ARCHIVES
Vol.
OF BIOCHEMISTRY
189, No. 1, July,
AND
pp. 161-171,
BIOPHYSICS
I978
Cellular Membrane Receptors for Oncovirus Envelope Glycoprotein: Properties of the Binding Reaction and Influence of Different Reagents on the Substrate and the Receptors SUBAL Department
BISHAYEE,’
of Pharmacology
METTE
STRAND,
and Experimental Therapeutics, Medicine, Baltimore, Maryland Received
November
14, 1977; revised
J. THOMAS
AND
The Johns 21205 March
Hopkins
AUGUST” University
School
of
14, 1978
Binding of purified envelope glycoprotein (gp69/71) of Rauscher murine type-C oncovirus to cellular membrane receptors has been analyzed with reaction systems using intact cells or membranes of disrupted cells. The reaction was highly specific; only cells permissive to infection by Rauscher virus bound the ‘““I-labeled viral glycoprotein. The specificity of binding was also demonstrated with respect to virus interference; cells productively infected with murine ecotropic type-C virus failed to bind the virus envelope glycoprotein whereas permissive cells infected with murine xenotropic virus continued to bind the Rauscher ecotropic virus glycoprotein. The reaction required the presence of Ca”, and was rapid and reversible with an equilibrium constant of about 8.3 X 10” Mm’ and an association rate constant of about 6.0 X 10” Mm’Sm’ at 24’C. The cellular receptors were saturated by about 4 x 10” molecules of glycoprotein per NIH 3T3 cell. No degradation or modification of bound glycoprotein was detected. The binding activity of the purified glycoprotein was susceptible to treatment with urea or guanidine-HCl; modification of tyrosyl, histidyl or tryptophan residues or the reduction of disulfide bonds also blocked the reaction. Sialic acid of the glycoprotein appeared not to be required. The receptor activity of a crude membrane fraction of disruated cells was abolished by treatment with trypsin, a-chymotrypsin, or phospholipase C.
Infection of a host cell by oncoviruses is known to involve a specific interaction between an envelope component of the virion and a receptor component of the cell membrane. The role of the envelope in this interaction was recognized chiefly by the correlation between host range, interference and neutralization properties within subgroups of avian oncoviruses, and by the observations that these properties were modified by phenotypic mixing and the formation of vii-ion pseudotypes (l-3). As proposed by Rubin (2) and Vogt and Ishizaki (3), the most obvious function of the envelope occurs at some early step of infection, during absorption or penetration. Early biological studies with the avian system by ’ Present address: Indian Medicine, 4, Raja Subodh 700032, India. ’ To whom correspondence
Institute Mullick should
Steck and Rubin (5, 6) and Piraino (7) suggested that the initial attachment of the virion to the cell may be random and that the specificity of infection is manifest at some later step in infection, such as the binding of specific receptors in the cell membrane or the penetration of the virus. These results were inconclusive, however, as it was found that specific interference by the infecting virus could be established under conditions whereby the virus particles responsible for interference appeared to remain at the cell surface (5). Progress in studies of the viral envelope has been achieved by the purification and characterization of an envelope glycoprotein of about 70,000 daltons from several murine viruses (8-10). This glycroprotein is the major constituent of the virion surface (11) and antibody prepared against it will neutralize the virus (12-14). Several studies have shown that the isolated glycoprotein
of Experimental Road, Calcutta be directed. 161
0003-9861/78/1891-0161$0200/O Copyright 0 1978 by Academic Press, Inc. All rights of reproduction in any form reserved.
