T-Cadherin Negatively Regulates the Proliferation of CutaneousSquamous Carcinoma Cells

Yohei Mukoyama,� Shuxia Zhou,w Yoshiki Miyachi,� and Norihisa Matsuyoshi��Department of Dermatology, Graduate School of Medicine, Kyoto University, Kyoto, Japan; wLife Sciences, Lawrence Berkeley National Laboratory, Berkeley,California, USA

T-cadherin is a unique member of the cadherin superfamily that lacks the transmembrane and cytoplasmic do-

mains, and is instead linked to the cell membrane via a glycosyl-phosphatidylinositol anchor. We previously

reported that T-cadherin was specifically expressed on the basal keratinocytes of the epidermis, and the expres-

sion of T-cadherin was significantly reduced in invasive cutaneous squamous cell carcinoma (SCC) and in the

lesional skin of psoriasis vulgaris. In this study, to obtain an insight into the role of T-cadherin in keratinocytes, we

used transfection methods and examined the effect of overexpression or knockdown of T-cadherin in immortalized

keratinocyte cell lines derived from SCC. T-cadherin overexpressed cells showed clearly reduced cell proliferation,

but the influence of cell–cell adhesiveness and cell mobility was not detected. Using a tetracycline-regulated

expression system, we also confirmed that the suppression of cell proliferation was dependent on the expression

level of T-cadherin. Cell cycle analysis demonstrated that over expression of T-cadherin induced a delay in the G2/M

phase. Our findings suggest that T-cadherin acts as an endogenous negative regulator of keratinocyte proliferation

and its inactivation is the cause for keratinocyte hyperproliferation in SCC or in psoriasis vulgaris.

Key words: aggregation/keratinocyte/migration/transfectionJ Invest Dermatol 124:833 –838, 2005

Cadherins are a family of cell adhesion molecules that ex-hibit calcium-dependent, homophilic binding. Each memberof the cadherins shows specific distribution in various kindsof organs and is known to be involved not only in cell ad-hesion but also in many biological processes, such as cellrecognition, cell signaling, cell communication, morpho-genesis, as well as pathological conditions such as carci-noma (Angst et al, 2001). It is easy to imagine that a reducedexpression of cadherin triggers the release of carcinomacells from the tumor mass. In fact, strong E-cadherin ex-pression is recognized in well-differentiated carcinomas,which maintain their cell–cell adhesiveness and are less in-vasive, but E-cadherin expression is reduced in undifferen-tiated carcinomas, which have lost their cell–cell adhesionand show a strong invasive tendency (Hirohashi and Kanai,2003).

T-cadherin (CDH 13, H-cadherin) is a unique member ofthe cadherin superfamily that lacks the transmembrane andcytoplasmic domains, and is instead linked to the plasmamembrane via a glycosyl-phosphatidylinositol (GPI) anchor(Ranscht and Dours-Zimmermann, 1991). The fact that thecytoplasmic domain of classical cadherin is crucial for itscell–cell adhesion suggests the existence of another func-tion of T-cadherin (Nagafuchi and Takeichi, 1988; Stappertand Kemler, 1994; Yap et al, 1998). Furthermore, T-cadherindoes not contain the highly conserved His–Ala–Val motif inthe extracellular domain (Ranscht and Dours-Zimmermann,

1991). Since this motif is crucial for the homophilic interac-tions of classical cadherin (Blaschuk et al, 1990; Nose et al,1990), the lack of this motif in T-cadherin also supports thishypothesis. Like other GPI-anchored proteins, T-cadherin isenriched in minor detergent-insoluble plasma membranedomains, called lipid rafts, and co-localizes with signalingmolecules such as Src family kinases (Doyle et al, 1998;Philippova et al, 1998). These features of T-cadherin indi-cate that T-cadherin may be involved in signal transductionrather than cell–cell adhesion.

We previously reported that T-cadherin is expressed onkeratinocytes and is specifically localized in the basal layerof the epidermis, where keratinocyte proliferation takesplace (Zhou et al, 2002). Furthermore, we also proved that T-cadherin expression is downregulated in invasive cutane-ous squamous cell carcinoma (SCC) and in the lesional skinof psoriasis vulgaris (Takeuchi et al, 2002; Zhou et al, 2003).Another study showed that T-cadherin suppressed the pro-liferation of neuroblastoma cells and glioma cells (Takeuchiet al, 2000; Huang et al, 2003). The function of T-cadherin inthe skin, however, still remains unclear; these results permitus to speculate that the role of T-cadherin in the skin is thenegative regulation of keratinocyte proliferation, and inac-tivation of T-cadherin is, conversely, the cause for keratin-ocyte hyperproliferation.

