Translocation of a hydrocarbon fluorescent probe between Epstein-Barr virus and lymphoid cells: An assay for

Transcription

1 Pro. Natl. Aad. Si. USA Vol. 75, No. 1, pp , Otober 1978 ell Biology Transloation of a hydroarbon fluoresent probe between Epstein-Barr virus and lymphoid ells: An assay for early events in viral infetion (liposomes/lymphoyte membranes/diphenylhexatriene/fluoresene polarization) KEN S. ROSENTHAL, SAUL YANOVIH, MIHAEL INBAR*, AND JAK L. STROMINGER Sidney Farber aner Institute, Harvard Medial Shool, Boston, Massahusetts 2115 ontributed by Jak L. Strominger, August 7, 1978 ABSTRAT Transloation of the hydroarbon fluoresent probe diphenylhexatriene (DPH) between membranes was studied by fluoresene polarization (P) analysis. First, using a model system, the high P value (.324) of DPH-labeled holesterol/phosphatidylholine liposomes and the low P value (.157) of DPH-Iabele phosphatidylholine liposomes allowed detetion of DPH transloation between interating liposomes. This was monitored by the hange in P in either diretion. Early events during ell-virus interations were similarly studied by monitoring DPH transloation. The P value of DPH-labeled Epstein-Barr Virus (EBV) was signifiantly higher ( ) than the P value of DPH-labele Iymphoid ells ( ). Hene, DPH transloation ould be deteted by hanges in P following inubation of DPH-labeled EBV and nonlabeled ells. A marked derease in P was observed after inubation of DPH-labeled EBV with either nonlabeled lymphoblastoid Raji ells or fresh human B lymphoytes. However, only a slight derease in P was obtained when DPH-labeled EBV was inubated with either nonlabeled fresh human T Ilmphoytes or fresh T or B rabbit lymphoytes. Moreover, inubation of fresh human B lymphoytes with the purified 3 omponent of omplement (a putative inhibitor for the EBV reeptor) prior to the addition of DPH-labeled EBV abolished the observed derease in the P value. Most of these experiments were arried out with both the P3HR-1 and the B95-8 strains of EBV. DPH transloation, as determined by fluoresene polarization analysis, is, therefore, measuring some early event during interation of this enveloped virus and mammalian ells. The potential appliability of this tehnique to other viruses is illustrated by an experiment with Semliki Forest virus. Epstein-Barr Virus (EBV) is an enveloped, DNA-ontaining human herpesvirus, whih is apable of interation with human B lymphoytes in vivo and in vitro (1). Although interations of EBV with some T lymphoblasts have been reported (2), the infetivity and transformation spetra of EBV are limited to B lymphoytes of humans and New World monkeys (3). Suh a limitation is presumably due to the presene of speifi EBV reeptors on B but not on T human lymphoytes. This reeptor is losely assoiated with the reeptor for the 3 omponent of omplement (4, 5). However, further studies on binding, fusion, and infetivity of EBV have been limited by the lak of an assay to determine early biohemial events during ell-virus interations. In this report suh an assay is desribed, as well as its appliation to measure events following interation of EBV with different types of lymphoid ells. The method is based on transloation of the hydroarbon fluoresent probe diphenylhexatriene (DPH) between viral envelopes and ellular membranes as monitored by fluoresene polarization analysis. Fluoresene polarization (P) analysis (6) is, in priniple, an indiret measurement of the thermal mobility of the DPH The publiation osts of this artile were defrayed in part by page harge payment. This artile must therefore be hereby marked "advertisement" in aordane with 18 U. S solely to indiate this fat. moleules inorporated into a lipid bilayer. This is determined to a large extent by the lipid omposition of the membrane (7). The degree of fluidity/rigidity of a membrane ould be envisaged as a mehanial barrier imposed by the lipid bilayer on the degree of rotational mobility of DPH moleules (8). Therefore a slow rotation of DPH in