162
BISHAYEE,
STRAND,
will react with cellular receptors. Hunsmann et al. (12) and Schafer et al. (15), using purified glycoprotein of Friend leukemia virus, and Hung et al. (16), using crude avian virus glycoprotein, demonstrated that these preparations added to normally permissive cells interfered with infection by the corresponding virus. Recently, a major advance in these studies was achieved by DeLarco and Todero (17) who developed an in vitro binding assay for virus envelope glycoprotein and carried out an initial characterization of the binding of glycoprotein to cells in tissue culture. These findings showed that NIH 3T3 cells bound approximately 5 X lo” molecules of glycoprotein per cell and that the ecotropic virus envelope bound specifically to mouse cells. In the present studies, we further characterize the binding of purified envelope glycoprotein (gp69/71) of Rauscher murine virus to intact cells and cell membranes. The findings agree closely with those of DeLarco and Todaro (17) and extend these studies to the kinetic properties of the reaction and to analysis of some of the properties of the substrate, receptor and reaction product. MATERIALS
AND
METHODS
Envelope glycoprotein. The envelope glycoprotein (gp69/71) of Rauscher murine virus was purified by phosphocellulose column chromatography as described (18). The purified protein was iodinated with Na’? (Amersham/Searle) according to the method of Hunter (19) as previously described (20). “‘I-labeled gp69/71 had a specific activity of 40 to 60 $i/pg. Cells. All cells were routinely propagated in monolayers at 37°C in Dulbecco’s modification of Eagle medium supplemented with 10% heat inactivated fetal bovine serum (Flow Laboratories, Rockville, Maryland) and equilibrated in air with 5% CO*. For binding studies, the cells were transferred to 16-mm disposable multi-dish trays (Costar), and propagated for 24-30 h prior to assay. NIH 3T3 cells (21), BALB/c 3T3 cells (22), a clonal line of wild mouse embryo cells (SC-I) (23), mink lung cells (24), rabbit cornea1 cells (SIRC) (25), and human fetal muscle-skin cells (Hu 8085) (26) were kindly provided by Dr. A. Hackett. SC-1 cells infected with AKR-Ll (27) or BALB/c mouse xenotropic virus (28) were kindly provided by Dr. J. Hartley. SC-1 cells and NIH 3T3 cells infected with Rauscher leukemia virus (29) were gifta from Dr. J. A. Bilello. Swiss 3T3FL cells and D56 S’L-, a cloned transformed derivative of this cell infected by Moloney sarcoma virus (30)
AND
AUGUST
were obtained from Dr. R. D. Bassin. NIH 3T3 cells infected with the endogenous BALB/c N-tropic virus (N-Cl-35) and BALB/c 3T3 cells infected with Btropic virus (B-Cl-11) (31) were provided by Dr. P. Besmer. Kirsten murine sarcoma virus (32) transformed nonproducer BALB/c 3T3 cells, BALB/c (KiMSV) (33) were provided by Dr. C. Croce. NIH 3T3 cells infected with Gross (34) type-C virus were described previously (35). Binding of purified envelope glycoprotein in NIH 3T3 cells. Binding of ““I-labeled gp69/71 to NIH 3T3 mouse cells in tissue culture was measured with an assay system using confluent monolayers of viable cells propagated for 24-30 h in 16-mm multi-dish trays. The cells were rinsed twice at 24°C with l.O-ml portions of the reaction buffer consisting of 116 mM NaCI, 53 mM KCl, 25 mM NaHCO:%, 55 mM glucose, 0.8 mM MgCL, 10 mM CaCb, 50 mu N,N-bis(2-hydroxyethyl)2-amino ethanesulfonic acid (BES), pH 7.4, and 1.0 mg/ml of bovine serum albumin. ““I-labeled gp69/71 (6-8 x lo4 cpm/ng), l-3 ng in 0.2 ml of reaction burfer, was added and the cells were incubated at 24°C for the desired time (30-120 min). The supernatant solution was then removed and the monolayers were rinsed 3 times with 3.0.ml portions of ice-cold reaction buffer. The cells were dispersed with 0.5 ml of 0.2 N NaOH for 15-20 min at 70°C and the solution was transferred with two 0.5-ml aliquots of 0.2 N HCl to glass tubes for counting (36). The background or nonspecific binding initially was determined either by preincubating the cells for 30 min with a 700-fold excess of unlabeled gp69/71 under standard assay conditions, or by the radioactivity bound to mink cells. All assays were carried out in duplicate or triplicate. Specific binding was determined by subtracting the background binding from the total radioactive uptake. Preparation of membranes. Cell monolayers were rinsed twice with 20 mu Tris-HCI, pH 7.4 containing 0.15 M NaCl, removed from the surface of the T-150 flasks (Costar) with the help of a rubber policeman rod, suspended in the above buffer, and then centrifuged at 4°C. The packed cells were washed with the same buffer, suspended in 2.0 ml of 20 mM Tris-HCl, pH 7.4 (1.25 x lo7 cells/ml), homogenized with 25 strokes using a tightly fitted Dounce homogenizer, and centrifuged at 23,000g for 30 min. The pellet was suspended in 2.0 ml of the above buffer and centrifuged at 1,OOOg for 30 s to remove the nuclei. Bovine serum albumin (10 mg/ml) was added to the supernatant membrane fraction which was then frozen at -20°C until use. Threefold freezing and thawing of this crude membrane preparation did not result in the loss of ‘?labeled gp69/71 binding. Nearly 90% of the gp69/71 binding activity of intact cells is present in this crude membrane preparation. Binding of ““I-labeled gp69/71 with isolated membranes. The binding assay with isolated membranes was carried out using Millipore EHWP filters (pore size 0.5 am). The standard reaction mixture, in a total
