[CANCER RESEARCH 63, 5850–5858, September 15, 2003]

Insulin-like Growth Factor-1 Induces Adhesion and Migration in Human MultipleMyeloma Cells via Activation of �1-Integrin and Phosphatidylinositol3�-Kinase/AKT Signaling1

Yu-Tzu Tai, Klaus Podar, Laurence Catley, Yu-Hua Tseng, Masaharu Akiyama, Reshma Shringarpure,Renate Burger, Teru Hideshima, Dharminder Chauhan, Nicholas Mitsiades, Paul Richardson, Nikhil C. Munshi,C. Ronald Kahn, Constantine Mitsiades, and Kenneth C. Anderson2

The Jerome Lipper Multiple Myeloma Center, Department of Medical Oncology, Dana-Farber; Cancer Institute, Boston, Massachusetts, 02115 [Y -T. T., K. P., L. C., M. A., R. S.,R. B., T. H., D. C., N. M., P. R., N. C. M., C. M., K. C. A.]; Department of Medicine, Harvard Medical School, Boston, Massachusetts 02115 [Y-T. T., K. P., L. C., M. A., R. S.,R. B., T. H., D. C., N. M., P. R., N. C. M., C. M., K. C. A.]; and Research Division, Joslin Diabetes Center, Department of Medicine, Harvard Medical School, Boston,Massachusetts 02215 [Y-H. T., C. R. K.]

ABSTRACT

Insulin-like growth factor-1 (IGF-I) is a growth and survival factor inhuman multiple myeloma (MM) cells. Here we examine the effect of IGF-Ion MM cell adhesion and migration, and define the role of �1 integrin inthese processes. IGF-I increases adhesion of MM.1S and OPM6 MM cellsto fibronectin (FN) in a time- and dose-dependent manner, as a conse-quence of IGF-IR activation. Conversely, blocking anti-�1 integrin mono-clonal antibody, RGD peptide, and cytochalasin D inhibit IGF-I-inducedcell adhesion to FN. IGF-I rapidly and transiently induces association ofIGF-IR and �1 integrin, with phosphorylation of IGF-IR, IRS-1, andp85PI3-K. IGF-I also triggers phosphorylation of AKT and ERK signifi-cantly. Both IGF-IR and �1 integrin colocalize to lipid rafts on the plasmamembrane after IGF-I stimulation. In addition, IGF-I triggers polymer-ization of F-actin, induces phosphorylation of p125FAK and paxillin, andenhances �1 integrin interaction with these focal adhesion proteins. Im-portantly, using pharmacological inhibitors of phosphatidylinositol 3�-kinase (PI3-K) (LY294002 and wortmannin) and extracellular signal-regulated kinase (PD98059), we demonstrate that IGF-I-induced MM celladhesion to FN is achieved only when PI3-K/AKT is activated. IGF-Iinduces a 1.7–2.2 (MM.1S) and 2–2.5-fold (OPM6) increase in migration,whereas blocking anti-IGF-I and anti-�1 integrin monoclonal antibodies,PI3-K inhibitors, as well as cytochalasin D abrogate IGF-I-induced MMcell transmigration. Finally, IGF-I induces adhesion of CD138� patientMM cells. Therefore, these studies suggest a role for IGF-I in traffickingand localization of MM cells in the bone marrow microenvironment.Moreover, they define the functional association of IGF-IR and �1 inte-grin in mediating MM cell homing, providing the preclinical rationale fornovel treatment strategies targeting IGF-I/IGF-IR in MM.

INTRODUCTION

MM3 is a B-cell malignancy characterized by excess expansion ofmalignant plasma cells in the BM in close association with stromalcells. Their localization in the BM results from migration or “homing”of MM cells from the vascular to the extravascular compartment ofthe BM. Little is known about the mechanisms of this homing;

nevertheless, their trafficking to the BM, as well as inside the BMmicroenvironment, is mediated by cell surface adhesion molecules atmultiple steps. A complex network of cytokines and cell adhesionmolecules is produced by BM stromal cells that regulate the prolif-eration, survival, and functional program of MM cells. For example,FN, a major ECM, is abundantly found within the BM microenviron-ment, and adhesion to FN in vitro can be inhibited by mAb to theintegrins and by synthetic peptides that mimic the FN-binding sites(e.g. RGD; Ref. 1).

Integrins are �/�-heterodimeric membrane proteins that mediatecell adhesion to the ECM. Integrin ligand binding by ECMs inducescytoskeletal rearrangement and cell motility in a variety of cell types.In the absence of �1 integrins, hematopoietic stem cells have impairedmigration (2). Although integrin-mediated adhesion is necessary fortumor motility, it is not sufficient. In human MM cells, the integrin�4�1 is one of the main adhesion receptors that mediate tumor cellbinding to FN and vascular cell adhesion molecule (3, 4). �1-Integrin-mediated adhesion of MM cells to FN confers protection againstdrug-induced apoptosis (5, 6) and triggers nuclear factor �B-depend-ent transcription, and secretion of the major MM growth and survivalfactor IL-6 (7). We demonstrated recently costimulation of humanMM cells via vascular endothelial growth factor and �1 integrin (8),supporting �1 integrin inside-out signaling.

The IGF-I is a low molecular weight, single chain polypeptide ofwhich the extensive local production is consistent with its mediatingautocrine or paracrine growth, in addition to its more classical endo-crine mechanism. IGF-I elicits its action on cells by binding to theIGF-IR, which consists of an �- and �-subunit heterodimer: ligand-dependent tyrosine kinase activity rests in the �-subunit. We andothers have shown that IGF-I is a potent growth and survival factor inhuman MM cells (9–14). Specifically, IGF-I activates at least 2distinct PI3-K and MAPK signaling pathways, leading to both pro-liferative and antiapoptotic effects (11–14). The increased growth ofthese cells in the presence of IGF-I requires IGF-IR and can beblocked by a neutralizing �IR3 mAb. IGF-I is also a BM stroma-derived chemoattractant factor for the murine 5T2 MM cells (15). Todate, however, whether IGF-I plays a role in regulating human MMcell adhesion and migration within the BM microenvironment remainsundetermined.

Cholesterol-rich microdomains of the plasma membrane, alsotermed lipid rafts, are implicated in the recruitment of essentialproteins for intracellular signal transduction. Therefore, they providesignaling platforms to coordinate cellular adhesion (16, 17) and trans-membrane signaling (18, 19). In human adenocarcinoma MCF-7 cells,transient redistribution of IGF-IR from nonraft to raft was observedafter IGF-I treatment (18, 20). Membrane compartmentalization be-tween rafts and nonrafts is also required for T-cell activation (21), andT-lymphocyte costimulation is mediated by reorganization of mem-brane raft microdomains (22). Importantly, integrins are lipid raft

Received 4/22/03; revised 6/12/03; accepted 7/3/03.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 with18 U.S.C. Section 1734 solely to indicate this fact.

1 Supported by a Multiple Myeloma Research Foundation Senior Research Award(Y-T. T., C. M., N. M., and T. H.), NIH Grants RO-1 50947 and PO1-78378, as well asthe Doris Duke Distinguished Clinical Research Scientist Award and the Cure forMyeloma Fund (K. C. A.).

2 To whom requests for reprints should be addressed, at Department of Adult Oncol-ogy, Dana-Farber Cancer Institute, M557, 44 Binney Street, Boston, MA 02115. Phone:(617) 632-2144; Fax: (617) 632-2140; E-mail: kenneth_anderson@dfci.harvard.edu.

3 The abbreviations used are: MM, multiple myeloma; BM, bone marrow; IGF,insulin-like growth factor; IGF-IR, type-1 IGF receptor; FN, fibronectin; ECM, extracel-lular matrix; IRS-1, insulin receptor substrate-1; PI3-K, phosphatidylinositol 3�-kinase;MAPK, mitogen-activated protein kinase; ERK, extracellular signal-regulated mitogen-activated protein kinase; mAb, monoclonal antibody; cyt D, cytochalasin D; PMA,phorbol myristate acetate; PLL, poly-L-Lysin; wort, wortmannin; Ab, antibody; chx,cycloheximide; F-actin, filamentous actin.

5850

Research. on July 3, 2021. © 2003 American Association for Cancercancerres.aacrjournals.org Downloaded from Research. on July 3, 2021. © 2003 American Association for Cancercancerres.aacrjournals.org Downloaded from Research. on July 3, 2021. © 2003 American Association for Cancercancerres.aacrjournals.org Downloaded from Research. on July 3, 2021. © 2003 American Association for Cancercancerres.aacrjournals.org Downloaded from Research. on July 3, 2021. © 2003 American Association for Cancercancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/http://cancerres.aacrjournals.org/http://cancerres.aacrjournals.org/http://cancerres.aacrjournals.org/http://cancerres.aacrjournals.org/

associated (16, 17, 23), although the functional relevance of thisassociation remains undefined.

