent deleterious actions of IGFBP-2, thus providing a novel therapeutic target. In conclusion, we have identified a novel interaction between αvβ3 and IGFBP-2 that represents a novel regulatory link between αvβ3 and the IGF system. Moreover, we postulate that this interaction may also be involved in IGFBP-2-mediated actions that are independent of the IGFs and play an important role in inhibiting or enhancing aspects of tumor growth and progression.

Grant support: AICR Grant 99-081 (to J. T. Price), the Susan G. Komen Foundation Grant BCTR 0100915 (to E. W. Thompson and J. T. Price), the Victorian Breast Cancer Research Consortium (to E. W. Thompson), the Cancer Council of Victoria Grant #65 (to J. Rossjohn), and a Wellcome Trust Senior Research Fellowship in Biomedical Science in Australia (to J. Rossjohn).

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.

Note: J. Rossjohn and J. T. Price contributed equally to this work.

Requests for reprints: John T. Price, Breast-Bone Metastasis/Cell Migration Laboratory, St. Vincent’s Institute of Medical Research, 41 Victoria Parade, Fitzroy, 3065, Australia. Phone: 61 3 9288 2569; Fax: 61 3 9416 2676; E-mail: [email protected]

Fig. 1.

Integrin expression on MCF-7puro and MCF-7β3 cells. In A, expression levels of αvβ3, αvβ5, αvβ6, αvβ8, αv, and β1 were examined by fluorescence-activated cell sorting with fluorescence intensity indicated in arbitrary units on the X axis. Expression of the β3 subunit increased αv at the cell surface, whereas other integrins remained unaltered. In B, Western blot analysis of cell lysates (20 μg) confirmed the expression of β3 and increased expression of αv in MCF-7β3 cells. The increased level of αv expression is as a result of the increased levels of β3 expression in the MCF-7 β3 cells; β-tubulin represents the control protein. In C, immunofluorescence analysis of the cells identified the localization of αvβ3 to the cell membrane within focal adhesion structures (arrows). A pattern similar to that of focal adhesion kinase was observed.

Fig. 1.

Integrin expression on MCF-7puro and MCF-7β3 cells. In A, expression levels of αvβ3, αvβ5, αvβ6, αvβ8, αv, and β1 were examined by fluorescence-activated cell sorting with fluorescence intensity indicated in arbitrary units on the X axis. Expression of the β3 subunit increased αv at the cell surface, whereas other integrins remained unaltered. In B, Western blot analysis of cell lysates (20 μg) confirmed the expression of β3 and increased expression of αv in MCF-7β3 cells. The increased level of αv expression is as a result of the increased levels of β3 expression in the MCF-7 β3 cells; β-tubulin represents the control protein. In C, immunofluorescence analysis of the cells identified the localization of αvβ3 to the cell membrane within focal adhesion structures (arrows). A pattern similar to that of focal adhesion kinase was observed.

Close modal
Fig. 2.

Expression of αvβ3 suppresses insulin-like growth factor (IGF)-I-mediated migration. In A, MCF-7puro cells display an enhanced migratory profile toward IGF-I (5–100 ng/ml) on collagen I when compared with that of the MCF-7β3 cells. In B, migration toward IGF-I (20 ng/ml) over different ECM substrates demonstrates that the inhibitory effect of αvβ3 expression on the MCF-7 cells can be overcome by the presence of vitronectin. Results are expressed as mean ± SD. ∗P < 0.05, ∗∗P < 0.01, Student’s t test.

Fig. 2.

Expression of αvβ3 suppresses insulin-like growth factor (IGF)-I-mediated migration. In A, MCF-7puro cells display an enhanced migratory profile toward IGF-I (5–100 ng/ml) on collagen I when compared with that of the MCF-7β3 cells. In B, migration toward IGF-I (20 ng/ml) over different ECM substrates demonstrates that the inhibitory effect of αvβ3 expression on the MCF-7 cells can be overcome by the presence of vitronectin. Results are expressed as mean ± SD. ∗P < 0.05, ∗∗P < 0.01, Student’s t test.

