Abstract. Introduction: ALCAM, activated leukocyte celladhesion molecule, is connected to the progression of certainsolid tumours and has been shown to be a prominent featurefor tumours that subsequently developed bone metastasis. Thepresent study investigated the biological influence of ALCAMon breast cancer cells in connection with bone biologicalenvironment. Μaterials and Methods: Suitable breast cancercells were transfected with either the ALCAM expressionconstruct or anti-ALCAM transgene, to create sublines thathad differential expression of ALCAM. The growth, migrationand invasion of the cells were evaluated in the presence orabsence of matrix proteins prepared from human bones.Results: ZR-751ΔALCAM (ALCAM knockdown) and MDA-MB-231ALCAMexp (overexpressing ALCAM) were constructed.MDA MB-231ALCAMexp cells showed a slower rate of growthcompared with control cells. However, in the presence of bonematrix proteins, MDA MB-231ALCAMexp showed asignificantly reduced rate of growth, p<0.01 vs. control cells.In contrast, ZR-751ΔALCAM cells grew faster compared withcontrol cells. MDA MB-231ALCAMexp displayed a significantlyreduced (p=0.012) and ZR-751ΔALCAM cells significantlyincreased invasiveness (p=0.02) vs. their respective controlscells. In an ECIS-based cell migration assay, MDA-MB-231ALCAMexp cells showed marked reduction in migration.Inclusion of bone matrix proteins therefore further reducedthe migration speed of MDA MB-231ALCAMexp cells.Conclusion: Loss of ALCAM in breast cancer cells facilitatesthe invasive behaviour of breast cancer and high levels ofALCAM in the cells have a suppressive role in the aggressivenature of breast cancer cells.

Activated leukocyte cell adhesion molecule (ALCAM), alsoknown as melanoma metastasis clone D (MEMD) or CD166,was first identified on activated leukocytes and in haemopoieticstem cells and myeloid progenitors (1, 2). The gene encodingALCAM is located on the long arm of human chromosome 3(3q13.1-q13.2) (1). It is organized into 16 exons spanningnearly 150 kb of DNA (3). ALCAM is a glycoprotein of theimmunoglobulin superfamily that is involved in homophilicadhesion and in heterotypic binding to the lymphocyte cellsurface receptor CD6 (4, 5). ALCAM- mediated cell adhesionis mediated by both homophilic (ALCAM-ALCAM) andheterophilic (ALCAM-CD6) (6) interactions between cells,with the heterophilic interactions being approximately 100times stronger than the homophilic interactions (7).

The influence of ALCAM on cell adhesion has inspired anumber of studies to evaluate its expression in human tumoursincluding melanoma, prostate cancer, breast cancer, colorectalcarcinoma, bladder cancer and oesophageal squamous cellcarcinoma. ALCAM expression has been shown to haveclinical implications. In malignant melanoma, approximatelyhalf of the cases investigated were found to have ALCAMpositivity and the expression is seen in the vertical growthphase of melanoma (6). In prostate cancer, colorectalcarcinoma and oesophageal squamous cell carcinoma,ALCAM expression was lost in high-grade tumours (8-10).ALCAM has also been shown to be involved in capillarytubule formation (11) and in vessel invasion into cartilage invitro (12). Degen et al. were the first to look for the presenceof ALCAM in breast cancer, and performed northern blotanalysis of MEMD (a partner to ALCAM) mRNA and foundexpression in the MCF-7 mammary carcinoma cell line.Studies have more recently shown ALCAM in resected breasttumours (13) and King et al. shown a decreased level ofALCAM expression correlating with the nodal involvement,the grade, the TMN stage, the Nottingham Prognostic IndexNPI and the clinical outcome (local recurrence and death)(13). These finding have recently been substantiated byJezierska et al., who similarly concluded that low ALCAMconcentrations correlated with an aggressive phenotype (14).

1163

Correspondence to: Wen G. Jiang, Metastasis and AngiogenesisResearch Group, Cardiff University School of Medicine, HeathPark, Cardiff CF14 4XN, U.K. Tel: +44 2920742895, Fax: +442920761623, e-mail: jiangw@cf.ac.uk

Key Words: ALCAM, CD166, breast cancer, bone metastasis, celladhesion.

