sign in sign up
<<
Home , Carcinoembryonic antigen, PSG2

Proc. Natl. Acad. Sci. USA Vol. 88, pp. 5533-5536, July 1991 Medical Sciences Receptor for mouse hepatitis virus is a member of the carcinoembryonic antigen family of (coronavirus/immunoglobulin superfamily/host resistance to virus/virus tissue tropism/virus receptor)

RICHARD K. WILLIAMS, GUI-SEN JIANG, AND KATHRYN V. HOLMES* Department of Pathology, Uniformed Services University of the Health Sciences, 4301 Jones Bridge Road, Bethesda, MD 20814 Communicated by Dorothy M. Horstmann, April 1, 1991 (receivedfor review February 28, 1991)

ABSTRACT The receptor for mouse hepatitis virus binds neither MHV-A59 virus nor anti-receptor mAb CC1, (MHV), a murine coronavirus, is a 110- to 120-kDa glycopro- we postulated that it has a mutation affecting the domain of tein on intestinal brush border membranes and hepatocyte the that normally would bind MHV and mAb membranes. The N-terminal 25-amino acid sequence of immu- CC1 (4). In this communication we demonstrate that the noaffinity-purified MHV receptor was identical to the pre- MHV receptor glycoprotein of BALB/c and Swiss Webster dicted mature N termini of two mouse related to human mice and its SJL/J homolog are members of the carcinoem- carcinoembryonic antigen (CEA) and was strongly homologous bryonic antigen (CEA) family of glycoproteins in the immu- to the N termini of members of the CEA family in humans and noglobulin superfamily. rats. Polyclonal to human CEA recognized the immunoaffinity-purified MHV receptor and the MHV receptor in liver membranes and intestinal brush border membranes MATERIALS AND METHODS from MHV-susceptible mouse strains. In membranes from Tissue Preparations and Tissue-Binding Assay. Intestinal MHV-resistant SJL/J mice, the anti-CEA antibodies recog- brush border membranes were prepared as described (3) from nized a homologous glycoprotein that failed to bind MHV. The adult female BALB/c and SJL/J mice. Mouse brain, colon, MHV receptor glycoprotein was detected in membranes of kidney, liver, lung, small intestine, spinal cord, and spleen BALB/c colon, small intestine, and liver, which are the were homogenized in phosphate-buffered saline (PBS; 0.137 principal targets for MHV replication in vivo. The MHV M NaCI/2.7 mM KCI/8.05 mM Na2HPO4.7H20/1.47 mM receptor glycoprotein resembled members of the human CEA KH2PO4) and centrifuged at 1000 x g for 5 min. The pellet family in molecular weight, acidic pl, extensive glycosylation, containing nuclei and debris was discarded, and particulate solubility in perchloric acid, and tissue distribution. Thus, the material was obtained by centrifuging the supernatant at MHV receptor is, to our knowledge, the first member of the 38,720 x g for 30 min. The pellet was washed by resuspending CEA family ofglycoproteins to be identified as a virus receptor. in PBS, centrifuging for 30 min at 38,720 x g, resuspended in PBS, and frozen at -70°C. Radiolabeled binding to Coronaviruses are common pathogens that cause respiratory these preparations was quantitated as described (10). illness in humans and respiratory and/or enteric diseases in Purification and Sequencing. MHV receptor from many animal species. The murine coronavirus, mouse hep- Swiss Webster mouse liver was immunoaffinity-purified on a atitis virus (MHV), infects mouse liver, epithelial cells of the mAb CC1 column as described (4) except that receptor was intestinal and respiratory tracts, macrophages, neural cells, eluted with a linear gradient of sodium thiocyanate from 1.0 and lymphocytes and is an excellent model for the study of to 5.0 M. MHV receptor protein from three 100-g batches viral infections in vivo (1, 2). Infection of cells is initiated were pooled. Concentrated receptor was electrophoresed on when the virus binds to a 110- to 120-kDa glycoprotein that an SDS/PAGE gel and blotted to poly (vinyl difluoride) is found on the plasma membranes of hepatocytes and membrane for N-terminal amino acid sequencing (11). Here, intestinal epithelial cells of MHV-susceptible BALB/c mice to maximize receptor on the membrane, 10 mM tetraborate (3). The MHV receptor has an acidic pI of 3.5-4.5 (4) and is (pH 9.6) was substituted for the standard 10 mM 3(Cyclo- highly glycosylated, with >35% of its total mass comprised of N-linked carbohydrate (R.K.W., unpublished data). hexylamino Propane Sulfonic acid (CAPS) (pH 11.0) buffer Monoclonal anti-receptor antibody CC1 (mAb CC1) blocks (G.S.J., unpublished data). Sequence was obtained from 10 binding of MHV to the receptor, prevents MHV infection of ,g of receptor glycoprotein at the Protein Sequencing Lab- cells in vitro (K.V.H., unpublished data), and inhibits virus oratory of the University of California at Davis. replication and dissemination in suckling mice in vivo (5). Antibodies and Antisera. The mAb CC1 directed against the The MHV receptor plays an important role in the patho- MHV receptor and the rabbit antibody directed against the genesis of MHV infections. Adult SJL/J mice are highly N-terminal 15 amino acids of the MHV receptor (anti-NTR) resistant to infection with MHV compared with other strains were produced as described (4). Rabbit antisera to human of mice such as BALB/c (6, 7). Resistance of SJL/J mice to CEA were from DAKO (Carpinteria, CA; A115, lot no. 067) MHV infection is a recessive trait mapped to the Hv-2 locus and Accurate Chemicals (Westbury, NY; AXL238, lot no. on mouse 7 (8, 9). Membranes from intestine 067G). Rabbit antiserum to human pregnancy-specific /3-1 and liver of SJL/J mice do not bind MHV or anti-receptor glycoprotein (PSG) was from Accurate Chemicals (West- mAb CC1, suggesting that resistance to infection is due to bury, NY; AXL 254, lot no. 24E). absence of a functional virus receptor (3, 4). SJL/J mice express a 100- to 110-kDa glycoprotein antigenically related Abbreviations: MHV, mouse hepatitis virus; CEA, carcinoembry- to the MHV receptor. Because this SJL/J receptor homolog onic antigen; mAb, ; NTR, N-terminal 15 amino acids of MHV receptor; NCA, normal cross-reacting antigen; mmCGM2 and mmCGM4, mouse CEA-related CDNA clones; The publication costs of this article were defrayed in part by page charge mCGM4, rat CEA-related 4; BGP, human biliary glycoprotein payment. This article must therefore be hereby marked "advertisement" I; PSG2, human pregnancy-specific /3-1 glycoprotein (E). in accordance with 18 U.S.C. §1734 solely to indicate this fact. *To whom reprint requests should be addressed.

