[CANCER RESEARCH 55, 5043-504K. November 1. 1995]

Regulation of the Expression of E-Cadherin on Human Cancer Cells by y-LinolenicAcid (GLA)1

Wen G. Jiang,2 Steve Hiscox, Maurice B. Hallett, David F. Horrobin, Robert E. Mansel, and Malcolm C. A. Puntis

University Department of Surgery. University of Wales College of Medicine. Heath Park, Cardiff CF4 4XN ¡W.G. J.. S. H., M. B H.. K. E. M.. M. C. A. PJ. and ScotiaPharmaceutical Ltd.. Cuildfard GUI IBA ¡D.F. HJ. Uniled Kingdom

ABSTRACT

E-cadherin is a cell to cell adhesion molecule which acts as a suppressorof metastasis. This study examined the effect of y-litmlmir acid (GLA), an-(>polyunsaturated fatty acid, on the expression of E-cadherin in human

cancer cells. Western blotting studies demonstrated that treatment of cellswith GLA for 24 h increased the expression of E-cadherin in lung, colon,

breast, melanoma, and liver cancer cells, but not in endothelial cells andfibroblasts. The results were confirmed by immunocytochemistry. Incontrast, two other n-6 fatty acids, linoleic acid and arachidonic acid,failed to induce these changes. The increased expression of E-cadherin was

correlated with reduced in vitro invasion and increased aggregation, indicating that the increased E-cadherin expression induced by GLA wasbiologically active. These data add GLA to the short list of E-cadherinup-regulatory factors. The up-regulation of E-cadherin expression in

human cancer cells may contribute to the anticancer properties of GLA.

INTRODUCTION

Certain n-6 PUFAs3 have been shown to have in vitro antiprolif-

eration and cytotoxic effects on breast, prostate, pancreatic cancer,and hepatoma cells (1-7). Inhibition of tumor cell growth by some

cytokines is also dependent on the presence of PUFAs (4). Theseanticancer effects are seen particularly with GLA, dihomo-GLA, and

EPA. We have recently reported another property of GLA that isdemonstrable on human colon cancer cells, namely, the inhibition ofmotility and in vitro invasive properties (8).

E-cadherin is a calcium-dependent transmembrane cell to cell adhesion molecule. E-cadherins bind to one another by homophilic

interactions and help to maintain tissue/organ structure. It appears thatE-cadherin is capable of inhibiting invasion and metastasis because(a) the metastatic potential of cell lines is inversely related to E-cadherin expression; (b) high E-cadherin levels in certain tumors arerelated to a less metastatic prognosis; (c) blocking E-cadherin function by antibody or other material, or complete removal of E-cadherin

by deletion of its gene, increases tumor cell motility, invasion, andmetastatic potential, while up-regulation of E-cadherin in breast cancer cells by insulin-like growth factor I (9) inhibits invasion andmotility; and (d) transfection of highly metastatic cells with E-cadherin produces cells with a low metastatic potential. Loss of E-

cadherin expression in some human cancers, or its mutation to aninactive form has been reported in gastric, gynecological, pancreatic,esophageal, lung, bladder, and breast cancers and other tumor types(10-15; for reviews, see Refs. 16-19). Recent work (20) also showsthat the expression of E-cadherin mRNA (in situ hybridization) incolorectal cancer is decreased, and that this is correlated with long-

term survival of the patients.These relationships between E-cadherin and tumor progression

Received 5/30/95; accepted 8/29/95.The costs of publication of this article were defrayed in part hy 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.

1This work was supported by the Welsh Scheme for the Development of Health and

Social Research.2 To whom requests for reprints should be addressed. Fax: 44 1222 761623.%The abbreviations used are: PUFA. polyunsaturated fatty acid; GLA, •¿�y-linolenic

acid; EPA. cicosapentaenoic acid; HGF/SF, hepatocyle growth factor/scatter factor; LA,linoleic acid; AA, arachidonic acid; ECL, enhanced chcmiluminescence.

prompted us to examine further the effects of GLA on E-cadherin

expression in a broader range of tumor cells. In this article, we reportthat GLA increased expression of E-cadherin in a range of human

cancer cells. This has been demonstrated by both Western blotting andimmunohistochemical studies. The raised levels of E-cadherin were

associated with increased cell aggregation and reduced in vitro invasion, suggesting that the expressed protein was functional.

