(CANCER RESEARCH57. 5272-5276. December 1. l997J

Advances in Brief

Galectin-3: A Novel Antiapoptotic Molecule with A Functional BH1 (NWGR)Domain of Bcl-2 Family'

Shiro Akahani, Pratima Nangia-Makker, Hidenori Inohara,2 Hyeong-Reh Choi Kim, and Avraham Raz3

Tumor Progression and Metastasis Program (S. A., P. N.M., H. 1., A. RI, Karmanos Cancer institute, and Departments of Pathology (H-R. C. K.] and Radiation Oncology.Wayne State University School of Medicine IA. RI, Detroit, Michigan 48201

Abstract

Galectin-3, a @-galactoside-bindingprotein, has been shown to beinvolved in tumor progression and metastasis. Here, we demonstrate thatexpression of galectin-3 in human breast carcinoma BT549 cells inhibitscis-dlamminedichloroplatinum (cisplatin)-induced poly(ADP-ribose) polymerase degradation and apoptosis, without altering Bcl-2, Bcl-XL, orBax expressions. Galectin-3 contains the NWGR amino acid sequencehighly conserved in the BH1 domain of the bcl-2 gene family, and asubstitution of glycine to alanine in this motif abrogated its antiapoptoticactivity. Our findings demonstrate that galectin-3 inhibits apoptosisthrough a cysteine protease pathway and highlight the functional significance of the NWGR motif in apoptosis resistance of a non-Bcl-2 protein.

Introduction

A number of antineoplastic agents have been developed for theeradication of malignancies, but successful chemotherapy still depends on the control of multidrug resistance, because failure may leadto mortality. For example, CDDP,4 a potent anticancer compound,which functions through interstrand DNA cross-links and the induction of apoptosis (1), has improved the outcome of many cancerpatients, but the mechanism(s) of CDDP resistance remains to bedefined. Expression of galectin-3, a Mr 31,000 carbohydrate-bindingprotein (2), correlates with neoplastic progression in head and neckcancer (3), thyroid cancer (4), gastric cancer (5), and colon cancer (6).In several experimental tumor systems, galectin-3 expression is related to the metastatic potential (7, 8). Recently, it has been suggestedthat galectin-3 may inhibit apoptosis through interactions with complementary carbohydrates (9) or with the antiapoptotic protein Bcl-2(10). Galectin-l and -9 have been reported to induce apoptosis (11—13), suggesting that some members of the galectin family are involvedin the regulation of apoptosis. Moreover, a domain in the COOH

terminus of galectin-3 was found to have a significant sequencesimilarity with the BH1 domain of the Bcl-2 family of proteinscontaining the NWGR motif (10), which is responsible for the antiapoptotic activity of Bcl-2. Further investigation was thus prompted toestablish a possible role of galectin-3 in the mechanism of druginduced apoptosis. To this end, we have used the recently identified

galectin-3-null cells, i.e., the human breast carcinoma BT549 cells. Inthis study, we demonstrate that expression of galectin-3 in human

Received 8/28/97; accepted 10/16/97.The costs of publication of this article were defrayed in part by the payment of page

charges. This article must therefore be hereby marked ads'ertisemen: in accordance with18 U.S.C. Section 1734 solely to indicate this fact.

I This work was supported in part by United States Army Grant DAMD 17-96.1-6181,

NIH Grants R29-CA64139 (to H-R. C. K.) and ROl-CA46120 (to A. R.), The Henry FordHealth Sciences Center and NIH Grant ROl-CA69480 (to A. R.), and the Paul ZuckermanSupport Foundation for Cancer Research (to A. R.).

2 Present address: Department of Otolaryngology, Osaka University Medical School,

2-2 Yamadaoka, Suita, Osaka 565, Japan.3 To whom requests for reprints should be addressed, at Tumor Progression and

Metastasis, Karmanos Cancer Institute, I 10 East Warren Avenue, Detroit, MI 48201.Phone:(313)833-0960;Fax:(313)831-7518.

4 The abbreviations used are: CDDP, cis-diamminedichloroplatinum (cisplatin); CMF

PBS, calcium-magnesium-free PBS; PARP, poly(ADP-ribose) polymerase.

breast carcinoma BT549 cells inhibits CDDP-induced PARP degradation and apoptosis, and that the NWGR motif of galectin-3 isrequired for its antiapoptotic activity, as determined by analysis of amutagenesis study showing that an amino acid substitution of glycineto alanine at position 182 abrogates its function.