PI3-K and p125FAK have been implicated in integrin-mediated cellmotility in breast cancer cells and smooth muscle cells triggered byIGF-I (24, 25). Recent reports have demonstrated tyrosine phospho-rylation of paxillin and p130cas, and association of Crk with p130cas

after IGF-I stimulation in Swiss 3T3 cells (26), suggesting that IGF-Icross-regulates integrin-dependent signaling pathways. In addition,tumor cell metastasis is regulated by the functional cooperation be-tween IGF-I signaling and integrin �v�5, independent of tumor cellgrowth (27). These findings suggest that integrin ligation, in conjunc-tion with cytokine activation, may play an important role in thedissemination of malignant tumor cells (27, 28). Because there is alack of studies of IGF-I/IGF-IR and its interaction with �1 integrins inhuman MM cells, we in the present study investigated the effect ofIGF-I on MM cell adhesion and migration, and defined IGF-I/IGF-IR-induced signaling pathways mediating these processes. We studiedinteractions between �1 integrin and IGF-I signaling, and showed thatcross-talk between downstream signals of integrin ligand binding andIGF-IR activation are essential for optimal MM cell adhesion andmigration. Furthermore, we defined the functional significance ofmembrane raft patching of IGF-IR and �1 integrin during attachmentand transmigration of MM cells induced by IGF-I.

MATERIALS AND METHODS

Cell Culture. The CD138� human MM-derived cell lines MM.1S andOPM6 were maintained as described (12, 29). Freshly isolated MM cells(CD138�) obtained after informed consent were prepared by positive selec-tion using CD138 microbeads (Mitenyi Biotech, Auburn, CA) according to themanufacturer’s protocol. The MM cells of the selected cell population ex-pressed 85–90% CD138, IGF-IR, and the �1 integrin (data not shown).

Reagents. Human plasma FN and neutralizing anti-IGF-IR mAb �IR3were obtained from Oncogene Research Products (San Diego, CA). IGF-Iwas obtained from PeproTech Inc. (Rocky Hill, NJ). Blocking anti-�1integrin mAb and anti-�-actinin mAb were obtained from Chemicon (Te-mecula, CA); anti-pAKT and PD98059 were purchased from Cell SignalingTechnology (Beverly, MA); anti-IRS-1, anti- p125FAK, anti-p85PI3-K, andanti-phosphotyrosine Ab (anti-pTyr, 4G10) were obtained from UpstateBiotechnology, Inc. (Lake Placid, NY); and anti-IGF-IR, anti-pERK, andanti-Src Abs for immunoblotting were purchased from Santa Cruz Biotech-nology (Santa Cruz, CA). All of the other reagents were purchased fromSigma Chemicals (St. Louis, MO).

Cell Adhesion Assays. Before adhesion, cell lines were starved overnightin RPMI 1640/0.5% BSA, without loss of viability. CD138� patient MM cellswere resuspended in RPMI1640/0.2%BSA (adhesion medium) and used di-rectly after their isolation. Cells (5 � 106/ml) were labeled with calcein-a.m.(Molecular Probes, Eugene, OR) for 30 min at 37°C, washed, and resuspendedin adhesion medium. Cells were stimulated with or without IGF-I at 0–400ng/ml for 15 min and added to FN (20 �g/ml) -coated 96-well plates for 45min. In some experiments, cells were incubated with IGF-I in the presence of100 �M Arg-Gly-Asp (RGD), a peptide that mimics the FN-binding site andblocks binding of �1 integrins to FN in vitro; or with Arg-Gly-Glu (RGE) asa control peptide. To identify the individual role of IGF-I/IGF-IR-signalingmolecules mediating MM adhesion, labeled cells were washed, preincubatedwith or without blocking mAbs against IGR-1R and anti-�1 integrin orinhibitors (wort, LY294002, cyt D, PD98059), and then treated with or withoutIGF-I (100 ng/ml). Treatments with these inhibitors alone produced no signif-icant toxicity, evidenced by trypan blue exclusion at the end of experiments.Cells were then added in triplicate to FN-coated 96-well plates at 37°C for 45min (MM cell lines) or 2 h (MM patient CD138� cells), and unbound cellswere removed by four washes with RPMI 1640. The absorbance of each wellwas measured using 492/520 nm filter set with a fluorescence plate reader(Wallac VICTOR2; Perkin-Elmer, Boston, MA).

Immunoblotting and Immunoprecipitation. Serum-starved MM cells(5 � 106/ml) were treated with or without IGF-I (100 ng/ml) for indicated time

intervals. Some cells were plated on dishes coated with FN (25 �g/ml) andthen treated with or without IGF-I (100 ng/ml) for 10 min. Cell extracts andimmunoblotting were performed as described previously (29). For immuno-precipitation, the extracts were precleared by incubation with 25 �l of normalrabbit IgG serum or normal mouse IgG serum, followed by incubation withprimary Abs diluted in lysis buffer for 2 h at 4°C.

Determination of the Distribution of Proteins between the Detergent-soluble and -insoluble Fractions. Serum-starved OPM6 cells (4 � 107/ml)were stimulated with IGF-I (100 ng/ml) for 0, 5, and 30 min before lysis in 200�l of ice-cold buffer [1% Brij 58, 20 mM Tris (pH 7.5), and 150 mM NaCl withprotease and phosphatase inhibitors; Ref. 30]. Brij 58, like Triton X-100, is arelatively mild, nonionic detergent. Membrane lipid rafts are insoluble innonionic detergents at low temperature. The lysates were centrifuged at14,000 � g for 25 min at 4°C, and the resultant supernatant was the detergent-soluble fraction (S). The detergent-insoluble pellets were resuspended andsonicated briefly in lysis buffer supplemented with 0.5% SDS and 1% �-mer-captoethanol. After centrifugation, the resulting supernatant was the detergent-insoluble fraction (I) of the cell, excluding the insoluble cytoskeletal fraction.This method separates proteins bound to membrane rafts (detergent-insoluble)from those that are not raft-associated (detergent-soluble; Refs. 18, 30). Theproteins were either separated directly by 8% SDS-PAGE and then visualizedby immunoblotting for IGF-IR, p85PI3-K, and Src, or were first immunopre-cipitated with an anti-�1 integrin mAb and then immunoblotted.

Copatching Experiments. Membrane lipid raft aggregation or patchingwas performed as described previously (18, 22). Serum-starved OPM6 cells,either treated with or without 100 ng/ml IGF-I for 10 min, were fixed in 2%paraformaldehyde for 20 min on ice, incubated with FITC-labeled choleratoxin B (FITC-CTx; 8 �g/ml) for 20 min, and then stained with either anti-�1integrin (10 �g/ml) or anti-IGF-IR at 4°C for 30 min. Primary Abs wereadditionally clustered by adding Alexa Fluor 568-conjugated secondary Ab for30 min on ice. After three washes, cells were layered on glass coverslips, fixedin 4% formaldehyde in PBS, and mounted onto slides in antifade solution(Molecular Probes). Cells were visualized using a Zeiss model LSM410confocal laser scanning microscope (Zeiss, New York, NY) equipped with anexternal argon-krypton laser (488 and 568 nm) to detect colocalization ofIGF-IR (red) and GM1 (green), as well as �1 integrin (red) and GM1 (green).Images of optical sections (512 � 512 pixels) were digitally recorded. Theresulting images were processed using Adobe PhotoShop software (MountainView, CA).

Actin Polymerization. Serum-starved cells were treated with or without100 ng/ml IGF-I, permeabilized, fixed, and stained in a single step by theaddition of a 1-ml solution containing 0.1 mg/ml L-lysophosphatidyl-choline,5% formaldehyde in PBS/1% BSA, and 5 units of Alexa Fluor 488 phalloidin(Molecular Probes). Cells were incubated for 20 min at 4°C, washed, andsubjected to flow cytometry. F-actin content induced by IGF-I was expressedas a percentage of control by dividing the mean fluorescence intensity of cellsat each time point to mean fluorescence intensity of the unstimulated cells attime 0.

Transmigration Assay. Chemicon QCM 96-well migration assay (8-�mpore size with FN-coated filter; Chemicon) was performed using cells pre-treated with indicated inhibitors for 60 min at 37°C. Cells (2 � 105/100 �l)were placed in the migration chamber, and 150 �l of adhesion medium with orwithout IGF-I (100 ng/ml) were added to the feeder tray. After 4-h incubation,150 �l of lysis buffer/dye solution was added to each well of the feeder trayand incubated for 15 min. Plates were measured with a fluorescence platereader using 480/520 nm filter set (Perkin-Elmer Life Sciences). Some serum-starved MM cell lines were preincubated for 1 h with or without mAbs againstIGF-IR, �1 integrin, and MOPC21 control IgG1, cyt D, as well as chx.Transmigration assays were also performed as described previously (29).