ANTICANCER RESEARCH 30: 1163-1168 (2010)

ALCAM, Activated Leukocyte Cell Adhesion Molecule,Influences the Aggressive Nature of Breast Cancer Cells,

a Potential Connection to Bone MetastasisSIMON DAVIES and WEN G. JIANG

Metastasis and Angiogenesis Research Group, Cardiff University School of Medicine Heath Park, Cardiff CF14 4XN, United Kingdom

0250-7005/2010 $2.00+.40

The published data on the function of ALCAM in breastcancer are not wholly consistent, there is however growingevidence in support of initial findings showing that a reducedlevel of ALCAM expression is an indicator of poor prognosisin breast cancer (13, 14). It has been reported recently thatALCAM expression is particularly aberrant in breast tumourswhich subsequently develop bone metastatis, and thatALCAM expression level is linked to both the clinicaloutcome and the presence of bone metastasis (15).

It is not clear how ALCAM expression may affect thecellular behaviour of breast cancer cells, and whether theexpression profile may influence osteotrophy of breast cancercells. In the present study, a series of breast cancer sublineswith different expression of ALCAM were developed andtheir biological behaviour was tested, including growth,migration and invasiveness. Furthermore, these parameterswere tested in the presence of bone matrix proteins.

Materials and Methods

Materials. Human breast cancer cell lines, MDA MB-231 and ZR 7-51, were obtained from ATCC (American Type Culture Collection,Maryland, USA). Other cells were obtained from ECACC (EuropeanCollection of Animal Cell Culture, Salisbury, England, UK). Theendothelial cell line HECV was from obtained from Interlab Cell lineCollection (ICLC, Naples, Italy). Recombinant human hepatocytegrowth factor/scatter factor (rhHGF/SF) was a gift from Dr. T.Nakamura, Osaka University Medical School, Osaka, Japan. Matrigel(reconstituted basement membrane) was obtained from CollaborativeResearch Products (Bedford, MA, USA). Transwell plates equippedwith a porous insert (pore size 8 μm) were obtained from BectonDickinson Labware (Oxford, UK). DNA gel extraction and plasmidextraction kits were obtained from Sigma (Poole, Dorset, England).Anti-actin and anti-GAPDH were from Santa Cruz BiotechnologiesInc., (Santa Cruz, CA, USA). Molecular biology grade agarose andDNA ladder were obtained from Invitrogen (Carlsbad, CA, USA).Master mix for routine polymerase chain reaction (PCR) and

quantitative PCR (Qabsolute) were obtained from AbGene, Surrey,England, UK). Mouse monoclonal antibody ALCAM/CD166 (forimmunohistochemistry) was obtained from Novocastra Laboratories(Newcastle upon Tyne, UK). Peroxidase conjugated anti-mouseantibodies were from Sigma (Poole, Dorset, UK) and a biotin universalkit was from Dako, Ltd. (Carpinteria, CA, USA). Bone matrix proteinswere prepared from fresh human bone tissues obtained immediatelyafter hip replacement, and collected under the ethical committeeapproval. Bones were crushed at ice cold temperature andsubsequently processed using a BioRuptor instrument (WolfLaboratories, York, UK), in order to extract matrix proteins.

Construction of hammerhead ribozyme transgenes targeting thehuman ALCAM and mammalian expression vector for ALCAM.Hammerhead ribozymes that specifically target ALCAM , based onthe secondary structure of human ALCAM, have been generated aspreviously described (16-18) (primer details shown in Table I). Anti-ALCAM transgenes were used to transfect ZR-751 cells which werestrongly positive for ALCAM (Figure 1A). Following selection oftransfected cells with blasticidin (used at 5 μg/ml) and verification,the following stably transfected cells were established: ALCAMknock-down cells (designated here as ZR751ALCAMrib), plasmid-onlycontrol cells (ZR751pEFa), and the wild type (ZR751WT). Full lengthhuman ALCAM coding region was amplified from a cDNA libraryof human breast tissues using the primers listed in Table I. Reversetranscription was carried out using a reverse transcription(RT) kit(Sigma, Poole, Dorset, England, UK) and amplification using anextensor PCR master mix which has an additional proof readingpolymerase (AbGene Ltd). The ALCAM full length coding productwas similarly cloned into the pEF6 vectors. MDA MB-231 cellswhich were negative for ALCAM were transfected with either thecontrol vector or ALCAM expression vector. Stably transfected cellswere designated as MDA231pEF/His and MDA231ALCAMexp, forcontrol transfection and ALCAM expression, respectively.