5533 Downloaded by guest on September 26, 2021 5534 Medical Sciences: Williams et al. Proc. Natl. Acad. Sci. USA 88 (1991) Biochemical Analysis. To test for solubility of receptor to the MHV receptor. PSG2 is a member activity in perchloric acid, colon from adult female BALB/c of a subgroup of the CEA family that contains several mice was sliced longitudinally, cleaned in ice-cold Dulbec- glycoproteins that are highly expressed in human placenta co's PBS, and homogenized with a motorized homogenizer (19, 20). CEA, the prototype molecule of the CEA family, is (Tekmar Tissuemizer) for 2 min in sterile distilled water. An an acidic (pI 3-4) 180-kDa cell-surface glycoprotein with equal volume of 1.2 M perchloric acid was added to the -50o of its molecular mass being carbohydrate and is an homogenate, and after 15 min, precipitated were important of human of the centrifuged for 20 min at 35,000 x g at 4(C to form a pellet. colon, , and (14, 25). CEA is also Supernatant material was collected, neutralized to pH 7 with found, in much lower abundance, on normal colonic epithe- Tris base, desalted, and concentrated by using centrifugal lium. NCA-95 and NCA-55 share several antigenic sites with microconcentrators (Centricon 30, Amicon Div., W. R. CEA and are acidic glycoproteins expressed in normal colon, Grace, Danvers, MA). liver, lung, and spleen and on polymorphonuclear neutrophils Immunoblotting and Virus Blotting. SDS/polyacrylamide and macrophages. BGP is a 90-kDa CEA-related protein in gels (12) were transferred to nitrocellulose membranes and normal adult human liver (26). The predicted N-terminal probed with polyclonal antibodies or mAbs, followed by amino acid sequences ofthe rat genomic sequences rnCGM1, radioiodinated staphylococcal protein A and autoradiogra- rnCGM2, rnCGM3, rnCGM4, and mCGM5, which encode a phy (13). For virus-overlay protein blots (3), membrane family of genes related to human CEA (17), were also proteins were electrophoresed, blotted the same as for im- homologous to the N terminus of the MHV receptor. Of the munoblots, and then incubated with the A59 strain of MHV. rat CEA-like sequences, rnCGM4 was the most closely Bound virus was detected with anti-viral antibody followed related to the N terminus of the MHV receptor, with 64% by radioiodinated staphylococcal protein A as described (3). identity in the first 25 amino acids (Fig. 1). Rat ecto-ATPase, a 95- to 110-kDa molecular mass cell-surface nucleotidase found on the bile canalicular surface of hepatocytes and on RESULTS intestine brush borde,r is highly similar to human BGP (18, Affinity purification of the MHV receptor from Swiss Web- 27). The N terminus ofthe MHV receptor shares 60o identity ster mouse liver membranes (4) was scaled up to yield with rat ecto-ATPase. The predicted N-terminal amino acid sufficient glycoprotein to permit sequencing of the first 25 sequences from the two mouse CEA-like transcripts, amino acids. We searched the National Biomedical Research mmCGM2 and mmCGM4 (15), were identical to the N-ter- Foundation/Protein Identification Resource protein se- minal sequence we obtained from the MHV receptor protein. quence data base for sequences related to this 25-amino acid CEA, NCA, and BGP have the unusual property of being sequence and found only members ofthe human CEA family soluble in high concentrations of perchloric acid (14). To of glycoproteins (14) in the best matching sequences. Rat