a "rigid" membrane domain is reorded by a high P value, whereas a fast rotation of DPH in a "fluid" membrane region is measured by a low P value (6-8). Based on this assumption, fluoresene polarization analysis of DPH has been extensively used, in the last few years, to analyze the dynami strutural organization of lipids in ellular membranes (9-11) and viral envelopes (12, 13). These studies have independently shown that the P values of DPHlabeled lymphoid ells are markedly lower (14, 15) than those of DPH-labeled enveloped viruses (12, 13). Moreover, it has also been shown that hydrophobi moleules suh as holesterol (16-19) and DPH (2, 21) an be transloated between interating membranes upon ollision of exposed lipid regions. We have therefore measured transloation of DPH between viral envelopes and ellular membranes by monitoring the appropriate hanges in the P values observed after inubation of DPH-labeled enveloped viruses and nonlabeled lymphoid ells. It appears that early events during ell-virus interations suh as binding to speifi reeptors or intermixing of hydrophobi moleules (fusion) an be quantitatively analyzed by DPH fluoresene polarization in a ell-virus mixture. MATERIALS AND METHODS hemials and Reagents. DPH was purhased from Koh and Light; tetrahydrofuran from Aldrih; holesterol from Sigma; egg leithin from Lipid Produts; dextran T-1 and Fioll/Hypaque from Pharmaia; nylon wool from Fenwal; Dulbeo's phosphate-buffered saline (Pi/Nal) and RPMI 164 medium from GIBO; and fetal alf serum from Mirobiologial Assoiates. Purified 3 omponent of omplement was kindly provided by R. Frade. Viruses. EBV produer ell lines P3HR-1 and B95-8 (22) were grown in RPMI 164 medium supplemented with 5% fetal alf serum. Viruses were purified aording to methods previously desribed (23), modified by the use of disontinuous dextran T-1 density gradient entrifugation from 1% to 3% in 5% steps. Purified Semliki Forest virus was kindly provided by A. Helenius and K. Simons (24). All viruses were kept frozen at -7. Abbreviations: P, fluoresene polarization; P25, P at 25 ; DPH, diphenylhexatriene; EBV, Epstein-Barr virus; SFV, Semliki Forest virus; Ptdho, phosphatidylholine; hol, holesterol; Pi/Nal, phosphate-buffered saline. * On leave of absene from the Weizmann Institute of Siene, Rehovot, Israel. 576

2 ell Biology: Rosenthal et al. ells. Raji, a Burkitt lymphoma ell line, and JY, a lymphoblastoid ell line established with EBV, were grown in RPMI 164 medium supplemented with 1% fetal alf serum. ells were used in the experiments 24 hr after the growth medium had been hanged. Fresh mixed populations of T and B lymphoytes were obtained from human and rabbit peripheral blood, by Fioll/Hypaque gradient entrifugation (25). T and B subpopulations were purified aording to methods previously desribed (26). Liposomes. Phosphatidylholine (Ptdho) and holesterol (hol) were kept as stok solutions in hloroform/methanol (2:1 vol/vol.) at -2 at onentrations of 1 and 5 mg/ml, respetively. For preparation of Ptdho liposomes,.1 ml of stok solution was evaporated with nitrogen and resuspended in 1 ml of Pi/Nal; for preparation of hol/ptdho liposomes,.1 ml of Ptdho and.1 ml of hol stok solutions were mixed, evaporated, and resuspended in 1 ml of Pi/Nal. Lipid dispersions were than soniated at 4 with a Branson soniator. Ptdho and hol/ptdho dispersions were soniated for 6 and 12 min, respetively, at maximum energy output (8), and only fresh liposomes were used in the experiments. Fluoresene Labeling with DPH. DPH (2 mm) in tetrahydrofuran was kept at 25 as a stok solution. For labeling of samples, a dispersion of DPH in Pi/Nal (2,gM) was prepared as previously desribed (7). Liposomes, viruses, and ells were labeled with DPH by inubating the samples in a 1-ml DPH/Pi/Nal dispersion for 3 min at 25. Free DPH was removed by washing the DPH-labeled samples with 3 vol of 1 ml Pi/NaI. No DPH was measurable in the final wash, in all ases. Fluoresene Polarization Analysis. The degree of fluoresene