Close modal
Fig. 3.

Expression of αvβ3 enhances localization of IGF binding protein (BP)-2 to MCF-7 cells. In A, Western blot analysis of MCF-7puro and MCF-7β3 cells demonstrates increased levels of IGFBP-2 and IGFBP-6. In B, real time quantitative reverse transcription-PCR of the IGFBPs (1–6) and IGF-I type I receptor (IGF-IR) showed no difference in gene expression. L32 was used for normalization. In C, membrane extracts were probed with anti-IGFBP-2, anti-αv, and anti-β3 showing increased levels of IGFBP-2 associating with the MCF-7β3 cell membrane.

Fig. 3.

Expression of αvβ3 enhances localization of IGF binding protein (BP)-2 to MCF-7 cells. In A, Western blot analysis of MCF-7puro and MCF-7β3 cells demonstrates increased levels of IGFBP-2 and IGFBP-6. In B, real time quantitative reverse transcription-PCR of the IGFBPs (1–6) and IGF-I type I receptor (IGF-IR) showed no difference in gene expression. L32 was used for normalization. In C, membrane extracts were probed with anti-IGFBP-2, anti-αv, and anti-β3 showing increased levels of IGFBP-2 associating with the MCF-7β3 cell membrane.

Close modal
Fig. 4.

Inhibition of IGF-I-mediated migration by αvβ3 is IGFBP-2 dependent. A, cell migration toward IGF-I and IGF-I (ER3) on collagen I. B, cells were preincubated with rIGFBP-2 (0–200 ng/ml) and washed in DMEM/0.1% BSA to remove unbound IGFBP-2 before migration analysis. Migration was toward IGF-I (20 ng/ml) on a vitronectin-coated filter. Results are expressed as mean ± SD. ∗∗P < 0.01, Student’s t test.

Fig. 4.

Inhibition of IGF-I-mediated migration by αvβ3 is IGFBP-2 dependent. A, cell migration toward IGF-I and IGF-I (ER3) on collagen I. B, cells were preincubated with rIGFBP-2 (0–200 ng/ml) and washed in DMEM/0.1% BSA to remove unbound IGFBP-2 before migration analysis. Migration was toward IGF-I (20 ng/ml) on a vitronectin-coated filter. Results are expressed as mean ± SD. ∗∗P < 0.01, Student’s t test.

Close modal
Fig. 5.

IGFBP-2 interaction with αvβ3 can be inhibited by vitronectin and 23C6. In A, cell lysates (30 μg) from cells cultured on tissue culture plastic and vitronectin were analyzed by Western blot for IGFBP-2. Reduction of IGFBP-2 localization was observed in the MCF-7β3 cells when cultured on vitronectin. A reduction was also seen in the MCF-7puro cells, suggesting the presence of another vitronectin receptor that may have the ability to bind IGFBP-2. Densitometry is represented graphically, and relative intensity is calculated with respect to β-tubulin. In B, cells at 70% confluence were serum starved overnight and lifted nonenzymatically, washed in serum-free media, and preincubated with 23C6 (50 μg/ml) or mouse IgG (50 μg/ml) for 30 min. Biotinylated rIGFBP-2 was then added and incubated with the cells for 1 h. Western blot analysis for αv, β3, and brIGFBP-2 was then performed on these samples. Preincubation of MCF-7β3 with 23C6 significantly reduced the ability of IGFBP-2 to associate with the cells (Lanes 4 and 8, brIGFBP-2), whereas 23C6 had no effect on IGFBP-2 association with the MCF-7puro cells.

Fig. 5.