compare the biochemical properties of the MHV receptor CEA-like DNA sequences were retrieved from the GenBank with the CEA family, mouse colon was homogenized in water DNA data base. Murine CEA-like CDNA clones mmCGM2 and treated with perchloric acid (0.6 M final acid concentra- and mmCGM4 isolated from a mouse colon cDNA library tion) to precipitate non-CEA-like proteins according to pro- probed with human CEA cDNA have been reported (15). The cedures for isolation of CEA and NCA (14). Perchloric acid alignment of the first 25 amino acids of the MHV receptor solubility of the MHV receptor was examined by blotting the glycoprotein with four members ofthe human CEA family of supernatant with MHV-A59 virus or with anti-receptor mAb glycoproteins is shown in Fig. 1. The basic structure of the CC1 (Fig. 2). MHV receptor glycoprotein was found in the CEA family of proteins is a 34-amino acid leader sequence perchloric acid supernatant, indicating that a substantial that is cleaved by signal peptidase and a 108-amino acid proportion of the MHV receptor glycoprotein in colon pos- N-terminal region that is followed by one or more repeated sesses CEA-like solubility characteristics. motifs containing two immunoglobulin-like domains. The The tissue distribution of the MHV receptor in the carboxyl terminus ofthe protein can be either attached to the BALB/c mouse was examined by binding "25I-labeled anti- membrane by phosphoinositol glycolipid or inserted into the receptor mAb CC1 to pellets containing cellular membranes membrane by a transmembrane domain (14). The N terminus (Fig. 3). To demonstrate the specificity of binding, we of the MHV receptor protein was homologous to the N included tissues from the SJL/J mouse that do not bind MHV termini of the mature CEA-related proteins beginning after or mAb CC1 (3, 4). The highest levels of MHV receptor the 34-amino acid leader sequence. Between the first 25 measured by the binding of radiolabeled mAb CC1 were amino acids ofthe MHV receptor and the first 25 amino acids found in BALB/c colon, small intestine, and liver. Receptor of the human CEA family members, 60-72% of the amino was not detectable in the remaining BALB/c tissues with this acid positions were identical, and 72-84% were conserved. assay. Thus, expression of the MHV receptor was similar to Of all the human proteins, PSG2 had the best amino acid that of human NCA and CEA, which are expressed prefer- percent percent iFEVTIEA V PPQVAEDNNVLLLVHNLP 25 MHV R identity similarity EVTIEAVPPQVAEDNNVLLLVHNLP mmCGM2,4 100 100 QV IVPFV[ V-E ENVLLLVHNLP rnCGM4 68 80 Q VT VAVPP N V V E EKS VLLLEH N LP EctoATPase 64 80 Q VT I E AQ P P K VSJE GKD V L L LVH N L P PSG 2 72 84 KLTIESTPFNVAEGKEVLLLVHNLP CEA 64 76 QLTESM PFNVAEGKE VLLLVHNLP BGP 60 80 KL TIESTPFN VAEGKEVLLLAHNLP NCA 60 72 FIG. 1. Alignment ofthe N-terminal amino acid sequence ofthe MHV receptor (MHV R) protein with members ofthe mouse, rat, and human CEA family of glycoproteins. Sequences were aligned by the WORDSEARCH and BESTFIT programs (16). mmCGM2,4 mouse mmCGM2 and mmCGM4 (15); mCGM4, rat CEA-related gene 4 (17); Ecto-ATPase, rat liver ecto-ATPase (18); PSG2, human pregnancy-specific P3-1 glycoprotein E (19, 20); NCA, normal cross-reacting antigen (21, 22); BGP, biliary glycoprotein 1 (23). The sequence of human CEA is from ref. 24. Downloaded by guest on September 26, 2021 Medical Sciences: Williams et al. Proc. Natl. Acad. Sci. USA 88 (1991) 5535