polarization (P) of the DPH-labeled liposomes, viruses, and ells was determined with the aid of the Elsint MV-1 mirovisosimeter as desribed (15, 2), and the P value was reorded diretly by the instrument aording to the equation P = [(III - I )/(III + Ii)], in whih III and I1 are the fluoresene intensities polarized parallel and perpendiular to the diretion of polarization of the exitation beam, respetively (6). All fluoresene measurements were arried out at 25, and the auray of the obtained P value was +.5. RESULTS Transloation of DPH upon Mixing of Liposomes. To analyze the possibility that DPH an be transloated between membranes, and that this transloation an be monitored by fluoresene polarization analysis, transloation of DPH was studied in an artifiial membrane model system. Previous studies have shown that inorporation of holesterol moleules into a phospholipid bilayer is assoiated with an inrease in the rigidity of the membrane (27). These hanges in the strutural organization of the membrane an be monitored by the differene in the degree of fluoresene polarization of DPH embedded in liposomes of pure phospholipids and in mixed liposomes of holesterol and phospholipids (28). A marked inrease in the P value results from inorporation of holesterol in leithin liposomes. The P values obtained with DPH-labeled Ptdho liposomes and DPH-labeled hol/ptdho liposomes at 25 were.157 and.324, respetively. On the basis of this large differene in the P values, transloation of DPH between the two liposomes was measured by fluoresene polarization analysis. Experiments summarized in Fig. 1 show a marked derease in the P value after inubation of DPH-labeled hol/ptdho liposomes with nonlabeled Ptdho liposomes and a signifiant inrease in the P value after inubation of DPH-labeled Ptdho liposomes with nonlabeled hol/ptdho liposomes. The results have also shown (Fig. 1) that partitioning.35 a.31 'U N as.27 'U.D U *.23 U in.1 UI..19 Pro. Natl. Aad. Si. USA 75 (1978) _B FIG. 1. Transloation of DPH between liposomes. P25 was determined for hol/ptdho (A) and Ptdho (D) liposomes by using the Elsint mirovisosimeter. Liposomes were prepared, labeled, and washed three times with 1 ml of Pi/Nal to eliminate free DPH. The absene of free DPH was determined by inubation of the wash supernatants with unlabeled liposomes and determination of the total fluoresene intensity after 3 min of inubation at 25. For DPH transloation experiments, 1.5 ml of DPH-labeled Ptdho liposomes was mixed with 1.5 ml of unlabeled hol/ptdho liposomes (B) or 1.5 ml of DPH-labeled hol/ptdho liposomes was mixed with 1.5 ml of unlabeled Ptdho liposomes (). The degree of fluoresene polarization of the mixtures was determined at different time intervals up to 3 min after mixing. of DPH is a rapid proess and that the reation reahes an equilibrium after less than 1 min of inubation at 25. Therefore, lipid-lipid interations during membrane intermixing an be studied by fluoresene polarization analysis of DPH provided that (i) the two interating membranes show a signifiant differene in their P values and (ii) no free DPH is assoiated with the labeled membranes. Transloation of DPH from EBV to Raji ells. Based on the results obtained with the membrane model system, an attempt was made to study ell-virus interation. A large differene in the P values of EBV and lymphoblastoid ells, inluding both EBV-transformed ell lines and nontransformed lymphoid target ells, provided the basis for DPH transloation studies. The envelopes of the B95-8 and P3HR-1 strains of EBV had P values of and.38 ±.4, respetively, while the membranes of the Raji, B95-8, and P3HR-1 ell lines had P values between.171 and.214, depending on the growth phase and serum onentration of the ulture (29). The differene in membrane fluidity of the envelopes of the P3HR-1 and B95-8 strains of EBV may be due to growth onditions of the host ells or inherent differenes in harateristis of the host ells or virus strain. Inubation of DPH-labeled EBV and