A Specific Heterotypic Cell Adhesion Activity between Members of ...

17 downloads 0 Views 2MB Size Report
13, Issue of May 5, pp. 7995-8001, 1991. Printed in U.S.A.. A Specific Heterotypic Cell Adhesion Activity between Members of. Carcinoembryonic Antigen Family ...
THEJOURNAL OF BIOLOGICAL CHEMISTRY 0 1991 by The American Society for Biochemistry and Molecular Biology, Inc

Vol . 266,NO. 13, Issue of May 5 , pp. 7995-8001, 1991 Printed in U.S.A.

A Specific Heterotypic Cell Adhesion Activity betweenMembers of Carcinoembryonic Antigen Family, W272 and NCA, Is Mediated by N-domains” (Received for publication, October 22, 1990)

Shinzo OikawaS, Chikako InuzukaS,Motomu Kurokis, Fumiko Arakawag,Yuji Matsuokag, Goro Kosakill, and Hiroshi NakazatoS11 From the $Laboratory of Molecular Biology, Suntory Institute for Biomedical Research, 1-1-1 Wakayamadai, Shimamoto-cho, Mishima-gun, Osaka 618, the §First Departmentof Biochemistry, School of Medicine, Fukuoka University, 7-45-1 Nanakuma, Jonan-ku. Fukuoka 814-01. and the TlTokvo Metrooolitan Komagome Hospital, 3-18-22, Honkomagome, Bunkyo-ku, Tokyo 113, Japan

Carcinoembryonic antigen (CEA)’ is one of the mostwidely The Ca2*-independenthomotypic andheterotypic cell adhesion activitiesof a carcinoembryonic antigen used tumor markers whose serum levels are used in the long (CEA) family member, W272 (CGMG), whose cDNA term monitoring of patients following the surgery mainly for has recently been isolated from libraries of human adenocarcinoma of the colon, breast, and lung(6). Following the success of the cloning of the CEA cDNA (7, 8), numbers peripheral leukocytes of apparently normal subjects (Arakawa, F., Kuroki, Mo., Misumi, Y., Oikawa, S., of clones of cDNAs (9-11) and partialgenomic sequences (12, Nakazato, H., and Matsuoka,Y. (1990)Biochem. Bio- 13) encoding the CEA-related antigens are isolated toreveal phys. Res. Commun. 166, 1063-1071) and spleen of that the CEA family belongs to the Ig superfamily (14-16) chronic myelogenous leukemia patients (Berling, B., and consists of more than 17 members which are subdivided Kolbinger, F., Grunert, F., Thompson, J. A., Brom- into theCEA and thepregnancy-specific pl-glycoprotein (PSG) bacher, F., Buchegger, F., von Kleist, S., and Zimmer- subfamilies (17). The PSG prototypeis known to circulate in mann, W. (1990)Cancer Res. 50,6534-6539) hasbeen large amounts in maternal serumin pregnancy (18). The finding that CEA belongs to the Ig superfamily (14examined.Chinesehamsterovary cells transfected 16) led us to propose that it may play an important role in with the cDNA for W272, CEA, nonspecificcrosshomotypic reacting antigen(NCA), and various antigens contain- cell-cell interactions (15), the Ca”-independent cell adhesion activity of CEA (19), andhomotypic and hetering chimericN-domain havebeen used. The W272 producers did not show homotypic binding at all but otypic adhesions of CEA and nonspecific cross-reacting antibound only to the cells expressing NCA and a chimeric gen (NCA) (20), a CEA family member, being substantiated CEA whose N-domain is substituted by that of NCA, later. We have recently constructeda cDNA libraryof leukocytes indicating the major contribution of N-domain of NCA in the specific binding. The importance of the N-ter- isolated from pooled peripheral blood of about 100 healthy minal region ofNCA N-domain for the W272-NCA volunteers and screened it for CEA-related sequences which yielded about 300 such clones. We have found the existence binding has beenshownbydetailedanalysisusing COS-1 cells producing various NCA whose N-domain of cDNAs for, among others, NCA and W272 (21) (CGM6 a r e chimera of that of NCA and CEA. The strict het- (22)) whereas no cDNAfor CEA was found.2 Theabsence of erotypic nature of the W272-NCA adhesion strongly CEA in the leukocytes has also been shown immunochemihomotypic and suggests that the cell adhesion activities exhibited by cally (23).Itis of importancetotestthe in order CEA family members are not the fortuitous activity heterotypic adhesion activities among these antigens physiological role of the CEA family members. but the specific one which have some important phys- to elucidate the iological roles. EXPERIMENTALPROCEDURES3