Materials and Methods

Cells and Culture Conditions. The human breast cancer cell line BT549was obtained from Dr. E. W. Thompson, Vincent T. Lombardi Cancer Research Center, Georgetown University Medical Center (Washington, D. C.).BT549 transfectants with plasmid DNA accompanying inserts in either thesense (11811, 11913, and 11914) or the antisense (41421) orientation encodingthe human galectin-3 cDNA were established as reported previously (8). Thesecell lines were cultured in DMEM (Life Technologies, Inc., Grand Island, NY)supplemented with 10% heat-inactivated fetal bovine serum, 2 nmi glutamine,nonessential amino acids, and antibiotics (Life Technologies, Inc.). The cultures were maintained at 37°C in a humidified atmosphere of7% CO2 and 93%

air. Cultures were used for the experiments within three passages after recovcry from frozen stocks.

Drugs. CDDP (Sigma Chemical Co., St. Louis, MO) was dissolved in thefiltrated normal saline and stored at 4°Cuntil use.

Galectin-3 and Bcl-2 Expressions in Untreated BT549 Cells and Tramfectants. Cells (5X 106)were pelleted by centrifugation, washed with CMFPBS (pH 7.4), and subsequently lysed in 1 ml of reducing SDS-PAGE samplebuffer containing 2% SDS, 62.5 mr@iTris-HC1(pH 6.8), 10%glycerol, and 5%f3-mercaptoethanol. Then, cell lysates were agitated overnight at room temperature and stored at —70°C until use. For Western blot analyses, approxi

mately 20 @xgof total proteins were boiled for S mm, separated on reducing15% SDS-PAGE, and blotted onto a nitrocellulose membrane. Blots wereprobed with rat anti-galectin-3 (American Type Culture Collection, Rockville,MD) or mouse anti-Bcl-2 antibody (DAKO Corp., Carpinteria, CA) accordingto the enhanced chemiluminescence (Amersham, Buckinghamshire, England).Western blot procedure.

Cell Viability. Cell viability was assessed by a trypanblue dye exclusiontest. Cells were cultured with or without CDDP and were collected as de

scribed by Huddart et a!. (14). Briefly, at the indicated times, cells weretrypsinized and centrifuged at 2000 rpm for S mm. Viability was evaluated onthe hemacytometer (Baxter, McGaw Park, IL) following the addition of 0.4%trypan blue (Matheson, Coleman, and Bell, Norwood, OH) dissolved in 0.85%saline.

DNA Degradation. DNA fragmentationinducedby CDDPwas detectedasan increased fluorescence in the hypoploid DNA region of a flow cytometriccell cycle histogram according to the method described by Telford et a!. (15).Briefly, cells with or without CDDP treatment were fixed with 80% ethanol at4°Cfor 30 mm, washed with CMF-PBS (pH 7.4), and resuspended with 50;xgbml propidium iodide (Coulter Immunology, Hialeah, FL) diluted withCMF-PBS (pH 7.4) containing 0.1% Triton X-l00, 0.1 mistEDTA (pH 8.0),and 50 i.tg/ml RNase A. Then, samples were agitated at 4°Cfor not less than15 mm prior to analysis with flow cytometry.

Cell Morphology. The change of cell morphologywith CDDP treatmentwas assessed by fluorescence microscopy using the DNA-binding fluorochrome bis-(benzimide)-trihydrochloride (Hoechst 33258; Sigma) as describedpreviously by Oberhammer et a!. (16). Briefly, 1X 106 CDDP-treated cell

pellets were collected by centrifugation at 2000 rpm for S mm and fixed with3% paraformaldehyde at room temperature for 10 mm. Then, cells were

5272

Research. on September 17, 2020. © 1997 American Association for Cancercancerres.aacrjournals.org Downloaded from

GALECTIN-3 IN APOPTOSIS

washed with CMF-PBS (pH 7.4) and stained in the dark with 8 @tg/m1Hoechst33258 in CMF-PBS (pH 7.4) at room temperature for 15 mm. Samples were

loaded on the glass slide coated with 3-aminopropyl-trimethoxysilane(Sigma),and chromatin condensation and nuclear fragmentation in apoptotic cells were

observed with a fluorescence microscope equipped with a UV filter.Activation of CPP32/Yama Protease and PARP Degradation. Before

and after the treatment with 25 @MCDDP for 72 h, cell lysates were electrophoresed on reducing 8 or 15% SDS-PAGE, transferred to nitrocellulose, andprobed with mouse anti-PARP antibody5 or goat anti-CPP32/Yama antibody(Santa Cruz Biotechnology, Santa Cruz, CA), respectively.