In this study, to obtain an insight into the role of T-cad-herin in keratinocytes, we used transfection methods andexamined the effect of overexpression or knockdown of T-cadherin in immortalized keratinocyte cell lines derived fromSCC. In T-cadherin overexpressing cells, we clearly ob-Abbreviations: SCC, squamous cell carcinoma; Tet, tetracycline

Copyright r 2005 by The Society for Investigative Dermatology, Inc.

833

served reduced cell proliferation, whereas the influence ofcell–cell adhesiveness and cell mobility was not detected.Using a tetracycline (Tet)-regulated expression system, wealso confirmed that the suppression of cell proliferation wasdependent on the expression level of T-cadherin. Cell cycleanalysis demonstrated that overexpression of T-cadherininduced a delay in the G2/M phase.

Results

Establishment of T-cadherin overexpressed and down-regulated cell lines We determined the expression level ofT-cadherin on native HSC-1 cells and the transfectants, theHSC-1 RNAI cell line (T-cadherin expression is downregu-lated by the RNAi method), the HSC-1 TCAD cell line (over-expression of T-cadherin, which is driven by thecytomegalovirus promoter), and the HSC-1 TCAD RNAIcell line (overexpressed T-cadherin is downregulated by theRNAi method), with western blotting (Fig 1a). The resultsdemonstrated that a weak band was detected from nativeHSC-1 cells and no signal was observed from HSC-1 RNAI(Fig 1a). HSC-1 TCAD showed two thick bands, which wereconsidered to be mature protein (lower band) and precursor(upper band) of T-cadherin. These two bands were thinner inHSC-1 TCAD RNAI, which possessed the characteristic thatthe overexpressed T-cadherin was downregulated by theRNAi method. The expression levels of E-and P-cadherinwere not changed by overexpression or knockdown of T-cadherin (Fig 1a).

Similar results were obtained from immunofluorescencestaining. HSC-1 cells showed a weak signal, no signal wasdetected from HSC-1 RNAI, a stronger signal than the na-tive HSC-1 cells was detected in HSC-1 TCAD, and aweaker signal than HSC-1 TCAD was observed in HSC-1TCAD RNAI (Fig 1b). In HSC-1 cells, it was also demon-strated that T-cadherin was diffusely distributed on bothcell–cell boundaries and the cell surface (Fig 1b1), whereasE-and P-cadherin were concentrated mainly at the cell–cellboundaries (Fig 1b5,6). These observations were similar totheir transfectants (data not shown).

Overexpression of T-cadherin suppressed the prolifer-ation of HSC-1 To analyze the effect of T-cadherin on cu-taneous squamous carcinoma cell proliferation, we countedthe native HSC-1 cells, HSC-1 RNAI, HSC-1 TCAD, andHSC-1 TCAD RNAI, which expressed different levels of T-cadherin, and compared the difference in proliferation ac-tivity. HSC-1 cells proliferated well and after 96 h the cellnumber increased 13.8 times (Fig 2a). On the other hand,the T-cadherin overexpressing cell line, HSC-1 TCAD, onlyproliferated up to 5.6 times (Fig 2a). HSC-1 TCAD RNAIrestored the proliferation ability of HSC-1 TCAD to 11.1times (Fig 2a). To know the effect of T-cadherin in nativeHSC-1 cells, we compared the proliferation curve of nativeHSC-1 cells and HSC-1 RNAI. But the proliferation curve ofHSC-1 RNAI was almost the same as that of native HSC-1cells (Fig 2a).

Figure 1The expression level of T-cadherin on HSC-1 and transfectants. (a)Western blotting of T-, E-, P-cadherin on native HSC-1 (lane 1), HSC-1RNAI (lane 2), HSC-1 TCAD (lane 3), and HSC-1 TCAD RNAI (lane 4) celllysate. Equal amounts of cell lysate were separated by SDS-PAGE, andthen transferred to nitrocellulose membrane. The membrane was in-cubated with rat monoclonal anti-T-cadherin antibody, TCD-1, mousemonoclonal anti-human E-cadherin, HECD-1, and mouse monoclonalanti-human P-cadherin, NCC-CAD-299, and then incubated with HRP-conjugated secondary antibody. Signals were detected by ECL sys-tems. (b) Immunofluorescence staining of cadherins on HSC-1 (1, 5, 6),HSC-1 RNAI (2), HSC-1 TCAD (3), and HSC-1 TCAD RNAI (4). Cellswere immunoreacted with TCD-1 (1–4), HECD-1 (5), and NCC-CAD-299 (6). Primary antibody was detected by Cy3-conjugated secondaryantibody and examined using a fluorescence microscope. Scalebar¼ 10 mm.