IGFBP-2 interaction with αvβ3 can be inhibited by vitronectin and 23C6. In A, cell lysates (30 μg) from cells cultured on tissue culture plastic and vitronectin were analyzed by Western blot for IGFBP-2. Reduction of IGFBP-2 localization was observed in the MCF-7β3 cells when cultured on vitronectin. A reduction was also seen in the MCF-7puro cells, suggesting the presence of another vitronectin receptor that may have the ability to bind IGFBP-2. Densitometry is represented graphically, and relative intensity is calculated with respect to β-tubulin. In B, cells at 70% confluence were serum starved overnight and lifted nonenzymatically, washed in serum-free media, and preincubated with 23C6 (50 μg/ml) or mouse IgG (50 μg/ml) for 30 min. Biotinylated rIGFBP-2 was then added and incubated with the cells for 1 h. Western blot analysis for αv, β3, and brIGFBP-2 was then performed on these samples. Preincubation of MCF-7β3 with 23C6 significantly reduced the ability of IGFBP-2 to associate with the cells (Lanes 4 and 8, brIGFBP-2), whereas 23C6 had no effect on IGFBP-2 association with the MCF-7puro cells.

Close modal
Fig. 6.

IGFBP-2 interacts with αvβ3 and is abrogated by 23C6 and bimolecular in nature. In A, brIGFBP-2 was incubated with MCF-7puro and MCF-7β3 cells for 6 h, and cell lysates were generated. Total cell lysate (200 μg) was incubated with streptavidin-coated beads, and complexes were separated by SDS-PAGE and probed for αv, β3, and streptavidin-horseradish peroxidase. Results demonstrate the association of brIGFBP-2 with αvβ3. In B, recombinant soluble αvβ3 (rsαvβ3) was coupled to NHS-Sepharose-coated agarose beads and used to probe MCF-7-conditioned media (CM) with or without IGF-I (20 ng/ml) and IGF-II (20 ng/ml; Lanes 4–9). Before the addition of rsαvβ3 to CM, rsαvβ3 was incubated with 23C6 (50 μg/ml) or IgG (50 μg/ml) for 30 min. As a positive control, an anti-IGFBP-2 antibody was conjugated to NHS-Sepharose beads (Lane 1) with beads alone acting as a negative control (Lane 2). In addition, vitronectin-coated beads were also added as a control (Lane 3). In C, rsαvβ3 was coupled to NHS-Sepharose-coated agarose beads and incubated with biotinylated recombinant IGFBP-2 overnight at 4°C in DMEM/0.1% BSA. The bimolecular complex of IGFBP-2 and αvβ3 was isolated by centrifugation, and biotinylated IGFBP-2 was identified by Western blot analysis via streptavidin-horseradish peroxidase detection. The negative control was blocked with NHS-Sepharose beads to control for nonspecific binding of IGFBP-2 to the beads. Biotinylated recombinant IGFBP-2 was run as a Western blot positive control.

Fig. 6.

IGFBP-2 interacts with αvβ3 and is abrogated by 23C6 and bimolecular in nature. In A, brIGFBP-2 was incubated with MCF-7puro and MCF-7β3 cells for 6 h, and cell lysates were generated. Total cell lysate (200 μg) was incubated with streptavidin-coated beads, and complexes were separated by SDS-PAGE and probed for αv, β3, and streptavidin-horseradish peroxidase. Results demonstrate the association of brIGFBP-2 with αvβ3. In B, recombinant soluble αvβ3 (rsαvβ3) was coupled to NHS-Sepharose-coated agarose beads and used to probe MCF-7-conditioned media (CM) with or without IGF-I (20 ng/ml) and IGF-II (20 ng/ml; Lanes 4–9). Before the addition of rsαvβ3 to CM, rsαvβ3 was incubated with 23C6 (50 μg/ml) or IgG (50 μg/ml) for 30 min. As a positive control, an anti-IGFBP-2 antibody was conjugated to NHS-Sepharose beads (Lane 1) with beads alone acting as a negative control (Lane 2). In addition, vitronectin-coated beads were also added as a control (Lane 3). In C, rsαvβ3 was coupled to NHS-Sepharose-coated agarose beads and incubated with biotinylated recombinant IGFBP-2 overnight at 4°C in DMEM/0.1% BSA. The bimolecular complex of IGFBP-2 and αvβ3 was isolated by centrifugation, and biotinylated IGFBP-2 was identified by Western blot analysis via streptavidin-horseradish peroxidase detection. The negative control was blocked with NHS-Sepharose beads to control for nonspecific binding of IGFBP-2 to the beads. Biotinylated recombinant IGFBP-2 was run as a Western blot positive control.