1 2 3 4 220 _ 220_.

100 _b.- 100_0 II"' 68_o 68_-

43 _ 43_-

1 2 3

FIG. 2. CEA-like solubility of the MHV receptor in perchloric FIG. 4. Antigenic cross-reactivity between the MHV receptor acid. Proteins in the SDS/polyacrylamide gels were blotted to and CEA. Affinity-purified MHV receptor from Swiss Webster nitrocellulose. Lanes: 1 and 4, 5 ,ug ofperchloric acid-soluble protein mouse liver (4 ng) run on an SDS/PAGE gel was blotted to nitro- from mouse colon; 2 and 3, 50 ,ug of total BALB/c mouse colon cellulose and probed with MHV-A59 in the virus blot assay (lane 1) homogenate protein. Lanes 1 and 2 were probed with MHV-A59 in or was immunoblotted with anti-receptor mAb CC1 (lane 2) or with the virus blot assay (3). Lanes 3 and 4 were probed with anti-receptor a 1:100 dilution of anti-CEA antiserum (lane 3, DAKO, Carpinteria, mAb CCM, rabbit anti-mouse IgG, and '251-labeled staphylococcal CA). protein A. expresses a receptor-related glycoprotein, which is 5-10 kDa entially in the colon. In addition, the MHV receptor was smaller than the 120-kDa MHV receptor from BALB/c expressed on intestine and liver as are human biliary glyco- intestine and does not bind MHV or anti-receptor mAb CC1 protein BGP (28) and rat ecto-ATPase (18, 27). but does bind anti-NTR (4). This SJL/J homolog ofthe MHV Antigenic cross-reactivity between CEA, NCA, and the receptor was recognized by polyclonal antisera to human MHV receptor was found when rabbit anti-CEA antiserum, CEA (Fig. 5). which recognizes both CEA and NCA, bound to the immu- noaffinity-purified MHV receptor from Swiss Webster DISCUSSION mouse liver (Fig. 4). The purified MHV receptor was also recognized by antiserum to human PSG2 but not by several The glycoprotein receptor for MHV was identified as a rabbit antisera raised against irrelevant antigens (data not murine member of the CEA family of glycoproteins in the shown). immunoglobulin superfamily by strong sequence homology MHV receptor from BALB/c mouse small intestine was of the first 25 amino acids at the N terminus of the proteins, also tested for cross-reactivity with anti-CEA antisera (Fig. common biochemical properties, and, serological cross- 5). In immunoblots with anti-receptor mAb CCI or anti-NTR, reactivity. The MHV receptor was expressed in a tissue- the MHV receptor in BALB/c intestinal brush border mem- dependent manner similar to that of several CEA family branes was the same size as the murine glycoprotein recog- members, including human BGP, NCA, and rat ecto-ATPase nized by the anti-CEA antisera. The 45- to 58-kDa BALB/c (14, 18, 25, 26), with highest levels ofexpression in the colon, intestinal glycoprotein recognized by anti-receptor mAb CC1 small intestine, and liver. These tissues are major targets for did not bind the anti-CEA antisera. SJL/J small intestine MHV replication in vivo. Thus, the tissue specificity of mouse coronavirus infection may be determined, at least in 1001 * BALB c 1 2 3 4 5 6 7 8 -m BALB.c unlabeled Mab CCI competition :80- 1 ~~~SJLJ! El SJL J unlabeled Mab CC0 competition 2200-