RNA preparation and RT-PCR. Total RNA was extracted from cellsusing an RNA extraction kit (AbGene Ltd) and the concentrationwas quantified using a spectrophotometer (Wolf Laboratories, York,England, UK). cDNA was synthesized using a first strand synthesiswith an oligodt primer (AbGene). PCR was performed using sets of

ANTICANCER RESEARCH 30: 1163-1168 (2010)

1164

Table I. Primers used in the present study.

Primers Sequence (5’-3’) Size (bp)

ALCAMrib1F CTG CAG CAC TTT TCC ATT CCT GTA CCA TGT CTG ATG AGT CCG TGA GGA 48ALCAMrib1R ACT AGT TAT CCA GTA TGT CAA TTT TCG TCC TCA CGG ACT 39ALCAMrib2F CTG CAG TAC GTC AAG TCG GAA AGG TCT CTG ATG CGT CCG TGA GGA 45ALCAMrib2R ACT AGT ATG GAG ATA CCA TTT TTC GTC CTC ACG GAC T 37ALCAMrib3F CTG CAG CTG ATC CTT GCA TTA CTC TGA TGA GTC CGT GAG GA 41ALCAMrib3R ACT AGT CAG AAA ACT ACA CTT TGT CTT TTC GTC CTC AGG ACT 42ALCAMF6 TTA TCA TAC CTT GCC GAC TT 19ALCAMR6 GGG TGG AAG TCA TGG TAT AG 20GAPDHF8 GGC TGC TTT TAA CTC TGG TA 20GAPDHR8 GAC TGT GGT CAT GAG TCC TT 20ALCAMExF1 AAT ATG GAA TCC AAG GGG G 19ALCAMExR1 GGC TTC AGT TTT GTG ATT GT 20ALCAMExF2 AAT ATG GAA TCC AAG GGG 18ALCAMExR2 GGC TTC AGT TTT GTG ATT G 19

primers (Table I) with the following conditions: 5 min at 95˚C, andthen 20 sec at 94˚C-25 s at 56˚C, 50 s at 72˚C for 36 cycles, andfinally 72˚C for 7 min. β-actin was amplified and used as a housekeeping control. PCR products were then separated on a 0.8%agarose gel, visualised under UV light, photographed using aUnisavetm camera (Wolf Laboratories) and documented withPhotoshop-CS4 (San Jose, CA, USA) software.

In vitro cell growth assay. This was based on a previously reportedmethod (20) Cells were plated into 96-well plated at 2,000 cells/wellfollowed by a period of up to to 5 day’s incubation. Cells were fixedin 10% formaldehyde on the day of plating and daily for thesubsequent 5 days. 0.5% Crystal violet (w/v) was used to stain cells.Following washing, the stained crystal violet was dissolved with10% (v/v) acetic acid and the absorbance was determined at awavelength of 540 nm using an ELx800 spectrophotometer (WolfLaboratories, York, UK). Absorbance represents the cell number.

Electric cell-substrate impedance sensing (ECIS)-based celladhesion assay. Two models of ECIS instrument were used: ECIS9600 for screening and ECIS 1600R for modelling. In both systems,8W10E arrays which have 10 electrodes in each well were used(Applied Biophysics Inc, NJ, USA) (21-23). Following treatment of

the array surface with a cysteine solution, the arrays were incubatedwith complete medium for 1 h. Respective cells (300,000 per well)in the same volume of medium were added to the arrays andincubated overnight to reach confluence. Baseline electric resistance(without addition of cancer cells) was first recorded, and then cancercell or control medium was added carefully to the cancer cells.Electrical changes were monitored continuously for up to 24 h. Inthe 9600 system, the monitoring was at fixed 30 Hz. In the 1600Rsystem, two conditions were recorded: 400 Hz, 4,000 Hz and 40,000Hz for screening the nature of endothelial changes and 4,000 Hz forfix frequency for cell modelling.

Statistical analysis was carried out using Minitab (version 12)(State College, PA, USA). TheAnderson-Darling test was used fornormality testing and the Student’s t-test for significance testing.

Results

Expression of ALCAM in breast cancer cell lines andestablishing stably transfected sublines. As shown in Figure1B, MDA MB-463, ZR751 and BT474 were positive, MDAMB-436, MCF-7, MDA MB-468 and BT482 were weaklypositive for ALCAM transcripts. In contrast, MDA MB-435,MDA MB-231, BT549 and MDA MB-157 were negative. ZR-751 and MDA MB-231 were selected for knocking down(using ribozyme transgenes) and over-expression, respectively.