Site-directed Mutagenesis of Galectin-3. Mutated galectin-3 (glycine182-alanine) was designed and obtained using the in vitro site-directed mutagenesis kit (Stratagene, La Jolla, CA) and cloned into the EcoRI site of thepCNC1O vector. For transfection, 2 @gof plasmid DNA were isolated andincubated with 10 @zlof lipofectamine (Life Technologies, Inc.), and thismixture was added onto the cells grown to 70% confluency. Clones wereselected at random using 800 @.tg/mlG4l8 (Life Technologies, Inc.) andmaintained in 10% complete DMEM containing 400 @g/mlG4l8. Oligonucleotideprimersfor PCR were sense (5'-TGGATAATAACTGGGCAAGGGAAGAAAG.3') and antisense (5'-ClTFC'fltCCUGCCCAGTTATTATCCA-3'), encoding 548—575nucleotides in human galectin-3 cDNA.

Results and Discussion

Human breast carcinoma BT549 cell clones transfected with plasmid DNA containing inserts in eitherthe sense (1 181 1, 11913, 11914)or the antisense control (41421) orientation encoding human galectin-3 were established from the galectin-3-null parental cells as reported previously (8). We have confirmed galectin-3 expression in theparental BT549 cells and its clones 41421 and 11914 by Western blotanalysis, and only sense-transfected cell clones expressed galectin-3,whereas the other two cell types remained galectin-3 null (Fig. 1A).

To test the effect of galectin-3 expression in CDDP-induced cytotoxicity, we treated all clones with CDDP and subsequently evaluatedcell viabilities by a trypan blue dye exclusion test. After an exposureto 25 @.LMCDDP, cell death was observed predominantly in parental

BT549 and 41421 cells when compared with 11811, 11913, and11914 cells (Fig. 1B). Parental BT549 and 41421 cells showed similarsensitivity to CDDP treatment; a 72-h exposure to CDDP exhibited adecrease in viability to less than 30% of the parental BT549 and41421 cells, whereas more than 60% of all galectin-3-expressingclones remained viable even after a 72-h exposure with the sameCDDP concentration.

To assess whether CDDP-treated cells underwent apoptosis, cellswere stained with the DNA-binding dye (propidium iodide) andsubjected to flow cytometric DNA analysis. The apoptotic subpopulation can be detected as the appearance of a discrete peak andincreased fluorescence in the@ cell cycle region in the DNAhistogram. It has been shown that the sub-01 peak was derived fromapoptotic cells in G@1/GJregion (15). The sub-G, population of thegalectin-3 null parental BT549 and 41421 cells occupied more than30% of the DNA histogram after a 72-h exposure to 25 @.LMof CDDP,whereas that of 11914 cells occupied not more than 20% (Fig. 2,D—F). To further investigate the apoptotic features of CDDP-treated

cells, including nuclear condensation and fragmentation, cells werevisualized under fluorescence microscopy after paraformaldehyde fixation and staining with the DNA-binding fluorochrome bis-(benzimide)-trihydrochloride (Hoechst 33258). As shown in Fig. 2, G—I,CDDP-induced cell death in all three types of cells involved apoptosis. More than 40% of CDDP-treated parental BT549 cells exhibited

nuclear morphologies that are indicative of apoptosis, i.e., DNAfragmentation and condensed chromatin (Fig. 2G). Similar patterns ofchromatin condensation and nuclear fragmentation were seen in

A-JI-C.)a)