m 601 o 1000-

~0 68

2 0 43

Liver S. Jnt. Colon Kidney Brain S. C. Lung Spleen 26 Tissue S B S B S B S B FIG. 3. Distribution of the MHV receptor in MHV-susceptible BALB/c and MHV-resistant SJL/J mouse tissues. Expression ofthe FIG. 5. Recognition by anti-CEA antisera of the MHV receptor MHV receptor glycoprotein was measured by binding of radioiodi- glycoprotein from BALB/c mice and the variant glycoprotein from nated anti-receptor mAb CCl in the tissue binding assay. Radioac- MHV-resistant SJL/J mice. Intestinal brush border membranes from tivity bound to 50 ,g ofprotein from BALB/c or SJL/J mouse tissues adult female BALB/c (lanes B) and SJL/J (lanes S) mice were was expressed as nanograms of labeled antibody. Data are averages immunoblotted with anti-receptor mAb CC1 (lanes 1 and 2) or with ofduplicates. To measure nonspecific binding ofradioiodinated mAb anti-CEA antisera (lanes 3 and 4, from DAKO, Carpinteria, CA; and CC1 (indicated as unlabeled mAb CC1 competition) tissues were lanes 5 and 6, from Accurate Chemicals, Westbury, NY). Antibody preincubated with excess unlabeled mAb CC1 before binding of the to the N-terminal peptide of the MHV receptor was used to probe radiolabeled antibody. S. int., small intestine; S. C., spinal cord. lanes 7 and 8. Downloaded by guest on September 26, 2021 5536 Medical Sciences: Williams et al. Proc. Natl. Acad Sci. USA 88 (1991) part, by tissue-dependent levels of expression of the CEA- 7. Barthold, S. W., Beck, D. S. & Smith, A. L. (1986) Arch. related receptor glycoprotein. Virol. 91, 247-256. Although CEA-related glycoproteins are found in humans, 8. Smith, M. S., Click, R. E. & Plagemann, P. G. (1984) J. rats, and MHV binds to intestinal brush border Immunol. 133, 428-432. mice, only 9. Knobler, R. L., Linthicum, D. S. & Cohn, M. (1985) J. Neu- membranes from mice and not to membranes from these roimmunol. 8, 15-28. other species (29). Therefore, we hypothesize that MHV 10. Williams, R. K., Kelly, P. T. & Akeson, R. A. (1985) Dev. recognizes a mouse-specific region of a murine CEA-related Brain Res. 19, 253-266. glycoprotein. 11. Matsudaira, P. (1987) J. Biol. Chem. 262, 10035-10038. The SJL/J mouse, which is resistant to MHV infection 12. Laemmli, U. K. (1970) Nature (London) 227, 680-685. when compared with susceptible strains such as BALB/c or 13. Towbin, H., Staehelin, T. & Gordon, J. (1979) Proc. Nati. Swiss Webster, expresses a glycoprotein that is homologous Acad. Sci. USA 76, 4350-4353. to the MHV from mice but 14. Yachi, A. & Shively, J. E., eds. (1989) The Carcinoembryonic receptor susceptible that lacks Antigen Gene Family (Elsevier, Amsterdam). virus-binding activity (3, 4). This SJL/J variant of the MHV 15. Beauchemin, N., Turbide, C., Huang, J. Q., Benchimol, S., receptor glycoprotein is slightly smaller than its BALB/c or Jothy, S., Shirota, K., Fuks, A. & Stanners, C. P. (1989) in