Davies and Jiang: ALCAM Influences the Aggressive Nature of Breast Cancer Cells

1165

Figure 1. A: Secondary structure of ALCAM mRNA used to identifytargeting site for ribozymes. B: Expression profile of ALCAM transcriptin human breast cancer cell lines. ZR751 and MDA MB 231 which werefound to be strongly positive and negative, respectively were chosen tobe suitable for transfection. C: Creation of sublines with differentialexpression of ALCAM. Left: ZR-751 transfected with anti-ALCAMribozyme transgenes showing that trangene-1 and -2 were more activethan transgene-3; Right: MDA MB-231 cells were transfected withALCAM expression vector. ALCAMexp4 vector appeared more effective.

Figure 2. A: Changes in absorption, indicating growth, of the differentMDA MB 231 cell lines after overexpressing ALCAM, in differentconcentrations of bone matrix protein (BMP). B: Changes in absorption,indicating growth, of the ZR-751 cell lines after ALCAM expression wasknocked down, in different concentrations of growth medium.

Figure 1C shows that transgene-1 and 2 and expressionconstruct-4 were effective in knocking down the ALCAMtranscript. Stable transfectants were established from thesestrains and used for subsequent studies.

Levels of ALCAM expression breast cancer cells directly affectthe rate of cell growth in bone matrix proteins. When ALCAMwas overexpressed MDA MB-231 cells showed substantiallyreduced rate of growth in the presence of bone matrix proteins(Figure 2A). This is particularly so when the concentration ofBMP was higher than 15%. In contrast, the change of growthpattern was not fully reproduced in ZR-751 cells afterknocking down ALCAM (Figure 2B), which may suggest thatthe incomplete knockdown as seen in Figure 1C may havecontributed to this action.

ALCAM expression impacts on the matrix adhesion of breastcancer cells. Overexpression of ALCAM in MDA MB-231

cells resulted in significant reduction of cell adhesion toextracellular matrix, Matrigel (Figure 3A). Using hepatocytegrowth factor (HGF) as a stimulus, wild-type and controlcells showed a significant increase in the adhesion. However,expression of ALCAM also substantially reduced theadhesion in the cell. In ZR-751 cells, knocking downALCAM increased the adhesion of the cells to Matrigel(Figure 3B). Using a BMP-coated surface, the same reductionof adhesion is shown as when MDA MB-231 cellsoverexpress ALCAM (Figure 3C). Using the ECIS method,the reduction of adhesion to culture surface (Figure 3D) andBMP-coated surface (Figure 3E) were similarly observed.

In vitro invasiveness of breast cancer cells is affected by theexpression profile of ALCAM. Two matrix models of the invitro invasiveness were used: Matrigel and Matrigel mixedwith BMP. As shown in Figure 4A, overexpression ofALCAM in MDA MB-231 cells resulted in significant

ANTICANCER RESEARCH 30: 1163-1168 (2010)

1166

Figure 3. A: Adherence of the different MDA MB-231 cell lines in standard and hepatocyte growth factor (HGF). B: Adherence of the different ZR-751 cell lines in standard and hepatocyte growth factor (HGF). C: Adherence of the different MDA MB-231 cell lines in bone matrix protein (BMP)and a combination of bone matrix protein (BMP) and hepatocyte growth factor (HGF). D: Attachment of MDA MB cell lines pEF6 and Exp 2 to theECIS central electrode over a 6-hour period. E: Attachment of each of the MDA MB cell lines to the ECIS central electrode over a 6-hour periodin standard medium, bone matrix protein (BMP), hepatocyte growth factor (HGF) or a combination of BMP and HGF.

reduction of invasiveness in the Matrigel model. A smallerbut nonetheless similar reduction is seen with the Matrigelmixture (Figure 4C). In contrast, knockdown ALCAM fromZR-751 resulted in an increase in invasiveness of invasioninto Matrigel (Figure 4B).

Discussion

This study shows that ALCAM expression is different indifferent breast cancer cell lines. It is further shown thatoverexpression of ALCAM in breast cancer cells results inmarked reduction of cell growth, adhesion to matrix andmigration. The opposite has been seen when ALCAM isknocked out of breast cancer cells. Data from the presentstudy support the clinical observation that ALCAMexpression in breast cancer may be linked to the likelihoodof the tumour developing bone metastasis (15).