MATERIALS AND METHODS

Cell Lines. Human hepatoma PLC/PRF/5, colon cancer HT29, HRT1K,and HT115. lung cancer SK LU 1, Cor L23, Cor L47, Cor LSI, Cor L677, andCor L88, melanoma G361, endothelial ECV304, and fibroblasl MRC5 cellswere obtained from the European Collection of Animal Cell Culture (Salisbury, UK). Human breast cancer cells MCF-7, MDA-MB-231, and ZR-751,

and the monocytic cell line U937 were obtained from American Type CultureCollection. The M19 human melanoma cells were a gift from D. Jones(Department of Surgery, University of Wales College of Medicine).

Materials. Matrigel (reconstituted basement membrane) was purchasedfrom Collaborative Research Product (Bedford, MA). A transwell plateequipped with a porous insert (pore size, 8.0 ¡j.m)was obtained from BectonDickinson (Oxford, UK) for the in vitro invasion study. A mouse antihumanE-cadherin (HECD-1) mAh was obtained from British Biotechnology (Oxford,UK), peroxidase-conjugated rabbit antimouse IgG for immunohistochemical

study was obatined from DAKO Ltd., and sheep antimouse IgG for Westernblotting was obtained from Amersham International (Little Chalfont, Buckinghamshire, UK). Recombinant human HGF/SF was a generous gift from Dr.T. Nakamura (Osaka, Japan). The cells were passaged three to five timesbefore assays were undertaken. Hoescht 33258, GLA, LA. AA, and EPA wereobtained from Sigma. Fatty acids were dissolved in ethanol and stored in liquidnitrogen. They were added to cells after dilution in complete medium containing albumin. All other materials were purchased from Sigma (UK) unlessotherwise stated.

Cell Invasion Assay. This was based on the methods of Albini et al. (21)and Parish et al. (22). Transwell chambers (Costar, Cambridge, MA) equippedwith 6.5-mm diameter polycarbonate membrane (pore size, 8 /urn) wereprecoated with a solubilized tissue basement membrane-Matrigel (50 /j.g/

membrane). After gel rehydration, 50,000 cells were added to each membranewith or without treatment. HGF/SF (20 ng/ml) was used in the lower chamberto induce invasion. In selected cultures, HECD-1 or EGTA was included. After

72 h of culture, the noninvasive cells were removed with a cotton swab, and thecells that had migrated through the membrane and stuck to the lower surfacewere fixed and stained with crystal violet. After extraction with 10% acetic-

acid, the absorbance was measured at 540 nm with a Titertek multiscanner.Invasion is shown here as invasion index: cell with treatment/control (8).

Tumor Cell Aggregation Assay. Cells treated with fatty acids or mediumonly (control) were first briefly washed with balanced salt solution and thenincubated in HCMF buffer (160 mM NaCl, 0.6 mM Na2HPO4, 0.1% w/vglucose, and 0.01 M HEPES, pH 7.4) containing 0.01% trypsin and 2 mMCaCU for 30 min at 37°C.Cells were washed and resuspended in HCMF with2 mM CaCl:, and the concentration was adjusted to 105cells/ml. They were

then added into a tube and put on a shaker. In selected cultures, HECD-1 or

EGTA was included. Proportions of this cell suspension were removed at 15,30, and 45 min and fixed in a 1.5 volume of 5% glutaraldehyde, a procedurewhich did not cause artificial aggregation or disaggregation as previouslyreported (23). The total numbers of particles (cells or cell aggregates) werecounted, and the aggregation index was calculated by (Nu —¿�N,)/N,„where Nn

is the number of particles before the experiment started, and N, is the numberat the chosen time (23, 24).

5043

on May 28, 2021. © 1995 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

REGULATION OF E-CADHERIN EXPRESSION BY OLA

Fig. 1. Expression of E-cadherin by Western

blotting. Cells were treated with medium only(control) or with LA, OLÀ,AA, or EPA for 24h. HRT18 colon cancer (A) and PLC/PRF/5liver (B) show increased E-cadherin after GLAtreatment. Tops, E-cadherin band stained withHECD-1 antibody and visualized with ECL;bottoms, relative density of the correspondingband for that gel.

A §o

s+

O Li!