Statistics. Data are mean � SE. Statistical analysis used the Student t test,with P � 0.05 considered significant.

RESULTS

Effect of IGF-I on MM Cell Adhesion to FN. We first investi-gated whether IGF-I has an effect on MM cell adhesion to FN. Inserum-starved MM cell lines MM.1S and OPM6, IGF-I (100 ng/ml)triggered increased MM cell adhesion as early as 2 min, which

5851

IGF-I-INDUCED CELL ADHESION AND MIGRATION

Research. on July 3, 2021. © 2003 American Association for Cancercancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

reached maximum at 30 min (Fig. 1A). IGF-I at 20–400 ng/ml,physiological achievable concentrations, was next used in dose-response adhesion experiments. IGF-I significantly enhanced celladhesion in a dose-dependent manner, with maximal adhesion at 100and 40 ng/ml of IGF-I for MM.1S (1.7–2.3 fold increase) and OPM6(2.5–3 fold increase) MM cells, respectively (Fig. 1B). As shown inFig. 1C, inhibition of IGF-IR by �IR3 mAb significantly decreasedadhesion compared with control mAb: 55–63% versus 0% inhibitionfor �IR3 mAb versus isotype control MOPC21 IgG. This confirmsthat IGF-I induces increased MM cell adhesion to FN via IGF-IR.

Anti-�1 Integrin mAb and RGD Peptide Block IGF-I-enhancedMM Adhesion to FN. Because �1 integrin mediates MM cell adhe-sion to FN and BM stromal cells, we next investigated the role of �1integrin in IGF-I-enhanced MM cell adhesion. To determine whether�1 integrin mediates IGF-I-induced cell-ECM adhesion, blockinganti-�1 integrin mAb or RGD peptide (100 �M), respectively, wereadded together with 100 ng/ml IGF-I. As shown in Fig. 2, anti-�1integrin mAb (5–20 �g/ml) and RGD peptide (100 �M) significantlydecreased IGF-I-stimulated MM cell adhesion, whereas isotype con-trol IgG and the mutant RGE peptide had no effect. In addition,pretreatment of MM cells with cyt D (0.5 �M), an inhibitor of actinfilament polymerization, completely blocked IGF-I-induced adhesion,indicating that cytoskeletal proteins are involved in �1 integrin-mediated MM cell adhesion induced by IGF-I.

IGF-I Induces Transient Association of IGF-IR and �1 Inte-grin. We next examined IGF-I/IGF-IR signaling in MM.1S andOPM6 cells. Specifically, serum-starved cells were treated with IGF-Ifor up to 1 h and analyzed for tyrosine phosphorylation, and activation

of IGF-IR and its substrate IRS-1, as well as activation of the PI3-K/AKT and MAPK pathways (Fig. 3A). Tyrosine phosphorylation ofIGF-IR was maximal within 2 min and decreased after 20 and 10 minin MM.1S and OPM6 cells, respectively, confirming receptor auto-phosphorylation and internalization induced by IGF-I. Tyrosine phos-phorylation of IRS-1, association of IRS-1 with p85PI3-K, and phos-phorylation of AKT were detectable within 2 min and persisted for atleast 60 min (Fig. 3A). Phosphorylation of ERK1/2 was increased at2 min and remained above basal levels after 60 min of IGF-I treat-

Fig. 1. Effects of IGF-I on adhesion of MM.1S and OPM6 cells to FN. A, serum-starved MM.1S and OPM6 MM cells were labeled, washed, incubated with (�) or without IGF-I(100 ng/ml; f), and added to FN-coated plates for the indicated times. After four washes, attached cells were quantified in a fluorescence analyzer. B, serum-starved MM.1S and OPM6MM cells were labeled, washed, and incubated with 0–400 ng/ml of IGF-I for 15 min. The mixture was then added to FN-coated 96-well plates and incubated for an additional 45min. C, MM.1S (f, �) and OPM6 (F, E) cells were preincubated with a neutralizing �IGF-IR mAb (�IR3; f, F) or isotype control MOPC21 (mouse IgG1; �, E) for 45 min beforeexposure to IGF-I (100 ng/ml). All adhesion data represent the mean of triplicate samples from one representative of three independent experiments; bars, �SE.

Fig. 2. A blocking anti-�1 mAb and RGD peptide block IGF-I-induced MM celladhesion. MM.1S (�) and OPM6 (f) cells were labeled, washed, and incubated with�IR3 (10 �g/ml), anti-�1 integrin (5, 20 �g/ml) mAb, or MOPC21 control IgG (20�g/ml), and cyt D (0.5 �g/ml) for 30 min before exposure to IGF-I (100 ng/ml). Somecells were incubated with RGD or control RGE peptide (100 �M) in the presence of IGF-I.Adhesion data represent the mean of triplicate samples from one representative of threeindependent experiments. �, statistical significance (P � 0.05) of inhibition comparedwith IGF-I treatment; bars, �SE.

5852

IGF-I-INDUCED CELL ADHESION AND MIGRATION

Research. on July 3, 2021. © 2003 American Association for Cancercancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

ment. Importantly, anti-IGF-IR immunoprecipitates from cell extractsimmunoblotted with anti-�1 integrin mAbs demonstrated IGF-IR and�1 integrin complex formation within 2 min. This interaction of �1integrin and IGF-IR was transient, occurring within 2 min and de-creasing after 20 min of IGF-I treatment (Fig. 3A). Because PMA isa strong agonist for the adhesiveness of �1 integrins and regulatesattachment of cells to FN (31, 32), we next prepared cell extracts fromMM cells after 10-min treatment with PMA (100 nM) and IGF-I (100ng/ml), and then immunoprecipitated with anti-�1 integrin mAb. Asshown in Fig. 3B, IGF-IR was detected in anti-�1 integrin immuno-precipitates after either treatment. In addition, pretreatment with �IR3mAb inhibits coimmunoprecipitation of �1 integrin with IGF-IR (Fig.3B). These data confirm that IGF-IR associates with �1 integrin inMM cells and additionally suggest that these receptors may cross-regulate one another in response to IGF-I or PMA.

IGF-IR and �1 Integrin are Colocalized in Membrane LipidRafts upon IGF-I Treatment. To evaluate the functional signifi-cance of the association of IGF-IR with �1 integrin, we next examinedwhether they are colocalized in membrane rafts after IGF-I treatmentin OPM6 MM cells. Membrane rafts, microdomains of the plasmamembrane enriched in cholesterol and sphingolipids, are insolubleafter treatment with nonionic detergents such as Triton X-100 andBrij58 at low temperatures (18, 23, 30). We examined the distributionof complex components between the detergent-soluble (S) and deter-gent-insoluble (I) fractions of OPM6 MM cells, before and after IGF-Istimulation and protein extraction at 4°C. As shown in Fig. 4A,IGF-IR and �1 integrin are present mainly in the S fraction inserum-starved OPM6 before IGF-I treatment, but are primarily in theI fraction after IGF-I stimulation. p85PI3-K, one of the earliest sub-strates activated by the IGF-IR, is similarly induced in the I fraction.Thirty min after IGF-I stimulation, these proteins return from mem-brane rafts (I fraction) to nonrafts (S fraction), demonstrating thetemporal sequence and dynamics of their association with membranerafts and their functional role in mediating IGF-I-stimulated MM celladhesion and subsequent migration. To confirm that proteins in Ifraction had been separated from proteins in S fraction, we immuno-blotted for Src using a pan-Src Ab. Because Src is associated exclu-sively with membrane rafts (33), constitutive expression of Src in theI fraction but not in the S fraction confirms the quality of thepreparation (Fig. 4A). These data demonstrate that membrane rafts are

essential for IGF-I/IGF-IR signaling (18, 19), and additionally showIGF-I-induced migration of �1 integrin from nonrafts to rafts on thecell membrane.