AND RESULTS4

Characterization of Transfectants-After transfection of the CHO dhfr- cells with the expression vector for W272, cells Cell-cell andcell-substrateinteractions play important roles in variousaspects of biological phenomena (1-5) including development,morphogenesis, andvariousimmuneresponses such as T-cell activation, antigen presentation, circulation, and homing of lymphocytes. The interactions are mediated by various groups of molecules including those belonging to Ig superfamily, which is a group of glycoproteins that share structural features representedby Ig (4).

* The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate thisfact. )I To whom correspondence should be addressed. Tel.: 075-9629283: Fax: 075-962-6448.

The abbreviations used are: CEA, carcinoembryonic antigen; AA, amino acid(s);bp, base pairs; dhfr,dihydrofolate reductase; FBS, fetal bovine serum; HBSS, Hanks’ balanced salt solution; HRP, horseradish peroxidase; ICAM, intercellular adhesion molecule; IgV, immunoglobulin variable region; kb, kilobases; LFA, lymphocyte function-associatedantigen; NCA,nonspecific cross-reactingantigen; NCAM, neural cell adhesion molecule; PBS, phosphate-buffered saline; PSG, pregnancy-specific01-glycoprotein; RIA, radioimmunoassay; UTR, untranslated region; CHO, Chinese Hamster ovary. Mo. Kuroki, F. Arakawa, Y. Matsuo, S. Oikawa, Y. Misumi, H. Nakazato, and Y. Matsuoka, sumitted for publication. Portions of thispaper(including“ExperimentalProcedures,” Table 1, and Figs. 1, 2, 5, 7, and 8) are presented in miniprint at the end of this paper. Miniprint iseasily read with the aidof a standard magnifying glass. Full size photocopies are included in the microfilm edition of the Journal that isavailable from Waverly Press. Part of “Results” was presentedin aCEAworkshopheld in Montreal, Canada, June 12-14, 1990.

7995

7996

Specific

Heterotypic

Cell Adhesion

Activity

were cultured in the selection culture medium and 13 independent dhfr’ clones were isolated as described under “Experimental Procedures.” When cell extracts of the 13 clones along with the other representative high producer clones of CEA (20), NCA (20), C/N, and N/C were subjected to sandwich RIA, five clones were found to be high producers of W272 (clones 1, 4, 5, 8, and 10) (Table 1). The values of the expressed antigen represent only relative estimates, because anti-NCA antibody cross-reactive with W272 and NCA as a standard were used in the RIA. The discrepancy between the values for N/C using two systems described under “Experimental Procedures” illustrates this relativity (Table 1). However, they should be comparable among transfectants producing the same antigen if measured by the same system. By immunoblot analysis, W272 producers represented by W5 were shown to produce the antigen of about 80 kDa (Fig. 2, lane 1) which was a little higher than our previous estimate of the molecular mass of W272 expressed in COS-1 cells (21). The difference might be due to the extent and types of glycosylation of the W272 in two cell lines, which are left for future studies. Similarly, CHO transfectants expressing C/N or N/C chimerit protein were established. The C/N antigen has Ndomain derived from CEA and the rest derived from NCA, the N/C antigen, vice versa (Fig. 3b). Each of the representative high producer cell lines (C/N 7 and N/C 15) produced the corresponding antigen which was reactive with both antiCEA and anti-NCA antibodies (Table 1). The chimeric antigens appeared to be almost of the same molecular mass as the recombinant CEA or NCA produced by clone CEA 1 or NCA

a

CEA NCA

N

1

I

@I

C/N N/C FIG. 3. Cell adhesion activities of W272 expressing CHO cells. a, monolayers of CHO cells expressing the antigens denoted under the columns are added with ‘Cr-labeled W272 producers Wl, W5, and W8 for 30 min at 37 “C and assayed for the adhesion of ‘ICr as described under “Experimental Procedures.” The values are the average of two experiments. C5 is the control CHO cells transfected with the vector. 6, schematic representation of antigens expected to be exoressed on the surface of CHO cells used in a. Bones indicate N, I. II, and III domains as labeled; it should be noted that gg& m, and res ectively. M .-stand for N-domain of W272, CEA, and NCA, iil , andI%!, domains (7, 17) for W272, CEA, and NCA area, respectively. See Fig. 1, legend, for the description of domains I, II, and III.

between

CEA

2-4 (20), respectively

Family

Members

(Fig. 2, lanes 2 and 3 versus lanes 5 and

4).