I - 205

'- 116

- 97

-205

Control LA GLA AA EPAControl LA QLA AA

C

I

|

TIi

e

l.1

I

E§I

SI

I

e£

I

II

- 205

'- 116

- 97

- 205

- 116

- 97

40

0-

relative density40-

20

relativ* dmilty

4 hr« 12 hr> 24 hri 48 hri4 hr« 12 hr» 24 hr« 48 hrs

Fig. 2. Kinetic changes in GLA-induced E-cadherinexpression demonstrated by Western blotting. A,PLC/PRF/5 cells; and B, Cor L23 cells. Tops, E-cadherin band stained with HECD-1 antibody and

visualized with ECL; bottoms, relative density ofthe corresponding band for that gel. Cells weretreated with GLA for 4, 12, 24, and 48 h. GLAincreased expression slightly at 12 h, with maximum changes seen after 24 h.

Table 1 Effects of GLA on the expression of E-cadherin and aggregation of the cells

E-cadherin expression

Western blotting" Immunohistochemical

Aggregation index(/ = 30 min)

CelllinesHRT18HT29HT115MCF-7MDA

MB231ZR751G361M19Cor

L23CorL47Cor

L51CorL88CorL677SKLU1PLC/PRF/5ECV304MRC5U937Control15402505507II330li4000With

GLA5270370781)1C07.05.50034000ControlPositivePositiveNo

stainPositiveNo

stainPositivePositiveNo

stainPositiveNo

stainWeakstainVery

weakstainNostainNostainPositiveNo

stainNostainNo

stainWith

GLAIncreased

expressionIncreasedexpressionVery

weakstainIncreasedexpressionNo

stainIncreasedexpressionIncreased

expressionNostainIncreased

expressionNostainIncreasedexpressionIncreased

expressionNostainNo

stainIncreasedexpressionNostainNostainNo

stainControl0.25

+0.100.34

±0.120.20±0.110.25

±0.100±0.050.55±0.070.06±0.09ND"0.30

±0.030.06±0.060.04±0.070.08

±0.100.04±0.060.27

±0.130.42±0.110.32

±0.11ND°ND"With

GLA0.53

±0.160.65±0.030.26±0.180.37±0.080.01

±0.110.76±0.080.25±0.12ND0.48

±0.030.09±0.020.03

±0.150.64±0.110.05±0.160.50±0.160.62

±0.080.34±0.05NDND

' Western blotting results shown as relative density of E-cadherin band scanned by a densitometer. ND: not done.1Aggregation was determined in a calcium-containing buffer (HCMF, with 2 mM Ca~+). Results shown are aggregation indices in 30 min.

5044

on May 28, 2021. © 1995 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

REGULATION OF E-CADHERIN EXPRESSION BY OLA

Fig. 3. Immunohistochemical staining of E-cadherin. A and ß,HRT18 cells; C and D, PLC/PRF/5 cells; and £and F, ZR751 cells. Cells were treated with OLA (50 UM;B. D. and F)for 24 h and E-cadherin was stained with HECD-1 antibody. Cells cultured with medium were used as controls (A, C, and £).Cells with GLA showed increased staining of E-cadherinon the surface, particularly in the cell to cell junctions.

SDS-PAGE and Western Blotting. Cells were pelleted and lysed inHCMF buffer containing 0.5% SDS, 0.5% Triton X-100, 2 min CaCl2, 100/j.g/ml phenylmethylsulfonyl fluoride, 1 /ng/ml leupeptin, and 1 fig/ml aproti-nin for 30 min. They were then boiled at 100°Cfor 5 min before clarification

at 13,000 X g for 10 min. Protein concentrations were measured using

fluorescamine and quantified by using a multifluoroscanner (Denly, Sussex,UK). Equal amounts of protein from each cell sample (controls and treated)were added onto an 8% polyacrylamide gel. Following electrophoresis, proteins were blotted onto nitrocellulose sheets and blocked in 10% skimmed milkfor 60 min before probing with HECD-1 (0.5 ju,g/ml) and peroxidase-conju-

gated secondary antibody (1:2000). Protein bands were visualized with an ECLsystem (Amersham, UK). The band densities of E-cadherin on the photo

graphic film were analyzed with a densitometcr and are shown here as relativevalues.