The growth of both MDA MB-231 and ZR-751 cells wasinvestigated in a variety of different concentrations of bothstandard medium and BMP. It was clear that high levels ofALCAM expression (in wild-type ZR-751 and in MDA MB-231ALCAMexp) are associated with cell growth under routineculture conditions and in the presence of BMP. Thus, thepresence of ALCAM reduces the growth rate in cancer cellsand the absence of ALCAM increases it. This conclusioncorrelates well with a clinical study where more aggressivefast growing tumors had very low levels of ALCAM (13).

The ability of a cancer cell to adhere to the either abasement membrane of standard medium or of BMP is thefirst stage in the spread of a cancer cell. As describedpreviously, the ability to adhere to the basement membrane ofa blood vessel wall allows extravasation and subsequentintravasation at a distant site. An increase in the ability of acancer cell to adhere would increase the ability of the cancerto metastasize. The presence of ALCAM is predicted to reducethe ability of cancer cells to adhere to a basement membrane,since the tumors with the lowest levels of ALCAM thatmetastasised in this study. In performing the adhesionexperiments, as well as using standard medium and BMP, thesame media were run with HGF added, since this has beenshown to increase the ability of cancer cells to adhere. Theresults obtained from the adhesion experiment showed thatMDA MB-231 and ZR-751, when manipulated to havedifferent expression pattern of ALCAM, showed a correlationbetween the levels of ALCAM and adhesiveness. The sameinverse correlation has been observed between the level ofALCAM and invasiveness of breast cancer. These results arein line with observations made in a previous clinical study, inwhich tumours which subsequently developed bone metastasistend to have low levels of ALCAM (15).

The present study thus provides new evidence that breastcancer cells expressing low levels of ALCAM are likely toadhere to bone matrix and survive in an environment that also

Davies and Jiang: ALCAM Influences the Aggressive Nature of Breast Cancer Cells

1167

Figure 4. A: Invasion of the MDA MB 231 cell lines through an artificialbasement membrane in the presence of standard medium and hepatocytegrowth factor (HGF). B: Invasion of the ZR-751 cell lines through anartificial basement membrane in the presence of standard medium andhepatocyte growth factor (HGF). C: Invasion of the MDA MB-231 celllines through an artificial basement membrane in the presence ofdifferent concentrations of bone matrix protein (BMP).

contain materials obtained from bone matrix. The method bywhich ALCAM interacts with specific protein(s) from thebone preparation remains to be elucidated further. However,these observations, together with recent reports from clinicalstudies, point strongly to an inhibitory role of ALCAM in thedevelopment of bone metastasis from breast cancer.

Acknowledgements

The authors would like to thank Cancer Research Wales and theAlbert Hung Foundation for supporting this work.

References

1 Bowen MA, DD Patel, Li X, Modrell B, Malacko AR, Wang W-C, Marquardt H, Neubauer M, Pesando JM, Francke U, HaynesBF and Aruffo A: Cloning, mapping, and characterization ofactivated leukocyte-cell adhesion molecule (ALCAM), a CD6ligand. J Exp Med 181: 2213-2220, 1995.

2 Uchida N, Yang Z, Combs J, Pourquie O, Nguyen M,Ramanathan R, Fu J, Welply A, Chen S, Weddell G, Sharma AK,Leiby KR, Karagogeos D, Hill B, Humeau L, Stallcup WB,Hoffman R, Tsukamoto AS, Gearing DP and Peault B: Thecharacterization, molecular cloning, and expression of a novelhematopoietic cell antigen from CD34_ human bone marrowcells. Blood 89: 2706-2716, 1997.

3 Ikeda K and Quertermous T: Molecular isolation andcharacterization of a soluble isoform of activated leukocyte celladhesion molecule that modulates endothelial cell function. JBiol Chem 279: 55315-55323, 2004

4 Burns F, von Kannen S, Guy L, Raper J, Kamholz J and ChangS: DM-GRASP, a novel immunoglobulin superfamily axonalsurface protein that supports neurite extension. Neuron 7: 209-220, 1991.

5 Peduzzi J, Irwin M and Geisert E: Distribution andcharacteristics of a 90 kDa protein, KG-CAM in the rat CNS.Brain Res 640: 296-307, 1994.

6 van Kempen LC, Nelissen JM, Degen WG, Torensma R, WeidleUH, Bloemers HP, Figdor CG and Swart GW: Molecular basisfor the homophilic activated leukocyte cell adhesion molecule(ALCAM)-ALCAM interaction. J Biol Chem 276: 25783-25790,2001.