Close modal
Fig. 7.

Expression of αvβ3 correlates with reduced tumor growth and localization of IGFBP-2 to tumors. Cells were inoculated into the mammary fat pad of 4-week-old BALB/c nu/nu mice. Tumors were measured over 50 days, after which time they were excised. Data points represent mean tumor volume + SE, ∗P < 0.05 (n = 9/group). Results are representative of three independent experiments. In A, MCF-7β3 tumor volume was found to be significantly lower than that of the MCF-7puro tumors. In B, tumors were examined for expression of IGFBP-2 and αvβ3. As expected, MCF-7β3 tumors expressed high levels of αvβ3, whereas MCF-7puro tumors were negative. MCF-7β3 tumors localized increased levels of IGFBP-2 when compared with that of MCF-7puro tumors. In C, MCF-7β3 tumor sections were dual stained with polyclonal anti-IGFBP-2 and 23C6 antibody directed against αvβ3. Green, αvβ3 expression; red, IGFBP-2 expression; yellow, colocalization.

Fig. 7.

Expression of αvβ3 correlates with reduced tumor growth and localization of IGFBP-2 to tumors. Cells were inoculated into the mammary fat pad of 4-week-old BALB/c nu/nu mice. Tumors were measured over 50 days, after which time they were excised. Data points represent mean tumor volume + SE, ∗P < 0.05 (n = 9/group). Results are representative of three independent experiments. In A, MCF-7β3 tumor volume was found to be significantly lower than that of the MCF-7puro tumors. In B, tumors were examined for expression of IGFBP-2 and αvβ3. As expected, MCF-7β3 tumors expressed high levels of αvβ3, whereas MCF-7puro tumors were negative. MCF-7β3 tumors localized increased levels of IGFBP-2 when compared with that of MCF-7puro tumors. In C, MCF-7β3 tumor sections were dual stained with polyclonal anti-IGFBP-2 and 23C6 antibody directed against αvβ3. Green, αvβ3 expression; red, IGFBP-2 expression; yellow, colocalization.

Close modal

We thank Drs. B. Forbes and S. Firth for the provision of antibodies and recombinant proteins and Prof. T. J. Martin, Dr. T. Tiganis, and Asst. Prof. J. Moseley for critical review of this manuscript.