We next validated IGF-I-triggered recruitment of these proteins torafts in intact cells using confocal microscopy. Aggregated lipid raftscan be detected as distinct patches by clustering of raft markers withAbs or other reagents (21). Upon coalescence of lipid rafts into largerdomains, other raft-associated proteins will colocalize with thesepatches. Nonraft-associated proteins do not colocalize with raftpatches, because of the solubility of the different lipid phases. There-fore, serum-starved and IGF-I-treated OPM6 cells were costainedwith anti-�1 integrin mAb (red) and with CTx (green). CTx, whichbinds specifically to raft-ganglioside GM1, was used to visualizemembrane rafts. Cells before and after IGF-I treatment were alsocostained with anti-IGF-IR mAb (red) and with CTx (green). Theseexperiments using anti-�1 integrin mAb and CTx showed little, if any,overlap between �1 integrin and GM1 in unstimulated OPM6 cells(Fig. 4B). However, stimulation with IGF-I increased �1 integrincostained with CTx (yellow), indicating that IGF-I stimulates �1integrin and raft clustering in the cell membrane (Fig. 4B). In parallel,IGF-IR colocalization with CTx after IGF-I stimulation is even moredistinct in IGF-I-treated versus control unstimulated cells (Fig. 4B).These results demonstrate the association of IGF-IR and �1 integrinwith membrane rafts after IGF-I stimulation in intact OPM6 MMcells. Furthermore, when lipid rafts were disrupted using methyl-�-cyclodextrin (5 mM) to deplete cholesterol, IGF-I-induced adhesionwas completely blocked to the baseline (data not shown). Conversely,replenishment with cholesterol (30 �g/ml) after methyl-�-cyclodex-trin treatment completely restores IGF-I-enhanced adhesion. There-fore, �1 integrin-mediated adhesion induced by IGF-I depends onintact membrane rafts.

IGF-I Induces �1 Integrin Interaction with Activated SignalingProteins Localized at Sites of Focal Adhesion. Because pretreat-ment with cyt D abrogates IGF-I-enhanced MM adhesion to FN (Fig.2), we next asked whether IGF-I increased levels of F-actin in OPM6cells. We observed that IGF-I triggers a rapid increase in the polym-erization of F-actin; it is detected as early at 2 min, persists for at least10 min, and decreases afterward (Fig. 5). The increased actin polym-erization occurs during IGF-I-enhanced adhesion.

We next examined whether focal adhesion formation because of �1

Fig. 3. IGF-I stimulates association of IGF-IR and �1 integrin, as wellas activation of PI3-K/AKT and ERK pathways. A, serum-starvedMM.1S or OPM6 cells were treated with IGF-I for the indicated times.Cell lysate (500 �g) was immunoprecipitated with anti-IGF-IR andanti-IRS-1 Abs and then immunoblotted with anti-pTyr mAb. Mem-branes were stripped and reprobed with anti-IGF-IR, anti-�1 integrin,and anti-p85PI3-K Abs. A total of 60 �g of each lysate was also resolvedby 8% SDS-PAGE, and immunoblotted with anti-pAKT and anti-pERKAbs. �-tubulin serves as loading controls. B, serum-starved cells wereincubated with IGF-I (100 ng/ml) or PMA (100 nM) for 10 min orpretreated with �IR3 (5 �g/ml) for 30 min before incubation with IGF-I(100 ng/ml) for 10 min. One mg of cell lysates was immunoprecipitatedwith 4 �g of anti-�1 integrin Abs. Immunoprecipitates were analyzed byimmunoblotting with anti-IGF-IR and anti-�1 integrin Abs.

5853

IGF-I-INDUCED CELL ADHESION AND MIGRATION

Research. on July 3, 2021. © 2003 American Association for Cancercancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

integrin-dependent intracellular signaling proteins (i.e. p125FAK andpaxillin), associated with actin polymerization, is influenced by IGF-I.Tyrosine phosphorylation of these proteins was evaluated by immu-noprecipitation of OPM6 cell lysates using indicated Abs, followed byanti-pTyr immunoblotting. Fig. 6A shows that maximum phosphoryl-

ation of p125FAK and paxillin is evident 10–20 min after IGF-Itreatment. There is a time lag of at least 10 min between maximalIGF-IR and p125FAK/paxillin phosphorylation. The temporal depend-ence of phosphorylation may represent sequential events during IGF-I-controlled MM adhesion and migration. Conversely, IGF-IR block-ing mAb �IR3 inhibits the IGF-I-induced tyrosine phosphorylation ofboth p125FAK and paxillin (Fig. 6B). In the whole cell extracts, theIGF-IR � subunit is clearly visible as Mr 97,000 band, and �IR3 mAbblocks IGF-I stimulation of IGF-IR tyrosine phosphorylation andsubsequent activation of p125FAK and paxillin (Fig. 6B).

To additionally examine whether IGF-I enhances physical interac-tion of �1-integrin with p125FAK, paxillin, and other focal adhesionproteins such as �-actinin, serum-starved OPM6 cells were culturedon FN-coated plates, and then treated with or without IGF-I for 10min. Cell extracts were prepared for immunoprecipitation withanti-�1 integrin mAb followed by immunoblotting with anti-p125FAK, anti-�-actinin, or anti-paxillin Abs. Fig. 6C shows that �1integrin interacts with p125FAK, �-actinin, and paxillin in OPM6 cellsadherent to FN. The interaction between �1 integrin and these focaladhesion proteins is not observed in cells adherent to PLL, which

Fig. 4. IGF-IR and �1 integrin transiently associate withmembrane rafts of OPM6 cells after IGF-I stimulation. A, serum-starved OPM6 cells were treated with IGF-I (100 ng/ml) for theindicated time intervals. Detergent-insoluble (proteins rich inmembrane raft, I) and detergent-soluble (S) cell lysates wereeither separated by 8% SDS-PAGE (IGF-IR, p85PI3-K, Src) orfirst immunoprecipitated with an anti-�1 integrin mAb, and thenimmunoblotted with individual Abs. Data are representative oftwo independent experiments. B, serum-starved (–) or IGF-I-stimulated OPM6 cells were fixed, costained with FITC-CTx �subunit (CTx, green), and either anti-IGF-IR or �1 integrin mAbs,stained with Alexa Fluor 568-conjugated secondary Ab (red), andthen analyzed by confocal microscopy. A representative area(�600) is shown.

Fig. 5. Changes in F-actin content after stimulation with IGF-I. Serum-starved OPM6cells were incubated for 0–15 min with 100 ng/ml of IGF-I (f) or adhesion medium (�),permeabilized, stained with Alexa Fluor 488 conjugated-phalloidin, and analyzed by flowcytometry. Results are mean of three independent experiments; bars, �SE.

5854

IGF-I-INDUCED CELL ADHESION AND MIGRATION

Research. on July 3, 2021. © 2003 American Association for Cancercancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

increases cell adhesion to plastics through nonspecific binding (datanot shown). Importantly, the interaction of �1 integrin with focaladhesion components is enhanced in OPM6 cells after treatment withIGF-I; by densitometric analysis, �1 integrin coimmunoprecipitationwith p125FAK (59%), �-actinin (47%), and paxillin (62%) increases,compared with adhesion of untreated MM to FN (Fig. 6D). Fig. 6Eadditionally shows that pretreatment with 0.5 �M cyt D for 30 mincompletely abrogates p125FAK/paxillin tyrosine phosphorylation in-duced by IGF-I and the baseline activation of these proteins in OPM6

cells adherent to FN. These data indicate that IGF-I cross-regulatesactivation of �1-integrin signaling, including phosphorylation ofp125FAK and paxillin.

Effects of Inhibitors of PI3-K and ERK on IGF-I-induced �1Integrin-dependent MM Adhesion and Transmigration. BecauseIGF-I predominantly induces PI3-K/AKT and ERK1/2 pathways inMM.1S and OPM6 cells (Fig. 3A), we next determined whetherthese pathways mediate IGF-I-enhanced adhesion and transmigra-tion of MM cells using PI3-K and ERK1/2 inhibitors. As shown in

Fig. 6. IGF-I enhances �1 integrin interaction with activated focaladhesion proteins. A, serum-starved OPM6 cells were treated for 0–40min with 100 ng/ml of IGF-I. Immunoprecipitation was performed usingindicated Abs followed by immunoblotting with anti-pTyr mAb. Directimmunoblotting with anti-p125FAK and anti-paxillin Abs serves equalloading controls. B, OPM6 cells were pretreated for 30 min, with orwithout 5 �g/ml �IR3. Tyrosine phosphorylation was assessed 10 minafter IGF-I (100 ng/ml) treatment. IGF-IR tyrosine phosphorylation wasassessed in whole-cell lysates because the addition of �IR3 mAb pre-cluded IGF-IR immunoprecipitation. C, serum-starved OPM6 cells wereplated on dishes coated with FN, and treated with or without IGF-I (100ng/ml) for 10 min. Attached cells were lysed and immunoprecipitatedwith anti-�1 integrin (�1) mAb and with normal IgG serum control.Immunoprecipitates were analyzed by immunoblotting with anti-p125FAK, anti-�-actinin, and anti-paxillin Abs. D, lysates from OPM6cells on FN-coated plates, pretreated with or without cyt D (0.5 �g/ml),and then stimulated with or without 100 ng/ml IGF-I for 10 min, wereimmunoprecipitated with anti-p125FAK and paxillin Abs, followed byimmunoblotting with anti-pTyr mAb. Results are representative of twoindependent experiments.