Figure2Overexpression of T-cadherin resulted in the suppression of cellproliferation. Cells were incubated in Dulbecco’s modified Eagle’smedium containing 1% fetal calf serum for the indicated time period. (a)Cells were harvested and counted by a Coulter Counter. Results are themeans � SD of three independent experiments performed in triplicate.��po0.01 versus HSC-1, ##po0.01 versus HSC-1 TCAD by the Kruskal–Wallis rank test with Scheffe’s multiple comparison test. (b, c) Cellswere incubated for 1 h in medium containing tetrazolium salt or BrdUfor WST-1 assay (b) or BrdU incorporation (c), respectively. Results arethe means � SD of five independent experiments. ��po0.01 versusHSC-1 by Student’s t test.

834 MUKOYAMA ET AL THE JOURNAL OF INVESTIGATIVE DERMATOLOGY

To confirm the difference in the proliferation betweennative HSC-1 and T-cadherin overexpressing HSC-1 TCAD,other proliferation assays were performed. Both the WST-1cell proliferation assay and BrdU incorporation assay re-vealed that overexpression of T-cadherin led to the sup-pression of cell proliferation (Figs 2b and 3c).

Suppression of cell proliferation mediated by T-cad-herin was dependent on its expression level To analyzethe relationship between cell proliferation and the T-cad-herin expression level, we used Tet-inducible T-cadherinoverexpressing cells. HSC-1 Trex TCAD showed threebands, which consisted of the precursor (upper), matureprotein (middle), and a degradation product (lower) of T-cadherin, and these bands were upregulated by Tet in aconcentration-dependent manner (Fig 3a). Although theaddition of Tet to native HSC-1 cells did not affect cell pro-liferation, its addition to HSC-1 Trex TCAD suppressed cellproliferation concentration-dependently, which means theexpression level of T-cadherin negatively regulates cell pro-liferation in HSC-1 cells. Similar results were obtained usinga different cutaneous squamous carcinoma cell line, DJM-1,except that only mature T-cadherin was detected in DJM-1Trex TCAD (Fig 3c,d ). The difference between HSC-1 TrexTCAD and DJM-1 Trex TCAD was probably due to theexpression efficacy of T-cadherin.

Overexpression of T-cadherin induced a delay in theG2/M phase of the cell cycle To explore the mechanism of

T-cadherin-mediated downregulation of proliferation, cellcycle analysis was performed. Both cell lines, native HSC-1cells and HSC-1 TCAD, showed G1/G0 (2N) synchronizationby serum starvation and a similar pattern of the cell cyclewithin 12 h of serum restimulation (Fig 4). Although HSC-1cells showed a continuous reduction in the number of cellsin the G2/M (4N) phase until 12–24 h after serum restimu-lation, a larger number of HSC-1 TCAD were still retained inthe G2/M phase at 24 h after serum restimulation (Fig 4).These results clearly showed that overexpression of T-cad-herin affects the progress of the cell cycle at the point of G2

or M phase, and consequently delayed the cell cycle.

T-cadherin did not affect either cell aggregation or cellmigration To investigate the effect of T-cadherin on cell–cell adhesiveness or cell mobility in cutaneous squamouscarcinoma cells, we used HSC-1, HSC-1 RNAI, and HSC-1TCAD, which expressed a different amount of T-cadherin,and performed a cell aggregation assay and cell migrationassay. HSC-1 TCAD tended to show a slightly higher ag-gregation index than native HSC-1, but the difference wasnot significant. HSC-1 RNAI exhibited almost the samerates as native HSC-1 (Fig 5a). In the cell mobility assay,there was no significant difference in the migration ratesbetween native HSC-1 and HSC-1 TCAD.

Discussion

The aim of this study was to obtain an insight into the role ofT-cadherin in keratinocytes. We examined the effect ofoverexpression or knockdown of T-cadherin in immortalizedkeratinocyte cell lines derived from SCC, and the resultsclearly demonstrate that T-cadherin negatively regulates theproliferation of cutaneous squamous carcinoma cells.

Figure 3T-cadherin suppressed the cell proliferation according to its ex-pression level. (a, c) Western blotting of T-cadherin on (a) HSC-1 andtetracycline (Tet)-treated HSC-1 Trex TCAD cell lysate or (c) DJM-1 andTet-treated DJM-1 Trex TCAD cell lysate. Cells were incubated in thepresence or absence of various concentrations of Tet for 24 h. Equalamounts of cell lysate were separated by SDS-PAGE, and then trans-ferred to a nitrocellulose membrane. The membrane was incubatedwith TCD-1 and then incubated with HRP-conjugated secondary an-tibody. Signals were detected by ECL systems. Tet-induced T-cadherinover expression in a concentration-dependent manner. (b, d) Effects ofTet on cell proliferation in (b) HSC-1 and HSC-1 Trex TCAD or (d) DJM-1and DJM-1 Trex TCAD. Cells were incubated in the presence or ab-sence of various concentrations of Tet for 96 h, and the number of cellswas counted by a Coulter Counter. Results are the means � SD ofthree independent experiments performed in triplicate. ��po0.01,�po0.05 versus absence of Tet by the Kruskal–Wallis rank test withDunnett’s multiple comparison test.