.-@@‘-

(1) It) ,@@01— ,-@-

.@ —Galectin-3—BcI•2

Fig. I . A. galectin-3 and Bcl-2 expressions in untreated parental BT549 and transfec(ants. Western blot analysis: samples of protein from cell lysates were subjected toreducing 15% SDS-PAGE, transferred to nitrocellulose, and probed with rat anti-galectin-3 antibody or mouse anti-Bcl-2 antibody. Human breast carcinoma MCF-7 andlymphoma WSU-CLL cells were used as positive controls of galectin-3 and Bcl-2,respectively. These results are representative of three separate experiments. Galectin-3and Bcl-2 expressions in untreated cells were also confirmed by Northern blot analyses(data not shown). B, cell viability of parental BT549 cells (U); antisense transfectant41421 cells (A); and sense transfectant I 181 1 (Li), 11913 (V), and I 1914 (V) cells with25 @zMCDDP treatment in a time-dependent manner. Cells were cultured with or withoutCDDP and harvested at the indicated times, and cell viability was determined using atrypan blue dye exclusion test. Data are means (bars, SD) of triplicate determinations fromthree separate experiments.

5 Biomol Research Meeting, Plymouth Meeting, PA.

5273

B 100.

@o

@ 50.

.@ 25.

>

CDDP-treated 41421 cells (Fig. 2H). However, I 1914 cells respondeddifferently to CDDP with more visible nucleoli, mostly without altered chromatin structure (Fig. 20. When considered together with theresult of flow cytometry, apoptosis was observed in galectin-3-nullparental BT-549 and 4142 1 cells, yet it was largely absent in galectin3-expressing 11914 cells.

We further evaluated whether the expression of galectin-3 inBT549 cells resulted in altered expression of the Bcl-2 family ofproteins, because previous studies had indicated that an elevated Bcl-2

level influenced the cell survival effect ( 17, 18). Galectin-3-null and-positive cell clones showed no difference in the level of Bcl-2 protein

expressions (Fig. lA); no significant difference in the level of theBcl-XL and Bax expressions among these three cell clones wasobserved (data not shown), and their expressions were not altered bya 72-h exposure to 25 @.tMCDDP (data not shown). We then investi

Time (hrs)

Research. on September 17, 2020. © 1997 American Association for Cancercancerres.aacrjournals.org Downloaded from

@,G:,.

@.. .

O.w@

•@H

a.i.;•

GALECTIN-3 IN APOVFOSIS

, A 8ubO1@O.31%! D $ub0131.31%

;@(f@@ .@..

SubGlaO.54% 7 E $ubG1@32.T7S

@ 1X@L

@. ‘J.

*

Fig. 2. Flow cytometric cell cycle histograms ofethanol-fixed parental BT549 and transfectantsstained with 50 g.@g/mlpropidium iodide before [parental BT549 (A), 41421 (B), and 11914 (C) cells(and after [parental BT549 (D), 41421 (E), and 11914(F) cells] treatment with 25 @MCDDP for 72 h ( I3).*. cells in the sub-G1 region that have undergone

chromatin degradation associated with apoptosis.G—I.morphological changes associated with apoptosis in parental BT549 cells (G) and transfectants[41421(H)and I1914(1)]treatedwith25 @zMCDDPfor 72 h. Cells were paraformaldehyde fixed, staineddark with Hoechst 33258, and visualized throughfluorescence microscopy with a UV filter. Magnification, X400.

.0

E

z

C.)

, C 8ub013.12%@ F $ubOllO.BS%

Channsl Number

gated the possibility that galectin-3 might regulate apoptosis throughinteractions with members of the Bcl-2 family of proteins, but acomplex formation between galectin-3 and Bcl-2 or Bax in 11914cells could not be established by coimmunprecipitation assay (data not

shown).PARP is a substrate for mammalian cell death proteases and is

degraded by activated cysteine proteases of the interleukin- 1 3 converting enzyme/ced-3 family, including CPP32IYama (caspase-3;Refs. 19 and 20). Degradation of PARP is a characteristic feature ofthe early stage of apoptosis and leads to inhibition of DNA repair;activation of an endonuclease; and, eventually, cell death (21 ). To

further study the involvement of galectin-3 in the apoptosis pathway,we examined expressions of intact and cleaved PARP before and aftera 72-h exposure to CDDP, showing that CDDP induced the cleavageof intact PARP (Mr 116,000) into the inactive fragment (Mr 85,000) ingalectin-3-null parental BT549 and 41421 cells but not in galectin-3-expressing 1 1914 cells (Fig. 3). Interestingly, the levels of precursor

CPP32/Yama in parental BT549 and 41421 cells attenuated after a72-h exposure to CDDP, when compared to expression in 1 1914 cells.