Fig. 7. IGF-I enhances �1 integrin-mediated MM adhesion in aPI3-K-dependent manner. A, MM.1S (�) and OPM6 (f) cells werepretreated with or without PI3-K inhibitors (LY294002 or wort) or ERKinhibitor (PD98005), and then incubated with or without IGF-I (100ng/ml, �) for 15 min. Adhesion data are the mean of triplicate samplesfrom one representative of three independent experiments; bars, �SE. �,statistical significance (P � 0.05) of inhibition of adhesion in IGF-I-treated cells pretreated without inhibitors versus controls. B, before cellattachment to uncoated-, PLL-coated, or FN-coated plates, MM.1S andOPM6 cells were preincubated with or without LY294002. Adhesionwas assessed in the presence or absence of IGF-I (100 ng/ml). Data arerepresentative of three independent experiments. �, statistical signifi-cance (P � 0.05) of inhibition of cells added to FN versus PLL.

5855

IGF-I-INDUCED CELL ADHESION AND MIGRATION

Research. on July 3, 2021. © 2003 American Association for Cancercancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

Fig. 7A, pretreatment with PI3-K inhibitors LY294002 (25–50 �M)or wort (0.1– 0.3 �M) for 1 h completely prevents IGF-I-inducedMM adhesion to FN, whereas pretreatment with ERK1/2 inhibitorPD98059 has no effect. In the presence of LY294002 (25–50 �M),PD98059 does not additionally inhibit IGF-I-induced MM celladhesion. Thus, IGF-I-induced MM adhesion depends on activa-tion of PI3-K/AKT signaling. We also examined the effect ofLY294002 on IGF-I-induced MM adhesion to uncoated, PLL-coated, or FN-coated plates. Preincubation of MM.1S and OPM6cells with LY294002 (20 –50 �M) completely abolishes the in-creased adhesion to FN induced by IGF-I (Fig. 7B). Importantly,treatment of MM cells with LY294002 (25 or 50 �M) does notaffect cell viability or cell adhesion on uncoated or PLL-coatedplates (Fig. 7B). These results demonstrate that IGF-I regulates �1integrin-mediated MM adhesion in a PI3-K-dependent manner.

We next measured the effect of LY294002 or wort and PD98059 onIGF-I-induced MM cell transmigration. Pretreatment with LY294002or wort for 1 h significantly inhibits the increase in MM cell trans-migration induced by IGF-I (Fig. 8A). In contrast, pretreatment withPD98059 has little effect on IGF-I-induced MM transmigration, con-firming that PI3-K/AKT is a key regulator in this process. In addition,pretreatment with �IR3 mAb, cytD, and anti-�1 integrin mAb abro-gates IGF-I-induced MM transmigration (Fig. 8B). Interestingly, IGF-I-induced MM transmigration does not require de novo protein syn-thesis, because chx pretreatment (chx; 5 �g/ml) for 30 min does notsignificantly abrogate the increased MM transmigration induced byIGF-I.

IGF-I Induces Adhesion of CD138� MM Cells to FN. Finally,we determined whether IGF-I induces adhesion of CD138� patientMM cells. CD138� MM cells from three patient samples wereincubated with or without IGF-I (200 ng/ml) and then added toFN-coated plates. IGF-I induces a 1.4–2.8-fold increase in adherenceto FN; conversely, pretreatment of �IR3 diminishes IGF-I-inducedadhesion (Fig. 9). Because we did not serum-starve freshly isolatedCD138� patient MM cells, �IR3 mAb does not completely abolishIGF-I-induced adhesion to FN. It is likely that other factors in theserum, i.e., vascular endothelial growth factor, contribute to MM celladhesion to FN.

DISCUSSION

We and others have defined the central role of PI3-K/AKT signal-ing in mediating IGF-I-induced MM cell proliferation and antiapo-ptosis (11, 13, 14). In the present study, we show that activation ofPI3-K/AKT and �1 integrin signaling cascades are key regulators inIGF-I-induced cell adhesion, and transmigration in MM.1S andOPM6 MM cells. We demonstrate that IGF-I stimulates adhesion andmigration of human MM cells via IGF-IR. We show that IGF-Iinduces rapid and transient association of activated IGF-IR and �1integrin, and additionally define that association of IGF-IR and �1integrin in membrane rafts is critical for IGF-I-induced cell adhesion.IGF-I activates PI3-K/AKT and ERK signaling, as well as actinpolymerization and focal adhesion complex formation includingp125FAK and paxillin. IGF-I additionally enhances the interaction of�1 integrin with these activated focal adhesion signaling molecules;conversely, inhibition of actin polymerization with cyt D abrogates

Fig. 8. PI3-K inhibitors block IGF-I-induced �1-integrin-dependentMM transmigration. A, effects of LY294002, wort, and PD98059 on thetransmigration of IGF-I-treated MM cells was performed in a 96-welltransmigration assay. Cells were preincubated with or without inhibitorsand allowed to migrate for 4 h to an IGF-I-containing lower chamber.Data represent the mean of triplicate samples from a representative of 2separate experiments; bars, �SE. B, serum-starved MM cells werepretreated with �IR3, anti-�1 integrin (�-�1) mAb, control IgG1 (20�g/ml in all cases), cyt D (0.5 �g/ml), or chx (5 �g/ml) for 30 min, andthen seeded in FN-coated filters separating transwells; IGF-I (100 ng/ml)was added to the lower chamber. The number of viable migrating cellsis the mean per well; bars, �SE.

Fig. 9. IGF-I enhances adhesion of CD138� patient MM cells to FN. CD138� MMcells were isolated from BM aspirates from 3 patients, resuspended in adhesion medium,and subject to adhesion assay as described above in the presence or absence of IGF-I (200ng/ml), with or without preincubation with �IR3 mAb. �, P � 0.05, data represent themean of duplicate wells; bars, �SE.

5856

IGF-I-INDUCED CELL ADHESION AND MIGRATION

Research. on July 3, 2021. © 2003 American Association for Cancercancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

such interactions, confirming the importance of concomitant activa-tion of IGF-I and �1 integrin signaling in mediating IGF-I-inducedcell adhesion and migration. Because IGF-I-induced adhesion andmigration is completely blocked in the presence of specific PI3-Kinhibitors, but not ERK inhibitors, we conclude that IGF-I-stimulatedadhesion and migration is dependent on PI3-K/AKT activity. Finally,IGF-I induces adhesion of CD138� patient MM cells, whereas �IR3mAb blocks this response. Taken together, our results suggest thatIGF-IR and �1 integrin play coordinated roles in enhancing MM celladhesion in the BM milieu, supporting the preclinical rationale fortargeting IGF-I/IGF-IR in novel treatment strategies for MM.

Our data extend a previous report of IGF-I as a chemoattractant inthe mouse 5T2 MM model (15, 34, 35) to human MM cells, providingnew insights into the role of IGF-I/IGF-IR in MM pathogenesis andclinical progression. This pleiotropic growth factor is produced in BMstromal cells, MM cells, osteoblasts, and endothelial cells in the BMmilieu. Our results support a role for IGF-I in MM cell progression, asproposed in a recent clinical study showing that MM patients withhigher serum IGF-I levels have inferior survival (36).

Because IGF-I induces MM cell adhesion very rapidly, earlyIGF-IR signaling events could be decisive. IGF-I stimulation inducestransient association between IGF-IR and �1 integrin (Fig. 3), whichis consistent with a previous report (37). This association could beeither direct or via integrin-associated protein (38). Importantly, ablocking anti-�1 integrin mAb completely blocks IGF-I-induced MMadhesion to FN, confirming a role for �1 integrin in these processes.PI3-K/AKT activity is induced after IGF-I stimulation; and con-versely, IGF-I-induced MM cell adhesion and migration is inhibitedby LY294002 and wort. Moreover, expression of dominant-negativeAKT using adenovirus vectors completely blocks IGF-I-induced MMadhesion and migration (data not shown). Therefore, PI3-K/AKTactivity mediates IGF-I-induced MM cell adhesion and migration. Inour previous study, LY294002 inhibits phosphorylation of AKT butnot of ERK1/2 induced by sCD40L, suggesting that PI3-K and MAPKpathways are independent (13, 29). Our present studies also indicatethe independence of these two pathways. Interestingly, IGF-I-inducedmigration is not dependent on de novo protein synthesis, becausecyclohexamide treatment has no effect on this response. The differ-ential effects of sCD40L and IGF-I on migration may be attributable,at least in part, to differential activation of nuclear factor �B and itstarget genes.