Figure4T-cadherin-overexpressing cells stayed in G2/M phase for a longerperiod than native cells. HSC-1 and HSC-1 Trex TCAD cells weresynchronized at the G1/G0 phase (2N) by serum starvation and thenstimulated with Dulbecco’s modified Eagle’s medium containing 1%serum for 0, 6, 12, 18, and 24 h. DNA content was measured by a flowcytometer at each time point. Both HSC-1 and HSC-1 TCAD showed aslightly reduced distribution of cells in the G1/G0 phase (2N), and in-creased distribution of cells in the G2/M phase (4N) between 0 and 12h. Thereafter, HSC-1 cells had a slightly decreased distribution in theG2/M phase, whereas HSC-1 TCAD cells still had an increased distri-bution in the G2/M phase between 18 and 24 h.

T-CADHERIN AS A GROWTH SUPPRESSOR IN SCC 835124 : 4 APRIL 2005

We previously reported the loss of T-cadherin expressionin SCC (Takeuchi et al, 2002) and hypothesized that T-cad-herin was one of the tumor suppressor factors. Similar re-sults were obtained from breast, lung, ovarian, and bladdercarcinomas (Lee, 1996; Kawakami et al, 1999; Maruyamaet al, 2001; Toyooka et al, 2001). Other studies showed thatT-cadherin suppressed proliferation of glioma cells and ne-uroblastoma cells (Takeuchi et al, 2000; Huang et al, 2003),but the function of T-cadherin in cutaneous tissue was stillunclear. In the previous study, we also demonstrated thatthe expression of T-cadherin was significantly downregu-lated in the lesional skin of psoriasis vulgaris (Zhou et al,2003), which does not exhibit a malignant phenotype, and isconsidered to be a T lymphocyte-mediated keratinocytehyperproliferative disease. Further, in normal skin tissue, T-cadherin was specifically localized in the basal layer, wherekeratinocyte proliferation takes place (Zhou et al, 2002).Here, we showed that T-cadherin negatively regulated thecell proliferation of at least two kinds of immortalized ker-atinocyte cell lines, HSC-1 and DJM-1, in accordance withthe T-cadherin expression level. This finding suggests thatT-cadherin acts as an endogenous negative regulator ofkeratinocyte proliferation at the basal layer of the epidermis.The epidermis is a constantly self-renewing tissue and bal-ancing the cell loss from the cornified layer and the celldivision in the basal layer is essential for the maintenance ofskin structure. This regulation is thought to be achieved bygrowth control at the basal layer cell including two differentkeratinocyte subpopulations: the stem cells and transientamplifying cells (TAC) (Lavker and Sun, 2000; Watt, 2001).The relation between T-cadherin and the stem cell or TAC isa very interesting issue that remains to be elucidated.

The next question is how T-cadherin suppresses the cellproliferation of cutaneous squamous carcinoma cells. Huanget al (2003) reported that T-cadherin overexpressing gliomaC6 cells resulted in p21CIP1/WAF1-mediated G2 phase arrestand aneuploidy (44N) in cells. They also showed that T-cad-herin overexpressing cells exhibited significantly increasednumbers of cells in the G2/M phase after synchronization byserum starvation, compared with the vector-transfected cellsthat were synchronized in the G1/G0 phase by serum star-vation. In our studies, T-cadherin overexpressing HSC-1 cellswere able to be induced into the G1/G0 phase by serumstarvation, similar to native HSC-1 cells. After serum restim-ulation, T-cadherin overexpressing cells resulted in a delaybut not in an arrest at the G2/M phase and there were noaneuploid cells (44N). Although the precise mechanism ofT-cadherin-mediated suppression of proliferation is still un-known, it may depend on the difference of cell type. Addi-tionally, we confirmed that overexpression of T-cadherininduced no apoptosis in HSC-1 cells (data not shown).