labeled Raji ells brought about a rapid derease in the P value (Fig. 2). The P value of DPH hanged from that of the virus to a value representative of Raji ells upon mixing of DPH-labeled EBV with unlabeled Raji. The half-time of the hange was 8-1 minutes at 25. This hange in P was due to transloation of DPH from the viral envelope to the ellular plasma membrane. This assumption is supported by the P value of the Raji ell-assoiated DPH after inubation and entrifugation and of the residual DPH-labeled EBV (Table 1). Similar results were obtained with the P3HR-1 and B95-8 strains of EBV, although only the former an superinfet Raji ells (Fig. 2). 3

3 578 ell Biology: Rosenthal et al. WI n% -.4 A B ',_ UL FIG. 2. Transloation of DPH from EBV to Raji. EBV was labeled with DPH by inubation of the purified virus with a PJ/Nal/ DPH dispersion for 3 zhin at 25. The labeled virus was washed three times with Pi/Nal; under these onditions all the fluoresene signal was assoiated with the virus envelope and no free DPH ould be deteted in the wash supernatants. A sample of Raji ells (5 X 16) was washed twie with P;/Nal, resuspended in 3 ml of Pi/Nal, and allowed to equilibrate to 25 in the Elsint mirovisosimeter. An aliquot (.1 ml) of the DPH-labeled P3HR-1 (A) or B95-8 (B) strain of EBV, orresponding to.5 to 1. liters of onentrated supernatant, was then added to the nonlabeled ells and the P25 values of the mixtures were reorded. Transloation of DPH from EBV to Lymphoid ells. To test the possibility that transloation of DPH from EBV to lymphoid ells an be related to the infetivity spetrum of EBV, DPH-labeled EBV was inubated with nonlabeled T and B subpopulations of fresh human lymphoytes. These experiments were possible due to the following observations: (i) the P value of DPH-labeled EBV was found to be muh higher that that of fresh human lymphoytes and (ii) the P values of DPH-labeled T and B subpopulations of fresh human lymphoytes.were very similar (Table 2). Results presented in Fig. 3 show a marked derease in the P value of DPH-labeled EBV after inubation with unlabeled human B lymphoytes for 3 min at 25. These results were observed for both the B95-8 and the PSHR-1 strains of EBV even though the P3HR-1 strain of EBV is not apable of transforming B lymphoytes in vitro (22). In ontrast, only a slight derease in the P value was observed when DPH-labeled EBV was inubated with unlabeled human T lymphoytes (Fig. 3 A and B). Results similar to those obtained with human T lymphoytes were also obtained when DPH-labeled EBV was inubated with unlabeled T and B subpopulations of rabbit lymphoytes (Fig. 3). On the basis of these observations it an be onluded that DPH transloation ours at a muh higher effiieny during speifi interations of EBV with either Raji ells or fresh human B lymphoytes than during nonspeifi interations of EBV with human T lymphoytes or rabbit T and B lymphoytes. Transloation of DPH upon mixing of dye-labeled EBV with human T lymphoytes or rabbit B and T lymphoytes ould be the result of (i) nonspeifi interations of the virus and ell leading to transient lipid bilayer ontat; (ii) the presene of some DPH in a form that is readily exhanged nonspeifially Table 1. Superinfetion of Raji ells with EBV* EBV Raji Raji ell Superstrain Virus ells + EBVt pellett natant P3HR P3HR ND ND B * All data are expressed as P25. ND, not determined. t Values are shown after 6-min inubations for P3HR-1 and 3-min inubation for B95-8 virus. After the inubation period, ells were pelleted at 5 X g for 1 min and resuspended to the original volume in Pi/Nal. Table 2. Pro. Natl. Aad. Si. USA 75 (1978) EBV interation with lymphoytes* EBV + ell Super- Exp. Virus ells ells pellett natant Human B ells with P3HR-1 or B95-8 virust Human T ells with P3HR-1 or B95-8 virus$ Rabbit B ells with B95-8 virus Rabbit T ells with B95-8 virus * All data are expressed as P25. t After 3 min, ells were pelleted at 5 X g for 5 min and resuspended to the original volume in PJ/Nal. Human lymphoytes