The Swiss Webster counterparts. The SJL/J variant was not Carcinoembryonic Antigen Gene Family, eds. Yachi, A. & recognized by anti-receptor mAb CC1 but cross-reacted with Shively, J. E. (Elsevier, Amsterdam), pp. 49-64. antibody to the N-terminal peptide of the Swiss Webster 16. Devereux, J. (1984) Nucleic Acids Res. 12, 387-395. MHV receptor (4) and with antisera to human CEA and 17. Kodelja, V., Lucas, K., Barnert, S., von Kleist, S., Thompson, NCA. the be an J. A. &Zimmermann, W. (1989)J. Biol. Chem. 264, 6906-6912. Therefore, SJL/J homolog may allelic 18. Lin, S. H. (1989) J. Biol. Chem. 264, 14403-14407. variant of mouse CEA-related glycoprotein, with a mutation 19. Rooney, B. C., Home, C. H. & Hardman, N. (1988) Gene 71, affecting the domain(s) of the glycoprotein that normally 439-449. would bind MHV and mAb CCL. 20. Streydio, C., Lacka, K., Swillens, S. & Vassart, G. (1988) CEA family members are large membrane glycoproteins Biochem. Biophys. Res. Commun. 154, 130-137. that may have several cellular functions associated with their 21. Tawaragi, Y., Oikawa, S., Matsuoka, Y., Kosaki, G. & Na- repeating immunoglobulin-like domains (21, 30, 31). CEA kazato, H. (1988) Biochem. Biophys. Res. Commun. 150, shares homology with the neural molecule, 89-96. N-CAM 22. Neumaier, M., Zimmermann, W., Shively, L., Hinoda, Y., (30, 32), and induces calcium-independent cell ag- Riggs, A. D. & Shively, J. E. (1988) J. Biol. Chem. 263, gregation in vitro (33). Murine mmCGM2, a 42-kDa glyco- 3202-3207. protein with an N-terminal amino acid sequence identical to 23. Hinoda, Y., Neumaier, M., Hefta, S. A., Drzeniek, Z., Wa- the MHV receptor protein, induces calcium-dependent cell gener, C., Shively, L., Hefta, L. J., Shively, J. E. & Paxton, aggregation (34). R. J. (1988) Proc. Natl. Acad. Sci. USA 85, 6959-6963. In addition to the CEA-related receptor for murine coro- 24. Paxton, R. J., Mooser, G., Pande, H., Lee, T. D. & Shively, navirus that we demonstrated here, several other members of J. E. (1987) Proc. NatI. Acad. Sci. USA 84, 920-924. the 25. Gold, P. & Freedman, S. 0. (1965) J. Exp. Med. 121, 439-462. immunoglobulin superfamily of glycoproteins have been 26. Svenberg, T., Hammerstrom, S. & Hedin, A. (1979) Mol. shown to serve as virus receptors, including CD4, ICAM-1, Immunol. 16, 245-252. and PVR, the receptors for HIV, human rhinoviruses, and 27. Lin, S. H. & Guidotti, G. (1989) J. Biol. Chem. 264, 14408- poliovirus, respectively (35-40). These immunoglobulin su- 14414. perfamily molecules have several common features which 28. Barnett, T. R., Kretschmer, A., Austen, D. A., Goebel, S. J., may make them particularly attractive targets for virus at- Hart, J. T., Elting, J. J. & Kamarck, M. E. (1989) J. Cell Biol. tachment and penetration: they show significant polymor- 108, 267-276. phism in a common tertiary structure (41), function in inter- 29. Compton, S. R. (1988) Doctoral dissertation (The Uniformed cellular adhesion and and Services Univ. of the Health Sciences, Bethesda, MD). (CD4 I-CAM-l), show tissue- 30. Beauchemin, N., Benchimol, S., Cournoyer, D., Fuks, A. & dependent expression on the plasma membrane. Stanners, C. P. (1987) Mol. Cell. Biol. 7, 3221-3230. 