Immunohistochemical Study. Cells were fixed with 4% formaldehydeand endogenous peroxidase blocked by methanol (0.3% H2O2) before blockingwith 2% BSA for 60 min. Cells were then incubated with HECD-1 (0.5 ng/ml)

for 60 min and extensively washed. For the immunohistochemical study,peroxidase-conjugated secondary antibody was added, and this was followedby color-developing agents (dimethylaminoazobenzene, 60 /Mg/ml). Slides

were mounted with Sterilyte.Cell Growth Monitoring. To eliminate the possibility that the effects seen

in the above assays might arise from the cytotoxicity of GLA, we choose lowGLA concentrations (<75 ¡J.M)in this study. Such concentrations have been

5045

on May 28, 2021. © 1995 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

REGULATION OF E-CADHERIN EXPRESSION BY OLA

Fig. 4. Effects of GLA on cell aggregation atvarious time points. Cells (left, HRT18; right,MCF-7) were treated with medium only (tops) or

GLA (bottoms} for 24 h before harvesting inHCMF-trypsin-calcium buffer. In each graph, dataare shown for medium alone, medium plusHECD-I (0.5 Mg/ml). and EGTA (1 mm). Theaggregation index was calculated as given in thetext. Both groups of cells showed increased aggregation in response to GLA treatment. These increases were minimized in the presence of HECD-1

and EGTA.

Without GLAWithout GLAMediumPlus HECD-1

Plus EGTA

With GLA (50JJM)MediumPlus HECD-1

Plus EGTA

With GLAMedium

* Plus HECD-1

•¿�Plus EGTA

10 20 30 40

Time (minutes)

10 20 30 40

Time (minutes)

previously shown by ourselves and by others to have no significant toxiceffects on these cells. In all of the studies, DNA contents from controls andfatty acid-treated cells were determined by using Hoescht 33258 (final con

centration, 1.0 fig/ml) as the fluorescent indicator and monitored by a Wellfluomultiscanner (Denly).

RESULTS

GLA Increased E-Cadherin Expression. Cells exposed to GLA(50 JXM)for 24 h had increased expression of E-cadherin. Fig. 1 showsE-cadherin levels by Western blotting of HRT18 colon cancer (Fig.

1/4) and PLC/PRF/5 liver (Fig. Iß)cancer cells in which expressionof E-cadherin was increased. A kinetic study revealed that the increased E-cadherin expression appeared as early as 12 h, but that

maximum changes were seen only after 24 h. The responses ofPLC/PRF/5 and Cor L23 cells are shown in Fig. 2. The rest of the dataare summarized in Table 1. Although most of the cells that expressdetectable E-cadherin, e.g., HRT18, HT29, MCF-7, ZR751, G361,

Cor L23, Cor LSI, Cor L88, and PLC/PRF/5, showed a positiveresponse to GLA, those with undetectable expression, including MDAMB 231 breast cancer, HT115 colon, SK LU 1, Cor L47 and Cor L677lung, and M19 melanoma cells, showed no response. Endothelial cellsECV304 and fibroblasts MRC5 also had no detectable E-cadherin

with this particular antibody (HECD1), and they did not show anyincrease with GLA.

Effects of Other Fatty Acids on E-Cadherin Expression. Otherfatty acids in the n-6 series (LA and AA, and a n-3 fatty acid, EPA)were tested for their effects on E-cadherin (Fig. 1). LA, AA, and EPAcaused either no significant increase in E-cadherin expression or verysmall increases. It is worth noting that in the MCF-7 cells, there wasa reduced expression of E-cadherin with LA.

Immunohistocheinical Detection of E-Cadherin. To determinethe location of the expression of E-cadherin, immunohistochemicaldetection was carried out, and this revealed that cell surface E-

cadherin staining was mainly in the cell to cell junctions, with PLC/

Fig. 5. Effects of GLA on tumor cell invasionthrough Matrigel. Left, MCF-7 cells; right. Cor L23

cells. HGF/SF promoted invasion was significantlyinhibited by GLA (50 JXM).The presence of bothHECD-1 (0.5 ng/ml) and EGTA (1 mM) in culturemedia also increased invasion into Matrigel. Theirexistence also reversed GLA-induced inhibition on

invasion. +, P < 0.05 v?rv».vcell with mediumalone; *, versus conditions with HECD-1 or EGTAalone; #, versus with HGF alone, using Mann-

Whitney U test.