were separated on a B ell immunoabsorbent olumn (26). T ells for exp. 2 were further purified by passage over a nylon wool olumn. Exps. 1 and 2 were arried out with P3HR- 1 virus and exp. 3 with B95-8 virus. Rabbit lymphoytes were separated on a nylon wool olumn. with a variety of ells-e.g., present in defetive virus partiles; (iii) the presene of DPH in ontaminating ellular membranes that non-speifially interat with ells; or (iv) in the ase of human lymphoytes, ontamination of the T ell subpopulation by a small fration of Ig-negative B lymphoytes that have EBV reeptors. It is therefore suggested that in order to alulate the speifi DPH transloation, the amount of DPH transloation between DPH-labeled EBV and human T lymphoytes an be subtrated from the amount of DPH transloation between DPH-labeled EBV and human B lymphoytes. Inhibition of DPH Transloation from EBV to B ells by the 3 omponent of omplement. The speifiity of DPH transloation upon fusion of EBV and human B lymphoytes was further analyzed by ompetition experiments with the S omponent of omplement. Reent reports (4, 5) have suggested that the omplement reeptors on the ell surfae of human B lymphoytes are assoiated with speifi binding of EBV partiles, suggesting that the omplement reeptor is part of the EBV reeptor. Therefore, the effet of the purified 3 omponent of omplement on DPH transloation was studied. Experiments summarized in Fig. 3D show, as expeted, a marked derease in the P value following inubation of DPH-labeled EBV and human B lymphoytes. However, the signifiant derease in the P value was markedly inhibited by a prior inubation of the B lymphoytes with purified 3 (Fig. 3D). This is indiative of a ompetition between EBV and 3 for the same reeptors on the ell surfae of the B-lymphoytes. Note that the derease in the P value obtained after inubation Table 3. SFV interation with JY ells* SFV Exp. virust JY ellst SFV + JY * All data are expressed as P25- t SFV labeled with DPH and washed free of DPH. JY ells labeled with DPH for 3 min at 25. SFV + JY omplex after 3 min of inubation of DPH-labeled SFV and nonlabeled JY ells.

4 ell Biology: Rosenthal et al. Pro. Natl. Aad. Si. USA 75 (1978) A,, N...34 U W ) ED Io.32[.3 T B T FIG. 3. Transloation of DPH from EBV to human and rabbit lymphoytes. EBV was labeled with DPH as desribed in the legend to Fig. 2. Subpopulations of normal human B and T lymphoytes were isolated from the peripheral blood of normal donors by Fioll/Hypaque density gradient entrifugation followed by a B ell immunoabsorbent olumn. Rabbit B and T lymphoyte subpopulations were separated on a nylon wool olumn. The lymphoytes were washed twie in Pi/Nal, and 4 x 16 ells in 3 ml of Pi/Nal were allowed to equilibrate to 25 in the mirovisosimeter. A.1-ml aliquot of DPH-labeled EBV, orresponding to.5 to 1. liter of onentrated supernatant, was added to the nonlabeled ells and the P25 value was reorded at different time intervals. (A) Addition of DPH-labeled P3HR-1 strain of EBV to human B (@) and T (o3) lymphoytes. (B) Addition of DPH-labeled B95-8 strain of EBV to human B () and T (3) lymphoytes. () Addition of DPH-labeled B95-8 strain of EBV to rabbit B () and T (3) lymphoytes. (D) Inhibition of EBV interation with human B lymphoytes by the 3 omponent of omplement. Two aliquots (4 X 16) of human B lymphoytes were suspended in 3 ml eah of Pi/Nal. Purified 3 (3,ug) was added to one sample and the two aliquots, were inubated for 3 min at 37. An aliquot (.1 ml) of DPH-labeled B95-8 was added to eah sample and the hanges in the P2Z value were monitored with the mirovisosimeter. () B ells with 3; () B ells without 3. Transloation of DPH from B95-8 to T ells of the same donor is shown in B. of DPH-labeled EBV and human B lymphoytes preinubated with 3 (Fig. 3D) is similar to the derease in the P value obtained when DPH-labeled EBV was inubated with human T lymphoytes (Fig. 3 A and B). If indeed the DPH transloation obtained with human T lymphoytes represents a nonspeifi interation, then