1
Hood J. D., Bednarski M., Frausto R., Guccione S., Reisfeld R. A., Xiang R., Cheresh D. A. Tumor regression by targeted gene delivery to the neovasculature.
Science (Wash. DC)
,
296
:
2404
-2407,  
2002
.
2
van der Flier A., Sonnenberg A. Function and interactions of integrins.
Cell Tissue Res.
,
305
:
285
-298,  
2001
.
3
Price J. T., Bonovich M. T., Kohn E. C. The biochemistry of cancer dissemination.
Crit. Rev. Biochem. Mol. Biol.
,
32
:
175
-253,  
1997
.
4
Eliceiri B. P., Cheresh D. A. Adhesion events in angiogenesis.
Curr. Opin. Cell Biol.
,
13
:
563
-568,  
2001
.
5
Miller W. H., Keenan R. M., Willette R. N., Lark M. W. Identification and in vivo efficacy of small-molecule antagonists of integrin αvβ3 (the vitronectin receptor).
Drug Discov. Today
,
5
:
397
-408,  
2000
.
6
Rusnati M., Tanghetti E., Dell’Era P., Gualandris A., Presta M. αvβ3 integrin mediates the cell-adhesive capacity and biological activity of basic fibroblast growth factor (FGF-2) in cultured endothelial cells.
Mol. Biol. Cell
,
8
:
2449
-2461,  
1997
.
7
Brooks P. C., Stromblad S., Sanders L. C., von Schalscha T. L., Aimes R. T., Stetler-Stevenson W. G., Quigley J. P., Cheresh D. A. Localization of matrix metalloproteinase MMP-2 to the surface of invasive cells by interaction with integrin α v β 3.
Cell
,
85
:
683
-693,  
1996
.
8
Kireeva M. L., Lam S. C., Lau L. F. Adhesion of human umbilical vein endothelial cells to the immediate-early gene product Cyr61 is mediated through integrin αvβ3.
J. Biol. Chem.
,
273
:
3090
-3096,  
1998
.
9
Silletti S., Kessler T., Goldberg J., Boger D. L., Cheresh D. A. Disruption of matrix metalloproteinase 2 binding to integrin α vβ 3 by an organic molecule inhibits angiogenesis and tumor growth in vivo.
Proc. Natl. Acad. Sci. USA
,
98
:
119
-124,  
2001
.
10
Vonlaufen A., Wiedle G., Borisch B., Birrer S., Luder P., Imhof B. A. Integrin α(v)β(3) expression in colon carcinoma correlates with survival.
Mod. Pathol.
,
14
:
1126
-1132,  
2001
.
11
Cooper C. R., Chay C. H., Pienta K. J. The role of α(v)β(3) in prostate cancer progression.
Neoplasia
,
4
:
191
-194,  
2002
.
12
Ria R., Vacca A., Ribatti D., Di Raimondo F., Merchionne F., Dammacco F. α(v)β(3) integrin engagement enhances cell invasiveness in human multiple myeloma.
Haematologica
,
87
:
836
-845,  
2002
.
13
Sturm R. A., Satyamoorthy K., Meier F., Gardiner B. B., Smit D. J., Vaidya B., Herlyn M. Osteonectin/SPARC induction by ectopic β(3) integrin in human radial growth phase primary melanoma cells.
Cancer Res.
,
62
:
226
-232,  
2002
.
14
Borges E., Jan Y., Ruoslahti E. Platelet-derived growth factor receptor β and vascular endothelial growth factor receptor 2 bind to the β 3 integrin through its extracellular domain.
J. Biol. Chem.
,
275
:
39867
-39873,  
2000
.
15
Giancotti F. G., Ruoslahti E. Integrin signaling.
Science (Wash. DC)
,
285
:
1028
-1032,  
1999
.
16
Jones P. L., Crack J., Rabinovitch M. Regulation of tenascin-C, a vascular smooth muscle cell survival factor that interacts with the α v β 3 integrin to promote epidermal growth factor receptor phosphorylation and growth.
J. Cell Biol.
,
139
:
279
-293,  
1997
.
17
Sachdev D., Yee D. The IGF system and breast cancer.
Endocrinol. Relat. Cancer
,
8
:
197
-209,  
2001
.
18
Jones J. I., Prevette T., Gockerman A., Clemmons D. R. Ligand occupancy of the α-V-β3 integrin is necessary for smooth muscle cells to migrate in response to insulin-like growth factor.