2.5Invasion index

1.5

0.5

Med i u niPlus HECD-1

Plus EGTA

Control HOF GLA + HGF GLA

5046

2.5Invasion index

1.5

0.5

I 1 MediumÜÕPlus HECD-1C] Plus EGTA

i

-i

i

Control HGF GLA + HGF GLA

on May 28, 2021. © 1995 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

KHifLATION Oí I -( ADHERIN liXI'KI SSION BY GI.A

PRF/5, HT29, HRT18, MCF-7, Cor L23, ZR-751, and G361 cells

staining strongly; Cor LSI and Cor L88 exhibiting weak staining; andM19, MDA MB 231, HT115, SK LU 1, and Cor L47 and Cor L677cells showing no significant staining. This was consistent with Western blotting. With GLA treatment, there was a marked increased instaining in PLC/PRF/5, HT29, HRT18, MCF-7, Cor L23, ZR-751,

and G361 cells. Cor LSI, L88, and HT115 also showed increasedstaining, whereas others (M19, MDA MB 231, SK LU 1, and Cor L47and Cor L677) showed no response. Fig. 3 shows examples of thisstaining, with the results for the rest of the cells summarized inTable 1.

Effects of GLA on Aggregation. An aggregation assay was performed to test the function of E-cadherin on the cell surface. Cells

were cultured with GLA for 24 h, and their aggregation ability wastested in a calcium-containing buffer (HCMF plus 2 mM Ca2+). After

GLA treatment, cells with increased expression of E-cadherin showed

increased aggregation compared with their control counterparts. Fig. 4shows examples of these cells at different time points (15, 30, and 45min). Inclusion of anti-E-cadherin antibody (HECD-1) or calcium-chelating agent (EGTA) abolished GLA-induced aggregation ofHRT18 cells and significantly reduced GLA-induced aggregation ofMCF-7 cells (Fig. 3). Table 1 summarizes the results from all of thecell types at / = 30 min. There was a correlation between theexpression of E-cadherin and aggregation in different cell types.

Effects of GLA on Invasion. It has been proposed that the level ofE-cadherin on the tumor cell surface is inversely correlated with theirinvasive behavior. We tested whether a change in E-cadherin expres

sion correlated with invasion by using a Matrigel invasion assay. Inthis study, HGF/SF (20 ng/ml) was placed in the lower chamber of theculture system as a chemoattractant (8). A marked reduction in cellinvasion into Matrigel was observed with those cells displaying anincrease in E-cadherin (Fig. 5). Both HECD-1 and EGTA preventedthe inhibitory effect of GLA on invasion. Furthermore, both HECD-1and EGTA increased invasion in unstimulated and HGF/SF-treated

cells (Fig. 5).Cell growth in the presence of GLA (75 /U.M)was determined by

measuring cellular DNA quantities with a Hoescht 33258 assay. Thisexperiment demonstrated that GLA, at the highest concentration usedin this study, did not change the quantity of DNA in the cultures,indicating that GLA at such concentrations was not toxic to the cells.

DISCUSSION

In this study, we have shown that GLA enhanced the expression ofE-cadhcrin, an effect which was correlated with increased aggregation

and reduced in vitro invasion of these cells. Furthermore, increasedaggregation and decreased invasion were dependent on extracellularcalcium and prevented by anti-E-cadherin antibody, demonstratingthat GLA induced E-cadherin was functionally active. Other n-6

PUFAs showed little or no effects.Metastasis is a key factor in determining the prognosis of the

patients with cancer. To establish a distant metastasis, a tumor cellmust detach from the primary tumor, migrate through the basementmembrane and extracellular matrix, intravasate, and travel in thecirculation to the new site before reattachment, extravasation, or thedevelopment of a new focus or neovascularization occurs (25-28).Essential and early events if a tumor is to initiate this "metastaticcascade" (28, 29) are the loss of cell to cell adhesion and modified

cell-matrix interactions, which allow the tumor cells to become motileand enhance their invasive potential. E-cadherin is a cell to cell

adhesion molecule which is linked to the cytoskeleton by a group ofproteins now known as catenins (17, 30). E-cadherin is now widely

accepted as a tumor/metastasis suppressor. Reduced expression, de

letion, or mutation of E-cadherin has been shown in a variety of

human cancers, and levels of the molecule are inversely correlatedwith disease stage, tumor invasiveness, and long-term prognosis of thepatients (10-12, 16-20).