Cell surface expression of each clone was confirmed by immunofluorescense staining or by using immunomagnet beads (Dynal, Norway) (data not shown). Aggregation of Transfectunts-Table 1 also shows the results of the aggregation assay on the producer of each antigen. As reported previously (20), CEA and NCA aggregated, respectively, extensively and weakly. C/N or N/C producer aggregated more or less than NCA or CEA producer, respectively. It should be noted here that the domain organization of C/N and NCA, and N/C and CEA is quite similar (Fig. 36). Interestingly, none of the W272 clones aggregated regardless of the amounts of the W272 expressed. These results were rather unexpected, for W272 is highly similar to NCA and CEA in primary structure and domain organization, except that CEA has extra domains II and III following domain I (Fig. 3b). Similarities at the AA level between W272 and NCA or CEA are 72 or 71% in N-domain and 84 or 78% in domain I, respectively (21). To analyze further the cell adhesion activity of W272, we have undertaken a more quantitative cell adhesion assay (20) as described under “Experimental Procedures.” Heterotypic Cell Adhesion of W272 Producers-As is shown in Fig. 3a, when three ‘Cr-labeled transfectants expressing W272 antigen were submitted to the adhesion assay, they did not bind to monolayers either of W272 (Wl, W5, and W8) or of CEA producers but bound to the monolayer of NCA producers. In order to study the domain(s) responsible for the specific binding, we have constructed expression vectors for the chimeric antigens which are illustrated in Fig. la, selected high CHO producers of the antigens as described above, and prepared monolayers of the cells. Interestingly, the W272 producers adhered only to the N/ C producer to the extent similar to or even a bit higher than that to the NCA producer and not at all to the C/N producer (Fig. 3a), clearly indicating that only the N-domain of NCA, but neither the N-domain of CEA nor the repetitive domain of NCA or CEA, mediates adhesion with W272. It should be noted that the heterotypic adhesions are very high, some of the binding being 50-60% (Fig. 3~) to which only the homotypic binding of the CEA producer is comparable (20). Adhesion of W272 Producers to COS-1 Cells Producing NCAs Carrying Chimeric N-domains-To analyze further the sequence(s) important for the specific binding, expression vectors for “NM” in which N-domain is a chimera of that of NCA and CEA have been constructed (Fig. lb) and transfected into COS-1 cells for the transient expression. The amounts of the antigens expressed were monitored by immunoblotting and RIA as described under “Experimental Procedures” and were about 600 ng/mg of protein for all cases except for the ABCd which expressed almost twice as much. Monolayers

of COS-1

transfectants

were

challenged

with

the

‘Cr-labeled CHO clone, W5, expressing W272. As shown in Fig. 4b, the binding to the positive control monolayer, NCA, was as high as 27% (lane 1) and that to the negative control, C/N, was only 2% (lane 8) indicating that the monolayers of transiently expressing cells could be used in our cell adhesion assay. As is shown in Fig. 4, the exchange of the 105th AA did not affect the adhesion activities indicating that Hislo of NCA is not responsible for the specific binding (lanes 2 and 7 versus lanes 1 and 8, respectively). We attributed the increase of binding in the case of ABCd (lane 2) to the almost 2-fold higher expression of the antigen as mentioned above. In