IGF-I induces an increase in actin polymerization, therefore en-hancing interactions among integrin, cytoskeletal, and signaling com-ponents. It contributes, at least in part, to IGF-I-induced MM adhe-sion. Increasing evidence supports a role for IGF-IR in regulatingfocal adhesion molecules, and we show here that p125FAK is tyrosinephosphorylated and activated by IGF-IR. Paxillin, which associateswith p125FAK and is tyrosine phosphorylated upon p125FAK activa-tion, is also tyrosine phosphorylated after IGF-I stimulation of MMcells. Because phosphorylation of these focal adhesion molecules istriggered by �1 integrin binding to FN, these results show that bothIGF-I and �1 integrin activate the focal adhesion pathway. Theseresults confirm that activated IGF-IR rapidly associates with �1integrin after IGF-I treatment (Fig. 3, A and B), demonstrating thefunctional sequelae of costimulation of both receptors. Because thereis a time lag between phosphorylation of p125FAK and paxillin,phosphorylation of p125FAK may be the initial event during celladhesion, with subsequent paxillin phosphorylation followed by de-phosphorylation of both molecules leading to cell migration. Further-more, we observed enhanced phosphorylation of focal adhesion pro-teins after IGF-I treatment of MM cells adherent to FN-coated plates,which was completely abolished by cyt D (Fig. 7D). Taken together,our data define a role for IGF-IR in the regulation of focal adhesion

proteins critical for MM cell adhesion and migration. In addition,these results are in concert with a recent study showing that IGF-I-induced tyrosine phosphorylation of focal adhesion proteins can bedissociated from the activation of the ERK pathway in Swiss 3T3 cells(26).

Segregation between raft and nonraft proteins in unstimulated cellslocalizes molecules during cell adhesion and migration. Our dataindicate that IGF-I triggers clustering of membrane rafts, which coa-lesce into large domains. Moreover, recruitment of activated integrinsinto lipid rafts facilitates upstream cell signaling. Although �1 inte-grins may associate with lipid rafts (16, 17, 39), they may be re-strained by cytoskeletal tethering in a manner that excludes them fromlipid rafts under unstimulated conditions. In our study, �1 integrin islocated predominantly in the soluble fraction of lysates from unstimu-lated cells and stains mainly in the nonrafts of unstimulated cellmembranes. IGF-I induces rapid and transient untethering of �1integrin from the cytoskeleton, its migration into membrane rafts, andits binding again to the cytoskeleton during MM cell adhesion andmigration. Because IGF-I also stimulates actin polymerization, opti-mal interaction of �1 integrin and activated focal adhesion signalingproteins is achieved to ensure these processes. These results demon-strate a complex dynamic sequence of events during cell migrationinduced by IGF-I in human MM cells, and confirm that cell adhesionis a prerequisite for cell migration. In addition, disruption of rafts bychemical depletion of membrane cholesterol impairs IGF-I-inducedcell adhesion, confirming a role of membrane rafts in IGF-I/IGF-IRand �1 integrin signaling. These functional sequelae of lipid rafts inMM cell adhesion are in concert with studies using T cells (16, 21, 22)and breast cancer cells (18, 20). Furthermore, Semac et al. (40)showed recently that Rituximab (mAb targeting CD20) in Burkittlymphoma-derived Raji cells, by redistributing CD20 to rafts, modi-fies their stability and organization, thereby modulating associatedsignaling pathways and C’ defense capacity. Our study suggests thattargeting IGF-I/IGF-IR using a humanized IGF-IR Ab or small mol-ecule may have similar functional sequelae.

In conclusion, our study identifies a chemotactic effect of IGF-I onhuman MM cells mediated via activation of PI3-K/AKT and �1integrin. We demonstrate for the first time that membrane raft asso-ciation with IGF-IR and �1 integrin regulates MM cell adhesion andmigration within the BM milieu. Importantly, these results suggest arole for IGF-I in MM disease progression and support targetingIGF-I/IGF-IR in novel MM treatment strategies.

ACKNOWLEDGMENTS

We thank Dr. Stine-Kathrein Kraeft in the laboratory of Dr. Lan Bo Chen(Department of Cancer Biology, Dana-Farber Cancer Institute) for excellentconfocal microscopy analysis.

REFERENCES

1. Craig, W. S., Cheng, S., Mullen, D. G., Blevitt, J., and Pierschbacher, M. D. Conceptand progress in the development of RGD-containing peptide pharmaceuticals.Biopolymers, 37: 157–175, 1995.

2. Hirsch, E., Iglesias, A., Potocnik, A. J., Hartmann, U., and Fassler, R. Impairedmigration but not differentiation of haematopoietic stem cells in the absence of �1integrins. Nature (Lond.), 380: 171–175, 1996.

3. Uchiyama, H., Barut, B. A., Mohrbacher, A. F., Chauhan, D., and Anderson, K. C.Adhesion of human myeloma-derived cell lines to bone marrow stromal cells stim-ulates interleukin-6 secretion. Blood, 82: 3712–3720, 1993.

4. Lokhorst, H. M., Lamme, T., de Smet, M., Klein, S., de Weger, R. A., van Oers, R.,and Bloem, A. C. Primary tumor cells of myeloma patients induce interleukin-6secretion in long-term bone marrow cultures. Blood, 84: 2269–2277, 1994.

5. Damiano, J. S., Cress, A. E., Hazlehurst, L. A., Shtil, A. A., and Dalton, W. S. Celladhesion mediated drug resistance (CAM-DR): role of integrins and resistance toapoptosis in human myeloma cell lines. Blood, 93: 1658–1667, 1999.

6. Damiano, J. S., and Dalton, W. S. Integrin-mediated drug resistance in multiplemyeloma. Leuk. Lymphoma, 38: 71–81, 2000.

5857

IGF-I-INDUCED CELL ADHESION AND MIGRATION

Research. on July 3, 2021. © 2003 American Association for Cancercancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

7. Chauhan, D., Uchiyama, H., Urashima, M., Yamamoto, K., and Anderson, K. C.Regulation of interleukin 6 in multiple myeloma and bone marrow stromal cells. StemCells, 13 (Suppl. 2): 35–39, 1995.

8. Podar, K., Tai, Y. T., Lin, B. K., Narsimhan, R. P., Sattler, M., Kijima, T., Salgia, R.,Gupta, D., Chauhan, D., and Anderson, K. C. Vascular endothelial growth factor-induced migration of multiple myeloma cells is associated with � 1 integrin- andphosphatidylinositol 3-kinase-dependent PKC � activation. J. Biol. Chem., 277:7875–7881, 2002.

9. Georgii-Hemming, P., Wiklund, H. J., Ljunggren, O., and Nilsson, K. Insulin-likegrowth factor I is a growth and survival factor in human multiple myeloma cell lines.Blood, 88: 2250–2258, 1996.

10. Ge, N. L., and Rudikoff, S. Insulin-like growth factor I is a dual effector of multiplemyeloma cell growth. Blood, 96: 2856–2861, 2000.

11. Mitsiades, C. S., Mitsiades, N., Poulaki, V., Schlossman, R., Akiyama, M., Chauhan,D., Hideshima, T., Treon, S. P., Munshi, N. C., Richardson, P. G., and Anderson,K. C. Activation of NF-�B and upregulation of intracellular anti-apoptotic proteinsvia the IGF-1/Akt signaling in human multiple myeloma cells: therapeutic implica-tions. Oncogene, 21: 5673–5683, 2002.

12. Ogawa, M., Nishiura, T., Oritani, K., Yoshida, H., Yoshimura, M., Okajima, Y.,Ishikawa, J., Hashimoto, K., Matsumura, I., Tomiyama, Y., and Matsuzawa, Y.Cytokines prevent dexamethasone-induced apoptosis via the activation of mitogen-activated protein kinase and phosphatidylinositol 3-kinase pathways in a new multiplemyeloma cell line. Cancer Res., 60: 4262–4269, 2000.

13. Pene, F., Claessens, Y. E., Muller, O., Viguie, F., Mayeux, P., Dreyfus, F., Lacombe,C., and Bouscary, D. Role of the phosphatidylinositol 3-kinase/Akt and mTOR/P70S6-kinase pathways in the proliferation and apoptosis in multiple myeloma.Oncogene, 21: 6587–6597, 2002.

14. Tu, Y., Gardner, A., and Lichtenstein, A. The phosphatidylinositol 3-kinase/AKTkinase pathway in multiple myeloma plasma cells: roles in cytokine-dependentsurvival and proliferative responses. Cancer Res., 60: 6763–6770, 2000.

15. Vanderkerken, K., Asosingh, K., Braet, F., Van Riet, I., and Van Camp, B. Insulin-like growth factor-1 acts as a chemoattractant factor for 5T2 multiple myeloma cells.Blood, 93: 235–241, 1999.