Such a relationship has raised interest in elucidating ways ofregulating E-cadherin expression in the hope that this may inhibit the

motile and invasive nature of tumor cells. Bracke et al. (31) havereported that tamoxifen restores the E-cadherin function in humanbreast cancer MCF-7/6 cells and suppresses their invasive phenotype.

Vandewalle et al. (32) have recently reported that a higher calciumenvironment may increase the expression of E-cadherin on MCF-7cells. Relaxin (33), 17-ßestradiol (34), and retinoic acids inducethe expression of E-cadherin (35). Here, our data add GLA to theshort list of E-cadherin up-regulatory factors. Of the variety of cell

types tested in this study, most of them responded to GLA. However, those with absent expression of E-cadherin showed no re

sponse. We currently have no evidence that the effects of GLA areeither on transcription or translation levels. Further work is nowrequired to establish the mechanism of regulation of this cell to celladhesion molecule.

In our study of the fatty acids tested, GLA showed the strongestup-regulation of E-cadherin expression, whereas LA and AA did not.Interestingly, LA slightly inhibited E-cadherin expression in MCF-7

cells. It has been previously reported in animal studies that LA maystimulate growth and metastasis of some breast cancer models (36,37), whereas oils containing both GLAs inhibit breast cancer growthin the nude mouse model (38). Our data may provide a partialexplanation for those phenomena.

The clinical importance of this result is worth exploring. GLA hasbeen used in pilot studies in patients with cancers, including breast,melanoma, and pancreatic cancer, as an adjuvant therapy. All of thesetumor types have been shown in vitro to be rather sensitive to GLA interms of cytoxicity. Patients with other tumors such as colon cancers,which are less sensitive to GLA, are not widely considered suitablecandidates for GLA cytotoxic treatment. Our study, along with previousreports on the GLA-induced inhibition of cell motility and invasion (8),

suggests that the cytotoxic effects of GLA and the effects on motility/invasion can be achieved independently. The latter effects can beachieved at much lower concentrations, making GLA of possible value asan adjuvant to surgery, chemotherapy, or radiotherapy.

This study reports that GLA, a n-6 series essential fatty acid,increases expression of E-cadherin in a range of human cancer cells.

The increased expression is associated with improved biological function as indicated by increased aggregation and reduced invasion ofthese cells. This may have importance in the development of futureclinical studies.

ACKNOWLEDGMENTS

We are grateful to Dr. T. Nakamura for providing recombinant human HGFand D. Jones for providing the M19 human melanoma cell line.

REFERENCES

1. Begin, M. E., Ells, G., Das, U. N., and Horrobin, D. F. Differential killing of humancarcinoma cells supplemented with n-3 and n-6 polyunsaturaled fatty acids. J. Nail.Cancer. Inst., 77: 1053-1062, 1986.

2. Tiwari, R. K., Mukhopadhyay, B., Telang, N. T., and Oshorne, M. P. Modulation ofgene expression by selected fatty acids in human breast cancer cells. Anticancer Res.,//: 1383-1388, 1991.

3. Rose, D. P., Connolly. J. M.. and Mcschtcr. C. L. Effect of dietary fat on human breastcancer growth and lung metastasis in nude mice. J. Nati. Cancer Inst., N3: 1491-1495,1991.

4. Newman. M. J. Inhibition of carcinoma and melanoma cell growth by type 1transforming growth factor beta is dependent on the presence of polyunsaturated tattyacids. Proc. Nati. Acad. Sci. USA, Ä7:5543-5547, 1990.

5. Takeda, S., Horrobin, D. F., Manku, M.. Sim, P. G.. Ells, G., and Simmons, V. Lipid

5047

on May 28, 2021. © 1995 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Rl.lil LATIOS 01 t: CADHHRIN liXI'RLSSION BY OLA

peroxidatiun in human breast cancer cells in response to y-linolenic acid and iron.Anticancer Res., 12: 329-333, 1992.

6. Horrobin, D. F. Essential fatty acids, lipid peroxidation, and cancer. In: D. F.Horrobin (ed.), Omega-6 essential fatty acids, pp. 351-378. New York: Wiley-Liss,

Inc., 1990.7. Hayashi, Y.. Fukushima, S.. Kishimoto, S., Kawaguchi, T.. Numata, M.. Isoda, Y.,

Mirano, J., and Nakano, M.. Anticancer effects of free polyunsaturated fatty acids inan oily lymphographic agent following intrahepatic arterial administration to a rabbitbearing VX-2 tumor. Cancer Res., 52: 400-405, 1992.