Proc. Natl. Acad. Sci. USA
,
93
:
2482
-2487,  
1996
.
19
Clemmons D. R., Horvitz G., Engleman W., Nichols T., Moralez A., Nickols G. A. Synthetic αVβ3 antagonists inhibit insulin-like growth factor-I-stimulated smooth muscle cell migration and replication.
Endocrinology
,
140
:
4616
-4621,  
1999
.
20
Furstenberger G., Senn H. J. Insulin-like growth factors and cancer.
Lancet Oncol.
,
3
:
298
-302,  
2002
.
21
Firth S. M., Baxter R. C. Cellular actions of the insulin-like growth factor binding proteins.
Endocr. Rev.
,
23
:
824
-854,  
2002
.
22
Zhang X., Yee D. Insulin-like growth factor binding protein-1 (IGFBP-1) inhibits breast cancer cell motility.
Cancer Res.
,
62
:
4369
-4375,  
2002
.
23
Metha R. J., Diefenbach B., Brown A., Cullen E., Jonczyk A., Gussow D., Luckenbach G. A., Goodman S. L. Transmembrane-truncated αvβ3 integrin retains high affinity for ligand binding: evidence for an ‘inside-out’ suppressor?.
Biochem. J.
,
330
:
861
-869,  
1998
.
24
Meyer T., Marshall J. F., Hart I. R. Expression of αv integrins and vitronectin receptor identity in breast cancer cells.
Br. J. Cancer
,
77
:
530
-536,  
1998
.
25
Hill M. M., Clark S. F., James D. E. Insulin-regulatable phosphoproteins in 3T3–L1 adipocytes form detergent-insoluble complexes not associated with caveolin.
Electrophoresis
,
18
:
2629
-2637,  
1997
.
26
Price J. T., Tiganis T., Agarwal A., Djakiew D., Thompson E. W. Epidermal growth factor promotes MDA-MB-231 breast cancer cell migration through a phosphatidylinositol 3′-kinase and phospholipase C-dependent mechanism.
Cancer Res.
,
59
:
5475
-5478,  
1999
.
27
Doerr M. E., Jones J. I. The roles of integrins and extracellular matrix proteins in the insulin-like growth factor I-stimulated chemotaxis of human breast cancer cells.
J. Biol. Chem.
,
271
:
2443
-2447,  
1996
.
28
Zheng B., Clemmons D. R. Blocking ligand occupancy of the αVβ3 integrin inhibits insulin-like growth factor I signaling in vascular smooth muscle cells.
Proc. Natl. Acad. Sci. USA
,
95
:
11217
-11222,  
1998
.
29
Reeve J. G., Morgan J., Schwander J., Bleehen N. M. Role for membrane and secreted insulin-like growth factor-binding protein-2 in the regulation of insulin-like growth factor action in lung tumors.
Cancer Res.
,
53
:
4680
-4685,  
1993
.
30
Maile L. A., Clemmons D. R. The αVβ3 integrin regulates insulin-like growth factor I (IGF-I) receptor phosphorylation by altering the rate of recruitment of the Src-homology 2-containing phosphotyrosine phosphatase-2 to the activated IGF-I receptor.
Endocrinology
,
143
:
4259
-4264,  
2002
.
31
Jones J. I., Gockerman A., Busby W. H., Jr., Wright G., Clemmons D. R. Insulin-like growth factor binding protein 1 stimulates cell migration and binds to the α 5 β 1 integrin by means of its Arg-Gly-Asp sequence.
Proc. Natl. Acad. Sci. USA
,
90
:
10553
-10557,  
1993
.
32
Hoeflich A., Reisinger R., Lahm H., Kiess W., Blum W. F., Kolb H. J., Weber M. M., Wolf E. Insulin-like growth factor-binding protein 2 in tumorigenesis: protector or promoter?.
Cancer Res.
,
61
:
8601
-8610,  
2001
.
33
Hoeflich A., Reisinger R., Vargas G. A., Elmlinger M. W., Schuett B., Jehle P. M., Renner-Muller I., Lahm H., Russo V. C., Wolf E. Mutation of the RGD sequence does not affect plasma membrane association and growth inhibitory effects of elevated IGFBP-2 in vivo.
FEBS Lett.
,
523
:
63
-67,  
2002
.
34
Lebrun P., Baron V., Hauck C. R., Schlaepfer D. D., Van Obberghen E. Cell adhesion and focal adhesion kinase regulate insulin receptor substrate-1 expression.