16. Krauss, K., and Altevogt, P. Integrin leukocyte function-associated antigen-1-medi-ated cell binding can be activated by clustering of membrane rafts. J. Biol. Chem.,274: 36921–36927, 1999.

17. Thorne, R. F., Marshall, J. F., Shafren, D. R., Gibson, P. G., Hart, I. R., and Burns,G. F. The integrins �3�1 and �6�1 physically and functionally associate with CD36in human melanoma cells. Requirement for the extracellular domain OF CD36.J. Biol. Chem., 275: 35264–35275, 2000.

18. Manes, S., Mira, E., Gomez-Mouton, C., Lacalle, R. A., Keller, P., Labrador, J. P.,and Martinez, A. C. Membrane raft microdomains mediate front-rear polarity inmigrating cells. EMBO J., 18: 6211–6220, 1999.

19. Podar, K., Tai, Y. T., Cole, C. E., Hideshima, T., Sattler, M., Hamblin, A., Mitsiades,N., Schlossman, R. L., Davies, F. E., Morgan, G. J., Munshi, N. C., Chauhan, D., andAnderson, K. C. Essential role of caveolae in interleukin-6- and insulin-like growthfactor I-triggered Akt-1-mediated survival of multiple myeloma cells. J. Biol. Chem.,278: 5794–5801, 2003.

20. Manes, S., Lacalle, R. A., Gomez-Mouton, C., del Real, G., Mira, E., and Martinez,A. C. Membrane raft microdomains in chemokine receptor function. Semin Immu-nol., 13: 147–157, 2001.

21. Janes, P. W., Ley, S. C., and Magee, A. I. Aggregation of lipid rafts accompaniessignaling via the T cell antigen receptor. J. Cell Biol., 147: 447–461, 1999.

22. Janes, P. W., Ley, S. C., Magee, A. I., and Kabouridis, P. S. The role of lipid rafts inT cell antigen receptor (TCR) signalling. Semin. Immunol., 12: 23–34, 2000.

23. Skubitz, K. M., Campbell, K. D., and Skubitz, A. P. CD63 associates with CD11/CD18 in large detergent-resistant complexes after translocation to the cell surface inhuman neutrophils. FEBS Lett., 469: 52–56, 2000.

24. Jones, J. I., Prevette, T., Gockerman, A., and Clemmons, D. R. Ligand occupancy ofthe �-V-�3 integrin is necessary for smooth muscle cells to migrate in response toinsulin-like growth factor. Proc. Natl. Acad. Sci. USA, 93: 2482–2487, 1996.

25. Doerr, M. E., and Jones, J. I. The roles of integrins and extracellular matrix proteinsin the insulin-like growth factor I-stimulated chemotaxis of human breast cancer cells.J. Biol. Chem., 271: 2443–2447, 1996.

26. Casamassima, A., and Rozengurt, E. Insulin-like growth factor I stimulates tyrosinephosphorylation of p130(Cas), focal adhesion kinase, and paxillin. Role of phosphati-dylinositol 3�-kinase and formation of a p130(Cas) Crk complex. J. Biol. Chem., 273:26149–26156, 1998.

27. Brooks, P. C., Klemke, R. L., Schon, S., Lewis, J. M., Schwartz, M. A., and Cheresh,D. A. Insulin-like growth factor receptor cooperates with integrin � v � 5 to promotetumor cell dissemination in vivo. J. Clin. Investig., 99: 1390–1398, 1997.

28. Arao, S., Masumoto, A., and Otsuki, M. �1 integrins play an essential role in adhesionand invasion of pancreatic carcinoma cells. Pancreas, 20: 129–137, 2000.

29. Tai, Y. T., Podar, K., Mitsiades, N., Lin, B., Mitsiades, C., Gupta, D., Akiyama, M.,Catley, L., Hideshima, T., Munshi, N. C., Treon, S. P., and Anderson, K. C. CD40induces human multiple myeloma cell migration via phosphatidylinositol 3-kinase/AKT/NF-� B signaling. Blood, 101: 2762–2769, 2003.

30. Hostager, B. S., Catlett, I. M., and Bishop, G. A. Recruitment of CD40 and tumornecrosis factor receptor-associated factors 2 and 3 to membrane microdomains duringCD40 signaling. J. Biol. Chem., 275: 15392–15398, 2000.

31. Bohnsack, J. F., Chang, J., Zhou, X., and Yednock, T. A. Mechanisms of � 1integrin-dependent adherence of granulocytic HL60 to fibronectin. J. Leukoc. Biol.,57: 592–599, 1995.

32. Hermanto, U., Zong, C. S., Li, W., and Wang, L. H. RACK1, an insulin-like growthfactor I (IGF-I) receptor-interacting protein, modulates IGF-I-dependent integrinsignaling and promotes cell spreading and contact with extracellular matrix. Mol.Cell. Biol., 22: 2345–2365, 2002.

33. Lacalle, R. A., Mira, E., Gomez-Mouton, C., Jimenez-Baranda, S., Martinez, A. C.,and Manes, S. Specific SHP-2 partitioning in raft domains triggers integrin-mediatedsignaling via Rho activation. J. Cell Biol., 157: 277–289, 2002.

34. Vanderkerken, K., De Greef, C., Asosingh, K., Arteta, B., De Veerman, M., VandeBroek, I., Van Riet, I., Kobayashi, M., Smedsrod, B., and Van Camp, B. Selectiveinitial in vivo homing pattern of 5T2 multiple myeloma cells in the C57BL/KalwRijmouse. Br. J. Cancer, 82: 953–959, 2000.

35. Asosingh, K., Gunthert, U., Bakkus, M. H., De Raeve, H., Goes, E., Van Riet, I., VanCamp, B., and Vanderkerken, K. In vivo induction of insulin-like growth factor-Ireceptor and CD44v6 confers homing and adhesion to murine multiple myelomacells. Cancer Res., 60: 3096–3104, 2000.

36. Standal, T., Borset, M., Lenhoff, S., Wisloff, F., Stordal, B., Sundan, A., Waage, A.,and Seidel, C. Serum insulinlike growth factor is not elevated in patients with multiplemyeloma but is still a prognostic factor. Blood, 100: 3925–3929, 2002.

37. Shakibaei, M., John, T., De Souza, P., Rahmanzadeh, R., and Merker, H. J. Signaltransduction by �1 integrin receptors in human chondrocytes in vitro: collaborationwith the insulin-like growth factor-I receptor. Biochem. J., 342 Pt 3: 615–623, 1999.

38. Maile, L. A., Imai, Y., Clarke, J. B., and Clemmons, D. R. Insulin-like growth factorI increases � v � 3 affinity by increasing the amount of integrin-associated proteinthat is associated with non-raft domains of the cellular membrane. J. Biol. Chem.,277: 1800–1805, 2002.

39. Claas, C., Stipp, C. S., and Hemler, M. E. Evaluation of prototype transmembrane 4superfamily protein complexes and their relation to lipid rafts. J. Biol. Chem., 276:7974–7984, 2001.

40. Semac, I., Palomba, C., Kulangara, K., Klages, N., van Echten-Deckert, G., Borisch,B., and Hoessli, D. C. Anti-CD20 therapeutic antibody rituximab modifies thefunctional organization of rafts/microdomains of B lymphoma cells. Cancer Res., 63:534–540, 2003.

5858

IGF-I-INDUCED CELL ADHESION AND MIGRATION

Research. on July 3, 2021. © 2003 American Association for Cancercancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

[CANCER RESEARCH 63, 7542–7544, November 1, 2003]

Announcements(Requests for announcements must be received at least three months before publication.)

FUTURE ANNUAL MEETINGS OF THE AMERICANASSOCIATION FOR CANCER RESEARCH

2004 March 27–31, Orlando, FL2005 April 16–20, Anaheim, CA

AACR SPECIAL CONFERENCES IN CANCER RESEARCH

A number of meetings are now being organized in the AACR’s series ofsmaller scientific meetings. Following are the topics, dates, locations, andprogram committees for these meetings. When full details of each meeting areavailable, AACR members will be the first to receive complete brochures andapplication forms for participation in these important conferences. Nonmem-bers may receive this information by sending their names and addresses toMeetings Mailing List, American Association for Cancer Research, 615 Chest-nut Street, 17th Floor, Philadelphia, PA 19106-4404. Up-to-date programinformation is also available via the Internet at the AACR’s website (http://www.aacr.org).