CELL
RECEPTORS
FOR
ONCOVIRUS
volume of 0.1 ml, containing 50 mu Tris-HCl, pH 7.4, 10 mg/ml bovine serum albumin, 10 mM CaC12, l-5 ng of ‘ZSI-labeled gp69/71 and the membrane fraction, was incubated at 24°C for 1 h. The reaction was terminated by adding 3 ml of ice-cold reaction buffer containing 0.1 mg/ml of bovine serum albumin. The membranes were collected immediately under reduced pressure on EHWP filters, presoaked in fetal bovine serum, and washed with 12-15 ml of buffer solution. The radioactivity retained in the filter was counted. Nonspecific binding was determined by preincubating the membrane fractions with a 700-fold excess of unlabeled gp69/71. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis. NIH 3T3 cells (1.3 x lo”) were incubated with ““I-labeled gp69/71. After washing, the cells were solubihzed with sample buffer containing 62.5 mM Tris-HCI, pH 6.8, 2% sodium dodecyl sulfate, 5% 2mercaptoethanol, and 10% glycerol. The proteins in the supernatant of the incubation mixture were precipitated with 7% trichloroacetic acid and the pellet was washed once with 0.5 ml of cold acetone. The precipitate was solubilized in the sample buffer. The samples were heated at 1oO’C for 1 min and then analyzed by electrophoresis in a 5-20% gradient polyacrylamide gel with an electrode buffer containing 0.1% sodium dodecyl sulfate in 25 mu Tris-192 mM glycine, pH 8.3 (37). Autoradiography of the gel, after fling and drying, was carried out with Kodak No Screen film (Kodak NS54T). Protein determination. Protein was determined according to the method of Lowry et al. (38). Reagents. The 30,000-dalton structural protein of Rauscher murine type-C virus (~30) was purified by ion-exchange column chromatography and Sephadex gel filtration as described elsewhere (18). Fetuin, ceruloplasmin, ovomucoid, soybean trypsin inhibitor, and bovine serum albumin were obtained from Sigma Chemical Company. Phospholipase C from Clostridiumperfringens (94 units/mg), trypsin (199 units/mg), a-chymotrypsin (61 units/mg), neuraminidase from C. perfringens (14 unita/mg), /3-galactosidase (360 units/mg), DNase (2300 units/mg), and RNase (2700 units/mg) were obtained from Worthington Biochemical Company. Goat anti-Rauscher gp69/71 serum was prepared and characterized as described elsewhere (18). All other reagents were of the highest purity available from commercial sources. RESULTS
Properties of gp69/ 71-cell interaction. Binding of lz51-labeled gp69/71 to cells was routinely measured using monolayers of cultured cells washed and incubated with a buffer solution containing NaCl, KCl, MgC12, CaC12, glucose and bovine serum albumin, as described under Materials and Methods. Of these constituents, the reac-
GLYCOPROTEIN
163
tion appeared to be strictly dependent only upon Ca’+; there was a 7-fold increase in the binding at 10 mu Ca2+. Ca2’ could not be replaced by Mg+, and NaCl and KC1 appeared not to be required, as their replacement by isotonic sucrose (0.25 M) had no significant effect on the reaction. The pH optima was broad with 90% of maximum binding observed between pH 6.8 to 8.0. Bovine serum albumin was added to the reaction in order to saturate nonspecific sites of protein binding to cells; its omission had no effect on specific gp69/71 binding but appeared to slightly increase background absorption of substrate. Several other proteins and glycoproteins such as the p30 structural protein of Rauscher virus, fetuin, ceruloplasmin or ovomucoid had no effect upon gp69/71 binding. Of the proteins tested, only unlabeled gp69/71 reduced the binding of ‘251-labeled gp69/71 in a fashion suggesting the chemical similarity of labeled and unlabeled substrates (data not shown). Kinetics of binding. Saturation of cellular binding sites was analyzed using 2.6 x lo4 NIH 3T3 cells and increasing concentrations of 1251-labeled gp69/71. The apparent saturation of receptor sites was achieved by the addition of approximately 30 ng of gp69/71, with the binding of 1.18 ng of gp69/71, or approximately 16.5 fmol (Fig. 1). Thus, the number of receptors per cell was approximately 4 X lo”. A Scatchard
”% ./ : 05 / ? i M’“k__-. I. ,< --&---.--, -I---&--, 20 40 FIG. 1. Binding as a function of “‘I-labeled gp69/71 concentration. NIH 3T3 or mink cells (2.6 x 104) were incubated with different amounts of ‘251-labeled gp69/71 (8 x lo4 cpm/ng) in a total volume of 0.2 ml for 2 h at 24°C. Assay system is described under Materials and Methods. 0, specific binding; A, nonspecific binding.