In the epidermis, E- and P-cadherin are the major com-ponents of intercellular adherence junctions (Nose andTakeichi, 1986), and are involved in morphogenesis andmaintaining the structure of skin (Hirai et al, 1989). It is re-ported that the expressions of E- and P-cadherin are re-duced in SCC, melanoma, and Paget’s disease andreduction of cadherin expression is considered to be relat-ed to the detachment of carcinoma cells from the tumormass (Shirahama et al, 1996; Furukawa, 1997). The contri-bution of T-cadherin to keratinocyte cell–cell adhesion, how-ever, is not yet known. Vestal and Ranscht (1992) showed,using T-cadherin cDNA-transfected CHO cells, that T-cad-herin induced calcium-dependent, homophilic adhesion, likeE- or P-cadherin. In our present data from the cell aggre-gation assay, overexpression of T-cadherin tended to showslightly increased cell–cell adhesiveness, but the differencewas not significant. In addition, T-cadherin was detected atboth the cell–cell borders and the cell surface, whereasE- and P-cadherin were concentrated mainly at the cell–cellborders. These results indicate that the homophilic bindingactivity of T-cadherin is relatively weaker than that of E- orP-cadherin. This phenomenon may be explained by thespecific structure of T-cadherin, which lacks the cytoplasmicdomain and His–Ala–Val motif. So, we concluded that, in theepidermis, the major function of E- and P-cadherin is cell–cell adhesion, and that of T-cadherin must be signal trans-duction or cell recognition, rather than cell–cell adhesion.

In conclusion, this study has provided new informationon the functional significance of T-cadherin in cutaneoussquamous carcinoma cells. T-cadherin negatively regulatesthe proliferation of cutaneous squamous carcinoma cells,and downregulation of T-cadherin may be involved in thepathogenesis of SCC and psoriasis vulgaris. Further studieson the expression and regulation of T-cadherin in hyper-proliferative skin diseases will provide insights into not onlythe pathogenesis but also the development of novel ther-apeutic targets.

Materials and Methods

Cell culture and cell proliferation assay Human cutaneoussquamous carcinoma cell lines, HSC-1 (Kondo and Aso, 1981)

Figure 5T-cadherin did not affect either cell aggregation or cell migration.(a) Aggregation assay of HSC-1 cells and transfectants. Cells weretreated with trypsin in the presence of Ca2þ . The cell suspension inHCMF containing Ca2þ was allowed to aggregate for 20 min at 371Con a gyratory shaker. Results are the means � SD of the aggregationindex from three independent experiments performed in triplicate.There was no significant difference between HSC-1 and HSC-1 TCADor HSC-1 and HSC-1 RNAI according to the Kruskal–Wallis rank test.(b) Migration assay of HSC-1 and HSC-1 TCAD cells. Cells were cul-tured on a colloidal gold particle-coated plate for 24 h. The area wheresingle cells had migrated was measured. Results are the means � SDof ten independent cells. There was no significant difference accordingto Student’s t test.

836 MUKOYAMA ET AL THE JOURNAL OF INVESTIGATIVE DERMATOLOGY

and DJM-1 (Kitajima et al, 1987), were used in this study. The cellswere routinely cultured in Dulbecco’s modified Eagle’s medium(DMEM) with 10% fetal bovine serum (FBS), 10 ng per mL strep-tomycin, and 100 units per mL penicillin G at 371C in a humidifiedatmosphere of 5% CO2 in air. Cells were trypsinized and subcul-tured when they were approaching subconfluency. The number ofcells was determined using a Coulter Counter (Beckman Coulter,Fullerton, California).

To assess the cell proliferation, 2 � 103 cells per cm2 wereseeded in DMEM containing 1% FBS onto culture dishes for directcell counting or 96-well plates for a WST-1 cell proliferation assayand BrdU incorporation assay. WST-1, which is one of the tetra-zolium salts, was used to determine cell proliferation according tothe procedure provided by the manufacturer (Takara bio, Shiga,Japan). The BrdU incorporation assay was performed using a CellProliferation Biotrack ELISA System, ver2 (Amersham Bioscience,Piscataway, New Jersey) according to the protocol provided withthe kit.

Antibodies The primary antibodies used in this study included therat anti-T-cadherin monoclonal antibody, TCD-1, generated in ourlaboratory as outlined below. Rats were immunized by injectingpurified glutathione S-transferase (GST) fusion protein containing ahuman T-cadherin extracellular domain (residues 210–350) to thefootpad. After the immunization, lymph node cells from the ratswere fused with myeloma cells P3U1 (Matsuyoshi et al, 1997).Screening was carried out by ELISA, and a hybridoma clone, des-ignated TCD-1, was established. Its culture supernatant was usedin the following experiments. The mouse anti-human E-cadherinmonoclonal antibody HECD-1 and mouse anti-human P-cadherinNCC-CAD-299 were also used (Shimoyama et al, 1989). HRP-conjugated goat anti-rat IgG (DakoCytomation, Glostrup, Den-mark), HRP-conjugated goat anti-mouse IgG (DakoCytomation),Cy3-conjugated goat anti-rat IgG (CHEMICON International, Teme-cula, California), and Cy3-conjugated goat anti-mouse IgG (CHEM-ICON International) were used as secondary antibodies.