8. Jiang, W. G., Hiscox, S., Halle«,M. B., Scott, C, Horrobin, D. F., and Puntis, M. C. A.Inhibition of hepatocyte growth factor/scatter factor induced motility and invasion ofhuman cancer cells by gamma linolenic acid. Br. J. Cancer, 71: 744-752, 1995.

9. Bracke, M. E., Vyncke, B. M., Bruynee, E. A., Vermeulen, S. J., De Bruyne,G. K., Van Larcbeke, N. A., Vleminckx, K., Van Roy, F. M., and Marcel. M. M.Insulin-like growth factor I activates the invasion supressor function of E-cadhcrin in MCF-7 human mammary carcinoma cells in vitro. Br. J. Cancer, 68:282-289, 1993.

10. Becker, K-F., Atkinson, M. J., Reich, U., Becker, I., Nekarka, H., Siewert, J. R.. andHöfler,H. E-cadherin gene mutations provide clues to diffuse type gastric carcinomas. Cancer Res., 54: 3845-3852, 1994.

11. Risinger, J, I., Berchuck, A., Kohler, M. F., and Boyd, J. Mutations of the E-cadheringene in human gynecologic cancers. Nat. Genet., 7: 98-102, 1994.

12. Bongiorno, P. F., Al-Kasspooles, M., Lee, S. W., Rachwal, W. J., Moore, J. H.,Whyte, R. 1., Orringer. M. B., and Beer, D. G. E-cadherin expression in primary andmetastati thoracic neoplasms and in Barrett's oesophagus. Br. J. Cancer, 71: 166-

172, 1995.13. Oka, H., Shiozaki. H., Kobayashi, K., Inoue, M., Tahara, H., Kobayashi, T.,

Takatsuka. Y., Matsuyoshi, N., Hirano, S., and Takeichi, M.. and Mori, T. Expressionof E-cadherin cell adhesion molecules in human breast cancer tissues and its relationship to metastasis. Cancer Res., 53: 1696-1701, 1993.

14. Palacios, J.. Benito, N., Pizarro, A., Suarez, A., Espada, J., Cano, A., andGamallo, C. Anomalous expression of P-cadherin in breast carcinoma. Correlation with E-cadherin expression and pathological features. Am. J. Pathol., 146:605-612, 1995.

15. Schipper, J. H., Frixen, U. H., Behrens, J., Unger, A., Jahnke, D., and Birchmeier, W.E-cadherin expression in squamous cell carcinomas of head and neck. Cancer Res..51: 6328-6337, 1991.

16. Takeichi, M. Cadherins in cancer: implications for invasion and metastasis. Curr.Opin. Cell Biol., 5. 806-811, 1993.

17. Kemler. R. From cadherins to catenis. TIG (Trends Genet.), 9: 317-321, 1993.

18. Birchmeier. W., and Behrens, J. Cadherin expression in carcinomas: role in theformation of cell junctions and the prvention of invasiveness. Biochim. Biophys.Acta, 1198: 11-29, 1994.

19. Marcel. M.. Bracke, M., and Van Roy, F. Down regulation in vivo of the invasionsupressor molecule E-cadhcrin in experimental and clinical cancer. In: Hirohashi et al.(eds.). Molecular and Cellular Basis for Cell to Cell Interaction: Us Significance inCancer. Proceedings of the 24th International Symposium of the Princess TakamatsuCancer Research Fund. pp. 63-80. Princeton: Princeton Scientific Publishing Co., 1994.

20. Doruki, S., Handy, A. M., Poulsom. R.. Northover, J., and Hart, I. R. Level ofexpressio of E-cadherin in colorcctal cancer correlates with clinical outcome. Br. J.Cancer. 71: 614-616, 1995.

21. Albini, A.. Iwamoto. Y.. Kleinman, H. K., Martin, G. R., Aaronson, S. A., Kozlowski.J. M., and McEwan, R. N. A rapid in vitro assay for quantitating the invasive potentialof tumor cells. Cancer Res., 47: 3239-3245, 1987.

22. Parish, C. R., Jakobsen, K. B., and Coombe, D. R. A basement membrane permeability assay which correlates with the mctastatic potential of tumour cells. Int. J.Cancer. 52: 378-383, 1992.