164
BISHAYEE,
STRAND,
plot (39) of this data, corrected for the active fraction of the substrate as described below, gave an equilibrium association constant, Ka, of 8.3 X 10' M-i, and the number of binding sites as 4.5 x lo5 per cell. Similar values were obtained when the same experiment was carried out by adding unlabeled gp69/71 to a constant amount of ““I-labeled gp69/71. The binding of 12”1-labeled gp69/71 to NIH 3T3 cells was time and temperature dependent; at 24°C the reaction was linear for 20 min and reached a maximum in 2 h (Fig. 2). The reaction was 6 times slower at 4°C (data not shown). Utilizing the secondorder equation Fzi= 2.303/t (P-R) log R (Py)/P (R-y), where P is the gp69/71 concentration, R is the receptor concentration as determined by Scatchard plot, y is the amount of P reacting in time t, and assuming that 20% of labeled gp69/71 was active (see below), the association rate constant was calculated to be 6 x lo5 M-IS-’ at 24’C. Using 3 ng of ‘251-labeled gp69/71, binding of the substrate was proportional to cell concentration below 2.6 x lo4 cells and reached a maximum at 1.3 x lo5 cells (Fig. 3). In this experiment, the maximum extent of glycoprotein binding was approximately 13% of the total gp69/71 added. This limi-
FIG. 2. Time course of the binding of ‘?-labeled gp69/71 with NIH 3T3 cells. NIH 3T3 or mink cells (1.3 x 1W5) were incubated in a total volume of 0.2 ml with 2.1 X IO-‘” M “‘I-labeled gp69/71 (3 ng, 8 X IO4 cpm/ng) at 24°C for the indicated periods of time. Binding activity was determined as indicated under Materials and Methods. Another set of cells were treated identically, except that after 2 h of incubation (as indicated by the arrow) the supernatant fluids were replaced with fresh reaction mixtures containing 2.1 x 10-l” M ‘251-labeled gp69/71 and the incubation was continued. 0, specific binding; A, nonspecific binding; 0, specific binding with fresh incubation mixture.
AND
AUGUST
NUMBER
OF CELLS
x IO-*
FIG. 3. Binding as a function of cell concentration. Different numbers of NIH 3T3 or mink cells were incubated with 2.1 x 10~“’ M %-labeled gp69/71 (3 ng, 8 x lo4 cpm/ng) for 2 h at 24’C, and in a total volume of 0.2 ml, as described under Materials and Methods. 0, specific binding; A, nonspecific binding.
tation in the amount of bound glycoprotein, despite an excess of cellular receptors, corresponded to that observed by analysis of the time course of the reaction and may be attributed to a depletion of active gp69/71. This was shown by the additional binding of gp69/71 when fresh substrate was added to a reaction mixture after two hi-s of incubation (Fig. 2). Moreover, in recent experiments, with reduced specific activity of labeling with iodide-125, a much greater proportion of 12”1-labeled glycoprotein will bind to cells, as much as 50% to 70% of total added substrate (Hughes, unpublished observations). Cell specificity of binding. The binding of radiolabeled gp69/71 was specific to cells permissive to Rauscher virus infection (NIH 3T3, BALB/c 3T3, and SC-l) (Table I). This included a transformed derivative of BALB/c 3T3 cells infected by Kirsten sarcoma virus. Mink, SIRC (rabbit corneal cell), and Hu 8085 (human muscle-skin) cells which are nonpermissive to Rauscher virus, all failed to bind ‘251-labeled gp69/71. The specificity of the reaction was also demonstrated with respect to known patterns of viral interference. Cells productively infected with a murine ecotropic virus failed to bind ‘251-labeled gp69/71; these included NIH 3T3 cells infected with either Rauscher, Gross or BALB/c N-tropic virus, BALB/c cells infected with the endogenous
CELL
RECEPTORS
FOR
ONCOVIRUS TABLE
CELL SPECIFICITY Cells
OF BINDING ““I-labeled Total
Uninfected, murine BALB/c 3T3 NIH 3T3 Swiss 3T3FL SC-1 Uninfected, non-murine Mink SIRC Hu 8085 Virus infected NIH (Rauscher) NIH (Gross) NIH (BALB N-tropic) NIH (Moloney sarcoma) BALB/c (BALB B-tropic) BALB/c (K&ten sarcoma) SC-1 (Rauscher) SC-1 (BALB X-tropic)
31,500 36,000 37,000 38,300
165
GLYCOPROTEIN
I OF?-LABELED gp69/71
bound
Nonspecific 900 l,ooo 900 1,250
gp69/71” Specific
binding fmol
cpm 30,600 35,000 36,100 37,050
l,W 900 1,100
l,W l,OQo 1,050
0 0 50
Loo0 1,100 WOO 30,000 1,250 27,000 Loo0 41,000
1,050 Loo0 1,050 l,W 1,050 MC@ l,W l,W
0 loo 0 29,000 200 26,000 0 40,000
5.3 6.1 6.3 6.5
Vol.