J. Biol. Chem.
,
275
:
38371
-38377,  
2000
.
35
Bentel J. M., Lebwohl D. E., Cullen K. J., Rubin M. S., Rosen N., Mendelsohn J., Miller W. H., Jr. Insulin-like growth factors modulate the growth inhibitory effects of retinoic acid on MCF-7 breast cancer cells.
J. Cell. Physiol.
,
165
:
212
-221,  
1995
.
36
Yee D., Jackson J. G., Kozelsky T. W., Figueroa J. A. Insulin-like growth factor binding protein 1 expression inhibits insulin-like growth factor I action in MCF-7 breast cancer cells.
Cell Growth Differ.
,
5
:
73
-77,  
1994
.
37
Li X., Regezi J., Ross F. P., Blystone S., Ilic D., Leong S. P., Ramos D. M. Integrin αvβ3 mediates K1735 murine melanoma cell motility in vivo and in vitro.
J. Cell Sci.
,
114
:
2665
-2672,  
2001
.
38
Li S. L., Liang S. J., Guo N., Wu A. M., Fujita-Yamaguchi Y. Single-chain antibodies against human insulin-like growth factor I receptor: expression, purification, and effect on tumor growth.
Cancer Immunol. Immunother.
,
49
:
243
-252,  
2000
.
39
Yu H., Rohan T. Role of the insulin-like growth factor family in cancer development and progression.
J. Natl. Cancer Inst. (Bethesda)
,
92
:
1472
-1489,  
2000
.
40
Zutter M. M., Mazoujian G., Santoro S. A. Decreased expression of integrin adhesive protein receptors in adenocarcinoma of the breast.
Am. J. Pathol.
,
137
:
863
-870,  
1990
.
41
Gui G. P., Wells C. A., Browne P. D., Yeomans P., Jordan S., Puddefoot J. R., Vinson G. P., Carpenter R. Integrin expression in primary breast cancer and its relation to axillary nodal status.
Surgery
,
117
:
102
-108,  
1995
.
42
Liapis H., Flath A., Kitazawa S. Integrin α V β 3 expression by bone-residing breast cancer metastases.
Diagn. Mol. Pathol.
,
5
:
127
-135,  
1996
.
43
Pekonen F., Nyman T., Ilvesmaki V., Partanen S. Insulin-like growth factor binding proteins in human breast cancer tissue.
Cancer Res.
,
52
:
5204
-5207,  
1992
.
44
Gebauer G., Jager W., Lang N. mRNA expression of components of the insulin-like growth factor system in breast cancer cell lines, tissues, and metastatic breast cancer cells.
Anticancer Res.
,
18
:
1191
-1195,  
1998
.
45
Krajcik R. A., Borofsky N. D., Massardo S., Orentreich N. Insulin-like growth factor I (IGF-I), IGF-binding proteins, and breast cancer.
Cancer Epidemiol. Biomark. Prev.
,
11
:
1566
-1573,  
2002
.
46
Yee D. The insulin-like growth factor system as a treatment target in breast cancer.
Semin. Oncol.
,
29
:
86
-95,  
2002
.
47
Bubendorf L., Kolmer M., Kononen J., Koivisto P., Mousses S., Chen Y., Mahlamaki E., Schraml P., Moch H., Willi N., Elkahloun A. G., Pretlow T. G., Gasser T. C., Mihatsch M. J., Sauter G., Kallioniemi O. P. Hormone therapy failure in human prostate cancer: analysis by complementary DNA and tissue microarrays.
J. Natl. Cancer Inst. (Bethesda)
,
91
:
1758
-1764,  
1999
.
48
Sallinen S. L., Sallinen P. K., Haapasalo H. K., Helin H. J., Helen P. T., Schraml P., Kallioniemi O. P., Kononen J. Identification of differentially expressed genes in human gliomas by DNA microarray and tissue chip techniques.
Cancer Res.
,
60
:
6617
-6622,  
2000
.
49
Fuller G. N., Rhee C. H., Hess K. R., Caskey L. S., Wang R., Bruner J. M., Yung W. K., Zhang W. Reactivation of insulin-like growth factor binding protein 2 expression in glioblastoma multiforme: a revelation by parallel gene expression profiling.
Cancer Res.
,
59
:
4228
-4232,  
1999
.