The human proto-oncogene bc!-2 and the Caenorhabditis e!egansgene ced-9, both of which protect cells from apoptosis (22—24),and asubstitution of glycine in the NWGR motif of Bcl-2 BH1 domain andits highly conserved motif of ced-9 lead to a loss of the antiapoptoticactivity of Bcl-2 and its binding activity to Bax (25—26). The BH1

domain in the Bcl-2 family of proteins and its partial homology withthe COOH terminus of galectin-3 suggest that the highly conservedNWGR motif might be also critical to the apoptotic function ofgalectin-3 (Fig. 4A). To determine the significance of the NWGRmotif in galectin-3, mutant galectin-3 (glycine 182 to alanine) wasgenerated and cloned into the expression vector, and the parentalBT-549 cells were transfected with it. Three stable transfectantsshowing differences in expression levels of the mutant galectin-3[clone 5 expressed mutant galectin-3 at the highest level (the most

among the three), and its expression was as strong as that of thewild-type 11914 cells (Fig. 4B)j were established in culture. Simultaneously, we examined in them the expression of Bcl-2, but noalteration in Bcl-2 expression was detected following the introductionof mutant galectin-3 (data not shown). To evaluate its function inCDDP-induced apoptosis, we have treated the clones with 25 p.MCDDP for up to 72 h, and a time-dependent cell viability was assessedand compared to that of the parental and the galectin-3 wild-typetransfected BT-549 cells (Fig. 4C). The three clones containing themutated galectin-3 and the parental BT-549 cells showed a similarsensitivity to CDDP, with clone 5 being the most susceptible to CDDPamong the clones. Next, we stained CDDP-treated cell clones withpropidium iodide for the analysis of sub-G1 content in a DNA histo

c@@_@

,@ c4@‘r-'@C'billn@ c@In@ a)I- @- @- I-' @-

m@ 1-me'-

:1,n8trt

@ -@ —P32

CDDP(—) CDDP(+)

Fig. 3. Cleavage of the DNA repair enzyme PARP and precursor CPP32/Yamaexpressions in parental BT549 cells and transfectants exposed to CDDP. Before and afterthe treatment with 25 p.MCDDP for 72 h, proteins from cell lysates were electrophoresedon reducing 8 or 15% SDS-PAGE, transferred to nitrocellulose, and probed with mouseanti-PARP antibody5 or goat anti-CPP32IYama antibody (Santa Cruz Biotech, SantaCruz, CA), respectively. The amount of protein used was the same as that described in theFig. lA legend.

5274

C. SI @e: l@t Z@C .i C

:L .1@ . . @aL 1t@ -@ @.a@ .h@@@ I@

Research. on September 17, 2020. © 1997 American Association for Cancercancerres.aacrjournals.org Downloaded from

GALECflN.3 IN APOPTOSIS

136 ELFRDGV-NWGRIVAFFEFGG

98 DMFSDGNFNWGRWALFYFAS

150 ELFRDGV-NWGRIVAFFSFGG174 NTKLD-N-NWGRE- ---ERQS

(“I N N@ .

(@) (@)S I S

l5(@ m

000

@.@ 0@,@@@

r- @-@

155

118169

188

ABcl-2Bax

Bcl-XL

Galectin- 3

Galectin—3/G182A

B

I-@

A

I “—GaI•ctln-3

Fig. 4. Stable clones expressing mutant galectin-3and comparison with parental BT549 cells and trans.fectantsof galectin-3cDNA.A.schematicrepresentation of the BH1 domain and its similarity with thecarbohydrate-binding domain of galectin-3. Bol4the NWGR motif, which is responsible for the antiapoptotic function of Bcl-2. An amino acid substitution of glycine 182 to alanine is shown in thebottom row (arrow). B, Western blot analysis:galectin-3 with an amino acid substitution of 182glycine for alanine was established by following theprotocol of an in vitro site-directed mutagenesis kit(Stratagene,La Jolla,CA)and transfectedintoparental BT549 cells. Mutant galectin-3 in stableclones was detected using rat anti-galectin-3, asshown in Fig. 1A. C. cell viability of clones expressing mutant galectin-3 ( •, clone 1; •,clone 3; and0, clone5) with25 @LMCDDPtreatmentin a timedependent manner and comparison with parentalBT549 (U) and wild-type galectin-3 transfectant11914 cells. Viability was determined as describedin the Fig. 2 legend.