OF BIOCHEMISTRY
189, No. 1, July,
AND
pp. 161-171,
BIOPHYSICS
I978
Cellular Membrane Receptors for Oncovirus Envelope Glycoprotein: Properties of the Binding Reaction and Influence of Different Reagents on the Substrate and the Receptors SUBAL Department
BISHAYEE,’
of Pharmacology
METTE
STRAND,
and Experimental Therapeutics, Medicine, Baltimore, Maryland Received
November
14, 1977; revised
J. THOMAS
AND
The Johns 21205 March
Hopkins
AUGUST” University
School
of
14, 1978
Binding of purified envelope glycoprotein (gp69/71) of Rauscher murine type-C oncovirus to cellular membrane receptors has been analyzed with reaction systems using intact cells or membranes of disrupted cells. The reaction was highly specific; only cells permissive to infection by Rauscher virus bound the ‘““I-labeled viral glycoprotein. The specificity of binding was also demonstrated with respect to virus interference; cells productively infected with murine ecotropic type-C virus failed to bind the virus envelope glycoprotein whereas permissive cells infected with murine xenotropic virus continued to bind the Rauscher ecotropic virus glycoprotein. The reaction required the presence of Ca”, and was rapid and reversible with an equilibrium constant of about 8.3 X 10” Mm’ and an association rate constant of about 6.0 X 10” Mm’Sm’ at 24’C. The cellular receptors were saturated by about 4 x 10” molecules of glycoprotein per NIH 3T3 cell. No degradation or modification of bound glycoprotein was detected. The binding activity of the purified glycoprotein was susceptible to treatment with urea or guanidine-HCl; modification of tyrosyl, histidyl or tryptophan residues or the reduction of disulfide bonds also blocked the reaction. Sialic acid of the glycoprotein appeared not to be required. The receptor activity of a crude membrane fraction of disruated cells was abolished by treatment with trypsin, a-chymotrypsin, or phospholipase C.
Infection of a host cell by oncoviruses is known to involve a specific interaction between an envelope component of the virion and a receptor component of the cell membrane. The role of the envelope in this interaction was recognized chiefly by the correlation between host range, interference and neutralization properties within subgroups of avian oncoviruses, and by the observations that these properties were modified by phenotypic mixing and the formation of vii-ion pseudotypes (l-3). As proposed by Rubin (2) and Vogt and Ishizaki (3), the most obvious function of the envelope occurs at some early step of infection, during absorption or penetration. Early biological studies with the avian system by ’ Present address: Indian Medicine, 4, Raja Subodh 700032, India. ’ To whom correspondence
Institute Mullick should
Steck and Rubin (5, 6) and Piraino (7) suggested that the initial attachment of the virion to the cell may be random and that the specificity of infection is manifest at some later step in infection, such as the binding of specific receptors in the cell membrane or the penetration of the virus. These results were inconclusive, however, as it was found that specific interference by the infecting virus could be established under conditions whereby the virus particles responsible for interference appeared to remain at the cell surface (5). Progress in studies of the viral envelope has been achieved by the purification and characterization of an envelope glycoprotein of about 70,000 daltons from several murine viruses (8-10). This glycroprotein is the major constituent of the virion surface (11) and antibody prepared against it will neutralize the virus (12-14). Several studies have shown that the isolated glycoprotein
of Experimental Road, Calcutta be directed. 161
0003-9861/78/1891-0161$0200/O Copyright 0 1978 by Academic Press, Inc. All rights of reproduction in any form reserved.