Plasmid constructs Total RNA from HSC-1 cells was purified us-ing ISOGEN (Nippon Gene, Tokyo, Japan), and cDNA was syn-thesized by reverse transcriptase Super Scripts II (Invitrogen,Carlsbad, California) using oligo dT primers. PCR was carriedout to obtain full-length cDNA of human T-cadherin using theforward primer, 50-cgg gcg ctt cta gtc gga caa aat gca gcc-30,and reverse primer, 50-agt caa gct tca gac gtc agg agt tct cac-30,and the product was subsequently cloned to the mammalianexpression vector pIRES neo2 (BD Biosciences Clontech, PaloAlto, California).

To construct the siRNA-human T-cadherin, the sense template,50-gat ccc ggt agt cga tag tga cag gtt caa gag acc tgt cac tat cgacta cct ttt ttg gaa a-30 (italic: BamHI cohesive end, bold: sensestrand, underline: antisense strand); and antisense template, 50-agc ttt tcc aaa aaa ggt agt cga tag tga cag gtc tct tga acc tgt cactat cga cta ccg g-30 (italic: HindIII cohesive end) were annealed andligated into a pSilencer 3.1-H1 hygro siRNA expression vector(Ambion, Austin, Texas).

The pIRESneo2-human T-cadherin was digested with EcoRVand NotI restriction endonucleases and the insert was ligated tothe appropriate sites of pcDNA 5/TO (Invitrogen), Tet-regulatedexpression vector.

All plasmid constructs were confirmed by sequencing using theABI PRISM BigDye Terminator cycle sequencing kit (Applied Bio-systems, Foster City, California).

Transfection The cells were transfected with various plasmidconstructs using lipofectamine plus reagent (Invitrogen).

HSC-1 cells were transfected with human T-cadherin-pIRESneo2 and selected with 250 mg per mL G418. Single cellclones were analyzed for T-cadherin expression using westernblotting, and HSC-1 TCAD expressing the highest level of T-cad-herin was established.

HSC-1 cells and HSC-1 TCAD cells were transfected with hu-man T-cadherin-siRNA-pSilencer. Cells were selected for 250 mgper mL hygromycin and screened with western blotting. HSC-1RNAI and HSC-1 TCAD RNAI with the lowest level of T-cadherinexpression were established.

HSC-1 cells and DJM-1 cells were co-transfected with humanT-cadherin-pcDNA5/TO and pcDNA6/TR (Invitrogen) that expressthe Tet repressor. Cells were selected with 250 mg per mLhygromycin and 5 mg per mL blasticidin double antibiotic resist-ance, and cultured in the presence or absence of 1 mg per mL Tetfor 48 h. Cell lysates were analyzed with western blotting for T-cadherin expression; HSC-1 Trex TCAD and DJM-1 Trex TCADwith Tet-inducible T-cadherin expression were established.

Cell cycle analysis Confluent cells were maintained for 48 h withserum free medium for G0 synchronization. Then, the cells weretrypsinized, cultured in DMEM containing 1% FBS for 6, 12, 18, 24h, and collected. The DNA content was determined by a Cycle-TEST PLUS DNA Reagent Kit (Beckton Dickinson, San Jose, Cal-ifornia), using the propidium iodide staining method, according tothe instructions provided by the manufacturer. Assays were per-formed with a flow cytometer (Epics XL-MCL, Beckman Coulter)and analyzed with EXPO32 software.

Cell aggregation assay Cadherin-mediated cell aggregation wasassayed as described (Matsuyoshi et al, 1992). Briefly, cells weretreated with 0.01% trypsin in the presence of 1 mM CaCl2 at 371Cfor 20 min, and then washed with Ca2þ - and Mg2þ -free Hepes-buffered (pH 7.4) HBSS (HMF) to obtain single-cell suspensions.5 � 104 cells in 0.5 mL HMF with 1 mM CaCl2 were incubated inalbumin-coated 24-well plates on a rotary shaker (80 r.p.m.) at371C for 20 min. The degree of cell aggregation was representedby the aggregation index (N0�N20)/N0, where N0 is the total cellnumber per well and N20 is the total particle number per well after20 min of incubation.

Cell migration assay Cover slips were coated with albumin andtreated with ethanol. After air drying, these cover slips were furthercoated with colloidal gold particles as described (Albrecht-Buehlerand Goldman, 1976). 1 � 103 cells were seeded on these coverslips placed into six-well plates and cultured for 24 h. Pictures ofthe migrated cells were taken and the areas where the cells hadremoved the colloidal gold particles were measured with AdobePhotoshop software (Adobe Systems, San Jose, California).