S ompletely abolished the speifi DPH transloation between EBV and human B lymphoytes. DPH Transloation from Semliki Forest Virus to JY ells. To see if DPH transloation an also be used to study interations of other enveloped viruses and target ells, an experiment was arried out with Semliki Forest virus (SFV) and the JY human lymphoblastoid ell line to whih SFV is known to bind (3). The results summarized in Table 3 learly indiate that the P values of DPH-labeled SFV are signifiantly higher ( ) than those of the DPH-labeled JY ells ( ). Inubation of DPH-labeled SFV with nonlabeled JY ells resulted in a marked derease in the P value, indiative of DPH transloation from SFV envelopes to JY ellular membranes. DISUSSION Interation of enveloped viruses and their target ells requires a speifi virus-reeptor interation, presumably a proteinprotein interation. Suh an interation brings the viral envelope and the target ell plasma membrane into lose proximity, perhaps allowing lipid-lipid interations followed by either fusion of the viral and ellular membranes or viropexis (31). In the ase of EBV, fusion of the viral envelope and the plasma membrane has been observed by eletron mirosopy (32). Inorporation of DPH in the viral envelope provides an assay for virus-ell interation upon transloation of the dye during ollision and intermixing of the ellular and viral membrane lipids. Aqueous transfer of the dye is unlikely due to the hydrophobiity of the ompound. Measurement of the DPH transfer is made possible by differenes in the fluoresene polarization of DPH inorporated into viral and ellular membranes. Transloation of DPH between membranes an be analyzed by hanges in the P value after o-inubation of virus and ells provided that (i) the initial P values of the membranes of the two systems are signifiantly different and (ii) that only one membrane is labeled with DPH. The envelopes of all viruses tested, inluding vesiular stomatitis virus (12, 13), Sindbis virus, (12) SFV (12), and both the P3HR-1 and B95-8 strains of EBV (this paper) have high P values indiative of rigid lipid domains. This is in ontrast to the muh more fluid membranes of their host and target ells. The feasibility of measuring interations of lipid domains by DPH transloation was illustrated by using liposomes of different P values. Upon mixing of DPH-labeled hol/ptdho liposomes (high P) with Ptdho liposomes (lower P), a rapid derease in polarization for the system was observed. Suh a hange ould only be due to the transloation of the fluoresent probe from a rigid to a more fluid membrane environment. In the liposomal model system, interation and possible fusion of liposomes takes plae readily upon mixing, allowing transloation of hydrophobi moleules suh as DPH. However, the transfer of hydrophobi moleules from the viral envelope to the plasma membrane of the ell, upon interation of lipid regions, an be prevented by ell surfae glyoproteins (17). The transfer of DPH from the lipid environment of an enveloped virus to its target ell most likely requires a speifi virus-reeptor interation either to bring the lipid phases into juxtaposition or to fuse them. The large derease in P upon interation of DPH-labeled EBV or SFV with its natural target ells indiated that a signifiant DPH exhange ourred. Suh an exhange was not observed upon inubation of EBV with ells that are not infetable by EBV, suh as human T ells or rabbit B and T ells. Nonspeifi interations of EBV, as with T ells or rabbit ells, does not allow a signifiant DPH transfer to our. The requirement for a speifi virus-ell reeptor interation to preede DPH transloation was supported by the inhibition of DPH transfer from EBV to human B ells by the purified 3 omponent of omplement. ompetition by 3 for binding of EBV to its reeptor prevented the lose ontat of lipid domains neessary for DPH transfer from the viral envelope to the target ell membrane. The omplement reeptor is either idential with or losely related to the EBV reeptor (2, 4, 5).