23. Takeichi, M. Functional correlation between cell adhesive properties and some cellsurface proteins. J. Cell Biol., 75: 464-474, 1977.

24. Vleminckx. K. L.. Deman, J. J.. Bruyneel, E. A., Vandcnbossche, G. M. R.,Keirsebilck, A. A., Mareel, M.. and van Roy, F. M. Enlarged cell-associated proteo-glycans abolish E-cadherin functionality in invasive tumor cells. Cancer Res., 54:873-877, 1994.

25. Liotta, L. A. Biochemical mechanisms of tumor invasion and métastases.Clin.Physiol. Biochem., 5: 190-199, 1987.

26. Jiang. W. G., Puntis. M. C. A., and Hallet!, M. B. Molecular and cellular basis ofcancer invasion and metastasis: implication for treatment. Br. J. Surg., 81: 1576-

1590, 1994.27. Miller, F. R. Immune mechanisms in the sequential steps of metastasis. Crit. Rev.

Oncog., 4: 293-311, 1988.28. Hart, I. R., Goode, N. T., and Wilson, R. E. Molecular aspects of the metastatic

cascade. Biochim. Biophys. Acta, 989: 65-84, 1989.

29. Van Roy, F., and Mareel, M. Tumor invasion: effects of cell adhesion and motility.Trends Cell Biol., 2: 163-169, 1992.

30. Reifer. M. Cancer, catenins, and cuticle pattern: a complex connection. Science(Washington DC), 262: 1667-1668, 1993.

31. Bracke. M. E., Charlier, C., Bruyneel, E. A.. Labit. C., Mareel, M. M., and Castronovo,V. Tamoxifen restores the E-cadherin function in human breast cancer MCF-7/6 cells andsuppresses their invasive phenotype. Cancer Res.. 54: 4607-4609. 1994.

32. Vandewalle, B., Revillion, F., Doutrelant, P., Granier, A. M., and Lefebvre, J. Effectof calcium on E-cadherin expression in breast tumor cells. Int. J. Oncol., 4: 181-185,

1994.33. Sacchi, T. B., Bani. D., Brandi. M. L., Palchetti, A., and Bigazzi, M. Relaxin

influences growth, differentiation and cell-cell adhesion of human breast-cancer cellsin culture. Int. J. Cancer, 57: 129-134, 1994.

34. MacCalman, C. D., Farookhi, R., and Blaschuk, O. W. Estradiol regulates E-cadherin

mRNA levels in the surface epithelium of the mouse ovary. Clin. Exp. Metastasis. 12:276-282, 1994.

35. Anzano. M. A., Byers, S. W., Smith, J. M., Peer, C. W., Mullen, L. T., Brown, C. C..Roberts, A. B., and Sporn, M. B. Prevention of breast cancer in the rat with9-ri.v-retinoic acid as a single agent and in combination with tamoxifen. Cancer Res.,54.- 4614-4617, 1994.

36. Hubbard, N. E., and Erickson, K. L. Enhancement of metastasis from a trans-

plantable mouse mammary tumor by dietary linoleic acid. Cancer Res., 47:6171-6175, 1987.

37. Rose, D. P., Connolly, J. M., and Liu. X. H. Effects of linolicc acid on the growth andmetastasis of two human breast cancer cell lines in nude mice and the invasivecapacity of these cell lines in vitro. Cancer Res., 54: 6557-6562, 1994.

38. Pritchard, G. A., and Mansel, R. E. The effects of essential fatty acids on the growthof breast cancer and melanoma. In: D. F. Horrobin (ed.), Omega-6 Essential Fatty A,pp. 379-390. New York: Wiley-Liss, Inc., 1990.

5048

on May 28, 2021. © 1995 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

1995;55:5043-5048. Cancer Res   Wen G. Jiang, Steve Hiscox, Maurice B. Hallett, et al.  

-Linolenic Acid (GLA)γCells by Regulation of the Expression of E-Cadherin on Human Cancer

  Updated version

  http://cancerres.aacrjournals.org/content/55/21/5043

Access the most recent version of this article at:

   

   

   

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

  Subscriptions

Reprints and

  .pubs@aacr.orgDepartment at

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

  Permissions

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

.http://cancerres.aacrjournals.org/content/55/21/5043To request permission to re-use all or part of this article, use this link

on May 28, 2021. © 1995 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from