31. Williams, A. F. (1987) Immunol. Today 8, 298-303. The authors gratefully acknowledge the excellent technical assis- 32. Thompson, J. A., Pande, H., Paxton, R. J., Shively, L., tance of Christine B. Cardellichio and Patrick Elia and helpful Padma, A., Simmer, R. L., Todd, C. W., Riggs, A. D. & discussions with Drs. Carl Dieffenbach, Michael Pensiero, John Shively, J. E. (1987) Proc. Natl. Acad. Sci. USA 84, 2965- Hay, and Gabriela Dveksler ofthe Uniformed Services University of 2969. the Health Sciences; Dr. J. Schlom of the National Cancer Institute, 33. Benchimol, S., Fuks, A., Jothy, S., Beauchemin, N., Shirota, Bethesda, MD; and Dr. S.-H. Lin of the University ofTexas, M. D. K. & Stanners, C. P. (1989) Cell 57, 327-334. Anderson Cancer Center, Houston, TX. This study was supported in 34. Turbide, C., Rojas, M., Stanners, C. P. & Beauchemin, N. part by National Institutes of Health Grants Al 25231 and AI 26075. (1991) J. Biol. Chem. 266, 309-315. R.K.W. was a recipient of National Research Service Award 1 F32 35. Dalgleish, A. G., Beverly, P., Clapham, P. R., Crawford, A107810 from the National Institute of Allergy and Infectious Dis- D. H., Greaves, M. F. & Weiss, R. A. (1984) Nature (London) eases. 312, 763-767. 36. Klatzmann, D., Champagne, E., Chamaret, S., Gruest, J., 1. Wege, H., Siddell, S. & ter Meulen, V. (1982) Curr. Top. Guetard, D., Hercend, T., Gluckman, J.-C. & Montagnier, L. Microbiol. Immunol. 99, 165-200. (1984) Nature (London) 312, 767-768. 2. Sturman, L. S. & Holmes, K. V. (1983) Adv. Virus Res. 28, 37. Staunton, D. E., Merluzzi, V. J., Rothlein, R., Barton, R., 35-112. Marlin, S. D. & Springer, T. A. (1989) Cell 56, 849-853. 3. Boyle, J. F., Weismiller, D. G. & Holmes, K. V. (1987) J. 38. Tomassini, J. E., Graham, D., DeWitt, C. M., Lineberger, Virol. 61, 185-189. D. W., Rodkey, J. A. & Colonno, R. J. (1989) Proc. Natl. 4. Williams, R. K., Jiang, G.-S., Snyder, S. W., Frana, M. & Acad. Sci. USA 86, 4907-4911. Holmes, K. V. (1990) J. Virol. 64, 3817-3823. 39. Greve, J. M., Davis, G., Meyer, A. M., Forte, C. P., Yost, 5. Smith, A. L., Cardellichio, C. B., Winograd, D. F., de Souza, S. C., Marlor, C. W., Kamarck, M. E. & McClelland, M. A. M. S., Barthold, S. W. & Holmes, K. V. (1991)J. Infect. Dis., (1989) Cell 56, 839-847. 163, 879-882. 40. Mendelsohn, C. L., Wimmer, E. & Racaniello, V. R. (1989) 6. Stohlman, S. A. & Frelinger, J. A. (1978) Immunogenetics 6, Cell 56, 855-865. 277-281. 41. White, J. M. & Littman, D. R. (1989) Cell 56, 725-728. Downloaded by guest on September 26, 2021