5 58 ell Biology: Rosenthal et al. The speifi interation of EBV with its reeptor was not strain speifi. Although the P3HR-1 strain of EBV is apable of superinfetion of Raji but does not transform peripheral B lymphoytes and the B95-8 strain of EBV transforms B lymphoytes but does not superinfet Raji, interation of both the B95-8 and P3HR-1 strains of EBV with both Raji and peripheral human B ells ould be observed by DPH transloation. Hene, the biologial distintion between the B95-8 and P3HR-1 strain of EBV does not seem to be the result of some phenomenon during interation of the virus and the ell reeptor. However, as an be seen in Figs. 2 and 3, the rate of transloation of DPH from the B95-8 strain of EBV to the target ell may be faster than for the P3HR-1 strain of EBV. Although the mehanism of DPH transloation is not learly understood, an experiment arried out with SFV orroborated the findings with EBV. Fluoresene polarization measurements of DPH transloation an serve as a method to study the early events of interation of ells with enveloped viruses. At this time, it annot be determined whether interation of viral and ellular membranes at a virus reeptor site is suffiient for DPH transloation or whether DPH transfer requires subsequent fusion of the ellular plasma membrane and viral envelope. The problem might be approahed by studying viral mutants that bind to ells but are not apable of infetion. This tehnique an also be used as an assay to study the properties of viral reeptors, as is illustrated in the ase of EBV by inhibition of DPH transloation with 3. Fluoresene polarization measurement of DPH transloation an also be used to study the lipid interations or fusion of other membrane systems differing in lipid fluidity, suh as interations of liposomes with ells, lipoproteins with ells, or ell to ell fusion. This study was supported in part by researh Grants A 2182 and A from the National Institutes of Health and by a fellowship to K.S.R. from the Amerian aner Soiety (PF 1369). 1. Epstein, M. A. & Ahong, B. G. (1977) Annu. Rev. Mirobiol. 31, Menezes, J., Seigneurin, J. M., Patel, P., Bourkos, A. & Lenoir, G. (1977) J. Virol. 22, Jondal, M. & Klein, G. (1973) J. Exp. Med. 138, Yefenof, E. & Klein, G. (1977) Int. J. aner 2, Yefenof, E., Klein, G., Jondal, M. & Oldstone, M. B. A. (1976) Int. J. aner 17, Weber, G. (1971) J. hem. Phys. 55, Shinitsky, M. & Inbar, M. (1974) J. Mol. Biol. 85, Pro. Nati. Aad. Si. USA 75 (1978) 8. Shinitsky, M. & Inbar, M. (1976) Biohim. Biophys. Ata 422, Fuhs, P., Parola, A., Robbins, P. W. & Blout, E. R. (1975) Pro. Nati. Aad. Si. USA 72, Inbar, M., Yuli, I. & Raz, A. (1977) Exp. ell Res. 15, de Laat, S. W., van der Saag, P. T. & Shinitzky, M. (1977) Pro. Natl. Aad. Si. USA 74, Moore, N. F., Barenholz, Y. & Wagner, R. R. (1976) J. Virol. 19, Barenholz, Y., Moore, N. F. & Wagner, R. R. (1976) Biohemistry 15, Inbar, M. & Ben Basset, H. (1976) Int. J. aner 18, Petitotou, M., Tuy, R., Rosenfeld,., Mishal, Z., Paintrand, M., Jasnin,., Mathe, G. & Inbar, M. (1978) Pro. Natl. Aad. Si. USA 75, Poznansky, M. & Lange, Y. (1978) Biohim. Biophys. Ata 56, Lenard, J. & Rothman, J. E. (1976) Pro. Nati. Aad. Si. USA 73, Inbar, M. & Shinitsky, M. (1974) Pro. Nati. Aad. Si. USA 71, Hagerman, J. S. & Gould, R. G. (1951) Pro. So. Exp. Biol. Med. 78, Berke, G., Tzur, R. & Inbar, M. (1978) J. Immunol. 12, ollard, J. G., de Wildt, A. & Inbar, M. (1978) FEBS Lett. 9, Miller, G., Robinson, J., Heston, L. & Lipman, M. (1974) Pro. Natl. Aad. Si. USA 71, Dolyniuk, M., Prithett, R. & Kieff, E. (1976) J. Virol. 17, Helenius, A. & Sbderlund, H. (1973) Biohim. Biophys. Ata 37, Bp6yum, A. (1968) Sand. J. lin. Lab. Invest. 21, Suppl hess, L. & Shlossman, S. F. (1976) in In Vitro Methods in ell Mediated and Tumor Immunity, eds. Bloom, G. & David, J. (Aademi, New York), Vol Yeagle, P. L., Martin, R. B., Lala, A. K., Lin, H. K. & Bloh, K. (1977) Pro. Natl. Aad. Si. USA 74, Lala, A. K., Lin, H. K. & Bloh, K. (1978) Bioorg. hem., in press. 29. Pessin, J. E., Salter, D. W. & Glaser, M. (1978) Biohemistry 17, Helenius, A., Morein, B., Fries, E., Simons, K., Robinson, P., Shirrmaher, V., Terhorst,. & Strominger, J. L. (1978) Pro. Natl. Aad. Si. USA 75, Dales, S. (1973) Bateriol. Rev. 37, Seigneurin, J. M., Vuillaume, M., Lenoir, G. & de The, G. (1977) J. Virol. 24,