Specific Heterotypic Cell Adhesion Activity

I

I N

I 1

I

M

between CEA Family Members

0

10

7997

I

I

I

20

30

40

% Cell Bound

FIG. 4. Adhesion between W272 and NCA carrying NCA-CEA chimeric N-domain transiently exp r e s s e d o n t h e s u r f a of c e COS- 1 cells. a, schematic representation of the antigens expressed on the surface of COS-1 cells. Boxes show domains N, I, and M from left to right as indicated. N-domain is further divided into four regions which are designated A , B, C, and D for NCA derived and a, b, c, and d for those derived from CEA and indicated by open and black bones, respectively. The antigens are:I , ABCD (i.e. NCA); 2, ABCd; 3, ABcd; 4 , Abcd 5, aBCD; 6, abCD; 7, abcD; 8, abcd (i.e.C/N) according to thenaming described under “ExperimentalProcedures.” 79, 89, and 105, in one-letter The letters in the boxes denote amino acid residues 21, 27, 28, 29, 43, 44, 46, 78,66, notation, that are different between NCA and CEA in the regions. b, binding of ”Cr-labeled W272 producers to the monolayer of COS-1 cells producing various chimeric NCA shown to theleft of the column in a. The resultsof the two experiments areshown. The average % of bound 51Crin two experiments is shown.

contrast, when the N-terminal A and a regions containing cations.Theactive Ca2+ depletion by treatingboth V r four different AA are exchanged, the binding characteristics labeled and monolayer cells with 5 mM EDTA to chelate Ca2+ are changed so as to reflect the origin of the sequence (lanes did not change the results (data not shown). 4 and 5 uersw lanes 1 and 8 ) , indicating the importance of DISCUSSION the four AA, i.e. Alaz1 and the consecutive Asn27-Arg28-Ile29, or some of them for the specific binding activity of NCA. The CEA family belongs to the Ig superfamily, and all its Especially noteworthy residues are AAZ8,positively charged, members so far identified including those of the PSG subfamhydrophilic arginine for NCA, andnonpolar, hydrophobic ily carry an IgV-like domain at the N-terminal end(17, 34). leucine for CEA. Apparently, however, the A region is not It is therefore conceivable that the IgV-like N-domain of the enough to represent the whole N-domain of NCA (compare CEAfamily members may convey a t least some of their clones expressing lanes 1 and 4 ) . Interestingly, B and C regions of NCA appar- functions. Our present study using CHO ently exert negative and positive effects, respectively, on the chimeric NCA and CEA in which N-domains are exchanged NCA for W272 binding(compare lanes 1, 3, and 4 or lanes 8, 6, and 5 ) . clearlyshowed that the binding site(s) on reside in the N-domain (Fig. 3), which may give us some clues Furthermore, the b region of CEA shows definitely positive effect (compare lanes 3 and 4 or lanes 5 and 6 ) . The positive useful in studying such functions. The importanceof the Ncharge and/or thehydrophilicity of the consecutive dipeptide domain for adhesion activity of CEA family members was CEA AN Arg43-Gln44 in CEA seemto beresponsible for theeffects. The alsosupported by ourunpublishedfindingthat of the N-domainof CEA) expressingCHO 46th AA in both CEA and NCA are hydrophobic to similar (lacking two-thirds extents and would not exert much effects. Taken together, cellshave neither homotypic nor heterotypic cell adhesion these results strongly suggest that the four AA, at most, in activity between NCA producers, and that anti-CEA monothe A region of NCA are mostresponsible for the heterotypic clonal antibodies which recognize the N-domain of CEA can adhesion of NCA to W272, although otherregions might have inhibit the CEA homotypic adhesion a ~ t i v i t i e s . ~ To further analyze the structural bases for the heterotypic some influence on the binding. The finer mapping of the AA residues that exertpositive and negative effectson the binding interaction between W272 and NCA, primary and secondary structures of the N-domains of W272, NCA, and CEA were activity is left for future studies. compared (Fig. 6). T h e similarity of the AA sequences between Heterotypic Adhesion Activity of W272 Does Not Require CEA and NCA is the highest, i.e. 89% or 12 AA differences Ca2+-Some adhesion molecules requirefor their activities homotypic in this domain (9). However, the domain of W272 showed divalent cations, e.g. cadherin requires Ca2+ for its relatively low similarity to thoseof NCA and CEA, i.e. 30 or binding activity (2), LFA-1 requires M$+ for the binding to 31 AA are different between W272 and NCA or CEA, respecICAM-1 (33), while the others, e.g. NCAM, do not (1).Aptively. Interestingly, the AA differences are reflected in the parently, CEA and NCA belong to the latter group (19, 20). differences of secondary structures that ispeculiarly characAs can be seen in Fig. 5 , W272 producers showed adhesion teristic to the Ig fold, which is the common structure for the activities notchanged in bothspecificity and extent when the experiment was done under divalent cation-poor conditions ’S. Oikawa, C. Inuzuka, Mo. Kuroki, F. Arakawa, Y. Matsuoka, showingclearly that W272 also does not require divalent G. Kosaki, and H. Nakazato, manuscript in preparation.