C I 00

gram and with Hoechst 33258 for the assessment of chromatin condensation and fragmentation. All galectin-3 mutant clones displayedsimilar features of apoptosis, as observed in parental BT-549 and41421 cells (data not shown). In addition, all of the galectin-3 mutantclones underwent activation of CPP32/Yama and proteolytic cleavageof PARP after 72 h of CDDP exposure (Fig. 5). Taken together, thesefindings indicate that the expression of the mutant galectin-3 resultedin failure to inhibit CDDP-induced apoptosis in BT-549 cells and thatthe NWGR motif of galectin-3 is critical for its antiapoptotic functionsimilar to that of Bcl-2 protein. It should be noted that the mutantgalectin-3 maintained activity similar to that of the wild-type asdetermined by homotypic aggregation process and similar cellulardistribution, as visualized by indirect immunofluorescence.

The results also indicate that galectin-3 confers apoptosis resistanceto CDDP in a cysteine protease-dependent manner, similarly to Bcl-2

(20). It remains possible that galectin-3 can replace or mimic Bcl-2,leading to the inhibition of CDDP-induced apoptosis involvingCPP32/Yama protease activation and PARP degradation.

Mutagenesis studies revealed that like Bcl-2 or ced-9, galectin-3regulates its antiapoptotic activity through the glycine residue of theNWGR motif. It has been demonstrated previously that syntheticglycoamines or glycosylated molecules, such as Mac-2 binding protein, could regulate homotypic aggregation through their interactionswith galectin-3 (27, 28), and it was suggested that the disruption ofgalectin-3-mediated cell-cell interactions on the cell surface resultedin induction of apoptosis in murine melanoma B16-Fl0 (9). Theresults from the COOH terminus mutation raises the possibility thatsubstitution of the glycine residue may change the affinity to carbohydrates on the cell surface, block galectin-3 mediated cell-cell interaction, and cause the loss of antiapoptotic function of galectin-3.

5275

Time(his)

Research. on September 17, 2020. © 1997 American Association for Cancercancerres.aacrjournals.org Downloaded from

GALECTIN.3 IN APOPTOSIS

P. W. J., Rini, J. M., and Wang. J. L. Galectins: a family of animal @3-galactosidebinding lectins. Cell, 76: 597—598,1994.

3. Gillenwater, A., Xu, XC., EI-Naggar, A. K., Clayman, G. L., and Lotan, R. Expression of galectins in head and neck squamous cell carcinoma. Head Neck, /8:422—432, 1996.

4. Xu, X. C., El-Naggar, A. K.. and Lotan, R. Differential expression of galectin-l andgalectin-3 in thyroid tumors. Am. J. Pathol., 147: 815—822, 1995.

5. Lotan, R., Ito, H., Yasui, W., Yokozaki, H., Lotan, D., and Tahara, E. Expression of3 l-KDa lactose-binding lectin in normal human gastric mucosa and primary andmetastatic gastric carcinomas. Int. J. Cancer, 56: 474—480, 1994.

6. Schoeppner. H. L., Raz, A., Ho, S. B., and Bresalier, R. S. Expression of anendogenous galactose-binding lectin correlates with neoplastic progression in thecolon. Cancer (Phila.), 75: 2818—2826,1995.

7. Raz, A., Thu. D., Hogan, V., Shah, N., Raz, T., Karkash, R., Pazerini, G., and Carmi,P. Evidence for the role of 34-KDa galactoside-binding lectin in transformation andmetastasis. Int. J. Cancer, 46: 871—877, 1990.

8. Nangia-Makker. P.. Thompson. E., Hogan, C., Ochieng, J., and Raz, A. Induction oftumorigenicity by galectin-3 in a non-tumorigenic human breast carcinoma cell line.Int. J. Oncol., 7: 1079—1087,1995.

9. Glinsky, G. V., and Glinsky, V. V. Apoptosis and metastasis: a superior resistance ofmetastatic cancer cells to programmed cell death. Cancer Lett., 101: 43—51,1996.

10. Yang, RY., Hsu, D. K., and Liu, F-T. Expression of galectin-3 modulates T-cellgrowth and apoptosis. Proc. NatI. Acad. Sci. USA., 93: 6737—6742, 1996.