162
BISHAYEE,
STRAND,
will react with cellular receptors. Hunsmann et al. (12) and Schafer et al. (15), using purified glycoprotein of Friend leukemia virus, and Hung et al. (16), using crude avian virus glycoprotein, demonstrated that these preparations added to normally permissive cells interfered with infection by the corresponding virus. Recently, a major advance in these studies was achieved by DeLarco and Todero (17) who developed an in vitro binding assay for virus envelope glycoprotein and carried out an initial characterization of the binding of glycoprotein to cells in tissue culture. These findings showed that NIH 3T3 cells bound approximately 5 X lo” molecules of glycoprotein per cell and that the ecotropic virus envelope bound specifically to mouse cells. In the present studies, we further characterize the binding of purified envelope glycoprotein (gp69/71) of Rauscher murine virus to intact cells and cell membranes. The findings agree closely with those of DeLarco and Todaro (17) and extend these studies to the kinetic properties of the reaction and to analysis of some of the properties of the substrate, receptor and reaction product. MATERIALS
AND
METHODS
Envelope glycoprotein. The envelope glycoprotein (gp69/71) of Rauscher murine virus was purified by phosphocellulose column chromatography as described (18). The purified protein was iodinated with Na’? (Amersham/Searle) according to the method of Hunter (19) as previously described (20). “‘I-labeled gp69/71 had a specific activity of 40 to 60 $i/pg. Cells. All cells were routinely propagated in monolayers at 37°C in Dulbecco’s modification of Eagle medium supplemented with 10% heat inactivated fetal bovine serum (Flow Laboratories, Rockville, Maryland) and equilibrated in air with 5% CO*. For binding studies, the cells were transferred to 16-mm disposable multi-dish trays (Costar), and propagated for 24-30 h prior to assay. NIH 3T3 cells (21), BALB/c 3T3 cells (22), a clonal line of wild mouse embryo cells (SC-I) (23), mink lung cells (24), rabbit cornea1 cells (SIRC) (25), and human fetal muscle-skin cells (Hu 8085) (26) were kindly provided by Dr. A. Hackett. SC-1 cells infected with AKR-Ll (27) or BALB/c mouse xenotropic virus (28) were kindly provided by Dr. J. Hartley. SC-1 cells and NIH 3T3 cells infected with Rauscher leukemia virus (29) were gifta from Dr. J. A. Bilello. Swiss 3T3FL cells and D56 S’L-, a cloned transformed derivative of this cell infected by Moloney sarcoma virus (30)
AND
AUGUST
were obtained from Dr. R. D. Bassin. NIH 3T3 cells infected with the endogenous BALB/c N-tropic virus (N-Cl-35) and BALB/c 3T3 cells infected with Btropic virus (B-Cl-11) (31) were provided by Dr. P. Besmer. Kirsten murine sarcoma virus (32) transformed nonproducer BALB/c 3T3 cells, BALB/c (KiMSV) (33) were provided by Dr. C. Croce. NIH 3T3 cells infected with Gross (34) type-C virus were described previously (35). Binding of purified envelope glycoprotein in NIH 3T3 cells. Binding of ““I-labeled gp69/71 to NIH 3T3 mouse cells in tissue culture was measured with an assay system using confluent monolayers of viable cells propagated for 24-30 h in 16-mm multi-dish trays. The cells were rinsed twice at 24°C with l.O-ml portions of the reaction buffer consisting of 116 mM NaCI, 53 mM KCl, 25 mM NaHCO:%, 55 mM glucose, 0.8 mM MgCL, 10 mM CaCb, 50 mu N,N-bis(2-hydroxyethyl)2-amino ethanesulfonic acid (BES), pH 7.4, and 1.0 mg/ml of bovine serum albumin. ““I-labeled gp69/71 (6-8 x lo4 cpm/ng), l-3 ng in 0.2 ml of reaction burfer, was added and the cells were incubated at 24°C for the desired time (30-120 min). The supernatant solution was then removed and the monolayers were rinsed 3 times with 3.0.ml portions of ice-cold reaction buffer. The cells were dispersed with 0.5 ml of 0.2 N NaOH for 15-20 min at 70°C and the solution was transferred with two 0.5-ml aliquots of 0.2 N HCl to glass tubes for counting (36). The background or nonspecific binding initially was determined either by preincubating the cells for 30 min with a 700-fold excess of unlabeled gp69/71 under standard assay conditions, or by the radioactivity bound to mink cells. All assays were carried out in duplicate or triplicate. Specific binding was determined by subtracting the background binding from the total radioactive uptake. Preparation of membranes. Cell monolayers were rinsed twice with 20 mu Tris-HCI, pH 7.4 containing 0.15 M NaCl, removed from the surface of the T-150 flasks (Costar) with the help of a rubber policeman rod, suspended in the above buffer, and then centrifuged at 4°C. The packed cells were washed with the same buffer, suspended in 2.0 ml of 20 mM Tris-HCl, pH 7.4 (1.25 x lo7 cells/ml), homogenized with 25 strokes using a tightly fitted Dounce homogenizer, and centrifuged at 23,000g for 30 min. The pellet was suspended in 2.0 ml of the above buffer and centrifuged at 1,OOOg for 30 s to remove the nuclei. Bovine serum albumin (10 mg/ml) was added to the supernatant membrane fraction which was then frozen at -20°C until use. Threefold freezing and thawing of this crude membrane preparation did not result in the loss of ‘?labeled gp69/71 binding. Nearly 90% of the gp69/71 binding activity of intact cells is present in this crude membrane preparation. Binding of ““I-labeled gp69/71 with isolated membranes. The binding assay with isolated membranes was carried out using Millipore EHWP filters (pore size 0.5 am). The standard reaction mixture, in a total