Specific Heterotypic Cell Adhesion Activity

7998

CEA NCA

between CEA Family Members

"

- -

"

A

B

"

C

C'

FIG. 6. Comparison c)f primary and secondary structures of N-domains of W272, CEA, and NCA. Comdete AA seauence of N-domain of CEA in one-letter notation is aligned with that of NCA and W272. Dashes indicate identity-and the residues different from that of CEA are shown; those different between CEA and NCA arein bold letters. A to D on top show the regions exchangedbetween CEA and NCA as described under "Experimental Procedures." Thick horizontal lines depict regions of 0 sheet structure whose naming in Ig (35) is shown on the bottom. Black boxes indicate probably most important residues.

some physiological conditions arevery large. Also, localization of the antigen at some part of the cell surface in vivo may well take place and render thelocal concentration very high. The isolation of cDNAs, for not only W272 but also NCA, from a cDNA library of leucocytes pooled from about 100 apparently healthyvolunteers: or demonstrationof the presence of mRNAs for both in the leucocytes of CML patients (22, 36) and the specific heterotypic adhesion between W272 and NCA presented here seem to suggest the involvement of W272 and NCA in some aspects of the immunological systems. In view of this, it should be noted CEA that is not expressed in leukocytes but expressed on epithelial cells (23, 37), NCA in both kindsof cells: and W272 not in epithelialcells but in leukocytes (22). In addition to the involvement of CEA in morphogenesis of colonic tissues and colonic tumors (191, CEA familymembers may be involved intheinteraction between circulating cells and/or between circulating cells and solid tissues. Furthermore, interactions of the members circulating in blood streams likesecretory PSG (18) or CEA 4). Fig. 7 summarizes schematically the results obtained in thisreleasedfrom tumors (38) with cells ormatrixshould be and our previous studies (20). We previously demonstrated considered. Apparently more works on the distributionof the that not only CEA but also NCA expressed on the CHOcell antigens and theregulation of expression, among others, will surface couldfunction in vitroas Ca2'-independent homotypic be needed to understand thephysiological roles of the memcell adhesion molecules. Also shown was the heterotypic adhe- bers of this family. Finally, possible binding models between W272 and NCA sion between NCA and CEA producers (notshown in Fig. 7). In the present study we have demonstrated that in contrast are schematically illustrated in Fig. 8. It has been taken into account that, although the presence of N-domain of NCA to CEA and NCA, W272 does not function as a homotypic adhesion molecule nor possess heterotypic adhesion activity seems to bea necessary and sufficient condition for NCA to between CEA. It has been shown, however, that W272 is a bind to W272 as discussed above, we do not know if either of W272 or heterotypic cell adhesion molecule interacting with NCA in the N-domain binds to N-domain or another part if the domain l a and/or 1bof NCA participate in the binding vitro. The presence of not only homotypic and heterotypic cell by stimulating or stabilizing the reaction. If they are coexadhesion activities (20) of rather loose specificities,6 but also pressed on a cell, the parallel bindingwill take place (Fig. 8). The same size and domain organization of the protein moiethe rigorously specific heterotypic one reported here, seems t o strongly suggest that CEA family members function ascell ties of both antigens may facilitate the binding. This type of adhesion molecules invivo. Although thisnotion may be heterodimer formation is knownfor major histocompatibility challenged on the basisof the possibility that the amountsof complex antigens, Ig, andT-cellreceptors (4). Whenthe interaction takes placebetween the antigens expressed on the antigens expressed on the surface of CHOorCOS-1 different cells, antiparallel interactions may take place. The transfectants are too large t o be physiological, it is quite possible that those expressed invivo on the surface of a binding between N-domain and N-domain (Fig. 8b, upper), l a (Fig. 8b, lower), orN-domain/ particular cell at a particular timeof its development or underN-domainanddomain molecules belonging to the Ig superfamily (4).IgV domains have two /3 sheets folded usually by aconserveddisulfide bond, with strands conventionally labeled ABCC'C"DEFG; strands ABED formingone sheet, strands GFCC'C" the other (35). When the Chou and Fasman (32) calculation was applied, seven to eight /3 strands (A-G) were predicted for the N-domain of the three CEA family members (Fig. 6). Analogous to IgV, and for convenience, we took the CEA as the standard and labeled its p strands ABCC'DEFG. The position and the length of the /3 strands are fairly similar among the three members withsome interesting differences. The pattern for W272 is most different lacking /3 strand A and having /3 strand C", which may explain the peculiar cell adhesion activitiesof W272-expressing cells. Another interesting feature is that p strand C, present in CEA and W272, is missing in NCA. This difference seems to result from the consecutive three-AA substitution (His-LeuPhe in CEA and Asn-Arg-Ile in NCA at position 27-29) which is found to be most important for the adhesion activity(Fig.