I I . Perillo, N. L., Pace, K. E., Seilhamer, J. J., and Baum, L. G. Apoptosis of T cellsmediated by galectin-l. Nature (Lond.), 378: 736—739, 1995.

12. Perillo, N. L., Uittenbogaart, C. H., Nguyen, J. T., and Baum, L. G. Galectin-l, anendogenous lectin produced by thymic epithelial cells. induces apoptosis of humanthymocytes. J. Exp. Med., 185: 1851—1858,1997.

13. Wada, J., Ota, K., Kumar, A., Wallner, E. I., and Kanwar, Y. S. Developmentalregulation, and apoptotic potential of galectin-9. a fJ-galactoside binding lectin.J. Clin. Invest., 99: 2452—2461,1997.

14. Huddart, R. A., Titley, J.. Robertson, G. T., Horwich, A., and Cooper, C. S.Programmed cell death in response to chemotherapeutic agents in human germ celltumor lines. Eur. J. Cancer, 31A: 739—746,1995.

15. Telford, W. G., King. L. E., and Fraker, P. J. Comparative evaluation of several DNAbinding dyes in the detection of apoptosis-associated chromatin degradation by flowcytometry. Cytometry, 13: 137—143,1992.

16. Oberhammer, F., Wilson, J. W., Dive, C., Morris, I. D., Hickman, J. A., Walkeling.A. E., Walker, P. R., and Sikorska, M. Apoptotic death in epithelial cells: cleavage ofDNA to 300 and/or kb fragments prior to or in the absence of internucleosomalfragmentation. EMBO J., 12: 3679—3684, 1993.

17. Korsmeyer, S. J. Regulators of cell death. Trends Genet., 11: 101—105,1995.18. Zhang, Z., Vuori, K., Reed, J. C., and Ruoslabti, E. The a5@l integrin supports

survival of cells on fibronectin and up-regulates Bcl-2 expression. Proc. NatI. Acad.Sd. USA, 92: 6161—6165, 1995.

19. Tewari, M., Quan, L. T., O'Rourke, K., Desnoyers, S., Zeng, S., Beidler, D. R.,Poirier, G. G., Salveson, G. S., and Dixit, V. M. Yama/CPP32@3, a mammalianhomolog of CED-3. is a CrmA- inhibitable protease that cleaves the death substratepoly(ADP-ribose) polymerase. Cell, 81: 801—809, 1995.

20. Ibrado, A. M., Huang. Y.. Fang, G., Liu, L., and Bhalla, K. Overexpression of Bcl-2or Bcl-xL inhibits Ara-C-induced CPP32/Yama protease activity and apoptosis ofhuman acute myelogenous leukemia HL-60 cells. Cancer Res., 56: 4743—4748, 1996.

21. Kaufman, S. H., Desnoyers, S., Ouaviano, Y., Davidson, N. E., and Poirier, G. G.Specific proteolytic cleavage of poly (ADP-ribose) polymerase: an early marker ofchemotherapy-induced apoptosis. Cancer Res.. 53: 3976—3985, 1993.

22. Reed, J. C. Bcl-2 and the regulation of programmed cell death. J. Cell. Biol., 124:1—6,1994.

23. Fraser, A., and Evan, G. A license to kill. Cell, 85: 781—784,1996.24. Hengartner, M. 0., and Horvitz, H. R. C. elegans cell survival gene ced-9 encodes a

functional homolog of mammalian proto-oncogene bcl-2. Cell. 76: 665—675, 1994.25. Hengartner, M. 0., Ellis, R. E., and Horvitz, H. R. Caenorhabditis elegans gene ced-9

protects cells from programmed cell death. Nature (Lond.), 369: 318—320, 1994.26. Yin, X. M., Oltvali, Z. N., and Korsmeyer, S. J. BHI and BH2 domains of Bcl-2 are

required for inhibition of apoptosis and heterodimerization with Bax. Nature (Lond.),369: 321—328,1994.

27. Glinsky, G. V., Mossine, V. V., Price, J. E., Bielenberg, D., and Glinsky, V. V.Inhibition of colony formation in agarose of metastatic human breast carcinoma andmelanoma cells by synthetic glycoamine analogs. Clin. Exp. Metastasis, 14: 253-267,1996.