Immunofluorescence staining The cells cultured on chamberslides were fixed with PLP (10 mM sodium periodate, 75 mMlysine, and 2.14 mg per mL paraformaldehyde) for 10 min at roomtemperature, followed by blocking with 5% skim milk in 0.01 M pH7.4 phosphate-buffered saline (PBS) for 30 min at room temper-ature. The chamber slides were incubated with primary antibodyfor 60 min at room temperature. After being washed with PBS threetimes, cells were incubated with Cy3-conjugated anti-rat or anti-mouse secondary antibody diluted with 5% skim milk in PBS for 60min. The slides were washed again with PBS three times, andmounted with glycerol. The slides were viewed on a Nikon mi-crophot-SA fluorescence microscope (Nikon, Tokyo, Japan).

Western blotting Cells were lysed with 2% SDS lysis buffer con-taining 5%-mercaptoethanol, and boiled for 5 min. Protein con-centrations were measured by means of the Bio-Rad DC proteinassay (BIO-RAD, Hercules, California). Samples were separated bySDS-PAGE, followed by transfer to a nitrocellulose membranesheet. The membranes were blocked with 5% skim milk in PBS.After incubation with the primary antibody, the membrane waswashed with PBS for 5 min, three times, and then incubated withthe HRP-conjugated secondary antibody. Signals were detectedby using the ECL system (Amersham Pharmacia Biotech, Buck-inghamshire, UK).

T-CADHERIN AS A GROWTH SUPPRESSOR IN SCC 837124 : 4 APRIL 2005

Statistical analysis Statistical analyses were performed usingStudent’s t test and the Kruskal–Wallis rank test with Scheffe’smultiple comparison test or Dunnett’s multiple comparison test.Significance was defined as a value of po0.05.

We are grateful to Professor S. Kondo for kindly providing us with thehuman cutaneous squamous carcinoma cell line, HSC-1. We alsowould like to thank Dr S. Hirohashi and Professor M. Takeichi for thegifts of the mouse anti-human E-cadherin antibody, HECD-1, and theanti-human P-cadherin antibody, NCC-CAD-299.

DOI: 10.1111/j.0022-202X.2005.23660.x

Manuscript received September 6, 2004; revised November 8, 2004;accepted for publication November 29, 2004

Address correspondence to: Dr Norihisa Matsuyoshi, Department ofDermatology, Graduate School of Medicine, Kyoto University, 54kawahara-cho, Shogo-in, Sakyo-ku, Kyoto 606-8317, Japan. Email:nori@kuhp.kyoto-u.ac.jp

References

Albrecht-Buehler G, Goldman RD: Microspike-mediated particle transport

towards the cell body during early spreading of 3T3 cells. Exp Cell Res

97:329–339, 1976

Angst BD, Marcozzi C, Magee AI: The cadherin superfamily: Diversity in form and

function. J Cell Sci 114:629–641, 2001

Blaschuk OW, Sullivan R, David S, Pouliot Y: Identification of a cadherin cell

adhesion recognition sequence. Dev Biol 139:227–229, 1990

Doyle DD, Goings GE, Upshaw-Earley J, Page E, Ranscht B, Palfrey HC:

T-cadherin is a major glycophosphoinositol-anchored protein associated

with noncaveolar detergent-insoluble domains of the cardiac sarcolem-

ma. J Biol Chem 273:6937–6943, 1998

Furukawa F, Fujii K, Horiguchi Y, et al: Roles of E- and P-cadherin in the human

skin. Microsc Res Technol 38:343–352, 1997

Hirai Y, Nose A, Kobayashi S, Takeichi M: Expression and role of E- and

P-cadherin adhesion molecules in embryonic histogenesis. II. Skin

morphogenesis. Development 105:271–277, 1989

Hirohashi S, Kanai Y: Cell adhesion system and human cancer morphogenesis.

Cancer Sci 94:575–581, 2003

Huang ZY, Wu Y, Hedrick N, Gutmann DH: T-cadherin-mediated cell growth reg-

ulation involves G2 phase arrest and requires p21CIP1/WAF1 expression.

Mol Cell Biol 23:566–578, 2003

Kawakami M, Staub J, Cliby W, Hartmann L, Smith DI, Shridhar V: Involvement of

H-cadherin (CDH13) on 16q in the region of frequent deletion in ovarian

cancer. Int J Oncol 15:715–720, 1999

Kitajima Y, Inoue S, Yaoita H: Effects of pemphigus antibody on the regeneration

of cell–cell contact in keratinocyte cultures grown in low to normal Caþ þ

concentration. J Invest Dermatol 89:167–171, 1987

Kondo S, Aso K: Establishment of a cell line of human skin squamous cell

carcinoma in vitro. Br J Dermatol 105:125–132, 1981

Lavker RM, Sun TT: Epidermal stem cells: Properties, markers, and location. Proc

Natl Acad Sci USA 97:13473–13475, 2000

Lee SW: H-cadherin, a novel cadherin with growth inhibitory functions and di-

minished expression in human breast cancer. Nat Med 2:776–782, 1996

Maruyama R, Toyooka S, Toyooka KO, et al: Aberrant promoter methylation

profile of bladder cancer and its relationship to clinicopathological fea-

tures. Cancer Res 61:8659–8663, 2001

Matsuyoshi N, Hamaguchi M, Taniguchi S, Nagafuchi A, Tsukita S, Takeichi M:

Cadherin-mediated cell–cell adhesion is perturbed by v-src tyrosine

phosphorylation in metastatic fibroblasts. J Cell Biol 118:703–714, 1992

Matsuyoshi N, Toda K, Horiguchi Y, Tanaka T, Nakagawa S, Takeichi M, Imamura

S: In vivo evidence of the critical role of cadherin-5 in murine vascular

integrity. Proc Assoc Am Physicians 109:362–371, 1997

Nagafuchi A, Takeichi M: Cell binding function of E-cadherin is regulated by the

cytoplasmic domain. EMBO J 7:3679–3684, 1988

Nose A, Takeichi M: A novel cadherin cell adhesion molecule: Its expression

patterns associated with implantation and organogenesis of mouse em-

bryos. J Cell Biol 103:2649–2658, 1986

Nose A, Tsuji K, Takeichi M: Localization of specificity determining sites in cad-

herin cell adhesion molecules. Cell 61:147–155, 1990

Philippova MP, Bochkov VN, Stambolsky DV, Tkachuk VA, Resink TJ: T-cadherin

and signal-transducing molecules co-localize in caveolin-rich membrane

domains of vascular smooth muscle cells. FEBS Lett 429:207–210, 1998

Ranscht B, Dours-Zimmermann MT: T-cadherin, a novel cadherin cell adhesion

molecule in the nervous system lacks the conserved cytoplasmic region.

Neuron 7:391–402, 1991

Shimoyama Y, Hirohashi S, Hirano S, Noguchi M, Shimosato Y, Takeichi M, Abe

O: Cadherin cell-adhesion molecules in human epithelial tissues and

carcinomas. Cancer Res 49:2128–2133, 1989

Shirahama S, Furukawa F, Wakita H, Takigawa M: E- and P-cadherin expression

in tumor tissues and soluble E-cadherin levels in sera of patients with skin

cancer. J Dermatol Sci 13:30–36, 1996

Stappert J, Kemler R: A short core region of E-cadherin is essential for catenin

binding and is highly phosphorylated. Cell Adhes Commun 2:319–327,

1994

Takeuchi T, Liang SB, Matsuyoshi N, Zhou S, Miyachi Y, Sonobe H, Ohtsuki Y:

Loss of T-cadherin (CDH13, H-cadherin) expression in cutaneous

squamous cell carcinoma. Lab Invest 82:1023–1029, 2002

Takeuchi T, Misaki A, Liang SB, Tachibana A, Hayashi N, Sonobe H, Ohtsuki Y:

Expression of T-cadherin (CDH13, H-Cadherin) in human brain and its

characteristics as a negative growth regulator of epidermal growth factor

in neuroblastoma cells. J Neurochem 74:1489–1497, 2000

Toyooka KO, Toyooka S, Virmani AK, et al: Loss of expression and aberrant

methylation of the CDH13 (H-cadherin) gene in breast and lung carcino-

mas. Cancer Res 61:4556–4560, 2001

Vestal DJ, Ranscht B: Glycosyl phosphatidylinositol-anchored T-cadherin medi-

ates calcium-dependent, homophilic cell adhesion. J Cell Biol 119:

451–461, 1992

Watt FM: Stem cell fate and patterning in mammalian epidermis. Curr Opin Genet

Dev 11:410–417, 2001

Yap AS, Niessen CM, Gumbiner BM: The juxtamembrane region of the cadherin

cytoplasmic tail supports lateral clustering, adhesive strengthening, and

interaction with p120ctn. J Cell Biol 141:779–789, 1998

Zhou S, Matsuyoshi N, Liang SB, Takeuchi T, Ohtsuki Y, Miyachi Y: Expression

of T-cadherin in Basal keratinocytes of skin. J Invest Dermatol 118:

1080–1084, 2002

Zhou S, Matsuyoshi N, Takeuchi T, Ohtsuki Y, Miyachi Y: Reciprocal altered ex-

pression of T-cadherin and P-cadherin in psoriasis vulgaris. Br J Dermatol

149:268–273, 2003

838 MUKOYAMA ET AL THE JOURNAL OF INVESTIGATIVE DERMATOLOGY