Specific Heterotypic

Cell Adhesion Activity

domain l b and domain la/domain l a (not shown) may be possible like inthe case of CD4-majorhistocompatibility complex antigens (39,40) etc. Further studiesusing recombinant antigensincluding crystallography are definitely needed to clarify these points. ~

~~

REFERENCES 1. Edelman, G. M. (1988) Biochemistry 27,3533-3543 2. Takeichi, M. (1988) Development 102, 639-655 3. McClay, D. R., and Ettensohn C. A. (1987) Annu. Reu. Cell. Biol. 3,319-345 4. Williams, A. F., and Barclay, A. N. (1988) Annu. Reu. Immunol. 6,381-405 5. Hemler, M. E. (1990) Annu. Reu. Immunol. 8,365-400 6. Goldenberg, D. M., Neville, A. M., Carter, A. C., Go, V. L. W., Holyoke, E. D., Isselbacher, K. J., Schein, P. S., Schwartz, M., McIntire, K. R., and Greenberg, L. P. (1981) Cancer Res. 41, 2017-2018 7. Oikawa, S., Nakazato, H., and Kosaki, G. (1987) Biochem. Biophys. Res. Commun. 142,511-518 8. Beauchemin, N., Benchimol, S., Cournoyer, D., Fuks, A., and Stanners, C. P. (1987) Mol. Cell. Biol. 7, 3221-3230 9. Tawaragi, Y., Oikawa, S., Matsuoka, Y., Kosaki, G., and Nakazato, H. (1988) Biochem. Biophys. Res. Commun. 150, 89-96 10. Neumaier, M., Zimmermann, W., Shivery, L., Hinoda, Y., Riggs, A. D., and Shivery, J. E. (1988) J . Biol. Chem. 263, 3202-3207 11. Hinoda, Y., Neumaier, M., Hefta, S., Drzeniek, Z., Wagener, C., Shivery, L., Hefta, L. J. F., Shivery, J . E., and Paxton, R. J. (1988) Proc. Natl. Acad. sci. U. s. A. 85,6959-6963 12. Oikawa, S., Kosaki, G., and Nakazato, H. (1987) Biochem. Biophys. Res. Commun. 146,464-469 13. Thompson, J. A,, Mauch, E., Chen, F., Hinoda, Y., Schrewe, H., Berling,B., Barnert, S., von Kleist, S., Shively, J. E.,and Zimmermann,W. (1989) Biochem. Biophys. Res. Commun. 158,996-1004 14. Paxton, R. J., Mooser, G., Pande, H., Lee, T. D., and Shively, J. E. (1987) Proc. Natl. A c Q ~Sci. . U. S. A. 84, 920-924 15. Oikawa, S., Imajo, S., Noguchi, T., Kosaki, G., and Nakazato, H. (1987) Biochem. Biophys. Res. Commun. 144, 634-642 16. Zimmermann, W., Ortlieb, B., Friedrich, R., and von Kleist, S. (1987) Proc. Natl. Acad. Sci. U. S. A. 84, 2960-2964 17. Barnett, T., and Zimmermann, W.(1990) Tumor Biol. 11,59-63 18. Tatarinov, Y. S., and Masyukevich, V. N. (1970) Byull. Elzsp. Biol. Med. U. S. S. R. 69, 66-68 19. Benchimol, S., Fuks, A,, Jothy, S., Beauchemin, N., Shirota, K., and Stanners, C. P. (1989) Cell 57, 327-334