28. Inohara, H., Akahani, S., Koths, K., and Raz, A. Interactions between galectin-3 andMac-2-binding protein mediate cell-cell adhesion. Cancer Res., 56: 4530—4534,1996.

29. Yang. J., Liu, X., Bhalla, K., Kim, C. N., Ibrado, A. M., Cai, J., Peng, T-I., Jones,D. P., and Wang, X. Prevention of apoptosis by Bcl-2: release of cytochrome c frommitochondria blocked. Science (Washington DC), 275: 1129—1132, 1997.

30. Kluck, R. M., Bossy-Wetzel, E., Green, D. R., and Newmeyer, D. D. The release ofcytochrome from mitochondria: a primary site for Bcl-2 regulation of apoptosis.Science (Washington DC), 275: 1132—1136, 1997.

1@ C@)t() 1- C@)Lfl

;h@<<< <<<N (‘4 (“Se4 C'l C',S

000000C1) 1@) (#) e'@c•) c@)

S S S S S S

C@ (@ (@ C@(5

000000@ @-@-.—.@-C) C) C) O)a) 0)

s@) It) I@, (flit) U@

@@@I4n@ct

@ —P32

CDDP(—) CDDP(+)

Fig. 5. Cleavage of PARP and precursor CPP32/Yama expressions in BT549 transfectants of mutant galectin-3 cDNA. Before and after a 72-h exposure to 25 @MCDDP.cleavage of PARP and CPP32/Yama in transfectants was examined by Western blot asdescribed in the Fig. 3 legend.

However, our preliminary study showed that in parental BT549 cells,CDDP activity was not inhibited by exogenously added recombinantwild-type galectin-3 with a concentration of 10 p.g/ml. This findingindicates that the role of galectin-3 in apoptosis is more likely toinvolve a mechanism in the intracellular compartment rather than onthe cell surface. It is now well accepted that Bcl-2 localized at themitochondrial membrane regulates apoptosis by blocking cytochromeC translocation (29, 30), whereas galectin-3 can be found at the

nuclease, the cytoplasm, the cell surface, or in a secreted form, but notat the Golgi apparatus or the mitochondria. This suggests that the twoproteins regulate their antiapoptotic activities at different cellular

compartments.It should be emphasized that galectin-3 does not belong to the Bc!-2

gene family, because they share less than a 30% homology (10).However, whether they have merged from a common ancestral genecontaining the NWGR motif is unknown; it should also be noted that

among the galectin family of proteins, only galectin-3 contains theBcl-2 NWGR motif. In summary, we show here that galectin-3 is anovel antiapoptotic molecule, which like Bcl-2 acts through cysteineprotease pathways; this finding may lead to the development of newreagents to induce apoptosis of tumor cells and contribute to theunderstanding of the role of galectin-3 in metastasis.

Acknowledgments

We thank E. W. Thompson for providing us with the BT549 cell line; F. G.Kern for the pCNCIO vector; E. Van Buren, R. Johnson, and L. Tait for theflow cytometry analyses; V. Powell for typing and editing; and Drs. W. Weiand R. Bright for their critical evaluation of the manuscript.

References

1. Cress, A. E., and Dalton, W. E. Multiple drug resistance and intermediate filaments.Cancer Metastasis Rev., 15: 499—506, 1996.

2. Barondes, S. H., Castronovo, V., Cooper, D. N. W., Cummings, R. D., Drickamer, K.,Fenzi, T., Gin. M. A., Hirabaya.shi, J., Hughes. C., Kasai, K., Leffler, H., Liu, F. T.,Lotan, R., Mercurio, A. M., Monsigny, M., Pillai, S.. Poirer, F., Raz, A., Rigby,

5276

Research. on September 17, 2020. © 1997 American Association for Cancercancerres.aacrjournals.org Downloaded from

1997;57:5272-5276. Cancer Res   Shiro Akahani, Pratima Nangia-Makker, Hidenori Inohara, et al.   BH1 (NWGR) Domain of Bcl-2 FamilyGalectin-3: A Novel Antiapoptotic Molecule with A Functional

  Updated version

  http://cancerres.aacrjournals.org/content/57/23/5272

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/57/23/5272To request permission to re-use all or part of this article, use this link

Research. on September 17, 2020. © 1997 American Association for Cancercancerres.aacrjournals.org Downloaded from