CELL
RECEPTORS
FOR
ONCOVIRUS
volume of 0.1 ml, containing 50 mu Tris-HCl, pH 7.4, 10 mg/ml bovine serum albumin, 10 mM CaC12, l-5 ng of ‘ZSI-labeled gp69/71 and the membrane fraction, was incubated at 24°C for 1 h. The reaction was terminated by adding 3 ml of ice-cold reaction buffer containing 0.1 mg/ml of bovine serum albumin. The membranes were collected immediately under reduced pressure on EHWP filters, presoaked in fetal bovine serum, and washed with 12-15 ml of buffer solution. The radioactivity retained in the filter was counted. Nonspecific binding was determined by preincubating the membrane fractions with a 700-fold excess of unlabeled gp69/71. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis. NIH 3T3 cells (1.3 x lo”) were incubated with ““I-labeled gp69/71. After washing, the cells were solubihzed with sample buffer containing 62.5 mM Tris-HCI, pH 6.8, 2% sodium dodecyl sulfate, 5% 2mercaptoethanol, and 10% glycerol. The proteins in the supernatant of the incubation mixture were precipitated with 7% trichloroacetic acid and the pellet was washed once with 0.5 ml of cold acetone. The precipitate was solubilized in the sample buffer. The samples were heated at 1oO’C for 1 min and then analyzed by electrophoresis in a 5-20% gradient polyacrylamide gel with an electrode buffer containing 0.1% sodium dodecyl sulfate in 25 mu Tris-192 mM glycine, pH 8.3 (37). Autoradiography of the gel, after fling and drying, was carried out with Kodak No Screen film (Kodak NS54T). Protein determination. Protein was determined according to the method of Lowry et al. (38). Reagents. The 30,000-dalton structural protein of Rauscher murine type-C virus (~30) was purified by ion-exchange column chromatography and Sephadex gel filtration as described elsewhere (18). Fetuin, ceruloplasmin, ovomucoid, soybean trypsin inhibitor, and bovine serum albumin were obtained from Sigma Chemical Company. Phospholipase C from Clostridiumperfringens (94 units/mg), trypsin (199 units/mg), a-chymotrypsin (61 units/mg), neuraminidase from C. perfringens (14 unita/mg), /3-galactosidase (360 units/mg), DNase (2300 units/mg), and RNase (2700 units/mg) were obtained from Worthington Biochemical Company. Goat anti-Rauscher gp69/71 serum was prepared and characterized as described elsewhere (18). All other reagents were of the highest purity available from commercial sources. RESULTS
Properties of gp69/ 71-cell interaction. Binding of lz51-labeled gp69/71 to cells was routinely measured using monolayers of cultured cells washed and incubated with a buffer solution containing NaCl, KCl, MgC12, CaC12, glucose and bovine serum albumin, as described under Materials and Methods. Of these constituents, the reac-
GLYCOPROTEIN
163
tion appeared to be strictly dependent only upon Ca’+; there was a 7-fold increase in the binding at 10 mu Ca2+. Ca2’ could not be replaced by Mg+, and NaCl and KC1 appeared not to be required, as their replacement by isotonic sucrose (0.25 M) had no significant effect on the reaction. The pH optima was broad with 90% of maximum binding observed between pH 6.8 to 8.0. Bovine serum albumin was added to the reaction in order to saturate nonspecific sites of protein binding to cells; its omission had no effect on specific gp69/71 binding but appeared to slightly increase background absorption of substrate. Several other proteins and glycoproteins such as the p30 structural protein of Rauscher virus, fetuin, ceruloplasmin or ovomucoid had no effect upon gp69/71 binding. Of the proteins tested, only unlabeled gp69/71 reduced the binding of ‘251-labeled gp69/71 in a fashion suggesting the chemical similarity of labeled and unlabeled substrates (data not shown). Kinetics of binding. Saturation of cellular binding sites was analyzed using 2.6 x lo4 NIH 3T3 cells and increasing concentrations of 1251-labeled gp69/71. The apparent saturation of receptor sites was achieved by the addition of approximately 30 ng of gp69/71, with the binding of 1.18 ng of gp69/71, or approximately 16.5 fmol (Fig. 1). Thus, the number of receptors per cell was approximately 4 X lo”. A Scatchard
”% ./ : 05 / ? i M’“k__-. I. ,< --&---.--, -I---&--, 20 40 FIG. 1. Binding as a function of “‘I-labeled gp69/71 concentration. NIH 3T3 or mink cells (2.6 x 104) were incubated with different amounts of ‘251-labeled gp69/71 (8 x lo4 cpm/ng) in a total volume of 0.2 ml for 2 h at 24°C. Assay system is described under Materials and Methods. 0, specific binding; A, nonspecific binding.