AACR-NCI-EORTC INTERNATIONAL CONFERENCE ONMOLECULAR TARGETS AND CANCER THERAPEUTICS

November 17–21, 2003Hynes Center, Boston, MA

ChairpersonsCharles L. Sawyers, Los Angeles, CAEdward A. Sausville, Bethesda, MDJaap Verweij, Rotterdam, The Netherlands

SIXTH JOINT CONFERENCE OF THE AACR AND JCA,ADVANCES IN CANCER RESEARCH

January 25–29, 2004Hilton Wai Koloa Village, Wai Koloa, Hawaii

ChairpersonsWaun Ki Hong, Houston, TXTakahashi Tsuruo, Tokyo, Japan

CALENDAR OF EVENTS

Lung Cancer Awareness Week, November 17–21, 2003. The Great AmericanSmokeout is Thursday, November 20. Toll-free patient support informationline: 1-877-646-LUNG (5864). Website: www.lungcancer.org.

10th Hong Kong International Cancer Congress, November 19–21, 2003,Faculty of Medicine Building, The University of Hong Kong, Hong Kong.Contact: 10th HKICC Congress Secretariat, Department of Surgery, Universityof Hong Kong Medical Centre, Queen Mary Hospital, Hong Kong. Phone:852.2818.0232 or 852.2855.4235; Fax: 852.2818.1186; E-mail: mededcon@hku.hk; Website: www.hkicc.org.

Third International Conference and 9th Annual Meeting of the InternationalSociety of Cancer Chemoprevention (ISCaC): Controversies in Tumor Preventionand Genetics, February 12–14, 2004, University of St. Gallen, Switzerland.E-mail: info@oncoconferences.ch; Website: www.oncoconferences.ch.

The National Comprehensive Cancer Network’s 9th Annual Conference: Clin-ical Practice Guidelines and Outcomes Data in Oncology, March 10–14, 2004.Westin Diplomat Resort and Spa, Hollywood, Florida. Website: www.nccn.org.

The UK Radiological Congress, run by The British Institute of Radiology, TheRoyal College of Radiologists, The Society and College of Radiographers andThe Institute of Physics and Engineering in Medicine, June 6–8, 2004, GMEXEt MICC, Manchester, UK. Abstract deadline: Feb. 2, 2004. Website: www.ukrc.org.uk.

11th Conference on Advances in Neuroblastoma Research, June 16–19, 2004,Genoa, Italy. E-mail: anr2004@neuroblastoma.org; Website: www.anr2004.org.

6th International Conference on Head and Neck Cancer, August 7–11, 2004,Marriott Wardman Park, Washington, DC. Contact: Concepts in Meeting &Events, 1805 Ardmore Boulevard, Pittsburgh, PA 15221. Phone:412.243.5156; Fax: 412.243.5160; E-mail: ssteighnercme@aol.com.

Molecular Targets for Cancer Therapy: 3rd Biennial Meeting, October1–5, 2004, Don Cesar Beach Resort & Spa, St. Petersburg Beach, FL.Contact: Ann Gordon. Phone: 813.903.4975; E-mail: gordonac@ moffitt.usf.edu.

7542

Corrections

In the article by T. Wissniowski et al., titled “Activation of Tumor-specificT Lymphocytes by Radio-Frequency Ablation of the VX2 Hepatoma in Rab-bits,” which appeared in the October 1, 2003 issue of Cancer Research (pp.6496–6500), the names of the first two authors were misspelled. The correctauthor list is as follows: Thaddäus Till Wissniowski, Johannes Hänsler, DanielNeureiter, Markus Frieser, Stefan Schaber, Birgit Esslinger, Reinhard Voll,Deike Strobel, Eckhart Georg Hahn, and Detlef Schuppan.

In the article by B. Ruggeri et al., titled “CEP-7055: A Novel, Orally ActivePan Inhibitor of Vascular Endothelial Growth Factor Receptor Tyrosine Ki-nases with Potent Antiangiogenic Activity and Antitumor Efficacy in Preclin-ical Models,” which appeared in the September 15, 2003 issue of CancerResearch (pp. 5978–5991), two co-authors, Shi Yang (Department of Bio-chemistry, Cephalon, Inc.) and Sonya Pritchard (Department of Oncology,Cephalon, Inc.), were inadvertently omitted from the list of authors. Thecorrect list is as follows: Bruce Ruggeri, Jasbir Singh, Diane Gingrich, ThelmaAngeles, Mark Albom, Shi Yang, Hong Chang, Candy Robinson, KathrynHunter, Pawel Dobrzanski, Susan Jones-Bolin, Sonya Pritchard, Lisa Aimone,Andres Klein-Szanto, Jean-Marc Herbert, Francoise Bono, Paul Schaeffer,Pierre Casellas, Bernard Bourie, Roberto Pili, John Isaacs, Mark Ator, RobertHudkins, Jeffrey Vaught, John Mallamo, and Craig Dionne.

In the article by Y-T Tai et al., titled “Insulin-like Growth Factor-1 InducesAdhesion and Migration in Human Multiple Myeloma Cells via Activation of�1-Integrin and Phosphatidylinositol 3’-Kinase/AKT Signaling,” which ap-peared in the September 15, 2003 issue of Cancer Research (pp. 5850–5858),the Western blotting in the bottom “IGF-1R” panel of figure 3B is incorrect.The correct figure appears below:

Fig. 3. IGF-I stimulates association of IGF-IR and �1 integrin, as wellas activation of PI3-K/AKT and ERK pathways. A, serum-starvedMM.1S or OPM6 cells were treated with IGF-I for the indicated times.Cell lysate (500 �g) was immunoprecipitated with anti-IGF-IR andanti-IRS-1 Abs and then immunoblotted with anti-pTyr mAb. Mem-branes were stripped and reprobed with anti-IGF-IR, anti-�1 integrin,and anti-p85PI3-K Abs. A total of 60 �g of each lysate was also resolvedby 8% SDS-PAGE, and immunoblotted with anti-pAKT and anti-pERKAbs. �-tubulin serves as loading controls. B, serum-starved cells wereincubated with IGF-I (100 ng/ml) or PMA (100 nM) for 10 min orpretreated with �IR3 (5 �g/ml) for 30 min before incubation with IGF-I(100 ng/ml) for 10 min. One mg of cell lysates was immunoprecipitatedwith 4 �g of anti-�1 integrin Abs. Immunoprecipitates were analyzed byimmunoblotting with anti-IGF-IR and anti-�1 integrin Abs.

7543

ANNOUNCEMENTS

In the article by G. Yousef et al., titled “Human Kallikrein 5: A PotentialNovel Serum Biomarker for Breast and Ovarian Cancer,” which appeared inthe July 15, 2003 issue of Cancer Research (pp. 3958–3965), figure 4 wasprinted incorrectly. Below is the correct figure.

Fig. 4. Fractionation of three biological fluids (serum, ascites fluid from an ovariancancer patient, and breast milk) by size-exclusion liquid chromatography. The elutionprofile of molecular mass standards is denoted by arrows. In serum, hK5 elutes as twoimmunoreactive peaks, one with an apparent molecular mass of 50 kDa (fractions 37–39)and one with an apparent molecular mass of approximately 150–180 kDa (fractions31–33). The elution profile of another kallikrein with a similar theoretical molecular mass,hK6, is also shown by dashed lines. This kallikrein elutes at a molecular mass of �35kDa, corresponding to free hK6. In ascites fluid, the same comments apply as for serum.In breast milk, hK5 elutes mainly as a single immunoreactive peak. hK6 elutes as twodistinct peaks, one at a molecular mass of �35 kDa and another one at a molecular massof �100 kDa.

7544

ANNOUNCEMENTS

2003;63:5850-5858. Cancer Res Yu-Tzu Tai, Klaus Podar, Laurence Catley, et al.

-Kinase/AKT Signaling′1-Integrin and Phosphatidylinositol 3βin Human Multiple Myeloma Cells via Activation of

Insulin-like Growth Factor-1 Induces Adhesion and Migration

Updated version

http://cancerres.aacrjournals.org/content/63/18/5850

Access the most recent version of this article at:

Cited articles

http://cancerres.aacrjournals.org/content/63/18/5850.full#ref-list-1

This article cites 38 articles, 26 of which you can access for free at:

Citing articles

http://cancerres.aacrjournals.org/content/63/18/5850.full#related-urls

This article has been cited by 40 HighWire-hosted articles. Access the articles at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

SubscriptionsReprints and

.pubs@aacr.orgDepartment at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. (CCC)Click on "Request Permissions" which will take you to the Copyright Clearance Center's

.http://cancerres.aacrjournals.org/content/63/18/5850To request permission to re-use all or part of this article, use this link

Research. on July 3, 2021. © 2003 American Association for Cancercancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/content/63/18/5850http://cancerres.aacrjournals.org/content/63/18/5850.full#ref-list-1http://cancerres.aacrjournals.org/content/63/18/5850.full#related-urlshttp://cancerres.aacrjournals.org/cgi/alertsmailto:pubs@aacr.orghttp://cancerres.aacrjournals.org/content/63/18/5850http://cancerres.aacrjournals.org/