between CEA Family Members

7999

20. Oikawa, S., Inuzuka, C., Kuroki, Mo., Matsuoka, Y., Kosaki, G., and Nakazato, H. (1989) Biochem. Biophys. Res. Commun. 164,39-45 21. Arakawa, F., Kuroki, Mo., Misumi, Y., Oikawa, S., Nakazato, H., andMatsuoka, Y. (1990) Biochem.Biophys.Res.Commun. 166. 1063-1071 22. Berling, B., Kolbinger, F., Grunert, F., Thompson, J. A., Brombacher, F., Buchegger, F., von Kleist, S., and Zimmermann, W. (1990) Cancer Res. 50, 6534-6539 23. Kuroki, Mo., Matsuo, Y., Kuroki, Ma., and Matsuoka, Y. (1990) Biochem. Biophys. Res. Commun. 166,701-708 24. Urlaub, G., and Chasin, L. A. (1980) Proc. Natl. Acad. Sci. U. S. A. 77, 4216-4220 25. Gluzman, Y. (1981) Cell 23, 175-182 26. Matsuoka, Y., Kuroki, Ma., Koga, Y., Ogawa, H., Nakazawa, N., Tachibana, S., and Minamizawa, T. (1983) J. Immunol. Meth. 58,31-47 27. Kuroki, Ma., Ichiki, S., Kuroki, Mo., and Matsuoka, Y. (1984) J . Natl. Cancer Znst. 69,401-408 28. Matsuoka, Y., Koga, Y., Maruta, H., Yoshino, M., and Tsuru, E. (1978) Int. J . Cancer 21, 604-610 29. Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) J . Biol. Chem. 193, 265-275 30. Laemmli, U. K. (1970) Nature 227, 680-685 31. Suzuki, R., Handa, K., Itoh,K.,andKumagai,K. (1983) J. Immunol. 130,981-987 32. Chou, P. Y., and Fasman, G. D. (1978) Adu. Enzymol. 47, 45148 33. Simmons, D., Makgoba, M. W., and Seed, B. (1988) Nature 331, 624-627 34. Nakazato, H., Oikawa, S., Kuroki, Mo., Matsuoka, Y., and Kosaki, G. (1989) in The Carcinoembryonic Antigen Gene Family (Yachi, A,, and Shivery, J. E., eds) pp. 75-86, Elsevier Science Publishers BV, Amsterdam 35. Lesk, A. M., and Chothia, C. (1982) J. Mol. Biol. 160, 325-342 36. Cournoyer, D., Beauchemin, N., Boucher, D., Benchimol, S., Fuks, A,, and Stanners, C. P. (1988) CancerRes. 48, 31533157 37. Kuroki, Mo., Kuroki, Ma., Moore, G., Ichiki, S., and Matsuoka, Y. (1988) Jpn. J . Cancer Res. 79, 82-90 38. Thomson, D. M. P., Krupey, J., Freedman, S. O., and Gold, P. (1969) Proc. Natl. Acad. sci. U. S. A,, 64, 161-167 39. Wang, J., Yan, Y., Garrett, T. P. J., Liu, J., Rodgers, D. W., Garlick, R. L., Tarr, G. E., Husain, Y., Reinherz, E. L., and Harrison, s. C. (1990) Nature 348,411-418 40. Ryu, S. E., Kwong, P. D., Truneh, A,, Porter, T. G., Arthos, J., Rosenberg, M., Dai, X., Xuong, N., Axel, R., Sweet, R. W., and Hendrickson, W. A. (1990) Nature 348, 419-426

Specific Heterotypic Cell Adhesion Activitybetween CEA Family Members

8000

KUCEAhlmVCan~hmlS(FI9.W

1:ng. I. Slrategm for the construction of cxpression VCCIO~S. Only the porttons of lnscrls hut not of Y E C I O ~ S arc