164
BISHAYEE,
STRAND,
plot (39) of this data, corrected for the active fraction of the substrate as described below, gave an equilibrium association constant, Ka, of 8.3 X 10' M-i, and the number of binding sites as 4.5 x lo5 per cell. Similar values were obtained when the same experiment was carried out by adding unlabeled gp69/71 to a constant amount of ““I-labeled gp69/71. The binding of 12”1-labeled gp69/71 to NIH 3T3 cells was time and temperature dependent; at 24°C the reaction was linear for 20 min and reached a maximum in 2 h (Fig. 2). The reaction was 6 times slower at 4°C (data not shown). Utilizing the secondorder equation Fzi= 2.303/t (P-R) log R (Py)/P (R-y), where P is the gp69/71 concentration, R is the receptor concentration as determined by Scatchard plot, y is the amount of P reacting in time t, and assuming that 20% of labeled gp69/71 was active (see below), the association rate constant was calculated to be 6 x lo5 M-IS-’ at 24’C. Using 3 ng of ‘251-labeled gp69/71, binding of the substrate was proportional to cell concentration below 2.6 x lo4 cells and reached a maximum at 1.3 x lo5 cells (Fig. 3). In this experiment, the maximum extent of glycoprotein binding was approximately 13% of the total gp69/71 added. This limi-
FIG. 2. Time course of the binding of ‘?-labeled gp69/71 with NIH 3T3 cells. NIH 3T3 or mink cells (1.3 x 1W5) were incubated in a total volume of 0.2 ml with 2.1 X IO-‘” M “‘I-labeled gp69/71 (3 ng, 8 X IO4 cpm/ng) at 24°C for the indicated periods of time. Binding activity was determined as indicated under Materials and Methods. Another set of cells were treated identically, except that after 2 h of incubation (as indicated by the arrow) the supernatant fluids were replaced with fresh reaction mixtures containing 2.1 x 10-l” M ‘251-labeled gp69/71 and the incubation was continued. 0, specific binding; A, nonspecific binding; 0, specific binding with fresh incubation mixture.
AND
AUGUST
NUMBER
OF CELLS
x IO-*
FIG. 3. Binding as a function of cell concentration. Different numbers of NIH 3T3 or mink cells were incubated with 2.1 x 10~“’ M %-labeled gp69/71 (3 ng, 8 x lo4 cpm/ng) for 2 h at 24’C, and in a total volume of 0.2 ml, as described under Materials and Methods. 0, specific binding; A, nonspecific binding.
tation in the amount of bound glycoprotein, despite an excess of cellular receptors, corresponded to that observed by analysis of the time course of the reaction and may be attributed to a depletion of active gp69/71. This was shown by the additional binding of gp69/71 when fresh substrate was added to a reaction mixture after two hi-s of incubation (Fig. 2). Moreover, in recent experiments, with reduced specific activity of labeling with iodide-125, a much greater proportion of 12”1-labeled glycoprotein will bind to cells, as much as 50% to 70% of total added substrate (Hughes, unpublished observations). Cell specificity of binding. The binding of radiolabeled gp69/71 was specific to cells permissive to Rauscher virus infection (NIH 3T3, BALB/c 3T3, and SC-l) (Table I). This included a transformed derivative of BALB/c 3T3 cells infected by Kirsten sarcoma virus. Mink, SIRC (rabbit corneal cell), and Hu 8085 (human muscle-skin) cells which are nonpermissive to Rauscher virus, all failed to bind ‘251-labeled gp69/71. The specificity of the reaction was also demonstrated with respect to known patterns of viral interference. Cells productively infected with a murine ecotropic virus failed to bind ‘251-labeled gp69/71; these included NIH 3T3 cells infected with either Rauscher, Gross or BALB/c N-tropic virus, BALB/c cells infected with the endogenous
CELL
RECEPTORS
FOR
ONCOVIRUS TABLE
CELL SPECIFICITY Cells
OF BINDING ““I-labeled Total
Uninfected, murine BALB/c 3T3 NIH 3T3 Swiss 3T3FL SC-1 Uninfected, non-murine Mink SIRC Hu 8085 Virus infected NIH (Rauscher) NIH (Gross) NIH (BALB N-tropic) NIH (Moloney sarcoma) BALB/c (BALB B-tropic) BALB/c (K&ten sarcoma) SC-1 (Rauscher) SC-1 (BALB X-tropic)
31,500 36,000 37,000 38,300
165
GLYCOPROTEIN
I OF?-LABELED gp69/71
bound
Nonspecific 900 l,ooo 900 1,250
gp69/71” Specific
binding fmol
cpm 30,600 35,000 36,100 37,050
l,W 900 1,100
l,W l,OQo 1,050
0 0 50
Loo0 1,100 WOO 30,000 1,250 27,000 Loo0 41,000
1,050 Loo0 1,050 l,W 1,050 MC@ l,W l,W
0 loo 0 29,000 200 26,000 0 40,000
5.3 6.1 6.3 6.5