7 Hassan, NJ, Barclay AN and Brown MH: Frontline: Optimal Tcell activation requires the engagement of CD6 and CD166. EurJ Immunol 34: 930-940, 2004.

8 Kristiansen G, Pilarsky C, Wissmann C, Stephan C, WeissbachL, Loy V, Loening S, Dietel M and Rosenthal A: ALCAM/CD166is up-regulated in low-grade prostate cancer and progressivelylost in high-grade lesions. Prostate 54: 34-43, 2003.

9 Weichert W, Knosel T, Bellach, J, Dieal M and Kistianse G:ALCAM/CD166 is overexpressed in colorectal carcinoma andcorrelates with shortened patient survival. J Clin Pathol 57:1160-1164, 2004.

10 Verma A, Shukla NK, Deo SVS, Gupta SD and Ralham R:MEMD/ALCAM: A potential marker for tumor invasion andnodal metastasis in esophageal squamous cell carcinoma.Oncology 68: 463-470, 2005.

11 Ohneda O, Ohneda K, Arai F, Lee J, Miyamoto T, Fukushima Y,Dowbenko D, Lasky LA and Suda T: ALCAM (CD166): its rolein hematopoietic and endothelial development. Blood 98: 2134-2142, 2001.

12 Arai F, Ohneda O, Miyamoto T, Zhang X and Suda T:Mesenchymal stem cells in perichondrium express activatedleukocyte cell adhesion molecule and participate in bone marrowformation. J Exp Med 195: 1549-1563, 2002.

13 King JA, Ofori-Acquah SF, Stevens T, Al-Mehdi AB, Fodstad Oand Jiang WG: Activated leukocyte cell adhesion molecule inbreast cancer: prognostic indicator. Breast Cancer Res 6: 478-487, 2004.

14 Jezierska A, Matysiak W and Motyl T: ALCAM/CD166 protectsbreast cancer cells against apoptosis and autophagy. Med SciMonit 12: 263-273, 2006.

15 Davies SR, Dent C, Watkins G, King JA, Mokbel K and JiangWG: Expression of the cell to cell adhesion molecule, ALCAM,in breast cancer patients and the potential link with skeletalmetastasis. Oncology Rep 19: 555-561, 2008.

16 Davies G, Jiang WG and Mason MD: Cell–cell adhesion moleculesand signaling intermediates and their role in the invasive potentialof prostate cancer cells. J Urol 163: 985-992, 2000.

17 Jiang WG, Davies G, Martin TA, Parcc C, Watkins G, MasonMD, Mokbel K and Mansel RE: Molecular targeting ofmatrilysin and its impact on tumour growth in vivo, the potentialimplications in breast cancer therapy. Clin Cancer Res 11: 6012-6019, 2005.

18 Jiang WG, Grimshaw D, Lane J, Martin TA, Parr C, Davies G,Laterra J and Mansel RE: Retroviral hammerhead transgenes tocMET and HGF/SF inhibited growth of breast tumour, inducedby fibroblasts. Clin Cancer Res 9: 4274-4281, 2003.

19 Jiang WG, Ablin RJ, Kynaston HG and Mason MD: TheProstate Transglutaminase (TGase-4, TGaseP) regulates theinteraction of prostate cancer and vascular endothelial cells, apotential role for the ROCK pathway. Microvascular Res 77:150-157, 2009.

20 Jiang WG, Hiscox S, Hallett MB, Horrobin DF, Mansel RE andPuntis MCA: Regulation of the expression of E-cadherin onhuman cancer cells by gamma linolenic acid. Cancer Res 55:5043-5048, 1995.

21 Giaever I and Keese CR: Monitoring fibroblast behavior in tissueculture with an applied electric field. Proc Natl Acad Sci USA81: 3761-3764, 1984.

22 Giaever I and Keese CR: Micromotion of mammalian cellsmeasured electrically. Proc Natl Acad Sci USA 88: 7896-7900,1991.

23 Jiang WG, Martin TA, Russell-Lewis J, Ye L, Douglas-Jones Aand Mansel RE: Eplin-alpha expression in human breast cancer,the impact on cellular migration and clinical outcome. MolCancer 7: 71, 2008.

Received February 7, 2010Revised March 26, 2010

Accepted March 26, 2010

ANTICANCER RESEARCH 30: 1163-1168 (2010)

1168